diff --git a/.mailmap b/.mailmap index 482dcfff40af1604aa78904a4db6871bca51a7cc..033955129e454c6f6f61cd0b174f55e97d6e53f6 100644 --- a/.mailmap +++ b/.mailmap @@ -233,6 +233,7 @@ Jisheng Zhang Johan Hovold Johan Hovold John Crispin +John Keeping John Paul Adrian Glaubitz John Stultz @@ -365,6 +366,11 @@ Nicolas Pitre Nicolas Saenz Julienne Nicolas Saenz Julienne Niklas Söderlund +Nikolay Aleksandrov +Nikolay Aleksandrov +Nikolay Aleksandrov +Nikolay Aleksandrov +Nikolay Aleksandrov Oleksandr Natalenko Oleksij Rempel Oleksij Rempel diff --git a/CREDITS b/CREDITS index 2d9da9a7defa666cbfcd2aab7fcca821f2027066..de7e4dbbc5991194ce9bcaeb94a368e79d79832a 100644 --- a/CREDITS +++ b/CREDITS @@ -1706,6 +1706,10 @@ S: Panoramastrasse 18 S: D-69126 Heidelberg S: Germany +N: Neil Horman +M: nhorman@tuxdriver.com +D: SCTP protocol maintainer. + N: Simon Horman M: horms@verge.net.au D: Renesas ARM/ARM64 SoC maintainer diff --git a/Documentation/admin-guide/cgroup-v2.rst b/Documentation/admin-guide/cgroup-v2.rst index f67c0829350b8e0ae9120cf873fedc118bab72e7..e592a93644739f0b1812fa5754c2f93e8b2bf08d 100644 --- a/Documentation/admin-guide/cgroup-v2.rst +++ b/Documentation/admin-guide/cgroup-v2.rst @@ -1213,23 +1213,25 @@ PAGE_SIZE multiple when read back. A read-write single value file which exists on non-root cgroups. The default is "max". - Memory usage throttle limit. This is the main mechanism to - control memory usage of a cgroup. If a cgroup's usage goes + Memory usage throttle limit. If a cgroup's usage goes over the high boundary, the processes of the cgroup are throttled and put under heavy reclaim pressure. Going over the high limit never invokes the OOM killer and - under extreme conditions the limit may be breached. + under extreme conditions the limit may be breached. The high + limit should be used in scenarios where an external process + monitors the limited cgroup to alleviate heavy reclaim + pressure. memory.max A read-write single value file which exists on non-root cgroups. The default is "max". - Memory usage hard limit. This is the final protection - mechanism. If a cgroup's memory usage reaches this limit and - can't be reduced, the OOM killer is invoked in the cgroup. - Under certain circumstances, the usage may go over the limit - temporarily. + Memory usage hard limit. This is the main mechanism to limit + memory usage of a cgroup. If a cgroup's memory usage reaches + this limit and can't be reduced, the OOM killer is invoked in + the cgroup. Under certain circumstances, the usage may go + over the limit temporarily. In default configuration regular 0-order allocations always succeed unless OOM killer chooses current task as a victim. @@ -1238,10 +1240,6 @@ PAGE_SIZE multiple when read back. Caller could retry them differently, return into userspace as -ENOMEM or silently ignore in cases like disk readahead. - This is the ultimate protection mechanism. As long as the - high limit is used and monitored properly, this limit's - utility is limited to providing the final safety net. - memory.reclaim A write-only nested-keyed file which exists for all cgroups. diff --git a/Documentation/admin-guide/cifs/changes.rst b/Documentation/admin-guide/cifs/changes.rst index 3147bbae9c43f21a41c5e110404821b5c2027e62..8c42c4de510b1046eeca71db2dd2e15bbde11897 100644 --- a/Documentation/admin-guide/cifs/changes.rst +++ b/Documentation/admin-guide/cifs/changes.rst @@ -5,5 +5,5 @@ Changes See https://wiki.samba.org/index.php/LinuxCIFSKernel for summary information about fixes/improvements to CIFS/SMB2/SMB3 support (changes to cifs.ko module) by kernel version (and cifs internal module version). -This may be easier to read than parsing the output of "git log fs/cifs" -by release. +This may be easier to read than parsing the output of +"git log fs/smb/client" by release. diff --git a/Documentation/admin-guide/cifs/usage.rst b/Documentation/admin-guide/cifs/usage.rst index 2e151cd8c2e4de65c13d5c1f51b346911bc45d98..5f936b4b601881d3e031b71a1cc258c5ddb9585e 100644 --- a/Documentation/admin-guide/cifs/usage.rst +++ b/Documentation/admin-guide/cifs/usage.rst @@ -45,7 +45,7 @@ Installation instructions If you have built the CIFS vfs as module (successfully) simply type ``make modules_install`` (or if you prefer, manually copy the file to -the modules directory e.g. /lib/modules/2.4.10-4GB/kernel/fs/cifs/cifs.ko). +the modules directory e.g. /lib/modules/6.3.0-060300-generic/kernel/fs/smb/client/cifs.ko). If you have built the CIFS vfs into the kernel itself, follow the instructions for your distribution on how to install a new kernel (usually you @@ -66,15 +66,15 @@ If cifs is built as a module, then the size and number of network buffers and maximum number of simultaneous requests to one server can be configured. Changing these from their defaults is not recommended. By executing modinfo:: - modinfo kernel/fs/cifs/cifs.ko + modinfo -on kernel/fs/cifs/cifs.ko the list of configuration changes that can be made +on kernel/fs/smb/client/cifs.ko the list of configuration changes that can be made at module initialization time (by running insmod cifs.ko) can be seen. Recommendations =============== -To improve security the SMB2.1 dialect or later (usually will get SMB3) is now +To improve security the SMB2.1 dialect or later (usually will get SMB3.1.1) is now the new default. To use old dialects (e.g. to mount Windows XP) use "vers=1.0" on mount (or vers=2.0 for Windows Vista). Note that the CIFS (vers=1.0) is much older and less secure than the default dialect SMB3 which includes diff --git a/Documentation/admin-guide/quickly-build-trimmed-linux.rst b/Documentation/admin-guide/quickly-build-trimmed-linux.rst index ff4f4cc8522bdbfa018269c1704a95827bd81c26..f08149bc53f84dd427e514d7eb898169030da8a7 100644 --- a/Documentation/admin-guide/quickly-build-trimmed-linux.rst +++ b/Documentation/admin-guide/quickly-build-trimmed-linux.rst @@ -215,12 +215,14 @@ again. reduce the compile time enormously, especially if you are running an universal kernel from a commodity Linux distribution. - There is a catch: the make target 'localmodconfig' will disable kernel - features you have not directly or indirectly through some program utilized - since you booted the system. You can reduce or nearly eliminate that risk by - using tricks outlined in the reference section; for quick testing purposes - that risk is often negligible, but it is an aspect you want to keep in mind - in case your kernel behaves oddly. + There is a catch: 'localmodconfig' is likely to disable kernel features you + did not use since you booted your Linux -- like drivers for currently + disconnected peripherals or a virtualization software not haven't used yet. + You can reduce or nearly eliminate that risk with tricks the reference + section outlines; but when building a kernel just for quick testing purposes + it is often negligible if such features are missing. But you should keep that + aspect in mind when using a kernel built with this make target, as it might + be the reason why something you only use occasionally stopped working. [:ref:`details`] @@ -271,6 +273,9 @@ again. does nothing at all; in that case you have to manually install your kernel, as outlined in the reference section. + If you are running a immutable Linux distribution, check its documentation + and the web to find out how to install your own kernel there. + [:ref:`details`] .. _another_sbs: @@ -291,29 +296,29 @@ again. version you care about, as git otherwise might retrieve the entire commit history:: - git fetch --shallow-exclude=v6.1 origin - - If you modified the sources (for example by applying a patch), you now need - to discard those modifications; that's because git otherwise will not be able - to switch to the sources of another version due to potential conflicting - changes:: - - git reset --hard + git fetch --shallow-exclude=v6.0 origin - Now checkout the version you are interested in, as explained above:: + Now switch to the version you are interested in -- but be aware the command + used here will discard any modifications you performed, as they would + conflict with the sources you want to checkout:: - git checkout --detach origin/master + git checkout --force --detach origin/master At this point you might want to patch the sources again or set/modify a build - tag, as explained earlier; afterwards adjust the build configuration to the - new codebase and build your next kernel:: + tag, as explained earlier. Afterwards adjust the build configuration to the + new codebase using olddefconfig, which will now adjust the configuration file + you prepared earlier using localmodconfig (~/linux/.config) for your next + kernel:: # reminder: if you want to apply patches, do it at this point # reminder: you might want to update your build tag at this point make olddefconfig + + Now build your kernel:: + make -j $(nproc --all) - Install the kernel as outlined above:: + Afterwards install the kernel as outlined above:: command -v installkernel && sudo make modules_install install @@ -584,11 +589,11 @@ versions and individual commits at hand at any time:: curl -L \ https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/clone.bundle \ -o linux-stable.git.bundle - git clone clone.bundle ~/linux/ + git clone linux-stable.git.bundle ~/linux/ rm linux-stable.git.bundle cd ~/linux/ - git remote set-url origin - https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git + git remote set-url origin \ + https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git git fetch origin git checkout --detach origin/master diff --git a/Documentation/cdrom/index.rst b/Documentation/cdrom/index.rst index e87a8785bc1a7172e7ed0ae897da3ba994b87587..e9b022d7093957d0e698cbc2ba0a303e3f6e4b4c 100644 --- a/Documentation/cdrom/index.rst +++ b/Documentation/cdrom/index.rst @@ -1,8 +1,8 @@ .. SPDX-License-Identifier: GPL-2.0 -===== -cdrom -===== +====== +CD-ROM +====== .. toctree:: :maxdepth: 1 diff --git a/Documentation/devicetree/bindings/ata/ahci-common.yaml b/Documentation/devicetree/bindings/ata/ahci-common.yaml index 7fdf40954a4c6798211d8b9178248f3d7f1a02b7..38770c4c85fd5defe674fca48d81e638eb2f1f97 100644 --- a/Documentation/devicetree/bindings/ata/ahci-common.yaml +++ b/Documentation/devicetree/bindings/ata/ahci-common.yaml @@ -8,7 +8,7 @@ title: Common Properties for Serial ATA AHCI controllers maintainers: - Hans de Goede - - Damien Le Moal + - Damien Le Moal description: This document defines device tree properties for a common AHCI SATA diff --git a/Documentation/devicetree/bindings/ata/ceva,ahci-1v84.yaml b/Documentation/devicetree/bindings/ata/ceva,ahci-1v84.yaml index 9b31f864e071ee7c53f63eb87db631eb20c1e075..71364c6081ff50d9573102aa48c9bbaa4ea75f1e 100644 --- a/Documentation/devicetree/bindings/ata/ceva,ahci-1v84.yaml +++ b/Documentation/devicetree/bindings/ata/ceva,ahci-1v84.yaml @@ -32,7 +32,7 @@ properties: maxItems: 1 iommus: - maxItems: 1 + maxItems: 4 power-domains: maxItems: 1 diff --git a/Documentation/devicetree/bindings/cache/qcom,llcc.yaml b/Documentation/devicetree/bindings/cache/qcom,llcc.yaml index d8b91944180ae1d03edd7fb99091bf7ccfe81b3c..44892aa589fd00215fc284077b95b3d0ed5a7e42 100644 --- a/Documentation/devicetree/bindings/cache/qcom,llcc.yaml +++ b/Documentation/devicetree/bindings/cache/qcom,llcc.yaml @@ -129,6 +129,7 @@ allOf: - qcom,sm8250-llcc - qcom,sm8350-llcc - qcom,sm8450-llcc + - qcom,sm8550-llcc then: properties: reg: diff --git a/Documentation/devicetree/bindings/clock/canaan,k210-clk.yaml b/Documentation/devicetree/bindings/clock/canaan,k210-clk.yaml index 998e5cce652f32085e5e75797d5d7df1874a92e2..380cb6d8002529cdc20d01811e7c6720ab4408f3 100644 --- a/Documentation/devicetree/bindings/clock/canaan,k210-clk.yaml +++ b/Documentation/devicetree/bindings/clock/canaan,k210-clk.yaml @@ -7,7 +7,7 @@ $schema: http://devicetree.org/meta-schemas/core.yaml# title: Canaan Kendryte K210 Clock maintainers: - - Damien Le Moal + - Damien Le Moal description: | Canaan Kendryte K210 SoC clocks driver bindings. The clock diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,aal.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,aal.yaml index 92741486c24dd301b9e4d2db93a94f1a34969fdc..7fd42c8fdc3282fdc87403a8591e817a85b7eff7 100644 --- a/Documentation/devicetree/bindings/display/mediatek/mediatek,aal.yaml +++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,aal.yaml @@ -27,6 +27,7 @@ properties: - items: - enum: - mediatek,mt2712-disp-aal + - mediatek,mt6795-disp-aal - const: mediatek,mt8173-disp-aal - items: - enum: diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,color.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,color.yaml index d0ea77fc4b069ba67b89331e2af59db6e2eaefa4..f21e4409204367dd3cfd946deada11d0c54965cf 100644 --- a/Documentation/devicetree/bindings/display/mediatek/mediatek,color.yaml +++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,color.yaml @@ -33,6 +33,7 @@ properties: - const: mediatek,mt2701-disp-color - items: - enum: + - mediatek,mt6795-disp-color - mediatek,mt8183-disp-color - mediatek,mt8186-disp-color - mediatek,mt8188-disp-color diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml index d976380801e3bf08078222053f1966c08035c84e..803c00f2620694f8a578617c628a547cad37cac5 100644 --- a/Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml +++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml @@ -17,15 +17,20 @@ description: | properties: compatible: - enum: - - mediatek,mt2701-dpi - - mediatek,mt7623-dpi - - mediatek,mt8173-dpi - - mediatek,mt8183-dpi - - mediatek,mt8186-dpi - - mediatek,mt8188-dp-intf - - mediatek,mt8192-dpi - - mediatek,mt8195-dp-intf + oneOf: + - enum: + - mediatek,mt2701-dpi + - mediatek,mt7623-dpi + - mediatek,mt8173-dpi + - mediatek,mt8183-dpi + - mediatek,mt8186-dpi + - mediatek,mt8188-dp-intf + - mediatek,mt8192-dpi + - mediatek,mt8195-dp-intf + - items: + - enum: + - mediatek,mt6795-dpi + - const: mediatek,mt8183-dpi reg: maxItems: 1 diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,dsi.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,dsi.yaml index 4707b60238b07b2d910da1c3e43cfbc8aa9e0df9..12441b937684d5924f6423b1d3460d70e26886fc 100644 --- a/Documentation/devicetree/bindings/display/mediatek/mediatek,dsi.yaml +++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,dsi.yaml @@ -22,13 +22,18 @@ allOf: properties: compatible: - enum: - - mediatek,mt2701-dsi - - mediatek,mt7623-dsi - - mediatek,mt8167-dsi - - mediatek,mt8173-dsi - - mediatek,mt8183-dsi - - mediatek,mt8186-dsi + oneOf: + - enum: + - mediatek,mt2701-dsi + - mediatek,mt7623-dsi + - mediatek,mt8167-dsi + - mediatek,mt8173-dsi + - mediatek,mt8183-dsi + - mediatek,mt8186-dsi + - items: + - enum: + - mediatek,mt6795-dsi + - const: mediatek,mt8173-dsi reg: maxItems: 1 diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,gamma.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,gamma.yaml index 6c2be9d6840b18a14030f52c48e4f144e84cc4e7..c6641acd75d6a0fb1c74a1269a1f3eedfbc3bb88 100644 --- a/Documentation/devicetree/bindings/display/mediatek/mediatek,gamma.yaml +++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,gamma.yaml @@ -24,6 +24,10 @@ properties: - enum: - mediatek,mt8173-disp-gamma - mediatek,mt8183-disp-gamma + - items: + - enum: + - mediatek,mt6795-disp-gamma + - const: mediatek,mt8173-disp-gamma - items: - enum: - mediatek,mt8186-disp-gamma diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,merge.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,merge.yaml index 2f8e2f4dc3b8258be604cec8811f05a3f9a0350b..eead5cb8636edbf3a64ea208fbdff9348a9f7984 100644 --- a/Documentation/devicetree/bindings/display/mediatek/mediatek,merge.yaml +++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,merge.yaml @@ -24,6 +24,9 @@ properties: - enum: - mediatek,mt8173-disp-merge - mediatek,mt8195-disp-merge + - items: + - const: mediatek,mt6795-disp-merge + - const: mediatek,mt8173-disp-merge reg: maxItems: 1 diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,od.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,od.yaml index 29f9fa8f82195d887e20c08d9fb94b5dfb0f2aca..831c653caffd7d6424b7f59ee7ae3d105f79d4b6 100644 --- a/Documentation/devicetree/bindings/display/mediatek/mediatek,od.yaml +++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,od.yaml @@ -24,6 +24,9 @@ properties: - enum: - mediatek,mt2712-disp-od - mediatek,mt8173-disp-od + - items: + - const: mediatek,mt6795-disp-od + - const: mediatek,mt8173-disp-od reg: maxItems: 1 diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,ovl.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,ovl.yaml index 92e320d54ba2aee02f9dbd5e748ca61caa039225..3e1069b00b56ef43a1c030eec4f5e4f7bd782887 100644 --- a/Documentation/devicetree/bindings/display/mediatek/mediatek,ovl.yaml +++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,ovl.yaml @@ -31,6 +31,10 @@ properties: - mediatek,mt7623-disp-ovl - mediatek,mt2712-disp-ovl - const: mediatek,mt2701-disp-ovl + - items: + - enum: + - mediatek,mt6795-disp-ovl + - const: mediatek,mt8173-disp-ovl - items: - enum: - mediatek,mt8188-disp-ovl diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,rdma.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,rdma.yaml index 42059efad45de0497388b03909af12adfebf7662..39dbb5c8bcf80d8b80dd0543807e5ec658980664 100644 --- a/Documentation/devicetree/bindings/display/mediatek/mediatek,rdma.yaml +++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,rdma.yaml @@ -37,6 +37,10 @@ properties: - mediatek,mt7623-disp-rdma - mediatek,mt2712-disp-rdma - const: mediatek,mt2701-disp-rdma + - items: + - enum: + - mediatek,mt6795-disp-rdma + - const: mediatek,mt8173-disp-rdma - items: - enum: - mediatek,mt8186-disp-rdma diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,split.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,split.yaml index 21a4e96ecd936f67d0992d8e46748e786ac75bf0..a8a5c9608598f04b22f78d0fa7dbc36c9f07326f 100644 --- a/Documentation/devicetree/bindings/display/mediatek/mediatek,split.yaml +++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,split.yaml @@ -23,6 +23,9 @@ properties: oneOf: - enum: - mediatek,mt8173-disp-split + - items: + - const: mediatek,mt6795-disp-split + - const: mediatek,mt8173-disp-split reg: maxItems: 1 diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,ufoe.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,ufoe.yaml index 62fad23a26f5a2a2ab859983f22a2e5986531a9d..39e3e2d4a0db8bc0674c71b86b90df5257cf626d 100644 --- a/Documentation/devicetree/bindings/display/mediatek/mediatek,ufoe.yaml +++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,ufoe.yaml @@ -24,6 +24,9 @@ properties: oneOf: - enum: - mediatek,mt8173-disp-ufoe + - items: + - const: mediatek,mt6795-disp-ufoe + - const: mediatek,mt8173-disp-ufoe reg: maxItems: 1 diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,wdma.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,wdma.yaml index 991183165d29f25ea7e99cd3dd65c887345870f4..a3a2b71a452309394b2401aa1607a10b8fa1175c 100644 --- a/Documentation/devicetree/bindings/display/mediatek/mediatek,wdma.yaml +++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,wdma.yaml @@ -23,6 +23,9 @@ properties: oneOf: - enum: - mediatek,mt8173-disp-wdma + - items: + - const: mediatek,mt6795-disp-wdma + - const: mediatek,mt8173-disp-wdma reg: maxItems: 1 diff --git a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml index f0c2237d5f82eb8d2d81e60d0ea7019fd3e12214..7a7cf3fb3e6da0012b57721d0e1d3965e26b41ff 100644 --- a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml +++ b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml @@ -29,6 +29,7 @@ properties: - items: - enum: - qcom,sm8450-dp + - qcom,sm8550-dp - const: qcom,sm8350-dp reg: diff --git a/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml b/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml index e6c1ebfe8a324773cab5a53a4a98fd294cdd9c99..01848bdd5873716b0d92a3b5e4d55c1249fbb29d 100644 --- a/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml +++ b/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml @@ -15,6 +15,7 @@ properties: - items: - enum: - qcom,apq8064-dsi-ctrl + - qcom,msm8226-dsi-ctrl - qcom,msm8916-dsi-ctrl - qcom,msm8953-dsi-ctrl - qcom,msm8974-dsi-ctrl @@ -26,6 +27,8 @@ properties: - qcom,sdm660-dsi-ctrl - qcom,sdm845-dsi-ctrl - qcom,sm6115-dsi-ctrl + - qcom,sm6350-dsi-ctrl + - qcom,sm6375-dsi-ctrl - qcom,sm8150-dsi-ctrl - qcom,sm8250-dsi-ctrl - qcom,sm8350-dsi-ctrl @@ -82,6 +85,18 @@ properties: Indicates if the DSI controller is driving a panel which needs 2 DSI links. + qcom,master-dsi: + type: boolean + description: | + Indicates if the DSI controller is the master DSI controller when + qcom,dual-dsi-mode enabled. + + qcom,sync-dual-dsi: + type: boolean + description: | + Indicates if the DSI controller needs to sync the other DSI controller + with MIPI DCS commands when qcom,dual-dsi-mode enabled. + assigned-clocks: minItems: 2 maxItems: 4 @@ -244,6 +259,7 @@ allOf: compatible: contains: enum: + - qcom,msm8226-dsi-ctrl - qcom,msm8974-dsi-ctrl then: properties: @@ -285,6 +301,7 @@ allOf: contains: enum: - qcom,msm8998-dsi-ctrl + - qcom,sm6350-dsi-ctrl then: properties: clocks: @@ -352,6 +369,7 @@ allOf: enum: - qcom,sdm845-dsi-ctrl - qcom,sm6115-dsi-ctrl + - qcom,sm6375-dsi-ctrl then: properties: clocks: diff --git a/Documentation/devicetree/bindings/display/msm/dsi-phy-28nm.yaml b/Documentation/devicetree/bindings/display/msm/dsi-phy-28nm.yaml index cf4a338c46610243a855f8b359e097d6f73ebde1..62fb3e484eb291adc7f4728737f592692c59b3fe 100644 --- a/Documentation/devicetree/bindings/display/msm/dsi-phy-28nm.yaml +++ b/Documentation/devicetree/bindings/display/msm/dsi-phy-28nm.yaml @@ -15,10 +15,11 @@ allOf: properties: compatible: enum: + - qcom,dsi-phy-28nm-8226 + - qcom,dsi-phy-28nm-8960 - qcom,dsi-phy-28nm-hpm - qcom,dsi-phy-28nm-hpm-fam-b - qcom,dsi-phy-28nm-lp - - qcom,dsi-phy-28nm-8960 reg: items: diff --git a/Documentation/devicetree/bindings/display/msm/gmu.yaml b/Documentation/devicetree/bindings/display/msm/gmu.yaml index 029d72822d8b69c0a51c1a967bcfe0a05a752592..5fc4106110ad270a85e564476e2c827f945dfa14 100644 --- a/Documentation/devicetree/bindings/display/msm/gmu.yaml +++ b/Documentation/devicetree/bindings/display/msm/gmu.yaml @@ -19,16 +19,18 @@ description: | properties: compatible: - items: - - pattern: '^qcom,adreno-gmu-6[0-9][0-9]\.[0-9]$' - - const: qcom,adreno-gmu + oneOf: + - items: + - pattern: '^qcom,adreno-gmu-6[0-9][0-9]\.[0-9]$' + - const: qcom,adreno-gmu + - const: qcom,adreno-gmu-wrapper reg: - minItems: 3 + minItems: 1 maxItems: 4 reg-names: - minItems: 3 + minItems: 1 maxItems: 4 clocks: @@ -44,7 +46,6 @@ properties: - description: GMU HFI interrupt - description: GMU interrupt - interrupt-names: items: - const: hfi @@ -72,14 +73,8 @@ required: - compatible - reg - reg-names - - clocks - - clock-names - - interrupts - - interrupt-names - power-domains - power-domain-names - - iommus - - operating-points-v2 additionalProperties: false @@ -122,6 +117,7 @@ allOf: contains: enum: - qcom,adreno-gmu-635.0 + - qcom,adreno-gmu-660.1 then: properties: reg: @@ -217,6 +213,28 @@ allOf: - const: axi - const: memnoc + - if: + properties: + compatible: + contains: + const: qcom,adreno-gmu-wrapper + then: + properties: + reg: + items: + - description: GMU wrapper register space + reg-names: + items: + - const: gmu + else: + required: + - clocks + - clock-names + - interrupts + - interrupt-names + - iommus + - operating-points-v2 + examples: - | #include @@ -249,3 +267,12 @@ examples: iommus = <&adreno_smmu 5>; operating-points-v2 = <&gmu_opp_table>; }; + + gmu_wrapper: gmu@596a000 { + compatible = "qcom,adreno-gmu-wrapper"; + reg = <0x0596a000 0x30000>; + reg-names = "gmu"; + power-domains = <&gpucc GPU_CX_GDSC>, + <&gpucc GPU_GX_GDSC>; + power-domain-names = "cx", "gx"; + }; diff --git a/Documentation/devicetree/bindings/display/msm/gpu.yaml b/Documentation/devicetree/bindings/display/msm/gpu.yaml index 5dabe7b6794b7eb010d4f980032f79e1a33d5a8f..58ca8912a8c343753a6d8704aaed1267a2babe35 100644 --- a/Documentation/devicetree/bindings/display/msm/gpu.yaml +++ b/Documentation/devicetree/bindings/display/msm/gpu.yaml @@ -36,10 +36,7 @@ properties: reg-names: minItems: 1 - items: - - const: kgsl_3d0_reg_memory - - const: cx_mem - - const: cx_dbgc + maxItems: 3 interrupts: maxItems: 1 @@ -157,16 +154,62 @@ allOf: required: - clocks - clock-names + - if: properties: compatible: contains: - pattern: '^qcom,adreno-6[0-9][0-9]\.[0-9]$' - - then: # Since Adreno 6xx series clocks should be defined in GMU + enum: + - qcom,adreno-610.0 + - qcom,adreno-619.1 + then: properties: - clocks: false - clock-names: false + clocks: + minItems: 6 + maxItems: 6 + + clock-names: + items: + - const: core + description: GPU Core clock + - const: iface + description: GPU Interface clock + - const: mem_iface + description: GPU Memory Interface clock + - const: alt_mem_iface + description: GPU Alternative Memory Interface clock + - const: gmu + description: CX GMU clock + - const: xo + description: GPUCC clocksource clock + + reg-names: + minItems: 1 + items: + - const: kgsl_3d0_reg_memory + - const: cx_dbgc + + required: + - clocks + - clock-names + else: + if: + properties: + compatible: + contains: + pattern: '^qcom,adreno-6[0-9][0-9]\.[0-9]$' + + then: # Starting with A6xx, the clocks are usually defined in the GMU node + properties: + clocks: false + clock-names: false + + reg-names: + minItems: 1 + items: + - const: kgsl_3d0_reg_memory + - const: cx_mem + - const: cx_dbgc examples: - | diff --git a/Documentation/devicetree/bindings/display/msm/qcom,mdp5.yaml b/Documentation/devicetree/bindings/display/msm/qcom,mdp5.yaml index a763cf8da122f9bfd252f98081d58855f6940524..2fe032d0e8f8a99f904ff6f7c6f3779d25262289 100644 --- a/Documentation/devicetree/bindings/display/msm/qcom,mdp5.yaml +++ b/Documentation/devicetree/bindings/display/msm/qcom,mdp5.yaml @@ -22,6 +22,7 @@ properties: - items: - enum: - qcom,apq8084-mdp5 + - qcom,msm8226-mdp5 - qcom,msm8916-mdp5 - qcom,msm8917-mdp5 - qcom,msm8953-mdp5 diff --git a/Documentation/devicetree/bindings/display/msm/qcom,mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,mdss.yaml index b0100105e428e77e11e409ee210a93f376a31786..db9f07c6142d80ba59b6a86d12ebb3a210a3c677 100644 --- a/Documentation/devicetree/bindings/display/msm/qcom,mdss.yaml +++ b/Documentation/devicetree/bindings/display/msm/qcom,mdss.yaml @@ -125,6 +125,7 @@ patternProperties: - qcom,dsi-phy-14nm-660 - qcom,dsi-phy-14nm-8953 - qcom,dsi-phy-20nm + - qcom,dsi-phy-28nm-8226 - qcom,dsi-phy-28nm-hpm - qcom,dsi-phy-28nm-lp - qcom,hdmi-phy-8084 diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sc7180-dpu.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sc7180-dpu.yaml index 1fb8321d9ee80cdbf7e3ef4b52c518beb3b464cd..630b11480496f194ad7c190a809cb6e35ccdc4f6 100644 --- a/Documentation/devicetree/bindings/display/msm/qcom,sc7180-dpu.yaml +++ b/Documentation/devicetree/bindings/display/msm/qcom,sc7180-dpu.yaml @@ -13,7 +13,10 @@ $ref: /schemas/display/msm/dpu-common.yaml# properties: compatible: - const: qcom,sc7180-dpu + enum: + - qcom,sc7180-dpu + - qcom,sm6350-dpu + - qcom,sm6375-dpu reg: items: @@ -26,6 +29,7 @@ properties: - const: vbif clocks: + minItems: 6 items: - description: Display hf axi clock - description: Display ahb clock @@ -33,8 +37,10 @@ properties: - description: Display lut clock - description: Display core clock - description: Display vsync clock + - description: Display core throttle clock clock-names: + minItems: 6 items: - const: bus - const: iface @@ -42,6 +48,7 @@ properties: - const: lut - const: core - const: vsync + - const: throttle required: - compatible @@ -52,6 +59,20 @@ required: unevaluatedProperties: false +allOf: + - if: + properties: + compatible: + const: qcom,sm6375-dpu + + then: + properties: + clocks: + minItems: 7 + + clock-names: + minItems: 7 + examples: - | #include diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sm6350-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sm6350-mdss.yaml new file mode 100644 index 0000000000000000000000000000000000000000..ed0ad194d4ce9ea5d610d66cf3739bd2a7b29ab2 --- /dev/null +++ b/Documentation/devicetree/bindings/display/msm/qcom,sm6350-mdss.yaml @@ -0,0 +1,213 @@ +# SPDX-License-Identifier: GPL-2.0-only or BSD-2-Clause +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/display/msm/qcom,sm6350-mdss.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Qualcomm SM6350 Display MDSS + +maintainers: + - Krishna Manikandan + +description: + SM6350 MSM Mobile Display Subsystem (MDSS), which encapsulates sub-blocks + like DPU display controller, DSI and DP interfaces etc. + +$ref: /schemas/display/msm/mdss-common.yaml# + +properties: + compatible: + const: qcom,sm6350-mdss + + clocks: + items: + - description: Display AHB clock from gcc + - description: Display AXI clock from gcc + - description: Display core clock + + clock-names: + items: + - const: iface + - const: bus + - const: core + + iommus: + maxItems: 1 + + interconnects: + maxItems: 2 + + interconnect-names: + maxItems: 2 + +patternProperties: + "^display-controller@[0-9a-f]+$": + type: object + properties: + compatible: + const: qcom,sm6350-dpu + + "^dsi@[0-9a-f]+$": + type: object + properties: + compatible: + items: + - const: qcom,sm6350-dsi-ctrl + - const: qcom,mdss-dsi-ctrl + + "^phy@[0-9a-f]+$": + type: object + properties: + compatible: + const: qcom,dsi-phy-10nm + +unevaluatedProperties: false + +examples: + - | + #include + #include + #include + #include + #include + + display-subsystem@ae00000 { + compatible = "qcom,sm6350-mdss"; + reg = <0x0ae00000 0x1000>; + reg-names = "mdss"; + + power-domains = <&dispcc MDSS_GDSC>; + + clocks = <&gcc GCC_DISP_AHB_CLK>, + <&gcc GCC_DISP_AXI_CLK>, + <&dispcc DISP_CC_MDSS_MDP_CLK>; + clock-names = "iface", "bus", "core"; + + interrupts = ; + interrupt-controller; + #interrupt-cells = <1>; + + iommus = <&apps_smmu 0x800 0x2>; + #address-cells = <1>; + #size-cells = <1>; + ranges; + + display-controller@ae01000 { + compatible = "qcom,sm6350-dpu"; + reg = <0x0ae01000 0x8f000>, + <0x0aeb0000 0x2008>; + reg-names = "mdp", "vbif"; + + clocks = <&gcc GCC_DISP_AXI_CLK>, + <&dispcc DISP_CC_MDSS_AHB_CLK>, + <&dispcc DISP_CC_MDSS_ROT_CLK>, + <&dispcc DISP_CC_MDSS_MDP_LUT_CLK>, + <&dispcc DISP_CC_MDSS_MDP_CLK>, + <&dispcc DISP_CC_MDSS_VSYNC_CLK>; + clock-names = "bus", "iface", "rot", "lut", "core", + "vsync"; + + assigned-clocks = <&dispcc DISP_CC_MDSS_MDP_CLK>, + <&dispcc DISP_CC_MDSS_VSYNC_CLK>, + <&dispcc DISP_CC_MDSS_ROT_CLK>, + <&dispcc DISP_CC_MDSS_AHB_CLK>; + assigned-clock-rates = <300000000>, + <19200000>, + <19200000>, + <19200000>; + + interrupt-parent = <&mdss>; + interrupts = <0>; + operating-points-v2 = <&mdp_opp_table>; + power-domains = <&rpmhpd SM6350_CX>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + + port@0 { + reg = <0>; + dpu_intf1_out: endpoint { + remote-endpoint = <&dsi0_in>; + }; + }; + + port@1 { + reg = <1>; + dpu_intf2_out: endpoint { + remote-endpoint = <&dsi1_in>; + }; + }; + }; + }; + + dsi@ae94000 { + compatible = "qcom,sm6350-dsi-ctrl", "qcom,mdss-dsi-ctrl"; + reg = <0x0ae94000 0x400>; + reg-names = "dsi_ctrl"; + + interrupt-parent = <&mdss>; + interrupts = <4>; + + clocks = <&dispcc DISP_CC_MDSS_BYTE0_CLK>, + <&dispcc DISP_CC_MDSS_BYTE0_INTF_CLK>, + <&dispcc DISP_CC_MDSS_PCLK0_CLK>, + <&dispcc DISP_CC_MDSS_ESC0_CLK>, + <&dispcc DISP_CC_MDSS_AHB_CLK>, + <&gcc GCC_DISP_AXI_CLK>; + clock-names = "byte", + "byte_intf", + "pixel", + "core", + "iface", + "bus"; + + assigned-clocks = <&dispcc DISP_CC_MDSS_BYTE0_CLK_SRC>, + <&dispcc DISP_CC_MDSS_PCLK0_CLK_SRC>; + assigned-clock-parents = <&dsi0_phy 0>, <&dsi0_phy 1>; + + operating-points-v2 = <&dsi_opp_table>; + power-domains = <&rpmhpd SM6350_MX>; + + phys = <&dsi0_phy>; + phy-names = "dsi"; + + #address-cells = <1>; + #size-cells = <0>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + + port@0 { + reg = <0>; + dsi0_in: endpoint { + remote-endpoint = <&dpu_intf1_out>; + }; + }; + + port@1 { + reg = <1>; + dsi0_out: endpoint { + }; + }; + }; + }; + + dsi0_phy: phy@ae94400 { + compatible = "qcom,dsi-phy-10nm"; + reg = <0x0ae94400 0x200>, + <0x0ae94600 0x280>, + <0x0ae94a00 0x1e0>; + reg-names = "dsi_phy", + "dsi_phy_lane", + "dsi_pll"; + + #clock-cells = <1>; + #phy-cells = <0>; + + clocks = <&dispcc DISP_CC_MDSS_AHB_CLK>, <&rpmhcc RPMH_CXO_CLK>; + clock-names = "iface", "ref"; + }; + }; +... diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sm6375-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sm6375-mdss.yaml new file mode 100644 index 0000000000000000000000000000000000000000..76369a4f7c4d449daf73b84f545e05889083448f --- /dev/null +++ b/Documentation/devicetree/bindings/display/msm/qcom,sm6375-mdss.yaml @@ -0,0 +1,215 @@ +# SPDX-License-Identifier: GPL-2.0-only or BSD-2-Clause +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/display/msm/qcom,sm6375-mdss.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Qualcomm SM6375 Display MDSS + +maintainers: + - Konrad Dybcio + +description: + SM6375 MSM Mobile Display Subsystem (MDSS), which encapsulates sub-blocks + like DPU display controller, DSI and DP interfaces etc. + +$ref: /schemas/display/msm/mdss-common.yaml# + +properties: + compatible: + const: qcom,sm6375-mdss + + clocks: + items: + - description: Display AHB clock from gcc + - description: Display AHB clock + - description: Display core clock + + clock-names: + items: + - const: iface + - const: ahb + - const: core + + iommus: + maxItems: 1 + + interconnects: + maxItems: 2 + + interconnect-names: + maxItems: 2 + +patternProperties: + "^display-controller@[0-9a-f]+$": + type: object + properties: + compatible: + const: qcom,sm6375-dpu + + "^dsi@[0-9a-f]+$": + type: object + properties: + compatible: + items: + - const: qcom,sm6375-dsi-ctrl + - const: qcom,mdss-dsi-ctrl + + "^phy@[0-9a-f]+$": + type: object + properties: + compatible: + const: qcom,sm6375-dsi-phy-7nm + +unevaluatedProperties: false + +examples: + - | + #include + #include + #include + #include + #include + + display-subsystem@5e00000 { + compatible = "qcom,sm6375-mdss"; + reg = <0x05e00000 0x1000>; + reg-names = "mdss"; + + power-domains = <&dispcc MDSS_GDSC>; + + clocks = <&gcc GCC_DISP_AHB_CLK>, + <&dispcc DISP_CC_MDSS_AHB_CLK>, + <&dispcc DISP_CC_MDSS_MDP_CLK>; + clock-names = "iface", "ahb", "core"; + + interrupts = ; + interrupt-controller; + #interrupt-cells = <1>; + + iommus = <&apps_smmu 0x820 0x2>; + #address-cells = <1>; + #size-cells = <1>; + ranges; + + display-controller@5e01000 { + compatible = "qcom,sm6375-dpu"; + reg = <0x05e01000 0x8e030>, + <0x05eb0000 0x2008>; + reg-names = "mdp", "vbif"; + + clocks = <&gcc GCC_DISP_HF_AXI_CLK>, + <&dispcc DISP_CC_MDSS_AHB_CLK>, + <&dispcc DISP_CC_MDSS_ROT_CLK>, + <&dispcc DISP_CC_MDSS_MDP_LUT_CLK>, + <&dispcc DISP_CC_MDSS_MDP_CLK>, + <&dispcc DISP_CC_MDSS_VSYNC_CLK>, + <&gcc GCC_DISP_THROTTLE_CORE_CLK>; + clock-names = "bus", + "iface", + "rot", + "lut", + "core", + "vsync", + "throttle"; + + assigned-clocks = <&dispcc DISP_CC_MDSS_VSYNC_CLK>; + assigned-clock-rates = <19200000>; + + operating-points-v2 = <&mdp_opp_table>; + power-domains = <&rpmpd SM6375_VDDCX>; + + interrupt-parent = <&mdss>; + interrupts = <0>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + + port@0 { + reg = <0>; + dpu_intf1_out: endpoint { + remote-endpoint = <&dsi0_in>; + }; + }; + + port@1 { + reg = <1>; + dpu_intf2_out: endpoint { + remote-endpoint = <&dsi1_in>; + }; + }; + }; + }; + + dsi@5e94000 { + compatible = "qcom,sm6375-dsi-ctrl", "qcom,mdss-dsi-ctrl"; + reg = <0x05e94000 0x400>; + reg-names = "dsi_ctrl"; + + interrupt-parent = <&mdss>; + interrupts = <4>; + + clocks = <&dispcc DISP_CC_MDSS_BYTE0_CLK>, + <&dispcc DISP_CC_MDSS_BYTE0_INTF_CLK>, + <&dispcc DISP_CC_MDSS_PCLK0_CLK>, + <&dispcc DISP_CC_MDSS_ESC0_CLK>, + <&dispcc DISP_CC_MDSS_AHB_CLK>, + <&gcc GCC_DISP_HF_AXI_CLK>; + clock-names = "byte", + "byte_intf", + "pixel", + "core", + "iface", + "bus"; + + assigned-clocks = <&dispcc DISP_CC_MDSS_BYTE0_CLK_SRC>, + <&dispcc DISP_CC_MDSS_PCLK0_CLK_SRC>; + assigned-clock-parents = <&mdss_dsi0_phy 0>, <&mdss_dsi0_phy 1>; + + operating-points-v2 = <&dsi_opp_table>; + power-domains = <&rpmpd SM6375_VDDMX>; + + phys = <&mdss_dsi0_phy>; + phy-names = "dsi"; + + #address-cells = <1>; + #size-cells = <0>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + + port@0 { + reg = <0>; + dsi0_in: endpoint { + remote-endpoint = <&dpu_intf1_out>; + }; + }; + + port@1 { + reg = <1>; + dsi0_out: endpoint { + }; + }; + }; + }; + + mdss_dsi0_phy: phy@5e94400 { + compatible = "qcom,sm6375-dsi-phy-7nm"; + reg = <0x05e94400 0x200>, + <0x05e94600 0x280>, + <0x05e94900 0x264>; + reg-names = "dsi_phy", + "dsi_phy_lane", + "dsi_pll"; + + #clock-cells = <1>; + #phy-cells = <0>; + + clocks = <&dispcc DISP_CC_MDSS_AHB_CLK>, + <&rpmcc RPM_SMD_XO_CLK_SRC>; + clock-names = "iface", "ref"; + }; + }; +... diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sm8350-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sm8350-mdss.yaml index 4d94dbff305415d9c02e3b03aacea1efc3dba5b9..79a226e4cc6a2027f642765dabf9850a707f856f 100644 --- a/Documentation/devicetree/bindings/display/msm/qcom,sm8350-mdss.yaml +++ b/Documentation/devicetree/bindings/display/msm/qcom,sm8350-mdss.yaml @@ -64,7 +64,7 @@ patternProperties: type: object properties: compatible: - const: qcom,dsi-phy-5nm-8350 + const: qcom,sm8350-dsi-phy-5nm unevaluatedProperties: false diff --git a/Documentation/devicetree/bindings/fpga/lattice,sysconfig.yaml b/Documentation/devicetree/bindings/fpga/lattice,sysconfig.yaml index 4fb05eb84e2af4488d06fc46345e24fb3f958af0..164331eb627502dae2545ab636f1e7ede6e45a66 100644 --- a/Documentation/devicetree/bindings/fpga/lattice,sysconfig.yaml +++ b/Documentation/devicetree/bindings/fpga/lattice,sysconfig.yaml @@ -7,7 +7,7 @@ $schema: http://devicetree.org/meta-schemas/core.yaml# title: Lattice Slave SPI sysCONFIG FPGA manager maintainers: - - Ivan Bornyakov + - Vladimir Georgiev description: | Lattice sysCONFIG port, which is used for FPGA configuration, among others, diff --git a/Documentation/devicetree/bindings/fpga/microchip,mpf-spi-fpga-mgr.yaml b/Documentation/devicetree/bindings/fpga/microchip,mpf-spi-fpga-mgr.yaml index 527532f039cecad9b39248b74fa9839804da6345..a157eecfb5fc333ced32389a02f79281b3d4161d 100644 --- a/Documentation/devicetree/bindings/fpga/microchip,mpf-spi-fpga-mgr.yaml +++ b/Documentation/devicetree/bindings/fpga/microchip,mpf-spi-fpga-mgr.yaml @@ -7,7 +7,7 @@ $schema: http://devicetree.org/meta-schemas/core.yaml# title: Microchip Polarfire FPGA manager. maintainers: - - Ivan Bornyakov + - Vladimir Georgiev description: Device Tree Bindings for Microchip Polarfire FPGA Manager using slave SPI to diff --git a/Documentation/devicetree/bindings/i3c/silvaco,i3c-master.yaml b/Documentation/devicetree/bindings/i3c/silvaco,i3c-master.yaml index 62f3ca66274feb7ea39b74f847ab773c5e1e0711..32c821f97779b7ea95ae2e16ab68698eff8ac8d2 100644 --- a/Documentation/devicetree/bindings/i3c/silvaco,i3c-master.yaml +++ b/Documentation/devicetree/bindings/i3c/silvaco,i3c-master.yaml @@ -44,7 +44,7 @@ required: - clock-names - clocks -additionalProperties: true +unevaluatedProperties: false examples: - | diff --git a/Documentation/devicetree/bindings/iio/adc/nxp,imx8qxp-adc.yaml b/Documentation/devicetree/bindings/iio/adc/nxp,imx8qxp-adc.yaml index 63369ba388e4738cd4934884a37fbc444749458b..0a192ca192c5b3221843afacfe1b60c06425423e 100644 --- a/Documentation/devicetree/bindings/iio/adc/nxp,imx8qxp-adc.yaml +++ b/Documentation/devicetree/bindings/iio/adc/nxp,imx8qxp-adc.yaml @@ -39,6 +39,12 @@ properties: power-domains: maxItems: 1 + vref-supply: + description: | + External ADC reference voltage supply on VREFH pad. If VERID[MVI] is + set, there are additional, internal reference voltages selectable. + VREFH1 is always from VREFH pad. + "#io-channel-cells": const: 1 @@ -72,6 +78,7 @@ examples: assigned-clocks = <&clk IMX_SC_R_ADC_0>; assigned-clock-rates = <24000000>; power-domains = <&pd IMX_SC_R_ADC_0>; + vref-supply = <®_1v8>; #io-channel-cells = <1>; }; }; diff --git a/Documentation/devicetree/bindings/iio/adc/renesas,rcar-gyroadc.yaml b/Documentation/devicetree/bindings/iio/adc/renesas,rcar-gyroadc.yaml index 1c7aee5ed3e0bfb208f5438503f7cccc5065d2a7..36dff3250ea76fe51b6ca32ed850617b1c58b2f0 100644 --- a/Documentation/devicetree/bindings/iio/adc/renesas,rcar-gyroadc.yaml +++ b/Documentation/devicetree/bindings/iio/adc/renesas,rcar-gyroadc.yaml @@ -90,7 +90,7 @@ patternProperties: of the MAX chips to the GyroADC, while MISO line of each Maxim ADC connects to a shared input pin of the GyroADC. enum: - - adi,7476 + - adi,ad7476 - fujitsu,mb88101a - maxim,max1162 - maxim,max11100 diff --git a/Documentation/devicetree/bindings/interrupt-controller/arm,gic-v3.yaml b/Documentation/devicetree/bindings/interrupt-controller/arm,gic-v3.yaml index 92117261e1e1d0e985e47fe6fca5814a2e4293db..39e64c7f6360cacfe615f0b65a2ec8e85482c17d 100644 --- a/Documentation/devicetree/bindings/interrupt-controller/arm,gic-v3.yaml +++ b/Documentation/devicetree/bindings/interrupt-controller/arm,gic-v3.yaml @@ -166,6 +166,12 @@ properties: resets: maxItems: 1 + mediatek,broken-save-restore-fw: + type: boolean + description: + Asserts that the firmware on this device has issues saving and restoring + GICR registers when the GIC redistributors are powered off. + dependencies: mbi-ranges: [ msi-controller ] msi-controller: [ mbi-ranges ] diff --git a/Documentation/devicetree/bindings/mfd/canaan,k210-sysctl.yaml b/Documentation/devicetree/bindings/mfd/canaan,k210-sysctl.yaml index 8459d36422050a14c227e34ba10e0f02fe54cd8c..3b3beab9db3f31a597fdff2b2e3135a1d7947de4 100644 --- a/Documentation/devicetree/bindings/mfd/canaan,k210-sysctl.yaml +++ b/Documentation/devicetree/bindings/mfd/canaan,k210-sysctl.yaml @@ -7,7 +7,7 @@ $schema: http://devicetree.org/meta-schemas/core.yaml# title: Canaan Kendryte K210 System Controller maintainers: - - Damien Le Moal + - Damien Le Moal description: Canaan Inc. Kendryte K210 SoC system controller which provides a diff --git a/Documentation/devicetree/bindings/net/can/st,stm32-bxcan.yaml b/Documentation/devicetree/bindings/net/can/st,stm32-bxcan.yaml index 769fa5c27b76ee427a46daeef2055a516f29268f..de1d4298893be8b7f3c2090f338484d1bb9d6a5d 100644 --- a/Documentation/devicetree/bindings/net/can/st,stm32-bxcan.yaml +++ b/Documentation/devicetree/bindings/net/can/st,stm32-bxcan.yaml @@ -21,11 +21,22 @@ properties: st,can-primary: description: - Primary and secondary mode of the bxCAN peripheral is only relevant - if the chip has two CAN peripherals. In that case they share some - of the required logic. + Primary mode of the bxCAN peripheral is only relevant if the chip has + two CAN peripherals in dual CAN configuration. In that case they share + some of the required logic. + Not to be used if the peripheral is in single CAN configuration. To avoid misunderstandings, it should be noted that ST documentation - uses the terms master/slave instead of primary/secondary. + uses the terms master instead of primary. + type: boolean + + st,can-secondary: + description: + Secondary mode of the bxCAN peripheral is only relevant if the chip + has two CAN peripherals in dual CAN configuration. In that case they + share some of the required logic. + Not to be used if the peripheral is in single CAN configuration. + To avoid misunderstandings, it should be noted that ST documentation + uses the terms slave instead of secondary. type: boolean reg: diff --git a/Documentation/devicetree/bindings/net/realtek-bluetooth.yaml b/Documentation/devicetree/bindings/net/realtek-bluetooth.yaml index 8cc2b9924680aa6614eccba2523665ae1dc84a4d..043e118c605c834cf652ab50a3a6a0e6f954fec8 100644 --- a/Documentation/devicetree/bindings/net/realtek-bluetooth.yaml +++ b/Documentation/devicetree/bindings/net/realtek-bluetooth.yaml @@ -11,7 +11,7 @@ maintainers: - Alistair Francis description: - RTL8723CS/RTL8723CS/RTL8821CS/RTL8822CS is a WiFi + BT chip. WiFi part + RTL8723BS/RTL8723CS/RTL8821CS/RTL8822CS is a WiFi + BT chip. WiFi part is connected over SDIO, while BT is connected over serial. It speaks H5 protocol with few extra commands to upload firmware and change module speed. @@ -27,7 +27,7 @@ properties: - items: - enum: - realtek,rtl8821cs-bt - - const: realtek,rtl8822cs-bt + - const: realtek,rtl8723bs-bt device-wake-gpios: maxItems: 1 diff --git a/Documentation/devicetree/bindings/pinctrl/canaan,k210-fpioa.yaml b/Documentation/devicetree/bindings/pinctrl/canaan,k210-fpioa.yaml index 7f4f36a58e56cba672fdd4b12ca3e496c6e4fc2e..739a08f00467787419ff7575d2e4846e0db30b9b 100644 --- a/Documentation/devicetree/bindings/pinctrl/canaan,k210-fpioa.yaml +++ b/Documentation/devicetree/bindings/pinctrl/canaan,k210-fpioa.yaml @@ -7,7 +7,7 @@ $schema: http://devicetree.org/meta-schemas/core.yaml# title: Canaan Kendryte K210 FPIOA maintainers: - - Damien Le Moal + - Damien Le Moal description: The Canaan Kendryte K210 SoC Fully Programmable IO Array (FPIOA) diff --git a/Documentation/devicetree/bindings/pinctrl/qcom,pmic-mpp.yaml b/Documentation/devicetree/bindings/pinctrl/qcom,pmic-mpp.yaml index c91d3e3a094b8f319785e2c0aefcf8d1761c31db..80f960671857042ca4018b6a495822dbb3056c6e 100644 --- a/Documentation/devicetree/bindings/pinctrl/qcom,pmic-mpp.yaml +++ b/Documentation/devicetree/bindings/pinctrl/qcom,pmic-mpp.yaml @@ -144,8 +144,9 @@ $defs: enum: [0, 1, 2, 3, 4, 5, 6, 7] qcom,paired: - - description: - Indicates that the pin should be operating in paired mode. + type: boolean + description: + Indicates that the pin should be operating in paired mode. required: - pins diff --git a/Documentation/devicetree/bindings/power/qcom,rpmpd.yaml b/Documentation/devicetree/bindings/power/qcom,rpmpd.yaml index afad3135ed67cc8f57adc269d19d27d5082d1ed3..f9c211a9a938e3e254b16070f37f1cb8053a7358 100644 --- a/Documentation/devicetree/bindings/power/qcom,rpmpd.yaml +++ b/Documentation/devicetree/bindings/power/qcom,rpmpd.yaml @@ -29,6 +29,7 @@ properties: - qcom,qcm2290-rpmpd - qcom,qcs404-rpmpd - qcom,qdu1000-rpmhpd + - qcom,sa8155p-rpmhpd - qcom,sa8540p-rpmhpd - qcom,sa8775p-rpmhpd - qcom,sdm660-rpmpd diff --git a/Documentation/devicetree/bindings/reset/canaan,k210-rst.yaml b/Documentation/devicetree/bindings/reset/canaan,k210-rst.yaml index ee8a2dcf5dfae9dea839339d8a75e8a68ac8188a..0c0135964b91d17c396e2ede223e066b843698ba 100644 --- a/Documentation/devicetree/bindings/reset/canaan,k210-rst.yaml +++ b/Documentation/devicetree/bindings/reset/canaan,k210-rst.yaml @@ -7,7 +7,7 @@ $schema: http://devicetree.org/meta-schemas/core.yaml# title: Canaan Kendryte K210 Reset Controller maintainers: - - Damien Le Moal + - Damien Le Moal description: | Canaan Kendryte K210 reset controller driver which supports the SoC diff --git a/Documentation/devicetree/bindings/riscv/canaan.yaml b/Documentation/devicetree/bindings/riscv/canaan.yaml index f8f3f286bd55924d8e1ce5b8f5f38bf4a461a56a..41fd11f70a49b26dd91b3dfd0e9cb13a069c120d 100644 --- a/Documentation/devicetree/bindings/riscv/canaan.yaml +++ b/Documentation/devicetree/bindings/riscv/canaan.yaml @@ -7,7 +7,7 @@ $schema: http://devicetree.org/meta-schemas/core.yaml# title: Canaan SoC-based boards maintainers: - - Damien Le Moal + - Damien Le Moal description: Canaan Kendryte K210 SoC-based boards diff --git a/Documentation/devicetree/bindings/serial/8250_omap.yaml b/Documentation/devicetree/bindings/serial/8250_omap.yaml index eb3488d8f9ee61854288c4b43a221fe03e5d7f6e..6a7be42da523c60e8b5d9c5ae8e3c6f2b1bb22db 100644 --- a/Documentation/devicetree/bindings/serial/8250_omap.yaml +++ b/Documentation/devicetree/bindings/serial/8250_omap.yaml @@ -70,6 +70,7 @@ properties: dsr-gpios: true rng-gpios: true dcd-gpios: true + rs485-rts-active-high: true rts-gpio: true power-domains: true clock-frequency: true diff --git a/Documentation/devicetree/bindings/sound/tas2562.yaml b/Documentation/devicetree/bindings/sound/tas2562.yaml index a5bb561bfcfbafdcd8f0d0c17be3d3ec1ba9c7bd..31a3024ea7898eba853ef98b300b97c1335a1a24 100644 --- a/Documentation/devicetree/bindings/sound/tas2562.yaml +++ b/Documentation/devicetree/bindings/sound/tas2562.yaml @@ -55,7 +55,9 @@ properties: description: TDM TX current sense time slot. '#sound-dai-cells': - const: 1 + # The codec has a single DAI, the #sound-dai-cells=<1>; case is left in for backward + # compatibility but is deprecated. + enum: [0, 1] required: - compatible @@ -72,7 +74,7 @@ examples: codec: codec@4c { compatible = "ti,tas2562"; reg = <0x4c>; - #sound-dai-cells = <1>; + #sound-dai-cells = <0>; interrupt-parent = <&gpio1>; interrupts = <14>; shutdown-gpios = <&gpio1 15 0>; diff --git a/Documentation/devicetree/bindings/sound/tas2770.yaml b/Documentation/devicetree/bindings/sound/tas2770.yaml index 26088adb9dc24073828412171f37b634d1fc5e44..8908bf1122e965fe8eaab5c2244ac8d5650407eb 100644 --- a/Documentation/devicetree/bindings/sound/tas2770.yaml +++ b/Documentation/devicetree/bindings/sound/tas2770.yaml @@ -57,7 +57,9 @@ properties: - 1 # Falling edge '#sound-dai-cells': - const: 1 + # The codec has a single DAI, the #sound-dai-cells=<1>; case is left in for backward + # compatibility but is deprecated. + enum: [0, 1] required: - compatible @@ -74,7 +76,7 @@ examples: codec: codec@41 { compatible = "ti,tas2770"; reg = <0x41>; - #sound-dai-cells = <1>; + #sound-dai-cells = <0>; interrupt-parent = <&gpio1>; interrupts = <14>; reset-gpio = <&gpio1 15 0>; diff --git a/Documentation/devicetree/bindings/sound/tas27xx.yaml b/Documentation/devicetree/bindings/sound/tas27xx.yaml index 8cba01316855d4ac1bf9b87ed3a975d2c13ac87a..a876545ec87d8c6bc766ee7513fb231d52fc848d 100644 --- a/Documentation/devicetree/bindings/sound/tas27xx.yaml +++ b/Documentation/devicetree/bindings/sound/tas27xx.yaml @@ -50,7 +50,9 @@ properties: description: TDM TX voltage sense time slot. '#sound-dai-cells': - const: 1 + # The codec has a single DAI, the #sound-dai-cells=<1>; case is left in for backward + # compatibility but is deprecated. + enum: [0, 1] required: - compatible @@ -67,7 +69,7 @@ examples: codec: codec@38 { compatible = "ti,tas2764"; reg = <0x38>; - #sound-dai-cells = <1>; + #sound-dai-cells = <0>; interrupt-parent = <&gpio1>; interrupts = <14>; reset-gpios = <&gpio1 15 0>; diff --git a/Documentation/devicetree/bindings/sound/tlv320aic32x4.txt b/Documentation/devicetree/bindings/sound/tlv320aic32x4.txt index f59125bc79d1bfb96cc91cee6f08d1b6785ba891..0b4e21bde5bc1c27fd39835517beb92004c743d0 100644 --- a/Documentation/devicetree/bindings/sound/tlv320aic32x4.txt +++ b/Documentation/devicetree/bindings/sound/tlv320aic32x4.txt @@ -8,7 +8,7 @@ Required properties: "ti,tlv320aic32x6" TLV320AIC3206, TLV320AIC3256 "ti,tas2505" TAS2505, TAS2521 - reg: I2C slave address - - supply-*: Required supply regulators are: + - *-supply: Required supply regulators are: "iov" - digital IO power supply "ldoin" - LDO power supply "dv" - Digital core power supply diff --git a/Documentation/devicetree/bindings/usb/cdns,usb3.yaml b/Documentation/devicetree/bindings/usb/cdns,usb3.yaml index cae46c4982adf69e3369fa450924df14e03fa804..69a93a0722f07f5c45ec6d5f817eed79231463ab 100644 --- a/Documentation/devicetree/bindings/usb/cdns,usb3.yaml +++ b/Documentation/devicetree/bindings/usb/cdns,usb3.yaml @@ -64,7 +64,7 @@ properties: description: size of memory intended as internal memory for endpoints buffers expressed in KB - $ref: /schemas/types.yaml#/definitions/uint32 + $ref: /schemas/types.yaml#/definitions/uint16 cdns,phyrst-a-enable: description: Enable resetting of PHY if Rx fail is detected diff --git a/Documentation/devicetree/bindings/usb/snps,dwc3.yaml b/Documentation/devicetree/bindings/usb/snps,dwc3.yaml index 50edc4da780e9caf68f7821bcff9c06ffa80bba2..4f7625955cccccff6964c98cfed9a1dc1f225326 100644 --- a/Documentation/devicetree/bindings/usb/snps,dwc3.yaml +++ b/Documentation/devicetree/bindings/usb/snps,dwc3.yaml @@ -287,7 +287,7 @@ properties: description: High-Speed PHY interface selection between UTMI+ and ULPI when the DWC_USB3_HSPHY_INTERFACE has value 3. - $ref: /schemas/types.yaml#/definitions/uint8 + $ref: /schemas/types.yaml#/definitions/string enum: [utmi, ulpi] snps,quirk-frame-length-adjustment: diff --git a/Documentation/devicetree/usage-model.rst b/Documentation/devicetree/usage-model.rst index b6a287955ee56ac14b9902c524e42cf496ad7a89..0717426856b229be1c52efc79b18dbc617d4c92f 100644 --- a/Documentation/devicetree/usage-model.rst +++ b/Documentation/devicetree/usage-model.rst @@ -415,6 +415,6 @@ When using the DT, this creates problems for of_platform_populate() because it must decide whether to register each node as either a platform_device or an amba_device. This unfortunately complicates the device creation model a little bit, but the solution turns out not to -be too invasive. If a node is compatible with "arm,amba-primecell", then +be too invasive. If a node is compatible with "arm,primecell", then of_platform_populate() will register it as an amba_device instead of a platform_device. diff --git a/Documentation/filesystems/index.rst b/Documentation/filesystems/index.rst index fbb2b5ada95b38838d8ae6ab97e476c28d3be341..eb252fc972aa95e8d731981a8ea3340b0da0160a 100644 --- a/Documentation/filesystems/index.rst +++ b/Documentation/filesystems/index.rst @@ -72,7 +72,6 @@ Documentation for filesystem implementations. befs bfs btrfs - cifs/index ceph coda configfs @@ -111,6 +110,7 @@ Documentation for filesystem implementations. ramfs-rootfs-initramfs relay romfs + smb/index spufs/index squashfs sysfs diff --git a/Documentation/filesystems/ramfs-rootfs-initramfs.rst b/Documentation/filesystems/ramfs-rootfs-initramfs.rst index 164960631925d05a4684fc5920e63cf204f57663..447f767c646276551570baa9de4dd19ccda95c61 100644 --- a/Documentation/filesystems/ramfs-rootfs-initramfs.rst +++ b/Documentation/filesystems/ramfs-rootfs-initramfs.rst @@ -6,8 +6,7 @@ Ramfs, rootfs and initramfs October 17, 2005 -Rob Landley -============================= +:Author: Rob Landley What is ramfs? -------------- diff --git a/Documentation/filesystems/sharedsubtree.rst b/Documentation/filesystems/sharedsubtree.rst index d83395354250d9ca02dc19a38cdf9762733ded04..1cf56489ed484dcf77581e6be1faa490e600a3ab 100644 --- a/Documentation/filesystems/sharedsubtree.rst +++ b/Documentation/filesystems/sharedsubtree.rst @@ -147,6 +147,7 @@ replicas continue to be exactly same. 3) Setting mount states +----------------------- The mount command (util-linux package) can be used to set mount states:: @@ -612,6 +613,7 @@ replicas continue to be exactly same. 6) Quiz +------- A. What is the result of the following command sequence? @@ -673,6 +675,7 @@ replicas continue to be exactly same. /mnt/1/test be? 7) FAQ +------ Q1. Why is bind mount needed? How is it different from symbolic links? symbolic links can get stale if the destination mount gets @@ -841,6 +844,7 @@ replicas continue to be exactly same. tmp usr tmp usr tmp usr 8) Implementation +----------------- 8A) Datastructure diff --git a/Documentation/filesystems/cifs/cifsroot.rst b/Documentation/filesystems/smb/cifsroot.rst similarity index 97% rename from Documentation/filesystems/cifs/cifsroot.rst rename to Documentation/filesystems/smb/cifsroot.rst index 4930bb443134aecdc4445b8058d8272bb11aa696..bf2d9db3acb924bb8e86c3ddc0bfdffbc098ed6e 100644 --- a/Documentation/filesystems/cifs/cifsroot.rst +++ b/Documentation/filesystems/smb/cifsroot.rst @@ -59,7 +59,7 @@ the root file system via SMB protocol. Enables the kernel to mount the root file system via SMB that are located in the and specified in this option. -The default mount options are set in fs/cifs/cifsroot.c. +The default mount options are set in fs/smb/client/cifsroot.c. server-ip IPv4 address of the server. diff --git a/Documentation/filesystems/cifs/index.rst b/Documentation/filesystems/smb/index.rst similarity index 100% rename from Documentation/filesystems/cifs/index.rst rename to Documentation/filesystems/smb/index.rst diff --git a/Documentation/filesystems/cifs/ksmbd.rst b/Documentation/filesystems/smb/ksmbd.rst similarity index 100% rename from Documentation/filesystems/cifs/ksmbd.rst rename to Documentation/filesystems/smb/ksmbd.rst diff --git a/Documentation/fpga/index.rst b/Documentation/fpga/index.rst index f80f95667ca217a0c87f68ad0a06cfcb1c2db87d..43c968871d995ad4e2ece499a15837a352e8e0ee 100644 --- a/Documentation/fpga/index.rst +++ b/Documentation/fpga/index.rst @@ -1,7 +1,7 @@ .. SPDX-License-Identifier: GPL-2.0 ==== -fpga +FPGA ==== .. toctree:: diff --git a/Documentation/gpu/amdgpu/apu-asic-info-table.csv b/Documentation/gpu/amdgpu/apu-asic-info-table.csv index 395a7b7bfaefb92ea106a378ae7f542b42cdac99..2e76b427ba1ee5f4806a3487ff929c4f252b4695 100644 --- a/Documentation/gpu/amdgpu/apu-asic-info-table.csv +++ b/Documentation/gpu/amdgpu/apu-asic-info-table.csv @@ -5,6 +5,8 @@ Ryzen 4000 series, RENOIR, DCN 2.1, 9.3, VCN 2.2, 4.1.2, 11.0.3 Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx, RAVEN2, DCN 1.0, 9.2.2, VCN 1.0.1, 4.1.1, 10.0.1 SteamDeck, VANGOGH, DCN 3.0.1, 10.3.1, VCN 3.1.0, 5.2.1, 11.5.0 Ryzen 5000 series / Ryzen 7x30 series, GREEN SARDINE / Cezanne / Barcelo / Barcelo-R, DCN 2.1, 9.3, VCN 2.2, 4.1.1, 12.0.1 -Ryzen 6000 series / Ryzen 7x35 series, YELLOW CARP / Rembrandt / Rembrandt+, 3.1.2, 10.3.3, VCN 3.1.1, 5.2.3, 13.0.3 +Ryzen 6000 series / Ryzen 7x35 series / Ryzen 7x36 series, YELLOW CARP / Rembrandt / Rembrandt-R, 3.1.2, 10.3.3, VCN 3.1.1, 5.2.3, 13.0.3 Ryzen 7000 series (AM5), Raphael, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5 +Ryzen 7x45 series (FL1), / Dragon Range, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5 Ryzen 7x20 series, Mendocino, 3.1.6, 10.3.7, 3.1.1, 5.2.7, 13.0.8 +Ryzen 7x40 series, Phoenix, 3.1.4, 11.0.1 / 11.0.4, 4.0.2, 6.0.1, 13.0.4 / 13.0.11 \ No newline at end of file diff --git a/Documentation/locking/index.rst b/Documentation/locking/index.rst index 7003bd5aeff4cf711f9fde4b9fdde5c120a79559..6a9ea96c8bcb7013ac97715550d6845bfb188593 100644 --- a/Documentation/locking/index.rst +++ b/Documentation/locking/index.rst @@ -1,7 +1,7 @@ .. SPDX-License-Identifier: GPL-2.0 ======= -locking +Locking ======= .. toctree:: diff --git a/Documentation/mm/page_table_check.rst b/Documentation/mm/page_table_check.rst index cfd8f4117cf3e9a1ef4800653b19d3711fe069d7..c12838ce6b8de2468c81b2d5d18b2fa74317e271 100644 --- a/Documentation/mm/page_table_check.rst +++ b/Documentation/mm/page_table_check.rst @@ -52,3 +52,22 @@ Build kernel with: Optionally, build kernel with PAGE_TABLE_CHECK_ENFORCED in order to have page table support without extra kernel parameter. + +Implementation notes +==================== + +We specifically decided not to use VMA information in order to avoid relying on +MM states (except for limited "struct page" info). The page table check is a +separate from Linux-MM state machine that verifies that the user accessible +pages are not falsely shared. + +PAGE_TABLE_CHECK depends on EXCLUSIVE_SYSTEM_RAM. The reason is that without +EXCLUSIVE_SYSTEM_RAM, users are allowed to map arbitrary physical memory +regions into the userspace via /dev/mem. At the same time, pages may change +their properties (e.g., from anonymous pages to named pages) while they are +still being mapped in the userspace, leading to "corruption" detected by the +page table check. + +Even with EXCLUSIVE_SYSTEM_RAM, I/O pages may be still allowed to be mapped via +/dev/mem. However, these pages are always considered as named pages, so they +won't break the logic used in the page table check. diff --git a/Documentation/netlink/specs/ethtool.yaml b/Documentation/netlink/specs/ethtool.yaml index 129f413ea3496fbd0bde88ce9bf786931d2a23ba..4846345bade4604d268a5ce5186378f4312d3dac 100644 --- a/Documentation/netlink/specs/ethtool.yaml +++ b/Documentation/netlink/specs/ethtool.yaml @@ -60,22 +60,6 @@ attribute-sets: type: nest nested-attributes: bitset-bits - - - name: u64-array - attributes: - - - name: u64 - type: nest - multi-attr: true - nested-attributes: u64 - - - name: s32-array - attributes: - - - name: s32 - type: nest - multi-attr: true - nested-attributes: s32 - name: string attributes: @@ -239,7 +223,7 @@ attribute-sets: name: tx-min-frag-size type: u32 - - name: tx-min-frag-size + name: rx-min-frag-size type: u32 - name: verify-enabled @@ -310,7 +294,7 @@ attribute-sets: name: master-slave-state type: u8 - - name: master-slave-lanes + name: lanes type: u32 - name: rate-matching @@ -338,7 +322,7 @@ attribute-sets: name: ext-substate type: u8 - - name: down-cnt + name: ext-down-cnt type: u32 - name: debug @@ -593,7 +577,7 @@ attribute-sets: name: phc-index type: u32 - - name: cable-test-nft-nest-result + name: cable-test-ntf-nest-result attributes: - name: pair @@ -602,7 +586,7 @@ attribute-sets: name: code type: u8 - - name: cable-test-nft-nest-fault-length + name: cable-test-ntf-nest-fault-length attributes: - name: pair @@ -611,16 +595,16 @@ attribute-sets: name: cm type: u32 - - name: cable-test-nft-nest + name: cable-test-ntf-nest attributes: - name: result type: nest - nested-attributes: cable-test-nft-nest-result + nested-attributes: cable-test-ntf-nest-result - name: fault-length type: nest - nested-attributes: cable-test-nft-nest-fault-length + nested-attributes: cable-test-ntf-nest-fault-length - name: cable-test attributes: @@ -634,7 +618,7 @@ attribute-sets: - name: nest type: nest - nested-attributes: cable-test-nft-nest + nested-attributes: cable-test-ntf-nest - name: cable-test-tdr-cfg attributes: @@ -705,16 +689,16 @@ attribute-sets: type: u8 - name: corrected - type: nest - nested-attributes: u64-array + type: binary + sub-type: u64 - name: uncorr - type: nest - nested-attributes: u64-array + type: binary + sub-type: u64 - name: corr-bits - type: nest - nested-attributes: u64-array + type: binary + sub-type: u64 - name: fec attributes: @@ -792,7 +776,7 @@ attribute-sets: name: hist-bkt-hi type: u32 - - name: hist-bkt-val + name: hist-val type: u64 - name: stats @@ -827,8 +811,8 @@ attribute-sets: type: u32 - name: index - type: nest - nested-attributes: s32-array + type: binary + sub-type: s32 - name: module attributes: @@ -981,7 +965,7 @@ operations: - duplex - master-slave-cfg - master-slave-state - - master-slave-lanes + - lanes - rate-matching dump: *linkmodes-get-op - @@ -1015,7 +999,7 @@ operations: - sqi-max - ext-state - ext-substate - - down-cnt + - ext-down-cnt dump: *linkstate-get-op - name: debug-get @@ -1367,7 +1351,7 @@ operations: reply: attributes: - header - - cable-test-nft-nest + - cable-test-ntf-nest - name: cable-test-tdr-act doc: Cable test TDR. @@ -1555,7 +1539,7 @@ operations: - hkey dump: *rss-get-op - - name: plca-get + name: plca-get-cfg doc: Get PLCA params. attribute-set: plca @@ -1577,7 +1561,7 @@ operations: - burst-tmr dump: *plca-get-op - - name: plca-set + name: plca-set-cfg doc: Set PLCA params. attribute-set: plca @@ -1601,7 +1585,7 @@ operations: - name: plca-ntf doc: Notification for change in PLCA params. - notify: plca-get + notify: plca-get-cfg - name: mm-get doc: Get MAC Merge configuration and state diff --git a/Documentation/netlink/specs/handshake.yaml b/Documentation/netlink/specs/handshake.yaml index 614f1a585511487a85339de90e2b675a800e5439..6d89e30f5fd52b1bfe7c1f36f9b5a859ccccfc2b 100644 --- a/Documentation/netlink/specs/handshake.yaml +++ b/Documentation/netlink/specs/handshake.yaml @@ -68,6 +68,9 @@ attribute-sets: type: nest nested-attributes: x509 multi-attr: true + - + name: peername + type: string - name: done attributes: @@ -105,6 +108,7 @@ operations: - auth-mode - peer-identity - certificate + - peername - name: done doc: Handler reports handshake completion diff --git a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/devlink.rst b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/devlink.rst index 3a7a714cc08f0a74dbc6de9dfd90f83a566b539e..3354ca3608ee671adc06939d78302420099b3110 100644 --- a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/devlink.rst +++ b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/devlink.rst @@ -40,6 +40,7 @@ flow_steering_mode: Device flow steering mode --------------------------------------------- The flow steering mode parameter controls the flow steering mode of the driver. Two modes are supported: + 1. 'dmfs' - Device managed flow steering. 2. 'smfs' - Software/Driver managed flow steering. @@ -99,6 +100,7 @@ between representors and stacked devices. By default metadata is enabled on the supported devices in E-switch. Metadata is applicable only for E-switch in switchdev mode and users may disable it when NONE of the below use cases will be in use: + 1. HCA is in Dual/multi-port RoCE mode. 2. VF/SF representor bonding (Usually used for Live migration) 3. Stacked devices @@ -180,7 +182,8 @@ User commands examples: $ devlink health diagnose pci/0000:82:00.0 reporter tx -NOTE: This command has valid output only when interface is up, otherwise the command has empty output. +.. note:: + This command has valid output only when interface is up, otherwise the command has empty output. - Show number of tx errors indicated, number of recover flows ended successfully, is autorecover enabled and graceful period from last recover:: @@ -232,8 +235,9 @@ User commands examples: $ devlink health dump show pci/0000:82:00.0 reporter fw -NOTE: This command can run only on the PF which has fw tracer ownership, -running it on other PF or any VF will return "Operation not permitted". +.. note:: + This command can run only on the PF which has fw tracer ownership, + running it on other PF or any VF will return "Operation not permitted". fw fatal reporter ----------------- @@ -256,7 +260,8 @@ User commands examples: $ devlink health dump show pci/0000:82:00.1 reporter fw_fatal -NOTE: This command can run only on PF. +.. note:: + This command can run only on PF. vnic reporter ------------- @@ -265,28 +270,37 @@ It is responsible for querying the vnic diagnostic counters from fw and displayi them in realtime. Description of the vnic counters: -total_q_under_processor_handle: number of queues in an error state due to -an async error or errored command. -send_queue_priority_update_flow: number of QP/SQ priority/SL update -events. -cq_overrun: number of times CQ entered an error state due to an -overflow. -async_eq_overrun: number of times an EQ mapped to async events was -overrun. -comp_eq_overrun: number of times an EQ mapped to completion events was -overrun. -quota_exceeded_command: number of commands issued and failed due to quota -exceeded. -invalid_command: number of commands issued and failed dues to any reason -other than quota exceeded. -nic_receive_steering_discard: number of packets that completed RX flow -steering but were discarded due to a mismatch in flow table. + +- total_q_under_processor_handle + number of queues in an error state due to + an async error or errored command. +- send_queue_priority_update_flow + number of QP/SQ priority/SL update events. +- cq_overrun + number of times CQ entered an error state due to an overflow. +- async_eq_overrun + number of times an EQ mapped to async events was overrun. + comp_eq_overrun number of times an EQ mapped to completion events was + overrun. +- quota_exceeded_command + number of commands issued and failed due to quota exceeded. +- invalid_command + number of commands issued and failed dues to any reason other than quota + exceeded. +- nic_receive_steering_discard + number of packets that completed RX flow + steering but were discarded due to a mismatch in flow table. User commands examples: -- Diagnose PF/VF vnic counters + +- Diagnose PF/VF vnic counters:: + $ devlink health diagnose pci/0000:82:00.1 reporter vnic + - Diagnose representor vnic counters (performed by supplying devlink port of the - representor, which can be obtained via devlink port command) + representor, which can be obtained via devlink port command):: + $ devlink health diagnose pci/0000:82:00.1/65537 reporter vnic -NOTE: This command can run over all interfaces such as PF/VF and representor ports. +.. note:: + This command can run over all interfaces such as PF/VF and representor ports. diff --git a/Documentation/networking/ip-sysctl.rst b/Documentation/networking/ip-sysctl.rst index 6ec06a33688a68c40dd882efedcdbe4480a1c5ec..80b8f73a0244f4682967e74bb2edec010bf0c27a 100644 --- a/Documentation/networking/ip-sysctl.rst +++ b/Documentation/networking/ip-sysctl.rst @@ -1352,8 +1352,8 @@ ping_group_range - 2 INTEGERS Restrict ICMP_PROTO datagram sockets to users in the group range. The default is "1 0", meaning, that nobody (not even root) may create ping sockets. Setting it to "100 100" would grant permissions - to the single group. "0 4294967295" would enable it for the world, "100 - 4294967295" would enable it for the users, but not daemons. + to the single group. "0 4294967294" would enable it for the world, "100 + 4294967294" would enable it for the users, but not daemons. tcp_early_demux - BOOLEAN Enable early demux for established TCP sockets. diff --git a/Documentation/networking/tls-handshake.rst b/Documentation/networking/tls-handshake.rst index a2817a88e905fa167244cebaf376684f3800faf2..6f5ea1646a4704653a41caee551ed85fdda12f0f 100644 --- a/Documentation/networking/tls-handshake.rst +++ b/Documentation/networking/tls-handshake.rst @@ -53,6 +53,7 @@ fills in a structure that contains the parameters of the request: struct socket *ta_sock; tls_done_func_t ta_done; void *ta_data; + const char *ta_peername; unsigned int ta_timeout_ms; key_serial_t ta_keyring; key_serial_t ta_my_cert; @@ -71,6 +72,10 @@ instantiated a struct file in sock->file. has completed. Further explanation of this function is in the "Handshake Completion" sesction below. +The consumer can provide a NUL-terminated hostname in the @ta_peername +field that is sent as part of ClientHello. If no peername is provided, +the DNS hostname associated with the server's IP address is used instead. + The consumer can fill in the @ta_timeout_ms field to force the servicing handshake agent to exit after a number of milliseconds. This enables the socket to be fully closed once both the kernel and the handshake agent diff --git a/Documentation/pcmcia/index.rst b/Documentation/pcmcia/index.rst index 7ae1f62fca14f8c5d06bd7d461d54853114063e3..8067236c51ab4208af1789eb907c479891685346 100644 --- a/Documentation/pcmcia/index.rst +++ b/Documentation/pcmcia/index.rst @@ -1,7 +1,7 @@ .. SPDX-License-Identifier: GPL-2.0 ====== -pcmcia +PCMCIA ====== .. toctree:: diff --git a/Documentation/process/maintainer-netdev.rst b/Documentation/process/maintainer-netdev.rst index f73ac9e175a8dd9e94ceeebbf5257018397bcbc7..83614cec9328b62b2333947a5254b1f57519545f 100644 --- a/Documentation/process/maintainer-netdev.rst +++ b/Documentation/process/maintainer-netdev.rst @@ -127,13 +127,32 @@ the value of ``Message-ID`` to the URL above. Updating patch status ~~~~~~~~~~~~~~~~~~~~~ -It may be tempting to help the maintainers and update the state of your -own patches when you post a new version or spot a bug. Please **do not** -do that. -Interfering with the patch status on patchwork will only cause confusion. Leave -it to the maintainer to figure out what is the most recent and current -version that should be applied. If there is any doubt, the maintainer -will reply and ask what should be done. +Contributors and reviewers do not have the permissions to update patch +state directly in patchwork. Patchwork doesn't expose much information +about the history of the state of patches, therefore having multiple +people update the state leads to confusion. + +Instead of delegating patchwork permissions netdev uses a simple mail +bot which looks for special commands/lines within the emails sent to +the mailing list. For example to mark a series as Changes Requested +one needs to send the following line anywhere in the email thread:: + + pw-bot: changes-requested + +As a result the bot will set the entire series to Changes Requested. +This may be useful when author discovers a bug in their own series +and wants to prevent it from getting applied. + +The use of the bot is entirely optional, if in doubt ignore its existence +completely. Maintainers will classify and update the state of the patches +themselves. No email should ever be sent to the list with the main purpose +of communicating with the bot, the bot commands should be seen as metadata. + +The use of the bot is restricted to authors of the patches (the ``From:`` +header on patch submission and command must match!), maintainers themselves +and a handful of senior reviewers. Bot records its activity here: + + https://patchwork.hopto.org/pw-bot.html Review timelines ~~~~~~~~~~~~~~~~ diff --git a/Documentation/riscv/patch-acceptance.rst b/Documentation/riscv/patch-acceptance.rst index 07d5a5623e2a3e87e5244b94b789c8174a983efe..634aa222b410c100945b9e954b333ce5b4871e57 100644 --- a/Documentation/riscv/patch-acceptance.rst +++ b/Documentation/riscv/patch-acceptance.rst @@ -16,6 +16,24 @@ tested code over experimental code. We wish to extend these same principles to the RISC-V-related code that will be accepted for inclusion in the kernel. +Patchwork +--------- + +RISC-V has a patchwork instance, where the status of patches can be checked: + + https://patchwork.kernel.org/project/linux-riscv/list/ + +If your patch does not appear in the default view, the RISC-V maintainers have +likely either requested changes, or expect it to be applied to another tree. + +Automation runs against this patchwork instance, building/testing patches as +they arrive. The automation applies patches against the current HEAD of the +RISC-V `for-next` and `fixes` branches, depending on whether the patch has been +detected as a fix. Failing those, it will use the RISC-V `master` branch. +The exact commit to which a series has been applied will be noted on patchwork. +Patches for which any of the checks fail are unlikely to be applied and in most +cases will need to be resubmitted. + Submit Checklist Addendum ------------------------- We'll only accept patches for new modules or extensions if the diff --git a/Documentation/s390/vfio-ap.rst b/Documentation/s390/vfio-ap.rst index d46e98c7c1ec6cabe1316c7122cea4a33933927f..bb3f4c4e288563dc1ca850b8f7f86b7b7347a56e 100644 --- a/Documentation/s390/vfio-ap.rst +++ b/Documentation/s390/vfio-ap.rst @@ -551,7 +551,6 @@ These are the steps: * IOMMU_SUPPORT * S390 * ZCRYPT - * S390_AP_IOMMU * VFIO * KVM diff --git a/Documentation/staging/crc32.rst b/Documentation/staging/crc32.rst index 8a6860f33b4e7622003687a018816bc06469d5f9..7542220967cb4c3366665ee7a7730b62ae44bcdb 100644 --- a/Documentation/staging/crc32.rst +++ b/Documentation/staging/crc32.rst @@ -1,5 +1,5 @@ ================================= -brief tutorial on CRC computation +Brief tutorial on CRC computation ================================= A CRC is a long-division remainder. You add the CRC to the message, diff --git a/Documentation/timers/index.rst b/Documentation/timers/index.rst index df510ad0c989bef451443dd503cd70947c378d35..983f91f8f023e073ee0513708126cdede8c745d6 100644 --- a/Documentation/timers/index.rst +++ b/Documentation/timers/index.rst @@ -1,7 +1,7 @@ .. SPDX-License-Identifier: GPL-2.0 ====== -timers +Timers ====== .. toctree:: diff --git a/Documentation/trace/histogram.rst b/Documentation/trace/histogram.rst index 479c9eac63352207d51859a65cc229f36022d361..3c9b263de9c24f6b3367bc1a8ae7cd80f06e993c 100644 --- a/Documentation/trace/histogram.rst +++ b/Documentation/trace/histogram.rst @@ -35,7 +35,7 @@ Documentation written by Tom Zanussi in place of an explicit value field - this is simply a count of event hits. If 'values' isn't specified, an implicit 'hitcount' value will be automatically created and used as the only value. - Keys can be any field, or the special string 'stacktrace', which + Keys can be any field, or the special string 'common_stacktrace', which will use the event's kernel stacktrace as the key. The keywords 'keys' or 'key' can be used to specify keys, and the keywords 'values', 'vals', or 'val' can be used to specify values. Compound @@ -54,7 +54,7 @@ Documentation written by Tom Zanussi 'compatible' if the fields named in the trigger share the same number and type of fields and those fields also have the same names. Note that any two events always share the compatible 'hitcount' and - 'stacktrace' fields and can therefore be combined using those + 'common_stacktrace' fields and can therefore be combined using those fields, however pointless that may be. 'hist' triggers add a 'hist' file to each event's subdirectory. @@ -547,9 +547,9 @@ Extended error information the hist trigger display symbolic call_sites, we can have the hist trigger additionally display the complete set of kernel stack traces that led to each call_site. To do that, we simply use the special - value 'stacktrace' for the key parameter:: + value 'common_stacktrace' for the key parameter:: - # echo 'hist:keys=stacktrace:values=bytes_req,bytes_alloc:sort=bytes_alloc' > \ + # echo 'hist:keys=common_stacktrace:values=bytes_req,bytes_alloc:sort=bytes_alloc' > \ /sys/kernel/tracing/events/kmem/kmalloc/trigger The above trigger will use the kernel stack trace in effect when an @@ -561,9 +561,9 @@ Extended error information every callpath to a kmalloc for a kernel compile):: # cat /sys/kernel/tracing/events/kmem/kmalloc/hist - # trigger info: hist:keys=stacktrace:vals=bytes_req,bytes_alloc:sort=bytes_alloc:size=2048 [active] + # trigger info: hist:keys=common_stacktrace:vals=bytes_req,bytes_alloc:sort=bytes_alloc:size=2048 [active] - { stacktrace: + { common_stacktrace: __kmalloc_track_caller+0x10b/0x1a0 kmemdup+0x20/0x50 hidraw_report_event+0x8a/0x120 [hid] @@ -581,7 +581,7 @@ Extended error information cpu_startup_entry+0x315/0x3e0 rest_init+0x7c/0x80 } hitcount: 3 bytes_req: 21 bytes_alloc: 24 - { stacktrace: + { common_stacktrace: __kmalloc_track_caller+0x10b/0x1a0 kmemdup+0x20/0x50 hidraw_report_event+0x8a/0x120 [hid] @@ -596,7 +596,7 @@ Extended error information do_IRQ+0x5a/0xf0 ret_from_intr+0x0/0x30 } hitcount: 3 bytes_req: 21 bytes_alloc: 24 - { stacktrace: + { common_stacktrace: kmem_cache_alloc_trace+0xeb/0x150 aa_alloc_task_context+0x27/0x40 apparmor_cred_prepare+0x1f/0x50 @@ -608,7 +608,7 @@ Extended error information . . . - { stacktrace: + { common_stacktrace: __kmalloc+0x11b/0x1b0 i915_gem_execbuffer2+0x6c/0x2c0 [i915] drm_ioctl+0x349/0x670 [drm] @@ -616,7 +616,7 @@ Extended error information SyS_ioctl+0x81/0xa0 system_call_fastpath+0x12/0x6a } hitcount: 17726 bytes_req: 13944120 bytes_alloc: 19593808 - { stacktrace: + { common_stacktrace: __kmalloc+0x11b/0x1b0 load_elf_phdrs+0x76/0xa0 load_elf_binary+0x102/0x1650 @@ -625,7 +625,7 @@ Extended error information SyS_execve+0x3a/0x50 return_from_execve+0x0/0x23 } hitcount: 33348 bytes_req: 17152128 bytes_alloc: 20226048 - { stacktrace: + { common_stacktrace: kmem_cache_alloc_trace+0xeb/0x150 apparmor_file_alloc_security+0x27/0x40 security_file_alloc+0x16/0x20 @@ -636,7 +636,7 @@ Extended error information SyS_open+0x1e/0x20 system_call_fastpath+0x12/0x6a } hitcount: 4766422 bytes_req: 9532844 bytes_alloc: 38131376 - { stacktrace: + { common_stacktrace: __kmalloc+0x11b/0x1b0 seq_buf_alloc+0x1b/0x50 seq_read+0x2cc/0x370 @@ -1026,7 +1026,7 @@ Extended error information First we set up an initially paused stacktrace trigger on the netif_receive_skb event:: - # echo 'hist:key=stacktrace:vals=len:pause' > \ + # echo 'hist:key=common_stacktrace:vals=len:pause' > \ /sys/kernel/tracing/events/net/netif_receive_skb/trigger Next, we set up an 'enable_hist' trigger on the sched_process_exec @@ -1060,9 +1060,9 @@ Extended error information $ wget https://www.kernel.org/pub/linux/kernel/v3.x/patch-3.19.xz # cat /sys/kernel/tracing/events/net/netif_receive_skb/hist - # trigger info: hist:keys=stacktrace:vals=len:sort=hitcount:size=2048 [paused] + # trigger info: hist:keys=common_stacktrace:vals=len:sort=hitcount:size=2048 [paused] - { stacktrace: + { common_stacktrace: __netif_receive_skb_core+0x46d/0x990 __netif_receive_skb+0x18/0x60 netif_receive_skb_internal+0x23/0x90 @@ -1079,7 +1079,7 @@ Extended error information kthread+0xd2/0xf0 ret_from_fork+0x42/0x70 } hitcount: 85 len: 28884 - { stacktrace: + { common_stacktrace: __netif_receive_skb_core+0x46d/0x990 __netif_receive_skb+0x18/0x60 netif_receive_skb_internal+0x23/0x90 @@ -1097,7 +1097,7 @@ Extended error information irq_thread+0x11f/0x150 kthread+0xd2/0xf0 } hitcount: 98 len: 664329 - { stacktrace: + { common_stacktrace: __netif_receive_skb_core+0x46d/0x990 __netif_receive_skb+0x18/0x60 process_backlog+0xa8/0x150 @@ -1115,7 +1115,7 @@ Extended error information inet_sendmsg+0x64/0xa0 sock_sendmsg+0x3d/0x50 } hitcount: 115 len: 13030 - { stacktrace: + { common_stacktrace: __netif_receive_skb_core+0x46d/0x990 __netif_receive_skb+0x18/0x60 netif_receive_skb_internal+0x23/0x90 @@ -1142,14 +1142,14 @@ Extended error information into the histogram. In order to avoid having to set everything up again, we can just clear the histogram first:: - # echo 'hist:key=stacktrace:vals=len:clear' >> \ + # echo 'hist:key=common_stacktrace:vals=len:clear' >> \ /sys/kernel/tracing/events/net/netif_receive_skb/trigger Just to verify that it is in fact cleared, here's what we now see in the hist file:: # cat /sys/kernel/tracing/events/net/netif_receive_skb/hist - # trigger info: hist:keys=stacktrace:vals=len:sort=hitcount:size=2048 [paused] + # trigger info: hist:keys=common_stacktrace:vals=len:sort=hitcount:size=2048 [paused] Totals: Hits: 0 @@ -1485,12 +1485,12 @@ Extended error information And here's an example that shows how to combine histogram data from any two events even if they don't share any 'compatible' fields - other than 'hitcount' and 'stacktrace'. These commands create a + other than 'hitcount' and 'common_stacktrace'. These commands create a couple of triggers named 'bar' using those fields:: - # echo 'hist:name=bar:key=stacktrace:val=hitcount' > \ + # echo 'hist:name=bar:key=common_stacktrace:val=hitcount' > \ /sys/kernel/tracing/events/sched/sched_process_fork/trigger - # echo 'hist:name=bar:key=stacktrace:val=hitcount' > \ + # echo 'hist:name=bar:key=common_stacktrace:val=hitcount' > \ /sys/kernel/tracing/events/net/netif_rx/trigger And displaying the output of either shows some interesting if @@ -1501,16 +1501,16 @@ Extended error information # event histogram # - # trigger info: hist:name=bar:keys=stacktrace:vals=hitcount:sort=hitcount:size=2048 [active] + # trigger info: hist:name=bar:keys=common_stacktrace:vals=hitcount:sort=hitcount:size=2048 [active] # - { stacktrace: + { common_stacktrace: kernel_clone+0x18e/0x330 kernel_thread+0x29/0x30 kthreadd+0x154/0x1b0 ret_from_fork+0x3f/0x70 } hitcount: 1 - { stacktrace: + { common_stacktrace: netif_rx_internal+0xb2/0xd0 netif_rx_ni+0x20/0x70 dev_loopback_xmit+0xaa/0xd0 @@ -1528,7 +1528,7 @@ Extended error information call_cpuidle+0x3b/0x60 cpu_startup_entry+0x22d/0x310 } hitcount: 1 - { stacktrace: + { common_stacktrace: netif_rx_internal+0xb2/0xd0 netif_rx_ni+0x20/0x70 dev_loopback_xmit+0xaa/0xd0 @@ -1543,7 +1543,7 @@ Extended error information SyS_sendto+0xe/0x10 entry_SYSCALL_64_fastpath+0x12/0x6a } hitcount: 2 - { stacktrace: + { common_stacktrace: netif_rx_internal+0xb2/0xd0 netif_rx+0x1c/0x60 loopback_xmit+0x6c/0xb0 @@ -1561,7 +1561,7 @@ Extended error information sock_sendmsg+0x38/0x50 ___sys_sendmsg+0x14e/0x270 } hitcount: 76 - { stacktrace: + { common_stacktrace: netif_rx_internal+0xb2/0xd0 netif_rx+0x1c/0x60 loopback_xmit+0x6c/0xb0 @@ -1579,7 +1579,7 @@ Extended error information sock_sendmsg+0x38/0x50 ___sys_sendmsg+0x269/0x270 } hitcount: 77 - { stacktrace: + { common_stacktrace: netif_rx_internal+0xb2/0xd0 netif_rx+0x1c/0x60 loopback_xmit+0x6c/0xb0 @@ -1597,7 +1597,7 @@ Extended error information sock_sendmsg+0x38/0x50 SYSC_sendto+0xef/0x170 } hitcount: 88 - { stacktrace: + { common_stacktrace: kernel_clone+0x18e/0x330 SyS_clone+0x19/0x20 entry_SYSCALL_64_fastpath+0x12/0x6a @@ -1949,7 +1949,7 @@ uninterruptible state:: # cd /sys/kernel/tracing # echo 's:block_lat pid_t pid; u64 delta; unsigned long[] stack;' > dynamic_events - # echo 'hist:keys=next_pid:ts=common_timestamp.usecs,st=stacktrace if prev_state == 2' >> events/sched/sched_switch/trigger + # echo 'hist:keys=next_pid:ts=common_timestamp.usecs,st=common_stacktrace if prev_state == 2' >> events/sched/sched_switch/trigger # echo 'hist:keys=prev_pid:delta=common_timestamp.usecs-$ts,s=$st:onmax($delta).trace(block_lat,prev_pid,$delta,$s)' >> events/sched/sched_switch/trigger # echo 1 > events/synthetic/block_lat/enable # cat trace diff --git a/Documentation/translations/zh_CN/devicetree/usage-model.rst b/Documentation/translations/zh_CN/devicetree/usage-model.rst index c6aee82c7e6e72535d12abcc117d601f8d7557ae..19ba4ae0cd811e055891452e349f55e77c3e9a6b 100644 --- a/Documentation/translations/zh_CN/devicetree/usage-model.rst +++ b/Documentation/translations/zh_CN/devicetree/usage-model.rst @@ -325,6 +325,6 @@ Primecell设备。然而，棘手的一点是，AMBA总线上的所有设备并 当使用DT时，这给of_platform_populate()带来了问题，因为它必须决定是否将 每个节点注册为platform_device或amba_device。不幸的是，这使设备创建模型 -变得有点复杂，但解决方案原来并不是太具有侵略性。如果一个节点与“arm,amba-primecell” +变得有点复杂，但解决方案原来并不是太具有侵略性。如果一个节点与“arm,primecell” 兼容，那么of_platform_populate()将把它注册为amba_device而不是 platform_device。 diff --git a/Documentation/userspace-api/ioctl/ioctl-number.rst b/Documentation/userspace-api/ioctl/ioctl-number.rst index 176e8fc3f31b9393421c9c75073d927911f1b4c2..4f7b23faebb98c74022604ff24084eb1e8257524 100644 --- a/Documentation/userspace-api/ioctl/ioctl-number.rst +++ b/Documentation/userspace-api/ioctl/ioctl-number.rst @@ -363,7 +363,7 @@ Code Seq# Include File Comments 0xCC 00-0F drivers/misc/ibmvmc.h pseries VMC driver 0xCD 01 linux/reiserfs_fs.h 0xCE 01-02 uapi/linux/cxl_mem.h Compute Express Link Memory Devices -0xCF 02 fs/cifs/ioctl.c +0xCF 02 fs/smb/client/cifs_ioctl.h 0xDB 00-0F drivers/char/mwave/mwavepub.h 0xDD 00-3F ZFCP device driver see drivers/s390/scsi/ diff --git a/MAINTAINERS b/MAINTAINERS index 29f242734eee356e0747e0aaec122cf79f75f0f7..44f5ad75cc7938deee93040127d18345b355bf94 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -956,7 +956,8 @@ F: Documentation/networking/device_drivers/ethernet/amazon/ena.rst F: drivers/net/ethernet/amazon/ AMAZON RDMA EFA DRIVER -M: Gal Pressman +M: Michael Margolin +R: Gal Pressman R: Yossi Leybovich L: linux-rdma@vger.kernel.org S: Supported @@ -1600,7 +1601,7 @@ F: drivers/media/i2c/ar0521.c ARASAN NAND CONTROLLER DRIVER M: Miquel Raynal -M: Naga Sureshkumar Relli +R: Michal Simek L: linux-mtd@lists.infradead.org S: Maintained F: Documentation/devicetree/bindings/mtd/arasan,nand-controller.yaml @@ -1677,10 +1678,7 @@ F: drivers/power/reset/arm-versatile-reboot.c F: drivers/soc/versatile/ ARM KOMEDA DRM-KMS DRIVER -M: James (Qian) Wang M: Liviu Dudau -M: Mihail Atanassov -L: Mali DP Maintainers S: Supported T: git git://anongit.freedesktop.org/drm/drm-misc F: Documentation/devicetree/bindings/display/arm,komeda.yaml @@ -1701,8 +1699,6 @@ F: include/uapi/drm/panfrost_drm.h ARM MALI-DP DRM DRIVER M: Liviu Dudau -M: Brian Starkey -L: Mali DP Maintainers S: Supported T: git git://anongit.freedesktop.org/drm/drm-misc F: Documentation/devicetree/bindings/display/arm,malidp.yaml @@ -1768,7 +1764,7 @@ F: include/linux/amba/mmci.h ARM PRIMECELL PL35X NAND CONTROLLER DRIVER M: Miquel Raynal -M: Naga Sureshkumar Relli +R: Michal Simek L: linux-mtd@lists.infradead.org S: Maintained F: Documentation/devicetree/bindings/mtd/arm,pl353-nand-r2p1.yaml @@ -1776,7 +1772,7 @@ F: drivers/mtd/nand/raw/pl35x-nand-controller.c ARM PRIMECELL PL35X SMC DRIVER M: Miquel Raynal -M: Naga Sureshkumar Relli +R: Michal Simek L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) S: Maintained F: Documentation/devicetree/bindings/memory-controllers/arm,pl35x-smc.yaml @@ -2434,6 +2430,15 @@ X: drivers/net/wireless/atmel/ N: at91 N: atmel +ARM/MICROCHIP (ARM64) SoC support +M: Conor Dooley +M: Nicolas Ferre +M: Claudiu Beznea +L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) +S: Supported +T: git https://git.kernel.org/pub/scm/linux/kernel/git/at91/linux.git +F: arch/arm64/boot/dts/microchip/ + ARM/Microchip Sparx5 SoC support M: Lars Povlsen M: Steen Hegelund @@ -2441,8 +2446,7 @@ M: Daniel Machon M: UNGLinuxDriver@microchip.com L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) S: Supported -T: git git://github.com/microchip-ung/linux-upstream.git -F: arch/arm64/boot/dts/microchip/ +F: arch/arm64/boot/dts/microchip/sparx* F: drivers/net/ethernet/microchip/vcap/ F: drivers/pinctrl/pinctrl-microchip-sgpio.c N: sparx5 @@ -3541,7 +3545,7 @@ F: Documentation/filesystems/befs.rst F: fs/befs/ BFQ I/O SCHEDULER -M: Paolo Valente +M: Paolo Valente M: Jens Axboe L: linux-block@vger.kernel.org S: Maintained @@ -4914,7 +4918,6 @@ F: drivers/media/cec/i2c/ch7322.c CIRRUS LOGIC AUDIO CODEC DRIVERS M: James Schulman M: David Rhodes -M: Lucas Tanure M: Richard Fitzgerald L: alsa-devel@alsa-project.org (moderated for non-subscribers) L: patches@opensource.cirrus.com @@ -5136,7 +5139,7 @@ X: drivers/clk/clkdev.c COMMON INTERNET FILE SYSTEM CLIENT (CIFS and SMB3) M: Steve French -R: Paulo Alcantara (DFS, global name space) +R: Paulo Alcantara (DFS, global name space) R: Ronnie Sahlberg (directory leases, sparse files) R: Shyam Prasad N (multichannel) R: Tom Talpey (RDMA, smbdirect) @@ -5146,8 +5149,8 @@ S: Supported W: https://wiki.samba.org/index.php/LinuxCIFS T: git git://git.samba.org/sfrench/cifs-2.6.git F: Documentation/admin-guide/cifs/ -F: fs/cifs/ -F: fs/smbfs_common/ +F: fs/smb/client/ +F: fs/smb/common/ F: include/uapi/linux/cifs COMPACTPCI HOTPLUG CORE @@ -5725,6 +5728,14 @@ F: include/linux/tfrc.h F: include/uapi/linux/dccp.h F: net/dccp/ +DEBUGOBJECTS: +M: Thomas Gleixner +L: linux-kernel@vger.kernel.org +S: Maintained +T: git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git core/debugobjects +F: lib/debugobjects.c +F: include/linux/debugobjects.h + DECSTATION PLATFORM SUPPORT M: "Maciej W. Rozycki" L: linux-mips@vger.kernel.org @@ -6012,7 +6023,7 @@ W: http://www.dialog-semiconductor.com/products F: Documentation/devicetree/bindings/input/da90??-onkey.txt F: Documentation/devicetree/bindings/input/dlg,da72??.txt F: Documentation/devicetree/bindings/mfd/da90*.txt -F: Documentation/devicetree/bindings/mfd/da90*.yaml +F: Documentation/devicetree/bindings/mfd/dlg,da90*.yaml F: Documentation/devicetree/bindings/regulator/da92*.txt F: Documentation/devicetree/bindings/regulator/dlg,da9*.yaml F: Documentation/devicetree/bindings/regulator/slg51000.txt @@ -6211,6 +6222,7 @@ X: Documentation/devicetree/ X: Documentation/driver-api/media/ X: Documentation/firmware-guide/acpi/ X: Documentation/i2c/ +X: Documentation/netlink/ X: Documentation/power/ X: Documentation/spi/ X: Documentation/userspace-api/media/ @@ -6571,6 +6583,7 @@ M: Rob Clark M: Abhinav Kumar M: Dmitry Baryshkov R: Sean Paul +R: Marijn Suijten L: linux-arm-msm@vger.kernel.org L: dri-devel@lists.freedesktop.org L: freedreno@lists.freedesktop.org @@ -6981,8 +6994,7 @@ F: Documentation/devicetree/bindings/display/bridge/renesas,dsi-csi2-tx.yaml F: Documentation/devicetree/bindings/display/bridge/renesas,dw-hdmi.yaml F: Documentation/devicetree/bindings/display/bridge/renesas,lvds.yaml F: Documentation/devicetree/bindings/display/renesas,du.yaml -F: drivers/gpu/drm/rcar-du/ -F: drivers/gpu/drm/shmobile/ +F: drivers/gpu/drm/renesas/ F: include/linux/platform_data/shmob_drm.h DRM DRIVERS FOR ROCKCHIP @@ -8168,6 +8180,7 @@ F: include/linux/spi/spi-fsl-dspi.h FREESCALE ENETC ETHERNET DRIVERS M: Claudiu Manoil +M: Vladimir Oltean L: netdev@vger.kernel.org S: Maintained F: drivers/net/ethernet/freescale/enetc/ @@ -8796,6 +8809,7 @@ F: include/linux/gpio/regmap.h GPIO SUBSYSTEM M: Linus Walleij M: Bartosz Golaszewski +R: Andy Shevchenko L: linux-gpio@vger.kernel.org S: Maintained T: git git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux.git @@ -9349,7 +9363,7 @@ F: include/linux/hisi_acc_qm.h HISILICON ROCE DRIVER M: Haoyue Xu -M: Wenpeng Liang +M: Junxian Huang L: linux-rdma@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/infiniband/hisilicon-hns-roce.txt @@ -9692,8 +9706,9 @@ F: include/uapi/linux/i2c-*.h F: include/uapi/linux/i2c.h I2C SUBSYSTEM HOST DRIVERS +M: Andi Shyti L: linux-i2c@vger.kernel.org -S: Odd Fixes +S: Maintained W: https://i2c.wiki.kernel.org/ Q: https://patchwork.ozlabs.org/project/linux-i2c/list/ T: git git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux.git @@ -10120,7 +10135,7 @@ S: Maintained F: Documentation/process/kernel-docs.rst INDUSTRY PACK SUBSYSTEM (IPACK) -M: Samuel Iglesias Gonsalvez +M: Vaibhav Gupta M: Jens Taprogge M: Greg Kroah-Hartman L: industrypack-devel@lists.sourceforge.net @@ -11315,9 +11330,9 @@ R: Tom Talpey L: linux-cifs@vger.kernel.org S: Maintained T: git git://git.samba.org/ksmbd.git -F: Documentation/filesystems/cifs/ksmbd.rst -F: fs/ksmbd/ -F: fs/smbfs_common/ +F: Documentation/filesystems/smb/ksmbd.rst +F: fs/smb/common/ +F: fs/smb/server/ KERNEL UNIT TESTING FRAMEWORK (KUnit) M: Brendan Higgins @@ -13842,7 +13857,7 @@ F: drivers/tty/serial/8250/8250_pci1xxxx.c MICROCHIP POLARFIRE FPGA DRIVERS M: Conor Dooley -R: Ivan Bornyakov +R: Vladimir Georgiev L: linux-fpga@vger.kernel.org S: Supported F: Documentation/devicetree/bindings/fpga/microchip,mpf-spi-fpga-mgr.yaml @@ -14576,6 +14591,7 @@ T: git git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next.git F: Documentation/devicetree/bindings/net/ F: drivers/connector/ F: drivers/net/ +X: drivers/net/wireless/ F: include/dt-bindings/net/ F: include/linux/etherdevice.h F: include/linux/fcdevice.h @@ -14625,6 +14641,7 @@ B: mailto:netdev@vger.kernel.org T: git git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net.git T: git git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next.git F: Documentation/core-api/netlink.rst +F: Documentation/netlink/ F: Documentation/networking/ F: Documentation/process/maintainer-netdev.rst F: Documentation/userspace-api/netlink/ @@ -14639,6 +14656,7 @@ F: include/uapi/linux/netdevice.h F: lib/net_utils.c F: lib/random32.c F: net/ +X: net/bluetooth/ F: tools/net/ F: tools/testing/selftests/net/ @@ -14938,6 +14956,7 @@ F: drivers/ntb/hw/intel/ NTFS FILESYSTEM M: Anton Altaparmakov +R: Namjae Jeon L: linux-ntfs-dev@lists.sourceforge.net S: Supported W: http://www.tuxera.com/ @@ -18041,6 +18060,14 @@ S: Maintained F: Documentation/devicetree/bindings/usb/renesas,rzn1-usbf.yaml F: drivers/usb/gadget/udc/renesas_usbf.c +RENESAS RZ/V2M I2C DRIVER +M: Fabrizio Castro +L: linux-i2c@vger.kernel.org +L: linux-renesas-soc@vger.kernel.org +S: Supported +F: Documentation/devicetree/bindings/i2c/renesas,rzv2m.yaml +F: drivers/i2c/busses/i2c-rzv2m.c + RENESAS USB PHY DRIVER M: Yoshihiro Shimoda L: linux-renesas-soc@vger.kernel.org @@ -18587,10 +18614,9 @@ F: Documentation/admin-guide/LSM/SafeSetID.rst F: security/safesetid/ SAMSUNG AUDIO (ASoC) DRIVERS -M: Krzysztof Kozlowski M: Sylwester Nawrocki L: alsa-devel@alsa-project.org (moderated for non-subscribers) -S: Supported +S: Maintained B: mailto:linux-samsung-soc@vger.kernel.org F: Documentation/devicetree/bindings/sound/samsung* F: sound/soc/samsung/ @@ -18718,7 +18744,6 @@ F: include/dt-bindings/clock/samsung,*.h F: include/linux/clk/samsung.h SAMSUNG SPI DRIVERS -M: Krzysztof Kozlowski M: Andi Shyti L: linux-spi@vger.kernel.org L: linux-samsung-soc@vger.kernel.org @@ -18854,12 +18879,11 @@ F: drivers/target/ F: include/target/ SCTP PROTOCOL -M: Neil Horman M: Marcelo Ricardo Leitner M: Xin Long L: linux-sctp@vger.kernel.org S: Maintained -W: http://lksctp.sourceforge.net +W: https://github.com/sctp/lksctp-tools/wiki F: Documentation/networking/sctp.rst F: include/linux/sctp.h F: include/net/sctp/ @@ -19128,6 +19152,9 @@ SHARED MEMORY COMMUNICATIONS (SMC) SOCKETS M: Karsten Graul M: Wenjia Zhang M: Jan Karcher +R: D. Wythe +R: Tony Lu +R: Wen Gu L: linux-s390@vger.kernel.org S: Supported F: net/smc/ diff --git a/Makefile b/Makefile index f836936fb4d8b919f0b1692bff41b4df61dfd0ec..b68b43c1907225428c34bc7b9022f6d8eb9416ce 100644 --- a/Makefile +++ b/Makefile @@ -2,7 +2,7 @@ VERSION = 6 PATCHLEVEL = 4 SUBLEVEL = 0 -EXTRAVERSION = -rc2 +EXTRAVERSION = -rc7 NAME = Hurr durr I'ma ninja sloth # *DOCUMENTATION* diff --git a/arch/arm/boot/dts/am57xx-cl-som-am57x.dts b/arch/arm/boot/dts/am57xx-cl-som-am57x.dts index 2fc9a5d5e0c0dea8668c230860c327b41134df60..625b9b311b49d2791f949ed06285838a6c36fca9 100644 --- a/arch/arm/boot/dts/am57xx-cl-som-am57x.dts +++ b/arch/arm/boot/dts/am57xx-cl-som-am57x.dts @@ -527,7 +527,7 @@ touchscreen@1 { interrupt-parent = <&gpio1>; interrupts = <31 0>; - pendown-gpio = <&gpio1 31 0>; + pendown-gpio = <&gpio1 31 GPIO_ACTIVE_LOW>; ti,x-min = /bits/ 16 <0x0>; diff --git a/arch/arm/boot/dts/at91-sama7g5ek.dts b/arch/arm/boot/dts/at91-sama7g5ek.dts index aa5cc0e98bbab2b5e6512eebbaf0292b2e3467cd..217e9b96c61e5dea644c6b2f46d5807f58312b38 100644 --- a/arch/arm/boot/dts/at91-sama7g5ek.dts +++ b/arch/arm/boot/dts/at91-sama7g5ek.dts @@ -792,7 +792,7 @@ &sdmmc2 { }; &shdwc { - atmel,shdwc-debouncer = <976>; + debounce-delay-us = <976>; status = "okay"; input@0 { diff --git a/arch/arm/boot/dts/at91sam9261ek.dts b/arch/arm/boot/dts/at91sam9261ek.dts index 88869ca874d1ae3cab71bb2ff98d8a739822877f..045cb253f23a6d07a39724f23be412927975e570 100644 --- a/arch/arm/boot/dts/at91sam9261ek.dts +++ b/arch/arm/boot/dts/at91sam9261ek.dts @@ -156,7 +156,7 @@ tsc2046@2 { compatible = "ti,ads7843"; interrupts-extended = <&pioC 2 IRQ_TYPE_EDGE_BOTH>; spi-max-frequency = <3000000>; - pendown-gpio = <&pioC 2 GPIO_ACTIVE_HIGH>; + pendown-gpio = <&pioC 2 GPIO_ACTIVE_LOW>; ti,x-min = /bits/ 16 <150>; ti,x-max = /bits/ 16 <3830>; diff --git a/arch/arm/boot/dts/imx6qdl-mba6.dtsi b/arch/arm/boot/dts/imx6qdl-mba6.dtsi index 78555a6188510f63da0bf1a2659423cbc56205de..7b7e6c2ad190c838b38c8e9a74ceea01992fdc1b 100644 --- a/arch/arm/boot/dts/imx6qdl-mba6.dtsi +++ b/arch/arm/boot/dts/imx6qdl-mba6.dtsi @@ -209,6 +209,7 @@ &pcie { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_pcie>; reset-gpio = <&gpio6 7 GPIO_ACTIVE_LOW>; + vpcie-supply = <®_pcie>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx6ull-dhcor-som.dtsi b/arch/arm/boot/dts/imx6ull-dhcor-som.dtsi index 5882c7565f6495b31c0d2c5f88431658e3ecf758..32a6022625d979041eb038c70213e696eb81b3ed 100644 --- a/arch/arm/boot/dts/imx6ull-dhcor-som.dtsi +++ b/arch/arm/boot/dts/imx6ull-dhcor-som.dtsi @@ -8,6 +8,7 @@ #include #include #include +#include #include "imx6ull.dtsi" / { @@ -84,16 +85,20 @@ onkey { regulators { vdd_soc_in_1v4: buck1 { + regulator-allowed-modes = ; /* PFM */ regulator-always-on; regulator-boot-on; + regulator-initial-mode = ; regulator-max-microvolt = <1400000>; regulator-min-microvolt = <1400000>; regulator-name = "vdd_soc_in_1v4"; }; vcc_3v3: buck2 { + regulator-allowed-modes = ; /* PWM */ regulator-always-on; regulator-boot-on; + regulator-initial-mode = ; regulator-max-microvolt = <3300000>; regulator-min-microvolt = <3300000>; regulator-name = "vcc_3v3"; @@ -106,8 +111,10 @@ vcc_3v3: buck2 { * the voltage is set to 1.5V. */ vcc_ddr_1v35: buck3 { + regulator-allowed-modes = ; /* PWM */ regulator-always-on; regulator-boot-on; + regulator-initial-mode = ; regulator-max-microvolt = <1500000>; regulator-min-microvolt = <1500000>; regulator-name = "vcc_ddr_1v35"; diff --git a/arch/arm/boot/dts/imx7d-pico-hobbit.dts b/arch/arm/boot/dts/imx7d-pico-hobbit.dts index d917dc4f2f22759bc546c18248bbff7fcc3d726f..6ad39dca70096601ff1b7b8065f040e7c1d711a9 100644 --- a/arch/arm/boot/dts/imx7d-pico-hobbit.dts +++ b/arch/arm/boot/dts/imx7d-pico-hobbit.dts @@ -64,7 +64,7 @@ ads7846@0 { interrupt-parent = <&gpio2>; interrupts = <7 0>; spi-max-frequency = <1000000>; - pendown-gpio = <&gpio2 7 0>; + pendown-gpio = <&gpio2 7 GPIO_ACTIVE_LOW>; vcc-supply = <®_3p3v>; ti,x-min = /bits/ 16 <0>; ti,x-max = /bits/ 16 <4095>; diff --git a/arch/arm/boot/dts/imx7d-sdb.dts b/arch/arm/boot/dts/imx7d-sdb.dts index f483bc0afe5ea75f8235f32ce8c06a60ba4e0eec..234e5fc647b22829891b0a38c2490ded483cabde 100644 --- a/arch/arm/boot/dts/imx7d-sdb.dts +++ b/arch/arm/boot/dts/imx7d-sdb.dts @@ -205,7 +205,7 @@ tsc2046@0 { pinctrl-0 = <&pinctrl_tsc2046_pendown>; interrupt-parent = <&gpio2>; interrupts = <29 0>; - pendown-gpio = <&gpio2 29 GPIO_ACTIVE_HIGH>; + pendown-gpio = <&gpio2 29 GPIO_ACTIVE_LOW>; touchscreen-max-pressure = <255>; wakeup-source; }; diff --git a/arch/arm/boot/dts/omap3-cm-t3x.dtsi b/arch/arm/boot/dts/omap3-cm-t3x.dtsi index e61b8a2bfb7deafd2d35da4a7b77c46ddba5db0b..51baedf1603bd43248ad4ed15b36dfdb5b6d6274 100644 --- a/arch/arm/boot/dts/omap3-cm-t3x.dtsi +++ b/arch/arm/boot/dts/omap3-cm-t3x.dtsi @@ -227,7 +227,7 @@ ads7846@0 { interrupt-parent = <&gpio2>; interrupts = <25 0>; /* gpio_57 */ - pendown-gpio = <&gpio2 25 GPIO_ACTIVE_HIGH>; + pendown-gpio = <&gpio2 25 GPIO_ACTIVE_LOW>; ti,x-min = /bits/ 16 <0x0>; ti,x-max = /bits/ 16 <0x0fff>; diff --git a/arch/arm/boot/dts/omap3-devkit8000-lcd-common.dtsi b/arch/arm/boot/dts/omap3-devkit8000-lcd-common.dtsi index 3decc2d78a6ca0776d16fd8bb74dbdefcd54466a..a7f99ae0c1fe9aca257d39458668640ea05ff222 100644 --- a/arch/arm/boot/dts/omap3-devkit8000-lcd-common.dtsi +++ b/arch/arm/boot/dts/omap3-devkit8000-lcd-common.dtsi @@ -54,7 +54,7 @@ ads7846@0 { interrupt-parent = <&gpio1>; interrupts = <27 0>; /* gpio_27 */ - pendown-gpio = <&gpio1 27 GPIO_ACTIVE_HIGH>; + pendown-gpio = <&gpio1 27 GPIO_ACTIVE_LOW>; ti,x-min = /bits/ 16 <0x0>; ti,x-max = /bits/ 16 <0x0fff>; diff --git a/arch/arm/boot/dts/omap3-lilly-a83x.dtsi b/arch/arm/boot/dts/omap3-lilly-a83x.dtsi index c595afe4181d7839139b620689575a8006a04093..d310b5c7bac362e8e2ae24e512d3667841fc9752 100644 --- a/arch/arm/boot/dts/omap3-lilly-a83x.dtsi +++ b/arch/arm/boot/dts/omap3-lilly-a83x.dtsi @@ -311,7 +311,7 @@ tsc2046@0 { interrupt-parent = <&gpio1>; interrupts = <8 0>; /* boot6 / gpio_8 */ spi-max-frequency = <1000000>; - pendown-gpio = <&gpio1 8 GPIO_ACTIVE_HIGH>; + pendown-gpio = <&gpio1 8 GPIO_ACTIVE_LOW>; vcc-supply = <®_vcc3>; pinctrl-names = "default"; pinctrl-0 = <&tsc2048_pins>; diff --git a/arch/arm/boot/dts/omap3-overo-common-lcd35.dtsi b/arch/arm/boot/dts/omap3-overo-common-lcd35.dtsi index 1d6e88f99eb317fcdfb117b3a8ac050ee8fb6d82..c3570acc35fadaabafeb7ea1433b6322535b322f 100644 --- a/arch/arm/boot/dts/omap3-overo-common-lcd35.dtsi +++ b/arch/arm/boot/dts/omap3-overo-common-lcd35.dtsi @@ -149,7 +149,7 @@ ads7846@0 { interrupt-parent = <&gpio4>; interrupts = <18 0>; /* gpio_114 */ - pendown-gpio = <&gpio4 18 GPIO_ACTIVE_HIGH>; + pendown-gpio = <&gpio4 18 GPIO_ACTIVE_LOW>; ti,x-min = /bits/ 16 <0x0>; ti,x-max = /bits/ 16 <0x0fff>; diff --git a/arch/arm/boot/dts/omap3-overo-common-lcd43.dtsi b/arch/arm/boot/dts/omap3-overo-common-lcd43.dtsi index 7e30f9d45790e7c804453837ac34aa6b47f31ce7..d95a0e130058cbe7c6ef3c1c3771970f77d48bea 100644 --- a/arch/arm/boot/dts/omap3-overo-common-lcd43.dtsi +++ b/arch/arm/boot/dts/omap3-overo-common-lcd43.dtsi @@ -160,7 +160,7 @@ ads7846@0 { interrupt-parent = <&gpio4>; interrupts = <18 0>; /* gpio_114 */ - pendown-gpio = <&gpio4 18 GPIO_ACTIVE_HIGH>; + pendown-gpio = <&gpio4 18 GPIO_ACTIVE_LOW>; ti,x-min = /bits/ 16 <0x0>; ti,x-max = /bits/ 16 <0x0fff>; diff --git a/arch/arm/boot/dts/omap3-pandora-common.dtsi b/arch/arm/boot/dts/omap3-pandora-common.dtsi index 559853764487f28f171708aba3e00dbb436eda38..4c3b6bab179cc1be81e22f5735c1a245efb32385 100644 --- a/arch/arm/boot/dts/omap3-pandora-common.dtsi +++ b/arch/arm/boot/dts/omap3-pandora-common.dtsi @@ -651,7 +651,7 @@ tsc2046@0 { pinctrl-0 = <&penirq_pins>; interrupt-parent = <&gpio3>; interrupts = <30 IRQ_TYPE_NONE>; /* GPIO_94 */ - pendown-gpio = <&gpio3 30 GPIO_ACTIVE_HIGH>; + pendown-gpio = <&gpio3 30 GPIO_ACTIVE_LOW>; vcc-supply = <&vaux4>; ti,x-min = /bits/ 16 <0>; diff --git a/arch/arm/boot/dts/omap5-cm-t54.dts b/arch/arm/boot/dts/omap5-cm-t54.dts index 2d87b9fc230ee20c1936f41b1c026691e2b56b1a..af288d63a26a464b456e5f353a0464d61715022e 100644 --- a/arch/arm/boot/dts/omap5-cm-t54.dts +++ b/arch/arm/boot/dts/omap5-cm-t54.dts @@ -354,7 +354,7 @@ ads7846@0 { interrupt-parent = <&gpio1>; interrupts = <15 0>; /* gpio1_wk15 */ - pendown-gpio = <&gpio1 15 GPIO_ACTIVE_HIGH>; + pendown-gpio = <&gpio1 15 GPIO_ACTIVE_LOW>; ti,x-min = /bits/ 16 <0x0>; diff --git a/arch/arm/boot/dts/qcom-apq8026-asus-sparrow.dts b/arch/arm/boot/dts/qcom-apq8026-asus-sparrow.dts index 7a80e1c9f126b8ca63f01c5b7dbf7e5305bb3a57..aa0e0e8d2a973ea75f3f86e938f0c909c4ea73dd 100644 --- a/arch/arm/boot/dts/qcom-apq8026-asus-sparrow.dts +++ b/arch/arm/boot/dts/qcom-apq8026-asus-sparrow.dts @@ -268,7 +268,6 @@ bluetooth_default_state: bluetooth-default-state { function = "gpio"; drive-strength = <8>; bias-disable; - input-enable; }; wlan_hostwake_default_state: wlan-hostwake-default-state { @@ -276,7 +275,6 @@ wlan_hostwake_default_state: wlan-hostwake-default-state { function = "gpio"; drive-strength = <2>; bias-disable; - input-enable; }; wlan_regulator_default_state: wlan-regulator-default-state { diff --git a/arch/arm/boot/dts/qcom-apq8026-huawei-sturgeon.dts b/arch/arm/boot/dts/qcom-apq8026-huawei-sturgeon.dts index d64096028ab13b5216308624e781b2cb6749fcd0..5593a3a60d6c4dc732db544bee5ea8a5c6a08a25 100644 --- a/arch/arm/boot/dts/qcom-apq8026-huawei-sturgeon.dts +++ b/arch/arm/boot/dts/qcom-apq8026-huawei-sturgeon.dts @@ -352,7 +352,6 @@ wlan_hostwake_default_state: wlan-hostwake-default-state { function = "gpio"; drive-strength = <2>; bias-disable; - input-enable; }; wlan_regulator_default_state: wlan-regulator-default-state { diff --git a/arch/arm/boot/dts/qcom-apq8026-lg-lenok.dts b/arch/arm/boot/dts/qcom-apq8026-lg-lenok.dts index b82381229adf6ebdb1533af643e781a6717e6791..b887e5361ec3a22b280554486a2a96216906f1fc 100644 --- a/arch/arm/boot/dts/qcom-apq8026-lg-lenok.dts +++ b/arch/arm/boot/dts/qcom-apq8026-lg-lenok.dts @@ -307,7 +307,6 @@ bluetooth_default_state: bluetooth-default-state { function = "gpio"; drive-strength = <2>; bias-disable; - input-enable; }; touch_pins: touch-state { @@ -317,7 +316,6 @@ irq-pins { drive-strength = <8>; bias-pull-down; - input-enable; }; reset-pins { @@ -335,7 +333,6 @@ wlan_hostwake_default_state: wlan-hostwake-default-state { function = "gpio"; drive-strength = <2>; bias-disable; - input-enable; }; wlan_regulator_default_state: wlan-regulator-default-state { diff --git a/arch/arm/boot/dts/qcom-apq8064.dtsi b/arch/arm/boot/dts/qcom-apq8064.dtsi index 672b246afbba27cd0474c60509b68dfdd0f64f6b..d2289205ff812844b99766781b5bb00f1752eb1b 100644 --- a/arch/arm/boot/dts/qcom-apq8064.dtsi +++ b/arch/arm/boot/dts/qcom-apq8064.dtsi @@ -83,6 +83,7 @@ CPU3: cpu@3 { L2: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; idle-states { diff --git a/arch/arm/boot/dts/qcom-apq8084.dtsi b/arch/arm/boot/dts/qcom-apq8084.dtsi index b653ea40c441c6c119e9d3d68d76120d5cfbe39d..83839e1ec4d16cc1534eb176c694e71111ba3213 100644 --- a/arch/arm/boot/dts/qcom-apq8084.dtsi +++ b/arch/arm/boot/dts/qcom-apq8084.dtsi @@ -74,6 +74,7 @@ cpu@3 { L2: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; qcom,saw = <&saw_l2>; }; diff --git a/arch/arm/boot/dts/qcom-ipq4019.dtsi b/arch/arm/boot/dts/qcom-ipq4019.dtsi index dfcfb3339c232e063502fac0feae3695b3559ada..f0ef86fadc9d9a70bc1a99b63a6b861a7ae90fbd 100644 --- a/arch/arm/boot/dts/qcom-ipq4019.dtsi +++ b/arch/arm/boot/dts/qcom-ipq4019.dtsi @@ -102,6 +102,7 @@ cpu@3 { L2: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; qcom,saw = <&saw_l2>; }; }; diff --git a/arch/arm/boot/dts/qcom-ipq8064.dtsi b/arch/arm/boot/dts/qcom-ipq8064.dtsi index af6764770fd15e9a12728878d393b9ae38d24419..7581845737a8b507fcfab4cfb5547bc89c382484 100644 --- a/arch/arm/boot/dts/qcom-ipq8064.dtsi +++ b/arch/arm/boot/dts/qcom-ipq8064.dtsi @@ -45,6 +45,7 @@ cpu1: cpu@1 { L2: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm/boot/dts/qcom-mdm9615-wp8548-mangoh-green.dts b/arch/arm/boot/dts/qcom-mdm9615-wp8548-mangoh-green.dts index a8304769b509a8945bcdf90a34b38d465bf21cc4..b269fdca1460c8401bd595d8b58bc8e01b1c9607 100644 --- a/arch/arm/boot/dts/qcom-mdm9615-wp8548-mangoh-green.dts +++ b/arch/arm/boot/dts/qcom-mdm9615-wp8548-mangoh-green.dts @@ -49,7 +49,6 @@ gpioext1_pins: gpioext1-state { gpioext1-pins { pins = "gpio2"; function = "gpio"; - input-enable; bias-disable; }; }; diff --git a/arch/arm/boot/dts/qcom-msm8660.dtsi b/arch/arm/boot/dts/qcom-msm8660.dtsi index f601b40ebcf4e9e75caa868c919592ca8af83f5f..78023ed2fdf71f5168692d43a8b9d80573666545 100644 --- a/arch/arm/boot/dts/qcom-msm8660.dtsi +++ b/arch/arm/boot/dts/qcom-msm8660.dtsi @@ -36,6 +36,7 @@ cpu@1 { L2: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm/boot/dts/qcom-msm8960.dtsi b/arch/arm/boot/dts/qcom-msm8960.dtsi index 2a668cd535ccebb9f18c23ad8544190ba9b3f3e5..616fef2ea682570c6c49634525a309bdc55a585d 100644 --- a/arch/arm/boot/dts/qcom-msm8960.dtsi +++ b/arch/arm/boot/dts/qcom-msm8960.dtsi @@ -42,6 +42,7 @@ cpu@1 { L2: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm/boot/dts/qcom-msm8974-lge-nexus5-hammerhead.dts b/arch/arm/boot/dts/qcom-msm8974-lge-nexus5-hammerhead.dts index ab35f2d644c0e83b284d87ea8230ba9d9dcc0f1d..861695cecf84e5f57b2ecbaff4895e124b7ad672 100644 --- a/arch/arm/boot/dts/qcom-msm8974-lge-nexus5-hammerhead.dts +++ b/arch/arm/boot/dts/qcom-msm8974-lge-nexus5-hammerhead.dts @@ -592,7 +592,6 @@ mpu6515_pin: mpu6515-state { pins = "gpio73"; function = "gpio"; bias-disable; - input-enable; }; touch_pin: touch-state { @@ -602,7 +601,6 @@ int-pins { drive-strength = <2>; bias-disable; - input-enable; }; reset-pins { diff --git a/arch/arm/boot/dts/qcom-msm8974-sony-xperia-rhine.dtsi b/arch/arm/boot/dts/qcom-msm8974-sony-xperia-rhine.dtsi index d3bec03b126c8b68cb404fa68ab3b60e9579ef99..68a2f9094e536f758b25ed0e97d97fc87257b71e 100644 --- a/arch/arm/boot/dts/qcom-msm8974-sony-xperia-rhine.dtsi +++ b/arch/arm/boot/dts/qcom-msm8974-sony-xperia-rhine.dtsi @@ -433,7 +433,6 @@ ts_int_pin: touch-int-state { function = "gpio"; drive-strength = <2>; bias-disable; - input-enable; }; sdc1_on: sdc1-on-state { diff --git a/arch/arm/boot/dts/qcom-msm8974.dtsi b/arch/arm/boot/dts/qcom-msm8974.dtsi index 8208012684d4a28edf627bdac4f9e17837b226ba..7ed0d925a4e9903a362d62010adba97aef0b7141 100644 --- a/arch/arm/boot/dts/qcom-msm8974.dtsi +++ b/arch/arm/boot/dts/qcom-msm8974.dtsi @@ -80,6 +80,7 @@ CPU3: cpu@3 { L2: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; qcom,saw = <&saw_l2>; }; diff --git a/arch/arm/boot/dts/qcom-msm8974pro-oneplus-bacon.dts b/arch/arm/boot/dts/qcom-msm8974pro-oneplus-bacon.dts index 8d2a054d8feed72afa744122f73b3ff94275a861..8230d0e1d95d1de1bff1ad1af990e1850b6b4246 100644 --- a/arch/arm/boot/dts/qcom-msm8974pro-oneplus-bacon.dts +++ b/arch/arm/boot/dts/qcom-msm8974pro-oneplus-bacon.dts @@ -461,7 +461,6 @@ int-pins { function = "gpio"; drive-strength = <2>; bias-disable; - input-enable; }; reset-pins { diff --git a/arch/arm/boot/dts/qcom-msm8974pro-samsung-klte.dts b/arch/arm/boot/dts/qcom-msm8974pro-samsung-klte.dts index b9698ffb66ca82c14211beb5f0a501f28a3060a0..eb505d6d7f31f656ec575e7bd89071c5306e89ae 100644 --- a/arch/arm/boot/dts/qcom-msm8974pro-samsung-klte.dts +++ b/arch/arm/boot/dts/qcom-msm8974pro-samsung-klte.dts @@ -704,7 +704,6 @@ hostwake-pins { pins = "gpio75"; function = "gpio"; drive-strength = <16>; - input-enable; }; devwake-pins { @@ -760,14 +759,12 @@ cmd-data-pins { i2c_touchkey_pins: i2c-touchkey-state { pins = "gpio95", "gpio96"; function = "gpio"; - input-enable; bias-pull-up; }; i2c_led_gpioex_pins: i2c-led-gpioex-state { pins = "gpio120", "gpio121"; function = "gpio"; - input-enable; bias-pull-down; }; @@ -781,7 +778,6 @@ gpioex_pin: gpioex-state { wifi_pin: wifi-state { pins = "gpio92"; function = "gpio"; - input-enable; bias-pull-down; }; diff --git a/arch/arm/boot/dts/qcom-msm8974pro-sony-xperia-shinano-castor.dts b/arch/arm/boot/dts/qcom-msm8974pro-sony-xperia-shinano-castor.dts index 04bc58d87abf9c4948a5d025882ce6ec1e511705..0f650ed3100523854453c66411b8c26606a41531 100644 --- a/arch/arm/boot/dts/qcom-msm8974pro-sony-xperia-shinano-castor.dts +++ b/arch/arm/boot/dts/qcom-msm8974pro-sony-xperia-shinano-castor.dts @@ -631,7 +631,6 @@ ts_int_pin: ts-int-pin-state { function = "gpio"; drive-strength = <2>; bias-disable; - input-enable; }; bt_host_wake_pin: bt-host-wake-state { diff --git a/arch/arm/boot/dts/stm32f429.dtsi b/arch/arm/boot/dts/stm32f429.dtsi index c9e05e3540d6408791be69435b23bb3b131d554b..00bf53f99c294c6c183366798bbd3bca7569456b 100644 --- a/arch/arm/boot/dts/stm32f429.dtsi +++ b/arch/arm/boot/dts/stm32f429.dtsi @@ -387,6 +387,7 @@ can2: can@40006800 { interrupt-names = "tx", "rx0", "rx1", "sce"; resets = <&rcc STM32F4_APB1_RESET(CAN2)>; clocks = <&rcc 0 STM32F4_APB1_CLOCK(CAN2)>; + st,can-secondary; st,gcan = <&gcan>; status = "disabled"; }; diff --git a/arch/arm/boot/dts/stm32f7-pinctrl.dtsi b/arch/arm/boot/dts/stm32f7-pinctrl.dtsi index c8e6c52fb248e0bb405472ceefa11828fa7e07a9..9f65403295ca059fc917e1c575a32b77cab96a56 100644 --- a/arch/arm/boot/dts/stm32f7-pinctrl.dtsi +++ b/arch/arm/boot/dts/stm32f7-pinctrl.dtsi @@ -283,6 +283,88 @@ pins2 { slew-rate = <2>; }; }; + + can1_pins_a: can1-0 { + pins1 { + pinmux = ; /* CAN1_TX */ + }; + pins2 { + pinmux = ; /* CAN1_RX */ + bias-pull-up; + }; + }; + + can1_pins_b: can1-1 { + pins1 { + pinmux = ; /* CAN1_TX */ + }; + pins2 { + pinmux = ; /* CAN1_RX */ + bias-pull-up; + }; + }; + + can1_pins_c: can1-2 { + pins1 { + pinmux = ; /* CAN1_TX */ + }; + pins2 { + pinmux = ; /* CAN1_RX */ + bias-pull-up; + + }; + }; + + can1_pins_d: can1-3 { + pins1 { + pinmux = ; /* CAN1_TX */ + }; + pins2 { + pinmux = ; /* CAN1_RX */ + bias-pull-up; + + }; + }; + + can2_pins_a: can2-0 { + pins1 { + pinmux = ; /* CAN2_TX */ + }; + pins2 { + pinmux = ; /* CAN2_RX */ + bias-pull-up; + }; + }; + + can2_pins_b: can2-1 { + pins1 { + pinmux = ; /* CAN2_TX */ + }; + pins2 { + pinmux = ; /* CAN2_RX */ + bias-pull-up; + }; + }; + + can3_pins_a: can3-0 { + pins1 { + pinmux = ; /* CAN3_TX */ + }; + pins2 { + pinmux = ; /* CAN3_RX */ + bias-pull-up; + }; + }; + + can3_pins_b: can3-1 { + pins1 { + pinmux = ; /* CAN3_TX */ + }; + pins2 { + pinmux = ; /* CAN3_RX */ + bias-pull-up; + }; + }; }; }; }; diff --git a/arch/arm/boot/dts/vexpress-v2p-ca5s.dts b/arch/arm/boot/dts/vexpress-v2p-ca5s.dts index 3b88209bacea2f8664dcc4653f6bbca2397de4cb..ff1f9a1bcfcfc3bdffe1234ae103d876dfb891e9 100644 --- a/arch/arm/boot/dts/vexpress-v2p-ca5s.dts +++ b/arch/arm/boot/dts/vexpress-v2p-ca5s.dts @@ -132,6 +132,7 @@ L2: cache-controller@2c0f0000 { reg = <0x2c0f0000 0x1000>; interrupts = <0 84 4>; cache-level = <2>; + cache-unified; }; pmu { diff --git a/arch/arm/include/asm/arm_pmuv3.h b/arch/arm/include/asm/arm_pmuv3.h index 78d3d4b82c6c2598d63d4c691ba0e5ae59b67e5c..f4db3e75d75f024ef41943729e4558d6eb3bf604 100644 --- a/arch/arm/include/asm/arm_pmuv3.h +++ b/arch/arm/include/asm/arm_pmuv3.h @@ -92,7 +92,7 @@ #define RETURN_READ_PMEVCNTRN(n) \ return read_sysreg(PMEVCNTR##n) -static unsigned long read_pmevcntrn(int n) +static inline unsigned long read_pmevcntrn(int n) { PMEVN_SWITCH(n, RETURN_READ_PMEVCNTRN); return 0; @@ -100,14 +100,14 @@ static unsigned long read_pmevcntrn(int n) #define WRITE_PMEVCNTRN(n) \ write_sysreg(val, PMEVCNTR##n) -static void write_pmevcntrn(int n, unsigned long val) +static inline void write_pmevcntrn(int n, unsigned long val) { PMEVN_SWITCH(n, WRITE_PMEVCNTRN); } #define WRITE_PMEVTYPERN(n) \ write_sysreg(val, PMEVTYPER##n) -static void write_pmevtypern(int n, unsigned long val) +static inline void write_pmevtypern(int n, unsigned long val) { PMEVN_SWITCH(n, WRITE_PMEVTYPERN); } diff --git a/arch/arm/mach-at91/pm.c b/arch/arm/mach-at91/pm.c index 60dc56d8acfb9aa08a289077e35e959621e2a1f3..437dd0352fd443a887f0a5db99b4d0919475e075 100644 --- a/arch/arm/mach-at91/pm.c +++ b/arch/arm/mach-at91/pm.c @@ -334,16 +334,14 @@ static bool at91_pm_eth_quirk_is_valid(struct at91_pm_quirk_eth *eth) pdev = of_find_device_by_node(eth->np); if (!pdev) return false; + /* put_device(eth->dev) is called at the end of suspend. */ eth->dev = &pdev->dev; } /* No quirks if device isn't a wakeup source. */ - if (!device_may_wakeup(eth->dev)) { - put_device(eth->dev); + if (!device_may_wakeup(eth->dev)) return false; - } - /* put_device(eth->dev) is called at the end of suspend. */ return true; } @@ -439,14 +437,14 @@ static int at91_pm_config_quirks(bool suspend) pr_err("AT91: PM: failed to enable %s clocks\n", j == AT91_PM_G_ETH ? "geth" : "eth"); } - } else { - /* - * Release the reference to eth->dev taken in - * at91_pm_eth_quirk_is_valid(). - */ - put_device(eth->dev); - eth->dev = NULL; } + + /* + * Release the reference to eth->dev taken in + * at91_pm_eth_quirk_is_valid(). + */ + put_device(eth->dev); + eth->dev = NULL; } return ret; diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index b1201d25a8a4ee75e6eeaddb46f626655883e6bf..343e1e1cae10a3d51098b2ad7b9f1d457fcca2a5 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -1516,7 +1516,7 @@ config XEN # 16K | 27 | 14 | 13 | 11 | # 64K | 29 | 16 | 13 | 13 | config ARCH_FORCE_MAX_ORDER - int "Order of maximal physically contiguous allocations" if EXPERT && (ARM64_4K_PAGES || ARM64_16K_PAGES) + int default "13" if ARM64_64K_PAGES default "11" if ARM64_16K_PAGES default "10" diff --git a/arch/arm64/boot/dts/arm/foundation-v8.dtsi b/arch/arm64/boot/dts/arm/foundation-v8.dtsi index 029578072d8fb495ce64cc86ee59e0dc3c36ab08..7b41537731a6aec17fee92655a52ab132aaefd36 100644 --- a/arch/arm64/boot/dts/arm/foundation-v8.dtsi +++ b/arch/arm64/boot/dts/arm/foundation-v8.dtsi @@ -59,6 +59,7 @@ cpu3: cpu@3 { L2_0: l2-cache0 { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm64/boot/dts/arm/rtsm_ve-aemv8a.dts b/arch/arm64/boot/dts/arm/rtsm_ve-aemv8a.dts index ef68f5aae7ddf7b7ac1d886ca30de613d08aebd7..afdf954206f1d1a895b6c0106b0b7afd7d9053b8 100644 --- a/arch/arm64/boot/dts/arm/rtsm_ve-aemv8a.dts +++ b/arch/arm64/boot/dts/arm/rtsm_ve-aemv8a.dts @@ -72,6 +72,7 @@ cpu@3 { L2_0: l2-cache0 { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm64/boot/dts/arm/vexpress-v2f-1xv7-ca53x2.dts b/arch/arm64/boot/dts/arm/vexpress-v2f-1xv7-ca53x2.dts index 796cd7d02eb55ac954a7d828000c706651250130..7bdeb965f0a964f3dd77d9c6d39717b4b9b00201 100644 --- a/arch/arm64/boot/dts/arm/vexpress-v2f-1xv7-ca53x2.dts +++ b/arch/arm64/boot/dts/arm/vexpress-v2f-1xv7-ca53x2.dts @@ -58,6 +58,7 @@ cpu@1 { L2_0: l2-cache0 { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm64/boot/dts/freescale/imx8-ss-conn.dtsi b/arch/arm64/boot/dts/freescale/imx8-ss-conn.dtsi index 2209c1ac6e9b011726b31b4269deac85b6445a92..e62a43591361b360a64f11998bda741708bbc1f0 100644 --- a/arch/arm64/boot/dts/freescale/imx8-ss-conn.dtsi +++ b/arch/arm64/boot/dts/freescale/imx8-ss-conn.dtsi @@ -171,6 +171,7 @@ usbotg3_cdns3: usb@5b120000 { interrupt-names = "host", "peripheral", "otg", "wakeup"; phys = <&usb3_phy>; phy-names = "cdns3,usb3-phy"; + cdns,on-chip-buff-size = /bits/ 16 <18>; status = "disabled"; }; }; diff --git a/arch/arm64/boot/dts/freescale/imx8-ss-dma.dtsi b/arch/arm64/boot/dts/freescale/imx8-ss-dma.dtsi index 2dce8f2ee3ea1f9b9133d1258e2b1a486d6e3c83..adb98a72bdfd911166c0bfdc62a4675240db356e 100644 --- a/arch/arm64/boot/dts/freescale/imx8-ss-dma.dtsi +++ b/arch/arm64/boot/dts/freescale/imx8-ss-dma.dtsi @@ -90,6 +90,8 @@ lpuart0: serial@5a060000 { clocks = <&uart0_lpcg IMX_LPCG_CLK_4>, <&uart0_lpcg IMX_LPCG_CLK_0>; clock-names = "ipg", "baud"; + assigned-clocks = <&clk IMX_SC_R_UART_0 IMX_SC_PM_CLK_PER>; + assigned-clock-rates = <80000000>; power-domains = <&pd IMX_SC_R_UART_0>; status = "disabled"; }; @@ -100,6 +102,8 @@ lpuart1: serial@5a070000 { clocks = <&uart1_lpcg IMX_LPCG_CLK_4>, <&uart1_lpcg IMX_LPCG_CLK_0>; clock-names = "ipg", "baud"; + assigned-clocks = <&clk IMX_SC_R_UART_1 IMX_SC_PM_CLK_PER>; + assigned-clock-rates = <80000000>; power-domains = <&pd IMX_SC_R_UART_1>; status = "disabled"; }; @@ -110,6 +114,8 @@ lpuart2: serial@5a080000 { clocks = <&uart2_lpcg IMX_LPCG_CLK_4>, <&uart2_lpcg IMX_LPCG_CLK_0>; clock-names = "ipg", "baud"; + assigned-clocks = <&clk IMX_SC_R_UART_2 IMX_SC_PM_CLK_PER>; + assigned-clock-rates = <80000000>; power-domains = <&pd IMX_SC_R_UART_2>; status = "disabled"; }; @@ -120,6 +126,8 @@ lpuart3: serial@5a090000 { clocks = <&uart3_lpcg IMX_LPCG_CLK_4>, <&uart3_lpcg IMX_LPCG_CLK_0>; clock-names = "ipg", "baud"; + assigned-clocks = <&clk IMX_SC_R_UART_3 IMX_SC_PM_CLK_PER>; + assigned-clock-rates = <80000000>; power-domains = <&pd IMX_SC_R_UART_3>; status = "disabled"; }; diff --git a/arch/arm64/boot/dts/freescale/imx8mn-beacon-baseboard.dtsi b/arch/arm64/boot/dts/freescale/imx8mn-beacon-baseboard.dtsi index 9e82069c941fa6ac78f6daeb4168e1e6e001813b..5a1f7c30afe57a09f16c658fc50ef228c3b86bd6 100644 --- a/arch/arm64/boot/dts/freescale/imx8mn-beacon-baseboard.dtsi +++ b/arch/arm64/boot/dts/freescale/imx8mn-beacon-baseboard.dtsi @@ -81,7 +81,7 @@ sound { &ecspi2 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_espi2>; - cs-gpios = <&gpio5 9 GPIO_ACTIVE_LOW>; + cs-gpios = <&gpio5 13 GPIO_ACTIVE_LOW>; status = "okay"; eeprom@0 { @@ -202,7 +202,7 @@ pinctrl_espi2: espi2grp { MX8MN_IOMUXC_ECSPI2_SCLK_ECSPI2_SCLK 0x82 MX8MN_IOMUXC_ECSPI2_MOSI_ECSPI2_MOSI 0x82 MX8MN_IOMUXC_ECSPI2_MISO_ECSPI2_MISO 0x82 - MX8MN_IOMUXC_ECSPI1_SS0_GPIO5_IO9 0x41 + MX8MN_IOMUXC_ECSPI2_SS0_GPIO5_IO13 0x41 >; }; diff --git a/arch/arm64/boot/dts/freescale/imx8mn-var-som.dtsi b/arch/arm64/boot/dts/freescale/imx8mn-var-som.dtsi index 67072e6c77d5fb7834b41a938881216d87939496..cbd9d124c80d09adcdc630a8cf4710e89201aa50 100644 --- a/arch/arm64/boot/dts/freescale/imx8mn-var-som.dtsi +++ b/arch/arm64/boot/dts/freescale/imx8mn-var-som.dtsi @@ -98,11 +98,17 @@ mdio { #address-cells = <1>; #size-cells = <0>; - ethphy: ethernet-phy@4 { + ethphy: ethernet-phy@4 { /* AR8033 or ADIN1300 */ compatible = "ethernet-phy-ieee802.3-c22"; reg = <4>; reset-gpios = <&gpio1 9 GPIO_ACTIVE_LOW>; reset-assert-us = <10000>; + /* + * Deassert delay: + * ADIN1300 requires 5ms. + * AR8033 requires 1ms. + */ + reset-deassert-us = <20000>; }; }; }; diff --git a/arch/arm64/boot/dts/freescale/imx8mn.dtsi b/arch/arm64/boot/dts/freescale/imx8mn.dtsi index bd84db55005313ac4e6c96b2929cf3b684e95c67..8be8f090e8b8e9e7c89a998463b23b067e40ff30 100644 --- a/arch/arm64/boot/dts/freescale/imx8mn.dtsi +++ b/arch/arm64/boot/dts/freescale/imx8mn.dtsi @@ -1069,13 +1069,6 @@ lcdif: lcdif@32e00000 { <&clk IMX8MN_CLK_DISP_APB_ROOT>, <&clk IMX8MN_CLK_DISP_AXI_ROOT>; clock-names = "pix", "axi", "disp_axi"; - assigned-clocks = <&clk IMX8MN_CLK_DISP_PIXEL_ROOT>, - <&clk IMX8MN_CLK_DISP_AXI>, - <&clk IMX8MN_CLK_DISP_APB>; - assigned-clock-parents = <&clk IMX8MN_CLK_DISP_PIXEL>, - <&clk IMX8MN_SYS_PLL2_1000M>, - <&clk IMX8MN_SYS_PLL1_800M>; - assigned-clock-rates = <594000000>, <500000000>, <200000000>; interrupts = ; power-domains = <&disp_blk_ctrl IMX8MN_DISPBLK_PD_LCDIF>; status = "disabled"; @@ -1093,12 +1086,6 @@ mipi_dsi: dsi@32e10000 { clocks = <&clk IMX8MN_CLK_DSI_CORE>, <&clk IMX8MN_CLK_DSI_PHY_REF>; clock-names = "bus_clk", "sclk_mipi"; - assigned-clocks = <&clk IMX8MN_CLK_DSI_CORE>, - <&clk IMX8MN_CLK_DSI_PHY_REF>; - assigned-clock-parents = <&clk IMX8MN_SYS_PLL1_266M>, - <&clk IMX8MN_CLK_24M>; - assigned-clock-rates = <266000000>, <24000000>; - samsung,pll-clock-frequency = <24000000>; interrupts = ; power-domains = <&disp_blk_ctrl IMX8MN_DISPBLK_PD_MIPI_DSI>; status = "disabled"; @@ -1142,6 +1129,21 @@ disp_blk_ctrl: blk-ctrl@32e28000 { "lcdif-axi", "lcdif-apb", "lcdif-pix", "dsi-pclk", "dsi-ref", "csi-aclk", "csi-pclk"; + assigned-clocks = <&clk IMX8MN_CLK_DSI_CORE>, + <&clk IMX8MN_CLK_DSI_PHY_REF>, + <&clk IMX8MN_CLK_DISP_PIXEL>, + <&clk IMX8MN_CLK_DISP_AXI>, + <&clk IMX8MN_CLK_DISP_APB>; + assigned-clock-parents = <&clk IMX8MN_SYS_PLL1_266M>, + <&clk IMX8MN_CLK_24M>, + <&clk IMX8MN_VIDEO_PLL1_OUT>, + <&clk IMX8MN_SYS_PLL2_1000M>, + <&clk IMX8MN_SYS_PLL1_800M>; + assigned-clock-rates = <266000000>, + <24000000>, + <594000000>, + <500000000>, + <200000000>; #power-domain-cells = <1>; }; diff --git a/arch/arm64/boot/dts/freescale/imx8mp.dtsi b/arch/arm64/boot/dts/freescale/imx8mp.dtsi index f81391993354f484aa60380a4b23545b3cfb724e..428c60462e3d6be7c09f62ea51f48be069af802f 100644 --- a/arch/arm64/boot/dts/freescale/imx8mp.dtsi +++ b/arch/arm64/boot/dts/freescale/imx8mp.dtsi @@ -1211,13 +1211,6 @@ lcdif1: display-controller@32e80000 { <&clk IMX8MP_CLK_MEDIA_APB_ROOT>, <&clk IMX8MP_CLK_MEDIA_AXI_ROOT>; clock-names = "pix", "axi", "disp_axi"; - assigned-clocks = <&clk IMX8MP_CLK_MEDIA_DISP1_PIX_ROOT>, - <&clk IMX8MP_CLK_MEDIA_AXI>, - <&clk IMX8MP_CLK_MEDIA_APB>; - assigned-clock-parents = <&clk IMX8MP_CLK_MEDIA_DISP1_PIX>, - <&clk IMX8MP_SYS_PLL2_1000M>, - <&clk IMX8MP_SYS_PLL1_800M>; - assigned-clock-rates = <594000000>, <500000000>, <200000000>; interrupts = ; power-domains = <&media_blk_ctrl IMX8MP_MEDIABLK_PD_LCDIF_1>; status = "disabled"; @@ -1237,11 +1230,6 @@ lcdif2: display-controller@32e90000 { <&clk IMX8MP_CLK_MEDIA_APB_ROOT>, <&clk IMX8MP_CLK_MEDIA_AXI_ROOT>; clock-names = "pix", "axi", "disp_axi"; - assigned-clocks = <&clk IMX8MP_CLK_MEDIA_DISP2_PIX>, - <&clk IMX8MP_VIDEO_PLL1>; - assigned-clock-parents = <&clk IMX8MP_VIDEO_PLL1_OUT>, - <&clk IMX8MP_VIDEO_PLL1_REF_SEL>; - assigned-clock-rates = <0>, <1039500000>; power-domains = <&media_blk_ctrl IMX8MP_MEDIABLK_PD_LCDIF_2>; status = "disabled"; @@ -1296,11 +1284,16 @@ media_blk_ctrl: blk-ctrl@32ec0000 { "disp1", "disp2", "isp", "phy"; assigned-clocks = <&clk IMX8MP_CLK_MEDIA_AXI>, - <&clk IMX8MP_CLK_MEDIA_APB>; + <&clk IMX8MP_CLK_MEDIA_APB>, + <&clk IMX8MP_CLK_MEDIA_DISP1_PIX>, + <&clk IMX8MP_CLK_MEDIA_DISP2_PIX>, + <&clk IMX8MP_VIDEO_PLL1>; assigned-clock-parents = <&clk IMX8MP_SYS_PLL2_1000M>, - <&clk IMX8MP_SYS_PLL1_800M>; - assigned-clock-rates = <500000000>, <200000000>; - + <&clk IMX8MP_SYS_PLL1_800M>, + <&clk IMX8MP_VIDEO_PLL1_OUT>, + <&clk IMX8MP_VIDEO_PLL1_OUT>; + assigned-clock-rates = <500000000>, <200000000>, + <0>, <0>, <1039500000>; #power-domain-cells = <1>; lvds_bridge: bridge@5c { diff --git a/arch/arm64/boot/dts/freescale/imx8qm-mek.dts b/arch/arm64/boot/dts/freescale/imx8qm-mek.dts index ce9d3f0b98fc0383667fe8ffd60b239eaab348a2..607cd6b4e9721f04b849347d6464efc8ba240753 100644 --- a/arch/arm64/boot/dts/freescale/imx8qm-mek.dts +++ b/arch/arm64/boot/dts/freescale/imx8qm-mek.dts @@ -82,8 +82,8 @@ &usdhc2 { pinctrl-0 = <&pinctrl_usdhc2>; bus-width = <4>; vmmc-supply = <®_usdhc2_vmmc>; - cd-gpios = <&lsio_gpio4 22 GPIO_ACTIVE_LOW>; - wp-gpios = <&lsio_gpio4 21 GPIO_ACTIVE_HIGH>; + cd-gpios = <&lsio_gpio5 22 GPIO_ACTIVE_LOW>; + wp-gpios = <&lsio_gpio5 21 GPIO_ACTIVE_HIGH>; status = "okay"; }; diff --git a/arch/arm64/boot/dts/freescale/imx8x-colibri-eval-v3.dtsi b/arch/arm64/boot/dts/freescale/imx8x-colibri-eval-v3.dtsi index 7264d784ae723cfa977b0953b4fed03c7d679b35..9af769ab8cebbdab47accd034fdc331d36e694f9 100644 --- a/arch/arm64/boot/dts/freescale/imx8x-colibri-eval-v3.dtsi +++ b/arch/arm64/boot/dts/freescale/imx8x-colibri-eval-v3.dtsi @@ -33,6 +33,12 @@ rtc_i2c: rtc@68 { }; }; +&iomuxc { + pinctrl-names = "default"; + pinctrl-0 = <&pinctrl_ext_io0>, <&pinctrl_hog0>, <&pinctrl_hog1>, + <&pinctrl_lpspi2_cs2>; +}; + /* Colibri SPI */ &lpspi2 { status = "okay"; diff --git a/arch/arm64/boot/dts/freescale/imx8x-colibri-iris.dtsi b/arch/arm64/boot/dts/freescale/imx8x-colibri-iris.dtsi index 5f30c88855e70769ebded3a2716eafa37f232cf0..f8953067bc3bb86096c83365f3f3a58327aa365d 100644 --- a/arch/arm64/boot/dts/freescale/imx8x-colibri-iris.dtsi +++ b/arch/arm64/boot/dts/freescale/imx8x-colibri-iris.dtsi @@ -48,8 +48,7 @@ pinctrl_gpio_iris: gpioirisgrp { , /* SODIMM 101 */ , /* SODIMM 97 */ , /* SODIMM 85 */ - , /* SODIMM 79 */ - ; /* SODIMM 45 */ + ; /* SODIMM 79 */ }; pinctrl_uart1_forceoff: uart1forceoffgrp { diff --git a/arch/arm64/boot/dts/freescale/imx8x-colibri.dtsi b/arch/arm64/boot/dts/freescale/imx8x-colibri.dtsi index 7cad79102e1afec6b06bd91fae24408832e180fd..49d105eb4769197af85317886bbb4acd13ef3248 100644 --- a/arch/arm64/boot/dts/freescale/imx8x-colibri.dtsi +++ b/arch/arm64/boot/dts/freescale/imx8x-colibri.dtsi @@ -363,10 +363,6 @@ &usdhc2 { /* TODO VPU Encoder/Decoder */ &iomuxc { - pinctrl-names = "default"; - pinctrl-0 = <&pinctrl_ext_io0>, <&pinctrl_hog0>, <&pinctrl_hog1>, - <&pinctrl_hog2>, <&pinctrl_lpspi2_cs2>; - /* On-module touch pen-down interrupt */ pinctrl_ad7879_int: ad7879intgrp { fsl,pins = ; @@ -499,8 +495,7 @@ pinctrl_hog0: hog0grp { }; pinctrl_hog1: hog1grp { - fsl,pins = , /* SODIMM 75 */ - ; /* SODIMM 93 */ + fsl,pins = ; /* SODIMM 93 */ }; pinctrl_hog2: hog2grp { @@ -774,3 +769,10 @@ pinctrl_wifi: wifigrp { fsl,pins = ; }; }; + +/* Delete peripherals which are not present on SOC, but are defined in imx8-ss-*.dtsi */ + +/delete-node/ &adc1; +/delete-node/ &adc1_lpcg; +/delete-node/ &dsp; +/delete-node/ &dsp_lpcg; diff --git a/arch/arm64/boot/dts/qcom/ipq5332.dtsi b/arch/arm64/boot/dts/qcom/ipq5332.dtsi index 12e0e179e139cc9d506f0a938dc0808ba6d36aae..af4d97143bcf53e45f8d67b836d15cbf8e8f8d10 100644 --- a/arch/arm64/boot/dts/qcom/ipq5332.dtsi +++ b/arch/arm64/boot/dts/qcom/ipq5332.dtsi @@ -73,6 +73,7 @@ CPU3: cpu@3 { L2_0: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm64/boot/dts/qcom/ipq6018.dtsi b/arch/arm64/boot/dts/qcom/ipq6018.dtsi index 9ff4e9d45065bfb8a72dc04507c45e5ab796c36a..f531797f26195f027205594083690eb7afb8c104 100644 --- a/arch/arm64/boot/dts/qcom/ipq6018.dtsi +++ b/arch/arm64/boot/dts/qcom/ipq6018.dtsi @@ -83,7 +83,8 @@ CPU3: cpu@3 { L2_0: l2-cache { compatible = "cache"; - cache-level = <0x2>; + cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm64/boot/dts/qcom/ipq8074.dtsi b/arch/arm64/boot/dts/qcom/ipq8074.dtsi index 84e715aa4310332c7e0ed7e2570a674f0bfe6fd6..5b2c1986c8f4f3c8696edd240edd63d69bd00169 100644 --- a/arch/arm64/boot/dts/qcom/ipq8074.dtsi +++ b/arch/arm64/boot/dts/qcom/ipq8074.dtsi @@ -66,7 +66,8 @@ CPU3: cpu@3 { L2_0: l2-cache { compatible = "cache"; - cache-level = <0x2>; + cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm64/boot/dts/qcom/ipq9574.dtsi b/arch/arm64/boot/dts/qcom/ipq9574.dtsi index 3bb7435f5e7f08a05faad0c9b270f020798e8def..0ed19fbf7d87d1c7abdb8db7258b39d25660020b 100644 --- a/arch/arm64/boot/dts/qcom/ipq9574.dtsi +++ b/arch/arm64/boot/dts/qcom/ipq9574.dtsi @@ -72,6 +72,7 @@ CPU3: cpu@3 { L2_0: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm64/boot/dts/qcom/msm8916.dtsi b/arch/arm64/boot/dts/qcom/msm8916.dtsi index 7e0fa37a3adf8d6c30d97d1915a2c8782352a8b0..834e0b66b7f2e68a477a9443894e80e9a1002dad 100644 --- a/arch/arm64/boot/dts/qcom/msm8916.dtsi +++ b/arch/arm64/boot/dts/qcom/msm8916.dtsi @@ -180,6 +180,7 @@ CPU3: cpu@3 { L2_0: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; idle-states { diff --git a/arch/arm64/boot/dts/qcom/msm8953.dtsi b/arch/arm64/boot/dts/qcom/msm8953.dtsi index 602cb188a6354266d9108316aeccf697271f9ac4..d44cfa0471e9a6f0e37c2244d0ff349aba44ddb3 100644 --- a/arch/arm64/boot/dts/qcom/msm8953.dtsi +++ b/arch/arm64/boot/dts/qcom/msm8953.dtsi @@ -153,11 +153,13 @@ core3 { L2_0: l2-cache-0 { compatible = "cache"; cache-level = <2>; + cache-unified; }; L2_1: l2-cache-1 { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm64/boot/dts/qcom/msm8976.dtsi b/arch/arm64/boot/dts/qcom/msm8976.dtsi index 1f0bd24a074a681de0beb52756b5f28614dfab41..f47fb8ea71e20049d5ca78996c14e7a31626065a 100644 --- a/arch/arm64/boot/dts/qcom/msm8976.dtsi +++ b/arch/arm64/boot/dts/qcom/msm8976.dtsi @@ -193,11 +193,13 @@ big_cpu_sleep_1: cpu-sleep-1-1 { l2_0: l2-cache0 { compatible = "cache"; cache-level = <2>; + cache-unified; }; l2_1: l2-cache1 { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm64/boot/dts/qcom/msm8994.dtsi b/arch/arm64/boot/dts/qcom/msm8994.dtsi index 2831966be960de32a651d01615685049a76f1014..bdc3f2ba1755e48b3e1b81cbd9ee0e790cfe2314 100644 --- a/arch/arm64/boot/dts/qcom/msm8994.dtsi +++ b/arch/arm64/boot/dts/qcom/msm8994.dtsi @@ -52,6 +52,7 @@ CPU0: cpu@0 { L2_0: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; @@ -88,6 +89,7 @@ CPU4: cpu@100 { L2_1: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm64/boot/dts/qcom/msm8996.dtsi b/arch/arm64/boot/dts/qcom/msm8996.dtsi index 2b35cb3f52921662ca823d3be5dd52017f1884f1..30257c07e1279f71a2536aafb34573a245c32ed3 100644 --- a/arch/arm64/boot/dts/qcom/msm8996.dtsi +++ b/arch/arm64/boot/dts/qcom/msm8996.dtsi @@ -53,8 +53,9 @@ CPU0: cpu@0 { #cooling-cells = <2>; next-level-cache = <&L2_0>; L2_0: l2-cache { - compatible = "cache"; - cache-level = <2>; + compatible = "cache"; + cache-level = <2>; + cache-unified; }; }; @@ -83,8 +84,9 @@ CPU2: cpu@100 { #cooling-cells = <2>; next-level-cache = <&L2_1>; L2_1: l2-cache { - compatible = "cache"; - cache-level = <2>; + compatible = "cache"; + cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm64/boot/dts/qcom/msm8998.dtsi b/arch/arm64/boot/dts/qcom/msm8998.dtsi index b150437a83558c440e2745397d873fe01b3fc88c..3ec941fed14fedcd6c8c36a829e01d562394d6d6 100644 --- a/arch/arm64/boot/dts/qcom/msm8998.dtsi +++ b/arch/arm64/boot/dts/qcom/msm8998.dtsi @@ -146,6 +146,7 @@ CPU0: cpu@0 { L2_0: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; @@ -190,6 +191,7 @@ CPU4: cpu@100 { L2_1: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm64/boot/dts/qcom/qcm2290.dtsi b/arch/arm64/boot/dts/qcom/qcm2290.dtsi index ae5abc76bcc785ab63c8b803338ad5f06916b10b..b29bc4e4b837c6f083aa48c4e13dbbd2df8d4947 100644 --- a/arch/arm64/boot/dts/qcom/qcm2290.dtsi +++ b/arch/arm64/boot/dts/qcom/qcm2290.dtsi @@ -51,6 +51,7 @@ CPU0: cpu@0 { L2_0: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm64/boot/dts/qcom/qcs404.dtsi b/arch/arm64/boot/dts/qcom/qcs404.dtsi index eefed585738cd7e5b250cbd892c6f132f2930002..972f753847e13316a9437a65f781b1f33f8be84f 100644 --- a/arch/arm64/boot/dts/qcom/qcs404.dtsi +++ b/arch/arm64/boot/dts/qcom/qcs404.dtsi @@ -95,6 +95,7 @@ CPU3: cpu@103 { L2_0: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; idle-states { diff --git a/arch/arm64/boot/dts/qcom/qdu1000.dtsi b/arch/arm64/boot/dts/qcom/qdu1000.dtsi index 734438113bbae71a57472f1ed20023717bd6df6d..fb553f0bb17aae36b16b0c08dd5241409cffcab3 100644 --- a/arch/arm64/boot/dts/qcom/qdu1000.dtsi +++ b/arch/arm64/boot/dts/qcom/qdu1000.dtsi @@ -35,9 +35,13 @@ CPU0: cpu@0 { next-level-cache = <&L2_0>; L2_0: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; L3_0: l3-cache { compatible = "cache"; + cache-level = <3>; + cache-unified; }; }; }; @@ -54,6 +58,8 @@ CPU1: cpu@100 { next-level-cache = <&L2_100>; L2_100: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -70,6 +76,8 @@ CPU2: cpu@200 { next-level-cache = <&L2_200>; L2_200: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -86,6 +94,8 @@ CPU3: cpu@300 { next-level-cache = <&L2_300>; L2_300: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; diff --git a/arch/arm64/boot/dts/qcom/sa8155p-adp.dts b/arch/arm64/boot/dts/qcom/sa8155p-adp.dts index 339fea522509fae5987378c949e15871a1b88f41..15e1ae1c1a9774302bf44490e99cbb515cb61208 100644 --- a/arch/arm64/boot/dts/qcom/sa8155p-adp.dts +++ b/arch/arm64/boot/dts/qcom/sa8155p-adp.dts @@ -7,7 +7,7 @@ #include #include -#include "sm8150.dtsi" +#include "sa8155p.dtsi" #include "pmm8155au_1.dtsi" #include "pmm8155au_2.dtsi" diff --git a/arch/arm64/boot/dts/qcom/sa8155p.dtsi b/arch/arm64/boot/dts/qcom/sa8155p.dtsi new file mode 100644 index 0000000000000000000000000000000000000000..ffb7ab695213a6009294420b150e476f604d614a --- /dev/null +++ b/arch/arm64/boot/dts/qcom/sa8155p.dtsi @@ -0,0 +1,40 @@ +// SPDX-License-Identifier: BSD-3-Clause +/* + * Copyright (c) 2023, Linaro Limited + * + * SA8155P is an automotive variant of SM8150, with some minor changes. + * Most notably, the RPMhPD setup differs: MMCX and LCX/LMX rails are gone, + * though the cmd-db doesn't reflect that and access attemps result in a bite. + */ + +#include "sm8150.dtsi" + +&dispcc { + power-domains = <&rpmhpd SA8155P_CX>; +}; + +&mdss_dsi0 { + power-domains = <&rpmhpd SA8155P_CX>; +}; + +&mdss_dsi1 { + power-domains = <&rpmhpd SA8155P_CX>; +}; + +&mdss_mdp { + power-domains = <&rpmhpd SA8155P_CX>; +}; + +&remoteproc_slpi { + power-domains = <&rpmhpd SA8155P_CX>, + <&rpmhpd SA8155P_MX>; +}; + +&rpmhpd { + /* + * The bindings were crafted such that SA8155P PDs match their + * SM8150 counterparts to make it more maintainable and only + * necessitate adjusting entries that actually differ + */ + compatible = "qcom,sa8155p-rpmhpd"; +}; diff --git a/arch/arm64/boot/dts/qcom/sa8775p.dtsi b/arch/arm64/boot/dts/qcom/sa8775p.dtsi index 2343df7e0ea4f36961ae305ddbaedcdc5157a287..c3310caf9f68c6451692efc1fba96f9bfc4b08e7 100644 --- a/arch/arm64/boot/dts/qcom/sa8775p.dtsi +++ b/arch/arm64/boot/dts/qcom/sa8775p.dtsi @@ -42,9 +42,13 @@ CPU0: cpu@0 { next-level-cache = <&L2_0>; L2_0: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; L3_0: l3-cache { compatible = "cache"; + cache-level = <3>; + cache-unified; }; }; }; @@ -58,6 +62,8 @@ CPU1: cpu@100 { next-level-cache = <&L2_1>; L2_1: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -71,6 +77,8 @@ CPU2: cpu@200 { next-level-cache = <&L2_2>; L2_2: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -84,6 +92,8 @@ CPU3: cpu@300 { next-level-cache = <&L2_3>; L2_3: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -97,9 +107,13 @@ CPU4: cpu@10000 { next-level-cache = <&L2_4>; L2_4: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_1>; L3_1: l3-cache { compatible = "cache"; + cache-level = <3>; + cache-unified; }; }; @@ -114,6 +128,8 @@ CPU5: cpu@10100 { next-level-cache = <&L2_5>; L2_5: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_1>; }; }; @@ -127,6 +143,8 @@ CPU6: cpu@10200 { next-level-cache = <&L2_6>; L2_6: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_1>; }; }; @@ -140,6 +158,8 @@ CPU7: cpu@10300 { next-level-cache = <&L2_7>; L2_7: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_1>; }; }; diff --git a/arch/arm64/boot/dts/qcom/sc7180-lite.dtsi b/arch/arm64/boot/dts/qcom/sc7180-lite.dtsi index d8ed1d7b4ec76224d87cadbb104c06ebb7bd7869..4b306a59d9bec16c42e0bc80d6f5191f49eab7fa 100644 --- a/arch/arm64/boot/dts/qcom/sc7180-lite.dtsi +++ b/arch/arm64/boot/dts/qcom/sc7180-lite.dtsi @@ -16,3 +16,11 @@ &cpu6_opp11 { &cpu6_opp12 { opp-peak-kBps = <8532000 23347200>; }; + +&cpu6_opp13 { + opp-peak-kBps = <8532000 23347200>; +}; + +&cpu6_opp14 { + opp-peak-kBps = <8532000 23347200>; +}; diff --git a/arch/arm64/boot/dts/qcom/sc7180.dtsi b/arch/arm64/boot/dts/qcom/sc7180.dtsi index ea1ffade1aa194ba5896680ac13083b7fd3cf24e..f479cab8ab45c44a3f3a45617908884df4a4ea6f 100644 --- a/arch/arm64/boot/dts/qcom/sc7180.dtsi +++ b/arch/arm64/boot/dts/qcom/sc7180.dtsi @@ -92,10 +92,12 @@ &LITTLE_CPU_SLEEP_1 L2_0: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; L3_0: l3-cache { compatible = "cache"; cache-level = <3>; + cache-unified; }; }; }; @@ -120,6 +122,7 @@ &LITTLE_CPU_SLEEP_1 L2_100: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -144,6 +147,7 @@ &LITTLE_CPU_SLEEP_1 L2_200: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -168,6 +172,7 @@ &LITTLE_CPU_SLEEP_1 L2_300: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -192,6 +197,7 @@ &LITTLE_CPU_SLEEP_1 L2_400: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -216,6 +222,7 @@ &LITTLE_CPU_SLEEP_1 L2_500: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -240,6 +247,7 @@ &BIG_CPU_SLEEP_1 L2_600: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -264,6 +272,7 @@ &BIG_CPU_SLEEP_1 L2_700: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; diff --git a/arch/arm64/boot/dts/qcom/sc7280-idp.dtsi b/arch/arm64/boot/dts/qcom/sc7280-idp.dtsi index c6dc200c00cea42c891e130beed7fe74522e220f..21027042cf1332481ae5934f957297420749c012 100644 --- a/arch/arm64/boot/dts/qcom/sc7280-idp.dtsi +++ b/arch/arm64/boot/dts/qcom/sc7280-idp.dtsi @@ -480,7 +480,6 @@ &swr0 { wcd_rx: codec@0,4 { compatible = "sdw20217010d00"; reg = <0 4>; - #sound-dai-cells = <1>; qcom,rx-port-mapping = <1 2 3 4 5>; }; }; @@ -491,7 +490,6 @@ &swr1 { wcd_tx: codec@0,3 { compatible = "sdw20217010d00"; reg = <0 3>; - #sound-dai-cells = <1>; qcom,tx-port-mapping = <1 2 3 4>; }; }; diff --git a/arch/arm64/boot/dts/qcom/sc7280-qcard.dtsi b/arch/arm64/boot/dts/qcom/sc7280-qcard.dtsi index 88b3586e389fad78a68342d11a4853352486b55a..9137db066d9e2581197b25edb4e7cad483b37513 100644 --- a/arch/arm64/boot/dts/qcom/sc7280-qcard.dtsi +++ b/arch/arm64/boot/dts/qcom/sc7280-qcard.dtsi @@ -414,7 +414,6 @@ &swr0 { wcd_rx: codec@0,4 { compatible = "sdw20217010d00"; reg = <0 4>; - #sound-dai-cells = <1>; qcom,rx-port-mapping = <1 2 3 4 5>; }; }; @@ -423,7 +422,6 @@ &swr1 { wcd_tx: codec@0,3 { compatible = "sdw20217010d00"; reg = <0 3>; - #sound-dai-cells = <1>; qcom,tx-port-mapping = <1 2 3 4>; }; }; diff --git a/arch/arm64/boot/dts/qcom/sc7280.dtsi b/arch/arm64/boot/dts/qcom/sc7280.dtsi index 31728f4614223b1b9ce440f709d16609e624a42f..2fd1d3c0eb345b1302e6ca89f62af4b875360747 100644 --- a/arch/arm64/boot/dts/qcom/sc7280.dtsi +++ b/arch/arm64/boot/dts/qcom/sc7280.dtsi @@ -182,10 +182,12 @@ &LITTLE_CPU_SLEEP_1 L2_0: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; L3_0: l3-cache { compatible = "cache"; cache-level = <3>; + cache-unified; }; }; }; @@ -208,6 +210,7 @@ &LITTLE_CPU_SLEEP_1 L2_100: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -230,6 +233,7 @@ &LITTLE_CPU_SLEEP_1 L2_200: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -252,6 +256,7 @@ &LITTLE_CPU_SLEEP_1 L2_300: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -274,6 +279,7 @@ &BIG_CPU_SLEEP_1 L2_400: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -296,6 +302,7 @@ &BIG_CPU_SLEEP_1 L2_500: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -318,6 +325,7 @@ &BIG_CPU_SLEEP_1 L2_600: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -340,6 +348,7 @@ &BIG_CPU_SLEEP_1 L2_700: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; diff --git a/arch/arm64/boot/dts/qcom/sc8280xp.dtsi b/arch/arm64/boot/dts/qcom/sc8280xp.dtsi index 8fa9fbfe5d0009b44b40e02f16e980dd848a8a2b..cc4aef21e6172e6b7399da43b09cd1896daa62e6 100644 --- a/arch/arm64/boot/dts/qcom/sc8280xp.dtsi +++ b/arch/arm64/boot/dts/qcom/sc8280xp.dtsi @@ -58,10 +58,12 @@ CPU0: cpu@0 { L2_0: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; L3_0: l3-cache { - compatible = "cache"; - cache-level = <3>; + compatible = "cache"; + cache-level = <3>; + cache-unified; }; }; }; @@ -83,6 +85,7 @@ CPU1: cpu@100 { L2_100: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -104,6 +107,7 @@ CPU2: cpu@200 { L2_200: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -125,6 +129,7 @@ CPU3: cpu@300 { L2_300: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -146,6 +151,7 @@ CPU4: cpu@400 { L2_400: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -167,6 +173,7 @@ CPU5: cpu@500 { L2_500: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -188,6 +195,7 @@ CPU6: cpu@600 { L2_600: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -209,6 +217,7 @@ CPU7: cpu@700 { L2_700: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -2726,6 +2735,7 @@ data-pins { pins = "gpio7"; function = "dmic1_data"; drive-strength = <8>; + input-enable; }; }; @@ -2743,6 +2753,7 @@ data-pins { function = "dmic1_data"; drive-strength = <2>; bias-pull-down; + input-enable; }; }; @@ -2758,6 +2769,7 @@ data-pins { pins = "gpio9"; function = "dmic2_data"; drive-strength = <8>; + input-enable; }; }; @@ -2775,6 +2787,7 @@ data-pins { function = "dmic2_data"; drive-strength = <2>; bias-pull-down; + input-enable; }; }; @@ -3982,6 +3995,7 @@ apps_rsc: rsc@18200000 { qcom,tcs-config = , , , ; label = "apps_rsc"; + power-domains = <&CLUSTER_PD>; apps_bcm_voter: bcm-voter { compatible = "qcom,bcm-voter"; diff --git a/arch/arm64/boot/dts/qcom/sdm630.dtsi b/arch/arm64/boot/dts/qcom/sdm630.dtsi index 37e72b1c56dce7430aeb63f481afb22a15e94f4b..eaead2f7beb4ea93a3e2c498c4455fc13fd0558c 100644 --- a/arch/arm64/boot/dts/qcom/sdm630.dtsi +++ b/arch/arm64/boot/dts/qcom/sdm630.dtsi @@ -63,6 +63,7 @@ &PERF_CLUSTER_SLEEP_1 L2_1: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; @@ -127,6 +128,7 @@ &PWR_CLUSTER_SLEEP_1 L2_0: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm64/boot/dts/qcom/sdm670.dtsi b/arch/arm64/boot/dts/qcom/sdm670.dtsi index c5f839dd1c6e3ab3fe9e44ba90da467db99f8e8a..b61e13db89bd58fcfb012d0c0d8e5f1e8ff7fb8a 100644 --- a/arch/arm64/boot/dts/qcom/sdm670.dtsi +++ b/arch/arm64/boot/dts/qcom/sdm670.dtsi @@ -41,8 +41,12 @@ CPU0: cpu@0 { L2_0: l2-cache { compatible = "cache"; next-level-cache = <&L3_0>; + cache-level = <2>; + cache-unified; L3_0: l3-cache { - compatible = "cache"; + compatible = "cache"; + cache-level = <3>; + cache-unified; }; }; }; @@ -57,6 +61,8 @@ CPU1: cpu@100 { next-level-cache = <&L2_100>; L2_100: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -71,6 +77,8 @@ CPU2: cpu@200 { next-level-cache = <&L2_200>; L2_200: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -85,6 +93,8 @@ CPU3: cpu@300 { next-level-cache = <&L2_300>; L2_300: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -99,6 +109,8 @@ CPU4: cpu@400 { next-level-cache = <&L2_400>; L2_400: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -113,6 +125,8 @@ CPU5: cpu@500 { next-level-cache = <&L2_500>; L2_500: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -127,6 +141,8 @@ CPU6: cpu@600 { next-level-cache = <&L2_600>; L2_600: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -141,6 +157,8 @@ CPU7: cpu@700 { next-level-cache = <&L2_700>; L2_700: l2-cache { compatible = "cache"; + cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; diff --git a/arch/arm64/boot/dts/qcom/sdm845.dtsi b/arch/arm64/boot/dts/qcom/sdm845.dtsi index 90424442bb4acab8149488c0f679032389361572..cdeb05e95674e9d490b37b26e008f98599b8476e 100644 --- a/arch/arm64/boot/dts/qcom/sdm845.dtsi +++ b/arch/arm64/boot/dts/qcom/sdm845.dtsi @@ -108,10 +108,12 @@ CPU0: cpu@0 { L2_0: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; L3_0: l3-cache { - compatible = "cache"; - cache-level = <3>; + compatible = "cache"; + cache-level = <3>; + cache-unified; }; }; }; @@ -135,6 +137,7 @@ CPU1: cpu@100 { L2_100: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -158,6 +161,7 @@ CPU2: cpu@200 { L2_200: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -181,6 +185,7 @@ CPU3: cpu@300 { L2_300: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -204,6 +209,7 @@ CPU4: cpu@400 { L2_400: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -227,6 +233,7 @@ CPU5: cpu@500 { L2_500: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -250,6 +257,7 @@ CPU6: cpu@600 { L2_600: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -273,6 +281,7 @@ CPU7: cpu@700 { L2_700: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; diff --git a/arch/arm64/boot/dts/qcom/sm6115.dtsi b/arch/arm64/boot/dts/qcom/sm6115.dtsi index 631ca327e064bc6c0a9269a2b3eaaf5170d43f34..43f31c1b9d5a74952bd0b324553506c601b76e3e 100644 --- a/arch/arm64/boot/dts/qcom/sm6115.dtsi +++ b/arch/arm64/boot/dts/qcom/sm6115.dtsi @@ -50,6 +50,7 @@ CPU0: cpu@0 { L2_0: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; @@ -102,6 +103,7 @@ CPU4: cpu@100 { L2_1: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm64/boot/dts/qcom/sm6125.dtsi b/arch/arm64/boot/dts/qcom/sm6125.dtsi index 9484752fb85066e373a86afea0c11ce3844ebe63..2aa093d16858a68e5c4aa1f1a663f850d2182930 100644 --- a/arch/arm64/boot/dts/qcom/sm6125.dtsi +++ b/arch/arm64/boot/dts/qcom/sm6125.dtsi @@ -47,6 +47,7 @@ CPU0: cpu@0 { L2_0: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; @@ -87,6 +88,7 @@ CPU4: cpu@100 { L2_1: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; }; }; diff --git a/arch/arm64/boot/dts/qcom/sm6350.dtsi b/arch/arm64/boot/dts/qcom/sm6350.dtsi index 18c4616848cefcb58ef8ab38fa0e2ee68e2294ef..ad34301f6cddf9fd62f1297e28d377b6bc398f1c 100644 --- a/arch/arm64/boot/dts/qcom/sm6350.dtsi +++ b/arch/arm64/boot/dts/qcom/sm6350.dtsi @@ -60,10 +60,12 @@ &clk_virt SLAVE_EBI_CH0 QCOM_ICC_TAG_ACTIVE_ONLY>, L2_0: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; L3_0: l3-cache { compatible = "cache"; cache-level = <3>; + cache-unified; }; }; }; @@ -86,6 +88,7 @@ &clk_virt SLAVE_EBI_CH0 QCOM_ICC_TAG_ACTIVE_ONLY>, L2_100: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -108,6 +111,7 @@ &clk_virt SLAVE_EBI_CH0 QCOM_ICC_TAG_ACTIVE_ONLY>, L2_200: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -130,6 +134,7 @@ &clk_virt SLAVE_EBI_CH0 QCOM_ICC_TAG_ACTIVE_ONLY>, L2_300: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -152,6 +157,7 @@ &clk_virt SLAVE_EBI_CH0 QCOM_ICC_TAG_ACTIVE_ONLY>, L2_400: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -174,6 +180,7 @@ &clk_virt SLAVE_EBI_CH0 QCOM_ICC_TAG_ACTIVE_ONLY>, L2_500: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -196,6 +203,7 @@ &clk_virt SLAVE_EBI_CH0 QCOM_ICC_TAG_ACTIVE_ONLY>, L2_600: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -218,6 +226,7 @@ &clk_virt SLAVE_EBI_CH0 QCOM_ICC_TAG_ACTIVE_ONLY>, L2_700: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; diff --git a/arch/arm64/boot/dts/qcom/sm6375-sony-xperia-murray-pdx225.dts b/arch/arm64/boot/dts/qcom/sm6375-sony-xperia-murray-pdx225.dts index 8220e6f441171166abff3656e3a31642ce8f145f..b2f1bb1d58e974080fc1d23f799a15aef0bdf8a6 100644 --- a/arch/arm64/boot/dts/qcom/sm6375-sony-xperia-murray-pdx225.dts +++ b/arch/arm64/boot/dts/qcom/sm6375-sony-xperia-murray-pdx225.dts @@ -178,12 +178,12 @@ &qupv3_id_1 { }; &remoteproc_adsp { - firmware-name = "qcom/Sony/murray/adsp.mbn"; + firmware-name = "qcom/sm6375/Sony/murray/adsp.mbn"; status = "okay"; }; &remoteproc_cdsp { - firmware-name = "qcom/Sony/murray/cdsp.mbn"; + firmware-name = "qcom/sm6375/Sony/murray/cdsp.mbn"; status = "okay"; }; diff --git a/arch/arm64/boot/dts/qcom/sm6375.dtsi b/arch/arm64/boot/dts/qcom/sm6375.dtsi index ae9b6bc446cbc1ae97a6d32919f17d8d03d381ee..f8d9c34d3b2f4f55e11cf3ba4beaa4ad0e0ac5b7 100644 --- a/arch/arm64/boot/dts/qcom/sm6375.dtsi +++ b/arch/arm64/boot/dts/qcom/sm6375.dtsi @@ -48,10 +48,14 @@ CPU0: cpu@0 { power-domain-names = "psci"; #cooling-cells = <2>; L2_0: l2-cache { - compatible = "cache"; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; L3_0: l3-cache { - compatible = "cache"; + compatible = "cache"; + cache-level = <3>; + cache-unified; }; }; }; @@ -68,8 +72,10 @@ CPU1: cpu@100 { power-domain-names = "psci"; #cooling-cells = <2>; L2_100: l2-cache { - compatible = "cache"; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -85,8 +91,10 @@ CPU2: cpu@200 { power-domain-names = "psci"; #cooling-cells = <2>; L2_200: l2-cache { - compatible = "cache"; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -102,8 +110,10 @@ CPU3: cpu@300 { power-domain-names = "psci"; #cooling-cells = <2>; L2_300: l2-cache { - compatible = "cache"; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -119,8 +129,10 @@ CPU4: cpu@400 { power-domain-names = "psci"; #cooling-cells = <2>; L2_400: l2-cache { - compatible = "cache"; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -136,8 +148,10 @@ CPU5: cpu@500 { power-domain-names = "psci"; #cooling-cells = <2>; L2_500: l2-cache { - compatible = "cache"; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -153,8 +167,10 @@ CPU6: cpu@600 { power-domain-names = "psci"; #cooling-cells = <2>; L2_600: l2-cache { - compatible = "cache"; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -170,8 +186,10 @@ CPU7: cpu@700 { power-domain-names = "psci"; #cooling-cells = <2>; L2_700: l2-cache { - compatible = "cache"; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; diff --git a/arch/arm64/boot/dts/qcom/sm8150.dtsi b/arch/arm64/boot/dts/qcom/sm8150.dtsi index 2273fa57198842ef4c700ea464bd24ed9fbde9bb..27dcda0d4288fcedcfff7b633761fc6d9ef567b8 100644 --- a/arch/arm64/boot/dts/qcom/sm8150.dtsi +++ b/arch/arm64/boot/dts/qcom/sm8150.dtsi @@ -63,10 +63,12 @@ CPU0: cpu@0 { L2_0: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; L3_0: l3-cache { - compatible = "cache"; - cache-level = <3>; + compatible = "cache"; + cache-level = <3>; + cache-unified; }; }; }; @@ -90,6 +92,7 @@ CPU1: cpu@100 { L2_100: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -113,6 +116,7 @@ CPU2: cpu@200 { L2_200: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -136,6 +140,7 @@ CPU3: cpu@300 { L2_300: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -159,6 +164,7 @@ CPU4: cpu@400 { L2_400: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -182,6 +188,7 @@ CPU5: cpu@500 { L2_500: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -205,6 +212,7 @@ CPU6: cpu@600 { L2_600: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -228,6 +236,7 @@ CPU7: cpu@700 { L2_700: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; diff --git a/arch/arm64/boot/dts/qcom/sm8250-xiaomi-elish-boe.dts b/arch/arm64/boot/dts/qcom/sm8250-xiaomi-elish-boe.dts index 8b2ae39950ffa75a60cde3e9c313fad34f980605..de6101ddebe7d552fed84a8e0526458158c46850 100644 --- a/arch/arm64/boot/dts/qcom/sm8250-xiaomi-elish-boe.dts +++ b/arch/arm64/boot/dts/qcom/sm8250-xiaomi-elish-boe.dts @@ -13,6 +13,6 @@ / { }; &display_panel { - compatible = "xiaomi,elish-boe-nt36523"; + compatible = "xiaomi,elish-boe-nt36523", "novatek,nt36523"; status = "okay"; }; diff --git a/arch/arm64/boot/dts/qcom/sm8250-xiaomi-elish-csot.dts b/arch/arm64/boot/dts/qcom/sm8250-xiaomi-elish-csot.dts index a4d5341495cf8e4746d57eda194c20ea9d9c96e8..4cffe9c703df872e56dea51f0629d5f8f2ae04e4 100644 --- a/arch/arm64/boot/dts/qcom/sm8250-xiaomi-elish-csot.dts +++ b/arch/arm64/boot/dts/qcom/sm8250-xiaomi-elish-csot.dts @@ -13,6 +13,6 @@ / { }; &display_panel { - compatible = "xiaomi,elish-csot-nt36523"; + compatible = "xiaomi,elish-csot-nt36523", "novatek,nt36523"; status = "okay"; }; diff --git a/arch/arm64/boot/dts/qcom/sm8350.dtsi b/arch/arm64/boot/dts/qcom/sm8350.dtsi index ebcb481571c28267f0938b6b97394140fbf5a36f..3efdc03ed0f1107860f6ef8aa95bcf6a7e31b4fc 100644 --- a/arch/arm64/boot/dts/qcom/sm8350.dtsi +++ b/arch/arm64/boot/dts/qcom/sm8350.dtsi @@ -58,12 +58,14 @@ CPU0: cpu@0 { power-domain-names = "psci"; #cooling-cells = <2>; L2_0: l2-cache { - compatible = "cache"; - cache-level = <2>; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; L3_0: l3-cache { - compatible = "cache"; - cache-level = <3>; + compatible = "cache"; + cache-level = <3>; + cache-unified; }; }; }; @@ -80,9 +82,10 @@ CPU1: cpu@100 { power-domain-names = "psci"; #cooling-cells = <2>; L2_100: l2-cache { - compatible = "cache"; - cache-level = <2>; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -98,9 +101,10 @@ CPU2: cpu@200 { power-domain-names = "psci"; #cooling-cells = <2>; L2_200: l2-cache { - compatible = "cache"; - cache-level = <2>; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -116,9 +120,10 @@ CPU3: cpu@300 { power-domain-names = "psci"; #cooling-cells = <2>; L2_300: l2-cache { - compatible = "cache"; - cache-level = <2>; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -134,9 +139,10 @@ CPU4: cpu@400 { power-domain-names = "psci"; #cooling-cells = <2>; L2_400: l2-cache { - compatible = "cache"; - cache-level = <2>; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -152,9 +158,10 @@ CPU5: cpu@500 { power-domain-names = "psci"; #cooling-cells = <2>; L2_500: l2-cache { - compatible = "cache"; - cache-level = <2>; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -170,9 +177,10 @@ CPU6: cpu@600 { power-domain-names = "psci"; #cooling-cells = <2>; L2_600: l2-cache { - compatible = "cache"; - cache-level = <2>; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -188,9 +196,10 @@ CPU7: cpu@700 { power-domain-names = "psci"; #cooling-cells = <2>; L2_700: l2-cache { - compatible = "cache"; - cache-level = <2>; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; diff --git a/arch/arm64/boot/dts/qcom/sm8450.dtsi b/arch/arm64/boot/dts/qcom/sm8450.dtsi index 595533aeafc406185ab470f3d740cdf6ccb263b2..d59ea8ee711143ad7aae301c3317d6b2bfb6f14a 100644 --- a/arch/arm64/boot/dts/qcom/sm8450.dtsi +++ b/arch/arm64/boot/dts/qcom/sm8450.dtsi @@ -57,12 +57,14 @@ CPU0: cpu@0 { #cooling-cells = <2>; clocks = <&cpufreq_hw 0>; L2_0: l2-cache { - compatible = "cache"; - cache-level = <2>; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; L3_0: l3-cache { - compatible = "cache"; - cache-level = <3>; + compatible = "cache"; + cache-level = <3>; + cache-unified; }; }; }; @@ -79,9 +81,10 @@ CPU1: cpu@100 { #cooling-cells = <2>; clocks = <&cpufreq_hw 0>; L2_100: l2-cache { - compatible = "cache"; - cache-level = <2>; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -97,9 +100,10 @@ CPU2: cpu@200 { #cooling-cells = <2>; clocks = <&cpufreq_hw 0>; L2_200: l2-cache { - compatible = "cache"; - cache-level = <2>; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -115,9 +119,10 @@ CPU3: cpu@300 { #cooling-cells = <2>; clocks = <&cpufreq_hw 0>; L2_300: l2-cache { - compatible = "cache"; - cache-level = <2>; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -133,9 +138,10 @@ CPU4: cpu@400 { #cooling-cells = <2>; clocks = <&cpufreq_hw 1>; L2_400: l2-cache { - compatible = "cache"; - cache-level = <2>; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -151,9 +157,10 @@ CPU5: cpu@500 { #cooling-cells = <2>; clocks = <&cpufreq_hw 1>; L2_500: l2-cache { - compatible = "cache"; - cache-level = <2>; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -169,9 +176,10 @@ CPU6: cpu@600 { #cooling-cells = <2>; clocks = <&cpufreq_hw 1>; L2_600: l2-cache { - compatible = "cache"; - cache-level = <2>; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; @@ -187,9 +195,10 @@ CPU7: cpu@700 { #cooling-cells = <2>; clocks = <&cpufreq_hw 2>; L2_700: l2-cache { - compatible = "cache"; - cache-level = <2>; - next-level-cache = <&L3_0>; + compatible = "cache"; + cache-level = <2>; + cache-unified; + next-level-cache = <&L3_0>; }; }; diff --git a/arch/arm64/boot/dts/qcom/sm8550.dtsi b/arch/arm64/boot/dts/qcom/sm8550.dtsi index 6e9bad8f6f33ebe74cfab95e457b18bd9af45fc2..558cbc4307080407e63a63533ed634b31c12bec6 100644 --- a/arch/arm64/boot/dts/qcom/sm8550.dtsi +++ b/arch/arm64/boot/dts/qcom/sm8550.dtsi @@ -80,10 +80,12 @@ CPU0: cpu@0 { L2_0: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; L3_0: l3-cache { compatible = "cache"; cache-level = <3>; + cache-unified; }; }; }; @@ -104,6 +106,7 @@ CPU1: cpu@100 { L2_100: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -124,6 +127,7 @@ CPU2: cpu@200 { L2_200: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -144,6 +148,7 @@ CPU3: cpu@300 { L2_300: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -164,6 +169,7 @@ CPU4: cpu@400 { L2_400: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -184,6 +190,7 @@ CPU5: cpu@500 { L2_500: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -204,6 +211,7 @@ CPU6: cpu@600 { L2_600: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -224,6 +232,7 @@ CPU7: cpu@700 { L2_700: l2-cache { compatible = "cache"; cache-level = <2>; + cache-unified; next-level-cache = <&L3_0>; }; }; @@ -2022,7 +2031,7 @@ swr3: soundwire-controller@6ab0000 { qcom,din-ports = <4>; qcom,dout-ports = <9>; - qcom,ports-sinterval = <0x07 0x1f 0x3f 0x07 0x1f 0x3f 0x18f 0xff 0xff 0x0f 0x0f 0xff 0x31f>; + qcom,ports-sinterval = /bits/ 16 <0x07 0x1f 0x3f 0x07 0x1f 0x3f 0x18f 0xff 0xff 0x0f 0x0f 0xff 0x31f>; qcom,ports-offset1 = /bits/ 8 <0x01 0x03 0x05 0x02 0x04 0x15 0x00 0xff 0xff 0x06 0x0d 0xff 0x00>; qcom,ports-offset2 = /bits/ 8 <0xff 0x07 0x1f 0xff 0x07 0x1f 0xff 0xff 0xff 0xff 0xff 0xff 0xff>; qcom,ports-hstart = /bits/ 8 <0xff 0xff 0xff 0xff 0xff 0xff 0x08 0xff 0xff 0xff 0xff 0xff 0x0f>; @@ -2068,7 +2077,7 @@ swr1: soundwire-controller@6ad0000 { qcom,din-ports = <0>; qcom,dout-ports = <10>; - qcom,ports-sinterval = <0x03 0x3f 0x1f 0x07 0x00 0x18f 0xff 0xff 0xff 0xff>; + qcom,ports-sinterval = /bits/ 16 <0x03 0x3f 0x1f 0x07 0x00 0x18f 0xff 0xff 0xff 0xff>; qcom,ports-offset1 = /bits/ 8 <0x00 0x00 0x0b 0x01 0x00 0x00 0xff 0xff 0xff 0xff>; qcom,ports-offset2 = /bits/ 8 <0x00 0x00 0x0b 0x00 0x00 0x00 0xff 0xff 0xff 0xff>; qcom,ports-hstart = /bits/ 8 <0xff 0x03 0xff 0xff 0xff 0x08 0xff 0xff 0xff 0xff>; @@ -2133,7 +2142,7 @@ swr0: soundwire-controller@6b10000 { qcom,din-ports = <4>; qcom,dout-ports = <9>; - qcom,ports-sinterval = <0x07 0x1f 0x3f 0x07 0x1f 0x3f 0x18f 0xff 0xff 0x0f 0x0f 0xff 0x31f>; + qcom,ports-sinterval = /bits/ 16 <0x07 0x1f 0x3f 0x07 0x1f 0x3f 0x18f 0xff 0xff 0x0f 0x0f 0xff 0x31f>; qcom,ports-offset1 = /bits/ 8 <0x01 0x03 0x05 0x02 0x04 0x15 0x00 0xff 0xff 0x06 0x0d 0xff 0x00>; qcom,ports-offset2 = /bits/ 8 <0xff 0x07 0x1f 0xff 0x07 0x1f 0xff 0xff 0xff 0xff 0xff 0xff 0xff>; qcom,ports-hstart = /bits/ 8 <0xff 0xff 0xff 0xff 0xff 0xff 0x08 0xff 0xff 0xff 0xff 0xff 0x0f>; @@ -3762,9 +3771,16 @@ gem_noc: interconnect@24100000 { system-cache-controller@25000000 { compatible = "qcom,sm8550-llcc"; - reg = <0 0x25000000 0 0x800000>, + reg = <0 0x25000000 0 0x200000>, + <0 0x25200000 0 0x200000>, + <0 0x25400000 0 0x200000>, + <0 0x25600000 0 0x200000>, <0 0x25800000 0 0x200000>; - reg-names = "llcc_base", "llcc_broadcast_base"; + reg-names = "llcc0_base", + "llcc1_base", + "llcc2_base", + "llcc3_base", + "llcc_broadcast_base"; interrupts = ; }; diff --git a/arch/arm64/include/asm/arm_pmuv3.h b/arch/arm64/include/asm/arm_pmuv3.h index d6b51deb7bf0ff2f28810cada684097be2666962..18dc2fb3d7b7b2d0633bbe2f4ef18ec2e8ce4f67 100644 --- a/arch/arm64/include/asm/arm_pmuv3.h +++ b/arch/arm64/include/asm/arm_pmuv3.h @@ -13,7 +13,7 @@ #define RETURN_READ_PMEVCNTRN(n) \ return read_sysreg(pmevcntr##n##_el0) -static unsigned long read_pmevcntrn(int n) +static inline unsigned long read_pmevcntrn(int n) { PMEVN_SWITCH(n, RETURN_READ_PMEVCNTRN); return 0; @@ -21,14 +21,14 @@ static unsigned long read_pmevcntrn(int n) #define WRITE_PMEVCNTRN(n) \ write_sysreg(val, pmevcntr##n##_el0) -static void write_pmevcntrn(int n, unsigned long val) +static inline void write_pmevcntrn(int n, unsigned long val) { PMEVN_SWITCH(n, WRITE_PMEVCNTRN); } #define WRITE_PMEVTYPERN(n) \ write_sysreg(val, pmevtyper##n##_el0) -static void write_pmevtypern(int n, unsigned long val) +static inline void write_pmevtypern(int n, unsigned long val) { PMEVN_SWITCH(n, WRITE_PMEVTYPERN); } diff --git a/arch/arm64/include/asm/cputype.h b/arch/arm64/include/asm/cputype.h index 683ca3af408485aa6cdf4aa44b5988d91cd6a0f3..5f6f84837a490375c8175626e3fec40057bb56d9 100644 --- a/arch/arm64/include/asm/cputype.h +++ b/arch/arm64/include/asm/cputype.h @@ -126,6 +126,10 @@ #define APPLE_CPU_PART_M1_FIRESTORM_MAX 0x029 #define APPLE_CPU_PART_M2_BLIZZARD 0x032 #define APPLE_CPU_PART_M2_AVALANCHE 0x033 +#define APPLE_CPU_PART_M2_BLIZZARD_PRO 0x034 +#define APPLE_CPU_PART_M2_AVALANCHE_PRO 0x035 +#define APPLE_CPU_PART_M2_BLIZZARD_MAX 0x038 +#define APPLE_CPU_PART_M2_AVALANCHE_MAX 0x039 #define AMPERE_CPU_PART_AMPERE1 0xAC3 @@ -181,6 +185,10 @@ #define MIDR_APPLE_M1_FIRESTORM_MAX MIDR_CPU_MODEL(ARM_CPU_IMP_APPLE, APPLE_CPU_PART_M1_FIRESTORM_MAX) #define MIDR_APPLE_M2_BLIZZARD MIDR_CPU_MODEL(ARM_CPU_IMP_APPLE, APPLE_CPU_PART_M2_BLIZZARD) #define MIDR_APPLE_M2_AVALANCHE MIDR_CPU_MODEL(ARM_CPU_IMP_APPLE, APPLE_CPU_PART_M2_AVALANCHE) +#define MIDR_APPLE_M2_BLIZZARD_PRO MIDR_CPU_MODEL(ARM_CPU_IMP_APPLE, APPLE_CPU_PART_M2_BLIZZARD_PRO) +#define MIDR_APPLE_M2_AVALANCHE_PRO MIDR_CPU_MODEL(ARM_CPU_IMP_APPLE, APPLE_CPU_PART_M2_AVALANCHE_PRO) +#define MIDR_APPLE_M2_BLIZZARD_MAX MIDR_CPU_MODEL(ARM_CPU_IMP_APPLE, APPLE_CPU_PART_M2_BLIZZARD_MAX) +#define MIDR_APPLE_M2_AVALANCHE_MAX MIDR_CPU_MODEL(ARM_CPU_IMP_APPLE, APPLE_CPU_PART_M2_AVALANCHE_MAX) #define MIDR_AMPERE1 MIDR_CPU_MODEL(ARM_CPU_IMP_AMPERE, AMPERE_CPU_PART_AMPERE1) /* Fujitsu Erratum 010001 affects A64FX 1.0 and 1.1, (v0r0 and v1r0) */ diff --git a/arch/arm64/include/asm/kvm_pgtable.h b/arch/arm64/include/asm/kvm_pgtable.h index 4cd6762bda805d16fe798a1db258155ced0d1a28..93bd0975b15f5545d5a408fd177ce35e780204ad 100644 --- a/arch/arm64/include/asm/kvm_pgtable.h +++ b/arch/arm64/include/asm/kvm_pgtable.h @@ -209,6 +209,7 @@ struct kvm_pgtable_visit_ctx { kvm_pte_t old; void *arg; struct kvm_pgtable_mm_ops *mm_ops; + u64 start; u64 addr; u64 end; u32 level; @@ -631,9 +632,9 @@ int kvm_pgtable_stage2_flush(struct kvm_pgtable *pgt, u64 addr, u64 size); * * The walker will walk the page-table entries corresponding to the input * address range specified, visiting entries according to the walker flags. - * Invalid entries are treated as leaf entries. Leaf entries are reloaded - * after invoking the walker callback, allowing the walker to descend into - * a newly installed table. + * Invalid entries are treated as leaf entries. The visited page table entry is + * reloaded after invoking the walker callback, allowing the walker to descend + * into a newly installed table. * * Returning a negative error code from the walker callback function will * terminate the walk immediately with the same error code. diff --git a/arch/arm64/include/asm/sysreg.h b/arch/arm64/include/asm/sysreg.h index e72d9aaab6b1b63461b6b061536486528c041514..eefd712f2430353f5800df0b06333dd0460f22c0 100644 --- a/arch/arm64/include/asm/sysreg.h +++ b/arch/arm64/include/asm/sysreg.h @@ -115,8 +115,14 @@ #define SB_BARRIER_INSN __SYS_BARRIER_INSN(0, 7, 31) #define SYS_DC_ISW sys_insn(1, 0, 7, 6, 2) +#define SYS_DC_IGSW sys_insn(1, 0, 7, 6, 4) +#define SYS_DC_IGDSW sys_insn(1, 0, 7, 6, 6) #define SYS_DC_CSW sys_insn(1, 0, 7, 10, 2) +#define SYS_DC_CGSW sys_insn(1, 0, 7, 10, 4) +#define SYS_DC_CGDSW sys_insn(1, 0, 7, 10, 6) #define SYS_DC_CISW sys_insn(1, 0, 7, 14, 2) +#define SYS_DC_CIGSW sys_insn(1, 0, 7, 14, 4) +#define SYS_DC_CIGDSW sys_insn(1, 0, 7, 14, 6) /* * Automatically generated definitions for system registers, the diff --git a/arch/arm64/kernel/mte.c b/arch/arm64/kernel/mte.c index f5bcb0dc6267385e3401ed2735429b45af56e8e7..7e89968bd28244ced303586a35bb2e81dc1ffad6 100644 --- a/arch/arm64/kernel/mte.c +++ b/arch/arm64/kernel/mte.c @@ -66,13 +66,10 @@ void mte_sync_tags(pte_t old_pte, pte_t pte) return; /* if PG_mte_tagged is set, tags have already been initialised */ - for (i = 0; i < nr_pages; i++, page++) { - if (!page_mte_tagged(page)) { + for (i = 0; i < nr_pages; i++, page++) + if (!page_mte_tagged(page)) mte_sync_page_tags(page, old_pte, check_swap, pte_is_tagged); - set_page_mte_tagged(page); - } - } /* ensure the tags are visible before the PTE is set */ smp_wmb(); diff --git a/arch/arm64/kernel/vdso.c b/arch/arm64/kernel/vdso.c index 0119dc91abb5dbc7567344aff13760454fbb24c3..d9e1355730ef5811163d173590970d79207a8c5b 100644 --- a/arch/arm64/kernel/vdso.c +++ b/arch/arm64/kernel/vdso.c @@ -288,7 +288,7 @@ static int aarch32_alloc_kuser_vdso_page(void) memcpy((void *)(vdso_page + 0x1000 - kuser_sz), __kuser_helper_start, kuser_sz); - aarch32_vectors_page = virt_to_page(vdso_page); + aarch32_vectors_page = virt_to_page((void *)vdso_page); return 0; } diff --git a/arch/arm64/kvm/fpsimd.c b/arch/arm64/kvm/fpsimd.c index 1279949599b5fc84948f4061d6571537883c54e9..4c9dcd8fc93950d2927725ef23e21275267dd0c0 100644 --- a/arch/arm64/kvm/fpsimd.c +++ b/arch/arm64/kvm/fpsimd.c @@ -81,26 +81,34 @@ void kvm_arch_vcpu_load_fp(struct kvm_vcpu *vcpu) fpsimd_kvm_prepare(); + /* + * We will check TIF_FOREIGN_FPSTATE just before entering the + * guest in kvm_arch_vcpu_ctxflush_fp() and override this to + * FP_STATE_FREE if the flag set. + */ vcpu->arch.fp_state = FP_STATE_HOST_OWNED; vcpu_clear_flag(vcpu, HOST_SVE_ENABLED); if (read_sysreg(cpacr_el1) & CPACR_EL1_ZEN_EL0EN) vcpu_set_flag(vcpu, HOST_SVE_ENABLED); - /* - * We don't currently support SME guests but if we leave - * things in streaming mode then when the guest starts running - * FPSIMD or SVE code it may generate SME traps so as a - * special case if we are in streaming mode we force the host - * state to be saved now and exit streaming mode so that we - * don't have to handle any SME traps for valid guest - * operations. Do this for ZA as well for now for simplicity. - */ if (system_supports_sme()) { vcpu_clear_flag(vcpu, HOST_SME_ENABLED); if (read_sysreg(cpacr_el1) & CPACR_EL1_SMEN_EL0EN) vcpu_set_flag(vcpu, HOST_SME_ENABLED); + /* + * If PSTATE.SM is enabled then save any pending FP + * state and disable PSTATE.SM. If we leave PSTATE.SM + * enabled and the guest does not enable SME via + * CPACR_EL1.SMEN then operations that should be valid + * may generate SME traps from EL1 to EL1 which we + * can't intercept and which would confuse the guest. + * + * Do the same for PSTATE.ZA in the case where there + * is state in the registers which has not already + * been saved, this is very unlikely to happen. + */ if (read_sysreg_s(SYS_SVCR) & (SVCR_SM_MASK | SVCR_ZA_MASK)) { vcpu->arch.fp_state = FP_STATE_FREE; fpsimd_save_and_flush_cpu_state(); diff --git a/arch/arm64/kvm/hyp/include/hyp/switch.h b/arch/arm64/kvm/hyp/include/hyp/switch.h index c41166f1a1dd7dd6c228dd7e0477424dd19be9ba..5c15c58f90ccebb1afe2c364a5fec23559dfad03 100644 --- a/arch/arm64/kvm/hyp/include/hyp/switch.h +++ b/arch/arm64/kvm/hyp/include/hyp/switch.h @@ -177,9 +177,17 @@ static bool kvm_hyp_handle_fpsimd(struct kvm_vcpu *vcpu, u64 *exit_code) sve_guest = vcpu_has_sve(vcpu); esr_ec = kvm_vcpu_trap_get_class(vcpu); - /* Don't handle SVE traps for non-SVE vcpus here: */ - if (!sve_guest && esr_ec != ESR_ELx_EC_FP_ASIMD) + /* Only handle traps the vCPU can support here: */ + switch (esr_ec) { + case ESR_ELx_EC_FP_ASIMD: + break; + case ESR_ELx_EC_SVE: + if (!sve_guest) + return false; + break; + default: return false; + } /* Valid trap. Switch the context: */ @@ -404,17 +412,21 @@ static bool kvm_hyp_handle_cp15_32(struct kvm_vcpu *vcpu, u64 *exit_code) return false; } -static bool kvm_hyp_handle_iabt_low(struct kvm_vcpu *vcpu, u64 *exit_code) +static bool kvm_hyp_handle_memory_fault(struct kvm_vcpu *vcpu, u64 *exit_code) { if (!__populate_fault_info(vcpu)) return true; return false; } +static bool kvm_hyp_handle_iabt_low(struct kvm_vcpu *vcpu, u64 *exit_code) + __alias(kvm_hyp_handle_memory_fault); +static bool kvm_hyp_handle_watchpt_low(struct kvm_vcpu *vcpu, u64 *exit_code) + __alias(kvm_hyp_handle_memory_fault); static bool kvm_hyp_handle_dabt_low(struct kvm_vcpu *vcpu, u64 *exit_code) { - if (!__populate_fault_info(vcpu)) + if (kvm_hyp_handle_memory_fault(vcpu, exit_code)) return true; if (static_branch_unlikely(&vgic_v2_cpuif_trap)) { diff --git a/arch/arm64/kvm/hyp/nvhe/mem_protect.c b/arch/arm64/kvm/hyp/nvhe/mem_protect.c index 2e9ec4a2a4a323d238e29f8ac199da8ea5301768..a8813b2129966d094e54154dec71c92781c032bb 100644 --- a/arch/arm64/kvm/hyp/nvhe/mem_protect.c +++ b/arch/arm64/kvm/hyp/nvhe/mem_protect.c @@ -575,7 +575,7 @@ struct pkvm_mem_donation { struct check_walk_data { enum pkvm_page_state desired; - enum pkvm_page_state (*get_page_state)(kvm_pte_t pte); + enum pkvm_page_state (*get_page_state)(kvm_pte_t pte, u64 addr); }; static int __check_page_state_visitor(const struct kvm_pgtable_visit_ctx *ctx, @@ -583,10 +583,7 @@ static int __check_page_state_visitor(const struct kvm_pgtable_visit_ctx *ctx, { struct check_walk_data *d = ctx->arg; - if (kvm_pte_valid(ctx->old) && !addr_is_allowed_memory(kvm_pte_to_phys(ctx->old))) - return -EINVAL; - - return d->get_page_state(ctx->old) == d->desired ? 0 : -EPERM; + return d->get_page_state(ctx->old, ctx->addr) == d->desired ? 0 : -EPERM; } static int check_page_state_range(struct kvm_pgtable *pgt, u64 addr, u64 size, @@ -601,8 +598,11 @@ static int check_page_state_range(struct kvm_pgtable *pgt, u64 addr, u64 size, return kvm_pgtable_walk(pgt, addr, size, &walker); } -static enum pkvm_page_state host_get_page_state(kvm_pte_t pte) +static enum pkvm_page_state host_get_page_state(kvm_pte_t pte, u64 addr) { + if (!addr_is_allowed_memory(addr)) + return PKVM_NOPAGE; + if (!kvm_pte_valid(pte) && pte) return PKVM_NOPAGE; @@ -709,7 +709,7 @@ static int host_complete_donation(u64 addr, const struct pkvm_mem_transition *tx return host_stage2_set_owner_locked(addr, size, host_id); } -static enum pkvm_page_state hyp_get_page_state(kvm_pte_t pte) +static enum pkvm_page_state hyp_get_page_state(kvm_pte_t pte, u64 addr) { if (!kvm_pte_valid(pte)) return PKVM_NOPAGE; diff --git a/arch/arm64/kvm/hyp/nvhe/switch.c b/arch/arm64/kvm/hyp/nvhe/switch.c index 71fa16a0dc775890adbcebe3a2ade963734ef445..77791495c9951d60e18fa8f80fd5c22385d65840 100644 --- a/arch/arm64/kvm/hyp/nvhe/switch.c +++ b/arch/arm64/kvm/hyp/nvhe/switch.c @@ -186,6 +186,7 @@ static const exit_handler_fn hyp_exit_handlers[] = { [ESR_ELx_EC_FP_ASIMD] = kvm_hyp_handle_fpsimd, [ESR_ELx_EC_IABT_LOW] = kvm_hyp_handle_iabt_low, [ESR_ELx_EC_DABT_LOW] = kvm_hyp_handle_dabt_low, + [ESR_ELx_EC_WATCHPT_LOW] = kvm_hyp_handle_watchpt_low, [ESR_ELx_EC_PAC] = kvm_hyp_handle_ptrauth, }; @@ -196,6 +197,7 @@ static const exit_handler_fn pvm_exit_handlers[] = { [ESR_ELx_EC_FP_ASIMD] = kvm_hyp_handle_fpsimd, [ESR_ELx_EC_IABT_LOW] = kvm_hyp_handle_iabt_low, [ESR_ELx_EC_DABT_LOW] = kvm_hyp_handle_dabt_low, + [ESR_ELx_EC_WATCHPT_LOW] = kvm_hyp_handle_watchpt_low, [ESR_ELx_EC_PAC] = kvm_hyp_handle_ptrauth, }; diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c index 3d61bd3e591d27e9858b028220a9eb81a7498f6d..95dae02ccc2e60c2a830061a23b2e9d8c16e6246 100644 --- a/arch/arm64/kvm/hyp/pgtable.c +++ b/arch/arm64/kvm/hyp/pgtable.c @@ -58,8 +58,9 @@ struct kvm_pgtable_walk_data { struct kvm_pgtable_walker *walker; + const u64 start; u64 addr; - u64 end; + const u64 end; }; static bool kvm_phys_is_valid(u64 phys) @@ -201,20 +202,33 @@ static inline int __kvm_pgtable_visit(struct kvm_pgtable_walk_data *data, .old = READ_ONCE(*ptep), .arg = data->walker->arg, .mm_ops = mm_ops, + .start = data->start, .addr = data->addr, .end = data->end, .level = level, .flags = flags, }; int ret = 0; + bool reload = false; kvm_pteref_t childp; bool table = kvm_pte_table(ctx.old, level); - if (table && (ctx.flags & KVM_PGTABLE_WALK_TABLE_PRE)) + if (table && (ctx.flags & KVM_PGTABLE_WALK_TABLE_PRE)) { ret = kvm_pgtable_visitor_cb(data, &ctx, KVM_PGTABLE_WALK_TABLE_PRE); + reload = true; + } if (!table && (ctx.flags & KVM_PGTABLE_WALK_LEAF)) { ret = kvm_pgtable_visitor_cb(data, &ctx, KVM_PGTABLE_WALK_LEAF); + reload = true; + } + + /* + * Reload the page table after invoking the walker callback for leaf + * entries or after pre-order traversal, to allow the walker to descend + * into a newly installed or replaced table. + */ + if (reload) { ctx.old = READ_ONCE(*ptep); table = kvm_pte_table(ctx.old, level); } @@ -293,6 +307,7 @@ int kvm_pgtable_walk(struct kvm_pgtable *pgt, u64 addr, u64 size, struct kvm_pgtable_walker *walker) { struct kvm_pgtable_walk_data walk_data = { + .start = ALIGN_DOWN(addr, PAGE_SIZE), .addr = ALIGN_DOWN(addr, PAGE_SIZE), .end = PAGE_ALIGN(walk_data.addr + size), .walker = walker, @@ -349,7 +364,7 @@ int kvm_pgtable_get_leaf(struct kvm_pgtable *pgt, u64 addr, } struct hyp_map_data { - u64 phys; + const u64 phys; kvm_pte_t attr; }; @@ -407,13 +422,12 @@ enum kvm_pgtable_prot kvm_pgtable_hyp_pte_prot(kvm_pte_t pte) static bool hyp_map_walker_try_leaf(const struct kvm_pgtable_visit_ctx *ctx, struct hyp_map_data *data) { + u64 phys = data->phys + (ctx->addr - ctx->start); kvm_pte_t new; - u64 granule = kvm_granule_size(ctx->level), phys = data->phys; if (!kvm_block_mapping_supported(ctx, phys)) return false; - data->phys += granule; new = kvm_init_valid_leaf_pte(phys, data->attr, ctx->level); if (ctx->old == new) return true; @@ -576,7 +590,7 @@ void kvm_pgtable_hyp_destroy(struct kvm_pgtable *pgt) } struct stage2_map_data { - u64 phys; + const u64 phys; kvm_pte_t attr; u8 owner_id; @@ -794,20 +808,43 @@ static bool stage2_pte_executable(kvm_pte_t pte) return !(pte & KVM_PTE_LEAF_ATTR_HI_S2_XN); } +static u64 stage2_map_walker_phys_addr(const struct kvm_pgtable_visit_ctx *ctx, + const struct stage2_map_data *data) +{ + u64 phys = data->phys; + + /* + * Stage-2 walks to update ownership data are communicated to the map + * walker using an invalid PA. Avoid offsetting an already invalid PA, + * which could overflow and make the address valid again. + */ + if (!kvm_phys_is_valid(phys)) + return phys; + + /* + * Otherwise, work out the correct PA based on how far the walk has + * gotten. + */ + return phys + (ctx->addr - ctx->start); +} + static bool stage2_leaf_mapping_allowed(const struct kvm_pgtable_visit_ctx *ctx, struct stage2_map_data *data) { + u64 phys = stage2_map_walker_phys_addr(ctx, data); + if (data->force_pte && (ctx->level < (KVM_PGTABLE_MAX_LEVELS - 1))) return false; - return kvm_block_mapping_supported(ctx, data->phys); + return kvm_block_mapping_supported(ctx, phys); } static int stage2_map_walker_try_leaf(const struct kvm_pgtable_visit_ctx *ctx, struct stage2_map_data *data) { kvm_pte_t new; - u64 granule = kvm_granule_size(ctx->level), phys = data->phys; + u64 phys = stage2_map_walker_phys_addr(ctx, data); + u64 granule = kvm_granule_size(ctx->level); struct kvm_pgtable *pgt = data->mmu->pgt; struct kvm_pgtable_mm_ops *mm_ops = ctx->mm_ops; @@ -841,8 +878,6 @@ static int stage2_map_walker_try_leaf(const struct kvm_pgtable_visit_ctx *ctx, stage2_make_pte(ctx, new); - if (kvm_phys_is_valid(phys)) - data->phys += granule; return 0; } @@ -1297,4 +1332,7 @@ void kvm_pgtable_stage2_free_removed(struct kvm_pgtable_mm_ops *mm_ops, void *pg }; WARN_ON(__kvm_pgtable_walk(&data, mm_ops, ptep, level + 1)); + + WARN_ON(mm_ops->page_count(pgtable) != 1); + mm_ops->put_page(pgtable); } diff --git a/arch/arm64/kvm/hyp/vhe/switch.c b/arch/arm64/kvm/hyp/vhe/switch.c index 3d868e84c7a0a33246f86b4ccadaed8b5abec548..7a1aa511e7da6f68fc88eb4a6b39dcb7927ded09 100644 --- a/arch/arm64/kvm/hyp/vhe/switch.c +++ b/arch/arm64/kvm/hyp/vhe/switch.c @@ -110,6 +110,7 @@ static const exit_handler_fn hyp_exit_handlers[] = { [ESR_ELx_EC_FP_ASIMD] = kvm_hyp_handle_fpsimd, [ESR_ELx_EC_IABT_LOW] = kvm_hyp_handle_iabt_low, [ESR_ELx_EC_DABT_LOW] = kvm_hyp_handle_dabt_low, + [ESR_ELx_EC_WATCHPT_LOW] = kvm_hyp_handle_watchpt_low, [ESR_ELx_EC_PAC] = kvm_hyp_handle_ptrauth, }; diff --git a/arch/arm64/kvm/inject_fault.c b/arch/arm64/kvm/inject_fault.c index 64c3aec0d937c36b84d49b7ca403933429b993c8..0bd93a5f21ce382506803d018ba7df7072ac0aa1 100644 --- a/arch/arm64/kvm/inject_fault.c +++ b/arch/arm64/kvm/inject_fault.c @@ -204,7 +204,7 @@ void kvm_inject_size_fault(struct kvm_vcpu *vcpu) * Size Fault at level 0, as if exceeding PARange. * * Non-LPAE guests will only get the external abort, as there - * is no way to to describe the ASF. + * is no way to describe the ASF. */ if (vcpu_el1_is_32bit(vcpu) && !(vcpu_read_sys_reg(vcpu, TCR_EL1) & TTBCR_EAE)) diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c index 45727d50d18dd90c842285a2d621628c77c9b0e6..491ca7eb2a4c6e186abbbae9112942e0420f0116 100644 --- a/arch/arm64/kvm/pmu-emul.c +++ b/arch/arm64/kvm/pmu-emul.c @@ -694,45 +694,23 @@ void kvm_host_pmu_init(struct arm_pmu *pmu) static struct arm_pmu *kvm_pmu_probe_armpmu(void) { - struct perf_event_attr attr = { }; - struct perf_event *event; - struct arm_pmu *pmu = NULL; - - /* - * Create a dummy event that only counts user cycles. As we'll never - * leave this function with the event being live, it will never - * count anything. But it allows us to probe some of the PMU - * details. Yes, this is terrible. - */ - attr.type = PERF_TYPE_RAW; - attr.size = sizeof(attr); - attr.pinned = 1; - attr.disabled = 0; - attr.exclude_user = 0; - attr.exclude_kernel = 1; - attr.exclude_hv = 1; - attr.exclude_host = 1; - attr.config = ARMV8_PMUV3_PERFCTR_CPU_CYCLES; - attr.sample_period = GENMASK(63, 0); + struct arm_pmu *tmp, *pmu = NULL; + struct arm_pmu_entry *entry; + int cpu; - event = perf_event_create_kernel_counter(&attr, -1, current, - kvm_pmu_perf_overflow, &attr); + mutex_lock(&arm_pmus_lock); - if (IS_ERR(event)) { - pr_err_once("kvm: pmu event creation failed %ld\n", - PTR_ERR(event)); - return NULL; - } + cpu = smp_processor_id(); + list_for_each_entry(entry, &arm_pmus, entry) { + tmp = entry->arm_pmu; - if (event->pmu) { - pmu = to_arm_pmu(event->pmu); - if (pmu->pmuver == ID_AA64DFR0_EL1_PMUVer_NI || - pmu->pmuver == ID_AA64DFR0_EL1_PMUVer_IMP_DEF) - pmu = NULL; + if (cpumask_test_cpu(cpu, &tmp->supported_cpus)) { + pmu = tmp; + break; + } } - perf_event_disable(event); - perf_event_release_kernel(event); + mutex_unlock(&arm_pmus_lock); return pmu; } @@ -912,7 +890,17 @@ int kvm_arm_pmu_v3_set_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr) return -EBUSY; if (!kvm->arch.arm_pmu) { - /* No PMU set, get the default one */ + /* + * No PMU set, get the default one. + * + * The observant among you will notice that the supported_cpus + * mask does not get updated for the default PMU even though it + * is quite possible the selected instance supports only a + * subset of cores in the system. This is intentional, and + * upholds the preexisting behavior on heterogeneous systems + * where vCPUs can be scheduled on any core but the guest + * counters could stop working. + */ kvm->arch.arm_pmu = kvm_pmu_probe_armpmu(); if (!kvm->arch.arm_pmu) return -ENODEV; diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c index 71b12094d6137008f65bf6e6297747e815bc4bd3..753aa741814977728778648fc7e3c7e05644ab51 100644 --- a/arch/arm64/kvm/sys_regs.c +++ b/arch/arm64/kvm/sys_regs.c @@ -211,6 +211,19 @@ static bool access_dcsw(struct kvm_vcpu *vcpu, return true; } +static bool access_dcgsw(struct kvm_vcpu *vcpu, + struct sys_reg_params *p, + const struct sys_reg_desc *r) +{ + if (!kvm_has_mte(vcpu->kvm)) { + kvm_inject_undefined(vcpu); + return false; + } + + /* Treat MTE S/W ops as we treat the classic ones: with contempt */ + return access_dcsw(vcpu, p, r); +} + static void get_access_mask(const struct sys_reg_desc *r, u64 *mask, u64 *shift) { switch (r->aarch32_map) { @@ -1756,8 +1769,14 @@ static bool access_spsr(struct kvm_vcpu *vcpu, */ static const struct sys_reg_desc sys_reg_descs[] = { { SYS_DESC(SYS_DC_ISW), access_dcsw }, + { SYS_DESC(SYS_DC_IGSW), access_dcgsw }, + { SYS_DESC(SYS_DC_IGDSW), access_dcgsw }, { SYS_DESC(SYS_DC_CSW), access_dcsw }, + { SYS_DESC(SYS_DC_CGSW), access_dcgsw }, + { SYS_DESC(SYS_DC_CGDSW), access_dcgsw }, { SYS_DESC(SYS_DC_CISW), access_dcsw }, + { SYS_DESC(SYS_DC_CIGSW), access_dcgsw }, + { SYS_DESC(SYS_DC_CIGDSW), access_dcgsw }, DBG_BCR_BVR_WCR_WVR_EL1(0), DBG_BCR_BVR_WCR_WVR_EL1(1), diff --git a/arch/arm64/kvm/vgic/vgic-init.c b/arch/arm64/kvm/vgic/vgic-init.c index 9d42c7cb2b588235abf9b314cf92bc52f61429b0..6eafc2c45cfcf73e1bb5d817545ec4f290a388e9 100644 --- a/arch/arm64/kvm/vgic/vgic-init.c +++ b/arch/arm64/kvm/vgic/vgic-init.c @@ -235,9 +235,9 @@ int kvm_vgic_vcpu_init(struct kvm_vcpu *vcpu) * KVM io device for the redistributor that belongs to this VCPU. */ if (dist->vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3) { - mutex_lock(&vcpu->kvm->arch.config_lock); + mutex_lock(&vcpu->kvm->slots_lock); ret = vgic_register_redist_iodev(vcpu); - mutex_unlock(&vcpu->kvm->arch.config_lock); + mutex_unlock(&vcpu->kvm->slots_lock); } return ret; } @@ -406,7 +406,7 @@ void kvm_vgic_destroy(struct kvm *kvm) /** * vgic_lazy_init: Lazy init is only allowed if the GIC exposed to the guest - * is a GICv2. A GICv3 must be explicitly initialized by the guest using the + * is a GICv2. A GICv3 must be explicitly initialized by userspace using the * KVM_DEV_ARM_VGIC_GRP_CTRL KVM_DEVICE group. * @kvm: kvm struct pointer */ @@ -446,11 +446,13 @@ int vgic_lazy_init(struct kvm *kvm) int kvm_vgic_map_resources(struct kvm *kvm) { struct vgic_dist *dist = &kvm->arch.vgic; + gpa_t dist_base; int ret = 0; if (likely(vgic_ready(kvm))) return 0; + mutex_lock(&kvm->slots_lock); mutex_lock(&kvm->arch.config_lock); if (vgic_ready(kvm)) goto out; @@ -463,13 +465,26 @@ int kvm_vgic_map_resources(struct kvm *kvm) else ret = vgic_v3_map_resources(kvm); - if (ret) + if (ret) { __kvm_vgic_destroy(kvm); - else - dist->ready = true; + goto out; + } + dist->ready = true; + dist_base = dist->vgic_dist_base; + mutex_unlock(&kvm->arch.config_lock); + + ret = vgic_register_dist_iodev(kvm, dist_base, + kvm_vgic_global_state.type); + if (ret) { + kvm_err("Unable to register VGIC dist MMIO regions\n"); + kvm_vgic_destroy(kvm); + } + mutex_unlock(&kvm->slots_lock); + return ret; out: mutex_unlock(&kvm->arch.config_lock); + mutex_unlock(&kvm->slots_lock); return ret; } diff --git a/arch/arm64/kvm/vgic/vgic-its.c b/arch/arm64/kvm/vgic/vgic-its.c index 750e51e3779a31a0eda2f8223be233fb10355664..5fe2365a629f25013fd9039ab8279fe76a31349c 100644 --- a/arch/arm64/kvm/vgic/vgic-its.c +++ b/arch/arm64/kvm/vgic/vgic-its.c @@ -1936,6 +1936,7 @@ void vgic_lpi_translation_cache_destroy(struct kvm *kvm) static int vgic_its_create(struct kvm_device *dev, u32 type) { + int ret; struct vgic_its *its; if (type != KVM_DEV_TYPE_ARM_VGIC_ITS) @@ -1945,9 +1946,12 @@ static int vgic_its_create(struct kvm_device *dev, u32 type) if (!its) return -ENOMEM; + mutex_lock(&dev->kvm->arch.config_lock); + if (vgic_initialized(dev->kvm)) { - int ret = vgic_v4_init(dev->kvm); + ret = vgic_v4_init(dev->kvm); if (ret < 0) { + mutex_unlock(&dev->kvm->arch.config_lock); kfree(its); return ret; } @@ -1960,12 +1964,10 @@ static int vgic_its_create(struct kvm_device *dev, u32 type) /* Yep, even more trickery for lock ordering... */ #ifdef CONFIG_LOCKDEP - mutex_lock(&dev->kvm->arch.config_lock); mutex_lock(&its->cmd_lock); mutex_lock(&its->its_lock); mutex_unlock(&its->its_lock); mutex_unlock(&its->cmd_lock); - mutex_unlock(&dev->kvm->arch.config_lock); #endif its->vgic_its_base = VGIC_ADDR_UNDEF; @@ -1986,7 +1988,11 @@ static int vgic_its_create(struct kvm_device *dev, u32 type) dev->private = its; - return vgic_its_set_abi(its, NR_ITS_ABIS - 1); + ret = vgic_its_set_abi(its, NR_ITS_ABIS - 1); + + mutex_unlock(&dev->kvm->arch.config_lock); + + return ret; } static void vgic_its_destroy(struct kvm_device *kvm_dev) diff --git a/arch/arm64/kvm/vgic/vgic-kvm-device.c b/arch/arm64/kvm/vgic/vgic-kvm-device.c index 35cfa268fd5de7a0e3d1e1fa7eaa864715f77103..212b73a715c1c245540e12636085bb0d4ac1d741 100644 --- a/arch/arm64/kvm/vgic/vgic-kvm-device.c +++ b/arch/arm64/kvm/vgic/vgic-kvm-device.c @@ -102,7 +102,11 @@ static int kvm_vgic_addr(struct kvm *kvm, struct kvm_device_attr *attr, bool wri if (get_user(addr, uaddr)) return -EFAULT; - mutex_lock(&kvm->arch.config_lock); + /* + * Since we can't hold config_lock while registering the redistributor + * iodevs, take the slots_lock immediately. + */ + mutex_lock(&kvm->slots_lock); switch (attr->attr) { case KVM_VGIC_V2_ADDR_TYPE_DIST: r = vgic_check_type(kvm, KVM_DEV_TYPE_ARM_VGIC_V2); @@ -182,6 +186,7 @@ static int kvm_vgic_addr(struct kvm *kvm, struct kvm_device_attr *attr, bool wri if (r) goto out; + mutex_lock(&kvm->arch.config_lock); if (write) { r = vgic_check_iorange(kvm, *addr_ptr, addr, alignment, size); if (!r) @@ -189,9 +194,10 @@ static int kvm_vgic_addr(struct kvm *kvm, struct kvm_device_attr *attr, bool wri } else { addr = *addr_ptr; } + mutex_unlock(&kvm->arch.config_lock); out: - mutex_unlock(&kvm->arch.config_lock); + mutex_unlock(&kvm->slots_lock); if (!r && !write) r = put_user(addr, uaddr); diff --git a/arch/arm64/kvm/vgic/vgic-mmio-v3.c b/arch/arm64/kvm/vgic/vgic-mmio-v3.c index 472b18ac92a242bd922cc1a3f58a56326de2c40b..188d2187eede935e43b31fcf4985205a1872d92f 100644 --- a/arch/arm64/kvm/vgic/vgic-mmio-v3.c +++ b/arch/arm64/kvm/vgic/vgic-mmio-v3.c @@ -769,10 +769,13 @@ int vgic_register_redist_iodev(struct kvm_vcpu *vcpu) struct vgic_io_device *rd_dev = &vcpu->arch.vgic_cpu.rd_iodev; struct vgic_redist_region *rdreg; gpa_t rd_base; - int ret; + int ret = 0; + + lockdep_assert_held(&kvm->slots_lock); + mutex_lock(&kvm->arch.config_lock); if (!IS_VGIC_ADDR_UNDEF(vgic_cpu->rd_iodev.base_addr)) - return 0; + goto out_unlock; /* * We may be creating VCPUs before having set the base address for the @@ -782,10 +785,12 @@ int vgic_register_redist_iodev(struct kvm_vcpu *vcpu) */ rdreg = vgic_v3_rdist_free_slot(&vgic->rd_regions); if (!rdreg) - return 0; + goto out_unlock; - if (!vgic_v3_check_base(kvm)) - return -EINVAL; + if (!vgic_v3_check_base(kvm)) { + ret = -EINVAL; + goto out_unlock; + } vgic_cpu->rdreg = rdreg; vgic_cpu->rdreg_index = rdreg->free_index; @@ -799,16 +804,20 @@ int vgic_register_redist_iodev(struct kvm_vcpu *vcpu) rd_dev->nr_regions = ARRAY_SIZE(vgic_v3_rd_registers); rd_dev->redist_vcpu = vcpu; - mutex_lock(&kvm->slots_lock); + mutex_unlock(&kvm->arch.config_lock); + ret = kvm_io_bus_register_dev(kvm, KVM_MMIO_BUS, rd_base, 2 * SZ_64K, &rd_dev->dev); - mutex_unlock(&kvm->slots_lock); - if (ret) return ret; + /* Protected by slots_lock */ rdreg->free_index++; return 0; + +out_unlock: + mutex_unlock(&kvm->arch.config_lock); + return ret; } static void vgic_unregister_redist_iodev(struct kvm_vcpu *vcpu) @@ -834,12 +843,10 @@ static int vgic_register_all_redist_iodevs(struct kvm *kvm) /* The current c failed, so iterate over the previous ones. */ int i; - mutex_lock(&kvm->slots_lock); for (i = 0; i < c; i++) { vcpu = kvm_get_vcpu(kvm, i); vgic_unregister_redist_iodev(vcpu); } - mutex_unlock(&kvm->slots_lock); } return ret; @@ -938,7 +945,9 @@ int vgic_v3_set_redist_base(struct kvm *kvm, u32 index, u64 addr, u32 count) { int ret; + mutex_lock(&kvm->arch.config_lock); ret = vgic_v3_alloc_redist_region(kvm, index, addr, count); + mutex_unlock(&kvm->arch.config_lock); if (ret) return ret; @@ -950,8 +959,10 @@ int vgic_v3_set_redist_base(struct kvm *kvm, u32 index, u64 addr, u32 count) if (ret) { struct vgic_redist_region *rdreg; + mutex_lock(&kvm->arch.config_lock); rdreg = vgic_v3_rdist_region_from_index(kvm, index); vgic_v3_free_redist_region(rdreg); + mutex_unlock(&kvm->arch.config_lock); return ret; } diff --git a/arch/arm64/kvm/vgic/vgic-mmio.c b/arch/arm64/kvm/vgic/vgic-mmio.c index 1939c94e0b248801800ea00c62f081f2b0c9f62d..ff558c05e990c728abd5361054cb04ef44083818 100644 --- a/arch/arm64/kvm/vgic/vgic-mmio.c +++ b/arch/arm64/kvm/vgic/vgic-mmio.c @@ -1096,7 +1096,6 @@ int vgic_register_dist_iodev(struct kvm *kvm, gpa_t dist_base_address, enum vgic_type type) { struct vgic_io_device *io_device = &kvm->arch.vgic.dist_iodev; - int ret = 0; unsigned int len; switch (type) { @@ -1114,10 +1113,6 @@ int vgic_register_dist_iodev(struct kvm *kvm, gpa_t dist_base_address, io_device->iodev_type = IODEV_DIST; io_device->redist_vcpu = NULL; - mutex_lock(&kvm->slots_lock); - ret = kvm_io_bus_register_dev(kvm, KVM_MMIO_BUS, dist_base_address, - len, &io_device->dev); - mutex_unlock(&kvm->slots_lock); - - return ret; + return kvm_io_bus_register_dev(kvm, KVM_MMIO_BUS, dist_base_address, + len, &io_device->dev); } diff --git a/arch/arm64/kvm/vgic/vgic-v2.c b/arch/arm64/kvm/vgic/vgic-v2.c index 645648349c99b122359535c36f221d78432e25c0..7e9cdb78f7ce885f9917d380d99fa18547af47e6 100644 --- a/arch/arm64/kvm/vgic/vgic-v2.c +++ b/arch/arm64/kvm/vgic/vgic-v2.c @@ -312,12 +312,6 @@ int vgic_v2_map_resources(struct kvm *kvm) return ret; } - ret = vgic_register_dist_iodev(kvm, dist->vgic_dist_base, VGIC_V2); - if (ret) { - kvm_err("Unable to register VGIC MMIO regions\n"); - return ret; - } - if (!static_branch_unlikely(&vgic_v2_cpuif_trap)) { ret = kvm_phys_addr_ioremap(kvm, dist->vgic_cpu_base, kvm_vgic_global_state.vcpu_base, diff --git a/arch/arm64/kvm/vgic/vgic-v3.c b/arch/arm64/kvm/vgic/vgic-v3.c index 469d816f356f3fe94b021f5e2fd14ea91310f114..c3b8e132d5992616ca1f308456ababce125c6c01 100644 --- a/arch/arm64/kvm/vgic/vgic-v3.c +++ b/arch/arm64/kvm/vgic/vgic-v3.c @@ -539,7 +539,6 @@ int vgic_v3_map_resources(struct kvm *kvm) { struct vgic_dist *dist = &kvm->arch.vgic; struct kvm_vcpu *vcpu; - int ret = 0; unsigned long c; kvm_for_each_vcpu(c, vcpu, kvm) { @@ -569,12 +568,6 @@ int vgic_v3_map_resources(struct kvm *kvm) return -EBUSY; } - ret = vgic_register_dist_iodev(kvm, dist->vgic_dist_base, VGIC_V3); - if (ret) { - kvm_err("Unable to register VGICv3 dist MMIO regions\n"); - return ret; - } - if (kvm_vgic_global_state.has_gicv4_1) vgic_v4_configure_vsgis(kvm); @@ -616,6 +609,10 @@ static const struct midr_range broken_seis[] = { MIDR_ALL_VERSIONS(MIDR_APPLE_M1_FIRESTORM_MAX), MIDR_ALL_VERSIONS(MIDR_APPLE_M2_BLIZZARD), MIDR_ALL_VERSIONS(MIDR_APPLE_M2_AVALANCHE), + MIDR_ALL_VERSIONS(MIDR_APPLE_M2_BLIZZARD_PRO), + MIDR_ALL_VERSIONS(MIDR_APPLE_M2_AVALANCHE_PRO), + MIDR_ALL_VERSIONS(MIDR_APPLE_M2_BLIZZARD_MAX), + MIDR_ALL_VERSIONS(MIDR_APPLE_M2_AVALANCHE_MAX), {}, }; diff --git a/arch/arm64/kvm/vgic/vgic-v4.c b/arch/arm64/kvm/vgic/vgic-v4.c index 3bb0034780605772859f329620765d742e3fdf59..c1c28fe680ba317e9b8712de9a04b264991cdd0a 100644 --- a/arch/arm64/kvm/vgic/vgic-v4.c +++ b/arch/arm64/kvm/vgic/vgic-v4.c @@ -184,13 +184,14 @@ static void vgic_v4_disable_vsgis(struct kvm_vcpu *vcpu) } } -/* Must be called with the kvm lock held */ void vgic_v4_configure_vsgis(struct kvm *kvm) { struct vgic_dist *dist = &kvm->arch.vgic; struct kvm_vcpu *vcpu; unsigned long i; + lockdep_assert_held(&kvm->arch.config_lock); + kvm_arm_halt_guest(kvm); kvm_for_each_vcpu(i, vcpu, kvm) { diff --git a/arch/arm64/kvm/vmid.c b/arch/arm64/kvm/vmid.c index 08978d0672e7e947b0afbca0eb7d5ae0d9be440a..7fe8ba1a2851c5b71acbf17075987b96436f1a4a 100644 --- a/arch/arm64/kvm/vmid.c +++ b/arch/arm64/kvm/vmid.c @@ -47,7 +47,7 @@ static void flush_context(void) int cpu; u64 vmid; - bitmap_clear(vmid_map, 0, NUM_USER_VMIDS); + bitmap_zero(vmid_map, NUM_USER_VMIDS); for_each_possible_cpu(cpu) { vmid = atomic64_xchg_relaxed(&per_cpu(active_vmids, cpu), 0); @@ -182,8 +182,7 @@ int __init kvm_arm_vmid_alloc_init(void) */ WARN_ON(NUM_USER_VMIDS - 1 <= num_possible_cpus()); atomic64_set(&vmid_generation, VMID_FIRST_VERSION); - vmid_map = kcalloc(BITS_TO_LONGS(NUM_USER_VMIDS), - sizeof(*vmid_map), GFP_KERNEL); + vmid_map = bitmap_zalloc(NUM_USER_VMIDS, GFP_KERNEL); if (!vmid_map) return -ENOMEM; @@ -192,5 +191,5 @@ int __init kvm_arm_vmid_alloc_init(void) void __init kvm_arm_vmid_alloc_free(void) { - kfree(vmid_map); + bitmap_free(vmid_map); } diff --git a/arch/arm64/mm/copypage.c b/arch/arm64/mm/copypage.c index 4aadcfb017545dbaef8fd7fb321ab2bdd9d565bd..a7bb20055ce0948a6f586e29fd1a05bbbc1cc75f 100644 --- a/arch/arm64/mm/copypage.c +++ b/arch/arm64/mm/copypage.c @@ -21,9 +21,10 @@ void copy_highpage(struct page *to, struct page *from) copy_page(kto, kfrom); + if (kasan_hw_tags_enabled()) + page_kasan_tag_reset(to); + if (system_supports_mte() && page_mte_tagged(from)) { - if (kasan_hw_tags_enabled()) - page_kasan_tag_reset(to); /* It's a new page, shouldn't have been tagged yet */ WARN_ON_ONCE(!try_page_mte_tagging(to)); mte_copy_page_tags(kto, kfrom); diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c index 9e0db5c387e3ae903ca5b6526b5ef6978eee22bf..6045a5117ac15bca1fe0cceba12509e13003df87 100644 --- a/arch/arm64/mm/fault.c +++ b/arch/arm64/mm/fault.c @@ -480,8 +480,8 @@ static void do_bad_area(unsigned long far, unsigned long esr, } } -#define VM_FAULT_BADMAP 0x010000 -#define VM_FAULT_BADACCESS 0x020000 +#define VM_FAULT_BADMAP ((__force vm_fault_t)0x010000) +#define VM_FAULT_BADACCESS ((__force vm_fault_t)0x020000) static vm_fault_t __do_page_fault(struct mm_struct *mm, unsigned long addr, unsigned int mm_flags, unsigned long vm_flags, @@ -600,8 +600,7 @@ static int __kprobes do_page_fault(unsigned long far, unsigned long esr, vma_end_read(vma); goto lock_mmap; } - fault = handle_mm_fault(vma, addr & PAGE_MASK, - mm_flags | FAULT_FLAG_VMA_LOCK, regs); + fault = handle_mm_fault(vma, addr, mm_flags | FAULT_FLAG_VMA_LOCK, regs); vma_end_read(vma); if (!(fault & VM_FAULT_RETRY)) { diff --git a/arch/loongarch/include/asm/loongarch.h b/arch/loongarch/include/asm/loongarch.h index b3323ab5b78d37d5acdea5ca9b3d3479eb2fdcc8..35e8a52fea11a71004e1a79e092f4ec6c0e2d10e 100644 --- a/arch/loongarch/include/asm/loongarch.h +++ b/arch/loongarch/include/asm/loongarch.h @@ -1496,7 +1496,7 @@ __BUILD_CSR_OP(tlbidx) #define write_fcsr(dest, val) \ do { \ __asm__ __volatile__( \ - " movgr2fcsr %0, "__stringify(dest)" \n" \ + " movgr2fcsr "__stringify(dest)", %0 \n" \ : : "r" (val)); \ } while (0) diff --git a/arch/loongarch/include/asm/pgtable-bits.h b/arch/loongarch/include/asm/pgtable-bits.h index 8b98d22a145b1d67974cffb52b116f8a626bf7ed..de46a6b1e9f11c87d68ef526e30c4cf816b17a38 100644 --- a/arch/loongarch/include/asm/pgtable-bits.h +++ b/arch/loongarch/include/asm/pgtable-bits.h @@ -22,12 +22,14 @@ #define _PAGE_PFN_SHIFT 12 #define _PAGE_SWP_EXCLUSIVE_SHIFT 23 #define _PAGE_PFN_END_SHIFT 48 +#define _PAGE_PRESENT_INVALID_SHIFT 60 #define _PAGE_NO_READ_SHIFT 61 #define _PAGE_NO_EXEC_SHIFT 62 #define _PAGE_RPLV_SHIFT 63 /* Used by software */ #define _PAGE_PRESENT (_ULCAST_(1) << _PAGE_PRESENT_SHIFT) +#define _PAGE_PRESENT_INVALID (_ULCAST_(1) << _PAGE_PRESENT_INVALID_SHIFT) #define _PAGE_WRITE (_ULCAST_(1) << _PAGE_WRITE_SHIFT) #define _PAGE_ACCESSED (_ULCAST_(1) << _PAGE_ACCESSED_SHIFT) #define _PAGE_MODIFIED (_ULCAST_(1) << _PAGE_MODIFIED_SHIFT) diff --git a/arch/loongarch/include/asm/pgtable.h b/arch/loongarch/include/asm/pgtable.h index d28fb9dbec5966c8dfc7ad07e5ae5908f61ed1af..9a9f9ff9b7098000002c3ed029e54c62a7ba05b1 100644 --- a/arch/loongarch/include/asm/pgtable.h +++ b/arch/loongarch/include/asm/pgtable.h @@ -213,7 +213,7 @@ static inline int pmd_bad(pmd_t pmd) static inline int pmd_present(pmd_t pmd) { if (unlikely(pmd_val(pmd) & _PAGE_HUGE)) - return !!(pmd_val(pmd) & (_PAGE_PRESENT | _PAGE_PROTNONE)); + return !!(pmd_val(pmd) & (_PAGE_PRESENT | _PAGE_PROTNONE | _PAGE_PRESENT_INVALID)); return pmd_val(pmd) != (unsigned long)invalid_pte_table; } @@ -558,6 +558,7 @@ static inline pmd_t pmd_modify(pmd_t pmd, pgprot_t newprot) static inline pmd_t pmd_mkinvalid(pmd_t pmd) { + pmd_val(pmd) |= _PAGE_PRESENT_INVALID; pmd_val(pmd) &= ~(_PAGE_PRESENT | _PAGE_VALID | _PAGE_DIRTY | _PAGE_PROTNONE); return pmd; diff --git a/arch/loongarch/kernel/hw_breakpoint.c b/arch/loongarch/kernel/hw_breakpoint.c index 2406c95b34cc4f0023f1db8755e60ff7ea0cc18a..021b59c248fac64413ad70a5387ee7a3f698b92d 100644 --- a/arch/loongarch/kernel/hw_breakpoint.c +++ b/arch/loongarch/kernel/hw_breakpoint.c @@ -396,6 +396,8 @@ int hw_breakpoint_arch_parse(struct perf_event *bp, if (hw->ctrl.type != LOONGARCH_BREAKPOINT_EXECUTE) alignment_mask = 0x7; + else + alignment_mask = 0x3; offset = hw->address & alignment_mask; hw->address &= ~alignment_mask; diff --git a/arch/loongarch/kernel/perf_event.c b/arch/loongarch/kernel/perf_event.c index ff28f99b47d7ebce47bdaf5b3a97b8b2cb49481b..0491bf453cd49601c4f8b7b35565ea4a2b83c689 100644 --- a/arch/loongarch/kernel/perf_event.c +++ b/arch/loongarch/kernel/perf_event.c @@ -271,7 +271,7 @@ static void loongarch_pmu_enable_event(struct hw_perf_event *evt, int idx) WARN_ON(idx < 0 || idx >= loongarch_pmu.num_counters); /* Make sure interrupt enabled. */ - cpuc->saved_ctrl[idx] = M_PERFCTL_EVENT(evt->event_base & 0xff) | + cpuc->saved_ctrl[idx] = M_PERFCTL_EVENT(evt->event_base) | (evt->config_base & M_PERFCTL_CONFIG_MASK) | CSR_PERFCTRL_IE; cpu = (event->cpu >= 0) ? event->cpu : smp_processor_id(); @@ -594,7 +594,7 @@ static struct pmu pmu = { static unsigned int loongarch_pmu_perf_event_encode(const struct loongarch_perf_event *pev) { - return (pev->event_id & 0xff); + return M_PERFCTL_EVENT(pev->event_id); } static const struct loongarch_perf_event *loongarch_pmu_map_general_event(int idx) @@ -849,7 +849,7 @@ static void resume_local_counters(void) static const struct loongarch_perf_event *loongarch_pmu_map_raw_event(u64 config) { - raw_event.event_id = config & 0xff; + raw_event.event_id = M_PERFCTL_EVENT(config); return &raw_event; } diff --git a/arch/loongarch/kernel/unaligned.c b/arch/loongarch/kernel/unaligned.c index bdff825d29ef4dd252430cbd0f5e923fb102cb5f..85fae3d2d71aebc5691686cfe67214561fd8d6fe 100644 --- a/arch/loongarch/kernel/unaligned.c +++ b/arch/loongarch/kernel/unaligned.c @@ -485,7 +485,7 @@ static int __init debugfs_unaligned(void) struct dentry *d; d = debugfs_create_dir("loongarch", NULL); - if (!d) + if (IS_ERR_OR_NULL(d)) return -ENOMEM; debugfs_create_u32("unaligned_instructions_user", diff --git a/arch/m68k/kernel/signal.c b/arch/m68k/kernel/signal.c index b9f6908a31bc39f94c0418a298aed87470755170..ba468b5f3f0b65bd57d74af263d19d9ba917f635 100644 --- a/arch/m68k/kernel/signal.c +++ b/arch/m68k/kernel/signal.c @@ -858,11 +858,17 @@ static inline int rt_setup_ucontext(struct ucontext __user *uc, struct pt_regs * } static inline void __user * -get_sigframe(struct ksignal *ksig, size_t frame_size) +get_sigframe(struct ksignal *ksig, struct pt_regs *tregs, size_t frame_size) { unsigned long usp = sigsp(rdusp(), ksig); + unsigned long gap = 0; - return (void __user *)((usp - frame_size) & -8UL); + if (CPU_IS_020_OR_030 && tregs->format == 0xb) { + /* USP is unreliable so use worst-case value */ + gap = 256; + } + + return (void __user *)((usp - gap - frame_size) & -8UL); } static int setup_frame(struct ksignal *ksig, sigset_t *set, @@ -880,7 +886,7 @@ static int setup_frame(struct ksignal *ksig, sigset_t *set, return -EFAULT; } - frame = get_sigframe(ksig, sizeof(*frame) + fsize); + frame = get_sigframe(ksig, tregs, sizeof(*frame) + fsize); if (fsize) err |= copy_to_user (frame + 1, regs + 1, fsize); @@ -952,7 +958,7 @@ static int setup_rt_frame(struct ksignal *ksig, sigset_t *set, return -EFAULT; } - frame = get_sigframe(ksig, sizeof(*frame)); + frame = get_sigframe(ksig, tregs, sizeof(*frame)); if (fsize) err |= copy_to_user (&frame->uc.uc_extra, regs + 1, fsize); diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index c2f5498d207fffd0f07bfff905ff75e05c504802..675a8660cb85a033792d9622040ea48a8b800cb0 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -79,6 +79,7 @@ config MIPS select HAVE_LD_DEAD_CODE_DATA_ELIMINATION select HAVE_MOD_ARCH_SPECIFIC select HAVE_NMI + select HAVE_PATA_PLATFORM select HAVE_PERF_EVENTS select HAVE_PERF_REGS select HAVE_PERF_USER_STACK_DUMP diff --git a/arch/mips/alchemy/common/dbdma.c b/arch/mips/alchemy/common/dbdma.c index 5ab0430004092900cb8a7de8013031ff18d862fe..6a3c890f7bbfedc43459edc3c69b6f8cae3c79dc 100644 --- a/arch/mips/alchemy/common/dbdma.c +++ b/arch/mips/alchemy/common/dbdma.c @@ -30,6 +30,7 @@ * */ +#include /* for dma_default_coherent */ #include #include #include @@ -623,17 +624,18 @@ u32 au1xxx_dbdma_put_source(u32 chanid, dma_addr_t buf, int nbytes, u32 flags) dp->dscr_cmd0 &= ~DSCR_CMD0_IE; /* - * There is an errata on the Au1200/Au1550 parts that could result - * in "stale" data being DMA'ed. It has to do with the snoop logic on - * the cache eviction buffer. DMA_NONCOHERENT is on by default for - * these parts. If it is fixed in the future, these dma_cache_inv will - * just be nothing more than empty macros. See io.h. + * There is an erratum on certain Au1200/Au1550 revisions that could + * result in "stale" data being DMA'ed. It has to do with the snoop + * logic on the cache eviction buffer. dma_default_coherent is set + * to false on these parts. */ - dma_cache_wback_inv((unsigned long)buf, nbytes); + if (!dma_default_coherent) + dma_cache_wback_inv(KSEG0ADDR(buf), nbytes); dp->dscr_cmd0 |= DSCR_CMD0_V; /* Let it rip */ wmb(); /* drain writebuffer */ dma_cache_wback_inv((unsigned long)dp, sizeof(*dp)); ctp->chan_ptr->ddma_dbell = 0; + wmb(); /* force doorbell write out to dma engine */ /* Get next descriptor pointer. */ ctp->put_ptr = phys_to_virt(DSCR_GET_NXTPTR(dp->dscr_nxtptr)); @@ -685,17 +687,18 @@ u32 au1xxx_dbdma_put_dest(u32 chanid, dma_addr_t buf, int nbytes, u32 flags) dp->dscr_source1, dp->dscr_dest0, dp->dscr_dest1); #endif /* - * There is an errata on the Au1200/Au1550 parts that could result in - * "stale" data being DMA'ed. It has to do with the snoop logic on the - * cache eviction buffer. DMA_NONCOHERENT is on by default for these - * parts. If it is fixed in the future, these dma_cache_inv will just - * be nothing more than empty macros. See io.h. + * There is an erratum on certain Au1200/Au1550 revisions that could + * result in "stale" data being DMA'ed. It has to do with the snoop + * logic on the cache eviction buffer. dma_default_coherent is set + * to false on these parts. */ - dma_cache_inv((unsigned long)buf, nbytes); + if (!dma_default_coherent) + dma_cache_inv(KSEG0ADDR(buf), nbytes); dp->dscr_cmd0 |= DSCR_CMD0_V; /* Let it rip */ wmb(); /* drain writebuffer */ dma_cache_wback_inv((unsigned long)dp, sizeof(*dp)); ctp->chan_ptr->ddma_dbell = 0; + wmb(); /* force doorbell write out to dma engine */ /* Get next descriptor pointer. */ ctp->put_ptr = phys_to_virt(DSCR_GET_NXTPTR(dp->dscr_nxtptr)); diff --git a/arch/mips/kernel/cpu-probe.c b/arch/mips/kernel/cpu-probe.c index 6d15a398d389e2bd269d47794edd1b37a370b956..e79adcb128e67df5a13dfc30740e555b4697554c 100644 --- a/arch/mips/kernel/cpu-probe.c +++ b/arch/mips/kernel/cpu-probe.c @@ -1502,6 +1502,10 @@ static inline void cpu_probe_alchemy(struct cpuinfo_mips *c, unsigned int cpu) break; } break; + case PRID_IMP_NETLOGIC_AU13XX: + c->cputype = CPU_ALCHEMY; + __cpu_name[cpu] = "Au1300"; + break; } } @@ -1863,6 +1867,7 @@ void cpu_probe(void) cpu_probe_mips(c, cpu); break; case PRID_COMP_ALCHEMY: + case PRID_COMP_NETLOGIC: cpu_probe_alchemy(c, cpu); break; case PRID_COMP_SIBYTE: diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c index febdc5564638e9b9bbb959791ecb75d6068b676c..c0e65135481b74bfb102f09dd56607c634f45b55 100644 --- a/arch/mips/kernel/setup.c +++ b/arch/mips/kernel/setup.c @@ -158,10 +158,6 @@ static unsigned long __init init_initrd(void) pr_err("initrd start must be page aligned\n"); goto disable; } - if (initrd_start < PAGE_OFFSET) { - pr_err("initrd start < PAGE_OFFSET\n"); - goto disable; - } /* * Sanitize initrd addresses. For example firmware @@ -174,6 +170,11 @@ static unsigned long __init init_initrd(void) initrd_end = (unsigned long)__va(end); initrd_start = (unsigned long)__va(__pa(initrd_start)); + if (initrd_start < PAGE_OFFSET) { + pr_err("initrd start < PAGE_OFFSET\n"); + goto disable; + } + ROOT_DEV = Root_RAM0; return PFN_UP(end); disable: diff --git a/arch/nios2/boot/dts/10m50_devboard.dts b/arch/nios2/boot/dts/10m50_devboard.dts index 56339bef3247d544fd34a101377b08d13d906959..0e7e5b0dd685c15abfae91f0e833a6035eed2c75 100644 --- a/arch/nios2/boot/dts/10m50_devboard.dts +++ b/arch/nios2/boot/dts/10m50_devboard.dts @@ -97,7 +97,7 @@ rgmii_0_eth_tse_0: ethernet@400 { rx-fifo-depth = <8192>; tx-fifo-depth = <8192>; address-bits = <48>; - max-frame-size = <1518>; + max-frame-size = <1500>; local-mac-address = [00 00 00 00 00 00]; altr,has-supplementary-unicast; altr,enable-sup-addr = <1>; diff --git a/arch/nios2/boot/dts/3c120_devboard.dts b/arch/nios2/boot/dts/3c120_devboard.dts index d10fb81686c7efb41dc1459700d04289bb1519f3..3ee316906379795493e85917ef6c2a94c5220144 100644 --- a/arch/nios2/boot/dts/3c120_devboard.dts +++ b/arch/nios2/boot/dts/3c120_devboard.dts @@ -106,7 +106,7 @@ tse_mac: ethernet@4000 { interrupt-names = "rx_irq", "tx_irq"; rx-fifo-depth = <8192>; tx-fifo-depth = <8192>; - max-frame-size = <1518>; + max-frame-size = <1500>; local-mac-address = [ 00 00 00 00 00 00 ]; phy-mode = "rgmii-id"; phy-handle = <&phy0>; diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig index 466a25525364d2d45e7ed5bf286856242fdc90d2..967bde65dd0ec895d78b2314a49aa8cb043e28c0 100644 --- a/arch/parisc/Kconfig +++ b/arch/parisc/Kconfig @@ -130,6 +130,10 @@ config PM config STACKTRACE_SUPPORT def_bool y +config LOCKDEP_SUPPORT + bool + default y + config ISA_DMA_API bool diff --git a/arch/parisc/Kconfig.debug b/arch/parisc/Kconfig.debug index f66554cd5c4518cf919abf321ce9137e8b2be59d..3a059cb5e112fb09fe701fe362c07c84cec247e2 100644 --- a/arch/parisc/Kconfig.debug +++ b/arch/parisc/Kconfig.debug @@ -1 +1,12 @@ # SPDX-License-Identifier: GPL-2.0 +# +config LIGHTWEIGHT_SPINLOCK_CHECK + bool "Enable lightweight spinlock checks" + depends on SMP && !DEBUG_SPINLOCK + default y + help + Add checks with low performance impact to the spinlock functions + to catch memory overwrites at runtime. For more advanced + spinlock debugging you should choose the DEBUG_SPINLOCK option + which will detect unitialized spinlocks too. + If unsure say Y here. diff --git a/arch/parisc/include/asm/assembly.h b/arch/parisc/include/asm/assembly.h index 0f0d4a496fef0ed5c4f54aae6fc1f75dbc2cb8a7..75677b526b2bb79f7bb531a1281e24296d3c3c7c 100644 --- a/arch/parisc/include/asm/assembly.h +++ b/arch/parisc/include/asm/assembly.h @@ -90,10 +90,6 @@ #include #include - sp = 30 - gp = 27 - ipsw = 22 - /* * We provide two versions of each macro to convert from physical * to virtual and vice versa. The "_r1" versions take one argument diff --git a/arch/parisc/include/asm/cacheflush.h b/arch/parisc/include/asm/cacheflush.h index 0bdee672413206aa9db8c23c4bd3fd3be2182660..c8b6928cee1ee442f4b9c9b0c996fe1801620f75 100644 --- a/arch/parisc/include/asm/cacheflush.h +++ b/arch/parisc/include/asm/cacheflush.h @@ -48,6 +48,10 @@ void flush_dcache_page(struct page *page); #define flush_dcache_mmap_lock(mapping) xa_lock_irq(&mapping->i_pages) #define flush_dcache_mmap_unlock(mapping) xa_unlock_irq(&mapping->i_pages) +#define flush_dcache_mmap_lock_irqsave(mapping, flags) \ + xa_lock_irqsave(&mapping->i_pages, flags) +#define flush_dcache_mmap_unlock_irqrestore(mapping, flags) \ + xa_unlock_irqrestore(&mapping->i_pages, flags) #define flush_icache_page(vma,page) do { \ flush_kernel_dcache_page_addr(page_address(page)); \ diff --git a/arch/parisc/include/asm/spinlock.h b/arch/parisc/include/asm/spinlock.h index a6e5d66a76562aaa8e89bab4b55e4aa39d13ad43..edfcb9858bcb7ca0bf2fee2cfeef304836beb4a5 100644 --- a/arch/parisc/include/asm/spinlock.h +++ b/arch/parisc/include/asm/spinlock.h @@ -7,10 +7,26 @@ #include #include +#define SPINLOCK_BREAK_INSN 0x0000c006 /* break 6,6 */ + +static inline void arch_spin_val_check(int lock_val) +{ + if (IS_ENABLED(CONFIG_LIGHTWEIGHT_SPINLOCK_CHECK)) + asm volatile( "andcm,= %0,%1,%%r0\n" + ".word %2\n" + : : "r" (lock_val), "r" (__ARCH_SPIN_LOCK_UNLOCKED_VAL), + "i" (SPINLOCK_BREAK_INSN)); +} + static inline int arch_spin_is_locked(arch_spinlock_t *x) { - volatile unsigned int *a = __ldcw_align(x); - return READ_ONCE(*a) == 0; + volatile unsigned int *a; + int lock_val; + + a = __ldcw_align(x); + lock_val = READ_ONCE(*a); + arch_spin_val_check(lock_val); + return (lock_val == 0); } static inline void arch_spin_lock(arch_spinlock_t *x) @@ -18,9 +34,18 @@ static inline void arch_spin_lock(arch_spinlock_t *x) volatile unsigned int *a; a = __ldcw_align(x); - while (__ldcw(a) == 0) + do { + int lock_val_old; + + lock_val_old = __ldcw(a); + arch_spin_val_check(lock_val_old); + if (lock_val_old) + return; /* got lock */ + + /* wait until we should try to get lock again */ while (*a == 0) continue; + } while (1); } static inline void arch_spin_unlock(arch_spinlock_t *x) @@ -29,15 +54,19 @@ static inline void arch_spin_unlock(arch_spinlock_t *x) a = __ldcw_align(x); /* Release with ordered store. */ - __asm__ __volatile__("stw,ma %0,0(%1)" : : "r"(1), "r"(a) : "memory"); + __asm__ __volatile__("stw,ma %0,0(%1)" + : : "r"(__ARCH_SPIN_LOCK_UNLOCKED_VAL), "r"(a) : "memory"); } static inline int arch_spin_trylock(arch_spinlock_t *x) { volatile unsigned int *a; + int lock_val; a = __ldcw_align(x); - return __ldcw(a) != 0; + lock_val = __ldcw(a); + arch_spin_val_check(lock_val); + return lock_val != 0; } /* diff --git a/arch/parisc/include/asm/spinlock_types.h b/arch/parisc/include/asm/spinlock_types.h index ca39ee350c3f422eee3ff28fa39cd5090600464e..d65934079ebdb032967f86976da3fc8963a84f80 100644 --- a/arch/parisc/include/asm/spinlock_types.h +++ b/arch/parisc/include/asm/spinlock_types.h @@ -2,13 +2,17 @@ #ifndef __ASM_SPINLOCK_TYPES_H #define __ASM_SPINLOCK_TYPES_H +#define __ARCH_SPIN_LOCK_UNLOCKED_VAL 0x1a46 + typedef struct { #ifdef CONFIG_PA20 volatile unsigned int slock; -# define __ARCH_SPIN_LOCK_UNLOCKED { 1 } +# define __ARCH_SPIN_LOCK_UNLOCKED { __ARCH_SPIN_LOCK_UNLOCKED_VAL } #else volatile unsigned int lock[4]; -# define __ARCH_SPIN_LOCK_UNLOCKED { { 1, 1, 1, 1 } } +# define __ARCH_SPIN_LOCK_UNLOCKED \ + { { __ARCH_SPIN_LOCK_UNLOCKED_VAL, __ARCH_SPIN_LOCK_UNLOCKED_VAL, \ + __ARCH_SPIN_LOCK_UNLOCKED_VAL, __ARCH_SPIN_LOCK_UNLOCKED_VAL } } #endif } arch_spinlock_t; diff --git a/arch/parisc/kernel/alternative.c b/arch/parisc/kernel/alternative.c index 66f5672c70bd469e6e1a261c4a7c080be1d808f5..25c4d6c3375db357f73c12e48826c941b92fc621 100644 --- a/arch/parisc/kernel/alternative.c +++ b/arch/parisc/kernel/alternative.c @@ -25,7 +25,7 @@ void __init_or_module apply_alternatives(struct alt_instr *start, { struct alt_instr *entry; int index = 0, applied = 0; - int num_cpus = num_online_cpus(); + int num_cpus = num_present_cpus(); u16 cond_check; cond_check = ALT_COND_ALWAYS | diff --git a/arch/parisc/kernel/cache.c b/arch/parisc/kernel/cache.c index 1d3b8bc8a6233641c4eeea6d8b775887b0de9a4e..ca4a302d4365fe9e7472d3c748b69369233610af 100644 --- a/arch/parisc/kernel/cache.c +++ b/arch/parisc/kernel/cache.c @@ -399,6 +399,7 @@ void flush_dcache_page(struct page *page) unsigned long offset; unsigned long addr, old_addr = 0; unsigned long count = 0; + unsigned long flags; pgoff_t pgoff; if (mapping && !mapping_mapped(mapping)) { @@ -420,7 +421,7 @@ void flush_dcache_page(struct page *page) * to flush one address here for them all to become coherent * on machines that support equivalent aliasing */ - flush_dcache_mmap_lock(mapping); + flush_dcache_mmap_lock_irqsave(mapping, flags); vma_interval_tree_foreach(mpnt, &mapping->i_mmap, pgoff, pgoff) { offset = (pgoff - mpnt->vm_pgoff) << PAGE_SHIFT; addr = mpnt->vm_start + offset; @@ -460,7 +461,7 @@ void flush_dcache_page(struct page *page) } WARN_ON(++count == 4096); } - flush_dcache_mmap_unlock(mapping); + flush_dcache_mmap_unlock_irqrestore(mapping, flags); } EXPORT_SYMBOL(flush_dcache_page); diff --git a/arch/parisc/kernel/pci-dma.c b/arch/parisc/kernel/pci-dma.c index ba87f791323be008d1e6abfcfbe846a55575cbf5..71ed5391f29d6f00ec85168b02cce2fe4eccb097 100644 --- a/arch/parisc/kernel/pci-dma.c +++ b/arch/parisc/kernel/pci-dma.c @@ -446,11 +446,27 @@ void arch_dma_free(struct device *dev, size_t size, void *vaddr, void arch_sync_dma_for_device(phys_addr_t paddr, size_t size, enum dma_data_direction dir) { + /* + * fdc: The data cache line is written back to memory, if and only if + * it is dirty, and then invalidated from the data cache. + */ flush_kernel_dcache_range((unsigned long)phys_to_virt(paddr), size); } void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size, enum dma_data_direction dir) { - flush_kernel_dcache_range((unsigned long)phys_to_virt(paddr), size); + unsigned long addr = (unsigned long) phys_to_virt(paddr); + + switch (dir) { + case DMA_TO_DEVICE: + case DMA_BIDIRECTIONAL: + flush_kernel_dcache_range(addr, size); + return; + case DMA_FROM_DEVICE: + purge_kernel_dcache_range_asm(addr, addr + size); + return; + default: + BUG(); + } } diff --git a/arch/parisc/kernel/process.c b/arch/parisc/kernel/process.c index 97c6f875bd0e12f188215afa07815356711593f5..24411ab79c307d4ab78f33303a7b1d8afb7d1c1d 100644 --- a/arch/parisc/kernel/process.c +++ b/arch/parisc/kernel/process.c @@ -122,13 +122,18 @@ void machine_power_off(void) /* It seems we have no way to power the system off via * software. The user has to press the button himself. */ - printk(KERN_EMERG "System shut down completed.\n" - "Please power this system off now."); + printk("Power off or press RETURN to reboot.\n"); /* prevent soft lockup/stalled CPU messages for endless loop. */ rcu_sysrq_start(); lockup_detector_soft_poweroff(); - for (;;); + while (1) { + /* reboot if user presses RETURN key */ + if (pdc_iodc_getc() == 13) { + printk("Rebooting...\n"); + machine_restart(NULL); + } + } } void (*pm_power_off)(void); diff --git a/arch/parisc/kernel/traps.c b/arch/parisc/kernel/traps.c index f9696fbf646c473cffbe26f316b1bcd289054431..304eebd1c83e7d802fd9b1d726d479fbaeefa175 100644 --- a/arch/parisc/kernel/traps.c +++ b/arch/parisc/kernel/traps.c @@ -47,6 +47,10 @@ #include #include +#if defined(CONFIG_LIGHTWEIGHT_SPINLOCK_CHECK) +#include +#endif + #include "../math-emu/math-emu.h" /* for handle_fpe() */ static void parisc_show_stack(struct task_struct *task, @@ -291,24 +295,30 @@ static void handle_break(struct pt_regs *regs) } #ifdef CONFIG_KPROBES - if (unlikely(iir == PARISC_KPROBES_BREAK_INSN)) { + if (unlikely(iir == PARISC_KPROBES_BREAK_INSN && !user_mode(regs))) { parisc_kprobe_break_handler(regs); return; } - if (unlikely(iir == PARISC_KPROBES_BREAK_INSN2)) { + if (unlikely(iir == PARISC_KPROBES_BREAK_INSN2 && !user_mode(regs))) { parisc_kprobe_ss_handler(regs); return; } #endif #ifdef CONFIG_KGDB - if (unlikely(iir == PARISC_KGDB_COMPILED_BREAK_INSN || - iir == PARISC_KGDB_BREAK_INSN)) { + if (unlikely((iir == PARISC_KGDB_COMPILED_BREAK_INSN || + iir == PARISC_KGDB_BREAK_INSN)) && !user_mode(regs)) { kgdb_handle_exception(9, SIGTRAP, 0, regs); return; } #endif +#ifdef CONFIG_LIGHTWEIGHT_SPINLOCK_CHECK + if ((iir == SPINLOCK_BREAK_INSN) && !user_mode(regs)) { + die_if_kernel("Spinlock was trashed", regs, 1); + } +#endif + if (unlikely(iir != GDB_BREAK_INSN)) parisc_printk_ratelimited(0, regs, KERN_DEBUG "break %d,%d: pid=%d command='%s'\n", diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 539d1f03ff42a5a60527836791f3b17cead05e05..bff5820b7cda14714f6b76da3bac2de92c8f740f 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -906,11 +906,17 @@ config DATA_SHIFT config ARCH_FORCE_MAX_ORDER int "Order of maximal physically contiguous allocations" + range 7 8 if PPC64 && PPC_64K_PAGES default "8" if PPC64 && PPC_64K_PAGES + range 12 12 if PPC64 && !PPC_64K_PAGES default "12" if PPC64 && !PPC_64K_PAGES + range 8 10 if PPC32 && PPC_16K_PAGES default "8" if PPC32 && PPC_16K_PAGES + range 6 10 if PPC32 && PPC_64K_PAGES default "6" if PPC32 && PPC_64K_PAGES + range 4 10 if PPC32 && PPC_256K_PAGES default "4" if PPC32 && PPC_256K_PAGES + range 10 10 default "10" help The kernel page allocator limits the size of maximal physically diff --git a/arch/powerpc/boot/Makefile b/arch/powerpc/boot/Makefile index 85cde5bf04b7ae7c3cfdb2394cbf67d4c7c96e0e..771b79423bbc2b4ea4ee4d5a5fa3c07b0a14d53c 100644 --- a/arch/powerpc/boot/Makefile +++ b/arch/powerpc/boot/Makefile @@ -34,8 +34,6 @@ endif BOOTCFLAGS := -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs \ -fno-strict-aliasing -O2 -msoft-float -mno-altivec -mno-vsx \ - $(call cc-option,-mno-prefixed) $(call cc-option,-mno-pcrel) \ - $(call cc-option,-mno-mma) \ $(call cc-option,-mno-spe) $(call cc-option,-mspe=no) \ -pipe -fomit-frame-pointer -fno-builtin -fPIC -nostdinc \ $(LINUXINCLUDE) @@ -71,6 +69,10 @@ BOOTAFLAGS := -D__ASSEMBLY__ $(BOOTCFLAGS) -nostdinc BOOTARFLAGS := -crD +BOOTCFLAGS += $(call cc-option,-mno-prefixed) \ + $(call cc-option,-mno-pcrel) \ + $(call cc-option,-mno-mma) + ifdef CONFIG_CC_IS_CLANG BOOTCFLAGS += $(CLANG_FLAGS) BOOTAFLAGS += $(CLANG_FLAGS) diff --git a/arch/powerpc/crypto/Kconfig b/arch/powerpc/crypto/Kconfig index 7113f9355165a663732964cdf1890541936a36a8..ad1872518992e0a074a250ff46997eb3d02740f4 100644 --- a/arch/powerpc/crypto/Kconfig +++ b/arch/powerpc/crypto/Kconfig @@ -96,7 +96,7 @@ config CRYPTO_AES_PPC_SPE config CRYPTO_AES_GCM_P10 tristate "Stitched AES/GCM acceleration support on P10 or later CPU (PPC)" - depends on PPC64 && CPU_LITTLE_ENDIAN + depends on PPC64 && CPU_LITTLE_ENDIAN && VSX select CRYPTO_LIB_AES select CRYPTO_ALGAPI select CRYPTO_AEAD diff --git a/arch/powerpc/crypto/Makefile b/arch/powerpc/crypto/Makefile index 05c7486f42c587b1bf1c59b3ad756c4a6f474609..7b4f516abec1d35b221b29ac6fec238f0eb55d99 100644 --- a/arch/powerpc/crypto/Makefile +++ b/arch/powerpc/crypto/Makefile @@ -22,15 +22,15 @@ sha1-ppc-spe-y := sha1-spe-asm.o sha1-spe-glue.o sha256-ppc-spe-y := sha256-spe-asm.o sha256-spe-glue.o crc32c-vpmsum-y := crc32c-vpmsum_asm.o crc32c-vpmsum_glue.o crct10dif-vpmsum-y := crct10dif-vpmsum_asm.o crct10dif-vpmsum_glue.o -aes-gcm-p10-crypto-y := aes-gcm-p10-glue.o aes-gcm-p10.o ghashp8-ppc.o aesp8-ppc.o +aes-gcm-p10-crypto-y := aes-gcm-p10-glue.o aes-gcm-p10.o ghashp10-ppc.o aesp10-ppc.o quiet_cmd_perl = PERL $@ cmd_perl = $(PERL) $< $(if $(CONFIG_CPU_LITTLE_ENDIAN), linux-ppc64le, linux-ppc64) > $@ -targets += aesp8-ppc.S ghashp8-ppc.S +targets += aesp10-ppc.S ghashp10-ppc.S -$(obj)/aesp8-ppc.S $(obj)/ghashp8-ppc.S: $(obj)/%.S: $(src)/%.pl FORCE +$(obj)/aesp10-ppc.S $(obj)/ghashp10-ppc.S: $(obj)/%.S: $(src)/%.pl FORCE $(call if_changed,perl) -OBJECT_FILES_NON_STANDARD_aesp8-ppc.o := y -OBJECT_FILES_NON_STANDARD_ghashp8-ppc.o := y +OBJECT_FILES_NON_STANDARD_aesp10-ppc.o := y +OBJECT_FILES_NON_STANDARD_ghashp10-ppc.o := y diff --git a/arch/powerpc/crypto/aes-gcm-p10-glue.c b/arch/powerpc/crypto/aes-gcm-p10-glue.c index bd3475f5348d99ec0620c968fd95bcffed2388cc..4b6e899895e7be8214d021559e9d0d7fbc8d65d2 100644 --- a/arch/powerpc/crypto/aes-gcm-p10-glue.c +++ b/arch/powerpc/crypto/aes-gcm-p10-glue.c @@ -30,15 +30,15 @@ MODULE_AUTHOR("Danny Tsen aadLen = alen; i = alen & ~0xf; if (i) { - gcm_ghash_p8(nXi, hash->Htable+32, aad, i); + gcm_ghash_p10(nXi, hash->Htable+32, aad, i); aad += i; alen -= i; } @@ -102,7 +102,7 @@ static void set_aad(struct gcm_ctx *gctx, struct Hash_ctx *hash, nXi[i] ^= aad[i]; memset(gctx->aad_hash, 0, 16); - gcm_ghash_p8(gctx->aad_hash, hash->Htable+32, nXi, 16); + gcm_ghash_p10(gctx->aad_hash, hash->Htable+32, nXi, 16); } else { memcpy(gctx->aad_hash, nXi, 16); } @@ -115,7 +115,7 @@ static void gcmp10_init(struct gcm_ctx *gctx, u8 *iv, unsigned char *rdkey, { __be32 counter = cpu_to_be32(1); - aes_p8_encrypt(hash->H, hash->H, rdkey); + aes_p10_encrypt(hash->H, hash->H, rdkey); set_subkey(hash->H); gcm_init_htable(hash->Htable+32, hash->H); @@ -126,7 +126,7 @@ static void gcmp10_init(struct gcm_ctx *gctx, u8 *iv, unsigned char *rdkey, /* * Encrypt counter vector as iv tag and increment counter. */ - aes_p8_encrypt(iv, gctx->ivtag, rdkey); + aes_p10_encrypt(iv, gctx->ivtag, rdkey); counter = cpu_to_be32(2); *((__be32 *)(iv+12)) = counter; @@ -160,7 +160,7 @@ static void finish_tag(struct gcm_ctx *gctx, struct Hash_ctx *hash, int len) /* * hash (AAD len and len) */ - gcm_ghash_p8(hash->Htable, hash->Htable+32, aclen, 16); + gcm_ghash_p10(hash->Htable, hash->Htable+32, aclen, 16); for (i = 0; i < 16; i++) hash->Htable[i] ^= gctx->ivtag[i]; @@ -192,7 +192,7 @@ static int p10_aes_gcm_setkey(struct crypto_aead *aead, const u8 *key, int ret; vsx_begin(); - ret = aes_p8_set_encrypt_key(key, keylen * 8, &ctx->enc_key); + ret = aes_p10_set_encrypt_key(key, keylen * 8, &ctx->enc_key); vsx_end(); return ret ? -EINVAL : 0; diff --git a/arch/powerpc/crypto/aesp8-ppc.pl b/arch/powerpc/crypto/aesp10-ppc.pl similarity index 99% rename from arch/powerpc/crypto/aesp8-ppc.pl rename to arch/powerpc/crypto/aesp10-ppc.pl index 1f22aec27d79970c5d1d7a4694cd6cb5416bc3d6..2c06ce2a2c7c02b159ac0108f8b380c3781f3403 100644 --- a/arch/powerpc/crypto/aesp8-ppc.pl +++ b/arch/powerpc/crypto/aesp10-ppc.pl @@ -110,7 +110,7 @@ die "can't locate ppc-xlate.pl"; open STDOUT,"| $^X $xlate $flavour ".shift || die "can't call $xlate: $!"; $FRAME=8*$SIZE_T; -$prefix="aes_p8"; +$prefix="aes_p10"; $sp="r1"; $vrsave="r12"; diff --git a/arch/powerpc/crypto/ghashp8-ppc.pl b/arch/powerpc/crypto/ghashp10-ppc.pl similarity index 97% rename from arch/powerpc/crypto/ghashp8-ppc.pl rename to arch/powerpc/crypto/ghashp10-ppc.pl index b56603b4a893c756217255a51fe90fdadae77d24..27a6b0bec645f9ae4e457fcbdbe0bccda07dc71e 100644 --- a/arch/powerpc/crypto/ghashp8-ppc.pl +++ b/arch/powerpc/crypto/ghashp10-ppc.pl @@ -64,7 +64,7 @@ $code=<<___; .text -.globl .gcm_init_p8 +.globl .gcm_init_p10 lis r0,0xfff0 li r8,0x10 mfspr $vrsave,256 @@ -110,7 +110,7 @@ $code=<<___; .long 0 .byte 0,12,0x14,0,0,0,2,0 .long 0 -.size .gcm_init_p8,.-.gcm_init_p8 +.size .gcm_init_p10,.-.gcm_init_p10 .globl .gcm_init_htable lis r0,0xfff0 @@ -237,7 +237,7 @@ $code=<<___; .long 0 .size .gcm_init_htable,.-.gcm_init_htable -.globl .gcm_gmult_p8 +.globl .gcm_gmult_p10 lis r0,0xfff8 li r8,0x10 mfspr $vrsave,256 @@ -283,9 +283,9 @@ $code=<<___; .long 0 .byte 0,12,0x14,0,0,0,2,0 .long 0 -.size .gcm_gmult_p8,.-.gcm_gmult_p8 +.size .gcm_gmult_p10,.-.gcm_gmult_p10 -.globl .gcm_ghash_p8 +.globl .gcm_ghash_p10 lis r0,0xfff8 li r8,0x10 mfspr $vrsave,256 @@ -350,7 +350,7 @@ Loop: .long 0 .byte 0,12,0x14,0,0,0,4,0 .long 0 -.size .gcm_ghash_p8,.-.gcm_ghash_p8 +.size .gcm_ghash_p10,.-.gcm_ghash_p10 .asciz "GHASH for PowerISA 2.07, CRYPTOGAMS by " .align 2 diff --git a/arch/powerpc/include/asm/iommu.h b/arch/powerpc/include/asm/iommu.h index 678b5bdc79b1f5df91372ce5c95fbde92a676918..34e14dfd8e042a679df1dd50cf6d6586aae0f099 100644 --- a/arch/powerpc/include/asm/iommu.h +++ b/arch/powerpc/include/asm/iommu.h @@ -205,7 +205,6 @@ extern void iommu_register_group(struct iommu_table_group *table_group, int pci_domain_number, unsigned long pe_num); extern int iommu_add_device(struct iommu_table_group *table_group, struct device *dev); -extern void iommu_del_device(struct device *dev); extern long iommu_tce_xchg(struct mm_struct *mm, struct iommu_table *tbl, unsigned long entry, unsigned long *hpa, enum dma_data_direction *direction); @@ -229,10 +228,6 @@ static inline int iommu_add_device(struct iommu_table_group *table_group, { return 0; } - -static inline void iommu_del_device(struct device *dev) -{ -} #endif /* !CONFIG_IOMMU_API */ u64 dma_iommu_get_required_mask(struct device *dev); diff --git a/arch/powerpc/kernel/dma-iommu.c b/arch/powerpc/kernel/dma-iommu.c index 038ce8d9061d166b8c85663adc2fa4dab08c23f1..8920862ffd7916ef3afcde6714a1676ae6a1f647 100644 --- a/arch/powerpc/kernel/dma-iommu.c +++ b/arch/powerpc/kernel/dma-iommu.c @@ -144,7 +144,7 @@ static bool dma_iommu_bypass_supported(struct device *dev, u64 mask) /* We support DMA to/from any memory page via the iommu */ int dma_iommu_dma_supported(struct device *dev, u64 mask) { - struct iommu_table *tbl = get_iommu_table_base(dev); + struct iommu_table *tbl; if (dev_is_pci(dev) && dma_iommu_bypass_supported(dev, mask)) { /* @@ -162,6 +162,8 @@ int dma_iommu_dma_supported(struct device *dev, u64 mask) return 1; } + tbl = get_iommu_table_base(dev); + if (!tbl) { dev_err(dev, "Warning: IOMMU dma not supported: mask 0x%08llx, table unavailable\n", mask); return 0; diff --git a/arch/powerpc/kernel/iommu.c b/arch/powerpc/kernel/iommu.c index 0089dd49b4cbf71a964aee0dd7ae5165853fbb70..67f0b01e6ff575d54d95532f04e70e735b3640ee 100644 --- a/arch/powerpc/kernel/iommu.c +++ b/arch/powerpc/kernel/iommu.c @@ -518,7 +518,7 @@ int ppc_iommu_map_sg(struct device *dev, struct iommu_table *tbl, /* Convert entry to a dma_addr_t */ entry += tbl->it_offset; dma_addr = entry << tbl->it_page_shift; - dma_addr |= (s->offset & ~IOMMU_PAGE_MASK(tbl)); + dma_addr |= (vaddr & ~IOMMU_PAGE_MASK(tbl)); DBG(" - %lu pages, entry: %lx, dma_addr: %lx\n", npages, entry, dma_addr); @@ -905,6 +905,7 @@ void *iommu_alloc_coherent(struct device *dev, struct iommu_table *tbl, unsigned int order; unsigned int nio_pages, io_order; struct page *page; + int tcesize = (1 << tbl->it_page_shift); size = PAGE_ALIGN(size); order = get_order(size); @@ -931,7 +932,8 @@ void *iommu_alloc_coherent(struct device *dev, struct iommu_table *tbl, memset(ret, 0, size); /* Set up tces to cover the allocated range */ - nio_pages = size >> tbl->it_page_shift; + nio_pages = IOMMU_PAGE_ALIGN(size, tbl) >> tbl->it_page_shift; + io_order = get_iommu_order(size, tbl); mapping = iommu_alloc(dev, tbl, ret, nio_pages, DMA_BIDIRECTIONAL, mask >> tbl->it_page_shift, io_order, 0); @@ -939,7 +941,8 @@ void *iommu_alloc_coherent(struct device *dev, struct iommu_table *tbl, free_pages((unsigned long)ret, order); return NULL; } - *dma_handle = mapping; + + *dma_handle = mapping | ((u64)ret & (tcesize - 1)); return ret; } @@ -950,7 +953,7 @@ void iommu_free_coherent(struct iommu_table *tbl, size_t size, unsigned int nio_pages; size = PAGE_ALIGN(size); - nio_pages = size >> tbl->it_page_shift; + nio_pages = IOMMU_PAGE_ALIGN(size, tbl) >> tbl->it_page_shift; iommu_free(tbl, dma_handle, nio_pages); size = PAGE_ALIGN(size); free_pages((unsigned long)vaddr, get_order(size)); @@ -1168,23 +1171,6 @@ int iommu_add_device(struct iommu_table_group *table_group, struct device *dev) } EXPORT_SYMBOL_GPL(iommu_add_device); -void iommu_del_device(struct device *dev) -{ - /* - * Some devices might not have IOMMU table and group - * and we needn't detach them from the associated - * IOMMU groups - */ - if (!device_iommu_mapped(dev)) { - pr_debug("iommu_tce: skipping device %s with no tbl\n", - dev_name(dev)); - return; - } - - iommu_group_remove_device(dev); -} -EXPORT_SYMBOL_GPL(iommu_del_device); - /* * A simple iommu_table_group_ops which only allows reusing the existing * iommu_table. This handles VFIO for POWER7 or the nested KVM. diff --git a/arch/powerpc/kernel/isa-bridge.c b/arch/powerpc/kernel/isa-bridge.c index 85bdd7d3652f84222f2cb6cdcc0be1d4a9664099..48e0eaf1ad61559a374fe83b33c65aee6ef35e7f 100644 --- a/arch/powerpc/kernel/isa-bridge.c +++ b/arch/powerpc/kernel/isa-bridge.c @@ -93,11 +93,12 @@ static int process_ISA_OF_ranges(struct device_node *isa_node, } inval_range: - if (!phb_io_base_phys) { + if (phb_io_base_phys) { pr_err("no ISA IO ranges or unexpected isa range, mapping 64k\n"); remap_isa_base(phb_io_base_phys, 0x10000); + return 0; } - return 0; + return -EINVAL; } diff --git a/arch/powerpc/mm/book3s64/radix_pgtable.c b/arch/powerpc/mm/book3s64/radix_pgtable.c index 26245aaf12b8bddb8f6082ce5b2b3c093f2e00cc..2297aa764ecdbcedbaf7958df234f5d5c4408470 100644 --- a/arch/powerpc/mm/book3s64/radix_pgtable.c +++ b/arch/powerpc/mm/book3s64/radix_pgtable.c @@ -1040,8 +1040,8 @@ void radix__ptep_set_access_flags(struct vm_area_struct *vma, pte_t *ptep, pte_t entry, unsigned long address, int psize) { struct mm_struct *mm = vma->vm_mm; - unsigned long set = pte_val(entry) & (_PAGE_DIRTY | _PAGE_ACCESSED | - _PAGE_RW | _PAGE_EXEC); + unsigned long set = pte_val(entry) & (_PAGE_DIRTY | _PAGE_SOFT_DIRTY | + _PAGE_ACCESSED | _PAGE_RW | _PAGE_EXEC); unsigned long change = pte_val(entry) ^ pte_val(*ptep); /* diff --git a/arch/powerpc/net/bpf_jit_comp.c b/arch/powerpc/net/bpf_jit_comp.c index e93aefcfb83f2663008833d9a845ae3a83b22556..37043dfc1adddf67b84f881fe751216db87162ec 100644 --- a/arch/powerpc/net/bpf_jit_comp.c +++ b/arch/powerpc/net/bpf_jit_comp.c @@ -101,6 +101,8 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp) bpf_hdr = jit_data->header; proglen = jit_data->proglen; extra_pass = true; + /* During extra pass, ensure index is reset before repopulating extable entries */ + cgctx.exentry_idx = 0; goto skip_init_ctx; } diff --git a/arch/powerpc/platforms/Kconfig b/arch/powerpc/platforms/Kconfig index 0d9b7609c7d5629404594eb9bec518795aae0763..3e2e252016f7a6e66de2e85a5c6bb548a36c1688 100644 --- a/arch/powerpc/platforms/Kconfig +++ b/arch/powerpc/platforms/Kconfig @@ -265,6 +265,7 @@ config CPM2 config FSL_ULI1575 bool "ULI1575 PCIe south bridge support" depends on FSL_SOC_BOOKE || PPC_86xx + depends on PCI select FSL_PCI select GENERIC_ISA_DMA help diff --git a/arch/powerpc/platforms/powernv/pci.c b/arch/powerpc/platforms/powernv/pci.c index 233a50e65fceddf8d641d71c40f989df7a219144..7725492097b627ff70b6026f0c9267bf94feb9be 100644 --- a/arch/powerpc/platforms/powernv/pci.c +++ b/arch/powerpc/platforms/powernv/pci.c @@ -865,28 +865,3 @@ void __init pnv_pci_init(void) /* Configure IOMMU DMA hooks */ set_pci_dma_ops(&dma_iommu_ops); } - -static int pnv_tce_iommu_bus_notifier(struct notifier_block *nb, - unsigned long action, void *data) -{ - struct device *dev = data; - - switch (action) { - case BUS_NOTIFY_DEL_DEVICE: - iommu_del_device(dev); - return 0; - default: - return 0; - } -} - -static struct notifier_block pnv_tce_iommu_bus_nb = { - .notifier_call = pnv_tce_iommu_bus_notifier, -}; - -static int __init pnv_tce_iommu_bus_notifier_init(void) -{ - bus_register_notifier(&pci_bus_type, &pnv_tce_iommu_bus_nb); - return 0; -} -machine_subsys_initcall_sync(powernv, pnv_tce_iommu_bus_notifier_init); diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c index 7464fa6e41455385b06885e5720bfcb1a5e9dc05..d59e8a98a200826c75a1602129312b8014f5f26d 100644 --- a/arch/powerpc/platforms/pseries/iommu.c +++ b/arch/powerpc/platforms/pseries/iommu.c @@ -91,19 +91,24 @@ static struct iommu_table_group *iommu_pseries_alloc_group(int node) static void iommu_pseries_free_group(struct iommu_table_group *table_group, const char *node_name) { - struct iommu_table *tbl; - if (!table_group) return; - tbl = table_group->tables[0]; #ifdef CONFIG_IOMMU_API if (table_group->group) { iommu_group_put(table_group->group); BUG_ON(table_group->group); } #endif - iommu_tce_table_put(tbl); + + /* Default DMA window table is at index 0, while DDW at 1. SR-IOV + * adapters only have table on index 1. + */ + if (table_group->tables[0]) + iommu_tce_table_put(table_group->tables[0]); + + if (table_group->tables[1]) + iommu_tce_table_put(table_group->tables[1]); kfree(table_group); } @@ -312,13 +317,22 @@ static void tce_free_pSeriesLP(unsigned long liobn, long tcenum, long tceshift, static void tce_freemulti_pSeriesLP(struct iommu_table *tbl, long tcenum, long npages) { u64 rc; + long rpages = npages; + unsigned long limit; if (!firmware_has_feature(FW_FEATURE_STUFF_TCE)) return tce_free_pSeriesLP(tbl->it_index, tcenum, tbl->it_page_shift, npages); - rc = plpar_tce_stuff((u64)tbl->it_index, - (u64)tcenum << tbl->it_page_shift, 0, npages); + do { + limit = min_t(unsigned long, rpages, 512); + + rc = plpar_tce_stuff((u64)tbl->it_index, + (u64)tcenum << tbl->it_page_shift, 0, limit); + + rpages -= limit; + tcenum += limit; + } while (rpages > 0 && !rc); if (rc && printk_ratelimit()) { printk("tce_freemulti_pSeriesLP: plpar_tce_stuff failed\n"); @@ -1695,31 +1709,6 @@ static int __init disable_multitce(char *str) __setup("multitce=", disable_multitce); -static int tce_iommu_bus_notifier(struct notifier_block *nb, - unsigned long action, void *data) -{ - struct device *dev = data; - - switch (action) { - case BUS_NOTIFY_DEL_DEVICE: - iommu_del_device(dev); - return 0; - default: - return 0; - } -} - -static struct notifier_block tce_iommu_bus_nb = { - .notifier_call = tce_iommu_bus_notifier, -}; - -static int __init tce_iommu_bus_notifier_init(void) -{ - bus_register_notifier(&pci_bus_type, &tce_iommu_bus_nb); - return 0; -} -machine_subsys_initcall_sync(pseries, tce_iommu_bus_notifier_init); - #ifdef CONFIG_SPAPR_TCE_IOMMU struct iommu_group *pSeries_pci_device_group(struct pci_controller *hose, struct pci_dev *pdev) diff --git a/arch/powerpc/purgatory/Makefile b/arch/powerpc/purgatory/Makefile index 6f5e2727963c4c18b28129d128e60371bb7f1aaf..78473d69cd2b6752e5cac6d02e7f0cc8affb829a 100644 --- a/arch/powerpc/purgatory/Makefile +++ b/arch/powerpc/purgatory/Makefile @@ -5,6 +5,11 @@ KCSAN_SANITIZE := n targets += trampoline_$(BITS).o purgatory.ro +# When profile-guided optimization is enabled, llvm emits two different +# overlapping text sections, which is not supported by kexec. Remove profile +# optimization flags. +KBUILD_CFLAGS := $(filter-out -fprofile-sample-use=% -fprofile-use=%,$(KBUILD_CFLAGS)) + LDFLAGS_purgatory.ro := -e purgatory_start -r --no-undefined $(obj)/purgatory.ro: $(obj)/trampoline_$(BITS).o FORCE diff --git a/arch/powerpc/xmon/xmon.c b/arch/powerpc/xmon/xmon.c index 728d3c257e4a33323bdca9716dd2658548fc693a..70c4c59a1a8f47f103ba180935974b6e71f6db31 100644 --- a/arch/powerpc/xmon/xmon.c +++ b/arch/powerpc/xmon/xmon.c @@ -88,7 +88,7 @@ static unsigned long ndump = 64; static unsigned long nidump = 16; static unsigned long ncsum = 4096; static int termch; -static char tmpstr[128]; +static char tmpstr[KSYM_NAME_LEN]; static int tracing_enabled; static long bus_error_jmp[JMP_BUF_LEN]; diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig index 348c0fa1fc8c7a8479c42c70bdede50a8a3c5126..5966ad97c30c3d8f157dad32f4da9a5ab07790a7 100644 --- a/arch/riscv/Kconfig +++ b/arch/riscv/Kconfig @@ -26,6 +26,7 @@ config RISCV select ARCH_HAS_GIGANTIC_PAGE select ARCH_HAS_KCOV select ARCH_HAS_MMIOWB + select ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE select ARCH_HAS_PMEM_API select ARCH_HAS_PTE_SPECIAL select ARCH_HAS_SET_DIRECT_MAP if MMU @@ -799,8 +800,11 @@ menu "Power management options" source "kernel/power/Kconfig" +# Hibernation is only possible on systems where the SBI implementation has +# marked its reserved memory as not accessible from, or does not run +# from the same memory as, Linux config ARCH_HIBERNATION_POSSIBLE - def_bool y + def_bool NONPORTABLE config ARCH_HIBERNATION_HEADER def_bool HIBERNATION diff --git a/arch/riscv/errata/Makefile b/arch/riscv/errata/Makefile index a1055965fbee6a6da0b173f9e894e79f8930dfd1..7b2637c8c332653f5c5551a88c0947f732233b23 100644 --- a/arch/riscv/errata/Makefile +++ b/arch/riscv/errata/Makefile @@ -1,2 +1,6 @@ +ifdef CONFIG_RELOCATABLE +KBUILD_CFLAGS += -fno-pie +endif + obj-$(CONFIG_ERRATA_SIFIVE) += sifive/ obj-$(CONFIG_ERRATA_THEAD) += thead/ diff --git a/arch/riscv/include/asm/hugetlb.h b/arch/riscv/include/asm/hugetlb.h index fe6f2300664162c1b0e233e1323aeb96c2d33746..ce1ebda1a49a79b41c240b033c94b2c47d90c465 100644 --- a/arch/riscv/include/asm/hugetlb.h +++ b/arch/riscv/include/asm/hugetlb.h @@ -36,6 +36,9 @@ int huge_ptep_set_access_flags(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep, pte_t pte, int dirty); +#define __HAVE_ARCH_HUGE_PTEP_GET +pte_t huge_ptep_get(pte_t *ptep); + pte_t arch_make_huge_pte(pte_t entry, unsigned int shift, vm_flags_t flags); #define arch_make_huge_pte arch_make_huge_pte diff --git a/arch/riscv/include/asm/kfence.h b/arch/riscv/include/asm/kfence.h index d887a54042aa417ae7152ea41ccb46addf8aea9d..0bbffd528096d97e02e070be23ae05271a283538 100644 --- a/arch/riscv/include/asm/kfence.h +++ b/arch/riscv/include/asm/kfence.h @@ -8,41 +8,8 @@ #include #include -static inline int split_pmd_page(unsigned long addr) -{ - int i; - unsigned long pfn = PFN_DOWN(__pa((addr & PMD_MASK))); - pmd_t *pmd = pmd_off_k(addr); - pte_t *pte = pte_alloc_one_kernel(&init_mm); - - if (!pte) - return -ENOMEM; - - for (i = 0; i < PTRS_PER_PTE; i++) - set_pte(pte + i, pfn_pte(pfn + i, PAGE_KERNEL)); - set_pmd(pmd, pfn_pmd(PFN_DOWN(__pa(pte)), PAGE_TABLE)); - - flush_tlb_kernel_range(addr, addr + PMD_SIZE); - return 0; -} - static inline bool arch_kfence_init_pool(void) { - int ret; - unsigned long addr; - pmd_t *pmd; - - for (addr = (unsigned long)__kfence_pool; is_kfence_address((void *)addr); - addr += PAGE_SIZE) { - pmd = pmd_off_k(addr); - - if (pmd_leaf(*pmd)) { - ret = split_pmd_page(addr); - if (ret) - return false; - } - } - return true; } diff --git a/arch/riscv/include/asm/perf_event.h b/arch/riscv/include/asm/perf_event.h index d42c901f9a9770985b53da11098d69cf7934cc62..665bbc9b2f840a5272eb493d799f21dab73dfd25 100644 --- a/arch/riscv/include/asm/perf_event.h +++ b/arch/riscv/include/asm/perf_event.h @@ -10,4 +10,11 @@ #include #define perf_arch_bpf_user_pt_regs(regs) (struct user_regs_struct *)regs + +#define perf_arch_fetch_caller_regs(regs, __ip) { \ + (regs)->epc = (__ip); \ + (regs)->s0 = (unsigned long) __builtin_frame_address(0); \ + (regs)->sp = current_stack_pointer; \ + (regs)->status = SR_PP; \ +} #endif /* _ASM_RISCV_PERF_EVENT_H */ diff --git a/arch/riscv/include/asm/pgtable.h b/arch/riscv/include/asm/pgtable.h index 2258b27173b02658209c5f260c051fe961d99fbd..75970ee2bda223be83a567adab6fa327e8e58cd1 100644 --- a/arch/riscv/include/asm/pgtable.h +++ b/arch/riscv/include/asm/pgtable.h @@ -165,8 +165,7 @@ extern struct pt_alloc_ops pt_ops __initdata; _PAGE_EXEC | _PAGE_WRITE) #define PAGE_COPY PAGE_READ -#define PAGE_COPY_EXEC PAGE_EXEC -#define PAGE_COPY_READ_EXEC PAGE_READ_EXEC +#define PAGE_COPY_EXEC PAGE_READ_EXEC #define PAGE_SHARED PAGE_WRITE #define PAGE_SHARED_EXEC PAGE_WRITE_EXEC diff --git a/arch/riscv/kernel/Makefile b/arch/riscv/kernel/Makefile index fbdccc21418a5b9bf0d53de537089b879a04a429..153864e4f3996dc3405a28fc024e0d76f0232028 100644 --- a/arch/riscv/kernel/Makefile +++ b/arch/riscv/kernel/Makefile @@ -23,6 +23,10 @@ ifdef CONFIG_FTRACE CFLAGS_REMOVE_alternative.o = $(CC_FLAGS_FTRACE) CFLAGS_REMOVE_cpufeature.o = $(CC_FLAGS_FTRACE) endif +ifdef CONFIG_RELOCATABLE +CFLAGS_alternative.o += -fno-pie +CFLAGS_cpufeature.o += -fno-pie +endif ifdef CONFIG_KASAN KASAN_SANITIZE_alternative.o := n KASAN_SANITIZE_cpufeature.o := n diff --git a/arch/riscv/kernel/probes/Makefile b/arch/riscv/kernel/probes/Makefile index c40139e9ca4781e414946bd9e58648f15c89d25a..8265ff497977cffb5559f25a2bd6adcce2eb1a60 100644 --- a/arch/riscv/kernel/probes/Makefile +++ b/arch/riscv/kernel/probes/Makefile @@ -4,3 +4,5 @@ obj-$(CONFIG_RETHOOK) += rethook.o rethook_trampoline.o obj-$(CONFIG_KPROBES_ON_FTRACE) += ftrace.o obj-$(CONFIG_UPROBES) += uprobes.o decode-insn.o simulate-insn.o CFLAGS_REMOVE_simulate-insn.o = $(CC_FLAGS_FTRACE) +CFLAGS_REMOVE_rethook.o = $(CC_FLAGS_FTRACE) +CFLAGS_REMOVE_rethook_trampoline.o = $(CC_FLAGS_FTRACE) diff --git a/arch/riscv/mm/hugetlbpage.c b/arch/riscv/mm/hugetlbpage.c index a163a3e0f0d46b92565e85318c105b9cf0891c29..e0ef56dc57b90a31a9ef95072f4360019ac84ba1 100644 --- a/arch/riscv/mm/hugetlbpage.c +++ b/arch/riscv/mm/hugetlbpage.c @@ -3,6 +3,30 @@ #include #ifdef CONFIG_RISCV_ISA_SVNAPOT +pte_t huge_ptep_get(pte_t *ptep) +{ + unsigned long pte_num; + int i; + pte_t orig_pte = ptep_get(ptep); + + if (!pte_present(orig_pte) || !pte_napot(orig_pte)) + return orig_pte; + + pte_num = napot_pte_num(napot_cont_order(orig_pte)); + + for (i = 0; i < pte_num; i++, ptep++) { + pte_t pte = ptep_get(ptep); + + if (pte_dirty(pte)) + orig_pte = pte_mkdirty(orig_pte); + + if (pte_young(pte)) + orig_pte = pte_mkyoung(orig_pte); + } + + return orig_pte; +} + pte_t *huge_pte_alloc(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long addr, @@ -218,6 +242,7 @@ void huge_ptep_set_wrprotect(struct mm_struct *mm, { pte_t pte = ptep_get(ptep); unsigned long order; + pte_t orig_pte; int i, pte_num; if (!pte_napot(pte)) { @@ -228,9 +253,12 @@ void huge_ptep_set_wrprotect(struct mm_struct *mm, order = napot_cont_order(pte); pte_num = napot_pte_num(order); ptep = huge_pte_offset(mm, addr, napot_cont_size(order)); + orig_pte = get_clear_contig_flush(mm, addr, ptep, pte_num); + + orig_pte = pte_wrprotect(orig_pte); for (i = 0; i < pte_num; i++, addr += PAGE_SIZE, ptep++) - ptep_set_wrprotect(mm, addr, ptep); + set_pte_at(mm, addr, ptep, orig_pte); } pte_t huge_ptep_clear_flush(struct vm_area_struct *vma, diff --git a/arch/riscv/mm/init.c b/arch/riscv/mm/init.c index 747e5b1ef02d35ebbc2fdaed3f6af793f0d2a5b5..4fa420faa780899b4802fc12eba567d03cd458e5 100644 --- a/arch/riscv/mm/init.c +++ b/arch/riscv/mm/init.c @@ -23,6 +23,7 @@ #ifdef CONFIG_RELOCATABLE #include #endif +#include #include #include @@ -293,7 +294,7 @@ static const pgprot_t protection_map[16] = { [VM_EXEC] = PAGE_EXEC, [VM_EXEC | VM_READ] = PAGE_READ_EXEC, [VM_EXEC | VM_WRITE] = PAGE_COPY_EXEC, - [VM_EXEC | VM_WRITE | VM_READ] = PAGE_COPY_READ_EXEC, + [VM_EXEC | VM_WRITE | VM_READ] = PAGE_COPY_EXEC, [VM_SHARED] = PAGE_NONE, [VM_SHARED | VM_READ] = PAGE_READ, [VM_SHARED | VM_WRITE] = PAGE_SHARED, @@ -659,18 +660,19 @@ void __init create_pgd_mapping(pgd_t *pgdp, create_pgd_next_mapping(nextp, va, pa, sz, prot); } -static uintptr_t __init best_map_size(phys_addr_t base, phys_addr_t size) +static uintptr_t __init best_map_size(phys_addr_t pa, uintptr_t va, + phys_addr_t size) { - if (!(base & (PGDIR_SIZE - 1)) && size >= PGDIR_SIZE) + if (!(pa & (PGDIR_SIZE - 1)) && !(va & (PGDIR_SIZE - 1)) && size >= PGDIR_SIZE) return PGDIR_SIZE; - if (!(base & (P4D_SIZE - 1)) && size >= P4D_SIZE) + if (!(pa & (P4D_SIZE - 1)) && !(va & (P4D_SIZE - 1)) && size >= P4D_SIZE) return P4D_SIZE; - if (!(base & (PUD_SIZE - 1)) && size >= PUD_SIZE) + if (!(pa & (PUD_SIZE - 1)) && !(va & (PUD_SIZE - 1)) && size >= PUD_SIZE) return PUD_SIZE; - if (!(base & (PMD_SIZE - 1)) && size >= PMD_SIZE) + if (!(pa & (PMD_SIZE - 1)) && !(va & (PMD_SIZE - 1)) && size >= PMD_SIZE) return PMD_SIZE; return PAGE_SIZE; @@ -922,9 +924,9 @@ static void __init create_kernel_page_table(pgd_t *pgdir, bool early) static void __init create_fdt_early_page_table(uintptr_t fix_fdt_va, uintptr_t dtb_pa) { +#ifndef CONFIG_BUILTIN_DTB uintptr_t pa = dtb_pa & ~(PMD_SIZE - 1); -#ifndef CONFIG_BUILTIN_DTB /* Make sure the fdt fixmap address is always aligned on PMD size */ BUILD_BUG_ON(FIX_FDT % (PMD_SIZE / PAGE_SIZE)); @@ -1167,14 +1169,16 @@ asmlinkage void __init setup_vm(uintptr_t dtb_pa) } static void __init create_linear_mapping_range(phys_addr_t start, - phys_addr_t end) + phys_addr_t end, + uintptr_t fixed_map_size) { phys_addr_t pa; uintptr_t va, map_size; for (pa = start; pa < end; pa += map_size) { va = (uintptr_t)__va(pa); - map_size = best_map_size(pa, end - pa); + map_size = fixed_map_size ? fixed_map_size : + best_map_size(pa, va, end - pa); create_pgd_mapping(swapper_pg_dir, va, pa, map_size, pgprot_from_va(va)); @@ -1184,6 +1188,7 @@ static void __init create_linear_mapping_range(phys_addr_t start, static void __init create_linear_mapping_page_table(void) { phys_addr_t start, end; + phys_addr_t kfence_pool __maybe_unused; u64 i; #ifdef CONFIG_STRICT_KERNEL_RWX @@ -1197,6 +1202,19 @@ static void __init create_linear_mapping_page_table(void) memblock_mark_nomap(krodata_start, krodata_size); #endif +#ifdef CONFIG_KFENCE + /* + * kfence pool must be backed by PAGE_SIZE mappings, so allocate it + * before we setup the linear mapping so that we avoid using hugepages + * for this region. + */ + kfence_pool = memblock_phys_alloc(KFENCE_POOL_SIZE, PAGE_SIZE); + BUG_ON(!kfence_pool); + + memblock_mark_nomap(kfence_pool, KFENCE_POOL_SIZE); + __kfence_pool = __va(kfence_pool); +#endif + /* Map all memory banks in the linear mapping */ for_each_mem_range(i, &start, &end) { if (start >= end) @@ -1207,17 +1225,25 @@ static void __init create_linear_mapping_page_table(void) if (end >= __pa(PAGE_OFFSET) + memory_limit) end = __pa(PAGE_OFFSET) + memory_limit; - create_linear_mapping_range(start, end); + create_linear_mapping_range(start, end, 0); } #ifdef CONFIG_STRICT_KERNEL_RWX - create_linear_mapping_range(ktext_start, ktext_start + ktext_size); + create_linear_mapping_range(ktext_start, ktext_start + ktext_size, 0); create_linear_mapping_range(krodata_start, - krodata_start + krodata_size); + krodata_start + krodata_size, 0); memblock_clear_nomap(ktext_start, ktext_size); memblock_clear_nomap(krodata_start, krodata_size); #endif + +#ifdef CONFIG_KFENCE + create_linear_mapping_range(kfence_pool, + kfence_pool + KFENCE_POOL_SIZE, + PAGE_SIZE); + + memblock_clear_nomap(kfence_pool, KFENCE_POOL_SIZE); +#endif } static void __init setup_vm_final(void) diff --git a/arch/riscv/purgatory/Makefile b/arch/riscv/purgatory/Makefile index 5730797a6b402c39c96cd0360486b3548631b389..bd2e27f82532681ff124a01b28c25dd346347afe 100644 --- a/arch/riscv/purgatory/Makefile +++ b/arch/riscv/purgatory/Makefile @@ -35,6 +35,11 @@ CFLAGS_sha256.o := -D__DISABLE_EXPORTS CFLAGS_string.o := -D__DISABLE_EXPORTS CFLAGS_ctype.o := -D__DISABLE_EXPORTS +# When profile-guided optimization is enabled, llvm emits two different +# overlapping text sections, which is not supported by kexec. Remove profile +# optimization flags. +KBUILD_CFLAGS := $(filter-out -fprofile-sample-use=% -fprofile-use=%,$(KBUILD_CFLAGS)) + # When linking purgatory.ro with -r unresolved symbols are not checked, # also link a purgatory.chk binary without -r to check for unresolved symbols. PURGATORY_LDFLAGS := -e purgatory_start -z nodefaultlib diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index db20c1589a98f3905d57f452ff19fc53380a45c6..6dab9c1be508b9138ef7dde50db6db2ef9ff5a3a 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -469,19 +469,11 @@ config SCHED_SMT config SCHED_MC def_bool n -config SCHED_BOOK - def_bool n - -config SCHED_DRAWER - def_bool n - config SCHED_TOPOLOGY def_bool y prompt "Topology scheduler support" select SCHED_SMT select SCHED_MC - select SCHED_BOOK - select SCHED_DRAWER help Topology scheduler support improves the CPU scheduler's decision making when dealing with machines that have multi-threading, @@ -716,7 +708,6 @@ config EADM_SCH config VFIO_CCW def_tristate n prompt "Support for VFIO-CCW subchannels" - depends on S390_CCW_IOMMU depends on VFIO select VFIO_MDEV help @@ -728,7 +719,7 @@ config VFIO_CCW config VFIO_AP def_tristate n prompt "VFIO support for AP devices" - depends on S390_AP_IOMMU && KVM + depends on KVM depends on VFIO depends on ZCRYPT select VFIO_MDEV diff --git a/arch/s390/configs/debug_defconfig b/arch/s390/configs/debug_defconfig index 4ccf66d29fc24b9cad9425fc0fcb9a01bfe4b3c0..be3bf03bf361850d9a0abd5cdb6644878ebec6a4 100644 --- a/arch/s390/configs/debug_defconfig +++ b/arch/s390/configs/debug_defconfig @@ -591,8 +591,6 @@ CONFIG_VIRTIO_BALLOON=m CONFIG_VIRTIO_INPUT=y CONFIG_VHOST_NET=m CONFIG_VHOST_VSOCK=m -CONFIG_S390_CCW_IOMMU=y -CONFIG_S390_AP_IOMMU=y CONFIG_EXT4_FS=y CONFIG_EXT4_FS_POSIX_ACL=y CONFIG_EXT4_FS_SECURITY=y @@ -703,6 +701,7 @@ CONFIG_IMA_DEFAULT_HASH_SHA256=y CONFIG_IMA_WRITE_POLICY=y CONFIG_IMA_APPRAISE=y CONFIG_LSM="yama,loadpin,safesetid,integrity,selinux,smack,tomoyo,apparmor" +CONFIG_INIT_STACK_NONE=y CONFIG_CRYPTO_USER=m # CONFIG_CRYPTO_MANAGER_DISABLE_TESTS is not set CONFIG_CRYPTO_PCRYPT=m diff --git a/arch/s390/configs/defconfig b/arch/s390/configs/defconfig index 693297a2e89733d888c30e78e9d95f981c6db015..769c7eed8b6ac67f01ac781b6eac7721c21ce3b8 100644 --- a/arch/s390/configs/defconfig +++ b/arch/s390/configs/defconfig @@ -580,8 +580,6 @@ CONFIG_VIRTIO_BALLOON=m CONFIG_VIRTIO_INPUT=y CONFIG_VHOST_NET=m CONFIG_VHOST_VSOCK=m -CONFIG_S390_CCW_IOMMU=y -CONFIG_S390_AP_IOMMU=y CONFIG_EXT4_FS=y CONFIG_EXT4_FS_POSIX_ACL=y CONFIG_EXT4_FS_SECURITY=y @@ -686,6 +684,7 @@ CONFIG_IMA_DEFAULT_HASH_SHA256=y CONFIG_IMA_WRITE_POLICY=y CONFIG_IMA_APPRAISE=y CONFIG_LSM="yama,loadpin,safesetid,integrity,selinux,smack,tomoyo,apparmor" +CONFIG_INIT_STACK_NONE=y CONFIG_CRYPTO_FIPS=y CONFIG_CRYPTO_USER=m # CONFIG_CRYPTO_MANAGER_DISABLE_TESTS is not set diff --git a/arch/s390/configs/zfcpdump_defconfig b/arch/s390/configs/zfcpdump_defconfig index 33a232bb68af95b460eb098136fbdd23e3045751..6f68b39817effcfc5f5a9612cd52ec465630be53 100644 --- a/arch/s390/configs/zfcpdump_defconfig +++ b/arch/s390/configs/zfcpdump_defconfig @@ -67,6 +67,7 @@ CONFIG_ZFCP=y # CONFIG_MISC_FILESYSTEMS is not set # CONFIG_NETWORK_FILESYSTEMS is not set CONFIG_LSM="yama,loadpin,safesetid,integrity" +CONFIG_INIT_STACK_NONE=y # CONFIG_ZLIB_DFLTCC is not set CONFIG_XZ_DEC_MICROLZMA=y CONFIG_PRINTK_TIME=y diff --git a/arch/s390/crypto/chacha-glue.c b/arch/s390/crypto/chacha-glue.c index 7752bd314558e50ce6c0142c2472cb60297799f7..5fae187f947a08325ccb19ff792118fdcc661277 100644 --- a/arch/s390/crypto/chacha-glue.c +++ b/arch/s390/crypto/chacha-glue.c @@ -82,7 +82,7 @@ void chacha_crypt_arch(u32 *state, u8 *dst, const u8 *src, * it cannot handle a block of data or less, but otherwise * it can handle data of arbitrary size */ - if (bytes <= CHACHA_BLOCK_SIZE || nrounds != 20) + if (bytes <= CHACHA_BLOCK_SIZE || nrounds != 20 || !MACHINE_HAS_VX) chacha_crypt_generic(state, dst, src, bytes, nrounds); else chacha20_crypt_s390(state, dst, src, bytes, diff --git a/arch/s390/include/asm/compat.h b/arch/s390/include/asm/compat.h index a386070f1d5653a97bf2aa075b21c23e83eac436..3cb9d813f022b9b8e565bca04f9ec1e805452e2f 100644 --- a/arch/s390/include/asm/compat.h +++ b/arch/s390/include/asm/compat.h @@ -112,7 +112,7 @@ struct compat_statfs64 { u32 f_namelen; u32 f_frsize; u32 f_flags; - u32 f_spare[4]; + u32 f_spare[5]; }; /* diff --git a/arch/s390/include/uapi/asm/statfs.h b/arch/s390/include/uapi/asm/statfs.h index 72604f7792c33642aed1907bb0a69effd8b06991..f85b50723dd355f04a66e3ef5ac07bf89de0c6ff 100644 --- a/arch/s390/include/uapi/asm/statfs.h +++ b/arch/s390/include/uapi/asm/statfs.h @@ -30,7 +30,7 @@ struct statfs { unsigned int f_namelen; unsigned int f_frsize; unsigned int f_flags; - unsigned int f_spare[4]; + unsigned int f_spare[5]; }; struct statfs64 { @@ -45,7 +45,7 @@ struct statfs64 { unsigned int f_namelen; unsigned int f_frsize; unsigned int f_flags; - unsigned int f_spare[4]; + unsigned int f_spare[5]; }; #endif diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile index 8983837b3565914b19451546d7b5a1ff61511c95..6b2a051e1f8a2a7fec822ab5b287a9b25fd33a84 100644 --- a/arch/s390/kernel/Makefile +++ b/arch/s390/kernel/Makefile @@ -10,6 +10,7 @@ CFLAGS_REMOVE_ftrace.o = $(CC_FLAGS_FTRACE) # Do not trace early setup code CFLAGS_REMOVE_early.o = $(CC_FLAGS_FTRACE) +CFLAGS_REMOVE_rethook.o = $(CC_FLAGS_FTRACE) endif diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c index 43de939b7af199757646a3ea82347897f6dd9980..f44f70de966115c64aa6286afb9bfe06f2543dc6 100644 --- a/arch/s390/kernel/ipl.c +++ b/arch/s390/kernel/ipl.c @@ -1935,14 +1935,13 @@ static struct shutdown_action __refdata dump_action = { static void dump_reipl_run(struct shutdown_trigger *trigger) { - unsigned long ipib = (unsigned long) reipl_block_actual; struct lowcore *abs_lc; unsigned int csum; csum = (__force unsigned int) csum_partial(reipl_block_actual, reipl_block_actual->hdr.len, 0); abs_lc = get_abs_lowcore(); - abs_lc->ipib = ipib; + abs_lc->ipib = __pa(reipl_block_actual); abs_lc->ipib_checksum = csum; put_abs_lowcore(abs_lc); dump_run(trigger); diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c index 9fd19530c9a5623d5901abb1dd9e2032238203ba..68adf1de8888b25fbfaea0a45537e8cd4fb21c2f 100644 --- a/arch/s390/kernel/topology.c +++ b/arch/s390/kernel/topology.c @@ -95,7 +95,7 @@ static void cpu_group_map(cpumask_t *dst, struct mask_info *info, unsigned int c static void cpu_thread_map(cpumask_t *dst, unsigned int cpu) { static cpumask_t mask; - int i; + unsigned int max_cpu; cpumask_clear(&mask); if (!cpumask_test_cpu(cpu, &cpu_setup_mask)) @@ -104,9 +104,10 @@ static void cpu_thread_map(cpumask_t *dst, unsigned int cpu) if (topology_mode != TOPOLOGY_MODE_HW) goto out; cpu -= cpu % (smp_cpu_mtid + 1); - for (i = 0; i <= smp_cpu_mtid; i++) { - if (cpumask_test_cpu(cpu + i, &cpu_setup_mask)) - cpumask_set_cpu(cpu + i, &mask); + max_cpu = min(cpu + smp_cpu_mtid, nr_cpu_ids - 1); + for (; cpu <= max_cpu; cpu++) { + if (cpumask_test_cpu(cpu, &cpu_setup_mask)) + cpumask_set_cpu(cpu, &mask); } out: cpumask_copy(dst, &mask); @@ -123,25 +124,26 @@ static void add_cpus_to_mask(struct topology_core *tl_core, unsigned int core; for_each_set_bit(core, &tl_core->mask, TOPOLOGY_CORE_BITS) { - unsigned int rcore; - int lcpu, i; + unsigned int max_cpu, rcore; + int cpu; rcore = TOPOLOGY_CORE_BITS - 1 - core + tl_core->origin; - lcpu = smp_find_processor_id(rcore << smp_cpu_mt_shift); - if (lcpu < 0) + cpu = smp_find_processor_id(rcore << smp_cpu_mt_shift); + if (cpu < 0) continue; - for (i = 0; i <= smp_cpu_mtid; i++) { - topo = &cpu_topology[lcpu + i]; + max_cpu = min(cpu + smp_cpu_mtid, nr_cpu_ids - 1); + for (; cpu <= max_cpu; cpu++) { + topo = &cpu_topology[cpu]; topo->drawer_id = drawer->id; topo->book_id = book->id; topo->socket_id = socket->id; topo->core_id = rcore; - topo->thread_id = lcpu + i; + topo->thread_id = cpu; topo->dedicated = tl_core->d; - cpumask_set_cpu(lcpu + i, &drawer->mask); - cpumask_set_cpu(lcpu + i, &book->mask); - cpumask_set_cpu(lcpu + i, &socket->mask); - smp_cpu_set_polarization(lcpu + i, tl_core->pp); + cpumask_set_cpu(cpu, &drawer->mask); + cpumask_set_cpu(cpu, &book->mask); + cpumask_set_cpu(cpu, &socket->mask); + smp_cpu_set_polarization(cpu, tl_core->pp); } } } diff --git a/arch/s390/purgatory/Makefile b/arch/s390/purgatory/Makefile index 32573b4f9bd20b13eba7bb51220e344dcf238178..cc8cf5abea158c63449fb130bd3f369a10884e5d 100644 --- a/arch/s390/purgatory/Makefile +++ b/arch/s390/purgatory/Makefile @@ -26,6 +26,7 @@ KBUILD_CFLAGS += -Wno-pointer-sign -Wno-sign-compare KBUILD_CFLAGS += -fno-zero-initialized-in-bss -fno-builtin -ffreestanding KBUILD_CFLAGS += -Os -m64 -msoft-float -fno-common KBUILD_CFLAGS += -fno-stack-protector +KBUILD_CFLAGS += -DDISABLE_BRANCH_PROFILING KBUILD_CFLAGS += $(CLANG_FLAGS) KBUILD_CFLAGS += $(call cc-option,-fno-PIE) KBUILD_AFLAGS := $(filter-out -DCC_USING_EXPOLINE,$(KBUILD_AFLAGS)) diff --git a/arch/um/drivers/Makefile b/arch/um/drivers/Makefile index dee6f66353b334c6b55de0097409f2ce11cb7a26..a461a950f0518d55fc5ad95175520c6b45e8efe4 100644 --- a/arch/um/drivers/Makefile +++ b/arch/um/drivers/Makefile @@ -16,7 +16,8 @@ mconsole-objs := mconsole_kern.o mconsole_user.o hostaudio-objs := hostaudio_kern.o ubd-objs := ubd_kern.o ubd_user.o port-objs := port_kern.o port_user.o -harddog-objs := harddog_kern.o harddog_user.o +harddog-objs := harddog_kern.o +harddog-builtin-$(CONFIG_UML_WATCHDOG) := harddog_user.o harddog_user_exp.o rtc-objs := rtc_kern.o rtc_user.o LDFLAGS_pcap.o = $(shell $(CC) $(KBUILD_CFLAGS) -print-file-name=libpcap.a) @@ -60,6 +61,7 @@ obj-$(CONFIG_PTY_CHAN) += pty.o obj-$(CONFIG_TTY_CHAN) += tty.o obj-$(CONFIG_XTERM_CHAN) += xterm.o xterm_kern.o obj-$(CONFIG_UML_WATCHDOG) += harddog.o +obj-y += $(harddog-builtin-y) $(harddog-builtin-m) obj-$(CONFIG_BLK_DEV_COW_COMMON) += cow_user.o obj-$(CONFIG_UML_RANDOM) += random.o obj-$(CONFIG_VIRTIO_UML) += virtio_uml.o diff --git a/arch/um/drivers/harddog.h b/arch/um/drivers/harddog.h new file mode 100644 index 0000000000000000000000000000000000000000..6d9ea60e7133e8598ae732888ad5bf77ea944458 --- /dev/null +++ b/arch/um/drivers/harddog.h @@ -0,0 +1,9 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef UM_WATCHDOG_H +#define UM_WATCHDOG_H + +int start_watchdog(int *in_fd_ret, int *out_fd_ret, char *sock); +void stop_watchdog(int in_fd, int out_fd); +int ping_watchdog(int fd); + +#endif /* UM_WATCHDOG_H */ diff --git a/arch/um/drivers/harddog_kern.c b/arch/um/drivers/harddog_kern.c index e6d4f43deba82a77e20eaf15ea81639f0ddb37d1..60d1c6cab8a95197e2c14cc789ca5a8f9fa47496 100644 --- a/arch/um/drivers/harddog_kern.c +++ b/arch/um/drivers/harddog_kern.c @@ -47,6 +47,7 @@ #include #include #include "mconsole.h" +#include "harddog.h" MODULE_LICENSE("GPL"); @@ -60,8 +61,6 @@ static int harddog_out_fd = -1; * Allow only one person to hold it open */ -extern int start_watchdog(int *in_fd_ret, int *out_fd_ret, char *sock); - static int harddog_open(struct inode *inode, struct file *file) { int err = -EBUSY; @@ -92,8 +91,6 @@ static int harddog_open(struct inode *inode, struct file *file) return err; } -extern void stop_watchdog(int in_fd, int out_fd); - static int harddog_release(struct inode *inode, struct file *file) { /* @@ -112,8 +109,6 @@ static int harddog_release(struct inode *inode, struct file *file) return 0; } -extern int ping_watchdog(int fd); - static ssize_t harddog_write(struct file *file, const char __user *data, size_t len, loff_t *ppos) { diff --git a/arch/um/drivers/harddog_user.c b/arch/um/drivers/harddog_user.c index 070468d22e3943f4d047aa8917776ee7f87b43db..9ed89304975ed351b264d5c068b70fd0094f44b2 100644 --- a/arch/um/drivers/harddog_user.c +++ b/arch/um/drivers/harddog_user.c @@ -7,6 +7,7 @@ #include #include #include +#include "harddog.h" struct dog_data { int stdin_fd; diff --git a/arch/um/drivers/harddog_user_exp.c b/arch/um/drivers/harddog_user_exp.c new file mode 100644 index 0000000000000000000000000000000000000000..c74d4b815d143d27e96f13672fe073faa5608cd0 --- /dev/null +++ b/arch/um/drivers/harddog_user_exp.c @@ -0,0 +1,9 @@ +// SPDX-License-Identifier: GPL-2.0 +#include +#include "harddog.h" + +#if IS_MODULE(CONFIG_UML_WATCHDOG) +EXPORT_SYMBOL(start_watchdog); +EXPORT_SYMBOL(stop_watchdog); +EXPORT_SYMBOL(ping_watchdog); +#endif diff --git a/arch/x86/crypto/aria-aesni-avx-asm_64.S b/arch/x86/crypto/aria-aesni-avx-asm_64.S index 7c1abc513f34621eff4d0ee72db6a6c3dbb291f5..9556dacd984154a2b6ede8808321b8c6c692f347 100644 --- a/arch/x86/crypto/aria-aesni-avx-asm_64.S +++ b/arch/x86/crypto/aria-aesni-avx-asm_64.S @@ -773,8 +773,6 @@ .octa 0x3F893781E95FE1576CDA64D2BA0CB204 #ifdef CONFIG_AS_GFNI -.section .rodata.cst8, "aM", @progbits, 8 -.align 8 /* AES affine: */ #define tf_aff_const BV8(1, 1, 0, 0, 0, 1, 1, 0) .Ltf_aff_bitmatrix: diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index 070cc4ef267210b4dc7a490f43cad45d5d3ab5c7..89b9c1cebb613a89865909cfe7e322f93f5abe10 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -4074,7 +4074,7 @@ static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr, void *data) if (x86_pmu.intel_cap.pebs_baseline) { arr[(*nr)++] = (struct perf_guest_switch_msr){ .msr = MSR_PEBS_DATA_CFG, - .host = cpuc->pebs_data_cfg, + .host = cpuc->active_pebs_data_cfg, .guest = kvm_pmu->pebs_data_cfg, }; } diff --git a/arch/x86/events/intel/uncore_snbep.c b/arch/x86/events/intel/uncore_snbep.c index fa9b209a11fabce2f57d585e9e08219182d947dd..d49e90dc04a4ccac259485982384f994c728c122 100644 --- a/arch/x86/events/intel/uncore_snbep.c +++ b/arch/x86/events/intel/uncore_snbep.c @@ -6150,6 +6150,7 @@ static struct intel_uncore_type spr_uncore_mdf = { }; #define UNCORE_SPR_NUM_UNCORE_TYPES 12 +#define UNCORE_SPR_CHA 0 #define UNCORE_SPR_IIO 1 #define UNCORE_SPR_IMC 6 #define UNCORE_SPR_UPI 8 @@ -6460,12 +6461,22 @@ static int uncore_type_max_boxes(struct intel_uncore_type **types, return max + 1; } +#define SPR_MSR_UNC_CBO_CONFIG 0x2FFE + void spr_uncore_cpu_init(void) { + struct intel_uncore_type *type; + u64 num_cbo; + uncore_msr_uncores = uncore_get_uncores(UNCORE_ACCESS_MSR, UNCORE_SPR_MSR_EXTRA_UNCORES, spr_msr_uncores); + type = uncore_find_type_by_id(uncore_msr_uncores, UNCORE_SPR_CHA); + if (type) { + rdmsrl(SPR_MSR_UNC_CBO_CONFIG, num_cbo); + type->num_boxes = num_cbo; + } spr_uncore_iio_free_running.num_boxes = uncore_type_max_boxes(uncore_msr_uncores, UNCORE_SPR_IIO); } diff --git a/arch/x86/include/asm/fpu/sched.h b/arch/x86/include/asm/fpu/sched.h index c2d6cd78ed0c2986181217a184e8d52f7e72c534..78fcde7b1f0700ee1e12cba4871edf7d3efed5f1 100644 --- a/arch/x86/include/asm/fpu/sched.h +++ b/arch/x86/include/asm/fpu/sched.h @@ -39,7 +39,7 @@ extern void fpu_flush_thread(void); static inline void switch_fpu_prepare(struct fpu *old_fpu, int cpu) { if (cpu_feature_enabled(X86_FEATURE_FPU) && - !(current->flags & (PF_KTHREAD | PF_IO_WORKER))) { + !(current->flags & (PF_KTHREAD | PF_USER_WORKER))) { save_fpregs_to_fpstate(old_fpu); /* * The save operation preserved register state, so the diff --git a/arch/x86/include/asm/vmx.h b/arch/x86/include/asm/vmx.h index 498dc600bd5c8ef55a4de29ecb87396d4d192859..0d02c4aafa6f49b2e981ac975610095844a815d3 100644 --- a/arch/x86/include/asm/vmx.h +++ b/arch/x86/include/asm/vmx.h @@ -13,7 +13,9 @@ #include +#include #include + #include #include diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index dd61752f4c966bf0db57abb92472f8c4ae4c0c82..4070a01c11b72af04f429dee966c495275a7dfb6 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -17,6 +17,7 @@ CFLAGS_REMOVE_ftrace.o = -pg CFLAGS_REMOVE_early_printk.o = -pg CFLAGS_REMOVE_head64.o = -pg CFLAGS_REMOVE_sev.o = -pg +CFLAGS_REMOVE_rethook.o = -pg endif KASAN_SANITIZE_head$(BITS).o := n diff --git a/arch/x86/kernel/cpu/topology.c b/arch/x86/kernel/cpu/topology.c index 5e868b62a7c4e4c466fadd8f858e004a9ba76e27..0270925fe013bcf9bbe76e589c93a35a6ddbfde5 100644 --- a/arch/x86/kernel/cpu/topology.c +++ b/arch/x86/kernel/cpu/topology.c @@ -79,7 +79,7 @@ int detect_extended_topology_early(struct cpuinfo_x86 *c) * initial apic id, which also represents 32-bit extended x2apic id. */ c->initial_apicid = edx; - smp_num_siblings = LEVEL_MAX_SIBLINGS(ebx); + smp_num_siblings = max_t(int, smp_num_siblings, LEVEL_MAX_SIBLINGS(ebx)); #endif return 0; } @@ -109,7 +109,8 @@ int detect_extended_topology(struct cpuinfo_x86 *c) */ cpuid_count(leaf, SMT_LEVEL, &eax, &ebx, &ecx, &edx); c->initial_apicid = edx; - core_level_siblings = smp_num_siblings = LEVEL_MAX_SIBLINGS(ebx); + core_level_siblings = LEVEL_MAX_SIBLINGS(ebx); + smp_num_siblings = max_t(int, smp_num_siblings, LEVEL_MAX_SIBLINGS(ebx)); core_plus_mask_width = ht_mask_width = BITS_SHIFT_NEXT_LEVEL(eax); die_level_siblings = LEVEL_MAX_SIBLINGS(ebx); pkg_mask_width = die_plus_mask_width = BITS_SHIFT_NEXT_LEVEL(eax); diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c index 0bf6779187dda158a8deba21e14215d2b0cbddb7..f18ca44c904b747e68a3fdf0f7926a7025da71b7 100644 --- a/arch/x86/kernel/dumpstack.c +++ b/arch/x86/kernel/dumpstack.c @@ -195,7 +195,6 @@ static void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, printk("%sCall Trace:\n", log_lvl); unwind_start(&state, task, regs, stack); - stack = stack ? : get_stack_pointer(task, regs); regs = unwind_get_entry_regs(&state, &partial); /* @@ -214,9 +213,13 @@ static void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, * - hardirq stack * - entry stack */ - for ( ; stack; stack = PTR_ALIGN(stack_info.next_sp, sizeof(long))) { + for (stack = stack ?: get_stack_pointer(task, regs); + stack; + stack = stack_info.next_sp) { const char *stack_name; + stack = PTR_ALIGN(stack, sizeof(long)); + if (get_stack_info(stack, task, &stack_info, &visit_mask)) { /* * We weren't on a valid stack. It's possible that diff --git a/arch/x86/kernel/fpu/context.h b/arch/x86/kernel/fpu/context.h index 9fcfa5c4dad799c99cafff1a0b6c4b9ad958af74..af5cbdd9bd29a2b496d349a439071b75b88f8d61 100644 --- a/arch/x86/kernel/fpu/context.h +++ b/arch/x86/kernel/fpu/context.h @@ -57,7 +57,7 @@ static inline void fpregs_restore_userregs(void) struct fpu *fpu = ¤t->thread.fpu; int cpu = smp_processor_id(); - if (WARN_ON_ONCE(current->flags & (PF_KTHREAD | PF_IO_WORKER))) + if (WARN_ON_ONCE(current->flags & (PF_KTHREAD | PF_USER_WORKER))) return; if (!fpregs_state_valid(fpu, cpu)) { diff --git a/arch/x86/kernel/fpu/core.c b/arch/x86/kernel/fpu/core.c index caf33486dc5ee5b39bc4bfb1032d233cc1f0e09d..1015af1ae562bff33925f2e3266517d8c365296c 100644 --- a/arch/x86/kernel/fpu/core.c +++ b/arch/x86/kernel/fpu/core.c @@ -426,7 +426,7 @@ void kernel_fpu_begin_mask(unsigned int kfpu_mask) this_cpu_write(in_kernel_fpu, true); - if (!(current->flags & (PF_KTHREAD | PF_IO_WORKER)) && + if (!(current->flags & (PF_KTHREAD | PF_USER_WORKER)) && !test_thread_flag(TIF_NEED_FPU_LOAD)) { set_thread_flag(TIF_NEED_FPU_LOAD); save_fpregs_to_fpstate(¤t->thread.fpu); diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S index a5df3e994f04f10f2c9fb814bf6f952a6452ae7a..113c13376e5125c05212dc1808ff44c9bf9df43c 100644 --- a/arch/x86/kernel/head_64.S +++ b/arch/x86/kernel/head_64.S @@ -77,6 +77,15 @@ SYM_CODE_START_NOALIGN(startup_64) call startup_64_setup_env popq %rsi + /* Now switch to __KERNEL_CS so IRET works reliably */ + pushq $__KERNEL_CS + leaq .Lon_kernel_cs(%rip), %rax + pushq %rax + lretq + +.Lon_kernel_cs: + UNWIND_HINT_END_OF_STACK + #ifdef CONFIG_AMD_MEM_ENCRYPT /* * Activate SEV/SME memory encryption if supported/enabled. This needs to @@ -90,15 +99,6 @@ SYM_CODE_START_NOALIGN(startup_64) popq %rsi #endif - /* Now switch to __KERNEL_CS so IRET works reliably */ - pushq $__KERNEL_CS - leaq .Lon_kernel_cs(%rip), %rax - pushq %rax - lretq - -.Lon_kernel_cs: - UNWIND_HINT_END_OF_STACK - /* Sanitize CPU configuration */ call verify_cpu diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c index 123bf8b97a4b21098a8271320f711cecf2eae5ec..0c9660a07b233ff404147fa6acda76668241f7ee 100644 --- a/arch/x86/kvm/cpuid.c +++ b/arch/x86/kvm/cpuid.c @@ -253,7 +253,6 @@ static void __kvm_update_cpuid_runtime(struct kvm_vcpu *vcpu, struct kvm_cpuid_e int nent) { struct kvm_cpuid_entry2 *best; - u64 guest_supported_xcr0 = cpuid_get_supported_xcr0(entries, nent); best = cpuid_entry2_find(entries, nent, 1, KVM_CPUID_INDEX_NOT_SIGNIFICANT); if (best) { @@ -292,21 +291,6 @@ static void __kvm_update_cpuid_runtime(struct kvm_vcpu *vcpu, struct kvm_cpuid_e vcpu->arch.ia32_misc_enable_msr & MSR_IA32_MISC_ENABLE_MWAIT); } - - /* - * Bits 127:0 of the allowed SECS.ATTRIBUTES (CPUID.0x12.0x1) enumerate - * the supported XSAVE Feature Request Mask (XFRM), i.e. the enclave's - * requested XCR0 value. The enclave's XFRM must be a subset of XCRO - * at the time of EENTER, thus adjust the allowed XFRM by the guest's - * supported XCR0. Similar to XCR0 handling, FP and SSE are forced to - * '1' even on CPUs that don't support XSAVE. - */ - best = cpuid_entry2_find(entries, nent, 0x12, 0x1); - if (best) { - best->ecx &= guest_supported_xcr0 & 0xffffffff; - best->edx &= guest_supported_xcr0 >> 32; - best->ecx |= XFEATURE_MASK_FPSSE; - } } void kvm_update_cpuid_runtime(struct kvm_vcpu *vcpu) diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index e542cf285b51842099332f594f4836836bda56d4..3c300a196bdf945ea7354cc370b4cd0aad515c61 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -228,6 +228,23 @@ static int kvm_recalculate_phys_map(struct kvm_apic_map *new, u32 xapic_id = kvm_xapic_id(apic); u32 physical_id; + /* + * For simplicity, KVM always allocates enough space for all possible + * xAPIC IDs. Yell, but don't kill the VM, as KVM can continue on + * without the optimized map. + */ + if (WARN_ON_ONCE(xapic_id > new->max_apic_id)) + return -EINVAL; + + /* + * Bail if a vCPU was added and/or enabled its APIC between allocating + * the map and doing the actual calculations for the map. Note, KVM + * hardcodes the x2APIC ID to vcpu_id, i.e. there's no TOCTOU bug if + * the compiler decides to reload x2apic_id after this check. + */ + if (x2apic_id > new->max_apic_id) + return -E2BIG; + /* * Deliberately truncate the vCPU ID when detecting a mismatched APIC * ID to avoid false positives if the vCPU ID, i.e. x2APIC ID, is a @@ -253,8 +270,7 @@ static int kvm_recalculate_phys_map(struct kvm_apic_map *new, */ if (vcpu->kvm->arch.x2apic_format) { /* See also kvm_apic_match_physical_addr(). */ - if ((apic_x2apic_mode(apic) || x2apic_id > 0xff) && - x2apic_id <= new->max_apic_id) + if (apic_x2apic_mode(apic) || x2apic_id > 0xff) new->phys_map[x2apic_id] = apic; if (!apic_x2apic_mode(apic) && !new->phys_map[xapic_id]) diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index c8961f45e3b1c1c01108f2830da0aad9624601b2..6eaa3d6994aebea7ecc344626adce9d0d062a31b 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -7091,7 +7091,10 @@ static void kvm_recover_nx_huge_pages(struct kvm *kvm) */ slot = NULL; if (atomic_read(&kvm->nr_memslots_dirty_logging)) { - slot = gfn_to_memslot(kvm, sp->gfn); + struct kvm_memslots *slots; + + slots = kvm_memslots_for_spte_role(kvm, sp->role); + slot = __gfn_to_memslot(slots, sp->gfn); WARN_ON_ONCE(!slot); } diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c index ca32389f3c36bae9d4c2d29c7d810fde12bb82ff..54089f990c8f83292eecede362b19269c8756900 100644 --- a/arch/x86/kvm/svm/svm.c +++ b/arch/x86/kvm/svm/svm.c @@ -3510,7 +3510,7 @@ static bool svm_is_vnmi_pending(struct kvm_vcpu *vcpu) if (!is_vnmi_enabled(svm)) return false; - return !!(svm->vmcb->control.int_ctl & V_NMI_BLOCKING_MASK); + return !!(svm->vmcb->control.int_ctl & V_NMI_PENDING_MASK); } static bool svm_set_vnmi_pending(struct kvm_vcpu *vcpu) diff --git a/arch/x86/kvm/vmx/sgx.c b/arch/x86/kvm/vmx/sgx.c index 0574030b071fbfea9ed7943da13c5d32942b4b14..2261b684a7d4475ab78107114aa5cb8dd5b2179e 100644 --- a/arch/x86/kvm/vmx/sgx.c +++ b/arch/x86/kvm/vmx/sgx.c @@ -170,12 +170,19 @@ static int __handle_encls_ecreate(struct kvm_vcpu *vcpu, return 1; } - /* Enforce CPUID restrictions on MISCSELECT, ATTRIBUTES and XFRM. */ + /* + * Enforce CPUID restrictions on MISCSELECT, ATTRIBUTES and XFRM. Note + * that the allowed XFRM (XFeature Request Mask) isn't strictly bound + * by the supported XCR0. FP+SSE *must* be set in XFRM, even if XSAVE + * is unsupported, i.e. even if XCR0 itself is completely unsupported. + */ if ((u32)miscselect & ~sgx_12_0->ebx || (u32)attributes & ~sgx_12_1->eax || (u32)(attributes >> 32) & ~sgx_12_1->ebx || (u32)xfrm & ~sgx_12_1->ecx || - (u32)(xfrm >> 32) & ~sgx_12_1->edx) { + (u32)(xfrm >> 32) & ~sgx_12_1->edx || + xfrm & ~(vcpu->arch.guest_supported_xcr0 | XFEATURE_MASK_FPSSE) || + (xfrm & XFEATURE_MASK_FPSSE) != XFEATURE_MASK_FPSSE) { kvm_inject_gp(vcpu, 0); return 1; } diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index ceb7c5e9cf9e9235a63cee5f909ba2a0ae41d5b3..04b57a336b34e5fc5224c47e5e8f23f794e98729 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -1446,7 +1446,7 @@ static const u32 msrs_to_save_base[] = { #endif MSR_IA32_TSC, MSR_IA32_CR_PAT, MSR_VM_HSAVE_PA, MSR_IA32_FEAT_CTL, MSR_IA32_BNDCFGS, MSR_TSC_AUX, - MSR_IA32_SPEC_CTRL, + MSR_IA32_SPEC_CTRL, MSR_IA32_TSX_CTRL, MSR_IA32_RTIT_CTL, MSR_IA32_RTIT_STATUS, MSR_IA32_RTIT_CR3_MATCH, MSR_IA32_RTIT_OUTPUT_BASE, MSR_IA32_RTIT_OUTPUT_MASK, MSR_IA32_RTIT_ADDR0_A, MSR_IA32_RTIT_ADDR0_B, @@ -7155,6 +7155,10 @@ static void kvm_probe_msr_to_save(u32 msr_index) if (!kvm_cpu_cap_has(X86_FEATURE_XFD)) return; break; + case MSR_IA32_TSX_CTRL: + if (!(kvm_get_arch_capabilities() & ARCH_CAP_TSX_CTRL_MSR)) + return; + break; default: break; } @@ -10754,6 +10758,9 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) exit_fastpath = EXIT_FASTPATH_EXIT_HANDLED; break; } + + /* Note, VM-Exits that go down the "slow" path are accounted below. */ + ++vcpu->stat.exits; } /* diff --git a/arch/x86/lib/copy_user_64.S b/arch/x86/lib/copy_user_64.S index 4fc5c2de2de467c689ddf91eb1684eeb00aeb9db..01c5de4c279b8fe8cf0fcd914f4cb4ec31945bc8 100644 --- a/arch/x86/lib/copy_user_64.S +++ b/arch/x86/lib/copy_user_64.S @@ -7,6 +7,8 @@ */ #include +#include +#include #include #include @@ -29,7 +31,7 @@ */ SYM_FUNC_START(rep_movs_alternative) cmpq $64,%rcx - jae .Lunrolled + jae .Llarge cmp $8,%ecx jae .Lword @@ -65,6 +67,12 @@ SYM_FUNC_START(rep_movs_alternative) _ASM_EXTABLE_UA( 2b, .Lcopy_user_tail) _ASM_EXTABLE_UA( 3b, .Lcopy_user_tail) +.Llarge: +0: ALTERNATIVE "jmp .Lunrolled", "rep movsb", X86_FEATURE_ERMS +1: RET + + _ASM_EXTABLE_UA( 0b, 1b) + .p2align 4 .Lunrolled: 10: movq (%rsi),%r8 diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index 3cdac0f0055deef186eb5a84d751f0ecb7058b9d..8192452d1d2d3569dddfb109e74e6b83d8b5570d 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -9,6 +9,7 @@ #include #include +#include #include #include #include @@ -261,6 +262,24 @@ static void __init probe_page_size_mask(void) } } +#define INTEL_MATCH(_model) { .vendor = X86_VENDOR_INTEL, \ + .family = 6, \ + .model = _model, \ + } +/* + * INVLPG may not properly flush Global entries + * on these CPUs when PCIDs are enabled. + */ +static const struct x86_cpu_id invlpg_miss_ids[] = { + INTEL_MATCH(INTEL_FAM6_ALDERLAKE ), + INTEL_MATCH(INTEL_FAM6_ALDERLAKE_L ), + INTEL_MATCH(INTEL_FAM6_ALDERLAKE_N ), + INTEL_MATCH(INTEL_FAM6_RAPTORLAKE ), + INTEL_MATCH(INTEL_FAM6_RAPTORLAKE_P), + INTEL_MATCH(INTEL_FAM6_RAPTORLAKE_S), + {} +}; + static void setup_pcid(void) { if (!IS_ENABLED(CONFIG_X86_64)) @@ -269,6 +288,12 @@ static void setup_pcid(void) if (!boot_cpu_has(X86_FEATURE_PCID)) return; + if (x86_match_cpu(invlpg_miss_ids)) { + pr_info("Incomplete global flushes, disabling PCID"); + setup_clear_cpu_cap(X86_FEATURE_PCID); + return; + } + if (boot_cpu_has(X86_FEATURE_PGE)) { /* * This can't be cr4_set_bits_and_update_boot() -- the diff --git a/arch/x86/pci/xen.c b/arch/x86/pci/xen.c index 8babce71915fe514aa24cfdb25f36c98ddc90369..014c508e914d9816bb062621554c65231468870a 100644 --- a/arch/x86/pci/xen.c +++ b/arch/x86/pci/xen.c @@ -198,7 +198,7 @@ static int xen_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) i++; } kfree(v); - return 0; + return msi_device_populate_sysfs(&dev->dev); error: if (ret == -ENOSYS) @@ -254,7 +254,7 @@ static int xen_hvm_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) dev_dbg(&dev->dev, "xen: msi --> pirq=%d --> irq=%d\n", pirq, irq); } - return 0; + return msi_device_populate_sysfs(&dev->dev); error: dev_err(&dev->dev, "Failed to create MSI%s! ret=%d!\n", @@ -346,7 +346,7 @@ static int xen_initdom_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) if (ret < 0) goto out; } - ret = 0; + ret = msi_device_populate_sysfs(&dev->dev); out: return ret; } @@ -394,6 +394,8 @@ static void xen_teardown_msi_irqs(struct pci_dev *dev) xen_destroy_irq(msidesc->irq + i); msidesc->irq = 0; } + + msi_device_destroy_sysfs(&dev->dev); } static void xen_pv_teardown_msi_irqs(struct pci_dev *dev) diff --git a/arch/x86/purgatory/Makefile b/arch/x86/purgatory/Makefile index 82fec66d46d29ee7e41e392fd5f22c524ad4f727..42abd6af119846977f7e8eea870bef761876a480 100644 --- a/arch/x86/purgatory/Makefile +++ b/arch/x86/purgatory/Makefile @@ -14,6 +14,11 @@ $(obj)/sha256.o: $(srctree)/lib/crypto/sha256.c FORCE CFLAGS_sha256.o := -D__DISABLE_EXPORTS +# When profile-guided optimization is enabled, llvm emits two different +# overlapping text sections, which is not supported by kexec. Remove profile +# optimization flags. +KBUILD_CFLAGS := $(filter-out -fprofile-sample-use=% -fprofile-use=%,$(KBUILD_CFLAGS)) + # When linking purgatory.ro with -r unresolved symbols are not checked, # also link a purgatory.chk binary without -r to check for unresolved symbols. PURGATORY_LDFLAGS := -e purgatory_start -z nodefaultlib diff --git a/arch/xtensa/kernel/signal.c b/arch/xtensa/kernel/signal.c index 876d5df157ed986ff3d919aba15fca177806d56e..5c01d7e70d90da4a0f495273180005cf235f4a7c 100644 --- a/arch/xtensa/kernel/signal.c +++ b/arch/xtensa/kernel/signal.c @@ -343,7 +343,19 @@ static int setup_frame(struct ksignal *ksig, sigset_t *set, struct rt_sigframe *frame; int err = 0, sig = ksig->sig; unsigned long sp, ra, tp, ps; + unsigned long handler = (unsigned long)ksig->ka.sa.sa_handler; + unsigned long handler_fdpic_GOT = 0; unsigned int base; + bool fdpic = IS_ENABLED(CONFIG_BINFMT_ELF_FDPIC) && + (current->personality & FDPIC_FUNCPTRS); + + if (fdpic) { + unsigned long __user *fdpic_func_desc = + (unsigned long __user *)handler; + if (__get_user(handler, &fdpic_func_desc[0]) || + __get_user(handler_fdpic_GOT, &fdpic_func_desc[1])) + return -EFAULT; + } sp = regs->areg[1]; @@ -373,20 +385,26 @@ static int setup_frame(struct ksignal *ksig, sigset_t *set, err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); if (ksig->ka.sa.sa_flags & SA_RESTORER) { - ra = (unsigned long)ksig->ka.sa.sa_restorer; + if (fdpic) { + unsigned long __user *fdpic_func_desc = + (unsigned long __user *)ksig->ka.sa.sa_restorer; + + err |= __get_user(ra, fdpic_func_desc); + } else { + ra = (unsigned long)ksig->ka.sa.sa_restorer; + } } else { /* Create sys_rt_sigreturn syscall in stack frame */ err |= gen_return_code(frame->retcode); - - if (err) { - return -EFAULT; - } ra = (unsigned long) frame->retcode; } - /* + if (err) + return -EFAULT; + + /* * Create signal handler execution context. * Return context not modified until this point. */ @@ -394,8 +412,7 @@ static int setup_frame(struct ksignal *ksig, sigset_t *set, /* Set up registers for signal handler; preserve the threadptr */ tp = regs->threadptr; ps = regs->ps; - start_thread(regs, (unsigned long) ksig->ka.sa.sa_handler, - (unsigned long) frame); + start_thread(regs, handler, (unsigned long)frame); /* Set up a stack frame for a call4 if userspace uses windowed ABI */ if (ps & PS_WOE_MASK) { @@ -413,6 +430,8 @@ static int setup_frame(struct ksignal *ksig, sigset_t *set, regs->areg[base + 4] = (unsigned long) &frame->uc; regs->threadptr = tp; regs->ps = ps; + if (fdpic) + regs->areg[base + 11] = handler_fdpic_GOT; pr_debug("SIG rt deliver (%s:%d): signal=%d sp=%p pc=%08lx\n", current->comm, current->pid, sig, frame, regs->pc); diff --git a/arch/xtensa/kernel/xtensa_ksyms.c b/arch/xtensa/kernel/xtensa_ksyms.c index 2a31b1ab0c9f20ed800b3f2490afe5e33b9bd579..17a7ef86fd0dd0c67c4ef0d25576ac1ce106620f 100644 --- a/arch/xtensa/kernel/xtensa_ksyms.c +++ b/arch/xtensa/kernel/xtensa_ksyms.c @@ -56,6 +56,8 @@ EXPORT_SYMBOL(empty_zero_page); */ extern long long __ashrdi3(long long, int); extern long long __ashldi3(long long, int); +extern long long __bswapdi2(long long); +extern int __bswapsi2(int); extern long long __lshrdi3(long long, int); extern int __divsi3(int, int); extern int __modsi3(int, int); @@ -66,6 +68,8 @@ extern unsigned long long __umulsidi3(unsigned int, unsigned int); EXPORT_SYMBOL(__ashldi3); EXPORT_SYMBOL(__ashrdi3); +EXPORT_SYMBOL(__bswapdi2); +EXPORT_SYMBOL(__bswapsi2); EXPORT_SYMBOL(__lshrdi3); EXPORT_SYMBOL(__divsi3); EXPORT_SYMBOL(__modsi3); diff --git a/arch/xtensa/lib/Makefile b/arch/xtensa/lib/Makefile index 7ecef0519a27caac48855896aa6042959f315c56..c9c2614188f74c788a566a1c58b24f3b68721163 100644 --- a/arch/xtensa/lib/Makefile +++ b/arch/xtensa/lib/Makefile @@ -4,7 +4,7 @@ # lib-y += memcopy.o memset.o checksum.o \ - ashldi3.o ashrdi3.o lshrdi3.o \ + ashldi3.o ashrdi3.o bswapdi2.o bswapsi2.o lshrdi3.o \ divsi3.o udivsi3.o modsi3.o umodsi3.o mulsi3.o umulsidi3.o \ usercopy.o strncpy_user.o strnlen_user.o lib-$(CONFIG_PCI) += pci-auto.o diff --git a/arch/xtensa/lib/bswapdi2.S b/arch/xtensa/lib/bswapdi2.S new file mode 100644 index 0000000000000000000000000000000000000000..d8e52e05eba664fe80a41a6ee20468eab878b1a2 --- /dev/null +++ b/arch/xtensa/lib/bswapdi2.S @@ -0,0 +1,21 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later WITH GCC-exception-2.0 */ +#include +#include +#include + +ENTRY(__bswapdi2) + + abi_entry_default + ssai 8 + srli a4, a2, 16 + src a4, a4, a2 + src a4, a4, a4 + src a4, a2, a4 + srli a2, a3, 16 + src a2, a2, a3 + src a2, a2, a2 + src a2, a3, a2 + mov a3, a4 + abi_ret_default + +ENDPROC(__bswapdi2) diff --git a/arch/xtensa/lib/bswapsi2.S b/arch/xtensa/lib/bswapsi2.S new file mode 100644 index 0000000000000000000000000000000000000000..9c1de1344f79ad7c71a6dcfcb73a0dd72c302328 --- /dev/null +++ b/arch/xtensa/lib/bswapsi2.S @@ -0,0 +1,16 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later WITH GCC-exception-2.0 */ +#include +#include +#include + +ENTRY(__bswapsi2) + + abi_entry_default + ssai 8 + srli a3, a2, 16 + src a3, a3, a2 + src a3, a3, a3 + src a2, a2, a3 + abi_ret_default + +ENDPROC(__bswapsi2) diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index 0ce64dd73cfe71489cce5dba5672a66b0052d4e0..f0b5c9c41cdee1c40cab795beca2e5e932ef0d00 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -34,6 +34,8 @@ #include "blk-ioprio.h" #include "blk-throttle.h" +static void __blkcg_rstat_flush(struct blkcg *blkcg, int cpu); + /* * blkcg_pol_mutex protects blkcg_policy[] and policy [de]activation. * blkcg_pol_register_mutex nests outside of it and synchronizes entire @@ -56,6 +58,8 @@ static LIST_HEAD(all_blkcgs); /* protected by blkcg_pol_mutex */ bool blkcg_debug_stats = false; +static DEFINE_RAW_SPINLOCK(blkg_stat_lock); + #define BLKG_DESTROY_BATCH_SIZE 64 /* @@ -163,10 +167,20 @@ static void blkg_free(struct blkcg_gq *blkg) static void __blkg_release(struct rcu_head *rcu) { struct blkcg_gq *blkg = container_of(rcu, struct blkcg_gq, rcu_head); + struct blkcg *blkcg = blkg->blkcg; + int cpu; #ifdef CONFIG_BLK_CGROUP_PUNT_BIO WARN_ON(!bio_list_empty(&blkg->async_bios)); #endif + /* + * Flush all the non-empty percpu lockless lists before releasing + * us, given these stat belongs to us. + * + * blkg_stat_lock is for serializing blkg stat update + */ + for_each_possible_cpu(cpu) + __blkcg_rstat_flush(blkcg, cpu); /* release the blkcg and parent blkg refs this blkg has been holding */ css_put(&blkg->blkcg->css); @@ -951,23 +965,26 @@ static void blkcg_iostat_update(struct blkcg_gq *blkg, struct blkg_iostat *cur, u64_stats_update_end_irqrestore(&blkg->iostat.sync, flags); } -static void blkcg_rstat_flush(struct cgroup_subsys_state *css, int cpu) +static void __blkcg_rstat_flush(struct blkcg *blkcg, int cpu) { - struct blkcg *blkcg = css_to_blkcg(css); struct llist_head *lhead = per_cpu_ptr(blkcg->lhead, cpu); struct llist_node *lnode; struct blkg_iostat_set *bisc, *next_bisc; - /* Root-level stats are sourced from system-wide IO stats */ - if (!cgroup_parent(css->cgroup)) - return; - rcu_read_lock(); lnode = llist_del_all(lhead); if (!lnode) goto out; + /* + * For covering concurrent parent blkg update from blkg_release(). + * + * When flushing from cgroup, cgroup_rstat_lock is always held, so + * this lock won't cause contention most of time. + */ + raw_spin_lock(&blkg_stat_lock); + /* * Iterate only the iostat_cpu's queued in the lockless list. */ @@ -991,13 +1008,19 @@ static void blkcg_rstat_flush(struct cgroup_subsys_state *css, int cpu) if (parent && parent->parent) blkcg_iostat_update(parent, &blkg->iostat.cur, &blkg->iostat.last); - percpu_ref_put(&blkg->refcnt); } - + raw_spin_unlock(&blkg_stat_lock); out: rcu_read_unlock(); } +static void blkcg_rstat_flush(struct cgroup_subsys_state *css, int cpu) +{ + /* Root-level stats are sourced from system-wide IO stats */ + if (cgroup_parent(css->cgroup)) + __blkcg_rstat_flush(css_to_blkcg(css), cpu); +} + /* * We source root cgroup stats from the system-wide stats to avoid * tracking the same information twice and incurring overhead when no @@ -2075,7 +2098,6 @@ void blk_cgroup_bio_start(struct bio *bio) llist_add(&bis->lnode, lhead); WRITE_ONCE(bis->lqueued, true); - percpu_ref_get(&bis->blkg->refcnt); } u64_stats_update_end_irqrestore(&bis->sync, flags); diff --git a/block/blk-core.c b/block/blk-core.c index 00c74330fa92c221422a629361607068a65eda99..1da77e7d62894651d3e307dbd1be1930983baad8 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -520,7 +520,7 @@ static inline int bio_check_eod(struct bio *bio) sector_t maxsector = bdev_nr_sectors(bio->bi_bdev); unsigned int nr_sectors = bio_sectors(bio); - if (nr_sectors && maxsector && + if (nr_sectors && (nr_sectors > maxsector || bio->bi_iter.bi_sector > maxsector - nr_sectors)) { pr_info_ratelimited("%s: attempt to access beyond end of device\n" diff --git a/block/blk-map.c b/block/blk-map.c index 04c55f1c492eb144b332139478b5cc390dbbf075..46eed2e627c363a4ab6a3797b538380be6a45ef6 100644 --- a/block/blk-map.c +++ b/block/blk-map.c @@ -248,7 +248,7 @@ static struct bio *blk_rq_map_bio_alloc(struct request *rq, { struct bio *bio; - if (rq->cmd_flags & REQ_ALLOC_CACHE) { + if (rq->cmd_flags & REQ_ALLOC_CACHE && (nr_vecs <= BIO_INLINE_VECS)) { bio = bio_alloc_bioset(NULL, nr_vecs, rq->cmd_flags, gfp_mask, &fs_bio_set); if (!bio) diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c index d6af9d431dc631989dce6709971d1fc042e89dc1..dfd81cab57888b33a39b7b8b8e7d8d1c3bae0506 100644 --- a/block/blk-mq-tag.c +++ b/block/blk-mq-tag.c @@ -39,16 +39,20 @@ void __blk_mq_tag_busy(struct blk_mq_hw_ctx *hctx) { unsigned int users; + /* + * calling test_bit() prior to test_and_set_bit() is intentional, + * it avoids dirtying the cacheline if the queue is already active. + */ if (blk_mq_is_shared_tags(hctx->flags)) { struct request_queue *q = hctx->queue; - if (test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) + if (test_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags) || + test_and_set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags)) return; - set_bit(QUEUE_FLAG_HCTX_ACTIVE, &q->queue_flags); } else { - if (test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) + if (test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) || + test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) return; - set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state); } users = atomic_inc_return(&hctx->tags->active_queues); diff --git a/block/blk-mq.c b/block/blk-mq.c index f6dad0886a2fa1bacac778c093717ba21e24b33e..850bfb844ed2f75b7c37774fe541a08cd59294c5 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -683,6 +683,10 @@ static void __blk_mq_free_request(struct request *rq) blk_crypto_free_request(rq); blk_pm_mark_last_busy(rq); rq->mq_hctx = NULL; + + if (rq->rq_flags & RQF_MQ_INFLIGHT) + __blk_mq_dec_active_requests(hctx); + if (rq->tag != BLK_MQ_NO_TAG) blk_mq_put_tag(hctx->tags, ctx, rq->tag); if (sched_tag != BLK_MQ_NO_TAG) @@ -694,15 +698,11 @@ static void __blk_mq_free_request(struct request *rq) void blk_mq_free_request(struct request *rq) { struct request_queue *q = rq->q; - struct blk_mq_hw_ctx *hctx = rq->mq_hctx; if ((rq->rq_flags & RQF_ELVPRIV) && q->elevator->type->ops.finish_request) q->elevator->type->ops.finish_request(rq); - if (rq->rq_flags & RQF_MQ_INFLIGHT) - __blk_mq_dec_active_requests(hctx); - if (unlikely(laptop_mode && !blk_rq_is_passthrough(rq))) laptop_io_completion(q->disk->bdi); diff --git a/block/blk-settings.c b/block/blk-settings.c index 896b4654ab00fa700146678f2a0c1c4371f2ed7b..4dd59059b788eb3b8b45cc555716eec11bfcd995 100644 --- a/block/blk-settings.c +++ b/block/blk-settings.c @@ -915,6 +915,7 @@ static bool disk_has_partitions(struct gendisk *disk) void disk_set_zoned(struct gendisk *disk, enum blk_zoned_model model) { struct request_queue *q = disk->queue; + unsigned int old_model = q->limits.zoned; switch (model) { case BLK_ZONED_HM: @@ -952,7 +953,7 @@ void disk_set_zoned(struct gendisk *disk, enum blk_zoned_model model) */ blk_queue_zone_write_granularity(q, queue_logical_block_size(q)); - } else { + } else if (old_model != BLK_ZONED_NONE) { disk_clear_zone_settings(disk); } } diff --git a/block/blk-wbt.c b/block/blk-wbt.c index e49a4868453276744e02561c6bfef1dfe68aa2c2..9ec2a2f1eda3882b8d174b90c2e5f14ae5d07679 100644 --- a/block/blk-wbt.c +++ b/block/blk-wbt.c @@ -730,14 +730,16 @@ void wbt_enable_default(struct gendisk *disk) { struct request_queue *q = disk->queue; struct rq_qos *rqos; - bool disable_flag = q->elevator && - test_bit(ELEVATOR_FLAG_DISABLE_WBT, &q->elevator->flags); + bool enable = IS_ENABLED(CONFIG_BLK_WBT_MQ); + + if (q->elevator && + test_bit(ELEVATOR_FLAG_DISABLE_WBT, &q->elevator->flags)) + enable = false; /* Throttling already enabled? */ rqos = wbt_rq_qos(q); if (rqos) { - if (!disable_flag && - RQWB(rqos)->enable_state == WBT_STATE_OFF_DEFAULT) + if (enable && RQWB(rqos)->enable_state == WBT_STATE_OFF_DEFAULT) RQWB(rqos)->enable_state = WBT_STATE_ON_DEFAULT; return; } @@ -746,7 +748,7 @@ void wbt_enable_default(struct gendisk *disk) if (!blk_queue_registered(q)) return; - if (queue_is_mq(q) && !disable_flag) + if (queue_is_mq(q) && enable) wbt_init(disk); } EXPORT_SYMBOL_GPL(wbt_enable_default); diff --git a/block/fops.c b/block/fops.c index d2e6be4e3d1c7da029d0a38bd18cf9b069f901e2..58d0aebc7313a838d4d895fd73f7c383290d75da 100644 --- a/block/fops.c +++ b/block/fops.c @@ -678,6 +678,16 @@ static long blkdev_fallocate(struct file *file, int mode, loff_t start, return error; } +static int blkdev_mmap(struct file *file, struct vm_area_struct *vma) +{ + struct inode *bd_inode = bdev_file_inode(file); + + if (bdev_read_only(I_BDEV(bd_inode))) + return generic_file_readonly_mmap(file, vma); + + return generic_file_mmap(file, vma); +} + const struct file_operations def_blk_fops = { .open = blkdev_open, .release = blkdev_close, @@ -685,7 +695,7 @@ const struct file_operations def_blk_fops = { .read_iter = blkdev_read_iter, .write_iter = blkdev_write_iter, .iopoll = iocb_bio_iopoll, - .mmap = generic_file_mmap, + .mmap = blkdev_mmap, .fsync = blkdev_fsync, .unlocked_ioctl = blkdev_ioctl, #ifdef CONFIG_COMPAT diff --git a/crypto/asymmetric_keys/public_key.c b/crypto/asymmetric_keys/public_key.c index eca5671ad3f2288eefe2a07b31d50b1a5728a8ed..50c933f86b218cca0ffe07756272dadbcf9854a2 100644 --- a/crypto/asymmetric_keys/public_key.c +++ b/crypto/asymmetric_keys/public_key.c @@ -380,9 +380,10 @@ int public_key_verify_signature(const struct public_key *pkey, struct crypto_wait cwait; struct crypto_akcipher *tfm; struct akcipher_request *req; - struct scatterlist src_sg[2]; + struct scatterlist src_sg; char alg_name[CRYPTO_MAX_ALG_NAME]; - char *key, *ptr; + char *buf, *ptr; + size_t buf_len; int ret; pr_devel("==>%s()\n", __func__); @@ -420,34 +421,37 @@ int public_key_verify_signature(const struct public_key *pkey, if (!req) goto error_free_tfm; - key = kmalloc(pkey->keylen + sizeof(u32) * 2 + pkey->paramlen, - GFP_KERNEL); - if (!key) + buf_len = max_t(size_t, pkey->keylen + sizeof(u32) * 2 + pkey->paramlen, + sig->s_size + sig->digest_size); + + buf = kmalloc(buf_len, GFP_KERNEL); + if (!buf) goto error_free_req; - memcpy(key, pkey->key, pkey->keylen); - ptr = key + pkey->keylen; + memcpy(buf, pkey->key, pkey->keylen); + ptr = buf + pkey->keylen; ptr = pkey_pack_u32(ptr, pkey->algo); ptr = pkey_pack_u32(ptr, pkey->paramlen); memcpy(ptr, pkey->params, pkey->paramlen); if (pkey->key_is_private) - ret = crypto_akcipher_set_priv_key(tfm, key, pkey->keylen); + ret = crypto_akcipher_set_priv_key(tfm, buf, pkey->keylen); else - ret = crypto_akcipher_set_pub_key(tfm, key, pkey->keylen); + ret = crypto_akcipher_set_pub_key(tfm, buf, pkey->keylen); if (ret) - goto error_free_key; + goto error_free_buf; if (strcmp(pkey->pkey_algo, "sm2") == 0 && sig->data_size) { ret = cert_sig_digest_update(sig, tfm); if (ret) - goto error_free_key; + goto error_free_buf; } - sg_init_table(src_sg, 2); - sg_set_buf(&src_sg[0], sig->s, sig->s_size); - sg_set_buf(&src_sg[1], sig->digest, sig->digest_size); - akcipher_request_set_crypt(req, src_sg, NULL, sig->s_size, + memcpy(buf, sig->s, sig->s_size); + memcpy(buf + sig->s_size, sig->digest, sig->digest_size); + + sg_init_one(&src_sg, buf, sig->s_size + sig->digest_size); + akcipher_request_set_crypt(req, &src_sg, NULL, sig->s_size, sig->digest_size); crypto_init_wait(&cwait); akcipher_request_set_callback(req, CRYPTO_TFM_REQ_MAY_BACKLOG | @@ -455,8 +459,8 @@ int public_key_verify_signature(const struct public_key *pkey, crypto_req_done, &cwait); ret = crypto_wait_req(crypto_akcipher_verify(req), &cwait); -error_free_key: - kfree(key); +error_free_buf: + kfree(buf); error_free_req: akcipher_request_free(req); error_free_tfm: diff --git a/drivers/accel/habanalabs/common/command_buffer.c b/drivers/accel/habanalabs/common/command_buffer.c index 6e09f48750a0279c89c4c850767a9dec6e506fcf..08f7aee426246b96f8d20103aac829ebe4c57034 100644 --- a/drivers/accel/habanalabs/common/command_buffer.c +++ b/drivers/accel/habanalabs/common/command_buffer.c @@ -27,12 +27,6 @@ static int cb_map_mem(struct hl_ctx *ctx, struct hl_cb *cb) return -EINVAL; } - if (!hdev->mmu_enable) { - dev_err_ratelimited(hdev->dev, - "Cannot map CB because MMU is disabled\n"); - return -EINVAL; - } - if (cb->is_mmu_mapped) return 0; diff --git a/drivers/accel/habanalabs/common/command_submission.c b/drivers/accel/habanalabs/common/command_submission.c index af9d2e22c6e7e08990a3cd4794a5a799203ff1ce..c23829dab97a1a26b2642fd5c7fb65047ee100c2 100644 --- a/drivers/accel/habanalabs/common/command_submission.c +++ b/drivers/accel/habanalabs/common/command_submission.c @@ -280,14 +280,8 @@ bool cs_needs_timeout(struct hl_cs *cs) static bool is_cb_patched(struct hl_device *hdev, struct hl_cs_job *job) { - /* - * Patched CB is created for external queues jobs, and for H/W queues - * jobs if the user CB was allocated by driver and MMU is disabled. - */ - return (job->queue_type == QUEUE_TYPE_EXT || - (job->queue_type == QUEUE_TYPE_HW && - job->is_kernel_allocated_cb && - !hdev->mmu_enable)); + /* Patched CB is created for external queues jobs */ + return (job->queue_type == QUEUE_TYPE_EXT); } /* @@ -363,14 +357,13 @@ static void hl_complete_job(struct hl_device *hdev, struct hl_cs_job *job) } } - /* For H/W queue jobs, if a user CB was allocated by driver and MMU is - * enabled, the user CB isn't released in cs_parser() and thus should be + /* For H/W queue jobs, if a user CB was allocated by driver, + * the user CB isn't released in cs_parser() and thus should be * released here. This is also true for INT queues jobs which were * allocated by driver. */ - if ((job->is_kernel_allocated_cb && - ((job->queue_type == QUEUE_TYPE_HW && hdev->mmu_enable) || - job->queue_type == QUEUE_TYPE_INT))) { + if (job->is_kernel_allocated_cb && + (job->queue_type == QUEUE_TYPE_HW || job->queue_type == QUEUE_TYPE_INT)) { atomic_dec(&job->user_cb->cs_cnt); hl_cb_put(job->user_cb); } @@ -804,12 +797,14 @@ static void cs_do_release(struct kref *ref) static void cs_timedout(struct work_struct *work) { + struct hl_cs *cs = container_of(work, struct hl_cs, work_tdr.work); + bool skip_reset_on_timeout, device_reset = false; struct hl_device *hdev; u64 event_mask = 0x0; + uint timeout_sec; int rc; - struct hl_cs *cs = container_of(work, struct hl_cs, - work_tdr.work); - bool skip_reset_on_timeout = cs->skip_reset_on_timeout, device_reset = false; + + skip_reset_on_timeout = cs->skip_reset_on_timeout; rc = cs_get_unless_zero(cs); if (!rc) @@ -840,29 +835,31 @@ static void cs_timedout(struct work_struct *work) event_mask |= HL_NOTIFIER_EVENT_CS_TIMEOUT; } + timeout_sec = jiffies_to_msecs(hdev->timeout_jiffies) / 1000; + switch (cs->type) { case CS_TYPE_SIGNAL: dev_err(hdev->dev, - "Signal command submission %llu has not finished in time!\n", - cs->sequence); + "Signal command submission %llu has not finished in %u seconds!\n", + cs->sequence, timeout_sec); break; case CS_TYPE_WAIT: dev_err(hdev->dev, - "Wait command submission %llu has not finished in time!\n", - cs->sequence); + "Wait command submission %llu has not finished in %u seconds!\n", + cs->sequence, timeout_sec); break; case CS_TYPE_COLLECTIVE_WAIT: dev_err(hdev->dev, - "Collective Wait command submission %llu has not finished in time!\n", - cs->sequence); + "Collective Wait command submission %llu has not finished in %u seconds!\n", + cs->sequence, timeout_sec); break; default: dev_err(hdev->dev, - "Command submission %llu has not finished in time!\n", - cs->sequence); + "Command submission %llu has not finished in %u seconds!\n", + cs->sequence, timeout_sec); break; } @@ -1139,11 +1136,10 @@ static void force_complete_cs(struct hl_device *hdev) spin_unlock(&hdev->cs_mirror_lock); } -void hl_abort_waitings_for_completion(struct hl_device *hdev) +void hl_abort_waiting_for_cs_completions(struct hl_device *hdev) { force_complete_cs(hdev); force_complete_multi_cs(hdev); - hl_release_pending_user_interrupts(hdev); } static void job_wq_completion(struct work_struct *work) @@ -1948,8 +1944,7 @@ static int cs_ioctl_signal_wait_create_jobs(struct hl_device *hdev, else cb_size = hdev->asic_funcs->get_signal_cb_size(hdev); - cb = hl_cb_kernel_create(hdev, cb_size, - q_type == QUEUE_TYPE_HW && hdev->mmu_enable); + cb = hl_cb_kernel_create(hdev, cb_size, q_type == QUEUE_TYPE_HW); if (!cb) { atomic64_inc(&ctx->cs_counters.out_of_mem_drop_cnt); atomic64_inc(&cntr->out_of_mem_drop_cnt); @@ -2152,7 +2147,7 @@ static int cs_ioctl_unreserve_signals(struct hl_fpriv *hpriv, u32 handle_id) hdev->asic_funcs->hw_queues_unlock(hdev); rc = -EINVAL; - goto out; + goto out_unlock; } /* @@ -2167,15 +2162,21 @@ static int cs_ioctl_unreserve_signals(struct hl_fpriv *hpriv, u32 handle_id) /* Release the id and free allocated memory of the handle */ idr_remove(&mgr->handles, handle_id); + + /* unlock before calling ctx_put, where we might sleep */ + spin_unlock(&mgr->lock); hl_ctx_put(encaps_sig_hdl->ctx); kfree(encaps_sig_hdl); + goto out; } else { rc = -EINVAL; dev_err(hdev->dev, "failed to unreserve signals, cannot find handler\n"); } -out: + +out_unlock: spin_unlock(&mgr->lock); +out: return rc; } diff --git a/drivers/accel/habanalabs/common/debugfs.c b/drivers/accel/habanalabs/common/debugfs.c index 22dd17c077c0037b5883c57e6aae6ead02791458..9e84a47a21dcf41f4f5ce26d3215cf2955ae3337 100644 --- a/drivers/accel/habanalabs/common/debugfs.c +++ b/drivers/accel/habanalabs/common/debugfs.c @@ -255,9 +255,6 @@ static int vm_show(struct seq_file *s, void *data) u64 j; int i; - if (!dev_entry->hdev->mmu_enable) - return 0; - mutex_lock(&dev_entry->ctx_mem_hash_mutex); list_for_each_entry(ctx, &dev_entry->ctx_mem_hash_list, debugfs_list) { @@ -436,9 +433,6 @@ static int mmu_show(struct seq_file *s, void *data) u64 virt_addr = dev_entry->mmu_addr, phys_addr; int i; - if (!hdev->mmu_enable) - return 0; - if (dev_entry->mmu_asid == HL_KERNEL_ASID_ID) ctx = hdev->kernel_ctx; else @@ -496,9 +490,6 @@ static ssize_t mmu_asid_va_write(struct file *file, const char __user *buf, char *c; ssize_t rc; - if (!hdev->mmu_enable) - return count; - if (count > sizeof(kbuf) - 1) goto err; if (copy_from_user(kbuf, buf, count)) @@ -535,9 +526,6 @@ static int mmu_ack_error(struct seq_file *s, void *data) struct hl_device *hdev = dev_entry->hdev; int rc; - if (!hdev->mmu_enable) - return 0; - if (!dev_entry->mmu_cap_mask) { dev_err(hdev->dev, "mmu_cap_mask is not set\n"); goto err; @@ -563,9 +551,6 @@ static ssize_t mmu_ack_error_value_write(struct file *file, char kbuf[MMU_KBUF_SIZE]; ssize_t rc; - if (!hdev->mmu_enable) - return count; - if (count > sizeof(kbuf) - 1) goto err; @@ -661,9 +646,6 @@ static bool hl_is_device_va(struct hl_device *hdev, u64 addr) { struct asic_fixed_properties *prop = &hdev->asic_prop; - if (!hdev->mmu_enable) - goto out; - if (prop->dram_supports_virtual_memory && (addr >= prop->dmmu.start_addr && addr < prop->dmmu.end_addr)) return true; @@ -675,7 +657,7 @@ static bool hl_is_device_va(struct hl_device *hdev, u64 addr) if (addr >= prop->pmmu_huge.start_addr && addr < prop->pmmu_huge.end_addr) return true; -out: + return false; } @@ -685,9 +667,6 @@ static bool hl_is_device_internal_memory_va(struct hl_device *hdev, u64 addr, struct asic_fixed_properties *prop = &hdev->asic_prop; u64 dram_start_addr, dram_end_addr; - if (!hdev->mmu_enable) - return false; - if (prop->dram_supports_virtual_memory) { dram_start_addr = prop->dmmu.start_addr; dram_end_addr = prop->dmmu.end_addr; @@ -1756,17 +1735,15 @@ static void add_files_to_device(struct hl_device *hdev, struct hl_dbg_device_ent } } -void hl_debugfs_add_device(struct hl_device *hdev) +int hl_debugfs_device_init(struct hl_device *hdev) { struct hl_dbg_device_entry *dev_entry = &hdev->hl_debugfs; int count = ARRAY_SIZE(hl_debugfs_list); dev_entry->hdev = hdev; - dev_entry->entry_arr = kmalloc_array(count, - sizeof(struct hl_debugfs_entry), - GFP_KERNEL); + dev_entry->entry_arr = kmalloc_array(count, sizeof(struct hl_debugfs_entry), GFP_KERNEL); if (!dev_entry->entry_arr) - return; + return -ENOMEM; dev_entry->data_dma_blob_desc.size = 0; dev_entry->data_dma_blob_desc.data = NULL; @@ -1787,21 +1764,14 @@ void hl_debugfs_add_device(struct hl_device *hdev) spin_lock_init(&dev_entry->userptr_spinlock); mutex_init(&dev_entry->ctx_mem_hash_mutex); - dev_entry->root = debugfs_create_dir(dev_name(hdev->dev), - hl_debug_root); - - add_files_to_device(hdev, dev_entry, dev_entry->root); - if (!hdev->asic_prop.fw_security_enabled) - add_secured_nodes(dev_entry, dev_entry->root); + return 0; } -void hl_debugfs_remove_device(struct hl_device *hdev) +void hl_debugfs_device_fini(struct hl_device *hdev) { struct hl_dbg_device_entry *entry = &hdev->hl_debugfs; int i; - debugfs_remove_recursive(entry->root); - mutex_destroy(&entry->ctx_mem_hash_mutex); mutex_destroy(&entry->file_mutex); @@ -1814,6 +1784,24 @@ void hl_debugfs_remove_device(struct hl_device *hdev) kfree(entry->entry_arr); } +void hl_debugfs_add_device(struct hl_device *hdev) +{ + struct hl_dbg_device_entry *dev_entry = &hdev->hl_debugfs; + + dev_entry->root = debugfs_create_dir(dev_name(hdev->dev), hl_debug_root); + + add_files_to_device(hdev, dev_entry, dev_entry->root); + if (!hdev->asic_prop.fw_security_enabled) + add_secured_nodes(dev_entry, dev_entry->root); +} + +void hl_debugfs_remove_device(struct hl_device *hdev) +{ + struct hl_dbg_device_entry *entry = &hdev->hl_debugfs; + + debugfs_remove_recursive(entry->root); +} + void hl_debugfs_add_file(struct hl_fpriv *hpriv) { struct hl_dbg_device_entry *dev_entry = &hpriv->hdev->hl_debugfs; diff --git a/drivers/accel/habanalabs/common/device.c b/drivers/accel/habanalabs/common/device.c index fabfc501ef5434c06f1ac8c5934011a6cee83888..b97339d1f7c6ea33e0428eb40bc86a1a31e320c5 100644 --- a/drivers/accel/habanalabs/common/device.c +++ b/drivers/accel/habanalabs/common/device.c @@ -674,7 +674,7 @@ static int device_init_cdev(struct hl_device *hdev, struct class *class, return 0; } -static int device_cdev_sysfs_add(struct hl_device *hdev) +static int cdev_sysfs_debugfs_add(struct hl_device *hdev) { int rc; @@ -699,7 +699,9 @@ static int device_cdev_sysfs_add(struct hl_device *hdev) goto delete_ctrl_cdev_device; } - hdev->cdev_sysfs_created = true; + hl_debugfs_add_device(hdev); + + hdev->cdev_sysfs_debugfs_created = true; return 0; @@ -710,11 +712,12 @@ static int device_cdev_sysfs_add(struct hl_device *hdev) return rc; } -static void device_cdev_sysfs_del(struct hl_device *hdev) +static void cdev_sysfs_debugfs_remove(struct hl_device *hdev) { - if (!hdev->cdev_sysfs_created) + if (!hdev->cdev_sysfs_debugfs_created) goto put_devices; + hl_debugfs_remove_device(hdev); hl_sysfs_fini(hdev); cdev_device_del(&hdev->cdev_ctrl, hdev->dev_ctrl); cdev_device_del(&hdev->cdev, hdev->dev); @@ -981,6 +984,18 @@ static void device_early_fini(struct hl_device *hdev) hdev->asic_funcs->early_fini(hdev); } +static bool is_pci_link_healthy(struct hl_device *hdev) +{ + u16 vendor_id; + + if (!hdev->pdev) + return false; + + pci_read_config_word(hdev->pdev, PCI_VENDOR_ID, &vendor_id); + + return (vendor_id == PCI_VENDOR_ID_HABANALABS); +} + static void hl_device_heartbeat(struct work_struct *work) { struct hl_device *hdev = container_of(work, struct hl_device, @@ -995,7 +1010,8 @@ static void hl_device_heartbeat(struct work_struct *work) goto reschedule; if (hl_device_operational(hdev, NULL)) - dev_err(hdev->dev, "Device heartbeat failed!\n"); + dev_err(hdev->dev, "Device heartbeat failed! PCI link is %s\n", + is_pci_link_healthy(hdev) ? "healthy" : "broken"); info.err_type = HL_INFO_FW_HEARTBEAT_ERR; info.event_mask = &event_mask; @@ -1157,6 +1173,16 @@ static void take_release_locks(struct hl_device *hdev) mutex_unlock(&hdev->fpriv_ctrl_list_lock); } +static void hl_abort_waiting_for_completions(struct hl_device *hdev) +{ + hl_abort_waiting_for_cs_completions(hdev); + + /* Release all pending user interrupts, each pending user interrupt + * holds a reference to a user context. + */ + hl_release_pending_user_interrupts(hdev); +} + static void cleanup_resources(struct hl_device *hdev, bool hard_reset, bool fw_reset, bool skip_wq_flush) { @@ -1176,10 +1202,7 @@ static void cleanup_resources(struct hl_device *hdev, bool hard_reset, bool fw_r /* flush the MMU prefetch workqueue */ flush_workqueue(hdev->prefetch_wq); - /* Release all pending user interrupts, each pending user interrupt - * holds a reference to user context - */ - hl_release_pending_user_interrupts(hdev); + hl_abort_waiting_for_completions(hdev); } /* @@ -1921,7 +1944,7 @@ int hl_device_cond_reset(struct hl_device *hdev, u32 flags, u64 event_mask) hl_ctx_put(ctx); - hl_abort_waitings_for_completion(hdev); + hl_abort_waiting_for_completions(hdev); return 0; @@ -2034,7 +2057,7 @@ static int create_cdev(struct hl_device *hdev) int hl_device_init(struct hl_device *hdev) { int i, rc, cq_cnt, user_interrupt_cnt, cq_ready_cnt; - bool add_cdev_sysfs_on_err = false; + bool expose_interfaces_on_err = false; rc = create_cdev(hdev); if (rc) @@ -2150,16 +2173,22 @@ int hl_device_init(struct hl_device *hdev) hdev->device_release_watchdog_timeout_sec = HL_DEVICE_RELEASE_WATCHDOG_TIMEOUT_SEC; hdev->memory_scrub_val = MEM_SCRUB_DEFAULT_VAL; - hl_debugfs_add_device(hdev); - /* debugfs nodes are created in hl_ctx_init so it must be called after - * hl_debugfs_add_device. + rc = hl_debugfs_device_init(hdev); + if (rc) { + dev_err(hdev->dev, "failed to initialize debugfs entry structure\n"); + kfree(hdev->kernel_ctx); + goto mmu_fini; + } + + /* The debugfs entry structure is accessed in hl_ctx_init(), so it must be called after + * hl_debugfs_device_init(). */ rc = hl_ctx_init(hdev, hdev->kernel_ctx, true); if (rc) { dev_err(hdev->dev, "failed to initialize kernel context\n"); kfree(hdev->kernel_ctx); - goto remove_device_from_debugfs; + goto debugfs_device_fini; } rc = hl_cb_pool_init(hdev); @@ -2175,11 +2204,10 @@ int hl_device_init(struct hl_device *hdev) } /* - * From this point, override rc (=0) in case of an error to allow - * debugging (by adding char devices and create sysfs nodes as part of - * the error flow). + * From this point, override rc (=0) in case of an error to allow debugging + * (by adding char devices and creating sysfs/debugfs files as part of the error flow). */ - add_cdev_sysfs_on_err = true; + expose_interfaces_on_err = true; /* Device is now enabled as part of the initialization requires * communication with the device firmware to get information that @@ -2221,15 +2249,13 @@ int hl_device_init(struct hl_device *hdev) } /* - * Expose devices and sysfs nodes to user. - * From here there is no need to add char devices and create sysfs nodes - * in case of an error. + * Expose devices and sysfs/debugfs files to user. + * From here there is no need to expose them in case of an error. */ - add_cdev_sysfs_on_err = false; - rc = device_cdev_sysfs_add(hdev); + expose_interfaces_on_err = false; + rc = cdev_sysfs_debugfs_add(hdev); if (rc) { - dev_err(hdev->dev, - "Failed to add char devices and sysfs nodes\n"); + dev_err(hdev->dev, "Failed to add char devices and sysfs/debugfs files\n"); rc = 0; goto out_disabled; } @@ -2275,8 +2301,8 @@ int hl_device_init(struct hl_device *hdev) if (hl_ctx_put(hdev->kernel_ctx) != 1) dev_err(hdev->dev, "kernel ctx is still alive on initialization failure\n"); -remove_device_from_debugfs: - hl_debugfs_remove_device(hdev); +debugfs_device_fini: + hl_debugfs_device_fini(hdev); mmu_fini: hl_mmu_fini(hdev); eq_fini: @@ -2300,15 +2326,11 @@ int hl_device_init(struct hl_device *hdev) put_device(hdev->dev); out_disabled: hdev->disabled = true; - if (add_cdev_sysfs_on_err) - device_cdev_sysfs_add(hdev); - if (hdev->pdev) - dev_err(&hdev->pdev->dev, - "Failed to initialize hl%d. Device %s is NOT usable !\n", - hdev->cdev_idx, dev_name(&(hdev)->pdev->dev)); - else - pr_err("Failed to initialize hl%d. Device %s is NOT usable !\n", - hdev->cdev_idx, dev_name(&(hdev)->pdev->dev)); + if (expose_interfaces_on_err) + cdev_sysfs_debugfs_add(hdev); + dev_err(&hdev->pdev->dev, + "Failed to initialize hl%d. Device %s is NOT usable !\n", + hdev->cdev_idx, dev_name(&hdev->pdev->dev)); return rc; } @@ -2427,8 +2449,6 @@ void hl_device_fini(struct hl_device *hdev) if ((hdev->kernel_ctx) && (hl_ctx_put(hdev->kernel_ctx) != 1)) dev_err(hdev->dev, "kernel ctx is still alive\n"); - hl_debugfs_remove_device(hdev); - hl_dec_fini(hdev); hl_vm_fini(hdev); @@ -2453,8 +2473,10 @@ void hl_device_fini(struct hl_device *hdev) device_early_fini(hdev); - /* Hide devices and sysfs nodes from user */ - device_cdev_sysfs_del(hdev); + /* Hide devices and sysfs/debugfs files from user */ + cdev_sysfs_debugfs_remove(hdev); + + hl_debugfs_device_fini(hdev); pr_info("removed device successfully\n"); } @@ -2667,3 +2689,11 @@ void hl_handle_fw_err(struct hl_device *hdev, struct hl_info_fw_err_info *info) if (info->event_mask) *info->event_mask |= HL_NOTIFIER_EVENT_CRITICL_FW_ERR; } + +void hl_enable_err_info_capture(struct hl_error_info *captured_err_info) +{ + vfree(captured_err_info->page_fault_info.user_mappings); + memset(captured_err_info, 0, sizeof(struct hl_error_info)); + atomic_set(&captured_err_info->cs_timeout.write_enable, 1); + captured_err_info->undef_opcode.write_enable = true; +} diff --git a/drivers/accel/habanalabs/common/firmware_if.c b/drivers/accel/habanalabs/common/firmware_if.c index 59f61ec66445a3e58d5a7cd337c1ec4156dde03a..acbc1a6b5cb1837ad3c25e5714ea87940e0cb936 100644 --- a/drivers/accel/habanalabs/common/firmware_if.c +++ b/drivers/accel/habanalabs/common/firmware_if.c @@ -71,38 +71,124 @@ static char *extract_fw_ver_from_str(const char *fw_str) return NULL; } -static int hl_get_preboot_major_minor(struct hl_device *hdev, char *preboot_ver) +/** + * extract_u32_until_given_char() - given a string of the format "*", extract the u32. + * @str: the given string + * @ver_num: the pointer to the extracted u32 to be returned to the caller. + * @given_char: the given char at the end of the u32 in the string + * + * Return: Upon success, return a pointer to the given_char in the string. Upon failure, return NULL + */ +static char *extract_u32_until_given_char(char *str, u32 *ver_num, char given_char) { - char major[8], minor[8], *first_dot, *second_dot; - int rc; + char num_str[8] = {}, *ch; - first_dot = strnstr(preboot_ver, ".", 10); - if (first_dot) { - strscpy(major, preboot_ver, first_dot - preboot_ver + 1); - rc = kstrtou32(major, 10, &hdev->fw_major_version); - } else { - rc = -EINVAL; + ch = strchrnul(str, given_char); + if (*ch == '\0' || ch == str || ch - str >= sizeof(num_str)) + return NULL; + + memcpy(num_str, str, ch - str); + if (kstrtou32(num_str, 10, ver_num)) + return NULL; + return ch; +} + +/** + * hl_get_sw_major_minor_subminor() - extract the FW's SW version major, minor, sub-minor + * from the version string + * @hdev: pointer to the hl_device + * @fw_str: the FW's version string + * + * The extracted version is set in the hdev fields: fw_sw_{major/minor/sub_minor}_ver. + * + * fw_str is expected to have one of two possible formats, examples: + * 1) 'Preboot version hl-gaudi2-1.9.0-fw-42.0.1-sec-3' + * 2) 'Preboot version hl-gaudi2-1.9.0-rc-fw-42.0.1-sec-3' + * In those examples, the SW major,minor,subminor are correspondingly: 1,9,0. + * + * Return: 0 for success or a negative error code for failure. + */ +static int hl_get_sw_major_minor_subminor(struct hl_device *hdev, const char *fw_str) +{ + char *end, *start; + + end = strnstr(fw_str, "-rc-", VERSION_MAX_LEN); + if (end == fw_str) + return -EINVAL; + + if (!end) + end = strnstr(fw_str, "-fw-", VERSION_MAX_LEN); + + if (end == fw_str) + return -EINVAL; + + if (!end) + return -EINVAL; + + for (start = end - 1; start != fw_str; start--) { + if (*start == '-') + break; } - if (rc) { - dev_err(hdev->dev, "Error %d parsing preboot major version\n", rc); - return rc; + if (start == fw_str) + return -EINVAL; + + /* start/end point each to the starting and ending hyphen of the sw version e.g. -1.9.0- */ + start++; + start = extract_u32_until_given_char(start, &hdev->fw_sw_major_ver, '.'); + if (!start) + goto err_zero_ver; + + start++; + start = extract_u32_until_given_char(start, &hdev->fw_sw_minor_ver, '.'); + if (!start) + goto err_zero_ver; + + start++; + start = extract_u32_until_given_char(start, &hdev->fw_sw_sub_minor_ver, '-'); + if (!start) + goto err_zero_ver; + + return 0; + +err_zero_ver: + hdev->fw_sw_major_ver = 0; + hdev->fw_sw_minor_ver = 0; + hdev->fw_sw_sub_minor_ver = 0; + return -EINVAL; +} + +/** + * hl_get_preboot_major_minor() - extract the FW's version major, minor from the version string. + * @hdev: pointer to the hl_device + * @preboot_ver: the FW's version string + * + * preboot_ver is expected to be the format of ..*, e.g: 42.0.1-sec-3 + * The extracted version is set in the hdev fields: fw_inner_{major/minor}_ver. + * + * Return: 0 on success, negative error code for failure. + */ +static int hl_get_preboot_major_minor(struct hl_device *hdev, char *preboot_ver) +{ + preboot_ver = extract_u32_until_given_char(preboot_ver, &hdev->fw_inner_major_ver, '.'); + if (!preboot_ver) { + dev_err(hdev->dev, "Error parsing preboot major version\n"); + goto err_zero_ver; } - /* skip the first dot */ - first_dot++; + preboot_ver++; - second_dot = strnstr(first_dot, ".", 10); - if (second_dot) { - strscpy(minor, first_dot, second_dot - first_dot + 1); - rc = kstrtou32(minor, 10, &hdev->fw_minor_version); - } else { - rc = -EINVAL; + preboot_ver = extract_u32_until_given_char(preboot_ver, &hdev->fw_inner_minor_ver, '.'); + if (!preboot_ver) { + dev_err(hdev->dev, "Error parsing preboot minor version\n"); + goto err_zero_ver; } + return 0; - if (rc) - dev_err(hdev->dev, "Error %d parsing preboot minor version\n", rc); - return rc; +err_zero_ver: + hdev->fw_inner_major_ver = 0; + hdev->fw_inner_minor_ver = 0; + return -EINVAL; } static int hl_request_fw(struct hl_device *hdev, @@ -505,6 +591,20 @@ void hl_fw_cpu_accessible_dma_pool_free(struct hl_device *hdev, size_t size, size); } +int hl_fw_send_soft_reset(struct hl_device *hdev) +{ + struct cpucp_packet pkt; + int rc; + + memset(&pkt, 0, sizeof(pkt)); + pkt.ctl = cpu_to_le32(CPUCP_PACKET_SOFT_RESET << CPUCP_PKT_CTL_OPCODE_SHIFT); + rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), 0, NULL); + if (rc) + dev_err(hdev->dev, "failed to send soft-reset msg (err = %d)\n", rc); + + return rc; +} + int hl_fw_send_device_activity(struct hl_device *hdev, bool open) { struct cpucp_packet pkt; @@ -1268,8 +1368,10 @@ void hl_fw_ask_hard_reset_without_linux(struct hl_device *hdev) void hl_fw_ask_halt_machine_without_linux(struct hl_device *hdev) { - struct static_fw_load_mgr *static_loader = - &hdev->fw_loader.static_loader; + struct fw_load_mgr *fw_loader = &hdev->fw_loader; + u32 status, cpu_boot_status_reg, cpu_timeout; + struct static_fw_load_mgr *static_loader; + struct pre_fw_load_props *pre_fw_load; int rc; if (hdev->device_cpu_is_halted) @@ -1277,12 +1379,28 @@ void hl_fw_ask_halt_machine_without_linux(struct hl_device *hdev) /* Stop device CPU to make sure nothing bad happens */ if (hdev->asic_prop.dynamic_fw_load) { + pre_fw_load = &fw_loader->pre_fw_load; + cpu_timeout = fw_loader->cpu_timeout; + cpu_boot_status_reg = pre_fw_load->cpu_boot_status_reg; + rc = hl_fw_dynamic_send_protocol_cmd(hdev, &hdev->fw_loader, - COMMS_GOTO_WFE, 0, false, - hdev->fw_loader.cpu_timeout); - if (rc) + COMMS_GOTO_WFE, 0, false, cpu_timeout); + if (rc) { dev_err(hdev->dev, "Failed sending COMMS_GOTO_WFE\n"); + } else { + rc = hl_poll_timeout( + hdev, + cpu_boot_status_reg, + status, + status == CPU_BOOT_STATUS_IN_WFE, + hdev->fw_poll_interval_usec, + cpu_timeout); + if (rc) + dev_err(hdev->dev, "Current status=%u. Timed-out updating to WFE\n", + status); + } } else { + static_loader = &hdev->fw_loader.static_loader; WREG32(static_loader->kmd_msg_to_cpu_reg, KMD_MSG_GOTO_WFE); msleep(static_loader->cpu_reset_wait_msec); @@ -2151,6 +2269,7 @@ static int hl_fw_dynamic_read_device_fw_version(struct hl_device *hdev, struct asic_fixed_properties *prop = &hdev->asic_prop; char *preboot_ver, *boot_ver; char btl_ver[32]; + int rc; switch (fwc) { case FW_COMP_BOOT_FIT: @@ -2164,20 +2283,20 @@ static int hl_fw_dynamic_read_device_fw_version(struct hl_device *hdev, break; case FW_COMP_PREBOOT: strscpy(prop->preboot_ver, fw_version, VERSION_MAX_LEN); - preboot_ver = strnstr(prop->preboot_ver, "Preboot", - VERSION_MAX_LEN); + preboot_ver = strnstr(prop->preboot_ver, "Preboot", VERSION_MAX_LEN); + dev_info(hdev->dev, "preboot full version: '%s'\n", preboot_ver); + if (preboot_ver && preboot_ver != prop->preboot_ver) { strscpy(btl_ver, prop->preboot_ver, min((int) (preboot_ver - prop->preboot_ver), 31)); dev_info(hdev->dev, "%s\n", btl_ver); } + rc = hl_get_sw_major_minor_subminor(hdev, preboot_ver); + if (rc) + return rc; preboot_ver = extract_fw_ver_from_str(prop->preboot_ver); if (preboot_ver) { - int rc; - - dev_info(hdev->dev, "preboot version %s\n", preboot_ver); - rc = hl_get_preboot_major_minor(hdev, preboot_ver); kfree(preboot_ver); if (rc) @@ -2367,16 +2486,6 @@ static int hl_fw_dynamic_load_image(struct hl_device *hdev, if (rc) goto release_fw; - /* update state according to boot stage */ - if (cur_fwc == FW_COMP_BOOT_FIT) { - struct cpu_dyn_regs *dyn_regs; - - dyn_regs = &fw_loader->dynamic_loader.comm_desc.cpu_dyn_regs; - hl_fw_boot_fit_update_state(hdev, - le32_to_cpu(dyn_regs->cpu_boot_dev_sts0), - le32_to_cpu(dyn_regs->cpu_boot_dev_sts1)); - } - /* copy boot fit to space allocated by FW */ rc = hl_fw_dynamic_copy_image(hdev, fw, fw_loader); if (rc) @@ -2679,6 +2788,14 @@ static int hl_fw_dynamic_init_cpu(struct hl_device *hdev, goto protocol_err; } + rc = hl_fw_dynamic_wait_for_boot_fit_active(hdev, fw_loader); + if (rc) + goto protocol_err; + + hl_fw_boot_fit_update_state(hdev, + le32_to_cpu(dyn_regs->cpu_boot_dev_sts0), + le32_to_cpu(dyn_regs->cpu_boot_dev_sts1)); + /* * when testing FW load (without Linux) on PLDM we don't want to * wait until boot fit is active as it may take several hours. @@ -2688,10 +2805,6 @@ static int hl_fw_dynamic_init_cpu(struct hl_device *hdev, if (hdev->pldm && !(hdev->fw_components & FW_TYPE_LINUX)) return 0; - rc = hl_fw_dynamic_wait_for_boot_fit_active(hdev, fw_loader); - if (rc) - goto protocol_err; - /* Enable DRAM scrambling before Linux boot and after successful * UBoot */ @@ -2725,7 +2838,8 @@ static int hl_fw_dynamic_init_cpu(struct hl_device *hdev, if (rc) goto protocol_err; - hl_fw_linux_update_state(hdev, le32_to_cpu(dyn_regs->cpu_boot_dev_sts0), + hl_fw_linux_update_state(hdev, + le32_to_cpu(dyn_regs->cpu_boot_dev_sts0), le32_to_cpu(dyn_regs->cpu_boot_dev_sts1)); hl_fw_dynamic_update_linux_interrupt_if(hdev); diff --git a/drivers/accel/habanalabs/common/habanalabs.h b/drivers/accel/habanalabs/common/habanalabs.h index eaae69a9f8178fb0ce9931f07fb8d1e0622dea41..d92ba2e30e3109db63b4055472f024a60cd3678a 100644 --- a/drivers/accel/habanalabs/common/habanalabs.h +++ b/drivers/accel/habanalabs/common/habanalabs.h @@ -36,6 +36,8 @@ struct hl_device; struct hl_fpriv; +#define PCI_VENDOR_ID_HABANALABS 0x1da3 + /* Use upper bits of mmap offset to store habana driver specific information. * bits[63:59] - Encode mmap type * bits[45:0] - mmap offset value @@ -113,18 +115,6 @@ enum hl_mmu_page_table_location { MMU_NUM_PGT_LOCATIONS /* num of PGT locations */ }; -/** - * enum hl_mmu_enablement - what mmu modules to enable - * @MMU_EN_NONE: mmu disabled. - * @MMU_EN_ALL: enable all. - * @MMU_EN_PMMU_ONLY: Enable only the PMMU leaving the DMMU disabled. - */ -enum hl_mmu_enablement { - MMU_EN_NONE = 0, - MMU_EN_ALL = 1, - MMU_EN_PMMU_ONLY = 3, /* N/A for Goya/Gaudi */ -}; - /* * HL_RSVD_SOBS 'sync stream' reserved sync objects per QMAN stream * HL_RSVD_MONS 'sync stream' reserved monitors per QMAN stream @@ -2568,12 +2558,7 @@ void hl_wreg(struct hl_device *hdev, u32 reg, u32 val); ktime_t __timeout; \ u32 __elbi_read; \ int __rc = 0; \ - if (hdev->pdev) \ - __timeout = ktime_add_us(ktime_get(), timeout_us); \ - else \ - __timeout = ktime_add_us(ktime_get(),\ - min((u64)(timeout_us * 10), \ - (u64) HL_SIM_MAX_TIMEOUT_US)); \ + __timeout = ktime_add_us(ktime_get(), timeout_us); \ might_sleep_if(sleep_us); \ for (;;) { \ if (elbi) { \ @@ -2625,13 +2610,7 @@ void hl_wreg(struct hl_device *hdev, u32 reg, u32 val); u8 __arr_idx; \ int __rc = 0; \ \ - if (hdev->pdev) \ - __timeout = ktime_add_us(ktime_get(), timeout_us); \ - else \ - __timeout = ktime_add_us(ktime_get(),\ - min(((u64)timeout_us * 10), \ - (u64) HL_SIM_MAX_TIMEOUT_US)); \ - \ + __timeout = ktime_add_us(ktime_get(), timeout_us); \ might_sleep_if(sleep_us); \ if (arr_size >= 64) \ __rc = -EINVAL; \ @@ -2689,12 +2668,8 @@ void hl_wreg(struct hl_device *hdev, u32 reg, u32 val); mem_written_by_device) \ ({ \ ktime_t __timeout; \ - if (hdev->pdev) \ - __timeout = ktime_add_us(ktime_get(), timeout_us); \ - else \ - __timeout = ktime_add_us(ktime_get(),\ - min((u64)(timeout_us * 100), \ - (u64) HL_SIM_MAX_TIMEOUT_US)); \ + \ + __timeout = ktime_add_us(ktime_get(), timeout_us); \ might_sleep_if(sleep_us); \ for (;;) { \ /* Verify we read updates done by other cores or by device */ \ @@ -3225,8 +3200,11 @@ struct hl_reset_info { * @captured_err_info: holds information about errors. * @reset_info: holds current device reset information. * @stream_master_qid_arr: pointer to array with QIDs of master streams. - * @fw_major_version: major version of current loaded preboot. - * @fw_minor_version: minor version of current loaded preboot. + * @fw_inner_major_ver: the major of current loaded preboot inner version. + * @fw_inner_minor_ver: the minor of current loaded preboot inner version. + * @fw_sw_major_ver: the major of current loaded preboot SW version. + * @fw_sw_minor_ver: the minor of current loaded preboot SW version. + * @fw_sw_sub_minor_ver: the sub-minor of current loaded preboot SW version. * @dram_used_mem: current DRAM memory consumption. * @memory_scrub_val: the value to which the dram will be scrubbed to using cb scrub_device_dram * @timeout_jiffies: device CS timeout value. @@ -3287,7 +3265,7 @@ struct hl_reset_info { * @in_debug: whether the device is in a state where the profiling/tracing infrastructure * can be used. This indication is needed because in some ASICs we need to do * specific operations to enable that infrastructure. - * @cdev_sysfs_created: were char devices and sysfs nodes created. + * @cdev_sysfs_debugfs_created: were char devices and sysfs/debugfs files created. * @stop_on_err: true if engines should stop on error. * @supports_sync_stream: is sync stream supported. * @sync_stream_queue_idx: helper index for sync stream queues initialization. @@ -3314,7 +3292,7 @@ struct hl_reset_info { * @nic_ports_mask: Controls which NIC ports are enabled. Used only for testing. * @fw_components: Controls which f/w components to load to the device. There are multiple f/w * stages and sometimes we want to stop at a certain stage. Used only for testing. - * @mmu_enable: Whether to enable or disable the device MMU(s). Used only for testing. + * @mmu_disable: Disable the device MMU(s). Used only for testing. * @cpu_queues_enable: Whether to enable queues communication vs. the f/w. Used only for testing. * @pldm: Whether we are running in Palladium environment. Used only for testing. * @hard_reset_on_fw_events: Whether to do device hard-reset when a fatal event is received from @@ -3412,8 +3390,11 @@ struct hl_device { struct hl_reset_info reset_info; u32 *stream_master_qid_arr; - u32 fw_major_version; - u32 fw_minor_version; + u32 fw_inner_major_ver; + u32 fw_inner_minor_ver; + u32 fw_sw_major_ver; + u32 fw_sw_minor_ver; + u32 fw_sw_sub_minor_ver; atomic64_t dram_used_mem; u64 memory_scrub_val; u64 timeout_jiffies; @@ -3451,7 +3432,7 @@ struct hl_device { u8 init_done; u8 device_cpu_disabled; u8 in_debug; - u8 cdev_sysfs_created; + u8 cdev_sysfs_debugfs_created; u8 stop_on_err; u8 supports_sync_stream; u8 sync_stream_queue_idx; @@ -3474,7 +3455,7 @@ struct hl_device { /* Parameters for bring-up to be upstreamed */ u64 nic_ports_mask; u64 fw_components; - u8 mmu_enable; + u8 mmu_disable; u8 cpu_queues_enable; u8 pldm; u8 hard_reset_on_fw_events; @@ -3547,9 +3528,15 @@ struct hl_ioctl_desc { hl_ioctl_t *func; }; -static inline bool hl_is_fw_ver_below_1_9(struct hl_device *hdev) +static inline bool hl_is_fw_sw_ver_below(struct hl_device *hdev, u32 fw_sw_major, u32 fw_sw_minor) { - return (hdev->fw_major_version < 42); + if (hdev->fw_sw_major_ver < fw_sw_major) + return true; + if (hdev->fw_sw_major_ver > fw_sw_major) + return false; + if (hdev->fw_sw_minor_ver < fw_sw_minor) + return true; + return false; } /* @@ -3813,8 +3800,6 @@ struct pgt_info *hl_mmu_hr_get_alloc_next_hop(struct hl_ctx *ctx, u64 curr_pte, bool *is_new_hop); int hl_mmu_hr_get_tlb_info(struct hl_ctx *ctx, u64 virt_addr, struct hl_mmu_hop_info *hops, struct hl_hr_mmu_funcs *hr_func); -void hl_mmu_swap_out(struct hl_ctx *ctx); -void hl_mmu_swap_in(struct hl_ctx *ctx); int hl_mmu_if_set_funcs(struct hl_device *hdev); void hl_mmu_v1_set_funcs(struct hl_device *hdev, struct hl_mmu_funcs *mmu); void hl_mmu_v2_hr_set_funcs(struct hl_device *hdev, struct hl_mmu_funcs *mmu); @@ -3872,6 +3857,7 @@ int hl_fw_dram_replaced_row_get(struct hl_device *hdev, int hl_fw_dram_pending_row_get(struct hl_device *hdev, u32 *pend_rows_num); int hl_fw_cpucp_engine_core_asid_set(struct hl_device *hdev, u32 asid); int hl_fw_send_device_activity(struct hl_device *hdev, bool open); +int hl_fw_send_soft_reset(struct hl_device *hdev); int hl_pci_bars_map(struct hl_device *hdev, const char * const name[3], bool is_wc[3]); int hl_pci_elbi_read(struct hl_device *hdev, u64 addr, u32 *data); @@ -3921,7 +3907,7 @@ void hl_dec_fini(struct hl_device *hdev); void hl_dec_ctx_fini(struct hl_ctx *ctx); void hl_release_pending_user_interrupts(struct hl_device *hdev); -void hl_abort_waitings_for_completion(struct hl_device *hdev); +void hl_abort_waiting_for_cs_completions(struct hl_device *hdev); int hl_cs_signal_sob_wraparound_handler(struct hl_device *hdev, u32 q_idx, struct hl_hw_sob **hw_sob, u32 count, bool encaps_sig); @@ -3958,11 +3944,14 @@ void hl_handle_page_fault(struct hl_device *hdev, u64 addr, u16 eng_id, bool is_ u64 *event_mask); void hl_handle_critical_hw_err(struct hl_device *hdev, u16 event_id, u64 *event_mask); void hl_handle_fw_err(struct hl_device *hdev, struct hl_info_fw_err_info *info); +void hl_enable_err_info_capture(struct hl_error_info *captured_err_info); #ifdef CONFIG_DEBUG_FS void hl_debugfs_init(void); void hl_debugfs_fini(void); +int hl_debugfs_device_init(struct hl_device *hdev); +void hl_debugfs_device_fini(struct hl_device *hdev); void hl_debugfs_add_device(struct hl_device *hdev); void hl_debugfs_remove_device(struct hl_device *hdev); void hl_debugfs_add_file(struct hl_fpriv *hpriv); diff --git a/drivers/accel/habanalabs/common/habanalabs_drv.c b/drivers/accel/habanalabs/common/habanalabs_drv.c index d9df64e75f33ad252adc59c1d0abe40cf9e74eeb..7263e84c1a4dc3916406d6be17b197af81d3ef00 100644 --- a/drivers/accel/habanalabs/common/habanalabs_drv.c +++ b/drivers/accel/habanalabs/common/habanalabs_drv.c @@ -13,6 +13,7 @@ #include #include +#include #define CREATE_TRACE_POINTS #include @@ -54,8 +55,6 @@ module_param(boot_error_status_mask, ulong, 0444); MODULE_PARM_DESC(boot_error_status_mask, "Mask of the error status during device CPU boot (If bitX is cleared then error X is masked. Default all 1's)"); -#define PCI_VENDOR_ID_HABANALABS 0x1da3 - #define PCI_IDS_GOYA 0x0001 #define PCI_IDS_GAUDI 0x1000 #define PCI_IDS_GAUDI_SEC 0x1010 @@ -220,9 +219,7 @@ int hl_device_open(struct inode *inode, struct file *filp) hl_debugfs_add_file(hpriv); - memset(&hdev->captured_err_info, 0, sizeof(hdev->captured_err_info)); - atomic_set(&hdev->captured_err_info.cs_timeout.write_enable, 1); - hdev->captured_err_info.undef_opcode.write_enable = true; + hl_enable_err_info_capture(&hdev->captured_err_info); hdev->open_counter++; hdev->last_successful_open_jif = jiffies; @@ -307,7 +304,6 @@ static void set_driver_behavior_per_device(struct hl_device *hdev) { hdev->nic_ports_mask = 0; hdev->fw_components = FW_TYPE_ALL_TYPES; - hdev->mmu_enable = MMU_EN_ALL; hdev->cpu_queues_enable = 1; hdev->pldm = 0; hdev->hard_reset_on_fw_events = 1; @@ -382,7 +378,6 @@ static int fixup_device_params(struct hl_device *hdev) /* If CPU queues not enabled, no way to do heartbeat */ if (!hdev->cpu_queues_enable) hdev->heartbeat = 0; - fixup_device_params_per_asic(hdev, tmp_timeout); return 0; diff --git a/drivers/accel/habanalabs/common/habanalabs_ioctl.c b/drivers/accel/habanalabs/common/habanalabs_ioctl.c index 203ee857810c7bd9c539041e309eb5e705fbe327..6a45a92344e9bde61baeed7fddfce295fb4b0086 100644 --- a/drivers/accel/habanalabs/common/habanalabs_ioctl.c +++ b/drivers/accel/habanalabs/common/habanalabs_ioctl.c @@ -62,7 +62,7 @@ static int hw_ip_info(struct hl_device *hdev, struct hl_info_args *args) hw_ip.device_id = hdev->asic_funcs->get_pci_id(hdev); hw_ip.sram_base_address = prop->sram_user_base_address; hw_ip.dram_base_address = - hdev->mmu_enable && prop->dram_supports_virtual_memory ? + prop->dram_supports_virtual_memory ? prop->dmmu.start_addr : prop->dram_user_base_address; hw_ip.tpc_enabled_mask = prop->tpc_enabled_mask & 0xFF; hw_ip.tpc_enabled_mask_ext = prop->tpc_enabled_mask; @@ -71,11 +71,8 @@ static int hw_ip_info(struct hl_device *hdev, struct hl_info_args *args) dram_available_size = prop->dram_size - dram_kmd_size; - if (hdev->mmu_enable == MMU_EN_ALL) - hw_ip.dram_size = DIV_ROUND_DOWN_ULL(dram_available_size, - prop->dram_page_size) * prop->dram_page_size; - else - hw_ip.dram_size = dram_available_size; + hw_ip.dram_size = DIV_ROUND_DOWN_ULL(dram_available_size, prop->dram_page_size) * + prop->dram_page_size; if (hw_ip.dram_size > PAGE_SIZE) hw_ip.dram_enabled = 1; @@ -842,15 +839,15 @@ static int hw_err_info(struct hl_fpriv *hpriv, struct hl_info_args *args) struct hw_err_info *info; int rc; - if ((!user_buf_size) || (!user_buf)) + if (!user_buf) return -EINVAL; - if (user_buf_size < sizeof(struct hl_info_hw_err_event)) - return -ENOMEM; - info = &hdev->captured_err_info.hw_err; if (!info->event_info_available) - return -ENOENT; + return 0; + + if (user_buf_size < sizeof(struct hl_info_hw_err_event)) + return -ENOMEM; rc = copy_to_user(user_buf, &info->event, sizeof(struct hl_info_hw_err_event)); return rc ? -EFAULT : 0; @@ -864,15 +861,15 @@ static int fw_err_info(struct hl_fpriv *hpriv, struct hl_info_args *args) struct fw_err_info *info; int rc; - if ((!user_buf_size) || (!user_buf)) + if (!user_buf) return -EINVAL; - if (user_buf_size < sizeof(struct hl_info_fw_err_event)) - return -ENOMEM; - info = &hdev->captured_err_info.fw_err; if (!info->event_info_available) - return -ENOENT; + return 0; + + if (user_buf_size < sizeof(struct hl_info_fw_err_event)) + return -ENOMEM; rc = copy_to_user(user_buf, &info->event, sizeof(struct hl_info_fw_err_event)); return rc ? -EFAULT : 0; @@ -1198,7 +1195,7 @@ static long _hl_ioctl(struct file *filep, unsigned int cmd, unsigned long arg, out_err: if (retcode) - dev_dbg(dev, "error in ioctl: pid=%d, cmd=0x%02x, nr=0x%02x\n", + dev_dbg_ratelimited(dev, "error in ioctl: pid=%d, cmd=0x%02x, nr=0x%02x\n", task_pid_nr(current), cmd, nr); if (kdata != stack_kdata) @@ -1222,7 +1219,7 @@ long hl_ioctl(struct file *filep, unsigned int cmd, unsigned long arg) if ((nr >= HL_COMMAND_START) && (nr < HL_COMMAND_END)) { ioctl = &hl_ioctls[nr]; } else { - dev_err(hdev->dev, "invalid ioctl: pid=%d, nr=0x%02x\n", + dev_dbg_ratelimited(hdev->dev, "invalid ioctl: pid=%d, nr=0x%02x\n", task_pid_nr(current), nr); return -ENOTTY; } @@ -1245,7 +1242,7 @@ long hl_ioctl_control(struct file *filep, unsigned int cmd, unsigned long arg) if (nr == _IOC_NR(HL_IOCTL_INFO)) { ioctl = &hl_ioctls_control[nr]; } else { - dev_err(hdev->dev_ctrl, "invalid ioctl: pid=%d, nr=0x%02x\n", + dev_dbg_ratelimited(hdev->dev_ctrl, "invalid ioctl: pid=%d, nr=0x%02x\n", task_pid_nr(current), nr); return -ENOTTY; } diff --git a/drivers/accel/habanalabs/common/irq.c b/drivers/accel/habanalabs/common/irq.c index c67895b1cdeb18b91222bff22647d03a8d2d81e7..b1010d206c2ef1fa95914fd1aa92b2d1772e7b0a 100644 --- a/drivers/accel/habanalabs/common/irq.c +++ b/drivers/accel/habanalabs/common/irq.c @@ -430,7 +430,7 @@ irqreturn_t hl_irq_handler_eq(int irq, void *arg) cur_eqe_index = FIELD_GET(EQ_CTL_INDEX_MASK, cur_eqe); if ((hdev->event_queue.check_eqe_index) && (((eq->prev_eqe_index + 1) & EQ_CTL_INDEX_MASK) != cur_eqe_index)) { - dev_dbg(hdev->dev, + dev_err(hdev->dev, "EQE %#x in queue is ready but index does not match %d!=%d", cur_eqe, ((eq->prev_eqe_index + 1) & EQ_CTL_INDEX_MASK), diff --git a/drivers/accel/habanalabs/common/memory.c b/drivers/accel/habanalabs/common/memory.c index a7b6a273ce21ca926b595ae4f2f717fa234bc933..4fc72a07d2f59a9933538faaad1bb0290fe790d4 100644 --- a/drivers/accel/habanalabs/common/memory.c +++ b/drivers/accel/habanalabs/common/memory.c @@ -1034,30 +1034,6 @@ static void unmap_phys_pg_pack(struct hl_ctx *ctx, u64 vaddr, } } -static int get_paddr_from_handle(struct hl_ctx *ctx, struct hl_mem_in *args, - u64 *paddr) -{ - struct hl_device *hdev = ctx->hdev; - struct hl_vm *vm = &hdev->vm; - struct hl_vm_phys_pg_pack *phys_pg_pack; - u32 handle; - - handle = lower_32_bits(args->map_device.handle); - spin_lock(&vm->idr_lock); - phys_pg_pack = idr_find(&vm->phys_pg_pack_handles, handle); - if (!phys_pg_pack) { - spin_unlock(&vm->idr_lock); - dev_err(hdev->dev, "no match for handle %u\n", handle); - return -EINVAL; - } - - *paddr = phys_pg_pack->pages[0]; - - spin_unlock(&vm->idr_lock); - - return 0; -} - /** * map_device_va() - map the given memory. * @ctx: pointer to the context structure. @@ -2094,76 +2070,6 @@ static int export_dmabuf_from_addr(struct hl_ctx *ctx, u64 addr, u64 size, u64 o return rc; } -static int mem_ioctl_no_mmu(struct hl_fpriv *hpriv, union hl_mem_args *args) -{ - struct hl_device *hdev = hpriv->hdev; - u64 block_handle, device_addr = 0; - struct hl_ctx *ctx = hpriv->ctx; - u32 handle = 0, block_size; - int rc; - - switch (args->in.op) { - case HL_MEM_OP_ALLOC: - if (args->in.alloc.mem_size == 0) { - dev_err(hdev->dev, "alloc size must be larger than 0\n"); - rc = -EINVAL; - goto out; - } - - /* Force contiguous as there are no real MMU - * translations to overcome physical memory gaps - */ - args->in.flags |= HL_MEM_CONTIGUOUS; - rc = alloc_device_memory(ctx, &args->in, &handle); - - memset(args, 0, sizeof(*args)); - args->out.handle = (__u64) handle; - break; - - case HL_MEM_OP_FREE: - rc = free_device_memory(ctx, &args->in); - break; - - case HL_MEM_OP_MAP: - if (args->in.flags & HL_MEM_USERPTR) { - dev_err(hdev->dev, "Failed to map host memory when MMU is disabled\n"); - rc = -EPERM; - } else { - rc = get_paddr_from_handle(ctx, &args->in, &device_addr); - memset(args, 0, sizeof(*args)); - args->out.device_virt_addr = device_addr; - } - - break; - - case HL_MEM_OP_UNMAP: - rc = 0; - break; - - case HL_MEM_OP_MAP_BLOCK: - rc = map_block(hdev, args->in.map_block.block_addr, &block_handle, &block_size); - args->out.block_handle = block_handle; - args->out.block_size = block_size; - break; - - case HL_MEM_OP_EXPORT_DMABUF_FD: - dev_err(hdev->dev, "Failed to export dma-buf object when MMU is disabled\n"); - rc = -EPERM; - break; - - case HL_MEM_OP_TS_ALLOC: - rc = allocate_timestamps_buffers(hpriv, &args->in, &args->out.handle); - break; - default: - dev_err(hdev->dev, "Unknown opcode for memory IOCTL\n"); - rc = -EINVAL; - break; - } - -out: - return rc; -} - static void ts_buff_release(struct hl_mmap_mem_buf *buf) { struct hl_ts_buff *ts_buff = buf->private; @@ -2282,9 +2188,6 @@ int hl_mem_ioctl(struct hl_fpriv *hpriv, void *data) return -EBUSY; } - if (!hdev->mmu_enable) - return mem_ioctl_no_mmu(hpriv, args); - switch (args->in.op) { case HL_MEM_OP_ALLOC: if (args->in.alloc.mem_size == 0) { @@ -2779,13 +2682,10 @@ int hl_vm_ctx_init(struct hl_ctx *ctx) atomic64_set(&ctx->dram_phys_mem, 0); /* - * - If MMU is enabled, init the ranges as usual. - * - If MMU is disabled, in case of host mapping, the returned address - * is the given one. * In case of DRAM mapping, the returned address is the physical * address of the memory related to the given handle. */ - if (!ctx->hdev->mmu_enable) + if (ctx->hdev->mmu_disable) return 0; dram_range_start = prop->dmmu.start_addr; @@ -2835,7 +2735,7 @@ void hl_vm_ctx_fini(struct hl_ctx *ctx) struct hl_mem_in args; int i; - if (!hdev->mmu_enable) + if (hdev->mmu_disable) return; hl_debugfs_remove_ctx_mem_hash(hdev, ctx); diff --git a/drivers/accel/habanalabs/common/mmu/mmu.c b/drivers/accel/habanalabs/common/mmu/mmu.c index f379e5b461a6df98572b555d711940721f734d1e..b2145716c605334edfe476a2fcf1656f6f19d764 100644 --- a/drivers/accel/habanalabs/common/mmu/mmu.c +++ b/drivers/accel/habanalabs/common/mmu/mmu.c @@ -44,7 +44,7 @@ int hl_mmu_init(struct hl_device *hdev) { int rc = -EOPNOTSUPP; - if (!hdev->mmu_enable) + if (hdev->mmu_disable) return 0; mutex_init(&hdev->mmu_lock); @@ -82,7 +82,7 @@ int hl_mmu_init(struct hl_device *hdev) */ void hl_mmu_fini(struct hl_device *hdev) { - if (!hdev->mmu_enable) + if (hdev->mmu_disable) return; if (hdev->mmu_func[MMU_DR_PGT].fini != NULL) @@ -107,7 +107,7 @@ int hl_mmu_ctx_init(struct hl_ctx *ctx) struct hl_device *hdev = ctx->hdev; int rc = -EOPNOTSUPP; - if (!hdev->mmu_enable) + if (hdev->mmu_disable) return 0; if (hdev->mmu_func[MMU_DR_PGT].ctx_init != NULL) { @@ -145,7 +145,7 @@ void hl_mmu_ctx_fini(struct hl_ctx *ctx) { struct hl_device *hdev = ctx->hdev; - if (!hdev->mmu_enable) + if (hdev->mmu_disable) return; if (hdev->mmu_func[MMU_DR_PGT].ctx_fini != NULL) @@ -233,7 +233,7 @@ int hl_mmu_unmap_page(struct hl_ctx *ctx, u64 virt_addr, u32 page_size, bool flu u64 real_virt_addr; bool is_dram_addr; - if (!hdev->mmu_enable) + if (hdev->mmu_disable) return 0; is_dram_addr = hl_is_dram_va(hdev, virt_addr); @@ -301,7 +301,7 @@ int hl_mmu_map_page(struct hl_ctx *ctx, u64 virt_addr, u64 phys_addr, u32 page_s bool is_dram_addr; - if (!hdev->mmu_enable) + if (hdev->mmu_disable) return 0; is_dram_addr = hl_is_dram_va(hdev, virt_addr); @@ -472,46 +472,6 @@ int hl_mmu_unmap_contiguous(struct hl_ctx *ctx, u64 virt_addr, u32 size) return rc; } -/* - * hl_mmu_swap_out - marks all mapping of the given ctx as swapped out - * - * @ctx: pointer to the context structure - * - */ -void hl_mmu_swap_out(struct hl_ctx *ctx) -{ - struct hl_device *hdev = ctx->hdev; - - if (!hdev->mmu_enable) - return; - - if (hdev->mmu_func[MMU_DR_PGT].swap_out != NULL) - hdev->mmu_func[MMU_DR_PGT].swap_out(ctx); - - if (hdev->mmu_func[MMU_HR_PGT].swap_out != NULL) - hdev->mmu_func[MMU_HR_PGT].swap_out(ctx); -} - -/* - * hl_mmu_swap_in - marks all mapping of the given ctx as swapped in - * - * @ctx: pointer to the context structure - * - */ -void hl_mmu_swap_in(struct hl_ctx *ctx) -{ - struct hl_device *hdev = ctx->hdev; - - if (!hdev->mmu_enable) - return; - - if (hdev->mmu_func[MMU_DR_PGT].swap_in != NULL) - hdev->mmu_func[MMU_DR_PGT].swap_in(ctx); - - if (hdev->mmu_func[MMU_HR_PGT].swap_in != NULL) - hdev->mmu_func[MMU_HR_PGT].swap_in(ctx); -} - static void hl_mmu_pa_page_with_offset(struct hl_ctx *ctx, u64 virt_addr, struct hl_mmu_hop_info *hops, u64 *phys_addr) @@ -594,7 +554,7 @@ int hl_mmu_get_tlb_info(struct hl_ctx *ctx, u64 virt_addr, int pgt_residency, rc; bool is_dram_addr; - if (!hdev->mmu_enable) + if (hdev->mmu_disable) return -EOPNOTSUPP; prop = &hdev->asic_prop; @@ -625,7 +585,7 @@ int hl_mmu_get_tlb_info(struct hl_ctx *ctx, u64 virt_addr, int hl_mmu_if_set_funcs(struct hl_device *hdev) { - if (!hdev->mmu_enable) + if (hdev->mmu_disable) return 0; switch (hdev->asic_type) { diff --git a/drivers/accel/habanalabs/common/security.c b/drivers/accel/habanalabs/common/security.c index 297e6e44fd0c45104d510cf115e953bf14722ac9..fe913965dbad7b4d9e5d3f95c30219011a26a811 100644 --- a/drivers/accel/habanalabs/common/security.c +++ b/drivers/accel/habanalabs/common/security.c @@ -284,14 +284,14 @@ void hl_secure_block(struct hl_device *hdev, * @instance_offset: offset between instances * @pb_blocks: blocks array * @blocks_array_size: blocks array size - * @regs_array: register array - * @regs_array_size: register array size + * @user_regs_array: unsecured register array + * @user_regs_array_size: unsecured register array size * @mask: enabled instances mask: 1- enabled, 0- disabled */ int hl_init_pb_with_mask(struct hl_device *hdev, u32 num_dcores, u32 dcore_offset, u32 num_instances, u32 instance_offset, const u32 pb_blocks[], u32 blocks_array_size, - const u32 *regs_array, u32 regs_array_size, u64 mask) + const u32 *user_regs_array, u32 user_regs_array_size, u64 mask) { int i, j; struct hl_block_glbl_sec *glbl_sec; @@ -303,8 +303,8 @@ int hl_init_pb_with_mask(struct hl_device *hdev, u32 num_dcores, return -ENOMEM; hl_secure_block(hdev, glbl_sec, blocks_array_size); - hl_unsecure_registers(hdev, regs_array, regs_array_size, 0, pb_blocks, - glbl_sec, blocks_array_size); + hl_unsecure_registers(hdev, user_regs_array, user_regs_array_size, 0, + pb_blocks, glbl_sec, blocks_array_size); /* Fill all blocks with the same configuration */ for (i = 0 ; i < num_dcores ; i++) { @@ -336,19 +336,19 @@ int hl_init_pb_with_mask(struct hl_device *hdev, u32 num_dcores, * @instance_offset: offset between instances * @pb_blocks: blocks array * @blocks_array_size: blocks array size - * @regs_array: register array - * @regs_array_size: register array size + * @user_regs_array: unsecured register array + * @user_regs_array_size: unsecured register array size * */ int hl_init_pb(struct hl_device *hdev, u32 num_dcores, u32 dcore_offset, u32 num_instances, u32 instance_offset, const u32 pb_blocks[], u32 blocks_array_size, - const u32 *regs_array, u32 regs_array_size) + const u32 *user_regs_array, u32 user_regs_array_size) { return hl_init_pb_with_mask(hdev, num_dcores, dcore_offset, num_instances, instance_offset, pb_blocks, - blocks_array_size, regs_array, regs_array_size, - ULLONG_MAX); + blocks_array_size, user_regs_array, + user_regs_array_size, ULLONG_MAX); } /** @@ -364,15 +364,15 @@ int hl_init_pb(struct hl_device *hdev, u32 num_dcores, u32 dcore_offset, * @instance_offset: offset between instances * @pb_blocks: blocks array * @blocks_array_size: blocks array size - * @regs_range_array: register range array - * @regs_range_array_size: register range array size + * @user_regs_range_array: unsecured register range array + * @user_regs_range_array_size: unsecured register range array size * @mask: enabled instances mask: 1- enabled, 0- disabled */ int hl_init_pb_ranges_with_mask(struct hl_device *hdev, u32 num_dcores, u32 dcore_offset, u32 num_instances, u32 instance_offset, const u32 pb_blocks[], u32 blocks_array_size, - const struct range *regs_range_array, u32 regs_range_array_size, - u64 mask) + const struct range *user_regs_range_array, + u32 user_regs_range_array_size, u64 mask) { int i, j, rc = 0; struct hl_block_glbl_sec *glbl_sec; @@ -384,8 +384,8 @@ int hl_init_pb_ranges_with_mask(struct hl_device *hdev, u32 num_dcores, return -ENOMEM; hl_secure_block(hdev, glbl_sec, blocks_array_size); - rc = hl_unsecure_registers_range(hdev, regs_range_array, - regs_range_array_size, 0, pb_blocks, glbl_sec, + rc = hl_unsecure_registers_range(hdev, user_regs_range_array, + user_regs_range_array_size, 0, pb_blocks, glbl_sec, blocks_array_size); if (rc) goto free_glbl_sec; @@ -422,19 +422,20 @@ int hl_init_pb_ranges_with_mask(struct hl_device *hdev, u32 num_dcores, * @instance_offset: offset between instances * @pb_blocks: blocks array * @blocks_array_size: blocks array size - * @regs_range_array: register range array - * @regs_range_array_size: register range array size + * @user_regs_range_array: unsecured register range array + * @user_regs_range_array_size: unsecured register range array size * */ int hl_init_pb_ranges(struct hl_device *hdev, u32 num_dcores, u32 dcore_offset, u32 num_instances, u32 instance_offset, const u32 pb_blocks[], u32 blocks_array_size, - const struct range *regs_range_array, u32 regs_range_array_size) + const struct range *user_regs_range_array, + u32 user_regs_range_array_size) { return hl_init_pb_ranges_with_mask(hdev, num_dcores, dcore_offset, num_instances, instance_offset, pb_blocks, - blocks_array_size, regs_range_array, - regs_range_array_size, ULLONG_MAX); + blocks_array_size, user_regs_range_array, + user_regs_range_array_size, ULLONG_MAX); } /** @@ -447,14 +448,14 @@ int hl_init_pb_ranges(struct hl_device *hdev, u32 num_dcores, * @instance_offset: offset between instances * @pb_blocks: blocks array * @blocks_array_size: blocks array size - * @regs_array: register array - * @regs_array_size: register array size + * @user_regs_array: unsecured register array + * @user_regs_array_size: unsecured register array size * */ int hl_init_pb_single_dcore(struct hl_device *hdev, u32 dcore_offset, u32 num_instances, u32 instance_offset, const u32 pb_blocks[], u32 blocks_array_size, - const u32 *regs_array, u32 regs_array_size) + const u32 *user_regs_array, u32 user_regs_array_size) { int i, rc = 0; struct hl_block_glbl_sec *glbl_sec; @@ -466,8 +467,8 @@ int hl_init_pb_single_dcore(struct hl_device *hdev, u32 dcore_offset, return -ENOMEM; hl_secure_block(hdev, glbl_sec, blocks_array_size); - rc = hl_unsecure_registers(hdev, regs_array, regs_array_size, 0, - pb_blocks, glbl_sec, blocks_array_size); + rc = hl_unsecure_registers(hdev, user_regs_array, user_regs_array_size, + 0, pb_blocks, glbl_sec, blocks_array_size); if (rc) goto free_glbl_sec; @@ -495,8 +496,8 @@ int hl_init_pb_single_dcore(struct hl_device *hdev, u32 dcore_offset, * @instance_offset: offset between instances * @pb_blocks: blocks array * @blocks_array_size: blocks array size - * @regs_range_array: register range array - * @regs_range_array_size: register range array size + * @user_regs_range_array: unsecured register range array + * @user_regs_range_array_size: unsecured register range array size * */ int hl_init_pb_ranges_single_dcore(struct hl_device *hdev, u32 dcore_offset, diff --git a/drivers/accel/habanalabs/gaudi/gaudi.c b/drivers/accel/habanalabs/gaudi/gaudi.c index a29aa8f7b6f31ea158afb155851a4797361186af..056e2ef44afb508a2644a804aecac202a01a6605 100644 --- a/drivers/accel/habanalabs/gaudi/gaudi.c +++ b/drivers/accel/habanalabs/gaudi/gaudi.c @@ -114,13 +114,6 @@ static u32 gaudi_stream_master[GAUDI_STREAM_MASTER_ARR_SIZE] = { GAUDI_QUEUE_ID_DMA_1_3 }; -static const char gaudi_irq_name[GAUDI_MSI_ENTRIES][GAUDI_MAX_STRING_LEN] = { - "gaudi cq 0_0", "gaudi cq 0_1", "gaudi cq 0_2", "gaudi cq 0_3", - "gaudi cq 1_0", "gaudi cq 1_1", "gaudi cq 1_2", "gaudi cq 1_3", - "gaudi cq 5_0", "gaudi cq 5_1", "gaudi cq 5_2", "gaudi cq 5_3", - "gaudi cpu eq" -}; - static const u8 gaudi_dma_assignment[GAUDI_DMA_MAX] = { [GAUDI_PCI_DMA_1] = GAUDI_ENGINE_ID_DMA_0, [GAUDI_PCI_DMA_2] = GAUDI_ENGINE_ID_DMA_1, @@ -1476,8 +1469,7 @@ static int gaudi_collective_wait_create_job(struct hl_device *hdev, } /* Allocate internal mapped CB for non patched CBs */ - cb = hl_cb_kernel_create(hdev, cb_size, - hdev->mmu_enable && !patched_cb); + cb = hl_cb_kernel_create(hdev, cb_size, !patched_cb); if (!cb) { atomic64_inc(&ctx->cs_counters.out_of_mem_drop_cnt); atomic64_inc(&cntr->out_of_mem_drop_cnt); @@ -3651,9 +3643,6 @@ static int gaudi_mmu_init(struct hl_device *hdev) u64 hop0_addr; int rc, i; - if (!hdev->mmu_enable) - return 0; - if (gaudi->hw_cap_initialized & HW_CAP_MMU) return 0; diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c index b778cf764a68a1e11b6bfea31d93cd8589c62c5f..20c4583f12b0d20cfc58207e2ba67bcef08e5ba1 100644 --- a/drivers/accel/habanalabs/gaudi2/gaudi2.c +++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c @@ -57,13 +57,13 @@ #define GAUDI2_NA_EVENT_CAUSE 0xFF #define GAUDI2_NUM_OF_QM_ERR_CAUSE 18 -#define GAUDI2_NUM_OF_QM_LCP_ERR_CAUSE 25 +#define GAUDI2_NUM_OF_LOWER_QM_ERR_CAUSE 25 #define GAUDI2_NUM_OF_QM_ARB_ERR_CAUSE 3 #define GAUDI2_NUM_OF_ARC_SEI_ERR_CAUSE 14 #define GAUDI2_NUM_OF_CPU_SEI_ERR_CAUSE 3 #define GAUDI2_NUM_OF_QM_SEI_ERR_CAUSE 2 #define GAUDI2_NUM_OF_ROT_ERR_CAUSE 22 -#define GAUDI2_NUM_OF_TPC_INTR_CAUSE 30 +#define GAUDI2_NUM_OF_TPC_INTR_CAUSE 31 #define GAUDI2_NUM_OF_DEC_ERR_CAUSE 25 #define GAUDI2_NUM_OF_MME_ERR_CAUSE 16 #define GAUDI2_NUM_OF_MME_SBTE_ERR_CAUSE 5 @@ -162,6 +162,9 @@ #define PSOC_RAZWI_ENG_STR_SIZE 128 #define PSOC_RAZWI_MAX_ENG_PER_RTR 5 +/* HW scrambles only bits 0-25 */ +#define HW_UNSCRAMBLED_BITS_MASK GENMASK_ULL(63, 26) + struct gaudi2_razwi_info { u32 axuser_xy; u32 rtr_ctrl; @@ -801,7 +804,7 @@ static const char * const gaudi2_qman_error_cause[GAUDI2_NUM_OF_QM_ERR_CAUSE] = "PQC L2H error" }; -static const char * const gaudi2_qman_lower_cp_error_cause[GAUDI2_NUM_OF_QM_LCP_ERR_CAUSE] = { +static const char * const gaudi2_lower_qman_error_cause[GAUDI2_NUM_OF_LOWER_QM_ERR_CAUSE] = { "RSVD0", "CQ AXI HBW error", "CP AXI HBW error", @@ -891,6 +894,7 @@ static const char * const gaudi2_tpc_interrupts_cause[GAUDI2_NUM_OF_TPC_INTR_CAU "invalid_lock_access", "LD_L protection violation", "ST_L protection violation", + "D$ L0CS mismatch", }; static const char * const guadi2_mme_error_cause[GAUDI2_NUM_OF_MME_ERR_CAUSE] = { @@ -3615,6 +3619,12 @@ static int gaudi2_sw_init(struct hl_device *hdev) prop->supports_compute_reset = true; + /* Event queue sanity check added in FW version 1.11 */ + if (hl_is_fw_sw_ver_below(hdev, 1, 11)) + hdev->event_queue.check_eqe_index = false; + else + hdev->event_queue.check_eqe_index = true; + hdev->asic_funcs->set_pci_memory_regions(hdev); rc = gaudi2_special_blocks_iterator_config(hdev); @@ -3630,8 +3640,8 @@ static int gaudi2_sw_init(struct hl_device *hdev) special_blocks_free: gaudi2_special_blocks_iterator_free(hdev); free_scratchpad_mem: - hl_asic_dma_pool_free(hdev, gaudi2->scratchpad_kernel_address, - gaudi2->scratchpad_bus_address); + hl_asic_dma_free_coherent(hdev, PAGE_SIZE, gaudi2->scratchpad_kernel_address, + gaudi2->scratchpad_bus_address); free_virt_msix_db_mem: hl_cpu_accessible_dma_pool_free(hdev, prop->pmmu.page_size, gaudi2->virt_msix_db_cpu_addr); free_cpu_accessible_dma_pool: @@ -4526,7 +4536,7 @@ static int gaudi2_set_tpc_engine_mode(struct hl_device *hdev, u32 engine_id, u32 reg_base = gaudi2_tpc_cfg_blocks_bases[tpc_id]; reg_addr = reg_base + TPC_CFG_STALL_OFFSET; reg_val = FIELD_PREP(DCORE0_TPC0_CFG_TPC_STALL_V_MASK, - !!(engine_command == HL_ENGINE_STALL)); + (engine_command == HL_ENGINE_STALL) ? 1 : 0); WREG32(reg_addr, reg_val); if (engine_command == HL_ENGINE_RESUME) { @@ -4550,7 +4560,7 @@ static int gaudi2_set_mme_engine_mode(struct hl_device *hdev, u32 engine_id, u32 reg_base = gaudi2_mme_ctrl_lo_blocks_bases[mme_id]; reg_addr = reg_base + MME_CTRL_LO_QM_STALL_OFFSET; reg_val = FIELD_PREP(DCORE0_MME_CTRL_LO_QM_STALL_V_MASK, - !!(engine_command == HL_ENGINE_STALL)); + (engine_command == HL_ENGINE_STALL) ? 1 : 0); WREG32(reg_addr, reg_val); return 0; @@ -4571,7 +4581,7 @@ static int gaudi2_set_edma_engine_mode(struct hl_device *hdev, u32 engine_id, u3 reg_base = gaudi2_dma_core_blocks_bases[edma_id]; reg_addr = reg_base + EDMA_CORE_CFG_STALL_OFFSET; reg_val = FIELD_PREP(DCORE0_EDMA0_CORE_CFG_1_HALT_MASK, - !!(engine_command == HL_ENGINE_STALL)); + (engine_command == HL_ENGINE_STALL) ? 1 : 0); WREG32(reg_addr, reg_val); if (engine_command == HL_ENGINE_STALL) { @@ -6148,18 +6158,24 @@ static int gaudi2_execute_soft_reset(struct hl_device *hdev, bool driver_perform u32 poll_timeout_us) { struct cpu_dyn_regs *dyn_regs = &hdev->fw_loader.dynamic_loader.comm_desc.cpu_dyn_regs; + int rc = 0; if (!driver_performs_reset) { - /* set SP to indicate reset request sent to FW */ - if (dyn_regs->cpu_rst_status) - WREG32(le32_to_cpu(dyn_regs->cpu_rst_status), CPU_RST_STATUS_NA); - else - WREG32(mmCPU_RST_STATUS_TO_HOST, CPU_RST_STATUS_NA); - - WREG32(le32_to_cpu(dyn_regs->gic_host_soft_rst_irq), - gaudi2_irq_map_table[GAUDI2_EVENT_CPU_SOFT_RESET].cpu_id); - - return gaudi2_get_soft_rst_done_indication(hdev, poll_timeout_us); + if (hl_is_fw_sw_ver_below(hdev, 1, 10)) { + /* set SP to indicate reset request sent to FW */ + if (dyn_regs->cpu_rst_status) + WREG32(le32_to_cpu(dyn_regs->cpu_rst_status), CPU_RST_STATUS_NA); + else + WREG32(mmCPU_RST_STATUS_TO_HOST, CPU_RST_STATUS_NA); + WREG32(le32_to_cpu(dyn_regs->gic_host_soft_rst_irq), + gaudi2_irq_map_table[GAUDI2_EVENT_CPU_SOFT_RESET].cpu_id); + + /* wait for f/w response */ + rc = gaudi2_get_soft_rst_done_indication(hdev, poll_timeout_us); + } else { + rc = hl_fw_send_soft_reset(hdev); + } + return rc; } /* Block access to engines, QMANs and SM during reset, these @@ -7231,7 +7247,7 @@ static bool gaudi2_get_tpc_idle_status(struct hl_device *hdev, u64 *mask_arr, u8 gaudi2_iterate_tpcs(hdev, &tpc_iter); - return tpc_idle_data.is_idle; + return *tpc_idle_data.is_idle; } static bool gaudi2_get_decoder_idle_status(struct hl_device *hdev, u64 *mask_arr, u8 mask_len, @@ -7737,137 +7753,28 @@ static bool gaudi2_handle_ecc_event(struct hl_device *hdev, u16 event_type, return !!ecc_data->is_critical; } -/* - * gaudi2_queue_idx_dec - decrement queue index (pi/ci) and handle wrap - * - * @idx: the current pi/ci value - * @q_len: the queue length (power of 2) - * - * @return the cyclically decremented index - */ -static inline u32 gaudi2_queue_idx_dec(u32 idx, u32 q_len) -{ - u32 mask = q_len - 1; - - /* - * modular decrement is equivalent to adding (queue_size -1) - * later we take LSBs to make sure the value is in the - * range [0, queue_len - 1] - */ - return (idx + q_len - 1) & mask; -} - -/** - * gaudi2_print_sw_config_stream_data - print SW config stream data - * - * @hdev: pointer to the habanalabs device structure - * @stream: the QMAN's stream - * @qman_base: base address of QMAN registers block - */ -static void gaudi2_print_sw_config_stream_data(struct hl_device *hdev, - u32 stream, u64 qman_base) -{ - u64 cq_ptr_lo, cq_ptr_hi, cq_tsize, cq_ptr; - u32 cq_ptr_lo_off, size; - - cq_ptr_lo_off = mmDCORE0_TPC0_QM_CQ_PTR_LO_1 - mmDCORE0_TPC0_QM_CQ_PTR_LO_0; - - cq_ptr_lo = qman_base + (mmDCORE0_TPC0_QM_CQ_PTR_LO_0 - mmDCORE0_TPC0_QM_BASE) + - stream * cq_ptr_lo_off; - - cq_ptr_hi = cq_ptr_lo + (mmDCORE0_TPC0_QM_CQ_PTR_HI_0 - mmDCORE0_TPC0_QM_CQ_PTR_LO_0); - - cq_tsize = cq_ptr_lo + (mmDCORE0_TPC0_QM_CQ_TSIZE_0 - mmDCORE0_TPC0_QM_CQ_PTR_LO_0); - - cq_ptr = (((u64) RREG32(cq_ptr_hi)) << 32) | RREG32(cq_ptr_lo); - size = RREG32(cq_tsize); - dev_info(hdev->dev, "stop on err: stream: %u, addr: %#llx, size: %x\n", - stream, cq_ptr, size); -} - -/** - * gaudi2_print_last_pqes_on_err - print last PQEs on error - * - * @hdev: pointer to the habanalabs device structure - * @qid_base: first QID of the QMAN (out of 4 streams) - * @stream: the QMAN's stream - * @qman_base: base address of QMAN registers block - * @pr_sw_conf: if true print the SW config stream data (CQ PTR and SIZE) - */ -static void gaudi2_print_last_pqes_on_err(struct hl_device *hdev, u32 qid_base, u32 stream, - u64 qman_base, bool pr_sw_conf) +static void print_lower_qman_data_on_err(struct hl_device *hdev, u64 qman_base) { - u32 ci, qm_ci_stream_off; - struct hl_hw_queue *q; - u64 pq_ci; - int i; - - q = &hdev->kernel_queues[qid_base + stream]; - - qm_ci_stream_off = mmDCORE0_TPC0_QM_PQ_CI_1 - mmDCORE0_TPC0_QM_PQ_CI_0; - pq_ci = qman_base + (mmDCORE0_TPC0_QM_PQ_CI_0 - mmDCORE0_TPC0_QM_BASE) + - stream * qm_ci_stream_off; - - hdev->asic_funcs->hw_queues_lock(hdev); - - if (pr_sw_conf) - gaudi2_print_sw_config_stream_data(hdev, stream, qman_base); - - ci = RREG32(pq_ci); - - /* we should start printing form ci -1 */ - ci = gaudi2_queue_idx_dec(ci, HL_QUEUE_LENGTH); - - for (i = 0; i < PQ_FETCHER_CACHE_SIZE; i++) { - struct hl_bd *bd; - u64 addr; - u32 len; - - bd = q->kernel_address; - bd += ci; - - len = le32_to_cpu(bd->len); - /* len 0 means uninitialized entry- break */ - if (!len) - break; - - addr = le64_to_cpu(bd->ptr); - - dev_info(hdev->dev, "stop on err PQE(stream %u): ci: %u, addr: %#llx, size: %x\n", - stream, ci, addr, len); - - /* get previous ci, wrap if needed */ - ci = gaudi2_queue_idx_dec(ci, HL_QUEUE_LENGTH); - } - - hdev->asic_funcs->hw_queues_unlock(hdev); -} + u32 lo, hi, cq_ptr_size, arc_cq_ptr_size; + u64 cq_ptr, arc_cq_ptr, cp_current_inst; -/** - * print_qman_data_on_err - extract QMAN data on error - * - * @hdev: pointer to the habanalabs device structure - * @qid_base: first QID of the QMAN (out of 4 streams) - * @stream: the QMAN's stream - * @qman_base: base address of QMAN registers block - * - * This function attempt to extract as much data as possible on QMAN error. - * On upper CP print the SW config stream data and last 8 PQEs. - * On lower CP print SW config data and last PQEs of ALL 4 upper CPs - */ -static void print_qman_data_on_err(struct hl_device *hdev, u32 qid_base, u32 stream, u64 qman_base) -{ - u32 i; + lo = RREG32(qman_base + QM_CQ_PTR_LO_4_OFFSET); + hi = RREG32(qman_base + QM_CQ_PTR_HI_4_OFFSET); + cq_ptr = ((u64) hi) << 32 | lo; + cq_ptr_size = RREG32(qman_base + QM_CQ_TSIZE_4_OFFSET); - if (stream != QMAN_STREAMS) { - gaudi2_print_last_pqes_on_err(hdev, qid_base, stream, qman_base, true); - return; - } + lo = RREG32(qman_base + QM_ARC_CQ_PTR_LO_OFFSET); + hi = RREG32(qman_base + QM_ARC_CQ_PTR_HI_OFFSET); + arc_cq_ptr = ((u64) hi) << 32 | lo; + arc_cq_ptr_size = RREG32(qman_base + QM_ARC_CQ_TSIZE_OFFSET); - gaudi2_print_sw_config_stream_data(hdev, stream, qman_base); + lo = RREG32(qman_base + QM_CP_CURRENT_INST_LO_4_OFFSET); + hi = RREG32(qman_base + QM_CP_CURRENT_INST_HI_4_OFFSET); + cp_current_inst = ((u64) hi) << 32 | lo; - for (i = 0 ; i < QMAN_STREAMS ; i++) - gaudi2_print_last_pqes_on_err(hdev, qid_base, i, qman_base, false); + dev_info(hdev->dev, + "LowerQM. CQ: {ptr %#llx, size %u}, ARC_CQ: {ptr %#llx, size %u}, CP: {instruction %#llx}\n", + cq_ptr, cq_ptr_size, arc_cq_ptr, arc_cq_ptr_size, cp_current_inst); } static int gaudi2_handle_qman_err_generic(struct hl_device *hdev, u16 event_type, @@ -7888,8 +7795,8 @@ static int gaudi2_handle_qman_err_generic(struct hl_device *hdev, u16 event_type continue; if (i == QMAN_STREAMS) { - snprintf(reg_desc, ARRAY_SIZE(reg_desc), "LowerCP"); - num_error_causes = GAUDI2_NUM_OF_QM_LCP_ERR_CAUSE; + snprintf(reg_desc, ARRAY_SIZE(reg_desc), "LowerQM"); + num_error_causes = GAUDI2_NUM_OF_LOWER_QM_ERR_CAUSE; } else { snprintf(reg_desc, ARRAY_SIZE(reg_desc), "stream%u", i); num_error_causes = GAUDI2_NUM_OF_QM_ERR_CAUSE; @@ -7900,12 +7807,13 @@ static int gaudi2_handle_qman_err_generic(struct hl_device *hdev, u16 event_type gaudi2_print_event(hdev, event_type, true, "%s. err cause: %s", reg_desc, i == QMAN_STREAMS ? - gaudi2_qman_lower_cp_error_cause[j] : + gaudi2_lower_qman_error_cause[j] : gaudi2_qman_error_cause[j]); error_count++; } - print_qman_data_on_err(hdev, qid_base, i, qman_base); + if (i == QMAN_STREAMS) + print_lower_qman_data_on_err(hdev, qman_base); } arb_err_val = RREG32(arb_err_addr); @@ -8033,7 +7941,7 @@ static void gaudi2_ack_module_razwi_event_handler(struct hl_device *hdev, u8 module_sub_idx, u64 *event_mask) { bool via_sft = false; - u32 hbw_rtr_id, lbw_rtr_id, dcore_id, dcore_rtr_id, eng_id; + u32 hbw_rtr_id, lbw_rtr_id, dcore_id, dcore_rtr_id, eng_id, binned_idx; u64 hbw_rtr_mstr_if_base_addr, lbw_rtr_mstr_if_base_addr; u32 hbw_shrd_aw = 0, hbw_shrd_ar = 0; u32 lbw_shrd_aw = 0, lbw_shrd_ar = 0; @@ -8041,15 +7949,21 @@ static void gaudi2_ack_module_razwi_event_handler(struct hl_device *hdev, switch (module) { case RAZWI_TPC: + sprintf(initiator_name, "TPC_%u", module_idx); + if (hdev->tpc_binning) { + binned_idx = __ffs(hdev->tpc_binning); + if (binned_idx == module_idx) + module_idx = TPC_ID_DCORE0_TPC6; + } + hbw_rtr_id = gaudi2_tpc_initiator_hbw_rtr_id[module_idx]; - if (hl_is_fw_ver_below_1_9(hdev) && + if (hl_is_fw_sw_ver_below(hdev, 1, 9) && !hdev->asic_prop.fw_security_enabled && ((module_idx == 0) || (module_idx == 1))) lbw_rtr_id = DCORE0_RTR0; else lbw_rtr_id = gaudi2_tpc_initiator_lbw_rtr_id[module_idx]; - sprintf(initiator_name, "TPC_%u", module_idx); break; case RAZWI_MME: sprintf(initiator_name, "MME_%u", module_idx); @@ -8108,9 +8022,14 @@ static void gaudi2_ack_module_razwi_event_handler(struct hl_device *hdev, sprintf(initiator_name, "NIC_%u", module_idx); break; case RAZWI_DEC: + sprintf(initiator_name, "DEC_%u", module_idx); + if (hdev->decoder_binning) { + binned_idx = __ffs(hdev->decoder_binning); + if (binned_idx == module_idx) + module_idx = DEC_ID_PCIE_VDEC1; + } hbw_rtr_id = gaudi2_dec_initiator_hbw_rtr_id[module_idx]; lbw_rtr_id = gaudi2_dec_initiator_lbw_rtr_id[module_idx]; - sprintf(initiator_name, "DEC_%u", module_idx); break; case RAZWI_ROT: hbw_rtr_id = gaudi2_rot_initiator_hbw_rtr_id[module_idx]; @@ -8251,6 +8170,7 @@ static bool gaudi2_handle_psoc_razwi_happened(struct hl_device *hdev, u32 razwi_ u16 num_of_eng, eng_id[PSOC_RAZWI_MAX_ENG_PER_RTR]; char eng_name_str[PSOC_RAZWI_ENG_STR_SIZE]; bool razwi_happened = false; + u64 addr; int i; num_of_eng = gaudi2_psoc_razwi_get_engines(common_razwi_info, ARRAY_SIZE(common_razwi_info), @@ -8269,43 +8189,53 @@ static bool gaudi2_handle_psoc_razwi_happened(struct hl_device *hdev, u32 razwi_ if (RREG32(base[i] + DEC_RAZWI_HBW_AW_SET)) { addr_hi = RREG32(base[i] + DEC_RAZWI_HBW_AW_ADDR_HI); addr_lo = RREG32(base[i] + DEC_RAZWI_HBW_AW_ADDR_LO); - dev_err(hdev->dev, + addr = ((u64)addr_hi << 32) + addr_lo; + if (addr) { + dev_err(hdev->dev, "PSOC HBW AW RAZWI: %s, address (aligned to 128 byte): 0x%llX\n", - eng_name_str, ((u64)addr_hi << 32) + addr_lo); - hl_handle_razwi(hdev, ((u64)addr_hi << 32) + addr_lo, &eng_id[0], + eng_name_str, addr); + hl_handle_razwi(hdev, addr, &eng_id[0], num_of_eng, HL_RAZWI_HBW | HL_RAZWI_WRITE, event_mask); - razwi_happened = true; + razwi_happened = true; + } } if (RREG32(base[i] + DEC_RAZWI_HBW_AR_SET)) { addr_hi = RREG32(base[i] + DEC_RAZWI_HBW_AR_ADDR_HI); addr_lo = RREG32(base[i] + DEC_RAZWI_HBW_AR_ADDR_LO); - dev_err(hdev->dev, + addr = ((u64)addr_hi << 32) + addr_lo; + if (addr) { + dev_err(hdev->dev, "PSOC HBW AR RAZWI: %s, address (aligned to 128 byte): 0x%llX\n", - eng_name_str, ((u64)addr_hi << 32) + addr_lo); - hl_handle_razwi(hdev, ((u64)addr_hi << 32) + addr_lo, &eng_id[0], + eng_name_str, addr); + hl_handle_razwi(hdev, addr, &eng_id[0], num_of_eng, HL_RAZWI_HBW | HL_RAZWI_READ, event_mask); - razwi_happened = true; + razwi_happened = true; + } } if (RREG32(base[i] + DEC_RAZWI_LBW_AW_SET)) { addr_lo = RREG32(base[i] + DEC_RAZWI_LBW_AW_ADDR); - dev_err(hdev->dev, + if (addr_lo) { + dev_err(hdev->dev, "PSOC LBW AW RAZWI: %s, address (aligned to 128 byte): 0x%X\n", eng_name_str, addr_lo); - hl_handle_razwi(hdev, addr_lo, &eng_id[0], + hl_handle_razwi(hdev, addr_lo, &eng_id[0], num_of_eng, HL_RAZWI_LBW | HL_RAZWI_WRITE, event_mask); - razwi_happened = true; + razwi_happened = true; + } } if (RREG32(base[i] + DEC_RAZWI_LBW_AR_SET)) { addr_lo = RREG32(base[i] + DEC_RAZWI_LBW_AR_ADDR); - dev_err(hdev->dev, - "PSOC LBW AR RAZWI: %s, address (aligned to 128 byte): 0x%X\n", - eng_name_str, addr_lo); - hl_handle_razwi(hdev, addr_lo, &eng_id[0], + if (addr_lo) { + dev_err(hdev->dev, + "PSOC LBW AR RAZWI: %s, address (aligned to 128 byte): 0x%X\n", + eng_name_str, addr_lo); + hl_handle_razwi(hdev, addr_lo, &eng_id[0], num_of_eng, HL_RAZWI_LBW | HL_RAZWI_READ, event_mask); - razwi_happened = true; + razwi_happened = true; + } } /* In common case the loop will break, when there is only one engine id, or * several engines with the same router. The exceptional case is with psoc razwi @@ -8789,13 +8719,13 @@ static int gaudi2_handle_kdma_core_event(struct hl_device *hdev, u16 event_type, return error_count; } -static int gaudi2_handle_dma_core_event(struct hl_device *hdev, u16 event_type, int sts_addr) +static int gaudi2_handle_dma_core_event(struct hl_device *hdev, u16 event_type, u64 intr_cause) { - u32 error_count = 0, sts_val = RREG32(sts_addr); + u32 error_count = 0; int i; for (i = 0 ; i < GAUDI2_NUM_OF_DMA_CORE_INTR_CAUSE ; i++) - if (sts_val & BIT(i)) { + if (intr_cause & BIT(i)) { gaudi2_print_event(hdev, event_type, true, "err cause: %s", gaudi2_dma_core_interrupts_cause[i]); error_count++; @@ -8806,27 +8736,6 @@ static int gaudi2_handle_dma_core_event(struct hl_device *hdev, u16 event_type, return error_count; } -static int gaudi2_handle_pdma_core_event(struct hl_device *hdev, u16 event_type, int pdma_idx) -{ - u32 sts_addr; - - sts_addr = mmPDMA0_CORE_ERR_CAUSE + pdma_idx * PDMA_OFFSET; - return gaudi2_handle_dma_core_event(hdev, event_type, sts_addr); -} - -static int gaudi2_handle_edma_core_event(struct hl_device *hdev, u16 event_type, int edma_idx) -{ - static const int edma_event_index_map[] = {2, 3, 0, 1, 6, 7, 4, 5}; - u32 sts_addr, index; - - index = edma_event_index_map[edma_idx]; - - sts_addr = mmDCORE0_EDMA0_CORE_ERR_CAUSE + - DCORE_OFFSET * (index / NUM_OF_EDMA_PER_DCORE) + - DCORE_EDMA_OFFSET * (index % NUM_OF_EDMA_PER_DCORE); - return gaudi2_handle_dma_core_event(hdev, event_type, sts_addr); -} - static void gaudi2_print_pcie_mstr_rr_mstr_if_razwi_info(struct hl_device *hdev, u64 *event_mask) { u32 mstr_if_base_addr = mmPCIE_MSTR_RR_MSTR_IF_RR_SHRD_HBW_BASE, razwi_happened_addr; @@ -8866,6 +8775,9 @@ static int gaudi2_print_pcie_addr_dec_info(struct hl_device *hdev, u16 event_typ u32 error_count = 0; int i; + gaudi2_print_event(hdev, event_type, true, + "intr_cause_data: %#llx", intr_cause_data); + for (i = 0 ; i < GAUDI2_NUM_OF_PCIE_ADDR_DEC_ERR_CAUSE ; i++) { if (!(intr_cause_data & BIT_ULL(i))) continue; @@ -8874,16 +8786,15 @@ static int gaudi2_print_pcie_addr_dec_info(struct hl_device *hdev, u16 event_typ "err cause: %s", gaudi2_pcie_addr_dec_error_cause[i]); error_count++; - switch (intr_cause_data & BIT_ULL(i)) { - case PCIE_WRAP_PCIE_IC_SEI_INTR_IND_AXI_LBW_ERR_INTR_MASK: - hl_check_for_glbl_errors(hdev); - break; - case PCIE_WRAP_PCIE_IC_SEI_INTR_IND_BAD_ACCESS_INTR_MASK: - gaudi2_print_pcie_mstr_rr_mstr_if_razwi_info(hdev, event_mask); - break; - } + /* + * Always check for LBW and HBW additional info as the indication itself is + * sometimes missing + */ } + hl_check_for_glbl_errors(hdev); + gaudi2_print_pcie_mstr_rr_mstr_if_razwi_info(hdev, event_mask); + return error_count; } @@ -8937,11 +8848,16 @@ static void gaudi2_handle_page_error(struct hl_device *hdev, u64 mmu_base, bool addr <<= 32; addr |= RREG32(mmu_base + MMU_OFFSET(mmDCORE0_HMMU0_MMU_PAGE_ERROR_CAPTURE_VA)); - if (!is_pmmu) + if (is_pmmu) { + dev_err_ratelimited(hdev->dev, "PMMU page fault on va 0x%llx\n", addr); + } else { + addr = gaudi2_mmu_descramble_addr(hdev, addr); + addr &= HW_UNSCRAMBLED_BITS_MASK; + dev_err_ratelimited(hdev->dev, "HMMU page fault on va range 0x%llx - 0x%llx\n", + addr, addr + ~HW_UNSCRAMBLED_BITS_MASK); + } - dev_err_ratelimited(hdev->dev, "%s page fault on va 0x%llx\n", - is_pmmu ? "PMMU" : "HMMU", addr); hl_handle_page_fault(hdev, addr, 0, is_pmmu, event_mask); WREG32(mmu_base + MMU_OFFSET(mmDCORE0_HMMU0_MMU_ACCESS_PAGE_ERROR_VALID), 0); @@ -9709,19 +9625,19 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent case GAUDI2_EVENT_KDMA_CH0_AXI_ERR_RSP: case GAUDI2_EVENT_KDMA0_CORE: error_count = gaudi2_handle_kdma_core_event(hdev, event_type, - le64_to_cpu(eq_entry->intr_cause.intr_cause_data)); + le64_to_cpu(eq_entry->intr_cause.intr_cause_data)); event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR; break; case GAUDI2_EVENT_HDMA2_CORE ... GAUDI2_EVENT_HDMA5_CORE: - index = event_type - GAUDI2_EVENT_HDMA2_CORE; - error_count = gaudi2_handle_edma_core_event(hdev, event_type, index); + error_count = gaudi2_handle_dma_core_event(hdev, event_type, + le64_to_cpu(eq_entry->intr_cause.intr_cause_data)); event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR; break; case GAUDI2_EVENT_PDMA0_CORE ... GAUDI2_EVENT_PDMA1_CORE: - index = event_type - GAUDI2_EVENT_PDMA0_CORE; - error_count = gaudi2_handle_pdma_core_event(hdev, event_type, index); + error_count = gaudi2_handle_dma_core_event(hdev, event_type, + le64_to_cpu(eq_entry->intr_cause.intr_cause_data)); event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR; break; diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2P.h b/drivers/accel/habanalabs/gaudi2/gaudi2P.h index 1cebe707772e75ba81b406dfb878385db252e131..5f3ce086928e3b143cd8dc1173522c80f19f40b5 100644 --- a/drivers/accel/habanalabs/gaudi2/gaudi2P.h +++ b/drivers/accel/habanalabs/gaudi2/gaudi2P.h @@ -98,7 +98,7 @@ #define GAUDI2_DEFAULT_CARD_NAME "HL225" #define QMAN_STREAMS 4 -#define PQ_FETCHER_CACHE_SIZE 8 + #define NUM_OF_MME_SBTE_PORTS 5 #define NUM_OF_MME_WB_PORTS 2 diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2_security.c b/drivers/accel/habanalabs/gaudi2/gaudi2_security.c index 694735f9e6e661dee4cea61ee24b994fbfd6d324..2742b1f801eb2a599bd08741b64267bb5d9a756d 100644 --- a/drivers/accel/habanalabs/gaudi2/gaudi2_security.c +++ b/drivers/accel/habanalabs/gaudi2/gaudi2_security.c @@ -479,6 +479,7 @@ static const u32 gaudi2_pb_dcr0_edma0_unsecured_regs[] = { mmDCORE0_EDMA0_CORE_CTX_TE_NUMROWS, mmDCORE0_EDMA0_CORE_CTX_IDX, mmDCORE0_EDMA0_CORE_CTX_IDX_INC, + mmDCORE0_EDMA0_CORE_RD_LBW_RATE_LIM_CFG, mmDCORE0_EDMA0_QM_CQ_CFG0_0, mmDCORE0_EDMA0_QM_CQ_CFG0_1, mmDCORE0_EDMA0_QM_CQ_CFG0_2, @@ -1533,6 +1534,10 @@ static const u32 gaudi2_pb_dcr0_tpc0_unsecured_regs[] = { mmDCORE0_TPC0_CFG_QM_KERNEL_CONFIG, mmDCORE0_TPC0_CFG_QM_KERNEL_ID, mmDCORE0_TPC0_CFG_QM_POWER_LOOP, + mmDCORE0_TPC0_CFG_TSB_CFG_MTRR_2_0, + mmDCORE0_TPC0_CFG_TSB_CFG_MTRR_2_1, + mmDCORE0_TPC0_CFG_TSB_CFG_MTRR_2_2, + mmDCORE0_TPC0_CFG_TSB_CFG_MTRR_2_3, mmDCORE0_TPC0_CFG_LUT_FUNC32_BASE2_ADDR_LO, mmDCORE0_TPC0_CFG_LUT_FUNC32_BASE2_ADDR_HI, mmDCORE0_TPC0_CFG_LUT_FUNC64_BASE2_ADDR_LO, @@ -1541,6 +1546,7 @@ static const u32 gaudi2_pb_dcr0_tpc0_unsecured_regs[] = { mmDCORE0_TPC0_CFG_LUT_FUNC128_BASE2_ADDR_HI, mmDCORE0_TPC0_CFG_LUT_FUNC256_BASE2_ADDR_LO, mmDCORE0_TPC0_CFG_LUT_FUNC256_BASE2_ADDR_HI, + mmDCORE0_TPC0_CFG_FP8_143_BIAS, mmDCORE0_TPC0_CFG_ROUND_CSR, mmDCORE0_TPC0_CFG_CONV_ROUND_CSR, mmDCORE0_TPC0_CFG_SEMAPHORE, @@ -3442,15 +3448,6 @@ static int gaudi2_init_protection_bits(struct hl_device *hdev) ARRAY_SIZE(gaudi2_pb_thermal_sensor0), NULL, HL_PB_NA); } - /* HBM */ - /* Temporarily skip until SW-63348 is solved - * instance_offset = mmHBM1_MC0_BASE - mmHBM0_MC0_BASE; - * rc |= hl_init_pb_with_mask(hdev, HL_PB_SHARED, HL_PB_NA, GAUDI2_HBM_NUM, - * instance_offset, gaudi2_pb_hbm, - * ARRAY_SIZE(gaudi2_pb_hbm), NULL, HL_PB_NA, - * prop->dram_enabled_mask); - */ - /* Scheduler ARCs */ instance_offset = mmARC_FARM_ARC1_AUX_BASE - mmARC_FARM_ARC0_AUX_BASE; rc |= hl_init_pb_ranges(hdev, HL_PB_SHARED, HL_PB_NA, diff --git a/drivers/accel/habanalabs/goya/goya.c b/drivers/accel/habanalabs/goya/goya.c index fb0ac9df841a7c46002a7e6b8f2aaff954d0085c..7c685e6075f6e54d3fba59742f605837cfda1a1c 100644 --- a/drivers/accel/habanalabs/goya/goya.c +++ b/drivers/accel/habanalabs/goya/goya.c @@ -2671,9 +2671,6 @@ int goya_mmu_init(struct hl_device *hdev) u64 hop0_addr; int rc, i; - if (!hdev->mmu_enable) - return 0; - if (goya->hw_cap_initialized & HW_CAP_MMU) return 0; diff --git a/drivers/accel/habanalabs/goya/goya_coresight.c b/drivers/accel/habanalabs/goya/goya_coresight.c index e7ac3046cfaae1196555be3c4f1968a9011a694c..a6d6cc38bcd845fb3c51a17f001940827105a911 100644 --- a/drivers/accel/habanalabs/goya/goya_coresight.c +++ b/drivers/accel/habanalabs/goya/goya_coresight.c @@ -371,13 +371,8 @@ static int goya_etr_validate_address(struct hl_device *hdev, u64 addr, return false; } - if (hdev->mmu_enable) { - range_start = prop->dmmu.start_addr; - range_end = prop->dmmu.end_addr; - } else { - range_start = prop->dram_user_base_address; - range_end = prop->dram_end_address; - } + range_start = prop->dmmu.start_addr; + range_end = prop->dmmu.end_addr; return hl_mem_area_inside_range(addr, size, range_start, range_end); } diff --git a/drivers/accel/habanalabs/include/common/cpucp_if.h b/drivers/accel/habanalabs/include/common/cpucp_if.h index 8bbe685458c49d0b234d60ba04734f5c58682312..33807b839c3727d56da93515c22f260b13752b7b 100644 --- a/drivers/accel/habanalabs/include/common/cpucp_if.h +++ b/drivers/accel/habanalabs/include/common/cpucp_if.h @@ -359,7 +359,7 @@ struct hl_eq_entry { union { __le64 data_placeholder; struct hl_eq_ecc_data ecc_data; - struct hl_eq_hbm_ecc_data hbm_ecc_data; /* Gaudi1 HBM */ + struct hl_eq_hbm_ecc_data hbm_ecc_data; /* Obsolete */ struct hl_eq_sm_sei_data sm_sei_data; struct cpucp_pkt_sync_err pkt_sync_err; struct hl_eq_fw_alive fw_alive; @@ -653,7 +653,7 @@ enum pq_init_status { * which address is passed via the CpuCp packet. In addition, the host's driver * passes the max size it allows the CpuCP to write to the structure, to prevent * data corruption in case of mismatched driver/FW versions. - * Relevant only to Gaudi. + * Obsolete. * * CPUCP_PACKET_GENERIC_PASSTHROUGH - * Generic opcode for all firmware info that is only passed to host @@ -665,6 +665,9 @@ enum pq_init_status { * * CPUCP_PACKET_REGISTER_INTERRUPTS - * Packet to register interrupts indicating LKD is ready to receive events from FW. + * + * CPUCP_PACKET_SOFT_RESET - + * Packet to perform soft-reset. */ enum cpucp_packet_id { @@ -731,6 +734,7 @@ enum cpucp_packet_id { CPUCP_PACKET_RESERVED11, /* not used */ CPUCP_PACKET_RESERVED12, /* internal */ CPUCP_PACKET_REGISTER_INTERRUPTS, /* internal */ + CPUCP_PACKET_SOFT_RESET, /* internal */ CPUCP_PACKET_ID_MAX /* must be last */ }; @@ -864,19 +868,19 @@ struct cpucp_array_data_packet { enum cpucp_led_index { CPUCP_LED0_INDEX = 0, CPUCP_LED1_INDEX, - CPUCP_LED2_INDEX + CPUCP_LED2_INDEX, + CPUCP_LED_MAX_INDEX = CPUCP_LED2_INDEX }; /* * enum cpucp_packet_rc - Error return code * @cpucp_packet_success -> in case of success. - * @cpucp_packet_invalid -> this is to support Goya and Gaudi platform. + * @cpucp_packet_invalid -> this is to support first generation platforms. * @cpucp_packet_fault -> in case of processing error like failing to * get device binding or semaphore etc. - * @cpucp_packet_invalid_pkt -> when cpucp packet is un-supported. This is - * supported Greco onwards. + * @cpucp_packet_invalid_pkt -> when cpucp packet is un-supported. * @cpucp_packet_invalid_params -> when checking parameter like length of buffer - * or attribute value etc. Supported Greco onwards. + * or attribute value etc. * @cpucp_packet_rc_max -> It indicates size of enum so should be at last. */ enum cpucp_packet_rc { @@ -1361,7 +1365,7 @@ struct cpucp_dev_info_signed { #define DCORE_MON_REGS_SZ 512 /* * struct dcore_monitor_regs_data - DCORE monitor regs data. - * the structure follows sync manager block layout. relevant only to Gaudi. + * the structure follows sync manager block layout. Obsolete. * @mon_pay_addrl: array of payload address low bits. * @mon_pay_addrh: array of payload address high bits. * @mon_pay_data: array of payload data. @@ -1376,7 +1380,7 @@ struct dcore_monitor_regs_data { __le32 mon_status[DCORE_MON_REGS_SZ]; }; -/* contains SM data for each SYNC_MNGR (relevant only to Gaudi) */ +/* contains SM data for each SYNC_MNGR (Obsolete) */ struct cpucp_monitor_dump { struct dcore_monitor_regs_data sync_mngr_w_s; struct dcore_monitor_regs_data sync_mngr_e_s; diff --git a/drivers/accel/habanalabs/include/common/hl_boot_if.h b/drivers/accel/habanalabs/include/common/hl_boot_if.h index c58d76a2705c5fb1b9c668898728421004956ac6..cff79f7f9f751fe9fcc33e586625a35e18f68854 100644 --- a/drivers/accel/habanalabs/include/common/hl_boot_if.h +++ b/drivers/accel/habanalabs/include/common/hl_boot_if.h @@ -35,6 +35,7 @@ enum cpu_boot_err { CPU_BOOT_ERR_TPM_FAIL = 20, CPU_BOOT_ERR_TMP_THRESH_INIT_FAIL = 21, CPU_BOOT_ERR_EEPROM_FAIL = 22, + CPU_BOOT_ERR_ENG_ARC_MEM_SCRUB_FAIL = 23, CPU_BOOT_ERR_ENABLED = 31, CPU_BOOT_ERR_SCND_EN = 63, CPU_BOOT_ERR_LAST = 64 /* we have 2 registers of 32 bits */ @@ -51,6 +52,7 @@ enum cpu_boot_err { (1 << CPU_BOOT_ERR_DEVICE_UNUSABLE_FAIL) | \ (1 << CPU_BOOT_ERR_BINNING_FAIL) | \ (1 << CPU_BOOT_ERR_DRAM_SKIPPED) | \ + (1 << CPU_BOOT_ERR_ENG_ARC_MEM_SCRUB_FAIL) | \ (1 << CPU_BOOT_ERR_EEPROM_FAIL)) /* @@ -132,6 +134,9 @@ enum cpu_boot_err { * CPU_BOOT_ERR_EEPROM_FAIL Failed reading EEPROM data. Defaults * are used. * + * CPU_BOOT_ERR_ENG_ARC_MEM_SCRUB_FAIL Failed scrubbing the Engines/ARCFarm + * memories. Boot disabled until reset. + * * CPU_BOOT_ERR0_ENABLED Error registers enabled. * This is a main indication that the * running FW populates the error @@ -157,6 +162,7 @@ enum cpu_boot_err { #define CPU_BOOT_ERR0_TPM_FAIL (1 << CPU_BOOT_ERR_TPM_FAIL) #define CPU_BOOT_ERR0_TMP_THRESH_INIT_FAIL (1 << CPU_BOOT_ERR_TMP_THRESH_INIT_FAIL) #define CPU_BOOT_ERR0_EEPROM_FAIL (1 << CPU_BOOT_ERR_EEPROM_FAIL) +#define CPU_BOOT_ERR0_ENG_ARC_MEM_SCRUB_FAIL (1 << CPU_BOOT_ERR_ENG_ARC_MEM_SCRUB_FAIL) #define CPU_BOOT_ERR0_ENABLED (1 << CPU_BOOT_ERR_ENABLED) #define CPU_BOOT_ERR1_ENABLED (1 << CPU_BOOT_ERR_ENABLED) @@ -744,36 +750,6 @@ struct comms_status { }; }; -/** - * HL_MODULES_MAX_NUM is determined by the size of modules_mask in struct - * hl_component_versions - */ -enum hl_modules { - HL_MODULES_BOOT_INFO = 0, - HL_MODULES_EEPROM, - HL_MODULES_FDT, - HL_MODULES_I2C, - HL_MODULES_LZ4, - HL_MODULES_MBEDTLS, - HL_MODULES_MAX_NUM = 16 -}; - -/** - * HL_COMPONENTS_MAX_NUM is determined by the size of components_mask in - * struct cpucp_versions - */ -enum hl_components { - HL_COMPONENTS_PID = 0, - HL_COMPONENTS_MGMT, - HL_COMPONENTS_PREBOOT, - HL_COMPONENTS_PPBOOT, - HL_COMPONENTS_ARMCP, - HL_COMPONENTS_CPLD, - HL_COMPONENTS_UBOOT, - HL_COMPONENTS_FUSE, - HL_COMPONENTS_MAX_NUM = 16 -}; - #define NAME_MAX_LEN 32 /* bytes */ struct hl_module_data { __u8 name[NAME_MAX_LEN]; @@ -787,8 +763,6 @@ struct hl_module_data { * @component: version of the component itself. * @fw_os: Firmware OS Version. * @comp_name: Name of the component. - * @modules_mask: i'th bit (from LSB) is a flag - on if module i in enum - * hl_modules is used. * @modules_counter: number of set bits in modules_mask. * @reserved: reserved for future use. * @modules: versions of the component's modules. Elborated explanation in @@ -800,9 +774,8 @@ struct hl_component_versions { __u8 component[VERSION_MAX_LEN]; __u8 fw_os[VERSION_MAX_LEN]; __u8 comp_name[NAME_MAX_LEN]; - __le16 modules_mask; __u8 modules_counter; - __u8 reserved[1]; + __u8 reserved[3]; struct hl_module_data modules[]; }; diff --git a/drivers/accel/habanalabs/include/gaudi2/asic_reg/gaudi2_regs.h b/drivers/accel/habanalabs/include/gaudi2/asic_reg/gaudi2_regs.h index 6c58af614236446929ccf3c5dfba52bb47b4ff5a..a08378d0802b559f58f472c90c1b3e49901412a5 100644 --- a/drivers/accel/habanalabs/include/gaudi2/asic_reg/gaudi2_regs.h +++ b/drivers/accel/habanalabs/include/gaudi2/asic_reg/gaudi2_regs.h @@ -242,6 +242,17 @@ #define QM_FENCE2_OFFSET (mmPDMA0_QM_CP_FENCE2_RDATA_0 - mmPDMA0_QM_BASE) #define QM_SEI_STATUS_OFFSET (mmPDMA0_QM_SEI_STATUS - mmPDMA0_QM_BASE) +#define QM_CQ_PTR_LO_4_OFFSET (mmPDMA0_QM_CQ_PTR_LO_4 - mmPDMA0_QM_BASE) +#define QM_CQ_PTR_HI_4_OFFSET (mmPDMA0_QM_CQ_PTR_HI_4 - mmPDMA0_QM_BASE) +#define QM_CQ_TSIZE_4_OFFSET (mmPDMA0_QM_CQ_TSIZE_4 - mmPDMA0_QM_BASE) + +#define QM_ARC_CQ_PTR_LO_OFFSET (mmPDMA0_QM_ARC_CQ_PTR_LO - mmPDMA0_QM_BASE) +#define QM_ARC_CQ_PTR_HI_OFFSET (mmPDMA0_QM_ARC_CQ_PTR_HI - mmPDMA0_QM_BASE) +#define QM_ARC_CQ_TSIZE_OFFSET (mmPDMA0_QM_ARC_CQ_TSIZE - mmPDMA0_QM_BASE) + +#define QM_CP_CURRENT_INST_LO_4_OFFSET (mmPDMA0_QM_CP_CURRENT_INST_LO_4 - mmPDMA0_QM_BASE) +#define QM_CP_CURRENT_INST_HI_4_OFFSET (mmPDMA0_QM_CP_CURRENT_INST_HI_4 - mmPDMA0_QM_BASE) + #define SFT_OFFSET (mmSFT1_HBW_RTR_IF0_RTR_H3_BASE - mmSFT0_HBW_RTR_IF0_RTR_H3_BASE) #define SFT_IF_RTR_OFFSET (mmSFT0_HBW_RTR_IF1_RTR_H3_BASE - mmSFT0_HBW_RTR_IF0_RTR_H3_BASE) diff --git a/drivers/accel/habanalabs/include/gaudi2/gaudi2_fw_if.h b/drivers/accel/habanalabs/include/gaudi2/gaudi2_fw_if.h index 8522f24deac036055aeeb1ec2f6d1474b4e9d205..18ca147b1c86001dc0896e074f0a3f1436a2b688 100644 --- a/drivers/accel/habanalabs/include/gaudi2/gaudi2_fw_if.h +++ b/drivers/accel/habanalabs/include/gaudi2/gaudi2_fw_if.h @@ -62,7 +62,7 @@ struct gaudi2_cold_rst_data { u32 fake_security_enable : 1; u32 fake_sig_validation_en : 1; u32 bist_skip_enable : 1; - u32 bist_need_iatu_config : 1; + u32 reserved1 : 1; u32 fake_bis_compliant : 1; u32 wd_rst_cause_arm : 1; u32 wd_rst_cause_arcpid : 1; diff --git a/drivers/accel/ivpu/Kconfig b/drivers/accel/ivpu/Kconfig index 9bdf168bf1d0e789236f19d48970d304db5c1392..1a4c4ed9d11368fffd381d66f486aa53aa5d440b 100644 --- a/drivers/accel/ivpu/Kconfig +++ b/drivers/accel/ivpu/Kconfig @@ -7,6 +7,7 @@ config DRM_ACCEL_IVPU depends on PCI && PCI_MSI select FW_LOADER select SHMEM + select GENERIC_ALLOCATOR help Choose this option if you have a system that has an 14th generation Intel CPU or newer. VPU stands for Versatile Processing Unit and it's a CPU-integrated diff --git a/drivers/accel/ivpu/ivpu_hw_mtl.c b/drivers/accel/ivpu/ivpu_hw_mtl.c index 3210f1b4a7dd3bf343c74bbf22273623f1405f28..3ff60fbbc8d9dba372e1bd866e59bfd08ce2098f 100644 --- a/drivers/accel/ivpu/ivpu_hw_mtl.c +++ b/drivers/accel/ivpu/ivpu_hw_mtl.c @@ -197,6 +197,11 @@ static void ivpu_pll_init_frequency_ratios(struct ivpu_device *vdev) hw->pll.pn_ratio = clamp_t(u8, fuse_pn_ratio, hw->pll.min_ratio, hw->pll.max_ratio); } +static int ivpu_hw_mtl_wait_for_vpuip_bar(struct ivpu_device *vdev) +{ + return REGV_POLL_FLD(MTL_VPU_HOST_SS_CPR_RST_CLR, AON, 0, 100); +} + static int ivpu_pll_drive(struct ivpu_device *vdev, bool enable) { struct ivpu_hw_info *hw = vdev->hw; @@ -239,6 +244,12 @@ static int ivpu_pll_drive(struct ivpu_device *vdev, bool enable) ivpu_err(vdev, "Timed out waiting for PLL ready status\n"); return ret; } + + ret = ivpu_hw_mtl_wait_for_vpuip_bar(vdev); + if (ret) { + ivpu_err(vdev, "Timed out waiting for VPUIP bar\n"); + return ret; + } } return 0; @@ -256,7 +267,7 @@ static int ivpu_pll_disable(struct ivpu_device *vdev) static void ivpu_boot_host_ss_rst_clr_assert(struct ivpu_device *vdev) { - u32 val = REGV_RD32(MTL_VPU_HOST_SS_CPR_RST_CLR); + u32 val = 0; val = REG_SET_FLD(MTL_VPU_HOST_SS_CPR_RST_CLR, TOP_NOC, val); val = REG_SET_FLD(MTL_VPU_HOST_SS_CPR_RST_CLR, DSS_MAS, val); @@ -743,9 +754,8 @@ static int ivpu_hw_mtl_power_down(struct ivpu_device *vdev) { int ret = 0; - if (ivpu_hw_mtl_reset(vdev)) { + if (!ivpu_hw_mtl_is_idle(vdev) && ivpu_hw_mtl_reset(vdev)) { ivpu_err(vdev, "Failed to reset the VPU\n"); - ret = -EIO; } if (ivpu_pll_disable(vdev)) { @@ -753,8 +763,10 @@ static int ivpu_hw_mtl_power_down(struct ivpu_device *vdev) ret = -EIO; } - if (ivpu_hw_mtl_d0i3_enable(vdev)) - ivpu_warn(vdev, "Failed to enable D0I3\n"); + if (ivpu_hw_mtl_d0i3_enable(vdev)) { + ivpu_err(vdev, "Failed to enter D0I3\n"); + ret = -EIO; + } return ret; } diff --git a/drivers/accel/ivpu/ivpu_hw_mtl_reg.h b/drivers/accel/ivpu/ivpu_hw_mtl_reg.h index d83ccfd9a871b6bde38b886797cc3b2a10f75060..593b8ff0741701a1006545882c37d0ff649c584c 100644 --- a/drivers/accel/ivpu/ivpu_hw_mtl_reg.h +++ b/drivers/accel/ivpu/ivpu_hw_mtl_reg.h @@ -91,6 +91,7 @@ #define MTL_VPU_HOST_SS_CPR_RST_SET_MSS_MAS_MASK BIT_MASK(11) #define MTL_VPU_HOST_SS_CPR_RST_CLR 0x00000098u +#define MTL_VPU_HOST_SS_CPR_RST_CLR_AON_MASK BIT_MASK(0) #define MTL_VPU_HOST_SS_CPR_RST_CLR_TOP_NOC_MASK BIT_MASK(1) #define MTL_VPU_HOST_SS_CPR_RST_CLR_DSS_MAS_MASK BIT_MASK(10) #define MTL_VPU_HOST_SS_CPR_RST_CLR_MSS_MAS_MASK BIT_MASK(11) diff --git a/drivers/accel/ivpu/ivpu_ipc.c b/drivers/accel/ivpu/ivpu_ipc.c index 3adcfa80fc0e556a9e48c08e1a8869474d5a18a0..fa0af59e39ab66f0a0c20d8e4bfc4892c2ff7426 100644 --- a/drivers/accel/ivpu/ivpu_ipc.c +++ b/drivers/accel/ivpu/ivpu_ipc.c @@ -183,9 +183,7 @@ ivpu_ipc_send(struct ivpu_device *vdev, struct ivpu_ipc_consumer *cons, struct v struct ivpu_ipc_info *ipc = vdev->ipc; int ret; - ret = mutex_lock_interruptible(&ipc->lock); - if (ret) - return ret; + mutex_lock(&ipc->lock); if (!ipc->on) { ret = -EAGAIN; diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c index 0a09bba8da2458e9e0aa6bc3de89ccb9570c768b..de9e69f70af7e215ff6f85812473158890ccef68 100644 --- a/drivers/accel/ivpu/ivpu_job.c +++ b/drivers/accel/ivpu/ivpu_job.c @@ -429,6 +429,7 @@ ivpu_job_prepare_bos_for_submit(struct drm_file *file, struct ivpu_job *job, u32 struct ivpu_file_priv *file_priv = file->driver_priv; struct ivpu_device *vdev = file_priv->vdev; struct ww_acquire_ctx acquire_ctx; + enum dma_resv_usage usage; struct ivpu_bo *bo; int ret; u32 i; @@ -459,22 +460,28 @@ ivpu_job_prepare_bos_for_submit(struct drm_file *file, struct ivpu_job *job, u32 job->cmd_buf_vpu_addr = bo->vpu_addr + commands_offset; - ret = drm_gem_lock_reservations((struct drm_gem_object **)job->bos, 1, &acquire_ctx); + ret = drm_gem_lock_reservations((struct drm_gem_object **)job->bos, buf_count, + &acquire_ctx); if (ret) { ivpu_warn(vdev, "Failed to lock reservations: %d\n", ret); return ret; } - ret = dma_resv_reserve_fences(bo->base.resv, 1); - if (ret) { - ivpu_warn(vdev, "Failed to reserve fences: %d\n", ret); - goto unlock_reservations; + for (i = 0; i < buf_count; i++) { + ret = dma_resv_reserve_fences(job->bos[i]->base.resv, 1); + if (ret) { + ivpu_warn(vdev, "Failed to reserve fences: %d\n", ret); + goto unlock_reservations; + } } - dma_resv_add_fence(bo->base.resv, job->done_fence, DMA_RESV_USAGE_WRITE); + for (i = 0; i < buf_count; i++) { + usage = (i == CMD_BUF_IDX) ? DMA_RESV_USAGE_WRITE : DMA_RESV_USAGE_BOOKKEEP; + dma_resv_add_fence(job->bos[i]->base.resv, job->done_fence, usage); + } unlock_reservations: - drm_gem_unlock_reservations((struct drm_gem_object **)job->bos, 1, &acquire_ctx); + drm_gem_unlock_reservations((struct drm_gem_object **)job->bos, buf_count, &acquire_ctx); wmb(); /* Flush write combining buffers */ diff --git a/drivers/accel/ivpu/ivpu_mmu.c b/drivers/accel/ivpu/ivpu_mmu.c index fa9a9ad5964357d6f73728f406c6d39889ccf0eb..53878e77aad3dbefa823f7741b0935014c7be8ad 100644 --- a/drivers/accel/ivpu/ivpu_mmu.c +++ b/drivers/accel/ivpu/ivpu_mmu.c @@ -597,16 +597,11 @@ static int ivpu_mmu_strtab_init(struct ivpu_device *vdev) int ivpu_mmu_invalidate_tlb(struct ivpu_device *vdev, u16 ssid) { struct ivpu_mmu_info *mmu = vdev->mmu; - int ret; - - ret = mutex_lock_interruptible(&mmu->lock); - if (ret) - return ret; + int ret = 0; - if (!mmu->on) { - ret = 0; + mutex_lock(&mmu->lock); + if (!mmu->on) goto unlock; - } ret = ivpu_mmu_cmdq_write_tlbi_nh_asid(vdev, ssid); if (ret) @@ -624,7 +619,7 @@ static int ivpu_mmu_cd_add(struct ivpu_device *vdev, u32 ssid, u64 cd_dma) struct ivpu_mmu_cdtab *cdtab = &mmu->cdtab; u64 *entry; u64 cd[4]; - int ret; + int ret = 0; if (ssid > IVPU_MMU_CDTAB_ENT_COUNT) return -EINVAL; @@ -665,14 +660,9 @@ static int ivpu_mmu_cd_add(struct ivpu_device *vdev, u32 ssid, u64 cd_dma) ivpu_dbg(vdev, MMU, "CDTAB %s entry (SSID=%u, dma=%pad): 0x%llx, 0x%llx, 0x%llx, 0x%llx\n", cd_dma ? "write" : "clear", ssid, &cd_dma, cd[0], cd[1], cd[2], cd[3]); - ret = mutex_lock_interruptible(&mmu->lock); - if (ret) - return ret; - - if (!mmu->on) { - ret = 0; + mutex_lock(&mmu->lock); + if (!mmu->on) goto unlock; - } ret = ivpu_mmu_cmdq_write_cfgi_all(vdev); if (ret) diff --git a/drivers/accel/qaic/qaic_control.c b/drivers/accel/qaic/qaic_control.c index 9f216eb6f76ed80eb0fa3706e83e291898374782..5c57f7b4494e482f46d920e34f265ae1cac0c1ba 100644 --- a/drivers/accel/qaic/qaic_control.c +++ b/drivers/accel/qaic/qaic_control.c @@ -997,14 +997,34 @@ static void *msg_xfer(struct qaic_device *qdev, struct wrapper_list *wrappers, u struct xfer_queue_elem elem; struct wire_msg *out_buf; struct wrapper_msg *w; + long ret = -EAGAIN; + int xfer_count = 0; int retry_count; - long ret; if (qdev->in_reset) { mutex_unlock(&qdev->cntl_mutex); return ERR_PTR(-ENODEV); } + /* Attempt to avoid a partial commit of a message */ + list_for_each_entry(w, &wrappers->list, list) + xfer_count++; + + for (retry_count = 0; retry_count < QAIC_MHI_RETRY_MAX; retry_count++) { + if (xfer_count <= mhi_get_free_desc_count(qdev->cntl_ch, DMA_TO_DEVICE)) { + ret = 0; + break; + } + msleep_interruptible(QAIC_MHI_RETRY_WAIT_MS); + if (signal_pending(current)) + break; + } + + if (ret) { + mutex_unlock(&qdev->cntl_mutex); + return ERR_PTR(ret); + } + elem.seq_num = seq_num; elem.buf = NULL; init_completion(&elem.xfer_done); @@ -1038,16 +1058,9 @@ static void *msg_xfer(struct qaic_device *qdev, struct wrapper_list *wrappers, u list_for_each_entry(w, &wrappers->list, list) { kref_get(&w->ref_count); retry_count = 0; -retry: ret = mhi_queue_buf(qdev->cntl_ch, DMA_TO_DEVICE, &w->msg, w->len, list_is_last(&w->list, &wrappers->list) ? MHI_EOT : MHI_CHAIN); if (ret) { - if (ret == -EAGAIN && retry_count++ < QAIC_MHI_RETRY_MAX) { - msleep_interruptible(QAIC_MHI_RETRY_WAIT_MS); - if (!signal_pending(current)) - goto retry; - } - qdev->cntl_lost_buf = true; kref_put(&w->ref_count, free_wrapper); mutex_unlock(&qdev->cntl_mutex); @@ -1249,7 +1262,7 @@ static int qaic_manage(struct qaic_device *qdev, struct qaic_user *usr, struct m int qaic_manage_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv) { - struct qaic_manage_msg *user_msg; + struct qaic_manage_msg *user_msg = data; struct qaic_device *qdev; struct manage_msg *msg; struct qaic_user *usr; @@ -1258,6 +1271,9 @@ int qaic_manage_ioctl(struct drm_device *dev, void *data, struct drm_file *file_ int usr_rcu_id; int ret; + if (user_msg->len > QAIC_MANAGE_MAX_MSG_LENGTH) + return -EINVAL; + usr = file_priv->driver_priv; usr_rcu_id = srcu_read_lock(&usr->qddev_lock); @@ -1275,13 +1291,6 @@ int qaic_manage_ioctl(struct drm_device *dev, void *data, struct drm_file *file_ return -ENODEV; } - user_msg = data; - - if (user_msg->len > QAIC_MANAGE_MAX_MSG_LENGTH) { - ret = -EINVAL; - goto out; - } - msg = kzalloc(QAIC_MANAGE_MAX_MSG_LENGTH + sizeof(*msg), GFP_KERNEL); if (!msg) { ret = -ENOMEM; diff --git a/drivers/accel/qaic/qaic_data.c b/drivers/accel/qaic/qaic_data.c index c0a574cd1b35c6b80696e0b3d46380eef5008e6f..e42c1f9ffff85eb72650b2bc75590d924d4f3d02 100644 --- a/drivers/accel/qaic/qaic_data.c +++ b/drivers/accel/qaic/qaic_data.c @@ -591,7 +591,7 @@ static int qaic_gem_object_mmap(struct drm_gem_object *obj, struct vm_area_struc struct qaic_bo *bo = to_qaic_bo(obj); unsigned long offset = 0; struct scatterlist *sg; - int ret; + int ret = 0; if (obj->import_attach) return -EINVAL; @@ -663,6 +663,10 @@ int qaic_create_bo_ioctl(struct drm_device *dev, void *data, struct drm_file *fi if (args->pad) return -EINVAL; + size = PAGE_ALIGN(args->size); + if (size == 0) + return -EINVAL; + usr = file_priv->driver_priv; usr_rcu_id = srcu_read_lock(&usr->qddev_lock); if (!usr->qddev) { @@ -677,12 +681,6 @@ int qaic_create_bo_ioctl(struct drm_device *dev, void *data, struct drm_file *fi goto unlock_dev_srcu; } - size = PAGE_ALIGN(args->size); - if (size == 0) { - ret = -EINVAL; - goto unlock_dev_srcu; - } - bo = qaic_alloc_init_bo(); if (IS_ERR(bo)) { ret = PTR_ERR(bo); @@ -926,8 +924,8 @@ int qaic_attach_slice_bo_ioctl(struct drm_device *dev, void *data, struct drm_fi { struct qaic_attach_slice_entry *slice_ent; struct qaic_attach_slice *args = data; + int rcu_id, usr_rcu_id, qdev_rcu_id; struct dma_bridge_chan *dbc; - int usr_rcu_id, qdev_rcu_id; struct drm_gem_object *obj; struct qaic_device *qdev; unsigned long arg_size; @@ -936,6 +934,22 @@ int qaic_attach_slice_bo_ioctl(struct drm_device *dev, void *data, struct drm_fi struct qaic_bo *bo; int ret; + if (args->hdr.count == 0) + return -EINVAL; + + arg_size = args->hdr.count * sizeof(*slice_ent); + if (arg_size / args->hdr.count != sizeof(*slice_ent)) + return -EINVAL; + + if (args->hdr.size == 0) + return -EINVAL; + + if (!(args->hdr.dir == DMA_TO_DEVICE || args->hdr.dir == DMA_FROM_DEVICE)) + return -EINVAL; + + if (args->data == 0) + return -EINVAL; + usr = file_priv->driver_priv; usr_rcu_id = srcu_read_lock(&usr->qddev_lock); if (!usr->qddev) { @@ -950,43 +964,11 @@ int qaic_attach_slice_bo_ioctl(struct drm_device *dev, void *data, struct drm_fi goto unlock_dev_srcu; } - if (args->hdr.count == 0) { - ret = -EINVAL; - goto unlock_dev_srcu; - } - - arg_size = args->hdr.count * sizeof(*slice_ent); - if (arg_size / args->hdr.count != sizeof(*slice_ent)) { - ret = -EINVAL; - goto unlock_dev_srcu; - } - if (args->hdr.dbc_id >= qdev->num_dbc) { ret = -EINVAL; goto unlock_dev_srcu; } - if (args->hdr.size == 0) { - ret = -EINVAL; - goto unlock_dev_srcu; - } - - if (!(args->hdr.dir == DMA_TO_DEVICE || args->hdr.dir == DMA_FROM_DEVICE)) { - ret = -EINVAL; - goto unlock_dev_srcu; - } - - dbc = &qdev->dbc[args->hdr.dbc_id]; - if (dbc->usr != usr) { - ret = -EINVAL; - goto unlock_dev_srcu; - } - - if (args->data == 0) { - ret = -EINVAL; - goto unlock_dev_srcu; - } - user_data = u64_to_user_ptr(args->data); slice_ent = kzalloc(arg_size, GFP_KERNEL); @@ -1013,9 +995,21 @@ int qaic_attach_slice_bo_ioctl(struct drm_device *dev, void *data, struct drm_fi bo = to_qaic_bo(obj); + if (bo->sliced) { + ret = -EINVAL; + goto put_bo; + } + + dbc = &qdev->dbc[args->hdr.dbc_id]; + rcu_id = srcu_read_lock(&dbc->ch_lock); + if (dbc->usr != usr) { + ret = -EINVAL; + goto unlock_ch_srcu; + } + ret = qaic_prepare_bo(qdev, bo, &args->hdr); if (ret) - goto put_bo; + goto unlock_ch_srcu; ret = qaic_attach_slicing_bo(qdev, bo, &args->hdr, slice_ent); if (ret) @@ -1025,6 +1019,7 @@ int qaic_attach_slice_bo_ioctl(struct drm_device *dev, void *data, struct drm_fi dma_sync_sgtable_for_cpu(&qdev->pdev->dev, bo->sgt, args->hdr.dir); bo->dbc = dbc; + srcu_read_unlock(&dbc->ch_lock, rcu_id); drm_gem_object_put(obj); srcu_read_unlock(&qdev->dev_lock, qdev_rcu_id); srcu_read_unlock(&usr->qddev_lock, usr_rcu_id); @@ -1033,6 +1028,8 @@ int qaic_attach_slice_bo_ioctl(struct drm_device *dev, void *data, struct drm_fi unprepare_bo: qaic_unprepare_bo(qdev, bo); +unlock_ch_srcu: + srcu_read_unlock(&dbc->ch_lock, rcu_id); put_bo: drm_gem_object_put(obj); free_slice_ent: @@ -1316,7 +1313,6 @@ static int __qaic_execute_bo_ioctl(struct drm_device *dev, void *data, struct dr received_ts = ktime_get_ns(); size = is_partial ? sizeof(*pexec) : sizeof(*exec); - n = (unsigned long)size * args->hdr.count; if (args->hdr.count == 0 || n / args->hdr.count != size) return -EINVAL; @@ -1665,6 +1661,9 @@ int qaic_wait_bo_ioctl(struct drm_device *dev, void *data, struct drm_file *file int rcu_id; int ret; + if (args->pad != 0) + return -EINVAL; + usr = file_priv->driver_priv; usr_rcu_id = srcu_read_lock(&usr->qddev_lock); if (!usr->qddev) { @@ -1679,11 +1678,6 @@ int qaic_wait_bo_ioctl(struct drm_device *dev, void *data, struct drm_file *file goto unlock_dev_srcu; } - if (args->pad != 0) { - ret = -EINVAL; - goto unlock_dev_srcu; - } - if (args->dbc_id >= qdev->num_dbc) { ret = -EINVAL; goto unlock_dev_srcu; @@ -1855,6 +1849,11 @@ void wakeup_dbc(struct qaic_device *qdev, u32 dbc_id) dbc->usr = NULL; empty_xfer_list(qdev, dbc); synchronize_srcu(&dbc->ch_lock); + /* + * Threads holding channel lock, may add more elements in the xfer_list. + * Flush out these elements from xfer_list. + */ + empty_xfer_list(qdev, dbc); } void release_dbc(struct qaic_device *qdev, u32 dbc_id) diff --git a/drivers/accel/qaic/qaic_drv.c b/drivers/accel/qaic/qaic_drv.c index ff80eb5717290c7d72ed3585bd0dd982dfe0f9ce..b5ba550a0c0400e03d3875bcbc5095c73582c06d 100644 --- a/drivers/accel/qaic/qaic_drv.c +++ b/drivers/accel/qaic/qaic_drv.c @@ -97,6 +97,7 @@ static int qaic_open(struct drm_device *dev, struct drm_file *file) cleanup_usr: cleanup_srcu_struct(&usr->qddev_lock); + ida_free(&qaic_usrs, usr->handle); free_usr: kfree(usr); dev_unlock: @@ -224,6 +225,9 @@ static void qaic_destroy_drm_device(struct qaic_device *qdev, s32 partition_id) struct qaic_user *usr; qddev = qdev->qddev; + qdev->qddev = NULL; + if (!qddev) + return; /* * Existing users get unresolvable errors till they close FDs. @@ -262,8 +266,8 @@ static void qaic_destroy_drm_device(struct qaic_device *qdev, s32 partition_id) static int qaic_mhi_probe(struct mhi_device *mhi_dev, const struct mhi_device_id *id) { + u16 major = -1, minor = -1; struct qaic_device *qdev; - u16 major, minor; int ret; /* diff --git a/drivers/acpi/apei/apei-internal.h b/drivers/acpi/apei/apei-internal.h index 1d6ef96547252d510b570858d937bc7c222586fe..67c2c3b959e1538a645d43bda3449378c0051ed8 100644 --- a/drivers/acpi/apei/apei-internal.h +++ b/drivers/acpi/apei/apei-internal.h @@ -7,7 +7,6 @@ #ifndef APEI_INTERNAL_H #define APEI_INTERNAL_H -#include #include struct apei_exec_context; @@ -130,10 +129,5 @@ static inline u32 cper_estatus_len(struct acpi_hest_generic_status *estatus) return sizeof(*estatus) + estatus->data_length; } -void cper_estatus_print(const char *pfx, - const struct acpi_hest_generic_status *estatus); -int cper_estatus_check_header(const struct acpi_hest_generic_status *estatus); -int cper_estatus_check(const struct acpi_hest_generic_status *estatus); - int apei_osc_setup(void); #endif diff --git a/drivers/acpi/apei/bert.c b/drivers/acpi/apei/bert.c index c23eb75866d0214fa98688e1206d106effb7cca7..7514e38d5640eed18483897f9ddfd867061d9fa9 100644 --- a/drivers/acpi/apei/bert.c +++ b/drivers/acpi/apei/bert.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include "apei-internal.h" diff --git a/drivers/acpi/resource.c b/drivers/acpi/resource.c index e8492b3a393ab6932c1f6c791bee1c442688fb64..0800a9d77558069edc16243098183b16f5f78e39 100644 --- a/drivers/acpi/resource.c +++ b/drivers/acpi/resource.c @@ -516,6 +516,17 @@ static const struct dmi_system_id maingear_laptop[] = { { } }; +static const struct dmi_system_id lg_laptop[] = { + { + .ident = "LG Electronics 17U70P", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "LG Electronics"), + DMI_MATCH(DMI_BOARD_NAME, "17U70P"), + }, + }, + { } +}; + struct irq_override_cmp { const struct dmi_system_id *system; unsigned char irq; @@ -532,6 +543,7 @@ static const struct irq_override_cmp override_table[] = { { lenovo_laptop, 10, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0, true }, { tongfang_gm_rg, 1, ACPI_EDGE_SENSITIVE, ACPI_ACTIVE_LOW, 1, true }, { maingear_laptop, 1, ACPI_EDGE_SENSITIVE, ACPI_ACTIVE_LOW, 1, true }, + { lg_laptop, 1, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0, false }, }; static bool acpi_dev_irq_override(u32 gsi, u8 triggering, u8 polarity, diff --git a/drivers/android/binder.c b/drivers/android/binder.c index fb56bfc45096d4a8a645dc30f7441b77c6c20c33..8fb7672021ee2d2fc06570c29dbcb2b6f09bc993 100644 --- a/drivers/android/binder.c +++ b/drivers/android/binder.c @@ -1934,24 +1934,23 @@ static void binder_deferred_fd_close(int fd) static void binder_transaction_buffer_release(struct binder_proc *proc, struct binder_thread *thread, struct binder_buffer *buffer, - binder_size_t failed_at, + binder_size_t off_end_offset, bool is_failure) { int debug_id = buffer->debug_id; - binder_size_t off_start_offset, buffer_offset, off_end_offset; + binder_size_t off_start_offset, buffer_offset; binder_debug(BINDER_DEBUG_TRANSACTION, "%d buffer release %d, size %zd-%zd, failed at %llx\n", proc->pid, buffer->debug_id, buffer->data_size, buffer->offsets_size, - (unsigned long long)failed_at); + (unsigned long long)off_end_offset); if (buffer->target_node) binder_dec_node(buffer->target_node, 1, 0); off_start_offset = ALIGN(buffer->data_size, sizeof(void *)); - off_end_offset = is_failure && failed_at ? failed_at : - off_start_offset + buffer->offsets_size; + for (buffer_offset = off_start_offset; buffer_offset < off_end_offset; buffer_offset += sizeof(binder_size_t)) { struct binder_object_header *hdr; @@ -2111,6 +2110,21 @@ static void binder_transaction_buffer_release(struct binder_proc *proc, } } +/* Clean up all the objects in the buffer */ +static inline void binder_release_entire_buffer(struct binder_proc *proc, + struct binder_thread *thread, + struct binder_buffer *buffer, + bool is_failure) +{ + binder_size_t off_end_offset; + + off_end_offset = ALIGN(buffer->data_size, sizeof(void *)); + off_end_offset += buffer->offsets_size; + + binder_transaction_buffer_release(proc, thread, buffer, + off_end_offset, is_failure); +} + static int binder_translate_binder(struct flat_binder_object *fp, struct binder_transaction *t, struct binder_thread *thread) @@ -2806,7 +2820,7 @@ static int binder_proc_transaction(struct binder_transaction *t, t_outdated->buffer = NULL; buffer->transaction = NULL; trace_binder_transaction_update_buffer_release(buffer); - binder_transaction_buffer_release(proc, NULL, buffer, 0, 0); + binder_release_entire_buffer(proc, NULL, buffer, false); binder_alloc_free_buf(&proc->alloc, buffer); kfree(t_outdated); binder_stats_deleted(BINDER_STAT_TRANSACTION); @@ -3775,7 +3789,7 @@ binder_free_buf(struct binder_proc *proc, binder_node_inner_unlock(buf_node); } trace_binder_transaction_buffer_release(buffer); - binder_transaction_buffer_release(proc, thread, buffer, 0, is_failure); + binder_release_entire_buffer(proc, thread, buffer, is_failure); binder_alloc_free_buf(&proc->alloc, buffer); } diff --git a/drivers/android/binder_alloc.c b/drivers/android/binder_alloc.c index 55a3c3c2409f0108ccbff79307e45036ad7c7701..662a2a2e2e84a77b6fc9294d34c376f3e27712f2 100644 --- a/drivers/android/binder_alloc.c +++ b/drivers/android/binder_alloc.c @@ -212,8 +212,8 @@ static int binder_update_page_range(struct binder_alloc *alloc, int allocate, mm = alloc->mm; if (mm) { - mmap_read_lock(mm); - vma = vma_lookup(mm, alloc->vma_addr); + mmap_write_lock(mm); + vma = alloc->vma; } if (!vma && need_mm) { @@ -270,7 +270,7 @@ static int binder_update_page_range(struct binder_alloc *alloc, int allocate, trace_binder_alloc_page_end(alloc, index); } if (mm) { - mmap_read_unlock(mm); + mmap_write_unlock(mm); mmput(mm); } return 0; @@ -303,21 +303,24 @@ static int binder_update_page_range(struct binder_alloc *alloc, int allocate, } err_no_vma: if (mm) { - mmap_read_unlock(mm); + mmap_write_unlock(mm); mmput(mm); } return vma ? -ENOMEM : -ESRCH; } +static inline void binder_alloc_set_vma(struct binder_alloc *alloc, + struct vm_area_struct *vma) +{ + /* pairs with smp_load_acquire in binder_alloc_get_vma() */ + smp_store_release(&alloc->vma, vma); +} + static inline struct vm_area_struct *binder_alloc_get_vma( struct binder_alloc *alloc) { - struct vm_area_struct *vma = NULL; - - if (alloc->vma_addr) - vma = vma_lookup(alloc->mm, alloc->vma_addr); - - return vma; + /* pairs with smp_store_release in binder_alloc_set_vma() */ + return smp_load_acquire(&alloc->vma); } static bool debug_low_async_space_locked(struct binder_alloc *alloc, int pid) @@ -380,15 +383,13 @@ static struct binder_buffer *binder_alloc_new_buf_locked( size_t size, data_offsets_size; int ret; - mmap_read_lock(alloc->mm); + /* Check binder_alloc is fully initialized */ if (!binder_alloc_get_vma(alloc)) { - mmap_read_unlock(alloc->mm); binder_alloc_debug(BINDER_DEBUG_USER_ERROR, "%d: binder_alloc_buf, no vma\n", alloc->pid); return ERR_PTR(-ESRCH); } - mmap_read_unlock(alloc->mm); data_offsets_size = ALIGN(data_size, sizeof(void *)) + ALIGN(offsets_size, sizeof(void *)); @@ -778,7 +779,9 @@ int binder_alloc_mmap_handler(struct binder_alloc *alloc, buffer->free = 1; binder_insert_free_buffer(alloc, buffer); alloc->free_async_space = alloc->buffer_size / 2; - alloc->vma_addr = vma->vm_start; + + /* Signal binder_alloc is fully initialized */ + binder_alloc_set_vma(alloc, vma); return 0; @@ -808,8 +811,7 @@ void binder_alloc_deferred_release(struct binder_alloc *alloc) buffers = 0; mutex_lock(&alloc->mutex); - BUG_ON(alloc->vma_addr && - vma_lookup(alloc->mm, alloc->vma_addr)); + BUG_ON(alloc->vma); while ((n = rb_first(&alloc->allocated_buffers))) { buffer = rb_entry(n, struct binder_buffer, rb_node); @@ -916,25 +918,17 @@ void binder_alloc_print_pages(struct seq_file *m, * Make sure the binder_alloc is fully initialized, otherwise we might * read inconsistent state. */ - - mmap_read_lock(alloc->mm); - if (binder_alloc_get_vma(alloc) == NULL) { - mmap_read_unlock(alloc->mm); - goto uninitialized; - } - - mmap_read_unlock(alloc->mm); - for (i = 0; i < alloc->buffer_size / PAGE_SIZE; i++) { - page = &alloc->pages[i]; - if (!page->page_ptr) - free++; - else if (list_empty(&page->lru)) - active++; - else - lru++; + if (binder_alloc_get_vma(alloc) != NULL) { + for (i = 0; i < alloc->buffer_size / PAGE_SIZE; i++) { + page = &alloc->pages[i]; + if (!page->page_ptr) + free++; + else if (list_empty(&page->lru)) + active++; + else + lru++; + } } - -uninitialized: mutex_unlock(&alloc->mutex); seq_printf(m, " pages: %d:%d:%d\n", active, lru, free); seq_printf(m, " pages high watermark: %zu\n", alloc->pages_high); @@ -969,7 +963,7 @@ int binder_alloc_get_allocated_count(struct binder_alloc *alloc) */ void binder_alloc_vma_close(struct binder_alloc *alloc) { - alloc->vma_addr = 0; + binder_alloc_set_vma(alloc, NULL); } /** diff --git a/drivers/android/binder_alloc.h b/drivers/android/binder_alloc.h index 0f811ac4bcffd1f922a6bd895cd5682ad9acd08c..138d1d5af9ce36051893ee50a8e728d7bfbc2940 100644 --- a/drivers/android/binder_alloc.h +++ b/drivers/android/binder_alloc.h @@ -75,7 +75,7 @@ struct binder_lru_page { /** * struct binder_alloc - per-binder proc state for binder allocator * @mutex: protects binder_alloc fields - * @vma_addr: vm_area_struct->vm_start passed to mmap_handler + * @vma: vm_area_struct passed to mmap_handler * (invariant after mmap) * @mm: copy of task->mm (invariant after open) * @buffer: base of per-proc address space mapped via mmap @@ -99,7 +99,7 @@ struct binder_lru_page { */ struct binder_alloc { struct mutex mutex; - unsigned long vma_addr; + struct vm_area_struct *vma; struct mm_struct *mm; void __user *buffer; struct list_head buffers; diff --git a/drivers/android/binder_alloc_selftest.c b/drivers/android/binder_alloc_selftest.c index 43a881073a42832feb96c906c53e0ad34d2b2970..c2b323bc3b3a53043fa5783fb4f809de93b24644 100644 --- a/drivers/android/binder_alloc_selftest.c +++ b/drivers/android/binder_alloc_selftest.c @@ -287,7 +287,7 @@ void binder_selftest_alloc(struct binder_alloc *alloc) if (!binder_selftest_run) return; mutex_lock(&binder_selftest_lock); - if (!binder_selftest_run || !alloc->vma_addr) + if (!binder_selftest_run || !alloc->vma) goto done; pr_info("STARTED\n"); binder_selftest_alloc_offset(alloc, end_offset, 0); diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 8bf612bdd61a5a1d8d04d70fd8579249ede7c728..b4f246f0cac7f05a6942d41526af51d6f65eb766 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -5348,7 +5348,7 @@ struct ata_port *ata_port_alloc(struct ata_host *host) mutex_init(&ap->scsi_scan_mutex); INIT_DELAYED_WORK(&ap->hotplug_task, ata_scsi_hotplug); - INIT_WORK(&ap->scsi_rescan_task, ata_scsi_dev_rescan); + INIT_DELAYED_WORK(&ap->scsi_rescan_task, ata_scsi_dev_rescan); INIT_LIST_HEAD(&ap->eh_done_q); init_waitqueue_head(&ap->eh_wait_q); init_completion(&ap->park_req_pending); @@ -5954,6 +5954,7 @@ static void ata_port_detach(struct ata_port *ap) WARN_ON(!(ap->pflags & ATA_PFLAG_UNLOADED)); cancel_delayed_work_sync(&ap->hotplug_task); + cancel_delayed_work_sync(&ap->scsi_rescan_task); skip_eh: /* clean up zpodd on port removal */ diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c index a6c90181180271bd0c435a658b4dc71a57f3968c..6f8d141915939441dafd1a9852e2172da128cadf 100644 --- a/drivers/ata/libata-eh.c +++ b/drivers/ata/libata-eh.c @@ -2984,7 +2984,7 @@ static int ata_eh_revalidate_and_attach(struct ata_link *link, ehc->i.flags |= ATA_EHI_SETMODE; /* schedule the scsi_rescan_device() here */ - schedule_work(&(ap->scsi_rescan_task)); + schedule_delayed_work(&ap->scsi_rescan_task, 0); } else if (dev->class == ATA_DEV_UNKNOWN && ehc->tries[dev->devno] && ata_class_enabled(ehc->classes[dev->devno])) { diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index 7bb12deab70c465bb2a701548bdc044cd8bcb11f..551077cea4e4e023e20e913e37ee7b939488083a 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -2694,18 +2694,36 @@ static unsigned int atapi_xlat(struct ata_queued_cmd *qc) return 0; } -static struct ata_device *ata_find_dev(struct ata_port *ap, int devno) +static struct ata_device *ata_find_dev(struct ata_port *ap, unsigned int devno) { - if (!sata_pmp_attached(ap)) { - if (likely(devno >= 0 && - devno < ata_link_max_devices(&ap->link))) + /* + * For the non-PMP case, ata_link_max_devices() returns 1 (SATA case), + * or 2 (IDE master + slave case). However, the former case includes + * libsas hosted devices which are numbered per scsi host, leading + * to devno potentially being larger than 0 but with each struct + * ata_device having its own struct ata_port and struct ata_link. + * To accommodate these, ignore devno and always use device number 0. + */ + if (likely(!sata_pmp_attached(ap))) { + int link_max_devices = ata_link_max_devices(&ap->link); + + if (link_max_devices == 1) + return &ap->link.device[0]; + + if (devno < link_max_devices) return &ap->link.device[devno]; - } else { - if (likely(devno >= 0 && - devno < ap->nr_pmp_links)) - return &ap->pmp_link[devno].device[0]; + + return NULL; } + /* + * For PMP-attached devices, the device number corresponds to C + * (channel) of SCSI [H:C:I:L], indicating the port pmp link + * for the device. + */ + if (devno < ap->nr_pmp_links) + return &ap->pmp_link[devno].device[0]; + return NULL; } @@ -4579,10 +4597,11 @@ int ata_scsi_user_scan(struct Scsi_Host *shost, unsigned int channel, void ata_scsi_dev_rescan(struct work_struct *work) { struct ata_port *ap = - container_of(work, struct ata_port, scsi_rescan_task); + container_of(work, struct ata_port, scsi_rescan_task.work); struct ata_link *link; struct ata_device *dev; unsigned long flags; + bool delay_rescan = false; mutex_lock(&ap->scsi_scan_mutex); spin_lock_irqsave(ap->lock, flags); @@ -4596,6 +4615,21 @@ void ata_scsi_dev_rescan(struct work_struct *work) if (scsi_device_get(sdev)) continue; + /* + * If the rescan work was scheduled because of a resume + * event, the port is already fully resumed, but the + * SCSI device may not yet be fully resumed. In such + * case, executing scsi_rescan_device() may cause a + * deadlock with the PM code on device_lock(). Prevent + * this by giving up and retrying rescan after a short + * delay. + */ + delay_rescan = sdev->sdev_gendev.power.is_suspended; + if (delay_rescan) { + scsi_device_put(sdev); + break; + } + spin_unlock_irqrestore(ap->lock, flags); scsi_rescan_device(&(sdev->sdev_gendev)); scsi_device_put(sdev); @@ -4605,4 +4639,8 @@ void ata_scsi_dev_rescan(struct work_struct *work) spin_unlock_irqrestore(ap->lock, flags); mutex_unlock(&ap->scsi_scan_mutex); + + if (delay_rescan) + schedule_delayed_work(&ap->scsi_rescan_task, + msecs_to_jiffies(5)); } diff --git a/drivers/base/cacheinfo.c b/drivers/base/cacheinfo.c index bba3482ddeb82e72b8578f4db36bdb5d78511a6f..cbae8be1fe52005553e3d9dded77a21595d2cafc 100644 --- a/drivers/base/cacheinfo.c +++ b/drivers/base/cacheinfo.c @@ -388,6 +388,16 @@ static int cache_shared_cpu_map_setup(unsigned int cpu) continue;/* skip if itself or no cacheinfo */ for (sib_index = 0; sib_index < cache_leaves(i); sib_index++) { sib_leaf = per_cpu_cacheinfo_idx(i, sib_index); + + /* + * Comparing cache IDs only makes sense if the leaves + * belong to the same cache level of same type. Skip + * the check if level and type do not match. + */ + if (sib_leaf->level != this_leaf->level || + sib_leaf->type != this_leaf->type) + continue; + if (cache_leaves_are_shared(this_leaf, sib_leaf)) { cpumask_set_cpu(cpu, &sib_leaf->shared_cpu_map); cpumask_set_cpu(i, &this_leaf->shared_cpu_map); @@ -400,11 +410,14 @@ static int cache_shared_cpu_map_setup(unsigned int cpu) coherency_max_size = this_leaf->coherency_line_size; } + /* shared_cpu_map is now populated for the cpu */ + this_cpu_ci->cpu_map_populated = true; return 0; } static void cache_shared_cpu_map_remove(unsigned int cpu) { + struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu); struct cacheinfo *this_leaf, *sib_leaf; unsigned int sibling, index, sib_index; @@ -419,6 +432,16 @@ static void cache_shared_cpu_map_remove(unsigned int cpu) for (sib_index = 0; sib_index < cache_leaves(sibling); sib_index++) { sib_leaf = per_cpu_cacheinfo_idx(sibling, sib_index); + + /* + * Comparing cache IDs only makes sense if the leaves + * belong to the same cache level of same type. Skip + * the check if level and type do not match. + */ + if (sib_leaf->level != this_leaf->level || + sib_leaf->type != this_leaf->type) + continue; + if (cache_leaves_are_shared(this_leaf, sib_leaf)) { cpumask_clear_cpu(cpu, &sib_leaf->shared_cpu_map); cpumask_clear_cpu(sibling, &this_leaf->shared_cpu_map); @@ -427,6 +450,9 @@ static void cache_shared_cpu_map_remove(unsigned int cpu) } } } + + /* cpu is no longer populated in the shared map */ + this_cpu_ci->cpu_map_populated = false; } static void free_cache_attributes(unsigned int cpu) diff --git a/drivers/base/class.c b/drivers/base/class.c index ac1808d1a2e8f0012beb816b97a7101b048b688c..05d9df90f621be89cdbd7d1ecee888c18a6b688b 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -320,6 +320,7 @@ void class_dev_iter_init(struct class_dev_iter *iter, const struct class *class, start_knode = &start->p->knode_class; klist_iter_init_node(&sp->klist_devices, &iter->ki, start_knode); iter->type = type; + iter->sp = sp; } EXPORT_SYMBOL_GPL(class_dev_iter_init); @@ -361,6 +362,7 @@ EXPORT_SYMBOL_GPL(class_dev_iter_next); void class_dev_iter_exit(struct class_dev_iter *iter) { klist_iter_exit(&iter->ki); + subsys_put(iter->sp); } EXPORT_SYMBOL_GPL(class_dev_iter_exit); diff --git a/drivers/base/firmware_loader/main.c b/drivers/base/firmware_loader/main.c index 9d79d5ad9102417386de44356c0d6d2af45f8e47..b58c42f1b1ce658157c2c832bca8ea929286b139 100644 --- a/drivers/base/firmware_loader/main.c +++ b/drivers/base/firmware_loader/main.c @@ -812,7 +812,7 @@ static void fw_log_firmware_info(const struct firmware *fw, const char *name, st char *outbuf; alg = crypto_alloc_shash("sha256", 0, 0); - if (!alg) + if (IS_ERR(alg)) return; sha256buf = kmalloc(SHA256_DIGEST_SIZE, GFP_KERNEL); diff --git a/drivers/base/regmap/Kconfig b/drivers/base/regmap/Kconfig index 33a8366e22a584a5681b9434cfac240e238d750c..0db2021f7477f2abd40ee69ff23dd759dbc6a1a7 100644 --- a/drivers/base/regmap/Kconfig +++ b/drivers/base/regmap/Kconfig @@ -4,16 +4,23 @@ # subsystems should select the appropriate symbols. config REGMAP + bool "Register Map support" if KUNIT_ALL_TESTS default y if (REGMAP_I2C || REGMAP_SPI || REGMAP_SPMI || REGMAP_W1 || REGMAP_AC97 || REGMAP_MMIO || REGMAP_IRQ || REGMAP_SOUNDWIRE || REGMAP_SOUNDWIRE_MBQ || REGMAP_SCCB || REGMAP_I3C || REGMAP_SPI_AVMM || REGMAP_MDIO || REGMAP_FSI) select IRQ_DOMAIN if REGMAP_IRQ select MDIO_BUS if REGMAP_MDIO - bool + help + Enable support for the Register Map (regmap) access API. + + Usually, this option is automatically selected when needed. + However, you may want to enable it manually for running the regmap + KUnit tests. + + If unsure, say N. config REGMAP_KUNIT tristate "KUnit tests for regmap" - depends on KUNIT + depends on KUNIT && REGMAP default KUNIT_ALL_TESTS - select REGMAP select REGMAP_RAM config REGMAP_AC97 diff --git a/drivers/base/regmap/regcache-maple.c b/drivers/base/regmap/regcache-maple.c index 9b1b559107ef4d8016536ab8ba589be4b02aa725..c2e3a0f6c2183bd2c3cc59561d162317dfea8d50 100644 --- a/drivers/base/regmap/regcache-maple.c +++ b/drivers/base/regmap/regcache-maple.c @@ -203,15 +203,18 @@ static int regcache_maple_sync(struct regmap *map, unsigned int min, mas_for_each(&mas, entry, max) { for (r = max(mas.index, lmin); r <= min(mas.last, lmax); r++) { + mas_pause(&mas); + rcu_read_unlock(); ret = regcache_sync_val(map, r, entry[r - mas.index]); if (ret != 0) goto out; + rcu_read_lock(); } } -out: rcu_read_unlock(); +out: map->cache_bypass = false; return ret; diff --git a/drivers/base/regmap/regcache.c b/drivers/base/regmap/regcache.c index 029564695dbbb1ba03964837efb4375eda41ea5b..97c681fcf9f6ee4cd4e5bb0c709fabb9faf07b50 100644 --- a/drivers/base/regmap/regcache.c +++ b/drivers/base/regmap/regcache.c @@ -284,6 +284,9 @@ static bool regcache_reg_needs_sync(struct regmap *map, unsigned int reg, { int ret; + if (!regmap_writeable(map, reg)) + return false; + /* If we don't know the chip just got reset, then sync everything. */ if (!map->no_sync_defaults) return true; diff --git a/drivers/base/regmap/regmap-sdw.c b/drivers/base/regmap/regmap-sdw.c index 09899ae99fc199b0585c7f417c1f76043dced7d0..159c0b740b001c18f1c9510d956c97316f9689b6 100644 --- a/drivers/base/regmap/regmap-sdw.c +++ b/drivers/base/regmap/regmap-sdw.c @@ -59,6 +59,10 @@ static int regmap_sdw_config_check(const struct regmap_config *config) if (config->pad_bits != 0) return -ENOTSUPP; + /* Only bulk writes are supported not multi-register writes */ + if (config->can_multi_write) + return -ENOTSUPP; + return 0; } diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index db7851f0e3b8cd7542b93a09d9a9224557868fd7..fa2d3fba6ac9d1d9df6f3a8ef63506c16f1e0cb0 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -2082,6 +2082,8 @@ int _regmap_raw_write(struct regmap *map, unsigned int reg, size_t val_count = val_len / val_bytes; size_t chunk_count, chunk_bytes; size_t chunk_regs = val_count; + size_t max_data = map->max_raw_write - map->format.reg_bytes - + map->format.pad_bytes; int ret, i; if (!val_count) @@ -2089,8 +2091,8 @@ int _regmap_raw_write(struct regmap *map, unsigned int reg, if (map->use_single_write) chunk_regs = 1; - else if (map->max_raw_write && val_len > map->max_raw_write) - chunk_regs = map->max_raw_write / val_bytes; + else if (map->max_raw_write && val_len > max_data) + chunk_regs = max_data / val_bytes; chunk_count = val_count / chunk_regs; chunk_bytes = chunk_regs * val_bytes; diff --git a/drivers/block/null_blk/main.c b/drivers/block/null_blk/main.c index b3fedafe301e837206d6a0c2a22f40231dc3aeb5..864013019d6b6508d8750c7f19c10a11b710e9ca 100644 --- a/drivers/block/null_blk/main.c +++ b/drivers/block/null_blk/main.c @@ -2244,6 +2244,7 @@ static void null_destroy_dev(struct nullb *nullb) struct nullb_device *dev = nullb->dev; null_del_dev(nullb); + null_free_device_storage(dev, false); null_free_dev(dev); } diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c index 84ad3b17956f67ddd55508e6eb7d9b23c67ecb1e..632751ddb2870846926391f77986ca1e44a0a5bc 100644 --- a/drivers/block/rbd.c +++ b/drivers/block/rbd.c @@ -1334,14 +1334,30 @@ static bool rbd_obj_is_tail(struct rbd_obj_request *obj_req) /* * Must be called after rbd_obj_calc_img_extents(). */ -static bool rbd_obj_copyup_enabled(struct rbd_obj_request *obj_req) +static void rbd_obj_set_copyup_enabled(struct rbd_obj_request *obj_req) { - if (!obj_req->num_img_extents || - (rbd_obj_is_entire(obj_req) && - !obj_req->img_request->snapc->num_snaps)) - return false; + rbd_assert(obj_req->img_request->snapc); - return true; + if (obj_req->img_request->op_type == OBJ_OP_DISCARD) { + dout("%s %p objno %llu discard\n", __func__, obj_req, + obj_req->ex.oe_objno); + return; + } + + if (!obj_req->num_img_extents) { + dout("%s %p objno %llu not overlapping\n", __func__, obj_req, + obj_req->ex.oe_objno); + return; + } + + if (rbd_obj_is_entire(obj_req) && + !obj_req->img_request->snapc->num_snaps) { + dout("%s %p objno %llu entire\n", __func__, obj_req, + obj_req->ex.oe_objno); + return; + } + + obj_req->flags |= RBD_OBJ_FLAG_COPYUP_ENABLED; } static u64 rbd_obj_img_extents_bytes(struct rbd_obj_request *obj_req) @@ -1442,6 +1458,7 @@ __rbd_obj_add_osd_request(struct rbd_obj_request *obj_req, static struct ceph_osd_request * rbd_obj_add_osd_request(struct rbd_obj_request *obj_req, int num_ops) { + rbd_assert(obj_req->img_request->snapc); return __rbd_obj_add_osd_request(obj_req, obj_req->img_request->snapc, num_ops); } @@ -1578,15 +1595,18 @@ static void rbd_img_request_init(struct rbd_img_request *img_request, mutex_init(&img_request->state_mutex); } +/* + * Only snap_id is captured here, for reads. For writes, snapshot + * context is captured in rbd_img_object_requests() after exclusive + * lock is ensured to be held. + */ static void rbd_img_capture_header(struct rbd_img_request *img_req) { struct rbd_device *rbd_dev = img_req->rbd_dev; lockdep_assert_held(&rbd_dev->header_rwsem); - if (rbd_img_is_write(img_req)) - img_req->snapc = ceph_get_snap_context(rbd_dev->header.snapc); - else + if (!rbd_img_is_write(img_req)) img_req->snap_id = rbd_dev->spec->snap_id; if (rbd_dev_parent_get(rbd_dev)) @@ -2233,9 +2253,6 @@ static int rbd_obj_init_write(struct rbd_obj_request *obj_req) if (ret) return ret; - if (rbd_obj_copyup_enabled(obj_req)) - obj_req->flags |= RBD_OBJ_FLAG_COPYUP_ENABLED; - obj_req->write_state = RBD_OBJ_WRITE_START; return 0; } @@ -2341,8 +2358,6 @@ static int rbd_obj_init_zeroout(struct rbd_obj_request *obj_req) if (ret) return ret; - if (rbd_obj_copyup_enabled(obj_req)) - obj_req->flags |= RBD_OBJ_FLAG_COPYUP_ENABLED; if (!obj_req->num_img_extents) { obj_req->flags |= RBD_OBJ_FLAG_NOOP_FOR_NONEXISTENT; if (rbd_obj_is_entire(obj_req)) @@ -3286,6 +3301,7 @@ static bool rbd_obj_advance_write(struct rbd_obj_request *obj_req, int *result) case RBD_OBJ_WRITE_START: rbd_assert(!*result); + rbd_obj_set_copyup_enabled(obj_req); if (rbd_obj_write_is_noop(obj_req)) return true; @@ -3472,9 +3488,19 @@ static int rbd_img_exclusive_lock(struct rbd_img_request *img_req) static void rbd_img_object_requests(struct rbd_img_request *img_req) { + struct rbd_device *rbd_dev = img_req->rbd_dev; struct rbd_obj_request *obj_req; rbd_assert(!img_req->pending.result && !img_req->pending.num_pending); + rbd_assert(!need_exclusive_lock(img_req) || + __rbd_is_lock_owner(rbd_dev)); + + if (rbd_img_is_write(img_req)) { + rbd_assert(!img_req->snapc); + down_read(&rbd_dev->header_rwsem); + img_req->snapc = ceph_get_snap_context(rbd_dev->header.snapc); + up_read(&rbd_dev->header_rwsem); + } for_each_obj_request(img_req, obj_req) { int result = 0; @@ -3492,7 +3518,6 @@ static void rbd_img_object_requests(struct rbd_img_request *img_req) static bool rbd_img_advance(struct rbd_img_request *img_req, int *result) { - struct rbd_device *rbd_dev = img_req->rbd_dev; int ret; again: @@ -3513,9 +3538,6 @@ static bool rbd_img_advance(struct rbd_img_request *img_req, int *result) if (*result) return true; - rbd_assert(!need_exclusive_lock(img_req) || - __rbd_is_lock_owner(rbd_dev)); - rbd_img_object_requests(img_req); if (!img_req->pending.num_pending) { *result = img_req->pending.result; @@ -3977,6 +3999,10 @@ static int rbd_post_acquire_action(struct rbd_device *rbd_dev) { int ret; + ret = rbd_dev_refresh(rbd_dev); + if (ret) + return ret; + if (rbd_dev->header.features & RBD_FEATURE_OBJECT_MAP) { ret = rbd_object_map_open(rbd_dev); if (ret) diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c index c7ed5d69e9eee2545c0f4458ce6f4a1847060cde..33d3298a0da16f1f70ecc87495f77746f2e6a4af 100644 --- a/drivers/block/ublk_drv.c +++ b/drivers/block/ublk_drv.c @@ -1120,6 +1120,11 @@ static inline bool ublk_queue_ready(struct ublk_queue *ubq) return ubq->nr_io_ready == ubq->q_depth; } +static void ublk_cmd_cancel_cb(struct io_uring_cmd *cmd, unsigned issue_flags) +{ + io_uring_cmd_done(cmd, UBLK_IO_RES_ABORT, 0, issue_flags); +} + static void ublk_cancel_queue(struct ublk_queue *ubq) { int i; @@ -1131,8 +1136,8 @@ static void ublk_cancel_queue(struct ublk_queue *ubq) struct ublk_io *io = &ubq->ios[i]; if (io->flags & UBLK_IO_FLAG_ACTIVE) - io_uring_cmd_done(io->cmd, UBLK_IO_RES_ABORT, 0, - IO_URING_F_UNLOCKED); + io_uring_cmd_complete_in_task(io->cmd, + ublk_cmd_cancel_cb); } /* all io commands are canceled */ diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index 23ed258b57f0e585e122a60d6d4dba7cc2972f02..c1890c8a9f6e7640d874237a988437ff4aa264d7 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -780,7 +780,8 @@ static int blkif_queue_rw_req(struct request *req, struct blkfront_ring_info *ri ring_req->u.rw.handle = info->handle; ring_req->operation = rq_data_dir(req) ? BLKIF_OP_WRITE : BLKIF_OP_READ; - if (req_op(req) == REQ_OP_FLUSH || req->cmd_flags & REQ_FUA) { + if (req_op(req) == REQ_OP_FLUSH || + (req_op(req) == REQ_OP_WRITE && (req->cmd_flags & REQ_FUA))) { /* * Ideally we can do an unordered flush-to-disk. * In case the backend onlysupports barriers, use that. diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c index 3a34d7c1475bceeeca3a3c5d10a10243d4d9e0f4..52ef44688d38a3685c7974feaa60af7ace37dac6 100644 --- a/drivers/bluetooth/btnxpuart.c +++ b/drivers/bluetooth/btnxpuart.c @@ -1319,17 +1319,17 @@ static void nxp_serdev_remove(struct serdev_device *serdev) hci_free_dev(hdev); } -static struct btnxpuart_data w8987_data = { +static struct btnxpuart_data w8987_data __maybe_unused = { .helper_fw_name = NULL, .fw_name = FIRMWARE_W8987, }; -static struct btnxpuart_data w8997_data = { +static struct btnxpuart_data w8997_data __maybe_unused = { .helper_fw_name = FIRMWARE_HELPER, .fw_name = FIRMWARE_W8997, }; -static const struct of_device_id nxpuart_of_match_table[] = { +static const struct of_device_id nxpuart_of_match_table[] __maybe_unused = { { .compatible = "nxp,88w8987-bt", .data = &w8987_data }, { .compatible = "nxp,88w8997-bt", .data = &w8997_data }, { } diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c index 1b064504b3887522f50ecc19b5d578863399baed..e30c979535b1dae8df06b4a4a158730bc83a9ba8 100644 --- a/drivers/bluetooth/hci_qca.c +++ b/drivers/bluetooth/hci_qca.c @@ -78,7 +78,8 @@ enum qca_flags { QCA_HW_ERROR_EVENT, QCA_SSR_TRIGGERED, QCA_BT_OFF, - QCA_ROM_FW + QCA_ROM_FW, + QCA_DEBUGFS_CREATED, }; enum qca_capabilities { @@ -635,6 +636,9 @@ static void qca_debugfs_init(struct hci_dev *hdev) if (!hdev->debugfs) return; + if (test_and_set_bit(QCA_DEBUGFS_CREATED, &qca->flags)) + return; + ibs_dir = debugfs_create_dir("ibs", hdev->debugfs); /* read only */ diff --git a/drivers/char/agp/parisc-agp.c b/drivers/char/agp/parisc-agp.c index d68d05d5d38388523b0fecfdd45cb6e32488c784..514f9f287a781dd910de9da4034a6ea74529bc11 100644 --- a/drivers/char/agp/parisc-agp.c +++ b/drivers/char/agp/parisc-agp.c @@ -90,6 +90,9 @@ parisc_agp_tlbflush(struct agp_memory *mem) { struct _parisc_agp_info *info = &parisc_agp_info; + /* force fdc ops to be visible to IOMMU */ + asm_io_sync(); + writeq(info->gart_base | ilog2(info->gart_size), info->ioc_regs+IOC_PCOM); readq(info->ioc_regs+IOC_PCOM); /* flush */ } @@ -158,6 +161,7 @@ parisc_agp_insert_memory(struct agp_memory *mem, off_t pg_start, int type) info->gatt[j] = parisc_agp_mask_memory(agp_bridge, paddr, type); + asm_io_fdc(&info->gatt[j]); } } @@ -191,7 +195,16 @@ static unsigned long parisc_agp_mask_memory(struct agp_bridge_data *bridge, dma_addr_t addr, int type) { - return SBA_PDIR_VALID_BIT | addr; + unsigned ci; /* coherent index */ + dma_addr_t pa; + + pa = addr & IOVP_MASK; + asm("lci 0(%1), %0" : "=r" (ci) : "r" (phys_to_virt(pa))); + + pa |= (ci >> PAGE_SHIFT) & 0xff;/* move CI (8 bits) into lowest byte */ + pa |= SBA_PDIR_VALID_BIT; /* set "valid" bit */ + + return cpu_to_le64(pa); } static void diff --git a/drivers/char/tpm/tpm-chip.c b/drivers/char/tpm/tpm-chip.c index c10a4aa9737366f949d3fb0602971ef85ae3d5a6..cd48033b804a3871529ce9d31acb34d7ec64819e 100644 --- a/drivers/char/tpm/tpm-chip.c +++ b/drivers/char/tpm/tpm-chip.c @@ -571,6 +571,10 @@ static int tpm_hwrng_read(struct hwrng *rng, void *data, size_t max, bool wait) { struct tpm_chip *chip = container_of(rng, struct tpm_chip, hwrng); + /* Give back zero bytes, as TPM chip has not yet fully resumed: */ + if (chip->flags & TPM_CHIP_FLAG_SUSPENDED) + return 0; + return tpm_get_random(chip, data, max); } diff --git a/drivers/char/tpm/tpm-interface.c b/drivers/char/tpm/tpm-interface.c index 4463d00182909b37773101c0b826b95e4e92c8bf..586ca10b0d72e5a0ed6fb84171b8ba1754def0d8 100644 --- a/drivers/char/tpm/tpm-interface.c +++ b/drivers/char/tpm/tpm-interface.c @@ -412,6 +412,8 @@ int tpm_pm_suspend(struct device *dev) } suspended: + chip->flags |= TPM_CHIP_FLAG_SUSPENDED; + if (rc) dev_err(dev, "Ignoring error %d while suspending\n", rc); return 0; @@ -429,6 +431,14 @@ int tpm_pm_resume(struct device *dev) if (chip == NULL) return -ENODEV; + chip->flags &= ~TPM_CHIP_FLAG_SUSPENDED; + + /* + * Guarantee that SUSPENDED is written last, so that hwrng does not + * activate before the chip has been fully resumed. + */ + wmb(); + return 0; } EXPORT_SYMBOL_GPL(tpm_pm_resume); diff --git a/drivers/char/tpm/tpm_tis.c b/drivers/char/tpm/tpm_tis.c index 7af389806643379fa76507755550fdae92871de9..7db3593941eaa0bce84ac5b5c217b814e8ae2772 100644 --- a/drivers/char/tpm/tpm_tis.c +++ b/drivers/char/tpm/tpm_tis.c @@ -122,6 +122,29 @@ static const struct dmi_system_id tpm_tis_dmi_table[] = { DMI_MATCH(DMI_PRODUCT_VERSION, "ThinkPad T490s"), }, }, + { + .callback = tpm_tis_disable_irq, + .ident = "ThinkStation P360 Tiny", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), + DMI_MATCH(DMI_PRODUCT_VERSION, "ThinkStation P360 Tiny"), + }, + }, + { + .callback = tpm_tis_disable_irq, + .ident = "ThinkPad L490", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), + DMI_MATCH(DMI_PRODUCT_VERSION, "ThinkPad L490"), + }, + }, + { + .callback = tpm_tis_disable_irq, + .ident = "UPX-TGL", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "AAEON"), + }, + }, {} }; diff --git a/drivers/char/tpm/tpm_tis_core.c b/drivers/char/tpm/tpm_tis_core.c index 02945d53fcefafae4e0c72ef3b2a4cf6a8cb85e5..558144fa707ae7ad0937d1ae2a16ac3afba7e310 100644 --- a/drivers/char/tpm/tpm_tis_core.c +++ b/drivers/char/tpm/tpm_tis_core.c @@ -1209,25 +1209,20 @@ static void tpm_tis_reenable_interrupts(struct tpm_chip *chip) u32 intmask; int rc; - if (chip->ops->clk_enable != NULL) - chip->ops->clk_enable(chip, true); - - /* reenable interrupts that device may have lost or - * BIOS/firmware may have disabled + /* + * Re-enable interrupts that device may have lost or BIOS/firmware may + * have disabled. */ rc = tpm_tis_write8(priv, TPM_INT_VECTOR(priv->locality), priv->irq); - if (rc < 0) - goto out; + if (rc < 0) { + dev_err(&chip->dev, "Setting IRQ failed.\n"); + return; + } intmask = priv->int_mask | TPM_GLOBAL_INT_ENABLE; - - tpm_tis_write32(priv, TPM_INT_ENABLE(priv->locality), intmask); - -out: - if (chip->ops->clk_enable != NULL) - chip->ops->clk_enable(chip, false); - - return; + rc = tpm_tis_write32(priv, TPM_INT_ENABLE(priv->locality), intmask); + if (rc < 0) + dev_err(&chip->dev, "Enabling interrupts failed.\n"); } int tpm_tis_resume(struct device *dev) @@ -1235,27 +1230,27 @@ int tpm_tis_resume(struct device *dev) struct tpm_chip *chip = dev_get_drvdata(dev); int ret; - ret = tpm_tis_request_locality(chip, 0); - if (ret < 0) + ret = tpm_chip_start(chip); + if (ret) return ret; if (chip->flags & TPM_CHIP_FLAG_IRQ) tpm_tis_reenable_interrupts(chip); - ret = tpm_pm_resume(dev); - if (ret) - goto out; - /* * TPM 1.2 requires self-test on resume. This function actually returns * an error code but for unknown reason it isn't handled. */ if (!(chip->flags & TPM_CHIP_FLAG_TPM2)) tpm1_do_selftest(chip); -out: - tpm_tis_relinquish_locality(chip, 0); - return ret; + tpm_chip_stop(chip); + + ret = tpm_pm_resume(dev); + if (ret) + return ret; + + return 0; } EXPORT_SYMBOL_GPL(tpm_tis_resume); #endif diff --git a/drivers/char/tpm/tpm_tis_core.h b/drivers/char/tpm/tpm_tis_core.h index e978f457fd4d4ce083a7a9f9689df5f2d7a8ed9d..610bfadb6acf1c1731575842eca18f1d8c6ea395 100644 --- a/drivers/char/tpm/tpm_tis_core.h +++ b/drivers/char/tpm/tpm_tis_core.h @@ -84,10 +84,10 @@ enum tis_defaults { #define ILB_REMAP_SIZE 0x100 enum tpm_tis_flags { - TPM_TIS_ITPM_WORKAROUND = BIT(0), - TPM_TIS_INVALID_STATUS = BIT(1), - TPM_TIS_DEFAULT_CANCELLATION = BIT(2), - TPM_TIS_IRQ_TESTED = BIT(3), + TPM_TIS_ITPM_WORKAROUND = 0, + TPM_TIS_INVALID_STATUS = 1, + TPM_TIS_DEFAULT_CANCELLATION = 2, + TPM_TIS_IRQ_TESTED = 3, }; struct tpm_tis_data { diff --git a/drivers/clk/clk-composite.c b/drivers/clk/clk-composite.c index edfa94641bbfed8f0a9d4e575f73c82b21598c2b..66759fe28fad67f98b6c723ce70849490a022e26 100644 --- a/drivers/clk/clk-composite.c +++ b/drivers/clk/clk-composite.c @@ -119,7 +119,10 @@ static int clk_composite_determine_rate(struct clk_hw *hw, if (ret) continue; - rate_diff = abs(req->rate - tmp_req.rate); + if (req->rate >= tmp_req.rate) + rate_diff = req->rate - tmp_req.rate; + else + rate_diff = tmp_req.rate - req->rate; if (!rate_diff || !req->best_parent_hw || best_rate_diff > rate_diff) { diff --git a/drivers/clk/clk-loongson2.c b/drivers/clk/clk-loongson2.c index 70ae1dd2e474cc401a270a8df3b26b4b63633f25..bacdcbb287ac618b57c236b1103ebbe2c31e7107 100644 --- a/drivers/clk/clk-loongson2.c +++ b/drivers/clk/clk-loongson2.c @@ -40,7 +40,7 @@ static struct clk_hw *loongson2_clk_register(struct device *dev, { int ret; struct clk_hw *hw; - struct clk_init_data init; + struct clk_init_data init = { }; hw = devm_kzalloc(dev, sizeof(*hw), GFP_KERNEL); if (!hw) diff --git a/drivers/clk/mediatek/clk-mt8365.c b/drivers/clk/mediatek/clk-mt8365.c index 6b4e193f648d81e3181abaef3fc80a7929ce8c14..c87a6c4a79678c1e29c2bd8e67eb984313f484f0 100644 --- a/drivers/clk/mediatek/clk-mt8365.c +++ b/drivers/clk/mediatek/clk-mt8365.c @@ -23,6 +23,7 @@ static DEFINE_SPINLOCK(mt8365_clk_lock); static const struct mtk_fixed_clk top_fixed_clks[] = { + FIXED_CLK(CLK_TOP_CLK_NULL, "clk_null", NULL, 0), FIXED_CLK(CLK_TOP_I2S0_BCK, "i2s0_bck", NULL, 26000000), FIXED_CLK(CLK_TOP_DSI0_LNTC_DSICK, "dsi0_lntc_dsick", "clk26m", 75000000), @@ -559,6 +560,14 @@ static const struct mtk_clk_divider top_adj_divs[] = { 0x324, 16, 8, CLK_DIVIDER_ROUND_CLOSEST), DIV_ADJ_F(CLK_TOP_APLL12_CK_DIV3, "apll12_ck_div3", "apll_i2s3_sel", 0x324, 24, 8, CLK_DIVIDER_ROUND_CLOSEST), + DIV_ADJ_F(CLK_TOP_APLL12_CK_DIV4, "apll12_ck_div4", "apll_tdmout_sel", + 0x328, 0, 8, CLK_DIVIDER_ROUND_CLOSEST), + DIV_ADJ_F(CLK_TOP_APLL12_CK_DIV4B, "apll12_ck_div4b", "apll_tdmout_sel", + 0x328, 8, 8, CLK_DIVIDER_ROUND_CLOSEST), + DIV_ADJ_F(CLK_TOP_APLL12_CK_DIV5, "apll12_ck_div5", "apll_tdmin_sel", + 0x328, 16, 8, CLK_DIVIDER_ROUND_CLOSEST), + DIV_ADJ_F(CLK_TOP_APLL12_CK_DIV5B, "apll12_ck_div5b", "apll_tdmin_sel", + 0x328, 24, 8, CLK_DIVIDER_ROUND_CLOSEST), DIV_ADJ_F(CLK_TOP_APLL12_CK_DIV6, "apll12_ck_div6", "apll_spdif_sel", 0x32c, 0, 8, CLK_DIVIDER_ROUND_CLOSEST), }; @@ -583,15 +592,15 @@ static const struct mtk_gate_regs top2_cg_regs = { #define GATE_TOP0(_id, _name, _parent, _shift) \ GATE_MTK(_id, _name, _parent, &top0_cg_regs, \ - _shift, &mtk_clk_gate_ops_no_setclr_inv) + _shift, &mtk_clk_gate_ops_no_setclr) #define GATE_TOP1(_id, _name, _parent, _shift) \ GATE_MTK(_id, _name, _parent, &top1_cg_regs, \ - _shift, &mtk_clk_gate_ops_no_setclr) + _shift, &mtk_clk_gate_ops_no_setclr_inv) #define GATE_TOP2(_id, _name, _parent, _shift) \ GATE_MTK(_id, _name, _parent, &top2_cg_regs, \ - _shift, &mtk_clk_gate_ops_no_setclr) + _shift, &mtk_clk_gate_ops_no_setclr_inv) static const struct mtk_gate top_clk_gates[] = { GATE_TOP0(CLK_TOP_CONN_32K, "conn_32k", "clk32k", 10), @@ -696,6 +705,7 @@ static const struct mtk_gate ifr_clks[] = { GATE_IFR3(CLK_IFR_GCPU, "ifr_gcpu", "axi_sel", 8), GATE_IFR3(CLK_IFR_TRNG, "ifr_trng", "axi_sel", 9), GATE_IFR3(CLK_IFR_AUXADC, "ifr_auxadc", "clk26m", 10), + GATE_IFR3(CLK_IFR_CPUM, "ifr_cpum", "clk26m", 11), GATE_IFR3(CLK_IFR_AUXADC_MD, "ifr_auxadc_md", "clk26m", 14), GATE_IFR3(CLK_IFR_AP_DMA, "ifr_ap_dma", "axi_sel", 18), GATE_IFR3(CLK_IFR_DEBUGSYS, "ifr_debugsys", "axi_sel", 24), @@ -717,6 +727,8 @@ static const struct mtk_gate ifr_clks[] = { GATE_IFR5(CLK_IFR_PWRAP_TMR, "ifr_pwrap_tmr", "clk26m", 12), GATE_IFR5(CLK_IFR_PWRAP_SPI, "ifr_pwrap_spi", "clk26m", 13), GATE_IFR5(CLK_IFR_PWRAP_SYS, "ifr_pwrap_sys", "clk26m", 14), + GATE_MTK_FLAGS(CLK_IFR_MCU_PM_BK, "ifr_mcu_pm_bk", NULL, &ifr5_cg_regs, + 17, &mtk_clk_gate_ops_setclr, CLK_IGNORE_UNUSED), GATE_IFR5(CLK_IFR_IRRX_26M, "ifr_irrx_26m", "clk26m", 22), GATE_IFR5(CLK_IFR_IRRX_32K, "ifr_irrx_32k", "clk32k", 23), GATE_IFR5(CLK_IFR_I2C0_AXI, "ifr_i2c0_axi", "i2c_sel", 24), diff --git a/drivers/clk/pxa/clk-pxa3xx.c b/drivers/clk/pxa/clk-pxa3xx.c index 42958a5426625a85af8935484bb13bd0bb576b20..621e298f101a039bcc99f14f42288951b55dae52 100644 --- a/drivers/clk/pxa/clk-pxa3xx.c +++ b/drivers/clk/pxa/clk-pxa3xx.c @@ -164,7 +164,7 @@ void pxa3xx_clk_update_accr(u32 disable, u32 enable, u32 xclkcfg, u32 mask) accr &= ~disable; accr |= enable; - writel(accr, ACCR); + writel(accr, clk_regs + ACCR); if (xclkcfg) __asm__("mcr p14, 0, %0, c6, c0, 0\n" : : "r"(xclkcfg)); diff --git a/drivers/cpufreq/acpi-cpufreq.c b/drivers/cpufreq/acpi-cpufreq.c index 29904395e95f9ca3b6c24d845584a5da141a99b0..b2f05d27167e393c92b63bda89b17a9b48cad923 100644 --- a/drivers/cpufreq/acpi-cpufreq.c +++ b/drivers/cpufreq/acpi-cpufreq.c @@ -975,7 +975,7 @@ static int __init acpi_cpufreq_probe(struct platform_device *pdev) /* don't keep reloading if cpufreq_driver exists */ if (cpufreq_get_current_driver()) - return -EEXIST; + return -ENODEV; pr_debug("%s\n", __func__); diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c index 5a3d4aa0f45a6a18e1abb3cceb6c972c4b3f3e19..ddd346a239e0b55acfaef158cb12ea1a8612f948 100644 --- a/drivers/cpufreq/amd-pstate.c +++ b/drivers/cpufreq/amd-pstate.c @@ -444,9 +444,8 @@ static int amd_pstate_verify(struct cpufreq_policy_data *policy) return 0; } -static int amd_pstate_target(struct cpufreq_policy *policy, - unsigned int target_freq, - unsigned int relation) +static int amd_pstate_update_freq(struct cpufreq_policy *policy, + unsigned int target_freq, bool fast_switch) { struct cpufreq_freqs freqs; struct amd_cpudata *cpudata = policy->driver_data; @@ -465,26 +464,51 @@ static int amd_pstate_target(struct cpufreq_policy *policy, des_perf = DIV_ROUND_CLOSEST(target_freq * cap_perf, cpudata->max_freq); - cpufreq_freq_transition_begin(policy, &freqs); + WARN_ON(fast_switch && !policy->fast_switch_enabled); + /* + * If fast_switch is desired, then there aren't any registered + * transition notifiers. See comment for + * cpufreq_enable_fast_switch(). + */ + if (!fast_switch) + cpufreq_freq_transition_begin(policy, &freqs); + amd_pstate_update(cpudata, min_perf, des_perf, - max_perf, false, policy->governor->flags); - cpufreq_freq_transition_end(policy, &freqs, false); + max_perf, fast_switch, policy->governor->flags); + + if (!fast_switch) + cpufreq_freq_transition_end(policy, &freqs, false); return 0; } +static int amd_pstate_target(struct cpufreq_policy *policy, + unsigned int target_freq, + unsigned int relation) +{ + return amd_pstate_update_freq(policy, target_freq, false); +} + +static unsigned int amd_pstate_fast_switch(struct cpufreq_policy *policy, + unsigned int target_freq) +{ + return amd_pstate_update_freq(policy, target_freq, true); +} + static void amd_pstate_adjust_perf(unsigned int cpu, unsigned long _min_perf, unsigned long target_perf, unsigned long capacity) { unsigned long max_perf, min_perf, des_perf, - cap_perf, lowest_nonlinear_perf; + cap_perf, lowest_nonlinear_perf, max_freq; struct cpufreq_policy *policy = cpufreq_cpu_get(cpu); struct amd_cpudata *cpudata = policy->driver_data; + unsigned int target_freq; cap_perf = READ_ONCE(cpudata->highest_perf); lowest_nonlinear_perf = READ_ONCE(cpudata->lowest_nonlinear_perf); + max_freq = READ_ONCE(cpudata->max_freq); des_perf = cap_perf; if (target_perf < capacity) @@ -501,6 +525,10 @@ static void amd_pstate_adjust_perf(unsigned int cpu, if (max_perf < min_perf) max_perf = min_perf; + des_perf = clamp_t(unsigned long, des_perf, min_perf, max_perf); + target_freq = div_u64(des_perf * max_freq, max_perf); + policy->cur = target_freq; + amd_pstate_update(cpudata, min_perf, des_perf, max_perf, true, policy->governor->flags); cpufreq_cpu_put(policy); @@ -715,6 +743,7 @@ static int amd_pstate_cpu_exit(struct cpufreq_policy *policy) freq_qos_remove_request(&cpudata->req[1]); freq_qos_remove_request(&cpudata->req[0]); + policy->fast_switch_possible = false; kfree(cpudata); return 0; @@ -1079,7 +1108,6 @@ static int amd_pstate_epp_cpu_init(struct cpufreq_policy *policy) policy->policy = CPUFREQ_POLICY_POWERSAVE; if (boot_cpu_has(X86_FEATURE_CPPC)) { - policy->fast_switch_possible = true; ret = rdmsrl_on_cpu(cpudata->cpu, MSR_AMD_CPPC_REQ, &value); if (ret) return ret; @@ -1102,7 +1130,6 @@ static int amd_pstate_epp_cpu_init(struct cpufreq_policy *policy) static int amd_pstate_epp_cpu_exit(struct cpufreq_policy *policy) { pr_debug("CPU %d exiting\n", policy->cpu); - policy->fast_switch_possible = false; return 0; } @@ -1309,6 +1336,7 @@ static struct cpufreq_driver amd_pstate_driver = { .flags = CPUFREQ_CONST_LOOPS | CPUFREQ_NEED_UPDATE_LIMITS, .verify = amd_pstate_verify, .target = amd_pstate_target, + .fast_switch = amd_pstate_fast_switch, .init = amd_pstate_cpu_init, .exit = amd_pstate_cpu_exit, .suspend = amd_pstate_cpu_suspend, diff --git a/drivers/cpufreq/pcc-cpufreq.c b/drivers/cpufreq/pcc-cpufreq.c index 1d2cfea9858afc121d8ecf4048d5bf2bfc9afe4a..73efbcf5513b2cde60faa7d3223479c1c5efd1bd 100644 --- a/drivers/cpufreq/pcc-cpufreq.c +++ b/drivers/cpufreq/pcc-cpufreq.c @@ -583,7 +583,7 @@ static int __init pcc_cpufreq_probe(struct platform_device *pdev) /* Skip initialization if another cpufreq driver is there. */ if (cpufreq_get_current_driver()) - return -EEXIST; + return -ENODEV; if (acpi_disabled) return -ENODEV; diff --git a/drivers/cxl/core/mbox.c b/drivers/cxl/core/mbox.c index 23b9ff920d7e85c8f347fc11bb8fe738635c31a4..bea9cf31a12df9a3f258e47e98ca5af3582ea856 100644 --- a/drivers/cxl/core/mbox.c +++ b/drivers/cxl/core/mbox.c @@ -1028,7 +1028,7 @@ static int cxl_mem_get_partition_info(struct cxl_dev_state *cxlds) * cxl_dev_state_identify() - Send the IDENTIFY command to the device. * @cxlds: The device data for the operation * - * Return: 0 if identify was executed successfully. + * Return: 0 if identify was executed successfully or media not ready. * * This will dispatch the identify command to the device and on success populate * structures to be exported to sysfs. @@ -1041,6 +1041,9 @@ int cxl_dev_state_identify(struct cxl_dev_state *cxlds) u32 val; int rc; + if (!cxlds->media_ready) + return 0; + mbox_cmd = (struct cxl_mbox_cmd) { .opcode = CXL_MBOX_OP_IDENTIFY, .size_out = sizeof(id), @@ -1102,6 +1105,13 @@ int cxl_mem_create_range_info(struct cxl_dev_state *cxlds) struct device *dev = cxlds->dev; int rc; + if (!cxlds->media_ready) { + cxlds->dpa_res = DEFINE_RES_MEM(0, 0); + cxlds->ram_res = DEFINE_RES_MEM(0, 0); + cxlds->pmem_res = DEFINE_RES_MEM(0, 0); + return 0; + } + cxlds->dpa_res = (struct resource)DEFINE_RES_MEM(0, cxlds->total_bytes); diff --git a/drivers/cxl/core/pci.c b/drivers/cxl/core/pci.c index f332fe7af92ba3f86833ca9f3517723b855e53a2..67f4ab6daa34f0144380c27d41d9ca252c52ef04 100644 --- a/drivers/cxl/core/pci.c +++ b/drivers/cxl/core/pci.c @@ -101,23 +101,57 @@ int devm_cxl_port_enumerate_dports(struct cxl_port *port) } EXPORT_SYMBOL_NS_GPL(devm_cxl_port_enumerate_dports, CXL); -/* - * Wait up to @media_ready_timeout for the device to report memory - * active. - */ -int cxl_await_media_ready(struct cxl_dev_state *cxlds) +static int cxl_dvsec_mem_range_valid(struct cxl_dev_state *cxlds, int id) +{ + struct pci_dev *pdev = to_pci_dev(cxlds->dev); + int d = cxlds->cxl_dvsec; + bool valid = false; + int rc, i; + u32 temp; + + if (id > CXL_DVSEC_RANGE_MAX) + return -EINVAL; + + /* Check MEM INFO VALID bit first, give up after 1s */ + i = 1; + do { + rc = pci_read_config_dword(pdev, + d + CXL_DVSEC_RANGE_SIZE_LOW(id), + &temp); + if (rc) + return rc; + + valid = FIELD_GET(CXL_DVSEC_MEM_INFO_VALID, temp); + if (valid) + break; + msleep(1000); + } while (i--); + + if (!valid) { + dev_err(&pdev->dev, + "Timeout awaiting memory range %d valid after 1s.\n", + id); + return -ETIMEDOUT; + } + + return 0; +} + +static int cxl_dvsec_mem_range_active(struct cxl_dev_state *cxlds, int id) { struct pci_dev *pdev = to_pci_dev(cxlds->dev); int d = cxlds->cxl_dvsec; bool active = false; - u64 md_status; int rc, i; + u32 temp; - for (i = media_ready_timeout; i; i--) { - u32 temp; + if (id > CXL_DVSEC_RANGE_MAX) + return -EINVAL; + /* Check MEM ACTIVE bit, up to 60s timeout by default */ + for (i = media_ready_timeout; i; i--) { rc = pci_read_config_dword( - pdev, d + CXL_DVSEC_RANGE_SIZE_LOW(0), &temp); + pdev, d + CXL_DVSEC_RANGE_SIZE_LOW(id), &temp); if (rc) return rc; @@ -134,6 +168,39 @@ int cxl_await_media_ready(struct cxl_dev_state *cxlds) return -ETIMEDOUT; } + return 0; +} + +/* + * Wait up to @media_ready_timeout for the device to report memory + * active. + */ +int cxl_await_media_ready(struct cxl_dev_state *cxlds) +{ + struct pci_dev *pdev = to_pci_dev(cxlds->dev); + int d = cxlds->cxl_dvsec; + int rc, i, hdm_count; + u64 md_status; + u16 cap; + + rc = pci_read_config_word(pdev, + d + CXL_DVSEC_CAP_OFFSET, &cap); + if (rc) + return rc; + + hdm_count = FIELD_GET(CXL_DVSEC_HDM_COUNT_MASK, cap); + for (i = 0; i < hdm_count; i++) { + rc = cxl_dvsec_mem_range_valid(cxlds, i); + if (rc) + return rc; + } + + for (i = 0; i < hdm_count; i++) { + rc = cxl_dvsec_mem_range_active(cxlds, i); + if (rc) + return rc; + } + md_status = readq(cxlds->regs.memdev + CXLMDEV_STATUS_OFFSET); if (!CXLMDEV_READY(md_status)) return -EIO; @@ -241,17 +308,36 @@ static void disable_hdm(void *_cxlhdm) hdm + CXL_HDM_DECODER_CTRL_OFFSET); } -static int devm_cxl_enable_hdm(struct device *host, struct cxl_hdm *cxlhdm) +int devm_cxl_enable_hdm(struct cxl_port *port, struct cxl_hdm *cxlhdm) { - void __iomem *hdm = cxlhdm->regs.hdm_decoder; + void __iomem *hdm; u32 global_ctrl; + /* + * If the hdm capability was not mapped there is nothing to enable and + * the caller is responsible for what happens next. For example, + * emulate a passthrough decoder. + */ + if (IS_ERR(cxlhdm)) + return 0; + + hdm = cxlhdm->regs.hdm_decoder; global_ctrl = readl(hdm + CXL_HDM_DECODER_CTRL_OFFSET); + + /* + * If the HDM decoder capability was enabled on entry, skip + * registering disable_hdm() since this decode capability may be + * owned by platform firmware. + */ + if (global_ctrl & CXL_HDM_DECODER_ENABLE) + return 0; + writel(global_ctrl | CXL_HDM_DECODER_ENABLE, hdm + CXL_HDM_DECODER_CTRL_OFFSET); - return devm_add_action_or_reset(host, disable_hdm, cxlhdm); + return devm_add_action_or_reset(&port->dev, disable_hdm, cxlhdm); } +EXPORT_SYMBOL_NS_GPL(devm_cxl_enable_hdm, CXL); int cxl_dvsec_rr_decode(struct device *dev, int d, struct cxl_endpoint_dvsec_info *info) @@ -425,7 +511,7 @@ int cxl_hdm_decode_init(struct cxl_dev_state *cxlds, struct cxl_hdm *cxlhdm, if (info->mem_enabled) return 0; - rc = devm_cxl_enable_hdm(&port->dev, cxlhdm); + rc = devm_cxl_enable_hdm(port, cxlhdm); if (rc) return rc; diff --git a/drivers/cxl/core/port.c b/drivers/cxl/core/port.c index da2068475fa26c5b4cea086e8433bcec77299f07..e7c284c890bc111eeefd19e369e996b4b6966f3f 100644 --- a/drivers/cxl/core/port.c +++ b/drivers/cxl/core/port.c @@ -750,11 +750,10 @@ struct cxl_port *devm_cxl_add_port(struct device *host, struct device *uport, parent_port = parent_dport ? parent_dport->port : NULL; if (IS_ERR(port)) { - dev_dbg(uport, "Failed to add %s%s%s%s: %ld\n", - dev_name(&port->dev), - parent_port ? " to " : "", + dev_dbg(uport, "Failed to add%s%s%s: %ld\n", + parent_port ? " port to " : "", parent_port ? dev_name(&parent_port->dev) : "", - parent_port ? "" : " (root port)", + parent_port ? "" : " root port", PTR_ERR(port)); } else { dev_dbg(uport, "%s added%s%s%s\n", diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h index 044a92d9813e239d1b97e7a4935f36850c30e154..f93a285389621cc77e86ae792c095b1671f30c85 100644 --- a/drivers/cxl/cxl.h +++ b/drivers/cxl/cxl.h @@ -710,6 +710,7 @@ struct cxl_endpoint_dvsec_info { struct cxl_hdm; struct cxl_hdm *devm_cxl_setup_hdm(struct cxl_port *port, struct cxl_endpoint_dvsec_info *info); +int devm_cxl_enable_hdm(struct cxl_port *port, struct cxl_hdm *cxlhdm); int devm_cxl_enumerate_decoders(struct cxl_hdm *cxlhdm, struct cxl_endpoint_dvsec_info *info); int devm_cxl_add_passthrough_decoder(struct cxl_port *port); diff --git a/drivers/cxl/cxlmem.h b/drivers/cxl/cxlmem.h index db12b6313afbf734fba7c38302a28c56c44950a1..a2845a7a69d820d9b4f04ae6a681839728d0f31c 100644 --- a/drivers/cxl/cxlmem.h +++ b/drivers/cxl/cxlmem.h @@ -266,6 +266,7 @@ struct cxl_poison_state { * @regs: Parsed register blocks * @cxl_dvsec: Offset to the PCIe device DVSEC * @rcd: operating in RCD mode (CXL 3.0 9.11.8 CXL Devices Attached to an RCH) + * @media_ready: Indicate whether the device media is usable * @payload_size: Size of space for payload * (CXL 2.0 8.2.8.4.3 Mailbox Capabilities Register) * @lsa_size: Size of Label Storage Area @@ -303,6 +304,7 @@ struct cxl_dev_state { int cxl_dvsec; bool rcd; + bool media_ready; size_t payload_size; size_t lsa_size; struct mutex mbox_mutex; /* Protects device mailbox and firmware */ diff --git a/drivers/cxl/cxlpci.h b/drivers/cxl/cxlpci.h index 0465ef963cd6a0b23dae3cfb7177d9a033dcde18..7c02e55b80429a9390d3322407701033bb6c01ce 100644 --- a/drivers/cxl/cxlpci.h +++ b/drivers/cxl/cxlpci.h @@ -31,6 +31,8 @@ #define CXL_DVSEC_RANGE_BASE_LOW(i) (0x24 + (i * 0x10)) #define CXL_DVSEC_MEM_BASE_LOW_MASK GENMASK(31, 28) +#define CXL_DVSEC_RANGE_MAX 2 + /* CXL 2.0 8.1.4: Non-CXL Function Map DVSEC */ #define CXL_DVSEC_FUNCTION_MAP 2 diff --git a/drivers/cxl/mem.c b/drivers/cxl/mem.c index 10caf180b3fa0bff3e1363c269ad4fe77cc7cff8..519edd0eb1967d8c0d0ec7bb5e934b267e37c1ff 100644 --- a/drivers/cxl/mem.c +++ b/drivers/cxl/mem.c @@ -124,6 +124,9 @@ static int cxl_mem_probe(struct device *dev) struct dentry *dentry; int rc; + if (!cxlds->media_ready) + return -EBUSY; + /* * Someone is trying to reattach this device after it lost its port * connection (an endpoint port previously registered by this memdev was diff --git a/drivers/cxl/pci.c b/drivers/cxl/pci.c index f7a5b8e9c102411ca5bfde2f1995488498aa7e91..0872f2233ed0c88aa9fec7e9e93935dfa1f0e740 100644 --- a/drivers/cxl/pci.c +++ b/drivers/cxl/pci.c @@ -708,6 +708,12 @@ static int cxl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) if (rc) dev_dbg(&pdev->dev, "Failed to map RAS capability.\n"); + rc = cxl_await_media_ready(cxlds); + if (rc == 0) + cxlds->media_ready = true; + else + dev_warn(&pdev->dev, "Media not active (%d)\n", rc); + rc = cxl_pci_setup_mailbox(cxlds); if (rc) return rc; diff --git a/drivers/cxl/port.c b/drivers/cxl/port.c index eb57324c4ad4a0e54c9a35500df9c3a216f603a2..c23b6164e1c0fa5c1e9a9c671c2e1d29da604c69 100644 --- a/drivers/cxl/port.c +++ b/drivers/cxl/port.c @@ -60,13 +60,17 @@ static int discover_region(struct device *dev, void *root) static int cxl_switch_port_probe(struct cxl_port *port) { struct cxl_hdm *cxlhdm; - int rc; + int rc, nr_dports; - rc = devm_cxl_port_enumerate_dports(port); - if (rc < 0) - return rc; + nr_dports = devm_cxl_port_enumerate_dports(port); + if (nr_dports < 0) + return nr_dports; cxlhdm = devm_cxl_setup_hdm(port, NULL); + rc = devm_cxl_enable_hdm(port, cxlhdm); + if (rc) + return rc; + if (!IS_ERR(cxlhdm)) return devm_cxl_enumerate_decoders(cxlhdm, NULL); @@ -75,7 +79,7 @@ static int cxl_switch_port_probe(struct cxl_port *port) return PTR_ERR(cxlhdm); } - if (rc == 1) { + if (nr_dports == 1) { dev_dbg(&port->dev, "Fallback to passthrough decoder\n"); return devm_cxl_add_passthrough_decoder(port); } @@ -113,12 +117,6 @@ static int cxl_endpoint_port_probe(struct cxl_port *port) if (rc) return rc; - rc = cxl_await_media_ready(cxlds); - if (rc) { - dev_err(&port->dev, "Media not active (%d)\n", rc); - return rc; - } - rc = devm_cxl_enumerate_decoders(cxlhdm, &info); if (rc) return rc; diff --git a/drivers/dma/at_hdmac.c b/drivers/dma/at_hdmac.c index 8858470246e1552e5e278a2b254f8ae874df894b..ee3a219e3a897c56ac69d7a902ef90ccc0262822 100644 --- a/drivers/dma/at_hdmac.c +++ b/drivers/dma/at_hdmac.c @@ -132,7 +132,7 @@ #define ATC_DST_PIP BIT(12) /* Destination Picture-in-Picture enabled */ #define ATC_SRC_DSCR_DIS BIT(16) /* Src Descriptor fetch disable */ #define ATC_DST_DSCR_DIS BIT(20) /* Dst Descriptor fetch disable */ -#define ATC_FC GENMASK(22, 21) /* Choose Flow Controller */ +#define ATC_FC GENMASK(23, 21) /* Choose Flow Controller */ #define ATC_FC_MEM2MEM 0x0 /* Mem-to-Mem (DMA) */ #define ATC_FC_MEM2PER 0x1 /* Mem-to-Periph (DMA) */ #define ATC_FC_PER2MEM 0x2 /* Periph-to-Mem (DMA) */ @@ -153,8 +153,6 @@ #define ATC_AUTO BIT(31) /* Auto multiple buffer tx enable */ /* Bitfields in CFG */ -#define ATC_PER_MSB(h) ((0x30U & (h)) >> 4) /* Extract most significant bits of a handshaking identifier */ - #define ATC_SRC_PER GENMASK(3, 0) /* Channel src rq associated with periph handshaking ifc h */ #define ATC_DST_PER GENMASK(7, 4) /* Channel dst rq associated with periph handshaking ifc h */ #define ATC_SRC_REP BIT(8) /* Source Replay Mod */ @@ -181,10 +179,15 @@ #define ATC_DPIP_HOLE GENMASK(15, 0) #define ATC_DPIP_BOUNDARY GENMASK(25, 16) -#define ATC_SRC_PER_ID(id) (FIELD_PREP(ATC_SRC_PER_MSB, (id)) | \ - FIELD_PREP(ATC_SRC_PER, (id))) -#define ATC_DST_PER_ID(id) (FIELD_PREP(ATC_DST_PER_MSB, (id)) | \ - FIELD_PREP(ATC_DST_PER, (id))) +#define ATC_PER_MSB GENMASK(5, 4) /* Extract MSBs of a handshaking identifier */ +#define ATC_SRC_PER_ID(id) \ + ({ typeof(id) _id = (id); \ + FIELD_PREP(ATC_SRC_PER_MSB, FIELD_GET(ATC_PER_MSB, _id)) | \ + FIELD_PREP(ATC_SRC_PER, _id); }) +#define ATC_DST_PER_ID(id) \ + ({ typeof(id) _id = (id); \ + FIELD_PREP(ATC_DST_PER_MSB, FIELD_GET(ATC_PER_MSB, _id)) | \ + FIELD_PREP(ATC_DST_PER, _id); }) diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c index 7da6d9b6098e719e531052c9aa938b669dfc3149..c3b37168b21f166e9690076a1692d965392b37f5 100644 --- a/drivers/dma/at_xdmac.c +++ b/drivers/dma/at_xdmac.c @@ -1102,6 +1102,8 @@ at_xdmac_prep_interleaved(struct dma_chan *chan, NULL, src_addr, dst_addr, xt, xt->sgl); + if (!first) + return NULL; /* Length of the block is (BLEN+1) microblocks. */ for (i = 0; i < xt->numf - 1; i++) @@ -1132,8 +1134,9 @@ at_xdmac_prep_interleaved(struct dma_chan *chan, src_addr, dst_addr, xt, chunk); if (!desc) { - list_splice_tail_init(&first->descs_list, - &atchan->free_descs_list); + if (first) + list_splice_tail_init(&first->descs_list, + &atchan->free_descs_list); return NULL; } diff --git a/drivers/dma/idxd/cdev.c b/drivers/dma/idxd/cdev.c index ecbf67c2ad2b07d51ec6b05764be8dc4c51068db..d32deb9b4e3dee466fd056230596aa42198d8285 100644 --- a/drivers/dma/idxd/cdev.c +++ b/drivers/dma/idxd/cdev.c @@ -277,7 +277,6 @@ static int idxd_cdev_open(struct inode *inode, struct file *filp) if (wq_dedicated(wq)) { rc = idxd_wq_set_pasid(wq, pasid); if (rc < 0) { - iommu_sva_unbind_device(sva); dev_err(dev, "wq set pasid failed: %d\n", rc); goto failed_set_pasid; } diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c index 0d9257fbdfb0d90d902761406de722eff92b1888..b4731fe6bbc14fc7360723726b535a6194383429 100644 --- a/drivers/dma/pl330.c +++ b/drivers/dma/pl330.c @@ -1050,7 +1050,7 @@ static bool _trigger(struct pl330_thread *thrd) return true; } -static bool _start(struct pl330_thread *thrd) +static bool pl330_start_thread(struct pl330_thread *thrd) { switch (_state(thrd)) { case PL330_STATE_FAULT_COMPLETING: @@ -1702,7 +1702,7 @@ static int pl330_update(struct pl330_dmac *pl330) thrd->req_running = -1; /* Get going again ASAP */ - _start(thrd); + pl330_start_thread(thrd); /* For now, just make a list of callbacks to be done */ list_add_tail(&descdone->rqd, &pl330->req_done); @@ -2089,7 +2089,7 @@ static void pl330_tasklet(struct tasklet_struct *t) } else { /* Make sure the PL330 Channel thread is active */ spin_lock(&pch->thread->dmac->lock); - _start(pch->thread); + pl330_start_thread(pch->thread); spin_unlock(&pch->thread->dmac->lock); } @@ -2107,7 +2107,7 @@ static void pl330_tasklet(struct tasklet_struct *t) if (power_down) { pch->active = true; spin_lock(&pch->thread->dmac->lock); - _start(pch->thread); + pl330_start_thread(pch->thread); spin_unlock(&pch->thread->dmac->lock); power_down = false; } diff --git a/drivers/dma/ti/k3-udma.c b/drivers/dma/ti/k3-udma.c index fc3a2a05ab7b70f33ab2c81e618a245941e1222a..b8329a23728d73f2a8a29e1c3965181b2712d635 100644 --- a/drivers/dma/ti/k3-udma.c +++ b/drivers/dma/ti/k3-udma.c @@ -5527,7 +5527,7 @@ static int udma_probe(struct platform_device *pdev) return ret; } -static int udma_pm_suspend(struct device *dev) +static int __maybe_unused udma_pm_suspend(struct device *dev) { struct udma_dev *ud = dev_get_drvdata(dev); struct dma_device *dma_dev = &ud->ddev; @@ -5549,7 +5549,7 @@ static int udma_pm_suspend(struct device *dev) return 0; } -static int udma_pm_resume(struct device *dev) +static int __maybe_unused udma_pm_resume(struct device *dev) { struct udma_dev *ud = dev_get_drvdata(dev); struct dma_device *dma_dev = &ud->ddev; diff --git a/drivers/edac/qcom_edac.c b/drivers/edac/qcom_edac.c index 265e0fb39bc7dcf87f682f5145d7c1274602f00a..b2db545c681031e649883f2c81cc33aecc9a31dd 100644 --- a/drivers/edac/qcom_edac.c +++ b/drivers/edac/qcom_edac.c @@ -21,30 +21,9 @@ #define TRP_SYN_REG_CNT 6 #define DRP_SYN_REG_CNT 8 -#define LLCC_COMMON_STATUS0 0x0003000c #define LLCC_LB_CNT_MASK GENMASK(31, 28) #define LLCC_LB_CNT_SHIFT 28 -/* Single & double bit syndrome register offsets */ -#define TRP_ECC_SB_ERR_SYN0 0x0002304c -#define TRP_ECC_DB_ERR_SYN0 0x00020370 -#define DRP_ECC_SB_ERR_SYN0 0x0004204c -#define DRP_ECC_DB_ERR_SYN0 0x00042070 - -/* Error register offsets */ -#define TRP_ECC_ERROR_STATUS1 0x00020348 -#define TRP_ECC_ERROR_STATUS0 0x00020344 -#define DRP_ECC_ERROR_STATUS1 0x00042048 -#define DRP_ECC_ERROR_STATUS0 0x00042044 - -/* TRP, DRP interrupt register offsets */ -#define DRP_INTERRUPT_STATUS 0x00041000 -#define TRP_INTERRUPT_0_STATUS 0x00020480 -#define DRP_INTERRUPT_CLEAR 0x00041008 -#define DRP_ECC_ERROR_CNTR_CLEAR 0x00040004 -#define TRP_INTERRUPT_0_CLEAR 0x00020484 -#define TRP_ECC_ERROR_CNTR_CLEAR 0x00020440 - /* Mask and shift macros */ #define ECC_DB_ERR_COUNT_MASK GENMASK(4, 0) #define ECC_DB_ERR_WAYS_MASK GENMASK(31, 16) @@ -60,15 +39,6 @@ #define DRP_TRP_INT_CLEAR GENMASK(1, 0) #define DRP_TRP_CNT_CLEAR GENMASK(1, 0) -/* Config registers offsets*/ -#define DRP_ECC_ERROR_CFG 0x00040000 - -/* Tag RAM, Data RAM interrupt register offsets */ -#define CMN_INTERRUPT_0_ENABLE 0x0003001c -#define CMN_INTERRUPT_2_ENABLE 0x0003003c -#define TRP_INTERRUPT_0_ENABLE 0x00020488 -#define DRP_INTERRUPT_ENABLE 0x0004100c - #define SB_ERROR_THRESHOLD 0x1 #define SB_ERROR_THRESHOLD_SHIFT 24 #define SB_DB_TRP_INTERRUPT_ENABLE 0x3 @@ -88,9 +58,6 @@ enum { static const struct llcc_edac_reg_data edac_reg_data[] = { [LLCC_DRAM_CE] = { .name = "DRAM Single-bit", - .synd_reg = DRP_ECC_SB_ERR_SYN0, - .count_status_reg = DRP_ECC_ERROR_STATUS1, - .ways_status_reg = DRP_ECC_ERROR_STATUS0, .reg_cnt = DRP_SYN_REG_CNT, .count_mask = ECC_SB_ERR_COUNT_MASK, .ways_mask = ECC_SB_ERR_WAYS_MASK, @@ -98,9 +65,6 @@ static const struct llcc_edac_reg_data edac_reg_data[] = { }, [LLCC_DRAM_UE] = { .name = "DRAM Double-bit", - .synd_reg = DRP_ECC_DB_ERR_SYN0, - .count_status_reg = DRP_ECC_ERROR_STATUS1, - .ways_status_reg = DRP_ECC_ERROR_STATUS0, .reg_cnt = DRP_SYN_REG_CNT, .count_mask = ECC_DB_ERR_COUNT_MASK, .ways_mask = ECC_DB_ERR_WAYS_MASK, @@ -108,9 +72,6 @@ static const struct llcc_edac_reg_data edac_reg_data[] = { }, [LLCC_TRAM_CE] = { .name = "TRAM Single-bit", - .synd_reg = TRP_ECC_SB_ERR_SYN0, - .count_status_reg = TRP_ECC_ERROR_STATUS1, - .ways_status_reg = TRP_ECC_ERROR_STATUS0, .reg_cnt = TRP_SYN_REG_CNT, .count_mask = ECC_SB_ERR_COUNT_MASK, .ways_mask = ECC_SB_ERR_WAYS_MASK, @@ -118,9 +79,6 @@ static const struct llcc_edac_reg_data edac_reg_data[] = { }, [LLCC_TRAM_UE] = { .name = "TRAM Double-bit", - .synd_reg = TRP_ECC_DB_ERR_SYN0, - .count_status_reg = TRP_ECC_ERROR_STATUS1, - .ways_status_reg = TRP_ECC_ERROR_STATUS0, .reg_cnt = TRP_SYN_REG_CNT, .count_mask = ECC_DB_ERR_COUNT_MASK, .ways_mask = ECC_DB_ERR_WAYS_MASK, @@ -128,7 +86,7 @@ static const struct llcc_edac_reg_data edac_reg_data[] = { }, }; -static int qcom_llcc_core_setup(struct regmap *llcc_bcast_regmap) +static int qcom_llcc_core_setup(struct llcc_drv_data *drv, struct regmap *llcc_bcast_regmap) { u32 sb_err_threshold; int ret; @@ -137,31 +95,31 @@ static int qcom_llcc_core_setup(struct regmap *llcc_bcast_regmap) * Configure interrupt enable registers such that Tag, Data RAM related * interrupts are propagated to interrupt controller for servicing */ - ret = regmap_update_bits(llcc_bcast_regmap, CMN_INTERRUPT_2_ENABLE, + ret = regmap_update_bits(llcc_bcast_regmap, drv->edac_reg_offset->cmn_interrupt_2_enable, TRP0_INTERRUPT_ENABLE, TRP0_INTERRUPT_ENABLE); if (ret) return ret; - ret = regmap_update_bits(llcc_bcast_regmap, TRP_INTERRUPT_0_ENABLE, + ret = regmap_update_bits(llcc_bcast_regmap, drv->edac_reg_offset->trp_interrupt_0_enable, SB_DB_TRP_INTERRUPT_ENABLE, SB_DB_TRP_INTERRUPT_ENABLE); if (ret) return ret; sb_err_threshold = (SB_ERROR_THRESHOLD << SB_ERROR_THRESHOLD_SHIFT); - ret = regmap_write(llcc_bcast_regmap, DRP_ECC_ERROR_CFG, + ret = regmap_write(llcc_bcast_regmap, drv->edac_reg_offset->drp_ecc_error_cfg, sb_err_threshold); if (ret) return ret; - ret = regmap_update_bits(llcc_bcast_regmap, CMN_INTERRUPT_2_ENABLE, + ret = regmap_update_bits(llcc_bcast_regmap, drv->edac_reg_offset->cmn_interrupt_2_enable, DRP0_INTERRUPT_ENABLE, DRP0_INTERRUPT_ENABLE); if (ret) return ret; - ret = regmap_write(llcc_bcast_regmap, DRP_INTERRUPT_ENABLE, + ret = regmap_write(llcc_bcast_regmap, drv->edac_reg_offset->drp_interrupt_enable, SB_DB_DRP_INTERRUPT_ENABLE); return ret; } @@ -170,29 +128,33 @@ static int qcom_llcc_core_setup(struct regmap *llcc_bcast_regmap) static int qcom_llcc_clear_error_status(int err_type, struct llcc_drv_data *drv) { - int ret = 0; + int ret; switch (err_type) { case LLCC_DRAM_CE: case LLCC_DRAM_UE: - ret = regmap_write(drv->bcast_regmap, DRP_INTERRUPT_CLEAR, + ret = regmap_write(drv->bcast_regmap, + drv->edac_reg_offset->drp_interrupt_clear, DRP_TRP_INT_CLEAR); if (ret) return ret; - ret = regmap_write(drv->bcast_regmap, DRP_ECC_ERROR_CNTR_CLEAR, + ret = regmap_write(drv->bcast_regmap, + drv->edac_reg_offset->drp_ecc_error_cntr_clear, DRP_TRP_CNT_CLEAR); if (ret) return ret; break; case LLCC_TRAM_CE: case LLCC_TRAM_UE: - ret = regmap_write(drv->bcast_regmap, TRP_INTERRUPT_0_CLEAR, + ret = regmap_write(drv->bcast_regmap, + drv->edac_reg_offset->trp_interrupt_0_clear, DRP_TRP_INT_CLEAR); if (ret) return ret; - ret = regmap_write(drv->bcast_regmap, TRP_ECC_ERROR_CNTR_CLEAR, + ret = regmap_write(drv->bcast_regmap, + drv->edac_reg_offset->trp_ecc_error_cntr_clear, DRP_TRP_CNT_CLEAR); if (ret) return ret; @@ -205,16 +167,54 @@ qcom_llcc_clear_error_status(int err_type, struct llcc_drv_data *drv) return ret; } +struct qcom_llcc_syn_regs { + u32 synd_reg; + u32 count_status_reg; + u32 ways_status_reg; +}; + +static void get_reg_offsets(struct llcc_drv_data *drv, int err_type, + struct qcom_llcc_syn_regs *syn_regs) +{ + const struct llcc_edac_reg_offset *edac_reg_offset = drv->edac_reg_offset; + + switch (err_type) { + case LLCC_DRAM_CE: + syn_regs->synd_reg = edac_reg_offset->drp_ecc_sb_err_syn0; + syn_regs->count_status_reg = edac_reg_offset->drp_ecc_error_status1; + syn_regs->ways_status_reg = edac_reg_offset->drp_ecc_error_status0; + break; + case LLCC_DRAM_UE: + syn_regs->synd_reg = edac_reg_offset->drp_ecc_db_err_syn0; + syn_regs->count_status_reg = edac_reg_offset->drp_ecc_error_status1; + syn_regs->ways_status_reg = edac_reg_offset->drp_ecc_error_status0; + break; + case LLCC_TRAM_CE: + syn_regs->synd_reg = edac_reg_offset->trp_ecc_sb_err_syn0; + syn_regs->count_status_reg = edac_reg_offset->trp_ecc_error_status1; + syn_regs->ways_status_reg = edac_reg_offset->trp_ecc_error_status0; + break; + case LLCC_TRAM_UE: + syn_regs->synd_reg = edac_reg_offset->trp_ecc_db_err_syn0; + syn_regs->count_status_reg = edac_reg_offset->trp_ecc_error_status1; + syn_regs->ways_status_reg = edac_reg_offset->trp_ecc_error_status0; + break; + } +} + /* Dump Syndrome registers data for Tag RAM, Data RAM bit errors*/ static int dump_syn_reg_values(struct llcc_drv_data *drv, u32 bank, int err_type) { struct llcc_edac_reg_data reg_data = edac_reg_data[err_type]; + struct qcom_llcc_syn_regs regs = { }; int err_cnt, err_ways, ret, i; u32 synd_reg, synd_val; + get_reg_offsets(drv, err_type, ®s); + for (i = 0; i < reg_data.reg_cnt; i++) { - synd_reg = reg_data.synd_reg + (i * 4); + synd_reg = regs.synd_reg + (i * 4); ret = regmap_read(drv->regmaps[bank], synd_reg, &synd_val); if (ret) @@ -224,7 +224,7 @@ dump_syn_reg_values(struct llcc_drv_data *drv, u32 bank, int err_type) reg_data.name, i, synd_val); } - ret = regmap_read(drv->regmaps[bank], reg_data.count_status_reg, + ret = regmap_read(drv->regmaps[bank], regs.count_status_reg, &err_cnt); if (ret) goto clear; @@ -234,7 +234,7 @@ dump_syn_reg_values(struct llcc_drv_data *drv, u32 bank, int err_type) edac_printk(KERN_CRIT, EDAC_LLCC, "%s: Error count: 0x%4x\n", reg_data.name, err_cnt); - ret = regmap_read(drv->regmaps[bank], reg_data.ways_status_reg, + ret = regmap_read(drv->regmaps[bank], regs.ways_status_reg, &err_ways); if (ret) goto clear; @@ -295,7 +295,7 @@ static irqreturn_t llcc_ecc_irq_handler(int irq, void *edev_ctl) /* Iterate over the banks and look for Tag RAM or Data RAM errors */ for (i = 0; i < drv->num_banks; i++) { - ret = regmap_read(drv->regmaps[i], DRP_INTERRUPT_STATUS, + ret = regmap_read(drv->regmaps[i], drv->edac_reg_offset->drp_interrupt_status, &drp_error); if (!ret && (drp_error & SB_ECC_ERROR)) { @@ -310,7 +310,7 @@ static irqreturn_t llcc_ecc_irq_handler(int irq, void *edev_ctl) if (!ret) irq_rc = IRQ_HANDLED; - ret = regmap_read(drv->regmaps[i], TRP_INTERRUPT_0_STATUS, + ret = regmap_read(drv->regmaps[i], drv->edac_reg_offset->trp_interrupt_0_status, &trp_error); if (!ret && (trp_error & SB_ECC_ERROR)) { @@ -342,7 +342,7 @@ static int qcom_llcc_edac_probe(struct platform_device *pdev) int ecc_irq; int rc; - rc = qcom_llcc_core_setup(llcc_driv_data->bcast_regmap); + rc = qcom_llcc_core_setup(llcc_driv_data, llcc_driv_data->bcast_regmap); if (rc) return rc; diff --git a/drivers/firmware/arm_ffa/bus.c b/drivers/firmware/arm_ffa/bus.c index f29d77ecf72db613724dc175cdba1a001fc0ac60..2b8bfcd010f5fd34e59b3eacdec0677e6692f020 100644 --- a/drivers/firmware/arm_ffa/bus.c +++ b/drivers/firmware/arm_ffa/bus.c @@ -15,6 +15,8 @@ #include "common.h" +static DEFINE_IDA(ffa_bus_id); + static int ffa_device_match(struct device *dev, struct device_driver *drv) { const struct ffa_device_id *id_table; @@ -53,7 +55,8 @@ static void ffa_device_remove(struct device *dev) { struct ffa_driver *ffa_drv = to_ffa_driver(dev->driver); - ffa_drv->remove(to_ffa_dev(dev)); + if (ffa_drv->remove) + ffa_drv->remove(to_ffa_dev(dev)); } static int ffa_device_uevent(const struct device *dev, struct kobj_uevent_env *env) @@ -130,6 +133,7 @@ static void ffa_release_device(struct device *dev) { struct ffa_device *ffa_dev = to_ffa_dev(dev); + ida_free(&ffa_bus_id, ffa_dev->id); kfree(ffa_dev); } @@ -170,18 +174,24 @@ bool ffa_device_is_valid(struct ffa_device *ffa_dev) struct ffa_device *ffa_device_register(const uuid_t *uuid, int vm_id, const struct ffa_ops *ops) { - int ret; + int id, ret; struct device *dev; struct ffa_device *ffa_dev; + id = ida_alloc_min(&ffa_bus_id, 1, GFP_KERNEL); + if (id < 0) + return NULL; + ffa_dev = kzalloc(sizeof(*ffa_dev), GFP_KERNEL); - if (!ffa_dev) + if (!ffa_dev) { + ida_free(&ffa_bus_id, id); return NULL; + } dev = &ffa_dev->dev; dev->bus = &ffa_bus_type; dev->release = ffa_release_device; - dev_set_name(&ffa_dev->dev, "arm-ffa-%04x", vm_id); + dev_set_name(&ffa_dev->dev, "arm-ffa-%d", id); ffa_dev->vm_id = vm_id; ffa_dev->ops = ops; @@ -217,4 +227,5 @@ void arm_ffa_bus_exit(void) { ffa_devices_unregister(); bus_unregister(&ffa_bus_type); + ida_destroy(&ffa_bus_id); } diff --git a/drivers/firmware/arm_ffa/driver.c b/drivers/firmware/arm_ffa/driver.c index fa85c64d3dede7edcc6c35b3d9959b868405d7a2..2109cd178ff706f3ec7b0db0735d18519ba639cb 100644 --- a/drivers/firmware/arm_ffa/driver.c +++ b/drivers/firmware/arm_ffa/driver.c @@ -193,7 +193,8 @@ __ffa_partition_info_get(u32 uuid0, u32 uuid1, u32 uuid2, u32 uuid3, int idx, count, flags = 0, sz, buf_sz; ffa_value_t partition_info; - if (!buffer || !num_partitions) /* Just get the count for now */ + if (drv_info->version > FFA_VERSION_1_0 && + (!buffer || !num_partitions)) /* Just get the count for now */ flags = PARTITION_INFO_GET_RETURN_COUNT_ONLY; mutex_lock(&drv_info->rx_lock); @@ -420,12 +421,18 @@ ffa_setup_and_transmit(u32 func_id, void *buffer, u32 max_fragsize, ep_mem_access->receiver = args->attrs[idx].receiver; ep_mem_access->attrs = args->attrs[idx].attrs; ep_mem_access->composite_off = COMPOSITE_OFFSET(args->nattrs); + ep_mem_access->flag = 0; + ep_mem_access->reserved = 0; } + mem_region->handle = 0; + mem_region->reserved_0 = 0; + mem_region->reserved_1 = 0; mem_region->ep_count = args->nattrs; composite = buffer + COMPOSITE_OFFSET(args->nattrs); composite->total_pg_cnt = ffa_get_num_pages_sg(args->sg); composite->addr_range_cnt = num_entries; + composite->reserved = 0; length = COMPOSITE_CONSTITUENTS_OFFSET(args->nattrs, num_entries); frag_len = COMPOSITE_CONSTITUENTS_OFFSET(args->nattrs, 0); @@ -460,6 +467,7 @@ ffa_setup_and_transmit(u32 func_id, void *buffer, u32 max_fragsize, constituents->address = sg_phys(args->sg); constituents->pg_cnt = args->sg->length / FFA_PAGE_SIZE; + constituents->reserved = 0; constituents++; frag_len += sizeof(struct ffa_mem_region_addr_range); } while ((args->sg = sg_next(args->sg))); diff --git a/drivers/firmware/arm_scmi/raw_mode.c b/drivers/firmware/arm_scmi/raw_mode.c index d40df099fd515276b87cdde144271c60954b46be..6971dcf72fb998425655865fd6ccf4e1ecb44976 100644 --- a/drivers/firmware/arm_scmi/raw_mode.c +++ b/drivers/firmware/arm_scmi/raw_mode.c @@ -1066,7 +1066,7 @@ static int scmi_xfer_raw_worker_init(struct scmi_raw_mode_info *raw) raw->wait_wq = alloc_workqueue("scmi-raw-wait-wq-%d", WQ_UNBOUND | WQ_FREEZABLE | - WQ_HIGHPRI, WQ_SYSFS, raw->id); + WQ_HIGHPRI | WQ_SYSFS, 0, raw->id); if (!raw->wait_wq) return -ENOMEM; diff --git a/drivers/firmware/cirrus/cs_dsp.c b/drivers/firmware/cirrus/cs_dsp.c index e4ccfb6a8fa52fefa852b009e161afb3ba23c264..ec056f6f40ce81ddd5a09578193811e9b9d1b9c1 100644 --- a/drivers/firmware/cirrus/cs_dsp.c +++ b/drivers/firmware/cirrus/cs_dsp.c @@ -2124,6 +2124,7 @@ static int cs_dsp_load_coeff(struct cs_dsp *dsp, const struct firmware *firmware file, blocks, le32_to_cpu(blk->len), type, le32_to_cpu(blk->id)); + region_name = cs_dsp_mem_region_name(type); mem = cs_dsp_find_region(dsp, type); if (!mem) { cs_dsp_err(dsp, "No base for region %x\n", type); @@ -2147,8 +2148,8 @@ static int cs_dsp_load_coeff(struct cs_dsp *dsp, const struct firmware *firmware reg = dsp->ops->region_to_reg(mem, reg); reg += offset; } else { - cs_dsp_err(dsp, "No %x for algorithm %x\n", - type, le32_to_cpu(blk->id)); + cs_dsp_err(dsp, "No %s for algorithm %x\n", + region_name, le32_to_cpu(blk->id)); } break; diff --git a/drivers/firmware/efi/libstub/Makefile.zboot b/drivers/firmware/efi/libstub/Makefile.zboot index 89ef820f3b34483a55cd6f9b30043910922da75e..2c489627a8078945f26071644cc5e5390c2ee0d9 100644 --- a/drivers/firmware/efi/libstub/Makefile.zboot +++ b/drivers/firmware/efi/libstub/Makefile.zboot @@ -32,7 +32,8 @@ zboot-size-len-$(CONFIG_KERNEL_GZIP) := 0 $(obj)/vmlinuz: $(obj)/vmlinux.bin FORCE $(call if_changed,$(zboot-method-y)) -OBJCOPYFLAGS_vmlinuz.o := -I binary -O $(EFI_ZBOOT_BFD_TARGET) $(EFI_ZBOOT_OBJCOPY_FLAGS) \ +# avoid eager evaluation to prevent references to non-existent build artifacts +OBJCOPYFLAGS_vmlinuz.o = -I binary -O $(EFI_ZBOOT_BFD_TARGET) $(EFI_ZBOOT_OBJCOPY_FLAGS) \ --rename-section .data=.gzdata,load,alloc,readonly,contents $(obj)/vmlinuz.o: $(obj)/vmlinuz FORCE $(call if_changed,objcopy) diff --git a/drivers/firmware/efi/libstub/efistub.h b/drivers/firmware/efi/libstub/efistub.h index 67d5a20802e0b7c6bff01a8da5c083f1768b6e58..54a2822cae7714584c668dd9e0650350e30922df 100644 --- a/drivers/firmware/efi/libstub/efistub.h +++ b/drivers/firmware/efi/libstub/efistub.h @@ -1133,4 +1133,7 @@ const u8 *__efi_get_smbios_string(const struct efi_smbios_record *record, void efi_remap_image(unsigned long image_base, unsigned alloc_size, unsigned long code_size); +asmlinkage efi_status_t __efiapi +efi_zboot_entry(efi_handle_t handle, efi_system_table_t *systab); + #endif diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig index 5521f060d58e9df3c45ab9bef89295eb3efac6aa..f45c6a36551c75533a92d6c6bf5897162b9ee5e8 100644 --- a/drivers/gpio/Kconfig +++ b/drivers/gpio/Kconfig @@ -897,7 +897,7 @@ config GPIO_F7188X help This option enables support for GPIOs found on Fintek Super-I/O chips F71869, F71869A, F71882FG, F71889F and F81866. - As well as Nuvoton Super-I/O chip NCT6116D. + As well as Nuvoton Super-I/O chip NCT6126D. To compile this driver as a module, choose M here: the module will be called f7188x-gpio. diff --git a/drivers/gpio/gpio-f7188x.c b/drivers/gpio/gpio-f7188x.c index 9effa7769bef5bcdea5865b964f78f261da40dfe..f54ca5a1775ea2178004e747286660bcf83900d1 100644 --- a/drivers/gpio/gpio-f7188x.c +++ b/drivers/gpio/gpio-f7188x.c @@ -48,7 +48,7 @@ /* * Nuvoton devices. */ -#define SIO_NCT6116D_ID 0xD283 /* NCT6116D chipset ID */ +#define SIO_NCT6126D_ID 0xD283 /* NCT6126D chipset ID */ #define SIO_LD_GPIO_NUVOTON 0x07 /* GPIO logical device */ @@ -62,7 +62,7 @@ enum chips { f81866, f81804, f81865, - nct6116d, + nct6126d, }; static const char * const f7188x_names[] = { @@ -74,7 +74,7 @@ static const char * const f7188x_names[] = { "f81866", "f81804", "f81865", - "nct6116d", + "nct6126d", }; struct f7188x_sio { @@ -187,8 +187,8 @@ static int f7188x_gpio_set_config(struct gpio_chip *chip, unsigned offset, /* Output mode register (0:open drain 1:push-pull). */ #define f7188x_gpio_out_mode(base) ((base) + 3) -#define f7188x_gpio_dir_invert(type) ((type) == nct6116d) -#define f7188x_gpio_data_single(type) ((type) == nct6116d) +#define f7188x_gpio_dir_invert(type) ((type) == nct6126d) +#define f7188x_gpio_data_single(type) ((type) == nct6126d) static struct f7188x_gpio_bank f71869_gpio_bank[] = { F7188X_GPIO_BANK(0, 6, 0xF0, DRVNAME "-0"), @@ -274,7 +274,7 @@ static struct f7188x_gpio_bank f81865_gpio_bank[] = { F7188X_GPIO_BANK(60, 5, 0x90, DRVNAME "-6"), }; -static struct f7188x_gpio_bank nct6116d_gpio_bank[] = { +static struct f7188x_gpio_bank nct6126d_gpio_bank[] = { F7188X_GPIO_BANK(0, 8, 0xE0, DRVNAME "-0"), F7188X_GPIO_BANK(10, 8, 0xE4, DRVNAME "-1"), F7188X_GPIO_BANK(20, 8, 0xE8, DRVNAME "-2"), @@ -282,7 +282,7 @@ static struct f7188x_gpio_bank nct6116d_gpio_bank[] = { F7188X_GPIO_BANK(40, 8, 0xF0, DRVNAME "-4"), F7188X_GPIO_BANK(50, 8, 0xF4, DRVNAME "-5"), F7188X_GPIO_BANK(60, 8, 0xF8, DRVNAME "-6"), - F7188X_GPIO_BANK(70, 1, 0xFC, DRVNAME "-7"), + F7188X_GPIO_BANK(70, 8, 0xFC, DRVNAME "-7"), }; static int f7188x_gpio_get_direction(struct gpio_chip *chip, unsigned offset) @@ -490,9 +490,9 @@ static int f7188x_gpio_probe(struct platform_device *pdev) data->nr_bank = ARRAY_SIZE(f81865_gpio_bank); data->bank = f81865_gpio_bank; break; - case nct6116d: - data->nr_bank = ARRAY_SIZE(nct6116d_gpio_bank); - data->bank = nct6116d_gpio_bank; + case nct6126d: + data->nr_bank = ARRAY_SIZE(nct6126d_gpio_bank); + data->bank = nct6126d_gpio_bank; break; default: return -ENODEV; @@ -559,9 +559,9 @@ static int __init f7188x_find(int addr, struct f7188x_sio *sio) case SIO_F81865_ID: sio->type = f81865; break; - case SIO_NCT6116D_ID: + case SIO_NCT6126D_ID: sio->device = SIO_LD_GPIO_NUVOTON; - sio->type = nct6116d; + sio->type = nct6126d; break; default: pr_info("Unsupported Fintek device 0x%04x\n", devid); @@ -569,7 +569,7 @@ static int __init f7188x_find(int addr, struct f7188x_sio *sio) } /* double check manufacturer where possible */ - if (sio->type != nct6116d) { + if (sio->type != nct6126d) { manid = superio_inw(addr, SIO_FINTEK_MANID); if (manid != SIO_FINTEK_ID) { pr_debug("Not a Fintek device at 0x%08x\n", addr); @@ -581,7 +581,7 @@ static int __init f7188x_find(int addr, struct f7188x_sio *sio) err = 0; pr_info("Found %s at %#x\n", f7188x_names[sio->type], (unsigned int)addr); - if (sio->type != nct6116d) + if (sio->type != nct6126d) pr_info(" revision %d\n", superio_inb(addr, SIO_FINTEK_DEVREV)); err: diff --git a/drivers/gpio/gpio-mockup.c b/drivers/gpio/gpio-mockup.c index e6a7049bef6417affd16c79982d3eeb7c4e6c255..b32063ac845a5073f67593c433f2fcc7694362c7 100644 --- a/drivers/gpio/gpio-mockup.c +++ b/drivers/gpio/gpio-mockup.c @@ -369,7 +369,7 @@ static void gpio_mockup_debugfs_setup(struct device *dev, priv->offset = i; priv->desc = gpiochip_get_desc(gc, i); - debugfs_create_file(name, 0200, chip->dbg_dir, priv, + debugfs_create_file(name, 0600, chip->dbg_dir, priv, &gpio_mockup_debugfs_ops); } } diff --git a/drivers/gpio/gpio-sim.c b/drivers/gpio/gpio-sim.c index a1c8702f362ce76ce7b1ae96ad0b3d378eec0ab6..8b49b0abacd518448ab7a49629a413d4388a35c1 100644 --- a/drivers/gpio/gpio-sim.c +++ b/drivers/gpio/gpio-sim.c @@ -696,6 +696,9 @@ static char **gpio_sim_make_line_names(struct gpio_sim_bank *bank, char **line_names; list_for_each_entry(line, &bank->line_list, siblings) { + if (line->offset >= bank->num_lines) + continue; + if (line->name) { if (line->offset > max_offset) max_offset = line->offset; @@ -721,8 +724,13 @@ static char **gpio_sim_make_line_names(struct gpio_sim_bank *bank, if (!line_names) return ERR_PTR(-ENOMEM); - list_for_each_entry(line, &bank->line_list, siblings) - line_names[line->offset] = line->name; + list_for_each_entry(line, &bank->line_list, siblings) { + if (line->offset >= bank->num_lines) + continue; + + if (line->name && (line->offset <= max_offset)) + line_names[line->offset] = line->name; + } return line_names; } @@ -754,6 +762,9 @@ static int gpio_sim_add_hogs(struct gpio_sim_device *dev) list_for_each_entry(bank, &dev->bank_list, siblings) { list_for_each_entry(line, &bank->line_list, siblings) { + if (line->offset >= bank->num_lines) + continue; + if (line->hog) num_hogs++; } @@ -769,6 +780,9 @@ static int gpio_sim_add_hogs(struct gpio_sim_device *dev) list_for_each_entry(bank, &dev->bank_list, siblings) { list_for_each_entry(line, &bank->line_list, siblings) { + if (line->offset >= bank->num_lines) + continue; + if (!line->hog) continue; diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c index 04fb05df805b016a08b8b70703ddcb7675795ff4..a7220e04a93e562dbc345705c36738a8464282f4 100644 --- a/drivers/gpio/gpiolib.c +++ b/drivers/gpio/gpiolib.c @@ -209,6 +209,8 @@ static int gpiochip_find_base(int ngpio) break; /* nope, check the space right after the chip */ base = gdev->base + gdev->ngpio; + if (base < GPIO_DYNAMIC_BASE) + base = GPIO_DYNAMIC_BASE; } if (gpio_is_valid(base)) { diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig index bb2e48cc6cd676a5b1fbeb71ba400fac1e86bf81..afb3b2f5f42530138e71eb0006193e26f1d728c3 100644 --- a/drivers/gpu/drm/Kconfig +++ b/drivers/gpu/drm/Kconfig @@ -289,9 +289,7 @@ source "drivers/gpu/drm/armada/Kconfig" source "drivers/gpu/drm/atmel-hlcdc/Kconfig" -source "drivers/gpu/drm/rcar-du/Kconfig" - -source "drivers/gpu/drm/shmobile/Kconfig" +source "drivers/gpu/drm/renesas/Kconfig" source "drivers/gpu/drm/sun4i/Kconfig" diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile index a33257d2bc7fd52f9822474917e0957cfe23dd71..7a09a89b493befd9d93d65cb4b303f1b57261c1a 100644 --- a/drivers/gpu/drm/Makefile +++ b/drivers/gpu/drm/Makefile @@ -140,6 +140,7 @@ obj-$(CONFIG_DRM_TTM) += ttm/ obj-$(CONFIG_DRM_SCHED) += scheduler/ obj-$(CONFIG_DRM_RADEON)+= radeon/ obj-$(CONFIG_DRM_AMDGPU)+= amd/amdgpu/ +obj-$(CONFIG_DRM_AMDGPU)+= amd/amdxcp/ obj-$(CONFIG_DRM_I915) += i915/ obj-$(CONFIG_DRM_KMB_DISPLAY) += kmb/ obj-$(CONFIG_DRM_MGAG200) += mgag200/ @@ -156,8 +157,7 @@ obj-$(CONFIG_DRM_UDL) += udl/ obj-$(CONFIG_DRM_AST) += ast/ obj-$(CONFIG_DRM_ARMADA) += armada/ obj-$(CONFIG_DRM_ATMEL_HLCDC) += atmel-hlcdc/ -obj-y += rcar-du/ -obj-$(CONFIG_DRM_SHMOBILE) +=shmobile/ +obj-y += renesas/ obj-y += omapdrm/ obj-$(CONFIG_DRM_SUN4I) += sun4i/ obj-y += tilcdc/ diff --git a/drivers/gpu/drm/amd/amdgpu/Kconfig b/drivers/gpu/drm/amd/amdgpu/Kconfig index 12adca8c7819c55ccb80f33a4e0d448df60332e6..b91e79c721e2bc62793f32997364f51c9fe3ba3d 100644 --- a/drivers/gpu/drm/amd/amdgpu/Kconfig +++ b/drivers/gpu/drm/amd/amdgpu/Kconfig @@ -69,6 +69,16 @@ config DRM_AMDGPU_USERPTR This option selects CONFIG_HMM and CONFIG_HMM_MIRROR if it isn't already selected to enabled full userptr support. +config DRM_AMDGPU_WERROR + bool "Force the compiler to throw an error instead of a warning when compiling" + depends on DRM_AMDGPU + depends on EXPERT + depends on !COMPILE_TEST + default n + help + Add -Werror to the build flags for amdgpu.ko. + Only enable this if you are warning code for amdgpu.ko. + source "drivers/gpu/drm/amd/acp/Kconfig" source "drivers/gpu/drm/amd/display/Kconfig" source "drivers/gpu/drm/amd/amdkfd/Kconfig" diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile index 415a7fa395c4c98993357ea62b7b41f430ae524b..8d16f280b6950357090ac7a91f622381429d8e26 100644 --- a/drivers/gpu/drm/amd/amdgpu/Makefile +++ b/drivers/gpu/drm/amd/amdgpu/Makefile @@ -39,6 +39,26 @@ ccflags-y := -I$(FULL_AMD_PATH)/include/asic_reg \ -I$(FULL_AMD_DISPLAY_PATH)/amdgpu_dm \ -I$(FULL_AMD_PATH)/amdkfd +subdir-ccflags-y := -Wextra +subdir-ccflags-y += -Wunused +subdir-ccflags-y += -Wmissing-prototypes +subdir-ccflags-y += -Wmissing-declarations +subdir-ccflags-y += -Wmissing-include-dirs +subdir-ccflags-y += -Wold-style-definition +subdir-ccflags-y += -Wmissing-format-attribute +# Need this to avoid recursive variable evaluation issues +cond-flags := $(call cc-option, -Wunused-but-set-variable) \ + $(call cc-option, -Wunused-const-variable) \ + $(call cc-option, -Wstringop-truncation) \ + $(call cc-option, -Wpacked-not-aligned) +subdir-ccflags-y += $(cond-flags) +subdir-ccflags-y += -Wno-unused-parameter +subdir-ccflags-y += -Wno-type-limits +subdir-ccflags-y += -Wno-sign-compare +subdir-ccflags-y += -Wno-missing-field-initializers +subdir-ccflags-y += -Wno-override-init +subdir-ccflags-$(CONFIG_DRM_AMDGPU_WERROR) += -Werror + amdgpu-y := amdgpu_drv.o # add KMS driver @@ -60,7 +80,7 @@ amdgpu-y += amdgpu_device.o amdgpu_kms.o \ amdgpu_umc.o smu_v11_0_i2c.o amdgpu_fru_eeprom.o amdgpu_rap.o \ amdgpu_fw_attestation.o amdgpu_securedisplay.o \ amdgpu_eeprom.o amdgpu_mca.o amdgpu_psp_ta.o amdgpu_lsdma.o \ - amdgpu_ring_mux.o + amdgpu_ring_mux.o amdgpu_xcp.o amdgpu-$(CONFIG_PROC_FS) += amdgpu_fdinfo.o @@ -78,7 +98,7 @@ amdgpu-y += \ vega20_reg_init.o nbio_v7_4.o nbio_v2_3.o nv.o arct_reg_init.o mxgpu_nv.o \ nbio_v7_2.o hdp_v4_0.o hdp_v5_0.o aldebaran_reg_init.o aldebaran.o soc21.o \ sienna_cichlid.o smu_v13_0_10.o nbio_v4_3.o hdp_v6_0.o nbio_v7_7.o hdp_v5_2.o lsdma_v6_0.o \ - nbio_v7_9.o + nbio_v7_9.o aqua_vanjaram_reg_init.o # add DF block amdgpu-y += \ @@ -183,12 +203,14 @@ amdgpu-y += \ vcn_v2_5.o \ vcn_v3_0.o \ vcn_v4_0.o \ + vcn_v4_0_3.o \ amdgpu_jpeg.o \ jpeg_v1_0.o \ jpeg_v2_0.o \ jpeg_v2_5.o \ jpeg_v3_0.o \ - jpeg_v4_0.o + jpeg_v4_0.o \ + jpeg_v4_0_3.o # add ATHUB block amdgpu-y += \ @@ -203,6 +225,7 @@ amdgpu-y += \ smuio_v11_0.o \ smuio_v11_0_6.o \ smuio_v13_0.o \ + smuio_v13_0_3.o \ smuio_v13_0_6.o # add reset block @@ -228,6 +251,7 @@ amdgpu-y += \ amdgpu_amdkfd_gfx_v9.o \ amdgpu_amdkfd_arcturus.o \ amdgpu_amdkfd_aldebaran.o \ + amdgpu_amdkfd_gc_9_4_3.o \ amdgpu_amdkfd_gfx_v10.o \ amdgpu_amdkfd_gfx_v10_3.o \ amdgpu_amdkfd_gfx_v11.o diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 02b827785e3998f7fa67e362cf4a87275115707f..a84bd4a0c42155b46ebbd0278e2588c6d0c4a08f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -107,8 +107,9 @@ #include "amdgpu_fdinfo.h" #include "amdgpu_mca.h" #include "amdgpu_ras.h" +#include "amdgpu_xcp.h" -#define MAX_GPU_INSTANCE 16 +#define MAX_GPU_INSTANCE 64 struct amdgpu_gpu_instance { @@ -212,6 +213,8 @@ extern int amdgpu_noretry; extern int amdgpu_force_asic_type; extern int amdgpu_smartshift_bias; extern int amdgpu_use_xgmi_p2p; +extern int amdgpu_mtype_local; +extern bool enforce_isolation; #ifdef CONFIG_HSA_AMD extern int sched_policy; extern bool debug_evictions; @@ -242,9 +245,10 @@ extern int amdgpu_num_kcq; extern int amdgpu_vcnfw_log; extern int amdgpu_sg_display; +extern int amdgpu_user_partt_mode; + #define AMDGPU_VM_MAX_NUM_CTX 4096 #define AMDGPU_SG_THRESHOLD (256*1024*1024) -#define AMDGPU_DEFAULT_GTT_SIZE_MB 3072ULL /* 3GB by default */ #define AMDGPU_WAIT_IDLE_TIMEOUT_IN_MS 3000 #define AMDGPU_MAX_USEC_TIMEOUT 100000 /* 100 ms */ #define AMDGPU_FENCE_JIFFIES_TIMEOUT (HZ / 2) @@ -282,6 +286,7 @@ extern int amdgpu_sg_display; #define AMDGPU_SMARTSHIFT_MAX_BIAS (100) #define AMDGPU_SMARTSHIFT_MIN_BIAS (-100) +struct amdgpu_xcp_mgr; struct amdgpu_device; struct amdgpu_irq_src; struct amdgpu_fpriv; @@ -463,6 +468,8 @@ struct amdgpu_fpriv { struct mutex bo_list_lock; struct idr bo_list_handles; struct amdgpu_ctx_mgr ctx_mgr; + /** GPU partition selection */ + uint32_t xcp_id; }; int amdgpu_file_to_fpriv(struct file *filp, struct amdgpu_fpriv **fpriv); @@ -573,6 +580,8 @@ struct amdgpu_asic_funcs { /* query video codecs */ int (*query_video_codecs)(struct amdgpu_device *adev, bool encode, const struct amdgpu_video_codecs **codecs); + /* encode "> 32bits" smn addressing */ + u64 (*encode_ext_smn_addressing)(int ext_id); }; /* @@ -607,6 +616,9 @@ void amdgpu_cgs_destroy_device(struct cgs_device *cgs_device); typedef uint32_t (*amdgpu_rreg_t)(struct amdgpu_device*, uint32_t); typedef void (*amdgpu_wreg_t)(struct amdgpu_device*, uint32_t, uint32_t); +typedef uint32_t (*amdgpu_rreg_ext_t)(struct amdgpu_device*, uint64_t); +typedef void (*amdgpu_wreg_ext_t)(struct amdgpu_device*, uint64_t, uint32_t); + typedef uint64_t (*amdgpu_rreg64_t)(struct amdgpu_device*, uint32_t); typedef void (*amdgpu_wreg64_t)(struct amdgpu_device*, uint32_t, uint64_t); @@ -657,7 +669,7 @@ enum amd_hw_ip_block_type { MAX_HWIP }; -#define HWIP_MAX_INSTANCE 28 +#define HWIP_MAX_INSTANCE 44 #define HW_ID_MAX 300 #define IP_VERSION(mj, mn, rv) (((mj) << 16) | ((mn) << 8) | (rv)) @@ -665,6 +677,17 @@ enum amd_hw_ip_block_type { #define IP_VERSION_MIN(ver) (((ver) >> 8) & 0xFF) #define IP_VERSION_REV(ver) ((ver) & 0xFF) +struct amdgpu_ip_map_info { + /* Map of logical to actual dev instances/mask */ + uint32_t dev_inst[MAX_HWIP][HWIP_MAX_INSTANCE]; + int8_t (*logical_to_dev_inst)(struct amdgpu_device *adev, + enum amd_hw_ip_block_type block, + int8_t inst); + uint32_t (*logical_to_dev_mask)(struct amdgpu_device *adev, + enum amd_hw_ip_block_type block, + uint32_t mask); +}; + struct amd_powerplay { void *pp_handle; const struct amd_pm_funcs *pp_funcs; @@ -750,6 +773,7 @@ struct amdgpu_device { struct amdgpu_acp acp; #endif struct amdgpu_hive_info *hive; + struct amdgpu_xcp_mgr *xcp_mgr; /* ASIC */ enum amd_asic_type asic_type; uint32_t family; @@ -797,6 +821,8 @@ struct amdgpu_device { amdgpu_wreg_t pcie_wreg; amdgpu_rreg_t pciep_rreg; amdgpu_wreg_t pciep_wreg; + amdgpu_rreg_ext_t pcie_rreg_ext; + amdgpu_wreg_ext_t pcie_wreg_ext; amdgpu_rreg64_t pcie_rreg64; amdgpu_wreg64_t pcie_wreg64; /* protects concurrent UVD register access */ @@ -830,7 +856,7 @@ struct amdgpu_device { dma_addr_t dummy_page_addr; struct amdgpu_vm_manager vm_manager; struct amdgpu_vmhub vmhub[AMDGPU_MAX_VMHUBS]; - unsigned num_vmhubs; + DECLARE_BITMAP(vmhubs_mask, AMDGPU_MAX_VMHUBS); /* memory management */ struct amdgpu_mman mman; @@ -962,6 +988,7 @@ struct amdgpu_device { /* soc15 register offset based on ip, instance and segment */ uint32_t *reg_offset[MAX_HWIP][HWIP_MAX_INSTANCE]; + struct amdgpu_ip_map_info ip_map; /* delayed work_func for deferring clockgating during resume */ struct delayed_work delayed_init_work; @@ -1020,6 +1047,9 @@ struct amdgpu_device { struct pci_saved_state *pci_state; pci_channel_state_t pci_channel_state; + /* Track auto wait count on s_barrier settings */ + bool barrier_has_auto_waitcnt; + struct amdgpu_reset_control *reset_cntl; uint32_t ip_versions[MAX_HWIP][HWIP_MAX_INSTANCE]; @@ -1050,6 +1080,8 @@ struct amdgpu_device { bool job_hang; bool dc_enabled; + /* Mask of active clusters */ + uint32_t aid_mask; }; static inline struct amdgpu_device *drm_to_adev(struct drm_device *ddev) @@ -1081,11 +1113,18 @@ size_t amdgpu_device_aper_access(struct amdgpu_device *adev, loff_t pos, void amdgpu_device_vram_access(struct amdgpu_device *adev, loff_t pos, void *buf, size_t size, bool write); +uint32_t amdgpu_device_wait_on_rreg(struct amdgpu_device *adev, + uint32_t inst, uint32_t reg_addr, char reg_name[], + uint32_t expected_value, uint32_t mask); uint32_t amdgpu_device_rreg(struct amdgpu_device *adev, uint32_t reg, uint32_t acc_flags); +u32 amdgpu_device_indirect_rreg_ext(struct amdgpu_device *adev, + u64 reg_addr); void amdgpu_device_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint32_t acc_flags); +void amdgpu_device_indirect_wreg_ext(struct amdgpu_device *adev, + u64 reg_addr, u32 reg_data); void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev, uint32_t reg, uint32_t v); void amdgpu_mm_wreg8(struct amdgpu_device *adev, uint32_t offset, uint8_t value); @@ -1137,6 +1176,8 @@ int emu_soc_asic_init(struct amdgpu_device *adev); #define WREG32_PCIE(reg, v) adev->pcie_wreg(adev, (reg), (v)) #define RREG32_PCIE_PORT(reg) adev->pciep_rreg(adev, (reg)) #define WREG32_PCIE_PORT(reg, v) adev->pciep_wreg(adev, (reg), (v)) +#define RREG32_PCIE_EXT(reg) adev->pcie_rreg_ext(adev, (reg)) +#define WREG32_PCIE_EXT(reg, v) adev->pcie_wreg_ext(adev, (reg), (v)) #define RREG64_PCIE(reg) adev->pcie_rreg64(adev, (reg)) #define WREG64_PCIE(reg, v) adev->pcie_wreg64(adev, (reg), (v)) #define RREG32_SMC(reg) adev->smc_rreg(adev, (reg)) @@ -1204,7 +1245,8 @@ int emu_soc_asic_init(struct amdgpu_device *adev); /* * ASICs macro. */ -#define amdgpu_asic_set_vga_state(adev, state) (adev)->asic_funcs->set_vga_state((adev), (state)) +#define amdgpu_asic_set_vga_state(adev, state) \ + ((adev)->asic_funcs->set_vga_state ? (adev)->asic_funcs->set_vga_state((adev), (state)) : 0) #define amdgpu_asic_reset(adev) (adev)->asic_funcs->reset((adev)) #define amdgpu_asic_reset_method(adev) (adev)->asic_funcs->reset_method((adev)) #define amdgpu_asic_get_xclk(adev) (adev)->asic_funcs->get_xclk((adev)) @@ -1235,6 +1277,10 @@ int emu_soc_asic_init(struct amdgpu_device *adev); #define amdgpu_inc_vram_lost(adev) atomic_inc(&((adev)->vram_lost_counter)); +#define for_each_inst(i, inst_mask) \ + for (i = ffs(inst_mask) - 1; inst_mask; \ + inst_mask &= ~(1U << i), i = ffs(inst_mask) - 1) + #define MIN(X, Y) ((X) < (Y) ? (X) : (Y)) /* Common functions */ @@ -1348,6 +1394,12 @@ struct amdgpu_afmt_acr amdgpu_afmt_acr(uint32_t clock); /* amdgpu_acpi.c */ +struct amdgpu_numa_info { + uint64_t size; + int pxm; + int nid; +}; + /* ATCS Device/Driver State */ #define AMDGPU_ATCS_PSC_DEV_STATE_D0 0 #define AMDGPU_ATCS_PSC_DEV_STATE_D3_HOT 3 @@ -1365,15 +1417,32 @@ int amdgpu_acpi_power_shift_control(struct amdgpu_device *adev, u8 dev_state, bool drv_state); int amdgpu_acpi_smart_shift_update(struct drm_device *dev, enum amdgpu_ss ss_state); int amdgpu_acpi_pcie_notify_device_ready(struct amdgpu_device *adev); +int amdgpu_acpi_get_tmr_info(struct amdgpu_device *adev, u64 *tmr_offset, + u64 *tmr_size); +int amdgpu_acpi_get_mem_info(struct amdgpu_device *adev, int xcc_id, + struct amdgpu_numa_info *numa_info); void amdgpu_acpi_get_backlight_caps(struct amdgpu_dm_backlight_caps *caps); bool amdgpu_acpi_should_gpu_reset(struct amdgpu_device *adev); void amdgpu_acpi_detect(void); +void amdgpu_acpi_release(void); #else static inline int amdgpu_acpi_init(struct amdgpu_device *adev) { return 0; } +static inline int amdgpu_acpi_get_tmr_info(struct amdgpu_device *adev, + u64 *tmr_offset, u64 *tmr_size) +{ + return -EINVAL; +} +static inline int amdgpu_acpi_get_mem_info(struct amdgpu_device *adev, + int xcc_id, + struct amdgpu_numa_info *numa_info) +{ + return -EINVAL; +} static inline void amdgpu_acpi_fini(struct amdgpu_device *adev) { } static inline bool amdgpu_acpi_should_gpu_reset(struct amdgpu_device *adev) { return false; } static inline void amdgpu_acpi_detect(void) { } +static inline void amdgpu_acpi_release(void) { } static inline bool amdgpu_acpi_is_power_shift_control_supported(void) { return false; } static inline int amdgpu_acpi_power_shift_control(struct amdgpu_device *adev, u8 dev_state, bool drv_state) { return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c index aeeec211861c434bae58aae748586e4a608360d5..385c6acb5728b7406f6211b3dab9166f1e48914a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include #include @@ -38,6 +39,45 @@ #include "amd_acpi.h" #include "atom.h" +/* Declare GUID for AMD _DSM method for XCCs */ +static const guid_t amd_xcc_dsm_guid = GUID_INIT(0x8267f5d5, 0xa556, 0x44f2, + 0xb8, 0xb4, 0x45, 0x56, 0x2e, + 0x8c, 0x5b, 0xec); + +#define AMD_XCC_HID_START 3000 +#define AMD_XCC_DSM_GET_NUM_FUNCS 0 +#define AMD_XCC_DSM_GET_SUPP_MODE 1 +#define AMD_XCC_DSM_GET_XCP_MODE 2 +#define AMD_XCC_DSM_GET_VF_XCC_MAPPING 4 +#define AMD_XCC_DSM_GET_TMR_INFO 5 +#define AMD_XCC_DSM_NUM_FUNCS 5 + +#define AMD_XCC_MAX_HID 24 + +struct xarray numa_info_xa; + +/* Encapsulates the XCD acpi object information */ +struct amdgpu_acpi_xcc_info { + struct list_head list; + struct amdgpu_numa_info *numa_info; + uint8_t xcp_node; + uint8_t phy_id; + acpi_handle handle; +}; + +struct amdgpu_acpi_dev_info { + struct list_head list; + struct list_head xcc_list; + uint16_t bdf; + uint16_t supp_xcp_mode; + uint16_t xcp_mode; + uint16_t mem_mode; + uint64_t tmr_base; + uint64_t tmr_size; +}; + +struct list_head amdgpu_acpi_dev_list; + struct amdgpu_atif_notification_cfg { bool enabled; int command_code; @@ -801,6 +841,343 @@ int amdgpu_acpi_smart_shift_update(struct drm_device *dev, enum amdgpu_ss ss_sta return r; } +#ifdef CONFIG_ACPI_NUMA +static inline uint64_t amdgpu_acpi_get_numa_size(int nid) +{ + /* This is directly using si_meminfo_node implementation as the + * function is not exported. + */ + int zone_type; + uint64_t managed_pages = 0; + + pg_data_t *pgdat = NODE_DATA(nid); + + for (zone_type = 0; zone_type < MAX_NR_ZONES; zone_type++) + managed_pages += + zone_managed_pages(&pgdat->node_zones[zone_type]); + return managed_pages * PAGE_SIZE; +} + +static struct amdgpu_numa_info *amdgpu_acpi_get_numa_info(uint32_t pxm) +{ + struct amdgpu_numa_info *numa_info; + int nid; + + numa_info = xa_load(&numa_info_xa, pxm); + + if (!numa_info) { + struct sysinfo info; + + numa_info = kzalloc(sizeof *numa_info, GFP_KERNEL); + if (!numa_info) + return NULL; + + nid = pxm_to_node(pxm); + numa_info->pxm = pxm; + numa_info->nid = nid; + + if (numa_info->nid == NUMA_NO_NODE) { + si_meminfo(&info); + numa_info->size = info.totalram * info.mem_unit; + } else { + numa_info->size = amdgpu_acpi_get_numa_size(nid); + } + xa_store(&numa_info_xa, numa_info->pxm, numa_info, GFP_KERNEL); + } + + return numa_info; +} +#endif + +/** + * amdgpu_acpi_get_node_id - obtain the NUMA node id for corresponding amdgpu + * acpi device handle + * + * @handle: acpi handle + * @numa_info: amdgpu_numa_info structure holding numa information + * + * Queries the ACPI interface to fetch the corresponding NUMA Node ID for a + * given amdgpu acpi device. + * + * Returns ACPI STATUS OK with Node ID on success or the corresponding failure reason + */ +static acpi_status amdgpu_acpi_get_node_id(acpi_handle handle, + struct amdgpu_numa_info **numa_info) +{ +#ifdef CONFIG_ACPI_NUMA + u64 pxm; + acpi_status status; + + if (!numa_info) + return_ACPI_STATUS(AE_ERROR); + + status = acpi_evaluate_integer(handle, "_PXM", NULL, &pxm); + + if (ACPI_FAILURE(status)) + return status; + + *numa_info = amdgpu_acpi_get_numa_info(pxm); + + if (!*numa_info) + return_ACPI_STATUS(AE_ERROR); + + return_ACPI_STATUS(AE_OK); +#else + return_ACPI_STATUS(AE_NOT_EXIST); +#endif +} + +static struct amdgpu_acpi_dev_info *amdgpu_acpi_get_dev(u16 bdf) +{ + struct amdgpu_acpi_dev_info *acpi_dev; + + if (list_empty(&amdgpu_acpi_dev_list)) + return NULL; + + list_for_each_entry(acpi_dev, &amdgpu_acpi_dev_list, list) + if (acpi_dev->bdf == bdf) + return acpi_dev; + + return NULL; +} + +static int amdgpu_acpi_dev_init(struct amdgpu_acpi_dev_info **dev_info, + struct amdgpu_acpi_xcc_info *xcc_info, u16 bdf) +{ + struct amdgpu_acpi_dev_info *tmp; + union acpi_object *obj; + int ret = -ENOENT; + + *dev_info = NULL; + tmp = kzalloc(sizeof(struct amdgpu_acpi_dev_info), GFP_KERNEL); + if (!tmp) + return -ENOMEM; + + INIT_LIST_HEAD(&tmp->xcc_list); + INIT_LIST_HEAD(&tmp->list); + tmp->bdf = bdf; + + obj = acpi_evaluate_dsm_typed(xcc_info->handle, &amd_xcc_dsm_guid, 0, + AMD_XCC_DSM_GET_SUPP_MODE, NULL, + ACPI_TYPE_INTEGER); + + if (!obj) { + acpi_handle_debug(xcc_info->handle, + "_DSM function %d evaluation failed", + AMD_XCC_DSM_GET_SUPP_MODE); + ret = -ENOENT; + goto out; + } + + tmp->supp_xcp_mode = obj->integer.value & 0xFFFF; + ACPI_FREE(obj); + + obj = acpi_evaluate_dsm_typed(xcc_info->handle, &amd_xcc_dsm_guid, 0, + AMD_XCC_DSM_GET_XCP_MODE, NULL, + ACPI_TYPE_INTEGER); + + if (!obj) { + acpi_handle_debug(xcc_info->handle, + "_DSM function %d evaluation failed", + AMD_XCC_DSM_GET_XCP_MODE); + ret = -ENOENT; + goto out; + } + + tmp->xcp_mode = obj->integer.value & 0xFFFF; + tmp->mem_mode = (obj->integer.value >> 32) & 0xFFFF; + ACPI_FREE(obj); + + /* Evaluate DSMs and fill XCC information */ + obj = acpi_evaluate_dsm_typed(xcc_info->handle, &amd_xcc_dsm_guid, 0, + AMD_XCC_DSM_GET_TMR_INFO, NULL, + ACPI_TYPE_PACKAGE); + + if (!obj || obj->package.count < 2) { + acpi_handle_debug(xcc_info->handle, + "_DSM function %d evaluation failed", + AMD_XCC_DSM_GET_TMR_INFO); + ret = -ENOENT; + goto out; + } + + tmp->tmr_base = obj->package.elements[0].integer.value; + tmp->tmr_size = obj->package.elements[1].integer.value; + ACPI_FREE(obj); + + DRM_DEBUG_DRIVER( + "New dev(%x): Supported xcp mode: %x curr xcp_mode : %x mem mode : %x, tmr base: %llx tmr size: %llx ", + tmp->bdf, tmp->supp_xcp_mode, tmp->xcp_mode, tmp->mem_mode, + tmp->tmr_base, tmp->tmr_size); + list_add_tail(&tmp->list, &amdgpu_acpi_dev_list); + *dev_info = tmp; + + return 0; + +out: + if (obj) + ACPI_FREE(obj); + kfree(tmp); + + return ret; +} + +static int amdgpu_acpi_get_xcc_info(struct amdgpu_acpi_xcc_info *xcc_info, + u16 *bdf) +{ + union acpi_object *obj; + acpi_status status; + int ret = -ENOENT; + + obj = acpi_evaluate_dsm_typed(xcc_info->handle, &amd_xcc_dsm_guid, 0, + AMD_XCC_DSM_GET_NUM_FUNCS, NULL, + ACPI_TYPE_INTEGER); + + if (!obj || obj->integer.value != AMD_XCC_DSM_NUM_FUNCS) + goto out; + ACPI_FREE(obj); + + /* Evaluate DSMs and fill XCC information */ + obj = acpi_evaluate_dsm_typed(xcc_info->handle, &amd_xcc_dsm_guid, 0, + AMD_XCC_DSM_GET_VF_XCC_MAPPING, NULL, + ACPI_TYPE_INTEGER); + + if (!obj) { + acpi_handle_debug(xcc_info->handle, + "_DSM function %d evaluation failed", + AMD_XCC_DSM_GET_VF_XCC_MAPPING); + ret = -EINVAL; + goto out; + } + + /* PF xcc id [39:32] */ + xcc_info->phy_id = (obj->integer.value >> 32) & 0xFF; + /* xcp node of this xcc [47:40] */ + xcc_info->xcp_node = (obj->integer.value >> 40) & 0xFF; + /* PF bus/dev/fn of this xcc [63:48] */ + *bdf = (obj->integer.value >> 48) & 0xFFFF; + ACPI_FREE(obj); + obj = NULL; + + status = + amdgpu_acpi_get_node_id(xcc_info->handle, &xcc_info->numa_info); + + /* TODO: check if this check is required */ + if (ACPI_SUCCESS(status)) + ret = 0; +out: + if (obj) + ACPI_FREE(obj); + + return ret; +} + +static int amdgpu_acpi_enumerate_xcc(void) +{ + struct amdgpu_acpi_dev_info *dev_info = NULL; + struct amdgpu_acpi_xcc_info *xcc_info; + struct acpi_device *acpi_dev; + char hid[ACPI_ID_LEN]; + int ret, id; + u16 bdf; + + INIT_LIST_HEAD(&amdgpu_acpi_dev_list); + xa_init(&numa_info_xa); + + for (id = 0; id < AMD_XCC_MAX_HID; id++) { + sprintf(hid, "%s%d", "AMD", AMD_XCC_HID_START + id); + acpi_dev = acpi_dev_get_first_match_dev(hid, NULL, -1); + /* These ACPI objects are expected to be in sequential order. If + * one is not found, no need to check the rest. + */ + if (!acpi_dev) { + DRM_DEBUG_DRIVER("No matching acpi device found for %s", + hid); + break; + } + + xcc_info = kzalloc(sizeof(struct amdgpu_acpi_xcc_info), + GFP_KERNEL); + if (!xcc_info) { + DRM_ERROR("Failed to allocate memory for xcc info\n"); + return -ENOMEM; + } + + INIT_LIST_HEAD(&xcc_info->list); + xcc_info->handle = acpi_device_handle(acpi_dev); + acpi_dev_put(acpi_dev); + + ret = amdgpu_acpi_get_xcc_info(xcc_info, &bdf); + if (ret) { + kfree(xcc_info); + continue; + } + + dev_info = amdgpu_acpi_get_dev(bdf); + + if (!dev_info) + ret = amdgpu_acpi_dev_init(&dev_info, xcc_info, bdf); + + if (ret == -ENOMEM) + return ret; + + if (!dev_info) { + kfree(xcc_info); + continue; + } + + list_add_tail(&xcc_info->list, &dev_info->xcc_list); + } + + return 0; +} + +int amdgpu_acpi_get_tmr_info(struct amdgpu_device *adev, u64 *tmr_offset, + u64 *tmr_size) +{ + struct amdgpu_acpi_dev_info *dev_info; + u16 bdf; + + if (!tmr_offset || !tmr_size) + return -EINVAL; + + bdf = (adev->pdev->bus->number << 8) | adev->pdev->devfn; + dev_info = amdgpu_acpi_get_dev(bdf); + if (!dev_info) + return -ENOENT; + + *tmr_offset = dev_info->tmr_base; + *tmr_size = dev_info->tmr_size; + + return 0; +} + +int amdgpu_acpi_get_mem_info(struct amdgpu_device *adev, int xcc_id, + struct amdgpu_numa_info *numa_info) +{ + struct amdgpu_acpi_dev_info *dev_info; + struct amdgpu_acpi_xcc_info *xcc_info; + u16 bdf; + + if (!numa_info) + return -EINVAL; + + bdf = (adev->pdev->bus->number << 8) | adev->pdev->devfn; + dev_info = amdgpu_acpi_get_dev(bdf); + if (!dev_info) + return -ENOENT; + + list_for_each_entry(xcc_info, &dev_info->xcc_list, list) { + if (xcc_info->phy_id == xcc_id) { + memcpy(numa_info, xcc_info->numa_info, + sizeof(*numa_info)); + return 0; + } + } + + return -ENOENT; +} + /** * amdgpu_acpi_event - handle notify events * @@ -1054,6 +1431,36 @@ void amdgpu_acpi_detect(void) } else { atif->backlight_caps.caps_valid = false; } + + amdgpu_acpi_enumerate_xcc(); +} + +void amdgpu_acpi_release(void) +{ + struct amdgpu_acpi_dev_info *dev_info, *dev_tmp; + struct amdgpu_acpi_xcc_info *xcc_info, *xcc_tmp; + struct amdgpu_numa_info *numa_info; + unsigned long index; + + xa_for_each(&numa_info_xa, index, numa_info) { + kfree(numa_info); + xa_erase(&numa_info_xa, index); + } + + if (list_empty(&amdgpu_acpi_dev_list)) + return; + + list_for_each_entry_safe(dev_info, dev_tmp, &amdgpu_acpi_dev_list, + list) { + list_for_each_entry_safe(xcc_info, xcc_tmp, &dev_info->xcc_list, + list) { + list_del(&xcc_info->list); + kfree(xcc_info); + } + + list_del(&dev_info->list); + kfree(dev_info); + } } #if IS_ENABLED(CONFIG_SUSPEND) @@ -1092,16 +1499,20 @@ bool amdgpu_acpi_is_s0ix_active(struct amdgpu_device *adev) * S0ix even though the system is suspending to idle, so return false * in that case. */ - if (!(acpi_gbl_FADT.flags & ACPI_FADT_LOW_POWER_S0)) - dev_warn_once(adev->dev, + if (!(acpi_gbl_FADT.flags & ACPI_FADT_LOW_POWER_S0)) { + dev_err_once(adev->dev, "Power consumption will be higher as BIOS has not been configured for suspend-to-idle.\n" "To use suspend-to-idle change the sleep mode in BIOS setup.\n"); + return false; + } #if !IS_ENABLED(CONFIG_AMD_PMC) - dev_warn_once(adev->dev, + dev_err_once(adev->dev, "Power consumption will be higher as the kernel has not been compiled with CONFIG_AMD_PMC.\n"); -#endif /* CONFIG_AMD_PMC */ + return false; +#else return true; +#endif /* CONFIG_AMD_PMC */ } #endif /* CONFIG_SUSPEND */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index 0385f7f692785efae9a7812903f63eab467f3aa4..b4fcad0e62f7ec23b22a1f27b3a8b352a3e0b1aa 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -53,7 +53,6 @@ int amdgpu_amdkfd_init(void) amdgpu_amdkfd_total_mem_size *= si.mem_unit; ret = kgd2kfd_init(); - amdgpu_amdkfd_gpuvm_init_mem_limits(); kfd_initialized = !ret; return ret; @@ -143,6 +142,8 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev) int i; int last_valid_bit; + amdgpu_amdkfd_gpuvm_init_mem_limits(); + if (adev->kfd.dev) { struct kgd2kfd_shared_resources gpu_resources = { .compute_vmid_bitmap = @@ -162,7 +163,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev) * clear */ bitmap_complement(gpu_resources.cp_queue_bitmap, - adev->gfx.mec.queue_bitmap, + adev->gfx.mec_bitmap[0].queue_bitmap, KGD_MAX_QUEUES); /* According to linux/bitmap.h we shouldn't use bitmap_clear if @@ -427,14 +428,23 @@ uint32_t amdgpu_amdkfd_get_fw_version(struct amdgpu_device *adev, } void amdgpu_amdkfd_get_local_mem_info(struct amdgpu_device *adev, - struct kfd_local_mem_info *mem_info) + struct kfd_local_mem_info *mem_info, + struct amdgpu_xcp *xcp) { memset(mem_info, 0, sizeof(*mem_info)); - mem_info->local_mem_size_public = adev->gmc.visible_vram_size; - mem_info->local_mem_size_private = adev->gmc.real_vram_size - + if (xcp) { + if (adev->gmc.real_vram_size == adev->gmc.visible_vram_size) + mem_info->local_mem_size_public = + KFD_XCP_MEMORY_SIZE(adev, xcp->id); + else + mem_info->local_mem_size_private = + KFD_XCP_MEMORY_SIZE(adev, xcp->id); + } else { + mem_info->local_mem_size_public = adev->gmc.visible_vram_size; + mem_info->local_mem_size_private = adev->gmc.real_vram_size - adev->gmc.visible_vram_size; - + } mem_info->vram_width = adev->gmc.vram_width; pr_debug("Address base: %pap public 0x%llx private 0x%llx\n", @@ -497,7 +507,7 @@ int amdgpu_amdkfd_get_dmabuf_info(struct amdgpu_device *adev, int dma_buf_fd, struct amdgpu_device **dmabuf_adev, uint64_t *bo_size, void *metadata_buffer, size_t buffer_size, uint32_t *metadata_size, - uint32_t *flags) + uint32_t *flags, int8_t *xcp_id) { struct dma_buf *dma_buf; struct drm_gem_object *obj; @@ -541,6 +551,8 @@ int amdgpu_amdkfd_get_dmabuf_info(struct amdgpu_device *adev, int dma_buf_fd, if (bo->flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) *flags |= KFD_IOC_ALLOC_MEM_FLAGS_PUBLIC; } + if (xcp_id) + *xcp_id = bo->xcp_id; out_put: dma_buf_put(dma_buf); @@ -732,17 +744,19 @@ int amdgpu_amdkfd_flush_gpu_tlb_vmid(struct amdgpu_device *adev, if (adev->family == AMDGPU_FAMILY_AI) { int i; - for (i = 0; i < adev->num_vmhubs; i++) + for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) amdgpu_gmc_flush_gpu_tlb(adev, vmid, i, 0); } else { - amdgpu_gmc_flush_gpu_tlb(adev, vmid, AMDGPU_GFXHUB_0, 0); + amdgpu_gmc_flush_gpu_tlb(adev, vmid, AMDGPU_GFXHUB(0), 0); } return 0; } int amdgpu_amdkfd_flush_gpu_tlb_pasid(struct amdgpu_device *adev, - uint16_t pasid, enum TLB_FLUSH_TYPE flush_type) + uint16_t pasid, + enum TLB_FLUSH_TYPE flush_type, + uint32_t inst) { bool all_hub = false; @@ -750,7 +764,7 @@ int amdgpu_amdkfd_flush_gpu_tlb_pasid(struct amdgpu_device *adev, adev->family == AMDGPU_FAMILY_RV) all_hub = true; - return amdgpu_gmc_flush_gpu_tlb_pasid(adev, pasid, flush_type, all_hub); + return amdgpu_gmc_flush_gpu_tlb_pasid(adev, pasid, flush_type, all_hub, inst); } bool amdgpu_amdkfd_have_atomics_support(struct amdgpu_device *adev) @@ -758,11 +772,32 @@ bool amdgpu_amdkfd_have_atomics_support(struct amdgpu_device *adev) return adev->have_atomics_support; } +void amdgpu_amdkfd_debug_mem_fence(struct amdgpu_device *adev) +{ + amdgpu_device_flush_hdp(adev, NULL); +} + void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev, bool reset) { amdgpu_umc_poison_handler(adev, reset); } +int amdgpu_amdkfd_send_close_event_drain_irq(struct amdgpu_device *adev, + uint32_t *payload) +{ + int ret; + + /* Device or IH ring is not ready so bail. */ + ret = amdgpu_ih_wait_on_checkpoint_process_ts(adev, &adev->irq.ih); + if (ret) + return ret; + + /* Send payload to fence KFD interrupts */ + amdgpu_amdkfd_interrupt(adev, payload); + + return 0; +} + bool amdgpu_amdkfd_ras_query_utcl2_poison_status(struct amdgpu_device *adev) { if (adev->gfx.ras && adev->gfx.ras->query_utcl2_poison_status) @@ -770,3 +805,28 @@ bool amdgpu_amdkfd_ras_query_utcl2_poison_status(struct amdgpu_device *adev) else return false; } + +int amdgpu_amdkfd_check_and_lock_kfd(struct amdgpu_device *adev) +{ + return kgd2kfd_check_and_lock_kfd(); +} + +void amdgpu_amdkfd_unlock_kfd(struct amdgpu_device *adev) +{ + kgd2kfd_unlock_kfd(); +} + + +u64 amdgpu_amdkfd_xcp_memory_size(struct amdgpu_device *adev, int xcp_id) +{ + u64 tmp; + s8 mem_id = KFD_XCP_MEM_ID(adev, xcp_id); + + if (adev->gmc.num_mem_partitions && xcp_id >= 0 && mem_id >= 0) { + tmp = adev->gmc.mem_partitions[mem_id].size; + do_div(tmp, adev->xcp_mgr->num_xcp_per_mem_partition); + return ALIGN_DOWN(tmp, PAGE_SIZE); + } else { + return adev->gmc.real_vram_size; + } +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index 01ba3589b60a0d6d643c42a0b32db517a2679eaa..2d0406bff84ecbeca3651d7e74fa3d588cfd161c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -30,10 +30,12 @@ #include #include #include +#include #include #include #include "amdgpu_sync.h" #include "amdgpu_vm.h" +#include "amdgpu_xcp.h" extern uint64_t amdgpu_amdkfd_total_mem_size; @@ -97,10 +99,13 @@ struct amdgpu_amdkfd_fence { struct amdgpu_kfd_dev { struct kfd_dev *dev; - int64_t vram_used; - uint64_t vram_used_aligned; + int64_t vram_used[MAX_XCP]; + uint64_t vram_used_aligned[MAX_XCP]; bool init_complete; struct work_struct reset_work; + + /* HMM page migration MEMORY_DEVICE_PRIVATE mapping */ + struct dev_pagemap pgmap; }; enum kgd_engine_type { @@ -151,6 +156,8 @@ void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev, void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev); void amdgpu_amdkfd_device_init(struct amdgpu_device *adev); void amdgpu_amdkfd_device_fini_sw(struct amdgpu_device *adev); +int amdgpu_amdkfd_check_and_lock_kfd(struct amdgpu_device *adev); +void amdgpu_amdkfd_unlock_kfd(struct amdgpu_device *adev); int amdgpu_amdkfd_submit_ib(struct amdgpu_device *adev, enum kgd_engine_type engine, uint32_t vmid, uint64_t gpu_addr, @@ -160,7 +167,8 @@ bool amdgpu_amdkfd_have_atomics_support(struct amdgpu_device *adev); int amdgpu_amdkfd_flush_gpu_tlb_vmid(struct amdgpu_device *adev, uint16_t vmid); int amdgpu_amdkfd_flush_gpu_tlb_pasid(struct amdgpu_device *adev, - uint16_t pasid, enum TLB_FLUSH_TYPE flush_type); + uint16_t pasid, enum TLB_FLUSH_TYPE flush_type, + uint32_t inst); bool amdgpu_amdkfd_is_kfd_vmid(struct amdgpu_device *adev, u32 vmid); @@ -224,7 +232,8 @@ int amdgpu_amdkfd_remove_gws_from_process(void *info, void *mem); uint32_t amdgpu_amdkfd_get_fw_version(struct amdgpu_device *adev, enum kgd_engine_type type); void amdgpu_amdkfd_get_local_mem_info(struct amdgpu_device *adev, - struct kfd_local_mem_info *mem_info); + struct kfd_local_mem_info *mem_info, + struct amdgpu_xcp *xcp); uint64_t amdgpu_amdkfd_get_gpu_clock_counter(struct amdgpu_device *adev); uint32_t amdgpu_amdkfd_get_max_engine_clock_in_mhz(struct amdgpu_device *adev); @@ -234,13 +243,15 @@ int amdgpu_amdkfd_get_dmabuf_info(struct amdgpu_device *adev, int dma_buf_fd, struct amdgpu_device **dmabuf_adev, uint64_t *bo_size, void *metadata_buffer, size_t buffer_size, uint32_t *metadata_size, - uint32_t *flags); + uint32_t *flags, int8_t *xcp_id); uint8_t amdgpu_amdkfd_get_xgmi_hops_count(struct amdgpu_device *dst, struct amdgpu_device *src); int amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(struct amdgpu_device *dst, struct amdgpu_device *src, bool is_min); int amdgpu_amdkfd_get_pcie_bandwidth_mbytes(struct amdgpu_device *adev, bool is_min); +int amdgpu_amdkfd_send_close_event_drain_irq(struct amdgpu_device *adev, + uint32_t *payload); /* Read user wptr from a specified user address space with page fault * disabled. The memory must be pinned and mapped to the hardware when @@ -279,7 +290,8 @@ int amdgpu_amdkfd_gpuvm_acquire_process_vm(struct amdgpu_device *adev, void amdgpu_amdkfd_gpuvm_release_process_vm(struct amdgpu_device *adev, void *drm_priv); uint64_t amdgpu_amdkfd_gpuvm_get_process_page_dir(void *drm_priv); -size_t amdgpu_amdkfd_get_available_memory(struct amdgpu_device *adev); +size_t amdgpu_amdkfd_get_available_memory(struct amdgpu_device *adev, + uint8_t xcp_id); int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( struct amdgpu_device *adev, uint64_t va, uint64_t size, void *drm_priv, struct kgd_mem **mem, @@ -310,6 +322,7 @@ int amdgpu_amdkfd_gpuvm_import_dmabuf(struct amdgpu_device *adev, uint64_t *mmap_offset); int amdgpu_amdkfd_gpuvm_export_dmabuf(struct kgd_mem *mem, struct dma_buf **dmabuf); +void amdgpu_amdkfd_debug_mem_fence(struct amdgpu_device *adev); int amdgpu_amdkfd_get_tile_config(struct amdgpu_device *adev, struct tile_config *config); void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev, @@ -319,9 +332,18 @@ void amdgpu_amdkfd_block_mmu_notifications(void *p); int amdgpu_amdkfd_criu_resume(void *p); bool amdgpu_amdkfd_ras_query_utcl2_poison_status(struct amdgpu_device *adev); int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev, - uint64_t size, u32 alloc_flag); + uint64_t size, u32 alloc_flag, int8_t xcp_id); void amdgpu_amdkfd_unreserve_mem_limit(struct amdgpu_device *adev, - uint64_t size, u32 alloc_flag); + uint64_t size, u32 alloc_flag, int8_t xcp_id); + +u64 amdgpu_amdkfd_xcp_memory_size(struct amdgpu_device *adev, int xcp_id); + +#define KFD_XCP_MEM_ID(adev, xcp_id) \ + ((adev)->xcp_mgr && (xcp_id) >= 0 ?\ + (adev)->xcp_mgr->xcp[(xcp_id)].mem_id : -1) + +#define KFD_XCP_MEMORY_SIZE(adev, xcp_id) amdgpu_amdkfd_xcp_memory_size((adev), (xcp_id)) + #if IS_ENABLED(CONFIG_HSA_AMD) void amdgpu_amdkfd_gpuvm_init_mem_limits(void); @@ -352,6 +374,17 @@ void amdgpu_amdkfd_release_notify(struct amdgpu_bo *bo) { } #endif + +#if IS_ENABLED(CONFIG_HSA_AMD_SVM) +int kgd2kfd_init_zone_device(struct amdgpu_device *adev); +#else +static inline +int kgd2kfd_init_zone_device(struct amdgpu_device *adev) +{ + return 0; +} +#endif + /* KGD2KFD callbacks */ int kgd2kfd_quiesce_mm(struct mm_struct *mm, uint32_t trigger); int kgd2kfd_resume_mm(struct mm_struct *mm); @@ -372,6 +405,8 @@ int kgd2kfd_post_reset(struct kfd_dev *kfd); void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry); void kgd2kfd_set_sram_ecc_flag(struct kfd_dev *kfd); void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint64_t throttle_bitmask); +int kgd2kfd_check_and_lock_kfd(void); +void kgd2kfd_unlock_kfd(void); #else static inline int kgd2kfd_init(void) { @@ -437,5 +472,14 @@ static inline void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint64_t throttle_bitmask) { } + +static inline int kgd2kfd_check_and_lock_kfd(void) +{ + return 0; +} + +static inline void kgd2kfd_unlock_kfd(void) +{ +} #endif #endif /* AMDGPU_AMDKFD_H_INCLUDED */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c index 4485bb29bec961f9b9ac33cca23d666ba8fd0f12..60f9e027fb6607bc1e444d0d03b0e9af10e63bee 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c @@ -23,6 +23,149 @@ #include "amdgpu_amdkfd.h" #include "amdgpu_amdkfd_arcturus.h" #include "amdgpu_amdkfd_gfx_v9.h" +#include "gc/gc_9_4_2_offset.h" +#include "gc/gc_9_4_2_sh_mask.h" +#include + +/* + * Returns TRAP_EN, EXCP_EN and EXCP_REPLACE. + * + * restore_dbg_registers is ignored here but is a general interface requirement + * for devices that support GFXOFF and where the RLC save/restore list + * does not support hw registers for debugging i.e. the driver has to manually + * initialize the debug mode registers after it has disabled GFX off during the + * debug session. + */ +static uint32_t kgd_aldebaran_enable_debug_trap(struct amdgpu_device *adev, + bool restore_dbg_registers, + uint32_t vmid) +{ + uint32_t data = 0; + + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, 0); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, 0); + + return data; +} + +/* returns TRAP_EN, EXCP_EN and EXCP_REPLACE. */ +static uint32_t kgd_aldebaran_disable_debug_trap(struct amdgpu_device *adev, + bool keep_trap_enabled, + uint32_t vmid) +{ + uint32_t data = 0; + + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, keep_trap_enabled); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, 0); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, 0); + + return data; +} + +static int kgd_aldebaran_validate_trap_override_request(struct amdgpu_device *adev, + uint32_t trap_override, + uint32_t *trap_mask_supported) +{ + *trap_mask_supported &= KFD_DBG_TRAP_MASK_FP_INVALID | + KFD_DBG_TRAP_MASK_FP_INPUT_DENORMAL | + KFD_DBG_TRAP_MASK_FP_DIVIDE_BY_ZERO | + KFD_DBG_TRAP_MASK_FP_OVERFLOW | + KFD_DBG_TRAP_MASK_FP_UNDERFLOW | + KFD_DBG_TRAP_MASK_FP_INEXACT | + KFD_DBG_TRAP_MASK_INT_DIVIDE_BY_ZERO | + KFD_DBG_TRAP_MASK_DBG_ADDRESS_WATCH | + KFD_DBG_TRAP_MASK_DBG_MEMORY_VIOLATION; + + if (trap_override != KFD_DBG_TRAP_OVERRIDE_OR && + trap_override != KFD_DBG_TRAP_OVERRIDE_REPLACE) + return -EPERM; + + return 0; +} + +/* returns TRAP_EN, EXCP_EN and EXCP_RPLACE. */ +static uint32_t kgd_aldebaran_set_wave_launch_trap_override(struct amdgpu_device *adev, + uint32_t vmid, + uint32_t trap_override, + uint32_t trap_mask_bits, + uint32_t trap_mask_request, + uint32_t *trap_mask_prev, + uint32_t kfd_dbg_trap_cntl_prev) + +{ + uint32_t data = 0; + + *trap_mask_prev = REG_GET_FIELD(kfd_dbg_trap_cntl_prev, SPI_GDBG_PER_VMID_CNTL, EXCP_EN); + trap_mask_bits = (trap_mask_bits & trap_mask_request) | + (*trap_mask_prev & ~trap_mask_request); + + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, trap_mask_bits); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, trap_override); + + return data; +} + +static uint32_t kgd_aldebaran_set_wave_launch_mode(struct amdgpu_device *adev, + uint8_t wave_launch_mode, + uint32_t vmid) +{ + uint32_t data = 0; + + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, LAUNCH_MODE, wave_launch_mode); + + return data; +} + +#define TCP_WATCH_STRIDE (regTCP_WATCH1_ADDR_H - regTCP_WATCH0_ADDR_H) +static uint32_t kgd_gfx_aldebaran_set_address_watch( + struct amdgpu_device *adev, + uint64_t watch_address, + uint32_t watch_address_mask, + uint32_t watch_id, + uint32_t watch_mode, + uint32_t debug_vmid) +{ + uint32_t watch_address_high; + uint32_t watch_address_low; + uint32_t watch_address_cntl; + + watch_address_cntl = 0; + watch_address_low = lower_32_bits(watch_address); + watch_address_high = upper_32_bits(watch_address) & 0xffff; + + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + MODE, + watch_mode); + + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + MASK, + watch_address_mask >> 6); + + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + VALID, + 1); + + WREG32_RLC((SOC15_REG_OFFSET(GC, 0, regTCP_WATCH0_ADDR_H) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_high); + + WREG32_RLC((SOC15_REG_OFFSET(GC, 0, regTCP_WATCH0_ADDR_L) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_low); + + return watch_address_cntl; +} + +static uint32_t kgd_gfx_aldebaran_clear_address_watch(struct amdgpu_device *adev, + uint32_t watch_id) +{ + return 0; +} const struct kfd2kgd_calls aldebaran_kfd2kgd = { .program_sh_mem_settings = kgd_gfx_v9_program_sh_mem_settings, @@ -42,5 +185,14 @@ const struct kfd2kgd_calls aldebaran_kfd2kgd = { kgd_gfx_v9_get_atc_vmid_pasid_mapping_info, .set_vm_context_page_table_base = kgd_gfx_v9_set_vm_context_page_table_base, .get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy, - .program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings + .enable_debug_trap = kgd_aldebaran_enable_debug_trap, + .disable_debug_trap = kgd_aldebaran_disable_debug_trap, + .validate_trap_override_request = kgd_aldebaran_validate_trap_override_request, + .set_wave_launch_trap_override = kgd_aldebaran_set_wave_launch_trap_override, + .set_wave_launch_mode = kgd_aldebaran_set_wave_launch_mode, + .set_address_watch = kgd_gfx_aldebaran_set_address_watch, + .clear_address_watch = kgd_gfx_aldebaran_clear_address_watch, + .get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times, + .build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info, + .program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c index 4191af5a3f132282dd01eb1ef1b7afdf967562f1..625db444df1cb60ddff16397f0714ef4238d91ad 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c @@ -26,6 +26,7 @@ #include "amdgpu.h" #include "amdgpu_amdkfd.h" #include "amdgpu_amdkfd_arcturus.h" +#include "amdgpu_reset.h" #include "sdma0/sdma0_4_2_2_offset.h" #include "sdma0/sdma0_4_2_2_sh_mask.h" #include "sdma1/sdma1_4_2_2_offset.h" @@ -48,6 +49,8 @@ #include "amdgpu_amdkfd_gfx_v9.h" #include "gfxhub_v1_0.h" #include "mmhub_v9_4.h" +#include "gc/gc_9_0_offset.h" +#include "gc/gc_9_0_sh_mask.h" #define HQD_N_REGS 56 #define DUMP_REG(addr) do { \ @@ -276,6 +279,117 @@ int kgd_arcturus_hqd_sdma_destroy(struct amdgpu_device *adev, void *mqd, return 0; } +/* + * Helper used to suspend/resume gfx pipe for image post process work to set + * barrier behaviour. + */ +static int suspend_resume_compute_scheduler(struct amdgpu_device *adev, bool suspend) +{ + int i, r = 0; + + for (i = 0; i < adev->gfx.num_compute_rings; i++) { + struct amdgpu_ring *ring = &adev->gfx.compute_ring[i]; + + if (!(ring && ring->sched.thread)) + continue; + + /* stop secheduler and drain ring. */ + if (suspend) { + drm_sched_stop(&ring->sched, NULL); + r = amdgpu_fence_wait_empty(ring); + if (r) + goto out; + } else { + drm_sched_start(&ring->sched, false); + } + } + +out: + /* return on resume or failure to drain rings. */ + if (!suspend || r) + return r; + + return amdgpu_device_ip_wait_for_idle(adev, AMD_IP_BLOCK_TYPE_GFX); +} + +static void set_barrier_auto_waitcnt(struct amdgpu_device *adev, bool enable_waitcnt) +{ + uint32_t data; + + WRITE_ONCE(adev->barrier_has_auto_waitcnt, enable_waitcnt); + + if (!down_read_trylock(&adev->reset_domain->sem)) + return; + + amdgpu_amdkfd_suspend(adev, false); + + if (suspend_resume_compute_scheduler(adev, true)) + goto out; + + data = RREG32(SOC15_REG_OFFSET(GC, 0, mmSQ_CONFIG)); + data = REG_SET_FIELD(data, SQ_CONFIG, DISABLE_BARRIER_WAITCNT, + !enable_waitcnt); + WREG32(SOC15_REG_OFFSET(GC, 0, mmSQ_CONFIG), data); + +out: + suspend_resume_compute_scheduler(adev, false); + + amdgpu_amdkfd_resume(adev, false); + + up_read(&adev->reset_domain->sem); +} + +/* + * restore_dbg_registers is ignored here but is a general interface requirement + * for devices that support GFXOFF and where the RLC save/restore list + * does not support hw registers for debugging i.e. the driver has to manually + * initialize the debug mode registers after it has disabled GFX off during the + * debug session. + */ +static uint32_t kgd_arcturus_enable_debug_trap(struct amdgpu_device *adev, + bool restore_dbg_registers, + uint32_t vmid) +{ + mutex_lock(&adev->grbm_idx_mutex); + + kgd_gfx_v9_set_wave_launch_stall(adev, vmid, true); + + set_barrier_auto_waitcnt(adev, true); + + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0); + + kgd_gfx_v9_set_wave_launch_stall(adev, vmid, false); + + mutex_unlock(&adev->grbm_idx_mutex); + + return 0; +} + +/* + * keep_trap_enabled is ignored here but is a general interface requirement + * for devices that support multi-process debugging where the performance + * overhead from trap temporary setup needs to be bypassed when the debug + * session has ended. + */ +static uint32_t kgd_arcturus_disable_debug_trap(struct amdgpu_device *adev, + bool keep_trap_enabled, + uint32_t vmid) +{ + + mutex_lock(&adev->grbm_idx_mutex); + + kgd_gfx_v9_set_wave_launch_stall(adev, vmid, true); + + set_barrier_auto_waitcnt(adev, false); + + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0); + + kgd_gfx_v9_set_wave_launch_stall(adev, vmid, false); + + mutex_unlock(&adev->grbm_idx_mutex); + + return 0; +} const struct kfd2kgd_calls arcturus_kfd2kgd = { .program_sh_mem_settings = kgd_gfx_v9_program_sh_mem_settings, .set_pasid_vmid_mapping = kgd_gfx_v9_set_pasid_vmid_mapping, @@ -294,6 +408,15 @@ const struct kfd2kgd_calls arcturus_kfd2kgd = { kgd_gfx_v9_get_atc_vmid_pasid_mapping_info, .set_vm_context_page_table_base = kgd_gfx_v9_set_vm_context_page_table_base, + .enable_debug_trap = kgd_arcturus_enable_debug_trap, + .disable_debug_trap = kgd_arcturus_disable_debug_trap, + .validate_trap_override_request = kgd_gfx_v9_validate_trap_override_request, + .set_wave_launch_trap_override = kgd_gfx_v9_set_wave_launch_trap_override, + .set_wave_launch_mode = kgd_gfx_v9_set_wave_launch_mode, + .set_address_watch = kgd_gfx_v9_set_address_watch, + .clear_address_watch = kgd_gfx_v9_clear_address_watch, + .get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times, + .build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info, .get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy, .program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c new file mode 100644 index 0000000000000000000000000000000000000000..5b4b7f8b92a51c830a0b44cf72cdc7cd10d9b04a --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c @@ -0,0 +1,384 @@ +/* + * Copyright 2021 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "amdgpu.h" +#include "amdgpu_amdkfd.h" +#include "amdgpu_amdkfd_gfx_v9.h" +#include "gc/gc_9_4_3_offset.h" +#include "gc/gc_9_4_3_sh_mask.h" +#include "athub/athub_1_8_0_offset.h" +#include "athub/athub_1_8_0_sh_mask.h" +#include "oss/osssys_4_4_2_offset.h" +#include "oss/osssys_4_4_2_sh_mask.h" +#include "v9_structs.h" +#include "soc15.h" +#include "sdma/sdma_4_4_2_offset.h" +#include "sdma/sdma_4_4_2_sh_mask.h" + +static inline struct v9_sdma_mqd *get_sdma_mqd(void *mqd) +{ + return (struct v9_sdma_mqd *)mqd; +} + +static uint32_t get_sdma_rlc_reg_offset(struct amdgpu_device *adev, + unsigned int engine_id, + unsigned int queue_id) +{ + uint32_t sdma_engine_reg_base = + SOC15_REG_OFFSET(SDMA0, GET_INST(SDMA0, engine_id), + regSDMA_RLC0_RB_CNTL) - + regSDMA_RLC0_RB_CNTL; + uint32_t retval = sdma_engine_reg_base + + queue_id * (regSDMA_RLC1_RB_CNTL - regSDMA_RLC0_RB_CNTL); + + pr_debug("RLC register offset for SDMA%d RLC%d: 0x%x\n", engine_id, + queue_id, retval); + return retval; +} + +static int kgd_gfx_v9_4_3_hqd_sdma_load(struct amdgpu_device *adev, void *mqd, + uint32_t __user *wptr, struct mm_struct *mm) +{ + struct v9_sdma_mqd *m; + uint32_t sdma_rlc_reg_offset; + unsigned long end_jiffies; + uint32_t data; + uint64_t data64; + uint64_t __user *wptr64 = (uint64_t __user *)wptr; + + m = get_sdma_mqd(mqd); + sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev, m->sdma_engine_id, + m->sdma_queue_id); + + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_CNTL, + m->sdmax_rlcx_rb_cntl & (~SDMA_RLC0_RB_CNTL__RB_ENABLE_MASK)); + + end_jiffies = msecs_to_jiffies(2000) + jiffies; + while (true) { + data = RREG32(sdma_rlc_reg_offset + regSDMA_RLC0_CONTEXT_STATUS); + if (data & SDMA_RLC0_CONTEXT_STATUS__IDLE_MASK) + break; + if (time_after(jiffies, end_jiffies)) { + pr_err("SDMA RLC not idle in %s\n", __func__); + return -ETIME; + } + usleep_range(500, 1000); + } + + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_DOORBELL_OFFSET, + m->sdmax_rlcx_doorbell_offset); + + data = REG_SET_FIELD(m->sdmax_rlcx_doorbell, SDMA_RLC0_DOORBELL, + ENABLE, 1); + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_DOORBELL, data); + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_RPTR, + m->sdmax_rlcx_rb_rptr); + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_RPTR_HI, + m->sdmax_rlcx_rb_rptr_hi); + + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_MINOR_PTR_UPDATE, 1); + if (read_user_wptr(mm, wptr64, data64)) { + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_WPTR, + lower_32_bits(data64)); + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_WPTR_HI, + upper_32_bits(data64)); + } else { + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_WPTR, + m->sdmax_rlcx_rb_rptr); + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_WPTR_HI, + m->sdmax_rlcx_rb_rptr_hi); + } + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_MINOR_PTR_UPDATE, 0); + + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_BASE, m->sdmax_rlcx_rb_base); + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_BASE_HI, + m->sdmax_rlcx_rb_base_hi); + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_RPTR_ADDR_LO, + m->sdmax_rlcx_rb_rptr_addr_lo); + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_RPTR_ADDR_HI, + m->sdmax_rlcx_rb_rptr_addr_hi); + + data = REG_SET_FIELD(m->sdmax_rlcx_rb_cntl, SDMA_RLC0_RB_CNTL, + RB_ENABLE, 1); + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_CNTL, data); + + return 0; +} + +static int kgd_gfx_v9_4_3_hqd_sdma_dump(struct amdgpu_device *adev, + uint32_t engine_id, uint32_t queue_id, + uint32_t (**dump)[2], uint32_t *n_regs) +{ + uint32_t sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev, + engine_id, queue_id); + uint32_t i = 0, reg; +#undef HQD_N_REGS +#define HQD_N_REGS (19+6+7+12) +#define DUMP_REG(addr) do { \ + if (WARN_ON_ONCE(i >= HQD_N_REGS)) \ + break; \ + (*dump)[i][0] = (addr) << 2; \ + (*dump)[i++][1] = RREG32(addr); \ + } while (0) + + *dump = kmalloc_array(HQD_N_REGS * 2, sizeof(uint32_t), GFP_KERNEL); + if (*dump == NULL) + return -ENOMEM; + + for (reg = regSDMA_RLC0_RB_CNTL; reg <= regSDMA_RLC0_DOORBELL; reg++) + DUMP_REG(sdma_rlc_reg_offset + reg); + for (reg = regSDMA_RLC0_STATUS; reg <= regSDMA_RLC0_CSA_ADDR_HI; reg++) + DUMP_REG(sdma_rlc_reg_offset + reg); + for (reg = regSDMA_RLC0_IB_SUB_REMAIN; + reg <= regSDMA_RLC0_MINOR_PTR_UPDATE; reg++) + DUMP_REG(sdma_rlc_reg_offset + reg); + for (reg = regSDMA_RLC0_MIDCMD_DATA0; + reg <= regSDMA_RLC0_MIDCMD_CNTL; reg++) + DUMP_REG(sdma_rlc_reg_offset + reg); + + WARN_ON_ONCE(i != HQD_N_REGS); + *n_regs = i; + + return 0; +} + +static bool kgd_gfx_v9_4_3_hqd_sdma_is_occupied(struct amdgpu_device *adev, void *mqd) +{ + struct v9_sdma_mqd *m; + uint32_t sdma_rlc_reg_offset; + uint32_t sdma_rlc_rb_cntl; + + m = get_sdma_mqd(mqd); + sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev, m->sdma_engine_id, + m->sdma_queue_id); + + sdma_rlc_rb_cntl = RREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_CNTL); + + if (sdma_rlc_rb_cntl & SDMA_RLC0_RB_CNTL__RB_ENABLE_MASK) + return true; + + return false; +} + +static int kgd_gfx_v9_4_3_hqd_sdma_destroy(struct amdgpu_device *adev, void *mqd, + unsigned int utimeout) +{ + struct v9_sdma_mqd *m; + uint32_t sdma_rlc_reg_offset; + uint32_t temp; + unsigned long end_jiffies = (utimeout * HZ / 1000) + jiffies; + + m = get_sdma_mqd(mqd); + sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(adev, m->sdma_engine_id, + m->sdma_queue_id); + + temp = RREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_CNTL); + temp = temp & ~SDMA_RLC0_RB_CNTL__RB_ENABLE_MASK; + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_CNTL, temp); + + while (true) { + temp = RREG32(sdma_rlc_reg_offset + regSDMA_RLC0_CONTEXT_STATUS); + if (temp & SDMA_RLC0_CONTEXT_STATUS__IDLE_MASK) + break; + if (time_after(jiffies, end_jiffies)) { + pr_err("SDMA RLC not idle in %s\n", __func__); + return -ETIME; + } + usleep_range(500, 1000); + } + + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_DOORBELL, 0); + WREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_CNTL, + RREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_CNTL) | + SDMA_RLC0_RB_CNTL__RB_ENABLE_MASK); + + m->sdmax_rlcx_rb_rptr = + RREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_RPTR); + m->sdmax_rlcx_rb_rptr_hi = + RREG32(sdma_rlc_reg_offset + regSDMA_RLC0_RB_RPTR_HI); + + return 0; +} + +static int kgd_gfx_v9_4_3_set_pasid_vmid_mapping(struct amdgpu_device *adev, + u32 pasid, unsigned int vmid, uint32_t xcc_inst) +{ + unsigned long timeout; + unsigned int reg; + unsigned int phy_inst = GET_INST(GC, xcc_inst); + /* Every two XCCs share one AID */ + unsigned int aid = phy_inst / 2; + + /* + * We have to assume that there is no outstanding mapping. + * The ATC_VMID_PASID_MAPPING_UPDATE_STATUS bit could be 0 because + * a mapping is in progress or because a mapping finished + * and the SW cleared it. + * So the protocol is to always wait & clear. + */ + uint32_t pasid_mapping = (pasid == 0) ? 0 : (uint32_t)pasid | + ATC_VMID0_PASID_MAPPING__VALID_MASK; + + WREG32(SOC15_REG_OFFSET(ATHUB, 0, + regATC_VMID0_PASID_MAPPING) + vmid, pasid_mapping); + + timeout = jiffies + msecs_to_jiffies(10); + while (!(RREG32(SOC15_REG_OFFSET(ATHUB, 0, + regATC_VMID_PASID_MAPPING_UPDATE_STATUS)) & + (1U << vmid))) { + if (time_after(jiffies, timeout)) { + pr_err("Fail to program VMID-PASID mapping\n"); + return -ETIME; + } + cpu_relax(); + } + + WREG32(SOC15_REG_OFFSET(ATHUB, 0, + regATC_VMID_PASID_MAPPING_UPDATE_STATUS), + 1U << vmid); + + reg = RREG32(SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_LUT_INDEX)); + /* Every 4 numbers is a cycle. 1st is AID, 2nd and 3rd are XCDs, + * and the 4th is reserved. Therefore "aid * 4 + (xcc_inst % 2) + 1" + * programs _LUT for XCC and "aid * 4" for AID where the XCC connects + * to. + */ + WREG32(SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_LUT_INDEX), + aid * 4 + (phy_inst % 2) + 1); + WREG32(SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_0_LUT) + vmid, + pasid_mapping); + WREG32(SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_LUT_INDEX), + aid * 4); + WREG32(SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_0_LUT_MM) + vmid, + pasid_mapping); + WREG32(SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_LUT_INDEX), reg); + + return 0; +} + +static inline struct v9_mqd *get_mqd(void *mqd) +{ + return (struct v9_mqd *)mqd; +} + +static int kgd_gfx_v9_4_3_hqd_load(struct amdgpu_device *adev, void *mqd, + uint32_t pipe_id, uint32_t queue_id, + uint32_t __user *wptr, uint32_t wptr_shift, + uint32_t wptr_mask, struct mm_struct *mm, uint32_t inst) +{ + struct v9_mqd *m; + uint32_t *mqd_hqd; + uint32_t reg, hqd_base, hqd_end, data; + + m = get_mqd(mqd); + + kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst); + + /* HQD registers extend to CP_HQD_AQL_DISPATCH_ID_HI */ + mqd_hqd = &m->cp_mqd_base_addr_lo; + hqd_base = SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_MQD_BASE_ADDR); + hqd_end = SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_AQL_DISPATCH_ID_HI); + + for (reg = hqd_base; reg <= hqd_end; reg++) + WREG32_RLC(reg, mqd_hqd[reg - hqd_base]); + + + /* Activate doorbell logic before triggering WPTR poll. */ + data = REG_SET_FIELD(m->cp_hqd_pq_doorbell_control, + CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 1); + WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_PQ_DOORBELL_CONTROL), + data); + + if (wptr) { + /* Don't read wptr with get_user because the user + * context may not be accessible (if this function + * runs in a work queue). Instead trigger a one-shot + * polling read from memory in the CP. This assumes + * that wptr is GPU-accessible in the queue's VMID via + * ATC or SVM. WPTR==RPTR before starting the poll so + * the CP starts fetching new commands from the right + * place. + * + * Guessing a 64-bit WPTR from a 32-bit RPTR is a bit + * tricky. Assume that the queue didn't overflow. The + * number of valid bits in the 32-bit RPTR depends on + * the queue size. The remaining bits are taken from + * the saved 64-bit WPTR. If the WPTR wrapped, add the + * queue size. + */ + uint32_t queue_size = + 2 << REG_GET_FIELD(m->cp_hqd_pq_control, + CP_HQD_PQ_CONTROL, QUEUE_SIZE); + uint64_t guessed_wptr = m->cp_hqd_pq_rptr & (queue_size - 1); + + if ((m->cp_hqd_pq_wptr_lo & (queue_size - 1)) < guessed_wptr) + guessed_wptr += queue_size; + guessed_wptr += m->cp_hqd_pq_wptr_lo & ~(queue_size - 1); + guessed_wptr += (uint64_t)m->cp_hqd_pq_wptr_hi << 32; + + WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_PQ_WPTR_LO), + lower_32_bits(guessed_wptr)); + WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_PQ_WPTR_HI), + upper_32_bits(guessed_wptr)); + WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_PQ_WPTR_POLL_ADDR), + lower_32_bits((uintptr_t)wptr)); + WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), + regCP_HQD_PQ_WPTR_POLL_ADDR_HI), + upper_32_bits((uintptr_t)wptr)); + WREG32(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_PQ_WPTR_POLL_CNTL1), + (uint32_t)kgd_gfx_v9_get_queue_mask(adev, pipe_id, + queue_id)); + } + + /* Start the EOP fetcher */ + WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_EOP_RPTR), + REG_SET_FIELD(m->cp_hqd_eop_rptr, + CP_HQD_EOP_RPTR, INIT_FETCHER, 1)); + + data = REG_SET_FIELD(m->cp_hqd_active, CP_HQD_ACTIVE, ACTIVE, 1); + WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_ACTIVE), data); + + kgd_gfx_v9_release_queue(adev, inst); + + return 0; +} + +const struct kfd2kgd_calls gc_9_4_3_kfd2kgd = { + .program_sh_mem_settings = kgd_gfx_v9_program_sh_mem_settings, + .set_pasid_vmid_mapping = kgd_gfx_v9_4_3_set_pasid_vmid_mapping, + .init_interrupts = kgd_gfx_v9_init_interrupts, + .hqd_load = kgd_gfx_v9_4_3_hqd_load, + .hiq_mqd_load = kgd_gfx_v9_hiq_mqd_load, + .hqd_sdma_load = kgd_gfx_v9_4_3_hqd_sdma_load, + .hqd_dump = kgd_gfx_v9_hqd_dump, + .hqd_sdma_dump = kgd_gfx_v9_4_3_hqd_sdma_dump, + .hqd_is_occupied = kgd_gfx_v9_hqd_is_occupied, + .hqd_sdma_is_occupied = kgd_gfx_v9_4_3_hqd_sdma_is_occupied, + .hqd_destroy = kgd_gfx_v9_hqd_destroy, + .hqd_sdma_destroy = kgd_gfx_v9_4_3_hqd_sdma_destroy, + .wave_control_execute = kgd_gfx_v9_wave_control_execute, + .get_atc_vmid_pasid_mapping_info = + kgd_gfx_v9_get_atc_vmid_pasid_mapping_info, + .set_vm_context_page_table_base = + kgd_gfx_v9_set_vm_context_page_table_base, + .program_trap_handler_settings = + kgd_gfx_v9_program_trap_handler_settings +}; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c index 9378fc79e9ea61ef9ad694298bb448bdd5838355..8ad7a7779e147ea2ed5b55020b5acb2bb33ce282 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c @@ -21,6 +21,7 @@ */ #include "amdgpu.h" #include "amdgpu_amdkfd.h" +#include "amdgpu_amdkfd_gfx_v10.h" #include "gc/gc_10_1_0_offset.h" #include "gc/gc_10_1_0_sh_mask.h" #include "athub/athub_2_0_0_offset.h" @@ -31,6 +32,7 @@ #include "v10_structs.h" #include "nv.h" #include "nvd.h" +#include enum hqd_dequeue_request_type { NO_ACTION = 0, @@ -79,7 +81,7 @@ static void kgd_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmi uint32_t sh_mem_config, uint32_t sh_mem_ape1_base, uint32_t sh_mem_ape1_limit, - uint32_t sh_mem_bases) + uint32_t sh_mem_bases, uint32_t inst) { lock_srbm(adev, 0, 0, 0, vmid); @@ -91,7 +93,7 @@ static void kgd_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmi } static int kgd_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid, - unsigned int vmid) + unsigned int vmid, uint32_t inst) { /* * We have to assume that there is no outstanding mapping. @@ -135,7 +137,8 @@ static int kgd_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid, * but still works */ -static int kgd_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id) +static int kgd_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id, + uint32_t inst) { uint32_t mec; uint32_t pipe; @@ -205,7 +208,7 @@ static inline struct v10_sdma_mqd *get_sdma_mqd(void *mqd) static int kgd_hqd_load(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, uint32_t queue_id, uint32_t __user *wptr, uint32_t wptr_shift, - uint32_t wptr_mask, struct mm_struct *mm) + uint32_t wptr_mask, struct mm_struct *mm, uint32_t inst) { struct v10_compute_mqd *m; uint32_t *mqd_hqd; @@ -286,9 +289,9 @@ static int kgd_hqd_load(struct amdgpu_device *adev, void *mqd, static int kgd_hiq_mqd_load(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, uint32_t queue_id, - uint32_t doorbell_off) + uint32_t doorbell_off, uint32_t inst) { - struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring; + struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring; struct v10_compute_mqd *m; uint32_t mec, pipe; int r; @@ -303,7 +306,7 @@ static int kgd_hiq_mqd_load(struct amdgpu_device *adev, void *mqd, pr_debug("kfd: set HIQ, mec:%d, pipe:%d, queue:%d.\n", mec, pipe, queue_id); - spin_lock(&adev->gfx.kiq.ring_lock); + spin_lock(&adev->gfx.kiq[0].ring_lock); r = amdgpu_ring_alloc(kiq_ring, 7); if (r) { pr_err("Failed to alloc KIQ (%d).\n", r); @@ -330,7 +333,7 @@ static int kgd_hiq_mqd_load(struct amdgpu_device *adev, void *mqd, amdgpu_ring_commit(kiq_ring); out_unlock: - spin_unlock(&adev->gfx.kiq.ring_lock); + spin_unlock(&adev->gfx.kiq[0].ring_lock); release_queue(adev); return r; @@ -338,7 +341,7 @@ static int kgd_hiq_mqd_load(struct amdgpu_device *adev, void *mqd, static int kgd_hqd_dump(struct amdgpu_device *adev, uint32_t pipe_id, uint32_t queue_id, - uint32_t (**dump)[2], uint32_t *n_regs) + uint32_t (**dump)[2], uint32_t *n_regs, uint32_t inst) { uint32_t i = 0, reg; #define HQD_N_REGS 56 @@ -469,7 +472,7 @@ static int kgd_hqd_sdma_dump(struct amdgpu_device *adev, static bool kgd_hqd_is_occupied(struct amdgpu_device *adev, uint64_t queue_address, uint32_t pipe_id, - uint32_t queue_id) + uint32_t queue_id, uint32_t inst) { uint32_t act; bool retval = false; @@ -510,7 +513,7 @@ static bool kgd_hqd_sdma_is_occupied(struct amdgpu_device *adev, void *mqd) static int kgd_hqd_destroy(struct amdgpu_device *adev, void *mqd, enum kfd_preempt_type reset_type, unsigned int utimeout, uint32_t pipe_id, - uint32_t queue_id) + uint32_t queue_id, uint32_t inst) { enum hqd_dequeue_request_type type; unsigned long end_jiffies; @@ -673,7 +676,7 @@ static bool get_atc_vmid_pasid_mapping_info(struct amdgpu_device *adev, static int kgd_wave_control_execute(struct amdgpu_device *adev, uint32_t gfx_index_val, - uint32_t sq_cmd) + uint32_t sq_cmd, uint32_t inst) { uint32_t data = 0; @@ -708,8 +711,295 @@ static void set_vm_context_page_table_base(struct amdgpu_device *adev, adev->gfxhub.funcs->setup_vm_pt_regs(adev, vmid, page_table_base); } +/* + * GFX10 helper for wave launch stall requirements on debug trap setting. + * + * vmid: + * Target VMID to stall/unstall. + * + * stall: + * 0-unstall wave launch (enable), 1-stall wave launch (disable). + * After wavefront launch has been stalled, allocated waves must drain from + * SPI in order for debug trap settings to take effect on those waves. + * This is roughly a ~3500 clock cycle wait on SPI where a read on + * SPI_GDBG_WAVE_CNTL translates to ~32 clock cycles. + * KGD_GFX_V10_WAVE_LAUNCH_SPI_DRAIN_LATENCY indicates the number of reads required. + * + * NOTE: We can afford to clear the entire STALL_VMID field on unstall + * because current GFX10 chips cannot support multi-process debugging due to + * trap configuration and masking being limited to global scope. Always + * assume single process conditions. + * + */ + +#define KGD_GFX_V10_WAVE_LAUNCH_SPI_DRAIN_LATENCY 110 +static void kgd_gfx_v10_set_wave_launch_stall(struct amdgpu_device *adev, uint32_t vmid, bool stall) +{ + uint32_t data = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL)); + int i; + + data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL, STALL_VMID, + stall ? 1 << vmid : 0); + + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), data); + + if (!stall) + return; + + for (i = 0; i < KGD_GFX_V10_WAVE_LAUNCH_SPI_DRAIN_LATENCY; i++) + RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL)); +} + +uint32_t kgd_gfx_v10_enable_debug_trap(struct amdgpu_device *adev, + bool restore_dbg_registers, + uint32_t vmid) +{ + + mutex_lock(&adev->grbm_idx_mutex); + + kgd_gfx_v10_set_wave_launch_stall(adev, vmid, true); + + /* assume gfx off is disabled for the debug session if rlc restore not supported. */ + if (restore_dbg_registers) { + uint32_t data = 0; + + data = REG_SET_FIELD(data, SPI_GDBG_TRAP_CONFIG, + VMID_SEL, 1 << vmid); + data = REG_SET_FIELD(data, SPI_GDBG_TRAP_CONFIG, + TRAP_EN, 1); + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_CONFIG), data); + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_DATA0), 0); + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_DATA1), 0); + + kgd_gfx_v10_set_wave_launch_stall(adev, vmid, false); + + mutex_unlock(&adev->grbm_idx_mutex); + + return 0; + } + + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0); + + kgd_gfx_v10_set_wave_launch_stall(adev, vmid, false); + + mutex_unlock(&adev->grbm_idx_mutex); + + return 0; +} + +uint32_t kgd_gfx_v10_disable_debug_trap(struct amdgpu_device *adev, + bool keep_trap_enabled, + uint32_t vmid) +{ + mutex_lock(&adev->grbm_idx_mutex); + + kgd_gfx_v10_set_wave_launch_stall(adev, vmid, true); + + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0); + + kgd_gfx_v10_set_wave_launch_stall(adev, vmid, false); + + mutex_unlock(&adev->grbm_idx_mutex); + + return 0; +} + +int kgd_gfx_v10_validate_trap_override_request(struct amdgpu_device *adev, + uint32_t trap_override, + uint32_t *trap_mask_supported) +{ + *trap_mask_supported &= KFD_DBG_TRAP_MASK_DBG_ADDRESS_WATCH; + + /* The SPI_GDBG_TRAP_MASK register is global and affects all + * processes. Only allow OR-ing the address-watch bit, since + * this only affects processes under the debugger. Other bits + * should stay 0 to avoid the debugger interfering with other + * processes. + */ + if (trap_override != KFD_DBG_TRAP_OVERRIDE_OR) + return -EINVAL; + + return 0; +} + +uint32_t kgd_gfx_v10_set_wave_launch_trap_override(struct amdgpu_device *adev, + uint32_t vmid, + uint32_t trap_override, + uint32_t trap_mask_bits, + uint32_t trap_mask_request, + uint32_t *trap_mask_prev, + uint32_t kfd_dbg_trap_cntl_prev) +{ + uint32_t data, wave_cntl_prev; + + mutex_lock(&adev->grbm_idx_mutex); + + wave_cntl_prev = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL)); + + kgd_gfx_v10_set_wave_launch_stall(adev, vmid, true); + + data = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK)); + *trap_mask_prev = REG_GET_FIELD(data, SPI_GDBG_TRAP_MASK, EXCP_EN); + + trap_mask_bits = (trap_mask_bits & trap_mask_request) | + (*trap_mask_prev & ~trap_mask_request); + + data = REG_SET_FIELD(data, SPI_GDBG_TRAP_MASK, EXCP_EN, trap_mask_bits); + data = REG_SET_FIELD(data, SPI_GDBG_TRAP_MASK, REPLACE, trap_override); + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), data); + + /* We need to preserve wave launch mode stall settings. */ + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), wave_cntl_prev); + + mutex_unlock(&adev->grbm_idx_mutex); + + return 0; +} + +uint32_t kgd_gfx_v10_set_wave_launch_mode(struct amdgpu_device *adev, + uint8_t wave_launch_mode, + uint32_t vmid) +{ + uint32_t data = 0; + bool is_mode_set = !!wave_launch_mode; + + mutex_lock(&adev->grbm_idx_mutex); + + kgd_gfx_v10_set_wave_launch_stall(adev, vmid, true); + + data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL2, + VMID_MASK, is_mode_set ? 1 << vmid : 0); + data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL2, + MODE, is_mode_set ? wave_launch_mode : 0); + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL2), data); + + kgd_gfx_v10_set_wave_launch_stall(adev, vmid, false); + + mutex_unlock(&adev->grbm_idx_mutex); + + return 0; +} + +#define TCP_WATCH_STRIDE (mmTCP_WATCH1_ADDR_H - mmTCP_WATCH0_ADDR_H) +uint32_t kgd_gfx_v10_set_address_watch(struct amdgpu_device *adev, + uint64_t watch_address, + uint32_t watch_address_mask, + uint32_t watch_id, + uint32_t watch_mode, + uint32_t debug_vmid) +{ + uint32_t watch_address_high; + uint32_t watch_address_low; + uint32_t watch_address_cntl; + + watch_address_cntl = 0; + + watch_address_low = lower_32_bits(watch_address); + watch_address_high = upper_32_bits(watch_address) & 0xffff; + + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + VMID, + debug_vmid); + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + MODE, + watch_mode); + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + MASK, + watch_address_mask >> 7); + + /* Turning off this watch point until we set all the registers */ + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + VALID, + 0); + + WREG32((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_CNTL) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_cntl); + + WREG32((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_ADDR_H) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_high); + + WREG32((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_ADDR_L) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_low); + + /* Enable the watch point */ + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + VALID, + 1); + + WREG32((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_CNTL) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_cntl); + + return 0; +} + +uint32_t kgd_gfx_v10_clear_address_watch(struct amdgpu_device *adev, + uint32_t watch_id) +{ + uint32_t watch_address_cntl; + + watch_address_cntl = 0; + + WREG32((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_CNTL) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_cntl); + + return 0; +} + + +/* kgd_gfx_v10_get_iq_wait_times: Returns the mmCP_IQ_WAIT_TIME1/2 values + * The values read are: + * ib_offload_wait_time -- Wait Count for Indirect Buffer Offloads. + * atomic_offload_wait_time -- Wait Count for L2 and GDS Atomics Offloads. + * wrm_offload_wait_time -- Wait Count for WAIT_REG_MEM Offloads. + * gws_wait_time -- Wait Count for Global Wave Syncs. + * que_sleep_wait_time -- Wait Count for Dequeue Retry. + * sch_wave_wait_time -- Wait Count for Scheduling Wave Message. + * sem_rearm_wait_time -- Wait Count for Semaphore re-arm. + * deq_retry_wait_time -- Wait Count for Global Wave Syncs. + */ +void kgd_gfx_v10_get_iq_wait_times(struct amdgpu_device *adev, + uint32_t *wait_times) + +{ + *wait_times = RREG32(SOC15_REG_OFFSET(GC, 0, mmCP_IQ_WAIT_TIME2)); +} + +void kgd_gfx_v10_build_grace_period_packet_info(struct amdgpu_device *adev, + uint32_t wait_times, + uint32_t grace_period, + uint32_t *reg_offset, + uint32_t *reg_data) +{ + *reg_data = wait_times; + + /* + * The CP cannont handle a 0 grace period input and will result in + * an infinite grace period being set so set to 1 to prevent this. + */ + if (grace_period == 0) + grace_period = 1; + + *reg_data = REG_SET_FIELD(*reg_data, + CP_IQ_WAIT_TIME2, + SCH_WAVE, + grace_period); + + *reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_IQ_WAIT_TIME2); +} + static void program_trap_handler_settings(struct amdgpu_device *adev, - uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr) + uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr, + uint32_t inst) { lock_srbm(adev, 0, 0, 0, vmid); @@ -750,5 +1040,14 @@ const struct kfd2kgd_calls gfx_v10_kfd2kgd = { .get_atc_vmid_pasid_mapping_info = get_atc_vmid_pasid_mapping_info, .set_vm_context_page_table_base = set_vm_context_page_table_base, + .enable_debug_trap = kgd_gfx_v10_enable_debug_trap, + .disable_debug_trap = kgd_gfx_v10_disable_debug_trap, + .validate_trap_override_request = kgd_gfx_v10_validate_trap_override_request, + .set_wave_launch_trap_override = kgd_gfx_v10_set_wave_launch_trap_override, + .set_wave_launch_mode = kgd_gfx_v10_set_wave_launch_mode, + .set_address_watch = kgd_gfx_v10_set_address_watch, + .clear_address_watch = kgd_gfx_v10_clear_address_watch, + .get_iq_wait_times = kgd_gfx_v10_get_iq_wait_times, + .build_grace_period_packet_info = kgd_gfx_v10_build_grace_period_packet_info, .program_trap_handler_settings = program_trap_handler_settings, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h new file mode 100644 index 0000000000000000000000000000000000000000..e6b70196071ac114fe03ab497c108296fb29ab43 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h @@ -0,0 +1,55 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ + +uint32_t kgd_gfx_v10_enable_debug_trap(struct amdgpu_device *adev, + bool restore_dbg_registers, + uint32_t vmid); +uint32_t kgd_gfx_v10_disable_debug_trap(struct amdgpu_device *adev, + bool keep_trap_enabled, + uint32_t vmid); +int kgd_gfx_v10_validate_trap_override_request(struct amdgpu_device *adev, + uint32_t trap_override, + uint32_t *trap_mask_supported); +uint32_t kgd_gfx_v10_set_wave_launch_trap_override(struct amdgpu_device *adev, + uint32_t vmid, + uint32_t trap_override, + uint32_t trap_mask_bits, + uint32_t trap_mask_request, + uint32_t *trap_mask_prev, + uint32_t kfd_dbg_trap_cntl_prev); +uint32_t kgd_gfx_v10_set_wave_launch_mode(struct amdgpu_device *adev, + uint8_t wave_launch_mode, + uint32_t vmid); +uint32_t kgd_gfx_v10_set_address_watch(struct amdgpu_device *adev, + uint64_t watch_address, + uint32_t watch_address_mask, + uint32_t watch_id, + uint32_t watch_mode, + uint32_t debug_vmid); +uint32_t kgd_gfx_v10_clear_address_watch(struct amdgpu_device *adev, + uint32_t watch_id); +void kgd_gfx_v10_get_iq_wait_times(struct amdgpu_device *adev, uint32_t *wait_times); +void kgd_gfx_v10_build_grace_period_packet_info(struct amdgpu_device *adev, + uint32_t wait_times, + uint32_t grace_period, + uint32_t *reg_offset, + uint32_t *reg_data); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c index ba21ec6b35e0caf72c409c3ac72a36ab7d51a895..8c8437a4383f7bf0aa85883a20ce50cf2a332b0d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c @@ -22,6 +22,7 @@ #include #include "amdgpu.h" #include "amdgpu_amdkfd.h" +#include "amdgpu_amdkfd_gfx_v10.h" #include "gc/gc_10_3_0_offset.h" #include "gc/gc_10_3_0_sh_mask.h" #include "oss/osssys_5_0_0_offset.h" @@ -80,7 +81,7 @@ static void program_sh_mem_settings_v10_3(struct amdgpu_device *adev, uint32_t v uint32_t sh_mem_config, uint32_t sh_mem_ape1_base, uint32_t sh_mem_ape1_limit, - uint32_t sh_mem_bases) + uint32_t sh_mem_bases, uint32_t inst) { lock_srbm(adev, 0, 0, 0, vmid); @@ -93,7 +94,7 @@ static void program_sh_mem_settings_v10_3(struct amdgpu_device *adev, uint32_t v /* ATC is defeatured on Sienna_Cichlid */ static int set_pasid_vmid_mapping_v10_3(struct amdgpu_device *adev, unsigned int pasid, - unsigned int vmid) + unsigned int vmid, uint32_t inst) { uint32_t value = pasid << IH_VMID_0_LUT__PASID__SHIFT; @@ -105,7 +106,8 @@ static int set_pasid_vmid_mapping_v10_3(struct amdgpu_device *adev, unsigned int return 0; } -static int init_interrupts_v10_3(struct amdgpu_device *adev, uint32_t pipe_id) +static int init_interrupts_v10_3(struct amdgpu_device *adev, uint32_t pipe_id, + uint32_t inst) { uint32_t mec; uint32_t pipe; @@ -177,7 +179,7 @@ static inline struct v10_sdma_mqd *get_sdma_mqd(void *mqd) static int hqd_load_v10_3(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, uint32_t queue_id, uint32_t __user *wptr, uint32_t wptr_shift, - uint32_t wptr_mask, struct mm_struct *mm) + uint32_t wptr_mask, struct mm_struct *mm, uint32_t inst) { struct v10_compute_mqd *m; uint32_t *mqd_hqd; @@ -273,9 +275,9 @@ static int hqd_load_v10_3(struct amdgpu_device *adev, void *mqd, static int hiq_mqd_load_v10_3(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, uint32_t queue_id, - uint32_t doorbell_off) + uint32_t doorbell_off, uint32_t inst) { - struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring; + struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring; struct v10_compute_mqd *m; uint32_t mec, pipe; int r; @@ -290,7 +292,7 @@ static int hiq_mqd_load_v10_3(struct amdgpu_device *adev, void *mqd, pr_debug("kfd: set HIQ, mec:%d, pipe:%d, queue:%d.\n", mec, pipe, queue_id); - spin_lock(&adev->gfx.kiq.ring_lock); + spin_lock(&adev->gfx.kiq[0].ring_lock); r = amdgpu_ring_alloc(kiq_ring, 7); if (r) { pr_err("Failed to alloc KIQ (%d).\n", r); @@ -317,7 +319,7 @@ static int hiq_mqd_load_v10_3(struct amdgpu_device *adev, void *mqd, amdgpu_ring_commit(kiq_ring); out_unlock: - spin_unlock(&adev->gfx.kiq.ring_lock); + spin_unlock(&adev->gfx.kiq[0].ring_lock); release_queue(adev); return r; @@ -325,7 +327,7 @@ static int hiq_mqd_load_v10_3(struct amdgpu_device *adev, void *mqd, static int hqd_dump_v10_3(struct amdgpu_device *adev, uint32_t pipe_id, uint32_t queue_id, - uint32_t (**dump)[2], uint32_t *n_regs) + uint32_t (**dump)[2], uint32_t *n_regs, uint32_t inst) { uint32_t i = 0, reg; #define HQD_N_REGS 56 @@ -456,7 +458,7 @@ static int hqd_sdma_dump_v10_3(struct amdgpu_device *adev, static bool hqd_is_occupied_v10_3(struct amdgpu_device *adev, uint64_t queue_address, uint32_t pipe_id, - uint32_t queue_id) + uint32_t queue_id, uint32_t inst) { uint32_t act; bool retval = false; @@ -498,7 +500,7 @@ static bool hqd_sdma_is_occupied_v10_3(struct amdgpu_device *adev, static int hqd_destroy_v10_3(struct amdgpu_device *adev, void *mqd, enum kfd_preempt_type reset_type, unsigned int utimeout, uint32_t pipe_id, - uint32_t queue_id) + uint32_t queue_id, uint32_t inst) { enum hqd_dequeue_request_type type; unsigned long end_jiffies; @@ -586,7 +588,7 @@ static int hqd_sdma_destroy_v10_3(struct amdgpu_device *adev, void *mqd, static int wave_control_execute_v10_3(struct amdgpu_device *adev, uint32_t gfx_index_val, - uint32_t sq_cmd) + uint32_t sq_cmd, uint32_t inst) { uint32_t data = 0; @@ -628,7 +630,8 @@ static void set_vm_context_page_table_base_v10_3(struct amdgpu_device *adev, } static void program_trap_handler_settings_v10_3(struct amdgpu_device *adev, - uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr) + uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr, + uint32_t inst) { lock_srbm(adev, 0, 0, 0, vmid); @@ -652,142 +655,6 @@ static void program_trap_handler_settings_v10_3(struct amdgpu_device *adev, unlock_srbm(adev); } -#if 0 -uint32_t enable_debug_trap_v10_3(struct amdgpu_device *adev, - uint32_t trap_debug_wave_launch_mode, - uint32_t vmid) -{ - uint32_t data = 0; - uint32_t orig_wave_cntl_value; - uint32_t orig_stall_vmid; - - mutex_lock(&adev->grbm_idx_mutex); - - orig_wave_cntl_value = RREG32(SOC15_REG_OFFSET(GC, - 0, - mmSPI_GDBG_WAVE_CNTL)); - orig_stall_vmid = REG_GET_FIELD(orig_wave_cntl_value, - SPI_GDBG_WAVE_CNTL, - STALL_VMID); - - data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL, STALL_RA, 1); - WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), data); - - data = 0; - WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), data); - - WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), orig_stall_vmid); - - mutex_unlock(&adev->grbm_idx_mutex); - - return 0; -} - -uint32_t disable_debug_trap_v10_3(struct amdgpu_device *adev) -{ - mutex_lock(&adev->grbm_idx_mutex); - - WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0); - - mutex_unlock(&adev->grbm_idx_mutex); - - return 0; -} - -uint32_t set_wave_launch_trap_override_v10_3(struct amdgpu_device *adev, - uint32_t trap_override, - uint32_t trap_mask) -{ - uint32_t data = 0; - - mutex_lock(&adev->grbm_idx_mutex); - - data = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL)); - data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL, STALL_RA, 1); - WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), data); - - data = 0; - data = REG_SET_FIELD(data, SPI_GDBG_TRAP_MASK, - EXCP_EN, trap_mask); - data = REG_SET_FIELD(data, SPI_GDBG_TRAP_MASK, - REPLACE, trap_override); - WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), data); - - data = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL)); - data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL, STALL_RA, 0); - WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), data); - - mutex_unlock(&adev->grbm_idx_mutex); - - return 0; -} - -uint32_t set_wave_launch_mode_v10_3(struct amdgpu_device *adev, - uint8_t wave_launch_mode, - uint32_t vmid) -{ - uint32_t data = 0; - bool is_stall_mode; - bool is_mode_set; - - is_stall_mode = (wave_launch_mode == 4); - is_mode_set = (wave_launch_mode != 0 && wave_launch_mode != 4); - - mutex_lock(&adev->grbm_idx_mutex); - - data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL2, - VMID_MASK, is_mode_set ? 1 << vmid : 0); - data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL2, - MODE, is_mode_set ? wave_launch_mode : 0); - WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL2), data); - - data = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL)); - data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL, - STALL_VMID, is_stall_mode ? 1 << vmid : 0); - data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL, - STALL_RA, is_stall_mode ? 1 : 0); - WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), data); - - mutex_unlock(&adev->grbm_idx_mutex); - - return 0; -} - -/* kgd_get_iq_wait_times: Returns the mmCP_IQ_WAIT_TIME1/2 values - * The values read are: - * ib_offload_wait_time -- Wait Count for Indirect Buffer Offloads. - * atomic_offload_wait_time -- Wait Count for L2 and GDS Atomics Offloads. - * wrm_offload_wait_time -- Wait Count for WAIT_REG_MEM Offloads. - * gws_wait_time -- Wait Count for Global Wave Syncs. - * que_sleep_wait_time -- Wait Count for Dequeue Retry. - * sch_wave_wait_time -- Wait Count for Scheduling Wave Message. - * sem_rearm_wait_time -- Wait Count for Semaphore re-arm. - * deq_retry_wait_time -- Wait Count for Global Wave Syncs. - */ -void get_iq_wait_times_v10_3(struct amdgpu_device *adev, - uint32_t *wait_times) - -{ - *wait_times = RREG32(SOC15_REG_OFFSET(GC, 0, mmCP_IQ_WAIT_TIME2)); -} - -void build_grace_period_packet_info_v10_3(struct amdgpu_device *adev, - uint32_t wait_times, - uint32_t grace_period, - uint32_t *reg_offset, - uint32_t *reg_data) -{ - *reg_data = wait_times; - - *reg_data = REG_SET_FIELD(*reg_data, - CP_IQ_WAIT_TIME2, - SCH_WAVE, - grace_period); - - *reg_offset = mmCP_IQ_WAIT_TIME2; -} -#endif - const struct kfd2kgd_calls gfx_v10_3_kfd2kgd = { .program_sh_mem_settings = program_sh_mem_settings_v10_3, .set_pasid_vmid_mapping = set_pasid_vmid_mapping_v10_3, @@ -805,12 +672,13 @@ const struct kfd2kgd_calls gfx_v10_3_kfd2kgd = { .get_atc_vmid_pasid_mapping_info = get_atc_vmid_pasid_mapping_info_v10_3, .set_vm_context_page_table_base = set_vm_context_page_table_base_v10_3, .program_trap_handler_settings = program_trap_handler_settings_v10_3, -#if 0 - .enable_debug_trap = enable_debug_trap_v10_3, - .disable_debug_trap = disable_debug_trap_v10_3, - .set_wave_launch_trap_override = set_wave_launch_trap_override_v10_3, - .set_wave_launch_mode = set_wave_launch_mode_v10_3, - .get_iq_wait_times = get_iq_wait_times_v10_3, - .build_grace_period_packet_info = build_grace_period_packet_info_v10_3, -#endif + .get_iq_wait_times = kgd_gfx_v10_get_iq_wait_times, + .build_grace_period_packet_info = kgd_gfx_v10_build_grace_period_packet_info, + .enable_debug_trap = kgd_gfx_v10_enable_debug_trap, + .disable_debug_trap = kgd_gfx_v10_disable_debug_trap, + .validate_trap_override_request = kgd_gfx_v10_validate_trap_override_request, + .set_wave_launch_trap_override = kgd_gfx_v10_set_wave_launch_trap_override, + .set_wave_launch_mode = kgd_gfx_v10_set_wave_launch_mode, + .set_address_watch = kgd_gfx_v10_set_address_watch, + .clear_address_watch = kgd_gfx_v10_clear_address_watch }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c index 7e80caa05060b90fd3a5ce6c9873792daca14bda..91c3574ebed303871a372b6d0d9841c34a578d3c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c @@ -30,6 +30,7 @@ #include "soc15d.h" #include "v11_structs.h" #include "soc21.h" +#include enum hqd_dequeue_request_type { NO_ACTION = 0, @@ -78,7 +79,7 @@ static void program_sh_mem_settings_v11(struct amdgpu_device *adev, uint32_t vmi uint32_t sh_mem_config, uint32_t sh_mem_ape1_base, uint32_t sh_mem_ape1_limit, - uint32_t sh_mem_bases) + uint32_t sh_mem_bases, uint32_t inst) { lock_srbm(adev, 0, 0, 0, vmid); @@ -89,7 +90,7 @@ static void program_sh_mem_settings_v11(struct amdgpu_device *adev, uint32_t vmi } static int set_pasid_vmid_mapping_v11(struct amdgpu_device *adev, unsigned int pasid, - unsigned int vmid) + unsigned int vmid, uint32_t inst) { uint32_t value = pasid << IH_VMID_0_LUT__PASID__SHIFT; @@ -101,7 +102,8 @@ static int set_pasid_vmid_mapping_v11(struct amdgpu_device *adev, unsigned int p return 0; } -static int init_interrupts_v11(struct amdgpu_device *adev, uint32_t pipe_id) +static int init_interrupts_v11(struct amdgpu_device *adev, uint32_t pipe_id, + uint32_t inst) { uint32_t mec; uint32_t pipe; @@ -162,7 +164,7 @@ static inline struct v11_sdma_mqd *get_sdma_mqd(void *mqd) static int hqd_load_v11(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, uint32_t queue_id, uint32_t __user *wptr, uint32_t wptr_shift, uint32_t wptr_mask, - struct mm_struct *mm) + struct mm_struct *mm, uint32_t inst) { struct v11_compute_mqd *m; uint32_t *mqd_hqd; @@ -258,9 +260,9 @@ static int hqd_load_v11(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, static int hiq_mqd_load_v11(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, uint32_t queue_id, - uint32_t doorbell_off) + uint32_t doorbell_off, uint32_t inst) { - struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring; + struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring; struct v11_compute_mqd *m; uint32_t mec, pipe; int r; @@ -275,7 +277,7 @@ static int hiq_mqd_load_v11(struct amdgpu_device *adev, void *mqd, pr_debug("kfd: set HIQ, mec:%d, pipe:%d, queue:%d.\n", mec, pipe, queue_id); - spin_lock(&adev->gfx.kiq.ring_lock); + spin_lock(&adev->gfx.kiq[0].ring_lock); r = amdgpu_ring_alloc(kiq_ring, 7); if (r) { pr_err("Failed to alloc KIQ (%d).\n", r); @@ -302,7 +304,7 @@ static int hiq_mqd_load_v11(struct amdgpu_device *adev, void *mqd, amdgpu_ring_commit(kiq_ring); out_unlock: - spin_unlock(&adev->gfx.kiq.ring_lock); + spin_unlock(&adev->gfx.kiq[0].ring_lock); release_queue(adev); return r; @@ -310,7 +312,7 @@ static int hiq_mqd_load_v11(struct amdgpu_device *adev, void *mqd, static int hqd_dump_v11(struct amdgpu_device *adev, uint32_t pipe_id, uint32_t queue_id, - uint32_t (**dump)[2], uint32_t *n_regs) + uint32_t (**dump)[2], uint32_t *n_regs, uint32_t inst) { uint32_t i = 0, reg; #define HQD_N_REGS 56 @@ -445,7 +447,7 @@ static int hqd_sdma_dump_v11(struct amdgpu_device *adev, } static bool hqd_is_occupied_v11(struct amdgpu_device *adev, uint64_t queue_address, - uint32_t pipe_id, uint32_t queue_id) + uint32_t pipe_id, uint32_t queue_id, uint32_t inst) { uint32_t act; bool retval = false; @@ -486,7 +488,7 @@ static bool hqd_sdma_is_occupied_v11(struct amdgpu_device *adev, void *mqd) static int hqd_destroy_v11(struct amdgpu_device *adev, void *mqd, enum kfd_preempt_type reset_type, unsigned int utimeout, uint32_t pipe_id, - uint32_t queue_id) + uint32_t queue_id, uint32_t inst) { enum hqd_dequeue_request_type type; unsigned long end_jiffies; @@ -571,7 +573,7 @@ static int hqd_sdma_destroy_v11(struct amdgpu_device *adev, void *mqd, static int wave_control_execute_v11(struct amdgpu_device *adev, uint32_t gfx_index_val, - uint32_t sq_cmd) + uint32_t sq_cmd, uint32_t inst) { uint32_t data = 0; @@ -606,6 +608,183 @@ static void set_vm_context_page_table_base_v11(struct amdgpu_device *adev, adev->gfxhub.funcs->setup_vm_pt_regs(adev, vmid, page_table_base); } +/* + * Returns TRAP_EN, EXCP_EN and EXCP_REPLACE. + * + * restore_dbg_registers is ignored here but is a general interface requirement + * for devices that support GFXOFF and where the RLC save/restore list + * does not support hw registers for debugging i.e. the driver has to manually + * initialize the debug mode registers after it has disabled GFX off during the + * debug session. + */ +static uint32_t kgd_gfx_v11_enable_debug_trap(struct amdgpu_device *adev, + bool restore_dbg_registers, + uint32_t vmid) +{ + uint32_t data = 0; + + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, 0); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, 0); + + return data; +} + +/* Returns TRAP_EN, EXCP_EN and EXCP_REPLACE. */ +static uint32_t kgd_gfx_v11_disable_debug_trap(struct amdgpu_device *adev, + bool keep_trap_enabled, + uint32_t vmid) +{ + uint32_t data = 0; + + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, keep_trap_enabled); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, 0); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, 0); + + return data; +} + +static int kgd_gfx_v11_validate_trap_override_request(struct amdgpu_device *adev, + uint32_t trap_override, + uint32_t *trap_mask_supported) +{ + *trap_mask_supported &= KFD_DBG_TRAP_MASK_FP_INVALID | + KFD_DBG_TRAP_MASK_FP_INPUT_DENORMAL | + KFD_DBG_TRAP_MASK_FP_DIVIDE_BY_ZERO | + KFD_DBG_TRAP_MASK_FP_OVERFLOW | + KFD_DBG_TRAP_MASK_FP_UNDERFLOW | + KFD_DBG_TRAP_MASK_FP_INEXACT | + KFD_DBG_TRAP_MASK_INT_DIVIDE_BY_ZERO | + KFD_DBG_TRAP_MASK_DBG_ADDRESS_WATCH | + KFD_DBG_TRAP_MASK_DBG_MEMORY_VIOLATION; + + if (adev->ip_versions[GC_HWIP][0] >= IP_VERSION(11, 0, 4)) + *trap_mask_supported |= KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_START | + KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_END; + + if (trap_override != KFD_DBG_TRAP_OVERRIDE_OR && + trap_override != KFD_DBG_TRAP_OVERRIDE_REPLACE) + return -EPERM; + + return 0; +} + +static uint32_t trap_mask_map_sw_to_hw(uint32_t mask) +{ + uint32_t trap_on_start = (mask & KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_START) ? 1 : 0; + uint32_t trap_on_end = (mask & KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_END) ? 1 : 0; + uint32_t excp_en = mask & (KFD_DBG_TRAP_MASK_FP_INVALID | + KFD_DBG_TRAP_MASK_FP_INPUT_DENORMAL | + KFD_DBG_TRAP_MASK_FP_DIVIDE_BY_ZERO | + KFD_DBG_TRAP_MASK_FP_OVERFLOW | + KFD_DBG_TRAP_MASK_FP_UNDERFLOW | + KFD_DBG_TRAP_MASK_FP_INEXACT | + KFD_DBG_TRAP_MASK_INT_DIVIDE_BY_ZERO | + KFD_DBG_TRAP_MASK_DBG_ADDRESS_WATCH | + KFD_DBG_TRAP_MASK_DBG_MEMORY_VIOLATION); + uint32_t ret; + + ret = REG_SET_FIELD(0, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, excp_en); + ret = REG_SET_FIELD(ret, SPI_GDBG_PER_VMID_CNTL, TRAP_ON_START, trap_on_start); + ret = REG_SET_FIELD(ret, SPI_GDBG_PER_VMID_CNTL, TRAP_ON_END, trap_on_end); + + return ret; +} + +static uint32_t trap_mask_map_hw_to_sw(uint32_t mask) +{ + uint32_t ret = REG_GET_FIELD(mask, SPI_GDBG_PER_VMID_CNTL, EXCP_EN); + + if (REG_GET_FIELD(mask, SPI_GDBG_PER_VMID_CNTL, TRAP_ON_START)) + ret |= KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_START; + + if (REG_GET_FIELD(mask, SPI_GDBG_PER_VMID_CNTL, TRAP_ON_END)) + ret |= KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_END; + + return ret; +} + +/* Returns TRAP_EN, EXCP_EN and EXCP_REPLACE. */ +static uint32_t kgd_gfx_v11_set_wave_launch_trap_override(struct amdgpu_device *adev, + uint32_t vmid, + uint32_t trap_override, + uint32_t trap_mask_bits, + uint32_t trap_mask_request, + uint32_t *trap_mask_prev, + uint32_t kfd_dbg_trap_cntl_prev) +{ + uint32_t data = 0; + + *trap_mask_prev = trap_mask_map_hw_to_sw(kfd_dbg_trap_cntl_prev); + + data = (trap_mask_bits & trap_mask_request) | (*trap_mask_prev & ~trap_mask_request); + data = trap_mask_map_sw_to_hw(data); + + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, trap_override); + + return data; +} + +static uint32_t kgd_gfx_v11_set_wave_launch_mode(struct amdgpu_device *adev, + uint8_t wave_launch_mode, + uint32_t vmid) +{ + uint32_t data = 0; + + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, LAUNCH_MODE, wave_launch_mode); + + return data; +} + +#define TCP_WATCH_STRIDE (regTCP_WATCH1_ADDR_H - regTCP_WATCH0_ADDR_H) +static uint32_t kgd_gfx_v11_set_address_watch(struct amdgpu_device *adev, + uint64_t watch_address, + uint32_t watch_address_mask, + uint32_t watch_id, + uint32_t watch_mode, + uint32_t debug_vmid) +{ + uint32_t watch_address_high; + uint32_t watch_address_low; + uint32_t watch_address_cntl; + + watch_address_cntl = 0; + watch_address_low = lower_32_bits(watch_address); + watch_address_high = upper_32_bits(watch_address) & 0xffff; + + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + MODE, + watch_mode); + + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + MASK, + watch_address_mask >> 7); + + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + VALID, + 1); + + WREG32_RLC((SOC15_REG_OFFSET(GC, 0, regTCP_WATCH0_ADDR_H) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_high); + + WREG32_RLC((SOC15_REG_OFFSET(GC, 0, regTCP_WATCH0_ADDR_L) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_low); + + return watch_address_cntl; +} + +static uint32_t kgd_gfx_v11_clear_address_watch(struct amdgpu_device *adev, + uint32_t watch_id) +{ + return 0; +} + const struct kfd2kgd_calls gfx_v11_kfd2kgd = { .program_sh_mem_settings = program_sh_mem_settings_v11, .set_pasid_vmid_mapping = set_pasid_vmid_mapping_v11, @@ -622,4 +801,11 @@ const struct kfd2kgd_calls gfx_v11_kfd2kgd = { .wave_control_execute = wave_control_execute_v11, .get_atc_vmid_pasid_mapping_info = NULL, .set_vm_context_page_table_base = set_vm_context_page_table_base_v11, + .enable_debug_trap = kgd_gfx_v11_enable_debug_trap, + .disable_debug_trap = kgd_gfx_v11_disable_debug_trap, + .validate_trap_override_request = kgd_gfx_v11_validate_trap_override_request, + .set_wave_launch_trap_override = kgd_gfx_v11_set_wave_launch_trap_override, + .set_wave_launch_mode = kgd_gfx_v11_set_wave_launch_mode, + .set_address_watch = kgd_gfx_v11_set_address_watch, + .clear_address_watch = kgd_gfx_v11_clear_address_watch }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c index e83cb1c09610c49f7ef8f020881567d79f62e254..6bf448ab3dffc1a9a40dd1658e19ae6cb8328fb4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c @@ -78,7 +78,7 @@ static void kgd_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmi uint32_t sh_mem_config, uint32_t sh_mem_ape1_base, uint32_t sh_mem_ape1_limit, - uint32_t sh_mem_bases) + uint32_t sh_mem_bases, uint32_t inst) { lock_srbm(adev, 0, 0, 0, vmid); @@ -91,7 +91,7 @@ static void kgd_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmi } static int kgd_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid, - unsigned int vmid) + unsigned int vmid, uint32_t inst) { /* * We have to assume that there is no outstanding mapping. @@ -114,7 +114,8 @@ static int kgd_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid, return 0; } -static int kgd_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id) +static int kgd_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id, + uint32_t inst) { uint32_t mec; uint32_t pipe; @@ -158,7 +159,7 @@ static inline struct cik_sdma_rlc_registers *get_sdma_mqd(void *mqd) static int kgd_hqd_load(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, uint32_t queue_id, uint32_t __user *wptr, uint32_t wptr_shift, - uint32_t wptr_mask, struct mm_struct *mm) + uint32_t wptr_mask, struct mm_struct *mm, uint32_t inst) { struct cik_mqd *m; uint32_t *mqd_hqd; @@ -202,7 +203,7 @@ static int kgd_hqd_load(struct amdgpu_device *adev, void *mqd, static int kgd_hqd_dump(struct amdgpu_device *adev, uint32_t pipe_id, uint32_t queue_id, - uint32_t (**dump)[2], uint32_t *n_regs) + uint32_t (**dump)[2], uint32_t *n_regs, uint32_t inst) { uint32_t i = 0, reg; #define HQD_N_REGS (35+4) @@ -318,7 +319,7 @@ static int kgd_hqd_sdma_dump(struct amdgpu_device *adev, static bool kgd_hqd_is_occupied(struct amdgpu_device *adev, uint64_t queue_address, uint32_t pipe_id, - uint32_t queue_id) + uint32_t queue_id, uint32_t inst) { uint32_t act; bool retval = false; @@ -358,7 +359,7 @@ static bool kgd_hqd_sdma_is_occupied(struct amdgpu_device *adev, void *mqd) static int kgd_hqd_destroy(struct amdgpu_device *adev, void *mqd, enum kfd_preempt_type reset_type, unsigned int utimeout, uint32_t pipe_id, - uint32_t queue_id) + uint32_t queue_id, uint32_t inst) { uint32_t temp; enum hqd_dequeue_request_type type; @@ -494,7 +495,7 @@ static int kgd_hqd_sdma_destroy(struct amdgpu_device *adev, void *mqd, static int kgd_wave_control_execute(struct amdgpu_device *adev, uint32_t gfx_index_val, - uint32_t sq_cmd) + uint32_t sq_cmd, uint32_t inst) { uint32_t data; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c index 870f352837fc6cdc4ea58f4e8d3412a97331cda6..cd06e4a6d1da4825e353752406c156337fa76577 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c @@ -72,7 +72,7 @@ static void kgd_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmi uint32_t sh_mem_config, uint32_t sh_mem_ape1_base, uint32_t sh_mem_ape1_limit, - uint32_t sh_mem_bases) + uint32_t sh_mem_bases, uint32_t inst) { lock_srbm(adev, 0, 0, 0, vmid); @@ -85,7 +85,7 @@ static void kgd_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmi } static int kgd_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid, - unsigned int vmid) + unsigned int vmid, uint32_t inst) { /* * We have to assume that there is no outstanding mapping. @@ -109,7 +109,8 @@ static int kgd_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid, return 0; } -static int kgd_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id) +static int kgd_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id, + uint32_t inst) { uint32_t mec; uint32_t pipe; @@ -153,7 +154,7 @@ static inline struct vi_sdma_mqd *get_sdma_mqd(void *mqd) static int kgd_hqd_load(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, uint32_t queue_id, uint32_t __user *wptr, uint32_t wptr_shift, - uint32_t wptr_mask, struct mm_struct *mm) + uint32_t wptr_mask, struct mm_struct *mm, uint32_t inst) { struct vi_mqd *m; uint32_t *mqd_hqd; @@ -226,7 +227,7 @@ static int kgd_hqd_load(struct amdgpu_device *adev, void *mqd, static int kgd_hqd_dump(struct amdgpu_device *adev, uint32_t pipe_id, uint32_t queue_id, - uint32_t (**dump)[2], uint32_t *n_regs) + uint32_t (**dump)[2], uint32_t *n_regs, uint32_t inst) { uint32_t i = 0, reg; #define HQD_N_REGS (54+4) @@ -350,7 +351,7 @@ static int kgd_hqd_sdma_dump(struct amdgpu_device *adev, static bool kgd_hqd_is_occupied(struct amdgpu_device *adev, uint64_t queue_address, uint32_t pipe_id, - uint32_t queue_id) + uint32_t queue_id, uint32_t inst) { uint32_t act; bool retval = false; @@ -390,7 +391,7 @@ static bool kgd_hqd_sdma_is_occupied(struct amdgpu_device *adev, void *mqd) static int kgd_hqd_destroy(struct amdgpu_device *adev, void *mqd, enum kfd_preempt_type reset_type, unsigned int utimeout, uint32_t pipe_id, - uint32_t queue_id) + uint32_t queue_id, uint32_t inst) { uint32_t temp; enum hqd_dequeue_request_type type; @@ -540,7 +541,7 @@ static bool get_atc_vmid_pasid_mapping_info(struct amdgpu_device *adev, static int kgd_wave_control_execute(struct amdgpu_device *adev, uint32_t gfx_index_val, - uint32_t sq_cmd) + uint32_t sq_cmd, uint32_t inst) { uint32_t data = 0; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c index e92b93557c13fa382a6ee4d66d4e43ef96b842d8..51d93fb13ea3bf95a4c3ab909d96fc5cc832a075 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c @@ -38,6 +38,7 @@ #include "soc15d.h" #include "gfx_v9_0.h" #include "amdgpu_amdkfd_gfx_v9.h" +#include enum hqd_dequeue_request_type { NO_ACTION = 0, @@ -46,29 +47,29 @@ enum hqd_dequeue_request_type { SAVE_WAVES }; -static void lock_srbm(struct amdgpu_device *adev, uint32_t mec, uint32_t pipe, - uint32_t queue, uint32_t vmid) +static void kgd_gfx_v9_lock_srbm(struct amdgpu_device *adev, uint32_t mec, uint32_t pipe, + uint32_t queue, uint32_t vmid, uint32_t inst) { mutex_lock(&adev->srbm_mutex); - soc15_grbm_select(adev, mec, pipe, queue, vmid); + soc15_grbm_select(adev, mec, pipe, queue, vmid, GET_INST(GC, inst)); } -static void unlock_srbm(struct amdgpu_device *adev) +static void kgd_gfx_v9_unlock_srbm(struct amdgpu_device *adev, uint32_t inst) { - soc15_grbm_select(adev, 0, 0, 0, 0); + soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, inst)); mutex_unlock(&adev->srbm_mutex); } -static void acquire_queue(struct amdgpu_device *adev, uint32_t pipe_id, - uint32_t queue_id) +void kgd_gfx_v9_acquire_queue(struct amdgpu_device *adev, uint32_t pipe_id, + uint32_t queue_id, uint32_t inst) { uint32_t mec = (pipe_id / adev->gfx.mec.num_pipe_per_mec) + 1; uint32_t pipe = (pipe_id % adev->gfx.mec.num_pipe_per_mec); - lock_srbm(adev, mec, pipe, queue_id, 0); + kgd_gfx_v9_lock_srbm(adev, mec, pipe, queue_id, 0, inst); } -static uint64_t get_queue_mask(struct amdgpu_device *adev, +uint64_t kgd_gfx_v9_get_queue_mask(struct amdgpu_device *adev, uint32_t pipe_id, uint32_t queue_id) { unsigned int bit = pipe_id * adev->gfx.mec.num_queue_per_pipe + @@ -77,28 +78,28 @@ static uint64_t get_queue_mask(struct amdgpu_device *adev, return 1ull << bit; } -static void release_queue(struct amdgpu_device *adev) +void kgd_gfx_v9_release_queue(struct amdgpu_device *adev, uint32_t inst) { - unlock_srbm(adev); + kgd_gfx_v9_unlock_srbm(adev, inst); } void kgd_gfx_v9_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmid, uint32_t sh_mem_config, uint32_t sh_mem_ape1_base, uint32_t sh_mem_ape1_limit, - uint32_t sh_mem_bases) + uint32_t sh_mem_bases, uint32_t inst) { - lock_srbm(adev, 0, 0, 0, vmid); + kgd_gfx_v9_lock_srbm(adev, 0, 0, 0, vmid, inst); - WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmSH_MEM_CONFIG), sh_mem_config); - WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmSH_MEM_BASES), sh_mem_bases); + WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmSH_MEM_CONFIG), sh_mem_config); + WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmSH_MEM_BASES), sh_mem_bases); /* APE1 no longer exists on GFX9 */ - unlock_srbm(adev); + kgd_gfx_v9_unlock_srbm(adev, inst); } int kgd_gfx_v9_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid, - unsigned int vmid) + unsigned int vmid, uint32_t inst) { /* * We have to assume that there is no outstanding mapping. @@ -156,7 +157,8 @@ int kgd_gfx_v9_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid, * but still works */ -int kgd_gfx_v9_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id) +int kgd_gfx_v9_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id, + uint32_t inst) { uint32_t mec; uint32_t pipe; @@ -164,13 +166,13 @@ int kgd_gfx_v9_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id) mec = (pipe_id / adev->gfx.mec.num_pipe_per_mec) + 1; pipe = (pipe_id % adev->gfx.mec.num_pipe_per_mec); - lock_srbm(adev, mec, pipe, 0, 0); + kgd_gfx_v9_lock_srbm(adev, mec, pipe, 0, 0, inst); - WREG32_SOC15(GC, 0, mmCPC_INT_CNTL, + WREG32_SOC15(GC, GET_INST(GC, inst), mmCPC_INT_CNTL, CP_INT_CNTL_RING0__TIME_STAMP_INT_ENABLE_MASK | CP_INT_CNTL_RING0__OPCODE_ERROR_INT_ENABLE_MASK); - unlock_srbm(adev); + kgd_gfx_v9_unlock_srbm(adev, inst); return 0; } @@ -220,7 +222,8 @@ static inline struct v9_sdma_mqd *get_sdma_mqd(void *mqd) int kgd_gfx_v9_hqd_load(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, uint32_t queue_id, uint32_t __user *wptr, uint32_t wptr_shift, - uint32_t wptr_mask, struct mm_struct *mm) + uint32_t wptr_mask, struct mm_struct *mm, + uint32_t inst) { struct v9_mqd *m; uint32_t *mqd_hqd; @@ -228,21 +231,22 @@ int kgd_gfx_v9_hqd_load(struct amdgpu_device *adev, void *mqd, m = get_mqd(mqd); - acquire_queue(adev, pipe_id, queue_id); + kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst); /* HQD registers extend from CP_MQD_BASE_ADDR to CP_HQD_EOP_WPTR_MEM. */ mqd_hqd = &m->cp_mqd_base_addr_lo; - hqd_base = SOC15_REG_OFFSET(GC, 0, mmCP_MQD_BASE_ADDR); + hqd_base = SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_MQD_BASE_ADDR); for (reg = hqd_base; - reg <= SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_HI); reg++) + reg <= SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_HI); reg++) WREG32_RLC(reg, mqd_hqd[reg - hqd_base]); /* Activate doorbell logic before triggering WPTR poll. */ data = REG_SET_FIELD(m->cp_hqd_pq_doorbell_control, CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 1); - WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_DOORBELL_CONTROL), data); + WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_DOORBELL_CONTROL), + data); if (wptr) { /* Don't read wptr with get_user because the user @@ -271,43 +275,43 @@ int kgd_gfx_v9_hqd_load(struct amdgpu_device *adev, void *mqd, guessed_wptr += m->cp_hqd_pq_wptr_lo & ~(queue_size - 1); guessed_wptr += (uint64_t)m->cp_hqd_pq_wptr_hi << 32; - WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_LO), + WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_LO), lower_32_bits(guessed_wptr)); - WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_HI), + WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_HI), upper_32_bits(guessed_wptr)); - WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_POLL_ADDR), + WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_POLL_ADDR), lower_32_bits((uintptr_t)wptr)); - WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_POLL_ADDR_HI), + WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_POLL_ADDR_HI), upper_32_bits((uintptr_t)wptr)); - WREG32_SOC15(GC, 0, mmCP_PQ_WPTR_POLL_CNTL1, - (uint32_t)get_queue_mask(adev, pipe_id, queue_id)); + WREG32_SOC15(GC, GET_INST(GC, inst), mmCP_PQ_WPTR_POLL_CNTL1, + (uint32_t)kgd_gfx_v9_get_queue_mask(adev, pipe_id, queue_id)); } /* Start the EOP fetcher */ - WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_EOP_RPTR), + WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_EOP_RPTR), REG_SET_FIELD(m->cp_hqd_eop_rptr, CP_HQD_EOP_RPTR, INIT_FETCHER, 1)); data = REG_SET_FIELD(m->cp_hqd_active, CP_HQD_ACTIVE, ACTIVE, 1); - WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_ACTIVE), data); + WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_ACTIVE), data); - release_queue(adev); + kgd_gfx_v9_release_queue(adev, inst); return 0; } int kgd_gfx_v9_hiq_mqd_load(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, uint32_t queue_id, - uint32_t doorbell_off) + uint32_t doorbell_off, uint32_t inst) { - struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring; + struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[inst].ring; struct v9_mqd *m; uint32_t mec, pipe; int r; m = get_mqd(mqd); - acquire_queue(adev, pipe_id, queue_id); + kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst); mec = (pipe_id / adev->gfx.mec.num_pipe_per_mec) + 1; pipe = (pipe_id % adev->gfx.mec.num_pipe_per_mec); @@ -315,7 +319,7 @@ int kgd_gfx_v9_hiq_mqd_load(struct amdgpu_device *adev, void *mqd, pr_debug("kfd: set HIQ, mec:%d, pipe:%d, queue:%d.\n", mec, pipe, queue_id); - spin_lock(&adev->gfx.kiq.ring_lock); + spin_lock(&adev->gfx.kiq[inst].ring_lock); r = amdgpu_ring_alloc(kiq_ring, 7); if (r) { pr_err("Failed to alloc KIQ (%d).\n", r); @@ -342,15 +346,15 @@ int kgd_gfx_v9_hiq_mqd_load(struct amdgpu_device *adev, void *mqd, amdgpu_ring_commit(kiq_ring); out_unlock: - spin_unlock(&adev->gfx.kiq.ring_lock); - release_queue(adev); + spin_unlock(&adev->gfx.kiq[inst].ring_lock); + kgd_gfx_v9_release_queue(adev, inst); return r; } int kgd_gfx_v9_hqd_dump(struct amdgpu_device *adev, uint32_t pipe_id, uint32_t queue_id, - uint32_t (**dump)[2], uint32_t *n_regs) + uint32_t (**dump)[2], uint32_t *n_regs, uint32_t inst) { uint32_t i = 0, reg; #define HQD_N_REGS 56 @@ -365,13 +369,13 @@ int kgd_gfx_v9_hqd_dump(struct amdgpu_device *adev, if (*dump == NULL) return -ENOMEM; - acquire_queue(adev, pipe_id, queue_id); + kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst); - for (reg = SOC15_REG_OFFSET(GC, 0, mmCP_MQD_BASE_ADDR); - reg <= SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_HI); reg++) + for (reg = SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_MQD_BASE_ADDR); + reg <= SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_HI); reg++) DUMP_REG(reg); - release_queue(adev); + kgd_gfx_v9_release_queue(adev, inst); WARN_ON_ONCE(i != HQD_N_REGS); *n_regs = i; @@ -481,23 +485,23 @@ static int kgd_hqd_sdma_dump(struct amdgpu_device *adev, bool kgd_gfx_v9_hqd_is_occupied(struct amdgpu_device *adev, uint64_t queue_address, uint32_t pipe_id, - uint32_t queue_id) + uint32_t queue_id, uint32_t inst) { uint32_t act; bool retval = false; uint32_t low, high; - acquire_queue(adev, pipe_id, queue_id); - act = RREG32_SOC15(GC, 0, mmCP_HQD_ACTIVE); + kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst); + act = RREG32_SOC15(GC, GET_INST(GC, inst), mmCP_HQD_ACTIVE); if (act) { low = lower_32_bits(queue_address >> 8); high = upper_32_bits(queue_address >> 8); - if (low == RREG32_SOC15(GC, 0, mmCP_HQD_PQ_BASE) && - high == RREG32_SOC15(GC, 0, mmCP_HQD_PQ_BASE_HI)) + if (low == RREG32_SOC15(GC, GET_INST(GC, inst), mmCP_HQD_PQ_BASE) && + high == RREG32_SOC15(GC, GET_INST(GC, inst), mmCP_HQD_PQ_BASE_HI)) retval = true; } - release_queue(adev); + kgd_gfx_v9_release_queue(adev, inst); return retval; } @@ -522,7 +526,7 @@ static bool kgd_hqd_sdma_is_occupied(struct amdgpu_device *adev, void *mqd) int kgd_gfx_v9_hqd_destroy(struct amdgpu_device *adev, void *mqd, enum kfd_preempt_type reset_type, unsigned int utimeout, uint32_t pipe_id, - uint32_t queue_id) + uint32_t queue_id, uint32_t inst) { enum hqd_dequeue_request_type type; unsigned long end_jiffies; @@ -532,10 +536,10 @@ int kgd_gfx_v9_hqd_destroy(struct amdgpu_device *adev, void *mqd, if (amdgpu_in_reset(adev)) return -EIO; - acquire_queue(adev, pipe_id, queue_id); + kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst); if (m->cp_hqd_vmid == 0) - WREG32_FIELD15_RLC(GC, 0, RLC_CP_SCHEDULERS, scheduler1, 0); + WREG32_FIELD15_RLC(GC, GET_INST(GC, inst), RLC_CP_SCHEDULERS, scheduler1, 0); switch (reset_type) { case KFD_PREEMPT_TYPE_WAVEFRONT_DRAIN: @@ -552,22 +556,22 @@ int kgd_gfx_v9_hqd_destroy(struct amdgpu_device *adev, void *mqd, break; } - WREG32_RLC(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_DEQUEUE_REQUEST), type); + WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_DEQUEUE_REQUEST), type); end_jiffies = (utimeout * HZ / 1000) + jiffies; while (true) { - temp = RREG32_SOC15(GC, 0, mmCP_HQD_ACTIVE); + temp = RREG32_SOC15(GC, GET_INST(GC, inst), mmCP_HQD_ACTIVE); if (!(temp & CP_HQD_ACTIVE__ACTIVE_MASK)) break; if (time_after(jiffies, end_jiffies)) { pr_err("cp queue preemption time out.\n"); - release_queue(adev); + kgd_gfx_v9_release_queue(adev, inst); return -ETIME; } usleep_range(500, 1000); } - release_queue(adev); + kgd_gfx_v9_release_queue(adev, inst); return 0; } @@ -624,14 +628,14 @@ bool kgd_gfx_v9_get_atc_vmid_pasid_mapping_info(struct amdgpu_device *adev, int kgd_gfx_v9_wave_control_execute(struct amdgpu_device *adev, uint32_t gfx_index_val, - uint32_t sq_cmd) + uint32_t sq_cmd, uint32_t inst) { uint32_t data = 0; mutex_lock(&adev->grbm_idx_mutex); - WREG32_SOC15_RLC_SHADOW(GC, 0, mmGRBM_GFX_INDEX, gfx_index_val); - WREG32_SOC15(GC, 0, mmSQ_CMD, sq_cmd); + WREG32_SOC15_RLC_SHADOW(GC, GET_INST(GC, inst), mmGRBM_GFX_INDEX, gfx_index_val); + WREG32_SOC15(GC, GET_INST(GC, inst), mmSQ_CMD, sq_cmd); data = REG_SET_FIELD(data, GRBM_GFX_INDEX, INSTANCE_BROADCAST_WRITES, 1); @@ -640,12 +644,271 @@ int kgd_gfx_v9_wave_control_execute(struct amdgpu_device *adev, data = REG_SET_FIELD(data, GRBM_GFX_INDEX, SE_BROADCAST_WRITES, 1); - WREG32_SOC15_RLC_SHADOW(GC, 0, mmGRBM_GFX_INDEX, data); + WREG32_SOC15_RLC_SHADOW(GC, GET_INST(GC, inst), mmGRBM_GFX_INDEX, data); mutex_unlock(&adev->grbm_idx_mutex); return 0; } +/* + * GFX9 helper for wave launch stall requirements on debug trap setting. + * + * vmid: + * Target VMID to stall/unstall. + * + * stall: + * 0-unstall wave launch (enable), 1-stall wave launch (disable). + * After wavefront launch has been stalled, allocated waves must drain from + * SPI in order for debug trap settings to take effect on those waves. + * This is roughly a ~96 clock cycle wait on SPI where a read on + * SPI_GDBG_WAVE_CNTL translates to ~32 clock cycles. + * KGD_GFX_V9_WAVE_LAUNCH_SPI_DRAIN_LATENCY indicates the number of reads required. + * + * NOTE: We can afford to clear the entire STALL_VMID field on unstall + * because GFX9.4.1 cannot support multi-process debugging due to trap + * configuration and masking being limited to global scope. Always assume + * single process conditions. + */ +#define KGD_GFX_V9_WAVE_LAUNCH_SPI_DRAIN_LATENCY 3 +void kgd_gfx_v9_set_wave_launch_stall(struct amdgpu_device *adev, + uint32_t vmid, + bool stall) +{ + int i; + uint32_t data = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL)); + + if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 1)) + data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL, STALL_VMID, + stall ? 1 << vmid : 0); + else + data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL, STALL_RA, + stall ? 1 : 0); + + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), data); + + if (!stall) + return; + + for (i = 0; i < KGD_GFX_V9_WAVE_LAUNCH_SPI_DRAIN_LATENCY; i++) + RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL)); +} + +/* + * restore_dbg_registers is ignored here but is a general interface requirement + * for devices that support GFXOFF and where the RLC save/restore list + * does not support hw registers for debugging i.e. the driver has to manually + * initialize the debug mode registers after it has disabled GFX off during the + * debug session. + */ +uint32_t kgd_gfx_v9_enable_debug_trap(struct amdgpu_device *adev, + bool restore_dbg_registers, + uint32_t vmid) +{ + mutex_lock(&adev->grbm_idx_mutex); + + kgd_gfx_v9_set_wave_launch_stall(adev, vmid, true); + + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0); + + kgd_gfx_v9_set_wave_launch_stall(adev, vmid, false); + + mutex_unlock(&adev->grbm_idx_mutex); + + return 0; +} + +/* + * keep_trap_enabled is ignored here but is a general interface requirement + * for devices that support multi-process debugging where the performance + * overhead from trap temporary setup needs to be bypassed when the debug + * session has ended. + */ +uint32_t kgd_gfx_v9_disable_debug_trap(struct amdgpu_device *adev, + bool keep_trap_enabled, + uint32_t vmid) +{ + mutex_lock(&adev->grbm_idx_mutex); + + kgd_gfx_v9_set_wave_launch_stall(adev, vmid, true); + + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0); + + kgd_gfx_v9_set_wave_launch_stall(adev, vmid, false); + + mutex_unlock(&adev->grbm_idx_mutex); + + return 0; +} + +int kgd_gfx_v9_validate_trap_override_request(struct amdgpu_device *adev, + uint32_t trap_override, + uint32_t *trap_mask_supported) +{ + *trap_mask_supported &= KFD_DBG_TRAP_MASK_DBG_ADDRESS_WATCH; + + /* The SPI_GDBG_TRAP_MASK register is global and affects all + * processes. Only allow OR-ing the address-watch bit, since + * this only affects processes under the debugger. Other bits + * should stay 0 to avoid the debugger interfering with other + * processes. + */ + if (trap_override != KFD_DBG_TRAP_OVERRIDE_OR) + return -EINVAL; + + return 0; +} + +uint32_t kgd_gfx_v9_set_wave_launch_trap_override(struct amdgpu_device *adev, + uint32_t vmid, + uint32_t trap_override, + uint32_t trap_mask_bits, + uint32_t trap_mask_request, + uint32_t *trap_mask_prev, + uint32_t kfd_dbg_cntl_prev) +{ + uint32_t data, wave_cntl_prev; + + mutex_lock(&adev->grbm_idx_mutex); + + wave_cntl_prev = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL)); + + kgd_gfx_v9_set_wave_launch_stall(adev, vmid, true); + + data = RREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK)); + *trap_mask_prev = REG_GET_FIELD(data, SPI_GDBG_TRAP_MASK, EXCP_EN); + + trap_mask_bits = (trap_mask_bits & trap_mask_request) | + (*trap_mask_prev & ~trap_mask_request); + + data = REG_SET_FIELD(data, SPI_GDBG_TRAP_MASK, EXCP_EN, trap_mask_bits); + data = REG_SET_FIELD(data, SPI_GDBG_TRAP_MASK, REPLACE, trap_override); + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), data); + + /* We need to preserve wave launch mode stall settings. */ + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL), wave_cntl_prev); + + mutex_unlock(&adev->grbm_idx_mutex); + + return 0; +} + +uint32_t kgd_gfx_v9_set_wave_launch_mode(struct amdgpu_device *adev, + uint8_t wave_launch_mode, + uint32_t vmid) +{ + uint32_t data = 0; + bool is_mode_set = !!wave_launch_mode; + + mutex_lock(&adev->grbm_idx_mutex); + + kgd_gfx_v9_set_wave_launch_stall(adev, vmid, true); + + data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL2, + VMID_MASK, is_mode_set ? 1 << vmid : 0); + data = REG_SET_FIELD(data, SPI_GDBG_WAVE_CNTL2, + MODE, is_mode_set ? wave_launch_mode : 0); + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_WAVE_CNTL2), data); + + kgd_gfx_v9_set_wave_launch_stall(adev, vmid, false); + + mutex_unlock(&adev->grbm_idx_mutex); + + return 0; +} + +#define TCP_WATCH_STRIDE (mmTCP_WATCH1_ADDR_H - mmTCP_WATCH0_ADDR_H) +uint32_t kgd_gfx_v9_set_address_watch(struct amdgpu_device *adev, + uint64_t watch_address, + uint32_t watch_address_mask, + uint32_t watch_id, + uint32_t watch_mode, + uint32_t debug_vmid) +{ + uint32_t watch_address_high; + uint32_t watch_address_low; + uint32_t watch_address_cntl; + + watch_address_cntl = 0; + + watch_address_low = lower_32_bits(watch_address); + watch_address_high = upper_32_bits(watch_address) & 0xffff; + + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + VMID, + debug_vmid); + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + MODE, + watch_mode); + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + MASK, + watch_address_mask >> 6); + + /* Turning off this watch point until we set all the registers */ + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + VALID, + 0); + + WREG32_RLC((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_CNTL) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_cntl); + + WREG32_RLC((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_ADDR_H) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_high); + + WREG32_RLC((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_ADDR_L) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_low); + + /* Enable the watch point */ + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + VALID, + 1); + + WREG32_RLC((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_CNTL) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_cntl); + + return 0; +} + +uint32_t kgd_gfx_v9_clear_address_watch(struct amdgpu_device *adev, + uint32_t watch_id) +{ + uint32_t watch_address_cntl; + + watch_address_cntl = 0; + + WREG32_RLC((SOC15_REG_OFFSET(GC, 0, mmTCP_WATCH0_CNTL) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_cntl); + + return 0; +} + +/* kgd_gfx_v9_get_iq_wait_times: Returns the mmCP_IQ_WAIT_TIME1/2 values + * The values read are: + * ib_offload_wait_time -- Wait Count for Indirect Buffer Offloads. + * atomic_offload_wait_time -- Wait Count for L2 and GDS Atomics Offloads. + * wrm_offload_wait_time -- Wait Count for WAIT_REG_MEM Offloads. + * gws_wait_time -- Wait Count for Global Wave Syncs. + * que_sleep_wait_time -- Wait Count for Dequeue Retry. + * sch_wave_wait_time -- Wait Count for Scheduling Wave Message. + * sem_rearm_wait_time -- Wait Count for Semaphore re-arm. + * deq_retry_wait_time -- Wait Count for Global Wave Syncs. + */ +void kgd_gfx_v9_get_iq_wait_times(struct amdgpu_device *adev, + uint32_t *wait_times) + +{ + *wait_times = RREG32(SOC15_REG_OFFSET(GC, 0, mmCP_IQ_WAIT_TIME2)); +} + void kgd_gfx_v9_set_vm_context_page_table_base(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { @@ -682,10 +945,11 @@ static void unlock_spi_csq_mutexes(struct amdgpu_device *adev) * @queue_idx: Index of queue in the queue-map bit-field * @wave_cnt: Output parameter updated with number of waves in flight * @vmid: Output parameter updated with VMID of queue whose wave count - * is being collected + * is being collected + * @inst: xcc's instance number on a multi-XCC setup */ static void get_wave_count(struct amdgpu_device *adev, int queue_idx, - int *wave_cnt, int *vmid) + int *wave_cnt, int *vmid, uint32_t inst) { int pipe_idx; int queue_slot; @@ -700,12 +964,12 @@ static void get_wave_count(struct amdgpu_device *adev, int queue_idx, *wave_cnt = 0; pipe_idx = queue_idx / adev->gfx.mec.num_queue_per_pipe; queue_slot = queue_idx % adev->gfx.mec.num_queue_per_pipe; - soc15_grbm_select(adev, 1, pipe_idx, queue_slot, 0); - reg_val = RREG32_SOC15_IP(GC, SOC15_REG_OFFSET(GC, 0, mmSPI_CSQ_WF_ACTIVE_COUNT_0) + + soc15_grbm_select(adev, 1, pipe_idx, queue_slot, 0, inst); + reg_val = RREG32_SOC15_IP(GC, SOC15_REG_OFFSET(GC, inst, mmSPI_CSQ_WF_ACTIVE_COUNT_0) + queue_slot); *wave_cnt = reg_val & SPI_CSQ_WF_ACTIVE_COUNT_0__COUNT_MASK; if (*wave_cnt != 0) - *vmid = (RREG32_SOC15(GC, 0, mmCP_HQD_VMID) & + *vmid = (RREG32_SOC15(GC, inst, mmCP_HQD_VMID) & CP_HQD_VMID__VMID_MASK) >> CP_HQD_VMID__VMID__SHIFT; } @@ -718,9 +982,10 @@ static void get_wave_count(struct amdgpu_device *adev, int queue_idx, * @adev: Handle of device from which to get number of waves in flight * @pasid: Identifies the process for which this query call is invoked * @pasid_wave_cnt: Output parameter updated with number of waves in flight that - * belong to process with given pasid + * belong to process with given pasid * @max_waves_per_cu: Output parameter updated with maximum number of waves - * possible per Compute Unit + * possible per Compute Unit + * @inst: xcc's instance number on a multi-XCC setup * * Note: It's possible that the device has too many queues (oversubscription) * in which case a VMID could be remapped to a different PASID. This could lead @@ -756,7 +1021,7 @@ static void get_wave_count(struct amdgpu_device *adev, int queue_idx, * Reading registers referenced above involves programming GRBM appropriately */ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid, - int *pasid_wave_cnt, int *max_waves_per_cu) + int *pasid_wave_cnt, int *max_waves_per_cu, uint32_t inst) { int qidx; int vmid; @@ -772,13 +1037,13 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid, DECLARE_BITMAP(cp_queue_bitmap, KGD_MAX_QUEUES); lock_spi_csq_mutexes(adev); - soc15_grbm_select(adev, 1, 0, 0, 0); + soc15_grbm_select(adev, 1, 0, 0, 0, inst); /* * Iterate through the shader engines and arrays of the device * to get number of waves in flight */ - bitmap_complement(cp_queue_bitmap, adev->gfx.mec.queue_bitmap, + bitmap_complement(cp_queue_bitmap, adev->gfx.mec_bitmap[0].queue_bitmap, KGD_MAX_QUEUES); max_queue_cnt = adev->gfx.mec.num_pipe_per_mec * adev->gfx.mec.num_queue_per_pipe; @@ -787,8 +1052,8 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid, for (se_idx = 0; se_idx < se_cnt; se_idx++) { for (sh_idx = 0; sh_idx < sh_cnt; sh_idx++) { - amdgpu_gfx_select_se_sh(adev, se_idx, sh_idx, 0xffffffff); - queue_map = RREG32_SOC15(GC, 0, mmSPI_CSQ_WF_ACTIVE_STATUS); + amdgpu_gfx_select_se_sh(adev, se_idx, sh_idx, 0xffffffff, inst); + queue_map = RREG32_SOC15(GC, inst, mmSPI_CSQ_WF_ACTIVE_STATUS); /* * Assumption: queue map encodes following schema: four @@ -808,10 +1073,11 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid, continue; /* Get number of waves in flight and aggregate them */ - get_wave_count(adev, qidx, &wave_cnt, &vmid); + get_wave_count(adev, qidx, &wave_cnt, &vmid, + inst); if (wave_cnt != 0) { pasid_tmp = - RREG32(SOC15_REG_OFFSET(OSSSYS, 0, + RREG32(SOC15_REG_OFFSET(OSSSYS, inst, mmIH_VMID_0_LUT) + vmid); if (pasid_tmp == pasid) vmid_wave_cnt += wave_cnt; @@ -820,8 +1086,8 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid, } } - amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); - soc15_grbm_select(adev, 0, 0, 0, 0); + amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, inst); + soc15_grbm_select(adev, 0, 0, 0, 0, inst); unlock_spi_csq_mutexes(adev); /* Update the output parameters and return */ @@ -830,28 +1096,51 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid, adev->gfx.cu_info.max_waves_per_simd; } +void kgd_gfx_v9_build_grace_period_packet_info(struct amdgpu_device *adev, + uint32_t wait_times, + uint32_t grace_period, + uint32_t *reg_offset, + uint32_t *reg_data) +{ + *reg_data = wait_times; + + /* + * The CP cannont handle a 0 grace period input and will result in + * an infinite grace period being set so set to 1 to prevent this. + */ + if (grace_period == 0) + grace_period = 1; + + *reg_data = REG_SET_FIELD(*reg_data, + CP_IQ_WAIT_TIME2, + SCH_WAVE, + grace_period); + + *reg_offset = SOC15_REG_OFFSET(GC, 0, mmCP_IQ_WAIT_TIME2); +} + void kgd_gfx_v9_program_trap_handler_settings(struct amdgpu_device *adev, - uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr) + uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr, uint32_t inst) { - lock_srbm(adev, 0, 0, 0, vmid); + kgd_gfx_v9_lock_srbm(adev, 0, 0, 0, vmid, inst); /* * Program TBA registers */ - WREG32_SOC15(GC, 0, mmSQ_SHADER_TBA_LO, + WREG32_SOC15(GC, GET_INST(GC, inst), mmSQ_SHADER_TBA_LO, lower_32_bits(tba_addr >> 8)); - WREG32_SOC15(GC, 0, mmSQ_SHADER_TBA_HI, + WREG32_SOC15(GC, GET_INST(GC, inst), mmSQ_SHADER_TBA_HI, upper_32_bits(tba_addr >> 8)); /* * Program TMA registers */ - WREG32_SOC15(GC, 0, mmSQ_SHADER_TMA_LO, + WREG32_SOC15(GC, GET_INST(GC, inst), mmSQ_SHADER_TMA_LO, lower_32_bits(tma_addr >> 8)); - WREG32_SOC15(GC, 0, mmSQ_SHADER_TMA_HI, + WREG32_SOC15(GC, GET_INST(GC, inst), mmSQ_SHADER_TMA_HI, upper_32_bits(tma_addr >> 8)); - unlock_srbm(adev); + kgd_gfx_v9_unlock_srbm(adev, inst); } const struct kfd2kgd_calls gfx_v9_kfd2kgd = { @@ -871,6 +1160,15 @@ const struct kfd2kgd_calls gfx_v9_kfd2kgd = { .get_atc_vmid_pasid_mapping_info = kgd_gfx_v9_get_atc_vmid_pasid_mapping_info, .set_vm_context_page_table_base = kgd_gfx_v9_set_vm_context_page_table_base, + .enable_debug_trap = kgd_gfx_v9_enable_debug_trap, + .disable_debug_trap = kgd_gfx_v9_disable_debug_trap, + .validate_trap_override_request = kgd_gfx_v9_validate_trap_override_request, + .set_wave_launch_trap_override = kgd_gfx_v9_set_wave_launch_trap_override, + .set_wave_launch_mode = kgd_gfx_v9_set_wave_launch_mode, + .set_address_watch = kgd_gfx_v9_set_address_watch, + .clear_address_watch = kgd_gfx_v9_clear_address_watch, + .get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times, + .build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info, .get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy, .program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h index c7ed3bc9053c58fc637ae39cac7a32008dedb388..5f54bff0db496c3fc75da01b8b5d76c9e241d78e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h @@ -20,41 +20,81 @@ * OTHER DEALINGS IN THE SOFTWARE. */ - - void kgd_gfx_v9_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmid, uint32_t sh_mem_config, uint32_t sh_mem_ape1_base, uint32_t sh_mem_ape1_limit, - uint32_t sh_mem_bases); + uint32_t sh_mem_bases, uint32_t inst); int kgd_gfx_v9_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid, - unsigned int vmid); -int kgd_gfx_v9_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id); + unsigned int vmid, uint32_t inst); +int kgd_gfx_v9_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id, + uint32_t inst); int kgd_gfx_v9_hqd_load(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, uint32_t queue_id, uint32_t __user *wptr, uint32_t wptr_shift, uint32_t wptr_mask, - struct mm_struct *mm); + struct mm_struct *mm, uint32_t inst); int kgd_gfx_v9_hiq_mqd_load(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, uint32_t queue_id, - uint32_t doorbell_off); + uint32_t doorbell_off, uint32_t inst); int kgd_gfx_v9_hqd_dump(struct amdgpu_device *adev, uint32_t pipe_id, uint32_t queue_id, - uint32_t (**dump)[2], uint32_t *n_regs); + uint32_t (**dump)[2], uint32_t *n_regs, uint32_t inst); bool kgd_gfx_v9_hqd_is_occupied(struct amdgpu_device *adev, uint64_t queue_address, uint32_t pipe_id, - uint32_t queue_id); + uint32_t queue_id, uint32_t inst); int kgd_gfx_v9_hqd_destroy(struct amdgpu_device *adev, void *mqd, enum kfd_preempt_type reset_type, unsigned int utimeout, uint32_t pipe_id, - uint32_t queue_id); + uint32_t queue_id, uint32_t inst); int kgd_gfx_v9_wave_control_execute(struct amdgpu_device *adev, uint32_t gfx_index_val, - uint32_t sq_cmd); + uint32_t sq_cmd, uint32_t inst); bool kgd_gfx_v9_get_atc_vmid_pasid_mapping_info(struct amdgpu_device *adev, uint8_t vmid, uint16_t *p_pasid); - void kgd_gfx_v9_set_vm_context_page_table_base(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base); void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid, - int *pasid_wave_cnt, int *max_waves_per_cu); + int *pasid_wave_cnt, int *max_waves_per_cu, uint32_t inst); void kgd_gfx_v9_program_trap_handler_settings(struct amdgpu_device *adev, - uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr); + uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr, + uint32_t inst); +void kgd_gfx_v9_acquire_queue(struct amdgpu_device *adev, uint32_t pipe_id, + uint32_t queue_id, uint32_t inst); +uint64_t kgd_gfx_v9_get_queue_mask(struct amdgpu_device *adev, + uint32_t pipe_id, uint32_t queue_id); +void kgd_gfx_v9_release_queue(struct amdgpu_device *adev, uint32_t inst); +void kgd_gfx_v9_set_wave_launch_stall(struct amdgpu_device *adev, + uint32_t vmid, + bool stall); +uint32_t kgd_gfx_v9_enable_debug_trap(struct amdgpu_device *adev, + bool restore_dbg_registers, + uint32_t vmid); +uint32_t kgd_gfx_v9_disable_debug_trap(struct amdgpu_device *adev, + bool keep_trap_enabled, + uint32_t vmid); +int kgd_gfx_v9_validate_trap_override_request(struct amdgpu_device *adev, + uint32_t trap_override, + uint32_t *trap_mask_supported); +uint32_t kgd_gfx_v9_set_wave_launch_mode(struct amdgpu_device *adev, + uint8_t wave_launch_mode, + uint32_t vmid); +uint32_t kgd_gfx_v9_set_wave_launch_trap_override(struct amdgpu_device *adev, + uint32_t vmid, + uint32_t trap_override, + uint32_t trap_mask_bits, + uint32_t trap_mask_request, + uint32_t *trap_mask_prev, + uint32_t kfd_dbg_trap_cntl_prev); +uint32_t kgd_gfx_v9_set_address_watch(struct amdgpu_device *adev, + uint64_t watch_address, + uint32_t watch_address_mask, + uint32_t watch_id, + uint32_t watch_mode, + uint32_t debug_vmid); +uint32_t kgd_gfx_v9_clear_address_watch(struct amdgpu_device *adev, + uint32_t watch_id); +void kgd_gfx_v9_get_iq_wait_times(struct amdgpu_device *adev, uint32_t *wait_times); +void kgd_gfx_v9_build_grace_period_packet_info(struct amdgpu_device *adev, + uint32_t wait_times, + uint32_t grace_period, + uint32_t *reg_offset, + uint32_t *reg_data); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index 83a83ced2439f5401bceb623476f57ee160580b4..f61527b800e62970b03ea769c643486eefc6e7be 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -35,7 +35,9 @@ #include "amdgpu_dma_buf.h" #include #include "amdgpu_xgmi.h" +#include "kfd_priv.h" #include "kfd_smi_events.h" +#include /* Userptr restore delay, just long enough to allow consecutive VM * changes to accumulate @@ -110,13 +112,16 @@ void amdgpu_amdkfd_gpuvm_init_mem_limits(void) struct sysinfo si; uint64_t mem; + if (kfd_mem_limit.max_system_mem_limit) + return; + si_meminfo(&si); mem = si.freeram - si.freehigh; mem *= si.mem_unit; spin_lock_init(&kfd_mem_limit.mem_limit_lock); kfd_mem_limit.max_system_mem_limit = mem - (mem >> 4); - kfd_mem_limit.max_ttm_mem_limit = (mem >> 1) - (mem >> 3); + kfd_mem_limit.max_ttm_mem_limit = ttm_tt_pages_limit() << PAGE_SHIFT; pr_debug("Kernel memory limit %lluM, TTM limit %lluM\n", (kfd_mem_limit.max_system_mem_limit >> 20), (kfd_mem_limit.max_ttm_mem_limit >> 20)); @@ -148,16 +153,20 @@ void amdgpu_amdkfd_reserve_system_mem(uint64_t size) * @size: Size of buffer, in bytes, encapsulated by B0. This should be * equivalent to amdgpu_bo_size(BO) * @alloc_flag: Flag used in allocating a BO as noted above + * @xcp_id: xcp_id is used to get xcp from xcp manager, one xcp is + * managed as one compute node in driver for app * - * Return: returns -ENOMEM in case of error, ZERO otherwise + * Return: + * returns -ENOMEM in case of error, ZERO otherwise */ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev, - uint64_t size, u32 alloc_flag) + uint64_t size, u32 alloc_flag, int8_t xcp_id) { uint64_t reserved_for_pt = ESTIMATE_PT_SIZE(amdgpu_amdkfd_total_mem_size); size_t system_mem_needed, ttm_mem_needed, vram_needed; int ret = 0; + uint64_t vram_size = 0; system_mem_needed = 0; ttm_mem_needed = 0; @@ -172,6 +181,17 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev, * 2M BO chunk. */ vram_needed = size; + /* + * For GFX 9.4.3, get the VRAM size from XCP structs + */ + if (WARN_ONCE(xcp_id < 0, "invalid XCP ID %d", xcp_id)) + return -EINVAL; + + vram_size = KFD_XCP_MEMORY_SIZE(adev, xcp_id); + if (adev->gmc.is_app_apu) { + system_mem_needed = size; + ttm_mem_needed = size; + } } else if (alloc_flag & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR) { system_mem_needed = size; } else if (!(alloc_flag & @@ -191,8 +211,8 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev, kfd_mem_limit.max_system_mem_limit && !no_system_mem_limit) || (kfd_mem_limit.ttm_mem_used + ttm_mem_needed > kfd_mem_limit.max_ttm_mem_limit) || - (adev && adev->kfd.vram_used + vram_needed > - adev->gmc.real_vram_size - reserved_for_pt)) { + (adev && xcp_id >= 0 && adev->kfd.vram_used[xcp_id] + vram_needed > + vram_size - reserved_for_pt)) { ret = -ENOMEM; goto release; } @@ -202,9 +222,11 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev, */ WARN_ONCE(vram_needed && !adev, "adev reference can't be null when vram is used"); - if (adev) { - adev->kfd.vram_used += vram_needed; - adev->kfd.vram_used_aligned += ALIGN(vram_needed, VRAM_AVAILABLITY_ALIGN); + if (adev && xcp_id >= 0) { + adev->kfd.vram_used[xcp_id] += vram_needed; + adev->kfd.vram_used_aligned[xcp_id] += adev->gmc.is_app_apu ? + vram_needed : + ALIGN(vram_needed, VRAM_AVAILABLITY_ALIGN); } kfd_mem_limit.system_mem_used += system_mem_needed; kfd_mem_limit.ttm_mem_used += ttm_mem_needed; @@ -215,7 +237,7 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev, } void amdgpu_amdkfd_unreserve_mem_limit(struct amdgpu_device *adev, - uint64_t size, u32 alloc_flag) + uint64_t size, u32 alloc_flag, int8_t xcp_id) { spin_lock(&kfd_mem_limit.mem_limit_lock); @@ -225,9 +247,19 @@ void amdgpu_amdkfd_unreserve_mem_limit(struct amdgpu_device *adev, } else if (alloc_flag & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) { WARN_ONCE(!adev, "adev reference can't be null when alloc mem flags vram is set"); + if (WARN_ONCE(xcp_id < 0, "invalid XCP ID %d", xcp_id)) + goto release; + if (adev) { - adev->kfd.vram_used -= size; - adev->kfd.vram_used_aligned -= ALIGN(size, VRAM_AVAILABLITY_ALIGN); + adev->kfd.vram_used[xcp_id] -= size; + if (adev->gmc.is_app_apu) { + adev->kfd.vram_used_aligned[xcp_id] -= size; + kfd_mem_limit.system_mem_used -= size; + kfd_mem_limit.ttm_mem_used -= size; + } else { + adev->kfd.vram_used_aligned[xcp_id] -= + ALIGN(size, VRAM_AVAILABLITY_ALIGN); + } } } else if (alloc_flag & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR) { kfd_mem_limit.system_mem_used -= size; @@ -237,8 +269,8 @@ void amdgpu_amdkfd_unreserve_mem_limit(struct amdgpu_device *adev, pr_err("%s: Invalid BO type %#x\n", __func__, alloc_flag); goto release; } - WARN_ONCE(adev && adev->kfd.vram_used < 0, - "KFD VRAM memory accounting unbalanced"); + WARN_ONCE(adev && xcp_id >= 0 && adev->kfd.vram_used[xcp_id] < 0, + "KFD VRAM memory accounting unbalanced for xcp: %d", xcp_id); WARN_ONCE(kfd_mem_limit.ttm_mem_used < 0, "KFD TTM memory accounting unbalanced"); WARN_ONCE(kfd_mem_limit.system_mem_used < 0, @@ -254,14 +286,16 @@ void amdgpu_amdkfd_release_notify(struct amdgpu_bo *bo) u32 alloc_flags = bo->kfd_bo->alloc_flags; u64 size = amdgpu_bo_size(bo); - amdgpu_amdkfd_unreserve_mem_limit(adev, size, alloc_flags); + amdgpu_amdkfd_unreserve_mem_limit(adev, size, alloc_flags, + bo->xcp_id); kfree(bo->kfd_bo); } /** - * @create_dmamap_sg_bo: Creates a amdgpu_bo object to reflect information + * create_dmamap_sg_bo() - Creates a amdgpu_bo object to reflect information * about USERPTR or DOOREBELL or MMIO BO. + * * @adev: Device for which dmamap BO is being created * @mem: BO of peer device that is being DMA mapped. Provides parameters * in building the dmamap BO @@ -285,7 +319,7 @@ create_dmamap_sg_bo(struct amdgpu_device *adev, ret = amdgpu_gem_object_create(adev, mem->bo->tbo.base.size, 1, AMDGPU_GEM_DOMAIN_CPU, AMDGPU_GEM_CREATE_PREEMPTIBLE | flags, - ttm_bo_type_sg, mem->bo->tbo.base.resv, &gem_obj); + ttm_bo_type_sg, mem->bo->tbo.base.resv, &gem_obj, 0); amdgpu_bo_unreserve(mem->bo); @@ -527,6 +561,12 @@ kfd_mem_dmamap_dmabuf(struct kfd_mem_attachment *attachment) { struct ttm_operation_ctx ctx = {.interruptible = true}; struct amdgpu_bo *bo = attachment->bo_va->base.bo; + int ret; + + amdgpu_bo_placement_from_domain(bo, AMDGPU_GEM_DOMAIN_CPU); + ret = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx); + if (ret) + return ret; amdgpu_bo_placement_from_domain(bo, AMDGPU_GEM_DOMAIN_GTT); return ttm_bo_validate(&bo->tbo, &bo->placement, &ctx); @@ -659,11 +699,10 @@ kfd_mem_dmaunmap_userptr(struct kgd_mem *mem, static void kfd_mem_dmaunmap_dmabuf(struct kfd_mem_attachment *attachment) { - struct ttm_operation_ctx ctx = {.interruptible = true}; - struct amdgpu_bo *bo = attachment->bo_va->base.bo; - - amdgpu_bo_placement_from_domain(bo, AMDGPU_GEM_DOMAIN_CPU); - ttm_bo_validate(&bo->tbo, &bo->placement, &ctx); + /* This is a no-op. We don't want to trigger eviction fences when + * unmapping DMABufs. Therefore the invalidation (moving to system + * domain) is done in kfd_mem_dmamap_dmabuf. + */ } /** @@ -804,7 +843,7 @@ static int kfd_mem_attach(struct amdgpu_device *adev, struct kgd_mem *mem, * if peer device has large BAR. In contrast, access over xGMI is * allowed for both small and large BAR configurations of peer device */ - if ((adev != bo_adev) && + if ((adev != bo_adev && !adev->gmc.is_app_apu) && ((mem->domain == AMDGPU_GEM_DOMAIN_VRAM) || (mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL) || (mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP))) { @@ -1599,23 +1638,42 @@ int amdgpu_amdkfd_criu_resume(void *p) return ret; } -size_t amdgpu_amdkfd_get_available_memory(struct amdgpu_device *adev) +size_t amdgpu_amdkfd_get_available_memory(struct amdgpu_device *adev, + uint8_t xcp_id) { uint64_t reserved_for_pt = ESTIMATE_PT_SIZE(amdgpu_amdkfd_total_mem_size); ssize_t available; + uint64_t vram_available, system_mem_available, ttm_mem_available; spin_lock(&kfd_mem_limit.mem_limit_lock); - available = adev->gmc.real_vram_size - - adev->kfd.vram_used_aligned + vram_available = KFD_XCP_MEMORY_SIZE(adev, xcp_id) + - adev->kfd.vram_used_aligned[xcp_id] - atomic64_read(&adev->vram_pin_size) - reserved_for_pt; + + if (adev->gmc.is_app_apu) { + system_mem_available = no_system_mem_limit ? + kfd_mem_limit.max_system_mem_limit : + kfd_mem_limit.max_system_mem_limit - + kfd_mem_limit.system_mem_used; + + ttm_mem_available = kfd_mem_limit.max_ttm_mem_limit - + kfd_mem_limit.ttm_mem_used; + + available = min3(system_mem_available, ttm_mem_available, + vram_available); + available = ALIGN_DOWN(available, PAGE_SIZE); + } else { + available = ALIGN_DOWN(vram_available, VRAM_AVAILABLITY_ALIGN); + } + spin_unlock(&kfd_mem_limit.mem_limit_lock); if (available < 0) available = 0; - return ALIGN_DOWN(available, VRAM_AVAILABLITY_ALIGN); + return available; } int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( @@ -1624,6 +1682,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( uint64_t *offset, uint32_t flags, bool criu_resume) { struct amdgpu_vm *avm = drm_priv_to_vm(drm_priv); + struct amdgpu_fpriv *fpriv = container_of(avm, struct amdgpu_fpriv, vm); enum ttm_bo_type bo_type = ttm_bo_type_device; struct sg_table *sg = NULL; uint64_t user_addr = 0; @@ -1631,6 +1690,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( struct drm_gem_object *gobj = NULL; u32 domain, alloc_domain; uint64_t aligned_size; + int8_t xcp_id = -1; u64 alloc_flags; int ret; @@ -1639,9 +1699,17 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( */ if (flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) { domain = alloc_domain = AMDGPU_GEM_DOMAIN_VRAM; - alloc_flags = AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE; - alloc_flags |= (flags & KFD_IOC_ALLOC_MEM_FLAGS_PUBLIC) ? + + if (adev->gmc.is_app_apu) { + domain = AMDGPU_GEM_DOMAIN_GTT; + alloc_domain = AMDGPU_GEM_DOMAIN_GTT; + alloc_flags = 0; + } else { + alloc_flags = AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE; + alloc_flags |= (flags & KFD_IOC_ALLOC_MEM_FLAGS_PUBLIC) ? AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED : 0; + } + xcp_id = fpriv->xcp_id == ~0 ? 0 : fpriv->xcp_id; } else if (flags & KFD_IOC_ALLOC_MEM_FLAGS_GTT) { domain = alloc_domain = AMDGPU_GEM_DOMAIN_GTT; alloc_flags = 0; @@ -1693,17 +1761,19 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( amdgpu_sync_create(&(*mem)->sync); - ret = amdgpu_amdkfd_reserve_mem_limit(adev, aligned_size, flags); + ret = amdgpu_amdkfd_reserve_mem_limit(adev, aligned_size, flags, + xcp_id); if (ret) { pr_debug("Insufficient memory\n"); goto err_reserve_limit; } - pr_debug("\tcreate BO VA 0x%llx size 0x%llx domain %s\n", - va, (*mem)->aql_queue ? size << 1 : size, domain_string(alloc_domain)); + pr_debug("\tcreate BO VA 0x%llx size 0x%llx domain %s xcp_id %d\n", + va, (*mem)->aql_queue ? size << 1 : size, + domain_string(alloc_domain), xcp_id); ret = amdgpu_gem_object_create(adev, aligned_size, 1, alloc_domain, alloc_flags, - bo_type, NULL, &gobj); + bo_type, NULL, &gobj, xcp_id + 1); if (ret) { pr_debug("Failed to create BO on domain %s. ret %d\n", domain_string(alloc_domain), ret); @@ -1728,6 +1798,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( (*mem)->domain = domain; (*mem)->mapped_to_gpu_memory = 0; (*mem)->process_info = avm->process_info; + add_kgd_mem_to_kfd_bo_list(*mem, avm->process_info, user_addr); if (user_addr) { @@ -1759,7 +1830,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( /* Don't unreserve system mem limit twice */ goto err_reserve_limit; err_bo_create: - amdgpu_amdkfd_unreserve_mem_limit(adev, aligned_size, flags); + amdgpu_amdkfd_unreserve_mem_limit(adev, aligned_size, flags, xcp_id); err_reserve_limit: mutex_destroy(&(*mem)->lock); if (gobj) @@ -1855,11 +1926,14 @@ int amdgpu_amdkfd_gpuvm_free_memory_of_gpu( } /* Update the size of the BO being freed if it was allocated from - * VRAM and is not imported. + * VRAM and is not imported. For APP APU VRAM allocations are done + * in GTT domain */ if (size) { - if ((mem->bo->preferred_domains == AMDGPU_GEM_DOMAIN_VRAM) && - (!is_imported)) + if (!is_imported && + (mem->bo->preferred_domains == AMDGPU_GEM_DOMAIN_VRAM || + (adev->gmc.is_app_apu && + mem->bo->preferred_domains == AMDGPU_GEM_DOMAIN_GTT))) *size = bo_size; else *size = 0; @@ -2282,8 +2356,9 @@ int amdgpu_amdkfd_gpuvm_import_dmabuf(struct amdgpu_device *adev, (*mem)->dmabuf = dma_buf; (*mem)->bo = bo; (*mem)->va = va; - (*mem)->domain = (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM) ? + (*mem)->domain = (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM) && !adev->gmc.is_app_apu ? AMDGPU_GEM_DOMAIN_VRAM : AMDGPU_GEM_DOMAIN_GTT; + (*mem)->mapped_to_gpu_memory = 0; (*mem)->process_info = avm->process_info; add_kgd_mem_to_kfd_bo_list(*mem, avm->process_info, false); @@ -2445,7 +2520,9 @@ static int update_invalid_user_pages(struct amdkfd_process_info *process_info, ret = -EAGAIN; goto unlock_out; } - mem->invalid = 0; + /* set mem valid if mem has hmm range associated */ + if (mem->range) + mem->invalid = 0; } unlock_out: @@ -2577,8 +2654,15 @@ static int confirm_valid_user_pages_locked(struct amdkfd_process_info *process_i list_for_each_entry_safe(mem, tmp_mem, &process_info->userptr_inval_list, validate_list.head) { - bool valid = amdgpu_ttm_tt_get_user_pages_done( - mem->bo->tbo.ttm, mem->range); + bool valid; + + /* keep mem without hmm range at userptr_inval_list */ + if (!mem->range) + continue; + + /* Only check mem with hmm range associated */ + valid = amdgpu_ttm_tt_get_user_pages_done( + mem->bo->tbo.ttm, mem->range); mem->range = NULL; if (!valid) { @@ -2586,7 +2670,12 @@ static int confirm_valid_user_pages_locked(struct amdkfd_process_info *process_i ret = -EAGAIN; continue; } - WARN(mem->invalid, "Valid BO is marked invalid"); + + if (mem->invalid) { + WARN(1, "Valid BO is marked invalid"); + ret = -EAGAIN; + continue; + } list_move_tail(&mem->validate_list.head, &process_info->userptr_valid_list); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c index ac6fe0ae4609f9dc79fa6ba9f17506b37612c199..ef4b9a41f20ad4b134bb565388be3779999d73a3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c @@ -272,6 +272,7 @@ static int convert_atom_mem_type_to_vram_type(struct amdgpu_device *adev, break; case ATOM_DGPU_VRAM_TYPE_HBM2: case ATOM_DGPU_VRAM_TYPE_HBM2E: + case ATOM_DGPU_VRAM_TYPE_HBM3: vram_type = AMDGPU_VRAM_TYPE_HBM; break; case ATOM_DGPU_VRAM_TYPE_GDDR6: diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c index 30c28a69e847d25cdf7461a184e896df497c6810..b582b83c4984fa7d833e1d21ce6800d5c1c78a47 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c @@ -104,9 +104,8 @@ static bool igp_read_bios_from_vram(struct amdgpu_device *adev) adev->bios = NULL; vram_base = pci_resource_start(adev->pdev, 0); bios = ioremap_wc(vram_base, size); - if (!bios) { + if (!bios) return false; - } adev->bios = kmalloc(size, GFP_KERNEL); if (!adev->bios) { @@ -133,9 +132,8 @@ bool amdgpu_read_bios(struct amdgpu_device *adev) adev->bios = NULL; /* XXX: some cards may return 0 for rom size? ddx has a workaround */ bios = pci_map_rom(adev->pdev, &size); - if (!bios) { + if (!bios) return false; - } adev->bios = kzalloc(size, GFP_KERNEL); if (adev->bios == NULL) { @@ -168,9 +166,9 @@ static bool amdgpu_read_bios_from_rom(struct amdgpu_device *adev) header[AMD_VBIOS_SIGNATURE_END] = 0; if ((!AMD_IS_VALID_VBIOS(header)) || - 0 != memcmp((char *)&header[AMD_VBIOS_SIGNATURE_OFFSET], - AMD_VBIOS_SIGNATURE, - strlen(AMD_VBIOS_SIGNATURE))) + memcmp((char *)&header[AMD_VBIOS_SIGNATURE_OFFSET], + AMD_VBIOS_SIGNATURE, + strlen(AMD_VBIOS_SIGNATURE)) != 0) return false; /* valid vbios, go on */ @@ -264,7 +262,7 @@ static int amdgpu_atrm_call(acpi_handle atrm_handle, uint8_t *bios, status = acpi_evaluate_object(atrm_handle, NULL, &atrm_arg, &buffer); if (ACPI_FAILURE(status)) { - printk("failed to evaluate ATRM got %s\n", acpi_format_exception(status)); + DRM_ERROR("failed to evaluate ATRM got %s\n", acpi_format_exception(status)); return -ENODEV; } @@ -363,7 +361,7 @@ static bool amdgpu_acpi_vfct_bios(struct amdgpu_device *adev) struct acpi_table_header *hdr; acpi_size tbl_size; UEFI_ACPI_VFCT *vfct; - unsigned offset; + unsigned int offset; if (!ACPI_SUCCESS(acpi_get_table("VFCT", 1, &hdr))) return false; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c index 6be30dcb029d579dd28995c10c0a8a27c762b81f..d34037b85cf8599d166f4981689611a951e8e82a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c @@ -593,11 +593,20 @@ static int amdgpu_connector_set_property(struct drm_connector *connector, switch (val) { default: - case DRM_MODE_SCALE_NONE: rmx_type = RMX_OFF; break; - case DRM_MODE_SCALE_CENTER: rmx_type = RMX_CENTER; break; - case DRM_MODE_SCALE_ASPECT: rmx_type = RMX_ASPECT; break; - case DRM_MODE_SCALE_FULLSCREEN: rmx_type = RMX_FULL; break; + case DRM_MODE_SCALE_NONE: + rmx_type = RMX_OFF; + break; + case DRM_MODE_SCALE_CENTER: + rmx_type = RMX_CENTER; + break; + case DRM_MODE_SCALE_ASPECT: + rmx_type = RMX_ASPECT; + break; + case DRM_MODE_SCALE_FULLSCREEN: + rmx_type = RMX_FULL; + break; } + if (amdgpu_encoder->rmx_type == rmx_type) return 0; @@ -799,12 +808,21 @@ static int amdgpu_connector_set_lcd_property(struct drm_connector *connector, } switch (value) { - case DRM_MODE_SCALE_NONE: rmx_type = RMX_OFF; break; - case DRM_MODE_SCALE_CENTER: rmx_type = RMX_CENTER; break; - case DRM_MODE_SCALE_ASPECT: rmx_type = RMX_ASPECT; break; + case DRM_MODE_SCALE_NONE: + rmx_type = RMX_OFF; + break; + case DRM_MODE_SCALE_CENTER: + rmx_type = RMX_CENTER; + break; + case DRM_MODE_SCALE_ASPECT: + rmx_type = RMX_ASPECT; + break; default: - case DRM_MODE_SCALE_FULLSCREEN: rmx_type = RMX_FULL; break; + case DRM_MODE_SCALE_FULLSCREEN: + rmx_type = RMX_FULL; + break; } + if (amdgpu_encoder->rmx_type == rmx_type) return 0; @@ -1127,7 +1145,8 @@ amdgpu_connector_dvi_detect(struct drm_connector *connector, bool force) /* assume digital unless load detected otherwise */ amdgpu_connector->use_digital = true; lret = encoder_funcs->detect(encoder, connector); - DRM_DEBUG_KMS("load_detect %x returned: %x\n",encoder->encoder_type,lret); + DRM_DEBUG_KMS("load_detect %x returned: %x\n", + encoder->encoder_type, lret); if (lret == connector_status_connected) amdgpu_connector->use_digital = false; } @@ -1991,7 +2010,7 @@ amdgpu_connector_add(struct amdgpu_device *adev, if (amdgpu_connector->hpd.hpd == AMDGPU_HPD_NONE) { if (i2c_bus->valid) { connector->polled = DRM_CONNECTOR_POLL_CONNECT | - DRM_CONNECTOR_POLL_DISCONNECT; + DRM_CONNECTOR_POLL_DISCONNECT; } } else connector->polled = DRM_CONNECTOR_POLL_HPD; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 2eb2c66843a8891a5bd49ebdd1defea630b7b99e..d9503882ea97276dedf96a357df76ec508293246 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -112,6 +112,9 @@ static int amdgpu_cs_p1_ib(struct amdgpu_cs_parser *p, if (r < 0) return r; + if (num_ibs[r] >= amdgpu_ring_max_ibs(chunk_ib->ip_type)) + return -EINVAL; + ++(num_ibs[r]); p->gang_leader_idx = r; return 0; @@ -192,7 +195,7 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p, uint64_t *chunk_array_user; uint64_t *chunk_array; uint32_t uf_offset = 0; - unsigned int size; + size_t size; int ret; int i; @@ -285,6 +288,7 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p, case AMDGPU_CHUNK_ID_SCHEDULED_DEPENDENCIES: case AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_WAIT: case AMDGPU_CHUNK_ID_SYNCOBJ_TIMELINE_SIGNAL: + case AMDGPU_CHUNK_ID_CP_GFX_SHADOW: break; default: @@ -305,7 +309,7 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p, } p->gang_leader = p->jobs[p->gang_leader_idx]; - if (p->ctx->vram_lost_counter != p->gang_leader->vram_lost_counter) { + if (p->ctx->generation != p->gang_leader->generation) { ret = -ECANCELED; goto free_all_kdata; } @@ -393,7 +397,7 @@ static int amdgpu_cs_p2_dependencies(struct amdgpu_cs_parser *p, { struct drm_amdgpu_cs_chunk_dep *deps = chunk->kdata; struct amdgpu_fpriv *fpriv = p->filp->driver_priv; - unsigned num_deps; + unsigned int num_deps; int i, r; num_deps = chunk->length_dw * 4 / @@ -464,7 +468,7 @@ static int amdgpu_cs_p2_syncobj_in(struct amdgpu_cs_parser *p, struct amdgpu_cs_chunk *chunk) { struct drm_amdgpu_cs_chunk_sem *deps = chunk->kdata; - unsigned num_deps; + unsigned int num_deps; int i, r; num_deps = chunk->length_dw * 4 / @@ -482,7 +486,7 @@ static int amdgpu_cs_p2_syncobj_timeline_wait(struct amdgpu_cs_parser *p, struct amdgpu_cs_chunk *chunk) { struct drm_amdgpu_cs_chunk_syncobj *syncobj_deps = chunk->kdata; - unsigned num_deps; + unsigned int num_deps; int i, r; num_deps = chunk->length_dw * 4 / @@ -502,7 +506,7 @@ static int amdgpu_cs_p2_syncobj_out(struct amdgpu_cs_parser *p, struct amdgpu_cs_chunk *chunk) { struct drm_amdgpu_cs_chunk_sem *deps = chunk->kdata; - unsigned num_deps; + unsigned int num_deps; int i; num_deps = chunk->length_dw * 4 / @@ -536,7 +540,7 @@ static int amdgpu_cs_p2_syncobj_timeline_signal(struct amdgpu_cs_parser *p, struct amdgpu_cs_chunk *chunk) { struct drm_amdgpu_cs_chunk_syncobj *syncobj_deps = chunk->kdata; - unsigned num_deps; + unsigned int num_deps; int i; num_deps = chunk->length_dw * 4 / @@ -575,6 +579,26 @@ static int amdgpu_cs_p2_syncobj_timeline_signal(struct amdgpu_cs_parser *p, return 0; } +static int amdgpu_cs_p2_shadow(struct amdgpu_cs_parser *p, + struct amdgpu_cs_chunk *chunk) +{ + struct drm_amdgpu_cs_chunk_cp_gfx_shadow *shadow = chunk->kdata; + int i; + + if (shadow->flags & ~AMDGPU_CS_CHUNK_CP_GFX_SHADOW_FLAGS_INIT_SHADOW) + return -EINVAL; + + for (i = 0; i < p->gang_size; ++i) { + p->jobs[i]->shadow_va = shadow->shadow_va; + p->jobs[i]->csa_va = shadow->csa_va; + p->jobs[i]->gds_va = shadow->gds_va; + p->jobs[i]->init_shadow = + shadow->flags & AMDGPU_CS_CHUNK_CP_GFX_SHADOW_FLAGS_INIT_SHADOW; + } + + return 0; +} + static int amdgpu_cs_pass2(struct amdgpu_cs_parser *p) { unsigned int ce_preempt = 0, de_preempt = 0; @@ -617,6 +641,11 @@ static int amdgpu_cs_pass2(struct amdgpu_cs_parser *p) if (r) return r; break; + case AMDGPU_CHUNK_ID_CP_GFX_SHADOW: + r = amdgpu_cs_p2_shadow(p, chunk); + if (r) + return r; + break; } } @@ -729,6 +758,7 @@ static void amdgpu_cs_get_threshold_for_moves(struct amdgpu_device *adev, if (used_vis_vram < total_vis_vram) { u64 free_vis_vram = total_vis_vram - used_vis_vram; + adev->mm_stats.accum_us_vis = min(adev->mm_stats.accum_us_vis + increment_us, us_upper_bound); @@ -1047,9 +1077,8 @@ static int amdgpu_cs_patch_ibs(struct amdgpu_cs_parser *p, /* the IB should be reserved at this point */ r = amdgpu_bo_kmap(aobj, (void **)&kptr); - if (r) { + if (r) return r; - } kptr += va_start - (m->start * AMDGPU_GPU_PAGE_SIZE); @@ -1356,7 +1385,7 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p, /* Cleanup the parser structure */ static void amdgpu_cs_parser_fini(struct amdgpu_cs_parser *parser) { - unsigned i; + unsigned int i; amdgpu_sync_free(&parser->sync); for (i = 0; i < parser->num_post_deps; i++) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c index c6d4d41c4393e18f5b3108d5cd90b6e084ef3ab1..23d054526e7c7ba21673d0b9b64e4e3bef718bbc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c @@ -106,3 +106,41 @@ int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm, ttm_eu_backoff_reservation(&ticket, &list); return 0; } + +int amdgpu_unmap_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm, + struct amdgpu_bo *bo, struct amdgpu_bo_va *bo_va, + uint64_t csa_addr) +{ + struct ww_acquire_ctx ticket; + struct list_head list; + struct amdgpu_bo_list_entry pd; + struct ttm_validate_buffer csa_tv; + int r; + + INIT_LIST_HEAD(&list); + INIT_LIST_HEAD(&csa_tv.head); + csa_tv.bo = &bo->tbo; + csa_tv.num_shared = 1; + + list_add(&csa_tv.head, &list); + amdgpu_vm_get_pd_bo(vm, &list, &pd); + + r = ttm_eu_reserve_buffers(&ticket, &list, true, NULL); + if (r) { + DRM_ERROR("failed to reserve CSA,PD BOs: err=%d\n", r); + return r; + } + + r = amdgpu_vm_bo_unmap(adev, bo_va, csa_addr); + if (r) { + DRM_ERROR("failed to do bo_unmap on static CSA, err=%d\n", r); + ttm_eu_backoff_reservation(&ticket, &list); + return r; + } + + amdgpu_vm_bo_del(adev, bo_va); + + ttm_eu_backoff_reservation(&ticket, &list); + + return 0; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.h index 524b4437a021781e4d95f98f36227850dd581211..7dfc1f2012ebf18f7d6f13d26797826e8ebe67c5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.h @@ -34,6 +34,9 @@ int amdgpu_allocate_static_csa(struct amdgpu_device *adev, struct amdgpu_bo **bo int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm, struct amdgpu_bo *bo, struct amdgpu_bo_va **bo_va, uint64_t csa_addr, uint32_t size); +int amdgpu_unmap_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm, + struct amdgpu_bo *bo, struct amdgpu_bo_va *bo_va, + uint64_t csa_addr); void amdgpu_free_static_csa(struct amdgpu_bo **bo); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index d2139ac1215950d742197c3f57ccb85bc812aa5f..0dc9c655c4fbdbd9dbd3224508fac806dbd99dfc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -222,8 +222,19 @@ static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, u32 hw_ip, drm_prio = amdgpu_ctx_to_drm_sched_prio(ctx_prio); hw_ip = array_index_nospec(hw_ip, AMDGPU_HW_IP_NUM); - scheds = adev->gpu_sched[hw_ip][hw_prio].sched; - num_scheds = adev->gpu_sched[hw_ip][hw_prio].num_scheds; + + if (!(adev)->xcp_mgr) { + scheds = adev->gpu_sched[hw_ip][hw_prio].sched; + num_scheds = adev->gpu_sched[hw_ip][hw_prio].num_scheds; + } else { + struct amdgpu_fpriv *fpriv; + + fpriv = container_of(ctx->ctx_mgr, struct amdgpu_fpriv, ctx_mgr); + r = amdgpu_xcp_select_scheds(adev, hw_ip, hw_prio, fpriv, + &num_scheds, &scheds); + if (r) + goto cleanup_entity; + } /* disable load balance if the hw engine retains context among dependent jobs */ if (hw_ip == AMDGPU_HW_IP_VCN_ENC || @@ -255,7 +266,8 @@ static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, u32 hw_ip, return r; } -static ktime_t amdgpu_ctx_fini_entity(struct amdgpu_ctx_entity *entity) +static ktime_t amdgpu_ctx_fini_entity(struct amdgpu_device *adev, + struct amdgpu_ctx_entity *entity) { ktime_t res = ns_to_ktime(0); int i; @@ -268,6 +280,8 @@ static ktime_t amdgpu_ctx_fini_entity(struct amdgpu_ctx_entity *entity) dma_fence_put(entity->fences[i]); } + amdgpu_xcp_release_sched(adev, entity); + kfree(entity); return res; } @@ -303,6 +317,7 @@ static int amdgpu_ctx_get_stable_pstate(struct amdgpu_ctx *ctx, static int amdgpu_ctx_init(struct amdgpu_ctx_mgr *mgr, int32_t priority, struct drm_file *filp, struct amdgpu_ctx *ctx) { + struct amdgpu_fpriv *fpriv = filp->driver_priv; u32 current_stable_pstate; int r; @@ -318,7 +333,7 @@ static int amdgpu_ctx_init(struct amdgpu_ctx_mgr *mgr, int32_t priority, ctx->reset_counter = atomic_read(&mgr->adev->gpu_reset_counter); ctx->reset_counter_query = ctx->reset_counter; - ctx->vram_lost_counter = atomic_read(&mgr->adev->vram_lost_counter); + ctx->generation = amdgpu_vm_generation(mgr->adev, &fpriv->vm); ctx->init_priority = priority; ctx->override_priority = AMDGPU_CTX_PRIORITY_UNSET; @@ -331,6 +346,7 @@ static int amdgpu_ctx_init(struct amdgpu_ctx_mgr *mgr, int32_t priority, else ctx->stable_pstate = current_stable_pstate; + ctx->ctx_mgr = &(fpriv->ctx_mgr); return 0; } @@ -399,7 +415,7 @@ static void amdgpu_ctx_fini(struct kref *ref) for (j = 0; j < AMDGPU_MAX_ENTITY_NUM; ++j) { ktime_t spend; - spend = amdgpu_ctx_fini_entity(ctx->entities[i][j]); + spend = amdgpu_ctx_fini_entity(adev, ctx->entities[i][j]); atomic64_add(ktime_to_ns(spend), &mgr->time_spend[i]); } } @@ -416,6 +432,7 @@ int amdgpu_ctx_get_entity(struct amdgpu_ctx *ctx, u32 hw_ip, u32 instance, u32 ring, struct drm_sched_entity **entity) { int r; + struct drm_sched_entity *ctx_entity; if (hw_ip >= AMDGPU_HW_IP_NUM) { DRM_ERROR("unknown HW IP type: %d\n", hw_ip); @@ -439,7 +456,14 @@ int amdgpu_ctx_get_entity(struct amdgpu_ctx *ctx, u32 hw_ip, u32 instance, return r; } - *entity = &ctx->entities[hw_ip][ring]->entity; + ctx_entity = &ctx->entities[hw_ip][ring]->entity; + r = drm_sched_entity_error(ctx_entity); + if (r) { + DRM_DEBUG("error entity %p\n", ctx_entity); + return r; + } + + *entity = ctx_entity; return 0; } @@ -570,12 +594,15 @@ static int amdgpu_ctx_query2(struct amdgpu_device *adev, if (ctx->reset_counter != atomic_read(&adev->gpu_reset_counter)) out->state.flags |= AMDGPU_CTX_QUERY2_FLAGS_RESET; - if (ctx->vram_lost_counter != atomic_read(&adev->vram_lost_counter)) + if (ctx->generation != amdgpu_vm_generation(adev, &fpriv->vm)) out->state.flags |= AMDGPU_CTX_QUERY2_FLAGS_VRAMLOST; if (atomic_read(&ctx->guilty)) out->state.flags |= AMDGPU_CTX_QUERY2_FLAGS_GUILTY; + if (amdgpu_in_reset(adev)) + out->state.flags |= AMDGPU_CTX_QUERY2_FLAGS_RESET_IN_PROGRESS; + if (adev->ras_enabled && con) { /* Return the cached values in O(1), * and schedule delayed work to cache diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h index 0fa0e56daf67e9167526268494a9c81840fbd7ad..85376baaa92f2bf6db23366d8d4d36b771923fa9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h @@ -47,7 +47,7 @@ struct amdgpu_ctx { struct amdgpu_ctx_mgr *mgr; unsigned reset_counter; unsigned reset_counter_query; - uint32_t vram_lost_counter; + uint64_t generation; spinlock_t ring_lock; struct amdgpu_ctx_entity *entities[AMDGPU_HW_IP_NUM][AMDGPU_MAX_ENTITY_NUM]; bool preamble_presented; @@ -57,6 +57,7 @@ struct amdgpu_ctx { unsigned long ras_counter_ce; unsigned long ras_counter_ue; uint32_t stable_pstate; + struct amdgpu_ctx_mgr *ctx_mgr; }; struct amdgpu_ctx_mgr { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c index f60753f97ac598bc04b617ba1e9ab7193cc58b64..56e89e76ff179a6eaa83c412e270328c0e143fb1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c @@ -56,14 +56,14 @@ * * Bit 62: Indicates a GRBM bank switch is needed * Bit 61: Indicates a SRBM bank switch is needed (implies bit 62 is - * zero) + * zero) * Bits 24..33: The SE or ME selector if needed * Bits 34..43: The SH (or SA) or PIPE selector if needed * Bits 44..53: The INSTANCE (or CU/WGP) or QUEUE selector if needed * * Bit 23: Indicates that the PM power gating lock should be held - * This is necessary to read registers that might be - * unreliable during a power gating transistion. + * This is necessary to read registers that might be + * unreliable during a power gating transistion. * * The lower bits are the BYTE offset of the register to read. This * allows reading multiple registers in a single call and having @@ -76,7 +76,7 @@ static int amdgpu_debugfs_process_reg_op(bool read, struct file *f, ssize_t result = 0; int r; bool pm_pg_lock, use_bank, use_ring; - unsigned instance_bank, sh_bank, se_bank, me, pipe, queue, vmid; + unsigned int instance_bank, sh_bank, se_bank, me, pipe, queue, vmid; pm_pg_lock = use_bank = use_ring = false; instance_bank = sh_bank = se_bank = me = pipe = queue = vmid = 0; @@ -136,10 +136,10 @@ static int amdgpu_debugfs_process_reg_op(bool read, struct file *f, } mutex_lock(&adev->grbm_idx_mutex); amdgpu_gfx_select_se_sh(adev, se_bank, - sh_bank, instance_bank); + sh_bank, instance_bank, 0); } else if (use_ring) { mutex_lock(&adev->srbm_mutex); - amdgpu_gfx_select_me_pipe_q(adev, me, pipe, queue, vmid); + amdgpu_gfx_select_me_pipe_q(adev, me, pipe, queue, vmid, 0); } if (pm_pg_lock) @@ -169,10 +169,10 @@ static int amdgpu_debugfs_process_reg_op(bool read, struct file *f, end: if (use_bank) { - amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); } else if (use_ring) { - amdgpu_gfx_select_me_pipe_q(adev, 0, 0, 0, 0); + amdgpu_gfx_select_me_pipe_q(adev, 0, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); } @@ -208,7 +208,7 @@ static int amdgpu_debugfs_regs2_open(struct inode *inode, struct file *file) { struct amdgpu_debugfs_regs2_data *rd; - rd = kzalloc(sizeof *rd, GFP_KERNEL); + rd = kzalloc(sizeof(*rd), GFP_KERNEL); if (!rd) return -ENOMEM; rd->adev = file_inode(file)->i_private; @@ -221,6 +221,7 @@ static int amdgpu_debugfs_regs2_open(struct inode *inode, struct file *file) static int amdgpu_debugfs_regs2_release(struct inode *inode, struct file *file) { struct amdgpu_debugfs_regs2_data *rd = file->private_data; + mutex_destroy(&rd->lock); kfree(file->private_data); return 0; @@ -262,14 +263,14 @@ static ssize_t amdgpu_debugfs_regs2_op(struct file *f, char __user *buf, u32 off } mutex_lock(&adev->grbm_idx_mutex); amdgpu_gfx_select_se_sh(adev, rd->id.grbm.se, - rd->id.grbm.sh, - rd->id.grbm.instance); + rd->id.grbm.sh, + rd->id.grbm.instance, rd->id.xcc_id); } if (rd->id.use_srbm) { mutex_lock(&adev->srbm_mutex); amdgpu_gfx_select_me_pipe_q(adev, rd->id.srbm.me, rd->id.srbm.pipe, - rd->id.srbm.queue, rd->id.srbm.vmid); + rd->id.srbm.queue, rd->id.srbm.vmid, rd->id.xcc_id); } if (rd->id.pg_lock) @@ -295,12 +296,12 @@ static ssize_t amdgpu_debugfs_regs2_op(struct file *f, char __user *buf, u32 off } end: if (rd->id.use_grbm) { - amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, rd->id.xcc_id); mutex_unlock(&adev->grbm_idx_mutex); } if (rd->id.use_srbm) { - amdgpu_gfx_select_me_pipe_q(adev, 0, 0, 0, 0); + amdgpu_gfx_select_me_pipe_q(adev, 0, 0, 0, 0, rd->id.xcc_id); mutex_unlock(&adev->srbm_mutex); } @@ -319,18 +320,45 @@ static ssize_t amdgpu_debugfs_regs2_op(struct file *f, char __user *buf, u32 off static long amdgpu_debugfs_regs2_ioctl(struct file *f, unsigned int cmd, unsigned long data) { struct amdgpu_debugfs_regs2_data *rd = f->private_data; + struct amdgpu_debugfs_regs2_iocdata v1_data; int r; + mutex_lock(&rd->lock); + switch (cmd) { + case AMDGPU_DEBUGFS_REGS2_IOC_SET_STATE_V2: + r = copy_from_user(&rd->id, (struct amdgpu_debugfs_regs2_iocdata_v2 *)data, + sizeof(rd->id)); + if (r) + r = -EINVAL; + goto done; case AMDGPU_DEBUGFS_REGS2_IOC_SET_STATE: - mutex_lock(&rd->lock); - r = copy_from_user(&rd->id, (struct amdgpu_debugfs_regs2_iocdata *)data, sizeof rd->id); - mutex_unlock(&rd->lock); - return r ? -EINVAL : 0; + r = copy_from_user(&v1_data, (struct amdgpu_debugfs_regs2_iocdata *)data, + sizeof(v1_data)); + if (r) { + r = -EINVAL; + goto done; + } + goto v1_copy; default: - return -EINVAL; - } - return 0; + r = -EINVAL; + goto done; + } + +v1_copy: + rd->id.use_srbm = v1_data.use_srbm; + rd->id.use_grbm = v1_data.use_grbm; + rd->id.pg_lock = v1_data.pg_lock; + rd->id.grbm.se = v1_data.grbm.se; + rd->id.grbm.sh = v1_data.grbm.sh; + rd->id.grbm.instance = v1_data.grbm.instance; + rd->id.srbm.me = v1_data.srbm.me; + rd->id.srbm.pipe = v1_data.srbm.pipe; + rd->id.srbm.queue = v1_data.srbm.queue; + rd->id.xcc_id = 0; +done: + mutex_unlock(&rd->lock); + return r; } static ssize_t amdgpu_debugfs_regs2_read(struct file *f, char __user *buf, size_t size, loff_t *pos) @@ -343,6 +371,136 @@ static ssize_t amdgpu_debugfs_regs2_write(struct file *f, const char __user *buf return amdgpu_debugfs_regs2_op(f, (char __user *)buf, *pos, size, 1); } +static int amdgpu_debugfs_gprwave_open(struct inode *inode, struct file *file) +{ + struct amdgpu_debugfs_gprwave_data *rd; + + rd = kzalloc(sizeof *rd, GFP_KERNEL); + if (!rd) + return -ENOMEM; + rd->adev = file_inode(file)->i_private; + file->private_data = rd; + mutex_init(&rd->lock); + + return 0; +} + +static int amdgpu_debugfs_gprwave_release(struct inode *inode, struct file *file) +{ + struct amdgpu_debugfs_gprwave_data *rd = file->private_data; + mutex_destroy(&rd->lock); + kfree(file->private_data); + return 0; +} + +static ssize_t amdgpu_debugfs_gprwave_read(struct file *f, char __user *buf, size_t size, loff_t *pos) +{ + struct amdgpu_debugfs_gprwave_data *rd = f->private_data; + struct amdgpu_device *adev = rd->adev; + ssize_t result = 0; + int r; + uint32_t *data, x; + + if (size & 0x3 || *pos & 0x3) + return -EINVAL; + + r = pm_runtime_get_sync(adev_to_drm(adev)->dev); + if (r < 0) { + pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + return r; + } + + r = amdgpu_virt_enable_access_debugfs(adev); + if (r < 0) { + pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + return r; + } + + data = kcalloc(1024, sizeof(*data), GFP_KERNEL); + if (!data) { + pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + amdgpu_virt_disable_access_debugfs(adev); + return -ENOMEM; + } + + /* switch to the specific se/sh/cu */ + mutex_lock(&adev->grbm_idx_mutex); + amdgpu_gfx_select_se_sh(adev, rd->id.se, rd->id.sh, rd->id.cu, rd->id.xcc_id); + + if (!rd->id.gpr_or_wave) { + x = 0; + if (adev->gfx.funcs->read_wave_data) + adev->gfx.funcs->read_wave_data(adev, rd->id.xcc_id, rd->id.simd, rd->id.wave, data, &x); + } else { + x = size >> 2; + if (rd->id.gpr.vpgr_or_sgpr) { + if (adev->gfx.funcs->read_wave_vgprs) + adev->gfx.funcs->read_wave_vgprs(adev, rd->id.xcc_id, rd->id.simd, rd->id.wave, rd->id.gpr.thread, *pos, size>>2, data); + } else { + if (adev->gfx.funcs->read_wave_sgprs) + adev->gfx.funcs->read_wave_sgprs(adev, rd->id.xcc_id, rd->id.simd, rd->id.wave, *pos, size>>2, data); + } + } + + amdgpu_gfx_select_se_sh(adev, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, rd->id.xcc_id); + mutex_unlock(&adev->grbm_idx_mutex); + + pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); + pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + + if (!x) { + result = -EINVAL; + goto done; + } + + while (size && (*pos < x * 4)) { + uint32_t value; + + value = data[*pos >> 2]; + r = put_user(value, (uint32_t *)buf); + if (r) { + result = r; + goto done; + } + + result += 4; + buf += 4; + *pos += 4; + size -= 4; + } + +done: + amdgpu_virt_disable_access_debugfs(adev); + kfree(data); + return result; +} + +static long amdgpu_debugfs_gprwave_ioctl(struct file *f, unsigned int cmd, unsigned long data) +{ + struct amdgpu_debugfs_gprwave_data *rd = f->private_data; + int r = 0; + + mutex_lock(&rd->lock); + + switch (cmd) { + case AMDGPU_DEBUGFS_GPRWAVE_IOC_SET_STATE: + if (copy_from_user(&rd->id, + (struct amdgpu_debugfs_gprwave_iocdata *)data, + sizeof(rd->id))) + r = -EFAULT; + goto done; + default: + r = -EINVAL; + goto done; + } + +done: + mutex_unlock(&rd->lock); + return r; +} + + + /** * amdgpu_debugfs_regs_pcie_read - Read from a PCIE register @@ -863,7 +1021,7 @@ static ssize_t amdgpu_debugfs_sensor_read(struct file *f, char __user *buf, * The offset being sought changes which wave that the status data * will be returned for. The bits are used as follows: * - * Bits 0..6: Byte offset into data + * Bits 0..6: Byte offset into data * Bits 7..14: SE selector * Bits 15..22: SH/SA selector * Bits 23..30: CU/{WGP+SIMD} selector @@ -907,13 +1065,13 @@ static ssize_t amdgpu_debugfs_wave_read(struct file *f, char __user *buf, /* switch to the specific se/sh/cu */ mutex_lock(&adev->grbm_idx_mutex); - amdgpu_gfx_select_se_sh(adev, se, sh, cu); + amdgpu_gfx_select_se_sh(adev, se, sh, cu, 0); x = 0; if (adev->gfx.funcs->read_wave_data) - adev->gfx.funcs->read_wave_data(adev, simd, wave, data, &x); + adev->gfx.funcs->read_wave_data(adev, 0, simd, wave, data, &x); - amdgpu_gfx_select_se_sh(adev, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF); + amdgpu_gfx_select_se_sh(adev, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0); mutex_unlock(&adev->grbm_idx_mutex); pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); @@ -1001,17 +1159,17 @@ static ssize_t amdgpu_debugfs_gpr_read(struct file *f, char __user *buf, /* switch to the specific se/sh/cu */ mutex_lock(&adev->grbm_idx_mutex); - amdgpu_gfx_select_se_sh(adev, se, sh, cu); + amdgpu_gfx_select_se_sh(adev, se, sh, cu, 0); if (bank == 0) { if (adev->gfx.funcs->read_wave_vgprs) - adev->gfx.funcs->read_wave_vgprs(adev, simd, wave, thread, offset, size>>2, data); + adev->gfx.funcs->read_wave_vgprs(adev, 0, simd, wave, thread, offset, size>>2, data); } else { if (adev->gfx.funcs->read_wave_sgprs) - adev->gfx.funcs->read_wave_sgprs(adev, simd, wave, offset, size>>2, data); + adev->gfx.funcs->read_wave_sgprs(adev, 0, simd, wave, offset, size>>2, data); } - amdgpu_gfx_select_se_sh(adev, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF); + amdgpu_gfx_select_se_sh(adev, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0); mutex_unlock(&adev->grbm_idx_mutex); pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); @@ -1339,6 +1497,15 @@ static const struct file_operations amdgpu_debugfs_regs2_fops = { .llseek = default_llseek }; +static const struct file_operations amdgpu_debugfs_gprwave_fops = { + .owner = THIS_MODULE, + .unlocked_ioctl = amdgpu_debugfs_gprwave_ioctl, + .read = amdgpu_debugfs_gprwave_read, + .open = amdgpu_debugfs_gprwave_open, + .release = amdgpu_debugfs_gprwave_release, + .llseek = default_llseek +}; + static const struct file_operations amdgpu_debugfs_regs_fops = { .owner = THIS_MODULE, .read = amdgpu_debugfs_regs_read, @@ -1416,6 +1583,7 @@ static const struct file_operations amdgpu_debugfs_gfxoff_residency_fops = { static const struct file_operations *debugfs_regs[] = { &amdgpu_debugfs_regs_fops, &amdgpu_debugfs_regs2_fops, + &amdgpu_debugfs_gprwave_fops, &amdgpu_debugfs_regs_didt_fops, &amdgpu_debugfs_regs_pcie_fops, &amdgpu_debugfs_regs_smc_fops, @@ -1429,9 +1597,10 @@ static const struct file_operations *debugfs_regs[] = { &amdgpu_debugfs_gfxoff_residency_fops, }; -static const char *debugfs_regs_names[] = { +static const char * const debugfs_regs_names[] = { "amdgpu_regs", "amdgpu_regs2", + "amdgpu_gprwave", "amdgpu_regs_didt", "amdgpu_regs_pcie", "amdgpu_regs_smc", @@ -1447,7 +1616,7 @@ static const char *debugfs_regs_names[] = { /** * amdgpu_debugfs_regs_init - Initialize debugfs entries that provide - * register access. + * register access. * * @adev: The device to attach the debugfs entries to */ @@ -1459,7 +1628,7 @@ int amdgpu_debugfs_regs_init(struct amdgpu_device *adev) for (i = 0; i < ARRAY_SIZE(debugfs_regs); i++) { ent = debugfs_create_file(debugfs_regs_names[i], - S_IFREG | S_IRUGO, root, + S_IFREG | 0444, root, adev, debugfs_regs[i]); if (!i && !IS_ERR_OR_NULL(ent)) i_size_write(ent->d_inode, adev->rmmio_size); @@ -1470,7 +1639,7 @@ int amdgpu_debugfs_regs_init(struct amdgpu_device *adev) static int amdgpu_debugfs_test_ib_show(struct seq_file *m, void *unused) { - struct amdgpu_device *adev = (struct amdgpu_device *)m->private; + struct amdgpu_device *adev = m->private; struct drm_device *dev = adev_to_drm(adev); int r = 0, i; @@ -1494,12 +1663,12 @@ static int amdgpu_debugfs_test_ib_show(struct seq_file *m, void *unused) kthread_park(ring->sched.thread); } - seq_printf(m, "run ib test:\n"); + seq_puts(m, "run ib test:\n"); r = amdgpu_ib_ring_tests(adev); if (r) seq_printf(m, "ib ring tests failed (%d).\n", r); else - seq_printf(m, "ib ring tests passed.\n"); + seq_puts(m, "ib ring tests passed.\n"); /* go on the scheduler */ for (i = 0; i < AMDGPU_MAX_RINGS; i++) { @@ -1581,7 +1750,7 @@ static int amdgpu_debugfs_benchmark(void *data, u64 val) static int amdgpu_debugfs_vm_info_show(struct seq_file *m, void *unused) { - struct amdgpu_device *adev = (struct amdgpu_device *)m->private; + struct amdgpu_device *adev = m->private; struct drm_device *dev = adev_to_drm(adev); struct drm_file *file; int r; @@ -1978,7 +2147,7 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev) amdgpu_debugfs_ring_init(adev, ring); } - for ( i = 0; i < adev->vcn.num_vcn_inst; i++) { + for (i = 0; i < adev->vcn.num_vcn_inst; i++) { if (!amdgpu_vcnfw_log) break; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 5c7d40873ee208b212b5d95b5d69bc6f303c193a..e25f085ee8867419605bcbe74f8ab869bf72b526 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -707,6 +707,48 @@ u32 amdgpu_device_indirect_rreg(struct amdgpu_device *adev, return r; } +u32 amdgpu_device_indirect_rreg_ext(struct amdgpu_device *adev, + u64 reg_addr) +{ + unsigned long flags, pcie_index, pcie_index_hi, pcie_data; + u32 r; + void __iomem *pcie_index_offset; + void __iomem *pcie_index_hi_offset; + void __iomem *pcie_data_offset; + + pcie_index = adev->nbio.funcs->get_pcie_index_offset(adev); + pcie_data = adev->nbio.funcs->get_pcie_data_offset(adev); + if (adev->nbio.funcs->get_pcie_index_hi_offset) + pcie_index_hi = adev->nbio.funcs->get_pcie_index_hi_offset(adev); + else + pcie_index_hi = 0; + + spin_lock_irqsave(&adev->pcie_idx_lock, flags); + pcie_index_offset = (void __iomem *)adev->rmmio + pcie_index * 4; + pcie_data_offset = (void __iomem *)adev->rmmio + pcie_data * 4; + if (pcie_index_hi != 0) + pcie_index_hi_offset = (void __iomem *)adev->rmmio + + pcie_index_hi * 4; + + writel(reg_addr, pcie_index_offset); + readl(pcie_index_offset); + if (pcie_index_hi != 0) { + writel((reg_addr >> 32) & 0xff, pcie_index_hi_offset); + readl(pcie_index_hi_offset); + } + r = readl(pcie_data_offset); + + /* clear the high bits */ + if (pcie_index_hi != 0) { + writel(0, pcie_index_hi_offset); + readl(pcie_index_hi_offset); + } + + spin_unlock_irqrestore(&adev->pcie_idx_lock, flags); + + return r; +} + /** * amdgpu_device_indirect_rreg64 - read a 64bits indirect register * @@ -747,8 +789,6 @@ u64 amdgpu_device_indirect_rreg64(struct amdgpu_device *adev, * amdgpu_device_indirect_wreg - write an indirect register address * * @adev: amdgpu_device pointer - * @pcie_index: mmio register offset - * @pcie_data: mmio register offset * @reg_addr: indirect register offset * @reg_data: indirect register data * @@ -774,12 +814,50 @@ void amdgpu_device_indirect_wreg(struct amdgpu_device *adev, spin_unlock_irqrestore(&adev->pcie_idx_lock, flags); } +void amdgpu_device_indirect_wreg_ext(struct amdgpu_device *adev, + u64 reg_addr, u32 reg_data) +{ + unsigned long flags, pcie_index, pcie_index_hi, pcie_data; + void __iomem *pcie_index_offset; + void __iomem *pcie_index_hi_offset; + void __iomem *pcie_data_offset; + + pcie_index = adev->nbio.funcs->get_pcie_index_offset(adev); + pcie_data = adev->nbio.funcs->get_pcie_data_offset(adev); + if (adev->nbio.funcs->get_pcie_index_hi_offset) + pcie_index_hi = adev->nbio.funcs->get_pcie_index_hi_offset(adev); + else + pcie_index_hi = 0; + + spin_lock_irqsave(&adev->pcie_idx_lock, flags); + pcie_index_offset = (void __iomem *)adev->rmmio + pcie_index * 4; + pcie_data_offset = (void __iomem *)adev->rmmio + pcie_data * 4; + if (pcie_index_hi != 0) + pcie_index_hi_offset = (void __iomem *)adev->rmmio + + pcie_index_hi * 4; + + writel(reg_addr, pcie_index_offset); + readl(pcie_index_offset); + if (pcie_index_hi != 0) { + writel((reg_addr >> 32) & 0xff, pcie_index_hi_offset); + readl(pcie_index_hi_offset); + } + writel(reg_data, pcie_data_offset); + readl(pcie_data_offset); + + /* clear the high bits */ + if (pcie_index_hi != 0) { + writel(0, pcie_index_hi_offset); + readl(pcie_index_hi_offset); + } + + spin_unlock_irqrestore(&adev->pcie_idx_lock, flags); +} + /** * amdgpu_device_indirect_wreg64 - write a 64bits indirect register address * * @adev: amdgpu_device pointer - * @pcie_index: mmio register offset - * @pcie_data: mmio register offset * @reg_addr: indirect register offset * @reg_data: indirect register data * @@ -840,6 +918,13 @@ static uint32_t amdgpu_invalid_rreg(struct amdgpu_device *adev, uint32_t reg) return 0; } +static uint32_t amdgpu_invalid_rreg_ext(struct amdgpu_device *adev, uint64_t reg) +{ + DRM_ERROR("Invalid callback to read register 0x%llX\n", reg); + BUG(); + return 0; +} + /** * amdgpu_invalid_wreg - dummy reg write function * @@ -857,6 +942,13 @@ static void amdgpu_invalid_wreg(struct amdgpu_device *adev, uint32_t reg, uint32 BUG(); } +static void amdgpu_invalid_wreg_ext(struct amdgpu_device *adev, uint64_t reg, uint32_t v) +{ + DRM_ERROR("Invalid callback to write register 0x%llX with 0x%08X\n", + reg, v); + BUG(); +} + /** * amdgpu_invalid_rreg64 - dummy 64 bit reg read function * @@ -942,7 +1034,8 @@ static int amdgpu_device_asic_init(struct amdgpu_device *adev) { amdgpu_asic_pre_asic_init(adev); - if (adev->ip_versions[GC_HWIP][0] >= IP_VERSION(11, 0, 0)) + if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3) || + adev->ip_versions[GC_HWIP][0] >= IP_VERSION(11, 0, 0)) return amdgpu_atomfirmware_asic_init(adev, true); else return amdgpu_atom_asic_init(adev->mode_info.atom_context); @@ -998,7 +1091,7 @@ void amdgpu_device_program_register_sequence(struct amdgpu_device *adev, if (array_size % 3) return; - for (i = 0; i < array_size; i +=3) { + for (i = 0; i < array_size; i += 3) { reg = registers[i + 0]; and_mask = registers[i + 1]; or_mask = registers[i + 2]; @@ -1090,7 +1183,8 @@ static int amdgpu_device_doorbell_init(struct amdgpu_device *adev) * doorbells are in the first page. So with paging queue enabled, * the max num_kernel_doorbells should + 1 page (0x400 in dword) */ - if (adev->asic_type >= CHIP_VEGA10) + if (adev->ip_versions[SDMA0_HWIP][0] >= IP_VERSION(4, 0, 0) && + adev->ip_versions[SDMA0_HWIP][0] < IP_VERSION(4, 2, 0)) adev->doorbell.num_kernel_doorbells += 0x400; } @@ -1291,6 +1385,15 @@ int amdgpu_device_resize_fb_bar(struct amdgpu_device *adev) return 0; } +static bool amdgpu_device_read_bios(struct amdgpu_device *adev) +{ + if (hweight32(adev->aid_mask) && (adev->flags & AMD_IS_APU)) { + return false; + } + + return true; +} + /* * GPU helpers function. */ @@ -1310,6 +1413,9 @@ bool amdgpu_device_need_post(struct amdgpu_device *adev) if (amdgpu_sriov_vf(adev)) return false; + if (!amdgpu_device_read_bios(adev)) + return false; + if (amdgpu_passthrough(adev)) { /* for FIJI: In whole GPU pass-through virtualization case, after VM reboot * some old smc fw still need driver do vPost otherwise gpu hang, while @@ -1547,7 +1653,7 @@ static int amdgpu_device_check_arguments(struct amdgpu_device *adev) dev_warn(adev->dev, "sched jobs (%d) must be at least 4\n", amdgpu_sched_jobs); amdgpu_sched_jobs = 4; - } else if (!is_power_of_2(amdgpu_sched_jobs)){ + } else if (!is_power_of_2(amdgpu_sched_jobs)) { dev_warn(adev->dev, "sched jobs (%d) must be a power of 2\n", amdgpu_sched_jobs); amdgpu_sched_jobs = roundup_pow_of_two(amdgpu_sched_jobs); @@ -2194,7 +2300,7 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev) total = true; for (i = 0; i < adev->num_ip_blocks; i++) { if ((amdgpu_ip_block_mask & (1 << i)) == 0) { - DRM_ERROR("disabled ip block: %d <%s>\n", + DRM_WARN("disabled ip block: %d <%s>\n", i, adev->ip_blocks[i].version->funcs->name); adev->ip_blocks[i].status.valid = false; } else { @@ -2220,14 +2326,16 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev) return r; /* Read BIOS */ - if (!amdgpu_get_bios(adev)) - return -EINVAL; + if (amdgpu_device_read_bios(adev)) { + if (!amdgpu_get_bios(adev)) + return -EINVAL; - r = amdgpu_atombios_init(adev); - if (r) { - dev_err(adev->dev, "amdgpu_atombios_init failed\n"); - amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_ATOMBIOS_INIT_FAIL, 0, 0); - return r; + r = amdgpu_atombios_init(adev); + if (r) { + dev_err(adev->dev, "amdgpu_atombios_init failed\n"); + amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_ATOMBIOS_INIT_FAIL, 0, 0); + return r; + } } /*get pf2vf msg info at it's earliest time*/ @@ -2376,6 +2484,8 @@ static int amdgpu_device_init_schedulers(struct amdgpu_device *adev) } } + amdgpu_xcp_update_partition_sched_list(adev); + return 0; } @@ -2533,8 +2643,10 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) goto init_failed; /* Don't init kfd if whole hive need to be reset during init */ - if (!adev->gmc.xgmi.pending_reset) + if (!adev->gmc.xgmi.pending_reset) { + kgd2kfd_init_zone_device(adev); amdgpu_amdkfd_device_init(adev); + } amdgpu_fru_get_product_info(adev); @@ -2759,8 +2871,9 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) DRM_ERROR("enable mgpu fan boost failed (%d).\n", r); /* For passthrough configuration on arcturus and aldebaran, enable special handling SBR */ - if (amdgpu_passthrough(adev) && ((adev->asic_type == CHIP_ARCTURUS && adev->gmc.xgmi.num_physical_nodes > 1)|| - adev->asic_type == CHIP_ALDEBARAN )) + if (amdgpu_passthrough(adev) && + ((adev->asic_type == CHIP_ARCTURUS && adev->gmc.xgmi.num_physical_nodes > 1) || + adev->asic_type == CHIP_ALDEBARAN)) amdgpu_dpm_handle_passthrough_sbr(adev, true); if (adev->gmc.xgmi.num_physical_nodes > 1) { @@ -3089,7 +3202,7 @@ static int amdgpu_device_ip_suspend_phase2(struct amdgpu_device *adev) } adev->ip_blocks[i].status.hw = false; /* handle putting the SMC in the appropriate state */ - if(!amdgpu_sriov_vf(adev)){ + if (!amdgpu_sriov_vf(adev)) { if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC) { r = amdgpu_dpm_set_mp1_state(adev, adev->mp1_state); if (r) { @@ -3608,6 +3721,8 @@ int amdgpu_device_init(struct amdgpu_device *adev, adev->smc_wreg = &amdgpu_invalid_wreg; adev->pcie_rreg = &amdgpu_invalid_rreg; adev->pcie_wreg = &amdgpu_invalid_wreg; + adev->pcie_rreg_ext = &amdgpu_invalid_rreg_ext; + adev->pcie_wreg_ext = &amdgpu_invalid_wreg_ext; adev->pciep_rreg = &amdgpu_invalid_rreg; adev->pciep_wreg = &amdgpu_invalid_wreg; adev->pcie_rreg64 = &amdgpu_invalid_rreg64; @@ -3633,6 +3748,7 @@ int amdgpu_device_init(struct amdgpu_device *adev, mutex_init(&adev->srbm_mutex); mutex_init(&adev->gfx.pipe_reserve_mutex); mutex_init(&adev->gfx.gfx_off_mutex); + mutex_init(&adev->gfx.partition_mutex); mutex_init(&adev->grbm_idx_mutex); mutex_init(&adev->mn_lock); mutex_init(&adev->virt.vf_errors.lock); @@ -3708,8 +3824,6 @@ int amdgpu_device_init(struct amdgpu_device *adev, DRM_INFO("register mmio base: 0x%08X\n", (uint32_t)adev->rmmio_base); DRM_INFO("register mmio size: %u\n", (unsigned)adev->rmmio_size); - amdgpu_device_get_pcie_info(adev); - if (amdgpu_mcbp) DRM_INFO("MCBP is enabled\n"); @@ -3725,6 +3839,8 @@ int amdgpu_device_init(struct amdgpu_device *adev, /* detect hw virtualization here */ amdgpu_detect_virtualization(adev); + amdgpu_device_get_pcie_info(adev); + r = amdgpu_device_get_job_timeout_settings(adev); if (r) { dev_err(adev->dev, "invalid lockup_timeout parameter syntax\n"); @@ -3753,21 +3869,24 @@ int amdgpu_device_init(struct amdgpu_device *adev, } /* enable PCIE atomic ops */ - if (amdgpu_sriov_vf(adev)) - adev->have_atomics_support = ((struct amd_sriov_msg_pf2vf_info *) - adev->virt.fw_reserve.p_pf2vf)->pcie_atomic_ops_support_flags == - (PCI_EXP_DEVCAP2_ATOMIC_COMP32 | PCI_EXP_DEVCAP2_ATOMIC_COMP64); + if (amdgpu_sriov_vf(adev)) { + if (adev->virt.fw_reserve.p_pf2vf) + adev->have_atomics_support = ((struct amd_sriov_msg_pf2vf_info *) + adev->virt.fw_reserve.p_pf2vf)->pcie_atomic_ops_support_flags == + (PCI_EXP_DEVCAP2_ATOMIC_COMP32 | PCI_EXP_DEVCAP2_ATOMIC_COMP64); /* APUs w/ gfx9 onwards doesn't reply on PCIe atomics, rather it is a * internal path natively support atomics, set have_atomics_support to true. */ - else if ((adev->flags & AMD_IS_APU) && - (adev->ip_versions[GC_HWIP][0] > IP_VERSION(9, 0, 0))) + } else if ((adev->flags & AMD_IS_APU) && + (adev->ip_versions[GC_HWIP][0] > IP_VERSION(9, 0, 0))) { adev->have_atomics_support = true; - else + } else { adev->have_atomics_support = !pci_enable_atomic_ops_to_root(adev->pdev, PCI_EXP_DEVCAP2_ATOMIC_COMP32 | PCI_EXP_DEVCAP2_ATOMIC_COMP64); + } + if (!adev->have_atomics_support) dev_info(adev->dev, "PCIE atomic ops is not supported\n"); @@ -3783,7 +3902,8 @@ int amdgpu_device_init(struct amdgpu_device *adev, amdgpu_reset_init(adev); /* detect if we are with an SRIOV vbios */ - amdgpu_device_detect_sriov_bios(adev); + if (adev->bios) + amdgpu_device_detect_sriov_bios(adev); /* check if we need to reset the asic * E.g., driver was not cleanly unloaded previously, etc. @@ -3835,25 +3955,27 @@ int amdgpu_device_init(struct amdgpu_device *adev, } } - if (adev->is_atom_fw) { - /* Initialize clocks */ - r = amdgpu_atomfirmware_get_clock_info(adev); - if (r) { - dev_err(adev->dev, "amdgpu_atomfirmware_get_clock_info failed\n"); - amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_ATOMBIOS_GET_CLOCK_FAIL, 0, 0); - goto failed; - } - } else { - /* Initialize clocks */ - r = amdgpu_atombios_get_clock_info(adev); - if (r) { - dev_err(adev->dev, "amdgpu_atombios_get_clock_info failed\n"); - amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_ATOMBIOS_GET_CLOCK_FAIL, 0, 0); - goto failed; + if (adev->bios) { + if (adev->is_atom_fw) { + /* Initialize clocks */ + r = amdgpu_atomfirmware_get_clock_info(adev); + if (r) { + dev_err(adev->dev, "amdgpu_atomfirmware_get_clock_info failed\n"); + amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_ATOMBIOS_GET_CLOCK_FAIL, 0, 0); + goto failed; + } + } else { + /* Initialize clocks */ + r = amdgpu_atombios_get_clock_info(adev); + if (r) { + dev_err(adev->dev, "amdgpu_atombios_get_clock_info failed\n"); + amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_ATOMBIOS_GET_CLOCK_FAIL, 0, 0); + goto failed; + } + /* init i2c buses */ + if (!amdgpu_device_has_dc_support(adev)) + amdgpu_atombios_i2c_init(adev); } - /* init i2c buses */ - if (!amdgpu_device_has_dc_support(adev)) - amdgpu_atombios_i2c_init(adev); } fence_driver_init: @@ -4019,7 +4141,7 @@ static void amdgpu_device_unmap_mmio(struct amdgpu_device *adev) adev->mman.aper_base_kaddr = NULL; /* Memory manager related */ - if (!adev->gmc.xgmi.connected_to_cpu) { + if (!adev->gmc.xgmi.connected_to_cpu && !adev->gmc.is_app_apu) { arch_phys_wc_del(adev->gmc.vram_mtrr); arch_io_free_memtype_wc(adev->gmc.aper_base, adev->gmc.aper_size); } @@ -4049,7 +4171,7 @@ void amdgpu_device_fini_hw(struct amdgpu_device *adev) /* disable all interrupts */ amdgpu_irq_disable_all(adev); - if (adev->mode_info.mode_config_initialized){ + if (adev->mode_info.mode_config_initialized) { if (!drm_drv_uses_atomic_modeset(adev_to_drm(adev))) drm_helper_force_disable_all(adev_to_drm(adev)); else @@ -4714,42 +4836,42 @@ bool amdgpu_device_should_recover_gpu(struct amdgpu_device *adev) int amdgpu_device_mode1_reset(struct amdgpu_device *adev) { - u32 i; - int ret = 0; + u32 i; + int ret = 0; - amdgpu_atombios_scratch_regs_engine_hung(adev, true); + amdgpu_atombios_scratch_regs_engine_hung(adev, true); - dev_info(adev->dev, "GPU mode1 reset\n"); + dev_info(adev->dev, "GPU mode1 reset\n"); - /* disable BM */ - pci_clear_master(adev->pdev); + /* disable BM */ + pci_clear_master(adev->pdev); - amdgpu_device_cache_pci_state(adev->pdev); + amdgpu_device_cache_pci_state(adev->pdev); - if (amdgpu_dpm_is_mode1_reset_supported(adev)) { - dev_info(adev->dev, "GPU smu mode1 reset\n"); - ret = amdgpu_dpm_mode1_reset(adev); - } else { - dev_info(adev->dev, "GPU psp mode1 reset\n"); - ret = psp_gpu_reset(adev); - } + if (amdgpu_dpm_is_mode1_reset_supported(adev)) { + dev_info(adev->dev, "GPU smu mode1 reset\n"); + ret = amdgpu_dpm_mode1_reset(adev); + } else { + dev_info(adev->dev, "GPU psp mode1 reset\n"); + ret = psp_gpu_reset(adev); + } - if (ret) - dev_err(adev->dev, "GPU mode1 reset failed\n"); + if (ret) + dev_err(adev->dev, "GPU mode1 reset failed\n"); - amdgpu_device_load_pci_state(adev->pdev); + amdgpu_device_load_pci_state(adev->pdev); - /* wait for asic to come out of reset */ - for (i = 0; i < adev->usec_timeout; i++) { - u32 memsize = adev->nbio.funcs->get_memsize(adev); + /* wait for asic to come out of reset */ + for (i = 0; i < adev->usec_timeout; i++) { + u32 memsize = adev->nbio.funcs->get_memsize(adev); - if (memsize != 0xffffffff) - break; - udelay(1); - } + if (memsize != 0xffffffff) + break; + udelay(1); + } - amdgpu_atombios_scratch_regs_engine_hung(adev, false); - return ret; + amdgpu_atombios_scratch_regs_engine_hung(adev, false); + return ret; } int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev, @@ -5478,7 +5600,7 @@ static void amdgpu_device_get_pcie_info(struct amdgpu_device *adev) adev->pm.pcie_mlw_mask = amdgpu_pcie_lane_cap; /* covers APUs as well */ - if (pci_is_root_bus(adev->pdev->bus)) { + if (pci_is_root_bus(adev->pdev->bus) && !amdgpu_passthrough(adev)) { if (adev->pm.pcie_gen_mask == 0) adev->pm.pcie_gen_mask = AMDGPU_DEFAULT_PCIE_GEN_MASK; if (adev->pm.pcie_mlw_mask == 0) @@ -5959,6 +6081,7 @@ void amdgpu_device_halt(struct amdgpu_device *adev) struct pci_dev *pdev = adev->pdev; struct drm_device *ddev = adev_to_drm(adev); + amdgpu_xcp_dev_unplug(adev); drm_dev_unplug(ddev); amdgpu_irq_disable_all(adev); @@ -6079,3 +6202,31 @@ bool amdgpu_device_has_display_hardware(struct amdgpu_device *adev) return true; } } + +uint32_t amdgpu_device_wait_on_rreg(struct amdgpu_device *adev, + uint32_t inst, uint32_t reg_addr, char reg_name[], + uint32_t expected_value, uint32_t mask) +{ + uint32_t ret = 0; + uint32_t old_ = 0; + uint32_t tmp_ = RREG32(reg_addr); + uint32_t loop = adev->usec_timeout; + + while ((tmp_ & (mask)) != (expected_value)) { + if (old_ != tmp_) { + loop = adev->usec_timeout; + old_ = tmp_; + } else + udelay(1); + tmp_ = RREG32(reg_addr); + loop--; + if (!loop) { + DRM_WARN("Register(%d) [%s] failed to reach value 0x%08x != 0x%08xn", + inst, reg_name, (uint32_t)expected_value, + (uint32_t)(tmp_ & (mask))); + ret = -ETIMEDOUT; + break; + } + } + return ret; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index 0ecce0b92b82e4a19eb2d093581b0a8d3c755a00..8e1cfc87122d675c46c06593f30cfc2c60901ca1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -30,6 +30,7 @@ #include "soc15.h" #include "gfx_v9_0.h" +#include "gfx_v9_4_3.h" #include "gmc_v9_0.h" #include "df_v1_7.h" #include "df_v3_6.h" @@ -76,12 +77,15 @@ #include "jpeg_v3_0.h" #include "vcn_v4_0.h" #include "jpeg_v4_0.h" +#include "vcn_v4_0_3.h" +#include "jpeg_v4_0_3.h" #include "amdgpu_vkms.h" #include "mes_v10_1.h" #include "mes_v11_0.h" #include "smuio_v11_0.h" #include "smuio_v11_0_6.h" #include "smuio_v13_0.h" +#include "smuio_v13_0_3.h" #include "smuio_v13_0_6.h" #define FIRMWARE_IP_DISCOVERY "amdgpu/ip_discovery.bin" @@ -200,14 +204,44 @@ static int hw_id_map[MAX_HWIP] = { [PCIE_HWIP] = PCIE_HWID, }; -static int amdgpu_discovery_read_binary_from_vram(struct amdgpu_device *adev, uint8_t *binary) +static int amdgpu_discovery_read_binary_from_sysmem(struct amdgpu_device *adev, uint8_t *binary) +{ + u64 tmr_offset, tmr_size, pos; + void *discv_regn; + int ret; + + ret = amdgpu_acpi_get_tmr_info(adev, &tmr_offset, &tmr_size); + if (ret) + return ret; + + pos = tmr_offset + tmr_size - DISCOVERY_TMR_OFFSET; + + /* This region is read-only and reserved from system use */ + discv_regn = memremap(pos, adev->mman.discovery_tmr_size, MEMREMAP_WC); + if (discv_regn) { + memcpy(binary, discv_regn, adev->mman.discovery_tmr_size); + memunmap(discv_regn); + return 0; + } + + return -ENOENT; +} + +static int amdgpu_discovery_read_binary_from_mem(struct amdgpu_device *adev, + uint8_t *binary) { uint64_t vram_size = (uint64_t)RREG32(mmRCC_CONFIG_MEMSIZE) << 20; - uint64_t pos = vram_size - DISCOVERY_TMR_OFFSET; + int ret = 0; - amdgpu_device_vram_access(adev, pos, (uint32_t *)binary, - adev->mman.discovery_tmr_size, false); - return 0; + if (vram_size) { + uint64_t pos = vram_size - DISCOVERY_TMR_OFFSET; + amdgpu_device_vram_access(adev, pos, (uint32_t *)binary, + adev->mman.discovery_tmr_size, false); + } else { + ret = amdgpu_discovery_read_binary_from_sysmem(adev, binary); + } + + return ret; } static int amdgpu_discovery_read_binary_from_file(struct amdgpu_device *adev, uint8_t *binary) @@ -280,6 +314,7 @@ static void amdgpu_discovery_harvest_config_quirk(struct amdgpu_device *adev) case 0xCF: case 0xDF: adev->vcn.harvest_config |= AMDGPU_VCN_HARVEST_VCN1; + adev->vcn.inst_mask &= ~AMDGPU_VCN_HARVEST_VCN1; break; default: break; @@ -301,33 +336,30 @@ static int amdgpu_discovery_init(struct amdgpu_device *adev) if (!adev->mman.discovery_bin) return -ENOMEM; - r = amdgpu_discovery_read_binary_from_vram(adev, adev->mman.discovery_bin); - if (r) { - dev_err(adev->dev, "failed to read ip discovery binary from vram\n"); - r = -EINVAL; - goto out; - } - - if (!amdgpu_discovery_verify_binary_signature(adev->mman.discovery_bin) || amdgpu_discovery == 2) { - /* ignore the discovery binary from vram if discovery=2 in kernel module parameter */ - if (amdgpu_discovery == 2) - dev_info(adev->dev,"force read ip discovery binary from file"); - else - dev_warn(adev->dev, "get invalid ip discovery binary signature from vram\n"); - - /* retry read ip discovery binary from file */ + /* Read from file if it is the preferred option */ + if (amdgpu_discovery == 2) { + dev_info(adev->dev, "use ip discovery information from file"); r = amdgpu_discovery_read_binary_from_file(adev, adev->mman.discovery_bin); + if (r) { dev_err(adev->dev, "failed to read ip discovery binary from file\n"); r = -EINVAL; goto out; } - /* check the ip discovery binary signature */ - if(!amdgpu_discovery_verify_binary_signature(adev->mman.discovery_bin)) { - dev_warn(adev->dev, "get invalid ip discovery binary signature from file\n"); - r = -EINVAL; + + } else { + r = amdgpu_discovery_read_binary_from_mem( + adev, adev->mman.discovery_bin); + if (r) goto out; - } + } + + /* check the ip discovery binary signature */ + if (!amdgpu_discovery_verify_binary_signature(adev->mman.discovery_bin)) { + dev_err(adev->dev, + "get invalid ip discovery binary signature\n"); + r = -EINVAL; + goto out; } bhdr = (struct binary_header *)adev->mman.discovery_bin; @@ -471,11 +503,11 @@ void amdgpu_discovery_fini(struct amdgpu_device *adev) adev->mman.discovery_bin = NULL; } -static int amdgpu_discovery_validate_ip(const struct ip *ip) +static int amdgpu_discovery_validate_ip(const struct ip_v4 *ip) { - if (ip->number_instance >= HWIP_MAX_INSTANCE) { - DRM_ERROR("Unexpected number_instance (%d) from ip discovery blob\n", - ip->number_instance); + if (ip->instance_number >= HWIP_MAX_INSTANCE) { + DRM_ERROR("Unexpected instance_number (%d) from ip discovery blob\n", + ip->instance_number); return -EINVAL; } if (le16_to_cpu(ip->hw_id) >= HW_ID_MAX) { @@ -493,7 +525,7 @@ static void amdgpu_discovery_read_harvest_bit_per_ip(struct amdgpu_device *adev, struct binary_header *bhdr; struct ip_discovery_header *ihdr; struct die_header *dhdr; - struct ip *ip; + struct ip_v4 *ip; uint16_t die_offset, ip_offset, num_dies, num_ips; int i, j; @@ -510,29 +542,41 @@ static void amdgpu_discovery_read_harvest_bit_per_ip(struct amdgpu_device *adev, ip_offset = die_offset + sizeof(*dhdr); for (j = 0; j < num_ips; j++) { - ip = (struct ip *)(adev->mman.discovery_bin + ip_offset); + ip = (struct ip_v4 *)(adev->mman.discovery_bin + ip_offset); if (amdgpu_discovery_validate_ip(ip)) goto next_ip; - if (le16_to_cpu(ip->harvest) == 1) { + if (le16_to_cpu(ip->variant) == 1) { switch (le16_to_cpu(ip->hw_id)) { case VCN_HWID: (*vcn_harvest_count)++; - if (ip->number_instance == 0) + if (ip->instance_number == 0) { adev->vcn.harvest_config |= AMDGPU_VCN_HARVEST_VCN0; - else + adev->vcn.inst_mask &= + ~AMDGPU_VCN_HARVEST_VCN0; + adev->jpeg.inst_mask &= + ~AMDGPU_VCN_HARVEST_VCN0; + } else { adev->vcn.harvest_config |= AMDGPU_VCN_HARVEST_VCN1; + adev->vcn.inst_mask &= + ~AMDGPU_VCN_HARVEST_VCN1; + adev->jpeg.inst_mask &= + ~AMDGPU_VCN_HARVEST_VCN1; + } break; case DMU_HWID: adev->harvest_ip_mask |= AMD_HARVEST_IP_DMU_MASK; break; default: break; - } - } + } + } next_ip: - ip_offset += struct_size(ip, base_address, ip->num_base_address); + if (ihdr->base_addr_64_bit) + ip_offset += struct_size(ip, base_address_64, ip->num_base_address); + else + ip_offset += struct_size(ip, base_address, ip->num_base_address); } } } @@ -564,10 +608,15 @@ static void amdgpu_discovery_read_from_harvest_table(struct amdgpu_device *adev, switch (le16_to_cpu(harvest_info->list[i].hw_id)) { case VCN_HWID: (*vcn_harvest_count)++; - if (harvest_info->list[i].number_instance == 0) - adev->vcn.harvest_config |= AMDGPU_VCN_HARVEST_VCN0; - else - adev->vcn.harvest_config |= AMDGPU_VCN_HARVEST_VCN1; + adev->vcn.harvest_config |= + (1 << harvest_info->list[i].number_instance); + adev->jpeg.harvest_config |= + (1 << harvest_info->list[i].number_instance); + + adev->vcn.inst_mask &= + ~(1U << harvest_info->list[i].number_instance); + adev->jpeg.inst_mask &= + ~(1U << harvest_info->list[i].number_instance); break; case DMU_HWID: adev->harvest_ip_mask |= AMD_HARVEST_IP_DMU_MASK; @@ -577,6 +626,14 @@ static void amdgpu_discovery_read_from_harvest_table(struct amdgpu_device *adev, 1 << (le16_to_cpu(harvest_info->list[i].number_instance)); (*umc_harvest_count)++; break; + case GC_HWID: + adev->gfx.xcc_mask &= + ~(1U << harvest_info->list[i].number_instance); + break; + case SDMA0_HWID: + adev->sdma.sdma_mask &= + ~(1U << harvest_info->list[i].number_instance); + break; default: break; } @@ -836,9 +893,40 @@ static void ip_disc_release(struct kobject *kobj) kfree(ip_top); } +static uint8_t amdgpu_discovery_get_harvest_info(struct amdgpu_device *adev, + uint16_t hw_id, uint8_t inst) +{ + uint8_t harvest = 0; + + /* Until a uniform way is figured, get mask based on hwid */ + switch (hw_id) { + case VCN_HWID: + harvest = ((1 << inst) & adev->vcn.inst_mask) == 0; + break; + case DMU_HWID: + if (adev->harvest_ip_mask & AMD_HARVEST_IP_DMU_MASK) + harvest = 0x1; + break; + case UMC_HWID: + /* TODO: It needs another parsing; for now, ignore.*/ + break; + case GC_HWID: + harvest = ((1 << inst) & adev->gfx.xcc_mask) == 0; + break; + case SDMA0_HWID: + harvest = ((1 << inst) & adev->sdma.sdma_mask) == 0; + break; + default: + break; + } + + return harvest; +} + static int amdgpu_discovery_sysfs_ips(struct amdgpu_device *adev, struct ip_die_entry *ip_die_entry, - const size_t _ip_offset, const int num_ips) + const size_t _ip_offset, const int num_ips, + bool reg_base_64) { int ii, jj, kk, res; @@ -852,10 +940,10 @@ static int amdgpu_discovery_sysfs_ips(struct amdgpu_device *adev, size_t ip_offset = _ip_offset; for (jj = 0; jj < num_ips; jj++) { - struct ip *ip; + struct ip_v4 *ip; struct ip_hw_instance *ip_hw_instance; - ip = (struct ip *)(adev->mman.discovery_bin + ip_offset); + ip = (struct ip_v4 *)(adev->mman.discovery_bin + ip_offset); if (amdgpu_discovery_validate_ip(ip) || le16_to_cpu(ip->hw_id) != ii) goto next_ip; @@ -903,22 +991,35 @@ static int amdgpu_discovery_sysfs_ips(struct amdgpu_device *adev, return -ENOMEM; } ip_hw_instance->hw_id = le16_to_cpu(ip->hw_id); /* == ii */ - ip_hw_instance->num_instance = ip->number_instance; + ip_hw_instance->num_instance = ip->instance_number; ip_hw_instance->major = ip->major; ip_hw_instance->minor = ip->minor; ip_hw_instance->revision = ip->revision; - ip_hw_instance->harvest = ip->harvest; + ip_hw_instance->harvest = + amdgpu_discovery_get_harvest_info( + adev, ip_hw_instance->hw_id, + ip_hw_instance->num_instance); ip_hw_instance->num_base_addresses = ip->num_base_address; - for (kk = 0; kk < ip_hw_instance->num_base_addresses; kk++) - ip_hw_instance->base_addr[kk] = ip->base_address[kk]; + for (kk = 0; kk < ip_hw_instance->num_base_addresses; kk++) { + if (reg_base_64) + ip_hw_instance->base_addr[kk] = + lower_32_bits(le64_to_cpu(ip->base_address_64[kk])) & 0x3FFFFFFF; + else + ip_hw_instance->base_addr[kk] = ip->base_address[kk]; + } kobject_init(&ip_hw_instance->kobj, &ip_hw_instance_ktype); ip_hw_instance->kobj.kset = &ip_hw_id->hw_id_kset; res = kobject_add(&ip_hw_instance->kobj, NULL, "%d", ip_hw_instance->num_instance); next_ip: - ip_offset += struct_size(ip, base_address, ip->num_base_address); + if (reg_base_64) + ip_offset += struct_size(ip, base_address_64, + ip->num_base_address); + else + ip_offset += struct_size(ip, base_address, + ip->num_base_address); } } @@ -972,7 +1073,7 @@ static int amdgpu_discovery_sysfs_recurse(struct amdgpu_device *adev) return res; } - amdgpu_discovery_sysfs_ips(adev, ip_die_entry, ip_offset, num_ips); + amdgpu_discovery_sysfs_ips(adev, ip_die_entry, ip_offset, num_ips, !!ihdr->base_addr_64_bit); } return 0; @@ -983,6 +1084,9 @@ static int amdgpu_discovery_sysfs_init(struct amdgpu_device *adev) struct kset *die_kset; int res, ii; + if (!adev->mman.discovery_bin) + return -EINVAL; + adev->ip_top = kzalloc(sizeof(*adev->ip_top), GFP_KERNEL); if (!adev->ip_top) return -ENOMEM; @@ -1082,7 +1186,7 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev) struct binary_header *bhdr; struct ip_discovery_header *ihdr; struct die_header *dhdr; - struct ip *ip; + struct ip_v4 *ip; uint16_t die_offset; uint16_t ip_offset; uint16_t num_dies; @@ -1098,6 +1202,10 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev) return r; } + adev->gfx.xcc_mask = 0; + adev->sdma.sdma_mask = 0; + adev->vcn.inst_mask = 0; + adev->jpeg.inst_mask = 0; bhdr = (struct binary_header *)adev->mman.discovery_bin; ihdr = (struct ip_discovery_header *)(adev->mman.discovery_bin + le16_to_cpu(bhdr->table_list[IP_DISCOVERY].offset)); @@ -1121,7 +1229,7 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev) le16_to_cpu(dhdr->die_id), num_ips); for (j = 0; j < num_ips; j++) { - ip = (struct ip *)(adev->mman.discovery_bin + ip_offset); + ip = (struct ip_v4 *)(adev->mman.discovery_bin + ip_offset); if (amdgpu_discovery_validate_ip(ip)) goto next_ip; @@ -1131,7 +1239,7 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev) DRM_DEBUG("%s(%d) #%d v%d.%d.%d:\n", hw_id_names[le16_to_cpu(ip->hw_id)], le16_to_cpu(ip->hw_id), - ip->number_instance, + ip->instance_number, ip->major, ip->minor, ip->revision); @@ -1145,23 +1253,33 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev) adev->vcn.vcn_config[adev->vcn.num_vcn_inst] = ip->revision & 0xc0; ip->revision &= ~0xc0; - if (adev->vcn.num_vcn_inst < AMDGPU_MAX_VCN_INSTANCES) + if (adev->vcn.num_vcn_inst < + AMDGPU_MAX_VCN_INSTANCES) { adev->vcn.num_vcn_inst++; - else + adev->vcn.inst_mask |= + (1U << ip->instance_number); + adev->jpeg.inst_mask |= + (1U << ip->instance_number); + } else { dev_err(adev->dev, "Too many VCN instances: %d vs %d\n", adev->vcn.num_vcn_inst + 1, AMDGPU_MAX_VCN_INSTANCES); + } } if (le16_to_cpu(ip->hw_id) == SDMA0_HWID || le16_to_cpu(ip->hw_id) == SDMA1_HWID || le16_to_cpu(ip->hw_id) == SDMA2_HWID || le16_to_cpu(ip->hw_id) == SDMA3_HWID) { - if (adev->sdma.num_instances < AMDGPU_MAX_SDMA_INSTANCES) + if (adev->sdma.num_instances < + AMDGPU_MAX_SDMA_INSTANCES) { adev->sdma.num_instances++; - else + adev->sdma.sdma_mask |= + (1U << ip->instance_number); + } else { dev_err(adev->dev, "Too many SDMA instances: %d vs %d\n", adev->sdma.num_instances + 1, AMDGPU_MAX_SDMA_INSTANCES); + } } if (le16_to_cpu(ip->hw_id) == UMC_HWID) { @@ -1169,20 +1287,38 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev) adev->umc.node_inst_num++; } + if (le16_to_cpu(ip->hw_id) == GC_HWID) + adev->gfx.xcc_mask |= + (1U << ip->instance_number); + for (k = 0; k < num_base_address; k++) { /* * convert the endianness of base addresses in place, * so that we don't need to convert them when accessing adev->reg_offset. */ - ip->base_address[k] = le32_to_cpu(ip->base_address[k]); + if (ihdr->base_addr_64_bit) + /* Truncate the 64bit base address from ip discovery + * and only store lower 32bit ip base in reg_offset[]. + * Bits > 32 follows ASIC specific format, thus just + * discard them and handle it within specific ASIC. + * By this way reg_offset[] and related helpers can + * stay unchanged. + * The base address is in dwords, thus clear the + * highest 2 bits to store. + */ + ip->base_address[k] = + lower_32_bits(le64_to_cpu(ip->base_address_64[k])) & 0x3FFFFFFF; + else + ip->base_address[k] = le32_to_cpu(ip->base_address[k]); DRM_DEBUG("\t0x%08x\n", ip->base_address[k]); } for (hw_ip = 0; hw_ip < MAX_HWIP; hw_ip++) { - if (hw_id_map[hw_ip] == le16_to_cpu(ip->hw_id)) { + if (hw_id_map[hw_ip] == le16_to_cpu(ip->hw_id) && + hw_id_map[hw_ip] != 0) { DRM_DEBUG("set register base offset for %s\n", hw_id_names[le16_to_cpu(ip->hw_id)]); - adev->reg_offset[hw_ip][ip->number_instance] = + adev->reg_offset[hw_ip][ip->instance_number] = ip->base_address; /* Instance support is somewhat inconsistent. * SDMA is a good example. Sienna cichlid has 4 total @@ -1193,69 +1329,22 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev) * example. On most chips there are multiple instances * with the same HWID. */ - adev->ip_versions[hw_ip][ip->number_instance] = + adev->ip_versions[hw_ip][ip->instance_number] = IP_VERSION(ip->major, ip->minor, ip->revision); } } next_ip: - ip_offset += struct_size(ip, base_address, ip->num_base_address); + if (ihdr->base_addr_64_bit) + ip_offset += struct_size(ip, base_address_64, ip->num_base_address); + else + ip_offset += struct_size(ip, base_address, ip->num_base_address); } } - amdgpu_discovery_sysfs_init(adev); - return 0; } -int amdgpu_discovery_get_ip_version(struct amdgpu_device *adev, int hw_id, int number_instance, - int *major, int *minor, int *revision) -{ - struct binary_header *bhdr; - struct ip_discovery_header *ihdr; - struct die_header *dhdr; - struct ip *ip; - uint16_t die_offset; - uint16_t ip_offset; - uint16_t num_dies; - uint16_t num_ips; - int i, j; - - if (!adev->mman.discovery_bin) { - DRM_ERROR("ip discovery uninitialized\n"); - return -EINVAL; - } - - bhdr = (struct binary_header *)adev->mman.discovery_bin; - ihdr = (struct ip_discovery_header *)(adev->mman.discovery_bin + - le16_to_cpu(bhdr->table_list[IP_DISCOVERY].offset)); - num_dies = le16_to_cpu(ihdr->num_dies); - - for (i = 0; i < num_dies; i++) { - die_offset = le16_to_cpu(ihdr->die_info[i].die_offset); - dhdr = (struct die_header *)(adev->mman.discovery_bin + die_offset); - num_ips = le16_to_cpu(dhdr->num_ips); - ip_offset = die_offset + sizeof(*dhdr); - - for (j = 0; j < num_ips; j++) { - ip = (struct ip *)(adev->mman.discovery_bin + ip_offset); - - if ((le16_to_cpu(ip->hw_id) == hw_id) && (ip->number_instance == number_instance)) { - if (major) - *major = ip->major; - if (minor) - *minor = ip->minor; - if (revision) - *revision = ip->revision; - return 0; - } - ip_offset += struct_size(ip, base_address, ip->num_base_address); - } - } - - return -EINVAL; -} - static void amdgpu_discovery_harvest_ip(struct amdgpu_device *adev) { int vcn_harvest_count = 0; @@ -1266,7 +1355,8 @@ static void amdgpu_discovery_harvest_ip(struct amdgpu_device *adev) * so read harvest bit per IP data structure to set * harvest configuration. */ - if (adev->ip_versions[GC_HWIP][0] < IP_VERSION(10, 2, 0)) { + if (adev->ip_versions[GC_HWIP][0] < IP_VERSION(10, 2, 0) && + adev->ip_versions[GC_HWIP][0] != IP_VERSION(9, 4, 3)) { if ((adev->pdev->device == 0x731E && (adev->pdev->revision == 0xC6 || adev->pdev->revision == 0xC7)) || @@ -1425,6 +1515,7 @@ static int amdgpu_discovery_get_mall_info(struct amdgpu_device *adev) mall_size += mall_size_per_umc; } adev->gmc.mall_size = mall_size; + adev->gmc.m_half_use = half_use; break; default: dev_err(adev->dev, @@ -1706,6 +1797,7 @@ static int amdgpu_discovery_set_smu_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(13, 0, 3): case IP_VERSION(13, 0, 4): case IP_VERSION(13, 0, 5): + case IP_VERSION(13, 0, 6): case IP_VERSION(13, 0, 7): case IP_VERSION(13, 0, 8): case IP_VERSION(13, 0, 10): @@ -1804,6 +1896,11 @@ static int amdgpu_discovery_set_gc_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(9, 4, 2): amdgpu_device_ip_block_add(adev, &gfx_v9_0_ip_block); break; + case IP_VERSION(9, 4, 3): + if (!amdgpu_exp_hw_support) + return -EINVAL; + amdgpu_device_ip_block_add(adev, &gfx_v9_4_3_ip_block); + break; case IP_VERSION(10, 1, 10): case IP_VERSION(10, 1, 2): case IP_VERSION(10, 1, 1): @@ -1939,7 +2036,6 @@ static int amdgpu_discovery_set_mm_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(3, 1, 1): case IP_VERSION(3, 1, 2): case IP_VERSION(3, 0, 2): - case IP_VERSION(3, 0, 192): amdgpu_device_ip_block_add(adev, &vcn_v3_0_ip_block); if (!amdgpu_sriov_vf(adev)) amdgpu_device_ip_block_add(adev, &jpeg_v3_0_ip_block); @@ -1952,7 +2048,11 @@ static int amdgpu_discovery_set_mm_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(4, 0, 4): amdgpu_device_ip_block_add(adev, &vcn_v4_0_ip_block); amdgpu_device_ip_block_add(adev, &jpeg_v4_0_ip_block); - return 0; + break; + case IP_VERSION(4, 0, 3): + amdgpu_device_ip_block_add(adev, &vcn_v4_0_3_ip_block); + amdgpu_device_ip_block_add(adev, &jpeg_v4_0_3_ip_block); + break; default: dev_err(adev->dev, "Failed to add vcn/jpeg ip block(UVD_HWIP:0x%x)\n", @@ -2000,6 +2100,17 @@ static int amdgpu_discovery_set_mes_ip_blocks(struct amdgpu_device *adev) return 0; } +static void amdgpu_discovery_init_soc_config(struct amdgpu_device *adev) +{ + switch (adev->ip_versions[GC_HWIP][0]) { + case IP_VERSION(9, 4, 3): + aqua_vanjaram_init_soc_config(adev); + break; + default: + break; + } +} + int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev) { int r; @@ -2177,6 +2288,9 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev) break; } + amdgpu_discovery_init_soc_config(adev); + amdgpu_discovery_sysfs_init(adev); + switch (adev->ip_versions[GC_HWIP][0]) { case IP_VERSION(9, 0, 1): case IP_VERSION(9, 2, 1): @@ -2387,6 +2501,12 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(13, 0, 2): adev->smuio.funcs = &smuio_v13_0_funcs; break; + case IP_VERSION(13, 0, 3): + adev->smuio.funcs = &smuio_v13_0_3_funcs; + if (adev->smuio.funcs->get_pkg_type(adev) == AMDGPU_PKG_TYPE_APU) { + adev->flags |= AMD_IS_APU; + } + break; case IP_VERSION(13, 0, 6): case IP_VERSION(13, 0, 8): adev->smuio.funcs = &smuio_v13_0_6_funcs; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h index 8563dd4a7dc2e525e3607f9dffa2a19db5b0c616..3a2f347bd50d5c7ee4199d49e4533b19452771e4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h @@ -24,12 +24,10 @@ #ifndef __AMDGPU_DISCOVERY__ #define __AMDGPU_DISCOVERY__ -#define DISCOVERY_TMR_SIZE (4 << 10) +#define DISCOVERY_TMR_SIZE (8 << 10) #define DISCOVERY_TMR_OFFSET (64 << 10) void amdgpu_discovery_fini(struct amdgpu_device *adev); -int amdgpu_discovery_get_ip_version(struct amdgpu_device *adev, int hw_id, int number_instance, - int *major, int *minor, int *revision); int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev); #endif /* __AMDGPU_DISCOVERY__ */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c index d60fe7eb5579aeae48961105b832e1a47fe1300c..b702f499f5fb3585f8db55ec91c7e0c14f0a4215 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c @@ -98,7 +98,7 @@ static void amdgpu_display_flip_callback(struct dma_fence *f, static bool amdgpu_display_flip_handle_fence(struct amdgpu_flip_work *work, struct dma_fence **f) { - struct dma_fence *fence= *f; + struct dma_fence *fence = *f; if (fence == NULL) return false; @@ -1252,21 +1252,21 @@ const struct drm_mode_config_funcs amdgpu_mode_funcs = { .fb_create = amdgpu_display_user_framebuffer_create, }; -static const struct drm_prop_enum_list amdgpu_underscan_enum_list[] = -{ { UNDERSCAN_OFF, "off" }, +static const struct drm_prop_enum_list amdgpu_underscan_enum_list[] = { + { UNDERSCAN_OFF, "off" }, { UNDERSCAN_ON, "on" }, { UNDERSCAN_AUTO, "auto" }, }; -static const struct drm_prop_enum_list amdgpu_audio_enum_list[] = -{ { AMDGPU_AUDIO_DISABLE, "off" }, +static const struct drm_prop_enum_list amdgpu_audio_enum_list[] = { + { AMDGPU_AUDIO_DISABLE, "off" }, { AMDGPU_AUDIO_ENABLE, "on" }, { AMDGPU_AUDIO_AUTO, "auto" }, }; /* XXX support different dither options? spatial, temporal, both, etc. */ -static const struct drm_prop_enum_list amdgpu_dither_enum_list[] = -{ { AMDGPU_FMT_DITHER_DISABLE, "off" }, +static const struct drm_prop_enum_list amdgpu_dither_enum_list[] = { + { AMDGPU_FMT_DITHER_DISABLE, "off" }, { AMDGPU_FMT_DITHER_ENABLE, "on" }, }; @@ -1496,8 +1496,7 @@ int amdgpu_display_get_crtc_scanoutpos(struct drm_device *dev, ret |= DRM_SCANOUTPOS_ACCURATE; vbl_start = vbl & 0x1fff; vbl_end = (vbl >> 16) & 0x1fff; - } - else { + } else { /* No: Fake something reasonable which gives at least ok results. */ vbl_start = mode->crtc_vdisplay; vbl_end = 0; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c index 0c001bb8fc2b03aa69c61d00048c0ba37c9ce2e2..12210598e5b8e64c8f88d50eaec69ff3f3161bac 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c @@ -149,7 +149,7 @@ static struct sg_table *amdgpu_dma_buf_map(struct dma_buf_attachment *attach, if (!bo->tbo.pin_count) { /* move buffer into GTT or VRAM */ struct ttm_operation_ctx ctx = { false, false }; - unsigned domains = AMDGPU_GEM_DOMAIN_GTT; + unsigned int domains = AMDGPU_GEM_DOMAIN_GTT; if (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM && attach->peer2peer) { @@ -336,7 +336,7 @@ amdgpu_dma_buf_create_obj(struct drm_device *dev, struct dma_buf *dma_buf) ret = amdgpu_gem_object_create(adev, dma_buf->size, PAGE_SIZE, AMDGPU_GEM_DOMAIN_CPU, flags, - ttm_bo_type_sg, resv, &gobj); + ttm_bo_type_sg, resv, &gobj, 0); if (ret) goto error; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h index 8fd11497fabaa61e48d9a15c2187357fca87fb62..f637574644c0ee16955365dc59cb4f6d4c82264f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h @@ -59,7 +59,7 @@ struct amdgpu_doorbell_index { uint32_t gfx_ring1; uint32_t gfx_userqueue_start; uint32_t gfx_userqueue_end; - uint32_t sdma_engine[8]; + uint32_t sdma_engine[16]; uint32_t mes_ring0; uint32_t mes_ring1; uint32_t ih; @@ -86,6 +86,8 @@ struct amdgpu_doorbell_index { uint32_t max_assignment; /* Per engine SDMA doorbell size in dword */ uint32_t sdma_doorbell_range; + /* Per xcc doorbell size for KIQ/KCQ */ + uint32_t xcc_doorbell_range; }; typedef enum _AMDGPU_DOORBELL_ASSIGNMENT @@ -164,7 +166,15 @@ typedef enum _AMDGPU_VEGA20_DOORBELL_ASSIGNMENT AMDGPU_VEGA20_DOORBELL64_FIRST_NON_CP = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE0, AMDGPU_VEGA20_DOORBELL64_LAST_NON_CP = AMDGPU_VEGA20_DOORBELL64_VCE_RING6_7, - AMDGPU_VEGA20_DOORBELL_MAX_ASSIGNMENT = 0x18F, + /* kiq/kcq from second XCD. Max 8 XCDs */ + AMDGPU_VEGA20_DOORBELL_XCC1_KIQ_START = 0x190, + /* 8 compute rings per GC. Max to 0x1CE */ + AMDGPU_VEGA20_DOORBELL_XCC1_MEC_RING0_START = 0x197, + + /* AID1 SDMA: 0x1D0 ~ 0x1F7 */ + AMDGPU_VEGA20_DOORBELL_AID1_sDMA_START = 0x1D0, + + AMDGPU_VEGA20_DOORBELL_MAX_ASSIGNMENT = 0x1F7, AMDGPU_VEGA20_DOORBELL_INVALID = 0xFFFF } AMDGPU_VEGA20_DOORBELL_ASSIGNMENT; @@ -301,6 +311,36 @@ typedef enum _AMDGPU_DOORBELL64_ASSIGNMENT AMDGPU_DOORBELL64_INVALID = 0xFFFF } AMDGPU_DOORBELL64_ASSIGNMENT; +typedef enum _AMDGPU_DOORBELL_ASSIGNMENT_LAYOUT1 { + /* XCC0: 0x00 ~20, XCC1: 20 ~ 2F ... */ + + /* KIQ/HIQ/DIQ */ + AMDGPU_DOORBELL_LAYOUT1_KIQ_START = 0x000, + AMDGPU_DOORBELL_LAYOUT1_HIQ = 0x001, + AMDGPU_DOORBELL_LAYOUT1_DIQ = 0x002, + /* Compute: 0x08 ~ 0x20 */ + AMDGPU_DOORBELL_LAYOUT1_MEC_RING_START = 0x008, + AMDGPU_DOORBELL_LAYOUT1_MEC_RING_END = 0x00F, + AMDGPU_DOORBELL_LAYOUT1_USERQUEUE_START = 0x010, + AMDGPU_DOORBELL_LAYOUT1_USERQUEUE_END = 0x01F, + AMDGPU_DOORBELL_LAYOUT1_XCC_RANGE = 0x020, + + /* SDMA: 0x100 ~ 0x19F */ + AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_START = 0x100, + AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_END = 0x19F, + /* IH: 0x1A0 ~ 0x1AF */ + AMDGPU_DOORBELL_LAYOUT1_IH = 0x1A0, + /* VCN: 0x1B0 ~ 0x1D4 */ + AMDGPU_DOORBELL_LAYOUT1_VCN_START = 0x1B0, + AMDGPU_DOORBELL_LAYOUT1_VCN_END = 0x1D4, + + AMDGPU_DOORBELL_LAYOUT1_FIRST_NON_CP = AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_START, + AMDGPU_DOORBELL_LAYOUT1_LAST_NON_CP = AMDGPU_DOORBELL_LAYOUT1_VCN_END, + + AMDGPU_DOORBELL_LAYOUT1_MAX_ASSIGNMENT = 0x1D4, + AMDGPU_DOORBELL_LAYOUT1_INVALID = 0xFFFF +} AMDGPU_DOORBELL_ASSIGNMENT_LAYOUT1; + u32 amdgpu_mm_rdoorbell(struct amdgpu_device *adev, u32 index); void amdgpu_mm_wdoorbell(struct amdgpu_device *adev, u32 index, u32 v); u64 amdgpu_mm_rdoorbell64(struct amdgpu_device *adev, u32 index); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 7681f79f462eb7c97ef6eb4e55001d7aca3dc954..43613569801b614c09832508c48c628d3fb01844 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -50,6 +50,7 @@ #include "amdgpu_ras.h" #include "amdgpu_xgmi.h" #include "amdgpu_reset.h" +#include "../amdxcp/amdgpu_xcp_drv.h" /* * KMS wrapper. @@ -110,9 +111,11 @@ * 3.52.0 - Add AMDGPU_IDS_FLAGS_CONFORMANT_TRUNC_COORD, add device_info fields: * tcp_cache_size, num_sqc_per_wgp, sqc_data_cache_size, sqc_inst_cache_size, * gl1c_cache_size, gl2c_cache_size, mall_size, enabled_rb_pipes_mask_hi + * 3.53.0 - Support for GFX11 CP GFX shadowing + * 3.54.0 - Add AMDGPU_CTX_QUERY2_FLAGS_RESET_IN_PROGRESS support */ #define KMS_DRIVER_MAJOR 3 -#define KMS_DRIVER_MINOR 52 +#define KMS_DRIVER_MINOR 54 #define KMS_DRIVER_PATCHLEVEL 0 unsigned int amdgpu_vram_limit = UINT_MAX; @@ -150,7 +153,7 @@ uint amdgpu_pg_mask = 0xffffffff; uint amdgpu_sdma_phase_quantum = 32; char *amdgpu_disable_cu; char *amdgpu_virtual_display; - +bool enforce_isolation; /* * OverDrive(bit 14) disabled by default * GFX DCS(bit 19) disabled by default @@ -191,6 +194,7 @@ int amdgpu_smartshift_bias; int amdgpu_use_xgmi_p2p = 1; int amdgpu_vcnfw_log; int amdgpu_sg_display = -1; /* auto */ +int amdgpu_user_partt_mode = AMDGPU_AUTO_COMPUTE_PARTITION_MODE; static void amdgpu_drv_delayed_reset_work_handler(struct work_struct *work); @@ -819,6 +823,13 @@ MODULE_PARM_DESC(no_queue_eviction_on_vm_fault, "No queue eviction on VM fault ( module_param_named(no_queue_eviction_on_vm_fault, amdgpu_no_queue_eviction_on_vm_fault, int, 0444); #endif +/** + * DOC: mtype_local (int) + */ +int amdgpu_mtype_local; +MODULE_PARM_DESC(mtype_local, "MTYPE for local memory (0 = MTYPE_RW (default), 1 = MTYPE_NC, 2 = MTYPE_CC)"); +module_param_named(mtype_local, amdgpu_mtype_local, int, 0444); + /** * DOC: pcie_p2p (bool) * Enable PCIe P2P (requires large-BAR). Default value: true (on) @@ -948,6 +959,28 @@ MODULE_PARM_DESC(smu_pptable_id, "specify pptable id to be used (-1 = auto(default) value, 0 = use pptable from vbios, > 0 = soft pptable id)"); module_param_named(smu_pptable_id, amdgpu_smu_pptable_id, int, 0444); +/** + * DOC: partition_mode (int) + * Used to override the default SPX mode. + */ +MODULE_PARM_DESC( + user_partt_mode, + "specify partition mode to be used (-2 = AMDGPU_AUTO_COMPUTE_PARTITION_MODE(default value) \ + 0 = AMDGPU_SPX_PARTITION_MODE, \ + 1 = AMDGPU_DPX_PARTITION_MODE, \ + 2 = AMDGPU_TPX_PARTITION_MODE, \ + 3 = AMDGPU_QPX_PARTITION_MODE, \ + 4 = AMDGPU_CPX_PARTITION_MODE)"); +module_param_named(user_partt_mode, amdgpu_user_partt_mode, uint, 0444); + + +/** + * DOC: enforce_isolation (bool) + * enforce process isolation between graphics and compute via using the same reserved vmid. + */ +module_param(enforce_isolation, bool, 0444); +MODULE_PARM_DESC(enforce_isolation, "enforce process isolation between graphics and compute . enforce_isolation = on"); + /* These devices are not supported by amdgpu. * They are supported by the mach64, r128, radeon drivers */ @@ -1615,6 +1648,7 @@ static const u16 amdgpu_unsupported_pciidlist[] = { 0x5874, 0x5940, 0x5941, + 0x5b70, 0x5b72, 0x5b73, 0x5b74, @@ -1660,7 +1694,7 @@ static const u16 amdgpu_unsupported_pciidlist[] = { }; static const struct pci_device_id pciidlist[] = { -#ifdef CONFIG_DRM_AMDGPU_SI +#ifdef CONFIG_DRM_AMDGPU_SI {0x1002, 0x6780, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TAHITI}, {0x1002, 0x6784, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TAHITI}, {0x1002, 0x6788, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TAHITI}, @@ -2017,6 +2051,11 @@ static const struct pci_device_id pciidlist[] = { .class_mask = 0xffffff, .driver_data = CHIP_IP_DISCOVERY }, + { PCI_DEVICE(0x1002, PCI_ANY_ID), + .class = PCI_CLASS_ACCELERATOR_PROCESSING << 8, + .class_mask = 0xffffff, + .driver_data = CHIP_IP_DISCOVERY }, + {0, 0, 0} }; @@ -2161,6 +2200,10 @@ static int amdgpu_pci_probe(struct pci_dev *pdev, goto err_pci; } + ret = amdgpu_xcp_dev_register(adev, ent); + if (ret) + goto err_pci; + /* * 1. don't init fbdev on hw without DCE * 2. don't init fbdev if there are no connectors @@ -2233,6 +2276,7 @@ amdgpu_pci_remove(struct pci_dev *pdev) struct drm_device *dev = pci_get_drvdata(pdev); struct amdgpu_device *adev = drm_to_adev(dev); + amdgpu_xcp_dev_unplug(adev); drm_dev_unplug(dev); if (adev->pm.rpm_mode != AMDGPU_RUNPM_NONE) { @@ -2818,6 +2862,33 @@ static const struct drm_driver amdgpu_kms_driver = { .patchlevel = KMS_DRIVER_PATCHLEVEL, }; +const struct drm_driver amdgpu_partition_driver = { + .driver_features = + DRIVER_GEM | DRIVER_RENDER | DRIVER_SYNCOBJ | + DRIVER_SYNCOBJ_TIMELINE, + .open = amdgpu_driver_open_kms, + .postclose = amdgpu_driver_postclose_kms, + .lastclose = amdgpu_driver_lastclose_kms, + .ioctls = amdgpu_ioctls_kms, + .num_ioctls = ARRAY_SIZE(amdgpu_ioctls_kms), + .dumb_create = amdgpu_mode_dumb_create, + .dumb_map_offset = amdgpu_mode_dumb_mmap, + .fops = &amdgpu_driver_kms_fops, + .release = &amdgpu_driver_release_kms, + + .prime_handle_to_fd = drm_gem_prime_handle_to_fd, + .prime_fd_to_handle = drm_gem_prime_fd_to_handle, + .gem_prime_import = amdgpu_gem_prime_import, + .gem_prime_mmap = drm_gem_prime_mmap, + + .name = DRIVER_NAME, + .desc = DRIVER_DESC, + .date = DRIVER_DATE, + .major = KMS_DRIVER_MAJOR, + .minor = KMS_DRIVER_MINOR, + .patchlevel = KMS_DRIVER_PATCHLEVEL, +}; + static struct pci_error_handlers amdgpu_pci_err_handler = { .error_detected = amdgpu_pci_error_detected, .mmio_enabled = amdgpu_pci_mmio_enabled, @@ -2885,9 +2956,11 @@ static void __exit amdgpu_exit(void) amdgpu_amdkfd_fini(); pci_unregister_driver(&amdgpu_kms_pci_driver); amdgpu_unregister_atpx_handler(); + amdgpu_acpi_release(); amdgpu_sync_fini(); amdgpu_fence_slab_fini(); mmu_notifier_synchronize(); + amdgpu_xcp_drv_release(); } module_init(amdgpu_init); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.h index 8178323e4beff1d900ef315df347bc810f00012b..5bc2cb661af7a0b9723c7af609ccf5d8d5b42c56 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.h @@ -42,6 +42,8 @@ #define DRIVER_DESC "AMD GPU" #define DRIVER_DATE "20150101" +extern const struct drm_driver amdgpu_partition_driver; + long amdgpu_drm_ioctl(struct file *filp, unsigned int cmd, unsigned long arg); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c index 27a782a9dc72817d25ae6ee4f4cf75d119ac6dbb..3aaeed2d356208d5601ccfdd84d2ca84ef79c9a9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c @@ -70,6 +70,7 @@ void amdgpu_encoder_set_active_device(struct drm_encoder *encoder) drm_for_each_connector_iter(connector, &iter) { if (connector->encoder == encoder) { struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector); + amdgpu_encoder->active_device = amdgpu_encoder->devices & amdgpu_connector->devices; DRM_DEBUG_KMS("setting active device to %08x from %08x %08x for encoder %d\n", amdgpu_encoder->active_device, amdgpu_encoder->devices, @@ -165,12 +166,12 @@ void amdgpu_panel_mode_fixup(struct drm_encoder *encoder, { struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder); struct drm_display_mode *native_mode = &amdgpu_encoder->native_mode; - unsigned hblank = native_mode->htotal - native_mode->hdisplay; - unsigned vblank = native_mode->vtotal - native_mode->vdisplay; - unsigned hover = native_mode->hsync_start - native_mode->hdisplay; - unsigned vover = native_mode->vsync_start - native_mode->vdisplay; - unsigned hsync_width = native_mode->hsync_end - native_mode->hsync_start; - unsigned vsync_width = native_mode->vsync_end - native_mode->vsync_start; + unsigned int hblank = native_mode->htotal - native_mode->hdisplay; + unsigned int vblank = native_mode->vtotal - native_mode->vdisplay; + unsigned int hover = native_mode->hsync_start - native_mode->hdisplay; + unsigned int vover = native_mode->vsync_start - native_mode->vdisplay; + unsigned int hsync_width = native_mode->hsync_end - native_mode->hsync_start; + unsigned int vsync_width = native_mode->vsync_end - native_mode->vsync_start; adjusted_mode->clock = native_mode->clock; adjusted_mode->flags = native_mode->flags; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c index f52d0ba91a770a4b437a1137aa88cf6a7113cf53..c694b41f6461b5e1495906a1d3bdd1af6a36b46f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c @@ -42,7 +42,6 @@ #include "amdgpu_reset.h" /* - * Fences * Fences mark an event in the GPUs pipeline and are used * for GPU/CPU synchronization. When the fence is written, * it is expected that all buffers associated with that fence @@ -140,7 +139,7 @@ static u32 amdgpu_fence_read(struct amdgpu_ring *ring) * Returns 0 on success, -ENOMEM on failure. */ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f, struct amdgpu_job *job, - unsigned flags) + unsigned int flags) { struct amdgpu_device *adev = ring->adev; struct dma_fence *fence; @@ -174,11 +173,11 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f, struct amd adev->fence_context + ring->idx, seq); /* Against remove in amdgpu_job_{free, free_cb} */ dma_fence_get(fence); - } - else + } else { dma_fence_init(fence, &amdgpu_fence_ops, &ring->fence_drv.lock, adev->fence_context + ring->idx, seq); + } } amdgpu_ring_emit_fence(ring, ring->fence_drv.gpu_addr, @@ -377,14 +376,11 @@ signed long amdgpu_fence_wait_polling(struct amdgpu_ring *ring, uint32_t wait_seq, signed long timeout) { - uint32_t seq; - - do { - seq = amdgpu_fence_read(ring); - udelay(5); - timeout -= 5; - } while ((int32_t)(wait_seq - seq) > 0 && timeout > 0); + while ((int32_t)(wait_seq - amdgpu_fence_read(ring)) > 0 && timeout > 0) { + udelay(2); + timeout -= 2; + } return timeout > 0 ? timeout : 0; } /** @@ -396,7 +392,7 @@ signed long amdgpu_fence_wait_polling(struct amdgpu_ring *ring, * Returns the number of emitted fences on the ring. Used by the * dynpm code to ring track activity. */ -unsigned amdgpu_fence_count_emitted(struct amdgpu_ring *ring) +unsigned int amdgpu_fence_count_emitted(struct amdgpu_ring *ring) { uint64_t emitted; @@ -475,7 +471,7 @@ void amdgpu_fence_update_start_timestamp(struct amdgpu_ring *ring, uint32_t seq, */ int amdgpu_fence_driver_start_ring(struct amdgpu_ring *ring, struct amdgpu_irq_src *irq_src, - unsigned irq_type) + unsigned int irq_type) { struct amdgpu_device *adev = ring->adev; uint64_t index; @@ -582,7 +578,8 @@ void amdgpu_fence_driver_hw_fini(struct amdgpu_device *adev) if (r) amdgpu_fence_driver_force_completion(ring); - if (ring->fence_drv.irq_src) + if (!drm_dev_is_unplugged(adev_to_drm(adev)) && + ring->fence_drv.irq_src) amdgpu_irq_put(adev, ring->fence_drv.irq_src, ring->fence_drv.irq_type); @@ -653,6 +650,7 @@ void amdgpu_fence_driver_hw_init(struct amdgpu_device *adev) for (i = 0; i < AMDGPU_MAX_RINGS; i++) { struct amdgpu_ring *ring = adev->rings[i]; + if (!ring || !ring->fence_drv.initialized) continue; @@ -693,6 +691,30 @@ void amdgpu_fence_driver_clear_job_fences(struct amdgpu_ring *ring) } } +/** + * amdgpu_fence_driver_set_error - set error code on fences + * @ring: the ring which contains the fences + * @error: the error code to set + * + * Set an error code to all the fences pending on the ring. + */ +void amdgpu_fence_driver_set_error(struct amdgpu_ring *ring, int error) +{ + struct amdgpu_fence_driver *drv = &ring->fence_drv; + unsigned long flags; + + spin_lock_irqsave(&drv->lock, flags); + for (unsigned int i = 0; i <= drv->num_fences_mask; ++i) { + struct dma_fence *fence; + + fence = rcu_dereference_protected(drv->fences[i], + lockdep_is_held(&drv->lock)); + if (fence && !dma_fence_is_signaled_locked(fence)) + dma_fence_set_error(fence, error); + } + spin_unlock_irqrestore(&drv->lock, flags); +} + /** * amdgpu_fence_driver_force_completion - force signal latest fence of ring * @@ -701,6 +723,7 @@ void amdgpu_fence_driver_clear_job_fences(struct amdgpu_ring *ring) */ void amdgpu_fence_driver_force_completion(struct amdgpu_ring *ring) { + amdgpu_fence_driver_set_error(ring, -ECANCELED); amdgpu_fence_write(ring, ring->fence_drv.sync_seq); amdgpu_fence_process(ring); } @@ -835,11 +858,12 @@ static const struct dma_fence_ops amdgpu_job_fence_ops = { #if defined(CONFIG_DEBUG_FS) static int amdgpu_debugfs_fence_info_show(struct seq_file *m, void *unused) { - struct amdgpu_device *adev = (struct amdgpu_device *)m->private; + struct amdgpu_device *adev = m->private; int i; for (i = 0; i < AMDGPU_MAX_RINGS; ++i) { struct amdgpu_ring *ring = adev->rings[i]; + if (!ring || !ring->fence_drv.initialized) continue; @@ -913,6 +937,7 @@ static void amdgpu_debugfs_reset_work(struct work_struct *work) reset_work); struct amdgpu_reset_context reset_context; + memset(&reset_context, 0, sizeof(reset_context)); reset_context.method = AMD_RESET_METHOD_NONE; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c index 01cb89ffbd566a5128dff3780c3f0017376571da..73b8cca35bab8780d1938a45d035d19648bdd081 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c @@ -35,6 +35,7 @@ #endif #include "amdgpu.h" #include +#include /* * GART @@ -102,6 +103,142 @@ void amdgpu_gart_dummy_page_fini(struct amdgpu_device *adev) adev->dummy_page_addr = 0; } +/** + * amdgpu_gart_table_ram_alloc - allocate system ram for gart page table + * + * @adev: amdgpu_device pointer + * + * Allocate system memory for GART page table for ASICs that don't have + * dedicated VRAM. + * Returns 0 for success, error for failure. + */ +int amdgpu_gart_table_ram_alloc(struct amdgpu_device *adev) +{ + unsigned int order = get_order(adev->gart.table_size); + gfp_t gfp_flags = GFP_KERNEL | __GFP_ZERO; + struct amdgpu_bo *bo = NULL; + struct sg_table *sg = NULL; + struct amdgpu_bo_param bp; + dma_addr_t dma_addr; + struct page *p; + int ret; + + if (adev->gart.bo != NULL) + return 0; + + p = alloc_pages(gfp_flags, order); + if (!p) + return -ENOMEM; + + /* If the hardware does not support UTCL2 snooping of the CPU caches + * then set_memory_wc() could be used as a workaround to mark the pages + * as write combine memory. + */ + dma_addr = dma_map_page(&adev->pdev->dev, p, 0, adev->gart.table_size, + DMA_BIDIRECTIONAL); + if (dma_mapping_error(&adev->pdev->dev, dma_addr)) { + dev_err(&adev->pdev->dev, "Failed to DMA MAP the GART BO page\n"); + __free_pages(p, order); + p = NULL; + return -EFAULT; + } + + dev_info(adev->dev, "%s dma_addr:%pad\n", __func__, &dma_addr); + /* Create SG table */ + sg = kmalloc(sizeof(*sg), GFP_KERNEL); + if (!sg) { + ret = -ENOMEM; + goto error; + } + ret = sg_alloc_table(sg, 1, GFP_KERNEL); + if (ret) + goto error; + + sg_dma_address(sg->sgl) = dma_addr; + sg->sgl->length = adev->gart.table_size; +#ifdef CONFIG_NEED_SG_DMA_LENGTH + sg->sgl->dma_length = adev->gart.table_size; +#endif + /* Create SG BO */ + memset(&bp, 0, sizeof(bp)); + bp.size = adev->gart.table_size; + bp.byte_align = PAGE_SIZE; + bp.domain = AMDGPU_GEM_DOMAIN_CPU; + bp.type = ttm_bo_type_sg; + bp.resv = NULL; + bp.bo_ptr_size = sizeof(struct amdgpu_bo); + bp.flags = 0; + ret = amdgpu_bo_create(adev, &bp, &bo); + if (ret) + goto error; + + bo->tbo.sg = sg; + bo->tbo.ttm->sg = sg; + bo->allowed_domains = AMDGPU_GEM_DOMAIN_GTT; + bo->preferred_domains = AMDGPU_GEM_DOMAIN_GTT; + + ret = amdgpu_bo_reserve(bo, true); + if (ret) { + dev_err(adev->dev, "(%d) failed to reserve bo for GART system bo\n", ret); + goto error; + } + + ret = amdgpu_bo_pin(bo, AMDGPU_GEM_DOMAIN_GTT); + WARN(ret, "Pinning the GART table failed"); + if (ret) + goto error_resv; + + adev->gart.bo = bo; + adev->gart.ptr = page_to_virt(p); + /* Make GART table accessible in VMID0 */ + ret = amdgpu_ttm_alloc_gart(&adev->gart.bo->tbo); + if (ret) + amdgpu_gart_table_ram_free(adev); + amdgpu_bo_unreserve(bo); + + return 0; + +error_resv: + amdgpu_bo_unreserve(bo); +error: + amdgpu_bo_unref(&bo); + if (sg) { + sg_free_table(sg); + kfree(sg); + } + __free_pages(p, order); + return ret; +} + +/** + * amdgpu_gart_table_ram_free - free gart page table system ram + * + * @adev: amdgpu_device pointer + * + * Free the system memory used for the GART page tableon ASICs that don't + * have dedicated VRAM. + */ +void amdgpu_gart_table_ram_free(struct amdgpu_device *adev) +{ + unsigned int order = get_order(adev->gart.table_size); + struct sg_table *sg = adev->gart.bo->tbo.sg; + struct page *p; + int ret; + + ret = amdgpu_bo_reserve(adev->gart.bo, false); + if (!ret) { + amdgpu_bo_unpin(adev->gart.bo); + amdgpu_bo_unreserve(adev->gart.bo); + } + amdgpu_bo_unref(&adev->gart.bo); + sg_free_table(sg); + kfree(sg); + p = virt_to_page(adev->gart.ptr); + __free_pages(p, order); + + adev->gart.ptr = NULL; +} + /** * amdgpu_gart_table_vram_alloc - allocate vram for gart page table * @@ -182,7 +319,7 @@ void amdgpu_gart_unbind(struct amdgpu_device *adev, uint64_t offset, } mb(); amdgpu_device_flush_hdp(adev, NULL); - for (i = 0; i < adev->num_vmhubs; i++) + for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) amdgpu_gmc_flush_gpu_tlb(adev, 0, i, 0); drm_dev_exit(idx); @@ -264,7 +401,7 @@ void amdgpu_gart_invalidate_tlb(struct amdgpu_device *adev) mb(); amdgpu_device_flush_hdp(adev, NULL); - for (i = 0; i < adev->num_vmhubs; i++) + for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) amdgpu_gmc_flush_gpu_tlb(adev, 0, i, 0); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h index 8fea3e04e4110696983ec6f4911424060300cea6..8283d682f543b9118294600dd009a57a87e31742 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h @@ -51,6 +51,8 @@ struct amdgpu_gart { uint64_t gart_pte_flags; }; +int amdgpu_gart_table_ram_alloc(struct amdgpu_device *adev); +void amdgpu_gart_table_ram_free(struct amdgpu_device *adev); int amdgpu_gart_table_vram_alloc(struct amdgpu_device *adev); void amdgpu_gart_table_vram_free(struct amdgpu_device *adev); int amdgpu_gart_table_vram_pin(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 863cb668e0005d5694b473ea7b72a906061a1b2b..74055cba3dc9aaf38ed74377c7ecf3ce5476749e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -98,7 +98,7 @@ int amdgpu_gem_object_create(struct amdgpu_device *adev, unsigned long size, int alignment, u32 initial_domain, u64 flags, enum ttm_bo_type type, struct dma_resv *resv, - struct drm_gem_object **obj) + struct drm_gem_object **obj, int8_t xcp_id_plus1) { struct amdgpu_bo *bo; struct amdgpu_bo_user *ubo; @@ -116,6 +116,7 @@ int amdgpu_gem_object_create(struct amdgpu_device *adev, unsigned long size, bp.flags = flags; bp.domain = initial_domain; bp.bo_ptr_size = sizeof(struct amdgpu_bo); + bp.xcp_id_plus1 = xcp_id_plus1; r = amdgpu_bo_create_user(adev, &bp, &ubo); if (r) @@ -336,7 +337,7 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data, retry: r = amdgpu_gem_object_create(adev, size, args->in.alignment, initial_domain, - flags, ttm_bo_type_device, resv, &gobj); + flags, ttm_bo_type_device, resv, &gobj, fpriv->xcp_id + 1); if (r && r != -ERESTARTSYS) { if (flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) { flags &= ~AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED; @@ -379,6 +380,7 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data, struct ttm_operation_ctx ctx = { true, false }; struct amdgpu_device *adev = drm_to_adev(dev); struct drm_amdgpu_gem_userptr *args = data; + struct amdgpu_fpriv *fpriv = filp->driver_priv; struct drm_gem_object *gobj; struct hmm_range *range; struct amdgpu_bo *bo; @@ -405,7 +407,7 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data, /* create a gem object to contain this object in */ r = amdgpu_gem_object_create(adev, args->size, 0, AMDGPU_GEM_DOMAIN_CPU, - 0, ttm_bo_type_device, NULL, &gobj); + 0, ttm_bo_type_device, NULL, &gobj, fpriv->xcp_id + 1); if (r) return r; @@ -908,6 +910,7 @@ int amdgpu_mode_dumb_create(struct drm_file *file_priv, struct drm_mode_create_dumb *args) { struct amdgpu_device *adev = drm_to_adev(dev); + struct amdgpu_fpriv *fpriv = file_priv->driver_priv; struct drm_gem_object *gobj; uint32_t handle; u64 flags = AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED | @@ -931,7 +934,7 @@ int amdgpu_mode_dumb_create(struct drm_file *file_priv, domain = amdgpu_bo_get_preferred_domain(adev, amdgpu_display_supported_domains(adev, flags)); r = amdgpu_gem_object_create(adev, args->size, 0, domain, flags, - ttm_bo_type_device, NULL, &gobj); + ttm_bo_type_device, NULL, &gobj, fpriv->xcp_id + 1); if (r) return -ENOMEM; @@ -948,7 +951,7 @@ int amdgpu_mode_dumb_create(struct drm_file *file_priv, #if defined(CONFIG_DEBUG_FS) static int amdgpu_debugfs_gem_info_show(struct seq_file *m, void *unused) { - struct amdgpu_device *adev = (struct amdgpu_device *)m->private; + struct amdgpu_device *adev = m->private; struct drm_device *dev = adev_to_drm(adev); struct drm_file *file; int r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h index 637bf51dbf06f0bf2f0701e4a54d4711f8598c77..f30264782ba27bfad89acbfbf52b503f81a41ee5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h @@ -43,8 +43,7 @@ int amdgpu_gem_object_create(struct amdgpu_device *adev, unsigned long size, int alignment, u32 initial_domain, u64 flags, enum ttm_bo_type type, struct dma_resv *resv, - struct drm_gem_object **obj); - + struct drm_gem_object **obj, int8_t xcp_id_plus1); int amdgpu_mode_dumb_create(struct drm_file *file_priv, struct drm_device *dev, struct drm_mode_create_dumb *args); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c index f3f541ba0acaa0fdf6761ad80b0e5fe10b27a1a6..a33d4bc34cee746cdee7e70103190fee0aa9f2cd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c @@ -28,6 +28,7 @@ #include "amdgpu_gfx.h" #include "amdgpu_rlc.h" #include "amdgpu_ras.h" +#include "amdgpu_xcp.h" /* delay 0.1 second to enable gfx off feature */ #define GFX_OFF_DELAY_ENABLE msecs_to_jiffies(100) @@ -63,10 +64,10 @@ void amdgpu_queue_mask_bit_to_mec_queue(struct amdgpu_device *adev, int bit, } bool amdgpu_gfx_is_mec_queue_enabled(struct amdgpu_device *adev, - int mec, int pipe, int queue) + int xcc_id, int mec, int pipe, int queue) { return test_bit(amdgpu_gfx_mec_queue_to_bit(adev, mec, pipe, queue), - adev->gfx.mec.queue_bitmap); + adev->gfx.mec_bitmap[xcc_id].queue_bitmap); } int amdgpu_gfx_me_queue_to_bit(struct amdgpu_device *adev, @@ -204,29 +205,38 @@ bool amdgpu_gfx_is_high_priority_compute_queue(struct amdgpu_device *adev, void amdgpu_gfx_compute_queue_acquire(struct amdgpu_device *adev) { - int i, queue, pipe; + int i, j, queue, pipe; bool multipipe_policy = amdgpu_gfx_is_compute_multipipe_capable(adev); int max_queues_per_mec = min(adev->gfx.mec.num_pipe_per_mec * adev->gfx.mec.num_queue_per_pipe, adev->gfx.num_compute_rings); + int num_xcc = adev->gfx.xcc_mask ? NUM_XCC(adev->gfx.xcc_mask) : 1; if (multipipe_policy) { - /* policy: make queues evenly cross all pipes on MEC1 only */ - for (i = 0; i < max_queues_per_mec; i++) { - pipe = i % adev->gfx.mec.num_pipe_per_mec; - queue = (i / adev->gfx.mec.num_pipe_per_mec) % - adev->gfx.mec.num_queue_per_pipe; - - set_bit(pipe * adev->gfx.mec.num_queue_per_pipe + queue, - adev->gfx.mec.queue_bitmap); + /* policy: make queues evenly cross all pipes on MEC1 only + * for multiple xcc, just use the original policy for simplicity */ + for (j = 0; j < num_xcc; j++) { + for (i = 0; i < max_queues_per_mec; i++) { + pipe = i % adev->gfx.mec.num_pipe_per_mec; + queue = (i / adev->gfx.mec.num_pipe_per_mec) % + adev->gfx.mec.num_queue_per_pipe; + + set_bit(pipe * adev->gfx.mec.num_queue_per_pipe + queue, + adev->gfx.mec_bitmap[j].queue_bitmap); + } } } else { /* policy: amdgpu owns all queues in the given pipe */ - for (i = 0; i < max_queues_per_mec; ++i) - set_bit(i, adev->gfx.mec.queue_bitmap); + for (j = 0; j < num_xcc; j++) { + for (i = 0; i < max_queues_per_mec; ++i) + set_bit(i, adev->gfx.mec_bitmap[j].queue_bitmap); + } } - dev_dbg(adev->dev, "mec queue bitmap weight=%d\n", bitmap_weight(adev->gfx.mec.queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES)); + for (j = 0; j < num_xcc; j++) { + dev_dbg(adev->dev, "mec queue bitmap weight=%d\n", + bitmap_weight(adev->gfx.mec_bitmap[j].queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES)); + } } void amdgpu_gfx_graphics_queue_acquire(struct amdgpu_device *adev) @@ -258,7 +268,7 @@ void amdgpu_gfx_graphics_queue_acquire(struct amdgpu_device *adev) } static int amdgpu_gfx_kiq_acquire(struct amdgpu_device *adev, - struct amdgpu_ring *ring) + struct amdgpu_ring *ring, int xcc_id) { int queue_bit; int mec, pipe, queue; @@ -268,7 +278,7 @@ static int amdgpu_gfx_kiq_acquire(struct amdgpu_device *adev, * adev->gfx.mec.num_queue_per_pipe; while (--queue_bit >= 0) { - if (test_bit(queue_bit, adev->gfx.mec.queue_bitmap)) + if (test_bit(queue_bit, adev->gfx.mec_bitmap[xcc_id].queue_bitmap)) continue; amdgpu_queue_mask_bit_to_mec_queue(adev, queue_bit, &mec, &pipe, &queue); @@ -294,9 +304,9 @@ static int amdgpu_gfx_kiq_acquire(struct amdgpu_device *adev, int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev, struct amdgpu_ring *ring, - struct amdgpu_irq_src *irq) + struct amdgpu_irq_src *irq, int xcc_id) { - struct amdgpu_kiq *kiq = &adev->gfx.kiq; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id]; int r = 0; spin_lock_init(&kiq->ring_lock); @@ -304,16 +314,20 @@ int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev, ring->adev = NULL; ring->ring_obj = NULL; ring->use_doorbell = true; - ring->doorbell_index = adev->doorbell_index.kiq; - ring->vm_hub = AMDGPU_GFXHUB_0; - - r = amdgpu_gfx_kiq_acquire(adev, ring); + ring->xcc_id = xcc_id; + ring->vm_hub = AMDGPU_GFXHUB(xcc_id); + ring->doorbell_index = + (adev->doorbell_index.kiq + + xcc_id * adev->doorbell_index.xcc_doorbell_range) + << 1; + + r = amdgpu_gfx_kiq_acquire(adev, ring, xcc_id); if (r) return r; ring->eop_gpu_addr = kiq->eop_gpu_addr; ring->no_scheduler = true; - sprintf(ring->name, "kiq_%d.%d.%d", ring->me, ring->pipe, ring->queue); + sprintf(ring->name, "kiq_%d.%d.%d.%d", xcc_id, ring->me, ring->pipe, ring->queue); r = amdgpu_ring_init(adev, ring, 1024, irq, AMDGPU_CP_KIQ_IRQ_DRIVER0, AMDGPU_RING_PRIO_DEFAULT, NULL); if (r) @@ -327,19 +341,19 @@ void amdgpu_gfx_kiq_free_ring(struct amdgpu_ring *ring) amdgpu_ring_fini(ring); } -void amdgpu_gfx_kiq_fini(struct amdgpu_device *adev) +void amdgpu_gfx_kiq_fini(struct amdgpu_device *adev, int xcc_id) { - struct amdgpu_kiq *kiq = &adev->gfx.kiq; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id]; amdgpu_bo_free_kernel(&kiq->eop_obj, &kiq->eop_gpu_addr, NULL); } int amdgpu_gfx_kiq_init(struct amdgpu_device *adev, - unsigned hpd_size) + unsigned hpd_size, int xcc_id) { int r; u32 *hpd; - struct amdgpu_kiq *kiq = &adev->gfx.kiq; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id]; r = amdgpu_bo_create_kernel(adev, hpd_size, PAGE_SIZE, AMDGPU_GEM_DOMAIN_GTT, &kiq->eop_obj, @@ -362,13 +376,18 @@ int amdgpu_gfx_kiq_init(struct amdgpu_device *adev, /* create MQD for each compute/gfx queue */ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev, - unsigned mqd_size) + unsigned mqd_size, int xcc_id) { - struct amdgpu_ring *ring = NULL; - int r, i; + int r, i, j; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id]; + struct amdgpu_ring *ring = &kiq->ring; + u32 domain = AMDGPU_GEM_DOMAIN_GTT; + + /* Only enable on gfx10 and 11 for now to avoid changing behavior on older chips */ + if (adev->ip_versions[GC_HWIP][0] >= IP_VERSION(10, 0, 0)) + domain |= AMDGPU_GEM_DOMAIN_VRAM; /* create MQD for KIQ */ - ring = &adev->gfx.kiq.ring; if (!adev->enable_mes_kiq && !ring->mqd_obj) { /* originaly the KIQ MQD is put in GTT domain, but for SRIOV VRAM domain is a must * otherwise hypervisor trigger SAVE_VF fail after driver unloaded which mean MQD @@ -387,8 +406,8 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev, } /* prepare MQD backup */ - adev->gfx.mec.mqd_backup[AMDGPU_MAX_COMPUTE_RINGS] = kmalloc(mqd_size, GFP_KERNEL); - if (!adev->gfx.mec.mqd_backup[AMDGPU_MAX_COMPUTE_RINGS]) + kiq->mqd_backup = kmalloc(mqd_size, GFP_KERNEL); + if (!kiq->mqd_backup) dev_warn(adev->dev, "no memory to create MQD backup for ring %s\n", ring->name); } @@ -398,13 +417,14 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev, ring = &adev->gfx.gfx_ring[i]; if (!ring->mqd_obj) { r = amdgpu_bo_create_kernel(adev, mqd_size, PAGE_SIZE, - AMDGPU_GEM_DOMAIN_GTT, &ring->mqd_obj, + domain, &ring->mqd_obj, &ring->mqd_gpu_addr, &ring->mqd_ptr); if (r) { dev_warn(adev->dev, "failed to create ring mqd bo (%d)", r); return r; } + ring->mqd_size = mqd_size; /* prepare MQD backup */ adev->gfx.me.mqd_backup[i] = kmalloc(mqd_size, GFP_KERNEL); if (!adev->gfx.me.mqd_backup[i]) @@ -415,19 +435,21 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev, /* create MQD for each KCQ */ for (i = 0; i < adev->gfx.num_compute_rings; i++) { - ring = &adev->gfx.compute_ring[i]; + j = i + xcc_id * adev->gfx.num_compute_rings; + ring = &adev->gfx.compute_ring[j]; if (!ring->mqd_obj) { r = amdgpu_bo_create_kernel(adev, mqd_size, PAGE_SIZE, - AMDGPU_GEM_DOMAIN_GTT, &ring->mqd_obj, + domain, &ring->mqd_obj, &ring->mqd_gpu_addr, &ring->mqd_ptr); if (r) { dev_warn(adev->dev, "failed to create ring mqd bo (%d)", r); return r; } + ring->mqd_size = mqd_size; /* prepare MQD backup */ - adev->gfx.mec.mqd_backup[i] = kmalloc(mqd_size, GFP_KERNEL); - if (!adev->gfx.mec.mqd_backup[i]) + adev->gfx.mec.mqd_backup[j] = kmalloc(mqd_size, GFP_KERNEL); + if (!adev->gfx.mec.mqd_backup[j]) dev_warn(adev->dev, "no memory to create MQD backup for ring %s\n", ring->name); } } @@ -435,10 +457,11 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev, return 0; } -void amdgpu_gfx_mqd_sw_fini(struct amdgpu_device *adev) +void amdgpu_gfx_mqd_sw_fini(struct amdgpu_device *adev, int xcc_id) { struct amdgpu_ring *ring = NULL; - int i; + int i, j; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id]; if (adev->asic_type >= CHIP_NAVI10 && amdgpu_async_gfx_ring) { for (i = 0; i < adev->gfx.num_gfx_rings; i++) { @@ -451,43 +474,81 @@ void amdgpu_gfx_mqd_sw_fini(struct amdgpu_device *adev) } for (i = 0; i < adev->gfx.num_compute_rings; i++) { - ring = &adev->gfx.compute_ring[i]; - kfree(adev->gfx.mec.mqd_backup[i]); + j = i + xcc_id * adev->gfx.num_compute_rings; + ring = &adev->gfx.compute_ring[j]; + kfree(adev->gfx.mec.mqd_backup[j]); amdgpu_bo_free_kernel(&ring->mqd_obj, &ring->mqd_gpu_addr, &ring->mqd_ptr); } - ring = &adev->gfx.kiq.ring; - kfree(adev->gfx.mec.mqd_backup[AMDGPU_MAX_COMPUTE_RINGS]); + ring = &kiq->ring; + kfree(kiq->mqd_backup); amdgpu_bo_free_kernel(&ring->mqd_obj, &ring->mqd_gpu_addr, &ring->mqd_ptr); } -int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev) +int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev, int xcc_id) { - struct amdgpu_kiq *kiq = &adev->gfx.kiq; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id]; struct amdgpu_ring *kiq_ring = &kiq->ring; int i, r = 0; + int j; if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) return -EINVAL; - spin_lock(&adev->gfx.kiq.ring_lock); + spin_lock(&kiq->ring_lock); if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size * adev->gfx.num_compute_rings)) { - spin_unlock(&adev->gfx.kiq.ring_lock); + spin_unlock(&kiq->ring_lock); return -ENOMEM; } - for (i = 0; i < adev->gfx.num_compute_rings; i++) - kiq->pmf->kiq_unmap_queues(kiq_ring, &adev->gfx.compute_ring[i], + for (i = 0; i < adev->gfx.num_compute_rings; i++) { + j = i + xcc_id * adev->gfx.num_compute_rings; + kiq->pmf->kiq_unmap_queues(kiq_ring, + &adev->gfx.compute_ring[j], RESET_QUEUES, 0, 0); + } - if (adev->gfx.kiq.ring.sched.ready && !adev->job_hang) + if (kiq_ring->sched.ready && !adev->job_hang) r = amdgpu_ring_test_helper(kiq_ring); - spin_unlock(&adev->gfx.kiq.ring_lock); + spin_unlock(&kiq->ring_lock); + + return r; +} + +int amdgpu_gfx_disable_kgq(struct amdgpu_device *adev, int xcc_id) +{ + struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id]; + struct amdgpu_ring *kiq_ring = &kiq->ring; + int i, r = 0; + int j; + + if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) + return -EINVAL; + + spin_lock(&kiq->ring_lock); + if (amdgpu_gfx_is_master_xcc(adev, xcc_id)) { + if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size * + adev->gfx.num_gfx_rings)) { + spin_unlock(&kiq->ring_lock); + return -ENOMEM; + } + + for (i = 0; i < adev->gfx.num_gfx_rings; i++) { + j = i + xcc_id * adev->gfx.num_gfx_rings; + kiq->pmf->kiq_unmap_queues(kiq_ring, + &adev->gfx.gfx_ring[j], + PREEMPT_QUEUES, 0, 0); + } + } + + if (adev->gfx.kiq[0].ring.sched.ready && !adev->job_hang) + r = amdgpu_ring_test_helper(kiq_ring); + spin_unlock(&kiq->ring_lock); return r; } @@ -505,18 +566,18 @@ int amdgpu_queue_mask_bit_to_set_resource_bit(struct amdgpu_device *adev, return set_resource_bit; } -int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev) +int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev, int xcc_id) { - struct amdgpu_kiq *kiq = &adev->gfx.kiq; - struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id]; + struct amdgpu_ring *kiq_ring = &kiq->ring; uint64_t queue_mask = 0; - int r, i; + int r, i, j; if (!kiq->pmf || !kiq->pmf->kiq_map_queues || !kiq->pmf->kiq_set_resources) return -EINVAL; for (i = 0; i < AMDGPU_MAX_COMPUTE_QUEUES; ++i) { - if (!test_bit(i, adev->gfx.mec.queue_bitmap)) + if (!test_bit(i, adev->gfx.mec_bitmap[xcc_id].queue_bitmap)) continue; /* This situation may be hit in the future if a new HW @@ -532,13 +593,15 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev) DRM_INFO("kiq ring mec %d pipe %d q %d\n", kiq_ring->me, kiq_ring->pipe, kiq_ring->queue); - spin_lock(&adev->gfx.kiq.ring_lock); + amdgpu_device_flush_hdp(adev, NULL); + + spin_lock(&kiq->ring_lock); r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size * adev->gfx.num_compute_rings + kiq->pmf->set_resources_size); if (r) { DRM_ERROR("Failed to lock KIQ (%d).\n", r); - spin_unlock(&adev->gfx.kiq.ring_lock); + spin_unlock(&kiq->ring_lock); return r; } @@ -546,11 +609,51 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev) queue_mask = ~0ULL; kiq->pmf->kiq_set_resources(kiq_ring, queue_mask); - for (i = 0; i < adev->gfx.num_compute_rings; i++) - kiq->pmf->kiq_map_queues(kiq_ring, &adev->gfx.compute_ring[i]); + for (i = 0; i < adev->gfx.num_compute_rings; i++) { + j = i + xcc_id * adev->gfx.num_compute_rings; + kiq->pmf->kiq_map_queues(kiq_ring, + &adev->gfx.compute_ring[j]); + } r = amdgpu_ring_test_helper(kiq_ring); - spin_unlock(&adev->gfx.kiq.ring_lock); + spin_unlock(&kiq->ring_lock); + if (r) + DRM_ERROR("KCQ enable failed\n"); + + return r; +} + +int amdgpu_gfx_enable_kgq(struct amdgpu_device *adev, int xcc_id) +{ + struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id]; + struct amdgpu_ring *kiq_ring = &kiq->ring; + int r, i, j; + + if (!kiq->pmf || !kiq->pmf->kiq_map_queues) + return -EINVAL; + + amdgpu_device_flush_hdp(adev, NULL); + + spin_lock(&kiq->ring_lock); + /* No need to map kcq on the slave */ + if (amdgpu_gfx_is_master_xcc(adev, xcc_id)) { + r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size * + adev->gfx.num_gfx_rings); + if (r) { + DRM_ERROR("Failed to lock KIQ (%d).\n", r); + spin_unlock(&kiq->ring_lock); + return r; + } + + for (i = 0; i < adev->gfx.num_gfx_rings; i++) { + j = i + xcc_id * adev->gfx.num_gfx_rings; + kiq->pmf->kiq_map_queues(kiq_ring, + &adev->gfx.gfx_ring[j]); + } + } + + r = amdgpu_ring_test_helper(kiq_ring); + spin_unlock(&kiq->ring_lock); if (r) DRM_ERROR("KCQ enable failed\n"); @@ -785,12 +888,31 @@ int amdgpu_gfx_cp_ecc_error_irq(struct amdgpu_device *adev, return 0; } +void amdgpu_gfx_ras_error_func(struct amdgpu_device *adev, + void *ras_error_status, + void (*func)(struct amdgpu_device *adev, void *ras_error_status, + int xcc_id)) +{ + int i; + int num_xcc = adev->gfx.xcc_mask ? NUM_XCC(adev->gfx.xcc_mask) : 1; + uint32_t xcc_mask = GENMASK(num_xcc - 1, 0); + struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; + + if (err_data) { + err_data->ue_count = 0; + err_data->ce_count = 0; + } + + for_each_inst(i, xcc_mask) + func(adev, ras_error_status, i); +} + uint32_t amdgpu_kiq_rreg(struct amdgpu_device *adev, uint32_t reg) { signed long r, cnt = 0; unsigned long flags; uint32_t seq, reg_val_offs = 0, value = 0; - struct amdgpu_kiq *kiq = &adev->gfx.kiq; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; struct amdgpu_ring *ring = &kiq->ring; if (amdgpu_device_skip_hw_access(adev)) @@ -858,7 +980,7 @@ void amdgpu_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v) signed long r, cnt = 0; unsigned long flags; uint32_t seq; - struct amdgpu_kiq *kiq = &adev->gfx.kiq; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; struct amdgpu_ring *ring = &kiq->ring; BUG_ON(!ring->funcs->emit_wreg); @@ -1062,3 +1184,125 @@ void amdgpu_gfx_cp_init_microcode(struct amdgpu_device *adev, adev->firmware.fw_size += ALIGN(fw_size, PAGE_SIZE); } } + +bool amdgpu_gfx_is_master_xcc(struct amdgpu_device *adev, int xcc_id) +{ + return !(xcc_id % (adev->gfx.num_xcc_per_xcp ? + adev->gfx.num_xcc_per_xcp : 1)); +} + +static ssize_t amdgpu_gfx_get_current_compute_partition(struct device *dev, + struct device_attribute *addr, + char *buf) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = drm_to_adev(ddev); + int mode; + + mode = amdgpu_xcp_query_partition_mode(adev->xcp_mgr, + AMDGPU_XCP_FL_NONE); + + return sysfs_emit(buf, "%s\n", amdgpu_gfx_compute_mode_desc(mode)); +} + +static ssize_t amdgpu_gfx_set_compute_partition(struct device *dev, + struct device_attribute *addr, + const char *buf, size_t count) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = drm_to_adev(ddev); + enum amdgpu_gfx_partition mode; + int ret = 0, num_xcc; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + if (num_xcc % 2 != 0) + return -EINVAL; + + if (!strncasecmp("SPX", buf, strlen("SPX"))) { + mode = AMDGPU_SPX_PARTITION_MODE; + } else if (!strncasecmp("DPX", buf, strlen("DPX"))) { + /* + * DPX mode needs AIDs to be in multiple of 2. + * Each AID connects 2 XCCs. + */ + if (num_xcc%4) + return -EINVAL; + mode = AMDGPU_DPX_PARTITION_MODE; + } else if (!strncasecmp("TPX", buf, strlen("TPX"))) { + if (num_xcc != 6) + return -EINVAL; + mode = AMDGPU_TPX_PARTITION_MODE; + } else if (!strncasecmp("QPX", buf, strlen("QPX"))) { + if (num_xcc != 8) + return -EINVAL; + mode = AMDGPU_QPX_PARTITION_MODE; + } else if (!strncasecmp("CPX", buf, strlen("CPX"))) { + mode = AMDGPU_CPX_PARTITION_MODE; + } else { + return -EINVAL; + } + + ret = amdgpu_xcp_switch_partition_mode(adev->xcp_mgr, mode); + + if (ret) + return ret; + + return count; +} + +static ssize_t amdgpu_gfx_get_available_compute_partition(struct device *dev, + struct device_attribute *addr, + char *buf) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = drm_to_adev(ddev); + char *supported_partition; + + /* TBD */ + switch (NUM_XCC(adev->gfx.xcc_mask)) { + case 8: + supported_partition = "SPX, DPX, QPX, CPX"; + break; + case 6: + supported_partition = "SPX, TPX, CPX"; + break; + case 4: + supported_partition = "SPX, DPX, CPX"; + break; + /* this seems only existing in emulation phase */ + case 2: + supported_partition = "SPX, CPX"; + break; + default: + supported_partition = "Not supported"; + break; + } + + return sysfs_emit(buf, "%s\n", supported_partition); +} + +static DEVICE_ATTR(current_compute_partition, S_IRUGO | S_IWUSR, + amdgpu_gfx_get_current_compute_partition, + amdgpu_gfx_set_compute_partition); + +static DEVICE_ATTR(available_compute_partition, S_IRUGO, + amdgpu_gfx_get_available_compute_partition, NULL); + +int amdgpu_gfx_sysfs_init(struct amdgpu_device *adev) +{ + int r; + + r = device_create_file(adev->dev, &dev_attr_current_compute_partition); + if (r) + return r; + + r = device_create_file(adev->dev, &dev_attr_available_compute_partition); + + return r; +} + +void amdgpu_gfx_sysfs_fini(struct amdgpu_device *adev) +{ + device_remove_file(adev->dev, &dev_attr_current_compute_partition); + device_remove_file(adev->dev, &dev_attr_available_compute_partition); +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h index bfabea76d166785c446a29117f20bc5599c1e569..ce0f7a8ad4b87355cfd8408fe3ad00475e10b1a7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h @@ -61,7 +61,42 @@ enum amdgpu_gfx_partition { AMDGPU_TPX_PARTITION_MODE = 2, AMDGPU_QPX_PARTITION_MODE = 3, AMDGPU_CPX_PARTITION_MODE = 4, - AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE, + AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE = -1, + /* Automatically choose the right mode */ + AMDGPU_AUTO_COMPUTE_PARTITION_MODE = -2, +}; + +#define NUM_XCC(x) hweight16(x) + +enum amdgpu_pkg_type { + AMDGPU_PKG_TYPE_APU = 2, + AMDGPU_PKG_TYPE_UNKNOWN, +}; + +enum amdgpu_gfx_ras_mem_id_type { + AMDGPU_GFX_CP_MEM = 0, + AMDGPU_GFX_GCEA_MEM, + AMDGPU_GFX_GC_CANE_MEM, + AMDGPU_GFX_GCUTCL2_MEM, + AMDGPU_GFX_GDS_MEM, + AMDGPU_GFX_LDS_MEM, + AMDGPU_GFX_RLC_MEM, + AMDGPU_GFX_SP_MEM, + AMDGPU_GFX_SPI_MEM, + AMDGPU_GFX_SQC_MEM, + AMDGPU_GFX_SQ_MEM, + AMDGPU_GFX_TA_MEM, + AMDGPU_GFX_TCC_MEM, + AMDGPU_GFX_TCA_MEM, + AMDGPU_GFX_TCI_MEM, + AMDGPU_GFX_TCP_MEM, + AMDGPU_GFX_TD_MEM, + AMDGPU_GFX_TCX_MEM, + AMDGPU_GFX_ATC_L2_MEM, + AMDGPU_GFX_UTCL2_MEM, + AMDGPU_GFX_VML2_MEM, + AMDGPU_GFX_VML2_WALKER_MEM, + AMDGPU_GFX_MEM_TYPE_NUM }; struct amdgpu_mec { @@ -75,8 +110,10 @@ struct amdgpu_mec { u32 num_mec; u32 num_pipe_per_mec; u32 num_queue_per_pipe; - void *mqd_backup[AMDGPU_MAX_COMPUTE_RINGS + 1]; + void *mqd_backup[AMDGPU_MAX_COMPUTE_RINGS * AMDGPU_MAX_GC_INSTANCES]; +}; +struct amdgpu_mec_bitmap { /* These are the resources for which amdgpu takes ownership */ DECLARE_BITMAP(queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES); }; @@ -120,6 +157,7 @@ struct amdgpu_kiq { struct amdgpu_ring ring; struct amdgpu_irq_src irq; const struct kiq_pm4_funcs *pmf; + void *mqd_backup; }; /* @@ -230,23 +268,37 @@ struct amdgpu_gfx_ras { struct amdgpu_iv_entry *entry); }; +struct amdgpu_gfx_shadow_info { + u32 shadow_size; + u32 shadow_alignment; + u32 csa_size; + u32 csa_alignment; +}; + struct amdgpu_gfx_funcs { /* get the gpu clock counter */ uint64_t (*get_gpu_clock_counter)(struct amdgpu_device *adev); void (*select_se_sh)(struct amdgpu_device *adev, u32 se_num, - u32 sh_num, u32 instance); - void (*read_wave_data)(struct amdgpu_device *adev, uint32_t simd, + u32 sh_num, u32 instance, int xcc_id); + void (*read_wave_data)(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields); - void (*read_wave_vgprs)(struct amdgpu_device *adev, uint32_t simd, + void (*read_wave_vgprs)(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t thread, uint32_t start, uint32_t size, uint32_t *dst); - void (*read_wave_sgprs)(struct amdgpu_device *adev, uint32_t simd, + void (*read_wave_sgprs)(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t start, uint32_t size, uint32_t *dst); void (*select_me_pipe_q)(struct amdgpu_device *adev, u32 me, u32 pipe, - u32 queue, u32 vmid); + u32 queue, u32 vmid, u32 xcc_id); void (*init_spm_golden)(struct amdgpu_device *adev); void (*update_perfmon_mgcg)(struct amdgpu_device *adev, bool enable); + int (*get_gfx_shadow_info)(struct amdgpu_device *adev, + struct amdgpu_gfx_shadow_info *shadow_info); + enum amdgpu_gfx_partition + (*query_partition_mode)(struct amdgpu_device *adev); + int (*switch_partition_mode)(struct amdgpu_device *adev, + int num_xccs_per_xcp); + int (*ih_node_to_logical_xcc)(struct amdgpu_device *adev, int ih_node); }; struct sq_work { @@ -296,7 +348,8 @@ struct amdgpu_gfx { struct amdgpu_ce ce; struct amdgpu_me me; struct amdgpu_mec mec; - struct amdgpu_kiq kiq; + struct amdgpu_mec_bitmap mec_bitmap[AMDGPU_MAX_GC_INSTANCES]; + struct amdgpu_kiq kiq[AMDGPU_MAX_GC_INSTANCES]; struct amdgpu_imu imu; bool rs64_enable; /* firmware format */ const struct firmware *me_fw; /* ME firmware */ @@ -376,15 +429,31 @@ struct amdgpu_gfx { struct amdgpu_ring sw_gfx_ring[AMDGPU_MAX_SW_GFX_RINGS]; struct amdgpu_ring_mux muxer; - enum amdgpu_gfx_partition partition_mode; - uint32_t num_xcd; + bool cp_gfx_shadow; /* for gfx11 */ + + uint16_t xcc_mask; uint32_t num_xcc_per_xcp; + struct mutex partition_mutex; }; +struct amdgpu_gfx_ras_reg_entry { + struct amdgpu_ras_err_status_reg_entry reg_entry; + enum amdgpu_gfx_ras_mem_id_type mem_id_type; + uint32_t se_num; +}; + +struct amdgpu_gfx_ras_mem_id_entry { + const struct amdgpu_ras_memory_id_entry *mem_id_ent; + uint32_t size; +}; + +#define AMDGPU_GFX_MEMID_ENT(x) {(x), ARRAY_SIZE(x)}, + #define amdgpu_gfx_get_gpu_clock_counter(adev) (adev)->gfx.funcs->get_gpu_clock_counter((adev)) -#define amdgpu_gfx_select_se_sh(adev, se, sh, instance) (adev)->gfx.funcs->select_se_sh((adev), (se), (sh), (instance)) -#define amdgpu_gfx_select_me_pipe_q(adev, me, pipe, q, vmid) (adev)->gfx.funcs->select_me_pipe_q((adev), (me), (pipe), (q), (vmid)) +#define amdgpu_gfx_select_se_sh(adev, se, sh, instance, xcc_id) ((adev)->gfx.funcs->select_se_sh((adev), (se), (sh), (instance), (xcc_id))) +#define amdgpu_gfx_select_me_pipe_q(adev, me, pipe, q, vmid, xcc_id) ((adev)->gfx.funcs->select_me_pipe_q((adev), (me), (pipe), (q), (vmid), (xcc_id))) #define amdgpu_gfx_init_spm_golden(adev) (adev)->gfx.funcs->init_spm_golden((adev)) +#define amdgpu_gfx_get_gfx_shadow_info(adev, si) ((adev)->gfx.funcs->get_gfx_shadow_info((adev), (si))) /** * amdgpu_gfx_create_bitmask - create a bitmask @@ -404,19 +473,21 @@ void amdgpu_gfx_parse_disable_cu(unsigned *mask, unsigned max_se, int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev, struct amdgpu_ring *ring, - struct amdgpu_irq_src *irq); + struct amdgpu_irq_src *irq, int xcc_id); void amdgpu_gfx_kiq_free_ring(struct amdgpu_ring *ring); -void amdgpu_gfx_kiq_fini(struct amdgpu_device *adev); +void amdgpu_gfx_kiq_fini(struct amdgpu_device *adev, int xcc_id); int amdgpu_gfx_kiq_init(struct amdgpu_device *adev, - unsigned hpd_size); + unsigned hpd_size, int xcc_id); int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev, - unsigned mqd_size); -void amdgpu_gfx_mqd_sw_fini(struct amdgpu_device *adev); -int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev); -int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev); + unsigned mqd_size, int xcc_id); +void amdgpu_gfx_mqd_sw_fini(struct amdgpu_device *adev, int xcc_id); +int amdgpu_gfx_disable_kcq(struct amdgpu_device *adev, int xcc_id); +int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev, int xcc_id); +int amdgpu_gfx_disable_kgq(struct amdgpu_device *adev, int xcc_id); +int amdgpu_gfx_enable_kgq(struct amdgpu_device *adev, int xcc_id); void amdgpu_gfx_compute_queue_acquire(struct amdgpu_device *adev); void amdgpu_gfx_graphics_queue_acquire(struct amdgpu_device *adev); @@ -425,8 +496,8 @@ int amdgpu_gfx_mec_queue_to_bit(struct amdgpu_device *adev, int mec, int pipe, int queue); void amdgpu_queue_mask_bit_to_mec_queue(struct amdgpu_device *adev, int bit, int *mec, int *pipe, int *queue); -bool amdgpu_gfx_is_mec_queue_enabled(struct amdgpu_device *adev, int mec, - int pipe, int queue); +bool amdgpu_gfx_is_mec_queue_enabled(struct amdgpu_device *adev, int xcc_id, + int mec, int pipe, int queue); bool amdgpu_gfx_is_high_priority_compute_queue(struct amdgpu_device *adev, struct amdgpu_ring *ring); bool amdgpu_gfx_is_high_priority_graphics_queue(struct amdgpu_device *adev, @@ -458,4 +529,33 @@ void amdgpu_gfx_cp_init_microcode(struct amdgpu_device *adev, uint32_t ucode_id) int amdgpu_gfx_ras_sw_init(struct amdgpu_device *adev); int amdgpu_gfx_poison_consumption_handler(struct amdgpu_device *adev, struct amdgpu_iv_entry *entry); + +bool amdgpu_gfx_is_master_xcc(struct amdgpu_device *adev, int xcc_id); +int amdgpu_gfx_sysfs_init(struct amdgpu_device *adev); +void amdgpu_gfx_sysfs_fini(struct amdgpu_device *adev); +void amdgpu_gfx_ras_error_func(struct amdgpu_device *adev, + void *ras_error_status, + void (*func)(struct amdgpu_device *adev, void *ras_error_status, + int xcc_id)); + +static inline const char *amdgpu_gfx_compute_mode_desc(int mode) +{ + switch (mode) { + case AMDGPU_SPX_PARTITION_MODE: + return "SPX"; + case AMDGPU_DPX_PARTITION_MODE: + return "DPX"; + case AMDGPU_TPX_PARTITION_MODE: + return "TPX"; + case AMDGPU_QPX_PARTITION_MODE: + return "QPX"; + case AMDGPU_CPX_PARTITION_MODE: + return "CPX"; + default: + return "UNKNOWN"; + } + + return "UNKNOWN"; +} + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index 4e2531758866c6e370622152661efcddf218a31b..d78bd97325434f7a6896a41a5436090dacc8cf46 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -534,22 +534,21 @@ void amdgpu_gmc_ras_fini(struct amdgpu_device *adev) * subject to change when ring number changes * Engine 17: Gart flushes */ -#define GFXHUB_FREE_VM_INV_ENGS_BITMAP 0x1FFF3 -#define MMHUB_FREE_VM_INV_ENGS_BITMAP 0x1FFF3 +#define AMDGPU_VMHUB_INV_ENG_BITMAP 0x1FFF3 int amdgpu_gmc_allocate_vm_inv_eng(struct amdgpu_device *adev) { struct amdgpu_ring *ring; - unsigned vm_inv_engs[AMDGPU_MAX_VMHUBS] = - {GFXHUB_FREE_VM_INV_ENGS_BITMAP, MMHUB_FREE_VM_INV_ENGS_BITMAP, - GFXHUB_FREE_VM_INV_ENGS_BITMAP}; + unsigned vm_inv_engs[AMDGPU_MAX_VMHUBS] = {0}; unsigned i; unsigned vmhub, inv_eng; - if (adev->enable_mes) { + /* init the vm inv eng for all vmhubs */ + for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) { + vm_inv_engs[i] = AMDGPU_VMHUB_INV_ENG_BITMAP; /* reserve engine 5 for firmware */ - for (vmhub = 0; vmhub < AMDGPU_MAX_VMHUBS; vmhub++) - vm_inv_engs[vmhub] &= ~(1 << 5); + if (adev->enable_mes) + vm_inv_engs[i] &= ~(1 << 5); } for (i = 0; i < adev->num_rings; ++i) { @@ -593,6 +592,8 @@ void amdgpu_gmc_tmz_set(struct amdgpu_device *adev) case IP_VERSION(9, 3, 0): /* GC 10.3.7 */ case IP_VERSION(10, 3, 7): + /* GC 11.0.1 */ + case IP_VERSION(11, 0, 1): if (amdgpu_tmz == 0) { adev->gmc.tmz_enabled = false; dev_info(adev->dev, @@ -616,7 +617,6 @@ void amdgpu_gmc_tmz_set(struct amdgpu_device *adev) case IP_VERSION(10, 3, 1): /* YELLOW_CARP*/ case IP_VERSION(10, 3, 3): - case IP_VERSION(11, 0, 1): case IP_VERSION(11, 0, 4): /* Don't enable it by default yet. */ @@ -670,7 +670,7 @@ void amdgpu_gmc_set_vm_fault_masks(struct amdgpu_device *adev, int hub_type, for (i = 0; i < 16; i++) { reg = hub->vm_context0_cntl + hub->ctx_distance * i; - tmp = (hub_type == AMDGPU_GFXHUB_0) ? + tmp = (hub_type == AMDGPU_GFXHUB(0)) ? RREG32_SOC15_IP(GC, reg) : RREG32_SOC15_IP(MMHUB, reg); @@ -679,7 +679,7 @@ void amdgpu_gmc_set_vm_fault_masks(struct amdgpu_device *adev, int hub_type, else tmp &= ~hub->vm_cntx_cntl_vm_fault; - (hub_type == AMDGPU_GFXHUB_0) ? + (hub_type == AMDGPU_GFXHUB(0)) ? WREG32_SOC15_IP(GC, reg, tmp) : WREG32_SOC15_IP(MMHUB, reg, tmp); } @@ -892,3 +892,47 @@ int amdgpu_gmc_vram_checking(struct amdgpu_device *adev) return 0; } + +static ssize_t current_memory_partition_show( + struct device *dev, struct device_attribute *addr, char *buf) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = drm_to_adev(ddev); + enum amdgpu_memory_partition mode; + + mode = adev->gmc.gmc_funcs->query_mem_partition_mode(adev); + switch (mode) { + case AMDGPU_NPS1_PARTITION_MODE: + return sysfs_emit(buf, "NPS1\n"); + case AMDGPU_NPS2_PARTITION_MODE: + return sysfs_emit(buf, "NPS2\n"); + case AMDGPU_NPS3_PARTITION_MODE: + return sysfs_emit(buf, "NPS3\n"); + case AMDGPU_NPS4_PARTITION_MODE: + return sysfs_emit(buf, "NPS4\n"); + case AMDGPU_NPS6_PARTITION_MODE: + return sysfs_emit(buf, "NPS6\n"); + case AMDGPU_NPS8_PARTITION_MODE: + return sysfs_emit(buf, "NPS8\n"); + default: + return sysfs_emit(buf, "UNKNOWN\n"); + } + + return sysfs_emit(buf, "UNKNOWN\n"); +} + +static DEVICE_ATTR_RO(current_memory_partition); + +int amdgpu_gmc_sysfs_init(struct amdgpu_device *adev) +{ + if (!adev->gmc.gmc_funcs->query_mem_partition_mode) + return 0; + + return device_create_file(adev->dev, + &dev_attr_current_memory_partition); +} + +void amdgpu_gmc_sysfs_fini(struct amdgpu_device *adev) +{ + device_remove_file(adev->dev, &dev_attr_current_memory_partition); +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h index 6d105d7fb98bb4e7d48749773c3d864724777997..56d73fade56850b5c0d9b45b450ef8bdf466332c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h @@ -63,6 +63,16 @@ struct firmware; +enum amdgpu_memory_partition { + UNKNOWN_MEMORY_PARTITION_MODE = 0, + AMDGPU_NPS1_PARTITION_MODE = 1, + AMDGPU_NPS2_PARTITION_MODE = 2, + AMDGPU_NPS3_PARTITION_MODE = 3, + AMDGPU_NPS4_PARTITION_MODE = 4, + AMDGPU_NPS6_PARTITION_MODE = 6, + AMDGPU_NPS8_PARTITION_MODE = 8, +}; + /* * GMC page fault information */ @@ -119,7 +129,8 @@ struct amdgpu_gmc_funcs { uint32_t vmhub, uint32_t flush_type); /* flush the vm tlb via pasid */ int (*flush_gpu_tlb_pasid)(struct amdgpu_device *adev, uint16_t pasid, - uint32_t flush_type, bool all_hub); + uint32_t flush_type, bool all_hub, + uint32_t inst); /* flush the vm tlb via ring */ uint64_t (*emit_flush_gpu_tlb)(struct amdgpu_ring *ring, unsigned vmid, uint64_t pd_addr); @@ -137,8 +148,15 @@ struct amdgpu_gmc_funcs { void (*get_vm_pte)(struct amdgpu_device *adev, struct amdgpu_bo_va_mapping *mapping, uint64_t *flags); + /* override per-page pte flags */ + void (*override_vm_pte_flags)(struct amdgpu_device *dev, + struct amdgpu_vm *vm, + uint64_t addr, uint64_t *flags); /* get the amount of memory used by the vbios for pre-OS console */ unsigned int (*get_vbios_fb_size)(struct amdgpu_device *adev); + + enum amdgpu_memory_partition (*query_mem_partition_mode)( + struct amdgpu_device *adev); }; struct amdgpu_xgmi_ras { @@ -164,6 +182,21 @@ struct amdgpu_xgmi { struct amdgpu_xgmi_ras *ras; }; +struct amdgpu_mem_partition_info { + union { + struct { + uint32_t fpfn; + uint32_t lpfn; + } range; + struct { + int node; + } numa; + }; + uint64_t size; +}; + +#define INVALID_PFN -1 + struct amdgpu_gmc { /* FB's physical address in MMIO space (for CPU to * map FB). This is different compared to the agp/ @@ -250,7 +283,10 @@ struct amdgpu_gmc { uint64_t last_fault:AMDGPU_GMC_FAULT_RING_ORDER; bool tmz_enabled; + bool is_app_apu; + struct amdgpu_mem_partition_info *mem_partitions; + uint8_t num_mem_partitions; const struct amdgpu_gmc_funcs *gmc_funcs; struct amdgpu_xgmi xgmi; @@ -265,6 +301,8 @@ struct amdgpu_gmc { /* MALL size */ u64 mall_size; + uint32_t m_half_use; + /* number of UMC instances */ int num_umc; /* mode2 save restore */ @@ -296,14 +334,17 @@ struct amdgpu_gmc { }; #define amdgpu_gmc_flush_gpu_tlb(adev, vmid, vmhub, type) ((adev)->gmc.gmc_funcs->flush_gpu_tlb((adev), (vmid), (vmhub), (type))) -#define amdgpu_gmc_flush_gpu_tlb_pasid(adev, pasid, type, allhub) \ +#define amdgpu_gmc_flush_gpu_tlb_pasid(adev, pasid, type, allhub, inst) \ ((adev)->gmc.gmc_funcs->flush_gpu_tlb_pasid \ - ((adev), (pasid), (type), (allhub))) + ((adev), (pasid), (type), (allhub), (inst))) #define amdgpu_gmc_emit_flush_gpu_tlb(r, vmid, addr) (r)->adev->gmc.gmc_funcs->emit_flush_gpu_tlb((r), (vmid), (addr)) #define amdgpu_gmc_emit_pasid_mapping(r, vmid, pasid) (r)->adev->gmc.gmc_funcs->emit_pasid_mapping((r), (vmid), (pasid)) #define amdgpu_gmc_map_mtype(adev, flags) (adev)->gmc.gmc_funcs->map_mtype((adev),(flags)) #define amdgpu_gmc_get_vm_pde(adev, level, dst, flags) (adev)->gmc.gmc_funcs->get_vm_pde((adev), (level), (dst), (flags)) #define amdgpu_gmc_get_vm_pte(adev, mapping, flags) (adev)->gmc.gmc_funcs->get_vm_pte((adev), (mapping), (flags)) +#define amdgpu_gmc_override_vm_pte_flags(adev, vm, addr, pte_flags) \ + (adev)->gmc.gmc_funcs->override_vm_pte_flags \ + ((adev), (vm), (addr), (pte_flags)) #define amdgpu_gmc_get_vbios_fb_size(adev) (adev)->gmc.gmc_funcs->get_vbios_fb_size((adev)) /** @@ -373,4 +414,7 @@ uint64_t amdgpu_gmc_vram_mc2pa(struct amdgpu_device *adev, uint64_t mc_addr); uint64_t amdgpu_gmc_vram_pa(struct amdgpu_device *adev, struct amdgpu_bo *bo); uint64_t amdgpu_gmc_vram_cpu_pa(struct amdgpu_device *adev, struct amdgpu_bo *bo); int amdgpu_gmc_vram_checking(struct amdgpu_device *adev); +int amdgpu_gmc_sysfs_init(struct amdgpu_device *adev); +void amdgpu_gmc_sysfs_fini(struct amdgpu_device *adev); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c index 2dadcfe43d03d253f8a5626770d869e731d2344e..081267161d40185d10d6292a2ad4f474ab554b95 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c @@ -190,8 +190,8 @@ int amdgpu_hmm_range_get_pages(struct mmu_interval_notifier *notifier, pr_debug("hmm range: start = 0x%lx, end = 0x%lx", hmm_range->start, hmm_range->end); - /* Assuming 512MB takes maxmium 1 second to fault page address */ - timeout = max((hmm_range->end - hmm_range->start) >> 29, 1UL); + /* Assuming 128MB takes maximum 1 second to fault page address */ + timeout = max((hmm_range->end - hmm_range->start) >> 27, 1UL); timeout *= HMM_RANGE_DEFAULT_TIMEOUT; timeout = jiffies + msecs_to_jiffies(timeout); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c index 4ff348e10e4d4bf430d641bdfcd7dce994988300..ebeddc9a37e9ba5070df80f09fdc6311095e6e6b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c @@ -136,7 +136,9 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, uint64_t fence_ctx; uint32_t status = 0, alloc_size; unsigned fence_flags = 0; - bool secure; + bool secure, init_shadow; + u64 shadow_va, csa_va, gds_va; + int vmid = AMDGPU_JOB_GET_VMID(job); unsigned i; int r = 0; @@ -150,9 +152,17 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, vm = job->vm; fence_ctx = job->base.s_fence ? job->base.s_fence->scheduled.context : 0; + shadow_va = job->shadow_va; + csa_va = job->csa_va; + gds_va = job->gds_va; + init_shadow = job->init_shadow; } else { vm = NULL; fence_ctx = 0; + shadow_va = 0; + csa_va = 0; + gds_va = 0; + init_shadow = false; } if (!ring->sched.ready && !ring->is_mes_queue) { @@ -212,7 +222,12 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, } amdgpu_ring_ib_begin(ring); - if (job && ring->funcs->init_cond_exec) + + if (ring->funcs->emit_gfx_shadow) + amdgpu_ring_emit_gfx_shadow(ring, shadow_va, csa_va, gds_va, + init_shadow, vmid); + + if (ring->funcs->init_cond_exec) patch_offset = amdgpu_ring_init_cond_exec(ring); amdgpu_device_flush_hdp(adev, ring); @@ -263,6 +278,18 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, fence_flags | AMDGPU_FENCE_FLAG_64BIT); } + if (ring->funcs->emit_gfx_shadow) { + amdgpu_ring_emit_gfx_shadow(ring, 0, 0, 0, false, 0); + + if (ring->funcs->init_cond_exec) { + unsigned ce_offset = ~0; + + ce_offset = amdgpu_ring_init_cond_exec(ring); + if (ce_offset != ~0 && ring->funcs->patch_cond_exec) + amdgpu_ring_patch_cond_exec(ring, ce_offset); + } + } + r = amdgpu_fence_emit(ring, f, job, fence_flags); if (r) { dev_err(adev->dev, "failed to emit fence (%d)\n", r); @@ -436,7 +463,7 @@ int amdgpu_ib_ring_tests(struct amdgpu_device *adev) static int amdgpu_debugfs_sa_info_show(struct seq_file *m, void *unused) { - struct amdgpu_device *adev = (struct amdgpu_device *)m->private; + struct amdgpu_device *adev = m->private; seq_printf(m, "--------------------- DELAYED --------------------- \n"); amdgpu_sa_bo_dump_debug_info(&adev->ib_pools[AMDGPU_IB_POOL_DELAYED], diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c index c991ca0b7a1c889d6ba705badb54c30126b15e5e..ff1ea99292fbf0185274b08c5b1b2c33689e0a44 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c @@ -409,7 +409,7 @@ int amdgpu_vmid_grab(struct amdgpu_vm *vm, struct amdgpu_ring *ring, if (r || !idle) goto error; - if (vm->reserved_vmid[vmhub]) { + if (vm->reserved_vmid[vmhub] || (enforce_isolation && (vmhub == AMDGPU_GFXHUB(0)))) { r = amdgpu_vmid_grab_reserved(vm, ring, job, &id, fence); if (r || !id) goto error; @@ -460,14 +460,11 @@ int amdgpu_vmid_grab(struct amdgpu_vm *vm, struct amdgpu_ring *ring, } int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev, - struct amdgpu_vm *vm, unsigned vmhub) { struct amdgpu_vmid_mgr *id_mgr = &adev->vm_manager.id_mgr[vmhub]; mutex_lock(&id_mgr->lock); - if (vm->reserved_vmid[vmhub]) - goto unlock; ++id_mgr->reserved_use_count; if (!id_mgr->reserved) { @@ -479,27 +476,23 @@ int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev, list_del_init(&id->list); id_mgr->reserved = id; } - vm->reserved_vmid[vmhub] = true; -unlock: mutex_unlock(&id_mgr->lock); return 0; } void amdgpu_vmid_free_reserved(struct amdgpu_device *adev, - struct amdgpu_vm *vm, unsigned vmhub) { struct amdgpu_vmid_mgr *id_mgr = &adev->vm_manager.id_mgr[vmhub]; mutex_lock(&id_mgr->lock); - if (vm->reserved_vmid[vmhub] && - !--id_mgr->reserved_use_count) { + if (!--id_mgr->reserved_use_count) { /* give the reserved ID back to normal round robin */ list_add(&id_mgr->reserved->list, &id_mgr->ids_lru); id_mgr->reserved = NULL; } - vm->reserved_vmid[vmhub] = false; + mutex_unlock(&id_mgr->lock); } @@ -578,6 +571,10 @@ void amdgpu_vmid_mgr_init(struct amdgpu_device *adev) list_add_tail(&id_mgr->ids[j].list, &id_mgr->ids_lru); } } + /* alloc a default reserved vmid to enforce isolation */ + if (enforce_isolation) + amdgpu_vmid_alloc_reserved(adev, AMDGPU_GFXHUB(0)); + } /** diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h index d1cc09b45da4a1783edfef7d9f182e247f4f3031..fa8c42c83d5d26bc6d90455b5e7b51b627c5b5d7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h @@ -79,11 +79,9 @@ void amdgpu_pasid_free_delayed(struct dma_resv *resv, bool amdgpu_vmid_had_gpu_reset(struct amdgpu_device *adev, struct amdgpu_vmid *id); int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev, - struct amdgpu_vm *vm, - unsigned vmhub); + unsigned vmhub); void amdgpu_vmid_free_reserved(struct amdgpu_device *adev, - struct amdgpu_vm *vm, - unsigned vmhub); + unsigned vmhub); int amdgpu_vmid_grab(struct amdgpu_vm *vm, struct amdgpu_ring *ring, struct amdgpu_job *job, struct dma_fence **fence); void amdgpu_vmid_reset(struct amdgpu_device *adev, unsigned vmhub, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c index 1d5af50331e41030625a6d983899b88f32240387..fceb3b384955ac023765d0616126b9c0e73a5f34 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c @@ -270,7 +270,7 @@ void amdgpu_ih_decode_iv_helper(struct amdgpu_device *adev, entry->timestamp = dw[1] | ((u64)(dw[2] & 0xffff) << 32); entry->timestamp_src = dw[2] >> 31; entry->pasid = dw[3] & 0xffff; - entry->pasid_src = dw[3] >> 31; + entry->node_id = (dw[3] >> 16) & 0xff; entry->src_data[0] = dw[4]; entry->src_data[1] = dw[5]; entry->src_data[2] = dw[6]; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c index fafebec5b7b66da8911ffb2f112817a1f1f626ae..5273decc5753b45fff9f12f21bc6e567b13d56f3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c @@ -99,6 +99,21 @@ const char *soc15_ih_clientid_name[] = { "MP1" }; +const int node_id_to_phys_map[NODEID_MAX] = { + [AID0_NODEID] = 0, + [XCD0_NODEID] = 0, + [XCD1_NODEID] = 1, + [AID1_NODEID] = 1, + [XCD2_NODEID] = 2, + [XCD3_NODEID] = 3, + [AID2_NODEID] = 2, + [XCD4_NODEID] = 4, + [XCD5_NODEID] = 5, + [AID3_NODEID] = 3, + [XCD6_NODEID] = 6, + [XCD7_NODEID] = 7, +}; + /** * amdgpu_irq_disable_all - disable *all* interrupts * @@ -109,7 +124,7 @@ const char *soc15_ih_clientid_name[] = { void amdgpu_irq_disable_all(struct amdgpu_device *adev) { unsigned long irqflags; - unsigned i, j, k; + unsigned int i, j, k; int r; spin_lock_irqsave(&adev->irq.lock, irqflags); @@ -124,7 +139,6 @@ void amdgpu_irq_disable_all(struct amdgpu_device *adev) continue; for (k = 0; k < src->num_types; ++k) { - atomic_set(&src->enabled_types[k], 0); r = src->funcs->set(adev, src, k, AMDGPU_IRQ_STATE_DISABLE); if (r) @@ -268,11 +282,11 @@ int amdgpu_irq_init(struct amdgpu_device *adev) int nvec = pci_msix_vec_count(adev->pdev); unsigned int flags; - if (nvec <= 0) { + if (nvec <= 0) flags = PCI_IRQ_MSI; - } else { + else flags = PCI_IRQ_MSI | PCI_IRQ_MSIX; - } + /* we only need one vector */ nvec = pci_alloc_irq_vectors(adev->pdev, 1, 1, flags); if (nvec > 0) { @@ -331,7 +345,7 @@ void amdgpu_irq_fini_hw(struct amdgpu_device *adev) */ void amdgpu_irq_fini_sw(struct amdgpu_device *adev) { - unsigned i, j; + unsigned int i, j; for (i = 0; i < AMDGPU_IRQ_CLIENTID_MAX; ++i) { if (!adev->irq.client[i].sources) @@ -365,7 +379,7 @@ void amdgpu_irq_fini_sw(struct amdgpu_device *adev) * 0 on success or error code otherwise */ int amdgpu_irq_add_id(struct amdgpu_device *adev, - unsigned client_id, unsigned src_id, + unsigned int client_id, unsigned int src_id, struct amdgpu_irq_src *source) { if (client_id >= AMDGPU_IRQ_CLIENTID_MAX) @@ -417,7 +431,7 @@ void amdgpu_irq_dispatch(struct amdgpu_device *adev, { u32 ring_index = ih->rptr >> 2; struct amdgpu_iv_entry entry; - unsigned client_id, src_id; + unsigned int client_id, src_id; struct amdgpu_irq_src *src; bool handled = false; int r; @@ -453,7 +467,8 @@ void amdgpu_irq_dispatch(struct amdgpu_device *adev, handled = true; } else { - DRM_DEBUG("Unhandled interrupt src_id: %d\n", src_id); + DRM_DEBUG("Unregistered interrupt src_id: %d of client_id:%d\n", + src_id, client_id); } /* Send it to amdkfd as well if it isn't already handled */ @@ -492,7 +507,7 @@ void amdgpu_irq_delegate(struct amdgpu_device *adev, * Updates interrupt state for the specific source (all ASICs). */ int amdgpu_irq_update(struct amdgpu_device *adev, - struct amdgpu_irq_src *src, unsigned type) + struct amdgpu_irq_src *src, unsigned int type) { unsigned long irqflags; enum amdgpu_interrupt_state state; @@ -501,7 +516,8 @@ int amdgpu_irq_update(struct amdgpu_device *adev, spin_lock_irqsave(&adev->irq.lock, irqflags); /* We need to determine after taking the lock, otherwise - we might disable just enabled interrupts again */ + * we might disable just enabled interrupts again + */ if (amdgpu_irq_enabled(adev, src, type)) state = AMDGPU_IRQ_STATE_ENABLE; else @@ -555,7 +571,7 @@ void amdgpu_irq_gpu_reset_resume_helper(struct amdgpu_device *adev) * 0 on success or error code otherwise */ int amdgpu_irq_get(struct amdgpu_device *adev, struct amdgpu_irq_src *src, - unsigned type) + unsigned int type) { if (!adev->irq.installed) return -ENOENT; @@ -585,7 +601,7 @@ int amdgpu_irq_get(struct amdgpu_device *adev, struct amdgpu_irq_src *src, * 0 on success or error code otherwise */ int amdgpu_irq_put(struct amdgpu_device *adev, struct amdgpu_irq_src *src, - unsigned type) + unsigned int type) { if (!adev->irq.installed) return -ENOENT; @@ -619,7 +635,7 @@ int amdgpu_irq_put(struct amdgpu_device *adev, struct amdgpu_irq_src *src, * invalid parameters */ bool amdgpu_irq_enabled(struct amdgpu_device *adev, struct amdgpu_irq_src *src, - unsigned type) + unsigned int type) { if (!adev->irq.installed) return false; @@ -732,7 +748,7 @@ void amdgpu_irq_remove_domain(struct amdgpu_device *adev) * Returns: * Linux IRQ */ -unsigned amdgpu_irq_create_mapping(struct amdgpu_device *adev, unsigned src_id) +unsigned int amdgpu_irq_create_mapping(struct amdgpu_device *adev, unsigned int src_id) { adev->irq.virq[src_id] = irq_create_mapping(adev->irq.domain, src_id); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h index be243adf3e6574852141b4a45441bff2beec0a8d..04c0b4fa17a4e7bf9628c52f9d03cbcd443f81ee 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h @@ -53,7 +53,7 @@ struct amdgpu_iv_entry { uint64_t timestamp; unsigned timestamp_src; unsigned pasid; - unsigned pasid_src; + unsigned node_id; unsigned src_data[AMDGPU_IRQ_SRC_DATA_MAX_SIZE_DW]; const uint32_t *iv_entry; }; @@ -102,6 +102,24 @@ struct amdgpu_irq { bool retry_cam_enabled; }; +enum interrupt_node_id_per_aid { + AID0_NODEID = 0, + XCD0_NODEID = 1, + XCD1_NODEID = 2, + AID1_NODEID = 4, + XCD2_NODEID = 5, + XCD3_NODEID = 6, + AID2_NODEID = 8, + XCD4_NODEID = 9, + XCD5_NODEID = 10, + AID3_NODEID = 12, + XCD6_NODEID = 13, + XCD7_NODEID = 14, + NODEID_MAX, +}; + +extern const int node_id_to_phys_map[NODEID_MAX]; + void amdgpu_irq_disable_all(struct amdgpu_device *adev); int amdgpu_irq_init(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c index c3d9d75143f4ffb8d1a9b3ac6c53b9ce4b84211a..78476bc75b4e1d151b4b5632126392dea6e984d9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c @@ -65,6 +65,8 @@ static enum drm_gpu_sched_stat amdgpu_job_timedout(struct drm_sched_job *s_job) DRM_ERROR("Process information: process %s pid %d thread %s pid %d\n", ti.process_name, ti.tgid, ti.task_name, ti.pid); + dma_fence_set_error(&s_job->s_fence->finished, -ETIME); + if (amdgpu_device_should_recover_gpu(ring->adev)) { struct amdgpu_reset_context reset_context; memset(&reset_context, 0, sizeof(reset_context)); @@ -107,7 +109,7 @@ int amdgpu_job_alloc(struct amdgpu_device *adev, struct amdgpu_vm *vm, (*job)->vm = vm; amdgpu_sync_create(&(*job)->explicit_sync); - (*job)->vram_lost_counter = atomic_read(&adev->vram_lost_counter); + (*job)->generation = amdgpu_vm_generation(adev, vm); (*job)->vm_pd_addr = AMDGPU_BO_INVALID_OFFSET; if (!entity) @@ -256,16 +258,27 @@ amdgpu_job_prepare_job(struct drm_sched_job *sched_job, struct dma_fence *fence = NULL; int r; + /* Ignore soft recovered fences here */ + r = drm_sched_entity_error(s_entity); + if (r && r != -ENODATA) + goto error; + if (!fence && job->gang_submit) fence = amdgpu_device_switch_gang(ring->adev, job->gang_submit); while (!fence && job->vm && !job->vmid) { r = amdgpu_vmid_grab(job->vm, ring, job, &fence); - if (r) + if (r) { DRM_ERROR("Error getting VM ID (%d)\n", r); + goto error; + } } return fence; + +error: + dma_fence_set_error(&job->base.s_fence->finished, r); + return NULL; } static struct dma_fence *amdgpu_job_run(struct drm_sched_job *sched_job) @@ -282,7 +295,7 @@ static struct dma_fence *amdgpu_job_run(struct drm_sched_job *sched_job) trace_amdgpu_sched_run_job(job); /* Skip job if VRAM is lost and never resubmit gangs */ - if (job->vram_lost_counter != atomic_read(&adev->vram_lost_counter) || + if (job->generation != amdgpu_vm_generation(adev, job->vm) || (job->job_run_counter && job->gang_submit)) dma_fence_set_error(finished, -ECANCELED); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h index 52f2e313ea17f299678b44ce17bfc36a48d2346b..a963a25ddd6209653fb3e70516d06b3798ec2042 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h @@ -61,12 +61,18 @@ struct amdgpu_job { uint32_t gds_base, gds_size; uint32_t gws_base, gws_size; uint32_t oa_base, oa_size; - uint32_t vram_lost_counter; + uint64_t generation; /* user fence handling */ uint64_t uf_addr; uint64_t uf_sequence; + /* virtual addresses for shadow/GDS/CSA */ + uint64_t shadow_va; + uint64_t csa_va; + uint64_t gds_va; + bool init_shadow; + /* job_run_counter >= 1 means a resubmit job */ uint32_t job_run_counter; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c index b07c000fc8ba39ee60540312eb8c0e6ee439d8e0..3add4b4f0667f4ace2696b9d46f9ea87dcc9f224 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c @@ -45,13 +45,14 @@ int amdgpu_jpeg_sw_init(struct amdgpu_device *adev) int amdgpu_jpeg_sw_fini(struct amdgpu_device *adev) { - int i; + int i, j; for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { if (adev->jpeg.harvest_config & (1 << i)) continue; - amdgpu_ring_fini(&adev->jpeg.inst[i].ring_dec); + for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) + amdgpu_ring_fini(&adev->jpeg.inst[i].ring_dec[j]); } mutex_destroy(&adev->jpeg.jpeg_pg_lock); @@ -76,13 +77,14 @@ static void amdgpu_jpeg_idle_work_handler(struct work_struct *work) struct amdgpu_device *adev = container_of(work, struct amdgpu_device, jpeg.idle_work.work); unsigned int fences = 0; - unsigned int i; + unsigned int i, j; for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { if (adev->jpeg.harvest_config & (1 << i)) continue; - fences += amdgpu_fence_count_emitted(&adev->jpeg.inst[i].ring_dec); + for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) + fences += amdgpu_fence_count_emitted(&adev->jpeg.inst[i].ring_dec[j]); } if (!fences && !atomic_read(&adev->jpeg.total_submission_cnt)) @@ -122,18 +124,21 @@ int amdgpu_jpeg_dec_ring_test_ring(struct amdgpu_ring *ring) if (amdgpu_sriov_vf(adev)) return 0; - WREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch, 0xCAFEDEAD); r = amdgpu_ring_alloc(ring, 3); if (r) return r; - amdgpu_ring_write(ring, PACKET0(adev->jpeg.internal.jpeg_pitch, 0)); - amdgpu_ring_write(ring, 0xDEADBEEF); + WREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch[ring->pipe], 0xCAFEDEAD); + /* Add a read register to make sure the write register is executed. */ + RREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch[ring->pipe]); + + amdgpu_ring_write(ring, PACKET0(adev->jpeg.internal.jpeg_pitch[ring->pipe], 0)); + amdgpu_ring_write(ring, 0xABADCAFE); amdgpu_ring_commit(ring); for (i = 0; i < adev->usec_timeout; i++) { - tmp = RREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch); - if (tmp == 0xDEADBEEF) + tmp = RREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch[ring->pipe]); + if (tmp == 0xABADCAFE) break; udelay(1); } @@ -161,8 +166,7 @@ static int amdgpu_jpeg_dec_set_reg(struct amdgpu_ring *ring, uint32_t handle, ib = &job->ibs[0]; - ib->ptr[0] = PACKETJ(adev->jpeg.internal.jpeg_pitch, 0, 0, - PACKETJ_TYPE0); + ib->ptr[0] = PACKETJ(adev->jpeg.internal.jpeg_pitch[ring->pipe], 0, 0, PACKETJ_TYPE0); ib->ptr[1] = 0xDEADBEEF; for (i = 2; i < 16; i += 2) { ib->ptr[i] = PACKETJ(0, 0, 0, PACKETJ_TYPE6); @@ -208,7 +212,7 @@ int amdgpu_jpeg_dec_ring_test_ib(struct amdgpu_ring *ring, long timeout) } if (!amdgpu_sriov_vf(adev)) { for (i = 0; i < adev->usec_timeout; i++) { - tmp = RREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch); + tmp = RREG32(adev->jpeg.inst[ring->me].external.jpeg_pitch[ring->pipe]); if (tmp == 0xDEADBEEF) break; udelay(1); @@ -241,6 +245,31 @@ int amdgpu_jpeg_process_poison_irq(struct amdgpu_device *adev, return 0; } +int amdgpu_jpeg_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block) +{ + int r, i; + + r = amdgpu_ras_block_late_init(adev, ras_block); + if (r) + return r; + + if (amdgpu_ras_is_supported(adev, ras_block->block)) { + for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { + if (adev->jpeg.harvest_config & (1 << i)) + continue; + + r = amdgpu_irq_get(adev, &adev->jpeg.inst[i].ras_poison_irq, 0); + if (r) + goto late_fini; + } + } + return 0; + +late_fini: + amdgpu_ras_block_late_fini(adev, ras_block); + return r; +} + int amdgpu_jpeg_ras_sw_init(struct amdgpu_device *adev) { int err; @@ -262,7 +291,7 @@ int amdgpu_jpeg_ras_sw_init(struct amdgpu_device *adev) adev->jpeg.ras_if = &ras->ras_block.ras_comm; if (!ras->ras_block.ras_late_init) - ras->ras_block.ras_late_init = amdgpu_ras_block_late_init; + ras->ras_block.ras_late_init = amdgpu_jpeg_ras_late_init; return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h index 0ca76f0f23e9c8a0b980050cfbf27afa7307ef34..ffe47e9f5bf2f6bf6728ed394330d521306e06b8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h @@ -26,19 +26,22 @@ #include "amdgpu_ras.h" -#define AMDGPU_MAX_JPEG_INSTANCES 2 +#define AMDGPU_MAX_JPEG_INSTANCES 4 +#define AMDGPU_MAX_JPEG_RINGS 8 #define AMDGPU_JPEG_HARVEST_JPEG0 (1 << 0) #define AMDGPU_JPEG_HARVEST_JPEG1 (1 << 1) struct amdgpu_jpeg_reg{ - unsigned jpeg_pitch; + unsigned jpeg_pitch[AMDGPU_MAX_JPEG_RINGS]; }; struct amdgpu_jpeg_inst { - struct amdgpu_ring ring_dec; + struct amdgpu_ring ring_dec[AMDGPU_MAX_JPEG_RINGS]; struct amdgpu_irq_src irq; + struct amdgpu_irq_src ras_poison_irq; struct amdgpu_jpeg_reg external; + uint8_t aid_id; }; struct amdgpu_jpeg_ras { @@ -48,6 +51,7 @@ struct amdgpu_jpeg_ras { struct amdgpu_jpeg { uint8_t num_jpeg_inst; struct amdgpu_jpeg_inst inst[AMDGPU_MAX_JPEG_INSTANCES]; + unsigned num_jpeg_rings; struct amdgpu_jpeg_reg internal; unsigned harvest_config; struct delayed_work idle_work; @@ -56,6 +60,9 @@ struct amdgpu_jpeg { atomic_t total_submission_cnt; struct ras_common_if *ras_if; struct amdgpu_jpeg_ras *ras; + + uint16_t inst_mask; + uint8_t num_inst_per_aid; }; int amdgpu_jpeg_sw_init(struct amdgpu_device *adev); @@ -72,6 +79,8 @@ int amdgpu_jpeg_dec_ring_test_ib(struct amdgpu_ring *ring, long timeout); int amdgpu_jpeg_process_poison_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry); +int amdgpu_jpeg_ras_late_init(struct amdgpu_device *adev, + struct ras_common_if *ras_block); int amdgpu_jpeg_ras_sw_init(struct amdgpu_device *adev); #endif /*__AMDGPU_JPEG_H__*/ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 0efb38539d70cc6dff5da3e222cf3b4a2ac3a25f..e3531aa3c8bd101935314b6214f8181c5a0c05aa 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -462,8 +462,9 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev, if (adev->jpeg.harvest_config & (1 << i)) continue; - if (adev->jpeg.inst[i].ring_dec.sched.ready) - ++num_rings; + for (j = 0; j < adev->jpeg.num_jpeg_rings; j++) + if (adev->jpeg.inst[i].ring_dec[j].sched.ready) + ++num_rings; } ib_start_alignment = 16; ib_size_alignment = 16; @@ -876,6 +877,19 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) dev_info->gl2c_cache_size = adev->gfx.config.gc_gl2c_per_gpu; dev_info->mall_size = adev->gmc.mall_size; + + if (adev->gfx.funcs->get_gfx_shadow_info) { + struct amdgpu_gfx_shadow_info shadow_info; + + ret = amdgpu_gfx_get_gfx_shadow_info(adev, &shadow_info); + if (!ret) { + dev_info->shadow_size = shadow_info.shadow_size; + dev_info->shadow_alignment = shadow_info.shadow_alignment; + dev_info->csa_size = shadow_info.csa_size; + dev_info->csa_alignment = shadow_info.csa_alignment; + } + } + ret = copy_to_user(out, dev_info, min((size_t)size, sizeof(*dev_info))) ? -EFAULT : 0; kfree(dev_info); @@ -1140,6 +1154,15 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) kfree(caps); return r; } + case AMDGPU_INFO_MAX_IBS: { + uint32_t max_ibs[AMDGPU_HW_IP_NUM]; + + for (i = 0; i < AMDGPU_HW_IP_NUM; ++i) + max_ibs[i] = amdgpu_ring_max_ibs(i); + + return copy_to_user(out, max_ibs, + min((size_t)size, sizeof(max_ibs))) ? -EFAULT : 0; + } default: DRM_DEBUG_KMS("Invalid request %d\n", info->query); return -EINVAL; @@ -1210,6 +1233,10 @@ int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv) if (r) goto error_pasid; + r = amdgpu_xcp_open_device(adev, fpriv, file_priv); + if (r) + goto error_vm; + r = amdgpu_vm_set_pasid(adev, &fpriv->vm, pasid); if (r) goto error_vm; @@ -1284,12 +1311,12 @@ void amdgpu_driver_postclose_kms(struct drm_device *dev, if (amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_VCE) != NULL) amdgpu_vce_free_handles(adev, file_priv); - if (amdgpu_mcbp) { - /* TODO: how to handle reserve failure */ - BUG_ON(amdgpu_bo_reserve(adev->virt.csa_obj, true)); - amdgpu_vm_bo_del(adev, fpriv->csa_va); + if (fpriv->csa_va) { + uint64_t csa_addr = amdgpu_csa_vaddr(adev) & AMDGPU_GMC_HOLE_MASK; + + WARN_ON(amdgpu_unmap_static_csa(adev, &fpriv->vm, adev->virt.csa_obj, + fpriv->csa_va, csa_addr)); fpriv->csa_va = NULL; - amdgpu_bo_unreserve(adev->virt.csa_obj); } pasid = fpriv->vm.pasid; @@ -1441,7 +1468,7 @@ void amdgpu_disable_vblank_kms(struct drm_crtc *crtc) static int amdgpu_debugfs_firmware_info_show(struct seq_file *m, void *unused) { - struct amdgpu_device *adev = (struct amdgpu_device *)m->private; + struct amdgpu_device *adev = m->private; struct drm_amdgpu_info_firmware fw_info; struct drm_amdgpu_query_fw query_fw; struct atom_context *ctx = adev->mode_info.atom_context; @@ -1449,7 +1476,7 @@ static int amdgpu_debugfs_firmware_info_show(struct seq_file *m, void *unused) int ret, i; static const char *ta_fw_name[TA_FW_TYPE_MAX_INDEX] = { -#define TA_FW_NAME(type) [TA_FW_TYPE_PSP_##type] = #type +#define TA_FW_NAME(type)[TA_FW_TYPE_PSP_##type] = #type TA_FW_NAME(XGMI), TA_FW_NAME(RAS), TA_FW_NAME(HDCP), @@ -1548,7 +1575,7 @@ static int amdgpu_debugfs_firmware_info_show(struct seq_file *m, void *unused) fw_info.feature, fw_info.ver); /* RLCV */ - query_fw.fw_type = AMDGPU_INFO_FW_GFX_RLCV; + query_fw.fw_type = AMDGPU_INFO_FW_GFX_RLCV; ret = amdgpu_firmware_info(&fw_info, &query_fw, adev); if (ret) return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c index f0f00466b59f2943dec6d8b067604f088367fdf5..e9091ebfe230db24072a7abc0dc9ff1effd9267f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c @@ -924,6 +924,43 @@ int amdgpu_mes_reg_wait(struct amdgpu_device *adev, uint32_t reg, return r; } +int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev, + uint64_t process_context_addr, + uint32_t spi_gdbg_per_vmid_cntl, + const uint32_t *tcp_watch_cntl, + uint32_t flags, + bool trap_en) +{ + struct mes_misc_op_input op_input = {0}; + int r; + + if (!adev->mes.funcs->misc_op) { + DRM_ERROR("mes set shader debugger is not supported!\n"); + return -EINVAL; + } + + op_input.op = MES_MISC_OP_SET_SHADER_DEBUGGER; + op_input.set_shader_debugger.process_context_addr = process_context_addr; + op_input.set_shader_debugger.flags.u32all = flags; + op_input.set_shader_debugger.spi_gdbg_per_vmid_cntl = spi_gdbg_per_vmid_cntl; + memcpy(op_input.set_shader_debugger.tcp_watch_cntl, tcp_watch_cntl, + sizeof(op_input.set_shader_debugger.tcp_watch_cntl)); + + if (((adev->mes.sched_version & AMDGPU_MES_API_VERSION_MASK) >> + AMDGPU_MES_API_VERSION_SHIFT) >= 14) + op_input.set_shader_debugger.trap_en = trap_en; + + amdgpu_mes_lock(&adev->mes); + + r = adev->mes.funcs->misc_op(&adev->mes, &op_input); + if (r) + DRM_ERROR("failed to set_shader_debugger\n"); + + amdgpu_mes_unlock(&adev->mes); + + return r; +} + static void amdgpu_mes_ring_to_queue_props(struct amdgpu_device *adev, struct amdgpu_ring *ring, @@ -1305,14 +1342,9 @@ static int amdgpu_mes_test_queues(struct amdgpu_ring **added_rings) if (!ring) continue; - r = amdgpu_ring_test_ring(ring); - if (r) { - DRM_DEV_ERROR(ring->adev->dev, - "ring %s test failed (%d)\n", - ring->name, r); + r = amdgpu_ring_test_helper(ring); + if (r) return r; - } else - DRM_INFO("ring %s test pass\n", ring->name); r = amdgpu_ring_test_ib(ring, 1000 * 10); if (r) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h index 547ec35691fac633e2e7f0256cfdbdadc583d5c8..2d6ac30b7135b894674224b826356c30ec3c8f3d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h @@ -219,6 +219,8 @@ struct mes_add_queue_input { uint32_t gws_size; uint64_t tba_addr; uint64_t tma_addr; + uint32_t trap_en; + uint32_t skip_process_ctx_clear; uint32_t is_kfd_process; uint32_t is_aql_queue; uint32_t queue_size; @@ -256,6 +258,7 @@ enum mes_misc_opcode { MES_MISC_OP_READ_REG, MES_MISC_OP_WRM_REG_WAIT, MES_MISC_OP_WRM_REG_WR_WAIT, + MES_MISC_OP_SET_SHADER_DEBUGGER, }; struct mes_misc_op_input { @@ -278,6 +281,21 @@ struct mes_misc_op_input { uint32_t reg0; uint32_t reg1; } wrm_reg; + + struct { + uint64_t process_context_addr; + union { + struct { + uint64_t single_memop : 1; + uint64_t single_alu_op : 1; + uint64_t reserved: 30; + }; + uint32_t u32all; + } flags; + uint32_t spi_gdbg_per_vmid_cntl; + uint32_t tcp_watch_cntl[4]; + uint32_t trap_en; + } set_shader_debugger; }; }; @@ -340,6 +358,12 @@ int amdgpu_mes_reg_wait(struct amdgpu_device *adev, uint32_t reg, int amdgpu_mes_reg_write_reg_wait(struct amdgpu_device *adev, uint32_t reg0, uint32_t reg1, uint32_t ref, uint32_t mask); +int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev, + uint64_t process_context_addr, + uint32_t spi_gdbg_per_vmid_cntl, + const uint32_t *tcp_watch_cntl, + uint32_t flags, + bool trap_en); int amdgpu_mes_add_ring(struct amdgpu_device *adev, int gang_id, int queue_type, int idx, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h index d21bb6dae56eb8d8812744330415db72a7709308..1ca9d4ed8063a61a6713bf5054273a333248e126 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h @@ -21,6 +21,29 @@ #ifndef __AMDGPU_MMHUB_H__ #define __AMDGPU_MMHUB_H__ +enum amdgpu_mmhub_ras_memory_id { + AMDGPU_MMHUB_WGMI_PAGEMEM = 0, + AMDGPU_MMHUB_RGMI_PAGEMEM = 1, + AMDGPU_MMHUB_WDRAM_PAGEMEM = 2, + AMDGPU_MMHUB_RDRAM_PAGEMEM = 3, + AMDGPU_MMHUB_WIO_CMDMEM = 4, + AMDGPU_MMHUB_RIO_CMDMEM = 5, + AMDGPU_MMHUB_WGMI_CMDMEM = 6, + AMDGPU_MMHUB_RGMI_CMDMEM = 7, + AMDGPU_MMHUB_WDRAM_CMDMEM = 8, + AMDGPU_MMHUB_RDRAM_CMDMEM = 9, + AMDGPU_MMHUB_MAM_DMEM0 = 10, + AMDGPU_MMHUB_MAM_DMEM1 = 11, + AMDGPU_MMHUB_MAM_DMEM2 = 12, + AMDGPU_MMHUB_MAM_DMEM3 = 13, + AMDGPU_MMHUB_WRET_TAGMEM = 19, + AMDGPU_MMHUB_RRET_TAGMEM = 20, + AMDGPU_MMHUB_WIO_DATAMEM = 21, + AMDGPU_MMHUB_WGMI_DATAMEM = 22, + AMDGPU_MMHUB_WDRAM_DATAMEM = 23, + AMDGPU_MMHUB_MEMORY_BLOCK_LAST, +}; + struct amdgpu_mmhub_ras { struct amdgpu_ras_block_object ras_block; }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h index c686ff4bcc393721957c10f97b26098df1013e89..8ab8ae01f87c3a0e64bb7871efc5fd37a15a52f3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h @@ -61,6 +61,7 @@ struct amdgpu_nbio_funcs { u32 (*get_hdp_flush_done_offset)(struct amdgpu_device *adev); u32 (*get_pcie_index_offset)(struct amdgpu_device *adev); u32 (*get_pcie_data_offset)(struct amdgpu_device *adev); + u32 (*get_pcie_index_hi_offset)(struct amdgpu_device *adev); u32 (*get_pcie_port_index_offset)(struct amdgpu_device *adev); u32 (*get_pcie_port_data_offset)(struct amdgpu_device *adev); u32 (*get_rev_id)(struct amdgpu_device *adev); @@ -95,6 +96,9 @@ struct amdgpu_nbio_funcs { void (*apply_l1_link_width_reconfig_wa)(struct amdgpu_device *adev); void (*clear_doorbell_interrupt)(struct amdgpu_device *adev); u32 (*get_rom_offset)(struct amdgpu_device *adev); + int (*get_compute_partition_mode)(struct amdgpu_device *adev); + u32 (*get_memory_partition_mode)(struct amdgpu_device *adev, + u32 *supp_modes); }; struct amdgpu_nbio { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index 2bd1a54ee86656bcf06c2e135c58a9f3a9b9d8ba..f7905bce0de15e49202dc23d022b9e68167bdcc8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -79,9 +79,10 @@ static void amdgpu_bo_user_destroy(struct ttm_buffer_object *tbo) static void amdgpu_bo_vm_destroy(struct ttm_buffer_object *tbo) { struct amdgpu_device *adev = amdgpu_ttm_adev(tbo->bdev); - struct amdgpu_bo *bo = ttm_to_amdgpu_bo(tbo); + struct amdgpu_bo *shadow_bo = ttm_to_amdgpu_bo(tbo), *bo; struct amdgpu_bo_vm *vmbo; + bo = shadow_bo->parent; vmbo = to_amdgpu_bo_vm(bo); /* in case amdgpu_device_recover_vram got NULL of bo->parent */ if (!list_empty(&vmbo->shadow_list)) { @@ -130,16 +131,26 @@ void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain) u32 c = 0; if (domain & AMDGPU_GEM_DOMAIN_VRAM) { - unsigned visible_pfn = adev->gmc.visible_vram_size >> PAGE_SHIFT; - - places[c].fpfn = 0; - places[c].lpfn = 0; + unsigned int visible_pfn = adev->gmc.visible_vram_size >> PAGE_SHIFT; + int8_t mem_id = KFD_XCP_MEM_ID(adev, abo->xcp_id); + + if (adev->gmc.mem_partitions && mem_id >= 0) { + places[c].fpfn = adev->gmc.mem_partitions[mem_id].range.fpfn; + /* + * memory partition range lpfn is inclusive start + size - 1 + * TTM place lpfn is exclusive start + size + */ + places[c].lpfn = adev->gmc.mem_partitions[mem_id].range.lpfn + 1; + } else { + places[c].fpfn = 0; + places[c].lpfn = 0; + } places[c].mem_type = TTM_PL_VRAM; places[c].flags = 0; if (flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) - places[c].lpfn = visible_pfn; - else if (adev->gmc.real_vram_size != adev->gmc.visible_vram_size) + places[c].lpfn = min_not_zero(places[c].lpfn, visible_pfn); + else places[c].flags |= TTM_PL_FLAG_TOPDOWN; if (flags & AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS) @@ -574,6 +585,13 @@ int amdgpu_bo_create(struct amdgpu_device *adev, bo->flags = bp->flags; + if (adev->gmc.mem_partitions) + /* For GPUs with spatial partitioning, bo->xcp_id=-1 means any partition */ + bo->xcp_id = bp->xcp_id_plus1 - 1; + else + /* For GPUs without spatial partitioning */ + bo->xcp_id = 0; + if (!amdgpu_bo_support_uswc(bo->flags)) bo->flags &= ~AMDGPU_GEM_CREATE_CPU_GTT_USWC; @@ -610,7 +628,7 @@ int amdgpu_bo_create(struct amdgpu_device *adev, bo->tbo.resource->mem_type == TTM_PL_VRAM) { struct dma_fence *fence; - r = amdgpu_fill_buffer(bo, 0, bo->tbo.base.resv, &fence); + r = amdgpu_fill_buffer(bo, 0, bo->tbo.base.resv, &fence, true); if (unlikely(r)) goto fail_unreserve; @@ -694,11 +712,6 @@ int amdgpu_bo_create_vm(struct amdgpu_device *adev, return r; *vmbo_ptr = to_amdgpu_bo_vm(bo_ptr); - INIT_LIST_HEAD(&(*vmbo_ptr)->shadow_list); - /* Set destroy callback to amdgpu_bo_vm_destroy after vmbo->shadow_list - * is initialized. - */ - bo_ptr->tbo.destroy = &amdgpu_bo_vm_destroy; return r; } @@ -715,6 +728,8 @@ void amdgpu_bo_add_to_shadow_list(struct amdgpu_bo_vm *vmbo) mutex_lock(&adev->shadow_list_lock); list_add_tail(&vmbo->shadow_list, &adev->shadow_list); + vmbo->shadow->parent = amdgpu_bo_ref(&vmbo->bo); + vmbo->shadow->tbo.destroy = &amdgpu_bo_vm_destroy; mutex_unlock(&adev->shadow_list_lock); } @@ -935,7 +950,7 @@ int amdgpu_bo_pin_restricted(struct amdgpu_bo *bo, u32 domain, bo->flags |= AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED; amdgpu_bo_placement_from_domain(bo, domain); for (i = 0; i < bo->placement.num_placement; i++) { - unsigned fpfn, lpfn; + unsigned int fpfn, lpfn; fpfn = min_offset >> PAGE_SHIFT; lpfn = max_offset >> PAGE_SHIFT; @@ -1016,7 +1031,7 @@ void amdgpu_bo_unpin(struct amdgpu_bo *bo) } } -static const char *amdgpu_vram_names[] = { +static const char * const amdgpu_vram_names[] = { "UNKNOWN", "GDDR1", "DDR2", @@ -1044,7 +1059,7 @@ static const char *amdgpu_vram_names[] = { int amdgpu_bo_init(struct amdgpu_device *adev) { /* On A+A platform, VRAM can be mapped as WB */ - if (!adev->gmc.xgmi.connected_to_cpu) { + if (!adev->gmc.xgmi.connected_to_cpu && !adev->gmc.is_app_apu) { /* reserve PAT memory space to WC for VRAM */ int r = arch_io_reserve_memtype_wc(adev->gmc.aper_base, adev->gmc.aper_size); @@ -1080,8 +1095,7 @@ void amdgpu_bo_fini(struct amdgpu_device *adev) amdgpu_ttm_fini(adev); if (drm_dev_enter(adev_to_drm(adev), &idx)) { - - if (!adev->gmc.xgmi.connected_to_cpu) { + if (!adev->gmc.xgmi.connected_to_cpu && !adev->gmc.is_app_apu) { arch_phys_wc_del(adev->gmc.vram_mtrr); arch_io_free_memtype_wc(adev->gmc.aper_base, adev->gmc.aper_size); } @@ -1148,8 +1162,8 @@ void amdgpu_bo_get_tiling_flags(struct amdgpu_bo *bo, u64 *tiling_flags) * Returns: * 0 for success or a negative error code on failure. */ -int amdgpu_bo_set_metadata (struct amdgpu_bo *bo, void *metadata, - uint32_t metadata_size, uint64_t flags) +int amdgpu_bo_set_metadata(struct amdgpu_bo *bo, void *metadata, + u32 metadata_size, uint64_t flags) { struct amdgpu_bo_user *ubo; void *buffer; @@ -1268,8 +1282,12 @@ void amdgpu_bo_move_notify(struct ttm_buffer_object *bo, void amdgpu_bo_get_memory(struct amdgpu_bo *bo, struct amdgpu_mem_stats *stats) { - unsigned int domain; uint64_t size = amdgpu_bo_size(bo); + unsigned int domain; + + /* Abort if the BO doesn't currently have a backing store */ + if (!bo->tbo.resource) + return; domain = amdgpu_mem_type_to_domain(bo->tbo.resource->mem_type); switch (domain) { @@ -1338,7 +1356,7 @@ void amdgpu_bo_release_notify(struct ttm_buffer_object *bo) if (WARN_ON_ONCE(!dma_resv_trylock(bo->base.resv))) return; - r = amdgpu_fill_buffer(abo, AMDGPU_POISON, bo->base.resv, &fence); + r = amdgpu_fill_buffer(abo, AMDGPU_POISON, bo->base.resv, &fence, true); if (!WARN_ON(r)) { amdgpu_bo_fence(abo, fence, false); dma_fence_put(fence); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h index 35b8106816a13cf6dababc862da9c268d7a7502a..05496b97ef930a0bb65821400859595c14799ef1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h @@ -56,6 +56,8 @@ struct amdgpu_bo_param { bool no_wait_gpu; struct dma_resv *resv; void (*destroy)(struct ttm_buffer_object *bo); + /* xcp partition number plus 1, 0 means any partition */ + int8_t xcp_id_plus1; }; /* bo virtual addresses in a vm */ @@ -108,6 +110,13 @@ struct amdgpu_bo { struct mmu_interval_notifier notifier; #endif struct kgd_mem *kfd_bo; + + /* + * For GPUs with spatial partitioning, xcp partition number, -1 means + * any partition. For other ASICs without spatial partition, always 0 + * for memory accounting. + */ + int8_t xcp_id; }; struct amdgpu_bo_user { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index 9d7e6e0e73edb9722233d3bff87769e92c8434a8..e15c27e05564c7347be5a5f3a70f0fa4c5c23f90 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -146,6 +146,9 @@ static int psp_init_sriov_microcode(struct psp_context *psp) case IP_VERSION(13, 0, 0): adev->virt.autoload_ucode_id = 0; break; + case IP_VERSION(13, 0, 6): + ret = psp_init_cap_microcode(psp, ucode_prefix); + break; case IP_VERSION(13, 0, 10): adev->virt.autoload_ucode_id = AMDGPU_UCODE_ID_CP_MES1_DATA; ret = psp_init_cap_microcode(psp, ucode_prefix); @@ -329,6 +332,9 @@ static bool psp_get_runtime_db_entry(struct amdgpu_device *adev, bool ret = false; int i; + if (adev->ip_versions[MP0_HWIP][0] == IP_VERSION(13, 0, 6)) + return false; + db_header_pos = adev->gmc.mc_vram_size - PSP_RUNTIME_DB_OFFSET; db_dir_pos = db_header_pos + sizeof(struct psp_runtime_data_header); @@ -411,7 +417,7 @@ static int psp_sw_init(void *handle) if ((psp_get_runtime_db_entry(adev, PSP_RUNTIME_ENTRY_TYPE_PPTABLE_ERR_STATUS, &scpm_entry)) && - (SCPM_DISABLE != scpm_entry.scpm_status)) { + (scpm_entry.scpm_status != SCPM_DISABLE)) { adev->scpm_enabled = true; adev->scpm_status = scpm_entry.scpm_status; } else { @@ -458,10 +464,9 @@ static int psp_sw_init(void *handle) if (adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 0) || adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 7)) { - ret= psp_sysfs_init(adev); - if (ret) { + ret = psp_sysfs_init(adev); + if (ret) return ret; - } } ret = amdgpu_bo_create_kernel(adev, PSP_1_MEG, PSP_1_MEG, @@ -474,7 +479,8 @@ static int psp_sw_init(void *handle) return ret; ret = amdgpu_bo_create_kernel(adev, PSP_FENCE_BUFFER_SIZE, PAGE_SIZE, - AMDGPU_GEM_DOMAIN_VRAM, + AMDGPU_GEM_DOMAIN_VRAM | + AMDGPU_GEM_DOMAIN_GTT, &psp->fence_buf_bo, &psp->fence_buf_mc_addr, &psp->fence_buf); @@ -482,7 +488,8 @@ static int psp_sw_init(void *handle) goto failed1; ret = amdgpu_bo_create_kernel(adev, PSP_CMD_BUFFER_SIZE, PAGE_SIZE, - AMDGPU_GEM_DOMAIN_VRAM, + AMDGPU_GEM_DOMAIN_VRAM | + AMDGPU_GEM_DOMAIN_GTT, &psp->cmd_buf_bo, &psp->cmd_buf_mc_addr, (void **)&psp->cmd_buf_mem); if (ret) @@ -520,6 +527,8 @@ static int psp_sw_fini(void *handle) kfree(cmd); cmd = NULL; + psp_free_shared_bufs(psp); + if (psp->km_ring.ring_mem) amdgpu_bo_free_kernel(&adev->firmware.rbuf, &psp->km_ring.ring_mem_mc_addr, @@ -560,6 +569,26 @@ int psp_wait_for(struct psp_context *psp, uint32_t reg_index, return -ETIME; } +int psp_wait_for_spirom_update(struct psp_context *psp, uint32_t reg_index, + uint32_t reg_val, uint32_t mask, uint32_t msec_timeout) +{ + uint32_t val; + int i; + struct amdgpu_device *adev = psp->adev; + + if (psp->adev->no_hw_access) + return 0; + + for (i = 0; i < msec_timeout; i++) { + val = RREG32(reg_index); + if ((val & mask) == reg_val) + return 0; + msleep(1); + } + + return -ETIME; +} + static const char *psp_gfx_cmd_name(enum psp_gfx_cmd_id cmd_id) { switch (cmd_id) { @@ -643,7 +672,7 @@ psp_cmd_submit_buf(struct psp_context *psp, skip_unsupport = (psp->cmd_buf_mem->resp.status == TEE_ERROR_NOT_SUPPORTED || psp->cmd_buf_mem->resp.status == PSP_ERR_UNKNOWN_COMMAND) && amdgpu_sriov_vf(psp->adev); - memcpy((void*)&cmd->resp, (void*)&psp->cmd_buf_mem->resp, sizeof(struct psp_gfx_resp)); + memcpy(&cmd->resp, &psp->cmd_buf_mem->resp, sizeof(struct psp_gfx_resp)); /* In some cases, psp response status is not 0 even there is no * problem while the command is submitted. Some version of PSP FW @@ -699,8 +728,13 @@ static void psp_prep_tmr_cmd_buf(struct psp_context *psp, uint64_t tmr_mc, struct amdgpu_bo *tmr_bo) { struct amdgpu_device *adev = psp->adev; - uint32_t size = amdgpu_bo_size(tmr_bo); - uint64_t tmr_pa = amdgpu_gmc_vram_pa(adev, tmr_bo); + uint32_t size = 0; + uint64_t tmr_pa = 0; + + if (tmr_bo) { + size = amdgpu_bo_size(tmr_bo); + tmr_pa = amdgpu_gmc_vram_pa(adev, tmr_bo); + } if (amdgpu_sriov_vf(psp->adev)) cmd->cmd_id = GFX_CMD_ID_SETUP_VMR; @@ -745,6 +779,16 @@ static int psp_load_toc(struct psp_context *psp, return ret; } +static bool psp_boottime_tmr(struct psp_context *psp) +{ + switch (psp->adev->ip_versions[MP0_HWIP][0]) { + case IP_VERSION(13, 0, 6): + return true; + default: + return false; + } +} + /* Set up Trusted Memory Region */ static int psp_tmr_init(struct psp_context *psp) { @@ -816,8 +860,9 @@ static int psp_tmr_load(struct psp_context *psp) cmd = acquire_psp_cmd_buf(psp); psp_prep_tmr_cmd_buf(psp, cmd, psp->tmr_mc_addr, psp->tmr_bo); - DRM_INFO("reserve 0x%lx from 0x%llx for PSP TMR\n", - amdgpu_bo_size(psp->tmr_bo), psp->tmr_mc_addr); + if (psp->tmr_bo) + DRM_INFO("reserve 0x%lx from 0x%llx for PSP TMR\n", + amdgpu_bo_size(psp->tmr_bo), psp->tmr_mc_addr); ret = psp_cmd_submit_buf(psp, NULL, cmd, psp->fence_buf_mc_addr); @@ -828,7 +873,7 @@ static int psp_tmr_load(struct psp_context *psp) } static void psp_prep_tmr_unload_cmd_buf(struct psp_context *psp, - struct psp_gfx_cmd_resp *cmd) + struct psp_gfx_cmd_resp *cmd) { if (amdgpu_sriov_vf(psp->adev)) cmd->cmd_id = GFX_CMD_ID_DESTROY_VMR; @@ -969,6 +1014,27 @@ static int psp_rl_load(struct amdgpu_device *adev) return ret; } +int psp_spatial_partition(struct psp_context *psp, int mode) +{ + struct psp_gfx_cmd_resp *cmd; + int ret; + + if (amdgpu_sriov_vf(psp->adev)) + return 0; + + cmd = acquire_psp_cmd_buf(psp); + + cmd->cmd_id = GFX_CMD_ID_SRIOV_SPATIAL_PART; + cmd->cmd.cmd_spatial_part.mode = mode; + + dev_info(psp->adev->dev, "Requesting %d partitions through PSP", mode); + ret = psp_cmd_submit_buf(psp, NULL, cmd, psp->fence_buf_mc_addr); + + release_psp_cmd_buf(psp); + + return ret; +} + static int psp_asd_initialize(struct psp_context *psp) { int ret; @@ -1065,7 +1131,7 @@ static void psp_prep_ta_load_cmd_buf(struct psp_gfx_cmd_resp *cmd, struct ta_context *context) { cmd->cmd_id = context->ta_load_type; - cmd->cmd.cmd_load_ta.app_phy_addr_lo = lower_32_bits(ta_bin_mc); + cmd->cmd.cmd_load_ta.app_phy_addr_lo = lower_32_bits(ta_bin_mc); cmd->cmd.cmd_load_ta.app_phy_addr_hi = upper_32_bits(ta_bin_mc); cmd->cmd.cmd_load_ta.app_len = context->bin_desc.size_bytes; @@ -1136,9 +1202,8 @@ int psp_ta_load(struct psp_context *psp, struct ta_context *context) context->resp_status = cmd->resp.status; - if (!ret) { + if (!ret) context->session_id = cmd->resp.session_id; - } release_psp_cmd_buf(psp); @@ -1254,8 +1319,9 @@ int psp_xgmi_get_node_id(struct psp_context *psp, uint64_t *node_id) static bool psp_xgmi_peer_link_info_supported(struct psp_context *psp) { - return psp->adev->ip_versions[MP0_HWIP][0] == IP_VERSION(13, 0, 2) && - psp->xgmi_context.context.bin_desc.fw_version >= 0x2000000b; + return (psp->adev->ip_versions[MP0_HWIP][0] == IP_VERSION(13, 0, 2) && + psp->xgmi_context.context.bin_desc.fw_version >= 0x2000000b) || + psp->adev->ip_versions[MP0_HWIP][0] >= IP_VERSION(13, 0, 6); } /* @@ -1363,6 +1429,9 @@ int psp_xgmi_get_topology_info(struct psp_context *psp, /* Invoke xgmi ta again to get the link information */ if (psp_xgmi_peer_link_info_supported(psp)) { struct ta_xgmi_cmd_get_peer_link_info_output *link_info_output; + bool requires_reflection = + (psp->xgmi_context.supports_extended_data && get_extended_data) || + psp->adev->ip_versions[MP0_HWIP][0] == IP_VERSION(13, 0, 6); xgmi_cmd->cmd_id = TA_COMMAND_XGMI__GET_PEER_LINKS; @@ -1377,11 +1446,11 @@ int psp_xgmi_get_topology_info(struct psp_context *psp, topology->nodes[i].num_links = get_extended_data ? topology->nodes[i].num_links + link_info_output->nodes[i].num_links : - link_info_output->nodes[i].num_links; + ((requires_reflection && topology->nodes[i].num_links) ? topology->nodes[i].num_links : + link_info_output->nodes[i].num_links); /* reflect the topology information for bi-directionality */ - if (psp->xgmi_context.supports_extended_data && - get_extended_data && topology->nodes[i].num_hops) + if (requires_reflection && topology->nodes[i].num_hops) psp_xgmi_reflect_topology_info(psp, topology->nodes[i]); } } @@ -1465,8 +1534,7 @@ int psp_ras_invoke(struct psp_context *psp, uint32_t ta_cmd_id) if (amdgpu_ras_intr_triggered()) return ret; - if (ras_cmd->if_version > RAS_TA_HOST_IF_VER) - { + if (ras_cmd->if_version > RAS_TA_HOST_IF_VER) { DRM_WARN("RAS: Unsupported Interface"); return -EINVAL; } @@ -1476,8 +1544,7 @@ int psp_ras_invoke(struct psp_context *psp, uint32_t ta_cmd_id) dev_warn(psp->adev->dev, "ECC switch disabled\n"); ras_cmd->ras_status = TA_RAS_STATUS__ERROR_RAS_NOT_AVAILABLE; - } - else if (ras_cmd->ras_out_message.flags.reg_access_failure_flag) + } else if (ras_cmd->ras_out_message.flags.reg_access_failure_flag) dev_warn(psp->adev->dev, "RAS internal register access blocked\n"); @@ -1573,11 +1640,10 @@ int psp_ras_initialize(struct psp_context *psp) if (ret) dev_warn(adev->dev, "PSP set boot config failed\n"); else - dev_warn(adev->dev, "GECC will be disabled in next boot cycle " - "if set amdgpu_ras_enable and/or amdgpu_ras_mask to 0x0\n"); + dev_warn(adev->dev, "GECC will be disabled in next boot cycle if set amdgpu_ras_enable and/or amdgpu_ras_mask to 0x0\n"); } } else { - if (1 == boot_cfg) { + if (boot_cfg == 1) { dev_info(adev->dev, "GECC is enabled\n"); } else { /* enable GECC in next boot cycle if it is disabled @@ -1607,8 +1673,11 @@ int psp_ras_initialize(struct psp_context *psp) if (amdgpu_ras_is_poison_mode_supported(adev)) ras_cmd->ras_in_message.init_flags.poison_mode_en = 1; - if (!adev->gmc.xgmi.connected_to_cpu) + if (!adev->gmc.xgmi.connected_to_cpu && !adev->gmc.is_app_apu) ras_cmd->ras_in_message.init_flags.dgpu_mode = 1; + ras_cmd->ras_in_message.init_flags.xcc_mask = + adev->gfx.xcc_mask; + ras_cmd->ras_in_message.init_flags.channel_dis_num = hweight32(adev->gmc.m_half_use) * 2; ret = psp_ta_load(psp, &psp->ras_context.context); @@ -1626,14 +1695,37 @@ int psp_ras_initialize(struct psp_context *psp) } int psp_ras_trigger_error(struct psp_context *psp, - struct ta_ras_trigger_error_input *info) + struct ta_ras_trigger_error_input *info, uint32_t instance_mask) { struct ta_ras_shared_memory *ras_cmd; + struct amdgpu_device *adev = psp->adev; int ret; + uint32_t dev_mask; if (!psp->ras_context.context.initialized) return -EINVAL; + switch (info->block_id) { + case TA_RAS_BLOCK__GFX: + dev_mask = GET_MASK(GC, instance_mask); + break; + case TA_RAS_BLOCK__SDMA: + dev_mask = GET_MASK(SDMA0, instance_mask); + break; + case TA_RAS_BLOCK__VCN: + case TA_RAS_BLOCK__JPEG: + dev_mask = GET_MASK(VCN, instance_mask); + break; + default: + dev_mask = instance_mask; + break; + } + + /* reuse sub_block_index for backward compatibility */ + dev_mask <<= AMDGPU_RAS_INST_SHIFT; + dev_mask &= AMDGPU_RAS_INST_MASK; + info->sub_block_index |= dev_mask; + ras_cmd = (struct ta_ras_shared_memory *)psp->ras_context.context.mem_context.shared_buf; memset(ras_cmd, 0, sizeof(struct ta_ras_shared_memory)); @@ -2077,10 +2169,12 @@ static int psp_hw_start(struct psp_context *psp) if (amdgpu_sriov_vf(adev) && amdgpu_in_reset(adev)) goto skip_pin_bo; - ret = psp_tmr_init(psp); - if (ret) { - DRM_ERROR("PSP tmr init failed!\n"); - return ret; + if (!psp_boottime_tmr(psp)) { + ret = psp_tmr_init(psp); + if (ret) { + DRM_ERROR("PSP tmr init failed!\n"); + return ret; + } } skip_pin_bo: @@ -2363,7 +2457,7 @@ static int psp_prep_load_ip_fw_cmd_buf(struct amdgpu_firmware_info *ucode, } static int psp_execute_non_psp_fw_load(struct psp_context *psp, - struct amdgpu_firmware_info *ucode) + struct amdgpu_firmware_info *ucode) { int ret = 0; struct psp_gfx_cmd_resp *cmd = acquire_psp_cmd_buf(psp); @@ -2402,9 +2496,8 @@ static int psp_load_smu_fw(struct psp_context *psp) (adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 4) || adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 2)))) { ret = amdgpu_dpm_set_mp1_state(adev, PP_MP1_STATE_UNLOAD); - if (ret) { + if (ret) DRM_WARN("Failed to set MP1 state prepare for reload\n"); - } } ret = psp_execute_non_psp_fw_load(psp, ucode); @@ -2655,8 +2748,6 @@ static int psp_hw_fini(void *handle) psp_ring_destroy(psp, PSP_RING_TYPE__KM); - psp_free_shared_bufs(psp); - return 0; } @@ -2716,9 +2807,8 @@ static int psp_suspend(void *handle) } ret = psp_ring_stop(psp, PSP_RING_TYPE__KM); - if (ret) { + if (ret) DRM_ERROR("PSP ring stop failed\n"); - } out: return ret; @@ -2967,7 +3057,7 @@ static int parse_sos_bin_descriptor(struct psp_context *psp, psp->sos.fw_version = le32_to_cpu(desc->fw_version); psp->sos.feature_version = le32_to_cpu(desc->fw_version); psp->sos.size_bytes = le32_to_cpu(desc->size_bytes); - psp->sos.start_addr = ucode_start_addr; + psp->sos.start_addr = ucode_start_addr; break; case PSP_FW_TYPE_PSP_SYS_DRV: psp->sys.fw_version = le32_to_cpu(desc->fw_version); @@ -3491,7 +3581,7 @@ void psp_copy_fw(struct psp_context *psp, uint8_t *start_addr, uint32_t bin_size drm_dev_exit(idx); } -static DEVICE_ATTR(usbc_pd_fw, S_IRUGO | S_IWUSR, +static DEVICE_ATTR(usbc_pd_fw, 0644, psp_usbc_pd_fw_sysfs_read, psp_usbc_pd_fw_sysfs_write); @@ -3548,6 +3638,9 @@ static ssize_t amdgpu_psp_vbflash_read(struct file *filp, struct kobject *kobj, void *fw_pri_cpu_addr; int ret; + if (adev->psp.vbflash_image_size == 0) + return -EINVAL; + dev_info(adev->dev, "VBIOS flash to PSP started"); ret = amdgpu_bo_create_kernel(adev, adev->psp.vbflash_image_size, @@ -3599,13 +3692,13 @@ static ssize_t amdgpu_psp_vbflash_status(struct device *dev, } static const struct bin_attribute psp_vbflash_bin_attr = { - .attr = {.name = "psp_vbflash", .mode = 0664}, + .attr = {.name = "psp_vbflash", .mode = 0660}, .size = 0, .write = amdgpu_psp_vbflash_write, .read = amdgpu_psp_vbflash_read, }; -static DEVICE_ATTR(psp_vbflash_status, 0444, amdgpu_psp_vbflash_status, NULL); +static DEVICE_ATTR(psp_vbflash_status, 0440, amdgpu_psp_vbflash_status, NULL); int amdgpu_psp_sysfs_init(struct amdgpu_device *adev) { @@ -3618,6 +3711,7 @@ int amdgpu_psp_sysfs_init(struct amdgpu_device *adev) switch (adev->ip_versions[MP0_HWIP][0]) { case IP_VERSION(13, 0, 0): case IP_VERSION(13, 0, 7): + case IP_VERSION(13, 0, 10): if (!psp->adev) { psp->adev = adev; psp_v13_0_set_psp_funcs(psp); @@ -3673,8 +3767,7 @@ static void psp_sysfs_fini(struct amdgpu_device *adev) device_remove_file(adev->dev, &dev_attr_usbc_pd_fw); } -const struct amdgpu_ip_block_version psp_v3_1_ip_block = -{ +const struct amdgpu_ip_block_version psp_v3_1_ip_block = { .type = AMD_IP_BLOCK_TYPE_PSP, .major = 3, .minor = 1, @@ -3682,8 +3775,7 @@ const struct amdgpu_ip_block_version psp_v3_1_ip_block = .funcs = &psp_ip_funcs, }; -const struct amdgpu_ip_block_version psp_v10_0_ip_block = -{ +const struct amdgpu_ip_block_version psp_v10_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_PSP, .major = 10, .minor = 0, @@ -3691,8 +3783,7 @@ const struct amdgpu_ip_block_version psp_v10_0_ip_block = .funcs = &psp_ip_funcs, }; -const struct amdgpu_ip_block_version psp_v11_0_ip_block = -{ +const struct amdgpu_ip_block_version psp_v11_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_PSP, .major = 11, .minor = 0, @@ -3708,8 +3799,7 @@ const struct amdgpu_ip_block_version psp_v11_0_8_ip_block = { .funcs = &psp_ip_funcs, }; -const struct amdgpu_ip_block_version psp_v12_0_ip_block = -{ +const struct amdgpu_ip_block_version psp_v12_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_PSP, .major = 12, .minor = 0, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h index cf4f60c661223c7562f73c05e2086b101aab17ec..2cae0b1a0b8ac720d2c84d7483be426474be5f1a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h @@ -455,6 +455,8 @@ extern const struct amdgpu_ip_block_version psp_v13_0_4_ip_block; extern int psp_wait_for(struct psp_context *psp, uint32_t reg_index, uint32_t field_val, uint32_t mask, bool check_changed); +extern int psp_wait_for_spirom_update(struct psp_context *psp, uint32_t reg_index, + uint32_t field_val, uint32_t mask, uint32_t msec_timeout); int psp_gpu_reset(struct amdgpu_device *adev); int psp_update_vcn_sram(struct amdgpu_device *adev, int inst_idx, @@ -486,7 +488,7 @@ int psp_ras_invoke(struct psp_context *psp, uint32_t ta_cmd_id); int psp_ras_enable_features(struct psp_context *psp, union ta_ras_cmd_input *info, bool enable); int psp_ras_trigger_error(struct psp_context *psp, - struct ta_ras_trigger_error_input *info); + struct ta_ras_trigger_error_input *info, uint32_t instance_mask); int psp_ras_terminate(struct psp_context *psp); int psp_hdcp_invoke(struct psp_context *psp, uint32_t ta_cmd_id); @@ -519,6 +521,8 @@ int psp_load_fw_list(struct psp_context *psp, struct amdgpu_firmware_info **ucode_list, int ucode_count); void psp_copy_fw(struct psp_context *psp, uint8_t *start_addr, uint32_t bin_size); +int psp_spatial_partition(struct psp_context *psp, int mode); + int is_psp_fw_valid(struct psp_bin_desc bin); int amdgpu_psp_sysfs_init(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 3ab8a88789c8fec5f82c5772e1e68b85ad6e0eda..4769a18304d7a4435403ffce859902b25e70f816 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -171,8 +171,7 @@ static int amdgpu_reserve_page_direct(struct amdgpu_device *adev, uint64_t addre memset(&err_rec, 0x0, sizeof(struct eeprom_table_record)); err_data.err_addr = &err_rec; - amdgpu_umc_fill_error_record(&err_data, address, - (address >> AMDGPU_GPU_PAGE_SHIFT), 0, 0); + amdgpu_umc_fill_error_record(&err_data, address, address, 0, 0); if (amdgpu_bad_page_threshold != 0) { amdgpu_ras_add_bad_pages(adev, err_data.err_addr, @@ -256,6 +255,8 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f, int block_id; uint32_t sub_block; u64 address, value; + /* default value is 0 if the mask is not set by user */ + u32 instance_mask = 0; if (*pos) return -EINVAL; @@ -306,7 +307,11 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f, data->op = op; if (op == 2) { - if (sscanf(str, "%*s %*s %*s 0x%x 0x%llx 0x%llx", + if (sscanf(str, "%*s %*s %*s 0x%x 0x%llx 0x%llx 0x%x", + &sub_block, &address, &value, &instance_mask) != 4 && + sscanf(str, "%*s %*s %*s %u %llu %llu %u", + &sub_block, &address, &value, &instance_mask) != 4 && + sscanf(str, "%*s %*s %*s 0x%x 0x%llx 0x%llx", &sub_block, &address, &value) != 3 && sscanf(str, "%*s %*s %*s %u %llu %llu", &sub_block, &address, &value) != 3) @@ -314,6 +319,7 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f, data->head.sub_block_index = sub_block; data->inject.address = address; data->inject.value = value; + data->inject.instance_mask = instance_mask; } } else { if (size < sizeof(*data)) @@ -326,6 +332,46 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f, return 0; } +static void amdgpu_ras_instance_mask_check(struct amdgpu_device *adev, + struct ras_debug_if *data) +{ + int num_xcc = adev->gfx.xcc_mask ? NUM_XCC(adev->gfx.xcc_mask) : 1; + uint32_t mask, inst_mask = data->inject.instance_mask; + + /* no need to set instance mask if there is only one instance */ + if (num_xcc <= 1 && inst_mask) { + data->inject.instance_mask = 0; + dev_dbg(adev->dev, + "RAS inject mask(0x%x) isn't supported and force it to 0.\n", + inst_mask); + + return; + } + + switch (data->head.block) { + case AMDGPU_RAS_BLOCK__GFX: + mask = GENMASK(num_xcc - 1, 0); + break; + case AMDGPU_RAS_BLOCK__SDMA: + mask = GENMASK(adev->sdma.num_instances - 1, 0); + break; + case AMDGPU_RAS_BLOCK__VCN: + case AMDGPU_RAS_BLOCK__JPEG: + mask = GENMASK(adev->vcn.num_vcn_inst - 1, 0); + break; + default: + mask = inst_mask; + break; + } + + /* remove invalid bits in instance mask */ + data->inject.instance_mask &= mask; + if (inst_mask != data->inject.instance_mask) + dev_dbg(adev->dev, + "Adjust RAS inject mask 0x%x to 0x%x\n", + inst_mask, data->inject.instance_mask); +} + /** * DOC: AMDGPU RAS debugfs control interface * @@ -341,7 +387,7 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f, * sub_block_index: some IPs have subcomponets. say, GFX, sDMA. * name: the name of IP. * - * inject has two more members than head, they are address, value. + * inject has three more members than head, they are address, value and mask. * As their names indicate, inject operation will write the * value to the address. * @@ -365,7 +411,7 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f, * * echo "disable " > /sys/kernel/debug/dri//ras/ras_ctrl * echo "enable " > /sys/kernel/debug/dri//ras/ras_ctrl - * echo "inject
> /sys/kernel/debug/dri//ras/ras_ctrl + * echo "inject
" > /sys/kernel/debug/dri//ras/ras_ctrl * * Where N, is the card which you want to affect. * @@ -382,13 +428,14 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f, * * The sub-block is a the sub-block index, pass 0 if there is no sub-block. * The address and value are hexadecimal numbers, leading 0x is optional. + * The mask means instance mask, is optional, default value is 0x1. * * For instance, * * .. code-block:: bash * * echo inject umc ue 0x0 0x0 0x0 > /sys/kernel/debug/dri/0/ras/ras_ctrl - * echo inject umc ce 0 0 0 > /sys/kernel/debug/dri/0/ras/ras_ctrl + * echo inject umc ce 0 0 0 3 > /sys/kernel/debug/dri/0/ras/ras_ctrl * echo disable umc > /sys/kernel/debug/dri/0/ras/ras_ctrl * * How to check the result of the operation? @@ -442,7 +489,8 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, ret = amdgpu_ras_feature_enable(adev, &data.head, 1); break; case 2: - if ((data.inject.address >= adev->gmc.mc_vram_size) || + if ((data.inject.address >= adev->gmc.mc_vram_size && + adev->gmc.mc_vram_size) || (data.inject.address >= RAS_UMC_INJECT_ADDR_LIMIT)) { dev_warn(adev->dev, "RAS WARN: input address " "0x%llx is invalid.", @@ -460,6 +508,8 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, break; } + amdgpu_ras_instance_mask_check(adev, &data); + /* data.inject.address is offset instead of absolute gpu address */ ret = amdgpu_ras_error_inject(adev, &data.inject); break; @@ -1115,15 +1165,15 @@ int amdgpu_ras_error_inject(struct amdgpu_device *adev, block_info.address); } - if (info->head.block == AMDGPU_RAS_BLOCK__GFX) { - if (block_obj->hw_ops->ras_error_inject) - ret = block_obj->hw_ops->ras_error_inject(adev, info); + if (block_obj->hw_ops->ras_error_inject) { + if (info->head.block == AMDGPU_RAS_BLOCK__GFX) + ret = block_obj->hw_ops->ras_error_inject(adev, info, info->instance_mask); + else /* Special ras_error_inject is defined (e.g: xgmi) */ + ret = block_obj->hw_ops->ras_error_inject(adev, &block_info, + info->instance_mask); } else { - /* If defined special ras_error_inject(e.g: xgmi), implement special ras_error_inject */ - if (block_obj->hw_ops->ras_error_inject) - ret = block_obj->hw_ops->ras_error_inject(adev, &block_info); - else /*If not defined .ras_error_inject, use default ras_error_inject*/ - ret = psp_ras_trigger_error(&adev->psp, &block_info); + /* default path */ + ret = psp_ras_trigger_error(&adev->psp, &block_info, info->instance_mask); } if (ret) @@ -1441,6 +1491,7 @@ static int amdgpu_ras_sysfs_remove_all(struct amdgpu_device *adev) static struct dentry *amdgpu_ras_debugfs_create_ctrl_node(struct amdgpu_device *adev) { struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + struct amdgpu_ras_eeprom_control *eeprom = &con->eeprom_control; struct drm_minor *minor = adev_to_drm(adev)->primary; struct dentry *dir; @@ -1451,6 +1502,7 @@ static struct dentry *amdgpu_ras_debugfs_create_ctrl_node(struct amdgpu_device * &amdgpu_ras_debugfs_eeprom_ops); debugfs_create_u32("bad_page_cnt_threshold", 0444, dir, &con->bad_page_cnt_threshold); + debugfs_create_u32("ras_num_recs", 0444, dir, &eeprom->ras_num_recs); debugfs_create_x32("ras_hw_enabled", 0444, dir, &adev->ras_hw_enabled); debugfs_create_x32("ras_enabled", 0444, dir, &adev->ras_enabled); debugfs_create_file("ras_eeprom_size", S_IRUGO, dir, adev, @@ -1597,8 +1649,7 @@ static int amdgpu_ras_fs_fini(struct amdgpu_device *adev) void amdgpu_ras_interrupt_fatal_error_handler(struct amdgpu_device *adev) { /* Fatal error events are handled on host side */ - if (amdgpu_sriov_vf(adev) || - !amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__PCIE_BIF)) + if (amdgpu_sriov_vf(adev)) return; if (adev->nbio.ras && @@ -1636,8 +1687,7 @@ static void amdgpu_ras_interrupt_poison_consumption_handler(struct ras_manager * } } - if (!adev->gmc.xgmi.connected_to_cpu) - amdgpu_umc_poison_handler(adev, false); + amdgpu_umc_poison_handler(adev, false); if (block_obj->hw_ops && block_obj->hw_ops->handle_poison_consumption) poison_stat = block_obj->hw_ops->handle_poison_consumption(adev); @@ -2008,9 +2058,15 @@ static void amdgpu_ras_do_recovery(struct work_struct *work) /* Perform full reset in fatal error mode */ if (!amdgpu_ras_is_poison_mode_supported(ras->adev)) set_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); - else + else { clear_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); + if (ras->gpu_reset_flags & AMDGPU_RAS_GPU_RESET_MODE2_RESET) { + ras->gpu_reset_flags &= ~AMDGPU_RAS_GPU_RESET_MODE2_RESET; + reset_context.method = AMD_RESET_METHOD_MODE2; + } + } + amdgpu_device_gpu_recover(ras->adev, NULL, &reset_context); } atomic_set(&ras->in_recovery, 0); @@ -2259,7 +2315,7 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev) atomic_set(&con->in_recovery, 0); con->eeprom_control.bad_channel_bitmap = 0; - max_eeprom_records_count = amdgpu_ras_eeprom_max_record_count(); + max_eeprom_records_count = amdgpu_ras_eeprom_max_record_count(&con->eeprom_control); amdgpu_ras_validate_threshold(adev, max_eeprom_records_count); /* Todo: During test the SMU might fail to read the eeprom through I2C @@ -2396,11 +2452,10 @@ static void amdgpu_ras_check_supported(struct amdgpu_device *adev) { adev->ras_hw_enabled = adev->ras_enabled = 0; - if (!adev->is_atom_fw || - !amdgpu_ras_asic_supported(adev)) + if (!amdgpu_ras_asic_supported(adev)) return; - if (!adev->gmc.xgmi.connected_to_cpu) { + if (!adev->gmc.xgmi.connected_to_cpu && !adev->gmc.is_app_apu) { if (amdgpu_atomfirmware_mem_ecc_supported(adev)) { dev_info(adev->dev, "MEM ECC is active.\n"); adev->ras_hw_enabled |= (1 << AMDGPU_RAS_BLOCK__UMC | @@ -2625,7 +2680,8 @@ int amdgpu_ras_init(struct amdgpu_device *adev) int amdgpu_persistent_edc_harvesting_supported(struct amdgpu_device *adev) { - if (adev->gmc.xgmi.connected_to_cpu) + if (adev->gmc.xgmi.connected_to_cpu || + adev->gmc.is_app_apu) return 1; return 0; } @@ -3104,3 +3160,143 @@ int amdgpu_ras_register_ras_block(struct amdgpu_device *adev, return 0; } + +void amdgpu_ras_get_error_type_name(uint32_t err_type, char *err_type_name) +{ + if (!err_type_name) + return; + + switch (err_type) { + case AMDGPU_RAS_ERROR__SINGLE_CORRECTABLE: + sprintf(err_type_name, "correctable"); + break; + case AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE: + sprintf(err_type_name, "uncorrectable"); + break; + default: + sprintf(err_type_name, "unknown"); + break; + } +} + +bool amdgpu_ras_inst_get_memory_id_field(struct amdgpu_device *adev, + const struct amdgpu_ras_err_status_reg_entry *reg_entry, + uint32_t instance, + uint32_t *memory_id) +{ + uint32_t err_status_lo_data, err_status_lo_offset; + + if (!reg_entry) + return false; + + err_status_lo_offset = + AMDGPU_RAS_REG_ENTRY_OFFSET(reg_entry->hwip, instance, + reg_entry->seg_lo, reg_entry->reg_lo); + err_status_lo_data = RREG32(err_status_lo_offset); + + if ((reg_entry->flags & AMDGPU_RAS_ERR_STATUS_VALID) && + !REG_GET_FIELD(err_status_lo_data, ERR_STATUS_LO, ERR_STATUS_VALID_FLAG)) + return false; + + *memory_id = REG_GET_FIELD(err_status_lo_data, ERR_STATUS_LO, MEMORY_ID); + + return true; +} + +bool amdgpu_ras_inst_get_err_cnt_field(struct amdgpu_device *adev, + const struct amdgpu_ras_err_status_reg_entry *reg_entry, + uint32_t instance, + unsigned long *err_cnt) +{ + uint32_t err_status_hi_data, err_status_hi_offset; + + if (!reg_entry) + return false; + + err_status_hi_offset = + AMDGPU_RAS_REG_ENTRY_OFFSET(reg_entry->hwip, instance, + reg_entry->seg_hi, reg_entry->reg_hi); + err_status_hi_data = RREG32(err_status_hi_offset); + + if ((reg_entry->flags & AMDGPU_RAS_ERR_INFO_VALID) && + !REG_GET_FIELD(err_status_hi_data, ERR_STATUS_HI, ERR_INFO_VALID_FLAG)) + /* keep the check here in case we need to refer to the result later */ + dev_dbg(adev->dev, "Invalid err_info field\n"); + + /* read err count */ + *err_cnt = REG_GET_FIELD(err_status_hi_data, ERR_STATUS, ERR_CNT); + + return true; +} + +void amdgpu_ras_inst_query_ras_error_count(struct amdgpu_device *adev, + const struct amdgpu_ras_err_status_reg_entry *reg_list, + uint32_t reg_list_size, + const struct amdgpu_ras_memory_id_entry *mem_list, + uint32_t mem_list_size, + uint32_t instance, + uint32_t err_type, + unsigned long *err_count) +{ + uint32_t memory_id; + unsigned long err_cnt; + char err_type_name[16]; + uint32_t i, j; + + for (i = 0; i < reg_list_size; i++) { + /* query memory_id from err_status_lo */ + if (!amdgpu_ras_inst_get_memory_id_field(adev, ®_list[i], + instance, &memory_id)) + continue; + + /* query err_cnt from err_status_hi */ + if (!amdgpu_ras_inst_get_err_cnt_field(adev, ®_list[i], + instance, &err_cnt) || + !err_cnt) + continue; + + *err_count += err_cnt; + + /* log the errors */ + amdgpu_ras_get_error_type_name(err_type, err_type_name); + if (!mem_list) { + /* memory_list is not supported */ + dev_info(adev->dev, + "%ld %s hardware errors detected in %s, instance: %d, memory_id: %d\n", + err_cnt, err_type_name, + reg_list[i].block_name, + instance, memory_id); + } else { + for (j = 0; j < mem_list_size; j++) { + if (memory_id == mem_list[j].memory_id) { + dev_info(adev->dev, + "%ld %s hardware errors detected in %s, instance: %d, memory block: %s\n", + err_cnt, err_type_name, + reg_list[i].block_name, + instance, mem_list[j].name); + break; + } + } + } + } +} + +void amdgpu_ras_inst_reset_ras_error_count(struct amdgpu_device *adev, + const struct amdgpu_ras_err_status_reg_entry *reg_list, + uint32_t reg_list_size, + uint32_t instance) +{ + uint32_t err_status_lo_offset, err_status_hi_offset; + uint32_t i; + + for (i = 0; i < reg_list_size; i++) { + err_status_lo_offset = + AMDGPU_RAS_REG_ENTRY_OFFSET(reg_list[i].hwip, instance, + reg_list[i].seg_lo, reg_list[i].reg_lo); + err_status_hi_offset = + AMDGPU_RAS_REG_ENTRY_OFFSET(reg_list[i].hwip, instance, + reg_list[i].seg_hi, reg_list[i].reg_hi); + WREG32(err_status_lo_offset, 0); + WREG32(err_status_hi_offset, 0); + } +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h index 17b3d1992e80879e7c8767fc8993b19814023847..46bf1889a9d753defcd5c8acace7b047bd4f98cf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h @@ -32,6 +32,11 @@ struct amdgpu_iv_entry; #define AMDGPU_RAS_FLAG_INIT_BY_VBIOS (0x1 << 0) +/* position of instance value in sub_block_index of + * ta_ras_trigger_error_input, the sub block uses lower 12 bits + */ +#define AMDGPU_RAS_INST_MASK 0xfffff000 +#define AMDGPU_RAS_INST_SHIFT 0xc enum amdgpu_ras_block { AMDGPU_RAS_BLOCK__UMC = 0, @@ -314,6 +319,45 @@ enum amdgpu_ras_ret { AMDGPU_RAS_PT, }; +/* ras error status reisger fields */ +#define ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +#define ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define ERR_STATUS__ERR_CNT__SHIFT 0x17 +#define ERR_STATUS__ERR_CNT_MASK 0x03800000L + +#define AMDGPU_RAS_REG_ENTRY(ip, inst, reg_lo, reg_hi) \ + ip##_HWIP, inst, reg_lo##_BASE_IDX, reg_lo, reg_hi##_BASE_IDX, reg_hi + +#define AMDGPU_RAS_REG_ENTRY_OFFSET(hwip, ip_inst, segment, reg) \ + (adev->reg_offset[hwip][ip_inst][segment] + (reg)) + +#define AMDGPU_RAS_ERR_INFO_VALID (1 << 0) +#define AMDGPU_RAS_ERR_STATUS_VALID (1 << 1) +#define AMDGPU_RAS_ERR_ADDRESS_VALID (1 << 2) + +#define AMDGPU_RAS_GPU_RESET_MODE2_RESET (0x1 << 0) + +struct amdgpu_ras_err_status_reg_entry { + uint32_t hwip; + uint32_t ip_inst; + uint32_t seg_lo; + uint32_t reg_lo; + uint32_t seg_hi; + uint32_t reg_hi; + uint32_t reg_inst; + uint32_t flags; + const char *block_name; +}; + +struct amdgpu_ras_memory_id_entry { + uint32_t memory_id; + const char *name; +}; + struct ras_common_if { enum amdgpu_ras_block block; enum amdgpu_ras_error_type type; @@ -385,6 +429,9 @@ struct amdgpu_ras { /* Indicates smu whether need update bad channel info */ bool update_channel_flag; + + /* Record special requirements of gpu reset caller */ + uint32_t gpu_reset_flags; }; struct ras_fs_data { @@ -471,6 +518,7 @@ struct ras_inject_if { struct ras_common_if head; uint64_t address; uint64_t value; + uint32_t instance_mask; }; struct ras_cure_if { @@ -508,7 +556,8 @@ struct amdgpu_ras_block_object { }; struct amdgpu_ras_block_hw_ops { - int (*ras_error_inject)(struct amdgpu_device *adev, void *inject_if); + int (*ras_error_inject)(struct amdgpu_device *adev, + void *inject_if, uint32_t instance_mask); void (*query_ras_error_count)(struct amdgpu_device *adev, void *ras_error_status); void (*query_ras_error_status)(struct amdgpu_device *adev); void (*query_ras_error_address)(struct amdgpu_device *adev, void *ras_error_status); @@ -696,4 +745,25 @@ int amdgpu_ras_set_context(struct amdgpu_device *adev, struct amdgpu_ras *ras_co int amdgpu_ras_register_ras_block(struct amdgpu_device *adev, struct amdgpu_ras_block_object *ras_block_obj); void amdgpu_ras_interrupt_fatal_error_handler(struct amdgpu_device *adev); +void amdgpu_ras_get_error_type_name(uint32_t err_type, char *err_type_name); +bool amdgpu_ras_inst_get_memory_id_field(struct amdgpu_device *adev, + const struct amdgpu_ras_err_status_reg_entry *reg_entry, + uint32_t instance, + uint32_t *memory_id); +bool amdgpu_ras_inst_get_err_cnt_field(struct amdgpu_device *adev, + const struct amdgpu_ras_err_status_reg_entry *reg_entry, + uint32_t instance, + unsigned long *err_cnt); +void amdgpu_ras_inst_query_ras_error_count(struct amdgpu_device *adev, + const struct amdgpu_ras_err_status_reg_entry *reg_list, + uint32_t reg_list_size, + const struct amdgpu_ras_memory_id_entry *mem_list, + uint32_t mem_list_size, + uint32_t instance, + uint32_t err_type, + unsigned long *err_count); +void amdgpu_ras_inst_reset_ras_error_count(struct amdgpu_device *adev, + const struct amdgpu_ras_err_status_reg_entry *reg_list, + uint32_t reg_list_size, + uint32_t instance); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c index c2c2a7718613c5a92024c084b963f228b7d8b191..0648dfe559afcffb738404a1d111a34bd20959d3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c @@ -68,11 +68,24 @@ /* Table hdr is 'AMDR' */ #define RAS_TABLE_HDR_VAL 0x414d4452 -#define RAS_TABLE_VER 0x00010000 /* Bad GPU tag ‘BADG’ */ #define RAS_TABLE_HDR_BAD 0x42414447 +/* + * EEPROM Table structure v1 + * --------------------------------- + * | | + * | EEPROM TABLE HEADER | + * | ( size 20 Bytes ) | + * | | + * --------------------------------- + * | | + * | BAD PAGE RECORD AREA | + * | | + * --------------------------------- + */ + /* Assume 2-Mbit size EEPROM and take up the whole space. */ #define RAS_TBL_SIZE_BYTES (256 * 1024) #define RAS_TABLE_START 0 @@ -81,6 +94,35 @@ #define RAS_MAX_RECORD_COUNT ((RAS_TBL_SIZE_BYTES - RAS_TABLE_HEADER_SIZE) \ / RAS_TABLE_RECORD_SIZE) +/* + * EEPROM Table structrue v2.1 + * --------------------------------- + * | | + * | EEPROM TABLE HEADER | + * | ( size 20 Bytes ) | + * | | + * --------------------------------- + * | | + * | EEPROM TABLE RAS INFO | + * | (available info size 4 Bytes) | + * | ( reserved size 252 Bytes ) | + * | | + * --------------------------------- + * | | + * | BAD PAGE RECORD AREA | + * | | + * --------------------------------- + */ + +/* EEPROM Table V2_1 */ +#define RAS_TABLE_V2_1_INFO_SIZE 256 +#define RAS_TABLE_V2_1_INFO_START RAS_TABLE_HEADER_SIZE +#define RAS_RECORD_START_V2_1 (RAS_HDR_START + RAS_TABLE_HEADER_SIZE + \ + RAS_TABLE_V2_1_INFO_SIZE) +#define RAS_MAX_RECORD_COUNT_V2_1 ((RAS_TBL_SIZE_BYTES - RAS_TABLE_HEADER_SIZE - \ + RAS_TABLE_V2_1_INFO_SIZE) \ + / RAS_TABLE_RECORD_SIZE) + /* Given a zero-based index of an EEPROM RAS record, yields the EEPROM * offset off of RAS_TABLE_START. That is, this is something you can * add to control->i2c_address, and then tell I2C layer to read @@ -103,6 +145,10 @@ #define RAS_NUM_RECS(_tbl_hdr) (((_tbl_hdr)->tbl_size - \ RAS_TABLE_HEADER_SIZE) / RAS_TABLE_RECORD_SIZE) +#define RAS_NUM_RECS_V2_1(_tbl_hdr) (((_tbl_hdr)->tbl_size - \ + RAS_TABLE_HEADER_SIZE - \ + RAS_TABLE_V2_1_INFO_SIZE) / RAS_TABLE_RECORD_SIZE) + #define to_amdgpu_device(x) (container_of(x, struct amdgpu_ras, eeprom_control))->adev static bool __is_ras_eeprom_supported(struct amdgpu_device *adev) @@ -230,6 +276,69 @@ static int __write_table_header(struct amdgpu_ras_eeprom_control *control) return res; } +static void +__encode_table_ras_info_to_buf(struct amdgpu_ras_eeprom_table_ras_info *rai, + unsigned char *buf) +{ + u32 *pp = (uint32_t *)buf; + u32 tmp; + + tmp = ((uint32_t)(rai->rma_status) & 0xFF) | + (((uint32_t)(rai->health_percent) << 8) & 0xFF00) | + (((uint32_t)(rai->ecc_page_threshold) << 16) & 0xFFFF0000); + pp[0] = cpu_to_le32(tmp); +} + +static void +__decode_table_ras_info_from_buf(struct amdgpu_ras_eeprom_table_ras_info *rai, + unsigned char *buf) +{ + u32 *pp = (uint32_t *)buf; + u32 tmp; + + tmp = le32_to_cpu(pp[0]); + rai->rma_status = tmp & 0xFF; + rai->health_percent = (tmp >> 8) & 0xFF; + rai->ecc_page_threshold = (tmp >> 16) & 0xFFFF; +} + +static int __write_table_ras_info(struct amdgpu_ras_eeprom_control *control) +{ + struct amdgpu_device *adev = to_amdgpu_device(control); + u8 *buf; + int res; + + buf = kzalloc(RAS_TABLE_V2_1_INFO_SIZE, GFP_KERNEL); + if (!buf) { + DRM_ERROR("Failed to alloc buf to write table ras info\n"); + return -ENOMEM; + } + + __encode_table_ras_info_to_buf(&control->tbl_rai, buf); + + /* i2c may be unstable in gpu reset */ + down_read(&adev->reset_domain->sem); + res = amdgpu_eeprom_write(adev->pm.ras_eeprom_i2c_bus, + control->i2c_address + + control->ras_info_offset, + buf, RAS_TABLE_V2_1_INFO_SIZE); + up_read(&adev->reset_domain->sem); + + if (res < 0) { + DRM_ERROR("Failed to write EEPROM table ras info:%d", res); + } else if (res < RAS_TABLE_V2_1_INFO_SIZE) { + DRM_ERROR("Short write:%d out of %d\n", + res, RAS_TABLE_V2_1_INFO_SIZE); + res = -EIO; + } else { + res = 0; + } + + kfree(buf); + + return res; +} + static u8 __calc_hdr_byte_sum(const struct amdgpu_ras_eeprom_control *control) { int ii; @@ -246,6 +355,21 @@ static u8 __calc_hdr_byte_sum(const struct amdgpu_ras_eeprom_control *control) return csum; } +static u8 __calc_ras_info_byte_sum(const struct amdgpu_ras_eeprom_control *control) +{ + int ii; + u8 *pp, csum; + size_t sz; + + sz = sizeof(control->tbl_rai); + pp = (u8 *) &control->tbl_rai; + csum = 0; + for (ii = 0; ii < sz; ii++, pp++) + csum += *pp; + + return csum; +} + static int amdgpu_ras_eeprom_correct_header_tag( struct amdgpu_ras_eeprom_control *control, uint32_t header) @@ -282,6 +406,7 @@ int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control) { struct amdgpu_device *adev = to_amdgpu_device(control); struct amdgpu_ras_eeprom_table_header *hdr = &control->tbl_hdr; + struct amdgpu_ras_eeprom_table_ras_info *rai = &control->tbl_rai; struct amdgpu_ras *con = amdgpu_ras_get_context(adev); u8 csum; int res; @@ -289,14 +414,37 @@ int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control) mutex_lock(&control->ras_tbl_mutex); hdr->header = RAS_TABLE_HDR_VAL; - hdr->version = RAS_TABLE_VER; - hdr->first_rec_offset = RAS_RECORD_START; - hdr->tbl_size = RAS_TABLE_HEADER_SIZE; + if (adev->umc.ras && + adev->umc.ras->set_eeprom_table_version) + adev->umc.ras->set_eeprom_table_version(hdr); + else + hdr->version = RAS_TABLE_VER_V1; + + if (hdr->version == RAS_TABLE_VER_V2_1) { + hdr->first_rec_offset = RAS_RECORD_START_V2_1; + hdr->tbl_size = RAS_TABLE_HEADER_SIZE + + RAS_TABLE_V2_1_INFO_SIZE; + rai->rma_status = GPU_HEALTH_USABLE; + /** + * GPU health represented as a percentage. + * 0 means worst health, 100 means fully health. + */ + rai->health_percent = 100; + /* ecc_page_threshold = 0 means disable bad page retirement */ + rai->ecc_page_threshold = con->bad_page_cnt_threshold; + } else { + hdr->first_rec_offset = RAS_RECORD_START; + hdr->tbl_size = RAS_TABLE_HEADER_SIZE; + } csum = __calc_hdr_byte_sum(control); + if (hdr->version == RAS_TABLE_VER_V2_1) + csum += __calc_ras_info_byte_sum(control); csum = -csum; hdr->checksum = csum; res = __write_table_header(control); + if (!res && hdr->version > RAS_TABLE_VER_V1) + res = __write_table_ras_info(control); control->ras_num_recs = 0; control->ras_fri = 0; @@ -573,11 +721,19 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control) "Saved bad pages %d reaches threshold value %d\n", control->ras_num_recs, ras->bad_page_cnt_threshold); control->tbl_hdr.header = RAS_TABLE_HDR_BAD; + if (control->tbl_hdr.version == RAS_TABLE_VER_V2_1) { + control->tbl_rai.rma_status = GPU_RETIRED__ECC_REACH_THRESHOLD; + control->tbl_rai.health_percent = 0; + } } - control->tbl_hdr.version = RAS_TABLE_VER; - control->tbl_hdr.first_rec_offset = RAS_INDEX_TO_OFFSET(control, control->ras_fri); - control->tbl_hdr.tbl_size = RAS_TABLE_HEADER_SIZE + control->ras_num_recs * RAS_TABLE_RECORD_SIZE; + if (control->tbl_hdr.version == RAS_TABLE_VER_V2_1) + control->tbl_hdr.tbl_size = RAS_TABLE_HEADER_SIZE + + RAS_TABLE_V2_1_INFO_SIZE + + control->ras_num_recs * RAS_TABLE_RECORD_SIZE; + else + control->tbl_hdr.tbl_size = RAS_TABLE_HEADER_SIZE + + control->ras_num_recs * RAS_TABLE_RECORD_SIZE; control->tbl_hdr.checksum = 0; buf_size = control->ras_num_recs * RAS_TABLE_RECORD_SIZE; @@ -606,6 +762,17 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control) goto Out; } + /** + * bad page records have been stored in eeprom, + * now calculate gpu health percent + */ + if (amdgpu_bad_page_threshold != 0 && + control->tbl_hdr.version == RAS_TABLE_VER_V2_1 && + control->ras_num_recs < ras->bad_page_cnt_threshold) + control->tbl_rai.health_percent = ((ras->bad_page_cnt_threshold - + control->ras_num_recs) * 100) / + ras->bad_page_cnt_threshold; + /* Recalc the checksum. */ csum = 0; @@ -613,10 +780,14 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control) csum += *pp; csum += __calc_hdr_byte_sum(control); + if (control->tbl_hdr.version == RAS_TABLE_VER_V2_1) + csum += __calc_ras_info_byte_sum(control); /* avoid sign extension when assigning to "checksum" */ csum = -csum; control->tbl_hdr.checksum = csum; res = __write_table_header(control); + if (!res && control->tbl_hdr.version > RAS_TABLE_VER_V1) + res = __write_table_ras_info(control); Out: kfree(buf); return res; @@ -807,9 +978,12 @@ int amdgpu_ras_eeprom_read(struct amdgpu_ras_eeprom_control *control, return res; } -uint32_t amdgpu_ras_eeprom_max_record_count(void) +uint32_t amdgpu_ras_eeprom_max_record_count(struct amdgpu_ras_eeprom_control *control) { - return RAS_MAX_RECORD_COUNT; + if (control->tbl_hdr.version == RAS_TABLE_VER_V2_1) + return RAS_MAX_RECORD_COUNT_V2_1; + else + return RAS_MAX_RECORD_COUNT; } static ssize_t @@ -1051,8 +1225,14 @@ static int __verify_ras_table_checksum(struct amdgpu_ras_eeprom_control *control int buf_size, res; u8 csum, *buf, *pp; - buf_size = RAS_TABLE_HEADER_SIZE + - control->ras_num_recs * RAS_TABLE_RECORD_SIZE; + if (control->tbl_hdr.version == RAS_TABLE_VER_V2_1) + buf_size = RAS_TABLE_HEADER_SIZE + + RAS_TABLE_V2_1_INFO_SIZE + + control->ras_num_recs * RAS_TABLE_RECORD_SIZE; + else + buf_size = RAS_TABLE_HEADER_SIZE + + control->ras_num_recs * RAS_TABLE_RECORD_SIZE; + buf = kzalloc(buf_size, GFP_KERNEL); if (!buf) { DRM_ERROR("Out of memory checking RAS table checksum.\n"); @@ -1080,6 +1260,39 @@ static int __verify_ras_table_checksum(struct amdgpu_ras_eeprom_control *control return res < 0 ? res : csum; } +static int __read_table_ras_info(struct amdgpu_ras_eeprom_control *control) +{ + struct amdgpu_ras_eeprom_table_ras_info *rai = &control->tbl_rai; + struct amdgpu_device *adev = to_amdgpu_device(control); + unsigned char *buf; + int res; + + buf = kzalloc(RAS_TABLE_V2_1_INFO_SIZE, GFP_KERNEL); + if (!buf) { + DRM_ERROR("Failed to alloc buf to read EEPROM table ras info\n"); + return -ENOMEM; + } + + /** + * EEPROM table V2_1 supports ras info, + * read EEPROM table ras info + */ + res = amdgpu_eeprom_read(adev->pm.ras_eeprom_i2c_bus, + control->i2c_address + control->ras_info_offset, + buf, RAS_TABLE_V2_1_INFO_SIZE); + if (res < RAS_TABLE_V2_1_INFO_SIZE) { + DRM_ERROR("Failed to read EEPROM table ras info, res:%d", res); + res = res >= 0 ? -EIO : res; + goto Out; + } + + __decode_table_ras_info_from_buf(rai, buf); + +Out: + kfree(buf); + return res == RAS_TABLE_V2_1_INFO_SIZE ? 0 : res; +} + int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control, bool *exceed_err_limit) { @@ -1102,8 +1315,7 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control, return -EINVAL; control->ras_header_offset = RAS_HDR_START; - control->ras_record_offset = RAS_RECORD_START; - control->ras_max_record_count = RAS_MAX_RECORD_COUNT; + control->ras_info_offset = RAS_TABLE_V2_1_INFO_START; mutex_init(&control->ras_tbl_mutex); /* Read the table header from EEPROM address */ @@ -1117,12 +1329,27 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control, __decode_table_header_from_buf(hdr, buf); - control->ras_num_recs = RAS_NUM_RECS(hdr); + if (hdr->version == RAS_TABLE_VER_V2_1) { + control->ras_num_recs = RAS_NUM_RECS_V2_1(hdr); + control->ras_record_offset = RAS_RECORD_START_V2_1; + control->ras_max_record_count = RAS_MAX_RECORD_COUNT_V2_1; + } else { + control->ras_num_recs = RAS_NUM_RECS(hdr); + control->ras_record_offset = RAS_RECORD_START; + control->ras_max_record_count = RAS_MAX_RECORD_COUNT; + } control->ras_fri = RAS_OFFSET_TO_INDEX(control, hdr->first_rec_offset); if (hdr->header == RAS_TABLE_HDR_VAL) { DRM_DEBUG_DRIVER("Found existing EEPROM table with %d records", control->ras_num_recs); + + if (hdr->version == RAS_TABLE_VER_V2_1) { + res = __read_table_ras_info(control); + if (res) + return res; + } + res = __verify_ras_table_checksum(control); if (res) DRM_ERROR("RAS table incorrect checksum or error:%d\n", @@ -1136,6 +1363,12 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control, ras->bad_page_cnt_threshold); } else if (hdr->header == RAS_TABLE_HDR_BAD && amdgpu_bad_page_threshold != 0) { + if (hdr->version == RAS_TABLE_VER_V2_1) { + res = __read_table_ras_info(control); + if (res) + return res; + } + res = __verify_ras_table_checksum(control); if (res) DRM_ERROR("RAS Table incorrect checksum or error:%d\n", diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h index 54d9bfe0881d9376471b82b218d9848f6c6b9bed..6dfd667f3013d0fd0c990bef7c819a8eeafa563f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h @@ -26,8 +26,16 @@ #include +#define RAS_TABLE_VER_V1 0x00010000 +#define RAS_TABLE_VER_V2_1 0x00021000 + struct amdgpu_device; +enum amdgpu_ras_gpu_health_status { + GPU_HEALTH_USABLE = 0, + GPU_RETIRED__ECC_REACH_THRESHOLD = 2, +}; + enum amdgpu_ras_eeprom_err_type { AMDGPU_RAS_EEPROM_ERR_NA, AMDGPU_RAS_EEPROM_ERR_RECOVERABLE, @@ -43,9 +51,18 @@ struct amdgpu_ras_eeprom_table_header { uint32_t checksum; } __packed; +struct amdgpu_ras_eeprom_table_ras_info { + u8 rma_status; + u8 health_percent; + u16 ecc_page_threshold; + u32 padding[64 - 1]; +} __packed; + struct amdgpu_ras_eeprom_control { struct amdgpu_ras_eeprom_table_header tbl_hdr; + struct amdgpu_ras_eeprom_table_ras_info tbl_rai; + /* Base I2C EEPPROM 19-bit memory address, * where the table is located. For more information, * see top of amdgpu_eeprom.c. @@ -58,6 +75,7 @@ struct amdgpu_ras_eeprom_control { * right after the header. */ u32 ras_header_offset; + u32 ras_info_offset; u32 ras_record_offset; /* Number of records in the table. @@ -124,7 +142,7 @@ int amdgpu_ras_eeprom_read(struct amdgpu_ras_eeprom_control *control, int amdgpu_ras_eeprom_append(struct amdgpu_ras_eeprom_control *control, struct eeprom_table_record *records, const u32 num); -uint32_t amdgpu_ras_eeprom_max_record_count(void); +uint32_t amdgpu_ras_eeprom_max_record_count(struct amdgpu_ras_eeprom_control *control); void amdgpu_ras_debugfs_set_ret_size(struct amdgpu_ras_eeprom_control *control); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c index 6437ead87e5fb9814315c00f589596ce391eefd9..eec41ad3040603622f4448458cd9f7342612acb4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c @@ -40,6 +40,7 @@ int amdgpu_reset_init(struct amdgpu_device *adev) switch (adev->ip_versions[MP1_HWIP][0]) { case IP_VERSION(13, 0, 2): + case IP_VERSION(13, 0, 6): ret = aldebaran_reset_init(adev); break; case IP_VERSION(11, 0, 7): @@ -61,6 +62,7 @@ int amdgpu_reset_fini(struct amdgpu_device *adev) switch (adev->ip_versions[MP1_HWIP][0]) { case IP_VERSION(13, 0, 2): + case IP_VERSION(13, 0, 6): ret = aldebaran_reset_fini(adev); break; case IP_VERSION(11, 0, 7): diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c index dc474b8096040bfdcc9c40dbf9426ca2ae24bdb6..80d6e132e4095d6b30811dca266bdfa88f77d240 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -49,6 +49,26 @@ * them until the pointers are equal again. */ +/** + * amdgpu_ring_max_ibs - Return max IBs that fit in a single submission. + * + * @type: ring type for which to return the limit. + */ +unsigned int amdgpu_ring_max_ibs(enum amdgpu_ring_type type) +{ + switch (type) { + case AMDGPU_RING_TYPE_GFX: + /* Need to keep at least 192 on GFX7+ for old radv. */ + return 192; + case AMDGPU_RING_TYPE_COMPUTE: + return 125; + case AMDGPU_RING_TYPE_VCN_JPEG: + return 16; + default: + return 49; + } +} + /** * amdgpu_ring_alloc - allocate space on the ring buffer * @@ -58,7 +78,7 @@ * Allocate @ndw dwords in the ring buffer (all asics). * Returns 0 on success, error on failure. */ -int amdgpu_ring_alloc(struct amdgpu_ring *ring, unsigned ndw) +int amdgpu_ring_alloc(struct amdgpu_ring *ring, unsigned int ndw) { /* Align requested size with padding so unlock_commit can * pad safely */ @@ -182,6 +202,7 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, int sched_hw_submission = amdgpu_sched_hw_submission; u32 *num_sched; u32 hw_ip; + unsigned int max_ibs_dw; /* Set the hw submission limit higher for KIQ because * it's used for a number of gfx/compute tasks by both @@ -290,6 +311,13 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, return r; } + max_ibs_dw = ring->funcs->emit_frame_size + + amdgpu_ring_max_ibs(ring->funcs->type) * ring->funcs->emit_ib_size; + max_ibs_dw = (max_ibs_dw + ring->funcs->align_mask) & ~ring->funcs->align_mask; + + if (WARN_ON(max_ibs_dw > max_dw)) + max_dw = max_ibs_dw; + ring->ring_size = roundup_pow_of_two(max_dw * 4 * sched_hw_submission); ring->buf_mask = (ring->ring_size / 4) - 1; @@ -361,6 +389,8 @@ void amdgpu_ring_fini(struct amdgpu_ring *ring) amdgpu_bo_free_kernel(&ring->ring_obj, &ring->gpu_addr, (void **)&ring->ring); + } else { + kfree(ring->fence_drv.fences); } dma_fence_put(ring->vmid_wait); @@ -403,11 +433,18 @@ void amdgpu_ring_emit_reg_write_reg_wait_helper(struct amdgpu_ring *ring, bool amdgpu_ring_soft_recovery(struct amdgpu_ring *ring, unsigned int vmid, struct dma_fence *fence) { + unsigned long flags; + ktime_t deadline = ktime_add_us(ktime_get(), 10000); if (amdgpu_sriov_vf(ring->adev) || !ring->funcs->soft_recovery || !fence) return false; + spin_lock_irqsave(fence->lock, flags); + if (!dma_fence_is_signaled_locked(fence)) + dma_fence_set_error(fence, -ENODATA); + spin_unlock_irqrestore(fence->lock, flags); + atomic_inc(&ring->adev->gpu_reset_counter); while (!dma_fence_is_signaled(fence) && ktime_to_ns(ktime_sub(deadline, ktime_get())) > 0) @@ -478,6 +515,70 @@ static const struct file_operations amdgpu_debugfs_ring_fops = { .llseek = default_llseek }; +static ssize_t amdgpu_debugfs_mqd_read(struct file *f, char __user *buf, + size_t size, loff_t *pos) +{ + struct amdgpu_ring *ring = file_inode(f)->i_private; + volatile u32 *mqd; + int r; + uint32_t value, result; + + if (*pos & 3 || size & 3) + return -EINVAL; + + result = 0; + + r = amdgpu_bo_reserve(ring->mqd_obj, false); + if (unlikely(r != 0)) + return r; + + r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&mqd); + if (r) { + amdgpu_bo_unreserve(ring->mqd_obj); + return r; + } + + while (size) { + if (*pos >= ring->mqd_size) + goto done; + + value = mqd[*pos/4]; + r = put_user(value, (uint32_t *)buf); + if (r) + goto done; + buf += 4; + result += 4; + size -= 4; + *pos += 4; + } + +done: + amdgpu_bo_kunmap(ring->mqd_obj); + mqd = NULL; + amdgpu_bo_unreserve(ring->mqd_obj); + if (r) + return r; + + return result; +} + +static const struct file_operations amdgpu_debugfs_mqd_fops = { + .owner = THIS_MODULE, + .read = amdgpu_debugfs_mqd_read, + .llseek = default_llseek +}; + +static int amdgpu_debugfs_ring_error(void *data, u64 val) +{ + struct amdgpu_ring *ring = data; + + amdgpu_fence_driver_set_error(ring, val); + return 0; +} + +DEFINE_DEBUGFS_ATTRIBUTE_SIGNED(amdgpu_debugfs_error_fops, NULL, + amdgpu_debugfs_ring_error, "%lld\n"); + #endif void amdgpu_debugfs_ring_init(struct amdgpu_device *adev, @@ -489,10 +590,21 @@ void amdgpu_debugfs_ring_init(struct amdgpu_device *adev, char name[32]; sprintf(name, "amdgpu_ring_%s", ring->name); - debugfs_create_file_size(name, S_IFREG | S_IRUGO, root, ring, + debugfs_create_file_size(name, S_IFREG | 0444, root, ring, &amdgpu_debugfs_ring_fops, ring->ring_size + 12); + if (ring->mqd_obj) { + sprintf(name, "amdgpu_mqd_%s", ring->name); + debugfs_create_file_size(name, S_IFREG | 0444, root, ring, + &amdgpu_debugfs_mqd_fops, + ring->mqd_size); + } + + sprintf(name, "amdgpu_error_%s", ring->name); + debugfs_create_file(name, 0200, root, ring, + &amdgpu_debugfs_error_fops); + #endif } @@ -581,3 +693,21 @@ void amdgpu_ring_ib_end(struct amdgpu_ring *ring) if (ring->is_sw_ring) amdgpu_sw_ring_ib_end(ring); } + +void amdgpu_ring_ib_on_emit_cntl(struct amdgpu_ring *ring) +{ + if (ring->is_sw_ring) + amdgpu_sw_ring_ib_mark_offset(ring, AMDGPU_MUX_OFFSET_TYPE_CONTROL); +} + +void amdgpu_ring_ib_on_emit_ce(struct amdgpu_ring *ring) +{ + if (ring->is_sw_ring) + amdgpu_sw_ring_ib_mark_offset(ring, AMDGPU_MUX_OFFSET_TYPE_CE); +} + +void amdgpu_ring_ib_on_emit_de(struct amdgpu_ring *ring) +{ + if (ring->is_sw_ring) + amdgpu_sw_ring_ib_mark_offset(ring, AMDGPU_MUX_OFFSET_TYPE_DE); +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h index d8749444b6891fb774b20df6f5a36b5a428ef21a..028ff075db5120ed165642d7da7d7adb7347f20c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h @@ -37,8 +37,8 @@ struct amdgpu_job; struct amdgpu_vm; /* max number of rings */ -#define AMDGPU_MAX_RINGS 28 -#define AMDGPU_MAX_HWIP_RINGS 8 +#define AMDGPU_MAX_RINGS 124 +#define AMDGPU_MAX_HWIP_RINGS 64 #define AMDGPU_MAX_GFX_RINGS 2 #define AMDGPU_MAX_SW_GFX_RINGS 2 #define AMDGPU_MAX_COMPUTE_RINGS 8 @@ -126,6 +126,7 @@ struct amdgpu_fence_driver { extern const struct drm_sched_backend_ops amdgpu_sched_ops; void amdgpu_fence_driver_clear_job_fences(struct amdgpu_ring *ring); +void amdgpu_fence_driver_set_error(struct amdgpu_ring *ring, int error); void amdgpu_fence_driver_force_completion(struct amdgpu_ring *ring); int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring); @@ -212,6 +213,8 @@ struct amdgpu_ring_funcs { void (*end_use)(struct amdgpu_ring *ring); void (*emit_switch_buffer) (struct amdgpu_ring *ring); void (*emit_cntxcntl) (struct amdgpu_ring *ring, uint32_t flags); + void (*emit_gfx_shadow)(struct amdgpu_ring *ring, u64 shadow_va, u64 csa_va, + u64 gds_va, bool init_shadow, int vmid); void (*emit_rreg)(struct amdgpu_ring *ring, uint32_t reg, uint32_t reg_val_offs); void (*emit_wreg)(struct amdgpu_ring *ring, uint32_t reg, uint32_t val); @@ -227,6 +230,9 @@ struct amdgpu_ring_funcs { int (*preempt_ib)(struct amdgpu_ring *ring); void (*emit_mem_sync)(struct amdgpu_ring *ring); void (*emit_wave_limit)(struct amdgpu_ring *ring, bool enable); + void (*patch_cntl)(struct amdgpu_ring *ring, unsigned offset); + void (*patch_ce)(struct amdgpu_ring *ring, unsigned offset); + void (*patch_de)(struct amdgpu_ring *ring, unsigned offset); }; struct amdgpu_ring { @@ -250,12 +256,14 @@ struct amdgpu_ring { uint32_t buf_mask; u32 idx; u32 xcc_id; + u32 xcp_id; u32 me; u32 pipe; u32 queue; struct amdgpu_bo *mqd_obj; uint64_t mqd_gpu_addr; void *mqd_ptr; + unsigned mqd_size; uint64_t eop_gpu_addr; u32 doorbell_index; bool use_doorbell; @@ -309,6 +317,7 @@ struct amdgpu_ring { #define amdgpu_ring_emit_hdp_flush(r) (r)->funcs->emit_hdp_flush((r)) #define amdgpu_ring_emit_switch_buffer(r) (r)->funcs->emit_switch_buffer((r)) #define amdgpu_ring_emit_cntxcntl(r, d) (r)->funcs->emit_cntxcntl((r), (d)) +#define amdgpu_ring_emit_gfx_shadow(r, s, c, g, i, v) ((r)->funcs->emit_gfx_shadow((r), (s), (c), (g), (i), (v))) #define amdgpu_ring_emit_rreg(r, d, o) (r)->funcs->emit_rreg((r), (d), (o)) #define amdgpu_ring_emit_wreg(r, d, v) (r)->funcs->emit_wreg((r), (d), (v)) #define amdgpu_ring_emit_reg_wait(r, d, v, m) (r)->funcs->emit_reg_wait((r), (d), (v), (m)) @@ -318,10 +327,17 @@ struct amdgpu_ring { #define amdgpu_ring_init_cond_exec(r) (r)->funcs->init_cond_exec((r)) #define amdgpu_ring_patch_cond_exec(r,o) (r)->funcs->patch_cond_exec((r),(o)) #define amdgpu_ring_preempt_ib(r) (r)->funcs->preempt_ib(r) +#define amdgpu_ring_patch_cntl(r, o) ((r)->funcs->patch_cntl((r), (o))) +#define amdgpu_ring_patch_ce(r, o) ((r)->funcs->patch_ce((r), (o))) +#define amdgpu_ring_patch_de(r, o) ((r)->funcs->patch_de((r), (o))) +unsigned int amdgpu_ring_max_ibs(enum amdgpu_ring_type type); int amdgpu_ring_alloc(struct amdgpu_ring *ring, unsigned ndw); void amdgpu_ring_ib_begin(struct amdgpu_ring *ring); void amdgpu_ring_ib_end(struct amdgpu_ring *ring); +void amdgpu_ring_ib_on_emit_cntl(struct amdgpu_ring *ring); +void amdgpu_ring_ib_on_emit_ce(struct amdgpu_ring *ring); +void amdgpu_ring_ib_on_emit_de(struct amdgpu_ring *ring); void amdgpu_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count); void amdgpu_ring_generic_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.c index 62079f0e3ee8f4f3cf0189b40b1401766766bf91..73516abef662f00f8b418a78a757adefb7bf7eef 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.c @@ -105,6 +105,16 @@ static void amdgpu_mux_resubmit_chunks(struct amdgpu_ring_mux *mux) amdgpu_fence_update_start_timestamp(e->ring, chunk->sync_seq, ktime_get()); + if (chunk->sync_seq == + le32_to_cpu(*(e->ring->fence_drv.cpu_addr + 2))) { + if (chunk->cntl_offset <= e->ring->buf_mask) + amdgpu_ring_patch_cntl(e->ring, + chunk->cntl_offset); + if (chunk->ce_offset <= e->ring->buf_mask) + amdgpu_ring_patch_ce(e->ring, chunk->ce_offset); + if (chunk->de_offset <= e->ring->buf_mask) + amdgpu_ring_patch_de(e->ring, chunk->de_offset); + } amdgpu_ring_mux_copy_pkt_from_sw_ring(mux, e->ring, chunk->start, chunk->end); @@ -407,6 +417,17 @@ void amdgpu_sw_ring_ib_end(struct amdgpu_ring *ring) amdgpu_ring_mux_end_ib(mux, ring); } +void amdgpu_sw_ring_ib_mark_offset(struct amdgpu_ring *ring, enum amdgpu_ring_mux_offset_type type) +{ + struct amdgpu_device *adev = ring->adev; + struct amdgpu_ring_mux *mux = &adev->gfx.muxer; + unsigned offset; + + offset = ring->wptr & ring->buf_mask; + + amdgpu_ring_mux_ib_mark_offset(mux, ring, offset, type); +} + void amdgpu_ring_mux_start_ib(struct amdgpu_ring_mux *mux, struct amdgpu_ring *ring) { struct amdgpu_mux_entry *e; @@ -429,6 +450,10 @@ void amdgpu_ring_mux_start_ib(struct amdgpu_ring_mux *mux, struct amdgpu_ring *r } chunk->start = ring->wptr; + /* the initialized value used to check if they are set by the ib submission*/ + chunk->cntl_offset = ring->buf_mask + 1; + chunk->de_offset = ring->buf_mask + 1; + chunk->ce_offset = ring->buf_mask + 1; list_add_tail(&chunk->entry, &e->list); } @@ -454,6 +479,41 @@ static void scan_and_remove_signaled_chunk(struct amdgpu_ring_mux *mux, struct a } } +void amdgpu_ring_mux_ib_mark_offset(struct amdgpu_ring_mux *mux, + struct amdgpu_ring *ring, u64 offset, + enum amdgpu_ring_mux_offset_type type) +{ + struct amdgpu_mux_entry *e; + struct amdgpu_mux_chunk *chunk; + + e = amdgpu_ring_mux_sw_entry(mux, ring); + if (!e) { + DRM_ERROR("cannot find entry!\n"); + return; + } + + chunk = list_last_entry(&e->list, struct amdgpu_mux_chunk, entry); + if (!chunk) { + DRM_ERROR("cannot find chunk!\n"); + return; + } + + switch (type) { + case AMDGPU_MUX_OFFSET_TYPE_CONTROL: + chunk->cntl_offset = offset; + break; + case AMDGPU_MUX_OFFSET_TYPE_DE: + chunk->de_offset = offset; + break; + case AMDGPU_MUX_OFFSET_TYPE_CE: + chunk->ce_offset = offset; + break; + default: + DRM_ERROR("invalid type (%d)\n", type); + break; + } +} + void amdgpu_ring_mux_end_ib(struct amdgpu_ring_mux *mux, struct amdgpu_ring *ring) { struct amdgpu_mux_entry *e; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.h index 4be45fc14954c9cf5a683641a4ef0d1a1e319eeb..b22d4fb2a8470e457ba5f0db4a9dcda352865716 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.h @@ -50,6 +50,12 @@ struct amdgpu_mux_entry { struct list_head list; }; +enum amdgpu_ring_mux_offset_type { + AMDGPU_MUX_OFFSET_TYPE_CONTROL, + AMDGPU_MUX_OFFSET_TYPE_DE, + AMDGPU_MUX_OFFSET_TYPE_CE, +}; + struct amdgpu_ring_mux { struct amdgpu_ring *real_ring; @@ -72,12 +78,18 @@ struct amdgpu_ring_mux { * @sync_seq: the fence seqno related with the saved IB. * @start:- start location on the software ring. * @end:- end location on the software ring. + * @control_offset:- the PRE_RESUME bit position used for resubmission. + * @de_offset:- the anchor in write_data for de meta of resubmission. + * @ce_offset:- the anchor in write_data for ce meta of resubmission. */ struct amdgpu_mux_chunk { struct list_head entry; uint32_t sync_seq; u64 start; u64 end; + u64 cntl_offset; + u64 de_offset; + u64 ce_offset; }; int amdgpu_ring_mux_init(struct amdgpu_ring_mux *mux, struct amdgpu_ring *ring, @@ -89,6 +101,8 @@ u64 amdgpu_ring_mux_get_wptr(struct amdgpu_ring_mux *mux, struct amdgpu_ring *ri u64 amdgpu_ring_mux_get_rptr(struct amdgpu_ring_mux *mux, struct amdgpu_ring *ring); void amdgpu_ring_mux_start_ib(struct amdgpu_ring_mux *mux, struct amdgpu_ring *ring); void amdgpu_ring_mux_end_ib(struct amdgpu_ring_mux *mux, struct amdgpu_ring *ring); +void amdgpu_ring_mux_ib_mark_offset(struct amdgpu_ring_mux *mux, struct amdgpu_ring *ring, + u64 offset, enum amdgpu_ring_mux_offset_type type); bool amdgpu_mcbp_handle_trailing_fence_irq(struct amdgpu_ring_mux *mux); u64 amdgpu_sw_ring_get_rptr_gfx(struct amdgpu_ring *ring); @@ -97,6 +111,7 @@ void amdgpu_sw_ring_set_wptr_gfx(struct amdgpu_ring *ring); void amdgpu_sw_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count); void amdgpu_sw_ring_ib_begin(struct amdgpu_ring *ring); void amdgpu_sw_ring_ib_end(struct amdgpu_ring *ring); +void amdgpu_sw_ring_ib_mark_offset(struct amdgpu_ring *ring, enum amdgpu_ring_mux_offset_type type); const char *amdgpu_sw_ring_name(int idx); unsigned int amdgpu_sw_ring_priority(int idx); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c index 85fb730d9fc84fba4dc41f6cade87172644ba938..35e0ae9acadcd616a056c73dd96960c3f8937569 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c @@ -31,12 +31,13 @@ * amdgpu_gfx_rlc_enter_safe_mode - Set RLC into safe mode * * @adev: amdgpu_device pointer + * @xcc_id: xcc accelerated compute core id * * Set RLC enter into safe mode if RLC is enabled and haven't in safe mode. */ -void amdgpu_gfx_rlc_enter_safe_mode(struct amdgpu_device *adev) +void amdgpu_gfx_rlc_enter_safe_mode(struct amdgpu_device *adev, int xcc_id) { - if (adev->gfx.rlc.in_safe_mode) + if (adev->gfx.rlc.in_safe_mode[xcc_id]) return; /* if RLC is not enabled, do nothing */ @@ -46,8 +47,8 @@ void amdgpu_gfx_rlc_enter_safe_mode(struct amdgpu_device *adev) if (adev->cg_flags & (AMD_CG_SUPPORT_GFX_CGCG | AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_3D_CGCG)) { - adev->gfx.rlc.funcs->set_safe_mode(adev); - adev->gfx.rlc.in_safe_mode = true; + adev->gfx.rlc.funcs->set_safe_mode(adev, xcc_id); + adev->gfx.rlc.in_safe_mode[xcc_id] = true; } } @@ -55,12 +56,13 @@ void amdgpu_gfx_rlc_enter_safe_mode(struct amdgpu_device *adev) * amdgpu_gfx_rlc_exit_safe_mode - Set RLC out of safe mode * * @adev: amdgpu_device pointer + * @xcc_id: xcc accelerated compute core id * * Set RLC exit safe mode if RLC is enabled and have entered into safe mode. */ -void amdgpu_gfx_rlc_exit_safe_mode(struct amdgpu_device *adev) +void amdgpu_gfx_rlc_exit_safe_mode(struct amdgpu_device *adev, int xcc_id) { - if (!(adev->gfx.rlc.in_safe_mode)) + if (!(adev->gfx.rlc.in_safe_mode[xcc_id])) return; /* if RLC is not enabled, do nothing */ @@ -70,8 +72,8 @@ void amdgpu_gfx_rlc_exit_safe_mode(struct amdgpu_device *adev) if (adev->cg_flags & (AMD_CG_SUPPORT_GFX_CGCG | AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_3D_CGCG)) { - adev->gfx.rlc.funcs->unset_safe_mode(adev); - adev->gfx.rlc.in_safe_mode = false; + adev->gfx.rlc.funcs->unset_safe_mode(adev, xcc_id); + adev->gfx.rlc.in_safe_mode[xcc_id] = false; } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h index 23f060db9255ce92479f112f02d00b520f8a1a77..80b263646966eef671ce23ee600e079350ebee82 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h @@ -157,8 +157,8 @@ typedef struct _RLC_TABLE_OF_CONTENT { struct amdgpu_rlc_funcs { bool (*is_rlc_enabled)(struct amdgpu_device *adev); - void (*set_safe_mode)(struct amdgpu_device *adev); - void (*unset_safe_mode)(struct amdgpu_device *adev); + void (*set_safe_mode)(struct amdgpu_device *adev, int xcc_id); + void (*unset_safe_mode)(struct amdgpu_device *adev, int xcc_id); int (*init)(struct amdgpu_device *adev); u32 (*get_csb_size)(struct amdgpu_device *adev); void (*get_csb_buffer)(struct amdgpu_device *adev, volatile u32 *buffer); @@ -201,7 +201,7 @@ struct amdgpu_rlc { u32 cp_table_size; /* safe mode for updating CG/PG state */ - bool in_safe_mode; + bool in_safe_mode[8]; const struct amdgpu_rlc_funcs *funcs; /* for firmware data */ @@ -260,8 +260,8 @@ struct amdgpu_rlc { struct amdgpu_rlcg_reg_access_ctrl reg_access_ctrl; }; -void amdgpu_gfx_rlc_enter_safe_mode(struct amdgpu_device *adev); -void amdgpu_gfx_rlc_exit_safe_mode(struct amdgpu_device *adev); +void amdgpu_gfx_rlc_enter_safe_mode(struct amdgpu_device *adev, int xcc_id); +void amdgpu_gfx_rlc_exit_safe_mode(struct amdgpu_device *adev, int xcc_id); int amdgpu_gfx_rlc_init_sr(struct amdgpu_device *adev, u32 dws); int amdgpu_gfx_rlc_init_csb(struct amdgpu_device *adev); int amdgpu_gfx_rlc_init_cpt(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c index 231ca06bc9c78ac57037e1e078c12bd7b6281c2c..78ec3420ef85307ba81e40d2876bab90ff20833c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c @@ -64,7 +64,7 @@ int amdgpu_sdma_get_index_from_ring(struct amdgpu_ring *ring, uint32_t *index) } uint64_t amdgpu_sdma_get_csa_mc_addr(struct amdgpu_ring *ring, - unsigned vmid) + unsigned int vmid) { struct amdgpu_device *adev = ring->adev; uint64_t csa_mc_addr; @@ -252,6 +252,13 @@ int amdgpu_sdma_init_microcode(struct amdgpu_device *adev, if (!duplicate && (instance != i)) continue; else { + /* Use a single copy per SDMA firmware type. PSP uses the same instance for all + * groups of SDMAs */ + if (adev->ip_versions[SDMA0_HWIP][0] == IP_VERSION(4, 4, 2) && + adev->firmware.load_type == AMDGPU_FW_LOAD_PSP && + adev->sdma.num_inst_per_aid == i) { + break; + } info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SDMA0 + i]; info->ucode_id = AMDGPU_UCODE_ID_SDMA0 + i; info->fw = adev->sdma.instance[i].fw; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h index fc8528812598b723b929031cc1deaf8e64073af9..513ac22120c1fa5da5731c8612fdf47f0b49e939 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h @@ -26,7 +26,7 @@ #include "amdgpu_ras.h" /* max number of IP instances */ -#define AMDGPU_MAX_SDMA_INSTANCES 8 +#define AMDGPU_MAX_SDMA_INSTANCES 16 enum amdgpu_sdma_irq { AMDGPU_SDMA_IRQ_INSTANCE0 = 0, @@ -37,9 +37,19 @@ enum amdgpu_sdma_irq { AMDGPU_SDMA_IRQ_INSTANCE5, AMDGPU_SDMA_IRQ_INSTANCE6, AMDGPU_SDMA_IRQ_INSTANCE7, + AMDGPU_SDMA_IRQ_INSTANCE8, + AMDGPU_SDMA_IRQ_INSTANCE9, + AMDGPU_SDMA_IRQ_INSTANCE10, + AMDGPU_SDMA_IRQ_INSTANCE11, + AMDGPU_SDMA_IRQ_INSTANCE12, + AMDGPU_SDMA_IRQ_INSTANCE13, + AMDGPU_SDMA_IRQ_INSTANCE14, + AMDGPU_SDMA_IRQ_INSTANCE15, AMDGPU_SDMA_IRQ_LAST }; +#define NUM_SDMA(x) hweight32(x) + struct amdgpu_sdma_instance { /* SDMA firmware */ const struct firmware *fw; @@ -49,6 +59,35 @@ struct amdgpu_sdma_instance { struct amdgpu_ring ring; struct amdgpu_ring page; bool burst_nop; + uint32_t aid_id; +}; + +enum amdgpu_sdma_ras_memory_id { + AMDGPU_SDMA_MBANK_DATA_BUF0 = 1, + AMDGPU_SDMA_MBANK_DATA_BUF1 = 2, + AMDGPU_SDMA_MBANK_DATA_BUF2 = 3, + AMDGPU_SDMA_MBANK_DATA_BUF3 = 4, + AMDGPU_SDMA_MBANK_DATA_BUF4 = 5, + AMDGPU_SDMA_MBANK_DATA_BUF5 = 6, + AMDGPU_SDMA_MBANK_DATA_BUF6 = 7, + AMDGPU_SDMA_MBANK_DATA_BUF7 = 8, + AMDGPU_SDMA_MBANK_DATA_BUF8 = 9, + AMDGPU_SDMA_MBANK_DATA_BUF9 = 10, + AMDGPU_SDMA_MBANK_DATA_BUF10 = 11, + AMDGPU_SDMA_MBANK_DATA_BUF11 = 12, + AMDGPU_SDMA_MBANK_DATA_BUF12 = 13, + AMDGPU_SDMA_MBANK_DATA_BUF13 = 14, + AMDGPU_SDMA_MBANK_DATA_BUF14 = 15, + AMDGPU_SDMA_MBANK_DATA_BUF15 = 16, + AMDGPU_SDMA_UCODE_BUF = 17, + AMDGPU_SDMA_RB_CMD_BUF = 18, + AMDGPU_SDMA_IB_CMD_BUF = 19, + AMDGPU_SDMA_UTCL1_RD_FIFO = 20, + AMDGPU_SDMA_UTCL1_RDBST_FIFO = 21, + AMDGPU_SDMA_UTCL1_WR_FIFO = 22, + AMDGPU_SDMA_DATA_LUT_FIFO = 23, + AMDGPU_SDMA_SPLIT_DAT_BUF = 24, + AMDGPU_SDMA_MEMORY_BLOCK_LAST, }; struct amdgpu_sdma_ras { @@ -66,6 +105,8 @@ struct amdgpu_sdma { struct amdgpu_irq_src srbm_write_irq; int num_instances; + uint32_t sdma_mask; + int num_inst_per_aid; uint32_t srbm_soft_reset; bool has_page_queue; struct ras_common_if *ras_if; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_smuio.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_smuio.h index c7a823f3f2c592fb8a7a6ca75a0ac4122090c909..89c38d8644718a664b580ce7cbec827753ff92f2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_smuio.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_smuio.h @@ -30,6 +30,7 @@ struct amdgpu_smuio_funcs { void (*get_clock_gating_state)(struct amdgpu_device *adev, u64 *flags); u32 (*get_die_id)(struct amdgpu_device *adev); u32 (*get_socket_id)(struct amdgpu_device *adev); + enum amdgpu_pkg_type (*get_pkg_type)(struct amdgpu_device *adev); bool (*is_host_gpu_xgmi_supported)(struct amdgpu_device *adev); }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 2cd081cbf70621d0159a1cff5499070241a8d927..0534ab7168094c273cd1deca72adfd32c87c0804 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -38,7 +38,6 @@ #include #include #include -#include #include #include #include @@ -65,7 +64,7 @@ MODULE_IMPORT_NS(DMA_BUF); -#define AMDGPU_TTM_VRAM_MAX_DW_READ (size_t)128 +#define AMDGPU_TTM_VRAM_MAX_DW_READ ((size_t)128) static int amdgpu_ttm_backend_bind(struct ttm_device *bdev, struct ttm_tt *ttm, @@ -184,11 +183,11 @@ static void amdgpu_evict_flags(struct ttm_buffer_object *bo, static int amdgpu_ttm_map_buffer(struct ttm_buffer_object *bo, struct ttm_resource *mem, struct amdgpu_res_cursor *mm_cur, - unsigned window, struct amdgpu_ring *ring, + unsigned int window, struct amdgpu_ring *ring, bool tmz, uint64_t *size, uint64_t *addr) { struct amdgpu_device *adev = ring->adev; - unsigned offset, num_pages, num_dw, num_bytes; + unsigned int offset, num_pages, num_dw, num_bytes; uint64_t src_addr, dst_addr; struct amdgpu_job *job; void *cpu_addr; @@ -229,7 +228,7 @@ static int amdgpu_ttm_map_buffer(struct ttm_buffer_object *bo, num_dw = ALIGN(adev->mman.buffer_funcs->copy_num_dw, 8); num_bytes = num_pages * 8 * AMDGPU_GPU_PAGES_IN_CPU_PAGE; - r = amdgpu_job_alloc_with_ib(adev, &adev->mman.entity, + r = amdgpu_job_alloc_with_ib(adev, &adev->mman.high_pr, AMDGPU_FENCE_OWNER_UNDEFINED, num_dw * 4 + num_bytes, AMDGPU_IB_POOL_DELAYED, &job); @@ -384,7 +383,8 @@ static int amdgpu_move_blit(struct ttm_buffer_object *bo, (abo->flags & AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE)) { struct dma_fence *wipe_fence = NULL; - r = amdgpu_fill_buffer(abo, AMDGPU_POISON, NULL, &wipe_fence); + r = amdgpu_fill_buffer(abo, AMDGPU_POISON, NULL, &wipe_fence, + false); if (r) { goto error; } else if (wipe_fence) { @@ -631,6 +631,7 @@ struct amdgpu_ttm_tt { struct task_struct *usertask; uint32_t userflags; bool bound; + int32_t pool_id; }; #define ttm_to_amdgpu_ttm_tt(ptr) container_of(ptr, struct amdgpu_ttm_tt, ttm) @@ -800,6 +801,44 @@ static void amdgpu_ttm_tt_unpin_userptr(struct ttm_device *bdev, sg_free_table(ttm->sg); } +/* + * total_pages is constructed as MQD0+CtrlStack0 + MQD1+CtrlStack1 + ... + * MQDn+CtrlStackn where n is the number of XCCs per partition. + * pages_per_xcc is the size of one MQD+CtrlStack. The first page is MQD + * and uses memory type default, UC. The rest of pages_per_xcc are + * Ctrl stack and modify their memory type to NC. + */ +static void amdgpu_ttm_gart_bind_gfx9_mqd(struct amdgpu_device *adev, + struct ttm_tt *ttm, uint64_t flags) +{ + struct amdgpu_ttm_tt *gtt = (void *)ttm; + uint64_t total_pages = ttm->num_pages; + int num_xcc = max(1U, adev->gfx.num_xcc_per_xcp); + uint64_t page_idx, pages_per_xcc; + int i; + uint64_t ctrl_flags = (flags & ~AMDGPU_PTE_MTYPE_VG10_MASK) | + AMDGPU_PTE_MTYPE_VG10(AMDGPU_MTYPE_NC); + + pages_per_xcc = total_pages; + do_div(pages_per_xcc, num_xcc); + + for (i = 0, page_idx = 0; i < num_xcc; i++, page_idx += pages_per_xcc) { + /* MQD page: use default flags */ + amdgpu_gart_bind(adev, + gtt->offset + (page_idx << PAGE_SHIFT), + 1, >t->ttm.dma_address[page_idx], flags); + /* + * Ctrl pages - modify the memory type to NC (ctrl_flags) from + * the second page of the BO onward. + */ + amdgpu_gart_bind(adev, + gtt->offset + ((page_idx + 1) << PAGE_SHIFT), + pages_per_xcc - 1, + >t->ttm.dma_address[page_idx + 1], + ctrl_flags); + } +} + static void amdgpu_ttm_gart_bind(struct amdgpu_device *adev, struct ttm_buffer_object *tbo, uint64_t flags) @@ -812,21 +851,7 @@ static void amdgpu_ttm_gart_bind(struct amdgpu_device *adev, flags |= AMDGPU_PTE_TMZ; if (abo->flags & AMDGPU_GEM_CREATE_CP_MQD_GFX9) { - uint64_t page_idx = 1; - - amdgpu_gart_bind(adev, gtt->offset, page_idx, - gtt->ttm.dma_address, flags); - - /* The memory type of the first page defaults to UC. Now - * modify the memory type to NC from the second page of - * the BO onward. - */ - flags &= ~AMDGPU_PTE_MTYPE_VG10_MASK; - flags |= AMDGPU_PTE_MTYPE_VG10(AMDGPU_MTYPE_NC); - - amdgpu_gart_bind(adev, gtt->offset + (page_idx << PAGE_SHIFT), - ttm->num_pages - page_idx, - &(gtt->ttm.dma_address[page_idx]), flags); + amdgpu_ttm_gart_bind_gfx9_mqd(adev, ttm, flags); } else { amdgpu_gart_bind(adev, gtt->offset, ttm->num_pages, gtt->ttm.dma_address, flags); @@ -1029,15 +1054,20 @@ static void amdgpu_ttm_backend_destroy(struct ttm_device *bdev, static struct ttm_tt *amdgpu_ttm_tt_create(struct ttm_buffer_object *bo, uint32_t page_flags) { + struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev); struct amdgpu_bo *abo = ttm_to_amdgpu_bo(bo); struct amdgpu_ttm_tt *gtt; enum ttm_caching caching; gtt = kzalloc(sizeof(struct amdgpu_ttm_tt), GFP_KERNEL); - if (gtt == NULL) { + if (!gtt) return NULL; - } + gtt->gobj = &bo->base; + if (adev->gmc.mem_partitions && abo->xcp_id >= 0) + gtt->pool_id = KFD_XCP_MEM_ID(adev, abo->xcp_id); + else + gtt->pool_id = abo->xcp_id; if (abo->flags & AMDGPU_GEM_CREATE_CPU_GTT_USWC) caching = ttm_write_combined; @@ -1064,6 +1094,7 @@ static int amdgpu_ttm_tt_populate(struct ttm_device *bdev, { struct amdgpu_device *adev = amdgpu_ttm_adev(bdev); struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); + struct ttm_pool *pool; pgoff_t i; int ret; @@ -1078,7 +1109,11 @@ static int amdgpu_ttm_tt_populate(struct ttm_device *bdev, if (ttm->page_flags & TTM_TT_FLAG_EXTERNAL) return 0; - ret = ttm_pool_alloc(&adev->mman.bdev.pool, ttm, ctx); + if (adev->mman.ttm_pools && gtt->pool_id >= 0) + pool = &adev->mman.ttm_pools[gtt->pool_id]; + else + pool = &adev->mman.bdev.pool; + ret = ttm_pool_alloc(pool, ttm, ctx); if (ret) return ret; @@ -1099,6 +1134,7 @@ static void amdgpu_ttm_tt_unpopulate(struct ttm_device *bdev, { struct amdgpu_ttm_tt *gtt = ttm_to_amdgpu_ttm_tt(ttm); struct amdgpu_device *adev; + struct ttm_pool *pool; pgoff_t i; amdgpu_ttm_backend_unbind(bdev, ttm); @@ -1117,7 +1153,13 @@ static void amdgpu_ttm_tt_unpopulate(struct ttm_device *bdev, ttm->pages[i]->mapping = NULL; adev = amdgpu_ttm_adev(bdev); - return ttm_pool_free(&adev->mman.bdev.pool, ttm); + + if (adev->mman.ttm_pools && gtt->pool_id >= 0) + pool = &adev->mman.ttm_pools[gtt->pool_id]; + else + pool = &adev->mman.bdev.pool; + + return ttm_pool_free(pool, ttm); } /** @@ -1414,7 +1456,7 @@ static int amdgpu_ttm_access_memory_sdma(struct ttm_buffer_object *bo, memcpy(adev->mman.sdma_access_ptr, buf, len); num_dw = ALIGN(adev->mman.buffer_funcs->copy_num_dw, 8); - r = amdgpu_job_alloc_with_ib(adev, &adev->mman.entity, + r = amdgpu_job_alloc_with_ib(adev, &adev->mman.high_pr, AMDGPU_FENCE_OWNER_UNDEFINED, num_dw * 4, AMDGPU_IB_POOL_DELAYED, &job); @@ -1623,14 +1665,15 @@ static int amdgpu_ttm_training_reserve_vram_fini(struct amdgpu_device *adev) return 0; } -static void amdgpu_ttm_training_data_block_init(struct amdgpu_device *adev) +static void amdgpu_ttm_training_data_block_init(struct amdgpu_device *adev, + uint32_t reserve_size) { struct psp_memory_training_context *ctx = &adev->psp.mem_train_ctx; memset(ctx, 0, sizeof(*ctx)); ctx->c2p_train_data_offset = - ALIGN((adev->gmc.mc_vram_size - adev->mman.discovery_tmr_size - SZ_1M), SZ_1M); + ALIGN((adev->gmc.mc_vram_size - reserve_size - SZ_1M), SZ_1M); ctx->p2c_train_data_offset = (adev->gmc.mc_vram_size - GDDR6_MEM_TRAINING_OFFSET); ctx->train_data_size = @@ -1648,11 +1691,12 @@ static void amdgpu_ttm_training_data_block_init(struct amdgpu_device *adev) */ static int amdgpu_ttm_reserve_tmr(struct amdgpu_device *adev) { - int ret; struct psp_memory_training_context *ctx = &adev->psp.mem_train_ctx; bool mem_train_support = false; + uint32_t reserve_size = 0; + int ret; - if (!amdgpu_sriov_vf(adev)) { + if (adev->bios && !amdgpu_sriov_vf(adev)) { if (amdgpu_atomfirmware_mem_training_supported(adev)) mem_train_support = true; else @@ -1666,14 +1710,18 @@ static int amdgpu_ttm_reserve_tmr(struct amdgpu_device *adev) * Otherwise, fallback to legacy approach to check and reserve tmr block for ip * discovery data and G6 memory training data respectively */ - adev->mman.discovery_tmr_size = - amdgpu_atomfirmware_get_fw_reserved_fb_size(adev); - if (!adev->mman.discovery_tmr_size) - adev->mman.discovery_tmr_size = DISCOVERY_TMR_OFFSET; + if (adev->bios) + reserve_size = + amdgpu_atomfirmware_get_fw_reserved_fb_size(adev); + + if (!adev->bios && adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) + reserve_size = max(reserve_size, (uint32_t)280 << 20); + else if (!reserve_size) + reserve_size = DISCOVERY_TMR_OFFSET; if (mem_train_support) { /* reserve vram for mem train according to TMR location */ - amdgpu_ttm_training_data_block_init(adev); + amdgpu_ttm_training_data_block_init(adev, reserve_size); ret = amdgpu_bo_create_kernel_at(adev, ctx->c2p_train_data_offset, ctx->train_data_size, @@ -1687,20 +1735,58 @@ static int amdgpu_ttm_reserve_tmr(struct amdgpu_device *adev) ctx->init = PSP_MEM_TRAIN_RESERVE_SUCCESS; } - ret = amdgpu_bo_create_kernel_at(adev, - adev->gmc.real_vram_size - adev->mman.discovery_tmr_size, - adev->mman.discovery_tmr_size, - &adev->mman.discovery_memory, - NULL); - if (ret) { - DRM_ERROR("alloc tmr failed(%d)!\n", ret); - amdgpu_bo_free_kernel(&adev->mman.discovery_memory, NULL, NULL); - return ret; + if (!adev->gmc.is_app_apu) { + ret = amdgpu_bo_create_kernel_at( + adev, adev->gmc.real_vram_size - reserve_size, + reserve_size, &adev->mman.fw_reserved_memory, NULL); + if (ret) { + DRM_ERROR("alloc tmr failed(%d)!\n", ret); + amdgpu_bo_free_kernel(&adev->mman.fw_reserved_memory, + NULL, NULL); + return ret; + } + } else { + DRM_DEBUG_DRIVER("backdoor fw loading path for PSP TMR, no reservation needed\n"); } return 0; } +static int amdgpu_ttm_pools_init(struct amdgpu_device *adev) +{ + int i; + + if (!adev->gmc.is_app_apu || !adev->gmc.num_mem_partitions) + return 0; + + adev->mman.ttm_pools = kcalloc(adev->gmc.num_mem_partitions, + sizeof(*adev->mman.ttm_pools), + GFP_KERNEL); + if (!adev->mman.ttm_pools) + return -ENOMEM; + + for (i = 0; i < adev->gmc.num_mem_partitions; i++) { + ttm_pool_init(&adev->mman.ttm_pools[i], adev->dev, + adev->gmc.mem_partitions[i].numa.node, + false, false); + } + return 0; +} + +static void amdgpu_ttm_pools_fini(struct amdgpu_device *adev) +{ + int i; + + if (!adev->gmc.is_app_apu || !adev->mman.ttm_pools) + return; + + for (i = 0; i < adev->gmc.num_mem_partitions; i++) + ttm_pool_fini(&adev->mman.ttm_pools[i]); + + kfree(adev->mman.ttm_pools); + adev->mman.ttm_pools = NULL; +} + /* * amdgpu_ttm_init - Init the memory management (ttm) as well as various * gtt/vram related fields. @@ -1727,6 +1813,12 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) DRM_ERROR("failed initializing buffer object driver(%d).\n", r); return r; } + + r = amdgpu_ttm_pools_init(adev); + if (r) { + DRM_ERROR("failed to init ttm pools(%d).\n", r); + return r; + } adev->mman.initialized = true; /* Initialize VRAM pool with all of VRAM divided into pages */ @@ -1744,6 +1836,9 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) adev->mman.aper_base_kaddr = ioremap_cache(adev->gmc.aper_base, adev->gmc.visible_vram_size); + else if (adev->gmc.is_app_apu) + DRM_DEBUG_DRIVER( + "No need to ioremap when real vram size is 0\n"); else #endif adev->mman.aper_base_kaddr = ioremap_wc(adev->gmc.aper_base, @@ -1755,9 +1850,8 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) *place on the VRAM, so reserve it early. */ r = amdgpu_ttm_fw_reserve_vram_init(adev); - if (r) { + if (r) return r; - } /* *The reserved vram for driver must be pinned to the specified @@ -1781,49 +1875,46 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) /* allocate memory as required for VGA * This is used for VGA emulation and pre-OS scanout buffers to * avoid display artifacts while transitioning between pre-OS - * and driver. */ - r = amdgpu_bo_create_kernel_at(adev, 0, adev->mman.stolen_vga_size, - &adev->mman.stolen_vga_memory, - NULL); - if (r) - return r; - r = amdgpu_bo_create_kernel_at(adev, adev->mman.stolen_vga_size, - adev->mman.stolen_extended_size, - &adev->mman.stolen_extended_memory, - NULL); - if (r) - return r; - r = amdgpu_bo_create_kernel_at(adev, adev->mman.stolen_reserved_offset, - adev->mman.stolen_reserved_size, - &adev->mman.stolen_reserved_memory, - NULL); - if (r) - return r; + * and driver. + */ + if (!adev->gmc.is_app_apu) { + r = amdgpu_bo_create_kernel_at(adev, 0, + adev->mman.stolen_vga_size, + &adev->mman.stolen_vga_memory, + NULL); + if (r) + return r; - DRM_INFO("amdgpu: %uM of VRAM memory ready\n", - (unsigned) (adev->gmc.real_vram_size / (1024 * 1024))); - - /* Compute GTT size, either based on 1/2 the size of RAM size - * or whatever the user passed on module init */ - if (amdgpu_gtt_size == -1) { - struct sysinfo si; - - si_meminfo(&si); - /* Certain GL unit tests for large textures can cause problems - * with the OOM killer since there is no way to link this memory - * to a process. This was originally mitigated (but not necessarily - * eliminated) by limiting the GTT size. The problem is this limit - * is often too low for many modern games so just make the limit 1/2 - * of system memory which aligns with TTM. The OOM accounting needs - * to be addressed, but we shouldn't prevent common 3D applications - * from being usable just to potentially mitigate that corner case. - */ - gtt_size = max((AMDGPU_DEFAULT_GTT_SIZE_MB << 20), - (u64)si.totalram * si.mem_unit / 2); + r = amdgpu_bo_create_kernel_at(adev, adev->mman.stolen_vga_size, + adev->mman.stolen_extended_size, + &adev->mman.stolen_extended_memory, + NULL); + + if (r) + return r; + + r = amdgpu_bo_create_kernel_at(adev, + adev->mman.stolen_reserved_offset, + adev->mman.stolen_reserved_size, + &adev->mman.stolen_reserved_memory, + NULL); + if (r) + return r; } else { - gtt_size = (uint64_t)amdgpu_gtt_size << 20; + DRM_DEBUG_DRIVER("Skipped stolen memory reservation\n"); } + DRM_INFO("amdgpu: %uM of VRAM memory ready\n", + (unsigned int)(adev->gmc.real_vram_size / (1024 * 1024))); + + /* Compute GTT size, either based on TTM limit + * or whatever the user passed on module init. + */ + if (amdgpu_gtt_size == -1) + gtt_size = ttm_tt_pages_limit() << PAGE_SHIFT; + else + gtt_size = (uint64_t)amdgpu_gtt_size << 20; + /* Initialize GTT memory pool */ r = amdgpu_gtt_mgr_init(adev, gtt_size); if (r) { @@ -1831,7 +1922,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) return r; } DRM_INFO("amdgpu: %uM of GTT memory ready.\n", - (unsigned)(gtt_size / (1024 * 1024))); + (unsigned int)(gtt_size / (1024 * 1024))); /* Initialize preemptible memory pool */ r = amdgpu_preempt_mgr_init(adev); @@ -1858,7 +1949,6 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) DRM_ERROR("Failed initializing oa heap.\n"); return r; } - if (amdgpu_bo_create_kernel(adev, PAGE_SIZE, PAGE_SIZE, AMDGPU_GEM_DOMAIN_GTT, &adev->mman.sdma_access_bo, NULL, @@ -1874,18 +1964,24 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) void amdgpu_ttm_fini(struct amdgpu_device *adev) { int idx; + if (!adev->mman.initialized) return; + amdgpu_ttm_pools_fini(adev); + amdgpu_ttm_training_reserve_vram_fini(adev); /* return the stolen vga memory back to VRAM */ - amdgpu_bo_free_kernel(&adev->mman.stolen_vga_memory, NULL, NULL); - amdgpu_bo_free_kernel(&adev->mman.stolen_extended_memory, NULL, NULL); - /* return the IP Discovery TMR memory back to VRAM */ - amdgpu_bo_free_kernel(&adev->mman.discovery_memory, NULL, NULL); - if (adev->mman.stolen_reserved_size) - amdgpu_bo_free_kernel(&adev->mman.stolen_reserved_memory, - NULL, NULL); + if (!adev->gmc.is_app_apu) { + amdgpu_bo_free_kernel(&adev->mman.stolen_vga_memory, NULL, NULL); + amdgpu_bo_free_kernel(&adev->mman.stolen_extended_memory, NULL, NULL); + /* return the FW reserved memory back to VRAM */ + amdgpu_bo_free_kernel(&adev->mman.fw_reserved_memory, NULL, + NULL); + if (adev->mman.stolen_reserved_size) + amdgpu_bo_free_kernel(&adev->mman.stolen_reserved_memory, + NULL, NULL); + } amdgpu_bo_free_kernel(&adev->mman.sdma_access_bo, NULL, &adev->mman.sdma_access_ptr); amdgpu_ttm_fw_reserve_vram_fini(adev); @@ -1927,7 +2023,7 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) int r; if (!adev->mman.initialized || amdgpu_in_reset(adev) || - adev->mman.buffer_funcs_enabled == enable) + adev->mman.buffer_funcs_enabled == enable || adev->gmc.is_app_apu) return; if (enable) { @@ -1936,7 +2032,7 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) ring = adev->mman.buffer_funcs_ring; sched = &ring->sched; - r = drm_sched_entity_init(&adev->mman.entity, + r = drm_sched_entity_init(&adev->mman.high_pr, DRM_SCHED_PRIORITY_KERNEL, &sched, 1, NULL); if (r) { @@ -1944,8 +2040,18 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) r); return; } + + r = drm_sched_entity_init(&adev->mman.low_pr, + DRM_SCHED_PRIORITY_NORMAL, &sched, + 1, NULL); + if (r) { + DRM_ERROR("Failed setting up TTM BO move entity (%d)\n", + r); + goto error_free_entity; + } } else { - drm_sched_entity_destroy(&adev->mman.entity); + drm_sched_entity_destroy(&adev->mman.high_pr); + drm_sched_entity_destroy(&adev->mman.low_pr); dma_fence_put(man->move); man->move = NULL; } @@ -1957,6 +2063,11 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) size = adev->gmc.visible_vram_size; man->size = size; adev->mman.buffer_funcs_enabled = enable; + + return; + +error_free_entity: + drm_sched_entity_destroy(&adev->mman.high_pr); } static int amdgpu_ttm_prepare_job(struct amdgpu_device *adev, @@ -1964,14 +2075,16 @@ static int amdgpu_ttm_prepare_job(struct amdgpu_device *adev, unsigned int num_dw, struct dma_resv *resv, bool vm_needs_flush, - struct amdgpu_job **job) + struct amdgpu_job **job, + bool delayed) { enum amdgpu_ib_pool_type pool = direct_submit ? AMDGPU_IB_POOL_DIRECT : AMDGPU_IB_POOL_DELAYED; int r; - - r = amdgpu_job_alloc_with_ib(adev, &adev->mman.entity, + struct drm_sched_entity *entity = delayed ? &adev->mman.low_pr : + &adev->mman.high_pr; + r = amdgpu_job_alloc_with_ib(adev, entity, AMDGPU_FENCE_OWNER_UNDEFINED, num_dw * 4, pool, job); if (r) @@ -1997,10 +2110,10 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, bool vm_needs_flush, bool tmz) { struct amdgpu_device *adev = ring->adev; - unsigned num_loops, num_dw; + unsigned int num_loops, num_dw; struct amdgpu_job *job; uint32_t max_bytes; - unsigned i; + unsigned int i; int r; if (!direct_submit && !ring->sched.ready) { @@ -2012,7 +2125,7 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, num_loops = DIV_ROUND_UP(byte_count, max_bytes); num_dw = ALIGN(num_loops * adev->mman.buffer_funcs->copy_num_dw, 8); r = amdgpu_ttm_prepare_job(adev, direct_submit, num_dw, - resv, vm_needs_flush, &job); + resv, vm_needs_flush, &job, false); if (r) return r; @@ -2048,7 +2161,7 @@ static int amdgpu_ttm_fill_mem(struct amdgpu_ring *ring, uint32_t src_data, uint64_t dst_addr, uint32_t byte_count, struct dma_resv *resv, struct dma_fence **fence, - bool vm_needs_flush) + bool vm_needs_flush, bool delayed) { struct amdgpu_device *adev = ring->adev; unsigned int num_loops, num_dw; @@ -2061,7 +2174,7 @@ static int amdgpu_ttm_fill_mem(struct amdgpu_ring *ring, uint32_t src_data, num_loops = DIV_ROUND_UP_ULL(byte_count, max_bytes); num_dw = ALIGN(num_loops * adev->mman.buffer_funcs->fill_num_dw, 8); r = amdgpu_ttm_prepare_job(adev, false, num_dw, resv, vm_needs_flush, - &job); + &job, delayed); if (r) return r; @@ -2084,7 +2197,8 @@ static int amdgpu_ttm_fill_mem(struct amdgpu_ring *ring, uint32_t src_data, int amdgpu_fill_buffer(struct amdgpu_bo *bo, uint32_t src_data, struct dma_resv *resv, - struct dma_fence **f) + struct dma_fence **f, + bool delayed) { struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev); struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring; @@ -2113,7 +2227,7 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo, goto error; r = amdgpu_ttm_fill_mem(ring, src_data, to, cur_size, resv, - &next, true); + &next, true, delayed); if (r) goto error; @@ -2164,7 +2278,7 @@ int amdgpu_ttm_evict_resources(struct amdgpu_device *adev, int mem_type) static int amdgpu_ttm_page_pool_show(struct seq_file *m, void *unused) { - struct amdgpu_device *adev = (struct amdgpu_device *)m->private; + struct amdgpu_device *adev = m->private; return ttm_pool_debugfs(&adev->mman.bdev.pool, m); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h index e2cd5894afc9d2b87726eaa24d9167843bfd9125..6d0d66e40db934abefa8c25e973ddcd283138c07 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h @@ -49,6 +49,7 @@ struct amdgpu_gtt_mgr { struct amdgpu_mman { struct ttm_device bdev; + struct ttm_pool *ttm_pools; bool initialized; void __iomem *aper_base_kaddr; @@ -58,8 +59,10 @@ struct amdgpu_mman { bool buffer_funcs_enabled; struct mutex gtt_window_lock; - /* Scheduler entity for buffer moves */ - struct drm_sched_entity entity; + /* High priority scheduler entity for buffer moves */ + struct drm_sched_entity high_pr; + /* Low priority scheduler entity for VRAM clearing */ + struct drm_sched_entity low_pr; struct amdgpu_vram_mgr vram_mgr; struct amdgpu_gtt_mgr gtt_mgr; @@ -78,7 +81,8 @@ struct amdgpu_mman { /* discovery */ uint8_t *discovery_bin; uint32_t discovery_tmr_size; - struct amdgpu_bo *discovery_memory; + /* fw reserved memory */ + struct amdgpu_bo *fw_reserved_memory; /* firmware VRAM reservation */ u64 fw_vram_usage_start_offset; @@ -150,7 +154,8 @@ int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev, int amdgpu_fill_buffer(struct amdgpu_bo *bo, uint32_t src_data, struct dma_resv *resv, - struct dma_fence **fence); + struct dma_fence **fence, + bool delayed); int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo); void amdgpu_ttm_recover_gart(struct ttm_buffer_object *tbo); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c index f76b1cb8baf85591670129ea231bc6138a6d43f1..16807ff96dc9b4b195c3d3b194bcbeb7599db2f3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c @@ -748,7 +748,7 @@ static int amdgpu_ucode_init_single_fw(struct amdgpu_device *adev, const struct imu_firmware_header_v1_0 *imu_hdr = NULL; u8 *ucode_addr; - if (NULL == ucode->fw) + if (!ucode->fw) return 0; ucode->mc_addr = mc_addr; @@ -972,7 +972,7 @@ static int amdgpu_ucode_patch_jt(struct amdgpu_firmware_info *ucode, uint8_t *src_addr = NULL; uint8_t *dst_addr = NULL; - if (NULL == ucode->fw) + if (!ucode->fw) return 0; comm_hdr = (const struct common_firmware_header *)ucode->fw->data; @@ -1043,6 +1043,7 @@ int amdgpu_ucode_init_bo(struct amdgpu_device *adev) if (i == AMDGPU_UCODE_ID_CP_MEC1 && adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) { const struct gfx_firmware_header_v1_0 *cp_hdr; + cp_hdr = (const struct gfx_firmware_header_v1_0 *)ucode->fw->data; amdgpu_ucode_patch_jt(ucode, adev->firmware.fw_buf_mc + fw_offset, adev->firmware.fw_buf_ptr + fw_offset); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c index 1edf8e6aeb1623df882d1139be1e49d4a849c268..db0d94ca4ffc5197c898113d8540ffe7db1a9e6f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c @@ -169,27 +169,31 @@ int amdgpu_umc_poison_handler(struct amdgpu_device *adev, bool reset) { int ret = AMDGPU_RAS_SUCCESS; - if (!amdgpu_sriov_vf(adev)) { - if (!adev->gmc.xgmi.connected_to_cpu) { - struct ras_err_data err_data = {0, 0, 0, NULL}; - struct ras_common_if head = { - .block = AMDGPU_RAS_BLOCK__UMC, - }; - struct ras_manager *obj = amdgpu_ras_find_obj(adev, &head); - - ret = amdgpu_umc_do_page_retirement(adev, &err_data, NULL, reset); - - if (ret == AMDGPU_RAS_SUCCESS && obj) { - obj->err_data.ue_count += err_data.ue_count; - obj->err_data.ce_count += err_data.ce_count; - } - } else if (reset) { + if (adev->gmc.xgmi.connected_to_cpu || + adev->gmc.is_app_apu) { + if (reset) { /* MCA poison handler is only responsible for GPU reset, * let MCA notifier do page retirement. */ kgd2kfd_set_sram_ecc_flag(adev->kfd.dev); amdgpu_ras_reset_gpu(adev); } + return ret; + } + + if (!amdgpu_sriov_vf(adev)) { + struct ras_err_data err_data = {0, 0, 0, NULL}; + struct ras_common_if head = { + .block = AMDGPU_RAS_BLOCK__UMC, + }; + struct ras_manager *obj = amdgpu_ras_find_obj(adev, &head); + + ret = amdgpu_umc_do_page_retirement(adev, &err_data, NULL, reset); + + if (ret == AMDGPU_RAS_SUCCESS && obj) { + obj->err_data.ue_count += err_data.ue_count; + obj->err_data.ce_count += err_data.ce_count; + } } else { if (adev->virt.ops && adev->virt.ops->ras_poison_handler) adev->virt.ops->ras_poison_handler(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h index 86133f77a9a42f6229c7d32843fbf43c001d9837..43321f57f557fcc1d9f4e0915985c0cb73938a88 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h @@ -59,6 +59,8 @@ struct amdgpu_umc_ras { void *ras_error_status); void (*ecc_info_query_ras_error_address)(struct amdgpu_device *adev, void *ras_error_status); + /* support different eeprom table version for different asic */ + void (*set_eeprom_table_version)(struct amdgpu_ras_eeprom_table_header *hdr); }; struct amdgpu_umc_funcs { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umr.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_umr.h index 919d9d4017509179ed31ec8e0698335cecef35d9..107f9bb0e24f7f2f8e810084cab0e43e8da9f020 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umr.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umr.h @@ -35,17 +35,51 @@ struct amdgpu_debugfs_regs2_iocdata { } srbm; }; +struct amdgpu_debugfs_regs2_iocdata_v2 { + __u32 use_srbm, use_grbm, pg_lock; + struct { + __u32 se, sh, instance; + } grbm; + struct { + __u32 me, pipe, queue, vmid; + } srbm; + u32 xcc_id; +}; + +struct amdgpu_debugfs_gprwave_iocdata { + u32 gpr_or_wave, se, sh, cu, wave, simd, xcc_id; + struct { + u32 thread, vpgr_or_sgpr; + } gpr; +}; + /* * MMIO debugfs state data (per file* handle) */ struct amdgpu_debugfs_regs2_data { struct amdgpu_device *adev; struct mutex lock; - struct amdgpu_debugfs_regs2_iocdata id; + struct amdgpu_debugfs_regs2_iocdata_v2 id; +}; + +struct amdgpu_debugfs_gprwave_data { + struct amdgpu_device *adev; + struct mutex lock; + struct amdgpu_debugfs_gprwave_iocdata id; }; enum AMDGPU_DEBUGFS_REGS2_CMDS { AMDGPU_DEBUGFS_REGS2_CMD_SET_STATE=0, + AMDGPU_DEBUGFS_REGS2_CMD_SET_STATE_V2, +}; + +enum AMDGPU_DEBUGFS_GPRWAVE_CMDS { + AMDGPU_DEBUGFS_GPRWAVE_CMD_SET_STATE=0, }; +//reg2 interface #define AMDGPU_DEBUGFS_REGS2_IOC_SET_STATE _IOWR(0x20, AMDGPU_DEBUGFS_REGS2_CMD_SET_STATE, struct amdgpu_debugfs_regs2_iocdata) +#define AMDGPU_DEBUGFS_REGS2_IOC_SET_STATE_V2 _IOWR(0x20, AMDGPU_DEBUGFS_REGS2_CMD_SET_STATE_V2, struct amdgpu_debugfs_regs2_iocdata_v2) + +//gprwave interface +#define AMDGPU_DEBUGFS_GPRWAVE_IOC_SET_STATE _IOWR(0x20, AMDGPU_DEBUGFS_GPRWAVE_CMD_SET_STATE, struct amdgpu_debugfs_gprwave_iocdata) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c index 6887109abb139e349ba47b5ed6919ba93852bf4a..b7441654e6fa730c344753eb0e121564449c7a40 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c @@ -96,16 +96,16 @@ */ struct amdgpu_uvd_cs_ctx { struct amdgpu_cs_parser *parser; - unsigned reg, count; - unsigned data0, data1; - unsigned idx; + unsigned int reg, count; + unsigned int data0, data1; + unsigned int idx; struct amdgpu_ib *ib; /* does the IB has a msg command */ bool has_msg_cmd; /* minimum buffer sizes */ - unsigned *buf_sizes; + unsigned int *buf_sizes; }; #ifdef CONFIG_DRM_AMDGPU_SI @@ -186,7 +186,7 @@ int amdgpu_uvd_sw_init(struct amdgpu_device *adev) unsigned long bo_size; const char *fw_name; const struct common_firmware_header *hdr; - unsigned family_id; + unsigned int family_id; int i, j, r; INIT_DELAYED_WORK(&adev->uvd.idle_work, amdgpu_uvd_idle_work_handler); @@ -275,7 +275,7 @@ int amdgpu_uvd_sw_init(struct amdgpu_device *adev) family_id = le32_to_cpu(hdr->ucode_version) & 0xff; if (adev->asic_type < CHIP_VEGA20) { - unsigned version_major, version_minor; + unsigned int version_major, version_minor; version_major = (le32_to_cpu(hdr->ucode_version) >> 24) & 0xff; version_minor = (le32_to_cpu(hdr->ucode_version) >> 8) & 0xff; @@ -420,7 +420,7 @@ int amdgpu_uvd_entity_init(struct amdgpu_device *adev) int amdgpu_uvd_suspend(struct amdgpu_device *adev) { - unsigned size; + unsigned int size; void *ptr; int i, j, idx; bool in_ras_intr = amdgpu_ras_intr_triggered(); @@ -469,7 +469,7 @@ int amdgpu_uvd_suspend(struct amdgpu_device *adev) int amdgpu_uvd_resume(struct amdgpu_device *adev) { - unsigned size; + unsigned int size; void *ptr; int i, idx; @@ -491,7 +491,7 @@ int amdgpu_uvd_resume(struct amdgpu_device *adev) adev->uvd.inst[i].saved_bo = NULL; } else { const struct common_firmware_header *hdr; - unsigned offset; + unsigned int offset; hdr = (const struct common_firmware_header *)adev->uvd.fw->data; if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) { @@ -542,6 +542,7 @@ void amdgpu_uvd_free_handles(struct amdgpu_device *adev, struct drm_file *filp) static void amdgpu_uvd_force_into_uvd_segment(struct amdgpu_bo *abo) { int i; + for (i = 0; i < abo->placement.num_placement; ++i) { abo->placements[i].fpfn = 0 >> PAGE_SHIFT; abo->placements[i].lpfn = (256 * 1024 * 1024) >> PAGE_SHIFT; @@ -579,7 +580,7 @@ static int amdgpu_uvd_cs_pass1(struct amdgpu_uvd_cs_ctx *ctx) r = amdgpu_cs_find_mapping(ctx->parser, addr, &bo, &mapping); if (r) { - DRM_ERROR("Can't find BO for addr 0x%08Lx\n", addr); + DRM_ERROR("Can't find BO for addr 0x%08llx\n", addr); return r; } @@ -589,6 +590,7 @@ static int amdgpu_uvd_cs_pass1(struct amdgpu_uvd_cs_ctx *ctx) if (cmd == 0x0 || cmd == 0x3) { /* yes, force it into VRAM */ uint32_t domain = AMDGPU_GEM_DOMAIN_VRAM; + amdgpu_bo_placement_from_domain(bo, domain); } amdgpu_uvd_force_into_uvd_segment(bo); @@ -609,21 +611,21 @@ static int amdgpu_uvd_cs_pass1(struct amdgpu_uvd_cs_ctx *ctx) * Peek into the decode message and calculate the necessary buffer sizes. */ static int amdgpu_uvd_cs_msg_decode(struct amdgpu_device *adev, uint32_t *msg, - unsigned buf_sizes[]) + unsigned int buf_sizes[]) { - unsigned stream_type = msg[4]; - unsigned width = msg[6]; - unsigned height = msg[7]; - unsigned dpb_size = msg[9]; - unsigned pitch = msg[28]; - unsigned level = msg[57]; + unsigned int stream_type = msg[4]; + unsigned int width = msg[6]; + unsigned int height = msg[7]; + unsigned int dpb_size = msg[9]; + unsigned int pitch = msg[28]; + unsigned int level = msg[57]; - unsigned width_in_mb = width / 16; - unsigned height_in_mb = ALIGN(height / 16, 2); - unsigned fs_in_mb = width_in_mb * height_in_mb; + unsigned int width_in_mb = width / 16; + unsigned int height_in_mb = ALIGN(height / 16, 2); + unsigned int fs_in_mb = width_in_mb * height_in_mb; - unsigned image_size, tmp, min_dpb_size, num_dpb_buffer; - unsigned min_ctx_size = ~0; + unsigned int image_size, tmp, min_dpb_size, num_dpb_buffer; + unsigned int min_ctx_size = ~0; image_size = width * height; image_size += image_size / 2; @@ -631,7 +633,7 @@ static int amdgpu_uvd_cs_msg_decode(struct amdgpu_device *adev, uint32_t *msg, switch (stream_type) { case 0: /* H264 */ - switch(level) { + switch (level) { case 30: num_dpb_buffer = 8100 / fs_in_mb; break; @@ -709,7 +711,7 @@ static int amdgpu_uvd_cs_msg_decode(struct amdgpu_device *adev, uint32_t *msg, break; case 7: /* H264 Perf */ - switch(level) { + switch (level) { case 30: num_dpb_buffer = 8100 / fs_in_mb; break; @@ -742,7 +744,7 @@ static int amdgpu_uvd_cs_msg_decode(struct amdgpu_device *adev, uint32_t *msg, /* reference picture buffer */ min_dpb_size = image_size * num_dpb_buffer; - if (!adev->uvd.use_ctx_buf){ + if (!adev->uvd.use_ctx_buf) { /* macroblock context buffer */ min_dpb_size += width_in_mb * height_in_mb * num_dpb_buffer * 192; @@ -805,7 +807,7 @@ static int amdgpu_uvd_cs_msg_decode(struct amdgpu_device *adev, uint32_t *msg, * Make sure that we don't open up to many sessions. */ static int amdgpu_uvd_cs_msg(struct amdgpu_uvd_cs_ctx *ctx, - struct amdgpu_bo *bo, unsigned offset) + struct amdgpu_bo *bo, unsigned int offset) { struct amdgpu_device *adev = ctx->parser->adev; int32_t *msg, msg_type, handle; @@ -911,7 +913,7 @@ static int amdgpu_uvd_cs_pass2(struct amdgpu_uvd_cs_ctx *ctx) r = amdgpu_cs_find_mapping(ctx->parser, addr, &bo, &mapping); if (r) { - DRM_ERROR("Can't find BO for addr 0x%08Lx\n", addr); + DRM_ERROR("Can't find BO for addr 0x%08llx\n", addr); return r; } @@ -930,7 +932,7 @@ static int amdgpu_uvd_cs_pass2(struct amdgpu_uvd_cs_ctx *ctx) if (cmd < 0x4) { if ((end - start) < ctx->buf_sizes[cmd]) { DRM_ERROR("buffer (%d) to small (%d / %d)!\n", cmd, - (unsigned)(end - start), + (unsigned int)(end - start), ctx->buf_sizes[cmd]); return -EINVAL; } @@ -938,7 +940,7 @@ static int amdgpu_uvd_cs_pass2(struct amdgpu_uvd_cs_ctx *ctx) } else if (cmd == 0x206) { if ((end - start) < ctx->buf_sizes[4]) { DRM_ERROR("buffer (%d) to small (%d / %d)!\n", cmd, - (unsigned)(end - start), + (unsigned int)(end - start), ctx->buf_sizes[4]); return -EINVAL; } @@ -949,14 +951,14 @@ static int amdgpu_uvd_cs_pass2(struct amdgpu_uvd_cs_ctx *ctx) if (!ctx->parser->adev->uvd.address_64_bit) { if ((start >> 28) != ((end - 1) >> 28)) { - DRM_ERROR("reloc %LX-%LX crossing 256MB boundary!\n", + DRM_ERROR("reloc %llx-%llx crossing 256MB boundary!\n", start, end); return -EINVAL; } if ((cmd == 0 || cmd == 0x3) && (start >> 28) != (ctx->parser->adev->uvd.inst->gpu_addr >> 28)) { - DRM_ERROR("msg/fb buffer %LX-%LX out of 256MB segment!\n", + DRM_ERROR("msg/fb buffer %llx-%llx out of 256MB segment!\n", start, end); return -EINVAL; } @@ -990,7 +992,7 @@ static int amdgpu_uvd_cs_reg(struct amdgpu_uvd_cs_ctx *ctx, ctx->idx++; for (i = 0; i <= ctx->count; ++i) { - unsigned reg = ctx->reg + i; + unsigned int reg = ctx->reg + i; if (ctx->idx >= ctx->ib->length_dw) { DRM_ERROR("Register command after end of CS!\n"); @@ -1036,7 +1038,8 @@ static int amdgpu_uvd_cs_packets(struct amdgpu_uvd_cs_ctx *ctx, for (ctx->idx = 0 ; ctx->idx < ctx->ib->length_dw; ) { uint32_t cmd = amdgpu_ib_get_value(ctx->ib, ctx->idx); - unsigned type = CP_PACKET_GET_TYPE(cmd); + unsigned int type = CP_PACKET_GET_TYPE(cmd); + switch (type) { case PACKET_TYPE0: ctx->reg = CP_PACKET0_GET_REG(cmd); @@ -1070,7 +1073,7 @@ int amdgpu_uvd_ring_parse_cs(struct amdgpu_cs_parser *parser, struct amdgpu_ib *ib) { struct amdgpu_uvd_cs_ctx ctx = {}; - unsigned buf_sizes[] = { + unsigned int buf_sizes[] = { [0x00000000] = 2048, [0x00000001] = 0xFFFFFFFF, [0x00000002] = 0xFFFFFFFF, @@ -1185,8 +1188,9 @@ static int amdgpu_uvd_send_msg(struct amdgpu_ring *ring, struct amdgpu_bo *bo, } /* multiple fence commands without any stream commands in between can - crash the vcpu so just try to emmit a dummy create/destroy msg to - avoid this */ + * crash the vcpu so just try to emmit a dummy create/destroy msg to + * avoid this + */ int amdgpu_uvd_get_create_msg(struct amdgpu_ring *ring, uint32_t handle, struct dma_fence **fence) { @@ -1252,15 +1256,14 @@ static void amdgpu_uvd_idle_work_handler(struct work_struct *work) { struct amdgpu_device *adev = container_of(work, struct amdgpu_device, uvd.idle_work.work); - unsigned fences = 0, i, j; + unsigned int fences = 0, i, j; for (i = 0; i < adev->uvd.num_uvd_inst; ++i) { if (adev->uvd.harvest_config & (1 << i)) continue; fences += amdgpu_fence_count_emitted(&adev->uvd.inst[i].ring); - for (j = 0; j < adev->uvd.num_enc_rings; ++j) { + for (j = 0; j < adev->uvd.num_enc_rings; ++j) fences += amdgpu_fence_count_emitted(&adev->uvd.inst[i].ring_enc[j]); - } } if (fences == 0) { @@ -1356,7 +1359,7 @@ int amdgpu_uvd_ring_test_ib(struct amdgpu_ring *ring, long timeout) */ uint32_t amdgpu_uvd_used_handles(struct amdgpu_device *adev) { - unsigned i; + unsigned int i; uint32_t used_handles = 0; for (i = 0; i < adev->uvd.max_handles; ++i) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c index e2b7324a70cbbdf29820b7d97141e9947e7ba25c..1904edf684071675cf5f10e6c354ba69d20f7afe 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c @@ -99,7 +99,7 @@ int amdgpu_vce_sw_init(struct amdgpu_device *adev, unsigned long size) { const char *fw_name; const struct common_firmware_header *hdr; - unsigned ucode_version, version_major, version_minor, binary_id; + unsigned int ucode_version, version_major, version_minor, binary_id; int i, r; switch (adev->asic_type) { @@ -207,7 +207,7 @@ int amdgpu_vce_sw_init(struct amdgpu_device *adev, unsigned long size) */ int amdgpu_vce_sw_fini(struct amdgpu_device *adev) { - unsigned i; + unsigned int i; if (adev->vce.vcpu_bo == NULL) return 0; @@ -286,7 +286,7 @@ int amdgpu_vce_resume(struct amdgpu_device *adev) { void *cpu_addr; const struct common_firmware_header *hdr; - unsigned offset; + unsigned int offset; int r, idx; if (adev->vce.vcpu_bo == NULL) @@ -332,7 +332,7 @@ static void amdgpu_vce_idle_work_handler(struct work_struct *work) { struct amdgpu_device *adev = container_of(work, struct amdgpu_device, vce.idle_work.work); - unsigned i, count = 0; + unsigned int i, count = 0; for (i = 0; i < adev->vce.num_rings; i++) count += amdgpu_fence_count_emitted(&adev->vce.ring[i]); @@ -409,6 +409,7 @@ void amdgpu_vce_free_handles(struct amdgpu_device *adev, struct drm_file *filp) { struct amdgpu_ring *ring = &adev->vce.ring[0]; int i, r; + for (i = 0; i < AMDGPU_MAX_VCE_HANDLES; ++i) { uint32_t handle = atomic_read(&adev->vce.handles[i]); @@ -436,7 +437,7 @@ void amdgpu_vce_free_handles(struct amdgpu_device *adev, struct drm_file *filp) static int amdgpu_vce_get_create_msg(struct amdgpu_ring *ring, uint32_t handle, struct dma_fence **fence) { - const unsigned ib_size_dw = 1024; + const unsigned int ib_size_dw = 1024; struct amdgpu_job *job; struct amdgpu_ib *ib; struct amdgpu_ib ib_msg; @@ -528,7 +529,7 @@ static int amdgpu_vce_get_create_msg(struct amdgpu_ring *ring, uint32_t handle, static int amdgpu_vce_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle, bool direct, struct dma_fence **fence) { - const unsigned ib_size_dw = 1024; + const unsigned int ib_size_dw = 1024; struct amdgpu_job *job; struct amdgpu_ib *ib; struct dma_fence *f = NULL; @@ -596,12 +597,12 @@ static int amdgpu_vce_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle, */ static int amdgpu_vce_validate_bo(struct amdgpu_cs_parser *p, struct amdgpu_ib *ib, int lo, int hi, - unsigned size, int32_t index) + unsigned int size, int32_t index) { int64_t offset = ((uint64_t)size) * ((int64_t)index); struct ttm_operation_ctx ctx = { false, false }; struct amdgpu_bo_va_mapping *mapping; - unsigned i, fpfn, lpfn; + unsigned int i, fpfn, lpfn; struct amdgpu_bo *bo; uint64_t addr; int r; @@ -619,7 +620,7 @@ static int amdgpu_vce_validate_bo(struct amdgpu_cs_parser *p, r = amdgpu_cs_find_mapping(p, addr, &bo, &mapping); if (r) { - DRM_ERROR("Can't find BO for addr 0x%010Lx %d %d %d %d\n", + DRM_ERROR("Can't find BO for addr 0x%010llx %d %d %d %d\n", addr, lo, hi, size, index); return r; } @@ -646,7 +647,7 @@ static int amdgpu_vce_validate_bo(struct amdgpu_cs_parser *p, * Patch relocation inside command stream with real buffer address */ static int amdgpu_vce_cs_reloc(struct amdgpu_cs_parser *p, struct amdgpu_ib *ib, - int lo, int hi, unsigned size, uint32_t index) + int lo, int hi, unsigned int size, uint32_t index) { struct amdgpu_bo_va_mapping *mapping; struct amdgpu_bo *bo; @@ -662,14 +663,14 @@ static int amdgpu_vce_cs_reloc(struct amdgpu_cs_parser *p, struct amdgpu_ib *ib, r = amdgpu_cs_find_mapping(p, addr, &bo, &mapping); if (r) { - DRM_ERROR("Can't find BO for addr 0x%010Lx %d %d %d %d\n", + DRM_ERROR("Can't find BO for addr 0x%010llx %d %d %d %d\n", addr, lo, hi, size, index); return r; } if ((addr + (uint64_t)size) > (mapping->last + 1) * AMDGPU_GPU_PAGE_SIZE) { - DRM_ERROR("BO too small for addr 0x%010Lx %d %d\n", + DRM_ERROR("BO too small for addr 0x%010llx %d %d\n", addr, lo, hi); return -EINVAL; } @@ -692,12 +693,12 @@ static int amdgpu_vce_cs_reloc(struct amdgpu_cs_parser *p, struct amdgpu_ib *ib, * @allocated: allocated a new handle? * * Validates the handle and return the found session index or -EINVAL - * we we don't have another free session index. + * we don't have another free session index. */ static int amdgpu_vce_validate_handle(struct amdgpu_cs_parser *p, uint32_t handle, uint32_t *allocated) { - unsigned i; + unsigned int i; /* validate the handle */ for (i = 0; i < AMDGPU_MAX_VCE_HANDLES; ++i) { @@ -735,14 +736,14 @@ int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, struct amdgpu_job *job, struct amdgpu_ib *ib) { - unsigned fb_idx = 0, bs_idx = 0; + unsigned int fb_idx = 0, bs_idx = 0; int session_idx = -1; uint32_t destroyed = 0; uint32_t created = 0; uint32_t allocated = 0; uint32_t tmp, handle = 0; uint32_t *size = &tmp; - unsigned idx; + unsigned int idx; int i, r = 0; job->vm = NULL; @@ -1084,7 +1085,7 @@ void amdgpu_vce_ring_emit_ib(struct amdgpu_ring *ring, * */ void amdgpu_vce_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq, - unsigned flags) + unsigned int flags) { WARN_ON(flags & AMDGPU_FENCE_FLAG_64BIT); @@ -1106,7 +1107,7 @@ int amdgpu_vce_ring_test_ring(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; uint32_t rptr; - unsigned i; + unsigned int i; int r, timeout = adev->usec_timeout; /* skip ring test for sriov*/ @@ -1171,7 +1172,7 @@ int amdgpu_vce_ring_test_ib(struct amdgpu_ring *ring, long timeout) enum amdgpu_ring_priority_level amdgpu_vce_get_ring_prio(int ring) { - switch(ring) { + switch (ring) { case 0: return AMDGPU_RING_PRIO_0; case 1: diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c index e63fcc58e8e0622f90599733b862a3953f902622..acbef1a24b9c2e3afc7e3e1654cdc2fa0ce65afb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c @@ -56,6 +56,7 @@ #define FIRMWARE_VCN_3_1_2 "amdgpu/vcn_3_1_2.bin" #define FIRMWARE_VCN4_0_0 "amdgpu/vcn_4_0_0.bin" #define FIRMWARE_VCN4_0_2 "amdgpu/vcn_4_0_2.bin" +#define FIRMWARE_VCN4_0_3 "amdgpu/vcn_4_0_3.bin" #define FIRMWARE_VCN4_0_4 "amdgpu/vcn_4_0_4.bin" MODULE_FIRMWARE(FIRMWARE_RAVEN); @@ -77,6 +78,7 @@ MODULE_FIRMWARE(FIRMWARE_YELLOW_CARP); MODULE_FIRMWARE(FIRMWARE_VCN_3_1_2); MODULE_FIRMWARE(FIRMWARE_VCN4_0_0); MODULE_FIRMWARE(FIRMWARE_VCN4_0_2); +MODULE_FIRMWARE(FIRMWARE_VCN4_0_3); MODULE_FIRMWARE(FIRMWARE_VCN4_0_4); static void amdgpu_vcn_idle_work_handler(struct work_struct *work); @@ -167,7 +169,7 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev) if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) bo_size += AMDGPU_GPU_PAGE_ALIGN(le32_to_cpu(hdr->ucode_size_bytes) + 8); - if (adev->ip_versions[UVD_HWIP][0] >= IP_VERSION(4, 0, 0)){ + if (adev->ip_versions[UVD_HWIP][0] >= IP_VERSION(4, 0, 0)) { fw_shared_size = AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_vcn4_fw_shared)); log_offset = offsetof(struct amdgpu_vcn4_fw_shared, fw_log); } else { @@ -233,11 +235,11 @@ int amdgpu_vcn_sw_fini(struct amdgpu_device *adev) if (adev->vcn.harvest_config & (1 << j)) continue; - if (adev->vcn.indirect_sram) { - amdgpu_bo_free_kernel(&adev->vcn.inst[j].dpg_sram_bo, - &adev->vcn.inst[j].dpg_sram_gpu_addr, - (void **)&adev->vcn.inst[j].dpg_sram_cpu_addr); - } + amdgpu_bo_free_kernel( + &adev->vcn.inst[j].dpg_sram_bo, + &adev->vcn.inst[j].dpg_sram_gpu_addr, + (void **)&adev->vcn.inst[j].dpg_sram_cpu_addr); + kvfree(adev->vcn.inst[j].saved_bo); amdgpu_bo_free_kernel(&adev->vcn.inst[j].vcpu_bo, @@ -274,20 +276,19 @@ bool amdgpu_vcn_is_disabled_vcn(struct amdgpu_device *adev, enum vcn_ring_type t bool ret = false; int vcn_config = adev->vcn.vcn_config[vcn_instance]; - if ((type == VCN_ENCODE_RING) && (vcn_config & VCN_BLOCK_ENCODE_DISABLE_MASK)) { + if ((type == VCN_ENCODE_RING) && (vcn_config & VCN_BLOCK_ENCODE_DISABLE_MASK)) ret = true; - } else if ((type == VCN_DECODE_RING) && (vcn_config & VCN_BLOCK_DECODE_DISABLE_MASK)) { + else if ((type == VCN_DECODE_RING) && (vcn_config & VCN_BLOCK_DECODE_DISABLE_MASK)) ret = true; - } else if ((type == VCN_UNIFIED_RING) && (vcn_config & VCN_BLOCK_QUEUE_DISABLE_MASK)) { + else if ((type == VCN_UNIFIED_RING) && (vcn_config & VCN_BLOCK_QUEUE_DISABLE_MASK)) ret = true; - } return ret; } int amdgpu_vcn_suspend(struct amdgpu_device *adev) { - unsigned size; + unsigned int size; void *ptr; int i, idx; @@ -316,7 +317,7 @@ int amdgpu_vcn_suspend(struct amdgpu_device *adev) int amdgpu_vcn_resume(struct amdgpu_device *adev) { - unsigned size; + unsigned int size; void *ptr; int i, idx; @@ -338,7 +339,7 @@ int amdgpu_vcn_resume(struct amdgpu_device *adev) adev->vcn.inst[i].saved_bo = NULL; } else { const struct common_firmware_header *hdr; - unsigned offset; + unsigned int offset; hdr = (const struct common_firmware_header *)adev->vcn.fw->data; if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) { @@ -369,9 +370,8 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work) if (adev->vcn.harvest_config & (1 << j)) continue; - for (i = 0; i < adev->vcn.num_enc_rings; ++i) { + for (i = 0; i < adev->vcn.num_enc_rings; ++i) fence[j] += amdgpu_fence_count_emitted(&adev->vcn.inst[j].ring_enc[i]); - } if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) { struct dpg_pause_state new_state; @@ -458,7 +458,7 @@ int amdgpu_vcn_dec_ring_test_ring(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; uint32_t tmp = 0; - unsigned i; + unsigned int i; int r; /* VCN in SRIOV does not support direct register read/write */ @@ -795,7 +795,7 @@ int amdgpu_vcn_enc_ring_test_ring(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; uint32_t rptr; - unsigned i; + unsigned int i; int r; if (amdgpu_sriov_vf(adev)) @@ -993,11 +993,14 @@ int amdgpu_vcn_enc_ring_test_ib(struct amdgpu_ring *ring, long timeout) int amdgpu_vcn_unified_ring_test_ib(struct amdgpu_ring *ring, long timeout) { + struct amdgpu_device *adev = ring->adev; long r; - r = amdgpu_vcn_enc_ring_test_ib(ring, timeout); - if (r) - goto error; + if (adev->ip_versions[UVD_HWIP][0] != IP_VERSION(4, 0, 3)) { + r = amdgpu_vcn_enc_ring_test_ib(ring, timeout); + if (r) + goto error; + } r = amdgpu_vcn_dec_sw_ring_test_ib(ring, timeout); @@ -1007,7 +1010,7 @@ int amdgpu_vcn_unified_ring_test_ib(struct amdgpu_ring *ring, long timeout) enum amdgpu_ring_priority_level amdgpu_vcn_get_enc_ring_prio(int ring) { - switch(ring) { + switch (ring) { case 0: return AMDGPU_RING_PRIO_0; case 1: @@ -1026,6 +1029,7 @@ void amdgpu_vcn_setup_ucode(struct amdgpu_device *adev) if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { const struct common_firmware_header *hdr; + hdr = (const struct common_firmware_header *)adev->vcn.fw->data; for (i = 0; i < adev->vcn.num_vcn_inst; i++) { @@ -1041,6 +1045,9 @@ void amdgpu_vcn_setup_ucode(struct amdgpu_device *adev) adev->firmware.ucode[idx].fw = adev->vcn.fw; adev->firmware.fw_size += ALIGN(le32_to_cpu(hdr->ucode_size_bytes), PAGE_SIZE); + + if (adev->ip_versions[UVD_HWIP][0] == IP_VERSION(4, 0, 3)) + break; } dev_info(adev->dev, "Will use PSP to load VCN firmware\n"); } @@ -1051,7 +1058,7 @@ void amdgpu_vcn_setup_ucode(struct amdgpu_device *adev) */ #if defined(CONFIG_DEBUG_FS) static ssize_t amdgpu_debugfs_vcn_fwlog_read(struct file *f, char __user *buf, - size_t size, loff_t *pos) + size_t size, loff_t *pos) { struct amdgpu_vcn_inst *vcn; void *log_buf; @@ -1097,7 +1104,7 @@ static ssize_t amdgpu_debugfs_vcn_fwlog_read(struct file *f, char __user *buf, if (read_pos == AMDGPU_VCNFW_LOG_SIZE) read_pos = plog->header_size; if (read_num[i] == copy_to_user((buf + read_bytes), - (log_buf + read_pos), read_num[i])) + (log_buf + read_pos), read_num[i])) return -EFAULT; read_bytes += read_num[i]; @@ -1118,7 +1125,7 @@ static const struct file_operations amdgpu_debugfs_vcnfwlog_fops = { #endif void amdgpu_debugfs_vcn_fwlog_init(struct amdgpu_device *adev, uint8_t i, - struct amdgpu_vcn_inst *vcn) + struct amdgpu_vcn_inst *vcn) { #if defined(CONFIG_DEBUG_FS) struct drm_minor *minor = adev_to_drm(adev)->primary; @@ -1126,7 +1133,7 @@ void amdgpu_debugfs_vcn_fwlog_init(struct amdgpu_device *adev, uint8_t i, char name[32]; sprintf(name, "amdgpu_vcn_%d_fwlog", i); - debugfs_create_file_size(name, S_IFREG | S_IRUGO, root, vcn, + debugfs_create_file_size(name, S_IFREG | 0444, root, vcn, &amdgpu_debugfs_vcnfwlog_fops, AMDGPU_VCNFW_LOG_SIZE); #endif @@ -1140,7 +1147,7 @@ void amdgpu_vcn_fwlog_init(struct amdgpu_vcn_inst *vcn) uint64_t fw_log_gpu_addr = vcn->fw_shared.gpu_addr + vcn->fw_shared.mem_size; volatile struct amdgpu_vcn_fwlog *log_buf = fw_log_cpu_addr; volatile struct amdgpu_fw_shared_fw_logging *fw_log = vcn->fw_shared.cpu_addr - + vcn->fw_shared.log_offset; + + vcn->fw_shared.log_offset; *flag |= cpu_to_le32(AMDGPU_VCN_FW_LOGGING_FLAG); fw_log->is_enabled = 1; fw_log->addr_lo = cpu_to_le32(fw_log_gpu_addr & 0xFFFFFFFF); @@ -1181,6 +1188,31 @@ int amdgpu_vcn_process_poison_irq(struct amdgpu_device *adev, return 0; } +int amdgpu_vcn_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block) +{ + int r, i; + + r = amdgpu_ras_block_late_init(adev, ras_block); + if (r) + return r; + + if (amdgpu_ras_is_supported(adev, ras_block->block)) { + for (i = 0; i < adev->vcn.num_vcn_inst; i++) { + if (adev->vcn.harvest_config & (1 << i)) + continue; + + r = amdgpu_irq_get(adev, &adev->vcn.inst[i].ras_poison_irq, 0); + if (r) + goto late_fini; + } + } + return 0; + +late_fini: + amdgpu_ras_block_late_fini(adev, ras_block); + return r; +} + int amdgpu_vcn_ras_sw_init(struct amdgpu_device *adev) { int err; @@ -1202,7 +1234,7 @@ int amdgpu_vcn_ras_sw_init(struct amdgpu_device *adev) adev->vcn.ras_if = &ras->ras_block.ras_comm; if (!ras->ras_block.ras_late_init) - ras->ras_block.ras_late_init = amdgpu_ras_block_late_init; + ras->ras_block.ras_late_init = amdgpu_vcn_ras_late_init; return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h index c730949ece7d9b27c62b483d1844ddb4aa6d0319..92d5534df5f42848b2a5f8f49b3298fd8c399137 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h @@ -32,7 +32,7 @@ #define AMDGPU_VCN_FIRMWARE_OFFSET 256 #define AMDGPU_VCN_MAX_ENC_RINGS 3 -#define AMDGPU_MAX_VCN_INSTANCES 2 +#define AMDGPU_MAX_VCN_INSTANCES 4 #define AMDGPU_MAX_VCN_ENC_RINGS AMDGPU_VCN_MAX_ENC_RINGS * AMDGPU_MAX_VCN_INSTANCES #define AMDGPU_VCN_HARVEST_VCN0 (1 << 0) @@ -141,18 +141,23 @@ RREG32_SOC15(VCN, inst_idx, mmUVD_DPG_LMA_DATA); \ }) -#define WREG32_SOC15_DPG_MODE(inst_idx, offset, value, mask_en, indirect) \ - do { \ - if (!indirect) { \ - WREG32_SOC15(VCN, inst_idx, mmUVD_DPG_LMA_DATA, value); \ - WREG32_SOC15(VCN, inst_idx, mmUVD_DPG_LMA_CTL, \ - (0x1 << UVD_DPG_LMA_CTL__READ_WRITE__SHIFT | \ - mask_en << UVD_DPG_LMA_CTL__MASK_EN__SHIFT | \ - offset << UVD_DPG_LMA_CTL__READ_WRITE_ADDR__SHIFT)); \ - } else { \ - *adev->vcn.inst[inst_idx].dpg_sram_curr_addr++ = offset; \ - *adev->vcn.inst[inst_idx].dpg_sram_curr_addr++ = value; \ - } \ +#define WREG32_SOC15_DPG_MODE(inst_idx, offset, value, mask_en, indirect) \ + do { \ + if (!indirect) { \ + WREG32_SOC15(VCN, GET_INST(VCN, inst_idx), \ + mmUVD_DPG_LMA_DATA, value); \ + WREG32_SOC15( \ + VCN, GET_INST(VCN, inst_idx), \ + mmUVD_DPG_LMA_CTL, \ + (0x1 << UVD_DPG_LMA_CTL__READ_WRITE__SHIFT | \ + mask_en << UVD_DPG_LMA_CTL__MASK_EN__SHIFT | \ + offset << UVD_DPG_LMA_CTL__READ_WRITE_ADDR__SHIFT)); \ + } else { \ + *adev->vcn.inst[inst_idx].dpg_sram_curr_addr++ = \ + offset; \ + *adev->vcn.inst[inst_idx].dpg_sram_curr_addr++ = \ + value; \ + } \ } while (0) #define AMDGPU_FW_SHARED_FLAG_0_UNIFIED_QUEUE (1 << 2) @@ -234,6 +239,7 @@ struct amdgpu_vcn_inst { struct amdgpu_ring ring_enc[AMDGPU_VCN_MAX_ENC_RINGS]; atomic_t sched_score; struct amdgpu_irq_src irq; + struct amdgpu_irq_src ras_poison_irq; struct amdgpu_vcn_reg external; struct amdgpu_bo *dpg_sram_bo; struct dpg_pause_state pause_state; @@ -242,6 +248,7 @@ struct amdgpu_vcn_inst { uint32_t *dpg_sram_curr_addr; atomic_t dpg_enc_submission_cnt; struct amdgpu_vcn_fw_shared fw_shared; + uint8_t aid_id; }; struct amdgpu_vcn_ras { @@ -271,6 +278,9 @@ struct amdgpu_vcn { struct ras_common_if *ras_if; struct amdgpu_vcn_ras *ras; + + uint16_t inst_mask; + uint8_t num_inst_per_aid; }; struct amdgpu_fw_shared_rb_ptrs_struct { @@ -400,6 +410,8 @@ void amdgpu_debugfs_vcn_fwlog_init(struct amdgpu_device *adev, int amdgpu_vcn_process_poison_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry); +int amdgpu_vcn_ras_late_init(struct amdgpu_device *adev, + struct ras_common_if *ras_block); int amdgpu_vcn_ras_sw_init(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c index f2e2cbaa7fde07b85f7691d0b31a45faa501b93e..25b4d7f0bd35996b4bb69d952dcb3ef2008bf57b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c @@ -56,7 +56,8 @@ void amdgpu_virt_init_setting(struct amdgpu_device *adev) /* enable virtual display */ if (adev->asic_type != CHIP_ALDEBARAN && - adev->asic_type != CHIP_ARCTURUS) { + adev->asic_type != CHIP_ARCTURUS && + ((adev->pdev->class >> 8) != PCI_CLASS_ACCELERATOR_PROCESSING)) { if (adev->mode_info.num_crtc == 0) adev->mode_info.num_crtc = 1; adev->enable_virtual_display = true; @@ -65,16 +66,19 @@ void amdgpu_virt_init_setting(struct amdgpu_device *adev) adev->cg_flags = 0; adev->pg_flags = 0; - /* enable mcbp for sriov asic_type before soc21 */ - amdgpu_mcbp = (adev->asic_type < CHIP_IP_DISCOVERY) ? 1 : 0; + /* enable mcbp for sriov */ + amdgpu_mcbp = 1; + /* Reduce kcq number to 2 to reduce latency */ + if (amdgpu_num_kcq == -1) + amdgpu_num_kcq = 2; } void amdgpu_virt_kiq_reg_write_reg_wait(struct amdgpu_device *adev, uint32_t reg0, uint32_t reg1, uint32_t ref, uint32_t mask) { - struct amdgpu_kiq *kiq = &adev->gfx.kiq; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; struct amdgpu_ring *ring = &kiq->ring; signed long r, cnt = 0; unsigned long flags; @@ -557,7 +561,6 @@ static void amdgpu_virt_populate_vf2pf_ucode_info(struct amdgpu_device *adev) POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_RLC_SRLS, adev->gfx.rlc_srls_fw_version); POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_MEC, adev->gfx.mec_fw_version); POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_MEC2, adev->gfx.mec2_fw_version); - POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_IMU, adev->gfx.imu_fw_version); POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_SOS, adev->psp.sos.fw_version); POPULATE_UCODE_INFO(vf2pf_info, AMD_SRIOV_UCODE_ID_ASD, adev->psp.asd_context.bin_desc.fw_version); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index 3c0310576b3bfb515d9dccc2cc3e6fffd935d80e..143d11afe0e5a97836e0566adea626e80742f07d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -266,6 +266,32 @@ static void amdgpu_vm_bo_done(struct amdgpu_vm_bo_base *vm_bo) spin_unlock(&vm_bo->vm->status_lock); } +/** + * amdgpu_vm_bo_reset_state_machine - reset the vm_bo state machine + * @vm: the VM which state machine to reset + * + * Move all vm_bo object in the VM into a state where they will be updated + * again during validation. + */ +static void amdgpu_vm_bo_reset_state_machine(struct amdgpu_vm *vm) +{ + struct amdgpu_vm_bo_base *vm_bo, *tmp; + + spin_lock(&vm->status_lock); + list_splice_init(&vm->done, &vm->invalidated); + list_for_each_entry(vm_bo, &vm->invalidated, vm_status) + vm_bo->moved = true; + list_for_each_entry_safe(vm_bo, tmp, &vm->idle, vm_status) { + struct amdgpu_bo *bo = vm_bo->bo; + + if (!bo || bo->tbo.type != ttm_bo_type_kernel) + list_move(&vm_bo->vm_status, &vm_bo->vm->moved); + else if (bo->parent) + list_move(&vm_bo->vm_status, &vm_bo->vm->relocated); + } + spin_unlock(&vm->status_lock); +} + /** * amdgpu_vm_bo_base_init - Adds bo to the list of bos associated with the vm * @@ -351,6 +377,58 @@ void amdgpu_vm_move_to_lru_tail(struct amdgpu_device *adev, spin_unlock(&adev->mman.bdev.lru_lock); } +/* Create scheduler entities for page table updates */ +static int amdgpu_vm_init_entities(struct amdgpu_device *adev, + struct amdgpu_vm *vm) +{ + int r; + + r = drm_sched_entity_init(&vm->immediate, DRM_SCHED_PRIORITY_NORMAL, + adev->vm_manager.vm_pte_scheds, + adev->vm_manager.vm_pte_num_scheds, NULL); + if (r) + goto error; + + return drm_sched_entity_init(&vm->delayed, DRM_SCHED_PRIORITY_NORMAL, + adev->vm_manager.vm_pte_scheds, + adev->vm_manager.vm_pte_num_scheds, NULL); + +error: + drm_sched_entity_destroy(&vm->immediate); + return r; +} + +/* Destroy the entities for page table updates again */ +static void amdgpu_vm_fini_entities(struct amdgpu_vm *vm) +{ + drm_sched_entity_destroy(&vm->immediate); + drm_sched_entity_destroy(&vm->delayed); +} + +/** + * amdgpu_vm_generation - return the page table re-generation counter + * @adev: the amdgpu_device + * @vm: optional VM to check, might be NULL + * + * Returns a page table re-generation token to allow checking if submissions + * are still valid to use this VM. The VM parameter might be NULL in which case + * just the VRAM lost counter will be used. + */ +uint64_t amdgpu_vm_generation(struct amdgpu_device *adev, struct amdgpu_vm *vm) +{ + uint64_t result = (u64)atomic_read(&adev->vram_lost_counter) << 32; + + if (!vm) + return result; + + result += vm->generation; + /* Add one if the page tables will be re-generated on next CS */ + if (drm_sched_entity_error(&vm->delayed)) + ++result; + + return result; +} + /** * amdgpu_vm_validate_pt_bos - validate the page table BOs * @@ -373,6 +451,15 @@ int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm, struct amdgpu_bo *bo; int r; + if (drm_sched_entity_error(&vm->delayed)) { + ++vm->generation; + amdgpu_vm_bo_reset_state_machine(vm); + amdgpu_vm_fini_entities(vm); + r = amdgpu_vm_init_entities(adev, vm); + if (r) + return r; + } + spin_lock(&vm->status_lock); while (!list_empty(&vm->evicted)) { bo_base = list_first_entry(&vm->evicted, @@ -920,42 +1007,51 @@ int amdgpu_vm_update_range(struct amdgpu_device *adev, struct amdgpu_vm *vm, return r; } +static void amdgpu_vm_bo_get_memory(struct amdgpu_bo_va *bo_va, + struct amdgpu_mem_stats *stats) +{ + struct amdgpu_vm *vm = bo_va->base.vm; + struct amdgpu_bo *bo = bo_va->base.bo; + + if (!bo) + return; + + /* + * For now ignore BOs which are currently locked and potentially + * changing their location. + */ + if (bo->tbo.base.resv != vm->root.bo->tbo.base.resv && + !dma_resv_trylock(bo->tbo.base.resv)) + return; + + amdgpu_bo_get_memory(bo, stats); + if (bo->tbo.base.resv != vm->root.bo->tbo.base.resv) + dma_resv_unlock(bo->tbo.base.resv); +} + void amdgpu_vm_get_memory(struct amdgpu_vm *vm, struct amdgpu_mem_stats *stats) { struct amdgpu_bo_va *bo_va, *tmp; spin_lock(&vm->status_lock); - list_for_each_entry_safe(bo_va, tmp, &vm->idle, base.vm_status) { - if (!bo_va->base.bo) - continue; - amdgpu_bo_get_memory(bo_va->base.bo, stats); - } - list_for_each_entry_safe(bo_va, tmp, &vm->evicted, base.vm_status) { - if (!bo_va->base.bo) - continue; - amdgpu_bo_get_memory(bo_va->base.bo, stats); - } - list_for_each_entry_safe(bo_va, tmp, &vm->relocated, base.vm_status) { - if (!bo_va->base.bo) - continue; - amdgpu_bo_get_memory(bo_va->base.bo, stats); - } - list_for_each_entry_safe(bo_va, tmp, &vm->moved, base.vm_status) { - if (!bo_va->base.bo) - continue; - amdgpu_bo_get_memory(bo_va->base.bo, stats); - } - list_for_each_entry_safe(bo_va, tmp, &vm->invalidated, base.vm_status) { - if (!bo_va->base.bo) - continue; - amdgpu_bo_get_memory(bo_va->base.bo, stats); - } - list_for_each_entry_safe(bo_va, tmp, &vm->done, base.vm_status) { - if (!bo_va->base.bo) - continue; - amdgpu_bo_get_memory(bo_va->base.bo, stats); - } + list_for_each_entry_safe(bo_va, tmp, &vm->idle, base.vm_status) + amdgpu_vm_bo_get_memory(bo_va, stats); + + list_for_each_entry_safe(bo_va, tmp, &vm->evicted, base.vm_status) + amdgpu_vm_bo_get_memory(bo_va, stats); + + list_for_each_entry_safe(bo_va, tmp, &vm->relocated, base.vm_status) + amdgpu_vm_bo_get_memory(bo_va, stats); + + list_for_each_entry_safe(bo_va, tmp, &vm->moved, base.vm_status) + amdgpu_vm_bo_get_memory(bo_va, stats); + + list_for_each_entry_safe(bo_va, tmp, &vm->invalidated, base.vm_status) + amdgpu_vm_bo_get_memory(bo_va, stats); + + list_for_each_entry_safe(bo_va, tmp, &vm->done, base.vm_status) + amdgpu_vm_bo_get_memory(bo_va, stats); spin_unlock(&vm->status_lock); } @@ -1358,6 +1454,7 @@ struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev, amdgpu_vm_bo_base_init(&bo_va->base, vm, bo); bo_va->ref_count = 1; + bo_va->last_pt_update = dma_fence_get_stub(); INIT_LIST_HEAD(&bo_va->valids); INIT_LIST_HEAD(&bo_va->invalids); @@ -1433,14 +1530,14 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev, uint64_t eaddr; /* validate the parameters */ - if (saddr & ~PAGE_MASK || offset & ~PAGE_MASK || - size == 0 || size & ~PAGE_MASK) + if (saddr & ~PAGE_MASK || offset & ~PAGE_MASK || size & ~PAGE_MASK) + return -EINVAL; + if (saddr + size <= saddr || offset + size <= offset) return -EINVAL; /* make sure object fit at this offset */ eaddr = saddr + size - 1; - if (saddr >= eaddr || - (bo && offset + size > amdgpu_bo_size(bo)) || + if ((bo && offset + size > amdgpu_bo_size(bo)) || (eaddr >= adev->vm_manager.max_pfn << AMDGPU_GPU_PAGE_SHIFT)) return -EINVAL; @@ -1499,14 +1596,14 @@ int amdgpu_vm_bo_replace_map(struct amdgpu_device *adev, int r; /* validate the parameters */ - if (saddr & ~PAGE_MASK || offset & ~PAGE_MASK || - size == 0 || size & ~PAGE_MASK) + if (saddr & ~PAGE_MASK || offset & ~PAGE_MASK || size & ~PAGE_MASK) + return -EINVAL; + if (saddr + size <= saddr || offset + size <= offset) return -EINVAL; /* make sure object fit at this offset */ eaddr = saddr + size - 1; - if (saddr >= eaddr || - (bo && offset + size > amdgpu_bo_size(bo)) || + if ((bo && offset + size > amdgpu_bo_size(bo)) || (eaddr >= adev->vm_manager.max_pfn << AMDGPU_GPU_PAGE_SHIFT)) return -EINVAL; @@ -2038,19 +2135,10 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm) INIT_LIST_HEAD(&vm->pt_freed); INIT_WORK(&vm->pt_free_work, amdgpu_vm_pt_free_work); - /* create scheduler entities for page table updates */ - r = drm_sched_entity_init(&vm->immediate, DRM_SCHED_PRIORITY_NORMAL, - adev->vm_manager.vm_pte_scheds, - adev->vm_manager.vm_pte_num_scheds, NULL); + r = amdgpu_vm_init_entities(adev, vm); if (r) return r; - r = drm_sched_entity_init(&vm->delayed, DRM_SCHED_PRIORITY_NORMAL, - adev->vm_manager.vm_pte_scheds, - adev->vm_manager.vm_pte_num_scheds, NULL); - if (r) - goto error_free_immediate; - vm->pte_support_ats = false; vm->is_compute_context = false; @@ -2067,9 +2155,11 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm) vm->update_funcs = &amdgpu_vm_cpu_funcs; else vm->update_funcs = &amdgpu_vm_sdma_funcs; - vm->last_update = NULL; + + vm->last_update = dma_fence_get_stub(); vm->last_unlocked = dma_fence_get_stub(); vm->last_tlb_flush = dma_fence_get_stub(); + vm->generation = 0; mutex_init(&vm->eviction_lock); vm->evicting = false; @@ -2110,10 +2200,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm) error_free_delayed: dma_fence_put(vm->last_tlb_flush); dma_fence_put(vm->last_unlocked); - drm_sched_entity_destroy(&vm->delayed); - -error_free_immediate: - drm_sched_entity_destroy(&vm->immediate); + amdgpu_vm_fini_entities(vm); return r; } @@ -2192,7 +2279,7 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm) goto unreserve_bo; dma_fence_put(vm->last_update); - vm->last_update = NULL; + vm->last_update = dma_fence_get_stub(); vm->is_compute_context = true; /* Free the shadow bo for compute VM */ @@ -2266,8 +2353,7 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm) amdgpu_bo_unref(&root); WARN_ON(vm->root.bo); - drm_sched_entity_destroy(&vm->immediate); - drm_sched_entity_destroy(&vm->delayed); + amdgpu_vm_fini_entities(vm); if (!RB_EMPTY_ROOT(&vm->va.rb_root)) { dev_err(adev->dev, "still active bo inside vm\n"); @@ -2282,8 +2368,14 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm) } dma_fence_put(vm->last_update); - for (i = 0; i < AMDGPU_MAX_VMHUBS; i++) - amdgpu_vmid_free_reserved(adev, vm, i); + + for (i = 0; i < AMDGPU_MAX_VMHUBS; i++) { + if (vm->reserved_vmid[i]) { + amdgpu_vmid_free_reserved(adev, i); + vm->reserved_vmid[i] = false; + } + } + } /** @@ -2366,18 +2458,25 @@ int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) union drm_amdgpu_vm *args = data; struct amdgpu_device *adev = drm_to_adev(dev); struct amdgpu_fpriv *fpriv = filp->driver_priv; - int r; + + /* No valid flags defined yet */ + if (args->in.flags) + return -EINVAL; switch (args->in.op) { case AMDGPU_VM_OP_RESERVE_VMID: /* We only have requirement to reserve vmid from gfxhub */ - r = amdgpu_vmid_alloc_reserved(adev, &fpriv->vm, - AMDGPU_GFXHUB_0); - if (r) - return r; + if (!fpriv->vm.reserved_vmid[AMDGPU_GFXHUB(0)]) { + amdgpu_vmid_alloc_reserved(adev, AMDGPU_GFXHUB(0)); + fpriv->vm.reserved_vmid[AMDGPU_GFXHUB(0)] = true; + } + break; case AMDGPU_VM_OP_UNRESERVE_VMID: - amdgpu_vmid_free_reserved(adev, &fpriv->vm, AMDGPU_GFXHUB_0); + if (fpriv->vm.reserved_vmid[AMDGPU_GFXHUB(0)]) { + amdgpu_vmid_free_reserved(adev, AMDGPU_GFXHUB(0)); + fpriv->vm.reserved_vmid[AMDGPU_GFXHUB(0)] = false; + } break; default: return -EINVAL; @@ -2432,6 +2531,9 @@ void amdgpu_vm_set_task_info(struct amdgpu_vm *vm) * amdgpu_vm_handle_fault - graceful handling of VM faults. * @adev: amdgpu device pointer * @pasid: PASID of the VM + * @vmid: VMID, only used for GFX 9.4.3. + * @node_id: Node_id received in IH cookie. Only applicable for + * GFX 9.4.3. * @addr: Address of the fault * @write_fault: true is write fault, false is read fault * @@ -2439,7 +2541,8 @@ void amdgpu_vm_set_task_info(struct amdgpu_vm *vm) * shouldn't be reported any more. */ bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid, - uint64_t addr, bool write_fault) + u32 vmid, u32 node_id, uint64_t addr, + bool write_fault) { bool is_compute_context = false; struct amdgpu_bo *root; @@ -2463,8 +2566,8 @@ bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid, addr /= AMDGPU_GPU_PAGE_SIZE; - if (is_compute_context && - !svm_range_restore_pages(adev, pasid, addr, write_fault)) { + if (is_compute_context && !svm_range_restore_pages(adev, pasid, vmid, + node_id, addr, write_fault)) { amdgpu_bo_unref(&root); return true; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h index 6f085f0b4ef3f818b449d2a09d2f5c5f0fff9786..9c85d494f2a24df5919c435728816bdfb3be8936 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h @@ -111,11 +111,14 @@ struct amdgpu_mem_stats; /* Reserve 4MB VRAM for page tables */ #define AMDGPU_VM_RESERVED_VRAM (8ULL << 20) -/* max number of VMHUB */ -#define AMDGPU_MAX_VMHUBS 3 -#define AMDGPU_GFXHUB_0 0 -#define AMDGPU_MMHUB_0 1 -#define AMDGPU_MMHUB_1 2 +/* + * max number of VMHUB + * layout: max 8 GFXHUB + 4 MMHUB0 + 1 MMHUB1 + */ +#define AMDGPU_MAX_VMHUBS 13 +#define AMDGPU_GFXHUB(x) (x) +#define AMDGPU_MMHUB0(x) (8 + x) +#define AMDGPU_MMHUB1(x) (8 + 4 + x) /* Reserve 2MB at top/bottom of address space for kernel use */ #define AMDGPU_VA_RESERVED_SIZE (2ULL << 20) @@ -292,6 +295,9 @@ struct amdgpu_vm { atomic64_t tlb_seq; struct dma_fence *last_tlb_flush; + /* How many times we had to re-generate the page tables */ + uint64_t generation; + /* Last unlocked submission to the scheduler entities */ struct dma_fence *last_unlocked; @@ -326,6 +332,9 @@ struct amdgpu_vm { struct ttm_lru_bulk_move lru_bulk_move; /* Flag to indicate if VM is used for compute */ bool is_compute_context; + + /* Memory partition number, -1 means any partition */ + int8_t mem_id; }; struct amdgpu_vm_manager { @@ -391,6 +400,7 @@ void amdgpu_vm_get_pd_bo(struct amdgpu_vm *vm, struct list_head *validated, struct amdgpu_bo_list_entry *entry); bool amdgpu_vm_ready(struct amdgpu_vm *vm); +uint64_t amdgpu_vm_generation(struct amdgpu_device *adev, struct amdgpu_vm *vm); int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm, int (*callback)(void *p, struct amdgpu_bo *bo), void *param); @@ -452,7 +462,8 @@ void amdgpu_vm_check_compute_bug(struct amdgpu_device *adev); void amdgpu_vm_get_task_info(struct amdgpu_device *adev, u32 pasid, struct amdgpu_task_info *task_info); bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid, - uint64_t addr, bool write_fault); + u32 vmid, u32 node_id, uint64_t addr, + bool write_fault); void amdgpu_vm_set_task_info(struct amdgpu_vm *vm); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c index df63dc3bca18cd2829575ef6a592e21dcaee0db0..dea1a64be44d092b78d34692968bc68aa9fb3c2c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c @@ -502,6 +502,7 @@ int amdgpu_vm_pt_clear(struct amdgpu_device *adev, struct amdgpu_vm *vm, int amdgpu_vm_pt_create(struct amdgpu_device *adev, struct amdgpu_vm *vm, int level, bool immediate, struct amdgpu_bo_vm **vmbo) { + struct amdgpu_fpriv *fpriv = container_of(vm, struct amdgpu_fpriv, vm); struct amdgpu_bo_param bp; struct amdgpu_bo *bo; struct dma_resv *resv; @@ -512,7 +513,12 @@ int amdgpu_vm_pt_create(struct amdgpu_device *adev, struct amdgpu_vm *vm, bp.size = amdgpu_vm_pt_size(adev, level); bp.byte_align = AMDGPU_GPU_PAGE_SIZE; - bp.domain = AMDGPU_GEM_DOMAIN_VRAM; + + if (!adev->gmc.is_app_apu) + bp.domain = AMDGPU_GEM_DOMAIN_VRAM; + else + bp.domain = AMDGPU_GEM_DOMAIN_GTT; + bp.domain = amdgpu_bo_get_preferred_domain(adev, bp.domain); bp.flags = AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS | AMDGPU_GEM_CREATE_CPU_GTT_USWC; @@ -529,6 +535,8 @@ int amdgpu_vm_pt_create(struct amdgpu_device *adev, struct amdgpu_vm *vm, bp.type = ttm_bo_type_kernel; bp.no_wait_gpu = immediate; + bp.xcp_id_plus1 = fpriv->xcp_id == ~0 ? 0 : fpriv->xcp_id + 1; + if (vm->root.bo) bp.resv = vm->root.bo->tbo.base.resv; @@ -553,6 +561,7 @@ int amdgpu_vm_pt_create(struct amdgpu_device *adev, struct amdgpu_vm *vm, bp.type = ttm_bo_type_kernel; bp.resv = bo->tbo.base.resv; bp.bo_ptr_size = sizeof(struct amdgpu_bo); + bp.xcp_id_plus1 = fpriv->xcp_id == ~0 ? 0 : fpriv->xcp_id + 1; r = amdgpu_bo_create(adev, &bp, &(*vmbo)->shadow); @@ -564,7 +573,6 @@ int amdgpu_vm_pt_create(struct amdgpu_device *adev, struct amdgpu_vm *vm, return r; } - (*vmbo)->shadow->parent = amdgpu_bo_ref(bo); amdgpu_bo_add_to_shadow_list(*vmbo); return 0; @@ -781,13 +789,14 @@ static void amdgpu_vm_pte_update_flags(struct amdgpu_vm_update_params *params, uint64_t pe, uint64_t addr, unsigned int count, uint32_t incr, uint64_t flags) - { + struct amdgpu_device *adev = params->adev; + if (level != AMDGPU_VM_PTB) { flags |= AMDGPU_PDE_PTE; - amdgpu_gmc_get_vm_pde(params->adev, level, &addr, &flags); + amdgpu_gmc_get_vm_pde(adev, level, &addr, &flags); - } else if (params->adev->asic_type >= CHIP_VEGA10 && + } else if (adev->asic_type >= CHIP_VEGA10 && !(flags & AMDGPU_PTE_VALID) && !(flags & AMDGPU_PTE_PRT)) { @@ -795,6 +804,21 @@ static void amdgpu_vm_pte_update_flags(struct amdgpu_vm_update_params *params, flags |= AMDGPU_PTE_EXECUTABLE; } + /* APUs mapping system memory may need different MTYPEs on different + * NUMA nodes. Only do this for contiguous ranges that can be assumed + * to be on the same NUMA node. + */ + if ((flags & AMDGPU_PTE_SYSTEM) && (adev->flags & AMD_IS_APU) && + adev->gmc.gmc_funcs->override_vm_pte_flags && + num_possible_nodes() > 1) { + if (!params->pages_addr) + amdgpu_gmc_override_vm_pte_flags(adev, params->vm, + addr, &flags); + else + dev_dbg(adev->dev, + "override_vm_pte_flags skipped: non-contiguous\n"); + } + params->vm->update_funcs->update(params, pt, pe, addr, count, incr, flags); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c index 43d6a9d6a5384f886b4460a539bb97450cf1a0e2..c7085a747b03b78012d33b9966eb2a10a6f558d6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c @@ -370,6 +370,45 @@ int amdgpu_vram_mgr_query_page_status(struct amdgpu_vram_mgr *mgr, return ret; } +static void amdgpu_dummy_vram_mgr_debug(struct ttm_resource_manager *man, + struct drm_printer *printer) +{ + DRM_DEBUG_DRIVER("Dummy vram mgr debug\n"); +} + +static bool amdgpu_dummy_vram_mgr_compatible(struct ttm_resource_manager *man, + struct ttm_resource *res, + const struct ttm_place *place, + size_t size) +{ + DRM_DEBUG_DRIVER("Dummy vram mgr compatible\n"); + return false; +} + +static bool amdgpu_dummy_vram_mgr_intersects(struct ttm_resource_manager *man, + struct ttm_resource *res, + const struct ttm_place *place, + size_t size) +{ + DRM_DEBUG_DRIVER("Dummy vram mgr intersects\n"); + return true; +} + +static void amdgpu_dummy_vram_mgr_del(struct ttm_resource_manager *man, + struct ttm_resource *res) +{ + DRM_DEBUG_DRIVER("Dummy vram mgr deleted\n"); +} + +static int amdgpu_dummy_vram_mgr_new(struct ttm_resource_manager *man, + struct ttm_buffer_object *tbo, + const struct ttm_place *place, + struct ttm_resource **res) +{ + DRM_DEBUG_DRIVER("Dummy vram mgr new\n"); + return -ENOSPC; +} + /** * amdgpu_vram_mgr_new - allocate new ranges * @@ -800,7 +839,7 @@ static void amdgpu_vram_mgr_debug(struct ttm_resource_manager *man, { struct amdgpu_vram_mgr *mgr = to_vram_mgr(man); struct drm_buddy *mm = &mgr->mm; - struct drm_buddy_block *block; + struct amdgpu_vram_reservation *rsv; drm_printf(printer, " vis usage:%llu\n", amdgpu_vram_mgr_vis_usage(mgr)); @@ -812,11 +851,20 @@ static void amdgpu_vram_mgr_debug(struct ttm_resource_manager *man, drm_buddy_print(mm, printer); drm_printf(printer, "reserved:\n"); - list_for_each_entry(block, &mgr->reserved_pages, link) - drm_buddy_block_print(mm, block, printer); + list_for_each_entry(rsv, &mgr->reserved_pages, blocks) + drm_printf(printer, "%#018llx-%#018llx: %llu\n", + rsv->start, rsv->start + rsv->size, rsv->size); mutex_unlock(&mgr->lock); } +static const struct ttm_resource_manager_func amdgpu_dummy_vram_mgr_func = { + .alloc = amdgpu_dummy_vram_mgr_new, + .free = amdgpu_dummy_vram_mgr_del, + .intersects = amdgpu_dummy_vram_mgr_intersects, + .compatible = amdgpu_dummy_vram_mgr_compatible, + .debug = amdgpu_dummy_vram_mgr_debug +}; + static const struct ttm_resource_manager_func amdgpu_vram_mgr_func = { .alloc = amdgpu_vram_mgr_new, .free = amdgpu_vram_mgr_del, @@ -841,17 +889,22 @@ int amdgpu_vram_mgr_init(struct amdgpu_device *adev) ttm_resource_manager_init(man, &adev->mman.bdev, adev->gmc.real_vram_size); - man->func = &amdgpu_vram_mgr_func; - - err = drm_buddy_init(&mgr->mm, man->size, PAGE_SIZE); - if (err) - return err; - mutex_init(&mgr->lock); INIT_LIST_HEAD(&mgr->reservations_pending); INIT_LIST_HEAD(&mgr->reserved_pages); mgr->default_page_size = PAGE_SIZE; + if (!adev->gmc.is_app_apu) { + man->func = &amdgpu_vram_mgr_func; + + err = drm_buddy_init(&mgr->mm, man->size, PAGE_SIZE); + if (err) + return err; + } else { + man->func = &amdgpu_dummy_vram_mgr_func; + DRM_INFO("Setup dummy vram mgr\n"); + } + ttm_set_driver_manager(&adev->mman.bdev, TTM_PL_VRAM, &mgr->manager); ttm_resource_manager_set_used(man, true); return 0; @@ -886,7 +939,8 @@ void amdgpu_vram_mgr_fini(struct amdgpu_device *adev) drm_buddy_free_list(&mgr->mm, &rsv->allocated); kfree(rsv); } - drm_buddy_fini(&mgr->mm); + if (!adev->gmc.is_app_apu) + drm_buddy_fini(&mgr->mm); mutex_unlock(&mgr->lock); ttm_resource_manager_cleanup(man); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c new file mode 100644 index 0000000000000000000000000000000000000000..d733fa6e7477e75f4f71c1895dd5e68953b868f8 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c @@ -0,0 +1,399 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +#include "amdgpu.h" +#include "amdgpu_xcp.h" +#include "amdgpu_drv.h" + +#include +#include "../amdxcp/amdgpu_xcp_drv.h" + +static int __amdgpu_xcp_run(struct amdgpu_xcp_mgr *xcp_mgr, + struct amdgpu_xcp_ip *xcp_ip, int xcp_state) +{ + int (*run_func)(void *handle, uint32_t inst_mask); + int ret = 0; + + if (!xcp_ip || !xcp_ip->valid || !xcp_ip->ip_funcs) + return 0; + + run_func = NULL; + + switch (xcp_state) { + case AMDGPU_XCP_PREPARE_SUSPEND: + run_func = xcp_ip->ip_funcs->prepare_suspend; + break; + case AMDGPU_XCP_SUSPEND: + run_func = xcp_ip->ip_funcs->suspend; + break; + case AMDGPU_XCP_PREPARE_RESUME: + run_func = xcp_ip->ip_funcs->prepare_resume; + break; + case AMDGPU_XCP_RESUME: + run_func = xcp_ip->ip_funcs->resume; + break; + } + + if (run_func) + ret = run_func(xcp_mgr->adev, xcp_ip->inst_mask); + + return ret; +} + +static int amdgpu_xcp_run_transition(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id, + int state) +{ + struct amdgpu_xcp_ip *xcp_ip; + struct amdgpu_xcp *xcp; + int i, ret; + + if (xcp_id >= MAX_XCP || !xcp_mgr->xcp[xcp_id].valid) + return -EINVAL; + + xcp = &xcp_mgr->xcp[xcp_id]; + for (i = 0; i < AMDGPU_XCP_MAX_BLOCKS; ++i) { + xcp_ip = &xcp->ip[i]; + ret = __amdgpu_xcp_run(xcp_mgr, xcp_ip, state); + if (ret) + break; + } + + return ret; +} + +int amdgpu_xcp_prepare_suspend(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id) +{ + return amdgpu_xcp_run_transition(xcp_mgr, xcp_id, + AMDGPU_XCP_PREPARE_SUSPEND); +} + +int amdgpu_xcp_suspend(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id) +{ + return amdgpu_xcp_run_transition(xcp_mgr, xcp_id, AMDGPU_XCP_SUSPEND); +} + +int amdgpu_xcp_prepare_resume(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id) +{ + return amdgpu_xcp_run_transition(xcp_mgr, xcp_id, + AMDGPU_XCP_PREPARE_RESUME); +} + +int amdgpu_xcp_resume(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id) +{ + return amdgpu_xcp_run_transition(xcp_mgr, xcp_id, AMDGPU_XCP_RESUME); +} + +static void __amdgpu_xcp_add_block(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id, + struct amdgpu_xcp_ip *ip) +{ + struct amdgpu_xcp *xcp; + + if (!ip) + return; + + xcp = &xcp_mgr->xcp[xcp_id]; + xcp->ip[ip->ip_id] = *ip; + xcp->ip[ip->ip_id].valid = true; + + xcp->valid = true; +} + +int amdgpu_xcp_init(struct amdgpu_xcp_mgr *xcp_mgr, int num_xcps, int mode) +{ + struct amdgpu_device *adev = xcp_mgr->adev; + struct amdgpu_xcp_ip ip; + uint8_t mem_id; + int i, j, ret; + + if (!num_xcps || num_xcps > MAX_XCP) + return -EINVAL; + + xcp_mgr->mode = mode; + + for (i = 0; i < MAX_XCP; ++i) + xcp_mgr->xcp[i].valid = false; + + for (i = 0; i < num_xcps; ++i) { + for (j = AMDGPU_XCP_GFXHUB; j < AMDGPU_XCP_MAX_BLOCKS; ++j) { + ret = xcp_mgr->funcs->get_ip_details(xcp_mgr, i, j, + &ip); + if (ret) + continue; + + __amdgpu_xcp_add_block(xcp_mgr, i, &ip); + } + + xcp_mgr->xcp[i].id = i; + + if (xcp_mgr->funcs->get_xcp_mem_id) { + ret = xcp_mgr->funcs->get_xcp_mem_id( + xcp_mgr, &xcp_mgr->xcp[i], &mem_id); + if (ret) + continue; + else + xcp_mgr->xcp[i].mem_id = mem_id; + } + } + + xcp_mgr->num_xcps = num_xcps; + amdgpu_xcp_update_partition_sched_list(adev); + + xcp_mgr->num_xcp_per_mem_partition = num_xcps / xcp_mgr->adev->gmc.num_mem_partitions; + return 0; +} + +int amdgpu_xcp_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, int mode) +{ + int ret, curr_mode, num_xcps = 0; + + if (!xcp_mgr || mode == AMDGPU_XCP_MODE_NONE) + return -EINVAL; + + if (xcp_mgr->mode == mode) + return 0; + + if (!xcp_mgr->funcs || !xcp_mgr->funcs->switch_partition_mode) + return 0; + + mutex_lock(&xcp_mgr->xcp_lock); + + curr_mode = xcp_mgr->mode; + /* State set to transient mode */ + xcp_mgr->mode = AMDGPU_XCP_MODE_TRANS; + + ret = xcp_mgr->funcs->switch_partition_mode(xcp_mgr, mode, &num_xcps); + + if (ret) { + /* Failed, get whatever mode it's at now */ + if (xcp_mgr->funcs->query_partition_mode) + xcp_mgr->mode = amdgpu_xcp_query_partition_mode( + xcp_mgr, AMDGPU_XCP_FL_LOCKED); + else + xcp_mgr->mode = curr_mode; + + goto out; + } + +out: + mutex_unlock(&xcp_mgr->xcp_lock); + + return ret; +} + +int amdgpu_xcp_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags) +{ + int mode; + + if (xcp_mgr->mode == AMDGPU_XCP_MODE_NONE) + return xcp_mgr->mode; + + if (!xcp_mgr->funcs || !xcp_mgr->funcs->query_partition_mode) + return xcp_mgr->mode; + + if (!(flags & AMDGPU_XCP_FL_LOCKED)) + mutex_lock(&xcp_mgr->xcp_lock); + mode = xcp_mgr->funcs->query_partition_mode(xcp_mgr); + if (xcp_mgr->mode != AMDGPU_XCP_MODE_TRANS && mode != xcp_mgr->mode) + dev_WARN( + xcp_mgr->adev->dev, + "Cached partition mode %d not matching with device mode %d", + xcp_mgr->mode, mode); + + if (!(flags & AMDGPU_XCP_FL_LOCKED)) + mutex_unlock(&xcp_mgr->xcp_lock); + + return mode; +} + +static int amdgpu_xcp_dev_alloc(struct amdgpu_device *adev) +{ + struct drm_device *p_ddev; + struct drm_device *ddev; + int i, ret; + + ddev = adev_to_drm(adev); + + for (i = 0; i < MAX_XCP; i++) { + ret = amdgpu_xcp_drm_dev_alloc(&p_ddev); + if (ret) + return ret; + + /* Redirect all IOCTLs to the primary device */ + adev->xcp_mgr->xcp[i].rdev = p_ddev->render->dev; + adev->xcp_mgr->xcp[i].pdev = p_ddev->primary->dev; + adev->xcp_mgr->xcp[i].driver = (struct drm_driver *)p_ddev->driver; + adev->xcp_mgr->xcp[i].vma_offset_manager = p_ddev->vma_offset_manager; + p_ddev->render->dev = ddev; + p_ddev->primary->dev = ddev; + p_ddev->vma_offset_manager = ddev->vma_offset_manager; + p_ddev->driver = &amdgpu_partition_driver; + adev->xcp_mgr->xcp[i].ddev = p_ddev; + } + + return 0; +} + +int amdgpu_xcp_mgr_init(struct amdgpu_device *adev, int init_mode, + int init_num_xcps, + struct amdgpu_xcp_mgr_funcs *xcp_funcs) +{ + struct amdgpu_xcp_mgr *xcp_mgr; + + if (!xcp_funcs || !xcp_funcs->switch_partition_mode || + !xcp_funcs->get_ip_details) + return -EINVAL; + + xcp_mgr = kzalloc(sizeof(*xcp_mgr), GFP_KERNEL); + + if (!xcp_mgr) + return -ENOMEM; + + xcp_mgr->adev = adev; + xcp_mgr->funcs = xcp_funcs; + xcp_mgr->mode = init_mode; + mutex_init(&xcp_mgr->xcp_lock); + + if (init_mode != AMDGPU_XCP_MODE_NONE) + amdgpu_xcp_init(xcp_mgr, init_num_xcps, init_mode); + + adev->xcp_mgr = xcp_mgr; + + return amdgpu_xcp_dev_alloc(adev); +} + +int amdgpu_xcp_get_partition(struct amdgpu_xcp_mgr *xcp_mgr, + enum AMDGPU_XCP_IP_BLOCK ip, int instance) +{ + struct amdgpu_xcp *xcp; + int i, id_mask = 0; + + if (ip >= AMDGPU_XCP_MAX_BLOCKS) + return -EINVAL; + + for (i = 0; i < xcp_mgr->num_xcps; ++i) { + xcp = &xcp_mgr->xcp[i]; + if ((xcp->valid) && (xcp->ip[ip].valid) && + (xcp->ip[ip].inst_mask & BIT(instance))) + id_mask |= BIT(i); + } + + if (!id_mask) + id_mask = -ENXIO; + + return id_mask; +} + +int amdgpu_xcp_get_inst_details(struct amdgpu_xcp *xcp, + enum AMDGPU_XCP_IP_BLOCK ip, + uint32_t *inst_mask) +{ + if (!xcp->valid || !inst_mask || !(xcp->ip[ip].valid)) + return -EINVAL; + + *inst_mask = xcp->ip[ip].inst_mask; + + return 0; +} + +int amdgpu_xcp_dev_register(struct amdgpu_device *adev, + const struct pci_device_id *ent) +{ + int i, ret; + + if (!adev->xcp_mgr) + return 0; + + for (i = 0; i < MAX_XCP; i++) { + ret = drm_dev_register(adev->xcp_mgr->xcp[i].ddev, ent->driver_data); + if (ret) + return ret; + } + + return 0; +} + +void amdgpu_xcp_dev_unplug(struct amdgpu_device *adev) +{ + struct drm_device *p_ddev; + int i; + + if (!adev->xcp_mgr) + return; + + for (i = 0; i < MAX_XCP; i++) { + p_ddev = adev->xcp_mgr->xcp[i].ddev; + drm_dev_unplug(p_ddev); + p_ddev->render->dev = adev->xcp_mgr->xcp[i].rdev; + p_ddev->primary->dev = adev->xcp_mgr->xcp[i].pdev; + p_ddev->driver = adev->xcp_mgr->xcp[i].driver; + p_ddev->vma_offset_manager = adev->xcp_mgr->xcp[i].vma_offset_manager; + } +} + +int amdgpu_xcp_open_device(struct amdgpu_device *adev, + struct amdgpu_fpriv *fpriv, + struct drm_file *file_priv) +{ + int i; + + if (!adev->xcp_mgr) + return 0; + + fpriv->xcp_id = ~0; + for (i = 0; i < MAX_XCP; ++i) { + if (!adev->xcp_mgr->xcp[i].ddev) + break; + + if (file_priv->minor == adev->xcp_mgr->xcp[i].ddev->render) { + if (adev->xcp_mgr->xcp[i].valid == FALSE) { + dev_err(adev->dev, "renderD%d partition %d not valid!", + file_priv->minor->index, i); + return -ENOENT; + } + dev_dbg(adev->dev, "renderD%d partition %d opened!", + file_priv->minor->index, i); + fpriv->xcp_id = i; + break; + } + } + + fpriv->vm.mem_id = fpriv->xcp_id == ~0 ? -1 : + adev->xcp_mgr->xcp[fpriv->xcp_id].mem_id; + return 0; +} + +void amdgpu_xcp_release_sched(struct amdgpu_device *adev, + struct amdgpu_ctx_entity *entity) +{ + struct drm_gpu_scheduler *sched; + struct amdgpu_ring *ring; + + if (!adev->xcp_mgr) + return; + + sched = entity->entity.rq->sched; + if (sched->ready) { + ring = to_amdgpu_ring(entity->entity.rq->sched); + atomic_dec(&adev->xcp_mgr->xcp[ring->xcp_id].ref_cnt); + } +} + diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h new file mode 100644 index 0000000000000000000000000000000000000000..0f8026d64ea518b7f7f390fd5373ac62b9b6c59e --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h @@ -0,0 +1,182 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef AMDGPU_XCP_H +#define AMDGPU_XCP_H + +#include +#include + +#include "amdgpu_ctx.h" + +#define MAX_XCP 8 + +#define AMDGPU_XCP_MODE_NONE -1 +#define AMDGPU_XCP_MODE_TRANS -2 + +#define AMDGPU_XCP_FL_NONE 0 +#define AMDGPU_XCP_FL_LOCKED (1 << 0) + +struct amdgpu_fpriv; + +enum AMDGPU_XCP_IP_BLOCK { + AMDGPU_XCP_GFXHUB, + AMDGPU_XCP_GFX, + AMDGPU_XCP_SDMA, + AMDGPU_XCP_VCN, + AMDGPU_XCP_MAX_BLOCKS +}; + +enum AMDGPU_XCP_STATE { + AMDGPU_XCP_PREPARE_SUSPEND, + AMDGPU_XCP_SUSPEND, + AMDGPU_XCP_PREPARE_RESUME, + AMDGPU_XCP_RESUME, +}; + +struct amdgpu_xcp_ip_funcs { + int (*prepare_suspend)(void *handle, uint32_t inst_mask); + int (*suspend)(void *handle, uint32_t inst_mask); + int (*prepare_resume)(void *handle, uint32_t inst_mask); + int (*resume)(void *handle, uint32_t inst_mask); +}; + +struct amdgpu_xcp_ip { + struct amdgpu_xcp_ip_funcs *ip_funcs; + uint32_t inst_mask; + + enum AMDGPU_XCP_IP_BLOCK ip_id; + bool valid; +}; + +struct amdgpu_xcp { + struct amdgpu_xcp_ip ip[AMDGPU_XCP_MAX_BLOCKS]; + + uint8_t id; + uint8_t mem_id; + bool valid; + atomic_t ref_cnt; + struct drm_device *ddev; + struct drm_device *rdev; + struct drm_device *pdev; + struct drm_driver *driver; + struct drm_vma_offset_manager *vma_offset_manager; + struct amdgpu_sched gpu_sched[AMDGPU_HW_IP_NUM][AMDGPU_RING_PRIO_MAX]; +}; + +struct amdgpu_xcp_mgr { + struct amdgpu_device *adev; + struct mutex xcp_lock; + struct amdgpu_xcp_mgr_funcs *funcs; + + struct amdgpu_xcp xcp[MAX_XCP]; + uint8_t num_xcps; + int8_t mode; + + /* Used to determine KFD memory size limits per XCP */ + unsigned int num_xcp_per_mem_partition; +}; + +struct amdgpu_xcp_mgr_funcs { + int (*switch_partition_mode)(struct amdgpu_xcp_mgr *xcp_mgr, int mode, + int *num_xcps); + int (*query_partition_mode)(struct amdgpu_xcp_mgr *xcp_mgr); + int (*get_ip_details)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id, + enum AMDGPU_XCP_IP_BLOCK ip_id, + struct amdgpu_xcp_ip *ip); + int (*get_xcp_mem_id)(struct amdgpu_xcp_mgr *xcp_mgr, + struct amdgpu_xcp *xcp, uint8_t *mem_id); + + int (*prepare_suspend)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); + int (*suspend)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); + int (*prepare_resume)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); + int (*resume)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); + int (*select_scheds)(struct amdgpu_device *adev, + u32 hw_ip, u32 hw_prio, struct amdgpu_fpriv *fpriv, + unsigned int *num_scheds, struct drm_gpu_scheduler ***scheds); + int (*update_partition_sched_list)(struct amdgpu_device *adev); +}; + +int amdgpu_xcp_prepare_suspend(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); +int amdgpu_xcp_suspend(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); +int amdgpu_xcp_prepare_resume(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); +int amdgpu_xcp_resume(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); + +int amdgpu_xcp_mgr_init(struct amdgpu_device *adev, int init_mode, + int init_xcps, struct amdgpu_xcp_mgr_funcs *xcp_funcs); +int amdgpu_xcp_init(struct amdgpu_xcp_mgr *xcp_mgr, int num_xcps, int mode); +int amdgpu_xcp_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags); +int amdgpu_xcp_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, int mode); +int amdgpu_xcp_get_partition(struct amdgpu_xcp_mgr *xcp_mgr, + enum AMDGPU_XCP_IP_BLOCK ip, int instance); + +int amdgpu_xcp_get_inst_details(struct amdgpu_xcp *xcp, + enum AMDGPU_XCP_IP_BLOCK ip, + uint32_t *inst_mask); + +int amdgpu_xcp_dev_register(struct amdgpu_device *adev, + const struct pci_device_id *ent); +void amdgpu_xcp_dev_unplug(struct amdgpu_device *adev); +int amdgpu_xcp_open_device(struct amdgpu_device *adev, + struct amdgpu_fpriv *fpriv, + struct drm_file *file_priv); +void amdgpu_xcp_release_sched(struct amdgpu_device *adev, + struct amdgpu_ctx_entity *entity); + +#define amdgpu_xcp_select_scheds(adev, e, c, d, x, y) \ + ((adev)->xcp_mgr && (adev)->xcp_mgr->funcs && \ + (adev)->xcp_mgr->funcs->select_scheds ? \ + (adev)->xcp_mgr->funcs->select_scheds((adev), (e), (c), (d), (x), (y)) : -ENOENT) +#define amdgpu_xcp_update_partition_sched_list(adev) \ + ((adev)->xcp_mgr && (adev)->xcp_mgr->funcs && \ + (adev)->xcp_mgr->funcs->update_partition_sched_list ? \ + (adev)->xcp_mgr->funcs->update_partition_sched_list(adev) : 0) + +static inline int amdgpu_xcp_get_num_xcp(struct amdgpu_xcp_mgr *xcp_mgr) +{ + if (!xcp_mgr) + return 1; + else + return xcp_mgr->num_xcps; +} + +static inline struct amdgpu_xcp * +amdgpu_get_next_xcp(struct amdgpu_xcp_mgr *xcp_mgr, int *from) +{ + if (!xcp_mgr) + return NULL; + + while (*from < MAX_XCP) { + if (xcp_mgr->xcp[*from].valid) + return &xcp_mgr->xcp[*from]; + ++(*from); + } + + return NULL; +} + +#define for_each_xcp(xcp_mgr, xcp, i) \ + for (i = 0, xcp = amdgpu_get_next_xcp(xcp_mgr, &i); xcp; \ + xcp = amdgpu_get_next_xcp(xcp_mgr, &i)) + +#endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index 439925477fb89f0e6ca7f1a499322fba85838b27..03dc59cbe8aacd62a21b8436f9350f524bea1578 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -325,6 +325,36 @@ static ssize_t amdgpu_xgmi_show_device_id(struct device *dev, } +static ssize_t amdgpu_xgmi_show_num_hops(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = drm_to_adev(ddev); + struct psp_xgmi_topology_info *top = &adev->psp.xgmi_context.top_info; + int i; + + for (i = 0; i < top->num_nodes; i++) + sprintf(buf + 3 * i, "%02x ", top->nodes[i].num_hops); + + return sysfs_emit(buf, "%s\n", buf); +} + +static ssize_t amdgpu_xgmi_show_num_links(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = drm_to_adev(ddev); + struct psp_xgmi_topology_info *top = &adev->psp.xgmi_context.top_info; + int i; + + for (i = 0; i < top->num_nodes; i++) + sprintf(buf + 3 * i, "%02x ", top->nodes[i].num_links); + + return sysfs_emit(buf, "%s\n", buf); +} + #define AMDGPU_XGMI_SET_FICAA(o) ((o) | 0x456801) static ssize_t amdgpu_xgmi_show_error(struct device *dev, struct device_attribute *attr, @@ -361,6 +391,8 @@ static ssize_t amdgpu_xgmi_show_error(struct device *dev, static DEVICE_ATTR(xgmi_device_id, S_IRUGO, amdgpu_xgmi_show_device_id, NULL); static DEVICE_ATTR(xgmi_error, S_IRUGO, amdgpu_xgmi_show_error, NULL); +static DEVICE_ATTR(xgmi_num_hops, S_IRUGO, amdgpu_xgmi_show_num_hops, NULL); +static DEVICE_ATTR(xgmi_num_links, S_IRUGO, amdgpu_xgmi_show_num_links, NULL); static int amdgpu_xgmi_sysfs_add_dev_info(struct amdgpu_device *adev, struct amdgpu_hive_info *hive) @@ -380,6 +412,15 @@ static int amdgpu_xgmi_sysfs_add_dev_info(struct amdgpu_device *adev, if (ret) pr_err("failed to create xgmi_error\n"); + /* Create xgmi num hops file */ + ret = device_create_file(adev->dev, &dev_attr_xgmi_num_hops); + if (ret) + pr_err("failed to create xgmi_num_hops\n"); + + /* Create xgmi num links file */ + ret = device_create_file(adev->dev, &dev_attr_xgmi_num_links); + if (ret) + pr_err("failed to create xgmi_num_links\n"); /* Create sysfs link to hive info folder on the first device */ if (hive->kobj.parent != (&adev->dev->kobj)) { @@ -407,6 +448,9 @@ static int amdgpu_xgmi_sysfs_add_dev_info(struct amdgpu_device *adev, remove_file: device_remove_file(adev->dev, &dev_attr_xgmi_device_id); + device_remove_file(adev->dev, &dev_attr_xgmi_error); + device_remove_file(adev->dev, &dev_attr_xgmi_num_hops); + device_remove_file(adev->dev, &dev_attr_xgmi_num_links); success: return ret; @@ -420,6 +464,8 @@ static void amdgpu_xgmi_sysfs_rem_dev_info(struct amdgpu_device *adev, device_remove_file(adev->dev, &dev_attr_xgmi_device_id); device_remove_file(adev->dev, &dev_attr_xgmi_error); + device_remove_file(adev->dev, &dev_attr_xgmi_num_hops); + device_remove_file(adev->dev, &dev_attr_xgmi_num_links); if (hive->kobj.parent != (&adev->dev->kobj)) sysfs_remove_link(&adev->dev->kobj,"xgmi_hive_info"); @@ -1014,7 +1060,8 @@ static void amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev, } /* Trigger XGMI/WAFL error */ -static int amdgpu_ras_error_inject_xgmi(struct amdgpu_device *adev, void *inject_if) +static int amdgpu_ras_error_inject_xgmi(struct amdgpu_device *adev, + void *inject_if, uint32_t instance_mask) { int ret = 0; struct ta_ras_trigger_error_input *block_info = @@ -1026,7 +1073,7 @@ static int amdgpu_ras_error_inject_xgmi(struct amdgpu_device *adev, void *injec if (amdgpu_dpm_allow_xgmi_power_down(adev, false)) dev_warn(adev->dev, "Failed to disallow XGMI power down"); - ret = psp_ras_trigger_error(&adev->psp, block_info); + ret = psp_ras_trigger_error(&adev->psp, block_info, instance_mask); if (amdgpu_ras_intr_triggered()) return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h b/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h index 24d42d24e6a01e033a84edbbfe561e0878dd6049..104a5ad8397da73126fb5e8aeb962542a552e642 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h @@ -70,7 +70,6 @@ enum amd_sriov_ucode_engine_id { AMD_SRIOV_UCODE_ID_RLC_SRLS, AMD_SRIOV_UCODE_ID_MEC, AMD_SRIOV_UCODE_ID_MEC2, - AMD_SRIOV_UCODE_ID_IMU, AMD_SRIOV_UCODE_ID_SOS, AMD_SRIOV_UCODE_ID_ASD, AMD_SRIOV_UCODE_ID_TA_RAS, diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c new file mode 100644 index 0000000000000000000000000000000000000000..16471b81a1f5d54c5f37b1bedbbd38e539e56c19 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c @@ -0,0 +1,658 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +#include "amdgpu.h" +#include "soc15.h" + +#include "soc15_common.h" +#include "amdgpu_xcp.h" +#include "gfx_v9_4_3.h" +#include "gfxhub_v1_2.h" +#include "sdma_v4_4_2.h" + +#define XCP_INST_MASK(num_inst, xcp_id) \ + (num_inst ? GENMASK(num_inst - 1, 0) << (xcp_id * num_inst) : 0) + +#define AMDGPU_XCP_OPS_KFD (1 << 0) + +void aqua_vanjaram_doorbell_index_init(struct amdgpu_device *adev) +{ + int i; + + adev->doorbell_index.kiq = AMDGPU_DOORBELL_LAYOUT1_KIQ_START; + + adev->doorbell_index.mec_ring0 = AMDGPU_DOORBELL_LAYOUT1_MEC_RING_START; + + adev->doorbell_index.userqueue_start = AMDGPU_DOORBELL_LAYOUT1_USERQUEUE_START; + adev->doorbell_index.userqueue_end = AMDGPU_DOORBELL_LAYOUT1_USERQUEUE_END; + adev->doorbell_index.xcc_doorbell_range = AMDGPU_DOORBELL_LAYOUT1_XCC_RANGE; + + adev->doorbell_index.sdma_doorbell_range = 20; + for (i = 0; i < adev->sdma.num_instances; i++) + adev->doorbell_index.sdma_engine[i] = + AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_START + + i * (adev->doorbell_index.sdma_doorbell_range >> 1); + + adev->doorbell_index.ih = AMDGPU_DOORBELL_LAYOUT1_IH; + adev->doorbell_index.vcn.vcn_ring0_1 = AMDGPU_DOORBELL_LAYOUT1_VCN_START; + + adev->doorbell_index.first_non_cp = AMDGPU_DOORBELL_LAYOUT1_FIRST_NON_CP; + adev->doorbell_index.last_non_cp = AMDGPU_DOORBELL_LAYOUT1_LAST_NON_CP; + + adev->doorbell_index.max_assignment = AMDGPU_DOORBELL_LAYOUT1_MAX_ASSIGNMENT << 1; +} + +static void aqua_vanjaram_set_xcp_id(struct amdgpu_device *adev, + uint32_t inst_idx, struct amdgpu_ring *ring) +{ + int xcp_id; + enum AMDGPU_XCP_IP_BLOCK ip_blk; + uint32_t inst_mask; + + ring->xcp_id = ~0; + if (adev->xcp_mgr->mode == AMDGPU_XCP_MODE_NONE) + return; + + inst_mask = 1 << inst_idx; + + switch (ring->funcs->type) { + case AMDGPU_HW_IP_GFX: + case AMDGPU_RING_TYPE_COMPUTE: + case AMDGPU_RING_TYPE_KIQ: + ip_blk = AMDGPU_XCP_GFX; + break; + case AMDGPU_RING_TYPE_SDMA: + ip_blk = AMDGPU_XCP_SDMA; + break; + case AMDGPU_RING_TYPE_VCN_ENC: + case AMDGPU_RING_TYPE_VCN_JPEG: + ip_blk = AMDGPU_XCP_VCN; + if (adev->xcp_mgr->mode == AMDGPU_CPX_PARTITION_MODE) + inst_mask = 1 << (inst_idx * 2); + break; + default: + DRM_ERROR("Not support ring type %d!", ring->funcs->type); + return; + } + + for (xcp_id = 0; xcp_id < adev->xcp_mgr->num_xcps; xcp_id++) { + if (adev->xcp_mgr->xcp[xcp_id].ip[ip_blk].inst_mask & inst_mask) { + ring->xcp_id = xcp_id; + break; + } + } +} + +static void aqua_vanjaram_xcp_gpu_sched_update( + struct amdgpu_device *adev, + struct amdgpu_ring *ring, + unsigned int sel_xcp_id) +{ + unsigned int *num_gpu_sched; + + num_gpu_sched = &adev->xcp_mgr->xcp[sel_xcp_id] + .gpu_sched[ring->funcs->type][ring->hw_prio].num_scheds; + adev->xcp_mgr->xcp[sel_xcp_id].gpu_sched[ring->funcs->type][ring->hw_prio] + .sched[(*num_gpu_sched)++] = &ring->sched; + DRM_DEBUG("%s :[%d] gpu_sched[%d][%d] = %d", ring->name, + sel_xcp_id, ring->funcs->type, + ring->hw_prio, *num_gpu_sched); +} + +static int aqua_vanjaram_xcp_sched_list_update( + struct amdgpu_device *adev) +{ + struct amdgpu_ring *ring; + int i; + + for (i = 0; i < MAX_XCP; i++) { + atomic_set(&adev->xcp_mgr->xcp[i].ref_cnt, 0); + memset(adev->xcp_mgr->xcp[i].gpu_sched, 0, sizeof(adev->xcp_mgr->xcp->gpu_sched)); + } + + if (adev->xcp_mgr->mode == AMDGPU_XCP_MODE_NONE) + return 0; + + for (i = 0; i < AMDGPU_MAX_RINGS; i++) { + ring = adev->rings[i]; + if (!ring || !ring->sched.ready) + continue; + + aqua_vanjaram_xcp_gpu_sched_update(adev, ring, ring->xcp_id); + + /* VCN is shared by two partitions under CPX MODE */ + if ((ring->funcs->type == AMDGPU_RING_TYPE_VCN_ENC || + ring->funcs->type == AMDGPU_RING_TYPE_VCN_JPEG) && + adev->xcp_mgr->mode == AMDGPU_CPX_PARTITION_MODE) + aqua_vanjaram_xcp_gpu_sched_update(adev, ring, ring->xcp_id + 1); + } + + return 0; +} + +static int aqua_vanjaram_update_partition_sched_list(struct amdgpu_device *adev) +{ + int i; + + for (i = 0; i < adev->num_rings; i++) { + struct amdgpu_ring *ring = adev->rings[i]; + + if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE || + ring->funcs->type == AMDGPU_RING_TYPE_KIQ) + aqua_vanjaram_set_xcp_id(adev, ring->xcc_id, ring); + else + aqua_vanjaram_set_xcp_id(adev, ring->me, ring); + } + + return aqua_vanjaram_xcp_sched_list_update(adev); +} + +static int aqua_vanjaram_select_scheds( + struct amdgpu_device *adev, + u32 hw_ip, + u32 hw_prio, + struct amdgpu_fpriv *fpriv, + unsigned int *num_scheds, + struct drm_gpu_scheduler ***scheds) +{ + u32 sel_xcp_id; + int i; + + if (fpriv->xcp_id == ~0) { + u32 least_ref_cnt = ~0; + + fpriv->xcp_id = 0; + for (i = 0; i < adev->xcp_mgr->num_xcps; i++) { + u32 total_ref_cnt; + + total_ref_cnt = atomic_read(&adev->xcp_mgr->xcp[i].ref_cnt); + if (total_ref_cnt < least_ref_cnt) { + fpriv->xcp_id = i; + least_ref_cnt = total_ref_cnt; + } + } + } + sel_xcp_id = fpriv->xcp_id; + + if (adev->xcp_mgr->xcp[sel_xcp_id].gpu_sched[hw_ip][hw_prio].num_scheds) { + *num_scheds = adev->xcp_mgr->xcp[fpriv->xcp_id].gpu_sched[hw_ip][hw_prio].num_scheds; + *scheds = adev->xcp_mgr->xcp[fpriv->xcp_id].gpu_sched[hw_ip][hw_prio].sched; + atomic_inc(&adev->xcp_mgr->xcp[sel_xcp_id].ref_cnt); + DRM_DEBUG("Selected partition #%d", sel_xcp_id); + } else { + DRM_ERROR("Failed to schedule partition #%d.", sel_xcp_id); + return -ENOENT; + } + + return 0; +} + +static int8_t aqua_vanjaram_logical_to_dev_inst(struct amdgpu_device *adev, + enum amd_hw_ip_block_type block, + int8_t inst) +{ + int8_t dev_inst; + + switch (block) { + case GC_HWIP: + case SDMA0_HWIP: + /* Both JPEG and VCN as JPEG is only alias of VCN */ + case VCN_HWIP: + dev_inst = adev->ip_map.dev_inst[block][inst]; + break; + default: + /* For rest of the IPs, no look up required. + * Assume 'logical instance == physical instance' for all configs. */ + dev_inst = inst; + break; + } + + return dev_inst; +} + +static uint32_t aqua_vanjaram_logical_to_dev_mask(struct amdgpu_device *adev, + enum amd_hw_ip_block_type block, + uint32_t mask) +{ + uint32_t dev_mask = 0; + int8_t log_inst, dev_inst; + + while (mask) { + log_inst = ffs(mask) - 1; + dev_inst = aqua_vanjaram_logical_to_dev_inst(adev, block, log_inst); + dev_mask |= (1 << dev_inst); + mask &= ~(1 << log_inst); + } + + return dev_mask; +} + +static void aqua_vanjaram_populate_ip_map(struct amdgpu_device *adev, + enum amd_hw_ip_block_type ip_block, + uint32_t inst_mask) +{ + int l = 0, i; + + while (inst_mask) { + i = ffs(inst_mask) - 1; + adev->ip_map.dev_inst[ip_block][l++] = i; + inst_mask &= ~(1 << i); + } + for (; l < HWIP_MAX_INSTANCE; l++) + adev->ip_map.dev_inst[ip_block][l] = -1; +} + +void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev) +{ + u32 ip_map[][2] = { + { GC_HWIP, adev->gfx.xcc_mask }, + { SDMA0_HWIP, adev->sdma.sdma_mask }, + { VCN_HWIP, adev->vcn.inst_mask }, + }; + int i; + + for (i = 0; i < ARRAY_SIZE(ip_map); ++i) + aqua_vanjaram_populate_ip_map(adev, ip_map[i][0], ip_map[i][1]); + + adev->ip_map.logical_to_dev_inst = aqua_vanjaram_logical_to_dev_inst; + adev->ip_map.logical_to_dev_mask = aqua_vanjaram_logical_to_dev_mask; +} + +/* Fixed pattern for smn addressing on different AIDs: + * bit[34]: indicate cross AID access + * bit[33:32]: indicate target AID id + * AID id range is 0 ~ 3 as maximum AID number is 4. + */ +u64 aqua_vanjaram_encode_ext_smn_addressing(int ext_id) +{ + u64 ext_offset; + + /* local routing and bit[34:32] will be zeros */ + if (ext_id == 0) + return 0; + + /* Initiated from host, accessing to all non-zero aids are cross traffic */ + ext_offset = ((u64)(ext_id & 0x3) << 32) | (1ULL << 34); + + return ext_offset; +} + +static int aqua_vanjaram_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr) +{ + enum amdgpu_gfx_partition mode = AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE; + struct amdgpu_device *adev = xcp_mgr->adev; + + if (adev->nbio.funcs->get_compute_partition_mode) + mode = adev->nbio.funcs->get_compute_partition_mode(adev); + + return mode; +} + +static int __aqua_vanjaram_get_xcc_per_xcp(struct amdgpu_xcp_mgr *xcp_mgr, int mode) +{ + int num_xcc, num_xcc_per_xcp = 0; + + num_xcc = NUM_XCC(xcp_mgr->adev->gfx.xcc_mask); + + switch (mode) { + case AMDGPU_SPX_PARTITION_MODE: + num_xcc_per_xcp = num_xcc; + break; + case AMDGPU_DPX_PARTITION_MODE: + num_xcc_per_xcp = num_xcc / 2; + break; + case AMDGPU_TPX_PARTITION_MODE: + num_xcc_per_xcp = num_xcc / 3; + break; + case AMDGPU_QPX_PARTITION_MODE: + num_xcc_per_xcp = num_xcc / 4; + break; + case AMDGPU_CPX_PARTITION_MODE: + num_xcc_per_xcp = 1; + break; + } + + return num_xcc_per_xcp; +} + +static int __aqua_vanjaram_get_xcp_ip_info(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id, + enum AMDGPU_XCP_IP_BLOCK ip_id, + struct amdgpu_xcp_ip *ip) +{ + struct amdgpu_device *adev = xcp_mgr->adev; + int num_xcc_xcp, num_sdma_xcp, num_vcn_xcp; + int num_sdma, num_vcn; + + num_sdma = adev->sdma.num_instances; + num_vcn = adev->vcn.num_vcn_inst; + + switch (xcp_mgr->mode) { + case AMDGPU_SPX_PARTITION_MODE: + num_sdma_xcp = num_sdma; + num_vcn_xcp = num_vcn; + break; + case AMDGPU_DPX_PARTITION_MODE: + num_sdma_xcp = num_sdma / 2; + num_vcn_xcp = num_vcn / 2; + break; + case AMDGPU_TPX_PARTITION_MODE: + num_sdma_xcp = num_sdma / 3; + num_vcn_xcp = num_vcn / 3; + break; + case AMDGPU_QPX_PARTITION_MODE: + num_sdma_xcp = num_sdma / 4; + num_vcn_xcp = num_vcn / 4; + break; + case AMDGPU_CPX_PARTITION_MODE: + num_sdma_xcp = 2; + num_vcn_xcp = num_vcn ? 1 : 0; + break; + default: + return -EINVAL; + } + + num_xcc_xcp = adev->gfx.num_xcc_per_xcp; + + switch (ip_id) { + case AMDGPU_XCP_GFXHUB: + ip->inst_mask = XCP_INST_MASK(num_xcc_xcp, xcp_id); + ip->ip_funcs = &gfxhub_v1_2_xcp_funcs; + break; + case AMDGPU_XCP_GFX: + ip->inst_mask = XCP_INST_MASK(num_xcc_xcp, xcp_id); + ip->ip_funcs = &gfx_v9_4_3_xcp_funcs; + break; + case AMDGPU_XCP_SDMA: + ip->inst_mask = XCP_INST_MASK(num_sdma_xcp, xcp_id); + ip->ip_funcs = &sdma_v4_4_2_xcp_funcs; + break; + case AMDGPU_XCP_VCN: + ip->inst_mask = XCP_INST_MASK(num_vcn_xcp, xcp_id); + /* TODO : Assign IP funcs */ + break; + default: + return -EINVAL; + } + + ip->ip_id = ip_id; + + return 0; +} + +static enum amdgpu_gfx_partition +__aqua_vanjaram_get_auto_mode(struct amdgpu_xcp_mgr *xcp_mgr) +{ + struct amdgpu_device *adev = xcp_mgr->adev; + int num_xcc; + + num_xcc = NUM_XCC(xcp_mgr->adev->gfx.xcc_mask); + + if (adev->gmc.num_mem_partitions == 1) + return AMDGPU_SPX_PARTITION_MODE; + + if (adev->gmc.num_mem_partitions == num_xcc) + return AMDGPU_CPX_PARTITION_MODE; + + if (adev->gmc.num_mem_partitions == num_xcc / 2) + return (adev->flags & AMD_IS_APU) ? AMDGPU_TPX_PARTITION_MODE : + AMDGPU_QPX_PARTITION_MODE; + + if (adev->gmc.num_mem_partitions == 2 && !(adev->flags & AMD_IS_APU)) + return AMDGPU_DPX_PARTITION_MODE; + + return AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE; +} + +static bool __aqua_vanjaram_is_valid_mode(struct amdgpu_xcp_mgr *xcp_mgr, + enum amdgpu_gfx_partition mode) +{ + struct amdgpu_device *adev = xcp_mgr->adev; + int num_xcc, num_xccs_per_xcp; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + switch (mode) { + case AMDGPU_SPX_PARTITION_MODE: + return adev->gmc.num_mem_partitions == 1 && num_xcc > 0; + case AMDGPU_DPX_PARTITION_MODE: + return adev->gmc.num_mem_partitions != 8 && (num_xcc % 4) == 0; + case AMDGPU_TPX_PARTITION_MODE: + return (adev->gmc.num_mem_partitions == 1 || + adev->gmc.num_mem_partitions == 3) && + ((num_xcc % 3) == 0); + case AMDGPU_QPX_PARTITION_MODE: + num_xccs_per_xcp = num_xcc / 4; + return (adev->gmc.num_mem_partitions == 1 || + adev->gmc.num_mem_partitions == 4) && + (num_xccs_per_xcp >= 2); + case AMDGPU_CPX_PARTITION_MODE: + return ((num_xcc > 1) && + (adev->gmc.num_mem_partitions == 1 || adev->gmc.num_mem_partitions == 4) && + (num_xcc % adev->gmc.num_mem_partitions) == 0); + default: + return false; + } + + return false; +} + +static int __aqua_vanjaram_pre_partition_switch(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags) +{ + /* TODO: + * Stop user queues and threads, and make sure GPU is empty of work. + */ + + if (flags & AMDGPU_XCP_OPS_KFD) + amdgpu_amdkfd_device_fini_sw(xcp_mgr->adev); + + return 0; +} + +static int __aqua_vanjaram_post_partition_switch(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags) +{ + int ret = 0; + + if (flags & AMDGPU_XCP_OPS_KFD) { + amdgpu_amdkfd_device_probe(xcp_mgr->adev); + amdgpu_amdkfd_device_init(xcp_mgr->adev); + /* If KFD init failed, return failure */ + if (!xcp_mgr->adev->kfd.init_complete) + ret = -EIO; + } + + return ret; +} + +static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, + int mode, int *num_xcps) +{ + int num_xcc_per_xcp, num_xcc, ret; + struct amdgpu_device *adev; + u32 flags = 0; + + adev = xcp_mgr->adev; + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + + if (mode == AMDGPU_AUTO_COMPUTE_PARTITION_MODE) { + mode = __aqua_vanjaram_get_auto_mode(xcp_mgr); + } else if (!__aqua_vanjaram_is_valid_mode(xcp_mgr, mode)) { + dev_err(adev->dev, + "Invalid compute partition mode requested, requested: %s, available memory partitions: %d", + amdgpu_gfx_compute_mode_desc(mode), adev->gmc.num_mem_partitions); + return -EINVAL; + } + + if (adev->kfd.init_complete) + flags |= AMDGPU_XCP_OPS_KFD; + + if (flags & AMDGPU_XCP_OPS_KFD) { + ret = amdgpu_amdkfd_check_and_lock_kfd(adev); + if (ret) + goto out; + } + + ret = __aqua_vanjaram_pre_partition_switch(xcp_mgr, flags); + if (ret) + goto unlock; + + num_xcc_per_xcp = __aqua_vanjaram_get_xcc_per_xcp(xcp_mgr, mode); + if (adev->gfx.funcs->switch_partition_mode) + adev->gfx.funcs->switch_partition_mode(xcp_mgr->adev, + num_xcc_per_xcp); + + /* Init info about new xcps */ + *num_xcps = num_xcc / num_xcc_per_xcp; + amdgpu_xcp_init(xcp_mgr, *num_xcps, mode); + + ret = __aqua_vanjaram_post_partition_switch(xcp_mgr, flags); +unlock: + if (flags & AMDGPU_XCP_OPS_KFD) + amdgpu_amdkfd_unlock_kfd(adev); +out: + return ret; +} + +static int __aqua_vanjaram_get_xcp_mem_id(struct amdgpu_device *adev, + int xcc_id, uint8_t *mem_id) +{ + /* memory/spatial modes validation check is already done */ + *mem_id = xcc_id / adev->gfx.num_xcc_per_xcp; + *mem_id /= adev->xcp_mgr->num_xcp_per_mem_partition; + + return 0; +} + +static int aqua_vanjaram_get_xcp_mem_id(struct amdgpu_xcp_mgr *xcp_mgr, + struct amdgpu_xcp *xcp, uint8_t *mem_id) +{ + struct amdgpu_numa_info numa_info; + struct amdgpu_device *adev; + uint32_t xcc_mask; + int r, i, xcc_id; + + adev = xcp_mgr->adev; + /* TODO: BIOS is not returning the right info now + * Check on this later + */ + /* + if (adev->gmc.gmc_funcs->query_mem_partition_mode) + mode = adev->gmc.gmc_funcs->query_mem_partition_mode(adev); + */ + if (adev->gmc.num_mem_partitions == 1) { + /* Only one range */ + *mem_id = 0; + return 0; + } + + r = amdgpu_xcp_get_inst_details(xcp, AMDGPU_XCP_GFX, &xcc_mask); + if (r || !xcc_mask) + return -EINVAL; + + xcc_id = ffs(xcc_mask) - 1; + if (!adev->gmc.is_app_apu) + return __aqua_vanjaram_get_xcp_mem_id(adev, xcc_id, mem_id); + + r = amdgpu_acpi_get_mem_info(adev, xcc_id, &numa_info); + + if (r) + return r; + + r = -EINVAL; + for (i = 0; i < adev->gmc.num_mem_partitions; ++i) { + if (adev->gmc.mem_partitions[i].numa.node == numa_info.nid) { + *mem_id = i; + r = 0; + break; + } + } + + return r; +} + +static int aqua_vanjaram_get_xcp_ip_details(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id, + enum AMDGPU_XCP_IP_BLOCK ip_id, + struct amdgpu_xcp_ip *ip) +{ + if (!ip) + return -EINVAL; + + return __aqua_vanjaram_get_xcp_ip_info(xcp_mgr, xcp_id, ip_id, ip); +} + +struct amdgpu_xcp_mgr_funcs aqua_vanjaram_xcp_funcs = { + .switch_partition_mode = &aqua_vanjaram_switch_partition_mode, + .query_partition_mode = &aqua_vanjaram_query_partition_mode, + .get_ip_details = &aqua_vanjaram_get_xcp_ip_details, + .get_xcp_mem_id = &aqua_vanjaram_get_xcp_mem_id, + .select_scheds = &aqua_vanjaram_select_scheds, + .update_partition_sched_list = &aqua_vanjaram_update_partition_sched_list +}; + +static int aqua_vanjaram_xcp_mgr_init(struct amdgpu_device *adev) +{ + int ret; + + ret = amdgpu_xcp_mgr_init(adev, AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE, 1, + &aqua_vanjaram_xcp_funcs); + if (ret) + return ret; + + /* TODO: Default memory node affinity init */ + + return ret; +} + +int aqua_vanjaram_init_soc_config(struct amdgpu_device *adev) +{ + u32 mask, inst_mask = adev->sdma.sdma_mask; + int ret, i; + + /* generally 1 AID supports 4 instances */ + adev->sdma.num_inst_per_aid = 4; + adev->sdma.num_instances = NUM_SDMA(adev->sdma.sdma_mask); + + adev->aid_mask = i = 1; + inst_mask >>= adev->sdma.num_inst_per_aid; + + for (mask = (1 << adev->sdma.num_inst_per_aid) - 1; inst_mask; + inst_mask >>= adev->sdma.num_inst_per_aid, ++i) { + if ((inst_mask & mask) == mask) + adev->aid_mask |= (1 << i); + } + + /* Harvest config is not used for aqua vanjaram. VCN and JPEGs will be + * addressed based on logical instance ids. + */ + adev->vcn.harvest_config = 0; + adev->vcn.num_inst_per_aid = 1; + adev->vcn.num_vcn_inst = hweight32(adev->vcn.inst_mask); + adev->jpeg.harvest_config = 0; + adev->jpeg.num_inst_per_aid = 1; + adev->jpeg.num_jpeg_inst = hweight32(adev->jpeg.inst_mask); + + ret = aqua_vanjaram_xcp_mgr_init(adev); + if (ret) + return ret; + + aqua_vanjaram_ip_map_init(adev); + + return 0; +} diff --git a/drivers/gpu/drm/amd/amdgpu/cik.c b/drivers/gpu/drm/amd/amdgpu/cik.c index de6d10390ab2f195fcb3a03091f904ce1edf2c0c..5641cf05d856b4253b9086376b720da92ad4ba64 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik.c +++ b/drivers/gpu/drm/amd/amdgpu/cik.c @@ -1141,12 +1141,12 @@ static uint32_t cik_get_register_value(struct amdgpu_device *adev, mutex_lock(&adev->grbm_idx_mutex); if (se_num != 0xffffffff || sh_num != 0xffffffff) - amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff, 0); val = RREG32(reg_offset); if (se_num != 0xffffffff || sh_num != 0xffffffff) - amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); return val; } else { diff --git a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c index 67d16236b2168ddc4b3e8ae66beaa638f30a13a9..52598fbc9b39db0c08705216b5e88bedf950c9cf 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c @@ -489,8 +489,6 @@ static int cik_sdma_gfx_resume(struct amdgpu_device *adev) #endif /* enable DMA IBs */ WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl); - - ring->sched.ready = true; } cik_sdma_enable(adev, true); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index f5b5ce1051a2864651245a776c7d8b7881dbfeb1..be984f8c71c7b1dca28bbe1188b4d87f847af1d9 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -3490,7 +3490,7 @@ static int gfx_v10_0_get_cu_info(struct amdgpu_device *adev, struct amdgpu_cu_info *cu_info); static uint64_t gfx_v10_0_get_gpu_clock_counter(struct amdgpu_device *adev); static void gfx_v10_0_select_se_sh(struct amdgpu_device *adev, u32 se_num, - u32 sh_num, u32 instance); + u32 sh_num, u32 instance, int xcc_id); static u32 gfx_v10_0_get_wgp_active_bitmap_per_sh(struct amdgpu_device *adev); static int gfx_v10_0_rlc_backdoor_autoload_buffer_init(struct amdgpu_device *adev); @@ -3568,7 +3568,7 @@ static void gfx10_kiq_unmap_queues(struct amdgpu_ring *kiq_ring, struct amdgpu_device *adev = kiq_ring->adev; uint32_t eng_sel = ring->funcs->type == AMDGPU_RING_TYPE_GFX ? 4 : 0; - if (adev->enable_mes && !adev->gfx.kiq.ring.sched.ready) { + if (adev->enable_mes && !adev->gfx.kiq[0].ring.sched.ready) { amdgpu_mes_unmap_legacy_queue(adev, ring, action, gpu_addr, seq); return; } @@ -3636,7 +3636,7 @@ static const struct kiq_pm4_funcs gfx_v10_0_kiq_pm4_funcs = { static void gfx_v10_0_set_kiq_pm4_funcs(struct amdgpu_device *adev) { - adev->gfx.kiq.pmf = &gfx_v10_0_kiq_pm4_funcs; + adev->gfx.kiq[0].pmf = &gfx_v10_0_kiq_pm4_funcs; } static void gfx_v10_0_init_spm_golden_registers(struct amdgpu_device *adev) @@ -4219,7 +4219,7 @@ static int gfx_v10_0_mec_init(struct amdgpu_device *adev) const struct gfx_firmware_header_v1_0 *mec_hdr = NULL; - bitmap_zero(adev->gfx.mec.queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES); + bitmap_zero(adev->gfx.mec_bitmap[0].queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES); /* take ownership of the relevant compute queues */ amdgpu_gfx_compute_queue_acquire(adev); @@ -4291,7 +4291,7 @@ static void wave_read_regs(struct amdgpu_device *adev, uint32_t wave, *(out++) = RREG32_SOC15(GC, 0, mmSQ_IND_DATA); } -static void gfx_v10_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields) +static void gfx_v10_0_read_wave_data(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields) { /* in gfx10 the SIMD_ID is specified as part of the INSTANCE * field when performing a select_se_sh so it should be @@ -4318,7 +4318,7 @@ static void gfx_v10_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, dst[(*no_fields)++] = wave_read_ind(adev, wave, ixSQ_WAVE_MODE); } -static void gfx_v10_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd, +static void gfx_v10_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t start, uint32_t size, uint32_t *dst) { @@ -4329,7 +4329,7 @@ static void gfx_v10_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd, dst); } -static void gfx_v10_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd, +static void gfx_v10_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t thread, uint32_t start, uint32_t size, uint32_t *dst) @@ -4340,7 +4340,7 @@ static void gfx_v10_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd, } static void gfx_v10_0_select_me_pipe_q(struct amdgpu_device *adev, - u32 me, u32 pipe, u32 q, u32 vm) + u32 me, u32 pipe, u32 q, u32 vm, u32 xcc_id) { nv_grbm_select(adev, me, pipe, q, vm); } @@ -4461,7 +4461,7 @@ static int gfx_v10_0_gfx_ring_init(struct amdgpu_device *adev, int ring_id, ring->doorbell_index = adev->doorbell_index.gfx_ring0 << 1; else ring->doorbell_index = adev->doorbell_index.gfx_ring1 << 1; - ring->vm_hub = AMDGPU_GFXHUB_0; + ring->vm_hub = AMDGPU_GFXHUB(0); sprintf(ring->name, "gfx_%d.%d.%d", ring->me, ring->pipe, ring->queue); irq_type = AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP + ring->pipe; @@ -4490,7 +4490,7 @@ static int gfx_v10_0_compute_ring_init(struct amdgpu_device *adev, int ring_id, ring->doorbell_index = (adev->doorbell_index.mec_ring0 + ring_id) << 1; ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr + (ring_id * GFX10_MEC_HPD_SIZE); - ring->vm_hub = AMDGPU_GFXHUB_0; + ring->vm_hub = AMDGPU_GFXHUB(0); sprintf(ring->name, "comp_%d.%d.%d", ring->me, ring->pipe, ring->queue); irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP @@ -4550,7 +4550,7 @@ static int gfx_v10_0_sw_init(void *handle) /* KIQ event */ r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_GRBM_CP, GFX_10_1__SRCID__CP_IB2_INTERRUPT_PKT, - &adev->gfx.kiq.irq); + &adev->gfx.kiq[0].irq); if (r) return r; @@ -4614,8 +4614,8 @@ static int gfx_v10_0_sw_init(void *handle) for (i = 0; i < adev->gfx.mec.num_mec; ++i) { for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) { for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) { - if (!amdgpu_gfx_is_mec_queue_enabled(adev, i, k, - j)) + if (!amdgpu_gfx_is_mec_queue_enabled(adev, 0, i, + k, j)) continue; r = gfx_v10_0_compute_ring_init(adev, ring_id, @@ -4629,19 +4629,19 @@ static int gfx_v10_0_sw_init(void *handle) } if (!adev->enable_mes_kiq) { - r = amdgpu_gfx_kiq_init(adev, GFX10_MEC_HPD_SIZE); + r = amdgpu_gfx_kiq_init(adev, GFX10_MEC_HPD_SIZE, 0); if (r) { DRM_ERROR("Failed to init KIQ BOs!\n"); return r; } - kiq = &adev->gfx.kiq; - r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq); + kiq = &adev->gfx.kiq[0]; + r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq, 0); if (r) return r; } - r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct v10_compute_mqd)); + r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct v10_compute_mqd), 0); if (r) return r; @@ -4690,11 +4690,11 @@ static int gfx_v10_0_sw_fini(void *handle) for (i = 0; i < adev->gfx.num_compute_rings; i++) amdgpu_ring_fini(&adev->gfx.compute_ring[i]); - amdgpu_gfx_mqd_sw_fini(adev); + amdgpu_gfx_mqd_sw_fini(adev, 0); if (!adev->enable_mes_kiq) { - amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq.ring); - amdgpu_gfx_kiq_fini(adev); + amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[0].ring); + amdgpu_gfx_kiq_fini(adev, 0); } gfx_v10_0_pfp_fini(adev); @@ -4712,7 +4712,7 @@ static int gfx_v10_0_sw_fini(void *handle) } static void gfx_v10_0_select_se_sh(struct amdgpu_device *adev, u32 se_num, - u32 sh_num, u32 instance) + u32 sh_num, u32 instance, int xcc_id) { u32 data; @@ -4772,13 +4772,13 @@ static void gfx_v10_0_setup_rb(struct amdgpu_device *adev) (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 6))) && ((gfx_v10_3_get_disabled_sa(adev) >> bitmap) & 1)) continue; - gfx_v10_0_select_se_sh(adev, i, j, 0xffffffff); + gfx_v10_0_select_se_sh(adev, i, j, 0xffffffff, 0); data = gfx_v10_0_get_rb_active_bitmap(adev); active_rbs |= data << ((i * adev->gfx.config.max_sh_per_se + j) * rb_bitmap_width_per_sh); } } - gfx_v10_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v10_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); adev->gfx.config.backend_enable_mask = active_rbs; @@ -4825,6 +4825,29 @@ static u32 gfx_v10_0_init_pa_sc_tile_steering_override(struct amdgpu_device *ade #define DEFAULT_SH_MEM_BASES (0x6000) +static void gfx_v10_0_debug_trap_config_init(struct amdgpu_device *adev, + uint32_t first_vmid, + uint32_t last_vmid) +{ + uint32_t data; + uint32_t trap_config_vmid_mask = 0; + int i; + + /* Calculate trap config vmid mask */ + for (i = first_vmid; i < last_vmid; i++) + trap_config_vmid_mask |= (1 << i); + + data = REG_SET_FIELD(0, SPI_GDBG_TRAP_CONFIG, + VMID_SEL, trap_config_vmid_mask); + data = REG_SET_FIELD(data, SPI_GDBG_TRAP_CONFIG, + TRAP_EN, 1); + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_CONFIG), data); + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0); + + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_DATA0), 0); + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_DATA1), 0); +} + static void gfx_v10_0_init_compute_vmid(struct amdgpu_device *adev) { int i; @@ -4856,6 +4879,9 @@ static void gfx_v10_0_init_compute_vmid(struct amdgpu_device *adev) WREG32_SOC15_OFFSET(GC, 0, mmGDS_GWS_VMID0, i, 0); WREG32_SOC15_OFFSET(GC, 0, mmGDS_OA_VMID0, i, 0); } + + gfx_v10_0_debug_trap_config_init(adev, adev->vm_manager.first_kfd_vmid, + AMDGPU_NUM_VMID); } static void gfx_v10_0_init_gds_vmid(struct amdgpu_device *adev) @@ -4907,7 +4933,7 @@ static void gfx_v10_0_tcp_harvest(struct amdgpu_device *adev) mutex_lock(&adev->grbm_idx_mutex); for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { - gfx_v10_0_select_se_sh(adev, i, j, 0xffffffff); + gfx_v10_0_select_se_sh(adev, i, j, 0xffffffff, 0); wgp_active_bitmap = gfx_v10_0_get_wgp_active_bitmap_per_sh(adev); /* * Set corresponding TCP bits for the inactive WGPs in @@ -4940,7 +4966,7 @@ static void gfx_v10_0_tcp_harvest(struct amdgpu_device *adev) } } - gfx_v10_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v10_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); } @@ -4978,7 +5004,7 @@ static void gfx_v10_0_constants_init(struct amdgpu_device *adev) /* XXX SH_MEM regs */ /* where to put LDS, scratch, GPUVM in FSA64 space */ mutex_lock(&adev->srbm_mutex); - for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB_0].num_ids; i++) { + for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB(0)].num_ids; i++) { nv_grbm_select(adev, 0, 0, 0, i); /* CP and shaders */ WREG32_SOC15(GC, 0, mmSH_MEM_CONFIG, DEFAULT_SH_MEM_CONFIG); @@ -6073,7 +6099,6 @@ static int gfx_v10_0_cp_gfx_resume(struct amdgpu_device *adev) u32 tmp; u32 rb_bufsz; u64 rb_addr, rptr_addr, wptr_gpu_addr; - u32 i; /* Set the write pointer delay */ WREG32_SOC15(GC, 0, mmCP_RB_WPTR_DELAY, 0); @@ -6168,11 +6193,6 @@ static int gfx_v10_0_cp_gfx_resume(struct amdgpu_device *adev) /* start the ring */ gfx_v10_0_cp_gfx_start(adev); - for (i = 0; i < adev->gfx.num_gfx_rings; i++) { - ring = &adev->gfx.gfx_ring[i]; - ring->sched.ready = true; - } - return 0; } @@ -6214,7 +6234,7 @@ static void gfx_v10_0_cp_compute_enable(struct amdgpu_device *adev, bool enable) CP_MEC_CNTL__MEC_ME2_HALT_MASK)); break; } - adev->gfx.kiq.ring.sched.ready = false; + adev->gfx.kiq[0].ring.sched.ready = false; } udelay(50); } @@ -6423,55 +6443,6 @@ static int gfx_v10_0_gfx_mqd_init(struct amdgpu_device *adev, void *m, return 0; } -#ifdef BRING_UP_DEBUG -static int gfx_v10_0_gfx_queue_init_register(struct amdgpu_ring *ring) -{ - struct amdgpu_device *adev = ring->adev; - struct v10_gfx_mqd *mqd = ring->mqd_ptr; - - /* set mmCP_GFX_HQD_WPTR/_HI to 0 */ - WREG32_SOC15(GC, 0, mmCP_GFX_HQD_WPTR, mqd->cp_gfx_hqd_wptr); - WREG32_SOC15(GC, 0, mmCP_GFX_HQD_WPTR_HI, mqd->cp_gfx_hqd_wptr_hi); - - /* set GFX_MQD_BASE */ - WREG32_SOC15(GC, 0, mmCP_MQD_BASE_ADDR, mqd->cp_mqd_base_addr); - WREG32_SOC15(GC, 0, mmCP_MQD_BASE_ADDR_HI, mqd->cp_mqd_base_addr_hi); - - /* set GFX_MQD_CONTROL */ - WREG32_SOC15(GC, 0, mmCP_GFX_MQD_CONTROL, mqd->cp_gfx_mqd_control); - - /* set GFX_HQD_VMID to 0 */ - WREG32_SOC15(GC, 0, mmCP_GFX_HQD_VMID, mqd->cp_gfx_hqd_vmid); - - WREG32_SOC15(GC, 0, mmCP_GFX_HQD_QUEUE_PRIORITY, - mqd->cp_gfx_hqd_queue_priority); - WREG32_SOC15(GC, 0, mmCP_GFX_HQD_QUANTUM, mqd->cp_gfx_hqd_quantum); - - /* set GFX_HQD_BASE, similar as CP_RB_BASE */ - WREG32_SOC15(GC, 0, mmCP_GFX_HQD_BASE, mqd->cp_gfx_hqd_base); - WREG32_SOC15(GC, 0, mmCP_GFX_HQD_BASE_HI, mqd->cp_gfx_hqd_base_hi); - - /* set GFX_HQD_RPTR_ADDR, similar as CP_RB_RPTR */ - WREG32_SOC15(GC, 0, mmCP_GFX_HQD_RPTR_ADDR, mqd->cp_gfx_hqd_rptr_addr); - WREG32_SOC15(GC, 0, mmCP_GFX_HQD_RPTR_ADDR_HI, mqd->cp_gfx_hqd_rptr_addr_hi); - - /* set GFX_HQD_CNTL, similar as CP_RB_CNTL */ - WREG32_SOC15(GC, 0, mmCP_GFX_HQD_CNTL, mqd->cp_gfx_hqd_cntl); - - /* set RB_WPTR_POLL_ADDR */ - WREG32_SOC15(GC, 0, mmCP_RB_WPTR_POLL_ADDR_LO, mqd->cp_rb_wptr_poll_addr_lo); - WREG32_SOC15(GC, 0, mmCP_RB_WPTR_POLL_ADDR_HI, mqd->cp_rb_wptr_poll_addr_hi); - - /* set RB_DOORBELL_CONTROL */ - WREG32_SOC15(GC, 0, mmCP_RB_DOORBELL_CONTROL, mqd->cp_rb_doorbell_control); - - /* active the queue */ - WREG32_SOC15(GC, 0, mmCP_GFX_HQD_ACTIVE, mqd->cp_gfx_hqd_active); - - return 0; -} -#endif - static int gfx_v10_0_gfx_init_queue(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; @@ -6492,59 +6463,23 @@ static int gfx_v10_0_gfx_init_queue(struct amdgpu_ring *ring) if (ring->doorbell_index == adev->doorbell_index.gfx_ring0 << 1) gfx_v10_0_cp_gfx_set_doorbell(adev, ring); -#ifdef BRING_UP_DEBUG - gfx_v10_0_gfx_queue_init_register(ring); -#endif nv_grbm_select(adev, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); if (adev->gfx.me.mqd_backup[mqd_idx]) memcpy(adev->gfx.me.mqd_backup[mqd_idx], mqd, sizeof(*mqd)); - } else if (amdgpu_in_reset(adev)) { - /* reset mqd with the backup copy */ + } else { + /* restore mqd with the backup copy */ if (adev->gfx.me.mqd_backup[mqd_idx]) memcpy(mqd, adev->gfx.me.mqd_backup[mqd_idx], sizeof(*mqd)); /* reset the ring */ ring->wptr = 0; *ring->wptr_cpu_addr = 0; amdgpu_ring_clear_ring(ring); -#ifdef BRING_UP_DEBUG - mutex_lock(&adev->srbm_mutex); - nv_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0); - gfx_v10_0_gfx_queue_init_register(ring); - nv_grbm_select(adev, 0, 0, 0, 0); - mutex_unlock(&adev->srbm_mutex); -#endif - } else { - amdgpu_ring_clear_ring(ring); } return 0; } -#ifndef BRING_UP_DEBUG -static int gfx_v10_0_kiq_enable_kgq(struct amdgpu_device *adev) -{ - struct amdgpu_kiq *kiq = &adev->gfx.kiq; - struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring; - int r, i; - - if (!kiq->pmf || !kiq->pmf->kiq_map_queues) - return -EINVAL; - - r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size * - adev->gfx.num_gfx_rings); - if (r) { - DRM_ERROR("Failed to lock KIQ (%d).\n", r); - return r; - } - - for (i = 0; i < adev->gfx.num_gfx_rings; i++) - kiq->pmf->kiq_map_queues(kiq_ring, &adev->gfx.gfx_ring[i]); - - return amdgpu_ring_test_helper(kiq_ring); -} -#endif - static int gfx_v10_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev) { int r, i; @@ -6555,7 +6490,7 @@ static int gfx_v10_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev) r = amdgpu_bo_reserve(ring->mqd_obj, false); if (unlikely(r != 0)) - goto done; + return r; r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr); if (!r) { @@ -6565,23 +6500,14 @@ static int gfx_v10_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev) } amdgpu_bo_unreserve(ring->mqd_obj); if (r) - goto done; + return r; } -#ifndef BRING_UP_DEBUG - r = gfx_v10_0_kiq_enable_kgq(adev); - if (r) - goto done; -#endif - r = gfx_v10_0_cp_gfx_start(adev); + + r = amdgpu_gfx_enable_kgq(adev, 0); if (r) - goto done; + return r; - for (i = 0; i < adev->gfx.num_gfx_rings; i++) { - ring = &adev->gfx.gfx_ring[i]; - ring->sched.ready = true; - } -done: - return r; + return gfx_v10_0_cp_gfx_start(adev); } static int gfx_v10_0_compute_mqd_init(struct amdgpu_device *adev, void *m, @@ -6812,14 +6738,13 @@ static int gfx_v10_0_kiq_init_queue(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; struct v10_compute_mqd *mqd = ring->mqd_ptr; - int mqd_idx = AMDGPU_MAX_COMPUTE_RINGS; gfx_v10_0_kiq_setting(ring); if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */ /* reset MQD to a clean status */ - if (adev->gfx.mec.mqd_backup[mqd_idx]) - memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(*mqd)); + if (adev->gfx.kiq[0].mqd_backup) + memcpy(mqd, adev->gfx.kiq[0].mqd_backup, sizeof(*mqd)); /* reset ring buffer */ ring->wptr = 0; @@ -6841,8 +6766,8 @@ static int gfx_v10_0_kiq_init_queue(struct amdgpu_ring *ring) nv_grbm_select(adev, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); - if (adev->gfx.mec.mqd_backup[mqd_idx]) - memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(*mqd)); + if (adev->gfx.kiq[0].mqd_backup) + memcpy(adev->gfx.kiq[0].mqd_backup, mqd, sizeof(*mqd)); } return 0; @@ -6864,17 +6789,14 @@ static int gfx_v10_0_kcq_init_queue(struct amdgpu_ring *ring) if (adev->gfx.mec.mqd_backup[mqd_idx]) memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(*mqd)); - } else if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */ - /* reset MQD to a clean status */ + } else { + /* restore MQD to a clean status */ if (adev->gfx.mec.mqd_backup[mqd_idx]) memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(*mqd)); - /* reset ring buffer */ ring->wptr = 0; atomic64_set((atomic64_t *)ring->wptr_cpu_addr, 0); amdgpu_ring_clear_ring(ring); - } else { - amdgpu_ring_clear_ring(ring); } return 0; @@ -6885,21 +6807,22 @@ static int gfx_v10_0_kiq_resume(struct amdgpu_device *adev) struct amdgpu_ring *ring; int r; - ring = &adev->gfx.kiq.ring; + ring = &adev->gfx.kiq[0].ring; r = amdgpu_bo_reserve(ring->mqd_obj, false); if (unlikely(r != 0)) return r; r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr); - if (unlikely(r != 0)) + if (unlikely(r != 0)) { + amdgpu_bo_unreserve(ring->mqd_obj); return r; + } gfx_v10_0_kiq_init_queue(ring); amdgpu_bo_kunmap(ring->mqd_obj); ring->mqd_ptr = NULL; amdgpu_bo_unreserve(ring->mqd_obj); - ring->sched.ready = true; return 0; } @@ -6927,7 +6850,7 @@ static int gfx_v10_0_kcq_resume(struct amdgpu_device *adev) goto done; } - r = amdgpu_gfx_enable_kcq(adev); + r = amdgpu_gfx_enable_kcq(adev, 0); done: return r; } @@ -7240,47 +7163,20 @@ static int gfx_v10_0_hw_init(void *handle) return r; } -#ifndef BRING_UP_DEBUG -static int gfx_v10_0_kiq_disable_kgq(struct amdgpu_device *adev) -{ - struct amdgpu_kiq *kiq = &adev->gfx.kiq; - struct amdgpu_ring *kiq_ring = &kiq->ring; - int i; - - if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) - return -EINVAL; - - if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size * - adev->gfx.num_gfx_rings)) - return -ENOMEM; - - for (i = 0; i < adev->gfx.num_gfx_rings; i++) - kiq->pmf->kiq_unmap_queues(kiq_ring, &adev->gfx.gfx_ring[i], - PREEMPT_QUEUES, 0, 0); - if (!adev->job_hang) - return amdgpu_ring_test_helper(kiq_ring); - else - return 0; -} -#endif - static int gfx_v10_0_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - int r; amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0); amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0); if (!adev->no_hw_access) { -#ifndef BRING_UP_DEBUG if (amdgpu_async_gfx_ring) { - r = gfx_v10_0_kiq_disable_kgq(adev); - if (r) + if (amdgpu_gfx_disable_kgq(adev, 0)) DRM_ERROR("KGQ disable failed\n"); } -#endif - if (amdgpu_gfx_disable_kcq(adev)) + + if (amdgpu_gfx_disable_kcq(adev, 0)) DRM_ERROR("KCQ disable failed\n"); } @@ -7572,7 +7468,7 @@ static bool gfx_v10_0_is_rlc_enabled(struct amdgpu_device *adev) return (REG_GET_FIELD(rlc_cntl, RLC_CNTL, RLC_ENABLE_F32)) ? true : false; } -static void gfx_v10_0_set_safe_mode(struct amdgpu_device *adev) +static void gfx_v10_0_set_safe_mode(struct amdgpu_device *adev, int xcc_id) { uint32_t data; unsigned i; @@ -7613,7 +7509,7 @@ static void gfx_v10_0_set_safe_mode(struct amdgpu_device *adev) } } -static void gfx_v10_0_unset_safe_mode(struct amdgpu_device *adev) +static void gfx_v10_0_unset_safe_mode(struct amdgpu_device *adev, int xcc_id) { uint32_t data; @@ -7960,7 +7856,7 @@ static void gfx_v10_0_apply_medium_grain_clock_gating_workaround(struct amdgpu_d static int gfx_v10_0_update_gfx_clock_gating(struct amdgpu_device *adev, bool enable) { - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); if (enable) { /* enable FGCG firstly*/ @@ -7999,7 +7895,7 @@ static int gfx_v10_0_update_gfx_clock_gating(struct amdgpu_device *adev, AMD_CG_SUPPORT_GFX_3D_CGLS)) gfx_v10_0_enable_gui_idle_interrupt(adev, enable); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); return 0; } @@ -8093,11 +7989,11 @@ static void gfx_v10_cntl_power_gating(struct amdgpu_device *adev, bool enable) static void gfx_v10_cntl_pg(struct amdgpu_device *adev, bool enable) { - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); gfx_v10_cntl_power_gating(adev, enable); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); } static const struct amdgpu_rlc_funcs gfx_v10_0_rlc_funcs = { @@ -8152,8 +8048,14 @@ static int gfx_v10_0_set_powergating_state(void *handle, case IP_VERSION(10, 3, 3): case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 7): + if (!enable) + amdgpu_gfx_off_ctrl(adev, false); + gfx_v10_cntl_pg(adev, enable); - amdgpu_gfx_off_ctrl(adev, enable); + + if (enable) + amdgpu_gfx_off_ctrl(adev, true); + break; default: break; @@ -8640,7 +8542,7 @@ static int gfx_v10_0_ring_preempt_ib(struct amdgpu_ring *ring) { int i, r = 0; struct amdgpu_device *adev = ring->adev; - struct amdgpu_kiq *kiq = &adev->gfx.kiq; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; struct amdgpu_ring *kiq_ring = &kiq->ring; unsigned long flags; @@ -9148,7 +9050,7 @@ static int gfx_v10_0_kiq_set_interrupt_state(struct amdgpu_device *adev, enum amdgpu_interrupt_state state) { uint32_t tmp, target; - struct amdgpu_ring *ring = &(adev->gfx.kiq.ring); + struct amdgpu_ring *ring = &(adev->gfx.kiq[0].ring); if (ring->me == 1) target = SOC15_REG_OFFSET(GC, 0, mmCP_ME1_PIPE0_INT_CNTL); @@ -9192,7 +9094,7 @@ static int gfx_v10_0_kiq_irq(struct amdgpu_device *adev, struct amdgpu_iv_entry *entry) { u8 me_id, pipe_id, queue_id; - struct amdgpu_ring *ring = &(adev->gfx.kiq.ring); + struct amdgpu_ring *ring = &(adev->gfx.kiq[0].ring); me_id = (entry->ring_id & 0x0c) >> 2; pipe_id = (entry->ring_id & 0x03) >> 0; @@ -9369,7 +9271,7 @@ static void gfx_v10_0_set_ring_funcs(struct amdgpu_device *adev) { int i; - adev->gfx.kiq.ring.funcs = &gfx_v10_0_ring_funcs_kiq; + adev->gfx.kiq[0].ring.funcs = &gfx_v10_0_ring_funcs_kiq; for (i = 0; i < adev->gfx.num_gfx_rings; i++) adev->gfx.gfx_ring[i].funcs = &gfx_v10_0_ring_funcs_gfx; @@ -9403,8 +9305,8 @@ static void gfx_v10_0_set_irq_funcs(struct amdgpu_device *adev) adev->gfx.eop_irq.num_types = AMDGPU_CP_IRQ_LAST; adev->gfx.eop_irq.funcs = &gfx_v10_0_eop_irq_funcs; - adev->gfx.kiq.irq.num_types = AMDGPU_CP_KIQ_IRQ_LAST; - adev->gfx.kiq.irq.funcs = &gfx_v10_0_kiq_irq_funcs; + adev->gfx.kiq[0].irq.num_types = AMDGPU_CP_KIQ_IRQ_LAST; + adev->gfx.kiq[0].irq.funcs = &gfx_v10_0_kiq_irq_funcs; adev->gfx.priv_reg_irq.num_types = 1; adev->gfx.priv_reg_irq.funcs = &gfx_v10_0_priv_reg_irq_funcs; @@ -9541,7 +9443,7 @@ static int gfx_v10_0_get_cu_info(struct amdgpu_device *adev, mask = 1; ao_bitmap = 0; counter = 0; - gfx_v10_0_select_se_sh(adev, i, j, 0xffffffff); + gfx_v10_0_select_se_sh(adev, i, j, 0xffffffff, 0); if (i < 4 && j < 2) gfx_v10_0_set_user_wgp_inactive_bitmap_per_sh( adev, disable_masks[i * 2 + j]); @@ -9562,7 +9464,7 @@ static int gfx_v10_0_get_cu_info(struct amdgpu_device *adev, cu_info->ao_cu_bitmap[i][j] = ao_bitmap; } } - gfx_v10_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v10_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); cu_info->number = active_cu_number; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c index f5c3762769843a19e9fdca3a6fdb85d22c6df57f..690e121d9ddaa7fed47a63cdfbaa80f5afe4f2d1 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c @@ -112,7 +112,7 @@ static int gfx_v11_0_get_cu_info(struct amdgpu_device *adev, struct amdgpu_cu_info *cu_info); static uint64_t gfx_v11_0_get_gpu_clock_counter(struct amdgpu_device *adev); static void gfx_v11_0_select_se_sh(struct amdgpu_device *adev, u32 se_num, - u32 sh_num, u32 instance); + u32 sh_num, u32 instance, int xcc_id); static u32 gfx_v11_0_get_wgp_active_bitmap_per_sh(struct amdgpu_device *adev); static void gfx_v11_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume); @@ -123,8 +123,8 @@ static int gfx_v11_0_wait_for_rlc_autoload_complete(struct amdgpu_device *adev); static void gfx_v11_0_ring_invalidate_tlbs(struct amdgpu_ring *ring, uint16_t pasid, uint32_t flush_type, bool all_hub, uint8_t dst_sel); -static void gfx_v11_0_set_safe_mode(struct amdgpu_device *adev); -static void gfx_v11_0_unset_safe_mode(struct amdgpu_device *adev); +static void gfx_v11_0_set_safe_mode(struct amdgpu_device *adev, int xcc_id); +static void gfx_v11_0_unset_safe_mode(struct amdgpu_device *adev, int xcc_id); static void gfx_v11_0_update_perf_clk(struct amdgpu_device *adev, bool enable); @@ -192,7 +192,7 @@ static void gfx11_kiq_unmap_queues(struct amdgpu_ring *kiq_ring, struct amdgpu_device *adev = kiq_ring->adev; uint32_t eng_sel = ring->funcs->type == AMDGPU_RING_TYPE_GFX ? 4 : 0; - if (adev->enable_mes && !adev->gfx.kiq.ring.sched.ready) { + if (adev->enable_mes && !adev->gfx.kiq[0].ring.sched.ready) { amdgpu_mes_unmap_legacy_queue(adev, ring, action, gpu_addr, seq); return; } @@ -260,7 +260,7 @@ static const struct kiq_pm4_funcs gfx_v11_0_kiq_pm4_funcs = { static void gfx_v11_0_set_kiq_pm4_funcs(struct amdgpu_device *adev) { - adev->gfx.kiq.pmf = &gfx_v11_0_kiq_pm4_funcs; + adev->gfx.kiq[0].pmf = &gfx_v11_0_kiq_pm4_funcs; } static void gfx_v11_0_init_golden_registers(struct amdgpu_device *adev) @@ -463,6 +463,23 @@ static int gfx_v11_0_init_toc_microcode(struct amdgpu_device *adev, const char * return err; } +static void gfx_v11_0_check_fw_cp_gfx_shadow(struct amdgpu_device *adev) +{ + switch (adev->ip_versions[GC_HWIP][0]) { + case IP_VERSION(11, 0, 0): + case IP_VERSION(11, 0, 2): + case IP_VERSION(11, 0, 3): + if ((adev->gfx.me_fw_version >= 1505) && + (adev->gfx.pfp_fw_version >= 1600) && + (adev->gfx.mec_fw_version >= 512)) + adev->gfx.cp_gfx_shadow = true; + break; + default: + adev->gfx.cp_gfx_shadow = false; + break; + } +} + static int gfx_v11_0_init_microcode(struct amdgpu_device *adev) { char fw_name[40]; @@ -539,6 +556,7 @@ static int gfx_v11_0_init_microcode(struct amdgpu_device *adev) /* only one MEC for gfx 11.0.0. */ adev->gfx.mec2_fw = NULL; + gfx_v11_0_check_fw_cp_gfx_shadow(adev); out: if (err) { amdgpu_ucode_release(&adev->gfx.pfp_fw); @@ -699,7 +717,7 @@ static int gfx_v11_0_mec_init(struct amdgpu_device *adev) u32 *hpd; size_t mec_hpd_size; - bitmap_zero(adev->gfx.mec.queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES); + bitmap_zero(adev->gfx.mec_bitmap[0].queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES); /* take ownership of the relevant compute queues */ amdgpu_gfx_compute_queue_acquire(adev); @@ -747,7 +765,7 @@ static void wave_read_regs(struct amdgpu_device *adev, uint32_t wave, *(out++) = RREG32_SOC15(GC, 0, regSQ_IND_DATA); } -static void gfx_v11_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields) +static void gfx_v11_0_read_wave_data(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields) { /* in gfx11 the SIMD_ID is specified as part of the INSTANCE * field when performing a select_se_sh so it should be @@ -773,7 +791,7 @@ static void gfx_v11_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, dst[(*no_fields)++] = wave_read_ind(adev, wave, ixSQ_WAVE_MODE); } -static void gfx_v11_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd, +static void gfx_v11_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t start, uint32_t size, uint32_t *dst) { @@ -784,7 +802,7 @@ static void gfx_v11_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd, dst); } -static void gfx_v11_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd, +static void gfx_v11_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t thread, uint32_t start, uint32_t size, uint32_t *dst) @@ -795,11 +813,32 @@ static void gfx_v11_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd, } static void gfx_v11_0_select_me_pipe_q(struct amdgpu_device *adev, - u32 me, u32 pipe, u32 q, u32 vm) + u32 me, u32 pipe, u32 q, u32 vm, u32 xcc_id) { soc21_grbm_select(adev, me, pipe, q, vm); } +/* all sizes are in bytes */ +#define MQD_SHADOW_BASE_SIZE 73728 +#define MQD_SHADOW_BASE_ALIGNMENT 256 +#define MQD_FWWORKAREA_SIZE 484 +#define MQD_FWWORKAREA_ALIGNMENT 256 + +static int gfx_v11_0_get_gfx_shadow_info(struct amdgpu_device *adev, + struct amdgpu_gfx_shadow_info *shadow_info) +{ + if (adev->gfx.cp_gfx_shadow) { + shadow_info->shadow_size = MQD_SHADOW_BASE_SIZE; + shadow_info->shadow_alignment = MQD_SHADOW_BASE_ALIGNMENT; + shadow_info->csa_size = MQD_FWWORKAREA_SIZE; + shadow_info->csa_alignment = MQD_FWWORKAREA_ALIGNMENT; + return 0; + } else { + memset(shadow_info, 0, sizeof(struct amdgpu_gfx_shadow_info)); + return -ENOTSUPP; + } +} + static const struct amdgpu_gfx_funcs gfx_v11_0_gfx_funcs = { .get_gpu_clock_counter = &gfx_v11_0_get_gpu_clock_counter, .select_se_sh = &gfx_v11_0_select_se_sh, @@ -808,6 +847,7 @@ static const struct amdgpu_gfx_funcs gfx_v11_0_gfx_funcs = { .read_wave_vgprs = &gfx_v11_0_read_wave_vgprs, .select_me_pipe_q = &gfx_v11_0_select_me_pipe_q, .update_perfmon_mgcg = &gfx_v11_0_update_perf_clk, + .get_gfx_shadow_info = &gfx_v11_0_get_gfx_shadow_info, }; static int gfx_v11_0_gpu_early_init(struct amdgpu_device *adev) @@ -866,7 +906,7 @@ static int gfx_v11_0_gfx_ring_init(struct amdgpu_device *adev, int ring_id, ring->doorbell_index = adev->doorbell_index.gfx_ring0 << 1; else ring->doorbell_index = adev->doorbell_index.gfx_ring1 << 1; - ring->vm_hub = AMDGPU_GFXHUB_0; + ring->vm_hub = AMDGPU_GFXHUB(0); sprintf(ring->name, "gfx_%d.%d.%d", ring->me, ring->pipe, ring->queue); irq_type = AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP + ring->pipe; @@ -897,7 +937,7 @@ static int gfx_v11_0_compute_ring_init(struct amdgpu_device *adev, int ring_id, ring->doorbell_index = (adev->doorbell_index.mec_ring0 + ring_id) << 1; ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr + (ring_id * GFX11_MEC_HPD_SIZE); - ring->vm_hub = AMDGPU_GFXHUB_0; + ring->vm_hub = AMDGPU_GFXHUB(0); sprintf(ring->name, "comp_%d.%d.%d", ring->me, ring->pipe, ring->queue); irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP @@ -1367,8 +1407,8 @@ static int gfx_v11_0_sw_init(void *handle) for (i = 0; i < adev->gfx.mec.num_mec; ++i) { for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) { for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) { - if (!amdgpu_gfx_is_mec_queue_enabled(adev, i, k, - j)) + if (!amdgpu_gfx_is_mec_queue_enabled(adev, 0, i, + k, j)) continue; r = gfx_v11_0_compute_ring_init(adev, ring_id, @@ -1382,19 +1422,19 @@ static int gfx_v11_0_sw_init(void *handle) } if (!adev->enable_mes_kiq) { - r = amdgpu_gfx_kiq_init(adev, GFX11_MEC_HPD_SIZE); + r = amdgpu_gfx_kiq_init(adev, GFX11_MEC_HPD_SIZE, 0); if (r) { DRM_ERROR("Failed to init KIQ BOs!\n"); return r; } - kiq = &adev->gfx.kiq; - r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq); + kiq = &adev->gfx.kiq[0]; + r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq, 0); if (r) return r; } - r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct v11_compute_mqd)); + r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct v11_compute_mqd), 0); if (r) return r; @@ -1456,11 +1496,11 @@ static int gfx_v11_0_sw_fini(void *handle) for (i = 0; i < adev->gfx.num_compute_rings; i++) amdgpu_ring_fini(&adev->gfx.compute_ring[i]); - amdgpu_gfx_mqd_sw_fini(adev); + amdgpu_gfx_mqd_sw_fini(adev, 0); if (!adev->enable_mes_kiq) { - amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq.ring); - amdgpu_gfx_kiq_fini(adev); + amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[0].ring); + amdgpu_gfx_kiq_fini(adev, 0); } gfx_v11_0_pfp_fini(adev); @@ -1477,7 +1517,7 @@ static int gfx_v11_0_sw_fini(void *handle) } static void gfx_v11_0_select_se_sh(struct amdgpu_device *adev, u32 se_num, - u32 sh_num, u32 instance) + u32 sh_num, u32 instance, int xcc_id) { u32 data; @@ -1598,6 +1638,7 @@ static void gfx_v11_0_init_compute_vmid(struct amdgpu_device *adev) /* Enable trap for each kfd vmid. */ data = RREG32_SOC15(GC, 0, regSPI_GDBG_PER_VMID_CNTL); data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1); + WREG32_SOC15(GC, 0, regSPI_GDBG_PER_VMID_CNTL, data); } soc21_grbm_select(adev, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); @@ -1667,7 +1708,7 @@ static void gfx_v11_0_constants_init(struct amdgpu_device *adev) /* XXX SH_MEM regs */ /* where to put LDS, scratch, GPUVM in FSA64 space */ mutex_lock(&adev->srbm_mutex); - for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB_0].num_ids; i++) { + for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB(0)].num_ids; i++) { soc21_grbm_select(adev, 0, 0, 0, i); /* CP and shaders */ WREG32_SOC15(GC, 0, regSH_MEM_CONFIG, DEFAULT_SH_MEM_CONFIG); @@ -3188,7 +3229,6 @@ static int gfx_v11_0_cp_gfx_resume(struct amdgpu_device *adev) u32 tmp; u32 rb_bufsz; u64 rb_addr, rptr_addr, wptr_gpu_addr; - u32 i; /* Set the write pointer delay */ WREG32_SOC15(GC, 0, regCP_RB_WPTR_DELAY, 0); @@ -3280,11 +3320,6 @@ static int gfx_v11_0_cp_gfx_resume(struct amdgpu_device *adev) /* start the ring */ gfx_v11_0_cp_gfx_start(adev); - for (i = 0; i < adev->gfx.num_gfx_rings; i++) { - ring = &adev->gfx.gfx_ring[i]; - ring->sched.ready = true; - } - return 0; } @@ -3330,8 +3365,6 @@ static void gfx_v11_0_cp_compute_enable(struct amdgpu_device *adev, bool enable) WREG32_SOC15(GC, 0, regCP_MEC_CNTL, data); } - adev->gfx.kiq.ring.sched.ready = enable; - udelay(50); } @@ -3633,55 +3666,6 @@ static int gfx_v11_0_gfx_mqd_init(struct amdgpu_device *adev, void *m, return 0; } -#ifdef BRING_UP_DEBUG -static int gfx_v11_0_gfx_queue_init_register(struct amdgpu_ring *ring) -{ - struct amdgpu_device *adev = ring->adev; - struct v11_gfx_mqd *mqd = ring->mqd_ptr; - - /* set mmCP_GFX_HQD_WPTR/_HI to 0 */ - WREG32_SOC15(GC, 0, regCP_GFX_HQD_WPTR, mqd->cp_gfx_hqd_wptr); - WREG32_SOC15(GC, 0, regCP_GFX_HQD_WPTR_HI, mqd->cp_gfx_hqd_wptr_hi); - - /* set GFX_MQD_BASE */ - WREG32_SOC15(GC, 0, regCP_MQD_BASE_ADDR, mqd->cp_mqd_base_addr); - WREG32_SOC15(GC, 0, regCP_MQD_BASE_ADDR_HI, mqd->cp_mqd_base_addr_hi); - - /* set GFX_MQD_CONTROL */ - WREG32_SOC15(GC, 0, regCP_GFX_MQD_CONTROL, mqd->cp_gfx_mqd_control); - - /* set GFX_HQD_VMID to 0 */ - WREG32_SOC15(GC, 0, regCP_GFX_HQD_VMID, mqd->cp_gfx_hqd_vmid); - - WREG32_SOC15(GC, 0, regCP_GFX_HQD_QUEUE_PRIORITY, - mqd->cp_gfx_hqd_queue_priority); - WREG32_SOC15(GC, 0, regCP_GFX_HQD_QUANTUM, mqd->cp_gfx_hqd_quantum); - - /* set GFX_HQD_BASE, similar as CP_RB_BASE */ - WREG32_SOC15(GC, 0, regCP_GFX_HQD_BASE, mqd->cp_gfx_hqd_base); - WREG32_SOC15(GC, 0, regCP_GFX_HQD_BASE_HI, mqd->cp_gfx_hqd_base_hi); - - /* set GFX_HQD_RPTR_ADDR, similar as CP_RB_RPTR */ - WREG32_SOC15(GC, 0, regCP_GFX_HQD_RPTR_ADDR, mqd->cp_gfx_hqd_rptr_addr); - WREG32_SOC15(GC, 0, regCP_GFX_HQD_RPTR_ADDR_HI, mqd->cp_gfx_hqd_rptr_addr_hi); - - /* set GFX_HQD_CNTL, similar as CP_RB_CNTL */ - WREG32_SOC15(GC, 0, regCP_GFX_HQD_CNTL, mqd->cp_gfx_hqd_cntl); - - /* set RB_WPTR_POLL_ADDR */ - WREG32_SOC15(GC, 0, regCP_RB_WPTR_POLL_ADDR_LO, mqd->cp_rb_wptr_poll_addr_lo); - WREG32_SOC15(GC, 0, regCP_RB_WPTR_POLL_ADDR_HI, mqd->cp_rb_wptr_poll_addr_hi); - - /* set RB_DOORBELL_CONTROL */ - WREG32_SOC15(GC, 0, regCP_RB_DOORBELL_CONTROL, mqd->cp_rb_doorbell_control); - - /* active the queue */ - WREG32_SOC15(GC, 0, regCP_GFX_HQD_ACTIVE, mqd->cp_gfx_hqd_active); - - return 0; -} -#endif - static int gfx_v11_0_gfx_init_queue(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; @@ -3693,59 +3677,23 @@ static int gfx_v11_0_gfx_init_queue(struct amdgpu_ring *ring) mutex_lock(&adev->srbm_mutex); soc21_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0); amdgpu_ring_init_mqd(ring); -#ifdef BRING_UP_DEBUG - gfx_v11_0_gfx_queue_init_register(ring); -#endif soc21_grbm_select(adev, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); if (adev->gfx.me.mqd_backup[mqd_idx]) memcpy(adev->gfx.me.mqd_backup[mqd_idx], mqd, sizeof(*mqd)); - } else if (amdgpu_in_reset(adev)) { - /* reset mqd with the backup copy */ + } else { + /* restore mqd with the backup copy */ if (adev->gfx.me.mqd_backup[mqd_idx]) memcpy(mqd, adev->gfx.me.mqd_backup[mqd_idx], sizeof(*mqd)); /* reset the ring */ ring->wptr = 0; *ring->wptr_cpu_addr = 0; amdgpu_ring_clear_ring(ring); -#ifdef BRING_UP_DEBUG - mutex_lock(&adev->srbm_mutex); - soc21_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0); - gfx_v11_0_gfx_queue_init_register(ring); - soc21_grbm_select(adev, 0, 0, 0, 0); - mutex_unlock(&adev->srbm_mutex); -#endif - } else { - amdgpu_ring_clear_ring(ring); } return 0; } -#ifndef BRING_UP_DEBUG -static int gfx_v11_0_kiq_enable_kgq(struct amdgpu_device *adev) -{ - struct amdgpu_kiq *kiq = &adev->gfx.kiq; - struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring; - int r, i; - - if (!kiq->pmf || !kiq->pmf->kiq_map_queues) - return -EINVAL; - - r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size * - adev->gfx.num_gfx_rings); - if (r) { - DRM_ERROR("Failed to lock KIQ (%d).\n", r); - return r; - } - - for (i = 0; i < adev->gfx.num_gfx_rings; i++) - kiq->pmf->kiq_map_queues(kiq_ring, &adev->gfx.gfx_ring[i]); - - return amdgpu_ring_test_helper(kiq_ring); -} -#endif - static int gfx_v11_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev) { int r, i; @@ -3756,7 +3704,7 @@ static int gfx_v11_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev) r = amdgpu_bo_reserve(ring->mqd_obj, false); if (unlikely(r != 0)) - goto done; + return r; r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr); if (!r) { @@ -3766,23 +3714,14 @@ static int gfx_v11_0_cp_async_gfx_ring_resume(struct amdgpu_device *adev) } amdgpu_bo_unreserve(ring->mqd_obj); if (r) - goto done; + return r; } -#ifndef BRING_UP_DEBUG - r = gfx_v11_0_kiq_enable_kgq(adev); - if (r) - goto done; -#endif - r = gfx_v11_0_cp_gfx_start(adev); + + r = amdgpu_gfx_enable_kgq(adev, 0); if (r) - goto done; + return r; - for (i = 0; i < adev->gfx.num_gfx_rings; i++) { - ring = &adev->gfx.gfx_ring[i]; - ring->sched.ready = true; - } -done: - return r; + return gfx_v11_0_cp_gfx_start(adev); } static int gfx_v11_0_compute_mqd_init(struct amdgpu_device *adev, void *m, @@ -4028,14 +3967,13 @@ static int gfx_v11_0_kiq_init_queue(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; struct v11_compute_mqd *mqd = ring->mqd_ptr; - int mqd_idx = AMDGPU_MAX_COMPUTE_RINGS; gfx_v11_0_kiq_setting(ring); if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */ /* reset MQD to a clean status */ - if (adev->gfx.mec.mqd_backup[mqd_idx]) - memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(*mqd)); + if (adev->gfx.kiq[0].mqd_backup) + memcpy(mqd, adev->gfx.kiq[0].mqd_backup, sizeof(*mqd)); /* reset ring buffer */ ring->wptr = 0; @@ -4057,8 +3995,8 @@ static int gfx_v11_0_kiq_init_queue(struct amdgpu_ring *ring) soc21_grbm_select(adev, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); - if (adev->gfx.mec.mqd_backup[mqd_idx]) - memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(*mqd)); + if (adev->gfx.kiq[0].mqd_backup) + memcpy(adev->gfx.kiq[0].mqd_backup, mqd, sizeof(*mqd)); } return 0; @@ -4080,17 +4018,14 @@ static int gfx_v11_0_kcq_init_queue(struct amdgpu_ring *ring) if (adev->gfx.mec.mqd_backup[mqd_idx]) memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(*mqd)); - } else if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */ - /* reset MQD to a clean status */ + } else { + /* restore MQD to a clean status */ if (adev->gfx.mec.mqd_backup[mqd_idx]) memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(*mqd)); - /* reset ring buffer */ ring->wptr = 0; atomic64_set((atomic64_t *)ring->wptr_cpu_addr, 0); amdgpu_ring_clear_ring(ring); - } else { - amdgpu_ring_clear_ring(ring); } return 0; @@ -4101,7 +4036,7 @@ static int gfx_v11_0_kiq_resume(struct amdgpu_device *adev) struct amdgpu_ring *ring; int r; - ring = &adev->gfx.kiq.ring; + ring = &adev->gfx.kiq[0].ring; r = amdgpu_bo_reserve(ring->mqd_obj, false); if (unlikely(r != 0)) @@ -4146,7 +4081,7 @@ static int gfx_v11_0_kcq_resume(struct amdgpu_device *adev) goto done; } - r = amdgpu_gfx_enable_kcq(adev); + r = amdgpu_gfx_enable_kcq(adev, 0); done: return r; } @@ -4239,7 +4174,7 @@ static int gfx_v11_0_gfxhub_enable(struct amdgpu_device *adev) false : true; adev->gfxhub.funcs->set_fault_enable_default(adev, value); - amdgpu_gmc_flush_gpu_tlb(adev, 0, AMDGPU_GFXHUB_0, 0); + amdgpu_gmc_flush_gpu_tlb(adev, 0, AMDGPU_GFXHUB(0), 0); return 0; } @@ -4407,48 +4342,20 @@ static int gfx_v11_0_hw_init(void *handle) return r; } -#ifndef BRING_UP_DEBUG -static int gfx_v11_0_kiq_disable_kgq(struct amdgpu_device *adev) -{ - struct amdgpu_kiq *kiq = &adev->gfx.kiq; - struct amdgpu_ring *kiq_ring = &kiq->ring; - int i, r = 0; - - if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) - return -EINVAL; - - if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size * - adev->gfx.num_gfx_rings)) - return -ENOMEM; - - for (i = 0; i < adev->gfx.num_gfx_rings; i++) - kiq->pmf->kiq_unmap_queues(kiq_ring, &adev->gfx.gfx_ring[i], - PREEMPT_QUEUES, 0, 0); - - if (adev->gfx.kiq.ring.sched.ready) - r = amdgpu_ring_test_helper(kiq_ring); - - return r; -} -#endif - static int gfx_v11_0_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - int r; amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0); amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0); if (!adev->no_hw_access) { -#ifndef BRING_UP_DEBUG if (amdgpu_async_gfx_ring) { - r = gfx_v11_0_kiq_disable_kgq(adev); - if (r) + if (amdgpu_gfx_disable_kgq(adev, 0)) DRM_ERROR("KGQ disable failed\n"); } -#endif - if (amdgpu_gfx_disable_kcq(adev)) + + if (amdgpu_gfx_disable_kcq(adev, 0)) DRM_ERROR("KCQ disable failed\n"); amdgpu_mes_kiq_hw_fini(adev); @@ -4525,7 +4432,7 @@ static int gfx_v11_0_soft_reset(void *handle) tmp = REG_SET_FIELD(tmp, CP_INT_CNTL, GFX_IDLE_INT_ENABLE, 0); WREG32_SOC15(GC, 0, regCP_INT_CNTL, tmp); - gfx_v11_0_set_safe_mode(adev); + gfx_v11_0_set_safe_mode(adev, 0); for (i = 0; i < adev->gfx.mec.num_mec; ++i) { for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) { @@ -4625,7 +4532,7 @@ static int gfx_v11_0_soft_reset(void *handle) tmp = REG_SET_FIELD(tmp, CP_INT_CNTL, GFX_IDLE_INT_ENABLE, 1); WREG32_SOC15(GC, 0, regCP_INT_CNTL, tmp); - gfx_v11_0_unset_safe_mode(adev); + gfx_v11_0_unset_safe_mode(adev, 0); return gfx_v11_0_cp_resume(adev); } @@ -4667,24 +4574,27 @@ static uint64_t gfx_v11_0_get_gpu_clock_counter(struct amdgpu_device *adev) uint64_t clock; uint64_t clock_counter_lo, clock_counter_hi_pre, clock_counter_hi_after; - amdgpu_gfx_off_ctrl(adev, false); - mutex_lock(&adev->gfx.gpu_clock_mutex); if (amdgpu_sriov_vf(adev)) { + amdgpu_gfx_off_ctrl(adev, false); + mutex_lock(&adev->gfx.gpu_clock_mutex); clock_counter_hi_pre = (uint64_t)RREG32_SOC15(GC, 0, regCP_MES_MTIME_HI); clock_counter_lo = (uint64_t)RREG32_SOC15(GC, 0, regCP_MES_MTIME_LO); clock_counter_hi_after = (uint64_t)RREG32_SOC15(GC, 0, regCP_MES_MTIME_HI); if (clock_counter_hi_pre != clock_counter_hi_after) clock_counter_lo = (uint64_t)RREG32_SOC15(GC, 0, regCP_MES_MTIME_LO); + mutex_unlock(&adev->gfx.gpu_clock_mutex); + amdgpu_gfx_off_ctrl(adev, true); } else { + preempt_disable(); clock_counter_hi_pre = (uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_UPPER); clock_counter_lo = (uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_LOWER); clock_counter_hi_after = (uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_UPPER); if (clock_counter_hi_pre != clock_counter_hi_after) clock_counter_lo = (uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_LOWER); + preempt_enable(); } clock = clock_counter_lo | (clock_counter_hi_after << 32ULL); - mutex_unlock(&adev->gfx.gpu_clock_mutex); - amdgpu_gfx_off_ctrl(adev, true); + return clock; } @@ -4791,7 +4701,7 @@ static bool gfx_v11_0_is_rlc_enabled(struct amdgpu_device *adev) return (REG_GET_FIELD(rlc_cntl, RLC_CNTL, RLC_ENABLE_F32)) ? true : false; } -static void gfx_v11_0_set_safe_mode(struct amdgpu_device *adev) +static void gfx_v11_0_set_safe_mode(struct amdgpu_device *adev, int xcc_id) { uint32_t data; unsigned i; @@ -4810,7 +4720,7 @@ static void gfx_v11_0_set_safe_mode(struct amdgpu_device *adev) } } -static void gfx_v11_0_unset_safe_mode(struct amdgpu_device *adev) +static void gfx_v11_0_unset_safe_mode(struct amdgpu_device *adev, int xcc_id) { WREG32_SOC15(GC, 0, regRLC_SAFE_MODE, RLC_SAFE_MODE__CMD_MASK); } @@ -5038,7 +4948,7 @@ static void gfx_v11_0_update_coarse_grain_clock_gating(struct amdgpu_device *ade static int gfx_v11_0_update_gfx_clock_gating(struct amdgpu_device *adev, bool enable) { - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); gfx_v11_0_update_coarse_grain_clock_gating(adev, enable); @@ -5058,7 +4968,7 @@ static int gfx_v11_0_update_gfx_clock_gating(struct amdgpu_device *adev, AMD_CG_SUPPORT_GFX_3D_CGLS)) gfx_v11_0_enable_gui_idle_interrupt(adev, enable); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); return 0; } @@ -5126,11 +5036,11 @@ static void gfx_v11_cntl_power_gating(struct amdgpu_device *adev, bool enable) static void gfx_v11_cntl_pg(struct amdgpu_device *adev, bool enable) { - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); gfx_v11_cntl_power_gating(adev, enable); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); } static int gfx_v11_0_set_powergating_state(void *handle, @@ -5150,8 +5060,14 @@ static int gfx_v11_0_set_powergating_state(void *handle, break; case IP_VERSION(11, 0, 1): case IP_VERSION(11, 0, 4): + if (!enable) + amdgpu_gfx_off_ctrl(adev, false); + gfx_v11_cntl_pg(adev, enable); - amdgpu_gfx_off_ctrl(adev, enable); + + if (enable) + amdgpu_gfx_off_ctrl(adev, true); + break; default: break; @@ -5583,6 +5499,29 @@ static void gfx_v11_0_ring_emit_cntxcntl(struct amdgpu_ring *ring, amdgpu_ring_write(ring, 0); } +static void gfx_v11_0_ring_emit_gfx_shadow(struct amdgpu_ring *ring, + u64 shadow_va, u64 csa_va, + u64 gds_va, bool init_shadow, + int vmid) +{ + struct amdgpu_device *adev = ring->adev; + + if (!adev->gfx.cp_gfx_shadow) + return; + + amdgpu_ring_write(ring, PACKET3(PACKET3_SET_Q_PREEMPTION_MODE, 7)); + amdgpu_ring_write(ring, lower_32_bits(shadow_va)); + amdgpu_ring_write(ring, upper_32_bits(shadow_va)); + amdgpu_ring_write(ring, lower_32_bits(gds_va)); + amdgpu_ring_write(ring, upper_32_bits(gds_va)); + amdgpu_ring_write(ring, lower_32_bits(csa_va)); + amdgpu_ring_write(ring, upper_32_bits(csa_va)); + amdgpu_ring_write(ring, shadow_va ? + PACKET3_SET_Q_PREEMPTION_MODE_IB_VMID(vmid) : 0); + amdgpu_ring_write(ring, init_shadow ? + PACKET3_SET_Q_PREEMPTION_MODE_INIT_SHADOW_MEM : 0); +} + static unsigned gfx_v11_0_ring_emit_init_cond_exec(struct amdgpu_ring *ring) { unsigned ret; @@ -5614,7 +5553,7 @@ static int gfx_v11_0_ring_preempt_ib(struct amdgpu_ring *ring) { int i, r = 0; struct amdgpu_device *adev = ring->adev; - struct amdgpu_kiq *kiq = &adev->gfx.kiq; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; struct amdgpu_ring *kiq_ring = &kiq->ring; unsigned long flags; @@ -6082,7 +6021,7 @@ static int gfx_v11_0_kiq_set_interrupt_state(struct amdgpu_device *adev, enum amdgpu_interrupt_state state) { uint32_t tmp, target; - struct amdgpu_ring *ring = &(adev->gfx.kiq.ring); + struct amdgpu_ring *ring = &(adev->gfx.kiq[0].ring); target = SOC15_REG_OFFSET(GC, 0, regCP_ME1_PIPE0_INT_CNTL); target += ring->pipe; @@ -6173,6 +6112,7 @@ static const struct amdgpu_ring_funcs gfx_v11_0_ring_funcs_gfx = { .set_wptr = gfx_v11_0_ring_set_wptr_gfx, .emit_frame_size = /* totally 242 maximum if 16 IBs */ 5 + /* COND_EXEC */ + 9 + /* SET_Q_PREEMPTION_MODE */ 7 + /* PIPELINE_SYNC */ SOC15_FLUSH_GPU_TLB_NUM_WREG * 5 + SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 7 + @@ -6199,6 +6139,7 @@ static const struct amdgpu_ring_funcs gfx_v11_0_ring_funcs_gfx = { .insert_nop = amdgpu_ring_insert_nop, .pad_ib = amdgpu_ring_generic_pad_ib, .emit_cntxcntl = gfx_v11_0_ring_emit_cntxcntl, + .emit_gfx_shadow = gfx_v11_0_ring_emit_gfx_shadow, .init_cond_exec = gfx_v11_0_ring_emit_init_cond_exec, .patch_cond_exec = gfx_v11_0_ring_emit_patch_cond_exec, .preempt_ib = gfx_v11_0_ring_preempt_ib, @@ -6279,7 +6220,7 @@ static void gfx_v11_0_set_ring_funcs(struct amdgpu_device *adev) { int i; - adev->gfx.kiq.ring.funcs = &gfx_v11_0_ring_funcs_kiq; + adev->gfx.kiq[0].ring.funcs = &gfx_v11_0_ring_funcs_kiq; for (i = 0; i < adev->gfx.num_gfx_rings; i++) adev->gfx.gfx_ring[i].funcs = &gfx_v11_0_ring_funcs_gfx; @@ -6428,7 +6369,7 @@ static int gfx_v11_0_get_cu_info(struct amdgpu_device *adev, for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { mask = 1; counter = 0; - gfx_v11_0_select_se_sh(adev, i, j, 0xffffffff); + gfx_v11_0_select_se_sh(adev, i, j, 0xffffffff, 0); if (i < 8 && j < 2) gfx_v11_0_set_user_wgp_inactive_bitmap_per_sh( adev, disable_masks[i * 2 + j]); @@ -6460,7 +6401,7 @@ static int gfx_v11_0_get_cu_info(struct amdgpu_device *adev, active_cu_number += counter; } } - gfx_v11_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v11_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); cu_info->number = active_cu_number; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c index 068b9586a223e004a183f335806a985ae14b0080..26d6286d86c9991f98c1ace2b9becb54eadeeae3 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c @@ -84,8 +84,20 @@ static int gfx_v11_0_3_poison_consumption_handler(struct amdgpu_device *adev, /* Workaround: when vmid and pasid are both zero, trigger gpu reset in KGD. */ if (entry && (entry->client_id == SOC21_IH_CLIENTID_GFX) && (entry->src_id == GFX_11_0_0__SRCID__RLC_GC_FED_INTERRUPT) && - !entry->vmid && !entry->pasid) + !entry->vmid && !entry->pasid) { + uint32_t rlc_status0 = 0; + + rlc_status0 = RREG32_SOC15(GC, 0, regRLC_RLCS_FED_STATUS_0); + + if (REG_GET_FIELD(rlc_status0, RLC_RLCS_FED_STATUS_0, SDMA0_FED_ERR) || + REG_GET_FIELD(rlc_status0, RLC_RLCS_FED_STATUS_0, SDMA1_FED_ERR)) { + struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); + + ras->gpu_reset_flags |= AMDGPU_RAS_GPU_RESET_MODE2_RESET; + } + amdgpu_ras_reset_gpu(adev); + } return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c index c41219e23151c925f022739d9439a8246dd9c24e..da6caff78c22be23ba918c49f4334c97236c0104 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c @@ -1285,7 +1285,7 @@ static void gfx_v6_0_tiling_mode_table_init(struct amdgpu_device *adev) } static void gfx_v6_0_select_se_sh(struct amdgpu_device *adev, u32 se_num, - u32 sh_num, u32 instance) + u32 sh_num, u32 instance, int xcc_id) { u32 data; @@ -1438,12 +1438,12 @@ static void gfx_v6_0_write_harvested_raster_configs(struct amdgpu_device *adev, } /* GRBM_GFX_INDEX has a different offset on SI */ - gfx_v6_0_select_se_sh(adev, se, 0xffffffff, 0xffffffff); + gfx_v6_0_select_se_sh(adev, se, 0xffffffff, 0xffffffff, 0); WREG32(mmPA_SC_RASTER_CONFIG, raster_config_se); } /* GRBM_GFX_INDEX has a different offset on SI */ - gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); } static void gfx_v6_0_setup_rb(struct amdgpu_device *adev) @@ -1459,14 +1459,14 @@ static void gfx_v6_0_setup_rb(struct amdgpu_device *adev) mutex_lock(&adev->grbm_idx_mutex); for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { - gfx_v6_0_select_se_sh(adev, i, j, 0xffffffff); + gfx_v6_0_select_se_sh(adev, i, j, 0xffffffff, 0); data = gfx_v6_0_get_rb_active_bitmap(adev); active_rbs |= data << ((i * adev->gfx.config.max_sh_per_se + j) * rb_bitmap_width_per_sh); } } - gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); adev->gfx.config.backend_enable_mask = active_rbs; adev->gfx.config.num_rbs = hweight32(active_rbs); @@ -1487,7 +1487,7 @@ static void gfx_v6_0_setup_rb(struct amdgpu_device *adev) /* cache the values for userspace */ for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { - gfx_v6_0_select_se_sh(adev, i, j, 0xffffffff); + gfx_v6_0_select_se_sh(adev, i, j, 0xffffffff, 0); adev->gfx.config.rb_config[i][j].rb_backend_disable = RREG32(mmCC_RB_BACKEND_DISABLE); adev->gfx.config.rb_config[i][j].user_rb_backend_disable = @@ -1496,7 +1496,7 @@ static void gfx_v6_0_setup_rb(struct amdgpu_device *adev) RREG32(mmPA_SC_RASTER_CONFIG); } } - gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); } @@ -1535,7 +1535,7 @@ static void gfx_v6_0_setup_spi(struct amdgpu_device *adev) mutex_lock(&adev->grbm_idx_mutex); for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { - gfx_v6_0_select_se_sh(adev, i, j, 0xffffffff); + gfx_v6_0_select_se_sh(adev, i, j, 0xffffffff, 0); data = RREG32(mmSPI_STATIC_THREAD_MGMT_3); active_cu = gfx_v6_0_get_cu_enabled(adev); @@ -1550,7 +1550,7 @@ static void gfx_v6_0_setup_spi(struct amdgpu_device *adev) } } } - gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); } @@ -2391,7 +2391,7 @@ static void gfx_v6_0_enable_lbpw(struct amdgpu_device *adev, bool enable) WREG32_FIELD(RLC_LB_CNTL, LOAD_BALANCE_ENABLE, enable ? 1 : 0); if (!enable) { - gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); WREG32(mmSPI_LB_CU_MASK, 0x00ff); } } @@ -2968,7 +2968,7 @@ static void wave_read_regs(struct amdgpu_device *adev, uint32_t simd, *(out++) = RREG32(mmSQ_IND_DATA); } -static void gfx_v6_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields) +static void gfx_v6_0_read_wave_data(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields) { /* type 0 wave data */ dst[(*no_fields)++] = 0; @@ -2993,7 +2993,7 @@ static void gfx_v6_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, u dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_MODE); } -static void gfx_v6_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd, +static void gfx_v6_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t start, uint32_t size, uint32_t *dst) { @@ -3003,7 +3003,7 @@ static void gfx_v6_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd, } static void gfx_v6_0_select_me_pipe_q(struct amdgpu_device *adev, - u32 me, u32 pipe, u32 q, u32 vm) + u32 me, u32 pipe, u32 q, u32 vm, u32 xcc_id) { DRM_INFO("Not implemented\n"); } @@ -3028,6 +3028,7 @@ static int gfx_v6_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + adev->gfx.xcc_mask = 1; adev->gfx.num_gfx_rings = GFX6_NUM_GFX_RINGS; adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev), GFX6_NUM_COMPUTE_RINGS); @@ -3073,7 +3074,7 @@ static int gfx_v6_0_sw_init(void *handle) ring = &adev->gfx.gfx_ring[i]; ring->ring_obj = NULL; sprintf(ring->name, "gfx"); - r = amdgpu_ring_init(adev, ring, 1024, + r = amdgpu_ring_init(adev, ring, 2048, &adev->gfx.eop_irq, AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP, AMDGPU_RING_PRIO_DEFAULT, NULL); @@ -3571,7 +3572,7 @@ static void gfx_v6_0_get_cu_info(struct amdgpu_device *adev) mask = 1; ao_bitmap = 0; counter = 0; - gfx_v6_0_select_se_sh(adev, i, j, 0xffffffff); + gfx_v6_0_select_se_sh(adev, i, j, 0xffffffff, 0); if (i < 4 && j < 2) gfx_v6_0_set_user_cu_inactive_bitmap( adev, disable_masks[i * 2 + j]); @@ -3593,7 +3594,7 @@ static void gfx_v6_0_get_cu_info(struct amdgpu_device *adev) } } - gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v6_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); cu_info->number = active_cu_number; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index 9d5c1e29b4a35816b22018c1ff75146af9244854..8c174c11eaee08910b53439d0727dfcf26031305 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -1548,11 +1548,12 @@ static void gfx_v7_0_tiling_mode_table_init(struct amdgpu_device *adev) * @sh_num: sh block to address * @instance: Certain registers are instanced per SE or SH. * 0xffffffff means broadcast to all SEs or SHs (CIK). - * + * @xcc_id: xcc accelerated compute core id * Select which SE, SH combinations to address. */ static void gfx_v7_0_select_se_sh(struct amdgpu_device *adev, - u32 se_num, u32 sh_num, u32 instance) + u32 se_num, u32 sh_num, u32 instance, + int xcc_id) { u32 data; @@ -1732,13 +1733,13 @@ gfx_v7_0_write_harvested_raster_configs(struct amdgpu_device *adev, } /* GRBM_GFX_INDEX has a different offset on CI+ */ - gfx_v7_0_select_se_sh(adev, se, 0xffffffff, 0xffffffff); + gfx_v7_0_select_se_sh(adev, se, 0xffffffff, 0xffffffff, 0); WREG32(mmPA_SC_RASTER_CONFIG, raster_config_se); WREG32(mmPA_SC_RASTER_CONFIG_1, raster_config_1); } /* GRBM_GFX_INDEX has a different offset on CI+ */ - gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); } /** @@ -1761,13 +1762,13 @@ static void gfx_v7_0_setup_rb(struct amdgpu_device *adev) mutex_lock(&adev->grbm_idx_mutex); for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { - gfx_v7_0_select_se_sh(adev, i, j, 0xffffffff); + gfx_v7_0_select_se_sh(adev, i, j, 0xffffffff, 0); data = gfx_v7_0_get_rb_active_bitmap(adev); active_rbs |= data << ((i * adev->gfx.config.max_sh_per_se + j) * rb_bitmap_width_per_sh); } } - gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); adev->gfx.config.backend_enable_mask = active_rbs; adev->gfx.config.num_rbs = hweight32(active_rbs); @@ -1790,7 +1791,7 @@ static void gfx_v7_0_setup_rb(struct amdgpu_device *adev) /* cache the values for userspace */ for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { - gfx_v7_0_select_se_sh(adev, i, j, 0xffffffff); + gfx_v7_0_select_se_sh(adev, i, j, 0xffffffff, 0); adev->gfx.config.rb_config[i][j].rb_backend_disable = RREG32(mmCC_RB_BACKEND_DISABLE); adev->gfx.config.rb_config[i][j].user_rb_backend_disable = @@ -1801,7 +1802,7 @@ static void gfx_v7_0_setup_rb(struct amdgpu_device *adev) RREG32(mmPA_SC_RASTER_CONFIG_1); } } - gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); } @@ -1911,7 +1912,7 @@ static void gfx_v7_0_constants_init(struct amdgpu_device *adev) * making sure that the following register writes will be broadcasted * to all the shaders */ - gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); /* XXX SH_MEM regs */ /* where to put LDS, scratch, GPUVM in FSA64 space */ @@ -2728,7 +2729,7 @@ static int gfx_v7_0_mec_init(struct amdgpu_device *adev) u32 *hpd; size_t mec_hpd_size; - bitmap_zero(adev->gfx.mec.queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES); + bitmap_zero(adev->gfx.mec_bitmap[0].queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES); /* take ownership of the relevant compute queues */ amdgpu_gfx_compute_queue_acquire(adev); @@ -3301,7 +3302,7 @@ static void gfx_v7_0_wait_for_rlc_serdes(struct amdgpu_device *adev) mutex_lock(&adev->grbm_idx_mutex); for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { - gfx_v7_0_select_se_sh(adev, i, j, 0xffffffff); + gfx_v7_0_select_se_sh(adev, i, j, 0xffffffff, 0); for (k = 0; k < adev->usec_timeout; k++) { if (RREG32(mmRLC_SERDES_CU_MASTER_BUSY) == 0) break; @@ -3309,7 +3310,7 @@ static void gfx_v7_0_wait_for_rlc_serdes(struct amdgpu_device *adev) } } } - gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); mask = RLC_SERDES_NONCU_MASTER_BUSY__SE_MASTER_BUSY_MASK | @@ -3361,7 +3362,7 @@ static bool gfx_v7_0_is_rlc_enabled(struct amdgpu_device *adev) return true; } -static void gfx_v7_0_set_safe_mode(struct amdgpu_device *adev) +static void gfx_v7_0_set_safe_mode(struct amdgpu_device *adev, int xcc_id) { u32 tmp, i, mask; @@ -3383,7 +3384,7 @@ static void gfx_v7_0_set_safe_mode(struct amdgpu_device *adev) } } -static void gfx_v7_0_unset_safe_mode(struct amdgpu_device *adev) +static void gfx_v7_0_unset_safe_mode(struct amdgpu_device *adev, int xcc_id) { u32 tmp; @@ -3474,7 +3475,7 @@ static int gfx_v7_0_rlc_resume(struct amdgpu_device *adev) WREG32(mmRLC_LB_CNTR_MAX, 0x00008000); mutex_lock(&adev->grbm_idx_mutex); - gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); WREG32(mmRLC_LB_INIT_CU_MASK, 0xffffffff); WREG32(mmRLC_LB_PARAMS, 0x00600408); WREG32(mmRLC_LB_CNTL, 0x80000004); @@ -3530,7 +3531,7 @@ static void gfx_v7_0_enable_cgcg(struct amdgpu_device *adev, bool enable) tmp = gfx_v7_0_halt_rlc(adev); mutex_lock(&adev->grbm_idx_mutex); - gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); WREG32(mmRLC_SERDES_WR_CU_MASTER_MASK, 0xffffffff); WREG32(mmRLC_SERDES_WR_NONCU_MASTER_MASK, 0xffffffff); tmp2 = RLC_SERDES_WR_CTRL__BPM_ADDR_MASK | @@ -3584,7 +3585,7 @@ static void gfx_v7_0_enable_mgcg(struct amdgpu_device *adev, bool enable) tmp = gfx_v7_0_halt_rlc(adev); mutex_lock(&adev->grbm_idx_mutex); - gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); WREG32(mmRLC_SERDES_WR_CU_MASTER_MASK, 0xffffffff); WREG32(mmRLC_SERDES_WR_NONCU_MASTER_MASK, 0xffffffff); data = RLC_SERDES_WR_CTRL__BPM_ADDR_MASK | @@ -3635,7 +3636,7 @@ static void gfx_v7_0_enable_mgcg(struct amdgpu_device *adev, bool enable) tmp = gfx_v7_0_halt_rlc(adev); mutex_lock(&adev->grbm_idx_mutex); - gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); WREG32(mmRLC_SERDES_WR_CU_MASTER_MASK, 0xffffffff); WREG32(mmRLC_SERDES_WR_NONCU_MASTER_MASK, 0xffffffff); data = RLC_SERDES_WR_CTRL__BPM_ADDR_MASK | RLC_SERDES_WR_CTRL__MGCG_OVERRIDE_1_MASK; @@ -4111,7 +4112,7 @@ static void wave_read_regs(struct amdgpu_device *adev, uint32_t simd, *(out++) = RREG32(mmSQ_IND_DATA); } -static void gfx_v7_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields) +static void gfx_v7_0_read_wave_data(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields) { /* type 0 wave data */ dst[(*no_fields)++] = 0; @@ -4136,7 +4137,7 @@ static void gfx_v7_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, u dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_MODE); } -static void gfx_v7_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd, +static void gfx_v7_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t start, uint32_t size, uint32_t *dst) { @@ -4146,7 +4147,7 @@ static void gfx_v7_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd, } static void gfx_v7_0_select_me_pipe_q(struct amdgpu_device *adev, - u32 me, u32 pipe, u32 q, u32 vm) + u32 me, u32 pipe, u32 q, u32 vm, u32 xcc_id) { cik_srbm_select(adev, me, pipe, q, vm); } @@ -4178,6 +4179,7 @@ static int gfx_v7_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + adev->gfx.xcc_mask = 1; adev->gfx.num_gfx_rings = GFX7_NUM_GFX_RINGS; adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev), AMDGPU_MAX_COMPUTE_RINGS); @@ -4456,7 +4458,8 @@ static int gfx_v7_0_sw_init(void *handle) for (i = 0; i < adev->gfx.mec.num_mec; ++i) { for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) { for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) { - if (!amdgpu_gfx_is_mec_queue_enabled(adev, i, k, j)) + if (!amdgpu_gfx_is_mec_queue_enabled(adev, 0, i, + k, j)) continue; r = gfx_v7_0_compute_ring_init(adev, @@ -5114,7 +5117,7 @@ static void gfx_v7_0_get_cu_info(struct amdgpu_device *adev) mask = 1; ao_bitmap = 0; counter = 0; - gfx_v7_0_select_se_sh(adev, i, j, 0xffffffff); + gfx_v7_0_select_se_sh(adev, i, j, 0xffffffff, 0); if (i < 4 && j < 2) gfx_v7_0_set_user_cu_inactive_bitmap( adev, disable_masks[i * 2 + j]); @@ -5135,7 +5138,7 @@ static void gfx_v7_0_get_cu_info(struct amdgpu_device *adev) cu_info->ao_cu_bitmap[i][j] = ao_bitmap; } } - gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); cu_info->number = active_cu_number; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index b1f2684d854ad3c614609f01a39512360c146bc1..51c1745c83697b04a7eab230f5a64847fceabed3 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -1304,7 +1304,7 @@ static int gfx_v8_0_mec_init(struct amdgpu_device *adev) u32 *hpd; size_t mec_hpd_size; - bitmap_zero(adev->gfx.mec.queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES); + bitmap_zero(adev->gfx.mec_bitmap[0].queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES); /* take ownership of the relevant compute queues */ amdgpu_gfx_compute_queue_acquire(adev); @@ -2001,7 +2001,8 @@ static int gfx_v8_0_sw_init(void *handle) for (i = 0; i < adev->gfx.mec.num_mec; ++i) { for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) { for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) { - if (!amdgpu_gfx_is_mec_queue_enabled(adev, i, k, j)) + if (!amdgpu_gfx_is_mec_queue_enabled(adev, 0, i, + k, j)) continue; r = gfx_v8_0_compute_ring_init(adev, @@ -2015,19 +2016,19 @@ static int gfx_v8_0_sw_init(void *handle) } } - r = amdgpu_gfx_kiq_init(adev, GFX8_MEC_HPD_SIZE); + r = amdgpu_gfx_kiq_init(adev, GFX8_MEC_HPD_SIZE, 0); if (r) { DRM_ERROR("Failed to init KIQ BOs!\n"); return r; } - kiq = &adev->gfx.kiq; - r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq); + kiq = &adev->gfx.kiq[0]; + r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq, 0); if (r) return r; /* create MQD for all compute queues as well as KIQ for SRIOV case */ - r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct vi_mqd_allocation)); + r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct vi_mqd_allocation), 0); if (r) return r; @@ -2050,9 +2051,9 @@ static int gfx_v8_0_sw_fini(void *handle) for (i = 0; i < adev->gfx.num_compute_rings; i++) amdgpu_ring_fini(&adev->gfx.compute_ring[i]); - amdgpu_gfx_mqd_sw_fini(adev); - amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq.ring); - amdgpu_gfx_kiq_fini(adev); + amdgpu_gfx_mqd_sw_fini(adev, 0); + amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[0].ring); + amdgpu_gfx_kiq_fini(adev, 0); gfx_v8_0_mec_fini(adev); amdgpu_gfx_rlc_fini(adev); @@ -3394,7 +3395,8 @@ static void gfx_v8_0_tiling_mode_table_init(struct amdgpu_device *adev) } static void gfx_v8_0_select_se_sh(struct amdgpu_device *adev, - u32 se_num, u32 sh_num, u32 instance) + u32 se_num, u32 sh_num, u32 instance, + int xcc_id) { u32 data; @@ -3417,7 +3419,7 @@ static void gfx_v8_0_select_se_sh(struct amdgpu_device *adev, } static void gfx_v8_0_select_me_pipe_q(struct amdgpu_device *adev, - u32 me, u32 pipe, u32 q, u32 vm) + u32 me, u32 pipe, u32 q, u32 vm, u32 xcc_id) { vi_srbm_select(adev, me, pipe, q, vm); } @@ -3578,13 +3580,13 @@ gfx_v8_0_write_harvested_raster_configs(struct amdgpu_device *adev, } /* GRBM_GFX_INDEX has a different offset on VI */ - gfx_v8_0_select_se_sh(adev, se, 0xffffffff, 0xffffffff); + gfx_v8_0_select_se_sh(adev, se, 0xffffffff, 0xffffffff, 0); WREG32(mmPA_SC_RASTER_CONFIG, raster_config_se); WREG32(mmPA_SC_RASTER_CONFIG_1, raster_config_1); } /* GRBM_GFX_INDEX has a different offset on VI */ - gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); } static void gfx_v8_0_setup_rb(struct amdgpu_device *adev) @@ -3600,13 +3602,13 @@ static void gfx_v8_0_setup_rb(struct amdgpu_device *adev) mutex_lock(&adev->grbm_idx_mutex); for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { - gfx_v8_0_select_se_sh(adev, i, j, 0xffffffff); + gfx_v8_0_select_se_sh(adev, i, j, 0xffffffff, 0); data = gfx_v8_0_get_rb_active_bitmap(adev); active_rbs |= data << ((i * adev->gfx.config.max_sh_per_se + j) * rb_bitmap_width_per_sh); } } - gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); adev->gfx.config.backend_enable_mask = active_rbs; adev->gfx.config.num_rbs = hweight32(active_rbs); @@ -3629,7 +3631,7 @@ static void gfx_v8_0_setup_rb(struct amdgpu_device *adev) /* cache the values for userspace */ for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { - gfx_v8_0_select_se_sh(adev, i, j, 0xffffffff); + gfx_v8_0_select_se_sh(adev, i, j, 0xffffffff, 0); adev->gfx.config.rb_config[i][j].rb_backend_disable = RREG32(mmCC_RB_BACKEND_DISABLE); adev->gfx.config.rb_config[i][j].user_rb_backend_disable = @@ -3640,7 +3642,7 @@ static void gfx_v8_0_setup_rb(struct amdgpu_device *adev) RREG32(mmPA_SC_RASTER_CONFIG_1); } } - gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); } @@ -3787,7 +3789,7 @@ static void gfx_v8_0_constants_init(struct amdgpu_device *adev) * making sure that the following register writes will be broadcasted * to all the shaders */ - gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); WREG32(mmPA_SC_FIFO_SIZE, (adev->gfx.config.sc_prim_fifo_size_frontend << @@ -3818,7 +3820,7 @@ static void gfx_v8_0_wait_for_rlc_serdes(struct amdgpu_device *adev) mutex_lock(&adev->grbm_idx_mutex); for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { - gfx_v8_0_select_se_sh(adev, i, j, 0xffffffff); + gfx_v8_0_select_se_sh(adev, i, j, 0xffffffff, 0); for (k = 0; k < adev->usec_timeout; k++) { if (RREG32(mmRLC_SERDES_CU_MASTER_BUSY) == 0) break; @@ -3826,7 +3828,7 @@ static void gfx_v8_0_wait_for_rlc_serdes(struct amdgpu_device *adev) } if (k == adev->usec_timeout) { gfx_v8_0_select_se_sh(adev, 0xffffffff, - 0xffffffff, 0xffffffff); + 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); DRM_INFO("Timeout wait for RLC serdes %u,%u\n", i, j); @@ -3834,7 +3836,7 @@ static void gfx_v8_0_wait_for_rlc_serdes(struct amdgpu_device *adev) } } } - gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); mask = RLC_SERDES_NONCU_MASTER_BUSY__SE_MASTER_BUSY_MASK | @@ -4281,7 +4283,6 @@ static int gfx_v8_0_cp_gfx_resume(struct amdgpu_device *adev) /* start the ring */ amdgpu_ring_clear_ring(ring); gfx_v8_0_cp_gfx_start(adev); - ring->sched.ready = true; return 0; } @@ -4292,7 +4293,7 @@ static void gfx_v8_0_cp_compute_enable(struct amdgpu_device *adev, bool enable) WREG32(mmCP_MEC_CNTL, 0); } else { WREG32(mmCP_MEC_CNTL, (CP_MEC_CNTL__MEC_ME1_HALT_MASK | CP_MEC_CNTL__MEC_ME2_HALT_MASK)); - adev->gfx.kiq.ring.sched.ready = false; + adev->gfx.kiq[0].ring.sched.ready = false; } udelay(50); } @@ -4314,12 +4315,12 @@ static void gfx_v8_0_kiq_setting(struct amdgpu_ring *ring) static int gfx_v8_0_kiq_kcq_enable(struct amdgpu_device *adev) { - struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring; + struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring; uint64_t queue_mask = 0; int r, i; for (i = 0; i < AMDGPU_MAX_COMPUTE_QUEUES; ++i) { - if (!test_bit(i, adev->gfx.mec.queue_bitmap)) + if (!test_bit(i, adev->gfx.mec_bitmap[0].queue_bitmap)) continue; /* This situation may be hit in the future if a new HW @@ -4595,14 +4596,13 @@ static int gfx_v8_0_kiq_init_queue(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; struct vi_mqd *mqd = ring->mqd_ptr; - int mqd_idx = AMDGPU_MAX_COMPUTE_RINGS; gfx_v8_0_kiq_setting(ring); if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */ /* reset MQD to a clean status */ - if (adev->gfx.mec.mqd_backup[mqd_idx]) - memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct vi_mqd_allocation)); + if (adev->gfx.kiq[0].mqd_backup) + memcpy(mqd, adev->gfx.kiq[0].mqd_backup, sizeof(struct vi_mqd_allocation)); /* reset ring buffer */ ring->wptr = 0; @@ -4625,8 +4625,8 @@ static int gfx_v8_0_kiq_init_queue(struct amdgpu_ring *ring) vi_srbm_select(adev, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); - if (adev->gfx.mec.mqd_backup[mqd_idx]) - memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(struct vi_mqd_allocation)); + if (adev->gfx.kiq[0].mqd_backup) + memcpy(adev->gfx.kiq[0].mqd_backup, mqd, sizeof(struct vi_mqd_allocation)); } return 0; @@ -4650,15 +4650,13 @@ static int gfx_v8_0_kcq_init_queue(struct amdgpu_ring *ring) if (adev->gfx.mec.mqd_backup[mqd_idx]) memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(struct vi_mqd_allocation)); - } else if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */ - /* reset MQD to a clean status */ + } else { + /* restore MQD to a clean status */ if (adev->gfx.mec.mqd_backup[mqd_idx]) memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct vi_mqd_allocation)); /* reset ring buffer */ ring->wptr = 0; amdgpu_ring_clear_ring(ring); - } else { - amdgpu_ring_clear_ring(ring); } return 0; } @@ -4678,21 +4676,22 @@ static int gfx_v8_0_kiq_resume(struct amdgpu_device *adev) struct amdgpu_ring *ring; int r; - ring = &adev->gfx.kiq.ring; + ring = &adev->gfx.kiq[0].ring; r = amdgpu_bo_reserve(ring->mqd_obj, false); if (unlikely(r != 0)) return r; r = amdgpu_bo_kmap(ring->mqd_obj, &ring->mqd_ptr); - if (unlikely(r != 0)) + if (unlikely(r != 0)) { + amdgpu_bo_unreserve(ring->mqd_obj); return r; + } gfx_v8_0_kiq_init_queue(ring); amdgpu_bo_kunmap(ring->mqd_obj); ring->mqd_ptr = NULL; amdgpu_bo_unreserve(ring->mqd_obj); - ring->sched.ready = true; return 0; } @@ -4741,7 +4740,7 @@ static int gfx_v8_0_cp_test_all_rings(struct amdgpu_device *adev) if (r) return r; - ring = &adev->gfx.kiq.ring; + ring = &adev->gfx.kiq[0].ring; r = amdgpu_ring_test_helper(ring); if (r) return r; @@ -4808,7 +4807,7 @@ static int gfx_v8_0_hw_init(void *handle) static int gfx_v8_0_kcq_disable(struct amdgpu_device *adev) { int r, i; - struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring; + struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring; r = amdgpu_ring_alloc(kiq_ring, 6 * adev->gfx.num_compute_rings); if (r) @@ -4902,7 +4901,7 @@ static int gfx_v8_0_hw_fini(void *handle) pr_debug("For SRIOV client, shouldn't do anything.\n"); return 0; } - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); if (!gfx_v8_0_wait_for_idle(adev)) gfx_v8_0_cp_enable(adev, false); else @@ -4911,7 +4910,7 @@ static int gfx_v8_0_hw_fini(void *handle) adev->gfx.rlc.funcs->stop(adev); else pr_err("rlc is busy, skip halt rlc\n"); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); return 0; } @@ -5216,7 +5215,7 @@ static void wave_read_regs(struct amdgpu_device *adev, uint32_t simd, *(out++) = RREG32(mmSQ_IND_DATA); } -static void gfx_v8_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields) +static void gfx_v8_0_read_wave_data(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields) { /* type 0 wave data */ dst[(*no_fields)++] = 0; @@ -5241,7 +5240,7 @@ static void gfx_v8_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, u dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_MODE); } -static void gfx_v8_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd, +static void gfx_v8_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t start, uint32_t size, uint32_t *dst) { @@ -5263,6 +5262,7 @@ static int gfx_v8_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + adev->gfx.xcc_mask = 1; adev->gfx.num_gfx_rings = GFX8_NUM_GFX_RINGS; adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev), AMDGPU_MAX_COMPUTE_RINGS); @@ -5376,7 +5376,7 @@ static int gfx_v8_0_set_powergating_state(void *handle, AMD_PG_SUPPORT_RLC_SMU_HS | AMD_PG_SUPPORT_CP | AMD_PG_SUPPORT_GFX_DMG)) - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); switch (adev->asic_type) { case CHIP_CARRIZO: case CHIP_STONEY: @@ -5430,7 +5430,7 @@ static int gfx_v8_0_set_powergating_state(void *handle, AMD_PG_SUPPORT_RLC_SMU_HS | AMD_PG_SUPPORT_CP | AMD_PG_SUPPORT_GFX_DMG)) - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); return 0; } @@ -5481,7 +5481,7 @@ static void gfx_v8_0_send_serdes_cmd(struct amdgpu_device *adev, { uint32_t data; - gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); WREG32(mmRLC_SERDES_WR_CU_MASTER_MASK, 0xffffffff); WREG32(mmRLC_SERDES_WR_NONCU_MASTER_MASK, 0xffffffff); @@ -5535,7 +5535,7 @@ static bool gfx_v8_0_is_rlc_enabled(struct amdgpu_device *adev) return true; } -static void gfx_v8_0_set_safe_mode(struct amdgpu_device *adev) +static void gfx_v8_0_set_safe_mode(struct amdgpu_device *adev, int xcc_id) { uint32_t data; unsigned i; @@ -5562,7 +5562,7 @@ static void gfx_v8_0_set_safe_mode(struct amdgpu_device *adev) } } -static void gfx_v8_0_unset_safe_mode(struct amdgpu_device *adev) +static void gfx_v8_0_unset_safe_mode(struct amdgpu_device *adev, int xcc_id) { uint32_t data; unsigned i; @@ -5621,7 +5621,7 @@ static void gfx_v8_0_update_medium_grain_clock_gating(struct amdgpu_device *adev { uint32_t temp, data; - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); /* It is disabled by HW by default */ if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGCG)) { @@ -5717,7 +5717,7 @@ static void gfx_v8_0_update_medium_grain_clock_gating(struct amdgpu_device *adev gfx_v8_0_wait_for_rlc_serdes(adev); } - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); } static void gfx_v8_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev, @@ -5727,7 +5727,7 @@ static void gfx_v8_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev temp = data = RREG32(mmRLC_CGCG_CGLS_CTRL); - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGCG)) { temp1 = data1 = RREG32(mmRLC_CGTT_MGCG_OVERRIDE); @@ -5810,7 +5810,7 @@ static void gfx_v8_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev gfx_v8_0_wait_for_rlc_serdes(adev); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); } static int gfx_v8_0_update_gfx_clock_gating(struct amdgpu_device *adev, bool enable) @@ -6723,11 +6723,11 @@ static void gfx_v8_0_parse_sq_irq(struct amdgpu_device *adev, unsigned ih_data, */ if (from_wq) { mutex_lock(&adev->grbm_idx_mutex); - gfx_v8_0_select_se_sh(adev, se_id, sh_id, cu_id); + gfx_v8_0_select_se_sh(adev, se_id, sh_id, cu_id, 0); sq_edc_source = REG_GET_FIELD(RREG32(mmSQ_EDC_INFO), SQ_EDC_INFO, SOURCE); - gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); } @@ -7001,7 +7001,7 @@ static void gfx_v8_0_set_ring_funcs(struct amdgpu_device *adev) { int i; - adev->gfx.kiq.ring.funcs = &gfx_v8_0_ring_funcs_kiq; + adev->gfx.kiq[0].ring.funcs = &gfx_v8_0_ring_funcs_kiq; for (i = 0; i < adev->gfx.num_gfx_rings; i++) adev->gfx.gfx_ring[i].funcs = &gfx_v8_0_ring_funcs_gfx; @@ -7116,7 +7116,7 @@ static void gfx_v8_0_get_cu_info(struct amdgpu_device *adev) mask = 1; ao_bitmap = 0; counter = 0; - gfx_v8_0_select_se_sh(adev, i, j, 0xffffffff); + gfx_v8_0_select_se_sh(adev, i, j, 0xffffffff, 0); if (i < 4 && j < 2) gfx_v8_0_set_user_cu_inactive_bitmap( adev, disable_masks[i * 2 + j]); @@ -7137,7 +7137,7 @@ static void gfx_v8_0_get_cu_info(struct amdgpu_device *adev) cu_info->ao_cu_bitmap[i][j] = ao_bitmap; } } - gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); cu_info->number = active_cu_number; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index f46d4b18a3fa35f9be4f18fcb0924e01f0beedde..65577eca58f1c85d44f55b2f10bf9300ee97f643 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -149,16 +149,6 @@ MODULE_FIRMWARE("amdgpu/aldebaran_sjt_mec2.bin"); #define mmGOLDEN_TSC_COUNT_LOWER_Renoir 0x0026 #define mmGOLDEN_TSC_COUNT_LOWER_Renoir_BASE_IDX 1 -#define mmGOLDEN_TSC_COUNT_UPPER_Raven 0x007a -#define mmGOLDEN_TSC_COUNT_UPPER_Raven_BASE_IDX 0 -#define mmGOLDEN_TSC_COUNT_LOWER_Raven 0x007b -#define mmGOLDEN_TSC_COUNT_LOWER_Raven_BASE_IDX 0 - -#define mmGOLDEN_TSC_COUNT_UPPER_Raven2 0x0068 -#define mmGOLDEN_TSC_COUNT_UPPER_Raven2_BASE_IDX 0 -#define mmGOLDEN_TSC_COUNT_LOWER_Raven2 0x0069 -#define mmGOLDEN_TSC_COUNT_LOWER_Raven2_BASE_IDX 0 - enum ta_ras_gfx_subblock { /*CPC*/ TA_RAS_BLOCK__GFX_CPC_INDEX_START = 0, @@ -765,12 +755,12 @@ static void gfx_v9_0_set_rlc_funcs(struct amdgpu_device *adev); static int gfx_v9_0_get_cu_info(struct amdgpu_device *adev, struct amdgpu_cu_info *cu_info); static uint64_t gfx_v9_0_get_gpu_clock_counter(struct amdgpu_device *adev); -static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume); +static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume, bool usegds); static u64 gfx_v9_0_ring_get_rptr_compute(struct amdgpu_ring *ring); static void gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev, void *ras_error_status); static int gfx_v9_0_ras_error_inject(struct amdgpu_device *adev, - void *inject_if); + void *inject_if, uint32_t instance_mask); static void gfx_v9_0_reset_ras_error_count(struct amdgpu_device *adev); static void gfx_v9_0_kiq_set_resources(struct amdgpu_ring *kiq_ring, @@ -898,7 +888,7 @@ static const struct kiq_pm4_funcs gfx_v9_0_kiq_pm4_funcs = { static void gfx_v9_0_set_kiq_pm4_funcs(struct amdgpu_device *adev) { - adev->gfx.kiq.pmf = &gfx_v9_0_kiq_pm4_funcs; + adev->gfx.kiq[0].pmf = &gfx_v9_0_kiq_pm4_funcs; } static void gfx_v9_0_init_golden_registers(struct amdgpu_device *adev) @@ -1504,7 +1494,7 @@ static void gfx_v9_0_init_always_on_cu_mask(struct amdgpu_device *adev) mask = 1; cu_bitmap = 0; counter = 0; - amdgpu_gfx_select_se_sh(adev, i, j, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, i, j, 0xffffffff, 0); for (k = 0; k < adev->gfx.config.max_cu_per_sh; k ++) { if (cu_info->bitmap[i][j] & mask) { @@ -1523,7 +1513,7 @@ static void gfx_v9_0_init_always_on_cu_mask(struct amdgpu_device *adev) cu_info->ao_cu_bitmap[i][j] = cu_bitmap; } } - amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); } @@ -1545,7 +1535,7 @@ static void gfx_v9_0_init_lbpw(struct amdgpu_device *adev) mutex_lock(&adev->grbm_idx_mutex); /* set mmRLC_LB_INIT_CU_MASK thru broadcast mode to enable all SE/SH*/ - amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); WREG32_SOC15(GC, 0, mmRLC_LB_INIT_CU_MASK, 0xffffffff); /* set mmRLC_LB_PARAMS = 0x003F_1006 */ @@ -1594,7 +1584,7 @@ static void gfx_v9_4_init_lbpw(struct amdgpu_device *adev) mutex_lock(&adev->grbm_idx_mutex); /* set mmRLC_LB_INIT_CU_MASK thru broadcast mode to enable all SE/SH*/ - amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); WREG32_SOC15(GC, 0, mmRLC_LB_INIT_CU_MASK, 0xffffffff); /* set mmRLC_LB_PARAMS = 0x003F_1006 */ @@ -1713,7 +1703,7 @@ static int gfx_v9_0_mec_init(struct amdgpu_device *adev) const struct gfx_firmware_header_v1_0 *mec_hdr; - bitmap_zero(adev->gfx.mec.queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES); + bitmap_zero(adev->gfx.mec_bitmap[0].queue_bitmap, AMDGPU_MAX_COMPUTE_QUEUES); /* take ownership of the relevant compute queues */ amdgpu_gfx_compute_queue_acquire(adev); @@ -1788,7 +1778,7 @@ static void wave_read_regs(struct amdgpu_device *adev, uint32_t simd, *(out++) = RREG32_SOC15(GC, 0, mmSQ_IND_DATA); } -static void gfx_v9_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields) +static void gfx_v9_0_read_wave_data(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields) { /* type 1 wave data */ dst[(*no_fields)++] = 1; @@ -1809,7 +1799,7 @@ static void gfx_v9_0_read_wave_data(struct amdgpu_device *adev, uint32_t simd, u dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_MODE); } -static void gfx_v9_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd, +static void gfx_v9_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t start, uint32_t size, uint32_t *dst) { @@ -1818,7 +1808,7 @@ static void gfx_v9_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd, start + SQIND_WAVE_SGPRS_OFFSET, size, dst); } -static void gfx_v9_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd, +static void gfx_v9_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t thread, uint32_t start, uint32_t size, uint32_t *dst) @@ -1829,9 +1819,9 @@ static void gfx_v9_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd, } static void gfx_v9_0_select_me_pipe_q(struct amdgpu_device *adev, - u32 me, u32 pipe, u32 q, u32 vm) + u32 me, u32 pipe, u32 q, u32 vm, u32 xcc_id) { - soc15_grbm_select(adev, me, pipe, q, vm); + soc15_grbm_select(adev, me, pipe, q, vm, 0); } static const struct amdgpu_gfx_funcs gfx_v9_0_gfx_funcs = { @@ -2005,7 +1995,7 @@ static int gfx_v9_0_compute_ring_init(struct amdgpu_device *adev, int ring_id, ring->doorbell_index = (adev->doorbell_index.mec_ring0 + ring_id) << 1; ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr + (ring_id * GFX9_MEC_HPD_SIZE); - ring->vm_hub = AMDGPU_GFXHUB_0; + ring->vm_hub = AMDGPU_GFXHUB(0); sprintf(ring->name, "comp_%d.%d.%d", ring->me, ring->pipe, ring->queue); irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP @@ -2105,7 +2095,7 @@ static int gfx_v9_0_sw_init(void *handle) /* disable scheduler on the real ring */ ring->no_scheduler = true; - ring->vm_hub = AMDGPU_GFXHUB_0; + ring->vm_hub = AMDGPU_GFXHUB(0); r = amdgpu_ring_init(adev, ring, 1024, &adev->gfx.eop_irq, AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP, AMDGPU_RING_PRIO_DEFAULT, NULL); @@ -2123,7 +2113,7 @@ static int gfx_v9_0_sw_init(void *handle) ring->doorbell_index = adev->doorbell_index.gfx_ring0 << 1; ring->is_sw_ring = true; hw_prio = amdgpu_sw_ring_priority(i); - ring->vm_hub = AMDGPU_GFXHUB_0; + ring->vm_hub = AMDGPU_GFXHUB(0); r = amdgpu_ring_init(adev, ring, 1024, &adev->gfx.eop_irq, AMDGPU_CP_IRQ_GFX_ME0_PIPE0_EOP, hw_prio, NULL); @@ -2154,7 +2144,8 @@ static int gfx_v9_0_sw_init(void *handle) for (i = 0; i < adev->gfx.mec.num_mec; ++i) { for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) { for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) { - if (!amdgpu_gfx_is_mec_queue_enabled(adev, i, k, j)) + if (!amdgpu_gfx_is_mec_queue_enabled(adev, 0, i, + k, j)) continue; r = gfx_v9_0_compute_ring_init(adev, @@ -2168,19 +2159,19 @@ static int gfx_v9_0_sw_init(void *handle) } } - r = amdgpu_gfx_kiq_init(adev, GFX9_MEC_HPD_SIZE); + r = amdgpu_gfx_kiq_init(adev, GFX9_MEC_HPD_SIZE, 0); if (r) { DRM_ERROR("Failed to init KIQ BOs!\n"); return r; } - kiq = &adev->gfx.kiq; - r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq); + kiq = &adev->gfx.kiq[0]; + r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq, 0); if (r) return r; /* create MQD for all compute queues as wel as KIQ for SRIOV case */ - r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct v9_mqd_allocation)); + r = amdgpu_gfx_mqd_sw_init(adev, sizeof(struct v9_mqd_allocation), 0); if (r) return r; @@ -2215,9 +2206,9 @@ static int gfx_v9_0_sw_fini(void *handle) for (i = 0; i < adev->gfx.num_compute_rings; i++) amdgpu_ring_fini(&adev->gfx.compute_ring[i]); - amdgpu_gfx_mqd_sw_fini(adev); - amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq.ring); - amdgpu_gfx_kiq_fini(adev); + amdgpu_gfx_mqd_sw_fini(adev, 0); + amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[0].ring); + amdgpu_gfx_kiq_fini(adev, 0); gfx_v9_0_mec_fini(adev); amdgpu_bo_free_kernel(&adev->gfx.rlc.clear_state_obj, @@ -2240,7 +2231,7 @@ static void gfx_v9_0_tiling_mode_table_init(struct amdgpu_device *adev) } void gfx_v9_0_select_se_sh(struct amdgpu_device *adev, u32 se_num, u32 sh_num, - u32 instance) + u32 instance, int xcc_id) { u32 data; @@ -2289,19 +2280,42 @@ static void gfx_v9_0_setup_rb(struct amdgpu_device *adev) mutex_lock(&adev->grbm_idx_mutex); for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { - amdgpu_gfx_select_se_sh(adev, i, j, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, i, j, 0xffffffff, 0); data = gfx_v9_0_get_rb_active_bitmap(adev); active_rbs |= data << ((i * adev->gfx.config.max_sh_per_se + j) * rb_bitmap_width_per_sh); } } - amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); adev->gfx.config.backend_enable_mask = active_rbs; adev->gfx.config.num_rbs = hweight32(active_rbs); } +static void gfx_v9_0_debug_trap_config_init(struct amdgpu_device *adev, + uint32_t first_vmid, + uint32_t last_vmid) +{ + uint32_t data; + uint32_t trap_config_vmid_mask = 0; + int i; + + /* Calculate trap config vmid mask */ + for (i = first_vmid; i < last_vmid; i++) + trap_config_vmid_mask |= (1 << i); + + data = REG_SET_FIELD(0, SPI_GDBG_TRAP_CONFIG, + VMID_SEL, trap_config_vmid_mask); + data = REG_SET_FIELD(data, SPI_GDBG_TRAP_CONFIG, + TRAP_EN, 1); + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_CONFIG), data); + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_MASK), 0); + + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_DATA0), 0); + WREG32(SOC15_REG_OFFSET(GC, 0, mmSPI_GDBG_TRAP_DATA1), 0); +} + #define DEFAULT_SH_MEM_BASES (0x6000) static void gfx_v9_0_init_compute_vmid(struct amdgpu_device *adev) { @@ -2323,12 +2337,12 @@ static void gfx_v9_0_init_compute_vmid(struct amdgpu_device *adev) mutex_lock(&adev->srbm_mutex); for (i = adev->vm_manager.first_kfd_vmid; i < AMDGPU_NUM_VMID; i++) { - soc15_grbm_select(adev, 0, 0, 0, i); + soc15_grbm_select(adev, 0, 0, 0, i, 0); /* CP and shaders */ WREG32_SOC15_RLC(GC, 0, mmSH_MEM_CONFIG, sh_mem_config); WREG32_SOC15_RLC(GC, 0, mmSH_MEM_BASES, sh_mem_bases); } - soc15_grbm_select(adev, 0, 0, 0, 0); + soc15_grbm_select(adev, 0, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); /* Initialize all compute VMIDs to have no GDS, GWS, or OA @@ -2366,8 +2380,8 @@ static void gfx_v9_0_init_sq_config(struct amdgpu_device *adev) switch (adev->ip_versions[GC_HWIP][0]) { case IP_VERSION(9, 4, 1): tmp = RREG32_SOC15(GC, 0, mmSQ_CONFIG); - tmp = REG_SET_FIELD(tmp, SQ_CONFIG, - DISABLE_BARRIER_WAITCNT, 1); + tmp = REG_SET_FIELD(tmp, SQ_CONFIG, DISABLE_BARRIER_WAITCNT, + !READ_ONCE(adev->barrier_has_auto_waitcnt)); WREG32_SOC15(GC, 0, mmSQ_CONFIG, tmp); break; default: @@ -2392,8 +2406,8 @@ static void gfx_v9_0_constants_init(struct amdgpu_device *adev) /* XXX SH_MEM regs */ /* where to put LDS, scratch, GPUVM in FSA64 space */ mutex_lock(&adev->srbm_mutex); - for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB_0].num_ids; i++) { - soc15_grbm_select(adev, 0, 0, 0, i); + for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB(0)].num_ids; i++) { + soc15_grbm_select(adev, 0, 0, 0, i, 0); /* CP and shaders */ if (i == 0) { tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE, @@ -2415,7 +2429,7 @@ static void gfx_v9_0_constants_init(struct amdgpu_device *adev) WREG32_SOC15_RLC(GC, 0, mmSH_MEM_BASES, tmp); } } - soc15_grbm_select(adev, 0, 0, 0, 0); + soc15_grbm_select(adev, 0, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); @@ -2432,7 +2446,7 @@ static void gfx_v9_0_wait_for_rlc_serdes(struct amdgpu_device *adev) mutex_lock(&adev->grbm_idx_mutex); for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { - amdgpu_gfx_select_se_sh(adev, i, j, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, i, j, 0xffffffff, 0); for (k = 0; k < adev->usec_timeout; k++) { if (RREG32_SOC15(GC, 0, mmRLC_SERDES_CU_MASTER_BUSY) == 0) break; @@ -2440,7 +2454,7 @@ static void gfx_v9_0_wait_for_rlc_serdes(struct amdgpu_device *adev) } if (k == adev->usec_timeout) { amdgpu_gfx_select_se_sh(adev, 0xffffffff, - 0xffffffff, 0xffffffff); + 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); DRM_INFO("Timeout wait for RLC serdes %u,%u\n", i, j); @@ -2448,7 +2462,7 @@ static void gfx_v9_0_wait_for_rlc_serdes(struct amdgpu_device *adev) } } } - amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); mask = RLC_SERDES_NONCU_MASTER_BUSY__SE_MASTER_BUSY_MASK | @@ -3143,7 +3157,6 @@ static int gfx_v9_0_cp_gfx_resume(struct amdgpu_device *adev) /* start the ring */ gfx_v9_0_cp_gfx_start(adev); - ring->sched.ready = true; return 0; } @@ -3155,7 +3168,7 @@ static void gfx_v9_0_cp_compute_enable(struct amdgpu_device *adev, bool enable) } else { WREG32_SOC15_RLC(GC, 0, mmCP_MEC_CNTL, (CP_MEC_CNTL__MEC_ME1_HALT_MASK | CP_MEC_CNTL__MEC_ME2_HALT_MASK)); - adev->gfx.kiq.ring.sched.ready = false; + adev->gfx.kiq[0].ring.sched.ready = false; } udelay(50); } @@ -3519,7 +3532,6 @@ static int gfx_v9_0_kiq_init_queue(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; struct v9_mqd *mqd = ring->mqd_ptr; - int mqd_idx = AMDGPU_MAX_COMPUTE_RINGS; struct v9_mqd *tmp_mqd; gfx_v9_0_kiq_setting(ring); @@ -3529,20 +3541,20 @@ static int gfx_v9_0_kiq_init_queue(struct amdgpu_ring *ring) * driver need to re-init the mqd. * check mqd->cp_hqd_pq_control since this value should not be 0 */ - tmp_mqd = (struct v9_mqd *)adev->gfx.mec.mqd_backup[mqd_idx]; + tmp_mqd = (struct v9_mqd *)adev->gfx.kiq[0].mqd_backup; if (amdgpu_in_reset(adev) && tmp_mqd->cp_hqd_pq_control){ /* for GPU_RESET case , reset MQD to a clean status */ - if (adev->gfx.mec.mqd_backup[mqd_idx]) - memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct v9_mqd_allocation)); + if (adev->gfx.kiq[0].mqd_backup) + memcpy(mqd, adev->gfx.kiq[0].mqd_backup, sizeof(struct v9_mqd_allocation)); /* reset ring buffer */ ring->wptr = 0; amdgpu_ring_clear_ring(ring); mutex_lock(&adev->srbm_mutex); - soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0); + soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, 0); gfx_v9_0_kiq_init_register(ring); - soc15_grbm_select(adev, 0, 0, 0, 0); + soc15_grbm_select(adev, 0, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); } else { memset((void *)mqd, 0, sizeof(struct v9_mqd_allocation)); @@ -3551,14 +3563,14 @@ static int gfx_v9_0_kiq_init_queue(struct amdgpu_ring *ring) if (amdgpu_sriov_vf(adev) && adev->in_suspend) amdgpu_ring_clear_ring(ring); mutex_lock(&adev->srbm_mutex); - soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0); + soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, 0); gfx_v9_0_mqd_init(ring); gfx_v9_0_kiq_init_register(ring); - soc15_grbm_select(adev, 0, 0, 0, 0); + soc15_grbm_select(adev, 0, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); - if (adev->gfx.mec.mqd_backup[mqd_idx]) - memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(struct v9_mqd_allocation)); + if (adev->gfx.kiq[0].mqd_backup) + memcpy(adev->gfx.kiq[0].mqd_backup, mqd, sizeof(struct v9_mqd_allocation)); } return 0; @@ -3582,24 +3594,21 @@ static int gfx_v9_0_kcq_init_queue(struct amdgpu_ring *ring) ((struct v9_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF; ((struct v9_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF; mutex_lock(&adev->srbm_mutex); - soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0); + soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, 0); gfx_v9_0_mqd_init(ring); - soc15_grbm_select(adev, 0, 0, 0, 0); + soc15_grbm_select(adev, 0, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); if (adev->gfx.mec.mqd_backup[mqd_idx]) memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(struct v9_mqd_allocation)); - } else if (amdgpu_in_reset(adev)) { /* for GPU_RESET case */ - /* reset MQD to a clean status */ + } else { + /* restore MQD to a clean status */ if (adev->gfx.mec.mqd_backup[mqd_idx]) memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct v9_mqd_allocation)); - /* reset ring buffer */ ring->wptr = 0; atomic64_set((atomic64_t *)ring->wptr_cpu_addr, 0); amdgpu_ring_clear_ring(ring); - } else { - amdgpu_ring_clear_ring(ring); } return 0; @@ -3610,21 +3619,22 @@ static int gfx_v9_0_kiq_resume(struct amdgpu_device *adev) struct amdgpu_ring *ring; int r; - ring = &adev->gfx.kiq.ring; + ring = &adev->gfx.kiq[0].ring; r = amdgpu_bo_reserve(ring->mqd_obj, false); if (unlikely(r != 0)) return r; r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr); - if (unlikely(r != 0)) + if (unlikely(r != 0)) { + amdgpu_bo_unreserve(ring->mqd_obj); return r; + } gfx_v9_0_kiq_init_queue(ring); amdgpu_bo_kunmap(ring->mqd_obj); ring->mqd_ptr = NULL; amdgpu_bo_unreserve(ring->mqd_obj); - ring->sched.ready = true; return 0; } @@ -3652,7 +3662,7 @@ static int gfx_v9_0_kcq_resume(struct amdgpu_device *adev) goto done; } - r = amdgpu_gfx_enable_kcq(adev); + r = amdgpu_gfx_enable_kcq(adev, 0); done: return r; } @@ -3772,7 +3782,7 @@ static int gfx_v9_0_hw_fini(void *handle) /* DF freeze and kcq disable will fail */ if (!amdgpu_ras_intr_triggered()) /* disable KCQ to avoid CPC touch memory not valid anymore */ - amdgpu_gfx_disable_kcq(adev); + amdgpu_gfx_disable_kcq(adev, 0); if (amdgpu_sriov_vf(adev)) { gfx_v9_0_cp_gfx_enable(adev, false); @@ -3790,11 +3800,11 @@ static int gfx_v9_0_hw_fini(void *handle) */ if (!amdgpu_in_reset(adev) && !adev->in_suspend) { mutex_lock(&adev->srbm_mutex); - soc15_grbm_select(adev, adev->gfx.kiq.ring.me, - adev->gfx.kiq.ring.pipe, - adev->gfx.kiq.ring.queue, 0); - gfx_v9_0_kiq_fini_register(&adev->gfx.kiq.ring); - soc15_grbm_select(adev, 0, 0, 0, 0); + soc15_grbm_select(adev, adev->gfx.kiq[0].ring.me, + adev->gfx.kiq[0].ring.pipe, + adev->gfx.kiq[0].ring.queue, 0, 0); + gfx_v9_0_kiq_fini_register(&adev->gfx.kiq[0].ring); + soc15_grbm_select(adev, 0, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); } @@ -3914,7 +3924,7 @@ static uint64_t gfx_v9_0_kiq_read_clock(struct amdgpu_device *adev) unsigned long flags; uint32_t seq, reg_val_offs = 0; uint64_t value = 0; - struct amdgpu_kiq *kiq = &adev->gfx.kiq; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; struct amdgpu_ring *ring = &kiq->ring; BUG_ON(!ring->funcs->emit_rreg); @@ -4002,36 +4012,6 @@ static uint64_t gfx_v9_0_get_gpu_clock_counter(struct amdgpu_device *adev) preempt_enable(); clock = clock_lo | (clock_hi << 32ULL); break; - case IP_VERSION(9, 1, 0): - preempt_disable(); - clock_hi = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven); - clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven); - hi_check = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven); - /* The PWR TSC clock frequency is 100MHz, which sets 32-bit carry over - * roughly every 42 seconds. - */ - if (hi_check != clock_hi) { - clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven); - clock_hi = hi_check; - } - preempt_enable(); - clock = clock_lo | (clock_hi << 32ULL); - break; - case IP_VERSION(9, 2, 2): - preempt_disable(); - clock_hi = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven2); - clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven2); - hi_check = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_UPPER_Raven2); - /* The PWR TSC clock frequency is 100MHz, which sets 32-bit carry over - * roughly every 42 seconds. - */ - if (hi_check != clock_hi) { - clock_lo = RREG32_SOC15_NO_KIQ(PWR, 0, mmGOLDEN_TSC_COUNT_LOWER_Raven2); - clock_hi = hi_check; - } - preempt_enable(); - clock = clock_lo | (clock_hi << 32ULL); - break; default: amdgpu_gfx_off_ctrl(adev, false); mutex_lock(&adev->gfx.gpu_clock_mutex); @@ -4544,6 +4524,7 @@ static int gfx_v9_0_early_init(void *handle) adev->gfx.num_gfx_rings = 0; else adev->gfx.num_gfx_rings = GFX9_NUM_GFX_RINGS; + adev->gfx.xcc_mask = 1; adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev), AMDGPU_MAX_COMPUTE_RINGS); gfx_v9_0_set_kiq_pm4_funcs(adev); @@ -4609,6 +4590,13 @@ static int gfx_v9_0_late_init(void *handle) if (r) return r; + if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2)) + gfx_v9_4_2_debug_trap_config_init(adev, + adev->vm_manager.first_kfd_vmid, AMDGPU_NUM_VMID); + else + gfx_v9_0_debug_trap_config_init(adev, + adev->vm_manager.first_kfd_vmid, AMDGPU_NUM_VMID); + return 0; } @@ -4624,7 +4612,7 @@ static bool gfx_v9_0_is_rlc_enabled(struct amdgpu_device *adev) return true; } -static void gfx_v9_0_set_safe_mode(struct amdgpu_device *adev) +static void gfx_v9_0_set_safe_mode(struct amdgpu_device *adev, int xcc_id) { uint32_t data; unsigned i; @@ -4641,7 +4629,7 @@ static void gfx_v9_0_set_safe_mode(struct amdgpu_device *adev) } } -static void gfx_v9_0_unset_safe_mode(struct amdgpu_device *adev) +static void gfx_v9_0_unset_safe_mode(struct amdgpu_device *adev, int xcc_id) { uint32_t data; @@ -4652,7 +4640,7 @@ static void gfx_v9_0_unset_safe_mode(struct amdgpu_device *adev) static void gfx_v9_0_update_gfx_cg_power_gating(struct amdgpu_device *adev, bool enable) { - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); if ((adev->pg_flags & AMD_PG_SUPPORT_GFX_PG) && enable) { gfx_v9_0_enable_gfx_cg_power_gating(adev, true); @@ -4664,7 +4652,7 @@ static void gfx_v9_0_update_gfx_cg_power_gating(struct amdgpu_device *adev, gfx_v9_0_enable_gfx_pipeline_powergating(adev, false); } - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); } static void gfx_v9_0_update_gfx_mg_power_gating(struct amdgpu_device *adev, @@ -4691,7 +4679,7 @@ static void gfx_v9_0_update_medium_grain_clock_gating(struct amdgpu_device *adev { uint32_t data, def; - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); /* It is disabled by HW by default */ if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGCG)) { @@ -4758,7 +4746,7 @@ static void gfx_v9_0_update_medium_grain_clock_gating(struct amdgpu_device *adev } } - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); } static void gfx_v9_0_update_3d_clock_gating(struct amdgpu_device *adev, @@ -4769,7 +4757,7 @@ static void gfx_v9_0_update_3d_clock_gating(struct amdgpu_device *adev, if (!adev->gfx.num_gfx_rings) return; - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); /* Enable 3D CGCG/CGLS */ if (enable) { @@ -4813,7 +4801,7 @@ static void gfx_v9_0_update_3d_clock_gating(struct amdgpu_device *adev, WREG32_SOC15(GC, 0, mmRLC_CGCG_CGLS_CTRL_3D, data); } - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); } static void gfx_v9_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev, @@ -4821,7 +4809,7 @@ static void gfx_v9_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev { uint32_t def, data; - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGCG)) { def = data = RREG32_SOC15(GC, 0, mmRLC_CGTT_MGCG_OVERRIDE); @@ -4865,7 +4853,7 @@ static void gfx_v9_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev WREG32_SOC15(GC, 0, mmRLC_CGCG_CGLS_CTRL, data); } - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); } static int gfx_v9_0_update_gfx_clock_gating(struct amdgpu_device *adev, @@ -5165,7 +5153,8 @@ static void gfx_v9_0_ring_emit_ib_gfx(struct amdgpu_ring *ring, gfx_v9_0_ring_emit_de_meta(ring, (!amdgpu_sriov_vf(ring->adev) && flags & AMDGPU_IB_PREEMPTED) ? - true : false); + true : false, + job->gds_size > 0 && job->gds_base != 0); } amdgpu_ring_write(ring, header); @@ -5176,9 +5165,83 @@ static void gfx_v9_0_ring_emit_ib_gfx(struct amdgpu_ring *ring, #endif lower_32_bits(ib->gpu_addr)); amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr)); + amdgpu_ring_ib_on_emit_cntl(ring); amdgpu_ring_write(ring, control); } +static void gfx_v9_0_ring_patch_cntl(struct amdgpu_ring *ring, + unsigned offset) +{ + u32 control = ring->ring[offset]; + + control |= INDIRECT_BUFFER_PRE_RESUME(1); + ring->ring[offset] = control; +} + +static void gfx_v9_0_ring_patch_ce_meta(struct amdgpu_ring *ring, + unsigned offset) +{ + struct amdgpu_device *adev = ring->adev; + void *ce_payload_cpu_addr; + uint64_t payload_offset, payload_size; + + payload_size = sizeof(struct v9_ce_ib_state); + + if (ring->is_mes_queue) { + payload_offset = offsetof(struct amdgpu_mes_ctx_meta_data, + gfx[0].gfx_meta_data) + + offsetof(struct v9_gfx_meta_data, ce_payload); + ce_payload_cpu_addr = + amdgpu_mes_ctx_get_offs_cpu_addr(ring, payload_offset); + } else { + payload_offset = offsetof(struct v9_gfx_meta_data, ce_payload); + ce_payload_cpu_addr = adev->virt.csa_cpu_addr + payload_offset; + } + + if (offset + (payload_size >> 2) <= ring->buf_mask + 1) { + memcpy((void *)&ring->ring[offset], ce_payload_cpu_addr, payload_size); + } else { + memcpy((void *)&ring->ring[offset], ce_payload_cpu_addr, + (ring->buf_mask + 1 - offset) << 2); + payload_size -= (ring->buf_mask + 1 - offset) << 2; + memcpy((void *)&ring->ring[0], + ce_payload_cpu_addr + ((ring->buf_mask + 1 - offset) << 2), + payload_size); + } +} + +static void gfx_v9_0_ring_patch_de_meta(struct amdgpu_ring *ring, + unsigned offset) +{ + struct amdgpu_device *adev = ring->adev; + void *de_payload_cpu_addr; + uint64_t payload_offset, payload_size; + + payload_size = sizeof(struct v9_de_ib_state); + + if (ring->is_mes_queue) { + payload_offset = offsetof(struct amdgpu_mes_ctx_meta_data, + gfx[0].gfx_meta_data) + + offsetof(struct v9_gfx_meta_data, de_payload); + de_payload_cpu_addr = + amdgpu_mes_ctx_get_offs_cpu_addr(ring, payload_offset); + } else { + payload_offset = offsetof(struct v9_gfx_meta_data, de_payload); + de_payload_cpu_addr = adev->virt.csa_cpu_addr + payload_offset; + } + + if (offset + (payload_size >> 2) <= ring->buf_mask + 1) { + memcpy((void *)&ring->ring[offset], de_payload_cpu_addr, payload_size); + } else { + memcpy((void *)&ring->ring[offset], de_payload_cpu_addr, + (ring->buf_mask + 1 - offset) << 2); + payload_size -= (ring->buf_mask + 1 - offset) << 2; + memcpy((void *)&ring->ring[0], + de_payload_cpu_addr + ((ring->buf_mask + 1 - offset) << 2), + payload_size); + } +} + static void gfx_v9_0_ring_emit_ib_compute(struct amdgpu_ring *ring, struct amdgpu_job *job, struct amdgpu_ib *ib, @@ -5374,6 +5437,8 @@ static void gfx_v9_0_ring_emit_ce_meta(struct amdgpu_ring *ring, bool resume) amdgpu_ring_write(ring, lower_32_bits(ce_payload_gpu_addr)); amdgpu_ring_write(ring, upper_32_bits(ce_payload_gpu_addr)); + amdgpu_ring_ib_on_emit_ce(ring); + if (resume) amdgpu_ring_write_multiple(ring, ce_payload_cpu_addr, sizeof(ce_payload) >> 2); @@ -5386,7 +5451,7 @@ static int gfx_v9_0_ring_preempt_ib(struct amdgpu_ring *ring) { int i, r = 0; struct amdgpu_device *adev = ring->adev; - struct amdgpu_kiq *kiq = &adev->gfx.kiq; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; struct amdgpu_ring *kiq_ring = &kiq->ring; unsigned long flags; @@ -5407,10 +5472,6 @@ static int gfx_v9_0_ring_preempt_ib(struct amdgpu_ring *ring) amdgpu_ring_alloc(ring, 13); gfx_v9_0_ring_emit_fence(ring, ring->trail_fence_gpu_addr, ring->trail_seq, AMDGPU_FENCE_FLAG_EXEC | AMDGPU_FENCE_FLAG_INT); - /*reset the CP_VMID_PREEMPT after trailing fence*/ - amdgpu_ring_emit_wreg(ring, - SOC15_REG_OFFSET(GC, 0, mmCP_VMID_PREEMPT), - 0x0); /* assert IB preemption, emit the trailing fence */ kiq->pmf->kiq_unmap_queues(kiq_ring, ring, PREEMPT_QUEUES_NO_UNMAP, @@ -5433,6 +5494,10 @@ static int gfx_v9_0_ring_preempt_ib(struct amdgpu_ring *ring) DRM_WARN("ring %d timeout to preempt ib\n", ring->idx); } + /*reset the CP_VMID_PREEMPT after trailing fence*/ + amdgpu_ring_emit_wreg(ring, + SOC15_REG_OFFSET(GC, 0, mmCP_VMID_PREEMPT), + 0x0); amdgpu_ring_commit(ring); /* deassert preemption condition */ @@ -5440,7 +5505,7 @@ static int gfx_v9_0_ring_preempt_ib(struct amdgpu_ring *ring) return r; } -static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume) +static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume, bool usegds) { struct amdgpu_device *adev = ring->adev; struct v9_de_ib_state de_payload = {0}; @@ -5471,8 +5536,10 @@ static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume) PAGE_SIZE); } - de_payload.gds_backup_addrlo = lower_32_bits(gds_addr); - de_payload.gds_backup_addrhi = upper_32_bits(gds_addr); + if (usegds) { + de_payload.gds_backup_addrlo = lower_32_bits(gds_addr); + de_payload.gds_backup_addrhi = upper_32_bits(gds_addr); + } cnt = (sizeof(de_payload) >> 2) + 4 - 2; amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, cnt)); @@ -5483,6 +5550,7 @@ static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring, bool resume) amdgpu_ring_write(ring, lower_32_bits(de_payload_gpu_addr)); amdgpu_ring_write(ring, upper_32_bits(de_payload_gpu_addr)); + amdgpu_ring_ib_on_emit_de(ring); if (resume) amdgpu_ring_write_multiple(ring, de_payload_cpu_addr, sizeof(de_payload) >> 2); @@ -6342,7 +6410,7 @@ static const struct soc15_ras_field_entry gfx_v9_0_ras_fields[] = { }; static int gfx_v9_0_ras_error_inject(struct amdgpu_device *adev, - void *inject_if) + void *inject_if, uint32_t instance_mask) { struct ras_inject_if *info = (struct ras_inject_if *)inject_if; int ret; @@ -6381,7 +6449,7 @@ static int gfx_v9_0_ras_error_inject(struct amdgpu_device *adev, block_info.value = info->value; mutex_lock(&adev->grbm_idx_mutex); - ret = psp_ras_trigger_error(&adev->psp, &block_info); + ret = psp_ras_trigger_error(&adev->psp, &block_info, instance_mask); mutex_unlock(&adev->grbm_idx_mutex); return ret; @@ -6609,7 +6677,7 @@ static void gfx_v9_0_reset_ras_error_count(struct amdgpu_device *adev) for (i = 0; i < ARRAY_SIZE(gfx_v9_0_edc_counter_regs); i++) { for (j = 0; j < gfx_v9_0_edc_counter_regs[i].se_num; j++) { for (k = 0; k < gfx_v9_0_edc_counter_regs[i].instance; k++) { - amdgpu_gfx_select_se_sh(adev, j, 0x0, k); + amdgpu_gfx_select_se_sh(adev, j, 0x0, k, 0); RREG32(SOC15_REG_ENTRY_OFFSET(gfx_v9_0_edc_counter_regs[i])); } } @@ -6671,7 +6739,7 @@ static void gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev, for (i = 0; i < ARRAY_SIZE(gfx_v9_0_edc_counter_regs); i++) { for (j = 0; j < gfx_v9_0_edc_counter_regs[i].se_num; j++) { for (k = 0; k < gfx_v9_0_edc_counter_regs[i].instance; k++) { - amdgpu_gfx_select_se_sh(adev, j, 0, k); + amdgpu_gfx_select_se_sh(adev, j, 0, k, 0); reg_value = RREG32(SOC15_REG_ENTRY_OFFSET(gfx_v9_0_edc_counter_regs[i])); if (reg_value) @@ -6686,7 +6754,7 @@ static void gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev, err_data->ce_count += sec_count; err_data->ue_count += ded_count; - amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); gfx_v9_0_query_utc_edc_status(adev, err_data); @@ -6893,6 +6961,9 @@ static const struct amdgpu_ring_funcs gfx_v9_0_sw_ring_funcs_gfx = { .emit_reg_write_reg_wait = gfx_v9_0_ring_emit_reg_write_reg_wait, .soft_recovery = gfx_v9_0_ring_soft_recovery, .emit_mem_sync = gfx_v9_0_emit_mem_sync, + .patch_cntl = gfx_v9_0_ring_patch_cntl, + .patch_de = gfx_v9_0_ring_patch_de_meta, + .patch_ce = gfx_v9_0_ring_patch_ce_meta, }; static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_compute = { @@ -6965,7 +7036,7 @@ static void gfx_v9_0_set_ring_funcs(struct amdgpu_device *adev) { int i; - adev->gfx.kiq.ring.funcs = &gfx_v9_0_ring_funcs_kiq; + adev->gfx.kiq[0].ring.funcs = &gfx_v9_0_ring_funcs_kiq; for (i = 0; i < adev->gfx.num_gfx_rings; i++) adev->gfx.gfx_ring[i].funcs = &gfx_v9_0_ring_funcs_gfx; @@ -7146,7 +7217,7 @@ static int gfx_v9_0_get_cu_info(struct amdgpu_device *adev, mask = 1; ao_bitmap = 0; counter = 0; - amdgpu_gfx_select_se_sh(adev, i, j, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, i, j, 0xffffffff, 0); gfx_v9_0_set_user_cu_inactive_bitmap( adev, disable_masks[i * adev->gfx.config.max_sh_per_se + j]); bitmap = gfx_v9_0_get_cu_active_bitmap(adev); @@ -7179,7 +7250,7 @@ static int gfx_v9_0_get_cu_info(struct amdgpu_device *adev, cu_info->ao_cu_bitmap[i % 4][j + i / 4] = ao_bitmap; } } - amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); cu_info->number = active_cu_number; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.h b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.h index dfe8d4841f582cb552bb3f09eb11c3585984ff2a..f9f6edc5e5589cfb798f724ec8531f3e46ee02bc 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.h +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.h @@ -27,6 +27,6 @@ extern const struct amdgpu_ip_block_version gfx_v9_0_ip_block; void gfx_v9_0_select_se_sh(struct amdgpu_device *adev, u32 se_num, u32 sh_num, - u32 instance); + u32 instance, int xcc_id); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c index c67e387a97f5c1c2c4949bca08affe6a3002bae9..bc8416afb62c5d5530ebea5b828b06f2974ed47c 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c @@ -970,29 +970,6 @@ static void gfx_v9_4_reset_ras_error_count(struct amdgpu_device *adev) WREG32_SOC15(GC, 0, mmATC_L2_CACHE_4K_DSM_INDEX, 255); } -static int gfx_v9_4_ras_error_inject(struct amdgpu_device *adev, - void *inject_if) -{ - struct ras_inject_if *info = (struct ras_inject_if *)inject_if; - int ret; - struct ta_ras_trigger_error_input block_info = { 0 }; - - if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX)) - return -EINVAL; - - block_info.block_id = amdgpu_ras_block_to_ta(info->head.block); - block_info.sub_block_index = info->head.sub_block_index; - block_info.inject_error_type = amdgpu_ras_error_to_ta(info->head.type); - block_info.address = info->address; - block_info.value = info->value; - - mutex_lock(&adev->grbm_idx_mutex); - ret = psp_ras_trigger_error(&adev->psp, &block_info); - mutex_unlock(&adev->grbm_idx_mutex); - - return ret; -} - static const struct soc15_reg_entry gfx_v9_4_ea_err_status_regs = { SOC15_REG_ENTRY(GC, 0, mmGCEA_ERR_STATUS), 0, 1, 32 }; @@ -1030,7 +1007,6 @@ static void gfx_v9_4_query_ras_error_status(struct amdgpu_device *adev) const struct amdgpu_ras_block_hw_ops gfx_v9_4_ras_ops = { - .ras_error_inject = &gfx_v9_4_ras_error_inject, .query_ras_error_count = &gfx_v9_4_query_ras_error_count, .reset_ras_error_count = &gfx_v9_4_reset_ras_error_count, .query_ras_error_status = &gfx_v9_4_query_ras_error_status, diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c index 3a797424579c5955ffe0f701cbec2344d7f2ef23..63f6843a069e2fbbb54efa3a1c53acf5a6f86d86 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c @@ -761,7 +761,7 @@ void gfx_v9_4_2_debug_trap_config_init(struct amdgpu_device *adev, for (i = first_vmid; i < last_vmid; i++) { data = 0; - soc15_grbm_select(adev, 0, 0, 0, i); + soc15_grbm_select(adev, 0, 0, 0, i, 0); data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1); data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, 0); data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, @@ -769,15 +769,18 @@ void gfx_v9_4_2_debug_trap_config_init(struct amdgpu_device *adev, WREG32(SOC15_REG_OFFSET(GC, 0, regSPI_GDBG_PER_VMID_CNTL), data); } - soc15_grbm_select(adev, 0, 0, 0, 0); + soc15_grbm_select(adev, 0, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); + + WREG32(SOC15_REG_OFFSET(GC, 0, regSPI_GDBG_TRAP_DATA0), 0); + WREG32(SOC15_REG_OFFSET(GC, 0, regSPI_GDBG_TRAP_DATA1), 0); } void gfx_v9_4_2_set_power_brake_sequence(struct amdgpu_device *adev) { u32 tmp; - gfx_v9_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + gfx_v9_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); tmp = 0; tmp = REG_SET_FIELD(tmp, GC_THROTTLE_CTRL, PATTERN_MODE, 1); @@ -1699,28 +1702,6 @@ static void gfx_v9_4_2_reset_ras_error_count(struct amdgpu_device *adev) gfx_v9_4_2_query_utc_edc_count(adev, NULL, NULL); } -static int gfx_v9_4_2_ras_error_inject(struct amdgpu_device *adev, void *inject_if) -{ - struct ras_inject_if *info = (struct ras_inject_if *)inject_if; - int ret; - struct ta_ras_trigger_error_input block_info = { 0 }; - - if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX)) - return -EINVAL; - - block_info.block_id = amdgpu_ras_block_to_ta(info->head.block); - block_info.sub_block_index = info->head.sub_block_index; - block_info.inject_error_type = amdgpu_ras_error_to_ta(info->head.type); - block_info.address = info->address; - block_info.value = info->value; - - mutex_lock(&adev->grbm_idx_mutex); - ret = psp_ras_trigger_error(&adev->psp, &block_info); - mutex_unlock(&adev->grbm_idx_mutex); - - return ret; -} - static void gfx_v9_4_2_query_ea_err_status(struct amdgpu_device *adev) { uint32_t i, j; @@ -1935,7 +1916,7 @@ static bool gfx_v9_4_2_query_uctl2_poison_status(struct amdgpu_device *adev) u32 status = 0; struct amdgpu_vmhub *hub; - hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; status = RREG32(hub->vm_l2_pro_fault_status); /* reset page fault status */ WREG32_P(hub->vm_l2_pro_fault_cntl, 1, ~1); @@ -1944,7 +1925,6 @@ static bool gfx_v9_4_2_query_uctl2_poison_status(struct amdgpu_device *adev) } struct amdgpu_ras_block_hw_ops gfx_v9_4_2_ras_ops = { - .ras_error_inject = &gfx_v9_4_2_ras_error_inject, .query_ras_error_count = &gfx_v9_4_2_query_ras_error_count, .reset_ras_error_count = &gfx_v9_4_2_reset_ras_error_count, .query_ras_error_status = &gfx_v9_4_2_query_ras_error_status, diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c index 5f8500577c021b1261ae2bff37b14aa8261c6936..c1ee54d4c3d3dc198b4983dbc6c9f690cf289cf8 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c @@ -25,35 +25,504 @@ #include "amdgpu.h" #include "amdgpu_gfx.h" #include "soc15.h" +#include "soc15d.h" #include "soc15_common.h" #include "vega10_enum.h" +#include "v9_structs.h" + +#include "ivsrcid/gfx/irqsrcs_gfx_9_0.h" + #include "gc/gc_9_4_3_offset.h" #include "gc/gc_9_4_3_sh_mask.h" #include "gfx_v9_4_3.h" +#include "amdgpu_xcp.h" +MODULE_FIRMWARE("amdgpu/gc_9_4_3_mec.bin"); +MODULE_FIRMWARE("amdgpu/gc_9_4_3_rlc.bin"); + +#define GFX9_MEC_HPD_SIZE 4096 #define RLCG_UCODE_LOADING_START_ADDRESS 0x00002000L +#define GOLDEN_GB_ADDR_CONFIG 0x2a114042 + +struct amdgpu_gfx_ras gfx_v9_4_3_ras; + +static void gfx_v9_4_3_set_ring_funcs(struct amdgpu_device *adev); +static void gfx_v9_4_3_set_irq_funcs(struct amdgpu_device *adev); +static void gfx_v9_4_3_set_gds_init(struct amdgpu_device *adev); +static void gfx_v9_4_3_set_rlc_funcs(struct amdgpu_device *adev); +static int gfx_v9_4_3_get_cu_info(struct amdgpu_device *adev, + struct amdgpu_cu_info *cu_info); + +static void gfx_v9_4_3_kiq_set_resources(struct amdgpu_ring *kiq_ring, + uint64_t queue_mask) +{ + amdgpu_ring_write(kiq_ring, PACKET3(PACKET3_SET_RESOURCES, 6)); + amdgpu_ring_write(kiq_ring, + PACKET3_SET_RESOURCES_VMID_MASK(0) | + /* vmid_mask:0* queue_type:0 (KIQ) */ + PACKET3_SET_RESOURCES_QUEUE_TYPE(0)); + amdgpu_ring_write(kiq_ring, + lower_32_bits(queue_mask)); /* queue mask lo */ + amdgpu_ring_write(kiq_ring, + upper_32_bits(queue_mask)); /* queue mask hi */ + amdgpu_ring_write(kiq_ring, 0); /* gws mask lo */ + amdgpu_ring_write(kiq_ring, 0); /* gws mask hi */ + amdgpu_ring_write(kiq_ring, 0); /* oac mask */ + amdgpu_ring_write(kiq_ring, 0); /* gds heap base:0, gds heap size:0 */ +} + +static void gfx_v9_4_3_kiq_map_queues(struct amdgpu_ring *kiq_ring, + struct amdgpu_ring *ring) +{ + struct amdgpu_device *adev = kiq_ring->adev; + uint64_t mqd_addr = amdgpu_bo_gpu_offset(ring->mqd_obj); + uint64_t wptr_addr = adev->wb.gpu_addr + (ring->wptr_offs * 4); + uint32_t eng_sel = ring->funcs->type == AMDGPU_RING_TYPE_GFX ? 4 : 0; + + amdgpu_ring_write(kiq_ring, PACKET3(PACKET3_MAP_QUEUES, 5)); + /* Q_sel:0, vmid:0, vidmem: 1, engine:0, num_Q:1*/ + amdgpu_ring_write(kiq_ring, /* Q_sel: 0, vmid: 0, engine: 0, num_Q: 1 */ + PACKET3_MAP_QUEUES_QUEUE_SEL(0) | /* Queue_Sel */ + PACKET3_MAP_QUEUES_VMID(0) | /* VMID */ + PACKET3_MAP_QUEUES_QUEUE(ring->queue) | + PACKET3_MAP_QUEUES_PIPE(ring->pipe) | + PACKET3_MAP_QUEUES_ME((ring->me == 1 ? 0 : 1)) | + /*queue_type: normal compute queue */ + PACKET3_MAP_QUEUES_QUEUE_TYPE(0) | + /* alloc format: all_on_one_pipe */ + PACKET3_MAP_QUEUES_ALLOC_FORMAT(0) | + PACKET3_MAP_QUEUES_ENGINE_SEL(eng_sel) | + /* num_queues: must be 1 */ + PACKET3_MAP_QUEUES_NUM_QUEUES(1)); + amdgpu_ring_write(kiq_ring, + PACKET3_MAP_QUEUES_DOORBELL_OFFSET(ring->doorbell_index)); + amdgpu_ring_write(kiq_ring, lower_32_bits(mqd_addr)); + amdgpu_ring_write(kiq_ring, upper_32_bits(mqd_addr)); + amdgpu_ring_write(kiq_ring, lower_32_bits(wptr_addr)); + amdgpu_ring_write(kiq_ring, upper_32_bits(wptr_addr)); +} + +static void gfx_v9_4_3_kiq_unmap_queues(struct amdgpu_ring *kiq_ring, + struct amdgpu_ring *ring, + enum amdgpu_unmap_queues_action action, + u64 gpu_addr, u64 seq) +{ + uint32_t eng_sel = ring->funcs->type == AMDGPU_RING_TYPE_GFX ? 4 : 0; + + amdgpu_ring_write(kiq_ring, PACKET3(PACKET3_UNMAP_QUEUES, 4)); + amdgpu_ring_write(kiq_ring, /* Q_sel: 0, vmid: 0, engine: 0, num_Q: 1 */ + PACKET3_UNMAP_QUEUES_ACTION(action) | + PACKET3_UNMAP_QUEUES_QUEUE_SEL(0) | + PACKET3_UNMAP_QUEUES_ENGINE_SEL(eng_sel) | + PACKET3_UNMAP_QUEUES_NUM_QUEUES(1)); + amdgpu_ring_write(kiq_ring, + PACKET3_UNMAP_QUEUES_DOORBELL_OFFSET0(ring->doorbell_index)); + + if (action == PREEMPT_QUEUES_NO_UNMAP) { + amdgpu_ring_write(kiq_ring, lower_32_bits(gpu_addr)); + amdgpu_ring_write(kiq_ring, upper_32_bits(gpu_addr)); + amdgpu_ring_write(kiq_ring, seq); + } else { + amdgpu_ring_write(kiq_ring, 0); + amdgpu_ring_write(kiq_ring, 0); + amdgpu_ring_write(kiq_ring, 0); + } +} + +static void gfx_v9_4_3_kiq_query_status(struct amdgpu_ring *kiq_ring, + struct amdgpu_ring *ring, + u64 addr, + u64 seq) +{ + uint32_t eng_sel = ring->funcs->type == AMDGPU_RING_TYPE_GFX ? 4 : 0; + + amdgpu_ring_write(kiq_ring, PACKET3(PACKET3_QUERY_STATUS, 5)); + amdgpu_ring_write(kiq_ring, + PACKET3_QUERY_STATUS_CONTEXT_ID(0) | + PACKET3_QUERY_STATUS_INTERRUPT_SEL(0) | + PACKET3_QUERY_STATUS_COMMAND(2)); + /* Q_sel: 0, vmid: 0, engine: 0, num_Q: 1 */ + amdgpu_ring_write(kiq_ring, + PACKET3_QUERY_STATUS_DOORBELL_OFFSET(ring->doorbell_index) | + PACKET3_QUERY_STATUS_ENG_SEL(eng_sel)); + amdgpu_ring_write(kiq_ring, lower_32_bits(addr)); + amdgpu_ring_write(kiq_ring, upper_32_bits(addr)); + amdgpu_ring_write(kiq_ring, lower_32_bits(seq)); + amdgpu_ring_write(kiq_ring, upper_32_bits(seq)); +} + +static void gfx_v9_4_3_kiq_invalidate_tlbs(struct amdgpu_ring *kiq_ring, + uint16_t pasid, uint32_t flush_type, + bool all_hub) +{ + amdgpu_ring_write(kiq_ring, PACKET3(PACKET3_INVALIDATE_TLBS, 0)); + amdgpu_ring_write(kiq_ring, + PACKET3_INVALIDATE_TLBS_DST_SEL(1) | + PACKET3_INVALIDATE_TLBS_ALL_HUB(all_hub) | + PACKET3_INVALIDATE_TLBS_PASID(pasid) | + PACKET3_INVALIDATE_TLBS_FLUSH_TYPE(flush_type)); +} + +static const struct kiq_pm4_funcs gfx_v9_4_3_kiq_pm4_funcs = { + .kiq_set_resources = gfx_v9_4_3_kiq_set_resources, + .kiq_map_queues = gfx_v9_4_3_kiq_map_queues, + .kiq_unmap_queues = gfx_v9_4_3_kiq_unmap_queues, + .kiq_query_status = gfx_v9_4_3_kiq_query_status, + .kiq_invalidate_tlbs = gfx_v9_4_3_kiq_invalidate_tlbs, + .set_resources_size = 8, + .map_queues_size = 7, + .unmap_queues_size = 6, + .query_status_size = 7, + .invalidate_tlbs_size = 2, +}; + +static void gfx_v9_4_3_set_kiq_pm4_funcs(struct amdgpu_device *adev) +{ + int i, num_xcc; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + for (i = 0; i < num_xcc; i++) + adev->gfx.kiq[i].pmf = &gfx_v9_4_3_kiq_pm4_funcs; +} + +static void gfx_v9_4_3_init_golden_registers(struct amdgpu_device *adev) +{ + int i, num_xcc, dev_inst; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + for (i = 0; i < num_xcc; i++) { + dev_inst = GET_INST(GC, i); + if (dev_inst >= 2) + WREG32_SOC15(GC, dev_inst, regGRBM_MCM_ADDR, 0x4); + + /* Golden settings applied by driver for ASIC with rev_id 0 */ + if (adev->rev_id == 0) { + WREG32_SOC15(GC, dev_inst, regGB_ADDR_CONFIG, + GOLDEN_GB_ADDR_CONFIG); + + WREG32_FIELD15_PREREG(GC, dev_inst, TCP_UTCL1_CNTL1, + REDUCE_FIFO_DEPTH_BY_2, 2); + } + } +} + +static void gfx_v9_4_3_write_data_to_reg(struct amdgpu_ring *ring, int eng_sel, + bool wc, uint32_t reg, uint32_t val) +{ + amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); + amdgpu_ring_write(ring, WRITE_DATA_ENGINE_SEL(eng_sel) | + WRITE_DATA_DST_SEL(0) | + (wc ? WR_CONFIRM : 0)); + amdgpu_ring_write(ring, reg); + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, val); +} + +static void gfx_v9_4_3_wait_reg_mem(struct amdgpu_ring *ring, int eng_sel, + int mem_space, int opt, uint32_t addr0, + uint32_t addr1, uint32_t ref, uint32_t mask, + uint32_t inv) +{ + amdgpu_ring_write(ring, PACKET3(PACKET3_WAIT_REG_MEM, 5)); + amdgpu_ring_write(ring, + /* memory (1) or register (0) */ + (WAIT_REG_MEM_MEM_SPACE(mem_space) | + WAIT_REG_MEM_OPERATION(opt) | /* wait */ + WAIT_REG_MEM_FUNCTION(3) | /* equal */ + WAIT_REG_MEM_ENGINE(eng_sel))); + + if (mem_space) + BUG_ON(addr0 & 0x3); /* Dword align */ + amdgpu_ring_write(ring, addr0); + amdgpu_ring_write(ring, addr1); + amdgpu_ring_write(ring, ref); + amdgpu_ring_write(ring, mask); + amdgpu_ring_write(ring, inv); /* poll interval */ +} + +static int gfx_v9_4_3_ring_test_ring(struct amdgpu_ring *ring) +{ + uint32_t scratch_reg0_offset, xcc_offset; + struct amdgpu_device *adev = ring->adev; + uint32_t tmp = 0; + unsigned i; + int r; + + /* Use register offset which is local to XCC in the packet */ + xcc_offset = SOC15_REG_OFFSET(GC, 0, regSCRATCH_REG0); + scratch_reg0_offset = SOC15_REG_OFFSET(GC, GET_INST(GC, ring->xcc_id), regSCRATCH_REG0); + WREG32(scratch_reg0_offset, 0xCAFEDEAD); + + r = amdgpu_ring_alloc(ring, 3); + if (r) + return r; + + amdgpu_ring_write(ring, PACKET3(PACKET3_SET_UCONFIG_REG, 1)); + amdgpu_ring_write(ring, xcc_offset - PACKET3_SET_UCONFIG_REG_START); + amdgpu_ring_write(ring, 0xDEADBEEF); + amdgpu_ring_commit(ring); + + for (i = 0; i < adev->usec_timeout; i++) { + tmp = RREG32(scratch_reg0_offset); + if (tmp == 0xDEADBEEF) + break; + udelay(1); + } + + if (i >= adev->usec_timeout) + r = -ETIMEDOUT; + return r; +} + +static int gfx_v9_4_3_ring_test_ib(struct amdgpu_ring *ring, long timeout) +{ + struct amdgpu_device *adev = ring->adev; + struct amdgpu_ib ib; + struct dma_fence *f = NULL; + + unsigned index; + uint64_t gpu_addr; + uint32_t tmp; + long r; + + r = amdgpu_device_wb_get(adev, &index); + if (r) + return r; + + gpu_addr = adev->wb.gpu_addr + (index * 4); + adev->wb.wb[index] = cpu_to_le32(0xCAFEDEAD); + memset(&ib, 0, sizeof(ib)); + r = amdgpu_ib_get(adev, NULL, 16, + AMDGPU_IB_POOL_DIRECT, &ib); + if (r) + goto err1; + + ib.ptr[0] = PACKET3(PACKET3_WRITE_DATA, 3); + ib.ptr[1] = WRITE_DATA_DST_SEL(5) | WR_CONFIRM; + ib.ptr[2] = lower_32_bits(gpu_addr); + ib.ptr[3] = upper_32_bits(gpu_addr); + ib.ptr[4] = 0xDEADBEEF; + ib.length_dw = 5; + + r = amdgpu_ib_schedule(ring, 1, &ib, NULL, &f); + if (r) + goto err2; + + r = dma_fence_wait_timeout(f, false, timeout); + if (r == 0) { + r = -ETIMEDOUT; + goto err2; + } else if (r < 0) { + goto err2; + } + + tmp = adev->wb.wb[index]; + if (tmp == 0xDEADBEEF) + r = 0; + else + r = -EINVAL; + +err2: + amdgpu_ib_free(adev, &ib, NULL); + dma_fence_put(f); +err1: + amdgpu_device_wb_free(adev, index); + return r; +} + + +/* This value might differs per partition */ static uint64_t gfx_v9_4_3_get_gpu_clock_counter(struct amdgpu_device *adev) { uint64_t clock; amdgpu_gfx_off_ctrl(adev, false); mutex_lock(&adev->gfx.gpu_clock_mutex); - WREG32_SOC15(GC, 0, regRLC_CAPTURE_GPU_CLOCK_COUNT, 1); - clock = (uint64_t)RREG32_SOC15(GC, 0, regRLC_GPU_CLOCK_COUNT_LSB) | - ((uint64_t)RREG32_SOC15(GC, 0, regRLC_GPU_CLOCK_COUNT_MSB) << 32ULL); + WREG32_SOC15(GC, GET_INST(GC, 0), regRLC_CAPTURE_GPU_CLOCK_COUNT, 1); + clock = (uint64_t)RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_GPU_CLOCK_COUNT_LSB) | + ((uint64_t)RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_GPU_CLOCK_COUNT_MSB) << 32ULL); mutex_unlock(&adev->gfx.gpu_clock_mutex); amdgpu_gfx_off_ctrl(adev, true); return clock; } -static void gfx_v9_4_3_select_se_sh(struct amdgpu_device *adev, - u32 se_num, - u32 sh_num, - u32 instance) +static void gfx_v9_4_3_free_microcode(struct amdgpu_device *adev) +{ + amdgpu_ucode_release(&adev->gfx.pfp_fw); + amdgpu_ucode_release(&adev->gfx.me_fw); + amdgpu_ucode_release(&adev->gfx.ce_fw); + amdgpu_ucode_release(&adev->gfx.rlc_fw); + amdgpu_ucode_release(&adev->gfx.mec_fw); + amdgpu_ucode_release(&adev->gfx.mec2_fw); + + kfree(adev->gfx.rlc.register_list_format); +} + +static int gfx_v9_4_3_init_rlc_microcode(struct amdgpu_device *adev, + const char *chip_name) +{ + char fw_name[30]; + int err; + const struct rlc_firmware_header_v2_0 *rlc_hdr; + uint16_t version_major; + uint16_t version_minor; + + snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_rlc.bin", chip_name); + + err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, fw_name); + if (err) + goto out; + rlc_hdr = (const struct rlc_firmware_header_v2_0 *)adev->gfx.rlc_fw->data; + + version_major = le16_to_cpu(rlc_hdr->header.header_version_major); + version_minor = le16_to_cpu(rlc_hdr->header.header_version_minor); + err = amdgpu_gfx_rlc_init_microcode(adev, version_major, version_minor); +out: + if (err) + amdgpu_ucode_release(&adev->gfx.rlc_fw); + + return err; +} + +static bool gfx_v9_4_3_should_disable_gfxoff(struct pci_dev *pdev) +{ + return true; +} + +static void gfx_v9_4_3_check_if_need_gfxoff(struct amdgpu_device *adev) +{ + if (gfx_v9_4_3_should_disable_gfxoff(adev->pdev)) + adev->pm.pp_feature &= ~PP_GFXOFF_MASK; +} + +static int gfx_v9_4_3_init_cp_compute_microcode(struct amdgpu_device *adev, + const char *chip_name) +{ + char fw_name[30]; + int err; + + snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec.bin", chip_name); + + err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw, fw_name); + if (err) + goto out; + amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_MEC1); + amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_MEC1_JT); + + adev->gfx.mec2_fw_version = adev->gfx.mec_fw_version; + adev->gfx.mec2_feature_version = adev->gfx.mec_feature_version; + + gfx_v9_4_3_check_if_need_gfxoff(adev); + +out: + if (err) + amdgpu_ucode_release(&adev->gfx.mec_fw); + return err; +} + +static int gfx_v9_4_3_init_microcode(struct amdgpu_device *adev) +{ + const char *chip_name; + int r; + + chip_name = "gc_9_4_3"; + + r = gfx_v9_4_3_init_rlc_microcode(adev, chip_name); + if (r) + return r; + + r = gfx_v9_4_3_init_cp_compute_microcode(adev, chip_name); + if (r) + return r; + + return r; +} + +static void gfx_v9_4_3_mec_fini(struct amdgpu_device *adev) +{ + amdgpu_bo_free_kernel(&adev->gfx.mec.hpd_eop_obj, NULL, NULL); + amdgpu_bo_free_kernel(&adev->gfx.mec.mec_fw_obj, NULL, NULL); +} + +static int gfx_v9_4_3_mec_init(struct amdgpu_device *adev) +{ + int r, i, num_xcc; + u32 *hpd; + const __le32 *fw_data; + unsigned fw_size; + u32 *fw; + size_t mec_hpd_size; + + const struct gfx_firmware_header_v1_0 *mec_hdr; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + for (i = 0; i < num_xcc; i++) + bitmap_zero(adev->gfx.mec_bitmap[i].queue_bitmap, + AMDGPU_MAX_COMPUTE_QUEUES); + + /* take ownership of the relevant compute queues */ + amdgpu_gfx_compute_queue_acquire(adev); + mec_hpd_size = + adev->gfx.num_compute_rings * num_xcc * GFX9_MEC_HPD_SIZE; + if (mec_hpd_size) { + r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE, + AMDGPU_GEM_DOMAIN_VRAM | + AMDGPU_GEM_DOMAIN_GTT, + &adev->gfx.mec.hpd_eop_obj, + &adev->gfx.mec.hpd_eop_gpu_addr, + (void **)&hpd); + if (r) { + dev_warn(adev->dev, "(%d) create HDP EOP bo failed\n", r); + gfx_v9_4_3_mec_fini(adev); + return r; + } + + if (amdgpu_emu_mode == 1) { + for (i = 0; i < mec_hpd_size / 4; i++) { + memset((void *)(hpd + i), 0, 4); + if (i % 50 == 0) + msleep(1); + } + } else { + memset(hpd, 0, mec_hpd_size); + } + + amdgpu_bo_kunmap(adev->gfx.mec.hpd_eop_obj); + amdgpu_bo_unreserve(adev->gfx.mec.hpd_eop_obj); + } + + mec_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec_fw->data; + + fw_data = (const __le32 *) + (adev->gfx.mec_fw->data + + le32_to_cpu(mec_hdr->header.ucode_array_offset_bytes)); + fw_size = le32_to_cpu(mec_hdr->header.ucode_size_bytes); + + r = amdgpu_bo_create_reserved(adev, mec_hdr->header.ucode_size_bytes, + PAGE_SIZE, AMDGPU_GEM_DOMAIN_GTT, + &adev->gfx.mec.mec_fw_obj, + &adev->gfx.mec.mec_fw_gpu_addr, + (void **)&fw); + if (r) { + dev_warn(adev->dev, "(%d) create mec firmware bo failed\n", r); + gfx_v9_4_3_mec_fini(adev); + return r; + } + + memcpy(fw, fw_data, fw_size); + + amdgpu_bo_kunmap(adev->gfx.mec.mec_fw_obj); + amdgpu_bo_unreserve(adev->gfx.mec.mec_fw_obj); + + return 0; +} + +static void gfx_v9_4_3_xcc_select_se_sh(struct amdgpu_device *adev, u32 se_num, + u32 sh_num, u32 instance, int xcc_id) { u32 data; @@ -76,24 +545,24 @@ static void gfx_v9_4_3_select_se_sh(struct amdgpu_device *adev, else data = REG_SET_FIELD(data, GRBM_GFX_INDEX, SH_INDEX, sh_num); - WREG32_SOC15_RLC_SHADOW_EX(reg, GC, 0, regGRBM_GFX_INDEX, data); + WREG32_SOC15_RLC_SHADOW_EX(reg, GC, GET_INST(GC, xcc_id), regGRBM_GFX_INDEX, data); } -static uint32_t wave_read_ind(struct amdgpu_device *adev, uint32_t simd, uint32_t wave, uint32_t address) +static uint32_t wave_read_ind(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t address) { - WREG32_SOC15_RLC(GC, 0, regSQ_IND_INDEX, + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regSQ_IND_INDEX, (wave << SQ_IND_INDEX__WAVE_ID__SHIFT) | (simd << SQ_IND_INDEX__SIMD_ID__SHIFT) | (address << SQ_IND_INDEX__INDEX__SHIFT) | (SQ_IND_INDEX__FORCE_READ_MASK)); - return RREG32_SOC15(GC, 0, regSQ_IND_DATA); + return RREG32_SOC15(GC, GET_INST(GC, xcc_id), regSQ_IND_DATA); } -static void wave_read_regs(struct amdgpu_device *adev, uint32_t simd, +static void wave_read_regs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t thread, uint32_t regno, uint32_t num, uint32_t *out) { - WREG32_SOC15_RLC(GC, 0, regSQ_IND_INDEX, + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regSQ_IND_INDEX, (wave << SQ_IND_INDEX__WAVE_ID__SHIFT) | (simd << SQ_IND_INDEX__SIMD_ID__SHIFT) | (regno << SQ_IND_INDEX__INDEX__SHIFT) | @@ -101,210 +570,672 @@ static void wave_read_regs(struct amdgpu_device *adev, uint32_t simd, (SQ_IND_INDEX__FORCE_READ_MASK) | (SQ_IND_INDEX__AUTO_INCR_MASK)); while (num--) - *(out++) = RREG32_SOC15(GC, 0, regSQ_IND_DATA); + *(out++) = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regSQ_IND_DATA); } static void gfx_v9_4_3_read_wave_data(struct amdgpu_device *adev, - uint32_t simd, uint32_t wave, + uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t *dst, int *no_fields) { /* type 1 wave data */ dst[(*no_fields)++] = 1; - dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_STATUS); - dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_PC_LO); - dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_PC_HI); - dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_EXEC_LO); - dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_EXEC_HI); - dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_HW_ID); - dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_INST_DW0); - dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_INST_DW1); - dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_GPR_ALLOC); - dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_LDS_ALLOC); - dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_TRAPSTS); - dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_IB_STS); - dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_IB_DBG0); - dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_M0); - dst[(*no_fields)++] = wave_read_ind(adev, simd, wave, ixSQ_WAVE_MODE); -} - -static void gfx_v9_4_3_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd, + dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_STATUS); + dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_PC_LO); + dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_PC_HI); + dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_EXEC_LO); + dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_EXEC_HI); + dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_HW_ID); + dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_INST_DW0); + dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_INST_DW1); + dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_GPR_ALLOC); + dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_LDS_ALLOC); + dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_TRAPSTS); + dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_IB_STS); + dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_IB_DBG0); + dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_M0); + dst[(*no_fields)++] = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_MODE); +} + +static void gfx_v9_4_3_read_wave_sgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t start, uint32_t size, uint32_t *dst) { - wave_read_regs(adev, simd, wave, 0, + wave_read_regs(adev, xcc_id, simd, wave, 0, start + SQIND_WAVE_SGPRS_OFFSET, size, dst); } -static void gfx_v9_4_3_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd, +static void gfx_v9_4_3_read_wave_vgprs(struct amdgpu_device *adev, uint32_t xcc_id, uint32_t simd, uint32_t wave, uint32_t thread, uint32_t start, uint32_t size, uint32_t *dst) { - wave_read_regs(adev, simd, wave, thread, + wave_read_regs(adev, xcc_id, simd, wave, thread, start + SQIND_WAVE_VGPRS_OFFSET, size, dst); } static void gfx_v9_4_3_select_me_pipe_q(struct amdgpu_device *adev, - u32 me, u32 pipe, u32 q, u32 vm) + u32 me, u32 pipe, u32 q, u32 vm, u32 xcc_id) { - soc15_grbm_select(adev, me, pipe, q, vm); + soc15_grbm_select(adev, me, pipe, q, vm, GET_INST(GC, xcc_id)); } -static bool gfx_v9_4_3_is_rlc_enabled(struct amdgpu_device *adev) + +static int gfx_v9_4_3_switch_compute_partition(struct amdgpu_device *adev, + int num_xccs_per_xcp) { - uint32_t rlc_setting; + int ret; - /* if RLC is not enabled, do nothing */ - rlc_setting = RREG32_SOC15(GC, 0, regRLC_CNTL); - if (!(rlc_setting & RLC_CNTL__RLC_ENABLE_F32_MASK)) - return false; + ret = psp_spatial_partition(&adev->psp, NUM_XCC(adev->gfx.xcc_mask) / + num_xccs_per_xcp); + if (ret) + return ret; - return true; + adev->gfx.num_xcc_per_xcp = num_xccs_per_xcp; + + return ret; } -static void gfx_v9_4_3_set_safe_mode(struct amdgpu_device *adev) +static int gfx_v9_4_3_ih_to_xcc_inst(struct amdgpu_device *adev, int ih_node) { - uint32_t data; - unsigned i; - - data = RLC_SAFE_MODE__CMD_MASK; - data |= (1 << RLC_SAFE_MODE__MESSAGE__SHIFT); - WREG32_SOC15(GC, 0, regRLC_SAFE_MODE, data); + int xcc; - /* wait for RLC_SAFE_MODE */ - for (i = 0; i < adev->usec_timeout; i++) { - if (!REG_GET_FIELD(RREG32_SOC15(GC, 0, regRLC_SAFE_MODE), RLC_SAFE_MODE, CMD)) - break; - udelay(1); + xcc = hweight8(adev->gfx.xcc_mask & GENMASK(ih_node / 2, 0)); + if (!xcc) { + dev_err(adev->dev, "Couldn't find xcc mapping from IH node"); + return -EINVAL; } + + return xcc - 1; } -static void gfx_v9_4_3_unset_safe_mode(struct amdgpu_device *adev) +static const struct amdgpu_gfx_funcs gfx_v9_4_3_gfx_funcs = { + .get_gpu_clock_counter = &gfx_v9_4_3_get_gpu_clock_counter, + .select_se_sh = &gfx_v9_4_3_xcc_select_se_sh, + .read_wave_data = &gfx_v9_4_3_read_wave_data, + .read_wave_sgprs = &gfx_v9_4_3_read_wave_sgprs, + .read_wave_vgprs = &gfx_v9_4_3_read_wave_vgprs, + .select_me_pipe_q = &gfx_v9_4_3_select_me_pipe_q, + .switch_partition_mode = &gfx_v9_4_3_switch_compute_partition, + .ih_node_to_logical_xcc = &gfx_v9_4_3_ih_to_xcc_inst, +}; + +static int gfx_v9_4_3_gpu_early_init(struct amdgpu_device *adev) { - uint32_t data; + u32 gb_addr_config; - data = RLC_SAFE_MODE__CMD_MASK; - WREG32_SOC15(GC, 0, regRLC_SAFE_MODE, data); + adev->gfx.funcs = &gfx_v9_4_3_gfx_funcs; + adev->gfx.ras = &gfx_v9_4_3_ras; + + switch (adev->ip_versions[GC_HWIP][0]) { + case IP_VERSION(9, 4, 3): + adev->gfx.config.max_hw_contexts = 8; + adev->gfx.config.sc_prim_fifo_size_frontend = 0x20; + adev->gfx.config.sc_prim_fifo_size_backend = 0x100; + adev->gfx.config.sc_hiz_tile_fifo_size = 0x30; + adev->gfx.config.sc_earlyz_tile_fifo_size = 0x4C0; + gb_addr_config = RREG32_SOC15(GC, GET_INST(GC, 0), regGB_ADDR_CONFIG); + break; + default: + BUG(); + break; + } + + adev->gfx.config.gb_addr_config = gb_addr_config; + + adev->gfx.config.gb_addr_config_fields.num_pipes = 1 << + REG_GET_FIELD( + adev->gfx.config.gb_addr_config, + GB_ADDR_CONFIG, + NUM_PIPES); + + adev->gfx.config.max_tile_pipes = + adev->gfx.config.gb_addr_config_fields.num_pipes; + + adev->gfx.config.gb_addr_config_fields.num_banks = 1 << + REG_GET_FIELD( + adev->gfx.config.gb_addr_config, + GB_ADDR_CONFIG, + NUM_BANKS); + adev->gfx.config.gb_addr_config_fields.max_compress_frags = 1 << + REG_GET_FIELD( + adev->gfx.config.gb_addr_config, + GB_ADDR_CONFIG, + MAX_COMPRESSED_FRAGS); + adev->gfx.config.gb_addr_config_fields.num_rb_per_se = 1 << + REG_GET_FIELD( + adev->gfx.config.gb_addr_config, + GB_ADDR_CONFIG, + NUM_RB_PER_SE); + adev->gfx.config.gb_addr_config_fields.num_se = 1 << + REG_GET_FIELD( + adev->gfx.config.gb_addr_config, + GB_ADDR_CONFIG, + NUM_SHADER_ENGINES); + adev->gfx.config.gb_addr_config_fields.pipe_interleave_size = 1 << (8 + + REG_GET_FIELD( + adev->gfx.config.gb_addr_config, + GB_ADDR_CONFIG, + PIPE_INTERLEAVE_SIZE)); + + return 0; } -static int gfx_v9_4_3_rlc_init(struct amdgpu_device *adev) +static int gfx_v9_4_3_compute_ring_init(struct amdgpu_device *adev, int ring_id, + int xcc_id, int mec, int pipe, int queue) { - /* init spm vmid with 0xf */ - if (adev->gfx.rlc.funcs->update_spm_vmid) - adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf); + unsigned irq_type; + struct amdgpu_ring *ring = &adev->gfx.compute_ring[ring_id]; + unsigned int hw_prio; + uint32_t xcc_doorbell_start; - return 0; + ring = &adev->gfx.compute_ring[xcc_id * adev->gfx.num_compute_rings + + ring_id]; + + /* mec0 is me1 */ + ring->xcc_id = xcc_id; + ring->me = mec + 1; + ring->pipe = pipe; + ring->queue = queue; + + ring->ring_obj = NULL; + ring->use_doorbell = true; + xcc_doorbell_start = adev->doorbell_index.mec_ring0 + + xcc_id * adev->doorbell_index.xcc_doorbell_range; + ring->doorbell_index = (xcc_doorbell_start + ring_id) << 1; + ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr + + (ring_id + xcc_id * adev->gfx.num_compute_rings) * + GFX9_MEC_HPD_SIZE; + ring->vm_hub = AMDGPU_GFXHUB(xcc_id); + sprintf(ring->name, "comp_%d.%d.%d.%d", + ring->xcc_id, ring->me, ring->pipe, ring->queue); + + irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP + + ((ring->me - 1) * adev->gfx.mec.num_pipe_per_mec) + + ring->pipe; + hw_prio = amdgpu_gfx_is_high_priority_compute_queue(adev, ring) ? + AMDGPU_GFX_PIPE_PRIO_HIGH : AMDGPU_GFX_PIPE_PRIO_NORMAL; + /* type-2 packets are deprecated on MEC, use type-3 instead */ + return amdgpu_ring_init(adev, ring, 1024, &adev->gfx.eop_irq, irq_type, + hw_prio, NULL); } -static void gfx_v9_4_3_wait_for_rlc_serdes(struct amdgpu_device *adev) +static int gfx_v9_4_3_sw_init(void *handle) { - u32 i, j, k; - u32 mask; + int i, j, k, r, ring_id, xcc_id, num_xcc; + struct amdgpu_kiq *kiq; + struct amdgpu_device *adev = (struct amdgpu_device *)handle; - mutex_lock(&adev->grbm_idx_mutex); - for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { - for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { - gfx_v9_4_3_select_se_sh(adev, i, j, 0xffffffff); - for (k = 0; k < adev->usec_timeout; k++) { - if (RREG32_SOC15(GC, 0, regRLC_SERDES_CU_MASTER_BUSY) == 0) - break; - udelay(1); - } - if (k == adev->usec_timeout) { - gfx_v9_4_3_select_se_sh(adev, 0xffffffff, - 0xffffffff, 0xffffffff); - mutex_unlock(&adev->grbm_idx_mutex); - DRM_INFO("Timeout wait for RLC serdes %u,%u\n", - i, j); - return; - } - } + adev->gfx.mec.num_mec = 2; + adev->gfx.mec.num_pipe_per_mec = 4; + adev->gfx.mec.num_queue_per_pipe = 8; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + + /* EOP Event */ + r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_GRBM_CP, GFX_9_0__SRCID__CP_EOP_INTERRUPT, &adev->gfx.eop_irq); + if (r) + return r; + + /* Privileged reg */ + r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_GRBM_CP, GFX_9_0__SRCID__CP_PRIV_REG_FAULT, + &adev->gfx.priv_reg_irq); + if (r) + return r; + + /* Privileged inst */ + r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_GRBM_CP, GFX_9_0__SRCID__CP_PRIV_INSTR_FAULT, + &adev->gfx.priv_inst_irq); + if (r) + return r; + + adev->gfx.gfx_current_status = AMDGPU_GFX_NORMAL_MODE; + + r = adev->gfx.rlc.funcs->init(adev); + if (r) { + DRM_ERROR("Failed to init rlc BOs!\n"); + return r; } - gfx_v9_4_3_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); - mutex_unlock(&adev->grbm_idx_mutex); - mask = RLC_SERDES_NONCU_MASTER_BUSY__SE_MASTER_BUSY_MASK | - RLC_SERDES_NONCU_MASTER_BUSY__GC_MASTER_BUSY_MASK | - RLC_SERDES_NONCU_MASTER_BUSY__TC0_MASTER_BUSY_MASK | - RLC_SERDES_NONCU_MASTER_BUSY__TC1_MASTER_BUSY_MASK; - for (k = 0; k < adev->usec_timeout; k++) { - if ((RREG32_SOC15(GC, 0, regRLC_SERDES_NONCU_MASTER_BUSY) & mask) == 0) - break; - udelay(1); + r = gfx_v9_4_3_mec_init(adev); + if (r) { + DRM_ERROR("Failed to init MEC BOs!\n"); + return r; } -} -static void gfx_v9_4_3_enable_gui_idle_interrupt(struct amdgpu_device *adev, - bool enable) -{ - u32 tmp; + /* set up the compute queues - allocate horizontally across pipes */ + for (xcc_id = 0; xcc_id < num_xcc; xcc_id++) { + ring_id = 0; + for (i = 0; i < adev->gfx.mec.num_mec; ++i) { + for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) { + for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; + k++) { + if (!amdgpu_gfx_is_mec_queue_enabled( + adev, xcc_id, i, k, j)) + continue; - /* These interrupts should be enabled to drive DS clock */ + r = gfx_v9_4_3_compute_ring_init(adev, + ring_id, + xcc_id, + i, k, j); + if (r) + return r; - tmp = RREG32_SOC15(GC, 0, regCP_INT_CNTL_RING0); + ring_id++; + } + } + } - tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CNTX_BUSY_INT_ENABLE, enable ? 1 : 0); - tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CNTX_EMPTY_INT_ENABLE, enable ? 1 : 0); - tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CMP_BUSY_INT_ENABLE, enable ? 1 : 0); - if (adev->gfx.num_gfx_rings) - tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, GFX_IDLE_INT_ENABLE, enable ? 1 : 0); + r = amdgpu_gfx_kiq_init(adev, GFX9_MEC_HPD_SIZE, xcc_id); + if (r) { + DRM_ERROR("Failed to init KIQ BOs!\n"); + return r; + } - WREG32_SOC15(GC, 0, regCP_INT_CNTL_RING0, tmp); -} + kiq = &adev->gfx.kiq[xcc_id]; + r = amdgpu_gfx_kiq_init_ring(adev, &kiq->ring, &kiq->irq, xcc_id); + if (r) + return r; -static void gfx_v9_4_3_rlc_stop(struct amdgpu_device *adev) -{ - WREG32_FIELD15_PREREG(GC, 0, RLC_CNTL, RLC_ENABLE_F32, 0); - gfx_v9_4_3_enable_gui_idle_interrupt(adev, false); - gfx_v9_4_3_wait_for_rlc_serdes(adev); + /* create MQD for all compute queues as wel as KIQ for SRIOV case */ + r = amdgpu_gfx_mqd_sw_init(adev, + sizeof(struct v9_mqd_allocation), xcc_id); + if (r) + return r; + } + + r = gfx_v9_4_3_gpu_early_init(adev); + if (r) + return r; + + r = amdgpu_gfx_sysfs_init(adev); + if (r) + return r; + + return amdgpu_gfx_ras_sw_init(adev); } -static void gfx_v9_4_3_rlc_reset(struct amdgpu_device *adev) +static int gfx_v9_4_3_sw_fini(void *handle) { - WREG32_FIELD15_PREREG(GC, 0, GRBM_SOFT_RESET, SOFT_RESET_RLC, 1); - udelay(50); - WREG32_FIELD15_PREREG(GC, 0, GRBM_SOFT_RESET, SOFT_RESET_RLC, 0); - udelay(50); + int i, num_xcc; + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + for (i = 0; i < adev->gfx.num_compute_rings * num_xcc; i++) + amdgpu_ring_fini(&adev->gfx.compute_ring[i]); + + for (i = 0; i < num_xcc; i++) { + amdgpu_gfx_mqd_sw_fini(adev, i); + amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[i].ring); + amdgpu_gfx_kiq_fini(adev, i); + } + + gfx_v9_4_3_mec_fini(adev); + amdgpu_bo_unref(&adev->gfx.rlc.clear_state_obj); + gfx_v9_4_3_free_microcode(adev); + amdgpu_gfx_sysfs_fini(adev); + + return 0; } -static void gfx_v9_4_3_rlc_start(struct amdgpu_device *adev) +#define DEFAULT_SH_MEM_BASES (0x6000) +static void gfx_v9_4_3_xcc_init_compute_vmid(struct amdgpu_device *adev, + int xcc_id) { -#ifdef AMDGPU_RLC_DEBUG_RETRY - u32 rlc_ucode_ver; -#endif + int i; + uint32_t sh_mem_config; + uint32_t sh_mem_bases; - WREG32_FIELD15_PREREG(GC, 0, RLC_CNTL, RLC_ENABLE_F32, 1); - udelay(50); + /* + * Configure apertures: + * LDS: 0x60000000'00000000 - 0x60000001'00000000 (4GB) + * Scratch: 0x60000001'00000000 - 0x60000002'00000000 (4GB) + * GPUVM: 0x60010000'00000000 - 0x60020000'00000000 (1TB) + */ + sh_mem_bases = DEFAULT_SH_MEM_BASES | (DEFAULT_SH_MEM_BASES << 16); - /* carrizo do enable cp interrupt after cp inited */ - if (!(adev->flags & AMD_IS_APU)) { - gfx_v9_4_3_enable_gui_idle_interrupt(adev, true); - udelay(50); + sh_mem_config = SH_MEM_ADDRESS_MODE_64 | + SH_MEM_ALIGNMENT_MODE_UNALIGNED << + SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT; + + mutex_lock(&adev->srbm_mutex); + for (i = adev->vm_manager.first_kfd_vmid; i < AMDGPU_NUM_VMID; i++) { + soc15_grbm_select(adev, 0, 0, 0, i, GET_INST(GC, xcc_id)); + /* CP and shaders */ + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regSH_MEM_CONFIG, sh_mem_config); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regSH_MEM_BASES, sh_mem_bases); } + soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id)); + mutex_unlock(&adev->srbm_mutex); -#ifdef AMDGPU_RLC_DEBUG_RETRY - /* RLC_GPM_GENERAL_6 : RLC Ucode version */ - rlc_ucode_ver = RREG32_SOC15(GC, 0, regRLC_GPM_GENERAL_6); - if (rlc_ucode_ver == 0x108) { - dev_info(adev->dev, - "Using rlc debug ucode. regRLC_GPM_GENERAL_6 ==0x08%x / fw_ver == %i \n", - rlc_ucode_ver, adev->gfx.rlc_fw_version); - /* RLC_GPM_TIMER_INT_3 : Timer interval in RefCLK cycles, - * default is 0x9C4 to create a 100us interval */ - WREG32_SOC15(GC, 0, regRLC_GPM_TIMER_INT_3, 0x9C4); - /* RLC_GPM_GENERAL_12 : Minimum gap between wptr and rptr - * to disable the page fault retry interrupts, default is - * 0x100 (256) */ - WREG32_SOC15(GC, 0, regRLC_GPM_GENERAL_12, 0x100); + /* Initialize all compute VMIDs to have no GDS, GWS, or OA + acccess. These should be enabled by FW for target VMIDs. */ + for (i = adev->vm_manager.first_kfd_vmid; i < AMDGPU_NUM_VMID; i++) { + WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_VMID0_BASE, 2 * i, 0); + WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_VMID0_SIZE, 2 * i, 0); + WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_GWS_VMID0, i, 0); + WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_OA_VMID0, i, 0); } -#endif } -static int gfx_v9_4_3_rlc_load_microcode(struct amdgpu_device *adev) +static void gfx_v9_4_3_xcc_init_gds_vmid(struct amdgpu_device *adev, int xcc_id) { - const struct rlc_firmware_header_v2_0 *hdr; - const __le32 *fw_data; - unsigned i, fw_size; + int vmid; + + /* + * Initialize all compute and user-gfx VMIDs to have no GDS, GWS, or OA + * access. Compute VMIDs should be enabled by FW for target VMIDs, + * the driver can enable them for graphics. VMID0 should maintain + * access so that HWS firmware can save/restore entries. + */ + for (vmid = 1; vmid < AMDGPU_NUM_VMID; vmid++) { + WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_VMID0_BASE, 2 * vmid, 0); + WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_VMID0_SIZE, 2 * vmid, 0); + WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_GWS_VMID0, vmid, 0); + WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_OA_VMID0, vmid, 0); + } +} + +static void gfx_v9_4_3_xcc_constants_init(struct amdgpu_device *adev, + int xcc_id) +{ + u32 tmp; + int i; + + /* XXX SH_MEM regs */ + /* where to put LDS, scratch, GPUVM in FSA64 space */ + mutex_lock(&adev->srbm_mutex); + for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB(0)].num_ids; i++) { + soc15_grbm_select(adev, 0, 0, 0, i, GET_INST(GC, xcc_id)); + /* CP and shaders */ + if (i == 0) { + tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE, + SH_MEM_ALIGNMENT_MODE_UNALIGNED); + tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE, + !!adev->gmc.noretry); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), + regSH_MEM_CONFIG, tmp); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), + regSH_MEM_BASES, 0); + } else { + tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE, + SH_MEM_ALIGNMENT_MODE_UNALIGNED); + tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE, + !!adev->gmc.noretry); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), + regSH_MEM_CONFIG, tmp); + tmp = REG_SET_FIELD(0, SH_MEM_BASES, PRIVATE_BASE, + (adev->gmc.private_aperture_start >> + 48)); + tmp = REG_SET_FIELD(tmp, SH_MEM_BASES, SHARED_BASE, + (adev->gmc.shared_aperture_start >> + 48)); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), + regSH_MEM_BASES, tmp); + } + } + soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, 0)); + + mutex_unlock(&adev->srbm_mutex); + + gfx_v9_4_3_xcc_init_compute_vmid(adev, xcc_id); + gfx_v9_4_3_xcc_init_gds_vmid(adev, xcc_id); +} + +static void gfx_v9_4_3_constants_init(struct amdgpu_device *adev) +{ + int i, num_xcc; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + + gfx_v9_4_3_get_cu_info(adev, &adev->gfx.cu_info); + adev->gfx.config.db_debug2 = + RREG32_SOC15(GC, GET_INST(GC, 0), regDB_DEBUG2); + + for (i = 0; i < num_xcc; i++) + gfx_v9_4_3_xcc_constants_init(adev, i); +} + +static void +gfx_v9_4_3_xcc_enable_save_restore_machine(struct amdgpu_device *adev, + int xcc_id) +{ + WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), RLC_SRM_CNTL, SRM_ENABLE, 1); +} + +static void gfx_v9_4_3_xcc_init_pg(struct amdgpu_device *adev, int xcc_id) +{ + /* + * Rlc save restore list is workable since v2_1. + * And it's needed by gfxoff feature. + */ + if (adev->gfx.rlc.is_rlc_v2_1) + gfx_v9_4_3_xcc_enable_save_restore_machine(adev, xcc_id); +} + +static void gfx_v9_4_3_xcc_disable_gpa_mode(struct amdgpu_device *adev, int xcc_id) +{ + uint32_t data; + + data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCPC_PSP_DEBUG); + data |= CPC_PSP_DEBUG__UTCL2IUGPAOVERRIDE_MASK; + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCPC_PSP_DEBUG, data); +} + +static void gfx_v9_4_3_xcc_program_xcc_id(struct amdgpu_device *adev, + int xcc_id) +{ + uint32_t tmp = 0; + int num_xcc; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + switch (num_xcc) { + /* directly config VIRTUAL_XCC_ID to 0 for 1-XCC */ + case 1: + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HYP_XCP_CTL, 0x8); + break; + case 2: + case 4: + case 6: + case 8: + tmp = (xcc_id % adev->gfx.num_xcc_per_xcp) << REG_FIELD_SHIFT(CP_HYP_XCP_CTL, VIRTUAL_XCC_ID); + tmp = tmp | (adev->gfx.num_xcc_per_xcp << REG_FIELD_SHIFT(CP_HYP_XCP_CTL, NUM_XCC_IN_XCP)); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HYP_XCP_CTL, tmp); + + break; + default: + break; + } +} + +static bool gfx_v9_4_3_is_rlc_enabled(struct amdgpu_device *adev) +{ + uint32_t rlc_setting; + + /* if RLC is not enabled, do nothing */ + rlc_setting = RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_CNTL); + if (!(rlc_setting & RLC_CNTL__RLC_ENABLE_F32_MASK)) + return false; + + return true; +} + +static void gfx_v9_4_3_xcc_set_safe_mode(struct amdgpu_device *adev, int xcc_id) +{ + uint32_t data; + unsigned i; + + data = RLC_SAFE_MODE__CMD_MASK; + data |= (1 << RLC_SAFE_MODE__MESSAGE__SHIFT); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_SAFE_MODE, data); + + /* wait for RLC_SAFE_MODE */ + for (i = 0; i < adev->usec_timeout; i++) { + if (!REG_GET_FIELD(RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_SAFE_MODE), RLC_SAFE_MODE, CMD)) + break; + udelay(1); + } +} + +static void gfx_v9_4_3_xcc_unset_safe_mode(struct amdgpu_device *adev, + int xcc_id) +{ + uint32_t data; + + data = RLC_SAFE_MODE__CMD_MASK; + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_SAFE_MODE, data); +} + +static int gfx_v9_4_3_rlc_init(struct amdgpu_device *adev) +{ + /* init spm vmid with 0xf */ + if (adev->gfx.rlc.funcs->update_spm_vmid) + adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf); + + return 0; +} + +static void gfx_v9_4_3_xcc_wait_for_rlc_serdes(struct amdgpu_device *adev, + int xcc_id) +{ + u32 i, j, k; + u32 mask; + + mutex_lock(&adev->grbm_idx_mutex); + for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { + for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { + gfx_v9_4_3_xcc_select_se_sh(adev, i, j, 0xffffffff, + xcc_id); + for (k = 0; k < adev->usec_timeout; k++) { + if (RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_SERDES_CU_MASTER_BUSY) == 0) + break; + udelay(1); + } + if (k == adev->usec_timeout) { + gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, + 0xffffffff, + 0xffffffff, xcc_id); + mutex_unlock(&adev->grbm_idx_mutex); + DRM_INFO("Timeout wait for RLC serdes %u,%u\n", + i, j); + return; + } + } + } + gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, + xcc_id); + mutex_unlock(&adev->grbm_idx_mutex); + + mask = RLC_SERDES_NONCU_MASTER_BUSY__SE_MASTER_BUSY_MASK | + RLC_SERDES_NONCU_MASTER_BUSY__GC_MASTER_BUSY_MASK | + RLC_SERDES_NONCU_MASTER_BUSY__TC0_MASTER_BUSY_MASK | + RLC_SERDES_NONCU_MASTER_BUSY__TC1_MASTER_BUSY_MASK; + for (k = 0; k < adev->usec_timeout; k++) { + if ((RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_SERDES_NONCU_MASTER_BUSY) & mask) == 0) + break; + udelay(1); + } +} + +static void gfx_v9_4_3_xcc_enable_gui_idle_interrupt(struct amdgpu_device *adev, + bool enable, int xcc_id) +{ + u32 tmp; + + /* These interrupts should be enabled to drive DS clock */ + + tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_INT_CNTL_RING0); + + tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CNTX_BUSY_INT_ENABLE, enable ? 1 : 0); + tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CNTX_EMPTY_INT_ENABLE, enable ? 1 : 0); + tmp = REG_SET_FIELD(tmp, CP_INT_CNTL_RING0, CMP_BUSY_INT_ENABLE, enable ? 1 : 0); + + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_INT_CNTL_RING0, tmp); +} + +static void gfx_v9_4_3_xcc_rlc_stop(struct amdgpu_device *adev, int xcc_id) +{ + WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), RLC_CNTL, + RLC_ENABLE_F32, 0); + gfx_v9_4_3_xcc_enable_gui_idle_interrupt(adev, false, xcc_id); + gfx_v9_4_3_xcc_wait_for_rlc_serdes(adev, xcc_id); +} + +static void gfx_v9_4_3_rlc_stop(struct amdgpu_device *adev) +{ + int i, num_xcc; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + for (i = 0; i < num_xcc; i++) + gfx_v9_4_3_xcc_rlc_stop(adev, i); +} + +static void gfx_v9_4_3_xcc_rlc_reset(struct amdgpu_device *adev, int xcc_id) +{ + WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), GRBM_SOFT_RESET, + SOFT_RESET_RLC, 1); + udelay(50); + WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), GRBM_SOFT_RESET, + SOFT_RESET_RLC, 0); + udelay(50); +} + +static void gfx_v9_4_3_rlc_reset(struct amdgpu_device *adev) +{ + int i, num_xcc; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + for (i = 0; i < num_xcc; i++) + gfx_v9_4_3_xcc_rlc_reset(adev, i); +} + +static void gfx_v9_4_3_xcc_rlc_start(struct amdgpu_device *adev, int xcc_id) +{ + WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), RLC_CNTL, + RLC_ENABLE_F32, 1); + udelay(50); + + /* carrizo do enable cp interrupt after cp inited */ + if (!(adev->flags & AMD_IS_APU)) { + gfx_v9_4_3_xcc_enable_gui_idle_interrupt(adev, true, xcc_id); + udelay(50); + } +} + +static void gfx_v9_4_3_rlc_start(struct amdgpu_device *adev) +{ +#ifdef AMDGPU_RLC_DEBUG_RETRY + u32 rlc_ucode_ver; +#endif + int i, num_xcc; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + for (i = 0; i < num_xcc; i++) { + gfx_v9_4_3_xcc_rlc_start(adev, i); +#ifdef AMDGPU_RLC_DEBUG_RETRY + /* RLC_GPM_GENERAL_6 : RLC Ucode version */ + rlc_ucode_ver = RREG32_SOC15(GC, GET_INST(GC, i), regRLC_GPM_GENERAL_6); + if (rlc_ucode_ver == 0x108) { + dev_info(adev->dev, + "Using rlc debug ucode. regRLC_GPM_GENERAL_6 ==0x08%x / fw_ver == %i \n", + rlc_ucode_ver, adev->gfx.rlc_fw_version); + /* RLC_GPM_TIMER_INT_3 : Timer interval in RefCLK cycles, + * default is 0x9C4 to create a 100us interval */ + WREG32_SOC15(GC, GET_INST(GC, i), regRLC_GPM_TIMER_INT_3, 0x9C4); + /* RLC_GPM_GENERAL_12 : Minimum gap between wptr and rptr + * to disable the page fault retry interrupts, default is + * 0x100 (256) */ + WREG32_SOC15(GC, GET_INST(GC, i), regRLC_GPM_GENERAL_12, 0x100); + } +#endif + } +} + +static int gfx_v9_4_3_xcc_rlc_load_microcode(struct amdgpu_device *adev, + int xcc_id) +{ + const struct rlc_firmware_header_v2_0 *hdr; + const __le32 *fw_data; + unsigned i, fw_size; if (!adev->gfx.rlc_fw) return -EINVAL; @@ -316,49 +1247,65 @@ static int gfx_v9_4_3_rlc_load_microcode(struct amdgpu_device *adev) le32_to_cpu(hdr->header.ucode_array_offset_bytes)); fw_size = le32_to_cpu(hdr->header.ucode_size_bytes) / 4; - WREG32_SOC15(GC, 0, regRLC_GPM_UCODE_ADDR, + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_GPM_UCODE_ADDR, RLCG_UCODE_LOADING_START_ADDRESS); for (i = 0; i < fw_size; i++) { if (amdgpu_emu_mode == 1 && i % 100 == 0) { dev_info(adev->dev, "Write RLC ucode data %u DWs\n", i); msleep(1); } - WREG32_SOC15(GC, 0, regRLC_GPM_UCODE_DATA, le32_to_cpup(fw_data++)); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_GPM_UCODE_DATA, le32_to_cpup(fw_data++)); } - WREG32_SOC15(GC, 0, regRLC_GPM_UCODE_ADDR, adev->gfx.rlc_fw_version); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_GPM_UCODE_ADDR, adev->gfx.rlc_fw_version); return 0; } -static int gfx_v9_4_3_rlc_resume(struct amdgpu_device *adev) +static int gfx_v9_4_3_xcc_rlc_resume(struct amdgpu_device *adev, int xcc_id) { int r; - adev->gfx.rlc.funcs->stop(adev); + if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) { + gfx_v9_4_3_xcc_rlc_stop(adev, xcc_id); + /* legacy rlc firmware loading */ + r = gfx_v9_4_3_xcc_rlc_load_microcode(adev, xcc_id); + if (r) + return r; + gfx_v9_4_3_xcc_rlc_start(adev, xcc_id); + } + amdgpu_gfx_rlc_enter_safe_mode(adev, xcc_id); /* disable CG */ - WREG32_SOC15(GC, 0, regRLC_CGCG_CGLS_CTRL, 0); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL, 0); + gfx_v9_4_3_xcc_init_pg(adev, xcc_id); + amdgpu_gfx_rlc_exit_safe_mode(adev, xcc_id); - /* TODO: revisit pg function */ - /* gfx_v9_4_3_init_pg(adev);*/ + return 0; +} - if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) { - /* legacy rlc firmware loading */ - r = gfx_v9_4_3_rlc_load_microcode(adev); +static int gfx_v9_4_3_rlc_resume(struct amdgpu_device *adev) +{ + int r, i, num_xcc; + + if (amdgpu_sriov_vf(adev)) + return 0; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + for (i = 0; i < num_xcc; i++) { + r = gfx_v9_4_3_xcc_rlc_resume(adev, i); if (r) return r; } - adev->gfx.rlc.funcs->start(adev); - return 0; } -static void gfx_v9_4_3_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) +static void gfx_v9_4_3_update_spm_vmid(struct amdgpu_device *adev, + unsigned vmid) { u32 reg, data; - reg = SOC15_REG_OFFSET(GC, 0, regRLC_SPM_MC_CNTL); + reg = SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regRLC_SPM_MC_CNTL); if (amdgpu_sriov_is_pp_one_vf(adev)) data = RREG32_NO_KIQ(reg); else @@ -368,9 +1315,9 @@ static void gfx_v9_4_3_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid data |= (vmid & RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK) << RLC_SPM_MC_CNTL__RLC_SPM_VMID__SHIFT; if (amdgpu_sriov_is_pp_one_vf(adev)) - WREG32_SOC15_NO_KIQ(GC, 0, regRLC_SPM_MC_CNTL, data); + WREG32_SOC15_NO_KIQ(GC, GET_INST(GC, 0), regRLC_SPM_MC_CNTL, data); else - WREG32_SOC15(GC, 0, regRLC_SPM_MC_CNTL, data); + WREG32_SOC15(GC, GET_INST(GC, 0), regRLC_SPM_MC_CNTL, data); } static const struct soc15_reg_rlcg rlcg_access_gc_9_4_3[] = { @@ -382,7 +1329,7 @@ static bool gfx_v9_4_3_check_rlcg_range(struct amdgpu_device *adev, uint32_t offset, struct soc15_reg_rlcg *entries, int arr_size) { - int i; + int i, inst; uint32_t reg; if (!entries) @@ -392,7 +1339,12 @@ static bool gfx_v9_4_3_check_rlcg_range(struct amdgpu_device *adev, const struct soc15_reg_rlcg *entry; entry = &entries[i]; - reg = adev->reg_offset[entry->hwip][entry->instance][entry->segment] + entry->reg; + inst = adev->ip_map.logical_to_dev_inst ? + adev->ip_map.logical_to_dev_inst( + adev, entry->hwip, entry->instance) : + entry->instance; + reg = adev->reg_offset[entry->hwip][inst][entry->segment] + + entry->reg; if (offset == reg) return true; } @@ -407,24 +1359,3009 @@ static bool gfx_v9_4_3_is_rlcg_access_range(struct amdgpu_device *adev, u32 offs ARRAY_SIZE(rlcg_access_gc_9_4_3)); } -const struct amdgpu_gfx_funcs gfx_v9_4_3_gfx_funcs = { - .get_gpu_clock_counter = &gfx_v9_4_3_get_gpu_clock_counter, - .select_se_sh = &gfx_v9_4_3_select_se_sh, - .read_wave_data = &gfx_v9_4_3_read_wave_data, - .read_wave_sgprs = &gfx_v9_4_3_read_wave_sgprs, - .read_wave_vgprs = &gfx_v9_4_3_read_wave_vgprs, - .select_me_pipe_q = &gfx_v9_4_3_select_me_pipe_q, -}; +static void gfx_v9_4_3_xcc_cp_compute_enable(struct amdgpu_device *adev, + bool enable, int xcc_id) +{ + if (enable) { + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_MEC_CNTL, 0); + } else { + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_MEC_CNTL, + (CP_MEC_CNTL__MEC_ME1_HALT_MASK | CP_MEC_CNTL__MEC_ME2_HALT_MASK)); + adev->gfx.kiq[xcc_id].ring.sched.ready = false; + } + udelay(50); +} -const struct amdgpu_rlc_funcs gfx_v9_4_3_rlc_funcs = { - .is_rlc_enabled = gfx_v9_4_3_is_rlc_enabled, - .set_safe_mode = gfx_v9_4_3_set_safe_mode, - .unset_safe_mode = gfx_v9_4_3_unset_safe_mode, - .init = gfx_v9_4_3_rlc_init, - .resume = gfx_v9_4_3_rlc_resume, - .stop = gfx_v9_4_3_rlc_stop, - .reset = gfx_v9_4_3_rlc_reset, - .start = gfx_v9_4_3_rlc_start, - .update_spm_vmid = gfx_v9_4_3_update_spm_vmid, - .is_rlcg_access_range = gfx_v9_4_3_is_rlcg_access_range, +static int gfx_v9_4_3_xcc_cp_compute_load_microcode(struct amdgpu_device *adev, + int xcc_id) +{ + const struct gfx_firmware_header_v1_0 *mec_hdr; + const __le32 *fw_data; + unsigned i; + u32 tmp; + u32 mec_ucode_addr_offset; + u32 mec_ucode_data_offset; + + if (!adev->gfx.mec_fw) + return -EINVAL; + + gfx_v9_4_3_xcc_cp_compute_enable(adev, false, xcc_id); + + mec_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec_fw->data; + amdgpu_ucode_print_gfx_hdr(&mec_hdr->header); + + fw_data = (const __le32 *) + (adev->gfx.mec_fw->data + + le32_to_cpu(mec_hdr->header.ucode_array_offset_bytes)); + tmp = 0; + tmp = REG_SET_FIELD(tmp, CP_CPC_IC_BASE_CNTL, VMID, 0); + tmp = REG_SET_FIELD(tmp, CP_CPC_IC_BASE_CNTL, CACHE_POLICY, 0); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_CPC_IC_BASE_CNTL, tmp); + + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_CPC_IC_BASE_LO, + adev->gfx.mec.mec_fw_gpu_addr & 0xFFFFF000); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_CPC_IC_BASE_HI, + upper_32_bits(adev->gfx.mec.mec_fw_gpu_addr)); + + mec_ucode_addr_offset = + SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regCP_MEC_ME1_UCODE_ADDR); + mec_ucode_data_offset = + SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regCP_MEC_ME1_UCODE_DATA); + + /* MEC1 */ + WREG32(mec_ucode_addr_offset, mec_hdr->jt_offset); + for (i = 0; i < mec_hdr->jt_size; i++) + WREG32(mec_ucode_data_offset, + le32_to_cpup(fw_data + mec_hdr->jt_offset + i)); + + WREG32(mec_ucode_addr_offset, adev->gfx.mec_fw_version); + /* Todo : Loading MEC2 firmware is only necessary if MEC2 should run different microcode than MEC1. */ + + return 0; +} + +/* KIQ functions */ +static void gfx_v9_4_3_xcc_kiq_setting(struct amdgpu_ring *ring, int xcc_id) +{ + uint32_t tmp; + struct amdgpu_device *adev = ring->adev; + + /* tell RLC which is KIQ queue */ + tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CP_SCHEDULERS); + tmp &= 0xffffff00; + tmp |= (ring->me << 5) | (ring->pipe << 3) | (ring->queue); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regRLC_CP_SCHEDULERS, tmp); + tmp |= 0x80; + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regRLC_CP_SCHEDULERS, tmp); +} + +static void gfx_v9_4_3_mqd_set_priority(struct amdgpu_ring *ring, struct v9_mqd *mqd) +{ + struct amdgpu_device *adev = ring->adev; + + if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) { + if (amdgpu_gfx_is_high_priority_compute_queue(adev, ring)) { + mqd->cp_hqd_pipe_priority = AMDGPU_GFX_PIPE_PRIO_HIGH; + mqd->cp_hqd_queue_priority = + AMDGPU_GFX_QUEUE_PRIORITY_MAXIMUM; + } + } +} + +static int gfx_v9_4_3_xcc_mqd_init(struct amdgpu_ring *ring, int xcc_id) +{ + struct amdgpu_device *adev = ring->adev; + struct v9_mqd *mqd = ring->mqd_ptr; + uint64_t hqd_gpu_addr, wb_gpu_addr, eop_base_addr; + uint32_t tmp; + + mqd->header = 0xC0310800; + mqd->compute_pipelinestat_enable = 0x00000001; + mqd->compute_static_thread_mgmt_se0 = 0xffffffff; + mqd->compute_static_thread_mgmt_se1 = 0xffffffff; + mqd->compute_static_thread_mgmt_se2 = 0xffffffff; + mqd->compute_static_thread_mgmt_se3 = 0xffffffff; + mqd->compute_misc_reserved = 0x00000003; + + mqd->dynamic_cu_mask_addr_lo = + lower_32_bits(ring->mqd_gpu_addr + + offsetof(struct v9_mqd_allocation, dynamic_cu_mask)); + mqd->dynamic_cu_mask_addr_hi = + upper_32_bits(ring->mqd_gpu_addr + + offsetof(struct v9_mqd_allocation, dynamic_cu_mask)); + + eop_base_addr = ring->eop_gpu_addr >> 8; + mqd->cp_hqd_eop_base_addr_lo = eop_base_addr; + mqd->cp_hqd_eop_base_addr_hi = upper_32_bits(eop_base_addr); + + /* set the EOP size, register value is 2^(EOP_SIZE+1) dwords */ + tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_EOP_CONTROL); + tmp = REG_SET_FIELD(tmp, CP_HQD_EOP_CONTROL, EOP_SIZE, + (order_base_2(GFX9_MEC_HPD_SIZE / 4) - 1)); + + mqd->cp_hqd_eop_control = tmp; + + /* enable doorbell? */ + tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_DOORBELL_CONTROL); + + if (ring->use_doorbell) { + tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL, + DOORBELL_OFFSET, ring->doorbell_index); + tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL, + DOORBELL_EN, 1); + tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL, + DOORBELL_SOURCE, 0); + tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL, + DOORBELL_HIT, 0); + } else { + tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL, + DOORBELL_EN, 0); + } + + mqd->cp_hqd_pq_doorbell_control = tmp; + + /* disable the queue if it's active */ + ring->wptr = 0; + mqd->cp_hqd_dequeue_request = 0; + mqd->cp_hqd_pq_rptr = 0; + mqd->cp_hqd_pq_wptr_lo = 0; + mqd->cp_hqd_pq_wptr_hi = 0; + + /* set the pointer to the MQD */ + mqd->cp_mqd_base_addr_lo = ring->mqd_gpu_addr & 0xfffffffc; + mqd->cp_mqd_base_addr_hi = upper_32_bits(ring->mqd_gpu_addr); + + /* set MQD vmid to 0 */ + tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MQD_CONTROL); + tmp = REG_SET_FIELD(tmp, CP_MQD_CONTROL, VMID, 0); + mqd->cp_mqd_control = tmp; + + /* set the pointer to the HQD, this is similar CP_RB0_BASE/_HI */ + hqd_gpu_addr = ring->gpu_addr >> 8; + mqd->cp_hqd_pq_base_lo = hqd_gpu_addr; + mqd->cp_hqd_pq_base_hi = upper_32_bits(hqd_gpu_addr); + + /* set up the HQD, this is similar to CP_RB0_CNTL */ + tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_CONTROL); + tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, QUEUE_SIZE, + (order_base_2(ring->ring_size / 4) - 1)); + tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, RPTR_BLOCK_SIZE, + ((order_base_2(AMDGPU_GPU_PAGE_SIZE / 4) - 1) << 8)); +#ifdef __BIG_ENDIAN + tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, ENDIAN_SWAP, 1); +#endif + tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, UNORD_DISPATCH, 0); + tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, ROQ_PQ_IB_FLIP, 0); + tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, PRIV_STATE, 1); + tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, KMD_QUEUE, 1); + mqd->cp_hqd_pq_control = tmp; + + /* set the wb address whether it's enabled or not */ + wb_gpu_addr = adev->wb.gpu_addr + (ring->rptr_offs * 4); + mqd->cp_hqd_pq_rptr_report_addr_lo = wb_gpu_addr & 0xfffffffc; + mqd->cp_hqd_pq_rptr_report_addr_hi = + upper_32_bits(wb_gpu_addr) & 0xffff; + + /* only used if CP_PQ_WPTR_POLL_CNTL.CP_PQ_WPTR_POLL_CNTL__EN_MASK=1 */ + wb_gpu_addr = adev->wb.gpu_addr + (ring->wptr_offs * 4); + mqd->cp_hqd_pq_wptr_poll_addr_lo = wb_gpu_addr & 0xfffffffc; + mqd->cp_hqd_pq_wptr_poll_addr_hi = upper_32_bits(wb_gpu_addr) & 0xffff; + + /* reset read and write pointers, similar to CP_RB0_WPTR/_RPTR */ + ring->wptr = 0; + mqd->cp_hqd_pq_rptr = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_RPTR); + + /* set the vmid for the queue */ + mqd->cp_hqd_vmid = 0; + + tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_PERSISTENT_STATE); + tmp = REG_SET_FIELD(tmp, CP_HQD_PERSISTENT_STATE, PRELOAD_SIZE, 0x53); + mqd->cp_hqd_persistent_state = tmp; + + /* set MIN_IB_AVAIL_SIZE */ + tmp = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_IB_CONTROL); + tmp = REG_SET_FIELD(tmp, CP_HQD_IB_CONTROL, MIN_IB_AVAIL_SIZE, 3); + mqd->cp_hqd_ib_control = tmp; + + /* set static priority for a queue/ring */ + gfx_v9_4_3_mqd_set_priority(ring, mqd); + mqd->cp_hqd_quantum = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_QUANTUM); + + /* map_queues packet doesn't need activate the queue, + * so only kiq need set this field. + */ + if (ring->funcs->type == AMDGPU_RING_TYPE_KIQ) + mqd->cp_hqd_active = 1; + + return 0; +} + +static int gfx_v9_4_3_xcc_kiq_init_register(struct amdgpu_ring *ring, + int xcc_id) +{ + struct amdgpu_device *adev = ring->adev; + struct v9_mqd *mqd = ring->mqd_ptr; + int j; + + /* disable wptr polling */ + WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), CP_PQ_WPTR_POLL_CNTL, EN, 0); + + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_EOP_BASE_ADDR, + mqd->cp_hqd_eop_base_addr_lo); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_EOP_BASE_ADDR_HI, + mqd->cp_hqd_eop_base_addr_hi); + + /* set the EOP size, register value is 2^(EOP_SIZE+1) dwords */ + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_EOP_CONTROL, + mqd->cp_hqd_eop_control); + + /* enable doorbell? */ + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_DOORBELL_CONTROL, + mqd->cp_hqd_pq_doorbell_control); + + /* disable the queue if it's active */ + if (RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_ACTIVE) & 1) { + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_DEQUEUE_REQUEST, 1); + for (j = 0; j < adev->usec_timeout; j++) { + if (!(RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_ACTIVE) & 1)) + break; + udelay(1); + } + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_DEQUEUE_REQUEST, + mqd->cp_hqd_dequeue_request); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_RPTR, + mqd->cp_hqd_pq_rptr); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_LO, + mqd->cp_hqd_pq_wptr_lo); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_HI, + mqd->cp_hqd_pq_wptr_hi); + } + + /* set the pointer to the MQD */ + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_MQD_BASE_ADDR, + mqd->cp_mqd_base_addr_lo); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_MQD_BASE_ADDR_HI, + mqd->cp_mqd_base_addr_hi); + + /* set MQD vmid to 0 */ + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_MQD_CONTROL, + mqd->cp_mqd_control); + + /* set the pointer to the HQD, this is similar CP_RB0_BASE/_HI */ + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_BASE, + mqd->cp_hqd_pq_base_lo); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_BASE_HI, + mqd->cp_hqd_pq_base_hi); + + /* set up the HQD, this is similar to CP_RB0_CNTL */ + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_CONTROL, + mqd->cp_hqd_pq_control); + + /* set the wb address whether it's enabled or not */ + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_RPTR_REPORT_ADDR, + mqd->cp_hqd_pq_rptr_report_addr_lo); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_RPTR_REPORT_ADDR_HI, + mqd->cp_hqd_pq_rptr_report_addr_hi); + + /* only used if CP_PQ_WPTR_POLL_CNTL.CP_PQ_WPTR_POLL_CNTL__EN_MASK=1 */ + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_POLL_ADDR, + mqd->cp_hqd_pq_wptr_poll_addr_lo); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_POLL_ADDR_HI, + mqd->cp_hqd_pq_wptr_poll_addr_hi); + + /* enable the doorbell if requested */ + if (ring->use_doorbell) { + WREG32_SOC15( + GC, GET_INST(GC, xcc_id), + regCP_MEC_DOORBELL_RANGE_LOWER, + ((adev->doorbell_index.kiq + + xcc_id * adev->doorbell_index.xcc_doorbell_range) * + 2) << 2); + WREG32_SOC15( + GC, GET_INST(GC, xcc_id), + regCP_MEC_DOORBELL_RANGE_UPPER, + ((adev->doorbell_index.userqueue_end + + xcc_id * adev->doorbell_index.xcc_doorbell_range) * + 2) << 2); + } + + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_DOORBELL_CONTROL, + mqd->cp_hqd_pq_doorbell_control); + + /* reset read and write pointers, similar to CP_RB0_WPTR/_RPTR */ + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_LO, + mqd->cp_hqd_pq_wptr_lo); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_HI, + mqd->cp_hqd_pq_wptr_hi); + + /* set the vmid for the queue */ + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_VMID, mqd->cp_hqd_vmid); + + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PERSISTENT_STATE, + mqd->cp_hqd_persistent_state); + + /* activate the queue */ + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_ACTIVE, + mqd->cp_hqd_active); + + if (ring->use_doorbell) + WREG32_FIELD15_PREREG(GC, GET_INST(GC, xcc_id), CP_PQ_STATUS, DOORBELL_ENABLE, 1); + + return 0; +} + +static int gfx_v9_4_3_xcc_q_fini_register(struct amdgpu_ring *ring, + int xcc_id) +{ + struct amdgpu_device *adev = ring->adev; + int j; + + /* disable the queue if it's active */ + if (RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_ACTIVE) & 1) { + + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_DEQUEUE_REQUEST, 1); + + for (j = 0; j < adev->usec_timeout; j++) { + if (!(RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HQD_ACTIVE) & 1)) + break; + udelay(1); + } + + if (j == AMDGPU_MAX_USEC_TIMEOUT) { + DRM_DEBUG("%s dequeue request failed.\n", ring->name); + + /* Manual disable if dequeue request times out */ + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_ACTIVE, 0); + } + + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_DEQUEUE_REQUEST, + 0); + } + + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_IQ_TIMER, 0); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_IB_CONTROL, 0); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PERSISTENT_STATE, 0); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_DOORBELL_CONTROL, 0x40000000); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_DOORBELL_CONTROL, 0); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_RPTR, 0); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_HI, 0); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_WPTR_LO, 0); + + return 0; +} + +static int gfx_v9_4_3_xcc_kiq_init_queue(struct amdgpu_ring *ring, int xcc_id) +{ + struct amdgpu_device *adev = ring->adev; + struct v9_mqd *mqd = ring->mqd_ptr; + struct v9_mqd *tmp_mqd; + + gfx_v9_4_3_xcc_kiq_setting(ring, xcc_id); + + /* GPU could be in bad state during probe, driver trigger the reset + * after load the SMU, in this case , the mqd is not be initialized. + * driver need to re-init the mqd. + * check mqd->cp_hqd_pq_control since this value should not be 0 + */ + tmp_mqd = (struct v9_mqd *)adev->gfx.kiq[xcc_id].mqd_backup; + if (amdgpu_in_reset(adev) && tmp_mqd->cp_hqd_pq_control) { + /* for GPU_RESET case , reset MQD to a clean status */ + if (adev->gfx.kiq[xcc_id].mqd_backup) + memcpy(mqd, adev->gfx.kiq[xcc_id].mqd_backup, sizeof(struct v9_mqd_allocation)); + + /* reset ring buffer */ + ring->wptr = 0; + amdgpu_ring_clear_ring(ring); + mutex_lock(&adev->srbm_mutex); + soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, GET_INST(GC, xcc_id)); + gfx_v9_4_3_xcc_kiq_init_register(ring, xcc_id); + soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id)); + mutex_unlock(&adev->srbm_mutex); + } else { + memset((void *)mqd, 0, sizeof(struct v9_mqd_allocation)); + ((struct v9_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF; + ((struct v9_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF; + mutex_lock(&adev->srbm_mutex); + soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, GET_INST(GC, xcc_id)); + gfx_v9_4_3_xcc_mqd_init(ring, xcc_id); + gfx_v9_4_3_xcc_kiq_init_register(ring, xcc_id); + soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id)); + mutex_unlock(&adev->srbm_mutex); + + if (adev->gfx.kiq[xcc_id].mqd_backup) + memcpy(adev->gfx.kiq[xcc_id].mqd_backup, mqd, sizeof(struct v9_mqd_allocation)); + } + + return 0; +} + +static int gfx_v9_4_3_xcc_kcq_init_queue(struct amdgpu_ring *ring, int xcc_id) +{ + struct amdgpu_device *adev = ring->adev; + struct v9_mqd *mqd = ring->mqd_ptr; + int mqd_idx = ring - &adev->gfx.compute_ring[0]; + struct v9_mqd *tmp_mqd; + + /* Same as above kiq init, driver need to re-init the mqd if mqd->cp_hqd_pq_control + * is not be initialized before + */ + tmp_mqd = (struct v9_mqd *)adev->gfx.mec.mqd_backup[mqd_idx]; + + if (!tmp_mqd->cp_hqd_pq_control || + (!amdgpu_in_reset(adev) && !adev->in_suspend)) { + memset((void *)mqd, 0, sizeof(struct v9_mqd_allocation)); + ((struct v9_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF; + ((struct v9_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF; + mutex_lock(&adev->srbm_mutex); + soc15_grbm_select(adev, ring->me, ring->pipe, ring->queue, 0, GET_INST(GC, xcc_id)); + gfx_v9_4_3_xcc_mqd_init(ring, xcc_id); + soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id)); + mutex_unlock(&adev->srbm_mutex); + + if (adev->gfx.mec.mqd_backup[mqd_idx]) + memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(struct v9_mqd_allocation)); + } else { + /* restore MQD to a clean status */ + if (adev->gfx.mec.mqd_backup[mqd_idx]) + memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct v9_mqd_allocation)); + /* reset ring buffer */ + ring->wptr = 0; + atomic64_set((atomic64_t *)&adev->wb.wb[ring->wptr_offs], 0); + amdgpu_ring_clear_ring(ring); + } + + return 0; +} + +static int gfx_v9_4_3_xcc_kcq_fini_register(struct amdgpu_device *adev, int xcc_id) +{ + struct amdgpu_ring *ring; + int j; + + for (j = 0; j < adev->gfx.num_compute_rings; j++) { + ring = &adev->gfx.compute_ring[j + xcc_id * adev->gfx.num_compute_rings]; + if (!amdgpu_in_reset(adev) && !adev->in_suspend) { + mutex_lock(&adev->srbm_mutex); + soc15_grbm_select(adev, ring->me, + ring->pipe, + ring->queue, 0, GET_INST(GC, xcc_id)); + gfx_v9_4_3_xcc_q_fini_register(ring, xcc_id); + soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id)); + mutex_unlock(&adev->srbm_mutex); + } + } + + return 0; +} + +static int gfx_v9_4_3_xcc_kiq_resume(struct amdgpu_device *adev, int xcc_id) +{ + struct amdgpu_ring *ring; + int r; + + ring = &adev->gfx.kiq[xcc_id].ring; + + r = amdgpu_bo_reserve(ring->mqd_obj, false); + if (unlikely(r != 0)) + return r; + + r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr); + if (unlikely(r != 0)) { + amdgpu_bo_unreserve(ring->mqd_obj); + return r; + } + + gfx_v9_4_3_xcc_kiq_init_queue(ring, xcc_id); + amdgpu_bo_kunmap(ring->mqd_obj); + ring->mqd_ptr = NULL; + amdgpu_bo_unreserve(ring->mqd_obj); + return 0; +} + +static int gfx_v9_4_3_xcc_kcq_resume(struct amdgpu_device *adev, int xcc_id) +{ + struct amdgpu_ring *ring = NULL; + int r = 0, i; + + gfx_v9_4_3_xcc_cp_compute_enable(adev, true, xcc_id); + + for (i = 0; i < adev->gfx.num_compute_rings; i++) { + ring = &adev->gfx.compute_ring[i + xcc_id * adev->gfx.num_compute_rings]; + + r = amdgpu_bo_reserve(ring->mqd_obj, false); + if (unlikely(r != 0)) + goto done; + r = amdgpu_bo_kmap(ring->mqd_obj, (void **)&ring->mqd_ptr); + if (!r) { + r = gfx_v9_4_3_xcc_kcq_init_queue(ring, xcc_id); + amdgpu_bo_kunmap(ring->mqd_obj); + ring->mqd_ptr = NULL; + } + amdgpu_bo_unreserve(ring->mqd_obj); + if (r) + goto done; + } + + r = amdgpu_gfx_enable_kcq(adev, xcc_id); +done: + return r; +} + +static int gfx_v9_4_3_xcc_cp_resume(struct amdgpu_device *adev, int xcc_id) +{ + struct amdgpu_ring *ring; + int r, j; + + gfx_v9_4_3_xcc_enable_gui_idle_interrupt(adev, false, xcc_id); + + if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) { + gfx_v9_4_3_xcc_disable_gpa_mode(adev, xcc_id); + + r = gfx_v9_4_3_xcc_cp_compute_load_microcode(adev, xcc_id); + if (r) + return r; + } + + /* set the virtual and physical id based on partition_mode */ + gfx_v9_4_3_xcc_program_xcc_id(adev, xcc_id); + + r = gfx_v9_4_3_xcc_kiq_resume(adev, xcc_id); + if (r) + return r; + + r = gfx_v9_4_3_xcc_kcq_resume(adev, xcc_id); + if (r) + return r; + + for (j = 0; j < adev->gfx.num_compute_rings; j++) { + ring = &adev->gfx.compute_ring + [j + xcc_id * adev->gfx.num_compute_rings]; + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + } + + gfx_v9_4_3_xcc_enable_gui_idle_interrupt(adev, true, xcc_id); + + return 0; +} + +static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev) +{ + int r = 0, i, num_xcc; + + if (amdgpu_xcp_query_partition_mode(adev->xcp_mgr, + AMDGPU_XCP_FL_NONE) == + AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE) + r = amdgpu_xcp_switch_partition_mode(adev->xcp_mgr, + amdgpu_user_partt_mode); + + if (r) + return r; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + for (i = 0; i < num_xcc; i++) { + r = gfx_v9_4_3_xcc_cp_resume(adev, i); + if (r) + return r; + } + + return 0; +} + +static void gfx_v9_4_3_xcc_cp_enable(struct amdgpu_device *adev, bool enable, + int xcc_id) +{ + gfx_v9_4_3_xcc_cp_compute_enable(adev, enable, xcc_id); +} + +static void gfx_v9_4_3_xcc_fini(struct amdgpu_device *adev, int xcc_id) +{ + if (amdgpu_gfx_disable_kcq(adev, xcc_id)) + DRM_ERROR("XCD %d KCQ disable failed\n", xcc_id); + + /* Use deinitialize sequence from CAIL when unbinding device + * from driver, otherwise KIQ is hanging when binding back + */ + if (!amdgpu_in_reset(adev) && !adev->in_suspend) { + mutex_lock(&adev->srbm_mutex); + soc15_grbm_select(adev, adev->gfx.kiq[xcc_id].ring.me, + adev->gfx.kiq[xcc_id].ring.pipe, + adev->gfx.kiq[xcc_id].ring.queue, 0, + GET_INST(GC, xcc_id)); + gfx_v9_4_3_xcc_q_fini_register(&adev->gfx.kiq[xcc_id].ring, + xcc_id); + soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id)); + mutex_unlock(&adev->srbm_mutex); + } + + gfx_v9_4_3_xcc_kcq_fini_register(adev, xcc_id); + gfx_v9_4_3_xcc_cp_enable(adev, false, xcc_id); +} + +static int gfx_v9_4_3_hw_init(void *handle) +{ + int r; + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + gfx_v9_4_3_init_golden_registers(adev); + + gfx_v9_4_3_constants_init(adev); + + r = adev->gfx.rlc.funcs->resume(adev); + if (r) + return r; + + r = gfx_v9_4_3_cp_resume(adev); + if (r) + return r; + + return r; +} + +static int gfx_v9_4_3_hw_fini(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int i, num_xcc; + + amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0); + amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0); + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + for (i = 0; i < num_xcc; i++) { + gfx_v9_4_3_xcc_fini(adev, i); + } + + return 0; +} + +static int gfx_v9_4_3_suspend(void *handle) +{ + return gfx_v9_4_3_hw_fini(handle); +} + +static int gfx_v9_4_3_resume(void *handle) +{ + return gfx_v9_4_3_hw_init(handle); +} + +static bool gfx_v9_4_3_is_idle(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int i, num_xcc; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + for (i = 0; i < num_xcc; i++) { + if (REG_GET_FIELD(RREG32_SOC15(GC, GET_INST(GC, i), regGRBM_STATUS), + GRBM_STATUS, GUI_ACTIVE)) + return false; + } + return true; +} + +static int gfx_v9_4_3_wait_for_idle(void *handle) +{ + unsigned i; + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + for (i = 0; i < adev->usec_timeout; i++) { + if (gfx_v9_4_3_is_idle(handle)) + return 0; + udelay(1); + } + return -ETIMEDOUT; +} + +static int gfx_v9_4_3_soft_reset(void *handle) +{ + u32 grbm_soft_reset = 0; + u32 tmp; + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + /* GRBM_STATUS */ + tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_STATUS); + if (tmp & (GRBM_STATUS__PA_BUSY_MASK | GRBM_STATUS__SC_BUSY_MASK | + GRBM_STATUS__BCI_BUSY_MASK | GRBM_STATUS__SX_BUSY_MASK | + GRBM_STATUS__TA_BUSY_MASK | GRBM_STATUS__VGT_BUSY_MASK | + GRBM_STATUS__DB_BUSY_MASK | GRBM_STATUS__CB_BUSY_MASK | + GRBM_STATUS__GDS_BUSY_MASK | GRBM_STATUS__SPI_BUSY_MASK | + GRBM_STATUS__IA_BUSY_MASK | GRBM_STATUS__IA_BUSY_NO_DMA_MASK)) { + grbm_soft_reset = REG_SET_FIELD(grbm_soft_reset, + GRBM_SOFT_RESET, SOFT_RESET_CP, 1); + grbm_soft_reset = REG_SET_FIELD(grbm_soft_reset, + GRBM_SOFT_RESET, SOFT_RESET_GFX, 1); + } + + if (tmp & (GRBM_STATUS__CP_BUSY_MASK | GRBM_STATUS__CP_COHERENCY_BUSY_MASK)) { + grbm_soft_reset = REG_SET_FIELD(grbm_soft_reset, + GRBM_SOFT_RESET, SOFT_RESET_CP, 1); + } + + /* GRBM_STATUS2 */ + tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_STATUS2); + if (REG_GET_FIELD(tmp, GRBM_STATUS2, RLC_BUSY)) + grbm_soft_reset = REG_SET_FIELD(grbm_soft_reset, + GRBM_SOFT_RESET, SOFT_RESET_RLC, 1); + + + if (grbm_soft_reset) { + /* stop the rlc */ + adev->gfx.rlc.funcs->stop(adev); + + /* Disable MEC parsing/prefetching */ + gfx_v9_4_3_xcc_cp_compute_enable(adev, false, 0); + + if (grbm_soft_reset) { + tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET); + tmp |= grbm_soft_reset; + dev_info(adev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp); + WREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET, tmp); + tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET); + + udelay(50); + + tmp &= ~grbm_soft_reset; + WREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET, tmp); + tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET); + } + + /* Wait a little for things to settle down */ + udelay(50); + } + return 0; +} + +static void gfx_v9_4_3_ring_emit_gds_switch(struct amdgpu_ring *ring, + uint32_t vmid, + uint32_t gds_base, uint32_t gds_size, + uint32_t gws_base, uint32_t gws_size, + uint32_t oa_base, uint32_t oa_size) +{ + struct amdgpu_device *adev = ring->adev; + + /* GDS Base */ + gfx_v9_4_3_write_data_to_reg(ring, 0, false, + SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regGDS_VMID0_BASE) + 2 * vmid, + gds_base); + + /* GDS Size */ + gfx_v9_4_3_write_data_to_reg(ring, 0, false, + SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regGDS_VMID0_SIZE) + 2 * vmid, + gds_size); + + /* GWS */ + gfx_v9_4_3_write_data_to_reg(ring, 0, false, + SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regGDS_GWS_VMID0) + vmid, + gws_size << GDS_GWS_VMID0__SIZE__SHIFT | gws_base); + + /* OA */ + gfx_v9_4_3_write_data_to_reg(ring, 0, false, + SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regGDS_OA_VMID0) + vmid, + (1 << (oa_size + oa_base)) - (1 << oa_base)); +} + +static int gfx_v9_4_3_early_init(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev), + AMDGPU_MAX_COMPUTE_RINGS); + gfx_v9_4_3_set_kiq_pm4_funcs(adev); + gfx_v9_4_3_set_ring_funcs(adev); + gfx_v9_4_3_set_irq_funcs(adev); + gfx_v9_4_3_set_gds_init(adev); + gfx_v9_4_3_set_rlc_funcs(adev); + + return gfx_v9_4_3_init_microcode(adev); +} + +static int gfx_v9_4_3_late_init(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int r; + + r = amdgpu_irq_get(adev, &adev->gfx.priv_reg_irq, 0); + if (r) + return r; + + r = amdgpu_irq_get(adev, &adev->gfx.priv_inst_irq, 0); + if (r) + return r; + + return 0; +} + +static void gfx_v9_4_3_xcc_update_sram_fgcg(struct amdgpu_device *adev, + bool enable, int xcc_id) +{ + uint32_t def, data; + + if (!(adev->cg_flags & AMD_CG_SUPPORT_GFX_FGCG)) + return; + + def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), + regRLC_CGTT_MGCG_OVERRIDE); + + if (enable) + data &= ~RLC_CGTT_MGCG_OVERRIDE__GFXIP_FGCG_OVERRIDE_MASK; + else + data |= RLC_CGTT_MGCG_OVERRIDE__GFXIP_FGCG_OVERRIDE_MASK; + + if (def != data) + WREG32_SOC15(GC, GET_INST(GC, xcc_id), + regRLC_CGTT_MGCG_OVERRIDE, data); + + def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CLK_CNTL); + + if (enable) + data &= ~RLC_CLK_CNTL__RLC_SRAM_CLK_GATER_OVERRIDE_MASK; + else + data |= RLC_CLK_CNTL__RLC_SRAM_CLK_GATER_OVERRIDE_MASK; + + if (def != data) + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CLK_CNTL, data); +} + +static void gfx_v9_4_3_xcc_update_repeater_fgcg(struct amdgpu_device *adev, + bool enable, int xcc_id) +{ + uint32_t def, data; + + if (!(adev->cg_flags & AMD_CG_SUPPORT_REPEATER_FGCG)) + return; + + def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), + regRLC_CGTT_MGCG_OVERRIDE); + + if (enable) + data &= ~RLC_CGTT_MGCG_OVERRIDE__GFXIP_REP_FGCG_OVERRIDE_MASK; + else + data |= RLC_CGTT_MGCG_OVERRIDE__GFXIP_REP_FGCG_OVERRIDE_MASK; + + if (def != data) + WREG32_SOC15(GC, GET_INST(GC, xcc_id), + regRLC_CGTT_MGCG_OVERRIDE, data); +} + +static void +gfx_v9_4_3_xcc_update_medium_grain_clock_gating(struct amdgpu_device *adev, + bool enable, int xcc_id) +{ + uint32_t data, def; + + /* It is disabled by HW by default */ + if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGCG)) { + /* 1 - RLC_CGTT_MGCG_OVERRIDE */ + def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE); + + data &= ~(RLC_CGTT_MGCG_OVERRIDE__GRBM_CGTT_SCLK_OVERRIDE_MASK | + RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGCG_OVERRIDE_MASK | + RLC_CGTT_MGCG_OVERRIDE__RLC_CGTT_SCLK_OVERRIDE_MASK | + RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGLS_OVERRIDE_MASK); + + if (def != data) + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE, data); + + /* MGLS is a global flag to control all MGLS in GFX */ + if (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGLS) { + /* 2 - RLC memory Light sleep */ + if (adev->cg_flags & AMD_CG_SUPPORT_GFX_RLC_LS) { + def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_MEM_SLP_CNTL); + data |= RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK; + if (def != data) + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_MEM_SLP_CNTL, data); + } + /* 3 - CP memory Light sleep */ + if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CP_LS) { + def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEM_SLP_CNTL); + data |= CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK; + if (def != data) + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEM_SLP_CNTL, data); + } + } + } else { + /* 1 - MGCG_OVERRIDE */ + def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE); + + data |= (RLC_CGTT_MGCG_OVERRIDE__RLC_CGTT_SCLK_OVERRIDE_MASK | + RLC_CGTT_MGCG_OVERRIDE__GRBM_CGTT_SCLK_OVERRIDE_MASK | + RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGCG_OVERRIDE_MASK | + RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGLS_OVERRIDE_MASK); + + if (def != data) + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE, data); + + /* 2 - disable MGLS in RLC */ + data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_MEM_SLP_CNTL); + if (data & RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK) { + data &= ~RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK; + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_MEM_SLP_CNTL, data); + } + + /* 3 - disable MGLS in CP */ + data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEM_SLP_CNTL); + if (data & CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK) { + data &= ~CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK; + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_MEM_SLP_CNTL, data); + } + } + +} + +static void +gfx_v9_4_3_xcc_update_coarse_grain_clock_gating(struct amdgpu_device *adev, + bool enable, int xcc_id) +{ + uint32_t def, data; + + if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGCG)) { + + def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE); + /* unset CGCG override */ + data &= ~RLC_CGTT_MGCG_OVERRIDE__GFXIP_CGCG_OVERRIDE_MASK; + if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGLS) + data &= ~RLC_CGTT_MGCG_OVERRIDE__GFXIP_CGLS_OVERRIDE_MASK; + else + data |= RLC_CGTT_MGCG_OVERRIDE__GFXIP_CGLS_OVERRIDE_MASK; + /* update CGCG and CGLS override bits */ + if (def != data) + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE, data); + + /* enable cgcg FSM(0x0000363F) */ + def = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL); + + data = (0x36 + << RLC_CGCG_CGLS_CTRL__CGCG_GFX_IDLE_THRESHOLD__SHIFT) | + RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK; + if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGLS) + data |= (0x000F << RLC_CGCG_CGLS_CTRL__CGLS_REP_COMPANSAT_DELAY__SHIFT) | + RLC_CGCG_CGLS_CTRL__CGLS_EN_MASK; + if (def != data) + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL, data); + + /* set IDLE_POLL_COUNT(0x00900100) */ + def = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_RB_WPTR_POLL_CNTL); + data = (0x0100 << CP_RB_WPTR_POLL_CNTL__POLL_FREQUENCY__SHIFT) | + (0x0090 << CP_RB_WPTR_POLL_CNTL__IDLE_POLL_COUNT__SHIFT); + if (def != data) + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_RB_WPTR_POLL_CNTL, data); + } else { + def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL); + /* reset CGCG/CGLS bits */ + data &= ~(RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK | RLC_CGCG_CGLS_CTRL__CGLS_EN_MASK); + /* disable cgcg and cgls in FSM */ + if (def != data) + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGCG_CGLS_CTRL, data); + } + +} + +static int gfx_v9_4_3_xcc_update_gfx_clock_gating(struct amdgpu_device *adev, + bool enable, int xcc_id) +{ + amdgpu_gfx_rlc_enter_safe_mode(adev, xcc_id); + + if (enable) { + /* FGCG */ + gfx_v9_4_3_xcc_update_sram_fgcg(adev, enable, xcc_id); + gfx_v9_4_3_xcc_update_repeater_fgcg(adev, enable, xcc_id); + + /* CGCG/CGLS should be enabled after MGCG/MGLS + * === MGCG + MGLS === + */ + gfx_v9_4_3_xcc_update_medium_grain_clock_gating(adev, enable, + xcc_id); + /* === CGCG + CGLS === */ + gfx_v9_4_3_xcc_update_coarse_grain_clock_gating(adev, enable, + xcc_id); + } else { + /* CGCG/CGLS should be disabled before MGCG/MGLS + * === CGCG + CGLS === + */ + gfx_v9_4_3_xcc_update_coarse_grain_clock_gating(adev, enable, + xcc_id); + /* === MGCG + MGLS === */ + gfx_v9_4_3_xcc_update_medium_grain_clock_gating(adev, enable, + xcc_id); + + /* FGCG */ + gfx_v9_4_3_xcc_update_sram_fgcg(adev, enable, xcc_id); + gfx_v9_4_3_xcc_update_repeater_fgcg(adev, enable, xcc_id); + } + + amdgpu_gfx_rlc_exit_safe_mode(adev, xcc_id); + + return 0; +} + +static const struct amdgpu_rlc_funcs gfx_v9_4_3_rlc_funcs = { + .is_rlc_enabled = gfx_v9_4_3_is_rlc_enabled, + .set_safe_mode = gfx_v9_4_3_xcc_set_safe_mode, + .unset_safe_mode = gfx_v9_4_3_xcc_unset_safe_mode, + .init = gfx_v9_4_3_rlc_init, + .resume = gfx_v9_4_3_rlc_resume, + .stop = gfx_v9_4_3_rlc_stop, + .reset = gfx_v9_4_3_rlc_reset, + .start = gfx_v9_4_3_rlc_start, + .update_spm_vmid = gfx_v9_4_3_update_spm_vmid, + .is_rlcg_access_range = gfx_v9_4_3_is_rlcg_access_range, +}; + +static int gfx_v9_4_3_set_powergating_state(void *handle, + enum amd_powergating_state state) +{ + return 0; +} + +static int gfx_v9_4_3_set_clockgating_state(void *handle, + enum amd_clockgating_state state) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int i, num_xcc; + + if (amdgpu_sriov_vf(adev)) + return 0; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + switch (adev->ip_versions[GC_HWIP][0]) { + case IP_VERSION(9, 4, 3): + for (i = 0; i < num_xcc; i++) + gfx_v9_4_3_xcc_update_gfx_clock_gating( + adev, state == AMD_CG_STATE_GATE, i); + break; + default: + break; + } + return 0; +} + +static void gfx_v9_4_3_get_clockgating_state(void *handle, u64 *flags) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int data; + + if (amdgpu_sriov_vf(adev)) + *flags = 0; + + /* AMD_CG_SUPPORT_GFX_MGCG */ + data = RREG32_KIQ(SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regRLC_CGTT_MGCG_OVERRIDE)); + if (!(data & RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGCG_OVERRIDE_MASK)) + *flags |= AMD_CG_SUPPORT_GFX_MGCG; + + /* AMD_CG_SUPPORT_GFX_CGCG */ + data = RREG32_KIQ(SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regRLC_CGCG_CGLS_CTRL)); + if (data & RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK) + *flags |= AMD_CG_SUPPORT_GFX_CGCG; + + /* AMD_CG_SUPPORT_GFX_CGLS */ + if (data & RLC_CGCG_CGLS_CTRL__CGLS_EN_MASK) + *flags |= AMD_CG_SUPPORT_GFX_CGLS; + + /* AMD_CG_SUPPORT_GFX_RLC_LS */ + data = RREG32_KIQ(SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regRLC_MEM_SLP_CNTL)); + if (data & RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK) + *flags |= AMD_CG_SUPPORT_GFX_RLC_LS | AMD_CG_SUPPORT_GFX_MGLS; + + /* AMD_CG_SUPPORT_GFX_CP_LS */ + data = RREG32_KIQ(SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regCP_MEM_SLP_CNTL)); + if (data & CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK) + *flags |= AMD_CG_SUPPORT_GFX_CP_LS | AMD_CG_SUPPORT_GFX_MGLS; +} + +static void gfx_v9_4_3_ring_emit_hdp_flush(struct amdgpu_ring *ring) +{ + struct amdgpu_device *adev = ring->adev; + u32 ref_and_mask, reg_mem_engine; + const struct nbio_hdp_flush_reg *nbio_hf_reg = adev->nbio.hdp_flush_reg; + + if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) { + switch (ring->me) { + case 1: + ref_and_mask = nbio_hf_reg->ref_and_mask_cp2 << ring->pipe; + break; + case 2: + ref_and_mask = nbio_hf_reg->ref_and_mask_cp6 << ring->pipe; + break; + default: + return; + } + reg_mem_engine = 0; + } else { + ref_and_mask = nbio_hf_reg->ref_and_mask_cp0; + reg_mem_engine = 1; /* pfp */ + } + + gfx_v9_4_3_wait_reg_mem(ring, reg_mem_engine, 0, 1, + adev->nbio.funcs->get_hdp_flush_req_offset(adev), + adev->nbio.funcs->get_hdp_flush_done_offset(adev), + ref_and_mask, ref_and_mask, 0x20); +} + +static void gfx_v9_4_3_ring_emit_ib_compute(struct amdgpu_ring *ring, + struct amdgpu_job *job, + struct amdgpu_ib *ib, + uint32_t flags) +{ + unsigned vmid = AMDGPU_JOB_GET_VMID(job); + u32 control = INDIRECT_BUFFER_VALID | ib->length_dw | (vmid << 24); + + /* Currently, there is a high possibility to get wave ID mismatch + * between ME and GDS, leading to a hw deadlock, because ME generates + * different wave IDs than the GDS expects. This situation happens + * randomly when at least 5 compute pipes use GDS ordered append. + * The wave IDs generated by ME are also wrong after suspend/resume. + * Those are probably bugs somewhere else in the kernel driver. + * + * Writing GDS_COMPUTE_MAX_WAVE_ID resets wave ID counters in ME and + * GDS to 0 for this ring (me/pipe). + */ + if (ib->flags & AMDGPU_IB_FLAG_RESET_GDS_MAX_WAVE_ID) { + amdgpu_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1)); + amdgpu_ring_write(ring, regGDS_COMPUTE_MAX_WAVE_ID); + amdgpu_ring_write(ring, ring->adev->gds.gds_compute_max_wave_id); + } + + amdgpu_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2)); + BUG_ON(ib->gpu_addr & 0x3); /* Dword align */ + amdgpu_ring_write(ring, +#ifdef __BIG_ENDIAN + (2 << 0) | +#endif + lower_32_bits(ib->gpu_addr)); + amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr)); + amdgpu_ring_write(ring, control); +} + +static void gfx_v9_4_3_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, + u64 seq, unsigned flags) +{ + bool write64bit = flags & AMDGPU_FENCE_FLAG_64BIT; + bool int_sel = flags & AMDGPU_FENCE_FLAG_INT; + bool writeback = flags & AMDGPU_FENCE_FLAG_TC_WB_ONLY; + + /* RELEASE_MEM - flush caches, send int */ + amdgpu_ring_write(ring, PACKET3(PACKET3_RELEASE_MEM, 6)); + amdgpu_ring_write(ring, ((writeback ? (EOP_TC_WB_ACTION_EN | + EOP_TC_NC_ACTION_EN) : + (EOP_TCL1_ACTION_EN | + EOP_TC_ACTION_EN | + EOP_TC_WB_ACTION_EN | + EOP_TC_MD_ACTION_EN)) | + EVENT_TYPE(CACHE_FLUSH_AND_INV_TS_EVENT) | + EVENT_INDEX(5))); + amdgpu_ring_write(ring, DATA_SEL(write64bit ? 2 : 1) | INT_SEL(int_sel ? 2 : 0)); + + /* + * the address should be Qword aligned if 64bit write, Dword + * aligned if only send 32bit data low (discard data high) + */ + if (write64bit) + BUG_ON(addr & 0x7); + else + BUG_ON(addr & 0x3); + amdgpu_ring_write(ring, lower_32_bits(addr)); + amdgpu_ring_write(ring, upper_32_bits(addr)); + amdgpu_ring_write(ring, lower_32_bits(seq)); + amdgpu_ring_write(ring, upper_32_bits(seq)); + amdgpu_ring_write(ring, 0); +} + +static void gfx_v9_4_3_ring_emit_pipeline_sync(struct amdgpu_ring *ring) +{ + int usepfp = (ring->funcs->type == AMDGPU_RING_TYPE_GFX); + uint32_t seq = ring->fence_drv.sync_seq; + uint64_t addr = ring->fence_drv.gpu_addr; + + gfx_v9_4_3_wait_reg_mem(ring, usepfp, 1, 0, + lower_32_bits(addr), upper_32_bits(addr), + seq, 0xffffffff, 4); +} + +static void gfx_v9_4_3_ring_emit_vm_flush(struct amdgpu_ring *ring, + unsigned vmid, uint64_t pd_addr) +{ + amdgpu_gmc_emit_flush_gpu_tlb(ring, vmid, pd_addr); +} + +static u64 gfx_v9_4_3_ring_get_rptr_compute(struct amdgpu_ring *ring) +{ + return ring->adev->wb.wb[ring->rptr_offs]; /* gfx9 hardware is 32bit rptr */ +} + +static u64 gfx_v9_4_3_ring_get_wptr_compute(struct amdgpu_ring *ring) +{ + u64 wptr; + + /* XXX check if swapping is necessary on BE */ + if (ring->use_doorbell) + wptr = atomic64_read((atomic64_t *)&ring->adev->wb.wb[ring->wptr_offs]); + else + BUG(); + return wptr; +} + +static void gfx_v9_4_3_ring_set_wptr_compute(struct amdgpu_ring *ring) +{ + struct amdgpu_device *adev = ring->adev; + + /* XXX check if swapping is necessary on BE */ + if (ring->use_doorbell) { + atomic64_set((atomic64_t *)&adev->wb.wb[ring->wptr_offs], ring->wptr); + WDOORBELL64(ring->doorbell_index, ring->wptr); + } else { + BUG(); /* only DOORBELL method supported on gfx9 now */ + } +} + +static void gfx_v9_4_3_ring_emit_fence_kiq(struct amdgpu_ring *ring, u64 addr, + u64 seq, unsigned int flags) +{ + struct amdgpu_device *adev = ring->adev; + + /* we only allocate 32bit for each seq wb address */ + BUG_ON(flags & AMDGPU_FENCE_FLAG_64BIT); + + /* write fence seq to the "addr" */ + amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); + amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) | + WRITE_DATA_DST_SEL(5) | WR_CONFIRM)); + amdgpu_ring_write(ring, lower_32_bits(addr)); + amdgpu_ring_write(ring, upper_32_bits(addr)); + amdgpu_ring_write(ring, lower_32_bits(seq)); + + if (flags & AMDGPU_FENCE_FLAG_INT) { + /* set register to trigger INT */ + amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); + amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) | + WRITE_DATA_DST_SEL(0) | WR_CONFIRM)); + amdgpu_ring_write(ring, SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regCPC_INT_STATUS)); + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, 0x20000000); /* src_id is 178 */ + } +} + +static void gfx_v9_4_3_ring_emit_rreg(struct amdgpu_ring *ring, uint32_t reg, + uint32_t reg_val_offs) +{ + struct amdgpu_device *adev = ring->adev; + + amdgpu_ring_write(ring, PACKET3(PACKET3_COPY_DATA, 4)); + amdgpu_ring_write(ring, 0 | /* src: register*/ + (5 << 8) | /* dst: memory */ + (1 << 20)); /* write confirm */ + amdgpu_ring_write(ring, reg); + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, lower_32_bits(adev->wb.gpu_addr + + reg_val_offs * 4)); + amdgpu_ring_write(ring, upper_32_bits(adev->wb.gpu_addr + + reg_val_offs * 4)); +} + +static void gfx_v9_4_3_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg, + uint32_t val) +{ + uint32_t cmd = 0; + + switch (ring->funcs->type) { + case AMDGPU_RING_TYPE_GFX: + cmd = WRITE_DATA_ENGINE_SEL(1) | WR_CONFIRM; + break; + case AMDGPU_RING_TYPE_KIQ: + cmd = (1 << 16); /* no inc addr */ + break; + default: + cmd = WR_CONFIRM; + break; + } + amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); + amdgpu_ring_write(ring, cmd); + amdgpu_ring_write(ring, reg); + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, val); +} + +static void gfx_v9_4_3_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg, + uint32_t val, uint32_t mask) +{ + gfx_v9_4_3_wait_reg_mem(ring, 0, 0, 0, reg, 0, val, mask, 0x20); +} + +static void gfx_v9_4_3_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring, + uint32_t reg0, uint32_t reg1, + uint32_t ref, uint32_t mask) +{ + amdgpu_ring_emit_reg_write_reg_wait_helper(ring, reg0, reg1, + ref, mask); +} + +static void gfx_v9_4_3_xcc_set_compute_eop_interrupt_state( + struct amdgpu_device *adev, int me, int pipe, + enum amdgpu_interrupt_state state, int xcc_id) +{ + u32 mec_int_cntl, mec_int_cntl_reg; + + /* + * amdgpu controls only the first MEC. That's why this function only + * handles the setting of interrupts for this specific MEC. All other + * pipes' interrupts are set by amdkfd. + */ + + if (me == 1) { + switch (pipe) { + case 0: + mec_int_cntl_reg = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regCP_ME1_PIPE0_INT_CNTL); + break; + case 1: + mec_int_cntl_reg = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regCP_ME1_PIPE1_INT_CNTL); + break; + case 2: + mec_int_cntl_reg = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regCP_ME1_PIPE2_INT_CNTL); + break; + case 3: + mec_int_cntl_reg = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regCP_ME1_PIPE3_INT_CNTL); + break; + default: + DRM_DEBUG("invalid pipe %d\n", pipe); + return; + } + } else { + DRM_DEBUG("invalid me %d\n", me); + return; + } + + switch (state) { + case AMDGPU_IRQ_STATE_DISABLE: + mec_int_cntl = RREG32(mec_int_cntl_reg); + mec_int_cntl = REG_SET_FIELD(mec_int_cntl, CP_ME1_PIPE0_INT_CNTL, + TIME_STAMP_INT_ENABLE, 0); + WREG32(mec_int_cntl_reg, mec_int_cntl); + break; + case AMDGPU_IRQ_STATE_ENABLE: + mec_int_cntl = RREG32(mec_int_cntl_reg); + mec_int_cntl = REG_SET_FIELD(mec_int_cntl, CP_ME1_PIPE0_INT_CNTL, + TIME_STAMP_INT_ENABLE, 1); + WREG32(mec_int_cntl_reg, mec_int_cntl); + break; + default: + break; + } +} + +static int gfx_v9_4_3_set_priv_reg_fault_state(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + unsigned type, + enum amdgpu_interrupt_state state) +{ + int i, num_xcc; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + switch (state) { + case AMDGPU_IRQ_STATE_DISABLE: + case AMDGPU_IRQ_STATE_ENABLE: + for (i = 0; i < num_xcc; i++) + WREG32_FIELD15_PREREG(GC, GET_INST(GC, i), CP_INT_CNTL_RING0, + PRIV_REG_INT_ENABLE, + state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0); + break; + default: + break; + } + + return 0; +} + +static int gfx_v9_4_3_set_priv_inst_fault_state(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + unsigned type, + enum amdgpu_interrupt_state state) +{ + int i, num_xcc; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + switch (state) { + case AMDGPU_IRQ_STATE_DISABLE: + case AMDGPU_IRQ_STATE_ENABLE: + for (i = 0; i < num_xcc; i++) + WREG32_FIELD15_PREREG(GC, GET_INST(GC, i), CP_INT_CNTL_RING0, + PRIV_INSTR_INT_ENABLE, + state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0); + break; + default: + break; + } + + return 0; +} + +static int gfx_v9_4_3_set_eop_interrupt_state(struct amdgpu_device *adev, + struct amdgpu_irq_src *src, + unsigned type, + enum amdgpu_interrupt_state state) +{ + int i, num_xcc; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + for (i = 0; i < num_xcc; i++) { + switch (type) { + case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP: + gfx_v9_4_3_xcc_set_compute_eop_interrupt_state( + adev, 1, 0, state, i); + break; + case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE1_EOP: + gfx_v9_4_3_xcc_set_compute_eop_interrupt_state( + adev, 1, 1, state, i); + break; + case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE2_EOP: + gfx_v9_4_3_xcc_set_compute_eop_interrupt_state( + adev, 1, 2, state, i); + break; + case AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE3_EOP: + gfx_v9_4_3_xcc_set_compute_eop_interrupt_state( + adev, 1, 3, state, i); + break; + case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE0_EOP: + gfx_v9_4_3_xcc_set_compute_eop_interrupt_state( + adev, 2, 0, state, i); + break; + case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE1_EOP: + gfx_v9_4_3_xcc_set_compute_eop_interrupt_state( + adev, 2, 1, state, i); + break; + case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE2_EOP: + gfx_v9_4_3_xcc_set_compute_eop_interrupt_state( + adev, 2, 2, state, i); + break; + case AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE3_EOP: + gfx_v9_4_3_xcc_set_compute_eop_interrupt_state( + adev, 2, 3, state, i); + break; + default: + break; + } + } + + return 0; +} + +static int gfx_v9_4_3_eop_irq(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + struct amdgpu_iv_entry *entry) +{ + int i, xcc_id; + u8 me_id, pipe_id, queue_id; + struct amdgpu_ring *ring; + + DRM_DEBUG("IH: CP EOP\n"); + me_id = (entry->ring_id & 0x0c) >> 2; + pipe_id = (entry->ring_id & 0x03) >> 0; + queue_id = (entry->ring_id & 0x70) >> 4; + + xcc_id = gfx_v9_4_3_ih_to_xcc_inst(adev, entry->node_id); + + if (xcc_id == -EINVAL) + return -EINVAL; + + switch (me_id) { + case 0: + case 1: + case 2: + for (i = 0; i < adev->gfx.num_compute_rings; i++) { + ring = &adev->gfx.compute_ring + [i + + xcc_id * adev->gfx.num_compute_rings]; + /* Per-queue interrupt is supported for MEC starting from VI. + * The interrupt can only be enabled/disabled per pipe instead of per queue. + */ + + if ((ring->me == me_id) && (ring->pipe == pipe_id) && (ring->queue == queue_id)) + amdgpu_fence_process(ring); + } + break; + } + return 0; +} + +static void gfx_v9_4_3_fault(struct amdgpu_device *adev, + struct amdgpu_iv_entry *entry) +{ + u8 me_id, pipe_id, queue_id; + struct amdgpu_ring *ring; + int i, xcc_id; + + me_id = (entry->ring_id & 0x0c) >> 2; + pipe_id = (entry->ring_id & 0x03) >> 0; + queue_id = (entry->ring_id & 0x70) >> 4; + + xcc_id = gfx_v9_4_3_ih_to_xcc_inst(adev, entry->node_id); + + if (xcc_id == -EINVAL) + return; + + switch (me_id) { + case 0: + case 1: + case 2: + for (i = 0; i < adev->gfx.num_compute_rings; i++) { + ring = &adev->gfx.compute_ring + [i + + xcc_id * adev->gfx.num_compute_rings]; + if (ring->me == me_id && ring->pipe == pipe_id && + ring->queue == queue_id) + drm_sched_fault(&ring->sched); + } + break; + } +} + +static int gfx_v9_4_3_priv_reg_irq(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + struct amdgpu_iv_entry *entry) +{ + DRM_ERROR("Illegal register access in command stream\n"); + gfx_v9_4_3_fault(adev, entry); + return 0; +} + +static int gfx_v9_4_3_priv_inst_irq(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + struct amdgpu_iv_entry *entry) +{ + DRM_ERROR("Illegal instruction in command stream\n"); + gfx_v9_4_3_fault(adev, entry); + return 0; +} + +static void gfx_v9_4_3_emit_mem_sync(struct amdgpu_ring *ring) +{ + const unsigned int cp_coher_cntl = + PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_SH_ICACHE_ACTION_ENA(1) | + PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_SH_KCACHE_ACTION_ENA(1) | + PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_TC_ACTION_ENA(1) | + PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_TCL1_ACTION_ENA(1) | + PACKET3_ACQUIRE_MEM_CP_COHER_CNTL_TC_WB_ACTION_ENA(1); + + /* ACQUIRE_MEM -make one or more surfaces valid for use by the subsequent operations */ + amdgpu_ring_write(ring, PACKET3(PACKET3_ACQUIRE_MEM, 5)); + amdgpu_ring_write(ring, cp_coher_cntl); /* CP_COHER_CNTL */ + amdgpu_ring_write(ring, 0xffffffff); /* CP_COHER_SIZE */ + amdgpu_ring_write(ring, 0xffffff); /* CP_COHER_SIZE_HI */ + amdgpu_ring_write(ring, 0); /* CP_COHER_BASE */ + amdgpu_ring_write(ring, 0); /* CP_COHER_BASE_HI */ + amdgpu_ring_write(ring, 0x0000000A); /* POLL_INTERVAL */ +} + +static void gfx_v9_4_3_emit_wave_limit_cs(struct amdgpu_ring *ring, + uint32_t pipe, bool enable) +{ + struct amdgpu_device *adev = ring->adev; + uint32_t val; + uint32_t wcl_cs_reg; + + /* regSPI_WCL_PIPE_PERCENT_CS[0-7]_DEFAULT values are same */ + val = enable ? 0x1 : 0x7f; + + switch (pipe) { + case 0: + wcl_cs_reg = SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regSPI_WCL_PIPE_PERCENT_CS0); + break; + case 1: + wcl_cs_reg = SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regSPI_WCL_PIPE_PERCENT_CS1); + break; + case 2: + wcl_cs_reg = SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regSPI_WCL_PIPE_PERCENT_CS2); + break; + case 3: + wcl_cs_reg = SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regSPI_WCL_PIPE_PERCENT_CS3); + break; + default: + DRM_DEBUG("invalid pipe %d\n", pipe); + return; + } + + amdgpu_ring_emit_wreg(ring, wcl_cs_reg, val); + +} +static void gfx_v9_4_3_emit_wave_limit(struct amdgpu_ring *ring, bool enable) +{ + struct amdgpu_device *adev = ring->adev; + uint32_t val; + int i; + + /* regSPI_WCL_PIPE_PERCENT_GFX is 7 bit multiplier register to limit + * number of gfx waves. Setting 5 bit will make sure gfx only gets + * around 25% of gpu resources. + */ + val = enable ? 0x1f : 0x07ffffff; + amdgpu_ring_emit_wreg(ring, + SOC15_REG_OFFSET(GC, GET_INST(GC, 0), regSPI_WCL_PIPE_PERCENT_GFX), + val); + + /* Restrict waves for normal/low priority compute queues as well + * to get best QoS for high priority compute jobs. + * + * amdgpu controls only 1st ME(0-3 CS pipes). + */ + for (i = 0; i < adev->gfx.mec.num_pipe_per_mec; i++) { + if (i != ring->pipe) + gfx_v9_4_3_emit_wave_limit_cs(ring, i, enable); + + } +} + +enum amdgpu_gfx_cp_ras_mem_id { + AMDGPU_GFX_CP_MEM1 = 1, + AMDGPU_GFX_CP_MEM2, + AMDGPU_GFX_CP_MEM3, + AMDGPU_GFX_CP_MEM4, + AMDGPU_GFX_CP_MEM5, +}; + +enum amdgpu_gfx_gcea_ras_mem_id { + AMDGPU_GFX_GCEA_IOWR_CMDMEM = 4, + AMDGPU_GFX_GCEA_IORD_CMDMEM, + AMDGPU_GFX_GCEA_GMIWR_CMDMEM, + AMDGPU_GFX_GCEA_GMIRD_CMDMEM, + AMDGPU_GFX_GCEA_DRAMWR_CMDMEM, + AMDGPU_GFX_GCEA_DRAMRD_CMDMEM, + AMDGPU_GFX_GCEA_MAM_DMEM0, + AMDGPU_GFX_GCEA_MAM_DMEM1, + AMDGPU_GFX_GCEA_MAM_DMEM2, + AMDGPU_GFX_GCEA_MAM_DMEM3, + AMDGPU_GFX_GCEA_MAM_AMEM0, + AMDGPU_GFX_GCEA_MAM_AMEM1, + AMDGPU_GFX_GCEA_MAM_AMEM2, + AMDGPU_GFX_GCEA_MAM_AMEM3, + AMDGPU_GFX_GCEA_MAM_AFLUSH_BUFFER, + AMDGPU_GFX_GCEA_WRET_TAGMEM, + AMDGPU_GFX_GCEA_RRET_TAGMEM, + AMDGPU_GFX_GCEA_IOWR_DATAMEM, + AMDGPU_GFX_GCEA_GMIWR_DATAMEM, + AMDGPU_GFX_GCEA_DRAM_DATAMEM, +}; + +enum amdgpu_gfx_gc_cane_ras_mem_id { + AMDGPU_GFX_GC_CANE_MEM0 = 0, +}; + +enum amdgpu_gfx_gcutcl2_ras_mem_id { + AMDGPU_GFX_GCUTCL2_MEM2P512X95 = 160, +}; + +enum amdgpu_gfx_gds_ras_mem_id { + AMDGPU_GFX_GDS_MEM0 = 0, +}; + +enum amdgpu_gfx_lds_ras_mem_id { + AMDGPU_GFX_LDS_BANK0 = 0, + AMDGPU_GFX_LDS_BANK1, + AMDGPU_GFX_LDS_BANK2, + AMDGPU_GFX_LDS_BANK3, + AMDGPU_GFX_LDS_BANK4, + AMDGPU_GFX_LDS_BANK5, + AMDGPU_GFX_LDS_BANK6, + AMDGPU_GFX_LDS_BANK7, + AMDGPU_GFX_LDS_BANK8, + AMDGPU_GFX_LDS_BANK9, + AMDGPU_GFX_LDS_BANK10, + AMDGPU_GFX_LDS_BANK11, + AMDGPU_GFX_LDS_BANK12, + AMDGPU_GFX_LDS_BANK13, + AMDGPU_GFX_LDS_BANK14, + AMDGPU_GFX_LDS_BANK15, + AMDGPU_GFX_LDS_BANK16, + AMDGPU_GFX_LDS_BANK17, + AMDGPU_GFX_LDS_BANK18, + AMDGPU_GFX_LDS_BANK19, + AMDGPU_GFX_LDS_BANK20, + AMDGPU_GFX_LDS_BANK21, + AMDGPU_GFX_LDS_BANK22, + AMDGPU_GFX_LDS_BANK23, + AMDGPU_GFX_LDS_BANK24, + AMDGPU_GFX_LDS_BANK25, + AMDGPU_GFX_LDS_BANK26, + AMDGPU_GFX_LDS_BANK27, + AMDGPU_GFX_LDS_BANK28, + AMDGPU_GFX_LDS_BANK29, + AMDGPU_GFX_LDS_BANK30, + AMDGPU_GFX_LDS_BANK31, + AMDGPU_GFX_LDS_SP_BUFFER_A, + AMDGPU_GFX_LDS_SP_BUFFER_B, +}; + +enum amdgpu_gfx_rlc_ras_mem_id { + AMDGPU_GFX_RLC_GPMF32 = 1, + AMDGPU_GFX_RLC_RLCVF32, + AMDGPU_GFX_RLC_SCRATCH, + AMDGPU_GFX_RLC_SRM_ARAM, + AMDGPU_GFX_RLC_SRM_DRAM, + AMDGPU_GFX_RLC_TCTAG, + AMDGPU_GFX_RLC_SPM_SE, + AMDGPU_GFX_RLC_SPM_GRBMT, +}; + +enum amdgpu_gfx_sp_ras_mem_id { + AMDGPU_GFX_SP_SIMDID0 = 0, +}; + +enum amdgpu_gfx_spi_ras_mem_id { + AMDGPU_GFX_SPI_MEM0 = 0, + AMDGPU_GFX_SPI_MEM1, + AMDGPU_GFX_SPI_MEM2, + AMDGPU_GFX_SPI_MEM3, +}; + +enum amdgpu_gfx_sqc_ras_mem_id { + AMDGPU_GFX_SQC_INST_CACHE_A = 100, + AMDGPU_GFX_SQC_INST_CACHE_B = 101, + AMDGPU_GFX_SQC_INST_CACHE_TAG_A = 102, + AMDGPU_GFX_SQC_INST_CACHE_TAG_B = 103, + AMDGPU_GFX_SQC_INST_CACHE_MISS_FIFO_A = 104, + AMDGPU_GFX_SQC_INST_CACHE_MISS_FIFO_B = 105, + AMDGPU_GFX_SQC_INST_CACHE_GATCL1_MISS_FIFO_A = 106, + AMDGPU_GFX_SQC_INST_CACHE_GATCL1_MISS_FIFO_B = 107, + AMDGPU_GFX_SQC_DATA_CACHE_A = 200, + AMDGPU_GFX_SQC_DATA_CACHE_B = 201, + AMDGPU_GFX_SQC_DATA_CACHE_TAG_A = 202, + AMDGPU_GFX_SQC_DATA_CACHE_TAG_B = 203, + AMDGPU_GFX_SQC_DATA_CACHE_MISS_FIFO_A = 204, + AMDGPU_GFX_SQC_DATA_CACHE_MISS_FIFO_B = 205, + AMDGPU_GFX_SQC_DATA_CACHE_HIT_FIFO_A = 206, + AMDGPU_GFX_SQC_DATA_CACHE_HIT_FIFO_B = 207, + AMDGPU_GFX_SQC_DIRTY_BIT_A = 208, + AMDGPU_GFX_SQC_DIRTY_BIT_B = 209, + AMDGPU_GFX_SQC_WRITE_DATA_BUFFER_CU0 = 210, + AMDGPU_GFX_SQC_WRITE_DATA_BUFFER_CU1 = 211, + AMDGPU_GFX_SQC_UTCL1_MISS_LFIFO_DATA_CACHE_A = 212, + AMDGPU_GFX_SQC_UTCL1_MISS_LFIFO_DATA_CACHE_B = 213, + AMDGPU_GFX_SQC_UTCL1_MISS_LFIFO_INST_CACHE = 108, +}; + +enum amdgpu_gfx_sq_ras_mem_id { + AMDGPU_GFX_SQ_SGPR_MEM0 = 0, + AMDGPU_GFX_SQ_SGPR_MEM1, + AMDGPU_GFX_SQ_SGPR_MEM2, + AMDGPU_GFX_SQ_SGPR_MEM3, +}; + +enum amdgpu_gfx_ta_ras_mem_id { + AMDGPU_GFX_TA_FS_AFIFO_RAM_LO = 1, + AMDGPU_GFX_TA_FS_AFIFO_RAM_HI, + AMDGPU_GFX_TA_FS_CFIFO_RAM, + AMDGPU_GFX_TA_FSX_LFIFO, + AMDGPU_GFX_TA_FS_DFIFO_RAM, +}; + +enum amdgpu_gfx_tcc_ras_mem_id { + AMDGPU_GFX_TCC_MEM1 = 1, +}; + +enum amdgpu_gfx_tca_ras_mem_id { + AMDGPU_GFX_TCA_MEM1 = 1, +}; + +enum amdgpu_gfx_tci_ras_mem_id { + AMDGPU_GFX_TCIW_MEM = 1, +}; + +enum amdgpu_gfx_tcp_ras_mem_id { + AMDGPU_GFX_TCP_LFIFO0 = 1, + AMDGPU_GFX_TCP_SET0BANK0_RAM, + AMDGPU_GFX_TCP_SET0BANK1_RAM, + AMDGPU_GFX_TCP_SET0BANK2_RAM, + AMDGPU_GFX_TCP_SET0BANK3_RAM, + AMDGPU_GFX_TCP_SET1BANK0_RAM, + AMDGPU_GFX_TCP_SET1BANK1_RAM, + AMDGPU_GFX_TCP_SET1BANK2_RAM, + AMDGPU_GFX_TCP_SET1BANK3_RAM, + AMDGPU_GFX_TCP_SET2BANK0_RAM, + AMDGPU_GFX_TCP_SET2BANK1_RAM, + AMDGPU_GFX_TCP_SET2BANK2_RAM, + AMDGPU_GFX_TCP_SET2BANK3_RAM, + AMDGPU_GFX_TCP_SET3BANK0_RAM, + AMDGPU_GFX_TCP_SET3BANK1_RAM, + AMDGPU_GFX_TCP_SET3BANK2_RAM, + AMDGPU_GFX_TCP_SET3BANK3_RAM, + AMDGPU_GFX_TCP_VM_FIFO, + AMDGPU_GFX_TCP_DB_TAGRAM0, + AMDGPU_GFX_TCP_DB_TAGRAM1, + AMDGPU_GFX_TCP_DB_TAGRAM2, + AMDGPU_GFX_TCP_DB_TAGRAM3, + AMDGPU_GFX_TCP_UTCL1_LFIFO_PROBE0, + AMDGPU_GFX_TCP_UTCL1_LFIFO_PROBE1, + AMDGPU_GFX_TCP_CMD_FIFO, +}; + +enum amdgpu_gfx_td_ras_mem_id { + AMDGPU_GFX_TD_UTD_CS_FIFO_MEM = 1, + AMDGPU_GFX_TD_UTD_SS_FIFO_LO_MEM, + AMDGPU_GFX_TD_UTD_SS_FIFO_HI_MEM, +}; + +enum amdgpu_gfx_tcx_ras_mem_id { + AMDGPU_GFX_TCX_FIFOD0 = 0, + AMDGPU_GFX_TCX_FIFOD1, + AMDGPU_GFX_TCX_FIFOD2, + AMDGPU_GFX_TCX_FIFOD3, + AMDGPU_GFX_TCX_FIFOD4, + AMDGPU_GFX_TCX_FIFOD5, + AMDGPU_GFX_TCX_FIFOD6, + AMDGPU_GFX_TCX_FIFOD7, + AMDGPU_GFX_TCX_FIFOB0, + AMDGPU_GFX_TCX_FIFOB1, + AMDGPU_GFX_TCX_FIFOB2, + AMDGPU_GFX_TCX_FIFOB3, + AMDGPU_GFX_TCX_FIFOB4, + AMDGPU_GFX_TCX_FIFOB5, + AMDGPU_GFX_TCX_FIFOB6, + AMDGPU_GFX_TCX_FIFOB7, + AMDGPU_GFX_TCX_FIFOA0, + AMDGPU_GFX_TCX_FIFOA1, + AMDGPU_GFX_TCX_FIFOA2, + AMDGPU_GFX_TCX_FIFOA3, + AMDGPU_GFX_TCX_FIFOA4, + AMDGPU_GFX_TCX_FIFOA5, + AMDGPU_GFX_TCX_FIFOA6, + AMDGPU_GFX_TCX_FIFOA7, + AMDGPU_GFX_TCX_CFIFO0, + AMDGPU_GFX_TCX_CFIFO1, + AMDGPU_GFX_TCX_CFIFO2, + AMDGPU_GFX_TCX_CFIFO3, + AMDGPU_GFX_TCX_CFIFO4, + AMDGPU_GFX_TCX_CFIFO5, + AMDGPU_GFX_TCX_CFIFO6, + AMDGPU_GFX_TCX_CFIFO7, + AMDGPU_GFX_TCX_FIFO_ACKB0, + AMDGPU_GFX_TCX_FIFO_ACKB1, + AMDGPU_GFX_TCX_FIFO_ACKB2, + AMDGPU_GFX_TCX_FIFO_ACKB3, + AMDGPU_GFX_TCX_FIFO_ACKB4, + AMDGPU_GFX_TCX_FIFO_ACKB5, + AMDGPU_GFX_TCX_FIFO_ACKB6, + AMDGPU_GFX_TCX_FIFO_ACKB7, + AMDGPU_GFX_TCX_FIFO_ACKD0, + AMDGPU_GFX_TCX_FIFO_ACKD1, + AMDGPU_GFX_TCX_FIFO_ACKD2, + AMDGPU_GFX_TCX_FIFO_ACKD3, + AMDGPU_GFX_TCX_FIFO_ACKD4, + AMDGPU_GFX_TCX_FIFO_ACKD5, + AMDGPU_GFX_TCX_FIFO_ACKD6, + AMDGPU_GFX_TCX_FIFO_ACKD7, + AMDGPU_GFX_TCX_DST_FIFOA0, + AMDGPU_GFX_TCX_DST_FIFOA1, + AMDGPU_GFX_TCX_DST_FIFOA2, + AMDGPU_GFX_TCX_DST_FIFOA3, + AMDGPU_GFX_TCX_DST_FIFOA4, + AMDGPU_GFX_TCX_DST_FIFOA5, + AMDGPU_GFX_TCX_DST_FIFOA6, + AMDGPU_GFX_TCX_DST_FIFOA7, + AMDGPU_GFX_TCX_DST_FIFOB0, + AMDGPU_GFX_TCX_DST_FIFOB1, + AMDGPU_GFX_TCX_DST_FIFOB2, + AMDGPU_GFX_TCX_DST_FIFOB3, + AMDGPU_GFX_TCX_DST_FIFOB4, + AMDGPU_GFX_TCX_DST_FIFOB5, + AMDGPU_GFX_TCX_DST_FIFOB6, + AMDGPU_GFX_TCX_DST_FIFOB7, + AMDGPU_GFX_TCX_DST_FIFOD0, + AMDGPU_GFX_TCX_DST_FIFOD1, + AMDGPU_GFX_TCX_DST_FIFOD2, + AMDGPU_GFX_TCX_DST_FIFOD3, + AMDGPU_GFX_TCX_DST_FIFOD4, + AMDGPU_GFX_TCX_DST_FIFOD5, + AMDGPU_GFX_TCX_DST_FIFOD6, + AMDGPU_GFX_TCX_DST_FIFOD7, + AMDGPU_GFX_TCX_DST_FIFO_ACKB0, + AMDGPU_GFX_TCX_DST_FIFO_ACKB1, + AMDGPU_GFX_TCX_DST_FIFO_ACKB2, + AMDGPU_GFX_TCX_DST_FIFO_ACKB3, + AMDGPU_GFX_TCX_DST_FIFO_ACKB4, + AMDGPU_GFX_TCX_DST_FIFO_ACKB5, + AMDGPU_GFX_TCX_DST_FIFO_ACKB6, + AMDGPU_GFX_TCX_DST_FIFO_ACKB7, + AMDGPU_GFX_TCX_DST_FIFO_ACKD0, + AMDGPU_GFX_TCX_DST_FIFO_ACKD1, + AMDGPU_GFX_TCX_DST_FIFO_ACKD2, + AMDGPU_GFX_TCX_DST_FIFO_ACKD3, + AMDGPU_GFX_TCX_DST_FIFO_ACKD4, + AMDGPU_GFX_TCX_DST_FIFO_ACKD5, + AMDGPU_GFX_TCX_DST_FIFO_ACKD6, + AMDGPU_GFX_TCX_DST_FIFO_ACKD7, +}; + +enum amdgpu_gfx_atc_l2_ras_mem_id { + AMDGPU_GFX_ATC_L2_MEM0 = 0, +}; + +enum amdgpu_gfx_utcl2_ras_mem_id { + AMDGPU_GFX_UTCL2_MEM0 = 0, +}; + +enum amdgpu_gfx_vml2_ras_mem_id { + AMDGPU_GFX_VML2_MEM0 = 0, +}; + +enum amdgpu_gfx_vml2_walker_ras_mem_id { + AMDGPU_GFX_VML2_WALKER_MEM0 = 0, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_cp_mem_list[] = { + {AMDGPU_GFX_CP_MEM1, "CP_MEM1"}, + {AMDGPU_GFX_CP_MEM2, "CP_MEM2"}, + {AMDGPU_GFX_CP_MEM3, "CP_MEM3"}, + {AMDGPU_GFX_CP_MEM4, "CP_MEM4"}, + {AMDGPU_GFX_CP_MEM5, "CP_MEM5"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_gcea_mem_list[] = { + {AMDGPU_GFX_GCEA_IOWR_CMDMEM, "GCEA_IOWR_CMDMEM"}, + {AMDGPU_GFX_GCEA_IORD_CMDMEM, "GCEA_IORD_CMDMEM"}, + {AMDGPU_GFX_GCEA_GMIWR_CMDMEM, "GCEA_GMIWR_CMDMEM"}, + {AMDGPU_GFX_GCEA_GMIRD_CMDMEM, "GCEA_GMIRD_CMDMEM"}, + {AMDGPU_GFX_GCEA_DRAMWR_CMDMEM, "GCEA_DRAMWR_CMDMEM"}, + {AMDGPU_GFX_GCEA_DRAMRD_CMDMEM, "GCEA_DRAMRD_CMDMEM"}, + {AMDGPU_GFX_GCEA_MAM_DMEM0, "GCEA_MAM_DMEM0"}, + {AMDGPU_GFX_GCEA_MAM_DMEM1, "GCEA_MAM_DMEM1"}, + {AMDGPU_GFX_GCEA_MAM_DMEM2, "GCEA_MAM_DMEM2"}, + {AMDGPU_GFX_GCEA_MAM_DMEM3, "GCEA_MAM_DMEM3"}, + {AMDGPU_GFX_GCEA_MAM_AMEM0, "GCEA_MAM_AMEM0"}, + {AMDGPU_GFX_GCEA_MAM_AMEM1, "GCEA_MAM_AMEM1"}, + {AMDGPU_GFX_GCEA_MAM_AMEM2, "GCEA_MAM_AMEM2"}, + {AMDGPU_GFX_GCEA_MAM_AMEM3, "GCEA_MAM_AMEM3"}, + {AMDGPU_GFX_GCEA_MAM_AFLUSH_BUFFER, "GCEA_MAM_AFLUSH_BUFFER"}, + {AMDGPU_GFX_GCEA_WRET_TAGMEM, "GCEA_WRET_TAGMEM"}, + {AMDGPU_GFX_GCEA_RRET_TAGMEM, "GCEA_RRET_TAGMEM"}, + {AMDGPU_GFX_GCEA_IOWR_DATAMEM, "GCEA_IOWR_DATAMEM"}, + {AMDGPU_GFX_GCEA_GMIWR_DATAMEM, "GCEA_GMIWR_DATAMEM"}, + {AMDGPU_GFX_GCEA_DRAM_DATAMEM, "GCEA_DRAM_DATAMEM"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_gc_cane_mem_list[] = { + {AMDGPU_GFX_GC_CANE_MEM0, "GC_CANE_MEM0"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_gcutcl2_mem_list[] = { + {AMDGPU_GFX_GCUTCL2_MEM2P512X95, "GCUTCL2_MEM2P512X95"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_gds_mem_list[] = { + {AMDGPU_GFX_GDS_MEM0, "GDS_MEM"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_lds_mem_list[] = { + {AMDGPU_GFX_LDS_BANK0, "LDS_BANK0"}, + {AMDGPU_GFX_LDS_BANK1, "LDS_BANK1"}, + {AMDGPU_GFX_LDS_BANK2, "LDS_BANK2"}, + {AMDGPU_GFX_LDS_BANK3, "LDS_BANK3"}, + {AMDGPU_GFX_LDS_BANK4, "LDS_BANK4"}, + {AMDGPU_GFX_LDS_BANK5, "LDS_BANK5"}, + {AMDGPU_GFX_LDS_BANK6, "LDS_BANK6"}, + {AMDGPU_GFX_LDS_BANK7, "LDS_BANK7"}, + {AMDGPU_GFX_LDS_BANK8, "LDS_BANK8"}, + {AMDGPU_GFX_LDS_BANK9, "LDS_BANK9"}, + {AMDGPU_GFX_LDS_BANK10, "LDS_BANK10"}, + {AMDGPU_GFX_LDS_BANK11, "LDS_BANK11"}, + {AMDGPU_GFX_LDS_BANK12, "LDS_BANK12"}, + {AMDGPU_GFX_LDS_BANK13, "LDS_BANK13"}, + {AMDGPU_GFX_LDS_BANK14, "LDS_BANK14"}, + {AMDGPU_GFX_LDS_BANK15, "LDS_BANK15"}, + {AMDGPU_GFX_LDS_BANK16, "LDS_BANK16"}, + {AMDGPU_GFX_LDS_BANK17, "LDS_BANK17"}, + {AMDGPU_GFX_LDS_BANK18, "LDS_BANK18"}, + {AMDGPU_GFX_LDS_BANK19, "LDS_BANK19"}, + {AMDGPU_GFX_LDS_BANK20, "LDS_BANK20"}, + {AMDGPU_GFX_LDS_BANK21, "LDS_BANK21"}, + {AMDGPU_GFX_LDS_BANK22, "LDS_BANK22"}, + {AMDGPU_GFX_LDS_BANK23, "LDS_BANK23"}, + {AMDGPU_GFX_LDS_BANK24, "LDS_BANK24"}, + {AMDGPU_GFX_LDS_BANK25, "LDS_BANK25"}, + {AMDGPU_GFX_LDS_BANK26, "LDS_BANK26"}, + {AMDGPU_GFX_LDS_BANK27, "LDS_BANK27"}, + {AMDGPU_GFX_LDS_BANK28, "LDS_BANK28"}, + {AMDGPU_GFX_LDS_BANK29, "LDS_BANK29"}, + {AMDGPU_GFX_LDS_BANK30, "LDS_BANK30"}, + {AMDGPU_GFX_LDS_BANK31, "LDS_BANK31"}, + {AMDGPU_GFX_LDS_SP_BUFFER_A, "LDS_SP_BUFFER_A"}, + {AMDGPU_GFX_LDS_SP_BUFFER_B, "LDS_SP_BUFFER_B"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_rlc_mem_list[] = { + {AMDGPU_GFX_RLC_GPMF32, "RLC_GPMF32"}, + {AMDGPU_GFX_RLC_RLCVF32, "RLC_RLCVF32"}, + {AMDGPU_GFX_RLC_SCRATCH, "RLC_SCRATCH"}, + {AMDGPU_GFX_RLC_SRM_ARAM, "RLC_SRM_ARAM"}, + {AMDGPU_GFX_RLC_SRM_DRAM, "RLC_SRM_DRAM"}, + {AMDGPU_GFX_RLC_TCTAG, "RLC_TCTAG"}, + {AMDGPU_GFX_RLC_SPM_SE, "RLC_SPM_SE"}, + {AMDGPU_GFX_RLC_SPM_GRBMT, "RLC_SPM_GRBMT"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_sp_mem_list[] = { + {AMDGPU_GFX_SP_SIMDID0, "SP_SIMDID0"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_spi_mem_list[] = { + {AMDGPU_GFX_SPI_MEM0, "SPI_MEM0"}, + {AMDGPU_GFX_SPI_MEM1, "SPI_MEM1"}, + {AMDGPU_GFX_SPI_MEM2, "SPI_MEM2"}, + {AMDGPU_GFX_SPI_MEM3, "SPI_MEM3"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_sqc_mem_list[] = { + {AMDGPU_GFX_SQC_INST_CACHE_A, "SQC_INST_CACHE_A"}, + {AMDGPU_GFX_SQC_INST_CACHE_B, "SQC_INST_CACHE_B"}, + {AMDGPU_GFX_SQC_INST_CACHE_TAG_A, "SQC_INST_CACHE_TAG_A"}, + {AMDGPU_GFX_SQC_INST_CACHE_TAG_B, "SQC_INST_CACHE_TAG_B"}, + {AMDGPU_GFX_SQC_INST_CACHE_MISS_FIFO_A, "SQC_INST_CACHE_MISS_FIFO_A"}, + {AMDGPU_GFX_SQC_INST_CACHE_MISS_FIFO_B, "SQC_INST_CACHE_MISS_FIFO_B"}, + {AMDGPU_GFX_SQC_INST_CACHE_GATCL1_MISS_FIFO_A, "SQC_INST_CACHE_GATCL1_MISS_FIFO_A"}, + {AMDGPU_GFX_SQC_INST_CACHE_GATCL1_MISS_FIFO_B, "SQC_INST_CACHE_GATCL1_MISS_FIFO_B"}, + {AMDGPU_GFX_SQC_DATA_CACHE_A, "SQC_DATA_CACHE_A"}, + {AMDGPU_GFX_SQC_DATA_CACHE_B, "SQC_DATA_CACHE_B"}, + {AMDGPU_GFX_SQC_DATA_CACHE_TAG_A, "SQC_DATA_CACHE_TAG_A"}, + {AMDGPU_GFX_SQC_DATA_CACHE_TAG_B, "SQC_DATA_CACHE_TAG_B"}, + {AMDGPU_GFX_SQC_DATA_CACHE_MISS_FIFO_A, "SQC_DATA_CACHE_MISS_FIFO_A"}, + {AMDGPU_GFX_SQC_DATA_CACHE_MISS_FIFO_B, "SQC_DATA_CACHE_MISS_FIFO_B"}, + {AMDGPU_GFX_SQC_DATA_CACHE_HIT_FIFO_A, "SQC_DATA_CACHE_HIT_FIFO_A"}, + {AMDGPU_GFX_SQC_DATA_CACHE_HIT_FIFO_B, "SQC_DATA_CACHE_HIT_FIFO_B"}, + {AMDGPU_GFX_SQC_DIRTY_BIT_A, "SQC_DIRTY_BIT_A"}, + {AMDGPU_GFX_SQC_DIRTY_BIT_B, "SQC_DIRTY_BIT_B"}, + {AMDGPU_GFX_SQC_WRITE_DATA_BUFFER_CU0, "SQC_WRITE_DATA_BUFFER_CU0"}, + {AMDGPU_GFX_SQC_WRITE_DATA_BUFFER_CU1, "SQC_WRITE_DATA_BUFFER_CU1"}, + {AMDGPU_GFX_SQC_UTCL1_MISS_LFIFO_DATA_CACHE_A, "SQC_UTCL1_MISS_LFIFO_DATA_CACHE_A"}, + {AMDGPU_GFX_SQC_UTCL1_MISS_LFIFO_DATA_CACHE_B, "SQC_UTCL1_MISS_LFIFO_DATA_CACHE_B"}, + {AMDGPU_GFX_SQC_UTCL1_MISS_LFIFO_INST_CACHE, "SQC_UTCL1_MISS_LFIFO_INST_CACHE"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_sq_mem_list[] = { + {AMDGPU_GFX_SQ_SGPR_MEM0, "SQ_SGPR_MEM0"}, + {AMDGPU_GFX_SQ_SGPR_MEM1, "SQ_SGPR_MEM1"}, + {AMDGPU_GFX_SQ_SGPR_MEM2, "SQ_SGPR_MEM2"}, + {AMDGPU_GFX_SQ_SGPR_MEM3, "SQ_SGPR_MEM3"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_ta_mem_list[] = { + {AMDGPU_GFX_TA_FS_AFIFO_RAM_LO, "TA_FS_AFIFO_RAM_LO"}, + {AMDGPU_GFX_TA_FS_AFIFO_RAM_HI, "TA_FS_AFIFO_RAM_HI"}, + {AMDGPU_GFX_TA_FS_CFIFO_RAM, "TA_FS_CFIFO_RAM"}, + {AMDGPU_GFX_TA_FSX_LFIFO, "TA_FSX_LFIFO"}, + {AMDGPU_GFX_TA_FS_DFIFO_RAM, "TA_FS_DFIFO_RAM"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_tcc_mem_list[] = { + {AMDGPU_GFX_TCC_MEM1, "TCC_MEM1"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_tca_mem_list[] = { + {AMDGPU_GFX_TCA_MEM1, "TCA_MEM1"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_tci_mem_list[] = { + {AMDGPU_GFX_TCIW_MEM, "TCIW_MEM"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_tcp_mem_list[] = { + {AMDGPU_GFX_TCP_LFIFO0, "TCP_LFIFO0"}, + {AMDGPU_GFX_TCP_SET0BANK0_RAM, "TCP_SET0BANK0_RAM"}, + {AMDGPU_GFX_TCP_SET0BANK1_RAM, "TCP_SET0BANK1_RAM"}, + {AMDGPU_GFX_TCP_SET0BANK2_RAM, "TCP_SET0BANK2_RAM"}, + {AMDGPU_GFX_TCP_SET0BANK3_RAM, "TCP_SET0BANK3_RAM"}, + {AMDGPU_GFX_TCP_SET1BANK0_RAM, "TCP_SET1BANK0_RAM"}, + {AMDGPU_GFX_TCP_SET1BANK1_RAM, "TCP_SET1BANK1_RAM"}, + {AMDGPU_GFX_TCP_SET1BANK2_RAM, "TCP_SET1BANK2_RAM"}, + {AMDGPU_GFX_TCP_SET1BANK3_RAM, "TCP_SET1BANK3_RAM"}, + {AMDGPU_GFX_TCP_SET2BANK0_RAM, "TCP_SET2BANK0_RAM"}, + {AMDGPU_GFX_TCP_SET2BANK1_RAM, "TCP_SET2BANK1_RAM"}, + {AMDGPU_GFX_TCP_SET2BANK2_RAM, "TCP_SET2BANK2_RAM"}, + {AMDGPU_GFX_TCP_SET2BANK3_RAM, "TCP_SET2BANK3_RAM"}, + {AMDGPU_GFX_TCP_SET3BANK0_RAM, "TCP_SET3BANK0_RAM"}, + {AMDGPU_GFX_TCP_SET3BANK1_RAM, "TCP_SET3BANK1_RAM"}, + {AMDGPU_GFX_TCP_SET3BANK2_RAM, "TCP_SET3BANK2_RAM"}, + {AMDGPU_GFX_TCP_SET3BANK3_RAM, "TCP_SET3BANK3_RAM"}, + {AMDGPU_GFX_TCP_VM_FIFO, "TCP_VM_FIFO"}, + {AMDGPU_GFX_TCP_DB_TAGRAM0, "TCP_DB_TAGRAM0"}, + {AMDGPU_GFX_TCP_DB_TAGRAM1, "TCP_DB_TAGRAM1"}, + {AMDGPU_GFX_TCP_DB_TAGRAM2, "TCP_DB_TAGRAM2"}, + {AMDGPU_GFX_TCP_DB_TAGRAM3, "TCP_DB_TAGRAM3"}, + {AMDGPU_GFX_TCP_UTCL1_LFIFO_PROBE0, "TCP_UTCL1_LFIFO_PROBE0"}, + {AMDGPU_GFX_TCP_UTCL1_LFIFO_PROBE1, "TCP_UTCL1_LFIFO_PROBE1"}, + {AMDGPU_GFX_TCP_CMD_FIFO, "TCP_CMD_FIFO"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_td_mem_list[] = { + {AMDGPU_GFX_TD_UTD_CS_FIFO_MEM, "TD_UTD_CS_FIFO_MEM"}, + {AMDGPU_GFX_TD_UTD_SS_FIFO_LO_MEM, "TD_UTD_SS_FIFO_LO_MEM"}, + {AMDGPU_GFX_TD_UTD_SS_FIFO_HI_MEM, "TD_UTD_SS_FIFO_HI_MEM"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_tcx_mem_list[] = { + {AMDGPU_GFX_TCX_FIFOD0, "TCX_FIFOD0"}, + {AMDGPU_GFX_TCX_FIFOD1, "TCX_FIFOD1"}, + {AMDGPU_GFX_TCX_FIFOD2, "TCX_FIFOD2"}, + {AMDGPU_GFX_TCX_FIFOD3, "TCX_FIFOD3"}, + {AMDGPU_GFX_TCX_FIFOD4, "TCX_FIFOD4"}, + {AMDGPU_GFX_TCX_FIFOD5, "TCX_FIFOD5"}, + {AMDGPU_GFX_TCX_FIFOD6, "TCX_FIFOD6"}, + {AMDGPU_GFX_TCX_FIFOD7, "TCX_FIFOD7"}, + {AMDGPU_GFX_TCX_FIFOB0, "TCX_FIFOB0"}, + {AMDGPU_GFX_TCX_FIFOB1, "TCX_FIFOB1"}, + {AMDGPU_GFX_TCX_FIFOB2, "TCX_FIFOB2"}, + {AMDGPU_GFX_TCX_FIFOB3, "TCX_FIFOB3"}, + {AMDGPU_GFX_TCX_FIFOB4, "TCX_FIFOB4"}, + {AMDGPU_GFX_TCX_FIFOB5, "TCX_FIFOB5"}, + {AMDGPU_GFX_TCX_FIFOB6, "TCX_FIFOB6"}, + {AMDGPU_GFX_TCX_FIFOB7, "TCX_FIFOB7"}, + {AMDGPU_GFX_TCX_FIFOA0, "TCX_FIFOA0"}, + {AMDGPU_GFX_TCX_FIFOA1, "TCX_FIFOA1"}, + {AMDGPU_GFX_TCX_FIFOA2, "TCX_FIFOA2"}, + {AMDGPU_GFX_TCX_FIFOA3, "TCX_FIFOA3"}, + {AMDGPU_GFX_TCX_FIFOA4, "TCX_FIFOA4"}, + {AMDGPU_GFX_TCX_FIFOA5, "TCX_FIFOA5"}, + {AMDGPU_GFX_TCX_FIFOA6, "TCX_FIFOA6"}, + {AMDGPU_GFX_TCX_FIFOA7, "TCX_FIFOA7"}, + {AMDGPU_GFX_TCX_CFIFO0, "TCX_CFIFO0"}, + {AMDGPU_GFX_TCX_CFIFO1, "TCX_CFIFO1"}, + {AMDGPU_GFX_TCX_CFIFO2, "TCX_CFIFO2"}, + {AMDGPU_GFX_TCX_CFIFO3, "TCX_CFIFO3"}, + {AMDGPU_GFX_TCX_CFIFO4, "TCX_CFIFO4"}, + {AMDGPU_GFX_TCX_CFIFO5, "TCX_CFIFO5"}, + {AMDGPU_GFX_TCX_CFIFO6, "TCX_CFIFO6"}, + {AMDGPU_GFX_TCX_CFIFO7, "TCX_CFIFO7"}, + {AMDGPU_GFX_TCX_FIFO_ACKB0, "TCX_FIFO_ACKB0"}, + {AMDGPU_GFX_TCX_FIFO_ACKB1, "TCX_FIFO_ACKB1"}, + {AMDGPU_GFX_TCX_FIFO_ACKB2, "TCX_FIFO_ACKB2"}, + {AMDGPU_GFX_TCX_FIFO_ACKB3, "TCX_FIFO_ACKB3"}, + {AMDGPU_GFX_TCX_FIFO_ACKB4, "TCX_FIFO_ACKB4"}, + {AMDGPU_GFX_TCX_FIFO_ACKB5, "TCX_FIFO_ACKB5"}, + {AMDGPU_GFX_TCX_FIFO_ACKB6, "TCX_FIFO_ACKB6"}, + {AMDGPU_GFX_TCX_FIFO_ACKB7, "TCX_FIFO_ACKB7"}, + {AMDGPU_GFX_TCX_FIFO_ACKD0, "TCX_FIFO_ACKD0"}, + {AMDGPU_GFX_TCX_FIFO_ACKD1, "TCX_FIFO_ACKD1"}, + {AMDGPU_GFX_TCX_FIFO_ACKD2, "TCX_FIFO_ACKD2"}, + {AMDGPU_GFX_TCX_FIFO_ACKD3, "TCX_FIFO_ACKD3"}, + {AMDGPU_GFX_TCX_FIFO_ACKD4, "TCX_FIFO_ACKD4"}, + {AMDGPU_GFX_TCX_FIFO_ACKD5, "TCX_FIFO_ACKD5"}, + {AMDGPU_GFX_TCX_FIFO_ACKD6, "TCX_FIFO_ACKD6"}, + {AMDGPU_GFX_TCX_FIFO_ACKD7, "TCX_FIFO_ACKD7"}, + {AMDGPU_GFX_TCX_DST_FIFOA0, "TCX_DST_FIFOA0"}, + {AMDGPU_GFX_TCX_DST_FIFOA1, "TCX_DST_FIFOA1"}, + {AMDGPU_GFX_TCX_DST_FIFOA2, "TCX_DST_FIFOA2"}, + {AMDGPU_GFX_TCX_DST_FIFOA3, "TCX_DST_FIFOA3"}, + {AMDGPU_GFX_TCX_DST_FIFOA4, "TCX_DST_FIFOA4"}, + {AMDGPU_GFX_TCX_DST_FIFOA5, "TCX_DST_FIFOA5"}, + {AMDGPU_GFX_TCX_DST_FIFOA6, "TCX_DST_FIFOA6"}, + {AMDGPU_GFX_TCX_DST_FIFOA7, "TCX_DST_FIFOA7"}, + {AMDGPU_GFX_TCX_DST_FIFOB0, "TCX_DST_FIFOB0"}, + {AMDGPU_GFX_TCX_DST_FIFOB1, "TCX_DST_FIFOB1"}, + {AMDGPU_GFX_TCX_DST_FIFOB2, "TCX_DST_FIFOB2"}, + {AMDGPU_GFX_TCX_DST_FIFOB3, "TCX_DST_FIFOB3"}, + {AMDGPU_GFX_TCX_DST_FIFOB4, "TCX_DST_FIFOB4"}, + {AMDGPU_GFX_TCX_DST_FIFOB5, "TCX_DST_FIFOB5"}, + {AMDGPU_GFX_TCX_DST_FIFOB6, "TCX_DST_FIFOB6"}, + {AMDGPU_GFX_TCX_DST_FIFOB7, "TCX_DST_FIFOB7"}, + {AMDGPU_GFX_TCX_DST_FIFOD0, "TCX_DST_FIFOD0"}, + {AMDGPU_GFX_TCX_DST_FIFOD1, "TCX_DST_FIFOD1"}, + {AMDGPU_GFX_TCX_DST_FIFOD2, "TCX_DST_FIFOD2"}, + {AMDGPU_GFX_TCX_DST_FIFOD3, "TCX_DST_FIFOD3"}, + {AMDGPU_GFX_TCX_DST_FIFOD4, "TCX_DST_FIFOD4"}, + {AMDGPU_GFX_TCX_DST_FIFOD5, "TCX_DST_FIFOD5"}, + {AMDGPU_GFX_TCX_DST_FIFOD6, "TCX_DST_FIFOD6"}, + {AMDGPU_GFX_TCX_DST_FIFOD7, "TCX_DST_FIFOD7"}, + {AMDGPU_GFX_TCX_DST_FIFO_ACKB0, "TCX_DST_FIFO_ACKB0"}, + {AMDGPU_GFX_TCX_DST_FIFO_ACKB1, "TCX_DST_FIFO_ACKB1"}, + {AMDGPU_GFX_TCX_DST_FIFO_ACKB2, "TCX_DST_FIFO_ACKB2"}, + {AMDGPU_GFX_TCX_DST_FIFO_ACKB3, "TCX_DST_FIFO_ACKB3"}, + {AMDGPU_GFX_TCX_DST_FIFO_ACKB4, "TCX_DST_FIFO_ACKB4"}, + {AMDGPU_GFX_TCX_DST_FIFO_ACKB5, "TCX_DST_FIFO_ACKB5"}, + {AMDGPU_GFX_TCX_DST_FIFO_ACKB6, "TCX_DST_FIFO_ACKB6"}, + {AMDGPU_GFX_TCX_DST_FIFO_ACKB7, "TCX_DST_FIFO_ACKB7"}, + {AMDGPU_GFX_TCX_DST_FIFO_ACKD0, "TCX_DST_FIFO_ACKD0"}, + {AMDGPU_GFX_TCX_DST_FIFO_ACKD1, "TCX_DST_FIFO_ACKD1"}, + {AMDGPU_GFX_TCX_DST_FIFO_ACKD2, "TCX_DST_FIFO_ACKD2"}, + {AMDGPU_GFX_TCX_DST_FIFO_ACKD3, "TCX_DST_FIFO_ACKD3"}, + {AMDGPU_GFX_TCX_DST_FIFO_ACKD4, "TCX_DST_FIFO_ACKD4"}, + {AMDGPU_GFX_TCX_DST_FIFO_ACKD5, "TCX_DST_FIFO_ACKD5"}, + {AMDGPU_GFX_TCX_DST_FIFO_ACKD6, "TCX_DST_FIFO_ACKD6"}, + {AMDGPU_GFX_TCX_DST_FIFO_ACKD7, "TCX_DST_FIFO_ACKD7"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_atc_l2_mem_list[] = { + {AMDGPU_GFX_ATC_L2_MEM, "ATC_L2_MEM"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_utcl2_mem_list[] = { + {AMDGPU_GFX_UTCL2_MEM, "UTCL2_MEM"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_vml2_mem_list[] = { + {AMDGPU_GFX_VML2_MEM, "VML2_MEM"}, +}; + +static const struct amdgpu_ras_memory_id_entry gfx_v9_4_3_ras_vml2_walker_mem_list[] = { + {AMDGPU_GFX_VML2_WALKER_MEM, "VML2_WALKER_MEM"}, +}; + +static const struct amdgpu_gfx_ras_mem_id_entry gfx_v9_4_3_ras_mem_list_array[AMDGPU_GFX_MEM_TYPE_NUM] = { + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_cp_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_gcea_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_gc_cane_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_gcutcl2_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_gds_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_lds_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_rlc_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_sp_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_spi_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_sqc_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_sq_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_ta_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_tcc_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_tca_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_tci_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_tcp_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_td_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_tcx_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_atc_l2_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_utcl2_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_vml2_mem_list) + AMDGPU_GFX_MEMID_ENT(gfx_v9_4_3_ras_vml2_walker_mem_list) +}; + +static const struct amdgpu_gfx_ras_reg_entry gfx_v9_4_3_ce_reg_list[] = { + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regRLC_CE_ERR_STATUS_LOW, regRLC_CE_ERR_STATUS_HIGH), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "RLC"}, + AMDGPU_GFX_RLC_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regCPC_CE_ERR_STATUS_LO, regCPC_CE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "CPC"}, + AMDGPU_GFX_CP_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regCPF_CE_ERR_STATUS_LO, regCPF_CE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "CPF"}, + AMDGPU_GFX_CP_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regCPG_CE_ERR_STATUS_LO, regCPG_CE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "CPG"}, + AMDGPU_GFX_CP_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regGDS_CE_ERR_STATUS_LO, regGDS_CE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "GDS"}, + AMDGPU_GFX_GDS_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regGC_CANE_CE_ERR_STATUS_LO, regGC_CANE_CE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "CANE"}, + AMDGPU_GFX_GC_CANE_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regSPI_CE_ERR_STATUS_LO, regSPI_CE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SPI"}, + AMDGPU_GFX_SPI_MEM, 8}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regSP0_CE_ERR_STATUS_LO, regSP0_CE_ERR_STATUS_HI), + 10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SP0"}, + AMDGPU_GFX_SP_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regSP1_CE_ERR_STATUS_LO, regSP1_CE_ERR_STATUS_HI), + 10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SP1"}, + AMDGPU_GFX_SP_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regSQ_CE_ERR_STATUS_LO, regSQ_CE_ERR_STATUS_HI), + 10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SQ"}, + AMDGPU_GFX_SQ_MEM, 8}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regSQC_CE_EDC_LO, regSQC_CE_EDC_HI), + 5, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SQC"}, + AMDGPU_GFX_SQC_MEM, 8}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCX_CE_ERR_STATUS_LO, regTCX_CE_ERR_STATUS_HI), + 2, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCX"}, + AMDGPU_GFX_TCX_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCC_CE_ERR_STATUS_LO, regTCC_CE_ERR_STATUS_HI), + 16, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCC"}, + AMDGPU_GFX_TCC_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regTA_CE_EDC_LO, regTA_CE_EDC_HI), + 10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TA"}, + AMDGPU_GFX_TA_MEM, 8}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCI_CE_EDC_LO_REG, regTCI_CE_EDC_HI_REG), + 31, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCI"}, + AMDGPU_GFX_TCI_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCP_CE_EDC_LO_REG, regTCP_CE_EDC_HI_REG), + 10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCP"}, + AMDGPU_GFX_TCP_MEM, 8}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regTD_CE_EDC_LO, regTD_CE_EDC_HI), + 10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TD"}, + AMDGPU_GFX_TD_MEM, 8}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regGCEA_CE_ERR_STATUS_LO, regGCEA_CE_ERR_STATUS_HI), + 16, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "GCEA"}, + AMDGPU_GFX_GCEA_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regLDS_CE_ERR_STATUS_LO, regLDS_CE_ERR_STATUS_HI), + 10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "LDS"}, + AMDGPU_GFX_LDS_MEM, 1}, +}; + +static const struct amdgpu_gfx_ras_reg_entry gfx_v9_4_3_ue_reg_list[] = { + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regRLC_UE_ERR_STATUS_LOW, regRLC_UE_ERR_STATUS_HIGH), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "RLC"}, + AMDGPU_GFX_RLC_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regCPC_UE_ERR_STATUS_LO, regCPC_UE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "CPC"}, + AMDGPU_GFX_CP_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regCPF_UE_ERR_STATUS_LO, regCPF_UE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "CPF"}, + AMDGPU_GFX_CP_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regCPG_UE_ERR_STATUS_LO, regCPG_UE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "CPG"}, + AMDGPU_GFX_CP_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regGDS_UE_ERR_STATUS_LO, regGDS_UE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "GDS"}, + AMDGPU_GFX_GDS_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regGC_CANE_UE_ERR_STATUS_LO, regGC_CANE_UE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "CANE"}, + AMDGPU_GFX_GC_CANE_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regSPI_UE_ERR_STATUS_LO, regSPI_UE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SPI"}, + AMDGPU_GFX_SPI_MEM, 8}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regSP0_UE_ERR_STATUS_LO, regSP0_UE_ERR_STATUS_HI), + 10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SP0"}, + AMDGPU_GFX_SP_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regSP1_UE_ERR_STATUS_LO, regSP1_UE_ERR_STATUS_HI), + 10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SP1"}, + AMDGPU_GFX_SP_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regSQ_UE_ERR_STATUS_LO, regSQ_UE_ERR_STATUS_HI), + 10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SQ"}, + AMDGPU_GFX_SQ_MEM, 8}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regSQC_UE_EDC_LO, regSQC_UE_EDC_HI), + 5, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SQC"}, + AMDGPU_GFX_SQC_MEM, 8}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCX_UE_ERR_STATUS_LO, regTCX_UE_ERR_STATUS_HI), + 2, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCX"}, + AMDGPU_GFX_TCX_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCC_UE_ERR_STATUS_LO, regTCC_UE_ERR_STATUS_HI), + 16, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCC"}, + AMDGPU_GFX_TCC_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regTA_UE_EDC_LO, regTA_UE_EDC_HI), + 10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TA"}, + AMDGPU_GFX_TA_MEM, 8}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCI_UE_EDC_LO_REG, regTCI_UE_EDC_HI_REG), + 31, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCI"}, + AMDGPU_GFX_TCI_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCP_UE_EDC_LO_REG, regTCP_UE_EDC_HI_REG), + 10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCP"}, + AMDGPU_GFX_TCP_MEM, 8}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regTD_UE_EDC_LO, regTD_UE_EDC_HI), + 10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TD"}, + AMDGPU_GFX_TD_MEM, 8}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regTCA_UE_ERR_STATUS_LO, regTCA_UE_ERR_STATUS_HI), + 2, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "TCA"}, + AMDGPU_GFX_TCA_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regGCEA_UE_ERR_STATUS_LO, regGCEA_UE_ERR_STATUS_HI), + 16, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "GCEA"}, + AMDGPU_GFX_GCEA_MEM, 1}, + {{AMDGPU_RAS_REG_ENTRY(GC, 0, regLDS_UE_ERR_STATUS_LO, regLDS_UE_ERR_STATUS_HI), + 10, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "LDS"}, + AMDGPU_GFX_LDS_MEM, 1}, +}; + +static const struct soc15_reg_entry gfx_v9_4_3_ea_err_status_regs = { + SOC15_REG_ENTRY(GC, 0, regGCEA_ERR_STATUS), 0, 1, 16 +}; + +static void gfx_v9_4_3_inst_query_ras_err_count(struct amdgpu_device *adev, + void *ras_error_status, int xcc_id) +{ + struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; + unsigned long ce_count = 0, ue_count = 0; + uint32_t i, j, k; + + mutex_lock(&adev->grbm_idx_mutex); + + for (i = 0; i < ARRAY_SIZE(gfx_v9_4_3_ce_reg_list); i++) { + for (j = 0; j < gfx_v9_4_3_ce_reg_list[i].se_num; j++) { + for (k = 0; k < gfx_v9_4_3_ce_reg_list[i].reg_entry.reg_inst; k++) { + /* no need to select if instance number is 1 */ + if (gfx_v9_4_3_ce_reg_list[i].se_num > 1 || + gfx_v9_4_3_ce_reg_list[i].reg_entry.reg_inst > 1) + gfx_v9_4_3_xcc_select_se_sh(adev, j, 0, k, xcc_id); + + amdgpu_ras_inst_query_ras_error_count(adev, + &(gfx_v9_4_3_ce_reg_list[i].reg_entry), + 1, + gfx_v9_4_3_ras_mem_list_array[gfx_v9_4_3_ce_reg_list[i].mem_id_type].mem_id_ent, + gfx_v9_4_3_ras_mem_list_array[gfx_v9_4_3_ce_reg_list[i].mem_id_type].size, + GET_INST(GC, xcc_id), + AMDGPU_RAS_ERROR__SINGLE_CORRECTABLE, + &ce_count); + + amdgpu_ras_inst_query_ras_error_count(adev, + &(gfx_v9_4_3_ue_reg_list[i].reg_entry), + 1, + gfx_v9_4_3_ras_mem_list_array[gfx_v9_4_3_ue_reg_list[i].mem_id_type].mem_id_ent, + gfx_v9_4_3_ras_mem_list_array[gfx_v9_4_3_ue_reg_list[i].mem_id_type].size, + GET_INST(GC, xcc_id), + AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE, + &ue_count); + } + } + } + + gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, + xcc_id); + mutex_unlock(&adev->grbm_idx_mutex); + + /* the caller should make sure initialize value of + * err_data->ue_count and err_data->ce_count + */ + err_data->ce_count += ce_count; + err_data->ue_count += ue_count; +} + +static void gfx_v9_4_3_inst_reset_ras_err_count(struct amdgpu_device *adev, + void *ras_error_status, int xcc_id) +{ + uint32_t i, j, k; + + mutex_lock(&adev->grbm_idx_mutex); + + for (i = 0; i < ARRAY_SIZE(gfx_v9_4_3_ce_reg_list); i++) { + for (j = 0; j < gfx_v9_4_3_ce_reg_list[i].se_num; j++) { + for (k = 0; k < gfx_v9_4_3_ce_reg_list[i].reg_entry.reg_inst; k++) { + /* no need to select if instance number is 1 */ + if (gfx_v9_4_3_ce_reg_list[i].se_num > 1 || + gfx_v9_4_3_ce_reg_list[i].reg_entry.reg_inst > 1) + gfx_v9_4_3_xcc_select_se_sh(adev, j, 0, k, xcc_id); + + amdgpu_ras_inst_reset_ras_error_count(adev, + &(gfx_v9_4_3_ce_reg_list[i].reg_entry), + 1, + GET_INST(GC, xcc_id)); + + amdgpu_ras_inst_reset_ras_error_count(adev, + &(gfx_v9_4_3_ue_reg_list[i].reg_entry), + 1, + GET_INST(GC, xcc_id)); + } + } + } + + gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, + xcc_id); + mutex_unlock(&adev->grbm_idx_mutex); +} + +static void gfx_v9_4_3_inst_query_ea_err_status(struct amdgpu_device *adev, + int xcc_id) +{ + uint32_t i, j; + uint32_t reg_value; + + mutex_lock(&adev->grbm_idx_mutex); + + for (i = 0; i < gfx_v9_4_3_ea_err_status_regs.se_num; i++) { + for (j = 0; j < gfx_v9_4_3_ea_err_status_regs.instance; j++) { + gfx_v9_4_3_xcc_select_se_sh(adev, i, 0, j, xcc_id); + reg_value = RREG32_SOC15(GC, GET_INST(GC, xcc_id), + regGCEA_ERR_STATUS); + if (REG_GET_FIELD(reg_value, GCEA_ERR_STATUS, SDP_RDRSP_STATUS) || + REG_GET_FIELD(reg_value, GCEA_ERR_STATUS, SDP_WRRSP_STATUS) || + REG_GET_FIELD(reg_value, GCEA_ERR_STATUS, SDP_RDRSP_DATAPARITY_ERROR)) { + dev_warn(adev->dev, + "GCEA err detected at instance: %d, status: 0x%x!\n", + j, reg_value); + } + /* clear after read */ + reg_value = REG_SET_FIELD(reg_value, GCEA_ERR_STATUS, + CLEAR_ERROR_STATUS, 0x1); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regGCEA_ERR_STATUS, + reg_value); + } + } + + gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, + xcc_id); + mutex_unlock(&adev->grbm_idx_mutex); +} + +static void gfx_v9_4_3_inst_query_utc_err_status(struct amdgpu_device *adev, + int xcc_id) +{ + uint32_t data; + + data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regUTCL2_MEM_ECC_STATUS); + if (data) { + dev_warn(adev->dev, "GFX UTCL2 Mem Ecc Status: 0x%x!\n", data); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regUTCL2_MEM_ECC_STATUS, 0x3); + } + + data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regVML2_MEM_ECC_STATUS); + if (data) { + dev_warn(adev->dev, "GFX VML2 Mem Ecc Status: 0x%x!\n", data); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regVML2_MEM_ECC_STATUS, 0x3); + } + + data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), + regVML2_WALKER_MEM_ECC_STATUS); + if (data) { + dev_warn(adev->dev, "GFX VML2 Walker Mem Ecc Status: 0x%x!\n", data); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regVML2_WALKER_MEM_ECC_STATUS, + 0x3); + } +} + +static void gfx_v9_4_3_log_cu_timeout_status(struct amdgpu_device *adev, + uint32_t status, int xcc_id) +{ + struct amdgpu_cu_info *cu_info = &adev->gfx.cu_info; + uint32_t i, simd, wave; + uint32_t wave_status; + uint32_t wave_pc_lo, wave_pc_hi; + uint32_t wave_exec_lo, wave_exec_hi; + uint32_t wave_inst_dw0, wave_inst_dw1; + uint32_t wave_ib_sts; + + for (i = 0; i < 32; i++) { + if (!((i << 1) & status)) + continue; + + simd = i / cu_info->max_waves_per_simd; + wave = i % cu_info->max_waves_per_simd; + + wave_status = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_STATUS); + wave_pc_lo = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_PC_LO); + wave_pc_hi = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_PC_HI); + wave_exec_lo = + wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_EXEC_LO); + wave_exec_hi = + wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_EXEC_HI); + wave_inst_dw0 = + wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_INST_DW0); + wave_inst_dw1 = + wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_INST_DW1); + wave_ib_sts = wave_read_ind(adev, xcc_id, simd, wave, ixSQ_WAVE_IB_STS); + + dev_info( + adev->dev, + "\t SIMD %d, Wave %d: status 0x%x, pc 0x%llx, exec 0x%llx, inst 0x%llx, ib_sts 0x%x\n", + simd, wave, wave_status, + ((uint64_t)wave_pc_hi << 32 | wave_pc_lo), + ((uint64_t)wave_exec_hi << 32 | wave_exec_lo), + ((uint64_t)wave_inst_dw1 << 32 | wave_inst_dw0), + wave_ib_sts); + } +} + +static void gfx_v9_4_3_inst_query_sq_timeout_status(struct amdgpu_device *adev, + int xcc_id) +{ + uint32_t se_idx, sh_idx, cu_idx; + uint32_t status; + + mutex_lock(&adev->grbm_idx_mutex); + for (se_idx = 0; se_idx < adev->gfx.config.max_shader_engines; se_idx++) { + for (sh_idx = 0; sh_idx < adev->gfx.config.max_sh_per_se; sh_idx++) { + for (cu_idx = 0; cu_idx < adev->gfx.config.max_cu_per_sh; cu_idx++) { + gfx_v9_4_3_xcc_select_se_sh(adev, se_idx, sh_idx, + cu_idx, xcc_id); + status = RREG32_SOC15(GC, GET_INST(GC, xcc_id), + regSQ_TIMEOUT_STATUS); + if (status != 0) { + dev_info( + adev->dev, + "GFX Watchdog Timeout: SE %d, SH %d, CU %d\n", + se_idx, sh_idx, cu_idx); + gfx_v9_4_3_log_cu_timeout_status( + adev, status, xcc_id); + } + /* clear old status */ + WREG32_SOC15(GC, GET_INST(GC, xcc_id), + regSQ_TIMEOUT_STATUS, 0); + } + } + } + gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, + xcc_id); + mutex_unlock(&adev->grbm_idx_mutex); +} + +static void gfx_v9_4_3_inst_query_ras_err_status(struct amdgpu_device *adev, + void *ras_error_status, int xcc_id) +{ + gfx_v9_4_3_inst_query_ea_err_status(adev, xcc_id); + gfx_v9_4_3_inst_query_utc_err_status(adev, xcc_id); + gfx_v9_4_3_inst_query_sq_timeout_status(adev, xcc_id); +} + +static void gfx_v9_4_3_inst_reset_utc_err_status(struct amdgpu_device *adev, + int xcc_id) +{ + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regUTCL2_MEM_ECC_STATUS, 0x3); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regVML2_MEM_ECC_STATUS, 0x3); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regVML2_WALKER_MEM_ECC_STATUS, 0x3); +} + +static void gfx_v9_4_3_inst_reset_ea_err_status(struct amdgpu_device *adev, + int xcc_id) +{ + uint32_t i, j; + uint32_t value; + + mutex_lock(&adev->grbm_idx_mutex); + for (i = 0; i < gfx_v9_4_3_ea_err_status_regs.se_num; i++) { + for (j = 0; j < gfx_v9_4_3_ea_err_status_regs.instance; j++) { + gfx_v9_4_3_xcc_select_se_sh(adev, i, 0, j, xcc_id); + value = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regGCEA_ERR_STATUS); + value = REG_SET_FIELD(value, GCEA_ERR_STATUS, + CLEAR_ERROR_STATUS, 0x1); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regGCEA_ERR_STATUS, value); + } + } + gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, + xcc_id); + mutex_unlock(&adev->grbm_idx_mutex); +} + +static void gfx_v9_4_3_inst_reset_sq_timeout_status(struct amdgpu_device *adev, + int xcc_id) +{ + uint32_t se_idx, sh_idx, cu_idx; + + mutex_lock(&adev->grbm_idx_mutex); + for (se_idx = 0; se_idx < adev->gfx.config.max_shader_engines; se_idx++) { + for (sh_idx = 0; sh_idx < adev->gfx.config.max_sh_per_se; sh_idx++) { + for (cu_idx = 0; cu_idx < adev->gfx.config.max_cu_per_sh; cu_idx++) { + gfx_v9_4_3_xcc_select_se_sh(adev, se_idx, sh_idx, + cu_idx, xcc_id); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), + regSQ_TIMEOUT_STATUS, 0); + } + } + } + gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, + xcc_id); + mutex_unlock(&adev->grbm_idx_mutex); +} + +static void gfx_v9_4_3_inst_reset_ras_err_status(struct amdgpu_device *adev, + void *ras_error_status, int xcc_id) +{ + gfx_v9_4_3_inst_reset_utc_err_status(adev, xcc_id); + gfx_v9_4_3_inst_reset_ea_err_status(adev, xcc_id); + gfx_v9_4_3_inst_reset_sq_timeout_status(adev, xcc_id); +} + +static void gfx_v9_4_3_query_ras_error_count(struct amdgpu_device *adev, + void *ras_error_status) +{ + amdgpu_gfx_ras_error_func(adev, ras_error_status, + gfx_v9_4_3_inst_query_ras_err_count); +} + +static void gfx_v9_4_3_reset_ras_error_count(struct amdgpu_device *adev) +{ + amdgpu_gfx_ras_error_func(adev, NULL, gfx_v9_4_3_inst_reset_ras_err_count); +} + +static void gfx_v9_4_3_query_ras_error_status(struct amdgpu_device *adev) +{ + amdgpu_gfx_ras_error_func(adev, NULL, gfx_v9_4_3_inst_query_ras_err_status); +} + +static void gfx_v9_4_3_reset_ras_error_status(struct amdgpu_device *adev) +{ + amdgpu_gfx_ras_error_func(adev, NULL, gfx_v9_4_3_inst_reset_ras_err_status); +} + +static const struct amd_ip_funcs gfx_v9_4_3_ip_funcs = { + .name = "gfx_v9_4_3", + .early_init = gfx_v9_4_3_early_init, + .late_init = gfx_v9_4_3_late_init, + .sw_init = gfx_v9_4_3_sw_init, + .sw_fini = gfx_v9_4_3_sw_fini, + .hw_init = gfx_v9_4_3_hw_init, + .hw_fini = gfx_v9_4_3_hw_fini, + .suspend = gfx_v9_4_3_suspend, + .resume = gfx_v9_4_3_resume, + .is_idle = gfx_v9_4_3_is_idle, + .wait_for_idle = gfx_v9_4_3_wait_for_idle, + .soft_reset = gfx_v9_4_3_soft_reset, + .set_clockgating_state = gfx_v9_4_3_set_clockgating_state, + .set_powergating_state = gfx_v9_4_3_set_powergating_state, + .get_clockgating_state = gfx_v9_4_3_get_clockgating_state, +}; + +static const struct amdgpu_ring_funcs gfx_v9_4_3_ring_funcs_compute = { + .type = AMDGPU_RING_TYPE_COMPUTE, + .align_mask = 0xff, + .nop = PACKET3(PACKET3_NOP, 0x3FFF), + .support_64bit_ptrs = true, + .get_rptr = gfx_v9_4_3_ring_get_rptr_compute, + .get_wptr = gfx_v9_4_3_ring_get_wptr_compute, + .set_wptr = gfx_v9_4_3_ring_set_wptr_compute, + .emit_frame_size = + 20 + /* gfx_v9_4_3_ring_emit_gds_switch */ + 7 + /* gfx_v9_4_3_ring_emit_hdp_flush */ + 5 + /* hdp invalidate */ + 7 + /* gfx_v9_4_3_ring_emit_pipeline_sync */ + SOC15_FLUSH_GPU_TLB_NUM_WREG * 5 + + SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 7 + + 2 + /* gfx_v9_4_3_ring_emit_vm_flush */ + 8 + 8 + 8 + /* gfx_v9_4_3_ring_emit_fence x3 for user fence, vm fence */ + 7 + /* gfx_v9_4_3_emit_mem_sync */ + 5 + /* gfx_v9_4_3_emit_wave_limit for updating regSPI_WCL_PIPE_PERCENT_GFX register */ + 15, /* for updating 3 regSPI_WCL_PIPE_PERCENT_CS registers */ + .emit_ib_size = 7, /* gfx_v9_4_3_ring_emit_ib_compute */ + .emit_ib = gfx_v9_4_3_ring_emit_ib_compute, + .emit_fence = gfx_v9_4_3_ring_emit_fence, + .emit_pipeline_sync = gfx_v9_4_3_ring_emit_pipeline_sync, + .emit_vm_flush = gfx_v9_4_3_ring_emit_vm_flush, + .emit_gds_switch = gfx_v9_4_3_ring_emit_gds_switch, + .emit_hdp_flush = gfx_v9_4_3_ring_emit_hdp_flush, + .test_ring = gfx_v9_4_3_ring_test_ring, + .test_ib = gfx_v9_4_3_ring_test_ib, + .insert_nop = amdgpu_ring_insert_nop, + .pad_ib = amdgpu_ring_generic_pad_ib, + .emit_wreg = gfx_v9_4_3_ring_emit_wreg, + .emit_reg_wait = gfx_v9_4_3_ring_emit_reg_wait, + .emit_reg_write_reg_wait = gfx_v9_4_3_ring_emit_reg_write_reg_wait, + .emit_mem_sync = gfx_v9_4_3_emit_mem_sync, + .emit_wave_limit = gfx_v9_4_3_emit_wave_limit, +}; + +static const struct amdgpu_ring_funcs gfx_v9_4_3_ring_funcs_kiq = { + .type = AMDGPU_RING_TYPE_KIQ, + .align_mask = 0xff, + .nop = PACKET3(PACKET3_NOP, 0x3FFF), + .support_64bit_ptrs = true, + .get_rptr = gfx_v9_4_3_ring_get_rptr_compute, + .get_wptr = gfx_v9_4_3_ring_get_wptr_compute, + .set_wptr = gfx_v9_4_3_ring_set_wptr_compute, + .emit_frame_size = + 20 + /* gfx_v9_4_3_ring_emit_gds_switch */ + 7 + /* gfx_v9_4_3_ring_emit_hdp_flush */ + 5 + /* hdp invalidate */ + 7 + /* gfx_v9_4_3_ring_emit_pipeline_sync */ + SOC15_FLUSH_GPU_TLB_NUM_WREG * 5 + + SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 7 + + 2 + /* gfx_v9_4_3_ring_emit_vm_flush */ + 8 + 8 + 8, /* gfx_v9_4_3_ring_emit_fence_kiq x3 for user fence, vm fence */ + .emit_ib_size = 7, /* gfx_v9_4_3_ring_emit_ib_compute */ + .emit_fence = gfx_v9_4_3_ring_emit_fence_kiq, + .test_ring = gfx_v9_4_3_ring_test_ring, + .insert_nop = amdgpu_ring_insert_nop, + .pad_ib = amdgpu_ring_generic_pad_ib, + .emit_rreg = gfx_v9_4_3_ring_emit_rreg, + .emit_wreg = gfx_v9_4_3_ring_emit_wreg, + .emit_reg_wait = gfx_v9_4_3_ring_emit_reg_wait, + .emit_reg_write_reg_wait = gfx_v9_4_3_ring_emit_reg_write_reg_wait, +}; + +static void gfx_v9_4_3_set_ring_funcs(struct amdgpu_device *adev) +{ + int i, j, num_xcc; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + for (i = 0; i < num_xcc; i++) { + adev->gfx.kiq[i].ring.funcs = &gfx_v9_4_3_ring_funcs_kiq; + + for (j = 0; j < adev->gfx.num_compute_rings; j++) + adev->gfx.compute_ring[j + i * adev->gfx.num_compute_rings].funcs + = &gfx_v9_4_3_ring_funcs_compute; + } +} + +static const struct amdgpu_irq_src_funcs gfx_v9_4_3_eop_irq_funcs = { + .set = gfx_v9_4_3_set_eop_interrupt_state, + .process = gfx_v9_4_3_eop_irq, +}; + +static const struct amdgpu_irq_src_funcs gfx_v9_4_3_priv_reg_irq_funcs = { + .set = gfx_v9_4_3_set_priv_reg_fault_state, + .process = gfx_v9_4_3_priv_reg_irq, +}; + +static const struct amdgpu_irq_src_funcs gfx_v9_4_3_priv_inst_irq_funcs = { + .set = gfx_v9_4_3_set_priv_inst_fault_state, + .process = gfx_v9_4_3_priv_inst_irq, +}; + +static void gfx_v9_4_3_set_irq_funcs(struct amdgpu_device *adev) +{ + adev->gfx.eop_irq.num_types = AMDGPU_CP_IRQ_LAST; + adev->gfx.eop_irq.funcs = &gfx_v9_4_3_eop_irq_funcs; + + adev->gfx.priv_reg_irq.num_types = 1; + adev->gfx.priv_reg_irq.funcs = &gfx_v9_4_3_priv_reg_irq_funcs; + + adev->gfx.priv_inst_irq.num_types = 1; + adev->gfx.priv_inst_irq.funcs = &gfx_v9_4_3_priv_inst_irq_funcs; +} + +static void gfx_v9_4_3_set_rlc_funcs(struct amdgpu_device *adev) +{ + adev->gfx.rlc.funcs = &gfx_v9_4_3_rlc_funcs; +} + + +static void gfx_v9_4_3_set_gds_init(struct amdgpu_device *adev) +{ + /* init asci gds info */ + switch (adev->ip_versions[GC_HWIP][0]) { + case IP_VERSION(9, 4, 3): + /* 9.4.3 removed all the GDS internal memory, + * only support GWS opcode in kernel, like barrier + * semaphore.etc */ + adev->gds.gds_size = 0; + break; + default: + adev->gds.gds_size = 0x10000; + break; + } + + switch (adev->ip_versions[GC_HWIP][0]) { + case IP_VERSION(9, 4, 3): + /* deprecated for 9.4.3, no usage at all */ + adev->gds.gds_compute_max_wave_id = 0; + break; + default: + /* this really depends on the chip */ + adev->gds.gds_compute_max_wave_id = 0x7ff; + break; + } + + adev->gds.gws_size = 64; + adev->gds.oa_size = 16; +} + +static void gfx_v9_4_3_set_user_cu_inactive_bitmap(struct amdgpu_device *adev, + u32 bitmap) +{ + u32 data; + + if (!bitmap) + return; + + data = bitmap << GC_USER_SHADER_ARRAY_CONFIG__INACTIVE_CUS__SHIFT; + data &= GC_USER_SHADER_ARRAY_CONFIG__INACTIVE_CUS_MASK; + + WREG32_SOC15(GC, GET_INST(GC, 0), regGC_USER_SHADER_ARRAY_CONFIG, data); +} + +static u32 gfx_v9_4_3_get_cu_active_bitmap(struct amdgpu_device *adev) +{ + u32 data, mask; + + data = RREG32_SOC15(GC, GET_INST(GC, 0), regCC_GC_SHADER_ARRAY_CONFIG); + data |= RREG32_SOC15(GC, GET_INST(GC, 0), regGC_USER_SHADER_ARRAY_CONFIG); + + data &= CC_GC_SHADER_ARRAY_CONFIG__INACTIVE_CUS_MASK; + data >>= CC_GC_SHADER_ARRAY_CONFIG__INACTIVE_CUS__SHIFT; + + mask = amdgpu_gfx_create_bitmask(adev->gfx.config.max_cu_per_sh); + + return (~data) & mask; +} + +static int gfx_v9_4_3_get_cu_info(struct amdgpu_device *adev, + struct amdgpu_cu_info *cu_info) +{ + int i, j, k, counter, active_cu_number = 0; + u32 mask, bitmap, ao_bitmap, ao_cu_mask = 0; + unsigned disable_masks[4 * 4]; + + if (!adev || !cu_info) + return -EINVAL; + + /* + * 16 comes from bitmap array size 4*4, and it can cover all gfx9 ASICs + */ + if (adev->gfx.config.max_shader_engines * + adev->gfx.config.max_sh_per_se > 16) + return -EINVAL; + + amdgpu_gfx_parse_disable_cu(disable_masks, + adev->gfx.config.max_shader_engines, + adev->gfx.config.max_sh_per_se); + + mutex_lock(&adev->grbm_idx_mutex); + for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { + for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { + mask = 1; + ao_bitmap = 0; + counter = 0; + gfx_v9_4_3_xcc_select_se_sh(adev, i, j, 0xffffffff, 0); + gfx_v9_4_3_set_user_cu_inactive_bitmap( + adev, disable_masks[i * adev->gfx.config.max_sh_per_se + j]); + bitmap = gfx_v9_4_3_get_cu_active_bitmap(adev); + + /* + * The bitmap(and ao_cu_bitmap) in cu_info structure is + * 4x4 size array, and it's usually suitable for Vega + * ASICs which has 4*2 SE/SH layout. + * But for Arcturus, SE/SH layout is changed to 8*1. + * To mostly reduce the impact, we make it compatible + * with current bitmap array as below: + * SE4,SH0 --> bitmap[0][1] + * SE5,SH0 --> bitmap[1][1] + * SE6,SH0 --> bitmap[2][1] + * SE7,SH0 --> bitmap[3][1] + */ + cu_info->bitmap[i % 4][j + i / 4] = bitmap; + + for (k = 0; k < adev->gfx.config.max_cu_per_sh; k++) { + if (bitmap & mask) { + if (counter < adev->gfx.config.max_cu_per_sh) + ao_bitmap |= mask; + counter++; + } + mask <<= 1; + } + active_cu_number += counter; + if (i < 2 && j < 2) + ao_cu_mask |= (ao_bitmap << (i * 16 + j * 8)); + cu_info->ao_cu_bitmap[i % 4][j + i / 4] = ao_bitmap; + } + } + gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, + 0); + mutex_unlock(&adev->grbm_idx_mutex); + + cu_info->number = active_cu_number; + cu_info->ao_cu_mask = ao_cu_mask; + cu_info->simd_per_cu = NUM_SIMD_PER_CU; + + return 0; +} + +const struct amdgpu_ip_block_version gfx_v9_4_3_ip_block = { + .type = AMD_IP_BLOCK_TYPE_GFX, + .major = 9, + .minor = 4, + .rev = 0, + .funcs = &gfx_v9_4_3_ip_funcs, +}; + +static int gfx_v9_4_3_xcp_resume(void *handle, uint32_t inst_mask) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + uint32_t tmp_mask; + int i, r; + + /* TODO : Initialize golden regs */ + /* gfx_v9_4_3_init_golden_registers(adev); */ + + tmp_mask = inst_mask; + for_each_inst(i, tmp_mask) + gfx_v9_4_3_xcc_constants_init(adev, i); + + if (!amdgpu_sriov_vf(adev)) { + tmp_mask = inst_mask; + for_each_inst(i, tmp_mask) { + r = gfx_v9_4_3_xcc_rlc_resume(adev, i); + if (r) + return r; + } + } + + tmp_mask = inst_mask; + for_each_inst(i, tmp_mask) { + r = gfx_v9_4_3_xcc_cp_resume(adev, i); + if (r) + return r; + } + + return 0; +} + +static int gfx_v9_4_3_xcp_suspend(void *handle, uint32_t inst_mask) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int i; + + for_each_inst(i, inst_mask) + gfx_v9_4_3_xcc_fini(adev, i); + + return 0; +} + +struct amdgpu_xcp_ip_funcs gfx_v9_4_3_xcp_funcs = { + .suspend = &gfx_v9_4_3_xcp_suspend, + .resume = &gfx_v9_4_3_xcp_resume +}; + +struct amdgpu_ras_block_hw_ops gfx_v9_4_3_ras_ops = { + .query_ras_error_count = &gfx_v9_4_3_query_ras_error_count, + .reset_ras_error_count = &gfx_v9_4_3_reset_ras_error_count, + .query_ras_error_status = &gfx_v9_4_3_query_ras_error_status, + .reset_ras_error_status = &gfx_v9_4_3_reset_ras_error_status, +}; + +struct amdgpu_gfx_ras gfx_v9_4_3_ras = { + .ras_block = { + .hw_ops = &gfx_v9_4_3_ras_ops, + }, }; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.h b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.h index 84e69701b81a6320b59b3452529e278da90e5a77..42d67ee0e7ef64c94b806834c8b98ac05e96465b 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.h +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.h @@ -24,7 +24,8 @@ #ifndef __GFX_V9_4_3_H__ #define __GFX_V9_4_3_H__ -extern const struct amdgpu_gfx_funcs gfx_v9_4_3_gfx_funcs; -extern const struct amdgpu_rlc_funcs gfx_v9_4_3_rlc_funcs; +extern const struct amdgpu_ip_block_version gfx_v9_4_3_ip_block; + +extern struct amdgpu_xcp_ip_funcs gfx_v9_4_3_xcp_funcs; #endif /* __GFX_V9_4_3_H__ */ diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c index ab2325f6c7ac5fc36d3e2f2bcabcc6f5175599b3..d94cc1ec7242db0ce8ca0f1b265bb9f4f64fe986 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c @@ -40,7 +40,7 @@ static void gfxhub_v1_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; WREG32_SOC15_OFFSET(GC, 0, mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, hub->ctx_addr_distance * vmid, @@ -247,7 +247,7 @@ static void gfxhub_v1_0_disable_identity_aperture(struct amdgpu_device *adev) static void gfxhub_v1_0_setup_vmid_config(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; unsigned num_level, block_size; uint32_t tmp; int i; @@ -307,7 +307,7 @@ static void gfxhub_v1_0_setup_vmid_config(struct amdgpu_device *adev) static void gfxhub_v1_0_program_invalidation(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; unsigned i; for (i = 0 ; i < 18; ++i) { @@ -338,7 +338,7 @@ static int gfxhub_v1_0_gart_enable(struct amdgpu_device *adev) static void gfxhub_v1_0_gart_disable(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; u32 tmp; u32 i; @@ -411,7 +411,7 @@ static void gfxhub_v1_0_set_fault_enable_default(struct amdgpu_device *adev, static void gfxhub_v1_0_init(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; hub->ctx0_ptb_addr_lo32 = SOC15_REG_OFFSET(GC, 0, diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c index c59c6c85fbff3d7c0147d71c605ea3730bb379ce..4dabf910334b7abd23989f0a55d82946b0831816 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c @@ -21,6 +21,7 @@ * */ #include "amdgpu.h" +#include "amdgpu_xcp.h" #include "gfxhub_v1_2.h" #include "gfxhub_v1_1.h" @@ -35,227 +36,288 @@ static u64 gfxhub_v1_2_get_mc_fb_offset(struct amdgpu_device *adev) { - return (u64)RREG32_SOC15(GC, 0, regMC_VM_FB_OFFSET) << 24; + return (u64)RREG32_SOC15(GC, GET_INST(GC, 0), regMC_VM_FB_OFFSET) << 24; +} + +static void gfxhub_v1_2_xcc_setup_vm_pt_regs(struct amdgpu_device *adev, + uint32_t vmid, + uint64_t page_table_base, + uint32_t xcc_mask) +{ + struct amdgpu_vmhub *hub; + int i; + + for_each_inst(i, xcc_mask) { + hub = &adev->vmhub[AMDGPU_GFXHUB(i)]; + WREG32_SOC15_OFFSET(GC, GET_INST(GC, i), + regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, + hub->ctx_addr_distance * vmid, + lower_32_bits(page_table_base)); + + WREG32_SOC15_OFFSET(GC, GET_INST(GC, i), + regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32, + hub->ctx_addr_distance * vmid, + upper_32_bits(page_table_base)); + } } static void gfxhub_v1_2_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; - - WREG32_SOC15_OFFSET(GC, 0, regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, - hub->ctx_addr_distance * vmid, - lower_32_bits(page_table_base)); + uint32_t xcc_mask; - WREG32_SOC15_OFFSET(GC, 0, regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32, - hub->ctx_addr_distance * vmid, - upper_32_bits(page_table_base)); + xcc_mask = GENMASK(NUM_XCC(adev->gfx.xcc_mask) - 1, 0); + gfxhub_v1_2_xcc_setup_vm_pt_regs(adev, vmid, page_table_base, xcc_mask); } -static void gfxhub_v1_2_init_gart_aperture_regs(struct amdgpu_device *adev) +static void gfxhub_v1_2_xcc_init_gart_aperture_regs(struct amdgpu_device *adev, + uint32_t xcc_mask) { uint64_t pt_base; + int i; if (adev->gmc.pdb0_bo) pt_base = amdgpu_gmc_pd_addr(adev->gmc.pdb0_bo); else pt_base = amdgpu_gmc_pd_addr(adev->gart.bo); - gfxhub_v1_2_setup_vm_pt_regs(adev, 0, pt_base); + gfxhub_v1_2_xcc_setup_vm_pt_regs(adev, 0, pt_base, xcc_mask); /* If use GART for FB translation, vmid0 page table covers both * vram and system memory (gart) */ - if (adev->gmc.pdb0_bo) { - WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32, - (u32)(adev->gmc.fb_start >> 12)); - WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32, - (u32)(adev->gmc.fb_start >> 44)); - - WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32, - (u32)(adev->gmc.gart_end >> 12)); - WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32, - (u32)(adev->gmc.gart_end >> 44)); - } else { - WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32, - (u32)(adev->gmc.gart_start >> 12)); - WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32, - (u32)(adev->gmc.gart_start >> 44)); - - WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32, - (u32)(adev->gmc.gart_end >> 12)); - WREG32_SOC15(GC, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32, - (u32)(adev->gmc.gart_end >> 44)); + for_each_inst(i, xcc_mask) { + if (adev->gmc.pdb0_bo) { + WREG32_SOC15(GC, GET_INST(GC, i), + regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32, + (u32)(adev->gmc.fb_start >> 12)); + WREG32_SOC15(GC, GET_INST(GC, i), + regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32, + (u32)(adev->gmc.fb_start >> 44)); + + WREG32_SOC15(GC, GET_INST(GC, i), + regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32, + (u32)(adev->gmc.gart_end >> 12)); + WREG32_SOC15(GC, GET_INST(GC, i), + regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32, + (u32)(adev->gmc.gart_end >> 44)); + } else { + WREG32_SOC15(GC, GET_INST(GC, i), + regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32, + (u32)(adev->gmc.gart_start >> 12)); + WREG32_SOC15(GC, GET_INST(GC, i), + regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32, + (u32)(adev->gmc.gart_start >> 44)); + + WREG32_SOC15(GC, GET_INST(GC, i), + regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32, + (u32)(adev->gmc.gart_end >> 12)); + WREG32_SOC15(GC, GET_INST(GC, i), + regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32, + (u32)(adev->gmc.gart_end >> 44)); + } } } -static void gfxhub_v1_2_init_system_aperture_regs(struct amdgpu_device *adev) +static void +gfxhub_v1_2_xcc_init_system_aperture_regs(struct amdgpu_device *adev, + uint32_t xcc_mask) { uint64_t value; uint32_t tmp; + int i; - /* Program the AGP BAR */ - WREG32_SOC15_RLC(GC, 0, regMC_VM_AGP_BASE, 0); - WREG32_SOC15_RLC(GC, 0, regMC_VM_AGP_BOT, adev->gmc.agp_start >> 24); - WREG32_SOC15_RLC(GC, 0, regMC_VM_AGP_TOP, adev->gmc.agp_end >> 24); - - if (!amdgpu_sriov_vf(adev) || adev->asic_type <= CHIP_VEGA10) { - /* Program the system aperture low logical page number. */ - WREG32_SOC15_RLC(GC, 0, regMC_VM_SYSTEM_APERTURE_LOW_ADDR, - min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18); - - if (adev->apu_flags & AMD_APU_IS_RAVEN2) - /* - * Raven2 has a HW issue that it is unable to use the - * vram which is out of MC_VM_SYSTEM_APERTURE_HIGH_ADDR. - * So here is the workaround that increase system - * aperture high address (add 1) to get rid of the VM - * fault and hardware hang. - */ - WREG32_SOC15_RLC(GC, 0, - regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, - max((adev->gmc.fb_end >> 18) + 0x1, - adev->gmc.agp_end >> 18)); - else - WREG32_SOC15_RLC(GC, 0, - regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, - max(adev->gmc.fb_end, adev->gmc.agp_end) >> 18); - - /* Set default page address. */ - value = amdgpu_gmc_vram_mc2pa(adev, adev->mem_scratch.gpu_addr); - WREG32_SOC15(GC, 0, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB, - (u32)(value >> 12)); - WREG32_SOC15(GC, 0, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, - (u32)(value >> 44)); - - /* Program "protection fault". */ - WREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_LO32, - (u32)(adev->dummy_page_addr >> 12)); - WREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_HI32, - (u32)((u64)adev->dummy_page_addr >> 44)); - - tmp = RREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_CNTL2); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL2, - ACTIVE_PAGE_MIGRATION_PTE_READ_RETRY, 1); - WREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_CNTL2, tmp); - } - - /* In the case squeezing vram into GART aperture, we don't use - * FB aperture and AGP aperture. Disable them. - */ - if (adev->gmc.pdb0_bo) { - WREG32_SOC15(GC, 0, regMC_VM_FB_LOCATION_TOP, 0); - WREG32_SOC15(GC, 0, regMC_VM_FB_LOCATION_BASE, 0x00FFFFFF); - WREG32_SOC15(GC, 0, regMC_VM_AGP_TOP, 0); - WREG32_SOC15(GC, 0, regMC_VM_AGP_BOT, 0xFFFFFF); - WREG32_SOC15(GC, 0, regMC_VM_SYSTEM_APERTURE_LOW_ADDR, 0x3FFFFFFF); - WREG32_SOC15(GC, 0, regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, 0); + for_each_inst(i, xcc_mask) { + /* Program the AGP BAR */ + WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_AGP_BASE, 0); + WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_AGP_BOT, adev->gmc.agp_start >> 24); + WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_AGP_TOP, adev->gmc.agp_end >> 24); + + if (!amdgpu_sriov_vf(adev) || adev->asic_type <= CHIP_VEGA10) { + /* Program the system aperture low logical page number. */ + WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_SYSTEM_APERTURE_LOW_ADDR, + min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18); + + if (adev->apu_flags & AMD_APU_IS_RAVEN2) + /* + * Raven2 has a HW issue that it is unable to use the + * vram which is out of MC_VM_SYSTEM_APERTURE_HIGH_ADDR. + * So here is the workaround that increase system + * aperture high address (add 1) to get rid of the VM + * fault and hardware hang. + */ + WREG32_SOC15_RLC(GC, GET_INST(GC, i), + regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, + max((adev->gmc.fb_end >> 18) + 0x1, + adev->gmc.agp_end >> 18)); + else + WREG32_SOC15_RLC(GC, GET_INST(GC, i), + regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, + max(adev->gmc.fb_end, adev->gmc.agp_end) >> 18); + + /* Set default page address. */ + value = amdgpu_gmc_vram_mc2pa(adev, adev->mem_scratch.gpu_addr); + WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB, + (u32)(value >> 12)); + WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, + (u32)(value >> 44)); + + /* Program "protection fault". */ + WREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_LO32, + (u32)(adev->dummy_page_addr >> 12)); + WREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_HI32, + (u32)((u64)adev->dummy_page_addr >> 44)); + + tmp = RREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_PROTECTION_FAULT_CNTL2); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL2, + ACTIVE_PAGE_MIGRATION_PTE_READ_RETRY, 1); + WREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_PROTECTION_FAULT_CNTL2, tmp); + } + + /* In the case squeezing vram into GART aperture, we don't use + * FB aperture and AGP aperture. Disable them. + */ + if (adev->gmc.pdb0_bo) { + WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_FB_LOCATION_TOP, 0); + WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_FB_LOCATION_BASE, 0x00FFFFFF); + WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_AGP_TOP, 0); + WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_AGP_BOT, 0xFFFFFF); + WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_SYSTEM_APERTURE_LOW_ADDR, 0x3FFFFFFF); + WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, 0); + } } } -static void gfxhub_v1_2_init_tlb_regs(struct amdgpu_device *adev) +static void gfxhub_v1_2_xcc_init_tlb_regs(struct amdgpu_device *adev, + uint32_t xcc_mask) { uint32_t tmp; + int i; - /* Setup TLB control */ - tmp = RREG32_SOC15(GC, 0, regMC_VM_MX_L1_TLB_CNTL); - - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 1); - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, SYSTEM_ACCESS_MODE, 3); - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, - ENABLE_ADVANCED_DRIVER_MODEL, 1); - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, - SYSTEM_APERTURE_UNMAPPED_ACCESS, 0); - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, - MTYPE, MTYPE_UC);/* XXX for emulation. */ - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ATC_EN, 1); - - WREG32_SOC15_RLC(GC, 0, regMC_VM_MX_L1_TLB_CNTL, tmp); + for_each_inst(i, xcc_mask) { + /* Setup TLB control */ + tmp = RREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_MX_L1_TLB_CNTL); + + tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, + ENABLE_L1_TLB, 1); + tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, + SYSTEM_ACCESS_MODE, 3); + tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, + ENABLE_ADVANCED_DRIVER_MODEL, 1); + tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, + SYSTEM_APERTURE_UNMAPPED_ACCESS, 0); + tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, + MTYPE, MTYPE_UC);/* XXX for emulation. */ + tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ATC_EN, 1); + + WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_MX_L1_TLB_CNTL, tmp); + } } -static void gfxhub_v1_2_init_cache_regs(struct amdgpu_device *adev) +static void gfxhub_v1_2_xcc_init_cache_regs(struct amdgpu_device *adev, + uint32_t xcc_mask) { uint32_t tmp; + int i; - /* Setup L2 cache */ - tmp = RREG32_SOC15(GC, 0, regVM_L2_CNTL); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 1); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_FRAGMENT_PROCESSING, 1); - /* XXX for emulation, Refer to closed source code.*/ - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, L2_PDE0_CACHE_TAG_GENERATION_MODE, - 0); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, PDE_FAULT_CLASSIFICATION, 0); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, CONTEXT1_IDENTITY_ACCESS_MODE, 1); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, IDENTITY_MODE_FRAGMENT_SIZE, 0); - WREG32_SOC15_RLC(GC, 0, regVM_L2_CNTL, tmp); - - tmp = RREG32_SOC15(GC, 0, regVM_L2_CNTL2); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS, 1); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1); - WREG32_SOC15_RLC(GC, 0, regVM_L2_CNTL2, tmp); - - tmp = regVM_L2_CNTL3_DEFAULT; - if (adev->gmc.translate_further) { - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 12); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, - L2_CACHE_BIGK_FRAGMENT_SIZE, 9); - } else { - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 9); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, - L2_CACHE_BIGK_FRAGMENT_SIZE, 6); + for_each_inst(i, xcc_mask) { + /* Setup L2 cache */ + tmp = RREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_CNTL); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 1); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_FRAGMENT_PROCESSING, 1); + /* XXX for emulation, Refer to closed source code.*/ + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, L2_PDE0_CACHE_TAG_GENERATION_MODE, + 0); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, PDE_FAULT_CLASSIFICATION, 0); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, CONTEXT1_IDENTITY_ACCESS_MODE, 1); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, IDENTITY_MODE_FRAGMENT_SIZE, 0); + WREG32_SOC15_RLC(GC, GET_INST(GC, i), regVM_L2_CNTL, tmp); + + tmp = RREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_CNTL2); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS, 1); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1); + WREG32_SOC15_RLC(GC, GET_INST(GC, i), regVM_L2_CNTL2, tmp); + + tmp = regVM_L2_CNTL3_DEFAULT; + if (adev->gmc.translate_further) { + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 12); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, + L2_CACHE_BIGK_FRAGMENT_SIZE, 9); + } else { + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 9); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, + L2_CACHE_BIGK_FRAGMENT_SIZE, 6); + } + WREG32_SOC15_RLC(GC, GET_INST(GC, i), regVM_L2_CNTL3, tmp); + + tmp = regVM_L2_CNTL4_DEFAULT; + /* For AMD APP APUs setup WC memory */ + if (adev->gmc.xgmi.connected_to_cpu || adev->gmc.is_app_apu) { + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PDE_REQUEST_PHYSICAL, 1); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PTE_REQUEST_PHYSICAL, 1); + } else { + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PDE_REQUEST_PHYSICAL, 0); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PTE_REQUEST_PHYSICAL, 0); + } + WREG32_SOC15_RLC(GC, GET_INST(GC, i), regVM_L2_CNTL4, tmp); } - WREG32_SOC15_RLC(GC, 0, regVM_L2_CNTL3, tmp); - - tmp = regVM_L2_CNTL4_DEFAULT; - if (adev->gmc.xgmi.connected_to_cpu) { - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PDE_REQUEST_PHYSICAL, 1); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PTE_REQUEST_PHYSICAL, 1); - } else { - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PDE_REQUEST_PHYSICAL, 0); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, VMC_TAP_PTE_REQUEST_PHYSICAL, 0); - } - WREG32_SOC15_RLC(GC, 0, regVM_L2_CNTL4, tmp); } -static void gfxhub_v1_2_enable_system_domain(struct amdgpu_device *adev) +static void gfxhub_v1_2_xcc_enable_system_domain(struct amdgpu_device *adev, + uint32_t xcc_mask) { uint32_t tmp; + int i; - tmp = RREG32_SOC15(GC, 0, regVM_CONTEXT0_CNTL); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, ENABLE_CONTEXT, 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_DEPTH, - adev->gmc.vmid0_page_table_depth); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_BLOCK_SIZE, - adev->gmc.vmid0_page_table_block_size); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, - RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0); - WREG32_SOC15(GC, 0, regVM_CONTEXT0_CNTL, tmp); + for_each_inst(i, xcc_mask) { + tmp = RREG32_SOC15(GC, GET_INST(GC, i), regVM_CONTEXT0_CNTL); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, ENABLE_CONTEXT, 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_DEPTH, + adev->gmc.vmid0_page_table_depth); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_BLOCK_SIZE, + adev->gmc.vmid0_page_table_block_size); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, + RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0); + WREG32_SOC15(GC, GET_INST(GC, i), regVM_CONTEXT0_CNTL, tmp); + } } -static void gfxhub_v1_2_disable_identity_aperture(struct amdgpu_device *adev) +static void +gfxhub_v1_2_xcc_disable_identity_aperture(struct amdgpu_device *adev, + uint32_t xcc_mask) { - WREG32_SOC15(GC, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_LO32, - 0XFFFFFFFF); - WREG32_SOC15(GC, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_HI32, - 0x0000000F); - - WREG32_SOC15(GC, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_LO32, - 0); - WREG32_SOC15(GC, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_HI32, - 0); - - WREG32_SOC15(GC, 0, regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_LO32, 0); - WREG32_SOC15(GC, 0, regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_HI32, 0); + int i; + for_each_inst(i, xcc_mask) { + WREG32_SOC15(GC, GET_INST(GC, i), + regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_LO32, + 0XFFFFFFFF); + WREG32_SOC15(GC, GET_INST(GC, i), + regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_HI32, + 0x0000000F); + + WREG32_SOC15(GC, GET_INST(GC, i), + regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_LO32, + 0); + WREG32_SOC15(GC, GET_INST(GC, i), + regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_HI32, + 0); + + WREG32_SOC15(GC, GET_INST(GC, i), + regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_LO32, 0); + WREG32_SOC15(GC, GET_INST(GC, i), + regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_HI32, 0); + } } -static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev) +static void gfxhub_v1_2_xcc_setup_vmid_config(struct amdgpu_device *adev, + uint32_t xcc_mask) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub; unsigned num_level, block_size; uint32_t tmp; - int i; + int i, j; num_level = adev->vm_manager.num_level; block_size = adev->vm_manager.block_size; @@ -264,124 +326,205 @@ static void gfxhub_v1_2_setup_vmid_config(struct amdgpu_device *adev) else block_size -= 9; - for (i = 0; i <= 14; i++) { - tmp = RREG32_SOC15_OFFSET(GC, 0, regVM_CONTEXT1_CNTL, i); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, ENABLE_CONTEXT, 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, PAGE_TABLE_DEPTH, - num_level); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, - 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - VALID_PROTECTION_FAULT_ENABLE_DEFAULT, 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - READ_PROTECTION_FAULT_ENABLE_DEFAULT, 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - PAGE_TABLE_BLOCK_SIZE, - block_size); - /* Send no-retry XNACK on fault to suppress VM fault storm. - * On Aldebaran, XNACK can be enabled in the SQ per-process. - * Retry faults need to be enabled for that to work. - */ - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, - !adev->gmc.noretry || - adev->asic_type == CHIP_ALDEBARAN); - WREG32_SOC15_OFFSET(GC, 0, regVM_CONTEXT1_CNTL, - i * hub->ctx_distance, tmp); - WREG32_SOC15_OFFSET(GC, 0, - regVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32, - i * hub->ctx_addr_distance, 0); - WREG32_SOC15_OFFSET(GC, 0, - regVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32, - i * hub->ctx_addr_distance, 0); - WREG32_SOC15_OFFSET(GC, 0, - regVM_CONTEXT1_PAGE_TABLE_END_ADDR_LO32, - i * hub->ctx_addr_distance, - lower_32_bits(adev->vm_manager.max_pfn - 1)); - WREG32_SOC15_OFFSET(GC, 0, - regVM_CONTEXT1_PAGE_TABLE_END_ADDR_HI32, - i * hub->ctx_addr_distance, - upper_32_bits(adev->vm_manager.max_pfn - 1)); + for_each_inst(j, xcc_mask) { + hub = &adev->vmhub[AMDGPU_GFXHUB(j)]; + for (i = 0; i <= 14; i++) { + tmp = RREG32_SOC15_OFFSET(GC, GET_INST(GC, j), regVM_CONTEXT1_CNTL, i); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, ENABLE_CONTEXT, 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, PAGE_TABLE_DEPTH, + num_level); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, + 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + VALID_PROTECTION_FAULT_ENABLE_DEFAULT, 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + READ_PROTECTION_FAULT_ENABLE_DEFAULT, 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + PAGE_TABLE_BLOCK_SIZE, + block_size); + /* Send no-retry XNACK on fault to suppress VM fault storm. + * On 9.4.2 and 9.4.3, XNACK can be enabled in + * the SQ per-process. + * Retry faults need to be enabled for that to work. + */ + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, + !adev->gmc.noretry || + adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2) || + adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)); + WREG32_SOC15_OFFSET(GC, GET_INST(GC, j), regVM_CONTEXT1_CNTL, + i * hub->ctx_distance, tmp); + WREG32_SOC15_OFFSET(GC, GET_INST(GC, j), + regVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32, + i * hub->ctx_addr_distance, 0); + WREG32_SOC15_OFFSET(GC, GET_INST(GC, j), + regVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32, + i * hub->ctx_addr_distance, 0); + WREG32_SOC15_OFFSET(GC, GET_INST(GC, j), + regVM_CONTEXT1_PAGE_TABLE_END_ADDR_LO32, + i * hub->ctx_addr_distance, + lower_32_bits(adev->vm_manager.max_pfn - 1)); + WREG32_SOC15_OFFSET(GC, GET_INST(GC, j), + regVM_CONTEXT1_PAGE_TABLE_END_ADDR_HI32, + i * hub->ctx_addr_distance, + upper_32_bits(adev->vm_manager.max_pfn - 1)); + } } } -static void gfxhub_v1_2_program_invalidation(struct amdgpu_device *adev) +static void gfxhub_v1_2_xcc_program_invalidation(struct amdgpu_device *adev, + uint32_t xcc_mask) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; - unsigned i; - - for (i = 0 ; i < 18; ++i) { - WREG32_SOC15_OFFSET(GC, 0, regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32, - i * hub->eng_addr_distance, 0xffffffff); - WREG32_SOC15_OFFSET(GC, 0, regVM_INVALIDATE_ENG0_ADDR_RANGE_HI32, - i * hub->eng_addr_distance, 0x1f); + struct amdgpu_vmhub *hub; + unsigned int i, j; + + for_each_inst(j, xcc_mask) { + hub = &adev->vmhub[AMDGPU_GFXHUB(j)]; + + for (i = 0 ; i < 18; ++i) { + WREG32_SOC15_OFFSET(GC, GET_INST(GC, j), regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32, + i * hub->eng_addr_distance, 0xffffffff); + WREG32_SOC15_OFFSET(GC, GET_INST(GC, j), regVM_INVALIDATE_ENG0_ADDR_RANGE_HI32, + i * hub->eng_addr_distance, 0x1f); + } } } -static int gfxhub_v1_2_gart_enable(struct amdgpu_device *adev) +static int gfxhub_v1_2_xcc_gart_enable(struct amdgpu_device *adev, + uint32_t xcc_mask) { - if (amdgpu_sriov_vf(adev) && adev->asic_type != CHIP_ARCTURUS) { - /* - * MC_VM_FB_LOCATION_BASE/TOP is NULL for VF, becuase they are - * VF copy registers so vbios post doesn't program them, for - * SRIOV driver need to program them - */ - WREG32_SOC15_RLC(GC, 0, regMC_VM_FB_LOCATION_BASE, - adev->gmc.vram_start >> 24); - WREG32_SOC15_RLC(GC, 0, regMC_VM_FB_LOCATION_TOP, - adev->gmc.vram_end >> 24); + uint32_t tmp_mask; + int i; + + tmp_mask = xcc_mask; + /* + * MC_VM_FB_LOCATION_BASE/TOP is NULL for VF, because they are + * VF copy registers so vbios post doesn't program them, for + * SRIOV driver need to program them + */ + if (amdgpu_sriov_vf(adev)) { + for_each_inst(i, tmp_mask) { + i = ffs(tmp_mask) - 1; + WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_FB_LOCATION_BASE, + adev->gmc.vram_start >> 24); + WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_FB_LOCATION_TOP, + adev->gmc.vram_end >> 24); + } } /* GART Enable. */ - gfxhub_v1_2_init_gart_aperture_regs(adev); - gfxhub_v1_2_init_system_aperture_regs(adev); - gfxhub_v1_2_init_tlb_regs(adev); + gfxhub_v1_2_xcc_init_gart_aperture_regs(adev, xcc_mask); + gfxhub_v1_2_xcc_init_system_aperture_regs(adev, xcc_mask); + gfxhub_v1_2_xcc_init_tlb_regs(adev, xcc_mask); if (!amdgpu_sriov_vf(adev)) - gfxhub_v1_2_init_cache_regs(adev); + gfxhub_v1_2_xcc_init_cache_regs(adev, xcc_mask); - gfxhub_v1_2_enable_system_domain(adev); + gfxhub_v1_2_xcc_enable_system_domain(adev, xcc_mask); if (!amdgpu_sriov_vf(adev)) - gfxhub_v1_2_disable_identity_aperture(adev); - gfxhub_v1_2_setup_vmid_config(adev); - gfxhub_v1_2_program_invalidation(adev); + gfxhub_v1_2_xcc_disable_identity_aperture(adev, xcc_mask); + gfxhub_v1_2_xcc_setup_vmid_config(adev, xcc_mask); + gfxhub_v1_2_xcc_program_invalidation(adev, xcc_mask); return 0; } +static int gfxhub_v1_2_gart_enable(struct amdgpu_device *adev) +{ + uint32_t xcc_mask; + + xcc_mask = GENMASK(NUM_XCC(adev->gfx.xcc_mask) - 1, 0); + return gfxhub_v1_2_xcc_gart_enable(adev, xcc_mask); +} + +static void gfxhub_v1_2_xcc_gart_disable(struct amdgpu_device *adev, + uint32_t xcc_mask) +{ + struct amdgpu_vmhub *hub; + u32 tmp; + u32 i, j; + + for_each_inst(j, xcc_mask) { + hub = &adev->vmhub[AMDGPU_GFXHUB(j)]; + /* Disable all tables */ + for (i = 0; i < 16; i++) + WREG32_SOC15_OFFSET(GC, GET_INST(GC, j), regVM_CONTEXT0_CNTL, + i * hub->ctx_distance, 0); + + /* Setup TLB control */ + tmp = RREG32_SOC15(GC, GET_INST(GC, j), regMC_VM_MX_L1_TLB_CNTL); + tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 0); + tmp = REG_SET_FIELD(tmp, + MC_VM_MX_L1_TLB_CNTL, + ENABLE_ADVANCED_DRIVER_MODEL, + 0); + WREG32_SOC15_RLC(GC, GET_INST(GC, j), regMC_VM_MX_L1_TLB_CNTL, tmp); + + /* Setup L2 cache */ + tmp = RREG32_SOC15(GC, GET_INST(GC, j), regVM_L2_CNTL); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 0); + WREG32_SOC15(GC, GET_INST(GC, j), regVM_L2_CNTL, tmp); + WREG32_SOC15(GC, GET_INST(GC, j), regVM_L2_CNTL3, 0); + } +} + static void gfxhub_v1_2_gart_disable(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + uint32_t xcc_mask; + + xcc_mask = GENMASK(NUM_XCC(adev->gfx.xcc_mask) - 1, 0); + gfxhub_v1_2_xcc_gart_disable(adev, xcc_mask); +} + +static void gfxhub_v1_2_xcc_set_fault_enable_default(struct amdgpu_device *adev, + bool value, + uint32_t xcc_mask) +{ u32 tmp; - u32 i; - - /* Disable all tables */ - for (i = 0; i < 16; i++) - WREG32_SOC15_OFFSET(GC, 0, regVM_CONTEXT0_CNTL, - i * hub->ctx_distance, 0); - - /* Setup TLB control */ - tmp = RREG32_SOC15(GC, 0, regMC_VM_MX_L1_TLB_CNTL); - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 0); - tmp = REG_SET_FIELD(tmp, - MC_VM_MX_L1_TLB_CNTL, - ENABLE_ADVANCED_DRIVER_MODEL, - 0); - WREG32_SOC15_RLC(GC, 0, regMC_VM_MX_L1_TLB_CNTL, tmp); - - /* Setup L2 cache */ - tmp = RREG32_SOC15(GC, 0, regVM_L2_CNTL); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 0); - WREG32_SOC15(GC, 0, regVM_L2_CNTL, tmp); - WREG32_SOC15(GC, 0, regVM_L2_CNTL3, 0); + int i; + + for_each_inst(i, xcc_mask) { + tmp = RREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_PROTECTION_FAULT_CNTL); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + PDE1_PROTECTION_FAULT_ENABLE_DEFAULT, value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + PDE2_PROTECTION_FAULT_ENABLE_DEFAULT, value); + tmp = REG_SET_FIELD(tmp, + VM_L2_PROTECTION_FAULT_CNTL, + TRANSLATE_FURTHER_PROTECTION_FAULT_ENABLE_DEFAULT, + value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + NACK_PROTECTION_FAULT_ENABLE_DEFAULT, value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + VALID_PROTECTION_FAULT_ENABLE_DEFAULT, value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + READ_PROTECTION_FAULT_ENABLE_DEFAULT, value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value); + if (!value) { + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_NO_RETRY_FAULT, 1); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_RETRY_FAULT, 1); + } + WREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_PROTECTION_FAULT_CNTL, tmp); + } } /** @@ -393,72 +536,100 @@ static void gfxhub_v1_2_gart_disable(struct amdgpu_device *adev) static void gfxhub_v1_2_set_fault_enable_default(struct amdgpu_device *adev, bool value) { - u32 tmp; - tmp = RREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_CNTL); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - PDE1_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - PDE2_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, - VM_L2_PROTECTION_FAULT_CNTL, - TRANSLATE_FURTHER_PROTECTION_FAULT_ENABLE_DEFAULT, - value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - NACK_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - VALID_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - READ_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value); - if (!value) { - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_NO_RETRY_FAULT, 1); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_RETRY_FAULT, 1); - } - WREG32_SOC15(GC, 0, regVM_L2_PROTECTION_FAULT_CNTL, tmp); + uint32_t xcc_mask; + + xcc_mask = GENMASK(NUM_XCC(adev->gfx.xcc_mask) - 1, 0); + gfxhub_v1_2_xcc_set_fault_enable_default(adev, value, xcc_mask); } -static void gfxhub_v1_2_init(struct amdgpu_device *adev) +static void gfxhub_v1_2_xcc_init(struct amdgpu_device *adev, uint32_t xcc_mask) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub; + int i; + + for_each_inst(i, xcc_mask) { + hub = &adev->vmhub[AMDGPU_GFXHUB(i)]; - hub->ctx0_ptb_addr_lo32 = - SOC15_REG_OFFSET(GC, 0, + hub->ctx0_ptb_addr_lo32 = + SOC15_REG_OFFSET(GC, GET_INST(GC, i), regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32); - hub->ctx0_ptb_addr_hi32 = - SOC15_REG_OFFSET(GC, 0, + hub->ctx0_ptb_addr_hi32 = + SOC15_REG_OFFSET(GC, GET_INST(GC, i), regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32); - hub->vm_inv_eng0_sem = - SOC15_REG_OFFSET(GC, 0, regVM_INVALIDATE_ENG0_SEM); - hub->vm_inv_eng0_req = - SOC15_REG_OFFSET(GC, 0, regVM_INVALIDATE_ENG0_REQ); - hub->vm_inv_eng0_ack = - SOC15_REG_OFFSET(GC, 0, regVM_INVALIDATE_ENG0_ACK); - hub->vm_context0_cntl = - SOC15_REG_OFFSET(GC, 0, regVM_CONTEXT0_CNTL); - hub->vm_l2_pro_fault_status = - SOC15_REG_OFFSET(GC, 0, regVM_L2_PROTECTION_FAULT_STATUS); - hub->vm_l2_pro_fault_cntl = - SOC15_REG_OFFSET(GC, 0, regVM_L2_PROTECTION_FAULT_CNTL); - - hub->ctx_distance = regVM_CONTEXT1_CNTL - regVM_CONTEXT0_CNTL; - hub->ctx_addr_distance = regVM_CONTEXT1_PAGE_TABLE_BASE_ADDR_LO32 - - regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32; - hub->eng_distance = regVM_INVALIDATE_ENG1_REQ - regVM_INVALIDATE_ENG0_REQ; - hub->eng_addr_distance = regVM_INVALIDATE_ENG1_ADDR_RANGE_LO32 - - regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32; + hub->vm_inv_eng0_sem = + SOC15_REG_OFFSET(GC, GET_INST(GC, i), regVM_INVALIDATE_ENG0_SEM); + hub->vm_inv_eng0_req = + SOC15_REG_OFFSET(GC, GET_INST(GC, i), regVM_INVALIDATE_ENG0_REQ); + hub->vm_inv_eng0_ack = + SOC15_REG_OFFSET(GC, GET_INST(GC, i), regVM_INVALIDATE_ENG0_ACK); + hub->vm_context0_cntl = + SOC15_REG_OFFSET(GC, GET_INST(GC, i), regVM_CONTEXT0_CNTL); + hub->vm_l2_pro_fault_status = + SOC15_REG_OFFSET(GC, GET_INST(GC, i), + regVM_L2_PROTECTION_FAULT_STATUS); + hub->vm_l2_pro_fault_cntl = + SOC15_REG_OFFSET(GC, GET_INST(GC, i), regVM_L2_PROTECTION_FAULT_CNTL); + + hub->ctx_distance = regVM_CONTEXT1_CNTL - + regVM_CONTEXT0_CNTL; + hub->ctx_addr_distance = + regVM_CONTEXT1_PAGE_TABLE_BASE_ADDR_LO32 - + regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32; + hub->eng_distance = regVM_INVALIDATE_ENG1_REQ - + regVM_INVALIDATE_ENG0_REQ; + hub->eng_addr_distance = + regVM_INVALIDATE_ENG1_ADDR_RANGE_LO32 - + regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32; + } } +static void gfxhub_v1_2_init(struct amdgpu_device *adev) +{ + uint32_t xcc_mask; + + xcc_mask = GENMASK(NUM_XCC(adev->gfx.xcc_mask) - 1, 0); + gfxhub_v1_2_xcc_init(adev, xcc_mask); +} + +static int gfxhub_v1_2_get_xgmi_info(struct amdgpu_device *adev) +{ + u32 max_num_physical_nodes; + u32 max_physical_node_id; + u32 xgmi_lfb_cntl; + u32 max_region; + u64 seg_size; + + xgmi_lfb_cntl = RREG32_SOC15(GC, GET_INST(GC, 0), regMC_VM_XGMI_LFB_CNTL); + seg_size = REG_GET_FIELD( + RREG32_SOC15(GC, GET_INST(GC, 0), regMC_VM_XGMI_LFB_SIZE), + MC_VM_XGMI_LFB_SIZE, PF_LFB_SIZE) << 24; + max_region = + REG_GET_FIELD(xgmi_lfb_cntl, MC_VM_XGMI_LFB_CNTL, PF_MAX_REGION); + + + + max_num_physical_nodes = 8; + max_physical_node_id = 7; + + /* PF_MAX_REGION=0 means xgmi is disabled */ + if (max_region || adev->gmc.xgmi.connected_to_cpu) { + adev->gmc.xgmi.num_physical_nodes = max_region + 1; + + if (adev->gmc.xgmi.num_physical_nodes > max_num_physical_nodes) + return -EINVAL; + + adev->gmc.xgmi.physical_node_id = + REG_GET_FIELD(xgmi_lfb_cntl, MC_VM_XGMI_LFB_CNTL, + PF_LFB_REGION); + + if (adev->gmc.xgmi.physical_node_id > max_physical_node_id) + return -EINVAL; + + adev->gmc.xgmi.node_segment_size = seg_size; + } + + return 0; +} const struct amdgpu_gfxhub_funcs gfxhub_v1_2_funcs = { .get_mc_fb_offset = gfxhub_v1_2_get_mc_fb_offset, @@ -467,5 +638,38 @@ const struct amdgpu_gfxhub_funcs gfxhub_v1_2_funcs = { .gart_disable = gfxhub_v1_2_gart_disable, .set_fault_enable_default = gfxhub_v1_2_set_fault_enable_default, .init = gfxhub_v1_2_init, - .get_xgmi_info = gfxhub_v1_1_get_xgmi_info, + .get_xgmi_info = gfxhub_v1_2_get_xgmi_info, +}; + +static int gfxhub_v1_2_xcp_resume(void *handle, uint32_t inst_mask) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + bool value; + + if (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) + value = false; + else + value = true; + + gfxhub_v1_2_xcc_set_fault_enable_default(adev, value, inst_mask); + + if (!amdgpu_sriov_vf(adev)) + return gfxhub_v1_2_xcc_gart_enable(adev, inst_mask); + + return 0; +} + +static int gfxhub_v1_2_xcp_suspend(void *handle, uint32_t inst_mask) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + if (!amdgpu_sriov_vf(adev)) + gfxhub_v1_2_xcc_gart_disable(adev, inst_mask); + + return 0; +} + +struct amdgpu_xcp_ip_funcs gfxhub_v1_2_xcp_funcs = { + .suspend = &gfxhub_v1_2_xcp_suspend, + .resume = &gfxhub_v1_2_xcp_resume }; diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.h b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.h index e2d508f5a7ee3b5b82f2c794448e7f99265afe2e..997e9f90c9900679ba07da37a2ec01461d5d189f 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.h +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.h @@ -26,4 +26,6 @@ extern const struct amdgpu_gfxhub_funcs gfxhub_v1_2_funcs; +extern struct amdgpu_xcp_ip_funcs gfxhub_v1_2_xcp_funcs; + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c index 9b3a0252731818dee104f1011f73e56c5603a37b..f173a61c6c15af3159c55fd3958fd0eb25ae0cfc 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c @@ -120,7 +120,7 @@ static u64 gfxhub_v2_0_get_mc_fb_offset(struct amdgpu_device *adev) static void gfxhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; WREG32_SOC15_OFFSET(GC, 0, mmGCVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, hub->ctx_addr_distance * vmid, @@ -282,7 +282,7 @@ static void gfxhub_v2_0_disable_identity_aperture(struct amdgpu_device *adev) static void gfxhub_v2_0_setup_vmid_config(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; int i; uint32_t tmp; @@ -331,7 +331,7 @@ static void gfxhub_v2_0_setup_vmid_config(struct amdgpu_device *adev) static void gfxhub_v2_0_program_invalidation(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; unsigned i; for (i = 0 ; i < 18; ++i) { @@ -360,7 +360,7 @@ static int gfxhub_v2_0_gart_enable(struct amdgpu_device *adev) static void gfxhub_v2_0_gart_disable(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; u32 tmp; u32 i; @@ -433,7 +433,7 @@ static const struct amdgpu_vmhub_funcs gfxhub_v2_0_vmhub_funcs = { static void gfxhub_v2_0_init(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; hub->ctx0_ptb_addr_lo32 = SOC15_REG_OFFSET(GC, 0, diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c index 4aacbbec31e28a85d918e5c4a10dab28a716587c..d8fc3e8088cd003104175a5bf23a9d2b63331431 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c @@ -123,7 +123,7 @@ static u64 gfxhub_v2_1_get_mc_fb_offset(struct amdgpu_device *adev) static void gfxhub_v2_1_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; WREG32_SOC15_OFFSET(GC, 0, mmGCVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, hub->ctx_addr_distance * vmid, @@ -291,7 +291,7 @@ static void gfxhub_v2_1_disable_identity_aperture(struct amdgpu_device *adev) static void gfxhub_v2_1_setup_vmid_config(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; int i; uint32_t tmp; @@ -340,7 +340,7 @@ static void gfxhub_v2_1_setup_vmid_config(struct amdgpu_device *adev) static void gfxhub_v2_1_program_invalidation(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; unsigned i; for (i = 0 ; i < 18; ++i) { @@ -381,7 +381,7 @@ static int gfxhub_v2_1_gart_enable(struct amdgpu_device *adev) static void gfxhub_v2_1_gart_disable(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; u32 tmp; u32 i; @@ -462,7 +462,7 @@ static const struct amdgpu_vmhub_funcs gfxhub_v2_1_vmhub_funcs = { static void gfxhub_v2_1_init(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; hub->ctx0_ptb_addr_lo32 = SOC15_REG_OFFSET(GC, 0, @@ -651,7 +651,7 @@ static void gfxhub_v2_1_restore_regs(struct amdgpu_device *adev) static void gfxhub_v2_1_halt(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; int i; uint32_t tmp; int time = 1000; diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c index 13712640fa46fa628f4cee14332705d7785da653..c53147f9c9fc0b2653fad912558a1d8848510fa9 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c @@ -119,7 +119,7 @@ static u64 gfxhub_v3_0_get_mc_fb_offset(struct amdgpu_device *adev) static void gfxhub_v3_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; WREG32_SOC15_OFFSET(GC, 0, regGCVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, hub->ctx_addr_distance * vmid, @@ -290,7 +290,7 @@ static void gfxhub_v3_0_disable_identity_aperture(struct amdgpu_device *adev) static void gfxhub_v3_0_setup_vmid_config(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; int i; uint32_t tmp; @@ -339,7 +339,7 @@ static void gfxhub_v3_0_setup_vmid_config(struct amdgpu_device *adev) static void gfxhub_v3_0_program_invalidation(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; unsigned i; for (i = 0 ; i < 18; ++i) { @@ -380,7 +380,7 @@ static int gfxhub_v3_0_gart_enable(struct amdgpu_device *adev) static void gfxhub_v3_0_gart_disable(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; u32 tmp; u32 i; @@ -463,7 +463,7 @@ static const struct amdgpu_vmhub_funcs gfxhub_v3_0_vmhub_funcs = { static void gfxhub_v3_0_init(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; hub->ctx0_ptb_addr_lo32 = SOC15_REG_OFFSET(GC, 0, diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c index 6e0bd628c8895a927017091d65f18ce326b3a5c9..ae777487d72efc3ef5871e037af03f02765e4ed8 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c @@ -122,7 +122,7 @@ static u64 gfxhub_v3_0_3_get_mc_fb_offset(struct amdgpu_device *adev) static void gfxhub_v3_0_3_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; WREG32_SOC15_OFFSET(GC, 0, regGCVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, hub->ctx_addr_distance * vmid, @@ -295,7 +295,7 @@ static void gfxhub_v3_0_3_disable_identity_aperture(struct amdgpu_device *adev) static void gfxhub_v3_0_3_setup_vmid_config(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; int i; uint32_t tmp; @@ -344,7 +344,7 @@ static void gfxhub_v3_0_3_setup_vmid_config(struct amdgpu_device *adev) static void gfxhub_v3_0_3_program_invalidation(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; unsigned i; for (i = 0 ; i < 18; ++i) { @@ -373,7 +373,7 @@ static int gfxhub_v3_0_3_gart_enable(struct amdgpu_device *adev) static void gfxhub_v3_0_3_gart_disable(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; u32 tmp; u32 i; @@ -451,7 +451,7 @@ static const struct amdgpu_vmhub_funcs gfxhub_v3_0_3_vmhub_funcs = { static void gfxhub_v3_0_3_init(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; hub->ctx0_ptb_addr_lo32 = SOC15_REG_OFFSET(GC, 0, diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index b213dcf8ca06bbd5abf3f930ce546400eeb1a14b..0c8a479895761e46bc2563da43a9fe318a9b5b4d 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -76,7 +76,7 @@ gmc_v10_0_vm_fault_interrupt_state(struct amdgpu_device *adev, switch (state) { case AMDGPU_IRQ_STATE_DISABLE: /* MM HUB */ - amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB_0, false); + amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB0(0), false); /* GFX HUB */ /* This works because this interrupt is only * enabled at init/resume and disabled in @@ -84,11 +84,11 @@ gmc_v10_0_vm_fault_interrupt_state(struct amdgpu_device *adev, * change over the course of suspend/resume. */ if (!adev->in_s0ix) - amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB_0, false); + amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB(0), false); break; case AMDGPU_IRQ_STATE_ENABLE: /* MM HUB */ - amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB_0, true); + amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB0(0), true); /* GFX HUB */ /* This works because this interrupt is only * enabled at init/resume and disabled in @@ -96,7 +96,7 @@ gmc_v10_0_vm_fault_interrupt_state(struct amdgpu_device *adev, * change over the course of suspend/resume. */ if (!adev->in_s0ix) - amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB_0, true); + amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB(0), true); break; default: break; @@ -139,7 +139,7 @@ static int gmc_v10_0_process_interrupt(struct amdgpu_device *adev, /* Try to handle the recoverable page faults by filling page * tables */ - if (amdgpu_vm_handle_fault(adev, entry->pasid, addr, write_fault)) + if (amdgpu_vm_handle_fault(adev, entry->pasid, 0, 0, addr, write_fault)) return 1; } @@ -149,7 +149,7 @@ static int gmc_v10_0_process_interrupt(struct amdgpu_device *adev, * be updated to avoid reading an incorrect value due to * the new fast GRBM interface. */ - if ((entry->vmid_src == AMDGPU_GFXHUB_0) && + if ((entry->vmid_src == AMDGPU_GFXHUB(0)) && (adev->ip_versions[GC_HWIP][0] < IP_VERSION(10, 3, 0))) RREG32(hub->vm_l2_pro_fault_status); @@ -212,8 +212,7 @@ static void gmc_v10_0_set_irq_funcs(struct amdgpu_device *adev) static bool gmc_v10_0_use_invalidate_semaphore(struct amdgpu_device *adev, uint32_t vmhub) { - return ((vmhub == AMDGPU_MMHUB_0 || - vmhub == AMDGPU_MMHUB_1) && + return ((vmhub == AMDGPU_MMHUB0(0)) && (!amdgpu_sriov_vf(adev))); } @@ -249,7 +248,7 @@ static void gmc_v10_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid, unsigned int i; unsigned char hub_ip = 0; - hub_ip = (vmhub == AMDGPU_GFXHUB_0) ? + hub_ip = (vmhub == AMDGPU_GFXHUB(0)) ? GC_HWIP : MMHUB_HWIP; spin_lock(&adev->gmc.invalidate_lock); @@ -284,7 +283,7 @@ static void gmc_v10_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid, * Issue a dummy read to wait for the ACK register to be cleared * to avoid a false ACK due to the new fast GRBM interface. */ - if ((vmhub == AMDGPU_GFXHUB_0) && + if ((vmhub == AMDGPU_GFXHUB(0)) && (adev->ip_versions[GC_HWIP][0] < IP_VERSION(10, 3, 0))) RREG32_RLC_NO_KIQ(hub->vm_inv_eng0_req + hub->eng_distance * eng, hub_ip); @@ -343,7 +342,7 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, /* For SRIOV run time, driver shouldn't access the register through MMIO * Directly use kiq to do the vm invalidation instead */ - if (adev->gfx.kiq.ring.sched.ready && !adev->enable_mes && + if (adev->gfx.kiq[0].ring.sched.ready && !adev->enable_mes && (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev)) && down_read_trylock(&adev->reset_domain->sem)) { struct amdgpu_vmhub *hub = &adev->vmhub[vmhub]; @@ -361,19 +360,19 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, mutex_lock(&adev->mman.gtt_window_lock); - if (vmhub == AMDGPU_MMHUB_0) { - gmc_v10_0_flush_vm_hub(adev, vmid, AMDGPU_MMHUB_0, 0); + if (vmhub == AMDGPU_MMHUB0(0)) { + gmc_v10_0_flush_vm_hub(adev, vmid, AMDGPU_MMHUB0(0), 0); mutex_unlock(&adev->mman.gtt_window_lock); return; } - BUG_ON(vmhub != AMDGPU_GFXHUB_0); + BUG_ON(vmhub != AMDGPU_GFXHUB(0)); if (!adev->mman.buffer_funcs_enabled || !adev->ib_pool_ready || amdgpu_in_reset(adev) || ring->sched.ready == false) { - gmc_v10_0_flush_vm_hub(adev, vmid, AMDGPU_GFXHUB_0, 0); + gmc_v10_0_flush_vm_hub(adev, vmid, AMDGPU_GFXHUB(0), 0); mutex_unlock(&adev->mman.gtt_window_lock); return; } @@ -383,7 +382,7 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, * translation. Avoid this by doing the invalidation from the SDMA * itself. */ - r = amdgpu_job_alloc_with_ib(ring->adev, &adev->mman.entity, + r = amdgpu_job_alloc_with_ib(ring->adev, &adev->mman.high_pr, AMDGPU_FENCE_OWNER_UNDEFINED, 16 * 4, AMDGPU_IB_POOL_IMMEDIATE, &job); @@ -415,12 +414,13 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, * @pasid: pasid to be flush * @flush_type: the flush type * @all_hub: Used with PACKET3_INVALIDATE_TLBS_ALL_HUB() + * @inst: is used to select which instance of KIQ to use for the invalidation * * Flush the TLB for the requested pasid. */ static int gmc_v10_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t pasid, uint32_t flush_type, - bool all_hub) + bool all_hub, uint32_t inst) { int vmid, i; signed long r; @@ -428,11 +428,11 @@ static int gmc_v10_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t queried_pasid; bool ret; u32 usec_timeout = amdgpu_sriov_vf(adev) ? SRIOV_USEC_TIMEOUT : adev->usec_timeout; - struct amdgpu_ring *ring = &adev->gfx.kiq.ring; - struct amdgpu_kiq *kiq = &adev->gfx.kiq; + struct amdgpu_ring *ring = &adev->gfx.kiq[0].ring; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; if (amdgpu_emu_mode == 0 && ring->sched.ready) { - spin_lock(&adev->gfx.kiq.ring_lock); + spin_lock(&adev->gfx.kiq[0].ring_lock); /* 2 dwords flush + 8 dwords fence */ amdgpu_ring_alloc(ring, kiq->pmf->invalidate_tlbs_size + 8); kiq->pmf->kiq_invalidate_tlbs(ring, @@ -440,12 +440,12 @@ static int gmc_v10_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT); if (r) { amdgpu_ring_undo(ring); - spin_unlock(&adev->gfx.kiq.ring_lock); + spin_unlock(&adev->gfx.kiq[0].ring_lock); return -ETIME; } amdgpu_ring_commit(ring); - spin_unlock(&adev->gfx.kiq.ring_lock); + spin_unlock(&adev->gfx.kiq[0].ring_lock); r = amdgpu_fence_wait_polling(ring, seq, usec_timeout); if (r < 1) { dev_err(adev->dev, "wait for kiq fence error: %ld.\n", r); @@ -461,12 +461,12 @@ static int gmc_v10_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, &queried_pasid); if (ret && queried_pasid == pasid) { if (all_hub) { - for (i = 0; i < adev->num_vmhubs; i++) + for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) gmc_v10_0_flush_gpu_tlb(adev, vmid, i, flush_type); } else { gmc_v10_0_flush_gpu_tlb(adev, vmid, - AMDGPU_GFXHUB_0, flush_type); + AMDGPU_GFXHUB(0), flush_type); } if (!adev->enable_mes) break; @@ -534,7 +534,7 @@ static void gmc_v10_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid if (ring->is_mes_queue) return; - if (ring->vm_hub == AMDGPU_GFXHUB_0) + if (ring->vm_hub == AMDGPU_GFXHUB(0)) reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_VMID_0_LUT) + vmid; else reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_VMID_0_LUT_MM) + vmid; @@ -929,7 +929,8 @@ static int gmc_v10_0_sw_init(void *handle) case IP_VERSION(10, 3, 6): case IP_VERSION(10, 3, 3): case IP_VERSION(10, 3, 7): - adev->num_vmhubs = 2; + set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask); + set_bit(AMDGPU_MMHUB0(0), adev->vmhubs_mask); /* * To fulfill 4-level page support, * vm size is 256TB (48bit), maximum size of Navi10/Navi14/Navi12, @@ -1075,9 +1076,9 @@ static int gmc_v10_0_gart_enable(struct amdgpu_device *adev) if (!adev->in_s0ix) adev->gfxhub.funcs->set_fault_enable_default(adev, value); adev->mmhub.funcs->set_fault_enable_default(adev, value); - gmc_v10_0_flush_gpu_tlb(adev, 0, AMDGPU_MMHUB_0, 0); + gmc_v10_0_flush_gpu_tlb(adev, 0, AMDGPU_MMHUB0(0), 0); if (!adev->in_s0ix) - gmc_v10_0_flush_gpu_tlb(adev, 0, AMDGPU_GFXHUB_0, 0); + gmc_v10_0_flush_gpu_tlb(adev, 0, AMDGPU_GFXHUB(0), 0); DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n", (unsigned)(adev->gmc.gart_size >> 20), diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c index d95f9fe8f1c54ef351d36a567fd8a4572038b339..c571f0d95994628916fd02e7db562f7a01cc98e9 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c @@ -31,6 +31,8 @@ #include "umc_v8_10.h" #include "athub/athub_3_0_0_sh_mask.h" #include "athub/athub_3_0_0_offset.h" +#include "dcn/dcn_3_2_0_offset.h" +#include "dcn/dcn_3_2_0_sh_mask.h" #include "oss/osssys_6_0_0_offset.h" #include "ivsrcid/vmc/irqsrcs_vmc_1_0.h" #include "navi10_enum.h" @@ -62,7 +64,7 @@ gmc_v11_0_vm_fault_interrupt_state(struct amdgpu_device *adev, switch (state) { case AMDGPU_IRQ_STATE_DISABLE: /* MM HUB */ - amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB_0, false); + amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB0(0), false); /* GFX HUB */ /* This works because this interrupt is only * enabled at init/resume and disabled in @@ -70,11 +72,11 @@ gmc_v11_0_vm_fault_interrupt_state(struct amdgpu_device *adev, * change over the course of suspend/resume. */ if (!adev->in_s0ix) - amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB_0, false); + amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB(0), false); break; case AMDGPU_IRQ_STATE_ENABLE: /* MM HUB */ - amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB_0, true); + amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_MMHUB0(0), true); /* GFX HUB */ /* This works because this interrupt is only * enabled at init/resume and disabled in @@ -82,7 +84,7 @@ gmc_v11_0_vm_fault_interrupt_state(struct amdgpu_device *adev, * change over the course of suspend/resume. */ if (!adev->in_s0ix) - amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB_0, true); + amdgpu_gmc_set_vm_fault_masks(adev, AMDGPU_GFXHUB(0), true); break; default: break; @@ -108,7 +110,7 @@ static int gmc_v11_0_process_interrupt(struct amdgpu_device *adev, * be updated to avoid reading an incorrect value due to * the new fast GRBM interface. */ - if (entry->vmid_src == AMDGPU_GFXHUB_0) + if (entry->vmid_src == AMDGPU_GFXHUB(0)) RREG32(hub->vm_l2_pro_fault_status); status = RREG32(hub->vm_l2_pro_fault_status); @@ -168,7 +170,7 @@ static void gmc_v11_0_set_irq_funcs(struct amdgpu_device *adev) static bool gmc_v11_0_use_invalidate_semaphore(struct amdgpu_device *adev, uint32_t vmhub) { - return ((vmhub == AMDGPU_MMHUB_0) && + return ((vmhub == AMDGPU_MMHUB0(0)) && (!amdgpu_sriov_vf(adev))); } @@ -200,7 +202,7 @@ static void gmc_v11_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid, unsigned int i; unsigned char hub_ip = 0; - hub_ip = (vmhub == AMDGPU_GFXHUB_0) ? + hub_ip = (vmhub == AMDGPU_GFXHUB(0)) ? GC_HWIP : MMHUB_HWIP; spin_lock(&adev->gmc.invalidate_lock); @@ -249,7 +251,7 @@ static void gmc_v11_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid, hub->eng_distance * eng, 0, hub_ip); /* Issue additional private vm invalidation to MMHUB */ - if ((vmhub != AMDGPU_GFXHUB_0) && + if ((vmhub != AMDGPU_GFXHUB(0)) && (hub->vm_l2_bank_select_reserved_cid2) && !amdgpu_sriov_vf(adev)) { inv_req = RREG32_NO_KIQ(hub->vm_l2_bank_select_reserved_cid2); @@ -282,7 +284,7 @@ static void gmc_v11_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid, static void gmc_v11_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, uint32_t vmhub, uint32_t flush_type) { - if ((vmhub == AMDGPU_GFXHUB_0) && !adev->gfx.is_poweron) + if ((vmhub == AMDGPU_GFXHUB(0)) && !adev->gfx.is_poweron) return; /* flush hdp cache */ @@ -291,7 +293,7 @@ static void gmc_v11_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, /* For SRIOV run time, driver shouldn't access the register through MMIO * Directly use kiq to do the vm invalidation instead */ - if ((adev->gfx.kiq.ring.sched.ready || adev->mes.ring.sched.ready) && + if ((adev->gfx.kiq[0].ring.sched.ready || adev->mes.ring.sched.ready) && (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev))) { struct amdgpu_vmhub *hub = &adev->vmhub[vmhub]; const unsigned eng = 17; @@ -317,23 +319,24 @@ static void gmc_v11_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, * @pasid: pasid to be flush * @flush_type: the flush type * @all_hub: flush all hubs + * @inst: is used to select which instance of KIQ to use for the invalidation * * Flush the TLB for the requested pasid. */ static int gmc_v11_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t pasid, uint32_t flush_type, - bool all_hub) + bool all_hub, uint32_t inst) { int vmid, i; signed long r; uint32_t seq; uint16_t queried_pasid; bool ret; - struct amdgpu_ring *ring = &adev->gfx.kiq.ring; - struct amdgpu_kiq *kiq = &adev->gfx.kiq; + struct amdgpu_ring *ring = &adev->gfx.kiq[0].ring; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; if (amdgpu_emu_mode == 0 && ring->sched.ready) { - spin_lock(&adev->gfx.kiq.ring_lock); + spin_lock(&adev->gfx.kiq[0].ring_lock); /* 2 dwords flush + 8 dwords fence */ amdgpu_ring_alloc(ring, kiq->pmf->invalidate_tlbs_size + 8); kiq->pmf->kiq_invalidate_tlbs(ring, @@ -341,12 +344,12 @@ static int gmc_v11_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT); if (r) { amdgpu_ring_undo(ring); - spin_unlock(&adev->gfx.kiq.ring_lock); + spin_unlock(&adev->gfx.kiq[0].ring_lock); return -ETIME; } amdgpu_ring_commit(ring); - spin_unlock(&adev->gfx.kiq.ring_lock); + spin_unlock(&adev->gfx.kiq[0].ring_lock); r = amdgpu_fence_wait_polling(ring, seq, adev->usec_timeout); if (r < 1) { dev_err(adev->dev, "wait for kiq fence error: %ld.\n", r); @@ -362,12 +365,12 @@ static int gmc_v11_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, &queried_pasid); if (ret && queried_pasid == pasid) { if (all_hub) { - for (i = 0; i < adev->num_vmhubs; i++) + for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) gmc_v11_0_flush_gpu_tlb(adev, vmid, i, flush_type); } else { gmc_v11_0_flush_gpu_tlb(adev, vmid, - AMDGPU_GFXHUB_0, flush_type); + AMDGPU_GFXHUB(0), flush_type); } } } @@ -433,7 +436,7 @@ static void gmc_v11_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid if (ring->is_mes_queue) return; - if (ring->vm_hub == AMDGPU_GFXHUB_0) + if (ring->vm_hub == AMDGPU_GFXHUB(0)) reg = SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_0_LUT) + vmid; else reg = SOC15_REG_OFFSET(OSSSYS, 0, regIH_VMID_0_LUT_MM) + vmid; @@ -546,7 +549,24 @@ static void gmc_v11_0_get_vm_pte(struct amdgpu_device *adev, static unsigned gmc_v11_0_get_vbios_fb_size(struct amdgpu_device *adev) { - return 0; + u32 d1vga_control = RREG32_SOC15(DCE, 0, regD1VGA_CONTROL); + unsigned size; + + if (REG_GET_FIELD(d1vga_control, D1VGA_CONTROL, D1VGA_MODE_ENABLE)) { + size = AMDGPU_VBIOS_VGA_ALLOCATION; + } else { + u32 viewport; + u32 pitch; + + viewport = RREG32_SOC15(DCE, 0, regHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION); + pitch = RREG32_SOC15(DCE, 0, regHUBPREQ0_DCSURF_SURFACE_PITCH); + size = (REG_GET_FIELD(viewport, + HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION, PRI_VIEWPORT_HEIGHT) * + REG_GET_FIELD(pitch, HUBPREQ0_DCSURF_SURFACE_PITCH, PITCH) * + 4); + } + + return size; } static const struct amdgpu_gmc_funcs gmc_v11_0_gmc_funcs = { @@ -760,7 +780,8 @@ static int gmc_v11_0_sw_init(void *handle) case IP_VERSION(11, 0, 2): case IP_VERSION(11, 0, 3): case IP_VERSION(11, 0, 4): - adev->num_vmhubs = 2; + set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask); + set_bit(AMDGPU_MMHUB0(0), adev->vmhubs_mask); /* * To fulfill 4-level page support, * vm size is 256TB (48bit), maximum size, @@ -867,7 +888,7 @@ static int gmc_v11_0_sw_fini(void *handle) static void gmc_v11_0_init_golden_registers(struct amdgpu_device *adev) { if (amdgpu_sriov_vf(adev)) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; WREG32(hub->vm_contexts_disable, 0); return; @@ -902,7 +923,7 @@ static int gmc_v11_0_gart_enable(struct amdgpu_device *adev) false : true; adev->mmhub.funcs->set_fault_enable_default(adev, value); - gmc_v11_0_flush_gpu_tlb(adev, 0, AMDGPU_MMHUB_0, 0); + gmc_v11_0_flush_gpu_tlb(adev, 0, AMDGPU_MMHUB0(0), 0); DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n", (unsigned)(adev->gmc.gart_size >> 20), diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c index b7dad4e678135434632ee942e1d580c0e0d559ff..aa754c95a0b3aaee667e16d0381ae1714f5e0fc1 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c @@ -808,7 +808,7 @@ static int gmc_v6_0_sw_init(void *handle) int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - adev->num_vmhubs = 1; + set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask); if (adev->flags & AMD_IS_APU) { adev->gmc.vram_type = AMDGPU_VRAM_TYPE_UNKNOWN; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c index 402960b0174e2b0d637c1ca115b2bf720e29879c..acd2b407860f7bc084250e35476100482c053e66 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c @@ -419,12 +419,13 @@ static int gmc_v7_0_mc_init(struct amdgpu_device *adev) * @pasid: pasid to be flush * @flush_type: type of flush * @all_hub: flush all hubs + * @inst: is used to select which instance of KIQ to use for the invalidation * * Flush the TLB for the requested pasid. */ static int gmc_v7_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t pasid, uint32_t flush_type, - bool all_hub) + bool all_hub, uint32_t inst) { int vmid; unsigned int tmp; @@ -977,7 +978,7 @@ static int gmc_v7_0_sw_init(void *handle) int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - adev->num_vmhubs = 1; + set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask); if (adev->flags & AMD_IS_APU) { adev->gmc.vram_type = AMDGPU_VRAM_TYPE_UNKNOWN; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index 504c1b34dab7de8f171d94e6723f46d2bf7f8608..85dead2a57021cf98ae6901465f781f3f4a13ffe 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -617,12 +617,13 @@ static int gmc_v8_0_mc_init(struct amdgpu_device *adev) * @pasid: pasid to be flush * @flush_type: type of flush * @all_hub: flush all hubs + * @inst: is used to select which instance of KIQ to use for the invalidation * * Flush the TLB for the requested pasid. */ static int gmc_v8_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t pasid, uint32_t flush_type, - bool all_hub) + bool all_hub, uint32_t inst) { int vmid; unsigned int tmp; @@ -1093,7 +1094,7 @@ static int gmc_v8_0_sw_init(void *handle) int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - adev->num_vmhubs = 1; + set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask); if (adev->flags & AMD_IS_APU) { adev->gmc.vram_type = AMDGPU_VRAM_TYPE_UNKNOWN; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 2fe21cefd772f35db714958fb782d05d722b2f43..67e669e0141cc0378a8620fb5f344b11ad368e6f 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -79,6 +79,7 @@ #define mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION_DCN2 0x05ea #define mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION_DCN2_BASE_IDX 2 +#define MAX_MEM_RANGES 8 static const char *gfxhub_client_ids[] = { "CB", @@ -481,7 +482,7 @@ static int gmc_v9_0_vm_fault_interrupt_state(struct amdgpu_device *adev, switch (state) { case AMDGPU_IRQ_STATE_DISABLE: - for (j = 0; j < adev->num_vmhubs; j++) { + for_each_set_bit(j, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) { hub = &adev->vmhub[j]; for (i = 0; i < 16; i++) { reg = hub->vm_context0_cntl + i; @@ -491,25 +492,25 @@ static int gmc_v9_0_vm_fault_interrupt_state(struct amdgpu_device *adev, * fini/suspend, so the overall state doesn't * change over the course of suspend/resume. */ - if (adev->in_s0ix && (j == AMDGPU_GFXHUB_0)) + if (adev->in_s0ix && (j == AMDGPU_GFXHUB(0))) continue; - if (j == AMDGPU_GFXHUB_0) - tmp = RREG32_SOC15_IP(GC, reg); - else + if (j >= AMDGPU_MMHUB0(0)) tmp = RREG32_SOC15_IP(MMHUB, reg); + else + tmp = RREG32_SOC15_IP(GC, reg); tmp &= ~bits; - if (j == AMDGPU_GFXHUB_0) - WREG32_SOC15_IP(GC, reg, tmp); - else + if (j >= AMDGPU_MMHUB0(0)) WREG32_SOC15_IP(MMHUB, reg, tmp); + else + WREG32_SOC15_IP(GC, reg, tmp); } } break; case AMDGPU_IRQ_STATE_ENABLE: - for (j = 0; j < adev->num_vmhubs; j++) { + for_each_set_bit(j, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) { hub = &adev->vmhub[j]; for (i = 0; i < 16; i++) { reg = hub->vm_context0_cntl + i; @@ -519,20 +520,20 @@ static int gmc_v9_0_vm_fault_interrupt_state(struct amdgpu_device *adev, * fini/suspend, so the overall state doesn't * change over the course of suspend/resume. */ - if (adev->in_s0ix && (j == AMDGPU_GFXHUB_0)) + if (adev->in_s0ix && (j == AMDGPU_GFXHUB(0))) continue; - if (j == AMDGPU_GFXHUB_0) - tmp = RREG32_SOC15_IP(GC, reg); - else + if (j >= AMDGPU_MMHUB0(0)) tmp = RREG32_SOC15_IP(MMHUB, reg); + else + tmp = RREG32_SOC15_IP(GC, reg); tmp |= bits; - if (j == AMDGPU_GFXHUB_0) - WREG32_SOC15_IP(GC, reg, tmp); - else + if (j >= AMDGPU_MMHUB0(0)) WREG32_SOC15_IP(MMHUB, reg, tmp); + else + WREG32_SOC15_IP(GC, reg, tmp); } } break; @@ -556,11 +557,31 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, const char *hub_name; u64 addr; uint32_t cam_index = 0; - int ret; + int ret, xcc_id = 0; + uint32_t node_id; + + node_id = entry->node_id; addr = (u64)entry->src_data[0] << 12; addr |= ((u64)entry->src_data[1] & 0xf) << 44; + if (entry->client_id == SOC15_IH_CLIENTID_VMC) { + hub_name = "mmhub0"; + hub = &adev->vmhub[AMDGPU_MMHUB0(node_id / 4)]; + } else if (entry->client_id == SOC15_IH_CLIENTID_VMC1) { + hub_name = "mmhub1"; + hub = &adev->vmhub[AMDGPU_MMHUB1(0)]; + } else { + hub_name = "gfxhub0"; + if (adev->gfx.funcs->ih_node_to_logical_xcc) { + xcc_id = adev->gfx.funcs->ih_node_to_logical_xcc(adev, + node_id); + if (xcc_id < 0) + xcc_id = 0; + } + hub = &adev->vmhub[xcc_id]; + } + if (retry_fault) { if (adev->irq.retry_cam_enabled) { /* Delegate it to a different ring if the hardware hasn't @@ -573,7 +594,8 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, cam_index = entry->src_data[2] & 0x3ff; - ret = amdgpu_vm_handle_fault(adev, entry->pasid, addr, write_fault); + ret = amdgpu_vm_handle_fault(adev, entry->pasid, entry->vmid, node_id, + addr, write_fault); WDOORBELL32(adev->irq.retry_cam_doorbell_index, cam_index); if (ret) return 1; @@ -595,7 +617,8 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, /* Try to handle the recoverable page faults by filling page * tables */ - if (amdgpu_vm_handle_fault(adev, entry->pasid, addr, write_fault)) + if (amdgpu_vm_handle_fault(adev, entry->pasid, entry->vmid, node_id, + addr, write_fault)) return 1; } } @@ -603,16 +626,6 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, if (!printk_ratelimit()) return 0; - if (entry->client_id == SOC15_IH_CLIENTID_VMC) { - hub_name = "mmhub0"; - hub = &adev->vmhub[AMDGPU_MMHUB_0]; - } else if (entry->client_id == SOC15_IH_CLIENTID_VMC1) { - hub_name = "mmhub1"; - hub = &adev->vmhub[AMDGPU_MMHUB_1]; - } else { - hub_name = "gfxhub0"; - hub = &adev->vmhub[AMDGPU_GFXHUB_0]; - } memset(&task_info, 0, sizeof(struct amdgpu_task_info)); amdgpu_vm_get_task_info(adev, entry->pasid, &task_info); @@ -628,6 +641,11 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, addr, entry->client_id, soc15_ih_clientid_name[entry->client_id]); + if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) + dev_err(adev->dev, " cookie node_id %d fault from die %s%d%s\n", + node_id, node_id % 4 == 3 ? "RSV" : "AID", node_id / 4, + node_id % 4 == 1 ? ".XCD0" : node_id % 4 == 2 ? ".XCD1" : ""); + if (amdgpu_sriov_vf(adev)) return 0; @@ -636,7 +654,7 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, * be updated to avoid reading an incorrect value due to * the new fast GRBM interface. */ - if ((entry->vmid_src == AMDGPU_GFXHUB_0) && + if ((entry->vmid_src == AMDGPU_GFXHUB(0)) && (adev->ip_versions[GC_HWIP][0] < IP_VERSION(9, 4, 2))) RREG32(hub->vm_l2_pro_fault_status); @@ -645,11 +663,10 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, rw = REG_GET_FIELD(status, VM_L2_PROTECTION_FAULT_STATUS, RW); WREG32_P(hub->vm_l2_pro_fault_cntl, 1, ~1); - dev_err(adev->dev, "VM_L2_PROTECTION_FAULT_STATUS:0x%08X\n", status); - if (hub == &adev->vmhub[AMDGPU_GFXHUB_0]) { + if (entry->vmid_src == AMDGPU_GFXHUB(0)) { dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", cid >= ARRAY_SIZE(gfxhub_client_ids) ? "unknown" : gfxhub_client_ids[cid], @@ -759,8 +776,8 @@ static bool gmc_v9_0_use_invalidate_semaphore(struct amdgpu_device *adev, adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) return false; - return ((vmhub == AMDGPU_MMHUB_0 || - vmhub == AMDGPU_MMHUB_1) && + return ((vmhub == AMDGPU_MMHUB0(0) || + vmhub == AMDGPU_MMHUB1(0)) && (!amdgpu_sriov_vf(adev)) && (!(!(adev->apu_flags & AMD_APU_IS_RAVEN2) && (adev->apu_flags & AMD_APU_IS_PICASSO)))); @@ -803,7 +820,7 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, u32 j, inv_req, inv_req2, tmp; struct amdgpu_vmhub *hub; - BUG_ON(vmhub >= adev->num_vmhubs); + BUG_ON(vmhub >= AMDGPU_MAX_VMHUBS); hub = &adev->vmhub[vmhub]; if (adev->gmc.xgmi.num_physical_nodes && @@ -816,6 +833,11 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, */ inv_req = gmc_v9_0_get_invalidate_req(vmid, 2); inv_req2 = gmc_v9_0_get_invalidate_req(vmid, flush_type); + } else if (flush_type == 2 && + adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3) && + adev->rev_id == 0) { + inv_req = gmc_v9_0_get_invalidate_req(vmid, 0); + inv_req2 = gmc_v9_0_get_invalidate_req(vmid, flush_type); } else { inv_req = gmc_v9_0_get_invalidate_req(vmid, flush_type); inv_req2 = 0; @@ -824,7 +846,7 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, /* This is necessary for a HW workaround under SRIOV as well * as GFXOFF under bare metal */ - if (adev->gfx.kiq.ring.sched.ready && + if (adev->gfx.kiq[0].ring.sched.ready && (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev)) && down_read_trylock(&adev->reset_domain->sem)) { uint32_t req = hub->vm_inv_eng0_req + hub->eng_distance * eng; @@ -849,11 +871,10 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, if (use_semaphore) { for (j = 0; j < adev->usec_timeout; j++) { /* a read return value of 1 means semaphore acquire */ - if (vmhub == AMDGPU_GFXHUB_0) - tmp = RREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_sem + hub->eng_distance * eng); - else + if (vmhub >= AMDGPU_MMHUB0(0)) tmp = RREG32_SOC15_IP_NO_KIQ(MMHUB, hub->vm_inv_eng0_sem + hub->eng_distance * eng); - + else + tmp = RREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_sem + hub->eng_distance * eng); if (tmp & 0x1) break; udelay(1); @@ -864,27 +885,26 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, } do { - if (vmhub == AMDGPU_GFXHUB_0) - WREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_req + hub->eng_distance * eng, inv_req); - else + if (vmhub >= AMDGPU_MMHUB0(0)) WREG32_SOC15_IP_NO_KIQ(MMHUB, hub->vm_inv_eng0_req + hub->eng_distance * eng, inv_req); + else + WREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_req + hub->eng_distance * eng, inv_req); /* * Issue a dummy read to wait for the ACK register to * be cleared to avoid a false ACK due to the new fast * GRBM interface. */ - if ((vmhub == AMDGPU_GFXHUB_0) && + if ((vmhub == AMDGPU_GFXHUB(0)) && (adev->ip_versions[GC_HWIP][0] < IP_VERSION(9, 4, 2))) RREG32_NO_KIQ(hub->vm_inv_eng0_req + hub->eng_distance * eng); for (j = 0; j < adev->usec_timeout; j++) { - if (vmhub == AMDGPU_GFXHUB_0) - tmp = RREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_ack + hub->eng_distance * eng); - else + if (vmhub >= AMDGPU_MMHUB0(0)) tmp = RREG32_SOC15_IP_NO_KIQ(MMHUB, hub->vm_inv_eng0_ack + hub->eng_distance * eng); - + else + tmp = RREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_ack + hub->eng_distance * eng); if (tmp & (1 << vmid)) break; udelay(1); @@ -900,10 +920,10 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, * add semaphore release after invalidation, * write with 0 means semaphore release */ - if (vmhub == AMDGPU_GFXHUB_0) - WREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_sem + hub->eng_distance * eng, 0); - else + if (vmhub >= AMDGPU_MMHUB0(0)) WREG32_SOC15_IP_NO_KIQ(MMHUB, hub->vm_inv_eng0_sem + hub->eng_distance * eng, 0); + else + WREG32_SOC15_IP_NO_KIQ(GC, hub->vm_inv_eng0_sem + hub->eng_distance * eng, 0); } spin_unlock(&adev->gmc.invalidate_lock); @@ -921,12 +941,13 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, * @pasid: pasid to be flush * @flush_type: the flush type * @all_hub: flush all hubs + * @inst: is used to select which instance of KIQ to use for the invalidation * * Flush the TLB for the requested pasid. */ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t pasid, uint32_t flush_type, - bool all_hub) + bool all_hub, uint32_t inst) { int vmid, i; signed long r; @@ -934,8 +955,8 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t queried_pasid; bool ret; u32 usec_timeout = amdgpu_sriov_vf(adev) ? SRIOV_USEC_TIMEOUT : adev->usec_timeout; - struct amdgpu_ring *ring = &adev->gfx.kiq.ring; - struct amdgpu_kiq *kiq = &adev->gfx.kiq; + struct amdgpu_ring *ring = &adev->gfx.kiq[inst].ring; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[inst]; if (amdgpu_in_reset(adev)) return -EIO; @@ -955,24 +976,31 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, if (vega20_xgmi_wa) ndw += kiq->pmf->invalidate_tlbs_size; - spin_lock(&adev->gfx.kiq.ring_lock); + spin_lock(&adev->gfx.kiq[inst].ring_lock); /* 2 dwords flush + 8 dwords fence */ amdgpu_ring_alloc(ring, ndw); if (vega20_xgmi_wa) kiq->pmf->kiq_invalidate_tlbs(ring, pasid, 2, all_hub); + + if (flush_type == 2 && + adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3) && + adev->rev_id == 0) + kiq->pmf->kiq_invalidate_tlbs(ring, + pasid, 0, all_hub); + kiq->pmf->kiq_invalidate_tlbs(ring, pasid, flush_type, all_hub); r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT); if (r) { amdgpu_ring_undo(ring); - spin_unlock(&adev->gfx.kiq.ring_lock); + spin_unlock(&adev->gfx.kiq[inst].ring_lock); up_read(&adev->reset_domain->sem); return -ETIME; } amdgpu_ring_commit(ring); - spin_unlock(&adev->gfx.kiq.ring_lock); + spin_unlock(&adev->gfx.kiq[inst].ring_lock); r = amdgpu_fence_wait_polling(ring, seq, usec_timeout); if (r < 1) { dev_err(adev->dev, "wait for kiq fence error: %ld.\n", r); @@ -989,12 +1017,12 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, &queried_pasid); if (ret && queried_pasid == pasid) { if (all_hub) { - for (i = 0; i < adev->num_vmhubs; i++) + for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) gmc_v9_0_flush_gpu_tlb(adev, vmid, i, flush_type); } else { gmc_v9_0_flush_gpu_tlb(adev, vmid, - AMDGPU_GFXHUB_0, flush_type); + AMDGPU_GFXHUB(0), flush_type); } break; } @@ -1060,10 +1088,10 @@ static void gmc_v9_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid, uint32_t reg; /* Do nothing because there's no lut register for mmhub1. */ - if (ring->vm_hub == AMDGPU_MMHUB_1) + if (ring->vm_hub == AMDGPU_MMHUB1(0)) return; - if (ring->vm_hub == AMDGPU_GFXHUB_0) + if (ring->vm_hub == AMDGPU_GFXHUB(0)) reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_VMID_0_LUT) + vmid; else reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_VMID_0_LUT_MM) + vmid; @@ -1159,13 +1187,14 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev, bool is_vram = bo->tbo.resource->mem_type == TTM_PL_VRAM; bool coherent = bo->flags & AMDGPU_GEM_CREATE_COHERENT; bool uncached = bo->flags & AMDGPU_GEM_CREATE_UNCACHED; - unsigned int mtype; + struct amdgpu_vm *vm = mapping->bo_va->base.vm; + unsigned int mtype_local, mtype; bool snoop = false; + bool is_local; switch (adev->ip_versions[GC_HWIP][0]) { case IP_VERSION(9, 4, 1): case IP_VERSION(9, 4, 2): - case IP_VERSION(9, 4, 3): if (is_vram) { if (bo_adev == adev) { if (uncached) @@ -1199,6 +1228,43 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev, */ snoop = true; } + break; + case IP_VERSION(9, 4, 3): + /* Only local VRAM BOs or system memory on non-NUMA APUs + * can be assumed to be local in their entirety. Choose + * MTYPE_NC as safe fallback for all system memory BOs on + * NUMA systems. Their MTYPE can be overridden per-page in + * gmc_v9_0_override_vm_pte_flags. + */ + mtype_local = MTYPE_RW; + if (amdgpu_mtype_local == 1) { + DRM_INFO_ONCE("Using MTYPE_NC for local memory\n"); + mtype_local = MTYPE_NC; + } else if (amdgpu_mtype_local == 2) { + DRM_INFO_ONCE("Using MTYPE_CC for local memory\n"); + mtype_local = MTYPE_CC; + } else { + DRM_INFO_ONCE("Using MTYPE_RW for local memory\n"); + } + is_local = (!is_vram && (adev->flags & AMD_IS_APU) && + num_possible_nodes() <= 1) || + (is_vram && adev == bo_adev && + KFD_XCP_MEM_ID(adev, bo->xcp_id) == vm->mem_id); + snoop = true; + if (uncached) { + mtype = MTYPE_UC; + } else if (adev->flags & AMD_IS_APU) { + mtype = is_local ? mtype_local : MTYPE_NC; + } else { + /* dGPU */ + if (is_local) + mtype = mtype_local; + else if (is_vram) + mtype = MTYPE_NC; + else + mtype = MTYPE_UC; + } + break; default: if (uncached || coherent) @@ -1241,6 +1307,72 @@ static void gmc_v9_0_get_vm_pte(struct amdgpu_device *adev, mapping, flags); } +static void gmc_v9_0_override_vm_pte_flags(struct amdgpu_device *adev, + struct amdgpu_vm *vm, + uint64_t addr, uint64_t *flags) +{ + int local_node, nid; + + /* Only GFX 9.4.3 APUs associate GPUs with NUMA nodes. Local system + * memory can use more efficient MTYPEs. + */ + if (adev->ip_versions[GC_HWIP][0] != IP_VERSION(9, 4, 3)) + return; + + /* Only direct-mapped memory allows us to determine the NUMA node from + * the DMA address. + */ + if (!adev->ram_is_direct_mapped) { + dev_dbg(adev->dev, "RAM is not direct mapped\n"); + return; + } + + /* Only override mappings with MTYPE_NC, which is the safe default for + * cacheable memory. + */ + if ((*flags & AMDGPU_PTE_MTYPE_VG10_MASK) != + AMDGPU_PTE_MTYPE_VG10(MTYPE_NC)) { + dev_dbg(adev->dev, "MTYPE is not NC\n"); + return; + } + + /* FIXME: Only supported on native mode for now. For carve-out, the + * NUMA affinity of the GPU/VM needs to come from the PCI info because + * memory partitions are not associated with different NUMA nodes. + */ + if (adev->gmc.is_app_apu && vm->mem_id >= 0) { + local_node = adev->gmc.mem_partitions[vm->mem_id].numa.node; + } else { + dev_dbg(adev->dev, "Only native mode APU is supported.\n"); + return; + } + + /* Only handle real RAM. Mappings of PCIe resources don't have struct + * page or NUMA nodes. + */ + if (!page_is_ram(addr >> PAGE_SHIFT)) { + dev_dbg(adev->dev, "Page is not RAM.\n"); + return; + } + nid = pfn_to_nid(addr >> PAGE_SHIFT); + dev_dbg(adev->dev, "vm->mem_id=%d, local_node=%d, nid=%d\n", + vm->mem_id, local_node, nid); + if (nid == local_node) { + uint64_t old_flags = *flags; + unsigned int mtype_local = MTYPE_RW; + + if (amdgpu_mtype_local == 1) + mtype_local = MTYPE_NC; + else if (amdgpu_mtype_local == 2) + mtype_local = MTYPE_CC; + + *flags = (*flags & ~AMDGPU_PTE_MTYPE_VG10_MASK) | + AMDGPU_PTE_MTYPE_VG10(mtype_local); + dev_dbg(adev->dev, "flags updated from %llx to %llx\n", + old_flags, *flags); + } +} + static unsigned gmc_v9_0_get_vbios_fb_size(struct amdgpu_device *adev) { u32 d1vga_control = RREG32_SOC15(DCE, 0, mmD1VGA_CONTROL); @@ -1283,6 +1415,27 @@ static unsigned gmc_v9_0_get_vbios_fb_size(struct amdgpu_device *adev) return size; } +static enum amdgpu_memory_partition +gmc_v9_0_get_memory_partition(struct amdgpu_device *adev, u32 *supp_modes) +{ + enum amdgpu_memory_partition mode = UNKNOWN_MEMORY_PARTITION_MODE; + + if (adev->nbio.funcs->get_memory_partition_mode) + mode = adev->nbio.funcs->get_memory_partition_mode(adev, + supp_modes); + + return mode; +} + +static enum amdgpu_memory_partition +gmc_v9_0_query_memory_partition(struct amdgpu_device *adev) +{ + if (amdgpu_sriov_vf(adev)) + return AMDGPU_NPS1_PARTITION_MODE; + + return gmc_v9_0_get_memory_partition(adev, NULL); +} + static const struct amdgpu_gmc_funcs gmc_v9_0_gmc_funcs = { .flush_gpu_tlb = gmc_v9_0_flush_gpu_tlb, .flush_gpu_tlb_pasid = gmc_v9_0_flush_gpu_tlb_pasid, @@ -1291,7 +1444,9 @@ static const struct amdgpu_gmc_funcs gmc_v9_0_gmc_funcs = { .map_mtype = gmc_v9_0_map_mtype, .get_vm_pde = gmc_v9_0_get_vm_pde, .get_vm_pte = gmc_v9_0_get_vm_pte, + .override_vm_pte_flags = gmc_v9_0_override_vm_pte_flags, .get_vbios_fb_size = gmc_v9_0_get_vbios_fb_size, + .query_mem_partition_mode = &gmc_v9_0_query_memory_partition, }; static void gmc_v9_0_set_gmc_funcs(struct amdgpu_device *adev) @@ -1372,6 +1527,9 @@ static void gmc_v9_0_set_mmhub_ras_funcs(struct amdgpu_device *adev) case IP_VERSION(9, 4, 2): adev->mmhub.ras = &mmhub_v1_7_ras; break; + case IP_VERSION(1, 8, 0): + adev->mmhub.ras = &mmhub_v1_8_ras; + break; default: /* mmhub ras is not available */ break; @@ -1419,9 +1577,13 @@ static int gmc_v9_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - /* ARCT and VEGA20 don't have XGMI defined in their IP discovery tables */ - if (adev->asic_type == CHIP_VEGA20 || - adev->asic_type == CHIP_ARCTURUS) + /* + * 9.4.0, 9.4.1 and 9.4.3 don't have XGMI defined + * in their IP discovery tables + */ + if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 0) || + adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 1) || + adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) adev->gmc.xgmi.supported = true; if (adev->ip_versions[XGMI_HWIP][0] == IP_VERSION(6, 1, 0)) { @@ -1430,6 +1592,20 @@ static int gmc_v9_0_early_init(void *handle) adev->smuio.funcs->is_host_gpu_xgmi_supported(adev); } + if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) { + enum amdgpu_pkg_type pkg_type = + adev->smuio.funcs->get_pkg_type(adev); + /* On GFXIP 9.4.3. APU, there is no physical VRAM domain present + * and the APU, can be in used two possible modes: + * - carveout mode + * - native APU mode + * "is_app_apu" can be used to identify the APU in the native + * mode. + */ + adev->gmc.is_app_apu = (pkg_type == AMDGPU_PKG_TYPE_APU && + !pci_resource_len(adev->pdev, 0)); + } + gmc_v9_0_set_gmc_funcs(adev); gmc_v9_0_set_irq_funcs(adev); gmc_v9_0_set_umc_funcs(adev); @@ -1525,8 +1701,13 @@ static int gmc_v9_0_mc_init(struct amdgpu_device *adev) int r; /* size in MB on si */ - adev->gmc.mc_vram_size = - adev->nbio.funcs->get_memsize(adev) * 1024ULL * 1024ULL; + if (!adev->gmc.is_app_apu) { + adev->gmc.mc_vram_size = + adev->nbio.funcs->get_memsize(adev) * 1024ULL * 1024ULL; + } else { + DRM_DEBUG("Set mc_vram_size = 0 for APP APU\n"); + adev->gmc.mc_vram_size = 0; + } adev->gmc.real_vram_size = adev->gmc.mc_vram_size; if (!(adev->flags & AMD_IS_APU) && @@ -1551,7 +1732,8 @@ static int gmc_v9_0_mc_init(struct amdgpu_device *adev) */ /* check whether both host-gpu and gpu-gpu xgmi links exist */ - if (((adev->flags & AMD_IS_APU) && !amdgpu_passthrough(adev)) || + if ((!amdgpu_sriov_vf(adev) && + (adev->flags & AMD_IS_APU) && !amdgpu_passthrough(adev)) || (adev->gmc.xgmi.supported && adev->gmc.xgmi.connected_to_cpu)) { adev->gmc.aper_base = @@ -1618,12 +1800,18 @@ static int gmc_v9_0_gart_init(struct amdgpu_device *adev) adev->gart.gart_pte_flags = AMDGPU_PTE_MTYPE_VG10(MTYPE_UC) | AMDGPU_PTE_EXECUTABLE; - r = amdgpu_gart_table_vram_alloc(adev); - if (r) - return r; + if (!adev->gmc.real_vram_size) { + dev_info(adev->dev, "Put GART in system memory for APU\n"); + r = amdgpu_gart_table_ram_alloc(adev); + if (r) + dev_err(adev->dev, "Failed to allocate GART in system memory\n"); + } else { + r = amdgpu_gart_table_vram_alloc(adev); + if (r) + return r; - if (adev->gmc.xgmi.connected_to_cpu) { - r = amdgpu_gmc_pdb0_alloc(adev); + if (adev->gmc.xgmi.connected_to_cpu) + r = amdgpu_gmc_pdb0_alloc(adev); } return r; @@ -1644,10 +1832,178 @@ static void gmc_v9_0_save_registers(struct amdgpu_device *adev) adev->gmc.sdpif_register = RREG32_SOC15(DCE, 0, mmDCHUBBUB_SDPIF_MMIO_CNTRL_0); } +static bool gmc_v9_0_validate_partition_info(struct amdgpu_device *adev) +{ + enum amdgpu_memory_partition mode; + u32 supp_modes; + bool valid; + + mode = gmc_v9_0_get_memory_partition(adev, &supp_modes); + + /* Mode detected by hardware not present in supported modes */ + if ((mode != UNKNOWN_MEMORY_PARTITION_MODE) && + !(BIT(mode - 1) & supp_modes)) + return false; + + switch (mode) { + case UNKNOWN_MEMORY_PARTITION_MODE: + case AMDGPU_NPS1_PARTITION_MODE: + valid = (adev->gmc.num_mem_partitions == 1); + break; + case AMDGPU_NPS2_PARTITION_MODE: + valid = (adev->gmc.num_mem_partitions == 2); + break; + case AMDGPU_NPS4_PARTITION_MODE: + valid = (adev->gmc.num_mem_partitions == 3 || + adev->gmc.num_mem_partitions == 4); + break; + default: + valid = false; + } + + return valid; +} + +static bool gmc_v9_0_is_node_present(int *node_ids, int num_ids, int nid) +{ + int i; + + /* Check if node with id 'nid' is present in 'node_ids' array */ + for (i = 0; i < num_ids; ++i) + if (node_ids[i] == nid) + return true; + + return false; +} + +static void +gmc_v9_0_init_acpi_mem_ranges(struct amdgpu_device *adev, + struct amdgpu_mem_partition_info *mem_ranges) +{ + int num_ranges = 0, ret, mem_groups; + struct amdgpu_numa_info numa_info; + int node_ids[MAX_MEM_RANGES]; + int num_xcc, xcc_id; + uint32_t xcc_mask; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + xcc_mask = (1U << num_xcc) - 1; + mem_groups = hweight32(adev->aid_mask); + + for_each_inst(xcc_id, xcc_mask) { + ret = amdgpu_acpi_get_mem_info(adev, xcc_id, &numa_info); + if (ret) + continue; + + if (numa_info.nid == NUMA_NO_NODE) { + mem_ranges[0].size = numa_info.size; + mem_ranges[0].numa.node = numa_info.nid; + num_ranges = 1; + break; + } + + if (gmc_v9_0_is_node_present(node_ids, num_ranges, + numa_info.nid)) + continue; + + node_ids[num_ranges] = numa_info.nid; + mem_ranges[num_ranges].numa.node = numa_info.nid; + mem_ranges[num_ranges].size = numa_info.size; + ++num_ranges; + } + + adev->gmc.num_mem_partitions = num_ranges; + + /* If there is only partition, don't use entire size */ + if (adev->gmc.num_mem_partitions == 1) { + mem_ranges[0].size = mem_ranges[0].size * (mem_groups - 1); + do_div(mem_ranges[0].size, mem_groups); + } +} + +static void +gmc_v9_0_init_sw_mem_ranges(struct amdgpu_device *adev, + struct amdgpu_mem_partition_info *mem_ranges) +{ + enum amdgpu_memory_partition mode; + u32 start_addr = 0, size; + int i; + + mode = gmc_v9_0_query_memory_partition(adev); + + switch (mode) { + case UNKNOWN_MEMORY_PARTITION_MODE: + case AMDGPU_NPS1_PARTITION_MODE: + adev->gmc.num_mem_partitions = 1; + break; + case AMDGPU_NPS2_PARTITION_MODE: + adev->gmc.num_mem_partitions = 2; + break; + case AMDGPU_NPS4_PARTITION_MODE: + if (adev->flags & AMD_IS_APU) + adev->gmc.num_mem_partitions = 3; + else + adev->gmc.num_mem_partitions = 4; + break; + default: + adev->gmc.num_mem_partitions = 1; + break; + } + + size = adev->gmc.real_vram_size >> AMDGPU_GPU_PAGE_SHIFT; + size /= adev->gmc.num_mem_partitions; + + for (i = 0; i < adev->gmc.num_mem_partitions; ++i) { + mem_ranges[i].range.fpfn = start_addr; + mem_ranges[i].size = ((u64)size << AMDGPU_GPU_PAGE_SHIFT); + mem_ranges[i].range.lpfn = start_addr + size - 1; + start_addr += size; + } + + /* Adjust the last one */ + mem_ranges[adev->gmc.num_mem_partitions - 1].range.lpfn = + (adev->gmc.real_vram_size >> AMDGPU_GPU_PAGE_SHIFT) - 1; + mem_ranges[adev->gmc.num_mem_partitions - 1].size = + adev->gmc.real_vram_size - + ((u64)mem_ranges[adev->gmc.num_mem_partitions - 1].range.fpfn + << AMDGPU_GPU_PAGE_SHIFT); +} + +static int gmc_v9_0_init_mem_ranges(struct amdgpu_device *adev) +{ + bool valid; + + adev->gmc.mem_partitions = kzalloc( + MAX_MEM_RANGES * sizeof(struct amdgpu_mem_partition_info), + GFP_KERNEL); + + if (!adev->gmc.mem_partitions) + return -ENOMEM; + + /* TODO : Get the range from PSP/Discovery for dGPU */ + if (adev->gmc.is_app_apu) + gmc_v9_0_init_acpi_mem_ranges(adev, adev->gmc.mem_partitions); + else + gmc_v9_0_init_sw_mem_ranges(adev, adev->gmc.mem_partitions); + + if (amdgpu_sriov_vf(adev)) + valid = true; + else + valid = gmc_v9_0_validate_partition_info(adev); + if (!valid) { + /* TODO: handle invalid case */ + dev_WARN(adev->dev, + "Mem ranges not matching with hardware config"); + } + + return 0; +} + static int gmc_v9_0_sw_init(void *handle) { int r, vram_width = 0, vram_type = 0, vram_vendor = 0, dma_addr_bits; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + unsigned long inst_mask = adev->aid_mask; adev->gfxhub.funcs->init(adev); @@ -1655,38 +2011,54 @@ static int gmc_v9_0_sw_init(void *handle) spin_lock_init(&adev->gmc.invalidate_lock); - r = amdgpu_atomfirmware_get_vram_info(adev, - &vram_width, &vram_type, &vram_vendor); - if (amdgpu_sriov_vf(adev)) - /* For Vega10 SR-IOV, vram_width can't be read from ATOM as RAVEN, - * and DF related registers is not readable, seems hardcord is the - * only way to set the correct vram_width - */ - adev->gmc.vram_width = 2048; - else if (amdgpu_emu_mode != 1) - adev->gmc.vram_width = vram_width; + if (!(adev->bios) || adev->gmc.is_app_apu) { + if (adev->flags & AMD_IS_APU) { + if (adev->gmc.is_app_apu) { + adev->gmc.vram_type = AMDGPU_VRAM_TYPE_HBM; + adev->gmc.vram_width = 128 * 64; + } else { + adev->gmc.vram_type = AMDGPU_VRAM_TYPE_DDR4; + adev->gmc.vram_width = 64 * 64; + } + } else { + adev->gmc.vram_type = AMDGPU_VRAM_TYPE_HBM; + adev->gmc.vram_width = 128 * 64; + } + } else { + r = amdgpu_atomfirmware_get_vram_info(adev, + &vram_width, &vram_type, &vram_vendor); + if (amdgpu_sriov_vf(adev)) + /* For Vega10 SR-IOV, vram_width can't be read from ATOM as RAVEN, + * and DF related registers is not readable, seems hardcord is the + * only way to set the correct vram_width + */ + adev->gmc.vram_width = 2048; + else if (amdgpu_emu_mode != 1) + adev->gmc.vram_width = vram_width; - if (!adev->gmc.vram_width) { - int chansize, numchan; + if (!adev->gmc.vram_width) { + int chansize, numchan; - /* hbm memory channel size */ - if (adev->flags & AMD_IS_APU) - chansize = 64; - else - chansize = 128; - if (adev->df.funcs && - adev->df.funcs->get_hbm_channel_number) { - numchan = adev->df.funcs->get_hbm_channel_number(adev); - adev->gmc.vram_width = numchan * chansize; + /* hbm memory channel size */ + if (adev->flags & AMD_IS_APU) + chansize = 64; + else + chansize = 128; + if (adev->df.funcs && + adev->df.funcs->get_hbm_channel_number) { + numchan = adev->df.funcs->get_hbm_channel_number(adev); + adev->gmc.vram_width = numchan * chansize; + } } - } - adev->gmc.vram_type = vram_type; - adev->gmc.vram_vendor = vram_vendor; + adev->gmc.vram_type = vram_type; + adev->gmc.vram_vendor = vram_vendor; + } switch (adev->ip_versions[GC_HWIP][0]) { case IP_VERSION(9, 1, 0): case IP_VERSION(9, 2, 2): - adev->num_vmhubs = 2; + set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask); + set_bit(AMDGPU_MMHUB0(0), adev->vmhubs_mask); if (adev->rev_id == 0x0 || adev->rev_id == 0x1) { amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48); @@ -1702,9 +2074,8 @@ static int gmc_v9_0_sw_init(void *handle) case IP_VERSION(9, 4, 0): case IP_VERSION(9, 3, 0): case IP_VERSION(9, 4, 2): - case IP_VERSION(9, 4, 3): - adev->num_vmhubs = 2; - + set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask); + set_bit(AMDGPU_MMHUB0(0), adev->vmhubs_mask); /* * To fulfill 4-level page support, @@ -1720,9 +2091,21 @@ static int gmc_v9_0_sw_init(void *handle) adev->gmc.translate_further = adev->vm_manager.num_level > 1; break; case IP_VERSION(9, 4, 1): - adev->num_vmhubs = 3; + set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask); + set_bit(AMDGPU_MMHUB0(0), adev->vmhubs_mask); + set_bit(AMDGPU_MMHUB1(0), adev->vmhubs_mask); /* Keep the vm size same with Vega20 */ + amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48); + adev->gmc.translate_further = adev->vm_manager.num_level > 1; + break; + case IP_VERSION(9, 4, 3): + bitmap_set(adev->vmhubs_mask, AMDGPU_GFXHUB(0), + NUM_XCC(adev->gfx.xcc_mask)); + + inst_mask <<= AMDGPU_MMHUB0(0); + bitmap_or(adev->vmhubs_mask, adev->vmhubs_mask, &inst_mask, 32); + amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48); adev->gmc.translate_further = adev->vm_manager.num_level > 1; break; @@ -1764,7 +2147,7 @@ static int gmc_v9_0_sw_init(void *handle) */ adev->gmc.mc_mask = 0xffffffffffffULL; /* 48 bit MC */ - dma_addr_bits = adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2) ? 48:44; + dma_addr_bits = adev->ip_versions[GC_HWIP][0] >= IP_VERSION(9, 4, 2) ? 48:44; r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(dma_addr_bits)); if (r) { printk(KERN_WARNING "amdgpu: No suitable DMA available.\n"); @@ -1778,6 +2161,12 @@ static int gmc_v9_0_sw_init(void *handle) amdgpu_gmc_get_vbios_allocations(adev); + if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) { + r = gmc_v9_0_init_mem_ranges(adev); + if (r) + return r; + } + /* Memory manager */ r = amdgpu_bo_init(adev); if (r) @@ -1810,6 +2199,9 @@ static int gmc_v9_0_sw_init(void *handle) if (r) return r; + if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) + amdgpu_gmc_sysfs_init(adev); + return 0; } @@ -1817,10 +2209,20 @@ static int gmc_v9_0_sw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) + amdgpu_gmc_sysfs_fini(adev); + adev->gmc.num_mem_partitions = 0; + kfree(adev->gmc.mem_partitions); + amdgpu_gmc_ras_fini(adev); amdgpu_gem_force_release(adev); amdgpu_vm_manager_fini(adev); - amdgpu_gart_table_vram_free(adev); + if (!adev->gmc.real_vram_size) { + dev_info(adev->dev, "Put GART in system memory for APU free\n"); + amdgpu_gart_table_ram_free(adev); + } else { + amdgpu_gart_table_vram_free(adev); + } amdgpu_bo_free_kernel(&adev->gmc.pdb0_bo, NULL, &adev->gmc.ptr_pdb0); amdgpu_bo_fini(adev); @@ -1946,8 +2348,8 @@ static int gmc_v9_0_hw_init(void *handle) adev->gfxhub.funcs->set_fault_enable_default(adev, value); adev->mmhub.funcs->set_fault_enable_default(adev, value); } - for (i = 0; i < adev->num_vmhubs; ++i) { - if (adev->in_s0ix && (i == AMDGPU_GFXHUB_0)) + for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) { + if (adev->in_s0ix && (i == AMDGPU_GFXHUB(0))) continue; gmc_v9_0_flush_gpu_tlb(adev, 0, i, 0); } diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c index a3076eb8af6acd14641e4a4588db13afe61ac41e..77595e9622da34204d3be939c2c1c74cfdf9bdf3 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c @@ -437,7 +437,7 @@ static int jpeg_v1_0_process_interrupt(struct amdgpu_device *adev, switch (entry->src_id) { case 126: - amdgpu_fence_process(&adev->jpeg.inst->ring_dec); + amdgpu_fence_process(adev->jpeg.inst->ring_dec); break; default: DRM_ERROR("Unhandled interrupt: %d %d\n", @@ -460,6 +460,7 @@ int jpeg_v1_0_early_init(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; adev->jpeg.num_jpeg_inst = 1; + adev->jpeg.num_jpeg_rings = 1; jpeg_v1_0_set_dec_ring_funcs(adev); jpeg_v1_0_set_irq_funcs(adev); @@ -484,15 +485,15 @@ int jpeg_v1_0_sw_init(void *handle) if (r) return r; - ring = &adev->jpeg.inst->ring_dec; - ring->vm_hub = AMDGPU_MMHUB_0; + ring = adev->jpeg.inst->ring_dec; + ring->vm_hub = AMDGPU_MMHUB0(0); sprintf(ring->name, "jpeg_dec"); r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0, AMDGPU_RING_PRIO_DEFAULT, NULL); if (r) return r; - adev->jpeg.internal.jpeg_pitch = adev->jpeg.inst->external.jpeg_pitch = + adev->jpeg.internal.jpeg_pitch[0] = adev->jpeg.inst->external.jpeg_pitch[0] = SOC15_REG_OFFSET(JPEG, 0, mmUVD_JPEG_PITCH); return 0; @@ -509,7 +510,7 @@ void jpeg_v1_0_sw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - amdgpu_ring_fini(&adev->jpeg.inst[0].ring_dec); + amdgpu_ring_fini(adev->jpeg.inst->ring_dec); } /** @@ -522,7 +523,7 @@ void jpeg_v1_0_sw_fini(void *handle) */ void jpeg_v1_0_start(struct amdgpu_device *adev, int mode) { - struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec; + struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec; if (mode == 0) { WREG32_SOC15(JPEG, 0, mmUVD_LMI_JRBC_RB_VMID, 0); @@ -579,7 +580,7 @@ static const struct amdgpu_ring_funcs jpeg_v1_0_decode_ring_vm_funcs = { static void jpeg_v1_0_set_dec_ring_funcs(struct amdgpu_device *adev) { - adev->jpeg.inst->ring_dec.funcs = &jpeg_v1_0_decode_ring_vm_funcs; + adev->jpeg.inst->ring_dec->funcs = &jpeg_v1_0_decode_ring_vm_funcs; DRM_INFO("JPEG decode is enabled in VM mode\n"); } diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c index 0eddf7c824a728e85e11ab9ad97ce5473245634a..c25d4a07350b76ad34d542177415503f4e7920a2 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c @@ -49,6 +49,7 @@ static int jpeg_v2_0_early_init(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; adev->jpeg.num_jpeg_inst = 1; + adev->jpeg.num_jpeg_rings = 1; jpeg_v2_0_set_dec_ring_funcs(adev); jpeg_v2_0_set_irq_funcs(adev); @@ -83,18 +84,18 @@ static int jpeg_v2_0_sw_init(void *handle) if (r) return r; - ring = &adev->jpeg.inst->ring_dec; + ring = adev->jpeg.inst->ring_dec; ring->use_doorbell = true; ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1; - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); sprintf(ring->name, "jpeg_dec"); r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0, AMDGPU_RING_PRIO_DEFAULT, NULL); if (r) return r; - adev->jpeg.internal.jpeg_pitch = mmUVD_JPEG_PITCH_INTERNAL_OFFSET; - adev->jpeg.inst->external.jpeg_pitch = SOC15_REG_OFFSET(JPEG, 0, mmUVD_JPEG_PITCH); + adev->jpeg.internal.jpeg_pitch[0] = mmUVD_JPEG_PITCH_INTERNAL_OFFSET; + adev->jpeg.inst->external.jpeg_pitch[0] = SOC15_REG_OFFSET(JPEG, 0, mmUVD_JPEG_PITCH); return 0; } @@ -129,7 +130,7 @@ static int jpeg_v2_0_sw_fini(void *handle) static int jpeg_v2_0_hw_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec; + struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec; int r; adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell, @@ -312,7 +313,7 @@ static void jpeg_v2_0_enable_clock_gating(struct amdgpu_device *adev) */ static int jpeg_v2_0_start(struct amdgpu_device *adev) { - struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec; + struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec; int r; if (adev->pm.dpm_enabled) @@ -729,7 +730,7 @@ static int jpeg_v2_0_process_interrupt(struct amdgpu_device *adev, switch (entry->src_id) { case VCN_2_0__SRCID__JPEG_DECODE: - amdgpu_fence_process(&adev->jpeg.inst->ring_dec); + amdgpu_fence_process(adev->jpeg.inst->ring_dec); break; default: DRM_ERROR("Unhandled interrupt: %d %d\n", @@ -791,7 +792,7 @@ static const struct amdgpu_ring_funcs jpeg_v2_0_dec_ring_vm_funcs = { static void jpeg_v2_0_set_dec_ring_funcs(struct amdgpu_device *adev) { - adev->jpeg.inst->ring_dec.funcs = &jpeg_v2_0_dec_ring_vm_funcs; + adev->jpeg.inst->ring_dec->funcs = &jpeg_v2_0_dec_ring_vm_funcs; DRM_INFO("JPEG decode is enabled in VM mode\n"); } diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c index b040f51d9aa9d2847ea7f0c1023a15de8e0ed381..aadb74de52bcf8e99d875727b9aa1dc4b6ecb511 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c @@ -60,6 +60,7 @@ static int jpeg_v2_5_early_init(void *handle) u32 harvest; int i; + adev->jpeg.num_jpeg_rings = 1; adev->jpeg.num_jpeg_inst = JPEG25_MAX_HW_INSTANCES_ARCTURUS; for (i = 0; i < adev->jpeg.num_jpeg_inst; i++) { harvest = RREG32_SOC15(JPEG, i, mmCC_UVD_HARVESTING); @@ -102,13 +103,13 @@ static int jpeg_v2_5_sw_init(void *handle) /* JPEG DJPEG POISON EVENT */ r = amdgpu_irq_add_id(adev, amdgpu_ih_clientid_jpeg[i], - VCN_2_6__SRCID_DJPEG0_POISON, &adev->jpeg.inst[i].irq); + VCN_2_6__SRCID_DJPEG0_POISON, &adev->jpeg.inst[i].ras_poison_irq); if (r) return r; /* JPEG EJPEG POISON EVENT */ r = amdgpu_irq_add_id(adev, amdgpu_ih_clientid_jpeg[i], - VCN_2_6__SRCID_EJPEG0_POISON, &adev->jpeg.inst[i].irq); + VCN_2_6__SRCID_EJPEG0_POISON, &adev->jpeg.inst[i].ras_poison_irq); if (r) return r; } @@ -125,12 +126,12 @@ static int jpeg_v2_5_sw_init(void *handle) if (adev->jpeg.harvest_config & (1 << i)) continue; - ring = &adev->jpeg.inst[i].ring_dec; + ring = adev->jpeg.inst[i].ring_dec; ring->use_doorbell = true; if (adev->ip_versions[UVD_HWIP][0] == IP_VERSION(2, 5, 0)) - ring->vm_hub = AMDGPU_MMHUB_1; + ring->vm_hub = AMDGPU_MMHUB1(0); else - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1 + 8 * i; sprintf(ring->name, "jpeg_dec_%d", i); r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst[i].irq, @@ -138,8 +139,8 @@ static int jpeg_v2_5_sw_init(void *handle) if (r) return r; - adev->jpeg.internal.jpeg_pitch = mmUVD_JPEG_PITCH_INTERNAL_OFFSET; - adev->jpeg.inst[i].external.jpeg_pitch = SOC15_REG_OFFSET(JPEG, i, mmUVD_JPEG_PITCH); + adev->jpeg.internal.jpeg_pitch[0] = mmUVD_JPEG_PITCH_INTERNAL_OFFSET; + adev->jpeg.inst[i].external.jpeg_pitch[0] = SOC15_REG_OFFSET(JPEG, i, mmUVD_JPEG_PITCH); } r = amdgpu_jpeg_ras_sw_init(adev); @@ -186,7 +187,7 @@ static int jpeg_v2_5_hw_init(void *handle) if (adev->jpeg.harvest_config & (1 << i)) continue; - ring = &adev->jpeg.inst[i].ring_dec; + ring = adev->jpeg.inst[i].ring_dec; adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell, (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 8 * i, i); @@ -221,6 +222,9 @@ static int jpeg_v2_5_hw_fini(void *handle) if (adev->jpeg.cur_state != AMD_PG_STATE_GATE && RREG32_SOC15(JPEG, i, mmUVD_JRBC_STATUS)) jpeg_v2_5_set_powergating_state(adev, AMD_PG_STATE_GATE); + + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG)) + amdgpu_irq_put(adev, &adev->jpeg.inst[i].ras_poison_irq, 0); } return 0; @@ -326,7 +330,7 @@ static int jpeg_v2_5_start(struct amdgpu_device *adev) if (adev->jpeg.harvest_config & (1 << i)) continue; - ring = &adev->jpeg.inst[i].ring_dec; + ring = adev->jpeg.inst[i].ring_dec; /* disable anti hang mechanism */ WREG32_P(SOC15_REG_OFFSET(JPEG, i, mmUVD_JPEG_POWER_STATUS), 0, ~UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK); @@ -569,6 +573,14 @@ static int jpeg_v2_5_set_interrupt_state(struct amdgpu_device *adev, return 0; } +static int jpeg_v2_6_set_ras_interrupt_state(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + unsigned int type, + enum amdgpu_interrupt_state state) +{ + return 0; +} + static int jpeg_v2_5_process_interrupt(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) @@ -591,11 +603,7 @@ static int jpeg_v2_5_process_interrupt(struct amdgpu_device *adev, switch (entry->src_id) { case VCN_2_0__SRCID__JPEG_DECODE: - amdgpu_fence_process(&adev->jpeg.inst[ip_instance].ring_dec); - break; - case VCN_2_6__SRCID_DJPEG0_POISON: - case VCN_2_6__SRCID_EJPEG0_POISON: - amdgpu_jpeg_process_poison_irq(adev, source, entry); + amdgpu_fence_process(adev->jpeg.inst[ip_instance].ring_dec); break; default: DRM_ERROR("Unhandled interrupt: %d %d\n", @@ -712,10 +720,10 @@ static void jpeg_v2_5_set_dec_ring_funcs(struct amdgpu_device *adev) if (adev->jpeg.harvest_config & (1 << i)) continue; if (adev->asic_type == CHIP_ARCTURUS) - adev->jpeg.inst[i].ring_dec.funcs = &jpeg_v2_5_dec_ring_vm_funcs; + adev->jpeg.inst[i].ring_dec->funcs = &jpeg_v2_5_dec_ring_vm_funcs; else /* CHIP_ALDEBARAN */ - adev->jpeg.inst[i].ring_dec.funcs = &jpeg_v2_6_dec_ring_vm_funcs; - adev->jpeg.inst[i].ring_dec.me = i; + adev->jpeg.inst[i].ring_dec->funcs = &jpeg_v2_6_dec_ring_vm_funcs; + adev->jpeg.inst[i].ring_dec->me = i; DRM_INFO("JPEG(%d) JPEG decode is enabled in VM mode\n", i); } } @@ -725,6 +733,11 @@ static const struct amdgpu_irq_src_funcs jpeg_v2_5_irq_funcs = { .process = jpeg_v2_5_process_interrupt, }; +static const struct amdgpu_irq_src_funcs jpeg_v2_6_ras_irq_funcs = { + .set = jpeg_v2_6_set_ras_interrupt_state, + .process = amdgpu_jpeg_process_poison_irq, +}; + static void jpeg_v2_5_set_irq_funcs(struct amdgpu_device *adev) { int i; @@ -735,6 +748,9 @@ static void jpeg_v2_5_set_irq_funcs(struct amdgpu_device *adev) adev->jpeg.inst[i].irq.num_types = 1; adev->jpeg.inst[i].irq.funcs = &jpeg_v2_5_irq_funcs; + + adev->jpeg.inst[i].ras_poison_irq.num_types = 1; + adev->jpeg.inst[i].ras_poison_irq.funcs = &jpeg_v2_6_ras_irq_funcs; } } @@ -800,6 +816,7 @@ const struct amdgpu_ras_block_hw_ops jpeg_v2_6_ras_hw_ops = { static struct amdgpu_jpeg_ras jpeg_v2_6_ras = { .ras_block = { .hw_ops = &jpeg_v2_6_ras_hw_ops, + .ras_late_init = amdgpu_jpeg_ras_late_init, }, }; diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c index 1c2292cc5f2ce2cf3836902dd5beb7db9997ef18..79791379fc2b09b9b77221f8710fa92484de8428 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c @@ -64,6 +64,7 @@ static int jpeg_v3_0_early_init(void *handle) } adev->jpeg.num_jpeg_inst = 1; + adev->jpeg.num_jpeg_rings = 1; jpeg_v3_0_set_dec_ring_funcs(adev); jpeg_v3_0_set_irq_funcs(adev); @@ -98,18 +99,18 @@ static int jpeg_v3_0_sw_init(void *handle) if (r) return r; - ring = &adev->jpeg.inst->ring_dec; + ring = adev->jpeg.inst->ring_dec; ring->use_doorbell = true; ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1; - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); sprintf(ring->name, "jpeg_dec"); r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0, AMDGPU_RING_PRIO_DEFAULT, NULL); if (r) return r; - adev->jpeg.internal.jpeg_pitch = mmUVD_JPEG_PITCH_INTERNAL_OFFSET; - adev->jpeg.inst->external.jpeg_pitch = SOC15_REG_OFFSET(JPEG, 0, mmUVD_JPEG_PITCH); + adev->jpeg.internal.jpeg_pitch[0] = mmUVD_JPEG_PITCH_INTERNAL_OFFSET; + adev->jpeg.inst->external.jpeg_pitch[0] = SOC15_REG_OFFSET(JPEG, 0, mmUVD_JPEG_PITCH); return 0; } @@ -144,7 +145,7 @@ static int jpeg_v3_0_sw_fini(void *handle) static int jpeg_v3_0_hw_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec; + struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec; int r; adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell, @@ -330,7 +331,7 @@ static int jpeg_v3_0_enable_static_power_gating(struct amdgpu_device *adev) */ static int jpeg_v3_0_start(struct amdgpu_device *adev) { - struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec; + struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec; int r; if (adev->pm.dpm_enabled) @@ -527,7 +528,7 @@ static int jpeg_v3_0_process_interrupt(struct amdgpu_device *adev, switch (entry->src_id) { case VCN_2_0__SRCID__JPEG_DECODE: - amdgpu_fence_process(&adev->jpeg.inst->ring_dec); + amdgpu_fence_process(adev->jpeg.inst->ring_dec); break; default: DRM_ERROR("Unhandled interrupt: %d %d\n", @@ -589,7 +590,7 @@ static const struct amdgpu_ring_funcs jpeg_v3_0_dec_ring_vm_funcs = { static void jpeg_v3_0_set_dec_ring_funcs(struct amdgpu_device *adev) { - adev->jpeg.inst->ring_dec.funcs = &jpeg_v3_0_dec_ring_vm_funcs; + adev->jpeg.inst->ring_dec->funcs = &jpeg_v3_0_dec_ring_vm_funcs; DRM_INFO("JPEG decode is enabled in VM mode\n"); } diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c index 77e1e64aa1d1c7b7987d15aa5a771b553754a230..a707d407fbd0d28edc26b2e5b2d8ba0cf332ee62 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c @@ -58,6 +58,7 @@ static int jpeg_v4_0_early_init(void *handle) adev->jpeg.num_jpeg_inst = 1; + adev->jpeg.num_jpeg_rings = 1; jpeg_v4_0_set_dec_ring_funcs(adev); jpeg_v4_0_set_irq_funcs(adev); @@ -87,13 +88,13 @@ static int jpeg_v4_0_sw_init(void *handle) /* JPEG DJPEG POISON EVENT */ r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN, - VCN_4_0__SRCID_DJPEG0_POISON, &adev->jpeg.inst->irq); + VCN_4_0__SRCID_DJPEG0_POISON, &adev->jpeg.inst->ras_poison_irq); if (r) return r; /* JPEG EJPEG POISON EVENT */ r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN, - VCN_4_0__SRCID_EJPEG0_POISON, &adev->jpeg.inst->irq); + VCN_4_0__SRCID_EJPEG0_POISON, &adev->jpeg.inst->ras_poison_irq); if (r) return r; @@ -105,10 +106,10 @@ static int jpeg_v4_0_sw_init(void *handle) if (r) return r; - ring = &adev->jpeg.inst->ring_dec; + ring = adev->jpeg.inst->ring_dec; ring->use_doorbell = true; ring->doorbell_index = amdgpu_sriov_vf(adev) ? (((adev->doorbell_index.vcn.vcn_ring0_1) << 1) + 4) : ((adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 1); - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); sprintf(ring->name, "jpeg_dec"); r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0, @@ -116,8 +117,8 @@ static int jpeg_v4_0_sw_init(void *handle) if (r) return r; - adev->jpeg.internal.jpeg_pitch = regUVD_JPEG_PITCH_INTERNAL_OFFSET; - adev->jpeg.inst->external.jpeg_pitch = SOC15_REG_OFFSET(JPEG, 0, regUVD_JPEG_PITCH); + adev->jpeg.internal.jpeg_pitch[0] = regUVD_JPEG_PITCH_INTERNAL_OFFSET; + adev->jpeg.inst->external.jpeg_pitch[0] = SOC15_REG_OFFSET(JPEG, 0, regUVD_JPEG_PITCH); r = amdgpu_jpeg_ras_sw_init(adev); if (r) @@ -156,7 +157,7 @@ static int jpeg_v4_0_sw_fini(void *handle) static int jpeg_v4_0_hw_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec; + struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec; int r; if (amdgpu_sriov_vf(adev)) { @@ -202,7 +203,8 @@ static int jpeg_v4_0_hw_fini(void *handle) RREG32_SOC15(JPEG, 0, regUVD_JRBC_STATUS)) jpeg_v4_0_set_powergating_state(adev, AMD_PG_STATE_GATE); } - amdgpu_irq_put(adev, &adev->jpeg.inst->irq, 0); + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG)) + amdgpu_irq_put(adev, &adev->jpeg.inst->ras_poison_irq, 0); return 0; } @@ -363,7 +365,7 @@ static int jpeg_v4_0_enable_static_power_gating(struct amdgpu_device *adev) */ static int jpeg_v4_0_start(struct amdgpu_device *adev) { - struct amdgpu_ring *ring = &adev->jpeg.inst->ring_dec; + struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec; int r; if (adev->pm.dpm_enabled) @@ -441,7 +443,7 @@ static int jpeg_v4_0_start_sriov(struct amdgpu_device *adev) table_size = 0; - ring = &adev->jpeg.inst->ring_dec; + ring = adev->jpeg.inst->ring_dec; MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(JPEG, 0, regUVD_LMI_JRBC_RB_64BIT_BAR_LOW), @@ -670,6 +672,14 @@ static int jpeg_v4_0_set_interrupt_state(struct amdgpu_device *adev, return 0; } +static int jpeg_v4_0_set_ras_interrupt_state(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + unsigned int type, + enum amdgpu_interrupt_state state) +{ + return 0; +} + static int jpeg_v4_0_process_interrupt(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) @@ -678,11 +688,7 @@ static int jpeg_v4_0_process_interrupt(struct amdgpu_device *adev, switch (entry->src_id) { case VCN_4_0__SRCID__JPEG_DECODE: - amdgpu_fence_process(&adev->jpeg.inst->ring_dec); - break; - case VCN_4_0__SRCID_DJPEG0_POISON: - case VCN_4_0__SRCID_EJPEG0_POISON: - amdgpu_jpeg_process_poison_irq(adev, source, entry); + amdgpu_fence_process(adev->jpeg.inst->ring_dec); break; default: DRM_DEV_ERROR(adev->dev, "Unhandled interrupt: %d %d\n", @@ -744,7 +750,7 @@ static const struct amdgpu_ring_funcs jpeg_v4_0_dec_ring_vm_funcs = { static void jpeg_v4_0_set_dec_ring_funcs(struct amdgpu_device *adev) { - adev->jpeg.inst->ring_dec.funcs = &jpeg_v4_0_dec_ring_vm_funcs; + adev->jpeg.inst->ring_dec->funcs = &jpeg_v4_0_dec_ring_vm_funcs; DRM_DEV_INFO(adev->dev, "JPEG decode is enabled in VM mode\n"); } @@ -753,10 +759,18 @@ static const struct amdgpu_irq_src_funcs jpeg_v4_0_irq_funcs = { .process = jpeg_v4_0_process_interrupt, }; +static const struct amdgpu_irq_src_funcs jpeg_v4_0_ras_irq_funcs = { + .set = jpeg_v4_0_set_ras_interrupt_state, + .process = amdgpu_jpeg_process_poison_irq, +}; + static void jpeg_v4_0_set_irq_funcs(struct amdgpu_device *adev) { adev->jpeg.inst->irq.num_types = 1; adev->jpeg.inst->irq.funcs = &jpeg_v4_0_irq_funcs; + + adev->jpeg.inst->ras_poison_irq.num_types = 1; + adev->jpeg.inst->ras_poison_irq.funcs = &jpeg_v4_0_ras_irq_funcs; } const struct amdgpu_ip_block_version jpeg_v4_0_ip_block = { @@ -811,6 +825,7 @@ const struct amdgpu_ras_block_hw_ops jpeg_v4_0_ras_hw_ops = { static struct amdgpu_jpeg_ras jpeg_v4_0_ras = { .ras_block = { .hw_ops = &jpeg_v4_0_ras_hw_ops, + .ras_late_init = amdgpu_jpeg_ras_late_init, }, }; diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c new file mode 100644 index 0000000000000000000000000000000000000000..ce2b22f7e4e4b8bf21860e5b4d724bcc16b2db7f --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c @@ -0,0 +1,1074 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#include "amdgpu.h" +#include "amdgpu_jpeg.h" +#include "soc15.h" +#include "soc15d.h" +#include "jpeg_v4_0_3.h" + +#include "vcn/vcn_4_0_3_offset.h" +#include "vcn/vcn_4_0_3_sh_mask.h" +#include "ivsrcid/vcn/irqsrcs_vcn_4_0.h" + +enum jpeg_engin_status { + UVD_PGFSM_STATUS__UVDJ_PWR_ON = 0, + UVD_PGFSM_STATUS__UVDJ_PWR_OFF = 2, +}; + +static void jpeg_v4_0_3_set_dec_ring_funcs(struct amdgpu_device *adev); +static void jpeg_v4_0_3_set_irq_funcs(struct amdgpu_device *adev); +static int jpeg_v4_0_3_set_powergating_state(void *handle, + enum amd_powergating_state state); +static void jpeg_v4_0_3_set_ras_funcs(struct amdgpu_device *adev); + +static int amdgpu_ih_srcid_jpeg[] = { + VCN_4_0__SRCID__JPEG_DECODE, + VCN_4_0__SRCID__JPEG1_DECODE, + VCN_4_0__SRCID__JPEG2_DECODE, + VCN_4_0__SRCID__JPEG3_DECODE, + VCN_4_0__SRCID__JPEG4_DECODE, + VCN_4_0__SRCID__JPEG5_DECODE, + VCN_4_0__SRCID__JPEG6_DECODE, + VCN_4_0__SRCID__JPEG7_DECODE +}; + +/** + * jpeg_v4_0_3_early_init - set function pointers + * + * @handle: amdgpu_device pointer + * + * Set ring and irq function pointers + */ +static int jpeg_v4_0_3_early_init(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + adev->jpeg.num_jpeg_rings = AMDGPU_MAX_JPEG_RINGS; + + jpeg_v4_0_3_set_dec_ring_funcs(adev); + jpeg_v4_0_3_set_irq_funcs(adev); + jpeg_v4_0_3_set_ras_funcs(adev); + + return 0; +} + +/** + * jpeg_v4_0_3_sw_init - sw init for JPEG block + * + * @handle: amdgpu_device pointer + * + * Load firmware and sw initialization + */ +static int jpeg_v4_0_3_sw_init(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_ring *ring; + int i, j, r, jpeg_inst; + + for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) { + /* JPEG TRAP */ + r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN, + amdgpu_ih_srcid_jpeg[j], &adev->jpeg.inst->irq); + if (r) + return r; + } + + r = amdgpu_jpeg_sw_init(adev); + if (r) + return r; + + r = amdgpu_jpeg_resume(adev); + if (r) + return r; + + for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { + jpeg_inst = GET_INST(JPEG, i); + + for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) { + ring = &adev->jpeg.inst[i].ring_dec[j]; + ring->use_doorbell = true; + ring->vm_hub = AMDGPU_MMHUB0(adev->jpeg.inst[i].aid_id); + ring->doorbell_index = + (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + + 1 + j + 9 * jpeg_inst; + sprintf(ring->name, "jpeg_dec_%d.%d", adev->jpeg.inst[i].aid_id, j); + r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0, + AMDGPU_RING_PRIO_DEFAULT, NULL); + if (r) + return r; + + adev->jpeg.internal.jpeg_pitch[j] = + regUVD_JRBC0_UVD_JRBC_SCRATCH0_INTERNAL_OFFSET; + adev->jpeg.inst[i].external.jpeg_pitch[j] = + SOC15_REG_OFFSET1( + JPEG, jpeg_inst, + regUVD_JRBC0_UVD_JRBC_SCRATCH0, + (j ? (0x40 * j - 0xc80) : 0)); + } + } + + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG)) { + r = amdgpu_jpeg_ras_sw_init(adev); + if (r) { + dev_err(adev->dev, "Failed to initialize jpeg ras block!\n"); + return r; + } + } + + return 0; +} + +/** + * jpeg_v4_0_3_sw_fini - sw fini for JPEG block + * + * @handle: amdgpu_device pointer + * + * JPEG suspend and free up sw allocation + */ +static int jpeg_v4_0_3_sw_fini(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int r; + + r = amdgpu_jpeg_suspend(adev); + if (r) + return r; + + r = amdgpu_jpeg_sw_fini(adev); + + return r; +} + +/** + * jpeg_v4_0_3_hw_init - start and test JPEG block + * + * @handle: amdgpu_device pointer + * + */ +static int jpeg_v4_0_3_hw_init(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_ring *ring; + int i, j, r, jpeg_inst; + + for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { + jpeg_inst = GET_INST(JPEG, i); + + ring = adev->jpeg.inst[i].ring_dec; + + if (ring->use_doorbell) + adev->nbio.funcs->vcn_doorbell_range( + adev, ring->use_doorbell, + (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + + 9 * jpeg_inst, + adev->jpeg.inst[i].aid_id); + + for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) { + ring = &adev->jpeg.inst[i].ring_dec[j]; + if (ring->use_doorbell) + WREG32_SOC15_OFFSET( + VCN, GET_INST(VCN, i), + regVCN_JPEG_DB_CTRL, + (ring->pipe ? (ring->pipe - 0x15) : 0), + ring->doorbell_index + << VCN_JPEG_DB_CTRL__OFFSET__SHIFT | + VCN_JPEG_DB_CTRL__EN_MASK); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + } + } + DRM_DEV_INFO(adev->dev, "JPEG decode initialized successfully.\n"); + + return 0; +} + +/** + * jpeg_v4_0_3_hw_fini - stop the hardware block + * + * @handle: amdgpu_device pointer + * + * Stop the JPEG block, mark ring as not ready any more + */ +static int jpeg_v4_0_3_hw_fini(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int ret = 0; + + cancel_delayed_work_sync(&adev->jpeg.idle_work); + + if (adev->jpeg.cur_state != AMD_PG_STATE_GATE) + ret = jpeg_v4_0_3_set_powergating_state(adev, AMD_PG_STATE_GATE); + + return ret; +} + +/** + * jpeg_v4_0_3_suspend - suspend JPEG block + * + * @handle: amdgpu_device pointer + * + * HW fini and suspend JPEG block + */ +static int jpeg_v4_0_3_suspend(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int r; + + r = jpeg_v4_0_3_hw_fini(adev); + if (r) + return r; + + r = amdgpu_jpeg_suspend(adev); + + return r; +} + +/** + * jpeg_v4_0_3_resume - resume JPEG block + * + * @handle: amdgpu_device pointer + * + * Resume firmware and hw init JPEG block + */ +static int jpeg_v4_0_3_resume(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int r; + + r = amdgpu_jpeg_resume(adev); + if (r) + return r; + + r = jpeg_v4_0_3_hw_init(adev); + + return r; +} + +static void jpeg_v4_0_3_disable_clock_gating(struct amdgpu_device *adev, int inst_idx) +{ + int i, jpeg_inst; + uint32_t data; + + jpeg_inst = GET_INST(JPEG, inst_idx); + data = RREG32_SOC15(JPEG, jpeg_inst, regJPEG_CGC_CTRL); + if (adev->cg_flags & AMD_CG_SUPPORT_JPEG_MGCG) { + data |= 1 << JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT; + data &= (~(JPEG_CGC_CTRL__JPEG0_DEC_MODE_MASK << 1)); + } else { + data &= ~JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT; + } + + data |= 1 << JPEG_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT; + data |= 4 << JPEG_CGC_CTRL__CLK_OFF_DELAY__SHIFT; + WREG32_SOC15(JPEG, jpeg_inst, regJPEG_CGC_CTRL, data); + + data = RREG32_SOC15(JPEG, jpeg_inst, regJPEG_CGC_GATE); + data &= ~(JPEG_CGC_GATE__JMCIF_MASK | JPEG_CGC_GATE__JRBBM_MASK); + for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) + data &= ~(JPEG_CGC_GATE__JPEG0_DEC_MASK << i); + WREG32_SOC15(JPEG, jpeg_inst, regJPEG_CGC_GATE, data); +} + +static void jpeg_v4_0_3_enable_clock_gating(struct amdgpu_device *adev, int inst_idx) +{ + int i, jpeg_inst; + uint32_t data; + + jpeg_inst = GET_INST(JPEG, inst_idx); + data = RREG32_SOC15(JPEG, jpeg_inst, regJPEG_CGC_CTRL); + if (adev->cg_flags & AMD_CG_SUPPORT_JPEG_MGCG) { + data |= 1 << JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT; + data |= (JPEG_CGC_CTRL__JPEG0_DEC_MODE_MASK << 1); + } else { + data &= ~JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT; + } + + data |= 1 << JPEG_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT; + data |= 4 << JPEG_CGC_CTRL__CLK_OFF_DELAY__SHIFT; + WREG32_SOC15(JPEG, jpeg_inst, regJPEG_CGC_CTRL, data); + + data = RREG32_SOC15(JPEG, jpeg_inst, regJPEG_CGC_GATE); + data |= (JPEG_CGC_GATE__JMCIF_MASK | JPEG_CGC_GATE__JRBBM_MASK); + for (i = 0; i < adev->jpeg.num_jpeg_rings; ++i) + data |= (JPEG_CGC_GATE__JPEG0_DEC_MASK << i); + WREG32_SOC15(JPEG, jpeg_inst, regJPEG_CGC_GATE, data); +} + +/** + * jpeg_v4_0_3_start - start JPEG block + * + * @adev: amdgpu_device pointer + * + * Setup and start the JPEG block + */ +static int jpeg_v4_0_3_start(struct amdgpu_device *adev) +{ + struct amdgpu_ring *ring; + int i, j, jpeg_inst; + + for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { + jpeg_inst = GET_INST(JPEG, i); + + WREG32_SOC15(JPEG, jpeg_inst, regUVD_PGFSM_CONFIG, + 1 << UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG__SHIFT); + SOC15_WAIT_ON_RREG( + JPEG, jpeg_inst, regUVD_PGFSM_STATUS, + UVD_PGFSM_STATUS__UVDJ_PWR_ON + << UVD_PGFSM_STATUS__UVDJ_PWR_STATUS__SHIFT, + UVD_PGFSM_STATUS__UVDJ_PWR_STATUS_MASK); + + /* disable anti hang mechanism */ + WREG32_P(SOC15_REG_OFFSET(JPEG, jpeg_inst, + regUVD_JPEG_POWER_STATUS), + 0, ~UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK); + + /* JPEG disable CGC */ + jpeg_v4_0_3_disable_clock_gating(adev, i); + + /* MJPEG global tiling registers */ + WREG32_SOC15(JPEG, jpeg_inst, regJPEG_DEC_GFX8_ADDR_CONFIG, + adev->gfx.config.gb_addr_config); + WREG32_SOC15(JPEG, jpeg_inst, regJPEG_DEC_GFX10_ADDR_CONFIG, + adev->gfx.config.gb_addr_config); + + /* enable JMI channel */ + WREG32_P(SOC15_REG_OFFSET(JPEG, jpeg_inst, regUVD_JMI_CNTL), 0, + ~UVD_JMI_CNTL__SOFT_RESET_MASK); + + for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) { + unsigned int reg_offset = (j?(0x40 * j - 0xc80):0); + + ring = &adev->jpeg.inst[i].ring_dec[j]; + + /* enable System Interrupt for JRBC */ + WREG32_P(SOC15_REG_OFFSET(JPEG, jpeg_inst, + regJPEG_SYS_INT_EN), + JPEG_SYS_INT_EN__DJRBC0_MASK << j, + ~(JPEG_SYS_INT_EN__DJRBC0_MASK << j)); + + WREG32_SOC15_OFFSET(JPEG, jpeg_inst, + regUVD_JMI0_UVD_LMI_JRBC_RB_VMID, + reg_offset, 0); + WREG32_SOC15_OFFSET(JPEG, jpeg_inst, + regUVD_JRBC0_UVD_JRBC_RB_CNTL, + reg_offset, + (0x00000001L | 0x00000002L)); + WREG32_SOC15_OFFSET( + JPEG, jpeg_inst, + regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW, + reg_offset, lower_32_bits(ring->gpu_addr)); + WREG32_SOC15_OFFSET( + JPEG, jpeg_inst, + regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH, + reg_offset, upper_32_bits(ring->gpu_addr)); + WREG32_SOC15_OFFSET(JPEG, jpeg_inst, + regUVD_JRBC0_UVD_JRBC_RB_RPTR, + reg_offset, 0); + WREG32_SOC15_OFFSET(JPEG, jpeg_inst, + regUVD_JRBC0_UVD_JRBC_RB_WPTR, + reg_offset, 0); + WREG32_SOC15_OFFSET(JPEG, jpeg_inst, + regUVD_JRBC0_UVD_JRBC_RB_CNTL, + reg_offset, 0x00000002L); + WREG32_SOC15_OFFSET(JPEG, jpeg_inst, + regUVD_JRBC0_UVD_JRBC_RB_SIZE, + reg_offset, ring->ring_size / 4); + ring->wptr = RREG32_SOC15_OFFSET( + JPEG, jpeg_inst, regUVD_JRBC0_UVD_JRBC_RB_WPTR, + reg_offset); + } + } + + return 0; +} + +/** + * jpeg_v4_0_3_stop - stop JPEG block + * + * @adev: amdgpu_device pointer + * + * stop the JPEG block + */ +static int jpeg_v4_0_3_stop(struct amdgpu_device *adev) +{ + int i, jpeg_inst; + + for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { + jpeg_inst = GET_INST(JPEG, i); + /* reset JMI */ + WREG32_P(SOC15_REG_OFFSET(JPEG, jpeg_inst, regUVD_JMI_CNTL), + UVD_JMI_CNTL__SOFT_RESET_MASK, + ~UVD_JMI_CNTL__SOFT_RESET_MASK); + + jpeg_v4_0_3_enable_clock_gating(adev, i); + + /* enable anti hang mechanism */ + WREG32_P(SOC15_REG_OFFSET(JPEG, jpeg_inst, + regUVD_JPEG_POWER_STATUS), + UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK, + ~UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK); + + WREG32_SOC15(JPEG, jpeg_inst, regUVD_PGFSM_CONFIG, + 2 << UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG__SHIFT); + SOC15_WAIT_ON_RREG( + JPEG, jpeg_inst, regUVD_PGFSM_STATUS, + UVD_PGFSM_STATUS__UVDJ_PWR_OFF + << UVD_PGFSM_STATUS__UVDJ_PWR_STATUS__SHIFT, + UVD_PGFSM_STATUS__UVDJ_PWR_STATUS_MASK); + } + + return 0; +} + +/** + * jpeg_v4_0_3_dec_ring_get_rptr - get read pointer + * + * @ring: amdgpu_ring pointer + * + * Returns the current hardware read pointer + */ +static uint64_t jpeg_v4_0_3_dec_ring_get_rptr(struct amdgpu_ring *ring) +{ + struct amdgpu_device *adev = ring->adev; + + return RREG32_SOC15_OFFSET( + JPEG, GET_INST(JPEG, ring->me), regUVD_JRBC0_UVD_JRBC_RB_RPTR, + ring->pipe ? (0x40 * ring->pipe - 0xc80) : 0); +} + +/** + * jpeg_v4_0_3_dec_ring_get_wptr - get write pointer + * + * @ring: amdgpu_ring pointer + * + * Returns the current hardware write pointer + */ +static uint64_t jpeg_v4_0_3_dec_ring_get_wptr(struct amdgpu_ring *ring) +{ + struct amdgpu_device *adev = ring->adev; + + if (ring->use_doorbell) + return adev->wb.wb[ring->wptr_offs]; + else + return RREG32_SOC15_OFFSET( + JPEG, GET_INST(JPEG, ring->me), + regUVD_JRBC0_UVD_JRBC_RB_WPTR, + ring->pipe ? (0x40 * ring->pipe - 0xc80) : 0); +} + +/** + * jpeg_v4_0_3_dec_ring_set_wptr - set write pointer + * + * @ring: amdgpu_ring pointer + * + * Commits the write pointer to the hardware + */ +static void jpeg_v4_0_3_dec_ring_set_wptr(struct amdgpu_ring *ring) +{ + struct amdgpu_device *adev = ring->adev; + + if (ring->use_doorbell) { + adev->wb.wb[ring->wptr_offs] = lower_32_bits(ring->wptr); + WDOORBELL32(ring->doorbell_index, lower_32_bits(ring->wptr)); + } else { + WREG32_SOC15_OFFSET(JPEG, GET_INST(JPEG, ring->me), + regUVD_JRBC0_UVD_JRBC_RB_WPTR, + (ring->pipe ? (0x40 * ring->pipe - 0xc80) : + 0), + lower_32_bits(ring->wptr)); + } +} + +/** + * jpeg_v4_0_3_dec_ring_insert_start - insert a start command + * + * @ring: amdgpu_ring pointer + * + * Write a start command to the ring. + */ +static void jpeg_v4_0_3_dec_ring_insert_start(struct amdgpu_ring *ring) +{ + amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, 0x62a04); /* PCTL0_MMHUB_DEEPSLEEP_IB */ + + amdgpu_ring_write(ring, PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, 0x80004000); +} + +/** + * jpeg_v4_0_3_dec_ring_insert_end - insert a end command + * + * @ring: amdgpu_ring pointer + * + * Write a end command to the ring. + */ +static void jpeg_v4_0_3_dec_ring_insert_end(struct amdgpu_ring *ring) +{ + amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, 0x62a04); + + amdgpu_ring_write(ring, PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, 0x00004000); +} + +/** + * jpeg_v4_0_3_dec_ring_emit_fence - emit an fence & trap command + * + * @ring: amdgpu_ring pointer + * @addr: address + * @seq: sequence number + * @flags: fence related flags + * + * Write a fence and a trap command to the ring. + */ +static void jpeg_v4_0_3_dec_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq, + unsigned int flags) +{ + WARN_ON(flags & AMDGPU_FENCE_FLAG_64BIT); + + amdgpu_ring_write(ring, PACKETJ(regUVD_JPEG_GPCOM_DATA0_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, seq); + + amdgpu_ring_write(ring, PACKETJ(regUVD_JPEG_GPCOM_DATA1_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, seq); + + amdgpu_ring_write(ring, PACKETJ(regUVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, lower_32_bits(addr)); + + amdgpu_ring_write(ring, PACKETJ(regUVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, upper_32_bits(addr)); + + amdgpu_ring_write(ring, PACKETJ(regUVD_JPEG_GPCOM_CMD_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, 0x8); + + amdgpu_ring_write(ring, PACKETJ(regUVD_JPEG_GPCOM_CMD_INTERNAL_OFFSET, + 0, PACKETJ_CONDITION_CHECK0, PACKETJ_TYPE4)); + amdgpu_ring_write(ring, 0); + + if (ring->adev->jpeg.inst[ring->me].aid_id) { + amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_MCM_ADDR_INTERNAL_OFFSET, + 0, PACKETJ_CONDITION_CHECK0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, 0x4); + } else { + amdgpu_ring_write(ring, PACKETJ(0, 0, 0, PACKETJ_TYPE6)); + amdgpu_ring_write(ring, 0); + } + + amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, 0x3fbc); + + if (ring->adev->jpeg.inst[ring->me].aid_id) { + amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_MCM_ADDR_INTERNAL_OFFSET, + 0, PACKETJ_CONDITION_CHECK0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, 0x0); + } else { + amdgpu_ring_write(ring, PACKETJ(0, 0, 0, PACKETJ_TYPE6)); + amdgpu_ring_write(ring, 0); + } + + amdgpu_ring_write(ring, PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, 0x1); + + amdgpu_ring_write(ring, PACKETJ(0, 0, 0, PACKETJ_TYPE7)); + amdgpu_ring_write(ring, 0); +} + +/** + * jpeg_v4_0_3_dec_ring_emit_ib - execute indirect buffer + * + * @ring: amdgpu_ring pointer + * @job: job to retrieve vmid from + * @ib: indirect buffer to execute + * @flags: unused + * + * Write ring commands to execute the indirect buffer. + */ +static void jpeg_v4_0_3_dec_ring_emit_ib(struct amdgpu_ring *ring, + struct amdgpu_job *job, + struct amdgpu_ib *ib, + uint32_t flags) +{ + unsigned int vmid = AMDGPU_JOB_GET_VMID(job); + + amdgpu_ring_write(ring, PACKETJ(regUVD_LMI_JRBC_IB_VMID_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, (vmid | (vmid << 4))); + + amdgpu_ring_write(ring, PACKETJ(regUVD_LMI_JPEG_VMID_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, (vmid | (vmid << 4))); + + amdgpu_ring_write(ring, PACKETJ(regUVD_LMI_JRBC_IB_64BIT_BAR_LOW_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, lower_32_bits(ib->gpu_addr)); + + amdgpu_ring_write(ring, PACKETJ(regUVD_LMI_JRBC_IB_64BIT_BAR_HIGH_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr)); + + amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_IB_SIZE_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, ib->length_dw); + + amdgpu_ring_write(ring, PACKETJ(regUVD_LMI_JRBC_RB_MEM_RD_64BIT_BAR_LOW_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, lower_32_bits(ring->gpu_addr)); + + amdgpu_ring_write(ring, PACKETJ(regUVD_LMI_JRBC_RB_MEM_RD_64BIT_BAR_HIGH_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, upper_32_bits(ring->gpu_addr)); + + amdgpu_ring_write(ring, PACKETJ(0, 0, PACKETJ_CONDITION_CHECK0, PACKETJ_TYPE2)); + amdgpu_ring_write(ring, 0); + + amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_RB_COND_RD_TIMER_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, 0x01400200); + + amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_RB_REF_DATA_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, 0x2); + + amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_STATUS_INTERNAL_OFFSET, + 0, PACKETJ_CONDITION_CHECK3, PACKETJ_TYPE3)); + amdgpu_ring_write(ring, 0x2); +} + +static void jpeg_v4_0_3_dec_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg, + uint32_t val, uint32_t mask) +{ + uint32_t reg_offset = (reg << 2); + + amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_RB_COND_RD_TIMER_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, 0x01400200); + + amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_RB_REF_DATA_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, val); + + amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + if (reg_offset >= 0x10000 && reg_offset <= 0x105ff) { + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, + PACKETJ((reg_offset >> 2), 0, 0, PACKETJ_TYPE3)); + } else { + amdgpu_ring_write(ring, reg_offset); + amdgpu_ring_write(ring, PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR, + 0, 0, PACKETJ_TYPE3)); + } + amdgpu_ring_write(ring, mask); +} + +static void jpeg_v4_0_3_dec_ring_emit_vm_flush(struct amdgpu_ring *ring, + unsigned int vmid, uint64_t pd_addr) +{ + struct amdgpu_vmhub *hub = &ring->adev->vmhub[ring->vm_hub]; + uint32_t data0, data1, mask; + + pd_addr = amdgpu_gmc_emit_flush_gpu_tlb(ring, vmid, pd_addr); + + /* wait for register write */ + data0 = hub->ctx0_ptb_addr_lo32 + vmid * hub->ctx_addr_distance; + data1 = lower_32_bits(pd_addr); + mask = 0xffffffff; + jpeg_v4_0_3_dec_ring_emit_reg_wait(ring, data0, data1, mask); +} + +static void jpeg_v4_0_3_dec_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg, uint32_t val) +{ + uint32_t reg_offset = (reg << 2); + + amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET, + 0, 0, PACKETJ_TYPE0)); + if (reg_offset >= 0x10000 && reg_offset <= 0x105ff) { + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, + PACKETJ((reg_offset >> 2), 0, 0, PACKETJ_TYPE0)); + } else { + amdgpu_ring_write(ring, reg_offset); + amdgpu_ring_write(ring, PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR, + 0, 0, PACKETJ_TYPE0)); + } + amdgpu_ring_write(ring, val); +} + +static void jpeg_v4_0_3_dec_ring_nop(struct amdgpu_ring *ring, uint32_t count) +{ + int i; + + WARN_ON(ring->wptr % 2 || count % 2); + + for (i = 0; i < count / 2; i++) { + amdgpu_ring_write(ring, PACKETJ(0, 0, 0, PACKETJ_TYPE6)); + amdgpu_ring_write(ring, 0); + } +} + +static bool jpeg_v4_0_3_is_idle(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + bool ret = false; + int i, j; + + for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { + for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) { + unsigned int reg_offset = (j?(0x40 * j - 0xc80):0); + + ret &= ((RREG32_SOC15_OFFSET( + JPEG, GET_INST(JPEG, i), + regUVD_JRBC0_UVD_JRBC_STATUS, + reg_offset) & + UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK) == + UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK); + } + } + + return ret; +} + +static int jpeg_v4_0_3_wait_for_idle(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int ret = 0; + int i, j; + + for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { + for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) { + unsigned int reg_offset = (j?(0x40 * j - 0xc80):0); + + ret &= SOC15_WAIT_ON_RREG_OFFSET( + JPEG, GET_INST(JPEG, i), + regUVD_JRBC0_UVD_JRBC_STATUS, reg_offset, + UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK, + UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK); + } + } + return ret; +} + +static int jpeg_v4_0_3_set_clockgating_state(void *handle, + enum amd_clockgating_state state) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + bool enable = (state == AMD_CG_STATE_GATE) ? true : false; + int i; + + for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { + if (enable) { + if (!jpeg_v4_0_3_is_idle(handle)) + return -EBUSY; + jpeg_v4_0_3_enable_clock_gating(adev, i); + } else { + jpeg_v4_0_3_disable_clock_gating(adev, i); + } + } + return 0; +} + +static int jpeg_v4_0_3_set_powergating_state(void *handle, + enum amd_powergating_state state) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int ret; + + if (state == adev->jpeg.cur_state) + return 0; + + if (state == AMD_PG_STATE_GATE) + ret = jpeg_v4_0_3_stop(adev); + else + ret = jpeg_v4_0_3_start(adev); + + if (!ret) + adev->jpeg.cur_state = state; + + return ret; +} + +static int jpeg_v4_0_3_set_interrupt_state(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + unsigned int type, + enum amdgpu_interrupt_state state) +{ + return 0; +} + +static int jpeg_v4_0_3_process_interrupt(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + struct amdgpu_iv_entry *entry) +{ + uint32_t i, inst; + + i = node_id_to_phys_map[entry->node_id]; + DRM_DEV_DEBUG(adev->dev, "IH: JPEG TRAP\n"); + + for (inst = 0; inst < adev->jpeg.num_jpeg_inst; ++inst) + if (adev->jpeg.inst[inst].aid_id == i) + break; + + if (inst >= adev->jpeg.num_jpeg_inst) { + dev_WARN_ONCE(adev->dev, 1, + "Interrupt received for unknown JPEG instance %d", + entry->node_id); + return 0; + } + + switch (entry->src_id) { + case VCN_4_0__SRCID__JPEG_DECODE: + amdgpu_fence_process(&adev->jpeg.inst[inst].ring_dec[0]); + break; + case VCN_4_0__SRCID__JPEG1_DECODE: + amdgpu_fence_process(&adev->jpeg.inst[inst].ring_dec[1]); + break; + case VCN_4_0__SRCID__JPEG2_DECODE: + amdgpu_fence_process(&adev->jpeg.inst[inst].ring_dec[2]); + break; + case VCN_4_0__SRCID__JPEG3_DECODE: + amdgpu_fence_process(&adev->jpeg.inst[inst].ring_dec[3]); + break; + case VCN_4_0__SRCID__JPEG4_DECODE: + amdgpu_fence_process(&adev->jpeg.inst[inst].ring_dec[4]); + break; + case VCN_4_0__SRCID__JPEG5_DECODE: + amdgpu_fence_process(&adev->jpeg.inst[inst].ring_dec[5]); + break; + case VCN_4_0__SRCID__JPEG6_DECODE: + amdgpu_fence_process(&adev->jpeg.inst[inst].ring_dec[6]); + break; + case VCN_4_0__SRCID__JPEG7_DECODE: + amdgpu_fence_process(&adev->jpeg.inst[inst].ring_dec[7]); + break; + default: + DRM_DEV_ERROR(adev->dev, "Unhandled interrupt: %d %d\n", + entry->src_id, entry->src_data[0]); + break; + } + + return 0; +} + +static const struct amd_ip_funcs jpeg_v4_0_3_ip_funcs = { + .name = "jpeg_v4_0_3", + .early_init = jpeg_v4_0_3_early_init, + .late_init = NULL, + .sw_init = jpeg_v4_0_3_sw_init, + .sw_fini = jpeg_v4_0_3_sw_fini, + .hw_init = jpeg_v4_0_3_hw_init, + .hw_fini = jpeg_v4_0_3_hw_fini, + .suspend = jpeg_v4_0_3_suspend, + .resume = jpeg_v4_0_3_resume, + .is_idle = jpeg_v4_0_3_is_idle, + .wait_for_idle = jpeg_v4_0_3_wait_for_idle, + .check_soft_reset = NULL, + .pre_soft_reset = NULL, + .soft_reset = NULL, + .post_soft_reset = NULL, + .set_clockgating_state = jpeg_v4_0_3_set_clockgating_state, + .set_powergating_state = jpeg_v4_0_3_set_powergating_state, +}; + +static const struct amdgpu_ring_funcs jpeg_v4_0_3_dec_ring_vm_funcs = { + .type = AMDGPU_RING_TYPE_VCN_JPEG, + .align_mask = 0xf, + .get_rptr = jpeg_v4_0_3_dec_ring_get_rptr, + .get_wptr = jpeg_v4_0_3_dec_ring_get_wptr, + .set_wptr = jpeg_v4_0_3_dec_ring_set_wptr, + .emit_frame_size = + SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 + + SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 + + 8 + /* jpeg_v4_0_3_dec_ring_emit_vm_flush */ + 22 + 22 + /* jpeg_v4_0_3_dec_ring_emit_fence x2 vm fence */ + 8 + 16, + .emit_ib_size = 22, /* jpeg_v4_0_3_dec_ring_emit_ib */ + .emit_ib = jpeg_v4_0_3_dec_ring_emit_ib, + .emit_fence = jpeg_v4_0_3_dec_ring_emit_fence, + .emit_vm_flush = jpeg_v4_0_3_dec_ring_emit_vm_flush, + .test_ring = amdgpu_jpeg_dec_ring_test_ring, + .test_ib = amdgpu_jpeg_dec_ring_test_ib, + .insert_nop = jpeg_v4_0_3_dec_ring_nop, + .insert_start = jpeg_v4_0_3_dec_ring_insert_start, + .insert_end = jpeg_v4_0_3_dec_ring_insert_end, + .pad_ib = amdgpu_ring_generic_pad_ib, + .begin_use = amdgpu_jpeg_ring_begin_use, + .end_use = amdgpu_jpeg_ring_end_use, + .emit_wreg = jpeg_v4_0_3_dec_ring_emit_wreg, + .emit_reg_wait = jpeg_v4_0_3_dec_ring_emit_reg_wait, + .emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper, +}; + +static void jpeg_v4_0_3_set_dec_ring_funcs(struct amdgpu_device *adev) +{ + int i, j, jpeg_inst; + + for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { + for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) { + adev->jpeg.inst[i].ring_dec[j].funcs = &jpeg_v4_0_3_dec_ring_vm_funcs; + adev->jpeg.inst[i].ring_dec[j].me = i; + adev->jpeg.inst[i].ring_dec[j].pipe = j; + } + jpeg_inst = GET_INST(JPEG, i); + adev->jpeg.inst[i].aid_id = + jpeg_inst / adev->jpeg.num_inst_per_aid; + } + DRM_DEV_INFO(adev->dev, "JPEG decode is enabled in VM mode\n"); +} + +static const struct amdgpu_irq_src_funcs jpeg_v4_0_3_irq_funcs = { + .set = jpeg_v4_0_3_set_interrupt_state, + .process = jpeg_v4_0_3_process_interrupt, +}; + +static void jpeg_v4_0_3_set_irq_funcs(struct amdgpu_device *adev) +{ + int i; + + for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { + adev->jpeg.inst->irq.num_types += adev->jpeg.num_jpeg_rings; + } + adev->jpeg.inst->irq.funcs = &jpeg_v4_0_3_irq_funcs; +} + +const struct amdgpu_ip_block_version jpeg_v4_0_3_ip_block = { + .type = AMD_IP_BLOCK_TYPE_JPEG, + .major = 4, + .minor = 0, + .rev = 3, + .funcs = &jpeg_v4_0_3_ip_funcs, +}; + +static const struct amdgpu_ras_err_status_reg_entry jpeg_v4_0_3_ue_reg_list[] = { + {AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG0S, regVCN_UE_ERR_STATUS_HI_JPEG0S), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG0S"}, + {AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG0D, regVCN_UE_ERR_STATUS_HI_JPEG0D), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG0D"}, + {AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG1S, regVCN_UE_ERR_STATUS_HI_JPEG1S), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG1S"}, + {AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG1D, regVCN_UE_ERR_STATUS_HI_JPEG1D), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG1D"}, + {AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG2S, regVCN_UE_ERR_STATUS_HI_JPEG2S), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG2S"}, + {AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG2D, regVCN_UE_ERR_STATUS_HI_JPEG2D), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG2D"}, + {AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG3S, regVCN_UE_ERR_STATUS_HI_JPEG3S), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG3S"}, + {AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG3D, regVCN_UE_ERR_STATUS_HI_JPEG3D), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG3D"}, + {AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG4S, regVCN_UE_ERR_STATUS_HI_JPEG4S), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG4S"}, + {AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG4D, regVCN_UE_ERR_STATUS_HI_JPEG4D), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG4D"}, + {AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG5S, regVCN_UE_ERR_STATUS_HI_JPEG5S), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG5S"}, + {AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG5D, regVCN_UE_ERR_STATUS_HI_JPEG5D), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG5D"}, + {AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG6S, regVCN_UE_ERR_STATUS_HI_JPEG6S), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG6S"}, + {AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG6D, regVCN_UE_ERR_STATUS_HI_JPEG6D), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG6D"}, + {AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG7S, regVCN_UE_ERR_STATUS_HI_JPEG7S), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG7S"}, + {AMDGPU_RAS_REG_ENTRY(JPEG, 0, regVCN_UE_ERR_STATUS_LO_JPEG7D, regVCN_UE_ERR_STATUS_HI_JPEG7D), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "JPEG7D"}, +}; + +static void jpeg_v4_0_3_inst_query_ras_error_count(struct amdgpu_device *adev, + uint32_t jpeg_inst, + void *ras_err_status) +{ + struct ras_err_data *err_data = (struct ras_err_data *)ras_err_status; + + /* jpeg v4_0_3 only support uncorrectable errors */ + amdgpu_ras_inst_query_ras_error_count(adev, + jpeg_v4_0_3_ue_reg_list, + ARRAY_SIZE(jpeg_v4_0_3_ue_reg_list), + NULL, 0, GET_INST(VCN, jpeg_inst), + AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE, + &err_data->ue_count); +} + +static void jpeg_v4_0_3_query_ras_error_count(struct amdgpu_device *adev, + void *ras_err_status) +{ + uint32_t i; + + if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG)) { + dev_warn(adev->dev, "JPEG RAS is not supported\n"); + return; + } + + for (i = 0; i < adev->jpeg.num_jpeg_inst; i++) + jpeg_v4_0_3_inst_query_ras_error_count(adev, i, ras_err_status); +} + +static void jpeg_v4_0_3_inst_reset_ras_error_count(struct amdgpu_device *adev, + uint32_t jpeg_inst) +{ + amdgpu_ras_inst_reset_ras_error_count(adev, + jpeg_v4_0_3_ue_reg_list, + ARRAY_SIZE(jpeg_v4_0_3_ue_reg_list), + GET_INST(VCN, jpeg_inst)); +} + +static void jpeg_v4_0_3_reset_ras_error_count(struct amdgpu_device *adev) +{ + uint32_t i; + + if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG)) { + dev_warn(adev->dev, "JPEG RAS is not supported\n"); + return; + } + + for (i = 0; i < adev->jpeg.num_jpeg_inst; i++) + jpeg_v4_0_3_inst_reset_ras_error_count(adev, i); +} + +static const struct amdgpu_ras_block_hw_ops jpeg_v4_0_3_ras_hw_ops = { + .query_ras_error_count = jpeg_v4_0_3_query_ras_error_count, + .reset_ras_error_count = jpeg_v4_0_3_reset_ras_error_count, +}; + +static struct amdgpu_jpeg_ras jpeg_v4_0_3_ras = { + .ras_block = { + .hw_ops = &jpeg_v4_0_3_ras_hw_ops, + }, +}; + +static void jpeg_v4_0_3_set_ras_funcs(struct amdgpu_device *adev) +{ + adev->jpeg.ras = &jpeg_v4_0_3_ras; +} diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h new file mode 100644 index 0000000000000000000000000000000000000000..22483dc663518f3a22572daf695dc37d17462f57 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.h @@ -0,0 +1,51 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef __JPEG_V4_0_3_H__ +#define __JPEG_V4_0_3_H__ + +#define regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET 0x1bfff +#define regUVD_JPEG_GPCOM_CMD_INTERNAL_OFFSET 0x404d +#define regUVD_JPEG_GPCOM_DATA0_INTERNAL_OFFSET 0x404e +#define regUVD_JPEG_GPCOM_DATA1_INTERNAL_OFFSET 0x404f +#define regUVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_INTERNAL_OFFSET 0x40ab +#define regUVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_INTERNAL_OFFSET 0x40ac +#define regUVD_LMI_JRBC_IB_VMID_INTERNAL_OFFSET 0x40a4 +#define regUVD_LMI_JPEG_VMID_INTERNAL_OFFSET 0x40a6 +#define regUVD_LMI_JRBC_IB_64BIT_BAR_LOW_INTERNAL_OFFSET 0x40b6 +#define regUVD_LMI_JRBC_IB_64BIT_BAR_HIGH_INTERNAL_OFFSET 0x40b7 +#define regUVD_JRBC_IB_SIZE_INTERNAL_OFFSET 0x4082 +#define regUVD_LMI_JRBC_RB_MEM_RD_64BIT_BAR_LOW_INTERNAL_OFFSET 0x42d4 +#define regUVD_LMI_JRBC_RB_MEM_RD_64BIT_BAR_HIGH_INTERNAL_OFFSET 0x42d5 +#define regUVD_JRBC_RB_COND_RD_TIMER_INTERNAL_OFFSET 0x4085 +#define regUVD_JRBC_RB_REF_DATA_INTERNAL_OFFSET 0x4084 +#define regUVD_JRBC_STATUS_INTERNAL_OFFSET 0x4089 +#define regUVD_JPEG_PITCH_INTERNAL_OFFSET 0x4043 +#define regUVD_JRBC0_UVD_JRBC_SCRATCH0_INTERNAL_OFFSET 0x4094 +#define regUVD_JRBC_EXTERNAL_MCM_ADDR_INTERNAL_OFFSET 0x1bffe + +#define JRBC_DEC_EXTERNAL_REG_WRITE_ADDR 0x18000 + +extern const struct amdgpu_ip_block_version jpeg_v4_0_3_ip_block; + +#endif /* __JPEG_V4_0_3_H__ */ diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c index 2e2062636d5f6882f8794b4cb242e0bc36590424..36a123e6c8ee752e7b1633506743044f24abdfc2 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c @@ -149,7 +149,7 @@ static int mes_v10_1_add_hw_queue(struct amdgpu_mes *mes, { struct amdgpu_device *adev = mes->adev; union MESAPI__ADD_QUEUE mes_add_queue_pkt; - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; uint32_t vm_cntx_cntl = hub->vm_cntx_cntl; memset(&mes_add_queue_pkt, 0, sizeof(mes_add_queue_pkt)); @@ -632,6 +632,8 @@ static int mes_v10_1_mqd_init(struct amdgpu_ring *ring) uint64_t hqd_gpu_addr, wb_gpu_addr, eop_base_addr; uint32_t tmp; + memset(mqd, 0, sizeof(*mqd)); + mqd->header = 0xC0310800; mqd->compute_pipelinestat_enable = 0x00000001; mqd->compute_static_thread_mgmt_se0 = 0xffffffff; @@ -728,6 +730,7 @@ static int mes_v10_1_mqd_init(struct amdgpu_ring *ring) /* offset: 184 - this is used for CP_HQD_GFX_CONTROL */ mqd->cp_hqd_suspend_cntl_stack_offset = tmp; + amdgpu_device_flush_hdp(ring->adev, NULL); return 0; } @@ -797,8 +800,8 @@ static void mes_v10_1_queue_init_register(struct amdgpu_ring *ring) static int mes_v10_1_kiq_enable_queue(struct amdgpu_device *adev) { - struct amdgpu_kiq *kiq = &adev->gfx.kiq; - struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; + struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring; int r; if (!kiq->pmf || !kiq->pmf->kiq_map_queues) @@ -812,13 +815,7 @@ static int mes_v10_1_kiq_enable_queue(struct amdgpu_device *adev) kiq->pmf->kiq_map_queues(kiq_ring, &adev->mes.ring); - r = amdgpu_ring_test_ring(kiq_ring); - if (r) { - DRM_ERROR("kfq enable failed\n"); - kiq_ring->sched.ready = false; - } - - return r; + return amdgpu_ring_test_helper(kiq_ring); } static int mes_v10_1_queue_init(struct amdgpu_device *adev) @@ -863,9 +860,9 @@ static int mes_v10_1_kiq_ring_init(struct amdgpu_device *adev) { struct amdgpu_ring *ring; - spin_lock_init(&adev->gfx.kiq.ring_lock); + spin_lock_init(&adev->gfx.kiq[0].ring_lock); - ring = &adev->gfx.kiq.ring; + ring = &adev->gfx.kiq[0].ring; ring->me = 3; ring->pipe = 1; @@ -891,7 +888,7 @@ static int mes_v10_1_mqd_sw_init(struct amdgpu_device *adev, struct amdgpu_ring *ring; if (pipe == AMDGPU_MES_KIQ_PIPE) - ring = &adev->gfx.kiq.ring; + ring = &adev->gfx.kiq[0].ring; else if (pipe == AMDGPU_MES_SCHED_PIPE) ring = &adev->mes.ring; else @@ -901,6 +898,7 @@ static int mes_v10_1_mqd_sw_init(struct amdgpu_device *adev, return 0; r = amdgpu_bo_create_kernel(adev, mqd_size, PAGE_SIZE, + AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT, &ring->mqd_obj, &ring->mqd_gpu_addr, &ring->mqd_ptr); if (r) { @@ -974,15 +972,15 @@ static int mes_v10_1_sw_fini(void *handle) amdgpu_ucode_release(&adev->mes.fw[pipe]); } - amdgpu_bo_free_kernel(&adev->gfx.kiq.ring.mqd_obj, - &adev->gfx.kiq.ring.mqd_gpu_addr, - &adev->gfx.kiq.ring.mqd_ptr); + amdgpu_bo_free_kernel(&adev->gfx.kiq[0].ring.mqd_obj, + &adev->gfx.kiq[0].ring.mqd_gpu_addr, + &adev->gfx.kiq[0].ring.mqd_ptr); amdgpu_bo_free_kernel(&adev->mes.ring.mqd_obj, &adev->mes.ring.mqd_gpu_addr, &adev->mes.ring.mqd_ptr); - amdgpu_ring_fini(&adev->gfx.kiq.ring); + amdgpu_ring_fini(&adev->gfx.kiq[0].ring); amdgpu_ring_fini(&adev->mes.ring); amdgpu_mes_fini(adev); @@ -1038,7 +1036,7 @@ static int mes_v10_1_kiq_hw_init(struct amdgpu_device *adev) mes_v10_1_enable(adev, true); - mes_v10_1_kiq_setting(&adev->gfx.kiq.ring); + mes_v10_1_kiq_setting(&adev->gfx.kiq[0].ring); r = mes_v10_1_queue_init(adev); if (r) @@ -1090,7 +1088,7 @@ static int mes_v10_1_hw_init(void *handle) * MES uses KIQ ring exclusively so driver cannot access KIQ ring * with MES enabled. */ - adev->gfx.kiq.ring.sched.ready = false; + adev->gfx.kiq[0].ring.sched.ready = false; adev->mes.ring.sched.ready = true; return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c index 45280f047180a8e05db79d9231aa983645535770..1bdaa00c0b466ff2aabfcf77ae41f781ad4a69ef 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c @@ -164,7 +164,7 @@ static int mes_v11_0_add_hw_queue(struct amdgpu_mes *mes, { struct amdgpu_device *adev = mes->adev; union MESAPI__ADD_QUEUE mes_add_queue_pkt; - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; uint32_t vm_cntx_cntl = hub->vm_cntx_cntl; memset(&mes_add_queue_pkt, 0, sizeof(mes_add_queue_pkt)); @@ -202,17 +202,14 @@ static int mes_v11_0_add_hw_queue(struct amdgpu_mes *mes, mes_add_queue_pkt.gws_size = input->gws_size; mes_add_queue_pkt.trap_handler_addr = input->tba_addr; mes_add_queue_pkt.tma_addr = input->tma_addr; + mes_add_queue_pkt.trap_en = input->trap_en; + mes_add_queue_pkt.skip_process_ctx_clear = input->skip_process_ctx_clear; mes_add_queue_pkt.is_kfd_process = input->is_kfd_process; /* For KFD, gds_size is re-used for queue size (needed in MES for AQL queues) */ mes_add_queue_pkt.is_aql_queue = input->is_aql_queue; mes_add_queue_pkt.gds_size = input->queue_size; - if (!(((adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 4) && - (adev->ip_versions[GC_HWIP][0] >= IP_VERSION(11, 0, 0)) && - (adev->ip_versions[GC_HWIP][0] <= IP_VERSION(11, 0, 3)))) - mes_add_queue_pkt.trap_en = 1; - /* For KFD, gds_size is re-used for queue size (needed in MES for AQL queues) */ mes_add_queue_pkt.is_aql_queue = input->is_aql_queue; mes_add_queue_pkt.gds_size = input->queue_size; @@ -339,6 +336,19 @@ static int mes_v11_0_misc_op(struct amdgpu_mes *mes, misc_pkt.wait_reg_mem.reg_offset1 = input->wrm_reg.reg0; misc_pkt.wait_reg_mem.reg_offset2 = input->wrm_reg.reg1; break; + case MES_MISC_OP_SET_SHADER_DEBUGGER: + misc_pkt.opcode = MESAPI_MISC__SET_SHADER_DEBUGGER; + misc_pkt.set_shader_debugger.process_context_addr = + input->set_shader_debugger.process_context_addr; + misc_pkt.set_shader_debugger.flags.u32all = + input->set_shader_debugger.flags.u32all; + misc_pkt.set_shader_debugger.spi_gdbg_per_vmid_cntl = + input->set_shader_debugger.spi_gdbg_per_vmid_cntl; + memcpy(misc_pkt.set_shader_debugger.tcp_watch_cntl, + input->set_shader_debugger.tcp_watch_cntl, + sizeof(misc_pkt.set_shader_debugger.tcp_watch_cntl)); + misc_pkt.set_shader_debugger.trap_en = input->set_shader_debugger.trap_en; + break; default: DRM_ERROR("unsupported misc op (%d) \n", input->op); return -EINVAL; @@ -704,6 +714,8 @@ static int mes_v11_0_mqd_init(struct amdgpu_ring *ring) uint64_t hqd_gpu_addr, wb_gpu_addr, eop_base_addr; uint32_t tmp; + memset(mqd, 0, sizeof(*mqd)); + mqd->header = 0xC0310800; mqd->compute_pipelinestat_enable = 0x00000001; mqd->compute_static_thread_mgmt_se0 = 0xffffffff; @@ -797,6 +809,7 @@ static int mes_v11_0_mqd_init(struct amdgpu_ring *ring) mqd->cp_hqd_iq_timer = regCP_HQD_IQ_TIMER_DEFAULT; mqd->cp_hqd_quantum = regCP_HQD_QUANTUM_DEFAULT; + amdgpu_device_flush_hdp(ring->adev, NULL); return 0; } @@ -864,8 +877,8 @@ static void mes_v11_0_queue_init_register(struct amdgpu_ring *ring) static int mes_v11_0_kiq_enable_queue(struct amdgpu_device *adev) { - struct amdgpu_kiq *kiq = &adev->gfx.kiq; - struct amdgpu_ring *kiq_ring = &adev->gfx.kiq.ring; + struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; + struct amdgpu_ring *kiq_ring = &adev->gfx.kiq[0].ring; int r; if (!kiq->pmf || !kiq->pmf->kiq_map_queues) @@ -879,12 +892,7 @@ static int mes_v11_0_kiq_enable_queue(struct amdgpu_device *adev) kiq->pmf->kiq_map_queues(kiq_ring, &adev->mes.ring); - r = amdgpu_ring_test_ring(kiq_ring); - if (r) { - DRM_ERROR("kfq enable failed\n"); - kiq_ring->sched.ready = false; - } - return r; + return amdgpu_ring_test_helper(kiq_ring); } static int mes_v11_0_queue_init(struct amdgpu_device *adev, @@ -894,7 +902,7 @@ static int mes_v11_0_queue_init(struct amdgpu_device *adev, int r; if (pipe == AMDGPU_MES_KIQ_PIPE) - ring = &adev->gfx.kiq.ring; + ring = &adev->gfx.kiq[0].ring; else if (pipe == AMDGPU_MES_SCHED_PIPE) ring = &adev->mes.ring; else @@ -961,9 +969,9 @@ static int mes_v11_0_kiq_ring_init(struct amdgpu_device *adev) { struct amdgpu_ring *ring; - spin_lock_init(&adev->gfx.kiq.ring_lock); + spin_lock_init(&adev->gfx.kiq[0].ring_lock); - ring = &adev->gfx.kiq.ring; + ring = &adev->gfx.kiq[0].ring; ring->me = 3; ring->pipe = 1; @@ -989,7 +997,7 @@ static int mes_v11_0_mqd_sw_init(struct amdgpu_device *adev, struct amdgpu_ring *ring; if (pipe == AMDGPU_MES_KIQ_PIPE) - ring = &adev->gfx.kiq.ring; + ring = &adev->gfx.kiq[0].ring; else if (pipe == AMDGPU_MES_SCHED_PIPE) ring = &adev->mes.ring; else @@ -999,6 +1007,7 @@ static int mes_v11_0_mqd_sw_init(struct amdgpu_device *adev, return 0; r = amdgpu_bo_create_kernel(adev, mqd_size, PAGE_SIZE, + AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT, &ring->mqd_obj, &ring->mqd_gpu_addr, &ring->mqd_ptr); if (r) { @@ -1074,15 +1083,15 @@ static int mes_v11_0_sw_fini(void *handle) amdgpu_ucode_release(&adev->mes.fw[pipe]); } - amdgpu_bo_free_kernel(&adev->gfx.kiq.ring.mqd_obj, - &adev->gfx.kiq.ring.mqd_gpu_addr, - &adev->gfx.kiq.ring.mqd_ptr); + amdgpu_bo_free_kernel(&adev->gfx.kiq[0].ring.mqd_obj, + &adev->gfx.kiq[0].ring.mqd_gpu_addr, + &adev->gfx.kiq[0].ring.mqd_ptr); amdgpu_bo_free_kernel(&adev->mes.ring.mqd_obj, &adev->mes.ring.mqd_gpu_addr, &adev->mes.ring.mqd_ptr); - amdgpu_ring_fini(&adev->gfx.kiq.ring); + amdgpu_ring_fini(&adev->gfx.kiq[0].ring); amdgpu_ring_fini(&adev->mes.ring); if (adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT) { @@ -1175,7 +1184,7 @@ static int mes_v11_0_kiq_hw_init(struct amdgpu_device *adev) mes_v11_0_enable(adev, true); - mes_v11_0_kiq_setting(&adev->gfx.kiq.ring); + mes_v11_0_kiq_setting(&adev->gfx.kiq[0].ring); r = mes_v11_0_queue_init(adev, AMDGPU_MES_KIQ_PIPE); if (r) @@ -1196,7 +1205,7 @@ static int mes_v11_0_kiq_hw_fini(struct amdgpu_device *adev) } if (amdgpu_sriov_vf(adev)) { - mes_v11_0_kiq_dequeue(&adev->gfx.kiq.ring); + mes_v11_0_kiq_dequeue(&adev->gfx.kiq[0].ring); mes_v11_0_kiq_clear(adev); } @@ -1244,7 +1253,7 @@ static int mes_v11_0_hw_init(void *handle) * MES uses KIQ ring exclusively so driver cannot access KIQ ring * with MES enabled. */ - adev->gfx.kiq.ring.sched.ready = false; + adev->gfx.kiq[0].ring.sched.ready = false; adev->mes.ring.sched.ready = true; return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c index 15e7cbeae75b815b88c9fab0842fcb302ed6d099..fb91b31056cae70585a082cc5f1ee1f92e5a541f 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c @@ -54,7 +54,7 @@ static u64 mmhub_v1_0_get_fb_location(struct amdgpu_device *adev) static void mmhub_v1_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; WREG32_SOC15_OFFSET(MMHUB, 0, mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, hub->ctx_addr_distance * vmid, @@ -229,7 +229,7 @@ static void mmhub_v1_0_disable_identity_aperture(struct amdgpu_device *adev) static void mmhub_v1_0_setup_vmid_config(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; unsigned num_level, block_size; uint32_t tmp; int i; @@ -285,7 +285,7 @@ static void mmhub_v1_0_setup_vmid_config(struct amdgpu_device *adev) static void mmhub_v1_0_program_invalidation(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; unsigned i; for (i = 0; i < 18; ++i) { @@ -338,7 +338,7 @@ static int mmhub_v1_0_gart_enable(struct amdgpu_device *adev) static void mmhub_v1_0_gart_disable(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; u32 tmp; u32 i; @@ -415,7 +415,7 @@ static void mmhub_v1_0_set_fault_enable_default(struct amdgpu_device *adev, bool static void mmhub_v1_0_init(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; hub->ctx0_ptb_addr_lo32 = SOC15_REG_OFFSET(MMHUB, 0, diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c index 73afbf2facc9e7f51015c4853082d04beb51c424..9086f2fdfaf422b4312d47c4fd3096e62cce1e60 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c @@ -54,7 +54,7 @@ static u64 mmhub_v1_7_get_fb_location(struct amdgpu_device *adev) static void mmhub_v1_7_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, hub->ctx_addr_distance * vmid, lower_32_bits(page_table_base)); @@ -261,7 +261,7 @@ static void mmhub_v1_7_disable_identity_aperture(struct amdgpu_device *adev) static void mmhub_v1_7_setup_vmid_config(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; unsigned num_level, block_size; uint32_t tmp; int i; @@ -319,7 +319,7 @@ static void mmhub_v1_7_setup_vmid_config(struct amdgpu_device *adev) static void mmhub_v1_7_program_invalidation(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; unsigned i; for (i = 0; i < 18; ++i) { @@ -348,7 +348,7 @@ static int mmhub_v1_7_gart_enable(struct amdgpu_device *adev) static void mmhub_v1_7_gart_disable(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; u32 tmp; u32 i; @@ -425,7 +425,7 @@ static void mmhub_v1_7_set_fault_enable_default(struct amdgpu_device *adev, bool static void mmhub_v1_7_init(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; hub->ctx0_ptb_addr_lo32 = SOC15_REG_OFFSET(MMHUB, 0, diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c index 342d1702104cb4e63c06ef614485efc6e536626b..5e8b493f86995457158ae1318aea2442211ec8e3 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c @@ -29,6 +29,7 @@ #include "soc15_common.h" #include "soc15.h" +#include "amdgpu_ras.h" #define regVM_L2_CNTL3_DEFAULT 0x80100007 #define regVM_L2_CNTL4_DEFAULT 0x000000c1 @@ -53,18 +54,30 @@ static u64 mmhub_v1_8_get_fb_location(struct amdgpu_device *adev) static void mmhub_v1_8_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; - - WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, - hub->ctx_addr_distance * vmid, lower_32_bits(page_table_base)); + struct amdgpu_vmhub *hub; + u32 inst_mask; + int i; - WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32, - hub->ctx_addr_distance * vmid, upper_32_bits(page_table_base)); + inst_mask = adev->aid_mask; + for_each_inst(i, inst_mask) { + hub = &adev->vmhub[AMDGPU_MMHUB0(i)]; + WREG32_SOC15_OFFSET(MMHUB, i, + regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, + hub->ctx_addr_distance * vmid, + lower_32_bits(page_table_base)); + + WREG32_SOC15_OFFSET(MMHUB, i, + regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32, + hub->ctx_addr_distance * vmid, + upper_32_bits(page_table_base)); + } } static void mmhub_v1_8_init_gart_aperture_regs(struct amdgpu_device *adev) { uint64_t pt_base; + u32 inst_mask; + int i; if (adev->gmc.pdb0_bo) pt_base = amdgpu_gmc_pd_addr(adev->gmc.pdb0_bo); @@ -76,187 +89,248 @@ static void mmhub_v1_8_init_gart_aperture_regs(struct amdgpu_device *adev) /* If use GART for FB translation, vmid0 page table covers both * vram and system memory (gart) */ - if (adev->gmc.pdb0_bo) { - WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32, - (u32)(adev->gmc.fb_start >> 12)); - WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32, - (u32)(adev->gmc.fb_start >> 44)); - - WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32, - (u32)(adev->gmc.gart_end >> 12)); - WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32, - (u32)(adev->gmc.gart_end >> 44)); - - } else { - WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32, - (u32)(adev->gmc.gart_start >> 12)); - WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32, - (u32)(adev->gmc.gart_start >> 44)); - - WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32, - (u32)(adev->gmc.gart_end >> 12)); - WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32, - (u32)(adev->gmc.gart_end >> 44)); + inst_mask = adev->aid_mask; + for_each_inst(i, inst_mask) { + if (adev->gmc.pdb0_bo) { + WREG32_SOC15(MMHUB, i, + regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32, + (u32)(adev->gmc.fb_start >> 12)); + WREG32_SOC15(MMHUB, i, + regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32, + (u32)(adev->gmc.fb_start >> 44)); + + WREG32_SOC15(MMHUB, i, + regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32, + (u32)(adev->gmc.gart_end >> 12)); + WREG32_SOC15(MMHUB, i, + regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32, + (u32)(adev->gmc.gart_end >> 44)); + + } else { + WREG32_SOC15(MMHUB, i, + regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32, + (u32)(adev->gmc.gart_start >> 12)); + WREG32_SOC15(MMHUB, i, + regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32, + (u32)(adev->gmc.gart_start >> 44)); + + WREG32_SOC15(MMHUB, i, + regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32, + (u32)(adev->gmc.gart_end >> 12)); + WREG32_SOC15(MMHUB, i, + regVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32, + (u32)(adev->gmc.gart_end >> 44)); + } } } static void mmhub_v1_8_init_system_aperture_regs(struct amdgpu_device *adev) { + uint32_t tmp, inst_mask; uint64_t value; - uint32_t tmp; + int i; - /* Program the AGP BAR */ - WREG32_SOC15(MMHUB, 0, regMC_VM_AGP_BASE, 0); - WREG32_SOC15(MMHUB, 0, regMC_VM_AGP_BOT, adev->gmc.agp_start >> 24); - WREG32_SOC15(MMHUB, 0, regMC_VM_AGP_TOP, adev->gmc.agp_end >> 24); + inst_mask = adev->aid_mask; + for_each_inst(i, inst_mask) { + /* Program the AGP BAR */ + WREG32_SOC15(MMHUB, i, regMC_VM_AGP_BASE, 0); + WREG32_SOC15(MMHUB, i, regMC_VM_AGP_BOT, + adev->gmc.agp_start >> 24); + WREG32_SOC15(MMHUB, i, regMC_VM_AGP_TOP, + adev->gmc.agp_end >> 24); - /* Program the system aperture low logical page number. */ - WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_LOW_ADDR, - min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18); + if (amdgpu_sriov_vf(adev)) + return; - WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, - max(adev->gmc.fb_end, adev->gmc.agp_end) >> 18); + /* Program the system aperture low logical page number. */ + WREG32_SOC15(MMHUB, i, regMC_VM_SYSTEM_APERTURE_LOW_ADDR, + min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18); - /* In the case squeezing vram into GART aperture, we don't use - * FB aperture and AGP aperture. Disable them. - */ - if (adev->gmc.pdb0_bo) { - WREG32_SOC15(MMHUB, 0, regMC_VM_AGP_BOT, 0xFFFFFF); - WREG32_SOC15(MMHUB, 0, regMC_VM_AGP_TOP, 0); - WREG32_SOC15(MMHUB, 0, regMC_VM_FB_LOCATION_TOP, 0); - WREG32_SOC15(MMHUB, 0, regMC_VM_FB_LOCATION_BASE, 0x00FFFFFF); - WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_LOW_ADDR, 0x3FFFFFFF); - WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, 0); - } - if (amdgpu_sriov_vf(adev)) - return; + WREG32_SOC15(MMHUB, i, regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, + max(adev->gmc.fb_end, adev->gmc.agp_end) >> 18); - /* Set default page address. */ - value = amdgpu_gmc_vram_mc2pa(adev, adev->mem_scratch.gpu_addr); - WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB, - (u32)(value >> 12)); - WREG32_SOC15(MMHUB, 0, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, - (u32)(value >> 44)); - - /* Program "protection fault". */ - WREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_LO32, - (u32)(adev->dummy_page_addr >> 12)); - WREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_HI32, - (u32)((u64)adev->dummy_page_addr >> 44)); - - tmp = RREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_CNTL2); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL2, - ACTIVE_PAGE_MIGRATION_PTE_READ_RETRY, 1); - WREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_CNTL2, tmp); + /* In the case squeezing vram into GART aperture, we don't use + * FB aperture and AGP aperture. Disable them. + */ + if (adev->gmc.pdb0_bo) { + WREG32_SOC15(MMHUB, i, regMC_VM_AGP_BOT, 0xFFFFFF); + WREG32_SOC15(MMHUB, i, regMC_VM_AGP_TOP, 0); + WREG32_SOC15(MMHUB, i, regMC_VM_FB_LOCATION_TOP, 0); + WREG32_SOC15(MMHUB, i, regMC_VM_FB_LOCATION_BASE, + 0x00FFFFFF); + WREG32_SOC15(MMHUB, i, + regMC_VM_SYSTEM_APERTURE_LOW_ADDR, + 0x3FFFFFFF); + WREG32_SOC15(MMHUB, i, + regMC_VM_SYSTEM_APERTURE_HIGH_ADDR, 0); + } + + /* Set default page address. */ + value = amdgpu_gmc_vram_mc2pa(adev, adev->mem_scratch.gpu_addr); + WREG32_SOC15(MMHUB, i, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB, + (u32)(value >> 12)); + WREG32_SOC15(MMHUB, i, regMC_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB, + (u32)(value >> 44)); + + /* Program "protection fault". */ + WREG32_SOC15(MMHUB, i, + regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_LO32, + (u32)(adev->dummy_page_addr >> 12)); + WREG32_SOC15(MMHUB, i, + regVM_L2_PROTECTION_FAULT_DEFAULT_ADDR_HI32, + (u32)((u64)adev->dummy_page_addr >> 44)); + + tmp = RREG32_SOC15(MMHUB, i, regVM_L2_PROTECTION_FAULT_CNTL2); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL2, + ACTIVE_PAGE_MIGRATION_PTE_READ_RETRY, 1); + WREG32_SOC15(MMHUB, i, regVM_L2_PROTECTION_FAULT_CNTL2, tmp); + } } static void mmhub_v1_8_init_tlb_regs(struct amdgpu_device *adev) { - uint32_t tmp; + uint32_t tmp, inst_mask; + int i; /* Setup TLB control */ - tmp = RREG32_SOC15(MMHUB, 0, regMC_VM_MX_L1_TLB_CNTL); - - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 1); - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, SYSTEM_ACCESS_MODE, 3); - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, - ENABLE_ADVANCED_DRIVER_MODEL, 1); - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, - SYSTEM_APERTURE_UNMAPPED_ACCESS, 0); - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, - MTYPE, MTYPE_UC);/* XXX for emulation. */ - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ATC_EN, 1); - - WREG32_SOC15(MMHUB, 0, regMC_VM_MX_L1_TLB_CNTL, tmp); + inst_mask = adev->aid_mask; + for_each_inst(i, inst_mask) { + tmp = RREG32_SOC15(MMHUB, i, regMC_VM_MX_L1_TLB_CNTL); + + tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, + 1); + tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, + SYSTEM_ACCESS_MODE, 3); + tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, + ENABLE_ADVANCED_DRIVER_MODEL, 1); + tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, + SYSTEM_APERTURE_UNMAPPED_ACCESS, 0); + tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, + MTYPE, MTYPE_UC);/* XXX for emulation. */ + tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ATC_EN, 1); + + WREG32_SOC15(MMHUB, i, regMC_VM_MX_L1_TLB_CNTL, tmp); + } } static void mmhub_v1_8_init_cache_regs(struct amdgpu_device *adev) { - uint32_t tmp; + uint32_t tmp, inst_mask; + int i; if (amdgpu_sriov_vf(adev)) return; /* Setup L2 cache */ - tmp = RREG32_SOC15(MMHUB, 0, regVM_L2_CNTL); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 1); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_FRAGMENT_PROCESSING, 1); - /* XXX for emulation, Refer to closed source code.*/ - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, L2_PDE0_CACHE_TAG_GENERATION_MODE, - 0); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, PDE_FAULT_CLASSIFICATION, 0); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, CONTEXT1_IDENTITY_ACCESS_MODE, 1); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, IDENTITY_MODE_FRAGMENT_SIZE, 0); - WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL, tmp); - - tmp = RREG32_SOC15(MMHUB, 0, regVM_L2_CNTL2); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS, 1); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1); - WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL2, tmp); - - tmp = regVM_L2_CNTL3_DEFAULT; - if (adev->gmc.translate_further) { - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 12); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, - L2_CACHE_BIGK_FRAGMENT_SIZE, 9); - } else { - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 9); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, - L2_CACHE_BIGK_FRAGMENT_SIZE, 6); - } - WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL3, tmp); - - tmp = regVM_L2_CNTL4_DEFAULT; - if (adev->gmc.xgmi.connected_to_cpu) { - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, - VMC_TAP_PDE_REQUEST_PHYSICAL, 1); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, - VMC_TAP_PTE_REQUEST_PHYSICAL, 1); - } else { - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, - VMC_TAP_PDE_REQUEST_PHYSICAL, 0); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, - VMC_TAP_PTE_REQUEST_PHYSICAL, 0); + inst_mask = adev->aid_mask; + for_each_inst(i, inst_mask) { + tmp = RREG32_SOC15(MMHUB, i, regVM_L2_CNTL); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 1); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, + ENABLE_L2_FRAGMENT_PROCESSING, 1); + /* XXX for emulation, Refer to closed source code.*/ + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, + L2_PDE0_CACHE_TAG_GENERATION_MODE, 0); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, PDE_FAULT_CLASSIFICATION, + 0); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, + CONTEXT1_IDENTITY_ACCESS_MODE, 1); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, + IDENTITY_MODE_FRAGMENT_SIZE, 0); + WREG32_SOC15(MMHUB, i, regVM_L2_CNTL, tmp); + + tmp = RREG32_SOC15(MMHUB, i, regVM_L2_CNTL2); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS, + 1); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1); + WREG32_SOC15(MMHUB, i, regVM_L2_CNTL2, tmp); + + tmp = regVM_L2_CNTL3_DEFAULT; + if (adev->gmc.translate_further) { + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 12); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, + L2_CACHE_BIGK_FRAGMENT_SIZE, 9); + } else { + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 9); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, + L2_CACHE_BIGK_FRAGMENT_SIZE, 6); + } + WREG32_SOC15(MMHUB, i, regVM_L2_CNTL3, tmp); + + tmp = regVM_L2_CNTL4_DEFAULT; + /* For AMD APP APUs setup WC memory */ + if (adev->gmc.xgmi.connected_to_cpu || adev->gmc.is_app_apu) { + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, + VMC_TAP_PDE_REQUEST_PHYSICAL, 1); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, + VMC_TAP_PTE_REQUEST_PHYSICAL, 1); + } else { + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, + VMC_TAP_PDE_REQUEST_PHYSICAL, 0); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL4, + VMC_TAP_PTE_REQUEST_PHYSICAL, 0); + } + WREG32_SOC15(MMHUB, i, regVM_L2_CNTL4, tmp); } - WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL4, tmp); } static void mmhub_v1_8_enable_system_domain(struct amdgpu_device *adev) { - uint32_t tmp; - - tmp = RREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_CNTL); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, ENABLE_CONTEXT, 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_DEPTH, - adev->gmc.vmid0_page_table_depth); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_BLOCK_SIZE, - adev->gmc.vmid0_page_table_block_size); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, - RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0); - WREG32_SOC15(MMHUB, 0, regVM_CONTEXT0_CNTL, tmp); + uint32_t tmp, inst_mask; + int i; + + inst_mask = adev->aid_mask; + for_each_inst(i, inst_mask) { + tmp = RREG32_SOC15(MMHUB, i, regVM_CONTEXT0_CNTL); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, ENABLE_CONTEXT, 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, PAGE_TABLE_DEPTH, + adev->gmc.vmid0_page_table_depth); + tmp = REG_SET_FIELD(tmp, + VM_CONTEXT0_CNTL, PAGE_TABLE_BLOCK_SIZE, + adev->gmc.vmid0_page_table_block_size); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT0_CNTL, + RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0); + WREG32_SOC15(MMHUB, i, regVM_CONTEXT0_CNTL, tmp); + } } static void mmhub_v1_8_disable_identity_aperture(struct amdgpu_device *adev) { + u32 inst_mask; + int i; + if (amdgpu_sriov_vf(adev)) return; - WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_LO32, 0xFFFFFFFF); - WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_HI32, 0x0000000F); - - WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_LO32, 0); - WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_HI32, 0); - - WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_LO32, 0); - WREG32_SOC15(MMHUB, 0, regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_HI32, 0); + inst_mask = adev->aid_mask; + for_each_inst(i, inst_mask) { + WREG32_SOC15(MMHUB, i, + regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_LO32, + 0XFFFFFFFF); + WREG32_SOC15(MMHUB, i, + regVM_L2_CONTEXT1_IDENTITY_APERTURE_LOW_ADDR_HI32, + 0x0000000F); + + WREG32_SOC15(MMHUB, i, + regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_LO32, + 0); + WREG32_SOC15(MMHUB, i, + regVM_L2_CONTEXT1_IDENTITY_APERTURE_HIGH_ADDR_HI32, + 0); + + WREG32_SOC15(MMHUB, i, + regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_LO32, 0); + WREG32_SOC15(MMHUB, i, + regVM_L2_CONTEXT_IDENTITY_PHYSICAL_OFFSET_HI32, 0); + } } static void mmhub_v1_8_setup_vmid_config(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; - unsigned num_level, block_size; - uint32_t tmp; - int i; + struct amdgpu_vmhub *hub; + unsigned int num_level, block_size; + uint32_t tmp, inst_mask; + int i, j; num_level = adev->vm_manager.num_level; block_size = adev->vm_manager.block_size; @@ -265,60 +339,75 @@ static void mmhub_v1_8_setup_vmid_config(struct amdgpu_device *adev) else block_size -= 9; - for (i = 0; i <= 14; i++) { - tmp = RREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_CNTL, i); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, ENABLE_CONTEXT, 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, PAGE_TABLE_DEPTH, - num_level); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, - 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - VALID_PROTECTION_FAULT_ENABLE_DEFAULT, 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - READ_PROTECTION_FAULT_ENABLE_DEFAULT, 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, 1); - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - PAGE_TABLE_BLOCK_SIZE, - block_size); - /* On Aldebaran, XNACK can be enabled in the SQ per-process. - * Retry faults need to be enabled for that to work. - */ - tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, - 1); - WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_CNTL, - i * hub->ctx_distance, tmp); - WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32, - i * hub->ctx_addr_distance, 0); - WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32, - i * hub->ctx_addr_distance, 0); - WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_PAGE_TABLE_END_ADDR_LO32, - i * hub->ctx_addr_distance, - lower_32_bits(adev->vm_manager.max_pfn - 1)); - WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT1_PAGE_TABLE_END_ADDR_HI32, - i * hub->ctx_addr_distance, - upper_32_bits(adev->vm_manager.max_pfn - 1)); + inst_mask = adev->aid_mask; + for_each_inst(j, inst_mask) { + hub = &adev->vmhub[AMDGPU_MMHUB0(j)]; + for (i = 0; i <= 14; i++) { + tmp = RREG32_SOC15_OFFSET(MMHUB, j, regVM_CONTEXT1_CNTL, + i); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + ENABLE_CONTEXT, 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + PAGE_TABLE_DEPTH, num_level); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + VALID_PROTECTION_FAULT_ENABLE_DEFAULT, 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + READ_PROTECTION_FAULT_ENABLE_DEFAULT, 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, 1); + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + PAGE_TABLE_BLOCK_SIZE, + block_size); + /* On 9.4.3, XNACK can be enabled in the SQ + * per-process. Retry faults need to be enabled for + * that to work. + */ + tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, + RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 1); + WREG32_SOC15_OFFSET(MMHUB, j, regVM_CONTEXT1_CNTL, + i * hub->ctx_distance, tmp); + WREG32_SOC15_OFFSET(MMHUB, j, + regVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32, + i * hub->ctx_addr_distance, 0); + WREG32_SOC15_OFFSET(MMHUB, j, + regVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32, + i * hub->ctx_addr_distance, 0); + WREG32_SOC15_OFFSET(MMHUB, j, + regVM_CONTEXT1_PAGE_TABLE_END_ADDR_LO32, + i * hub->ctx_addr_distance, + lower_32_bits(adev->vm_manager.max_pfn - 1)); + WREG32_SOC15_OFFSET(MMHUB, j, + regVM_CONTEXT1_PAGE_TABLE_END_ADDR_HI32, + i * hub->ctx_addr_distance, + upper_32_bits(adev->vm_manager.max_pfn - 1)); + } } } static void mmhub_v1_8_program_invalidation(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; - unsigned i; - - for (i = 0; i < 18; ++i) { - WREG32_SOC15_OFFSET(MMHUB, 0, regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32, - i * hub->eng_addr_distance, 0xffffffff); - WREG32_SOC15_OFFSET(MMHUB, 0, regVM_INVALIDATE_ENG0_ADDR_RANGE_HI32, - i * hub->eng_addr_distance, 0x1f); + struct amdgpu_vmhub *hub; + u32 i, j, inst_mask; + + inst_mask = adev->aid_mask; + for_each_inst(j, inst_mask) { + hub = &adev->vmhub[AMDGPU_MMHUB0(j)]; + for (i = 0; i < 18; ++i) { + WREG32_SOC15_OFFSET(MMHUB, j, + regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32, + i * hub->eng_addr_distance, 0xffffffff); + WREG32_SOC15_OFFSET(MMHUB, j, + regVM_INVALIDATE_ENG0_ADDR_RANGE_HI32, + i * hub->eng_addr_distance, 0x1f); + } } } @@ -352,28 +441,34 @@ static int mmhub_v1_8_gart_enable(struct amdgpu_device *adev) static void mmhub_v1_8_gart_disable(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub; u32 tmp; - u32 i; + u32 i, j, inst_mask; /* Disable all tables */ - for (i = 0; i < 16; i++) - WREG32_SOC15_OFFSET(MMHUB, 0, regVM_CONTEXT0_CNTL, - i * hub->ctx_distance, 0); - - /* Setup TLB control */ - tmp = RREG32_SOC15(MMHUB, 0, regMC_VM_MX_L1_TLB_CNTL); - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, 0); - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, - ENABLE_ADVANCED_DRIVER_MODEL, 0); - WREG32_SOC15(MMHUB, 0, regMC_VM_MX_L1_TLB_CNTL, tmp); - - if (!amdgpu_sriov_vf(adev)) { - /* Setup L2 cache */ - tmp = RREG32_SOC15(MMHUB, 0, regVM_L2_CNTL); - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, 0); - WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL, tmp); - WREG32_SOC15(MMHUB, 0, regVM_L2_CNTL3, 0); + inst_mask = adev->aid_mask; + for_each_inst(j, inst_mask) { + hub = &adev->vmhub[AMDGPU_MMHUB0(j)]; + for (i = 0; i < 16; i++) + WREG32_SOC15_OFFSET(MMHUB, j, regVM_CONTEXT0_CNTL, + i * hub->ctx_distance, 0); + + /* Setup TLB control */ + tmp = RREG32_SOC15(MMHUB, j, regMC_VM_MX_L1_TLB_CNTL); + tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ENABLE_L1_TLB, + 0); + tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, + ENABLE_ADVANCED_DRIVER_MODEL, 0); + WREG32_SOC15(MMHUB, j, regMC_VM_MX_L1_TLB_CNTL, tmp); + + if (!amdgpu_sriov_vf(adev)) { + /* Setup L2 cache */ + tmp = RREG32_SOC15(MMHUB, j, regVM_L2_CNTL); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL, ENABLE_L2_CACHE, + 0); + WREG32_SOC15(MMHUB, j, regVM_L2_CNTL, tmp); + WREG32_SOC15(MMHUB, j, regVM_L2_CNTL3, 0); + } } } @@ -385,73 +480,83 @@ static void mmhub_v1_8_gart_disable(struct amdgpu_device *adev) */ static void mmhub_v1_8_set_fault_enable_default(struct amdgpu_device *adev, bool value) { - u32 tmp; + u32 tmp, inst_mask; + int i; if (amdgpu_sriov_vf(adev)) return; - tmp = RREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_CNTL); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - PDE1_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - PDE2_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - TRANSLATE_FURTHER_PROTECTION_FAULT_ENABLE_DEFAULT, - value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - NACK_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - VALID_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - READ_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value); - if (!value) { + inst_mask = adev->aid_mask; + for_each_inst(i, inst_mask) { + tmp = RREG32_SOC15(MMHUB, i, regVM_L2_PROTECTION_FAULT_CNTL); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + PDE1_PROTECTION_FAULT_ENABLE_DEFAULT, value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + PDE2_PROTECTION_FAULT_ENABLE_DEFAULT, value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + TRANSLATE_FURTHER_PROTECTION_FAULT_ENABLE_DEFAULT, + value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + NACK_PROTECTION_FAULT_ENABLE_DEFAULT, value); tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_NO_RETRY_FAULT, 1); + DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, value); tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_RETRY_FAULT, 1); + VALID_PROTECTION_FAULT_ENABLE_DEFAULT, value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + READ_PROTECTION_FAULT_ENABLE_DEFAULT, value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value); + if (!value) { + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_NO_RETRY_FAULT, 1); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_RETRY_FAULT, 1); + } + + WREG32_SOC15(MMHUB, i, regVM_L2_PROTECTION_FAULT_CNTL, tmp); } - - WREG32_SOC15(MMHUB, 0, regVM_L2_PROTECTION_FAULT_CNTL, tmp); } static void mmhub_v1_8_init(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; - - hub->ctx0_ptb_addr_lo32 = - SOC15_REG_OFFSET(MMHUB, 0, - regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32); - hub->ctx0_ptb_addr_hi32 = - SOC15_REG_OFFSET(MMHUB, 0, - regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32); - hub->vm_inv_eng0_req = - SOC15_REG_OFFSET(MMHUB, 0, regVM_INVALIDATE_ENG0_REQ); - hub->vm_inv_eng0_ack = - SOC15_REG_OFFSET(MMHUB, 0, regVM_INVALIDATE_ENG0_ACK); - hub->vm_context0_cntl = - SOC15_REG_OFFSET(MMHUB, 0, regVM_CONTEXT0_CNTL); - hub->vm_l2_pro_fault_status = - SOC15_REG_OFFSET(MMHUB, 0, regVM_L2_PROTECTION_FAULT_STATUS); - hub->vm_l2_pro_fault_cntl = - SOC15_REG_OFFSET(MMHUB, 0, regVM_L2_PROTECTION_FAULT_CNTL); - - hub->ctx_distance = regVM_CONTEXT1_CNTL - regVM_CONTEXT0_CNTL; - hub->ctx_addr_distance = regVM_CONTEXT1_PAGE_TABLE_BASE_ADDR_LO32 - - regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32; - hub->eng_distance = regVM_INVALIDATE_ENG1_REQ - regVM_INVALIDATE_ENG0_REQ; - hub->eng_addr_distance = regVM_INVALIDATE_ENG1_ADDR_RANGE_LO32 - - regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32; + struct amdgpu_vmhub *hub; + u32 inst_mask; + int i; + inst_mask = adev->aid_mask; + for_each_inst(i, inst_mask) { + hub = &adev->vmhub[AMDGPU_MMHUB0(i)]; + + hub->ctx0_ptb_addr_lo32 = SOC15_REG_OFFSET(MMHUB, i, + regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32); + hub->ctx0_ptb_addr_hi32 = SOC15_REG_OFFSET(MMHUB, i, + regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32); + hub->vm_inv_eng0_req = + SOC15_REG_OFFSET(MMHUB, i, regVM_INVALIDATE_ENG0_REQ); + hub->vm_inv_eng0_ack = + SOC15_REG_OFFSET(MMHUB, i, regVM_INVALIDATE_ENG0_ACK); + hub->vm_context0_cntl = + SOC15_REG_OFFSET(MMHUB, i, regVM_CONTEXT0_CNTL); + hub->vm_l2_pro_fault_status = SOC15_REG_OFFSET(MMHUB, i, + regVM_L2_PROTECTION_FAULT_STATUS); + hub->vm_l2_pro_fault_cntl = SOC15_REG_OFFSET(MMHUB, i, + regVM_L2_PROTECTION_FAULT_CNTL); + + hub->ctx_distance = regVM_CONTEXT1_CNTL - regVM_CONTEXT0_CNTL; + hub->ctx_addr_distance = + regVM_CONTEXT1_PAGE_TABLE_BASE_ADDR_LO32 - + regVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32; + hub->eng_distance = regVM_INVALIDATE_ENG1_REQ - + regVM_INVALIDATE_ENG0_REQ; + hub->eng_addr_distance = regVM_INVALIDATE_ENG1_ADDR_RANGE_LO32 - + regVM_INVALIDATE_ENG0_ADDR_RANGE_LO32; + } } static int mmhub_v1_8_set_clockgating(struct amdgpu_device *adev, @@ -475,3 +580,277 @@ const struct amdgpu_mmhub_funcs mmhub_v1_8_funcs = { .set_clockgating = mmhub_v1_8_set_clockgating, .get_clockgating = mmhub_v1_8_get_clockgating, }; + +static const struct amdgpu_ras_err_status_reg_entry mmhub_v1_8_ce_reg_list[] = { + {AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA0_CE_ERR_STATUS_LO, regMMEA0_CE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA0"}, + {AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA1_CE_ERR_STATUS_LO, regMMEA1_CE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA1"}, + {AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA2_CE_ERR_STATUS_LO, regMMEA2_CE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA2"}, + {AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA3_CE_ERR_STATUS_LO, regMMEA3_CE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA3"}, + {AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA4_CE_ERR_STATUS_LO, regMMEA4_CE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA4"}, + {AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMM_CANE_CE_ERR_STATUS_LO, regMM_CANE_CE_ERR_STATUS_HI), + 1, 0, "MM_CANE"}, +}; + +static const struct amdgpu_ras_err_status_reg_entry mmhub_v1_8_ue_reg_list[] = { + {AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA0_UE_ERR_STATUS_LO, regMMEA0_UE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA0"}, + {AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA1_UE_ERR_STATUS_LO, regMMEA1_UE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA1"}, + {AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA2_UE_ERR_STATUS_LO, regMMEA2_UE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA2"}, + {AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA3_UE_ERR_STATUS_LO, regMMEA3_UE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA3"}, + {AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMMEA4_UE_ERR_STATUS_LO, regMMEA4_UE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "MMEA4"}, + {AMDGPU_RAS_REG_ENTRY(MMHUB, 0, regMM_CANE_UE_ERR_STATUS_LO, regMM_CANE_UE_ERR_STATUS_HI), + 1, 0, "MM_CANE"}, +}; + +static const struct amdgpu_ras_memory_id_entry mmhub_v1_8_ras_memory_list[] = { + {AMDGPU_MMHUB_WGMI_PAGEMEM, "MMEA_WGMI_PAGEMEM"}, + {AMDGPU_MMHUB_RGMI_PAGEMEM, "MMEA_RGMI_PAGEMEM"}, + {AMDGPU_MMHUB_WDRAM_PAGEMEM, "MMEA_WDRAM_PAGEMEM"}, + {AMDGPU_MMHUB_RDRAM_PAGEMEM, "MMEA_RDRAM_PAGEMEM"}, + {AMDGPU_MMHUB_WIO_CMDMEM, "MMEA_WIO_CMDMEM"}, + {AMDGPU_MMHUB_RIO_CMDMEM, "MMEA_RIO_CMDMEM"}, + {AMDGPU_MMHUB_WGMI_CMDMEM, "MMEA_WGMI_CMDMEM"}, + {AMDGPU_MMHUB_RGMI_CMDMEM, "MMEA_RGMI_CMDMEM"}, + {AMDGPU_MMHUB_WDRAM_CMDMEM, "MMEA_WDRAM_CMDMEM"}, + {AMDGPU_MMHUB_RDRAM_CMDMEM, "MMEA_RDRAM_CMDMEM"}, + {AMDGPU_MMHUB_MAM_DMEM0, "MMEA_MAM_DMEM0"}, + {AMDGPU_MMHUB_MAM_DMEM1, "MMEA_MAM_DMEM1"}, + {AMDGPU_MMHUB_MAM_DMEM2, "MMEA_MAM_DMEM2"}, + {AMDGPU_MMHUB_MAM_DMEM3, "MMEA_MAM_DMEM3"}, + {AMDGPU_MMHUB_WRET_TAGMEM, "MMEA_WRET_TAGMEM"}, + {AMDGPU_MMHUB_RRET_TAGMEM, "MMEA_RRET_TAGMEM"}, + {AMDGPU_MMHUB_WIO_DATAMEM, "MMEA_WIO_DATAMEM"}, + {AMDGPU_MMHUB_WGMI_DATAMEM, "MMEA_WGMI_DATAMEM"}, + {AMDGPU_MMHUB_WDRAM_DATAMEM, "MMEA_WDRAM_DATAMEM"}, +}; + +static void mmhub_v1_8_inst_query_ras_error_count(struct amdgpu_device *adev, + uint32_t mmhub_inst, + void *ras_err_status) +{ + struct ras_err_data *err_data = (struct ras_err_data *)ras_err_status; + + amdgpu_ras_inst_query_ras_error_count(adev, + mmhub_v1_8_ce_reg_list, + ARRAY_SIZE(mmhub_v1_8_ce_reg_list), + mmhub_v1_8_ras_memory_list, + ARRAY_SIZE(mmhub_v1_8_ras_memory_list), + mmhub_inst, + AMDGPU_RAS_ERROR__SINGLE_CORRECTABLE, + &err_data->ce_count); + amdgpu_ras_inst_query_ras_error_count(adev, + mmhub_v1_8_ue_reg_list, + ARRAY_SIZE(mmhub_v1_8_ue_reg_list), + mmhub_v1_8_ras_memory_list, + ARRAY_SIZE(mmhub_v1_8_ras_memory_list), + mmhub_inst, + AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE, + &err_data->ue_count); +} + +static void mmhub_v1_8_query_ras_error_count(struct amdgpu_device *adev, + void *ras_err_status) +{ + uint32_t inst_mask; + uint32_t i; + + if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__MMHUB)) { + dev_warn(adev->dev, "MMHUB RAS is not supported\n"); + return; + } + + inst_mask = adev->aid_mask; + for_each_inst(i, inst_mask) + mmhub_v1_8_inst_query_ras_error_count(adev, i, ras_err_status); +} + +static void mmhub_v1_8_inst_reset_ras_error_count(struct amdgpu_device *adev, + uint32_t mmhub_inst) +{ + amdgpu_ras_inst_reset_ras_error_count(adev, + mmhub_v1_8_ce_reg_list, + ARRAY_SIZE(mmhub_v1_8_ce_reg_list), + mmhub_inst); + amdgpu_ras_inst_reset_ras_error_count(adev, + mmhub_v1_8_ue_reg_list, + ARRAY_SIZE(mmhub_v1_8_ue_reg_list), + mmhub_inst); +} + +static void mmhub_v1_8_reset_ras_error_count(struct amdgpu_device *adev) +{ + uint32_t inst_mask; + uint32_t i; + + if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__MMHUB)) { + dev_warn(adev->dev, "MMHUB RAS is not supported\n"); + return; + } + + inst_mask = adev->aid_mask; + for_each_inst(i, inst_mask) + mmhub_v1_8_inst_reset_ras_error_count(adev, i); +} + +static const u32 mmhub_v1_8_mmea_err_status_reg[] __maybe_unused = { + regMMEA0_ERR_STATUS, + regMMEA1_ERR_STATUS, + regMMEA2_ERR_STATUS, + regMMEA3_ERR_STATUS, + regMMEA4_ERR_STATUS, +}; + +static void mmhub_v1_8_inst_query_ras_err_status(struct amdgpu_device *adev, + uint32_t mmhub_inst) +{ + uint32_t reg_value; + uint32_t mmea_err_status_addr_dist; + uint32_t i; + + /* query mmea ras err status */ + mmea_err_status_addr_dist = regMMEA1_ERR_STATUS - regMMEA0_ERR_STATUS; + for (i = 0; i < ARRAY_SIZE(mmhub_v1_8_mmea_err_status_reg); i++) { + reg_value = RREG32_SOC15_OFFSET(MMHUB, mmhub_inst, + regMMEA0_ERR_STATUS, + i * mmea_err_status_addr_dist); + if (REG_GET_FIELD(reg_value, MMEA0_ERR_STATUS, SDP_RDRSP_STATUS) || + REG_GET_FIELD(reg_value, MMEA0_ERR_STATUS, SDP_WRRSP_STATUS) || + REG_GET_FIELD(reg_value, MMEA0_ERR_STATUS, SDP_RDRSP_DATAPARITY_ERROR)) { + dev_warn(adev->dev, + "Detected MMEA%d err in MMHUB%d, status: 0x%x\n", + i, mmhub_inst, reg_value); + } + } + + /* query mm_cane ras err status */ + reg_value = RREG32_SOC15(MMHUB, mmhub_inst, regMM_CANE_ERR_STATUS); + if (REG_GET_FIELD(reg_value, MM_CANE_ERR_STATUS, SDPM_RDRSP_STATUS) || + REG_GET_FIELD(reg_value, MM_CANE_ERR_STATUS, SDPM_WRRSP_STATUS) || + REG_GET_FIELD(reg_value, MM_CANE_ERR_STATUS, SDPM_RDRSP_DATAPARITY_ERROR)) { + dev_warn(adev->dev, + "Detected MM CANE err in MMHUB%d, status: 0x%x\n", + mmhub_inst, reg_value); + } +} + +static void mmhub_v1_8_query_ras_error_status(struct amdgpu_device *adev) +{ + uint32_t inst_mask; + uint32_t i; + + if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__MMHUB)) { + dev_warn(adev->dev, "MMHUB RAS is not supported\n"); + return; + } + + inst_mask = adev->aid_mask; + for_each_inst(i, inst_mask) + mmhub_v1_8_inst_query_ras_err_status(adev, i); +} + +static void mmhub_v1_8_inst_reset_ras_err_status(struct amdgpu_device *adev, + uint32_t mmhub_inst) +{ + uint32_t mmea_cgtt_clk_cntl_addr_dist; + uint32_t mmea_err_status_addr_dist; + uint32_t reg_value; + uint32_t i; + + /* reset mmea ras err status */ + mmea_cgtt_clk_cntl_addr_dist = regMMEA1_CGTT_CLK_CTRL - regMMEA0_CGTT_CLK_CTRL; + mmea_err_status_addr_dist = regMMEA1_ERR_STATUS - regMMEA0_ERR_STATUS; + for (i = 0; i < ARRAY_SIZE(mmhub_v1_8_mmea_err_status_reg); i++) { + /* force clk branch on for response path + * set MMEA0_CGTT_CLK_CTRL.SOFT_OVERRIDE_RETURN = 1 + */ + reg_value = RREG32_SOC15_OFFSET(MMHUB, mmhub_inst, + regMMEA0_CGTT_CLK_CTRL, + i * mmea_cgtt_clk_cntl_addr_dist); + reg_value = REG_SET_FIELD(reg_value, MMEA0_CGTT_CLK_CTRL, + SOFT_OVERRIDE_RETURN, 1); + WREG32_SOC15_OFFSET(MMHUB, mmhub_inst, + regMMEA0_CGTT_CLK_CTRL, + i * mmea_cgtt_clk_cntl_addr_dist, + reg_value); + + /* set MMEA0_ERR_STATUS.CLEAR_ERROR_STATUS = 1 */ + reg_value = RREG32_SOC15_OFFSET(MMHUB, mmhub_inst, + regMMEA0_ERR_STATUS, + i * mmea_err_status_addr_dist); + reg_value = REG_SET_FIELD(reg_value, MMEA0_ERR_STATUS, + CLEAR_ERROR_STATUS, 1); + WREG32_SOC15_OFFSET(MMHUB, mmhub_inst, + regMMEA0_ERR_STATUS, + i * mmea_err_status_addr_dist, + reg_value); + + /* set MMEA0_CGTT_CLK_CTRL.SOFT_OVERRIDE_RETURN = 0 */ + reg_value = RREG32_SOC15_OFFSET(MMHUB, mmhub_inst, + regMMEA0_CGTT_CLK_CTRL, + i * mmea_cgtt_clk_cntl_addr_dist); + reg_value = REG_SET_FIELD(reg_value, MMEA0_CGTT_CLK_CTRL, + SOFT_OVERRIDE_RETURN, 0); + WREG32_SOC15_OFFSET(MMHUB, mmhub_inst, + regMMEA0_CGTT_CLK_CTRL, + i * mmea_cgtt_clk_cntl_addr_dist, + reg_value); + } + + /* reset mm_cane ras err status + * force clk branch on for response path + * set MM_CANE_ICG_CTRL.SOFT_OVERRIDE_ATRET = 1 + */ + reg_value = RREG32_SOC15(MMHUB, mmhub_inst, regMM_CANE_ICG_CTRL); + reg_value = REG_SET_FIELD(reg_value, MM_CANE_ICG_CTRL, + SOFT_OVERRIDE_ATRET, 1); + WREG32_SOC15(MMHUB, mmhub_inst, regMM_CANE_ICG_CTRL, reg_value); + + /* set MM_CANE_ERR_STATUS.CLEAR_ERROR_STATUS = 1 */ + reg_value = RREG32_SOC15(MMHUB, mmhub_inst, regMM_CANE_ERR_STATUS); + reg_value = REG_SET_FIELD(reg_value, MM_CANE_ERR_STATUS, + CLEAR_ERROR_STATUS, 1); + WREG32_SOC15(MMHUB, mmhub_inst, regMM_CANE_ERR_STATUS, reg_value); + + /* set MM_CANE_ICG_CTRL.SOFT_OVERRIDE_ATRET = 0 */ + reg_value = RREG32_SOC15(MMHUB, mmhub_inst, regMM_CANE_ICG_CTRL); + reg_value = REG_SET_FIELD(reg_value, MM_CANE_ICG_CTRL, + SOFT_OVERRIDE_ATRET, 0); + WREG32_SOC15(MMHUB, mmhub_inst, regMM_CANE_ICG_CTRL, reg_value); +} + +static void mmhub_v1_8_reset_ras_error_status(struct amdgpu_device *adev) +{ + uint32_t inst_mask; + uint32_t i; + + if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__MMHUB)) { + dev_warn(adev->dev, "MMHUB RAS is not supported\n"); + return; + } + + inst_mask = adev->aid_mask; + for_each_inst(i, inst_mask) + mmhub_v1_8_inst_reset_ras_err_status(adev, i); +} + +static const struct amdgpu_ras_block_hw_ops mmhub_v1_8_ras_hw_ops = { + .query_ras_error_count = mmhub_v1_8_query_ras_error_count, + .reset_ras_error_count = mmhub_v1_8_reset_ras_error_count, + .query_ras_error_status = mmhub_v1_8_query_ras_error_status, + .reset_ras_error_status = mmhub_v1_8_reset_ras_error_status, +}; + +struct amdgpu_mmhub_ras mmhub_v1_8_ras = { + .ras_block = { + .hw_ops = &mmhub_v1_8_ras_hw_ops, + }, +}; diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.h b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.h index 0bb36200e4e5b10a4cf06302141265c7c16c987f..126f0075ac5083e1010dba5bcc06dae0728ee208 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.h +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.h @@ -24,5 +24,6 @@ #define __MMHUB_V1_8_H__ extern const struct amdgpu_mmhub_funcs mmhub_v1_8_funcs; +extern struct amdgpu_mmhub_ras mmhub_v1_8_ras; #endif diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c index 278e32db878d7f987d36656091c278a2d1d1d911..8f76c6ecf50a86481a1f02a265f73ce154192bb9 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c @@ -187,7 +187,7 @@ mmhub_v2_0_print_l2_protection_fault_status(struct amdgpu_device *adev, static void mmhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; WREG32_SOC15_OFFSET_RLC(MMHUB, 0, mmMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, hub->ctx_addr_distance * vmid, @@ -362,7 +362,7 @@ static void mmhub_v2_0_disable_identity_aperture(struct amdgpu_device *adev) static void mmhub_v2_0_setup_vmid_config(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; int i; uint32_t tmp; @@ -412,7 +412,7 @@ static void mmhub_v2_0_setup_vmid_config(struct amdgpu_device *adev) static void mmhub_v2_0_program_invalidation(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; unsigned i; for (i = 0; i < 18; ++i) { @@ -441,7 +441,7 @@ static int mmhub_v2_0_gart_enable(struct amdgpu_device *adev) static void mmhub_v2_0_gart_disable(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; u32 tmp; u32 i; @@ -520,7 +520,7 @@ static const struct amdgpu_vmhub_funcs mmhub_v2_0_vmhub_funcs = { static void mmhub_v2_0_init(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; hub->ctx0_ptb_addr_lo32 = SOC15_REG_OFFSET(MMHUB, 0, diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c index fcf2813e70db80850535c2eb724e0a06318d22e6..8bd0fc8d9d25ed13ca9181805771b17b2f536433 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c @@ -121,7 +121,7 @@ static void mmhub_v2_3_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; WREG32_SOC15_OFFSET(MMHUB, 0, mmMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, hub->ctx_addr_distance * vmid, lower_32_bits(page_table_base)); @@ -280,7 +280,7 @@ static void mmhub_v2_3_disable_identity_aperture(struct amdgpu_device *adev) static void mmhub_v2_3_setup_vmid_config(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; int i; uint32_t tmp; @@ -330,7 +330,7 @@ static void mmhub_v2_3_setup_vmid_config(struct amdgpu_device *adev) static void mmhub_v2_3_program_invalidation(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; unsigned i; for (i = 0; i < 18; ++i) { @@ -373,7 +373,7 @@ static int mmhub_v2_3_gart_enable(struct amdgpu_device *adev) static void mmhub_v2_3_gart_disable(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; u32 tmp; u32 i; @@ -446,7 +446,7 @@ static const struct amdgpu_vmhub_funcs mmhub_v2_3_vmhub_funcs = { static void mmhub_v2_3_init(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; hub->ctx0_ptb_addr_lo32 = SOC15_REG_OFFSET(MMHUB, 0, diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0.c index 17a792616979ce5aca2dbd2919bddf1b27d31e51..441379e91cfa8e02cc9bf203171d348a28e63692 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0.c @@ -136,7 +136,7 @@ mmhub_v3_0_print_l2_protection_fault_status(struct amdgpu_device *adev, static void mmhub_v3_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; WREG32_SOC15_OFFSET(MMHUB, 0, regMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, hub->ctx_addr_distance * vmid, @@ -319,7 +319,7 @@ static void mmhub_v3_0_disable_identity_aperture(struct amdgpu_device *adev) static void mmhub_v3_0_setup_vmid_config(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; int i; uint32_t tmp; @@ -369,7 +369,7 @@ static void mmhub_v3_0_setup_vmid_config(struct amdgpu_device *adev) static void mmhub_v3_0_program_invalidation(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; unsigned i; for (i = 0; i < 18; ++i) { @@ -398,7 +398,7 @@ static int mmhub_v3_0_gart_enable(struct amdgpu_device *adev) static void mmhub_v3_0_gart_disable(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; u32 tmp; u32 i; @@ -477,7 +477,7 @@ static const struct amdgpu_vmhub_funcs mmhub_v3_0_vmhub_funcs = { static void mmhub_v3_0_init(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; hub->ctx0_ptb_addr_lo32 = SOC15_REG_OFFSET(MMHUB, 0, diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c index 26509b6b8c2402611c17eca9200d2a6e1d06592e..12c7f4b46ea94f69ceeae1f7a015ba5c91e7ebc4 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c @@ -138,7 +138,7 @@ static void mmhub_v3_0_1_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; WREG32_SOC15_OFFSET(MMHUB, 0, regMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, hub->ctx_addr_distance * vmid, @@ -306,7 +306,7 @@ static void mmhub_v3_0_1_disable_identity_aperture(struct amdgpu_device *adev) static void mmhub_v3_0_1_setup_vmid_config(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; int i; uint32_t tmp; @@ -356,7 +356,7 @@ static void mmhub_v3_0_1_setup_vmid_config(struct amdgpu_device *adev) static void mmhub_v3_0_1_program_invalidation(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; unsigned i; for (i = 0; i < 18; ++i) { @@ -385,7 +385,7 @@ static int mmhub_v3_0_1_gart_enable(struct amdgpu_device *adev) static void mmhub_v3_0_1_gart_disable(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; u32 tmp; u32 i; @@ -459,7 +459,7 @@ static const struct amdgpu_vmhub_funcs mmhub_v3_0_1_vmhub_funcs = { static void mmhub_v3_0_1_init(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; hub->ctx0_ptb_addr_lo32 = SOC15_REG_OFFSET(MMHUB, 0, diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_2.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_2.c index 26abbc6a47ab282cdb7bcfe9890fe490eacdd763..5dadc85abf7ef337ebbe6e68ba9fb1b849f72d84 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_2.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_2.c @@ -129,7 +129,7 @@ mmhub_v3_0_2_print_l2_protection_fault_status(struct amdgpu_device *adev, static void mmhub_v3_0_2_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; WREG32_SOC15_OFFSET(MMHUB, 0, regMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, hub->ctx_addr_distance * vmid, @@ -311,7 +311,7 @@ static void mmhub_v3_0_2_disable_identity_aperture(struct amdgpu_device *adev) static void mmhub_v3_0_2_setup_vmid_config(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; int i; uint32_t tmp; @@ -361,7 +361,7 @@ static void mmhub_v3_0_2_setup_vmid_config(struct amdgpu_device *adev) static void mmhub_v3_0_2_program_invalidation(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; unsigned i; for (i = 0; i < 18; ++i) { @@ -390,7 +390,7 @@ static int mmhub_v3_0_2_gart_enable(struct amdgpu_device *adev) static void mmhub_v3_0_2_gart_disable(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; u32 tmp; u32 i; @@ -469,7 +469,7 @@ static const struct amdgpu_vmhub_funcs mmhub_v3_0_2_vmhub_funcs = { static void mmhub_v3_0_2_init(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; hub->ctx0_ptb_addr_lo32 = SOC15_REG_OFFSET(MMHUB, 0, diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c index 72083e96222f03f05e108faeebf42d73055eecda..e790f890aec656b196cd7e010a4a011edb4ed309 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c @@ -57,7 +57,7 @@ static u64 mmhub_v9_4_get_fb_location(struct amdgpu_device *adev) static void mmhub_v9_4_setup_hubid_vm_pt_regs(struct amdgpu_device *adev, int hubid, uint32_t vmid, uint64_t value) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; WREG32_SOC15_OFFSET(MMHUB, 0, mmVML2VC0_VM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, @@ -294,7 +294,7 @@ static void mmhub_v9_4_disable_identity_aperture(struct amdgpu_device *adev, static void mmhub_v9_4_setup_vmid_config(struct amdgpu_device *adev, int hubid) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; unsigned int num_level, block_size; uint32_t tmp; int i; @@ -363,7 +363,7 @@ static void mmhub_v9_4_setup_vmid_config(struct amdgpu_device *adev, int hubid) static void mmhub_v9_4_program_invalidation(struct amdgpu_device *adev, int hubid) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; unsigned i; for (i = 0; i < 18; ++i) { @@ -404,7 +404,7 @@ static int mmhub_v9_4_gart_enable(struct amdgpu_device *adev) static void mmhub_v9_4_gart_disable(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB_0]; + struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; u32 tmp; u32 i, j; @@ -507,8 +507,8 @@ static void mmhub_v9_4_set_fault_enable_default(struct amdgpu_device *adev, bool static void mmhub_v9_4_init(struct amdgpu_device *adev) { - struct amdgpu_vmhub *hub[MMHUB_NUM_INSTANCES] = - {&adev->vmhub[AMDGPU_MMHUB_0], &adev->vmhub[AMDGPU_MMHUB_1]}; + struct amdgpu_vmhub *hub[MMHUB_NUM_INSTANCES] = { + &adev->vmhub[AMDGPU_MMHUB0(0)], &adev->vmhub[AMDGPU_MMHUB1(0)]}; int i; for (i = 0; i < MMHUB_NUM_INSTANCES; i++) { diff --git a/drivers/gpu/drm/amd/amdgpu/mmsch_v3_0.h b/drivers/gpu/drm/amd/amdgpu/mmsch_v3_0.h index 3e4e858a69652fbeea520519cbb8a72ffaf84819..a773ef61b78c5d79b2bc31eb23011b7b36194b93 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmsch_v3_0.h +++ b/drivers/gpu/drm/amd/amdgpu/mmsch_v3_0.h @@ -30,6 +30,8 @@ #define MMSCH_VERSION_MINOR 0 #define MMSCH_VERSION (MMSCH_VERSION_MAJOR << 16 | MMSCH_VERSION_MINOR) +#define MMSCH_V3_0_VCN_INSTANCES 0x2 + enum mmsch_v3_0_command_type { MMSCH_COMMAND__DIRECT_REG_WRITE = 0, MMSCH_COMMAND__DIRECT_REG_POLLING = 2, @@ -47,7 +49,7 @@ struct mmsch_v3_0_table_info { struct mmsch_v3_0_init_header { uint32_t version; uint32_t total_size; - struct mmsch_v3_0_table_info inst[AMDGPU_MAX_VCN_INSTANCES]; + struct mmsch_v3_0_table_info inst[MMSCH_V3_0_VCN_INSTANCES]; }; struct mmsch_v3_0_cmd_direct_reg_header { diff --git a/drivers/gpu/drm/amd/amdgpu/mmsch_v4_0.h b/drivers/gpu/drm/amd/amdgpu/mmsch_v4_0.h index 83653a50a1a219298555d92cc6927ba9fae8b00c..796d4f8791e5f5f8727296794d595efe5adb6a4b 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmsch_v4_0.h +++ b/drivers/gpu/drm/amd/amdgpu/mmsch_v4_0.h @@ -43,6 +43,8 @@ #define MMSCH_VF_MAILBOX_RESP__OK 0x1 #define MMSCH_VF_MAILBOX_RESP__INCOMPLETE 0x2 +#define MMSCH_V4_0_VCN_INSTANCES 0x2 + enum mmsch_v4_0_command_type { MMSCH_COMMAND__DIRECT_REG_WRITE = 0, MMSCH_COMMAND__DIRECT_REG_POLLING = 2, @@ -60,7 +62,7 @@ struct mmsch_v4_0_table_info { struct mmsch_v4_0_init_header { uint32_t version; uint32_t total_size; - struct mmsch_v4_0_table_info inst[AMDGPU_MAX_VCN_INSTANCES]; + struct mmsch_v4_0_table_info inst[MMSCH_V4_0_VCN_INSTANCES]; struct mmsch_v4_0_table_info jpegdec; }; diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c index 24d12075ca3aa86f7c8d010d84e72ee60e59afdb..cd1a02d30420ea94d729850cbee7c24b4fafbfa4 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c @@ -30,6 +30,8 @@ #include "ivsrcid/nbio/irqsrcs_nbif_7_4.h" #include +#define NPS_MODE_MASK 0x000000FFL + static void nbio_v7_9_remap_hdp_registers(struct amdgpu_device *adev) { WREG32_SOC15(NBIO, 0, regBIF_BX0_REMAP_HDP_MEM_FLUSH_CNTL, @@ -66,6 +68,13 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan bool use_doorbell, int doorbell_index, int doorbell_size) { u32 doorbell_range = 0, doorbell_ctrl = 0; + int aid_id, dev_inst; + + dev_inst = GET_INST(SDMA0, instance); + aid_id = adev->sdma.instance[instance].aid_id; + + if (use_doorbell == false) + return; doorbell_range = REG_SET_FIELD(doorbell_range, DOORBELL0_CTRL_ENTRY_0, @@ -80,9 +89,10 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan REG_SET_FIELD(doorbell_ctrl, S2A_DOORBELL_ENTRY_1_CTRL, S2A_DOORBELL_PORT1_RANGE_SIZE, doorbell_size); - switch (instance) { + switch (dev_inst % adev->sdma.num_inst_per_aid) { case 0: - WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_1, doorbell_range); + WREG32_SOC15_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_1, + 4 * aid_id, doorbell_range); doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl, S2A_DOORBELL_ENTRY_1_CTRL, @@ -94,10 +104,12 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan S2A_DOORBELL_ENTRY_1_CTRL, S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE, 0x1); - WREG32_SOC15(NBIO, 0, regS2A_DOORBELL_ENTRY_1_CTRL, doorbell_ctrl); + WREG32_SOC15_EXT(NBIO, aid_id, regS2A_DOORBELL_ENTRY_1_CTRL, + aid_id, doorbell_ctrl); break; case 1: - WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_2, doorbell_range); + WREG32_SOC15_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_2, + 4 * aid_id, doorbell_range); doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl, S2A_DOORBELL_ENTRY_1_CTRL, @@ -109,10 +121,12 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan S2A_DOORBELL_ENTRY_1_CTRL, S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE, 0x2); - WREG32_SOC15(NBIO, 0, regS2A_DOORBELL_ENTRY_2_CTRL, doorbell_ctrl); + WREG32_SOC15_EXT(NBIO, aid_id, regS2A_DOORBELL_ENTRY_2_CTRL, + aid_id, doorbell_ctrl); break; case 2: - WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_3, doorbell_range); + WREG32_SOC15_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_3, + 4 * aid_id, doorbell_range); doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl, S2A_DOORBELL_ENTRY_1_CTRL, @@ -124,10 +138,12 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan S2A_DOORBELL_ENTRY_1_CTRL, S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE, 0x8); - WREG32_SOC15(NBIO, 0, regS2A_DOORBELL_ENTRY_5_CTRL, doorbell_ctrl); + WREG32_SOC15_EXT(NBIO, aid_id, regS2A_DOORBELL_ENTRY_5_CTRL, + aid_id, doorbell_ctrl); break; case 3: - WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_4, doorbell_range); + WREG32_SOC15_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_4, + 4 * aid_id, doorbell_range); doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl, S2A_DOORBELL_ENTRY_1_CTRL, @@ -139,11 +155,12 @@ static void nbio_v7_9_sdma_doorbell_range(struct amdgpu_device *adev, int instan S2A_DOORBELL_ENTRY_1_CTRL, S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE, 0x9); - WREG32_SOC15(NBIO, 0, regS2A_DOORBELL_ENTRY_6_CTRL, doorbell_ctrl); + WREG32_SOC15_EXT(NBIO, aid_id, regS2A_DOORBELL_ENTRY_6_CTRL, + aid_id, doorbell_ctrl); break; default: break; - }; + } return; } @@ -152,6 +169,7 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do int doorbell_index, int instance) { u32 doorbell_range = 0, doorbell_ctrl = 0; + u32 aid_id = instance; if (use_doorbell) { doorbell_range = REG_SET_FIELD(doorbell_range, @@ -161,7 +179,12 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do doorbell_range = REG_SET_FIELD(doorbell_range, DOORBELL0_CTRL_ENTRY_0, BIF_DOORBELL0_RANGE_SIZE_ENTRY, - 0x8); + 0x9); + if (aid_id) + doorbell_range = REG_SET_FIELD(doorbell_range, + DOORBELL0_CTRL_ENTRY_0, + DOORBELL0_FENCE_ENABLE_ENTRY, + 0x4); doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl, S2A_DOORBELL_ENTRY_1_CTRL, @@ -174,10 +197,15 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do S2A_DOORBELL_PORT1_RANGE_OFFSET, 0x4); doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl, S2A_DOORBELL_ENTRY_1_CTRL, - S2A_DOORBELL_PORT1_RANGE_SIZE, 0x8); + S2A_DOORBELL_PORT1_RANGE_SIZE, 0x9); doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl, S2A_DOORBELL_ENTRY_1_CTRL, S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE, 0x4); + + WREG32_SOC15_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_17, + aid_id, doorbell_range); + WREG32_SOC15_EXT(NBIO, aid_id, regS2A_DOORBELL_ENTRY_4_CTRL, + aid_id, doorbell_ctrl); } else { doorbell_range = REG_SET_FIELD(doorbell_range, DOORBELL0_CTRL_ENTRY_0, @@ -185,10 +213,12 @@ static void nbio_v7_9_vcn_doorbell_range(struct amdgpu_device *adev, bool use_do doorbell_ctrl = REG_SET_FIELD(doorbell_ctrl, S2A_DOORBELL_ENTRY_1_CTRL, S2A_DOORBELL_PORT1_RANGE_SIZE, 0); - } - WREG32_SOC15(NBIO, 0, regDOORBELL0_CTRL_ENTRY_17, doorbell_range); - WREG32_SOC15(NBIO, 0, regS2A_DOORBELL_ENTRY_4_CTRL, doorbell_ctrl); + WREG32_SOC15_OFFSET(NBIO, 0, regDOORBELL0_CTRL_ENTRY_17, + aid_id, doorbell_range); + WREG32_SOC15_EXT(NBIO, aid_id, regS2A_DOORBELL_ENTRY_4_CTRL, + aid_id, doorbell_ctrl); + } } static void nbio_v7_9_enable_doorbell_aperture(struct amdgpu_device *adev, @@ -235,7 +265,7 @@ static void nbio_v7_9_ih_doorbell_range(struct amdgpu_device *adev, ih_doorbell_range = REG_SET_FIELD(ih_doorbell_range, DOORBELL0_CTRL_ENTRY_0, BIF_DOORBELL0_RANGE_SIZE_ENTRY, - 0x4); + 0x8); ih_doorbell_ctrl = REG_SET_FIELD(ih_doorbell_ctrl, S2A_DOORBELL_ENTRY_1_CTRL, @@ -248,7 +278,7 @@ static void nbio_v7_9_ih_doorbell_range(struct amdgpu_device *adev, S2A_DOORBELL_PORT1_RANGE_OFFSET, 0); ih_doorbell_ctrl = REG_SET_FIELD(ih_doorbell_ctrl, S2A_DOORBELL_ENTRY_1_CTRL, - S2A_DOORBELL_PORT1_RANGE_SIZE, 0x4); + S2A_DOORBELL_PORT1_RANGE_SIZE, 0x8); ih_doorbell_ctrl = REG_SET_FIELD(ih_doorbell_ctrl, S2A_DOORBELL_ENTRY_1_CTRL, S2A_DOORBELL_PORT1_AWADDR_31_28_VALUE, 0); @@ -319,6 +349,11 @@ static u32 nbio_v7_9_get_pcie_data_offset(struct amdgpu_device *adev) return SOC15_REG_OFFSET(NBIO, 0, regBIF_BX0_PCIE_DATA2); } +static u32 nbio_v7_9_get_pcie_index_hi_offset(struct amdgpu_device *adev) +{ + return SOC15_REG_OFFSET(NBIO, 0, regBIF_BX0_PCIE_INDEX2_HI); +} + const struct nbio_hdp_flush_reg nbio_v7_9_hdp_flush_reg = { .ref_and_mask_cp0 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__CP0_MASK, .ref_and_mask_cp1 = BIF_BX_PF0_GPU_HDP_FLUSH_DONE__CP1_MASK, @@ -347,11 +382,57 @@ static void nbio_v7_9_enable_doorbell_interrupt(struct amdgpu_device *adev, DOORBELL_INTERRUPT_DISABLE, enable ? 0 : 1); } +static int nbio_v7_9_get_compute_partition_mode(struct amdgpu_device *adev) +{ + u32 tmp, px; + + tmp = RREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_COMPUTE_STATUS); + px = REG_GET_FIELD(tmp, BIF_BX_PF0_PARTITION_COMPUTE_STATUS, + PARTITION_MODE); + + return px; +} + +static u32 nbio_v7_9_get_memory_partition_mode(struct amdgpu_device *adev, + u32 *supp_modes) +{ + u32 tmp; + + tmp = RREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_MEM_STATUS); + tmp = REG_GET_FIELD(tmp, BIF_BX_PF0_PARTITION_MEM_STATUS, NPS_MODE); + + if (supp_modes) { + *supp_modes = + RREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_MEM_CAP); + } + + return ffs(tmp); +} + +static void nbio_v7_9_init_registers(struct amdgpu_device *adev) +{ + u32 inst_mask; + int i; + + WREG32_SOC15(NBIO, 0, regXCC_DOORBELL_FENCE, + 0xff & ~(adev->gfx.xcc_mask)); + + WREG32_SOC15(NBIO, 0, regBIFC_GFX_INT_MONITOR_MASK, 0x7ff); + + inst_mask = adev->aid_mask & ~1U; + for_each_inst(i, inst_mask) { + WREG32_SOC15_EXT(NBIO, i, regXCC_DOORBELL_FENCE, i, + XCC_DOORBELL_FENCE__SHUB_SLV_MODE_MASK); + + } +} + const struct amdgpu_nbio_funcs nbio_v7_9_funcs = { .get_hdp_flush_req_offset = nbio_v7_9_get_hdp_flush_req_offset, .get_hdp_flush_done_offset = nbio_v7_9_get_hdp_flush_done_offset, .get_pcie_index_offset = nbio_v7_9_get_pcie_index_offset, .get_pcie_data_offset = nbio_v7_9_get_pcie_data_offset, + .get_pcie_index_hi_offset = nbio_v7_9_get_pcie_index_hi_offset, .get_rev_id = nbio_v7_9_get_rev_id, .mc_access_enable = nbio_v7_9_mc_access_enable, .get_memsize = nbio_v7_9_get_memsize, @@ -366,4 +447,7 @@ const struct amdgpu_nbio_funcs nbio_v7_9_funcs = { .get_clockgating_state = nbio_v7_9_get_clockgating_state, .ih_control = nbio_v7_9_ih_control, .remap_hdp_registers = nbio_v7_9_remap_hdp_registers, + .get_compute_partition_mode = nbio_v7_9_get_compute_partition_mode, + .get_memory_partition_mode = nbio_v7_9_get_memory_partition_mode, + .init_registers = nbio_v7_9_init_registers, }; diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c index 0fb6013441f07e32e9abc51465b6318fea4c5d9a..51523b27a186ca6cc3a5889b2ab60e493e803f85 100644 --- a/drivers/gpu/drm/amd/amdgpu/nv.c +++ b/drivers/gpu/drm/amd/amdgpu/nv.c @@ -341,11 +341,6 @@ void nv_grbm_select(struct amdgpu_device *adev, WREG32_SOC15(GC, 0, mmGRBM_GFX_CNTL, grbm_gfx_cntl); } -static void nv_vga_set_state(struct amdgpu_device *adev, bool state) -{ - /* todo */ -} - static bool nv_read_disabled_bios(struct amdgpu_device *adev) { /* todo */ @@ -381,12 +376,12 @@ static uint32_t nv_read_indexed_register(struct amdgpu_device *adev, u32 se_num, mutex_lock(&adev->grbm_idx_mutex); if (se_num != 0xffffffff || sh_num != 0xffffffff) - amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff, 0); val = RREG32(reg_offset); if (se_num != 0xffffffff || sh_num != 0xffffffff) - amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); return val; } @@ -632,9 +627,9 @@ static int nv_update_umd_stable_pstate(struct amdgpu_device *adev, bool enter) { if (enter) - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); else - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); if (adev->gfx.funcs->update_perfmon_mgcg) adev->gfx.funcs->update_perfmon_mgcg(adev, !enter); @@ -654,7 +649,6 @@ static const struct amdgpu_asic_funcs nv_asic_funcs = .read_register = &nv_read_register, .reset = &nv_asic_reset, .reset_method = &nv_asic_reset_method, - .set_vga_state = &nv_vga_set_state, .get_xclk = &nv_get_xclk, .set_uvd_clocks = &nv_set_uvd_clocks, .set_vce_clocks = &nv_set_vce_clocks, diff --git a/drivers/gpu/drm/amd/amdgpu/nvd.h b/drivers/gpu/drm/amd/amdgpu/nvd.h index fd6b58243b03279b52d5f042e39d2727370924e9..631dafb9229900474845cd50a3a3144a8f748299 100644 --- a/drivers/gpu/drm/amd/amdgpu/nvd.h +++ b/drivers/gpu/drm/amd/amdgpu/nvd.h @@ -462,6 +462,9 @@ # define PACKET3_QUERY_STATUS_ENG_SEL(x) ((x) << 25) #define PACKET3_RUN_LIST 0xA5 #define PACKET3_MAP_PROCESS_VM 0xA6 - +/* GFX11 */ +#define PACKET3_SET_Q_PREEMPTION_MODE 0xF0 +# define PACKET3_SET_Q_PREEMPTION_MODE_IB_VMID(x) ((x) << 0) +# define PACKET3_SET_Q_PREEMPTION_MODE_INIT_SHADOW_MEM (1 << 0) #endif diff --git a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h index 22c775f39119b7be8584c312661a045fc2ab7848..18917df785ecafc76e08dc99a78973be53df220c 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h +++ b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h @@ -102,6 +102,7 @@ enum psp_gfx_cmd_id GFX_CMD_ID_LOAD_TOC = 0x00000020, /* Load TOC and obtain TMR size */ GFX_CMD_ID_AUTOLOAD_RLC = 0x00000021, /* Indicates all graphics fw loaded, start RLC autoload */ GFX_CMD_ID_BOOT_CFG = 0x00000022, /* Boot Config */ + GFX_CMD_ID_SRIOV_SPATIAL_PART = 0x00000027, /* Configure spatial partitioning mode */ }; /* PSP boot config sub-commands */ @@ -338,6 +339,13 @@ struct psp_gfx_cmd_boot_cfg uint32_t boot_config_valid; /* dynamic boot configuration valid bits bitmask */ }; +struct psp_gfx_cmd_sriov_spatial_part { + uint32_t mode; + uint32_t override_ips; + uint32_t override_xcds_avail; + uint32_t override_this_aid; +}; + /* All GFX ring buffer commands. */ union psp_gfx_commands { @@ -351,6 +359,7 @@ union psp_gfx_commands struct psp_gfx_cmd_setup_tmr cmd_setup_vmr; struct psp_gfx_cmd_load_toc cmd_load_toc; struct psp_gfx_cmd_boot_cfg boot_cfg; + struct psp_gfx_cmd_sriov_spatial_part cmd_spatial_part; }; struct psp_gfx_uresp_reserved diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c index e1b7fca096660a99387079c8fd73a99053de916b..5f10883da6a2388ef7f8b1354f5e3eddb0d970dd 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c @@ -57,7 +57,13 @@ static int psp_v10_0_init_microcode(struct psp_context *psp) if (err) return err; - return psp_init_ta_microcode(psp, ucode_prefix); + err = psp_init_ta_microcode(psp, ucode_prefix); + if ((adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 1, 0)) && + (adev->pdev->revision == 0xa1) && + (psp->securedisplay_context.context.bin_desc.fw_version >= 0x27000008)) { + adev->psp.securedisplay_context.context.bin_desc.size_bytes = 0; + } + return err; } static int psp_v10_0_ring_create(struct psp_context *psp, diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c index caee76ab71105f2b93a98c68497c9ecfaf9dcfad..f9cb0d2c89d15bf2b36e72e73392ad446e929faa 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c @@ -624,10 +624,11 @@ static int psp_v13_0_exec_spi_cmd(struct psp_context *psp, int cmd) WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_73, 1); if (cmd == C2PMSG_CMD_SPI_UPDATE_FLASH_IMAGE) - return 0; - - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_115), - MBOX_READY_FLAG, MBOX_READY_MASK, false); + ret = psp_wait_for_spirom_update(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_115), + MBOX_READY_FLAG, MBOX_READY_MASK, PSP_SPIROM_UPDATE_TIMEOUT); + else + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_115), + MBOX_READY_FLAG, MBOX_READY_MASK, false); if (ret) { dev_err(adev->dev, "SPI cmd %x timed out, ret = %d", cmd, ret); return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.h b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.h index b2414a729ca17890765a5d4fb0a307dc980f24da..de5677ce4330dd33d2a65b75dc396b6bfecca9fb 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.h +++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.h @@ -25,6 +25,8 @@ #include "amdgpu_psp.h" +#define PSP_SPIROM_UPDATE_TIMEOUT 60000 /* 60s */ + void psp_v13_0_set_psp_funcs(struct psp_context *psp); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c index fd2a7b66ac56216ea3b2ba2cfb18089befb2d221..51afc92994a80d1ee525067102287913ce82bf86 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c @@ -466,8 +466,6 @@ static int sdma_v2_4_gfx_resume(struct amdgpu_device *adev) #endif /* enable DMA IBs */ WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl); - - ring->sched.ready = true; } sdma_v2_4_enable(adev, true); diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c index e572389089d249ae1e2c8f0ca76bf483b08f8e82..344202870aebec2d3a24d6c23c4f94c26019686a 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c @@ -734,8 +734,6 @@ static int sdma_v3_0_gfx_resume(struct amdgpu_device *adev) #endif /* enable DMA IBs */ WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl); - - ring->sched.ready = true; } /* unhalt the MEs */ diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c index 9295ac7edd5650adf959e1240ada69bef3357032..cd37f45e01a119d824b808dca66e04727de25199 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c @@ -1114,8 +1114,6 @@ static void sdma_v4_0_gfx_resume(struct amdgpu_device *adev, unsigned int i) #endif /* enable DMA IBs */ WREG32_SDMA(i, mmSDMA0_GFX_IB_CNTL, ib_cntl); - - ring->sched.ready = true; } /** @@ -1202,8 +1200,6 @@ static void sdma_v4_0_page_resume(struct amdgpu_device *adev, unsigned int i) #endif /* enable DMA IBs */ WREG32_SDMA(i, mmSDMA0_PAGE_IB_CNTL, ib_cntl); - - ring->sched.ready = true; } static void @@ -1825,12 +1821,12 @@ static int sdma_v4_0_sw_init(void *handle) /* * On Arcturus, SDMA instance 5~7 has a different vmhub - * type(AMDGPU_MMHUB_1). + * type(AMDGPU_MMHUB1). */ if (adev->ip_versions[SDMA0_HWIP][0] == IP_VERSION(4, 2, 2) && i >= 5) - ring->vm_hub = AMDGPU_MMHUB_1; + ring->vm_hub = AMDGPU_MMHUB1(0); else - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); sprintf(ring->name, "sdma%d", i); r = amdgpu_ring_init(adev, ring, 1024, &adev->sdma.trap_irq, @@ -1847,13 +1843,23 @@ static int sdma_v4_0_sw_init(void *handle) /* paging queue use same doorbell index/routing as gfx queue * with 0x400 (4096 dwords) offset on second doorbell page */ - ring->doorbell_index = adev->doorbell_index.sdma_engine[i] << 1; - ring->doorbell_index += 0x400; + if (adev->ip_versions[SDMA0_HWIP][0] >= IP_VERSION(4, 0, 0) && + adev->ip_versions[SDMA0_HWIP][0] < IP_VERSION(4, 2, 0)) { + ring->doorbell_index = + adev->doorbell_index.sdma_engine[i] << 1; + ring->doorbell_index += 0x400; + } else { + /* From vega20, the sdma_doorbell_range in 1st + * doorbell page is reserved for page queue. + */ + ring->doorbell_index = + (adev->doorbell_index.sdma_engine[i] + 1) << 1; + } if (adev->ip_versions[SDMA0_HWIP][0] == IP_VERSION(4, 2, 2) && i >= 5) - ring->vm_hub = AMDGPU_MMHUB_1; + ring->vm_hub = AMDGPU_MMHUB1(0); else - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); sprintf(ring->name, "page%d", i); r = amdgpu_ring_init(adev, ring, 1024, @@ -2306,7 +2312,7 @@ const struct amd_ip_funcs sdma_v4_0_ip_funcs = { static const struct amdgpu_ring_funcs sdma_v4_0_ring_funcs = { .type = AMDGPU_RING_TYPE_SDMA, - .align_mask = 0xf, + .align_mask = 0xff, .nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), .support_64bit_ptrs = true, .secure_submission_supported = true, @@ -2338,7 +2344,7 @@ static const struct amdgpu_ring_funcs sdma_v4_0_ring_funcs = { static const struct amdgpu_ring_funcs sdma_v4_0_page_ring_funcs = { .type = AMDGPU_RING_TYPE_SDMA, - .align_mask = 0xf, + .align_mask = 0xff, .nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), .support_64bit_ptrs = true, .secure_submission_supported = true, diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4.c index 6f9895cdddb10e41b010c128da121376fb8ec504..0ddb6955a6d3be5f32322c72207e82463e68d320 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4.c @@ -141,6 +141,10 @@ static const struct soc15_ras_field_entry sdma_v4_4_ras_fields[] = { SOC15_REG_FIELD(SDMA0_EDC_COUNTER2, SDMA_UTCL1_RDBST_FIFO_SED), 0, 0, }, + { "SDMA_UTCL1_WR_FIFO_SED", SOC15_REG_ENTRY(SDMA0, 0, regSDMA0_EDC_COUNTER2), + SOC15_REG_FIELD(SDMA0_EDC_COUNTER2, SDMA_UTCL1_WR_FIFO_SED), + 0, 0, + }, { "SDMA_DATA_LUT_FIFO_SED", SOC15_REG_ENTRY(SDMA0, 0, regSDMA0_EDC_COUNTER2), SOC15_REG_FIELD(SDMA0_EDC_COUNTER2, SDMA_DATA_LUT_FIFO_SED), 0, 0, diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c index 64dcaa2670dd155cc00fe74126b37f94aa2f9508..ea5e12390d182830f3bcce6a531154b082a00479 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c @@ -27,6 +27,7 @@ #include #include "amdgpu.h" +#include "amdgpu_xcp.h" #include "amdgpu_ucode.h" #include "amdgpu_trace.h" @@ -53,11 +54,14 @@ static void sdma_v4_4_2_set_ring_funcs(struct amdgpu_device *adev); static void sdma_v4_4_2_set_buffer_funcs(struct amdgpu_device *adev); static void sdma_v4_4_2_set_vm_pte_funcs(struct amdgpu_device *adev); static void sdma_v4_4_2_set_irq_funcs(struct amdgpu_device *adev); +static void sdma_v4_4_2_set_ras_funcs(struct amdgpu_device *adev); static u32 sdma_v4_4_2_get_reg_offset(struct amdgpu_device *adev, u32 instance, u32 offset) { - return (adev->reg_offset[SDMA0_HWIP][instance][0] + offset); + u32 dev_inst = GET_INST(SDMA0, instance); + + return (adev->reg_offset[SDMA0_HWIP][dev_inst][0] + offset); } static unsigned sdma_v4_4_2_seq_to_irq_id(int seq_num) @@ -92,13 +96,25 @@ static int sdma_v4_4_2_irq_id_to_seq(unsigned client_id) } } -static void sdma_v4_4_2_init_golden_registers(struct amdgpu_device *adev) +static void sdma_v4_4_2_inst_init_golden_registers(struct amdgpu_device *adev, + uint32_t inst_mask) { - switch (adev->ip_versions[SDMA0_HWIP][0]) { - case IP_VERSION(4, 4, 2): - break; - default: - break; + u32 val; + int i; + + for (i = 0; i < adev->sdma.num_instances; i++) { + val = RREG32_SDMA(i, regSDMA_GB_ADDR_CONFIG); + val = REG_SET_FIELD(val, SDMA_GB_ADDR_CONFIG, NUM_BANKS, 4); + val = REG_SET_FIELD(val, SDMA_GB_ADDR_CONFIG, + PIPE_INTERLEAVE_SIZE, 0); + WREG32_SDMA(i, regSDMA_GB_ADDR_CONFIG, val); + + val = RREG32_SDMA(i, regSDMA_GB_ADDR_CONFIG_READ); + val = REG_SET_FIELD(val, SDMA_GB_ADDR_CONFIG_READ, NUM_BANKS, + 4); + val = REG_SET_FIELD(val, SDMA_GB_ADDR_CONFIG_READ, + PIPE_INTERLEAVE_SIZE, 0); + WREG32_SDMA(i, regSDMA_GB_ADDR_CONFIG_READ, val); } } @@ -399,19 +415,21 @@ static void sdma_v4_4_2_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 /** - * sdma_v4_4_2_gfx_stop - stop the gfx async dma engines + * sdma_v4_4_2_inst_gfx_stop - stop the gfx async dma engines * * @adev: amdgpu_device pointer + * @inst_mask: mask of dma engine instances to be disabled * * Stop the gfx async dma ring buffers. */ -static void sdma_v4_4_2_gfx_stop(struct amdgpu_device *adev) +static void sdma_v4_4_2_inst_gfx_stop(struct amdgpu_device *adev, + uint32_t inst_mask) { struct amdgpu_ring *sdma[AMDGPU_MAX_SDMA_INSTANCES]; u32 rb_cntl, ib_cntl; int i, unset = 0; - for (i = 0; i < adev->sdma.num_instances; i++) { + for_each_inst(i, inst_mask) { sdma[i] = &adev->sdma.instance[i].ring; if ((adev->mman.buffer_funcs_ring == sdma[i]) && unset != 1) { @@ -429,32 +447,36 @@ static void sdma_v4_4_2_gfx_stop(struct amdgpu_device *adev) } /** - * sdma_v4_4_2_rlc_stop - stop the compute async dma engines + * sdma_v4_4_2_inst_rlc_stop - stop the compute async dma engines * * @adev: amdgpu_device pointer + * @inst_mask: mask of dma engine instances to be disabled * * Stop the compute async dma queues. */ -static void sdma_v4_4_2_rlc_stop(struct amdgpu_device *adev) +static void sdma_v4_4_2_inst_rlc_stop(struct amdgpu_device *adev, + uint32_t inst_mask) { /* XXX todo */ } /** - * sdma_v4_4_2_page_stop - stop the page async dma engines + * sdma_v4_4_2_inst_page_stop - stop the page async dma engines * * @adev: amdgpu_device pointer + * @inst_mask: mask of dma engine instances to be disabled * * Stop the page async dma ring buffers. */ -static void sdma_v4_4_2_page_stop(struct amdgpu_device *adev) +static void sdma_v4_4_2_inst_page_stop(struct amdgpu_device *adev, + uint32_t inst_mask) { struct amdgpu_ring *sdma[AMDGPU_MAX_SDMA_INSTANCES]; u32 rb_cntl, ib_cntl; int i; bool unset = false; - for (i = 0; i < adev->sdma.num_instances; i++) { + for_each_inst(i, inst_mask) { sdma[i] = &adev->sdma.instance[i].page; if ((adev->mman.buffer_funcs_ring == sdma[i]) && @@ -475,14 +497,16 @@ static void sdma_v4_4_2_page_stop(struct amdgpu_device *adev) } /** - * sdma_v4_4_2_ctx_switch_enable - stop the async dma engines context switch + * sdma_v4_4_2_inst_ctx_switch_enable - stop the async dma engines context switch * * @adev: amdgpu_device pointer * @enable: enable/disable the DMA MEs context switch. + * @inst_mask: mask of dma engine instances to be enabled * * Halt or unhalt the async dma engines context switch. */ -static void sdma_v4_4_2_ctx_switch_enable(struct amdgpu_device *adev, bool enable) +static void sdma_v4_4_2_inst_ctx_switch_enable(struct amdgpu_device *adev, + bool enable, uint32_t inst_mask) { u32 f32_cntl, phase_quantum = 0; int i; @@ -511,7 +535,7 @@ static void sdma_v4_4_2_ctx_switch_enable(struct amdgpu_device *adev, bool enabl unit << SDMA_PHASE0_QUANTUM__UNIT__SHIFT; } - for (i = 0; i < adev->sdma.num_instances; i++) { + for_each_inst(i, inst_mask) { f32_cntl = RREG32_SDMA(i, regSDMA_CNTL); f32_cntl = REG_SET_FIELD(f32_cntl, SDMA_CNTL, AUTO_CTXSW_ENABLE, enable ? 1 : 0); @@ -525,30 +549,36 @@ static void sdma_v4_4_2_ctx_switch_enable(struct amdgpu_device *adev, bool enabl /* Extend page fault timeout to avoid interrupt storm */ WREG32_SDMA(i, regSDMA_UTCL1_TIMEOUT, 0x00800080); } - } /** - * sdma_v4_4_2_enable - stop the async dma engines + * sdma_v4_4_2_inst_enable - stop the async dma engines * * @adev: amdgpu_device pointer * @enable: enable/disable the DMA MEs. + * @inst_mask: mask of dma engine instances to be enabled * * Halt or unhalt the async dma engines. */ -static void sdma_v4_4_2_enable(struct amdgpu_device *adev, bool enable) +static void sdma_v4_4_2_inst_enable(struct amdgpu_device *adev, bool enable, + uint32_t inst_mask) { u32 f32_cntl; int i; if (!enable) { - sdma_v4_4_2_gfx_stop(adev); - sdma_v4_4_2_rlc_stop(adev); + sdma_v4_4_2_inst_gfx_stop(adev, inst_mask); + sdma_v4_4_2_inst_rlc_stop(adev, inst_mask); if (adev->sdma.has_page_queue) - sdma_v4_4_2_page_stop(adev); + sdma_v4_4_2_inst_page_stop(adev, inst_mask); + + /* SDMA FW needs to respond to FREEZE requests during reset. + * Keep it running during reset */ + if (!amdgpu_sriov_vf(adev) && amdgpu_in_reset(adev)) + return; } - for (i = 0; i < adev->sdma.num_instances; i++) { + for_each_inst(i, inst_mask) { f32_cntl = RREG32_SDMA(i, regSDMA_F32_CNTL); f32_cntl = REG_SET_FIELD(f32_cntl, SDMA_F32_CNTL, HALT, enable ? 0 : 1); WREG32_SDMA(i, regSDMA_F32_CNTL, f32_cntl); @@ -659,8 +689,6 @@ static void sdma_v4_4_2_gfx_resume(struct amdgpu_device *adev, unsigned int i) #endif /* enable DMA IBs */ WREG32_SDMA(i, regSDMA_GFX_IB_CNTL, ib_cntl); - - ring->sched.ready = true; } /** @@ -750,8 +778,6 @@ static void sdma_v4_4_2_page_resume(struct amdgpu_device *adev, unsigned int i) #endif /* enable DMA IBs */ WREG32_SDMA(i, regSDMA_PAGE_IB_CNTL, ib_cntl); - - ring->sched.ready = true; } static void sdma_v4_4_2_init_pg(struct amdgpu_device *adev) @@ -760,14 +786,16 @@ static void sdma_v4_4_2_init_pg(struct amdgpu_device *adev) } /** - * sdma_v4_4_2_rlc_resume - setup and start the async dma engines + * sdma_v4_4_2_inst_rlc_resume - setup and start the async dma engines * * @adev: amdgpu_device pointer + * @inst_mask: mask of dma engine instances to be enabled * * Set up the compute DMA queues and enable them. * Returns 0 for success, error for failure. */ -static int sdma_v4_4_2_rlc_resume(struct amdgpu_device *adev) +static int sdma_v4_4_2_inst_rlc_resume(struct amdgpu_device *adev, + uint32_t inst_mask) { sdma_v4_4_2_init_pg(adev); @@ -775,14 +803,16 @@ static int sdma_v4_4_2_rlc_resume(struct amdgpu_device *adev) } /** - * sdma_v4_4_2_load_microcode - load the sDMA ME ucode + * sdma_v4_4_2_inst_load_microcode - load the sDMA ME ucode * * @adev: amdgpu_device pointer + * @inst_mask: mask of dma engine instances to be enabled * * Loads the sDMA0/1 ucode. * Returns 0 for success, -EINVAL if the ucode is not available. */ -static int sdma_v4_4_2_load_microcode(struct amdgpu_device *adev) +static int sdma_v4_4_2_inst_load_microcode(struct amdgpu_device *adev, + uint32_t inst_mask) { const struct sdma_firmware_header_v1_0 *hdr; const __le32 *fw_data; @@ -790,9 +820,9 @@ static int sdma_v4_4_2_load_microcode(struct amdgpu_device *adev) int i, j; /* halt the MEs */ - sdma_v4_4_2_enable(adev, false); + sdma_v4_4_2_inst_enable(adev, false, inst_mask); - for (i = 0; i < adev->sdma.num_instances; i++) { + for_each_inst(i, inst_mask) { if (!adev->sdma.instance[i].fw) return -EINVAL; @@ -818,38 +848,42 @@ static int sdma_v4_4_2_load_microcode(struct amdgpu_device *adev) } /** - * sdma_v4_4_2_start - setup and start the async dma engines + * sdma_v4_4_2_inst_start - setup and start the async dma engines * * @adev: amdgpu_device pointer + * @inst_mask: mask of dma engine instances to be enabled * * Set up the DMA engines and enable them. * Returns 0 for success, error for failure. */ -static int sdma_v4_4_2_start(struct amdgpu_device *adev) +static int sdma_v4_4_2_inst_start(struct amdgpu_device *adev, + uint32_t inst_mask) { struct amdgpu_ring *ring; + uint32_t tmp_mask; int i, r = 0; if (amdgpu_sriov_vf(adev)) { - sdma_v4_4_2_ctx_switch_enable(adev, false); - sdma_v4_4_2_enable(adev, false); + sdma_v4_4_2_inst_ctx_switch_enable(adev, false, inst_mask); + sdma_v4_4_2_inst_enable(adev, false, inst_mask); } else { /* bypass sdma microcode loading on Gopher */ if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP && - !(adev->pdev->device == 0x49) && !(adev->pdev->device == 0x50)) { - r = sdma_v4_4_2_load_microcode(adev); + adev->sdma.instance[0].fw) { + r = sdma_v4_4_2_inst_load_microcode(adev, inst_mask); if (r) return r; } /* unhalt the MEs */ - sdma_v4_4_2_enable(adev, true); + sdma_v4_4_2_inst_enable(adev, true, inst_mask); /* enable sdma ring preemption */ - sdma_v4_4_2_ctx_switch_enable(adev, true); + sdma_v4_4_2_inst_ctx_switch_enable(adev, true, inst_mask); } /* start the gfx rings and rlc compute queues */ - for (i = 0; i < adev->sdma.num_instances; i++) { + tmp_mask = inst_mask; + for_each_inst(i, tmp_mask) { uint32_t temp; WREG32_SDMA(i, regSDMA_SEM_WAIT_FAIL_TIMER_CNTL, 0); @@ -860,6 +894,8 @@ static int sdma_v4_4_2_start(struct amdgpu_device *adev) /* set utc l1 enable flag always to 1 */ temp = RREG32_SDMA(i, regSDMA_CNTL); temp = REG_SET_FIELD(temp, SDMA_CNTL, UTC_L1_ENABLE, 1); + /* enable context empty interrupt during initialization */ + temp = REG_SET_FIELD(temp, SDMA_CNTL, CTXEMPTY_INT_ENABLE, 1); WREG32_SDMA(i, regSDMA_CNTL, temp); if (!amdgpu_sriov_vf(adev)) { @@ -876,15 +912,16 @@ static int sdma_v4_4_2_start(struct amdgpu_device *adev) } if (amdgpu_sriov_vf(adev)) { - sdma_v4_4_2_ctx_switch_enable(adev, true); - sdma_v4_4_2_enable(adev, true); + sdma_v4_4_2_inst_ctx_switch_enable(adev, true, inst_mask); + sdma_v4_4_2_inst_enable(adev, true, inst_mask); } else { - r = sdma_v4_4_2_rlc_resume(adev); + r = sdma_v4_4_2_inst_rlc_resume(adev, inst_mask); if (r) return r; } - for (i = 0; i < adev->sdma.num_instances; i++) { + tmp_mask = inst_mask; + for_each_inst(i, tmp_mask) { ring = &adev->sdma.instance[i].ring; r = amdgpu_ring_test_helper(ring); @@ -1221,6 +1258,7 @@ static int sdma_v4_4_2_early_init(void *handle) sdma_v4_4_2_set_buffer_funcs(adev); sdma_v4_4_2_set_vm_pte_funcs(adev); sdma_v4_4_2_set_irq_funcs(adev); + sdma_v4_4_2_set_ras_funcs(adev); return 0; } @@ -1253,9 +1291,10 @@ static int sdma_v4_4_2_sw_init(void *handle) struct amdgpu_ring *ring; int r, i; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + u32 aid_id; /* SDMA trap event */ - for (i = 0; i < adev->sdma.num_instances; i++) { + for (i = 0; i < adev->sdma.num_inst_per_aid; i++) { r = amdgpu_irq_add_id(adev, sdma_v4_4_2_seq_to_irq_id(i), SDMA0_4_0__SRCID__SDMA_TRAP, &adev->sdma.trap_irq); @@ -1264,7 +1303,7 @@ static int sdma_v4_4_2_sw_init(void *handle) } /* SDMA SRAM ECC event */ - for (i = 0; i < adev->sdma.num_instances; i++) { + for (i = 0; i < adev->sdma.num_inst_per_aid; i++) { r = amdgpu_irq_add_id(adev, sdma_v4_4_2_seq_to_irq_id(i), SDMA0_4_0__SRCID__SDMA_SRAM_ECC, &adev->sdma.ecc_irq); @@ -1273,7 +1312,7 @@ static int sdma_v4_4_2_sw_init(void *handle) } /* SDMA VM_HOLE/DOORBELL_INV/POLL_TIMEOUT/SRBM_WRITE_PROTECTION event*/ - for (i = 0; i < adev->sdma.num_instances; i++) { + for (i = 0; i < adev->sdma.num_inst_per_aid; i++) { r = amdgpu_irq_add_id(adev, sdma_v4_4_2_seq_to_irq_id(i), SDMA0_4_0__SRCID__SDMA_VM_HOLE, &adev->sdma.vm_hole_irq); @@ -1303,15 +1342,17 @@ static int sdma_v4_4_2_sw_init(void *handle) ring = &adev->sdma.instance[i].ring; ring->ring_obj = NULL; ring->use_doorbell = true; + aid_id = adev->sdma.instance[i].aid_id; DRM_DEBUG("SDMA %d use_doorbell being set to: [%s]\n", i, ring->use_doorbell?"true":"false"); /* doorbell size is 2 dwords, get DWORD offset */ ring->doorbell_index = adev->doorbell_index.sdma_engine[i] << 1; - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(aid_id); - sprintf(ring->name, "sdma%d", i); + sprintf(ring->name, "sdma%d.%d", aid_id, + i % adev->sdma.num_inst_per_aid); r = amdgpu_ring_init(adev, ring, 1024, &adev->sdma.trap_irq, AMDGPU_SDMA_IRQ_INSTANCE0 + i, AMDGPU_RING_PRIO_DEFAULT, NULL); @@ -1323,14 +1364,15 @@ static int sdma_v4_4_2_sw_init(void *handle) ring->ring_obj = NULL; ring->use_doorbell = true; - /* paging queue use same doorbell index/routing as gfx queue - * with 0x400 (4096 dwords) offset on second doorbell page + /* doorbell index of page queue is assigned right after + * gfx queue on the same instance */ - ring->doorbell_index = adev->doorbell_index.sdma_engine[i] << 1; - ring->doorbell_index += 0x400; - ring->vm_hub = AMDGPU_MMHUB_0; + ring->doorbell_index = + (adev->doorbell_index.sdma_engine[i] + 1) << 1; + ring->vm_hub = AMDGPU_MMHUB0(aid_id); - sprintf(ring->name, "page%d", i); + sprintf(ring->name, "page%d.%d", aid_id, + i % adev->sdma.num_inst_per_aid); r = amdgpu_ring_init(adev, ring, 1024, &adev->sdma.trap_irq, AMDGPU_SDMA_IRQ_INSTANCE0 + i, @@ -1340,6 +1382,11 @@ static int sdma_v4_4_2_sw_init(void *handle) } } + if (amdgpu_sdma_ras_sw_init(adev)) { + dev_err(adev->dev, "fail to initialize sdma ras block\n"); + return -EINVAL; + } + return r; } @@ -1366,14 +1413,13 @@ static int sdma_v4_4_2_hw_init(void *handle) { int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + uint32_t inst_mask; - if (adev->flags & AMD_IS_APU) - amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_SDMA, false); - + inst_mask = GENMASK(adev->sdma.num_instances - 1, 0); if (!amdgpu_sriov_vf(adev)) - sdma_v4_4_2_init_golden_registers(adev); + sdma_v4_4_2_inst_init_golden_registers(adev, inst_mask); - r = sdma_v4_4_2_start(adev); + r = sdma_v4_4_2_inst_start(adev, inst_mask); return r; } @@ -1381,26 +1427,36 @@ static int sdma_v4_4_2_hw_init(void *handle) static int sdma_v4_4_2_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + uint32_t inst_mask; int i; if (amdgpu_sriov_vf(adev)) return 0; - for (i = 0; i < adev->sdma.num_instances; i++) { - amdgpu_irq_put(adev, &adev->sdma.ecc_irq, - AMDGPU_SDMA_IRQ_INSTANCE0 + i); + inst_mask = GENMASK(adev->sdma.num_instances - 1, 0); + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA)) { + for (i = 0; i < adev->sdma.num_instances; i++) { + amdgpu_irq_put(adev, &adev->sdma.ecc_irq, + AMDGPU_SDMA_IRQ_INSTANCE0 + i); + } } - sdma_v4_4_2_ctx_switch_enable(adev, false); - sdma_v4_4_2_enable(adev, false); + sdma_v4_4_2_inst_ctx_switch_enable(adev, false, inst_mask); + sdma_v4_4_2_inst_enable(adev, false, inst_mask); return 0; } +static int sdma_v4_4_2_set_clockgating_state(void *handle, + enum amd_clockgating_state state); + static int sdma_v4_4_2_suspend(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + if (amdgpu_in_reset(adev)) + sdma_v4_4_2_set_clockgating_state(adev, AMD_CG_STATE_UNGATE); + return sdma_v4_4_2_hw_fini(adev); } @@ -1471,13 +1527,31 @@ static int sdma_v4_4_2_process_trap_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { - uint32_t instance; + uint32_t instance, i; DRM_DEBUG("IH: SDMA trap\n"); instance = sdma_v4_4_2_irq_id_to_seq(entry->client_id); + + /* Client id gives the SDMA instance in AID. To know the exact SDMA + * instance, interrupt entry gives the node id which corresponds to the AID instance. + * Match node id with the AID id associated with the SDMA instance. */ + for (i = instance; i < adev->sdma.num_instances; + i += adev->sdma.num_inst_per_aid) { + if (adev->sdma.instance[i].aid_id == + node_id_to_phys_map[entry->node_id]) + break; + } + + if (i >= adev->sdma.num_instances) { + dev_WARN_ONCE( + adev->dev, 1, + "Couldn't find the right sdma instance in trap handler"); + return 0; + } + switch (entry->ring_id) { case 0: - amdgpu_fence_process(&adev->sdma.instance[instance].ring); + amdgpu_fence_process(&adev->sdma.instance[i].ring); break; default: break; @@ -1496,7 +1570,7 @@ static int sdma_v4_4_2_process_ras_data_cb(struct amdgpu_device *adev, * be disabled and the driver should only look for the aggregated * interrupt via sync flood */ - if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX)) + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA)) goto out; instance = sdma_v4_4_2_irq_id_to_seq(entry->client_id); @@ -1535,15 +1609,22 @@ static int sdma_v4_4_2_set_ecc_irq_state(struct amdgpu_device *adev, unsigned type, enum amdgpu_interrupt_state state) { - u32 sdma_edc_config; + u32 sdma_cntl; - sdma_edc_config = RREG32_SDMA(type, regCC_SDMA_EDC_CONFIG); - /* - * FIXME: This was inherited from Aldebaran, but no this field - * definition in the regspec of both Aldebaran and SDMA 4.4.2 - */ - sdma_edc_config |= (state == AMDGPU_IRQ_STATE_ENABLE) ? (1 << 2) : 0; - WREG32_SDMA(type, regCC_SDMA_EDC_CONFIG, sdma_edc_config); + sdma_cntl = RREG32_SDMA(type, regSDMA_CNTL); + switch (state) { + case AMDGPU_IRQ_STATE_DISABLE: + sdma_cntl = REG_SET_FIELD(sdma_cntl, SDMA_CNTL, + DRAM_ECC_INT_ENABLE, 0); + WREG32_SDMA(type, regSDMA_CNTL, sdma_cntl); + break; + /* sdma ecc interrupt is enabled by default + * driver doesn't need to do anything to + * enable the interrupt */ + case AMDGPU_IRQ_STATE_ENABLE: + default: + break; + } return 0; } @@ -1615,19 +1696,49 @@ static int sdma_v4_4_2_process_srbm_write_irq(struct amdgpu_device *adev, return 0; } -static void sdma_v4_4_2_update_medium_grain_clock_gating( - struct amdgpu_device *adev, - bool enable) +static void sdma_v4_4_2_inst_update_medium_grain_light_sleep( + struct amdgpu_device *adev, bool enable, uint32_t inst_mask) { uint32_t data, def; int i; - if (enable && (adev->cg_flags & AMD_CG_SUPPORT_SDMA_MGCG)) { - for (i = 0; i < adev->sdma.num_instances; i++) { + /* leave as default if it is not driver controlled */ + if (!(adev->cg_flags & AMD_CG_SUPPORT_SDMA_LS)) + return; + + if (enable) { + for_each_inst(i, inst_mask) { + /* 1-not override: enable sdma mem light sleep */ + def = data = RREG32_SDMA(i, regSDMA_POWER_CNTL); + data |= SDMA_POWER_CNTL__MEM_POWER_OVERRIDE_MASK; + if (def != data) + WREG32_SDMA(i, regSDMA_POWER_CNTL, data); + } + } else { + for_each_inst(i, inst_mask) { + /* 0-override:disable sdma mem light sleep */ + def = data = RREG32_SDMA(i, regSDMA_POWER_CNTL); + data &= ~SDMA_POWER_CNTL__MEM_POWER_OVERRIDE_MASK; + if (def != data) + WREG32_SDMA(i, regSDMA_POWER_CNTL, data); + } + } +} + +static void sdma_v4_4_2_inst_update_medium_grain_clock_gating( + struct amdgpu_device *adev, bool enable, uint32_t inst_mask) +{ + uint32_t data, def; + int i; + + /* leave as default if it is not driver controlled */ + if (!(adev->cg_flags & AMD_CG_SUPPORT_SDMA_MGCG)) + return; + + if (enable) { + for_each_inst(i, inst_mask) { def = data = RREG32_SDMA(i, regSDMA_CLK_CTRL); - data &= ~(SDMA_CLK_CTRL__SOFT_OVERRIDE7_MASK | - SDMA_CLK_CTRL__SOFT_OVERRIDE6_MASK | - SDMA_CLK_CTRL__SOFT_OVERRIDE5_MASK | + data &= ~(SDMA_CLK_CTRL__SOFT_OVERRIDE5_MASK | SDMA_CLK_CTRL__SOFT_OVERRIDE4_MASK | SDMA_CLK_CTRL__SOFT_OVERRIDE3_MASK | SDMA_CLK_CTRL__SOFT_OVERRIDE2_MASK | @@ -1637,11 +1748,9 @@ static void sdma_v4_4_2_update_medium_grain_clock_gating( WREG32_SDMA(i, regSDMA_CLK_CTRL, data); } } else { - for (i = 0; i < adev->sdma.num_instances; i++) { + for_each_inst(i, inst_mask) { def = data = RREG32_SDMA(i, regSDMA_CLK_CTRL); - data |= (SDMA_CLK_CTRL__SOFT_OVERRIDE7_MASK | - SDMA_CLK_CTRL__SOFT_OVERRIDE6_MASK | - SDMA_CLK_CTRL__SOFT_OVERRIDE5_MASK | + data |= (SDMA_CLK_CTRL__SOFT_OVERRIDE5_MASK | SDMA_CLK_CTRL__SOFT_OVERRIDE4_MASK | SDMA_CLK_CTRL__SOFT_OVERRIDE3_MASK | SDMA_CLK_CTRL__SOFT_OVERRIDE2_MASK | @@ -1653,45 +1762,21 @@ static void sdma_v4_4_2_update_medium_grain_clock_gating( } } - -static void sdma_v4_4_2_update_medium_grain_light_sleep( - struct amdgpu_device *adev, - bool enable) -{ - uint32_t data, def; - int i; - - if (enable && (adev->cg_flags & AMD_CG_SUPPORT_SDMA_LS)) { - for (i = 0; i < adev->sdma.num_instances; i++) { - /* 1-not override: enable sdma mem light sleep */ - def = data = RREG32_SDMA(0, regSDMA_POWER_CNTL); - data |= SDMA_POWER_CNTL__MEM_POWER_OVERRIDE_MASK; - if (def != data) - WREG32_SDMA(0, regSDMA_POWER_CNTL, data); - } - } else { - for (i = 0; i < adev->sdma.num_instances; i++) { - /* 0-override:disable sdma mem light sleep */ - def = data = RREG32_SDMA(0, regSDMA_POWER_CNTL); - data &= ~SDMA_POWER_CNTL__MEM_POWER_OVERRIDE_MASK; - if (def != data) - WREG32_SDMA(0, regSDMA_POWER_CNTL, data); - } - } -} - static int sdma_v4_4_2_set_clockgating_state(void *handle, enum amd_clockgating_state state) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + uint32_t inst_mask; if (amdgpu_sriov_vf(adev)) return 0; - sdma_v4_4_2_update_medium_grain_clock_gating(adev, - state == AMD_CG_STATE_GATE); - sdma_v4_4_2_update_medium_grain_light_sleep(adev, - state == AMD_CG_STATE_GATE); + inst_mask = GENMASK(adev->sdma.num_instances - 1, 0); + + sdma_v4_4_2_inst_update_medium_grain_clock_gating( + adev, state == AMD_CG_STATE_GATE, inst_mask); + sdma_v4_4_2_inst_update_medium_grain_light_sleep( + adev, state == AMD_CG_STATE_GATE, inst_mask); return 0; } @@ -1710,12 +1795,12 @@ static void sdma_v4_4_2_get_clockgating_state(void *handle, u64 *flags) *flags = 0; /* AMD_CG_SUPPORT_SDMA_MGCG */ - data = RREG32(SOC15_REG_OFFSET(SDMA0, 0, regSDMA_CLK_CTRL)); - if (!(data & SDMA_CLK_CTRL__SOFT_OVERRIDE7_MASK)) + data = RREG32(SOC15_REG_OFFSET(SDMA0, GET_INST(SDMA0, 0), regSDMA_CLK_CTRL)); + if (!(data & SDMA_CLK_CTRL__SOFT_OVERRIDE5_MASK)) *flags |= AMD_CG_SUPPORT_SDMA_MGCG; /* AMD_CG_SUPPORT_SDMA_LS */ - data = RREG32(SOC15_REG_OFFSET(SDMA0, 0, regSDMA_POWER_CNTL)); + data = RREG32(SOC15_REG_OFFSET(SDMA0, GET_INST(SDMA0, 0), regSDMA_POWER_CNTL)); if (data & SDMA_POWER_CNTL__MEM_POWER_OVERRIDE_MASK) *flags |= AMD_CG_SUPPORT_SDMA_LS; } @@ -1740,7 +1825,7 @@ const struct amd_ip_funcs sdma_v4_4_2_ip_funcs = { static const struct amdgpu_ring_funcs sdma_v4_4_2_ring_funcs = { .type = AMDGPU_RING_TYPE_SDMA, - .align_mask = 0xf, + .align_mask = 0xff, .nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), .support_64bit_ptrs = true, .get_rptr = sdma_v4_4_2_ring_get_rptr, @@ -1771,7 +1856,7 @@ static const struct amdgpu_ring_funcs sdma_v4_4_2_ring_funcs = { static const struct amdgpu_ring_funcs sdma_v4_4_2_page_ring_funcs = { .type = AMDGPU_RING_TYPE_SDMA, - .align_mask = 0xf, + .align_mask = 0xff, .nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), .support_64bit_ptrs = true, .get_rptr = sdma_v4_4_2_ring_get_rptr, @@ -1802,7 +1887,7 @@ static const struct amdgpu_ring_funcs sdma_v4_4_2_page_ring_funcs = { static void sdma_v4_4_2_set_ring_funcs(struct amdgpu_device *adev) { - int i; + int i, dev_inst; for (i = 0; i < adev->sdma.num_instances; i++) { adev->sdma.instance[i].ring.funcs = &sdma_v4_4_2_ring_funcs; @@ -1812,6 +1897,11 @@ static void sdma_v4_4_2_set_ring_funcs(struct amdgpu_device *adev) &sdma_v4_4_2_page_ring_funcs; adev->sdma.instance[i].page.me = i; } + + dev_inst = GET_INST(SDMA0, i); + /* AID to which SDMA belongs depends on physical instance */ + adev->sdma.instance[i].aid_id = + dev_inst / adev->sdma.num_inst_per_aid; } } @@ -1965,3 +2055,146 @@ const struct amdgpu_ip_block_version sdma_v4_4_2_ip_block = { .rev = 0, .funcs = &sdma_v4_4_2_ip_funcs, }; + +static int sdma_v4_4_2_xcp_resume(void *handle, uint32_t inst_mask) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int r; + + if (!amdgpu_sriov_vf(adev)) + sdma_v4_4_2_inst_init_golden_registers(adev, inst_mask); + + r = sdma_v4_4_2_inst_start(adev, inst_mask); + + return r; +} + +static int sdma_v4_4_2_xcp_suspend(void *handle, uint32_t inst_mask) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + uint32_t tmp_mask = inst_mask; + int i; + + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA)) { + for_each_inst(i, tmp_mask) { + amdgpu_irq_put(adev, &adev->sdma.ecc_irq, + AMDGPU_SDMA_IRQ_INSTANCE0 + i); + } + } + + sdma_v4_4_2_inst_ctx_switch_enable(adev, false, inst_mask); + sdma_v4_4_2_inst_enable(adev, false, inst_mask); + + return 0; +} + +struct amdgpu_xcp_ip_funcs sdma_v4_4_2_xcp_funcs = { + .suspend = &sdma_v4_4_2_xcp_suspend, + .resume = &sdma_v4_4_2_xcp_resume +}; + +static const struct amdgpu_ras_err_status_reg_entry sdma_v4_2_2_ue_reg_list[] = { + {AMDGPU_RAS_REG_ENTRY(SDMA0, 0, regSDMA_UE_ERR_STATUS_LO, regSDMA_UE_ERR_STATUS_HI), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "SDMA"}, +}; + +static const struct amdgpu_ras_memory_id_entry sdma_v4_4_2_ras_memory_list[] = { + {AMDGPU_SDMA_MBANK_DATA_BUF0, "SDMA_MBANK_DATA_BUF0"}, + {AMDGPU_SDMA_MBANK_DATA_BUF1, "SDMA_MBANK_DATA_BUF1"}, + {AMDGPU_SDMA_MBANK_DATA_BUF2, "SDMA_MBANK_DATA_BUF2"}, + {AMDGPU_SDMA_MBANK_DATA_BUF3, "SDMA_MBANK_DATA_BUF3"}, + {AMDGPU_SDMA_MBANK_DATA_BUF4, "SDMA_MBANK_DATA_BUF4"}, + {AMDGPU_SDMA_MBANK_DATA_BUF5, "SDMA_MBANK_DATA_BUF5"}, + {AMDGPU_SDMA_MBANK_DATA_BUF6, "SDMA_MBANK_DATA_BUF6"}, + {AMDGPU_SDMA_MBANK_DATA_BUF7, "SDMA_MBANK_DATA_BUF7"}, + {AMDGPU_SDMA_MBANK_DATA_BUF8, "SDMA_MBANK_DATA_BUF8"}, + {AMDGPU_SDMA_MBANK_DATA_BUF9, "SDMA_MBANK_DATA_BUF9"}, + {AMDGPU_SDMA_MBANK_DATA_BUF10, "SDMA_MBANK_DATA_BUF10"}, + {AMDGPU_SDMA_MBANK_DATA_BUF11, "SDMA_MBANK_DATA_BUF11"}, + {AMDGPU_SDMA_MBANK_DATA_BUF12, "SDMA_MBANK_DATA_BUF12"}, + {AMDGPU_SDMA_MBANK_DATA_BUF13, "SDMA_MBANK_DATA_BUF13"}, + {AMDGPU_SDMA_MBANK_DATA_BUF14, "SDMA_MBANK_DATA_BUF14"}, + {AMDGPU_SDMA_MBANK_DATA_BUF15, "SDMA_MBANK_DATA_BUF15"}, + {AMDGPU_SDMA_UCODE_BUF, "SDMA_UCODE_BUF"}, + {AMDGPU_SDMA_RB_CMD_BUF, "SDMA_RB_CMD_BUF"}, + {AMDGPU_SDMA_IB_CMD_BUF, "SDMA_IB_CMD_BUF"}, + {AMDGPU_SDMA_UTCL1_RD_FIFO, "SDMA_UTCL1_RD_FIFO"}, + {AMDGPU_SDMA_UTCL1_RDBST_FIFO, "SDMA_UTCL1_RDBST_FIFO"}, + {AMDGPU_SDMA_UTCL1_WR_FIFO, "SDMA_UTCL1_WR_FIFO"}, + {AMDGPU_SDMA_DATA_LUT_FIFO, "SDMA_DATA_LUT_FIFO"}, + {AMDGPU_SDMA_SPLIT_DAT_BUF, "SDMA_SPLIT_DAT_BUF"}, +}; + +static void sdma_v4_4_2_inst_query_ras_error_count(struct amdgpu_device *adev, + uint32_t sdma_inst, + void *ras_err_status) +{ + struct ras_err_data *err_data = (struct ras_err_data *)ras_err_status; + uint32_t sdma_dev_inst = GET_INST(SDMA0, sdma_inst); + + /* sdma v4_4_2 doesn't support query ce counts */ + amdgpu_ras_inst_query_ras_error_count(adev, + sdma_v4_2_2_ue_reg_list, + ARRAY_SIZE(sdma_v4_2_2_ue_reg_list), + sdma_v4_4_2_ras_memory_list, + ARRAY_SIZE(sdma_v4_4_2_ras_memory_list), + sdma_dev_inst, + AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE, + &err_data->ue_count); +} + +static void sdma_v4_4_2_query_ras_error_count(struct amdgpu_device *adev, + void *ras_err_status) +{ + uint32_t inst_mask; + int i = 0; + + inst_mask = GENMASK(adev->sdma.num_instances - 1, 0); + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA)) { + for_each_inst(i, inst_mask) + sdma_v4_4_2_inst_query_ras_error_count(adev, i, ras_err_status); + } else { + dev_warn(adev->dev, "SDMA RAS is not supported\n"); + } +} + +static void sdma_v4_4_2_inst_reset_ras_error_count(struct amdgpu_device *adev, + uint32_t sdma_inst) +{ + uint32_t sdma_dev_inst = GET_INST(SDMA0, sdma_inst); + + amdgpu_ras_inst_reset_ras_error_count(adev, + sdma_v4_2_2_ue_reg_list, + ARRAY_SIZE(sdma_v4_2_2_ue_reg_list), + sdma_dev_inst); +} + +static void sdma_v4_4_2_reset_ras_error_count(struct amdgpu_device *adev) +{ + uint32_t inst_mask; + int i = 0; + + inst_mask = GENMASK(adev->sdma.num_instances - 1, 0); + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__SDMA)) { + for_each_inst(i, inst_mask) + sdma_v4_4_2_inst_reset_ras_error_count(adev, i); + } else { + dev_warn(adev->dev, "SDMA RAS is not supported\n"); + } +} + +static const struct amdgpu_ras_block_hw_ops sdma_v4_4_2_ras_hw_ops = { + .query_ras_error_count = sdma_v4_4_2_query_ras_error_count, + .reset_ras_error_count = sdma_v4_4_2_reset_ras_error_count, +}; + +static struct amdgpu_sdma_ras sdma_v4_4_2_ras = { + .ras_block = { + .hw_ops = &sdma_v4_4_2_ras_hw_ops, + }, +}; + +static void sdma_v4_4_2_set_ras_funcs(struct amdgpu_device *adev) +{ + adev->sdma.ras = &sdma_v4_4_2_ras; +} diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.h b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.h index 4814e8a074d615e8bfbe1794b98f2ec8be2d0d06..d516145529bb8dcd4874a761e26595f7290cb853 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.h +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.h @@ -27,4 +27,6 @@ extern const struct amd_ip_funcs sdma_v4_4_2_ip_funcs; extern const struct amdgpu_ip_block_version sdma_v4_4_2_ip_block; +extern struct amdgpu_xcp_ip_funcs sdma_v4_4_2_xcp_funcs; + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c index 92e1299be021950146fd99bf530644b5b063a3a1..5c4d4df9cf94ccb075c605ddf85ac714bd8df397 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c @@ -819,8 +819,6 @@ static int sdma_v5_0_gfx_resume(struct amdgpu_device *adev) /* enable DMA IBs */ WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl); - ring->sched.ready = true; - if (amdgpu_sriov_vf(adev)) { /* bare-metal sequence doesn't need below to lines */ sdma_v5_0_ctx_switch_enable(adev, true); sdma_v5_0_enable(adev, true); @@ -1389,7 +1387,7 @@ static int sdma_v5_0_sw_init(void *handle) (adev->doorbell_index.sdma_engine[0] << 1) //get DWORD offset : (adev->doorbell_index.sdma_engine[1] << 1); // get DWORD offset - ring->vm_hub = AMDGPU_GFXHUB_0; + ring->vm_hub = AMDGPU_GFXHUB(0); sprintf(ring->name, "sdma%d", i); r = amdgpu_ring_init(adev, ring, 1024, &adev->sdma.trap_irq, (i == 0) ? AMDGPU_SDMA_IRQ_INSTANCE0 : diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c index ca7e8757d78e5eba40acf7c9522d556774ea18e1..a7b230e5a26d6f988a2dc2c5cf5f6abdfa406e77 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c @@ -617,18 +617,14 @@ static int sdma_v5_2_gfx_resume(struct amdgpu_device *adev) /* enable DMA IBs */ WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl); - ring->sched.ready = true; - if (amdgpu_sriov_vf(adev)) { /* bare-metal sequence doesn't need below to lines */ sdma_v5_2_ctx_switch_enable(adev, true); sdma_v5_2_enable(adev, true); } - r = amdgpu_ring_test_ring(ring); - if (r) { - ring->sched.ready = false; + r = amdgpu_ring_test_helper(ring); + if (r) return r; - } if (adev->mman.buffer_funcs_ring == ring) amdgpu_ttm_set_buffer_funcs_status(adev, true); @@ -1253,7 +1249,7 @@ static int sdma_v5_2_sw_init(void *handle) ring->doorbell_index = (adev->doorbell_index.sdma_engine[i] << 1); //get DWORD offset - ring->vm_hub = AMDGPU_GFXHUB_0; + ring->vm_hub = AMDGPU_GFXHUB(0); sprintf(ring->name, "sdma%d", i); r = amdgpu_ring_init(adev, ring, 1024, &adev->sdma.trap_irq, AMDGPU_SDMA_IRQ_INSTANCE0 + i, diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c index 3d9a80511a45083bca937e7d8752be72b4d54371..3b03dda854fdc178d8ab954173e5791f486bdbdf 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c @@ -238,6 +238,8 @@ static void sdma_v6_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count) * * @ring: amdgpu ring pointer * @ib: IB object to schedule + * @flags: unused + * @job: job to retrieve vmid from * * Schedule an IB in the DMA ring. */ @@ -585,16 +587,12 @@ static int sdma_v6_0_gfx_resume(struct amdgpu_device *adev) /* enable DMA IBs */ WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_IB_CNTL), ib_cntl); - ring->sched.ready = true; - if (amdgpu_sriov_vf(adev)) sdma_v6_0_enable(adev, true); r = amdgpu_ring_test_helper(ring); - if (r) { - ring->sched.ready = false; + if (r) return r; - } if (adev->mman.buffer_funcs_ring == ring) amdgpu_ttm_set_buffer_funcs_status(adev, true); @@ -942,6 +940,7 @@ static int sdma_v6_0_ring_test_ring(struct amdgpu_ring *ring) * sdma_v6_0_ring_test_ib - test an IB on the DMA engine * * @ring: amdgpu_ring structure holding ring information + * @timeout: timeout value in jiffies, or MAX_SCHEDULE_TIMEOUT * * Test a simple IB in the DMA ring. * Returns 0 on success, error on failure. @@ -1122,6 +1121,7 @@ static void sdma_v6_0_vm_set_pte_pde(struct amdgpu_ib *ib, /** * sdma_v6_0_ring_pad_ib - pad the IB * @ib: indirect buffer to fill with padding + * @ring: amdgpu ring pointer * * Pad the IB with NOPs to a boundary multiple of 8. */ @@ -1171,6 +1171,8 @@ static void sdma_v6_0_ring_emit_pipeline_sync(struct amdgpu_ring *ring) * sdma_v6_0_ring_emit_vm_flush - vm flush using sDMA * * @ring: amdgpu_ring pointer + * @vmid: vmid number to use + * @pd_addr: address * * Update the page table base and flush the VM TLB * using sDMA. @@ -1298,7 +1300,7 @@ static int sdma_v6_0_sw_init(void *handle) ring->doorbell_index = (adev->doorbell_index.sdma_engine[i] << 1); // get DWORD offset - ring->vm_hub = AMDGPU_GFXHUB_0; + ring->vm_hub = AMDGPU_GFXHUB(0); sprintf(ring->name, "sdma%d", i); r = amdgpu_ring_init(adev, ring, 1024, &adev->sdma.trap_irq, diff --git a/drivers/gpu/drm/amd/amdgpu/si.c b/drivers/gpu/drm/amd/amdgpu/si.c index 7f99e130acd0663eec48e3e6149468fd2b833c45..f64b87b11b1b5aa7be6c5d7cad97b004e86b0ffc 100644 --- a/drivers/gpu/drm/amd/amdgpu/si.c +++ b/drivers/gpu/drm/amd/amdgpu/si.c @@ -1181,12 +1181,12 @@ static uint32_t si_get_register_value(struct amdgpu_device *adev, mutex_lock(&adev->grbm_idx_mutex); if (se_num != 0xffffffff || sh_num != 0xffffffff) - amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff, 0); val = RREG32(reg_offset); if (se_num != 0xffffffff || sh_num != 0xffffffff) - amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); return val; } else { diff --git a/drivers/gpu/drm/amd/amdgpu/si_dma.c b/drivers/gpu/drm/amd/amdgpu/si_dma.c index abca8b529721ebcdcbe9aae406590081f95b22bb..42c4547f32ec9cc9f3eba7582f8f7a953a84897b 100644 --- a/drivers/gpu/drm/amd/amdgpu/si_dma.c +++ b/drivers/gpu/drm/amd/amdgpu/si_dma.c @@ -174,8 +174,6 @@ static int si_dma_start(struct amdgpu_device *adev) WREG32(DMA_RB_WPTR + sdma_offsets[i], ring->wptr << 2); WREG32(DMA_RB_CNTL + sdma_offsets[i], rb_cntl | DMA_RB_ENABLE); - ring->sched.ready = true; - r = amdgpu_ring_test_helper(ring); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.c b/drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.c new file mode 100644 index 0000000000000000000000000000000000000000..4368a5891eebd24ae824fa55b627a85d8fddb080 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.c @@ -0,0 +1,103 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +#include "amdgpu.h" +#include "smuio_v13_0_3.h" +#include "soc15_common.h" +#include "smuio/smuio_13_0_3_offset.h" +#include "smuio/smuio_13_0_3_sh_mask.h" + +#define PKG_TYPE_MASK 0x00000003L + +/** + * smuio_v13_0_3_get_die_id - query die id from FCH. + * + * @adev: amdgpu device pointer + * + * Returns die id + */ +static u32 smuio_v13_0_3_get_die_id(struct amdgpu_device *adev) +{ + u32 data, die_id; + + data = RREG32_SOC15(SMUIO, 0, regSMUIO_MCM_CONFIG); + die_id = REG_GET_FIELD(data, SMUIO_MCM_CONFIG, DIE_ID); + + return die_id; +} + +/** + * smuio_v13_0_3_get_socket_id - query socket id from FCH + * + * @adev: amdgpu device pointer + * + * Returns socket id + */ +static u32 smuio_v13_0_3_get_socket_id(struct amdgpu_device *adev) +{ + u32 data, socket_id; + + data = RREG32_SOC15(SMUIO, 0, regSMUIO_MCM_CONFIG); + socket_id = REG_GET_FIELD(data, SMUIO_MCM_CONFIG, SOCKET_ID); + + return socket_id; +} + +/** + * smuio_v13_0_3_get_pkg_type - query package type set by MP1/bootcode + * + * @adev: amdgpu device pointer + * + * Returns package type + */ + +static enum amdgpu_pkg_type smuio_v13_0_3_get_pkg_type(struct amdgpu_device *adev) +{ + enum amdgpu_pkg_type pkg_type; + u32 data; + + data = RREG32_SOC15(SMUIO, 0, regSMUIO_MCM_CONFIG); + data = REG_GET_FIELD(data, SMUIO_MCM_CONFIG, PKG_TYPE); + /* pkg_type[4:0] + * + * bit 1 == 1 APU form factor + * + * b0100 - b1111 - Reserved + */ + switch (data & PKG_TYPE_MASK) { + case 0x2: + pkg_type = AMDGPU_PKG_TYPE_APU; + break; + default: + pkg_type = AMDGPU_PKG_TYPE_UNKNOWN; + break; + } + + return pkg_type; +} + + +const struct amdgpu_smuio_funcs smuio_v13_0_3_funcs = { + .get_die_id = smuio_v13_0_3_get_die_id, + .get_socket_id = smuio_v13_0_3_get_socket_id, + .get_pkg_type = smuio_v13_0_3_get_pkg_type, +}; diff --git a/drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.h b/drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.h new file mode 100644 index 0000000000000000000000000000000000000000..795f66c5a58b8fdf9da2435190e6f6a90623518d --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/smuio_v13_0_3.h @@ -0,0 +1,30 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +#ifndef __SMUIO_V13_0_3_H__ +#define __SMUIO_V13_0_3_H__ + +#include "soc15_common.h" + +extern const struct amdgpu_smuio_funcs smuio_v13_0_3_funcs; + +#endif /* __SMUIO_V13_0_3_H__ */ diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index 6d15d5cd9e07f64cac8c6961f65b68e55ad86409..afcaeadda4c74545171b15597c601fb76506059d 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -153,6 +153,24 @@ static const struct amdgpu_video_codecs rn_video_codecs_decode = .codec_array = rn_video_codecs_decode_array, }; +static const struct amdgpu_video_codec_info vcn_4_0_3_video_codecs_decode_array[] = { + {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4_AVC, 4096, 4096, 52)}, + {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_HEVC, 8192, 4352, 186)}, + {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_JPEG, 4096, 4096, 0)}, + {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_VP9, 8192, 4352, 0)}, + {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_AV1, 8192, 4352, 0)}, +}; + +static const struct amdgpu_video_codecs vcn_4_0_3_video_codecs_decode = { + .codec_count = ARRAY_SIZE(vcn_4_0_3_video_codecs_decode_array), + .codec_array = vcn_4_0_3_video_codecs_decode_array, +}; + +static const struct amdgpu_video_codecs vcn_4_0_3_video_codecs_encode = { + .codec_count = 0, + .codec_array = NULL, +}; + static int soc15_query_video_codecs(struct amdgpu_device *adev, bool encode, const struct amdgpu_video_codecs **codecs) { @@ -185,6 +203,12 @@ static int soc15_query_video_codecs(struct amdgpu_device *adev, bool encode, else *codecs = &rn_video_codecs_decode; return 0; + case IP_VERSION(4, 0, 3): + if (encode) + *codecs = &vcn_4_0_3_video_codecs_encode; + else + *codecs = &vcn_4_0_3_video_codecs_decode; + return 0; default: return -EINVAL; } @@ -301,17 +325,18 @@ static u32 soc15_get_xclk(struct amdgpu_device *adev) u32 reference_clock = adev->clock.spll.reference_freq; if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(12, 0, 0) || - adev->ip_versions[MP1_HWIP][0] == IP_VERSION(12, 0, 1) || - adev->ip_versions[MP1_HWIP][0] == IP_VERSION(10, 0, 0) || - adev->ip_versions[MP1_HWIP][0] == IP_VERSION(10, 0, 1)) + adev->ip_versions[MP1_HWIP][0] == IP_VERSION(12, 0, 1)) return 10000; + if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(10, 0, 0) || + adev->ip_versions[MP1_HWIP][0] == IP_VERSION(10, 0, 1)) + return reference_clock / 4; return reference_clock; } void soc15_grbm_select(struct amdgpu_device *adev, - u32 me, u32 pipe, u32 queue, u32 vmid) + u32 me, u32 pipe, u32 queue, u32 vmid, int xcc_id) { u32 grbm_gfx_cntl = 0; grbm_gfx_cntl = REG_SET_FIELD(grbm_gfx_cntl, GRBM_GFX_CNTL, PIPEID, pipe); @@ -319,12 +344,7 @@ void soc15_grbm_select(struct amdgpu_device *adev, grbm_gfx_cntl = REG_SET_FIELD(grbm_gfx_cntl, GRBM_GFX_CNTL, VMID, vmid); grbm_gfx_cntl = REG_SET_FIELD(grbm_gfx_cntl, GRBM_GFX_CNTL, QUEUEID, queue); - WREG32_SOC15_RLC_SHADOW(GC, 0, mmGRBM_GFX_CNTL, grbm_gfx_cntl); -} - -static void soc15_vga_set_state(struct amdgpu_device *adev, bool state) -{ - /* todo */ + WREG32_SOC15_RLC_SHADOW(GC, xcc_id, mmGRBM_GFX_CNTL, grbm_gfx_cntl); } static bool soc15_read_disabled_bios(struct amdgpu_device *adev) @@ -363,12 +383,12 @@ static uint32_t soc15_read_indexed_register(struct amdgpu_device *adev, u32 se_n mutex_lock(&adev->grbm_idx_mutex); if (se_num != 0xffffffff || sh_num != 0xffffffff) - amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff, 0); val = RREG32(reg_offset); if (se_num != 0xffffffff || sh_num != 0xffffffff) - amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); return val; } @@ -532,6 +552,15 @@ soc15_asic_reset_method(struct amdgpu_device *adev) if (connected_to_cpu) return AMD_RESET_METHOD_MODE2; break; + case IP_VERSION(13, 0, 6): + /* Use gpu_recovery param to target a reset method. + * Enable triggering of GPU reset only if specified + * by module parameter. + */ + if (amdgpu_gpu_recovery == 4 || amdgpu_gpu_recovery == 5) + return AMD_RESET_METHOD_MODE2; + else + return AMD_RESET_METHOD_NONE; default: break; } @@ -816,7 +845,6 @@ static const struct amdgpu_asic_funcs soc15_asic_funcs = .read_register = &soc15_read_register, .reset = &soc15_asic_reset, .reset_method = &soc15_asic_reset_method, - .set_vga_state = &soc15_vga_set_state, .get_xclk = &soc15_get_xclk, .set_uvd_clocks = &soc15_set_uvd_clocks, .set_vce_clocks = &soc15_set_vce_clocks, @@ -838,7 +866,6 @@ static const struct amdgpu_asic_funcs vega20_asic_funcs = .read_register = &soc15_read_register, .reset = &soc15_asic_reset, .reset_method = &soc15_asic_reset_method, - .set_vga_state = &soc15_vga_set_state, .get_xclk = &soc15_get_xclk, .set_uvd_clocks = &soc15_set_uvd_clocks, .set_vce_clocks = &soc15_set_vce_clocks, @@ -853,6 +880,28 @@ static const struct amdgpu_asic_funcs vega20_asic_funcs = .query_video_codecs = &soc15_query_video_codecs, }; +static const struct amdgpu_asic_funcs aqua_vanjaram_asic_funcs = +{ + .read_disabled_bios = &soc15_read_disabled_bios, + .read_bios_from_rom = &amdgpu_soc15_read_bios_from_rom, + .read_register = &soc15_read_register, + .reset = &soc15_asic_reset, + .reset_method = &soc15_asic_reset_method, + .get_xclk = &soc15_get_xclk, + .set_uvd_clocks = &soc15_set_uvd_clocks, + .set_vce_clocks = &soc15_set_vce_clocks, + .get_config_memsize = &soc15_get_config_memsize, + .need_full_reset = &soc15_need_full_reset, + .init_doorbell_index = &aqua_vanjaram_doorbell_index_init, + .get_pcie_usage = &vega20_get_pcie_usage, + .need_reset_on_init = &soc15_need_reset_on_init, + .get_pcie_replay_count = &soc15_get_pcie_replay_count, + .supports_baco = &soc15_supports_baco, + .pre_asic_init = &soc15_pre_asic_init, + .query_video_codecs = &soc15_query_video_codecs, + .encode_ext_smn_addressing = &aqua_vanjaram_encode_ext_smn_addressing, +}; + static int soc15_common_early_init(void *handle) { #define MMIO_REG_HOLE_OFFSET (0x80000 - PAGE_SIZE) @@ -866,6 +915,8 @@ static int soc15_common_early_init(void *handle) adev->smc_wreg = NULL; adev->pcie_rreg = &amdgpu_device_indirect_rreg; adev->pcie_wreg = &amdgpu_device_indirect_wreg; + adev->pcie_rreg_ext = &amdgpu_device_indirect_rreg_ext; + adev->pcie_wreg_ext = &amdgpu_device_indirect_wreg_ext; adev->pcie_rreg64 = &amdgpu_device_indirect_rreg64; adev->pcie_wreg64 = &amdgpu_device_indirect_wreg64; adev->uvd_ctx_rreg = &soc15_uvd_ctx_rreg; @@ -1094,9 +1145,18 @@ static int soc15_common_early_init(void *handle) adev->external_rev_id = adev->rev_id + 0x3c; break; case IP_VERSION(9, 4, 3): - adev->asic_funcs = &vega20_asic_funcs; - adev->cg_flags = 0; - adev->pg_flags = 0; + adev->asic_funcs = &aqua_vanjaram_asic_funcs; + adev->cg_flags = + AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_CGCG | + AMD_CG_SUPPORT_GFX_CGLS | AMD_CG_SUPPORT_SDMA_MGCG | + AMD_CG_SUPPORT_GFX_FGCG | AMD_CG_SUPPORT_REPEATER_FGCG | + AMD_CG_SUPPORT_VCN_MGCG | AMD_CG_SUPPORT_JPEG_MGCG | + AMD_CG_SUPPORT_IH_CG; + adev->pg_flags = + AMD_PG_SUPPORT_VCN | + AMD_PG_SUPPORT_VCN_DPG | + AMD_PG_SUPPORT_JPEG; + adev->external_rev_id = adev->rev_id + 0x46; break; default: /* FIXME: not supported yet */ diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.h b/drivers/gpu/drm/amd/amdgpu/soc15.h index efc2a253e8dbe19b093791dff706c8c99e6fad2a..eac54042c6c0e424629cf6c2c3cbbe8824ad3246 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.h +++ b/drivers/gpu/drm/amd/amdgpu/soc15.h @@ -100,7 +100,7 @@ struct soc15_ras_field_entry { #define SOC15_RAS_REG_FIELD_VAL(val, entry, field) SOC15_REG_FIELD_VAL((val), (entry).field##_count_mask, (entry).field##_count_shift) void soc15_grbm_select(struct amdgpu_device *adev, - u32 me, u32 pipe, u32 queue, u32 vmid); + u32 me, u32 pipe, u32 queue, u32 vmid, int xcc_id); void soc15_set_virt_ops(struct amdgpu_device *adev); void soc15_program_register_sequence(struct amdgpu_device *adev, @@ -111,7 +111,11 @@ int vega10_reg_base_init(struct amdgpu_device *adev); int vega20_reg_base_init(struct amdgpu_device *adev); int arct_reg_base_init(struct amdgpu_device *adev); int aldebaran_reg_base_init(struct amdgpu_device *adev); +void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev); +u64 aqua_vanjaram_encode_ext_smn_addressing(int ext_id); +int aqua_vanjaram_init_soc_config(struct amdgpu_device *adev); void vega10_doorbell_index_init(struct amdgpu_device *adev); void vega20_doorbell_index_init(struct amdgpu_device *adev); +void aqua_vanjaram_doorbell_index_init(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h b/drivers/gpu/drm/amd/amdgpu/soc15_common.h index 9fefd403e14fb11775acb55e67092c3405756258..96948a59f8dd556455d1f6ab0bd484285852b6e7 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15_common.h +++ b/drivers/gpu/drm/amd/amdgpu/soc15_common.h @@ -24,8 +24,18 @@ #ifndef __SOC15_COMMON_H__ #define __SOC15_COMMON_H__ +/* GET_INST returns the physical instance corresponding to a logical instance */ +#define GET_INST(ip, inst) \ + (adev->ip_map.logical_to_dev_inst ? \ + adev->ip_map.logical_to_dev_inst(adev, ip##_HWIP, inst) : inst) +#define GET_MASK(ip, mask) \ + (adev->ip_map.logical_to_dev_mask ? \ + adev->ip_map.logical_to_dev_mask(adev, ip##_HWIP, mask) : mask) + /* Register Access Macros */ #define SOC15_REG_OFFSET(ip, inst, reg) (adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) +#define SOC15_REG_OFFSET1(ip, inst, reg, offset) \ + (adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + (reg)+(offset)) #define __WREG32_SOC15_RLC__(reg, value, flag, hwip) \ ((amdgpu_sriov_vf(adev) && adev->gfx.rlc.funcs && adev->gfx.rlc.rlcg_reg_access_supported) ? \ @@ -66,7 +76,8 @@ AMDGPU_REGS_NO_KIQ, ip##_HWIP) #define RREG32_SOC15_OFFSET(ip, inst, reg, offset) \ - __RREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) + offset, 0, ip##_HWIP) + __RREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + (reg)) + \ + (offset), 0, ip##_HWIP) #define WREG32_SOC15(ip, inst, reg, value) \ __WREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg), \ @@ -86,31 +97,15 @@ __WREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) + offset, \ value, 0, ip##_HWIP) -#define SOC15_WAIT_ON_RREG(ip, inst, reg, expected_value, mask) \ -({ int ret = 0; \ - do { \ - uint32_t old_ = 0; \ - uint32_t tmp_ = RREG32(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg); \ - uint32_t loop = adev->usec_timeout; \ - ret = 0; \ - while ((tmp_ & (mask)) != (expected_value)) { \ - if (old_ != tmp_) { \ - loop = adev->usec_timeout; \ - old_ = tmp_; \ - } else \ - udelay(1); \ - tmp_ = RREG32(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg); \ - loop--; \ - if (!loop) { \ - DRM_WARN("Register(%d) [%s] failed to reach value 0x%08x != 0x%08x\n", \ - inst, #reg, (unsigned)expected_value, (unsigned)(tmp_ & (mask))); \ - ret = -ETIMEDOUT; \ - break; \ - } \ - } \ - } while (0); \ - ret; \ -}) +#define SOC15_WAIT_ON_RREG(ip, inst, reg, expected_value, mask) \ + amdgpu_device_wait_on_rreg(adev, inst, \ + (adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + (reg)), \ + #reg, expected_value, mask) + +#define SOC15_WAIT_ON_RREG_OFFSET(ip, inst, reg, offset, expected_value, mask) \ + amdgpu_device_wait_on_rreg(adev, inst, \ + (adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + (reg) + (offset)), \ + #reg, expected_value, mask) #define WREG32_RLC(reg, value) \ __WREG32_SOC15_RLC__(reg, value, AMDGPU_REGS_RLC, GC_HWIP) @@ -157,10 +152,10 @@ do { \ uint32_t target_reg = adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg;\ if (amdgpu_sriov_fullaccess(adev)) { \ - uint32_t r2 = adev->reg_offset[GC_HWIP][0][prefix##SCRATCH_REG1_BASE_IDX] + prefix##SCRATCH_REG2; \ - uint32_t r3 = adev->reg_offset[GC_HWIP][0][prefix##SCRATCH_REG1_BASE_IDX] + prefix##SCRATCH_REG3; \ - uint32_t grbm_cntl = adev->reg_offset[GC_HWIP][0][prefix##GRBM_GFX_CNTL_BASE_IDX] + prefix##GRBM_GFX_CNTL; \ - uint32_t grbm_idx = adev->reg_offset[GC_HWIP][0][prefix##GRBM_GFX_INDEX_BASE_IDX] + prefix##GRBM_GFX_INDEX; \ + uint32_t r2 = adev->reg_offset[GC_HWIP][inst][prefix##SCRATCH_REG1_BASE_IDX] + prefix##SCRATCH_REG2; \ + uint32_t r3 = adev->reg_offset[GC_HWIP][inst][prefix##SCRATCH_REG1_BASE_IDX] + prefix##SCRATCH_REG3; \ + uint32_t grbm_cntl = adev->reg_offset[GC_HWIP][inst][prefix##GRBM_GFX_CNTL_BASE_IDX] + prefix##GRBM_GFX_CNTL; \ + uint32_t grbm_idx = adev->reg_offset[GC_HWIP][inst][prefix##GRBM_GFX_INDEX_BASE_IDX] + prefix##GRBM_GFX_INDEX; \ if (target_reg == grbm_cntl) \ WREG32(r2, value); \ else if (target_reg == grbm_idx) \ @@ -176,13 +171,13 @@ #define WREG32_SOC15_RLC(ip, inst, reg, value) \ do { \ - uint32_t target_reg = adev->reg_offset[ip##_HWIP][0][reg##_BASE_IDX] + reg;\ + uint32_t target_reg = adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg;\ __WREG32_SOC15_RLC__(target_reg, value, AMDGPU_REGS_RLC, ip##_HWIP); \ } while (0) #define WREG32_SOC15_RLC_EX(prefix, ip, inst, reg, value) \ do { \ - uint32_t target_reg = adev->reg_offset[GC_HWIP][0][reg##_BASE_IDX] + reg;\ + uint32_t target_reg = adev->reg_offset[GC_HWIP][inst][reg##_BASE_IDX] + reg;\ WREG32_RLC_EX(prefix, target_reg, value); \ } while (0) @@ -199,4 +194,14 @@ #define RREG32_SOC15_OFFSET_RLC(ip, inst, reg, offset) \ __RREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) + offset, AMDGPU_REGS_RLC, ip##_HWIP) +/* inst equals to ext for some IPs */ +#define RREG32_SOC15_EXT(ip, inst, reg, ext) \ + RREG32_PCIE_EXT((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) * 4 \ + + adev->asic_funcs->encode_ext_smn_addressing(ext)) \ + +#define WREG32_SOC15_EXT(ip, inst, reg, ext, value) \ + WREG32_PCIE_EXT((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) * 4 \ + + adev->asic_funcs->encode_ext_smn_addressing(ext), \ + value) \ + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/soc21.c b/drivers/gpu/drm/amd/amdgpu/soc21.c index d7716253651440186ce6c0c95df74361505209a2..e5e5d68a4d702a2e2b9df53ecb8ad141abda254f 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc21.c +++ b/drivers/gpu/drm/amd/amdgpu/soc21.c @@ -248,11 +248,6 @@ void soc21_grbm_select(struct amdgpu_device *adev, WREG32_SOC15(GC, 0, regGRBM_GFX_CNTL, grbm_gfx_cntl); } -static void soc21_vga_set_state(struct amdgpu_device *adev, bool state) -{ - /* todo */ -} - static bool soc21_read_disabled_bios(struct amdgpu_device *adev) { /* todo */ @@ -288,12 +283,12 @@ static uint32_t soc21_read_indexed_register(struct amdgpu_device *adev, u32 se_n mutex_lock(&adev->grbm_idx_mutex); if (se_num != 0xffffffff || sh_num != 0xffffffff) - amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff, 0); val = RREG32(reg_offset); if (se_num != 0xffffffff || sh_num != 0xffffffff) - amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); return val; } @@ -542,9 +537,9 @@ static int soc21_update_umd_stable_pstate(struct amdgpu_device *adev, bool enter) { if (enter) - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); else - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); if (adev->gfx.funcs->update_perfmon_mgcg) adev->gfx.funcs->update_perfmon_mgcg(adev, !enter); @@ -559,7 +554,6 @@ static const struct amdgpu_asic_funcs soc21_asic_funcs = .read_register = &soc21_read_register, .reset = &soc21_asic_reset, .reset_method = &soc21_asic_reset_method, - .set_vga_state = &soc21_vga_set_state, .get_xclk = &soc21_get_xclk, .set_uvd_clocks = &soc21_set_uvd_clocks, .set_vce_clocks = &soc21_set_vce_clocks, diff --git a/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h b/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h index 30d0482ac466ccbdf4dd9658cd74da24e388be5b..879bb7af297c7bda447b199a2165a03b4c691562 100644 --- a/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h +++ b/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h @@ -129,6 +129,8 @@ struct ta_ras_trigger_error_input { struct ta_ras_init_flags { uint8_t poison_mode_en; uint8_t dgpu_mode; + uint16_t xcc_mask; + uint8_t channel_dis_num; }; struct ta_ras_output_flags { diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c b/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c index d51ae0bc36f797f899fb6b05c31070a1d3b97672..46bfdee79bfd2aeda2d1dbbcbb55b81dc4a04e85 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c +++ b/drivers/gpu/drm/amd/amdgpu/umc_v8_10.c @@ -444,6 +444,11 @@ static void umc_v8_10_ecc_info_query_ras_error_address(struct amdgpu_device *ade umc_v8_10_ecc_info_query_error_address, ras_error_status); } +static void umc_v8_10_set_eeprom_table_version(struct amdgpu_ras_eeprom_table_header *hdr) +{ + hdr->version = RAS_TABLE_VER_V2_1; +} + const struct amdgpu_ras_block_hw_ops umc_v8_10_ras_hw_ops = { .query_ras_error_count = umc_v8_10_query_ras_error_count, .query_ras_error_address = umc_v8_10_query_ras_error_address, @@ -457,4 +462,5 @@ struct amdgpu_umc_ras umc_v8_10_ras = { .query_ras_poison_mode = umc_v8_10_query_ras_poison_mode, .ecc_info_query_ras_error_count = umc_v8_10_ecc_info_query_ras_error_count, .ecc_info_query_ras_error_address = umc_v8_10_ecc_info_query_ras_error_address, + .set_eeprom_table_version = umc_v8_10_set_eeprom_table_version, }; diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v8_10.h b/drivers/gpu/drm/amd/amdgpu/umc_v8_10.h index c6dfd433fec7bc4f85338f8bffbb19f6b5fd4e82..dc12e0af5451e9d728109ea3d878297ec39eb7f9 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v8_10.h +++ b/drivers/gpu/drm/amd/amdgpu/umc_v8_10.h @@ -33,7 +33,8 @@ /* Total channel instances for all available umc nodes */ #define UMC_V8_10_TOTAL_CHANNEL_NUM(adev) \ - (UMC_V8_10_CHANNEL_INSTANCE_NUM * UMC_V8_10_UMC_INSTANCE_NUM * (adev)->gmc.num_umc) + (UMC_V8_10_CHANNEL_INSTANCE_NUM * UMC_V8_10_UMC_INSTANCE_NUM * \ + (adev)->gmc.num_umc - hweight32((adev)->gmc.m_half_use) * 2) /* UMC regiser per channel offset */ #define UMC_V8_10_PER_CHANNEL_OFFSET 0x400 diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c index e32b656b3dab8ea280109bf16207fff040a70a78..abaa4463e906c988b2a252f23582d58f8833edb9 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c @@ -444,7 +444,7 @@ static int uvd_v7_0_sw_init(void *handle) continue; if (!amdgpu_sriov_vf(adev)) { ring = &adev->uvd.inst[j].ring; - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); sprintf(ring->name, "uvd_%d", ring->me); r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst[j].irq, 0, @@ -455,7 +455,7 @@ static int uvd_v7_0_sw_init(void *handle) for (i = 0; i < adev->uvd.num_enc_rings; ++i) { ring = &adev->uvd.inst[j].ring_enc[i]; - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); sprintf(ring->name, "uvd_enc_%d.%d", ring->me, i); if (amdgpu_sriov_vf(adev)) { ring->use_doorbell = true; diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c index 57b85bb6a1e491f37ef40cecf23b00f7f8b337bd..e0b70cd3b697c53de7128f5bc28f5924e8848ba7 100644 --- a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c @@ -466,7 +466,7 @@ static int vce_v4_0_sw_init(void *handle) enum amdgpu_ring_priority_level hw_prio = amdgpu_vce_get_ring_prio(i); ring = &adev->vce.ring[i]; - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); sprintf(ring->name, "vce%d", i); if (amdgpu_sriov_vf(adev)) { /* DOORBELL only works under SRIOV */ diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c index 761c28fa6ec14cd5294fe0b5a239f8b879086b22..16feb491adf5de18f28b6dab2e056d61287040df 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c @@ -120,7 +120,7 @@ static int vcn_v1_0_sw_init(void *handle) return r; ring = &adev->vcn.inst->ring_dec; - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); sprintf(ring->name, "vcn_dec"); r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0, AMDGPU_RING_PRIO_DEFAULT, NULL); @@ -142,7 +142,7 @@ static int vcn_v1_0_sw_init(void *handle) enum amdgpu_ring_priority_level hw_prio = amdgpu_vcn_get_enc_ring_prio(i); ring = &adev->vcn.inst->ring_enc[i]; - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); sprintf(ring->name, "vcn_enc%d", i); r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0, hw_prio, NULL); @@ -211,7 +211,7 @@ static int vcn_v1_0_hw_init(void *handle) goto done; } - ring = &adev->jpeg.inst->ring_dec; + ring = adev->jpeg.inst->ring_dec; r = amdgpu_ring_test_helper(ring); if (r) goto done; @@ -1304,7 +1304,7 @@ static int vcn_v1_0_pause_dpg_mode(struct amdgpu_device *adev, UVD_DPG_PAUSE__JPEG_PAUSE_DPG_ACK_MASK); /* Restore */ - ring = &adev->jpeg.inst->ring_dec; + ring = adev->jpeg.inst->ring_dec; WREG32_SOC15(UVD, 0, mmUVD_LMI_JRBC_RB_VMID, 0); WREG32_SOC15(UVD, 0, mmUVD_JRBC_RB_CNTL, UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK | @@ -1802,7 +1802,7 @@ static void vcn_v1_0_idle_work_handler(struct work_struct *work) else new_state.fw_based = VCN_DPG_STATE__UNPAUSE; - if (amdgpu_fence_count_emitted(&adev->jpeg.inst->ring_dec)) + if (amdgpu_fence_count_emitted(adev->jpeg.inst->ring_dec)) new_state.jpeg = VCN_DPG_STATE__PAUSE; else new_state.jpeg = VCN_DPG_STATE__UNPAUSE; @@ -1810,7 +1810,7 @@ static void vcn_v1_0_idle_work_handler(struct work_struct *work) adev->vcn.pause_dpg_mode(adev, 0, &new_state); } - fences += amdgpu_fence_count_emitted(&adev->jpeg.inst->ring_dec); + fences += amdgpu_fence_count_emitted(adev->jpeg.inst->ring_dec); fences += amdgpu_fence_count_emitted(&adev->vcn.inst->ring_dec); if (fences == 0) { @@ -1832,7 +1832,7 @@ static void vcn_v1_0_ring_begin_use(struct amdgpu_ring *ring) mutex_lock(&adev->vcn.vcn1_jpeg1_workaround); - if (amdgpu_fence_wait_empty(&ring->adev->jpeg.inst->ring_dec)) + if (amdgpu_fence_wait_empty(ring->adev->jpeg.inst->ring_dec)) DRM_ERROR("VCN dec: jpeg dec ring may not be empty\n"); vcn_v1_0_set_pg_for_begin_use(ring, set_clocks); @@ -1864,7 +1864,7 @@ void vcn_v1_0_set_pg_for_begin_use(struct amdgpu_ring *ring, bool set_clocks) else new_state.fw_based = VCN_DPG_STATE__UNPAUSE; - if (amdgpu_fence_count_emitted(&adev->jpeg.inst->ring_dec)) + if (amdgpu_fence_count_emitted(adev->jpeg.inst->ring_dec)) new_state.jpeg = VCN_DPG_STATE__PAUSE; else new_state.jpeg = VCN_DPG_STATE__UNPAUSE; diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c index 7c2b3aa480836c502f74413e729915f3faa99588..c975aed2f6c7845f480e0a9b17db92a7110ba217 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c @@ -129,7 +129,7 @@ static int vcn_v2_0_sw_init(void *handle) ring->use_doorbell = true; ring->doorbell_index = adev->doorbell_index.vcn.vcn_ring0_1 << 1; - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); sprintf(ring->name, "vcn_dec"); r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0, @@ -160,7 +160,7 @@ static int vcn_v2_0_sw_init(void *handle) ring = &adev->vcn.inst->ring_enc[i]; ring->use_doorbell = true; - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); if (!amdgpu_sriov_vf(adev)) ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 2 + i; else diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c index ab0b45d0ead18ae6f315a8b59be901247f05aefe..bb1875f926f19d78d16a08538df7232ed0f1c5be 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c @@ -143,7 +143,7 @@ static int vcn_v2_5_sw_init(void *handle) /* VCN POISON TRAP */ r = amdgpu_irq_add_id(adev, amdgpu_ih_clientid_vcns[j], - VCN_2_6__SRCID_UVD_POISON, &adev->vcn.inst[j].irq); + VCN_2_6__SRCID_UVD_POISON, &adev->vcn.inst[j].ras_poison_irq); if (r) return r; } @@ -188,9 +188,9 @@ static int vcn_v2_5_sw_init(void *handle) (amdgpu_sriov_vf(adev) ? 2*j : 8*j); if (adev->ip_versions[UVD_HWIP][0] == IP_VERSION(2, 5, 0)) - ring->vm_hub = AMDGPU_MMHUB_1; + ring->vm_hub = AMDGPU_MMHUB1(0); else - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); sprintf(ring->name, "vcn_dec_%d", j); r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst[j].irq, @@ -208,9 +208,9 @@ static int vcn_v2_5_sw_init(void *handle) (amdgpu_sriov_vf(adev) ? (1 + i + 2*j) : (2 + i + 8*j)); if (adev->ip_versions[UVD_HWIP][0] == IP_VERSION(2, 5, 0)) - ring->vm_hub = AMDGPU_MMHUB_1; + ring->vm_hub = AMDGPU_MMHUB1(0); else - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); sprintf(ring->name, "vcn_enc_%d.%d", j, i); r = amdgpu_ring_init(adev, ring, 512, @@ -354,6 +354,9 @@ static int vcn_v2_5_hw_fini(void *handle) (adev->vcn.cur_state != AMD_PG_STATE_GATE && RREG32_SOC15(VCN, i, mmUVD_STATUS))) vcn_v2_5_set_powergating_state(adev, AMD_PG_STATE_GATE); + + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN)) + amdgpu_irq_put(adev, &adev->vcn.inst[i].ras_poison_irq, 0); } return 0; @@ -1807,6 +1810,14 @@ static int vcn_v2_5_set_interrupt_state(struct amdgpu_device *adev, return 0; } +static int vcn_v2_6_set_ras_interrupt_state(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + unsigned int type, + enum amdgpu_interrupt_state state) +{ + return 0; +} + static int vcn_v2_5_process_interrupt(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) @@ -1837,9 +1848,6 @@ static int vcn_v2_5_process_interrupt(struct amdgpu_device *adev, case VCN_2_0__SRCID__UVD_ENC_LOW_LATENCY: amdgpu_fence_process(&adev->vcn.inst[ip_instance].ring_enc[1]); break; - case VCN_2_6__SRCID_UVD_POISON: - amdgpu_vcn_process_poison_irq(adev, source, entry); - break; default: DRM_ERROR("Unhandled interrupt: %d %d\n", entry->src_id, entry->src_data[0]); @@ -1854,6 +1862,11 @@ static const struct amdgpu_irq_src_funcs vcn_v2_5_irq_funcs = { .process = vcn_v2_5_process_interrupt, }; +static const struct amdgpu_irq_src_funcs vcn_v2_6_ras_irq_funcs = { + .set = vcn_v2_6_set_ras_interrupt_state, + .process = amdgpu_vcn_process_poison_irq, +}; + static void vcn_v2_5_set_irq_funcs(struct amdgpu_device *adev) { int i; @@ -1863,6 +1876,9 @@ static void vcn_v2_5_set_irq_funcs(struct amdgpu_device *adev) continue; adev->vcn.inst[i].irq.num_types = adev->vcn.num_enc_rings + 1; adev->vcn.inst[i].irq.funcs = &vcn_v2_5_irq_funcs; + + adev->vcn.inst[i].ras_poison_irq.num_types = adev->vcn.num_enc_rings + 1; + adev->vcn.inst[i].ras_poison_irq.funcs = &vcn_v2_6_ras_irq_funcs; } } @@ -1965,6 +1981,7 @@ const struct amdgpu_ras_block_hw_ops vcn_v2_6_ras_hw_ops = { static struct amdgpu_vcn_ras vcn_v2_6_ras = { .ras_block = { .hw_ops = &vcn_v2_6_ras_hw_ops, + .ras_late_init = amdgpu_vcn_ras_late_init, }, }; diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c index 3eab186261aabd0a8960684a59bf68d80d6ffdf5..c8f63b3c6f69d945388187d41c65482686e3b354 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c @@ -189,7 +189,7 @@ static int vcn_v3_0_sw_init(void *handle) } else { ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 8 * i; } - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); sprintf(ring->name, "vcn_dec_%d", i); r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst[i].irq, 0, AMDGPU_RING_PRIO_DEFAULT, @@ -213,7 +213,7 @@ static int vcn_v3_0_sw_init(void *handle) } else { ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 2 + j + 8 * i; } - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); sprintf(ring->name, "vcn_enc_%d.%d", i, j); r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst[i].irq, 0, hw_prio, &adev->vcn.inst[i].sched_score); @@ -1313,7 +1313,7 @@ static int vcn_v3_0_start_sriov(struct amdgpu_device *adev) header.version = MMSCH_VERSION; header.total_size = sizeof(struct mmsch_v3_0_init_header) >> 2; - for (i = 0; i < AMDGPU_MAX_VCN_INSTANCES; i++) { + for (i = 0; i < MMSCH_V3_0_VCN_INSTANCES; i++) { header.inst[i].init_status = 0; header.inst[i].table_offset = 0; header.inst[i].table_size = 0; diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c index bf0674039598d3154776c682400d10dfd5db5f29..b48bb52124884a90aab32eaeb3209d861f63866d 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c @@ -129,7 +129,11 @@ static int vcn_v4_0_sw_init(void *handle) if (adev->vcn.harvest_config & (1 << i)) continue; - atomic_set(&adev->vcn.inst[i].sched_score, 0); + /* Init instance 0 sched_score to 1, so it's scheduled after other instances */ + if (i == 0) + atomic_set(&adev->vcn.inst[i].sched_score, 1); + else + atomic_set(&adev->vcn.inst[i].sched_score, 0); /* VCN UNIFIED TRAP */ r = amdgpu_irq_add_id(adev, amdgpu_ih_clientid_vcns[i], @@ -139,7 +143,7 @@ static int vcn_v4_0_sw_init(void *handle) /* VCN POISON TRAP */ r = amdgpu_irq_add_id(adev, amdgpu_ih_clientid_vcns[i], - VCN_4_0__SRCID_UVD_POISON, &adev->vcn.inst[i].irq); + VCN_4_0__SRCID_UVD_POISON, &adev->vcn.inst[i].ras_poison_irq); if (r) return r; @@ -149,7 +153,7 @@ static int vcn_v4_0_sw_init(void *handle) ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + i * (adev->vcn.num_enc_rings + 1) + 1; else ring->doorbell_index = (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + 2 + 8 * i; - ring->vm_hub = AMDGPU_MMHUB_0; + ring->vm_hub = AMDGPU_MMHUB0(0); sprintf(ring->name, "vcn_unified_%d", i); r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst[i].irq, 0, @@ -305,8 +309,8 @@ static int vcn_v4_0_hw_fini(void *handle) vcn_v4_0_set_powergating_state(adev, AMD_PG_STATE_GATE); } } - - amdgpu_irq_put(adev, &adev->vcn.inst[i].irq, 0); + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN)) + amdgpu_irq_put(adev, &adev->vcn.inst[i].ras_poison_irq, 0); } return 0; @@ -1239,7 +1243,7 @@ static int vcn_v4_0_start_sriov(struct amdgpu_device *adev) header.version = MMSCH_VERSION; header.total_size = sizeof(struct mmsch_v4_0_init_header) >> 2; - for (i = 0; i < AMDGPU_MAX_VCN_INSTANCES; i++) { + for (i = 0; i < MMSCH_V4_0_VCN_INSTANCES; i++) { header.inst[i].init_status = 0; header.inst[i].table_offset = 0; header.inst[i].table_size = 0; @@ -1975,6 +1979,24 @@ static int vcn_v4_0_set_interrupt_state(struct amdgpu_device *adev, struct amdgp return 0; } +/** + * vcn_v4_0_set_ras_interrupt_state - set VCN block RAS interrupt state + * + * @adev: amdgpu_device pointer + * @source: interrupt sources + * @type: interrupt types + * @state: interrupt states + * + * Set VCN block RAS interrupt state + */ +static int vcn_v4_0_set_ras_interrupt_state(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + unsigned int type, + enum amdgpu_interrupt_state state) +{ + return 0; +} + /** * vcn_v4_0_process_interrupt - process VCN block interrupt * @@ -2007,9 +2029,6 @@ static int vcn_v4_0_process_interrupt(struct amdgpu_device *adev, struct amdgpu_ case VCN_4_0__SRCID__UVD_ENC_GENERAL_PURPOSE: amdgpu_fence_process(&adev->vcn.inst[ip_instance].ring_enc[0]); break; - case VCN_4_0__SRCID_UVD_POISON: - amdgpu_vcn_process_poison_irq(adev, source, entry); - break; default: DRM_ERROR("Unhandled interrupt: %d %d\n", entry->src_id, entry->src_data[0]); @@ -2024,6 +2043,11 @@ static const struct amdgpu_irq_src_funcs vcn_v4_0_irq_funcs = { .process = vcn_v4_0_process_interrupt, }; +static const struct amdgpu_irq_src_funcs vcn_v4_0_ras_irq_funcs = { + .set = vcn_v4_0_set_ras_interrupt_state, + .process = amdgpu_vcn_process_poison_irq, +}; + /** * vcn_v4_0_set_irq_funcs - set VCN block interrupt irq functions * @@ -2041,6 +2065,9 @@ static void vcn_v4_0_set_irq_funcs(struct amdgpu_device *adev) adev->vcn.inst[i].irq.num_types = adev->vcn.num_enc_rings + 1; adev->vcn.inst[i].irq.funcs = &vcn_v4_0_irq_funcs; + + adev->vcn.inst[i].ras_poison_irq.num_types = adev->vcn.num_enc_rings + 1; + adev->vcn.inst[i].ras_poison_irq.funcs = &vcn_v4_0_ras_irq_funcs; } } @@ -2114,6 +2141,7 @@ const struct amdgpu_ras_block_hw_ops vcn_v4_0_ras_hw_ops = { static struct amdgpu_vcn_ras vcn_v4_0_ras = { .ras_block = { .hw_ops = &vcn_v4_0_ras_hw_ops, + .ras_late_init = amdgpu_vcn_ras_late_init, }, }; diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c new file mode 100644 index 0000000000000000000000000000000000000000..5d67b8b8a3d6bc03a566c62dd56cb9b785df3dc7 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c @@ -0,0 +1,1541 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#include +#include + +#include "amdgpu.h" +#include "amdgpu_vcn.h" +#include "amdgpu_pm.h" +#include "soc15.h" +#include "soc15d.h" +#include "soc15_hw_ip.h" +#include "vcn_v2_0.h" + +#include "vcn/vcn_4_0_3_offset.h" +#include "vcn/vcn_4_0_3_sh_mask.h" +#include "ivsrcid/vcn/irqsrcs_vcn_4_0.h" + +#define mmUVD_DPG_LMA_CTL regUVD_DPG_LMA_CTL +#define mmUVD_DPG_LMA_CTL_BASE_IDX regUVD_DPG_LMA_CTL_BASE_IDX +#define mmUVD_DPG_LMA_DATA regUVD_DPG_LMA_DATA +#define mmUVD_DPG_LMA_DATA_BASE_IDX regUVD_DPG_LMA_DATA_BASE_IDX + +#define VCN_VID_SOC_ADDRESS_2_0 0x1fb00 +#define VCN1_VID_SOC_ADDRESS_3_0 0x48300 + +static void vcn_v4_0_3_set_unified_ring_funcs(struct amdgpu_device *adev); +static void vcn_v4_0_3_set_irq_funcs(struct amdgpu_device *adev); +static int vcn_v4_0_3_set_powergating_state(void *handle, + enum amd_powergating_state state); +static int vcn_v4_0_3_pause_dpg_mode(struct amdgpu_device *adev, + int inst_idx, struct dpg_pause_state *new_state); +static void vcn_v4_0_3_unified_ring_set_wptr(struct amdgpu_ring *ring); +static void vcn_v4_0_3_set_ras_funcs(struct amdgpu_device *adev); +static void vcn_v4_0_3_enable_ras(struct amdgpu_device *adev, + int inst_idx, bool indirect); +/** + * vcn_v4_0_3_early_init - set function pointers + * + * @handle: amdgpu_device pointer + * + * Set ring and irq function pointers + */ +static int vcn_v4_0_3_early_init(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + /* re-use enc ring as unified ring */ + adev->vcn.num_enc_rings = 1; + + vcn_v4_0_3_set_unified_ring_funcs(adev); + vcn_v4_0_3_set_irq_funcs(adev); + vcn_v4_0_3_set_ras_funcs(adev); + + return amdgpu_vcn_early_init(adev); +} + +/** + * vcn_v4_0_3_sw_init - sw init for VCN block + * + * @handle: amdgpu_device pointer + * + * Load firmware and sw initialization + */ +static int vcn_v4_0_3_sw_init(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_ring *ring; + int i, r, vcn_inst; + + r = amdgpu_vcn_sw_init(adev); + if (r) + return r; + + amdgpu_vcn_setup_ucode(adev); + + r = amdgpu_vcn_resume(adev); + if (r) + return r; + + /* VCN DEC TRAP */ + r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN, + VCN_4_0__SRCID__UVD_ENC_GENERAL_PURPOSE, &adev->vcn.inst->irq); + if (r) + return r; + + for (i = 0; i < adev->vcn.num_vcn_inst; i++) { + volatile struct amdgpu_vcn4_fw_shared *fw_shared; + + vcn_inst = GET_INST(VCN, i); + + ring = &adev->vcn.inst[i].ring_enc[0]; + ring->use_doorbell = true; + ring->doorbell_index = + (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + + 9 * vcn_inst; + ring->vm_hub = AMDGPU_MMHUB0(adev->vcn.inst[i].aid_id); + sprintf(ring->name, "vcn_unified_%d", adev->vcn.inst[i].aid_id); + r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0, + AMDGPU_RING_PRIO_DEFAULT, + &adev->vcn.inst[i].sched_score); + if (r) + return r; + + fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr; + fw_shared->present_flag_0 = cpu_to_le32(AMDGPU_FW_SHARED_FLAG_0_UNIFIED_QUEUE); + fw_shared->sq.is_enabled = true; + + if (amdgpu_vcnfw_log) + amdgpu_vcn_fwlog_init(&adev->vcn.inst[i]); + } + + if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) + adev->vcn.pause_dpg_mode = vcn_v4_0_3_pause_dpg_mode; + + if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN)) { + r = amdgpu_vcn_ras_sw_init(adev); + if (r) { + dev_err(adev->dev, "Failed to initialize vcn ras block!\n"); + return r; + } + } + + return 0; +} + +/** + * vcn_v4_0_3_sw_fini - sw fini for VCN block + * + * @handle: amdgpu_device pointer + * + * VCN suspend and free up sw allocation + */ +static int vcn_v4_0_3_sw_fini(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int i, r, idx; + + if (drm_dev_enter(&adev->ddev, &idx)) { + for (i = 0; i < adev->vcn.num_vcn_inst; i++) { + volatile struct amdgpu_vcn4_fw_shared *fw_shared; + + fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr; + fw_shared->present_flag_0 = 0; + fw_shared->sq.is_enabled = cpu_to_le32(false); + } + drm_dev_exit(idx); + } + + r = amdgpu_vcn_suspend(adev); + if (r) + return r; + + r = amdgpu_vcn_sw_fini(adev); + + return r; +} + +/** + * vcn_v4_0_3_hw_init - start and test VCN block + * + * @handle: amdgpu_device pointer + * + * Initialize the hardware, boot up the VCPU and do some testing + */ +static int vcn_v4_0_3_hw_init(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_ring *ring; + int i, r, vcn_inst; + + for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { + vcn_inst = GET_INST(VCN, i); + ring = &adev->vcn.inst[i].ring_enc[0]; + + if (ring->use_doorbell) { + adev->nbio.funcs->vcn_doorbell_range( + adev, ring->use_doorbell, + (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + + 9 * vcn_inst, + adev->vcn.inst[i].aid_id); + + WREG32_SOC15( + VCN, GET_INST(VCN, ring->me), + regVCN_RB1_DB_CTRL, + ring->doorbell_index + << VCN_RB1_DB_CTRL__OFFSET__SHIFT | + VCN_RB1_DB_CTRL__EN_MASK); + + /* Read DB_CTRL to flush the write DB_CTRL command. */ + RREG32_SOC15( + VCN, GET_INST(VCN, ring->me), + regVCN_RB1_DB_CTRL); + } + + r = amdgpu_ring_test_helper(ring); + if (r) + goto done; + } + +done: + if (!r) + DRM_DEV_INFO(adev->dev, "VCN decode initialized successfully(under %s).\n", + (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)?"DPG Mode":"SPG Mode"); + + return r; +} + +/** + * vcn_v4_0_3_hw_fini - stop the hardware block + * + * @handle: amdgpu_device pointer + * + * Stop the VCN block, mark ring as not ready any more + */ +static int vcn_v4_0_3_hw_fini(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + cancel_delayed_work_sync(&adev->vcn.idle_work); + + if (adev->vcn.cur_state != AMD_PG_STATE_GATE) + vcn_v4_0_3_set_powergating_state(adev, AMD_PG_STATE_GATE); + + return 0; +} + +/** + * vcn_v4_0_3_suspend - suspend VCN block + * + * @handle: amdgpu_device pointer + * + * HW fini and suspend VCN block + */ +static int vcn_v4_0_3_suspend(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int r; + + r = vcn_v4_0_3_hw_fini(adev); + if (r) + return r; + + r = amdgpu_vcn_suspend(adev); + + return r; +} + +/** + * vcn_v4_0_3_resume - resume VCN block + * + * @handle: amdgpu_device pointer + * + * Resume firmware and hw init VCN block + */ +static int vcn_v4_0_3_resume(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int r; + + r = amdgpu_vcn_resume(adev); + if (r) + return r; + + r = vcn_v4_0_3_hw_init(adev); + + return r; +} + +/** + * vcn_v4_0_3_mc_resume - memory controller programming + * + * @adev: amdgpu_device pointer + * @inst_idx: instance number + * + * Let the VCN memory controller know it's offsets + */ +static void vcn_v4_0_3_mc_resume(struct amdgpu_device *adev, int inst_idx) +{ + uint32_t offset, size, vcn_inst; + const struct common_firmware_header *hdr; + + hdr = (const struct common_firmware_header *)adev->vcn.fw->data; + size = AMDGPU_GPU_PAGE_ALIGN(le32_to_cpu(hdr->ucode_size_bytes) + 8); + + vcn_inst = GET_INST(VCN, inst_idx); + /* cache window 0: fw */ + if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { + WREG32_SOC15( + VCN, vcn_inst, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW, + (adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + inst_idx] + .tmr_mc_addr_lo)); + WREG32_SOC15( + VCN, vcn_inst, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH, + (adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + inst_idx] + .tmr_mc_addr_hi)); + WREG32_SOC15(VCN, vcn_inst, regUVD_VCPU_CACHE_OFFSET0, 0); + offset = 0; + } else { + WREG32_SOC15(VCN, vcn_inst, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW, + lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr)); + WREG32_SOC15(VCN, vcn_inst, + regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH, + upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr)); + offset = size; + WREG32_SOC15(VCN, vcn_inst, regUVD_VCPU_CACHE_OFFSET0, + AMDGPU_UVD_FIRMWARE_OFFSET >> 3); + } + WREG32_SOC15(VCN, vcn_inst, regUVD_VCPU_CACHE_SIZE0, size); + + /* cache window 1: stack */ + WREG32_SOC15(VCN, vcn_inst, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW, + lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset)); + WREG32_SOC15(VCN, vcn_inst, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH, + upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset)); + WREG32_SOC15(VCN, vcn_inst, regUVD_VCPU_CACHE_OFFSET1, 0); + WREG32_SOC15(VCN, vcn_inst, regUVD_VCPU_CACHE_SIZE1, + AMDGPU_VCN_STACK_SIZE); + + /* cache window 2: context */ + WREG32_SOC15(VCN, vcn_inst, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW, + lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset + + AMDGPU_VCN_STACK_SIZE)); + WREG32_SOC15(VCN, vcn_inst, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH, + upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset + + AMDGPU_VCN_STACK_SIZE)); + WREG32_SOC15(VCN, vcn_inst, regUVD_VCPU_CACHE_OFFSET2, 0); + WREG32_SOC15(VCN, vcn_inst, regUVD_VCPU_CACHE_SIZE2, + AMDGPU_VCN_CONTEXT_SIZE); + + /* non-cache window */ + WREG32_SOC15( + VCN, vcn_inst, regUVD_LMI_VCPU_NC0_64BIT_BAR_LOW, + lower_32_bits(adev->vcn.inst[inst_idx].fw_shared.gpu_addr)); + WREG32_SOC15( + VCN, vcn_inst, regUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH, + upper_32_bits(adev->vcn.inst[inst_idx].fw_shared.gpu_addr)); + WREG32_SOC15(VCN, vcn_inst, regUVD_VCPU_NONCACHE_OFFSET0, 0); + WREG32_SOC15( + VCN, vcn_inst, regUVD_VCPU_NONCACHE_SIZE0, + AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_vcn4_fw_shared))); +} + +/** + * vcn_v4_0_3_mc_resume_dpg_mode - memory controller programming for dpg mode + * + * @adev: amdgpu_device pointer + * @inst_idx: instance number index + * @indirect: indirectly write sram + * + * Let the VCN memory controller know it's offsets with dpg mode + */ +static void vcn_v4_0_3_mc_resume_dpg_mode(struct amdgpu_device *adev, int inst_idx, bool indirect) +{ + uint32_t offset, size; + const struct common_firmware_header *hdr; + + hdr = (const struct common_firmware_header *)adev->vcn.fw->data; + size = AMDGPU_GPU_PAGE_ALIGN(le32_to_cpu(hdr->ucode_size_bytes) + 8); + + /* cache window 0: fw */ + if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { + if (!indirect) { + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), + (adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + + inst_idx].tmr_mc_addr_lo), 0, indirect); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), + (adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + + inst_idx].tmr_mc_addr_hi), 0, indirect); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_VCPU_CACHE_OFFSET0), 0, 0, indirect); + } else { + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), 0, 0, indirect); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), 0, 0, indirect); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_VCPU_CACHE_OFFSET0), 0, 0, indirect); + } + offset = 0; + } else { + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), + lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr), 0, indirect); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), + upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr), 0, indirect); + offset = size; + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_VCPU_CACHE_OFFSET0), + AMDGPU_UVD_FIRMWARE_OFFSET >> 3, 0, indirect); + } + + if (!indirect) + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_VCPU_CACHE_SIZE0), size, 0, indirect); + else + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_VCPU_CACHE_SIZE0), 0, 0, indirect); + + /* cache window 1: stack */ + if (!indirect) { + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW), + lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset), 0, indirect); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH), + upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset), 0, indirect); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_VCPU_CACHE_OFFSET1), 0, 0, indirect); + } else { + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW), 0, 0, indirect); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH), 0, 0, indirect); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_VCPU_CACHE_OFFSET1), 0, 0, indirect); + } + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_VCPU_CACHE_SIZE1), AMDGPU_VCN_STACK_SIZE, 0, indirect); + + /* cache window 2: context */ + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW), + lower_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset + + AMDGPU_VCN_STACK_SIZE), 0, indirect); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH), + upper_32_bits(adev->vcn.inst[inst_idx].gpu_addr + offset + + AMDGPU_VCN_STACK_SIZE), 0, indirect); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_VCPU_CACHE_OFFSET2), 0, 0, indirect); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_VCPU_CACHE_SIZE2), AMDGPU_VCN_CONTEXT_SIZE, 0, indirect); + + /* non-cache window */ + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_LMI_VCPU_NC0_64BIT_BAR_LOW), + lower_32_bits(adev->vcn.inst[inst_idx].fw_shared.gpu_addr), 0, indirect); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH), + upper_32_bits(adev->vcn.inst[inst_idx].fw_shared.gpu_addr), 0, indirect); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_VCPU_NONCACHE_OFFSET0), 0, 0, indirect); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_VCPU_NONCACHE_SIZE0), + AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_vcn4_fw_shared)), 0, indirect); + + /* VCN global tiling registers */ + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_GFX8_ADDR_CONFIG), adev->gfx.config.gb_addr_config, 0, indirect); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_GFX10_ADDR_CONFIG), adev->gfx.config.gb_addr_config, 0, indirect); +} + +/** + * vcn_v4_0_3_disable_clock_gating - disable VCN clock gating + * + * @adev: amdgpu_device pointer + * @inst_idx: instance number + * + * Disable clock gating for VCN block + */ +static void vcn_v4_0_3_disable_clock_gating(struct amdgpu_device *adev, int inst_idx) +{ + uint32_t data; + int vcn_inst; + + if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG) + return; + + vcn_inst = GET_INST(VCN, inst_idx); + + /* VCN disable CGC */ + data = RREG32_SOC15(VCN, vcn_inst, regUVD_CGC_CTRL); + data &= ~UVD_CGC_CTRL__DYN_CLOCK_MODE_MASK; + data |= 1 << UVD_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT; + data |= 4 << UVD_CGC_CTRL__CLK_OFF_DELAY__SHIFT; + WREG32_SOC15(VCN, vcn_inst, regUVD_CGC_CTRL, data); + + data = RREG32_SOC15(VCN, vcn_inst, regUVD_CGC_GATE); + data &= ~(UVD_CGC_GATE__SYS_MASK + | UVD_CGC_GATE__MPEG2_MASK + | UVD_CGC_GATE__REGS_MASK + | UVD_CGC_GATE__RBC_MASK + | UVD_CGC_GATE__LMI_MC_MASK + | UVD_CGC_GATE__LMI_UMC_MASK + | UVD_CGC_GATE__MPC_MASK + | UVD_CGC_GATE__LBSI_MASK + | UVD_CGC_GATE__LRBBM_MASK + | UVD_CGC_GATE__WCB_MASK + | UVD_CGC_GATE__VCPU_MASK + | UVD_CGC_GATE__MMSCH_MASK); + + WREG32_SOC15(VCN, vcn_inst, regUVD_CGC_GATE, data); + SOC15_WAIT_ON_RREG(VCN, vcn_inst, regUVD_CGC_GATE, 0, 0xFFFFFFFF); + + data = RREG32_SOC15(VCN, vcn_inst, regUVD_CGC_CTRL); + data &= ~(UVD_CGC_CTRL__SYS_MODE_MASK + | UVD_CGC_CTRL__MPEG2_MODE_MASK + | UVD_CGC_CTRL__REGS_MODE_MASK + | UVD_CGC_CTRL__RBC_MODE_MASK + | UVD_CGC_CTRL__LMI_MC_MODE_MASK + | UVD_CGC_CTRL__LMI_UMC_MODE_MASK + | UVD_CGC_CTRL__MPC_MODE_MASK + | UVD_CGC_CTRL__LBSI_MODE_MASK + | UVD_CGC_CTRL__LRBBM_MODE_MASK + | UVD_CGC_CTRL__WCB_MODE_MASK + | UVD_CGC_CTRL__VCPU_MODE_MASK + | UVD_CGC_CTRL__MMSCH_MODE_MASK); + WREG32_SOC15(VCN, vcn_inst, regUVD_CGC_CTRL, data); + + data = RREG32_SOC15(VCN, vcn_inst, regUVD_SUVD_CGC_GATE); + data |= (UVD_SUVD_CGC_GATE__SRE_MASK + | UVD_SUVD_CGC_GATE__SIT_MASK + | UVD_SUVD_CGC_GATE__SMP_MASK + | UVD_SUVD_CGC_GATE__SCM_MASK + | UVD_SUVD_CGC_GATE__SDB_MASK + | UVD_SUVD_CGC_GATE__SRE_H264_MASK + | UVD_SUVD_CGC_GATE__SRE_HEVC_MASK + | UVD_SUVD_CGC_GATE__SIT_H264_MASK + | UVD_SUVD_CGC_GATE__SIT_HEVC_MASK + | UVD_SUVD_CGC_GATE__SCM_H264_MASK + | UVD_SUVD_CGC_GATE__SCM_HEVC_MASK + | UVD_SUVD_CGC_GATE__SDB_H264_MASK + | UVD_SUVD_CGC_GATE__SDB_HEVC_MASK + | UVD_SUVD_CGC_GATE__ENT_MASK + | UVD_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK + | UVD_SUVD_CGC_GATE__SITE_MASK + | UVD_SUVD_CGC_GATE__SRE_VP9_MASK + | UVD_SUVD_CGC_GATE__SCM_VP9_MASK + | UVD_SUVD_CGC_GATE__SIT_VP9_DEC_MASK + | UVD_SUVD_CGC_GATE__SDB_VP9_MASK + | UVD_SUVD_CGC_GATE__IME_HEVC_MASK); + WREG32_SOC15(VCN, vcn_inst, regUVD_SUVD_CGC_GATE, data); + + data = RREG32_SOC15(VCN, vcn_inst, regUVD_SUVD_CGC_CTRL); + data &= ~(UVD_SUVD_CGC_CTRL__SRE_MODE_MASK + | UVD_SUVD_CGC_CTRL__SIT_MODE_MASK + | UVD_SUVD_CGC_CTRL__SMP_MODE_MASK + | UVD_SUVD_CGC_CTRL__SCM_MODE_MASK + | UVD_SUVD_CGC_CTRL__SDB_MODE_MASK + | UVD_SUVD_CGC_CTRL__ENT_MODE_MASK + | UVD_SUVD_CGC_CTRL__IME_MODE_MASK + | UVD_SUVD_CGC_CTRL__SITE_MODE_MASK); + WREG32_SOC15(VCN, vcn_inst, regUVD_SUVD_CGC_CTRL, data); +} + +/** + * vcn_v4_0_3_disable_clock_gating_dpg_mode - disable VCN clock gating dpg mode + * + * @adev: amdgpu_device pointer + * @sram_sel: sram select + * @inst_idx: instance number index + * @indirect: indirectly write sram + * + * Disable clock gating for VCN block with dpg mode + */ +static void vcn_v4_0_3_disable_clock_gating_dpg_mode(struct amdgpu_device *adev, uint8_t sram_sel, + int inst_idx, uint8_t indirect) +{ + uint32_t reg_data = 0; + + if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG) + return; + + /* enable sw clock gating control */ + reg_data = 0 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT; + reg_data |= 1 << UVD_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT; + reg_data |= 4 << UVD_CGC_CTRL__CLK_OFF_DELAY__SHIFT; + reg_data &= ~(UVD_CGC_CTRL__SYS_MODE_MASK | + UVD_CGC_CTRL__MPEG2_MODE_MASK | + UVD_CGC_CTRL__REGS_MODE_MASK | + UVD_CGC_CTRL__RBC_MODE_MASK | + UVD_CGC_CTRL__LMI_MC_MODE_MASK | + UVD_CGC_CTRL__LMI_UMC_MODE_MASK | + UVD_CGC_CTRL__IDCT_MODE_MASK | + UVD_CGC_CTRL__MPRD_MODE_MASK | + UVD_CGC_CTRL__MPC_MODE_MASK | + UVD_CGC_CTRL__LBSI_MODE_MASK | + UVD_CGC_CTRL__LRBBM_MODE_MASK | + UVD_CGC_CTRL__WCB_MODE_MASK | + UVD_CGC_CTRL__VCPU_MODE_MASK); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_CGC_CTRL), reg_data, sram_sel, indirect); + + /* turn off clock gating */ + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_CGC_GATE), 0, sram_sel, indirect); + + /* turn on SUVD clock gating */ + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_SUVD_CGC_GATE), 1, sram_sel, indirect); + + /* turn on sw mode in UVD_SUVD_CGC_CTRL */ + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_SUVD_CGC_CTRL), 0, sram_sel, indirect); +} + +/** + * vcn_v4_0_3_enable_clock_gating - enable VCN clock gating + * + * @adev: amdgpu_device pointer + * @inst_idx: instance number + * + * Enable clock gating for VCN block + */ +static void vcn_v4_0_3_enable_clock_gating(struct amdgpu_device *adev, int inst_idx) +{ + uint32_t data; + int vcn_inst; + + if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG) + return; + + vcn_inst = GET_INST(VCN, inst_idx); + + /* enable VCN CGC */ + data = RREG32_SOC15(VCN, vcn_inst, regUVD_CGC_CTRL); + data |= 0 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT; + data |= 1 << UVD_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT; + data |= 4 << UVD_CGC_CTRL__CLK_OFF_DELAY__SHIFT; + WREG32_SOC15(VCN, vcn_inst, regUVD_CGC_CTRL, data); + + data = RREG32_SOC15(VCN, vcn_inst, regUVD_CGC_CTRL); + data |= (UVD_CGC_CTRL__SYS_MODE_MASK + | UVD_CGC_CTRL__MPEG2_MODE_MASK + | UVD_CGC_CTRL__REGS_MODE_MASK + | UVD_CGC_CTRL__RBC_MODE_MASK + | UVD_CGC_CTRL__LMI_MC_MODE_MASK + | UVD_CGC_CTRL__LMI_UMC_MODE_MASK + | UVD_CGC_CTRL__MPC_MODE_MASK + | UVD_CGC_CTRL__LBSI_MODE_MASK + | UVD_CGC_CTRL__LRBBM_MODE_MASK + | UVD_CGC_CTRL__WCB_MODE_MASK + | UVD_CGC_CTRL__VCPU_MODE_MASK); + WREG32_SOC15(VCN, vcn_inst, regUVD_CGC_CTRL, data); + + data = RREG32_SOC15(VCN, vcn_inst, regUVD_SUVD_CGC_CTRL); + data |= (UVD_SUVD_CGC_CTRL__SRE_MODE_MASK + | UVD_SUVD_CGC_CTRL__SIT_MODE_MASK + | UVD_SUVD_CGC_CTRL__SMP_MODE_MASK + | UVD_SUVD_CGC_CTRL__SCM_MODE_MASK + | UVD_SUVD_CGC_CTRL__SDB_MODE_MASK + | UVD_SUVD_CGC_CTRL__ENT_MODE_MASK + | UVD_SUVD_CGC_CTRL__IME_MODE_MASK + | UVD_SUVD_CGC_CTRL__SITE_MODE_MASK); + WREG32_SOC15(VCN, vcn_inst, regUVD_SUVD_CGC_CTRL, data); +} + +/** + * vcn_v4_0_3_start_dpg_mode - VCN start with dpg mode + * + * @adev: amdgpu_device pointer + * @inst_idx: instance number index + * @indirect: indirectly write sram + * + * Start VCN block with dpg mode + */ +static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, bool indirect) +{ + volatile struct amdgpu_vcn4_fw_shared *fw_shared = + adev->vcn.inst[inst_idx].fw_shared.cpu_addr; + struct amdgpu_ring *ring; + int vcn_inst; + uint32_t tmp; + + vcn_inst = GET_INST(VCN, inst_idx); + /* disable register anti-hang mechanism */ + WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_POWER_STATUS), 1, + ~UVD_POWER_STATUS__UVD_POWER_STATUS_MASK); + /* enable dynamic power gating mode */ + tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_POWER_STATUS); + tmp |= UVD_POWER_STATUS__UVD_PG_MODE_MASK; + tmp |= UVD_POWER_STATUS__UVD_PG_EN_MASK; + WREG32_SOC15(VCN, vcn_inst, regUVD_POWER_STATUS, tmp); + + if (indirect) { + DRM_DEV_DEBUG(adev->dev, "VCN %d start: on AID %d", + inst_idx, adev->vcn.inst[inst_idx].aid_id); + adev->vcn.inst[inst_idx].dpg_sram_curr_addr = + (uint32_t *)adev->vcn.inst[inst_idx].dpg_sram_cpu_addr; + /* Use dummy register 0xDEADBEEF passing AID selection to PSP FW */ + WREG32_SOC15_DPG_MODE(inst_idx, 0xDEADBEEF, + adev->vcn.inst[inst_idx].aid_id, 0, true); + } + + /* enable clock gating */ + vcn_v4_0_3_disable_clock_gating_dpg_mode(adev, 0, inst_idx, indirect); + + /* enable VCPU clock */ + tmp = (0xFF << UVD_VCPU_CNTL__PRB_TIMEOUT_VAL__SHIFT); + tmp |= UVD_VCPU_CNTL__CLK_EN_MASK; + tmp |= UVD_VCPU_CNTL__BLK_RST_MASK; + + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_VCPU_CNTL), tmp, 0, indirect); + + /* disable master interrupt */ + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_MASTINT_EN), 0, 0, indirect); + + /* setup regUVD_LMI_CTRL */ + tmp = (UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN_MASK | + UVD_LMI_CTRL__REQ_MODE_MASK | + UVD_LMI_CTRL__CRC_RESET_MASK | + UVD_LMI_CTRL__MASK_MC_URGENT_MASK | + UVD_LMI_CTRL__DATA_COHERENCY_EN_MASK | + UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN_MASK | + (8 << UVD_LMI_CTRL__WRITE_CLEAN_TIMER__SHIFT) | + 0x00100000L); + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_LMI_CTRL), tmp, 0, indirect); + + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_MPC_CNTL), + 0x2 << UVD_MPC_CNTL__REPLACEMENT_MODE__SHIFT, 0, indirect); + + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_MPC_SET_MUXA0), + ((0x1 << UVD_MPC_SET_MUXA0__VARA_1__SHIFT) | + (0x2 << UVD_MPC_SET_MUXA0__VARA_2__SHIFT) | + (0x3 << UVD_MPC_SET_MUXA0__VARA_3__SHIFT) | + (0x4 << UVD_MPC_SET_MUXA0__VARA_4__SHIFT)), 0, indirect); + + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_MPC_SET_MUXB0), + ((0x1 << UVD_MPC_SET_MUXB0__VARB_1__SHIFT) | + (0x2 << UVD_MPC_SET_MUXB0__VARB_2__SHIFT) | + (0x3 << UVD_MPC_SET_MUXB0__VARB_3__SHIFT) | + (0x4 << UVD_MPC_SET_MUXB0__VARB_4__SHIFT)), 0, indirect); + + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_MPC_SET_MUX), + ((0x0 << UVD_MPC_SET_MUX__SET_0__SHIFT) | + (0x1 << UVD_MPC_SET_MUX__SET_1__SHIFT) | + (0x2 << UVD_MPC_SET_MUX__SET_2__SHIFT)), 0, indirect); + + vcn_v4_0_3_mc_resume_dpg_mode(adev, inst_idx, indirect); + + tmp = (0xFF << UVD_VCPU_CNTL__PRB_TIMEOUT_VAL__SHIFT); + tmp |= UVD_VCPU_CNTL__CLK_EN_MASK; + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_VCPU_CNTL), tmp, 0, indirect); + + /* enable LMI MC and UMC channels */ + tmp = 0x1f << UVD_LMI_CTRL2__RE_OFLD_MIF_WR_REQ_NUM__SHIFT; + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_LMI_CTRL2), tmp, 0, indirect); + + vcn_v4_0_3_enable_ras(adev, inst_idx, indirect); + + /* enable master interrupt */ + WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET( + VCN, 0, regUVD_MASTINT_EN), + UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect); + + if (indirect) + psp_update_vcn_sram(adev, 0, adev->vcn.inst[inst_idx].dpg_sram_gpu_addr, + (uint32_t)((uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_curr_addr - + (uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_cpu_addr)); + + ring = &adev->vcn.inst[inst_idx].ring_enc[0]; + + /* program the RB_BASE for ring buffer */ + WREG32_SOC15(VCN, vcn_inst, regUVD_RB_BASE_LO, + lower_32_bits(ring->gpu_addr)); + WREG32_SOC15(VCN, vcn_inst, regUVD_RB_BASE_HI, + upper_32_bits(ring->gpu_addr)); + + WREG32_SOC15(VCN, vcn_inst, regUVD_RB_SIZE, + ring->ring_size / sizeof(uint32_t)); + + /* resetting ring, fw should not check RB ring */ + tmp = RREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE); + tmp &= ~(VCN_RB_ENABLE__RB_EN_MASK); + WREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE, tmp); + fw_shared->sq.queue_mode |= FW_QUEUE_RING_RESET; + + /* Initialize the ring buffer's read and write pointers */ + WREG32_SOC15(VCN, vcn_inst, regUVD_RB_RPTR, 0); + WREG32_SOC15(VCN, vcn_inst, regUVD_RB_WPTR, 0); + ring->wptr = RREG32_SOC15(VCN, vcn_inst, regUVD_RB_WPTR); + + tmp = RREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE); + tmp |= VCN_RB_ENABLE__RB_EN_MASK; + WREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE, tmp); + fw_shared->sq.queue_mode &= ~(FW_QUEUE_RING_RESET | FW_QUEUE_DPG_HOLD_OFF); + + /*resetting done, fw can check RB ring */ + fw_shared->sq.queue_mode &= cpu_to_le32(~FW_QUEUE_RING_RESET); + + return 0; +} + +/** + * vcn_v4_0_3_start - VCN start + * + * @adev: amdgpu_device pointer + * + * Start VCN block + */ +static int vcn_v4_0_3_start(struct amdgpu_device *adev) +{ + volatile struct amdgpu_vcn4_fw_shared *fw_shared; + struct amdgpu_ring *ring; + int i, j, k, r, vcn_inst; + uint32_t tmp; + + if (adev->pm.dpm_enabled) + amdgpu_dpm_enable_uvd(adev, true); + + for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { + if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) { + r = vcn_v4_0_3_start_dpg_mode(adev, i, adev->vcn.indirect_sram); + continue; + } + + vcn_inst = GET_INST(VCN, i); + /* set VCN status busy */ + tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_STATUS) | + UVD_STATUS__UVD_BUSY; + WREG32_SOC15(VCN, vcn_inst, regUVD_STATUS, tmp); + + /*SW clock gating */ + vcn_v4_0_3_disable_clock_gating(adev, i); + + /* enable VCPU clock */ + WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_VCPU_CNTL), + UVD_VCPU_CNTL__CLK_EN_MASK, + ~UVD_VCPU_CNTL__CLK_EN_MASK); + + /* disable master interrupt */ + WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_MASTINT_EN), 0, + ~UVD_MASTINT_EN__VCPU_EN_MASK); + + /* enable LMI MC and UMC channels */ + WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_LMI_CTRL2), 0, + ~UVD_LMI_CTRL2__STALL_ARB_UMC_MASK); + + tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_SOFT_RESET); + tmp &= ~UVD_SOFT_RESET__LMI_SOFT_RESET_MASK; + tmp &= ~UVD_SOFT_RESET__LMI_UMC_SOFT_RESET_MASK; + WREG32_SOC15(VCN, vcn_inst, regUVD_SOFT_RESET, tmp); + + /* setup regUVD_LMI_CTRL */ + tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_LMI_CTRL); + WREG32_SOC15(VCN, vcn_inst, regUVD_LMI_CTRL, + tmp | UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN_MASK | + UVD_LMI_CTRL__MASK_MC_URGENT_MASK | + UVD_LMI_CTRL__DATA_COHERENCY_EN_MASK | + UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN_MASK); + + /* setup regUVD_MPC_CNTL */ + tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_MPC_CNTL); + tmp &= ~UVD_MPC_CNTL__REPLACEMENT_MODE_MASK; + tmp |= 0x2 << UVD_MPC_CNTL__REPLACEMENT_MODE__SHIFT; + WREG32_SOC15(VCN, vcn_inst, regUVD_MPC_CNTL, tmp); + + /* setup UVD_MPC_SET_MUXA0 */ + WREG32_SOC15(VCN, vcn_inst, regUVD_MPC_SET_MUXA0, + ((0x1 << UVD_MPC_SET_MUXA0__VARA_1__SHIFT) | + (0x2 << UVD_MPC_SET_MUXA0__VARA_2__SHIFT) | + (0x3 << UVD_MPC_SET_MUXA0__VARA_3__SHIFT) | + (0x4 << UVD_MPC_SET_MUXA0__VARA_4__SHIFT))); + + /* setup UVD_MPC_SET_MUXB0 */ + WREG32_SOC15(VCN, vcn_inst, regUVD_MPC_SET_MUXB0, + ((0x1 << UVD_MPC_SET_MUXB0__VARB_1__SHIFT) | + (0x2 << UVD_MPC_SET_MUXB0__VARB_2__SHIFT) | + (0x3 << UVD_MPC_SET_MUXB0__VARB_3__SHIFT) | + (0x4 << UVD_MPC_SET_MUXB0__VARB_4__SHIFT))); + + /* setup UVD_MPC_SET_MUX */ + WREG32_SOC15(VCN, vcn_inst, regUVD_MPC_SET_MUX, + ((0x0 << UVD_MPC_SET_MUX__SET_0__SHIFT) | + (0x1 << UVD_MPC_SET_MUX__SET_1__SHIFT) | + (0x2 << UVD_MPC_SET_MUX__SET_2__SHIFT))); + + vcn_v4_0_3_mc_resume(adev, i); + + /* VCN global tiling registers */ + WREG32_SOC15(VCN, vcn_inst, regUVD_GFX8_ADDR_CONFIG, + adev->gfx.config.gb_addr_config); + WREG32_SOC15(VCN, vcn_inst, regUVD_GFX10_ADDR_CONFIG, + adev->gfx.config.gb_addr_config); + + /* unblock VCPU register access */ + WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_RB_ARB_CTRL), 0, + ~UVD_RB_ARB_CTRL__VCPU_DIS_MASK); + + /* release VCPU reset to boot */ + WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_VCPU_CNTL), 0, + ~UVD_VCPU_CNTL__BLK_RST_MASK); + + for (j = 0; j < 10; ++j) { + uint32_t status; + + for (k = 0; k < 100; ++k) { + status = RREG32_SOC15(VCN, vcn_inst, + regUVD_STATUS); + if (status & 2) + break; + mdelay(10); + } + r = 0; + if (status & 2) + break; + + DRM_DEV_ERROR(adev->dev, + "VCN decode not responding, trying to reset the VCPU!!!\n"); + WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, + regUVD_VCPU_CNTL), + UVD_VCPU_CNTL__BLK_RST_MASK, + ~UVD_VCPU_CNTL__BLK_RST_MASK); + mdelay(10); + WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, + regUVD_VCPU_CNTL), + 0, ~UVD_VCPU_CNTL__BLK_RST_MASK); + + mdelay(10); + r = -1; + } + + if (r) { + DRM_DEV_ERROR(adev->dev, "VCN decode not responding, giving up!!!\n"); + return r; + } + + /* enable master interrupt */ + WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_MASTINT_EN), + UVD_MASTINT_EN__VCPU_EN_MASK, + ~UVD_MASTINT_EN__VCPU_EN_MASK); + + /* clear the busy bit of VCN_STATUS */ + WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_STATUS), 0, + ~(2 << UVD_STATUS__VCPU_REPORT__SHIFT)); + + ring = &adev->vcn.inst[i].ring_enc[0]; + fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr; + + /* program the RB_BASE for ring buffer */ + WREG32_SOC15(VCN, vcn_inst, regUVD_RB_BASE_LO, + lower_32_bits(ring->gpu_addr)); + WREG32_SOC15(VCN, vcn_inst, regUVD_RB_BASE_HI, + upper_32_bits(ring->gpu_addr)); + + WREG32_SOC15(VCN, vcn_inst, regUVD_RB_SIZE, + ring->ring_size / sizeof(uint32_t)); + + /* resetting ring, fw should not check RB ring */ + tmp = RREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE); + tmp &= ~(VCN_RB_ENABLE__RB_EN_MASK); + WREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE, tmp); + + /* Initialize the ring buffer's read and write pointers */ + WREG32_SOC15(VCN, vcn_inst, regUVD_RB_RPTR, 0); + WREG32_SOC15(VCN, vcn_inst, regUVD_RB_WPTR, 0); + + tmp = RREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE); + tmp |= VCN_RB_ENABLE__RB_EN_MASK; + WREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE, tmp); + + ring->wptr = RREG32_SOC15(VCN, vcn_inst, regUVD_RB_WPTR); + fw_shared->sq.queue_mode &= + cpu_to_le32(~(FW_QUEUE_RING_RESET | FW_QUEUE_DPG_HOLD_OFF)); + + } + return 0; +} + +/** + * vcn_v4_0_3_stop_dpg_mode - VCN stop with dpg mode + * + * @adev: amdgpu_device pointer + * @inst_idx: instance number index + * + * Stop VCN block with dpg mode + */ +static int vcn_v4_0_3_stop_dpg_mode(struct amdgpu_device *adev, int inst_idx) +{ + uint32_t tmp; + int vcn_inst; + + vcn_inst = GET_INST(VCN, inst_idx); + + /* Wait for power status to be 1 */ + SOC15_WAIT_ON_RREG(VCN, vcn_inst, regUVD_POWER_STATUS, 1, + UVD_POWER_STATUS__UVD_POWER_STATUS_MASK); + + /* wait for read ptr to be equal to write ptr */ + tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_RB_WPTR); + SOC15_WAIT_ON_RREG(VCN, vcn_inst, regUVD_RB_RPTR, tmp, 0xFFFFFFFF); + + SOC15_WAIT_ON_RREG(VCN, vcn_inst, regUVD_POWER_STATUS, 1, + UVD_POWER_STATUS__UVD_POWER_STATUS_MASK); + + /* disable dynamic power gating mode */ + WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_POWER_STATUS), 0, + ~UVD_POWER_STATUS__UVD_PG_MODE_MASK); + return 0; +} + +/** + * vcn_v4_0_3_stop - VCN stop + * + * @adev: amdgpu_device pointer + * + * Stop VCN block + */ +static int vcn_v4_0_3_stop(struct amdgpu_device *adev) +{ + volatile struct amdgpu_vcn4_fw_shared *fw_shared; + int i, r = 0, vcn_inst; + uint32_t tmp; + + for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { + vcn_inst = GET_INST(VCN, i); + + fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr; + fw_shared->sq.queue_mode |= FW_QUEUE_DPG_HOLD_OFF; + + if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) { + vcn_v4_0_3_stop_dpg_mode(adev, i); + continue; + } + + /* wait for vcn idle */ + r = SOC15_WAIT_ON_RREG(VCN, vcn_inst, regUVD_STATUS, + UVD_STATUS__IDLE, 0x7); + if (r) + goto Done; + + tmp = UVD_LMI_STATUS__VCPU_LMI_WRITE_CLEAN_MASK | + UVD_LMI_STATUS__READ_CLEAN_MASK | + UVD_LMI_STATUS__WRITE_CLEAN_MASK | + UVD_LMI_STATUS__WRITE_CLEAN_RAW_MASK; + r = SOC15_WAIT_ON_RREG(VCN, vcn_inst, regUVD_LMI_STATUS, tmp, + tmp); + if (r) + goto Done; + + /* stall UMC channel */ + tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_LMI_CTRL2); + tmp |= UVD_LMI_CTRL2__STALL_ARB_UMC_MASK; + WREG32_SOC15(VCN, vcn_inst, regUVD_LMI_CTRL2, tmp); + tmp = UVD_LMI_STATUS__UMC_READ_CLEAN_RAW_MASK | + UVD_LMI_STATUS__UMC_WRITE_CLEAN_RAW_MASK; + r = SOC15_WAIT_ON_RREG(VCN, vcn_inst, regUVD_LMI_STATUS, tmp, + tmp); + if (r) + goto Done; + + /* Unblock VCPU Register access */ + WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_RB_ARB_CTRL), + UVD_RB_ARB_CTRL__VCPU_DIS_MASK, + ~UVD_RB_ARB_CTRL__VCPU_DIS_MASK); + + /* release VCPU reset to boot */ + WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_VCPU_CNTL), + UVD_VCPU_CNTL__BLK_RST_MASK, + ~UVD_VCPU_CNTL__BLK_RST_MASK); + + /* disable VCPU clock */ + WREG32_P(SOC15_REG_OFFSET(VCN, vcn_inst, regUVD_VCPU_CNTL), 0, + ~(UVD_VCPU_CNTL__CLK_EN_MASK)); + + /* reset LMI UMC/LMI/VCPU */ + tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_SOFT_RESET); + tmp |= UVD_SOFT_RESET__LMI_UMC_SOFT_RESET_MASK; + WREG32_SOC15(VCN, vcn_inst, regUVD_SOFT_RESET, tmp); + + tmp = RREG32_SOC15(VCN, vcn_inst, regUVD_SOFT_RESET); + tmp |= UVD_SOFT_RESET__LMI_SOFT_RESET_MASK; + WREG32_SOC15(VCN, vcn_inst, regUVD_SOFT_RESET, tmp); + + /* clear VCN status */ + WREG32_SOC15(VCN, vcn_inst, regUVD_STATUS, 0); + + /* apply HW clock gating */ + vcn_v4_0_3_enable_clock_gating(adev, i); + } +Done: + if (adev->pm.dpm_enabled) + amdgpu_dpm_enable_uvd(adev, false); + + return 0; +} + +/** + * vcn_v4_0_3_pause_dpg_mode - VCN pause with dpg mode + * + * @adev: amdgpu_device pointer + * @inst_idx: instance number index + * @new_state: pause state + * + * Pause dpg mode for VCN block + */ +static int vcn_v4_0_3_pause_dpg_mode(struct amdgpu_device *adev, int inst_idx, + struct dpg_pause_state *new_state) +{ + + return 0; +} + +/** + * vcn_v4_0_3_unified_ring_get_rptr - get unified read pointer + * + * @ring: amdgpu_ring pointer + * + * Returns the current hardware unified read pointer + */ +static uint64_t vcn_v4_0_3_unified_ring_get_rptr(struct amdgpu_ring *ring) +{ + struct amdgpu_device *adev = ring->adev; + + if (ring != &adev->vcn.inst[ring->me].ring_enc[0]) + DRM_ERROR("wrong ring id is identified in %s", __func__); + + return RREG32_SOC15(VCN, GET_INST(VCN, ring->me), regUVD_RB_RPTR); +} + +/** + * vcn_v4_0_3_unified_ring_get_wptr - get unified write pointer + * + * @ring: amdgpu_ring pointer + * + * Returns the current hardware unified write pointer + */ +static uint64_t vcn_v4_0_3_unified_ring_get_wptr(struct amdgpu_ring *ring) +{ + struct amdgpu_device *adev = ring->adev; + + if (ring != &adev->vcn.inst[ring->me].ring_enc[0]) + DRM_ERROR("wrong ring id is identified in %s", __func__); + + if (ring->use_doorbell) + return *ring->wptr_cpu_addr; + else + return RREG32_SOC15(VCN, GET_INST(VCN, ring->me), + regUVD_RB_WPTR); +} + +/** + * vcn_v4_0_3_unified_ring_set_wptr - set enc write pointer + * + * @ring: amdgpu_ring pointer + * + * Commits the enc write pointer to the hardware + */ +static void vcn_v4_0_3_unified_ring_set_wptr(struct amdgpu_ring *ring) +{ + struct amdgpu_device *adev = ring->adev; + + if (ring != &adev->vcn.inst[ring->me].ring_enc[0]) + DRM_ERROR("wrong ring id is identified in %s", __func__); + + if (ring->use_doorbell) { + *ring->wptr_cpu_addr = lower_32_bits(ring->wptr); + WDOORBELL32(ring->doorbell_index, lower_32_bits(ring->wptr)); + } else { + WREG32_SOC15(VCN, GET_INST(VCN, ring->me), regUVD_RB_WPTR, + lower_32_bits(ring->wptr)); + } +} + +static const struct amdgpu_ring_funcs vcn_v4_0_3_unified_ring_vm_funcs = { + .type = AMDGPU_RING_TYPE_VCN_ENC, + .align_mask = 0x3f, + .nop = VCN_ENC_CMD_NO_OP, + .get_rptr = vcn_v4_0_3_unified_ring_get_rptr, + .get_wptr = vcn_v4_0_3_unified_ring_get_wptr, + .set_wptr = vcn_v4_0_3_unified_ring_set_wptr, + .emit_frame_size = + SOC15_FLUSH_GPU_TLB_NUM_WREG * 3 + + SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 4 + + 4 + /* vcn_v2_0_enc_ring_emit_vm_flush */ + 5 + 5 + /* vcn_v2_0_enc_ring_emit_fence x2 vm fence */ + 1, /* vcn_v2_0_enc_ring_insert_end */ + .emit_ib_size = 5, /* vcn_v2_0_enc_ring_emit_ib */ + .emit_ib = vcn_v2_0_enc_ring_emit_ib, + .emit_fence = vcn_v2_0_enc_ring_emit_fence, + .emit_vm_flush = vcn_v2_0_enc_ring_emit_vm_flush, + .test_ring = amdgpu_vcn_enc_ring_test_ring, + .test_ib = amdgpu_vcn_unified_ring_test_ib, + .insert_nop = amdgpu_ring_insert_nop, + .insert_end = vcn_v2_0_enc_ring_insert_end, + .pad_ib = amdgpu_ring_generic_pad_ib, + .begin_use = amdgpu_vcn_ring_begin_use, + .end_use = amdgpu_vcn_ring_end_use, + .emit_wreg = vcn_v2_0_enc_ring_emit_wreg, + .emit_reg_wait = vcn_v2_0_enc_ring_emit_reg_wait, + .emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper, +}; + +/** + * vcn_v4_0_3_set_unified_ring_funcs - set unified ring functions + * + * @adev: amdgpu_device pointer + * + * Set unified ring functions + */ +static void vcn_v4_0_3_set_unified_ring_funcs(struct amdgpu_device *adev) +{ + int i, vcn_inst; + + for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { + adev->vcn.inst[i].ring_enc[0].funcs = &vcn_v4_0_3_unified_ring_vm_funcs; + adev->vcn.inst[i].ring_enc[0].me = i; + vcn_inst = GET_INST(VCN, i); + adev->vcn.inst[i].aid_id = + vcn_inst / adev->vcn.num_inst_per_aid; + } + DRM_DEV_INFO(adev->dev, "VCN decode is enabled in VM mode\n"); +} + +/** + * vcn_v4_0_3_is_idle - check VCN block is idle + * + * @handle: amdgpu_device pointer + * + * Check whether VCN block is idle + */ +static bool vcn_v4_0_3_is_idle(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int i, ret = 1; + + for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { + ret &= (RREG32_SOC15(VCN, GET_INST(VCN, i), regUVD_STATUS) == + UVD_STATUS__IDLE); + } + + return ret; +} + +/** + * vcn_v4_0_3_wait_for_idle - wait for VCN block idle + * + * @handle: amdgpu_device pointer + * + * Wait for VCN block idle + */ +static int vcn_v4_0_3_wait_for_idle(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int i, ret = 0; + + for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { + ret = SOC15_WAIT_ON_RREG(VCN, GET_INST(VCN, i), regUVD_STATUS, + UVD_STATUS__IDLE, UVD_STATUS__IDLE); + if (ret) + return ret; + } + + return ret; +} + +/* vcn_v4_0_3_set_clockgating_state - set VCN block clockgating state + * + * @handle: amdgpu_device pointer + * @state: clock gating state + * + * Set VCN block clockgating state + */ +static int vcn_v4_0_3_set_clockgating_state(void *handle, + enum amd_clockgating_state state) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + bool enable = (state == AMD_CG_STATE_GATE) ? true : false; + int i; + + for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { + if (enable) { + if (RREG32_SOC15(VCN, GET_INST(VCN, i), + regUVD_STATUS) != UVD_STATUS__IDLE) + return -EBUSY; + vcn_v4_0_3_enable_clock_gating(adev, i); + } else { + vcn_v4_0_3_disable_clock_gating(adev, i); + } + } + return 0; +} + +/** + * vcn_v4_0_3_set_powergating_state - set VCN block powergating state + * + * @handle: amdgpu_device pointer + * @state: power gating state + * + * Set VCN block powergating state + */ +static int vcn_v4_0_3_set_powergating_state(void *handle, + enum amd_powergating_state state) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int ret; + + if (state == adev->vcn.cur_state) + return 0; + + if (state == AMD_PG_STATE_GATE) + ret = vcn_v4_0_3_stop(adev); + else + ret = vcn_v4_0_3_start(adev); + + if (!ret) + adev->vcn.cur_state = state; + + return ret; +} + +/** + * vcn_v4_0_3_set_interrupt_state - set VCN block interrupt state + * + * @adev: amdgpu_device pointer + * @source: interrupt sources + * @type: interrupt types + * @state: interrupt states + * + * Set VCN block interrupt state + */ +static int vcn_v4_0_3_set_interrupt_state(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + unsigned int type, + enum amdgpu_interrupt_state state) +{ + return 0; +} + +/** + * vcn_v4_0_3_process_interrupt - process VCN block interrupt + * + * @adev: amdgpu_device pointer + * @source: interrupt sources + * @entry: interrupt entry from clients and sources + * + * Process VCN block interrupt + */ +static int vcn_v4_0_3_process_interrupt(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + struct amdgpu_iv_entry *entry) +{ + uint32_t i, inst; + + i = node_id_to_phys_map[entry->node_id]; + + DRM_DEV_DEBUG(adev->dev, "IH: VCN TRAP\n"); + + for (inst = 0; inst < adev->vcn.num_vcn_inst; ++inst) + if (adev->vcn.inst[inst].aid_id == i) + break; + + if (inst >= adev->vcn.num_vcn_inst) { + dev_WARN_ONCE(adev->dev, 1, + "Interrupt received for unknown VCN instance %d", + entry->node_id); + return 0; + } + + switch (entry->src_id) { + case VCN_4_0__SRCID__UVD_ENC_GENERAL_PURPOSE: + amdgpu_fence_process(&adev->vcn.inst[inst].ring_enc[0]); + break; + default: + DRM_DEV_ERROR(adev->dev, "Unhandled interrupt: %d %d\n", + entry->src_id, entry->src_data[0]); + break; + } + + return 0; +} + +static const struct amdgpu_irq_src_funcs vcn_v4_0_3_irq_funcs = { + .set = vcn_v4_0_3_set_interrupt_state, + .process = vcn_v4_0_3_process_interrupt, +}; + +/** + * vcn_v4_0_3_set_irq_funcs - set VCN block interrupt irq functions + * + * @adev: amdgpu_device pointer + * + * Set VCN block interrupt irq functions + */ +static void vcn_v4_0_3_set_irq_funcs(struct amdgpu_device *adev) +{ + int i; + + for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { + adev->vcn.inst->irq.num_types++; + } + adev->vcn.inst->irq.funcs = &vcn_v4_0_3_irq_funcs; +} + +static const struct amd_ip_funcs vcn_v4_0_3_ip_funcs = { + .name = "vcn_v4_0_3", + .early_init = vcn_v4_0_3_early_init, + .late_init = NULL, + .sw_init = vcn_v4_0_3_sw_init, + .sw_fini = vcn_v4_0_3_sw_fini, + .hw_init = vcn_v4_0_3_hw_init, + .hw_fini = vcn_v4_0_3_hw_fini, + .suspend = vcn_v4_0_3_suspend, + .resume = vcn_v4_0_3_resume, + .is_idle = vcn_v4_0_3_is_idle, + .wait_for_idle = vcn_v4_0_3_wait_for_idle, + .check_soft_reset = NULL, + .pre_soft_reset = NULL, + .soft_reset = NULL, + .post_soft_reset = NULL, + .set_clockgating_state = vcn_v4_0_3_set_clockgating_state, + .set_powergating_state = vcn_v4_0_3_set_powergating_state, +}; + +const struct amdgpu_ip_block_version vcn_v4_0_3_ip_block = { + .type = AMD_IP_BLOCK_TYPE_VCN, + .major = 4, + .minor = 0, + .rev = 3, + .funcs = &vcn_v4_0_3_ip_funcs, +}; + +static const struct amdgpu_ras_err_status_reg_entry vcn_v4_0_3_ue_reg_list[] = { + {AMDGPU_RAS_REG_ENTRY(VCN, 0, regVCN_UE_ERR_STATUS_LO_VIDD, regVCN_UE_ERR_STATUS_HI_VIDD), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "VIDD"}, + {AMDGPU_RAS_REG_ENTRY(VCN, 0, regVCN_UE_ERR_STATUS_LO_VIDV, regVCN_UE_ERR_STATUS_HI_VIDV), + 1, (AMDGPU_RAS_ERR_INFO_VALID | AMDGPU_RAS_ERR_STATUS_VALID), "VIDV"}, +}; + +static void vcn_v4_0_3_inst_query_ras_error_count(struct amdgpu_device *adev, + uint32_t vcn_inst, + void *ras_err_status) +{ + struct ras_err_data *err_data = (struct ras_err_data *)ras_err_status; + + /* vcn v4_0_3 only support query uncorrectable errors */ + amdgpu_ras_inst_query_ras_error_count(adev, + vcn_v4_0_3_ue_reg_list, + ARRAY_SIZE(vcn_v4_0_3_ue_reg_list), + NULL, 0, GET_INST(VCN, vcn_inst), + AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE, + &err_data->ue_count); +} + +static void vcn_v4_0_3_query_ras_error_count(struct amdgpu_device *adev, + void *ras_err_status) +{ + uint32_t i; + + if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN)) { + dev_warn(adev->dev, "VCN RAS is not supported\n"); + return; + } + + for (i = 0; i < adev->vcn.num_vcn_inst; i++) + vcn_v4_0_3_inst_query_ras_error_count(adev, i, ras_err_status); +} + +static void vcn_v4_0_3_inst_reset_ras_error_count(struct amdgpu_device *adev, + uint32_t vcn_inst) +{ + amdgpu_ras_inst_reset_ras_error_count(adev, + vcn_v4_0_3_ue_reg_list, + ARRAY_SIZE(vcn_v4_0_3_ue_reg_list), + GET_INST(VCN, vcn_inst)); +} + +static void vcn_v4_0_3_reset_ras_error_count(struct amdgpu_device *adev) +{ + uint32_t i; + + if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN)) { + dev_warn(adev->dev, "VCN RAS is not supported\n"); + return; + } + + for (i = 0; i < adev->vcn.num_vcn_inst; i++) + vcn_v4_0_3_inst_reset_ras_error_count(adev, i); +} + +static const struct amdgpu_ras_block_hw_ops vcn_v4_0_3_ras_hw_ops = { + .query_ras_error_count = vcn_v4_0_3_query_ras_error_count, + .reset_ras_error_count = vcn_v4_0_3_reset_ras_error_count, +}; + +static struct amdgpu_vcn_ras vcn_v4_0_3_ras = { + .ras_block = { + .hw_ops = &vcn_v4_0_3_ras_hw_ops, + }, +}; + +static void vcn_v4_0_3_set_ras_funcs(struct amdgpu_device *adev) +{ + adev->vcn.ras = &vcn_v4_0_3_ras; +} + +static void vcn_v4_0_3_enable_ras(struct amdgpu_device *adev, + int inst_idx, bool indirect) +{ + uint32_t tmp; + + if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN)) + return; + + tmp = VCN_RAS_CNTL__VCPU_VCODEC_REARM_MASK | + VCN_RAS_CNTL__VCPU_VCODEC_IH_EN_MASK | + VCN_RAS_CNTL__VCPU_VCODEC_PMI_EN_MASK | + VCN_RAS_CNTL__VCPU_VCODEC_STALL_EN_MASK; + WREG32_SOC15_DPG_MODE(inst_idx, + SOC15_DPG_MODE_OFFSET(VCN, 0, regVCN_RAS_CNTL), + tmp, 0, indirect); + + tmp = UVD_SYS_INT_EN__RASCNTL_VCPU_VCODEC_EN_MASK; + WREG32_SOC15_DPG_MODE(inst_idx, + SOC15_DPG_MODE_OFFSET(VCN, 0, regUVD_SYS_INT_EN), + tmp, 0, indirect); +} diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.h b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.h new file mode 100644 index 0000000000000000000000000000000000000000..0b046114373ae2798da83284b9f7074e800bff5c --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.h @@ -0,0 +1,29 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef __VCN_V4_0_3_H__ +#define __VCN_V4_0_3_H__ + +extern const struct amdgpu_ip_block_version vcn_v4_0_3_ip_block; + +#endif /* __VCN_V4_0_3_H__ */ diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c index 536128447b71d6e82fd26ff84ab749ad75963731..4d719df376a7224dba72c2cf880cd18e5e8d2044 100644 --- a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c @@ -334,7 +334,8 @@ static int vega20_ih_irq_init(struct amdgpu_device *adev) vega20_setup_retry_doorbell(adev->irq.retry_cam_doorbell_index)); /* Enable IH Retry CAM */ - if (adev->ip_versions[OSSSYS_HWIP][0] == IP_VERSION(4, 4, 0)) + if (adev->ip_versions[OSSSYS_HWIP][0] == IP_VERSION(4, 4, 0) || + adev->ip_versions[OSSSYS_HWIP][0] == IP_VERSION(4, 4, 2)) WREG32_FIELD15(OSSSYS, 0, IH_RETRY_INT_CAM_CNTL_ALDEBARAN, ENABLE, 1); else @@ -526,6 +527,7 @@ static int vega20_ih_early_init(void *handle) static int vega20_ih_sw_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + bool use_bus_addr = true; int r; r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_IH, 0, @@ -533,14 +535,18 @@ static int vega20_ih_sw_init(void *handle) if (r) return r; - r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 256 * 1024, true); + if ((adev->flags & AMD_IS_APU) && + (adev->ip_versions[OSSSYS_HWIP][0] == IP_VERSION(4, 4, 2))) + use_bus_addr = false; + + r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 256 * 1024, use_bus_addr); if (r) return r; adev->irq.ih.use_doorbell = true; adev->irq.ih.doorbell_index = adev->doorbell_index.ih << 1; - r = amdgpu_ih_ring_init(adev, &adev->irq.ih1, PAGE_SIZE, true); + r = amdgpu_ih_ring_init(adev, &adev->irq.ih1, PAGE_SIZE, use_bus_addr); if (r) return r; @@ -559,7 +565,7 @@ static int vega20_ih_sw_init(void *handle) /* initialize ih control registers offset */ vega20_ih_init_register_offset(adev); - r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, PAGE_SIZE, true); + r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, PAGE_SIZE, use_bus_addr); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c index 531f173ade2d62c1f1ced0ef3a83596b181efb87..6a8494f98d3ef488659479270b8a2a5464585868 100644 --- a/drivers/gpu/drm/amd/amdgpu/vi.c +++ b/drivers/gpu/drm/amd/amdgpu/vi.c @@ -542,8 +542,15 @@ static u32 vi_get_xclk(struct amdgpu_device *adev) u32 reference_clock = adev->clock.spll.reference_freq; u32 tmp; - if (adev->flags & AMD_IS_APU) - return reference_clock; + if (adev->flags & AMD_IS_APU) { + switch (adev->asic_type) { + case CHIP_STONEY: + /* vbios says 48Mhz, but the actual freq is 100Mhz */ + return 10000; + default: + return reference_clock; + } + } tmp = RREG32_SMC(ixCG_CLKPIN_CNTL_2); if (REG_GET_FIELD(tmp, CG_CLKPIN_CNTL_2, MUX_TCLK_TO_XCLK)) @@ -580,11 +587,6 @@ void vi_srbm_select(struct amdgpu_device *adev, WREG32(mmSRBM_GFX_CNTL, srbm_gfx_cntl); } -static void vi_vga_set_state(struct amdgpu_device *adev, bool state) -{ - /* todo */ -} - static bool vi_read_disabled_bios(struct amdgpu_device *adev) { u32 bus_cntl; @@ -762,12 +764,12 @@ static uint32_t vi_get_register_value(struct amdgpu_device *adev, mutex_lock(&adev->grbm_idx_mutex); if (se_num != 0xffffffff || sh_num != 0xffffffff) - amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff, 0); val = RREG32(reg_offset); if (se_num != 0xffffffff || sh_num != 0xffffffff) - amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); + amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, 0); mutex_unlock(&adev->grbm_idx_mutex); return val; } else { @@ -1435,7 +1437,6 @@ static const struct amdgpu_asic_funcs vi_asic_funcs = .read_register = &vi_read_register, .reset = &vi_asic_reset, .reset_method = &vi_asic_reset_method, - .set_vga_state = &vi_vga_set_state, .get_xclk = &vi_get_xclk, .set_uvd_clocks = &vi_set_uvd_clocks, .set_vce_clocks = &vi_set_vce_clocks, diff --git a/drivers/gpu/drm/amd/amdkfd/Makefile b/drivers/gpu/drm/amd/amdkfd/Makefile index e758c2a24cd08c4ff21bf2339bcd4a6f54e639e5..2ec8f27c5366cac1a9fac0936c439bb4545184d3 100644 --- a/drivers/gpu/drm/amd/amdkfd/Makefile +++ b/drivers/gpu/drm/amd/amdkfd/Makefile @@ -53,9 +53,11 @@ AMDKFD_FILES := $(AMDKFD_PATH)/kfd_module.o \ $(AMDKFD_PATH)/kfd_events.o \ $(AMDKFD_PATH)/cik_event_interrupt.o \ $(AMDKFD_PATH)/kfd_int_process_v9.o \ + $(AMDKFD_PATH)/kfd_int_process_v10.o \ $(AMDKFD_PATH)/kfd_int_process_v11.o \ $(AMDKFD_PATH)/kfd_smi_events.o \ - $(AMDKFD_PATH)/kfd_crat.o + $(AMDKFD_PATH)/kfd_crat.o \ + $(AMDKFD_PATH)/kfd_debug.o ifneq ($(CONFIG_AMD_IOMMU_V2),) AMDKFD_FILES += $(AMDKFD_PATH)/kfd_iommu.o diff --git a/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c b/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c index 5c8023cba1961f40a78f22de007031090fa84efd..795382b55e0a916cf61207622a994ac55c7971b1 100644 --- a/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c +++ b/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c @@ -26,7 +26,7 @@ #include "amdgpu_amdkfd.h" #include "kfd_smi_events.h" -static bool cik_event_interrupt_isr(struct kfd_dev *dev, +static bool cik_event_interrupt_isr(struct kfd_node *dev, const uint32_t *ih_ring_entry, uint32_t *patched_ihre, bool *patched_flag) @@ -85,7 +85,7 @@ static bool cik_event_interrupt_isr(struct kfd_dev *dev, !amdgpu_no_queue_eviction_on_vm_fault); } -static void cik_event_interrupt_wq(struct kfd_dev *dev, +static void cik_event_interrupt_wq(struct kfd_node *dev, const uint32_t *ih_ring_entry) { const struct cik_ih_ring_entry *ihre = @@ -118,9 +118,9 @@ static void cik_event_interrupt_wq(struct kfd_dev *dev, return; if (info.vmid == vmid) - kfd_signal_vm_fault_event(dev, pasid, &info); + kfd_signal_vm_fault_event(dev, pasid, &info, NULL); else - kfd_signal_vm_fault_event(dev, pasid, NULL); + kfd_signal_vm_fault_event(dev, pasid, NULL, NULL); } } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index 1b54a9aaae70cca9111cfe8d37cd2355c62f3dc7..6a27b000a246e7d04dca0a8d42b92a5a0ff96d8e 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -44,6 +44,7 @@ #include "amdgpu_amdkfd.h" #include "kfd_smi_events.h" #include "amdgpu_dma_buf.h" +#include "kfd_debug.h" static long kfd_ioctl(struct file *, unsigned int, unsigned long); static int kfd_open(struct inode *, struct file *); @@ -142,15 +143,13 @@ static int kfd_open(struct inode *inode, struct file *filep) return -EPERM; } - process = kfd_create_process(filep); + process = kfd_create_process(current); if (IS_ERR(process)) return PTR_ERR(process); - if (kfd_is_locked()) { - dev_dbg(kfd_device, "kfd is locked!\n" - "process %d unreferenced", process->pasid); + if (kfd_process_init_cwsr_apu(process, filep)) { kfd_unref_process(process); - return -EAGAIN; + return -EFAULT; } /* filep now owns the reference returned by kfd_create_process */ @@ -186,7 +185,12 @@ static int kfd_ioctl_get_version(struct file *filep, struct kfd_process *p, static int set_queue_properties_from_user(struct queue_properties *q_properties, struct kfd_ioctl_create_queue_args *args) { - if (args->queue_percentage > KFD_MAX_QUEUE_PERCENTAGE) { + /* + * Repurpose queue percentage to accommodate new features: + * bit 0-7: queue percentage + * bit 8-15: pm4_target_xcc + */ + if ((args->queue_percentage & 0xFF) > KFD_MAX_QUEUE_PERCENTAGE) { pr_err("Queue percentage must be between 0 to KFD_MAX_QUEUE_PERCENTAGE\n"); return -EINVAL; } @@ -236,7 +240,9 @@ static int set_queue_properties_from_user(struct queue_properties *q_properties, q_properties->is_interop = false; q_properties->is_gws = false; - q_properties->queue_percent = args->queue_percentage; + q_properties->queue_percent = args->queue_percentage & 0xFF; + /* bit 8-15 are repurposed to be PM4 target XCC */ + q_properties->pm4_target_xcc = (args->queue_percentage >> 8) & 0xFF; q_properties->priority = args->queue_priority; q_properties->queue_address = args->ring_base_address; q_properties->queue_size = args->ring_size; @@ -293,7 +299,7 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p, void *data) { struct kfd_ioctl_create_queue_args *args = data; - struct kfd_dev *dev; + struct kfd_node *dev; int err = 0; unsigned int queue_id; struct kfd_process_device *pdd; @@ -328,7 +334,7 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p, } if (!pdd->doorbell_index && - kfd_alloc_process_doorbells(dev, &pdd->doorbell_index) < 0) { + kfd_alloc_process_doorbells(dev->kfd, &pdd->doorbell_index) < 0) { err = -ENOMEM; goto err_alloc_doorbells; } @@ -336,7 +342,7 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p, /* Starting with GFX11, wptr BOs must be mapped to GART for MES to determine work * on unmapped queues for usermode queue oversubscription (no aggregated doorbell) */ - if (dev->shared_resources.enable_mes && + if (dev->kfd->shared_resources.enable_mes && ((dev->adev->mes.sched_version & AMDGPU_MES_API_VERSION_MASK) >> AMDGPU_MES_API_VERSION_SHIFT) >= 2) { struct amdgpu_bo_va_mapping *wptr_mapping; @@ -404,6 +410,7 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p, pr_debug("Write ptr address == 0x%016llX\n", args->write_pointer_address); + kfd_dbg_ev_raise(KFD_EC_MASK(EC_QUEUE_NEW), p, dev, queue_id, false, NULL, 0); return 0; err_create_queue: @@ -442,7 +449,12 @@ static int kfd_ioctl_update_queue(struct file *filp, struct kfd_process *p, struct kfd_ioctl_update_queue_args *args = data; struct queue_properties properties; - if (args->queue_percentage > KFD_MAX_QUEUE_PERCENTAGE) { + /* + * Repurpose queue percentage to accommodate new features: + * bit 0-7: queue percentage + * bit 8-15: pm4_target_xcc + */ + if ((args->queue_percentage & 0xFF) > KFD_MAX_QUEUE_PERCENTAGE) { pr_err("Queue percentage must be between 0 to KFD_MAX_QUEUE_PERCENTAGE\n"); return -EINVAL; } @@ -466,7 +478,9 @@ static int kfd_ioctl_update_queue(struct file *filp, struct kfd_process *p, properties.queue_address = args->ring_base_address; properties.queue_size = args->ring_size; - properties.queue_percent = args->queue_percentage; + properties.queue_percent = args->queue_percentage & 0xFF; + /* bit 8-15 are repurposed to be PM4 target XCC */ + properties.pm4_target_xcc = (args->queue_percentage >> 8) & 0xFF; properties.priority = args->queue_priority; pr_debug("Updating queue id %d for pasid 0x%x\n", @@ -524,8 +538,6 @@ static int kfd_ioctl_set_cu_mask(struct file *filp, struct kfd_process *p, goto out; } - minfo.update_flag = UPDATE_FLAG_CU_MASK; - mutex_lock(&p->mutex); retval = pqm_update_mqd(&p->pqm, args->queue_id, &minfo); @@ -887,7 +899,7 @@ static int kfd_ioctl_set_scratch_backing_va(struct file *filep, { struct kfd_ioctl_set_scratch_backing_va_args *args = data; struct kfd_process_device *pdd; - struct kfd_dev *dev; + struct kfd_node *dev; long err; mutex_lock(&p->mutex); @@ -1006,19 +1018,26 @@ static int kfd_ioctl_acquire_vm(struct file *filep, struct kfd_process *p, return ret; } -bool kfd_dev_is_large_bar(struct kfd_dev *dev) +bool kfd_dev_is_large_bar(struct kfd_node *dev) { if (debug_largebar) { pr_debug("Simulate large-bar allocation on non large-bar machine\n"); return true; } - if (dev->use_iommu_v2) + if (dev->kfd->use_iommu_v2) return false; if (dev->local_mem_info.local_mem_size_private == 0 && - dev->local_mem_info.local_mem_size_public > 0) + dev->local_mem_info.local_mem_size_public > 0) + return true; + + if (dev->local_mem_info.local_mem_size_public == 0 && + dev->kfd->adev->gmc.is_app_apu) { + pr_debug("APP APU, Consider like a large bar system\n"); return true; + } + return false; } @@ -1030,7 +1049,8 @@ static int kfd_ioctl_get_available_memory(struct file *filep, if (!pdd) return -EINVAL; - args->available = amdgpu_amdkfd_get_available_memory(pdd->dev->adev); + args->available = amdgpu_amdkfd_get_available_memory(pdd->dev->adev, + pdd->dev->node_id); kfd_unlock_pdd(pdd); return 0; } @@ -1041,7 +1061,7 @@ static int kfd_ioctl_alloc_memory_of_gpu(struct file *filep, struct kfd_ioctl_alloc_memory_of_gpu_args *args = data; struct kfd_process_device *pdd; void *mem; - struct kfd_dev *dev; + struct kfd_node *dev; int idr_handle; long err; uint64_t offset = args->mmap_offset; @@ -1105,7 +1125,7 @@ static int kfd_ioctl_alloc_memory_of_gpu(struct file *filep, } if (flags & KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL) { - if (args->size != kfd_doorbell_process_slice(dev)) { + if (args->size != kfd_doorbell_process_slice(dev->kfd)) { err = -EINVAL; goto err_unlock; } @@ -1231,7 +1251,7 @@ static int kfd_ioctl_map_memory_to_gpu(struct file *filep, struct kfd_ioctl_map_memory_to_gpu_args *args = data; struct kfd_process_device *pdd, *peer_pdd; void *mem; - struct kfd_dev *dev; + struct kfd_node *dev; long err = 0; int i; uint32_t *devices_arr = NULL; @@ -1405,7 +1425,7 @@ static int kfd_ioctl_unmap_memory_from_gpu(struct file *filep, args->n_success = i+1; } - flush_tlb = kfd_flush_tlb_after_unmap(pdd->dev); + flush_tlb = kfd_flush_tlb_after_unmap(pdd->dev->kfd); if (flush_tlb) { err = amdgpu_amdkfd_gpuvm_sync_memory(pdd->dev->adev, (struct kgd_mem *) mem, true); @@ -1445,7 +1465,7 @@ static int kfd_ioctl_alloc_queue_gws(struct file *filep, int retval; struct kfd_ioctl_alloc_queue_gws_args *args = data; struct queue *q; - struct kfd_dev *dev; + struct kfd_node *dev; mutex_lock(&p->mutex); q = pqm_get_user_queue(&p->pqm, args->queue_id); @@ -1467,6 +1487,11 @@ static int kfd_ioctl_alloc_queue_gws(struct file *filep, goto out_unlock; } + if (!kfd_dbg_has_gws_support(dev) && p->debug_trap_enabled) { + retval = -EBUSY; + goto out_unlock; + } + retval = pqm_set_gws(&p->pqm, args->queue_id, args->num_gws ? dev->gws : NULL); mutex_unlock(&p->mutex); @@ -1482,10 +1507,11 @@ static int kfd_ioctl_get_dmabuf_info(struct file *filep, struct kfd_process *p, void *data) { struct kfd_ioctl_get_dmabuf_info_args *args = data; - struct kfd_dev *dev = NULL; + struct kfd_node *dev = NULL; struct amdgpu_device *dmabuf_adev; void *metadata_buffer = NULL; uint32_t flags; + int8_t xcp_id; unsigned int i; int r; @@ -1506,17 +1532,14 @@ static int kfd_ioctl_get_dmabuf_info(struct file *filep, r = amdgpu_amdkfd_get_dmabuf_info(dev->adev, args->dmabuf_fd, &dmabuf_adev, &args->size, metadata_buffer, args->metadata_size, - &args->metadata_size, &flags); + &args->metadata_size, &flags, &xcp_id); if (r) goto exit; - /* Reverse-lookup gpu_id from kgd pointer */ - dev = kfd_device_by_adev(dmabuf_adev); - if (!dev) { - r = -EINVAL; - goto exit; - } - args->gpu_id = dev->id; + if (xcp_id >= 0) + args->gpu_id = dmabuf_adev->kfd.dev->nodes[xcp_id]->id; + else + args->gpu_id = dmabuf_adev->kfd.dev->nodes[0]->id; args->flags = flags; /* Copy metadata buffer to user mode */ @@ -1596,7 +1619,7 @@ static int kfd_ioctl_export_dmabuf(struct file *filep, struct kfd_ioctl_export_dmabuf_args *args = data; struct kfd_process_device *pdd; struct dma_buf *dmabuf; - struct kfd_dev *dev; + struct kfd_node *dev; void *mem; int ret = 0; @@ -2178,7 +2201,7 @@ static int criu_restore_devices(struct kfd_process *p, } for (i = 0; i < args->num_devices; i++) { - struct kfd_dev *dev; + struct kfd_node *dev; struct kfd_process_device *pdd; struct file *drm_file; @@ -2240,7 +2263,7 @@ static int criu_restore_devices(struct kfd_process *p, } if (!pdd->doorbell_index && - kfd_alloc_process_doorbells(pdd->dev, &pdd->doorbell_index) < 0) { + kfd_alloc_process_doorbells(pdd->dev->kfd, &pdd->doorbell_index) < 0) { ret = -ENOMEM; goto exit; } @@ -2268,7 +2291,8 @@ static int criu_restore_memory_of_gpu(struct kfd_process_device *pdd, u64 offset; if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL) { - if (bo_bucket->size != kfd_doorbell_process_slice(pdd->dev)) + if (bo_bucket->size != + kfd_doorbell_process_slice(pdd->dev->kfd)) return -EINVAL; offset = kfd_get_process_doorbells(pdd); @@ -2350,7 +2374,7 @@ static int criu_restore_bo(struct kfd_process *p, /* now map these BOs to GPU/s */ for (j = 0; j < p->n_pdds; j++) { - struct kfd_dev *peer; + struct kfd_node *peer; struct kfd_process_device *peer_pdd; if (!bo_priv->mapped_gpuids[j]) @@ -2715,6 +2739,356 @@ static int kfd_ioctl_criu(struct file *filep, struct kfd_process *p, void *data) return ret; } +static int runtime_enable(struct kfd_process *p, uint64_t r_debug, + bool enable_ttmp_setup) +{ + int i = 0, ret = 0; + + if (p->is_runtime_retry) + goto retry; + + if (p->runtime_info.runtime_state != DEBUG_RUNTIME_STATE_DISABLED) + return -EBUSY; + + for (i = 0; i < p->n_pdds; i++) { + struct kfd_process_device *pdd = p->pdds[i]; + + if (pdd->qpd.queue_count) + return -EEXIST; + } + + p->runtime_info.runtime_state = DEBUG_RUNTIME_STATE_ENABLED; + p->runtime_info.r_debug = r_debug; + p->runtime_info.ttmp_setup = enable_ttmp_setup; + + if (p->runtime_info.ttmp_setup) { + for (i = 0; i < p->n_pdds; i++) { + struct kfd_process_device *pdd = p->pdds[i]; + + if (!kfd_dbg_is_rlc_restore_supported(pdd->dev)) { + amdgpu_gfx_off_ctrl(pdd->dev->adev, false); + pdd->dev->kfd2kgd->enable_debug_trap( + pdd->dev->adev, + true, + pdd->dev->vm_info.last_vmid_kfd); + } else if (kfd_dbg_is_per_vmid_supported(pdd->dev)) { + pdd->spi_dbg_override = pdd->dev->kfd2kgd->enable_debug_trap( + pdd->dev->adev, + false, + 0); + } + } + } + +retry: + if (p->debug_trap_enabled) { + if (!p->is_runtime_retry) { + kfd_dbg_trap_activate(p); + kfd_dbg_ev_raise(KFD_EC_MASK(EC_PROCESS_RUNTIME), + p, NULL, 0, false, NULL, 0); + } + + mutex_unlock(&p->mutex); + ret = down_interruptible(&p->runtime_enable_sema); + mutex_lock(&p->mutex); + + p->is_runtime_retry = !!ret; + } + + return ret; +} + +static int runtime_disable(struct kfd_process *p) +{ + int i = 0, ret; + bool was_enabled = p->runtime_info.runtime_state == DEBUG_RUNTIME_STATE_ENABLED; + + p->runtime_info.runtime_state = DEBUG_RUNTIME_STATE_DISABLED; + p->runtime_info.r_debug = 0; + + if (p->debug_trap_enabled) { + if (was_enabled) + kfd_dbg_trap_deactivate(p, false, 0); + + if (!p->is_runtime_retry) + kfd_dbg_ev_raise(KFD_EC_MASK(EC_PROCESS_RUNTIME), + p, NULL, 0, false, NULL, 0); + + mutex_unlock(&p->mutex); + ret = down_interruptible(&p->runtime_enable_sema); + mutex_lock(&p->mutex); + + p->is_runtime_retry = !!ret; + if (ret) + return ret; + } + + if (was_enabled && p->runtime_info.ttmp_setup) { + for (i = 0; i < p->n_pdds; i++) { + struct kfd_process_device *pdd = p->pdds[i]; + + if (!kfd_dbg_is_rlc_restore_supported(pdd->dev)) + amdgpu_gfx_off_ctrl(pdd->dev->adev, true); + } + } + + p->runtime_info.ttmp_setup = false; + + /* disable ttmp setup */ + for (i = 0; i < p->n_pdds; i++) { + struct kfd_process_device *pdd = p->pdds[i]; + + if (kfd_dbg_is_per_vmid_supported(pdd->dev)) { + pdd->spi_dbg_override = + pdd->dev->kfd2kgd->disable_debug_trap( + pdd->dev->adev, + false, + pdd->dev->vm_info.last_vmid_kfd); + + if (!pdd->dev->kfd->shared_resources.enable_mes) + debug_refresh_runlist(pdd->dev->dqm); + else + kfd_dbg_set_mes_debug_mode(pdd); + } + } + + return 0; +} + +static int kfd_ioctl_runtime_enable(struct file *filep, struct kfd_process *p, void *data) +{ + struct kfd_ioctl_runtime_enable_args *args = data; + int r; + + mutex_lock(&p->mutex); + + if (args->mode_mask & KFD_RUNTIME_ENABLE_MODE_ENABLE_MASK) + r = runtime_enable(p, args->r_debug, + !!(args->mode_mask & KFD_RUNTIME_ENABLE_MODE_TTMP_SAVE_MASK)); + else + r = runtime_disable(p); + + mutex_unlock(&p->mutex); + + return r; +} + +static int kfd_ioctl_set_debug_trap(struct file *filep, struct kfd_process *p, void *data) +{ + struct kfd_ioctl_dbg_trap_args *args = data; + struct task_struct *thread = NULL; + struct mm_struct *mm = NULL; + struct pid *pid = NULL; + struct kfd_process *target = NULL; + struct kfd_process_device *pdd = NULL; + int r = 0; + + if (sched_policy == KFD_SCHED_POLICY_NO_HWS) { + pr_err("Debugging does not support sched_policy %i", sched_policy); + return -EINVAL; + } + + pid = find_get_pid(args->pid); + if (!pid) { + pr_debug("Cannot find pid info for %i\n", args->pid); + r = -ESRCH; + goto out; + } + + thread = get_pid_task(pid, PIDTYPE_PID); + if (!thread) { + r = -ESRCH; + goto out; + } + + mm = get_task_mm(thread); + if (!mm) { + r = -ESRCH; + goto out; + } + + if (args->op == KFD_IOC_DBG_TRAP_ENABLE) { + bool create_process; + + rcu_read_lock(); + create_process = thread && thread != current && ptrace_parent(thread) == current; + rcu_read_unlock(); + + target = create_process ? kfd_create_process(thread) : + kfd_lookup_process_by_pid(pid); + } else { + target = kfd_lookup_process_by_pid(pid); + } + + if (IS_ERR_OR_NULL(target)) { + pr_debug("Cannot find process PID %i to debug\n", args->pid); + r = target ? PTR_ERR(target) : -ESRCH; + goto out; + } + + /* Check if target is still PTRACED. */ + rcu_read_lock(); + if (target != p && args->op != KFD_IOC_DBG_TRAP_DISABLE + && ptrace_parent(target->lead_thread) != current) { + pr_err("PID %i is not PTRACED and cannot be debugged\n", args->pid); + r = -EPERM; + } + rcu_read_unlock(); + + if (r) + goto out; + + mutex_lock(&target->mutex); + + if (args->op != KFD_IOC_DBG_TRAP_ENABLE && !target->debug_trap_enabled) { + pr_err("PID %i not debug enabled for op %i\n", args->pid, args->op); + r = -EINVAL; + goto unlock_out; + } + + if (target->runtime_info.runtime_state != DEBUG_RUNTIME_STATE_ENABLED && + (args->op == KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_OVERRIDE || + args->op == KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_MODE || + args->op == KFD_IOC_DBG_TRAP_SUSPEND_QUEUES || + args->op == KFD_IOC_DBG_TRAP_RESUME_QUEUES || + args->op == KFD_IOC_DBG_TRAP_SET_NODE_ADDRESS_WATCH || + args->op == KFD_IOC_DBG_TRAP_CLEAR_NODE_ADDRESS_WATCH || + args->op == KFD_IOC_DBG_TRAP_SET_FLAGS)) { + r = -EPERM; + goto unlock_out; + } + + if (args->op == KFD_IOC_DBG_TRAP_SET_NODE_ADDRESS_WATCH || + args->op == KFD_IOC_DBG_TRAP_CLEAR_NODE_ADDRESS_WATCH) { + int user_gpu_id = kfd_process_get_user_gpu_id(target, + args->op == KFD_IOC_DBG_TRAP_SET_NODE_ADDRESS_WATCH ? + args->set_node_address_watch.gpu_id : + args->clear_node_address_watch.gpu_id); + + pdd = kfd_process_device_data_by_id(target, user_gpu_id); + if (user_gpu_id == -EINVAL || !pdd) { + r = -ENODEV; + goto unlock_out; + } + } + + switch (args->op) { + case KFD_IOC_DBG_TRAP_ENABLE: + if (target != p) + target->debugger_process = p; + + r = kfd_dbg_trap_enable(target, + args->enable.dbg_fd, + (void __user *)args->enable.rinfo_ptr, + &args->enable.rinfo_size); + if (!r) + target->exception_enable_mask = args->enable.exception_mask; + + break; + case KFD_IOC_DBG_TRAP_DISABLE: + r = kfd_dbg_trap_disable(target); + break; + case KFD_IOC_DBG_TRAP_SEND_RUNTIME_EVENT: + r = kfd_dbg_send_exception_to_runtime(target, + args->send_runtime_event.gpu_id, + args->send_runtime_event.queue_id, + args->send_runtime_event.exception_mask); + break; + case KFD_IOC_DBG_TRAP_SET_EXCEPTIONS_ENABLED: + kfd_dbg_set_enabled_debug_exception_mask(target, + args->set_exceptions_enabled.exception_mask); + break; + case KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_OVERRIDE: + r = kfd_dbg_trap_set_wave_launch_override(target, + args->launch_override.override_mode, + args->launch_override.enable_mask, + args->launch_override.support_request_mask, + &args->launch_override.enable_mask, + &args->launch_override.support_request_mask); + break; + case KFD_IOC_DBG_TRAP_SET_WAVE_LAUNCH_MODE: + r = kfd_dbg_trap_set_wave_launch_mode(target, + args->launch_mode.launch_mode); + break; + case KFD_IOC_DBG_TRAP_SUSPEND_QUEUES: + r = suspend_queues(target, + args->suspend_queues.num_queues, + args->suspend_queues.grace_period, + args->suspend_queues.exception_mask, + (uint32_t *)args->suspend_queues.queue_array_ptr); + + break; + case KFD_IOC_DBG_TRAP_RESUME_QUEUES: + r = resume_queues(target, args->resume_queues.num_queues, + (uint32_t *)args->resume_queues.queue_array_ptr); + break; + case KFD_IOC_DBG_TRAP_SET_NODE_ADDRESS_WATCH: + r = kfd_dbg_trap_set_dev_address_watch(pdd, + args->set_node_address_watch.address, + args->set_node_address_watch.mask, + &args->set_node_address_watch.id, + args->set_node_address_watch.mode); + break; + case KFD_IOC_DBG_TRAP_CLEAR_NODE_ADDRESS_WATCH: + r = kfd_dbg_trap_clear_dev_address_watch(pdd, + args->clear_node_address_watch.id); + break; + case KFD_IOC_DBG_TRAP_SET_FLAGS: + r = kfd_dbg_trap_set_flags(target, &args->set_flags.flags); + break; + case KFD_IOC_DBG_TRAP_QUERY_DEBUG_EVENT: + r = kfd_dbg_ev_query_debug_event(target, + &args->query_debug_event.queue_id, + &args->query_debug_event.gpu_id, + args->query_debug_event.exception_mask, + &args->query_debug_event.exception_mask); + break; + case KFD_IOC_DBG_TRAP_QUERY_EXCEPTION_INFO: + r = kfd_dbg_trap_query_exception_info(target, + args->query_exception_info.source_id, + args->query_exception_info.exception_code, + args->query_exception_info.clear_exception, + (void __user *)args->query_exception_info.info_ptr, + &args->query_exception_info.info_size); + break; + case KFD_IOC_DBG_TRAP_GET_QUEUE_SNAPSHOT: + r = pqm_get_queue_snapshot(&target->pqm, + args->queue_snapshot.exception_mask, + (void __user *)args->queue_snapshot.snapshot_buf_ptr, + &args->queue_snapshot.num_queues, + &args->queue_snapshot.entry_size); + break; + case KFD_IOC_DBG_TRAP_GET_DEVICE_SNAPSHOT: + r = kfd_dbg_trap_device_snapshot(target, + args->device_snapshot.exception_mask, + (void __user *)args->device_snapshot.snapshot_buf_ptr, + &args->device_snapshot.num_devices, + &args->device_snapshot.entry_size); + break; + default: + pr_err("Invalid option: %i\n", args->op); + r = -EINVAL; + } + +unlock_out: + mutex_unlock(&target->mutex); + +out: + if (thread) + put_task_struct(thread); + + if (mm) + mmput(mm); + + if (pid) + put_pid(pid); + + if (target) + kfd_unref_process(target); + + return r; +} + #define AMDKFD_IOCTL_DEF(ioctl, _func, _flags) \ [_IOC_NR(ioctl)] = {.cmd = ioctl, .func = _func, .flags = _flags, \ .cmd_drv = 0, .name = #ioctl} @@ -2827,6 +3201,12 @@ static const struct amdkfd_ioctl_desc amdkfd_ioctls[] = { AMDKFD_IOCTL_DEF(AMDKFD_IOC_EXPORT_DMABUF, kfd_ioctl_export_dmabuf, 0), + + AMDKFD_IOCTL_DEF(AMDKFD_IOC_RUNTIME_ENABLE, + kfd_ioctl_runtime_enable, 0), + + AMDKFD_IOCTL_DEF(AMDKFD_IOC_DBG_TRAP, + kfd_ioctl_set_debug_trap, 0), }; #define AMDKFD_CORE_IOCTL_COUNT ARRAY_SIZE(amdkfd_ioctls) @@ -2947,7 +3327,7 @@ static long kfd_ioctl(struct file *filep, unsigned int cmd, unsigned long arg) return retcode; } -static int kfd_mmio_mmap(struct kfd_dev *dev, struct kfd_process *process, +static int kfd_mmio_mmap(struct kfd_node *dev, struct kfd_process *process, struct vm_area_struct *vma) { phys_addr_t address; @@ -2981,7 +3361,7 @@ static int kfd_mmio_mmap(struct kfd_dev *dev, struct kfd_process *process, static int kfd_mmap(struct file *filp, struct vm_area_struct *vma) { struct kfd_process *process; - struct kfd_dev *dev = NULL; + struct kfd_node *dev = NULL; unsigned long mmap_offset; unsigned int gpu_id; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c index 475e4702735407b52513a81edf4415ac7f9afd53..49f40d9f16e8652e79582d13d5bcc95e26f4a5ad 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c @@ -1166,7 +1166,7 @@ static int kfd_parse_subtype_iolink(struct crat_subtype_iolink *iolink, if (props->iolink_type == CRAT_IOLINK_TYPE_PCIEXPRESS) props->weight = 20; else if (props->iolink_type == CRAT_IOLINK_TYPE_XGMI) - props->weight = 15 * iolink->num_hops_xgmi; + props->weight = iolink->weight_xgmi; else props->weight = node_distance(id_from, id_to); @@ -1405,7 +1405,7 @@ static int kfd_fill_gpu_cache_info_from_gfx_config(struct kfd_dev *kdev, return i; } -int kfd_get_gpu_cache_info(struct kfd_dev *kdev, struct kfd_gpu_cache_info **pcache_info) +int kfd_get_gpu_cache_info(struct kfd_node *kdev, struct kfd_gpu_cache_info **pcache_info) { int num_of_cache_types = 0; @@ -1524,7 +1524,7 @@ int kfd_get_gpu_cache_info(struct kfd_dev *kdev, struct kfd_gpu_cache_info **pca case IP_VERSION(11, 0, 3): case IP_VERSION(11, 0, 4): num_of_cache_types = - kfd_fill_gpu_cache_info_from_gfx_config(kdev, *pcache_info); + kfd_fill_gpu_cache_info_from_gfx_config(kdev->kfd, *pcache_info); break; default: *pcache_info = dummy_cache_info; @@ -1858,7 +1858,7 @@ static int kfd_create_vcrat_image_cpu(void *pcrat_image, size_t *size) } static int kfd_fill_gpu_memory_affinity(int *avail_size, - struct kfd_dev *kdev, uint8_t type, uint64_t size, + struct kfd_node *kdev, uint8_t type, uint64_t size, struct crat_subtype_memory *sub_type_hdr, uint32_t proximity_domain, const struct kfd_local_mem_info *local_mem_info) @@ -1887,7 +1887,7 @@ static int kfd_fill_gpu_memory_affinity(int *avail_size, } #ifdef CONFIG_ACPI_NUMA -static void kfd_find_numa_node_in_srat(struct kfd_dev *kdev) +static void kfd_find_numa_node_in_srat(struct kfd_node *kdev) { struct acpi_table_header *table_header = NULL; struct acpi_subtable_header *sub_header = NULL; @@ -1972,6 +1972,9 @@ static void kfd_find_numa_node_in_srat(struct kfd_dev *kdev) } #endif +#define KFD_CRAT_INTRA_SOCKET_WEIGHT 13 +#define KFD_CRAT_XGMI_WEIGHT 15 + /* kfd_fill_gpu_direct_io_link - Fill in direct io link from GPU * to its NUMA node * @avail_size: Available size in the memory @@ -1982,7 +1985,7 @@ static void kfd_find_numa_node_in_srat(struct kfd_dev *kdev) * Return 0 if successful else return -ve value */ static int kfd_fill_gpu_direct_io_link_to_cpu(int *avail_size, - struct kfd_dev *kdev, + struct kfd_node *kdev, struct crat_subtype_iolink *sub_type_hdr, uint32_t proximity_domain) { @@ -2002,7 +2005,16 @@ static int kfd_fill_gpu_direct_io_link_to_cpu(int *avail_size, /* Fill in IOLINK subtype. * TODO: Fill-in other fields of iolink subtype */ - if (kdev->adev->gmc.xgmi.connected_to_cpu) { + if (kdev->adev->gmc.xgmi.connected_to_cpu || + (KFD_GC_VERSION(kdev) == IP_VERSION(9, 4, 3) && + kdev->adev->smuio.funcs->get_pkg_type(kdev->adev) == + AMDGPU_PKG_TYPE_APU)) { + bool ext_cpu = KFD_GC_VERSION(kdev) != IP_VERSION(9, 4, 3); + int mem_bw = 819200, weight = ext_cpu ? KFD_CRAT_XGMI_WEIGHT : + KFD_CRAT_INTRA_SOCKET_WEIGHT; + uint32_t bandwidth = ext_cpu ? amdgpu_amdkfd_get_xgmi_bandwidth_mbytes( + kdev->adev, NULL, true) : mem_bw; + /* * with host gpu xgmi link, host can access gpu memory whether * or not pcie bar type is large, so always create bidirectional @@ -2010,14 +2022,9 @@ static int kfd_fill_gpu_direct_io_link_to_cpu(int *avail_size, */ sub_type_hdr->flags |= CRAT_IOLINK_FLAGS_BI_DIRECTIONAL; sub_type_hdr->io_interface_type = CRAT_IOLINK_TYPE_XGMI; - sub_type_hdr->num_hops_xgmi = 1; - if (KFD_GC_VERSION(kdev) == IP_VERSION(9, 4, 2)) { - sub_type_hdr->minimum_bandwidth_mbs = - amdgpu_amdkfd_get_xgmi_bandwidth_mbytes( - kdev->adev, NULL, true); - sub_type_hdr->maximum_bandwidth_mbs = - sub_type_hdr->minimum_bandwidth_mbs; - } + sub_type_hdr->weight_xgmi = weight; + sub_type_hdr->minimum_bandwidth_mbs = bandwidth; + sub_type_hdr->maximum_bandwidth_mbs = bandwidth; } else { sub_type_hdr->io_interface_type = CRAT_IOLINK_TYPE_PCIEXPRESS; sub_type_hdr->minimum_bandwidth_mbs = @@ -2029,7 +2036,8 @@ static int kfd_fill_gpu_direct_io_link_to_cpu(int *avail_size, sub_type_hdr->proximity_domain_from = proximity_domain; #ifdef CONFIG_ACPI_NUMA - if (kdev->adev->pdev->dev.numa_node == NUMA_NO_NODE) + if (kdev->adev->pdev->dev.numa_node == NUMA_NO_NODE && + num_possible_nodes() > 1) kfd_find_numa_node_in_srat(kdev); #endif #ifdef CONFIG_NUMA @@ -2044,12 +2052,14 @@ static int kfd_fill_gpu_direct_io_link_to_cpu(int *avail_size, } static int kfd_fill_gpu_xgmi_link_to_gpu(int *avail_size, - struct kfd_dev *kdev, - struct kfd_dev *peer_kdev, + struct kfd_node *kdev, + struct kfd_node *peer_kdev, struct crat_subtype_iolink *sub_type_hdr, uint32_t proximity_domain_from, uint32_t proximity_domain_to) { + bool use_ta_info = kdev->kfd->num_nodes == 1; + *avail_size -= sizeof(struct crat_subtype_iolink); if (*avail_size < 0) return -ENOMEM; @@ -2064,12 +2074,25 @@ static int kfd_fill_gpu_xgmi_link_to_gpu(int *avail_size, sub_type_hdr->io_interface_type = CRAT_IOLINK_TYPE_XGMI; sub_type_hdr->proximity_domain_from = proximity_domain_from; sub_type_hdr->proximity_domain_to = proximity_domain_to; - sub_type_hdr->num_hops_xgmi = - amdgpu_amdkfd_get_xgmi_hops_count(kdev->adev, peer_kdev->adev); - sub_type_hdr->maximum_bandwidth_mbs = - amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(kdev->adev, peer_kdev->adev, false); - sub_type_hdr->minimum_bandwidth_mbs = sub_type_hdr->maximum_bandwidth_mbs ? - amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(kdev->adev, NULL, true) : 0; + + if (use_ta_info) { + sub_type_hdr->weight_xgmi = KFD_CRAT_XGMI_WEIGHT * + amdgpu_amdkfd_get_xgmi_hops_count(kdev->adev, peer_kdev->adev); + sub_type_hdr->maximum_bandwidth_mbs = + amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(kdev->adev, + peer_kdev->adev, false); + sub_type_hdr->minimum_bandwidth_mbs = sub_type_hdr->maximum_bandwidth_mbs ? + amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(kdev->adev, NULL, true) : 0; + } else { + bool is_single_hop = kdev->kfd == peer_kdev->kfd; + int weight = is_single_hop ? KFD_CRAT_INTRA_SOCKET_WEIGHT : + (2 * KFD_CRAT_INTRA_SOCKET_WEIGHT) + KFD_CRAT_XGMI_WEIGHT; + int mem_bw = 819200; + + sub_type_hdr->weight_xgmi = weight; + sub_type_hdr->maximum_bandwidth_mbs = is_single_hop ? mem_bw : 0; + sub_type_hdr->minimum_bandwidth_mbs = is_single_hop ? mem_bw : 0; + } return 0; } @@ -2081,7 +2104,7 @@ static int kfd_fill_gpu_xgmi_link_to_gpu(int *avail_size, * [OUT] actual size of data filled in crat_image */ static int kfd_create_vcrat_image_gpu(void *pcrat_image, - size_t *size, struct kfd_dev *kdev, + size_t *size, struct kfd_node *kdev, uint32_t proximity_domain) { struct crat_header *crat_table = (struct crat_header *)pcrat_image; @@ -2153,7 +2176,7 @@ static int kfd_create_vcrat_image_gpu(void *pcrat_image, /* Check if this node supports IOMMU. During parsing this flag will * translate to HSA_CAP_ATS_PRESENT */ - if (!kfd_iommu_check_device(kdev)) + if (!kfd_iommu_check_device(kdev->kfd)) cu->hsa_capability |= CRAT_CU_FLAGS_IOMMU_PRESENT; crat_table->length += sub_type_hdr->length; @@ -2216,12 +2239,12 @@ static int kfd_create_vcrat_image_gpu(void *pcrat_image, * (from other GPU to this GPU) will be added * in kfd_parse_subtype_iolink. */ - if (kdev->hive_id) { + if (kdev->kfd->hive_id) { for (nid = 0; nid < proximity_domain; ++nid) { peer_dev = kfd_topology_device_by_proximity_domain_no_lock(nid); if (!peer_dev->gpu) continue; - if (peer_dev->gpu->hive_id != kdev->hive_id) + if (peer_dev->gpu->kfd->hive_id != kdev->kfd->hive_id) continue; sub_type_hdr = (typeof(sub_type_hdr))( (char *)sub_type_hdr + @@ -2255,12 +2278,12 @@ static int kfd_create_vcrat_image_gpu(void *pcrat_image, * (COMPUTE_UNIT_CPU | COMPUTE_UNIT_GPU) - Create VCRAT for APU * -- this option is not currently implemented. * The assumption is that all AMD APUs will have CRAT - * @kdev: Valid kfd_device required if flags contain COMPUTE_UNIT_GPU + * @kdev: Valid kfd_node required if flags contain COMPUTE_UNIT_GPU * * Return 0 if successful else return -ve value */ int kfd_create_crat_image_virtual(void **crat_image, size_t *size, - int flags, struct kfd_dev *kdev, + int flags, struct kfd_node *kdev, uint32_t proximity_domain) { void *pcrat_image = NULL; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.h b/drivers/gpu/drm/amd/amdkfd/kfd_crat.h index 8d1e8ba58dee80e817a8d606ff302caf2b167d0f..fc719389b5d653bacd56c4d7df90e4c4aecabdc5 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.h @@ -275,7 +275,7 @@ struct crat_subtype_iolink { uint32_t maximum_bandwidth_mbs; uint32_t recommended_transfer_size; uint8_t reserved2[CRAT_IOLINK_RESERVED_LENGTH - 1]; - uint8_t num_hops_xgmi; + uint8_t weight_xgmi; }; /* @@ -293,7 +293,7 @@ struct crat_subtype_generic { #pragma pack() -struct kfd_dev; +struct kfd_node; /* Static table to describe GPU Cache information */ struct kfd_gpu_cache_info { @@ -305,14 +305,14 @@ struct kfd_gpu_cache_info { */ uint32_t num_cu_shared; }; -int kfd_get_gpu_cache_info(struct kfd_dev *kdev, struct kfd_gpu_cache_info **pcache_info); +int kfd_get_gpu_cache_info(struct kfd_node *kdev, struct kfd_gpu_cache_info **pcache_info); int kfd_create_crat_image_acpi(void **crat_image, size_t *size); void kfd_destroy_crat_image(void *crat_image); int kfd_parse_crat_table(void *crat_image, struct list_head *device_list, uint32_t proximity_domain); int kfd_create_crat_image_virtual(void **crat_image, size_t *size, - int flags, struct kfd_dev *kdev, + int flags, struct kfd_node *kdev, uint32_t proximity_domain); #endif /* KFD_CRAT_H_INCLUDED */ diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c new file mode 100644 index 0000000000000000000000000000000000000000..fff3ccc04fa94f13042e1ddcb69c1ee7c709bc7e --- /dev/null +++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c @@ -0,0 +1,1118 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ + +#include "kfd_debug.h" +#include "kfd_device_queue_manager.h" +#include "kfd_topology.h" +#include +#include + +#define MAX_WATCH_ADDRESSES 4 + +int kfd_dbg_ev_query_debug_event(struct kfd_process *process, + unsigned int *queue_id, + unsigned int *gpu_id, + uint64_t exception_clear_mask, + uint64_t *event_status) +{ + struct process_queue_manager *pqm; + struct process_queue_node *pqn; + int i; + + if (!(process && process->debug_trap_enabled)) + return -ENODATA; + + mutex_lock(&process->event_mutex); + *event_status = 0; + *queue_id = 0; + *gpu_id = 0; + + /* find and report queue events */ + pqm = &process->pqm; + list_for_each_entry(pqn, &pqm->queues, process_queue_list) { + uint64_t tmp = process->exception_enable_mask; + + if (!pqn->q) + continue; + + tmp &= pqn->q->properties.exception_status; + + if (!tmp) + continue; + + *event_status = pqn->q->properties.exception_status; + *queue_id = pqn->q->properties.queue_id; + *gpu_id = pqn->q->device->id; + pqn->q->properties.exception_status &= ~exception_clear_mask; + goto out; + } + + /* find and report device events */ + for (i = 0; i < process->n_pdds; i++) { + struct kfd_process_device *pdd = process->pdds[i]; + uint64_t tmp = process->exception_enable_mask + & pdd->exception_status; + + if (!tmp) + continue; + + *event_status = pdd->exception_status; + *gpu_id = pdd->dev->id; + pdd->exception_status &= ~exception_clear_mask; + goto out; + } + + /* report process events */ + if (process->exception_enable_mask & process->exception_status) { + *event_status = process->exception_status; + process->exception_status &= ~exception_clear_mask; + } + +out: + mutex_unlock(&process->event_mutex); + return *event_status ? 0 : -EAGAIN; +} + +void debug_event_write_work_handler(struct work_struct *work) +{ + struct kfd_process *process; + + static const char write_data = '.'; + loff_t pos = 0; + + process = container_of(work, + struct kfd_process, + debug_event_workarea); + + kernel_write(process->dbg_ev_file, &write_data, 1, &pos); +} + +/* update process/device/queue exception status, write to descriptor + * only if exception_status is enabled. + */ +bool kfd_dbg_ev_raise(uint64_t event_mask, + struct kfd_process *process, struct kfd_node *dev, + unsigned int source_id, bool use_worker, + void *exception_data, size_t exception_data_size) +{ + struct process_queue_manager *pqm; + struct process_queue_node *pqn; + int i; + static const char write_data = '.'; + loff_t pos = 0; + bool is_subscribed = true; + + if (!(process && process->debug_trap_enabled)) + return false; + + mutex_lock(&process->event_mutex); + + if (event_mask & KFD_EC_MASK_DEVICE) { + for (i = 0; i < process->n_pdds; i++) { + struct kfd_process_device *pdd = process->pdds[i]; + + if (pdd->dev != dev) + continue; + + pdd->exception_status |= event_mask & KFD_EC_MASK_DEVICE; + + if (event_mask & KFD_EC_MASK(EC_DEVICE_MEMORY_VIOLATION)) { + if (!pdd->vm_fault_exc_data) { + pdd->vm_fault_exc_data = kmemdup( + exception_data, + exception_data_size, + GFP_KERNEL); + if (!pdd->vm_fault_exc_data) + pr_debug("Failed to allocate exception data memory"); + } else { + pr_debug("Debugger exception data not saved\n"); + print_hex_dump_bytes("exception data: ", + DUMP_PREFIX_OFFSET, + exception_data, + exception_data_size); + } + } + break; + } + } else if (event_mask & KFD_EC_MASK_PROCESS) { + process->exception_status |= event_mask & KFD_EC_MASK_PROCESS; + } else { + pqm = &process->pqm; + list_for_each_entry(pqn, &pqm->queues, + process_queue_list) { + int target_id; + + if (!pqn->q) + continue; + + target_id = event_mask & KFD_EC_MASK(EC_QUEUE_NEW) ? + pqn->q->properties.queue_id : + pqn->q->doorbell_id; + + if (pqn->q->device != dev || target_id != source_id) + continue; + + pqn->q->properties.exception_status |= event_mask; + break; + } + } + + if (process->exception_enable_mask & event_mask) { + if (use_worker) + schedule_work(&process->debug_event_workarea); + else + kernel_write(process->dbg_ev_file, + &write_data, + 1, + &pos); + } else { + is_subscribed = false; + } + + mutex_unlock(&process->event_mutex); + + return is_subscribed; +} + +/* set pending event queue entry from ring entry */ +bool kfd_set_dbg_ev_from_interrupt(struct kfd_node *dev, + unsigned int pasid, + uint32_t doorbell_id, + uint64_t trap_mask, + void *exception_data, + size_t exception_data_size) +{ + struct kfd_process *p; + bool signaled_to_debugger_or_runtime = false; + + p = kfd_lookup_process_by_pasid(pasid); + + if (!p) + return false; + + if (!kfd_dbg_ev_raise(trap_mask, p, dev, doorbell_id, true, + exception_data, exception_data_size)) { + struct process_queue_manager *pqm; + struct process_queue_node *pqn; + + if (!!(trap_mask & KFD_EC_MASK_QUEUE) && + p->runtime_info.runtime_state == DEBUG_RUNTIME_STATE_ENABLED) { + mutex_lock(&p->mutex); + + pqm = &p->pqm; + list_for_each_entry(pqn, &pqm->queues, + process_queue_list) { + + if (!(pqn->q && pqn->q->device == dev && + pqn->q->doorbell_id == doorbell_id)) + continue; + + kfd_send_exception_to_runtime(p, pqn->q->properties.queue_id, + trap_mask); + + signaled_to_debugger_or_runtime = true; + + break; + } + + mutex_unlock(&p->mutex); + } else if (trap_mask & KFD_EC_MASK(EC_DEVICE_MEMORY_VIOLATION)) { + kfd_dqm_evict_pasid(dev->dqm, p->pasid); + kfd_signal_vm_fault_event(dev, p->pasid, NULL, + exception_data); + + signaled_to_debugger_or_runtime = true; + } + } else { + signaled_to_debugger_or_runtime = true; + } + + kfd_unref_process(p); + + return signaled_to_debugger_or_runtime; +} + +int kfd_dbg_send_exception_to_runtime(struct kfd_process *p, + unsigned int dev_id, + unsigned int queue_id, + uint64_t error_reason) +{ + if (error_reason & KFD_EC_MASK(EC_DEVICE_MEMORY_VIOLATION)) { + struct kfd_process_device *pdd = NULL; + struct kfd_hsa_memory_exception_data *data; + int i; + + for (i = 0; i < p->n_pdds; i++) { + if (p->pdds[i]->dev->id == dev_id) { + pdd = p->pdds[i]; + break; + } + } + + if (!pdd) + return -ENODEV; + + data = (struct kfd_hsa_memory_exception_data *) + pdd->vm_fault_exc_data; + + kfd_dqm_evict_pasid(pdd->dev->dqm, p->pasid); + kfd_signal_vm_fault_event(pdd->dev, p->pasid, NULL, data); + error_reason &= ~KFD_EC_MASK(EC_DEVICE_MEMORY_VIOLATION); + } + + if (error_reason & (KFD_EC_MASK(EC_PROCESS_RUNTIME))) { + /* + * block should only happen after the debugger receives runtime + * enable notice. + */ + up(&p->runtime_enable_sema); + error_reason &= ~KFD_EC_MASK(EC_PROCESS_RUNTIME); + } + + if (error_reason) + return kfd_send_exception_to_runtime(p, queue_id, error_reason); + + return 0; +} + +static int kfd_dbg_set_queue_workaround(struct queue *q, bool enable) +{ + struct mqd_update_info minfo = {0}; + int err; + + if (!q) + return 0; + + if (KFD_GC_VERSION(q->device) < IP_VERSION(11, 0, 0) || + KFD_GC_VERSION(q->device) >= IP_VERSION(12, 0, 0)) + return 0; + + if (enable && q->properties.is_user_cu_masked) + return -EBUSY; + + minfo.update_flag = enable ? UPDATE_FLAG_DBG_WA_ENABLE : UPDATE_FLAG_DBG_WA_DISABLE; + + q->properties.is_dbg_wa = enable; + err = q->device->dqm->ops.update_queue(q->device->dqm, q, &minfo); + if (err) + q->properties.is_dbg_wa = false; + + return err; +} + +static int kfd_dbg_set_workaround(struct kfd_process *target, bool enable) +{ + struct process_queue_manager *pqm = &target->pqm; + struct process_queue_node *pqn; + int r = 0; + + list_for_each_entry(pqn, &pqm->queues, process_queue_list) { + r = kfd_dbg_set_queue_workaround(pqn->q, enable); + if (enable && r) + goto unwind; + } + + return 0; + +unwind: + list_for_each_entry(pqn, &pqm->queues, process_queue_list) + kfd_dbg_set_queue_workaround(pqn->q, false); + + if (enable) + target->runtime_info.runtime_state = r == -EBUSY ? + DEBUG_RUNTIME_STATE_ENABLED_BUSY : + DEBUG_RUNTIME_STATE_ENABLED_ERROR; + + return r; +} + +int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd) +{ + uint32_t spi_dbg_cntl = pdd->spi_dbg_override | pdd->spi_dbg_launch_mode; + uint32_t flags = pdd->process->dbg_flags; + bool sq_trap_en = !!spi_dbg_cntl; + + if (!kfd_dbg_is_per_vmid_supported(pdd->dev)) + return 0; + + return amdgpu_mes_set_shader_debugger(pdd->dev->adev, pdd->proc_ctx_gpu_addr, spi_dbg_cntl, + pdd->watch_points, flags, sq_trap_en); +} + +#define KFD_DEBUGGER_INVALID_WATCH_POINT_ID -1 +static int kfd_dbg_get_dev_watch_id(struct kfd_process_device *pdd, int *watch_id) +{ + int i; + + *watch_id = KFD_DEBUGGER_INVALID_WATCH_POINT_ID; + + spin_lock(&pdd->dev->kfd->watch_points_lock); + + for (i = 0; i < MAX_WATCH_ADDRESSES; i++) { + /* device watchpoint in use so skip */ + if ((pdd->dev->kfd->alloc_watch_ids >> i) & 0x1) + continue; + + pdd->alloc_watch_ids |= 0x1 << i; + pdd->dev->kfd->alloc_watch_ids |= 0x1 << i; + *watch_id = i; + spin_unlock(&pdd->dev->kfd->watch_points_lock); + return 0; + } + + spin_unlock(&pdd->dev->kfd->watch_points_lock); + + return -ENOMEM; +} + +static void kfd_dbg_clear_dev_watch_id(struct kfd_process_device *pdd, int watch_id) +{ + spin_lock(&pdd->dev->kfd->watch_points_lock); + + /* process owns device watch point so safe to clear */ + if ((pdd->alloc_watch_ids >> watch_id) & 0x1) { + pdd->alloc_watch_ids &= ~(0x1 << watch_id); + pdd->dev->kfd->alloc_watch_ids &= ~(0x1 << watch_id); + } + + spin_unlock(&pdd->dev->kfd->watch_points_lock); +} + +static bool kfd_dbg_owns_dev_watch_id(struct kfd_process_device *pdd, int watch_id) +{ + bool owns_watch_id = false; + + spin_lock(&pdd->dev->kfd->watch_points_lock); + owns_watch_id = watch_id < MAX_WATCH_ADDRESSES && + ((pdd->alloc_watch_ids >> watch_id) & 0x1); + + spin_unlock(&pdd->dev->kfd->watch_points_lock); + + return owns_watch_id; +} + +int kfd_dbg_trap_clear_dev_address_watch(struct kfd_process_device *pdd, + uint32_t watch_id) +{ + int r; + + if (!kfd_dbg_owns_dev_watch_id(pdd, watch_id)) + return -EINVAL; + + if (!pdd->dev->kfd->shared_resources.enable_mes) { + r = debug_lock_and_unmap(pdd->dev->dqm); + if (r) + return r; + } + + amdgpu_gfx_off_ctrl(pdd->dev->adev, false); + pdd->watch_points[watch_id] = pdd->dev->kfd2kgd->clear_address_watch( + pdd->dev->adev, + watch_id); + amdgpu_gfx_off_ctrl(pdd->dev->adev, true); + + if (!pdd->dev->kfd->shared_resources.enable_mes) + r = debug_map_and_unlock(pdd->dev->dqm); + else + r = kfd_dbg_set_mes_debug_mode(pdd); + + kfd_dbg_clear_dev_watch_id(pdd, watch_id); + + return r; +} + +int kfd_dbg_trap_set_dev_address_watch(struct kfd_process_device *pdd, + uint64_t watch_address, + uint32_t watch_address_mask, + uint32_t *watch_id, + uint32_t watch_mode) +{ + int r = kfd_dbg_get_dev_watch_id(pdd, watch_id); + + if (r) + return r; + + if (!pdd->dev->kfd->shared_resources.enable_mes) { + r = debug_lock_and_unmap(pdd->dev->dqm); + if (r) { + kfd_dbg_clear_dev_watch_id(pdd, *watch_id); + return r; + } + } + + amdgpu_gfx_off_ctrl(pdd->dev->adev, false); + pdd->watch_points[*watch_id] = pdd->dev->kfd2kgd->set_address_watch( + pdd->dev->adev, + watch_address, + watch_address_mask, + *watch_id, + watch_mode, + pdd->dev->vm_info.last_vmid_kfd); + amdgpu_gfx_off_ctrl(pdd->dev->adev, true); + + if (!pdd->dev->kfd->shared_resources.enable_mes) + r = debug_map_and_unlock(pdd->dev->dqm); + else + r = kfd_dbg_set_mes_debug_mode(pdd); + + /* HWS is broken so no point in HW rollback but release the watchpoint anyways */ + if (r) + kfd_dbg_clear_dev_watch_id(pdd, *watch_id); + + return 0; +} + +static void kfd_dbg_clear_process_address_watch(struct kfd_process *target) +{ + int i, j; + + for (i = 0; i < target->n_pdds; i++) + for (j = 0; j < MAX_WATCH_ADDRESSES; j++) + kfd_dbg_trap_clear_dev_address_watch(target->pdds[i], j); +} + +int kfd_dbg_trap_set_flags(struct kfd_process *target, uint32_t *flags) +{ + uint32_t prev_flags = target->dbg_flags; + int i, r = 0, rewind_count = 0; + + for (i = 0; i < target->n_pdds; i++) { + if (!kfd_dbg_is_per_vmid_supported(target->pdds[i]->dev) && + (*flags & KFD_DBG_TRAP_FLAG_SINGLE_MEM_OP)) { + *flags = prev_flags; + return -EACCES; + } + } + + target->dbg_flags = *flags & KFD_DBG_TRAP_FLAG_SINGLE_MEM_OP; + *flags = prev_flags; + for (i = 0; i < target->n_pdds; i++) { + struct kfd_process_device *pdd = target->pdds[i]; + + if (!kfd_dbg_is_per_vmid_supported(pdd->dev)) + continue; + + if (!pdd->dev->kfd->shared_resources.enable_mes) + r = debug_refresh_runlist(pdd->dev->dqm); + else + r = kfd_dbg_set_mes_debug_mode(pdd); + + if (r) { + target->dbg_flags = prev_flags; + break; + } + + rewind_count++; + } + + /* Rewind flags */ + if (r) { + target->dbg_flags = prev_flags; + + for (i = 0; i < rewind_count; i++) { + struct kfd_process_device *pdd = target->pdds[i]; + + if (!kfd_dbg_is_per_vmid_supported(pdd->dev)) + continue; + + if (!pdd->dev->kfd->shared_resources.enable_mes) + debug_refresh_runlist(pdd->dev->dqm); + else + kfd_dbg_set_mes_debug_mode(pdd); + } + } + + return r; +} + +/* kfd_dbg_trap_deactivate: + * target: target process + * unwind: If this is unwinding a failed kfd_dbg_trap_enable() + * unwind_count: + * If unwind == true, how far down the pdd list we need + * to unwind + * else: ignored + */ +void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind_count) +{ + int i; + + if (!unwind) { + uint32_t flags = 0; + int resume_count = resume_queues(target, 0, NULL); + + if (resume_count) + pr_debug("Resumed %d queues\n", resume_count); + + cancel_work_sync(&target->debug_event_workarea); + kfd_dbg_clear_process_address_watch(target); + kfd_dbg_trap_set_wave_launch_mode(target, 0); + + kfd_dbg_trap_set_flags(target, &flags); + } + + for (i = 0; i < target->n_pdds; i++) { + struct kfd_process_device *pdd = target->pdds[i]; + + /* If this is an unwind, and we have unwound the required + * enable calls on the pdd list, we need to stop now + * otherwise we may mess up another debugger session. + */ + if (unwind && i == unwind_count) + break; + + kfd_process_set_trap_debug_flag(&pdd->qpd, false); + + /* GFX off is already disabled by debug activate if not RLC restore supported. */ + if (kfd_dbg_is_rlc_restore_supported(pdd->dev)) + amdgpu_gfx_off_ctrl(pdd->dev->adev, false); + pdd->spi_dbg_override = + pdd->dev->kfd2kgd->disable_debug_trap( + pdd->dev->adev, + target->runtime_info.ttmp_setup, + pdd->dev->vm_info.last_vmid_kfd); + amdgpu_gfx_off_ctrl(pdd->dev->adev, true); + + if (!kfd_dbg_is_per_vmid_supported(pdd->dev) && + release_debug_trap_vmid(pdd->dev->dqm, &pdd->qpd)) + pr_err("Failed to release debug vmid on [%i]\n", pdd->dev->id); + + if (!pdd->dev->kfd->shared_resources.enable_mes) + debug_refresh_runlist(pdd->dev->dqm); + else + kfd_dbg_set_mes_debug_mode(pdd); + } + + kfd_dbg_set_workaround(target, false); +} + +static void kfd_dbg_clean_exception_status(struct kfd_process *target) +{ + struct process_queue_manager *pqm; + struct process_queue_node *pqn; + int i; + + for (i = 0; i < target->n_pdds; i++) { + struct kfd_process_device *pdd = target->pdds[i]; + + kfd_process_drain_interrupts(pdd); + + pdd->exception_status = 0; + } + + pqm = &target->pqm; + list_for_each_entry(pqn, &pqm->queues, process_queue_list) { + if (!pqn->q) + continue; + + pqn->q->properties.exception_status = 0; + } + + target->exception_status = 0; +} + +int kfd_dbg_trap_disable(struct kfd_process *target) +{ + if (!target->debug_trap_enabled) + return 0; + + /* + * Defer deactivation to runtime if runtime not enabled otherwise reset + * attached running target runtime state to enable for re-attach. + */ + if (target->runtime_info.runtime_state == DEBUG_RUNTIME_STATE_ENABLED) + kfd_dbg_trap_deactivate(target, false, 0); + else if (target->runtime_info.runtime_state != DEBUG_RUNTIME_STATE_DISABLED) + target->runtime_info.runtime_state = DEBUG_RUNTIME_STATE_ENABLED; + + fput(target->dbg_ev_file); + target->dbg_ev_file = NULL; + + if (target->debugger_process) { + atomic_dec(&target->debugger_process->debugged_process_count); + target->debugger_process = NULL; + } + + target->debug_trap_enabled = false; + kfd_dbg_clean_exception_status(target); + kfd_unref_process(target); + + return 0; +} + +int kfd_dbg_trap_activate(struct kfd_process *target) +{ + int i, r = 0; + + r = kfd_dbg_set_workaround(target, true); + if (r) + return r; + + for (i = 0; i < target->n_pdds; i++) { + struct kfd_process_device *pdd = target->pdds[i]; + + if (!kfd_dbg_is_per_vmid_supported(pdd->dev)) { + r = reserve_debug_trap_vmid(pdd->dev->dqm, &pdd->qpd); + + if (r) { + target->runtime_info.runtime_state = (r == -EBUSY) ? + DEBUG_RUNTIME_STATE_ENABLED_BUSY : + DEBUG_RUNTIME_STATE_ENABLED_ERROR; + + goto unwind_err; + } + } + + /* Disable GFX OFF to prevent garbage read/writes to debug registers. + * If RLC restore of debug registers is not supported and runtime enable + * hasn't done so already on ttmp setup request, restore the trap config registers. + * + * If RLC restore of debug registers is not supported, keep gfx off disabled for + * the debug session. + */ + amdgpu_gfx_off_ctrl(pdd->dev->adev, false); + if (!(kfd_dbg_is_rlc_restore_supported(pdd->dev) || + target->runtime_info.ttmp_setup)) + pdd->dev->kfd2kgd->enable_debug_trap(pdd->dev->adev, true, + pdd->dev->vm_info.last_vmid_kfd); + + pdd->spi_dbg_override = pdd->dev->kfd2kgd->enable_debug_trap( + pdd->dev->adev, + false, + pdd->dev->vm_info.last_vmid_kfd); + + if (kfd_dbg_is_rlc_restore_supported(pdd->dev)) + amdgpu_gfx_off_ctrl(pdd->dev->adev, true); + + /* + * Setting the debug flag in the trap handler requires that the TMA has been + * allocated, which occurs during CWSR initialization. + * In the event that CWSR has not been initialized at this point, setting the + * flag will be called again during CWSR initialization if the target process + * is still debug enabled. + */ + kfd_process_set_trap_debug_flag(&pdd->qpd, true); + + if (!pdd->dev->kfd->shared_resources.enable_mes) + r = debug_refresh_runlist(pdd->dev->dqm); + else + r = kfd_dbg_set_mes_debug_mode(pdd); + + if (r) { + target->runtime_info.runtime_state = + DEBUG_RUNTIME_STATE_ENABLED_ERROR; + goto unwind_err; + } + } + + return 0; + +unwind_err: + /* Enabling debug failed, we need to disable on + * all GPUs so the enable is all or nothing. + */ + kfd_dbg_trap_deactivate(target, true, i); + return r; +} + +int kfd_dbg_trap_enable(struct kfd_process *target, uint32_t fd, + void __user *runtime_info, uint32_t *runtime_size) +{ + struct file *f; + uint32_t copy_size; + int i, r = 0; + + if (target->debug_trap_enabled) + return -EALREADY; + + /* Enable pre-checks */ + for (i = 0; i < target->n_pdds; i++) { + struct kfd_process_device *pdd = target->pdds[i]; + + if (!KFD_IS_SOC15(pdd->dev)) + return -ENODEV; + + if (!kfd_dbg_has_gws_support(pdd->dev) && pdd->qpd.num_gws) + return -EBUSY; + } + + copy_size = min((size_t)(*runtime_size), sizeof(target->runtime_info)); + + f = fget(fd); + if (!f) { + pr_err("Failed to get file for (%i)\n", fd); + return -EBADF; + } + + target->dbg_ev_file = f; + + /* defer activation to runtime if not runtime enabled */ + if (target->runtime_info.runtime_state == DEBUG_RUNTIME_STATE_ENABLED) + kfd_dbg_trap_activate(target); + + /* We already hold the process reference but hold another one for the + * debug session. + */ + kref_get(&target->ref); + target->debug_trap_enabled = true; + + if (target->debugger_process) + atomic_inc(&target->debugger_process->debugged_process_count); + + if (copy_to_user(runtime_info, (void *)&target->runtime_info, copy_size)) { + kfd_dbg_trap_deactivate(target, false, 0); + r = -EFAULT; + } + + *runtime_size = sizeof(target->runtime_info); + + return r; +} + +static int kfd_dbg_validate_trap_override_request(struct kfd_process *p, + uint32_t trap_override, + uint32_t trap_mask_request, + uint32_t *trap_mask_supported) +{ + int i = 0; + + *trap_mask_supported = 0xffffffff; + + for (i = 0; i < p->n_pdds; i++) { + struct kfd_process_device *pdd = p->pdds[i]; + int err = pdd->dev->kfd2kgd->validate_trap_override_request( + pdd->dev->adev, + trap_override, + trap_mask_supported); + + if (err) + return err; + } + + if (trap_mask_request & ~*trap_mask_supported) + return -EACCES; + + return 0; +} + +int kfd_dbg_trap_set_wave_launch_override(struct kfd_process *target, + uint32_t trap_override, + uint32_t trap_mask_bits, + uint32_t trap_mask_request, + uint32_t *trap_mask_prev, + uint32_t *trap_mask_supported) +{ + int r = 0, i; + + r = kfd_dbg_validate_trap_override_request(target, + trap_override, + trap_mask_request, + trap_mask_supported); + + if (r) + return r; + + for (i = 0; i < target->n_pdds; i++) { + struct kfd_process_device *pdd = target->pdds[i]; + + amdgpu_gfx_off_ctrl(pdd->dev->adev, false); + pdd->spi_dbg_override = pdd->dev->kfd2kgd->set_wave_launch_trap_override( + pdd->dev->adev, + pdd->dev->vm_info.last_vmid_kfd, + trap_override, + trap_mask_bits, + trap_mask_request, + trap_mask_prev, + pdd->spi_dbg_override); + amdgpu_gfx_off_ctrl(pdd->dev->adev, true); + + if (!pdd->dev->kfd->shared_resources.enable_mes) + r = debug_refresh_runlist(pdd->dev->dqm); + else + r = kfd_dbg_set_mes_debug_mode(pdd); + + if (r) + break; + } + + return r; +} + +int kfd_dbg_trap_set_wave_launch_mode(struct kfd_process *target, + uint8_t wave_launch_mode) +{ + int r = 0, i; + + if (wave_launch_mode != KFD_DBG_TRAP_WAVE_LAUNCH_MODE_NORMAL && + wave_launch_mode != KFD_DBG_TRAP_WAVE_LAUNCH_MODE_HALT && + wave_launch_mode != KFD_DBG_TRAP_WAVE_LAUNCH_MODE_DEBUG) + return -EINVAL; + + for (i = 0; i < target->n_pdds; i++) { + struct kfd_process_device *pdd = target->pdds[i]; + + amdgpu_gfx_off_ctrl(pdd->dev->adev, false); + pdd->spi_dbg_launch_mode = pdd->dev->kfd2kgd->set_wave_launch_mode( + pdd->dev->adev, + wave_launch_mode, + pdd->dev->vm_info.last_vmid_kfd); + amdgpu_gfx_off_ctrl(pdd->dev->adev, true); + + if (!pdd->dev->kfd->shared_resources.enable_mes) + r = debug_refresh_runlist(pdd->dev->dqm); + else + r = kfd_dbg_set_mes_debug_mode(pdd); + + if (r) + break; + } + + return r; +} + +int kfd_dbg_trap_query_exception_info(struct kfd_process *target, + uint32_t source_id, + uint32_t exception_code, + bool clear_exception, + void __user *info, + uint32_t *info_size) +{ + bool found = false; + int r = 0; + uint32_t copy_size, actual_info_size = 0; + uint64_t *exception_status_ptr = NULL; + + if (!target) + return -EINVAL; + + if (!info || !info_size) + return -EINVAL; + + mutex_lock(&target->event_mutex); + + if (KFD_DBG_EC_TYPE_IS_QUEUE(exception_code)) { + /* Per queue exceptions */ + struct queue *queue = NULL; + int i; + + for (i = 0; i < target->n_pdds; i++) { + struct kfd_process_device *pdd = target->pdds[i]; + struct qcm_process_device *qpd = &pdd->qpd; + + list_for_each_entry(queue, &qpd->queues_list, list) { + if (!found && queue->properties.queue_id == source_id) { + found = true; + break; + } + } + if (found) + break; + } + + if (!found) { + r = -EINVAL; + goto out; + } + + if (!(queue->properties.exception_status & KFD_EC_MASK(exception_code))) { + r = -ENODATA; + goto out; + } + exception_status_ptr = &queue->properties.exception_status; + } else if (KFD_DBG_EC_TYPE_IS_DEVICE(exception_code)) { + /* Per device exceptions */ + struct kfd_process_device *pdd = NULL; + int i; + + for (i = 0; i < target->n_pdds; i++) { + pdd = target->pdds[i]; + if (pdd->dev->id == source_id) { + found = true; + break; + } + } + + if (!found) { + r = -EINVAL; + goto out; + } + + if (!(pdd->exception_status & KFD_EC_MASK(exception_code))) { + r = -ENODATA; + goto out; + } + + if (exception_code == EC_DEVICE_MEMORY_VIOLATION) { + copy_size = min((size_t)(*info_size), pdd->vm_fault_exc_data_size); + + if (copy_to_user(info, pdd->vm_fault_exc_data, copy_size)) { + r = -EFAULT; + goto out; + } + actual_info_size = pdd->vm_fault_exc_data_size; + if (clear_exception) { + kfree(pdd->vm_fault_exc_data); + pdd->vm_fault_exc_data = NULL; + pdd->vm_fault_exc_data_size = 0; + } + } + exception_status_ptr = &pdd->exception_status; + } else if (KFD_DBG_EC_TYPE_IS_PROCESS(exception_code)) { + /* Per process exceptions */ + if (!(target->exception_status & KFD_EC_MASK(exception_code))) { + r = -ENODATA; + goto out; + } + + if (exception_code == EC_PROCESS_RUNTIME) { + copy_size = min((size_t)(*info_size), sizeof(target->runtime_info)); + + if (copy_to_user(info, (void *)&target->runtime_info, copy_size)) { + r = -EFAULT; + goto out; + } + + actual_info_size = sizeof(target->runtime_info); + } + + exception_status_ptr = &target->exception_status; + } else { + pr_debug("Bad exception type [%i]\n", exception_code); + r = -EINVAL; + goto out; + } + + *info_size = actual_info_size; + if (clear_exception) + *exception_status_ptr &= ~KFD_EC_MASK(exception_code); +out: + mutex_unlock(&target->event_mutex); + return r; +} + +int kfd_dbg_trap_device_snapshot(struct kfd_process *target, + uint64_t exception_clear_mask, + void __user *user_info, + uint32_t *number_of_device_infos, + uint32_t *entry_size) +{ + struct kfd_dbg_device_info_entry device_info; + uint32_t tmp_entry_size = *entry_size, tmp_num_devices; + int i, r = 0; + + if (!(target && user_info && number_of_device_infos && entry_size)) + return -EINVAL; + + tmp_num_devices = min_t(size_t, *number_of_device_infos, target->n_pdds); + *number_of_device_infos = target->n_pdds; + *entry_size = min_t(size_t, *entry_size, sizeof(device_info)); + + if (!tmp_num_devices) + return 0; + + memset(&device_info, 0, sizeof(device_info)); + + mutex_lock(&target->event_mutex); + + /* Run over all pdd of the process */ + for (i = 0; i < tmp_num_devices; i++) { + struct kfd_process_device *pdd = target->pdds[i]; + struct kfd_topology_device *topo_dev = kfd_topology_device_by_id(pdd->dev->id); + + device_info.gpu_id = pdd->dev->id; + device_info.exception_status = pdd->exception_status; + device_info.lds_base = pdd->lds_base; + device_info.lds_limit = pdd->lds_limit; + device_info.scratch_base = pdd->scratch_base; + device_info.scratch_limit = pdd->scratch_limit; + device_info.gpuvm_base = pdd->gpuvm_base; + device_info.gpuvm_limit = pdd->gpuvm_limit; + device_info.location_id = topo_dev->node_props.location_id; + device_info.vendor_id = topo_dev->node_props.vendor_id; + device_info.device_id = topo_dev->node_props.device_id; + device_info.revision_id = pdd->dev->adev->pdev->revision; + device_info.subsystem_vendor_id = pdd->dev->adev->pdev->subsystem_vendor; + device_info.subsystem_device_id = pdd->dev->adev->pdev->subsystem_device; + device_info.fw_version = pdd->dev->kfd->mec_fw_version; + device_info.gfx_target_version = + topo_dev->node_props.gfx_target_version; + device_info.simd_count = topo_dev->node_props.simd_count; + device_info.max_waves_per_simd = + topo_dev->node_props.max_waves_per_simd; + device_info.array_count = topo_dev->node_props.array_count; + device_info.simd_arrays_per_engine = + topo_dev->node_props.simd_arrays_per_engine; + device_info.num_xcc = NUM_XCC(pdd->dev->xcc_mask); + device_info.capability = topo_dev->node_props.capability; + device_info.debug_prop = topo_dev->node_props.debug_prop; + + if (exception_clear_mask) + pdd->exception_status &= ~exception_clear_mask; + + if (copy_to_user(user_info, &device_info, *entry_size)) { + r = -EFAULT; + break; + } + + user_info += tmp_entry_size; + } + + mutex_unlock(&target->event_mutex); + + return r; +} + +void kfd_dbg_set_enabled_debug_exception_mask(struct kfd_process *target, + uint64_t exception_set_mask) +{ + uint64_t found_mask = 0; + struct process_queue_manager *pqm; + struct process_queue_node *pqn; + static const char write_data = '.'; + loff_t pos = 0; + int i; + + mutex_lock(&target->event_mutex); + + found_mask |= target->exception_status; + + pqm = &target->pqm; + list_for_each_entry(pqn, &pqm->queues, process_queue_list) { + if (!pqn->q) + continue; + + found_mask |= pqn->q->properties.exception_status; + } + + for (i = 0; i < target->n_pdds; i++) { + struct kfd_process_device *pdd = target->pdds[i]; + + found_mask |= pdd->exception_status; + } + + if (exception_set_mask & found_mask) + kernel_write(target->dbg_ev_file, &write_data, 1, &pos); + + target->exception_enable_mask = exception_set_mask; + + mutex_unlock(&target->event_mutex); +} diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h new file mode 100644 index 0000000000000000000000000000000000000000..a289e59ceb797525be1879a8366cdcb2a09a0fbd --- /dev/null +++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h @@ -0,0 +1,123 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ + +#ifndef KFD_DEBUG_EVENTS_H_INCLUDED +#define KFD_DEBUG_EVENTS_H_INCLUDED + +#include "kfd_priv.h" + +void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind_count); +int kfd_dbg_trap_activate(struct kfd_process *target); +int kfd_dbg_ev_query_debug_event(struct kfd_process *process, + unsigned int *queue_id, + unsigned int *gpu_id, + uint64_t exception_clear_mask, + uint64_t *event_status); +bool kfd_set_dbg_ev_from_interrupt(struct kfd_node *dev, + unsigned int pasid, + uint32_t doorbell_id, + uint64_t trap_mask, + void *exception_data, + size_t exception_data_size); +bool kfd_dbg_ev_raise(uint64_t event_mask, + struct kfd_process *process, struct kfd_node *dev, + unsigned int source_id, bool use_worker, + void *exception_data, + size_t exception_data_size); +int kfd_dbg_trap_disable(struct kfd_process *target); +int kfd_dbg_trap_enable(struct kfd_process *target, uint32_t fd, + void __user *runtime_info, + uint32_t *runtime_info_size); +int kfd_dbg_trap_set_wave_launch_override(struct kfd_process *target, + uint32_t trap_override, + uint32_t trap_mask_bits, + uint32_t trap_mask_request, + uint32_t *trap_mask_prev, + uint32_t *trap_mask_supported); +int kfd_dbg_trap_set_wave_launch_mode(struct kfd_process *target, + uint8_t wave_launch_mode); +int kfd_dbg_trap_clear_dev_address_watch(struct kfd_process_device *pdd, + uint32_t watch_id); +int kfd_dbg_trap_set_dev_address_watch(struct kfd_process_device *pdd, + uint64_t watch_address, + uint32_t watch_address_mask, + uint32_t *watch_id, + uint32_t watch_mode); +int kfd_dbg_trap_set_flags(struct kfd_process *target, uint32_t *flags); +int kfd_dbg_trap_query_exception_info(struct kfd_process *target, + uint32_t source_id, + uint32_t exception_code, + bool clear_exception, + void __user *info, + uint32_t *info_size); +int kfd_dbg_send_exception_to_runtime(struct kfd_process *p, + unsigned int dev_id, + unsigned int queue_id, + uint64_t error_reason); + +static inline bool kfd_dbg_is_per_vmid_supported(struct kfd_node *dev) +{ + return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2) || + KFD_GC_VERSION(dev) >= IP_VERSION(11, 0, 0); +} + +void debug_event_write_work_handler(struct work_struct *work); +int kfd_dbg_trap_device_snapshot(struct kfd_process *target, + uint64_t exception_clear_mask, + void __user *user_info, + uint32_t *number_of_device_infos, + uint32_t *entry_size); + +void kfd_dbg_set_enabled_debug_exception_mask(struct kfd_process *target, + uint64_t exception_set_mask); +/* + * If GFX off is enabled, chips that do not support RLC restore for the debug + * registers will disable GFX off temporarily for the entire debug session. + * See disable_on_trap_action_entry and enable_on_trap_action_exit for details. + */ +static inline bool kfd_dbg_is_rlc_restore_supported(struct kfd_node *dev) +{ + return !(KFD_GC_VERSION(dev) == IP_VERSION(10, 1, 10) || + KFD_GC_VERSION(dev) == IP_VERSION(10, 1, 1)); +} + +static inline bool kfd_dbg_has_gws_support(struct kfd_node *dev) +{ + if ((KFD_GC_VERSION(dev) == IP_VERSION(9, 0, 1) + && dev->kfd->mec2_fw_version < 0x81b6) || + (KFD_GC_VERSION(dev) >= IP_VERSION(9, 1, 0) + && KFD_GC_VERSION(dev) <= IP_VERSION(9, 2, 2) + && dev->kfd->mec2_fw_version < 0x1b6) || + (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 0) + && dev->kfd->mec2_fw_version < 0x1b6) || + (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 1) + && dev->kfd->mec2_fw_version < 0x30) || + (KFD_GC_VERSION(dev) >= IP_VERSION(11, 0, 0) && + KFD_GC_VERSION(dev) < IP_VERSION(12, 0, 0))) + return false; + + /* Assume debugging and cooperative launch supported otherwise. */ + return true; +} + +int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd); +#endif diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debugfs.c b/drivers/gpu/drm/amd/amdkfd/kfd_debugfs.c index ad5a40a685aca87dab92a71e184ea60fbac7421f..4a5a0a4e00f217dc26d6ecf8d210135576e066cd 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_debugfs.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_debugfs.c @@ -43,7 +43,7 @@ static int kfd_debugfs_hang_hws_read(struct seq_file *m, void *data) static ssize_t kfd_debugfs_hang_hws_write(struct file *file, const char __user *user_buf, size_t size, loff_t *ppos) { - struct kfd_dev *dev; + struct kfd_node *dev; char tmp[16]; uint32_t gpu_id; int ret = -EINVAL; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c index 00f528eb981263203ebd65b1d60a7745a0bd3bbc..9d4abfd8b55ebb70d979e0362488c1cfc2a8e17b 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c @@ -32,8 +32,10 @@ #include "kfd_iommu.h" #include "amdgpu_amdkfd.h" #include "kfd_smi_events.h" +#include "kfd_svm.h" #include "kfd_migrate.h" #include "amdgpu.h" +#include "amdgpu_xcp.h" #define MQD_SIZE_ALIGNED 768 @@ -42,7 +44,7 @@ * once locked, kfd driver will stop any further GPU execution. * create process (open) will return -EAGAIN. */ -static atomic_t kfd_locked = ATOMIC_INIT(0); +static int kfd_locked; #ifdef CONFIG_DRM_AMDGPU_CIK extern const struct kfd2kgd_calls gfx_v7_kfd2kgd; @@ -51,6 +53,7 @@ extern const struct kfd2kgd_calls gfx_v8_kfd2kgd; extern const struct kfd2kgd_calls gfx_v9_kfd2kgd; extern const struct kfd2kgd_calls arcturus_kfd2kgd; extern const struct kfd2kgd_calls aldebaran_kfd2kgd; +extern const struct kfd2kgd_calls gc_9_4_3_kfd2kgd; extern const struct kfd2kgd_calls gfx_v10_kfd2kgd; extern const struct kfd2kgd_calls gfx_v10_3_kfd2kgd; extern const struct kfd2kgd_calls gfx_v11_kfd2kgd; @@ -60,7 +63,7 @@ static int kfd_gtt_sa_init(struct kfd_dev *kfd, unsigned int buf_size, static void kfd_gtt_sa_fini(struct kfd_dev *kfd); static int kfd_resume_iommu(struct kfd_dev *kfd); -static int kfd_resume(struct kfd_dev *kfd); +static int kfd_resume(struct kfd_node *kfd); static void kfd_device_info_set_sdma_info(struct kfd_dev *kfd) { @@ -81,6 +84,7 @@ static void kfd_device_info_set_sdma_info(struct kfd_dev *kfd) case IP_VERSION(4, 2, 0):/* VEGA20 */ case IP_VERSION(4, 2, 2):/* ARCTURUS */ case IP_VERSION(4, 4, 0):/* ALDEBARAN */ + case IP_VERSION(4, 4, 2): case IP_VERSION(5, 0, 0):/* NAVI10 */ case IP_VERSION(5, 0, 1):/* CYAN_SKILLFISH */ case IP_VERSION(5, 0, 2):/* NAVI14 */ @@ -102,20 +106,19 @@ static void kfd_device_info_set_sdma_info(struct kfd_dev *kfd) kfd->device_info.num_sdma_queues_per_engine = 8; } + bitmap_zero(kfd->device_info.reserved_sdma_queues_bitmap, KFD_MAX_SDMA_QUEUES); + switch (sdma_version) { case IP_VERSION(6, 0, 0): + case IP_VERSION(6, 0, 1): case IP_VERSION(6, 0, 2): case IP_VERSION(6, 0, 3): /* Reserve 1 for paging and 1 for gfx */ kfd->device_info.num_reserved_sdma_queues_per_engine = 2; /* BIT(0)=engine-0 queue-0; BIT(1)=engine-1 queue-0; BIT(2)=engine-0 queue-1; ... */ - kfd->device_info.reserved_sdma_queues_bitmap = 0xFULL; - break; - case IP_VERSION(6, 0, 1): - /* Reserve 1 for paging and 1 for gfx */ - kfd->device_info.num_reserved_sdma_queues_per_engine = 2; - /* BIT(0)=engine-0 queue-0; BIT(1)=engine-0 queue-1; ... */ - kfd->device_info.reserved_sdma_queues_bitmap = 0x3ULL; + bitmap_set(kfd->device_info.reserved_sdma_queues_bitmap, 0, + kfd->adev->sdma.num_instances * + kfd->device_info.num_reserved_sdma_queues_per_engine); break; default: break; @@ -135,6 +138,9 @@ static void kfd_device_info_set_event_interrupt_class(struct kfd_dev *kfd) case IP_VERSION(9, 4, 0): /* VEGA20 */ case IP_VERSION(9, 4, 1): /* ARCTURUS */ case IP_VERSION(9, 4, 2): /* ALDEBARAN */ + case IP_VERSION(9, 4, 3): /* GC 9.4.3 */ + kfd->device_info.event_interrupt_class = &event_interrupt_class_v9; + break; case IP_VERSION(10, 3, 1): /* VANGOGH */ case IP_VERSION(10, 3, 3): /* YELLOW_CARP */ case IP_VERSION(10, 3, 6): /* GC 10.3.6 */ @@ -148,7 +154,7 @@ static void kfd_device_info_set_event_interrupt_class(struct kfd_dev *kfd) case IP_VERSION(10, 3, 2): /* NAVY_FLOUNDER */ case IP_VERSION(10, 3, 4): /* DIMGREY_CAVEFISH */ case IP_VERSION(10, 3, 5): /* BEIGE_GOBY */ - kfd->device_info.event_interrupt_class = &event_interrupt_class_v9; + kfd->device_info.event_interrupt_class = &event_interrupt_class_v10; break; case IP_VERSION(11, 0, 0): case IP_VERSION(11, 0, 1): @@ -327,8 +333,10 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf) f2g = &aldebaran_kfd2kgd; break; case IP_VERSION(9, 4, 3): - gfx_target_version = 90400; - f2g = &aldebaran_kfd2kgd; + gfx_target_version = adev->rev_id >= 1 ? 90402 + : adev->flags & AMD_IS_APU ? 90400 + : 90401; + f2g = &gc_9_4_3_kfd2kgd; break; /* Navi10 */ case IP_VERSION(10, 1, 10): @@ -406,8 +414,15 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf) f2g = &gfx_v11_kfd2kgd; break; case IP_VERSION(11, 0, 3): - /* Note: Compiler version is 11.0.1 while HW version is 11.0.3 */ - gfx_target_version = 110001; + if ((adev->pdev->device == 0x7460 && + adev->pdev->revision == 0x00) || + (adev->pdev->device == 0x7461 && + adev->pdev->revision == 0x00)) + /* Note: Compiler version is 11.0.5 while HW version is 11.0.3 */ + gfx_target_version = 110005; + else + /* Note: Compiler version is 11.0.1 while HW version is 11.0.3 */ + gfx_target_version = 110001; f2g = &gfx_v11_kfd2kgd; break; default: @@ -440,8 +455,6 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf) memset(&kfd->doorbell_available_index, 0, sizeof(kfd->doorbell_available_index)); - atomic_set(&kfd->sram_ecc_flag, 0); - ida_init(&kfd->doorbell_ida); return kfd; @@ -488,41 +501,112 @@ static void kfd_cwsr_init(struct kfd_dev *kfd) } } -static int kfd_gws_init(struct kfd_dev *kfd) +static int kfd_gws_init(struct kfd_node *node) { int ret = 0; + struct kfd_dev *kfd = node->kfd; - if (kfd->dqm->sched_policy == KFD_SCHED_POLICY_NO_HWS) + if (node->dqm->sched_policy == KFD_SCHED_POLICY_NO_HWS) return 0; - if (hws_gws_support || (KFD_IS_SOC15(kfd) && - ((KFD_GC_VERSION(kfd) == IP_VERSION(9, 0, 1) + if (hws_gws_support || (KFD_IS_SOC15(node) && + ((KFD_GC_VERSION(node) == IP_VERSION(9, 0, 1) && kfd->mec2_fw_version >= 0x81b3) || - (KFD_GC_VERSION(kfd) <= IP_VERSION(9, 4, 0) + (KFD_GC_VERSION(node) <= IP_VERSION(9, 4, 0) && kfd->mec2_fw_version >= 0x1b3) || - (KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 1) + (KFD_GC_VERSION(node) == IP_VERSION(9, 4, 1) && kfd->mec2_fw_version >= 0x30) || - (KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 2) + (KFD_GC_VERSION(node) == IP_VERSION(9, 4, 2) && kfd->mec2_fw_version >= 0x28) || - (KFD_GC_VERSION(kfd) >= IP_VERSION(10, 3, 0) - && KFD_GC_VERSION(kfd) < IP_VERSION(11, 0, 0) + (KFD_GC_VERSION(node) >= IP_VERSION(10, 3, 0) + && KFD_GC_VERSION(node) < IP_VERSION(11, 0, 0) && kfd->mec2_fw_version >= 0x6b)))) - ret = amdgpu_amdkfd_alloc_gws(kfd->adev, - kfd->adev->gds.gws_size, &kfd->gws); + ret = amdgpu_amdkfd_alloc_gws(node->adev, + node->adev->gds.gws_size, &node->gws); return ret; } -static void kfd_smi_init(struct kfd_dev *dev) +static void kfd_smi_init(struct kfd_node *dev) { INIT_LIST_HEAD(&dev->smi_clients); spin_lock_init(&dev->smi_lock); } +static int kfd_init_node(struct kfd_node *node) +{ + int err = -1; + + if (kfd_interrupt_init(node)) { + dev_err(kfd_device, "Error initializing interrupts\n"); + goto kfd_interrupt_error; + } + + node->dqm = device_queue_manager_init(node); + if (!node->dqm) { + dev_err(kfd_device, "Error initializing queue manager\n"); + goto device_queue_manager_error; + } + + if (kfd_gws_init(node)) { + dev_err(kfd_device, "Could not allocate %d gws\n", + node->adev->gds.gws_size); + goto gws_error; + } + + if (kfd_resume(node)) + goto kfd_resume_error; + + if (kfd_topology_add_device(node)) { + dev_err(kfd_device, "Error adding device to topology\n"); + goto kfd_topology_add_device_error; + } + + kfd_smi_init(node); + + return 0; + +kfd_topology_add_device_error: +kfd_resume_error: +gws_error: + device_queue_manager_uninit(node->dqm); +device_queue_manager_error: + kfd_interrupt_exit(node); +kfd_interrupt_error: + if (node->gws) + amdgpu_amdkfd_free_gws(node->adev, node->gws); + + /* Cleanup the node memory here */ + kfree(node); + return err; +} + +static void kfd_cleanup_nodes(struct kfd_dev *kfd, unsigned int num_nodes) +{ + struct kfd_node *knode; + unsigned int i; + + for (i = 0; i < num_nodes; i++) { + knode = kfd->nodes[i]; + device_queue_manager_uninit(knode->dqm); + kfd_interrupt_exit(knode); + kfd_topology_remove_device(knode); + if (knode->gws) + amdgpu_amdkfd_free_gws(knode->adev, knode->gws); + kfree(knode); + kfd->nodes[i] = NULL; + } +} + bool kgd2kfd_device_init(struct kfd_dev *kfd, const struct kgd2kfd_shared_resources *gpu_resources) { - unsigned int size, map_process_packet_size; + unsigned int size, map_process_packet_size, i; + struct kfd_node *node; + uint32_t first_vmid_kfd, last_vmid_kfd, vmid_num_kfd; + unsigned int max_proc_per_quantum; + int partition_mode; + int xcp_idx; kfd->mec_fw_version = amdgpu_amdkfd_get_fw_version(kfd->adev, KGD_ENGINE_MEC1); @@ -532,10 +616,14 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, KGD_ENGINE_SDMA1); kfd->shared_resources = *gpu_resources; - kfd->vm_info.first_vmid_kfd = ffs(gpu_resources->compute_vmid_bitmap)-1; - kfd->vm_info.last_vmid_kfd = fls(gpu_resources->compute_vmid_bitmap)-1; - kfd->vm_info.vmid_num_kfd = kfd->vm_info.last_vmid_kfd - - kfd->vm_info.first_vmid_kfd + 1; + kfd->num_nodes = amdgpu_xcp_get_num_xcp(kfd->adev->xcp_mgr); + + if (kfd->num_nodes == 0) { + dev_err(kfd_device, + "KFD num nodes cannot be 0, num_xcc_in_node: %d\n", + kfd->adev->gfx.num_xcc_per_xcp); + goto out; + } /* Allow BIF to recode atomics to PCIe 3.0 AtomicOps. * 32 and 64-bit requests are possible and must be @@ -554,11 +642,34 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, return false; } + first_vmid_kfd = ffs(gpu_resources->compute_vmid_bitmap)-1; + last_vmid_kfd = fls(gpu_resources->compute_vmid_bitmap)-1; + vmid_num_kfd = last_vmid_kfd - first_vmid_kfd + 1; + + /* For GFX9.4.3, we need special handling for VMIDs depending on + * partition mode. + * In CPX mode, the VMID range needs to be shared between XCDs. + * Additionally, there are 13 VMIDs (3-15) available for KFD. To + * divide them equally, we change starting VMID to 4 and not use + * VMID 3. + * If the VMID range changes for GFX9.4.3, then this code MUST be + * revisited. + */ + if (kfd->adev->xcp_mgr) { + partition_mode = amdgpu_xcp_query_partition_mode(kfd->adev->xcp_mgr, + AMDGPU_XCP_FL_LOCKED); + if (partition_mode == AMDGPU_CPX_PARTITION_MODE && + kfd->num_nodes != 1) { + vmid_num_kfd /= 2; + first_vmid_kfd = last_vmid_kfd + 1 - vmid_num_kfd*2; + } + } + /* Verify module parameters regarding mapped process number*/ if (hws_max_conc_proc >= 0) - kfd->max_proc_per_quantum = min((u32)hws_max_conc_proc, kfd->vm_info.vmid_num_kfd); + max_proc_per_quantum = min((u32)hws_max_conc_proc, vmid_num_kfd); else - kfd->max_proc_per_quantum = kfd->vm_info.vmid_num_kfd; + max_proc_per_quantum = vmid_num_kfd; /* calculate max size of mqds needed for queues */ size = max_num_of_queues_per_device * @@ -606,27 +717,15 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, if (amdgpu_use_xgmi_p2p) kfd->hive_id = kfd->adev->gmc.xgmi.hive_id; - kfd->noretry = kfd->adev->gmc.noretry; - - if (kfd_interrupt_init(kfd)) { - dev_err(kfd_device, "Error initializing interrupts\n"); - goto kfd_interrupt_error; - } - - kfd->dqm = device_queue_manager_init(kfd); - if (!kfd->dqm) { - dev_err(kfd_device, "Error initializing queue manager\n"); - goto device_queue_manager_error; - } - - /* If supported on this device, allocate global GWS that is shared - * by all KFD processes + /* + * For GFX9.4.3, the KFD abstracts all partitions within a socket as + * xGMI connected in the topology so assign a unique hive id per + * device based on the pci device location if device is in PCIe mode. */ - if (kfd_gws_init(kfd)) { - dev_err(kfd_device, "Could not allocate %d gws\n", - kfd->adev->gds.gws_size); - goto gws_error; - } + if (!kfd->hive_id && (KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 3)) && kfd->num_nodes > 1) + kfd->hive_id = pci_dev_id(kfd->adev->pdev); + + kfd->noretry = kfd->adev->gmc.noretry; /* If CRAT is broken, won't set iommu enabled */ kfd_double_confirm_iommu_support(kfd); @@ -639,48 +738,100 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, kfd_cwsr_init(kfd); - svm_migrate_init(kfd->adev); + dev_info(kfd_device, "Total number of KFD nodes to be created: %d\n", + kfd->num_nodes); + + /* Allocate the KFD nodes */ + for (i = 0, xcp_idx = 0; i < kfd->num_nodes; i++) { + node = kzalloc(sizeof(struct kfd_node), GFP_KERNEL); + if (!node) + goto node_alloc_error; + + node->node_id = i; + node->adev = kfd->adev; + node->kfd = kfd; + node->kfd2kgd = kfd->kfd2kgd; + node->vm_info.vmid_num_kfd = vmid_num_kfd; + node->xcp = amdgpu_get_next_xcp(kfd->adev->xcp_mgr, &xcp_idx); + /* TODO : Check if error handling is needed */ + if (node->xcp) { + amdgpu_xcp_get_inst_details(node->xcp, AMDGPU_XCP_GFX, + &node->xcc_mask); + ++xcp_idx; + } else { + node->xcc_mask = + (1U << NUM_XCC(kfd->adev->gfx.xcc_mask)) - 1; + } - if (kfd_resume_iommu(kfd)) - goto device_iommu_error; + if (node->xcp) { + dev_info(kfd_device, "KFD node %d partition %d size %lldM\n", + node->node_id, node->xcp->mem_id, + KFD_XCP_MEMORY_SIZE(node->adev, node->node_id) >> 20); + } - if (kfd_resume(kfd)) - goto kfd_resume_error; + if (KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 3) && + partition_mode == AMDGPU_CPX_PARTITION_MODE && + kfd->num_nodes != 1) { + /* For GFX9.4.3 and CPX mode, first XCD gets VMID range + * 4-9 and second XCD gets VMID range 10-15. + */ - amdgpu_amdkfd_get_local_mem_info(kfd->adev, &kfd->local_mem_info); + node->vm_info.first_vmid_kfd = (i%2 == 0) ? + first_vmid_kfd : + first_vmid_kfd+vmid_num_kfd; + node->vm_info.last_vmid_kfd = (i%2 == 0) ? + last_vmid_kfd-vmid_num_kfd : + last_vmid_kfd; + node->compute_vmid_bitmap = + ((0x1 << (node->vm_info.last_vmid_kfd + 1)) - 1) - + ((0x1 << (node->vm_info.first_vmid_kfd)) - 1); + } else { + node->vm_info.first_vmid_kfd = first_vmid_kfd; + node->vm_info.last_vmid_kfd = last_vmid_kfd; + node->compute_vmid_bitmap = + gpu_resources->compute_vmid_bitmap; + } + node->max_proc_per_quantum = max_proc_per_quantum; + atomic_set(&node->sram_ecc_flag, 0); - if (kfd_topology_add_device(kfd)) { - dev_err(kfd_device, "Error adding device to topology\n"); - goto kfd_topology_add_device_error; + amdgpu_amdkfd_get_local_mem_info(kfd->adev, + &node->local_mem_info, node->xcp); + + /* Initialize the KFD node */ + if (kfd_init_node(node)) { + dev_err(kfd_device, "Error initializing KFD node\n"); + goto node_init_error; + } + kfd->nodes[i] = node; } - kfd_smi_init(kfd); + svm_range_set_max_pages(kfd->adev); + + if (kfd_resume_iommu(kfd)) + goto kfd_resume_iommu_error; + + spin_lock_init(&kfd->watch_points_lock); kfd->init_complete = true; dev_info(kfd_device, "added device %x:%x\n", kfd->adev->pdev->vendor, kfd->adev->pdev->device); pr_debug("Starting kfd with the following scheduling policy %d\n", - kfd->dqm->sched_policy); + node->dqm->sched_policy); goto out; -kfd_topology_add_device_error: -kfd_resume_error: +kfd_resume_iommu_error: +node_init_error: +node_alloc_error: + kfd_cleanup_nodes(kfd, i); device_iommu_error: -gws_error: - device_queue_manager_uninit(kfd->dqm); -device_queue_manager_error: - kfd_interrupt_exit(kfd); -kfd_interrupt_error: kfd_doorbell_fini(kfd); kfd_doorbell_error: kfd_gtt_sa_fini(kfd); kfd_gtt_sa_init_error: amdgpu_amdkfd_free_gtt_mem(kfd->adev, kfd->gtt_mem); alloc_gtt_mem_failure: - if (kfd->gws) - amdgpu_amdkfd_free_gws(kfd->adev, kfd->gws); dev_err(kfd_device, "device %x:%x NOT added due to errors\n", kfd->adev->pdev->vendor, kfd->adev->pdev->device); @@ -691,15 +842,13 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, void kgd2kfd_device_exit(struct kfd_dev *kfd) { if (kfd->init_complete) { - device_queue_manager_uninit(kfd->dqm); - kfd_interrupt_exit(kfd); - kfd_topology_remove_device(kfd); + /* Cleanup KFD nodes */ + kfd_cleanup_nodes(kfd, kfd->num_nodes); + /* Cleanup common/shared resources */ kfd_doorbell_fini(kfd); ida_destroy(&kfd->doorbell_ida); kfd_gtt_sa_fini(kfd); amdgpu_amdkfd_free_gtt_mem(kfd->adev, kfd->gtt_mem); - if (kfd->gws) - amdgpu_amdkfd_free_gws(kfd->adev, kfd->gws); } kfree(kfd); @@ -707,16 +856,23 @@ void kgd2kfd_device_exit(struct kfd_dev *kfd) int kgd2kfd_pre_reset(struct kfd_dev *kfd) { + struct kfd_node *node; + int i; + if (!kfd->init_complete) return 0; - kfd_smi_event_update_gpu_reset(kfd, false); - - kfd->dqm->ops.pre_reset(kfd->dqm); + for (i = 0; i < kfd->num_nodes; i++) { + node = kfd->nodes[i]; + kfd_smi_event_update_gpu_reset(node, false); + node->dqm->ops.pre_reset(node->dqm); + } kgd2kfd_suspend(kfd, false); - kfd_signal_reset_event(kfd); + for (i = 0; i < kfd->num_nodes; i++) + kfd_signal_reset_event(kfd->nodes[i]); + return 0; } @@ -729,57 +885,83 @@ int kgd2kfd_pre_reset(struct kfd_dev *kfd) int kgd2kfd_post_reset(struct kfd_dev *kfd) { int ret; + struct kfd_node *node; + int i; if (!kfd->init_complete) return 0; - ret = kfd_resume(kfd); - if (ret) - return ret; - atomic_dec(&kfd_locked); + for (i = 0; i < kfd->num_nodes; i++) { + ret = kfd_resume(kfd->nodes[i]); + if (ret) + return ret; + } - atomic_set(&kfd->sram_ecc_flag, 0); + mutex_lock(&kfd_processes_mutex); + --kfd_locked; + mutex_unlock(&kfd_processes_mutex); - kfd_smi_event_update_gpu_reset(kfd, true); + for (i = 0; i < kfd->num_nodes; i++) { + node = kfd->nodes[i]; + atomic_set(&node->sram_ecc_flag, 0); + kfd_smi_event_update_gpu_reset(node, true); + } return 0; } bool kfd_is_locked(void) { - return (atomic_read(&kfd_locked) > 0); + lockdep_assert_held(&kfd_processes_mutex); + return (kfd_locked > 0); } void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm) { + struct kfd_node *node; + int i; + int count; + if (!kfd->init_complete) return; /* for runtime suspend, skip locking kfd */ if (!run_pm) { + mutex_lock(&kfd_processes_mutex); + count = ++kfd_locked; + mutex_unlock(&kfd_processes_mutex); + /* For first KFD device suspend all the KFD processes */ - if (atomic_inc_return(&kfd_locked) == 1) + if (count == 1) kfd_suspend_all_processes(); } - kfd->dqm->ops.stop(kfd->dqm); + for (i = 0; i < kfd->num_nodes; i++) { + node = kfd->nodes[i]; + node->dqm->ops.stop(node->dqm); + } kfd_iommu_suspend(kfd); } int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm) { - int ret, count; + int ret, count, i; if (!kfd->init_complete) return 0; - ret = kfd_resume(kfd); - if (ret) - return ret; + for (i = 0; i < kfd->num_nodes; i++) { + ret = kfd_resume(kfd->nodes[i]); + if (ret) + return ret; + } /* for runtime resume, skip unlocking kfd */ if (!run_pm) { - count = atomic_dec_return(&kfd_locked); + mutex_lock(&kfd_processes_mutex); + count = --kfd_locked; + mutex_unlock(&kfd_processes_mutex); + WARN_ONCE(count < 0, "KFD suspend / resume ref. error"); if (count == 0) ret = kfd_resume_all_processes(); @@ -808,15 +990,15 @@ static int kfd_resume_iommu(struct kfd_dev *kfd) return err; } -static int kfd_resume(struct kfd_dev *kfd) +static int kfd_resume(struct kfd_node *node) { int err = 0; - err = kfd->dqm->ops.start(kfd->dqm); + err = node->dqm->ops.start(node->dqm); if (err) dev_err(kfd_device, "Error starting queue manager for device %x:%x\n", - kfd->adev->pdev->vendor, kfd->adev->pdev->device); + node->adev->pdev->vendor, node->adev->pdev->device); return err; } @@ -839,9 +1021,10 @@ static inline void kfd_queue_work(struct workqueue_struct *wq, /* This is called directly from KGD at ISR. */ void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry) { - uint32_t patched_ihre[KFD_MAX_RING_ENTRY_SIZE]; + uint32_t patched_ihre[KFD_MAX_RING_ENTRY_SIZE], i; bool is_patched = false; unsigned long flags; + struct kfd_node *node; if (!kfd->init_complete) return; @@ -851,16 +1034,22 @@ void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry) return; } - spin_lock_irqsave(&kfd->interrupt_lock, flags); - - if (kfd->interrupts_active - && interrupt_is_wanted(kfd, ih_ring_entry, - patched_ihre, &is_patched) - && enqueue_ih_ring_entry(kfd, - is_patched ? patched_ihre : ih_ring_entry)) - kfd_queue_work(kfd->ih_wq, &kfd->interrupt_work); + for (i = 0; i < kfd->num_nodes; i++) { + node = kfd->nodes[i]; + spin_lock_irqsave(&node->interrupt_lock, flags); + + if (node->interrupts_active + && interrupt_is_wanted(node, ih_ring_entry, + patched_ihre, &is_patched) + && enqueue_ih_ring_entry(node, + is_patched ? patched_ihre : ih_ring_entry)) { + kfd_queue_work(node->ih_wq, &node->interrupt_work); + spin_unlock_irqrestore(&node->interrupt_lock, flags); + return; + } + spin_unlock_irqrestore(&node->interrupt_lock, flags); + } - spin_unlock_irqrestore(&kfd->interrupt_lock, flags); } int kgd2kfd_quiesce_mm(struct mm_struct *mm, uint32_t trigger) @@ -998,10 +1187,11 @@ static inline uint32_t *kfd_gtt_sa_calc_cpu_addr(void *start_addr, return (uint32_t *) ((uint64_t) start_addr + bit_num * chunk_size); } -int kfd_gtt_sa_allocate(struct kfd_dev *kfd, unsigned int size, +int kfd_gtt_sa_allocate(struct kfd_node *node, unsigned int size, struct kfd_mem_obj **mem_obj) { unsigned int found, start_search, cur_size; + struct kfd_dev *kfd = node->kfd; if (size == 0) return -EINVAL; @@ -1101,8 +1291,10 @@ int kfd_gtt_sa_allocate(struct kfd_dev *kfd, unsigned int size, return -ENOMEM; } -int kfd_gtt_sa_free(struct kfd_dev *kfd, struct kfd_mem_obj *mem_obj) +int kfd_gtt_sa_free(struct kfd_node *node, struct kfd_mem_obj *mem_obj) { + struct kfd_dev *kfd = node->kfd; + /* Act like kfree when trying to free a NULL object */ if (!mem_obj) return 0; @@ -1124,29 +1316,40 @@ int kfd_gtt_sa_free(struct kfd_dev *kfd, struct kfd_mem_obj *mem_obj) void kgd2kfd_set_sram_ecc_flag(struct kfd_dev *kfd) { + /* + * TODO: Currently update SRAM ECC flag for first node. + * This needs to be updated later when we can + * identify SRAM ECC error on other nodes also. + */ if (kfd) - atomic_inc(&kfd->sram_ecc_flag); + atomic_inc(&kfd->nodes[0]->sram_ecc_flag); } -void kfd_inc_compute_active(struct kfd_dev *kfd) +void kfd_inc_compute_active(struct kfd_node *node) { - if (atomic_inc_return(&kfd->compute_profile) == 1) - amdgpu_amdkfd_set_compute_idle(kfd->adev, false); + if (atomic_inc_return(&node->kfd->compute_profile) == 1) + amdgpu_amdkfd_set_compute_idle(node->adev, false); } -void kfd_dec_compute_active(struct kfd_dev *kfd) +void kfd_dec_compute_active(struct kfd_node *node) { - int count = atomic_dec_return(&kfd->compute_profile); + int count = atomic_dec_return(&node->kfd->compute_profile); if (count == 0) - amdgpu_amdkfd_set_compute_idle(kfd->adev, true); + amdgpu_amdkfd_set_compute_idle(node->adev, true); WARN_ONCE(count < 0, "Compute profile ref. count error"); } void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint64_t throttle_bitmask) { + /* + * TODO: For now, raise the throttling event only on first node. + * This will need to change after we are able to determine + * which node raised the throttling event. + */ if (kfd && kfd->init_complete) - kfd_smi_event_update_thermal_throttling(kfd, throttle_bitmask); + kfd_smi_event_update_thermal_throttling(kfd->nodes[0], + throttle_bitmask); } /* kfd_get_num_sdma_engines returns the number of PCIe optimized SDMA and @@ -1154,19 +1357,41 @@ void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint64_t throttle_bitmask) * When the device has more than two engines, we reserve two for PCIe to enable * full-duplex and the rest are used as XGMI. */ -unsigned int kfd_get_num_sdma_engines(struct kfd_dev *kdev) +unsigned int kfd_get_num_sdma_engines(struct kfd_node *node) { /* If XGMI is not supported, all SDMA engines are PCIe */ - if (!kdev->adev->gmc.xgmi.supported) - return kdev->adev->sdma.num_instances; + if (!node->adev->gmc.xgmi.supported) + return node->adev->sdma.num_instances/(int)node->kfd->num_nodes; - return min(kdev->adev->sdma.num_instances, 2); + return min(node->adev->sdma.num_instances/(int)node->kfd->num_nodes, 2); } -unsigned int kfd_get_num_xgmi_sdma_engines(struct kfd_dev *kdev) +unsigned int kfd_get_num_xgmi_sdma_engines(struct kfd_node *node) { /* After reserved for PCIe, the rest of engines are XGMI */ - return kdev->adev->sdma.num_instances - kfd_get_num_sdma_engines(kdev); + return node->adev->sdma.num_instances/(int)node->kfd->num_nodes - + kfd_get_num_sdma_engines(node); +} + +int kgd2kfd_check_and_lock_kfd(void) +{ + mutex_lock(&kfd_processes_mutex); + if (!hash_empty(kfd_processes_table) || kfd_is_locked()) { + mutex_unlock(&kfd_processes_mutex); + return -EBUSY; + } + + ++kfd_locked; + mutex_unlock(&kfd_processes_mutex); + + return 0; +} + +void kgd2kfd_unlock_kfd(void) +{ + mutex_lock(&kfd_processes_mutex); + --kfd_locked; + mutex_unlock(&kfd_processes_mutex); } #if defined(CONFIG_DEBUG_FS) @@ -1174,7 +1399,7 @@ unsigned int kfd_get_num_xgmi_sdma_engines(struct kfd_dev *kdev) /* This function will send a package to HIQ to hang the HWS * which will trigger a GPU reset and bring the HWS back to normal state */ -int kfd_debugfs_hang_hws(struct kfd_dev *dev) +int kfd_debugfs_hang_hws(struct kfd_node *dev) { if (dev->dqm->sched_policy != KFD_SCHED_POLICY_HWS) { pr_err("HWS is not enabled"); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index 7a95698d83f73e82519d3a92e65885997162bbc4..f515cb8f30caf01858901a4138dcfc52ca663fda 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -36,6 +36,7 @@ #include "kfd_kernel_queue.h" #include "amdgpu_amdkfd.h" #include "mes_api_def.h" +#include "kfd_debug.h" /* Size of the per-pipe EOP queue */ #define CIK_HPD_EOP_BYTES_LOG2 11 @@ -46,10 +47,13 @@ static int set_pasid_vmid_mapping(struct device_queue_manager *dqm, static int execute_queues_cpsch(struct device_queue_manager *dqm, enum kfd_unmap_queues_filter filter, - uint32_t filter_param); + uint32_t filter_param, + uint32_t grace_period); static int unmap_queues_cpsch(struct device_queue_manager *dqm, enum kfd_unmap_queues_filter filter, - uint32_t filter_param, bool reset); + uint32_t filter_param, + uint32_t grace_period, + bool reset); static int map_queues_cpsch(struct device_queue_manager *dqm); @@ -74,31 +78,31 @@ enum KFD_MQD_TYPE get_mqd_type_from_queue_type(enum kfd_queue_type type) static bool is_pipe_enabled(struct device_queue_manager *dqm, int mec, int pipe) { int i; - int pipe_offset = (mec * dqm->dev->shared_resources.num_pipe_per_mec - + pipe) * dqm->dev->shared_resources.num_queue_per_pipe; + int pipe_offset = (mec * dqm->dev->kfd->shared_resources.num_pipe_per_mec + + pipe) * dqm->dev->kfd->shared_resources.num_queue_per_pipe; /* queue is available for KFD usage if bit is 1 */ - for (i = 0; i < dqm->dev->shared_resources.num_queue_per_pipe; ++i) + for (i = 0; i < dqm->dev->kfd->shared_resources.num_queue_per_pipe; ++i) if (test_bit(pipe_offset + i, - dqm->dev->shared_resources.cp_queue_bitmap)) + dqm->dev->kfd->shared_resources.cp_queue_bitmap)) return true; return false; } unsigned int get_cp_queues_num(struct device_queue_manager *dqm) { - return bitmap_weight(dqm->dev->shared_resources.cp_queue_bitmap, + return bitmap_weight(dqm->dev->kfd->shared_resources.cp_queue_bitmap, KGD_MAX_QUEUES); } unsigned int get_queues_per_pipe(struct device_queue_manager *dqm) { - return dqm->dev->shared_resources.num_queue_per_pipe; + return dqm->dev->kfd->shared_resources.num_queue_per_pipe; } unsigned int get_pipes_per_mec(struct device_queue_manager *dqm) { - return dqm->dev->shared_resources.num_pipe_per_mec; + return dqm->dev->kfd->shared_resources.num_pipe_per_mec; } static unsigned int get_num_all_sdma_engines(struct device_queue_manager *dqm) @@ -110,29 +114,40 @@ static unsigned int get_num_all_sdma_engines(struct device_queue_manager *dqm) unsigned int get_num_sdma_queues(struct device_queue_manager *dqm) { return kfd_get_num_sdma_engines(dqm->dev) * - dqm->dev->device_info.num_sdma_queues_per_engine; + dqm->dev->kfd->device_info.num_sdma_queues_per_engine; } unsigned int get_num_xgmi_sdma_queues(struct device_queue_manager *dqm) { return kfd_get_num_xgmi_sdma_engines(dqm->dev) * - dqm->dev->device_info.num_sdma_queues_per_engine; + dqm->dev->kfd->device_info.num_sdma_queues_per_engine; } -static inline uint64_t get_reserved_sdma_queues_bitmap(struct device_queue_manager *dqm) +static void init_sdma_bitmaps(struct device_queue_manager *dqm) { - return dqm->dev->device_info.reserved_sdma_queues_bitmap; + bitmap_zero(dqm->sdma_bitmap, KFD_MAX_SDMA_QUEUES); + bitmap_set(dqm->sdma_bitmap, 0, get_num_sdma_queues(dqm)); + + bitmap_zero(dqm->xgmi_sdma_bitmap, KFD_MAX_SDMA_QUEUES); + bitmap_set(dqm->xgmi_sdma_bitmap, 0, get_num_xgmi_sdma_queues(dqm)); + + /* Mask out the reserved queues */ + bitmap_andnot(dqm->sdma_bitmap, dqm->sdma_bitmap, + dqm->dev->kfd->device_info.reserved_sdma_queues_bitmap, + KFD_MAX_SDMA_QUEUES); } void program_sh_mem_settings(struct device_queue_manager *dqm, struct qcm_process_device *qpd) { - return dqm->dev->kfd2kgd->program_sh_mem_settings( - dqm->dev->adev, qpd->vmid, - qpd->sh_mem_config, - qpd->sh_mem_ape1_base, - qpd->sh_mem_ape1_limit, - qpd->sh_mem_bases); + uint32_t xcc_mask = dqm->dev->xcc_mask; + int xcc_id; + + for_each_inst(xcc_id, xcc_mask) + dqm->dev->kfd2kgd->program_sh_mem_settings( + dqm->dev->adev, qpd->vmid, qpd->sh_mem_config, + qpd->sh_mem_ape1_base, qpd->sh_mem_ape1_limit, + qpd->sh_mem_bases, xcc_id); } static void kfd_hws_hang(struct device_queue_manager *dqm) @@ -211,6 +226,9 @@ static int add_queue_mes(struct device_queue_manager *dqm, struct queue *q, queue_input.paging = false; queue_input.tba_addr = qpd->tba_addr; queue_input.tma_addr = qpd->tma_addr; + queue_input.trap_en = KFD_GC_VERSION(q->device) < IP_VERSION(11, 0, 0) || + KFD_GC_VERSION(q->device) > IP_VERSION(11, 0, 3); + queue_input.skip_process_ctx_clear = qpd->pqm->process->debug_trap_enabled; queue_type = convert_to_mes_queue_type(q->properties.type); if (queue_type < 0) { @@ -330,7 +348,7 @@ static int allocate_doorbell(struct qcm_process_device *qpd, struct queue *q, uint32_t const *restore_id) { - struct kfd_dev *dev = qpd->dqm->dev; + struct kfd_node *dev = qpd->dqm->dev; if (!KFD_IS_SOC15(dev)) { /* On pre-SOC15 chips we need to use the queue ID to @@ -349,8 +367,17 @@ static int allocate_doorbell(struct qcm_process_device *qpd, * for a SDMA engine is 512. */ - uint32_t *idx_offset = dev->shared_resources.sdma_doorbell_idx; - uint32_t valid_id = idx_offset[q->properties.sdma_engine_id] + uint32_t *idx_offset = dev->kfd->shared_resources.sdma_doorbell_idx; + + /* + * q->properties.sdma_engine_id corresponds to the virtual + * sdma engine number. However, for doorbell allocation, + * we need the physical sdma engine id in order to get the + * correct doorbell offset. + */ + uint32_t valid_id = idx_offset[qpd->dqm->dev->node_id * + get_num_all_sdma_engines(qpd->dqm) + + q->properties.sdma_engine_id] + (q->properties.sdma_queue_id & 1) * KFD_QUEUE_DOORBELL_MIRROR_OFFSET + (q->properties.sdma_queue_id >> 1); @@ -382,7 +409,7 @@ static int allocate_doorbell(struct qcm_process_device *qpd, } q->properties.doorbell_off = - kfd_get_doorbell_dw_offset_in_bar(dev, qpd_to_pdd(qpd), + kfd_get_doorbell_dw_offset_in_bar(dev->kfd, qpd_to_pdd(qpd), q->doorbell_id); return 0; } @@ -391,7 +418,7 @@ static void deallocate_doorbell(struct qcm_process_device *qpd, struct queue *q) { unsigned int old; - struct kfd_dev *dev = qpd->dqm->dev; + struct kfd_node *dev = qpd->dqm->dev; if (!KFD_IS_SOC15(dev) || q->properties.type == KFD_QUEUE_TYPE_SDMA || @@ -405,10 +432,14 @@ static void deallocate_doorbell(struct qcm_process_device *qpd, static void program_trap_handler_settings(struct device_queue_manager *dqm, struct qcm_process_device *qpd) { + uint32_t xcc_mask = dqm->dev->xcc_mask; + int xcc_id; + if (dqm->dev->kfd2kgd->program_trap_handler_settings) - dqm->dev->kfd2kgd->program_trap_handler_settings( - dqm->dev->adev, qpd->vmid, - qpd->tba_addr, qpd->tma_addr); + for_each_inst(xcc_id, xcc_mask) + dqm->dev->kfd2kgd->program_trap_handler_settings( + dqm->dev->adev, qpd->vmid, qpd->tba_addr, + qpd->tma_addr, xcc_id); } static int allocate_vmid(struct device_queue_manager *dqm, @@ -441,7 +472,7 @@ static int allocate_vmid(struct device_queue_manager *dqm, program_sh_mem_settings(dqm, qpd); - if (KFD_IS_SOC15(dqm->dev) && dqm->dev->cwsr_enabled) + if (KFD_IS_SOC15(dqm->dev) && dqm->dev->kfd->cwsr_enabled) program_trap_handler_settings(dqm, qpd); /* qpd->page_table_base is set earlier when register_process() @@ -460,7 +491,7 @@ static int allocate_vmid(struct device_queue_manager *dqm, return 0; } -static int flush_texture_cache_nocpsch(struct kfd_dev *kdev, +static int flush_texture_cache_nocpsch(struct kfd_node *kdev, struct qcm_process_device *qpd) { const struct packet_manager_funcs *pmf = qpd->dqm->packet_mgr.pmf; @@ -661,7 +692,7 @@ static inline void deallocate_hqd(struct device_queue_manager *dqm, #define SQ_IND_CMD_CMD_KILL 0x00000003 #define SQ_IND_CMD_MODE_BROADCAST 0x00000001 -static int dbgdev_wave_reset_wavefronts(struct kfd_dev *dev, struct kfd_process *p) +static int dbgdev_wave_reset_wavefronts(struct kfd_node *dev, struct kfd_process *p) { int status = 0; unsigned int vmid; @@ -671,6 +702,8 @@ static int dbgdev_wave_reset_wavefronts(struct kfd_dev *dev, struct kfd_process struct kfd_process_device *pdd; int first_vmid_to_scan = dev->vm_info.first_vmid_kfd; int last_vmid_to_scan = dev->vm_info.last_vmid_kfd; + uint32_t xcc_mask = dev->xcc_mask; + int xcc_id; reg_sq_cmd.u32All = 0; reg_gfx_index.u32All = 0; @@ -715,9 +748,10 @@ static int dbgdev_wave_reset_wavefronts(struct kfd_dev *dev, struct kfd_process reg_sq_cmd.bits.cmd = SQ_IND_CMD_CMD_KILL; reg_sq_cmd.bits.vm_id = vmid; - dev->kfd2kgd->wave_control_execute(dev->adev, - reg_gfx_index.u32All, - reg_sq_cmd.u32All); + for_each_inst(xcc_id, xcc_mask) + dev->kfd2kgd->wave_control_execute( + dev->adev, reg_gfx_index.u32All, + reg_sq_cmd.u32All, xcc_id); return 0; } @@ -837,9 +871,9 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q, /* Make sure the queue is unmapped before updating the MQD */ if (dqm->sched_policy != KFD_SCHED_POLICY_NO_HWS) { - if (!dqm->dev->shared_resources.enable_mes) + if (!dqm->dev->kfd->shared_resources.enable_mes) retval = unmap_queues_cpsch(dqm, - KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0, false); + KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0, USE_DEFAULT_GRACE_PERIOD, false); else if (prev_active) retval = remove_queue_mes(dqm, q, &pdd->qpd); @@ -858,7 +892,7 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q, } retval = mqd_mgr->destroy_mqd(mqd_mgr, q->mqd, - (dqm->dev->cwsr_enabled ? + (dqm->dev->kfd->cwsr_enabled ? KFD_PREEMPT_TYPE_WAVEFRONT_SAVE : KFD_PREEMPT_TYPE_WAVEFRONT_DRAIN), KFD_UNMAP_LATENCY_MS, q->pipe, q->queue); @@ -895,7 +929,7 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q, } if (dqm->sched_policy != KFD_SCHED_POLICY_NO_HWS) { - if (!dqm->dev->shared_resources.enable_mes) + if (!dqm->dev->kfd->shared_resources.enable_mes) retval = map_queues_cpsch(dqm); else if (q->properties.is_active) retval = add_queue_mes(dqm, q, &pdd->qpd); @@ -917,6 +951,92 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q, return retval; } +/* suspend_single_queue does not lock the dqm like the + * evict_process_queues_cpsch or evict_process_queues_nocpsch. You should + * lock the dqm before calling, and unlock after calling. + * + * The reason we don't lock the dqm is because this function may be + * called on multiple queues in a loop, so rather than locking/unlocking + * multiple times, we will just keep the dqm locked for all of the calls. + */ +static int suspend_single_queue(struct device_queue_manager *dqm, + struct kfd_process_device *pdd, + struct queue *q) +{ + bool is_new; + + if (q->properties.is_suspended) + return 0; + + pr_debug("Suspending PASID %u queue [%i]\n", + pdd->process->pasid, + q->properties.queue_id); + + is_new = q->properties.exception_status & KFD_EC_MASK(EC_QUEUE_NEW); + + if (is_new || q->properties.is_being_destroyed) { + pr_debug("Suspend: skip %s queue id %i\n", + is_new ? "new" : "destroyed", + q->properties.queue_id); + return -EBUSY; + } + + q->properties.is_suspended = true; + if (q->properties.is_active) { + if (dqm->dev->kfd->shared_resources.enable_mes) { + int r = remove_queue_mes(dqm, q, &pdd->qpd); + + if (r) + return r; + } + + decrement_queue_count(dqm, &pdd->qpd, q); + q->properties.is_active = false; + } + + return 0; +} + +/* resume_single_queue does not lock the dqm like the functions + * restore_process_queues_cpsch or restore_process_queues_nocpsch. You should + * lock the dqm before calling, and unlock after calling. + * + * The reason we don't lock the dqm is because this function may be + * called on multiple queues in a loop, so rather than locking/unlocking + * multiple times, we will just keep the dqm locked for all of the calls. + */ +static int resume_single_queue(struct device_queue_manager *dqm, + struct qcm_process_device *qpd, + struct queue *q) +{ + struct kfd_process_device *pdd; + + if (!q->properties.is_suspended) + return 0; + + pdd = qpd_to_pdd(qpd); + + pr_debug("Restoring from suspend PASID %u queue [%i]\n", + pdd->process->pasid, + q->properties.queue_id); + + q->properties.is_suspended = false; + + if (QUEUE_IS_ACTIVE(q->properties)) { + if (dqm->dev->kfd->shared_resources.enable_mes) { + int r = add_queue_mes(dqm, q, &pdd->qpd); + + if (r) + return r; + } + + q->properties.is_active = true; + increment_queue_count(dqm, qpd, q); + } + + return 0; +} + static int evict_process_queues_nocpsch(struct device_queue_manager *dqm, struct qcm_process_device *qpd) { @@ -951,7 +1071,7 @@ static int evict_process_queues_nocpsch(struct device_queue_manager *dqm, continue; retval = mqd_mgr->destroy_mqd(mqd_mgr, q->mqd, - (dqm->dev->cwsr_enabled ? + (dqm->dev->kfd->cwsr_enabled ? KFD_PREEMPT_TYPE_WAVEFRONT_SAVE : KFD_PREEMPT_TYPE_WAVEFRONT_DRAIN), KFD_UNMAP_LATENCY_MS, q->pipe, q->queue); @@ -979,6 +1099,14 @@ static int evict_process_queues_cpsch(struct device_queue_manager *dqm, goto out; pdd = qpd_to_pdd(qpd); + + /* The debugger creates processes that temporarily have not acquired + * all VMs for all devices and has no VMs itself. + * Skip queue eviction on process eviction. + */ + if (!pdd->drm_priv) + goto out; + pr_debug_ratelimited("Evicting PASID 0x%x queues\n", pdd->process->pasid); @@ -993,7 +1121,7 @@ static int evict_process_queues_cpsch(struct device_queue_manager *dqm, q->properties.is_active = false; decrement_queue_count(dqm, qpd, q); - if (dqm->dev->shared_resources.enable_mes) { + if (dqm->dev->kfd->shared_resources.enable_mes) { retval = remove_queue_mes(dqm, q, qpd); if (retval) { pr_err("Failed to evict queue %d\n", @@ -1003,11 +1131,12 @@ static int evict_process_queues_cpsch(struct device_queue_manager *dqm, } } pdd->last_evict_timestamp = get_jiffies_64(); - if (!dqm->dev->shared_resources.enable_mes) + if (!dqm->dev->kfd->shared_resources.enable_mes) retval = execute_queues_cpsch(dqm, qpd->is_debug ? KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES : - KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0); + KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0, + USE_DEFAULT_GRACE_PERIOD); out: dqm_unlock(dqm); @@ -1100,13 +1229,10 @@ static int restore_process_queues_cpsch(struct device_queue_manager *dqm, { struct queue *q; struct kfd_process_device *pdd; - uint64_t pd_base; uint64_t eviction_duration; int retval = 0; pdd = qpd_to_pdd(qpd); - /* Retrieve PD base */ - pd_base = amdgpu_amdkfd_gpuvm_get_process_page_dir(pdd->drm_priv); dqm_lock(dqm); if (WARN_ON_ONCE(!qpd->evicted)) /* already restored, do nothing */ @@ -1116,12 +1242,19 @@ static int restore_process_queues_cpsch(struct device_queue_manager *dqm, goto out; } + /* The debugger creates processes that temporarily have not acquired + * all VMs for all devices and has no VMs itself. + * Skip queue restore on process restore. + */ + if (!pdd->drm_priv) + goto vm_not_acquired; + pr_debug_ratelimited("Restoring PASID 0x%x queues\n", pdd->process->pasid); /* Update PD Base in QPD */ - qpd->page_table_base = pd_base; - pr_debug("Updated PD address to 0x%llx\n", pd_base); + qpd->page_table_base = amdgpu_amdkfd_gpuvm_get_process_page_dir(pdd->drm_priv); + pr_debug("Updated PD address to 0x%llx\n", qpd->page_table_base); /* activate all active queues on the qpd */ list_for_each_entry(q, &qpd->queues_list, list) { @@ -1132,7 +1265,7 @@ static int restore_process_queues_cpsch(struct device_queue_manager *dqm, q->properties.is_active = true; increment_queue_count(dqm, &pdd->qpd, q); - if (dqm->dev->shared_resources.enable_mes) { + if (dqm->dev->kfd->shared_resources.enable_mes) { retval = add_queue_mes(dqm, q, qpd); if (retval) { pr_err("Failed to restore queue %d\n", @@ -1141,12 +1274,13 @@ static int restore_process_queues_cpsch(struct device_queue_manager *dqm, } } } - if (!dqm->dev->shared_resources.enable_mes) + if (!dqm->dev->kfd->shared_resources.enable_mes) retval = execute_queues_cpsch(dqm, - KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0); - qpd->evicted = 0; + KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0, USE_DEFAULT_GRACE_PERIOD); eviction_duration = get_jiffies_64() - pdd->last_evict_timestamp; atomic64_add(eviction_duration, &pdd->evict_duration_counter); +vm_not_acquired: + qpd->evicted = 0; out: dqm_unlock(dqm); return retval; @@ -1229,35 +1363,32 @@ static int set_pasid_vmid_mapping(struct device_queue_manager *dqm, u32 pasid, unsigned int vmid) { - return dqm->dev->kfd2kgd->set_pasid_vmid_mapping( - dqm->dev->adev, pasid, vmid); -} + uint32_t xcc_mask = dqm->dev->xcc_mask; + int xcc_id, ret; -static void init_interrupts(struct device_queue_manager *dqm) -{ - unsigned int i; + for_each_inst(xcc_id, xcc_mask) { + ret = dqm->dev->kfd2kgd->set_pasid_vmid_mapping( + dqm->dev->adev, pasid, vmid, xcc_id); + if (ret) + break; + } - for (i = 0 ; i < get_pipes_per_mec(dqm) ; i++) - if (is_pipe_enabled(dqm, 0, i)) - dqm->dev->kfd2kgd->init_interrupts(dqm->dev->adev, i); + return ret; } -static void init_sdma_bitmaps(struct device_queue_manager *dqm) +static void init_interrupts(struct device_queue_manager *dqm) { - unsigned int num_sdma_queues = - min_t(unsigned int, sizeof(dqm->sdma_bitmap)*8, - get_num_sdma_queues(dqm)); - unsigned int num_xgmi_sdma_queues = - min_t(unsigned int, sizeof(dqm->xgmi_sdma_bitmap)*8, - get_num_xgmi_sdma_queues(dqm)); - - if (num_sdma_queues) - dqm->sdma_bitmap = GENMASK_ULL(num_sdma_queues-1, 0); - if (num_xgmi_sdma_queues) - dqm->xgmi_sdma_bitmap = GENMASK_ULL(num_xgmi_sdma_queues-1, 0); + uint32_t xcc_mask = dqm->dev->xcc_mask; + unsigned int i, xcc_id; - dqm->sdma_bitmap &= ~get_reserved_sdma_queues_bitmap(dqm); - pr_info("sdma_bitmap: %llx\n", dqm->sdma_bitmap); + for_each_inst(xcc_id, xcc_mask) { + for (i = 0 ; i < get_pipes_per_mec(dqm) ; i++) { + if (is_pipe_enabled(dqm, 0, i)) { + dqm->dev->kfd2kgd->init_interrupts( + dqm->dev->adev, i, xcc_id); + } + } + } } static int initialize_nocpsch(struct device_queue_manager *dqm) @@ -1282,7 +1413,7 @@ static int initialize_nocpsch(struct device_queue_manager *dqm) for (queue = 0; queue < get_queues_per_pipe(dqm); queue++) if (test_bit(pipe_offset + queue, - dqm->dev->shared_resources.cp_queue_bitmap)) + dqm->dev->kfd->shared_resources.cp_queue_bitmap)) dqm->allocated_queues[pipe] |= 1 << queue; } @@ -1322,9 +1453,16 @@ static int start_nocpsch(struct device_queue_manager *dqm) static int stop_nocpsch(struct device_queue_manager *dqm) { + dqm_lock(dqm); + if (!dqm->sched_running) { + dqm_unlock(dqm); + return 0; + } + if (dqm->dev->adev->asic_type == CHIP_HAWAII) pm_uninit(&dqm->packet_mgr, false); dqm->sched_running = false; + dqm_unlock(dqm); return 0; } @@ -1342,46 +1480,48 @@ static int allocate_sdma_queue(struct device_queue_manager *dqm, int bit; if (q->properties.type == KFD_QUEUE_TYPE_SDMA) { - if (dqm->sdma_bitmap == 0) { + if (bitmap_empty(dqm->sdma_bitmap, KFD_MAX_SDMA_QUEUES)) { pr_err("No more SDMA queue to allocate\n"); return -ENOMEM; } if (restore_sdma_id) { /* Re-use existing sdma_id */ - if (!(dqm->sdma_bitmap & (1ULL << *restore_sdma_id))) { + if (!test_bit(*restore_sdma_id, dqm->sdma_bitmap)) { pr_err("SDMA queue already in use\n"); return -EBUSY; } - dqm->sdma_bitmap &= ~(1ULL << *restore_sdma_id); + clear_bit(*restore_sdma_id, dqm->sdma_bitmap); q->sdma_id = *restore_sdma_id; } else { /* Find first available sdma_id */ - bit = __ffs64(dqm->sdma_bitmap); - dqm->sdma_bitmap &= ~(1ULL << bit); + bit = find_first_bit(dqm->sdma_bitmap, + get_num_sdma_queues(dqm)); + clear_bit(bit, dqm->sdma_bitmap); q->sdma_id = bit; } - q->properties.sdma_engine_id = q->sdma_id % - kfd_get_num_sdma_engines(dqm->dev); + q->properties.sdma_engine_id = + q->sdma_id % kfd_get_num_sdma_engines(dqm->dev); q->properties.sdma_queue_id = q->sdma_id / kfd_get_num_sdma_engines(dqm->dev); } else if (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) { - if (dqm->xgmi_sdma_bitmap == 0) { + if (bitmap_empty(dqm->xgmi_sdma_bitmap, KFD_MAX_SDMA_QUEUES)) { pr_err("No more XGMI SDMA queue to allocate\n"); return -ENOMEM; } if (restore_sdma_id) { /* Re-use existing sdma_id */ - if (!(dqm->xgmi_sdma_bitmap & (1ULL << *restore_sdma_id))) { + if (!test_bit(*restore_sdma_id, dqm->xgmi_sdma_bitmap)) { pr_err("SDMA queue already in use\n"); return -EBUSY; } - dqm->xgmi_sdma_bitmap &= ~(1ULL << *restore_sdma_id); + clear_bit(*restore_sdma_id, dqm->xgmi_sdma_bitmap); q->sdma_id = *restore_sdma_id; } else { - bit = __ffs64(dqm->xgmi_sdma_bitmap); - dqm->xgmi_sdma_bitmap &= ~(1ULL << bit); + bit = find_first_bit(dqm->xgmi_sdma_bitmap, + get_num_xgmi_sdma_queues(dqm)); + clear_bit(bit, dqm->xgmi_sdma_bitmap); q->sdma_id = bit; } /* sdma_engine_id is sdma id including @@ -1409,11 +1549,11 @@ static void deallocate_sdma_queue(struct device_queue_manager *dqm, if (q->properties.type == KFD_QUEUE_TYPE_SDMA) { if (q->sdma_id >= get_num_sdma_queues(dqm)) return; - dqm->sdma_bitmap |= (1ULL << q->sdma_id); + set_bit(q->sdma_id, dqm->sdma_bitmap); } else if (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) { if (q->sdma_id >= get_num_xgmi_sdma_queues(dqm)) return; - dqm->xgmi_sdma_bitmap |= (1ULL << q->sdma_id); + set_bit(q->sdma_id, dqm->xgmi_sdma_bitmap); } } @@ -1426,14 +1566,14 @@ static int set_sched_resources(struct device_queue_manager *dqm) int i, mec; struct scheduling_resources res; - res.vmid_mask = dqm->dev->shared_resources.compute_vmid_bitmap; + res.vmid_mask = dqm->dev->compute_vmid_bitmap; res.queue_mask = 0; for (i = 0; i < KGD_MAX_QUEUES; ++i) { - mec = (i / dqm->dev->shared_resources.num_queue_per_pipe) - / dqm->dev->shared_resources.num_pipe_per_mec; + mec = (i / dqm->dev->kfd->shared_resources.num_queue_per_pipe) + / dqm->dev->kfd->shared_resources.num_pipe_per_mec; - if (!test_bit(i, dqm->dev->shared_resources.cp_queue_bitmap)) + if (!test_bit(i, dqm->dev->kfd->shared_resources.cp_queue_bitmap)) continue; /* only acquire queues from the first MEC */ @@ -1475,9 +1615,13 @@ static int initialize_cpsch(struct device_queue_manager *dqm) dqm->gws_queue_count = 0; dqm->active_runlist = false; INIT_WORK(&dqm->hw_exception_work, kfd_process_hw_exception); + dqm->trap_debug_vmid = 0; init_sdma_bitmaps(dqm); + if (dqm->dev->kfd2kgd->get_iq_wait_times) + dqm->dev->kfd2kgd->get_iq_wait_times(dqm->dev->adev, + &dqm->wait_times); return 0; } @@ -1489,7 +1633,7 @@ static int start_cpsch(struct device_queue_manager *dqm) dqm_lock(dqm); - if (!dqm->dev->shared_resources.enable_mes) { + if (!dqm->dev->kfd->shared_resources.enable_mes) { retval = pm_init(&dqm->packet_mgr, dqm); if (retval) goto fail_packet_manager_init; @@ -1516,14 +1660,15 @@ static int start_cpsch(struct device_queue_manager *dqm) dqm->is_hws_hang = false; dqm->is_resetting = false; dqm->sched_running = true; - if (!dqm->dev->shared_resources.enable_mes) - execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0); + + if (!dqm->dev->kfd->shared_resources.enable_mes) + execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0, USE_DEFAULT_GRACE_PERIOD); dqm_unlock(dqm); return 0; fail_allocate_vidmem: fail_set_sched_resources: - if (!dqm->dev->shared_resources.enable_mes) + if (!dqm->dev->kfd->shared_resources.enable_mes) pm_uninit(&dqm->packet_mgr, false); fail_packet_manager_init: dqm_unlock(dqm); @@ -1541,8 +1686,8 @@ static int stop_cpsch(struct device_queue_manager *dqm) } if (!dqm->is_hws_hang) { - if (!dqm->dev->shared_resources.enable_mes) - unmap_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0, false); + if (!dqm->dev->kfd->shared_resources.enable_mes) + unmap_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0, USE_DEFAULT_GRACE_PERIOD, false); else remove_all_queues_mes(dqm); } @@ -1550,11 +1695,11 @@ static int stop_cpsch(struct device_queue_manager *dqm) hanging = dqm->is_hws_hang || dqm->is_resetting; dqm->sched_running = false; - if (!dqm->dev->shared_resources.enable_mes) + if (!dqm->dev->kfd->shared_resources.enable_mes) pm_release_ib(&dqm->packet_mgr); kfd_gtt_sa_free(dqm->dev, dqm->fence_mem); - if (!dqm->dev->shared_resources.enable_mes) + if (!dqm->dev->kfd->shared_resources.enable_mes) pm_uninit(&dqm->packet_mgr, hanging); dqm_unlock(dqm); @@ -1584,7 +1729,8 @@ static int create_kernel_queue_cpsch(struct device_queue_manager *dqm, list_add(&kq->list, &qpd->priv_queue_list); increment_queue_count(dqm, qpd, kq->queue); qpd->is_debug = true; - execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0); + execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0, + USE_DEFAULT_GRACE_PERIOD); dqm_unlock(dqm); return 0; @@ -1598,7 +1744,8 @@ static void destroy_kernel_queue_cpsch(struct device_queue_manager *dqm, list_del(&kq->list); decrement_queue_count(dqm, qpd, kq->queue); qpd->is_debug = false; - execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0); + execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0, + USE_DEFAULT_GRACE_PERIOD); /* * Unconditionally decrement this counter, regardless of the queue's * type. @@ -1658,6 +1805,9 @@ static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q, * updates the is_evicted flag but is a no-op otherwise. */ q->properties.is_evicted = !!qpd->evicted; + q->properties.is_dbg_wa = qpd->pqm->process->debug_trap_enabled && + KFD_GC_VERSION(q->device) >= IP_VERSION(11, 0, 0) && + KFD_GC_VERSION(q->device) <= IP_VERSION(11, 0, 3); if (qd) mqd_mgr->restore_mqd(mqd_mgr, &q->mqd, q->mqd_mem_obj, &q->gart_mqd_addr, @@ -1673,9 +1823,9 @@ static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q, if (q->properties.is_active) { increment_queue_count(dqm, qpd, q); - if (!dqm->dev->shared_resources.enable_mes) + if (!dqm->dev->kfd->shared_resources.enable_mes) retval = execute_queues_cpsch(dqm, - KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0); + KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0, USE_DEFAULT_GRACE_PERIOD); else retval = add_queue_mes(dqm, q, qpd); if (retval) @@ -1764,7 +1914,9 @@ static int map_queues_cpsch(struct device_queue_manager *dqm) /* dqm->lock mutex has to be locked before calling this function */ static int unmap_queues_cpsch(struct device_queue_manager *dqm, enum kfd_unmap_queues_filter filter, - uint32_t filter_param, bool reset) + uint32_t filter_param, + uint32_t grace_period, + bool reset) { int retval = 0; struct mqd_manager *mqd_mgr; @@ -1776,6 +1928,12 @@ static int unmap_queues_cpsch(struct device_queue_manager *dqm, if (!dqm->active_runlist) return retval; + if (grace_period != USE_DEFAULT_GRACE_PERIOD) { + retval = pm_update_grace_period(&dqm->packet_mgr, grace_period); + if (retval) + return retval; + } + retval = pm_send_unmap_queue(&dqm->packet_mgr, filter, filter_param, reset); if (retval) return retval; @@ -1808,6 +1966,13 @@ static int unmap_queues_cpsch(struct device_queue_manager *dqm, return -ETIME; } + /* We need to reset the grace period value for this device */ + if (grace_period != USE_DEFAULT_GRACE_PERIOD) { + if (pm_update_grace_period(&dqm->packet_mgr, + USE_DEFAULT_GRACE_PERIOD)) + pr_err("Failed to reset grace period\n"); + } + pm_release_ib(&dqm->packet_mgr); dqm->active_runlist = false; @@ -1823,7 +1988,7 @@ static int reset_queues_cpsch(struct device_queue_manager *dqm, dqm_lock(dqm); retval = unmap_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_BY_PASID, - pasid, true); + pasid, USE_DEFAULT_GRACE_PERIOD, true); dqm_unlock(dqm); return retval; @@ -1832,19 +1997,45 @@ static int reset_queues_cpsch(struct device_queue_manager *dqm, /* dqm->lock mutex has to be locked before calling this function */ static int execute_queues_cpsch(struct device_queue_manager *dqm, enum kfd_unmap_queues_filter filter, - uint32_t filter_param) + uint32_t filter_param, + uint32_t grace_period) { int retval; if (dqm->is_hws_hang) return -EIO; - retval = unmap_queues_cpsch(dqm, filter, filter_param, false); + retval = unmap_queues_cpsch(dqm, filter, filter_param, grace_period, false); if (retval) return retval; return map_queues_cpsch(dqm); } +static int wait_on_destroy_queue(struct device_queue_manager *dqm, + struct queue *q) +{ + struct kfd_process_device *pdd = kfd_get_process_device_data(q->device, + q->process); + int ret = 0; + + if (pdd->qpd.is_debug) + return ret; + + q->properties.is_being_destroyed = true; + + if (pdd->process->debug_trap_enabled && q->properties.is_suspended) { + dqm_unlock(dqm); + mutex_unlock(&q->process->mutex); + ret = wait_event_interruptible(dqm->destroy_wait, + !q->properties.is_suspended); + + mutex_lock(&q->process->mutex); + dqm_lock(dqm); + } + + return ret; +} + static int destroy_queue_cpsch(struct device_queue_manager *dqm, struct qcm_process_device *qpd, struct queue *q) @@ -1864,11 +2055,16 @@ static int destroy_queue_cpsch(struct device_queue_manager *dqm, q->properties.queue_id); } - retval = 0; - /* remove queue from list to prevent rescheduling after preemption */ dqm_lock(dqm); + retval = wait_on_destroy_queue(dqm, q); + + if (retval) { + dqm_unlock(dqm); + return retval; + } + if (qpd->is_debug) { /* * error, currently we do not allow to destroy a queue @@ -1893,10 +2089,11 @@ static int destroy_queue_cpsch(struct device_queue_manager *dqm, list_del(&q->list); qpd->queue_count--; if (q->properties.is_active) { - if (!dqm->dev->shared_resources.enable_mes) { - decrement_queue_count(dqm, qpd, q); + decrement_queue_count(dqm, qpd, q); + if (!dqm->dev->kfd->shared_resources.enable_mes) { retval = execute_queues_cpsch(dqm, - KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0); + KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0, + USE_DEFAULT_GRACE_PERIOD); if (retval == -ETIME) qpd->reset_wavefronts = true; } else { @@ -1914,7 +2111,14 @@ static int destroy_queue_cpsch(struct device_queue_manager *dqm, dqm_unlock(dqm); - /* Do free_mqd after dqm_unlock(dqm) to avoid circular locking */ + /* + * Do free_mqd and raise delete event after dqm_unlock(dqm) to avoid + * circular locking + */ + kfd_dbg_ev_raise(KFD_EC_MASK(EC_DEVICE_QUEUE_DELETE), + qpd->pqm->process, q->device, + -1, false, NULL, 0); + mqd_mgr->free_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj); return retval; @@ -2056,7 +2260,7 @@ static int get_wave_state(struct device_queue_manager *dqm, mqd_mgr = dqm->mqd_mgrs[KFD_MQD_TYPE_CP]; if (q->properties.type != KFD_QUEUE_TYPE_COMPUTE || - q->properties.is_active || !q->device->cwsr_enabled || + q->properties.is_active || !q->device->kfd->cwsr_enabled || !mqd_mgr->get_wave_state) { dqm_unlock(dqm); return -EINVAL; @@ -2069,8 +2273,8 @@ static int get_wave_state(struct device_queue_manager *dqm, * and the queue should be protected against destruction by the process * lock. */ - return mqd_mgr->get_wave_state(mqd_mgr, q->mqd, ctl_stack, - ctl_stack_used_size, save_area_used_size); + return mqd_mgr->get_wave_state(mqd_mgr, q->mqd, &q->properties, + ctl_stack, ctl_stack_used_size, save_area_used_size); } static void get_queue_checkpoint_info(struct device_queue_manager *dqm, @@ -2105,7 +2309,7 @@ static int checkpoint_mqd(struct device_queue_manager *dqm, dqm_lock(dqm); - if (q->properties.is_active || !q->device->cwsr_enabled) { + if (q->properties.is_active || !q->device->kfd->cwsr_enabled) { r = -EINVAL; goto dqm_unlock; } @@ -2158,7 +2362,7 @@ static int process_termination_cpsch(struct device_queue_manager *dqm, if (q->properties.is_active) { decrement_queue_count(dqm, qpd, q); - if (dqm->dev->shared_resources.enable_mes) { + if (dqm->dev->kfd->shared_resources.enable_mes) { retval = remove_queue_mes(dqm, q, qpd); if (retval) pr_err("Failed to remove queue %d\n", @@ -2180,8 +2384,8 @@ static int process_termination_cpsch(struct device_queue_manager *dqm, } } - if (!dqm->dev->shared_resources.enable_mes) - retval = execute_queues_cpsch(dqm, filter, 0); + if (!dqm->dev->kfd->shared_resources.enable_mes) + retval = execute_queues_cpsch(dqm, filter, 0, USE_DEFAULT_GRACE_PERIOD); if ((!dqm->is_hws_hang) && (retval || qpd->reset_wavefronts)) { pr_warn("Resetting wave fronts (cpsch) on dev %p\n", dqm->dev); @@ -2242,12 +2446,13 @@ static int init_mqd_managers(struct device_queue_manager *dqm) static int allocate_hiq_sdma_mqd(struct device_queue_manager *dqm) { int retval; - struct kfd_dev *dev = dqm->dev; + struct kfd_node *dev = dqm->dev; struct kfd_mem_obj *mem_obj = &dqm->hiq_sdma_mqd; uint32_t size = dqm->mqd_mgrs[KFD_MQD_TYPE_SDMA]->mqd_size * get_num_all_sdma_engines(dqm) * - dev->device_info.num_sdma_queues_per_engine + - dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ]->mqd_size; + dev->kfd->device_info.num_sdma_queues_per_engine + + (dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ]->mqd_size * + NUM_XCC(dqm->dev->xcc_mask)); retval = amdgpu_amdkfd_alloc_gtt_mem(dev->adev, size, &(mem_obj->gtt_mem), &(mem_obj->gpu_addr), @@ -2256,7 +2461,7 @@ static int allocate_hiq_sdma_mqd(struct device_queue_manager *dqm) return retval; } -struct device_queue_manager *device_queue_manager_init(struct kfd_dev *dev) +struct device_queue_manager *device_queue_manager_init(struct kfd_node *dev) { struct device_queue_manager *dqm; @@ -2373,20 +2578,22 @@ struct device_queue_manager *device_queue_manager_init(struct kfd_dev *dev) if (init_mqd_managers(dqm)) goto out_free; - if (!dev->shared_resources.enable_mes && allocate_hiq_sdma_mqd(dqm)) { + if (!dev->kfd->shared_resources.enable_mes && allocate_hiq_sdma_mqd(dqm)) { pr_err("Failed to allocate hiq sdma mqd trunk buffer\n"); goto out_free; } - if (!dqm->ops.initialize(dqm)) + if (!dqm->ops.initialize(dqm)) { + init_waitqueue_head(&dqm->destroy_wait); return dqm; + } out_free: kfree(dqm); return NULL; } -static void deallocate_hiq_sdma_mqd(struct kfd_dev *dev, +static void deallocate_hiq_sdma_mqd(struct kfd_node *dev, struct kfd_mem_obj *mqd) { WARN(!mqd, "No hiq sdma mqd trunk to free"); @@ -2396,8 +2603,9 @@ static void deallocate_hiq_sdma_mqd(struct kfd_dev *dev, void device_queue_manager_uninit(struct device_queue_manager *dqm) { + dqm->ops.stop(dqm); dqm->ops.uninitialize(dqm); - if (!dqm->dev->shared_resources.enable_mes) + if (!dqm->dev->kfd->shared_resources.enable_mes) deallocate_hiq_sdma_mqd(dqm->dev, &dqm->hiq_sdma_mqd); kfree(dqm); } @@ -2426,6 +2634,498 @@ static void kfd_process_hw_exception(struct work_struct *work) amdgpu_amdkfd_gpu_reset(dqm->dev->adev); } +int reserve_debug_trap_vmid(struct device_queue_manager *dqm, + struct qcm_process_device *qpd) +{ + int r; + int updated_vmid_mask; + + if (dqm->sched_policy == KFD_SCHED_POLICY_NO_HWS) { + pr_err("Unsupported on sched_policy: %i\n", dqm->sched_policy); + return -EINVAL; + } + + dqm_lock(dqm); + + if (dqm->trap_debug_vmid != 0) { + pr_err("Trap debug id already reserved\n"); + r = -EBUSY; + goto out_unlock; + } + + r = unmap_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0, + USE_DEFAULT_GRACE_PERIOD, false); + if (r) + goto out_unlock; + + updated_vmid_mask = dqm->dev->kfd->shared_resources.compute_vmid_bitmap; + updated_vmid_mask &= ~(1 << dqm->dev->vm_info.last_vmid_kfd); + + dqm->dev->kfd->shared_resources.compute_vmid_bitmap = updated_vmid_mask; + dqm->trap_debug_vmid = dqm->dev->vm_info.last_vmid_kfd; + r = set_sched_resources(dqm); + if (r) + goto out_unlock; + + r = map_queues_cpsch(dqm); + if (r) + goto out_unlock; + + pr_debug("Reserved VMID for trap debug: %i\n", dqm->trap_debug_vmid); + +out_unlock: + dqm_unlock(dqm); + return r; +} + +/* + * Releases vmid for the trap debugger + */ +int release_debug_trap_vmid(struct device_queue_manager *dqm, + struct qcm_process_device *qpd) +{ + int r; + int updated_vmid_mask; + uint32_t trap_debug_vmid; + + if (dqm->sched_policy == KFD_SCHED_POLICY_NO_HWS) { + pr_err("Unsupported on sched_policy: %i\n", dqm->sched_policy); + return -EINVAL; + } + + dqm_lock(dqm); + trap_debug_vmid = dqm->trap_debug_vmid; + if (dqm->trap_debug_vmid == 0) { + pr_err("Trap debug id is not reserved\n"); + r = -EINVAL; + goto out_unlock; + } + + r = unmap_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0, + USE_DEFAULT_GRACE_PERIOD, false); + if (r) + goto out_unlock; + + updated_vmid_mask = dqm->dev->kfd->shared_resources.compute_vmid_bitmap; + updated_vmid_mask |= (1 << dqm->dev->vm_info.last_vmid_kfd); + + dqm->dev->kfd->shared_resources.compute_vmid_bitmap = updated_vmid_mask; + dqm->trap_debug_vmid = 0; + r = set_sched_resources(dqm); + if (r) + goto out_unlock; + + r = map_queues_cpsch(dqm); + if (r) + goto out_unlock; + + pr_debug("Released VMID for trap debug: %i\n", trap_debug_vmid); + +out_unlock: + dqm_unlock(dqm); + return r; +} + +#define QUEUE_NOT_FOUND -1 +/* invalidate queue operation in array */ +static void q_array_invalidate(uint32_t num_queues, uint32_t *queue_ids) +{ + int i; + + for (i = 0; i < num_queues; i++) + queue_ids[i] |= KFD_DBG_QUEUE_INVALID_MASK; +} + +/* find queue index in array */ +static int q_array_get_index(unsigned int queue_id, + uint32_t num_queues, + uint32_t *queue_ids) +{ + int i; + + for (i = 0; i < num_queues; i++) + if (queue_id == (queue_ids[i] & ~KFD_DBG_QUEUE_INVALID_MASK)) + return i; + + return QUEUE_NOT_FOUND; +} + +struct copy_context_work_handler_workarea { + struct work_struct copy_context_work; + struct kfd_process *p; +}; + +static void copy_context_work_handler (struct work_struct *work) +{ + struct copy_context_work_handler_workarea *workarea; + struct mqd_manager *mqd_mgr; + struct queue *q; + struct mm_struct *mm; + struct kfd_process *p; + uint32_t tmp_ctl_stack_used_size, tmp_save_area_used_size; + int i; + + workarea = container_of(work, + struct copy_context_work_handler_workarea, + copy_context_work); + + p = workarea->p; + mm = get_task_mm(p->lead_thread); + + if (!mm) + return; + + kthread_use_mm(mm); + for (i = 0; i < p->n_pdds; i++) { + struct kfd_process_device *pdd = p->pdds[i]; + struct device_queue_manager *dqm = pdd->dev->dqm; + struct qcm_process_device *qpd = &pdd->qpd; + + list_for_each_entry(q, &qpd->queues_list, list) { + mqd_mgr = dqm->mqd_mgrs[KFD_MQD_TYPE_CP]; + + /* We ignore the return value from get_wave_state + * because + * i) right now, it always returns 0, and + * ii) if we hit an error, we would continue to the + * next queue anyway. + */ + mqd_mgr->get_wave_state(mqd_mgr, + q->mqd, + &q->properties, + (void __user *) q->properties.ctx_save_restore_area_address, + &tmp_ctl_stack_used_size, + &tmp_save_area_used_size); + } + } + kthread_unuse_mm(mm); + mmput(mm); +} + +static uint32_t *get_queue_ids(uint32_t num_queues, uint32_t *usr_queue_id_array) +{ + size_t array_size = num_queues * sizeof(uint32_t); + uint32_t *queue_ids = NULL; + + if (!usr_queue_id_array) + return NULL; + + queue_ids = kzalloc(array_size, GFP_KERNEL); + if (!queue_ids) + return ERR_PTR(-ENOMEM); + + if (copy_from_user(queue_ids, usr_queue_id_array, array_size)) + return ERR_PTR(-EFAULT); + + return queue_ids; +} + +int resume_queues(struct kfd_process *p, + uint32_t num_queues, + uint32_t *usr_queue_id_array) +{ + uint32_t *queue_ids = NULL; + int total_resumed = 0; + int i; + + if (usr_queue_id_array) { + queue_ids = get_queue_ids(num_queues, usr_queue_id_array); + + if (IS_ERR(queue_ids)) + return PTR_ERR(queue_ids); + + /* mask all queues as invalid. unmask per successful request */ + q_array_invalidate(num_queues, queue_ids); + } + + for (i = 0; i < p->n_pdds; i++) { + struct kfd_process_device *pdd = p->pdds[i]; + struct device_queue_manager *dqm = pdd->dev->dqm; + struct qcm_process_device *qpd = &pdd->qpd; + struct queue *q; + int r, per_device_resumed = 0; + + dqm_lock(dqm); + + /* unmask queues that resume or already resumed as valid */ + list_for_each_entry(q, &qpd->queues_list, list) { + int q_idx = QUEUE_NOT_FOUND; + + if (queue_ids) + q_idx = q_array_get_index( + q->properties.queue_id, + num_queues, + queue_ids); + + if (!queue_ids || q_idx != QUEUE_NOT_FOUND) { + int err = resume_single_queue(dqm, &pdd->qpd, q); + + if (queue_ids) { + if (!err) { + queue_ids[q_idx] &= + ~KFD_DBG_QUEUE_INVALID_MASK; + } else { + queue_ids[q_idx] |= + KFD_DBG_QUEUE_ERROR_MASK; + break; + } + } + + if (dqm->dev->kfd->shared_resources.enable_mes) { + wake_up_all(&dqm->destroy_wait); + if (!err) + total_resumed++; + } else { + per_device_resumed++; + } + } + } + + if (!per_device_resumed) { + dqm_unlock(dqm); + continue; + } + + r = execute_queues_cpsch(dqm, + KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, + 0, + USE_DEFAULT_GRACE_PERIOD); + if (r) { + pr_err("Failed to resume process queues\n"); + if (queue_ids) { + list_for_each_entry(q, &qpd->queues_list, list) { + int q_idx = q_array_get_index( + q->properties.queue_id, + num_queues, + queue_ids); + + /* mask queue as error on resume fail */ + if (q_idx != QUEUE_NOT_FOUND) + queue_ids[q_idx] |= + KFD_DBG_QUEUE_ERROR_MASK; + } + } + } else { + wake_up_all(&dqm->destroy_wait); + total_resumed += per_device_resumed; + } + + dqm_unlock(dqm); + } + + if (queue_ids) { + if (copy_to_user((void __user *)usr_queue_id_array, queue_ids, + num_queues * sizeof(uint32_t))) + pr_err("copy_to_user failed on queue resume\n"); + + kfree(queue_ids); + } + + return total_resumed; +} + +int suspend_queues(struct kfd_process *p, + uint32_t num_queues, + uint32_t grace_period, + uint64_t exception_clear_mask, + uint32_t *usr_queue_id_array) +{ + uint32_t *queue_ids = get_queue_ids(num_queues, usr_queue_id_array); + int total_suspended = 0; + int i; + + if (IS_ERR(queue_ids)) + return PTR_ERR(queue_ids); + + /* mask all queues as invalid. umask on successful request */ + q_array_invalidate(num_queues, queue_ids); + + for (i = 0; i < p->n_pdds; i++) { + struct kfd_process_device *pdd = p->pdds[i]; + struct device_queue_manager *dqm = pdd->dev->dqm; + struct qcm_process_device *qpd = &pdd->qpd; + struct queue *q; + int r, per_device_suspended = 0; + + mutex_lock(&p->event_mutex); + dqm_lock(dqm); + + /* unmask queues that suspend or already suspended */ + list_for_each_entry(q, &qpd->queues_list, list) { + int q_idx = q_array_get_index(q->properties.queue_id, + num_queues, + queue_ids); + + if (q_idx != QUEUE_NOT_FOUND) { + int err = suspend_single_queue(dqm, pdd, q); + bool is_mes = dqm->dev->kfd->shared_resources.enable_mes; + + if (!err) { + queue_ids[q_idx] &= ~KFD_DBG_QUEUE_INVALID_MASK; + if (exception_clear_mask && is_mes) + q->properties.exception_status &= + ~exception_clear_mask; + + if (is_mes) + total_suspended++; + else + per_device_suspended++; + } else if (err != -EBUSY) { + r = err; + queue_ids[q_idx] |= KFD_DBG_QUEUE_ERROR_MASK; + break; + } + } + } + + if (!per_device_suspended) { + dqm_unlock(dqm); + mutex_unlock(&p->event_mutex); + if (total_suspended) + amdgpu_amdkfd_debug_mem_fence(dqm->dev->adev); + continue; + } + + r = execute_queues_cpsch(dqm, + KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0, + grace_period); + + if (r) + pr_err("Failed to suspend process queues.\n"); + else + total_suspended += per_device_suspended; + + list_for_each_entry(q, &qpd->queues_list, list) { + int q_idx = q_array_get_index(q->properties.queue_id, + num_queues, queue_ids); + + if (q_idx == QUEUE_NOT_FOUND) + continue; + + /* mask queue as error on suspend fail */ + if (r) + queue_ids[q_idx] |= KFD_DBG_QUEUE_ERROR_MASK; + else if (exception_clear_mask) + q->properties.exception_status &= + ~exception_clear_mask; + } + + dqm_unlock(dqm); + mutex_unlock(&p->event_mutex); + amdgpu_device_flush_hdp(dqm->dev->adev, NULL); + } + + if (total_suspended) { + struct copy_context_work_handler_workarea copy_context_worker; + + INIT_WORK_ONSTACK( + ©_context_worker.copy_context_work, + copy_context_work_handler); + + copy_context_worker.p = p; + + schedule_work(©_context_worker.copy_context_work); + + + flush_work(©_context_worker.copy_context_work); + destroy_work_on_stack(©_context_worker.copy_context_work); + } + + if (copy_to_user((void __user *)usr_queue_id_array, queue_ids, + num_queues * sizeof(uint32_t))) + pr_err("copy_to_user failed on queue suspend\n"); + + kfree(queue_ids); + + return total_suspended; +} + +static uint32_t set_queue_type_for_user(struct queue_properties *q_props) +{ + switch (q_props->type) { + case KFD_QUEUE_TYPE_COMPUTE: + return q_props->format == KFD_QUEUE_FORMAT_PM4 + ? KFD_IOC_QUEUE_TYPE_COMPUTE + : KFD_IOC_QUEUE_TYPE_COMPUTE_AQL; + case KFD_QUEUE_TYPE_SDMA: + return KFD_IOC_QUEUE_TYPE_SDMA; + case KFD_QUEUE_TYPE_SDMA_XGMI: + return KFD_IOC_QUEUE_TYPE_SDMA_XGMI; + default: + WARN_ONCE(true, "queue type not recognized!"); + return 0xffffffff; + }; +} + +void set_queue_snapshot_entry(struct queue *q, + uint64_t exception_clear_mask, + struct kfd_queue_snapshot_entry *qss_entry) +{ + qss_entry->ring_base_address = q->properties.queue_address; + qss_entry->write_pointer_address = (uint64_t)q->properties.write_ptr; + qss_entry->read_pointer_address = (uint64_t)q->properties.read_ptr; + qss_entry->ctx_save_restore_address = + q->properties.ctx_save_restore_area_address; + qss_entry->ctx_save_restore_area_size = + q->properties.ctx_save_restore_area_size; + qss_entry->exception_status = q->properties.exception_status; + qss_entry->queue_id = q->properties.queue_id; + qss_entry->gpu_id = q->device->id; + qss_entry->ring_size = (uint32_t)q->properties.queue_size; + qss_entry->queue_type = set_queue_type_for_user(&q->properties); + q->properties.exception_status &= ~exception_clear_mask; +} + +int debug_lock_and_unmap(struct device_queue_manager *dqm) +{ + int r; + + if (dqm->sched_policy == KFD_SCHED_POLICY_NO_HWS) { + pr_err("Unsupported on sched_policy: %i\n", dqm->sched_policy); + return -EINVAL; + } + + if (!kfd_dbg_is_per_vmid_supported(dqm->dev)) + return 0; + + dqm_lock(dqm); + + r = unmap_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0, 0, false); + if (r) + dqm_unlock(dqm); + + return r; +} + +int debug_map_and_unlock(struct device_queue_manager *dqm) +{ + int r; + + if (dqm->sched_policy == KFD_SCHED_POLICY_NO_HWS) { + pr_err("Unsupported on sched_policy: %i\n", dqm->sched_policy); + return -EINVAL; + } + + if (!kfd_dbg_is_per_vmid_supported(dqm->dev)) + return 0; + + r = map_queues_cpsch(dqm); + + dqm_unlock(dqm); + + return r; +} + +int debug_refresh_runlist(struct device_queue_manager *dqm) +{ + int r = debug_lock_and_unmap(dqm); + + if (r) + return r; + + return debug_map_and_unlock(dqm); +} + #if defined(CONFIG_DEBUG_FS) static void seq_reg_dump(struct seq_file *m, @@ -2452,52 +3152,66 @@ static void seq_reg_dump(struct seq_file *m, int dqm_debugfs_hqds(struct seq_file *m, void *data) { struct device_queue_manager *dqm = data; + uint32_t xcc_mask = dqm->dev->xcc_mask; uint32_t (*dump)[2], n_regs; int pipe, queue; - int r = 0; + int r = 0, xcc_id; + uint32_t sdma_engine_start; if (!dqm->sched_running) { seq_puts(m, " Device is stopped\n"); return 0; } - r = dqm->dev->kfd2kgd->hqd_dump(dqm->dev->adev, - KFD_CIK_HIQ_PIPE, KFD_CIK_HIQ_QUEUE, - &dump, &n_regs); - if (!r) { - seq_printf(m, " HIQ on MEC %d Pipe %d Queue %d\n", - KFD_CIK_HIQ_PIPE/get_pipes_per_mec(dqm)+1, - KFD_CIK_HIQ_PIPE%get_pipes_per_mec(dqm), - KFD_CIK_HIQ_QUEUE); - seq_reg_dump(m, dump, n_regs); + for_each_inst(xcc_id, xcc_mask) { + r = dqm->dev->kfd2kgd->hqd_dump(dqm->dev->adev, + KFD_CIK_HIQ_PIPE, + KFD_CIK_HIQ_QUEUE, &dump, + &n_regs, xcc_id); + if (!r) { + seq_printf( + m, + " Inst %d, HIQ on MEC %d Pipe %d Queue %d\n", + xcc_id, + KFD_CIK_HIQ_PIPE / get_pipes_per_mec(dqm) + 1, + KFD_CIK_HIQ_PIPE % get_pipes_per_mec(dqm), + KFD_CIK_HIQ_QUEUE); + seq_reg_dump(m, dump, n_regs); - kfree(dump); - } + kfree(dump); + } - for (pipe = 0; pipe < get_pipes_per_mec(dqm); pipe++) { - int pipe_offset = pipe * get_queues_per_pipe(dqm); + for (pipe = 0; pipe < get_pipes_per_mec(dqm); pipe++) { + int pipe_offset = pipe * get_queues_per_pipe(dqm); - for (queue = 0; queue < get_queues_per_pipe(dqm); queue++) { - if (!test_bit(pipe_offset + queue, - dqm->dev->shared_resources.cp_queue_bitmap)) - continue; + for (queue = 0; queue < get_queues_per_pipe(dqm); queue++) { + if (!test_bit(pipe_offset + queue, + dqm->dev->kfd->shared_resources.cp_queue_bitmap)) + continue; - r = dqm->dev->kfd2kgd->hqd_dump( - dqm->dev->adev, pipe, queue, &dump, &n_regs); - if (r) - break; + r = dqm->dev->kfd2kgd->hqd_dump(dqm->dev->adev, + pipe, queue, + &dump, &n_regs, + xcc_id); + if (r) + break; - seq_printf(m, " CP Pipe %d, Queue %d\n", - pipe, queue); - seq_reg_dump(m, dump, n_regs); + seq_printf(m, + " Inst %d, CP Pipe %d, Queue %d\n", + xcc_id, pipe, queue); + seq_reg_dump(m, dump, n_regs); - kfree(dump); + kfree(dump); + } } } - for (pipe = 0; pipe < get_num_all_sdma_engines(dqm); pipe++) { + sdma_engine_start = dqm->dev->node_id * get_num_all_sdma_engines(dqm); + for (pipe = sdma_engine_start; + pipe < (sdma_engine_start + get_num_all_sdma_engines(dqm)); + pipe++) { for (queue = 0; - queue < dqm->dev->device_info.num_sdma_queues_per_engine; + queue < dqm->dev->kfd->device_info.num_sdma_queues_per_engine; queue++) { r = dqm->dev->kfd2kgd->hqd_sdma_dump( dqm->dev->adev, pipe, queue, &dump, &n_regs); @@ -2526,7 +3240,8 @@ int dqm_debugfs_hang_hws(struct device_queue_manager *dqm) return r; } dqm->active_runlist = true; - r = execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0); + r = execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, + 0, USE_DEFAULT_GRACE_PERIOD); dqm_unlock(dqm); return r; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h index a537b9ef3e1679e34cd5b890dd221d8f6375ab62..7dd4b177219dec5c36931c1a65c3f36f6d4609b4 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h @@ -37,6 +37,7 @@ #define KFD_MES_PROCESS_QUANTUM 100000 #define KFD_MES_GANG_QUANTUM 10000 +#define USE_DEFAULT_GRACE_PERIOD 0xffffffff struct device_process_node { struct qcm_process_device *qpd; @@ -207,7 +208,7 @@ struct device_queue_manager_asic_ops { struct queue *q, struct qcm_process_device *qpd); struct mqd_manager * (*mqd_manager_init)(enum KFD_MQD_TYPE type, - struct kfd_dev *dev); + struct kfd_node *dev); }; /** @@ -228,7 +229,7 @@ struct device_queue_manager { struct mqd_manager *mqd_mgrs[KFD_MQD_TYPE_MAX]; struct packet_manager packet_mgr; - struct kfd_dev *dev; + struct kfd_node *dev; struct mutex lock_hidden; /* use dqm_lock/unlock(dqm) */ struct list_head queues; unsigned int saved_flags; @@ -239,8 +240,8 @@ struct device_queue_manager { unsigned int total_queue_count; unsigned int next_pipe_to_allocate; unsigned int *allocated_queues; - uint64_t sdma_bitmap; - uint64_t xgmi_sdma_bitmap; + DECLARE_BITMAP(sdma_bitmap, KFD_MAX_SDMA_QUEUES); + DECLARE_BITMAP(xgmi_sdma_bitmap, KFD_MAX_SDMA_QUEUES); /* the pasid mapping for each kfd vmid */ uint16_t vmid_pasid[VMID_NUM]; uint64_t pipelines_addr; @@ -249,6 +250,7 @@ struct device_queue_manager { struct kfd_mem_obj *fence_mem; bool active_runlist; int sched_policy; + uint32_t trap_debug_vmid; /* hw exception */ bool is_hws_hang; @@ -256,6 +258,13 @@ struct device_queue_manager { struct work_struct hw_exception_work; struct kfd_mem_obj hiq_sdma_mqd; bool sched_running; + + /* used for GFX 9.4.3 only */ + uint32_t current_logical_xcc_start; + + uint32_t wait_times; + + wait_queue_head_t destroy_wait; }; void device_queue_manager_init_cik( @@ -279,6 +288,24 @@ unsigned int get_queues_per_pipe(struct device_queue_manager *dqm); unsigned int get_pipes_per_mec(struct device_queue_manager *dqm); unsigned int get_num_sdma_queues(struct device_queue_manager *dqm); unsigned int get_num_xgmi_sdma_queues(struct device_queue_manager *dqm); +int reserve_debug_trap_vmid(struct device_queue_manager *dqm, + struct qcm_process_device *qpd); +int release_debug_trap_vmid(struct device_queue_manager *dqm, + struct qcm_process_device *qpd); +int suspend_queues(struct kfd_process *p, + uint32_t num_queues, + uint32_t grace_period, + uint64_t exception_clear_mask, + uint32_t *usr_queue_id_array); +int resume_queues(struct kfd_process *p, + uint32_t num_queues, + uint32_t *usr_queue_id_array); +void set_queue_snapshot_entry(struct queue *q, + uint64_t exception_clear_mask, + struct kfd_queue_snapshot_entry *qss_entry); +int debug_lock_and_unmap(struct device_queue_manager *dqm); +int debug_map_and_unlock(struct device_queue_manager *dqm); +int debug_refresh_runlist(struct device_queue_manager *dqm); static inline unsigned int get_sh_mem_bases_32(struct kfd_process_device *pdd) { diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c index 8b2dd2670ab7fad43093ca897ea5501001c270ef..8af64338876877f10e20e525de8f02a74db3bb20 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c @@ -24,9 +24,7 @@ #include "kfd_device_queue_manager.h" #include "vega10_enum.h" -#include "gc/gc_9_0_offset.h" -#include "gc/gc_9_0_sh_mask.h" -#include "sdma0/sdma0_4_0_sh_mask.h" +#include "gc/gc_9_4_3_sh_mask.h" static int update_qpd_v9(struct device_queue_manager *dqm, struct qcm_process_device *qpd); @@ -62,9 +60,13 @@ static int update_qpd_v9(struct device_queue_manager *dqm, qpd->sh_mem_config = SH_MEM_ALIGNMENT_MODE_UNALIGNED << SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT; - if (dqm->dev->noretry && !dqm->dev->use_iommu_v2) + if (dqm->dev->kfd->noretry && !dqm->dev->kfd->use_iommu_v2) qpd->sh_mem_config |= 1 << SH_MEM_CONFIG__RETRY_DISABLE__SHIFT; + if (KFD_GC_VERSION(dqm->dev->kfd) == IP_VERSION(9, 4, 3)) + qpd->sh_mem_config |= + (1 << SH_MEM_CONFIG__F8_MODE__SHIFT); + qpd->sh_mem_ape1_limit = 0; qpd->sh_mem_ape1_base = 0; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c b/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c index 38c9e1ca66913006ed8b62b3e215f9828dd212c5..6421b620388decdb9849fd4d970f20ebf610fc74 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c @@ -138,7 +138,7 @@ void kfd_doorbell_fini(struct kfd_dev *kfd) iounmap(kfd->doorbell_kernel_ptr); } -int kfd_doorbell_mmap(struct kfd_dev *dev, struct kfd_process *process, +int kfd_doorbell_mmap(struct kfd_node *dev, struct kfd_process *process, struct vm_area_struct *vma) { phys_addr_t address; @@ -148,7 +148,7 @@ int kfd_doorbell_mmap(struct kfd_dev *dev, struct kfd_process *process, * For simplicitly we only allow mapping of the entire doorbell * allocation of a single device & process. */ - if (vma->vm_end - vma->vm_start != kfd_doorbell_process_slice(dev)) + if (vma->vm_end - vma->vm_start != kfd_doorbell_process_slice(dev->kfd)) return -EINVAL; pdd = kfd_get_process_device_data(dev, process); @@ -170,13 +170,13 @@ int kfd_doorbell_mmap(struct kfd_dev *dev, struct kfd_process *process, " vm_flags == 0x%04lX\n" " size == 0x%04lX\n", (unsigned long long) vma->vm_start, address, vma->vm_flags, - kfd_doorbell_process_slice(dev)); + kfd_doorbell_process_slice(dev->kfd)); return io_remap_pfn_range(vma, vma->vm_start, address >> PAGE_SHIFT, - kfd_doorbell_process_slice(dev), + kfd_doorbell_process_slice(dev->kfd), vma->vm_page_prot); } @@ -278,14 +278,14 @@ uint64_t kfd_get_number_elems(struct kfd_dev *kfd) phys_addr_t kfd_get_process_doorbells(struct kfd_process_device *pdd) { if (!pdd->doorbell_index) { - int r = kfd_alloc_process_doorbells(pdd->dev, + int r = kfd_alloc_process_doorbells(pdd->dev->kfd, &pdd->doorbell_index); if (r < 0) return 0; } - return pdd->dev->doorbell_base + - pdd->doorbell_index * kfd_doorbell_process_slice(pdd->dev); + return pdd->dev->kfd->doorbell_base + + pdd->doorbell_index * kfd_doorbell_process_slice(pdd->dev->kfd); } int kfd_alloc_process_doorbells(struct kfd_dev *kfd, unsigned int *doorbell_index) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c index c894cf8f7c50d468c6ef57a6a93b1d07dbb15975..8081a9408006e388d4ea5dc3ebad422159717983 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c @@ -41,6 +41,7 @@ struct kfd_event_waiter { wait_queue_entry_t wait; struct kfd_event *event; /* Event to wait for */ bool activated; /* Becomes true when event is signaled */ + bool event_age_enabled; /* set to true when last_event_age is non-zero */ }; /* @@ -348,7 +349,7 @@ static int kfd_event_page_set(struct kfd_process *p, void *kernel_address, int kfd_kmap_event_page(struct kfd_process *p, uint64_t event_page_offset) { - struct kfd_dev *kfd; + struct kfd_node *kfd; struct kfd_process_device *pdd; void *mem, *kern_addr; uint64_t size; @@ -431,6 +432,7 @@ int kfd_event_create(struct file *devkfd, struct kfd_process *p, if (!ret) { *event_id = ev->event_id; *event_trigger_data = ev->event_id; + ev->event_age = 1; } else { kfree(ev); } @@ -629,6 +631,11 @@ static void set_event(struct kfd_event *ev) * updating the wait queues in kfd_wait_on_events. */ ev->signaled = !ev->auto_reset || !waitqueue_active(&ev->wq); + if (!(++ev->event_age)) { + /* Never wrap back to reserved/default event age 0/1 */ + ev->event_age = 2; + WARN_ONCE(1, "event_age wrap back!"); + } list_for_each_entry(waiter, &ev->wq.head, wait.entry) WRITE_ONCE(waiter->activated, true); @@ -791,9 +798,9 @@ static struct kfd_event_waiter *alloc_event_waiters(uint32_t num_events) static int init_event_waiter(struct kfd_process *p, struct kfd_event_waiter *waiter, - uint32_t event_id) + struct kfd_event_data *event_data) { - struct kfd_event *ev = lookup_event_by_id(p, event_id); + struct kfd_event *ev = lookup_event_by_id(p, event_data->event_id); if (!ev) return -EINVAL; @@ -802,6 +809,15 @@ static int init_event_waiter(struct kfd_process *p, waiter->event = ev; waiter->activated = ev->signaled; ev->signaled = ev->signaled && !ev->auto_reset; + + /* last_event_age = 0 reserved for backward compatible */ + if (waiter->event->type == KFD_EVENT_TYPE_SIGNAL && + event_data->signal_event_data.last_event_age) { + waiter->event_age_enabled = true; + if (ev->event_age != event_data->signal_event_data.last_event_age) + waiter->activated = true; + } + if (!waiter->activated) add_wait_queue(&ev->wq, &waiter->wait); spin_unlock(&ev->lock); @@ -849,22 +865,29 @@ static int copy_signaled_event_data(uint32_t num_events, struct kfd_event_waiter *event_waiters, struct kfd_event_data __user *data) { - struct kfd_hsa_memory_exception_data *src; - struct kfd_hsa_memory_exception_data __user *dst; + void *src; + void __user *dst; struct kfd_event_waiter *waiter; struct kfd_event *event; - uint32_t i; + uint32_t i, size = 0; for (i = 0; i < num_events; i++) { waiter = &event_waiters[i]; event = waiter->event; if (!event) return -EINVAL; /* event was destroyed */ - if (waiter->activated && event->type == KFD_EVENT_TYPE_MEMORY) { - dst = &data[i].memory_exception_data; - src = &event->memory_exception_data; - if (copy_to_user(dst, src, - sizeof(struct kfd_hsa_memory_exception_data))) + if (waiter->activated) { + if (event->type == KFD_EVENT_TYPE_MEMORY) { + dst = &data[i].memory_exception_data; + src = &event->memory_exception_data; + size = sizeof(struct kfd_hsa_memory_exception_data); + } else if (event->type == KFD_EVENT_TYPE_SIGNAL && + waiter->event_age_enabled) { + dst = &data[i].signal_event_data.last_event_age; + src = &event->event_age; + size = sizeof(u64); + } + if (size && copy_to_user(dst, src, size)) return -EFAULT; } } @@ -942,8 +965,7 @@ int kfd_wait_on_events(struct kfd_process *p, goto out_unlock; } - ret = init_event_waiter(p, &event_waiters[i], - event_data.event_id); + ret = init_event_waiter(p, &event_waiters[i], &event_data); if (ret) goto out_unlock; } @@ -1125,7 +1147,7 @@ static void lookup_events_by_type_and_signal(struct kfd_process *p, } #ifdef KFD_SUPPORT_IOMMU_V2 -void kfd_signal_iommu_event(struct kfd_dev *dev, u32 pasid, +void kfd_signal_iommu_event(struct kfd_node *dev, u32 pasid, unsigned long address, bool is_write_requested, bool is_execute_requested) { @@ -1221,8 +1243,9 @@ void kfd_signal_hw_exception_event(u32 pasid) kfd_unref_process(p); } -void kfd_signal_vm_fault_event(struct kfd_dev *dev, u32 pasid, - struct kfd_vm_fault_info *info) +void kfd_signal_vm_fault_event(struct kfd_node *dev, u32 pasid, + struct kfd_vm_fault_info *info, + struct kfd_hsa_memory_exception_data *data) { struct kfd_event *ev; uint32_t id; @@ -1239,19 +1262,24 @@ void kfd_signal_vm_fault_event(struct kfd_dev *dev, u32 pasid, return; } - memset(&memory_exception_data, 0, sizeof(memory_exception_data)); - memory_exception_data.gpu_id = user_gpu_id; - memory_exception_data.failure.imprecise = true; - /* Set failure reason */ - if (info) { - memory_exception_data.va = (info->page_addr) << PAGE_SHIFT; - memory_exception_data.failure.NotPresent = - info->prot_valid ? 1 : 0; - memory_exception_data.failure.NoExecute = - info->prot_exec ? 1 : 0; - memory_exception_data.failure.ReadOnly = - info->prot_write ? 1 : 0; - memory_exception_data.failure.imprecise = 0; + /* SoC15 chips and onwards will pass in data from now on. */ + if (!data) { + memset(&memory_exception_data, 0, sizeof(memory_exception_data)); + memory_exception_data.gpu_id = user_gpu_id; + memory_exception_data.failure.imprecise = true; + + /* Set failure reason */ + if (info) { + memory_exception_data.va = (info->page_addr) << + PAGE_SHIFT; + memory_exception_data.failure.NotPresent = + info->prot_valid ? 1 : 0; + memory_exception_data.failure.NoExecute = + info->prot_exec ? 1 : 0; + memory_exception_data.failure.ReadOnly = + info->prot_write ? 1 : 0; + memory_exception_data.failure.imprecise = 0; + } } rcu_read_lock(); @@ -1260,7 +1288,8 @@ void kfd_signal_vm_fault_event(struct kfd_dev *dev, u32 pasid, idr_for_each_entry_continue(&p->event_idr, ev, id) if (ev->type == KFD_EVENT_TYPE_MEMORY) { spin_lock(&ev->lock); - ev->memory_exception_data = memory_exception_data; + ev->memory_exception_data = data ? *data : + memory_exception_data; set_event(ev); spin_unlock(&ev->lock); } @@ -1269,7 +1298,7 @@ void kfd_signal_vm_fault_event(struct kfd_dev *dev, u32 pasid, kfd_unref_process(p); } -void kfd_signal_reset_event(struct kfd_dev *dev) +void kfd_signal_reset_event(struct kfd_node *dev) { struct kfd_hsa_hw_exception_data hw_exception_data; struct kfd_hsa_memory_exception_data memory_exception_data; @@ -1325,7 +1354,7 @@ void kfd_signal_reset_event(struct kfd_dev *dev) srcu_read_unlock(&kfd_processes_srcu, idx); } -void kfd_signal_poison_consumed_event(struct kfd_dev *dev, u32 pasid) +void kfd_signal_poison_consumed_event(struct kfd_node *dev, u32 pasid) { struct kfd_process *p = kfd_lookup_process_by_pasid(pasid); struct kfd_hsa_memory_exception_data memory_exception_data; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.h b/drivers/gpu/drm/amd/amdkfd/kfd_events.h index 1c62c8dd6460182ac26d2c0c94f18382ad081e92..52ccfd397c2b35b7158aa460037c7fb484d34f27 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_events.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.h @@ -53,6 +53,7 @@ struct signal_page; struct kfd_event { u32 event_id; + u64 event_age; bool signaled; bool auto_reset; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c index 8aebe408c54495057fe3339d2090527ab7e8b26f..da2ca00d79e5e6e291939195ca346b807d3f3e5d 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c @@ -322,21 +322,21 @@ static void kfd_init_apertures_vi(struct kfd_process_device *pdd, uint8_t id) pdd->lds_base = MAKE_LDS_APP_BASE_VI(); pdd->lds_limit = MAKE_LDS_APP_LIMIT(pdd->lds_base); - if (!pdd->dev->use_iommu_v2) { + if (!pdd->dev->kfd->use_iommu_v2) { /* dGPUs: SVM aperture starting at 0 * with small reserved space for kernel. * Set them to CANONICAL addresses. */ pdd->gpuvm_base = SVM_USER_BASE; pdd->gpuvm_limit = - pdd->dev->shared_resources.gpuvm_size - 1; + pdd->dev->kfd->shared_resources.gpuvm_size - 1; } else { /* set them to non CANONICAL addresses, and no SVM is * allocated. */ pdd->gpuvm_base = MAKE_GPUVM_APP_BASE_VI(id + 1); pdd->gpuvm_limit = MAKE_GPUVM_APP_LIMIT(pdd->gpuvm_base, - pdd->dev->shared_resources.gpuvm_size); + pdd->dev->kfd->shared_resources.gpuvm_size); } pdd->scratch_base = MAKE_SCRATCH_APP_BASE_VI(); @@ -356,7 +356,7 @@ static void kfd_init_apertures_v9(struct kfd_process_device *pdd, uint8_t id) */ pdd->gpuvm_base = SVM_USER_BASE; pdd->gpuvm_limit = - pdd->dev->shared_resources.gpuvm_size - 1; + pdd->dev->kfd->shared_resources.gpuvm_size - 1; pdd->scratch_base = MAKE_SCRATCH_APP_BASE_V9(); pdd->scratch_limit = MAKE_SCRATCH_APP_LIMIT(pdd->scratch_base); @@ -365,7 +365,7 @@ static void kfd_init_apertures_v9(struct kfd_process_device *pdd, uint8_t id) int kfd_init_apertures(struct kfd_process *process) { uint8_t id = 0; - struct kfd_dev *dev; + struct kfd_node *dev; struct kfd_process_device *pdd; /*Iterating over all devices*/ @@ -417,7 +417,7 @@ int kfd_init_apertures(struct kfd_process *process) } } - if (!dev->use_iommu_v2) { + if (!dev->kfd->use_iommu_v2) { /* dGPUs: the reserved space for kernel * before SVM */ diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c new file mode 100644 index 0000000000000000000000000000000000000000..c7991e07b6be56d97b85ced8508d29e988519b0f --- /dev/null +++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c @@ -0,0 +1,405 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ + +#include "kfd_events.h" +#include "kfd_debug.h" +#include "soc15_int.h" +#include "kfd_device_queue_manager.h" + +/* + * GFX10 SQ Interrupts + * + * There are 3 encoding types of interrupts sourced from SQ sent as a 44-bit + * packet to the Interrupt Handler: + * Auto - Generated by the SQG (various cmd overflows, timestamps etc) + * Wave - Generated by S_SENDMSG through a shader program + * Error - HW generated errors (Illegal instructions, Memviols, EDC etc) + * + * The 44-bit packet is mapped as {context_id1[7:0],context_id0[31:0]} plus + * 4-bits for VMID (SOC15_VMID_FROM_IH_ENTRY) as such: + * + * - context_id1[7:6] + * Encoding type (0 = Auto, 1 = Wave, 2 = Error) + * + * - context_id0[24] + * PRIV bit indicates that Wave S_SEND or error occurred within trap + * + * - context_id0[22:0] + * 23-bit data with the following layout per encoding type: + * Auto - only context_id0[8:0] is used, which reports various interrupts + * generated by SQG. The rest is 0. + * Wave - user data sent from m0 via S_SENDMSG + * Error - Error type (context_id0[22:19]), Error Details (rest of bits) + * + * The other context_id bits show coordinates (SE/SH/CU/SIMD/WGP) for wave + * S_SENDMSG and Errors. These are 0 for Auto. + */ + +enum SQ_INTERRUPT_WORD_ENCODING { + SQ_INTERRUPT_WORD_ENCODING_AUTO = 0x0, + SQ_INTERRUPT_WORD_ENCODING_INST, + SQ_INTERRUPT_WORD_ENCODING_ERROR, +}; + +enum SQ_INTERRUPT_ERROR_TYPE { + SQ_INTERRUPT_ERROR_TYPE_EDC_FUE = 0x0, + SQ_INTERRUPT_ERROR_TYPE_ILLEGAL_INST, + SQ_INTERRUPT_ERROR_TYPE_MEMVIOL, + SQ_INTERRUPT_ERROR_TYPE_EDC_FED, +}; + +/* SQ_INTERRUPT_WORD_AUTO_CTXID */ +#define SQ_INTERRUPT_WORD_AUTO_CTXID0__THREAD_TRACE__SHIFT 0 +#define SQ_INTERRUPT_WORD_AUTO_CTXID0__WLT__SHIFT 1 +#define SQ_INTERRUPT_WORD_AUTO_CTXID0__THREAD_TRACE_BUF0_FULL__SHIFT 2 +#define SQ_INTERRUPT_WORD_AUTO_CTXID0__THREAD_TRACE_BUF1_FULL__SHIFT 3 +#define SQ_INTERRUPT_WORD_AUTO_CTXID0__THREAD_TRACE_UTC_ERROR__SHIFT 7 +#define SQ_INTERRUPT_WORD_AUTO_CTXID1__SE_ID__SHIFT 4 +#define SQ_INTERRUPT_WORD_AUTO_CTXID1__ENCODING__SHIFT 6 + +#define SQ_INTERRUPT_WORD_AUTO_CTXID0__THREAD_TRACE_MASK 0x00000001 +#define SQ_INTERRUPT_WORD_AUTO_CTXID0__WLT_MASK 0x00000002 +#define SQ_INTERRUPT_WORD_AUTO_CTXID0__THREAD_TRACE_BUF0_FULL_MASK 0x00000004 +#define SQ_INTERRUPT_WORD_AUTO_CTXID0__THREAD_TRACE_BUF1_FULL_MASK 0x00000008 +#define SQ_INTERRUPT_WORD_AUTO_CTXID0__THREAD_TRACE_UTC_ERROR_MASK 0x00000080 +#define SQ_INTERRUPT_WORD_AUTO_CTXID1__SE_ID_MASK 0x030 +#define SQ_INTERRUPT_WORD_AUTO_CTXID1__ENCODING_MASK 0x0c0 + +/* SQ_INTERRUPT_WORD_WAVE_CTXID */ +#define SQ_INTERRUPT_WORD_WAVE_CTXID0__DATA__SHIFT 0 +#define SQ_INTERRUPT_WORD_WAVE_CTXID0__SA_ID__SHIFT 23 +#define SQ_INTERRUPT_WORD_WAVE_CTXID0__PRIV__SHIFT 24 +#define SQ_INTERRUPT_WORD_WAVE_CTXID0__WAVE_ID__SHIFT 25 +#define SQ_INTERRUPT_WORD_WAVE_CTXID0__SIMD_ID__SHIFT 30 +#define SQ_INTERRUPT_WORD_WAVE_CTXID1__WGP_ID__SHIFT 0 +#define SQ_INTERRUPT_WORD_WAVE_CTXID1__SE_ID__SHIFT 4 +#define SQ_INTERRUPT_WORD_WAVE_CTXID1__ENCODING__SHIFT 6 + +#define SQ_INTERRUPT_WORD_WAVE_CTXID0__DATA_MASK 0x000007fffff +#define SQ_INTERRUPT_WORD_WAVE_CTXID0__SA_ID_MASK 0x0000800000 +#define SQ_INTERRUPT_WORD_WAVE_CTXID0__PRIV_MASK 0x00001000000 +#define SQ_INTERRUPT_WORD_WAVE_CTXID0__WAVE_ID_MASK 0x0003e000000 +#define SQ_INTERRUPT_WORD_WAVE_CTXID0__SIMD_ID_MASK 0x000c0000000 +#define SQ_INTERRUPT_WORD_WAVE_CTXID1__WGP_ID_MASK 0x00f +#define SQ_INTERRUPT_WORD_WAVE_CTXID1__SE_ID_MASK 0x030 +#define SQ_INTERRUPT_WORD_WAVE_CTXID1__ENCODING_MASK 0x0c0 + +#define KFD_CTXID0__ERR_TYPE_MASK 0x780000 +#define KFD_CTXID0__ERR_TYPE__SHIFT 19 + +/* GFX10 SQ interrupt ENC type bit (context_id1[7:6]) for wave s_sendmsg */ +#define KFD_CONTEXT_ID1_ENC_TYPE_WAVE_MASK 0x40 +/* GFX10 SQ interrupt PRIV bit (context_id0[24]) for s_sendmsg inside trap */ +#define KFD_CONTEXT_ID0_PRIV_MASK 0x1000000 +/* + * The debugger will send user data(m0) with PRIV=1 to indicate it requires + * notification from the KFD with the following queue id (DOORBELL_ID) and + * trap code (TRAP_CODE). + */ +#define KFD_CONTEXT_ID0_DEBUG_DOORBELL_MASK 0x0003ff +#define KFD_CONTEXT_ID0_DEBUG_TRAP_CODE_SHIFT 10 +#define KFD_CONTEXT_ID0_DEBUG_TRAP_CODE_MASK 0x07fc00 +#define KFD_DEBUG_DOORBELL_ID(ctxid0) ((ctxid0) & \ + KFD_CONTEXT_ID0_DEBUG_DOORBELL_MASK) +#define KFD_DEBUG_TRAP_CODE(ctxid0) (((ctxid0) & \ + KFD_CONTEXT_ID0_DEBUG_TRAP_CODE_MASK) \ + >> KFD_CONTEXT_ID0_DEBUG_TRAP_CODE_SHIFT) +#define KFD_DEBUG_CP_BAD_OP_ECODE_MASK 0x3fffc00 +#define KFD_DEBUG_CP_BAD_OP_ECODE_SHIFT 10 +#define KFD_DEBUG_CP_BAD_OP_ECODE(ctxid0) (((ctxid0) & \ + KFD_DEBUG_CP_BAD_OP_ECODE_MASK) \ + >> KFD_DEBUG_CP_BAD_OP_ECODE_SHIFT) + +static void event_interrupt_poison_consumption(struct kfd_node *dev, + uint16_t pasid, uint16_t client_id) +{ + int old_poison, ret = -EINVAL; + struct kfd_process *p = kfd_lookup_process_by_pasid(pasid); + + if (!p) + return; + + /* all queues of a process will be unmapped in one time */ + old_poison = atomic_cmpxchg(&p->poison, 0, 1); + kfd_unref_process(p); + if (old_poison) + return; + + switch (client_id) { + case SOC15_IH_CLIENTID_SE0SH: + case SOC15_IH_CLIENTID_SE1SH: + case SOC15_IH_CLIENTID_SE2SH: + case SOC15_IH_CLIENTID_SE3SH: + case SOC15_IH_CLIENTID_UTCL2: + ret = kfd_dqm_evict_pasid(dev->dqm, pasid); + break; + case SOC15_IH_CLIENTID_SDMA0: + case SOC15_IH_CLIENTID_SDMA1: + case SOC15_IH_CLIENTID_SDMA2: + case SOC15_IH_CLIENTID_SDMA3: + case SOC15_IH_CLIENTID_SDMA4: + break; + default: + break; + } + + kfd_signal_poison_consumed_event(dev, pasid); + + /* resetting queue passes, do page retirement without gpu reset + * resetting queue fails, fallback to gpu reset solution + */ + if (!ret) { + dev_warn(dev->adev->dev, + "RAS poison consumption, unmap queue flow succeeded: client id %d\n", + client_id); + amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, false); + } else { + dev_warn(dev->adev->dev, + "RAS poison consumption, fall back to gpu reset flow: client id %d\n", + client_id); + amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, true); + } +} + +static bool event_interrupt_isr_v10(struct kfd_node *dev, + const uint32_t *ih_ring_entry, + uint32_t *patched_ihre, + bool *patched_flag) +{ + uint16_t source_id, client_id, pasid, vmid; + const uint32_t *data = ih_ring_entry; + + source_id = SOC15_SOURCE_ID_FROM_IH_ENTRY(ih_ring_entry); + client_id = SOC15_CLIENT_ID_FROM_IH_ENTRY(ih_ring_entry); + + /* Only handle interrupts from KFD VMIDs */ + vmid = SOC15_VMID_FROM_IH_ENTRY(ih_ring_entry); + if (!KFD_IRQ_IS_FENCE(client_id, source_id) && + (vmid < dev->vm_info.first_vmid_kfd || + vmid > dev->vm_info.last_vmid_kfd)) + return false; + + pasid = SOC15_PASID_FROM_IH_ENTRY(ih_ring_entry); + + /* Only handle clients we care about */ + if (client_id != SOC15_IH_CLIENTID_GRBM_CP && + client_id != SOC15_IH_CLIENTID_SDMA0 && + client_id != SOC15_IH_CLIENTID_SDMA1 && + client_id != SOC15_IH_CLIENTID_SDMA2 && + client_id != SOC15_IH_CLIENTID_SDMA3 && + client_id != SOC15_IH_CLIENTID_SDMA4 && + client_id != SOC15_IH_CLIENTID_SDMA5 && + client_id != SOC15_IH_CLIENTID_SDMA6 && + client_id != SOC15_IH_CLIENTID_SDMA7 && + client_id != SOC15_IH_CLIENTID_VMC && + client_id != SOC15_IH_CLIENTID_VMC1 && + client_id != SOC15_IH_CLIENTID_UTCL2 && + client_id != SOC15_IH_CLIENTID_SE0SH && + client_id != SOC15_IH_CLIENTID_SE1SH && + client_id != SOC15_IH_CLIENTID_SE2SH && + client_id != SOC15_IH_CLIENTID_SE3SH) + return false; + + pr_debug("client id 0x%x, source id %d, vmid %d, pasid 0x%x. raw data:\n", + client_id, source_id, vmid, pasid); + pr_debug("%8X, %8X, %8X, %8X, %8X, %8X, %8X, %8X.\n", + data[0], data[1], data[2], data[3], + data[4], data[5], data[6], data[7]); + + /* If there is no valid PASID, it's likely a bug */ + if (WARN_ONCE(pasid == 0, "Bug: No PASID in KFD interrupt")) + return 0; + + /* Interrupt types we care about: various signals and faults. + * They will be forwarded to a work queue (see below). + */ + return source_id == SOC15_INTSRC_CP_END_OF_PIPE || + source_id == SOC15_INTSRC_SDMA_TRAP || + source_id == SOC15_INTSRC_SQ_INTERRUPT_MSG || + source_id == SOC15_INTSRC_CP_BAD_OPCODE || + client_id == SOC15_IH_CLIENTID_VMC || + client_id == SOC15_IH_CLIENTID_VMC1 || + client_id == SOC15_IH_CLIENTID_UTCL2 || + KFD_IRQ_IS_FENCE(client_id, source_id); +} + +static void event_interrupt_wq_v10(struct kfd_node *dev, + const uint32_t *ih_ring_entry) +{ + uint16_t source_id, client_id, pasid, vmid; + uint32_t context_id0, context_id1; + uint32_t encoding, sq_intr_err_type; + + source_id = SOC15_SOURCE_ID_FROM_IH_ENTRY(ih_ring_entry); + client_id = SOC15_CLIENT_ID_FROM_IH_ENTRY(ih_ring_entry); + pasid = SOC15_PASID_FROM_IH_ENTRY(ih_ring_entry); + vmid = SOC15_VMID_FROM_IH_ENTRY(ih_ring_entry); + context_id0 = SOC15_CONTEXT_ID0_FROM_IH_ENTRY(ih_ring_entry); + context_id1 = SOC15_CONTEXT_ID1_FROM_IH_ENTRY(ih_ring_entry); + + if (client_id == SOC15_IH_CLIENTID_GRBM_CP || + client_id == SOC15_IH_CLIENTID_SE0SH || + client_id == SOC15_IH_CLIENTID_SE1SH || + client_id == SOC15_IH_CLIENTID_SE2SH || + client_id == SOC15_IH_CLIENTID_SE3SH) { + if (source_id == SOC15_INTSRC_CP_END_OF_PIPE) + kfd_signal_event_interrupt(pasid, context_id0, 32); + else if (source_id == SOC15_INTSRC_SQ_INTERRUPT_MSG) { + encoding = REG_GET_FIELD(context_id1, + SQ_INTERRUPT_WORD_WAVE_CTXID1, ENCODING); + switch (encoding) { + case SQ_INTERRUPT_WORD_ENCODING_AUTO: + pr_debug( + "sq_intr: auto, se %d, ttrace %d, wlt %d, ttrac_buf0_full %d, ttrac_buf1_full %d, ttrace_utc_err %d\n", + REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_AUTO_CTXID1, + SE_ID), + REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0, + THREAD_TRACE), + REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0, + WLT), + REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0, + THREAD_TRACE_BUF0_FULL), + REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0, + THREAD_TRACE_BUF1_FULL), + REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_AUTO_CTXID0, + THREAD_TRACE_UTC_ERROR)); + break; + case SQ_INTERRUPT_WORD_ENCODING_INST: + pr_debug("sq_intr: inst, se %d, data 0x%x, sa %d, priv %d, wave_id %d, simd_id %d, wgp_id %d\n", + REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_WAVE_CTXID1, + SE_ID), + REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0, + DATA), + REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0, + SA_ID), + REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0, + PRIV), + REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0, + WAVE_ID), + REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0, + SIMD_ID), + REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_WAVE_CTXID1, + WGP_ID)); + if (context_id0 & SQ_INTERRUPT_WORD_WAVE_CTXID0__PRIV_MASK) { + if (kfd_set_dbg_ev_from_interrupt(dev, pasid, + KFD_DEBUG_DOORBELL_ID(context_id0), + KFD_DEBUG_TRAP_CODE(context_id0), + NULL, 0)) + return; + } + break; + case SQ_INTERRUPT_WORD_ENCODING_ERROR: + sq_intr_err_type = REG_GET_FIELD(context_id0, KFD_CTXID0, + ERR_TYPE); + pr_warn("sq_intr: error, se %d, data 0x%x, sa %d, priv %d, wave_id %d, simd_id %d, wgp_id %d, err_type %d\n", + REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_WAVE_CTXID1, + SE_ID), + REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0, + DATA), + REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0, + SA_ID), + REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0, + PRIV), + REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0, + WAVE_ID), + REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0, + SIMD_ID), + REG_GET_FIELD(context_id1, SQ_INTERRUPT_WORD_WAVE_CTXID1, + WGP_ID), + sq_intr_err_type); + if (sq_intr_err_type != SQ_INTERRUPT_ERROR_TYPE_ILLEGAL_INST && + sq_intr_err_type != SQ_INTERRUPT_ERROR_TYPE_MEMVIOL) { + event_interrupt_poison_consumption(dev, pasid, source_id); + return; + } + break; + default: + break; + } + kfd_signal_event_interrupt(pasid, context_id0 & 0x7fffff, 23); + } else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE) { + kfd_set_dbg_ev_from_interrupt(dev, pasid, + KFD_DEBUG_DOORBELL_ID(context_id0), + KFD_EC_MASK(KFD_DEBUG_CP_BAD_OP_ECODE(context_id0)), + NULL, + 0); + } + } else if (client_id == SOC15_IH_CLIENTID_SDMA0 || + client_id == SOC15_IH_CLIENTID_SDMA1 || + client_id == SOC15_IH_CLIENTID_SDMA2 || + client_id == SOC15_IH_CLIENTID_SDMA3 || + (client_id == SOC15_IH_CLIENTID_SDMA3_Sienna_Cichlid && + KFD_GC_VERSION(dev) == IP_VERSION(10, 3, 0)) || + client_id == SOC15_IH_CLIENTID_SDMA4 || + client_id == SOC15_IH_CLIENTID_SDMA5 || + client_id == SOC15_IH_CLIENTID_SDMA6 || + client_id == SOC15_IH_CLIENTID_SDMA7) { + if (source_id == SOC15_INTSRC_SDMA_TRAP) { + kfd_signal_event_interrupt(pasid, context_id0 & 0xfffffff, 28); + } else if (source_id == SOC15_INTSRC_SDMA_ECC) { + event_interrupt_poison_consumption(dev, pasid, source_id); + return; + } + } else if (client_id == SOC15_IH_CLIENTID_VMC || + client_id == SOC15_IH_CLIENTID_VMC1 || + client_id == SOC15_IH_CLIENTID_UTCL2) { + struct kfd_vm_fault_info info = {0}; + uint16_t ring_id = SOC15_RING_ID_FROM_IH_ENTRY(ih_ring_entry); + struct kfd_hsa_memory_exception_data exception_data; + + if (client_id == SOC15_IH_CLIENTID_UTCL2 && + amdgpu_amdkfd_ras_query_utcl2_poison_status(dev->adev)) { + event_interrupt_poison_consumption(dev, pasid, client_id); + return; + } + + info.vmid = vmid; + info.mc_id = client_id; + info.page_addr = ih_ring_entry[4] | + (uint64_t)(ih_ring_entry[5] & 0xf) << 32; + info.prot_valid = ring_id & 0x08; + info.prot_read = ring_id & 0x10; + info.prot_write = ring_id & 0x20; + + memset(&exception_data, 0, sizeof(exception_data)); + exception_data.gpu_id = dev->id; + exception_data.va = (info.page_addr) << PAGE_SHIFT; + exception_data.failure.NotPresent = info.prot_valid ? 1 : 0; + exception_data.failure.NoExecute = info.prot_exec ? 1 : 0; + exception_data.failure.ReadOnly = info.prot_write ? 1 : 0; + exception_data.failure.imprecise = 0; + + kfd_set_dbg_ev_from_interrupt(dev, + pasid, + -1, + KFD_EC_MASK(EC_DEVICE_MEMORY_VIOLATION), + &exception_data, + sizeof(exception_data)); + } else if (KFD_IRQ_IS_FENCE(client_id, source_id)) { + kfd_process_close_interrupt_drain(pasid); + } +} + +const struct kfd_event_interrupt_class event_interrupt_class_v10 = { + .interrupt_isr = event_interrupt_isr_v10, + .interrupt_wq = event_interrupt_wq_v10, +}; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c index 0d53f60674226b6abe0b1ec4d1377051a60e91ff..f933bd231fb9cea8e7ce125d80cd0e31c9e06112 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c @@ -26,6 +26,7 @@ #include "kfd_device_queue_manager.h" #include "ivsrcid/vmc/irqsrcs_vmc_1_0.h" #include "kfd_smi_events.h" +#include "kfd_debug.h" /* * GFX11 SQ Interrupts @@ -187,7 +188,7 @@ static void print_sq_intr_info_error(uint32_t context_id0, uint32_t context_id1) REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_ERROR_CTXID1, WGP_ID)); } -static void event_interrupt_poison_consumption_v11(struct kfd_dev *dev, +static void event_interrupt_poison_consumption_v11(struct kfd_node *dev, uint16_t pasid, uint16_t source_id) { int ret = -EINVAL; @@ -225,7 +226,7 @@ static void event_interrupt_poison_consumption_v11(struct kfd_dev *dev, amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, true); } -static bool event_interrupt_isr_v11(struct kfd_dev *dev, +static bool event_interrupt_isr_v11(struct kfd_node *dev, const uint32_t *ih_ring_entry, uint32_t *patched_ihre, bool *patched_flag) @@ -238,7 +239,7 @@ static bool event_interrupt_isr_v11(struct kfd_dev *dev, client_id = SOC15_CLIENT_ID_FROM_IH_ENTRY(ih_ring_entry); /* Only handle interrupts from KFD VMIDs */ vmid = SOC15_VMID_FROM_IH_ENTRY(ih_ring_entry); - if (/*!KFD_IRQ_IS_FENCE(client_id, source_id) &&*/ + if (!KFD_IRQ_IS_FENCE(client_id, source_id) && (vmid < dev->vm_info.first_vmid_kfd || vmid > dev->vm_info.last_vmid_kfd)) return false; @@ -267,19 +268,19 @@ static bool event_interrupt_isr_v11(struct kfd_dev *dev, source_id == SOC15_INTSRC_SQ_INTERRUPT_MSG || source_id == SOC15_INTSRC_CP_BAD_OPCODE || source_id == SOC21_INTSRC_SDMA_TRAP || - /* KFD_IRQ_IS_FENCE(client_id, source_id) || */ + KFD_IRQ_IS_FENCE(client_id, source_id) || (((client_id == SOC21_IH_CLIENTID_VMC) || ((client_id == SOC21_IH_CLIENTID_GFX) && (source_id == UTCL2_1_0__SRCID__FAULT))) && !amdgpu_no_queue_eviction_on_vm_fault); } -static void event_interrupt_wq_v11(struct kfd_dev *dev, +static void event_interrupt_wq_v11(struct kfd_node *dev, const uint32_t *ih_ring_entry) { uint16_t source_id, client_id, ring_id, pasid, vmid; uint32_t context_id0, context_id1; - uint8_t sq_int_enc, sq_int_errtype, sq_int_priv; + uint8_t sq_int_enc, sq_int_priv, sq_int_errtype; struct kfd_vm_fault_info info = {0}; struct kfd_hsa_memory_exception_data exception_data; @@ -312,9 +313,9 @@ static void event_interrupt_wq_v11(struct kfd_dev *dev, exception_data.failure.ReadOnly = info.prot_write ? 1 : 0; exception_data.failure.imprecise = 0; - /*kfd_set_dbg_ev_from_interrupt(dev, pasid, -1, + kfd_set_dbg_ev_from_interrupt(dev, pasid, -1, KFD_EC_MASK(EC_DEVICE_MEMORY_VIOLATION), - &exception_data, sizeof(exception_data));*/ + &exception_data, sizeof(exception_data)); kfd_smi_event_update_vmfault(dev, pasid); /* GRBM, SDMA, SE, PMM */ @@ -324,11 +325,11 @@ static void event_interrupt_wq_v11(struct kfd_dev *dev, /* CP */ if (source_id == SOC15_INTSRC_CP_END_OF_PIPE) kfd_signal_event_interrupt(pasid, context_id0, 32); - /*else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE) + else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE) kfd_set_dbg_ev_from_interrupt(dev, pasid, KFD_CTXID0_DOORBELL_ID(context_id0), KFD_EC_MASK(KFD_CTXID0_CP_BAD_OP_ECODE(context_id0)), - NULL, 0);*/ + NULL, 0); /* SDMA */ else if (source_id == SOC21_INTSRC_SDMA_TRAP) @@ -350,11 +351,11 @@ static void event_interrupt_wq_v11(struct kfd_dev *dev, print_sq_intr_info_inst(context_id0, context_id1); sq_int_priv = REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID0, PRIV); - /*if (sq_int_priv && (kfd_set_dbg_ev_from_interrupt(dev, pasid, + if (sq_int_priv && (kfd_set_dbg_ev_from_interrupt(dev, pasid, KFD_CTXID0_DOORBELL_ID(context_id0), KFD_CTXID0_TRAP_CODE(context_id0), NULL, 0))) - return;*/ + return; break; case SQ_INTERRUPT_WORD_ENCODING_ERROR: print_sq_intr_info_error(context_id0, context_id1); @@ -373,8 +374,8 @@ static void event_interrupt_wq_v11(struct kfd_dev *dev, kfd_signal_event_interrupt(pasid, context_id0 & 0xffffff, 24); } - /*} else if (KFD_IRQ_IS_FENCE(client_id, source_id)) { - kfd_process_close_interrupt_drain(pasid);*/ + } else if (KFD_IRQ_IS_FENCE(client_id, source_id)) { + kfd_process_close_interrupt_drain(pasid); } } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c index 0b75a37b689b8c27ec0a0ee8df6d0bc78f68c92a..d5c9f30552e3bd17e7621b33896ebe6a1f9b92ef 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c @@ -23,10 +23,40 @@ #include "kfd_priv.h" #include "kfd_events.h" +#include "kfd_debug.h" #include "soc15_int.h" #include "kfd_device_queue_manager.h" #include "kfd_smi_events.h" +/* + * GFX9 SQ Interrupts + * + * There are 3 encoding types of interrupts sourced from SQ sent as a 44-bit + * packet to the Interrupt Handler: + * Auto - Generated by the SQG (various cmd overflows, timestamps etc) + * Wave - Generated by S_SENDMSG through a shader program + * Error - HW generated errors (Illegal instructions, Memviols, EDC etc) + * + * The 44-bit packet is mapped as {context_id1[7:0],context_id0[31:0]} plus + * 4-bits for VMID (SOC15_VMID_FROM_IH_ENTRY) as such: + * + * - context_id0[27:26] + * Encoding type (0 = Auto, 1 = Wave, 2 = Error) + * + * - context_id0[13] + * PRIV bit indicates that Wave S_SEND or error occurred within trap + * + * - {context_id1[7:0],context_id0[31:28],context_id0[11:0]} + * 24-bit data with the following layout per encoding type: + * Auto - only context_id0[8:0] is used, which reports various interrupts + * generated by SQG. The rest is 0. + * Wave - user data sent from m0 via S_SENDMSG + * Error - Error type (context_id1[7:4]), Error Details (rest of bits) + * + * The other context_id bits show coordinates (SE/SH/CU/SIMD/WAVE) for wave + * S_SENDMSG and Errors. These are 0 for Auto. + */ + enum SQ_INTERRUPT_WORD_ENCODING { SQ_INTERRUPT_WORD_ENCODING_AUTO = 0x0, SQ_INTERRUPT_WORD_ENCODING_INST, @@ -84,13 +114,33 @@ enum SQ_INTERRUPT_ERROR_TYPE { #define SQ_INTERRUPT_WORD_WAVE_CTXID__SE_ID_MASK 0x03000000 #define SQ_INTERRUPT_WORD_WAVE_CTXID__ENCODING_MASK 0x0c000000 +/* GFX9 SQ interrupt 24-bit data from context_id<0,1> */ #define KFD_CONTEXT_ID_GET_SQ_INT_DATA(ctx0, ctx1) \ ((ctx0 & 0xfff) | ((ctx0 >> 16) & 0xf000) | ((ctx1 << 16) & 0xff0000)) #define KFD_SQ_INT_DATA__ERR_TYPE_MASK 0xF00000 #define KFD_SQ_INT_DATA__ERR_TYPE__SHIFT 20 -static void event_interrupt_poison_consumption_v9(struct kfd_dev *dev, +/* + * The debugger will send user data(m0) with PRIV=1 to indicate it requires + * notification from the KFD with the following queue id (DOORBELL_ID) and + * trap code (TRAP_CODE). + */ +#define KFD_INT_DATA_DEBUG_DOORBELL_MASK 0x0003ff +#define KFD_INT_DATA_DEBUG_TRAP_CODE_SHIFT 10 +#define KFD_INT_DATA_DEBUG_TRAP_CODE_MASK 0x07fc00 +#define KFD_DEBUG_DOORBELL_ID(sq_int_data) ((sq_int_data) & \ + KFD_INT_DATA_DEBUG_DOORBELL_MASK) +#define KFD_DEBUG_TRAP_CODE(sq_int_data) (((sq_int_data) & \ + KFD_INT_DATA_DEBUG_TRAP_CODE_MASK) \ + >> KFD_INT_DATA_DEBUG_TRAP_CODE_SHIFT) +#define KFD_DEBUG_CP_BAD_OP_ECODE_MASK 0x3fffc00 +#define KFD_DEBUG_CP_BAD_OP_ECODE_SHIFT 10 +#define KFD_DEBUG_CP_BAD_OP_ECODE(ctxid0) (((ctxid0) & \ + KFD_DEBUG_CP_BAD_OP_ECODE_MASK) \ + >> KFD_DEBUG_CP_BAD_OP_ECODE_SHIFT) + +static void event_interrupt_poison_consumption_v9(struct kfd_node *dev, uint16_t pasid, uint16_t client_id) { int old_poison, ret = -EINVAL; @@ -160,7 +210,7 @@ static bool context_id_expected(struct kfd_dev *dev) } } -static bool event_interrupt_isr_v9(struct kfd_dev *dev, +static bool event_interrupt_isr_v9(struct kfd_node *dev, const uint32_t *ih_ring_entry, uint32_t *patched_ihre, bool *patched_flag) @@ -168,14 +218,16 @@ static bool event_interrupt_isr_v9(struct kfd_dev *dev, uint16_t source_id, client_id, pasid, vmid; const uint32_t *data = ih_ring_entry; + source_id = SOC15_SOURCE_ID_FROM_IH_ENTRY(ih_ring_entry); + client_id = SOC15_CLIENT_ID_FROM_IH_ENTRY(ih_ring_entry); + /* Only handle interrupts from KFD VMIDs */ vmid = SOC15_VMID_FROM_IH_ENTRY(ih_ring_entry); - if (vmid < dev->vm_info.first_vmid_kfd || - vmid > dev->vm_info.last_vmid_kfd) + if (!KFD_IRQ_IS_FENCE(client_id, source_id) && + (vmid < dev->vm_info.first_vmid_kfd || + vmid > dev->vm_info.last_vmid_kfd)) return false; - source_id = SOC15_SOURCE_ID_FROM_IH_ENTRY(ih_ring_entry); - client_id = SOC15_CLIENT_ID_FROM_IH_ENTRY(ih_ring_entry); pasid = SOC15_PASID_FROM_IH_ENTRY(ih_ring_entry); /* Only handle clients we care about */ @@ -194,7 +246,8 @@ static bool event_interrupt_isr_v9(struct kfd_dev *dev, client_id != SOC15_IH_CLIENTID_SE0SH && client_id != SOC15_IH_CLIENTID_SE1SH && client_id != SOC15_IH_CLIENTID_SE2SH && - client_id != SOC15_IH_CLIENTID_SE3SH) + client_id != SOC15_IH_CLIENTID_SE3SH && + !KFD_IRQ_IS_FENCE(client_id, source_id)) return false; /* This is a known issue for gfx9. Under non HWS, pasid is not set @@ -206,7 +259,7 @@ static bool event_interrupt_isr_v9(struct kfd_dev *dev, *patched_flag = true; memcpy(patched_ihre, ih_ring_entry, - dev->device_info.ih_ring_entry_size); + dev->kfd->device_info.ih_ring_entry_size); pasid = dev->dqm->vmid_pasid[vmid]; @@ -235,7 +288,7 @@ static bool event_interrupt_isr_v9(struct kfd_dev *dev, uint32_t context_id = SOC15_CONTEXT_ID0_FROM_IH_ENTRY(ih_ring_entry); - if (context_id == 0 && context_id_expected(dev)) + if (context_id == 0 && context_id_expected(dev->kfd)) return false; } @@ -247,13 +300,14 @@ static bool event_interrupt_isr_v9(struct kfd_dev *dev, source_id == SOC15_INTSRC_SDMA_ECC || source_id == SOC15_INTSRC_SQ_INTERRUPT_MSG || source_id == SOC15_INTSRC_CP_BAD_OPCODE || + KFD_IRQ_IS_FENCE(client_id, source_id) || ((client_id == SOC15_IH_CLIENTID_VMC || client_id == SOC15_IH_CLIENTID_VMC1 || client_id == SOC15_IH_CLIENTID_UTCL2) && !amdgpu_no_queue_eviction_on_vm_fault); } -static void event_interrupt_wq_v9(struct kfd_dev *dev, +static void event_interrupt_wq_v9(struct kfd_node *dev, const uint32_t *ih_ring_entry) { uint16_t source_id, client_id, pasid, vmid; @@ -302,6 +356,13 @@ static void event_interrupt_wq_v9(struct kfd_dev *dev, REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, SIMD_ID), REG_GET_FIELD(context_id0, SQ_INTERRUPT_WORD_WAVE_CTXID, CU_ID), sq_int_data); + if (context_id0 & SQ_INTERRUPT_WORD_WAVE_CTXID__PRIV_MASK) { + if (kfd_set_dbg_ev_from_interrupt(dev, pasid, + KFD_DEBUG_DOORBELL_ID(sq_int_data), + KFD_DEBUG_TRAP_CODE(sq_int_data), + NULL, 0)) + return; + } break; case SQ_INTERRUPT_WORD_ENCODING_ERROR: sq_intr_err = REG_GET_FIELD(sq_int_data, KFD_SQ_INT_DATA, ERR_TYPE); @@ -324,8 +385,12 @@ static void event_interrupt_wq_v9(struct kfd_dev *dev, break; } kfd_signal_event_interrupt(pasid, context_id0 & 0xffffff, 24); - } else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE) - kfd_signal_hw_exception_event(pasid); + } else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE) { + kfd_set_dbg_ev_from_interrupt(dev, pasid, + KFD_DEBUG_DOORBELL_ID(context_id0), + KFD_EC_MASK(KFD_DEBUG_CP_BAD_OP_ECODE(context_id0)), + NULL, 0); + } } else if (client_id == SOC15_IH_CLIENTID_SDMA0 || client_id == SOC15_IH_CLIENTID_SDMA1 || client_id == SOC15_IH_CLIENTID_SDMA2 || @@ -345,6 +410,7 @@ static void event_interrupt_wq_v9(struct kfd_dev *dev, client_id == SOC15_IH_CLIENTID_UTCL2) { struct kfd_vm_fault_info info = {0}; uint16_t ring_id = SOC15_RING_ID_FROM_IH_ENTRY(ih_ring_entry); + struct kfd_hsa_memory_exception_data exception_data; if (client_id == SOC15_IH_CLIENTID_UTCL2 && amdgpu_amdkfd_ras_query_utcl2_poison_status(dev->adev)) { @@ -360,9 +426,23 @@ static void event_interrupt_wq_v9(struct kfd_dev *dev, info.prot_read = ring_id & 0x10; info.prot_write = ring_id & 0x20; + memset(&exception_data, 0, sizeof(exception_data)); + exception_data.gpu_id = dev->id; + exception_data.va = (info.page_addr) << PAGE_SHIFT; + exception_data.failure.NotPresent = info.prot_valid ? 1 : 0; + exception_data.failure.NoExecute = info.prot_exec ? 1 : 0; + exception_data.failure.ReadOnly = info.prot_write ? 1 : 0; + exception_data.failure.imprecise = 0; + + kfd_set_dbg_ev_from_interrupt(dev, + pasid, + -1, + KFD_EC_MASK(EC_DEVICE_MEMORY_VIOLATION), + &exception_data, + sizeof(exception_data)); kfd_smi_event_update_vmfault(dev, pasid); - kfd_dqm_evict_pasid(dev->dqm, pasid); - kfd_signal_vm_fault_event(dev, pasid, &info); + } else if (KFD_IRQ_IS_FENCE(client_id, source_id)) { + kfd_process_close_interrupt_drain(pasid); } } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c b/drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c index 34772fe74296043df6a0adbcbda3a28978f7ad91..dd3c43c1ad70bdb0022f9d3ab533070c5edb87c8 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c @@ -50,29 +50,29 @@ static void interrupt_wq(struct work_struct *); -int kfd_interrupt_init(struct kfd_dev *kfd) +int kfd_interrupt_init(struct kfd_node *node) { int r; - r = kfifo_alloc(&kfd->ih_fifo, - KFD_IH_NUM_ENTRIES * kfd->device_info.ih_ring_entry_size, + r = kfifo_alloc(&node->ih_fifo, + KFD_IH_NUM_ENTRIES * node->kfd->device_info.ih_ring_entry_size, GFP_KERNEL); if (r) { - dev_err(kfd->adev->dev, "Failed to allocate IH fifo\n"); + dev_err(node->adev->dev, "Failed to allocate IH fifo\n"); return r; } - kfd->ih_wq = alloc_workqueue("KFD IH", WQ_HIGHPRI, 1); - if (unlikely(!kfd->ih_wq)) { - kfifo_free(&kfd->ih_fifo); - dev_err(kfd->adev->dev, "Failed to allocate KFD IH workqueue\n"); + node->ih_wq = alloc_workqueue("KFD IH", WQ_HIGHPRI, 1); + if (unlikely(!node->ih_wq)) { + kfifo_free(&node->ih_fifo); + dev_err(node->adev->dev, "Failed to allocate KFD IH workqueue\n"); return -ENOMEM; } - spin_lock_init(&kfd->interrupt_lock); + spin_lock_init(&node->interrupt_lock); - INIT_WORK(&kfd->interrupt_work, interrupt_wq); + INIT_WORK(&node->interrupt_work, interrupt_wq); - kfd->interrupts_active = true; + node->interrupts_active = true; /* * After this function returns, the interrupt will be enabled. This @@ -84,7 +84,7 @@ int kfd_interrupt_init(struct kfd_dev *kfd) return 0; } -void kfd_interrupt_exit(struct kfd_dev *kfd) +void kfd_interrupt_exit(struct kfd_node *node) { /* * Stop the interrupt handler from writing to the ring and scheduling @@ -93,31 +93,31 @@ void kfd_interrupt_exit(struct kfd_dev *kfd) */ unsigned long flags; - spin_lock_irqsave(&kfd->interrupt_lock, flags); - kfd->interrupts_active = false; - spin_unlock_irqrestore(&kfd->interrupt_lock, flags); + spin_lock_irqsave(&node->interrupt_lock, flags); + node->interrupts_active = false; + spin_unlock_irqrestore(&node->interrupt_lock, flags); /* * flush_work ensures that there are no outstanding * work-queue items that will access interrupt_ring. New work items * can't be created because we stopped interrupt handling above. */ - flush_workqueue(kfd->ih_wq); + flush_workqueue(node->ih_wq); - kfifo_free(&kfd->ih_fifo); + kfifo_free(&node->ih_fifo); } /* * Assumption: single reader/writer. This function is not re-entrant */ -bool enqueue_ih_ring_entry(struct kfd_dev *kfd, const void *ih_ring_entry) +bool enqueue_ih_ring_entry(struct kfd_node *node, const void *ih_ring_entry) { int count; - count = kfifo_in(&kfd->ih_fifo, ih_ring_entry, - kfd->device_info.ih_ring_entry_size); - if (count != kfd->device_info.ih_ring_entry_size) { - dev_dbg_ratelimited(kfd->adev->dev, + count = kfifo_in(&node->ih_fifo, ih_ring_entry, + node->kfd->device_info.ih_ring_entry_size); + if (count != node->kfd->device_info.ih_ring_entry_size) { + dev_dbg_ratelimited(node->adev->dev, "Interrupt ring overflow, dropping interrupt %d\n", count); return false; @@ -129,32 +129,32 @@ bool enqueue_ih_ring_entry(struct kfd_dev *kfd, const void *ih_ring_entry) /* * Assumption: single reader/writer. This function is not re-entrant */ -static bool dequeue_ih_ring_entry(struct kfd_dev *kfd, void *ih_ring_entry) +static bool dequeue_ih_ring_entry(struct kfd_node *node, void *ih_ring_entry) { int count; - count = kfifo_out(&kfd->ih_fifo, ih_ring_entry, - kfd->device_info.ih_ring_entry_size); + count = kfifo_out(&node->ih_fifo, ih_ring_entry, + node->kfd->device_info.ih_ring_entry_size); - WARN_ON(count && count != kfd->device_info.ih_ring_entry_size); + WARN_ON(count && count != node->kfd->device_info.ih_ring_entry_size); - return count == kfd->device_info.ih_ring_entry_size; + return count == node->kfd->device_info.ih_ring_entry_size; } static void interrupt_wq(struct work_struct *work) { - struct kfd_dev *dev = container_of(work, struct kfd_dev, + struct kfd_node *dev = container_of(work, struct kfd_node, interrupt_work); uint32_t ih_ring_entry[KFD_MAX_RING_ENTRY_SIZE]; unsigned long start_jiffies = jiffies; - if (dev->device_info.ih_ring_entry_size > sizeof(ih_ring_entry)) { + if (dev->kfd->device_info.ih_ring_entry_size > sizeof(ih_ring_entry)) { dev_err_once(dev->adev->dev, "Ring entry too small\n"); return; } while (dequeue_ih_ring_entry(dev, ih_ring_entry)) { - dev->device_info.event_interrupt_class->interrupt_wq(dev, + dev->kfd->device_info.event_interrupt_class->interrupt_wq(dev, ih_ring_entry); if (time_is_before_jiffies(start_jiffies + HZ)) { /* If we spent more than a second processing signals, @@ -166,14 +166,14 @@ static void interrupt_wq(struct work_struct *work) } } -bool interrupt_is_wanted(struct kfd_dev *dev, +bool interrupt_is_wanted(struct kfd_node *dev, const uint32_t *ih_ring_entry, uint32_t *patched_ihre, bool *flag) { /* integer and bitwise OR so there is no boolean short-circuiting */ unsigned int wanted = 0; - wanted |= dev->device_info.event_interrupt_class->interrupt_isr(dev, + wanted |= dev->kfd->device_info.event_interrupt_class->interrupt_isr(dev, ih_ring_entry, patched_ihre, flag); return wanted != 0; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c index ec1bf611624ef25d47c03c18ae4ffa66d77232a5..808ee010520a242b42300bb7ca3bf453879d88a1 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c @@ -109,11 +109,11 @@ int kfd_iommu_device_init(struct kfd_dev *kfd) */ int kfd_iommu_bind_process_to_device(struct kfd_process_device *pdd) { - struct kfd_dev *dev = pdd->dev; + struct kfd_node *dev = pdd->dev; struct kfd_process *p = pdd->process; int err; - if (!dev->use_iommu_v2 || pdd->bound == PDD_BOUND) + if (!dev->kfd->use_iommu_v2 || pdd->bound == PDD_BOUND) return 0; if (unlikely(pdd->bound == PDD_BOUND_SUSPENDED)) { @@ -121,6 +121,12 @@ int kfd_iommu_bind_process_to_device(struct kfd_process_device *pdd) return -EINVAL; } + if (!kfd_is_first_node(dev)) { + dev_warn_once(kfd_device, + "IOMMU supported only on first node\n"); + return 0; + } + err = amd_iommu_bind_pasid(dev->adev->pdev, p->pasid, p->lead_thread); if (!err) pdd->bound = PDD_BOUND; @@ -138,7 +144,8 @@ void kfd_iommu_unbind_process(struct kfd_process *p) int i; for (i = 0; i < p->n_pdds; i++) - if (p->pdds[i]->bound == PDD_BOUND) + if ((p->pdds[i]->bound == PDD_BOUND) && + (kfd_is_first_node((p->pdds[i]->dev)))) amd_iommu_unbind_pasid(p->pdds[i]->dev->adev->pdev, p->pasid); } @@ -146,7 +153,7 @@ void kfd_iommu_unbind_process(struct kfd_process *p) /* Callback for process shutdown invoked by the IOMMU driver */ static void iommu_pasid_shutdown_callback(struct pci_dev *pdev, u32 pasid) { - struct kfd_dev *dev = kfd_device_by_pci_dev(pdev); + struct kfd_node *dev = kfd_device_by_pci_dev(pdev); struct kfd_process *p; struct kfd_process_device *pdd; @@ -182,7 +189,7 @@ static void iommu_pasid_shutdown_callback(struct pci_dev *pdev, u32 pasid) static int iommu_invalid_ppr_cb(struct pci_dev *pdev, u32 pasid, unsigned long address, u16 flags) { - struct kfd_dev *dev; + struct kfd_node *dev; dev_warn_ratelimited(kfd_device, "Invalid PPR device %x:%x.%x pasid 0x%x address 0x%lX flags 0x%X", @@ -205,7 +212,7 @@ static int iommu_invalid_ppr_cb(struct pci_dev *pdev, u32 pasid, * Bind processes do the device that have been temporarily unbound * (PDD_BOUND_SUSPENDED) in kfd_unbind_processes_from_device. */ -static int kfd_bind_processes_to_device(struct kfd_dev *kfd) +static int kfd_bind_processes_to_device(struct kfd_node *knode) { struct kfd_process_device *pdd; struct kfd_process *p; @@ -216,14 +223,14 @@ static int kfd_bind_processes_to_device(struct kfd_dev *kfd) hash_for_each_rcu(kfd_processes_table, temp, p, kfd_processes) { mutex_lock(&p->mutex); - pdd = kfd_get_process_device_data(kfd, p); + pdd = kfd_get_process_device_data(knode, p); if (WARN_ON(!pdd) || pdd->bound != PDD_BOUND_SUSPENDED) { mutex_unlock(&p->mutex); continue; } - err = amd_iommu_bind_pasid(kfd->adev->pdev, p->pasid, + err = amd_iommu_bind_pasid(knode->adev->pdev, p->pasid, p->lead_thread); if (err < 0) { pr_err("Unexpected pasid 0x%x binding failure\n", @@ -246,7 +253,7 @@ static int kfd_bind_processes_to_device(struct kfd_dev *kfd) * processes will be restored to PDD_BOUND state in * kfd_bind_processes_to_device. */ -static void kfd_unbind_processes_from_device(struct kfd_dev *kfd) +static void kfd_unbind_processes_from_device(struct kfd_node *knode) { struct kfd_process_device *pdd; struct kfd_process *p; @@ -256,7 +263,7 @@ static void kfd_unbind_processes_from_device(struct kfd_dev *kfd) hash_for_each_rcu(kfd_processes_table, temp, p, kfd_processes) { mutex_lock(&p->mutex); - pdd = kfd_get_process_device_data(kfd, p); + pdd = kfd_get_process_device_data(knode, p); if (WARN_ON(!pdd)) { mutex_unlock(&p->mutex); @@ -281,7 +288,7 @@ void kfd_iommu_suspend(struct kfd_dev *kfd) if (!kfd->use_iommu_v2) return; - kfd_unbind_processes_from_device(kfd); + kfd_unbind_processes_from_device(kfd->nodes[0]); amd_iommu_set_invalidate_ctx_cb(kfd->adev->pdev, NULL); amd_iommu_set_invalid_ppr_cb(kfd->adev->pdev, NULL); @@ -312,7 +319,7 @@ int kfd_iommu_resume(struct kfd_dev *kfd) amd_iommu_set_invalid_ppr_cb(kfd->adev->pdev, iommu_invalid_ppr_cb); - err = kfd_bind_processes_to_device(kfd); + err = kfd_bind_processes_to_device(kfd->nodes[0]); if (err) { amd_iommu_set_invalidate_ctx_cb(kfd->adev->pdev, NULL); amd_iommu_set_invalid_ppr_cb(kfd->adev->pdev, NULL); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c index bcf7bc3302c9f65e12bb0eeef992ab1e7f061410..1bea629c49ca027d7c11f80c6a9bb7105b7a47f7 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c @@ -38,7 +38,7 @@ /* Initialize a kernel queue, including allocations of GART memory * needed for the queue. */ -static bool kq_initialize(struct kernel_queue *kq, struct kfd_dev *dev, +static bool kq_initialize(struct kernel_queue *kq, struct kfd_node *dev, enum kfd_queue_type type, unsigned int queue_size) { struct queue_properties prop; @@ -75,7 +75,7 @@ static bool kq_initialize(struct kernel_queue *kq, struct kfd_dev *dev, if (!kq->mqd_mgr) return false; - prop.doorbell_ptr = kfd_get_kernel_doorbell(dev, &prop.doorbell_off); + prop.doorbell_ptr = kfd_get_kernel_doorbell(dev->kfd, &prop.doorbell_off); if (!prop.doorbell_ptr) { pr_err("Failed to initialize doorbell"); @@ -112,7 +112,7 @@ static bool kq_initialize(struct kernel_queue *kq, struct kfd_dev *dev, kq->rptr_kernel = kq->rptr_mem->cpu_ptr; kq->rptr_gpu_addr = kq->rptr_mem->gpu_addr; - retval = kfd_gtt_sa_allocate(dev, dev->device_info.doorbell_size, + retval = kfd_gtt_sa_allocate(dev, dev->kfd->device_info.doorbell_size, &kq->wptr_mem); if (retval != 0) @@ -189,7 +189,7 @@ static bool kq_initialize(struct kernel_queue *kq, struct kfd_dev *dev, err_eop_allocate_vidmem: kfd_gtt_sa_free(dev, kq->pq); err_pq_allocate_vidmem: - kfd_release_kernel_doorbell(dev, prop.doorbell_ptr); + kfd_release_kernel_doorbell(dev->kfd, prop.doorbell_ptr); err_get_kernel_doorbell: return false; @@ -220,7 +220,7 @@ static void kq_uninitialize(struct kernel_queue *kq, bool hanging) kfd_gtt_sa_free(kq->dev, kq->eop_mem); kfd_gtt_sa_free(kq->dev, kq->pq); - kfd_release_kernel_doorbell(kq->dev, + kfd_release_kernel_doorbell(kq->dev->kfd, kq->queue->properties.doorbell_ptr); uninit_queue(kq->queue); } @@ -298,7 +298,7 @@ void kq_submit_packet(struct kernel_queue *kq) } pr_debug("\n"); #endif - if (kq->dev->device_info.doorbell_size == 8) { + if (kq->dev->kfd->device_info.doorbell_size == 8) { *kq->wptr64_kernel = kq->pending_wptr64; write_kernel_doorbell64(kq->queue->properties.doorbell_ptr, kq->pending_wptr64); @@ -311,7 +311,7 @@ void kq_submit_packet(struct kernel_queue *kq) void kq_rollback_packet(struct kernel_queue *kq) { - if (kq->dev->device_info.doorbell_size == 8) { + if (kq->dev->kfd->device_info.doorbell_size == 8) { kq->pending_wptr64 = *kq->wptr64_kernel; kq->pending_wptr = *kq->wptr_kernel % (kq->queue->properties.queue_size / 4); @@ -320,7 +320,7 @@ void kq_rollback_packet(struct kernel_queue *kq) } } -struct kernel_queue *kernel_queue_init(struct kfd_dev *dev, +struct kernel_queue *kernel_queue_init(struct kfd_node *dev, enum kfd_queue_type type) { struct kernel_queue *kq; @@ -345,7 +345,7 @@ void kernel_queue_uninit(struct kernel_queue *kq, bool hanging) } /* FIXME: Can this test be removed? */ -static __attribute__((unused)) void test_kq(struct kfd_dev *dev) +static __attribute__((unused)) void test_kq(struct kfd_node *dev) { struct kernel_queue *kq; uint32_t *buffer, i; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.h b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.h index 383202fd1ea283529cac9d8b205bb6bb32f85f5e..9a624443084543f4b294b377e7c15d46e0efb7a6 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.h @@ -53,7 +53,7 @@ void kq_rollback_packet(struct kernel_queue *kq); struct kernel_queue { /* data */ - struct kfd_dev *dev; + struct kfd_node *dev; struct mqd_manager *mqd_mgr; struct queue *queue; uint64_t pending_wptr64; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c index 54933903bcb8a3a2547dc31d8354c7b1a31b3850..709ac885ca6d7b725266632c6a6ab9d2c2dc893d 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c @@ -64,7 +64,7 @@ svm_migrate_gart_map(struct amdgpu_ring *ring, uint64_t npages, num_dw = ALIGN(adev->mman.buffer_funcs->copy_num_dw, 8); num_bytes = npages * 8; - r = amdgpu_job_alloc_with_ib(adev, &adev->mman.entity, + r = amdgpu_job_alloc_with_ib(adev, &adev->mman.high_pr, AMDGPU_FENCE_OWNER_UNDEFINED, num_dw * 4 + num_bytes, AMDGPU_IB_POOL_DELAYED, @@ -206,7 +206,7 @@ svm_migrate_copy_done(struct amdgpu_device *adev, struct dma_fence *mfence) unsigned long svm_migrate_addr_to_pfn(struct amdgpu_device *adev, unsigned long addr) { - return (addr + adev->kfd.dev->pgmap.range.start) >> PAGE_SHIFT; + return (addr + adev->kfd.pgmap.range.start) >> PAGE_SHIFT; } static void @@ -236,7 +236,7 @@ svm_migrate_addr(struct amdgpu_device *adev, struct page *page) unsigned long addr; addr = page_to_pfn(page) << PAGE_SHIFT; - return (addr - adev->kfd.dev->pgmap.range.start); + return (addr - adev->kfd.pgmap.range.start); } static struct page * @@ -287,11 +287,12 @@ static unsigned long svm_migrate_unsuccessful_pages(struct migrate_vma *migrate) } static int -svm_migrate_copy_to_vram(struct amdgpu_device *adev, struct svm_range *prange, +svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange, struct migrate_vma *migrate, struct dma_fence **mfence, dma_addr_t *scratch, uint64_t ttm_res_offset) { - uint64_t npages = migrate->npages; + uint64_t npages = migrate->cpages; + struct amdgpu_device *adev = node->adev; struct device *dev = adev->dev; struct amdgpu_res_cursor cursor; dma_addr_t *src; @@ -321,7 +322,7 @@ svm_migrate_copy_to_vram(struct amdgpu_device *adev, struct svm_range *prange, DMA_TO_DEVICE); r = dma_mapping_error(dev, src[i]); if (r) { - dev_err(adev->dev, "%s: fail %d dma_map_page\n", + dev_err(dev, "%s: fail %d dma_map_page\n", __func__, r); goto out_free_vram_pages; } @@ -390,12 +391,13 @@ svm_migrate_copy_to_vram(struct amdgpu_device *adev, struct svm_range *prange, } static long -svm_migrate_vma_to_vram(struct amdgpu_device *adev, struct svm_range *prange, +svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange, struct vm_area_struct *vma, uint64_t start, uint64_t end, uint32_t trigger, uint64_t ttm_res_offset) { struct kfd_process *p = container_of(prange->svms, struct kfd_process, svms); uint64_t npages = (end - start) >> PAGE_SHIFT; + struct amdgpu_device *adev = node->adev; struct kfd_process_device *pdd; struct dma_fence *mfence = NULL; struct migrate_vma migrate = { 0 }; @@ -421,9 +423,9 @@ svm_migrate_vma_to_vram(struct amdgpu_device *adev, struct svm_range *prange, migrate.dst = migrate.src + npages; scratch = (dma_addr_t *)(migrate.dst + npages); - kfd_smi_event_migration_start(adev->kfd.dev, p->lead_thread->pid, + kfd_smi_event_migration_start(node, p->lead_thread->pid, start >> PAGE_SHIFT, end >> PAGE_SHIFT, - 0, adev->kfd.dev->id, prange->prefetch_loc, + 0, node->id, prange->prefetch_loc, prange->preferred_loc, trigger); r = migrate_vma_setup(&migrate); @@ -445,7 +447,7 @@ svm_migrate_vma_to_vram(struct amdgpu_device *adev, struct svm_range *prange, else pr_debug("0x%lx pages migrated\n", cpages); - r = svm_migrate_copy_to_vram(adev, prange, &migrate, &mfence, scratch, ttm_res_offset); + r = svm_migrate_copy_to_vram(node, prange, &migrate, &mfence, scratch, ttm_res_offset); migrate_vma_pages(&migrate); pr_debug("successful/cpages/npages 0x%lx/0x%lx/0x%lx\n", @@ -454,9 +456,9 @@ svm_migrate_vma_to_vram(struct amdgpu_device *adev, struct svm_range *prange, svm_migrate_copy_done(adev, mfence); migrate_vma_finalize(&migrate); - kfd_smi_event_migration_end(adev->kfd.dev, p->lead_thread->pid, + kfd_smi_event_migration_end(node, p->lead_thread->pid, start >> PAGE_SHIFT, end >> PAGE_SHIFT, - 0, adev->kfd.dev->id, trigger); + 0, node->id, trigger); svm_range_dma_unmap(adev->dev, scratch, 0, npages); svm_range_free_dma_mappings(prange); @@ -465,7 +467,7 @@ svm_migrate_vma_to_vram(struct amdgpu_device *adev, struct svm_range *prange, kvfree(buf); out: if (!r && cpages) { - pdd = svm_range_get_pdd_by_adev(prange, adev); + pdd = svm_range_get_pdd_by_node(prange, node); if (pdd) WRITE_ONCE(pdd->page_in, pdd->page_in + cpages); @@ -492,8 +494,8 @@ svm_migrate_ram_to_vram(struct svm_range *prange, uint32_t best_loc, { unsigned long addr, start, end; struct vm_area_struct *vma; - struct amdgpu_device *adev; uint64_t ttm_res_offset; + struct kfd_node *node; unsigned long cpages = 0; long r = 0; @@ -503,9 +505,9 @@ svm_migrate_ram_to_vram(struct svm_range *prange, uint32_t best_loc, return 0; } - adev = svm_range_get_adev_by_id(prange, best_loc); - if (!adev) { - pr_debug("failed to get device by id 0x%x\n", best_loc); + node = svm_range_get_node_by_id(prange, best_loc); + if (!node) { + pr_debug("failed to get kfd node by id 0x%x\n", best_loc); return -ENODEV; } @@ -515,9 +517,9 @@ svm_migrate_ram_to_vram(struct svm_range *prange, uint32_t best_loc, start = prange->start << PAGE_SHIFT; end = (prange->last + 1) << PAGE_SHIFT; - r = svm_range_vram_node_new(adev, prange, true); + r = svm_range_vram_node_new(node, prange, true); if (r) { - dev_dbg(adev->dev, "fail %ld to alloc vram\n", r); + dev_dbg(node->adev->dev, "fail %ld to alloc vram\n", r); return r; } ttm_res_offset = prange->offset << PAGE_SHIFT; @@ -530,7 +532,7 @@ svm_migrate_ram_to_vram(struct svm_range *prange, uint32_t best_loc, break; next = min(vma->vm_end, end); - r = svm_migrate_vma_to_vram(adev, prange, vma, addr, next, trigger, ttm_res_offset); + r = svm_migrate_vma_to_vram(node, prange, vma, addr, next, trigger, ttm_res_offset); if (r < 0) { pr_debug("failed %ld to migrate\n", r); break; @@ -649,11 +651,13 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange, /** * svm_migrate_vma_to_ram - migrate range inside one vma from device to system * - * @adev: amdgpu device to migrate from * @prange: svm range structure * @vma: vm_area_struct that range [start, end] belongs to * @start: range start virtual address in pages * @end: range end virtual address in pages + * @node: kfd node device to migrate from + * @trigger: reason of migration + * @fault_page: is from vmf->page, svm_migrate_to_ram(), this is CPU page fault callback * * Context: Process context, caller hold mmap read lock, prange->migrate_mutex * @@ -663,7 +667,7 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange, * positive values - partial migration, number of pages not migrated */ static long -svm_migrate_vma_to_ram(struct amdgpu_device *adev, struct svm_range *prange, +svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange, struct vm_area_struct *vma, uint64_t start, uint64_t end, uint32_t trigger, struct page *fault_page) { @@ -671,6 +675,7 @@ svm_migrate_vma_to_ram(struct amdgpu_device *adev, struct svm_range *prange, uint64_t npages = (end - start) >> PAGE_SHIFT; unsigned long upages = npages; unsigned long cpages = 0; + struct amdgpu_device *adev = node->adev; struct kfd_process_device *pdd; struct dma_fence *mfence = NULL; struct migrate_vma migrate = { 0 }; @@ -699,9 +704,9 @@ svm_migrate_vma_to_ram(struct amdgpu_device *adev, struct svm_range *prange, migrate.fault_page = fault_page; scratch = (dma_addr_t *)(migrate.dst + npages); - kfd_smi_event_migration_start(adev->kfd.dev, p->lead_thread->pid, + kfd_smi_event_migration_start(node, p->lead_thread->pid, start >> PAGE_SHIFT, end >> PAGE_SHIFT, - adev->kfd.dev->id, 0, prange->prefetch_loc, + node->id, 0, prange->prefetch_loc, prange->preferred_loc, trigger); r = migrate_vma_setup(&migrate); @@ -735,9 +740,9 @@ svm_migrate_vma_to_ram(struct amdgpu_device *adev, struct svm_range *prange, svm_migrate_copy_done(adev, mfence); migrate_vma_finalize(&migrate); - kfd_smi_event_migration_end(adev->kfd.dev, p->lead_thread->pid, + kfd_smi_event_migration_end(node, p->lead_thread->pid, start >> PAGE_SHIFT, end >> PAGE_SHIFT, - adev->kfd.dev->id, 0, trigger); + node->id, 0, trigger); svm_range_dma_unmap(adev->dev, scratch, 0, npages); @@ -745,7 +750,7 @@ svm_migrate_vma_to_ram(struct amdgpu_device *adev, struct svm_range *prange, kvfree(buf); out: if (!r && cpages) { - pdd = svm_range_get_pdd_by_adev(prange, adev); + pdd = svm_range_get_pdd_by_node(prange, node); if (pdd) WRITE_ONCE(pdd->page_out, pdd->page_out + cpages); } @@ -757,6 +762,7 @@ svm_migrate_vma_to_ram(struct amdgpu_device *adev, struct svm_range *prange, * @prange: range structure * @mm: process mm, use current->mm if NULL * @trigger: reason of migration + * @fault_page: is from vmf->page, svm_migrate_to_ram(), this is CPU page fault callback * * Context: Process context, caller hold mmap read lock, prange->migrate_mutex * @@ -766,7 +772,7 @@ svm_migrate_vma_to_ram(struct amdgpu_device *adev, struct svm_range *prange, int svm_migrate_vram_to_ram(struct svm_range *prange, struct mm_struct *mm, uint32_t trigger, struct page *fault_page) { - struct amdgpu_device *adev; + struct kfd_node *node; struct vm_area_struct *vma; unsigned long addr; unsigned long start; @@ -780,13 +786,11 @@ int svm_migrate_vram_to_ram(struct svm_range *prange, struct mm_struct *mm, return 0; } - adev = svm_range_get_adev_by_id(prange, prange->actual_loc); - if (!adev) { - pr_debug("failed to get device by id 0x%x\n", - prange->actual_loc); + node = svm_range_get_node_by_id(prange, prange->actual_loc); + if (!node) { + pr_debug("failed to get kfd node by id 0x%x\n", prange->actual_loc); return -ENODEV; } - pr_debug("svms 0x%p prange 0x%p [0x%lx 0x%lx] from gpu 0x%x to ram\n", prange->svms, prange, prange->start, prange->last, prange->actual_loc); @@ -805,7 +809,7 @@ int svm_migrate_vram_to_ram(struct svm_range *prange, struct mm_struct *mm, } next = min(vma->vm_end, end); - r = svm_migrate_vma_to_ram(adev, prange, vma, addr, next, trigger, + r = svm_migrate_vma_to_ram(node, prange, vma, addr, next, trigger, fault_page); if (r < 0) { pr_debug("failed %ld to migrate prange %p\n", r, prange); @@ -987,18 +991,21 @@ static const struct dev_pagemap_ops svm_migrate_pgmap_ops = { /* Each VRAM page uses sizeof(struct page) on system memory */ #define SVM_HMM_PAGE_STRUCT_SIZE(size) ((size)/PAGE_SIZE * sizeof(struct page)) -int svm_migrate_init(struct amdgpu_device *adev) +int kgd2kfd_init_zone_device(struct amdgpu_device *adev) { - struct kfd_dev *kfddev = adev->kfd.dev; + struct amdgpu_kfd_dev *kfddev = &adev->kfd; struct dev_pagemap *pgmap; struct resource *res = NULL; unsigned long size; void *r; - /* Page migration works on Vega10 or newer */ - if (!KFD_IS_SOC15(kfddev)) + /* Page migration works on gfx9 or newer */ + if (adev->ip_versions[GC_HWIP][0] < IP_VERSION(9, 0, 1)) return -EINVAL; + if (adev->gmc.is_app_apu) + return 0; + pgmap = &kfddev->pgmap; memset(pgmap, 0, sizeof(*pgmap)); @@ -1041,8 +1048,6 @@ int svm_migrate_init(struct amdgpu_device *adev) amdgpu_amdkfd_reserve_system_mem(SVM_HMM_PAGE_STRUCT_SIZE(size)); - svm_range_set_max_pages(adev); - pr_info("HMM registered %ldMB device memory\n", size >> 20); return 0; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.h b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.h index a5d7e6d2226469a4684cea39e9158836b29c0d71..487f26368164831ccc155fe630bdbaaaddf81d62 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.h @@ -47,15 +47,6 @@ int svm_migrate_vram_to_ram(struct svm_range *prange, struct mm_struct *mm, unsigned long svm_migrate_addr_to_pfn(struct amdgpu_device *adev, unsigned long addr); -int svm_migrate_init(struct amdgpu_device *adev); - -#else - -static inline int svm_migrate_init(struct amdgpu_device *adev) -{ - return 0; -} - #endif /* IS_ENABLED(CONFIG_HSA_AMD_SVM) */ #endif /* KFD_MIGRATE_H_ */ diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c index 623ccd227b7de058263941c7a300c6379c42458a..863cf060af484dde217d24fa86074789d79accfe 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c @@ -46,7 +46,7 @@ int pipe_priority_map[] = { KFD_PIPE_PRIORITY_CS_HIGH }; -struct kfd_mem_obj *allocate_hiq_mqd(struct kfd_dev *dev, struct queue_properties *q) +struct kfd_mem_obj *allocate_hiq_mqd(struct kfd_node *dev, struct queue_properties *q) { struct kfd_mem_obj *mqd_mem_obj = NULL; @@ -61,7 +61,7 @@ struct kfd_mem_obj *allocate_hiq_mqd(struct kfd_dev *dev, struct queue_propertie return mqd_mem_obj; } -struct kfd_mem_obj *allocate_sdma_mqd(struct kfd_dev *dev, +struct kfd_mem_obj *allocate_sdma_mqd(struct kfd_node *dev, struct queue_properties *q) { struct kfd_mem_obj *mqd_mem_obj = NULL; @@ -72,11 +72,12 @@ struct kfd_mem_obj *allocate_sdma_mqd(struct kfd_dev *dev, return NULL; offset = (q->sdma_engine_id * - dev->device_info.num_sdma_queues_per_engine + + dev->kfd->device_info.num_sdma_queues_per_engine + q->sdma_queue_id) * dev->dqm->mqd_mgrs[KFD_MQD_TYPE_SDMA]->mqd_size; - offset += dev->dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ]->mqd_size; + offset += dev->dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ]->mqd_size * + NUM_XCC(dev->xcc_mask); mqd_mem_obj->gtt_mem = (void *)((uint64_t)dev->dqm->hiq_sdma_mqd.gtt_mem + offset); @@ -189,7 +190,7 @@ int kfd_hiq_load_mqd_kiq(struct mqd_manager *mm, void *mqd, struct queue_properties *p, struct mm_struct *mms) { return mm->dev->kfd2kgd->hiq_mqd_load(mm->dev->adev, mqd, pipe_id, - queue_id, p->doorbell_off); + queue_id, p->doorbell_off, 0); } int kfd_destroy_mqd_cp(struct mqd_manager *mm, void *mqd, @@ -197,7 +198,7 @@ int kfd_destroy_mqd_cp(struct mqd_manager *mm, void *mqd, uint32_t pipe_id, uint32_t queue_id) { return mm->dev->kfd2kgd->hqd_destroy(mm->dev->adev, mqd, type, timeout, - pipe_id, queue_id); + pipe_id, queue_id, 0); } void kfd_free_mqd_cp(struct mqd_manager *mm, void *mqd, @@ -216,7 +217,7 @@ bool kfd_is_occupied_cp(struct mqd_manager *mm, void *mqd, uint32_t queue_id) { return mm->dev->kfd2kgd->hqd_is_occupied(mm->dev->adev, queue_address, - pipe_id, queue_id); + pipe_id, queue_id, 0); } int kfd_load_mqd_sdma(struct mqd_manager *mm, void *mqd, @@ -246,3 +247,28 @@ bool kfd_is_occupied_sdma(struct mqd_manager *mm, void *mqd, { return mm->dev->kfd2kgd->hqd_sdma_is_occupied(mm->dev->adev, mqd); } + +uint64_t kfd_hiq_mqd_stride(struct kfd_node *dev) +{ + return dev->dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ]->mqd_size; +} + +void kfd_get_hiq_xcc_mqd(struct kfd_node *dev, struct kfd_mem_obj *mqd_mem_obj, + uint32_t virtual_xcc_id) +{ + uint64_t offset; + + offset = kfd_hiq_mqd_stride(dev) * virtual_xcc_id; + + mqd_mem_obj->gtt_mem = (virtual_xcc_id == 0) ? + dev->dqm->hiq_sdma_mqd.gtt_mem : NULL; + mqd_mem_obj->gpu_addr = dev->dqm->hiq_sdma_mqd.gpu_addr + offset; + mqd_mem_obj->cpu_ptr = (uint32_t *)((uintptr_t) + dev->dqm->hiq_sdma_mqd.cpu_ptr + offset); +} + +uint64_t kfd_mqd_stride(struct mqd_manager *mm, + struct queue_properties *q) +{ + return mm->mqd_size; +} diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h index 57f900ccaa10d396bd395f790d5ee21a3ee65aa8..23158db7da035d9820b8b14007826f75f69f3cc1 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h @@ -68,7 +68,7 @@ */ extern int pipe_priority_map[]; struct mqd_manager { - struct kfd_mem_obj* (*allocate_mqd)(struct kfd_dev *kfd, + struct kfd_mem_obj* (*allocate_mqd)(struct kfd_node *kfd, struct queue_properties *q); void (*init_mqd)(struct mqd_manager *mm, void **mqd, @@ -97,6 +97,7 @@ struct mqd_manager { uint32_t queue_id); int (*get_wave_state)(struct mqd_manager *mm, void *mqd, + struct queue_properties *q, void __user *ctl_stack, u32 *ctl_stack_used_size, u32 *save_area_used_size); @@ -119,16 +120,18 @@ struct mqd_manager { int (*debugfs_show_mqd)(struct seq_file *m, void *data); #endif uint32_t (*read_doorbell_id)(void *mqd); + uint64_t (*mqd_stride)(struct mqd_manager *mm, + struct queue_properties *p); struct mutex mqd_mutex; - struct kfd_dev *dev; + struct kfd_node *dev; uint32_t mqd_size; }; -struct kfd_mem_obj *allocate_hiq_mqd(struct kfd_dev *dev, +struct kfd_mem_obj *allocate_hiq_mqd(struct kfd_node *dev, struct queue_properties *q); -struct kfd_mem_obj *allocate_sdma_mqd(struct kfd_dev *dev, +struct kfd_mem_obj *allocate_sdma_mqd(struct kfd_node *dev, struct queue_properties *q); void free_mqd_hiq_sdma(struct mqd_manager *mm, void *mqd, struct kfd_mem_obj *mqd_mem_obj); @@ -164,4 +167,10 @@ bool kfd_is_occupied_sdma(struct mqd_manager *mm, void *mqd, uint64_t queue_address, uint32_t pipe_id, uint32_t queue_id); +void kfd_get_hiq_xcc_mqd(struct kfd_node *dev, + struct kfd_mem_obj *mqd_mem_obj, uint32_t virtual_xcc_id); + +uint64_t kfd_hiq_mqd_stride(struct kfd_node *dev); +uint64_t kfd_mqd_stride(struct mqd_manager *mm, + struct queue_properties *q); #endif /* KFD_MQD_MANAGER_H_ */ diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c index 4889865c725cef3e65e648b7432f37f2c1aa2627..65c9f01a1f86c1d60cb30c7a3524ab0e137cd40d 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c @@ -48,8 +48,7 @@ static void update_cu_mask(struct mqd_manager *mm, void *mqd, struct cik_mqd *m; uint32_t se_mask[4] = {0}; /* 4 is the max # of SEs */ - if (!minfo || (minfo->update_flag != UPDATE_FLAG_CU_MASK) || - !minfo->cu_mask.ptr) + if (!minfo || !minfo->cu_mask.ptr) return; mqd_symmetrically_map_cu_mask(mm, @@ -74,7 +73,7 @@ static void set_priority(struct cik_mqd *m, struct queue_properties *q) m->cp_hqd_queue_priority = q->priority; } -static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd, +static struct kfd_mem_obj *allocate_mqd(struct kfd_node *kfd, struct queue_properties *q) { struct kfd_mem_obj *mqd_mem_obj; @@ -167,7 +166,7 @@ static int load_mqd(struct mqd_manager *mm, void *mqd, uint32_t pipe_id, return mm->dev->kfd2kgd->hqd_load(mm->dev->adev, mqd, pipe_id, queue_id, (uint32_t __user *)p->write_ptr, - wptr_shift, wptr_mask, mms); + wptr_shift, wptr_mask, mms, 0); } static void __update_mqd(struct mqd_manager *mm, void *mqd, @@ -390,7 +389,7 @@ static int debugfs_show_mqd_sdma(struct seq_file *m, void *data) struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type, - struct kfd_dev *dev) + struct kfd_node *dev) { struct mqd_manager *mqd; @@ -428,6 +427,7 @@ struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type, mqd->destroy_mqd = kfd_destroy_mqd_cp; mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct cik_mqd); + mqd->mqd_stride = kfd_mqd_stride; #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; #endif @@ -442,6 +442,7 @@ struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type, mqd->destroy_mqd = kfd_destroy_mqd_cp; mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct cik_mqd); + mqd->mqd_stride = kfd_mqd_stride; #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; #endif @@ -457,6 +458,7 @@ struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type, mqd->checkpoint_mqd = checkpoint_mqd_sdma; mqd->restore_mqd = restore_mqd_sdma; mqd->mqd_size = sizeof(struct cik_sdma_rlc_registers); + mqd->mqd_stride = kfd_mqd_stride; #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd_sdma; #endif @@ -470,7 +472,7 @@ struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type, } struct mqd_manager *mqd_manager_init_cik_hawaii(enum KFD_MQD_TYPE type, - struct kfd_dev *dev) + struct kfd_node *dev) { struct mqd_manager *mqd; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c index d3e2b6a599a40e1a3cc8b6ca5624eee3ee4e1aa4..94c0fc2e57b7f8e6afafe438e1d75ee4f05eae75 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c @@ -48,8 +48,7 @@ static void update_cu_mask(struct mqd_manager *mm, void *mqd, struct v10_compute_mqd *m; uint32_t se_mask[4] = {0}; /* 4 is the max # of SEs */ - if (!minfo || (minfo->update_flag != UPDATE_FLAG_CU_MASK) || - !minfo->cu_mask.ptr) + if (!minfo || !minfo->cu_mask.ptr) return; mqd_symmetrically_map_cu_mask(mm, @@ -74,7 +73,7 @@ static void set_priority(struct v10_compute_mqd *m, struct queue_properties *q) m->cp_hqd_queue_priority = q->priority; } -static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd, +static struct kfd_mem_obj *allocate_mqd(struct kfd_node *kfd, struct queue_properties *q) { struct kfd_mem_obj *mqd_mem_obj; @@ -117,12 +116,17 @@ static void init_mqd(struct mqd_manager *mm, void **mqd, 1 << CP_HQD_QUANTUM__QUANTUM_SCALE__SHIFT | 1 << CP_HQD_QUANTUM__QUANTUM_DURATION__SHIFT; + /* Set cp_hqd_hq_scheduler0 bit 14 to 1 to have the CP set up the + * DISPATCH_PTR. This is required for the kfd debugger + */ + m->cp_hqd_hq_scheduler0 = 1 << 14; + if (q->format == KFD_QUEUE_FORMAT_AQL) { m->cp_hqd_aql_control = 1 << CP_HQD_AQL_CONTROL__CONTROL0__SHIFT; } - if (mm->dev->cwsr_enabled) { + if (mm->dev->kfd->cwsr_enabled) { m->cp_hqd_persistent_state |= (1 << CP_HQD_PERSISTENT_STATE__QSWITCH_MODE__SHIFT); m->cp_hqd_ctx_save_base_addr_lo = @@ -151,7 +155,7 @@ static int load_mqd(struct mqd_manager *mm, void *mqd, r = mm->dev->kfd2kgd->hqd_load(mm->dev->adev, mqd, pipe_id, queue_id, (uint32_t __user *)p->write_ptr, - wptr_shift, 0, mms); + wptr_shift, 0, mms, 0); return r; } @@ -210,7 +214,7 @@ static void update_mqd(struct mqd_manager *mm, void *mqd, m->cp_hqd_pq_doorbell_control |= 1 << CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_BIF_DROP__SHIFT; } - if (mm->dev->cwsr_enabled) + if (mm->dev->kfd->cwsr_enabled) m->cp_hqd_ctx_save_control = 0; update_cu_mask(mm, mqd, minfo); @@ -227,11 +231,13 @@ static uint32_t read_doorbell_id(void *mqd) } static int get_wave_state(struct mqd_manager *mm, void *mqd, + struct queue_properties *q, void __user *ctl_stack, u32 *ctl_stack_used_size, u32 *save_area_used_size) { struct v10_compute_mqd *m; + struct kfd_context_save_area_header header; m = get_mqd(mqd); @@ -250,6 +256,15 @@ static int get_wave_state(struct mqd_manager *mm, void *mqd, * accessible to user mode */ + header.wave_state.control_stack_size = *ctl_stack_used_size; + header.wave_state.wave_state_size = *save_area_used_size; + + header.wave_state.wave_state_offset = m->cp_hqd_wg_state_offset; + header.wave_state.control_stack_offset = m->cp_hqd_cntl_stack_offset; + + if (copy_to_user(ctl_stack, &header, sizeof(header.wave_state))) + return -EFAULT; + return 0; } @@ -405,7 +420,7 @@ static int debugfs_show_mqd_sdma(struct seq_file *m, void *data) #endif struct mqd_manager *mqd_manager_init_v10(enum KFD_MQD_TYPE type, - struct kfd_dev *dev) + struct kfd_node *dev) { struct mqd_manager *mqd; @@ -432,6 +447,7 @@ struct mqd_manager *mqd_manager_init_v10(enum KFD_MQD_TYPE type, mqd->get_wave_state = get_wave_state; mqd->checkpoint_mqd = checkpoint_mqd; mqd->restore_mqd = restore_mqd; + mqd->mqd_stride = kfd_mqd_stride; #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; #endif @@ -447,6 +463,7 @@ struct mqd_manager *mqd_manager_init_v10(enum KFD_MQD_TYPE type, mqd->destroy_mqd = kfd_destroy_mqd_cp; mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct v10_compute_mqd); + mqd->mqd_stride = kfd_mqd_stride; #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; #endif @@ -478,6 +495,7 @@ struct mqd_manager *mqd_manager_init_v10(enum KFD_MQD_TYPE type, mqd->checkpoint_mqd = checkpoint_mqd_sdma; mqd->restore_mqd = restore_mqd_sdma; mqd->mqd_size = sizeof(struct v10_sdma_mqd); + mqd->mqd_stride = kfd_mqd_stride; #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd_sdma; #endif diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c index 5aa75f72caa1efc9aac1f393b6b69bdfdb995a01..31fec5e70d13523a36781a8f78b0d32df5b2b537 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c @@ -46,15 +46,33 @@ static void update_cu_mask(struct mqd_manager *mm, void *mqd, { struct v11_compute_mqd *m; uint32_t se_mask[KFD_MAX_NUM_SE] = {0}; + bool has_wa_flag = minfo && (minfo->update_flag & (UPDATE_FLAG_DBG_WA_ENABLE | + UPDATE_FLAG_DBG_WA_DISABLE)); - if (!minfo || (minfo->update_flag != UPDATE_FLAG_CU_MASK) || - !minfo->cu_mask.ptr) + if (!minfo || !(has_wa_flag || minfo->cu_mask.ptr)) return; + m = get_mqd(mqd); + + if (has_wa_flag) { + uint32_t wa_mask = minfo->update_flag == UPDATE_FLAG_DBG_WA_ENABLE ? + 0xffff : 0xffffffff; + + m->compute_static_thread_mgmt_se0 = wa_mask; + m->compute_static_thread_mgmt_se1 = wa_mask; + m->compute_static_thread_mgmt_se2 = wa_mask; + m->compute_static_thread_mgmt_se3 = wa_mask; + m->compute_static_thread_mgmt_se4 = wa_mask; + m->compute_static_thread_mgmt_se5 = wa_mask; + m->compute_static_thread_mgmt_se6 = wa_mask; + m->compute_static_thread_mgmt_se7 = wa_mask; + + return; + } + mqd_symmetrically_map_cu_mask(mm, minfo->cu_mask.ptr, minfo->cu_mask.count, se_mask); - m = get_mqd(mqd); m->compute_static_thread_mgmt_se0 = se_mask[0]; m->compute_static_thread_mgmt_se1 = se_mask[1]; m->compute_static_thread_mgmt_se2 = se_mask[2]; @@ -81,7 +99,7 @@ static void set_priority(struct v11_compute_mqd *m, struct queue_properties *q) m->cp_hqd_queue_priority = q->priority; } -static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd, +static struct kfd_mem_obj *allocate_mqd(struct kfd_node *node, struct queue_properties *q) { struct kfd_mem_obj *mqd_mem_obj; @@ -91,12 +109,12 @@ static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd, * MES write to areas beyond MQD size. So allocate * 1 PAGE_SIZE memory for MQD is MES is enabled. */ - if (kfd->shared_resources.enable_mes) + if (node->kfd->shared_resources.enable_mes) size = PAGE_SIZE; else size = sizeof(struct v11_compute_mqd); - if (kfd_gtt_sa_allocate(kfd, size, &mqd_mem_obj)) + if (kfd_gtt_sa_allocate(node, size, &mqd_mem_obj)) return NULL; return mqd_mem_obj; @@ -109,11 +127,12 @@ static void init_mqd(struct mqd_manager *mm, void **mqd, uint64_t addr; struct v11_compute_mqd *m; int size; + uint32_t wa_mask = q->is_dbg_wa ? 0xffff : 0xffffffff; m = (struct v11_compute_mqd *) mqd_mem_obj->cpu_ptr; addr = mqd_mem_obj->gpu_addr; - if (mm->dev->shared_resources.enable_mes) + if (mm->dev->kfd->shared_resources.enable_mes) size = PAGE_SIZE; else size = sizeof(struct v11_compute_mqd); @@ -122,14 +141,15 @@ static void init_mqd(struct mqd_manager *mm, void **mqd, m->header = 0xC0310800; m->compute_pipelinestat_enable = 1; - m->compute_static_thread_mgmt_se0 = 0xFFFFFFFF; - m->compute_static_thread_mgmt_se1 = 0xFFFFFFFF; - m->compute_static_thread_mgmt_se2 = 0xFFFFFFFF; - m->compute_static_thread_mgmt_se3 = 0xFFFFFFFF; - m->compute_static_thread_mgmt_se4 = 0xFFFFFFFF; - m->compute_static_thread_mgmt_se5 = 0xFFFFFFFF; - m->compute_static_thread_mgmt_se6 = 0xFFFFFFFF; - m->compute_static_thread_mgmt_se7 = 0xFFFFFFFF; + + m->compute_static_thread_mgmt_se0 = wa_mask; + m->compute_static_thread_mgmt_se1 = wa_mask; + m->compute_static_thread_mgmt_se2 = wa_mask; + m->compute_static_thread_mgmt_se3 = wa_mask; + m->compute_static_thread_mgmt_se4 = wa_mask; + m->compute_static_thread_mgmt_se5 = wa_mask; + m->compute_static_thread_mgmt_se6 = wa_mask; + m->compute_static_thread_mgmt_se7 = wa_mask; m->cp_hqd_persistent_state = CP_HQD_PERSISTENT_STATE__PRELOAD_REQ_MASK | 0x55 << CP_HQD_PERSISTENT_STATE__PRELOAD_SIZE__SHIFT; @@ -143,6 +163,11 @@ static void init_mqd(struct mqd_manager *mm, void **mqd, 1 << CP_HQD_QUANTUM__QUANTUM_SCALE__SHIFT | 1 << CP_HQD_QUANTUM__QUANTUM_DURATION__SHIFT; + /* Set cp_hqd_hq_scheduler0 bit 14 to 1 to have the CP set up the + * DISPATCH_PTR. This is required for the kfd debugger + */ + m->cp_hqd_hq_status0 = 1 << 14; + /* * GFX11 RS64 CPFW version >= 509 supports PCIe atomics support * acknowledgment. @@ -155,7 +180,7 @@ static void init_mqd(struct mqd_manager *mm, void **mqd, 1 << CP_HQD_AQL_CONTROL__CONTROL0__SHIFT; } - if (mm->dev->cwsr_enabled) { + if (mm->dev->kfd->cwsr_enabled) { m->cp_hqd_persistent_state |= (1 << CP_HQD_PERSISTENT_STATE__QSWITCH_MODE__SHIFT); m->cp_hqd_ctx_save_base_addr_lo = @@ -184,7 +209,7 @@ static int load_mqd(struct mqd_manager *mm, void *mqd, r = mm->dev->kfd2kgd->hqd_load(mm->dev->adev, mqd, pipe_id, queue_id, (uint32_t __user *)p->write_ptr, - wptr_shift, 0, mms); + wptr_shift, 0, mms, 0); return r; } @@ -243,7 +268,7 @@ static void update_mqd(struct mqd_manager *mm, void *mqd, m->cp_hqd_pq_doorbell_control |= 1 << CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_BIF_DROP__SHIFT; } - if (mm->dev->cwsr_enabled) + if (mm->dev->kfd->cwsr_enabled) m->cp_hqd_ctx_save_control = 0; update_cu_mask(mm, mqd, minfo); @@ -260,12 +285,13 @@ static uint32_t read_doorbell_id(void *mqd) } static int get_wave_state(struct mqd_manager *mm, void *mqd, + struct queue_properties *q, void __user *ctl_stack, u32 *ctl_stack_used_size, u32 *save_area_used_size) { struct v11_compute_mqd *m; - /*struct mqd_user_context_save_area_header header;*/ + struct kfd_context_save_area_header header; m = get_mqd(mqd); @@ -283,16 +309,15 @@ static int get_wave_state(struct mqd_manager *mm, void *mqd, * it's part of the context save area that is already * accessible to user mode */ -/* - header.control_stack_size = *ctl_stack_used_size; - header.wave_state_size = *save_area_used_size; + header.wave_state.control_stack_size = *ctl_stack_used_size; + header.wave_state.wave_state_size = *save_area_used_size; - header.wave_state_offset = m->cp_hqd_wg_state_offset; - header.control_stack_offset = m->cp_hqd_cntl_stack_offset; + header.wave_state.wave_state_offset = m->cp_hqd_wg_state_offset; + header.wave_state.control_stack_offset = m->cp_hqd_cntl_stack_offset; - if (copy_to_user(ctl_stack, &header, sizeof(header))) + if (copy_to_user(ctl_stack, &header, sizeof(header.wave_state))) return -EFAULT; -*/ + return 0; } @@ -319,7 +344,7 @@ static void init_mqd_sdma(struct mqd_manager *mm, void **mqd, m = (struct v11_sdma_mqd *) mqd_mem_obj->cpu_ptr; - if (mm->dev->shared_resources.enable_mes) + if (mm->dev->kfd->shared_resources.enable_mes) size = PAGE_SIZE; else size = sizeof(struct v11_sdma_mqd); @@ -387,7 +412,7 @@ static int debugfs_show_mqd_sdma(struct seq_file *m, void *data) #endif struct mqd_manager *mqd_manager_init_v11(enum KFD_MQD_TYPE type, - struct kfd_dev *dev) + struct kfd_node *dev) { struct mqd_manager *mqd; @@ -463,7 +488,7 @@ struct mqd_manager *mqd_manager_init_v11(enum KFD_MQD_TYPE type, * To allocate SDMA MQDs by generic functions * when MES is enabled. */ - if (dev->shared_resources.enable_mes) { + if (dev->kfd->shared_resources.enable_mes) { mqd->allocate_mqd = allocate_mqd; mqd->free_mqd = kfd_free_mqd_cp; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c index fdbfd725841ffe60889f620c36b2f8c63fff10ab..601bb9f68048ccfc13f7cc5dcab641e7ca727e73 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c @@ -32,6 +32,22 @@ #include "gc/gc_9_0_sh_mask.h" #include "sdma0/sdma0_4_0_sh_mask.h" #include "amdgpu_amdkfd.h" +#include "kfd_device_queue_manager.h" + +static void update_mqd(struct mqd_manager *mm, void *mqd, + struct queue_properties *q, + struct mqd_update_info *minfo); + +static uint64_t mqd_stride_v9(struct mqd_manager *mm, + struct queue_properties *q) +{ + if (mm->dev->kfd->cwsr_enabled && + q->type == KFD_QUEUE_TYPE_COMPUTE) + return ALIGN(q->ctl_stack_size, PAGE_SIZE) + + ALIGN(sizeof(struct v9_mqd), PAGE_SIZE); + + return mm->mqd_size; +} static inline struct v9_mqd *get_mqd(void *mqd) { @@ -49,8 +65,7 @@ static void update_cu_mask(struct mqd_manager *mm, void *mqd, struct v9_mqd *m; uint32_t se_mask[KFD_MAX_NUM_SE] = {0}; - if (!minfo || (minfo->update_flag != UPDATE_FLAG_CU_MASK) || - !minfo->cu_mask.ptr) + if (!minfo || !minfo->cu_mask.ptr) return; mqd_symmetrically_map_cu_mask(mm, @@ -83,7 +98,7 @@ static void set_priority(struct v9_mqd *m, struct queue_properties *q) m->cp_hqd_queue_priority = q->priority; } -static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd, +static struct kfd_mem_obj *allocate_mqd(struct kfd_node *node, struct queue_properties *q) { int retval; @@ -105,28 +120,30 @@ static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd, * pass a special bo flag AMDGPU_GEM_CREATE_CP_MQD_GFX9 to instruct * amdgpu memory functions to do so. */ - if (kfd->cwsr_enabled && (q->type == KFD_QUEUE_TYPE_COMPUTE)) { + if (node->kfd->cwsr_enabled && (q->type == KFD_QUEUE_TYPE_COMPUTE)) { mqd_mem_obj = kzalloc(sizeof(struct kfd_mem_obj), GFP_KERNEL); if (!mqd_mem_obj) return NULL; - retval = amdgpu_amdkfd_alloc_gtt_mem(kfd->adev, - ALIGN(q->ctl_stack_size, PAGE_SIZE) + - ALIGN(sizeof(struct v9_mqd), PAGE_SIZE), + retval = amdgpu_amdkfd_alloc_gtt_mem(node->adev, + (ALIGN(q->ctl_stack_size, PAGE_SIZE) + + ALIGN(sizeof(struct v9_mqd), PAGE_SIZE)) * + NUM_XCC(node->xcc_mask), &(mqd_mem_obj->gtt_mem), &(mqd_mem_obj->gpu_addr), (void *)&(mqd_mem_obj->cpu_ptr), true); + + if (retval) { + kfree(mqd_mem_obj); + return NULL; + } } else { - retval = kfd_gtt_sa_allocate(kfd, sizeof(struct v9_mqd), + retval = kfd_gtt_sa_allocate(node, sizeof(struct v9_mqd), &mqd_mem_obj); - } - - if (retval) { - kfree(mqd_mem_obj); - return NULL; + if (retval) + return NULL; } return mqd_mem_obj; - } static void init_mqd(struct mqd_manager *mm, void **mqd, @@ -135,7 +152,6 @@ static void init_mqd(struct mqd_manager *mm, void **mqd, { uint64_t addr; struct v9_mqd *m; - struct amdgpu_device *adev = (struct amdgpu_device *)mm->dev->adev; m = (struct v9_mqd *) mqd_mem_obj->cpu_ptr; addr = mqd_mem_obj->gpu_addr; @@ -165,31 +181,21 @@ static void init_mqd(struct mqd_manager *mm, void **mqd, 1 << CP_HQD_QUANTUM__QUANTUM_SCALE__SHIFT | 1 << CP_HQD_QUANTUM__QUANTUM_DURATION__SHIFT; - if (q->format == KFD_QUEUE_FORMAT_AQL) { + /* Set cp_hqd_hq_scheduler0 bit 14 to 1 to have the CP set up the + * DISPATCH_PTR. This is required for the kfd debugger + */ + m->cp_hqd_hq_status0 = 1 << 14; + + if (q->format == KFD_QUEUE_FORMAT_AQL) m->cp_hqd_aql_control = 1 << CP_HQD_AQL_CONTROL__CONTROL0__SHIFT; - if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) { - /* On GC 9.4.3, DW 41 is re-purposed as - * compute_tg_chunk_size. - * TODO: review this setting when active CUs in the - * partition play a role - */ - m->compute_static_thread_mgmt_se6 = 1; - } - } else { - /* PM4 queue */ - if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) { - m->compute_static_thread_mgmt_se6 = 0; - /* TODO: program pm4_target_xcc */ - } - } if (q->tba_addr) { m->compute_pgm_rsrc2 |= (1 << COMPUTE_PGM_RSRC2__TRAP_PRESENT__SHIFT); } - if (mm->dev->cwsr_enabled && q->ctx_save_restore_area_address) { + if (mm->dev->kfd->cwsr_enabled && q->ctx_save_restore_area_address) { m->cp_hqd_persistent_state |= (1 << CP_HQD_PERSISTENT_STATE__QSWITCH_MODE__SHIFT); m->cp_hqd_ctx_save_base_addr_lo = @@ -205,7 +211,7 @@ static void init_mqd(struct mqd_manager *mm, void **mqd, *mqd = m; if (gart_addr) *gart_addr = addr; - mm->update_mqd(mm, m, q, NULL); + update_mqd(mm, m, q, NULL); } static int load_mqd(struct mqd_manager *mm, void *mqd, @@ -217,14 +223,13 @@ static int load_mqd(struct mqd_manager *mm, void *mqd, return mm->dev->kfd2kgd->hqd_load(mm->dev->adev, mqd, pipe_id, queue_id, (uint32_t __user *)p->write_ptr, - wptr_shift, 0, mms); + wptr_shift, 0, mms, 0); } static void update_mqd(struct mqd_manager *mm, void *mqd, struct queue_properties *q, struct mqd_update_info *minfo) { - struct amdgpu_device *adev = (struct amdgpu_device *)mm->dev->adev; struct v9_mqd *m; m = get_mqd(mqd); @@ -257,9 +262,14 @@ static void update_mqd(struct mqd_manager *mm, void *mqd, * Limit is 0xFF EOP entries (= 0x7F8 dwords). CP will not submit * more than (EOP entry count - 1) so a queue size of 0x800 dwords * is safe, giving a maximum field value of 0xA. + * + * Also, do calculation only if EOP is used (size > 0), otherwise + * the order_base_2 calculation provides incorrect result. + * */ - m->cp_hqd_eop_control = min(0xA, - order_base_2(q->eop_ring_buffer_size / 4) - 1); + m->cp_hqd_eop_control = q->eop_ring_buffer_size ? + min(0xA, order_base_2(q->eop_ring_buffer_size / 4) - 1) : 0; + m->cp_hqd_eop_base_addr_lo = lower_32_bits(q->eop_ring_buffer_address >> 8); m->cp_hqd_eop_base_addr_hi = @@ -270,17 +280,14 @@ static void update_mqd(struct mqd_manager *mm, void *mqd, m->cp_hqd_vmid = q->vmid; if (q->format == KFD_QUEUE_FORMAT_AQL) { - m->cp_hqd_pq_control |= + m->cp_hqd_pq_control |= CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR_MASK | 2 << CP_HQD_PQ_CONTROL__SLOT_BASED_WPTR__SHIFT | 1 << CP_HQD_PQ_CONTROL__QUEUE_FULL_EN__SHIFT | 1 << CP_HQD_PQ_CONTROL__WPP_CLAMP_EN__SHIFT; - if (adev->ip_versions[GC_HWIP][0] != IP_VERSION(9, 4, 3)) - m->cp_hqd_pq_control |= - CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR_MASK; m->cp_hqd_pq_doorbell_control |= 1 << CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_BIF_DROP__SHIFT; } - if (mm->dev->cwsr_enabled && q->ctx_save_restore_area_address) + if (mm->dev->kfd->cwsr_enabled && q->ctx_save_restore_area_address) m->cp_hqd_ctx_save_control = 0; update_cu_mask(mm, mqd, minfo); @@ -298,11 +305,13 @@ static uint32_t read_doorbell_id(void *mqd) } static int get_wave_state(struct mqd_manager *mm, void *mqd, + struct queue_properties *q, void __user *ctl_stack, u32 *ctl_stack_used_size, u32 *save_area_used_size) { struct v9_mqd *m; + struct kfd_context_save_area_header header; /* Control stack is located one page after MQD. */ void *mqd_ctl_stack = (void *)((uintptr_t)mqd + PAGE_SIZE); @@ -314,7 +323,18 @@ static int get_wave_state(struct mqd_manager *mm, void *mqd, *save_area_used_size = m->cp_hqd_wg_state_offset - m->cp_hqd_cntl_stack_size; - if (copy_to_user(ctl_stack, mqd_ctl_stack, m->cp_hqd_cntl_stack_size)) + header.wave_state.control_stack_size = *ctl_stack_used_size; + header.wave_state.wave_state_size = *save_area_used_size; + + header.wave_state.wave_state_offset = m->cp_hqd_wg_state_offset; + header.wave_state.control_stack_offset = m->cp_hqd_cntl_stack_offset; + + if (copy_to_user(ctl_stack, &header, sizeof(header.wave_state))) + return -EFAULT; + + if (copy_to_user(ctl_stack + m->cp_hqd_cntl_stack_offset, + mqd_ctl_stack + m->cp_hqd_cntl_stack_offset, + *ctl_stack_used_size)) return -EFAULT; return 0; @@ -467,6 +487,288 @@ static void restore_mqd_sdma(struct mqd_manager *mm, void **mqd, qp->is_active = 0; } +static void init_mqd_hiq_v9_4_3(struct mqd_manager *mm, void **mqd, + struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, + struct queue_properties *q) +{ + struct v9_mqd *m; + int xcc = 0; + struct kfd_mem_obj xcc_mqd_mem_obj; + uint64_t xcc_gart_addr = 0; + + memset(&xcc_mqd_mem_obj, 0x0, sizeof(struct kfd_mem_obj)); + + for (xcc = 0; xcc < NUM_XCC(mm->dev->xcc_mask); xcc++) { + kfd_get_hiq_xcc_mqd(mm->dev, &xcc_mqd_mem_obj, xcc); + + init_mqd(mm, (void **)&m, &xcc_mqd_mem_obj, &xcc_gart_addr, q); + + m->cp_hqd_pq_control |= CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR_MASK | + 1 << CP_HQD_PQ_CONTROL__PRIV_STATE__SHIFT | + 1 << CP_HQD_PQ_CONTROL__KMD_QUEUE__SHIFT; + m->cp_mqd_stride_size = kfd_hiq_mqd_stride(mm->dev); + if (xcc == 0) { + /* Set no_update_rptr = 0 in Master XCC */ + m->cp_hqd_pq_control &= ~CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR_MASK; + + /* Set the MQD pointer and gart address to XCC0 MQD */ + *mqd = m; + *gart_addr = xcc_gart_addr; + } + } +} + +static int hiq_load_mqd_kiq_v9_4_3(struct mqd_manager *mm, void *mqd, + uint32_t pipe_id, uint32_t queue_id, + struct queue_properties *p, struct mm_struct *mms) +{ + uint32_t xcc_mask = mm->dev->xcc_mask; + int xcc_id, err, inst = 0; + void *xcc_mqd; + uint64_t hiq_mqd_size = kfd_hiq_mqd_stride(mm->dev); + + for_each_inst(xcc_id, xcc_mask) { + xcc_mqd = mqd + hiq_mqd_size * inst; + err = mm->dev->kfd2kgd->hiq_mqd_load(mm->dev->adev, xcc_mqd, + pipe_id, queue_id, + p->doorbell_off, xcc_id); + if (err) { + pr_debug("Failed to load HIQ MQD for XCC: %d\n", inst); + break; + } + ++inst; + } + + return err; +} + +static int destroy_hiq_mqd_v9_4_3(struct mqd_manager *mm, void *mqd, + enum kfd_preempt_type type, unsigned int timeout, + uint32_t pipe_id, uint32_t queue_id) +{ + uint32_t xcc_mask = mm->dev->xcc_mask; + int xcc_id, err, inst = 0; + void *xcc_mqd; + uint64_t hiq_mqd_size = kfd_hiq_mqd_stride(mm->dev); + + for_each_inst(xcc_id, xcc_mask) { + xcc_mqd = mqd + hiq_mqd_size * inst; + err = mm->dev->kfd2kgd->hqd_destroy(mm->dev->adev, xcc_mqd, + type, timeout, pipe_id, + queue_id, xcc_id); + if (err) { + pr_debug("Destroy MQD failed for xcc: %d\n", inst); + break; + } + ++inst; + } + + return err; +} + +static void get_xcc_mqd(struct kfd_mem_obj *mqd_mem_obj, + struct kfd_mem_obj *xcc_mqd_mem_obj, + uint64_t offset) +{ + xcc_mqd_mem_obj->gtt_mem = (offset == 0) ? + mqd_mem_obj->gtt_mem : NULL; + xcc_mqd_mem_obj->gpu_addr = mqd_mem_obj->gpu_addr + offset; + xcc_mqd_mem_obj->cpu_ptr = (uint32_t *)((uintptr_t)mqd_mem_obj->cpu_ptr + + offset); +} + +static void init_mqd_v9_4_3(struct mqd_manager *mm, void **mqd, + struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, + struct queue_properties *q) +{ + struct v9_mqd *m; + int xcc = 0; + struct kfd_mem_obj xcc_mqd_mem_obj; + uint64_t xcc_gart_addr = 0; + uint64_t xcc_ctx_save_restore_area_address; + uint64_t offset = mm->mqd_stride(mm, q); + uint32_t local_xcc_start = mm->dev->dqm->current_logical_xcc_start++; + + memset(&xcc_mqd_mem_obj, 0x0, sizeof(struct kfd_mem_obj)); + for (xcc = 0; xcc < NUM_XCC(mm->dev->xcc_mask); xcc++) { + get_xcc_mqd(mqd_mem_obj, &xcc_mqd_mem_obj, offset*xcc); + + init_mqd(mm, (void **)&m, &xcc_mqd_mem_obj, &xcc_gart_addr, q); + + m->cp_mqd_stride_size = offset; + + /* + * Update the CWSR address for each XCC if CWSR is enabled + * and CWSR area is allocated in thunk + */ + if (mm->dev->kfd->cwsr_enabled && + q->ctx_save_restore_area_address) { + xcc_ctx_save_restore_area_address = + q->ctx_save_restore_area_address + + (xcc * q->ctx_save_restore_area_size); + + m->cp_hqd_ctx_save_base_addr_lo = + lower_32_bits(xcc_ctx_save_restore_area_address); + m->cp_hqd_ctx_save_base_addr_hi = + upper_32_bits(xcc_ctx_save_restore_area_address); + } + + if (q->format == KFD_QUEUE_FORMAT_AQL) { + m->compute_tg_chunk_size = 1; + m->compute_current_logic_xcc_id = + (local_xcc_start + xcc) % + NUM_XCC(mm->dev->xcc_mask); + + switch (xcc) { + case 0: + /* Master XCC */ + m->cp_hqd_pq_control &= + ~CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR_MASK; + break; + default: + break; + } + } else { + /* PM4 Queue */ + m->compute_current_logic_xcc_id = 0; + m->compute_tg_chunk_size = 0; + m->pm4_target_xcc_in_xcp = q->pm4_target_xcc; + } + + if (xcc == 0) { + /* Set the MQD pointer and gart address to XCC0 MQD */ + *mqd = m; + *gart_addr = xcc_gart_addr; + } + } +} + +static void update_mqd_v9_4_3(struct mqd_manager *mm, void *mqd, + struct queue_properties *q, struct mqd_update_info *minfo) +{ + struct v9_mqd *m; + int xcc = 0; + uint64_t size = mm->mqd_stride(mm, q); + + for (xcc = 0; xcc < NUM_XCC(mm->dev->xcc_mask); xcc++) { + m = get_mqd(mqd + size * xcc); + update_mqd(mm, m, q, minfo); + + if (q->format == KFD_QUEUE_FORMAT_AQL) { + switch (xcc) { + case 0: + /* Master XCC */ + m->cp_hqd_pq_control &= + ~CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR_MASK; + break; + default: + break; + } + m->compute_tg_chunk_size = 1; + } else { + /* PM4 Queue */ + m->compute_current_logic_xcc_id = 0; + m->compute_tg_chunk_size = 0; + m->pm4_target_xcc_in_xcp = q->pm4_target_xcc; + } + } +} + +static int destroy_mqd_v9_4_3(struct mqd_manager *mm, void *mqd, + enum kfd_preempt_type type, unsigned int timeout, + uint32_t pipe_id, uint32_t queue_id) +{ + uint32_t xcc_mask = mm->dev->xcc_mask; + int xcc_id, err, inst = 0; + void *xcc_mqd; + struct v9_mqd *m; + uint64_t mqd_offset; + + m = get_mqd(mqd); + mqd_offset = m->cp_mqd_stride_size; + + for_each_inst(xcc_id, xcc_mask) { + xcc_mqd = mqd + mqd_offset * inst; + err = mm->dev->kfd2kgd->hqd_destroy(mm->dev->adev, xcc_mqd, + type, timeout, pipe_id, + queue_id, xcc_id); + if (err) { + pr_debug("Destroy MQD failed for xcc: %d\n", inst); + break; + } + ++inst; + } + + return err; +} + +static int load_mqd_v9_4_3(struct mqd_manager *mm, void *mqd, + uint32_t pipe_id, uint32_t queue_id, + struct queue_properties *p, struct mm_struct *mms) +{ + /* AQL write pointer counts in 64B packets, PM4/CP counts in dwords. */ + uint32_t wptr_shift = (p->format == KFD_QUEUE_FORMAT_AQL ? 4 : 0); + uint32_t xcc_mask = mm->dev->xcc_mask; + int xcc_id, err, inst = 0; + void *xcc_mqd; + uint64_t mqd_stride_size = mm->mqd_stride(mm, p); + + for_each_inst(xcc_id, xcc_mask) { + xcc_mqd = mqd + mqd_stride_size * inst; + err = mm->dev->kfd2kgd->hqd_load( + mm->dev->adev, xcc_mqd, pipe_id, queue_id, + (uint32_t __user *)p->write_ptr, wptr_shift, 0, mms, + xcc_id); + if (err) { + pr_debug("Load MQD failed for xcc: %d\n", inst); + break; + } + ++inst; + } + + return err; +} + +static int get_wave_state_v9_4_3(struct mqd_manager *mm, void *mqd, + struct queue_properties *q, + void __user *ctl_stack, + u32 *ctl_stack_used_size, + u32 *save_area_used_size) +{ + int xcc, err = 0; + void *xcc_mqd; + void __user *xcc_ctl_stack; + uint64_t mqd_stride_size = mm->mqd_stride(mm, q); + u32 tmp_ctl_stack_used_size = 0, tmp_save_area_used_size = 0; + + for (xcc = 0; xcc < NUM_XCC(mm->dev->xcc_mask); xcc++) { + xcc_mqd = mqd + mqd_stride_size * xcc; + xcc_ctl_stack = (void __user *)((uintptr_t)ctl_stack + + q->ctx_save_restore_area_size * xcc); + + err = get_wave_state(mm, xcc_mqd, q, xcc_ctl_stack, + &tmp_ctl_stack_used_size, + &tmp_save_area_used_size); + if (err) + break; + + /* + * Set the ctl_stack_used_size and save_area_used_size to + * ctl_stack_used_size and save_area_used_size of XCC 0 when + * passing the info the user-space. + * For multi XCC, user-space would have to look at the header + * info of each Control stack area to determine the control + * stack size and save area used. + */ + if (xcc == 0) { + *ctl_stack_used_size = tmp_ctl_stack_used_size; + *save_area_used_size = tmp_save_area_used_size; + } + } + + return err; +} + #if defined(CONFIG_DEBUG_FS) static int debugfs_show_mqd(struct seq_file *m, void *data) @@ -486,7 +788,7 @@ static int debugfs_show_mqd_sdma(struct seq_file *m, void *data) #endif struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type, - struct kfd_dev *dev) + struct kfd_node *dev) { struct mqd_manager *mqd; @@ -502,34 +804,50 @@ struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type, switch (type) { case KFD_MQD_TYPE_CP: mqd->allocate_mqd = allocate_mqd; - mqd->init_mqd = init_mqd; mqd->free_mqd = kfd_free_mqd_cp; - mqd->load_mqd = load_mqd; - mqd->update_mqd = update_mqd; - mqd->destroy_mqd = kfd_destroy_mqd_cp; mqd->is_occupied = kfd_is_occupied_cp; - mqd->get_wave_state = get_wave_state; mqd->get_checkpoint_info = get_checkpoint_info; mqd->checkpoint_mqd = checkpoint_mqd; mqd->restore_mqd = restore_mqd; mqd->mqd_size = sizeof(struct v9_mqd); + mqd->mqd_stride = mqd_stride_v9; #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; #endif + if (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 3)) { + mqd->init_mqd = init_mqd_v9_4_3; + mqd->load_mqd = load_mqd_v9_4_3; + mqd->update_mqd = update_mqd_v9_4_3; + mqd->destroy_mqd = destroy_mqd_v9_4_3; + mqd->get_wave_state = get_wave_state_v9_4_3; + } else { + mqd->init_mqd = init_mqd; + mqd->load_mqd = load_mqd; + mqd->update_mqd = update_mqd; + mqd->destroy_mqd = kfd_destroy_mqd_cp; + mqd->get_wave_state = get_wave_state; + } break; case KFD_MQD_TYPE_HIQ: mqd->allocate_mqd = allocate_hiq_mqd; - mqd->init_mqd = init_mqd_hiq; mqd->free_mqd = free_mqd_hiq_sdma; - mqd->load_mqd = kfd_hiq_load_mqd_kiq; mqd->update_mqd = update_mqd; - mqd->destroy_mqd = kfd_destroy_mqd_cp; mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct v9_mqd); + mqd->mqd_stride = kfd_mqd_stride; #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; #endif mqd->read_doorbell_id = read_doorbell_id; + if (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 3)) { + mqd->init_mqd = init_mqd_hiq_v9_4_3; + mqd->load_mqd = hiq_load_mqd_kiq_v9_4_3; + mqd->destroy_mqd = destroy_hiq_mqd_v9_4_3; + } else { + mqd->init_mqd = init_mqd_hiq; + mqd->load_mqd = kfd_hiq_load_mqd_kiq; + mqd->destroy_mqd = kfd_destroy_mqd_cp; + } break; case KFD_MQD_TYPE_DIQ: mqd->allocate_mqd = allocate_mqd; @@ -555,6 +873,7 @@ struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type, mqd->checkpoint_mqd = checkpoint_mqd_sdma; mqd->restore_mqd = restore_mqd_sdma; mqd->mqd_size = sizeof(struct v9_sdma_mqd); + mqd->mqd_stride = kfd_mqd_stride; #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd_sdma; #endif diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c index 530ba6f5b57eaf289adf432caa59e84263dcf6d4..d1e962da51dd3196155faa2cd6a84bf2feb11684 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c @@ -51,8 +51,7 @@ static void update_cu_mask(struct mqd_manager *mm, void *mqd, struct vi_mqd *m; uint32_t se_mask[4] = {0}; /* 4 is the max # of SEs */ - if (!minfo || (minfo->update_flag != UPDATE_FLAG_CU_MASK) || - !minfo->cu_mask.ptr) + if (!minfo || !minfo->cu_mask.ptr) return; mqd_symmetrically_map_cu_mask(mm, @@ -77,7 +76,7 @@ static void set_priority(struct vi_mqd *m, struct queue_properties *q) m->cp_hqd_queue_priority = q->priority; } -static struct kfd_mem_obj *allocate_mqd(struct kfd_dev *kfd, +static struct kfd_mem_obj *allocate_mqd(struct kfd_node *kfd, struct queue_properties *q) { struct kfd_mem_obj *mqd_mem_obj; @@ -136,7 +135,7 @@ static void init_mqd(struct mqd_manager *mm, void **mqd, (1 << COMPUTE_PGM_RSRC2__TRAP_PRESENT__SHIFT); } - if (mm->dev->cwsr_enabled && q->ctx_save_restore_area_address) { + if (mm->dev->kfd->cwsr_enabled && q->ctx_save_restore_area_address) { m->cp_hqd_persistent_state |= (1 << CP_HQD_PERSISTENT_STATE__QSWITCH_MODE__SHIFT); m->cp_hqd_ctx_save_base_addr_lo = @@ -165,7 +164,7 @@ static int load_mqd(struct mqd_manager *mm, void *mqd, return mm->dev->kfd2kgd->hqd_load(mm->dev->adev, mqd, pipe_id, queue_id, (uint32_t __user *)p->write_ptr, - wptr_shift, wptr_mask, mms); + wptr_shift, wptr_mask, mms, 0); } static void __update_mqd(struct mqd_manager *mm, void *mqd, @@ -227,7 +226,7 @@ static void __update_mqd(struct mqd_manager *mm, void *mqd, 2 << CP_HQD_PQ_CONTROL__SLOT_BASED_WPTR__SHIFT; } - if (mm->dev->cwsr_enabled && q->ctx_save_restore_area_address) + if (mm->dev->kfd->cwsr_enabled && q->ctx_save_restore_area_address) m->cp_hqd_ctx_save_control = atc_bit << CP_HQD_CTX_SAVE_CONTROL__ATC__SHIFT | mtype << CP_HQD_CTX_SAVE_CONTROL__MTYPE__SHIFT; @@ -261,6 +260,7 @@ static void update_mqd_tonga(struct mqd_manager *mm, void *mqd, } static int get_wave_state(struct mqd_manager *mm, void *mqd, + struct queue_properties *q, void __user *ctl_stack, u32 *ctl_stack_used_size, u32 *save_area_used_size) @@ -446,7 +446,7 @@ static int debugfs_show_mqd_sdma(struct seq_file *m, void *data) #endif struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type, - struct kfd_dev *dev) + struct kfd_node *dev) { struct mqd_manager *mqd; @@ -486,6 +486,7 @@ struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type, mqd->destroy_mqd = kfd_destroy_mqd_cp; mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct vi_mqd); + mqd->mqd_stride = kfd_mqd_stride; #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; #endif @@ -500,6 +501,7 @@ struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type, mqd->destroy_mqd = kfd_destroy_mqd_cp; mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct vi_mqd); + mqd->mqd_stride = kfd_mqd_stride; #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; #endif @@ -515,6 +517,7 @@ struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type, mqd->checkpoint_mqd = checkpoint_mqd_sdma; mqd->restore_mqd = restore_mqd_sdma; mqd->mqd_size = sizeof(struct vi_sdma_mqd); + mqd->mqd_stride = kfd_mqd_stride; #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd_sdma; #endif @@ -528,7 +531,7 @@ struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type, } struct mqd_manager *mqd_manager_init_vi_tonga(enum KFD_MQD_TYPE type, - struct kfd_dev *dev) + struct kfd_node *dev) { struct mqd_manager *mqd; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c index f612325241aa4fe5d303cc07ec6c5db5bd4a032f..401096c103b2f1e9d51d0cea56089eb94195eb5c 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c @@ -45,7 +45,7 @@ static void pm_calc_rlib_size(struct packet_manager *pm, unsigned int process_count, queue_count, compute_queue_count, gws_queue_count; unsigned int map_queue_size; unsigned int max_proc_per_quantum = 1; - struct kfd_dev *dev = pm->dqm->dev; + struct kfd_node *dev = pm->dqm->dev; process_count = pm->dqm->processes_count; queue_count = pm->dqm->active_queue_count; @@ -370,6 +370,38 @@ int pm_send_query_status(struct packet_manager *pm, uint64_t fence_address, return retval; } +int pm_update_grace_period(struct packet_manager *pm, uint32_t grace_period) +{ + int retval = 0; + uint32_t *buffer, size; + + size = pm->pmf->set_grace_period_size; + + mutex_lock(&pm->lock); + + if (size) { + kq_acquire_packet_buffer(pm->priv_queue, + size / sizeof(uint32_t), + (unsigned int **)&buffer); + + if (!buffer) { + pr_err("Failed to allocate buffer on kernel queue\n"); + retval = -ENOMEM; + goto out; + } + + retval = pm->pmf->set_grace_period(pm, buffer, grace_period); + if (!retval) + kq_submit_packet(pm->priv_queue); + else + kq_rollback_packet(pm->priv_queue); + } + +out: + mutex_unlock(&pm->lock); + return retval; +} + int pm_send_unmap_queue(struct packet_manager *pm, enum kfd_unmap_queues_filter filter, uint32_t filter_param, bool reset) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c index 18250845a9890a155be4c15917b713af919b8f6e..29a2d0499b674f293127830f69a238be599061ba 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c @@ -34,6 +34,9 @@ static int pm_map_process_v9(struct packet_manager *pm, { struct pm4_mes_map_process *packet; uint64_t vm_page_table_base_addr = qpd->page_table_base; + struct kfd_node *kfd = pm->dqm->dev; + struct kfd_process_device *pdd = + container_of(qpd, struct kfd_process_device, qpd); packet = (struct pm4_mes_map_process *)buffer; memset(buffer, 0, sizeof(struct pm4_mes_map_process)); @@ -49,6 +52,12 @@ static int pm_map_process_v9(struct packet_manager *pm, packet->bitfields14.sdma_enable = 1; packet->bitfields14.num_queues = (qpd->is_debug) ? 0 : qpd->queue_count; + if (kfd->dqm->trap_debug_vmid && pdd->process->debug_trap_enabled && + pdd->process->runtime_info.runtime_state == DEBUG_RUNTIME_STATE_ENABLED) { + packet->bitfields2.debug_vmid = kfd->dqm->trap_debug_vmid; + packet->bitfields2.new_debug = 1; + } + packet->sh_mem_config = qpd->sh_mem_config; packet->sh_mem_bases = qpd->sh_mem_bases; if (qpd->tba_addr) { @@ -79,6 +88,10 @@ static int pm_map_process_aldebaran(struct packet_manager *pm, { struct pm4_mes_map_process_aldebaran *packet; uint64_t vm_page_table_base_addr = qpd->page_table_base; + struct kfd_dev *kfd = pm->dqm->dev->kfd; + struct kfd_process_device *pdd = + container_of(qpd, struct kfd_process_device, qpd); + int i; packet = (struct pm4_mes_map_process_aldebaran *)buffer; memset(buffer, 0, sizeof(struct pm4_mes_map_process_aldebaran)); @@ -93,6 +106,16 @@ static int pm_map_process_aldebaran(struct packet_manager *pm, packet->bitfields14.num_oac = qpd->num_oac; packet->bitfields14.sdma_enable = 1; packet->bitfields14.num_queues = (qpd->is_debug) ? 0 : qpd->queue_count; + packet->spi_gdbg_per_vmid_cntl = pdd->spi_dbg_override | + pdd->spi_dbg_launch_mode; + + if (pdd->process->debug_trap_enabled) { + for (i = 0; i < kfd->device_info.num_of_watch_points; i++) + packet->tcp_watch_cntl[i] = pdd->watch_points[i]; + + packet->bitfields2.single_memops = + !!(pdd->process->dbg_flags & KFD_DBG_TRAP_FLAG_SINGLE_MEM_OP); + } packet->sh_mem_config = qpd->sh_mem_config; packet->sh_mem_bases = qpd->sh_mem_bases; @@ -119,7 +142,7 @@ static int pm_runlist_v9(struct packet_manager *pm, uint32_t *buffer, struct pm4_mes_runlist *packet; int concurrent_proc_cnt = 0; - struct kfd_dev *kfd = pm->dqm->dev; + struct kfd_node *kfd = pm->dqm->dev; /* Determine the number of processes to map together to HW: * it can not exceed the number of VMIDs available to the @@ -220,13 +243,24 @@ static int pm_map_queues_v9(struct packet_manager *pm, uint32_t *buffer, case KFD_QUEUE_TYPE_SDMA: case KFD_QUEUE_TYPE_SDMA_XGMI: use_static = false; /* no static queues under SDMA */ - if (q->properties.sdma_engine_id < 2 && !pm_use_ext_eng(q->device)) + if (q->properties.sdma_engine_id < 2 && + !pm_use_ext_eng(q->device->kfd)) packet->bitfields2.engine_sel = q->properties.sdma_engine_id + engine_sel__mes_map_queues__sdma0_vi; else { - packet->bitfields2.extended_engine_sel = - extended_engine_sel__mes_map_queues__sdma0_to_7_sel; - packet->bitfields2.engine_sel = q->properties.sdma_engine_id; + /* + * For GFX9.4.3, SDMA engine id can be greater than 8. + * For such cases, set extended_engine_sel to 2 and + * ensure engine_sel lies between 0-7. + */ + if (q->properties.sdma_engine_id >= 8) + packet->bitfields2.extended_engine_sel = + extended_engine_sel__mes_map_queues__sdma8_to_15_sel; + else + packet->bitfields2.extended_engine_sel = + extended_engine_sel__mes_map_queues__sdma0_to_7_sel; + + packet->bitfields2.engine_sel = q->properties.sdma_engine_id % 8; } break; default: @@ -251,6 +285,41 @@ static int pm_map_queues_v9(struct packet_manager *pm, uint32_t *buffer, return 0; } +static int pm_set_grace_period_v9(struct packet_manager *pm, + uint32_t *buffer, + uint32_t grace_period) +{ + struct pm4_mec_write_data_mmio *packet; + uint32_t reg_offset = 0; + uint32_t reg_data = 0; + + pm->dqm->dev->kfd2kgd->build_grace_period_packet_info( + pm->dqm->dev->adev, + pm->dqm->wait_times, + grace_period, + ®_offset, + ®_data); + + if (grace_period == USE_DEFAULT_GRACE_PERIOD) + reg_data = pm->dqm->wait_times; + + packet = (struct pm4_mec_write_data_mmio *)buffer; + memset(buffer, 0, sizeof(struct pm4_mec_write_data_mmio)); + + packet->header.u32All = pm_build_pm4_header(IT_WRITE_DATA, + sizeof(struct pm4_mec_write_data_mmio)); + + packet->bitfields2.dst_sel = dst_sel___write_data__mem_mapped_register; + packet->bitfields2.addr_incr = + addr_incr___write_data__do_not_increment_address; + + packet->bitfields3.dst_mmreg_addr = reg_offset; + + packet->data = reg_data; + + return 0; +} + static int pm_unmap_queues_v9(struct packet_manager *pm, uint32_t *buffer, enum kfd_unmap_queues_filter filter, uint32_t filter_param, bool reset) @@ -263,7 +332,8 @@ static int pm_unmap_queues_v9(struct packet_manager *pm, uint32_t *buffer, packet->header.u32All = pm_build_pm4_header(IT_UNMAP_QUEUES, sizeof(struct pm4_mes_unmap_queues)); - packet->bitfields2.extended_engine_sel = pm_use_ext_eng(pm->dqm->dev) ? + packet->bitfields2.extended_engine_sel = + pm_use_ext_eng(pm->dqm->dev->kfd) ? extended_engine_sel__mes_unmap_queues__sdma0_to_7_sel : extended_engine_sel__mes_unmap_queues__legacy_engine_sel; @@ -333,6 +403,7 @@ const struct packet_manager_funcs kfd_v9_pm_funcs = { .set_resources = pm_set_resources_v9, .map_queues = pm_map_queues_v9, .unmap_queues = pm_unmap_queues_v9, + .set_grace_period = pm_set_grace_period_v9, .query_status = pm_query_status_v9, .release_mem = NULL, .map_process_size = sizeof(struct pm4_mes_map_process), @@ -340,6 +411,7 @@ const struct packet_manager_funcs kfd_v9_pm_funcs = { .set_resources_size = sizeof(struct pm4_mes_set_resources), .map_queues_size = sizeof(struct pm4_mes_map_queues), .unmap_queues_size = sizeof(struct pm4_mes_unmap_queues), + .set_grace_period_size = sizeof(struct pm4_mec_write_data_mmio), .query_status_size = sizeof(struct pm4_mes_query_status), .release_mem_size = 0, }; @@ -350,6 +422,7 @@ const struct packet_manager_funcs kfd_aldebaran_pm_funcs = { .set_resources = pm_set_resources_v9, .map_queues = pm_map_queues_v9, .unmap_queues = pm_unmap_queues_v9, + .set_grace_period = pm_set_grace_period_v9, .query_status = pm_query_status_v9, .release_mem = NULL, .map_process_size = sizeof(struct pm4_mes_map_process_aldebaran), @@ -357,6 +430,7 @@ const struct packet_manager_funcs kfd_aldebaran_pm_funcs = { .set_resources_size = sizeof(struct pm4_mes_set_resources), .map_queues_size = sizeof(struct pm4_mes_map_queues), .unmap_queues_size = sizeof(struct pm4_mes_unmap_queues), + .set_grace_period_size = sizeof(struct pm4_mec_write_data_mmio), .query_status_size = sizeof(struct pm4_mes_query_status), .release_mem_size = 0, }; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_vi.c index 4f951eaa6ee854c2bdca252d99ae9849b6e9b368..c1199d06d131b6ef8db8e8034f433dd38454cc4f 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_vi.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_vi.c @@ -77,7 +77,7 @@ static int pm_runlist_vi(struct packet_manager *pm, uint32_t *buffer, { struct pm4_mes_runlist *packet; int concurrent_proc_cnt = 0; - struct kfd_dev *kfd = pm->dqm->dev; + struct kfd_node *kfd = pm->dqm->dev; if (WARN_ON(!ib)) return -EFAULT; @@ -303,6 +303,7 @@ const struct packet_manager_funcs kfd_vi_pm_funcs = { .set_resources = pm_set_resources_vi, .map_queues = pm_map_queues_vi, .unmap_queues = pm_unmap_queues_vi, + .set_grace_period = NULL, .query_status = pm_query_status_vi, .release_mem = pm_release_mem_vi, .map_process_size = sizeof(struct pm4_mes_map_process), @@ -310,6 +311,7 @@ const struct packet_manager_funcs kfd_vi_pm_funcs = { .set_resources_size = sizeof(struct pm4_mes_set_resources), .map_queues_size = sizeof(struct pm4_mes_map_queues), .unmap_queues_size = sizeof(struct pm4_mes_unmap_queues), + .set_grace_period_size = 0, .query_status_size = sizeof(struct pm4_mes_query_status), .release_mem_size = sizeof(struct pm4_mec_release_mem) }; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h b/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h index a666710ed40359d50a3ffe383d16ad7ec84ac50d..8b6b2bd5c148fdbc3b35055c30da55c7e165ade9 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_pm4_headers_ai.h @@ -146,7 +146,10 @@ struct pm4_mes_map_process { union { struct { uint32_t pasid:16; - uint32_t reserved1:8; + uint32_t reserved1:2; + uint32_t debug_vmid:4; + uint32_t new_debug:1; + uint32_t reserved2:1; uint32_t diq_enable:1; uint32_t process_quantum:7; } bitfields2; @@ -263,7 +266,8 @@ enum mes_map_queues_engine_sel_enum { enum mes_map_queues_extended_engine_sel_enum { extended_engine_sel__mes_map_queues__legacy_engine_sel = 0, - extended_engine_sel__mes_map_queues__sdma0_to_7_sel = 1 + extended_engine_sel__mes_map_queues__sdma0_to_7_sel = 1, + extended_engine_sel__mes_map_queues__sdma8_to_15_sel = 2 }; struct pm4_mes_map_queues { @@ -583,6 +587,71 @@ struct pm4_mec_release_mem { #endif +#ifndef PM4_MEC_WRITE_DATA_DEFINED +#define PM4_MEC_WRITE_DATA_DEFINED + +enum WRITE_DATA_dst_sel_enum { + dst_sel___write_data__mem_mapped_register = 0, + dst_sel___write_data__tc_l2 = 2, + dst_sel___write_data__gds = 3, + dst_sel___write_data__memory = 5, + dst_sel___write_data__memory_mapped_adc_persistent_state = 6, +}; + +enum WRITE_DATA_addr_incr_enum { + addr_incr___write_data__increment_address = 0, + addr_incr___write_data__do_not_increment_address = 1 +}; + +enum WRITE_DATA_wr_confirm_enum { + wr_confirm___write_data__do_not_wait_for_write_confirmation = 0, + wr_confirm___write_data__wait_for_write_confirmation = 1 +}; + +enum WRITE_DATA_cache_policy_enum { + cache_policy___write_data__lru = 0, + cache_policy___write_data__stream = 1 +}; + + +struct pm4_mec_write_data_mmio { + union { + union PM4_MES_TYPE_3_HEADER header; /*header */ + unsigned int ordinal1; + }; + + union { + struct { + unsigned int reserved1:8; + unsigned int dst_sel:4; + unsigned int reserved2:4; + unsigned int addr_incr:1; + unsigned int reserved3:2; + unsigned int resume_vf:1; + unsigned int wr_confirm:1; + unsigned int reserved4:4; + unsigned int cache_policy:2; + unsigned int reserved5:5; + } bitfields2; + unsigned int ordinal2; + }; + + union { + struct { + unsigned int dst_mmreg_addr:18; + unsigned int reserved6:14; + } bitfields3; + unsigned int ordinal3; + }; + + uint32_t reserved7; + + uint32_t data; + +}; + +#endif + enum { CACHE_FLUSH_AND_INV_TS_EVENT = 0x00000014 }; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index 94a438956868663c99ba83947d030444e9257e4c..7364a5d77c6ee065487df119c817bed2921821a5 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -113,6 +113,8 @@ #define KFD_UNMAP_LATENCY_MS (4000) +#define KFD_MAX_SDMA_QUEUES 128 + /* * 512 = 0x200 * The doorbell index distance between SDMA RLC (2*i) and (2*i+1) in the @@ -199,6 +201,8 @@ extern int amdgpu_no_queue_eviction_on_vm_fault; /* Enable eviction debug messages */ extern bool debug_evictions; +extern struct mutex kfd_processes_mutex; + enum cache_policy { cache_policy_coherent, cache_policy_noncoherent @@ -210,11 +214,13 @@ enum cache_policy { ((KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2)) || \ (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 3))) +struct kfd_node; + struct kfd_event_interrupt_class { - bool (*interrupt_isr)(struct kfd_dev *dev, + bool (*interrupt_isr)(struct kfd_node *dev, const uint32_t *ih_ring_entry, uint32_t *patched_ihre, bool *patched_flag); - void (*interrupt_wq)(struct kfd_dev *dev, + void (*interrupt_wq)(struct kfd_node *dev, const uint32_t *ih_ring_entry); }; @@ -233,11 +239,11 @@ struct kfd_device_info { uint32_t no_atomic_fw_version; unsigned int num_sdma_queues_per_engine; unsigned int num_reserved_sdma_queues_per_engine; - uint64_t reserved_sdma_queues_bitmap; + DECLARE_BITMAP(reserved_sdma_queues_bitmap, KFD_MAX_SDMA_QUEUES); }; -unsigned int kfd_get_num_sdma_engines(struct kfd_dev *kdev); -unsigned int kfd_get_num_xgmi_sdma_engines(struct kfd_dev *kdev); +unsigned int kfd_get_num_sdma_engines(struct kfd_node *kdev); +unsigned int kfd_get_num_xgmi_sdma_engines(struct kfd_node *kdev); struct kfd_mem_obj { uint32_t range_start; @@ -253,13 +259,70 @@ struct kfd_vmid_info { uint32_t vmid_num_kfd; }; +#define MAX_KFD_NODES 8 + +struct kfd_dev; + +struct kfd_node { + unsigned int node_id; + struct amdgpu_device *adev; /* Duplicated here along with keeping + * a copy in kfd_dev to save a hop + */ + const struct kfd2kgd_calls *kfd2kgd; /* Duplicated here along with + * keeping a copy in kfd_dev to + * save a hop + */ + struct kfd_vmid_info vm_info; + unsigned int id; /* topology stub index */ + uint32_t xcc_mask; /* Instance mask of XCCs present */ + struct amdgpu_xcp *xcp; + + /* Interrupts */ + struct kfifo ih_fifo; + struct workqueue_struct *ih_wq; + struct work_struct interrupt_work; + spinlock_t interrupt_lock; + + /* + * Interrupts of interest to KFD are copied + * from the HW ring into a SW ring. + */ + bool interrupts_active; + uint32_t interrupt_bitmap; /* Only used for GFX 9.4.3 */ + + /* QCM Device instance */ + struct device_queue_manager *dqm; + + /* Global GWS resource shared between processes */ + void *gws; + bool gws_debug_workaround; + + /* Clients watching SMI events */ + struct list_head smi_clients; + spinlock_t smi_lock; + uint32_t reset_seq_num; + + /* SRAM ECC flag */ + atomic_t sram_ecc_flag; + + /*spm process id */ + unsigned int spm_pasid; + + /* Maximum process number mapped to HW scheduler */ + unsigned int max_proc_per_quantum; + + unsigned int compute_vmid_bitmap; + + struct kfd_local_mem_info local_mem_info; + + struct kfd_dev *kfd; +}; + struct kfd_dev { struct amdgpu_device *adev; struct kfd_device_info device_info; - unsigned int id; /* topology stub index */ - phys_addr_t doorbell_base; /* Start of actual doorbells used by * KFD. It is aligned for mapping * into user mode @@ -274,8 +337,6 @@ struct kfd_dev { */ struct kgd2kfd_shared_resources shared_resources; - struct kfd_vmid_info vm_info; - struct kfd_local_mem_info local_mem_info; const struct kfd2kgd_calls *kfd2kgd; struct mutex doorbell_mutex; @@ -290,30 +351,13 @@ struct kfd_dev { unsigned int gtt_sa_chunk_size; unsigned int gtt_sa_num_of_chunks; - /* Interrupts */ - struct kfifo ih_fifo; - struct workqueue_struct *ih_wq; - struct work_struct interrupt_work; - spinlock_t interrupt_lock; - - /* QCM Device instance */ - struct device_queue_manager *dqm; - bool init_complete; - /* - * Interrupts of interest to KFD are copied - * from the HW ring into a SW ring. - */ - bool interrupts_active; /* Firmware versions */ uint16_t mec_fw_version; uint16_t mec2_fw_version; uint16_t sdma_fw_version; - /* Maximum process number mapped to HW scheduler */ - unsigned int max_proc_per_quantum; - /* CWSR */ bool cwsr_enabled; const void *cwsr_isa; @@ -327,28 +371,20 @@ struct kfd_dev { /* Use IOMMU v2 flag */ bool use_iommu_v2; - /* SRAM ECC flag */ - atomic_t sram_ecc_flag; - /* Compute Profile ref. count */ atomic_t compute_profile; - /* Global GWS resource shared between processes */ - void *gws; - - /* Clients watching SMI events */ - struct list_head smi_clients; - spinlock_t smi_lock; - - uint32_t reset_seq_num; - struct ida doorbell_ida; unsigned int max_doorbell_slices; int noretry; - /* HMM page migration MEMORY_DEVICE_PRIVATE mapping */ - struct dev_pagemap pgmap; + struct kfd_node *nodes[MAX_KFD_NODES]; + unsigned int num_nodes; + + /* Track per device allocated watch points */ + uint32_t alloc_watch_ids; + spinlock_t watch_points_lock; }; enum kfd_mempool { @@ -478,8 +514,13 @@ struct queue_properties { uint32_t doorbell_off; bool is_interop; bool is_evicted; + bool is_suspended; + bool is_being_destroyed; bool is_active; bool is_gws; + uint32_t pm4_target_xcc; + bool is_dbg_wa; + bool is_user_cu_masked; /* Not relevant for user mode queues in cp scheduling */ unsigned int vmid; /* Relevant only for sdma queues*/ @@ -494,15 +535,18 @@ struct queue_properties { uint32_t ctl_stack_size; uint64_t tba_addr; uint64_t tma_addr; + uint64_t exception_status; }; #define QUEUE_IS_ACTIVE(q) ((q).queue_size > 0 && \ (q).queue_address != 0 && \ (q).queue_percent > 0 && \ - !(q).is_evicted) + !(q).is_evicted && \ + !(q).is_suspended) enum mqd_update_flag { - UPDATE_FLAG_CU_MASK = 0, + UPDATE_FLAG_DBG_WA_ENABLE = 1, + UPDATE_FLAG_DBG_WA_DISABLE = 2, }; struct mqd_update_info { @@ -563,7 +607,7 @@ struct queue { unsigned int doorbell_id; struct kfd_process *process; - struct kfd_dev *device; + struct kfd_node *device; void *gws; /* procfs */ @@ -697,7 +741,7 @@ enum kfd_pdd_bound { /* Data that is per-process-per device. */ struct kfd_process_device { /* The device that owns this data. */ - struct kfd_dev *dev; + struct kfd_node *dev; /* The process that owns this kfd_process_device. */ struct kfd_process *process; @@ -783,6 +827,18 @@ struct kfd_process_device { uint64_t faults; uint64_t page_in; uint64_t page_out; + + /* Exception code status*/ + uint64_t exception_status; + void *vm_fault_exc_data; + size_t vm_fault_exc_data_size; + + /* Tracks debug per-vmid request settings */ + uint32_t spi_dbg_override; + uint32_t spi_dbg_launch_mode; + uint32_t watch_points[4]; + uint32_t alloc_watch_ids; + /* * If this process has been checkpointed before, then the user * application will use the original gpu_id on the @@ -887,19 +943,57 @@ struct kfd_process { */ unsigned long last_restore_timestamp; + /* Indicates device process is debug attached with reserved vmid. */ + bool debug_trap_enabled; + + /* per-process-per device debug event fd file */ + struct file *dbg_ev_file; + + /* If the process is a kfd debugger, we need to know so we can clean + * up at exit time. If a process enables debugging on itself, it does + * its own clean-up, so we don't set the flag here. We track this by + * counting the number of processes this process is debugging. + */ + atomic_t debugged_process_count; + + /* If the process is a debugged, this is the debugger process */ + struct kfd_process *debugger_process; + /* Kobj for our procfs */ struct kobject *kobj; struct kobject *kobj_queues; struct attribute attr_pasid; + /* Keep track cwsr init */ + bool has_cwsr; + + /* Exception code enable mask and status */ + uint64_t exception_enable_mask; + uint64_t exception_status; + + /* Used to drain stale interrupts */ + wait_queue_head_t wait_irq_drain; + bool irq_drain_is_open; + /* shared virtual memory registered by this process */ struct svm_range_list svms; bool xnack_enabled; + /* Work area for debugger event writer worker. */ + struct work_struct debug_event_workarea; + + /* Tracks debug per-vmid request for debug flags */ + bool dbg_flags; + atomic_t poison; /* Queues are in paused stated because we are in the process of doing a CRIU checkpoint */ bool queues_paused; + + /* Tracks runtime enable status */ + struct semaphore runtime_enable_sema; + bool is_runtime_retry; + struct kfd_runtime_info runtime_info; }; #define KFD_PROCESS_TABLE_SIZE 5 /* bits: 32 entries */ @@ -925,20 +1019,19 @@ struct amdkfd_ioctl_desc { unsigned int cmd_drv; const char *name; }; -bool kfd_dev_is_large_bar(struct kfd_dev *dev); +bool kfd_dev_is_large_bar(struct kfd_node *dev); int kfd_process_create_wq(void); void kfd_process_destroy_wq(void); void kfd_cleanup_processes(void); -struct kfd_process *kfd_create_process(struct file *filep); +struct kfd_process *kfd_create_process(struct task_struct *thread); struct kfd_process *kfd_get_process(const struct task_struct *task); struct kfd_process *kfd_lookup_process_by_pasid(u32 pasid); struct kfd_process *kfd_lookup_process_by_mm(const struct mm_struct *mm); int kfd_process_gpuidx_from_gpuid(struct kfd_process *p, uint32_t gpu_id); -int kfd_process_gpuid_from_adev(struct kfd_process *p, - struct amdgpu_device *adev, uint32_t *gpuid, - uint32_t *gpuidx); +int kfd_process_gpuid_from_node(struct kfd_process *p, struct kfd_node *node, + uint32_t *gpuid, uint32_t *gpuidx); static inline int kfd_process_gpuid_from_gpuidx(struct kfd_process *p, uint32_t gpuidx, uint32_t *gpuid) { return gpuidx < p->n_pdds ? p->pdds[gpuidx]->dev->id : -EINVAL; @@ -961,16 +1054,16 @@ int kfd_process_get_user_gpu_id(struct kfd_process *p, uint32_t actual_gpu_id); int kfd_process_device_init_vm(struct kfd_process_device *pdd, struct file *drm_file); -struct kfd_process_device *kfd_bind_process_to_device(struct kfd_dev *dev, +struct kfd_process_device *kfd_bind_process_to_device(struct kfd_node *dev, struct kfd_process *p); -struct kfd_process_device *kfd_get_process_device_data(struct kfd_dev *dev, +struct kfd_process_device *kfd_get_process_device_data(struct kfd_node *dev, struct kfd_process *p); -struct kfd_process_device *kfd_create_process_device_data(struct kfd_dev *dev, +struct kfd_process_device *kfd_create_process_device_data(struct kfd_node *dev, struct kfd_process *p); bool kfd_process_xnack_mode(struct kfd_process *p, bool supported); -int kfd_reserved_mem_mmap(struct kfd_dev *dev, struct kfd_process *process, +int kfd_reserved_mem_mmap(struct kfd_node *dev, struct kfd_process *process, struct vm_area_struct *vma); /* KFD process API for creating and translating handles */ @@ -994,7 +1087,7 @@ void kfd_pasid_free(u32 pasid); size_t kfd_doorbell_process_slice(struct kfd_dev *kfd); int kfd_doorbell_init(struct kfd_dev *kfd); void kfd_doorbell_fini(struct kfd_dev *kfd); -int kfd_doorbell_mmap(struct kfd_dev *dev, struct kfd_process *process, +int kfd_doorbell_mmap(struct kfd_node *dev, struct kfd_process *process, struct vm_area_struct *vma); void __iomem *kfd_get_kernel_doorbell(struct kfd_dev *kfd, unsigned int *doorbell_off); @@ -1012,10 +1105,10 @@ void kfd_free_process_doorbells(struct kfd_dev *kfd, unsigned int doorbell_index); /* GTT Sub-Allocator */ -int kfd_gtt_sa_allocate(struct kfd_dev *kfd, unsigned int size, +int kfd_gtt_sa_allocate(struct kfd_node *node, unsigned int size, struct kfd_mem_obj **mem_obj); -int kfd_gtt_sa_free(struct kfd_dev *kfd, struct kfd_mem_obj *mem_obj); +int kfd_gtt_sa_free(struct kfd_node *node, struct kfd_mem_obj *mem_obj); extern struct device *kfd_device; @@ -1028,27 +1121,53 @@ void kfd_procfs_del_queue(struct queue *q); /* Topology */ int kfd_topology_init(void); void kfd_topology_shutdown(void); -int kfd_topology_add_device(struct kfd_dev *gpu); -int kfd_topology_remove_device(struct kfd_dev *gpu); +int kfd_topology_add_device(struct kfd_node *gpu); +int kfd_topology_remove_device(struct kfd_node *gpu); struct kfd_topology_device *kfd_topology_device_by_proximity_domain( uint32_t proximity_domain); struct kfd_topology_device *kfd_topology_device_by_proximity_domain_no_lock( uint32_t proximity_domain); struct kfd_topology_device *kfd_topology_device_by_id(uint32_t gpu_id); -struct kfd_dev *kfd_device_by_id(uint32_t gpu_id); -struct kfd_dev *kfd_device_by_pci_dev(const struct pci_dev *pdev); -struct kfd_dev *kfd_device_by_adev(const struct amdgpu_device *adev); -int kfd_topology_enum_kfd_devices(uint8_t idx, struct kfd_dev **kdev); +struct kfd_node *kfd_device_by_id(uint32_t gpu_id); +struct kfd_node *kfd_device_by_pci_dev(const struct pci_dev *pdev); +static inline bool kfd_irq_is_from_node(struct kfd_node *node, uint32_t node_id, + uint32_t vmid) +{ + return (node->interrupt_bitmap & (1 << node_id)) != 0 && + (node->compute_vmid_bitmap & (1 << vmid)) != 0; +} +static inline struct kfd_node *kfd_node_by_irq_ids(struct amdgpu_device *adev, + uint32_t node_id, uint32_t vmid) { + struct kfd_dev *dev = adev->kfd.dev; + uint32_t i; + + if (adev->ip_versions[GC_HWIP][0] != IP_VERSION(9, 4, 3)) + return dev->nodes[0]; + + for (i = 0; i < dev->num_nodes; i++) + if (kfd_irq_is_from_node(dev->nodes[i], node_id, vmid)) + return dev->nodes[i]; + + return NULL; +} +int kfd_topology_enum_kfd_devices(uint8_t idx, struct kfd_node **kdev); int kfd_numa_node_to_apic_id(int numa_node_id); void kfd_double_confirm_iommu_support(struct kfd_dev *gpu); /* Interrupts */ -int kfd_interrupt_init(struct kfd_dev *dev); -void kfd_interrupt_exit(struct kfd_dev *dev); -bool enqueue_ih_ring_entry(struct kfd_dev *kfd, const void *ih_ring_entry); -bool interrupt_is_wanted(struct kfd_dev *dev, +#define KFD_IRQ_FENCE_CLIENTID 0xff +#define KFD_IRQ_FENCE_SOURCEID 0xff +#define KFD_IRQ_IS_FENCE(client, source) \ + ((client) == KFD_IRQ_FENCE_CLIENTID && \ + (source) == KFD_IRQ_FENCE_SOURCEID) +int kfd_interrupt_init(struct kfd_node *dev); +void kfd_interrupt_exit(struct kfd_node *dev); +bool enqueue_ih_ring_entry(struct kfd_node *kfd, const void *ih_ring_entry); +bool interrupt_is_wanted(struct kfd_node *dev, const uint32_t *ih_ring_entry, uint32_t *patched_ihre, bool *flag); +int kfd_process_drain_interrupts(struct kfd_process_device *pdd); +void kfd_process_close_interrupt_drain(unsigned int pasid); /* amdkfd Apertures */ int kfd_init_apertures(struct kfd_process *process); @@ -1056,6 +1175,11 @@ int kfd_init_apertures(struct kfd_process *process); void kfd_process_set_trap_handler(struct qcm_process_device *qpd, uint64_t tba_addr, uint64_t tma_addr); +void kfd_process_set_trap_debug_flag(struct qcm_process_device *qpd, + bool enabled); + +/* CWSR initialization */ +int kfd_process_init_cwsr_apu(struct kfd_process *process, struct file *filep); /* CRIU */ /* @@ -1174,22 +1298,22 @@ void print_queue_properties(struct queue_properties *q); void print_queue(struct queue *q); struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type, - struct kfd_dev *dev); + struct kfd_node *dev); struct mqd_manager *mqd_manager_init_cik_hawaii(enum KFD_MQD_TYPE type, - struct kfd_dev *dev); + struct kfd_node *dev); struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type, - struct kfd_dev *dev); + struct kfd_node *dev); struct mqd_manager *mqd_manager_init_vi_tonga(enum KFD_MQD_TYPE type, - struct kfd_dev *dev); + struct kfd_node *dev); struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type, - struct kfd_dev *dev); + struct kfd_node *dev); struct mqd_manager *mqd_manager_init_v10(enum KFD_MQD_TYPE type, - struct kfd_dev *dev); + struct kfd_node *dev); struct mqd_manager *mqd_manager_init_v11(enum KFD_MQD_TYPE type, - struct kfd_dev *dev); -struct device_queue_manager *device_queue_manager_init(struct kfd_dev *dev); + struct kfd_node *dev); +struct device_queue_manager *device_queue_manager_init(struct kfd_node *dev); void device_queue_manager_uninit(struct device_queue_manager *dqm); -struct kernel_queue *kernel_queue_init(struct kfd_dev *dev, +struct kernel_queue *kernel_queue_init(struct kfd_node *dev, enum kfd_queue_type type); void kernel_queue_uninit(struct kernel_queue *kq, bool hanging); int kfd_dqm_evict_pasid(struct device_queue_manager *dqm, u32 pasid); @@ -1206,7 +1330,7 @@ void kfd_process_dequeue_from_all_devices(struct kfd_process *p); int pqm_init(struct process_queue_manager *pqm, struct kfd_process *p); void pqm_uninit(struct process_queue_manager *pqm); int pqm_create_queue(struct process_queue_manager *pqm, - struct kfd_dev *dev, + struct kfd_node *dev, struct file *f, struct queue_properties *properties, unsigned int *qid, @@ -1231,6 +1355,11 @@ int pqm_get_wave_state(struct process_queue_manager *pqm, void __user *ctl_stack, u32 *ctl_stack_used_size, u32 *save_area_used_size); +int pqm_get_queue_snapshot(struct process_queue_manager *pqm, + uint64_t exception_clear_mask, + void __user *buf, + int *num_qss_entries, + uint32_t *entry_size); int amdkfd_fence_wait_timeout(uint64_t *fence_addr, uint64_t fence_value, @@ -1270,6 +1399,8 @@ struct packet_manager_funcs { int (*unmap_queues)(struct packet_manager *pm, uint32_t *buffer, enum kfd_unmap_queues_filter mode, uint32_t filter_param, bool reset); + int (*set_grace_period)(struct packet_manager *pm, uint32_t *buffer, + uint32_t grace_period); int (*query_status)(struct packet_manager *pm, uint32_t *buffer, uint64_t fence_address, uint64_t fence_value); int (*release_mem)(uint64_t gpu_addr, uint32_t *buffer); @@ -1280,6 +1411,7 @@ struct packet_manager_funcs { int set_resources_size; int map_queues_size; int unmap_queues_size; + int set_grace_period_size; int query_status_size; int release_mem_size; }; @@ -1302,6 +1434,8 @@ int pm_send_unmap_queue(struct packet_manager *pm, void pm_release_ib(struct packet_manager *pm); +int pm_update_grace_period(struct packet_manager *pm, uint32_t grace_period); + /* Following PM funcs can be shared among VI and AI */ unsigned int pm_build_pm4_header(unsigned int opcode, size_t packet_size); @@ -1310,6 +1444,7 @@ uint64_t kfd_get_number_elems(struct kfd_dev *kfd); /* Events */ extern const struct kfd_event_interrupt_class event_interrupt_class_cik; extern const struct kfd_event_interrupt_class event_interrupt_class_v9; +extern const struct kfd_event_interrupt_class event_interrupt_class_v10; extern const struct kfd_event_interrupt_class event_interrupt_class_v11; extern const struct kfd_device_global_init_class device_global_init_class_cik; @@ -1323,7 +1458,7 @@ int kfd_wait_on_events(struct kfd_process *p, uint32_t *wait_result); void kfd_signal_event_interrupt(u32 pasid, uint32_t partial_id, uint32_t valid_id_bits); -void kfd_signal_iommu_event(struct kfd_dev *dev, +void kfd_signal_iommu_event(struct kfd_node *dev, u32 pasid, unsigned long address, bool is_write_requested, bool is_execute_requested); void kfd_signal_hw_exception_event(u32 pasid); @@ -1339,32 +1474,36 @@ int kfd_event_create(struct file *devkfd, struct kfd_process *p, int kfd_get_num_events(struct kfd_process *p); int kfd_event_destroy(struct kfd_process *p, uint32_t event_id); -void kfd_signal_vm_fault_event(struct kfd_dev *dev, u32 pasid, - struct kfd_vm_fault_info *info); +void kfd_signal_vm_fault_event(struct kfd_node *dev, u32 pasid, + struct kfd_vm_fault_info *info, + struct kfd_hsa_memory_exception_data *data); -void kfd_signal_reset_event(struct kfd_dev *dev); +void kfd_signal_reset_event(struct kfd_node *dev); -void kfd_signal_poison_consumed_event(struct kfd_dev *dev, u32 pasid); +void kfd_signal_poison_consumed_event(struct kfd_node *dev, u32 pasid); void kfd_flush_tlb(struct kfd_process_device *pdd, enum TLB_FLUSH_TYPE type); static inline bool kfd_flush_tlb_after_unmap(struct kfd_dev *dev) { - return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2) || - (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 1) && - dev->adev->sdma.instance[0].fw_version >= 18) || + return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 3) || + KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2) || + (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 1) && dev->sdma_fw_version >= 18) || KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 0); } +int kfd_send_exception_to_runtime(struct kfd_process *p, + unsigned int queue_id, + uint64_t error_reason); bool kfd_is_locked(void); /* Compute profile */ -void kfd_inc_compute_active(struct kfd_dev *dev); -void kfd_dec_compute_active(struct kfd_dev *dev); +void kfd_inc_compute_active(struct kfd_node *dev); +void kfd_dec_compute_active(struct kfd_node *dev); /* Cgroup Support */ /* Check with device cgroup if @kfd device is accessible */ -static inline int kfd_devcgroup_check_permission(struct kfd_dev *kfd) +static inline int kfd_devcgroup_check_permission(struct kfd_node *kfd) { #if defined(CONFIG_CGROUP_DEVICE) || defined(CONFIG_CGROUP_BPF) struct drm_device *ddev = adev_to_drm(kfd->adev); @@ -1377,6 +1516,11 @@ static inline int kfd_devcgroup_check_permission(struct kfd_dev *kfd) #endif } +static inline bool kfd_is_first_node(struct kfd_node *node) +{ + return (node == node->kfd->nodes[0]); +} + /* Debugfs */ #if defined(CONFIG_DEBUG_FS) @@ -1389,7 +1533,7 @@ int dqm_debugfs_hqds(struct seq_file *m, void *data); int kfd_debugfs_rls_by_device(struct seq_file *m, void *data); int pm_debugfs_runlist(struct seq_file *m, void *data); -int kfd_debugfs_hang_hws(struct kfd_dev *dev); +int kfd_debugfs_hang_hws(struct kfd_node *dev); int pm_debugfs_hang_hws(struct packet_manager *pm); int dqm_debugfs_hang_hws(struct device_queue_manager *dqm); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index 07a9eaf9b7d8e36c8226363a11cdeb1cda45b243..3d3611705d4199b90c8cd9b249b9d5b594852e5f 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -44,13 +44,14 @@ struct mm_struct; #include "kfd_iommu.h" #include "kfd_svm.h" #include "kfd_smi_events.h" +#include "kfd_debug.h" /* * List of struct kfd_process (field kfd_process). * Unique/indexed by mm_struct* */ DEFINE_HASHTABLE(kfd_processes_table, KFD_PROCESS_TABLE_SIZE); -static DEFINE_MUTEX(kfd_processes_mutex); +DEFINE_MUTEX(kfd_processes_mutex); DEFINE_SRCU(kfd_processes_srcu); @@ -69,7 +70,6 @@ static struct kfd_process *find_process(const struct task_struct *thread, bool ref); static void kfd_process_ref_release(struct kref *ref); static struct kfd_process *create_process(const struct task_struct *thread); -static int kfd_process_init_cwsr_apu(struct kfd_process *p, struct file *filep); static void evict_process_worker(struct work_struct *work); static void restore_process_worker(struct work_struct *work); @@ -269,7 +269,7 @@ static int kfd_get_cu_occupancy(struct attribute *attr, char *buffer) int cu_cnt; int wave_cnt; int max_waves_per_cu; - struct kfd_dev *dev = NULL; + struct kfd_node *dev = NULL; struct kfd_process *proc = NULL; struct kfd_process_device *pdd = NULL; @@ -290,7 +290,7 @@ static int kfd_get_cu_occupancy(struct attribute *attr, char *buffer) wave_cnt = 0; max_waves_per_cu = 0; dev->kfd2kgd->get_cu_occupancy(dev->adev, proc->pasid, &wave_cnt, - &max_waves_per_cu); + &max_waves_per_cu, 0); /* Translate wave count to number of compute units */ cu_cnt = (wave_cnt + (max_waves_per_cu - 1)) / max_waves_per_cu; @@ -691,7 +691,7 @@ void kfd_process_destroy_wq(void) static void kfd_process_free_gpuvm(struct kgd_mem *mem, struct kfd_process_device *pdd, void **kptr) { - struct kfd_dev *dev = pdd->dev; + struct kfd_node *dev = pdd->dev; if (kptr && *kptr) { amdgpu_amdkfd_gpuvm_unmap_gtt_bo_from_kernel(mem); @@ -713,7 +713,7 @@ static int kfd_process_alloc_gpuvm(struct kfd_process_device *pdd, uint64_t gpu_va, uint32_t size, uint32_t flags, struct kgd_mem **mem, void **kptr) { - struct kfd_dev *kdev = pdd->dev; + struct kfd_node *kdev = pdd->dev; int err; err = amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(kdev->adev, gpu_va, size, @@ -798,18 +798,19 @@ static void kfd_process_device_destroy_ib_mem(struct kfd_process_device *pdd) kfd_process_free_gpuvm(qpd->ib_mem, pdd, &qpd->ib_kaddr); } -struct kfd_process *kfd_create_process(struct file *filep) +struct kfd_process *kfd_create_process(struct task_struct *thread) { struct kfd_process *process; - struct task_struct *thread = current; int ret; - if (!thread->mm) + if (!(thread->mm && mmget_not_zero(thread->mm))) return ERR_PTR(-EINVAL); /* Only the pthreads threading model is supported. */ - if (thread->group_leader->mm != thread->mm) + if (thread->group_leader->mm != thread->mm) { + mmput(thread->mm); return ERR_PTR(-EINVAL); + } /* * take kfd processes mutex before starting of process creation @@ -818,6 +819,12 @@ struct kfd_process *kfd_create_process(struct file *filep) */ mutex_lock(&kfd_processes_mutex); + if (kfd_is_locked()) { + mutex_unlock(&kfd_processes_mutex); + pr_debug("KFD is locked! Cannot create process"); + return ERR_PTR(-EINVAL); + } + /* A prior open of /dev/kfd could have already created the process. */ process = find_process(thread, false); if (process) { @@ -827,10 +834,6 @@ struct kfd_process *kfd_create_process(struct file *filep) if (IS_ERR(process)) goto out; - ret = kfd_process_init_cwsr_apu(process, filep); - if (ret) - goto out_destroy; - if (!procfs.kobj) goto out; @@ -859,21 +862,16 @@ struct kfd_process *kfd_create_process(struct file *filep) kfd_procfs_add_sysfs_stats(process); kfd_procfs_add_sysfs_files(process); kfd_procfs_add_sysfs_counters(process); + + init_waitqueue_head(&process->wait_irq_drain); } out: if (!IS_ERR(process)) kref_get(&process->ref); mutex_unlock(&kfd_processes_mutex); + mmput(thread->mm); return process; - -out_destroy: - hash_del_rcu(&process->kfd_processes); - mutex_unlock(&kfd_processes_mutex); - synchronize_srcu(&kfd_processes_srcu); - /* kfd_process_free_notifier will trigger the cleanup */ - mmu_notifier_put(&process->mmu_notifier); - return ERR_PTR(ret); } struct kfd_process *kfd_get_process(const struct task_struct *thread) @@ -982,7 +980,7 @@ static void kfd_process_device_free_bos(struct kfd_process_device *pdd) static void kfd_process_kunmap_signal_bo(struct kfd_process *p) { struct kfd_process_device *pdd; - struct kfd_dev *kdev; + struct kfd_node *kdev; void *mem; kdev = kfd_device_by_id(GET_GPU_ID(p->signal_handle)); @@ -1040,9 +1038,9 @@ static void kfd_process_destroy_pdds(struct kfd_process *p) bitmap_free(pdd->qpd.doorbell_bitmap); idr_destroy(&pdd->alloc_idr); - kfd_free_process_doorbells(pdd->dev, pdd->doorbell_index); + kfd_free_process_doorbells(pdd->dev->kfd, pdd->doorbell_index); - if (pdd->dev->shared_resources.enable_mes) + if (pdd->dev->kfd->shared_resources.enable_mes) amdgpu_amdkfd_free_gtt_mem(pdd->dev->adev, pdd->proc_ctx_bo); /* @@ -1169,11 +1167,40 @@ static void kfd_process_free_notifier(struct mmu_notifier *mn) static void kfd_process_notifier_release_internal(struct kfd_process *p) { + int i; + cancel_delayed_work_sync(&p->eviction_work); cancel_delayed_work_sync(&p->restore_work); + for (i = 0; i < p->n_pdds; i++) { + struct kfd_process_device *pdd = p->pdds[i]; + + /* re-enable GFX OFF since runtime enable with ttmp setup disabled it. */ + if (!kfd_dbg_is_rlc_restore_supported(pdd->dev) && p->runtime_info.ttmp_setup) + amdgpu_gfx_off_ctrl(pdd->dev->adev, true); + } + /* Indicate to other users that MM is no longer valid */ p->mm = NULL; + kfd_dbg_trap_disable(p); + + if (atomic_read(&p->debugged_process_count) > 0) { + struct kfd_process *target; + unsigned int temp; + int idx = srcu_read_lock(&kfd_processes_srcu); + + hash_for_each_rcu(kfd_processes_table, temp, target, kfd_processes) { + if (target->debugger_process && target->debugger_process == p) { + mutex_lock_nested(&target->mutex, 1); + kfd_dbg_trap_disable(target); + mutex_unlock(&target->mutex); + if (atomic_read(&p->debugged_process_count) == 0) + break; + } + } + + srcu_read_unlock(&kfd_processes_srcu, idx); + } mmu_notifier_put(&p->mmu_notifier); } @@ -1253,16 +1280,19 @@ void kfd_cleanup_processes(void) mmu_notifier_synchronize(); } -static int kfd_process_init_cwsr_apu(struct kfd_process *p, struct file *filep) +int kfd_process_init_cwsr_apu(struct kfd_process *p, struct file *filep) { unsigned long offset; int i; + if (p->has_cwsr) + return 0; + for (i = 0; i < p->n_pdds; i++) { - struct kfd_dev *dev = p->pdds[i]->dev; + struct kfd_node *dev = p->pdds[i]->dev; struct qcm_process_device *qpd = &p->pdds[i]->qpd; - if (!dev->cwsr_enabled || qpd->cwsr_kaddr || qpd->cwsr_base) + if (!dev->kfd->cwsr_enabled || qpd->cwsr_kaddr || qpd->cwsr_base) continue; offset = KFD_MMAP_TYPE_RESERVED_MEM | KFD_MMAP_GPU_ID(dev->id); @@ -1279,19 +1309,23 @@ static int kfd_process_init_cwsr_apu(struct kfd_process *p, struct file *filep) return err; } - memcpy(qpd->cwsr_kaddr, dev->cwsr_isa, dev->cwsr_isa_size); + memcpy(qpd->cwsr_kaddr, dev->kfd->cwsr_isa, dev->kfd->cwsr_isa_size); + + kfd_process_set_trap_debug_flag(qpd, p->debug_trap_enabled); qpd->tma_addr = qpd->tba_addr + KFD_CWSR_TMA_OFFSET; pr_debug("set tba :0x%llx, tma:0x%llx, cwsr_kaddr:%p for pqm.\n", qpd->tba_addr, qpd->tma_addr, qpd->cwsr_kaddr); } + p->has_cwsr = true; + return 0; } static int kfd_process_device_init_cwsr_dgpu(struct kfd_process_device *pdd) { - struct kfd_dev *dev = pdd->dev; + struct kfd_node *dev = pdd->dev; struct qcm_process_device *qpd = &pdd->qpd; uint32_t flags = KFD_IOC_ALLOC_MEM_FLAGS_GTT | KFD_IOC_ALLOC_MEM_FLAGS_NO_SUBSTITUTE @@ -1300,7 +1334,7 @@ static int kfd_process_device_init_cwsr_dgpu(struct kfd_process_device *pdd) void *kaddr; int ret; - if (!dev->cwsr_enabled || qpd->cwsr_kaddr || !qpd->cwsr_base) + if (!dev->kfd->cwsr_enabled || qpd->cwsr_kaddr || !qpd->cwsr_base) return 0; /* cwsr_base is only set for dGPU */ @@ -1313,7 +1347,10 @@ static int kfd_process_device_init_cwsr_dgpu(struct kfd_process_device *pdd) qpd->cwsr_kaddr = kaddr; qpd->tba_addr = qpd->cwsr_base; - memcpy(qpd->cwsr_kaddr, dev->cwsr_isa, dev->cwsr_isa_size); + memcpy(qpd->cwsr_kaddr, dev->kfd->cwsr_isa, dev->kfd->cwsr_isa_size); + + kfd_process_set_trap_debug_flag(&pdd->qpd, + pdd->process->debug_trap_enabled); qpd->tma_addr = qpd->tba_addr + KFD_CWSR_TMA_OFFSET; pr_debug("set tba :0x%llx, tma:0x%llx, cwsr_kaddr:%p for pqm.\n", @@ -1324,10 +1361,10 @@ static int kfd_process_device_init_cwsr_dgpu(struct kfd_process_device *pdd) static void kfd_process_device_destroy_cwsr_dgpu(struct kfd_process_device *pdd) { - struct kfd_dev *dev = pdd->dev; + struct kfd_node *dev = pdd->dev; struct qcm_process_device *qpd = &pdd->qpd; - if (!dev->cwsr_enabled || !qpd->cwsr_kaddr || !qpd->cwsr_base) + if (!dev->kfd->cwsr_enabled || !qpd->cwsr_kaddr || !qpd->cwsr_base) return; kfd_process_free_gpuvm(qpd->cwsr_mem, pdd, &qpd->cwsr_kaddr); @@ -1371,7 +1408,7 @@ bool kfd_process_xnack_mode(struct kfd_process *p, bool supported) * support retry. */ for (i = 0; i < p->n_pdds; i++) { - struct kfd_dev *dev = p->pdds[i]->dev; + struct kfd_node *dev = p->pdds[i]->dev; /* Only consider GFXv9 and higher GPUs. Older GPUs don't * support the SVM APIs and don't need to be considered @@ -1394,13 +1431,23 @@ bool kfd_process_xnack_mode(struct kfd_process *p, bool supported) if (KFD_GC_VERSION(dev) >= IP_VERSION(10, 1, 1)) return false; - if (dev->noretry) + if (dev->kfd->noretry) return false; } return true; } +void kfd_process_set_trap_debug_flag(struct qcm_process_device *qpd, + bool enabled) +{ + if (qpd->cwsr_kaddr) { + uint64_t *tma = + (uint64_t *)(qpd->cwsr_kaddr + KFD_CWSR_TMA_OFFSET); + tma[2] = enabled; + } +} + /* * On return the kfd_process is fully operational and will be freed when the * mm is released @@ -1428,6 +1475,11 @@ static struct kfd_process *create_process(const struct task_struct *thread) if (err) goto err_event_init; process->is_32bit_user_mode = in_compat_syscall(); + process->debug_trap_enabled = false; + process->debugger_process = NULL; + process->exception_enable_mask = 0; + atomic_set(&process->debugged_process_count, 0); + sema_init(&process->runtime_enable_sema, 0); process->pasid = kfd_pasid_alloc(); if (process->pasid == 0) { @@ -1475,6 +1527,8 @@ static struct kfd_process *create_process(const struct task_struct *thread) kfd_unref_process(process); get_task_struct(process->lead_thread); + INIT_WORK(&process->debug_event_workarea, debug_event_write_work_handler); + return process; err_register_notifier: @@ -1528,7 +1582,7 @@ static int init_doorbell_bitmap(struct qcm_process_device *qpd, return 0; } -struct kfd_process_device *kfd_get_process_device_data(struct kfd_dev *dev, +struct kfd_process_device *kfd_get_process_device_data(struct kfd_node *dev, struct kfd_process *p) { int i; @@ -1540,7 +1594,7 @@ struct kfd_process_device *kfd_get_process_device_data(struct kfd_dev *dev, return NULL; } -struct kfd_process_device *kfd_create_process_device_data(struct kfd_dev *dev, +struct kfd_process_device *kfd_create_process_device_data(struct kfd_node *dev, struct kfd_process *p) { struct kfd_process_device *pdd = NULL; @@ -1552,7 +1606,7 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_dev *dev, if (!pdd) return NULL; - if (init_doorbell_bitmap(&pdd->qpd, dev)) { + if (init_doorbell_bitmap(&pdd->qpd, dev->kfd)) { pr_err("Failed to init doorbell for process\n"); goto err_free_pdd; } @@ -1573,7 +1627,7 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_dev *dev, pdd->user_gpu_id = dev->id; atomic64_set(&pdd->evict_duration_counter, 0); - if (dev->shared_resources.enable_mes) { + if (dev->kfd->shared_resources.enable_mes) { retval = amdgpu_amdkfd_alloc_gtt_mem(dev->adev, AMDGPU_MES_PROC_CTX_SIZE, &pdd->proc_ctx_bo, @@ -1588,6 +1642,11 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_dev *dev, } p->pdds[p->n_pdds++] = pdd; + if (kfd_dbg_is_per_vmid_supported(pdd->dev)) + pdd->spi_dbg_override = pdd->dev->kfd2kgd->disable_debug_trap( + pdd->dev->adev, + false, + 0); /* Init idr used for memory handle translation */ idr_init(&pdd->alloc_idr); @@ -1619,7 +1678,7 @@ int kfd_process_device_init_vm(struct kfd_process_device *pdd, struct amdgpu_fpriv *drv_priv; struct amdgpu_vm *avm; struct kfd_process *p; - struct kfd_dev *dev; + struct kfd_node *dev; int ret; if (!drm_file) @@ -1679,7 +1738,7 @@ int kfd_process_device_init_vm(struct kfd_process_device *pdd, * * Assumes that the process lock is held. */ -struct kfd_process_device *kfd_bind_process_to_device(struct kfd_dev *dev, +struct kfd_process_device *kfd_bind_process_to_device(struct kfd_node *dev, struct kfd_process *p) { struct kfd_process_device *pdd; @@ -1885,13 +1944,13 @@ int kfd_process_gpuidx_from_gpuid(struct kfd_process *p, uint32_t gpu_id) } int -kfd_process_gpuid_from_adev(struct kfd_process *p, struct amdgpu_device *adev, - uint32_t *gpuid, uint32_t *gpuidx) +kfd_process_gpuid_from_node(struct kfd_process *p, struct kfd_node *node, + uint32_t *gpuid, uint32_t *gpuidx) { int i; for (i = 0; i < p->n_pdds; i++) - if (p->pdds[i] && p->pdds[i]->dev->adev == adev) { + if (p->pdds[i] && p->pdds[i]->dev == node) { *gpuid = p->pdds[i]->user_gpu_id; *gpuidx = i; return 0; @@ -1961,8 +2020,10 @@ static void restore_process_worker(struct work_struct *work) */ p->last_restore_timestamp = get_jiffies_64(); - ret = amdgpu_amdkfd_gpuvm_restore_process_bos(p->kgd_process_info, - &p->ef); + /* VMs may not have been acquired yet during debugging. */ + if (p->kgd_process_info) + ret = amdgpu_amdkfd_gpuvm_restore_process_bos(p->kgd_process_info, + &p->ef); if (ret) { pr_debug("Failed to restore BOs of pasid 0x%x, retry after %d ms\n", p->pasid, PROCESS_BACK_OFF_TIME_MS); @@ -1988,7 +2049,7 @@ void kfd_suspend_all_processes(void) WARN(debug_evictions, "Evicting all processes"); hash_for_each_rcu(kfd_processes_table, temp, p, kfd_processes) { cancel_delayed_work_sync(&p->eviction_work); - cancel_delayed_work_sync(&p->restore_work); + flush_delayed_work(&p->restore_work); if (kfd_process_evict_queues(p, KFD_QUEUE_EVICTION_TRIGGER_SUSPEND)) pr_err("Failed to suspend process 0x%x\n", p->pasid); @@ -2016,7 +2077,7 @@ int kfd_resume_all_processes(void) return ret; } -int kfd_reserved_mem_mmap(struct kfd_dev *dev, struct kfd_process *process, +int kfd_reserved_mem_mmap(struct kfd_node *dev, struct kfd_process *process, struct vm_area_struct *vma) { struct kfd_process_device *pdd; @@ -2051,7 +2112,9 @@ void kfd_flush_tlb(struct kfd_process_device *pdd, enum TLB_FLUSH_TYPE type) { struct amdgpu_vm *vm = drm_priv_to_vm(pdd->drm_priv); uint64_t tlb_seq = amdgpu_vm_tlb_seq(vm); - struct kfd_dev *dev = pdd->dev; + struct kfd_node *dev = pdd->dev; + uint32_t xcc_mask = dev->xcc_mask; + int xcc = 0; /* * It can be that we race and lose here, but that is extremely unlikely @@ -2069,11 +2132,126 @@ void kfd_flush_tlb(struct kfd_process_device *pdd, enum TLB_FLUSH_TYPE type) amdgpu_amdkfd_flush_gpu_tlb_vmid(dev->adev, pdd->qpd.vmid); } else { - amdgpu_amdkfd_flush_gpu_tlb_pasid(dev->adev, - pdd->process->pasid, type); + for_each_inst(xcc, xcc_mask) + amdgpu_amdkfd_flush_gpu_tlb_pasid( + dev->adev, pdd->process->pasid, type, xcc); } } +/* assumes caller holds process lock. */ +int kfd_process_drain_interrupts(struct kfd_process_device *pdd) +{ + uint32_t irq_drain_fence[8]; + int r = 0; + + if (!KFD_IS_SOC15(pdd->dev)) + return 0; + + pdd->process->irq_drain_is_open = true; + + memset(irq_drain_fence, 0, sizeof(irq_drain_fence)); + irq_drain_fence[0] = (KFD_IRQ_FENCE_SOURCEID << 8) | + KFD_IRQ_FENCE_CLIENTID; + irq_drain_fence[3] = pdd->process->pasid; + + /* ensure stale irqs scheduled KFD interrupts and send drain fence. */ + if (amdgpu_amdkfd_send_close_event_drain_irq(pdd->dev->adev, + irq_drain_fence)) { + pdd->process->irq_drain_is_open = false; + return 0; + } + + r = wait_event_interruptible(pdd->process->wait_irq_drain, + !READ_ONCE(pdd->process->irq_drain_is_open)); + if (r) + pdd->process->irq_drain_is_open = false; + + return r; +} + +void kfd_process_close_interrupt_drain(unsigned int pasid) +{ + struct kfd_process *p; + + p = kfd_lookup_process_by_pasid(pasid); + + if (!p) + return; + + WRITE_ONCE(p->irq_drain_is_open, false); + wake_up_all(&p->wait_irq_drain); + kfd_unref_process(p); +} + +struct send_exception_work_handler_workarea { + struct work_struct work; + struct kfd_process *p; + unsigned int queue_id; + uint64_t error_reason; +}; + +static void send_exception_work_handler(struct work_struct *work) +{ + struct send_exception_work_handler_workarea *workarea; + struct kfd_process *p; + struct queue *q; + struct mm_struct *mm; + struct kfd_context_save_area_header __user *csa_header; + uint64_t __user *err_payload_ptr; + uint64_t cur_err; + uint32_t ev_id; + + workarea = container_of(work, + struct send_exception_work_handler_workarea, + work); + p = workarea->p; + + mm = get_task_mm(p->lead_thread); + + if (!mm) + return; + + kthread_use_mm(mm); + + q = pqm_get_user_queue(&p->pqm, workarea->queue_id); + + if (!q) + goto out; + + csa_header = (void __user *)q->properties.ctx_save_restore_area_address; + + get_user(err_payload_ptr, (uint64_t __user **)&csa_header->err_payload_addr); + get_user(cur_err, err_payload_ptr); + cur_err |= workarea->error_reason; + put_user(cur_err, err_payload_ptr); + get_user(ev_id, &csa_header->err_event_id); + + kfd_set_event(p, ev_id); + +out: + kthread_unuse_mm(mm); + mmput(mm); +} + +int kfd_send_exception_to_runtime(struct kfd_process *p, + unsigned int queue_id, + uint64_t error_reason) +{ + struct send_exception_work_handler_workarea worker; + + INIT_WORK_ONSTACK(&worker.work, send_exception_work_handler); + + worker.p = p; + worker.queue_id = queue_id; + worker.error_reason = error_reason; + + schedule_work(&worker.work); + flush_work(&worker.work); + destroy_work_on_stack(&worker.work); + + return 0; +} + struct kfd_process_device *kfd_process_device_data_by_id(struct kfd_process *p, uint32_t gpu_id) { int i; @@ -2133,4 +2311,3 @@ int kfd_debugfs_mqds_by_process(struct seq_file *m, void *data) } #endif - diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c index 4236539d9f932e34c62357767972c53c9498f92a..9ad1a2186a245ef74c07681d9324648d2dfb9628 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c @@ -81,7 +81,7 @@ static int find_available_queue_slot(struct process_queue_manager *pqm, void kfd_process_dequeue_from_device(struct kfd_process_device *pdd) { - struct kfd_dev *dev = pdd->dev; + struct kfd_node *dev = pdd->dev; if (pdd->already_dequeued) return; @@ -93,7 +93,7 @@ void kfd_process_dequeue_from_device(struct kfd_process_device *pdd) int pqm_set_gws(struct process_queue_manager *pqm, unsigned int qid, void *gws) { - struct kfd_dev *dev = NULL; + struct kfd_node *dev = NULL; struct process_queue_node *pqn; struct kfd_process_device *pdd; struct kgd_mem *mem = NULL; @@ -178,7 +178,7 @@ void pqm_uninit(struct process_queue_manager *pqm) } static int init_user_queue(struct process_queue_manager *pqm, - struct kfd_dev *dev, struct queue **q, + struct kfd_node *dev, struct queue **q, struct queue_properties *q_properties, struct file *f, struct amdgpu_bo *wptr_bo, unsigned int qid) @@ -187,6 +187,7 @@ static int init_user_queue(struct process_queue_manager *pqm, /* Doorbell initialized in user space*/ q_properties->doorbell_ptr = NULL; + q_properties->exception_status = KFD_EC_MASK(EC_QUEUE_NEW); /* let DQM handle it*/ q_properties->vmid = 0; @@ -199,7 +200,7 @@ static int init_user_queue(struct process_queue_manager *pqm, (*q)->device = dev; (*q)->process = pqm->process; - if (dev->shared_resources.enable_mes) { + if (dev->kfd->shared_resources.enable_mes) { retval = amdgpu_amdkfd_alloc_gtt_mem(dev->adev, AMDGPU_MES_GANG_CTX_SIZE, &(*q)->gang_ctx_bo, @@ -224,7 +225,7 @@ static int init_user_queue(struct process_queue_manager *pqm, } int pqm_create_queue(struct process_queue_manager *pqm, - struct kfd_dev *dev, + struct kfd_node *dev, struct file *f, struct queue_properties *properties, unsigned int *qid, @@ -242,6 +243,13 @@ int pqm_create_queue(struct process_queue_manager *pqm, enum kfd_queue_type type = properties->type; unsigned int max_queues = 127; /* HWS limit */ + /* + * On GFX 9.4.3, increase the number of queues that + * can be created to 255. No HWS limit on GFX 9.4.3. + */ + if (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 3)) + max_queues = 255; + q = NULL; kq = NULL; @@ -258,7 +266,7 @@ int pqm_create_queue(struct process_queue_manager *pqm, * Hence we also check the type as well */ if ((pdd->qpd.is_debug) || (type == KFD_QUEUE_TYPE_DIQ)) - max_queues = dev->device_info.max_no_of_hqd/2; + max_queues = dev->kfd->device_info.max_no_of_hqd/2; if (pdd->qpd.queue_count >= max_queues) return -ENOSPC; @@ -330,6 +338,10 @@ int pqm_create_queue(struct process_queue_manager *pqm, kq->queue->properties.queue_id = *qid; pqn->kq = kq; pqn->q = NULL; + retval = kfd_process_drain_interrupts(pdd); + if (retval) + break; + retval = dev->dqm->ops.create_kernel_queue(dev->dqm, kq, &pdd->qpd); break; @@ -354,7 +366,7 @@ int pqm_create_queue(struct process_queue_manager *pqm, */ *p_doorbell_offset_in_process = (q->properties.doorbell_off * sizeof(uint32_t)) & - (kfd_doorbell_process_slice(dev) - 1); + (kfd_doorbell_process_slice(dev->kfd) - 1); pr_debug("PQM After DQM create queue\n"); @@ -387,7 +399,7 @@ int pqm_destroy_queue(struct process_queue_manager *pqm, unsigned int qid) struct process_queue_node *pqn; struct kfd_process_device *pdd; struct device_queue_manager *dqm; - struct kfd_dev *dev; + struct kfd_node *dev; int retval; dqm = NULL; @@ -439,7 +451,7 @@ int pqm_destroy_queue(struct process_queue_manager *pqm, unsigned int qid) pdd->qpd.num_gws = 0; } - if (dev->shared_resources.enable_mes) { + if (dev->kfd->shared_resources.enable_mes) { amdgpu_amdkfd_free_gtt_mem(dev->adev, pqn->q->gang_ctx_bo); if (pqn->q->wptr_bo) @@ -477,6 +489,7 @@ int pqm_update_queue_properties(struct process_queue_manager *pqm, pqn->q->properties.queue_size = p->queue_size; pqn->q->properties.queue_percent = p->queue_percent; pqn->q->properties.priority = p->priority; + pqn->q->properties.pm4_target_xcc = p->pm4_target_xcc; retval = pqn->q->device->dqm->ops.update_queue(pqn->q->device->dqm, pqn->q, NULL); @@ -498,8 +511,12 @@ int pqm_update_mqd(struct process_queue_manager *pqm, return -EFAULT; } + /* CUs are masked for debugger requirements so deny user mask */ + if (pqn->q->properties.is_dbg_wa && minfo && minfo->cu_mask.ptr) + return -EBUSY; + /* ASICs that have WGPs must enforce pairwise enabled mask checks. */ - if (minfo && minfo->update_flag == UPDATE_FLAG_CU_MASK && minfo->cu_mask.ptr && + if (minfo && minfo->cu_mask.ptr && KFD_GC_VERSION(pqn->q->device) >= IP_VERSION(10, 0, 0)) { int i; @@ -518,6 +535,9 @@ int pqm_update_mqd(struct process_queue_manager *pqm, if (retval != 0) return retval; + if (minfo && minfo->cu_mask.ptr) + pqn->q->properties.is_user_cu_masked = true; + return 0; } @@ -565,6 +585,46 @@ int pqm_get_wave_state(struct process_queue_manager *pqm, save_area_used_size); } +int pqm_get_queue_snapshot(struct process_queue_manager *pqm, + uint64_t exception_clear_mask, + void __user *buf, + int *num_qss_entries, + uint32_t *entry_size) +{ + struct process_queue_node *pqn; + struct kfd_queue_snapshot_entry src; + uint32_t tmp_entry_size = *entry_size, tmp_qss_entries = *num_qss_entries; + int r = 0; + + *num_qss_entries = 0; + if (!(*entry_size)) + return -EINVAL; + + *entry_size = min_t(size_t, *entry_size, sizeof(struct kfd_queue_snapshot_entry)); + mutex_lock(&pqm->process->event_mutex); + + memset(&src, 0, sizeof(src)); + + list_for_each_entry(pqn, &pqm->queues, process_queue_list) { + if (!pqn->q) + continue; + + if (*num_qss_entries < tmp_qss_entries) { + set_queue_snapshot_entry(pqn->q, exception_clear_mask, &src); + + if (copy_to_user(buf, &src, *entry_size)) { + r = -EFAULT; + break; + } + buf += tmp_entry_size; + } + *num_qss_entries += 1; + } + + mutex_unlock(&pqm->process->event_mutex); + return r; +} + static int get_queue_data_sizes(struct kfd_process_device *pdd, struct queue *q, uint32_t *mqd_size, @@ -859,7 +919,7 @@ int kfd_criu_restore_queue(struct kfd_process *p, } if (!pdd->doorbell_index && - kfd_alloc_process_doorbells(pdd->dev, &pdd->doorbell_index) < 0) { + kfd_alloc_process_doorbells(pdd->dev->kfd, &pdd->doorbell_index) < 0) { ret = -ENOMEM; goto exit; } @@ -927,7 +987,9 @@ int pqm_debugfs_mqds(struct seq_file *m, void *data) struct queue *q; enum KFD_MQD_TYPE mqd_type; struct mqd_manager *mqd_mgr; - int r = 0; + int r = 0, xcc, num_xccs = 1; + void *mqd; + uint64_t size = 0; list_for_each_entry(pqn, &pqm->queues, process_queue_list) { if (pqn->q) { @@ -943,6 +1005,7 @@ int pqm_debugfs_mqds(struct seq_file *m, void *data) seq_printf(m, " Compute queue on device %x\n", q->device->id); mqd_type = KFD_MQD_TYPE_CP; + num_xccs = NUM_XCC(q->device->xcc_mask); break; default: seq_printf(m, @@ -951,6 +1014,8 @@ int pqm_debugfs_mqds(struct seq_file *m, void *data) continue; } mqd_mgr = q->device->dqm->mqd_mgrs[mqd_type]; + size = mqd_mgr->mqd_stride(mqd_mgr, + &q->properties); } else if (pqn->kq) { q = pqn->kq->queue; mqd_mgr = pqn->kq->mqd_mgr; @@ -972,9 +1037,12 @@ int pqm_debugfs_mqds(struct seq_file *m, void *data) continue; } - r = mqd_mgr->debugfs_show_mqd(m, q->mqd); - if (r != 0) - break; + for (xcc = 0; xcc < num_xccs; xcc++) { + mqd = q->mqd + size * xcc; + r = mqd_mgr->debugfs_show_mqd(m, mqd); + if (r != 0) + break; + } } return r; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c index 0472b56de2457ee98d451ec69c6242129c94c9ed..d9953c2b2661445524c6444118f8c31822831f46 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c @@ -36,7 +36,7 @@ struct kfd_smi_client { wait_queue_head_t wait_queue; /* events enabled */ uint64_t events; - struct kfd_dev *dev; + struct kfd_node *dev; spinlock_t lock; struct rcu_head rcu; pid_t pid; @@ -149,7 +149,7 @@ static void kfd_smi_ev_client_free(struct rcu_head *p) static int kfd_smi_ev_release(struct inode *inode, struct file *filep) { struct kfd_smi_client *client = filep->private_data; - struct kfd_dev *dev = client->dev; + struct kfd_node *dev = client->dev; spin_lock(&dev->smi_lock); list_del_rcu(&client->list); @@ -171,7 +171,7 @@ static bool kfd_smi_ev_enabled(pid_t pid, struct kfd_smi_client *client, return events & KFD_SMI_EVENT_MASK_FROM_INDEX(event); } -static void add_event_to_kfifo(pid_t pid, struct kfd_dev *dev, +static void add_event_to_kfifo(pid_t pid, struct kfd_node *dev, unsigned int smi_event, char *event_msg, int len) { struct kfd_smi_client *client; @@ -196,7 +196,7 @@ static void add_event_to_kfifo(pid_t pid, struct kfd_dev *dev, } __printf(4, 5) -static void kfd_smi_event_add(pid_t pid, struct kfd_dev *dev, +static void kfd_smi_event_add(pid_t pid, struct kfd_node *dev, unsigned int event, char *fmt, ...) { char fifo_in[KFD_SMI_EVENT_MSG_SIZE]; @@ -215,7 +215,7 @@ static void kfd_smi_event_add(pid_t pid, struct kfd_dev *dev, add_event_to_kfifo(pid, dev, event, fifo_in, len); } -void kfd_smi_event_update_gpu_reset(struct kfd_dev *dev, bool post_reset) +void kfd_smi_event_update_gpu_reset(struct kfd_node *dev, bool post_reset) { unsigned int event; @@ -228,7 +228,7 @@ void kfd_smi_event_update_gpu_reset(struct kfd_dev *dev, bool post_reset) kfd_smi_event_add(0, dev, event, "%x\n", dev->reset_seq_num); } -void kfd_smi_event_update_thermal_throttling(struct kfd_dev *dev, +void kfd_smi_event_update_thermal_throttling(struct kfd_node *dev, uint64_t throttle_bitmask) { kfd_smi_event_add(0, dev, KFD_SMI_EVENT_THERMAL_THROTTLE, "%llx:%llx\n", @@ -236,7 +236,7 @@ void kfd_smi_event_update_thermal_throttling(struct kfd_dev *dev, amdgpu_dpm_get_thermal_throttling_counter(dev->adev)); } -void kfd_smi_event_update_vmfault(struct kfd_dev *dev, uint16_t pasid) +void kfd_smi_event_update_vmfault(struct kfd_node *dev, uint16_t pasid) { struct amdgpu_task_info task_info; @@ -250,58 +250,58 @@ void kfd_smi_event_update_vmfault(struct kfd_dev *dev, uint16_t pasid) task_info.pid, task_info.task_name); } -void kfd_smi_event_page_fault_start(struct kfd_dev *dev, pid_t pid, +void kfd_smi_event_page_fault_start(struct kfd_node *node, pid_t pid, unsigned long address, bool write_fault, ktime_t ts) { - kfd_smi_event_add(pid, dev, KFD_SMI_EVENT_PAGE_FAULT_START, + kfd_smi_event_add(pid, node, KFD_SMI_EVENT_PAGE_FAULT_START, "%lld -%d @%lx(%x) %c\n", ktime_to_ns(ts), pid, - address, dev->id, write_fault ? 'W' : 'R'); + address, node->id, write_fault ? 'W' : 'R'); } -void kfd_smi_event_page_fault_end(struct kfd_dev *dev, pid_t pid, +void kfd_smi_event_page_fault_end(struct kfd_node *node, pid_t pid, unsigned long address, bool migration) { - kfd_smi_event_add(pid, dev, KFD_SMI_EVENT_PAGE_FAULT_END, + kfd_smi_event_add(pid, node, KFD_SMI_EVENT_PAGE_FAULT_END, "%lld -%d @%lx(%x) %c\n", ktime_get_boottime_ns(), - pid, address, dev->id, migration ? 'M' : 'U'); + pid, address, node->id, migration ? 'M' : 'U'); } -void kfd_smi_event_migration_start(struct kfd_dev *dev, pid_t pid, +void kfd_smi_event_migration_start(struct kfd_node *node, pid_t pid, unsigned long start, unsigned long end, uint32_t from, uint32_t to, uint32_t prefetch_loc, uint32_t preferred_loc, uint32_t trigger) { - kfd_smi_event_add(pid, dev, KFD_SMI_EVENT_MIGRATE_START, + kfd_smi_event_add(pid, node, KFD_SMI_EVENT_MIGRATE_START, "%lld -%d @%lx(%lx) %x->%x %x:%x %d\n", ktime_get_boottime_ns(), pid, start, end - start, from, to, prefetch_loc, preferred_loc, trigger); } -void kfd_smi_event_migration_end(struct kfd_dev *dev, pid_t pid, +void kfd_smi_event_migration_end(struct kfd_node *node, pid_t pid, unsigned long start, unsigned long end, uint32_t from, uint32_t to, uint32_t trigger) { - kfd_smi_event_add(pid, dev, KFD_SMI_EVENT_MIGRATE_END, + kfd_smi_event_add(pid, node, KFD_SMI_EVENT_MIGRATE_END, "%lld -%d @%lx(%lx) %x->%x %d\n", ktime_get_boottime_ns(), pid, start, end - start, from, to, trigger); } -void kfd_smi_event_queue_eviction(struct kfd_dev *dev, pid_t pid, +void kfd_smi_event_queue_eviction(struct kfd_node *node, pid_t pid, uint32_t trigger) { - kfd_smi_event_add(pid, dev, KFD_SMI_EVENT_QUEUE_EVICTION, + kfd_smi_event_add(pid, node, KFD_SMI_EVENT_QUEUE_EVICTION, "%lld -%d %x %d\n", ktime_get_boottime_ns(), pid, - dev->id, trigger); + node->id, trigger); } -void kfd_smi_event_queue_restore(struct kfd_dev *dev, pid_t pid) +void kfd_smi_event_queue_restore(struct kfd_node *node, pid_t pid) { - kfd_smi_event_add(pid, dev, KFD_SMI_EVENT_QUEUE_RESTORE, + kfd_smi_event_add(pid, node, KFD_SMI_EVENT_QUEUE_RESTORE, "%lld -%d %x\n", ktime_get_boottime_ns(), pid, - dev->id); + node->id); } void kfd_smi_event_queue_restore_rescheduled(struct mm_struct *mm) @@ -324,16 +324,16 @@ void kfd_smi_event_queue_restore_rescheduled(struct mm_struct *mm) kfd_unref_process(p); } -void kfd_smi_event_unmap_from_gpu(struct kfd_dev *dev, pid_t pid, +void kfd_smi_event_unmap_from_gpu(struct kfd_node *node, pid_t pid, unsigned long address, unsigned long last, uint32_t trigger) { - kfd_smi_event_add(pid, dev, KFD_SMI_EVENT_UNMAP_FROM_GPU, + kfd_smi_event_add(pid, node, KFD_SMI_EVENT_UNMAP_FROM_GPU, "%lld -%d @%lx(%lx) %x %d\n", ktime_get_boottime_ns(), - pid, address, last - address + 1, dev->id, trigger); + pid, address, last - address + 1, node->id, trigger); } -int kfd_smi_event_open(struct kfd_dev *dev, uint32_t *fd) +int kfd_smi_event_open(struct kfd_node *dev, uint32_t *fd) { struct kfd_smi_client *client; int ret; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h index 76fe4e0ec2d2379e8e946adafc5d50e7c3d9c64d..fa95c2dfd587ffaf3b8cd536dca2e4f7615fb8d9 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h @@ -24,29 +24,29 @@ #ifndef KFD_SMI_EVENTS_H_INCLUDED #define KFD_SMI_EVENTS_H_INCLUDED -int kfd_smi_event_open(struct kfd_dev *dev, uint32_t *fd); -void kfd_smi_event_update_vmfault(struct kfd_dev *dev, uint16_t pasid); -void kfd_smi_event_update_thermal_throttling(struct kfd_dev *dev, +int kfd_smi_event_open(struct kfd_node *dev, uint32_t *fd); +void kfd_smi_event_update_vmfault(struct kfd_node *dev, uint16_t pasid); +void kfd_smi_event_update_thermal_throttling(struct kfd_node *dev, uint64_t throttle_bitmask); -void kfd_smi_event_update_gpu_reset(struct kfd_dev *dev, bool post_reset); -void kfd_smi_event_page_fault_start(struct kfd_dev *dev, pid_t pid, +void kfd_smi_event_update_gpu_reset(struct kfd_node *dev, bool post_reset); +void kfd_smi_event_page_fault_start(struct kfd_node *node, pid_t pid, unsigned long address, bool write_fault, ktime_t ts); -void kfd_smi_event_page_fault_end(struct kfd_dev *dev, pid_t pid, +void kfd_smi_event_page_fault_end(struct kfd_node *node, pid_t pid, unsigned long address, bool migration); -void kfd_smi_event_migration_start(struct kfd_dev *dev, pid_t pid, +void kfd_smi_event_migration_start(struct kfd_node *node, pid_t pid, unsigned long start, unsigned long end, uint32_t from, uint32_t to, uint32_t prefetch_loc, uint32_t preferred_loc, uint32_t trigger); -void kfd_smi_event_migration_end(struct kfd_dev *dev, pid_t pid, +void kfd_smi_event_migration_end(struct kfd_node *node, pid_t pid, unsigned long start, unsigned long end, uint32_t from, uint32_t to, uint32_t trigger); -void kfd_smi_event_queue_eviction(struct kfd_dev *dev, pid_t pid, +void kfd_smi_event_queue_eviction(struct kfd_node *node, pid_t pid, uint32_t trigger); -void kfd_smi_event_queue_restore(struct kfd_dev *dev, pid_t pid); +void kfd_smi_event_queue_restore(struct kfd_node *node, pid_t pid); void kfd_smi_event_queue_restore_rescheduled(struct mm_struct *mm); -void kfd_smi_event_unmap_from_gpu(struct kfd_dev *dev, pid_t pid, +void kfd_smi_event_unmap_from_gpu(struct kfd_node *node, pid_t pid, unsigned long address, unsigned long last, uint32_t trigger); #endif diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c index 96a138a395150247a56bf5c111d41fb029247869..5ff1a5a89d96818d643e625b9ab594b8a742db32 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c @@ -170,12 +170,11 @@ svm_range_dma_map_dev(struct amdgpu_device *adev, struct svm_range *prange, page = hmm_pfn_to_page(hmm_pfns[i]); if (is_zone_device_page(page)) { - struct amdgpu_device *bo_adev = - amdgpu_ttm_adev(prange->svm_bo->bo->tbo.bdev); + struct amdgpu_device *bo_adev = prange->svm_bo->node->adev; addr[i] = (hmm_pfns[i] << PAGE_SHIFT) + bo_adev->vm_manager.vram_base_offset - - bo_adev->kfd.dev->pgmap.range.start; + bo_adev->kfd.pgmap.range.start; addr[i] |= SVM_RANGE_VRAM_DOMAIN; pr_debug_ratelimited("vram address: 0x%llx\n", addr[i]); continue; @@ -281,7 +280,7 @@ static void svm_range_free(struct svm_range *prange, bool update_mem_usage) if (update_mem_usage && !p->xnack_enabled) { pr_debug("unreserve prange 0x%p size: 0x%llx\n", prange, size); amdgpu_amdkfd_unreserve_mem_limit(NULL, size, - KFD_IOC_ALLOC_MEM_FLAGS_USERPTR); + KFD_IOC_ALLOC_MEM_FLAGS_USERPTR, 0); } mutex_destroy(&prange->lock); mutex_destroy(&prange->migrate_mutex); @@ -314,7 +313,7 @@ svm_range *svm_range_new(struct svm_range_list *svms, uint64_t start, p = container_of(svms, struct kfd_process, svms); if (!p->xnack_enabled && update_mem_usage && amdgpu_amdkfd_reserve_mem_limit(NULL, size << PAGE_SHIFT, - KFD_IOC_ALLOC_MEM_FLAGS_USERPTR)) { + KFD_IOC_ALLOC_MEM_FLAGS_USERPTR, 0)) { pr_info("SVM mapping failed, exceeds resident system memory limit\n"); kfree(prange); return NULL; @@ -424,10 +423,8 @@ static void svm_range_bo_unref(struct svm_range_bo *svm_bo) } static bool -svm_range_validate_svm_bo(struct amdgpu_device *adev, struct svm_range *prange) +svm_range_validate_svm_bo(struct kfd_node *node, struct svm_range *prange) { - struct amdgpu_device *bo_adev; - mutex_lock(&prange->lock); if (!prange->svm_bo) { mutex_unlock(&prange->lock); @@ -440,12 +437,11 @@ svm_range_validate_svm_bo(struct amdgpu_device *adev, struct svm_range *prange) } if (svm_bo_ref_unless_zero(prange->svm_bo)) { /* - * Migrate from GPU to GPU, remove range from source bo_adev - * svm_bo range list, and return false to allocate svm_bo from - * destination adev. + * Migrate from GPU to GPU, remove range from source svm_bo->node + * range list, and return false to allocate svm_bo from destination + * node. */ - bo_adev = amdgpu_ttm_adev(prange->svm_bo->bo->tbo.bdev); - if (bo_adev != adev) { + if (prange->svm_bo->node != node) { mutex_unlock(&prange->lock); spin_lock(&prange->svm_bo->list_lock); @@ -513,7 +509,7 @@ static struct svm_range_bo *svm_range_bo_new(void) } int -svm_range_vram_node_new(struct amdgpu_device *adev, struct svm_range *prange, +svm_range_vram_node_new(struct kfd_node *node, struct svm_range *prange, bool clear) { struct amdgpu_bo_param bp; @@ -528,7 +524,7 @@ svm_range_vram_node_new(struct amdgpu_device *adev, struct svm_range *prange, pr_debug("pasid: %x svms 0x%p [0x%lx 0x%lx]\n", p->pasid, prange->svms, prange->start, prange->last); - if (svm_range_validate_svm_bo(adev, prange)) + if (svm_range_validate_svm_bo(node, prange)) return 0; svm_bo = svm_range_bo_new(); @@ -542,6 +538,7 @@ svm_range_vram_node_new(struct amdgpu_device *adev, struct svm_range *prange, kfree(svm_bo); return -ESRCH; } + svm_bo->node = node; svm_bo->eviction_fence = amdgpu_amdkfd_fence_create(dma_fence_context_alloc(1), mm, @@ -558,13 +555,20 @@ svm_range_vram_node_new(struct amdgpu_device *adev, struct svm_range *prange, bp.flags |= AMDGPU_GEM_CREATE_DISCARDABLE; bp.type = ttm_bo_type_device; bp.resv = NULL; + if (node->xcp) + bp.xcp_id_plus1 = node->xcp->id + 1; - r = amdgpu_bo_create_user(adev, &bp, &ubo); + r = amdgpu_bo_create_user(node->adev, &bp, &ubo); if (r) { pr_debug("failed %d to create bo\n", r); goto create_bo_failed; } bo = &ubo->bo; + + pr_debug("alloc bo at offset 0x%lx size 0x%lx on partition %d\n", + bo->tbo.resource->start << PAGE_SHIFT, bp.size, + bp.xcp_id_plus1 - 1); + r = amdgpu_bo_reserve(bo, true); if (r) { pr_debug("failed %d to reserve bo\n", r); @@ -617,45 +621,30 @@ void svm_range_vram_node_free(struct svm_range *prange) prange->ttm_res = NULL; } -struct amdgpu_device * -svm_range_get_adev_by_id(struct svm_range *prange, uint32_t gpu_id) +struct kfd_node * +svm_range_get_node_by_id(struct svm_range *prange, uint32_t gpu_id) { - struct kfd_process_device *pdd; struct kfd_process *p; - int32_t gpu_idx; + struct kfd_process_device *pdd; p = container_of(prange->svms, struct kfd_process, svms); - - gpu_idx = kfd_process_gpuidx_from_gpuid(p, gpu_id); - if (gpu_idx < 0) { - pr_debug("failed to get device by id 0x%x\n", gpu_id); - return NULL; - } - pdd = kfd_process_device_from_gpuidx(p, gpu_idx); + pdd = kfd_process_device_data_by_id(p, gpu_id); if (!pdd) { - pr_debug("failed to get device by idx 0x%x\n", gpu_idx); + pr_debug("failed to get kfd process device by id 0x%x\n", gpu_id); return NULL; } - return pdd->dev->adev; + return pdd->dev; } struct kfd_process_device * -svm_range_get_pdd_by_adev(struct svm_range *prange, struct amdgpu_device *adev) +svm_range_get_pdd_by_node(struct svm_range *prange, struct kfd_node *node) { struct kfd_process *p; - int32_t gpu_idx, gpuid; - int r; p = container_of(prange->svms, struct kfd_process, svms); - r = kfd_process_gpuid_from_adev(p, adev, &gpuid, &gpu_idx); - if (r) { - pr_debug("failed to get device id by adev %p\n", adev); - return NULL; - } - - return kfd_process_device_from_gpuidx(p, gpu_idx); + return kfd_get_process_device_data(node, p); } static int svm_range_bo_validate(void *param, struct amdgpu_bo *bo) @@ -735,7 +724,9 @@ svm_range_apply_attrs(struct kfd_process *p, struct svm_range *prange, case KFD_IOCTL_SVM_ATTR_ACCESS: case KFD_IOCTL_SVM_ATTR_ACCESS_IN_PLACE: case KFD_IOCTL_SVM_ATTR_NO_ACCESS: - *update_mapping = true; + if (!p->xnack_enabled) + *update_mapping = true; + gpuidx = kfd_process_gpuidx_from_gpuid(p, attrs[i].value); if (attrs[i].type == KFD_IOCTL_SVM_ATTR_NO_ACCESS) { @@ -818,7 +809,7 @@ svm_range_is_same_attrs(struct kfd_process *p, struct svm_range *prange, } } - return true; + return !prange->is_error_flag; } /** @@ -1146,31 +1137,39 @@ svm_range_split_by_granularity(struct kfd_process *p, struct mm_struct *mm, } return 0; } +static bool +svm_nodes_in_same_hive(struct kfd_node *node_a, struct kfd_node *node_b) +{ + return (node_a->adev == node_b->adev || + amdgpu_xgmi_same_hive(node_a->adev, node_b->adev)); +} static uint64_t -svm_range_get_pte_flags(struct amdgpu_device *adev, struct svm_range *prange, - int domain) +svm_range_get_pte_flags(struct kfd_node *node, + struct svm_range *prange, int domain) { - struct amdgpu_device *bo_adev; + struct kfd_node *bo_node; uint32_t flags = prange->flags; uint32_t mapping_flags = 0; uint64_t pte_flags; bool snoop = (domain != SVM_RANGE_VRAM_DOMAIN); bool coherent = flags & KFD_IOCTL_SVM_FLAG_COHERENT; + bool uncached = false; /*flags & KFD_IOCTL_SVM_FLAG_UNCACHED;*/ + unsigned int mtype_local; if (domain == SVM_RANGE_VRAM_DOMAIN) - bo_adev = amdgpu_ttm_adev(prange->svm_bo->bo->tbo.bdev); + bo_node = prange->svm_bo->node; - switch (KFD_GC_VERSION(adev->kfd.dev)) { + switch (node->adev->ip_versions[GC_HWIP][0]) { case IP_VERSION(9, 4, 1): if (domain == SVM_RANGE_VRAM_DOMAIN) { - if (bo_adev == adev) { + if (bo_node == node) { mapping_flags |= coherent ? AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_RW; } else { mapping_flags |= coherent ? AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC; - if (amdgpu_xgmi_same_hive(adev, bo_adev)) + if (svm_nodes_in_same_hive(node, bo_node)) snoop = true; } } else { @@ -1180,15 +1179,15 @@ svm_range_get_pte_flags(struct amdgpu_device *adev, struct svm_range *prange, break; case IP_VERSION(9, 4, 2): if (domain == SVM_RANGE_VRAM_DOMAIN) { - if (bo_adev == adev) { + if (bo_node == node) { mapping_flags |= coherent ? AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_RW; - if (adev->gmc.xgmi.connected_to_cpu) + if (node->adev->gmc.xgmi.connected_to_cpu) snoop = true; } else { mapping_flags |= coherent ? AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC; - if (amdgpu_xgmi_same_hive(adev, bo_adev)) + if (svm_nodes_in_same_hive(node, bo_node)) snoop = true; } } else { @@ -1196,6 +1195,37 @@ svm_range_get_pte_flags(struct amdgpu_device *adev, struct svm_range *prange, AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC; } break; + case IP_VERSION(9, 4, 3): + mtype_local = amdgpu_mtype_local == 1 ? AMDGPU_VM_MTYPE_NC : + (amdgpu_mtype_local == 2 ? AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_RW); + snoop = true; + if (uncached) { + mapping_flags |= AMDGPU_VM_MTYPE_UC; + } else if (domain == SVM_RANGE_VRAM_DOMAIN) { + /* local HBM region close to partition */ + if (bo_node->adev == node->adev && + (!bo_node->xcp || !node->xcp || bo_node->xcp->mem_id == node->xcp->mem_id)) + mapping_flags |= mtype_local; + /* local HBM region far from partition or remote XGMI GPU */ + else if (svm_nodes_in_same_hive(bo_node, node)) + mapping_flags |= AMDGPU_VM_MTYPE_NC; + /* PCIe P2P */ + else + mapping_flags |= AMDGPU_VM_MTYPE_UC; + /* system memory accessed by the APU */ + } else if (node->adev->flags & AMD_IS_APU) { + /* On NUMA systems, locality is determined per-page + * in amdgpu_gmc_override_vm_pte_flags + */ + if (num_possible_nodes() <= 1) + mapping_flags |= mtype_local; + else + mapping_flags |= AMDGPU_VM_MTYPE_NC; + /* system memory accessed by the dGPU */ + } else { + mapping_flags |= AMDGPU_VM_MTYPE_UC; + } + break; default: mapping_flags |= coherent ? AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC; @@ -1212,7 +1242,7 @@ svm_range_get_pte_flags(struct amdgpu_device *adev, struct svm_range *prange, pte_flags |= (domain == SVM_RANGE_VRAM_DOMAIN) ? 0 : AMDGPU_PTE_SYSTEM; pte_flags |= snoop ? AMDGPU_PTE_SNOOPED : 0; - pte_flags |= amdgpu_gem_va_map_flags(adev, mapping_flags); + pte_flags |= amdgpu_gem_va_map_flags(node->adev, mapping_flags); return pte_flags; } @@ -1319,7 +1349,7 @@ svm_range_map_to_gpu(struct kfd_process_device *pdd, struct svm_range *prange, pr_debug("Mapping range [0x%lx 0x%llx] on domain: %s\n", last_start, prange->start + i, last_domain ? "GPU" : "CPU"); - pte_flags = svm_range_get_pte_flags(adev, prange, last_domain); + pte_flags = svm_range_get_pte_flags(pdd->dev, prange, last_domain); if (readonly) pte_flags &= ~AMDGPU_PTE_WRITEABLE; @@ -1328,6 +1358,10 @@ svm_range_map_to_gpu(struct kfd_process_device *pdd, struct svm_range *prange, (last_domain == SVM_RANGE_VRAM_DOMAIN) ? 1 : 0, pte_flags); + /* For dGPU mode, we use same vm_manager to allocate VRAM for + * different memory partition based on fpfn/lpfn, we should use + * same vm_manager.vram_base_offset regardless memory partition. + */ r = amdgpu_vm_update_range(adev, vm, false, false, flush_tlb, NULL, last_start, prange->start + i, pte_flags, @@ -1365,16 +1399,14 @@ svm_range_map_to_gpus(struct svm_range *prange, unsigned long offset, unsigned long *bitmap, bool wait, bool flush_tlb) { struct kfd_process_device *pdd; - struct amdgpu_device *bo_adev; + struct amdgpu_device *bo_adev = NULL; struct kfd_process *p; struct dma_fence *fence = NULL; uint32_t gpuidx; int r = 0; if (prange->svm_bo && prange->ttm_res) - bo_adev = amdgpu_ttm_adev(prange->svm_bo->bo->tbo.bdev); - else - bo_adev = NULL; + bo_adev = prange->svm_bo->node->adev; p = container_of(prange->svms, struct kfd_process, svms); for_each_set_bit(gpuidx, bitmap, MAX_GPU_INSTANCE) { @@ -1522,48 +1554,54 @@ static int svm_range_validate_and_map(struct mm_struct *mm, struct svm_range *prange, int32_t gpuidx, bool intr, bool wait, bool flush_tlb) { - struct svm_validate_context ctx; + struct svm_validate_context *ctx; unsigned long start, end, addr; struct kfd_process *p; void *owner; int32_t idx; int r = 0; - ctx.process = container_of(prange->svms, struct kfd_process, svms); - ctx.prange = prange; - ctx.intr = intr; + ctx = kzalloc(sizeof(struct svm_validate_context), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + ctx->process = container_of(prange->svms, struct kfd_process, svms); + ctx->prange = prange; + ctx->intr = intr; if (gpuidx < MAX_GPU_INSTANCE) { - bitmap_zero(ctx.bitmap, MAX_GPU_INSTANCE); - bitmap_set(ctx.bitmap, gpuidx, 1); - } else if (ctx.process->xnack_enabled) { - bitmap_copy(ctx.bitmap, prange->bitmap_aip, MAX_GPU_INSTANCE); + bitmap_zero(ctx->bitmap, MAX_GPU_INSTANCE); + bitmap_set(ctx->bitmap, gpuidx, 1); + } else if (ctx->process->xnack_enabled) { + bitmap_copy(ctx->bitmap, prange->bitmap_aip, MAX_GPU_INSTANCE); /* If prefetch range to GPU, or GPU retry fault migrate range to * GPU, which has ACCESS attribute to the range, create mapping * on that GPU. */ if (prange->actual_loc) { - gpuidx = kfd_process_gpuidx_from_gpuid(ctx.process, + gpuidx = kfd_process_gpuidx_from_gpuid(ctx->process, prange->actual_loc); if (gpuidx < 0) { WARN_ONCE(1, "failed get device by id 0x%x\n", prange->actual_loc); - return -EINVAL; + r = -EINVAL; + goto free_ctx; } if (test_bit(gpuidx, prange->bitmap_access)) - bitmap_set(ctx.bitmap, gpuidx, 1); + bitmap_set(ctx->bitmap, gpuidx, 1); } } else { - bitmap_or(ctx.bitmap, prange->bitmap_access, + bitmap_or(ctx->bitmap, prange->bitmap_access, prange->bitmap_aip, MAX_GPU_INSTANCE); } - if (bitmap_empty(ctx.bitmap, MAX_GPU_INSTANCE)) { - if (!prange->mapped_to_gpu) - return 0; + if (bitmap_empty(ctx->bitmap, MAX_GPU_INSTANCE)) { + if (!prange->mapped_to_gpu) { + r = 0; + goto free_ctx; + } - bitmap_copy(ctx.bitmap, prange->bitmap_access, MAX_GPU_INSTANCE); + bitmap_copy(ctx->bitmap, prange->bitmap_access, MAX_GPU_INSTANCE); } if (prange->actual_loc && !prange->ttm_res) { @@ -1571,15 +1609,16 @@ static int svm_range_validate_and_map(struct mm_struct *mm, * svm_migrate_ram_to_vram after allocating a BO. */ WARN_ONCE(1, "VRAM BO missing during validation\n"); - return -EINVAL; + r = -EINVAL; + goto free_ctx; } - svm_range_reserve_bos(&ctx); + svm_range_reserve_bos(ctx); p = container_of(prange->svms, struct kfd_process, svms); - owner = kfd_svm_page_owner(p, find_first_bit(ctx.bitmap, + owner = kfd_svm_page_owner(p, find_first_bit(ctx->bitmap, MAX_GPU_INSTANCE)); - for_each_set_bit(idx, ctx.bitmap, MAX_GPU_INSTANCE) { + for_each_set_bit(idx, ctx->bitmap, MAX_GPU_INSTANCE) { if (kfd_svm_page_owner(p, idx) != owner) { owner = NULL; break; @@ -1616,7 +1655,7 @@ static int svm_range_validate_and_map(struct mm_struct *mm, } offset = (addr - start) >> PAGE_SHIFT; - r = svm_range_dma_map(prange, ctx.bitmap, offset, npages, + r = svm_range_dma_map(prange, ctx->bitmap, offset, npages, hmm_range->hmm_pfns); if (r) { pr_debug("failed %d to dma map range\n", r); @@ -1636,7 +1675,7 @@ static int svm_range_validate_and_map(struct mm_struct *mm, } r = svm_range_map_to_gpus(prange, offset, npages, readonly, - ctx.bitmap, wait, flush_tlb); + ctx->bitmap, wait, flush_tlb); unlock_out: svm_range_unlock(prange); @@ -1650,11 +1689,15 @@ static int svm_range_validate_and_map(struct mm_struct *mm, } unreserve_out: - svm_range_unreserve_bos(&ctx); + svm_range_unreserve_bos(ctx); + prange->is_error_flag = !!r; if (!r) prange->validate_timestamp = ktime_get_boottime(); +free_ctx: + kfree(ctx); + return r; } @@ -1783,6 +1826,7 @@ static void svm_range_restore_work(struct work_struct *work) * @mm: current process mm_struct * @start: starting process queue number * @last: last process queue number + * @event: mmu notifier event when range is evicted or migrated * * Stop all queues of the process to ensure GPU doesn't access the memory, then * return to let CPU evict the buffer and proceed CPU pagetable update. @@ -1906,14 +1950,23 @@ void svm_range_set_max_pages(struct amdgpu_device *adev) { uint64_t max_pages; uint64_t pages, _pages; + uint64_t min_pages = 0; + int i, id; + + for (i = 0; i < adev->kfd.dev->num_nodes; i++) { + if (adev->kfd.dev->nodes[i]->xcp) + id = adev->kfd.dev->nodes[i]->xcp->id; + else + id = -1; + pages = KFD_XCP_MEMORY_SIZE(adev, id) >> 17; + pages = clamp(pages, 1ULL << 9, 1ULL << 18); + pages = rounddown_pow_of_two(pages); + min_pages = min_not_zero(min_pages, pages); + } - /* 1/32 VRAM size in pages */ - pages = adev->gmc.real_vram_size >> 17; - pages = clamp(pages, 1ULL << 9, 1ULL << 18); - pages = rounddown_pow_of_two(pages); do { max_pages = READ_ONCE(max_svm_range_pages); - _pages = min_not_zero(max_pages, pages); + _pages = min_not_zero(max_pages, min_pages); } while (cmpxchg(&max_svm_range_pages, max_pages, _pages) != max_pages); } @@ -2507,29 +2560,31 @@ svm_range_from_addr(struct svm_range_list *svms, unsigned long addr, */ static int32_t svm_range_best_restore_location(struct svm_range *prange, - struct amdgpu_device *adev, + struct kfd_node *node, int32_t *gpuidx) { - struct amdgpu_device *bo_adev, *preferred_adev; + struct kfd_node *bo_node, *preferred_node; struct kfd_process *p; uint32_t gpuid; int r; p = container_of(prange->svms, struct kfd_process, svms); - r = kfd_process_gpuid_from_adev(p, adev, &gpuid, gpuidx); + r = kfd_process_gpuid_from_node(p, node, &gpuid, gpuidx); if (r < 0) { pr_debug("failed to get gpuid from kgd\n"); return -1; } + if (node->adev->gmc.is_app_apu) + return 0; + if (prange->preferred_loc == gpuid || prange->preferred_loc == KFD_IOCTL_SVM_LOCATION_SYSMEM) { return prange->preferred_loc; } else if (prange->preferred_loc != KFD_IOCTL_SVM_LOCATION_UNDEFINED) { - preferred_adev = svm_range_get_adev_by_id(prange, - prange->preferred_loc); - if (amdgpu_xgmi_same_hive(adev, preferred_adev)) + preferred_node = svm_range_get_node_by_id(prange, prange->preferred_loc); + if (preferred_node && svm_nodes_in_same_hive(node, preferred_node)) return prange->preferred_loc; /* fall through */ } @@ -2541,8 +2596,8 @@ svm_range_best_restore_location(struct svm_range *prange, if (!prange->actual_loc) return 0; - bo_adev = svm_range_get_adev_by_id(prange, prange->actual_loc); - if (amdgpu_xgmi_same_hive(adev, bo_adev)) + bo_node = svm_range_get_node_by_id(prange, prange->actual_loc); + if (bo_node && svm_nodes_in_same_hive(node, bo_node)) return prange->actual_loc; else return 0; @@ -2659,7 +2714,7 @@ svm_range_check_vm_userptr(struct kfd_process *p, uint64_t start, uint64_t last, } static struct -svm_range *svm_range_create_unregistered_range(struct amdgpu_device *adev, +svm_range *svm_range_create_unregistered_range(struct kfd_node *node, struct kfd_process *p, struct mm_struct *mm, int64_t addr) @@ -2694,7 +2749,7 @@ svm_range *svm_range_create_unregistered_range(struct amdgpu_device *adev, pr_debug("Failed to create prange in address [0x%llx]\n", addr); return NULL; } - if (kfd_process_gpuid_from_adev(p, adev, &gpuid, &gpuidx)) { + if (kfd_process_gpuid_from_node(p, node, &gpuid, &gpuidx)) { pr_debug("failed to get gpuid from kgd\n"); svm_range_free(prange, true); return NULL; @@ -2748,7 +2803,7 @@ static bool svm_range_skip_recover(struct svm_range *prange) } static void -svm_range_count_fault(struct amdgpu_device *adev, struct kfd_process *p, +svm_range_count_fault(struct kfd_node *node, struct kfd_process *p, int32_t gpuidx) { struct kfd_process_device *pdd; @@ -2761,7 +2816,7 @@ svm_range_count_fault(struct amdgpu_device *adev, struct kfd_process *p, uint32_t gpuid; int r; - r = kfd_process_gpuid_from_adev(p, adev, &gpuid, &gpuidx); + r = kfd_process_gpuid_from_node(p, node, &gpuid, &gpuidx); if (r < 0) return; } @@ -2789,6 +2844,7 @@ svm_fault_allowed(struct vm_area_struct *vma, bool write_fault) int svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid, + uint32_t vmid, uint32_t node_id, uint64_t addr, bool write_fault) { struct mm_struct *mm = NULL; @@ -2796,6 +2852,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid, struct svm_range *prange; struct kfd_process *p; ktime_t timestamp = ktime_get_boottime(); + struct kfd_node *node; int32_t best_loc; int32_t gpuidx = MAX_GPU_INSTANCE; bool write_locked = false; @@ -2803,7 +2860,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid, bool migration = false; int r = 0; - if (!KFD_IS_SVM_API_SUPPORTED(adev->kfd.dev)) { + if (!KFD_IS_SVM_API_SUPPORTED(adev)) { pr_debug("device does not support SVM\n"); return -EFAULT; } @@ -2839,6 +2896,13 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid, goto out; } + node = kfd_node_by_irq_ids(adev, node_id, vmid); + if (!node) { + pr_debug("kfd node does not exist node_id: %d, vmid: %d\n", node_id, + vmid); + r = -EFAULT; + goto out; + } mmap_read_lock(mm); retry_write_locked: mutex_lock(&svms->lock); @@ -2857,7 +2921,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid, write_locked = true; goto retry_write_locked; } - prange = svm_range_create_unregistered_range(adev, p, mm, addr); + prange = svm_range_create_unregistered_range(node, p, mm, addr); if (!prange) { pr_debug("failed to create unregistered range svms 0x%p address [0x%llx]\n", svms, addr); @@ -2872,7 +2936,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid, mutex_lock(&prange->migrate_mutex); if (svm_range_skip_recover(prange)) { - amdgpu_gmc_filter_faults_remove(adev, addr, pasid); + amdgpu_gmc_filter_faults_remove(node->adev, addr, pasid); r = 0; goto out_unlock_range; } @@ -2903,7 +2967,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid, goto out_unlock_range; } - best_loc = svm_range_best_restore_location(prange, adev, &gpuidx); + best_loc = svm_range_best_restore_location(prange, node, &gpuidx); if (best_loc == -1) { pr_debug("svms %p failed get best restore loc [0x%lx 0x%lx]\n", svms, prange->start, prange->last); @@ -2915,7 +2979,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid, svms, prange->start, prange->last, best_loc, prange->actual_loc); - kfd_smi_event_page_fault_start(adev->kfd.dev, p->lead_thread->pid, addr, + kfd_smi_event_page_fault_start(node, p->lead_thread->pid, addr, write_fault, timestamp); if (prange->actual_loc != best_loc) { @@ -2953,7 +3017,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid, pr_debug("failed %d to map svms 0x%p [0x%lx 0x%lx] to gpus\n", r, svms, prange->start, prange->last); - kfd_smi_event_page_fault_end(adev->kfd.dev, p->lead_thread->pid, addr, + kfd_smi_event_page_fault_end(node, p->lead_thread->pid, addr, migration); out_unlock_range: @@ -2962,7 +3026,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid, mutex_unlock(&svms->lock); mmap_read_unlock(mm); - svm_range_count_fault(adev, p, gpuidx); + svm_range_count_fault(node, p, gpuidx); mmput(mm); out: @@ -2970,7 +3034,7 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid, if (r == -EAGAIN) { pr_debug("recover vm fault later\n"); - amdgpu_gmc_filter_faults_remove(adev, addr, pasid); + amdgpu_gmc_filter_faults_remove(node->adev, addr, pasid); r = 0; } return r; @@ -2994,10 +3058,10 @@ svm_range_switch_xnack_reserve_mem(struct kfd_process *p, bool xnack_enabled) size = (pchild->last - pchild->start + 1) << PAGE_SHIFT; if (xnack_enabled) { amdgpu_amdkfd_unreserve_mem_limit(NULL, size, - KFD_IOC_ALLOC_MEM_FLAGS_USERPTR); + KFD_IOC_ALLOC_MEM_FLAGS_USERPTR, 0); } else { r = amdgpu_amdkfd_reserve_mem_limit(NULL, size, - KFD_IOC_ALLOC_MEM_FLAGS_USERPTR); + KFD_IOC_ALLOC_MEM_FLAGS_USERPTR, 0); if (r) goto out_unlock; reserved_size += size; @@ -3007,10 +3071,10 @@ svm_range_switch_xnack_reserve_mem(struct kfd_process *p, bool xnack_enabled) size = (prange->last - prange->start + 1) << PAGE_SHIFT; if (xnack_enabled) { amdgpu_amdkfd_unreserve_mem_limit(NULL, size, - KFD_IOC_ALLOC_MEM_FLAGS_USERPTR); + KFD_IOC_ALLOC_MEM_FLAGS_USERPTR, 0); } else { r = amdgpu_amdkfd_reserve_mem_limit(NULL, size, - KFD_IOC_ALLOC_MEM_FLAGS_USERPTR); + KFD_IOC_ALLOC_MEM_FLAGS_USERPTR, 0); if (r) goto out_unlock; reserved_size += size; @@ -3023,7 +3087,7 @@ svm_range_switch_xnack_reserve_mem(struct kfd_process *p, bool xnack_enabled) if (r) amdgpu_amdkfd_unreserve_mem_limit(NULL, reserved_size, - KFD_IOC_ALLOC_MEM_FLAGS_USERPTR); + KFD_IOC_ALLOC_MEM_FLAGS_USERPTR, 0); else /* Change xnack mode must be inside svms lock, to avoid race with * svm_range_deferred_list_work unreserve memory in parallel. @@ -3081,7 +3145,7 @@ int svm_range_list_init(struct kfd_process *p) spin_lock_init(&svms->deferred_list_lock); for (i = 0; i < p->n_pdds; i++) - if (KFD_IS_SVM_API_SUPPORTED(p->pdds[i]->dev)) + if (KFD_IS_SVM_API_SUPPORTED(p->pdds[i]->dev->adev)) bitmap_set(svms->bitmap_supported, i, 1); return 0; @@ -3212,7 +3276,7 @@ svm_range_best_prefetch_location(struct svm_range *prange) DECLARE_BITMAP(bitmap, MAX_GPU_INSTANCE); uint32_t best_loc = prange->prefetch_loc; struct kfd_process_device *pdd; - struct amdgpu_device *bo_adev; + struct kfd_node *bo_node; struct kfd_process *p; uint32_t gpuidx; @@ -3221,9 +3285,14 @@ svm_range_best_prefetch_location(struct svm_range *prange) if (!best_loc || best_loc == KFD_IOCTL_SVM_LOCATION_UNDEFINED) goto out; - bo_adev = svm_range_get_adev_by_id(prange, best_loc); - if (!bo_adev) { - WARN_ONCE(1, "failed to get device by id 0x%x\n", best_loc); + bo_node = svm_range_get_node_by_id(prange, best_loc); + if (!bo_node) { + WARN_ONCE(1, "failed to get valid kfd node at id%x\n", best_loc); + best_loc = 0; + goto out; + } + + if (bo_node->adev->gmc.is_app_apu) { best_loc = 0; goto out; } @@ -3241,10 +3310,10 @@ svm_range_best_prefetch_location(struct svm_range *prange) continue; } - if (pdd->dev->adev == bo_adev) + if (pdd->dev->adev == bo_node->adev) continue; - if (!amdgpu_xgmi_same_hive(pdd->dev->adev, bo_adev)) { + if (!svm_nodes_in_same_hive(pdd->dev, bo_node)) { best_loc = 0; break; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h index 7a33b93f9df6fbb80a7628946f678bc48368f1a2..21b14510882b89364a0b2dce47745b4242cb0338 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h @@ -48,6 +48,7 @@ struct svm_range_bo { struct work_struct eviction_work; uint32_t evicting; struct work_struct release_work; + struct kfd_node *node; }; enum svm_work_list_ops { @@ -133,6 +134,7 @@ struct svm_range { DECLARE_BITMAP(bitmap_aip, MAX_GPU_INSTANCE); bool validated_once; bool mapped_to_gpu; + bool is_error_flag; }; static inline void svm_range_lock(struct svm_range *prange) @@ -163,16 +165,17 @@ int svm_ioctl(struct kfd_process *p, enum kfd_ioctl_svm_op op, uint64_t start, struct svm_range *svm_range_from_addr(struct svm_range_list *svms, unsigned long addr, struct svm_range **parent); -struct amdgpu_device *svm_range_get_adev_by_id(struct svm_range *prange, - uint32_t id); -int svm_range_vram_node_new(struct amdgpu_device *adev, - struct svm_range *prange, bool clear); +struct kfd_node *svm_range_get_node_by_id(struct svm_range *prange, + uint32_t gpu_id); +int svm_range_vram_node_new(struct kfd_node *node, struct svm_range *prange, + bool clear); void svm_range_vram_node_free(struct svm_range *prange); int svm_range_split_by_granularity(struct kfd_process *p, struct mm_struct *mm, unsigned long addr, struct svm_range *parent, struct svm_range *prange); -int svm_range_restore_pages(struct amdgpu_device *adev, - unsigned int pasid, uint64_t addr, bool write_fault); +int svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid, + uint32_t vmid, uint32_t node_id, uint64_t addr, + bool write_fault); int svm_range_schedule_evict_svm_bo(struct amdgpu_amdkfd_fence *fence); void svm_range_add_list_work(struct svm_range_list *svms, struct svm_range *prange, struct mm_struct *mm, @@ -192,13 +195,14 @@ int kfd_criu_restore_svm(struct kfd_process *p, uint64_t max_priv_data_size); int kfd_criu_resume_svm(struct kfd_process *p); struct kfd_process_device * -svm_range_get_pdd_by_adev(struct svm_range *prange, struct amdgpu_device *adev); +svm_range_get_pdd_by_node(struct svm_range *prange, struct kfd_node *node); void svm_range_list_lock_and_flush_work(struct svm_range_list *svms, struct mm_struct *mm); /* SVM API and HMM page migration work together, device memory type * is initialized to not 0 when page migration register device memory. */ -#define KFD_IS_SVM_API_SUPPORTED(dev) ((dev)->pgmap.type != 0) +#define KFD_IS_SVM_API_SUPPORTED(adev) ((adev)->kfd.pgmap.type != 0 ||\ + (adev)->gmc.is_app_apu) void svm_range_bo_unref_async(struct svm_range_bo *svm_bo); @@ -219,8 +223,9 @@ static inline void svm_range_list_fini(struct kfd_process *p) } static inline int svm_range_restore_pages(struct amdgpu_device *adev, - unsigned int pasid, uint64_t addr, - bool write_fault) + unsigned int pasid, + uint32_t client_id, uint32_t node_id, + uint64_t addr, bool write_fault) { return -EFAULT; } @@ -261,6 +266,10 @@ static inline int kfd_criu_resume_svm(struct kfd_process *p) return 0; } +static inline void svm_range_set_max_pages(struct amdgpu_device *adev) +{ +} + #define KFD_IS_SVM_API_SUPPORTED(dev) false #endif /* IS_ENABLED(CONFIG_HSA_AMD_SVM) */ diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index 8e4124dcb6e4c532840f50e3054565c3e772fb58..90b86a6ac7bd63f030a1ccef73fa0e582cb19ea3 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -96,7 +96,7 @@ struct kfd_topology_device *kfd_topology_device_by_id(uint32_t gpu_id) return ret; } -struct kfd_dev *kfd_device_by_id(uint32_t gpu_id) +struct kfd_node *kfd_device_by_id(uint32_t gpu_id) { struct kfd_topology_device *top_dev; @@ -107,10 +107,10 @@ struct kfd_dev *kfd_device_by_id(uint32_t gpu_id) return top_dev->gpu; } -struct kfd_dev *kfd_device_by_pci_dev(const struct pci_dev *pdev) +struct kfd_node *kfd_device_by_pci_dev(const struct pci_dev *pdev) { struct kfd_topology_device *top_dev; - struct kfd_dev *device = NULL; + struct kfd_node *device = NULL; down_read(&topology_lock); @@ -125,24 +125,6 @@ struct kfd_dev *kfd_device_by_pci_dev(const struct pci_dev *pdev) return device; } -struct kfd_dev *kfd_device_by_adev(const struct amdgpu_device *adev) -{ - struct kfd_topology_device *top_dev; - struct kfd_dev *device = NULL; - - down_read(&topology_lock); - - list_for_each_entry(top_dev, &topology_device_list, list) - if (top_dev->gpu && top_dev->gpu->adev == adev) { - device = top_dev->gpu; - break; - } - - up_read(&topology_lock); - - return device; -} - /* Called with write topology_lock acquired */ static void kfd_release_topology_device(struct kfd_topology_device *dev) { @@ -468,7 +450,8 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, sysfs_show_32bit_prop(buffer, offs, "cpu_cores_count", dev->node_props.cpu_cores_count); sysfs_show_32bit_prop(buffer, offs, "simd_count", - dev->gpu ? dev->node_props.simd_count : 0); + dev->gpu ? (dev->node_props.simd_count * + NUM_XCC(dev->gpu->xcc_mask)) : 0); sysfs_show_32bit_prop(buffer, offs, "mem_banks_count", dev->node_props.mem_banks_count); sysfs_show_32bit_prop(buffer, offs, "caches_count", @@ -492,7 +475,8 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, sysfs_show_32bit_prop(buffer, offs, "wave_front_size", dev->node_props.wave_front_size); sysfs_show_32bit_prop(buffer, offs, "array_count", - dev->node_props.array_count); + dev->gpu ? (dev->node_props.array_count * + NUM_XCC(dev->gpu->xcc_mask)) : 0); sysfs_show_32bit_prop(buffer, offs, "simd_arrays_per_engine", dev->node_props.simd_arrays_per_engine); sysfs_show_32bit_prop(buffer, offs, "cu_per_simd_array", @@ -526,7 +510,7 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, if (dev->gpu) { log_max_watch_addr = - __ilog2_u32(dev->gpu->device_info.num_of_watch_points); + __ilog2_u32(dev->gpu->kfd->device_info.num_of_watch_points); if (log_max_watch_addr) { dev->node_props.capability |= @@ -548,14 +532,17 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, sysfs_show_64bit_prop(buffer, offs, "local_mem_size", 0ULL); sysfs_show_32bit_prop(buffer, offs, "fw_version", - dev->gpu->mec_fw_version); + dev->gpu->kfd->mec_fw_version); sysfs_show_32bit_prop(buffer, offs, "capability", dev->node_props.capability); + sysfs_show_64bit_prop(buffer, offs, "debug_prop", + dev->node_props.debug_prop); sysfs_show_32bit_prop(buffer, offs, "sdma_fw_version", - dev->gpu->sdma_fw_version); + dev->gpu->kfd->sdma_fw_version); sysfs_show_64bit_prop(buffer, offs, "unique_id", dev->gpu->adev->unique_id); - + sysfs_show_32bit_prop(buffer, offs, "num_xcc", + NUM_XCC(dev->gpu->xcc_mask)); } return sysfs_show_32bit_prop(buffer, offs, "max_engine_clk_ccompute", @@ -1157,10 +1144,10 @@ void kfd_topology_shutdown(void) up_write(&topology_lock); } -static uint32_t kfd_generate_gpu_id(struct kfd_dev *gpu) +static uint32_t kfd_generate_gpu_id(struct kfd_node *gpu) { uint32_t hashout; - uint32_t buf[7]; + uint32_t buf[8]; uint64_t local_mem_size; int i; @@ -1177,8 +1164,9 @@ static uint32_t kfd_generate_gpu_id(struct kfd_dev *gpu) buf[4] = gpu->adev->pdev->bus->number; buf[5] = lower_32_bits(local_mem_size); buf[6] = upper_32_bits(local_mem_size); + buf[7] = (ffs(gpu->xcc_mask) - 1) | (NUM_XCC(gpu->xcc_mask) << 16); - for (i = 0, hashout = 0; i < 7; i++) + for (i = 0, hashout = 0; i < 8; i++) hashout ^= hash_32(buf[i], KFD_GPU_ID_HASH_WIDTH); return hashout; @@ -1188,7 +1176,7 @@ static uint32_t kfd_generate_gpu_id(struct kfd_dev *gpu) * list then return NULL. This means a new topology device has to * be created for this GPU. */ -static struct kfd_topology_device *kfd_assign_gpu(struct kfd_dev *gpu) +static struct kfd_topology_device *kfd_assign_gpu(struct kfd_node *gpu) { struct kfd_topology_device *dev; struct kfd_topology_device *out_dev = NULL; @@ -1201,7 +1189,7 @@ static struct kfd_topology_device *kfd_assign_gpu(struct kfd_dev *gpu) /* Discrete GPUs need their own topology device list * entries. Don't assign them to CPU/APU nodes. */ - if (!gpu->use_iommu_v2 && + if (!gpu->kfd->use_iommu_v2 && dev->node_props.cpu_cores_count) continue; @@ -1248,7 +1236,8 @@ static void kfd_fill_mem_clk_max_info(struct kfd_topology_device *dev) * for APUs - If CRAT from ACPI reports more than one bank, then * all the banks will report the same mem_clk_max information */ - amdgpu_amdkfd_get_local_mem_info(dev->gpu->adev, &local_mem_info); + amdgpu_amdkfd_get_local_mem_info(dev->gpu->adev, &local_mem_info, + dev->gpu->xcp); list_for_each_entry(mem, &dev->mem_props, list) mem->mem_clk_max = local_mem_info.mem_clk_max; @@ -1275,7 +1264,7 @@ static void kfd_set_iolink_no_atomics(struct kfd_topology_device *dev, CRAT_IOLINK_FLAGS_NO_ATOMICS_64_BIT; /* set gpu (dev) flags. */ } else { - if (!dev->gpu->pci_atomic_requested || + if (!dev->gpu->kfd->pci_atomic_requested || dev->gpu->adev->asic_type == CHIP_HAWAII) link->flags |= CRAT_IOLINK_FLAGS_NO_ATOMICS_32_BIT | CRAT_IOLINK_FLAGS_NO_ATOMICS_64_BIT; @@ -1323,10 +1312,16 @@ static void kfd_fill_iolink_non_crat_info(struct kfd_topology_device *dev) continue; /* Include the CPU peer in GPU hive if connected over xGMI. */ - if (!peer_dev->gpu && !peer_dev->node_props.hive_id && - dev->node_props.hive_id && - dev->gpu->adev->gmc.xgmi.connected_to_cpu) + if (!peer_dev->gpu && + link->iolink_type == CRAT_IOLINK_TYPE_XGMI) { + /* + * If the GPU is not part of a GPU hive, use its pci + * device location as the hive ID to bind with the CPU. + */ + if (!dev->node_props.hive_id) + dev->node_props.hive_id = pci_dev_id(dev->gpu->adev->pdev); peer_dev->node_props.hive_id = dev->node_props.hive_id; + } list_for_each_entry(inbound_link, &peer_dev->io_link_props, list) { @@ -1569,8 +1564,8 @@ static int kfd_dev_create_p2p_links(void) if (dev == new_dev) break; if (!dev->gpu || !dev->gpu->adev || - (dev->gpu->hive_id && - dev->gpu->hive_id == new_dev->gpu->hive_id)) + (dev->gpu->kfd->hive_id && + dev->gpu->kfd->hive_id == new_dev->gpu->kfd->hive_id)) goto next; /* check if node(s) is/are peer accessible in one direction or bi-direction */ @@ -1590,7 +1585,6 @@ static int kfd_dev_create_p2p_links(void) return ret; } - /* Helper function. See kfd_fill_gpu_cache_info for parameter description */ static int fill_in_l1_pcache(struct kfd_cache_properties **props_ext, struct kfd_gpu_cache_info *pcache_info, @@ -1723,7 +1717,7 @@ static int fill_in_l2_l3_pcache(struct kfd_cache_properties **props_ext, /* kfd_fill_cache_non_crat_info - Fill GPU cache info using kfd_gpu_cache_info * tables */ -static void kfd_fill_cache_non_crat_info(struct kfd_topology_device *dev, struct kfd_dev *kdev) +static void kfd_fill_cache_non_crat_info(struct kfd_topology_device *dev, struct kfd_node *kdev) { struct kfd_gpu_cache_info *pcache_info = NULL; int i, j, k; @@ -1805,7 +1799,7 @@ static void kfd_fill_cache_non_crat_info(struct kfd_topology_device *dev, struct pr_debug("Added [%d] GPU cache entries\n", num_of_entries); } -static int kfd_topology_add_device_locked(struct kfd_dev *gpu, uint32_t gpu_id, +static int kfd_topology_add_device_locked(struct kfd_node *gpu, uint32_t gpu_id, struct kfd_topology_device **dev) { int proximity_domain = ++topology_crat_proximity_domain; @@ -1865,7 +1859,103 @@ static int kfd_topology_add_device_locked(struct kfd_dev *gpu, uint32_t gpu_id, return res; } -int kfd_topology_add_device(struct kfd_dev *gpu) +static void kfd_topology_set_dbg_firmware_support(struct kfd_topology_device *dev) +{ + bool firmware_supported = true; + + if (KFD_GC_VERSION(dev->gpu) >= IP_VERSION(11, 0, 0) && + KFD_GC_VERSION(dev->gpu) < IP_VERSION(12, 0, 0)) { + uint32_t mes_api_rev = (dev->gpu->adev->mes.sched_version & + AMDGPU_MES_API_VERSION_MASK) >> + AMDGPU_MES_API_VERSION_SHIFT; + uint32_t mes_rev = dev->gpu->adev->mes.sched_version & + AMDGPU_MES_VERSION_MASK; + + firmware_supported = (mes_api_rev >= 14) && (mes_rev >= 64); + goto out; + } + + /* + * Note: Any unlisted devices here are assumed to support exception handling. + * Add additional checks here as needed. + */ + switch (KFD_GC_VERSION(dev->gpu)) { + case IP_VERSION(9, 0, 1): + firmware_supported = dev->gpu->kfd->mec_fw_version >= 459 + 32768; + break; + case IP_VERSION(9, 1, 0): + case IP_VERSION(9, 2, 1): + case IP_VERSION(9, 2, 2): + case IP_VERSION(9, 3, 0): + case IP_VERSION(9, 4, 0): + firmware_supported = dev->gpu->kfd->mec_fw_version >= 459; + break; + case IP_VERSION(9, 4, 1): + firmware_supported = dev->gpu->kfd->mec_fw_version >= 60; + break; + case IP_VERSION(9, 4, 2): + firmware_supported = dev->gpu->kfd->mec_fw_version >= 51; + break; + case IP_VERSION(10, 1, 10): + case IP_VERSION(10, 1, 2): + case IP_VERSION(10, 1, 1): + firmware_supported = dev->gpu->kfd->mec_fw_version >= 144; + break; + case IP_VERSION(10, 3, 0): + case IP_VERSION(10, 3, 2): + case IP_VERSION(10, 3, 1): + case IP_VERSION(10, 3, 4): + case IP_VERSION(10, 3, 5): + firmware_supported = dev->gpu->kfd->mec_fw_version >= 89; + break; + case IP_VERSION(10, 1, 3): + case IP_VERSION(10, 3, 3): + firmware_supported = false; + break; + default: + break; + } + +out: + if (firmware_supported) + dev->node_props.capability |= HSA_CAP_TRAP_DEBUG_FIRMWARE_SUPPORTED; +} + +static void kfd_topology_set_capabilities(struct kfd_topology_device *dev) +{ + dev->node_props.capability |= ((HSA_CAP_DOORBELL_TYPE_2_0 << + HSA_CAP_DOORBELL_TYPE_TOTALBITS_SHIFT) & + HSA_CAP_DOORBELL_TYPE_TOTALBITS_MASK); + + dev->node_props.capability |= HSA_CAP_TRAP_DEBUG_SUPPORT | + HSA_CAP_TRAP_DEBUG_WAVE_LAUNCH_TRAP_OVERRIDE_SUPPORTED | + HSA_CAP_TRAP_DEBUG_WAVE_LAUNCH_MODE_SUPPORTED; + + if (KFD_GC_VERSION(dev->gpu) < IP_VERSION(10, 0, 0)) { + dev->node_props.debug_prop |= HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX9 | + HSA_DBG_WATCH_ADDR_MASK_HI_BIT; + + if (KFD_GC_VERSION(dev->gpu) < IP_VERSION(9, 4, 2)) + dev->node_props.debug_prop |= + HSA_DBG_DISPATCH_INFO_ALWAYS_VALID; + else + dev->node_props.capability |= + HSA_CAP_TRAP_DEBUG_PRECISE_MEMORY_OPERATIONS_SUPPORTED; + } else { + dev->node_props.debug_prop |= HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX10 | + HSA_DBG_WATCH_ADDR_MASK_HI_BIT; + + if (KFD_GC_VERSION(dev->gpu) < IP_VERSION(11, 0, 0)) + dev->node_props.debug_prop |= HSA_DBG_DISPATCH_INFO_ALWAYS_VALID; + else + dev->node_props.capability |= + HSA_CAP_TRAP_DEBUG_PRECISE_MEMORY_OPERATIONS_SUPPORTED; + } + + kfd_topology_set_dbg_firmware_support(dev); +} + +int kfd_topology_add_device(struct kfd_node *gpu) { uint32_t gpu_id; struct kfd_topology_device *dev; @@ -1916,28 +2006,37 @@ int kfd_topology_add_device(struct kfd_dev *gpu) dev->node_props.simd_arrays_per_engine = cu_info.num_shader_arrays_per_engine; - dev->node_props.gfx_target_version = gpu->device_info.gfx_target_version; + dev->node_props.gfx_target_version = + gpu->kfd->device_info.gfx_target_version; dev->node_props.vendor_id = gpu->adev->pdev->vendor; dev->node_props.device_id = gpu->adev->pdev->device; dev->node_props.capability |= ((dev->gpu->adev->rev_id << HSA_CAP_ASIC_REVISION_SHIFT) & HSA_CAP_ASIC_REVISION_MASK); + dev->node_props.location_id = pci_dev_id(gpu->adev->pdev); + if (KFD_GC_VERSION(dev->gpu->kfd) == IP_VERSION(9, 4, 3)) + dev->node_props.location_id |= dev->gpu->node_id; + dev->node_props.domain = pci_domain_nr(gpu->adev->pdev->bus); dev->node_props.max_engine_clk_fcompute = amdgpu_amdkfd_get_max_engine_clock_in_mhz(dev->gpu->adev); dev->node_props.max_engine_clk_ccompute = cpufreq_quick_get_max(0) / 1000; - dev->node_props.drm_render_minor = - gpu->shared_resources.drm_render_minor; - dev->node_props.hive_id = gpu->hive_id; + if (gpu->xcp) + dev->node_props.drm_render_minor = gpu->xcp->ddev->render->index; + else + dev->node_props.drm_render_minor = + gpu->kfd->shared_resources.drm_render_minor; + + dev->node_props.hive_id = gpu->kfd->hive_id; dev->node_props.num_sdma_engines = kfd_get_num_sdma_engines(gpu); dev->node_props.num_sdma_xgmi_engines = kfd_get_num_xgmi_sdma_engines(gpu); dev->node_props.num_sdma_queues_per_engine = - gpu->device_info.num_sdma_queues_per_engine - - gpu->device_info.num_reserved_sdma_queues_per_engine; + gpu->kfd->device_info.num_sdma_queues_per_engine - + gpu->kfd->device_info.num_reserved_sdma_queues_per_engine; dev->node_props.num_gws = (dev->gpu->gws && dev->gpu->dqm->sched_policy != KFD_SCHED_POLICY_NO_HWS) ? dev->gpu->adev->gds.gws_size : 0; @@ -1966,20 +2065,18 @@ int kfd_topology_add_device(struct kfd_dev *gpu) HSA_CAP_DOORBELL_TYPE_TOTALBITS_MASK); break; default: - if (KFD_GC_VERSION(dev->gpu) >= IP_VERSION(9, 0, 1)) - dev->node_props.capability |= ((HSA_CAP_DOORBELL_TYPE_2_0 << - HSA_CAP_DOORBELL_TYPE_TOTALBITS_SHIFT) & - HSA_CAP_DOORBELL_TYPE_TOTALBITS_MASK); - else + if (KFD_GC_VERSION(dev->gpu) < IP_VERSION(9, 0, 1)) WARN(1, "Unexpected ASIC family %u", dev->gpu->adev->asic_type); + else + kfd_topology_set_capabilities(dev); } /* * Overwrite ATS capability according to needs_iommu_device to fix * potential missing corresponding bit in CRAT of BIOS. */ - if (dev->gpu->use_iommu_v2) + if (dev->gpu->kfd->use_iommu_v2) dev->node_props.capability |= HSA_CAP_ATS_PRESENT; else dev->node_props.capability &= ~HSA_CAP_ATS_PRESENT; @@ -2007,7 +2104,7 @@ int kfd_topology_add_device(struct kfd_dev *gpu) dev->node_props.capability |= (dev->gpu->adev->ras_enabled != 0) ? HSA_CAP_RASEVENTNOTIFY : 0; - if (KFD_IS_SVM_API_SUPPORTED(dev->gpu->adev->kfd.dev)) + if (KFD_IS_SVM_API_SUPPORTED(dev->gpu->adev)) dev->node_props.capability |= HSA_CAP_SVMAPI_SUPPORTED; kfd_debug_print_topology(); @@ -2079,7 +2176,7 @@ static void kfd_topology_update_io_links(int proximity_domain) } } -int kfd_topology_remove_device(struct kfd_dev *gpu) +int kfd_topology_remove_device(struct kfd_node *gpu) { struct kfd_topology_device *dev, *tmp; uint32_t gpu_id; @@ -2119,7 +2216,7 @@ int kfd_topology_remove_device(struct kfd_dev *gpu) * Return - 0: On success (@kdev will be NULL for non GPU nodes) * -1: If end of list */ -int kfd_topology_enum_kfd_devices(uint8_t idx, struct kfd_dev **kdev) +int kfd_topology_enum_kfd_devices(uint8_t idx, struct kfd_node **kdev) { struct kfd_topology_device *top_dev; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h index fca30d00a9bb3e3c7b09fab0fb8c2974c528ee41..cba2cd5ed9d19cbda41dcfd07cd0eb44ed8f9e48 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h @@ -31,6 +31,11 @@ #define KFD_TOPOLOGY_PUBLIC_NAME_SIZE 32 +#define HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX9 6 +#define HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX10 7 +#define HSA_DBG_WATCH_ADDR_MASK_HI_BIT \ + (29 << HSA_DBG_WATCH_ADDR_MASK_HI_BIT_SHIFT) + struct kfd_node_properties { uint64_t hive_id; uint32_t cpu_cores_count; @@ -42,6 +47,7 @@ struct kfd_node_properties { uint32_t cpu_core_id_base; uint32_t simd_id_base; uint32_t capability; + uint64_t debug_prop; uint32_t max_waves_per_simd; uint32_t lds_size_in_kb; uint32_t gds_size_in_kb; @@ -75,7 +81,7 @@ struct kfd_mem_properties { uint32_t flags; uint32_t width; uint32_t mem_clk_max; - struct kfd_dev *gpu; + struct kfd_node *gpu; struct kobject *kobj; struct attribute attr; }; @@ -93,7 +99,7 @@ struct kfd_cache_properties { uint32_t cache_latency; uint32_t cache_type; uint8_t sibling_map[CACHE_SIBLINGMAP_SIZE]; - struct kfd_dev *gpu; + struct kfd_node *gpu; struct kobject *kobj; struct attribute attr; uint32_t sibling_map_size; @@ -113,7 +119,7 @@ struct kfd_iolink_properties { uint32_t max_bandwidth; uint32_t rec_transfer_size; uint32_t flags; - struct kfd_dev *gpu; + struct kfd_node *gpu; struct kobject *kobj; struct attribute attr; }; @@ -135,7 +141,7 @@ struct kfd_topology_device { struct list_head io_link_props; struct list_head p2p_link_props; struct list_head perf_props; - struct kfd_dev *gpu; + struct kfd_node *gpu; struct kobject *kobj_node; struct kobject *kobj_mem; struct kobject *kobj_cache; diff --git a/drivers/gpu/drm/amd/amdxcp/Makefile b/drivers/gpu/drm/amd/amdxcp/Makefile new file mode 100644 index 0000000000000000000000000000000000000000..870501a4bb8c044a2e57dd1bf96bc9f3473128bb --- /dev/null +++ b/drivers/gpu/drm/amd/amdxcp/Makefile @@ -0,0 +1,25 @@ +# +# Copyright 2023 Advanced Micro Devices, Inc. +# +# Permission is hereby granted, free of charge, to any person obtaining a +# copy of this software and associated documentation files (the "Software"), +# to deal in the Software without restriction, including without limitation +# the rights to use, copy, modify, merge, publish, distribute, sublicense, +# and/or sell copies of the Software, and to permit persons to whom the +# Software is furnished to do so, subject to the following conditions: +# +# The above copyright notice and this permission notice shall be included in +# all copies or substantial portions of the Software. +# +# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL +# THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR +# OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, +# ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR +# OTHER DEALINGS IN THE SOFTWARE. +# + +amdxcp-y := amdgpu_xcp_drv.o + +obj-$(CONFIG_DRM_AMDGPU) += amdxcp.o diff --git a/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c new file mode 100644 index 0000000000000000000000000000000000000000..353597fc908d9ffc384f3d7181e4c757549220d6 --- /dev/null +++ b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c @@ -0,0 +1,110 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#include +#include +#include + +#include + +#include "amdgpu_xcp_drv.h" + +#define MAX_XCP_PLATFORM_DEVICE 64 + +struct xcp_device { + struct drm_device drm; + struct platform_device *pdev; +}; + +static const struct drm_driver amdgpu_xcp_driver = { + .driver_features = DRIVER_GEM | DRIVER_RENDER, + .name = "amdgpu_xcp_drv", + .major = 1, + .minor = 0, +}; + +static int pdev_num; +static struct xcp_device *xcp_dev[MAX_XCP_PLATFORM_DEVICE]; + +int amdgpu_xcp_drm_dev_alloc(struct drm_device **ddev) +{ + struct platform_device *pdev; + struct xcp_device *pxcp_dev; + int ret; + + if (pdev_num >= MAX_XCP_PLATFORM_DEVICE) + return -ENODEV; + + pdev = platform_device_register_simple("amdgpu_xcp", pdev_num, NULL, 0); + if (IS_ERR(pdev)) + return PTR_ERR(pdev); + + if (!devres_open_group(&pdev->dev, NULL, GFP_KERNEL)) { + ret = -ENOMEM; + goto out_unregister; + } + + pxcp_dev = devm_drm_dev_alloc(&pdev->dev, &amdgpu_xcp_driver, struct xcp_device, drm); + if (IS_ERR(pxcp_dev)) { + ret = PTR_ERR(pxcp_dev); + goto out_devres; + } + + xcp_dev[pdev_num] = pxcp_dev; + xcp_dev[pdev_num]->pdev = pdev; + *ddev = &pxcp_dev->drm; + pdev_num++; + + return 0; + +out_devres: + devres_release_group(&pdev->dev, NULL); +out_unregister: + platform_device_unregister(pdev); + + return ret; +} +EXPORT_SYMBOL(amdgpu_xcp_drm_dev_alloc); + +void amdgpu_xcp_drv_release(void) +{ + for (--pdev_num; pdev_num >= 0; --pdev_num) { + devres_release_group(&xcp_dev[pdev_num]->pdev->dev, NULL); + platform_device_unregister(xcp_dev[pdev_num]->pdev); + xcp_dev[pdev_num]->pdev = NULL; + xcp_dev[pdev_num] = NULL; + } + pdev_num = 0; +} +EXPORT_SYMBOL(amdgpu_xcp_drv_release); + +static void __exit amdgpu_xcp_drv_exit(void) +{ + amdgpu_xcp_drv_release(); +} + +module_exit(amdgpu_xcp_drv_exit); + +MODULE_AUTHOR("AMD linux driver team"); +MODULE_DESCRIPTION("AMD XCP PLATFORM DEVICES"); +MODULE_LICENSE("GPL and additional rights"); diff --git a/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.h b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.h new file mode 100644 index 0000000000000000000000000000000000000000..c1c4b679bf95c875478a1204a32bba17d326af02 --- /dev/null +++ b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.h @@ -0,0 +1,29 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef _AMDGPU_XCP_DRV_H_ +#define _AMDGPU_XCP_DRV_H_ + +int amdgpu_xcp_drm_dev_alloc(struct drm_device **ddev); +void amdgpu_xcp_drv_release(void); +#endif /* _AMDGPU_XCP_DRV_H_ */ diff --git a/drivers/gpu/drm/amd/display/Kconfig b/drivers/gpu/drm/amd/display/Kconfig index 2d8e55e29637fcb45371367e6c672f0fa6c26bd9..bf0a655d009e60e78235ca542706ca3f23440df9 100644 --- a/drivers/gpu/drm/amd/display/Kconfig +++ b/drivers/gpu/drm/amd/display/Kconfig @@ -8,7 +8,7 @@ config DRM_AMD_DC depends on BROKEN || !CC_IS_CLANG || X86_64 || SPARC64 || ARM64 select SND_HDA_COMPONENT if SND_HDA_CORE # !CC_IS_CLANG: https://github.com/ClangBuiltLinux/linux/issues/1752 - select DRM_AMD_DC_FP if (X86 || (PPC64 && ALTIVEC) || (ARM64 && KERNEL_MODE_NEON && !CC_IS_CLANG)) + select DRM_AMD_DC_FP if (X86 || LOONGARCH || (PPC64 && ALTIVEC) || (ARM64 && KERNEL_MODE_NEON && !CC_IS_CLANG)) help Choose this option if you want to use the new display engine support for AMDGPU. This adds required support for Vega and @@ -42,16 +42,13 @@ config DEBUG_KERNEL_DC Choose this option if you want to hit kdgb_break in assert. config DRM_AMD_SECURE_DISPLAY - bool "Enable secure display support" - depends on DEBUG_FS - depends on DRM_AMD_DC_FP - help - Choose this option if you want to - support secure display - - This option enables the calculation - of crc of specific region via debugfs. - Cooperate with specific DMCU FW. + bool "Enable secure display support" + depends on DEBUG_FS + depends on DRM_AMD_DC_FP + help + Choose this option if you want to support secure display + This option enables the calculation of crc of specific region via + debugfs. Cooperate with specific DMCU FW. endmenu diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 8b4b186c57f515fd541eaa15ce55e0b3ebf2dadb..514f6785a02093ea3b19221d6fac4ad18e078cee 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -365,6 +365,14 @@ static inline void reverse_planes_order(struct dc_surface_update *array_of_surfa * adjustments and preparation before calling it. This function is a wrapper * for the dc_update_planes_and_stream that does any required configuration * before passing control to DC. + * + * @dc: Display Core control structure + * @update_type: specify whether it is FULL/MEDIUM/FAST update + * @planes_count: planes count to update + * @stream: stream state + * @stream_update: stream update + * @array_of_surface_update: dc surface update pointer + * */ static inline bool update_planes_and_stream_adapter(struct dc *dc, int update_type, @@ -1646,11 +1654,6 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) if (amdgpu_dc_feature_mask & DC_DISABLE_LTTPR_DP2_0) init_data.flags.allow_lttpr_non_transparent_mode.bits.DP2_0 = true; - /* Disable SubVP + DRR config by default */ - init_data.flags.disable_subvp_drr = true; - if (amdgpu_dc_feature_mask & DC_ENABLE_SUBVP_DRR) - init_data.flags.disable_subvp_drr = false; - init_data.flags.seamless_boot_edp_requested = false; if (check_seamless_boot_capability(adev)) { @@ -1672,9 +1675,11 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) adev->dm.dc = dc_create(&init_data); if (adev->dm.dc) { - DRM_INFO("Display Core initialized with v%s!\n", DC_VER); + DRM_INFO("Display Core v%s initialized on %s\n", DC_VER, + dce_version_to_string(adev->dm.dc->ctx->dce_version)); } else { - DRM_INFO("Display Core failed to initialize with v%s!\n", DC_VER); + DRM_INFO("Display Core v%s failed to initialize on %s\n", DC_VER, + dce_version_to_string(adev->dm.dc->ctx->dce_version)); goto error; } @@ -1776,12 +1781,6 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) dc_init_callbacks(adev->dm.dc, &init_params); } -#if defined(CONFIG_DRM_AMD_SECURE_DISPLAY) - adev->dm.secure_display_ctxs = amdgpu_dm_crtc_secure_display_create_contexts(adev); - if (!adev->dm.secure_display_ctxs) { - DRM_ERROR("amdgpu: failed to initialize secure_display_ctxs.\n"); - } -#endif if (dc_is_dmub_outbox_supported(adev->dm.dc)) { init_completion(&adev->dm.dmub_aux_transfer_done); adev->dm.dmub_notify = kzalloc(sizeof(struct dmub_notification), GFP_KERNEL); @@ -1840,6 +1839,11 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) goto error; } +#if defined(CONFIG_DRM_AMD_SECURE_DISPLAY) + adev->dm.secure_display_ctxs = amdgpu_dm_crtc_secure_display_create_contexts(adev); + if (!adev->dm.secure_display_ctxs) + DRM_ERROR("amdgpu: failed to initialize secure display contexts.\n"); +#endif DRM_DEBUG_DRIVER("KMS initialized.\n"); @@ -2479,20 +2483,25 @@ static void dm_gpureset_toggle_interrupts(struct amdgpu_device *adev, if (acrtc && state->stream_status[i].plane_count != 0) { irq_source = IRQ_TYPE_PFLIP + acrtc->otg_inst; rc = dc_interrupt_set(adev->dm.dc, irq_source, enable) ? 0 : -EBUSY; - DRM_DEBUG_VBL("crtc %d - vupdate irq %sabling: r=%d\n", - acrtc->crtc_id, enable ? "en" : "dis", rc); if (rc) DRM_WARN("Failed to %s pflip interrupts\n", enable ? "enable" : "disable"); if (enable) { - rc = amdgpu_dm_crtc_enable_vblank(&acrtc->base); - if (rc) - DRM_WARN("Failed to enable vblank interrupts\n"); - } else { - amdgpu_dm_crtc_disable_vblank(&acrtc->base); - } + if (amdgpu_dm_crtc_vrr_active(to_dm_crtc_state(acrtc->base.state))) + rc = amdgpu_dm_crtc_set_vupdate_irq(&acrtc->base, true); + } else + rc = amdgpu_dm_crtc_set_vupdate_irq(&acrtc->base, false); + + if (rc) + DRM_WARN("Failed to %sable vupdate interrupt\n", enable ? "en" : "dis"); + irq_source = IRQ_TYPE_VBLANK + acrtc->otg_inst; + /* During gpu-reset we disable and then enable vblank irq, so + * don't use amdgpu_irq_get/put() to avoid refcount change. + */ + if (!dc_interrupt_set(adev->dm.dc, irq_source, enable)) + DRM_WARN("Failed to %sable vblank interrupt\n", enable ? "en" : "dis"); } } @@ -2852,7 +2861,7 @@ static int dm_resume(void *handle) * this is the case when traversing through already created * MST connectors, should be skipped */ - if (aconnector->dc_link->type == dc_connection_mst_branch) + if (aconnector && aconnector->mst_root) continue; mutex_lock(&aconnector->hpd_lock); @@ -3258,6 +3267,7 @@ static void dm_handle_mst_sideband_msg(struct amdgpu_dm_connector *aconnector) while (dret == dpcd_bytes_to_read && process_count < max_process_count) { + u8 ack[DP_PSR_ERROR_STATUS - DP_SINK_COUNT_ESI] = {}; u8 retry; dret = 0; @@ -3266,28 +3276,29 @@ static void dm_handle_mst_sideband_msg(struct amdgpu_dm_connector *aconnector) DRM_DEBUG_DRIVER("ESI %02x %02x %02x\n", esi[0], esi[1], esi[2]); /* handle HPD short pulse irq */ if (aconnector->mst_mgr.mst_state) - drm_dp_mst_hpd_irq( - &aconnector->mst_mgr, - esi, - &new_irq_handled); + drm_dp_mst_hpd_irq_handle_event(&aconnector->mst_mgr, + esi, + ack, + &new_irq_handled); if (new_irq_handled) { /* ACK at DPCD to notify down stream */ - const int ack_dpcd_bytes_to_write = - dpcd_bytes_to_read - 1; - for (retry = 0; retry < 3; retry++) { - u8 wret; - - wret = drm_dp_dpcd_write( - &aconnector->dm_dp_aux.aux, - dpcd_addr + 1, - &esi[1], - ack_dpcd_bytes_to_write); - if (wret == ack_dpcd_bytes_to_write) + ssize_t wret; + + wret = drm_dp_dpcd_writeb(&aconnector->dm_dp_aux.aux, + dpcd_addr + 1, + ack[1]); + if (wret == 1) break; } + if (retry == 3) { + DRM_ERROR("Failed to ack MST event.\n"); + return; + } + + drm_dp_mst_hpd_irq_send_new_request(&aconnector->mst_mgr); /* check if there is new irq to be handled */ dret = drm_dp_dpcd_read( &aconnector->dm_dp_aux.aux, @@ -5326,21 +5337,44 @@ get_aspect_ratio(const struct drm_display_mode *mode_in) } static enum dc_color_space -get_output_color_space(const struct dc_crtc_timing *dc_crtc_timing) +get_output_color_space(const struct dc_crtc_timing *dc_crtc_timing, + const struct drm_connector_state *connector_state) { enum dc_color_space color_space = COLOR_SPACE_SRGB; - switch (dc_crtc_timing->pixel_encoding) { - case PIXEL_ENCODING_YCBCR422: - case PIXEL_ENCODING_YCBCR444: - case PIXEL_ENCODING_YCBCR420: - { + switch (connector_state->colorspace) { + case DRM_MODE_COLORIMETRY_BT601_YCC: + if (dc_crtc_timing->flags.Y_ONLY) + color_space = COLOR_SPACE_YCBCR601_LIMITED; + else + color_space = COLOR_SPACE_YCBCR601; + break; + case DRM_MODE_COLORIMETRY_BT709_YCC: + if (dc_crtc_timing->flags.Y_ONLY) + color_space = COLOR_SPACE_YCBCR709_LIMITED; + else + color_space = COLOR_SPACE_YCBCR709; + break; + case DRM_MODE_COLORIMETRY_OPRGB: + color_space = COLOR_SPACE_ADOBERGB; + break; + case DRM_MODE_COLORIMETRY_BT2020_RGB: + case DRM_MODE_COLORIMETRY_BT2020_YCC: + if (dc_crtc_timing->pixel_encoding == PIXEL_ENCODING_RGB) + color_space = COLOR_SPACE_2020_RGB_FULLRANGE; + else + color_space = COLOR_SPACE_2020_YCBCR; + break; + case DRM_MODE_COLORIMETRY_DEFAULT: // ITU601 + default: + if (dc_crtc_timing->pixel_encoding == PIXEL_ENCODING_RGB) { + color_space = COLOR_SPACE_SRGB; /* * 27030khz is the separation point between HDTV and SDTV * according to HDMI spec, we use YCbCr709 and YCbCr601 * respectively */ - if (dc_crtc_timing->pix_clk_100hz > 270300) { + } else if (dc_crtc_timing->pix_clk_100hz > 270300) { if (dc_crtc_timing->flags.Y_ONLY) color_space = COLOR_SPACE_YCBCR709_LIMITED; @@ -5353,15 +5387,6 @@ get_output_color_space(const struct dc_crtc_timing *dc_crtc_timing) else color_space = COLOR_SPACE_YCBCR601; } - - } - break; - case PIXEL_ENCODING_RGB: - color_space = COLOR_SPACE_SRGB; - break; - - default: - WARN_ON(1); break; } @@ -5500,7 +5525,7 @@ static void fill_stream_properties_from_drm_display_mode( } } - stream->output_color_space = get_output_color_space(timing_out); + stream->output_color_space = get_output_color_space(timing_out, connector_state); } static void fill_audio_info(struct audio_info *audio_info, @@ -5942,15 +5967,14 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, { struct drm_display_mode *preferred_mode = NULL; struct drm_connector *drm_connector; - const struct drm_connector_state *con_state = - dm_state ? &dm_state->base : NULL; + const struct drm_connector_state *con_state = &dm_state->base; struct dc_stream_state *stream = NULL; struct drm_display_mode mode; struct drm_display_mode saved_mode; struct drm_display_mode *freesync_mode = NULL; bool native_mode_found = false; bool recalculate_timing = false; - bool scale = dm_state ? (dm_state->scaling != RMX_OFF) : false; + bool scale = dm_state->scaling != RMX_OFF; int mode_refresh; int preferred_refresh = 0; enum color_transfer_func tf = TRANSFER_FUNC_UNKNOWN; @@ -6013,8 +6037,7 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, */ DRM_DEBUG_DRIVER("No preferred mode found\n"); } else { - recalculate_timing = amdgpu_freesync_vid_mode && - is_freesync_video_mode(&mode, aconnector); + recalculate_timing = is_freesync_video_mode(&mode, aconnector); if (recalculate_timing) { freesync_mode = get_highest_refresh_rate_mode(aconnector, false); drm_mode_copy(&saved_mode, &mode); @@ -6029,7 +6052,7 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, if (recalculate_timing) drm_mode_set_crtcinfo(&saved_mode, 0); - else if (!dm_state) + else drm_mode_set_crtcinfo(&mode, 0); /* @@ -6342,6 +6365,31 @@ amdgpu_dm_connector_late_register(struct drm_connector *connector) return 0; } +static void amdgpu_dm_connector_funcs_force(struct drm_connector *connector) +{ + struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); + struct dc_link *dc_link = aconnector->dc_link; + struct dc_sink *dc_em_sink = aconnector->dc_em_sink; + struct edid *edid; + + if (!connector->edid_override) + return; + + drm_edid_override_connector_update(&aconnector->base); + edid = aconnector->base.edid_blob_ptr->data; + aconnector->edid = edid; + + /* Update emulated (virtual) sink's EDID */ + if (dc_em_sink && dc_link) { + memset(&dc_em_sink->edid_caps, 0, sizeof(struct dc_edid_caps)); + memmove(dc_em_sink->dc_edid.raw_edid, edid, (edid->extensions + 1) * EDID_LENGTH); + dm_helpers_parse_edid_caps( + dc_link, + &dc_em_sink->dc_edid, + &dc_em_sink->edid_caps); + } +} + static const struct drm_connector_funcs amdgpu_dm_connector_funcs = { .reset = amdgpu_dm_connector_funcs_reset, .detect = amdgpu_dm_connector_detect, @@ -6352,7 +6400,8 @@ static const struct drm_connector_funcs amdgpu_dm_connector_funcs = { .atomic_set_property = amdgpu_dm_connector_atomic_set_property, .atomic_get_property = amdgpu_dm_connector_atomic_get_property, .late_register = amdgpu_dm_connector_late_register, - .early_unregister = amdgpu_dm_connector_unregister + .early_unregister = amdgpu_dm_connector_unregister, + .force = amdgpu_dm_connector_funcs_force }; static int get_modes(struct drm_connector *connector) @@ -6369,11 +6418,19 @@ static void create_eml_sink(struct amdgpu_dm_connector *aconnector) struct edid *edid; if (!aconnector->base.edid_blob_ptr) { - DRM_ERROR("No EDID firmware found on connector: %s ,forcing to OFF!\n", - aconnector->base.name); + /* if connector->edid_override valid, pass + * it to edid_override to edid_blob_ptr + */ - aconnector->base.force = DRM_FORCE_OFF; - return; + drm_edid_override_connector_update(&aconnector->base); + + if (!aconnector->base.edid_blob_ptr) { + DRM_ERROR("No EDID firmware found on connector: %s ,forcing to OFF!\n", + aconnector->base.name); + + aconnector->base.force = DRM_FORCE_OFF; + return; + } } edid = (struct edid *) aconnector->base.edid_blob_ptr->data; @@ -6558,7 +6615,9 @@ enum drm_mode_status amdgpu_dm_connector_mode_valid(struct drm_connector *connec goto fail; } - stream = create_validate_stream_for_sink(aconnector, mode, NULL, NULL); + stream = create_validate_stream_for_sink(aconnector, mode, + to_dm_connector_state(connector->state), + NULL); if (stream) { dc_stream_release(stream); result = MODE_OK; @@ -6652,6 +6711,14 @@ amdgpu_dm_connector_atomic_check(struct drm_connector *conn, if (!crtc) return 0; + if (new_con_state->colorspace != old_con_state->colorspace) { + new_crtc_state = drm_atomic_get_crtc_state(state, crtc); + if (IS_ERR(new_crtc_state)) + return PTR_ERR(new_crtc_state); + + new_crtc_state->mode_changed = true; + } + if (!drm_connector_atomic_hdr_metadata_equal(old_con_state, new_con_state)) { struct dc_info_packet hdr_infopacket; @@ -6674,7 +6741,7 @@ amdgpu_dm_connector_atomic_check(struct drm_connector *conn, * set is permissible, however. So only force a * modeset if we're entering or exiting HDR. */ - new_crtc_state->mode_changed = + new_crtc_state->mode_changed = new_crtc_state->mode_changed || !old_con_state->hdr_output_metadata || !new_con_state->hdr_output_metadata; } @@ -6737,7 +6804,7 @@ static int dm_encoder_helper_atomic_check(struct drm_encoder *encoder, int clock, bpp = 0; bool is_y420 = false; - if (!aconnector->mst_output_port || !aconnector->dc_sink) + if (!aconnector->mst_output_port) return 0; mst_port = aconnector->mst_output_port; @@ -7163,7 +7230,7 @@ static void amdgpu_dm_connector_add_freesync_modes(struct drm_connector *connect struct amdgpu_dm_connector *amdgpu_dm_connector = to_amdgpu_dm_connector(connector); - if (!(amdgpu_freesync_vid_mode && edid)) + if (!edid) return; if (amdgpu_dm_connector->max_vfreq - amdgpu_dm_connector->min_vfreq > 10) @@ -7191,7 +7258,13 @@ static int amdgpu_dm_connector_get_modes(struct drm_connector *connector) drm_add_modes_noedid(connector, 1920, 1080); } else { amdgpu_dm_connector_ddc_get_modes(connector, edid); - amdgpu_dm_connector_add_common_modes(encoder, connector); + /* most eDP supports only timings from its edid, + * usually only detailed timings are available + * from eDP edid. timings which are not from edid + * may damage eDP + */ + if (connector->connector_type != DRM_MODE_CONNECTOR_eDP) + amdgpu_dm_connector_add_common_modes(encoder, connector); amdgpu_dm_connector_add_freesync_modes(connector, edid); } amdgpu_dm_fbc_init(connector); @@ -7199,6 +7272,12 @@ static int amdgpu_dm_connector_get_modes(struct drm_connector *connector) return amdgpu_dm_connector->num_modes; } +static const u32 supported_colorspaces = + BIT(DRM_MODE_COLORIMETRY_BT709_YCC) | + BIT(DRM_MODE_COLORIMETRY_OPRGB) | + BIT(DRM_MODE_COLORIMETRY_BT2020_RGB) | + BIT(DRM_MODE_COLORIMETRY_BT2020_YCC); + void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm, struct amdgpu_dm_connector *aconnector, int connector_type, @@ -7279,6 +7358,15 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm, adev->mode_info.abm_level_property, 0); } + if (connector_type == DRM_MODE_CONNECTOR_HDMIA) { + if (!drm_mode_create_hdmi_colorspace_property(&aconnector->base, supported_colorspaces)) + drm_connector_attach_colorspace_property(&aconnector->base); + } else if (connector_type == DRM_MODE_CONNECTOR_DisplayPort || + connector_type == DRM_MODE_CONNECTOR_eDP) { + if (!drm_mode_create_dp_colorspace_property(&aconnector->base, supported_colorspaces)) + drm_connector_attach_colorspace_property(&aconnector->base); + } + if (connector_type == DRM_MODE_CONNECTOR_HDMIA || connector_type == DRM_MODE_CONNECTOR_DisplayPort || connector_type == DRM_MODE_CONNECTOR_eDP) { @@ -8193,6 +8281,12 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, if (acrtc_state->abm_level != dm_old_crtc_state->abm_level) bundle->stream_update.abm_level = &acrtc_state->abm_level; + mutex_lock(&dm->dc_lock); + if ((acrtc_state->update_type > UPDATE_TYPE_FAST) && + acrtc_state->stream->link->psr_settings.psr_allow_active) + amdgpu_dm_psr_disable(acrtc_state->stream); + mutex_unlock(&dm->dc_lock); + /* * If FreeSync state on the stream has changed then we need to * re-adjust the min/max bounds now that DC doesn't handle this @@ -8206,10 +8300,6 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, spin_unlock_irqrestore(&pcrtc->dev->event_lock, flags); } mutex_lock(&dm->dc_lock); - if ((acrtc_state->update_type > UPDATE_TYPE_FAST) && - acrtc_state->stream->link->psr_settings.psr_allow_active) - amdgpu_dm_psr_disable(acrtc_state->stream); - update_planes_and_stream_adapter(dm->dc, acrtc_state->update_type, planes_count, @@ -8871,10 +8961,17 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) drm_atomic_helper_cleanup_planes(dev, state); - /* return the stolen vga memory back to VRAM */ - if (!adev->mman.keep_stolen_vga_memory) - amdgpu_bo_free_kernel(&adev->mman.stolen_vga_memory, NULL, NULL); - amdgpu_bo_free_kernel(&adev->mman.stolen_extended_memory, NULL, NULL); + /* Don't free the memory if we are hitting this as part of suspend. + * This way we don't free any memory during suspend; see + * amdgpu_bo_free_kernel(). The memory will be freed in the first + * non-suspend modeset or when the driver is torn down. + */ + if (!adev->in_suspend) { + /* return the stolen vga memory back to VRAM */ + if (!adev->mman.keep_stolen_vga_memory) + amdgpu_bo_free_kernel(&adev->mman.stolen_vga_memory, NULL, NULL); + amdgpu_bo_free_kernel(&adev->mman.stolen_extended_memory, NULL, NULL); + } /* * Finally, drop a runtime PM reference for each newly disabled CRTC, @@ -9208,8 +9305,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm, * TODO: Refactor this function to allow this check to work * in all conditions. */ - if (amdgpu_freesync_vid_mode && - dm_new_crtc_state->stream && + if (dm_new_crtc_state->stream && is_timing_unchanged_for_freesync(new_crtc_state, old_crtc_state)) goto skip_modeset; @@ -9251,7 +9347,9 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm, } /* Now check if we should set freesync video mode */ - if (amdgpu_freesync_vid_mode && dm_new_crtc_state->stream && + if (dm_new_crtc_state->stream && + dc_is_stream_unchanged(new_stream, dm_old_crtc_state->stream) && + dc_is_stream_scaling_unchanged(new_stream, dm_old_crtc_state->stream) && is_timing_unchanged_for_freesync(new_crtc_state, old_crtc_state)) { new_crtc_state->mode_changed = false; @@ -9263,7 +9361,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm, set_freesync_fixed_config(dm_new_crtc_state); goto skip_modeset; - } else if (amdgpu_freesync_vid_mode && aconnector && + } else if (aconnector && is_freesync_video_mode(&new_crtc_state->mode, aconnector)) { struct drm_display_mode *high_mode; @@ -10323,7 +10421,7 @@ static bool dm_edid_parser_send_cea(struct amdgpu_display_manager *dm, input->cea_total_length = total_length; memcpy(input->payload, data, length); - res = dc_dmub_srv_cmd_with_reply_data(dm->dc->ctx->dmub_srv, &cmd); + res = dm_execute_dmub_cmd(dm->dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY); if (!res) { DRM_ERROR("EDID CEA parser failed\n"); return false; @@ -10773,3 +10871,13 @@ bool check_seamless_boot_capability(struct amdgpu_device *adev) return false; } + +bool dm_execute_dmub_cmd(const struct dc_context *ctx, union dmub_rb_cmd *cmd, enum dm_dmub_wait_type wait_type) +{ + return dc_dmub_srv_cmd_run(ctx->dmub_srv, cmd, wait_type); +} + +bool dm_execute_dmub_cmd_list(const struct dc_context *ctx, unsigned int count, union dmub_rb_cmd *cmd, enum dm_dmub_wait_type wait_type) +{ + return dc_dmub_srv_cmd_run_list(ctx->dmub_srv, count, cmd, wait_type); +} diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index 2e2413fd73a4f93682589042ae40ba08377667cd..4561f55afa99d157874207d84c9bee7a2f8e3b0b 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -661,10 +661,6 @@ struct amdgpu_dm_connector { struct mutex hpd_lock; bool fake_enable; -#ifdef CONFIG_DEBUG_FS - uint32_t debugfs_dpcd_address; - uint32_t debugfs_dpcd_size; -#endif bool force_yuv420_output; struct dsc_preferred_settings dsc_settings; union dp_downstream_port_present mst_downstream_port_present; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c index 27711743c22ce6d7cfe1576d3a6e2903cf798920..0802f8e8fac5f07f1f53d9fc85986154e499a45a 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c @@ -83,12 +83,15 @@ const char *const *amdgpu_dm_crtc_get_crc_sources(struct drm_crtc *crtc, } #ifdef CONFIG_DRM_AMD_SECURE_DISPLAY -static void amdgpu_dm_set_crc_window_default(struct drm_crtc *crtc) +static void amdgpu_dm_set_crc_window_default(struct drm_crtc *crtc, struct dc_stream_state *stream) { struct drm_device *drm_dev = crtc->dev; + struct amdgpu_display_manager *dm = &drm_to_adev(drm_dev)->dm; struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc); + bool was_activated; spin_lock_irq(&drm_dev->event_lock); + was_activated = acrtc->dm_irq_params.window_param.activated; acrtc->dm_irq_params.window_param.x_start = 0; acrtc->dm_irq_params.window_param.y_start = 0; acrtc->dm_irq_params.window_param.x_end = 0; @@ -97,6 +100,14 @@ static void amdgpu_dm_set_crc_window_default(struct drm_crtc *crtc) acrtc->dm_irq_params.window_param.update_win = false; acrtc->dm_irq_params.window_param.skip_frame_cnt = 0; spin_unlock_irq(&drm_dev->event_lock); + + /* Disable secure_display if it was enabled */ + if (was_activated) { + /* stop ROI update on this crtc */ + flush_work(&dm->secure_display_ctxs[crtc->index].notify_ta_work); + flush_work(&dm->secure_display_ctxs[crtc->index].forward_roi_work); + dc_stream_forward_crc_window(stream, NULL, true); + } } static void amdgpu_dm_crtc_notify_ta_to_read(struct work_struct *work) @@ -204,9 +215,6 @@ int amdgpu_dm_crtc_configure_crc_source(struct drm_crtc *crtc, struct dm_crtc_state *dm_crtc_state, enum amdgpu_dm_pipe_crc_source source) { -#if defined(CONFIG_DRM_AMD_SECURE_DISPLAY) - int i; -#endif struct amdgpu_device *adev = drm_to_adev(crtc->dev); struct dc_stream_state *stream_state = dm_crtc_state->stream; bool enable = amdgpu_dm_is_valid_crc_source(source); @@ -220,19 +228,6 @@ int amdgpu_dm_crtc_configure_crc_source(struct drm_crtc *crtc, /* Enable or disable CRTC CRC generation */ if (dm_is_crc_source_crtc(source) || source == AMDGPU_DM_PIPE_CRC_SOURCE_NONE) { -#if defined(CONFIG_DRM_AMD_SECURE_DISPLAY) - /* Disable secure_display if it was enabled */ - if (!enable) { - for (i = 0; i < adev->mode_info.num_crtc; i++) { - if (adev->dm.secure_display_ctxs[i].crtc == crtc) { - /* stop ROI update on this crtc */ - flush_work(&adev->dm.secure_display_ctxs[i].notify_ta_work); - flush_work(&adev->dm.secure_display_ctxs[i].forward_roi_work); - dc_stream_forward_crc_window(stream_state, NULL, true); - } - } - } -#endif if (!dc_stream_configure_crc(stream_state->ctx->dc, stream_state, NULL, enable, enable)) { ret = -EINVAL; @@ -363,7 +358,7 @@ int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name) #if defined(CONFIG_DRM_AMD_SECURE_DISPLAY) /* Reset secure_display when we change crc source from debugfs */ - amdgpu_dm_set_crc_window_default(crtc); + amdgpu_dm_set_crc_window_default(crtc, crtc_state->stream); #endif if (amdgpu_dm_crtc_configure_crc_source(crtc, crtc_state, source)) { diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h index 935adca6f0486aa65797653d1f011143ff3f7594..748e80ef40d0a541433a36c774bb1d6868367805 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h @@ -100,7 +100,7 @@ struct secure_display_context *amdgpu_dm_crtc_secure_display_create_contexts( #else #define amdgpu_dm_crc_window_is_activated(x) #define amdgpu_dm_crtc_handle_crc_window_irq(x) -#define amdgpu_dm_crtc_secure_display_create_contexts() +#define amdgpu_dm_crtc_secure_display_create_contexts(x) #endif #endif /* AMD_DAL_DEV_AMDGPU_DM_AMDGPU_DM_CRC_H_ */ diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c index e3762e806617c11a724e34e3954ecbedfea49352..440fc0869a34ba631c84df42df35318ed37aeabb 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c @@ -146,7 +146,6 @@ static void vblank_control_worker(struct work_struct *work) static inline int dm_set_vblank(struct drm_crtc *crtc, bool enable) { - enum dc_irq_source irq_source; struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc); struct amdgpu_device *adev = drm_to_adev(crtc->dev); struct dm_crtc_state *acrtc_state = to_dm_crtc_state(crtc->state); @@ -169,18 +168,9 @@ static inline int dm_set_vblank(struct drm_crtc *crtc, bool enable) if (rc) return rc; - if (amdgpu_in_reset(adev)) { - irq_source = IRQ_TYPE_VBLANK + acrtc->otg_inst; - /* During gpu-reset we disable and then enable vblank irq, so - * don't use amdgpu_irq_get/put() to avoid refcount change. - */ - if (!dc_interrupt_set(adev->dm.dc, irq_source, enable)) - rc = -EBUSY; - } else { - rc = (enable) - ? amdgpu_irq_get(adev, &adev->crtc_irq, acrtc->crtc_id) - : amdgpu_irq_put(adev, &adev->crtc_irq, acrtc->crtc_id); - } + rc = (enable) + ? amdgpu_irq_get(adev, &adev->crtc_irq, acrtc->crtc_id) + : amdgpu_irq_put(adev, &adev->crtc_irq, acrtc->crtc_id); if (rc) return rc; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c index 827fcb4fb3b3b27ebdf7faf6d6f21f09063552f3..5ea3284b2b77f0ce189e89d279f3b23c4004c4fd 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c @@ -906,6 +906,61 @@ static int amdgpu_current_bpc_show(struct seq_file *m, void *data) } DEFINE_SHOW_ATTRIBUTE(amdgpu_current_bpc); +/* + * Returns the current colorspace for the crtc. + * Example usage: cat /sys/kernel/debug/dri/0/crtc-0/amdgpu_current_colorspace + */ +static int amdgpu_current_colorspace_show(struct seq_file *m, void *data) +{ + struct drm_crtc *crtc = m->private; + struct drm_device *dev = crtc->dev; + struct dm_crtc_state *dm_crtc_state = NULL; + int res = -ENODEV; + + mutex_lock(&dev->mode_config.mutex); + drm_modeset_lock(&crtc->mutex, NULL); + if (crtc->state == NULL) + goto unlock; + + dm_crtc_state = to_dm_crtc_state(crtc->state); + if (dm_crtc_state->stream == NULL) + goto unlock; + + switch (dm_crtc_state->stream->output_color_space) { + case COLOR_SPACE_SRGB: + seq_printf(m, "sRGB"); + break; + case COLOR_SPACE_YCBCR601: + case COLOR_SPACE_YCBCR601_LIMITED: + seq_printf(m, "BT601_YCC"); + break; + case COLOR_SPACE_YCBCR709: + case COLOR_SPACE_YCBCR709_LIMITED: + seq_printf(m, "BT709_YCC"); + break; + case COLOR_SPACE_ADOBERGB: + seq_printf(m, "opRGB"); + break; + case COLOR_SPACE_2020_RGB_FULLRANGE: + seq_printf(m, "BT2020_RGB"); + break; + case COLOR_SPACE_2020_YCBCR: + seq_printf(m, "BT2020_YCC"); + break; + default: + goto unlock; + } + res = 0; + +unlock: + drm_modeset_unlock(&crtc->mutex); + mutex_unlock(&dev->mode_config.mutex); + + return res; +} +DEFINE_SHOW_ATTRIBUTE(amdgpu_current_colorspace); + + /* * Example usage: * Disable dsc passthrough, i.e.,: have dsc decoding at converver, not external RX @@ -1039,88 +1094,6 @@ static ssize_t dp_sdp_message_debugfs_write(struct file *f, const char __user *b return write_size; } -static ssize_t dp_dpcd_address_write(struct file *f, const char __user *buf, - size_t size, loff_t *pos) -{ - int r; - struct amdgpu_dm_connector *connector = file_inode(f)->i_private; - - if (size < sizeof(connector->debugfs_dpcd_address)) - return -EINVAL; - - r = copy_from_user(&connector->debugfs_dpcd_address, - buf, sizeof(connector->debugfs_dpcd_address)); - - return size - r; -} - -static ssize_t dp_dpcd_size_write(struct file *f, const char __user *buf, - size_t size, loff_t *pos) -{ - int r; - struct amdgpu_dm_connector *connector = file_inode(f)->i_private; - - if (size < sizeof(connector->debugfs_dpcd_size)) - return -EINVAL; - - r = copy_from_user(&connector->debugfs_dpcd_size, - buf, sizeof(connector->debugfs_dpcd_size)); - - if (connector->debugfs_dpcd_size > 256) - connector->debugfs_dpcd_size = 0; - - return size - r; -} - -static ssize_t dp_dpcd_data_write(struct file *f, const char __user *buf, - size_t size, loff_t *pos) -{ - int r; - char *data; - struct amdgpu_dm_connector *connector = file_inode(f)->i_private; - struct dc_link *link = connector->dc_link; - uint32_t write_size = connector->debugfs_dpcd_size; - - if (!write_size || size < write_size) - return -EINVAL; - - data = kzalloc(write_size, GFP_KERNEL); - if (!data) - return 0; - - r = copy_from_user(data, buf, write_size); - - dm_helpers_dp_write_dpcd(link->ctx, link, - connector->debugfs_dpcd_address, data, write_size - r); - kfree(data); - return write_size - r; -} - -static ssize_t dp_dpcd_data_read(struct file *f, char __user *buf, - size_t size, loff_t *pos) -{ - int r; - char *data; - struct amdgpu_dm_connector *connector = file_inode(f)->i_private; - struct dc_link *link = connector->dc_link; - uint32_t read_size = connector->debugfs_dpcd_size; - - if (!read_size || size < read_size) - return 0; - - data = kzalloc(read_size, GFP_KERNEL); - if (!data) - return 0; - - dm_helpers_dp_read_dpcd(link->ctx, link, - connector->debugfs_dpcd_address, data, read_size); - - r = copy_to_user(buf, data, read_size); - - kfree(data); - return read_size - r; -} - /* function: Read link's DSC & FEC capabilities * * @@ -2682,25 +2655,6 @@ static const struct file_operations sdp_message_fops = { .llseek = default_llseek }; -static const struct file_operations dp_dpcd_address_debugfs_fops = { - .owner = THIS_MODULE, - .write = dp_dpcd_address_write, - .llseek = default_llseek -}; - -static const struct file_operations dp_dpcd_size_debugfs_fops = { - .owner = THIS_MODULE, - .write = dp_dpcd_size_write, - .llseek = default_llseek -}; - -static const struct file_operations dp_dpcd_data_debugfs_fops = { - .owner = THIS_MODULE, - .read = dp_dpcd_data_read, - .write = dp_dpcd_data_write, - .llseek = default_llseek -}; - static const struct file_operations dp_max_bpc_debugfs_fops = { .owner = THIS_MODULE, .read = dp_max_bpc_read, @@ -2724,9 +2678,6 @@ static const struct { {"test_pattern", &dp_phy_test_pattern_fops}, {"hdcp_sink_capability", &hdcp_sink_capability_fops}, {"sdp_message", &sdp_message_fops}, - {"aux_dpcd_address", &dp_dpcd_address_debugfs_fops}, - {"aux_dpcd_size", &dp_dpcd_size_debugfs_fops}, - {"aux_dpcd_data", &dp_dpcd_data_debugfs_fops}, {"dsc_clock_en", &dp_dsc_clock_en_debugfs_fops}, {"dsc_slice_width", &dp_dsc_slice_width_debugfs_fops}, {"dsc_slice_height", &dp_dsc_slice_height_debugfs_fops}, @@ -2809,6 +2760,32 @@ static int psr_read_residency(void *data, u64 *val) return 0; } +/* read allow_edp_hotplug_detection */ +static int allow_edp_hotplug_detection_get(void *data, u64 *val) +{ + struct amdgpu_dm_connector *aconnector = data; + struct drm_connector *connector = &aconnector->base; + struct drm_device *dev = connector->dev; + struct amdgpu_device *adev = drm_to_adev(dev); + + *val = adev->dm.dc->config.allow_edp_hotplug_detection; + + return 0; +} + +/* set allow_edp_hotplug_detection */ +static int allow_edp_hotplug_detection_set(void *data, u64 val) +{ + struct amdgpu_dm_connector *aconnector = data; + struct drm_connector *connector = &aconnector->base; + struct drm_device *dev = connector->dev; + struct amdgpu_device *adev = drm_to_adev(dev); + + adev->dm.dc->config.allow_edp_hotplug_detection = (uint32_t) val; + + return 0; +} + /* * Set dmcub trace event IRQ enable or disable. * Usage to enable dmcub trace event IRQ: echo 1 > /sys/kernel/debug/dri/0/amdgpu_dm_dmcub_trace_event_en @@ -2847,6 +2824,10 @@ DEFINE_DEBUGFS_ATTRIBUTE(psr_fops, psr_get, NULL, "%llu\n"); DEFINE_DEBUGFS_ATTRIBUTE(psr_residency_fops, psr_read_residency, NULL, "%llu\n"); +DEFINE_DEBUGFS_ATTRIBUTE(allow_edp_hotplug_detection_fops, + allow_edp_hotplug_detection_get, + allow_edp_hotplug_detection_set, "%llu\n"); + DEFINE_SHOW_ATTRIBUTE(current_backlight); DEFINE_SHOW_ATTRIBUTE(target_backlight); @@ -3017,6 +2998,8 @@ void connector_debugfs_init(struct amdgpu_dm_connector *connector) &target_backlight_fops); debugfs_create_file("ilr_setting", 0644, dir, connector, &edp_ilr_debugfs_fops); + debugfs_create_file("allow_edp_hotplug_detection", 0644, dir, connector, + &allow_edp_hotplug_detection_fops); } for (i = 0; i < ARRAY_SIZE(connector_debugfs_entries); i++) { @@ -3025,9 +3008,6 @@ void connector_debugfs_init(struct amdgpu_dm_connector *connector) connector_debugfs_entries[i].fops); } - connector->debugfs_dpcd_address = 0; - connector->debugfs_dpcd_size = 0; - if (connector->base.connector_type == DRM_MODE_CONNECTOR_HDMIA) { for (i = 0; i < ARRAY_SIZE(hdmi_debugfs_entries); i++) { debugfs_create_file(hdmi_debugfs_entries[i].name, @@ -3246,6 +3226,8 @@ void crtc_debugfs_init(struct drm_crtc *crtc) #endif debugfs_create_file("amdgpu_current_bpc", 0644, crtc->debugfs_entry, crtc, &amdgpu_current_bpc_fops); + debugfs_create_file("amdgpu_current_colorspace", 0644, crtc->debugfs_entry, + crtc, &amdgpu_current_colorspace_fops); } /* diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c index c6ce2b7123b795439b01936e18a96406399cfbe4..cd20cfc049969bfbb9b74d9d05863fda8364666a 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c @@ -44,9 +44,6 @@ #include "dm_helpers.h" #include "ddc_service_types.h" -/* MST Dock */ -static const uint8_t SYNAPTICS_DEVICE_ID[] = "SYNA"; - /* dm_helpers_parse_edid_caps * * Parse edid caps @@ -702,6 +699,9 @@ static void apply_synaptics_fifo_reset_wa(struct drm_dp_aux *aux) DC_LOG_DC("Done apply_synaptics_fifo_reset_wa\n"); } +/* MST Dock */ +static const uint8_t SYNAPTICS_DEVICE_ID[] = "SYNA"; + static uint8_t write_dsc_enable_synaptics_non_virtual_dpcd_mst( struct drm_dp_aux *aux, const struct dc_stream_state *stream, @@ -885,10 +885,34 @@ enum dc_edid_status dm_helpers_read_local_edid( DRM_ERROR("EDID err: %d, on connector: %s", edid_status, aconnector->base.name); + if (link->aux_mode) { + union test_request test_request = {0}; + union test_response test_response = {0}; - /* DP Compliance Test 4.2.2.3 */ - if (link->aux_mode) - drm_dp_send_real_edid_checksum(&aconnector->dm_dp_aux.aux, sink->dc_edid.raw_edid[sink->dc_edid.length-1]); + dm_helpers_dp_read_dpcd(ctx, + link, + DP_TEST_REQUEST, + &test_request.raw, + sizeof(union test_request)); + + if (!test_request.bits.EDID_READ) + return edid_status; + + test_response.bits.EDID_CHECKSUM_WRITE = 1; + + dm_helpers_dp_write_dpcd(ctx, + link, + DP_TEST_EDID_CHECKSUM, + &sink->dc_edid.raw_edid[sink->dc_edid.length-1], + 1); + + dm_helpers_dp_write_dpcd(ctx, + link, + DP_TEST_RESPONSE, + &test_response.raw, + sizeof(test_response)); + + } return edid_status; } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index 810ab682f424f94e7f450e82651a18b61016c7b7..46d0a8f57e552b87ef1f8f7159b19af6641b21f2 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -45,8 +45,7 @@ #endif #include "dc/dcn20/dcn20_resource.h" -bool is_timing_changed(struct dc_stream_state *cur_stream, - struct dc_stream_state *new_stream); + #define PEAK_FACTOR_X1000 1006 static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux, @@ -1422,7 +1421,7 @@ int pre_validate_dsc(struct drm_atomic_state *state, struct dc_stream_state *stream = dm_state->context->streams[i]; if (local_dc_state->streams[i] && - is_timing_changed(stream, local_dc_state->streams[i])) { + dc_is_timing_changed(stream, local_dc_state->streams[i])) { DRM_INFO_ONCE("crtc[%d] needs mode_changed\n", i); } else { int ind = find_crtc_index_in_state_by_stream(state, stream); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c b/drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c index c42aa947c969d101767ab829714732974fe5a158..172aa10a8800f3683b49356ca867761cf301bdb3 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/dc_fpu.c @@ -33,6 +33,8 @@ #include #elif defined(CONFIG_ARM64) #include +#elif defined(CONFIG_LOONGARCH) +#include #endif /** @@ -88,7 +90,7 @@ void dc_fpu_begin(const char *function_name, const int line) *pcpu += 1; if (*pcpu == 1) { -#if defined(CONFIG_X86) +#if defined(CONFIG_X86) || defined(CONFIG_LOONGARCH) migrate_disable(); kernel_fpu_begin(); #elif defined(CONFIG_PPC64) @@ -128,7 +130,7 @@ void dc_fpu_end(const char *function_name, const int line) pcpu = get_cpu_ptr(&fpu_recursion_depth); *pcpu -= 1; if (*pcpu <= 0) { -#if defined(CONFIG_X86) +#if defined(CONFIG_X86) || defined(CONFIG_LOONGARCH) kernel_fpu_end(); migrate_enable(); #elif defined(CONFIG_PPC64) diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c index 1ef9e4053bb717fb712cff95c4642a1533c84d4a..90a02d7bd3da3f6e194dafcdc1c3b852c923b51f 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c @@ -123,9 +123,7 @@ static void encoder_control_dmcub( sizeof(cmd.digx_encoder_control.header); cmd.digx_encoder_control.encoder_control.dig.stream_param = *dig; - dc_dmub_srv_cmd_queue(dmcub, &cmd); - dc_dmub_srv_cmd_execute(dmcub); - dc_dmub_srv_wait_idle(dmcub); + dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static enum bp_result encoder_control_digx_v1_5( @@ -261,9 +259,7 @@ static void transmitter_control_dmcub( sizeof(cmd.dig1_transmitter_control.header); cmd.dig1_transmitter_control.transmitter_control.dig = *dig; - dc_dmub_srv_cmd_queue(dmcub, &cmd); - dc_dmub_srv_cmd_execute(dmcub); - dc_dmub_srv_wait_idle(dmcub); + dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static enum bp_result transmitter_control_v1_6( @@ -325,9 +321,7 @@ static void transmitter_control_dmcub_v1_7( sizeof(cmd.dig1_transmitter_control.header); cmd.dig1_transmitter_control.transmitter_control.dig_v1_7 = *dig; - dc_dmub_srv_cmd_queue(dmcub, &cmd); - dc_dmub_srv_cmd_execute(dmcub); - dc_dmub_srv_wait_idle(dmcub); + dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static enum bp_result transmitter_control_v1_7( @@ -435,9 +429,7 @@ static void set_pixel_clock_dmcub( sizeof(cmd.set_pixel_clock.header); cmd.set_pixel_clock.pixel_clock.clk = *clk; - dc_dmub_srv_cmd_queue(dmcub, &cmd); - dc_dmub_srv_cmd_execute(dmcub); - dc_dmub_srv_wait_idle(dmcub); + dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static enum bp_result set_pixel_clock_v7( @@ -804,9 +796,7 @@ static void enable_disp_power_gating_dmcub( sizeof(cmd.enable_disp_power_gating.header); cmd.enable_disp_power_gating.power_gating.pwr = *pwr; - dc_dmub_srv_cmd_queue(dmcub, &cmd); - dc_dmub_srv_cmd_execute(dmcub); - dc_dmub_srv_wait_idle(dmcub); + dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static enum bp_result enable_disp_power_gating_v2_1( @@ -1016,10 +1006,7 @@ static void enable_lvtma_control_dmcub( panel_instance; cmd.lvtma_control.data.bypass_panel_control_wait = bypass_panel_control_wait; - dc_dmub_srv_cmd_queue(dmcub, &cmd); - dc_dmub_srv_cmd_execute(dmcub); - dc_dmub_srv_wait_idle(dmcub); - + dm_execute_dmub_cmd(dmcub->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static enum bp_result enable_lvtma_control( diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c index 934e6423dc1ad213e1da362b6d7e880ea9edd64e..1f36ad8a7de462878c250c2bafaf70894a911fc8 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c @@ -111,12 +111,10 @@ int dce112_set_clock(struct clk_mgr *clk_mgr_base, int requested_clk_khz) bp->funcs->set_dce_clock(bp, &dce_clk_params); - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { - if (clk_mgr_dce->dfs_bypass_disp_clk != actual_clock) - dmcu->funcs->set_psr_wait_loop(dmcu, - actual_clock / 1000 / 7); - } + if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { + if (clk_mgr_dce->dfs_bypass_disp_clk != actual_clock) + dmcu->funcs->set_psr_wait_loop(dmcu, + actual_clock / 1000 / 7); } clk_mgr_dce->dfs_bypass_disp_clk = actual_clock; @@ -153,12 +151,10 @@ int dce112_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_clk_khz) clk_mgr->cur_min_clks_state = DM_PP_CLOCKS_STATE_NOMINAL; - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { - if (clk_mgr->dfs_bypass_disp_clk != actual_clock) - dmcu->funcs->set_psr_wait_loop(dmcu, - actual_clock / 1000 / 7); - } + if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { + if (clk_mgr->dfs_bypass_disp_clk != actual_clock) + dmcu->funcs->set_psr_wait_loop(dmcu, + actual_clock / 1000 / 7); } clk_mgr->dfs_bypass_disp_clk = actual_clock; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c index 450eaead4f20c9acb24ddf9024c8ebcb08e3e645..89b79dd396283c7d3b8017ce582149f10ec8e370 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c @@ -135,12 +135,10 @@ int rv1_vbios_smu_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_di VBIOSSMC_MSG_SetDispclkFreq, khz_to_mhz_ceil(requested_dispclk_khz)); - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { - if (clk_mgr->dfs_bypass_disp_clk != actual_dispclk_set_mhz) - dmcu->funcs->set_psr_wait_loop(dmcu, - actual_dispclk_set_mhz / 7); - } + if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { + if (clk_mgr->dfs_bypass_disp_clk != actual_dispclk_set_mhz) + dmcu->funcs->set_psr_wait_loop(dmcu, + actual_dispclk_set_mhz / 7); } return actual_dispclk_set_mhz * 1000; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c index 650f3b4b562e90eea549935db4ed52a542c0c1bb..c435f7632e8e8b76b868a26fec65d40f938ee6b6 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c @@ -531,6 +531,11 @@ void dcn20_clk_mgr_construct( struct pp_smu_funcs *pp_smu, struct dccg *dccg) { + int dprefclk_did; + int target_div; + uint32_t pll_req_reg; + struct fixed31_32 pll_req; + clk_mgr->base.ctx = ctx; clk_mgr->pp_smu = pp_smu; clk_mgr->base.funcs = &dcn2_funcs; @@ -547,42 +552,34 @@ void dcn20_clk_mgr_construct( clk_mgr->base.dprefclk_khz = 700000; // 700 MHz planned if VCO is 3.85 GHz, will be retrieved - if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) { - dcn2_funcs.update_clocks = dcn2_update_clocks_fpga; - clk_mgr->base.dentist_vco_freq_khz = 3850000; + /* DFS Slice 2 should be used for DPREFCLK */ + dprefclk_did = REG_READ(CLK3_CLK2_DFS_CNTL); + /* Convert DPREFCLK DFS Slice DID to actual divider */ + target_div = dentist_get_divider_from_did(dprefclk_did); + /* get FbMult value */ + pll_req_reg = REG_READ(CLK3_CLK_PLL_REQ); - } else { - /* DFS Slice 2 should be used for DPREFCLK */ - int dprefclk_did = REG_READ(CLK3_CLK2_DFS_CNTL); - /* Convert DPREFCLK DFS Slice DID to actual divider*/ - int target_div = dentist_get_divider_from_did(dprefclk_did); - - /* get FbMult value */ - uint32_t pll_req_reg = REG_READ(CLK3_CLK_PLL_REQ); - struct fixed31_32 pll_req; - - /* set up a fixed-point number - * this works because the int part is on the right edge of the register - * and the frac part is on the left edge - */ + /* set up a fixed-point number + * this works because the int part is on the right edge of the register + * and the frac part is on the left edge + */ - pll_req = dc_fixpt_from_int(pll_req_reg & clk_mgr->clk_mgr_mask->FbMult_int); - pll_req.value |= pll_req_reg & clk_mgr->clk_mgr_mask->FbMult_frac; + pll_req = dc_fixpt_from_int(pll_req_reg & clk_mgr->clk_mgr_mask->FbMult_int); + pll_req.value |= pll_req_reg & clk_mgr->clk_mgr_mask->FbMult_frac; - /* multiply by REFCLK period */ - pll_req = dc_fixpt_mul_int(pll_req, 100000); + /* multiply by REFCLK period */ + pll_req = dc_fixpt_mul_int(pll_req, 100000); - /* integer part is now VCO frequency in kHz */ - clk_mgr->base.dentist_vco_freq_khz = dc_fixpt_floor(pll_req); + /* integer part is now VCO frequency in kHz */ + clk_mgr->base.dentist_vco_freq_khz = dc_fixpt_floor(pll_req); - /* in case we don't get a value from the register, use default */ - if (clk_mgr->base.dentist_vco_freq_khz == 0) - clk_mgr->base.dentist_vco_freq_khz = 3850000; + /* in case we don't get a value from the register, use default */ + if (clk_mgr->base.dentist_vco_freq_khz == 0) + clk_mgr->base.dentist_vco_freq_khz = 3850000; - /* Calculate the DPREFCLK in kHz.*/ - clk_mgr->base.dprefclk_khz = (DENTIST_DIVIDER_RANGE_SCALE_FACTOR - * clk_mgr->base.dentist_vco_freq_khz) / target_div; - } + /* Calculate the DPREFCLK in kHz.*/ + clk_mgr->base.dprefclk_khz = (DENTIST_DIVIDER_RANGE_SCALE_FACTOR + * clk_mgr->base.dentist_vco_freq_khz) / target_div; //Integrated_info table does not exist on dGPU projects so should not be referenced //anywhere in code for dGPUs. //Also there is no plan for now that DFS BYPASS will be used on NV10/12/14. @@ -590,4 +587,3 @@ void dcn20_clk_mgr_construct( dce_clock_read_ss_info(clk_mgr); } - diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn201/dcn201_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn201/dcn201_clk_mgr.c index 811720749faf8032807891b8799b52937b79bc03..694fe4271b4d9f502946fd13ebc8c66f14a6aa94 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn201/dcn201_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn201/dcn201_clk_mgr.c @@ -190,23 +190,17 @@ void dcn201_clk_mgr_construct(struct dc_context *ctx, clk_mgr->dprefclk_ss_divider = 1000; clk_mgr->ss_on_dprefclk = false; - if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) { - dcn201_funcs.update_clocks = dcn2_update_clocks_fpga; - clk_mgr->base.dprefclk_khz = 600000; - clk_mgr->base.dentist_vco_freq_khz = 3000000; - } else { - clk_mgr->base.dprefclk_khz = REG_READ(CLK4_CLK2_CURRENT_CNT); - clk_mgr->base.dprefclk_khz *= 100; + clk_mgr->base.dprefclk_khz = REG_READ(CLK4_CLK2_CURRENT_CNT); + clk_mgr->base.dprefclk_khz *= 100; - if (clk_mgr->base.dprefclk_khz == 0) - clk_mgr->base.dprefclk_khz = 600000; + if (clk_mgr->base.dprefclk_khz == 0) + clk_mgr->base.dprefclk_khz = 600000; - REG_GET(CLK4_CLK_PLL_REQ, FbMult_int, &clk_mgr->base.dentist_vco_freq_khz); - clk_mgr->base.dentist_vco_freq_khz *= 100000; + REG_GET(CLK4_CLK_PLL_REQ, FbMult_int, &clk_mgr->base.dentist_vco_freq_khz); + clk_mgr->base.dentist_vco_freq_khz *= 100000; - if (clk_mgr->base.dentist_vco_freq_khz == 0) - clk_mgr->base.dentist_vco_freq_khz = 3000000; - } + if (clk_mgr->base.dentist_vco_freq_khz == 0) + clk_mgr->base.dentist_vco_freq_khz = 3000000; if (!debug->disable_dfs_bypass && bp->integrated_info) if (bp->integrated_info->gpu_cap_info & DFS_BYPASS_ENABLE) diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c index bd9fd0b54f46ae994b3d0bc7e1322cb32afed8b4..0c6a4ab72b1d298e00a856c640e936fac2f89a92 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c @@ -705,6 +705,7 @@ void rn_clk_mgr_construct( struct dpm_clocks clock_table = { 0 }; enum pp_smu_status status = 0; int is_green_sardine = 0; + struct clk_log_info log_info = {0}; #if defined(CONFIG_DRM_AMD_DC_FP) is_green_sardine = ASICREV_IS_GREEN_SARDINE(ctx->asic_id.hw_internal_rev); @@ -725,48 +726,41 @@ void rn_clk_mgr_construct( clk_mgr->smu_ver = rn_vbios_smu_get_smu_version(clk_mgr); - if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) { - dcn21_funcs.update_clocks = dcn2_update_clocks_fpga; + clk_mgr->periodic_retraining_disabled = rn_vbios_smu_is_periodic_retraining_disabled(clk_mgr); + + /* SMU Version 55.51.0 and up no longer have an issue + * that needs to limit minimum dispclk */ + if (clk_mgr->smu_ver >= SMU_VER_55_51_0) + debug->min_disp_clk_khz = 0; + + /* TODO: Check we get what we expect during bringup */ + clk_mgr->base.dentist_vco_freq_khz = get_vco_frequency_from_reg(clk_mgr); + + /* in case we don't get a value from the register, use default */ + if (clk_mgr->base.dentist_vco_freq_khz == 0) clk_mgr->base.dentist_vco_freq_khz = 3600000; - } else { - struct clk_log_info log_info = {0}; - - clk_mgr->periodic_retraining_disabled = rn_vbios_smu_is_periodic_retraining_disabled(clk_mgr); - - /* SMU Version 55.51.0 and up no longer have an issue - * that needs to limit minimum dispclk */ - if (clk_mgr->smu_ver >= SMU_VER_55_51_0) - debug->min_disp_clk_khz = 0; - - /* TODO: Check we get what we expect during bringup */ - clk_mgr->base.dentist_vco_freq_khz = get_vco_frequency_from_reg(clk_mgr); - - /* in case we don't get a value from the register, use default */ - if (clk_mgr->base.dentist_vco_freq_khz == 0) - clk_mgr->base.dentist_vco_freq_khz = 3600000; - - if (ctx->dc_bios->integrated_info->memory_type == LpDdr4MemType) { - if (clk_mgr->periodic_retraining_disabled) { - rn_bw_params.wm_table = lpddr4_wm_table_with_disabled_ppt; - } else { - if (is_green_sardine) - rn_bw_params.wm_table = lpddr4_wm_table_gs; - else - rn_bw_params.wm_table = lpddr4_wm_table_rn; - } + + if (ctx->dc_bios->integrated_info->memory_type == LpDdr4MemType) { + if (clk_mgr->periodic_retraining_disabled) { + rn_bw_params.wm_table = lpddr4_wm_table_with_disabled_ppt; } else { if (is_green_sardine) - rn_bw_params.wm_table = ddr4_wm_table_gs; - else { - if (ctx->dc->config.is_single_rank_dimm) - rn_bw_params.wm_table = ddr4_1R_wm_table_rn; - else - rn_bw_params.wm_table = ddr4_wm_table_rn; - } + rn_bw_params.wm_table = lpddr4_wm_table_gs; + else + rn_bw_params.wm_table = lpddr4_wm_table_rn; + } + } else { + if (is_green_sardine) + rn_bw_params.wm_table = ddr4_wm_table_gs; + else { + if (ctx->dc->config.is_single_rank_dimm) + rn_bw_params.wm_table = ddr4_1R_wm_table_rn; + else + rn_bw_params.wm_table = ddr4_wm_table_rn; } - /* Saved clocks configured at boot for debug purposes */ - rn_dump_clk_registers(&clk_mgr->base.boot_snapshot, &clk_mgr->base, &log_info); } + /* Saved clocks configured at boot for debug purposes */ + rn_dump_clk_registers(&clk_mgr->base.boot_snapshot, &clk_mgr->base, &log_info); clk_mgr->base.dprefclk_khz = 600000; dce_clock_read_ss_info(clk_mgr); @@ -786,9 +780,8 @@ void rn_clk_mgr_construct( } } - if (!IS_FPGA_MAXIMUS_DC(ctx->dce_environment) && clk_mgr->smu_ver >= 0x00371500) { - /* enable powerfeatures when displaycount goes to 0 */ + /* enable powerfeatures when displaycount goes to 0 */ + if (clk_mgr->smu_ver >= 0x00371500) rn_vbios_smu_enable_48mhz_tmdp_refclk_pwrdwn(clk_mgr, !debug->disable_48mhz_pwrdwn); - } } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c index 27fbe906682f9f8ebc1ac770d7981ea8c8e7a45f..8c9d45e5b13b8399984cf35c7b22e486d74d4552 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c @@ -147,17 +147,14 @@ int rn_vbios_smu_set_dispclk(struct clk_mgr_internal *clk_mgr, int requested_dis VBIOSSMC_MSG_SetDispclkFreq, khz_to_mhz_ceil(requested_dispclk_khz)); - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { - if (clk_mgr->dfs_bypass_disp_clk != actual_dispclk_set_mhz) - dmcu->funcs->set_psr_wait_loop(dmcu, - actual_dispclk_set_mhz / 7); - } + if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { + if (clk_mgr->dfs_bypass_disp_clk != actual_dispclk_set_mhz) + dmcu->funcs->set_psr_wait_loop(dmcu, + actual_dispclk_set_mhz / 7); } // pmfw always set clock more than or equal requested clock - if (!IS_DIAG_DC(dc->ctx->dce_environment)) - ASSERT(actual_dispclk_set_mhz >= khz_to_mhz_ceil(requested_dispclk_khz)); + ASSERT(actual_dispclk_set_mhz >= khz_to_mhz_ceil(requested_dispclk_khz)); return actual_dispclk_set_mhz * 1000; } @@ -221,15 +218,13 @@ void rn_vbios_smu_set_phyclk(struct clk_mgr_internal *clk_mgr, int requested_phy int rn_vbios_smu_set_dppclk(struct clk_mgr_internal *clk_mgr, int requested_dpp_khz) { int actual_dppclk_set_mhz = -1; - struct dc *dc = clk_mgr->base.ctx->dc; actual_dppclk_set_mhz = rn_vbios_smu_send_msg_with_param( clk_mgr, VBIOSSMC_MSG_SetDppclkFreq, khz_to_mhz_ceil(requested_dpp_khz)); - if (!IS_DIAG_DC(dc->ctx->dce_environment)) - ASSERT(actual_dppclk_set_mhz >= khz_to_mhz_ceil(requested_dpp_khz)); + ASSERT(actual_dppclk_set_mhz >= khz_to_mhz_ceil(requested_dpp_khz)); return actual_dppclk_set_mhz * 1000; } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c index 694a9d3d92aee48715f6c5d843d3c378cabd84e8..3271c8c7905ddc4bf9a7e800b9fcef449e49dae6 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c @@ -206,7 +206,6 @@ static void dcn3_update_clocks(struct clk_mgr *clk_mgr_base, bool force_reset = false; bool update_uclk = false; bool p_state_change_support; - int total_plane_count; if (dc->work_arounds.skip_clock_update || !clk_mgr->smu_present) return; @@ -247,8 +246,7 @@ static void dcn3_update_clocks(struct clk_mgr *clk_mgr_base, clk_mgr_base->clks.socclk_khz = new_clocks->socclk_khz; clk_mgr_base->clks.prev_p_state_change_support = clk_mgr_base->clks.p_state_change_support; - total_plane_count = clk_mgr_helper_get_active_plane_cnt(dc, context); - p_state_change_support = new_clocks->p_state_change_support || (total_plane_count == 0); + p_state_change_support = new_clocks->p_state_change_support; // invalidate the current P-State forced min in certain dc_mode_softmax situations if (dc->clk_mgr->dc_mode_softmax_enabled && safe_to_lower && !p_state_change_support) { @@ -523,6 +521,8 @@ void dcn3_clk_mgr_construct( struct pp_smu_funcs *pp_smu, struct dccg *dccg) { + struct clk_state_registers_and_bypass s = { 0 }; + clk_mgr->base.ctx = ctx; clk_mgr->base.funcs = &dcn3_funcs; clk_mgr->regs = &clk_mgr_regs; @@ -539,27 +539,19 @@ void dcn3_clk_mgr_construct( clk_mgr->base.dprefclk_khz = 730000; // 700 MHz planned if VCO is 3.85 GHz, will be retrieved - if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) { - clk_mgr->base.funcs = &dcn3_fpga_funcs; - clk_mgr->base.dentist_vco_freq_khz = 3650000; - - } else { - struct clk_state_registers_and_bypass s = { 0 }; + /* integer part is now VCO frequency in kHz */ + clk_mgr->base.dentist_vco_freq_khz = dcn30_get_vco_frequency_from_reg(clk_mgr); - /* integer part is now VCO frequency in kHz */ - clk_mgr->base.dentist_vco_freq_khz = dcn30_get_vco_frequency_from_reg(clk_mgr); - - /* in case we don't get a value from the register, use default */ - if (clk_mgr->base.dentist_vco_freq_khz == 0) - clk_mgr->base.dentist_vco_freq_khz = 3650000; - /* Convert dprefclk units from MHz to KHz */ - /* Value already divided by 10, some resolution lost */ + /* in case we don't get a value from the register, use default */ + if (clk_mgr->base.dentist_vco_freq_khz == 0) + clk_mgr->base.dentist_vco_freq_khz = 3650000; + /* Convert dprefclk units from MHz to KHz */ + /* Value already divided by 10, some resolution lost */ - /*TODO: uncomment assert once dcn3_dump_clk_registers is implemented */ - //ASSERT(s.dprefclk != 0); - if (s.dprefclk != 0) - clk_mgr->base.dprefclk_khz = s.dprefclk * 1000; - } + /*TODO: uncomment assert once dcn3_dump_clk_registers is implemented */ + //ASSERT(s.dprefclk != 0); + if (s.dprefclk != 0) + clk_mgr->base.dprefclk_khz = s.dprefclk * 1000; clk_mgr->dfs_bypass_enabled = false; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c index 1fbf1c105dc12da3e6b934e1faab8157e335d19c..bdbf183066981aefdcf5cf3dfd6bee85d2bbfcbc 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c @@ -312,6 +312,9 @@ void dcn30_smu_set_display_refresh_from_mall(struct clk_mgr_internal *clk_mgr, b /* bits 8:7 for cache timer scale, bits 6:1 for cache timer delay, bit 0 = 1 for enable, = 0 for disable */ uint32_t param = (cache_timer_scale << 7) | (cache_timer_delay << 1) | (enable ? 1 : 0); + smu_print("SMU Set display refresh from mall: enable = %d, cache_timer_delay = %d, cache_timer_scale = %d\n", + enable, cache_timer_delay, cache_timer_scale); + dcn30_smu_send_msg_with_param(clk_mgr, DALSMC_MSG_SetDisplayRefreshFromMall, param, NULL); } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c index 01383aac6b4190f84e12765e6782c209b8309ed3..a5489fe6875f453149d622d59e9b6417b4db616c 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c @@ -117,7 +117,7 @@ static void vg_update_clocks(struct clk_mgr *clk_mgr_base, display_count = vg_get_active_display_cnt_wa(dc, context); /* if we can go lower, go lower */ - if (display_count == 0 && !IS_DIAG_DC(dc->ctx->dce_environment)) { + if (display_count == 0) { union display_idle_optimization_u idle_info = { 0 }; idle_info.idle_info.df_request_disabled = 1; @@ -151,10 +151,8 @@ static void vg_update_clocks(struct clk_mgr *clk_mgr_base, } // workaround: Limit dppclk to 100Mhz to avoid lower eDP panel switch to plus 4K monitor underflow. - if (!IS_DIAG_DC(dc->ctx->dce_environment)) { - if (new_clocks->dppclk_khz < 100000) - new_clocks->dppclk_khz = 100000; - } + if (new_clocks->dppclk_khz < 100000) + new_clocks->dppclk_khz = 100000; if (should_set_clock(safe_to_lower, new_clocks->dppclk_khz, clk_mgr->base.clks.dppclk_khz)) { if (clk_mgr->base.clks.dppclk_khz > new_clocks->dppclk_khz) @@ -664,6 +662,7 @@ void vg_clk_mgr_construct( struct dccg *dccg) { struct smu_dpm_clks smu_dpm_clks = { 0 }; + struct clk_log_info log_info = {0}; clk_mgr->base.base.ctx = ctx; clk_mgr->base.base.funcs = &vg_funcs; @@ -703,32 +702,25 @@ void vg_clk_mgr_construct( ASSERT(smu_dpm_clks.dpm_clks); - if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) { - vg_funcs.update_clocks = dcn2_update_clocks_fpga; - clk_mgr->base.base.dentist_vco_freq_khz = 3600000; - } else { - struct clk_log_info log_info = {0}; + clk_mgr->base.smu_ver = dcn301_smu_get_smu_version(&clk_mgr->base); - clk_mgr->base.smu_ver = dcn301_smu_get_smu_version(&clk_mgr->base); + if (clk_mgr->base.smu_ver) + clk_mgr->base.smu_present = true; - if (clk_mgr->base.smu_ver) - clk_mgr->base.smu_present = true; + /* TODO: Check we get what we expect during bringup */ + clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base); - /* TODO: Check we get what we expect during bringup */ - clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base); - - /* in case we don't get a value from the register, use default */ - if (clk_mgr->base.base.dentist_vco_freq_khz == 0) - clk_mgr->base.base.dentist_vco_freq_khz = 3600000; + /* in case we don't get a value from the register, use default */ + if (clk_mgr->base.base.dentist_vco_freq_khz == 0) + clk_mgr->base.base.dentist_vco_freq_khz = 3600000; - if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) { - vg_bw_params.wm_table = lpddr5_wm_table; - } else { - vg_bw_params.wm_table = ddr4_wm_table; - } - /* Saved clocks configured at boot for debug purposes */ - vg_dump_clk_registers(&clk_mgr->base.base.boot_snapshot, &clk_mgr->base.base, &log_info); + if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) { + vg_bw_params.wm_table = lpddr5_wm_table; + } else { + vg_bw_params.wm_table = ddr4_wm_table; } + /* Saved clocks configured at boot for debug purposes */ + vg_dump_clk_registers(&clk_mgr->base.base.boot_snapshot, &clk_mgr->base.base, &log_info); clk_mgr->base.base.dprefclk_khz = 600000; dce_clock_read_ss_info(&clk_mgr->base); @@ -746,12 +738,6 @@ void vg_clk_mgr_construct( if (smu_dpm_clks.dpm_clks && smu_dpm_clks.mc_address.quad_part != 0) dm_helpers_free_gpu_mem(clk_mgr->base.base.ctx, DC_MEM_ALLOC_TYPE_FRAME_BUFFER, smu_dpm_clks.dpm_clks); -/* - if (!IS_FPGA_MAXIMUS_DC(ctx->dce_environment) && clk_mgr->base.smu_ver) { - enable powerfeatures when displaycount goes to 0 - dcn301_smu_enable_phy_refclk_pwrdwn(clk_mgr, !debug->disable_48mhz_pwrdwn); - } -*/ } void vg_clk_mgr_destroy(struct clk_mgr_internal *clk_mgr_int) diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c index f9e2e0c3095e7c652fae5c70823a8531a15218e3..7ccd96959256d06aa9ec1f70e74784362c103e4e 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c @@ -205,10 +205,8 @@ void dcn31_update_clocks(struct clk_mgr *clk_mgr_base, } // workaround: Limit dppclk to 100Mhz to avoid lower eDP panel switch to plus 4K monitor underflow. - if (!IS_DIAG_DC(dc->ctx->dce_environment)) { - if (new_clocks->dppclk_khz < 100000) - new_clocks->dppclk_khz = 100000; - } + if (new_clocks->dppclk_khz < 100000) + new_clocks->dppclk_khz = 100000; if (should_set_clock(safe_to_lower, new_clocks->dppclk_khz, clk_mgr->base.clks.dppclk_khz)) { if (clk_mgr->base.clks.dppclk_khz > new_clocks->dppclk_khz) @@ -250,9 +248,7 @@ void dcn31_update_clocks(struct clk_mgr *clk_mgr_base, cmd.notify_clocks.clocks.dispclk_khz = clk_mgr_base->clks.dispclk_khz; cmd.notify_clocks.clocks.dppclk_khz = clk_mgr_base->clks.dppclk_khz; - dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv); - dc_dmub_srv_wait_idle(dc->ctx->dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static int get_vco_frequency_from_reg(struct clk_mgr_internal *clk_mgr) @@ -674,6 +670,7 @@ void dcn31_clk_mgr_construct( struct dccg *dccg) { struct dcn31_smu_dpm_clks smu_dpm_clks = { 0 }; + struct clk_log_info log_info = {0}; clk_mgr->base.base.ctx = ctx; clk_mgr->base.base.funcs = &dcn31_funcs; @@ -713,29 +710,22 @@ void dcn31_clk_mgr_construct( ASSERT(smu_dpm_clks.dpm_clks); - if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) { - clk_mgr->base.base.funcs = &dcn3_fpga_funcs; - } else { - struct clk_log_info log_info = {0}; - - clk_mgr->base.smu_ver = dcn31_smu_get_smu_version(&clk_mgr->base); + clk_mgr->base.smu_ver = dcn31_smu_get_smu_version(&clk_mgr->base); - if (clk_mgr->base.smu_ver) - clk_mgr->base.smu_present = true; + if (clk_mgr->base.smu_ver) + clk_mgr->base.smu_present = true; - /* TODO: Check we get what we expect during bringup */ - clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base); - - if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) { - dcn31_bw_params.wm_table = lpddr5_wm_table; - } else { - dcn31_bw_params.wm_table = ddr5_wm_table; - } - /* Saved clocks configured at boot for debug purposes */ - dcn31_dump_clk_registers(&clk_mgr->base.base.boot_snapshot, - &clk_mgr->base.base, &log_info); + /* TODO: Check we get what we expect during bringup */ + clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base); + if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) { + dcn31_bw_params.wm_table = lpddr5_wm_table; + } else { + dcn31_bw_params.wm_table = ddr5_wm_table; } + /* Saved clocks configured at boot for debug purposes */ + dcn31_dump_clk_registers(&clk_mgr->base.base.boot_snapshot, + &clk_mgr->base.base, &log_info); clk_mgr->base.base.dprefclk_khz = 600000; clk_mgr->base.base.clks.ref_dtbclk_khz = 600000; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c index 0827c7df28557920d6bd34adf4aa1b7aafef071f..32279c5db72483f24106a5b282622903baaaa6ac 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c @@ -130,7 +130,7 @@ static int dcn31_smu_send_msg_with_param(struct clk_mgr_internal *clk_mgr, if (result == VBIOSSMC_Result_Failed) { if (msg_id == VBIOSSMC_MSG_TransferTableDram2Smu && param == TABLE_WATERMARKS) - DC_LOG_WARNING("Watermarks table not configured properly by SMU"); + DC_LOG_DEBUG("Watermarks table not configured properly by SMU"); else ASSERT(0); REG_WRITE(MP1_SMN_C2PMSG_91, VBIOSSMC_Result_OK); diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c index 5cb44f838bde5ba8af586c08871925b4e8c6ffcc..2f7c8996b19d33b4785fa721de3f4d82f1421fa7 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c @@ -241,10 +241,8 @@ void dcn314_update_clocks(struct clk_mgr *clk_mgr_base, } // workaround: Limit dppclk to 100Mhz to avoid lower eDP panel switch to plus 4K monitor underflow. - if (!IS_DIAG_DC(dc->ctx->dce_environment)) { - if (new_clocks->dppclk_khz < 100000) - new_clocks->dppclk_khz = 100000; - } + if (new_clocks->dppclk_khz < 100000) + new_clocks->dppclk_khz = 100000; if (should_set_clock(safe_to_lower, new_clocks->dppclk_khz, clk_mgr->base.clks.dppclk_khz)) { if (clk_mgr->base.clks.dppclk_khz > new_clocks->dppclk_khz) @@ -286,9 +284,7 @@ void dcn314_update_clocks(struct clk_mgr *clk_mgr_base, cmd.notify_clocks.clocks.dispclk_khz = clk_mgr_base->clks.dispclk_khz; cmd.notify_clocks.clocks.dppclk_khz = clk_mgr_base->clks.dppclk_khz; - dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv); - dc_dmub_srv_wait_idle(dc->ctx->dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static int get_vco_frequency_from_reg(struct clk_mgr_internal *clk_mgr) @@ -405,32 +401,32 @@ static struct wm_table lpddr5_wm_table = { .wm_inst = WM_A, .wm_type = WM_TYPE_PSTATE_CHG, .pstate_latency_us = 11.65333, - .sr_exit_time_us = 16.5, - .sr_enter_plus_exit_time_us = 18.5, + .sr_exit_time_us = 30.0, + .sr_enter_plus_exit_time_us = 32.0, .valid = true, }, { .wm_inst = WM_B, .wm_type = WM_TYPE_PSTATE_CHG, .pstate_latency_us = 11.65333, - .sr_exit_time_us = 16.5, - .sr_enter_plus_exit_time_us = 18.5, + .sr_exit_time_us = 30.0, + .sr_enter_plus_exit_time_us = 32.0, .valid = true, }, { .wm_inst = WM_C, .wm_type = WM_TYPE_PSTATE_CHG, .pstate_latency_us = 11.65333, - .sr_exit_time_us = 16.5, - .sr_enter_plus_exit_time_us = 18.5, + .sr_exit_time_us = 30.0, + .sr_enter_plus_exit_time_us = 32.0, .valid = true, }, { .wm_inst = WM_D, .wm_type = WM_TYPE_PSTATE_CHG, .pstate_latency_us = 11.65333, - .sr_exit_time_us = 16.5, - .sr_enter_plus_exit_time_us = 18.5, + .sr_exit_time_us = 30.0, + .sr_enter_plus_exit_time_us = 32.0, .valid = true, }, } @@ -726,6 +722,7 @@ void dcn314_clk_mgr_construct( struct dccg *dccg) { struct dcn314_smu_dpm_clks smu_dpm_clks = { 0 }; + struct clk_log_info log_info = {0}; clk_mgr->base.base.ctx = ctx; clk_mgr->base.base.funcs = &dcn314_funcs; @@ -765,29 +762,22 @@ void dcn314_clk_mgr_construct( ASSERT(smu_dpm_clks.dpm_clks); - if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) { - clk_mgr->base.base.funcs = &dcn3_fpga_funcs; - } else { - struct clk_log_info log_info = {0}; - - clk_mgr->base.smu_ver = dcn314_smu_get_smu_version(&clk_mgr->base); + clk_mgr->base.smu_ver = dcn314_smu_get_smu_version(&clk_mgr->base); - if (clk_mgr->base.smu_ver) - clk_mgr->base.smu_present = true; + if (clk_mgr->base.smu_ver) + clk_mgr->base.smu_present = true; - /* TODO: Check we get what we expect during bringup */ - clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base); + /* TODO: Check we get what we expect during bringup */ + clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base); - if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) - dcn314_bw_params.wm_table = lpddr5_wm_table; - else - dcn314_bw_params.wm_table = ddr5_wm_table; + if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) + dcn314_bw_params.wm_table = lpddr5_wm_table; + else + dcn314_bw_params.wm_table = ddr5_wm_table; - /* Saved clocks configured at boot for debug purposes */ - dcn314_dump_clk_registers(&clk_mgr->base.base.boot_snapshot, - &clk_mgr->base.base, &log_info); - - } + /* Saved clocks configured at boot for debug purposes */ + dcn314_dump_clk_registers(&clk_mgr->base.base.boot_snapshot, + &clk_mgr->base.base, &log_info); clk_mgr->base.base.dprefclk_khz = 600000; clk_mgr->base.base.clks.ref_dtbclk_khz = 600000; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_smu.c index 0765334f0825985811d8a3cf9997034c0db0c291..07baa10a86473d64ed06efbb5d0b3ecf92af8aa2 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_smu.c @@ -145,7 +145,7 @@ static int dcn314_smu_send_msg_with_param(struct clk_mgr_internal *clk_mgr, if (result == VBIOSSMC_Result_Failed) { if (msg_id == VBIOSSMC_MSG_TransferTableDram2Smu && param == TABLE_WATERMARKS) - DC_LOG_WARNING("Watermarks table not configured properly by SMU"); + DC_LOG_DEBUG("Watermarks table not configured properly by SMU"); else if (msg_id == VBIOSSMC_MSG_SetHardMinDcfclkByFreq || msg_id == VBIOSSMC_MSG_SetMinDeepSleepDcfclk) DC_LOG_WARNING("DCFCLK_DPM is not enabled by BIOS"); diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c index b737cbc468f55237bc1336a56f495066ed0eec8e..b2c4f97afc8b4c0852199e0ce9209646c7ef8d7b 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c @@ -184,12 +184,10 @@ static void dcn315_update_clocks(struct clk_mgr *clk_mgr_base, } // workaround: Limit dppclk to 100Mhz to avoid lower eDP panel switch to plus 4K monitor underflow. - if (!IS_DIAG_DC(dc->ctx->dce_environment)) { - if (new_clocks->dppclk_khz < MIN_DPP_DISP_CLK) - new_clocks->dppclk_khz = MIN_DPP_DISP_CLK; - if (new_clocks->dispclk_khz < MIN_DPP_DISP_CLK) - new_clocks->dispclk_khz = MIN_DPP_DISP_CLK; - } + if (new_clocks->dppclk_khz < MIN_DPP_DISP_CLK) + new_clocks->dppclk_khz = MIN_DPP_DISP_CLK; + if (new_clocks->dispclk_khz < MIN_DPP_DISP_CLK) + new_clocks->dispclk_khz = MIN_DPP_DISP_CLK; if (should_set_clock(safe_to_lower, new_clocks->dppclk_khz, clk_mgr->base.clks.dppclk_khz)) { if (clk_mgr->base.clks.dppclk_khz > new_clocks->dppclk_khz) @@ -234,9 +232,7 @@ static void dcn315_update_clocks(struct clk_mgr *clk_mgr_base, cmd.notify_clocks.clocks.dispclk_khz = clk_mgr_base->clks.dispclk_khz; cmd.notify_clocks.clocks.dppclk_khz = clk_mgr_base->clks.dppclk_khz; - dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv); - dc_dmub_srv_wait_idle(dc->ctx->dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static void dcn315_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass, @@ -602,6 +598,7 @@ void dcn315_clk_mgr_construct( struct dccg *dccg) { struct dcn315_smu_dpm_clks smu_dpm_clks = { 0 }; + struct clk_log_info log_info = {0}; clk_mgr->base.base.ctx = ctx; clk_mgr->base.base.funcs = &dcn315_funcs; @@ -641,26 +638,19 @@ void dcn315_clk_mgr_construct( ASSERT(smu_dpm_clks.dpm_clks); - if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) { - clk_mgr->base.base.funcs = &dcn3_fpga_funcs; - } else { - struct clk_log_info log_info = {0}; - - clk_mgr->base.smu_ver = dcn315_smu_get_smu_version(&clk_mgr->base); + clk_mgr->base.smu_ver = dcn315_smu_get_smu_version(&clk_mgr->base); - if (clk_mgr->base.smu_ver > 0) - clk_mgr->base.smu_present = true; - - if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) { - dcn315_bw_params.wm_table = lpddr5_wm_table; - } else { - dcn315_bw_params.wm_table = ddr5_wm_table; - } - /* Saved clocks configured at boot for debug purposes */ - dcn315_dump_clk_registers(&clk_mgr->base.base.boot_snapshot, - &clk_mgr->base.base, &log_info); + if (clk_mgr->base.smu_ver > 0) + clk_mgr->base.smu_present = true; + if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) { + dcn315_bw_params.wm_table = lpddr5_wm_table; + } else { + dcn315_bw_params.wm_table = ddr5_wm_table; } + /* Saved clocks configured at boot for debug purposes */ + dcn315_dump_clk_registers(&clk_mgr->base.base.boot_snapshot, + &clk_mgr->base.base, &log_info); clk_mgr->base.base.dprefclk_khz = 600000; clk_mgr->base.base.dprefclk_khz = dcn315_smu_get_dpref_clk(&clk_mgr->base); diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c index 93db4dbee713e5f1de73571777b37053d1a8f386..d7de756301cf7346e19b84a330a46a88de28ebf9 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c @@ -207,12 +207,10 @@ static void dcn316_update_clocks(struct clk_mgr *clk_mgr_base, } // workaround: Limit dppclk to 100Mhz to avoid lower eDP panel switch to plus 4K monitor underflow. - if (!IS_DIAG_DC(dc->ctx->dce_environment)) { - if (new_clocks->dppclk_khz < 100000) - new_clocks->dppclk_khz = 100000; - if (new_clocks->dispclk_khz < 100000) - new_clocks->dispclk_khz = 100000; - } + if (new_clocks->dppclk_khz < 100000) + new_clocks->dppclk_khz = 100000; + if (new_clocks->dispclk_khz < 100000) + new_clocks->dispclk_khz = 100000; if (should_set_clock(safe_to_lower, new_clocks->dppclk_khz, clk_mgr->base.clks.dppclk_khz)) { if (clk_mgr->base.clks.dppclk_khz > new_clocks->dppclk_khz) @@ -254,9 +252,7 @@ static void dcn316_update_clocks(struct clk_mgr *clk_mgr_base, cmd.notify_clocks.clocks.dispclk_khz = clk_mgr_base->clks.dispclk_khz; cmd.notify_clocks.clocks.dppclk_khz = clk_mgr_base->clks.dppclk_khz; - dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv); - dc_dmub_srv_wait_idle(dc->ctx->dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static void dcn316_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass, @@ -618,6 +614,7 @@ void dcn316_clk_mgr_construct( struct dccg *dccg) { struct dcn316_smu_dpm_clks smu_dpm_clks = { 0 }; + struct clk_log_info log_info = {0}; clk_mgr->base.base.ctx = ctx; clk_mgr->base.base.funcs = &dcn316_funcs; @@ -657,35 +654,27 @@ void dcn316_clk_mgr_construct( ASSERT(smu_dpm_clks.dpm_clks); - if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) { - clk_mgr->base.base.funcs = &dcn3_fpga_funcs; - clk_mgr->base.base.dentist_vco_freq_khz = 2500000; - } else { - struct clk_log_info log_info = {0}; - - clk_mgr->base.smu_ver = dcn316_smu_get_smu_version(&clk_mgr->base); + clk_mgr->base.smu_ver = dcn316_smu_get_smu_version(&clk_mgr->base); - if (clk_mgr->base.smu_ver > 0) - clk_mgr->base.smu_present = true; + if (clk_mgr->base.smu_ver > 0) + clk_mgr->base.smu_present = true; - // Skip this for now as it did not work on DCN315, renable during bring up - clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base); + // Skip this for now as it did not work on DCN315, renable during bring up + clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base); - /* in case we don't get a value from the register, use default */ - if (clk_mgr->base.base.dentist_vco_freq_khz == 0) - clk_mgr->base.base.dentist_vco_freq_khz = 2500000; /* 2400MHz */ + /* in case we don't get a value from the register, use default */ + if (clk_mgr->base.base.dentist_vco_freq_khz == 0) + clk_mgr->base.base.dentist_vco_freq_khz = 2500000; /* 2400MHz */ - if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) { - dcn316_bw_params.wm_table = lpddr5_wm_table; - } else { - dcn316_bw_params.wm_table = ddr4_wm_table; - } - /* Saved clocks configured at boot for debug purposes */ - dcn316_dump_clk_registers(&clk_mgr->base.base.boot_snapshot, - &clk_mgr->base.base, &log_info); - + if (ctx->dc_bios->integrated_info->memory_type == LpDdr5MemType) { + dcn316_bw_params.wm_table = lpddr5_wm_table; + } else { + dcn316_bw_params.wm_table = ddr4_wm_table; } + /* Saved clocks configured at boot for debug purposes */ + dcn316_dump_clk_registers(&clk_mgr->base.base.boot_snapshot, + &clk_mgr->base.base, &log_info); clk_mgr->base.base.dprefclk_khz = 600000; clk_mgr->base.base.dprefclk_khz = dcn316_smu_get_dpref_clk(&clk_mgr->base); diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c index 8d9444db092abecbdfa504ad7f904680294f7b8b..6a811755e2e6f4fa01763f6dd66d1ee02da69efb 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c @@ -182,23 +182,32 @@ void dcn32_init_clocks(struct clk_mgr *clk_mgr_base) dcn32_init_single_clock(clk_mgr, PPCLK_DCFCLK, &clk_mgr_base->bw_params->clk_table.entries[0].dcfclk_mhz, &num_entries_per_clk->num_dcfclk_levels); + clk_mgr_base->bw_params->dc_mode_limit.dcfclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DCFCLK); /* SOCCLK */ dcn32_init_single_clock(clk_mgr, PPCLK_SOCCLK, &clk_mgr_base->bw_params->clk_table.entries[0].socclk_mhz, &num_entries_per_clk->num_socclk_levels); + clk_mgr_base->bw_params->dc_mode_limit.socclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_SOCCLK); /* DTBCLK */ - if (!clk_mgr->base.ctx->dc->debug.disable_dtb_ref_clk_switch) + if (!clk_mgr->base.ctx->dc->debug.disable_dtb_ref_clk_switch) { dcn32_init_single_clock(clk_mgr, PPCLK_DTBCLK, &clk_mgr_base->bw_params->clk_table.entries[0].dtbclk_mhz, &num_entries_per_clk->num_dtbclk_levels); + clk_mgr_base->bw_params->dc_mode_limit.dtbclk_mhz = + dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DTBCLK); + } /* DISPCLK */ dcn32_init_single_clock(clk_mgr, PPCLK_DISPCLK, &clk_mgr_base->bw_params->clk_table.entries[0].dispclk_mhz, &num_entries_per_clk->num_dispclk_levels); num_levels = num_entries_per_clk->num_dispclk_levels; + clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DISPCLK); + //HW recommends limit of 1950 MHz in display clock for all DCN3.2.x + if (clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz > 1950) + clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz = 1950; if (num_entries_per_clk->num_dcfclk_levels && num_entries_per_clk->num_dtbclk_levels && @@ -233,6 +242,32 @@ void dcn32_init_clocks(struct clk_mgr *clk_mgr_base) DC_FP_END(); } +static void dcn32_update_clocks_update_dtb_dto(struct clk_mgr_internal *clk_mgr, + struct dc_state *context, + int ref_dtbclk_khz) +{ + struct dccg *dccg = clk_mgr->dccg; + uint32_t tg_mask = 0; + int i; + + for (i = 0; i < clk_mgr->base.ctx->dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + struct dtbclk_dto_params dto_params = {0}; + + /* use mask to program DTO once per tg */ + if (pipe_ctx->stream_res.tg && + !(tg_mask & (1 << pipe_ctx->stream_res.tg->inst))) { + tg_mask |= (1 << pipe_ctx->stream_res.tg->inst); + + dto_params.otg_inst = pipe_ctx->stream_res.tg->inst; + dto_params.ref_dtbclk_khz = ref_dtbclk_khz; + + dccg->funcs->set_dtbclk_dto(clk_mgr->dccg, &dto_params); + //dccg->funcs->set_audio_dtbclk_dto(clk_mgr->dccg, &dto_params); + } + } +} + /* Since DPPCLK request to PMFW needs to be exact (due to DPP DTO programming), * update DPPCLK to be the exact frequency that will be set after the DPPCLK * divider is updated. This will prevent rounding issues that could cause DPP @@ -433,10 +468,6 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base, bool update_uclk = false, update_fclk = false; bool p_state_change_support; bool fclk_p_state_change_support; - int total_plane_count; - - if (dc->work_arounds.skip_clock_update) - return; if (clk_mgr_base->clks.dispclk_khz == 0 || (dc->debug.force_clock_mode & 0x1)) { @@ -462,10 +493,10 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base, clk_mgr_base->clks.fclk_prev_p_state_change_support = clk_mgr_base->clks.fclk_p_state_change_support; - total_plane_count = clk_mgr_helper_get_active_plane_cnt(dc, context); - fclk_p_state_change_support = new_clocks->fclk_p_state_change_support || (total_plane_count == 0); + fclk_p_state_change_support = new_clocks->fclk_p_state_change_support; - if (should_update_pstate_support(safe_to_lower, fclk_p_state_change_support, clk_mgr_base->clks.fclk_p_state_change_support)) { + if (should_update_pstate_support(safe_to_lower, fclk_p_state_change_support, clk_mgr_base->clks.fclk_p_state_change_support) && + !dc->work_arounds.clock_update_disable_mask.fclk) { clk_mgr_base->clks.fclk_p_state_change_support = fclk_p_state_change_support; /* To enable FCLK P-state switching, send FCLK_PSTATE_SUPPORTED message to PMFW */ @@ -479,12 +510,14 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base, new_clocks->dcfclk_khz = (new_clocks->dcfclk_khz > (dc->debug.force_min_dcfclk_mhz * 1000)) ? new_clocks->dcfclk_khz : (dc->debug.force_min_dcfclk_mhz * 1000); - if (should_set_clock(safe_to_lower, new_clocks->dcfclk_khz, clk_mgr_base->clks.dcfclk_khz)) { + if (should_set_clock(safe_to_lower, new_clocks->dcfclk_khz, clk_mgr_base->clks.dcfclk_khz) && + !dc->work_arounds.clock_update_disable_mask.dcfclk) { clk_mgr_base->clks.dcfclk_khz = new_clocks->dcfclk_khz; dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_DCFCLK, khz_to_mhz_ceil(clk_mgr_base->clks.dcfclk_khz)); } - if (should_set_clock(safe_to_lower, new_clocks->dcfclk_deep_sleep_khz, clk_mgr_base->clks.dcfclk_deep_sleep_khz)) { + if (should_set_clock(safe_to_lower, new_clocks->dcfclk_deep_sleep_khz, clk_mgr_base->clks.dcfclk_deep_sleep_khz) && + !dc->work_arounds.clock_update_disable_mask.dcfclk_ds) { clk_mgr_base->clks.dcfclk_deep_sleep_khz = new_clocks->dcfclk_deep_sleep_khz; dcn30_smu_set_min_deep_sleep_dcef_clk(clk_mgr, khz_to_mhz_ceil(clk_mgr_base->clks.dcfclk_deep_sleep_khz)); } @@ -502,9 +535,9 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base, dcn32_smu_send_cab_for_uclk_message(clk_mgr, clk_mgr_base->clks.num_ways); } - - p_state_change_support = new_clocks->p_state_change_support || (total_plane_count == 0); - if (should_update_pstate_support(safe_to_lower, p_state_change_support, clk_mgr_base->clks.p_state_change_support)) { + p_state_change_support = new_clocks->p_state_change_support; + if (should_update_pstate_support(safe_to_lower, p_state_change_support, clk_mgr_base->clks.p_state_change_support) && + !dc->work_arounds.clock_update_disable_mask.uclk) { clk_mgr_base->clks.p_state_change_support = p_state_change_support; /* to disable P-State switching, set UCLK min = max */ @@ -518,20 +551,23 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base, update_fclk = true; } - if (clk_mgr_base->ctx->dce_version != DCN_VERSION_3_21 && !clk_mgr_base->clks.fclk_p_state_change_support && update_fclk) { + if (clk_mgr_base->ctx->dce_version != DCN_VERSION_3_21 && !clk_mgr_base->clks.fclk_p_state_change_support && update_fclk && + !dc->work_arounds.clock_update_disable_mask.fclk) { /* Handle code for sending a message to PMFW that FCLK P-state change is not supported */ dcn32_smu_send_fclk_pstate_message(clk_mgr, FCLK_PSTATE_NOTSUPPORTED); } /* Always update saved value, even if new value not set due to P-State switching unsupported */ - if (should_set_clock(safe_to_lower, new_clocks->dramclk_khz, clk_mgr_base->clks.dramclk_khz)) { + if (should_set_clock(safe_to_lower, new_clocks->dramclk_khz, clk_mgr_base->clks.dramclk_khz) && + !dc->work_arounds.clock_update_disable_mask.uclk) { clk_mgr_base->clks.dramclk_khz = new_clocks->dramclk_khz; update_uclk = true; } /* set UCLK to requested value if P-State switching is supported, or to re-enable P-State switching */ if (clk_mgr_base->clks.p_state_change_support && - (update_uclk || !clk_mgr_base->clks.prev_p_state_change_support)) + (update_uclk || !clk_mgr_base->clks.prev_p_state_change_support) && + !dc->work_arounds.clock_update_disable_mask.uclk) dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_UCLK, khz_to_mhz_ceil(clk_mgr_base->clks.dramclk_khz)); if (clk_mgr_base->clks.num_ways != new_clocks->num_ways && @@ -570,6 +606,7 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base, /* DCCG requires KHz precision for DTBCLK */ clk_mgr_base->clks.ref_dtbclk_khz = dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_DTBCLK, khz_to_mhz_ceil(new_clocks->ref_dtbclk_khz)); + dcn32_update_clocks_update_dtb_dto(clk_mgr, context, clk_mgr_base->clks.ref_dtbclk_khz); } if (dc->config.forced_clocks == false || (force_reset && safe_to_lower)) { @@ -789,6 +826,7 @@ static void dcn32_get_memclk_states_from_smu(struct clk_mgr *clk_mgr_base) dcn32_init_single_clock(clk_mgr, PPCLK_UCLK, &clk_mgr_base->bw_params->clk_table.entries[0].memclk_mhz, &num_entries_per_clk->num_memclk_levels); + clk_mgr_base->bw_params->dc_mode_limit.memclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_UCLK); /* memclk must have at least one level */ num_entries_per_clk->num_memclk_levels = num_entries_per_clk->num_memclk_levels ? num_entries_per_clk->num_memclk_levels : 1; @@ -796,6 +834,7 @@ static void dcn32_get_memclk_states_from_smu(struct clk_mgr *clk_mgr_base) dcn32_init_single_clock(clk_mgr, PPCLK_FCLK, &clk_mgr_base->bw_params->clk_table.entries[0].fclk_mhz, &num_entries_per_clk->num_fclk_levels); + clk_mgr_base->bw_params->dc_mode_limit.fclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_FCLK); if (num_entries_per_clk->num_memclk_levels >= num_entries_per_clk->num_fclk_levels) { num_levels = num_entries_per_clk->num_memclk_levels; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 52564b93f7eb2dc7539bd184efeea5031e9a149a..dd3a9d06c6e21127e6791219cd0b51eafb8f96f9 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -515,8 +515,7 @@ dc_stream_forward_dmub_crc_window(struct dc_dmub_srv *dmub_srv, cmd.secure_display.roi_info.y_end = rect->y + rect->height; } - dc_dmub_srv_cmd_queue(dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dmub_srv); + dm_execute_dmub_cmd(dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); } static inline void @@ -858,7 +857,6 @@ static bool dc_construct_ctx(struct dc *dc, const struct dc_init_data *init_params) { struct dc_context *dc_ctx; - enum dce_version dc_version = DCE_VERSION_UNKNOWN; dc_ctx = kzalloc(sizeof(*dc_ctx), GFP_KERNEL); if (!dc_ctx) @@ -876,8 +874,7 @@ static bool dc_construct_ctx(struct dc *dc, /* Create logger */ - dc_version = resource_parse_asic_id(init_params->asic_id); - dc_ctx->dce_version = dc_version; + dc_ctx->dce_version = resource_parse_asic_id(init_params->asic_id); dc_ctx->perf_trace = dc_perf_trace_create(); if (!dc_ctx->perf_trace) { @@ -1120,6 +1117,33 @@ static void phantom_pipe_blank( hws->funcs.wait_for_blank_complete(opp); } +static void dc_update_viusal_confirm_color(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe_ctx) +{ + if (dc->ctx->dce_version >= DCN_VERSION_1_0) { + memset(&pipe_ctx->visual_confirm_color, 0, sizeof(struct tg_color)); + + if (dc->debug.visual_confirm == VISUAL_CONFIRM_HDR) + get_hdr_visual_confirm_color(pipe_ctx, &(pipe_ctx->visual_confirm_color)); + else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SURFACE) + get_surface_visual_confirm_color(pipe_ctx, &(pipe_ctx->visual_confirm_color)); + else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SWIZZLE) + get_surface_tile_visual_confirm_color(pipe_ctx, &(pipe_ctx->visual_confirm_color)); + else { + if (dc->ctx->dce_version < DCN_VERSION_2_0) + color_space_to_black_color( + dc, pipe_ctx->stream->output_color_space, &(pipe_ctx->visual_confirm_color)); + } + if (dc->ctx->dce_version >= DCN_VERSION_2_0) { + if (dc->debug.visual_confirm == VISUAL_CONFIRM_MPCTREE) + get_mpctree_visual_confirm_color(pipe_ctx, &(pipe_ctx->visual_confirm_color)); + else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SUBVP) + get_subvp_visual_confirm_color(dc, context, pipe_ctx, &(pipe_ctx->visual_confirm_color)); + else if (dc->debug.visual_confirm == VISUAL_CONFIRM_MCLK_SWITCH) + get_mclk_switch_visual_confirm_color(dc, context, pipe_ctx, &(pipe_ctx->visual_confirm_color)); + } + } +} + static void disable_dangling_plane(struct dc *dc, struct dc_state *context) { int i, j; @@ -1190,6 +1214,9 @@ static void disable_dangling_plane(struct dc *dc, struct dc_state *context) dc_rem_all_planes_for_stream(dc, old_stream, dangling_context); disable_all_writeback_pipes_for_stream(dc, old_stream, dangling_context); + if (pipe->stream && pipe->plane_state) + dc_update_viusal_confirm_color(dc, context, pipe); + if (dc->hwss.apply_ctx_for_surface) { apply_ctx_interdependent_lock(dc, dc->current_state, old_stream, true); dc->hwss.apply_ctx_for_surface(dc, old_stream, 0, dangling_context); @@ -1981,6 +2008,9 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c return result; } +static bool commit_minimal_transition_state(struct dc *dc, + struct dc_state *transition_base_context); + /** * dc_commit_streams - Commit current stream state * @@ -2002,6 +2032,8 @@ enum dc_status dc_commit_streams(struct dc *dc, struct dc_state *context; enum dc_status res = DC_OK; struct dc_validation_set set[MAX_STREAMS] = {0}; + struct pipe_ctx *pipe; + bool handle_exit_odm2to1 = false; if (dc->ctx->dce_environment == DCE_ENV_VIRTUAL_HW) return res; @@ -2026,6 +2058,22 @@ enum dc_status dc_commit_streams(struct dc *dc, } } + /* Check for case where we are going from odm 2:1 to max + * pipe scenario. For these cases, we will call + * commit_minimal_transition_state() to exit out of odm 2:1 + * first before processing new streams + */ + if (stream_count == dc->res_pool->pipe_count) { + for (i = 0; i < dc->res_pool->pipe_count; i++) { + pipe = &dc->current_state->res_ctx.pipe_ctx[i]; + if (pipe->next_odm_pipe) + handle_exit_odm2to1 = true; + } + } + + if (handle_exit_odm2to1) + res = commit_minimal_transition_state(dc, dc->current_state); + context = dc_create_state(dc); if (!context) goto context_alloc_fail; @@ -2483,9 +2531,6 @@ static enum surface_update_type det_surface_update(const struct dc *dc, enum surface_update_type overall_type = UPDATE_TYPE_FAST; union surface_update_flags *update_flags = &u->surface->update_flags; - if (u->flip_addr) - update_flags->bits.addr_update = 1; - if (!is_surface_in_context(context, u->surface) || u->surface->force_full_update) { update_flags->raw = 0xFFFFFFFF; return UPDATE_TYPE_FULL; @@ -2544,15 +2589,19 @@ static enum surface_update_type det_surface_update(const struct dc *dc, elevate_update_type(&overall_type, type); } - if (update_flags->bits.input_csc_change - || update_flags->bits.coeff_reduction_change - || update_flags->bits.lut_3d - || update_flags->bits.gamma_change - || update_flags->bits.gamut_remap_change) { + if (update_flags->bits.lut_3d) { type = UPDATE_TYPE_FULL; elevate_update_type(&overall_type, type); } + if (dc->debug.enable_legacy_fast_update && + (update_flags->bits.gamma_change || + update_flags->bits.gamut_remap_change || + update_flags->bits.input_csc_change || + update_flags->bits.coeff_reduction_change)) { + type = UPDATE_TYPE_FULL; + elevate_update_type(&overall_type, type); + } return overall_type; } @@ -2585,7 +2634,7 @@ static enum surface_update_type check_update_surfaces_for_stream( stream_update->integer_scaling_update) su_flags->bits.scaling = 1; - if (stream_update->out_transfer_func) + if (dc->debug.enable_legacy_fast_update && stream_update->out_transfer_func) su_flags->bits.out_tf = 1; if (stream_update->abm_level) @@ -2605,14 +2654,23 @@ static enum surface_update_type check_update_surfaces_for_stream( if (stream_update->mst_bw_update) su_flags->bits.mst_bw = 1; - if (stream_update->crtc_timing_adjust && dc_extended_blank_supported(dc)) - su_flags->bits.crtc_timing_adjust = 1; + + if (stream_update->stream && stream_update->stream->freesync_on_desktop && + (stream_update->vrr_infopacket || stream_update->allow_freesync || + stream_update->vrr_active_variable || stream_update->vrr_active_fixed)) + su_flags->bits.fams_changed = 1; if (su_flags->raw != 0) overall_type = UPDATE_TYPE_FULL; if (stream_update->output_csc_transform || stream_update->output_color_space) su_flags->bits.out_csc = 1; + + /* Output transfer function changes do not require bandwidth recalculation, + * so don't trigger a full update + */ + if (!dc->debug.enable_legacy_fast_update && stream_update->out_transfer_func) + su_flags->bits.out_tf = 1; } for (i = 0 ; i < surface_count; i++) { @@ -2965,6 +3023,9 @@ static void copy_stream_update_to_stream(struct dc *dc, if (update->vrr_active_variable) stream->vrr_active_variable = *update->vrr_active_variable; + if (update->vrr_active_fixed) + stream->vrr_active_fixed = *update->vrr_active_fixed; + if (update->crtc_timing_adjust) stream->adjust = *update->crtc_timing_adjust; @@ -3269,6 +3330,13 @@ static void commit_planes_do_stream_update(struct dc *dc, dc->hwss.prepare_bandwidth(dc, dc->current_state); dc->link_srv->set_dpms_on(dc->current_state, pipe_ctx); } + } else if (pipe_ctx->stream->link->wa_flags.blank_stream_on_ocs_change && stream_update->output_color_space + && !stream->dpms_off && dc_is_dp_signal(pipe_ctx->stream->signal)) { + /* + * Workaround for firmware issue in some receivers where they don't pick up + * correct output color space unless DP link is disabled/re-enabled + */ + dc->link_srv->set_dpms_on(dc->current_state, pipe_ctx); } if (stream_update->abm_level && pipe_ctx->stream_res.abm) { @@ -3309,7 +3377,6 @@ void dc_dmub_update_dirty_rect(struct dc *dc, struct dc_state *context) { union dmub_rb_cmd cmd; - struct dc_context *dc_ctx = dc->ctx; struct dmub_cmd_update_dirty_rect_data *update_dirty_rect; unsigned int i, j; unsigned int panel_inst = 0; @@ -3350,10 +3417,166 @@ void dc_dmub_update_dirty_rect(struct dc *dc, update_dirty_rect->panel_inst = panel_inst; update_dirty_rect->pipe_idx = j; - dc_dmub_srv_cmd_queue(dc_ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc_ctx->dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); + } + } +} + +static void build_dmub_update_dirty_rect( + struct dc *dc, + int surface_count, + struct dc_stream_state *stream, + struct dc_surface_update *srf_updates, + struct dc_state *context, + struct dc_dmub_cmd dc_dmub_cmd[], + unsigned int *dmub_cmd_count) +{ + union dmub_rb_cmd cmd; + struct dmub_cmd_update_dirty_rect_data *update_dirty_rect; + unsigned int i, j; + unsigned int panel_inst = 0; + + if (!dc_dmub_should_send_dirty_rect_cmd(dc, stream)) + return; + + if (!dc_get_edp_link_panel_inst(dc, stream->link, &panel_inst)) + return; + + memset(&cmd, 0x0, sizeof(cmd)); + cmd.update_dirty_rect.header.type = DMUB_CMD__UPDATE_DIRTY_RECT; + cmd.update_dirty_rect.header.sub_type = 0; + cmd.update_dirty_rect.header.payload_bytes = + sizeof(cmd.update_dirty_rect) - + sizeof(cmd.update_dirty_rect.header); + update_dirty_rect = &cmd.update_dirty_rect.update_dirty_rect_data; + for (i = 0; i < surface_count; i++) { + struct dc_plane_state *plane_state = srf_updates[i].surface; + const struct dc_flip_addrs *flip_addr = srf_updates[i].flip_addr; + + if (!srf_updates[i].surface || !flip_addr) + continue; + /* Do not send in immediate flip mode */ + if (srf_updates[i].surface->flip_immediate) + continue; + update_dirty_rect->cmd_version = DMUB_CMD_PSR_CONTROL_VERSION_1; + update_dirty_rect->dirty_rect_count = flip_addr->dirty_rect_count; + memcpy(update_dirty_rect->src_dirty_rects, flip_addr->dirty_rects, + sizeof(flip_addr->dirty_rects)); + for (j = 0; j < dc->res_pool->pipe_count; j++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; + + if (pipe_ctx->stream != stream) + continue; + if (pipe_ctx->plane_state != plane_state) + continue; + update_dirty_rect->panel_inst = panel_inst; + update_dirty_rect->pipe_idx = j; + dc_dmub_cmd[*dmub_cmd_count].dmub_cmd = cmd; + dc_dmub_cmd[*dmub_cmd_count].wait_type = DM_DMUB_WAIT_TYPE_NO_WAIT; + (*dmub_cmd_count)++; + } + } +} + + +/** + * build_dmub_cmd_list() - Build an array of DMCUB commands to be sent to DMCUB + * + * @dc: Current DC state + * @srf_updates: Array of surface updates + * @surface_count: Number of surfaces that have an updated + * @stream: Corresponding stream to be updated in the current flip + * @context: New DC state to be programmed + * + * @dc_dmub_cmd: Array of DMCUB commands to be sent to DMCUB + * @dmub_cmd_count: Count indicating the number of DMCUB commands in dc_dmub_cmd array + * + * This function builds an array of DMCUB commands to be sent to DMCUB. This function is required + * to build an array of commands and have them sent while the OTG lock is acquired. + * + * Return: void + */ +static void build_dmub_cmd_list(struct dc *dc, + struct dc_surface_update *srf_updates, + int surface_count, + struct dc_stream_state *stream, + struct dc_state *context, + struct dc_dmub_cmd dc_dmub_cmd[], + unsigned int *dmub_cmd_count) +{ + // Initialize cmd count to 0 + *dmub_cmd_count = 0; + build_dmub_update_dirty_rect(dc, surface_count, stream, srf_updates, context, dc_dmub_cmd, dmub_cmd_count); +} + +static void commit_planes_for_stream_fast(struct dc *dc, + struct dc_surface_update *srf_updates, + int surface_count, + struct dc_stream_state *stream, + struct dc_stream_update *stream_update, + enum surface_update_type update_type, + struct dc_state *context) +{ + int i, j; + struct pipe_ctx *top_pipe_to_program = NULL; + dc_z10_restore(dc); + + for (j = 0; j < dc->res_pool->pipe_count; j++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; + + if (!pipe_ctx->top_pipe && + !pipe_ctx->prev_odm_pipe && + pipe_ctx->stream && + pipe_ctx->stream == stream) { + top_pipe_to_program = pipe_ctx; + } + } + + if (dc->debug.visual_confirm) { + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + + if (pipe->stream && pipe->plane_state) + dc_update_viusal_confirm_color(dc, context, pipe); + } + } + + for (i = 0; i < surface_count; i++) { + struct dc_plane_state *plane_state = srf_updates[i].surface; + /*set logical flag for lock/unlock use*/ + for (j = 0; j < dc->res_pool->pipe_count; j++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; + + if (!pipe_ctx->plane_state) + continue; + if (should_update_pipe_for_plane(context, pipe_ctx, plane_state)) + continue; + pipe_ctx->plane_state->triplebuffer_flips = false; + if (update_type == UPDATE_TYPE_FAST && + dc->hwss.program_triplebuffer && + !pipe_ctx->plane_state->flip_immediate && dc->debug.enable_tri_buf) { + /*triple buffer for VUpdate only*/ + pipe_ctx->plane_state->triplebuffer_flips = true; + } } } + + build_dmub_cmd_list(dc, + srf_updates, + surface_count, + stream, + context, + context->dc_dmub_cmd, + &(context->dmub_cmd_count)); + hwss_build_fast_sequence(dc, + context->dc_dmub_cmd, + context->dmub_cmd_count, + context->block_sequence, + &(context->block_sequence_steps), + top_pipe_to_program); + hwss_execute_sequence(dc, + context->block_sequence, + context->block_sequence_steps); } static void commit_planes_for_stream(struct dc *dc, @@ -3393,21 +3616,6 @@ static void commit_planes_for_stream(struct dc *dc, } } - if (get_seamless_boot_stream_count(context) > 0 && surface_count > 0) { - /* Optimize seamless boot flag keeps clocks and watermarks high until - * first flip. After first flip, optimization is required to lower - * bandwidth. Important to note that it is expected UEFI will - * only light up a single display on POST, therefore we only expect - * one stream with seamless boot flag set. - */ - if (stream->apply_seamless_boot_optimization) { - stream->apply_seamless_boot_optimization = false; - - if (get_seamless_boot_stream_count(context) == 0) - dc->optimized_required = true; - } - } - if (update_type == UPDATE_TYPE_FULL) { dc_allow_idle_optimizations(dc, false); @@ -3449,6 +3657,14 @@ static void commit_planes_for_stream(struct dc *dc, } } + if (dc->debug.visual_confirm) + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + + if (pipe->stream && pipe->plane_state) + dc_update_viusal_confirm_color(dc, context, pipe); + } + if (stream->test_pattern.type != DP_TEST_PATTERN_VIDEO_MODE) { struct pipe_ctx *mpcc_pipe; struct pipe_ctx *odm_pipe; @@ -3532,43 +3748,40 @@ static void commit_planes_for_stream(struct dc *dc, for (j = 0; j < dc->res_pool->pipe_count; j++) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; - if (dc->debug.visual_confirm == VISUAL_CONFIRM_SUBVP && + if ((dc->debug.visual_confirm == VISUAL_CONFIRM_SUBVP || + dc->debug.visual_confirm == VISUAL_CONFIRM_MCLK_SWITCH) && pipe_ctx->stream && pipe_ctx->plane_state) { - /* Only update visual confirm for SUBVP here. + /* Only update visual confirm for SUBVP and Mclk switching here. * The bar appears on all pipes, so we need to update the bar on all displays, * so the information doesn't get stale. */ - struct mpcc_blnd_cfg blnd_cfg = { 0 }; - - dc->hwss.update_visual_confirm_color(dc, pipe_ctx, &blnd_cfg.black_color, + dc->hwss.update_visual_confirm_color(dc, pipe_ctx, pipe_ctx->plane_res.hubp->inst); } } } - if (!IS_DIAG_DC(dc->ctx->dce_environment)) { - for (i = 0; i < surface_count; i++) { - struct dc_plane_state *plane_state = srf_updates[i].surface; - /*set logical flag for lock/unlock use*/ - for (j = 0; j < dc->res_pool->pipe_count; j++) { - struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; - if (!pipe_ctx->plane_state) - continue; - if (should_update_pipe_for_plane(context, pipe_ctx, plane_state)) - continue; - pipe_ctx->plane_state->triplebuffer_flips = false; - if (update_type == UPDATE_TYPE_FAST && - dc->hwss.program_triplebuffer != NULL && - !pipe_ctx->plane_state->flip_immediate && dc->debug.enable_tri_buf) { - /*triple buffer for VUpdate only*/ - pipe_ctx->plane_state->triplebuffer_flips = true; - } - } - if (update_type == UPDATE_TYPE_FULL) { - /* force vsync flip when reconfiguring pipes to prevent underflow */ - plane_state->flip_immediate = false; + for (i = 0; i < surface_count; i++) { + struct dc_plane_state *plane_state = srf_updates[i].surface; + /*set logical flag for lock/unlock use*/ + for (j = 0; j < dc->res_pool->pipe_count; j++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; + if (!pipe_ctx->plane_state) + continue; + if (should_update_pipe_for_plane(context, pipe_ctx, plane_state)) + continue; + pipe_ctx->plane_state->triplebuffer_flips = false; + if (update_type == UPDATE_TYPE_FAST && + dc->hwss.program_triplebuffer != NULL && + !pipe_ctx->plane_state->flip_immediate && dc->debug.enable_tri_buf) { + /*triple buffer for VUpdate only*/ + pipe_ctx->plane_state->triplebuffer_flips = true; } } + if (update_type == UPDATE_TYPE_FULL) { + /* force vsync flip when reconfiguring pipes to prevent underflow */ + plane_state->flip_immediate = false; + } } // Update Type FULL, Surface updates @@ -3872,6 +4085,7 @@ static bool commit_minimal_transition_state(struct dc *dc, unsigned int i, j; unsigned int pipe_in_use = 0; bool subvp_in_use = false; + bool odm_in_use = false; if (!transition_context) return false; @@ -3900,6 +4114,18 @@ static bool commit_minimal_transition_state(struct dc *dc, } } + /* If ODM is enabled and we are adding or removing planes from any ODM + * pipe, we must use the minimal transition. + */ + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe = &dc->current_state->res_ctx.pipe_ctx[i]; + + if (pipe->stream && pipe->next_odm_pipe) { + odm_in_use = true; + break; + } + } + /* When the OS add a new surface if we have been used all of pipes with odm combine * and mpc split feature, it need use commit_minimal_transition_state to transition safely. * After OS exit MPO, it will back to use odm and mpc split with all of pipes, we need @@ -3908,7 +4134,7 @@ static bool commit_minimal_transition_state(struct dc *dc, * Reduce the scenarios to use dc_commit_state_no_check in the stage of flip. Especially * enter/exit MPO when DCN still have enough resources. */ - if (pipe_in_use != dc->res_pool->pipe_count && !subvp_in_use) { + if (pipe_in_use != dc->res_pool->pipe_count && !subvp_in_use && !odm_in_use) { dc_release_state(transition_context); return true; } @@ -3972,6 +4198,41 @@ static bool commit_minimal_transition_state(struct dc *dc, return true; } +/** + * update_seamless_boot_flags() - Helper function for updating seamless boot flags + * + * @dc: Current DC state + * @context: New DC state to be programmed + * @surface_count: Number of surfaces that have an updated + * @stream: Corresponding stream to be updated in the current flip + * + * Updating seamless boot flags do not need to be part of the commit sequence. This + * helper function will update the seamless boot flags on each flip (if required) + * outside of the HW commit sequence (fast or slow). + * + * Return: void + */ +static void update_seamless_boot_flags(struct dc *dc, + struct dc_state *context, + int surface_count, + struct dc_stream_state *stream) +{ + if (get_seamless_boot_stream_count(context) > 0 && surface_count > 0) { + /* Optimize seamless boot flag keeps clocks and watermarks high until + * first flip. After first flip, optimization is required to lower + * bandwidth. Important to note that it is expected UEFI will + * only light up a single display on POST, therefore we only expect + * one stream with seamless boot flag set. + */ + if (stream->apply_seamless_boot_optimization) { + stream->apply_seamless_boot_optimization = false; + + if (get_seamless_boot_stream_count(context) == 0) + dc->optimized_required = true; + } + } +} + bool dc_update_planes_and_stream(struct dc *dc, struct dc_surface_update *srf_updates, int surface_count, struct dc_stream_state *stream, @@ -4038,14 +4299,25 @@ bool dc_update_planes_and_stream(struct dc *dc, update_type = UPDATE_TYPE_FULL; } - commit_planes_for_stream( - dc, - srf_updates, - surface_count, - stream, - stream_update, - update_type, - context); + update_seamless_boot_flags(dc, context, surface_count, stream); + if (!dc->debug.enable_legacy_fast_update && update_type == UPDATE_TYPE_FAST) { + commit_planes_for_stream_fast(dc, + srf_updates, + surface_count, + stream, + stream_update, + update_type, + context); + } else { + commit_planes_for_stream( + dc, + srf_updates, + surface_count, + stream, + stream_update, + update_type, + context); + } if (dc->current_state != context) { @@ -4170,7 +4442,17 @@ void dc_commit_updates_for_stream(struct dc *dc, TRACE_DC_PIPE_STATE(pipe_ctx, i, MAX_PIPES); - commit_planes_for_stream( + update_seamless_boot_flags(dc, context, surface_count, stream); + if (!dc->debug.enable_legacy_fast_update && update_type == UPDATE_TYPE_FAST) { + commit_planes_for_stream_fast(dc, + srf_updates, + surface_count, + stream, + stream_update, + update_type, + context); + } else { + commit_planes_for_stream( dc, srf_updates, surface_count, @@ -4178,6 +4460,7 @@ void dc_commit_updates_for_stream(struct dc *dc, stream_update, update_type, context); + } /*update current_State*/ if (dc->current_state != context) { @@ -4264,9 +4547,6 @@ void dc_set_power_state( dc_z10_restore(dc); - if (dc->ctx->dmub_srv) - dc_dmub_srv_wait_phy_init(dc->ctx->dmub_srv); - dc->hwss.init_hw(dc); if (dc->hwss.init_sys_ctx != NULL && @@ -4606,7 +4886,6 @@ bool dc_process_dmub_aux_transfer_async(struct dc *dc, { uint8_t action; union dmub_rb_cmd cmd = {0}; - struct dc_dmub_srv *dmub_srv = dc->ctx->dmub_srv; ASSERT(payload->length <= 16); @@ -4654,9 +4933,7 @@ bool dc_process_dmub_aux_transfer_async(struct dc *dc, ); } - dc_dmub_srv_cmd_queue(dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dmub_srv); - dc_dmub_srv_wait_idle(dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } @@ -4700,7 +4977,6 @@ bool dc_process_dmub_set_config_async(struct dc *dc, struct dmub_notification *notify) { union dmub_rb_cmd cmd = {0}; - struct dc_dmub_srv *dmub_srv = dc->ctx->dmub_srv; bool is_cmd_complete = true; /* prepare SET_CONFIG command */ @@ -4711,7 +4987,7 @@ bool dc_process_dmub_set_config_async(struct dc *dc, cmd.set_config_access.set_config_control.cmd_pkt.msg_type = payload->msg_type; cmd.set_config_access.set_config_control.cmd_pkt.msg_data = payload->msg_data; - if (!dc_dmub_srv_cmd_with_reply_data(dmub_srv, &cmd)) { + if (!dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) { /* command is not processed by dmub */ notify->sc_status = SET_CONFIG_UNKNOWN_ERROR; return is_cmd_complete; @@ -4746,7 +5022,6 @@ enum dc_status dc_process_dmub_set_mst_slots(const struct dc *dc, uint8_t *mst_slots_in_use) { union dmub_rb_cmd cmd = {0}; - struct dc_dmub_srv *dmub_srv = dc->ctx->dmub_srv; /* prepare MST_ALLOC_SLOTS command */ cmd.set_mst_alloc_slots.header.type = DMUB_CMD__DPIA; @@ -4755,7 +5030,7 @@ enum dc_status dc_process_dmub_set_mst_slots(const struct dc *dc, cmd.set_mst_alloc_slots.mst_slots_control.instance = dc->links[link_index]->ddc_hw_inst; cmd.set_mst_alloc_slots.mst_slots_control.mst_alloc_slots = mst_alloc_slots; - if (!dc_dmub_srv_cmd_with_reply_data(dmub_srv, &cmd)) + if (!dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) /* command is not processed by dmub */ return DC_ERROR_UNEXPECTED; @@ -4789,18 +5064,27 @@ void dc_process_dmub_dpia_hpd_int_enable(const struct dc *dc, uint32_t hpd_int_enable) { union dmub_rb_cmd cmd = {0}; - struct dc_dmub_srv *dmub_srv = dc->ctx->dmub_srv; cmd.dpia_hpd_int_enable.header.type = DMUB_CMD__DPIA_HPD_INT_ENABLE; cmd.dpia_hpd_int_enable.enable = hpd_int_enable; - dc_dmub_srv_cmd_queue(dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dmub_srv); - dc_dmub_srv_wait_idle(dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); DC_LOG_DEBUG("%s: hpd_int_enable(%d)\n", __func__, hpd_int_enable); } +/** + * dc_print_dmub_diagnostic_data - Print DMUB diagnostic data for debugging + * + * @dc: [in] dc structure + * + * + */ +void dc_print_dmub_diagnostic_data(const struct dc *dc) +{ + dc_dmub_srv_log_diagnostic_data(dc->ctx->dmub_srv); +} + /** * dc_disable_accelerated_mode - disable accelerated mode * @dc: dc structure @@ -4860,21 +5144,3 @@ void dc_notify_vsync_int_state(struct dc *dc, struct dc_stream_state *stream, bo if (pipe->stream_res.abm && pipe->stream_res.abm->funcs->set_abm_pause) pipe->stream_res.abm->funcs->set_abm_pause(pipe->stream_res.abm, !enable, i, pipe->stream_res.tg->inst); } - -/** - * dc_extended_blank_supported - Decide whether extended blank is supported - * - * @dc: [in] Current DC state - * - * Extended blank is a freesync optimization feature to be enabled in the - * future. During the extra vblank period gained from freesync, we have the - * ability to enter z9/z10. - * - * Return: - * Indicate whether extended blank is supported (%true or %false) - */ -bool dc_extended_blank_supported(struct dc *dc) -{ - return dc->debug.extended_blank_optimization && !dc->debug.disable_z10 - && dc->caps.zstate_support && dc->caps.is_apu; -} diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c index 2acbf692193f7728f0ef17ad78022002dffb9791..d7d00fefaab9be09343c8356d221317b1680d851 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c @@ -27,6 +27,8 @@ #include "core_types.h" #include "timing_generator.h" #include "hw_sequencer.h" +#include "hw_sequencer_private.h" +#include "basics/dc_common.h" #define NUM_ELEMENTS(a) (sizeof(a) / sizeof((a)[0])) @@ -421,6 +423,7 @@ void get_hdr_visual_confirm_color( void get_subvp_visual_confirm_color( struct dc *dc, + struct dc_state *context, struct pipe_ctx *pipe_ctx, struct tg_color *color) { @@ -428,15 +431,17 @@ void get_subvp_visual_confirm_color( bool enable_subvp = false; int i; - if (!dc->ctx || !dc->ctx->dmub_srv || !pipe_ctx) + if (!dc->ctx || !dc->ctx->dmub_srv || !pipe_ctx || !context) return; for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *pipe = &dc->current_state->res_ctx.pipe_ctx[i]; + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; if (pipe->stream && pipe->stream->mall_stream_config.paired_stream && pipe->stream->mall_stream_config.type == SUBVP_MAIN) { /* SubVP enable - red */ + color->color_g_y = 0; + color->color_b_cb = 0; color->color_r_cr = color_value; enable_subvp = true; @@ -448,12 +453,374 @@ void get_subvp_visual_confirm_color( if (enable_subvp && pipe_ctx->stream->mall_stream_config.type == SUBVP_NONE) { color->color_r_cr = 0; - if (pipe_ctx->stream->ignore_msa_timing_param == 1) + if (pipe_ctx->stream->allow_freesync == 1) { /* SubVP enable and DRR on - green */ + color->color_b_cb = 0; color->color_g_y = color_value; - else + } else { /* SubVP enable and No DRR - blue */ + color->color_g_y = 0; color->color_b_cb = color_value; + } + } +} + +void hwss_build_fast_sequence(struct dc *dc, + struct dc_dmub_cmd *dc_dmub_cmd, + unsigned int dmub_cmd_count, + struct block_sequence block_sequence[], + int *num_steps, + struct pipe_ctx *pipe_ctx) +{ + struct dc_plane_state *plane = pipe_ctx->plane_state; + struct dc_stream_state *stream = pipe_ctx->stream; + struct dce_hwseq *hws = dc->hwseq; + struct pipe_ctx *current_pipe = NULL; + struct pipe_ctx *current_mpc_pipe = NULL; + unsigned int i = 0; + + *num_steps = 0; // Initialize to 0 + + if (!plane || !stream) + return; + + if (dc->hwss.subvp_pipe_control_lock_fast) { + block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.dc = dc; + block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.lock = true; + block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.pipe_ctx = pipe_ctx; + block_sequence[*num_steps].func = DMUB_SUBVP_PIPE_CONTROL_LOCK_FAST; + (*num_steps)++; + } + if (dc->hwss.pipe_control_lock) { + block_sequence[*num_steps].params.pipe_control_lock_params.dc = dc; + block_sequence[*num_steps].params.pipe_control_lock_params.lock = true; + block_sequence[*num_steps].params.pipe_control_lock_params.pipe_ctx = pipe_ctx; + block_sequence[*num_steps].func = OPTC_PIPE_CONTROL_LOCK; + (*num_steps)++; + } + + for (i = 0; i < dmub_cmd_count; i++) { + block_sequence[*num_steps].params.send_dmcub_cmd_params.ctx = dc->ctx; + block_sequence[*num_steps].params.send_dmcub_cmd_params.cmd = &(dc_dmub_cmd[i].dmub_cmd); + block_sequence[*num_steps].params.send_dmcub_cmd_params.wait_type = dc_dmub_cmd[i].wait_type; + block_sequence[*num_steps].func = DMUB_SEND_DMCUB_CMD; + (*num_steps)++; + } + + current_pipe = pipe_ctx; + while (current_pipe) { + current_mpc_pipe = current_pipe; + while (current_mpc_pipe) { + if (dc->hwss.set_flip_control_gsl && current_mpc_pipe->plane_state && current_mpc_pipe->plane_state->update_flags.raw) { + block_sequence[*num_steps].params.set_flip_control_gsl_params.pipe_ctx = current_mpc_pipe; + block_sequence[*num_steps].params.set_flip_control_gsl_params.flip_immediate = current_mpc_pipe->plane_state->flip_immediate; + block_sequence[*num_steps].func = HUBP_SET_FLIP_CONTROL_GSL; + (*num_steps)++; + } + if (dc->hwss.program_triplebuffer && dc->debug.enable_tri_buf && current_mpc_pipe->plane_state->update_flags.raw) { + block_sequence[*num_steps].params.program_triplebuffer_params.dc = dc; + block_sequence[*num_steps].params.program_triplebuffer_params.pipe_ctx = current_mpc_pipe; + block_sequence[*num_steps].params.program_triplebuffer_params.enableTripleBuffer = current_mpc_pipe->plane_state->triplebuffer_flips; + block_sequence[*num_steps].func = HUBP_PROGRAM_TRIPLEBUFFER; + (*num_steps)++; + } + if (dc->hwss.update_plane_addr && current_mpc_pipe->plane_state->update_flags.bits.addr_update) { + block_sequence[*num_steps].params.update_plane_addr_params.dc = dc; + block_sequence[*num_steps].params.update_plane_addr_params.pipe_ctx = current_mpc_pipe; + block_sequence[*num_steps].func = HUBP_UPDATE_PLANE_ADDR; + (*num_steps)++; + } + + if (hws->funcs.set_input_transfer_func && current_mpc_pipe->plane_state->update_flags.bits.gamma_change) { + block_sequence[*num_steps].params.set_input_transfer_func_params.dc = dc; + block_sequence[*num_steps].params.set_input_transfer_func_params.pipe_ctx = current_mpc_pipe; + block_sequence[*num_steps].params.set_input_transfer_func_params.plane_state = current_mpc_pipe->plane_state; + block_sequence[*num_steps].func = DPP_SET_INPUT_TRANSFER_FUNC; + (*num_steps)++; + } + + if (dc->hwss.program_gamut_remap && current_mpc_pipe->plane_state->update_flags.bits.gamut_remap_change) { + block_sequence[*num_steps].params.program_gamut_remap_params.pipe_ctx = current_mpc_pipe; + block_sequence[*num_steps].func = DPP_PROGRAM_GAMUT_REMAP; + (*num_steps)++; + } + if (current_mpc_pipe->plane_state->update_flags.bits.input_csc_change) { + block_sequence[*num_steps].params.setup_dpp_params.pipe_ctx = current_mpc_pipe; + block_sequence[*num_steps].func = DPP_SETUP_DPP; + (*num_steps)++; + } + if (current_mpc_pipe->plane_state->update_flags.bits.coeff_reduction_change) { + block_sequence[*num_steps].params.program_bias_and_scale_params.pipe_ctx = current_mpc_pipe; + block_sequence[*num_steps].func = DPP_PROGRAM_BIAS_AND_SCALE; + (*num_steps)++; + } + if (hws->funcs.set_output_transfer_func && current_mpc_pipe->stream->update_flags.bits.out_tf) { + block_sequence[*num_steps].params.set_output_transfer_func_params.dc = dc; + block_sequence[*num_steps].params.set_output_transfer_func_params.pipe_ctx = current_mpc_pipe; + block_sequence[*num_steps].params.set_output_transfer_func_params.stream = current_mpc_pipe->stream; + block_sequence[*num_steps].func = DPP_SET_OUTPUT_TRANSFER_FUNC; + (*num_steps)++; + } + + if (current_mpc_pipe->stream->update_flags.bits.out_csc) { + block_sequence[*num_steps].params.power_on_mpc_mem_pwr_params.mpc = dc->res_pool->mpc; + block_sequence[*num_steps].params.power_on_mpc_mem_pwr_params.mpcc_id = current_mpc_pipe->plane_res.hubp->inst; + block_sequence[*num_steps].params.power_on_mpc_mem_pwr_params.power_on = true; + block_sequence[*num_steps].func = MPC_POWER_ON_MPC_MEM_PWR; + (*num_steps)++; + + if (current_mpc_pipe->stream->csc_color_matrix.enable_adjustment == true) { + block_sequence[*num_steps].params.set_output_csc_params.mpc = dc->res_pool->mpc; + block_sequence[*num_steps].params.set_output_csc_params.opp_id = current_mpc_pipe->stream_res.opp->inst; + block_sequence[*num_steps].params.set_output_csc_params.regval = current_mpc_pipe->stream->csc_color_matrix.matrix; + block_sequence[*num_steps].params.set_output_csc_params.ocsc_mode = MPC_OUTPUT_CSC_COEF_A; + block_sequence[*num_steps].func = MPC_SET_OUTPUT_CSC; + (*num_steps)++; + } else { + block_sequence[*num_steps].params.set_ocsc_default_params.mpc = dc->res_pool->mpc; + block_sequence[*num_steps].params.set_ocsc_default_params.opp_id = current_mpc_pipe->stream_res.opp->inst; + block_sequence[*num_steps].params.set_ocsc_default_params.color_space = current_mpc_pipe->stream->output_color_space; + block_sequence[*num_steps].params.set_ocsc_default_params.ocsc_mode = MPC_OUTPUT_CSC_COEF_A; + block_sequence[*num_steps].func = MPC_SET_OCSC_DEFAULT; + (*num_steps)++; + } + } + current_mpc_pipe = current_mpc_pipe->bottom_pipe; + } + current_pipe = current_pipe->next_odm_pipe; + } + + if (dc->hwss.pipe_control_lock) { + block_sequence[*num_steps].params.pipe_control_lock_params.dc = dc; + block_sequence[*num_steps].params.pipe_control_lock_params.lock = false; + block_sequence[*num_steps].params.pipe_control_lock_params.pipe_ctx = pipe_ctx; + block_sequence[*num_steps].func = OPTC_PIPE_CONTROL_LOCK; + (*num_steps)++; + } + if (dc->hwss.subvp_pipe_control_lock_fast) { + block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.dc = dc; + block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.lock = false; + block_sequence[*num_steps].params.subvp_pipe_control_lock_fast_params.pipe_ctx = pipe_ctx; + block_sequence[*num_steps].func = DMUB_SUBVP_PIPE_CONTROL_LOCK_FAST; + (*num_steps)++; + } + + current_pipe = pipe_ctx; + while (current_pipe) { + current_mpc_pipe = current_pipe; + + while (current_mpc_pipe) { + if (!current_mpc_pipe->bottom_pipe && !pipe_ctx->next_odm_pipe && + current_mpc_pipe->stream && current_mpc_pipe->plane_state && + current_mpc_pipe->plane_state->update_flags.bits.addr_update && + !current_mpc_pipe->plane_state->skip_manual_trigger) { + block_sequence[*num_steps].params.program_manual_trigger_params.pipe_ctx = current_mpc_pipe; + block_sequence[*num_steps].func = OPTC_PROGRAM_MANUAL_TRIGGER; + (*num_steps)++; + } + current_mpc_pipe = current_mpc_pipe->bottom_pipe; + } + current_pipe = current_pipe->next_odm_pipe; + } +} + +void hwss_execute_sequence(struct dc *dc, + struct block_sequence block_sequence[], + int num_steps) +{ + unsigned int i; + union block_sequence_params *params; + struct dce_hwseq *hws = dc->hwseq; + + for (i = 0; i < num_steps; i++) { + params = &(block_sequence[i].params); + switch (block_sequence[i].func) { + + case DMUB_SUBVP_PIPE_CONTROL_LOCK_FAST: + dc->hwss.subvp_pipe_control_lock_fast(params); + break; + case OPTC_PIPE_CONTROL_LOCK: + dc->hwss.pipe_control_lock(params->pipe_control_lock_params.dc, + params->pipe_control_lock_params.pipe_ctx, + params->pipe_control_lock_params.lock); + break; + case HUBP_SET_FLIP_CONTROL_GSL: + dc->hwss.set_flip_control_gsl(params->set_flip_control_gsl_params.pipe_ctx, + params->set_flip_control_gsl_params.flip_immediate); + break; + case HUBP_PROGRAM_TRIPLEBUFFER: + dc->hwss.program_triplebuffer(params->program_triplebuffer_params.dc, + params->program_triplebuffer_params.pipe_ctx, + params->program_triplebuffer_params.enableTripleBuffer); + break; + case HUBP_UPDATE_PLANE_ADDR: + dc->hwss.update_plane_addr(params->update_plane_addr_params.dc, + params->update_plane_addr_params.pipe_ctx); + break; + case DPP_SET_INPUT_TRANSFER_FUNC: + hws->funcs.set_input_transfer_func(params->set_input_transfer_func_params.dc, + params->set_input_transfer_func_params.pipe_ctx, + params->set_input_transfer_func_params.plane_state); + break; + case DPP_PROGRAM_GAMUT_REMAP: + dc->hwss.program_gamut_remap(params->program_gamut_remap_params.pipe_ctx); + break; + case DPP_SETUP_DPP: + hwss_setup_dpp(params); + break; + case DPP_PROGRAM_BIAS_AND_SCALE: + hwss_program_bias_and_scale(params); + break; + case OPTC_PROGRAM_MANUAL_TRIGGER: + hwss_program_manual_trigger(params); + break; + case DPP_SET_OUTPUT_TRANSFER_FUNC: + hws->funcs.set_output_transfer_func(params->set_output_transfer_func_params.dc, + params->set_output_transfer_func_params.pipe_ctx, + params->set_output_transfer_func_params.stream); + break; + case MPC_UPDATE_VISUAL_CONFIRM: + dc->hwss.update_visual_confirm_color(params->update_visual_confirm_params.dc, + params->update_visual_confirm_params.pipe_ctx, + params->update_visual_confirm_params.mpcc_id); + break; + case MPC_POWER_ON_MPC_MEM_PWR: + hwss_power_on_mpc_mem_pwr(params); + break; + case MPC_SET_OUTPUT_CSC: + hwss_set_output_csc(params); + break; + case MPC_SET_OCSC_DEFAULT: + hwss_set_ocsc_default(params); + break; + case DMUB_SEND_DMCUB_CMD: + hwss_send_dmcub_cmd(params); + break; + default: + ASSERT(false); + break; + } + } +} + +void hwss_send_dmcub_cmd(union block_sequence_params *params) +{ + struct dc_context *ctx = params->send_dmcub_cmd_params.ctx; + union dmub_rb_cmd *cmd = params->send_dmcub_cmd_params.cmd; + enum dm_dmub_wait_type wait_type = params->send_dmcub_cmd_params.wait_type; + + dm_execute_dmub_cmd(ctx, cmd, wait_type); +} + +void hwss_program_manual_trigger(union block_sequence_params *params) +{ + struct pipe_ctx *pipe_ctx = params->program_manual_trigger_params.pipe_ctx; + + if (pipe_ctx->stream_res.tg->funcs->program_manual_trigger) + pipe_ctx->stream_res.tg->funcs->program_manual_trigger(pipe_ctx->stream_res.tg); +} + +void hwss_setup_dpp(union block_sequence_params *params) +{ + struct pipe_ctx *pipe_ctx = params->setup_dpp_params.pipe_ctx; + struct dpp *dpp = pipe_ctx->plane_res.dpp; + struct dc_plane_state *plane_state = pipe_ctx->plane_state; + + if (dpp && dpp->funcs->dpp_setup) { + // program the input csc + dpp->funcs->dpp_setup(dpp, + plane_state->format, + EXPANSION_MODE_ZERO, + plane_state->input_csc_color_matrix, + plane_state->color_space, + NULL); + } +} + +void hwss_program_bias_and_scale(union block_sequence_params *params) +{ + struct pipe_ctx *pipe_ctx = params->program_bias_and_scale_params.pipe_ctx; + struct dpp *dpp = pipe_ctx->plane_res.dpp; + struct dc_plane_state *plane_state = pipe_ctx->plane_state; + struct dc_bias_and_scale bns_params = {0}; + + //TODO :for CNVC set scale and bias registers if necessary + build_prescale_params(&bns_params, plane_state); + if (dpp->funcs->dpp_program_bias_and_scale) + dpp->funcs->dpp_program_bias_and_scale(dpp, &bns_params); +} + +void hwss_power_on_mpc_mem_pwr(union block_sequence_params *params) +{ + struct mpc *mpc = params->power_on_mpc_mem_pwr_params.mpc; + int mpcc_id = params->power_on_mpc_mem_pwr_params.mpcc_id; + bool power_on = params->power_on_mpc_mem_pwr_params.power_on; + + if (mpc->funcs->power_on_mpc_mem_pwr) + mpc->funcs->power_on_mpc_mem_pwr(mpc, mpcc_id, power_on); +} + +void hwss_set_output_csc(union block_sequence_params *params) +{ + struct mpc *mpc = params->set_output_csc_params.mpc; + int opp_id = params->set_output_csc_params.opp_id; + const uint16_t *matrix = params->set_output_csc_params.regval; + enum mpc_output_csc_mode ocsc_mode = params->set_output_csc_params.ocsc_mode; + + if (mpc->funcs->set_output_csc != NULL) + mpc->funcs->set_output_csc(mpc, + opp_id, + matrix, + ocsc_mode); +} + +void hwss_set_ocsc_default(union block_sequence_params *params) +{ + struct mpc *mpc = params->set_ocsc_default_params.mpc; + int opp_id = params->set_ocsc_default_params.opp_id; + enum dc_color_space colorspace = params->set_ocsc_default_params.color_space; + enum mpc_output_csc_mode ocsc_mode = params->set_ocsc_default_params.ocsc_mode; + + if (mpc->funcs->set_ocsc_default != NULL) + mpc->funcs->set_ocsc_default(mpc, + opp_id, + colorspace, + ocsc_mode); +} + +void get_mclk_switch_visual_confirm_color( + struct dc *dc, + struct dc_state *context, + struct pipe_ctx *pipe_ctx, + struct tg_color *color) +{ + uint32_t color_value = MAX_TG_COLOR_VALUE; + struct vba_vars_st *vba = &context->bw_ctx.dml.vba; + + if (!dc->ctx || !dc->ctx->dmub_srv || !pipe_ctx || !vba || !context) + return; + + if (vba->DRAMClockChangeSupport[vba->VoltageLevel][vba->maxMpcComb] != + dm_dram_clock_change_unsupported) { + /* MCLK switching is supported */ + if (!pipe_ctx->has_vactive_margin) { + /* In Vblank - yellow */ + color->color_r_cr = color_value; + color->color_g_y = color_value; + + if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) { + /* FPO + Vblank - cyan */ + color->color_r_cr = 0; + color->color_g_y = color_value; + color->color_b_cb = color_value; + } + } else { + /* In Vactive - pink */ + color->color_r_cr = color_value; + color->color_b_cb = color_value; + } + /* SubVP */ + get_subvp_visual_confirm_color(dc, context, pipe_ctx, color); } } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index 117d80cb36fbb5e245ce0dd8b27e076cc3dc75fd..2f3d9a698486d2d4461bb2da749a64f082c1a5f9 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -69,6 +69,7 @@ #include "../dcn32/dcn32_resource.h" #include "../dcn321/dcn321_resource.h" + #define DC_LOGGER_INIT(logger) enum dce_version resource_parse_asic_id(struct hw_asic_id asic_id) @@ -1444,6 +1445,26 @@ static int acquire_first_split_pipe( split_pipe->plane_res.mpcc_inst = pool->dpps[i]->inst; split_pipe->pipe_idx = i; + split_pipe->stream = stream; + return i; + } else if (split_pipe->prev_odm_pipe && + split_pipe->prev_odm_pipe->plane_state == split_pipe->plane_state) { + split_pipe->prev_odm_pipe->next_odm_pipe = split_pipe->next_odm_pipe; + if (split_pipe->next_odm_pipe) + split_pipe->next_odm_pipe->prev_odm_pipe = split_pipe->prev_odm_pipe; + + if (split_pipe->prev_odm_pipe->plane_state) + resource_build_scaling_params(split_pipe->prev_odm_pipe); + + memset(split_pipe, 0, sizeof(*split_pipe)); + split_pipe->stream_res.tg = pool->timing_generators[i]; + split_pipe->plane_res.hubp = pool->hubps[i]; + split_pipe->plane_res.ipp = pool->ipps[i]; + split_pipe->plane_res.dpp = pool->dpps[i]; + split_pipe->stream_res.opp = pool->opps[i]; + split_pipe->plane_res.mpcc_inst = pool->dpps[i]->inst; + split_pipe->pipe_idx = i; + split_pipe->stream = stream; return i; } @@ -1858,7 +1879,7 @@ bool dc_add_all_planes_for_stream( return add_all_planes_for_stream(dc, stream, &set, 1, context); } -bool is_timing_changed(struct dc_stream_state *cur_stream, +bool dc_is_timing_changed(struct dc_stream_state *cur_stream, struct dc_stream_state *new_stream) { if (cur_stream == NULL) @@ -1883,7 +1904,7 @@ static bool are_stream_backends_same( if (stream_a == NULL || stream_b == NULL) return false; - if (is_timing_changed(stream_a, stream_b)) + if (dc_is_timing_changed(stream_a, stream_b)) return false; if (stream_a->signal != stream_b->signal) @@ -3014,23 +3035,29 @@ static void set_avi_info_frame( hdmi_info.bits.S0_S1 = scan_type; /* C0, C1 : Colorimetry */ - if (color_space == COLOR_SPACE_YCBCR709 || - color_space == COLOR_SPACE_YCBCR709_LIMITED) + switch (color_space) { + case COLOR_SPACE_YCBCR709: + case COLOR_SPACE_YCBCR709_LIMITED: hdmi_info.bits.C0_C1 = COLORIMETRY_ITU709; - else if (color_space == COLOR_SPACE_YCBCR601 || - color_space == COLOR_SPACE_YCBCR601_LIMITED) + break; + case COLOR_SPACE_YCBCR601: + case COLOR_SPACE_YCBCR601_LIMITED: hdmi_info.bits.C0_C1 = COLORIMETRY_ITU601; - else { - hdmi_info.bits.C0_C1 = COLORIMETRY_NO_DATA; - } - if (color_space == COLOR_SPACE_2020_RGB_FULLRANGE || - color_space == COLOR_SPACE_2020_RGB_LIMITEDRANGE || - color_space == COLOR_SPACE_2020_YCBCR) { + break; + case COLOR_SPACE_2020_RGB_FULLRANGE: + case COLOR_SPACE_2020_RGB_LIMITEDRANGE: + case COLOR_SPACE_2020_YCBCR: hdmi_info.bits.EC0_EC2 = COLORIMETRYEX_BT2020RGBYCBCR; hdmi_info.bits.C0_C1 = COLORIMETRY_EXTENDED; - } else if (color_space == COLOR_SPACE_ADOBERGB) { + break; + case COLOR_SPACE_ADOBERGB: hdmi_info.bits.EC0_EC2 = COLORIMETRYEX_ADOBERGB; hdmi_info.bits.C0_C1 = COLORIMETRY_EXTENDED; + break; + case COLOR_SPACE_SRGB: + default: + hdmi_info.bits.C0_C1 = COLORIMETRY_NO_DATA; + break; } if (pixel_encoding && color_space == COLOR_SPACE_2020_YCBCR && @@ -3508,7 +3535,7 @@ bool pipe_need_reprogram( if (pipe_ctx_old->stream_res.stream_enc != pipe_ctx->stream_res.stream_enc) return true; - if (is_timing_changed(pipe_ctx_old->stream, pipe_ctx->stream)) + if (dc_is_timing_changed(pipe_ctx_old->stream, pipe_ctx->stream)) return true; if (pipe_ctx_old->stream->dpms_off != pipe_ctx->stream->dpms_off) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c index 72b261ad95870d646fb06f8de7a83d2659aae1bc..6e11d2b701f82fc5e437eda80dbed72d6d21a998 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c @@ -276,8 +276,8 @@ static void program_cursor_attributes( } dc->hwss.set_cursor_attribute(pipe_ctx); - - dc_send_update_cursor_info_to_dmu(pipe_ctx, i); + if (dc->ctx->dmub_srv) + dc_send_update_cursor_info_to_dmu(pipe_ctx, i); if (dc->hwss.set_cursor_sdr_white_level) dc->hwss.set_cursor_sdr_white_level(pipe_ctx); } @@ -396,8 +396,8 @@ static void program_cursor_position( } dc->hwss.set_cursor_position(pipe_ctx); - - dc_send_update_cursor_info_to_dmu(pipe_ctx, i); + if (dc->ctx->dmub_srv) + dc_send_update_cursor_info_to_dmu(pipe_ctx, i); } if (pipe_to_program) @@ -490,25 +490,6 @@ bool dc_stream_add_writeback(struct dc *dc, struct dwbc *dwb = dc->res_pool->dwbc[wb_info->dwb_pipe_inst]; dwb->otg_inst = stream_status->primary_otg_inst; } - if (IS_DIAG_DC(dc->ctx->dce_environment)) { - if (!dc->hwss.update_bandwidth(dc, dc->current_state)) { - dm_error("DC: update_bandwidth failed!\n"); - return false; - } - - /* enable writeback */ - if (dc->hwss.enable_writeback) { - struct dwbc *dwb = dc->res_pool->dwbc[wb_info->dwb_pipe_inst]; - - if (dwb->funcs->is_enabled(dwb)) { - /* writeback pipe already enabled, only need to update */ - dc->hwss.update_writeback(dc, wb_info, dc->current_state); - } else { - /* Enable writeback pipe from scratch*/ - dc->hwss.enable_writeback(dc, wb_info, dc->current_state); - } - } - } return true; } @@ -553,17 +534,6 @@ bool dc_stream_remove_writeback(struct dc *dc, } stream->num_wb_info = j; - if (IS_DIAG_DC(dc->ctx->dce_environment)) { - /* recalculate and apply DML parameters */ - if (!dc->hwss.update_bandwidth(dc, dc->current_state)) { - dm_error("DC: update_bandwidth failed!\n"); - return false; - } - - /* disable writeback */ - if (dc->hwss.disable_writeback) - dc->hwss.disable_writeback(dc, dwb_pipe_inst); - } return true; } diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index 30f0ba05a6e6c3043b9652aa0ed54c39130fccfd..26d05e2250888f8df3603d14330734eba8484259 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -1,5 +1,5 @@ /* - * Copyright 2012-14 Advanced Micro Devices, Inc. + * Copyright 2012-2023 Advanced Micro Devices, Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -45,7 +45,7 @@ struct aux_payload; struct set_config_cmd_payload; struct dmub_notification; -#define DC_VER "3.2.230" +#define DC_VER "3.2.239" #define MAX_SURFACES 3 #define MAX_PLANES 6 @@ -60,7 +60,9 @@ struct dc_versions { }; enum dp_protocol_version { - DP_VERSION_1_4, + DP_VERSION_1_4 = 0, + DP_VERSION_2_1, + DP_VERSION_UNKNOWN, }; enum dc_plane_type { @@ -209,6 +211,8 @@ struct dc_color_caps { struct dc_dmub_caps { bool psr; bool mclk_sw; + bool subvp_psr; + bool gecc_enable; }; struct dc_caps { @@ -262,6 +266,7 @@ struct dc_caps { uint16_t subvp_pstate_allow_width_us; uint16_t subvp_vertical_int_margin_us; bool seamless_odm; + uint32_t max_v_total; uint8_t subvp_drr_vblank_start_margin_us; }; @@ -270,8 +275,13 @@ struct dc_bug_wa { bool dedcn20_305_wa; bool skip_clock_update; bool lt_early_cr_pattern; + struct { + uint8_t uclk : 1; + uint8_t fclk : 1; + uint8_t dcfclk : 1; + uint8_t dcfclk_ds: 1; + } clock_update_disable_mask; }; - struct dc_dcc_surface_param { struct dc_size surface_size; enum surface_pixel_format format; @@ -419,6 +429,7 @@ enum visual_confirm { VISUAL_CONFIRM_FAMS = 7, VISUAL_CONFIRM_SWIZZLE = 9, VISUAL_CONFIRM_SUBVP = 14, + VISUAL_CONFIRM_MCLK_SWITCH = 16, }; enum dc_psr_power_opts { @@ -698,6 +709,8 @@ struct dc_virtual_addr_space_config { struct dc_bounding_box_overrides { int sr_exit_time_ns; int sr_enter_plus_exit_time_ns; + int sr_exit_z8_time_ns; + int sr_enter_plus_exit_z8_time_ns; int urgent_latency_ns; int percent_of_ideal_drambw; int dram_clock_change_latency_ns; @@ -767,6 +780,8 @@ struct dc_debug_options { int sr_enter_plus_exit_time_dpm0_ns; int sr_exit_time_ns; int sr_enter_plus_exit_time_ns; + int sr_exit_z8_time_ns; + int sr_enter_plus_exit_z8_time_ns; int urgent_latency_ns; uint32_t underflow_assert_delay_us; int percent_of_ideal_drambw; @@ -855,7 +870,6 @@ struct dc_debug_options { bool force_usr_allow; /* uses value at boot and disables switch */ bool disable_dtb_ref_clk_switch; - uint32_t fixed_vs_aux_delay_config_wa; bool extended_blank_optimization; union aux_wake_wa_options aux_wake_wa; uint32_t mst_start_top_delay; @@ -879,6 +893,14 @@ struct dc_debug_options { uint32_t fpo_vactive_margin_us; bool disable_fpo_vactive; bool disable_boot_optimizations; + bool override_odm_optimization; + bool minimize_dispclk_using_odm; + bool disable_subvp_high_refresh; + bool disable_dp_plus_plus_wa; + uint32_t fpo_vactive_min_active_margin_us; + uint32_t fpo_vactive_max_blank_us; + bool enable_legacy_fast_update; + bool disable_dc_mode_overwrite; }; struct gpu_info_soc_bounding_box_v1_0; @@ -1502,6 +1524,7 @@ struct dc_link { /* Forced DPIA into TBT3 compatibility mode. */ bool dpia_forced_tbt3_mode; bool dongle_mode_timing_override; + bool blank_stream_on_ocs_change; } wa_flags; struct link_mst_stream_allocation_table mst_stream_alloc_table; @@ -2126,8 +2149,6 @@ struct dc_sink_init_data { bool converter_disable_audio; }; -bool dc_extended_blank_supported(struct dc *dc); - struct dc_sink *dc_sink_create(const struct dc_sink_init_data *init_params); /* Newer interfaces */ @@ -2220,10 +2241,15 @@ enum dc_status dc_process_dmub_set_mst_slots(const struct dc *dc, void dc_process_dmub_dpia_hpd_int_enable(const struct dc *dc, uint32_t hpd_int_enable); +void dc_print_dmub_diagnostic_data(const struct dc *dc); + /* DSC Interfaces */ #include "dc_dsc.h" /* Disable acc mode Interfaces */ void dc_disable_accelerated_mode(struct dc *dc); +bool dc_is_timing_changed(struct dc_stream_state *cur_stream, + struct dc_stream_state *new_stream); + #endif /* DC_INTERFACE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c index a9b9490a532c2f9872e06d07d330572bbca52425..c52c40b1638722e10a4703de3e808eca0d83a919 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c @@ -65,47 +65,6 @@ void dc_dmub_srv_destroy(struct dc_dmub_srv **dmub_srv) } } -void dc_dmub_srv_cmd_queue(struct dc_dmub_srv *dc_dmub_srv, - union dmub_rb_cmd *cmd) -{ - struct dmub_srv *dmub = dc_dmub_srv->dmub; - struct dc_context *dc_ctx = dc_dmub_srv->ctx; - enum dmub_status status; - - status = dmub_srv_cmd_queue(dmub, cmd); - if (status == DMUB_STATUS_OK) - return; - - if (status != DMUB_STATUS_QUEUE_FULL) - goto error; - - /* Execute and wait for queue to become empty again. */ - dc_dmub_srv_cmd_execute(dc_dmub_srv); - dc_dmub_srv_wait_idle(dc_dmub_srv); - - /* Requeue the command. */ - status = dmub_srv_cmd_queue(dmub, cmd); - if (status == DMUB_STATUS_OK) - return; - -error: - DC_ERROR("Error queuing DMUB command: status=%d\n", status); - dc_dmub_srv_log_diagnostic_data(dc_dmub_srv); -} - -void dc_dmub_srv_cmd_execute(struct dc_dmub_srv *dc_dmub_srv) -{ - struct dmub_srv *dmub = dc_dmub_srv->dmub; - struct dc_context *dc_ctx = dc_dmub_srv->ctx; - enum dmub_status status; - - status = dmub_srv_cmd_execute(dmub); - if (status != DMUB_STATUS_OK) { - DC_ERROR("Error starting DMUB execution: status=%d\n", status); - dc_dmub_srv_log_diagnostic_data(dc_dmub_srv); - } -} - void dc_dmub_srv_wait_idle(struct dc_dmub_srv *dc_dmub_srv) { struct dmub_srv *dmub = dc_dmub_srv->dmub; @@ -159,50 +118,89 @@ void dc_dmub_srv_send_inbox0_cmd(struct dc_dmub_srv *dmub_srv, } } -bool dc_dmub_srv_cmd_with_reply_data(struct dc_dmub_srv *dc_dmub_srv, union dmub_rb_cmd *cmd) +bool dc_dmub_srv_cmd_run(struct dc_dmub_srv *dc_dmub_srv, union dmub_rb_cmd *cmd, enum dm_dmub_wait_type wait_type) +{ + return dc_dmub_srv_cmd_run_list(dc_dmub_srv, 1, cmd, wait_type); +} + +bool dc_dmub_srv_cmd_run_list(struct dc_dmub_srv *dc_dmub_srv, unsigned int count, union dmub_rb_cmd *cmd_list, enum dm_dmub_wait_type wait_type) { + struct dc_context *dc_ctx; struct dmub_srv *dmub; enum dmub_status status; + int i; if (!dc_dmub_srv || !dc_dmub_srv->dmub) return false; + dc_ctx = dc_dmub_srv->ctx; dmub = dc_dmub_srv->dmub; - status = dmub_srv_cmd_with_reply_data(dmub, cmd); + for (i = 0 ; i < count; i++) { + // Queue command + status = dmub_srv_cmd_queue(dmub, &cmd_list[i]); + + if (status == DMUB_STATUS_QUEUE_FULL) { + /* Execute and wait for queue to become empty again. */ + dmub_srv_cmd_execute(dmub); + dmub_srv_wait_for_idle(dmub, 100000); + + /* Requeue the command. */ + status = dmub_srv_cmd_queue(dmub, &cmd_list[i]); + } + + if (status != DMUB_STATUS_OK) { + DC_ERROR("Error queueing DMUB command: status=%d\n", status); + dc_dmub_srv_log_diagnostic_data(dc_dmub_srv); + return false; + } + } + + status = dmub_srv_cmd_execute(dmub); if (status != DMUB_STATUS_OK) { - DC_LOG_DEBUG("No reply for DMUB command: status=%d\n", status); + DC_ERROR("Error starting DMUB execution: status=%d\n", status); + dc_dmub_srv_log_diagnostic_data(dc_dmub_srv); return false; } + // Wait for DMUB to process command + if (wait_type != DM_DMUB_WAIT_TYPE_NO_WAIT) { + status = dmub_srv_wait_for_idle(dmub, 100000); + + if (status != DMUB_STATUS_OK) { + DC_LOG_DEBUG("No reply for DMUB command: status=%d\n", status); + dc_dmub_srv_log_diagnostic_data(dc_dmub_srv); + return false; + } + + // Copy data back from ring buffer into command + if (wait_type == DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) + dmub_rb_get_return_data(&dmub->inbox1_rb, cmd_list); + } + return true; } -void dc_dmub_srv_wait_phy_init(struct dc_dmub_srv *dc_dmub_srv) +bool dc_dmub_srv_optimized_init_done(struct dc_dmub_srv *dc_dmub_srv) { - struct dmub_srv *dmub = dc_dmub_srv->dmub; - struct dc_context *dc_ctx = dc_dmub_srv->ctx; + struct dmub_srv *dmub; + struct dc_context *dc_ctx; + union dmub_fw_boot_status boot_status; enum dmub_status status; - for (;;) { - /* Wait up to a second for PHY init. */ - status = dmub_srv_wait_for_phy_init(dmub, 1000000); - if (status == DMUB_STATUS_OK) - /* Initialization OK */ - break; - - DC_ERROR("DMCUB PHY init failed: status=%d\n", status); - ASSERT(0); + if (!dc_dmub_srv || !dc_dmub_srv->dmub) + return false; - if (status != DMUB_STATUS_TIMEOUT) - /* - * Server likely initialized or we don't have - * DMCUB HW support - this won't end. - */ - break; + dmub = dc_dmub_srv->dmub; + dc_ctx = dc_dmub_srv->ctx; - /* Continue spinning so we don't hang the ASIC. */ + status = dmub_srv_get_fw_boot_status(dmub, &boot_status); + if (status != DMUB_STATUS_OK) { + DC_ERROR("Error querying DMUB boot status: error=%d\n", status); + return false; } + + return boot_status.bits.optimized_init_done; } bool dc_dmub_srv_notify_stream_mask(struct dc_dmub_srv *dc_dmub_srv, @@ -267,9 +265,7 @@ void dc_dmub_srv_drr_update_cmd(struct dc *dc, uint32_t tg_inst, uint32_t vtotal cmd.drr_update.header.payload_bytes = sizeof(cmd.drr_update) - sizeof(cmd.drr_update.header); // Send the command to the DMCUB. - dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv); - dc_dmub_srv_wait_idle(dc->ctx->dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } void dc_dmub_srv_set_drr_manual_trigger_cmd(struct dc *dc, uint32_t tg_inst) @@ -283,9 +279,7 @@ void dc_dmub_srv_set_drr_manual_trigger_cmd(struct dc *dc, uint32_t tg_inst) cmd.drr_update.header.payload_bytes = sizeof(cmd.drr_update) - sizeof(cmd.drr_update.header); // Send the command to the DMCUB. - dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv); - dc_dmub_srv_wait_idle(dc->ctx->dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } static uint8_t dc_dmub_srv_get_pipes_for_stream(struct dc *dc, struct dc_stream_state *stream) @@ -378,21 +372,14 @@ bool dc_dmub_srv_p_state_delegate(struct dc *dc, bool should_manage_pstate, stru sizeof(cmd.fw_assisted_mclk_switch) - sizeof(cmd.fw_assisted_mclk_switch.header); // Send the command to the DMCUB. - dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv); - dc_dmub_srv_wait_idle(dc->ctx->dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } -void dc_dmub_srv_query_caps_cmd(struct dmub_srv *dmub) +void dc_dmub_srv_query_caps_cmd(struct dc_dmub_srv *dc_dmub_srv) { union dmub_rb_cmd cmd = { 0 }; - enum dmub_status status; - - if (!dmub) { - return; - } memset(&cmd, 0, sizeof(cmd)); @@ -402,15 +389,10 @@ void dc_dmub_srv_query_caps_cmd(struct dmub_srv *dmub) cmd.query_feature_caps.header.ret_status = 1; cmd.query_feature_caps.header.payload_bytes = sizeof(struct dmub_cmd_query_feature_caps_data); - /* Send command to fw */ - status = dmub_srv_cmd_with_reply_data(dmub, &cmd); - - ASSERT(status == DMUB_STATUS_OK); - /* If command was processed, copy feature caps to dmub srv */ - if (status == DMUB_STATUS_OK && + if (dm_execute_dmub_cmd(dc_dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) && cmd.query_feature_caps.header.ret_status == 0) { - memcpy(&dmub->feature_caps, + memcpy(&dc_dmub_srv->dmub->feature_caps, &cmd.query_feature_caps.query_feature_caps_data, sizeof(struct dmub_feature_caps)); } @@ -419,7 +401,6 @@ void dc_dmub_srv_query_caps_cmd(struct dmub_srv *dmub) void dc_dmub_srv_get_visual_confirm_color_cmd(struct dc *dc, struct pipe_ctx *pipe_ctx) { union dmub_rb_cmd cmd = { 0 }; - enum dmub_status status; unsigned int panel_inst = 0; dc_get_edp_link_panel_inst(dc, pipe_ctx->stream->link, &panel_inst); @@ -433,13 +414,8 @@ void dc_dmub_srv_get_visual_confirm_color_cmd(struct dc *dc, struct pipe_ctx *pi cmd.visual_confirm_color.header.payload_bytes = sizeof(struct dmub_cmd_visual_confirm_color_data); cmd.visual_confirm_color.visual_confirm_color_data.visual_confirm_color.panel_inst = panel_inst; - // Send command to fw - status = dmub_srv_cmd_with_reply_data(dc->ctx->dmub_srv->dmub, &cmd); - - ASSERT(status == DMUB_STATUS_OK); - // If command was processed, copy feature caps to dmub srv - if (status == DMUB_STATUS_OK && + if (dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) && cmd.visual_confirm_color.header.ret_status == 0) { memcpy(&dc->ctx->dmub_srv->dmub->visual_confirm_color, &cmd.visual_confirm_color.visual_confirm_color_data, @@ -797,9 +773,8 @@ void dc_dmub_setup_subvp_dmub_command(struct dc *dc, cmd.fw_assisted_mclk_switch_v2.config_data.watermark_a_cache = wm_val_refclk < 0xFFFF ? wm_val_refclk : 0xFFFF; } - dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv); - dc_dmub_srv_wait_idle(dc->ctx->dmub_srv); + + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } bool dc_dmub_srv_get_diagnostic_data(struct dc_dmub_srv *dc_dmub_srv, struct dmub_diagnostic_data *diag_data) @@ -823,74 +798,40 @@ void dc_dmub_srv_log_diagnostic_data(struct dc_dmub_srv *dc_dmub_srv) return; } - DC_LOG_DEBUG( - "DMCUB STATE\n" - " dmcub_version : %08x\n" - " scratch [0] : %08x\n" - " scratch [1] : %08x\n" - " scratch [2] : %08x\n" - " scratch [3] : %08x\n" - " scratch [4] : %08x\n" - " scratch [5] : %08x\n" - " scratch [6] : %08x\n" - " scratch [7] : %08x\n" - " scratch [8] : %08x\n" - " scratch [9] : %08x\n" - " scratch [10] : %08x\n" - " scratch [11] : %08x\n" - " scratch [12] : %08x\n" - " scratch [13] : %08x\n" - " scratch [14] : %08x\n" - " scratch [15] : %08x\n" - " pc : %08x\n" - " unk_fault_addr : %08x\n" - " inst_fault_addr : %08x\n" - " data_fault_addr : %08x\n" - " inbox1_rptr : %08x\n" - " inbox1_wptr : %08x\n" - " inbox1_size : %08x\n" - " inbox0_rptr : %08x\n" - " inbox0_wptr : %08x\n" - " inbox0_size : %08x\n" - " is_enabled : %d\n" - " is_soft_reset : %d\n" - " is_secure_reset : %d\n" - " is_traceport_en : %d\n" - " is_cw0_en : %d\n" - " is_cw6_en : %d\n", - diag_data.dmcub_version, - diag_data.scratch[0], - diag_data.scratch[1], - diag_data.scratch[2], - diag_data.scratch[3], - diag_data.scratch[4], - diag_data.scratch[5], - diag_data.scratch[6], - diag_data.scratch[7], - diag_data.scratch[8], - diag_data.scratch[9], - diag_data.scratch[10], - diag_data.scratch[11], - diag_data.scratch[12], - diag_data.scratch[13], - diag_data.scratch[14], - diag_data.scratch[15], - diag_data.pc, - diag_data.undefined_address_fault_addr, - diag_data.inst_fetch_fault_addr, - diag_data.data_write_fault_addr, - diag_data.inbox1_rptr, - diag_data.inbox1_wptr, - diag_data.inbox1_size, - diag_data.inbox0_rptr, - diag_data.inbox0_wptr, - diag_data.inbox0_size, - diag_data.is_dmcub_enabled, - diag_data.is_dmcub_soft_reset, - diag_data.is_dmcub_secure_reset, - diag_data.is_traceport_en, - diag_data.is_cw0_enabled, - diag_data.is_cw6_enabled); + DC_LOG_DEBUG("DMCUB STATE:"); + DC_LOG_DEBUG(" dmcub_version : %08x", diag_data.dmcub_version); + DC_LOG_DEBUG(" scratch [0] : %08x", diag_data.scratch[0]); + DC_LOG_DEBUG(" scratch [1] : %08x", diag_data.scratch[1]); + DC_LOG_DEBUG(" scratch [2] : %08x", diag_data.scratch[2]); + DC_LOG_DEBUG(" scratch [3] : %08x", diag_data.scratch[3]); + DC_LOG_DEBUG(" scratch [4] : %08x", diag_data.scratch[4]); + DC_LOG_DEBUG(" scratch [5] : %08x", diag_data.scratch[5]); + DC_LOG_DEBUG(" scratch [6] : %08x", diag_data.scratch[6]); + DC_LOG_DEBUG(" scratch [7] : %08x", diag_data.scratch[7]); + DC_LOG_DEBUG(" scratch [8] : %08x", diag_data.scratch[8]); + DC_LOG_DEBUG(" scratch [9] : %08x", diag_data.scratch[9]); + DC_LOG_DEBUG(" scratch [10] : %08x", diag_data.scratch[10]); + DC_LOG_DEBUG(" scratch [11] : %08x", diag_data.scratch[11]); + DC_LOG_DEBUG(" scratch [12] : %08x", diag_data.scratch[12]); + DC_LOG_DEBUG(" scratch [13] : %08x", diag_data.scratch[13]); + DC_LOG_DEBUG(" scratch [14] : %08x", diag_data.scratch[14]); + DC_LOG_DEBUG(" scratch [15] : %08x", diag_data.scratch[15]); + DC_LOG_DEBUG(" pc : %08x", diag_data.pc); + DC_LOG_DEBUG(" unk_fault_addr : %08x", diag_data.undefined_address_fault_addr); + DC_LOG_DEBUG(" inst_fault_addr : %08x", diag_data.inst_fetch_fault_addr); + DC_LOG_DEBUG(" data_fault_addr : %08x", diag_data.data_write_fault_addr); + DC_LOG_DEBUG(" inbox1_rptr : %08x", diag_data.inbox1_rptr); + DC_LOG_DEBUG(" inbox1_wptr : %08x", diag_data.inbox1_wptr); + DC_LOG_DEBUG(" inbox1_size : %08x", diag_data.inbox1_size); + DC_LOG_DEBUG(" inbox0_rptr : %08x", diag_data.inbox0_rptr); + DC_LOG_DEBUG(" inbox0_wptr : %08x", diag_data.inbox0_wptr); + DC_LOG_DEBUG(" inbox0_size : %08x", diag_data.inbox0_size); + DC_LOG_DEBUG(" is_enabled : %d", diag_data.is_dmcub_enabled); + DC_LOG_DEBUG(" is_soft_reset : %d", diag_data.is_dmcub_soft_reset); + DC_LOG_DEBUG(" is_secure_reset : %d", diag_data.is_dmcub_secure_reset); + DC_LOG_DEBUG(" is_traceport_en : %d", diag_data.is_traceport_en); + DC_LOG_DEBUG(" is_cw0_en : %d", diag_data.is_cw0_enabled); + DC_LOG_DEBUG(" is_cw6_en : %d", diag_data.is_cw6_enabled); } static bool dc_can_pipe_disable_cursor(struct pipe_ctx *pipe_ctx) @@ -982,14 +923,6 @@ static void dc_build_cursor_update_payload0( payload->panel_inst = panel_inst; } -static void dc_send_cmd_to_dmu(struct dc_dmub_srv *dmub_srv, - union dmub_rb_cmd *cmd) -{ - dc_dmub_srv_cmd_queue(dmub_srv, cmd); - dc_dmub_srv_cmd_execute(dmub_srv); - dc_dmub_srv_wait_idle(dmub_srv); -} - static void dc_build_cursor_position_update_payload0( struct dmub_cmd_update_cursor_payload0 *pl, const uint8_t p_idx, const struct hubp *hubp, const struct dpp *dpp) @@ -1032,9 +965,11 @@ static void dc_build_cursor_attribute_update_payload1( void dc_send_update_cursor_info_to_dmu( struct pipe_ctx *pCtx, uint8_t pipe_idx) { - union dmub_rb_cmd cmd = { 0 }; - union dmub_cmd_update_cursor_info_data *update_cursor_info = - &cmd.update_cursor_info.update_cursor_info_data; + union dmub_rb_cmd cmd[2]; + union dmub_cmd_update_cursor_info_data *update_cursor_info_0 = + &cmd[0].update_cursor_info.update_cursor_info_data; + + memset(cmd, 0, sizeof(cmd)); if (!dc_dmub_should_update_cursor_data(pCtx)) return; @@ -1051,31 +986,28 @@ void dc_send_update_cursor_info_to_dmu( { /* Build Payload#0 Header */ - cmd.update_cursor_info.header.type = DMUB_CMD__UPDATE_CURSOR_INFO; - cmd.update_cursor_info.header.payload_bytes = - sizeof(cmd.update_cursor_info.update_cursor_info_data); - cmd.update_cursor_info.header.multi_cmd_pending = 1; /* To combine multi dmu cmd, 1st cmd */ + cmd[0].update_cursor_info.header.type = DMUB_CMD__UPDATE_CURSOR_INFO; + cmd[0].update_cursor_info.header.payload_bytes = + sizeof(cmd[0].update_cursor_info.update_cursor_info_data); + cmd[0].update_cursor_info.header.multi_cmd_pending = 1; //To combine multi dmu cmd, 1st cmd /* Prepare Payload */ - dc_build_cursor_update_payload0(pCtx, pipe_idx, &update_cursor_info->payload0); + dc_build_cursor_update_payload0(pCtx, pipe_idx, &update_cursor_info_0->payload0); - dc_build_cursor_position_update_payload0(&update_cursor_info->payload0, pipe_idx, + dc_build_cursor_position_update_payload0(&update_cursor_info_0->payload0, pipe_idx, pCtx->plane_res.hubp, pCtx->plane_res.dpp); - /* Send update_curosr_info to queue */ - dc_dmub_srv_cmd_queue(pCtx->stream->ctx->dmub_srv, &cmd); - } + } { /* Build Payload#1 Header */ - memset(update_cursor_info, 0, sizeof(union dmub_cmd_update_cursor_info_data)); - cmd.update_cursor_info.header.type = DMUB_CMD__UPDATE_CURSOR_INFO; - cmd.update_cursor_info.header.payload_bytes = sizeof(struct cursor_attributes_cfg); - cmd.update_cursor_info.header.multi_cmd_pending = 0; /* Indicate it's the last command. */ + cmd[1].update_cursor_info.header.type = DMUB_CMD__UPDATE_CURSOR_INFO; + cmd[1].update_cursor_info.header.payload_bytes = sizeof(struct cursor_attributes_cfg); + cmd[1].update_cursor_info.header.multi_cmd_pending = 0; //Indicate it's the last command. dc_build_cursor_attribute_update_payload1( - &cmd.update_cursor_info.update_cursor_info_data.payload1.attribute_cfg, + &cmd[1].update_cursor_info.update_cursor_info_data.payload1.attribute_cfg, pipe_idx, pCtx->plane_res.hubp, pCtx->plane_res.dpp); /* Combine 2nd cmds update_curosr_info to DMU */ - dc_send_cmd_to_dmu(pCtx->stream->ctx->dmub_srv, &cmd); + dm_execute_dmub_cmd_list(pCtx->stream->ctx, 2, cmd, DM_DMUB_WAIT_TYPE_WAIT); } } diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h index d34f5563df2ec374f6c1b0543afc61ddc42b6c28..a5196a9292b356576b4bc0ef596da76b45da26b7 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h @@ -26,7 +26,7 @@ #ifndef _DMUB_DC_SRV_H_ #define _DMUB_DC_SRV_H_ -#include "os_types.h" +#include "dm_services_types.h" #include "dmub/dmub_srv.h" struct dmub_srv; @@ -52,16 +52,13 @@ struct dc_dmub_srv { void *dm; }; -void dc_dmub_srv_cmd_queue(struct dc_dmub_srv *dc_dmub_srv, - union dmub_rb_cmd *cmd); - -void dc_dmub_srv_cmd_execute(struct dc_dmub_srv *dc_dmub_srv); - void dc_dmub_srv_wait_idle(struct dc_dmub_srv *dc_dmub_srv); -void dc_dmub_srv_wait_phy_init(struct dc_dmub_srv *dc_dmub_srv); +bool dc_dmub_srv_optimized_init_done(struct dc_dmub_srv *dc_dmub_srv); + +bool dc_dmub_srv_cmd_run(struct dc_dmub_srv *dc_dmub_srv, union dmub_rb_cmd *cmd, enum dm_dmub_wait_type wait_type); -bool dc_dmub_srv_cmd_with_reply_data(struct dc_dmub_srv *dc_dmub_srv, union dmub_rb_cmd *cmd); +bool dc_dmub_srv_cmd_run_list(struct dc_dmub_srv *dc_dmub_srv, unsigned int count, union dmub_rb_cmd *cmd_list, enum dm_dmub_wait_type wait_type); bool dc_dmub_srv_notify_stream_mask(struct dc_dmub_srv *dc_dmub_srv, unsigned int stream_mask); @@ -77,7 +74,7 @@ void dc_dmub_srv_drr_update_cmd(struct dc *dc, uint32_t tg_inst, uint32_t vtotal void dc_dmub_srv_set_drr_manual_trigger_cmd(struct dc *dc, uint32_t tg_inst); bool dc_dmub_srv_p_state_delegate(struct dc *dc, bool enable_pstate, struct dc_state *context); -void dc_dmub_srv_query_caps_cmd(struct dmub_srv *dmub); +void dc_dmub_srv_query_caps_cmd(struct dc_dmub_srv *dc_dmub_srv); void dc_dmub_srv_get_visual_confirm_color_cmd(struct dc *dc, struct pipe_ctx *pipe_ctx); void dc_dmub_srv_clear_inbox0_ack(struct dc_dmub_srv *dmub_srv); void dc_dmub_srv_wait_for_inbox0_ack(struct dc_dmub_srv *dmub_srv); diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h index 49aab1924665a750eeb71d6992537d62534f4f75..55139d7bf422cb2dcb388d599308f4e304269f3d 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h @@ -61,7 +61,7 @@ enum dc_link_rate { */ LINK_RATE_UHBR10 = 1000, // UHBR10 - 10.0 Gbps/Lane LINK_RATE_UHBR13_5 = 1350, // UHBR13.5 - 13.5 Gbps/Lane - LINK_RATE_UHBR20 = 2000, // UHBR10 - 20.0 Gbps/Lane + LINK_RATE_UHBR20 = 2000, // UHBR20 - 20.0 Gbps/Lane }; enum dc_link_spread { @@ -566,6 +566,12 @@ struct dpcd_amd_device_id { uint8_t dal_version_byte2; }; +struct target_luminance_value { + uint8_t byte0; + uint8_t byte1; + uint8_t byte2; +}; + struct dpcd_source_backlight_set { struct { uint8_t byte0; @@ -1225,6 +1231,7 @@ struct dpcd_caps { union dp_main_line_channel_coding_cap channel_coding_cap; union dp_sink_video_fallback_formats fallback_formats; union dp_fec_capability1 fec_cap1; + bool panel_luminance_control; union dp_cable_id cable_id; uint8_t edp_rev; union edp_alpm_caps alpm_caps; diff --git a/drivers/gpu/drm/amd/display/dc/dc_dsc.h b/drivers/gpu/drm/amd/display/dc/dc_dsc.h index 0e92a322c2ed3e534f5ca123a32710a4620fc36d..9491b76d61f58f572a3c4a48100d082a5d3045bc 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dsc.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dsc.h @@ -58,6 +58,7 @@ struct dc_dsc_config_options { uint32_t dsc_min_slice_height_override; uint32_t max_target_bpp_limit_override_x16; uint32_t slice_height_granularity; + uint32_t dsc_force_odm_hslice_override; }; bool dc_dsc_parse_dsc_dpcd(const struct dc *dc, diff --git a/drivers/gpu/drm/amd/display/dc/dc_helper.c b/drivers/gpu/drm/amd/display/dc/dc_helper.c index f43cce16bb6ce78288e64efc5a835988637bcab3..3907eeff560ce7880ae6571db17b82154c76123e 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dc_helper.c @@ -41,19 +41,13 @@ static inline void submit_dmub_read_modify_write( const struct dc_context *ctx) { struct dmub_rb_cmd_read_modify_write *cmd_buf = &offload->cmd_data.read_modify_write; - bool gather = false; offload->should_burst_write = (offload->same_addr_count == (DMUB_READ_MODIFY_WRITE_SEQ__MAX - 1)); cmd_buf->header.payload_bytes = sizeof(struct dmub_cmd_read_modify_write_sequence) * offload->reg_seq_count; - gather = ctx->dmub_srv->reg_helper_offload.gather_in_progress; - ctx->dmub_srv->reg_helper_offload.gather_in_progress = false; - - dc_dmub_srv_cmd_queue(ctx->dmub_srv, &offload->cmd_data); - - ctx->dmub_srv->reg_helper_offload.gather_in_progress = gather; + dm_execute_dmub_cmd(ctx, &offload->cmd_data, DM_DMUB_WAIT_TYPE_NO_WAIT); memset(cmd_buf, 0, sizeof(*cmd_buf)); @@ -66,17 +60,11 @@ static inline void submit_dmub_burst_write( const struct dc_context *ctx) { struct dmub_rb_cmd_burst_write *cmd_buf = &offload->cmd_data.burst_write; - bool gather = false; cmd_buf->header.payload_bytes = sizeof(uint32_t) * offload->reg_seq_count; - gather = ctx->dmub_srv->reg_helper_offload.gather_in_progress; - ctx->dmub_srv->reg_helper_offload.gather_in_progress = false; - - dc_dmub_srv_cmd_queue(ctx->dmub_srv, &offload->cmd_data); - - ctx->dmub_srv->reg_helper_offload.gather_in_progress = gather; + dm_execute_dmub_cmd(ctx, &offload->cmd_data, DM_DMUB_WAIT_TYPE_NO_WAIT); memset(cmd_buf, 0, sizeof(*cmd_buf)); @@ -88,17 +76,11 @@ static inline void submit_dmub_reg_wait( const struct dc_context *ctx) { struct dmub_rb_cmd_reg_wait *cmd_buf = &offload->cmd_data.reg_wait; - bool gather = false; - - gather = ctx->dmub_srv->reg_helper_offload.gather_in_progress; - ctx->dmub_srv->reg_helper_offload.gather_in_progress = false; - dc_dmub_srv_cmd_queue(ctx->dmub_srv, &offload->cmd_data); + dm_execute_dmub_cmd(ctx, &offload->cmd_data, DM_DMUB_WAIT_TYPE_NO_WAIT); memset(cmd_buf, 0, sizeof(*cmd_buf)); offload->reg_seq_count = 0; - - ctx->dmub_srv->reg_helper_offload.gather_in_progress = gather; } struct dc_reg_value_masks { @@ -151,7 +133,6 @@ static void dmub_flush_buffer_execute( const struct dc_context *ctx) { submit_dmub_read_modify_write(offload, ctx); - dc_dmub_srv_cmd_execute(ctx->dmub_srv); } static void dmub_flush_burst_write_buffer_execute( @@ -159,7 +140,6 @@ static void dmub_flush_burst_write_buffer_execute( const struct dc_context *ctx) { submit_dmub_burst_write(offload, ctx); - dc_dmub_srv_cmd_execute(ctx->dmub_srv); } static bool dmub_reg_value_burst_set_pack(const struct dc_context *ctx, uint32_t addr, @@ -484,8 +464,7 @@ void generic_reg_wait(const struct dc_context *ctx, field_value = get_reg_field_value_ex(reg_val, mask, shift); if (field_value == condition_value) { - if (i * delay_between_poll_us > 1000 && - !IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) + if (i * delay_between_poll_us > 1000) DC_LOG_DC("REG_WAIT taking a while: %dms in %s line:%d\n", delay_between_poll_us * i / 1000, func_name, line); @@ -497,8 +476,7 @@ void generic_reg_wait(const struct dc_context *ctx, delay_between_poll_us, time_out_num_tries, func_name, line); - if (!IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) - BREAK_TO_DEBUGGER(); + BREAK_TO_DEBUGGER(); } void generic_write_indirect_reg(const struct dc_context *ctx, @@ -691,8 +669,6 @@ void reg_sequence_start_execute(const struct dc_context *ctx) default: return; } - - dc_dmub_srv_cmd_execute(ctx->dmub_srv); } } @@ -712,3 +688,59 @@ void reg_sequence_wait_done(const struct dc_context *ctx) dc_dmub_srv_wait_idle(ctx->dmub_srv); } } + +char *dce_version_to_string(const int version) +{ + switch (version) { + case DCE_VERSION_8_0: + return "DCE 8.0"; + case DCE_VERSION_8_1: + return "DCE 8.1"; + case DCE_VERSION_8_3: + return "DCE 8.3"; + case DCE_VERSION_10_0: + return "DCE 10.0"; + case DCE_VERSION_11_0: + return "DCE 11.0"; + case DCE_VERSION_11_2: + return "DCE 11.2"; + case DCE_VERSION_11_22: + return "DCE 11.22"; + case DCE_VERSION_12_0: + return "DCE 12.0"; + case DCE_VERSION_12_1: + return "DCE 12.1"; + case DCN_VERSION_1_0: + return "DCN 1.0"; + case DCN_VERSION_1_01: + return "DCN 1.0.1"; + case DCN_VERSION_2_0: + return "DCN 2.0"; + case DCN_VERSION_2_1: + return "DCN 2.1"; + case DCN_VERSION_2_01: + return "DCN 2.0.1"; + case DCN_VERSION_3_0: + return "DCN 3.0"; + case DCN_VERSION_3_01: + return "DCN 3.0.1"; + case DCN_VERSION_3_02: + return "DCN 3.0.2"; + case DCN_VERSION_3_03: + return "DCN 3.0.3"; + case DCN_VERSION_3_1: + return "DCN 3.1"; + case DCN_VERSION_3_14: + return "DCN 3.1.4"; + case DCN_VERSION_3_15: + return "DCN 3.1.5"; + case DCN_VERSION_3_16: + return "DCN 3.1.6"; + case DCN_VERSION_3_2: + return "DCN 3.2"; + case DCN_VERSION_3_21: + return "DCN 3.2.1"; + default: + return "Unknown"; + } +} diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h index 25284006019c3b6d9fdb5c881ea5314ea601632f..3697ea1d14c1bf5f219f40e1cbc7e71cc8a36799 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_stream.h +++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h @@ -131,6 +131,7 @@ union stream_update_flags { uint32_t dsc_changed : 1; uint32_t mst_bw : 1; uint32_t crtc_timing_adjust : 1; + uint32_t fams_changed : 1; } bits; uint32_t raw; @@ -171,6 +172,10 @@ struct mall_temp_config { bool is_phantom_plane[MAX_PIPES]; }; +struct dc_stream_debug_options { + char force_odm_combine_segments; +}; + struct dc_stream_state { // sink is deprecated, new code should not reference // this pointer @@ -181,6 +186,7 @@ struct dc_stream_state { * a stream via the volatile dc_state rather than the static dc_link. */ struct link_encoder *link_enc; + struct dc_stream_debug_options debug; struct dc_panel_patch sink_patches; union display_content_support content_support; struct dc_crtc_timing timing; @@ -227,6 +233,7 @@ struct dc_stream_state { */ bool vrr_active_variable; bool freesync_on_desktop; + bool vrr_active_fixed; bool converter_disable_audio; uint8_t qs_bit; @@ -295,6 +302,7 @@ struct dc_stream_state { bool vblank_synchronized; bool fpo_in_use; struct mall_stream_config mall_stream_config; + bool skip_edp_power_down; }; #define ABM_LEVEL_IMMEDIATE_DISABLE 255 @@ -320,6 +328,7 @@ struct dc_stream_update { bool integer_scaling_update; bool *allow_freesync; bool *vrr_active_variable; + bool *vrr_active_fixed; struct colorspace_transform *gamut_remap; enum dc_color_space *output_color_space; diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index 45ab48fe5d004b9c02d81458814e741a2437b905..0ce7728a5a4be5671279de5300b1e569539490ee 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -69,13 +69,6 @@ enum dce_environment { DCE_ENV_VIRTUAL_HW }; -/* Note: use these macro definitions instead of direct comparison! */ -#define IS_FPGA_MAXIMUS_DC(dce_environment) \ - (dce_environment == DCE_ENV_FPGA_MAXIMUS) - -#define IS_DIAG_DC(dce_environment) \ - (IS_FPGA_MAXIMUS_DC(dce_environment) || (dce_environment == DCE_ENV_DIAG)) - struct dc_perf_trace { unsigned long read_count; unsigned long write_count; @@ -83,7 +76,7 @@ struct dc_perf_trace { unsigned long last_entry_write; }; -#define MAX_SURFACE_NUM 4 +#define MAX_SURFACE_NUM 6 #define NUM_PIXEL_FORMATS 10 enum tiling_mode { @@ -196,6 +189,7 @@ struct dc_panel_patch { unsigned int disable_fams; unsigned int skip_avmute; unsigned int mst_start_top_delay; + unsigned int delay_disable_aux_intercept_ms; }; struct dc_edid_caps { @@ -603,6 +597,7 @@ enum dc_psr_state { PSR_STATE4b_FULL_FRAME, PSR_STATE4c_FULL_FRAME, PSR_STATE4_FULL_FRAME_POWERUP, + PSR_STATE4_FULL_FRAME_HW_LOCK, PSR_STATE5, PSR_STATE5a, PSR_STATE5b, diff --git a/drivers/gpu/drm/amd/display/dc/dce/Makefile b/drivers/gpu/drm/amd/display/dc/dce/Makefile index 0d7db132a20fe048106129d743bd7eb3613285a6..01490c9ba95878de06c6e4099a1a9aa9cb2ff70c 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dce/Makefile @@ -29,7 +29,7 @@ DCE = dce_audio.o dce_stream_encoder.o dce_link_encoder.o dce_hwseq.o \ dce_mem_input.o dce_clock_source.o dce_scl_filters.o dce_transform.o \ dce_opp.o dce_dmcu.o dce_abm.o dce_ipp.o dce_aux.o \ -dce_i2c.o dce_i2c_hw.o dce_i2c_sw.o dmub_psr.o dmub_abm.o dce_panel_cntl.o \ +dce_i2c.o dce_i2c_hw.o dce_i2c_sw.o dmub_psr.o dmub_abm.o dmub_abm_lcd.o dce_panel_cntl.o \ dmub_hw_lock_mgr.o dmub_outbox.o AMD_DAL_DCE = $(addprefix $(AMDDALPATH)/dc/dce/,$(DCE)) diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c index 462c7a3ec3ccbc2ac8413f917f500d86e1eb0c4c..ed8936405dfa5f324e991f5b5f3d11853c25c9c6 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c @@ -920,25 +920,6 @@ static bool dce112_program_pix_clk( struct dce110_clk_src *clk_src = TO_DCE110_CLK_SRC(clock_source); struct bp_pixel_clock_parameters bp_pc_params = {0}; - if (IS_FPGA_MAXIMUS_DC(clock_source->ctx->dce_environment)) { - unsigned int inst = pix_clk_params->controller_id - CONTROLLER_ID_D0; - unsigned dp_dto_ref_100hz = 7000000; - unsigned clock_100hz = pll_settings->actual_pix_clk_100hz; - - /* Set DTO values: phase = target clock, modulo = reference clock */ - REG_WRITE(PHASE[inst], clock_100hz); - REG_WRITE(MODULO[inst], dp_dto_ref_100hz); - - /* Enable DTO */ - if (clk_src->cs_mask->PIPE0_DTO_SRC_SEL) - REG_UPDATE_2(PIXEL_RATE_CNTL[inst], - DP_DTO0_ENABLE, 1, - PIPE0_DTO_SRC_SEL, 1); - else - REG_UPDATE(PIXEL_RATE_CNTL[inst], - DP_DTO0_ENABLE, 1); - return true; - } /* First disable SS * ATOMBIOS will enable by default SS on PLL for DP, * do not disable it here @@ -1015,25 +996,6 @@ static bool dcn31_program_pix_clk( REG_UPDATE(PIXEL_RATE_CNTL[inst], DP_DTO0_ENABLE, 1); } else { - if (IS_FPGA_MAXIMUS_DC(clock_source->ctx->dce_environment)) { - unsigned int inst = pix_clk_params->controller_id - CONTROLLER_ID_D0; - unsigned dp_dto_ref_100hz = 7000000; - unsigned clock_100hz = pll_settings->actual_pix_clk_100hz; - - /* Set DTO values: phase = target clock, modulo = reference clock */ - REG_WRITE(PHASE[inst], clock_100hz); - REG_WRITE(MODULO[inst], dp_dto_ref_100hz); - - /* Enable DTO */ - if (clk_src->cs_mask->PIPE0_DTO_SRC_SEL) - REG_UPDATE_2(PIXEL_RATE_CNTL[inst], - DP_DTO0_ENABLE, 1, - PIPE0_DTO_SRC_SEL, 1); - else - REG_UPDATE(PIXEL_RATE_CNTL[inst], - DP_DTO0_ENABLE, 1); - return true; - } if (clk_src->cs_mask->PIPE0_DTO_SRC_SEL) REG_UPDATE(PIXEL_RATE_CNTL[inst], diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c index e74266cc0098649a58ca39acacf1f13f0b682169..63009db8b5a7237476ea690032c7ad6004e775ba 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c @@ -1093,11 +1093,9 @@ static void dcn21_dmcu_construct( dce_dmcu_construct(dmcu_dce, ctx, regs, dmcu_shift, dmcu_mask); - if (!IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) { - psp_version = dm_read_reg(ctx, mmMP0_SMN_C2PMSG_58); - dmcu_dce->base.auto_load_dmcu = ((psp_version & 0x00FF00FF) > 0x00110029); - dmcu_dce->base.psp_version = psp_version; - } + psp_version = dm_read_reg(ctx, mmMP0_SMN_C2PMSG_58); + dmcu_dce->base.auto_load_dmcu = ((psp_version & 0x00FF00FF) > 0x00110029); + dmcu_dce->base.psp_version = psp_version; } struct dmcu *dce_dmcu_create( diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c index 9fc48208c2e42ab19afb3584f9d7ca93a3a4bdd9..2fb9572ce25dbb80f8a9e2432542a217f0bd415c 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c @@ -24,212 +24,139 @@ */ #include "dmub_abm.h" -#include "dce_abm.h" +#include "dmub_abm_lcd.h" #include "dc.h" -#include "dc_dmub_srv.h" -#include "dmub/dmub_srv.h" #include "core_types.h" -#include "dm_services.h" -#include "reg_helper.h" -#include "fixed31_32.h" - -#include "atom.h" #define TO_DMUB_ABM(abm)\ container_of(abm, struct dce_abm, base) -#define REG(reg) \ - (dce_abm->regs->reg) - -#undef FN -#define FN(reg_name, field_name) \ - dce_abm->abm_shift->field_name, dce_abm->abm_mask->field_name - -#define CTX \ - dce_abm->base.ctx - -#define DISABLE_ABM_IMMEDIATELY 255 - - +#define ABM_FEATURE_NO_SUPPORT 0 +#define ABM_LCD_SUPPORT 1 -static void dmub_abm_enable_fractional_pwm(struct dc_context *dc) +static unsigned int abm_feature_support(struct abm *abm, unsigned int panel_inst) { - union dmub_rb_cmd cmd; - uint32_t fractional_pwm = (dc->dc->config.disable_fractional_pwm == false) ? 1 : 0; - uint32_t edp_id_count = dc->dc_edp_id_count; + struct dc_context *dc = abm->ctx; + struct dc_link *edp_links[MAX_NUM_EDP]; int i; - uint8_t panel_mask = 0; - - for (i = 0; i < edp_id_count; i++) - panel_mask |= 0x01 << i; - - memset(&cmd, 0, sizeof(cmd)); - cmd.abm_set_pwm_frac.header.type = DMUB_CMD__ABM; - cmd.abm_set_pwm_frac.header.sub_type = DMUB_CMD__ABM_SET_PWM_FRAC; - cmd.abm_set_pwm_frac.abm_set_pwm_frac_data.fractional_pwm = fractional_pwm; - cmd.abm_set_pwm_frac.abm_set_pwm_frac_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1; - cmd.abm_set_pwm_frac.abm_set_pwm_frac_data.panel_mask = panel_mask; - cmd.abm_set_pwm_frac.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_pwm_frac_data); - - dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->dmub_srv); - dc_dmub_srv_wait_idle(dc->dmub_srv); -} - -static void dmub_abm_init(struct abm *abm, uint32_t backlight) -{ - struct dce_abm *dce_abm = TO_DMUB_ABM(abm); - - REG_WRITE(DC_ABM1_HG_SAMPLE_RATE, 0x3); - REG_WRITE(DC_ABM1_HG_SAMPLE_RATE, 0x1); - REG_WRITE(DC_ABM1_LS_SAMPLE_RATE, 0x3); - REG_WRITE(DC_ABM1_LS_SAMPLE_RATE, 0x1); - REG_WRITE(BL1_PWM_BL_UPDATE_SAMPLE_RATE, 0x1); - - REG_SET_3(DC_ABM1_HG_MISC_CTRL, 0, - ABM1_HG_NUM_OF_BINS_SEL, 0, - ABM1_HG_VMAX_SEL, 1, - ABM1_HG_BIN_BITWIDTH_SIZE_SEL, 0); - - REG_SET_3(DC_ABM1_IPCSC_COEFF_SEL, 0, - ABM1_IPCSC_COEFF_SEL_R, 2, - ABM1_IPCSC_COEFF_SEL_G, 4, - ABM1_IPCSC_COEFF_SEL_B, 2); - - REG_UPDATE(BL1_PWM_CURRENT_ABM_LEVEL, - BL1_PWM_CURRENT_ABM_LEVEL, backlight); - - REG_UPDATE(BL1_PWM_TARGET_ABM_LEVEL, - BL1_PWM_TARGET_ABM_LEVEL, backlight); + int edp_num; + unsigned int ret = ABM_FEATURE_NO_SUPPORT; - REG_UPDATE(BL1_PWM_USER_LEVEL, - BL1_PWM_USER_LEVEL, backlight); + dc_get_edp_links(dc->dc, edp_links, &edp_num); - REG_UPDATE_2(DC_ABM1_LS_MIN_MAX_PIXEL_VALUE_THRES, - ABM1_LS_MIN_PIXEL_VALUE_THRES, 0, - ABM1_LS_MAX_PIXEL_VALUE_THRES, 1000); + for (i = 0; i < edp_num; i++) { + if (panel_inst == i) + break; + } - REG_SET_3(DC_ABM1_HGLS_REG_READ_PROGRESS, 0, - ABM1_HG_REG_READ_MISSED_FRAME_CLEAR, 1, - ABM1_LS_REG_READ_MISSED_FRAME_CLEAR, 1, - ABM1_BL_REG_READ_MISSED_FRAME_CLEAR, 1); + if (i < edp_num) { + ret = ABM_LCD_SUPPORT; + } - dmub_abm_enable_fractional_pwm(abm->ctx); + return ret; } -static unsigned int dmub_abm_get_current_backlight(struct abm *abm) +static void dmub_abm_init_ex(struct abm *abm, uint32_t backlight) { - struct dce_abm *dce_abm = TO_DMUB_ABM(abm); - unsigned int backlight = REG_READ(BL1_PWM_CURRENT_ABM_LEVEL); - - /* return backlight in hardware format which is unsigned 17 bits, with - * 1 bit integer and 16 bit fractional - */ - return backlight; + dmub_abm_init(abm, backlight); } -static unsigned int dmub_abm_get_target_backlight(struct abm *abm) +static unsigned int dmub_abm_get_current_backlight_ex(struct abm *abm) { - struct dce_abm *dce_abm = TO_DMUB_ABM(abm); - unsigned int backlight = REG_READ(BL1_PWM_TARGET_ABM_LEVEL); + return dmub_abm_get_current_backlight(abm); +} - /* return backlight in hardware format which is unsigned 17 bits, with - * 1 bit integer and 16 bit fractional - */ - return backlight; +static unsigned int dmub_abm_get_target_backlight_ex(struct abm *abm) +{ + return dmub_abm_get_target_backlight(abm); } -static bool dmub_abm_set_level(struct abm *abm, uint32_t level) +static bool dmub_abm_set_level_ex(struct abm *abm, uint32_t level) { - union dmub_rb_cmd cmd; - struct dc_context *dc = abm->ctx; - struct dc_link *edp_links[MAX_NUM_EDP]; - int i; - int edp_num; - uint8_t panel_mask = 0; + bool ret = false; + unsigned int feature_support, i; + uint8_t panel_mask0 = 0; - dc_get_edp_links(dc->dc, edp_links, &edp_num); + for (i = 0; i < MAX_NUM_EDP; i++) { + feature_support = abm_feature_support(abm, i); - for (i = 0; i < edp_num; i++) { - if (edp_links[i]->link_status.link_active) - panel_mask |= (0x01 << i); + if (feature_support == ABM_LCD_SUPPORT) + panel_mask0 |= (0x01 << i); } - memset(&cmd, 0, sizeof(cmd)); - cmd.abm_set_level.header.type = DMUB_CMD__ABM; - cmd.abm_set_level.header.sub_type = DMUB_CMD__ABM_SET_LEVEL; - cmd.abm_set_level.abm_set_level_data.level = level; - cmd.abm_set_level.abm_set_level_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1; - cmd.abm_set_level.abm_set_level_data.panel_mask = panel_mask; - cmd.abm_set_level.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_level_data); + if (panel_mask0) + ret = dmub_abm_set_level(abm, level, panel_mask0); - dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->dmub_srv); - dc_dmub_srv_wait_idle(dc->dmub_srv); - - return true; + return ret; } -static bool dmub_abm_init_config(struct abm *abm, +static bool dmub_abm_init_config_ex(struct abm *abm, const char *src, unsigned int bytes, unsigned int inst) { - union dmub_rb_cmd cmd; - struct dc_context *dc = abm->ctx; - uint8_t panel_mask = 0x01 << inst; + unsigned int feature_support; + + feature_support = abm_feature_support(abm, inst); + + if (feature_support == ABM_LCD_SUPPORT) + dmub_abm_init_config(abm, src, bytes, inst); + + return true; +} - // TODO: Optimize by only reading back final 4 bytes - dmub_flush_buffer_mem(&dc->dmub_srv->dmub->scratch_mem_fb); +static bool dmub_abm_set_pause_ex(struct abm *abm, bool pause, unsigned int panel_inst, unsigned int stream_inst) +{ + bool ret = false; + unsigned int feature_support; - // Copy iramtable into cw7 - memcpy(dc->dmub_srv->dmub->scratch_mem_fb.cpu_addr, (void *)src, bytes); + feature_support = abm_feature_support(abm, panel_inst); - memset(&cmd, 0, sizeof(cmd)); - // Fw will copy from cw7 to fw_state - cmd.abm_init_config.header.type = DMUB_CMD__ABM; - cmd.abm_init_config.header.sub_type = DMUB_CMD__ABM_INIT_CONFIG; - cmd.abm_init_config.abm_init_config_data.src.quad_part = dc->dmub_srv->dmub->scratch_mem_fb.gpu_addr; - cmd.abm_init_config.abm_init_config_data.bytes = bytes; - cmd.abm_init_config.abm_init_config_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1; - cmd.abm_init_config.abm_init_config_data.panel_mask = panel_mask; + if (feature_support == ABM_LCD_SUPPORT) + ret = dmub_abm_set_pause(abm, pause, panel_inst, stream_inst); - cmd.abm_init_config.header.payload_bytes = sizeof(struct dmub_cmd_abm_init_config_data); + return ret; +} - dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->dmub_srv); - dc_dmub_srv_wait_idle(dc->dmub_srv); +static bool dmub_abm_set_pipe_ex(struct abm *abm, uint32_t otg_inst, uint32_t option, uint32_t panel_inst) +{ + bool ret = false; + unsigned int feature_support; - return true; + feature_support = abm_feature_support(abm, panel_inst); + + if (feature_support == ABM_LCD_SUPPORT) + ret = dmub_abm_set_pipe(abm, otg_inst, option, panel_inst); + + return ret; } -static bool dmub_abm_set_pause(struct abm *abm, bool pause, unsigned int panel_inst, unsigned int stream_inst) +static bool dmub_abm_set_backlight_level_pwm_ex(struct abm *abm, + unsigned int backlight_pwm_u16_16, + unsigned int frame_ramp, + unsigned int controller_id, + unsigned int panel_inst) { - union dmub_rb_cmd cmd; - struct dc_context *dc = abm->ctx; - uint8_t panel_mask = 0x01 << panel_inst; + bool ret = false; + unsigned int feature_support; - memset(&cmd, 0, sizeof(cmd)); - cmd.abm_pause.header.type = DMUB_CMD__ABM; - cmd.abm_pause.header.sub_type = DMUB_CMD__ABM_PAUSE; - cmd.abm_pause.abm_pause_data.enable = pause; - cmd.abm_pause.abm_pause_data.panel_mask = panel_mask; - cmd.abm_set_level.header.payload_bytes = sizeof(struct dmub_cmd_abm_pause_data); + feature_support = abm_feature_support(abm, panel_inst); - dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->dmub_srv); - dc_dmub_srv_wait_idle(dc->dmub_srv); + if (feature_support == ABM_LCD_SUPPORT) + ret = dmub_abm_set_backlight_level(abm, backlight_pwm_u16_16, frame_ramp, panel_inst); - return true; + return ret; } static const struct abm_funcs abm_funcs = { - .abm_init = dmub_abm_init, - .set_abm_level = dmub_abm_set_level, - .get_current_backlight = dmub_abm_get_current_backlight, - .get_target_backlight = dmub_abm_get_target_backlight, - .init_abm_config = dmub_abm_init_config, - .set_abm_pause = dmub_abm_set_pause, + .abm_init = dmub_abm_init_ex, + .set_abm_level = dmub_abm_set_level_ex, + .get_current_backlight = dmub_abm_get_current_backlight_ex, + .get_target_backlight = dmub_abm_get_target_backlight_ex, + .init_abm_config = dmub_abm_init_config_ex, + .set_abm_pause = dmub_abm_set_pause_ex, + .set_pipe_ex = dmub_abm_set_pipe_ex, + .set_backlight_level_pwm = dmub_abm_set_backlight_level_pwm_ex, }; static void dmub_abm_construct( @@ -256,16 +183,19 @@ struct abm *dmub_abm_create( const struct dce_abm_shift *abm_shift, const struct dce_abm_mask *abm_mask) { - struct dce_abm *abm_dce = kzalloc(sizeof(*abm_dce), GFP_KERNEL); + if (ctx->dc->caps.dmcub_support) { + struct dce_abm *abm_dce = kzalloc(sizeof(*abm_dce), GFP_KERNEL); - if (abm_dce == NULL) { - BREAK_TO_DEBUGGER(); - return NULL; - } + if (abm_dce == NULL) { + BREAK_TO_DEBUGGER(); + return NULL; + } - dmub_abm_construct(abm_dce, ctx, regs, abm_shift, abm_mask); + dmub_abm_construct(abm_dce, ctx, regs, abm_shift, abm_mask); - return &abm_dce->base; + return &abm_dce->base; + } + return NULL; } void dmub_abm_destroy(struct abm **abm) diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c new file mode 100644 index 0000000000000000000000000000000000000000..39da73eba86e503416c7fc90c975e2fad58d9414 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c @@ -0,0 +1,252 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "dmub_abm.h" +#include "dmub_abm_lcd.h" +#include "dce_abm.h" +#include "dc.h" +#include "dc_dmub_srv.h" +#include "dmub/dmub_srv.h" +#include "core_types.h" +#include "dm_services.h" +#include "reg_helper.h" +#include "fixed31_32.h" + +#ifdef _WIN32 +#include "atombios.h" +#else +#include "atom.h" +#endif + +#define TO_DMUB_ABM(abm)\ + container_of(abm, struct dce_abm, base) + +#define REG(reg) \ + (dce_abm->regs->reg) + +#undef FN +#define FN(reg_name, field_name) \ + dce_abm->abm_shift->field_name, dce_abm->abm_mask->field_name + +#define CTX \ + dce_abm->base.ctx + +#define DISABLE_ABM_IMMEDIATELY 255 + + + +static void dmub_abm_enable_fractional_pwm(struct dc_context *dc) +{ + union dmub_rb_cmd cmd; + uint32_t fractional_pwm = (dc->dc->config.disable_fractional_pwm == false) ? 1 : 0; + uint32_t edp_id_count = dc->dc_edp_id_count; + int i; + uint8_t panel_mask = 0; + + for (i = 0; i < edp_id_count; i++) + panel_mask |= 0x01 << i; + + memset(&cmd, 0, sizeof(cmd)); + cmd.abm_set_pwm_frac.header.type = DMUB_CMD__ABM; + cmd.abm_set_pwm_frac.header.sub_type = DMUB_CMD__ABM_SET_PWM_FRAC; + cmd.abm_set_pwm_frac.abm_set_pwm_frac_data.fractional_pwm = fractional_pwm; + cmd.abm_set_pwm_frac.abm_set_pwm_frac_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1; + cmd.abm_set_pwm_frac.abm_set_pwm_frac_data.panel_mask = panel_mask; + cmd.abm_set_pwm_frac.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_pwm_frac_data); + + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); +} + +void dmub_abm_init(struct abm *abm, uint32_t backlight) +{ + struct dce_abm *dce_abm = TO_DMUB_ABM(abm); + + REG_WRITE(DC_ABM1_HG_SAMPLE_RATE, 0x3); + REG_WRITE(DC_ABM1_HG_SAMPLE_RATE, 0x1); + REG_WRITE(DC_ABM1_LS_SAMPLE_RATE, 0x3); + REG_WRITE(DC_ABM1_LS_SAMPLE_RATE, 0x1); + REG_WRITE(BL1_PWM_BL_UPDATE_SAMPLE_RATE, 0x1); + + REG_SET_3(DC_ABM1_HG_MISC_CTRL, 0, + ABM1_HG_NUM_OF_BINS_SEL, 0, + ABM1_HG_VMAX_SEL, 1, + ABM1_HG_BIN_BITWIDTH_SIZE_SEL, 0); + + REG_SET_3(DC_ABM1_IPCSC_COEFF_SEL, 0, + ABM1_IPCSC_COEFF_SEL_R, 2, + ABM1_IPCSC_COEFF_SEL_G, 4, + ABM1_IPCSC_COEFF_SEL_B, 2); + + REG_UPDATE(BL1_PWM_CURRENT_ABM_LEVEL, + BL1_PWM_CURRENT_ABM_LEVEL, backlight); + + REG_UPDATE(BL1_PWM_TARGET_ABM_LEVEL, + BL1_PWM_TARGET_ABM_LEVEL, backlight); + + REG_UPDATE(BL1_PWM_USER_LEVEL, + BL1_PWM_USER_LEVEL, backlight); + + REG_UPDATE_2(DC_ABM1_LS_MIN_MAX_PIXEL_VALUE_THRES, + ABM1_LS_MIN_PIXEL_VALUE_THRES, 0, + ABM1_LS_MAX_PIXEL_VALUE_THRES, 1000); + + REG_SET_3(DC_ABM1_HGLS_REG_READ_PROGRESS, 0, + ABM1_HG_REG_READ_MISSED_FRAME_CLEAR, 1, + ABM1_LS_REG_READ_MISSED_FRAME_CLEAR, 1, + ABM1_BL_REG_READ_MISSED_FRAME_CLEAR, 1); + + dmub_abm_enable_fractional_pwm(abm->ctx); +} + +unsigned int dmub_abm_get_current_backlight(struct abm *abm) +{ + struct dce_abm *dce_abm = TO_DMUB_ABM(abm); + unsigned int backlight = REG_READ(BL1_PWM_CURRENT_ABM_LEVEL); + + /* return backlight in hardware format which is unsigned 17 bits, with + * 1 bit integer and 16 bit fractional + */ + return backlight; +} + +unsigned int dmub_abm_get_target_backlight(struct abm *abm) +{ + struct dce_abm *dce_abm = TO_DMUB_ABM(abm); + unsigned int backlight = REG_READ(BL1_PWM_TARGET_ABM_LEVEL); + + /* return backlight in hardware format which is unsigned 17 bits, with + * 1 bit integer and 16 bit fractional + */ + return backlight; +} + +bool dmub_abm_set_level(struct abm *abm, uint32_t level, uint8_t panel_mask) +{ + union dmub_rb_cmd cmd; + struct dc_context *dc = abm->ctx; + + memset(&cmd, 0, sizeof(cmd)); + cmd.abm_set_level.header.type = DMUB_CMD__ABM; + cmd.abm_set_level.header.sub_type = DMUB_CMD__ABM_SET_LEVEL; + cmd.abm_set_level.abm_set_level_data.level = level; + cmd.abm_set_level.abm_set_level_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1; + cmd.abm_set_level.abm_set_level_data.panel_mask = panel_mask; + cmd.abm_set_level.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_level_data); + + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + + return true; +} + +void dmub_abm_init_config(struct abm *abm, + const char *src, + unsigned int bytes, + unsigned int inst) +{ + union dmub_rb_cmd cmd; + struct dc_context *dc = abm->ctx; + uint8_t panel_mask = 0x01 << inst; + + // TODO: Optimize by only reading back final 4 bytes + dmub_flush_buffer_mem(&dc->dmub_srv->dmub->scratch_mem_fb); + + // Copy iramtable into cw7 + memcpy(dc->dmub_srv->dmub->scratch_mem_fb.cpu_addr, (void *)src, bytes); + + memset(&cmd, 0, sizeof(cmd)); + // Fw will copy from cw7 to fw_state + cmd.abm_init_config.header.type = DMUB_CMD__ABM; + cmd.abm_init_config.header.sub_type = DMUB_CMD__ABM_INIT_CONFIG; + cmd.abm_init_config.abm_init_config_data.src.quad_part = dc->dmub_srv->dmub->scratch_mem_fb.gpu_addr; + cmd.abm_init_config.abm_init_config_data.bytes = bytes; + cmd.abm_init_config.abm_init_config_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1; + cmd.abm_init_config.abm_init_config_data.panel_mask = panel_mask; + + cmd.abm_init_config.header.payload_bytes = sizeof(struct dmub_cmd_abm_init_config_data); + + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + +} + +bool dmub_abm_set_pause(struct abm *abm, bool pause, unsigned int panel_inst, unsigned int stream_inst) +{ + union dmub_rb_cmd cmd; + struct dc_context *dc = abm->ctx; + uint8_t panel_mask = 0x01 << panel_inst; + + memset(&cmd, 0, sizeof(cmd)); + cmd.abm_pause.header.type = DMUB_CMD__ABM; + cmd.abm_pause.header.sub_type = DMUB_CMD__ABM_PAUSE; + cmd.abm_pause.abm_pause_data.enable = pause; + cmd.abm_pause.abm_pause_data.panel_mask = panel_mask; + cmd.abm_set_level.header.payload_bytes = sizeof(struct dmub_cmd_abm_pause_data); + + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + + return true; +} + +bool dmub_abm_set_pipe(struct abm *abm, uint32_t otg_inst, uint32_t option, uint32_t panel_inst) +{ + union dmub_rb_cmd cmd; + struct dc_context *dc = abm->ctx; + uint32_t ramping_boundary = 0xFFFF; + + memset(&cmd, 0, sizeof(cmd)); + cmd.abm_set_pipe.header.type = DMUB_CMD__ABM; + cmd.abm_set_pipe.header.sub_type = DMUB_CMD__ABM_SET_PIPE; + cmd.abm_set_pipe.abm_set_pipe_data.otg_inst = otg_inst; + cmd.abm_set_pipe.abm_set_pipe_data.set_pipe_option = option; + cmd.abm_set_pipe.abm_set_pipe_data.panel_inst = panel_inst; + cmd.abm_set_pipe.abm_set_pipe_data.ramping_boundary = ramping_boundary; + cmd.abm_set_pipe.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_pipe_data); + + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + + return true; +} + +bool dmub_abm_set_backlight_level(struct abm *abm, + unsigned int backlight_pwm_u16_16, + unsigned int frame_ramp, + unsigned int panel_inst) +{ + union dmub_rb_cmd cmd; + struct dc_context *dc = abm->ctx; + + memset(&cmd, 0, sizeof(cmd)); + cmd.abm_set_backlight.header.type = DMUB_CMD__ABM; + cmd.abm_set_backlight.header.sub_type = DMUB_CMD__ABM_SET_BACKLIGHT; + cmd.abm_set_backlight.abm_set_backlight_data.frame_ramp = frame_ramp; + cmd.abm_set_backlight.abm_set_backlight_data.backlight_user_level = backlight_pwm_u16_16; + cmd.abm_set_backlight.abm_set_backlight_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1; + cmd.abm_set_backlight.abm_set_backlight_data.panel_mask = (0x01 << panel_inst); + cmd.abm_set_backlight.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_backlight_data); + + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + + return true; +} + diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.h b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.h new file mode 100644 index 0000000000000000000000000000000000000000..00b4e268768ef20f36eeaa5a10399b0a0b69a59f --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.h @@ -0,0 +1,46 @@ +/* + * Copyright 2019 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DMUB_ABM_LCD_H__ +#define __DMUB_ABM_LCD_H__ + +#include "abm.h" + +void dmub_abm_init(struct abm *abm, uint32_t backlight); +bool dmub_abm_set_level(struct abm *abm, uint32_t level, uint8_t panel_mask); +unsigned int dmub_abm_get_current_backlight(struct abm *abm); +unsigned int dmub_abm_get_target_backlight(struct abm *abm); +void dmub_abm_init_config(struct abm *abm, + const char *src, + unsigned int bytes, + unsigned int inst); + +bool dmub_abm_set_pause(struct abm *abm, bool pause, unsigned int panel_inst, unsigned int stream_inst); +bool dmub_abm_set_pipe(struct abm *abm, uint32_t otg_inst, uint32_t option, uint32_t panel_inst); +bool dmub_abm_set_backlight_level(struct abm *abm, + unsigned int backlight_pwm_u16_16, + unsigned int frame_ramp, + unsigned int panel_inst); +#endif diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c index 3f32e9c3fbaf4cdb1020990fa0b315e579043bc9..2aa0e01a6891b07e00143d7e803774a822616abf 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c @@ -47,9 +47,7 @@ void dmub_hw_lock_mgr_cmd(struct dc_dmub_srv *dmub_srv, if (!lock) cmd.lock_hw.lock_hw_data.should_release = 1; - dc_dmub_srv_cmd_queue(dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dmub_srv); - dc_dmub_srv_wait_idle(dmub_srv); + dm_execute_dmub_cmd(dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } void dmub_hw_lock_mgr_inbox0_cmd(struct dc_dmub_srv *dmub_srv, diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_outbox.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_outbox.c index fff1d07d865d7e006d262ad4fafe986616f9acab..d8009b2dc56a068ae42e088b52cacbf4f9b0f053 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_outbox.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_outbox.c @@ -48,7 +48,5 @@ void dmub_enable_outbox_notification(struct dc_dmub_srv *dmub_srv) sizeof(cmd.outbox1_enable.header); cmd.outbox1_enable.enable = true; - dc_dmub_srv_cmd_queue(dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dmub_srv); - dc_dmub_srv_wait_idle(dmub_srv); + dm_execute_dmub_cmd(dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c index 9705d8f88382589fe05131aa62d3e89984f9479b..0f24b6fbd22013a86ad8f516d256aa9c9e8643e4 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c @@ -87,6 +87,8 @@ static enum dc_psr_state convert_psr_state(uint32_t raw_state) state = PSR_STATE4c_FULL_FRAME; else if (raw_state == 0x4E) state = PSR_STATE4_FULL_FRAME_POWERUP; + else if (raw_state == 0x4F) + state = PSR_STATE4_FULL_FRAME_HW_LOCK; else if (raw_state == 0x60) state = PSR_STATE_HWLOCK_MGR; else if (raw_state == 0x61) @@ -168,9 +170,7 @@ static bool dmub_psr_set_version(struct dmub_psr *dmub, struct dc_stream_state * cmd.psr_set_version.psr_set_version_data.panel_inst = panel_inst; cmd.psr_set_version.header.payload_bytes = sizeof(struct dmub_cmd_psr_set_version_data); - dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->dmub_srv); - dc_dmub_srv_wait_idle(dc->dmub_srv); + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } @@ -198,9 +198,7 @@ static void dmub_psr_enable(struct dmub_psr *dmub, bool enable, bool wait, uint8 cmd.psr_enable.header.payload_bytes = 0; // Send header only - dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->dmub_srv); - dc_dmub_srv_wait_idle(dc->dmub_srv); + dm_execute_dmub_cmd(dc->dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); /* Below loops 1000 x 500us = 500 ms. * Exit PSR may need to wait 1-2 frames to power up. Timeout after at @@ -248,9 +246,7 @@ static void dmub_psr_set_level(struct dmub_psr *dmub, uint16_t psr_level, uint8_ cmd.psr_set_level.psr_set_level_data.psr_level = psr_level; cmd.psr_set_level.psr_set_level_data.cmd_version = DMUB_CMD_PSR_CONTROL_VERSION_1; cmd.psr_set_level.psr_set_level_data.panel_inst = panel_inst; - dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->dmub_srv); - dc_dmub_srv_wait_idle(dc->dmub_srv); + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } /* @@ -269,9 +265,7 @@ static void dmub_psr_set_sink_vtotal_in_psr_active(struct dmub_psr *dmub, cmd.psr_set_vtotal.psr_set_vtotal_data.psr_vtotal_idle = psr_vtotal_idle; cmd.psr_set_vtotal.psr_set_vtotal_data.psr_vtotal_su = psr_vtotal_su; - dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->dmub_srv); - dc_dmub_srv_wait_idle(dc->dmub_srv); + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } /* @@ -290,9 +284,7 @@ static void dmub_psr_set_power_opt(struct dmub_psr *dmub, unsigned int power_opt cmd.psr_set_power_opt.psr_set_power_opt_data.power_opt = power_opt; cmd.psr_set_power_opt.psr_set_power_opt_data.panel_inst = panel_inst; - dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->dmub_srv); - dc_dmub_srv_wait_idle(dc->dmub_srv); + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } /* @@ -422,9 +414,7 @@ static bool dmub_psr_copy_settings(struct dmub_psr *dmub, copy_settings_data->relock_delay_frame_cnt = 2; copy_settings_data->dsc_slice_height = psr_context->dsc_slice_height; - dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->dmub_srv); - dc_dmub_srv_wait_idle(dc->dmub_srv); + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } @@ -445,9 +435,7 @@ static void dmub_psr_force_static(struct dmub_psr *dmub, uint8_t panel_inst) cmd.psr_force_static.header.sub_type = DMUB_CMD__PSR_FORCE_STATIC; cmd.psr_enable.header.payload_bytes = 0; - dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->dmub_srv); - dc_dmub_srv_wait_idle(dc->dmub_srv); + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } /* diff --git a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c index 54805802cbd5a6020556accbb813b3e9ed39649c..42e9b6a529f6f881b3107b419eaf9fa00e89c30e 100644 --- a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c @@ -401,6 +401,10 @@ static const struct dc_plane_cap plane_cap = { } }; +static const struct dc_debug_options debug_defaults = { + .enable_legacy_fast_update = true, +}; + #define CTX ctx #define REG(reg) mm ## reg @@ -1071,6 +1075,7 @@ static bool dce100_resource_construct( dc->caps.dual_link_dvi = true; dc->caps.disable_dp_clk_share = true; dc->caps.extended_aux_timeout_support = false; + dc->debug = debug_defaults; for (i = 0; i < pool->base.pipe_count; i++) { pool->base.timing_generators[i] = diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c index 8d2460d06bced340f6d6b2d2edfaa06d6f55de8c..6c9ca43d1040b3610699ec94615b40356cee3ceb 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c @@ -209,9 +209,6 @@ static bool dce110_enable_display_power_gating( struct dc_context *ctx = dc->ctx; unsigned int underlay_idx = dc->res_pool->underlay_pipe_index; - if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) - return true; - if (power_gating == PIPE_GATING_CONTROL_INIT) cntl = ASIC_PIPE_INIT; else if (power_gating == PIPE_GATING_CONTROL_ENABLE) @@ -1219,7 +1216,8 @@ void dce110_blank_stream(struct pipe_ctx *pipe_ctx) struct dce_hwseq *hws = link->dc->hwseq; if (link->local_sink && link->local_sink->sink_signal == SIGNAL_TYPE_EDP) { - hws->funcs.edp_backlight_control(link, false); + if (!stream->skip_edp_power_down) + hws->funcs.edp_backlight_control(link, false); link->dc->hwss.set_abm_immediate_disable(pipe_ctx); } @@ -2291,6 +2289,11 @@ enum dc_status dce110_apply_ctx_to_hw( if (DC_OK != status) return status; + +#ifdef CONFIG_DRM_AMD_DC_FP + if (hws->funcs.resync_fifo_dccg_dio) + hws->funcs.resync_fifo_dccg_dio(hws, dc, context); +#endif } if (dc->fbc_compressor) diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c index a4a45a6ce61e4658fdb3f39b049abcd936221c1d..46eca5a21e1cedcef9d84e75e865585d1638c3f0 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c @@ -424,6 +424,10 @@ static const struct dc_plane_cap plane_cap = { 64 }; +static const struct dc_debug_options debug_defaults = { + .enable_legacy_fast_update = true, +}; + static const struct dc_plane_cap underlay_plane_cap = { .type = DC_PLANE_TYPE_DCE_UNDERLAY, .per_pixel_alpha = 1, @@ -1368,6 +1372,7 @@ static bool dce110_resource_construct( dc->caps.min_horizontal_blanking_period = 80; dc->caps.is_apu = true; dc->caps.extended_aux_timeout_support = false; + dc->debug = debug_defaults; /************************************************* * Create resources * diff --git a/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.c index 19873ee1f78dd72665ad670c13730edc04eaed6e..690caaaff019409732bee7bd7676c8ff7420e62d 100644 --- a/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.c @@ -120,9 +120,6 @@ static bool dce112_enable_display_power_gating( enum bp_pipe_control_action cntl; struct dc_context *ctx = dc->ctx; - if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) - return true; - if (power_gating == PIPE_GATING_CONTROL_INIT) cntl = ASIC_PIPE_INIT; else if (power_gating == PIPE_GATING_CONTROL_ENABLE) diff --git a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c index e179e80667d1c84675f7bfe3b2282c64376f323a..8088558861836a73d07a41f8ffba46f7fd55892a 100644 --- a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c @@ -429,6 +429,10 @@ static const struct dc_plane_cap plane_cap = { 64 }; +static const struct dc_debug_options debug_defaults = { + .enable_legacy_fast_update = true, +}; + #define CTX ctx #define REG(reg) mm ## reg @@ -1239,6 +1243,7 @@ static bool dce112_resource_construct( dc->caps.min_horizontal_blanking_period = 80; dc->caps.dual_link_dvi = true; dc->caps.extended_aux_timeout_support = false; + dc->debug = debug_defaults; /************************************************* * Create resources * diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.c index d4afe6c824d2cb75974056f11ff21f874bdfb05b..45e08c4d58618194fd820690038c165dbe81f9b5 100644 --- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_hw_sequencer.c @@ -159,9 +159,6 @@ static bool dce120_enable_display_power_gating( enum bp_pipe_control_action cntl; struct dc_context *ctx = dc->ctx; - if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) - return true; - if (power_gating == PIPE_GATING_CONTROL_INIT) cntl = ASIC_PIPE_INIT; else if (power_gating == PIPE_GATING_CONTROL_ENABLE) diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c index af631085e88c56937ac4d1a119167bdf46f3d75f..18c5a86d2d614be6242397b73600fb33d46afac8 100644 --- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c @@ -526,6 +526,7 @@ static const struct dc_plane_cap plane_cap = { static const struct dc_debug_options debug_defaults = { .disable_clock_gate = true, + .enable_legacy_fast_update = true, }; static struct clock_source *dce120_clock_source_create( diff --git a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c index 5825e6f412bd7642f8264c4b5c3440be1b8c73e0..3935fd455f0f9117fe4e6e8fa3f0c3126c904ce2 100644 --- a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c @@ -418,6 +418,10 @@ static const struct dc_plane_cap plane_cap = { } }; +static const struct dc_debug_options debug_defaults = { + .enable_legacy_fast_update = true, +}; + static const struct dce_dmcu_registers dmcu_regs = { DMCU_DCE80_REG_LIST() }; @@ -969,6 +973,7 @@ static bool dce80_construct( dc->caps.min_horizontal_blanking_period = 80; dc->caps.dual_link_dvi = true; dc->caps.extended_aux_timeout_support = false; + dc->debug = debug_defaults; /************************************************* * Create resources * @@ -1369,6 +1374,7 @@ static bool dce83_construct( dc->caps.max_cursor_size = 128; dc->caps.min_horizontal_blanking_period = 80; dc->caps.is_apu = true; + dc->debug = debug_defaults; /************************************************* * Create resources * diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h index 0b17c2993ca5b5b3fd2187f150a29a6d86b89540..09784222cc0336ae87c5b3387f3155c2c2abee75 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h @@ -690,6 +690,8 @@ struct dcn_hubp_state { uint32_t primary_surface_addr_hi; uint32_t primary_meta_addr_lo; uint32_t primary_meta_addr_hi; + uint32_t uclk_pstate_force; + uint32_t hubp_cntl; }; struct dcn10_hubp { diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c index 1c3b6f25a7825167741039e4bcbddf3e01f53eb0..20a1582be0b13e77819283d2650be3e0aca8bc74 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c @@ -1012,31 +1012,29 @@ static void dcn10_reset_back_end_for_pipe( return; } - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - link = pipe_ctx->stream->link; - /* DPMS may already disable or */ - /* dpms_off status is incorrect due to fastboot - * feature. When system resume from S4 with second - * screen only, the dpms_off would be true but - * VBIOS lit up eDP, so check link status too. - */ - if (!pipe_ctx->stream->dpms_off || link->link_status.link_active) - dc->link_srv->set_dpms_off(pipe_ctx); - else if (pipe_ctx->stream_res.audio) - dc->hwss.disable_audio_stream(pipe_ctx); - - if (pipe_ctx->stream_res.audio) { - /*disable az_endpoint*/ - pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio); - - /*free audio*/ - if (dc->caps.dynamic_audio == true) { - /*we have to dynamic arbitrate the audio endpoints*/ - /*we free the resource, need reset is_audio_acquired*/ - update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, - pipe_ctx->stream_res.audio, false); - pipe_ctx->stream_res.audio = NULL; - } + link = pipe_ctx->stream->link; + /* DPMS may already disable or */ + /* dpms_off status is incorrect due to fastboot + * feature. When system resume from S4 with second + * screen only, the dpms_off would be true but + * VBIOS lit up eDP, so check link status too. + */ + if (!pipe_ctx->stream->dpms_off || link->link_status.link_active) + dc->link_srv->set_dpms_off(pipe_ctx); + else if (pipe_ctx->stream_res.audio) + dc->hwss.disable_audio_stream(pipe_ctx); + + if (pipe_ctx->stream_res.audio) { + /*disable az_endpoint*/ + pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio); + + /*free audio*/ + if (dc->caps.dynamic_audio == true) { + /*we have to dynamic arbitrate the audio endpoints*/ + /*we free the resource, need reset is_audio_acquired*/ + update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, + pipe_ctx->stream_res.audio, false); + pipe_ctx->stream_res.audio = NULL; } } @@ -1499,54 +1497,32 @@ void dcn10_init_hw(struct dc *dc) if (dc->res_pool->dccg && dc->res_pool->dccg->funcs->dccg_init) dc->res_pool->dccg->funcs->dccg_init(res_pool->dccg); - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - - REG_WRITE(REFCLK_CNTL, 0); - REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1); - REG_WRITE(DIO_MEM_PWR_CTRL, 0); - - if (!dc->debug.disable_clock_gate) { - /* enable all DCN clock gating */ - REG_WRITE(DCCG_GATE_DISABLE_CNTL, 0); - - REG_WRITE(DCCG_GATE_DISABLE_CNTL2, 0); - - REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0); - } - - //Enable ability to power gate / don't force power on permanently - if (hws->funcs.enable_power_gating_plane) - hws->funcs.enable_power_gating_plane(hws, true); - - return; - } - if (!dcb->funcs->is_accelerated_mode(dcb)) hws->funcs.disable_vga(dc->hwseq); - hws->funcs.bios_golden_init(dc); + if (!dc_dmub_srv_optimized_init_done(dc->ctx->dmub_srv)) + hws->funcs.bios_golden_init(dc); + if (dc->ctx->dc_bios->fw_info_valid) { res_pool->ref_clocks.xtalin_clock_inKhz = dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency; - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - if (res_pool->dccg && res_pool->hubbub) { + if (res_pool->dccg && res_pool->hubbub) { - (res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg, - dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency, - &res_pool->ref_clocks.dccg_ref_clock_inKhz); + (res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg, + dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency, + &res_pool->ref_clocks.dccg_ref_clock_inKhz); - (res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub, - res_pool->ref_clocks.dccg_ref_clock_inKhz, - &res_pool->ref_clocks.dchub_ref_clock_inKhz); - } else { - // Not all ASICs have DCCG sw component - res_pool->ref_clocks.dccg_ref_clock_inKhz = - res_pool->ref_clocks.xtalin_clock_inKhz; - res_pool->ref_clocks.dchub_ref_clock_inKhz = - res_pool->ref_clocks.xtalin_clock_inKhz; - } + (res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub, + res_pool->ref_clocks.dccg_ref_clock_inKhz, + &res_pool->ref_clocks.dchub_ref_clock_inKhz); + } else { + // Not all ASICs have DCCG sw component + res_pool->ref_clocks.dccg_ref_clock_inKhz = + res_pool->ref_clocks.xtalin_clock_inKhz; + res_pool->ref_clocks.dchub_ref_clock_inKhz = + res_pool->ref_clocks.xtalin_clock_inKhz; } } else ASSERT_CRITICAL(false); @@ -1923,6 +1899,11 @@ void dcn10_pipe_control_lock( * * TODO: Optimize cursor programming to be once per frame before VUPDATE * to avoid the need for this workaround. + * + * @dc: Current DC state + * @pipe_ctx: Pipe_ctx pointer for delayed cursor update + * + * Return: void */ static void delay_cursor_until_vupdate(struct dc *dc, struct pipe_ctx *pipe_ctx) { @@ -2600,23 +2581,15 @@ static void dcn10_update_dpp(struct dpp *dpp, struct dc_plane_state *plane_state dpp->funcs->dpp_program_bias_and_scale(dpp, &bns_params); } -void dcn10_update_visual_confirm_color(struct dc *dc, struct pipe_ctx *pipe_ctx, struct tg_color *color, int mpcc_id) +void dcn10_update_visual_confirm_color(struct dc *dc, + struct pipe_ctx *pipe_ctx, + int mpcc_id) { struct mpc *mpc = dc->res_pool->mpc; - if (dc->debug.visual_confirm == VISUAL_CONFIRM_HDR) - get_hdr_visual_confirm_color(pipe_ctx, color); - else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SURFACE) - get_surface_visual_confirm_color(pipe_ctx, color); - else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SWIZZLE) - get_surface_tile_visual_confirm_color(pipe_ctx, color); - else - color_space_to_black_color( - dc, pipe_ctx->stream->output_color_space, color); - if (mpc->funcs->set_bg_color) { - memcpy(&pipe_ctx->plane_state->visual_confirm_color, color, sizeof(struct tg_color)); - mpc->funcs->set_bg_color(mpc, color, mpcc_id); + memcpy(&pipe_ctx->plane_state->visual_confirm_color, &(pipe_ctx->visual_confirm_color), sizeof(struct tg_color)); + mpc->funcs->set_bg_color(mpc, &(pipe_ctx->visual_confirm_color), mpcc_id); } } @@ -2669,7 +2642,7 @@ void dcn10_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) /* If there is no full update, don't need to touch MPC tree*/ if (!pipe_ctx->plane_state->update_flags.bits.full_update) { mpc->funcs->update_blending(mpc, &blnd_cfg, mpcc_id); - dc->hwss.update_visual_confirm_color(dc, pipe_ctx, &blnd_cfg.black_color, mpcc_id); + dc->hwss.update_visual_confirm_color(dc, pipe_ctx, mpcc_id); return; } @@ -2691,7 +2664,7 @@ void dcn10_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) NULL, hubp->inst, mpcc_id); - dc->hwss.update_visual_confirm_color(dc, pipe_ctx, &blnd_cfg.black_color, mpcc_id); + dc->hwss.update_visual_confirm_color(dc, pipe_ctx, mpcc_id); ASSERT(new_mpcc != NULL); hubp->opp_id = pipe_ctx->stream_res.opp->inst; @@ -3076,15 +3049,13 @@ void dcn10_prepare_bandwidth( if (dc->debug.sanity_checks) hws->funcs.verify_allow_pstate_change_high(dc); - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - if (context->stream_count == 0) - context->bw_ctx.bw.dcn.clk.phyclk_khz = 0; + if (context->stream_count == 0) + context->bw_ctx.bw.dcn.clk.phyclk_khz = 0; - dc->clk_mgr->funcs->update_clocks( - dc->clk_mgr, - context, - false); - } + dc->clk_mgr->funcs->update_clocks( + dc->clk_mgr, + context, + false); dc->wm_optimized_required = hubbub->funcs->program_watermarks(hubbub, &context->bw_ctx.bw.dcn.watermarks, @@ -3116,15 +3087,13 @@ void dcn10_optimize_bandwidth( if (dc->debug.sanity_checks) hws->funcs.verify_allow_pstate_change_high(dc); - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - if (context->stream_count == 0) - context->bw_ctx.bw.dcn.clk.phyclk_khz = 0; + if (context->stream_count == 0) + context->bw_ctx.bw.dcn.clk.phyclk_khz = 0; - dc->clk_mgr->funcs->update_clocks( - dc->clk_mgr, - context, - true); - } + dc->clk_mgr->funcs->update_clocks( + dc->clk_mgr, + context, + true); hubbub->funcs->program_watermarks(hubbub, &context->bw_ctx.bw.dcn.watermarks, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h index 0ef7bf7ddb75e918472f92577837de7907bdebfb..ef6d56da417cdffb6b15de4529972d8ef3795bc4 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h @@ -202,7 +202,6 @@ void dcn10_get_dcc_en_bits(struct dc *dc, int *dcc_en_bits); void dcn10_update_visual_confirm_color( struct dc *dc, struct pipe_ctx *pipe_ctx, - struct tg_color *color, int mpcc_id); #endif /* __DC_HWSS_DCN10_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c index a0f8e31d2adc91a683340faf9d372011e06efc5c..46a2ebcabd1ac5c4a65baf539cb09f6f2dfa7cb8 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c @@ -45,7 +45,8 @@ #include "dcn10_cm_common.h" #include "clk_mgr.h" -unsigned int snprintf_count(char *pBuf, unsigned int bufSize, char *fmt, ...) +__printf(3, 4) +unsigned int snprintf_count(char *pbuf, unsigned int bufsize, char *fmt, ...) { int ret_vsnprintf; unsigned int chars_printed; @@ -53,15 +54,15 @@ unsigned int snprintf_count(char *pBuf, unsigned int bufSize, char *fmt, ...) va_list args; va_start(args, fmt); - ret_vsnprintf = vsnprintf(pBuf, bufSize, fmt, args); + ret_vsnprintf = vsnprintf(pbuf, bufsize, fmt, args); va_end(args); if (ret_vsnprintf > 0) { - if (ret_vsnprintf < bufSize) + if (ret_vsnprintf < bufsize) chars_printed = ret_vsnprintf; else - chars_printed = bufSize - 1; + chars_printed = bufsize - 1; } else chars_printed = 0; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c index 41cec7acf51fe1301d6ee67120a8be1fbdc5e3cc..0dec57679269b0da486b00b328ad1377494ca1e5 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c @@ -37,14 +37,14 @@ #define CTX \ oppn10->base.ctx - -/************* FORMATTER ************/ - /** - * set_truncation + * opp1_set_truncation(): * 1) set truncation depth: 0 for 18 bpp or 1 for 24 bpp * 2) enable truncation * 3) HW remove 12bit FMT support for DCE11 power saving reason. + * + * @oppn10: output_pixel_processor struct instance for dcn10. + * @params: pointer to bit_depth_reduction_params. */ static void opp1_set_truncation( struct dcn10_opp *oppn10, @@ -149,11 +149,12 @@ void opp1_program_bit_depth_reduction( } /** - * set_pixel_encoding - * - * Set Pixel Encoding + * opp1_set_pixel_encoding(): * 0: RGB 4:4:4 or YCbCr 4:4:4 or YOnly * 1: YCbCr 4:2:2 + * + * @oppn10: output_pixel_processor struct instance for dcn10. + * @params: pointer to clamping_and_pixel_encoding_params. */ static void opp1_set_pixel_encoding( struct dcn10_opp *oppn10, @@ -180,13 +181,16 @@ static void opp1_set_pixel_encoding( } /** - * Set Clamping + * opp1_set_clamping(): * 1) Set clamping format based on bpc - 0 for 6bpc (No clamping) * 1 for 8 bpc * 2 for 10 bpc * 3 for 12 bpc * 7 for programable * 2) Enable clamp if Limited range requested + * + * @oppn10: output_pixel_processor struct instance for dcn10. + * @params: pointer to clamping_and_pixel_encoding_params. */ static void opp1_set_clamping( struct dcn10_opp *oppn10, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c index c9e53dc49c92b3dcb899f4658c7d05558686b3f7..0e8f4f36c87c517dac7544afd815c32650fc12a7 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c @@ -42,11 +42,13 @@ #define STATIC_SCREEN_EVENT_MASK_RANGETIMING_DOUBLE_BUFFER_UPDATE_EN 0x100 /** -* apply_front_porch_workaround TODO FPGA still need? -* -* This is a workaround for a bug that has existed since R5xx and has not been -* fixed keep Front porch at minimum 2 for Interlaced mode or 1 for progressive. -*/ + * apply_front_porch_workaround() - This is a workaround for a bug that has + * existed since R5xx and has not been fixed + * keep Front porch at minimum 2 for Interlaced + * mode or 1 for progressive. + * + * @timing: Timing parameters used to configure DCN blocks. + */ static void apply_front_porch_workaround(struct dc_crtc_timing *timing) { if (timing->flags.INTERLACE == 1) { @@ -133,9 +135,20 @@ void optc1_setup_vertical_interrupt2( } /** - * program_timing_generator used by mode timing set - * Program CRTC Timing Registers - OTG_H_*, OTG_V_*, Pixel repetition. - * Including SYNC. Call BIOS command table to program Timings. + * optc1_program_timing() - used by mode timing set Program + * CRTC Timing Registers - OTG_H_*, + * OTG_V_*, Pixel repetition. + * Including SYNC. Call BIOS command table to program Timings. + * + * @optc: timing_generator instance. + * @dc_crtc_timing: Timing parameters used to configure DCN blocks. + * @vready_offset: Vready's starting position. + * @vstartup_start: Vstartup period. + * @vupdate_offset: Vupdate starting position. + * @vupdate_width: Vupdate duration. + * @signal: DC signal types. + * @use_vbios: to program timings from BIOS command table. + * */ void optc1_program_timing( struct timing_generator *optc, @@ -385,6 +398,9 @@ void optc1_set_blank_data_double_buffer(struct timing_generator *optc, bool enab * Sets double buffer point for V_TOTAL, H_TOTAL, VTOTAL_MIN, * VTOTAL_MAX, VTOTAL_MIN_SEL and VTOTAL_MAX_SEL registers. * + * @optc: timing_generator instance. + * @enable: Enable DRR double buffering control if true, disable otherwise. + * * Options: any time, start of frame, dp start of frame (range timing) */ void optc1_set_timing_double_buffer(struct timing_generator *optc, bool enable) @@ -397,8 +413,9 @@ void optc1_set_timing_double_buffer(struct timing_generator *optc, bool enable) } /** - * unblank_crtc - * Call ASIC Control Object to UnBlank CRTC. + * optc1_unblank_crtc() - Call ASIC Control Object to UnBlank CRTC. + * + * @optc: timing_generator instance. */ static void optc1_unblank_crtc(struct timing_generator *optc) { @@ -419,8 +436,9 @@ static void optc1_unblank_crtc(struct timing_generator *optc) } /** - * blank_crtc - * Call ASIC Control Object to Blank CRTC. + * optc1_blank_crtc() - Call ASIC Control Object to Blank CRTC. + * + * @optc: timing_generator instance. */ static void optc1_blank_crtc(struct timing_generator *optc) @@ -493,8 +511,9 @@ void optc1_enable_optc_clock(struct timing_generator *optc, bool enable) } /** - * Enable CRTC - * Enable CRTC - call ASIC Control Object to enable Timing generator. + * optc1_enable_crtc() - Enable CRTC - call ASIC Control Object to enable Timing generator. + * + * @optc: timing_generator instance. */ static bool optc1_enable_crtc(struct timing_generator *optc) { @@ -653,11 +672,9 @@ void optc1_lock(struct timing_generator *optc) REG_SET(OTG_MASTER_UPDATE_LOCK, 0, OTG_MASTER_UPDATE_LOCK, 1); - /* Should be fast, status does not update on maximus */ - if (optc->ctx->dce_environment != DCE_ENV_FPGA_MAXIMUS) - REG_WAIT(OTG_MASTER_UPDATE_LOCK, - UPDATE_LOCK_STATUS, 1, - 1, 10); + REG_WAIT(OTG_MASTER_UPDATE_LOCK, + UPDATE_LOCK_STATUS, 1, + 1, 10); TRACE_OPTC_LOCK_UNLOCK_STATE(optc1, optc->inst, true); } @@ -892,15 +909,11 @@ static void optc1_program_manual_trigger(struct timing_generator *optc) MANUAL_FLOW_CONTROL, 0); } - /** - ***************************************************************************** - * Function: set_drr + * optc1_set_drr() - Program dynamic refresh rate registers m_OTGx_OTG_V_TOTAL_*. * - * @brief - * Program dynamic refresh rate registers m_OTGx_OTG_V_TOTAL_*. - * - ***************************************************************************** + * @optc: timing_generator instance. + * @params: parameters used for Dynamic Refresh Rate. */ void optc1_set_drr( struct timing_generator *optc, @@ -932,19 +945,10 @@ void optc1_set_drr( OTG_FORCE_LOCK_ON_EVENT, 0, OTG_SET_V_TOTAL_MIN_MASK_EN, 0, OTG_SET_V_TOTAL_MIN_MASK, 0); - - // Setup manual flow control for EOF via TRIG_A - optc->funcs->setup_manual_trigger(optc); - - } else { - REG_UPDATE_4(OTG_V_TOTAL_CONTROL, - OTG_SET_V_TOTAL_MIN_MASK, 0, - OTG_V_TOTAL_MIN_SEL, 0, - OTG_V_TOTAL_MAX_SEL, 0, - OTG_FORCE_LOCK_ON_EVENT, 0); - - optc->funcs->set_vtotal_min_max(optc, 0, 0); } + + // Setup manual flow control for EOF via TRIG_A + optc->funcs->setup_manual_trigger(optc); } void optc1_set_vtotal_min_max(struct timing_generator *optc, int vtotal_min, int vtotal_max) diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c index 21ec1ba5ed75698452fdc2bcbff3f37d01aa1665..4b02f8443534e59e7462e99290d7498a2b54ce5a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c @@ -553,6 +553,7 @@ static const struct dc_debug_options debug_defaults_drv = { .recovery_enabled = false, /*enable this by default after testing.*/ .max_downscale_src_width = 3840, .underflow_assert_delay_us = 0xFFFFFFFF, + .enable_legacy_fast_update = true, }; static const struct dc_debug_options debug_defaults_diags = { @@ -886,13 +887,6 @@ static const struct resource_create_funcs res_create_funcs = { .create_hwseq = dcn10_hwseq_create, }; -static const struct resource_create_funcs res_create_maximus_funcs = { - .read_dce_straps = NULL, - .create_audio = NULL, - .create_stream_encoder = NULL, - .create_hwseq = dcn10_hwseq_create, -}; - static void dcn10_clock_source_destroy(struct clock_source **clk_src) { kfree(TO_DCE110_CLK_SRC(*clk_src)); @@ -1651,9 +1645,8 @@ static bool dcn10_resource_construct( } if (!resource_construct(num_virtual_links, dc, &pool->base, - (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ? - &res_create_funcs : &res_create_maximus_funcs))) - goto fail; + &res_create_funcs)) + goto fail; dcn10_hw_sequencer_construct(dc); dc->caps.max_planes = pool->base.pipe_count; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h index 7bdc146f7cb59810209f058e11c17be633b09762..c8602bcfa393aaf7a959686ac1dd009bf988707d 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h @@ -208,7 +208,9 @@ #define DCCG314_REG_FIELD_LIST(type) \ type DSCCLK3_DTO_PHASE;\ type DSCCLK3_DTO_MODULO;\ - type DSCCLK3_DTO_ENABLE; + type DSCCLK3_DTO_ENABLE;\ + type DENTIST_DISPCLK_RDIVIDER;\ + type DENTIST_DISPCLK_WDIVIDER; #define DCCG32_REG_FIELD_LIST(type) \ type DPSTREAMCLK0_EN;\ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c index 5bd698cd6d20b7c9f34021ac7987a05099522ee2..5eebe7f03ddc72a8b060ec4768b5e7753b827942 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c @@ -30,22 +30,13 @@ #include "dsc/dscc_types.h" #include "dsc/rc_calc.h" -static void dsc_log_pps(struct display_stream_compressor *dsc, struct drm_dsc_config *pps); -static bool dsc_prepare_config(const struct dsc_config *dsc_cfg, struct dsc_reg_values *dsc_reg_vals, - struct dsc_optc_config *dsc_optc_cfg); -static void dsc_init_reg_values(struct dsc_reg_values *reg_vals); -static void dsc_update_from_dsc_parameters(struct dsc_reg_values *reg_vals, const struct dsc_parameters *dsc_params); static void dsc_write_to_registers(struct display_stream_compressor *dsc, const struct dsc_reg_values *reg_vals); -static enum dsc_pixel_format dsc_dc_pixel_encoding_to_dsc_pixel_format(enum dc_pixel_encoding dc_pix_enc, bool is_ycbcr422_simple); -static enum dsc_bits_per_comp dsc_dc_color_depth_to_dsc_bits_per_comp(enum dc_color_depth); /* Object I/F functions */ -static void dsc2_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz); static void dsc2_read_state(struct display_stream_compressor *dsc, struct dcn_dsc_state *s); static bool dsc2_validate_stream(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg); static void dsc2_set_config(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, struct dsc_optc_config *dsc_optc_cfg); -static bool dsc2_get_packed_pps(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, uint8_t *dsc_packed_pps); static void dsc2_enable(struct display_stream_compressor *dsc, int opp_pipe); static void dsc2_disable(struct display_stream_compressor *dsc); static void dsc2_disconnect(struct display_stream_compressor *dsc); @@ -108,7 +99,7 @@ void dsc2_construct(struct dcn20_dsc *dsc, /* This returns the capabilities for a single DSC encoder engine. Number of slices and total throughput * can be doubled, tripled etc. by using additional DSC engines. */ -static void dsc2_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz) +void dsc2_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz) { dsc_enc_caps->dsc_version = 0x21; /* v1.2 - DP spec defined it in reverse order and we kept it */ @@ -184,7 +175,7 @@ static bool dsc2_validate_stream(struct display_stream_compressor *dsc, const st } -static void dsc_config_log(struct display_stream_compressor *dsc, const struct dsc_config *config) +void dsc_config_log(struct display_stream_compressor *dsc, const struct dsc_config *config) { DC_LOG_DSC("\tnum_slices_h %d", config->dc_dsc_cfg.num_slices_h); DC_LOG_DSC("\tnum_slices_v %d", config->dc_dsc_cfg.num_slices_v); @@ -211,7 +202,7 @@ static void dsc2_set_config(struct display_stream_compressor *dsc, const struct } -static bool dsc2_get_packed_pps(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, uint8_t *dsc_packed_pps) +bool dsc2_get_packed_pps(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, uint8_t *dsc_packed_pps) { bool is_config_ok; struct dsc_reg_values dsc_reg_vals; @@ -291,7 +282,7 @@ static void dsc2_disconnect(struct display_stream_compressor *dsc) } /* This module's internal functions */ -static void dsc_log_pps(struct display_stream_compressor *dsc, struct drm_dsc_config *pps) +void dsc_log_pps(struct display_stream_compressor *dsc, struct drm_dsc_config *pps) { int i; int bits_per_pixel = pps->bits_per_pixel; @@ -345,7 +336,7 @@ static void dsc_log_pps(struct display_stream_compressor *dsc, struct drm_dsc_co } } -static void dsc_override_rc_params(struct rc_params *rc, const struct dc_dsc_rc_params_override *override) +void dsc_override_rc_params(struct rc_params *rc, const struct dc_dsc_rc_params_override *override) { uint8_t i; @@ -372,7 +363,7 @@ static void dsc_override_rc_params(struct rc_params *rc, const struct dc_dsc_rc_ rc->flatness_det_thresh = override->flatness_det_thresh; } -static bool dsc_prepare_config(const struct dsc_config *dsc_cfg, struct dsc_reg_values *dsc_reg_vals, +bool dsc_prepare_config(const struct dsc_config *dsc_cfg, struct dsc_reg_values *dsc_reg_vals, struct dsc_optc_config *dsc_optc_cfg) { struct dsc_parameters dsc_params; @@ -463,7 +454,7 @@ static bool dsc_prepare_config(const struct dsc_config *dsc_cfg, struct dsc_reg_ } -static enum dsc_pixel_format dsc_dc_pixel_encoding_to_dsc_pixel_format(enum dc_pixel_encoding dc_pix_enc, bool is_ycbcr422_simple) +enum dsc_pixel_format dsc_dc_pixel_encoding_to_dsc_pixel_format(enum dc_pixel_encoding dc_pix_enc, bool is_ycbcr422_simple) { enum dsc_pixel_format dsc_pix_fmt = DSC_PIXFMT_UNKNOWN; @@ -495,7 +486,7 @@ static enum dsc_pixel_format dsc_dc_pixel_encoding_to_dsc_pixel_format(enum dc_p } -static enum dsc_bits_per_comp dsc_dc_color_depth_to_dsc_bits_per_comp(enum dc_color_depth dc_color_depth) +enum dsc_bits_per_comp dsc_dc_color_depth_to_dsc_bits_per_comp(enum dc_color_depth dc_color_depth) { enum dsc_bits_per_comp bpc = DSC_BPC_UNKNOWN; @@ -518,7 +509,7 @@ static enum dsc_bits_per_comp dsc_dc_color_depth_to_dsc_bits_per_comp(enum dc_co } -static void dsc_init_reg_values(struct dsc_reg_values *reg_vals) +void dsc_init_reg_values(struct dsc_reg_values *reg_vals) { int i; @@ -574,7 +565,7 @@ static void dsc_init_reg_values(struct dsc_reg_values *reg_vals) * This is required because dscc_compute_dsc_parameters returns a modified PPS, which in turn * affects non-PPS register values. */ -static void dsc_update_from_dsc_parameters(struct dsc_reg_values *reg_vals, const struct dsc_parameters *dsc_params) +void dsc_update_from_dsc_parameters(struct dsc_reg_values *reg_vals, const struct dsc_parameters *dsc_params) { int i; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h index 7ce64a3c1b02b4364e3bc9387550e76e06a81e60..ba869387c3c5837b16f23380fc57d46c9c7468d8 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.h @@ -549,6 +549,27 @@ struct dcn20_dsc { int max_image_width; }; +void dsc_config_log(struct display_stream_compressor *dsc, + const struct dsc_config *config); + +void dsc_log_pps(struct display_stream_compressor *dsc, + struct drm_dsc_config *pps); + +void dsc_override_rc_params(struct rc_params *rc, + const struct dc_dsc_rc_params_override *override); + +bool dsc_prepare_config(const struct dsc_config *dsc_cfg, + struct dsc_reg_values *dsc_reg_vals, + struct dsc_optc_config *dsc_optc_cfg); + +enum dsc_pixel_format dsc_dc_pixel_encoding_to_dsc_pixel_format(enum dc_pixel_encoding dc_pix_enc, + bool is_ycbcr422_simple); + +enum dsc_bits_per_comp dsc_dc_color_depth_to_dsc_bits_per_comp(enum dc_color_depth dc_color_depth); + +void dsc_init_reg_values(struct dsc_reg_values *reg_vals); + +void dsc_update_from_dsc_parameters(struct dsc_reg_values *reg_vals, const struct dsc_parameters *dsc_params); void dsc2_construct(struct dcn20_dsc *dsc, struct dc_context *ctx, @@ -557,5 +578,12 @@ void dsc2_construct(struct dcn20_dsc *dsc, const struct dcn20_dsc_shift *dsc_shift, const struct dcn20_dsc_mask *dsc_mask); +void dsc2_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, + int pixel_clock_100Hz); + +bool dsc2_get_packed_pps(struct display_stream_compressor *dsc, + const struct dsc_config *dsc_cfg, + uint8_t *dsc_packed_pps); + #endif diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c index 24bd932199366e328323af92b290eda437121c21..6eebcb22e31739916a77ef9024da1f6ded89dcd7 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubbub.c @@ -623,6 +623,17 @@ void hubbub2_read_state(struct hubbub *hubbub, struct dcn_hubbub_state *hubbub_s REG_GET(DCN_VM_FAULT_STATUS, DCN_VM_ERROR_VMID, &hubbub_state->vm_error_vmid); REG_GET(DCN_VM_FAULT_STATUS, DCN_VM_ERROR_PIPE, &hubbub_state->vm_error_pipe); } + + if (REG(DCHUBBUB_TEST_DEBUG_INDEX) && REG(DCHUBBUB_TEST_DEBUG_DATA)) { + REG_WRITE(DCHUBBUB_TEST_DEBUG_INDEX, 0x6); + hubbub_state->test_debug_data = REG_READ(DCHUBBUB_TEST_DEBUG_DATA); + } + + if (REG(DCHUBBUB_ARB_WATERMARK_CHANGE_CNTL)) + hubbub_state->watermark_change_cntl = REG_READ(DCHUBBUB_ARB_WATERMARK_CHANGE_CNTL); + + if (REG(DCHUBBUB_ARB_DRAM_STATE_CNTL)) + hubbub_state->dram_state_cntl = REG_READ(DCHUBBUB_ARB_DRAM_STATE_CNTL); } static const struct hubbub_funcs hubbub2_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c index 422fbf79da64fb95eda7627455870a0dbdb351cb..eaf9e9ccad2a7089c2637fcb82b501fde81569bf 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c @@ -313,6 +313,10 @@ void dcn20_init_blank( } opp = dc->res_pool->opps[opp_id_src0]; + /* don't override the blank pattern if already enabled with the correct one. */ + if (opp->funcs->dpg_is_blanked && opp->funcs->dpg_is_blanked(opp)) + return; + if (num_opps == 2) { otg_active_width = otg_active_width / 2; @@ -1357,6 +1361,7 @@ static void dcn20_detect_pipe_changes(struct pipe_ctx *old_pipe, struct pipe_ctx new_pipe->update_flags.bits.dppclk = 1; new_pipe->update_flags.bits.hubp_interdependent = 1; new_pipe->update_flags.bits.hubp_rq_dlg_ttu = 1; + new_pipe->update_flags.bits.unbounded_req = 1; new_pipe->update_flags.bits.gamut_remap = 1; new_pipe->update_flags.bits.scaler = 1; new_pipe->update_flags.bits.viewport = 1; @@ -1500,6 +1505,9 @@ static void dcn20_detect_pipe_changes(struct pipe_ctx *old_pipe, struct pipe_ctx memcmp(&old_pipe->rq_regs, &new_pipe->rq_regs, sizeof(old_pipe->rq_regs))) new_pipe->update_flags.bits.hubp_rq_dlg_ttu = 1; } + + if (old_pipe->unbounded_req != new_pipe->unbounded_req) + new_pipe->update_flags.bits.unbounded_req = 1; } static void dcn20_update_dchubp_dpp( @@ -1533,10 +1541,11 @@ static void dcn20_update_dchubp_dpp( &pipe_ctx->ttu_regs, &pipe_ctx->rq_regs, &pipe_ctx->pipe_dlg_param); - - if (hubp->funcs->set_unbounded_requesting) - hubp->funcs->set_unbounded_requesting(hubp, pipe_ctx->unbounded_req); } + + if (pipe_ctx->update_flags.bits.unbounded_req && hubp->funcs->set_unbounded_requesting) + hubp->funcs->set_unbounded_requesting(hubp, pipe_ctx->unbounded_req); + if (pipe_ctx->update_flags.bits.hubp_interdependent) hubp->funcs->hubp_setup_interdependent( hubp, @@ -2113,20 +2122,11 @@ void dcn20_optimize_bandwidth( if (hubbub->funcs->program_compbuf_size) hubbub->funcs->program_compbuf_size(hubbub, context->bw_ctx.bw.dcn.compbuf_size_kb, true); - if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) { - dc_dmub_srv_p_state_delegate(dc, - true, context); - context->bw_ctx.bw.dcn.clk.p_state_change_support = true; - dc->clk_mgr->clks.fw_based_mclk_switching = true; - } else { - dc->clk_mgr->clks.fw_based_mclk_switching = false; - } - dc->clk_mgr->funcs->update_clocks( dc->clk_mgr, context, true); - if (dc_extended_blank_supported(dc) && context->bw_ctx.bw.dcn.clk.zstate_support == DCN_ZSTATE_SUPPORT_ALLOW) { + if (context->bw_ctx.bw.dcn.clk.zstate_support == DCN_ZSTATE_SUPPORT_ALLOW) { for (i = 0; i < dc->res_pool->pipe_count; ++i) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; @@ -2134,7 +2134,7 @@ void dcn20_optimize_bandwidth( && pipe_ctx->stream->adjust.v_total_min == pipe_ctx->stream->adjust.v_total_max && pipe_ctx->stream->adjust.v_total_max > pipe_ctx->stream->timing.v_total) pipe_ctx->plane_res.hubp->funcs->program_extended_blank(pipe_ctx->plane_res.hubp, - pipe_ctx->dlg_regs.optimized_min_dst_y_next_start); + pipe_ctx->dlg_regs.min_dst_y_next_start); } } } @@ -2471,36 +2471,31 @@ static void dcn20_reset_back_end_for_pipe( return; } - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - /* DPMS may already disable or */ - /* dpms_off status is incorrect due to fastboot - * feature. When system resume from S4 with second - * screen only, the dpms_off would be true but - * VBIOS lit up eDP, so check link status too. - */ - if (!pipe_ctx->stream->dpms_off || link->link_status.link_active) - dc->link_srv->set_dpms_off(pipe_ctx); - else if (pipe_ctx->stream_res.audio) - dc->hwss.disable_audio_stream(pipe_ctx); - - /* free acquired resources */ - if (pipe_ctx->stream_res.audio) { - /*disable az_endpoint*/ - pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio); - - /*free audio*/ - if (dc->caps.dynamic_audio == true) { - /*we have to dynamic arbitrate the audio endpoints*/ - /*we free the resource, need reset is_audio_acquired*/ - update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, - pipe_ctx->stream_res.audio, false); - pipe_ctx->stream_res.audio = NULL; - } + /* DPMS may already disable or */ + /* dpms_off status is incorrect due to fastboot + * feature. When system resume from S4 with second + * screen only, the dpms_off would be true but + * VBIOS lit up eDP, so check link status too. + */ + if (!pipe_ctx->stream->dpms_off || link->link_status.link_active) + dc->link_srv->set_dpms_off(pipe_ctx); + else if (pipe_ctx->stream_res.audio) + dc->hwss.disable_audio_stream(pipe_ctx); + + /* free acquired resources */ + if (pipe_ctx->stream_res.audio) { + /*disable az_endpoint*/ + pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio); + + /*free audio*/ + if (dc->caps.dynamic_audio == true) { + /*we have to dynamic arbitrate the audio endpoints*/ + /*we free the resource, need reset is_audio_acquired*/ + update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, + pipe_ctx->stream_res.audio, false); + pipe_ctx->stream_res.audio = NULL; } } - else if (pipe_ctx->stream_res.dsc) { - dc->link_srv->set_dsc_enable(pipe_ctx, false); - } /* by upper caller loop, parent pipe: pipe0, will be reset last. * back end share by all pipes and will be disable only when disable @@ -2576,28 +2571,6 @@ void dcn20_reset_hw_ctx_wrap( } } -void dcn20_update_visual_confirm_color(struct dc *dc, struct pipe_ctx *pipe_ctx, struct tg_color *color, int mpcc_id) -{ - struct mpc *mpc = dc->res_pool->mpc; - - // input to MPCC is always RGB, by default leave black_color at 0 - if (dc->debug.visual_confirm == VISUAL_CONFIRM_HDR) - get_hdr_visual_confirm_color(pipe_ctx, color); - else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SURFACE) - get_surface_visual_confirm_color(pipe_ctx, color); - else if (dc->debug.visual_confirm == VISUAL_CONFIRM_MPCTREE) - get_mpctree_visual_confirm_color(pipe_ctx, color); - else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SWIZZLE) - get_surface_tile_visual_confirm_color(pipe_ctx, color); - else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SUBVP) - get_subvp_visual_confirm_color(dc, pipe_ctx, color); - - if (mpc->funcs->set_bg_color) { - memcpy(&pipe_ctx->plane_state->visual_confirm_color, color, sizeof(struct tg_color)); - mpc->funcs->set_bg_color(mpc, color, mpcc_id); - } -} - void dcn20_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) { struct hubp *hubp = pipe_ctx->plane_res.hubp; @@ -2653,7 +2626,7 @@ void dcn20_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) if (!pipe_ctx->plane_state->update_flags.bits.full_update && !pipe_ctx->update_flags.bits.mpcc) { mpc->funcs->update_blending(mpc, &blnd_cfg, mpcc_id); - dc->hwss.update_visual_confirm_color(dc, pipe_ctx, &blnd_cfg.black_color, mpcc_id); + dc->hwss.update_visual_confirm_color(dc, pipe_ctx, mpcc_id); return; } @@ -2675,7 +2648,7 @@ void dcn20_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) NULL, hubp->inst, mpcc_id); - dc->hwss.update_visual_confirm_color(dc, pipe_ctx, &blnd_cfg.black_color, mpcc_id); + dc->hwss.update_visual_confirm_color(dc, pipe_ctx, mpcc_id); ASSERT(new_mpcc != NULL); hubp->opp_id = pipe_ctx->stream_res.opp->inst; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h index 33a36c02b2f8f6b09032e738de53d2cbc5ae180a..01901b08644c584b539d6e9718b1bf8ec72f47f6 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h @@ -150,10 +150,5 @@ void dcn20_set_disp_pattern_generator(const struct dc *dc, const struct tg_color *solid_color, int width, int height, int offset); -void dcn20_update_visual_confirm_color(struct dc *dc, - struct pipe_ctx *pipe_ctx, - struct tg_color *color, - int mpcc_id); - #endif /* __DC_HWSS_DCN20_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c index 7c5817c426faa78c55da31ac1eacf83c315fb737..e4b44e691ce664dd20b5a2958c8acb1ac56c03be 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c @@ -102,7 +102,7 @@ static const struct hw_sequencer_funcs dcn20_funcs = { .disable_link_output = dce110_disable_link_output, .set_disp_pattern_generator = dcn20_set_disp_pattern_generator, .get_dcc_en_bits = dcn10_get_dcc_en_bits, - .update_visual_confirm_color = dcn20_update_visual_confirm_color + .update_visual_confirm_color = dcn10_update_visual_confirm_color, }; static const struct hwseq_private_funcs dcn20_private_funcs = { @@ -145,8 +145,4 @@ void dcn20_hw_sequencer_construct(struct dc *dc) dc->hwss = dcn20_funcs; dc->hwseq->funcs = dcn20_private_funcs; - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - dc->hwss.init_hw = dcn20_fpga_init_hw; - dc->hwseq->funcs.init_pipes = NULL; - } } diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c index a08c335b738386e53aee90ea88b71abcb7d2304a..58bdbd859bf9b0d4c95e992a355a434944deac78 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c @@ -38,8 +38,12 @@ optc1->tg_shift->field_name, optc1->tg_mask->field_name /** - * Enable CRTC - * Enable CRTC - call ASIC Control Object to enable Timing generator. + * optc2_enable_crtc() - Enable CRTC - call ASIC Control Object to enable Timing generator. + * + * @optc: timing_generator instance. + * + * Return: If CRTC is enabled, return true. + * */ bool optc2_enable_crtc(struct timing_generator *optc) { @@ -73,15 +77,18 @@ bool optc2_enable_crtc(struct timing_generator *optc) } /** - *For the below, I'm not sure how your GSL parameters are stored in your env, - * so I will assume a gsl_params struct for now + * optc2_set_gsl() - Assign OTG to GSL groups, + * set one of the OTGs to be master & rest are slaves + * + * @optc: timing_generator instance. + * @params: pointer to gsl_params */ void optc2_set_gsl(struct timing_generator *optc, const struct gsl_params *params) { struct optc *optc1 = DCN10TG_FROM_TG(optc); -/** +/* * There are (MAX_OPTC+1)/2 gsl groups available for use. * In each group (assign an OTG to a group by setting OTG_GSLX_EN = 1, * set one of the OTGs to be the master (OTG_GSL_MASTER_EN = 1) and the rest are slaves. @@ -391,10 +398,9 @@ void optc2_triplebuffer_lock(struct timing_generator *optc) REG_SET(OTG_MASTER_UPDATE_LOCK, 0, OTG_MASTER_UPDATE_LOCK, 1); - if (optc->ctx->dce_environment != DCE_ENV_FPGA_MAXIMUS) - REG_WAIT(OTG_MASTER_UPDATE_LOCK, - UPDATE_LOCK_STATUS, 1, - 1, 10); + REG_WAIT(OTG_MASTER_UPDATE_LOCK, + UPDATE_LOCK_STATUS, 1, + 1, 10); } void optc2_triplebuffer_unlock(struct timing_generator *optc) @@ -456,6 +462,16 @@ void optc2_setup_manual_trigger(struct timing_generator *optc) { struct optc *optc1 = DCN10TG_FROM_TG(optc); + /* Set the min/max selectors unconditionally so that + * DMCUB fw may change OTG timings when necessary + * TODO: Remove the w/a after fixing the issue in DMCUB firmware + */ + REG_UPDATE_4(OTG_V_TOTAL_CONTROL, + OTG_V_TOTAL_MIN_SEL, 1, + OTG_V_TOTAL_MAX_SEL, 1, + OTG_FORCE_LOCK_ON_EVENT, 0, + OTG_SET_V_TOTAL_MIN_MASK, (1 << 1)); /* TRIGA */ + REG_SET_8(OTG_TRIGA_CNTL, 0, OTG_TRIGA_SOURCE_SELECT, 21, OTG_TRIGA_SOURCE_PIPE_SELECT, optc->inst, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c index 1d8c5805ef20c1cb580fe5b7a6d82eae12a11eb2..4cc8de2627ce92bee5a3b7f98e81b9f2b6670d96 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c @@ -722,22 +722,7 @@ static const struct dc_debug_options debug_defaults_drv = { .scl_reset_length10 = true, .sanity_checks = false, .underflow_assert_delay_us = 0xFFFFFFFF, -}; - -static const struct dc_debug_options debug_defaults_diags = { - .disable_dmcu = false, - .force_abm_enable = false, - .timing_trace = true, - .clock_trace = true, - .disable_dpp_power_gate = true, - .disable_hubp_power_gate = true, - .disable_clock_gate = true, - .disable_pplib_clock_request = true, - .disable_pplib_wm_range = true, - .disable_stutter = true, - .scl_reset_length10 = true, - .underflow_assert_delay_us = 0xFFFFFFFF, - .enable_tri_buf = true, + .enable_legacy_fast_update = true, }; void dcn20_dpp_destroy(struct dpp **dpp) @@ -1066,13 +1051,6 @@ static const struct resource_create_funcs res_create_funcs = { .create_hwseq = dcn20_hwseq_create, }; -static const struct resource_create_funcs res_create_maximus_funcs = { - .read_dce_straps = NULL, - .create_audio = NULL, - .create_stream_encoder = NULL, - .create_hwseq = dcn20_hwseq_create, -}; - static void dcn20_pp_smu_destroy(struct pp_smu_funcs **pp_smu); void dcn20_clock_source_destroy(struct clock_source **clk_src) @@ -2488,15 +2466,9 @@ static bool dcn20_resource_construct( dc->caps.dp_hdmi21_pcon_support = true; - if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) { + if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) dc->debug = debug_defaults_drv; - } else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) { - pool->base.pipe_count = 4; - pool->base.mpcc_count = pool->base.pipe_count; - dc->debug = debug_defaults_diags; - } else { - dc->debug = debug_defaults_diags; - } + //dcn2.0x dc->work_arounds.dedcn20_305_wa = true; @@ -2734,9 +2706,8 @@ static bool dcn20_resource_construct( } if (!resource_construct(num_virtual_links, dc, &pool->base, - (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ? - &res_create_funcs : &res_create_maximus_funcs))) - goto create_fail; + &res_create_funcs)) + goto create_fail; dcn20_hw_sequencer_construct(dc); diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_hwseq.c index 1aeb04fbd89d322fa0ba48fe67788f2709c0d8b3..9e027db6d752ca8d9693f0638c958598bebfcbf5 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_hwseq.c @@ -231,52 +231,39 @@ void dcn201_init_hw(struct dc *dc) if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks) dc->clk_mgr->funcs->init_clocks(dc->clk_mgr); - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - REG_WRITE(RBBMIF_TIMEOUT_DIS, 0xFFFFFFFF); - REG_WRITE(RBBMIF_TIMEOUT_DIS_2, 0xFFFFFFFF); - - hws->funcs.dccg_init(hws); - - REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_REFDIV, 2); - REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1); - REG_WRITE(REFCLK_CNTL, 0); - } else { - hws->funcs.bios_golden_init(dc); - - if (dc->ctx->dc_bios->fw_info_valid) { - res_pool->ref_clocks.xtalin_clock_inKhz = - dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency; - - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - if (res_pool->dccg && res_pool->hubbub) { - (res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg, - dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency, - &res_pool->ref_clocks.dccg_ref_clock_inKhz); - - (res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub, - res_pool->ref_clocks.dccg_ref_clock_inKhz, - &res_pool->ref_clocks.dchub_ref_clock_inKhz); - } else { - res_pool->ref_clocks.dccg_ref_clock_inKhz = - res_pool->ref_clocks.xtalin_clock_inKhz; - res_pool->ref_clocks.dchub_ref_clock_inKhz = - res_pool->ref_clocks.xtalin_clock_inKhz; - } - } - } else - ASSERT_CRITICAL(false); - for (i = 0; i < dc->link_count; i++) { - /* Power up AND update implementation according to the - * required signal (which may be different from the - * default signal on connector). - */ - struct dc_link *link = dc->links[i]; - - link->link_enc->funcs->hw_init(link->link_enc); + hws->funcs.bios_golden_init(dc); + + if (dc->ctx->dc_bios->fw_info_valid) { + res_pool->ref_clocks.xtalin_clock_inKhz = + dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency; + + if (res_pool->dccg && res_pool->hubbub) { + (res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg, + dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency, + &res_pool->ref_clocks.dccg_ref_clock_inKhz); + + (res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub, + res_pool->ref_clocks.dccg_ref_clock_inKhz, + &res_pool->ref_clocks.dchub_ref_clock_inKhz); + } else { + res_pool->ref_clocks.dccg_ref_clock_inKhz = + res_pool->ref_clocks.xtalin_clock_inKhz; + res_pool->ref_clocks.dchub_ref_clock_inKhz = + res_pool->ref_clocks.xtalin_clock_inKhz; } - if (hws->fb_offset.quad_part == 0) - read_mmhub_vm_setup(hws); + } else + ASSERT_CRITICAL(false); + for (i = 0; i < dc->link_count; i++) { + /* Power up AND update implementation according to the + * required signal (which may be different from the + * default signal on connector). + */ + struct dc_link *link = dc->links[i]; + + link->link_enc->funcs->hw_init(link->link_enc); } + if (hws->fb_offset.quad_part == 0) + read_mmhub_vm_setup(hws); /* Blank pixel data with OPP DPG */ for (i = 0; i < res_pool->timing_generator_count; i++) { @@ -362,10 +349,6 @@ void dcn201_init_hw(struct dc *dc) tg->funcs->tg_init(tg); } - /* end of FPGA. Below if real ASIC */ - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) - return; - for (i = 0; i < res_pool->audio_count; i++) { struct audio *audio = res_pool->audios[i]; @@ -496,7 +479,7 @@ void dcn201_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) /* If there is no full update, don't need to touch MPC tree*/ if (!pipe_ctx->plane_state->update_flags.bits.full_update) { - dc->hwss.update_visual_confirm_color(dc, pipe_ctx, &blnd_cfg.black_color, mpcc_id); + dc->hwss.update_visual_confirm_color(dc, pipe_ctx, mpcc_id); mpc->funcs->update_blending(mpc, &blnd_cfg, mpcc_id); return; } @@ -521,7 +504,7 @@ void dcn201_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx) dc->res_pool->mpc, mpcc_id); /* Call MPC to insert new plane */ - dc->hwss.update_visual_confirm_color(dc, pipe_ctx, &blnd_cfg.black_color, mpcc_id); + dc->hwss.update_visual_confirm_color(dc, pipe_ctx, mpcc_id); new_mpcc = mpc->funcs->insert_plane(dc->res_pool->mpc, mpc_tree_params, &blnd_cfg, diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_init.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_init.c index 9c16633e473a0ff87de4589e60a7e88adbd7aa2f..92dd4cddbab8aa143ed8ad91e20f2dc6fe720e2d 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_init.c @@ -91,7 +91,7 @@ static const struct hw_sequencer_funcs dcn201_funcs = { .enable_dp_link_output = dce110_enable_dp_link_output, .disable_link_output = dce110_disable_link_output, .set_disp_pattern_generator = dcn20_set_disp_pattern_generator, - .update_visual_confirm_color = dcn20_update_visual_confirm_color, + .update_visual_confirm_color = dcn10_update_visual_confirm_color, }; static const struct hwseq_private_funcs dcn201_private_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_optc.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_optc.c index 730875dfd8b44a86b996913f4c9c4e4ab1d454da..70fcbec03fb6b7b067207fd236343cc4adf81587 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_optc.c @@ -55,10 +55,9 @@ static void optc201_triplebuffer_lock(struct timing_generator *optc) REG_SET(OTG_MASTER_UPDATE_LOCK, 0, OTG_MASTER_UPDATE_LOCK, 1); - if (optc->ctx->dce_environment != DCE_ENV_FPGA_MAXIMUS) - REG_WAIT(OTG_MASTER_UPDATE_LOCK, - UPDATE_LOCK_STATUS, 1, - 1, 10); + REG_WAIT(OTG_MASTER_UPDATE_LOCK, + UPDATE_LOCK_STATUS, 1, + 1, 10); } static void optc201_triplebuffer_unlock(struct timing_generator *optc) diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c index 6ea70da28aaaf529ccb5406257b6834c470f8dda..fdba8a9f5c30191260a683e44661b74cac4edd19 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c @@ -613,6 +613,7 @@ static const struct dc_debug_options debug_defaults_drv = { .sanity_checks = false, .underflow_assert_delay_us = 0xFFFFFFFF, .enable_tri_buf = false, + .enable_legacy_fast_update = true, }; static void dcn201_dpp_destroy(struct dpp **dpp) @@ -896,13 +897,6 @@ static const struct resource_create_funcs res_create_funcs = { .create_hwseq = dcn201_hwseq_create, }; -static const struct resource_create_funcs res_create_maximus_funcs = { - .read_dce_straps = NULL, - .create_audio = NULL, - .create_stream_encoder = NULL, - .create_hwseq = dcn201_hwseq_create, -}; - static void dcn201_clock_source_destroy(struct clock_source **clk_src) { kfree(TO_DCE110_CLK_SRC(*clk_src)); @@ -1272,9 +1266,8 @@ static bool dcn201_resource_construct( } if (!resource_construct(num_virtual_links, dc, &pool->base, - (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ? - &res_create_funcs : &res_create_maximus_funcs))) - goto create_fail; + &res_create_funcs)) + goto create_fail; dcn201_hw_sequencer_construct(dc); diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c index 58e459c7e7d35006ea76e710b732993b7e2b227c..f976fac8dc3f6fa1891b9953ab94aa55386c3729 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c @@ -667,7 +667,6 @@ static void program_surface_flip_and_addr(struct hubp *hubp, struct surface_flip static void dmcub_PLAT_54186_wa(struct hubp *hubp, struct surface_flip_registers *flip_regs) { - struct dc_dmub_srv *dmcub = hubp->ctx->dmub_srv; struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); union dmub_rb_cmd cmd; @@ -690,11 +689,7 @@ static void dmcub_PLAT_54186_wa(struct hubp *hubp, cmd.PLAT_54186_wa.flip.flip_params.vmid = flip_regs->vmid; PERF_TRACE(); // TODO: remove after performance is stable. - dc_dmub_srv_cmd_queue(dmcub, &cmd); - PERF_TRACE(); // TODO: remove after performance is stable. - dc_dmub_srv_cmd_execute(dmcub); - PERF_TRACE(); // TODO: remove after performance is stable. - dc_dmub_srv_wait_idle(dmcub); + dm_execute_dmub_cmd(hubp->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); PERF_TRACE(); // TODO: remove after performance is stable. } diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c index 2a182c2f57d6f11e34317a04902a11fe92f1b3a4..43463d08f21ba995187474f74d45935b23fe7366 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hwseq.c @@ -152,13 +152,28 @@ static bool dmub_abm_set_pipe(struct abm *abm, uint32_t otg_inst, uint32_t optio cmd.abm_set_pipe.abm_set_pipe_data.ramping_boundary = ramping_boundary; cmd.abm_set_pipe.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_pipe_data); - dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->dmub_srv); - dc_dmub_srv_wait_idle(dc->dmub_srv); + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } +static void dmub_abm_set_backlight(struct dc_context *dc, uint32_t backlight_pwm_u16_16, + uint32_t frame_ramp, uint32_t panel_inst) +{ + union dmub_rb_cmd cmd; + + memset(&cmd, 0, sizeof(cmd)); + cmd.abm_set_backlight.header.type = DMUB_CMD__ABM; + cmd.abm_set_backlight.header.sub_type = DMUB_CMD__ABM_SET_BACKLIGHT; + cmd.abm_set_backlight.abm_set_backlight_data.frame_ramp = frame_ramp; + cmd.abm_set_backlight.abm_set_backlight_data.backlight_user_level = backlight_pwm_u16_16; + cmd.abm_set_backlight.abm_set_backlight_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1; + cmd.abm_set_backlight.abm_set_backlight_data.panel_mask = (0x01 << panel_inst); + cmd.abm_set_backlight.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_backlight_data); + + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); +} + void dcn21_set_abm_immediate_disable(struct pipe_ctx *pipe_ctx) { struct abm *abm = pipe_ctx->stream_res.abm; @@ -173,8 +188,12 @@ void dcn21_set_abm_immediate_disable(struct pipe_ctx *pipe_ctx) } if (abm && panel_cntl) { - dmub_abm_set_pipe(abm, otg_inst, SET_ABM_PIPE_IMMEDIATELY_DISABLE, - panel_cntl->inst); + if (abm->funcs && abm->funcs->set_pipe_ex) { + abm->funcs->set_pipe_ex(abm, otg_inst, SET_ABM_PIPE_IMMEDIATELY_DISABLE, + panel_cntl->inst); + } else { + dmub_abm_set_pipe(abm, otg_inst, SET_ABM_PIPE_IMMEDIATELY_DISABLE, panel_cntl->inst); + } panel_cntl->funcs->store_backlight_level(panel_cntl); } } @@ -191,18 +210,21 @@ void dcn21_set_pipe(struct pipe_ctx *pipe_ctx) return; } - if (abm && panel_cntl) - dmub_abm_set_pipe(abm, otg_inst, SET_ABM_PIPE_NORMAL, panel_cntl->inst); + if (abm && panel_cntl) { + if (abm->funcs && abm->funcs->set_pipe_ex) { + abm->funcs->set_pipe_ex(abm, otg_inst, SET_ABM_PIPE_NORMAL, panel_cntl->inst); + } else { + dmub_abm_set_pipe(abm, otg_inst, SET_ABM_PIPE_NORMAL, panel_cntl->inst); + } + } } bool dcn21_set_backlight_level(struct pipe_ctx *pipe_ctx, uint32_t backlight_pwm_u16_16, uint32_t frame_ramp) { - union dmub_rb_cmd cmd; struct dc_context *dc = pipe_ctx->stream->ctx; struct abm *abm = pipe_ctx->stream_res.abm; - uint32_t otg_inst = pipe_ctx->stream_res.tg->inst; struct panel_cntl *panel_cntl = pipe_ctx->stream->link->panel_cntl; if (dc->dc->res_pool->dmcu) { @@ -210,21 +232,23 @@ bool dcn21_set_backlight_level(struct pipe_ctx *pipe_ctx, return true; } - if (abm && panel_cntl) - dmub_abm_set_pipe(abm, otg_inst, SET_ABM_PIPE_NORMAL, panel_cntl->inst); + if (abm != NULL) { + uint32_t otg_inst = pipe_ctx->stream_res.tg->inst; - memset(&cmd, 0, sizeof(cmd)); - cmd.abm_set_backlight.header.type = DMUB_CMD__ABM; - cmd.abm_set_backlight.header.sub_type = DMUB_CMD__ABM_SET_BACKLIGHT; - cmd.abm_set_backlight.abm_set_backlight_data.frame_ramp = frame_ramp; - cmd.abm_set_backlight.abm_set_backlight_data.backlight_user_level = backlight_pwm_u16_16; - cmd.abm_set_backlight.abm_set_backlight_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1; - cmd.abm_set_backlight.abm_set_backlight_data.panel_mask = (0x01 << panel_cntl->inst); - cmd.abm_set_backlight.header.payload_bytes = sizeof(struct dmub_cmd_abm_set_backlight_data); + if (abm && panel_cntl) { + if (abm->funcs && abm->funcs->set_pipe_ex) { + abm->funcs->set_pipe_ex(abm, otg_inst, SET_ABM_PIPE_NORMAL, panel_cntl->inst); + } else { + dmub_abm_set_pipe(abm, otg_inst, SET_ABM_PIPE_NORMAL, panel_cntl->inst); + } + } + } - dc_dmub_srv_cmd_queue(dc->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->dmub_srv); - dc_dmub_srv_wait_idle(dc->dmub_srv); + if (abm && abm->funcs && abm->funcs->set_backlight_level_pwm) + abm->funcs->set_backlight_level_pwm(abm, backlight_pwm_u16_16, + frame_ramp, 0, panel_cntl->inst); + else + dmub_abm_set_backlight(dc, backlight_pwm_u16_16, frame_ramp, panel_cntl->inst); return true; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c index fe1a8e2e08ef1bf689ce33c3f99a810c68802a02..f024157bd6eb6407a741292b007e7b0c7e9d0536 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c @@ -106,7 +106,7 @@ static const struct hw_sequencer_funcs dcn21_funcs = { .is_abm_supported = dcn21_is_abm_supported, .set_disp_pattern_generator = dcn20_set_disp_pattern_generator, .get_dcc_en_bits = dcn10_get_dcc_en_bits, - .update_visual_confirm_color = dcn20_update_visual_confirm_color, + .update_visual_confirm_color = dcn10_update_visual_confirm_color, }; static const struct hwseq_private_funcs dcn21_private_funcs = { @@ -151,8 +151,4 @@ void dcn21_hw_sequencer_construct(struct dc *dc) dc->hwss = dcn21_funcs; dc->hwseq->funcs = dcn21_private_funcs; - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - dc->hwss.init_hw = dcn20_fpga_init_hw; - dc->hwseq->funcs.init_pipes = NULL; - } } diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c index 19aaa557b2db009ffcf39c4ea935e261f2657223..d693ea42d033b9e9259ccd4ce9e43a53e075caa7 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c @@ -653,22 +653,7 @@ static const struct dc_debug_options debug_defaults_drv = { .usbc_combo_phy_reset_wa = true, .dmub_command_table = true, .use_max_lb = true, -}; - -static const struct dc_debug_options debug_defaults_diags = { - .disable_dmcu = false, - .force_abm_enable = false, - .timing_trace = true, - .clock_trace = true, - .disable_dpp_power_gate = true, - .disable_hubp_power_gate = true, - .disable_clock_gate = true, - .disable_pplib_clock_request = true, - .disable_pplib_wm_range = true, - .disable_stutter = true, - .disable_48mhz_pwrdwn = true, - .enable_tri_buf = true, - .use_max_lb = true + .enable_legacy_fast_update = true, }; static const struct dc_panel_config panel_config_defaults = { @@ -1219,13 +1204,6 @@ static const struct resource_create_funcs res_create_funcs = { .create_hwseq = dcn21_hwseq_create, }; -static const struct resource_create_funcs res_create_maximus_funcs = { - .read_dce_straps = NULL, - .create_audio = NULL, - .create_stream_encoder = NULL, - .create_hwseq = dcn21_hwseq_create, -}; - static const struct encoder_feature_support link_enc_feature = { .max_hdmi_deep_color = COLOR_DEPTH_121212, .max_hdmi_pixel_clock = 600000, @@ -1503,11 +1481,6 @@ static bool dcn21_resource_construct( if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) dc->debug = debug_defaults_drv; - else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) { - pool->base.pipe_count = 4; - dc->debug = debug_defaults_diags; - } else - dc->debug = debug_defaults_diags; // Init the vm_helper if (dc->vm_helper) @@ -1721,9 +1694,8 @@ static bool dcn21_resource_construct( } if (!resource_construct(num_virtual_links, dc, &pool->base, - (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ? - &res_create_funcs : &res_create_maximus_funcs))) - goto create_fail; + &res_create_funcs)) + goto create_fail; dcn21_hw_sequencer_construct(dc); diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/Makefile b/drivers/gpu/drm/amd/display/dc/dcn30/Makefile index b7c2ae9ddfda3a9ed8b552be2a40c3966f6284ec..4a3e9e47b6b6fb7d5fa43a806ab086461e7ea9e2 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn30/Makefile @@ -1,16 +1,16 @@ -# +# # Copyright 2020 Advanced Micro Devices, Inc. -# +# # Permission is hereby granted, free of charge, to any person obtaining a # copy of this software and associated documentation files (the "Software"), # to deal in the Software without restriction, including without limitation # the rights to use, copy, modify, merge, publish, distribute, sublicense, # and/or sell copies of the Software, and to permit persons to whom the # Software is furnished to do so, subject to the following conditions: -# +# # The above copyright notice and this permission notice shall be included in # all copies or substantial portions of the Software. -# +# # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL @@ -18,17 +18,31 @@ # OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, # ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR # OTHER DEALINGS IN THE SOFTWARE. -# +# # Authors: AMD -# -# +# +# + +DCN30 := \ + dcn30_init.o \ + dcn30_hubbub.o \ + dcn30_hubp.o \ + dcn30_dpp.o \ + dcn30_optc.o \ + dcn30_dccg.o \ + dcn30_hwseq.o \ + dcn30_mpc.o dcn30_vpg.o \ + dcn30_afmt.o \ + dcn30_dio_stream_encoder.o \ + dcn30_dwb.o \ + dcn30_dpp_cm.o \ + dcn30_dwb_cm.o \ + dcn30_cm_common.o \ + dcn30_mmhubbub.o \ + dcn30_resource.o \ + dcn30_dio_link_encoder.o -DCN30 = dcn30_init.o dcn30_hubbub.o dcn30_hubp.o dcn30_dpp.o dcn30_optc.o \ - dcn30_dccg.o dcn30_hwseq.o dcn30_mpc.o dcn30_vpg.o \ - dcn30_afmt.o dcn30_dio_stream_encoder.o dcn30_dwb.o \ - dcn30_dpp_cm.o dcn30_dwb_cm.o dcn30_cm_common.o dcn30_mmhubbub.o \ - dcn30_dio_link_encoder.o dcn30_resource.o AMD_DAL_DCN30 = $(addprefix $(AMDDALPATH)/dc/dcn30/,$(DCN30)) diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dio_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dio_stream_encoder.c index 9d08127d209b8e0e9b974ba31ae045311211bdd7..005dbe099a7a7f10e9bf15218a0a4ae13988e52e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dio_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dio_stream_encoder.c @@ -436,6 +436,21 @@ void enc3_stream_encoder_update_dp_info_packets( &info_frame->vsc, true); } + /* TODO: VSC SDP at packetIndex 1 should be retricted only if PSR-SU on. + * There should have another Infopacket type (e.g. vsc_psrsu) for PSR_SU. + * In addition, currently the driver check the valid bit then update and + * send the corresponding Infopacket. For PSR-SU, the SDP only be sent + * while entering PSR-SU mode. So we need another parameter(e.g. send) + * in dc_info_packet to indicate which infopacket should be enabled by + * default here. + */ + if (info_frame->vsc.valid) { + enc->vpg->funcs->update_generic_info_packet( + enc->vpg, + 1, /* packetIndex */ + &info_frame->vsc, + true); + } /* TODO: VSC SDP at packetIndex 1 should be restricted only if PSR-SU on. * There should have another Infopacket type (e.g. vsc_psrsu) for PSR_SU. * In addition, currently the driver check the valid bit then update and diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c index e46bbe7ddcc910d5b8f562bce1669667b8a2b75a..2861d974fcf62a0465ab6a39a6e65c249888ec27 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c @@ -449,6 +449,12 @@ void hubp3_read_state(struct hubp *hubp) SWATH_HEIGHT_C, &rq_regs->rq_regs_c.swath_height, PTE_ROW_HEIGHT_LINEAR_C, &rq_regs->rq_regs_c.pte_row_height_linear); + if (REG(UCLK_PSTATE_FORCE)) + s->uclk_pstate_force = REG_READ(UCLK_PSTATE_FORCE); + + if (REG(DCHUBP_CNTL)) + s->hubp_cntl = REG_READ(DCHUBP_CNTL); + } void hubp3_setup( diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c index 8263a07f265f2157cb72afce81dcddb0e9af572f..b9753867d97b00313566d82081dbcaa582b30bac 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c @@ -330,10 +330,6 @@ void dcn30_enable_writeback( DC_LOG_DWB("%s dwb_pipe_inst = %d, mpcc_inst = %d",\ __func__, wb_info->dwb_pipe_inst,\ wb_info->mpcc_inst); - if (IS_DIAG_DC(dc->ctx->dce_environment)) { - /*till diags switch to warmup interface*/ - dcn30_mmhubbub_warmup(dc, 1, wb_info); - } /* Update writeback pipe */ dcn30_set_writeback(dc, wb_info, context); @@ -447,28 +443,6 @@ void dcn30_init_hw(struct dc *dc) if (res_pool->dccg->funcs->dccg_init) res_pool->dccg->funcs->dccg_init(res_pool->dccg); - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - - REG_WRITE(REFCLK_CNTL, 0); - REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1); - REG_WRITE(DIO_MEM_PWR_CTRL, 0); - - if (!dc->debug.disable_clock_gate) { - /* enable all DCN clock gating */ - REG_WRITE(DCCG_GATE_DISABLE_CNTL, 0); - - REG_WRITE(DCCG_GATE_DISABLE_CNTL2, 0); - - REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0); - } - - //Enable ability to power gate / don't force power on permanently - if (hws->funcs.enable_power_gating_plane) - hws->funcs.enable_power_gating_plane(hws, true); - - return; - } - if (!dcb->funcs->is_accelerated_mode(dcb)) { hws->funcs.bios_golden_init(dc); hws->funcs.disable_vga(dc->hwseq); @@ -491,23 +465,21 @@ void dcn30_init_hw(struct dc *dc) res_pool->ref_clocks.xtalin_clock_inKhz = dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency; - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - if (res_pool->dccg && res_pool->hubbub) { + if (res_pool->dccg && res_pool->hubbub) { - (res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg, - dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency, - &res_pool->ref_clocks.dccg_ref_clock_inKhz); + (res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg, + dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency, + &res_pool->ref_clocks.dccg_ref_clock_inKhz); - (res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub, - res_pool->ref_clocks.dccg_ref_clock_inKhz, - &res_pool->ref_clocks.dchub_ref_clock_inKhz); - } else { - // Not all ASICs have DCCG sw component - res_pool->ref_clocks.dccg_ref_clock_inKhz = - res_pool->ref_clocks.xtalin_clock_inKhz; - res_pool->ref_clocks.dchub_ref_clock_inKhz = - res_pool->ref_clocks.xtalin_clock_inKhz; - } + (res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub, + res_pool->ref_clocks.dccg_ref_clock_inKhz, + &res_pool->ref_clocks.dchub_ref_clock_inKhz); + } else { + // Not all ASICs have DCCG sw component + res_pool->ref_clocks.dccg_ref_clock_inKhz = + res_pool->ref_clocks.xtalin_clock_inKhz; + res_pool->ref_clocks.dchub_ref_clock_inKhz = + res_pool->ref_clocks.xtalin_clock_inKhz; } } else ASSERT_CRITICAL(false); @@ -632,7 +604,7 @@ void dcn30_init_hw(struct dc *dc) dc->res_pool->hubbub->funcs->init_crb(dc->res_pool->hubbub); // Get DMCUB capabilities - dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv->dmub); + dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv); dc->caps.dmub_caps.psr = dc->ctx->dmub_srv->dmub->feature_caps.psr; dc->caps.dmub_caps.mclk_sw = dc->ctx->dmub_srv->dmub->feature_caps.fw_assisted_mclk_switch; } @@ -736,8 +708,7 @@ bool dcn30_apply_idle_power_optimizations(struct dc *dc, bool enable) cmd.mall.header.sub_type = DMUB_CMD__MALL_ACTION_NO_DF_REQ; cmd.mall.header.payload_bytes = sizeof(cmd.mall) - sizeof(cmd.mall.header); - dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); return true; } @@ -859,9 +830,7 @@ bool dcn30_apply_idle_power_optimizations(struct dc *dc, bool enable) cmd.mall.cursor_height = cursor_attr.height; cmd.mall.cursor_pitch = cursor_attr.pitch; - dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv); - dc_dmub_srv_wait_idle(dc->ctx->dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); /* Use copied cursor, and it's okay to not switch back */ cursor_attr.address.quad_part = cmd.mall.cursor_copy_dst.quad_part; @@ -877,8 +846,7 @@ bool dcn30_apply_idle_power_optimizations(struct dc *dc, bool enable) cmd.mall.tmr_scale = tmr_scale; cmd.mall.debug_bits = dc->debug.mall_error_as_fatal; - dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); return true; } @@ -895,9 +863,7 @@ bool dcn30_apply_idle_power_optimizations(struct dc *dc, bool enable) cmd.mall.header.payload_bytes = sizeof(cmd.mall) - sizeof(cmd.mall.header); - dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv); - dc_dmub_srv_wait_idle(dc->ctx->dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } @@ -983,36 +949,13 @@ void dcn30_set_disp_pattern_generator(const struct dc *dc, } void dcn30_prepare_bandwidth(struct dc *dc, - struct dc_state *context) + struct dc_state *context) { - bool p_state_change_support = context->bw_ctx.bw.dcn.clk.p_state_change_support; - /* Any transition into an FPO config should disable MCLK switching first to avoid - * driver and FW P-State synchronization issues. - */ - if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching || dc->clk_mgr->clks.fw_based_mclk_switching) { - dc->optimized_required = true; - context->bw_ctx.bw.dcn.clk.p_state_change_support = false; - } - if (dc->clk_mgr->dc_mode_softmax_enabled) if (dc->clk_mgr->clks.dramclk_khz <= dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000 && context->bw_ctx.bw.dcn.clk.dramclk_khz > dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000) dc->clk_mgr->funcs->set_max_memclk(dc->clk_mgr, dc->clk_mgr->bw_params->clk_table.entries[dc->clk_mgr->bw_params->clk_table.num_entries - 1].memclk_mhz); dcn20_prepare_bandwidth(dc, context); - /* - * enabled -> enabled: do not disable - * enabled -> disabled: disable - * disabled -> enabled: don't care - * disabled -> disabled: don't care - */ - if (!context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) - dc_dmub_srv_p_state_delegate(dc, false, context); - - if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching || dc->clk_mgr->clks.fw_based_mclk_switching) { - /* After disabling P-State, restore the original value to ensure we get the correct P-State - * on the next optimize. */ - context->bw_ctx.bw.dcn.clk.p_state_change_support = p_state_change_support; - } } diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c index 3216d10c58ba71d31bfd992e7f28e30769a5194b..3d19acaa12f3ce3e5b20092fa44767379a7fe97e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c @@ -106,7 +106,7 @@ static const struct hw_sequencer_funcs dcn30_funcs = { .disable_link_output = dce110_disable_link_output, .set_disp_pattern_generator = dcn30_set_disp_pattern_generator, .get_dcc_en_bits = dcn10_get_dcc_en_bits, - .update_visual_confirm_color = dcn20_update_visual_confirm_color, + .update_visual_confirm_color = dcn10_update_visual_confirm_color, .is_abm_supported = dcn21_is_abm_supported }; @@ -151,8 +151,4 @@ void dcn30_hw_sequencer_construct(struct dc *dc) dc->hwss = dcn30_funcs; dc->hwseq->funcs = dcn30_private_funcs; - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - dc->hwss.init_hw = dcn20_fpga_init_hw; - dc->hwseq->funcs.init_pipes = NULL; - } } diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c index c95f000b63b28d7e5eeb0580fca4d6e0a3063a00..dfb8f62765f28e6f52110355cd9edd560184e246 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c @@ -55,10 +55,9 @@ void optc3_triplebuffer_lock(struct timing_generator *optc) REG_SET(OTG_MASTER_UPDATE_LOCK, 0, OTG_MASTER_UPDATE_LOCK, 1); - if (optc->ctx->dce_environment != DCE_ENV_FPGA_MAXIMUS) - REG_WAIT(OTG_MASTER_UPDATE_LOCK, - UPDATE_LOCK_STATUS, 1, - 1, 10); + REG_WAIT(OTG_MASTER_UPDATE_LOCK, + UPDATE_LOCK_STATUS, 1, + 1, 10); TRACE_OPTC_LOCK_UNLOCK_STATE(optc1, optc->inst, true); } @@ -280,6 +279,9 @@ static void optc3_set_odm_combine(struct timing_generator *optc, int *opp_id, in * Sets double buffer point for V_TOTAL, H_TOTAL, VTOTAL_MIN, * VTOTAL_MAX, VTOTAL_MIN_SEL and VTOTAL_MAX_SEL registers. * + * @optc: timing_generator instance. + * @enable: Enable DRR double buffering control if true, disable otherwise. + * * Options: any time, start of frame, dp start of frame (range timing) */ static void optc3_set_timing_double_buffer(struct timing_generator *optc, bool enable) @@ -301,7 +303,12 @@ static void optc3_wait_drr_doublebuffer_pending_clear(struct timing_generator *o void optc3_set_vtotal_min_max(struct timing_generator *optc, int vtotal_min, int vtotal_max) { - optc1_set_vtotal_min_max(optc, vtotal_min, vtotal_max); + struct dc *dc = optc->ctx->dc; + + if (dc->caps.dmub_caps.mclk_sw && !dc->debug.disable_fams) + dc_dmub_srv_drr_update_cmd(dc, optc->inst, vtotal_min, vtotal_max); + else + optc1_set_vtotal_min_max(optc, vtotal_min, vtotal_max); } void optc3_tg_init(struct timing_generator *optc) diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c index 67a34cda3774ae46afafa0c00b5a7b5989f0e9ce..1a0284a068b2a74e24a96d79af6f6272dc303ac5 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c @@ -728,24 +728,6 @@ static const struct dc_debug_options debug_defaults_drv = { .exit_idle_opt_for_cursor_updates = true }; -static const struct dc_debug_options debug_defaults_diags = { - .disable_dmcu = true, //No dmcu on DCN30 - .force_abm_enable = false, - .timing_trace = true, - .clock_trace = true, - .disable_dpp_power_gate = true, - .disable_hubp_power_gate = true, - .disable_clock_gate = true, - .disable_pplib_clock_request = true, - .disable_pplib_wm_range = true, - .disable_stutter = false, - .scl_reset_length10 = true, - .dwb_fi_phase = -1, // -1 = disable - .dmub_command_table = true, - .enable_tri_buf = true, - .use_max_lb = true -}; - static const struct dc_panel_config panel_config_defaults = { .psr = { .disable_psr = false, @@ -1076,13 +1058,6 @@ static const struct resource_create_funcs res_create_funcs = { .create_hwseq = dcn30_hwseq_create, }; -static const struct resource_create_funcs res_create_maximus_funcs = { - .read_dce_straps = NULL, - .create_audio = NULL, - .create_stream_encoder = NULL, - .create_hwseq = dcn30_hwseq_create, -}; - static void dcn30_resource_destruct(struct dcn30_resource_pool *pool) { unsigned int i; @@ -2353,6 +2328,7 @@ static bool dcn30_resource_construct( dc->caps.color.mpc.ocsc = 1; dc->caps.dp_hdmi21_pcon_support = true; + dc->caps.max_v_total = (1 << 15) - 1; /* read VBIOS LTTPR caps */ { @@ -2376,10 +2352,7 @@ static bool dcn30_resource_construct( if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) dc->debug = debug_defaults_drv; - else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) { - dc->debug = debug_defaults_diags; - } else - dc->debug = debug_defaults_diags; + // Init the vm_helper if (dc->vm_helper) vm_helper_init(dc->vm_helper, 16); @@ -2577,8 +2550,7 @@ static bool dcn30_resource_construct( /* Audio, Stream Encoders including DIG and virtual, MPC 3D LUTs */ if (!resource_construct(num_virtual_links, dc, &pool->base, - (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ? - &res_create_funcs : &res_create_maximus_funcs))) + &res_create_funcs)) goto create_fail; /* HW Sequencer and Plane caps */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c index 6192851c59ed8950653ba4668177417eb2cf1a3d..257df8660b4cafef2a64231487729bbd2d4b3486 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c @@ -107,7 +107,7 @@ static const struct hw_sequencer_funcs dcn301_funcs = { .get_dcc_en_bits = dcn10_get_dcc_en_bits, .optimize_pwr_state = dcn21_optimize_pwr_state, .exit_optimized_pwr_state = dcn21_exit_optimized_pwr_state, - .update_visual_confirm_color = dcn20_update_visual_confirm_color, + .update_visual_confirm_color = dcn10_update_visual_confirm_color, }; static const struct hwseq_private_funcs dcn301_private_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c index 5ac2a272c380fb968e6768f10f49dcf988f7a441..3485fbb1093efbafa1673692e3bdcf07a729a469 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c @@ -702,23 +702,6 @@ static const struct dc_debug_options debug_defaults_drv = { .exit_idle_opt_for_cursor_updates = true }; -static const struct dc_debug_options debug_defaults_diags = { - .disable_dmcu = true, - .force_abm_enable = false, - .timing_trace = true, - .clock_trace = true, - .disable_dpp_power_gate = false, - .disable_hubp_power_gate = false, - .disable_clock_gate = true, - .disable_pplib_clock_request = true, - .disable_pplib_wm_range = true, - .disable_stutter = true, - .scl_reset_length10 = true, - .dwb_fi_phase = -1, // -1 = disable - .dmub_command_table = true, - .use_max_lb = false, -}; - static void dcn301_dpp_destroy(struct dpp **dpp) { kfree(TO_DCN20_DPP(*dpp)); @@ -1047,13 +1030,6 @@ static const struct resource_create_funcs res_create_funcs = { .create_hwseq = dcn301_hwseq_create, }; -static const struct resource_create_funcs res_create_maximus_funcs = { - .read_dce_straps = NULL, - .create_audio = NULL, - .create_stream_encoder = NULL, - .create_hwseq = dcn301_hwseq_create, -}; - static void dcn301_destruct(struct dcn301_resource_pool *pool) { unsigned int i; @@ -1513,10 +1489,7 @@ static bool dcn301_resource_construct( if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) dc->debug = debug_defaults_drv; - else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) { - dc->debug = debug_defaults_diags; - } else - dc->debug = debug_defaults_diags; + // Init the vm_helper if (dc->vm_helper) vm_helper_init(dc->vm_helper, 16); @@ -1710,9 +1683,8 @@ static bool dcn301_resource_construct( /* Audio, Stream Encoders including HPO and virtual, MPC 3D LUTs */ if (!resource_construct(num_virtual_links, dc, &pool->base, - (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ? - &res_create_funcs : &res_create_maximus_funcs))) - goto create_fail; + &res_create_funcs)) + goto create_fail; /* HW Sequencer and Plane caps */ dcn301_hw_sequencer_construct(dc); diff --git a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c index 9f93c43115ba316c514cb6027b808de421428d10..7dc065ea247a1824175ef8bf0fcf94334d86f977 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c @@ -98,24 +98,6 @@ static const struct dc_debug_options debug_defaults_drv = { .exit_idle_opt_for_cursor_updates = true }; -static const struct dc_debug_options debug_defaults_diags = { - .disable_dmcu = true, - .force_abm_enable = false, - .timing_trace = true, - .clock_trace = true, - .disable_dpp_power_gate = true, - .disable_hubp_power_gate = true, - .disable_clock_gate = true, - .disable_pplib_clock_request = true, - .disable_pplib_wm_range = true, - .disable_stutter = false, - .scl_reset_length10 = true, - .dwb_fi_phase = -1, // -1 = disable - .dmub_command_table = true, - .enable_tri_buf = true, - .use_max_lb = true -}; - static const struct dc_panel_config panel_config_defaults = { .psr = { .disable_psr = false, @@ -954,13 +936,6 @@ static const struct resource_create_funcs res_create_funcs = { .create_hwseq = dcn302_hwseq_create, }; -static const struct resource_create_funcs res_create_maximus_funcs = { - .read_dce_straps = NULL, - .create_audio = NULL, - .create_stream_encoder = NULL, - .create_hwseq = dcn302_hwseq_create, -}; - static bool is_soc_bounding_box_valid(struct dc *dc) { uint32_t hw_internal_rev = dc->ctx->asic_id.hw_internal_rev; @@ -1252,6 +1227,7 @@ static bool dcn302_resource_construct( dc->caps.force_dp_tps4_for_cp2520 = true; dc->caps.extended_aux_timeout_support = true; dc->caps.dmcub_support = true; + dc->caps.max_v_total = (1 << 15) - 1; /* Color pipeline capabilities */ dc->caps.color.dpp.dcn_arch = 1; @@ -1309,8 +1285,6 @@ static bool dcn302_resource_construct( if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) dc->debug = debug_defaults_drv; - else - dc->debug = debug_defaults_diags; // Init the vm_helper if (dc->vm_helper) @@ -1489,8 +1463,7 @@ static bool dcn302_resource_construct( /* Audio, Stream Encoders including HPO and virtual, MPC 3D LUTs */ if (!resource_construct(num_virtual_links, dc, pool, - (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ? - &res_create_funcs : &res_create_maximus_funcs))) + &res_create_funcs)) goto create_fail; /* HW Sequencer and Plane caps */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c index 7f72ef882ca417618d2ea77223ceb9b433442972..6d9761395288db268ca69df92c8ce9a7a557885e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c @@ -81,23 +81,6 @@ static const struct dc_debug_options debug_defaults_drv = { .disable_idle_power_optimizations = false, }; -static const struct dc_debug_options debug_defaults_diags = { - .disable_dmcu = true, - .force_abm_enable = false, - .timing_trace = true, - .clock_trace = true, - .disable_dpp_power_gate = true, - .disable_hubp_power_gate = true, - .disable_clock_gate = true, - .disable_pplib_clock_request = true, - .disable_pplib_wm_range = true, - .disable_stutter = false, - .scl_reset_length10 = true, - .dwb_fi_phase = -1, // -1 = disable - .dmub_command_table = true, - .enable_tri_buf = true, -}; - static const struct dc_panel_config panel_config_defaults = { .psr = { .disable_psr = false, @@ -881,13 +864,6 @@ static const struct resource_create_funcs res_create_funcs = { .create_hwseq = dcn303_hwseq_create, }; -static const struct resource_create_funcs res_create_maximus_funcs = { - .read_dce_straps = NULL, - .create_audio = NULL, - .create_stream_encoder = NULL, - .create_hwseq = dcn303_hwseq_create, -}; - static bool is_soc_bounding_box_valid(struct dc *dc) { uint32_t hw_internal_rev = dc->ctx->asic_id.hw_internal_rev; @@ -1176,6 +1152,7 @@ static bool dcn303_resource_construct( dc->caps.force_dp_tps4_for_cp2520 = true; dc->caps.extended_aux_timeout_support = true; dc->caps.dmcub_support = true; + dc->caps.max_v_total = (1 << 15) - 1; /* Color pipeline capabilities */ dc->caps.color.dpp.dcn_arch = 1; @@ -1232,8 +1209,6 @@ static bool dcn303_resource_construct( if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) dc->debug = debug_defaults_drv; - else - dc->debug = debug_defaults_diags; // Init the vm_helper if (dc->vm_helper) @@ -1400,8 +1375,7 @@ static bool dcn303_resource_construct( /* Audio, Stream Encoders including HPO and virtual, MPC 3D LUTs */ if (!resource_construct(num_virtual_links, dc, pool, - (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ? - &res_create_funcs : &res_create_maximus_funcs))) + &res_create_funcs)) goto create_fail; /* HW Sequencer and Plane caps */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c index 4c2fdfea162f55cb567e2ac702936221d70eec1a..65c1d754e2d6b1fb173f86d2603fb1a17550b809 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c @@ -47,6 +47,14 @@ void dccg31_update_dpp_dto(struct dccg *dccg, int dpp_inst, int req_dppclk) { struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg); + if (dccg->dpp_clock_gated[dpp_inst]) { + /* + * Do not update the DPPCLK DTO if the clock is stopped. + * It is treated the same as if the pipe itself were in PG. + */ + return; + } + if (dccg->ref_dppclk && req_dppclk) { int ref_dppclk = dccg->ref_dppclk; int modulo, phase; diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c index 745a5d187a98d7e372c3e2920f22ea548336650b..bd62502380d8da7961634084f8627f9cc2e18437 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c @@ -117,7 +117,6 @@ static bool query_dp_alt_from_dmub(struct link_encoder *enc, union dmub_rb_cmd *cmd) { struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); - struct dc_dmub_srv *dc_dmub_srv = enc->ctx->dmub_srv; memset(cmd, 0, sizeof(*cmd)); cmd->query_dp_alt.header.type = DMUB_CMD__VBIOS; @@ -126,7 +125,7 @@ static bool query_dp_alt_from_dmub(struct link_encoder *enc, cmd->query_dp_alt.header.payload_bytes = sizeof(cmd->query_dp_alt.data); cmd->query_dp_alt.data.phy_id = phy_id_from_transmitter(enc10->base.transmitter); - if (!dc_dmub_srv_cmd_with_reply_data(dc_dmub_srv, cmd)) + if (!dm_execute_dmub_cmd(enc->ctx, cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) return false; return true; @@ -425,7 +424,6 @@ static bool link_dpia_control(struct dc_context *dc_ctx, struct dmub_cmd_dig_dpia_control_data *dpia_control) { union dmub_rb_cmd cmd; - struct dc_dmub_srv *dmub = dc_ctx->dmub_srv; memset(&cmd, 0, sizeof(cmd)); @@ -438,9 +436,7 @@ static bool link_dpia_control(struct dc_context *dc_ctx, cmd.dig1_dpia_control.dpia_control = *dpia_control; - dc_dmub_srv_cmd_queue(dmub, &cmd); - dc_dmub_srv_cmd_execute(dmub); - dc_dmub_srv_wait_idle(dmub); + dm_execute_dmub_cmd(dc_ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hubbub.c index 7e7cd5b64e6a1b46ab2f7e650ad05292dfc3727f..7445ed27852a1f3e32ea968662a171f61ccc123b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hubbub.c @@ -103,6 +103,7 @@ static void dcn31_program_det_size(struct hubbub *hubbub, int hubp_inst, unsigne default: break; } + DC_LOG_DEBUG("Set DET%d to %d segments\n", hubp_inst, det_size_segments); /* Should never be hit, if it is we have an erroneous hw config*/ ASSERT(hubbub2->det0_size + hubbub2->det1_size + hubbub2->det2_size + hubbub2->det3_size + hubbub2->compbuf_size_segments <= hubbub2->crb_size_segs); diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c index 62ce36c75c4d24375dfd8e6484ae3c516491b80f..2a7f47642a4479aedecda90136412ac44bf3be72 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c @@ -117,28 +117,6 @@ void dcn31_init_hw(struct dc *dc) if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks) dc->clk_mgr->funcs->init_clocks(dc->clk_mgr); - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - - REG_WRITE(REFCLK_CNTL, 0); - REG_UPDATE(DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBBUB_GLOBAL_TIMER_ENABLE, 1); - REG_WRITE(DIO_MEM_PWR_CTRL, 0); - - if (!dc->debug.disable_clock_gate) { - /* enable all DCN clock gating */ - REG_WRITE(DCCG_GATE_DISABLE_CNTL, 0); - - REG_WRITE(DCCG_GATE_DISABLE_CNTL2, 0); - - REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0); - } - - //Enable ability to power gate / don't force power on permanently - if (hws->funcs.enable_power_gating_plane) - hws->funcs.enable_power_gating_plane(hws, true); - - return; - } - if (!dcb->funcs->is_accelerated_mode(dcb)) { hws->funcs.bios_golden_init(dc); if (hws->funcs.disable_vga) @@ -154,23 +132,21 @@ void dcn31_init_hw(struct dc *dc) res_pool->ref_clocks.xtalin_clock_inKhz = dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency; - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - if (res_pool->dccg && res_pool->hubbub) { - - (res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg, - dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency, - &res_pool->ref_clocks.dccg_ref_clock_inKhz); - - (res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub, - res_pool->ref_clocks.dccg_ref_clock_inKhz, - &res_pool->ref_clocks.dchub_ref_clock_inKhz); - } else { - // Not all ASICs have DCCG sw component - res_pool->ref_clocks.dccg_ref_clock_inKhz = - res_pool->ref_clocks.xtalin_clock_inKhz; - res_pool->ref_clocks.dchub_ref_clock_inKhz = - res_pool->ref_clocks.xtalin_clock_inKhz; - } + if (res_pool->dccg && res_pool->hubbub) { + + (res_pool->dccg->funcs->get_dccg_ref_freq)(res_pool->dccg, + dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency, + &res_pool->ref_clocks.dccg_ref_clock_inKhz); + + (res_pool->hubbub->funcs->get_dchub_ref_freq)(res_pool->hubbub, + res_pool->ref_clocks.dccg_ref_clock_inKhz, + &res_pool->ref_clocks.dchub_ref_clock_inKhz); + } else { + // Not all ASICs have DCCG sw component + res_pool->ref_clocks.dccg_ref_clock_inKhz = + res_pool->ref_clocks.xtalin_clock_inKhz; + res_pool->ref_clocks.dchub_ref_clock_inKhz = + res_pool->ref_clocks.xtalin_clock_inKhz; } } else ASSERT_CRITICAL(false); @@ -197,10 +173,6 @@ void dcn31_init_hw(struct dc *dc) } } - /* Enables outbox notifications for usb4 dpia */ - if (dc->res_pool->usb4_dpia_count) - dmub_enable_outbox_notification(dc->ctx->dmub_srv); - /* we want to turn off all dp displays before doing detection */ dc->link_srv->blank_all_dp_displays(dc); @@ -297,8 +269,9 @@ void dcn31_init_hw(struct dc *dc) #endif // Get DMCUB capabilities - dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv->dmub); + dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv); dc->caps.dmub_caps.psr = dc->ctx->dmub_srv->dmub->feature_caps.psr; + dc->caps.dmub_caps.mclk_sw = dc->ctx->dmub_srv->dmub->feature_caps.fw_assisted_mclk_switch; } void dcn31_dsc_pg_control( @@ -442,9 +415,7 @@ void dcn31_z10_save_init(struct dc *dc) cmd.dcn_restore.header.type = DMUB_CMD__IDLE_OPT; cmd.dcn_restore.header.sub_type = DMUB_CMD__IDLE_OPT_DCN_SAVE_INIT; - dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv); - dc_dmub_srv_wait_idle(dc->ctx->dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } void dcn31_z10_restore(const struct dc *dc) @@ -462,9 +433,7 @@ void dcn31_z10_restore(const struct dc *dc) cmd.dcn_restore.header.type = DMUB_CMD__IDLE_OPT; cmd.dcn_restore.header.sub_type = DMUB_CMD__IDLE_OPT_DCN_RESTORE; - dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv); - dc_dmub_srv_wait_idle(dc->ctx->dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } void dcn31_hubp_pg_control(struct dce_hwseq *hws, unsigned int hubp_inst, bool power_on) @@ -560,35 +529,31 @@ static void dcn31_reset_back_end_for_pipe( pipe_ctx->stream_res.tg->funcs->set_drr( pipe_ctx->stream_res.tg, NULL); - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - link = pipe_ctx->stream->link; - /* DPMS may already disable or */ - /* dpms_off status is incorrect due to fastboot - * feature. When system resume from S4 with second - * screen only, the dpms_off would be true but - * VBIOS lit up eDP, so check link status too. - */ - if (!pipe_ctx->stream->dpms_off || link->link_status.link_active) - dc->link_srv->set_dpms_off(pipe_ctx); - else if (pipe_ctx->stream_res.audio) - dc->hwss.disable_audio_stream(pipe_ctx); - - /* free acquired resources */ - if (pipe_ctx->stream_res.audio) { - /*disable az_endpoint*/ - pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio); - - /*free audio*/ - if (dc->caps.dynamic_audio == true) { - /*we have to dynamic arbitrate the audio endpoints*/ - /*we free the resource, need reset is_audio_acquired*/ - update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, - pipe_ctx->stream_res.audio, false); - pipe_ctx->stream_res.audio = NULL; - } + link = pipe_ctx->stream->link; + /* DPMS may already disable or */ + /* dpms_off status is incorrect due to fastboot + * feature. When system resume from S4 with second + * screen only, the dpms_off would be true but + * VBIOS lit up eDP, so check link status too. + */ + if (!pipe_ctx->stream->dpms_off || link->link_status.link_active) + dc->link_srv->set_dpms_off(pipe_ctx); + else if (pipe_ctx->stream_res.audio) + dc->hwss.disable_audio_stream(pipe_ctx); + + /* free acquired resources */ + if (pipe_ctx->stream_res.audio) { + /*disable az_endpoint*/ + pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio); + + /*free audio*/ + if (dc->caps.dynamic_audio == true) { + /*we have to dynamic arbitrate the audio endpoints*/ + /*we free the resource, need reset is_audio_acquired*/ + update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, + pipe_ctx->stream_res.audio, false); + pipe_ctx->stream_res.audio = NULL; } - } else if (pipe_ctx->stream_res.dsc) { - dc->link_srv->set_dsc_enable(pipe_ctx, false); } pipe_ctx->stream = NULL; diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c index 3a32810bbe382d45a01fdfc0ab80d367c8aee14c..fc25cc300a175b980daa8e7646541342e06b0a9f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c @@ -58,6 +58,7 @@ static const struct hw_sequencer_funcs dcn31_funcs = { .enable_audio_stream = dce110_enable_audio_stream, .disable_audio_stream = dce110_disable_audio_stream, .disable_plane = dcn20_disable_plane, + .disable_pixel_data = dcn20_disable_pixel_data, .pipe_control_lock = dcn20_pipe_control_lock, .interdependent_update_lock = dcn10_lock_all_pipes, .cursor_lock = dcn10_cursor_lock, @@ -109,7 +110,7 @@ static const struct hw_sequencer_funcs dcn31_funcs = { .set_disp_pattern_generator = dcn30_set_disp_pattern_generator, .optimize_pwr_state = dcn21_optimize_pwr_state, .exit_optimized_pwr_state = dcn21_exit_optimized_pwr_state, - .update_visual_confirm_color = dcn20_update_visual_confirm_color, + .update_visual_confirm_color = dcn10_update_visual_confirm_color, }; static const struct hwseq_private_funcs dcn31_private_funcs = { @@ -153,8 +154,4 @@ void dcn31_hw_sequencer_construct(struct dc *dc) dc->hwss = dcn31_funcs; dc->hwseq->funcs = dcn31_private_funcs; - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - dc->hwss.init_hw = dcn20_fpga_init_hw; - dc->hwseq->funcs.init_pipes = NULL; - } } diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c index 11ea9d13e3128f0c68e5bd46188a01537accb19c..217acd4e292a30225a06cabbeb60d33fdbf2ec97 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c @@ -52,7 +52,7 @@ static bool dcn31_query_backlight_info(struct panel_cntl *panel_cntl, union dmub cmd->panel_cntl.header.payload_bytes = sizeof(cmd->panel_cntl.data); cmd->panel_cntl.data.inst = dcn31_panel_cntl->base.inst; - return dc_dmub_srv_cmd_with_reply_data(dc_dmub_srv, cmd); + return dm_execute_dmub_cmd(dc_dmub_srv->ctx, cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY); } static uint32_t dcn31_get_16_bit_backlight_from_pwm(struct panel_cntl *panel_cntl) @@ -85,7 +85,7 @@ static uint32_t dcn31_panel_cntl_hw_init(struct panel_cntl *panel_cntl) panel_cntl->stored_backlight_registers.LVTMA_PWRSEQ_REF_DIV_BL_PWM_REF_DIV; cmd.panel_cntl.data.bl_pwm_ref_div2 = panel_cntl->stored_backlight_registers.PANEL_PWRSEQ_REF_DIV2; - if (!dc_dmub_srv_cmd_with_reply_data(dc_dmub_srv, &cmd)) + if (!dm_execute_dmub_cmd(dc_dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) return 0; panel_cntl->stored_backlight_registers.BL_PWM_CNTL = cmd.panel_cntl.data.bl_pwm_cntl; diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c index ff8cd50764348ca98891ba9b3bd0534d9e47132e..fc33b5fcabe1094c8d374bdfb274d86a614160ca 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c @@ -887,28 +887,11 @@ static const struct dc_debug_options debug_defaults_drv = { } }, .disable_z10 = true, + .enable_legacy_fast_update = true, .enable_z9_disable_interface = true, /* Allow support for the PMFW interface for disable Z9*/ .dml_hostvm_override = DML_HOSTVM_OVERRIDE_FALSE, }; -static const struct dc_debug_options debug_defaults_diags = { - .disable_dmcu = true, - .force_abm_enable = false, - .timing_trace = true, - .clock_trace = true, - .disable_dpp_power_gate = true, - .disable_hubp_power_gate = true, - .disable_clock_gate = true, - .disable_pplib_clock_request = true, - .disable_pplib_wm_range = true, - .disable_stutter = false, - .scl_reset_length10 = true, - .dwb_fi_phase = -1, // -1 = disable - .dmub_command_table = true, - .enable_tri_buf = true, - .use_max_lb = true -}; - static const struct dc_panel_config panel_config_defaults = { .psr = { .disable_psr = false, @@ -1341,13 +1324,6 @@ static struct dce_hwseq *dcn31_hwseq_create( hws->regs = &hwseq_reg; hws->shifts = &hwseq_shift; hws->masks = &hwseq_mask; - /* DCN3.1 FPGA Workaround - * Need to enable HPO DP Stream Encoder before setting OTG master enable. - * To do so, move calling function enable_stream_timing to only be done AFTER calling - * function core_link_enable_stream - */ - if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) - hws->wa.dp_hpo_and_otg_sequence = true; } return hws; } @@ -1360,15 +1336,6 @@ static const struct resource_create_funcs res_create_funcs = { .create_hwseq = dcn31_hwseq_create, }; -static const struct resource_create_funcs res_create_maximus_funcs = { - .read_dce_straps = NULL, - .create_audio = NULL, - .create_stream_encoder = NULL, - .create_hpo_dp_stream_encoder = dcn31_hpo_dp_stream_encoder_create, - .create_hpo_dp_link_encoder = dcn31_hpo_dp_link_encoder_create, - .create_hwseq = dcn31_hwseq_create, -}; - static void dcn31_resource_destruct(struct dcn31_resource_pool *pool) { unsigned int i; @@ -1988,10 +1955,7 @@ static bool dcn31_resource_construct( if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) dc->debug = debug_defaults_drv; - else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) { - dc->debug = debug_defaults_diags; - } else - dc->debug = debug_defaults_diags; + // Init the vm_helper if (dc->vm_helper) vm_helper_init(dc->vm_helper, 16); @@ -2195,9 +2159,8 @@ static bool dcn31_resource_construct( /* Audio, Stream Encoders including HPO and virtual, MPC 3D LUTs */ if (!resource_construct(num_virtual_links, dc, &pool->base, - (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ? - &res_create_funcs : &res_create_maximus_funcs))) - goto create_fail; + &res_create_funcs)) + goto create_fail; /* HW Sequencer and Plane caps */ dcn31_hw_sequencer_construct(dc); diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c index de7bfba2c179802b70f214a1e96b75946ce23847..cf23d7bc560a6c92aef84c00f1b081a59b21a6bd 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c @@ -45,6 +45,16 @@ #define DC_LOGGER \ dccg->ctx->logger +static void dccg314_trigger_dio_fifo_resync( + struct dccg *dccg) +{ + struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg); + uint32_t dispclk_rdivider_value = 0; + + REG_GET(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_RDIVIDER, &dispclk_rdivider_value); + REG_UPDATE(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, dispclk_rdivider_value); +} + static void dccg314_get_pixel_rate_div( struct dccg *dccg, uint32_t otg_inst, @@ -322,6 +332,9 @@ static void dccg314_dpp_root_clock_control( { struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg); + if (dccg->dpp_clock_gated[dpp_inst] == clock_on) + return; + if (clock_on) { /* turn off the DTO and leave phase/modulo at max */ REG_UPDATE(DPPCLK_DTO_CTRL, DPPCLK_DTO_ENABLE[dpp_inst], 0); @@ -335,6 +348,8 @@ static void dccg314_dpp_root_clock_control( DPPCLK0_DTO_PHASE, 0, DPPCLK0_DTO_MODULO, 1); } + + dccg->dpp_clock_gated[dpp_inst] = !clock_on; } static const struct dccg_funcs dccg314_funcs = { @@ -357,6 +372,7 @@ static const struct dccg_funcs dccg314_funcs = { .disable_dsc = dccg31_disable_dscclk, .enable_dsc = dccg31_enable_dscclk, .set_pixel_rate_div = dccg314_set_pixel_rate_div, + .trigger_dio_fifo_resync = dccg314_trigger_dio_fifo_resync, .set_valid_pixel_rate = dccg314_set_valid_pixel_rate, }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.h b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.h index 90687a9e8fdddf451fc9b8363ae7dced0065331b..8e07d3151f915dc964545d446eb5cedfea48a523 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.h +++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.h @@ -192,7 +192,10 @@ DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYBSYMCLK_GATE_DISABLE, mask_sh),\ DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYCSYMCLK_GATE_DISABLE, mask_sh),\ DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYDSYMCLK_GATE_DISABLE, mask_sh),\ - DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYESYMCLK_GATE_DISABLE, mask_sh) + DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYESYMCLK_GATE_DISABLE, mask_sh),\ + DCCG_SF(HDMISTREAMCLK0_DTO_PARAM, HDMISTREAMCLK0_DTO_MODULO, mask_sh),\ + DCCG_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_RDIVIDER, mask_sh),\ + DCCG_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, mask_sh) struct dccg *dccg314_create( struct dc_context *ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c index cc3fe9cac5b530f1afbf20889dab11714fae5fb9..7a43f88685006395a9f7f30aabb44414b15d39de 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c @@ -337,14 +337,13 @@ void dcn314_enable_power_gating_plane(struct dce_hwseq *hws, bool enable) REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 0); } -unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div) +void dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div) { struct dc_stream_state *stream = pipe_ctx->stream; - unsigned int odm_combine_factor = 0; bool two_pix_per_container = false; two_pix_per_container = optc2_is_two_pixels_per_containter(&stream->timing); - odm_combine_factor = get_odm_config(pipe_ctx, NULL); + get_odm_config(pipe_ctx, NULL); if (stream->ctx->dc->link_srv->dp_is_128b_132b_signal(pipe_ctx)) { *k1_div = PIXEL_RATE_DIV_BY_1; @@ -362,15 +361,11 @@ unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsig } else { *k1_div = PIXEL_RATE_DIV_BY_1; *k2_div = PIXEL_RATE_DIV_BY_4; - if (odm_combine_factor == 2) - *k2_div = PIXEL_RATE_DIV_BY_2; } } if ((*k1_div == PIXEL_RATE_DIV_NA) && (*k2_div == PIXEL_RATE_DIV_NA)) ASSERT(false); - - return odm_combine_factor; } void dcn314_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx) @@ -390,6 +385,35 @@ void dcn314_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx) pix_per_cycle); } +void dcn314_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc_state *context) +{ + unsigned int i; + struct pipe_ctx *pipe = NULL; + bool otg_disabled[MAX_PIPES] = {false}; + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + pipe = &dc->current_state->res_ctx.pipe_ctx[i]; + + if (pipe->top_pipe || pipe->prev_odm_pipe) + continue; + + if (pipe->stream && (pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal))) { + pipe->stream_res.tg->funcs->disable_crtc(pipe->stream_res.tg); + reset_sync_context_for_pipe(dc, context, i); + otg_disabled[i] = true; + } + } + + hws->ctx->dc->res_pool->dccg->funcs->trigger_dio_fifo_resync(hws->ctx->dc->res_pool->dccg); + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + pipe = &dc->current_state->res_ctx.pipe_ctx[i]; + + if (otg_disabled[i]) + pipe->stream_res.tg->funcs->enable_crtc(pipe->stream_res.tg); + } +} + void dcn314_dpp_root_clock_control(struct dce_hwseq *hws, unsigned int dpp_inst, bool clock_on) { if (!hws->ctx->dc->debug.root_clock_optimization.bits.dpp) @@ -417,9 +441,7 @@ void dcn314_hubp_pg_control(struct dce_hwseq *hws, unsigned int hubp_inst, bool cmd.domain_control.data.inst = hubp_inst; cmd.domain_control.data.power_gate = !power_on; - dc_dmub_srv_cmd_queue(ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(ctx->dmub_srv); - dc_dmub_srv_wait_idle(ctx->dmub_srv); + dm_execute_dmub_cmd(ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); PERF_TRACE(); } diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h index 6d0b62503caa6aeb3e6a8b2f22870fecd31cfcae..96035c75e0dfe20d3d713a6d9f33b916a86d8fc2 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h @@ -37,10 +37,12 @@ void dcn314_dsc_pg_control(struct dce_hwseq *hws, unsigned int dsc_inst, bool po void dcn314_enable_power_gating_plane(struct dce_hwseq *hws, bool enable); -unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div); +void dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div); void dcn314_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx); +void dcn314_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc_state *context); + void dcn314_hubp_pg_control(struct dce_hwseq *hws, unsigned int hubp_inst, bool power_on); void dcn314_dpp_root_clock_control(struct dce_hwseq *hws, unsigned int dpp_inst, bool clock_on); diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c index a588f46b166f4945315c55be21de9db01a150827..86d6a514dec0b42beeda042220636fa5df276c37 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c @@ -60,6 +60,7 @@ static const struct hw_sequencer_funcs dcn314_funcs = { .enable_audio_stream = dce110_enable_audio_stream, .disable_audio_stream = dce110_disable_audio_stream, .disable_plane = dcn20_disable_plane, + .disable_pixel_data = dcn20_disable_pixel_data, .pipe_control_lock = dcn20_pipe_control_lock, .interdependent_update_lock = dcn10_lock_all_pipes, .cursor_lock = dcn10_cursor_lock, @@ -111,7 +112,7 @@ static const struct hw_sequencer_funcs dcn314_funcs = { .set_disp_pattern_generator = dcn30_set_disp_pattern_generator, .optimize_pwr_state = dcn21_optimize_pwr_state, .exit_optimized_pwr_state = dcn21_exit_optimized_pwr_state, - .update_visual_confirm_color = dcn20_update_visual_confirm_color, + .update_visual_confirm_color = dcn10_update_visual_confirm_color, }; static const struct hwseq_private_funcs dcn314_private_funcs = { @@ -151,6 +152,7 @@ static const struct hwseq_private_funcs dcn314_private_funcs = { .setup_hpo_hw_control = dcn31_setup_hpo_hw_control, .calculate_dccg_k1_k2_values = dcn314_calculate_dccg_k1_k2_values, .set_pixels_per_cycle = dcn314_set_pixels_per_cycle, + .resync_fifo_dccg_dio = dcn314_resync_fifo_dccg_dio, }; void dcn314_hw_sequencer_construct(struct dc *dc) @@ -158,8 +160,4 @@ void dcn314_hw_sequencer_construct(struct dc *dc) dc->hwss = dcn314_funcs; dc->hwseq->funcs = dcn314_private_funcs; - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - dc->hwss.init_hw = dcn20_fpga_init_hw; - dc->hwseq->funcs.init_pipes = NULL; - } } diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c index abeeede38fb3912f904a864ec08320e7cadba93e..a840b008d660305436f80fcd4cf0ff3ed73147e9 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c @@ -117,23 +117,6 @@ #define regBIF_BX2_BIOS_SCRATCH_6 0x003e #define regBIF_BX2_BIOS_SCRATCH_6_BASE_IDX 1 -struct IP_BASE_INSTANCE { - unsigned int segment[MAX_SEGMENT]; -}; - -struct IP_BASE { - struct IP_BASE_INSTANCE instance[MAX_INSTANCE]; -}; - -static const struct IP_BASE DCN_BASE = { { { { 0x00000012, 0x000000C0, 0x000034C0, 0x00009000, 0x02403C00, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0, 0, 0 } } } }; - - #define DC_LOGGER_INIT(logger) enum dcn31_clk_src_array_id { @@ -891,8 +874,8 @@ static const struct dc_debug_options debug_defaults_drv = { .force_abm_enable = false, .timing_trace = false, .clock_trace = true, - .disable_dpp_power_gate = true, - .disable_hubp_power_gate = true, + .disable_dpp_power_gate = false, + .disable_hubp_power_gate = false, .disable_pplib_clock_request = false, .pipe_split_policy = MPC_SPLIT_DYNAMIC, .force_single_disp_pipe_split = false, @@ -921,6 +904,22 @@ static const struct dc_debug_options debug_defaults_drv = { .afmt = true, } }, + + .root_clock_optimization = { + .bits = { + .dpp = true, + .dsc = false, + .hdmistream = false, + .hdmichar = false, + .dpstream = false, + .symclk32_se = false, + .symclk32_le = false, + .symclk_fe = false, + .physymclk = false, + .dpiasymclk = false, + } + }, + .seamless_boot_odm_combine = true }; @@ -1375,13 +1374,6 @@ static struct dce_hwseq *dcn314_hwseq_create( hws->regs = &hwseq_reg; hws->shifts = &hwseq_shift; hws->masks = &hwseq_mask; - /* DCN3.1 FPGA Workaround - * Need to enable HPO DP Stream Encoder before setting OTG master enable. - * To do so, move calling function enable_stream_timing to only be done AFTER calling - * function core_link_enable_stream - */ - if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) - hws->wa.dp_hpo_and_otg_sequence = true; } return hws; } @@ -1394,15 +1386,6 @@ static const struct resource_create_funcs res_create_funcs = { .create_hwseq = dcn314_hwseq_create, }; -static const struct resource_create_funcs res_create_maximus_funcs = { - .read_dce_straps = NULL, - .create_audio = NULL, - .create_stream_encoder = NULL, - .create_hpo_dp_stream_encoder = dcn31_hpo_dp_stream_encoder_create, - .create_hpo_dp_link_encoder = dcn31_hpo_dp_link_encoder_create, - .create_hwseq = dcn314_hwseq_create, -}; - static void dcn314_resource_destruct(struct dcn314_resource_pool *pool) { unsigned int i; @@ -1900,6 +1883,13 @@ static bool dcn314_resource_construct( /* Use pipe context based otg sync logic */ dc->config.use_pipe_ctx_sync_logic = true; + /* Disable pipe power gating when unsupported */ + if (ctx->asic_id.hw_internal_rev == 0x01 || + ctx->asic_id.hw_internal_rev == 0x80) { + dc->debug.disable_dpp_power_gate = true; + dc->debug.disable_hubp_power_gate = true; + } + /* read VBIOS LTTPR caps */ { if (ctx->dc_bios->funcs->get_lttpr_caps) { @@ -2101,8 +2091,7 @@ static bool dcn314_resource_construct( /* Audio, Stream Encoders including HPO and virtual, MPC 3D LUTs */ if (!resource_construct(num_virtual_links, dc, &pool->base, - (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ? - &res_create_funcs : &res_create_maximus_funcs))) + &res_create_funcs)) goto create_fail; /* HW Sequencer and Plane caps */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c index 41c972c8eb19816fd4cf496e691750dc741b7259..f1153941907e6a14dd243cc4b17964bed5e8c625 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c @@ -136,6 +136,9 @@ #define DCN3_15_MAX_DET_SIZE 384 #define DCN3_15_CRB_SEGMENT_SIZE_KB 64 +#define DCN3_15_MAX_DET_SEGS (DCN3_15_MAX_DET_SIZE / DCN3_15_CRB_SEGMENT_SIZE_KB) +/* Minimum 2 extra segments need to be in compbuf and claimable to guarantee seamless mpo transitions */ +#define MIN_RESERVED_DET_SEGS 2 enum dcn31_clk_src_array_id { DCN31_CLK_SRC_PLL0, @@ -884,27 +887,10 @@ static const struct dc_debug_options debug_defaults_drv = { .afmt = true, } }, + .enable_legacy_fast_update = true, .psr_power_use_phy_fsm = 0, }; -static const struct dc_debug_options debug_defaults_diags = { - .disable_dmcu = true, - .force_abm_enable = false, - .timing_trace = true, - .clock_trace = true, - .disable_dpp_power_gate = true, - .disable_hubp_power_gate = true, - .disable_clock_gate = true, - .disable_pplib_clock_request = true, - .disable_pplib_wm_range = true, - .disable_stutter = false, - .scl_reset_length10 = true, - .dwb_fi_phase = -1, // -1 = disable - .dmub_command_table = true, - .enable_tri_buf = true, - .use_max_lb = true -}; - static const struct dc_panel_config panel_config_defaults = { .psr = { .disable_psr = false, @@ -1339,13 +1325,6 @@ static struct dce_hwseq *dcn31_hwseq_create( hws->regs = &hwseq_reg; hws->shifts = &hwseq_shift; hws->masks = &hwseq_mask; - /* DCN3.1 FPGA Workaround - * Need to enable HPO DP Stream Encoder before setting OTG master enable. - * To do so, move calling function enable_stream_timing to only be done AFTER calling - * function core_link_enable_stream - */ - if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) - hws->wa.dp_hpo_and_otg_sequence = true; } return hws; } @@ -1358,15 +1337,6 @@ static const struct resource_create_funcs res_create_funcs = { .create_hwseq = dcn31_hwseq_create, }; -static const struct resource_create_funcs res_create_maximus_funcs = { - .read_dce_straps = NULL, - .create_audio = NULL, - .create_stream_encoder = NULL, - .create_hpo_dp_stream_encoder = dcn31_hpo_dp_stream_encoder_create, - .create_hpo_dp_link_encoder = dcn31_hpo_dp_link_encoder_create, - .create_hwseq = dcn31_hwseq_create, -}; - static void dcn315_resource_destruct(struct dcn315_resource_pool *pool) { unsigned int i; @@ -1636,21 +1606,69 @@ static bool is_dual_plane(enum surface_pixel_format format) return format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN || format == SURFACE_PIXEL_FORMAT_GRPH_RGBE_ALPHA; } +static int source_format_to_bpp (enum source_format_class SourcePixelFormat) +{ + if (SourcePixelFormat == dm_444_64) + return 8; + else if (SourcePixelFormat == dm_444_16 || SourcePixelFormat == dm_444_16) + return 2; + else if (SourcePixelFormat == dm_444_8) + return 1; + else if (SourcePixelFormat == dm_rgbe_alpha) + return 5; + else if (SourcePixelFormat == dm_420_8) + return 3; + else if (SourcePixelFormat == dm_420_12) + return 6; + else + return 4; +} + +static bool allow_pixel_rate_crb(struct dc *dc, struct dc_state *context) +{ + int i; + struct resource_context *res_ctx = &context->res_ctx; + + /*Don't apply for single stream*/ + if (context->stream_count < 2) + return false; + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + if (!res_ctx->pipe_ctx[i].stream) + continue; + + /*Don't apply if scaling*/ + if (res_ctx->pipe_ctx[i].stream->src.width != res_ctx->pipe_ctx[i].stream->dst.width || + res_ctx->pipe_ctx[i].stream->src.height != res_ctx->pipe_ctx[i].stream->dst.height || + (res_ctx->pipe_ctx[i].plane_state && (res_ctx->pipe_ctx[i].plane_state->src_rect.width + != res_ctx->pipe_ctx[i].plane_state->dst_rect.width || + res_ctx->pipe_ctx[i].plane_state->src_rect.height + != res_ctx->pipe_ctx[i].plane_state->dst_rect.height))) + return false; + /*Don't apply if MPO to avoid transition issues*/ + if (res_ctx->pipe_ctx[i].top_pipe && res_ctx->pipe_ctx[i].top_pipe->plane_state != res_ctx->pipe_ctx[i].plane_state) + return false; + } + return true; +} + static int dcn315_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, bool fast_validate) { - int i, pipe_cnt; + int i, pipe_cnt, crb_idx, crb_pipes; struct resource_context *res_ctx = &context->res_ctx; struct pipe_ctx *pipe; const int max_usable_det = context->bw_ctx.dml.ip.config_return_buffer_size_in_kbytes - DCN3_15_MIN_COMPBUF_SIZE_KB; + int remaining_det_segs = max_usable_det / DCN3_15_CRB_SEGMENT_SIZE_KB; + bool pixel_rate_crb = allow_pixel_rate_crb(dc, context); DC_FP_START(); dcn31x_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); DC_FP_END(); - for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { + for (i = 0, pipe_cnt = 0, crb_pipes = 0; i < dc->res_pool->pipe_count; i++) { struct dc_crtc_timing *timing; if (!res_ctx->pipe_ctx[i].stream) @@ -1671,6 +1689,28 @@ static int dcn315_populate_dml_pipes_from_context( pipes[pipe_cnt].dout.dsc_input_bpc = 0; DC_FP_START(); dcn31_zero_pipe_dcc_fraction(pipes, pipe_cnt); + if (pixel_rate_crb && !pipe->top_pipe && !pipe->prev_odm_pipe) { + int bpp = source_format_to_bpp(pipes[pipe_cnt].pipe.src.source_format); + /* Ceil to crb segment size */ + int approx_det_segs_required_for_pstate = dcn_get_approx_det_segs_required_for_pstate( + &context->bw_ctx.dml.soc, timing->pix_clk_100hz, bpp, DCN3_15_CRB_SEGMENT_SIZE_KB); + + if (approx_det_segs_required_for_pstate <= 2 * DCN3_15_MAX_DET_SEGS) { + bool split_required = approx_det_segs_required_for_pstate > DCN3_15_MAX_DET_SEGS; + split_required = split_required || timing->pix_clk_100hz >= dcn_get_max_non_odm_pix_rate_100hz(&dc->dml.soc); + split_required = split_required || (pipe->plane_state && pipe->plane_state->src_rect.width > 5120); + + /* Minimum 2 segments to allow mpc/odm combine if its used later */ + if (approx_det_segs_required_for_pstate < 2) + approx_det_segs_required_for_pstate = 2; + if (split_required) + approx_det_segs_required_for_pstate += approx_det_segs_required_for_pstate % 2; + pipes[pipe_cnt].pipe.src.det_size_override = approx_det_segs_required_for_pstate; + remaining_det_segs -= approx_det_segs_required_for_pstate; + } else + remaining_det_segs = -1; + crb_pipes++; + } DC_FP_END(); if (pipes[pipe_cnt].dout.dsc_enable) { @@ -1689,16 +1729,55 @@ static int dcn315_populate_dml_pipes_from_context( break; } } - pipe_cnt++; } + /* Spread remaining unreserved crb evenly among all pipes*/ + if (pixel_rate_crb) { + for (i = 0, pipe_cnt = 0, crb_idx = 0; i < dc->res_pool->pipe_count; i++) { + pipe = &res_ctx->pipe_ctx[i]; + if (!pipe->stream) + continue; + + /* Do not use asymetric crb if not enough for pstate support */ + if (remaining_det_segs < 0) { + pipes[pipe_cnt].pipe.src.det_size_override = 0; + pipe_cnt++; + continue; + } + + if (!pipe->top_pipe && !pipe->prev_odm_pipe) { + bool split_required = pipe->stream->timing.pix_clk_100hz >= dcn_get_max_non_odm_pix_rate_100hz(&dc->dml.soc) + || (pipe->plane_state && pipe->plane_state->src_rect.width > 5120); + + if (remaining_det_segs > MIN_RESERVED_DET_SEGS) + pipes[pipe_cnt].pipe.src.det_size_override += (remaining_det_segs - MIN_RESERVED_DET_SEGS) / crb_pipes + + (crb_idx < (remaining_det_segs - MIN_RESERVED_DET_SEGS) % crb_pipes ? 1 : 0); + if (pipes[pipe_cnt].pipe.src.det_size_override > 2 * DCN3_15_MAX_DET_SEGS) { + /* Clamp to 2 pipe split max det segments */ + remaining_det_segs += pipes[pipe_cnt].pipe.src.det_size_override - 2 * (DCN3_15_MAX_DET_SEGS); + pipes[pipe_cnt].pipe.src.det_size_override = 2 * DCN3_15_MAX_DET_SEGS; + } + if (pipes[pipe_cnt].pipe.src.det_size_override > DCN3_15_MAX_DET_SEGS || split_required) { + /* If we are splitting we must have an even number of segments */ + remaining_det_segs += pipes[pipe_cnt].pipe.src.det_size_override % 2; + pipes[pipe_cnt].pipe.src.det_size_override -= pipes[pipe_cnt].pipe.src.det_size_override % 2; + } + /* Convert segments into size for DML use */ + pipes[pipe_cnt].pipe.src.det_size_override *= DCN3_15_CRB_SEGMENT_SIZE_KB; + + crb_idx++; + } + pipe_cnt++; + } + } + if (pipe_cnt) context->bw_ctx.dml.ip.det_buffer_size_kbytes = (max_usable_det / DCN3_15_CRB_SEGMENT_SIZE_KB / pipe_cnt) * DCN3_15_CRB_SEGMENT_SIZE_KB; if (context->bw_ctx.dml.ip.det_buffer_size_kbytes > DCN3_15_MAX_DET_SIZE) context->bw_ctx.dml.ip.det_buffer_size_kbytes = DCN3_15_MAX_DET_SIZE; - ASSERT(context->bw_ctx.dml.ip.det_buffer_size_kbytes >= DCN3_15_DEFAULT_DET_SIZE); + dc->config.enable_4to1MPC = false; if (pipe_cnt == 1 && pipe->plane_state && !dc->debug.disable_z9_mpc) { if (is_dual_plane(pipe->plane_state->format) @@ -1845,10 +1924,7 @@ static bool dcn315_resource_construct( if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) dc->debug = debug_defaults_drv; - else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) { - dc->debug = debug_defaults_diags; - } else - dc->debug = debug_defaults_diags; + // Init the vm_helper if (dc->vm_helper) vm_helper_init(dc->vm_helper, 16); @@ -2029,9 +2105,8 @@ static bool dcn315_resource_construct( /* Audio, Stream Encoders including HPO and virtual, MPC 3D LUTs */ if (!resource_construct(num_virtual_links, dc, &pool->base, - (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ? - &res_create_funcs : &res_create_maximus_funcs))) - goto create_fail; + &res_create_funcs)) + goto create_fail; /* HW Sequencer and Plane caps */ dcn31_hw_sequencer_construct(dc); diff --git a/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c b/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c index 9ead347a33e93846ba523d520779f1e181aa8289..707cf28bbceb14ff88019f8537be078353f8dc26 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c @@ -884,24 +884,7 @@ static const struct dc_debug_options debug_defaults_drv = { .afmt = true, } }, -}; - -static const struct dc_debug_options debug_defaults_diags = { - .disable_dmcu = true, - .force_abm_enable = false, - .timing_trace = true, - .clock_trace = true, - .disable_dpp_power_gate = true, - .disable_hubp_power_gate = true, - .disable_clock_gate = true, - .disable_pplib_clock_request = true, - .disable_pplib_wm_range = true, - .disable_stutter = false, - .scl_reset_length10 = true, - .dwb_fi_phase = -1, // -1 = disable - .dmub_command_table = true, - .enable_tri_buf = true, - .use_max_lb = true + .enable_legacy_fast_update = true, }; static const struct dc_panel_config panel_config_defaults = { @@ -1340,13 +1323,6 @@ static struct dce_hwseq *dcn31_hwseq_create( hws->regs = &hwseq_reg; hws->shifts = &hwseq_shift; hws->masks = &hwseq_mask; - /* DCN3.1 FPGA Workaround - * Need to enable HPO DP Stream Encoder before setting OTG master enable. - * To do so, move calling function enable_stream_timing to only be done AFTER calling - * function core_link_enable_stream - */ - if (IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) - hws->wa.dp_hpo_and_otg_sequence = true; } return hws; } @@ -1359,15 +1335,6 @@ static const struct resource_create_funcs res_create_funcs = { .create_hwseq = dcn31_hwseq_create, }; -static const struct resource_create_funcs res_create_maximus_funcs = { - .read_dce_straps = NULL, - .create_audio = NULL, - .create_stream_encoder = NULL, - .create_hpo_dp_stream_encoder = dcn31_hpo_dp_stream_encoder_create, - .create_hpo_dp_link_encoder = dcn31_hpo_dp_link_encoder_create, - .create_hwseq = dcn31_hwseq_create, -}; - static void dcn316_resource_destruct(struct dcn316_resource_pool *pool) { unsigned int i; @@ -1844,10 +1811,7 @@ static bool dcn316_resource_construct( if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) dc->debug = debug_defaults_drv; - else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) { - dc->debug = debug_defaults_diags; - } else - dc->debug = debug_defaults_diags; + // Init the vm_helper if (dc->vm_helper) vm_helper_init(dc->vm_helper, 16); @@ -2028,9 +1992,8 @@ static bool dcn316_resource_construct( /* Audio, Stream Encoders including HPO and virtual, MPC 3D LUTs */ if (!resource_construct(num_virtual_links, dc, &pool->base, - (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ? - &res_create_funcs : &res_create_maximus_funcs))) - goto create_fail; + &res_create_funcs)) + goto create_fail; /* HW Sequencer and Plane caps */ dcn31_hw_sequencer_construct(dc); diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c index ffbb739d85b6923aacf49024b2925928709a1149..11e28e056cf7c61e780755e53af2a9bbb82a88d3 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c @@ -42,18 +42,14 @@ #define DC_LOGGER \ dccg->ctx->logger -/* This function is a workaround for writing to OTG_PIXEL_RATE_DIV - * without the probability of causing a DIG FIFO error. - */ -static void dccg32_wait_for_dentist_change_done( +static void dccg32_trigger_dio_fifo_resync( struct dccg *dccg) { struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg); + uint32_t dispclk_rdivider_value = 0; - uint32_t dentist_dispclk_value = REG_READ(DENTIST_DISPCLK_CNTL); - - REG_WRITE(DENTIST_DISPCLK_CNTL, dentist_dispclk_value); - REG_WAIT(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_CHG_DONE, 1, 50, 2000); + REG_GET(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_RDIVIDER, &dispclk_rdivider_value); + REG_UPDATE(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, dispclk_rdivider_value); } static void dccg32_get_pixel_rate_div( @@ -124,29 +120,21 @@ static void dccg32_set_pixel_rate_div( REG_UPDATE_2(OTG_PIXEL_RATE_DIV, OTG0_PIXEL_RATE_DIVK1, k1, OTG0_PIXEL_RATE_DIVK2, k2); - - dccg32_wait_for_dentist_change_done(dccg); break; case 1: REG_UPDATE_2(OTG_PIXEL_RATE_DIV, OTG1_PIXEL_RATE_DIVK1, k1, OTG1_PIXEL_RATE_DIVK2, k2); - - dccg32_wait_for_dentist_change_done(dccg); break; case 2: REG_UPDATE_2(OTG_PIXEL_RATE_DIV, OTG2_PIXEL_RATE_DIVK1, k1, OTG2_PIXEL_RATE_DIVK2, k2); - - dccg32_wait_for_dentist_change_done(dccg); break; case 3: REG_UPDATE_2(OTG_PIXEL_RATE_DIV, OTG3_PIXEL_RATE_DIVK1, k1, OTG3_PIXEL_RATE_DIVK2, k2); - - dccg32_wait_for_dentist_change_done(dccg); break; default: BREAK_TO_DEBUGGER(); @@ -352,6 +340,7 @@ static const struct dccg_funcs dccg32_funcs = { .otg_add_pixel = dccg32_otg_add_pixel, .otg_drop_pixel = dccg32_otg_drop_pixel, .set_pixel_rate_div = dccg32_set_pixel_rate_div, + .trigger_dio_fifo_resync = dccg32_trigger_dio_fifo_resync, }; struct dccg *dccg32_create( diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.h index 8071ab98d7084b4b9a19629785e0dbf4459107bc..cf5508718122248a3c1488f54b418bdbd47e9eba 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.h +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.h @@ -112,8 +112,9 @@ DCCG_SF(DTBCLK_P_CNTL, DTBCLK_P3_EN, mask_sh),\ DCCG_SF(DCCG_AUDIO_DTO_SOURCE, DCCG_AUDIO_DTO_SEL, mask_sh),\ DCCG_SF(DCCG_AUDIO_DTO_SOURCE, DCCG_AUDIO_DTO0_SOURCE_SEL, mask_sh),\ - DCCG_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_CHG_DONE, mask_sh) - + DCCG_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_CHG_DONE, mask_sh),\ + DCCG_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_RDIVIDER, mask_sh),\ + DCCG_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, mask_sh) struct dccg *dccg32_create( struct dc_context *ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c index eb08ccc38e798ce43b17fcac4defda9769664c6f..a18b9c0c5709ca3575e3f8a420b58a1e6adece8a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c @@ -42,8 +42,8 @@ hubbub2->shifts->field_name, hubbub2->masks->field_name /** - * @DCN32_CRB_SEGMENT_SIZE_KB: Maximum Configurable Return Buffer size for - * DCN32 + * DCN32_CRB_SEGMENT_SIZE_KB: Maximum Configurable Return Buffer size for + * DCN32 */ #define DCN32_CRB_SEGMENT_SIZE_KB 64 diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c index 1f5ee5cde6e1c91fcc53d9e238d9e70e81e40c14..c586468872e2a0fc5bbd4d78fb0d954c830bf476 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c @@ -274,8 +274,7 @@ bool dcn32_apply_idle_power_optimizations(struct dc *dc, bool enable) cmd.cab.header.sub_type = DMUB_CMD__CAB_NO_DCN_REQ; cmd.cab.header.payload_bytes = sizeof(cmd.cab) - sizeof(cmd.cab.header); - dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); return true; } @@ -309,8 +308,7 @@ bool dcn32_apply_idle_power_optimizations(struct dc *dc, bool enable) cmd.cab.header.payload_bytes = sizeof(cmd.cab) - sizeof(cmd.cab.header); cmd.cab.cab_alloc_ways = ways; - dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); return true; } @@ -326,9 +324,7 @@ bool dcn32_apply_idle_power_optimizations(struct dc *dc, bool enable) cmd.cab.header.payload_bytes = sizeof(cmd.cab) - sizeof(cmd.cab.header); - dc_dmub_srv_cmd_queue(dc->ctx->dmub_srv, &cmd); - dc_dmub_srv_cmd_execute(dc->ctx->dmub_srv); - dc_dmub_srv_wait_idle(dc->ctx->dmub_srv); + dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT); return true; } @@ -413,6 +409,30 @@ void dcn32_subvp_pipe_control_lock(struct dc *dc, } } +void dcn32_subvp_pipe_control_lock_fast(union block_sequence_params *params) +{ + struct dc *dc = params->subvp_pipe_control_lock_fast_params.dc; + bool lock = params->subvp_pipe_control_lock_fast_params.lock; + struct pipe_ctx *pipe_ctx = params->subvp_pipe_control_lock_fast_params.pipe_ctx; + bool subvp_immediate_flip = false; + + if (pipe_ctx && pipe_ctx->stream && pipe_ctx->plane_state) { + if (pipe_ctx->stream->mall_stream_config.type == SUBVP_MAIN && + pipe_ctx->plane_state->flip_immediate) + subvp_immediate_flip = true; + } + + // Don't need to lock for DRR VSYNC flips -- FW will wait for DRR pending update cleared. + if (subvp_immediate_flip) { + union dmub_inbox0_cmd_lock_hw hw_lock_cmd = { 0 }; + + hw_lock_cmd.bits.command_code = DMUB_INBOX0_CMD__HW_LOCK; + hw_lock_cmd.bits.hw_lock_client = HW_LOCK_CLIENT_DRIVER; + hw_lock_cmd.bits.lock = lock; + hw_lock_cmd.bits.should_release = !lock; + dmub_hw_lock_mgr_inbox0_cmd(dc->ctx->dmub_srv, hw_lock_cmd); + } +} bool dcn32_set_mpc_shaper_3dlut( struct pipe_ctx *pipe_ctx, const struct dc_stream_state *stream) @@ -587,8 +607,8 @@ void dcn32_update_force_pstate(struct dc *dc, struct dc_state *context) struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; struct hubp *hubp = pipe->plane_res.hubp; - if (!pipe->stream || (pipe->stream && !(pipe->stream->mall_stream_config.type == SUBVP_MAIN || - pipe->stream->fpo_in_use))) { + if (!pipe->stream || !(pipe->stream->mall_stream_config.type == SUBVP_MAIN || + pipe->stream->fpo_in_use)) { if (hubp && hubp->funcs->hubp_update_force_pstate_disallow) hubp->funcs->hubp_update_force_pstate_disallow(hubp, false); } @@ -596,7 +616,7 @@ void dcn32_update_force_pstate(struct dc *dc, struct dc_state *context) /* Today only FPO uses cursor P-State force. Only clear cursor P-State force * if it's not FPO. */ - if (!pipe->stream || (pipe->stream && !pipe->stream->fpo_in_use)) { + if (!pipe->stream || !pipe->stream->fpo_in_use) { if (hubp && hubp->funcs->hubp_update_force_cursor_pstate_disallow) hubp->funcs->hubp_update_force_cursor_pstate_disallow(hubp, false); } @@ -721,6 +741,9 @@ static void dcn32_initialize_min_clocks(struct dc *dc) clocks->socclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].socclk_mhz * 1000; clocks->dramclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].memclk_mhz * 1000; clocks->dppclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].dppclk_mhz * 1000; + clocks->ref_dtbclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].dtbclk_mhz * 1000; + clocks->fclk_p_state_change_support = true; + clocks->p_state_change_support = true; if (dc->debug.disable_boot_optimizations) { clocks->dispclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].dispclk_mhz * 1000; } else { @@ -730,9 +753,6 @@ static void dcn32_initialize_min_clocks(struct dc *dc) * freq to ensure that the timing is valid and unchanged. */ clocks->dispclk_khz = dc->clk_mgr->funcs->get_dispclk_from_dentist(dc->clk_mgr); - clocks->ref_dtbclk_khz = dc->clk_mgr->bw_params->clk_table.entries[0].dtbclk_mhz * 1000; - clocks->fclk_p_state_change_support = true; - clocks->p_state_change_support = true; } dc->clk_mgr->funcs->update_clocks( @@ -946,8 +966,10 @@ void dcn32_init_hw(struct dc *dc) // Get DMCUB capabilities if (dc->ctx->dmub_srv) { - dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv->dmub); + dc_dmub_srv_query_caps_cmd(dc->ctx->dmub_srv); dc->caps.dmub_caps.psr = dc->ctx->dmub_srv->dmub->feature_caps.psr; + dc->caps.dmub_caps.subvp_psr = dc->ctx->dmub_srv->dmub->feature_caps.subvp_psr_support; + dc->caps.dmub_caps.gecc_enable = dc->ctx->dmub_srv->dmub->feature_caps.gecc_enable; dc->caps.dmub_caps.mclk_sw = dc->ctx->dmub_srv->dmub->feature_caps.fw_assisted_mclk_switch; } } @@ -1119,10 +1141,9 @@ void dcn32_update_odm(struct dc *dc, struct dc_state *context, struct pipe_ctx * } } -unsigned int dcn32_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div) +void dcn32_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div) { struct dc_stream_state *stream = pipe_ctx->stream; - unsigned int odm_combine_factor = 0; bool two_pix_per_container = false; // For phantom pipes, use the same programming as the main pipes @@ -1130,7 +1151,6 @@ unsigned int dcn32_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsign stream = pipe_ctx->stream->mall_stream_config.paired_stream; } two_pix_per_container = optc2_is_two_pixels_per_containter(&stream->timing); - odm_combine_factor = get_odm_config(pipe_ctx, NULL); if (stream->ctx->dc->link_srv->dp_is_128b_132b_signal(pipe_ctx)) { *k1_div = PIXEL_RATE_DIV_BY_1; @@ -1148,15 +1168,13 @@ unsigned int dcn32_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsign } else { *k1_div = PIXEL_RATE_DIV_BY_1; *k2_div = PIXEL_RATE_DIV_BY_4; - if ((odm_combine_factor == 2) || dcn32_is_dp_dig_pixel_rate_div_policy(pipe_ctx)) + if (dcn32_is_dp_dig_pixel_rate_div_policy(pipe_ctx)) *k2_div = PIXEL_RATE_DIV_BY_2; } } if ((*k1_div == PIXEL_RATE_DIV_NA) && (*k2_div == PIXEL_RATE_DIV_NA)) ASSERT(false); - - return odm_combine_factor; } void dcn32_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx) @@ -1177,6 +1195,36 @@ void dcn32_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx) pix_per_cycle); } +void dcn32_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc_state *context) +{ + unsigned int i; + struct pipe_ctx *pipe = NULL; + bool otg_disabled[MAX_PIPES] = {false}; + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + pipe = &dc->current_state->res_ctx.pipe_ctx[i]; + + if (pipe->top_pipe || pipe->prev_odm_pipe) + continue; + + if (pipe->stream && (pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal)) + && pipe->stream->mall_stream_config.type != SUBVP_PHANTOM) { + pipe->stream_res.tg->funcs->disable_crtc(pipe->stream_res.tg); + reset_sync_context_for_pipe(dc, context, i); + otg_disabled[i] = true; + } + } + + hws->ctx->dc->res_pool->dccg->funcs->trigger_dio_fifo_resync(hws->ctx->dc->res_pool->dccg); + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + pipe = &dc->current_state->res_ctx.pipe_ctx[i]; + + if (otg_disabled[i]) + pipe->stream_res.tg->funcs->enable_crtc(pipe->stream_res.tg); + } +} + void dcn32_unblank_stream(struct pipe_ctx *pipe_ctx, struct dc_link_settings *link_settings) { diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h index 6694c1d14aa31d97db0734e5af9649eb8ea3fbb6..bf9bffabe0c03133eee8e568a3b9aa358b2f5e9a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.h @@ -71,10 +71,12 @@ void dcn32_update_force_pstate(struct dc *dc, struct dc_state *context); void dcn32_update_odm(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe_ctx); -unsigned int dcn32_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div); +void dcn32_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div); void dcn32_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx); +void dcn32_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc_state *context); + void dcn32_subvp_pipe_control_lock(struct dc *dc, struct dc_state *context, bool lock, @@ -82,6 +84,8 @@ void dcn32_subvp_pipe_control_lock(struct dc *dc, struct pipe_ctx *top_pipe_to_program, bool subvp_prev_use); +void dcn32_subvp_pipe_control_lock_fast(union block_sequence_params *params); + void dcn32_unblank_stream(struct pipe_ctx *pipe_ctx, struct dc_link_settings *link_settings); diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c index 8085f2acb1a96f0d2906bba1a24ff30f33038205..c2490e16a66a6c99cfb268226ee9acae54607e68 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c @@ -109,7 +109,8 @@ static const struct hw_sequencer_funcs dcn32_funcs = { .commit_subvp_config = dcn32_commit_subvp_config, .enable_phantom_streams = dcn32_enable_phantom_streams, .subvp_pipe_control_lock = dcn32_subvp_pipe_control_lock, - .update_visual_confirm_color = dcn20_update_visual_confirm_color, + .update_visual_confirm_color = dcn10_update_visual_confirm_color, + .subvp_pipe_control_lock_fast = dcn32_subvp_pipe_control_lock_fast, .update_phantom_vp_position = dcn32_update_phantom_vp_position, .update_dsc_pg = dcn32_update_dsc_pg, .apply_update_flags_for_phantom = dcn32_apply_update_flags_for_phantom, @@ -153,6 +154,7 @@ static const struct hwseq_private_funcs dcn32_private_funcs = { .update_mall_sel = dcn32_update_mall_sel, .calculate_dccg_k1_k2_values = dcn32_calculate_dccg_k1_k2_values, .set_pixels_per_cycle = dcn32_set_pixels_per_cycle, + .resync_fifo_dccg_dio = dcn32_resync_fifo_dccg_dio, .is_dp_dig_pixel_rate_div_policy = dcn32_is_dp_dig_pixel_rate_div_policy, }; @@ -161,8 +163,4 @@ void dcn32_hw_sequencer_init_functions(struct dc *dc) dc->hwss = dcn32_funcs; dc->hwseq->funcs = dcn32_private_funcs; - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - dc->hwss.init_hw = dcn20_fpga_init_hw; - dc->hwseq->funcs.init_pipes = NULL; - } } diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c index 2ee798965bc2b326aa67fc29cb0feec4ecf99c65..8abb94f60078fc637a2e80087d8ac39c9c6c954c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.c @@ -98,7 +98,7 @@ static void optc32_set_odm_combine(struct timing_generator *optc, int *opp_id, i optc1->opp_count = opp_cnt; } -static void optc32_set_h_timing_div_manual_mode(struct timing_generator *optc, bool manual_mode) +void optc32_set_h_timing_div_manual_mode(struct timing_generator *optc, bool manual_mode) { struct optc *optc1 = DCN10TG_FROM_TG(optc); @@ -106,8 +106,11 @@ static void optc32_set_h_timing_div_manual_mode(struct timing_generator *optc, b OTG_H_TIMING_DIV_MODE_MANUAL, manual_mode ? 1 : 0); } /** - * Enable CRTC - * Enable CRTC - call ASIC Control Object to enable Timing generator. + * optc32_enable_crtc() - Enable CRTC - call ASIC Control Object to enable Timing generator. + * + * @optc: timing_generator instance. + * + * Return: If CRTC is enabled, return true. */ static bool optc32_enable_crtc(struct timing_generator *optc) { @@ -245,16 +248,9 @@ static void optc32_set_drr( } optc->funcs->set_vtotal_min_max(optc, params->vertical_total_min - 1, params->vertical_total_max - 1); - optc32_setup_manual_trigger(optc); - } else { - REG_UPDATE_4(OTG_V_TOTAL_CONTROL, - OTG_SET_V_TOTAL_MIN_MASK, 0, - OTG_V_TOTAL_MIN_SEL, 0, - OTG_V_TOTAL_MAX_SEL, 0, - OTG_FORCE_LOCK_ON_EVENT, 0); - - optc->funcs->set_vtotal_min_max(optc, 0, 0); } + + optc32_setup_manual_trigger(optc); } static struct timing_generator_funcs dcn32_tg_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.h index b92ba8c75694034c10c310c6284a9e0c949fb03c..abf0121a1006010e4ab7bc09f9c414201977d970 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_optc.h @@ -179,5 +179,6 @@ SF(OTG0_OTG_DRR_CONTROL, OTG_V_TOTAL_LAST_USED_BY_DRR, mask_sh) void dcn32_timing_generator_init(struct optc *optc1); +void optc32_set_h_timing_div_manual_mode(struct timing_generator *optc, bool manual_mode); #endif /* __DC_OPTC_DCN32_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c index 22dd1ebea618b583ab560511e1fa4a2f6eae2e91..19f134caa8adba9125f2f886c6b6639a982d95a2 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c @@ -726,28 +726,12 @@ static const struct dc_debug_options debug_defaults_drv = { .override_dispclk_programming = true, .disable_fpo_optimizations = false, .fpo_vactive_margin_us = 2000, // 2000us - .disable_fpo_vactive = true, + .disable_fpo_vactive = false, .disable_boot_optimizations = false, -}; - -static const struct dc_debug_options debug_defaults_diags = { - .disable_dmcu = true, - .force_abm_enable = false, - .timing_trace = true, - .clock_trace = true, - .disable_dpp_power_gate = true, - .disable_hubp_power_gate = true, - .disable_dsc_power_gate = true, - .disable_clock_gate = true, - .disable_pplib_clock_request = true, - .disable_pplib_wm_range = true, - .disable_stutter = false, - .scl_reset_length10 = true, - .dwb_fi_phase = -1, // -1 = disable - .dmub_command_table = true, - .enable_tri_buf = true, - .use_max_lb = true, - .force_disable_subvp = true + .disable_subvp_high_refresh = false, + .disable_dp_plus_plus_wa = true, + .fpo_vactive_min_active_margin_us = 200, + .fpo_vactive_max_blank_us = 1000, }; static struct dce_aux *dcn32_aux_engine_create( @@ -1353,15 +1337,6 @@ static const struct resource_create_funcs res_create_funcs = { .create_hwseq = dcn32_hwseq_create, }; -static const struct resource_create_funcs res_create_maximus_funcs = { - .read_dce_straps = NULL, - .create_audio = NULL, - .create_stream_encoder = NULL, - .create_hpo_dp_stream_encoder = dcn32_hpo_dp_stream_encoder_create, - .create_hpo_dp_link_encoder = dcn32_hpo_dp_link_encoder_create, - .create_hwseq = dcn32_hwseq_create, -}; - static void dcn32_resource_destruct(struct dcn32_resource_pool *pool) { unsigned int i; @@ -1888,6 +1863,8 @@ bool dcn32_validate_bandwidth(struct dc *dc, dc->res_pool->funcs->calculate_wm_and_dlg(dc, context, pipes, pipe_cnt, vlevel); + dcn32_override_min_req_memclk(dc, context); + BW_VAL_TRACE_END_WATERMARKS(); goto validate_out; @@ -2198,6 +2175,7 @@ static bool dcn32_resource_construct( dc->caps.extended_aux_timeout_support = true; dc->caps.dmcub_support = true; dc->caps.seamless_odm = true; + dc->caps.max_v_total = (1 << 15) - 1; /* Color pipeline capabilities */ dc->caps.color.dpp.dcn_arch = 1; @@ -2254,10 +2232,7 @@ static bool dcn32_resource_construct( if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) dc->debug = debug_defaults_drv; - else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) { - dc->debug = debug_defaults_diags; - } else - dc->debug = debug_defaults_diags; + // Init the vm_helper if (dc->vm_helper) vm_helper_init(dc->vm_helper, 16); @@ -2313,8 +2288,7 @@ static bool dcn32_resource_construct( } /* DML */ - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) - dml_init_instance(&dc->dml, &dcn3_2_soc, &dcn3_2_ip, DML_PROJECT_DCN32); + dml_init_instance(&dc->dml, &dcn3_2_soc, &dcn3_2_ip, DML_PROJECT_DCN32); /* IRQ Service */ init_data.ctx = dc->ctx; @@ -2451,9 +2425,8 @@ static bool dcn32_resource_construct( /* Audio, HWSeq, Stream Encoders including HPO and virtual, MPC 3D LUTs */ if (!resource_construct(num_virtual_links, dc, &pool->base, - (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ? - &res_create_funcs : &res_create_maximus_funcs))) - goto create_fail; + &res_create_funcs)) + goto create_fail; /* HW Sequencer init functions and Plane caps */ dcn32_hw_sequencer_init_functions(dc); diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h index 3937dbc1e552634c830d0f991d597c4f67070093..81e4431708295d4077a755a2e15e157caee44185 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h @@ -37,9 +37,10 @@ #define DCN3_2_MBLK_WIDTH 128 #define DCN3_2_MBLK_HEIGHT_4BPE 128 #define DCN3_2_MBLK_HEIGHT_8BPE 64 -#define DCN3_2_VMIN_DISPCLK_HZ 717000000 #define DCN3_2_DCFCLK_DS_INIT_KHZ 10000 // Choose 10Mhz for init DCFCLK DS freq -#define DCN3_2_MIN_ACTIVE_SWITCH_MARGIN_FPO_US 100 // Only allow FPO + Vactive if active margin >= 100 +#define SUBVP_HIGH_REFRESH_LIST_LEN 3 +#define DCN3_2_MAX_SUBVP_PIXEL_RATE_MHZ 1800 +#define DCN3_2_VMIN_DISPCLK_HZ 717000000 #define TO_DCN32_RES_POOL(pool)\ container_of(pool, struct dcn32_resource_pool, base) @@ -47,6 +48,15 @@ extern struct _vcs_dpi_ip_params_st dcn3_2_ip; extern struct _vcs_dpi_soc_bounding_box_st dcn3_2_soc; +struct subvp_high_refresh_list { + int min_refresh; + int max_refresh; + struct resolution { + int width; + int height; + } res[SUBVP_HIGH_REFRESH_LIST_LEN]; +}; + struct dcn32_resource_pool { struct resource_pool base; }; @@ -151,10 +161,18 @@ struct dc_stream_state *dcn32_can_support_mclk_switch_using_fw_based_vblank_stre bool dcn32_allow_subvp_with_active_margin(struct pipe_ctx *pipe); +bool dcn32_allow_subvp_high_refresh_rate(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe); + unsigned int dcn32_calc_num_avail_chans_for_mall(struct dc *dc, int num_chans); double dcn32_determine_max_vratio_prefetch(struct dc *dc, struct dc_state *context); +bool dcn32_check_native_scaling_for_res(struct pipe_ctx *pipe, unsigned int width, unsigned int height); + +bool dcn32_subvp_drr_admissable(struct dc *dc, struct dc_state *context); + +bool dcn32_subvp_vblank_admissable(struct dc *dc, struct dc_state *context, int vlevel); + /* definitions for run time init of reg offsets */ /* CLK SRC */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c index eeca16faf31af29b9749dca9b0866c6b671db5f7..a9c41ef0751fbf3f89f8fb49e3ffc59b8f58872c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c @@ -94,18 +94,15 @@ uint32_t dcn32_helper_calculate_mall_bytes_for_cursor( } /** - * ******************************************************************************************** - * dcn32_helper_calculate_num_ways_for_subvp: Calculate number of ways needed for SubVP + * dcn32_helper_calculate_num_ways_for_subvp(): Calculate number of ways needed for SubVP * * Gets total allocation required for the phantom viewport calculated by DML in bytes and * converts to number of cache ways. * - * @param [in] dc: current dc state - * @param [in] context: new dc state + * @dc: current dc state + * @context: new dc state * - * @return: number of ways required for SubVP - * - * ******************************************************************************************** + * Return: number of ways required for SubVP */ uint32_t dcn32_helper_calculate_num_ways_for_subvp( struct dc *dc, @@ -258,11 +255,8 @@ bool dcn32_is_psr_capable(struct pipe_ctx *pipe) return psr_capable; } -#define DCN3_2_NEW_DET_OVERRIDE_MIN_MULTIPLIER 7 - /** - * ******************************************************************************************* - * dcn32_determine_det_override: Determine DET allocation for each pipe + * dcn32_determine_det_override(): Determine DET allocation for each pipe * * This function determines how much DET to allocate for each pipe. The total number of * DET segments will be split equally among each of the streams, and after that the DET @@ -271,6 +265,7 @@ bool dcn32_is_psr_capable(struct pipe_ctx *pipe) * If there is a plane that's driven by more than 1 pipe (i.e. pipe split), then the * number of DET for that given plane will be split among the pipes driving that plane. * + * * High level algorithm: * 1. Split total DET among number of streams * 2. For each stream, split DET among the planes @@ -278,25 +273,11 @@ bool dcn32_is_psr_capable(struct pipe_ctx *pipe) * among those pipes. * 4. Assign the DET override to the DML pipes. * - * Special cases: - * - * For two displays that have a large difference in pixel rate, we may experience - * underflow on the larger display when we divide the DET equally. For this, we - * will implement a modified algorithm to assign more DET to larger display. - * - * 1. Calculate difference in pixel rates ( multiplier ) between two displays - * 2. If the multiplier exceeds DCN3_2_NEW_DET_OVERRIDE_MIN_MULTIPLIER, then - * implement the modified DET override algorithm. - * 3. Assign smaller DET size for lower pixel display and higher DET size for - * higher pixel display + * @dc: Current DC state + * @context: New DC state to be programmed + * @pipes: Array of DML pipes * - * @param [in]: dc: Current DC state - * @param [in]: context: New DC state to be programmed - * @param [in]: pipes: Array of DML pipes - * - * @return: void - * - * ******************************************************************************************* + * Return: void */ void dcn32_determine_det_override(struct dc *dc, struct dc_state *context, @@ -309,31 +290,10 @@ void dcn32_determine_det_override(struct dc *dc, struct dc_plane_state *current_plane = NULL; uint8_t stream_count = 0; - int phy_pix_clk_mult, lower_mode_stream_index; - int phy_pix_clk[MAX_PIPES] = {0}; - bool use_new_det_override_algorithm = false; - for (i = 0; i < context->stream_count; i++) { /* Don't count SubVP streams for DET allocation */ - if (context->streams[i]->mall_stream_config.type != SUBVP_PHANTOM) { - phy_pix_clk[i] = context->streams[i]->phy_pix_clk; + if (context->streams[i]->mall_stream_config.type != SUBVP_PHANTOM) stream_count++; - } - } - - /* Check for special case with two displays, one with much higher pixel rate */ - if (stream_count == 2) { - ASSERT((phy_pix_clk[0] > 0) && (phy_pix_clk[1] > 0)); - if (phy_pix_clk[0] < phy_pix_clk[1]) { - lower_mode_stream_index = 0; - phy_pix_clk_mult = phy_pix_clk[1] / phy_pix_clk[0]; - } else { - lower_mode_stream_index = 1; - phy_pix_clk_mult = phy_pix_clk[0] / phy_pix_clk[1]; - } - - if (phy_pix_clk_mult >= DCN3_2_NEW_DET_OVERRIDE_MIN_MULTIPLIER) - use_new_det_override_algorithm = true; } if (stream_count > 0) { @@ -342,13 +302,6 @@ void dcn32_determine_det_override(struct dc *dc, if (context->streams[i]->mall_stream_config.type == SUBVP_PHANTOM) continue; - if (use_new_det_override_algorithm) { - if (i == lower_mode_stream_index) - stream_segments = 4; - else - stream_segments = 14; - } - if (context->stream_status[i].plane_count > 0) plane_segments = stream_segments / context->stream_status[i].plane_count; else @@ -432,8 +385,7 @@ void dcn32_set_det_allocations(struct dc *dc, struct dc_state *context, } /** - * ******************************************************************************************* - * dcn32_save_mall_state: Save MALL (SubVP) state for fast validation cases + * dcn32_save_mall_state(): Save MALL (SubVP) state for fast validation cases * * This function saves the MALL (SubVP) case for fast validation cases. For fast validation, * there are situations where a shallow copy of the dc->current_state is created for the @@ -446,13 +398,11 @@ void dcn32_set_det_allocations(struct dc *dc, struct dc_state *context, * NOTE: This function ONLY works if the streams are not moved to a different pipe in the * validation. We don't expect this to happen in fast_validation=1 cases. * - * @param [in]: dc: Current DC state - * @param [in]: context: New DC state to be programmed - * @param [out]: temp_config: struct used to cache the existing MALL state - * - * @return: void + * @dc: Current DC state + * @context: New DC state to be programmed + * @temp_config: struct used to cache the existing MALL state * - * ******************************************************************************************* + * Return: void */ void dcn32_save_mall_state(struct dc *dc, struct dc_state *context, @@ -472,18 +422,15 @@ void dcn32_save_mall_state(struct dc *dc, } /** - * ******************************************************************************************* - * dcn32_restore_mall_state: Restore MALL (SubVP) state for fast validation cases + * dcn32_restore_mall_state(): Restore MALL (SubVP) state for fast validation cases * * Restore the MALL state based on the previously saved state from dcn32_save_mall_state * - * @param [in]: dc: Current DC state - * @param [in/out]: context: New DC state to be programmed, restore MALL state into here - * @param [in]: temp_config: struct that has the cached MALL state + * @dc: Current DC state + * @context: New DC state to be programmed, restore MALL state into here + * @temp_config: struct that has the cached MALL state * - * @return: void - * - * ******************************************************************************************* + * Return: void */ void dcn32_restore_mall_state(struct dc *dc, struct dc_state *context, @@ -588,10 +535,11 @@ static int get_refresh_rate(struct dc_stream_state *fpo_candidate_stream) } /** - * dcn32_can_support_mclk_switch_using_fw_based_vblank_stretch - Determines if config can support FPO + * dcn32_can_support_mclk_switch_using_fw_based_vblank_stretch() - Determines if config can + * support FPO * - * @param [in]: dc - current dc state - * @param [in]: context - new dc state + * @dc: current dc state + * @context: new dc state * * Return: Pointer to FPO stream candidate if config can support FPO, otherwise NULL */ @@ -626,7 +574,7 @@ struct dc_stream_state *dcn32_can_support_mclk_switch_using_fw_based_vblank_stre DC_FP_END(); DC_FP_START(); - is_fpo_vactive = dcn32_find_vactive_pipe(dc, context, DCN3_2_MIN_ACTIVE_SWITCH_MARGIN_FPO_US); + is_fpo_vactive = dcn32_find_vactive_pipe(dc, context, dc->debug.fpo_vactive_min_active_margin_us); DC_FP_END(); if (!is_fpo_vactive || dc->debug.disable_fpo_vactive) return NULL; @@ -656,3 +604,120 @@ struct dc_stream_state *dcn32_can_support_mclk_switch_using_fw_based_vblank_stre return fpo_candidate_stream; } + +bool dcn32_check_native_scaling_for_res(struct pipe_ctx *pipe, unsigned int width, unsigned int height) +{ + bool is_native_scaling = false; + + if (pipe->stream->timing.h_addressable == width && + pipe->stream->timing.v_addressable == height && + pipe->plane_state->src_rect.width == width && + pipe->plane_state->src_rect.height == height && + pipe->plane_state->dst_rect.width == width && + pipe->plane_state->dst_rect.height == height) + is_native_scaling = true; + + return is_native_scaling; +} + +/** + * dcn32_subvp_drr_admissable() - Determine if SubVP + DRR config is admissible + * + * @dc: Current DC state + * @context: New DC state to be programmed + * + * SubVP + DRR is admissible under the following conditions: + * - Config must have 2 displays (i.e., 2 non-phantom master pipes) + * - One display is SubVP + * - Other display must have Freesync enabled + * - The potential DRR display must not be PSR capable + * + * Return: True if admissible, false otherwise + */ +bool dcn32_subvp_drr_admissable(struct dc *dc, struct dc_state *context) +{ + bool result = false; + uint32_t i; + uint8_t subvp_count = 0; + uint8_t non_subvp_pipes = 0; + bool drr_pipe_found = false; + bool drr_psr_capable = false; + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + + if (!pipe->stream) + continue; + + if (pipe->plane_state && !pipe->top_pipe) { + if (pipe->stream->mall_stream_config.type == SUBVP_MAIN) + subvp_count++; + if (pipe->stream->mall_stream_config.type == SUBVP_NONE) { + non_subvp_pipes++; + drr_psr_capable = (drr_psr_capable || dcn32_is_psr_capable(pipe)); + if (pipe->stream->ignore_msa_timing_param && + (pipe->stream->allow_freesync || pipe->stream->vrr_active_variable)) { + drr_pipe_found = true; + } + } + } + } + + if (subvp_count == 1 && non_subvp_pipes == 1 && drr_pipe_found && !drr_psr_capable) + result = true; + + return result; +} + +/** + * dcn32_subvp_vblank_admissable() - Determine if SubVP + Vblank config is admissible + * + * @dc: Current DC state + * @context: New DC state to be programmed + * @vlevel: Voltage level calculated by DML + * + * SubVP + Vblank is admissible under the following conditions: + * - Config must have 2 displays (i.e., 2 non-phantom master pipes) + * - One display is SubVP + * - Other display must not have Freesync capability + * - DML must have output DRAM clock change support as SubVP + Vblank + * - The potential vblank display must not be PSR capable + * + * Return: True if admissible, false otherwise + */ +bool dcn32_subvp_vblank_admissable(struct dc *dc, struct dc_state *context, int vlevel) +{ + bool result = false; + uint32_t i; + uint8_t subvp_count = 0; + uint8_t non_subvp_pipes = 0; + bool drr_pipe_found = false; + struct vba_vars_st *vba = &context->bw_ctx.dml.vba; + bool vblank_psr_capable = false; + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + + if (!pipe->stream) + continue; + + if (pipe->plane_state && !pipe->top_pipe) { + if (pipe->stream->mall_stream_config.type == SUBVP_MAIN) + subvp_count++; + if (pipe->stream->mall_stream_config.type == SUBVP_NONE) { + non_subvp_pipes++; + vblank_psr_capable = (vblank_psr_capable || dcn32_is_psr_capable(pipe)); + if (pipe->stream->ignore_msa_timing_param && + (pipe->stream->allow_freesync || pipe->stream->vrr_active_variable)) { + drr_pipe_found = true; + } + } + } + } + + if (subvp_count == 1 && non_subvp_pipes == 1 && !drr_pipe_found && !vblank_psr_capable && + vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_vblank_w_mall_sub_vp) + result = true; + + return result; +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c index a60ddb343d13b53739b93e640e600cdf1612b9f8..ea204742ad35455baa093e38d6fd7945aaa16c9c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c @@ -725,31 +725,13 @@ static const struct dc_debug_options debug_defaults_drv = { .override_dispclk_programming = true, .disable_fpo_optimizations = false, .fpo_vactive_margin_us = 2000, // 2000us - .disable_fpo_vactive = true, + .disable_fpo_vactive = false, .disable_boot_optimizations = false, + .disable_subvp_high_refresh = false, + .fpo_vactive_min_active_margin_us = 200, + .fpo_vactive_max_blank_us = 1000, }; -static const struct dc_debug_options debug_defaults_diags = { - .disable_dmcu = true, - .force_abm_enable = false, - .timing_trace = true, - .clock_trace = true, - .disable_dpp_power_gate = true, - .disable_hubp_power_gate = true, - .disable_dsc_power_gate = true, - .disable_clock_gate = true, - .disable_pplib_clock_request = true, - .disable_pplib_wm_range = true, - .disable_stutter = false, - .scl_reset_length10 = true, - .dwb_fi_phase = -1, // -1 = disable - .dmub_command_table = true, - .enable_tri_buf = true, - .use_max_lb = true, - .force_disable_subvp = true, -}; - - static struct dce_aux *dcn321_aux_engine_create( struct dc_context *ctx, uint32_t inst) @@ -1340,15 +1322,6 @@ static const struct resource_create_funcs res_create_funcs = { .create_hwseq = dcn321_hwseq_create, }; -static const struct resource_create_funcs res_create_maximus_funcs = { - .read_dce_straps = NULL, - .create_audio = NULL, - .create_stream_encoder = NULL, - .create_hpo_dp_stream_encoder = dcn321_hpo_dp_stream_encoder_create, - .create_hpo_dp_link_encoder = dcn321_hpo_dp_link_encoder_create, - .create_hwseq = dcn321_hwseq_create, -}; - static void dcn321_resource_destruct(struct dcn321_resource_pool *pool) { unsigned int i; @@ -1735,9 +1708,9 @@ static bool dcn321_resource_construct( dc->caps.subvp_pstate_allow_width_us = 20; dc->caps.subvp_vertical_int_margin_us = 30; dc->caps.subvp_drr_vblank_start_margin_us = 100; // 100us margin - dc->caps.max_slave_planes = 1; - dc->caps.max_slave_yuv_planes = 1; - dc->caps.max_slave_rgb_planes = 1; + dc->caps.max_slave_planes = 2; + dc->caps.max_slave_yuv_planes = 2; + dc->caps.max_slave_rgb_planes = 2; dc->caps.post_blend_color_processing = true; dc->caps.force_dp_tps4_for_cp2520 = true; dc->caps.dp_hpo = true; @@ -1745,6 +1718,7 @@ static bool dcn321_resource_construct( dc->caps.edp_dsc_support = true; dc->caps.extended_aux_timeout_support = true; dc->caps.dmcub_support = true; + dc->caps.max_v_total = (1 << 15) - 1; /* Color pipeline capabilities */ dc->caps.color.dpp.dcn_arch = 1; @@ -1798,10 +1772,7 @@ static bool dcn321_resource_construct( if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) dc->debug = debug_defaults_drv; - else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) { - dc->debug = debug_defaults_diags; - } else - dc->debug = debug_defaults_diags; + // Init the vm_helper if (dc->vm_helper) vm_helper_init(dc->vm_helper, 16); @@ -1857,8 +1828,7 @@ static bool dcn321_resource_construct( } /* DML */ - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) - dml_init_instance(&dc->dml, &dcn3_21_soc, &dcn3_21_ip, DML_PROJECT_DCN32); + dml_init_instance(&dc->dml, &dcn3_21_soc, &dcn3_21_ip, DML_PROJECT_DCN32); /* IRQ Service */ init_data.ctx = dc->ctx; @@ -1990,9 +1960,8 @@ static bool dcn321_resource_construct( /* Audio, HWSeq, Stream Encoders including HPO and virtual, MPC 3D LUTs */ if (!resource_construct(num_virtual_links, dc, &pool->base, - (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) ? - &res_create_funcs : &res_create_maximus_funcs))) - goto create_fail; + &res_create_funcs)) + goto create_fail; /* HW Sequencer init functions and Plane caps */ dcn32_hw_sequencer_init_functions(dc); diff --git a/drivers/gpu/drm/amd/display/dc/dm_services.h b/drivers/gpu/drm/amd/display/dc/dm_services.h index 9a3f2a44f882458cf35acaa779fa83c22295c2cd..d0eed3b4771e6f1a2e4097db8bb145894c9a0c16 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_services.h +++ b/drivers/gpu/drm/amd/display/dc/dm_services.h @@ -40,6 +40,7 @@ struct dmub_srv; struct dc_dmub_srv; +union dmub_rb_cmd; irq_handler_idx dm_register_interrupt( struct dc_context *ctx, @@ -273,6 +274,12 @@ void dm_perf_trace_timestamp(const char *func_name, unsigned int line, struct dc #define PERF_TRACE() dm_perf_trace_timestamp(__func__, __LINE__, CTX) #define PERF_TRACE_CTX(__CTX) dm_perf_trace_timestamp(__func__, __LINE__, __CTX) +/* + * DMUB Interfaces + */ +bool dm_execute_dmub_cmd(const struct dc_context *ctx, union dmub_rb_cmd *cmd, enum dm_dmub_wait_type wait_type); +bool dm_execute_dmub_cmd_list(const struct dc_context *ctx, unsigned int count, union dmub_rb_cmd *cmd, enum dm_dmub_wait_type wait_type); + /* * Debug and verification hooks */ @@ -285,4 +292,6 @@ void dm_dtn_log_append_v(struct dc_context *ctx, void dm_dtn_log_end(struct dc_context *ctx, struct dc_log_buffer_ctx *log_ctx); +char *dce_version_to_string(const int version); + #endif /* __DM_SERVICES_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dm_services_types.h b/drivers/gpu/drm/amd/display/dc/dm_services_types.h index b52ba6ffabe1c6c2a7309073d5b2560c10e4d6e6..facf269c4326dc55d25b8d44716322bd6ddc07ed 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_services_types.h +++ b/drivers/gpu/drm/amd/display/dc/dm_services_types.h @@ -269,4 +269,10 @@ struct dtn_min_clk_info { uint32_t min_memory_clock_khz; }; +enum dm_dmub_wait_type { + DM_DMUB_WAIT_TYPE_NO_WAIT, + DM_DMUB_WAIT_TYPE_WAIT, + DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY, +}; + #endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile index 01db035589c59674a9c3458365727b0a2bbb7c88..77cf5545c94cc8e827ada222c844ba530a6ea0cf 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile @@ -38,6 +38,11 @@ ifdef CONFIG_ARM64 dml_rcflags := -mgeneral-regs-only endif +ifdef CONFIG_LOONGARCH +dml_ccflags := -mfpu=64 +dml_rcflags := -msoft-float +endif + ifdef CONFIG_CC_IS_GCC ifneq ($(call gcc-min-version, 70100),y) IS_OLD_GCC = 1 diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c index f1c1a4b5fcac33532b8eb0911d344a2b13051d10..8ae5ddbd1b271b6148a30a3870a28d26f8b3f87b 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c @@ -948,10 +948,10 @@ static enum dcn_zstate_support_state decide_zstate_support(struct dc *dc, struc { int plane_count; int i; - unsigned int optimized_min_dst_y_next_start_us; + unsigned int min_dst_y_next_start_us; plane_count = 0; - optimized_min_dst_y_next_start_us = 0; + min_dst_y_next_start_us = 0; for (i = 0; i < dc->res_pool->pipe_count; i++) { if (context->res_ctx.pipe_ctx[i].plane_state) plane_count++; @@ -973,19 +973,18 @@ static enum dcn_zstate_support_state decide_zstate_support(struct dc *dc, struc else if (context->stream_count == 1 && context->streams[0]->signal == SIGNAL_TYPE_EDP) { struct dc_link *link = context->streams[0]->sink->link; struct dc_stream_status *stream_status = &context->stream_status[0]; + struct dc_stream_state *current_stream = context->streams[0]; int minmum_z8_residency = dc->debug.minimum_z8_residency_time > 0 ? dc->debug.minimum_z8_residency_time : 1000; bool allow_z8 = context->bw_ctx.dml.vba.StutterPeriod > (double)minmum_z8_residency; bool is_pwrseq0 = link->link_index == 0; + bool isFreesyncVideo; - if (dc_extended_blank_supported(dc)) { - for (i = 0; i < dc->res_pool->pipe_count; i++) { - if (context->res_ctx.pipe_ctx[i].stream == context->streams[0] - && context->res_ctx.pipe_ctx[i].stream->adjust.v_total_min == context->res_ctx.pipe_ctx[i].stream->adjust.v_total_max - && context->res_ctx.pipe_ctx[i].stream->adjust.v_total_min > context->res_ctx.pipe_ctx[i].stream->timing.v_total) { - optimized_min_dst_y_next_start_us = - context->res_ctx.pipe_ctx[i].dlg_regs.optimized_min_dst_y_next_start_us; - break; - } + isFreesyncVideo = current_stream->adjust.v_total_min == current_stream->adjust.v_total_max; + isFreesyncVideo = isFreesyncVideo && current_stream->timing.v_total < current_stream->adjust.v_total_min; + for (i = 0; i < dc->res_pool->pipe_count; i++) { + if (context->res_ctx.pipe_ctx[i].stream == current_stream && isFreesyncVideo) { + min_dst_y_next_start_us = context->res_ctx.pipe_ctx[i].dlg_regs.min_dst_y_next_start_us; + break; } } @@ -993,7 +992,7 @@ static enum dcn_zstate_support_state decide_zstate_support(struct dc *dc, struc if (stream_status->plane_count > 1) return DCN_ZSTATE_SUPPORT_DISALLOW; - if (is_pwrseq0 && (context->bw_ctx.dml.vba.StutterPeriod > 5000.0 || optimized_min_dst_y_next_start_us > 5000)) + if (is_pwrseq0 && (context->bw_ctx.dml.vba.StutterPeriod > 5000.0 || min_dst_y_next_start_us > 5000)) return DCN_ZSTATE_SUPPORT_ALLOW; else if (is_pwrseq0 && link->psr_settings.psr_version == DC_PSR_VERSION_1 && !link->panel_config.psr.disable_psr) return allow_z8 ? DCN_ZSTATE_SUPPORT_ALLOW_Z8_Z10_ONLY : DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY; @@ -1043,7 +1042,7 @@ void dcn20_calculate_dlg_params(struct dc *dc, int pipe_cnt, int vlevel) { - int i, pipe_idx; + int i, pipe_idx, active_hubp_count = 0; dc_assert_fp_enabled(); @@ -1079,6 +1078,8 @@ void dcn20_calculate_dlg_params(struct dc *dc, for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { if (!context->res_ctx.pipe_ctx[i].stream) continue; + if (context->res_ctx.pipe_ctx[i].plane_state) + active_hubp_count++; pipes[pipe_idx].pipe.dest.vstartup_start = get_vstartup(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); pipes[pipe_idx].pipe.dest.vupdate_offset = get_vupdate_offset(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); pipes[pipe_idx].pipe.dest.vupdate_width = get_vupdate_width(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); @@ -1098,13 +1099,13 @@ void dcn20_calculate_dlg_params(struct dc *dc, context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz = pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000; context->res_ctx.pipe_ctx[i].pipe_dlg_param = pipes[pipe_idx].pipe.dest; - if (context->res_ctx.pipe_ctx[i].stream->adaptive_sync_infopacket.valid) - dcn20_adjust_freesync_v_startup( - &context->res_ctx.pipe_ctx[i].stream->timing, - &context->res_ctx.pipe_ctx[i].pipe_dlg_param.vstartup_start); pipe_idx++; } + /* If DCN isn't making memory requests we can allow pstate change */ + if (!active_hubp_count) { + context->bw_ctx.bw.dcn.clk.p_state_change_support = true; + } /*save a original dppclock copy*/ context->bw_ctx.bw.dcn.clk.bw_dppclk_khz = context->bw_ctx.bw.dcn.clk.dppclk_khz; context->bw_ctx.bw.dcn.clk.bw_dispclk_khz = context->bw_ctx.bw.dcn.clk.dispclk_khz; @@ -1885,6 +1886,17 @@ void dcn20_patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; } + if ((int)(bb->sr_exit_z8_time_us * 1000) + != dc->bb_overrides.sr_exit_z8_time_ns + && dc->bb_overrides.sr_exit_z8_time_ns) { + bb->sr_exit_z8_time_us = dc->bb_overrides.sr_exit_z8_time_ns / 1000.0; + } + + if ((int)(bb->sr_enter_plus_exit_z8_time_us * 1000) + != dc->bb_overrides.sr_enter_plus_exit_z8_time_ns + && dc->bb_overrides.sr_enter_plus_exit_z8_time_ns) { + bb->sr_enter_plus_exit_z8_time_us = dc->bb_overrides.sr_enter_plus_exit_z8_time_ns / 1000.0; + } if ((int)(bb->urgent_latency_us * 1000) != dc->bb_overrides.urgent_latency_ns && dc->bb_overrides.urgent_latency_ns) { bb->urgent_latency_us = dc->bb_overrides.urgent_latency_ns / 1000.0; @@ -1915,6 +1927,7 @@ static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *co int vlevel = 0; int pipe_split_from[MAX_PIPES]; int pipe_cnt = 0; + int i = 0; display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_ATOMIC); DC_LOGGER_INIT(dc->ctx->logger); @@ -1938,6 +1951,15 @@ static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *co dcn20_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, fast_validate); dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); + for (i = 0; i < dc->res_pool->pipe_count; i++) { + if (!context->res_ctx.pipe_ctx[i].stream) + continue; + if (context->res_ctx.pipe_ctx[i].stream->adaptive_sync_infopacket.valid) + dcn20_adjust_freesync_v_startup( + &context->res_ctx.pipe_ctx[i].stream->timing, + &context->res_ctx.pipe_ctx[i].pipe_dlg_param.vstartup_start); + } + BW_VAL_TRACE_END_WATERMARKS(); goto validate_out; @@ -2210,6 +2232,7 @@ bool dcn21_validate_bandwidth_fp(struct dc *dc, int vlevel = 0; int pipe_split_from[MAX_PIPES]; int pipe_cnt = 0; + int i = 0; display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_ATOMIC); DC_LOGGER_INIT(dc->ctx->logger); @@ -2238,6 +2261,15 @@ bool dcn21_validate_bandwidth_fp(struct dc *dc, dcn21_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, fast_validate); dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); + for (i = 0; i < dc->res_pool->pipe_count; i++) { + if (!context->res_ctx.pipe_ctx[i].stream) + continue; + if (context->res_ctx.pipe_ctx[i].stream->adaptive_sync_infopacket.valid) + dcn20_adjust_freesync_v_startup( + &context->res_ctx.pipe_ctx[i].stream->timing, + &context->res_ctx.pipe_ctx[i].pipe_dlg_param.vstartup_start); + } + BW_VAL_TRACE_END_WATERMARKS(); goto validate_out; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c index b7c2844d0cbee29dfc4e7a4230008723fd48ba43..f294f2f8c75bcf6dd102ab6ac199b96e5a9a0c10 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c @@ -810,7 +810,7 @@ static bool CalculatePrefetchSchedule( *swath_width_chroma_ub = dml_ceil(SwathWidthY / 2 - 1, myPipe->BlockWidth256BytesC) + myPipe->BlockWidth256BytesC; } else { *swath_width_luma_ub = dml_ceil(SwathWidthY - 1, myPipe->BlockHeight256BytesY) + myPipe->BlockHeight256BytesY; - if (myPipe->BlockWidth256BytesC > 0) + if (myPipe->BlockHeight256BytesC > 0) *swath_width_chroma_ub = dml_ceil(SwathWidthY / 2 - 1, myPipe->BlockHeight256BytesC) + myPipe->BlockHeight256BytesC; } diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c index a352c703e2589cd4eaa9e52f4715cc5863c4e7f0..ccb4ad78f6670657df8df469fd3a449ba1ea89b4 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c @@ -674,10 +674,19 @@ void dcn30_fpu_update_bw_bounding_box(struct dc *dc, } /** - * Finds dummy_latency_index when MCLK switching using firmware based - * vblank stretch is enabled. This function will iterate through the - * table of dummy pstate latencies until the lowest value that allows + * dcn30_find_dummy_latency_index_for_fw_based_mclk_switch() - Finds + * dummy_latency_index when MCLK switching using firmware based vblank stretch + * is enabled. This function will iterate through the table of dummy pstate + * latencies until the lowest value that allows * dm_allow_self_refresh_and_mclk_switch to happen is found + * + * @dc: Current DC state + * @context: new dc state + * @pipes: DML pipe params + * @pipe_cnt: number of DML pipes + * @vlevel: Voltage level calculated by DML + * + * Return: lowest dummy_latency_index value */ int dcn30_find_dummy_latency_index_for_fw_based_mclk_switch(struct dc *dc, struct dc_state *context, diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c index 7d0626e42ea68995edff1a9c3ee45fa4a4b71acd..9af1a43c042bc8dce9cbaa08906266c7abfda778 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c @@ -4939,8 +4939,8 @@ void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l } v->TotImmediateFlipBytes = 0.0; for (k = 0; k <= v->NumberOfActivePlanes - 1; k++) { - v->TotImmediateFlipBytes = v->TotImmediateFlipBytes + v->NoOfDPP[i][j][k] * v->PDEAndMetaPTEBytesPerFrame[i][j][k] - + v->MetaRowBytes[i][j][k] + v->DPTEBytesPerRow[i][j][k]; + v->TotImmediateFlipBytes = v->TotImmediateFlipBytes + v->NoOfDPP[i][j][k] * (v->PDEAndMetaPTEBytesPerFrame[i][j][k] + + v->MetaRowBytes[i][j][k] + v->DPTEBytesPerRow[i][j][k]); } for (k = 0; k <= v->NumberOfActivePlanes - 1; k++) { @@ -5130,7 +5130,7 @@ void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l ViewportExceedsSurface = true; if (v->SourcePixelFormat[k] != dm_444_64 && v->SourcePixelFormat[k] != dm_444_32 && v->SourcePixelFormat[k] != dm_444_16 - && v->SourcePixelFormat[k] != dm_444_8 && v->SourcePixelFormat[k] != dm_rgbe) { + && v->SourcePixelFormat[k] != dm_444_16 && v->SourcePixelFormat[k] != dm_444_8 && v->SourcePixelFormat[k] != dm_rgbe) { if (v->ViewportWidthChroma[k] > v->SurfaceWidthC[k] || v->ViewportHeightChroma[k] > v->SurfaceHeightC[k]) { ViewportExceedsSurface = true; } diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c index cd3cfcb2a2b01b44697503d911671bc493662ccf..0497a5d74a62f67492e76bec43ad282e7c863193 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c @@ -980,7 +980,7 @@ static void dml_rq_dlg_get_dlg_params(struct display_mode_lib *mode_lib, unsigned int vstartup_start = 0; unsigned int dst_x_after_scaler = 0; - unsigned int dst_y_after_scaler = 0; + int dst_y_after_scaler = 0; double line_wait = 0; double dst_y_prefetch = 0; double dst_y_per_vm_vblank = 0; @@ -1171,6 +1171,8 @@ static void dml_rq_dlg_get_dlg_params(struct display_mode_lib *mode_lib, dst_x_after_scaler = get_dst_x_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); dst_y_after_scaler = get_dst_y_after_scaler(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); + if (dst_y_after_scaler < 0) + dst_y_after_scaler = 0; // do some adjustment on the dst_after scaler to account for odm combine mode dml_print("DML_DLG: %s: input dst_x_after_scaler = %d\n", diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c index 422f17aefd4a6cc4695468d84781a15c30911fd6..6ce90678b33c0368ecf334d4d98e5e78f8e846ff 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c @@ -333,45 +333,43 @@ void dcn301_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_param memcpy(s, dcn3_01_soc.clock_limits, sizeof(dcn3_01_soc.clock_limits)); /* Default clock levels are used for diags, which may lead to overclocking. */ - if (!IS_DIAG_DC(dc->ctx->dce_environment)) { - dcn3_01_ip.max_num_otg = pool->base.res_cap->num_timing_generator; - dcn3_01_ip.max_num_dpp = pool->base.pipe_count; - dcn3_01_soc.num_chans = bw_params->num_channels; - - ASSERT(clk_table->num_entries); - for (i = 0; i < clk_table->num_entries; i++) { - /* loop backwards*/ - for (closest_clk_lvl = 0, j = dcn3_01_soc.num_states - 1; j >= 0; j--) { - if ((unsigned int) dcn3_01_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) { - closest_clk_lvl = j; - break; - } + dcn3_01_ip.max_num_otg = pool->base.res_cap->num_timing_generator; + dcn3_01_ip.max_num_dpp = pool->base.pipe_count; + dcn3_01_soc.num_chans = bw_params->num_channels; + + ASSERT(clk_table->num_entries); + for (i = 0; i < clk_table->num_entries; i++) { + /* loop backwards*/ + for (closest_clk_lvl = 0, j = dcn3_01_soc.num_states - 1; j >= 0; j--) { + if ((unsigned int) dcn3_01_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) { + closest_clk_lvl = j; + break; } - - s[i].state = i; - s[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz; - s[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz; - s[i].socclk_mhz = clk_table->entries[i].socclk_mhz; - s[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2; - - s[i].dispclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dispclk_mhz; - s[i].dppclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dppclk_mhz; - s[i].dram_bw_per_chan_gbps = - dcn3_01_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps; - s[i].dscclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dscclk_mhz; - s[i].dtbclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dtbclk_mhz; - s[i].phyclk_d18_mhz = - dcn3_01_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz; - s[i].phyclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].phyclk_mhz; } - if (clk_table->num_entries) { - dcn3_01_soc.num_states = clk_table->num_entries; - /* duplicate last level */ - s[dcn3_01_soc.num_states] = - dcn3_01_soc.clock_limits[dcn3_01_soc.num_states - 1]; - s[dcn3_01_soc.num_states].state = dcn3_01_soc.num_states; - } + s[i].state = i; + s[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz; + s[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz; + s[i].socclk_mhz = clk_table->entries[i].socclk_mhz; + s[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2; + + s[i].dispclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dispclk_mhz; + s[i].dppclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dppclk_mhz; + s[i].dram_bw_per_chan_gbps = + dcn3_01_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps; + s[i].dscclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dscclk_mhz; + s[i].dtbclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dtbclk_mhz; + s[i].phyclk_d18_mhz = + dcn3_01_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz; + s[i].phyclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].phyclk_mhz; + } + + if (clk_table->num_entries) { + dcn3_01_soc.num_states = clk_table->num_entries; + /* duplicate last level */ + s[dcn3_01_soc.num_states] = + dcn3_01_soc.clock_limits[dcn3_01_soc.num_states - 1]; + s[dcn3_01_soc.num_states].state = dcn3_01_soc.num_states; } memcpy(dcn3_01_soc.clock_limits, s, sizeof(dcn3_01_soc.clock_limits)); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c index 59836570603ac6e3aef195b9d62aa8eea92f3b50..deb6d162a2d5c00df00b069a57e885b34a18e939 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.c @@ -483,7 +483,7 @@ void dcn31_calculate_wm_and_dlg_fp( int pipe_cnt, int vlevel) { - int i, pipe_idx, active_hubp_count = 0; + int i, pipe_idx, total_det = 0, active_hubp_count = 0; double dcfclk = context->bw_ctx.dml.vba.DCFCLKState[vlevel][context->bw_ctx.dml.vba.maxMpcComb]; dc_assert_fp_enabled(); @@ -563,6 +563,18 @@ void dcn31_calculate_wm_and_dlg_fp( if (context->res_ctx.pipe_ctx[i].stream) context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz = 0; } + for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { + if (!context->res_ctx.pipe_ctx[i].stream) + continue; + + context->res_ctx.pipe_ctx[i].det_buffer_size_kb = + get_det_buffer_size_kbytes(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); + if (context->res_ctx.pipe_ctx[i].det_buffer_size_kb > 384) + context->res_ctx.pipe_ctx[i].det_buffer_size_kb /= 2; + total_det += context->res_ctx.pipe_ctx[i].det_buffer_size_kb; + pipe_idx++; + } + context->bw_ctx.bw.dcn.compbuf_size_kb = context->bw_ctx.dml.ip.config_return_buffer_size_in_kbytes - total_det; } void dcn31_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) @@ -570,6 +582,7 @@ void dcn31_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params struct _vcs_dpi_voltage_scaling_st *s = dc->scratch.update_bw_bounding_box.clock_limits; struct clk_limit_table *clk_table = &bw_params->clk_table; unsigned int i, closest_clk_lvl; + int max_dispclk_mhz = 0, max_dppclk_mhz = 0; int j; dc_assert_fp_enabled(); @@ -577,59 +590,55 @@ void dcn31_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params memcpy(s, dcn3_1_soc.clock_limits, sizeof(dcn3_1_soc.clock_limits)); // Default clock levels are used for diags, which may lead to overclocking. - if (!IS_DIAG_DC(dc->ctx->dce_environment)) { - int max_dispclk_mhz = 0, max_dppclk_mhz = 0; + dcn3_1_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator; + dcn3_1_ip.max_num_dpp = dc->res_pool->pipe_count; + dcn3_1_soc.num_chans = bw_params->num_channels; - dcn3_1_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator; - dcn3_1_ip.max_num_dpp = dc->res_pool->pipe_count; - dcn3_1_soc.num_chans = bw_params->num_channels; + ASSERT(clk_table->num_entries); - ASSERT(clk_table->num_entries); + /* Prepass to find max clocks independent of voltage level. */ + for (i = 0; i < clk_table->num_entries; ++i) { + if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz) + max_dispclk_mhz = clk_table->entries[i].dispclk_mhz; + if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz) + max_dppclk_mhz = clk_table->entries[i].dppclk_mhz; + } - /* Prepass to find max clocks independent of voltage level. */ - for (i = 0; i < clk_table->num_entries; ++i) { - if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz) - max_dispclk_mhz = clk_table->entries[i].dispclk_mhz; - if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz) - max_dppclk_mhz = clk_table->entries[i].dppclk_mhz; + for (i = 0; i < clk_table->num_entries; i++) { + /* loop backwards*/ + for (closest_clk_lvl = 0, j = dcn3_1_soc.num_states - 1; j >= 0; j--) { + if ((unsigned int) dcn3_1_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) { + closest_clk_lvl = j; + break; + } } - for (i = 0; i < clk_table->num_entries; i++) { - /* loop backwards*/ - for (closest_clk_lvl = 0, j = dcn3_1_soc.num_states - 1; j >= 0; j--) { - if ((unsigned int) dcn3_1_soc.clock_limits[j].dcfclk_mhz <= clk_table->entries[i].dcfclk_mhz) { - closest_clk_lvl = j; - break; - } - } + s[i].state = i; - s[i].state = i; - - /* Clocks dependent on voltage level. */ - s[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz; - s[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz; - s[i].socclk_mhz = clk_table->entries[i].socclk_mhz; - s[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * - 2 * clk_table->entries[i].wck_ratio; - - /* Clocks independent of voltage level. */ - s[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz : - dcn3_1_soc.clock_limits[closest_clk_lvl].dispclk_mhz; - - s[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz : - dcn3_1_soc.clock_limits[closest_clk_lvl].dppclk_mhz; - - s[i].dram_bw_per_chan_gbps = - dcn3_1_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps; - s[i].dscclk_mhz = dcn3_1_soc.clock_limits[closest_clk_lvl].dscclk_mhz; - s[i].dtbclk_mhz = dcn3_1_soc.clock_limits[closest_clk_lvl].dtbclk_mhz; - s[i].phyclk_d18_mhz = - dcn3_1_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz; - s[i].phyclk_mhz = dcn3_1_soc.clock_limits[closest_clk_lvl].phyclk_mhz; - } - if (clk_table->num_entries) { - dcn3_1_soc.num_states = clk_table->num_entries; - } + /* Clocks dependent on voltage level. */ + s[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz; + s[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz; + s[i].socclk_mhz = clk_table->entries[i].socclk_mhz; + s[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * + 2 * clk_table->entries[i].wck_ratio; + + /* Clocks independent of voltage level. */ + s[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz : + dcn3_1_soc.clock_limits[closest_clk_lvl].dispclk_mhz; + + s[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz : + dcn3_1_soc.clock_limits[closest_clk_lvl].dppclk_mhz; + + s[i].dram_bw_per_chan_gbps = + dcn3_1_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps; + s[i].dscclk_mhz = dcn3_1_soc.clock_limits[closest_clk_lvl].dscclk_mhz; + s[i].dtbclk_mhz = dcn3_1_soc.clock_limits[closest_clk_lvl].dtbclk_mhz; + s[i].phyclk_d18_mhz = + dcn3_1_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz; + s[i].phyclk_mhz = dcn3_1_soc.clock_limits[closest_clk_lvl].phyclk_mhz; + } + if (clk_table->num_entries) { + dcn3_1_soc.num_states = clk_table->num_entries; } memcpy(dcn3_1_soc.clock_limits, s, sizeof(dcn3_1_soc.clock_limits)); @@ -643,10 +652,7 @@ void dcn31_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params dcn3_1_soc.dram_clock_change_latency_us = dc->debug.dram_clock_change_latency_ns / 1000; } - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) - dml_init_instance(&dc->dml, &dcn3_1_soc, &dcn3_1_ip, DML_PROJECT_DCN31); - else - dml_init_instance(&dc->dml, &dcn3_1_soc, &dcn3_1_ip, DML_PROJECT_DCN31_FPGA); + dml_init_instance(&dc->dml, &dcn3_1_soc, &dcn3_1_ip, DML_PROJECT_DCN31); } void dcn315_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) @@ -707,10 +713,7 @@ void dcn315_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_param dcn3_15_soc.dram_clock_change_latency_us = dc->debug.dram_clock_change_latency_ns / 1000; } - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) - dml_init_instance(&dc->dml, &dcn3_15_soc, &dcn3_15_ip, DML_PROJECT_DCN315); - else - dml_init_instance(&dc->dml, &dcn3_15_soc, &dcn3_15_ip, DML_PROJECT_DCN31_FPGA); + dml_init_instance(&dc->dml, &dcn3_15_soc, &dcn3_15_ip, DML_PROJECT_DCN315); } void dcn316_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) @@ -726,71 +729,68 @@ void dcn316_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_param memcpy(s, dcn3_16_soc.clock_limits, sizeof(dcn3_16_soc.clock_limits)); // Default clock levels are used for diags, which may lead to overclocking. - if (!IS_DIAG_DC(dc->ctx->dce_environment)) { + dcn3_16_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator; + dcn3_16_ip.max_num_dpp = dc->res_pool->pipe_count; + dcn3_16_soc.num_chans = bw_params->num_channels; - dcn3_16_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator; - dcn3_16_ip.max_num_dpp = dc->res_pool->pipe_count; - dcn3_16_soc.num_chans = bw_params->num_channels; - - ASSERT(clk_table->num_entries); + ASSERT(clk_table->num_entries); - /* Prepass to find max clocks independent of voltage level. */ - for (i = 0; i < clk_table->num_entries; ++i) { - if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz) - max_dispclk_mhz = clk_table->entries[i].dispclk_mhz; - if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz) - max_dppclk_mhz = clk_table->entries[i].dppclk_mhz; - } + /* Prepass to find max clocks independent of voltage level. */ + for (i = 0; i < clk_table->num_entries; ++i) { + if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz) + max_dispclk_mhz = clk_table->entries[i].dispclk_mhz; + if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz) + max_dppclk_mhz = clk_table->entries[i].dppclk_mhz; + } - for (i = 0; i < clk_table->num_entries; i++) { - /* loop backwards*/ - for (closest_clk_lvl = 0, j = dcn3_16_soc.num_states - 1; j >= 0; j--) { - if ((unsigned int) dcn3_16_soc.clock_limits[j].dcfclk_mhz <= - clk_table->entries[i].dcfclk_mhz) { - closest_clk_lvl = j; - break; - } - } - // Ported from DCN315 - if (clk_table->num_entries == 1) { - /*smu gives one DPM level, let's take the highest one*/ - closest_clk_lvl = dcn3_16_soc.num_states - 1; + for (i = 0; i < clk_table->num_entries; i++) { + /* loop backwards*/ + for (closest_clk_lvl = 0, j = dcn3_16_soc.num_states - 1; j >= 0; j--) { + if ((unsigned int) dcn3_16_soc.clock_limits[j].dcfclk_mhz <= + clk_table->entries[i].dcfclk_mhz) { + closest_clk_lvl = j; + break; } + } + // Ported from DCN315 + if (clk_table->num_entries == 1) { + /*smu gives one DPM level, let's take the highest one*/ + closest_clk_lvl = dcn3_16_soc.num_states - 1; + } - s[i].state = i; + s[i].state = i; - /* Clocks dependent on voltage level. */ - s[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz; - if (clk_table->num_entries == 1 && - s[i].dcfclk_mhz < - dcn3_16_soc.clock_limits[closest_clk_lvl].dcfclk_mhz) { - /*SMU fix not released yet*/ - s[i].dcfclk_mhz = - dcn3_16_soc.clock_limits[closest_clk_lvl].dcfclk_mhz; - } - s[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz; - s[i].socclk_mhz = clk_table->entries[i].socclk_mhz; - s[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * - 2 * clk_table->entries[i].wck_ratio; - - /* Clocks independent of voltage level. */ - s[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz : - dcn3_16_soc.clock_limits[closest_clk_lvl].dispclk_mhz; - - s[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz : - dcn3_16_soc.clock_limits[closest_clk_lvl].dppclk_mhz; - - s[i].dram_bw_per_chan_gbps = - dcn3_16_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps; - s[i].dscclk_mhz = dcn3_16_soc.clock_limits[closest_clk_lvl].dscclk_mhz; - s[i].dtbclk_mhz = dcn3_16_soc.clock_limits[closest_clk_lvl].dtbclk_mhz; - s[i].phyclk_d18_mhz = - dcn3_16_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz; - s[i].phyclk_mhz = dcn3_16_soc.clock_limits[closest_clk_lvl].phyclk_mhz; - } - if (clk_table->num_entries) { - dcn3_16_soc.num_states = clk_table->num_entries; + /* Clocks dependent on voltage level. */ + s[i].dcfclk_mhz = clk_table->entries[i].dcfclk_mhz; + if (clk_table->num_entries == 1 && + s[i].dcfclk_mhz < + dcn3_16_soc.clock_limits[closest_clk_lvl].dcfclk_mhz) { + /*SMU fix not released yet*/ + s[i].dcfclk_mhz = + dcn3_16_soc.clock_limits[closest_clk_lvl].dcfclk_mhz; } + s[i].fabricclk_mhz = clk_table->entries[i].fclk_mhz; + s[i].socclk_mhz = clk_table->entries[i].socclk_mhz; + s[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * + 2 * clk_table->entries[i].wck_ratio; + + /* Clocks independent of voltage level. */ + s[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz : + dcn3_16_soc.clock_limits[closest_clk_lvl].dispclk_mhz; + + s[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz : + dcn3_16_soc.clock_limits[closest_clk_lvl].dppclk_mhz; + + s[i].dram_bw_per_chan_gbps = + dcn3_16_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps; + s[i].dscclk_mhz = dcn3_16_soc.clock_limits[closest_clk_lvl].dscclk_mhz; + s[i].dtbclk_mhz = dcn3_16_soc.clock_limits[closest_clk_lvl].dtbclk_mhz; + s[i].phyclk_d18_mhz = + dcn3_16_soc.clock_limits[closest_clk_lvl].phyclk_d18_mhz; + s[i].phyclk_mhz = dcn3_16_soc.clock_limits[closest_clk_lvl].phyclk_mhz; + } + if (clk_table->num_entries) { + dcn3_16_soc.num_states = clk_table->num_entries; } memcpy(dcn3_16_soc.clock_limits, s, sizeof(dcn3_16_soc.clock_limits)); @@ -805,13 +805,21 @@ void dcn316_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_param dcn3_16_soc.dram_clock_change_latency_us = dc->debug.dram_clock_change_latency_ns / 1000; } - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) - dml_init_instance(&dc->dml, &dcn3_16_soc, &dcn3_16_ip, DML_PROJECT_DCN31); - else - dml_init_instance(&dc->dml, &dcn3_16_soc, &dcn3_16_ip, DML_PROJECT_DCN31_FPGA); + dml_init_instance(&dc->dml, &dcn3_16_soc, &dcn3_16_ip, DML_PROJECT_DCN31); } int dcn_get_max_non_odm_pix_rate_100hz(struct _vcs_dpi_soc_bounding_box_st *soc) { return soc->clock_limits[0].dispclk_mhz * 10000.0 / (1.0 + soc->dcn_downspread_percent / 100.0); } + +int dcn_get_approx_det_segs_required_for_pstate( + struct _vcs_dpi_soc_bounding_box_st *soc, + int pix_clk_100hz, int bpp, int seg_size_kb) +{ + /* Roughly calculate required crb to hide latency. In practice there is slightly + * more buffer available for latency hiding + */ + return (int)(soc->dram_clock_change_latency_us * pix_clk_100hz * bpp + / 10240000 + seg_size_kb - 1) / seg_size_kb; +} diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h index 687d3522cc33e6bc016a04982df30021949f3535..8f9c8faed26059ae5aed1e03949ea3b0fe4e13d1 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h @@ -47,6 +47,9 @@ void dcn31_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params void dcn315_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params); void dcn316_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params); int dcn_get_max_non_odm_pix_rate_100hz(struct _vcs_dpi_soc_bounding_box_st *soc); +int dcn_get_approx_det_segs_required_for_pstate( + struct _vcs_dpi_soc_bounding_box_st *soc, + int pix_clk_100hz, int bpp, int seg_size_kb); int dcn31x_populate_dml_pipes_from_context(struct dc *dc, struct dc_state *context, diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c index bd674dc30df333059ccbfed4561d63003fb63c21..43016c462251f48a9d85d4a6a2f3134263c61cb7 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c @@ -532,7 +532,8 @@ static void CalculateStutterEfficiency( static void CalculateSwathAndDETConfiguration( bool ForceSingleDPP, int NumberOfActivePlanes, - unsigned int DETBufferSizeInKByte, + bool DETSharedByAllDPP, + unsigned int DETBufferSizeInKByte[], double MaximumSwathWidthLuma[], double MaximumSwathWidthChroma[], enum scan_direction_class SourceScan[], @@ -3118,7 +3119,7 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman v->SurfaceWidthC[k], v->SurfaceHeightY[k], v->SurfaceHeightC[k], - v->DETBufferSizeInKByte[0] * 1024, + v->DETBufferSizeInKByte[k] * 1024, v->BlockHeight256BytesY[k], v->BlockHeight256BytesC[k], v->SurfaceTiling[k], @@ -3313,7 +3314,8 @@ static void DisplayPipeConfiguration(struct display_mode_lib *mode_lib) CalculateSwathAndDETConfiguration( false, v->NumberOfActivePlanes, - v->DETBufferSizeInKByte[0], + mode_lib->project == DML_PROJECT_DCN315 && v->DETSizeOverride[0], + v->DETBufferSizeInKByte, dummy1, dummy2, v->SourceScan, @@ -3779,14 +3781,16 @@ static noinline void CalculatePrefetchSchedulePerPlane( &v->VReadyOffsetPix[k]); } -static void PatchDETBufferSizeInKByte(unsigned int NumberOfActivePlanes, int NoOfDPPThisState[], unsigned int config_return_buffer_size_in_kbytes, unsigned int *DETBufferSizeInKByte) +static void PatchDETBufferSizeInKByte(unsigned int NumberOfActivePlanes, int NoOfDPPThisState[], unsigned int config_return_buffer_size_in_kbytes, unsigned int DETBufferSizeInKByte[]) { int i, total_pipes = 0; for (i = 0; i < NumberOfActivePlanes; i++) total_pipes += NoOfDPPThisState[i]; - *DETBufferSizeInKByte = ((config_return_buffer_size_in_kbytes - DCN3_15_MIN_COMPBUF_SIZE_KB) / 64 / total_pipes) * 64; - if (*DETBufferSizeInKByte > DCN3_15_MAX_DET_SIZE) - *DETBufferSizeInKByte = DCN3_15_MAX_DET_SIZE; + DETBufferSizeInKByte[0] = ((config_return_buffer_size_in_kbytes - DCN3_15_MIN_COMPBUF_SIZE_KB) / 64 / total_pipes) * 64; + if (DETBufferSizeInKByte[0] > DCN3_15_MAX_DET_SIZE) + DETBufferSizeInKByte[0] = DCN3_15_MAX_DET_SIZE; + for (i = 1; i < NumberOfActivePlanes; i++) + DETBufferSizeInKByte[i] = DETBufferSizeInKByte[0]; } @@ -4026,7 +4030,8 @@ void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l CalculateSwathAndDETConfiguration( true, v->NumberOfActivePlanes, - v->DETBufferSizeInKByte[0], + mode_lib->project == DML_PROJECT_DCN315 && v->DETSizeOverride[0], + v->DETBufferSizeInKByte, v->MaximumSwathWidthLuma, v->MaximumSwathWidthChroma, v->SourceScan, @@ -4166,6 +4171,10 @@ void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l || (v->PlaneRequiredDISPCLK > v->MaxDispclkRoundedDownToDFSGranularity)) { v->DISPCLK_DPPCLK_Support[i][j] = false; } + if (mode_lib->project == DML_PROJECT_DCN315 && v->DETSizeOverride[k] > DCN3_15_MAX_DET_SIZE && v->NoOfDPP[i][j][k] < 2) { + v->MPCCombine[i][j][k] = true; + v->NoOfDPP[i][j][k] = 2; + } } v->TotalNumberOfActiveDPP[i][j] = 0; v->TotalNumberOfSingleDPPPlanes[i][j] = 0; @@ -4642,12 +4651,13 @@ void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l v->ODMCombineEnableThisState[k] = v->ODMCombineEnablePerState[i][k]; } - if (v->NumberOfActivePlanes > 1 && mode_lib->project == DML_PROJECT_DCN315) - PatchDETBufferSizeInKByte(v->NumberOfActivePlanes, v->NoOfDPPThisState, v->ip.config_return_buffer_size_in_kbytes, &v->DETBufferSizeInKByte[0]); + if (v->NumberOfActivePlanes > 1 && mode_lib->project == DML_PROJECT_DCN315 && !v->DETSizeOverride[0]) + PatchDETBufferSizeInKByte(v->NumberOfActivePlanes, v->NoOfDPPThisState, v->ip.config_return_buffer_size_in_kbytes, v->DETBufferSizeInKByte); CalculateSwathAndDETConfiguration( false, v->NumberOfActivePlanes, - v->DETBufferSizeInKByte[0], + mode_lib->project == DML_PROJECT_DCN315 && v->DETSizeOverride[0], + v->DETBufferSizeInKByte, v->MaximumSwathWidthLuma, v->MaximumSwathWidthChroma, v->SourceScan, @@ -5274,8 +5284,8 @@ void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l v->TotImmediateFlipBytes = 0.0; for (k = 0; k < v->NumberOfActivePlanes; k++) { v->TotImmediateFlipBytes = v->TotImmediateFlipBytes - + v->NoOfDPP[i][j][k] * v->PDEAndMetaPTEBytesPerFrame[i][j][k] + v->MetaRowBytes[i][j][k] - + v->DPTEBytesPerRow[i][j][k]; + + v->NoOfDPP[i][j][k] * (v->PDEAndMetaPTEBytesPerFrame[i][j][k] + v->MetaRowBytes[i][j][k] + + v->DPTEBytesPerRow[i][j][k]); } for (k = 0; k < v->NumberOfActivePlanes; k++) { @@ -6611,7 +6621,8 @@ static void CalculateStutterEfficiency( static void CalculateSwathAndDETConfiguration( bool ForceSingleDPP, int NumberOfActivePlanes, - unsigned int DETBufferSizeInKByte, + bool DETSharedByAllDPP, + unsigned int DETBufferSizeInKByteA[], double MaximumSwathWidthLuma[], double MaximumSwathWidthChroma[], enum scan_direction_class SourceScan[], @@ -6695,6 +6706,10 @@ static void CalculateSwathAndDETConfiguration( *ViewportSizeSupport = true; for (k = 0; k < NumberOfActivePlanes; ++k) { + unsigned int DETBufferSizeInKByte = DETBufferSizeInKByteA[k]; + + if (DETSharedByAllDPP && DPPPerPlane[k]) + DETBufferSizeInKByte /= DPPPerPlane[k]; if ((SourcePixelFormat[k] == dm_444_64 || SourcePixelFormat[k] == dm_444_32 || SourcePixelFormat[k] == dm_444_16 || SourcePixelFormat[k] == dm_mono_16 || SourcePixelFormat[k] == dm_mono_8 || SourcePixelFormat[k] == dm_rgbe)) { if (SurfaceTiling[k] == dm_sw_linear @@ -7017,7 +7032,7 @@ static double CalculateUrgentLatency( return ret; } -static void UseMinimumDCFCLK( +static noinline_for_stack void UseMinimumDCFCLK( struct display_mode_lib *mode_lib, int MaxPrefetchMode, int ReorderingBytes) diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c index 2244e4fb8c96da195ce1c1b1d3771c13de5dc576..4113ce79c4aff59b9f8ba6f5381ffc7e53459d6e 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c @@ -987,8 +987,7 @@ static void dml_rq_dlg_get_dlg_params( dlg_vblank_start = interlaced ? (vblank_start / 2) : vblank_start; disp_dlg_regs->min_dst_y_next_start = (unsigned int) (((double) dlg_vblank_start) * dml_pow(2, 2)); - disp_dlg_regs->optimized_min_dst_y_next_start_us = 0; - disp_dlg_regs->optimized_min_dst_y_next_start = disp_dlg_regs->min_dst_y_next_start; + disp_dlg_regs->min_dst_y_next_start_us = 0; ASSERT(disp_dlg_regs->min_dst_y_next_start < (unsigned int)dml_pow(2, 18)); dml_print("DML_DLG: %s: min_ttu_vblank (us) = %3.2f\n", __func__, min_ttu_vblank); @@ -1433,14 +1432,6 @@ static void dml_rq_dlg_get_dlg_params( dml_print("DML_DLG: %s: disp_dlg_regs->dst_y_per_vm_flip = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_vm_flip); dml_print("DML_DLG: %s: disp_dlg_regs->dst_y_per_row_flip = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_row_flip); - // hack for FPGA - if (mode_lib->project == DML_PROJECT_DCN31_FPGA) { - if (disp_dlg_regs->vratio_prefetch >= (unsigned int) dml_pow(2, 22)) { - disp_dlg_regs->vratio_prefetch = (unsigned int) dml_pow(2, 22) - 1; - dml_print("vratio_prefetch exceed the max value, the register field is [21:0]\n"); - } - } - disp_dlg_regs->refcyc_per_pte_group_vblank_l = (unsigned int) (dst_y_per_row_vblank * (double) htotal * ref_freq_to_pix_freq / (double) dpte_groups_per_row_ub_l); ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_l < (unsigned int)dml_pow(2, 13)); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c index 9e54e3d0eb7808f2ee3bebeab8e12769be6a9ecc..c9afddd11589b5ec60a2f313d8747309c604a151 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c @@ -33,7 +33,7 @@ #include "dml/display_mode_vba.h" struct _vcs_dpi_ip_params_st dcn3_14_ip = { - .VBlankNomDefaultUS = 668, + .VBlankNomDefaultUS = 800, .gpuvm_enable = 1, .gpuvm_max_page_table_levels = 1, .hostvm_enable = 1, @@ -190,8 +190,7 @@ void dcn314_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p dc_assert_fp_enabled(); // Default clock levels are used for diags, which may lead to overclocking. - if (!IS_DIAG_DC(dc->ctx->dce_environment) && dc->config.use_default_clock_table == false) { - + if (dc->config.use_default_clock_table == false) { dcn3_14_ip.max_num_otg = dc->res_pool->res_cap->num_timing_generator; dcn3_14_ip.max_num_dpp = dc->res_pool->pipe_count; @@ -266,11 +265,7 @@ void dcn314_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p } dcn20_patch_bounding_box(dc, &dcn3_14_soc); - - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) - dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN314); - else - dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN31_FPGA); + dml_init_instance(&dc->dml, &dcn3_14_soc, &dcn3_14_ip, DML_PROJECT_DCN314); } static bool is_dual_plane(enum surface_pixel_format format) @@ -286,6 +281,7 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c struct resource_context *res_ctx = &context->res_ctx; struct pipe_ctx *pipe; bool upscaled = false; + const unsigned int max_allowed_vblank_nom = 1023; dc_assert_fp_enabled(); @@ -299,9 +295,11 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c pipe = &res_ctx->pipe_ctx[i]; timing = &pipe->stream->timing; - if (dc_extended_blank_supported(dc) && pipe->stream->adjust.v_total_max == pipe->stream->adjust.v_total_min - && pipe->stream->adjust.v_total_min > timing->v_total) - pipes[pipe_cnt].pipe.dest.vtotal = pipe->stream->adjust.v_total_min; + pipes[pipe_cnt].pipe.dest.vtotal = pipe->stream->adjust.v_total_min; + pipes[pipe_cnt].pipe.dest.vblank_nom = timing->v_total - pipes[pipe_cnt].pipe.dest.vactive; + pipes[pipe_cnt].pipe.dest.vblank_nom = min(pipes[pipe_cnt].pipe.dest.vblank_nom, dcn3_14_ip.VBlankNomDefaultUS); + pipes[pipe_cnt].pipe.dest.vblank_nom = max(pipes[pipe_cnt].pipe.dest.vblank_nom, timing->v_sync_width); + pipes[pipe_cnt].pipe.dest.vblank_nom = min(pipes[pipe_cnt].pipe.dest.vblank_nom, max_allowed_vblank_nom); if (pipe->plane_state && (pipe->plane_state->src_rect.height < pipe->plane_state->dst_rect.height || @@ -323,8 +321,6 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_luma = 0; pipes[pipe_cnt].pipe.src.dcc_fraction_of_zs_req_chroma = 0; pipes[pipe_cnt].pipe.dest.vfront_porch = timing->v_front_porch; - pipes[pipe_cnt].pipe.dest.vblank_nom = - dcn3_14_ip.VBlankNomDefaultUS / (timing->h_total / (timing->pix_clk_100hz / 10000.0)); pipes[pipe_cnt].pipe.src.dcc_rate = 3; pipes[pipe_cnt].dout.dsc_input_bpc = 0; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c index 7eb2173b7691e8bc5ab733911fc622c3b3ec394f..9010c47476e923e6f8d1f95aca7de12fe9e47165 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c @@ -5371,8 +5371,8 @@ void dml314_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_ v->TotImmediateFlipBytes = 0.0; for (k = 0; k < v->NumberOfActivePlanes; k++) { v->TotImmediateFlipBytes = v->TotImmediateFlipBytes - + v->NoOfDPP[i][j][k] * v->PDEAndMetaPTEBytesPerFrame[i][j][k] + v->MetaRowBytes[i][j][k] - + v->DPTEBytesPerRow[i][j][k]; + + v->NoOfDPP[i][j][k] * (v->PDEAndMetaPTEBytesPerFrame[i][j][k] + v->MetaRowBytes[i][j][k] + + v->DPTEBytesPerRow[i][j][k]); } for (k = 0; k < v->NumberOfActivePlanes; k++) { @@ -5557,6 +5557,65 @@ void dml314_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_ } } } + for (i = v->soc.num_states; i >= 0; i--) { + for (j = 0; j < 2; j++) { + enum dm_validation_status status = DML_VALIDATION_OK; + + if (!v->ScaleRatioAndTapsSupport) { + status = DML_FAIL_SCALE_RATIO_TAP; + } else if (!v->SourceFormatPixelAndScanSupport) { + status = DML_FAIL_SOURCE_PIXEL_FORMAT; + } else if (!v->ViewportSizeSupport[i][j]) { + status = DML_FAIL_VIEWPORT_SIZE; + } else if (P2IWith420) { + status = DML_FAIL_P2I_WITH_420; + } else if (DSCOnlyIfNecessaryWithBPP) { + status = DML_FAIL_DSC_ONLY_IF_NECESSARY_WITH_BPP; + } else if (DSC422NativeNotSupported) { + status = DML_FAIL_NOT_DSC422_NATIVE; + } else if (!v->ODMCombine4To1SupportCheckOK[i]) { + status = DML_FAIL_ODM_COMBINE4TO1; + } else if (v->NotEnoughDSCUnits[i]) { + status = DML_FAIL_NOT_ENOUGH_DSC; + } else if (!v->ROBSupport[i][j]) { + status = DML_FAIL_REORDERING_BUFFER; + } else if (!v->DISPCLK_DPPCLK_Support[i][j]) { + status = DML_FAIL_DISPCLK_DPPCLK; + } else if (!v->TotalAvailablePipesSupport[i][j]) { + status = DML_FAIL_TOTAL_AVAILABLE_PIPES; + } else if (!EnoughWritebackUnits) { + status = DML_FAIL_ENOUGH_WRITEBACK_UNITS; + } else if (!v->WritebackLatencySupport) { + status = DML_FAIL_WRITEBACK_LATENCY; + } else if (!v->WritebackScaleRatioAndTapsSupport) { + status = DML_FAIL_WRITEBACK_SCALE_RATIO_TAP; + } else if (!v->CursorSupport) { + status = DML_FAIL_CURSOR_SUPPORT; + } else if (!v->PitchSupport) { + status = DML_FAIL_PITCH_SUPPORT; + } else if (ViewportExceedsSurface) { + status = DML_FAIL_VIEWPORT_EXCEEDS_SURFACE; + } else if (!v->PrefetchSupported[i][j]) { + status = DML_FAIL_PREFETCH_SUPPORT; + } else if (!v->DynamicMetadataSupported[i][j]) { + status = DML_FAIL_DYNAMIC_METADATA; + } else if (!v->TotalVerticalActiveBandwidthSupport[i][j]) { + status = DML_FAIL_TOTAL_V_ACTIVE_BW; + } else if (!v->VRatioInPrefetchSupported[i][j]) { + status = DML_FAIL_V_RATIO_PREFETCH; + } else if (!v->PTEBufferSizeNotExceeded[i][j]) { + status = DML_FAIL_PTE_BUFFER_SIZE; + } else if (v->NonsupportedDSCInputBPC) { + status = DML_FAIL_DSC_INPUT_BPC; + } else if ((v->HostVMEnable + && !v->ImmediateFlipSupportedForState[i][j])) { + status = DML_FAIL_HOST_VM_IMMEDIATE_FLIP; + } else if (FMTBufferExceeded) { + status = DML_FAIL_FMT_BUFFER_EXCEEDED; + } + mode_lib->vba.ValidationStatus[i] = status; + } + } { unsigned int MaximumMPCCombine = 0; @@ -7061,7 +7120,7 @@ static double CalculateUrgentLatency( return ret; } -static void UseMinimumDCFCLK( +static noinline_for_stack void UseMinimumDCFCLK( struct display_mode_lib *mode_lib, int MaxPrefetchMode, int ReorderingBytes) diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c index ea4eb66066c42bc00767f27bf7e7a692e04c0de7..b3e8dc08030cb75cfbfc88e7d4b520dc86472e9f 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c @@ -951,7 +951,6 @@ static void dml_rq_dlg_get_dlg_params( { const display_pipe_source_params_st *src = &e2e_pipe_param[pipe_idx].pipe.src; const display_pipe_dest_params_st *dst = &e2e_pipe_param[pipe_idx].pipe.dest; - const display_output_params_st *dout = &e2e_pipe_param[pipe_idx].dout; const display_clocks_and_cfg_st *clks = &e2e_pipe_param[pipe_idx].clks_cfg; const scaler_ratio_depth_st *scl = &e2e_pipe_param[pipe_idx].pipe.scale_ratio_depth; const scaler_taps_st *taps = &e2e_pipe_param[pipe_idx].pipe.scale_taps; @@ -1000,8 +999,6 @@ static void dml_rq_dlg_get_dlg_params( unsigned int vupdate_width; unsigned int vready_offset; - unsigned int dispclk_delay_subtotal; - unsigned int vstartup_start; unsigned int dst_x_after_scaler; unsigned int dst_y_after_scaler; @@ -1051,7 +1048,6 @@ static void dml_rq_dlg_get_dlg_params( float vba__refcyc_per_req_delivery_pre_l = get_refcyc_per_req_delivery_pre_l_in_us(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz; // From VBA float vba__refcyc_per_req_delivery_l = get_refcyc_per_req_delivery_l_in_us(mode_lib, e2e_pipe_param, num_pipes, pipe_idx) * refclk_freq_in_mhz; // From VBA - int blank_lines = 0; memset(disp_dlg_regs, 0, sizeof(*disp_dlg_regs)); memset(disp_ttu_regs, 0, sizeof(*disp_ttu_regs)); @@ -1075,17 +1071,10 @@ static void dml_rq_dlg_get_dlg_params( min_ttu_vblank = get_min_ttu_vblank_in_us(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); // From VBA dlg_vblank_start = interlaced ? (vblank_start / 2) : vblank_start; - disp_dlg_regs->optimized_min_dst_y_next_start = disp_dlg_regs->min_dst_y_next_start; - disp_dlg_regs->optimized_min_dst_y_next_start_us = 0; - disp_dlg_regs->min_dst_y_next_start = (unsigned int) (((double) dlg_vblank_start) * dml_pow(2, 2)); - blank_lines = (dst->vblank_end + dst->vtotal_min - dst->vblank_start - dst->vstartup_start - 1); - if (blank_lines < 0) - blank_lines = 0; - if (blank_lines != 0) { - disp_dlg_regs->optimized_min_dst_y_next_start = vba__min_dst_y_next_start; - disp_dlg_regs->optimized_min_dst_y_next_start_us = (disp_dlg_regs->optimized_min_dst_y_next_start * dst->hactive) / (unsigned int) dst->pixel_rate_mhz; - disp_dlg_regs->min_dst_y_next_start = disp_dlg_regs->optimized_min_dst_y_next_start; - } + disp_dlg_regs->min_dst_y_next_start_us = + (vba__min_dst_y_next_start * dst->hactive) / (unsigned int) dst->pixel_rate_mhz; + disp_dlg_regs->min_dst_y_next_start = vba__min_dst_y_next_start * dml_pow(2, 2); + ASSERT(disp_dlg_regs->min_dst_y_next_start < (unsigned int)dml_pow(2, 18)); dml_print("DML_DLG: %s: min_ttu_vblank (us) = %3.2f\n", __func__, min_ttu_vblank); @@ -1127,13 +1116,6 @@ static void dml_rq_dlg_get_dlg_params( vupdate_offset = dst->vupdate_offset; vupdate_width = dst->vupdate_width; vready_offset = dst->vready_offset; - dispclk_delay_subtotal = mode_lib->ip.dispclk_delay_subtotal; - - if (dout->dsc_enable) { - double dsc_delay = get_dsc_delay(mode_lib, e2e_pipe_param, num_pipes, pipe_idx); // FROM VBA - - dispclk_delay_subtotal += dsc_delay; - } vstartup_start = dst->vstartup_start; if (interlaced) { @@ -1538,14 +1520,6 @@ static void dml_rq_dlg_get_dlg_params( dml_print("DML_DLG: %s: disp_dlg_regs->dst_y_per_vm_flip = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_vm_flip); dml_print("DML_DLG: %s: disp_dlg_regs->dst_y_per_row_flip = 0x%x\n", __func__, disp_dlg_regs->dst_y_per_row_flip); - // hack for FPGA - if (mode_lib->project == DML_PROJECT_DCN31_FPGA) { - if (disp_dlg_regs->vratio_prefetch >= (unsigned int) dml_pow(2, 22)) { - disp_dlg_regs->vratio_prefetch = (unsigned int) dml_pow(2, 22) - 1; - dml_print("vratio_prefetch exceed the max value, the register field is [21:0]\n"); - } - } - disp_dlg_regs->refcyc_per_pte_group_vblank_l = (unsigned int) (dst_y_per_row_vblank * (double) htotal * ref_freq_to_pix_freq / (double) dpte_groups_per_row_ub_l); ASSERT(disp_dlg_regs->refcyc_per_pte_group_vblank_l < (unsigned int)dml_pow(2, 13)); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c index 47beb4ea779d30df03b6e5daf638eee79f9e7f05..e2bb2b9971f304f3c6bf5699e9e54f0b9cde21cf 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c @@ -35,6 +35,15 @@ #define DC_LOGGER_INIT(logger) +static const struct subvp_high_refresh_list subvp_high_refresh_list = { + .min_refresh = 120, + .max_refresh = 165, + .res = { + {.width = 3840, .height = 2160, }, + {.width = 3440, .height = 1440, }, + {.width = 2560, .height = 1440, }}, +}; + struct _vcs_dpi_ip_params_st dcn3_2_ip = { .gpuvm_enable = 0, .gpuvm_max_page_table_levels = 4, @@ -138,7 +147,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_2_soc = { .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, - .pct_ideal_sdp_bw_after_urgent = 100.0, + .pct_ideal_sdp_bw_after_urgent = 90.0, .pct_ideal_fabric_bw_after_urgent = 67.0, .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 20.0, .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0, // N/A, for now keep as is until DML implemented @@ -670,7 +679,6 @@ static bool dcn32_assign_subvp_pipe(struct dc *dc, unsigned int max_frame_time = 0; bool valid_assignment_found = false; unsigned int free_pipes = dcn32_get_num_free_pipes(dc, context); - bool current_assignment_freesync = false; struct vba_vars_st *vba = &context->bw_ctx.dml.vba; for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { @@ -692,8 +700,12 @@ static bool dcn32_assign_subvp_pipe(struct dc *dc, * to combine this with SubVP can cause issues with the scheduling). * - Not TMZ surface */ - if (pipe->plane_state && !pipe->top_pipe && !dcn32_is_center_timing(pipe) && !dcn32_is_psr_capable(pipe) && - pipe->stream->mall_stream_config.type == SUBVP_NONE && refresh_rate < 120 && !pipe->plane_state->address.tmz_surface && + if (pipe->plane_state && !pipe->top_pipe && !dcn32_is_center_timing(pipe) && + !(pipe->stream->timing.pix_clk_100hz / 10000 > DCN3_2_MAX_SUBVP_PIXEL_RATE_MHZ) && + (!dcn32_is_psr_capable(pipe) || (context->stream_count == 1 && dc->caps.dmub_caps.subvp_psr)) && + pipe->stream->mall_stream_config.type == SUBVP_NONE && + (refresh_rate < 120 || dcn32_allow_subvp_high_refresh_rate(dc, context, pipe)) && + !pipe->plane_state->address.tmz_surface && (vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] <= 0 || (vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] > 0 && dcn32_allow_subvp_with_active_margin(pipe)))) { @@ -707,19 +719,10 @@ static bool dcn32_assign_subvp_pipe(struct dc *dc, struct dc_stream_state *stream = pipe->stream; unsigned int frame_us = (stream->timing.v_total * stream->timing.h_total / (double)(stream->timing.pix_clk_100hz * 100)) * 1000000; - if (frame_us > max_frame_time && !stream->ignore_msa_timing_param) { + if (frame_us > max_frame_time) { *index = i; max_frame_time = frame_us; valid_assignment_found = true; - current_assignment_freesync = false; - /* For the 2-Freesync display case, still choose the one with the - * longest frame time - */ - } else if (stream->ignore_msa_timing_param && (!valid_assignment_found || - (current_assignment_freesync && frame_us > max_frame_time))) { - *index = i; - valid_assignment_found = true; - current_assignment_freesync = true; } } } @@ -851,10 +854,9 @@ static bool subvp_subvp_schedulable(struct dc *dc, struct dc_state *context) } /** - * subvp_drr_schedulable - Determine if SubVP + DRR config is schedulable + * subvp_drr_schedulable() - Determine if SubVP + DRR config is schedulable * @dc: current dc state * @context: new dc state - * @drr_pipe: DRR pipe_ctx for the SubVP + DRR config * * High level algorithm: * 1. Get timing for SubVP pipe, phantom pipe, and DRR pipe @@ -865,11 +867,12 @@ static bool subvp_subvp_schedulable(struct dc *dc, struct dc_state *context) * * Return: True if the SubVP + DRR config is schedulable, false otherwise */ -static bool subvp_drr_schedulable(struct dc *dc, struct dc_state *context, struct pipe_ctx *drr_pipe) +static bool subvp_drr_schedulable(struct dc *dc, struct dc_state *context) { bool schedulable = false; uint32_t i; struct pipe_ctx *pipe = NULL; + struct pipe_ctx *drr_pipe = NULL; struct dc_crtc_timing *main_timing = NULL; struct dc_crtc_timing *phantom_timing = NULL; struct dc_crtc_timing *drr_timing = NULL; @@ -880,10 +883,6 @@ static bool subvp_drr_schedulable(struct dc *dc, struct dc_state *context, struc int16_t stretched_drr_us = 0; int16_t drr_stretched_vblank_us = 0; int16_t max_vblank_mallregion = 0; - const struct dc_config *config = &dc->config; - - if (config->disable_subvp_drr) - return false; // Find SubVP pipe for (i = 0; i < dc->res_pool->pipe_count; i++) { @@ -899,6 +898,19 @@ static bool subvp_drr_schedulable(struct dc *dc, struct dc_state *context, struc break; } + // Find the DRR pipe + for (i = 0; i < dc->res_pool->pipe_count; i++) { + drr_pipe = &context->res_ctx.pipe_ctx[i]; + + // We check for master pipe only + if (!drr_pipe->stream || !drr_pipe->plane_state || drr_pipe->top_pipe || drr_pipe->prev_odm_pipe) + continue; + + if (drr_pipe->stream->mall_stream_config.type == SUBVP_NONE && drr_pipe->stream->ignore_msa_timing_param && + (drr_pipe->stream->allow_freesync || drr_pipe->stream->vrr_active_variable)) + break; + } + main_timing = &pipe->stream->timing; phantom_timing = &pipe->stream->mall_stream_config.paired_stream->timing; drr_timing = &drr_pipe->stream->timing; @@ -984,13 +996,7 @@ static bool subvp_vblank_schedulable(struct dc *dc, struct dc_state *context) if (!subvp_pipe && pipe->stream->mall_stream_config.type == SUBVP_MAIN) subvp_pipe = pipe; } - // Use ignore_msa_timing_param and VRR active, or Freesync flag to identify as DRR On - if (found && context->res_ctx.pipe_ctx[vblank_index].stream->ignore_msa_timing_param && - (context->res_ctx.pipe_ctx[vblank_index].stream->allow_freesync || - context->res_ctx.pipe_ctx[vblank_index].stream->vrr_active_variable)) { - // SUBVP + DRR case -- only allowed if run through DRR validation path - schedulable = false; - } else if (found) { + if (found) { main_timing = &subvp_pipe->stream->timing; phantom_timing = &subvp_pipe->stream->mall_stream_config.paired_stream->timing; vblank_timing = &context->res_ctx.pipe_ctx[vblank_index].stream->timing; @@ -1019,6 +1025,53 @@ static bool subvp_vblank_schedulable(struct dc *dc, struct dc_state *context) return schedulable; } +/** + * subvp_subvp_admissable() - Determine if subvp + subvp config is admissible + * + * @dc: Current DC state + * @context: New DC state to be programmed + * + * SubVP + SubVP is admissible under the following conditions: + * - All SubVP pipes are < 120Hz OR + * - All SubVP pipes are >= 120hz + * + * Return: True if admissible, false otherwise + */ +static bool subvp_subvp_admissable(struct dc *dc, + struct dc_state *context) +{ + bool result = false; + uint32_t i; + uint8_t subvp_count = 0; + uint32_t min_refresh = subvp_high_refresh_list.min_refresh, max_refresh = 0; + uint32_t refresh_rate = 0; + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + + if (!pipe->stream) + continue; + + if (pipe->plane_state && !pipe->top_pipe && + pipe->stream->mall_stream_config.type == SUBVP_MAIN) { + refresh_rate = (pipe->stream->timing.pix_clk_100hz * 100 + + pipe->stream->timing.v_total * pipe->stream->timing.h_total - 1) + / (double)(pipe->stream->timing.v_total * pipe->stream->timing.h_total); + if (refresh_rate < min_refresh) + min_refresh = refresh_rate; + if (refresh_rate > max_refresh) + max_refresh = refresh_rate; + subvp_count++; + } + } + + if (subvp_count == 2 && ((min_refresh < 120 && max_refresh < 120) || + (min_refresh >= 120 && max_refresh >= 120))) + result = true; + + return result; +} + /** * subvp_validate_static_schedulability - Check which SubVP case is calculated * and handle static analysis based on the case. @@ -1037,11 +1090,12 @@ static bool subvp_validate_static_schedulability(struct dc *dc, struct dc_state *context, int vlevel) { - bool schedulable = true; // true by default for single display case + bool schedulable = false; struct vba_vars_st *vba = &context->bw_ctx.dml.vba; uint32_t i, pipe_idx; uint8_t subvp_count = 0; uint8_t vactive_count = 0; + uint8_t non_subvp_pipes = 0; for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; @@ -1049,14 +1103,18 @@ static bool subvp_validate_static_schedulability(struct dc *dc, if (!pipe->stream) continue; - if (pipe->plane_state && !pipe->top_pipe && - pipe->stream->mall_stream_config.type == SUBVP_MAIN) - subvp_count++; + if (pipe->plane_state && !pipe->top_pipe) { + if (pipe->stream->mall_stream_config.type == SUBVP_MAIN) + subvp_count++; + if (pipe->stream->mall_stream_config.type == SUBVP_NONE) { + non_subvp_pipes++; + } + } // Count how many planes that aren't SubVP/phantom are capable of VACTIVE // switching (SubVP + VACTIVE unsupported). In situations where we force // SubVP for a VACTIVE plane, we don't want to increment the vactive_count. - if (vba->ActiveDRAMClockChangeLatencyMargin[vba->pipe_plane[pipe_idx]] > 0 && + if (vba->ActiveDRAMClockChangeLatencyMarginPerState[vlevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] > 0 && pipe->stream->mall_stream_config.type == SUBVP_NONE) { vactive_count++; } @@ -1065,13 +1123,14 @@ static bool subvp_validate_static_schedulability(struct dc *dc, if (subvp_count == 2) { // Static schedulability check for SubVP + SubVP case - schedulable = subvp_subvp_schedulable(dc, context); - } else if (vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_vblank_w_mall_sub_vp) { - // Static schedulability check for SubVP + VBLANK case. Also handle the case where - // DML outputs SubVP + VBLANK + VACTIVE (DML will report as SubVP + VBLANK) - if (vactive_count > 0) - schedulable = false; - else + schedulable = subvp_subvp_admissable(dc, context) && subvp_subvp_schedulable(dc, context); + } else if (subvp_count == 1 && non_subvp_pipes == 0) { + // Single SubVP configs will be supported by default as long as it's suppported by DML + schedulable = true; + } else if (subvp_count == 1 && non_subvp_pipes == 1) { + if (dcn32_subvp_drr_admissable(dc, context)) + schedulable = subvp_drr_schedulable(dc, context); + else if (dcn32_subvp_vblank_admissable(dc, context, vlevel)) schedulable = subvp_vblank_schedulable(dc, context); } else if (vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_vactive_w_mall_sub_vp && vactive_count > 0) { @@ -1095,10 +1154,6 @@ static void dcn32_full_validate_bw_helper(struct dc *dc, unsigned int dc_pipe_idx = 0; int i = 0; bool found_supported_config = false; - struct pipe_ctx *pipe = NULL; - uint32_t non_subvp_pipes = 0; - bool drr_pipe_found = false; - uint32_t drr_pipe_index = 0; dc_assert_fp_enabled(); @@ -1129,7 +1184,7 @@ static void dcn32_full_validate_bw_helper(struct dc *dc, * 4. Display configuration passes validation * 5. (Config doesn't support MCLK in VACTIVE/VBLANK || dc->debug.force_subvp_mclk_switch) */ - if (!dc->debug.force_disable_subvp && dcn32_all_pipes_have_stream_and_plane(dc, context) && + if (!dc->debug.force_disable_subvp && !dc->caps.dmub_caps.gecc_enable && dcn32_all_pipes_have_stream_and_plane(dc, context) && !dcn32_mpo_in_use(context) && !dcn32_any_surfaces_rotated(dc, context) && (*vlevel == context->bw_ctx.dml.soc.num_states || vba->DRAMClockChangeSupport[*vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported || @@ -1188,31 +1243,12 @@ static void dcn32_full_validate_bw_helper(struct dc *dc, } } - if (*vlevel < context->bw_ctx.dml.soc.num_states && - vba->DRAMClockChangeSupport[*vlevel][vba->maxMpcComb] != dm_dram_clock_change_unsupported - && subvp_validate_static_schedulability(dc, context, *vlevel)) { + if (*vlevel < context->bw_ctx.dml.soc.num_states + && subvp_validate_static_schedulability(dc, context, *vlevel)) found_supported_config = true; - } else if (*vlevel < context->bw_ctx.dml.soc.num_states) { - /* Case where 1 SubVP is added, and DML reports MCLK unsupported or DRR is allowed. - * This handles the case for SubVP + DRR, where the DRR display does not support MCLK - * switch at it's native refresh rate / timing, or DRR is allowed for the non-subvp - * display. - */ - for (i = 0; i < dc->res_pool->pipe_count; i++) { - pipe = &context->res_ctx.pipe_ctx[i]; - if (pipe->stream && pipe->plane_state && !pipe->top_pipe && - pipe->stream->mall_stream_config.type == SUBVP_NONE) { - non_subvp_pipes++; - // Use ignore_msa_timing_param flag to identify as DRR - if (pipe->stream->ignore_msa_timing_param && pipe->stream->allow_freesync) { - drr_pipe_found = true; - drr_pipe_index = i; - } - } - } - // If there is only 1 remaining non SubVP pipe that is DRR, check static - // schedulability for SubVP + DRR. - if (non_subvp_pipes == 1 && drr_pipe_found) { + if (found_supported_config) { + // For SubVP + DRR cases, we can force the lowest vlevel that supports the mode + if (dcn32_subvp_drr_admissable(dc, context) && subvp_drr_schedulable(dc, context)) { /* find lowest vlevel that supports the config */ for (i = *vlevel; i >= 0; i--) { if (vba->ModeSupport[i][vba->maxMpcComb]) { @@ -1221,9 +1257,6 @@ static void dcn32_full_validate_bw_helper(struct dc *dc, break; } } - - found_supported_config = subvp_drr_schedulable(dc, context, - &context->res_ctx.pipe_ctx[drr_pipe_index]); } } } @@ -1315,6 +1348,7 @@ static void dcn32_calculate_dlg_params(struct dc *dc, struct dc_state *context, int i, pipe_idx, active_hubp_count = 0; bool usr_retraining_support = false; bool unbounded_req_enabled = false; + struct vba_vars_st *vba = &context->bw_ctx.dml.vba; dc_assert_fp_enabled(); @@ -1396,6 +1430,11 @@ static void dcn32_calculate_dlg_params(struct dc *dc, struct dc_state *context, context->res_ctx.pipe_ctx[i].surface_size_in_mall_bytes = get_surface_size_in_mall(&context->bw_ctx.dml, pipes, pipe_cnt, pipe_idx); + if (vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] > 0) + context->res_ctx.pipe_ctx[i].has_vactive_margin = true; + else + context->res_ctx.pipe_ctx[i].has_vactive_margin = false; + /* MALL Allocation Sizes */ /* count from active, top pipes per plane only */ if (context->res_ctx.pipe_ctx[i].stream && context->res_ctx.pipe_ctx[i].plane_state && @@ -1432,6 +1471,7 @@ static void dcn32_calculate_dlg_params(struct dc *dc, struct dc_state *context, context->bw_ctx.bw.dcn.clk.dramclk_khz = 0; context->bw_ctx.bw.dcn.clk.fclk_khz = 0; context->bw_ctx.bw.dcn.clk.p_state_change_support = true; + context->bw_ctx.bw.dcn.clk.fclk_p_state_change_support = true; } /*save a original dppclock copy*/ context->bw_ctx.bw.dcn.clk.bw_dppclk_khz = context->bw_ctx.bw.dcn.clk.dppclk_khz; @@ -2005,6 +2045,7 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context, maxMpcComb = context->bw_ctx.dml.vba.maxMpcComb; dcfclk_from_fw_based_mclk_switching = context->bw_ctx.dml.vba.DCFCLKState[vlevel][context->bw_ctx.dml.vba.maxMpcComb]; pstate_en = true; + context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][maxMpcComb] = dm_dram_clock_change_vblank; } else { /* Restore FCLK latency and re-run validation to go back to original validation * output if we find that enabling FPO does not give us any benefit (i.e. lower @@ -2062,6 +2103,7 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context, * sr_enter_exit/sr_exit should be lower than used for DRAM (TBD after bringup or later, use as decided in Clk Mgr) */ + /* if (dcn3_2_soc.num_states > 2) { vlevel_temp = 0; dcfclk = dc->clk_mgr->bw_params->clk_table.entries[0].dcfclk_mhz; @@ -2088,6 +2130,7 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context, context->bw_ctx.bw.dcn.watermarks.d.urgent_latency_ns = get_urgent_latency(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; context->bw_ctx.bw.dcn.watermarks.d.cstate_pstate.fclk_pstate_change_ns = get_fclk_watermark(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; context->bw_ctx.bw.dcn.watermarks.d.usr_retraining_ns = get_usr_retraining_watermark(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; + */ /* Set C, for Dummy P-State: * All clocks min. @@ -2189,6 +2232,9 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context, context->bw_ctx.bw.dcn.watermarks.a.usr_retraining_ns = get_usr_retraining_watermark(&context->bw_ctx.dml, pipes, pipe_cnt) * 1000; } + /* Make set D = set A since we do not optimized watermarks for MALL */ + context->bw_ctx.bw.dcn.watermarks.d = context->bw_ctx.bw.dcn.watermarks.a; + for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { if (!context->res_ctx.pipe_ctx[i].stream) continue; @@ -2303,14 +2349,48 @@ void dcn32_patch_dpm_table(struct clk_bw_params *bw_params) bw_params->clk_table.entries[0].memclk_mhz = dcn3_2_soc.clock_limits[0].dram_speed_mts / 16; } -static int build_synthetic_soc_states(struct clk_bw_params *bw_params, +/* + * override_max_clk_values - Overwrite the max clock frequencies with the max DC mode timings + * Input: + * max_clk_limit - struct containing the desired clock timings + * Output: + * curr_clk_limit - struct containing the timings that need to be overwritten + * Return: 0 upon success, non-zero for failure + */ +static int override_max_clk_values(struct clk_limit_table_entry *max_clk_limit, + struct clk_limit_table_entry *curr_clk_limit) +{ + if (NULL == max_clk_limit || NULL == curr_clk_limit) + return -1; //invalid parameters + + //only overwrite if desired max clock frequency is initialized + if (max_clk_limit->dcfclk_mhz != 0) + curr_clk_limit->dcfclk_mhz = max_clk_limit->dcfclk_mhz; + + if (max_clk_limit->fclk_mhz != 0) + curr_clk_limit->fclk_mhz = max_clk_limit->fclk_mhz; + + if (max_clk_limit->memclk_mhz != 0) + curr_clk_limit->memclk_mhz = max_clk_limit->memclk_mhz; + + if (max_clk_limit->socclk_mhz != 0) + curr_clk_limit->socclk_mhz = max_clk_limit->socclk_mhz; + + if (max_clk_limit->dtbclk_mhz != 0) + curr_clk_limit->dtbclk_mhz = max_clk_limit->dtbclk_mhz; + + if (max_clk_limit->dispclk_mhz != 0) + curr_clk_limit->dispclk_mhz = max_clk_limit->dispclk_mhz; + + return 0; +} + +static int build_synthetic_soc_states(bool disable_dc_mode_overwrite, struct clk_bw_params *bw_params, struct _vcs_dpi_voltage_scaling_st *table, unsigned int *num_entries) { int i, j; struct _vcs_dpi_voltage_scaling_st entry = {0}; - - unsigned int max_dcfclk_mhz = 0, max_dispclk_mhz = 0, max_dppclk_mhz = 0, - max_phyclk_mhz = 0, max_dtbclk_mhz = 0, max_fclk_mhz = 0, max_uclk_mhz = 0; + struct clk_limit_table_entry max_clk_data = {0}; unsigned int min_dcfclk_mhz = 199, min_fclk_mhz = 299; @@ -2321,51 +2401,76 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params, unsigned int num_fclk_dpms = 0; unsigned int num_dcfclk_dpms = 0; - for (i = 0; i < MAX_NUM_DPM_LVL; i++) { - if (bw_params->clk_table.entries[i].dcfclk_mhz > max_dcfclk_mhz) - max_dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz; - if (bw_params->clk_table.entries[i].fclk_mhz > max_fclk_mhz) - max_fclk_mhz = bw_params->clk_table.entries[i].fclk_mhz; - if (bw_params->clk_table.entries[i].memclk_mhz > max_uclk_mhz) - max_uclk_mhz = bw_params->clk_table.entries[i].memclk_mhz; - if (bw_params->clk_table.entries[i].dispclk_mhz > max_dispclk_mhz) - max_dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz; - if (bw_params->clk_table.entries[i].dppclk_mhz > max_dppclk_mhz) - max_dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz; - if (bw_params->clk_table.entries[i].phyclk_mhz > max_phyclk_mhz) - max_phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz; - if (bw_params->clk_table.entries[i].dtbclk_mhz > max_dtbclk_mhz) - max_dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz; + unsigned int num_dc_uclk_dpms = 0; + unsigned int num_dc_fclk_dpms = 0; + unsigned int num_dc_dcfclk_dpms = 0; - if (bw_params->clk_table.entries[i].memclk_mhz > 0) + for (i = 0; i < MAX_NUM_DPM_LVL; i++) { + if (bw_params->clk_table.entries[i].dcfclk_mhz > max_clk_data.dcfclk_mhz) + max_clk_data.dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz; + if (bw_params->clk_table.entries[i].fclk_mhz > max_clk_data.fclk_mhz) + max_clk_data.fclk_mhz = bw_params->clk_table.entries[i].fclk_mhz; + if (bw_params->clk_table.entries[i].memclk_mhz > max_clk_data.memclk_mhz) + max_clk_data.memclk_mhz = bw_params->clk_table.entries[i].memclk_mhz; + if (bw_params->clk_table.entries[i].dispclk_mhz > max_clk_data.dispclk_mhz) + max_clk_data.dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz; + if (bw_params->clk_table.entries[i].dppclk_mhz > max_clk_data.dppclk_mhz) + max_clk_data.dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz; + if (bw_params->clk_table.entries[i].phyclk_mhz > max_clk_data.phyclk_mhz) + max_clk_data.phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz; + if (bw_params->clk_table.entries[i].dtbclk_mhz > max_clk_data.dtbclk_mhz) + max_clk_data.dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz; + + if (bw_params->clk_table.entries[i].memclk_mhz > 0) { num_uclk_dpms++; - if (bw_params->clk_table.entries[i].fclk_mhz > 0) + if (bw_params->clk_table.entries[i].memclk_mhz <= bw_params->dc_mode_limit.memclk_mhz) + num_dc_uclk_dpms++; + } + if (bw_params->clk_table.entries[i].fclk_mhz > 0) { num_fclk_dpms++; - if (bw_params->clk_table.entries[i].dcfclk_mhz > 0) + if (bw_params->clk_table.entries[i].fclk_mhz <= bw_params->dc_mode_limit.fclk_mhz) + num_dc_fclk_dpms++; + } + if (bw_params->clk_table.entries[i].dcfclk_mhz > 0) { num_dcfclk_dpms++; + if (bw_params->clk_table.entries[i].dcfclk_mhz <= bw_params->dc_mode_limit.dcfclk_mhz) + num_dc_dcfclk_dpms++; + } + } + + if (!disable_dc_mode_overwrite) { + //Overwrite max frequencies with max DC mode frequencies for DC mode systems + override_max_clk_values(&bw_params->dc_mode_limit, &max_clk_data); + num_uclk_dpms = num_dc_uclk_dpms; + num_fclk_dpms = num_dc_fclk_dpms; + num_dcfclk_dpms = num_dc_dcfclk_dpms; + bw_params->clk_table.num_entries_per_clk.num_memclk_levels = num_uclk_dpms; + bw_params->clk_table.num_entries_per_clk.num_fclk_levels = num_fclk_dpms; } if (num_dcfclk_dpms > 0 && bw_params->clk_table.entries[0].fclk_mhz > min_fclk_mhz) min_fclk_mhz = bw_params->clk_table.entries[0].fclk_mhz; - if (!max_dcfclk_mhz || !max_dispclk_mhz || !max_dtbclk_mhz) + if (!max_clk_data.dcfclk_mhz || !max_clk_data.dispclk_mhz || !max_clk_data.dtbclk_mhz) return -1; - if (max_dppclk_mhz == 0) - max_dppclk_mhz = max_dispclk_mhz; + if (max_clk_data.dppclk_mhz == 0) + max_clk_data.dppclk_mhz = max_clk_data.dispclk_mhz; - if (max_fclk_mhz == 0) - max_fclk_mhz = max_dcfclk_mhz * dcn3_2_soc.pct_ideal_sdp_bw_after_urgent / dcn3_2_soc.pct_ideal_fabric_bw_after_urgent; + if (max_clk_data.fclk_mhz == 0) + max_clk_data.fclk_mhz = max_clk_data.dcfclk_mhz * + dcn3_2_soc.pct_ideal_sdp_bw_after_urgent / + dcn3_2_soc.pct_ideal_fabric_bw_after_urgent; - if (max_phyclk_mhz == 0) - max_phyclk_mhz = dcn3_2_soc.clock_limits[0].phyclk_mhz; + if (max_clk_data.phyclk_mhz == 0) + max_clk_data.phyclk_mhz = dcn3_2_soc.clock_limits[0].phyclk_mhz; *num_entries = 0; - entry.dispclk_mhz = max_dispclk_mhz; - entry.dscclk_mhz = max_dispclk_mhz / 3; - entry.dppclk_mhz = max_dppclk_mhz; - entry.dtbclk_mhz = max_dtbclk_mhz; - entry.phyclk_mhz = max_phyclk_mhz; + entry.dispclk_mhz = max_clk_data.dispclk_mhz; + entry.dscclk_mhz = max_clk_data.dispclk_mhz / 3; + entry.dppclk_mhz = max_clk_data.dppclk_mhz; + entry.dtbclk_mhz = max_clk_data.dtbclk_mhz; + entry.phyclk_mhz = max_clk_data.phyclk_mhz; entry.phyclk_d18_mhz = dcn3_2_soc.clock_limits[0].phyclk_d18_mhz; entry.phyclk_d32_mhz = dcn3_2_soc.clock_limits[0].phyclk_d32_mhz; @@ -2379,7 +2484,7 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params, } // Insert the max DCFCLK - entry.dcfclk_mhz = max_dcfclk_mhz; + entry.dcfclk_mhz = max_clk_data.dcfclk_mhz; entry.fabricclk_mhz = 0; entry.dram_speed_mts = 0; @@ -2407,7 +2512,7 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params, // If FCLK fine grained, only insert max else { entry.dcfclk_mhz = 0; - entry.fabricclk_mhz = max_fclk_mhz; + entry.fabricclk_mhz = max_clk_data.fclk_mhz; entry.dram_speed_mts = 0; insert_entry_into_table_sorted(table, num_entries, &entry); @@ -2419,9 +2524,9 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params, // Remove states that require higher clocks than are supported for (i = *num_entries - 1; i >= 0 ; i--) { - if (table[i].dcfclk_mhz > max_dcfclk_mhz || - table[i].fabricclk_mhz > max_fclk_mhz || - table[i].dram_speed_mts > max_uclk_mhz * 16) + if (table[i].dcfclk_mhz > max_clk_data.dcfclk_mhz || + table[i].fabricclk_mhz > max_clk_data.fclk_mhz || + table[i].dram_speed_mts > max_clk_data.memclk_mhz * 16) remove_entry_from_table_at_index(table, num_entries, i); } @@ -2508,80 +2613,78 @@ void dcn32_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_pa { dc_assert_fp_enabled(); - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - /* Overrides from dc->config options */ - dcn3_2_ip.clamp_min_dcfclk = dc->config.clamp_min_dcfclk; + /* Overrides from dc->config options */ + dcn3_2_ip.clamp_min_dcfclk = dc->config.clamp_min_dcfclk; - /* Override from passed dc->bb_overrides if available*/ - if ((int)(dcn3_2_soc.sr_exit_time_us * 1000) != dc->bb_overrides.sr_exit_time_ns - && dc->bb_overrides.sr_exit_time_ns) { - dcn3_2_soc.sr_exit_time_us = dc->bb_overrides.sr_exit_time_ns / 1000.0; - } - - if ((int)(dcn3_2_soc.sr_enter_plus_exit_time_us * 1000) - != dc->bb_overrides.sr_enter_plus_exit_time_ns - && dc->bb_overrides.sr_enter_plus_exit_time_ns) { - dcn3_2_soc.sr_enter_plus_exit_time_us = - dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; - } + /* Override from passed dc->bb_overrides if available*/ + if ((int)(dcn3_2_soc.sr_exit_time_us * 1000) != dc->bb_overrides.sr_exit_time_ns + && dc->bb_overrides.sr_exit_time_ns) { + dcn3_2_soc.sr_exit_time_us = dc->bb_overrides.sr_exit_time_ns / 1000.0; + } - if ((int)(dcn3_2_soc.urgent_latency_us * 1000) != dc->bb_overrides.urgent_latency_ns - && dc->bb_overrides.urgent_latency_ns) { - dcn3_2_soc.urgent_latency_us = dc->bb_overrides.urgent_latency_ns / 1000.0; - dcn3_2_soc.urgent_latency_pixel_data_only_us = dc->bb_overrides.urgent_latency_ns / 1000.0; - } + if ((int)(dcn3_2_soc.sr_enter_plus_exit_time_us * 1000) + != dc->bb_overrides.sr_enter_plus_exit_time_ns + && dc->bb_overrides.sr_enter_plus_exit_time_ns) { + dcn3_2_soc.sr_enter_plus_exit_time_us = + dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; + } - if ((int)(dcn3_2_soc.dram_clock_change_latency_us * 1000) - != dc->bb_overrides.dram_clock_change_latency_ns - && dc->bb_overrides.dram_clock_change_latency_ns) { - dcn3_2_soc.dram_clock_change_latency_us = - dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; - } + if ((int)(dcn3_2_soc.urgent_latency_us * 1000) != dc->bb_overrides.urgent_latency_ns + && dc->bb_overrides.urgent_latency_ns) { + dcn3_2_soc.urgent_latency_us = dc->bb_overrides.urgent_latency_ns / 1000.0; + dcn3_2_soc.urgent_latency_pixel_data_only_us = dc->bb_overrides.urgent_latency_ns / 1000.0; + } - if ((int)(dcn3_2_soc.fclk_change_latency_us * 1000) - != dc->bb_overrides.fclk_clock_change_latency_ns - && dc->bb_overrides.fclk_clock_change_latency_ns) { - dcn3_2_soc.fclk_change_latency_us = - dc->bb_overrides.fclk_clock_change_latency_ns / 1000; - } + if ((int)(dcn3_2_soc.dram_clock_change_latency_us * 1000) + != dc->bb_overrides.dram_clock_change_latency_ns + && dc->bb_overrides.dram_clock_change_latency_ns) { + dcn3_2_soc.dram_clock_change_latency_us = + dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; + } - if ((int)(dcn3_2_soc.dummy_pstate_latency_us * 1000) - != dc->bb_overrides.dummy_clock_change_latency_ns - && dc->bb_overrides.dummy_clock_change_latency_ns) { - dcn3_2_soc.dummy_pstate_latency_us = - dc->bb_overrides.dummy_clock_change_latency_ns / 1000.0; - } + if ((int)(dcn3_2_soc.fclk_change_latency_us * 1000) + != dc->bb_overrides.fclk_clock_change_latency_ns + && dc->bb_overrides.fclk_clock_change_latency_ns) { + dcn3_2_soc.fclk_change_latency_us = + dc->bb_overrides.fclk_clock_change_latency_ns / 1000; + } - /* Override from VBIOS if VBIOS bb_info available */ - if (dc->ctx->dc_bios->funcs->get_soc_bb_info) { - struct bp_soc_bb_info bb_info = {0}; + if ((int)(dcn3_2_soc.dummy_pstate_latency_us * 1000) + != dc->bb_overrides.dummy_clock_change_latency_ns + && dc->bb_overrides.dummy_clock_change_latency_ns) { + dcn3_2_soc.dummy_pstate_latency_us = + dc->bb_overrides.dummy_clock_change_latency_ns / 1000.0; + } - if (dc->ctx->dc_bios->funcs->get_soc_bb_info(dc->ctx->dc_bios, &bb_info) == BP_RESULT_OK) { - if (bb_info.dram_clock_change_latency_100ns > 0) - dcn3_2_soc.dram_clock_change_latency_us = - bb_info.dram_clock_change_latency_100ns * 10; + /* Override from VBIOS if VBIOS bb_info available */ + if (dc->ctx->dc_bios->funcs->get_soc_bb_info) { + struct bp_soc_bb_info bb_info = {0}; - if (bb_info.dram_sr_enter_exit_latency_100ns > 0) - dcn3_2_soc.sr_enter_plus_exit_time_us = - bb_info.dram_sr_enter_exit_latency_100ns * 10; + if (dc->ctx->dc_bios->funcs->get_soc_bb_info(dc->ctx->dc_bios, &bb_info) == BP_RESULT_OK) { + if (bb_info.dram_clock_change_latency_100ns > 0) + dcn3_2_soc.dram_clock_change_latency_us = + bb_info.dram_clock_change_latency_100ns * 10; - if (bb_info.dram_sr_exit_latency_100ns > 0) - dcn3_2_soc.sr_exit_time_us = - bb_info.dram_sr_exit_latency_100ns * 10; - } - } + if (bb_info.dram_sr_enter_exit_latency_100ns > 0) + dcn3_2_soc.sr_enter_plus_exit_time_us = + bb_info.dram_sr_enter_exit_latency_100ns * 10; - /* Override from VBIOS for num_chan */ - if (dc->ctx->dc_bios->vram_info.num_chans) { - dcn3_2_soc.num_chans = dc->ctx->dc_bios->vram_info.num_chans; - dcn3_2_soc.mall_allocated_for_dcn_mbytes = (double)(dcn32_calc_num_avail_chans_for_mall(dc, - dc->ctx->dc_bios->vram_info.num_chans) * dc->caps.mall_size_per_mem_channel); + if (bb_info.dram_sr_exit_latency_100ns > 0) + dcn3_2_soc.sr_exit_time_us = + bb_info.dram_sr_exit_latency_100ns * 10; } + } - if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes) - dcn3_2_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes; + /* Override from VBIOS for num_chan */ + if (dc->ctx->dc_bios->vram_info.num_chans) { + dcn3_2_soc.num_chans = dc->ctx->dc_bios->vram_info.num_chans; + dcn3_2_soc.mall_allocated_for_dcn_mbytes = (double)(dcn32_calc_num_avail_chans_for_mall(dc, + dc->ctx->dc_bios->vram_info.num_chans) * dc->caps.mall_size_per_mem_channel); } + if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes) + dcn3_2_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes; + /* DML DSC delay factor workaround */ dcn3_2_ip.dsc_delay_factor_wa = dc->debug.dsc_delay_factor_wa_x1000 / 1000.0; @@ -2592,7 +2695,7 @@ void dcn32_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_pa dc->dml.soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0; /* Overrides Clock levelsfrom CLK Mgr table entries as reported by PM FW */ - if ((!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) && (bw_params->clk_table.entries[0].memclk_mhz)) { + if (bw_params->clk_table.entries[0].memclk_mhz) { if (dc->debug.use_legacy_soc_bb_mechanism) { unsigned int i = 0, j = 0, num_states = 0; @@ -2736,7 +2839,8 @@ void dcn32_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_pa dcn3_2_soc.clock_limits[i].phyclk_d32_mhz = dcn3_2_soc.clock_limits[0].phyclk_d32_mhz; } } else { - build_synthetic_soc_states(bw_params, dcn3_2_soc.clock_limits, &dcn3_2_soc.num_states); + build_synthetic_soc_states(dc->debug.disable_dc_mode_overwrite, bw_params, + dcn3_2_soc.clock_limits, &dcn3_2_soc.num_states); } /* Re-init DML with updated bb */ @@ -2783,15 +2887,76 @@ bool dcn32_allow_subvp_with_active_margin(struct pipe_ctx *pipe) } /** - * ******************************************************************************************* - * dcn32_determine_max_vratio_prefetch: Determine max Vratio for prefetch by driver policy + * dcn32_allow_subvp_high_refresh_rate: Determine if the high refresh rate config will allow subvp * - * @param [in]: dc: Current DC state - * @param [in]: context: New DC state to be programmed + * @dc: Current DC state + * @context: New DC state to be programmed + * @pipe: Pipe to be considered for use in subvp * - * @return: Max vratio for prefetch + * On high refresh rate display configs, we will allow subvp under the following conditions: + * 1. Resolution is 3840x2160, 3440x1440, or 2560x1440 + * 2. Refresh rate is between 120hz - 165hz + * 3. No scaling + * 4. Freesync is inactive + * 5. For single display cases, freesync must be disabled * - * ******************************************************************************************* + * Return: True if pipe can be used for subvp, false otherwise + */ +bool dcn32_allow_subvp_high_refresh_rate(struct dc *dc, struct dc_state *context, struct pipe_ctx *pipe) +{ + bool allow = false; + uint32_t refresh_rate = 0; + uint32_t subvp_min_refresh = subvp_high_refresh_list.min_refresh; + uint32_t subvp_max_refresh = subvp_high_refresh_list.max_refresh; + uint32_t min_refresh = subvp_max_refresh; + uint32_t i; + + /* Only allow SubVP on high refresh displays if all connected displays + * are considered "high refresh" (i.e. >= 120hz). We do not want to + * allow combinations such as 120hz (SubVP) + 60hz (SubVP). + */ + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + + if (!pipe_ctx->stream) + continue; + refresh_rate = (pipe_ctx->stream->timing.pix_clk_100hz * 100 + + pipe_ctx->stream->timing.v_total * pipe_ctx->stream->timing.h_total - 1) + / (double)(pipe_ctx->stream->timing.v_total * pipe_ctx->stream->timing.h_total); + + if (refresh_rate < min_refresh) + min_refresh = refresh_rate; + } + + if (!dc->debug.disable_subvp_high_refresh && min_refresh >= subvp_min_refresh && pipe->stream && + pipe->plane_state && !(pipe->stream->vrr_active_variable || pipe->stream->vrr_active_fixed)) { + refresh_rate = (pipe->stream->timing.pix_clk_100hz * 100 + + pipe->stream->timing.v_total * pipe->stream->timing.h_total - 1) + / (double)(pipe->stream->timing.v_total * pipe->stream->timing.h_total); + if (refresh_rate >= subvp_min_refresh && refresh_rate <= subvp_max_refresh) { + for (i = 0; i < SUBVP_HIGH_REFRESH_LIST_LEN; i++) { + uint32_t width = subvp_high_refresh_list.res[i].width; + uint32_t height = subvp_high_refresh_list.res[i].height; + + if (dcn32_check_native_scaling_for_res(pipe, width, height)) { + if ((context->stream_count == 1 && !pipe->stream->allow_freesync) || context->stream_count > 1) { + allow = true; + break; + } + } + } + } + } + return allow; +} + +/** + * dcn32_determine_max_vratio_prefetch: Determine max Vratio for prefetch by driver policy + * + * @dc: Current DC state + * @context: New DC state to be programmed + * + * Return: Max vratio for prefetch */ double dcn32_determine_max_vratio_prefetch(struct dc *dc, struct dc_state *context) { @@ -2821,9 +2986,9 @@ double dcn32_determine_max_vratio_prefetch(struct dc *dc, struct dc_state *conte * ActiveMargin <= 0 to be the FPO stream candidate if found. * * - * @param [in]: dc - current dc state - * @param [in]: context - new dc state - * @param [out]: fpo_candidate_stream - pointer to FPO stream candidate if one is found + * @dc: current dc state + * @context: new dc state + * @fpo_candidate_stream: pointer to FPO stream candidate if one is found * * Return: void */ @@ -2849,10 +3014,9 @@ void dcn32_assign_fpo_vactive_candidate(struct dc *dc, const struct dc_state *co /** * dcn32_find_vactive_pipe - Determines if the config has a pipe that can switch in VACTIVE * - * @param [in]: dc - current dc state - * @param [in]: context - new dc state - * @param [in]: vactive_margin_req_us - The vactive marign required for a vactive pipe to be - * considered "found" + * @dc: current dc state + * @context: new dc state + * @vactive_margin_req_us: The vactive marign required for a vactive pipe to be considered "found" * * Return: True if VACTIVE display is found, false otherwise */ @@ -2861,6 +3025,7 @@ bool dcn32_find_vactive_pipe(struct dc *dc, const struct dc_state *context, uint unsigned int i, pipe_idx; const struct vba_vars_st *vba = &context->bw_ctx.dml.vba; bool vactive_found = false; + unsigned int blank_us = 0; for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) { const struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; @@ -2868,7 +3033,10 @@ bool dcn32_find_vactive_pipe(struct dc *dc, const struct dc_state *context, uint if (!pipe->stream) continue; - if (vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] >= vactive_margin_req_us) { + blank_us = ((pipe->stream->timing.v_total - pipe->stream->timing.v_addressable) * pipe->stream->timing.h_total / + (double)(pipe->stream->timing.pix_clk_100hz * 100)) * 1000000; + if (vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] >= vactive_margin_req_us && + !(pipe->stream->vrr_active_variable || pipe->stream->vrr_active_fixed) && blank_us < dc->debug.fpo_vactive_max_blank_us) { vactive_found = true; break; } @@ -2882,3 +3050,18 @@ void dcn32_set_clock_limits(const struct _vcs_dpi_soc_bounding_box_st *soc_bb) dc_assert_fp_enabled(); dcn3_2_soc.clock_limits[0].dcfclk_mhz = 1200.0; } + +void dcn32_override_min_req_memclk(struct dc *dc, struct dc_state *context) +{ + // WA: restrict FPO and SubVP to use first non-strobe mode (DCN32 BW issue) + if ((context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching || dcn32_subvp_in_use(dc, context)) && + dc->dml.soc.num_chans <= 8) { + int num_mclk_levels = dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_memclk_levels; + + if (context->bw_ctx.dml.vba.DRAMSpeed <= dc->clk_mgr->bw_params->clk_table.entries[0].memclk_mhz * 16 && + num_mclk_levels > 1) { + context->bw_ctx.dml.vba.DRAMSpeed = dc->clk_mgr->bw_params->clk_table.entries[1].memclk_mhz * 16; + context->bw_ctx.bw.dcn.clk.dramclk_khz = context->bw_ctx.dml.vba.DRAMSpeed * 1000 / 16; + } + } +} diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h index dcf512cd30721645033a86dae9be51ed44d7aa16..a4206b71d650a99c24f7c543972618132f67aaca 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h @@ -80,6 +80,8 @@ void dcn32_assign_fpo_vactive_candidate(struct dc *dc, const struct dc_state *co bool dcn32_find_vactive_pipe(struct dc *dc, const struct dc_state *context, uint32_t vactive_margin_req); +void dcn32_override_min_req_memclk(struct dc *dc, struct dc_state *context); + void dcn32_set_clock_limits(const struct _vcs_dpi_soc_bounding_box_st *soc_bb); #endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c index d75248b6cae997cbe359b8ea9ed551a3c7f0687d..cbdfb762c10c58427c70d2e618b38527eafc9878 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c @@ -811,7 +811,7 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman v->SwathHeightC[k], TWait, (v->DRAMSpeedPerState[mode_lib->vba.VoltageLevel] <= MEM_STROBE_FREQ_MHZ || - v->DCFCLKPerState[mode_lib->vba.VoltageLevel] <= MIN_DCFCLK_FREQ_MHZ) ? + v->DCFCLKPerState[mode_lib->vba.VoltageLevel] <= DCFCLK_FREQ_EXTRA_PREFETCH_REQ_MHZ) ? mode_lib->vba.ip.min_prefetch_in_strobe_us : 0, /* Output */ &v->DSTXAfterScaler[k], @@ -2323,10 +2323,14 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l mode_lib->vba.LinkCapacitySupport[i] = true; for (k = 0; k < mode_lib->vba.NumberOfActiveSurfaces; ++k) { if (mode_lib->vba.BlendingAndTiming[k] == k - && (mode_lib->vba.Output[k] == dm_dp || mode_lib->vba.Output[k] == dm_dp2p0 - || mode_lib->vba.Output[k] == dm_edp - || mode_lib->vba.Output[k] == dm_hdmi) - && mode_lib->vba.OutputBppPerState[i][k] == 0) { + && (mode_lib->vba.Output[k] == dm_dp || mode_lib->vba.Output[k] == dm_dp2p0 + || mode_lib->vba.Output[k] == dm_edp + || mode_lib->vba.Output[k] == dm_hdmi) + && mode_lib->vba.OutputBppPerState[i][k] == 0 && + (mode_lib->vba.UsesMALLForPStateChange[k] != dm_use_mall_pstate_change_phantom_pipe)) { + /* Phantom pipes don't consider DSC in DML, so it could fail link check. + * However, we don't care about the link for phantom pipes. + */ mode_lib->vba.LinkCapacitySupport[i] = false; } } @@ -3311,7 +3315,7 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l v->swath_width_chroma_ub_this_state[k], v->SwathHeightYThisState[k], v->SwathHeightCThisState[k], v->TWait, - (v->DRAMSpeedPerState[i] <= MEM_STROBE_FREQ_MHZ || v->DCFCLKState[i][j] <= MIN_DCFCLK_FREQ_MHZ) ? + (v->DRAMSpeedPerState[i] <= MEM_STROBE_FREQ_MHZ || v->DCFCLKState[i][j] <= DCFCLK_FREQ_EXTRA_PREFETCH_REQ_MHZ) ? mode_lib->vba.ip.min_prefetch_in_strobe_us : 0, /* Output */ diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h index d98e36a9a09ccfba3267e18bf4bcc5e024b2307c..c4745d63039bb4069b547b85b1baedb612a02c11 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h @@ -53,7 +53,7 @@ #define BPP_BLENDED_PIPE 0xffffffff #define MEM_STROBE_FREQ_MHZ 1600 -#define MIN_DCFCLK_FREQ_MHZ 200 +#define DCFCLK_FREQ_EXTRA_PREFETCH_REQ_MHZ 300 #define MEM_STROBE_MAX_DELIVERY_TIME_US 60.0 struct display_mode_lib; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c index 61cc4904ade4192c8d1e3fe9ca8fc02848d953af..a50e7f4dce4215f1b5f537186f9185dbd7f146b6 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c @@ -1595,7 +1595,6 @@ double dml32_TruncToValidBPP( unsigned int NonDSCBPP0; unsigned int NonDSCBPP1; unsigned int NonDSCBPP2; - unsigned int NonDSCBPP3; if (Format == dm_420) { NonDSCBPP0 = 12; @@ -1604,10 +1603,9 @@ double dml32_TruncToValidBPP( MinDSCBPP = 6; MaxDSCBPP = 1.5 * DSCInputBitPerComponent - 1 / 16; } else if (Format == dm_444) { - NonDSCBPP0 = 18; - NonDSCBPP1 = 24; - NonDSCBPP2 = 30; - NonDSCBPP3 = 36; + NonDSCBPP0 = 24; + NonDSCBPP1 = 30; + NonDSCBPP2 = 36; MinDSCBPP = 8; MaxDSCBPP = 3 * DSCInputBitPerComponent - 1.0 / 16; } else { @@ -1661,9 +1659,7 @@ double dml32_TruncToValidBPP( else return dml_floor(16.0 * MaxLinkBPP, 1.0) / 16.0; } else { - if (MaxLinkBPP >= NonDSCBPP3) - return NonDSCBPP3; - else if (MaxLinkBPP >= NonDSCBPP2) + if (MaxLinkBPP >= NonDSCBPP2) return NonDSCBPP2; else if (MaxLinkBPP >= NonDSCBPP1) return NonDSCBPP1; @@ -1674,7 +1670,7 @@ double dml32_TruncToValidBPP( } } else { if (!((DSCEnable == false && (DesiredBPP == NonDSCBPP2 || DesiredBPP == NonDSCBPP1 || - DesiredBPP == NonDSCBPP0 || DesiredBPP == NonDSCBPP3)) || + DesiredBPP <= NonDSCBPP0)) || (DSCEnable && DesiredBPP >= MinDSCBPP && DesiredBPP <= MaxDSCBPP))) return BPP_INVALID; else @@ -4342,7 +4338,7 @@ void dml32_CalculateWatermarksMALLUseAndDRAMSpeedChangeSupport( + v->WritebackChunkSize * 1024.0 / 32.0 / SOCCLK; } if (v->USRRetrainingRequiredFinal) - v->Watermark.WritebackUrgentWatermark = v->Watermark.WritebackUrgentWatermark + v->Watermark.WritebackDRAMClockChangeWatermark = v->Watermark.WritebackDRAMClockChangeWatermark + mmSOCParameters.USRRetrainingLatency; if (TotalActiveWriteback <= 1) { @@ -4660,6 +4656,10 @@ void dml32_CalculateMinAndMaxPrefetchMode( } else if (AllowForPStateChangeOrStutterInVBlankFinal == dm_prefetch_support_uclk_fclk_and_stutter) { *MinPrefetchMode = 0; *MaxPrefetchMode = 0; + } else if (AllowForPStateChangeOrStutterInVBlankFinal == + dm_prefetch_support_uclk_fclk_and_stutter_if_possible) { + *MinPrefetchMode = 0; + *MaxPrefetchMode = 3; } else { *MinPrefetchMode = 0; *MaxPrefetchMode = 3; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c index 395ae8761980ff1fe3f73a9d1d088559ac9c6b1c..9ba6cb67655f4aff436432ef9df5f2e2d37952f3 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c @@ -116,7 +116,7 @@ void dml32_rq_dlg_get_rq_reg(display_rq_regs_st *rq_regs, else rq_regs->rq_regs_l.min_meta_chunk_size = dml_log2(min_meta_chunk_bytes) - 6 + 1; - if (min_meta_chunk_bytes == 0) + if (p1_min_meta_chunk_bytes == 0) rq_regs->rq_regs_c.min_meta_chunk_size = 0; else rq_regs->rq_regs_c.min_meta_chunk_size = dml_log2(p1_min_meta_chunk_bytes) - 6 + 1; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c index 342a1bcb492735d1c776f512103297316a6eeace..f0683fd9d3f06d2b78f109558b363d4072d3eaa6 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c @@ -252,14 +252,48 @@ static void remove_entry_from_table_at_index(struct _vcs_dpi_voltage_scaling_st memset(&table[--(*num_entries)], 0, sizeof(struct _vcs_dpi_voltage_scaling_st)); } -static int build_synthetic_soc_states(struct clk_bw_params *bw_params, +/* + * override_max_clk_values - Overwrite the max clock frequencies with the max DC mode timings + * Input: + * max_clk_limit - struct containing the desired clock timings + * Output: + * curr_clk_limit - struct containing the timings that need to be overwritten + * Return: 0 upon success, non-zero for failure + */ +static int override_max_clk_values(struct clk_limit_table_entry *max_clk_limit, + struct clk_limit_table_entry *curr_clk_limit) +{ + if (NULL == max_clk_limit || NULL == curr_clk_limit) + return -1; //invalid parameters + + //only overwrite if desired max clock frequency is initialized + if (max_clk_limit->dcfclk_mhz != 0) + curr_clk_limit->dcfclk_mhz = max_clk_limit->dcfclk_mhz; + + if (max_clk_limit->fclk_mhz != 0) + curr_clk_limit->fclk_mhz = max_clk_limit->fclk_mhz; + + if (max_clk_limit->memclk_mhz != 0) + curr_clk_limit->memclk_mhz = max_clk_limit->memclk_mhz; + + if (max_clk_limit->socclk_mhz != 0) + curr_clk_limit->socclk_mhz = max_clk_limit->socclk_mhz; + + if (max_clk_limit->dtbclk_mhz != 0) + curr_clk_limit->dtbclk_mhz = max_clk_limit->dtbclk_mhz; + + if (max_clk_limit->dispclk_mhz != 0) + curr_clk_limit->dispclk_mhz = max_clk_limit->dispclk_mhz; + + return 0; +} + +static int build_synthetic_soc_states(bool disable_dc_mode_overwrite, struct clk_bw_params *bw_params, struct _vcs_dpi_voltage_scaling_st *table, unsigned int *num_entries) { int i, j; struct _vcs_dpi_voltage_scaling_st entry = {0}; - - unsigned int max_dcfclk_mhz = 0, max_dispclk_mhz = 0, max_dppclk_mhz = 0, - max_phyclk_mhz = 0, max_dtbclk_mhz = 0, max_fclk_mhz = 0, max_uclk_mhz = 0; + struct clk_limit_table_entry max_clk_data = {0}; unsigned int min_dcfclk_mhz = 199, min_fclk_mhz = 299; @@ -270,53 +304,78 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params, unsigned int num_fclk_dpms = 0; unsigned int num_dcfclk_dpms = 0; + unsigned int num_dc_uclk_dpms = 0; + unsigned int num_dc_fclk_dpms = 0; + unsigned int num_dc_dcfclk_dpms = 0; + for (i = 0; i < MAX_NUM_DPM_LVL; i++) { - if (bw_params->clk_table.entries[i].dcfclk_mhz > max_dcfclk_mhz) - max_dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz; - if (bw_params->clk_table.entries[i].fclk_mhz > max_fclk_mhz) - max_fclk_mhz = bw_params->clk_table.entries[i].fclk_mhz; - if (bw_params->clk_table.entries[i].memclk_mhz > max_uclk_mhz) - max_uclk_mhz = bw_params->clk_table.entries[i].memclk_mhz; - if (bw_params->clk_table.entries[i].dispclk_mhz > max_dispclk_mhz) - max_dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz; - if (bw_params->clk_table.entries[i].dppclk_mhz > max_dppclk_mhz) - max_dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz; - if (bw_params->clk_table.entries[i].phyclk_mhz > max_phyclk_mhz) - max_phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz; - if (bw_params->clk_table.entries[i].dtbclk_mhz > max_dtbclk_mhz) - max_dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz; - - if (bw_params->clk_table.entries[i].memclk_mhz > 0) + if (bw_params->clk_table.entries[i].dcfclk_mhz > max_clk_data.dcfclk_mhz) + max_clk_data.dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz; + if (bw_params->clk_table.entries[i].fclk_mhz > max_clk_data.fclk_mhz) + max_clk_data.fclk_mhz = bw_params->clk_table.entries[i].fclk_mhz; + if (bw_params->clk_table.entries[i].memclk_mhz > max_clk_data.memclk_mhz) + max_clk_data.memclk_mhz = bw_params->clk_table.entries[i].memclk_mhz; + if (bw_params->clk_table.entries[i].dispclk_mhz > max_clk_data.dispclk_mhz) + max_clk_data.dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz; + if (bw_params->clk_table.entries[i].dppclk_mhz > max_clk_data.dppclk_mhz) + max_clk_data.dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz; + if (bw_params->clk_table.entries[i].phyclk_mhz > max_clk_data.phyclk_mhz) + max_clk_data.phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz; + if (bw_params->clk_table.entries[i].dtbclk_mhz > max_clk_data.dtbclk_mhz) + max_clk_data.dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz; + + if (bw_params->clk_table.entries[i].memclk_mhz > 0) { num_uclk_dpms++; - if (bw_params->clk_table.entries[i].fclk_mhz > 0) + if (bw_params->clk_table.entries[i].memclk_mhz <= bw_params->dc_mode_limit.memclk_mhz) + num_dc_uclk_dpms++; + } + if (bw_params->clk_table.entries[i].fclk_mhz > 0) { num_fclk_dpms++; - if (bw_params->clk_table.entries[i].dcfclk_mhz > 0) + if (bw_params->clk_table.entries[i].fclk_mhz <= bw_params->dc_mode_limit.fclk_mhz) + num_dc_fclk_dpms++; + } + if (bw_params->clk_table.entries[i].dcfclk_mhz > 0) { num_dcfclk_dpms++; + if (bw_params->clk_table.entries[i].dcfclk_mhz <= bw_params->dc_mode_limit.dcfclk_mhz) + num_dc_dcfclk_dpms++; + } + } + + if (!disable_dc_mode_overwrite) { + //Overwrite max frequencies with max DC mode frequencies for DC mode systems + override_max_clk_values(&bw_params->dc_mode_limit, &max_clk_data); + num_uclk_dpms = num_dc_uclk_dpms; + num_fclk_dpms = num_dc_fclk_dpms; + num_dcfclk_dpms = num_dc_dcfclk_dpms; + bw_params->clk_table.num_entries_per_clk.num_memclk_levels = num_uclk_dpms; + bw_params->clk_table.num_entries_per_clk.num_fclk_levels = num_fclk_dpms; } if (num_dcfclk_dpms > 0 && bw_params->clk_table.entries[0].fclk_mhz > min_fclk_mhz) min_fclk_mhz = bw_params->clk_table.entries[0].fclk_mhz; - if (!max_dcfclk_mhz || !max_dispclk_mhz || !max_dtbclk_mhz) + if (!max_clk_data.dcfclk_mhz || !max_clk_data.dispclk_mhz || !max_clk_data.dtbclk_mhz) return -1; - if (max_dppclk_mhz == 0) - max_dppclk_mhz = max_dispclk_mhz; + if (max_clk_data.dppclk_mhz == 0) + max_clk_data.dppclk_mhz = max_clk_data.dispclk_mhz; - if (max_fclk_mhz == 0) - max_fclk_mhz = max_dcfclk_mhz * dcn3_21_soc.pct_ideal_sdp_bw_after_urgent / dcn3_21_soc.pct_ideal_fabric_bw_after_urgent; + if (max_clk_data.fclk_mhz == 0) + max_clk_data.fclk_mhz = max_clk_data.dcfclk_mhz * + dcn3_2_soc.pct_ideal_sdp_bw_after_urgent / + dcn3_2_soc.pct_ideal_fabric_bw_after_urgent; - if (max_phyclk_mhz == 0) - max_phyclk_mhz = dcn3_21_soc.clock_limits[0].phyclk_mhz; + if (max_clk_data.phyclk_mhz == 0) + max_clk_data.phyclk_mhz = dcn3_2_soc.clock_limits[0].phyclk_mhz; *num_entries = 0; - entry.dispclk_mhz = max_dispclk_mhz; - entry.dscclk_mhz = max_dispclk_mhz / 3; - entry.dppclk_mhz = max_dppclk_mhz; - entry.dtbclk_mhz = max_dtbclk_mhz; - entry.phyclk_mhz = max_phyclk_mhz; - entry.phyclk_d18_mhz = dcn3_21_soc.clock_limits[0].phyclk_d18_mhz; - entry.phyclk_d32_mhz = dcn3_21_soc.clock_limits[0].phyclk_d32_mhz; + entry.dispclk_mhz = max_clk_data.dispclk_mhz; + entry.dscclk_mhz = max_clk_data.dispclk_mhz / 3; + entry.dppclk_mhz = max_clk_data.dppclk_mhz; + entry.dtbclk_mhz = max_clk_data.dtbclk_mhz; + entry.phyclk_mhz = max_clk_data.phyclk_mhz; + entry.phyclk_d18_mhz = dcn3_2_soc.clock_limits[0].phyclk_d18_mhz; + entry.phyclk_d32_mhz = dcn3_2_soc.clock_limits[0].phyclk_d32_mhz; // Insert all the DCFCLK STAs for (i = 0; i < num_dcfclk_stas; i++) { @@ -328,7 +387,7 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params, } // Insert the max DCFCLK - entry.dcfclk_mhz = max_dcfclk_mhz; + entry.dcfclk_mhz = max_clk_data.dcfclk_mhz; entry.fabricclk_mhz = 0; entry.dram_speed_mts = 0; @@ -356,7 +415,7 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params, // If FCLK fine grained, only insert max else { entry.dcfclk_mhz = 0; - entry.fabricclk_mhz = max_fclk_mhz; + entry.fabricclk_mhz = max_clk_data.fclk_mhz; entry.dram_speed_mts = 0; dcn321_insert_entry_into_table_sorted(table, num_entries, &entry); @@ -368,9 +427,9 @@ static int build_synthetic_soc_states(struct clk_bw_params *bw_params, // Remove states that require higher clocks than are supported for (i = *num_entries - 1; i >= 0 ; i--) { - if (table[i].dcfclk_mhz > max_dcfclk_mhz || - table[i].fabricclk_mhz > max_fclk_mhz || - table[i].dram_speed_mts > max_uclk_mhz * 16) + if (table[i].dcfclk_mhz > max_clk_data.dcfclk_mhz || + table[i].fabricclk_mhz > max_clk_data.fclk_mhz || + table[i].dram_speed_mts > max_clk_data.memclk_mhz * 16) remove_entry_from_table_at_index(table, num_entries, i); } @@ -471,80 +530,78 @@ static void dcn321_get_optimal_dcfclk_fclk_for_uclk(unsigned int uclk_mts, void dcn321_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params) { dc_assert_fp_enabled(); - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - /* Overrides from dc->config options */ - dcn3_21_ip.clamp_min_dcfclk = dc->config.clamp_min_dcfclk; - - /* Override from passed dc->bb_overrides if available*/ - if ((int)(dcn3_21_soc.sr_exit_time_us * 1000) != dc->bb_overrides.sr_exit_time_ns - && dc->bb_overrides.sr_exit_time_ns) { - dcn3_21_soc.sr_exit_time_us = dc->bb_overrides.sr_exit_time_ns / 1000.0; - } + /* Overrides from dc->config options */ + dcn3_21_ip.clamp_min_dcfclk = dc->config.clamp_min_dcfclk; - if ((int)(dcn3_21_soc.sr_enter_plus_exit_time_us * 1000) - != dc->bb_overrides.sr_enter_plus_exit_time_ns - && dc->bb_overrides.sr_enter_plus_exit_time_ns) { - dcn3_21_soc.sr_enter_plus_exit_time_us = - dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; - } + /* Override from passed dc->bb_overrides if available*/ + if ((int)(dcn3_21_soc.sr_exit_time_us * 1000) != dc->bb_overrides.sr_exit_time_ns + && dc->bb_overrides.sr_exit_time_ns) { + dcn3_21_soc.sr_exit_time_us = dc->bb_overrides.sr_exit_time_ns / 1000.0; + } - if ((int)(dcn3_21_soc.urgent_latency_us * 1000) != dc->bb_overrides.urgent_latency_ns - && dc->bb_overrides.urgent_latency_ns) { - dcn3_21_soc.urgent_latency_us = dc->bb_overrides.urgent_latency_ns / 1000.0; - dcn3_21_soc.urgent_latency_pixel_data_only_us = dc->bb_overrides.urgent_latency_ns / 1000.0; - } + if ((int)(dcn3_21_soc.sr_enter_plus_exit_time_us * 1000) + != dc->bb_overrides.sr_enter_plus_exit_time_ns + && dc->bb_overrides.sr_enter_plus_exit_time_ns) { + dcn3_21_soc.sr_enter_plus_exit_time_us = + dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; + } - if ((int)(dcn3_21_soc.dram_clock_change_latency_us * 1000) - != dc->bb_overrides.dram_clock_change_latency_ns - && dc->bb_overrides.dram_clock_change_latency_ns) { - dcn3_21_soc.dram_clock_change_latency_us = - dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; - } + if ((int)(dcn3_21_soc.urgent_latency_us * 1000) != dc->bb_overrides.urgent_latency_ns + && dc->bb_overrides.urgent_latency_ns) { + dcn3_21_soc.urgent_latency_us = dc->bb_overrides.urgent_latency_ns / 1000.0; + dcn3_21_soc.urgent_latency_pixel_data_only_us = dc->bb_overrides.urgent_latency_ns / 1000.0; + } - if ((int)(dcn3_21_soc.fclk_change_latency_us * 1000) - != dc->bb_overrides.fclk_clock_change_latency_ns - && dc->bb_overrides.fclk_clock_change_latency_ns) { - dcn3_21_soc.fclk_change_latency_us = - dc->bb_overrides.fclk_clock_change_latency_ns / 1000; - } + if ((int)(dcn3_21_soc.dram_clock_change_latency_us * 1000) + != dc->bb_overrides.dram_clock_change_latency_ns + && dc->bb_overrides.dram_clock_change_latency_ns) { + dcn3_21_soc.dram_clock_change_latency_us = + dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; + } - if ((int)(dcn3_21_soc.dummy_pstate_latency_us * 1000) - != dc->bb_overrides.dummy_clock_change_latency_ns - && dc->bb_overrides.dummy_clock_change_latency_ns) { - dcn3_21_soc.dummy_pstate_latency_us = - dc->bb_overrides.dummy_clock_change_latency_ns / 1000.0; - } + if ((int)(dcn3_21_soc.fclk_change_latency_us * 1000) + != dc->bb_overrides.fclk_clock_change_latency_ns + && dc->bb_overrides.fclk_clock_change_latency_ns) { + dcn3_21_soc.fclk_change_latency_us = + dc->bb_overrides.fclk_clock_change_latency_ns / 1000; + } - /* Override from VBIOS if VBIOS bb_info available */ - if (dc->ctx->dc_bios->funcs->get_soc_bb_info) { - struct bp_soc_bb_info bb_info = {0}; + if ((int)(dcn3_21_soc.dummy_pstate_latency_us * 1000) + != dc->bb_overrides.dummy_clock_change_latency_ns + && dc->bb_overrides.dummy_clock_change_latency_ns) { + dcn3_21_soc.dummy_pstate_latency_us = + dc->bb_overrides.dummy_clock_change_latency_ns / 1000.0; + } - if (dc->ctx->dc_bios->funcs->get_soc_bb_info(dc->ctx->dc_bios, &bb_info) == BP_RESULT_OK) { - if (bb_info.dram_clock_change_latency_100ns > 0) - dcn3_21_soc.dram_clock_change_latency_us = - bb_info.dram_clock_change_latency_100ns * 10; + /* Override from VBIOS if VBIOS bb_info available */ + if (dc->ctx->dc_bios->funcs->get_soc_bb_info) { + struct bp_soc_bb_info bb_info = {0}; - if (bb_info.dram_sr_enter_exit_latency_100ns > 0) - dcn3_21_soc.sr_enter_plus_exit_time_us = - bb_info.dram_sr_enter_exit_latency_100ns * 10; + if (dc->ctx->dc_bios->funcs->get_soc_bb_info(dc->ctx->dc_bios, &bb_info) == BP_RESULT_OK) { + if (bb_info.dram_clock_change_latency_100ns > 0) + dcn3_21_soc.dram_clock_change_latency_us = + bb_info.dram_clock_change_latency_100ns * 10; - if (bb_info.dram_sr_exit_latency_100ns > 0) - dcn3_21_soc.sr_exit_time_us = - bb_info.dram_sr_exit_latency_100ns * 10; - } - } + if (bb_info.dram_sr_enter_exit_latency_100ns > 0) + dcn3_21_soc.sr_enter_plus_exit_time_us = + bb_info.dram_sr_enter_exit_latency_100ns * 10; - /* Override from VBIOS for num_chan */ - if (dc->ctx->dc_bios->vram_info.num_chans) { - dcn3_21_soc.num_chans = dc->ctx->dc_bios->vram_info.num_chans; - dcn3_21_soc.mall_allocated_for_dcn_mbytes = (double)(dcn32_calc_num_avail_chans_for_mall(dc, - dc->ctx->dc_bios->vram_info.num_chans) * dc->caps.mall_size_per_mem_channel); + if (bb_info.dram_sr_exit_latency_100ns > 0) + dcn3_21_soc.sr_exit_time_us = + bb_info.dram_sr_exit_latency_100ns * 10; } + } - if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes) - dcn3_21_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes; + /* Override from VBIOS for num_chan */ + if (dc->ctx->dc_bios->vram_info.num_chans) { + dcn3_21_soc.num_chans = dc->ctx->dc_bios->vram_info.num_chans; + dcn3_21_soc.mall_allocated_for_dcn_mbytes = (double)(dcn32_calc_num_avail_chans_for_mall(dc, + dc->ctx->dc_bios->vram_info.num_chans) * dc->caps.mall_size_per_mem_channel); } + if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes) + dcn3_21_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes; + /* DML DSC delay factor workaround */ dcn3_21_ip.dsc_delay_factor_wa = dc->debug.dsc_delay_factor_wa_x1000 / 1000.0; @@ -555,150 +612,149 @@ void dcn321_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p dc->dml.soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0; /* Overrides Clock levelsfrom CLK Mgr table entries as reported by PM FW */ - if ((!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) && (bw_params->clk_table.entries[0].memclk_mhz)) { - if (dc->debug.use_legacy_soc_bb_mechanism) { - unsigned int i = 0, j = 0, num_states = 0; - - unsigned int dcfclk_mhz[DC__VOLTAGE_STATES] = {0}; - unsigned int dram_speed_mts[DC__VOLTAGE_STATES] = {0}; - unsigned int optimal_uclk_for_dcfclk_sta_targets[DC__VOLTAGE_STATES] = {0}; - unsigned int optimal_dcfclk_for_uclk[DC__VOLTAGE_STATES] = {0}; - - unsigned int dcfclk_sta_targets[DC__VOLTAGE_STATES] = {615, 906, 1324, 1564}; - unsigned int num_dcfclk_sta_targets = 4, num_uclk_states = 0; - unsigned int max_dcfclk_mhz = 0, max_dispclk_mhz = 0, max_dppclk_mhz = 0, max_phyclk_mhz = 0; - - for (i = 0; i < MAX_NUM_DPM_LVL; i++) { - if (bw_params->clk_table.entries[i].dcfclk_mhz > max_dcfclk_mhz) - max_dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz; - if (bw_params->clk_table.entries[i].dispclk_mhz > max_dispclk_mhz) - max_dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz; - if (bw_params->clk_table.entries[i].dppclk_mhz > max_dppclk_mhz) - max_dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz; - if (bw_params->clk_table.entries[i].phyclk_mhz > max_phyclk_mhz) - max_phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz; - } - if (!max_dcfclk_mhz) - max_dcfclk_mhz = dcn3_21_soc.clock_limits[0].dcfclk_mhz; - if (!max_dispclk_mhz) - max_dispclk_mhz = dcn3_21_soc.clock_limits[0].dispclk_mhz; - if (!max_dppclk_mhz) - max_dppclk_mhz = dcn3_21_soc.clock_limits[0].dppclk_mhz; - if (!max_phyclk_mhz) - max_phyclk_mhz = dcn3_21_soc.clock_limits[0].phyclk_mhz; - - if (max_dcfclk_mhz > dcfclk_sta_targets[num_dcfclk_sta_targets-1]) { - // If max DCFCLK is greater than the max DCFCLK STA target, insert into the DCFCLK STA target array - dcfclk_sta_targets[num_dcfclk_sta_targets] = max_dcfclk_mhz; - num_dcfclk_sta_targets++; - } else if (max_dcfclk_mhz < dcfclk_sta_targets[num_dcfclk_sta_targets-1]) { - // If max DCFCLK is less than the max DCFCLK STA target, cap values and remove duplicates - for (i = 0; i < num_dcfclk_sta_targets; i++) { - if (dcfclk_sta_targets[i] > max_dcfclk_mhz) { - dcfclk_sta_targets[i] = max_dcfclk_mhz; - break; - } + if (dc->debug.use_legacy_soc_bb_mechanism) { + unsigned int i = 0, j = 0, num_states = 0; + + unsigned int dcfclk_mhz[DC__VOLTAGE_STATES] = {0}; + unsigned int dram_speed_mts[DC__VOLTAGE_STATES] = {0}; + unsigned int optimal_uclk_for_dcfclk_sta_targets[DC__VOLTAGE_STATES] = {0}; + unsigned int optimal_dcfclk_for_uclk[DC__VOLTAGE_STATES] = {0}; + + unsigned int dcfclk_sta_targets[DC__VOLTAGE_STATES] = {615, 906, 1324, 1564}; + unsigned int num_dcfclk_sta_targets = 4, num_uclk_states = 0; + unsigned int max_dcfclk_mhz = 0, max_dispclk_mhz = 0, max_dppclk_mhz = 0, max_phyclk_mhz = 0; + + for (i = 0; i < MAX_NUM_DPM_LVL; i++) { + if (bw_params->clk_table.entries[i].dcfclk_mhz > max_dcfclk_mhz) + max_dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz; + if (bw_params->clk_table.entries[i].dispclk_mhz > max_dispclk_mhz) + max_dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz; + if (bw_params->clk_table.entries[i].dppclk_mhz > max_dppclk_mhz) + max_dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz; + if (bw_params->clk_table.entries[i].phyclk_mhz > max_phyclk_mhz) + max_phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz; + } + if (!max_dcfclk_mhz) + max_dcfclk_mhz = dcn3_21_soc.clock_limits[0].dcfclk_mhz; + if (!max_dispclk_mhz) + max_dispclk_mhz = dcn3_21_soc.clock_limits[0].dispclk_mhz; + if (!max_dppclk_mhz) + max_dppclk_mhz = dcn3_21_soc.clock_limits[0].dppclk_mhz; + if (!max_phyclk_mhz) + max_phyclk_mhz = dcn3_21_soc.clock_limits[0].phyclk_mhz; + + if (max_dcfclk_mhz > dcfclk_sta_targets[num_dcfclk_sta_targets-1]) { + // If max DCFCLK is greater than the max DCFCLK STA target, insert into the DCFCLK STA target array + dcfclk_sta_targets[num_dcfclk_sta_targets] = max_dcfclk_mhz; + num_dcfclk_sta_targets++; + } else if (max_dcfclk_mhz < dcfclk_sta_targets[num_dcfclk_sta_targets-1]) { + // If max DCFCLK is less than the max DCFCLK STA target, cap values and remove duplicates + for (i = 0; i < num_dcfclk_sta_targets; i++) { + if (dcfclk_sta_targets[i] > max_dcfclk_mhz) { + dcfclk_sta_targets[i] = max_dcfclk_mhz; + break; } - // Update size of array since we "removed" duplicates - num_dcfclk_sta_targets = i + 1; } + // Update size of array since we "removed" duplicates + num_dcfclk_sta_targets = i + 1; + } - num_uclk_states = bw_params->clk_table.num_entries; + num_uclk_states = bw_params->clk_table.num_entries; - // Calculate optimal dcfclk for each uclk - for (i = 0; i < num_uclk_states; i++) { - dcn321_get_optimal_dcfclk_fclk_for_uclk(bw_params->clk_table.entries[i].memclk_mhz * 16, - &optimal_dcfclk_for_uclk[i], NULL); - if (optimal_dcfclk_for_uclk[i] < bw_params->clk_table.entries[0].dcfclk_mhz) { - optimal_dcfclk_for_uclk[i] = bw_params->clk_table.entries[0].dcfclk_mhz; - } + // Calculate optimal dcfclk for each uclk + for (i = 0; i < num_uclk_states; i++) { + dcn321_get_optimal_dcfclk_fclk_for_uclk(bw_params->clk_table.entries[i].memclk_mhz * 16, + &optimal_dcfclk_for_uclk[i], NULL); + if (optimal_dcfclk_for_uclk[i] < bw_params->clk_table.entries[0].dcfclk_mhz) { + optimal_dcfclk_for_uclk[i] = bw_params->clk_table.entries[0].dcfclk_mhz; } + } - // Calculate optimal uclk for each dcfclk sta target - for (i = 0; i < num_dcfclk_sta_targets; i++) { - for (j = 0; j < num_uclk_states; j++) { - if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) { - optimal_uclk_for_dcfclk_sta_targets[i] = - bw_params->clk_table.entries[j].memclk_mhz * 16; - break; - } + // Calculate optimal uclk for each dcfclk sta target + for (i = 0; i < num_dcfclk_sta_targets; i++) { + for (j = 0; j < num_uclk_states; j++) { + if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) { + optimal_uclk_for_dcfclk_sta_targets[i] = + bw_params->clk_table.entries[j].memclk_mhz * 16; + break; } } + } - i = 0; - j = 0; - // create the final dcfclk and uclk table - while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) { - if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) { - dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; - dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; + i = 0; + j = 0; + // create the final dcfclk and uclk table + while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) { + if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) { + dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; + dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; + } else { + if (j < num_uclk_states && optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) { + dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j]; + dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16; } else { - if (j < num_uclk_states && optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) { - dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j]; - dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16; - } else { - j = num_uclk_states; - } + j = num_uclk_states; } } + } - while (i < num_dcfclk_sta_targets && num_states < DC__VOLTAGE_STATES) { - dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; - dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; - } + while (i < num_dcfclk_sta_targets && num_states < DC__VOLTAGE_STATES) { + dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; + dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; + } - while (j < num_uclk_states && num_states < DC__VOLTAGE_STATES && - optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) { - dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j]; - dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16; - } + while (j < num_uclk_states && num_states < DC__VOLTAGE_STATES && + optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) { + dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j]; + dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16; + } - dcn3_21_soc.num_states = num_states; - for (i = 0; i < dcn3_21_soc.num_states; i++) { - dcn3_21_soc.clock_limits[i].state = i; - dcn3_21_soc.clock_limits[i].dcfclk_mhz = dcfclk_mhz[i]; - dcn3_21_soc.clock_limits[i].fabricclk_mhz = dcfclk_mhz[i]; - - /* Fill all states with max values of all these clocks */ - dcn3_21_soc.clock_limits[i].dispclk_mhz = max_dispclk_mhz; - dcn3_21_soc.clock_limits[i].dppclk_mhz = max_dppclk_mhz; - dcn3_21_soc.clock_limits[i].phyclk_mhz = max_phyclk_mhz; - dcn3_21_soc.clock_limits[i].dscclk_mhz = max_dispclk_mhz / 3; - - /* Populate from bw_params for DTBCLK, SOCCLK */ - if (i > 0) { - if (!bw_params->clk_table.entries[i].dtbclk_mhz) { - dcn3_21_soc.clock_limits[i].dtbclk_mhz = dcn3_21_soc.clock_limits[i-1].dtbclk_mhz; - } else { - dcn3_21_soc.clock_limits[i].dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz; - } - } else if (bw_params->clk_table.entries[i].dtbclk_mhz) { + dcn3_21_soc.num_states = num_states; + for (i = 0; i < dcn3_21_soc.num_states; i++) { + dcn3_21_soc.clock_limits[i].state = i; + dcn3_21_soc.clock_limits[i].dcfclk_mhz = dcfclk_mhz[i]; + dcn3_21_soc.clock_limits[i].fabricclk_mhz = dcfclk_mhz[i]; + + /* Fill all states with max values of all these clocks */ + dcn3_21_soc.clock_limits[i].dispclk_mhz = max_dispclk_mhz; + dcn3_21_soc.clock_limits[i].dppclk_mhz = max_dppclk_mhz; + dcn3_21_soc.clock_limits[i].phyclk_mhz = max_phyclk_mhz; + dcn3_21_soc.clock_limits[i].dscclk_mhz = max_dispclk_mhz / 3; + + /* Populate from bw_params for DTBCLK, SOCCLK */ + if (i > 0) { + if (!bw_params->clk_table.entries[i].dtbclk_mhz) { + dcn3_21_soc.clock_limits[i].dtbclk_mhz = dcn3_21_soc.clock_limits[i-1].dtbclk_mhz; + } else { dcn3_21_soc.clock_limits[i].dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz; } + } else if (bw_params->clk_table.entries[i].dtbclk_mhz) { + dcn3_21_soc.clock_limits[i].dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz; + } - if (!bw_params->clk_table.entries[i].socclk_mhz && i > 0) - dcn3_21_soc.clock_limits[i].socclk_mhz = dcn3_21_soc.clock_limits[i-1].socclk_mhz; - else - dcn3_21_soc.clock_limits[i].socclk_mhz = bw_params->clk_table.entries[i].socclk_mhz; + if (!bw_params->clk_table.entries[i].socclk_mhz && i > 0) + dcn3_21_soc.clock_limits[i].socclk_mhz = dcn3_21_soc.clock_limits[i-1].socclk_mhz; + else + dcn3_21_soc.clock_limits[i].socclk_mhz = bw_params->clk_table.entries[i].socclk_mhz; - if (!dram_speed_mts[i] && i > 0) - dcn3_21_soc.clock_limits[i].dram_speed_mts = dcn3_21_soc.clock_limits[i-1].dram_speed_mts; - else - dcn3_21_soc.clock_limits[i].dram_speed_mts = dram_speed_mts[i]; + if (!dram_speed_mts[i] && i > 0) + dcn3_21_soc.clock_limits[i].dram_speed_mts = dcn3_21_soc.clock_limits[i-1].dram_speed_mts; + else + dcn3_21_soc.clock_limits[i].dram_speed_mts = dram_speed_mts[i]; - /* These clocks cannot come from bw_params, always fill from dcn3_21_soc[0] */ - /* PHYCLK_D18, PHYCLK_D32 */ - dcn3_21_soc.clock_limits[i].phyclk_d18_mhz = dcn3_21_soc.clock_limits[0].phyclk_d18_mhz; - dcn3_21_soc.clock_limits[i].phyclk_d32_mhz = dcn3_21_soc.clock_limits[0].phyclk_d32_mhz; - } - } else { - build_synthetic_soc_states(bw_params, dcn3_21_soc.clock_limits, &dcn3_21_soc.num_states); + /* These clocks cannot come from bw_params, always fill from dcn3_21_soc[0] */ + /* PHYCLK_D18, PHYCLK_D32 */ + dcn3_21_soc.clock_limits[i].phyclk_d18_mhz = dcn3_21_soc.clock_limits[0].phyclk_d18_mhz; + dcn3_21_soc.clock_limits[i].phyclk_d32_mhz = dcn3_21_soc.clock_limits[0].phyclk_d32_mhz; } - - /* Re-init DML with updated bb */ - dml_init_instance(&dc->dml, &dcn3_21_soc, &dcn3_21_ip, DML_PROJECT_DCN32); - if (dc->current_state) - dml_init_instance(&dc->current_state->bw_ctx.dml, &dcn3_21_soc, &dcn3_21_ip, DML_PROJECT_DCN32); + } else { + build_synthetic_soc_states(dc->debug.disable_dc_mode_overwrite, bw_params, + dcn3_21_soc.clock_limits, &dcn3_21_soc.num_states); } + + /* Re-init DML with updated bb */ + dml_init_instance(&dc->dml, &dcn3_21_soc, &dcn3_21_ip, DML_PROJECT_DCN32); + if (dc->current_state) + dml_init_instance(&dc->current_state->bw_ctx.dml, &dcn3_21_soc, &dcn3_21_ip, DML_PROJECT_DCN32); } diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h index 0bffae95f3a297e9b8398e46ed960b6ebec25fb8..d5831a34f5a196a14af062826e30530e293f4d6f 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_enums.h @@ -190,6 +190,14 @@ enum dm_validation_status { DML_FAIL_DSC_INPUT_BPC, DML_FAIL_PREFETCH_SUPPORT, DML_FAIL_V_RATIO_PREFETCH, + DML_FAIL_P2I_WITH_420, + DML_FAIL_DSC_ONLY_IF_NECESSARY_WITH_BPP, + DML_FAIL_NOT_DSC422_NATIVE, + DML_FAIL_ODM_COMBINE4TO1, + DML_FAIL_ENOUGH_WRITEBACK_UNITS, + DML_FAIL_VIEWPORT_EXCEEDS_SURFACE, + DML_FAIL_DYNAMIC_METADATA, + DML_FAIL_FMT_BUFFER_EXCEEDED, }; enum writeback_config { diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c index bdf3ac6cadd5561f2dcbde2a199f594ebdf10679..da0cfbb071e6a4bf67327c29eb48e50663a8950e 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.c @@ -113,7 +113,6 @@ void dml_init_instance(struct display_mode_lib *lib, lib->funcs = dml30_funcs; break; case DML_PROJECT_DCN31: - case DML_PROJECT_DCN31_FPGA: case DML_PROJECT_DCN315: lib->funcs = dml31_funcs; break; diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h index a9d49ef58fb59c07daac02ae46ae8274199a8157..5edf69fa40d11daf836ab3c9905f7fb04e980e88 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_lib.h @@ -41,7 +41,6 @@ enum dml_project { DML_PROJECT_DCN30, DML_PROJECT_DCN31, DML_PROJECT_DCN315, - DML_PROJECT_DCN31_FPGA, DML_PROJECT_DCN314, DML_PROJECT_DCN32, }; diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h index 3c077164f36203400f38d9910058e4004cca12bc..ff0246a9458fdc895ef443766cbce0f156237fe7 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h @@ -619,8 +619,7 @@ struct _vcs_dpi_display_dlg_regs_st { unsigned int refcyc_h_blank_end; unsigned int dlg_vblank_end; unsigned int min_dst_y_next_start; - unsigned int optimized_min_dst_y_next_start; - unsigned int optimized_min_dst_y_next_start_us; + unsigned int min_dst_y_next_start_us; unsigned int refcyc_per_htotal; unsigned int refcyc_x_after_scaler; unsigned int dst_y_after_scaler; diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c index f9653f511baa32a1dd7555743258b573b9c6c832..9a3ded31119529d550385c41543b77cc025c6789 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c @@ -571,6 +571,10 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) mode_lib->vba.OutputLinkDPRate[mode_lib->vba.NumberOfActivePlanes] = dout->dp_rate; mode_lib->vba.ODMUse[mode_lib->vba.NumberOfActivePlanes] = dst->odm_combine_policy; mode_lib->vba.DETSizeOverride[mode_lib->vba.NumberOfActivePlanes] = src->det_size_override; + if (src->det_size_override) + mode_lib->vba.DETBufferSizeInKByte[mode_lib->vba.NumberOfActivePlanes] = src->det_size_override; + else + mode_lib->vba.DETBufferSizeInKByte[mode_lib->vba.NumberOfActivePlanes] = ip->det_buffer_size_kbytes; //TODO: Need to assign correct values to dp_multistream vars mode_lib->vba.OutputMultistreamEn[mode_lib->vba.NumberOfActiveSurfaces] = dout->dp_multistream_en; mode_lib->vba.OutputMultistreamId[mode_lib->vba.NumberOfActiveSurfaces] = dout->dp_multistream_id; @@ -785,6 +789,8 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) mode_lib->vba.pipe_plane[k] = mode_lib->vba.NumberOfActivePlanes; mode_lib->vba.DPPPerPlane[mode_lib->vba.NumberOfActivePlanes]++; + if (src_k->det_size_override) + mode_lib->vba.DETBufferSizeInKByte[mode_lib->vba.NumberOfActivePlanes] = src_k->det_size_override; if (mode_lib->vba.SourceScan[mode_lib->vba.NumberOfActivePlanes] == dm_horz) { mode_lib->vba.ViewportWidth[mode_lib->vba.NumberOfActivePlanes] += @@ -927,18 +933,16 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) } /** - * ******************************************************************************************** * cache_debug_params: Cache any params that needed to be maintained from the initial validation * for debug purposes. * * The DML getters can modify some of the VBA params that we are interested in (for example when * calculating with dummy p-state latency), so cache any params here that we want for debugging * - * @param [in] mode_lib: mode_lib input/output of validate call + * @mode_lib: mode_lib input/output of validate call * - * @return: void + * Return: void * - * ******************************************************************************************** */ static void cache_debug_params(struct display_mode_lib *mode_lib) { diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c index 2bdc47615543c57de17602860cadfc7cb0fe0d5f..58dd62cce4bb9b473cb3197b8b9f7302dacde8a0 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c +++ b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c @@ -645,8 +645,6 @@ static int get_available_dsc_slices(union dsc_enc_slice_caps slice_caps, int *av { int idx = 0; - memset(available_slices, -1, MIN_AVAILABLE_SLICES_SIZE); - if (slice_caps.bits.NUM_SLICES_1) available_slices[idx++] = 1; @@ -700,7 +698,7 @@ static int inc_num_slices(union dsc_enc_slice_caps slice_caps, int num_slices) } } - if (new_num_slices == num_slices) // No biger number of slices found + if (new_num_slices == num_slices) // No bigger number of slices found new_num_slices++; return new_num_slices; @@ -952,6 +950,13 @@ static bool setup_dsc_config( else is_dsc_possible = false; } + // When we force 2:1 ODM, we can't have 1 slice to divide amongst 2 separate DSC instances + // need to enforce at minimum 2 horizontal slices + if (options->dsc_force_odm_hslice_override) { + num_slices_h = fit_num_slices_up(dsc_common_caps.slice_caps, 2); + if (num_slices_h == 0) + is_dsc_possible = false; + } if (!is_dsc_possible) goto done; @@ -1163,6 +1168,7 @@ void dc_dsc_policy_set_disable_dsc_stream_overhead(bool disable) void dc_dsc_get_default_config_option(const struct dc *dc, struct dc_dsc_config_options *options) { options->dsc_min_slice_height_override = dc->debug.dsc_min_slice_height_override; + options->dsc_force_odm_hslice_override = dc->debug.force_odm_combine; options->max_target_bpp_limit_override_x16 = 0; options->slice_height_granularity = 1; } diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index 2eb597a24425e9cd666926474501b8071421c125..034610b74a37e6270587f840f64629c98e86eb51 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -37,6 +37,7 @@ #include "dwb.h" #include "mcif_wb.h" #include "panel_cntl.h" +#include "dmub/inc/dmub_cmd.h" #define MAX_CLOCK_SOURCES 7 #define MAX_SVP_PHANTOM_STREAMS 2 @@ -374,6 +375,7 @@ union pipe_update_flags { uint32_t viewport : 1; uint32_t plane_changed : 1; uint32_t det_size : 1; + uint32_t unbounded_req : 1; } bits; uint32_t raw; }; @@ -426,6 +428,8 @@ struct pipe_ctx { struct dwbc *dwbc; struct mcif_wb *mcif_wb; union pipe_update_flags update_flags; + struct tg_color visual_confirm_color; + bool has_vactive_margin; }; /* Data used for dynamic link encoder assignment. @@ -496,6 +500,11 @@ struct bw_context { struct display_mode_lib dml; }; +struct dc_dmub_cmd { + union dmub_rb_cmd dmub_cmd; + enum dm_dmub_wait_type wait_type; +}; + /** * struct dc_state - The full description of a state requested by users */ @@ -544,6 +553,11 @@ struct dc_state { */ struct bw_context bw_ctx; + struct block_sequence block_sequence[50]; + unsigned int block_sequence_steps; + struct dc_dmub_cmd dc_dmub_cmd[10]; + unsigned int dmub_cmd_count; + /** * @refcount: refcount reference * diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h b/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h index ecb4191b6e64f45cceb2af602c59a5d9f2ad85b9..d2190a3320f64cd8e9336f7cf4ba53384e1934ac 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h @@ -55,6 +55,10 @@ struct abm_funcs { unsigned int bytes, unsigned int inst); bool (*set_abm_pause)(struct abm *abm, bool pause, unsigned int panel_inst, unsigned int otg_inst); + bool (*set_pipe_ex)(struct abm *abm, + unsigned int otg_inst, + unsigned int option, + unsigned int panel_inst); }; #endif diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h index bef843cc32a1aeb3c9b7bc2cf83e76d492fefed3..6faf40fa5c695ecef2dc295b5cc04d684be7febf 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h @@ -233,6 +233,7 @@ struct clk_bw_params { struct clk_limit_table clk_table; struct wm_table wm_table; struct dummy_pstate_entry dummy_pstate_table[4]; + struct clk_limit_table_entry dc_mode_limit; }; /* Public interfaces */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h index ad6acd1b34e1dd33eedc072b7db807ff3a40ffd3..8dc804bbe98bd11e89b1cb9578be40935399b317 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h @@ -68,6 +68,7 @@ struct dccg { const struct dccg_funcs *funcs; int pipe_dppclk_khz[MAX_PIPES]; int ref_dppclk; + bool dpp_clock_gated[MAX_PIPES]; //int dtbclk_khz[MAX_PIPES];/* TODO needs to be removed */ //int audio_dtbclk_khz;/* TODO needs to be removed */ //int ref_dtbclk_khz;/* TODO needs to be removed */ @@ -159,6 +160,9 @@ struct dccg_funcs { int otg_inst, int pixclk_khz); + void (*trigger_dio_fifo_resync)( + struct dccg *dccg); + void (*dpp_root_clock_control)( struct dccg *dccg, unsigned int dpp_inst, diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h index beb26dc8a07fadf0f66901a7d051414a887a20e3..aaa29361384672d3886005c67f96c7aba5aa39f7 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h @@ -111,6 +111,9 @@ struct dcn_hubbub_state { uint32_t vm_error_vmid; uint32_t vm_error_pipe; uint32_t vm_error_mode; + uint32_t test_debug_data; + uint32_t watermark_change_cntl; + uint32_t dram_state_cntl; }; struct hubbub_funcs { diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h index 88ac723d10aa72340fbff0c0903cf28fa1de713e..02ff99f7bec2baec1d0532e6fd1220bc7783830c 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h @@ -44,6 +44,138 @@ struct dc_virtual_addr_space_config; struct dpp; struct dce_hwseq; struct link_resource; +struct dc_dmub_cmd; + +struct subvp_pipe_control_lock_fast_params { + struct dc *dc; + bool lock; + struct pipe_ctx *pipe_ctx; +}; + +struct pipe_control_lock_params { + struct dc *dc; + struct pipe_ctx *pipe_ctx; + bool lock; +}; + +struct set_flip_control_gsl_params { + struct pipe_ctx *pipe_ctx; + bool flip_immediate; +}; + +struct program_triplebuffer_params { + const struct dc *dc; + struct pipe_ctx *pipe_ctx; + bool enableTripleBuffer; +}; + +struct update_plane_addr_params { + struct dc *dc; + struct pipe_ctx *pipe_ctx; +}; + +struct set_input_transfer_func_params { + struct dc *dc; + struct pipe_ctx *pipe_ctx; + struct dc_plane_state *plane_state; +}; + +struct program_gamut_remap_params { + struct pipe_ctx *pipe_ctx; +}; + +struct program_manual_trigger_params { + struct pipe_ctx *pipe_ctx; +}; + +struct send_dmcub_cmd_params { + struct dc_context *ctx; + union dmub_rb_cmd *cmd; + enum dm_dmub_wait_type wait_type; +}; + +struct setup_dpp_params { + struct pipe_ctx *pipe_ctx; +}; + +struct program_bias_and_scale_params { + struct pipe_ctx *pipe_ctx; +}; + +struct set_output_transfer_func_params { + struct dc *dc; + struct pipe_ctx *pipe_ctx; + const struct dc_stream_state *stream; +}; + +struct update_visual_confirm_params { + struct dc *dc; + struct pipe_ctx *pipe_ctx; + int mpcc_id; +}; + +struct power_on_mpc_mem_pwr_params { + struct mpc *mpc; + int mpcc_id; + bool power_on; +}; + +struct set_output_csc_params { + struct mpc *mpc; + int opp_id; + const uint16_t *regval; + enum mpc_output_csc_mode ocsc_mode; +}; + +struct set_ocsc_default_params { + struct mpc *mpc; + int opp_id; + enum dc_color_space color_space; + enum mpc_output_csc_mode ocsc_mode; +}; + +union block_sequence_params { + struct update_plane_addr_params update_plane_addr_params; + struct subvp_pipe_control_lock_fast_params subvp_pipe_control_lock_fast_params; + struct pipe_control_lock_params pipe_control_lock_params; + struct set_flip_control_gsl_params set_flip_control_gsl_params; + struct program_triplebuffer_params program_triplebuffer_params; + struct set_input_transfer_func_params set_input_transfer_func_params; + struct program_gamut_remap_params program_gamut_remap_params; + struct program_manual_trigger_params program_manual_trigger_params; + struct send_dmcub_cmd_params send_dmcub_cmd_params; + struct setup_dpp_params setup_dpp_params; + struct program_bias_and_scale_params program_bias_and_scale_params; + struct set_output_transfer_func_params set_output_transfer_func_params; + struct update_visual_confirm_params update_visual_confirm_params; + struct power_on_mpc_mem_pwr_params power_on_mpc_mem_pwr_params; + struct set_output_csc_params set_output_csc_params; + struct set_ocsc_default_params set_ocsc_default_params; +}; + +enum block_sequence_func { + DMUB_SUBVP_PIPE_CONTROL_LOCK_FAST = 0, + OPTC_PIPE_CONTROL_LOCK, + HUBP_SET_FLIP_CONTROL_GSL, + HUBP_PROGRAM_TRIPLEBUFFER, + HUBP_UPDATE_PLANE_ADDR, + DPP_SET_INPUT_TRANSFER_FUNC, + DPP_PROGRAM_GAMUT_REMAP, + OPTC_PROGRAM_MANUAL_TRIGGER, + DMUB_SEND_DMCUB_CMD, + DPP_SETUP_DPP, + DPP_PROGRAM_BIAS_AND_SCALE, + DPP_SET_OUTPUT_TRANSFER_FUNC, + MPC_UPDATE_VISUAL_CONFIRM, + MPC_POWER_ON_MPC_MEM_PWR, + MPC_SET_OUTPUT_CSC, + MPC_SET_OCSC_DEFAULT, +}; + +struct block_sequence { + union block_sequence_params params; + enum block_sequence_func func; +}; struct hw_sequencer_funcs { void (*hardware_release)(struct dc *dc); @@ -252,12 +384,12 @@ struct hw_sequencer_funcs { const struct tg_color *solid_color, int width, int height, int offset); + void (*subvp_pipe_control_lock_fast)(union block_sequence_params *params); void (*z10_restore)(const struct dc *dc); void (*z10_save_init)(struct dc *dc); void (*update_visual_confirm_color)(struct dc *dc, struct pipe_ctx *pipe_ctx, - struct tg_color *color, int mpcc_id); void (*update_phantom_vp_position)(struct dc *dc, @@ -294,6 +426,7 @@ void get_surface_visual_confirm_color( void get_subvp_visual_confirm_color( struct dc *dc, + struct dc_state *context, struct pipe_ctx *pipe_ctx, struct tg_color *color); @@ -306,4 +439,36 @@ void get_mpctree_visual_confirm_color( void get_surface_tile_visual_confirm_color( struct pipe_ctx *pipe_ctx, struct tg_color *color); + +void get_mclk_switch_visual_confirm_color( + struct dc *dc, + struct dc_state *context, + struct pipe_ctx *pipe_ctx, + struct tg_color *color); + +void hwss_execute_sequence(struct dc *dc, + struct block_sequence block_sequence[], + int num_steps); + +void hwss_build_fast_sequence(struct dc *dc, + struct dc_dmub_cmd *dc_dmub_cmd, + unsigned int dmub_cmd_count, + struct block_sequence block_sequence[], + int *num_steps, + struct pipe_ctx *pipe_ctx); + +void hwss_send_dmcub_cmd(union block_sequence_params *params); + +void hwss_program_manual_trigger(union block_sequence_params *params); + +void hwss_setup_dpp(union block_sequence_params *params); + +void hwss_program_bias_and_scale(union block_sequence_params *params); + +void hwss_power_on_mpc_mem_pwr(union block_sequence_params *params); + +void hwss_set_output_csc(union block_sequence_params *params); + +void hwss_set_ocsc_default(union block_sequence_params *params); + #endif /* __DC_HW_SEQUENCER_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h index 4513544559be2019a2c327b8011cca3d6afb8dfe..a151865a3a20d8509c63994ca03e3ec2a0a73260 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h @@ -156,10 +156,12 @@ struct hwseq_private_funcs { void (*program_mall_pipe_config)(struct dc *dc, struct dc_state *context); void (*update_force_pstate)(struct dc *dc, struct dc_state *context); void (*update_mall_sel)(struct dc *dc, struct dc_state *context); - unsigned int (*calculate_dccg_k1_k2_values)(struct pipe_ctx *pipe_ctx, + void (*calculate_dccg_k1_k2_values)(struct pipe_ctx *pipe_ctx, unsigned int *k1_div, unsigned int *k2_div); void (*set_pixels_per_cycle)(struct pipe_ctx *pipe_ctx); + void (*resync_fifo_dccg_dio)(struct dce_hwseq *hws, struct dc *dc, + struct dc_state *context); bool (*is_dp_dig_pixel_rate_div_policy)(struct pipe_ctx *pipe_ctx); #endif }; diff --git a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.c b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.c index edd7d026a762a2273ebb5ec4461d83806dcc33de..586fe25c170295af0aa43742d32823088c9eaa95 100644 --- a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.c +++ b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.c @@ -28,24 +28,6 @@ #include "dccg.h" #include "clk_mgr.h" -static enum phyd32clk_clock_source get_phyd32clk_src(struct dc_link *link) -{ - switch (link->link_enc->transmitter) { - case TRANSMITTER_UNIPHY_A: - return PHYD32CLKA; - case TRANSMITTER_UNIPHY_B: - return PHYD32CLKB; - case TRANSMITTER_UNIPHY_C: - return PHYD32CLKC; - case TRANSMITTER_UNIPHY_D: - return PHYD32CLKD; - case TRANSMITTER_UNIPHY_E: - return PHYD32CLKE; - default: - return PHYD32CLKA; - } -} - static void set_hpo_dp_throttled_vcp_size(struct pipe_ctx *pipe_ctx, struct fixed31_32 throttled_vcp_size) { @@ -120,81 +102,26 @@ static void setup_hpo_dp_stream_attribute(struct pipe_ctx *pipe_ctx) DPCD_SOURCE_SEQ_AFTER_DP_STREAM_ATTR); } -static void enable_hpo_dp_fpga_link_output(struct dc_link *link, - const struct link_resource *link_res, - enum signal_type signal, - enum clock_source_id clock_source, - const struct dc_link_settings *link_settings) -{ - const struct dc *dc = link->dc; - enum phyd32clk_clock_source phyd32clk = get_phyd32clk_src(link); - int phyd32clk_freq_khz = link_settings->link_rate == LINK_RATE_UHBR10 ? 312500 : - link_settings->link_rate == LINK_RATE_UHBR13_5 ? 412875 : - link_settings->link_rate == LINK_RATE_UHBR20 ? 625000 : 0; - - dm_set_phyd32clk(dc->ctx, phyd32clk_freq_khz); - dc->res_pool->dccg->funcs->set_physymclk( - dc->res_pool->dccg, - link->link_enc_hw_inst, - PHYSYMCLK_FORCE_SRC_PHYD32CLK, - true); - dc->res_pool->dccg->funcs->enable_symclk32_le( - dc->res_pool->dccg, - link_res->hpo_dp_link_enc->inst, - phyd32clk); - link_res->hpo_dp_link_enc->funcs->link_enable( - link_res->hpo_dp_link_enc, - link_settings->lane_count); - -} - static void enable_hpo_dp_link_output(struct dc_link *link, const struct link_resource *link_res, enum signal_type signal, enum clock_source_id clock_source, const struct dc_link_settings *link_settings) { - if (IS_FPGA_MAXIMUS_DC(link->dc->ctx->dce_environment)) - enable_hpo_dp_fpga_link_output(link, link_res, signal, - clock_source, link_settings); - else - link_res->hpo_dp_link_enc->funcs->enable_link_phy( - link_res->hpo_dp_link_enc, - link_settings, - link->link_enc->transmitter, - link->link_enc->hpd_source); -} - - -static void disable_hpo_dp_fpga_link_output(struct dc_link *link, - const struct link_resource *link_res, - enum signal_type signal) -{ - const struct dc *dc = link->dc; - - link_res->hpo_dp_link_enc->funcs->link_disable(link_res->hpo_dp_link_enc); - dc->res_pool->dccg->funcs->disable_symclk32_le( - dc->res_pool->dccg, - link_res->hpo_dp_link_enc->inst); - dc->res_pool->dccg->funcs->set_physymclk( - dc->res_pool->dccg, - link->link_enc_hw_inst, - PHYSYMCLK_FORCE_SRC_SYMCLK, - false); - dm_set_phyd32clk(dc->ctx, 0); + link_res->hpo_dp_link_enc->funcs->enable_link_phy( + link_res->hpo_dp_link_enc, + link_settings, + link->link_enc->transmitter, + link->link_enc->hpd_source); } static void disable_hpo_dp_link_output(struct dc_link *link, const struct link_resource *link_res, enum signal_type signal) { - if (IS_FPGA_MAXIMUS_DC(link->dc->ctx->dce_environment)) { - disable_hpo_dp_fpga_link_output(link, link_res, signal); - } else { link_res->hpo_dp_link_enc->funcs->link_disable(link_res->hpo_dp_link_enc); link_res->hpo_dp_link_enc->funcs->disable_link_phy( link_res->hpo_dp_link_enc, signal); - } } static void set_hpo_dp_link_test_pattern(struct dc_link *link, diff --git a/drivers/gpu/drm/amd/display/dc/link/link_detection.c b/drivers/gpu/drm/amd/display/dc/link/link_detection.c index a131e30fd7d6a01036b55598662b50345dbc34e1..8041b8369e45f1d11d7b9929a71afcb066fc8f69 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_detection.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_detection.c @@ -593,6 +593,10 @@ static bool detect_dp(struct dc_link *link, /* DP SST branch */ link->type = dc_connection_sst_branch; } else { + if (link->dc->debug.disable_dp_plus_plus_wa && + link->link_enc->features.flags.bits.IS_UHBR20_CAPABLE) + return false; + /* DP passive dongles */ sink_caps->signal = dp_passive_dongle_detection(link->ddc, sink_caps, @@ -980,6 +984,11 @@ static bool detect_link_and_local_sink(struct dc_link *link, (link->dpcd_caps.dongle_type != DISPLAY_DONGLE_DP_HDMI_CONVERTER)) converter_disable_audio = true; + + /* limited link rate to HBR3 for DPIA until we implement USB4 V2 */ + if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA && + link->reported_link_cap.link_rate > LINK_RATE_HIGH3) + link->reported_link_cap.link_rate = LINK_RATE_HIGH3; break; } diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c index 2267fb09783057d996e53d5837026192a4b0f595..1a7b93e41e352adb0e67a11d01a36e7d76eaffbc 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c @@ -765,7 +765,7 @@ static bool dp_set_dsc_on_rx(struct pipe_ctx *pipe_ctx, bool enable) struct dc_stream_state *stream = pipe_ctx->stream; bool result = false; - if (dc_is_virtual_signal(stream->signal) || IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) + if (dc_is_virtual_signal(stream->signal)) result = true; else result = dm_helpers_dp_write_dsc_enable(dc->ctx, stream, enable); @@ -778,7 +778,6 @@ static bool dp_set_dsc_on_rx(struct pipe_ctx *pipe_ctx, bool enable) void link_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) { struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc; - struct dc *dc = pipe_ctx->stream->ctx->dc; struct dc_stream_state *stream = pipe_ctx->stream; struct pipe_ctx *odm_pipe; int opp_cnt = 1; @@ -816,8 +815,7 @@ void link_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) optc_dsc_mode = dsc_optc_cfg.is_pixel_format_444 ? OPTC_DSC_ENABLED_444 : OPTC_DSC_ENABLED_NATIVE_SUBSAMPLED; /* Enable DSC in encoder */ - if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) - && !dp_is_128b_132b_signal(pipe_ctx)) { + if (dc_is_dp_signal(stream->signal) && !dp_is_128b_132b_signal(pipe_ctx)) { DC_LOG_DSC("Setting stream encoder DSC config for engine %d:", (int)pipe_ctx->stream_res.stream_enc->id); dsc_optc_config_log(dsc, &dsc_optc_cfg); pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config(pipe_ctx->stream_res.stream_enc, @@ -849,7 +847,7 @@ void link_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) false, NULL, true); - else if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { + else { pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config( pipe_ctx->stream_res.stream_enc, OPTC_DSC_DISABLED, 0, 0); @@ -2209,9 +2207,8 @@ static enum dc_status enable_link( * link settings. Need to call disable first before enabling at * new link settings. */ - if (link->link_status.link_active) { + if (link->link_status.link_active && !stream->skip_edp_power_down) disable_link(link, &pipe_ctx->link_res, pipe_ctx->stream->signal); - } switch (pipe_ctx->stream->signal) { case SIGNAL_TYPE_DISPLAY_PORT: @@ -2271,8 +2268,7 @@ void link_set_dpms_off(struct pipe_ctx *pipe_ctx) } } - if (!IS_DIAG_DC(dc->ctx->dce_environment) && - dc_is_virtual_signal(pipe_ctx->stream->signal)) + if (dc_is_virtual_signal(pipe_ctx->stream->signal)) return; if (!pipe_ctx->stream->sink->edid_caps.panel_patch.skip_avmute) { @@ -2330,7 +2326,9 @@ void link_set_dpms_off(struct pipe_ctx *pipe_ctx) dc->hwss.disable_stream(pipe_ctx); } else { dc->hwss.disable_stream(pipe_ctx); - disable_link(pipe_ctx->stream->link, &pipe_ctx->link_res, pipe_ctx->stream->signal); + if (!pipe_ctx->stream->skip_edp_power_down) { + disable_link(pipe_ctx->stream->link, &pipe_ctx->link_res, pipe_ctx->stream->signal); + } } if (pipe_ctx->stream->timing.flags.DSC) { @@ -2358,6 +2356,8 @@ void link_set_dpms_on( enum otg_out_mux_dest otg_out_dest = OUT_MUX_DIO; struct vpg *vpg = pipe_ctx->stream_res.stream_enc->vpg; const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); + bool apply_edp_fast_boot_optimization = + pipe_ctx->stream->apply_edp_fast_boot_optimization; ASSERT(is_master_pipe_for_link(link, pipe_ctx)); @@ -2375,8 +2375,7 @@ void link_set_dpms_on( } } - if (!IS_DIAG_DC(dc->ctx->dce_environment) && - dc_is_virtual_signal(pipe_ctx->stream->signal)) + if (dc_is_virtual_signal(pipe_ctx->stream->signal)) return; link_enc = link_enc_cfg_get_link_enc(link); @@ -2402,138 +2401,126 @@ void link_set_dpms_on( link_hwss->setup_stream_attribute(pipe_ctx); - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - bool apply_edp_fast_boot_optimization = - pipe_ctx->stream->apply_edp_fast_boot_optimization; - - pipe_ctx->stream->apply_edp_fast_boot_optimization = false; + pipe_ctx->stream->apply_edp_fast_boot_optimization = false; - // Enable VPG before building infoframe - if (vpg && vpg->funcs->vpg_poweron) - vpg->funcs->vpg_poweron(vpg); + // Enable VPG before building infoframe + if (vpg && vpg->funcs->vpg_poweron) + vpg->funcs->vpg_poweron(vpg); - resource_build_info_frame(pipe_ctx); - dc->hwss.update_info_frame(pipe_ctx); + resource_build_info_frame(pipe_ctx); + dc->hwss.update_info_frame(pipe_ctx); - if (dc_is_dp_signal(pipe_ctx->stream->signal)) - dp_trace_source_sequence(link, DPCD_SOURCE_SEQ_AFTER_UPDATE_INFO_FRAME); - - /* Do not touch link on seamless boot optimization. */ - if (pipe_ctx->stream->apply_seamless_boot_optimization) { - pipe_ctx->stream->dpms_off = false; + if (dc_is_dp_signal(pipe_ctx->stream->signal)) + dp_trace_source_sequence(link, DPCD_SOURCE_SEQ_AFTER_UPDATE_INFO_FRAME); - /* Still enable stream features & audio on seamless boot for DP external displays */ - if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT) { - enable_stream_features(pipe_ctx); - dc->hwss.enable_audio_stream(pipe_ctx); - } + /* Do not touch link on seamless boot optimization. */ + if (pipe_ctx->stream->apply_seamless_boot_optimization) { + pipe_ctx->stream->dpms_off = false; - update_psp_stream_config(pipe_ctx, false); - return; - } - - /* eDP lit up by bios already, no need to enable again. */ - if (pipe_ctx->stream->signal == SIGNAL_TYPE_EDP && - apply_edp_fast_boot_optimization && - !pipe_ctx->stream->timing.flags.DSC && - !pipe_ctx->next_odm_pipe) { - pipe_ctx->stream->dpms_off = false; - update_psp_stream_config(pipe_ctx, false); - return; + /* Still enable stream features & audio on seamless boot for DP external displays */ + if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT) { + enable_stream_features(pipe_ctx); + dc->hwss.enable_audio_stream(pipe_ctx); } - if (pipe_ctx->stream->dpms_off) - return; + update_psp_stream_config(pipe_ctx, false); + return; + } - /* Have to setup DSC before DIG FE and BE are connected (which happens before the - * link training). This is to make sure the bandwidth sent to DIG BE won't be - * bigger than what the link and/or DIG BE can handle. VBID[6]/CompressedStream_flag - * will be automatically set at a later time when the video is enabled - * (DP_VID_STREAM_EN = 1). - */ - if (pipe_ctx->stream->timing.flags.DSC) { - if (dc_is_dp_signal(pipe_ctx->stream->signal) || - dc_is_virtual_signal(pipe_ctx->stream->signal)) - link_set_dsc_enable(pipe_ctx, true); + /* eDP lit up by bios already, no need to enable again. */ + if (pipe_ctx->stream->signal == SIGNAL_TYPE_EDP && + apply_edp_fast_boot_optimization && + !pipe_ctx->stream->timing.flags.DSC && + !pipe_ctx->next_odm_pipe) { + pipe_ctx->stream->dpms_off = false; + update_psp_stream_config(pipe_ctx, false); + return; + } - } + if (pipe_ctx->stream->dpms_off) + return; - status = enable_link(state, pipe_ctx); + /* Have to setup DSC before DIG FE and BE are connected (which happens before the + * link training). This is to make sure the bandwidth sent to DIG BE won't be + * bigger than what the link and/or DIG BE can handle. VBID[6]/CompressedStream_flag + * will be automatically set at a later time when the video is enabled + * (DP_VID_STREAM_EN = 1). + */ + if (pipe_ctx->stream->timing.flags.DSC) { + if (dc_is_dp_signal(pipe_ctx->stream->signal) || + dc_is_virtual_signal(pipe_ctx->stream->signal)) + link_set_dsc_enable(pipe_ctx, true); - if (status != DC_OK) { - DC_LOG_WARNING("enabling link %u failed: %d\n", - pipe_ctx->stream->link->link_index, - status); + } - /* Abort stream enable *unless* the failure was due to - * DP link training - some DP monitors will recover and - * show the stream anyway. But MST displays can't proceed - * without link training. - */ - if (status != DC_FAIL_DP_LINK_TRAINING || - pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) { - if (false == stream->link->link_status.link_active) - disable_link(stream->link, &pipe_ctx->link_res, - pipe_ctx->stream->signal); - BREAK_TO_DEBUGGER(); - return; - } - } + status = enable_link(state, pipe_ctx); - /* turn off otg test pattern if enable */ - if (pipe_ctx->stream_res.tg->funcs->set_test_pattern) - pipe_ctx->stream_res.tg->funcs->set_test_pattern(pipe_ctx->stream_res.tg, - CONTROLLER_DP_TEST_PATTERN_VIDEOMODE, - COLOR_DEPTH_UNDEFINED); + if (status != DC_OK) { + DC_LOG_WARNING("enabling link %u failed: %d\n", + pipe_ctx->stream->link->link_index, + status); - /* This second call is needed to reconfigure the DIG - * as a workaround for the incorrect value being applied - * from transmitter control. + /* Abort stream enable *unless* the failure was due to + * DP link training - some DP monitors will recover and + * show the stream anyway. But MST displays can't proceed + * without link training. */ - if (!(dc_is_virtual_signal(pipe_ctx->stream->signal) || - dp_is_128b_132b_signal(pipe_ctx))) { - if (link_enc) - link_enc->funcs->setup( - link_enc, + if (status != DC_FAIL_DP_LINK_TRAINING || + pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) { + if (false == stream->link->link_status.link_active) + disable_link(stream->link, &pipe_ctx->link_res, pipe_ctx->stream->signal); - } + BREAK_TO_DEBUGGER(); + return; + } + } - dc->hwss.enable_stream(pipe_ctx); + /* turn off otg test pattern if enable */ + if (pipe_ctx->stream_res.tg->funcs->set_test_pattern) + pipe_ctx->stream_res.tg->funcs->set_test_pattern(pipe_ctx->stream_res.tg, + CONTROLLER_DP_TEST_PATTERN_VIDEOMODE, + COLOR_DEPTH_UNDEFINED); - /* Set DPS PPS SDP (AKA "info frames") */ - if (pipe_ctx->stream->timing.flags.DSC) { - if (dc_is_dp_signal(pipe_ctx->stream->signal) || - dc_is_virtual_signal(pipe_ctx->stream->signal)) { - dp_set_dsc_on_rx(pipe_ctx, true); - link_set_dsc_pps_packet(pipe_ctx, true, true); - } + /* This second call is needed to reconfigure the DIG + * as a workaround for the incorrect value being applied + * from transmitter control. + */ + if (!(dc_is_virtual_signal(pipe_ctx->stream->signal) || + dp_is_128b_132b_signal(pipe_ctx))) { + if (link_enc) + link_enc->funcs->setup( + link_enc, + pipe_ctx->stream->signal); } - if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) - allocate_mst_payload(pipe_ctx); - else if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT && - dp_is_128b_132b_signal(pipe_ctx)) - update_sst_payload(pipe_ctx, true); + dc->hwss.enable_stream(pipe_ctx); + + /* Set DPS PPS SDP (AKA "info frames") */ + if (pipe_ctx->stream->timing.flags.DSC) { + if (dc_is_dp_signal(pipe_ctx->stream->signal) || + dc_is_virtual_signal(pipe_ctx->stream->signal)) { + dp_set_dsc_on_rx(pipe_ctx, true); + link_set_dsc_pps_packet(pipe_ctx, true, true); + } + } - dc->hwss.unblank_stream(pipe_ctx, - &pipe_ctx->stream->link->cur_link_settings); + if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) + allocate_mst_payload(pipe_ctx); + else if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT && + dp_is_128b_132b_signal(pipe_ctx)) + update_sst_payload(pipe_ctx, true); - if (stream->sink_patches.delay_ignore_msa > 0) - msleep(stream->sink_patches.delay_ignore_msa); + dc->hwss.unblank_stream(pipe_ctx, + &pipe_ctx->stream->link->cur_link_settings); - if (dc_is_dp_signal(pipe_ctx->stream->signal)) - enable_stream_features(pipe_ctx); - update_psp_stream_config(pipe_ctx, false); + if (stream->sink_patches.delay_ignore_msa > 0) + msleep(stream->sink_patches.delay_ignore_msa); - dc->hwss.enable_audio_stream(pipe_ctx); + if (dc_is_dp_signal(pipe_ctx->stream->signal)) + enable_stream_features(pipe_ctx); + update_psp_stream_config(pipe_ctx, false); - } else { // if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) - if (dp_is_128b_132b_signal(pipe_ctx)) - dp_fpga_hpo_enable_link_and_stream(state, pipe_ctx); - if (dc_is_dp_signal(pipe_ctx->stream->signal) || - dc_is_virtual_signal(pipe_ctx->stream->signal)) - link_set_dsc_enable(pipe_ctx, true); - } + dc->hwss.enable_audio_stream(pipe_ctx); if (dc_is_hdmi_signal(pipe_ctx->stream->signal)) { set_avmute(pipe_ctx, false); diff --git a/drivers/gpu/drm/amd/display/dc/link/link_factory.c b/drivers/gpu/drm/amd/display/dc/link/link_factory.c index 1515c817f03be244f7107766f80210a2d1a7d8b8..ac1c3e2e7c1d667747a0366dab47c7b0c3eb2cab 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_factory.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_factory.c @@ -563,11 +563,9 @@ static bool construct_phy(struct dc_link *link, goto create_fail; } - /* TODO: #DAL3 Implement id to str function.*/ - LINK_INFO("Connector[%d] description:" - "signal %d\n", + LINK_INFO("Connector[%d] description: signal: %s\n", init_params->connector_index, - link->connector_signal); + signal_type_to_string(link->connector_signal)); ddc_service_init_data.ctx = link->ctx; ddc_service_init_data.id = link->link_id; diff --git a/drivers/gpu/drm/amd/display/dc/link/link_validation.c b/drivers/gpu/drm/amd/display/dc/link/link_validation.c index d4b7da526f0a55eeb7736edb5f99a84826110126..e8b2fc4002a52d07dc91e7cee0c3250acd23dc51 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_validation.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_validation.c @@ -359,5 +359,8 @@ bool link_validate_dpia_bandwidth(const struct dc_stream_state *stream, const un link[i] = stream[i].link; bw_needed[i] = dc_bandwidth_in_kbps_from_timing(&stream[i].timing); } + + ret = dpia_validate_usb4_bw(link, bw_needed, num_streams); + return ret; } diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c index ba98013fecd00fc74f452d9a1bfc1ab04c3bbce5..3a5e80b5771169a79c94c076f1a945634e156b57 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c @@ -326,8 +326,7 @@ bool dp_is_fec_supported(const struct dc_link *link) return (dc_is_dp_signal(link->connector_signal) && link_enc && link_enc->features.fec_supported && - link->dpcd_caps.fec_cap.bits.FEC_CAPABLE && - !IS_FPGA_MAXIMUS_DC(link->ctx->dce_environment)); + link->dpcd_caps.fec_cap.bits.FEC_CAPABLE); } bool dp_should_enable_fec(const struct dc_link *link) @@ -1043,9 +1042,7 @@ static enum dc_status wake_up_aux_channel(struct dc_link *link) DP_SET_POWER, &dpcd_power_state, sizeof(dpcd_power_state)); - if (status < 0) - DC_LOG_DC("%s: Failed to power up sink: %s\n", __func__, - dpcd_power_state == DP_SET_POWER_D0 ? "D0" : "D3"); + DC_LOG_DC("%s: Failed to power up sink\n", __func__); return DC_ERROR_UNEXPECTED; } @@ -1396,7 +1393,7 @@ static bool get_usbc_cable_id(struct dc_link *link, union dp_cable_id *cable_id) cmd.cable_id.header.payload_bytes = sizeof(cmd.cable_id.data); cmd.cable_id.data.input.phy_inst = resource_transmitter_to_phy_idx( link->dc, link->link_enc->transmitter); - if (dc_dmub_srv_cmd_with_reply_data(link->ctx->dmub_srv, &cmd) && + if (dm_execute_dmub_cmd(link->dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) && cmd.cable_id.header.ret_status == 1) { cable_id->raw = cmd.cable_id.data.output_raw; DC_LOG_DC("usbc_cable_id = %d.\n", cable_id->raw); @@ -1452,7 +1449,8 @@ bool read_is_mst_supported(struct dc_link *link) */ static bool dpcd_read_sink_ext_caps(struct dc_link *link) { - uint8_t dpcd_data; + uint8_t dpcd_data = 0; + uint8_t edp_general_cap2 = 0; if (!link) return false; @@ -1461,6 +1459,12 @@ static bool dpcd_read_sink_ext_caps(struct dc_link *link) return false; link->dpcd_sink_ext_caps.raw = dpcd_data; + + if (core_link_read_dpcd(link, DP_EDP_GENERAL_CAP_2, &edp_general_cap2, 1) != DC_OK) + return false; + + link->dpcd_caps.panel_luminance_control = (edp_general_cap2 & DP_EDP_PANEL_LUMINANCE_CONTROL_CAPABLE) != 0; + return true; } @@ -1554,6 +1558,9 @@ static bool retrieve_link_cap(struct dc_link *link) int i; struct dp_sink_hw_fw_revision dp_hw_fw_revision; const uint32_t post_oui_delay = 30; // 30ms + bool is_fec_supported = false; + bool is_dsc_basic_supported = false; + bool is_dsc_passthrough_supported = false; memset(dpcd_data, '\0', sizeof(dpcd_data)); memset(&down_strm_port_count, @@ -1696,6 +1703,7 @@ static bool retrieve_link_cap(struct dc_link *link) /* TODO - decouple raw mst capability from policy decision */ link->dpcd_caps.is_mst_capable = read_is_mst_supported(link); + DC_LOG_DC("%s: MST_Support: %s\n", __func__, str_yes_no(link->dpcd_caps.is_mst_capable)); get_active_converter_info(ds_port.byte, link); @@ -1803,6 +1811,17 @@ static bool retrieve_link_cap(struct dc_link *link) DP_DSC_SUPPORT, link->dpcd_caps.dsc_caps.dsc_basic_caps.raw, sizeof(link->dpcd_caps.dsc_caps.dsc_basic_caps.raw)); + if (status == DC_OK) { + is_fec_supported = link->dpcd_caps.fec_cap.bits.FEC_CAPABLE; + is_dsc_basic_supported = link->dpcd_caps.dsc_caps.dsc_basic_caps.fields.dsc_support.DSC_SUPPORT; + is_dsc_passthrough_supported = link->dpcd_caps.dsc_caps.dsc_basic_caps.fields.dsc_support.DSC_PASSTHROUGH_SUPPORT; + DC_LOG_DC("%s: FEC_Sink_Support: %s\n", __func__, + str_yes_no(is_fec_supported)); + DC_LOG_DC("%s: DSC_Basic_Sink_Support: %s\n", __func__, + str_yes_no(is_dsc_basic_supported)); + DC_LOG_DC("%s: DSC_Passthrough_Sink_Support: %s\n", __func__, + str_yes_no(is_dsc_passthrough_supported)); + } if (link->dpcd_caps.dongle_type != DISPLAY_DONGLE_NONE) { status = core_link_read_dpcd( link, @@ -1931,6 +1950,9 @@ void detect_edp_sink_caps(struct dc_link *link) link_rate_in_khz = (supported_link_rates[entry+1] * 0x100 + supported_link_rates[entry]) * 200; + DC_LOG_DC("%s: eDP v1.4 supported sink rates: [%d] %d kHz\n", __func__, + entry / 2, link_rate_in_khz); + if (link_rate_in_khz != 0) { link_rate = linkRateInKHzToLinkRateMultiplier(link_rate_in_khz); link->dpcd_caps.edp_supported_link_rates[link->dpcd_caps.edp_supported_link_rates_count] = link_rate; diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c index 4626fabc0a96bdf5014c91f92bb50daaf6c6f2ae..0bb7491339098a5ee99ae8e210dc04f6fb4c61c2 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c @@ -90,7 +90,7 @@ bool dpia_query_hpd_status(struct dc_link *link) cmd.query_hpd.data.ch_type = AUX_CHANNEL_DPIA; /* Return HPD status reported by DMUB if query successfully executed. */ - if (dc_dmub_srv_cmd_with_reply_data(dmub_srv, &cmd) && cmd.query_hpd.data.status == AUX_RET_SUCCESS) + if (dm_execute_dmub_cmd(dmub_srv->ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) && cmd.query_hpd.data.status == AUX_RET_SUCCESS) is_hpd_high = cmd.query_hpd.data.result; DC_LOG_DEBUG("%s: link(%d) dpia(%d) cmd_status(%d) result(%d)\n", diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c index 579fa222810d4ad0a7508d35275649ad4c2e1b4d..e011df4bdaf216a872a6c6fee711faaed645e523 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c @@ -1653,10 +1653,19 @@ bool perform_link_training_with_retries( break; } - DC_LOG_WARNING("%s: Link(%d) training attempt %u of %d failed @ rate(%d) x lane(%d) @ spread = %x : fail reason:(%d)\n", - __func__, link->link_index, (unsigned int)j + 1, attempts, - cur_link_settings.link_rate, cur_link_settings.lane_count, - cur_link_settings.link_spread, status); + if (j == (attempts - 1)) { + DC_LOG_WARNING( + "%s: Link(%d) training attempt %u of %d failed @ rate(%d) x lane(%d) @ spread = %x : fail reason:(%d)\n", + __func__, link->link_index, (unsigned int)j + 1, attempts, + cur_link_settings.link_rate, cur_link_settings.lane_count, + cur_link_settings.link_spread, status); + } else { + DC_LOG_HW_LINK_TRAINING( + "%s: Link(%d) training attempt %u of %d failed @ rate(%d) x lane(%d) @ spread = %x : fail reason:(%d)\n", + __func__, link->link_index, (unsigned int)j + 1, attempts, + cur_link_settings.link_rate, cur_link_settings.lane_count, + cur_link_settings.link_spread, status); + } dp_disable_link_phy(link, &pipe_ctx->link_res, signal); diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_128b_132b.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_128b_132b.c index 23d380f09a21c8a0371e31207d475e59035e2de0..db87cfe37b5c952db91a1d49738fc27fd3171c63 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_128b_132b.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_128b_132b.c @@ -211,11 +211,17 @@ enum link_training_result dp_perform_128b_132b_link_training( dpcd_set_link_settings(link, lt_settings); - if (result == LINK_TRAINING_SUCCESS) + if (result == LINK_TRAINING_SUCCESS) { result = dp_perform_128b_132b_channel_eq_done_sequence(link, link_res, lt_settings); + if (result == LINK_TRAINING_SUCCESS) + DC_LOG_HW_LINK_TRAINING("%s: Channel EQ done.\n", __func__); + } - if (result == LINK_TRAINING_SUCCESS) + if (result == LINK_TRAINING_SUCCESS) { result = dp_perform_128b_132b_cds_done_sequence(link, link_res, lt_settings); + if (result == LINK_TRAINING_SUCCESS) + DC_LOG_HW_LINK_TRAINING("%s: CDS done.\n", __func__); + } return result; } diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_8b_10b.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_8b_10b.c index 3889ebb2256bf384708bddad569f464c3c3712af..2b4c15b0b407084618748f64ce7d31a12cfc4338 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_8b_10b.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_8b_10b.c @@ -388,6 +388,8 @@ enum link_training_result dp_perform_8b_10b_link_training( link_res, lt_settings, repeater_id); + if (status == LINK_TRAINING_SUCCESS) + DC_LOG_HW_LINK_TRAINING("%s: Channel EQ done.\n", __func__); repeater_training_done(link, repeater_id); @@ -409,6 +411,8 @@ enum link_training_result dp_perform_8b_10b_link_training( link_res, lt_settings, DPRX); + if (status == LINK_TRAINING_SUCCESS) + DC_LOG_HW_LINK_TRAINING("%s: Channel EQ done.\n", __func__); } } diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c index 5731c4b61f9f020c2769dd7784736fe7a12b1149..15faaf645b145dfbac9c2ee8c690cd1b46ba1106 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c @@ -233,7 +233,7 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy( link->dpcd_caps.lttpr_caps.phy_repeater_cnt); const uint8_t vendor_lttpr_write_data_intercept_en[4] = {0x1, 0x55, 0x63, 0x0}; const uint8_t vendor_lttpr_write_data_intercept_dis[4] = {0x1, 0x55, 0x63, 0x68}; - uint32_t pre_disable_intercept_delay_ms = link->dc->debug.fixed_vs_aux_delay_config_wa; + uint32_t pre_disable_intercept_delay_ms = 0; uint8_t vendor_lttpr_write_data_vs[4] = {0x1, 0x51, 0x63, 0x0}; uint8_t vendor_lttpr_write_data_pe[4] = {0x1, 0x52, 0x63, 0x0}; uint32_t vendor_lttpr_write_address = 0xF004F; @@ -244,6 +244,10 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy( uint8_t toggle_rate; uint8_t rate; + if (link->local_sink) + pre_disable_intercept_delay_ms = + link->local_sink->edid_caps.panel_patch.delay_disable_aux_intercept_ms; + /* Only 8b/10b is supported */ ASSERT(link_dp_get_encoding_format(<_settings->link_settings) == DP_8b_10b_ENCODING); @@ -259,7 +263,7 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy( /* Certain display and cable configuration require extra delay */ if (offset > 2) - pre_disable_intercept_delay_ms = link->dc->debug.fixed_vs_aux_delay_config_wa * 2; + pre_disable_intercept_delay_ms = pre_disable_intercept_delay_ms * 2; } /* Vendor specific: Reset lane settings */ @@ -380,7 +384,8 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy( 0); /* Vendor specific: Disable intercept */ for (i = 0; i < max_vendor_dpcd_retries; i++) { - msleep(pre_disable_intercept_delay_ms); + if (pre_disable_intercept_delay_ms != 0) + msleep(pre_disable_intercept_delay_ms); dpcd_status = core_link_write_dpcd( link, vendor_lttpr_write_address, @@ -591,10 +596,9 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence( const uint8_t vendor_lttpr_write_data_adicora_eq1[4] = {0x1, 0x55, 0x63, 0x2E}; const uint8_t vendor_lttpr_write_data_adicora_eq2[4] = {0x1, 0x55, 0x63, 0x01}; const uint8_t vendor_lttpr_write_data_adicora_eq3[4] = {0x1, 0x55, 0x63, 0x68}; - uint32_t pre_disable_intercept_delay_ms = link->dc->debug.fixed_vs_aux_delay_config_wa; uint8_t vendor_lttpr_write_data_vs[4] = {0x1, 0x51, 0x63, 0x0}; uint8_t vendor_lttpr_write_data_pe[4] = {0x1, 0x52, 0x63, 0x0}; - + uint32_t pre_disable_intercept_delay_ms = 0; uint32_t vendor_lttpr_write_address = 0xF004F; enum link_training_result status = LINK_TRAINING_SUCCESS; uint8_t lane = 0; @@ -603,6 +607,10 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence( uint8_t toggle_rate; uint8_t rate; + if (link->local_sink) + pre_disable_intercept_delay_ms = + link->local_sink->edid_caps.panel_patch.delay_disable_aux_intercept_ms; + /* Only 8b/10b is supported */ ASSERT(link_dp_get_encoding_format(<_settings->link_settings) == DP_8b_10b_ENCODING); @@ -618,7 +626,7 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence( /* Certain display and cable configuration require extra delay */ if (offset > 2) - pre_disable_intercept_delay_ms = link->dc->debug.fixed_vs_aux_delay_config_wa * 2; + pre_disable_intercept_delay_ms = pre_disable_intercept_delay_ms * 2; } /* Vendor specific: Reset lane settings */ @@ -739,7 +747,8 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence( 0); /* Vendor specific: Disable intercept */ for (i = 0; i < max_vendor_dpcd_retries; i++) { - msleep(pre_disable_intercept_delay_ms); + if (pre_disable_intercept_delay_ms != 0) + msleep(pre_disable_intercept_delay_ms); dpcd_status = core_link_write_dpcd( link, vendor_lttpr_write_address, diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c index 8d1df863659ce6e26b2c2352ed14993522e7a0c8..2039a345f23a174f4689008365bd573c03ee3b40 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c @@ -165,14 +165,35 @@ bool edp_set_backlight_level_nits(struct dc_link *link, *(uint16_t *)&dpcd_backlight_set.backlight_transition_time_ms = (uint16_t)transition_time_in_ms; - if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_LEVEL, + if (!link->dpcd_caps.panel_luminance_control) { + if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_LEVEL, (uint8_t *)(&dpcd_backlight_set), sizeof(dpcd_backlight_set)) != DC_OK) - return false; + return false; - if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_CONTROL, + if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_CONTROL, &backlight_control, 1) != DC_OK) - return false; + return false; + } else { + const uint8_t backlight_enable = DP_EDP_PANEL_LUMINANCE_CONTROL_ENABLE; + struct target_luminance_value *target_luminance = NULL; + + //if target luminance value is greater than 24 bits, clip the value to 24 bits + if (backlight_millinits > 0xFFFFFF) + backlight_millinits = 0xFFFFFF; + + target_luminance = (struct target_luminance_value *)&backlight_millinits; + + if (core_link_write_dpcd(link, DP_EDP_BACKLIGHT_MODE_SET_REGISTER, + &backlight_enable, + sizeof(backlight_enable)) != DC_OK) + return false; + + if (core_link_write_dpcd(link, DP_EDP_PANEL_TARGET_LUMINANCE_VALUE, + (uint8_t *)(target_luminance), + sizeof(struct target_luminance_value)) != DC_OK) + return false; + } return true; } diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h index ba1715e2d25a9e6a7905aa600d9400f230e49fd0..7c9a2b34bd05cda5b5112a6f876dd5847a4591b8 100644 --- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h @@ -271,7 +271,7 @@ struct dmub_srv_hw_params { */ struct dmub_diagnostic_data { uint32_t dmcub_version; - uint32_t scratch[16]; + uint32_t scratch[17]; uint32_t pc; uint32_t undefined_address_fault_addr; uint32_t inst_fetch_fault_addr; @@ -282,6 +282,7 @@ struct dmub_diagnostic_data { uint32_t inbox0_rptr; uint32_t inbox0_wptr; uint32_t inbox0_size; + uint32_t gpint_datain0; uint8_t is_dmcub_enabled : 1; uint8_t is_dmcub_soft_reset : 1; uint8_t is_dmcub_secure_reset : 1; @@ -340,6 +341,8 @@ struct dmub_srv_hw_funcs { void (*setup_mailbox)(struct dmub_srv *dmub, const struct dmub_region *inbox1); + uint32_t (*get_inbox1_wptr)(struct dmub_srv *dmub); + uint32_t (*get_inbox1_rptr)(struct dmub_srv *dmub); void (*set_inbox1_wptr)(struct dmub_srv *dmub, uint32_t wptr_offset); @@ -366,7 +369,6 @@ struct dmub_srv_hw_funcs { bool (*is_hw_init)(struct dmub_srv *dmub); - bool (*is_phy_init)(struct dmub_srv *dmub); void (*enable_dmub_boot_options)(struct dmub_srv *dmub, const struct dmub_srv_hw_params *params); @@ -601,6 +603,18 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub, */ enum dmub_status dmub_srv_hw_reset(struct dmub_srv *dmub); +/** + * dmub_srv_sync_inbox1() - sync sw state with hw state + * @dmub: the dmub service + * + * Sync sw state with hw state when resume from S0i3 + * + * Return: + * DMUB_STATUS_OK - success + * DMUB_STATUS_INVALID - unspecified error + */ +enum dmub_status dmub_srv_sync_inbox1(struct dmub_srv *dmub); + /** * dmub_srv_cmd_queue() - queues a command to the DMUB * @dmub: the dmub service diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h index 598fa1de54ce3ac72545fbdf1e04a63501b1aa11..af1f50742371f83fe02ad22a787e6cb5ecfabf05 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h @@ -257,7 +257,9 @@ struct dmub_feature_caps { */ uint8_t psr; uint8_t fw_assisted_mclk_switch; - uint8_t reserved[6]; + uint8_t reserved[4]; + uint8_t subvp_psr_support; + uint8_t gecc_enable; }; struct dmub_visual_confirm_color { @@ -360,7 +362,7 @@ union dmub_fw_boot_status { uint32_t optimized_init_done : 1; /**< 1 if optimized init done */ uint32_t restore_required : 1; /**< 1 if driver should call restore */ uint32_t defer_load : 1; /**< 1 if VBIOS data is deferred programmed */ - uint32_t reserved : 1; + uint32_t fams_enabled : 1; /**< 1 if VBIOS data is deferred programmed */ uint32_t detection_required: 1; /**< if detection need to be triggered by driver */ uint32_t hw_power_init_done: 1; /**< 1 if hw power init is completed */ } bits; /**< status bits */ @@ -376,6 +378,7 @@ enum dmub_fw_boot_status_bit { DMUB_FW_BOOT_STATUS_BIT_OPTIMIZED_INIT_DONE = (1 << 2), /**< 1 if init done */ DMUB_FW_BOOT_STATUS_BIT_RESTORE_REQUIRED = (1 << 3), /**< 1 if driver should call restore */ DMUB_FW_BOOT_STATUS_BIT_DEFERRED_LOADED = (1 << 4), /**< 1 if VBIOS data is deferred programmed */ + DMUB_FW_BOOT_STATUS_BIT_FAMS_ENABLED = (1 << 5), /**< 1 if FAMS is enabled*/ DMUB_FW_BOOT_STATUS_BIT_DETECTION_REQUIRED = (1 << 6), /**< 1 if detection need to be triggered by driver*/ DMUB_FW_BOOT_STATUS_BIT_HW_POWER_INIT_DONE = (1 << 7), /**< 1 if hw power init is completed */ }; @@ -395,6 +398,12 @@ enum dmub_lvtma_status_bit { DMUB_LVTMA_STATUS_BIT_EDP_ON = (1 << 1), }; +enum dmub_ips_disable_type { + DMUB_IPS_DISABLE_IPS1 = 1, + DMUB_IPS_DISABLE_IPS2 = 2, + DMUB_IPS_DISABLE_IPS2_Z10 = 3, +}; + /** * union dmub_fw_boot_options - Boot option definitions for SCRATCH14 */ @@ -419,7 +428,10 @@ union dmub_fw_boot_options { uint32_t dpia_hpd_int_enable_supported: 1; /* 1 if dpia hpd int enable supported */ uint32_t usb4_dpia_bw_alloc_supported: 1; /* 1 if USB4 dpia BW allocation supported */ uint32_t disable_clk_ds: 1; /* 1 if disallow dispclk_ds and dppclk_ds*/ - uint32_t reserved : 14; /**< reserved */ + uint32_t disable_timeout_recovery : 1; /* 1 if timeout recovery should be disabled */ + uint32_t ips_pg_disable: 1; /* 1 to disable ONO domains power gating*/ + uint32_t ips_disable: 2; /* options to disable ips support*/ + uint32_t reserved : 10; /**< reserved */ } bits; /**< boot bits */ uint32_t all; /**< 32-bit access to bits */ }; @@ -988,16 +1000,25 @@ struct dmub_rb_cmd_mall { }; /** - * enum dmub_cmd_cab_type - TODO: + * enum dmub_cmd_cab_type - CAB command data. */ enum dmub_cmd_cab_type { + /** + * No idle optimizations (i.e. no CAB) + */ DMUB_CMD__CAB_NO_IDLE_OPTIMIZATION = 0, + /** + * No DCN requests for memory + */ DMUB_CMD__CAB_NO_DCN_REQ = 1, + /** + * Fit surfaces in CAB (i.e. CAB enable) + */ DMUB_CMD__CAB_DCN_SS_FIT_IN_CAB = 2, }; /** - * struct dmub_rb_cmd_cab_for_ss - TODO: + * struct dmub_rb_cmd_cab - CAB command data. */ struct dmub_rb_cmd_cab_for_ss { struct dmub_cmd_header header; @@ -1005,6 +1026,9 @@ struct dmub_rb_cmd_cab_for_ss { uint8_t debug_bits; /* debug bits */ }; +/** + * Enum for indicating which MCLK switch mode per pipe + */ enum mclk_switch_mode { NONE = 0, FPO = 1, @@ -1125,8 +1149,6 @@ struct dmub_rb_cmd_idle_opt_dcn_restore { */ struct dmub_dcn_notify_idle_cntl_data { uint8_t driver_idle; - uint8_t d3_entry; - uint8_t trigger; uint8_t pad[1]; }; @@ -3550,6 +3572,10 @@ union dmub_rb_cmd { * Definition of a DMUB_CMD__DPIA_HPD_INT_ENABLE command. */ struct dmub_rb_cmd_dpia_hpd_int_enable dpia_hpd_int_enable; + /** + * Definition of a DMUB_CMD__IDLE_OPT_DCN_NOTIFY_IDLE command. + */ + struct dmub_rb_cmd_idle_opt_dcn_notify_idle idle_opt_notify_idle; }; /** diff --git a/drivers/gpu/drm/amd/display/dmub/src/Makefile b/drivers/gpu/drm/amd/display/dmub/src/Makefile index 0589ad4778eea3d3c4c2153aeb31e4d9107d0296..caf095aca8f3f1379d177080222d8d01a38bf827 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/Makefile +++ b/drivers/gpu/drm/amd/display/dmub/src/Makefile @@ -22,7 +22,7 @@ DMUB = dmub_srv.o dmub_srv_stat.o dmub_reg.o dmub_dcn20.o dmub_dcn21.o DMUB += dmub_dcn30.o dmub_dcn301.o dmub_dcn302.o dmub_dcn303.o -DMUB += dmub_dcn31.o dmub_dcn315.o dmub_dcn316.o +DMUB += dmub_dcn31.o dmub_dcn314.o dmub_dcn315.o dmub_dcn316.o DMUB += dmub_dcn32.o AMD_DAL_DMUB = $(addprefix $(AMDDALPATH)/dmub/src/,$(DMUB)) diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c index a6540e27044d289c67d93eb4389282c8828b27b5..98dad0d47e72cfd87f012b34e61067ee64df9936 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.c @@ -282,6 +282,11 @@ void dmub_dcn20_setup_mailbox(struct dmub_srv *dmub, REG_WRITE(DMCUB_INBOX1_SIZE, inbox1->top - inbox1->base); } +uint32_t dmub_dcn20_get_inbox1_wptr(struct dmub_srv *dmub) +{ + return REG_READ(DMCUB_INBOX1_WPTR); +} + uint32_t dmub_dcn20_get_inbox1_rptr(struct dmub_srv *dmub) { return REG_READ(DMCUB_INBOX1_RPTR); diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.h index c2e5831ac52cc017454587b4c23d423d40ef591c..1df128e57ed3bc4c610242bb91dcdfc9755a612e 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.h +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn20.h @@ -202,6 +202,8 @@ void dmub_dcn20_setup_windows(struct dmub_srv *dmub, void dmub_dcn20_setup_mailbox(struct dmub_srv *dmub, const struct dmub_region *inbox1); +uint32_t dmub_dcn20_get_inbox1_wptr(struct dmub_srv *dmub); + uint32_t dmub_dcn20_get_inbox1_rptr(struct dmub_srv *dmub); void dmub_dcn20_set_inbox1_wptr(struct dmub_srv *dmub, uint32_t wptr_offset); diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.c index 51bb9bceb1b101f164da6bd8141dc37753afaf1b..2d212bc974ccab11ef1de6fb88418ea7077235f8 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.c @@ -54,9 +54,3 @@ const struct dmub_srv_common_regs dmub_srv_dcn21_regs = { #undef DMUB_SF }; -/* Shared functions. */ - -bool dmub_dcn21_is_phy_init(struct dmub_srv *dmub) -{ - return REG_READ(DMCUB_SCRATCH10) == 0; -} diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.h index 6fd5b0cd4ef34aa9c00db6d3e908660b571ed98a..8c4033ae4007dd8eb29b034b76d0e7fc556222b1 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.h +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn21.h @@ -32,8 +32,4 @@ extern const struct dmub_srv_common_regs dmub_srv_dcn21_regs; -/* Hardware functions. */ - -bool dmub_dcn21_is_phy_init(struct dmub_srv *dmub); - #endif /* _DMUB_DCN21_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c index c90b9ee42e126d67b225eed32b759e0287c8e704..ebf7aeec4029cb8a7b03bcf0c6a705a846990c1b 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c @@ -242,6 +242,11 @@ void dmub_dcn31_setup_mailbox(struct dmub_srv *dmub, REG_WRITE(DMCUB_INBOX1_SIZE, inbox1->top - inbox1->base); } +uint32_t dmub_dcn31_get_inbox1_wptr(struct dmub_srv *dmub) +{ + return REG_READ(DMCUB_INBOX1_WPTR); +} + uint32_t dmub_dcn31_get_inbox1_rptr(struct dmub_srv *dmub) { return REG_READ(DMCUB_INBOX1_RPTR); diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.h index f6db6f89d45dc74b6c17a6c506c53a8a0df5ed00..7d5c10ee539b413f4b4c34c52c3e30c92def6584 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.h +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.h @@ -204,6 +204,8 @@ void dmub_dcn31_setup_windows(struct dmub_srv *dmub, void dmub_dcn31_setup_mailbox(struct dmub_srv *dmub, const struct dmub_region *inbox1); +uint32_t dmub_dcn31_get_inbox1_wptr(struct dmub_srv *dmub); + uint32_t dmub_dcn31_get_inbox1_rptr(struct dmub_srv *dmub); void dmub_dcn31_set_inbox1_wptr(struct dmub_srv *dmub, uint32_t wptr_offset); diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn314.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn314.c new file mode 100644 index 0000000000000000000000000000000000000000..48a06dbd9be78687a12f1574ba43906ccf35c13a --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn314.c @@ -0,0 +1,62 @@ +/* + * Copyright 2021 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "../dmub_srv.h" +#include "dmub_reg.h" +#include "dmub_dcn314.h" + +#include "dcn/dcn_3_1_4_offset.h" +#include "dcn/dcn_3_1_4_sh_mask.h" + +#define DCN_BASE__INST0_SEG0 0x00000012 +#define DCN_BASE__INST0_SEG1 0x000000C0 +#define DCN_BASE__INST0_SEG2 0x000034C0 +#define DCN_BASE__INST0_SEG3 0x00009000 +#define DCN_BASE__INST0_SEG4 0x02403C00 +#define DCN_BASE__INST0_SEG5 0 + +#define BASE_INNER(seg) DCN_BASE__INST0_SEG##seg +#define CTX dmub +#define REGS dmub->regs_dcn31 +#define REG_OFFSET_EXP(reg_name) (BASE(reg##reg_name##_BASE_IDX) + reg##reg_name) + +/* Registers. */ + +const struct dmub_srv_dcn31_regs dmub_srv_dcn314_regs = { +#define DMUB_SR(reg) REG_OFFSET_EXP(reg), + { + DMUB_DCN31_REGS() + DMCUB_INTERNAL_REGS() + }, +#undef DMUB_SR + +#define DMUB_SF(reg, field) FD_MASK(reg, field), + { DMUB_DCN31_FIELDS() }, +#undef DMUB_SF + +#define DMUB_SF(reg, field) FD_SHIFT(reg, field), + { DMUB_DCN31_FIELDS() }, +#undef DMUB_SF +}; diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn314.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn314.h new file mode 100644 index 0000000000000000000000000000000000000000..674267a2940e91cb1e5d971dac573da313b9e3d7 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn314.h @@ -0,0 +1,33 @@ +/* + * Copyright 2021 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DMUB_DCN314_H_ +#define _DMUB_DCN314_H_ + +#include "dmub_dcn31.h" + +extern const struct dmub_srv_dcn31_regs dmub_srv_dcn314_regs; + +#endif /* _DMUB_DCN314_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c index 9c20516be066ccd6b1e620b7b61fc78d41ce544f..bf5994e292d9467f533ccf75eab21e0bb45b906e 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c @@ -116,10 +116,6 @@ void dmub_dcn32_reset(struct dmub_srv *dmub) break; } - /* Clear the GPINT command manually so we don't reset again. */ - cmd.all = 0; - dmub->hw_funcs.set_gpint(dmub, cmd); - /* Force reset in case we timed out, DMCUB is likely hung. */ } @@ -133,6 +129,10 @@ void dmub_dcn32_reset(struct dmub_srv *dmub) REG_WRITE(DMCUB_OUTBOX0_RPTR, 0); REG_WRITE(DMCUB_OUTBOX0_WPTR, 0); REG_WRITE(DMCUB_SCRATCH0, 0); + + /* Clear the GPINT command manually so we don't reset again. */ + cmd.all = 0; + dmub->hw_funcs.set_gpint(dmub, cmd); } void dmub_dcn32_reset_release(struct dmub_srv *dmub) @@ -266,6 +266,11 @@ void dmub_dcn32_setup_mailbox(struct dmub_srv *dmub, REG_WRITE(DMCUB_INBOX1_SIZE, inbox1->top - inbox1->base); } +uint32_t dmub_dcn32_get_inbox1_wptr(struct dmub_srv *dmub) +{ + return REG_READ(DMCUB_INBOX1_WPTR); +} + uint32_t dmub_dcn32_get_inbox1_rptr(struct dmub_srv *dmub) { return REG_READ(DMCUB_INBOX1_RPTR); @@ -434,6 +439,7 @@ void dmub_dcn32_get_diagnostic_data(struct dmub_srv *dmub, struct dmub_diagnosti diag_data->scratch[13] = REG_READ(DMCUB_SCRATCH13); diag_data->scratch[14] = REG_READ(DMCUB_SCRATCH14); diag_data->scratch[15] = REG_READ(DMCUB_SCRATCH15); + diag_data->scratch[16] = REG_READ(DMCUB_SCRATCH16); diag_data->undefined_address_fault_addr = REG_READ(DMCUB_UNDEFINED_ADDRESS_FAULT_ADDR); diag_data->inst_fetch_fault_addr = REG_READ(DMCUB_INST_FETCH_FAULT_ADDR); @@ -464,6 +470,8 @@ void dmub_dcn32_get_diagnostic_data(struct dmub_srv *dmub, struct dmub_diagnosti REG_GET(DMCUB_REGION3_CW6_TOP_ADDRESS, DMCUB_REGION3_CW6_ENABLE, &is_cw6_enabled); diag_data->is_cw6_enabled = is_cw6_enabled; + + diag_data->gpint_datain0 = REG_READ(DMCUB_GPINT_DATAIN0); } void dmub_dcn32_configure_dmub_in_system_memory(struct dmub_srv *dmub) { diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.h index 7d1a6eb4d6657869b2307641531d1b8ac48f36fe..d58a1e4b9f1c12340e3afe8235bfed9a509ce914 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.h +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.h @@ -107,6 +107,7 @@ struct dmub_srv; DMUB_SR(DMCUB_SCRATCH15) \ DMUB_SR(DMCUB_SCRATCH16) \ DMUB_SR(DMCUB_SCRATCH17) \ + DMUB_SR(DMCUB_GPINT_DATAIN0) \ DMUB_SR(DMCUB_GPINT_DATAIN1) \ DMUB_SR(DMCUB_GPINT_DATAOUT) \ DMUB_SR(CC_DC_PIPE_DIS) \ @@ -206,6 +207,8 @@ void dmub_dcn32_setup_windows(struct dmub_srv *dmub, void dmub_dcn32_setup_mailbox(struct dmub_srv *dmub, const struct dmub_region *inbox1); +uint32_t dmub_dcn32_get_inbox1_wptr(struct dmub_srv *dmub); + uint32_t dmub_dcn32_get_inbox1_rptr(struct dmub_srv *dmub); void dmub_dcn32_set_inbox1_wptr(struct dmub_srv *dmub, uint32_t wptr_offset); diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c index 92c18bfb98b3badee51b02a0f2f8109971bda035..9e9a6a44a7ac4b4a949a2e8261d21dd3f218a15a 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c @@ -32,6 +32,7 @@ #include "dmub_dcn302.h" #include "dmub_dcn303.h" #include "dmub_dcn31.h" +#include "dmub_dcn314.h" #include "dmub_dcn315.h" #include "dmub_dcn316.h" #include "dmub_dcn32.h" @@ -166,6 +167,7 @@ static bool dmub_srv_hw_setup(struct dmub_srv *dmub, enum dmub_asic asic) funcs->backdoor_load = dmub_dcn20_backdoor_load; funcs->setup_windows = dmub_dcn20_setup_windows; funcs->setup_mailbox = dmub_dcn20_setup_mailbox; + funcs->get_inbox1_wptr = dmub_dcn20_get_inbox1_wptr; funcs->get_inbox1_rptr = dmub_dcn20_get_inbox1_rptr; funcs->set_inbox1_wptr = dmub_dcn20_set_inbox1_wptr; funcs->is_supported = dmub_dcn20_is_supported; @@ -190,11 +192,9 @@ static bool dmub_srv_hw_setup(struct dmub_srv *dmub, enum dmub_asic asic) funcs->get_diagnostic_data = dmub_dcn20_get_diagnostic_data; - if (asic == DMUB_ASIC_DCN21) { + if (asic == DMUB_ASIC_DCN21) dmub->regs = &dmub_srv_dcn21_regs; - funcs->is_phy_init = dmub_dcn21_is_phy_init; - } if (asic == DMUB_ASIC_DCN30) { dmub->regs = &dmub_srv_dcn30_regs; @@ -226,7 +226,9 @@ static bool dmub_srv_hw_setup(struct dmub_srv *dmub, enum dmub_asic asic) case DMUB_ASIC_DCN314: case DMUB_ASIC_DCN315: case DMUB_ASIC_DCN316: - if (asic == DMUB_ASIC_DCN315) + if (asic == DMUB_ASIC_DCN314) + dmub->regs_dcn31 = &dmub_srv_dcn314_regs; + else if (asic == DMUB_ASIC_DCN315) dmub->regs_dcn31 = &dmub_srv_dcn315_regs; else if (asic == DMUB_ASIC_DCN316) dmub->regs_dcn31 = &dmub_srv_dcn316_regs; @@ -237,6 +239,7 @@ static bool dmub_srv_hw_setup(struct dmub_srv *dmub, enum dmub_asic asic) funcs->backdoor_load = dmub_dcn31_backdoor_load; funcs->setup_windows = dmub_dcn31_setup_windows; funcs->setup_mailbox = dmub_dcn31_setup_mailbox; + funcs->get_inbox1_wptr = dmub_dcn31_get_inbox1_wptr; funcs->get_inbox1_rptr = dmub_dcn31_get_inbox1_rptr; funcs->set_inbox1_wptr = dmub_dcn31_set_inbox1_wptr; funcs->setup_out_mailbox = dmub_dcn31_setup_out_mailbox; @@ -275,6 +278,7 @@ static bool dmub_srv_hw_setup(struct dmub_srv *dmub, enum dmub_asic asic) funcs->backdoor_load_zfb_mode = dmub_dcn32_backdoor_load_zfb_mode; funcs->setup_windows = dmub_dcn32_setup_windows; funcs->setup_mailbox = dmub_dcn32_setup_mailbox; + funcs->get_inbox1_wptr = dmub_dcn32_get_inbox1_wptr; funcs->get_inbox1_rptr = dmub_dcn32_get_inbox1_rptr; funcs->set_inbox1_wptr = dmub_dcn32_set_inbox1_wptr; funcs->setup_out_mailbox = dmub_dcn32_setup_out_mailbox; @@ -644,6 +648,20 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub, return DMUB_STATUS_OK; } +enum dmub_status dmub_srv_sync_inbox1(struct dmub_srv *dmub) +{ + if (!dmub->sw_init) + return DMUB_STATUS_INVALID; + + if (dmub->hw_funcs.get_inbox1_rptr && dmub->hw_funcs.get_inbox1_wptr) { + dmub->inbox1_rb.rptr = dmub->hw_funcs.get_inbox1_rptr(dmub); + dmub->inbox1_rb.wrpt = dmub->hw_funcs.get_inbox1_wptr(dmub); + dmub->inbox1_last_wptr = dmub->inbox1_rb.wrpt; + } + + return DMUB_STATUS_OK; +} + enum dmub_status dmub_srv_hw_reset(struct dmub_srv *dmub) { if (!dmub->sw_init) @@ -721,27 +739,6 @@ enum dmub_status dmub_srv_wait_for_auto_load(struct dmub_srv *dmub, return DMUB_STATUS_TIMEOUT; } -enum dmub_status dmub_srv_wait_for_phy_init(struct dmub_srv *dmub, - uint32_t timeout_us) -{ - uint32_t i = 0; - - if (!dmub->hw_init) - return DMUB_STATUS_INVALID; - - if (!dmub->hw_funcs.is_phy_init) - return DMUB_STATUS_OK; - - for (i = 0; i <= timeout_us; i += 10) { - if (dmub->hw_funcs.is_phy_init(dmub)) - return DMUB_STATUS_OK; - - udelay(10); - } - - return DMUB_STATUS_TIMEOUT; -} - enum dmub_status dmub_srv_wait_for_idle(struct dmub_srv *dmub, uint32_t timeout_us) { diff --git a/drivers/gpu/drm/amd/display/include/fixed31_32.h b/drivers/gpu/drm/amd/display/include/fixed31_32.h index ece97ae0e826c746e4a85d0375a649f246b9ad08..d4cf7ead1d877e7bfd1a75fa651c81e6062cefc7 100644 --- a/drivers/gpu/drm/amd/display/include/fixed31_32.h +++ b/drivers/gpu/drm/amd/display/include/fixed31_32.h @@ -525,7 +525,7 @@ static inline struct fixed31_32 dc_fixpt_truncate(struct fixed31_32 arg, unsigne if (negative) arg.value = -arg.value; - arg.value &= (~0LL) << (FIXED31_32_BITS_PER_FRACTIONAL_PART - frac_bits); + arg.value &= (~0ULL) << (FIXED31_32_BITS_PER_FRACTIONAL_PART - frac_bits); if (negative) arg.value = -arg.value; return arg; diff --git a/drivers/gpu/drm/amd/display/include/signal_types.h b/drivers/gpu/drm/amd/display/include/signal_types.h index 23a308c3eccbf6362c356d0ad9f8a029cb1feac3..325c5ba4c82a60030ea2ef4b93269375946bd36d 100644 --- a/drivers/gpu/drm/amd/display/include/signal_types.h +++ b/drivers/gpu/drm/amd/display/include/signal_types.h @@ -44,6 +44,34 @@ enum signal_type { SIGNAL_TYPE_VIRTUAL = (1 << 9), /* Virtual Display */ }; +static inline const char *signal_type_to_string(const int type) +{ + switch (type) { + case SIGNAL_TYPE_NONE: + return "No signal"; + case SIGNAL_TYPE_DVI_SINGLE_LINK: + return "DVI: Single Link"; + case SIGNAL_TYPE_DVI_DUAL_LINK: + return "DVI: Dual Link"; + case SIGNAL_TYPE_HDMI_TYPE_A: + return "HDMI: TYPE A"; + case SIGNAL_TYPE_LVDS: + return "LVDS"; + case SIGNAL_TYPE_RGB: + return "RGB"; + case SIGNAL_TYPE_DISPLAY_PORT: + return "Display Port"; + case SIGNAL_TYPE_DISPLAY_PORT_MST: + return "Display Port: MST"; + case SIGNAL_TYPE_EDP: + return "Embedded Display Port"; + case SIGNAL_TYPE_VIRTUAL: + return "Virtual"; + default: + return "Unknown"; + } +} + /* help functions for signal types manipulation */ static inline bool dc_is_hdmi_tmds_signal(enum signal_type signal) { diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c index 5c41a4751db445c114167fdff7d0b63375bc11e0..dbd60811f95da4a8d17f76784758ce363f3e0058 100644 --- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c +++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c @@ -1,5 +1,5 @@ /* - * Copyright 2016 Advanced Micro Devices, Inc. + * Copyright 2016-2023 Advanced Micro Devices, Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -989,6 +989,7 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, unsigned int refresh_range = 0; unsigned long long min_refresh_in_uhz = 0; unsigned long long max_refresh_in_uhz = 0; + unsigned long long min_hardware_refresh_in_uhz = 0; if (mod_freesync == NULL) return; @@ -999,7 +1000,13 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, nominal_field_rate_in_uhz = mod_freesync_calc_nominal_field_rate(stream); - min_refresh_in_uhz = in_config->min_refresh_in_uhz; + if (stream->ctx->dc->caps.max_v_total != 0 && stream->timing.h_total != 0) { + min_hardware_refresh_in_uhz = div64_u64((stream->timing.pix_clk_100hz * 100000000ULL), + (stream->timing.h_total * stream->ctx->dc->caps.max_v_total)); + } + /* Limit minimum refresh rate to what can be supported by hardware */ + min_refresh_in_uhz = min_hardware_refresh_in_uhz > in_config->min_refresh_in_uhz ? + min_hardware_refresh_in_uhz : in_config->min_refresh_in_uhz; max_refresh_in_uhz = in_config->max_refresh_in_uhz; /* Full range may be larger than current video timing, so cap at nominal */ @@ -1137,10 +1144,6 @@ void mod_freesync_handle_preflip(struct mod_freesync *mod_freesync, if (in_out_vrr->supported && in_out_vrr->state == VRR_STATE_ACTIVE_VARIABLE) { - unsigned int oldest_index = plane->time.index + 1; - - if (oldest_index >= DC_PLANE_UPDATE_TIMES_MAX) - oldest_index = 0; last_render_time_in_us = curr_time_stamp_in_us - plane->time.prev_update_time_in_us; diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c index 51e76bce92eaa24f92adf41a7bab973428154b16..30349881a283aeeb4e13ba660608dc9387a2bf41 100644 --- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c +++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c @@ -116,6 +116,27 @@ static const struct abm_parameters * const abm_settings[] = { abm_settings_config2, }; +static const struct dm_bl_data_point custom_backlight_curve0[] = { + {2, 14}, {4, 16}, {6, 18}, {8, 21}, {10, 23}, {12, 26}, {14, 29}, {16, 32}, {18, 35}, + {20, 38}, {22, 41}, {24, 44}, {26, 48}, {28, 52}, {30, 55}, {32, 59}, {34, 62}, + {36, 67}, {38, 71}, {40, 75}, {42, 80}, {44, 84}, {46, 88}, {48, 93}, {50, 98}, + {52, 103}, {54, 108}, {56, 113}, {58, 118}, {60, 123}, {62, 129}, {64, 135}, {66, 140}, + {68, 146}, {70, 152}, {72, 158}, {74, 164}, {76, 171}, {78, 177}, {80, 183}, {82, 190}, + {84, 197}, {86, 204}, {88, 211}, {90, 218}, {92, 225}, {94, 232}, {96, 240}, {98, 247}}; + +struct custom_backlight_profile { + uint8_t ac_level_percentage; + uint8_t dc_level_percentage; + uint8_t min_input_signal; + uint8_t max_input_signal; + uint8_t num_data_points; + const struct dm_bl_data_point *data_points; +}; + +static const struct custom_backlight_profile custom_backlight_profiles[] = { + {100, 32, 12, 255, ARRAY_SIZE(custom_backlight_curve0), custom_backlight_curve0}, +}; + #define NUM_AMBI_LEVEL 5 #define NUM_AGGR_LEVEL 4 #define NUM_POWER_FN_SEGS 8 @@ -944,3 +965,25 @@ bool psr_su_set_dsc_slice_height(struct dc *dc, struct dc_link *link, return true; } + +bool fill_custom_backlight_caps(unsigned int config_no, struct dm_acpi_atif_backlight_caps *caps) +{ + unsigned int data_points_size; + + if (config_no >= ARRAY_SIZE(custom_backlight_profiles)) + return false; + + data_points_size = custom_backlight_profiles[config_no].num_data_points + * sizeof(custom_backlight_profiles[config_no].data_points[0]); + + caps->size = sizeof(struct dm_acpi_atif_backlight_caps) - sizeof(caps->data_points) + data_points_size; + caps->flags = 0; + caps->error_code = 0; + caps->ac_level_percentage = custom_backlight_profiles[config_no].ac_level_percentage; + caps->dc_level_percentage = custom_backlight_profiles[config_no].dc_level_percentage; + caps->min_input_signal = custom_backlight_profiles[config_no].min_input_signal; + caps->max_input_signal = custom_backlight_profiles[config_no].max_input_signal; + caps->num_data_points = custom_backlight_profiles[config_no].num_data_points; + memcpy(caps->data_points, custom_backlight_profiles[config_no].data_points, data_points_size); + return true; +} diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.h b/drivers/gpu/drm/amd/display/modules/power/power_helpers.h index 1d3079e56799f3daf4c9d119873a036d52f93e13..ffc924c9991bfe2644963a0213decddf93e988b7 100644 --- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.h +++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.h @@ -62,4 +62,7 @@ bool mod_power_only_edp(const struct dc_state *context, bool psr_su_set_dsc_slice_height(struct dc *dc, struct dc_link *link, struct dc_stream_state *stream, struct psr_config *config); + +bool fill_custom_backlight_caps(unsigned int config_no, + struct dm_acpi_atif_backlight_caps *caps); #endif /* MODULES_POWER_POWER_HELPERS_H_ */ diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h index e4a22c68517d1dc2be04b452d70ac3b096f5aa86..f175e65b853a001ea1bcc66d8edfd5b726cbc952 100644 --- a/drivers/gpu/drm/amd/include/amd_shared.h +++ b/drivers/gpu/drm/amd/include/amd_shared.h @@ -240,7 +240,6 @@ enum DC_FEATURE_MASK { DC_DISABLE_LTTPR_DP2_0 = (1 << 6), //0x40, disabled by default DC_PSR_ALLOW_SMU_OPT = (1 << 7), //0x80, disabled by default DC_PSR_ALLOW_MULTI_DISP_OPT = (1 << 8), //0x100, disabled by default - DC_ENABLE_SUBVP_DRR = (1 << 9), // 0x200, disabled by default }; enum DC_DEBUG_MASK { diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_offset.h index 79c41004c0b62da870d85c598e136ea121543664..4908044f74090b7aa55e152ccea58508cd9d6c34 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_offset.h @@ -5194,6 +5194,20 @@ #define mmSPI_WCL_PIPE_PERCENT_CS6_BASE_IDX 0 #define mmSPI_WCL_PIPE_PERCENT_CS7 0x1f70 #define mmSPI_WCL_PIPE_PERCENT_CS7_BASE_IDX 0 +#define mmSPI_GDBG_WAVE_CNTL 0x1f71 +#define mmSPI_GDBG_WAVE_CNTL_BASE_IDX 0 +#define mmSPI_GDBG_TRAP_CONFIG 0x1f72 +#define mmSPI_GDBG_TRAP_CONFIG_BASE_IDX 0 +#define mmSPI_GDBG_TRAP_MASK 0x1f73 +#define mmSPI_GDBG_TRAP_MASK_BASE_IDX 0 +#define mmSPI_GDBG_WAVE_CNTL2 0x1f74 +#define mmSPI_GDBG_WAVE_CNTL2_BASE_IDX 0 +#define mmSPI_GDBG_WAVE_CNTL3 0x1f75 +#define mmSPI_GDBG_WAVE_CNTL3_BASE_IDX 0 +#define mmSPI_GDBG_TRAP_DATA0 0x1f78 +#define mmSPI_GDBG_TRAP_DATA0_BASE_IDX 0 +#define mmSPI_GDBG_TRAP_DATA1 0x1f79 +#define mmSPI_GDBG_TRAP_DATA1_BASE_IDX 0 #define mmSPI_COMPUTE_QUEUE_RESET 0x1f7b #define mmSPI_COMPUTE_QUEUE_RESET_BASE_IDX 0 #define mmSPI_RESOURCE_RESERVE_CU_0 0x1f7c diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_sh_mask.h index 52043e14306787568d319eb8774ddb3d4cad61cf..9b7d219e7954f2bab5a69a8d7eecb67d1f0a547b 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_1_0_sh_mask.h @@ -19700,6 +19700,75 @@ //SPI_WCL_PIPE_PERCENT_CS7 #define SPI_WCL_PIPE_PERCENT_CS7__VALUE__SHIFT 0x0 #define SPI_WCL_PIPE_PERCENT_CS7__VALUE_MASK 0x7FL +//SPI_GDBG_WAVE_CNTL +#define SPI_GDBG_WAVE_CNTL__STALL_RA__SHIFT 0x0 +#define SPI_GDBG_WAVE_CNTL__STALL_VMID__SHIFT 0x1 +#define SPI_GDBG_WAVE_CNTL__STALL_RA_MASK 0x00000001L +#define SPI_GDBG_WAVE_CNTL__STALL_VMID_MASK 0x0001FFFEL +//SPI_GDBG_TRAP_CONFIG +#define SPI_GDBG_TRAP_CONFIG__ME_SEL__SHIFT 0x0 +#define SPI_GDBG_TRAP_CONFIG__PIPE_SEL__SHIFT 0x2 +#define SPI_GDBG_TRAP_CONFIG__QUEUE_SEL__SHIFT 0x4 +#define SPI_GDBG_TRAP_CONFIG__ME_MATCH__SHIFT 0x7 +#define SPI_GDBG_TRAP_CONFIG__PIPE_MATCH__SHIFT 0x8 +#define SPI_GDBG_TRAP_CONFIG__QUEUE_MATCH__SHIFT 0x9 +#define SPI_GDBG_TRAP_CONFIG__TRAP_EN__SHIFT 0xf +#define SPI_GDBG_TRAP_CONFIG__VMID_SEL__SHIFT 0x10 +#define SPI_GDBG_TRAP_CONFIG__ME_SEL_MASK 0x00000003L +#define SPI_GDBG_TRAP_CONFIG__PIPE_SEL_MASK 0x0000000CL +#define SPI_GDBG_TRAP_CONFIG__QUEUE_SEL_MASK 0x00000070L +#define SPI_GDBG_TRAP_CONFIG__ME_MATCH_MASK 0x00000080L +#define SPI_GDBG_TRAP_CONFIG__PIPE_MATCH_MASK 0x00000100L +#define SPI_GDBG_TRAP_CONFIG__QUEUE_MATCH_MASK 0x00000200L +#define SPI_GDBG_TRAP_CONFIG__TRAP_EN_MASK 0x00008000L +#define SPI_GDBG_TRAP_CONFIG__VMID_SEL_MASK 0xFFFF0000L +//SPI_GDBG_TRAP_MASK +#define SPI_GDBG_TRAP_MASK__EXCP_EN__SHIFT 0x0 +#define SPI_GDBG_TRAP_MASK__REPLACE__SHIFT 0x9 +#define SPI_GDBG_TRAP_MASK__EXCP_EN_MASK 0x01FFL +#define SPI_GDBG_TRAP_MASK__REPLACE_MASK 0x0200L +//SPI_GDBG_WAVE_CNTL2 +#define SPI_GDBG_WAVE_CNTL2__VMID_MASK__SHIFT 0x0 +#define SPI_GDBG_WAVE_CNTL2__MODE__SHIFT 0x10 +#define SPI_GDBG_WAVE_CNTL2__VMID_MASK_MASK 0x0000FFFFL +#define SPI_GDBG_WAVE_CNTL2__MODE_MASK 0x00030000L +//SPI_GDBG_WAVE_CNTL3 +#define SPI_GDBG_WAVE_CNTL3__STALL_PS__SHIFT 0x0 +#define SPI_GDBG_WAVE_CNTL3__STALL_VS__SHIFT 0x1 +#define SPI_GDBG_WAVE_CNTL3__STALL_GS__SHIFT 0x2 +#define SPI_GDBG_WAVE_CNTL3__STALL_HS__SHIFT 0x3 +#define SPI_GDBG_WAVE_CNTL3__STALL_CSG__SHIFT 0x4 +#define SPI_GDBG_WAVE_CNTL3__STALL_CS0__SHIFT 0x5 +#define SPI_GDBG_WAVE_CNTL3__STALL_CS1__SHIFT 0x6 +#define SPI_GDBG_WAVE_CNTL3__STALL_CS2__SHIFT 0x7 +#define SPI_GDBG_WAVE_CNTL3__STALL_CS3__SHIFT 0x8 +#define SPI_GDBG_WAVE_CNTL3__STALL_CS4__SHIFT 0x9 +#define SPI_GDBG_WAVE_CNTL3__STALL_CS5__SHIFT 0xa +#define SPI_GDBG_WAVE_CNTL3__STALL_CS6__SHIFT 0xb +#define SPI_GDBG_WAVE_CNTL3__STALL_CS7__SHIFT 0xc +#define SPI_GDBG_WAVE_CNTL3__STALL_DURATION__SHIFT 0xd +#define SPI_GDBG_WAVE_CNTL3__STALL_MULT__SHIFT 0x1c +#define SPI_GDBG_WAVE_CNTL3__STALL_PS_MASK 0x00000001L +#define SPI_GDBG_WAVE_CNTL3__STALL_VS_MASK 0x00000002L +#define SPI_GDBG_WAVE_CNTL3__STALL_GS_MASK 0x00000004L +#define SPI_GDBG_WAVE_CNTL3__STALL_HS_MASK 0x00000008L +#define SPI_GDBG_WAVE_CNTL3__STALL_CSG_MASK 0x00000010L +#define SPI_GDBG_WAVE_CNTL3__STALL_CS0_MASK 0x00000020L +#define SPI_GDBG_WAVE_CNTL3__STALL_CS1_MASK 0x00000040L +#define SPI_GDBG_WAVE_CNTL3__STALL_CS2_MASK 0x00000080L +#define SPI_GDBG_WAVE_CNTL3__STALL_CS3_MASK 0x00000100L +#define SPI_GDBG_WAVE_CNTL3__STALL_CS4_MASK 0x00000200L +#define SPI_GDBG_WAVE_CNTL3__STALL_CS5_MASK 0x00000400L +#define SPI_GDBG_WAVE_CNTL3__STALL_CS6_MASK 0x00000800L +#define SPI_GDBG_WAVE_CNTL3__STALL_CS7_MASK 0x00001000L +#define SPI_GDBG_WAVE_CNTL3__STALL_DURATION_MASK 0x0FFFE000L +#define SPI_GDBG_WAVE_CNTL3__STALL_MULT_MASK 0x10000000L +//SPI_GDBG_TRAP_DATA0 +#define SPI_GDBG_TRAP_DATA0__DATA__SHIFT 0x0 +#define SPI_GDBG_TRAP_DATA0__DATA_MASK 0xFFFFFFFFL +//SPI_GDBG_TRAP_DATA1 +#define SPI_GDBG_TRAP_DATA1__DATA__SHIFT 0x0 +#define SPI_GDBG_TRAP_DATA1__DATA_MASK 0xFFFFFFFFL //SPI_COMPUTE_QUEUE_RESET #define SPI_COMPUTE_QUEUE_RESET__RESET__SHIFT 0x0 #define SPI_COMPUTE_QUEUE_RESET__RESET_MASK 0x01L diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_offset.h index a734abaa91a59feefd2c10a6d035f7f443c81e73..5e15ac14b63c8e84c7a483b195f92bd8d3cbfa98 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_offset.h @@ -26,6 +26,8 @@ #define mmSQ_DEBUG_STS_GLOBAL_BASE_IDX 0 #define mmSQ_DEBUG_STS_GLOBAL2 0x10B0 #define mmSQ_DEBUG_STS_GLOBAL2_BASE_IDX 0 +#define mmSQ_DEBUG 0x10B1 +#define mmSQ_DEBUG_BASE_IDX 0 // addressBlock: gc_sdma0_sdma0dec // base address: 0x4980 @@ -4853,10 +4855,18 @@ #define mmSPI_WCL_PIPE_PERCENT_CS3_BASE_IDX 0 #define mmSPI_GDBG_WAVE_CNTL 0x1f71 #define mmSPI_GDBG_WAVE_CNTL_BASE_IDX 0 +#define mmSPI_GDBG_TRAP_CONFIG 0x1f72 +#define mmSPI_GDBG_TRAP_CONFIG_BASE_IDX 0 #define mmSPI_GDBG_TRAP_MASK 0x1f73 #define mmSPI_GDBG_TRAP_MASK_BASE_IDX 0 #define mmSPI_GDBG_WAVE_CNTL2 0x1f74 #define mmSPI_GDBG_WAVE_CNTL2_BASE_IDX 0 +#define mmSPI_GDBG_WAVE_CNTL3 0x1f75 +#define mmSPI_GDBG_WAVE_CNTL3_BASE_IDX 0 +#define mmSPI_GDBG_TRAP_DATA0 0x1f78 +#define mmSPI_GDBG_TRAP_DATA0_BASE_IDX 0 +#define mmSPI_GDBG_TRAP_DATA1 0x1f79 +#define mmSPI_GDBG_TRAP_DATA1_BASE_IDX 0 #define mmSPI_COMPUTE_QUEUE_RESET 0x1f7b #define mmSPI_COMPUTE_QUEUE_RESET_BASE_IDX 0 #define mmSPI_RESOURCE_RESERVE_CU_0 0x1f7c diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_sh_mask.h index d7a17bae2584649fa1551e65a0e78beca4669704..e4ecd6c2d20e2e2d91c7d47ef3d543eba91b13e1 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_10_3_0_sh_mask.h @@ -47907,6 +47907,10 @@ // addressBlock: sqind +//SQ_DEBUG +#define SQ_DEBUG__SINGLE_MEMOP_MASK 0x00000001L +#define SQ_DEBUG__SINGLE_MEMOP__SHIFT 0x00000000 + //SQ_DEBUG_STS_GLOBAL #define SQ_DEBUG_STS_GLOBAL2__FIFO_LEVEL_GFX0_MASK 0x000000ffL #define SQ_DEBUG_STS_GLOBAL2__FIFO_LEVEL_GFX0__SHIFT 0x00000000 diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_11_0_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_11_0_0_sh_mask.h index 4f08f90856fc70f801adbc574e31fce0d0f02a81..3088a4a13cb50b918dd190711ec2a84532a28200 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_11_0_0_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_11_0_0_sh_mask.h @@ -17216,11 +17216,15 @@ #define SPI_GDBG_PER_VMID_CNTL__TRAP_EN__SHIFT 0x3 #define SPI_GDBG_PER_VMID_CNTL__EXCP_EN__SHIFT 0x4 #define SPI_GDBG_PER_VMID_CNTL__EXCP_REPLACE__SHIFT 0xd +#define SPI_GDBG_PER_VMID_CNTL__TRAP_ON_START__SHIFT 0xe +#define SPI_GDBG_PER_VMID_CNTL__TRAP_ON_END__SHIFT 0xf #define SPI_GDBG_PER_VMID_CNTL__STALL_VMID_MASK 0x00000001L #define SPI_GDBG_PER_VMID_CNTL__LAUNCH_MODE_MASK 0x00000006L #define SPI_GDBG_PER_VMID_CNTL__TRAP_EN_MASK 0x00000008L #define SPI_GDBG_PER_VMID_CNTL__EXCP_EN_MASK 0x00001FF0L #define SPI_GDBG_PER_VMID_CNTL__EXCP_REPLACE_MASK 0x00002000L +#define SPI_GDBG_PER_VMID_CNTL__TRAP_ON_START_MASK 0x00004000L +#define SPI_GDBG_PER_VMID_CNTL__TRAP_ON_END_MASK 0x00008000L //SPI_COMPUTE_QUEUE_RESET #define SPI_COMPUTE_QUEUE_RESET__RESET__SHIFT 0x0 #define SPI_COMPUTE_QUEUE_RESET__RESET_MASK 0x01L diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_3_offset.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_3_offset.h index 3100de8b3881011f86defa3e0aee79868b104c73..393963502b7a6646ce8f72932661f89e891fb4aa 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_3_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_3_offset.h @@ -705,6 +705,46 @@ #define regSQC_ICACHE_UTCL1_STATUS_BASE_IDX 0 #define regSQC_DCACHE_UTCL1_STATUS 0x03d8 #define regSQC_DCACHE_UTCL1_STATUS_BASE_IDX 0 +#define regSQC_UE_EDC_LO 0x03d9 +#define regSQC_UE_EDC_LO_BASE_IDX 0 +#define regSQC_UE_EDC_HI 0x03da +#define regSQC_UE_EDC_HI_BASE_IDX 0 +#define regSQC_CE_EDC_LO 0x03db +#define regSQC_CE_EDC_LO_BASE_IDX 0 +#define regSQC_CE_EDC_HI 0x03dc +#define regSQC_CE_EDC_HI_BASE_IDX 0 +#define regSQ_UE_ERR_STATUS_LO 0x03dd +#define regSQ_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regSQ_UE_ERR_STATUS_HI 0x03de +#define regSQ_UE_ERR_STATUS_HI_BASE_IDX 0 +#define regSQ_CE_ERR_STATUS_LO 0x03df +#define regSQ_CE_ERR_STATUS_LO_BASE_IDX 0 +#define regSQ_CE_ERR_STATUS_HI 0x03e0 +#define regSQ_CE_ERR_STATUS_HI_BASE_IDX 0 +#define regLDS_UE_ERR_STATUS_LO 0x03e1 +#define regLDS_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regLDS_UE_ERR_STATUS_HI 0x03e2 +#define regLDS_UE_ERR_STATUS_HI_BASE_IDX 0 +#define regLDS_CE_ERR_STATUS_LO 0x03e3 +#define regLDS_CE_ERR_STATUS_LO_BASE_IDX 0 +#define regLDS_CE_ERR_STATUS_HI 0x03e4 +#define regLDS_CE_ERR_STATUS_HI_BASE_IDX 0 +#define regSP0_UE_ERR_STATUS_LO 0x03e5 +#define regSP0_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regSP0_UE_ERR_STATUS_HI 0x03e6 +#define regSP0_UE_ERR_STATUS_HI_BASE_IDX 0 +#define regSP0_CE_ERR_STATUS_LO 0x03e7 +#define regSP0_CE_ERR_STATUS_LO_BASE_IDX 0 +#define regSP0_CE_ERR_STATUS_HI 0x03e8 +#define regSP0_CE_ERR_STATUS_HI_BASE_IDX 0 +#define regSP1_UE_ERR_STATUS_LO 0x03e9 +#define regSP1_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regSP1_UE_ERR_STATUS_HI 0x03ea +#define regSP1_UE_ERR_STATUS_HI_BASE_IDX 0 +#define regSP1_CE_ERR_STATUS_LO 0x03eb +#define regSP1_CE_ERR_STATUS_LO_BASE_IDX 0 +#define regSP1_CE_ERR_STATUS_HI 0x03ec +#define regSP1_CE_ERR_STATUS_HI_BASE_IDX 0 // addressBlock: xcd0_gc_shsdec @@ -727,6 +767,14 @@ #define regSPI_DSM_CNTL2_BASE_IDX 0 #define regSPI_EDC_CNT 0x0445 #define regSPI_EDC_CNT_BASE_IDX 0 +#define regSPI_UE_ERR_STATUS_LO 0x0446 +#define regSPI_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regSPI_UE_ERR_STATUS_HI 0x0447 +#define regSPI_UE_ERR_STATUS_HI_BASE_IDX 0 +#define regSPI_CE_ERR_STATUS_LO 0x0448 +#define regSPI_CE_ERR_STATUS_LO_BASE_IDX 0 +#define regSPI_CE_ERR_STATUS_HI 0x0449 +#define regSPI_CE_ERR_STATUS_HI_BASE_IDX 0 #define regSPI_DEBUG_BUSY 0x0450 #define regSPI_DEBUG_BUSY_BASE_IDX 0 #define regSPI_CONFIG_PS_CU_EN 0x0452 @@ -871,6 +919,14 @@ #define regTD_STATUS_BASE_IDX 0 #define regTD_POWER_CNTL 0x052a #define regTD_POWER_CNTL_BASE_IDX 0 +#define regTD_UE_EDC_LO 0x052b +#define regTD_UE_EDC_LO_BASE_IDX 0 +#define regTD_UE_EDC_HI 0x052c +#define regTD_UE_EDC_HI_BASE_IDX 0 +#define regTD_CE_EDC_LO 0x052d +#define regTD_CE_EDC_LO_BASE_IDX 0 +#define regTD_CE_EDC_HI 0x052e +#define regTD_CE_EDC_HI_BASE_IDX 0 #define regTD_DSM_CNTL 0x052f #define regTD_DSM_CNTL_BASE_IDX 0 #define regTD_DSM_CNTL2 0x0530 @@ -893,6 +949,14 @@ #define regTA_DSM_CNTL_BASE_IDX 0 #define regTA_DSM_CNTL2 0x0585 #define regTA_DSM_CNTL2_BASE_IDX 0 +#define regTA_UE_EDC_LO 0x0587 +#define regTA_UE_EDC_LO_BASE_IDX 0 +#define regTA_UE_EDC_HI 0x0588 +#define regTA_UE_EDC_HI_BASE_IDX 0 +#define regTA_CE_EDC_LO 0x0589 +#define regTA_CE_EDC_LO_BASE_IDX 0 +#define regTA_CE_EDC_HI 0x058a +#define regTA_CE_EDC_HI_BASE_IDX 0 // addressBlock: xcd0_gc_gdsdec @@ -923,6 +987,14 @@ #define regGDS_DSM_CNTL2_BASE_IDX 0 #define regGDS_WD_GDS_CSB 0x05ce #define regGDS_WD_GDS_CSB_BASE_IDX 0 +#define regGDS_UE_ERR_STATUS_LO 0x05cf +#define regGDS_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regGDS_UE_ERR_STATUS_HI 0x05d0 +#define regGDS_UE_ERR_STATUS_HI_BASE_IDX 0 +#define regGDS_CE_ERR_STATUS_LO 0x05d1 +#define regGDS_CE_ERR_STATUS_LO_BASE_IDX 0 +#define regGDS_CE_ERR_STATUS_HI 0x05d2 +#define regGDS_CE_ERR_STATUS_HI_BASE_IDX 0 // addressBlock: xcd0_gc_rbdec @@ -1243,6 +1315,10 @@ #define regGCEA_MAM_CTRL_BASE_IDX 0 #define regGCEA_MAM_CTRL2 0x0702 #define regGCEA_MAM_CTRL2_BASE_IDX 0 +#define regGCEA_UE_ERR_STATUS_LO 0x0706 +#define regGCEA_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regGCEA_UE_ERR_STATUS_HI 0x0707 +#define regGCEA_UE_ERR_STATUS_HI_BASE_IDX 0 #define regGCEA_DSM_CNTL 0x0708 #define regGCEA_DSM_CNTL_BASE_IDX 0 #define regGCEA_DSM_CNTLA 0x0709 @@ -1277,6 +1353,10 @@ #define regGCEA_SDP_BACKDOOR_DATACREDITS1_BASE_IDX 0 #define regGCEA_SDP_BACKDOOR_MISCCREDITS 0x0719 #define regGCEA_SDP_BACKDOOR_MISCCREDITS_BASE_IDX 0 +#define regGCEA_CE_ERR_STATUS_LO 0x071b +#define regGCEA_CE_ERR_STATUS_LO_BASE_IDX 0 +#define regGCEA_CE_ERR_STATUS_HI 0x071d +#define regGCEA_CE_ERR_STATUS_HI_BASE_IDX 0 #define regGCEA_SDP_ENABLE 0x071f #define regGCEA_SDP_ENABLE_BASE_IDX 0 @@ -1389,6 +1469,14 @@ #define regATC_L2_CNTL4_BASE_IDX 0 #define regATC_L2_MM_GROUP_RT_CLASSES 0x0816 #define regATC_L2_MM_GROUP_RT_CLASSES_BASE_IDX 0 +#define regATC_L2_UE_ERR_STATUS_LO 0x081a +#define regATC_L2_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regATC_L2_UE_ERR_STATUS_HI 0x081b +#define regATC_L2_UE_ERR_STATUS_HI_BASE_IDX 0 +#define regATC_L2_CE_ERR_STATUS_LO 0x081c +#define regATC_L2_CE_ERR_STATUS_LO_BASE_IDX 0 +#define regATC_L2_CE_ERR_STATUS_HI 0x081d +#define regATC_L2_CE_ERR_STATUS_HI_BASE_IDX 0 // addressBlock: xcd0_gc_utcl2_vml2pfdec @@ -1475,6 +1563,30 @@ #define regUTCL2_EDC_MODE_BASE_IDX 0 #define regUTCL2_EDC_CONFIG 0x084c #define regUTCL2_EDC_CONFIG_BASE_IDX 0 +#define regVML2_UE_ERR_STATUS_LO 0x084d +#define regVML2_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regVML2_WALKER_UE_ERR_STATUS_LO 0x084e +#define regVML2_WALKER_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regUTCL2_UE_ERR_STATUS_LO 0x084f +#define regUTCL2_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regVML2_UE_ERR_STATUS_HI 0x0850 +#define regVML2_UE_ERR_STATUS_HI_BASE_IDX 0 +#define regVML2_WALKER_UE_ERR_STATUS_HI 0x0851 +#define regVML2_WALKER_UE_ERR_STATUS_HI_BASE_IDX 0 +#define regUTCL2_UE_ERR_STATUS_HI 0x0852 +#define regUTCL2_UE_ERR_STATUS_HI_BASE_IDX 0 +#define regVML2_CE_ERR_STATUS_LO 0x0853 +#define regVML2_CE_ERR_STATUS_LO_BASE_IDX 0 +#define regVML2_WALKER_CE_ERR_STATUS_LO 0x0854 +#define regVML2_WALKER_CE_ERR_STATUS_LO_BASE_IDX 0 +#define regUTCL2_CE_ERR_STATUS_LO 0x0855 +#define regUTCL2_CE_ERR_STATUS_LO_BASE_IDX 0 +#define regVML2_CE_ERR_STATUS_HI 0x0856 +#define regVML2_CE_ERR_STATUS_HI_BASE_IDX 0 +#define regVML2_WALKER_CE_ERR_STATUS_HI 0x0857 +#define regVML2_WALKER_CE_ERR_STATUS_HI_BASE_IDX 0 +#define regUTCL2_CE_ERR_STATUS_HI 0x0858 +#define regUTCL2_CE_ERR_STATUS_HI_BASE_IDX 0 // addressBlock: xcd0_gc_utcl2_vml2vcdec @@ -2011,6 +2123,22 @@ #define regTC_CFG_L1_VOLATILE_BASE_IDX 0 #define regTC_CFG_L2_VOLATILE 0x0b23 #define regTC_CFG_L2_VOLATILE_BASE_IDX 0 +#define regTCP_UE_EDC_HI_REG 0x0b54 +#define regTCP_UE_EDC_HI_REG_BASE_IDX 0 +#define regTCP_UE_EDC_LO_REG 0x0b55 +#define regTCP_UE_EDC_LO_REG_BASE_IDX 0 +#define regTCP_CE_EDC_HI_REG 0x0b56 +#define regTCP_CE_EDC_HI_REG_BASE_IDX 0 +#define regTCP_CE_EDC_LO_REG 0x0b57 +#define regTCP_CE_EDC_LO_REG_BASE_IDX 0 +#define regTCI_UE_EDC_HI_REG 0x0b58 +#define regTCI_UE_EDC_HI_REG_BASE_IDX 0 +#define regTCI_UE_EDC_LO_REG 0x0b59 +#define regTCI_UE_EDC_LO_REG_BASE_IDX 0 +#define regTCI_CE_EDC_HI_REG 0x0b5a +#define regTCI_CE_EDC_HI_REG_BASE_IDX 0 +#define regTCI_CE_EDC_LO_REG 0x0b5b +#define regTCI_CE_EDC_LO_REG_BASE_IDX 0 #define regTCI_MISC 0x0b5c #define regTCI_MISC_BASE_IDX 0 #define regTCI_CNTL_3 0x0b5d @@ -2061,6 +2189,26 @@ #define regTCX_DSM_CNTL_BASE_IDX 0 #define regTCX_DSM_CNTL2 0x0bc8 #define regTCX_DSM_CNTL2_BASE_IDX 0 +#define regTCA_UE_ERR_STATUS_LO 0x0bc9 +#define regTCA_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regTCA_UE_ERR_STATUS_HI 0x0bca +#define regTCA_UE_ERR_STATUS_HI_BASE_IDX 0 +#define regTCX_UE_ERR_STATUS_LO 0x0bcb +#define regTCX_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regTCX_UE_ERR_STATUS_HI 0x0bcc +#define regTCX_UE_ERR_STATUS_HI_BASE_IDX 0 +#define regTCX_CE_ERR_STATUS_LO 0x0bcd +#define regTCX_CE_ERR_STATUS_LO_BASE_IDX 0 +#define regTCX_CE_ERR_STATUS_HI 0x0bce +#define regTCX_CE_ERR_STATUS_HI_BASE_IDX 0 +#define regTCC_UE_ERR_STATUS_LO 0x0bcf +#define regTCC_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regTCC_UE_ERR_STATUS_HI 0x0bd0 +#define regTCC_UE_ERR_STATUS_HI_BASE_IDX 0 +#define regTCC_CE_ERR_STATUS_LO 0x0bd1 +#define regTCC_CE_ERR_STATUS_LO_BASE_IDX 0 +#define regTCC_CE_ERR_STATUS_HI 0x0bd2 +#define regTCC_CE_ERR_STATUS_HI_BASE_IDX 0 // addressBlock: xcd0_gc_shdec @@ -2905,6 +3053,30 @@ #define regCP_MEC2_F32_INT_DIS_BASE_IDX 0 #define regCP_VMID_STATUS 0x10bf #define regCP_VMID_STATUS_BASE_IDX 0 +#define regCPC_UE_ERR_STATUS_LO 0x10e0 +#define regCPC_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regCPC_UE_ERR_STATUS_HI 0x10e1 +#define regCPC_UE_ERR_STATUS_HI_BASE_IDX 0 +#define regCPC_CE_ERR_STATUS_LO 0x10e2 +#define regCPC_CE_ERR_STATUS_LO_BASE_IDX 0 +#define regCPC_CE_ERR_STATUS_HI 0x10e3 +#define regCPC_CE_ERR_STATUS_HI_BASE_IDX 0 +#define regCPF_UE_ERR_STATUS_LO 0x10e4 +#define regCPF_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regCPF_UE_ERR_STATUS_HI 0x10e5 +#define regCPF_UE_ERR_STATUS_HI_BASE_IDX 0 +#define regCPF_CE_ERR_STATUS_LO 0x10e6 +#define regCPF_CE_ERR_STATUS_LO_BASE_IDX 0 +#define regCPF_CE_ERR_STATUS_HI 0x10e7 +#define regCPF_CE_ERR_STATUS_HI_BASE_IDX 0 +#define regCPG_UE_ERR_STATUS_LO 0x10e8 +#define regCPG_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regCPG_UE_ERR_STATUS_HI 0x10e9 +#define regCPG_UE_ERR_STATUS_HI_BASE_IDX 0 +#define regCPG_CE_ERR_STATUS_LO 0x10ea +#define regCPG_CE_ERR_STATUS_LO_BASE_IDX 0 +#define regCPG_CE_ERR_STATUS_HI 0x10eb +#define regCPG_CE_ERR_STATUS_HI_BASE_IDX 0 // addressBlock: xcd0_gc_cppdec2 @@ -5364,6 +5536,18 @@ #define regSPI_WAVE_LIMIT_CNTL 0x2443 #define regSPI_WAVE_LIMIT_CNTL_BASE_IDX 1 +// addressBlock: xcd0_gc_gccanedec +// base address: 0x33d00 +#define regGC_CANE_ERR_STATUS 0x2f4d +#define regGC_CANE_ERR_STATUS_BASE_IDX 1 +#define regGC_CANE_UE_ERR_STATUS_LO 0x2f4e +#define regGC_CANE_UE_ERR_STATUS_LO_BASE_IDX 1 +#define regGC_CANE_UE_ERR_STATUS_HI 0x2f4f +#define regGC_CANE_UE_ERR_STATUS_HI_BASE_IDX 1 +#define regGC_CANE_CE_ERR_STATUS_LO 0x2f50 +#define regGC_CANE_CE_ERR_STATUS_LO_BASE_IDX 1 +#define regGC_CANE_CE_ERR_STATUS_HI 0x2f51 +#define regGC_CANE_CE_ERR_STATUS_HI_BASE_IDX 1 // addressBlock: xcd0_gc_perfddec // base address: 0x34000 @@ -6583,6 +6767,10 @@ #define regRLC_CAPTURE_GPU_CLOCK_COUNT_2_BASE_IDX 1 #define regRLC_CPG_STAT_INVAL 0x4d09 #define regRLC_CPG_STAT_INVAL_BASE_IDX 1 +#define regRLC_UE_ERR_STATUS_LOW 0x4d40 +#define regRLC_UE_ERR_STATUS_LOW_BASE_IDX 1 +#define regRLC_UE_ERR_STATUS_HIGH 0x4d41 +#define regRLC_UE_ERR_STATUS_HIGH_BASE_IDX 1 #define regRLC_DSM_CNTL 0x4d42 #define regRLC_DSM_CNTL_BASE_IDX 1 #define regRLC_DSM_CNTLA 0x4d43 @@ -6591,6 +6779,10 @@ #define regRLC_DSM_CNTL2_BASE_IDX 1 #define regRLC_DSM_CNTL2A 0x4d45 #define regRLC_DSM_CNTL2A_BASE_IDX 1 +#define regRLC_CE_ERR_STATUS_LOW 0x4d49 +#define regRLC_CE_ERR_STATUS_LOW_BASE_IDX 1 +#define regRLC_CE_ERR_STATUS_HIGH 0x4d4a +#define regRLC_CE_ERR_STATUS_HIGH_BASE_IDX 1 #define regRLC_RLCV_SPARE_INT 0x4f30 #define regRLC_RLCV_SPARE_INT_BASE_IDX 1 #define regRLC_SMU_CLK_REQ 0x4f97 diff --git a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_3_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_3_sh_mask.h index 84a75b58347f7c078f0b317254a390c257cfb1df..2bd9f3f1026f9cc61dbd9dd1c2a637eb27831ee9 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_3_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/gc/gc_9_4_3_sh_mask.h @@ -4129,6 +4129,240 @@ #define SQC_DCACHE_UTCL1_STATUS__FAULT_DETECTED_MASK 0x00000001L #define SQC_DCACHE_UTCL1_STATUS__RETRY_DETECTED_MASK 0x00000002L #define SQC_DCACHE_UTCL1_STATUS__PRT_DETECTED_MASK 0x00000004L +//SQC_UE_EDC_LO +#define SQC_UE_EDC_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define SQC_UE_EDC_LO__ADDRESS_INFO_VALID_FLAG__SHIFT 0x1 +#define SQC_UE_EDC_LO__ADDRESS__SHIFT 0x2 +#define SQC_UE_EDC_LO__MEM_ID__SHIFT 0x18 +#define SQC_UE_EDC_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define SQC_UE_EDC_LO__ADDRESS_INFO_VALID_FLAG_MASK 0x00000002L +#define SQC_UE_EDC_LO__ADDRESS_MASK 0x00FFFFFCL +#define SQC_UE_EDC_LO__MEM_ID_MASK 0xFF000000L +//SQC_UE_EDC_HI +#define SQC_UE_EDC_HI__ECC__SHIFT 0x0 +#define SQC_UE_EDC_HI__PARITY__SHIFT 0x1 +#define SQC_UE_EDC_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define SQC_UE_EDC_HI__ERR_INFO__SHIFT 0x3 +#define SQC_UE_EDC_HI__UE_CNT__SHIFT 0x17 +#define SQC_UE_EDC_HI__FED_CNT__SHIFT 0x1a +#define SQC_UE_EDC_HI__ECC_MASK 0x00000001L +#define SQC_UE_EDC_HI__PARITY_MASK 0x00000002L +#define SQC_UE_EDC_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define SQC_UE_EDC_HI__ERR_INFO_MASK 0x007FFFF8L +#define SQC_UE_EDC_HI__UE_CNT_MASK 0x03800000L +#define SQC_UE_EDC_HI__FED_CNT_MASK 0x1C000000L +//SQC_CE_EDC_LO +#define SQC_CE_EDC_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define SQC_CE_EDC_LO__ADDRESS_INFO_VALID_FLAG__SHIFT 0x1 +#define SQC_CE_EDC_LO__ADDRESS__SHIFT 0x2 +#define SQC_CE_EDC_LO__MEM_ID__SHIFT 0x18 +#define SQC_CE_EDC_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define SQC_CE_EDC_LO__ADDRESS_INFO_VALID_FLAG_MASK 0x00000002L +#define SQC_CE_EDC_LO__ADDRESS_MASK 0x00FFFFFCL +#define SQC_CE_EDC_LO__MEM_ID_MASK 0xFF000000L +//SQC_CE_EDC_HI +#define SQC_CE_EDC_HI__ECC__SHIFT 0x0 +#define SQC_CE_EDC_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define SQC_CE_EDC_HI__ERR_INFO__SHIFT 0x3 +#define SQC_CE_EDC_HI__CE_CNT__SHIFT 0x17 +#define SQC_CE_EDC_HI__POSION__SHIFT 0x1a +#define SQC_CE_EDC_HI__ECC_MASK 0x00000001L +#define SQC_CE_EDC_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define SQC_CE_EDC_HI__ERR_INFO_MASK 0x007FFFF8L +#define SQC_CE_EDC_HI__CE_CNT_MASK 0x03800000L +#define SQC_CE_EDC_HI__POSION_MASK 0x04000000L +//SQ_UE_ERR_STATUS_LO +#define SQ_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define SQ_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define SQ_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define SQ_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define SQ_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define SQ_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define SQ_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define SQ_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//SQ_UE_ERR_STATUS_HI +#define SQ_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define SQ_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define SQ_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define SQ_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define SQ_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define SQ_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define SQ_UE_ERR_STATUS_HI__RESERVED__SHIFT 0x1d +#define SQ_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define SQ_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define SQ_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define SQ_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define SQ_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define SQ_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define SQ_UE_ERR_STATUS_HI__RESERVED_MASK 0xE0000000L +//SQ_CE_ERR_STATUS_LO +#define SQ_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define SQ_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define SQ_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define SQ_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define SQ_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define SQ_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define SQ_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define SQ_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//SQ_CE_ERR_STATUS_HI +#define SQ_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define SQ_CE_ERR_STATUS_HI__OTHER__SHIFT 0x1 +#define SQ_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define SQ_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define SQ_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define SQ_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define SQ_CE_ERR_STATUS_HI__RESERVED__SHIFT 0x1b +#define SQ_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define SQ_CE_ERR_STATUS_HI__OTHER_MASK 0x00000002L +#define SQ_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define SQ_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define SQ_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define SQ_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define SQ_CE_ERR_STATUS_HI__RESERVED_MASK 0xF8000000L +//LDS_UE_ERR_STATUS_LO +#define LDS_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define LDS_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define LDS_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define LDS_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define LDS_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define LDS_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define LDS_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define LDS_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//LDS_UE_ERR_STATUS_HI +#define LDS_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define LDS_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define LDS_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define LDS_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define LDS_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define LDS_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define LDS_UE_ERR_STATUS_HI__RESERVED__SHIFT 0x1d +#define LDS_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define LDS_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define LDS_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define LDS_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define LDS_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define LDS_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define LDS_UE_ERR_STATUS_HI__RESERVED_MASK 0xE0000000L +//LDS_CE_ERR_STATUS_LO +#define LDS_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define LDS_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define LDS_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define LDS_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define LDS_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define LDS_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define LDS_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define LDS_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//LDS_CE_ERR_STATUS_HI +#define LDS_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define LDS_CE_ERR_STATUS_HI__OTHER__SHIFT 0x1 +#define LDS_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define LDS_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define LDS_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define LDS_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define LDS_CE_ERR_STATUS_HI__RESERVED__SHIFT 0x1b +#define LDS_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define LDS_CE_ERR_STATUS_HI__OTHER_MASK 0x00000002L +#define LDS_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define LDS_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define LDS_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define LDS_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define LDS_CE_ERR_STATUS_HI__RESERVED_MASK 0xF8000000L +//SP0_UE_ERR_STATUS_LO +#define SP0_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define SP0_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define SP0_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define SP0_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define SP0_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define SP0_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define SP0_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define SP0_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//SP0_UE_ERR_STATUS_HI +#define SP0_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define SP0_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define SP0_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define SP0_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define SP0_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define SP0_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define SP0_UE_ERR_STATUS_HI__RESERVED__SHIFT 0x1d +#define SP0_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define SP0_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define SP0_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define SP0_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define SP0_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define SP0_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define SP0_UE_ERR_STATUS_HI__RESERVED_MASK 0xE0000000L +//SP0_CE_ERR_STATUS_LO +#define SP0_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define SP0_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define SP0_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define SP0_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define SP0_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define SP0_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define SP0_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define SP0_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//SP0_CE_ERR_STATUS_HI +#define SP0_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define SP0_CE_ERR_STATUS_HI__OTHER__SHIFT 0x1 +#define SP0_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define SP0_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define SP0_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define SP0_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define SP0_CE_ERR_STATUS_HI__RESERVED__SHIFT 0x1b +#define SP0_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define SP0_CE_ERR_STATUS_HI__OTHER_MASK 0x00000002L +#define SP0_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define SP0_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define SP0_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define SP0_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define SP0_CE_ERR_STATUS_HI__RESERVED_MASK 0xF8000000L +//SP1_UE_ERR_STATUS_LO +#define SP1_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define SP1_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define SP1_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define SP1_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define SP1_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define SP1_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define SP1_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define SP1_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//SP1_UE_ERR_STATUS_HI +#define SP1_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define SP1_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define SP1_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define SP1_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define SP1_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define SP1_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define SP1_UE_ERR_STATUS_HI__RESERVED__SHIFT 0x1d +#define SP1_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define SP1_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define SP1_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define SP1_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define SP1_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define SP1_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define SP1_UE_ERR_STATUS_HI__RESERVED_MASK 0xE0000000L +//SP1_CE_ERR_STATUS_LO +#define SP1_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define SP1_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define SP1_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define SP1_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define SP1_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define SP1_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define SP1_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define SP1_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//SP1_CE_ERR_STATUS_HI +#define SP1_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define SP1_CE_ERR_STATUS_HI__OTHER__SHIFT 0x1 +#define SP1_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define SP1_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define SP1_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define SP1_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define SP1_CE_ERR_STATUS_HI__RESERVED__SHIFT 0x1b +#define SP1_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define SP1_CE_ERR_STATUS_HI__OTHER_MASK 0x00000002L +#define SP1_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define SP1_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define SP1_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define SP1_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define SP1_CE_ERR_STATUS_HI__RESERVED_MASK 0xF8000000L // addressBlock: xcd0_gc_shsdec @@ -4235,6 +4469,54 @@ #define SPI_EDC_CNT__SPI_LIFE_CNT_SEC_COUNT_MASK 0x00030000L #define SPI_EDC_CNT__SPI_LIFE_CNT_DED_COUNT_MASK 0x000C0000L #define SPI_EDC_CNT__UNUSED_MASK 0xFFF00000L +//SPI_UE_ERR_STATUS_LO +#define SPI_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define SPI_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define SPI_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define SPI_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define SPI_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define SPI_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define SPI_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define SPI_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//SPI_UE_ERR_STATUS_HI +#define SPI_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define SPI_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define SPI_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define SPI_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define SPI_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define SPI_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define SPI_UE_ERR_STATUS_HI__RESERVED__SHIFT 0x1d +#define SPI_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define SPI_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define SPI_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define SPI_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define SPI_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define SPI_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define SPI_UE_ERR_STATUS_HI__RESERVED_MASK 0xE0000000L +//SPI_CE_ERR_STATUS_LO +#define SPI_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define SPI_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define SPI_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define SPI_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define SPI_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define SPI_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define SPI_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define SPI_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//SPI_CE_ERR_STATUS_HI +#define SPI_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define SPI_CE_ERR_STATUS_HI__OTHER__SHIFT 0x1 +#define SPI_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define SPI_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define SPI_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define SPI_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define SPI_CE_ERR_STATUS_HI__RESERVED__SHIFT 0x1b +#define SPI_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define SPI_CE_ERR_STATUS_HI__OTHER_MASK 0x00000002L +#define SPI_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define SPI_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define SPI_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define SPI_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define SPI_CE_ERR_STATUS_HI__RESERVED_MASK 0xF8000000L //SPI_DEBUG_BUSY #define SPI_DEBUG_BUSY__HS_BUSY__SHIFT 0x0 #define SPI_DEBUG_BUSY__GS_BUSY__SHIFT 0x1 @@ -4622,6 +4904,48 @@ #define TD_POWER_CNTL__MGCG_OUTPUTSTAGE_MASK 0x00000002L #define TD_POWER_CNTL__MID0_THREAD_DATA_MASK 0x00000004L #define TD_POWER_CNTL__MID2_ACCUM_DATA_MASK 0x00000008L +//TD_UE_EDC_LO +#define TD_UE_EDC_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define TD_UE_EDC_LO__ADDRESS_INFO_VALID_FLAG__SHIFT 0x1 +#define TD_UE_EDC_LO__ADDRESS__SHIFT 0x2 +#define TD_UE_EDC_LO__MEM_ID__SHIFT 0x18 +#define TD_UE_EDC_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define TD_UE_EDC_LO__ADDRESS_INFO_VALID_FLAG_MASK 0x00000002L +#define TD_UE_EDC_LO__ADDRESS_MASK 0x00FFFFFCL +#define TD_UE_EDC_LO__MEM_ID_MASK 0xFF000000L +//TD_UE_EDC_HI +#define TD_UE_EDC_HI__ECC__SHIFT 0x0 +#define TD_UE_EDC_HI__PARITY__SHIFT 0x1 +#define TD_UE_EDC_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define TD_UE_EDC_HI__ERR_INFO__SHIFT 0x3 +#define TD_UE_EDC_HI__UE_CNT__SHIFT 0x17 +#define TD_UE_EDC_HI__FED_CNT__SHIFT 0x1a +#define TD_UE_EDC_HI__ECC_MASK 0x00000001L +#define TD_UE_EDC_HI__PARITY_MASK 0x00000002L +#define TD_UE_EDC_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define TD_UE_EDC_HI__ERR_INFO_MASK 0x007FFFF8L +#define TD_UE_EDC_HI__UE_CNT_MASK 0x03800000L +#define TD_UE_EDC_HI__FED_CNT_MASK 0x1C000000L +//TD_CE_EDC_LO +#define TD_CE_EDC_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define TD_CE_EDC_LO__ADDRESS_INFO_VALID_FLAG__SHIFT 0x1 +#define TD_CE_EDC_LO__ADDRESS__SHIFT 0x2 +#define TD_CE_EDC_LO__MEM_ID__SHIFT 0x18 +#define TD_CE_EDC_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define TD_CE_EDC_LO__ADDRESS_INFO_VALID_FLAG_MASK 0x00000002L +#define TD_CE_EDC_LO__ADDRESS_MASK 0x00FFFFFCL +#define TD_CE_EDC_LO__MEM_ID_MASK 0xFF000000L +//TD_CE_EDC_HI +#define TD_CE_EDC_HI__ECC__SHIFT 0x0 +#define TD_CE_EDC_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define TD_CE_EDC_HI__ERR_INFO__SHIFT 0x3 +#define TD_CE_EDC_HI__CE_CNT__SHIFT 0x17 +#define TD_CE_EDC_HI__POISON__SHIFT 0x1a +#define TD_CE_EDC_HI__ECC_MASK 0x00000001L +#define TD_CE_EDC_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define TD_CE_EDC_HI__ERR_INFO_MASK 0x007FFFF8L +#define TD_CE_EDC_HI__CE_CNT_MASK 0x03800000L +#define TD_CE_EDC_HI__POISON_MASK 0x04000000L //TD_DSM_CNTL #define TD_DSM_CNTL__TD_SS_FIFO_LO_DSM_IRRITATOR_DATA__SHIFT 0x0 #define TD_DSM_CNTL__TD_SS_FIFO_LO_ENABLE_SINGLE_WRITE__SHIFT 0x2 @@ -4771,6 +5095,48 @@ #define TA_DSM_CNTL2__TA_FS_AFIFO_HI_ENABLE_ERROR_INJECT_MASK 0x000C0000L #define TA_DSM_CNTL2__TA_FS_AFIFO_HI_SELECT_INJECT_DELAY_MASK 0x00100000L #define TA_DSM_CNTL2__TA_INJECT_DELAY_MASK 0xFC000000L +//TA_UE_EDC_LO +#define TA_UE_EDC_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define TA_UE_EDC_LO__ADDRESS_INFO_VALID_FLAG__SHIFT 0x1 +#define TA_UE_EDC_LO__ADDRESS__SHIFT 0x2 +#define TA_UE_EDC_LO__MEM_ID__SHIFT 0x18 +#define TA_UE_EDC_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define TA_UE_EDC_LO__ADDRESS_INFO_VALID_FLAG_MASK 0x00000002L +#define TA_UE_EDC_LO__ADDRESS_MASK 0x00FFFFFCL +#define TA_UE_EDC_LO__MEM_ID_MASK 0xFF000000L +//TA_UE_EDC_HI +#define TA_UE_EDC_HI__ECC__SHIFT 0x0 +#define TA_UE_EDC_HI__PARITY__SHIFT 0x1 +#define TA_UE_EDC_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define TA_UE_EDC_HI__ERR_INFO__SHIFT 0x3 +#define TA_UE_EDC_HI__UE_CNT__SHIFT 0x17 +#define TA_UE_EDC_HI__FED_CNT__SHIFT 0x1a +#define TA_UE_EDC_HI__ECC_MASK 0x00000001L +#define TA_UE_EDC_HI__PARITY_MASK 0x00000002L +#define TA_UE_EDC_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define TA_UE_EDC_HI__ERR_INFO_MASK 0x007FFFF8L +#define TA_UE_EDC_HI__UE_CNT_MASK 0x03800000L +#define TA_UE_EDC_HI__FED_CNT_MASK 0x1C000000L +//TA_CE_EDC_LO +#define TA_CE_EDC_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define TA_CE_EDC_LO__ADDRESS_INFO_VALID_FLAG__SHIFT 0x1 +#define TA_CE_EDC_LO__ADDRESS__SHIFT 0x2 +#define TA_CE_EDC_LO__MEM_ID__SHIFT 0x18 +#define TA_CE_EDC_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define TA_CE_EDC_LO__ADDRESS_INFO_VALID_FLAG_MASK 0x00000002L +#define TA_CE_EDC_LO__ADDRESS_MASK 0x00FFFFFCL +#define TA_CE_EDC_LO__MEM_ID_MASK 0xFF000000L +//TA_CE_EDC_HI +#define TA_CE_EDC_HI__ECC__SHIFT 0x0 +#define TA_CE_EDC_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define TA_CE_EDC_HI__ERR_INFO__SHIFT 0x3 +#define TA_CE_EDC_HI__CE_CNT__SHIFT 0x17 +#define TA_CE_EDC_HI__POISON__SHIFT 0x1a +#define TA_CE_EDC_HI__ECC_MASK 0x00000001L +#define TA_CE_EDC_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define TA_CE_EDC_HI__ERR_INFO_MASK 0x007FFFF8L +#define TA_CE_EDC_HI__CE_CNT_MASK 0x03800000L +#define TA_CE_EDC_HI__POISON_MASK 0x04000000L // addressBlock: xcd0_gc_gdsdec @@ -5015,6 +5381,54 @@ #define GDS_WD_GDS_CSB__UNUSED__SHIFT 0xd #define GDS_WD_GDS_CSB__COUNTER_MASK 0x00001FFFL #define GDS_WD_GDS_CSB__UNUSED_MASK 0xFFFFE000L +//GDS_UE_ERR_STATUS_LO +#define GDS_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define GDS_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define GDS_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define GDS_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define GDS_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define GDS_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define GDS_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define GDS_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//GDS_UE_ERR_STATUS_HI +#define GDS_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define GDS_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define GDS_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define GDS_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define GDS_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define GDS_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define GDS_UE_ERR_STATUS_HI__RESERVED__SHIFT 0x1d +#define GDS_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define GDS_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define GDS_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define GDS_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define GDS_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define GDS_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define GDS_UE_ERR_STATUS_HI__RESERVED_MASK 0xE0000000L +//GDS_CE_ERR_STATUS_LO +#define GDS_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define GDS_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define GDS_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define GDS_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define GDS_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define GDS_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define GDS_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define GDS_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//GDS_CE_ERR_STATUS_HI +#define GDS_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define GDS_CE_ERR_STATUS_HI__OTHER__SHIFT 0x1 +#define GDS_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define GDS_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define GDS_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define GDS_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define GDS_CE_ERR_STATUS_HI__RESERVED__SHIFT 0x1b +#define GDS_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define GDS_CE_ERR_STATUS_HI__OTHER_MASK 0x00000002L +#define GDS_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define GDS_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define GDS_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define GDS_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define GDS_CE_ERR_STATUS_HI__RESERVED_MASK 0xF8000000L // addressBlock: xcd0_gc_rbdec @@ -7370,6 +7784,30 @@ #define GCEA_MAM_CTRL2__ARAM_FLUSH_NOALLOC_MASK 0x00000040L #define GCEA_MAM_CTRL2__RESERVED_FIELD_MASK 0x00FFFF80L #define GCEA_MAM_CTRL2__ADDR_HI_MASK 0xFF000000L +//GCEA_UE_ERR_STATUS_LO +#define GCEA_UE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define GCEA_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define GCEA_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define GCEA_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define GCEA_UE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define GCEA_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define GCEA_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define GCEA_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//GCEA_UE_ERR_STATUS_HI +#define GCEA_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define GCEA_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define GCEA_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define GCEA_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define GCEA_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define GCEA_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define GCEA_UE_ERR_STATUS_HI__RESERVED_FIELD__SHIFT 0x1d +#define GCEA_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define GCEA_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define GCEA_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define GCEA_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define GCEA_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define GCEA_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define GCEA_UE_ERR_STATUS_HI__RESERVED_FIELD_MASK 0xE0000000L //GCEA_DSM_CNTL #define GCEA_DSM_CNTL__DRAMRD_CMDMEM_DSM_IRRITATOR_DATA__SHIFT 0x0 #define GCEA_DSM_CNTL__DRAMRD_CMDMEM_ENABLE_SINGLE_WRITE__SHIFT 0x2 @@ -7745,6 +8183,30 @@ #define GCEA_SDP_BACKDOOR_MISCCREDITS__WRRSP_CREDITS_RELEASED_MASK 0x0000FF00L #define GCEA_SDP_BACKDOOR_MISCCREDITS__PRB_REQ_CREDITS_RELEASED_MASK 0x007F0000L #define GCEA_SDP_BACKDOOR_MISCCREDITS__PRB_RSP_CREDITS_RECEIVED_MASK 0x3F800000L +//GCEA_CE_ERR_STATUS_LO +#define GCEA_CE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define GCEA_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define GCEA_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define GCEA_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define GCEA_CE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define GCEA_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define GCEA_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define GCEA_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//GCEA_CE_ERR_STATUS_HI +#define GCEA_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define GCEA_CE_ERR_STATUS_HI__RESERVED_FIELD0__SHIFT 0x1 +#define GCEA_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define GCEA_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define GCEA_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define GCEA_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define GCEA_CE_ERR_STATUS_HI__RESERVED_FIELD1__SHIFT 0x1b +#define GCEA_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define GCEA_CE_ERR_STATUS_HI__RESERVED_FIELD0_MASK 0x00000002L +#define GCEA_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define GCEA_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define GCEA_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define GCEA_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define GCEA_CE_ERR_STATUS_HI__RESERVED_FIELD1_MASK 0xF8000000L //GCEA_SDP_ENABLE #define GCEA_SDP_ENABLE__ENABLE__SHIFT 0x0 #define GCEA_SDP_ENABLE__ENABLE_MASK 0x00000001L @@ -8440,6 +8902,54 @@ //ATC_L2_MM_GROUP_RT_CLASSES #define ATC_L2_MM_GROUP_RT_CLASSES__GROUP_RT_CLASS__SHIFT 0x0 #define ATC_L2_MM_GROUP_RT_CLASSES__GROUP_RT_CLASS_MASK 0xFFFFFFFFL +//ATC_L2_UE_ERR_STATUS_LO +#define ATC_L2_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define ATC_L2_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define ATC_L2_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define ATC_L2_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define ATC_L2_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define ATC_L2_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define ATC_L2_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define ATC_L2_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//ATC_L2_UE_ERR_STATUS_HI +#define ATC_L2_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define ATC_L2_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define ATC_L2_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define ATC_L2_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define ATC_L2_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define ATC_L2_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define ATC_L2_UE_ERR_STATUS_HI__RESERVED__SHIFT 0x1d +#define ATC_L2_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define ATC_L2_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define ATC_L2_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define ATC_L2_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define ATC_L2_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define ATC_L2_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define ATC_L2_UE_ERR_STATUS_HI__RESERVED_MASK 0x60000000L +//ATC_L2_CE_ERR_STATUS_LO +#define ATC_L2_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define ATC_L2_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define ATC_L2_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define ATC_L2_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define ATC_L2_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define ATC_L2_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define ATC_L2_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define ATC_L2_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//ATC_L2_CE_ERR_STATUS_HI +#define ATC_L2_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define ATC_L2_CE_ERR_STATUS_HI__OTHER__SHIFT 0x1 +#define ATC_L2_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define ATC_L2_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define ATC_L2_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define ATC_L2_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define ATC_L2_CE_ERR_STATUS_HI__RESERVED__SHIFT 0x1b +#define ATC_L2_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define ATC_L2_CE_ERR_STATUS_HI__OTHER_MASK 0x00000002L +#define ATC_L2_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define ATC_L2_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define ATC_L2_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define ATC_L2_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define ATC_L2_CE_ERR_STATUS_HI__RESERVED_MASK 0xF8000000L // addressBlock: xcd0_gc_utcl2_vml2pfdec @@ -8888,6 +9398,150 @@ #define UTCL2_EDC_CONFIG__DIS_EDC__SHIFT 0x1 #define UTCL2_EDC_CONFIG__WRITE_DIS_MASK 0x00000001L #define UTCL2_EDC_CONFIG__DIS_EDC_MASK 0x00000002L +//VML2_UE_ERR_STATUS_LO +#define VML2_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define VML2_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define VML2_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define VML2_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define VML2_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define VML2_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define VML2_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define VML2_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//VML2_WALKER_UE_ERR_STATUS_LO +#define VML2_WALKER_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define VML2_WALKER_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define VML2_WALKER_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define VML2_WALKER_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define VML2_WALKER_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define VML2_WALKER_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define VML2_WALKER_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define VML2_WALKER_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//UTCL2_UE_ERR_STATUS_LO +#define UTCL2_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define UTCL2_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define UTCL2_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define UTCL2_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define UTCL2_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define UTCL2_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define UTCL2_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define UTCL2_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//VML2_UE_ERR_STATUS_HI +#define VML2_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define VML2_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define VML2_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define VML2_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define VML2_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define VML2_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define VML2_UE_ERR_STATUS_HI__RESERVED__SHIFT 0x1d +#define VML2_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define VML2_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define VML2_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define VML2_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define VML2_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define VML2_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define VML2_UE_ERR_STATUS_HI__RESERVED_MASK 0xE0000000L +//VML2_WALKER_UE_ERR_STATUS_HI +#define VML2_WALKER_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define VML2_WALKER_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define VML2_WALKER_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define VML2_WALKER_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define VML2_WALKER_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define VML2_WALKER_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define VML2_WALKER_UE_ERR_STATUS_HI__RESERVED__SHIFT 0x1d +#define VML2_WALKER_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define VML2_WALKER_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define VML2_WALKER_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define VML2_WALKER_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define VML2_WALKER_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define VML2_WALKER_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define VML2_WALKER_UE_ERR_STATUS_HI__RESERVED_MASK 0xE0000000L +//UTCL2_UE_ERR_STATUS_HI +#define UTCL2_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define UTCL2_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define UTCL2_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define UTCL2_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define UTCL2_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define UTCL2_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define UTCL2_UE_ERR_STATUS_HI__RESERVED__SHIFT 0x1d +#define UTCL2_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define UTCL2_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define UTCL2_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define UTCL2_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define UTCL2_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define UTCL2_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define UTCL2_UE_ERR_STATUS_HI__RESERVED_MASK 0xE0000000L +//VML2_CE_ERR_STATUS_LO +#define VML2_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define VML2_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define VML2_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define VML2_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define VML2_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define VML2_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define VML2_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define VML2_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//VML2_WALKER_CE_ERR_STATUS_LO +#define VML2_WALKER_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define VML2_WALKER_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define VML2_WALKER_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define VML2_WALKER_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define VML2_WALKER_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define VML2_WALKER_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define VML2_WALKER_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define VML2_WALKER_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//UTCL2_CE_ERR_STATUS_LO +#define UTCL2_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define UTCL2_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define UTCL2_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define UTCL2_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define UTCL2_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define UTCL2_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define UTCL2_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define UTCL2_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//VML2_CE_ERR_STATUS_HI +#define VML2_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define VML2_CE_ERR_STATUS_HI__OTHER__SHIFT 0x1 +#define VML2_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define VML2_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define VML2_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define VML2_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define VML2_CE_ERR_STATUS_HI__RESERVED__SHIFT 0x1b +#define VML2_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define VML2_CE_ERR_STATUS_HI__OTHER_MASK 0x00000002L +#define VML2_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define VML2_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define VML2_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define VML2_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define VML2_CE_ERR_STATUS_HI__RESERVED_MASK 0xF8000000L +//VML2_WALKER_CE_ERR_STATUS_HI +#define VML2_WALKER_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define VML2_WALKER_CE_ERR_STATUS_HI__OTHER__SHIFT 0x1 +#define VML2_WALKER_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define VML2_WALKER_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define VML2_WALKER_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define VML2_WALKER_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define VML2_WALKER_CE_ERR_STATUS_HI__RESERVED__SHIFT 0x1b +#define VML2_WALKER_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define VML2_WALKER_CE_ERR_STATUS_HI__OTHER_MASK 0x00000002L +#define VML2_WALKER_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define VML2_WALKER_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define VML2_WALKER_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define VML2_WALKER_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define VML2_WALKER_CE_ERR_STATUS_HI__RESERVED_MASK 0xF8000000L +//UTCL2_CE_ERR_STATUS_HI +#define UTCL2_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define UTCL2_CE_ERR_STATUS_HI__OTHER__SHIFT 0x1 +#define UTCL2_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define UTCL2_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define UTCL2_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define UTCL2_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define UTCL2_CE_ERR_STATUS_HI__RESERVED__SHIFT 0x1b +#define UTCL2_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define UTCL2_CE_ERR_STATUS_HI__OTHER_MASK 0x00000002L +#define UTCL2_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define UTCL2_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define UTCL2_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define UTCL2_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define UTCL2_CE_ERR_STATUS_HI__RESERVED_MASK 0xF8000000L // addressBlock: xcd0_gc_utcl2_vml2vcdec @@ -11139,6 +11793,98 @@ //TC_CFG_L2_VOLATILE #define TC_CFG_L2_VOLATILE__VOL__SHIFT 0x0 #define TC_CFG_L2_VOLATILE__VOL_MASK 0x0000000FL +//TCP_UE_EDC_HI_REG +#define TCP_UE_EDC_HI_REG__ECC__SHIFT 0x0 +#define TCP_UE_EDC_HI_REG__PARITY__SHIFT 0x1 +#define TCP_UE_EDC_HI_REG__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define TCP_UE_EDC_HI_REG__ERR_INFO__SHIFT 0x3 +#define TCP_UE_EDC_HI_REG__UE_CNT__SHIFT 0x17 +#define TCP_UE_EDC_HI_REG__FED_CNT__SHIFT 0x1a +#define TCP_UE_EDC_HI_REG__RESERVED__SHIFT 0x1d +#define TCP_UE_EDC_HI_REG__ECC_MASK 0x00000001L +#define TCP_UE_EDC_HI_REG__PARITY_MASK 0x00000002L +#define TCP_UE_EDC_HI_REG__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define TCP_UE_EDC_HI_REG__ERR_INFO_MASK 0x007FFFF8L +#define TCP_UE_EDC_HI_REG__UE_CNT_MASK 0x03800000L +#define TCP_UE_EDC_HI_REG__FED_CNT_MASK 0x1C000000L +#define TCP_UE_EDC_HI_REG__RESERVED_MASK 0xE0000000L +//TCP_UE_EDC_LO_REG +#define TCP_UE_EDC_LO_REG__STATUS_VALID_FLAG__SHIFT 0x0 +#define TCP_UE_EDC_LO_REG__ADDRESS_INFO_VALID_FLAG__SHIFT 0x1 +#define TCP_UE_EDC_LO_REG__ADDRESS__SHIFT 0x2 +#define TCP_UE_EDC_LO_REG__MEM_ID__SHIFT 0x18 +#define TCP_UE_EDC_LO_REG__STATUS_VALID_FLAG_MASK 0x00000001L +#define TCP_UE_EDC_LO_REG__ADDRESS_INFO_VALID_FLAG_MASK 0x00000002L +#define TCP_UE_EDC_LO_REG__ADDRESS_MASK 0x00FFFFFCL +#define TCP_UE_EDC_LO_REG__MEM_ID_MASK 0xFF000000L +//TCP_CE_EDC_HI_REG +#define TCP_CE_EDC_HI_REG__ECC__SHIFT 0x0 +#define TCP_CE_EDC_HI_REG__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define TCP_CE_EDC_HI_REG__ERR_INFO__SHIFT 0x3 +#define TCP_CE_EDC_HI_REG__CE_CNT__SHIFT 0x17 +#define TCP_CE_EDC_HI_REG__POISON__SHIFT 0x1a +#define TCP_CE_EDC_HI_REG__RESERVED__SHIFT 0x1b +#define TCP_CE_EDC_HI_REG__ECC_MASK 0x00000001L +#define TCP_CE_EDC_HI_REG__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define TCP_CE_EDC_HI_REG__ERR_INFO_MASK 0x007FFFF8L +#define TCP_CE_EDC_HI_REG__CE_CNT_MASK 0x03800000L +#define TCP_CE_EDC_HI_REG__POISON_MASK 0x04000000L +#define TCP_CE_EDC_HI_REG__RESERVED_MASK 0xF8000000L +//TCP_CE_EDC_LO_REG +#define TCP_CE_EDC_LO_REG__STATUS_VALID_FLAG__SHIFT 0x0 +#define TCP_CE_EDC_LO_REG__ADDRESS_INFO_VALID_FLAG__SHIFT 0x1 +#define TCP_CE_EDC_LO_REG__ADDRESS__SHIFT 0x2 +#define TCP_CE_EDC_LO_REG__MEM_ID__SHIFT 0x18 +#define TCP_CE_EDC_LO_REG__STATUS_VALID_FLAG_MASK 0x00000001L +#define TCP_CE_EDC_LO_REG__ADDRESS_INFO_VALID_FLAG_MASK 0x00000002L +#define TCP_CE_EDC_LO_REG__ADDRESS_MASK 0x00FFFFFCL +#define TCP_CE_EDC_LO_REG__MEM_ID_MASK 0xFF000000L +//TCI_UE_EDC_HI_REG +#define TCI_UE_EDC_HI_REG__ECC__SHIFT 0x0 +#define TCI_UE_EDC_HI_REG__PARITY__SHIFT 0x1 +#define TCI_UE_EDC_HI_REG__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define TCI_UE_EDC_HI_REG__ERR_INFO__SHIFT 0x3 +#define TCI_UE_EDC_HI_REG__UE_CNT__SHIFT 0x17 +#define TCI_UE_EDC_HI_REG__FED_CNT__SHIFT 0x1a +#define TCI_UE_EDC_HI_REG__RESERVED__SHIFT 0x1d +#define TCI_UE_EDC_HI_REG__ECC_MASK 0x00000001L +#define TCI_UE_EDC_HI_REG__PARITY_MASK 0x00000002L +#define TCI_UE_EDC_HI_REG__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define TCI_UE_EDC_HI_REG__ERR_INFO_MASK 0x007FFFF8L +#define TCI_UE_EDC_HI_REG__UE_CNT_MASK 0x03800000L +#define TCI_UE_EDC_HI_REG__FED_CNT_MASK 0x1C000000L +#define TCI_UE_EDC_HI_REG__RESERVED_MASK 0xE0000000L +//TCI_UE_EDC_LO_REG +#define TCI_UE_EDC_LO_REG__STATUS_VALID_FLAG__SHIFT 0x0 +#define TCI_UE_EDC_LO_REG__ADDRESS_INFO_VALID_FLAG__SHIFT 0x1 +#define TCI_UE_EDC_LO_REG__ADDRESS__SHIFT 0x2 +#define TCI_UE_EDC_LO_REG__MEM_ID__SHIFT 0x18 +#define TCI_UE_EDC_LO_REG__STATUS_VALID_FLAG_MASK 0x00000001L +#define TCI_UE_EDC_LO_REG__ADDRESS_INFO_VALID_FLAG_MASK 0x00000002L +#define TCI_UE_EDC_LO_REG__ADDRESS_MASK 0x00FFFFFCL +#define TCI_UE_EDC_LO_REG__MEM_ID_MASK 0xFF000000L +//TCI_CE_EDC_HI_REG +#define TCI_CE_EDC_HI_REG__ECC__SHIFT 0x0 +#define TCI_CE_EDC_HI_REG__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define TCI_CE_EDC_HI_REG__ERR_INFO__SHIFT 0x3 +#define TCI_CE_EDC_HI_REG__CE_CNT__SHIFT 0x17 +#define TCI_CE_EDC_HI_REG__POISON__SHIFT 0x1a +#define TCI_CE_EDC_HI_REG__RESERVED__SHIFT 0x1b +#define TCI_CE_EDC_HI_REG__ECC_MASK 0x00000001L +#define TCI_CE_EDC_HI_REG__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define TCI_CE_EDC_HI_REG__ERR_INFO_MASK 0x007FFFF8L +#define TCI_CE_EDC_HI_REG__CE_CNT_MASK 0x03800000L +#define TCI_CE_EDC_HI_REG__POISON_MASK 0x04000000L +#define TCI_CE_EDC_HI_REG__RESERVED_MASK 0xF8000000L +//TCI_CE_EDC_LO_REG +#define TCI_CE_EDC_LO_REG__STATUS_VALID_FLAG__SHIFT 0x0 +#define TCI_CE_EDC_LO_REG__ADDRESS_INFO_VALID_FLAG__SHIFT 0x1 +#define TCI_CE_EDC_LO_REG__ADDRESS__SHIFT 0x2 +#define TCI_CE_EDC_LO_REG__MEM_ID__SHIFT 0x18 +#define TCI_CE_EDC_LO_REG__STATUS_VALID_FLAG_MASK 0x00000001L +#define TCI_CE_EDC_LO_REG__ADDRESS_INFO_VALID_FLAG_MASK 0x00000002L +#define TCI_CE_EDC_LO_REG__ADDRESS_MASK 0x00FFFFFCL +#define TCI_CE_EDC_LO_REG__MEM_ID_MASK 0xFF000000L //TCI_MISC #define TCI_MISC__FGCG_REPEATER_DISABLE__SHIFT 0x0 #define TCI_MISC__LEGACY_MGCG_DISABLE__SHIFT 0x1 @@ -11560,6 +12306,112 @@ #define TCX_DSM_CNTL2__SED_ENABLE_ERROR_INJECT_MASK 0x00000003L #define TCX_DSM_CNTL2__SED_SELECT_INJECT_DELAY_MASK 0x00000004L #define TCX_DSM_CNTL2__INJECT_DELAY_MASK 0xFC000000L +//TCA_UE_ERR_STATUS_LO +#define TCA_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define TCA_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define TCA_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define TCA_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define TCA_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define TCA_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define TCA_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define TCA_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//TCA_UE_ERR_STATUS_HI +#define TCA_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define TCA_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define TCA_UE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG__SHIFT 0x2 +#define TCA_UE_ERR_STATUS_HI__ERROR_INFO__SHIFT 0x3 +#define TCA_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define TCA_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define TCA_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define TCA_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define TCA_UE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG_MASK 0x00000004L +#define TCA_UE_ERR_STATUS_HI__ERROR_INFO_MASK 0x007FFFF8L +#define TCA_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define TCA_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +//TCX_UE_ERR_STATUS_LO +#define TCX_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define TCX_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define TCX_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define TCX_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define TCX_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define TCX_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define TCX_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define TCX_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//TCX_UE_ERR_STATUS_HI +#define TCX_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define TCX_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define TCX_UE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG__SHIFT 0x2 +#define TCX_UE_ERR_STATUS_HI__ERROR_INFO__SHIFT 0x3 +#define TCX_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define TCX_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define TCX_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define TCX_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define TCX_UE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG_MASK 0x00000004L +#define TCX_UE_ERR_STATUS_HI__ERROR_INFO_MASK 0x007FFFF8L +#define TCX_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define TCX_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +//TCX_CE_ERR_STATUS_LO +#define TCX_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define TCX_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define TCX_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define TCX_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define TCX_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define TCX_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define TCX_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define TCX_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//TCX_CE_ERR_STATUS_HI +#define TCX_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define TCX_CE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG__SHIFT 0x2 +#define TCX_CE_ERR_STATUS_HI__ERROR_INFO__SHIFT 0x3 +#define TCX_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define TCX_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define TCX_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define TCX_CE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG_MASK 0x00000004L +#define TCX_CE_ERR_STATUS_HI__ERROR_INFO_MASK 0x007FFFF8L +#define TCX_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define TCX_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +//TCC_UE_ERR_STATUS_LO +#define TCC_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define TCC_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define TCC_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define TCC_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define TCC_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define TCC_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define TCC_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define TCC_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//TCC_UE_ERR_STATUS_HI +#define TCC_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define TCC_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define TCC_UE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG__SHIFT 0x2 +#define TCC_UE_ERR_STATUS_HI__ERROR_INFO__SHIFT 0x3 +#define TCC_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define TCC_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define TCC_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define TCC_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define TCC_UE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG_MASK 0x00000004L +#define TCC_UE_ERR_STATUS_HI__ERROR_INFO_MASK 0x007FFFF8L +#define TCC_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define TCC_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +//TCC_CE_ERR_STATUS_LO +#define TCC_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define TCC_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define TCC_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define TCC_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define TCC_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define TCC_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define TCC_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define TCC_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//TCC_CE_ERR_STATUS_HI +#define TCC_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define TCC_CE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG__SHIFT 0x2 +#define TCC_CE_ERR_STATUS_HI__ERROR_INFO__SHIFT 0x3 +#define TCC_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define TCC_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define TCC_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define TCC_CE_ERR_STATUS_HI__ERROR_INFO_VALID_FLAG_MASK 0x00000004L +#define TCC_CE_ERR_STATUS_HI__ERROR_INFO_MASK 0x007FFFF8L +#define TCC_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define TCC_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L // addressBlock: xcd0_gc_shdec @@ -14384,6 +15236,150 @@ #define CP_VMID_STATUS__PREEMPT_CE_STATUS__SHIFT 0x10 #define CP_VMID_STATUS__PREEMPT_DE_STATUS_MASK 0x0000FFFFL #define CP_VMID_STATUS__PREEMPT_CE_STATUS_MASK 0xFFFF0000L +//CPC_UE_ERR_STATUS_LO +#define CPC_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define CPC_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define CPC_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define CPC_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define CPC_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define CPC_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define CPC_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define CPC_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//CPC_UE_ERR_STATUS_HI +#define CPC_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define CPC_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define CPC_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define CPC_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define CPC_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define CPC_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define CPC_UE_ERR_STATUS_HI__RESERVED__SHIFT 0x1d +#define CPC_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define CPC_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define CPC_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define CPC_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define CPC_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define CPC_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define CPC_UE_ERR_STATUS_HI__RESERVED_MASK 0xE0000000L +//CPC_CE_ERR_STATUS_LO +#define CPC_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define CPC_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define CPC_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define CPC_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define CPC_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define CPC_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define CPC_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define CPC_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//CPC_CE_ERR_STATUS_HI +#define CPC_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define CPC_CE_ERR_STATUS_HI__OTHER__SHIFT 0x1 +#define CPC_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define CPC_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define CPC_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define CPC_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define CPC_CE_ERR_STATUS_HI__RESERVED__SHIFT 0x1b +#define CPC_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define CPC_CE_ERR_STATUS_HI__OTHER_MASK 0x00000002L +#define CPC_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define CPC_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define CPC_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define CPC_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define CPC_CE_ERR_STATUS_HI__RESERVED_MASK 0xF8000000L +//CPF_UE_ERR_STATUS_LO +#define CPF_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define CPF_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define CPF_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define CPF_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define CPF_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define CPF_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define CPF_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define CPF_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//CPF_UE_ERR_STATUS_HI +#define CPF_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define CPF_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define CPF_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define CPF_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define CPF_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define CPF_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define CPF_UE_ERR_STATUS_HI__RESERVED__SHIFT 0x1d +#define CPF_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define CPF_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define CPF_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define CPF_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define CPF_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define CPF_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define CPF_UE_ERR_STATUS_HI__RESERVED_MASK 0xE0000000L +//CPF_CE_ERR_STATUS_LO +#define CPF_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define CPF_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define CPF_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define CPF_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define CPF_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define CPF_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define CPF_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define CPF_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//CPF_CE_ERR_STATUS_HI +#define CPF_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define CPF_CE_ERR_STATUS_HI__OTHER__SHIFT 0x1 +#define CPF_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define CPF_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define CPF_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define CPF_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define CPF_CE_ERR_STATUS_HI__RESERVED__SHIFT 0x1b +#define CPF_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define CPF_CE_ERR_STATUS_HI__OTHER_MASK 0x00000002L +#define CPF_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define CPF_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define CPF_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define CPF_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define CPF_CE_ERR_STATUS_HI__RESERVED_MASK 0xF8000000L +//CPG_UE_ERR_STATUS_LO +#define CPG_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define CPG_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define CPG_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define CPG_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define CPG_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define CPG_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define CPG_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define CPG_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//CPG_UE_ERR_STATUS_HI +#define CPG_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define CPG_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define CPG_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define CPG_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define CPG_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define CPG_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define CPG_UE_ERR_STATUS_HI__RESERVED__SHIFT 0x1d +#define CPG_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define CPG_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define CPG_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define CPG_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define CPG_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define CPG_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define CPG_UE_ERR_STATUS_HI__RESERVED_MASK 0xE0000000L +//CPG_CE_ERR_STATUS_LO +#define CPG_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define CPG_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define CPG_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define CPG_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define CPG_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define CPG_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define CPG_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define CPG_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//CPG_CE_ERR_STATUS_HI +#define CPG_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define CPG_CE_ERR_STATUS_HI__OTHER__SHIFT 0x1 +#define CPG_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define CPG_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define CPG_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define CPG_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define CPG_CE_ERR_STATUS_HI__RESERVED__SHIFT 0x1b +#define CPG_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define CPG_CE_ERR_STATUS_HI__OTHER_MASK 0x00000002L +#define CPG_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define CPG_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define CPG_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define CPG_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define CPG_CE_ERR_STATUS_HI__RESERVED_MASK 0xF8000000L // addressBlock: xcd0_gc_cppdec2 @@ -22764,6 +23760,74 @@ #define SPI_WAVE_LIMIT_CNTL__GS_WAVE_GRAN_MASK 0x00000030L #define SPI_WAVE_LIMIT_CNTL__HS_WAVE_GRAN_MASK 0x000000C0L +// addressBlock: xcd0_gc_gccanedec +//GC_CANE_ERR_STATUS +#define GC_CANE_ERR_STATUS__SDPM_RDRSP_STATUS__SHIFT 0x0 +#define GC_CANE_ERR_STATUS__SDPM_WRRSP_STATUS__SHIFT 0x4 +#define GC_CANE_ERR_STATUS__SDPM_RDRSP_DATASTATUS__SHIFT 0x8 +#define GC_CANE_ERR_STATUS__SDPM_RDRSP_DATAPARITY_ERROR__SHIFT 0xa +#define GC_CANE_ERR_STATUS__SDPS_DAT_ERROR__SHIFT 0xb +#define GC_CANE_ERR_STATUS__SDPS_DAT_PARITY_ERROR__SHIFT 0xc +#define GC_CANE_ERR_STATUS__CLEAR_ERROR_STATUS__SHIFT 0xd +#define GC_CANE_ERR_STATUS__BUSY_ON_ERROR__SHIFT 0xe +#define GC_CANE_ERR_STATUS__BUSY_ON_UER_ERROR__SHIFT 0xf +#define GC_CANE_ERR_STATUS__FUE_FLAG__SHIFT 0x10 +#define GC_CANE_ERR_STATUS__INTERRUPT_ON_FATAL__SHIFT 0x11 +#define GC_CANE_ERR_STATUS__LEVEL_INTERRUPT__SHIFT 0x12 +#define GC_CANE_ERR_STATUS__SDPM_RDRSP_STATUS_MASK 0x0000000FL +#define GC_CANE_ERR_STATUS__SDPM_WRRSP_STATUS_MASK 0x000000F0L +#define GC_CANE_ERR_STATUS__SDPM_RDRSP_DATASTATUS_MASK 0x00000300L +#define GC_CANE_ERR_STATUS__SDPM_RDRSP_DATAPARITY_ERROR_MASK 0x00000400L +#define GC_CANE_ERR_STATUS__SDPS_DAT_ERROR_MASK 0x00000800L +#define GC_CANE_ERR_STATUS__SDPS_DAT_PARITY_ERROR_MASK 0x00001000L +#define GC_CANE_ERR_STATUS__CLEAR_ERROR_STATUS_MASK 0x00002000L +#define GC_CANE_ERR_STATUS__BUSY_ON_ERROR_MASK 0x00004000L +#define GC_CANE_ERR_STATUS__BUSY_ON_UER_ERROR_MASK 0x00008000L +#define GC_CANE_ERR_STATUS__FUE_FLAG_MASK 0x00010000L +#define GC_CANE_ERR_STATUS__INTERRUPT_ON_FATAL_MASK 0x00020000L +#define GC_CANE_ERR_STATUS__LEVEL_INTERRUPT_MASK 0x00040000L +//GC_CANE_UE_ERR_STATUS_LO +#define GC_CANE_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define GC_CANE_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define GC_CANE_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define GC_CANE_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define GC_CANE_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define GC_CANE_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define GC_CANE_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define GC_CANE_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//GC_CANE_UE_ERR_STATUS_HI +#define GC_CANE_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define GC_CANE_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define GC_CANE_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define GC_CANE_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define GC_CANE_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define GC_CANE_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define GC_CANE_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define GC_CANE_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define GC_CANE_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define GC_CANE_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define GC_CANE_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define GC_CANE_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +//GC_CANE_CE_ERR_STATUS_LO +#define GC_CANE_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define GC_CANE_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define GC_CANE_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define GC_CANE_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define GC_CANE_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define GC_CANE_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define GC_CANE_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define GC_CANE_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//GC_CANE_CE_ERR_STATUS_HI +#define GC_CANE_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define GC_CANE_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define GC_CANE_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define GC_CANE_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define GC_CANE_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define GC_CANE_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define GC_CANE_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define GC_CANE_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define GC_CANE_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define GC_CANE_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L // addressBlock: xcd0_gc_perfddec //CPG_PERFCOUNTER1_LO @@ -26471,6 +27535,30 @@ //RLC_CPG_STAT_INVAL #define RLC_CPG_STAT_INVAL__CPG_stat_inval__SHIFT 0x0 #define RLC_CPG_STAT_INVAL__CPG_stat_inval_MASK 0x00000001L +//RLC_UE_ERR_STATUS_LOW +#define RLC_UE_ERR_STATUS_LOW__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define RLC_UE_ERR_STATUS_LOW__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define RLC_UE_ERR_STATUS_LOW__ADDRESS__SHIFT 0x2 +#define RLC_UE_ERR_STATUS_LOW__MEMORY_ID__SHIFT 0x18 +#define RLC_UE_ERR_STATUS_LOW__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define RLC_UE_ERR_STATUS_LOW__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define RLC_UE_ERR_STATUS_LOW__ADDRESS_MASK 0x00FFFFFCL +#define RLC_UE_ERR_STATUS_LOW__MEMORY_ID_MASK 0xFF000000L +//RLC_UE_ERR_STATUS_HIGH +#define RLC_UE_ERR_STATUS_HIGH__ECC__SHIFT 0x0 +#define RLC_UE_ERR_STATUS_HIGH__PARITY__SHIFT 0x1 +#define RLC_UE_ERR_STATUS_HIGH__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define RLC_UE_ERR_STATUS_HIGH__ERR_INFO__SHIFT 0x3 +#define RLC_UE_ERR_STATUS_HIGH__UE_CNT__SHIFT 0x17 +#define RLC_UE_ERR_STATUS_HIGH__FED_CNT__SHIFT 0x1a +#define RLC_UE_ERR_STATUS_HIGH__RESERVED__SHIFT 0x1d +#define RLC_UE_ERR_STATUS_HIGH__ECC_MASK 0x00000001L +#define RLC_UE_ERR_STATUS_HIGH__PARITY_MASK 0x00000002L +#define RLC_UE_ERR_STATUS_HIGH__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define RLC_UE_ERR_STATUS_HIGH__ERR_INFO_MASK 0x007FFFF8L +#define RLC_UE_ERR_STATUS_HIGH__UE_CNT_MASK 0x03800000L +#define RLC_UE_ERR_STATUS_HIGH__FED_CNT_MASK 0x1C000000L +#define RLC_UE_ERR_STATUS_HIGH__RESERVED_MASK 0xE0000000L //RLC_DSM_CNTL #define RLC_DSM_CNTL__RLCG_INSTR_RAM_IRRITATOR_DATA_SEL__SHIFT 0x0 #define RLC_DSM_CNTL__RLCG_INSTR_RAM_IRRITATOR_SINGLE_WRITE__SHIFT 0x2 @@ -26573,6 +27661,30 @@ #define RLC_DSM_CNTL2A__RLC_SPM_SE2_SCRATCH_RAM_SELECT_INJECT_DELAY_MASK 0x00000100L #define RLC_DSM_CNTL2A__RLC_SPM_SE3_SCRATCH_RAM_ENABLE_ERROR_INJECT_MASK 0x00000600L #define RLC_DSM_CNTL2A__RLC_SPM_SE3_SCRATCH_RAM_SELECT_INJECT_DELAY_MASK 0x00000800L +//RLC_CE_ERR_STATUS_LOW +#define RLC_CE_ERR_STATUS_LOW__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define RLC_CE_ERR_STATUS_LOW__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define RLC_CE_ERR_STATUS_LOW__ADDRESS__SHIFT 0x2 +#define RLC_CE_ERR_STATUS_LOW__MEMORY_ID__SHIFT 0x18 +#define RLC_CE_ERR_STATUS_LOW__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define RLC_CE_ERR_STATUS_LOW__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define RLC_CE_ERR_STATUS_LOW__ADDRESS_MASK 0x00FFFFFCL +#define RLC_CE_ERR_STATUS_LOW__MEMORY_ID_MASK 0xFF000000L +//RLC_CE_ERR_STATUS_HIGH +#define RLC_CE_ERR_STATUS_HIGH__ECC__SHIFT 0x0 +#define RLC_CE_ERR_STATUS_HIGH__OTHER__SHIFT 0x1 +#define RLC_CE_ERR_STATUS_HIGH__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define RLC_CE_ERR_STATUS_HIGH__ERR_INFO__SHIFT 0x3 +#define RLC_CE_ERR_STATUS_HIGH__CE_CNT__SHIFT 0x17 +#define RLC_CE_ERR_STATUS_HIGH__POISON__SHIFT 0x1a +#define RLC_CE_ERR_STATUS_HIGH__RESERVED__SHIFT 0x1b +#define RLC_CE_ERR_STATUS_HIGH__ECC_MASK 0x00000001L +#define RLC_CE_ERR_STATUS_HIGH__OTHER_MASK 0x00000002L +#define RLC_CE_ERR_STATUS_HIGH__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define RLC_CE_ERR_STATUS_HIGH__ERR_INFO_MASK 0x007FFFF8L +#define RLC_CE_ERR_STATUS_HIGH__CE_CNT_MASK 0x03800000L +#define RLC_CE_ERR_STATUS_HIGH__POISON_MASK 0x04000000L +#define RLC_CE_ERR_STATUS_HIGH__RESERVED_MASK 0xF8000000L //RLC_RLCV_SPARE_INT #define RLC_RLCV_SPARE_INT__INTERRUPT__SHIFT 0x0 #define RLC_RLCV_SPARE_INT__RESERVED__SHIFT 0x1 diff --git a/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_8_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_8_0_offset.h index 8bcc81f2dfc0b890acc1846dc95e9cac7ee8bbc3..879ee9de3ff310be0b41effea2cfb8ada0cbb631 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_8_0_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_8_0_offset.h @@ -1491,6 +1491,10 @@ #define regMMEA0_PERFCOUNTER1_CFG_BASE_IDX 0 #define regMMEA0_PERFCOUNTER_RSLT_CNTL 0x0400 #define regMMEA0_PERFCOUNTER_RSLT_CNTL_BASE_IDX 0 +#define regMMEA0_UE_ERR_STATUS_LO 0x0406 +#define regMMEA0_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regMMEA0_UE_ERR_STATUS_HI 0x0407 +#define regMMEA0_UE_ERR_STATUS_HI_BASE_IDX 0 #define regMMEA0_DSM_CNTL 0x0408 #define regMMEA0_DSM_CNTL_BASE_IDX 0 #define regMMEA0_DSM_CNTLA 0x0409 @@ -1511,8 +1515,12 @@ #define regMMEA0_ERR_STATUS_BASE_IDX 0 #define regMMEA0_MISC2 0x0412 #define regMMEA0_MISC2_BASE_IDX 0 +#define regMMEA0_CE_ERR_STATUS_LO 0x0414 +#define regMMEA0_CE_ERR_STATUS_LO_BASE_IDX 0 #define regMMEA0_MISC_AON 0x0415 #define regMMEA0_MISC_AON_BASE_IDX 0 +#define regMMEA0_CE_ERR_STATUS_HI 0x0416 +#define regMMEA0_CE_ERR_STATUS_HI_BASE_IDX 0 // addressBlock: aid_mmhub_ea_mmeadec1 @@ -1709,6 +1717,10 @@ #define regMMEA1_PERFCOUNTER1_CFG_BASE_IDX 0 #define regMMEA1_PERFCOUNTER_RSLT_CNTL 0x0540 #define regMMEA1_PERFCOUNTER_RSLT_CNTL_BASE_IDX 0 +#define regMMEA1_UE_ERR_STATUS_LO 0x0546 +#define regMMEA1_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regMMEA1_UE_ERR_STATUS_HI 0x0547 +#define regMMEA1_UE_ERR_STATUS_HI_BASE_IDX 0 #define regMMEA1_DSM_CNTL 0x0548 #define regMMEA1_DSM_CNTL_BASE_IDX 0 #define regMMEA1_DSM_CNTLA 0x0549 @@ -1729,8 +1741,12 @@ #define regMMEA1_ERR_STATUS_BASE_IDX 0 #define regMMEA1_MISC2 0x0552 #define regMMEA1_MISC2_BASE_IDX 0 +#define regMMEA1_CE_ERR_STATUS_LO 0x0554 +#define regMMEA1_CE_ERR_STATUS_LO_BASE_IDX 0 #define regMMEA1_MISC_AON 0x0555 #define regMMEA1_MISC_AON_BASE_IDX 0 +#define regMMEA1_CE_ERR_STATUS_HI 0x0556 +#define regMMEA1_CE_ERR_STATUS_HI_BASE_IDX 0 // addressBlock: aid_mmhub_ea_mmeadec2 @@ -1927,6 +1943,10 @@ #define regMMEA2_PERFCOUNTER1_CFG_BASE_IDX 0 #define regMMEA2_PERFCOUNTER_RSLT_CNTL 0x0680 #define regMMEA2_PERFCOUNTER_RSLT_CNTL_BASE_IDX 0 +#define regMMEA2_UE_ERR_STATUS_LO 0x0686 +#define regMMEA2_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regMMEA2_UE_ERR_STATUS_HI 0x0687 +#define regMMEA2_UE_ERR_STATUS_HI_BASE_IDX 0 #define regMMEA2_DSM_CNTL 0x0688 #define regMMEA2_DSM_CNTL_BASE_IDX 0 #define regMMEA2_DSM_CNTLA 0x0689 @@ -1947,8 +1967,12 @@ #define regMMEA2_ERR_STATUS_BASE_IDX 0 #define regMMEA2_MISC2 0x0692 #define regMMEA2_MISC2_BASE_IDX 0 +#define regMMEA2_CE_ERR_STATUS_LO 0x0694 +#define regMMEA2_CE_ERR_STATUS_LO_BASE_IDX 0 #define regMMEA2_MISC_AON 0x0695 #define regMMEA2_MISC_AON_BASE_IDX 0 +#define regMMEA2_CE_ERR_STATUS_HI 0x0696 +#define regMMEA2_CE_ERR_STATUS_HI_BASE_IDX 0 // addressBlock: aid_mmhub_ea_mmeadec3 @@ -2145,6 +2169,10 @@ #define regMMEA3_PERFCOUNTER1_CFG_BASE_IDX 0 #define regMMEA3_PERFCOUNTER_RSLT_CNTL 0x07c0 #define regMMEA3_PERFCOUNTER_RSLT_CNTL_BASE_IDX 0 +#define regMMEA3_UE_ERR_STATUS_LO 0x07c6 +#define regMMEA3_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regMMEA3_UE_ERR_STATUS_HI 0x07c7 +#define regMMEA3_UE_ERR_STATUS_HI_BASE_IDX 0 #define regMMEA3_DSM_CNTL 0x07c8 #define regMMEA3_DSM_CNTL_BASE_IDX 0 #define regMMEA3_DSM_CNTLA 0x07c9 @@ -2165,9 +2193,12 @@ #define regMMEA3_ERR_STATUS_BASE_IDX 0 #define regMMEA3_MISC2 0x07d2 #define regMMEA3_MISC2_BASE_IDX 0 +#define regMMEA3_CE_ERR_STATUS_LO 0x07d4 +#define regMMEA3_CE_ERR_STATUS_LO_BASE_IDX 0 #define regMMEA3_MISC_AON 0x07d5 #define regMMEA3_MISC_AON_BASE_IDX 0 - +#define regMMEA3_CE_ERR_STATUS_HI 0x07d6 +#define regMMEA3_CE_ERR_STATUS_HI_BASE_IDX 0 // addressBlock: aid_mmhub_ea_mmeadec4 // base address: 0x62000 @@ -2363,6 +2394,10 @@ #define regMMEA4_PERFCOUNTER1_CFG_BASE_IDX 0 #define regMMEA4_PERFCOUNTER_RSLT_CNTL 0x0900 #define regMMEA4_PERFCOUNTER_RSLT_CNTL_BASE_IDX 0 +#define regMMEA4_UE_ERR_STATUS_LO 0x0906 +#define regMMEA4_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regMMEA4_UE_ERR_STATUS_HI 0x0907 +#define regMMEA4_UE_ERR_STATUS_HI_BASE_IDX 0 #define regMMEA4_DSM_CNTL 0x0908 #define regMMEA4_DSM_CNTL_BASE_IDX 0 #define regMMEA4_DSM_CNTLA 0x0909 @@ -2383,9 +2418,12 @@ #define regMMEA4_ERR_STATUS_BASE_IDX 0 #define regMMEA4_MISC2 0x0912 #define regMMEA4_MISC2_BASE_IDX 0 +#define regMMEA4_CE_ERR_STATUS_LO 0x0914 +#define regMMEA4_CE_ERR_STATUS_LO_BASE_IDX 0 #define regMMEA4_MISC_AON 0x0915 #define regMMEA4_MISC_AON_BASE_IDX 0 - +#define regMMEA4_CE_ERR_STATUS_HI 0x0916 +#define regMMEA4_CE_ERR_STATUS_HI_BASE_IDX 0 // addressBlock: aid_mmhub_pctldec0 // base address: 0x62a00 @@ -3310,5 +3348,19 @@ #define regL2TLB_PERFCOUNTER_HI 0x0d2d #define regL2TLB_PERFCOUNTER_HI_BASE_IDX 0 +// addressBlock: aid_mmhub_mm_cane_mmcanedec +// base address: 0x635f0 +#define regMM_CANE_ICG_CTRL 0x0d8a +#define regMM_CANE_ICG_CTRL_BASE_IDX 0 +#define regMM_CANE_ERR_STATUS 0x0d8c +#define regMM_CANE_ERR_STATUS_BASE_IDX 0 +#define regMM_CANE_UE_ERR_STATUS_LO 0x0d8d +#define regMM_CANE_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regMM_CANE_UE_ERR_STATUS_HI 0x0d8e +#define regMM_CANE_UE_ERR_STATUS_HI_BASE_IDX 0 +#define regMM_CANE_CE_ERR_STATUS_LO 0x0d8f +#define regMM_CANE_CE_ERR_STATUS_LO_BASE_IDX 0 +#define regMM_CANE_CE_ERR_STATUS_HI 0x0d90 +#define regMM_CANE_CE_ERR_STATUS_HI_BASE_IDX 0 #endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_8_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_8_0_sh_mask.h index af41468ce69ffbf92f98d992def946114ea645fd..088c1f02aa433b4728e0322fce53e1c197c6456f 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_8_0_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_8_0_sh_mask.h @@ -10470,6 +10470,30 @@ #define MMEA0_PERFCOUNTER_RSLT_CNTL__ENABLE_ANY_MASK 0x01000000L #define MMEA0_PERFCOUNTER_RSLT_CNTL__CLEAR_ALL_MASK 0x02000000L #define MMEA0_PERFCOUNTER_RSLT_CNTL__STOP_ALL_ON_SATURATE_MASK 0x04000000L +//MMEA0_UE_ERR_STATUS_LO +#define MMEA0_UE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define MMEA0_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define MMEA0_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define MMEA0_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define MMEA0_UE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define MMEA0_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define MMEA0_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define MMEA0_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//MMEA0_UE_ERR_STATUS_HI +#define MMEA0_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define MMEA0_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define MMEA0_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define MMEA0_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define MMEA0_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define MMEA0_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define MMEA0_UE_ERR_STATUS_HI__RESERVED_FIELD__SHIFT 0x1d +#define MMEA0_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define MMEA0_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define MMEA0_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define MMEA0_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define MMEA0_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define MMEA0_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define MMEA0_UE_ERR_STATUS_HI__RESERVED_FIELD_MASK 0xE0000000L //MMEA0_DSM_CNTL #define MMEA0_DSM_CNTL__DRAMRD_CMDMEM_DSM_IRRITATOR_DATA__SHIFT 0x0 #define MMEA0_DSM_CNTL__DRAMRD_CMDMEM_ENABLE_SINGLE_WRITE__SHIFT 0x2 @@ -10718,12 +10742,35 @@ #define MMEA0_MISC2__DRAM_WR_THROTTLE_MASK 0x00020000L #define MMEA0_MISC2__GMI_RD_THROTTLE_MASK 0x00040000L #define MMEA0_MISC2__GMI_WR_THROTTLE_MASK 0x00080000L +//MMEA0_CE_ERR_STATUS_LO +#define MMEA0_CE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define MMEA0_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define MMEA0_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define MMEA0_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define MMEA0_CE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define MMEA0_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define MMEA0_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define MMEA0_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L //MMEA0_MISC_AON #define MMEA0_MISC_AON__LINKMGR_PARTACK_HYSTERESIS__SHIFT 0x0 #define MMEA0_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE__SHIFT 0x2 #define MMEA0_MISC_AON__LINKMGR_PARTACK_HYSTERESIS_MASK 0x00000003L #define MMEA0_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE_MASK 0x00000004L - +//MMEA0_CE_ERR_STATUS_HI +#define MMEA0_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define MMEA0_CE_ERR_STATUS_HI__RESERVED_FIELD0__SHIFT 0x1 +#define MMEA0_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define MMEA0_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define MMEA0_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define MMEA0_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define MMEA0_CE_ERR_STATUS_HI__RESERVED_FIELD1__SHIFT 0x1b +#define MMEA0_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define MMEA0_CE_ERR_STATUS_HI__RESERVED_FIELD0_MASK 0x00000002L +#define MMEA0_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define MMEA0_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define MMEA0_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define MMEA0_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define MMEA0_CE_ERR_STATUS_HI__RESERVED_FIELD1_MASK 0xF8000000L // addressBlock: aid_mmhub_ea_mmeadec1 //MMEA1_DRAM_RD_CLI2GRP_MAP0 @@ -12418,6 +12465,30 @@ #define MMEA1_PERFCOUNTER_RSLT_CNTL__ENABLE_ANY_MASK 0x01000000L #define MMEA1_PERFCOUNTER_RSLT_CNTL__CLEAR_ALL_MASK 0x02000000L #define MMEA1_PERFCOUNTER_RSLT_CNTL__STOP_ALL_ON_SATURATE_MASK 0x04000000L +//MMEA1_UE_ERR_STATUS_LO +#define MMEA1_UE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define MMEA1_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define MMEA1_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define MMEA1_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define MMEA1_UE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define MMEA1_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define MMEA1_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define MMEA1_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//MMEA1_UE_ERR_STATUS_HI +#define MMEA1_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define MMEA1_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define MMEA1_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define MMEA1_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define MMEA1_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define MMEA1_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define MMEA1_UE_ERR_STATUS_HI__RESERVED_FIELD__SHIFT 0x1d +#define MMEA1_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define MMEA1_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define MMEA1_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define MMEA1_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define MMEA1_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define MMEA1_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define MMEA1_UE_ERR_STATUS_HI__RESERVED_FIELD_MASK 0xE0000000L //MMEA1_DSM_CNTL #define MMEA1_DSM_CNTL__DRAMRD_CMDMEM_DSM_IRRITATOR_DATA__SHIFT 0x0 #define MMEA1_DSM_CNTL__DRAMRD_CMDMEM_ENABLE_SINGLE_WRITE__SHIFT 0x2 @@ -12666,12 +12737,35 @@ #define MMEA1_MISC2__DRAM_WR_THROTTLE_MASK 0x00020000L #define MMEA1_MISC2__GMI_RD_THROTTLE_MASK 0x00040000L #define MMEA1_MISC2__GMI_WR_THROTTLE_MASK 0x00080000L +//MMEA1_CE_ERR_STATUS_LO +#define MMEA1_CE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define MMEA1_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define MMEA1_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define MMEA1_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define MMEA1_CE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define MMEA1_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define MMEA1_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define MMEA1_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L //MMEA1_MISC_AON #define MMEA1_MISC_AON__LINKMGR_PARTACK_HYSTERESIS__SHIFT 0x0 #define MMEA1_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE__SHIFT 0x2 #define MMEA1_MISC_AON__LINKMGR_PARTACK_HYSTERESIS_MASK 0x00000003L #define MMEA1_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE_MASK 0x00000004L - +//MMEA1_CE_ERR_STATUS_HI +#define MMEA1_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define MMEA1_CE_ERR_STATUS_HI__RESERVED_FIELD0__SHIFT 0x1 +#define MMEA1_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define MMEA1_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define MMEA1_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define MMEA1_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define MMEA1_CE_ERR_STATUS_HI__RESERVED_FIELD1__SHIFT 0x1b +#define MMEA1_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define MMEA1_CE_ERR_STATUS_HI__RESERVED_FIELD0_MASK 0x00000002L +#define MMEA1_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define MMEA1_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define MMEA1_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define MMEA1_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define MMEA1_CE_ERR_STATUS_HI__RESERVED_FIELD1_MASK 0xF8000000L // addressBlock: aid_mmhub_ea_mmeadec2 //MMEA2_DRAM_RD_CLI2GRP_MAP0 @@ -14366,6 +14460,30 @@ #define MMEA2_PERFCOUNTER_RSLT_CNTL__ENABLE_ANY_MASK 0x01000000L #define MMEA2_PERFCOUNTER_RSLT_CNTL__CLEAR_ALL_MASK 0x02000000L #define MMEA2_PERFCOUNTER_RSLT_CNTL__STOP_ALL_ON_SATURATE_MASK 0x04000000L +//MMEA2_UE_ERR_STATUS_LO +#define MMEA2_UE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define MMEA2_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define MMEA2_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define MMEA2_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define MMEA2_UE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define MMEA2_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define MMEA2_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define MMEA2_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//MMEA2_UE_ERR_STATUS_HI +#define MMEA2_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define MMEA2_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define MMEA2_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define MMEA2_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define MMEA2_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define MMEA2_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define MMEA2_UE_ERR_STATUS_HI__RESERVED_FIELD__SHIFT 0x1d +#define MMEA2_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define MMEA2_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define MMEA2_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define MMEA2_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define MMEA2_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define MMEA2_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define MMEA2_UE_ERR_STATUS_HI__RESERVED_FIELD_MASK 0xE0000000L //MMEA2_DSM_CNTL #define MMEA2_DSM_CNTL__DRAMRD_CMDMEM_DSM_IRRITATOR_DATA__SHIFT 0x0 #define MMEA2_DSM_CNTL__DRAMRD_CMDMEM_ENABLE_SINGLE_WRITE__SHIFT 0x2 @@ -14614,12 +14732,35 @@ #define MMEA2_MISC2__DRAM_WR_THROTTLE_MASK 0x00020000L #define MMEA2_MISC2__GMI_RD_THROTTLE_MASK 0x00040000L #define MMEA2_MISC2__GMI_WR_THROTTLE_MASK 0x00080000L +//MMEA2_CE_ERR_STATUS_LO +#define MMEA2_CE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define MMEA2_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define MMEA2_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define MMEA2_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define MMEA2_CE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define MMEA2_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define MMEA2_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define MMEA2_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L //MMEA2_MISC_AON #define MMEA2_MISC_AON__LINKMGR_PARTACK_HYSTERESIS__SHIFT 0x0 #define MMEA2_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE__SHIFT 0x2 #define MMEA2_MISC_AON__LINKMGR_PARTACK_HYSTERESIS_MASK 0x00000003L #define MMEA2_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE_MASK 0x00000004L - +//MMEA2_CE_ERR_STATUS_HI +#define MMEA2_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define MMEA2_CE_ERR_STATUS_HI__RESERVED_FIELD0__SHIFT 0x1 +#define MMEA2_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define MMEA2_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define MMEA2_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define MMEA2_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define MMEA2_CE_ERR_STATUS_HI__RESERVED_FIELD1__SHIFT 0x1b +#define MMEA2_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define MMEA2_CE_ERR_STATUS_HI__RESERVED_FIELD0_MASK 0x00000002L +#define MMEA2_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define MMEA2_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define MMEA2_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define MMEA2_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define MMEA2_CE_ERR_STATUS_HI__RESERVED_FIELD1_MASK 0xF8000000L // addressBlock: aid_mmhub_ea_mmeadec3 //MMEA3_DRAM_RD_CLI2GRP_MAP0 @@ -16314,6 +16455,30 @@ #define MMEA3_PERFCOUNTER_RSLT_CNTL__ENABLE_ANY_MASK 0x01000000L #define MMEA3_PERFCOUNTER_RSLT_CNTL__CLEAR_ALL_MASK 0x02000000L #define MMEA3_PERFCOUNTER_RSLT_CNTL__STOP_ALL_ON_SATURATE_MASK 0x04000000L +//MMEA3_UE_ERR_STATUS_LO +#define MMEA3_UE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define MMEA3_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define MMEA3_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define MMEA3_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define MMEA3_UE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define MMEA3_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define MMEA3_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define MMEA3_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//MMEA3_UE_ERR_STATUS_HI +#define MMEA3_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define MMEA3_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define MMEA3_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define MMEA3_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define MMEA3_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define MMEA3_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define MMEA3_UE_ERR_STATUS_HI__RESERVED_FIELD__SHIFT 0x1d +#define MMEA3_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define MMEA3_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define MMEA3_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define MMEA3_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define MMEA3_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define MMEA3_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define MMEA3_UE_ERR_STATUS_HI__RESERVED_FIELD_MASK 0xE0000000L //MMEA3_DSM_CNTL #define MMEA3_DSM_CNTL__DRAMRD_CMDMEM_DSM_IRRITATOR_DATA__SHIFT 0x0 #define MMEA3_DSM_CNTL__DRAMRD_CMDMEM_ENABLE_SINGLE_WRITE__SHIFT 0x2 @@ -16562,12 +16727,35 @@ #define MMEA3_MISC2__DRAM_WR_THROTTLE_MASK 0x00020000L #define MMEA3_MISC2__GMI_RD_THROTTLE_MASK 0x00040000L #define MMEA3_MISC2__GMI_WR_THROTTLE_MASK 0x00080000L +//MMEA3_CE_ERR_STATUS_LO +#define MMEA3_CE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define MMEA3_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define MMEA3_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define MMEA3_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define MMEA3_CE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define MMEA3_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define MMEA3_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define MMEA3_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L //MMEA3_MISC_AON #define MMEA3_MISC_AON__LINKMGR_PARTACK_HYSTERESIS__SHIFT 0x0 #define MMEA3_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE__SHIFT 0x2 #define MMEA3_MISC_AON__LINKMGR_PARTACK_HYSTERESIS_MASK 0x00000003L #define MMEA3_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE_MASK 0x00000004L - +//MMEA3_CE_ERR_STATUS_HI +#define MMEA3_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define MMEA3_CE_ERR_STATUS_HI__RESERVED_FIELD0__SHIFT 0x1 +#define MMEA3_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define MMEA3_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define MMEA3_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define MMEA3_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define MMEA3_CE_ERR_STATUS_HI__RESERVED_FIELD1__SHIFT 0x1b +#define MMEA3_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define MMEA3_CE_ERR_STATUS_HI__RESERVED_FIELD0_MASK 0x00000002L +#define MMEA3_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define MMEA3_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define MMEA3_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define MMEA3_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define MMEA3_CE_ERR_STATUS_HI__RESERVED_FIELD1_MASK 0xF8000000L // addressBlock: aid_mmhub_ea_mmeadec4 //MMEA4_DRAM_RD_CLI2GRP_MAP0 @@ -18262,6 +18450,30 @@ #define MMEA4_PERFCOUNTER_RSLT_CNTL__ENABLE_ANY_MASK 0x01000000L #define MMEA4_PERFCOUNTER_RSLT_CNTL__CLEAR_ALL_MASK 0x02000000L #define MMEA4_PERFCOUNTER_RSLT_CNTL__STOP_ALL_ON_SATURATE_MASK 0x04000000L +//MMEA4_UE_ERR_STATUS_LO +#define MMEA4_UE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define MMEA4_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define MMEA4_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define MMEA4_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define MMEA4_UE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define MMEA4_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define MMEA4_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define MMEA4_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//MMEA4_UE_ERR_STATUS_HI +#define MMEA4_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define MMEA4_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define MMEA4_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define MMEA4_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define MMEA4_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define MMEA4_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define MMEA4_UE_ERR_STATUS_HI__RESERVED_FIELD__SHIFT 0x1d +#define MMEA4_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define MMEA4_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define MMEA4_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define MMEA4_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define MMEA4_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define MMEA4_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define MMEA4_UE_ERR_STATUS_HI__RESERVED_FIELD_MASK 0xE0000000L //MMEA4_DSM_CNTL #define MMEA4_DSM_CNTL__DRAMRD_CMDMEM_DSM_IRRITATOR_DATA__SHIFT 0x0 #define MMEA4_DSM_CNTL__DRAMRD_CMDMEM_ENABLE_SINGLE_WRITE__SHIFT 0x2 @@ -18510,12 +18722,35 @@ #define MMEA4_MISC2__DRAM_WR_THROTTLE_MASK 0x00020000L #define MMEA4_MISC2__GMI_RD_THROTTLE_MASK 0x00040000L #define MMEA4_MISC2__GMI_WR_THROTTLE_MASK 0x00080000L +//MMEA4_CE_ERR_STATUS_LO +#define MMEA4_CE_ERR_STATUS_LO__STATUS_VALID_FLAG__SHIFT 0x0 +#define MMEA4_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define MMEA4_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define MMEA4_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define MMEA4_CE_ERR_STATUS_LO__STATUS_VALID_FLAG_MASK 0x00000001L +#define MMEA4_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define MMEA4_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define MMEA4_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L //MMEA4_MISC_AON #define MMEA4_MISC_AON__LINKMGR_PARTACK_HYSTERESIS__SHIFT 0x0 #define MMEA4_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE__SHIFT 0x2 #define MMEA4_MISC_AON__LINKMGR_PARTACK_HYSTERESIS_MASK 0x00000003L #define MMEA4_MISC_AON__LINKMGR_PARTACK_DEASSERT_MODE_MASK 0x00000004L - +//MMEA4_CE_ERR_STATUS_HI +#define MMEA4_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define MMEA4_CE_ERR_STATUS_HI__RESERVED_FIELD0__SHIFT 0x1 +#define MMEA4_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define MMEA4_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define MMEA4_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define MMEA4_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define MMEA4_CE_ERR_STATUS_HI__RESERVED_FIELD1__SHIFT 0x1b +#define MMEA4_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define MMEA4_CE_ERR_STATUS_HI__RESERVED_FIELD0_MASK 0x00000002L +#define MMEA4_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define MMEA4_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define MMEA4_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define MMEA4_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L +#define MMEA4_CE_ERR_STATUS_HI__RESERVED_FIELD1_MASK 0xF8000000L // addressBlock: aid_mmhub_pctldec0 //PCTL0_CTRL @@ -22311,5 +22546,83 @@ #define L2TLB_PERFCOUNTER_HI__COUNTER_HI_MASK 0x0000FFFFL #define L2TLB_PERFCOUNTER_HI__COMPARE_VALUE_MASK 0xFFFF0000L - +// addressBlock: aid_mmhub_mm_cane_mmcanedec +//MM_CANE_ICG_CTRL +#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_IREQ0__SHIFT 0x0 +#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_ATRET__SHIFT 0x1 +#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_OREQ__SHIFT 0x2 +#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_REGISTER__SHIFT 0x3 +#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_SDPM_RETURN__SHIFT 0x4 +#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_IREQ0_MASK 0x00000001L +#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_ATRET_MASK 0x00000002L +#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_OREQ_MASK 0x00000004L +#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_REGISTER_MASK 0x00000008L +#define MM_CANE_ICG_CTRL__SOFT_OVERRIDE_SDPM_RETURN_MASK 0x00000010L +//MM_CANE_ERR_STATUS +#define MM_CANE_ERR_STATUS__SDPM_RDRSP_STATUS__SHIFT 0x0 +#define MM_CANE_ERR_STATUS__SDPM_WRRSP_STATUS__SHIFT 0x4 +#define MM_CANE_ERR_STATUS__SDPM_RDRSP_DATASTATUS__SHIFT 0x8 +#define MM_CANE_ERR_STATUS__SDPM_RDRSP_DATAPARITY_ERROR__SHIFT 0xa +#define MM_CANE_ERR_STATUS__SDPS_DAT_ERROR__SHIFT 0xb +#define MM_CANE_ERR_STATUS__SDPS_DAT_PARITY_ERROR__SHIFT 0xc +#define MM_CANE_ERR_STATUS__CLEAR_ERROR_STATUS__SHIFT 0xd +#define MM_CANE_ERR_STATUS__BUSY_ON_ERROR__SHIFT 0xe +#define MM_CANE_ERR_STATUS__BUSY_ON_UER_ERROR__SHIFT 0xf +#define MM_CANE_ERR_STATUS__FUE_FLAG__SHIFT 0x10 +#define MM_CANE_ERR_STATUS__INTERRUPT_ON_FATAL__SHIFT 0x11 +#define MM_CANE_ERR_STATUS__LEVEL_INTERRUPT__SHIFT 0x12 +#define MM_CANE_ERR_STATUS__SDPM_RDRSP_STATUS_MASK 0x0000000FL +#define MM_CANE_ERR_STATUS__SDPM_WRRSP_STATUS_MASK 0x000000F0L +#define MM_CANE_ERR_STATUS__SDPM_RDRSP_DATASTATUS_MASK 0x00000300L +#define MM_CANE_ERR_STATUS__SDPM_RDRSP_DATAPARITY_ERROR_MASK 0x00000400L +#define MM_CANE_ERR_STATUS__SDPS_DAT_ERROR_MASK 0x00000800L +#define MM_CANE_ERR_STATUS__SDPS_DAT_PARITY_ERROR_MASK 0x00001000L +#define MM_CANE_ERR_STATUS__CLEAR_ERROR_STATUS_MASK 0x00002000L +#define MM_CANE_ERR_STATUS__BUSY_ON_ERROR_MASK 0x00004000L +#define MM_CANE_ERR_STATUS__BUSY_ON_UER_ERROR_MASK 0x00008000L +#define MM_CANE_ERR_STATUS__FUE_FLAG_MASK 0x00010000L +#define MM_CANE_ERR_STATUS__INTERRUPT_ON_FATAL_MASK 0x00020000L +#define MM_CANE_ERR_STATUS__LEVEL_INTERRUPT_MASK 0x00040000L +//MM_CANE_UE_ERR_STATUS_LO +#define MM_CANE_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define MM_CANE_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define MM_CANE_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define MM_CANE_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define MM_CANE_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define MM_CANE_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define MM_CANE_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define MM_CANE_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//MM_CANE_UE_ERR_STATUS_HI +#define MM_CANE_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define MM_CANE_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define MM_CANE_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define MM_CANE_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define MM_CANE_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define MM_CANE_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define MM_CANE_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define MM_CANE_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define MM_CANE_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define MM_CANE_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define MM_CANE_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define MM_CANE_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +//MM_CANE_CE_ERR_STATUS_LO +#define MM_CANE_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define MM_CANE_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define MM_CANE_CE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define MM_CANE_CE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define MM_CANE_CE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define MM_CANE_CE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define MM_CANE_CE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define MM_CANE_CE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//MM_CANE_CE_ERR_STATUS_HI +#define MM_CANE_CE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define MM_CANE_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define MM_CANE_CE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define MM_CANE_CE_ERR_STATUS_HI__CE_CNT__SHIFT 0x17 +#define MM_CANE_CE_ERR_STATUS_HI__POISON__SHIFT 0x1a +#define MM_CANE_CE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define MM_CANE_CE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define MM_CANE_CE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define MM_CANE_CE_ERR_STATUS_HI__CE_CNT_MASK 0x03800000L +#define MM_CANE_CE_ERR_STATUS_HI__POISON_MASK 0x04000000L #endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_9_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_9_0_offset.h index 033f2796c1e37208026ac2d838287e0f48bd2b73..c8a15c8f4822808757380f6504fd431219cc449e 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_9_0_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_9_0_offset.h @@ -6201,6 +6201,8 @@ #define regNBIF_SHUB_TODET_SYNCFLOOD_CTRL2_BASE_IDX 8 #define regBIFC_BME_ERR_LOG_HB 0xe8ab #define regBIFC_BME_ERR_LOG_HB_BASE_IDX 8 +#define regBIFC_GFX_INT_MONITOR_MASK 0xe8ad +#define regBIFC_GFX_INT_MONITOR_MASK_BASE_IDX 8 #define regBIFC_HRP_SDP_WRRSP_POOLCRED_ALLOC 0xe8c0 #define regBIFC_HRP_SDP_WRRSP_POOLCRED_ALLOC_BASE_IDX 8 #define regBIFC_HRP_SDP_RDRSP_POOLCRED_ALLOC 0xe8c1 diff --git a/drivers/gpu/drm/amd/include/asic_reg/sdma/sdma_4_4_2_offset.h b/drivers/gpu/drm/amd/include/asic_reg/sdma/sdma_4_4_2_offset.h index 31bef0776ded568e70f9db4891f2636c9630afa2..ead81aeffd67903976b78fa124c994da95bf64ea 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/sdma/sdma_4_4_2_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/sdma/sdma_4_4_2_offset.h @@ -211,6 +211,10 @@ #define regSDMA_RAS_STATUS_BASE_IDX 0 #define regSDMA_CLK_STATUS 0x0068 #define regSDMA_CLK_STATUS_BASE_IDX 0 +#define regSDMA_UE_ERR_STATUS_LO 0x0069 +#define regSDMA_UE_ERR_STATUS_LO_BASE_IDX 0 +#define regSDMA_UE_ERR_STATUS_HI 0x006a +#define regSDMA_UE_ERR_STATUS_HI_BASE_IDX 0 #define regSDMA_POWER_CNTL 0x006b #define regSDMA_POWER_CNTL_BASE_IDX 0 #define regSDMA_CLK_CTRL 0x006c diff --git a/drivers/gpu/drm/amd/include/asic_reg/sdma/sdma_4_4_2_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/sdma/sdma_4_4_2_sh_mask.h index e46cb33393554f8b627336389b90298e51cbabcb..290953bdf1d699a21ed72bbf135f53c85e325e8c 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/sdma/sdma_4_4_2_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/sdma/sdma_4_4_2_sh_mask.h @@ -1171,6 +1171,30 @@ #define SDMA_CLK_STATUS__F32_CLK_MASK 0x00000008L #define SDMA_CLK_STATUS__CE_CLK_MASK 0x00000010L #define SDMA_CLK_STATUS__PERF_CLK_MASK 0x00000020L +//SDMA_UE_ERR_STATUS_LO +#define SDMA_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT 0x0 +#define SDMA_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG__SHIFT 0x1 +#define SDMA_UE_ERR_STATUS_LO__ADDRESS__SHIFT 0x2 +#define SDMA_UE_ERR_STATUS_LO__MEMORY_ID__SHIFT 0x18 +#define SDMA_UE_ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK 0x00000001L +#define SDMA_UE_ERR_STATUS_LO__ADDRESS_VALID_FLAG_MASK 0x00000002L +#define SDMA_UE_ERR_STATUS_LO__ADDRESS_MASK 0x00FFFFFCL +#define SDMA_UE_ERR_STATUS_LO__MEMORY_ID_MASK 0xFF000000L +//SDMA_UE_ERR_STATUS_HI +#define SDMA_UE_ERR_STATUS_HI__ECC__SHIFT 0x0 +#define SDMA_UE_ERR_STATUS_HI__PARITY__SHIFT 0x1 +#define SDMA_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG__SHIFT 0x2 +#define SDMA_UE_ERR_STATUS_HI__ERR_INFO__SHIFT 0x3 +#define SDMA_UE_ERR_STATUS_HI__UE_CNT__SHIFT 0x17 +#define SDMA_UE_ERR_STATUS_HI__FED_CNT__SHIFT 0x1a +#define SDMA_UE_ERR_STATUS_HI__RESERVED__SHIFT 0x1d +#define SDMA_UE_ERR_STATUS_HI__ECC_MASK 0x00000001L +#define SDMA_UE_ERR_STATUS_HI__PARITY_MASK 0x00000002L +#define SDMA_UE_ERR_STATUS_HI__ERR_INFO_VALID_FLAG_MASK 0x00000004L +#define SDMA_UE_ERR_STATUS_HI__ERR_INFO_MASK 0x007FFFF8L +#define SDMA_UE_ERR_STATUS_HI__UE_CNT_MASK 0x03800000L +#define SDMA_UE_ERR_STATUS_HI__FED_CNT_MASK 0x1C000000L +#define SDMA_UE_ERR_STATUS_HI__RESERVED_MASK 0xE0000000L //SDMA_POWER_CNTL #define SDMA_POWER_CNTL__PG_CNTL_ENABLE__SHIFT 0x0 #define SDMA_POWER_CNTL__EXT_PG_POWER_ON_REQ__SHIFT 0x1 diff --git a/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_13_0_3_offset.h b/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_13_0_3_offset.h new file mode 100644 index 0000000000000000000000000000000000000000..b62b489402c54e5acc5267f5c0a50b0aaae7eb75 --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_13_0_3_offset.h @@ -0,0 +1,177 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +#ifndef _smuio_13_0_3_OFFSET_HEADER +#define _smuio_13_0_3_OFFSET_HEADER + + + +// addressBlock: aid_smuio_smuio_reset_SmuSmuioDec +// base address: 0x5a300 +#define regSMUIO_MP_RESET_INTR 0x00c1 +#define regSMUIO_MP_RESET_INTR_BASE_IDX 1 +#define regSMUIO_SOC_HALT 0x00c2 +#define regSMUIO_SOC_HALT_BASE_IDX 1 + + +// addressBlock: aid_smuio_smuio_tsc_SmuSmuioDec +// base address: 0x5a8a0 +#define regPWROK_REFCLK_GAP_CYCLES 0x0028 +#define regPWROK_REFCLK_GAP_CYCLES_BASE_IDX 2 +#define regGOLDEN_TSC_INCREMENT_UPPER 0x002b +#define regGOLDEN_TSC_INCREMENT_UPPER_BASE_IDX 2 +#define regGOLDEN_TSC_INCREMENT_LOWER 0x002c +#define regGOLDEN_TSC_INCREMENT_LOWER_BASE_IDX 2 +#define regGOLDEN_TSC_COUNT_UPPER 0x002d +#define regGOLDEN_TSC_COUNT_UPPER_BASE_IDX 2 +#define regGOLDEN_TSC_COUNT_LOWER 0x002e +#define regGOLDEN_TSC_COUNT_LOWER_BASE_IDX 2 +#define regSOC_GOLDEN_TSC_SHADOW_UPPER 0x002f +#define regSOC_GOLDEN_TSC_SHADOW_UPPER_BASE_IDX 2 +#define regSOC_GOLDEN_TSC_SHADOW_LOWER 0x0030 +#define regSOC_GOLDEN_TSC_SHADOW_LOWER_BASE_IDX 2 +#define regSOC_GAP_PWROK 0x0031 +#define regSOC_GAP_PWROK_BASE_IDX 2 + + +// addressBlock: aid_smuio_smuio_swtimer_SmuSmuioDec +// base address: 0x5ac70 +#define regPWR_VIRT_RESET_REQ 0x011c +#define regPWR_VIRT_RESET_REQ_BASE_IDX 2 +#define regPWR_DISP_TIMER_CONTROL 0x011d +#define regPWR_DISP_TIMER_CONTROL_BASE_IDX 2 +#define regPWR_DISP_TIMER_DEBUG 0x011e +#define regPWR_DISP_TIMER_DEBUG_BASE_IDX 2 +#define regPWR_DISP_TIMER2_CONTROL 0x011f +#define regPWR_DISP_TIMER2_CONTROL_BASE_IDX 2 +#define regPWR_DISP_TIMER2_DEBUG 0x0120 +#define regPWR_DISP_TIMER2_DEBUG_BASE_IDX 2 +#define regPWR_DISP_TIMER_GLOBAL_CONTROL 0x0121 +#define regPWR_DISP_TIMER_GLOBAL_CONTROL_BASE_IDX 2 +#define regPWR_IH_CONTROL 0x0122 +#define regPWR_IH_CONTROL_BASE_IDX 2 + + +// addressBlock: aid_smuio_smuio_misc_SmuSmuioDec +// base address: 0x5a000 +#define regSMUIO_MCM_CONFIG 0x0023 +#define regSMUIO_MCM_CONFIG_BASE_IDX 1 +#define regIP_DISCOVERY_VERSION 0x0000 +#define regIP_DISCOVERY_VERSION_BASE_IDX 2 +#define regSCRATCH_REGISTER0 0x01bd +#define regSCRATCH_REGISTER0_BASE_IDX 2 +#define regSCRATCH_REGISTER1 0x01be +#define regSCRATCH_REGISTER1_BASE_IDX 2 +#define regSCRATCH_REGISTER2 0x01bf +#define regSCRATCH_REGISTER2_BASE_IDX 2 +#define regSCRATCH_REGISTER3 0x01c0 +#define regSCRATCH_REGISTER3_BASE_IDX 2 +#define regSCRATCH_REGISTER4 0x01c1 +#define regSCRATCH_REGISTER4_BASE_IDX 2 +#define regSCRATCH_REGISTER5 0x01c2 +#define regSCRATCH_REGISTER5_BASE_IDX 2 +#define regSCRATCH_REGISTER6 0x01c3 +#define regSCRATCH_REGISTER6_BASE_IDX 2 +#define regSCRATCH_REGISTER7 0x01c4 +#define regSCRATCH_REGISTER7_BASE_IDX 2 + + +// addressBlock: aid_smuio_smuio_gpio_SmuSmuioDec +// base address: 0x5a500 +#define regSMU_GPIOPAD_SW_INT_STAT 0x0140 +#define regSMU_GPIOPAD_SW_INT_STAT_BASE_IDX 1 +#define regSMU_GPIOPAD_MASK 0x0141 +#define regSMU_GPIOPAD_MASK_BASE_IDX 1 +#define regSMU_GPIOPAD_A 0x0142 +#define regSMU_GPIOPAD_A_BASE_IDX 1 +#define regSMU_GPIOPAD_TXIMPSEL 0x0143 +#define regSMU_GPIOPAD_TXIMPSEL_BASE_IDX 1 +#define regSMU_GPIOPAD_EN 0x0144 +#define regSMU_GPIOPAD_EN_BASE_IDX 1 +#define regSMU_GPIOPAD_Y 0x0145 +#define regSMU_GPIOPAD_Y_BASE_IDX 1 +#define regSMU_GPIOPAD_RXEN 0x0146 +#define regSMU_GPIOPAD_RXEN_BASE_IDX 1 +#define regSMU_GPIOPAD_RCVR_SEL0 0x0147 +#define regSMU_GPIOPAD_RCVR_SEL0_BASE_IDX 1 +#define regSMU_GPIOPAD_RCVR_SEL1 0x0148 +#define regSMU_GPIOPAD_RCVR_SEL1_BASE_IDX 1 +#define regSMU_GPIOPAD_PU_EN 0x0149 +#define regSMU_GPIOPAD_PU_EN_BASE_IDX 1 +#define regSMU_GPIOPAD_PD_EN 0x014a +#define regSMU_GPIOPAD_PD_EN_BASE_IDX 1 +#define regSMU_GPIOPAD_PINSTRAPS 0x014b +#define regSMU_GPIOPAD_PINSTRAPS_BASE_IDX 1 +#define regDFT_PINSTRAPS 0x014c +#define regDFT_PINSTRAPS_BASE_IDX 1 +#define regSMU_GPIOPAD_INT_STAT_EN 0x014d +#define regSMU_GPIOPAD_INT_STAT_EN_BASE_IDX 1 +#define regSMU_GPIOPAD_INT_STAT 0x014e +#define regSMU_GPIOPAD_INT_STAT_BASE_IDX 1 +#define regSMU_GPIOPAD_INT_STAT_AK 0x014f +#define regSMU_GPIOPAD_INT_STAT_AK_BASE_IDX 1 +#define regSMU_GPIOPAD_INT_EN 0x0150 +#define regSMU_GPIOPAD_INT_EN_BASE_IDX 1 +#define regSMU_GPIOPAD_INT_TYPE 0x0151 +#define regSMU_GPIOPAD_INT_TYPE_BASE_IDX 1 +#define regSMU_GPIOPAD_INT_POLARITY 0x0152 +#define regSMU_GPIOPAD_INT_POLARITY_BASE_IDX 1 +#define regSMUIO_PCC_GPIO_SELECT 0x0155 +#define regSMUIO_PCC_GPIO_SELECT_BASE_IDX 1 +#define regSMU_GPIOPAD_S0 0x0156 +#define regSMU_GPIOPAD_S0_BASE_IDX 1 +#define regSMU_GPIOPAD_S1 0x0157 +#define regSMU_GPIOPAD_S1_BASE_IDX 1 +#define regSMU_GPIOPAD_SCHMEN 0x0158 +#define regSMU_GPIOPAD_SCHMEN_BASE_IDX 1 +#define regSMU_GPIOPAD_SCL_EN 0x0159 +#define regSMU_GPIOPAD_SCL_EN_BASE_IDX 1 +#define regSMU_GPIOPAD_SDA_EN 0x015a +#define regSMU_GPIOPAD_SDA_EN_BASE_IDX 1 +#define regSMUIO_GPIO_INT0_SELECT 0x015b +#define regSMUIO_GPIO_INT0_SELECT_BASE_IDX 1 +#define regSMUIO_GPIO_INT1_SELECT 0x015c +#define regSMUIO_GPIO_INT1_SELECT_BASE_IDX 1 +#define regSMUIO_GPIO_INT2_SELECT 0x015d +#define regSMUIO_GPIO_INT2_SELECT_BASE_IDX 1 +#define regSMUIO_GPIO_INT3_SELECT 0x015e +#define regSMUIO_GPIO_INT3_SELECT_BASE_IDX 1 +#define regSMU_GPIOPAD_MP_INT0_STAT 0x015f +#define regSMU_GPIOPAD_MP_INT0_STAT_BASE_IDX 1 +#define regSMU_GPIOPAD_MP_INT1_STAT 0x0160 +#define regSMU_GPIOPAD_MP_INT1_STAT_BASE_IDX 1 +#define regSMU_GPIOPAD_MP_INT2_STAT 0x0161 +#define regSMU_GPIOPAD_MP_INT2_STAT_BASE_IDX 1 +#define regSMU_GPIOPAD_MP_INT3_STAT 0x0162 +#define regSMU_GPIOPAD_MP_INT3_STAT_BASE_IDX 1 +#define regSMIO_INDEX 0x0163 +#define regSMIO_INDEX_BASE_IDX 1 +#define regS0_VID_SMIO_CNTL 0x0164 +#define regS0_VID_SMIO_CNTL_BASE_IDX 1 +#define regS1_VID_SMIO_CNTL 0x0165 +#define regS1_VID_SMIO_CNTL_BASE_IDX 1 +#define regOPEN_DRAIN_SELECT 0x0166 +#define regOPEN_DRAIN_SELECT_BASE_IDX 1 +#define regSMIO_ENABLE 0x0167 +#define regSMIO_ENABLE_BASE_IDX 1 + +#endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_13_0_3_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_13_0_3_sh_mask.h new file mode 100644 index 0000000000000000000000000000000000000000..be896f3089fe25060a976539c7b01222a69ce770 --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/smuio/smuio_13_0_3_sh_mask.h @@ -0,0 +1,428 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +#ifndef _smuio_13_0_3_SH_MASK_HEADER +#define _smuio_13_0_3_SH_MASK_HEADER + + +// addressBlock: aid_smuio_smuio_reset_SmuSmuioDec +//SMUIO_MP_RESET_INTR +#define SMUIO_MP_RESET_INTR__SMUIO_MP_RESET_INTR__SHIFT 0x0 +#define SMUIO_MP_RESET_INTR__SMUIO_MP_RESET_INTR_MASK 0x00000001L +//SMUIO_SOC_HALT +#define SMUIO_SOC_HALT__WDT_FORCE_PWROK_EN__SHIFT 0x2 +#define SMUIO_SOC_HALT__WDT_FORCE_RESETn_EN__SHIFT 0x3 +#define SMUIO_SOC_HALT__WDT_FORCE_PWROK_EN_MASK 0x00000004L +#define SMUIO_SOC_HALT__WDT_FORCE_RESETn_EN_MASK 0x00000008L + + +// addressBlock: aid_smuio_smuio_tsc_SmuSmuioDec +//PWROK_REFCLK_GAP_CYCLES +#define PWROK_REFCLK_GAP_CYCLES__Pwrok_PreAssertion_clkgap_cycles__SHIFT 0x0 +#define PWROK_REFCLK_GAP_CYCLES__Pwrok_PostAssertion_clkgap_cycles__SHIFT 0x8 +#define PWROK_REFCLK_GAP_CYCLES__Pwrok_PreAssertion_clkgap_cycles_MASK 0x000000FFL +#define PWROK_REFCLK_GAP_CYCLES__Pwrok_PostAssertion_clkgap_cycles_MASK 0x0000FF00L +//GOLDEN_TSC_INCREMENT_UPPER +#define GOLDEN_TSC_INCREMENT_UPPER__GoldenTscIncrementUpper__SHIFT 0x0 +#define GOLDEN_TSC_INCREMENT_UPPER__GoldenTscIncrementUpper_MASK 0x00FFFFFFL +//GOLDEN_TSC_INCREMENT_LOWER +#define GOLDEN_TSC_INCREMENT_LOWER__GoldenTscIncrementLower__SHIFT 0x0 +#define GOLDEN_TSC_INCREMENT_LOWER__GoldenTscIncrementLower_MASK 0xFFFFFFFFL +//GOLDEN_TSC_COUNT_UPPER +#define GOLDEN_TSC_COUNT_UPPER__GoldenTscCountUpper__SHIFT 0x0 +#define GOLDEN_TSC_COUNT_UPPER__GoldenTscCountUpper_MASK 0x00FFFFFFL +//GOLDEN_TSC_COUNT_LOWER +#define GOLDEN_TSC_COUNT_LOWER__GoldenTscCountLower__SHIFT 0x0 +#define GOLDEN_TSC_COUNT_LOWER__GoldenTscCountLower_MASK 0xFFFFFFFFL +//SOC_GOLDEN_TSC_SHADOW_UPPER +#define SOC_GOLDEN_TSC_SHADOW_UPPER__SocGoldenTscShadowUpper__SHIFT 0x0 +#define SOC_GOLDEN_TSC_SHADOW_UPPER__SocGoldenTscShadowUpper_MASK 0x00FFFFFFL +//SOC_GOLDEN_TSC_SHADOW_LOWER +#define SOC_GOLDEN_TSC_SHADOW_LOWER__SocGoldenTscShadowLower__SHIFT 0x0 +#define SOC_GOLDEN_TSC_SHADOW_LOWER__SocGoldenTscShadowLower_MASK 0xFFFFFFFFL +//SOC_GAP_PWROK +#define SOC_GAP_PWROK__soc_gap_pwrok__SHIFT 0x0 +#define SOC_GAP_PWROK__soc_gap_pwrok_MASK 0x00000001L + + +// addressBlock: aid_smuio_smuio_swtimer_SmuSmuioDec +//PWR_VIRT_RESET_REQ +#define PWR_VIRT_RESET_REQ__VF_FLR__SHIFT 0x0 +#define PWR_VIRT_RESET_REQ__PF_FLR__SHIFT 0x1f +#define PWR_VIRT_RESET_REQ__VF_FLR_MASK 0x7FFFFFFFL +#define PWR_VIRT_RESET_REQ__PF_FLR_MASK 0x80000000L +//PWR_DISP_TIMER_CONTROL +#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_COUNT__SHIFT 0x0 +#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_ENABLE__SHIFT 0x19 +#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_DISABLE__SHIFT 0x1a +#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_MASK__SHIFT 0x1b +#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_STAT_AK__SHIFT 0x1c +#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_TYPE__SHIFT 0x1d +#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_MODE__SHIFT 0x1e +#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_COUNT_MASK 0x01FFFFFFL +#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_ENABLE_MASK 0x02000000L +#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_DISABLE_MASK 0x04000000L +#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_MASK_MASK 0x08000000L +#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_STAT_AK_MASK 0x10000000L +#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_TYPE_MASK 0x20000000L +#define PWR_DISP_TIMER_CONTROL__DISP_TIMER_INT_MODE_MASK 0x40000000L +//PWR_DISP_TIMER_DEBUG +#define PWR_DISP_TIMER_DEBUG__DISP_TIMER_INT_RUNNING__SHIFT 0x0 +#define PWR_DISP_TIMER_DEBUG__DISP_TIMER_INT_STAT__SHIFT 0x1 +#define PWR_DISP_TIMER_DEBUG__DISP_TIMER_INT__SHIFT 0x2 +#define PWR_DISP_TIMER_DEBUG__DISP_TIMER_RUN_VAL__SHIFT 0x7 +#define PWR_DISP_TIMER_DEBUG__DISP_TIMER_INT_RUNNING_MASK 0x00000001L +#define PWR_DISP_TIMER_DEBUG__DISP_TIMER_INT_STAT_MASK 0x00000002L +#define PWR_DISP_TIMER_DEBUG__DISP_TIMER_INT_MASK 0x00000004L +#define PWR_DISP_TIMER_DEBUG__DISP_TIMER_RUN_VAL_MASK 0xFFFFFF80L +//PWR_DISP_TIMER2_CONTROL +#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_COUNT__SHIFT 0x0 +#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_ENABLE__SHIFT 0x19 +#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_DISABLE__SHIFT 0x1a +#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_MASK__SHIFT 0x1b +#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_STAT_AK__SHIFT 0x1c +#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_TYPE__SHIFT 0x1d +#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_MODE__SHIFT 0x1e +#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_COUNT_MASK 0x01FFFFFFL +#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_ENABLE_MASK 0x02000000L +#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_DISABLE_MASK 0x04000000L +#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_MASK_MASK 0x08000000L +#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_STAT_AK_MASK 0x10000000L +#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_TYPE_MASK 0x20000000L +#define PWR_DISP_TIMER2_CONTROL__DISP_TIMER_INT_MODE_MASK 0x40000000L +//PWR_DISP_TIMER2_DEBUG +#define PWR_DISP_TIMER2_DEBUG__DISP_TIMER_INT_RUNNING__SHIFT 0x0 +#define PWR_DISP_TIMER2_DEBUG__DISP_TIMER_INT_STAT__SHIFT 0x1 +#define PWR_DISP_TIMER2_DEBUG__DISP_TIMER_INT__SHIFT 0x2 +#define PWR_DISP_TIMER2_DEBUG__DISP_TIMER_RUN_VAL__SHIFT 0x7 +#define PWR_DISP_TIMER2_DEBUG__DISP_TIMER_INT_RUNNING_MASK 0x00000001L +#define PWR_DISP_TIMER2_DEBUG__DISP_TIMER_INT_STAT_MASK 0x00000002L +#define PWR_DISP_TIMER2_DEBUG__DISP_TIMER_INT_MASK 0x00000004L +#define PWR_DISP_TIMER2_DEBUG__DISP_TIMER_RUN_VAL_MASK 0xFFFFFF80L +//PWR_DISP_TIMER_GLOBAL_CONTROL +#define PWR_DISP_TIMER_GLOBAL_CONTROL__DISP_TIMER_PULSE_WIDTH__SHIFT 0x0 +#define PWR_DISP_TIMER_GLOBAL_CONTROL__DISP_TIMER_PULSE_EN__SHIFT 0xa +#define PWR_DISP_TIMER_GLOBAL_CONTROL__DISP_TIMER_PULSE_WIDTH_MASK 0x000003FFL +#define PWR_DISP_TIMER_GLOBAL_CONTROL__DISP_TIMER_PULSE_EN_MASK 0x00000400L +//PWR_IH_CONTROL +#define PWR_IH_CONTROL__MAX_CREDIT__SHIFT 0x0 +#define PWR_IH_CONTROL__DISP_TIMER_TRIGGER_MASK__SHIFT 0x5 +#define PWR_IH_CONTROL__DISP_TIMER2_TRIGGER_MASK__SHIFT 0x6 +#define PWR_IH_CONTROL__PWR_IH_CLK_GATE_EN__SHIFT 0x1f +#define PWR_IH_CONTROL__MAX_CREDIT_MASK 0x0000001FL +#define PWR_IH_CONTROL__DISP_TIMER_TRIGGER_MASK_MASK 0x00000020L +#define PWR_IH_CONTROL__DISP_TIMER2_TRIGGER_MASK_MASK 0x00000040L +#define PWR_IH_CONTROL__PWR_IH_CLK_GATE_EN_MASK 0x80000000L + + +// addressBlock: aid_smuio_smuio_misc_SmuSmuioDec +//SMUIO_MCM_CONFIG +#define SMUIO_MCM_CONFIG__DIE_ID__SHIFT 0x0 +#define SMUIO_MCM_CONFIG__PKG_TYPE__SHIFT 0x2 +#define SMUIO_MCM_CONFIG__SOCKET_ID__SHIFT 0x8 +#define SMUIO_MCM_CONFIG__PKG_SUBTYPE__SHIFT 0xc +#define SMUIO_MCM_CONFIG__CONSOLE_K__SHIFT 0x10 +#define SMUIO_MCM_CONFIG__CONSOLE_A__SHIFT 0x11 +#define SMUIO_MCM_CONFIG__TOPOLOGY_ID__SHIFT 0x12 +#define SMUIO_MCM_CONFIG__DIE_ID_MASK 0x00000003L +#define SMUIO_MCM_CONFIG__PKG_TYPE_MASK 0x0000003CL +#define SMUIO_MCM_CONFIG__SOCKET_ID_MASK 0x00000F00L +#define SMUIO_MCM_CONFIG__PKG_SUBTYPE_MASK 0x00001000L +#define SMUIO_MCM_CONFIG__CONSOLE_K_MASK 0x00010000L +#define SMUIO_MCM_CONFIG__CONSOLE_A_MASK 0x00020000L +#define SMUIO_MCM_CONFIG__TOPOLOGY_ID_MASK 0x007C0000L +//IP_DISCOVERY_VERSION +#define IP_DISCOVERY_VERSION__IP_DISCOVERY_VERSION__SHIFT 0x0 +#define IP_DISCOVERY_VERSION__IP_DISCOVERY_VERSION_MASK 0xFFFFFFFFL +//SCRATCH_REGISTER0 +#define SCRATCH_REGISTER0__ScratchPad0__SHIFT 0x0 +#define SCRATCH_REGISTER0__ScratchPad0_MASK 0xFFFFFFFFL +//SCRATCH_REGISTER1 +#define SCRATCH_REGISTER1__ScratchPad1__SHIFT 0x0 +#define SCRATCH_REGISTER1__ScratchPad1_MASK 0xFFFFFFFFL +//SCRATCH_REGISTER2 +#define SCRATCH_REGISTER2__ScratchPad2__SHIFT 0x0 +#define SCRATCH_REGISTER2__ScratchPad2_MASK 0xFFFFFFFFL +//SCRATCH_REGISTER3 +#define SCRATCH_REGISTER3__ScratchPad3__SHIFT 0x0 +#define SCRATCH_REGISTER3__ScratchPad3_MASK 0xFFFFFFFFL +//SCRATCH_REGISTER4 +#define SCRATCH_REGISTER4__ScratchPad4__SHIFT 0x0 +#define SCRATCH_REGISTER4__ScratchPad4_MASK 0xFFFFFFFFL +//SCRATCH_REGISTER5 +#define SCRATCH_REGISTER5__ScratchPad5__SHIFT 0x0 +#define SCRATCH_REGISTER5__ScratchPad5_MASK 0xFFFFFFFFL +//SCRATCH_REGISTER6 +#define SCRATCH_REGISTER6__ScratchPad6__SHIFT 0x0 +#define SCRATCH_REGISTER6__ScratchPad6_MASK 0xFFFFFFFFL +//SCRATCH_REGISTER7 +#define SCRATCH_REGISTER7__ScratchPad7__SHIFT 0x0 +#define SCRATCH_REGISTER7__ScratchPad7_MASK 0xFFFFFFFFL + + +// addressBlock: aid_smuio_smuio_gpio_SmuSmuioDec +//SMU_GPIOPAD_SW_INT_STAT +#define SMU_GPIOPAD_SW_INT_STAT__SW_INT_STAT__SHIFT 0x0 +#define SMU_GPIOPAD_SW_INT_STAT__SW_INT_STAT_MASK 0x00000001L +//SMU_GPIOPAD_MASK +#define SMU_GPIOPAD_MASK__GPIO_MASK__SHIFT 0x0 +#define SMU_GPIOPAD_MASK__GPIO_MASK_MASK 0x7FFFFFFFL +//SMU_GPIOPAD_A +#define SMU_GPIOPAD_A__GPIO_A__SHIFT 0x0 +#define SMU_GPIOPAD_A__GPIO_A_MASK 0x7FFFFFFFL +//SMU_GPIOPAD_TXIMPSEL +#define SMU_GPIOPAD_TXIMPSEL__GPIO_TXIMPSEL__SHIFT 0x0 +#define SMU_GPIOPAD_TXIMPSEL__GPIO_TXIMPSEL_MASK 0x7FFFFFFFL +//SMU_GPIOPAD_EN +#define SMU_GPIOPAD_EN__GPIO_EN__SHIFT 0x0 +#define SMU_GPIOPAD_EN__GPIO_EN_MASK 0x7FFFFFFFL +//SMU_GPIOPAD_Y +#define SMU_GPIOPAD_Y__GPIO_Y__SHIFT 0x0 +#define SMU_GPIOPAD_Y__GPIO_Y_MASK 0x7FFFFFFFL +//SMU_GPIOPAD_RXEN +#define SMU_GPIOPAD_RXEN__GPIO_RXEN__SHIFT 0x0 +#define SMU_GPIOPAD_RXEN__GPIO_RXEN_MASK 0x7FFFFFFFL +//SMU_GPIOPAD_RCVR_SEL0 +#define SMU_GPIOPAD_RCVR_SEL0__GPIO_RCVR_SEL0__SHIFT 0x0 +#define SMU_GPIOPAD_RCVR_SEL0__GPIO_RCVR_SEL0_MASK 0x7FFFFFFFL +//SMU_GPIOPAD_RCVR_SEL1 +#define SMU_GPIOPAD_RCVR_SEL1__GPIO_RCVR_SEL1__SHIFT 0x0 +#define SMU_GPIOPAD_RCVR_SEL1__GPIO_RCVR_SEL1_MASK 0x7FFFFFFFL +//SMU_GPIOPAD_PU_EN +#define SMU_GPIOPAD_PU_EN__GPIO_PU_EN__SHIFT 0x0 +#define SMU_GPIOPAD_PU_EN__GPIO_PU_EN_MASK 0x7FFFFFFFL +//SMU_GPIOPAD_PD_EN +#define SMU_GPIOPAD_PD_EN__GPIO_PD_EN__SHIFT 0x0 +#define SMU_GPIOPAD_PD_EN__GPIO_PD_EN_MASK 0x7FFFFFFFL +//SMU_GPIOPAD_PINSTRAPS +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_0__SHIFT 0x0 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_1__SHIFT 0x1 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_2__SHIFT 0x2 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_3__SHIFT 0x3 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_4__SHIFT 0x4 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_5__SHIFT 0x5 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_6__SHIFT 0x6 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_7__SHIFT 0x7 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_8__SHIFT 0x8 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_9__SHIFT 0x9 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_10__SHIFT 0xa +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_11__SHIFT 0xb +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_12__SHIFT 0xc +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_13__SHIFT 0xd +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_14__SHIFT 0xe +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_15__SHIFT 0xf +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_16__SHIFT 0x10 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_17__SHIFT 0x11 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_18__SHIFT 0x12 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_19__SHIFT 0x13 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_20__SHIFT 0x14 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_21__SHIFT 0x15 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_22__SHIFT 0x16 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_23__SHIFT 0x17 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_24__SHIFT 0x18 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_25__SHIFT 0x19 +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_26__SHIFT 0x1a +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_27__SHIFT 0x1b +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_28__SHIFT 0x1c +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_29__SHIFT 0x1d +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_30__SHIFT 0x1e +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_0_MASK 0x00000001L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_1_MASK 0x00000002L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_2_MASK 0x00000004L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_3_MASK 0x00000008L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_4_MASK 0x00000010L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_5_MASK 0x00000020L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_6_MASK 0x00000040L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_7_MASK 0x00000080L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_8_MASK 0x00000100L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_9_MASK 0x00000200L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_10_MASK 0x00000400L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_11_MASK 0x00000800L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_12_MASK 0x00001000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_13_MASK 0x00002000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_14_MASK 0x00004000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_15_MASK 0x00008000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_16_MASK 0x00010000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_17_MASK 0x00020000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_18_MASK 0x00040000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_19_MASK 0x00080000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_20_MASK 0x00100000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_21_MASK 0x00200000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_22_MASK 0x00400000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_23_MASK 0x00800000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_24_MASK 0x01000000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_25_MASK 0x02000000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_26_MASK 0x04000000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_27_MASK 0x08000000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_28_MASK 0x10000000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_29_MASK 0x20000000L +#define SMU_GPIOPAD_PINSTRAPS__GPIO_PINSTRAP_30_MASK 0x40000000L +//DFT_PINSTRAPS +#define DFT_PINSTRAPS__DFT_PINSTRAPS__SHIFT 0x0 +#define DFT_PINSTRAPS__DFT_PINSTRAPS_MASK 0x000003FFL +//SMU_GPIOPAD_INT_STAT_EN +#define SMU_GPIOPAD_INT_STAT_EN__GPIO_INT_STAT_EN__SHIFT 0x0 +#define SMU_GPIOPAD_INT_STAT_EN__SW_INITIATED_INT_STAT_EN__SHIFT 0x1f +#define SMU_GPIOPAD_INT_STAT_EN__GPIO_INT_STAT_EN_MASK 0x1FFFFFFFL +#define SMU_GPIOPAD_INT_STAT_EN__SW_INITIATED_INT_STAT_EN_MASK 0x80000000L +//SMU_GPIOPAD_INT_STAT +#define SMU_GPIOPAD_INT_STAT__GPIO_INT_STAT__SHIFT 0x0 +#define SMU_GPIOPAD_INT_STAT__SW_INITIATED_INT_STAT__SHIFT 0x1f +#define SMU_GPIOPAD_INT_STAT__GPIO_INT_STAT_MASK 0x1FFFFFFFL +#define SMU_GPIOPAD_INT_STAT__SW_INITIATED_INT_STAT_MASK 0x80000000L +//SMU_GPIOPAD_INT_STAT_AK +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_0__SHIFT 0x0 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_1__SHIFT 0x1 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_2__SHIFT 0x2 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_3__SHIFT 0x3 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_4__SHIFT 0x4 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_5__SHIFT 0x5 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_6__SHIFT 0x6 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_7__SHIFT 0x7 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_8__SHIFT 0x8 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_9__SHIFT 0x9 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_10__SHIFT 0xa +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_11__SHIFT 0xb +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_12__SHIFT 0xc +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_13__SHIFT 0xd +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_14__SHIFT 0xe +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_15__SHIFT 0xf +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_16__SHIFT 0x10 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_17__SHIFT 0x11 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_18__SHIFT 0x12 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_19__SHIFT 0x13 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_20__SHIFT 0x14 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_21__SHIFT 0x15 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_22__SHIFT 0x16 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_23__SHIFT 0x17 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_24__SHIFT 0x18 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_25__SHIFT 0x19 +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_26__SHIFT 0x1a +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_27__SHIFT 0x1b +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_28__SHIFT 0x1c +#define SMU_GPIOPAD_INT_STAT_AK__SW_INITIATED_INT_STAT_AK__SHIFT 0x1f +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_0_MASK 0x00000001L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_1_MASK 0x00000002L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_2_MASK 0x00000004L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_3_MASK 0x00000008L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_4_MASK 0x00000010L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_5_MASK 0x00000020L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_6_MASK 0x00000040L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_7_MASK 0x00000080L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_8_MASK 0x00000100L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_9_MASK 0x00000200L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_10_MASK 0x00000400L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_11_MASK 0x00000800L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_12_MASK 0x00001000L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_13_MASK 0x00002000L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_14_MASK 0x00004000L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_15_MASK 0x00008000L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_16_MASK 0x00010000L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_17_MASK 0x00020000L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_18_MASK 0x00040000L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_19_MASK 0x00080000L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_20_MASK 0x00100000L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_21_MASK 0x00200000L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_22_MASK 0x00400000L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_23_MASK 0x00800000L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_24_MASK 0x01000000L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_25_MASK 0x02000000L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_26_MASK 0x04000000L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_27_MASK 0x08000000L +#define SMU_GPIOPAD_INT_STAT_AK__GPIO_INT_STAT_AK_28_MASK 0x10000000L +#define SMU_GPIOPAD_INT_STAT_AK__SW_INITIATED_INT_STAT_AK_MASK 0x80000000L +//SMU_GPIOPAD_INT_EN +#define SMU_GPIOPAD_INT_EN__GPIO_INT_EN__SHIFT 0x0 +#define SMU_GPIOPAD_INT_EN__SW_INITIATED_INT_EN__SHIFT 0x1f +#define SMU_GPIOPAD_INT_EN__GPIO_INT_EN_MASK 0x1FFFFFFFL +#define SMU_GPIOPAD_INT_EN__SW_INITIATED_INT_EN_MASK 0x80000000L +//SMU_GPIOPAD_INT_TYPE +#define SMU_GPIOPAD_INT_TYPE__GPIO_INT_TYPE__SHIFT 0x0 +#define SMU_GPIOPAD_INT_TYPE__SW_INITIATED_INT_TYPE__SHIFT 0x1f +#define SMU_GPIOPAD_INT_TYPE__GPIO_INT_TYPE_MASK 0x1FFFFFFFL +#define SMU_GPIOPAD_INT_TYPE__SW_INITIATED_INT_TYPE_MASK 0x80000000L +//SMU_GPIOPAD_INT_POLARITY +#define SMU_GPIOPAD_INT_POLARITY__GPIO_INT_POLARITY__SHIFT 0x0 +#define SMU_GPIOPAD_INT_POLARITY__SW_INITIATED_INT_POLARITY__SHIFT 0x1f +#define SMU_GPIOPAD_INT_POLARITY__GPIO_INT_POLARITY_MASK 0x1FFFFFFFL +#define SMU_GPIOPAD_INT_POLARITY__SW_INITIATED_INT_POLARITY_MASK 0x80000000L +//SMUIO_PCC_GPIO_SELECT +#define SMUIO_PCC_GPIO_SELECT__GPIO__SHIFT 0x0 +#define SMUIO_PCC_GPIO_SELECT__GPIO_MASK 0xFFFFFFFFL +//SMU_GPIOPAD_S0 +#define SMU_GPIOPAD_S0__GPIO_S0__SHIFT 0x0 +#define SMU_GPIOPAD_S0__GPIO_S0_MASK 0x7FFFFFFFL +//SMU_GPIOPAD_S1 +#define SMU_GPIOPAD_S1__GPIO_S1__SHIFT 0x0 +#define SMU_GPIOPAD_S1__GPIO_S1_MASK 0x7FFFFFFFL +//SMU_GPIOPAD_SCHMEN +#define SMU_GPIOPAD_SCHMEN__GPIO_SCHMEN__SHIFT 0x0 +#define SMU_GPIOPAD_SCHMEN__GPIO_SCHMEN_MASK 0x7FFFFFFFL +//SMU_GPIOPAD_SCL_EN +#define SMU_GPIOPAD_SCL_EN__GPIO_SCL_EN__SHIFT 0x0 +#define SMU_GPIOPAD_SCL_EN__GPIO_SCL_EN_MASK 0x7FFFFFFFL +//SMU_GPIOPAD_SDA_EN +#define SMU_GPIOPAD_SDA_EN__GPIO_SDA_EN__SHIFT 0x0 +#define SMU_GPIOPAD_SDA_EN__GPIO_SDA_EN_MASK 0x7FFFFFFFL +//SMUIO_GPIO_INT0_SELECT +#define SMUIO_GPIO_INT0_SELECT__GPIO_INT0_SELECT__SHIFT 0x0 +#define SMUIO_GPIO_INT0_SELECT__GPIO_INT0_SELECT_MASK 0xFFFFFFFFL +//SMUIO_GPIO_INT1_SELECT +#define SMUIO_GPIO_INT1_SELECT__GPIO_INT1_SELECT__SHIFT 0x0 +#define SMUIO_GPIO_INT1_SELECT__GPIO_INT1_SELECT_MASK 0xFFFFFFFFL +//SMUIO_GPIO_INT2_SELECT +#define SMUIO_GPIO_INT2_SELECT__GPIO_INT2_SELECT__SHIFT 0x0 +#define SMUIO_GPIO_INT2_SELECT__GPIO_INT2_SELECT_MASK 0xFFFFFFFFL +//SMUIO_GPIO_INT3_SELECT +#define SMUIO_GPIO_INT3_SELECT__GPIO_INT3_SELECT__SHIFT 0x0 +#define SMUIO_GPIO_INT3_SELECT__GPIO_INT3_SELECT_MASK 0xFFFFFFFFL +//SMU_GPIOPAD_MP_INT0_STAT +#define SMU_GPIOPAD_MP_INT0_STAT__GPIO_MP_INT0_STAT__SHIFT 0x0 +#define SMU_GPIOPAD_MP_INT0_STAT__GPIO_MP_INT0_STAT_MASK 0x1FFFFFFFL +//SMU_GPIOPAD_MP_INT1_STAT +#define SMU_GPIOPAD_MP_INT1_STAT__GPIO_MP_INT1_STAT__SHIFT 0x0 +#define SMU_GPIOPAD_MP_INT1_STAT__GPIO_MP_INT1_STAT_MASK 0x1FFFFFFFL +//SMU_GPIOPAD_MP_INT2_STAT +#define SMU_GPIOPAD_MP_INT2_STAT__GPIO_MP_INT2_STAT__SHIFT 0x0 +#define SMU_GPIOPAD_MP_INT2_STAT__GPIO_MP_INT2_STAT_MASK 0x1FFFFFFFL +//SMU_GPIOPAD_MP_INT3_STAT +#define SMU_GPIOPAD_MP_INT3_STAT__GPIO_MP_INT3_STAT__SHIFT 0x0 +#define SMU_GPIOPAD_MP_INT3_STAT__GPIO_MP_INT3_STAT_MASK 0x1FFFFFFFL +//SMIO_INDEX +#define SMIO_INDEX__SW_SMIO_INDEX__SHIFT 0x0 +#define SMIO_INDEX__SW_SMIO_INDEX_MASK 0x00000001L +//S0_VID_SMIO_CNTL +#define S0_VID_SMIO_CNTL__S0_SMIO_VALUES__SHIFT 0x0 +#define S0_VID_SMIO_CNTL__S0_SMIO_VALUES_MASK 0xFFFFFFFFL +//S1_VID_SMIO_CNTL +#define S1_VID_SMIO_CNTL__S1_SMIO_VALUES__SHIFT 0x0 +#define S1_VID_SMIO_CNTL__S1_SMIO_VALUES_MASK 0xFFFFFFFFL +//OPEN_DRAIN_SELECT +#define OPEN_DRAIN_SELECT__OPEN_DRAIN_SELECT__SHIFT 0x0 +#define OPEN_DRAIN_SELECT__RESERVED__SHIFT 0x1f +#define OPEN_DRAIN_SELECT__OPEN_DRAIN_SELECT_MASK 0x7FFFFFFFL +#define OPEN_DRAIN_SELECT__RESERVED_MASK 0x80000000L +//SMIO_ENABLE +#define SMIO_ENABLE__SMIO_ENABLE__SHIFT 0x0 +#define SMIO_ENABLE__SMIO_ENABLE_MASK 0xFFFFFFFFL + +#endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_offset.h b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_offset.h new file mode 100644 index 0000000000000000000000000000000000000000..e9742d10de1c6c411e4e66b5e7f6851841bbd389 --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_offset.h @@ -0,0 +1,2332 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +#ifndef _vcn_4_0_3_OFFSET_HEADER +#define _vcn_4_0_3_OFFSET_HEADER + + + +// addressBlock: aid_uvd0_uvddec +// base address: 0x1fb00 +#define regUVD_TOP_CTRL 0x00c0 +#define regUVD_TOP_CTRL_BASE_IDX 1 +#define regUVD_CGC_GATE 0x00c1 +#define regUVD_CGC_GATE_BASE_IDX 1 +#define regUVD_CGC_CTRL 0x00c2 +#define regUVD_CGC_CTRL_BASE_IDX 1 +#define regAVM_SUVD_CGC_GATE 0x00c4 +#define regAVM_SUVD_CGC_GATE_BASE_IDX 1 +#define regCDEFE_SUVD_CGC_GATE 0x00c4 +#define regCDEFE_SUVD_CGC_GATE_BASE_IDX 1 +#define regEFC_SUVD_CGC_GATE 0x00c4 +#define regEFC_SUVD_CGC_GATE_BASE_IDX 1 +#define regENT_SUVD_CGC_GATE 0x00c4 +#define regENT_SUVD_CGC_GATE_BASE_IDX 1 +#define regIME_SUVD_CGC_GATE 0x00c4 +#define regIME_SUVD_CGC_GATE_BASE_IDX 1 +#define regPPU_SUVD_CGC_GATE 0x00c4 +#define regPPU_SUVD_CGC_GATE_BASE_IDX 1 +#define regSAOE_SUVD_CGC_GATE 0x00c4 +#define regSAOE_SUVD_CGC_GATE_BASE_IDX 1 +#define regSCM_SUVD_CGC_GATE 0x00c4 +#define regSCM_SUVD_CGC_GATE_BASE_IDX 1 +#define regSDB_SUVD_CGC_GATE 0x00c4 +#define regSDB_SUVD_CGC_GATE_BASE_IDX 1 +#define regSIT0_NXT_SUVD_CGC_GATE 0x00c4 +#define regSIT0_NXT_SUVD_CGC_GATE_BASE_IDX 1 +#define regSIT1_NXT_SUVD_CGC_GATE 0x00c4 +#define regSIT1_NXT_SUVD_CGC_GATE_BASE_IDX 1 +#define regSIT2_NXT_SUVD_CGC_GATE 0x00c4 +#define regSIT2_NXT_SUVD_CGC_GATE_BASE_IDX 1 +#define regSIT_SUVD_CGC_GATE 0x00c4 +#define regSIT_SUVD_CGC_GATE_BASE_IDX 1 +#define regSMPA_SUVD_CGC_GATE 0x00c4 +#define regSMPA_SUVD_CGC_GATE_BASE_IDX 1 +#define regSMP_SUVD_CGC_GATE 0x00c4 +#define regSMP_SUVD_CGC_GATE_BASE_IDX 1 +#define regSRE_SUVD_CGC_GATE 0x00c4 +#define regSRE_SUVD_CGC_GATE_BASE_IDX 1 +#define regUVD_MPBE0_SUVD_CGC_GATE 0x00c4 +#define regUVD_MPBE0_SUVD_CGC_GATE_BASE_IDX 1 +#define regUVD_MPBE1_SUVD_CGC_GATE 0x00c4 +#define regUVD_MPBE1_SUVD_CGC_GATE_BASE_IDX 1 +#define regUVD_SUVD_CGC_GATE 0x00c4 +#define regUVD_SUVD_CGC_GATE_BASE_IDX 1 +#define regAVM_SUVD_CGC_GATE2 0x00c5 +#define regAVM_SUVD_CGC_GATE2_BASE_IDX 1 +#define regCDEFE_SUVD_CGC_GATE2 0x00c5 +#define regCDEFE_SUVD_CGC_GATE2_BASE_IDX 1 +#define regDBR_SUVD_CGC_GATE2 0x00c5 +#define regDBR_SUVD_CGC_GATE2_BASE_IDX 1 +#define regENT_SUVD_CGC_GATE2 0x00c5 +#define regENT_SUVD_CGC_GATE2_BASE_IDX 1 +#define regIME_SUVD_CGC_GATE2 0x00c5 +#define regIME_SUVD_CGC_GATE2_BASE_IDX 1 +#define regMPC1_SUVD_CGC_GATE2 0x00c5 +#define regMPC1_SUVD_CGC_GATE2_BASE_IDX 1 +#define regSAOE_SUVD_CGC_GATE2 0x00c5 +#define regSAOE_SUVD_CGC_GATE2_BASE_IDX 1 +#define regSDB_SUVD_CGC_GATE2 0x00c5 +#define regSDB_SUVD_CGC_GATE2_BASE_IDX 1 +#define regSIT0_NXT_SUVD_CGC_GATE2 0x00c5 +#define regSIT0_NXT_SUVD_CGC_GATE2_BASE_IDX 1 +#define regSIT1_NXT_SUVD_CGC_GATE2 0x00c5 +#define regSIT1_NXT_SUVD_CGC_GATE2_BASE_IDX 1 +#define regSIT2_NXT_SUVD_CGC_GATE2 0x00c5 +#define regSIT2_NXT_SUVD_CGC_GATE2_BASE_IDX 1 +#define regSIT_SUVD_CGC_GATE2 0x00c5 +#define regSIT_SUVD_CGC_GATE2_BASE_IDX 1 +#define regSMPA_SUVD_CGC_GATE2 0x00c5 +#define regSMPA_SUVD_CGC_GATE2_BASE_IDX 1 +#define regSMP_SUVD_CGC_GATE2 0x00c5 +#define regSMP_SUVD_CGC_GATE2_BASE_IDX 1 +#define regSRE_SUVD_CGC_GATE2 0x00c5 +#define regSRE_SUVD_CGC_GATE2_BASE_IDX 1 +#define regUVD_MPBE0_SUVD_CGC_GATE2 0x00c5 +#define regUVD_MPBE0_SUVD_CGC_GATE2_BASE_IDX 1 +#define regUVD_MPBE1_SUVD_CGC_GATE2 0x00c5 +#define regUVD_MPBE1_SUVD_CGC_GATE2_BASE_IDX 1 +#define regUVD_SUVD_CGC_GATE2 0x00c5 +#define regUVD_SUVD_CGC_GATE2_BASE_IDX 1 +#define regAVM_SUVD_CGC_CTRL 0x00c6 +#define regAVM_SUVD_CGC_CTRL_BASE_IDX 1 +#define regCDEFE_SUVD_CGC_CTRL 0x00c6 +#define regCDEFE_SUVD_CGC_CTRL_BASE_IDX 1 +#define regDBR_SUVD_CGC_CTRL 0x00c6 +#define regDBR_SUVD_CGC_CTRL_BASE_IDX 1 +#define regEFC_SUVD_CGC_CTRL 0x00c6 +#define regEFC_SUVD_CGC_CTRL_BASE_IDX 1 +#define regENT_SUVD_CGC_CTRL 0x00c6 +#define regENT_SUVD_CGC_CTRL_BASE_IDX 1 +#define regIME_SUVD_CGC_CTRL 0x00c6 +#define regIME_SUVD_CGC_CTRL_BASE_IDX 1 +#define regMPC1_SUVD_CGC_CTRL 0x00c6 +#define regMPC1_SUVD_CGC_CTRL_BASE_IDX 1 +#define regPPU_SUVD_CGC_CTRL 0x00c6 +#define regPPU_SUVD_CGC_CTRL_BASE_IDX 1 +#define regSAOE_SUVD_CGC_CTRL 0x00c6 +#define regSAOE_SUVD_CGC_CTRL_BASE_IDX 1 +#define regSCM_SUVD_CGC_CTRL 0x00c6 +#define regSCM_SUVD_CGC_CTRL_BASE_IDX 1 +#define regSDB_SUVD_CGC_CTRL 0x00c6 +#define regSDB_SUVD_CGC_CTRL_BASE_IDX 1 +#define regSIT0_NXT_SUVD_CGC_CTRL 0x00c6 +#define regSIT0_NXT_SUVD_CGC_CTRL_BASE_IDX 1 +#define regSIT1_NXT_SUVD_CGC_CTRL 0x00c6 +#define regSIT1_NXT_SUVD_CGC_CTRL_BASE_IDX 1 +#define regSIT2_NXT_SUVD_CGC_CTRL 0x00c6 +#define regSIT2_NXT_SUVD_CGC_CTRL_BASE_IDX 1 +#define regSIT_SUVD_CGC_CTRL 0x00c6 +#define regSIT_SUVD_CGC_CTRL_BASE_IDX 1 +#define regSMPA_SUVD_CGC_CTRL 0x00c6 +#define regSMPA_SUVD_CGC_CTRL_BASE_IDX 1 +#define regSMP_SUVD_CGC_CTRL 0x00c6 +#define regSMP_SUVD_CGC_CTRL_BASE_IDX 1 +#define regSRE_SUVD_CGC_CTRL 0x00c6 +#define regSRE_SUVD_CGC_CTRL_BASE_IDX 1 +#define regUVD_MPBE0_SUVD_CGC_CTRL 0x00c6 +#define regUVD_MPBE0_SUVD_CGC_CTRL_BASE_IDX 1 +#define regUVD_MPBE1_SUVD_CGC_CTRL 0x00c6 +#define regUVD_MPBE1_SUVD_CGC_CTRL_BASE_IDX 1 +#define regUVD_SUVD_CGC_CTRL 0x00c6 +#define regUVD_SUVD_CGC_CTRL_BASE_IDX 1 +#define regUVD_CGC_CTRL3 0x00ca +#define regUVD_CGC_CTRL3_BASE_IDX 1 +#define regUVD_GPCOM_VCPU_DATA0 0x00d0 +#define regUVD_GPCOM_VCPU_DATA0_BASE_IDX 1 +#define regUVD_GPCOM_VCPU_DATA1 0x00d1 +#define regUVD_GPCOM_VCPU_DATA1_BASE_IDX 1 +#define regUVD_GPCOM_SYS_CMD 0x00d2 +#define regUVD_GPCOM_SYS_CMD_BASE_IDX 1 +#define regUVD_GPCOM_SYS_DATA0 0x00d3 +#define regUVD_GPCOM_SYS_DATA0_BASE_IDX 1 +#define regUVD_GPCOM_SYS_DATA1 0x00d4 +#define regUVD_GPCOM_SYS_DATA1_BASE_IDX 1 +#define regUVD_VCPU_INT_EN 0x00d5 +#define regUVD_VCPU_INT_EN_BASE_IDX 1 +#define regUVD_VCPU_INT_STATUS 0x00d6 +#define regUVD_VCPU_INT_STATUS_BASE_IDX 1 +#define regUVD_VCPU_INT_ACK 0x00d7 +#define regUVD_VCPU_INT_ACK_BASE_IDX 1 +#define regUVD_VCPU_INT_ROUTE 0x00d8 +#define regUVD_VCPU_INT_ROUTE_BASE_IDX 1 +#define regUVD_DRV_FW_MSG 0x00d9 +#define regUVD_DRV_FW_MSG_BASE_IDX 1 +#define regUVD_FW_DRV_MSG_ACK 0x00da +#define regUVD_FW_DRV_MSG_ACK_BASE_IDX 1 +#define regUVD_SUVD_INT_EN 0x00db +#define regUVD_SUVD_INT_EN_BASE_IDX 1 +#define regUVD_SUVD_INT_STATUS 0x00dc +#define regUVD_SUVD_INT_STATUS_BASE_IDX 1 +#define regUVD_SUVD_INT_ACK 0x00dd +#define regUVD_SUVD_INT_ACK_BASE_IDX 1 +#define regUVD_ENC_VCPU_INT_EN 0x00de +#define regUVD_ENC_VCPU_INT_EN_BASE_IDX 1 +#define regUVD_ENC_VCPU_INT_STATUS 0x00df +#define regUVD_ENC_VCPU_INT_STATUS_BASE_IDX 1 +#define regUVD_ENC_VCPU_INT_ACK 0x00e0 +#define regUVD_ENC_VCPU_INT_ACK_BASE_IDX 1 +#define regUVD_MASTINT_EN 0x00e1 +#define regUVD_MASTINT_EN_BASE_IDX 1 +#define regUVD_SYS_INT_EN 0x00e2 +#define regUVD_SYS_INT_EN_BASE_IDX 1 +#define regUVD_SYS_INT_STATUS 0x00e3 +#define regUVD_SYS_INT_STATUS_BASE_IDX 1 +#define regUVD_SYS_INT_ACK 0x00e4 +#define regUVD_SYS_INT_ACK_BASE_IDX 1 +#define regUVD_JOB_DONE 0x00e5 +#define regUVD_JOB_DONE_BASE_IDX 1 +#define regUVD_CBUF_ID 0x00e6 +#define regUVD_CBUF_ID_BASE_IDX 1 +#define regUVD_CONTEXT_ID 0x00e7 +#define regUVD_CONTEXT_ID_BASE_IDX 1 +#define regUVD_CONTEXT_ID2 0x00e8 +#define regUVD_CONTEXT_ID2_BASE_IDX 1 +#define regUVD_NO_OP 0x00e9 +#define regUVD_NO_OP_BASE_IDX 1 +#define regUVD_RB_BASE_LO 0x00ea +#define regUVD_RB_BASE_LO_BASE_IDX 1 +#define regUVD_RB_BASE_HI 0x00eb +#define regUVD_RB_BASE_HI_BASE_IDX 1 +#define regUVD_RB_SIZE 0x00ec +#define regUVD_RB_SIZE_BASE_IDX 1 +#define regUVD_RB_BASE_LO2 0x00ef +#define regUVD_RB_BASE_LO2_BASE_IDX 1 +#define regUVD_RB_BASE_HI2 0x00f0 +#define regUVD_RB_BASE_HI2_BASE_IDX 1 +#define regUVD_RB_SIZE2 0x00f1 +#define regUVD_RB_SIZE2_BASE_IDX 1 +#define regUVD_RB_BASE_LO3 0x00f4 +#define regUVD_RB_BASE_LO3_BASE_IDX 1 +#define regUVD_RB_BASE_HI3 0x00f5 +#define regUVD_RB_BASE_HI3_BASE_IDX 1 +#define regUVD_RB_SIZE3 0x00f6 +#define regUVD_RB_SIZE3_BASE_IDX 1 +#define regUVD_RB_BASE_LO4 0x00f9 +#define regUVD_RB_BASE_LO4_BASE_IDX 1 +#define regUVD_RB_BASE_HI4 0x00fa +#define regUVD_RB_BASE_HI4_BASE_IDX 1 +#define regUVD_RB_SIZE4 0x00fb +#define regUVD_RB_SIZE4_BASE_IDX 1 +#define regUVD_OUT_RB_BASE_LO 0x00fe +#define regUVD_OUT_RB_BASE_LO_BASE_IDX 1 +#define regUVD_OUT_RB_BASE_HI 0x00ff +#define regUVD_OUT_RB_BASE_HI_BASE_IDX 1 +#define regUVD_OUT_RB_SIZE 0x0100 +#define regUVD_OUT_RB_SIZE_BASE_IDX 1 +#define regUVD_IOV_ACTIVE_FCN_ID 0x0103 +#define regUVD_IOV_ACTIVE_FCN_ID_BASE_IDX 1 +#define regUVD_IOV_MAILBOX 0x0104 +#define regUVD_IOV_MAILBOX_BASE_IDX 1 +#define regUVD_IOV_MAILBOX_RESP 0x0105 +#define regUVD_IOV_MAILBOX_RESP_BASE_IDX 1 +#define regUVD_RB_ARB_CTRL 0x0106 +#define regUVD_RB_ARB_CTRL_BASE_IDX 1 +#define regUVD_CTX_INDEX 0x0107 +#define regUVD_CTX_INDEX_BASE_IDX 1 +#define regUVD_CTX_DATA 0x0108 +#define regUVD_CTX_DATA_BASE_IDX 1 +#define regUVD_CXW_WR 0x0109 +#define regUVD_CXW_WR_BASE_IDX 1 +#define regUVD_CXW_WR_INT_ID 0x010a +#define regUVD_CXW_WR_INT_ID_BASE_IDX 1 +#define regUVD_CXW_WR_INT_CTX_ID 0x010b +#define regUVD_CXW_WR_INT_CTX_ID_BASE_IDX 1 +#define regUVD_CXW_INT_ID 0x010c +#define regUVD_CXW_INT_ID_BASE_IDX 1 +#define regUVD_MPEG2_ERROR 0x010d +#define regUVD_MPEG2_ERROR_BASE_IDX 1 +#define regUVD_YBASE 0x0110 +#define regUVD_YBASE_BASE_IDX 1 +#define regUVD_UVBASE 0x0111 +#define regUVD_UVBASE_BASE_IDX 1 +#define regUVD_PITCH 0x0112 +#define regUVD_PITCH_BASE_IDX 1 +#define regUVD_WIDTH 0x0113 +#define regUVD_WIDTH_BASE_IDX 1 +#define regUVD_HEIGHT 0x0114 +#define regUVD_HEIGHT_BASE_IDX 1 +#define regUVD_PICCOUNT 0x0115 +#define regUVD_PICCOUNT_BASE_IDX 1 +#define regUVD_MPRD_INITIAL_XY 0x0116 +#define regUVD_MPRD_INITIAL_XY_BASE_IDX 1 +#define regUVD_MPEG2_CTRL 0x0117 +#define regUVD_MPEG2_CTRL_BASE_IDX 1 +#define regUVD_MB_CTL_BUF_BASE 0x0118 +#define regUVD_MB_CTL_BUF_BASE_BASE_IDX 1 +#define regUVD_PIC_CTL_BUF_BASE 0x0119 +#define regUVD_PIC_CTL_BUF_BASE_BASE_IDX 1 +#define regUVD_DXVA_BUF_SIZE 0x011a +#define regUVD_DXVA_BUF_SIZE_BASE_IDX 1 +#define regUVD_SCRATCH_NP 0x011b +#define regUVD_SCRATCH_NP_BASE_IDX 1 +#define regUVD_CLK_SWT_HANDSHAKE 0x011c +#define regUVD_CLK_SWT_HANDSHAKE_BASE_IDX 1 +#define regUVD_GP_SCRATCH0 0x011e +#define regUVD_GP_SCRATCH0_BASE_IDX 1 +#define regUVD_GP_SCRATCH1 0x011f +#define regUVD_GP_SCRATCH1_BASE_IDX 1 +#define regUVD_GP_SCRATCH2 0x0120 +#define regUVD_GP_SCRATCH2_BASE_IDX 1 +#define regUVD_GP_SCRATCH3 0x0121 +#define regUVD_GP_SCRATCH3_BASE_IDX 1 +#define regUVD_GP_SCRATCH4 0x0122 +#define regUVD_GP_SCRATCH4_BASE_IDX 1 +#define regUVD_GP_SCRATCH5 0x0123 +#define regUVD_GP_SCRATCH5_BASE_IDX 1 +#define regUVD_GP_SCRATCH6 0x0124 +#define regUVD_GP_SCRATCH6_BASE_IDX 1 +#define regUVD_GP_SCRATCH7 0x0125 +#define regUVD_GP_SCRATCH7_BASE_IDX 1 +#define regUVD_GP_SCRATCH8 0x0126 +#define regUVD_GP_SCRATCH8_BASE_IDX 1 +#define regUVD_GP_SCRATCH9 0x0127 +#define regUVD_GP_SCRATCH9_BASE_IDX 1 +#define regUVD_GP_SCRATCH10 0x0128 +#define regUVD_GP_SCRATCH10_BASE_IDX 1 +#define regUVD_GP_SCRATCH11 0x0129 +#define regUVD_GP_SCRATCH11_BASE_IDX 1 +#define regUVD_GP_SCRATCH12 0x012a +#define regUVD_GP_SCRATCH12_BASE_IDX 1 +#define regUVD_GP_SCRATCH13 0x012b +#define regUVD_GP_SCRATCH13_BASE_IDX 1 +#define regUVD_GP_SCRATCH14 0x012c +#define regUVD_GP_SCRATCH14_BASE_IDX 1 +#define regUVD_GP_SCRATCH15 0x012d +#define regUVD_GP_SCRATCH15_BASE_IDX 1 +#define regUVD_GP_SCRATCH16 0x012e +#define regUVD_GP_SCRATCH16_BASE_IDX 1 +#define regUVD_GP_SCRATCH17 0x012f +#define regUVD_GP_SCRATCH17_BASE_IDX 1 +#define regUVD_GP_SCRATCH18 0x0130 +#define regUVD_GP_SCRATCH18_BASE_IDX 1 +#define regUVD_GP_SCRATCH19 0x0131 +#define regUVD_GP_SCRATCH19_BASE_IDX 1 +#define regUVD_GP_SCRATCH20 0x0132 +#define regUVD_GP_SCRATCH20_BASE_IDX 1 +#define regUVD_GP_SCRATCH21 0x0133 +#define regUVD_GP_SCRATCH21_BASE_IDX 1 +#define regUVD_GP_SCRATCH22 0x0134 +#define regUVD_GP_SCRATCH22_BASE_IDX 1 +#define regUVD_GP_SCRATCH23 0x0135 +#define regUVD_GP_SCRATCH23_BASE_IDX 1 +#define regUVD_AUDIO_RB_BASE_LO 0x0136 +#define regUVD_AUDIO_RB_BASE_LO_BASE_IDX 1 +#define regUVD_AUDIO_RB_BASE_HI 0x0137 +#define regUVD_AUDIO_RB_BASE_HI_BASE_IDX 1 +#define regUVD_AUDIO_RB_SIZE 0x0138 +#define regUVD_AUDIO_RB_SIZE_BASE_IDX 1 +#define regUVD_VCPU_INT_STATUS2 0x013b +#define regUVD_VCPU_INT_STATUS2_BASE_IDX 1 +#define regUVD_VCPU_INT_ACK2 0x013c +#define regUVD_VCPU_INT_ACK2_BASE_IDX 1 +#define regUVD_VCPU_INT_EN2 0x013d +#define regUVD_VCPU_INT_EN2_BASE_IDX 1 +#define regUVD_SUVD_CGC_STATUS2 0x013e +#define regUVD_SUVD_CGC_STATUS2_BASE_IDX 1 +#define regUVD_SUVD_INT_STATUS2 0x0140 +#define regUVD_SUVD_INT_STATUS2_BASE_IDX 1 +#define regUVD_SUVD_INT_EN2 0x0141 +#define regUVD_SUVD_INT_EN2_BASE_IDX 1 +#define regUVD_SUVD_INT_ACK2 0x0142 +#define regUVD_SUVD_INT_ACK2_BASE_IDX 1 +#define regUVD_STATUS 0x0143 +#define regUVD_STATUS_BASE_IDX 1 +#define regUVD_ENC_PIPE_BUSY 0x0144 +#define regUVD_ENC_PIPE_BUSY_BASE_IDX 1 +#define regUVD_FW_POWER_STATUS 0x0145 +#define regUVD_FW_POWER_STATUS_BASE_IDX 1 +#define regUVD_CNTL 0x0146 +#define regUVD_CNTL_BASE_IDX 1 +#define regUVD_SOFT_RESET 0x0147 +#define regUVD_SOFT_RESET_BASE_IDX 1 +#define regUVD_SOFT_RESET2 0x0148 +#define regUVD_SOFT_RESET2_BASE_IDX 1 +#define regUVD_MMSCH_SOFT_RESET 0x0149 +#define regUVD_MMSCH_SOFT_RESET_BASE_IDX 1 +#define regUVD_WIG_CTRL 0x014a +#define regUVD_WIG_CTRL_BASE_IDX 1 +#define regUVD_CGC_STATUS 0x014c +#define regUVD_CGC_STATUS_BASE_IDX 1 +#define regUVD_CGC_UDEC_STATUS 0x014e +#define regUVD_CGC_UDEC_STATUS_BASE_IDX 1 +#define regUVD_SUVD_CGC_STATUS 0x0150 +#define regUVD_SUVD_CGC_STATUS_BASE_IDX 1 +#define regUVD_GPCOM_VCPU_CMD 0x0152 +#define regUVD_GPCOM_VCPU_CMD_BASE_IDX 1 + + +// addressBlock: aid_uvd0_ecpudec +// base address: 0x1fe00 +#define regUVD_VCPU_CACHE_OFFSET0 0x0180 +#define regUVD_VCPU_CACHE_OFFSET0_BASE_IDX 1 +#define regUVD_VCPU_CACHE_SIZE0 0x0181 +#define regUVD_VCPU_CACHE_SIZE0_BASE_IDX 1 +#define regUVD_VCPU_CACHE_OFFSET1 0x0182 +#define regUVD_VCPU_CACHE_OFFSET1_BASE_IDX 1 +#define regUVD_VCPU_CACHE_SIZE1 0x0183 +#define regUVD_VCPU_CACHE_SIZE1_BASE_IDX 1 +#define regUVD_VCPU_CACHE_OFFSET2 0x0184 +#define regUVD_VCPU_CACHE_OFFSET2_BASE_IDX 1 +#define regUVD_VCPU_CACHE_SIZE2 0x0185 +#define regUVD_VCPU_CACHE_SIZE2_BASE_IDX 1 +#define regUVD_VCPU_CACHE_OFFSET3 0x0186 +#define regUVD_VCPU_CACHE_OFFSET3_BASE_IDX 1 +#define regUVD_VCPU_CACHE_SIZE3 0x0187 +#define regUVD_VCPU_CACHE_SIZE3_BASE_IDX 1 +#define regUVD_VCPU_CACHE_OFFSET4 0x0188 +#define regUVD_VCPU_CACHE_OFFSET4_BASE_IDX 1 +#define regUVD_VCPU_CACHE_SIZE4 0x0189 +#define regUVD_VCPU_CACHE_SIZE4_BASE_IDX 1 +#define regUVD_VCPU_CACHE_OFFSET5 0x018a +#define regUVD_VCPU_CACHE_OFFSET5_BASE_IDX 1 +#define regUVD_VCPU_CACHE_SIZE5 0x018b +#define regUVD_VCPU_CACHE_SIZE5_BASE_IDX 1 +#define regUVD_VCPU_CACHE_OFFSET6 0x018c +#define regUVD_VCPU_CACHE_OFFSET6_BASE_IDX 1 +#define regUVD_VCPU_CACHE_SIZE6 0x018d +#define regUVD_VCPU_CACHE_SIZE6_BASE_IDX 1 +#define regUVD_VCPU_CACHE_OFFSET7 0x018e +#define regUVD_VCPU_CACHE_OFFSET7_BASE_IDX 1 +#define regUVD_VCPU_CACHE_SIZE7 0x018f +#define regUVD_VCPU_CACHE_SIZE7_BASE_IDX 1 +#define regUVD_VCPU_CACHE_OFFSET8 0x0190 +#define regUVD_VCPU_CACHE_OFFSET8_BASE_IDX 1 +#define regUVD_VCPU_CACHE_SIZE8 0x0191 +#define regUVD_VCPU_CACHE_SIZE8_BASE_IDX 1 +#define regUVD_VCPU_NONCACHE_OFFSET0 0x0192 +#define regUVD_VCPU_NONCACHE_OFFSET0_BASE_IDX 1 +#define regUVD_VCPU_NONCACHE_SIZE0 0x0193 +#define regUVD_VCPU_NONCACHE_SIZE0_BASE_IDX 1 +#define regUVD_VCPU_NONCACHE_OFFSET1 0x0194 +#define regUVD_VCPU_NONCACHE_OFFSET1_BASE_IDX 1 +#define regUVD_VCPU_NONCACHE_SIZE1 0x0195 +#define regUVD_VCPU_NONCACHE_SIZE1_BASE_IDX 1 +#define regUVD_VCPU_CNTL 0x0196 +#define regUVD_VCPU_CNTL_BASE_IDX 1 +#define regUVD_VCPU_PRID 0x0197 +#define regUVD_VCPU_PRID_BASE_IDX 1 +#define regUVD_VCPU_TRCE 0x0198 +#define regUVD_VCPU_TRCE_BASE_IDX 1 +#define regUVD_VCPU_TRCE_RD 0x0199 +#define regUVD_VCPU_TRCE_RD_BASE_IDX 1 +#define regUVD_VCPU_IND_INDEX 0x019b +#define regUVD_VCPU_IND_INDEX_BASE_IDX 1 +#define regUVD_VCPU_IND_DATA 0x019c +#define regUVD_VCPU_IND_DATA_BASE_IDX 1 + + +// addressBlock: aid_uvd0_uvd_mpcdec +// base address: 0x1ff30 +#define regUVD_MP_SWAP_CNTL 0x01cc +#define regUVD_MP_SWAP_CNTL_BASE_IDX 1 +#define regUVD_MP_SWAP_CNTL2 0x01cd +#define regUVD_MP_SWAP_CNTL2_BASE_IDX 1 +#define regUVD_MPC_LUMA_SRCH 0x01ce +#define regUVD_MPC_LUMA_SRCH_BASE_IDX 1 +#define regUVD_MPC_LUMA_HIT 0x01cf +#define regUVD_MPC_LUMA_HIT_BASE_IDX 1 +#define regUVD_MPC_LUMA_HITPEND 0x01d0 +#define regUVD_MPC_LUMA_HITPEND_BASE_IDX 1 +#define regUVD_MPC_CHROMA_SRCH 0x01d1 +#define regUVD_MPC_CHROMA_SRCH_BASE_IDX 1 +#define regUVD_MPC_CHROMA_HIT 0x01d2 +#define regUVD_MPC_CHROMA_HIT_BASE_IDX 1 +#define regUVD_MPC_CHROMA_HITPEND 0x01d3 +#define regUVD_MPC_CHROMA_HITPEND_BASE_IDX 1 +#define regUVD_MPC_CNTL 0x01d4 +#define regUVD_MPC_CNTL_BASE_IDX 1 +#define regUVD_MPC_PITCH 0x01d5 +#define regUVD_MPC_PITCH_BASE_IDX 1 +#define regUVD_MPC_SET_MUXA0 0x01d6 +#define regUVD_MPC_SET_MUXA0_BASE_IDX 1 +#define regUVD_MPC_SET_MUXA1 0x01d7 +#define regUVD_MPC_SET_MUXA1_BASE_IDX 1 +#define regUVD_MPC_SET_MUXB0 0x01d8 +#define regUVD_MPC_SET_MUXB0_BASE_IDX 1 +#define regUVD_MPC_SET_MUXB1 0x01d9 +#define regUVD_MPC_SET_MUXB1_BASE_IDX 1 +#define regUVD_MPC_SET_MUX 0x01da +#define regUVD_MPC_SET_MUX_BASE_IDX 1 +#define regUVD_MPC_SET_ALU 0x01db +#define regUVD_MPC_SET_ALU_BASE_IDX 1 +#define regUVD_MPC_PERF0 0x01dc +#define regUVD_MPC_PERF0_BASE_IDX 1 +#define regUVD_MPC_PERF1 0x01dd +#define regUVD_MPC_PERF1_BASE_IDX 1 +#define regUVD_MPC_IND_INDEX 0x01de +#define regUVD_MPC_IND_INDEX_BASE_IDX 1 +#define regUVD_MPC_IND_DATA 0x01df +#define regUVD_MPC_IND_DATA_BASE_IDX 1 + + +// addressBlock: aid_uvd0_uvd_rbcdec +// base address: 0x1ff90 +#define regUVD_RBC_IB_SIZE 0x01e4 +#define regUVD_RBC_IB_SIZE_BASE_IDX 1 +#define regUVD_RBC_IB_SIZE_UPDATE 0x01e5 +#define regUVD_RBC_IB_SIZE_UPDATE_BASE_IDX 1 +#define regUVD_RBC_RB_CNTL 0x01e6 +#define regUVD_RBC_RB_CNTL_BASE_IDX 1 +#define regUVD_RBC_RB_RPTR_ADDR 0x01e7 +#define regUVD_RBC_RB_RPTR_ADDR_BASE_IDX 1 +#define regUVD_RBC_VCPU_ACCESS 0x01ea +#define regUVD_RBC_VCPU_ACCESS_BASE_IDX 1 +#define regUVD_FW_SEMAPHORE_CNTL 0x01eb +#define regUVD_FW_SEMAPHORE_CNTL_BASE_IDX 1 +#define regUVD_RBC_READ_REQ_URGENT_CNTL 0x01ed +#define regUVD_RBC_READ_REQ_URGENT_CNTL_BASE_IDX 1 +#define regUVD_RBC_RB_WPTR_CNTL 0x01ee +#define regUVD_RBC_RB_WPTR_CNTL_BASE_IDX 1 +#define regUVD_RBC_WPTR_STATUS 0x01ef +#define regUVD_RBC_WPTR_STATUS_BASE_IDX 1 +#define regUVD_RBC_WPTR_POLL_CNTL 0x01f0 +#define regUVD_RBC_WPTR_POLL_CNTL_BASE_IDX 1 +#define regUVD_RBC_WPTR_POLL_ADDR 0x01f1 +#define regUVD_RBC_WPTR_POLL_ADDR_BASE_IDX 1 +#define regUVD_SEMA_CMD 0x01f2 +#define regUVD_SEMA_CMD_BASE_IDX 1 +#define regUVD_SEMA_ADDR_LOW 0x01f3 +#define regUVD_SEMA_ADDR_LOW_BASE_IDX 1 +#define regUVD_SEMA_ADDR_HIGH 0x01f4 +#define regUVD_SEMA_ADDR_HIGH_BASE_IDX 1 +#define regUVD_ENGINE_CNTL 0x01f5 +#define regUVD_ENGINE_CNTL_BASE_IDX 1 +#define regUVD_SEMA_TIMEOUT_STATUS 0x01f6 +#define regUVD_SEMA_TIMEOUT_STATUS_BASE_IDX 1 +#define regUVD_SEMA_CNTL 0x01f7 +#define regUVD_SEMA_CNTL_BASE_IDX 1 +#define regUVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL 0x01f8 +#define regUVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL_BASE_IDX 1 +#define regUVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL 0x01f9 +#define regUVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL_BASE_IDX 1 +#define regUVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL 0x01fa +#define regUVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL_BASE_IDX 1 +#define regUVD_JOB_START 0x01fb +#define regUVD_JOB_START_BASE_IDX 1 +#define regUVD_RBC_BUF_STATUS 0x01fc +#define regUVD_RBC_BUF_STATUS_BASE_IDX 1 +#define regUVD_RBC_SWAP_CNTL 0x01fd +#define regUVD_RBC_SWAP_CNTL_BASE_IDX 1 + + +// addressBlock: aid_uvd0_lmi_adpdec +// base address: 0x20090 +#define regUVD_LMI_RE_64BIT_BAR_LOW 0x0224 +#define regUVD_LMI_RE_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_RE_64BIT_BAR_HIGH 0x0225 +#define regUVD_LMI_RE_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_IT_64BIT_BAR_LOW 0x0226 +#define regUVD_LMI_IT_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_IT_64BIT_BAR_HIGH 0x0227 +#define regUVD_LMI_IT_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MP_64BIT_BAR_LOW 0x0228 +#define regUVD_LMI_MP_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MP_64BIT_BAR_HIGH 0x0229 +#define regUVD_LMI_MP_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_CM_64BIT_BAR_LOW 0x022a +#define regUVD_LMI_CM_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_CM_64BIT_BAR_HIGH 0x022b +#define regUVD_LMI_CM_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_DB_64BIT_BAR_LOW 0x022c +#define regUVD_LMI_DB_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_DB_64BIT_BAR_HIGH 0x022d +#define regUVD_LMI_DB_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_DBW_64BIT_BAR_LOW 0x022e +#define regUVD_LMI_DBW_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_DBW_64BIT_BAR_HIGH 0x022f +#define regUVD_LMI_DBW_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_IDCT_64BIT_BAR_LOW 0x0230 +#define regUVD_LMI_IDCT_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_IDCT_64BIT_BAR_HIGH 0x0231 +#define regUVD_LMI_IDCT_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MPRD_S0_64BIT_BAR_LOW 0x0232 +#define regUVD_LMI_MPRD_S0_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MPRD_S0_64BIT_BAR_HIGH 0x0233 +#define regUVD_LMI_MPRD_S0_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MPRD_S1_64BIT_BAR_LOW 0x0234 +#define regUVD_LMI_MPRD_S1_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MPRD_S1_64BIT_BAR_HIGH 0x0235 +#define regUVD_LMI_MPRD_S1_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MPRD_DBW_64BIT_BAR_LOW 0x0236 +#define regUVD_LMI_MPRD_DBW_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MPRD_DBW_64BIT_BAR_HIGH 0x0237 +#define regUVD_LMI_MPRD_DBW_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MPC_64BIT_BAR_LOW 0x0238 +#define regUVD_LMI_MPC_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MPC_64BIT_BAR_HIGH 0x0239 +#define regUVD_LMI_MPC_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_RBC_RB_64BIT_BAR_LOW 0x023a +#define regUVD_LMI_RBC_RB_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_RBC_RB_64BIT_BAR_HIGH 0x023b +#define regUVD_LMI_RBC_RB_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_RBC_IB_64BIT_BAR_LOW 0x023c +#define regUVD_LMI_RBC_IB_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_RBC_IB_64BIT_BAR_HIGH 0x023d +#define regUVD_LMI_RBC_IB_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_LBSI_64BIT_BAR_LOW 0x023e +#define regUVD_LMI_LBSI_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_LBSI_64BIT_BAR_HIGH 0x023f +#define regUVD_LMI_LBSI_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_VCPU_NC0_64BIT_BAR_LOW 0x0240 +#define regUVD_LMI_VCPU_NC0_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH 0x0241 +#define regUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_VCPU_NC1_64BIT_BAR_LOW 0x0242 +#define regUVD_LMI_VCPU_NC1_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_VCPU_NC1_64BIT_BAR_HIGH 0x0243 +#define regUVD_LMI_VCPU_NC1_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW 0x0244 +#define regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH 0x0245 +#define regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_CENC_64BIT_BAR_LOW 0x0246 +#define regUVD_LMI_CENC_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_CENC_64BIT_BAR_HIGH 0x0247 +#define regUVD_LMI_CENC_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_SRE_64BIT_BAR_LOW 0x0248 +#define regUVD_LMI_SRE_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_SRE_64BIT_BAR_HIGH 0x0249 +#define regUVD_LMI_SRE_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_GPGPU_64BIT_BAR_LOW 0x024a +#define regUVD_LMI_MIF_GPGPU_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_GPGPU_64BIT_BAR_HIGH 0x024b +#define regUVD_LMI_MIF_GPGPU_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_CURR_LUMA_64BIT_BAR_LOW 0x024c +#define regUVD_LMI_MIF_CURR_LUMA_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_CURR_LUMA_64BIT_BAR_HIGH 0x024d +#define regUVD_LMI_MIF_CURR_LUMA_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_LOW 0x024e +#define regUVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_HIGH 0x024f +#define regUVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_REF_64BIT_BAR_LOW 0x0250 +#define regUVD_LMI_MIF_REF_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_REF_64BIT_BAR_HIGH 0x0251 +#define regUVD_LMI_MIF_REF_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_DBW_64BIT_BAR_LOW 0x0252 +#define regUVD_LMI_MIF_DBW_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_DBW_64BIT_BAR_HIGH 0x0253 +#define regUVD_LMI_MIF_DBW_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_CM_COLOC_64BIT_BAR_LOW 0x0254 +#define regUVD_LMI_MIF_CM_COLOC_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_CM_COLOC_64BIT_BAR_HIGH 0x0255 +#define regUVD_LMI_MIF_CM_COLOC_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_BSP0_64BIT_BAR_LOW 0x0256 +#define regUVD_LMI_MIF_BSP0_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_BSP0_64BIT_BAR_HIGH 0x0257 +#define regUVD_LMI_MIF_BSP0_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_BSP1_64BIT_BAR_LOW 0x0258 +#define regUVD_LMI_MIF_BSP1_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_BSP1_64BIT_BAR_HIGH 0x0259 +#define regUVD_LMI_MIF_BSP1_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_BSP2_64BIT_BAR_LOW 0x025a +#define regUVD_LMI_MIF_BSP2_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_BSP2_64BIT_BAR_HIGH 0x025b +#define regUVD_LMI_MIF_BSP2_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_BSP3_64BIT_BAR_LOW 0x025c +#define regUVD_LMI_MIF_BSP3_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_BSP3_64BIT_BAR_HIGH 0x025d +#define regUVD_LMI_MIF_BSP3_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_BSD0_64BIT_BAR_LOW 0x025e +#define regUVD_LMI_MIF_BSD0_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_BSD0_64BIT_BAR_HIGH 0x025f +#define regUVD_LMI_MIF_BSD0_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_BSD1_64BIT_BAR_LOW 0x0260 +#define regUVD_LMI_MIF_BSD1_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_BSD1_64BIT_BAR_HIGH 0x0261 +#define regUVD_LMI_MIF_BSD1_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_BSD2_64BIT_BAR_LOW 0x0262 +#define regUVD_LMI_MIF_BSD2_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_BSD2_64BIT_BAR_HIGH 0x0263 +#define regUVD_LMI_MIF_BSD2_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_BSD3_64BIT_BAR_LOW 0x0264 +#define regUVD_LMI_MIF_BSD3_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_BSD3_64BIT_BAR_HIGH 0x0265 +#define regUVD_LMI_MIF_BSD3_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_BSD4_64BIT_BAR_LOW 0x0266 +#define regUVD_LMI_MIF_BSD4_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_BSD4_64BIT_BAR_HIGH 0x0267 +#define regUVD_LMI_MIF_BSD4_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW 0x0270 +#define regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH 0x0271 +#define regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE8_64BIT_BAR_LOW 0x0272 +#define regUVD_LMI_VCPU_CACHE8_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE8_64BIT_BAR_HIGH 0x0273 +#define regUVD_LMI_VCPU_CACHE8_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW 0x0274 +#define regUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH 0x0275 +#define regUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE3_64BIT_BAR_LOW 0x0276 +#define regUVD_LMI_VCPU_CACHE3_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE3_64BIT_BAR_HIGH 0x0277 +#define regUVD_LMI_VCPU_CACHE3_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE4_64BIT_BAR_LOW 0x0278 +#define regUVD_LMI_VCPU_CACHE4_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE4_64BIT_BAR_HIGH 0x0279 +#define regUVD_LMI_VCPU_CACHE4_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE5_64BIT_BAR_LOW 0x027a +#define regUVD_LMI_VCPU_CACHE5_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE5_64BIT_BAR_HIGH 0x027b +#define regUVD_LMI_VCPU_CACHE5_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE6_64BIT_BAR_LOW 0x027c +#define regUVD_LMI_VCPU_CACHE6_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE6_64BIT_BAR_HIGH 0x027d +#define regUVD_LMI_VCPU_CACHE6_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE7_64BIT_BAR_LOW 0x027e +#define regUVD_LMI_VCPU_CACHE7_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE7_64BIT_BAR_HIGH 0x027f +#define regUVD_LMI_VCPU_CACHE7_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_SCLR_64BIT_BAR_LOW 0x0280 +#define regUVD_LMI_MIF_SCLR_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_SCLR_64BIT_BAR_HIGH 0x0281 +#define regUVD_LMI_MIF_SCLR_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_SCLR2_64BIT_BAR_LOW 0x0282 +#define regUVD_LMI_MIF_SCLR2_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_SCLR2_64BIT_BAR_HIGH 0x0283 +#define regUVD_LMI_MIF_SCLR2_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_SPH_64BIT_BAR_HIGH 0x0284 +#define regUVD_LMI_SPH_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_LOW 0x0298 +#define regUVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_HIGH 0x0299 +#define regUVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_LOW 0x029a +#define regUVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_HIGH 0x029b +#define regUVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_LOW 0x029c +#define regUVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_HIGH 0x029d +#define regUVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_LOW 0x029e +#define regUVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_HIGH 0x029f +#define regUVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_ADP_ATOMIC_CONFIG 0x02a1 +#define regUVD_ADP_ATOMIC_CONFIG_BASE_IDX 1 +#define regUVD_LMI_ARB_CTRL2 0x02a2 +#define regUVD_LMI_ARB_CTRL2_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE_VMIDS_MULTI 0x02a7 +#define regUVD_LMI_VCPU_CACHE_VMIDS_MULTI_BASE_IDX 1 +#define regUVD_LMI_VCPU_NC_VMIDS_MULTI 0x02a8 +#define regUVD_LMI_VCPU_NC_VMIDS_MULTI_BASE_IDX 1 +#define regUVD_LMI_LAT_CTRL 0x02a9 +#define regUVD_LMI_LAT_CTRL_BASE_IDX 1 +#define regUVD_LMI_LAT_CNTR 0x02aa +#define regUVD_LMI_LAT_CNTR_BASE_IDX 1 +#define regUVD_LMI_AVG_LAT_CNTR 0x02ab +#define regUVD_LMI_AVG_LAT_CNTR_BASE_IDX 1 +#define regUVD_LMI_SPH 0x02ac +#define regUVD_LMI_SPH_BASE_IDX 1 +#define regUVD_LMI_VCPU_CACHE_VMID 0x02ad +#define regUVD_LMI_VCPU_CACHE_VMID_BASE_IDX 1 +#define regUVD_LMI_CTRL2 0x02ae +#define regUVD_LMI_CTRL2_BASE_IDX 1 +#define regUVD_LMI_URGENT_CTRL 0x02af +#define regUVD_LMI_URGENT_CTRL_BASE_IDX 1 +#define regUVD_LMI_CTRL 0x02b0 +#define regUVD_LMI_CTRL_BASE_IDX 1 +#define regUVD_LMI_STATUS 0x02b1 +#define regUVD_LMI_STATUS_BASE_IDX 1 +#define regUVD_LMI_PERFMON_CTRL 0x02b4 +#define regUVD_LMI_PERFMON_CTRL_BASE_IDX 1 +#define regUVD_LMI_PERFMON_COUNT_LO 0x02b5 +#define regUVD_LMI_PERFMON_COUNT_LO_BASE_IDX 1 +#define regUVD_LMI_PERFMON_COUNT_HI 0x02b6 +#define regUVD_LMI_PERFMON_COUNT_HI_BASE_IDX 1 +#define regUVD_LMI_ADP_SWAP_CNTL 0x02b7 +#define regUVD_LMI_ADP_SWAP_CNTL_BASE_IDX 1 +#define regUVD_LMI_RBC_RB_VMID 0x02b8 +#define regUVD_LMI_RBC_RB_VMID_BASE_IDX 1 +#define regUVD_LMI_RBC_IB_VMID 0x02b9 +#define regUVD_LMI_RBC_IB_VMID_BASE_IDX 1 +#define regUVD_LMI_MC_CREDITS 0x02ba +#define regUVD_LMI_MC_CREDITS_BASE_IDX 1 +#define regUVD_LMI_ADP_IND_INDEX 0x02be +#define regUVD_LMI_ADP_IND_INDEX_BASE_IDX 1 +#define regUVD_LMI_ADP_IND_DATA 0x02bf +#define regUVD_LMI_ADP_IND_DATA_BASE_IDX 1 +#define regUVD_LMI_ADP_PF_EN 0x02c0 +#define regUVD_LMI_ADP_PF_EN_BASE_IDX 1 +#define regUVD_LMI_PREF_CTRL 0x02c2 +#define regUVD_LMI_PREF_CTRL_BASE_IDX 1 +#define regUVD_LMI_MIF_REF_LUMA_64BIT_BAR_LOW 0x02dd +#define regUVD_LMI_MIF_REF_LUMA_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MIF_REF_LUMA_64BIT_BAR_HIGH 0x02de +#define regUVD_LMI_MIF_REF_LUMA_64BIT_BAR_HIGH_BASE_IDX 1 +#define regVCN_RAS_CNTL 0x02df +#define regVCN_RAS_CNTL_BASE_IDX 1 + + +// addressBlock: aid_uvd0_uvd_jpeg0_jpegnpdec +// base address: 0x20f00 +#define regUVD_JPEG_CNTL 0x05c0 +#define regUVD_JPEG_CNTL_BASE_IDX 1 +#define regUVD_JPEG_RB_BASE 0x05c1 +#define regUVD_JPEG_RB_BASE_BASE_IDX 1 +#define regUVD_JPEG_RB_WPTR 0x05c2 +#define regUVD_JPEG_RB_WPTR_BASE_IDX 1 +#define regUVD_JPEG_RB_RPTR 0x05c3 +#define regUVD_JPEG_RB_RPTR_BASE_IDX 1 +#define regUVD_JPEG_RB_SIZE 0x05c4 +#define regUVD_JPEG_RB_SIZE_BASE_IDX 1 +#define regUVD_JPEG_DEC_CNT 0x05c5 +#define regUVD_JPEG_DEC_CNT_BASE_IDX 1 +#define regUVD_JPEG_SPS_INFO 0x05c6 +#define regUVD_JPEG_SPS_INFO_BASE_IDX 1 +#define regUVD_JPEG_SPS1_INFO 0x05c7 +#define regUVD_JPEG_SPS1_INFO_BASE_IDX 1 +#define regUVD_JPEG_RE_TIMER 0x05c8 +#define regUVD_JPEG_RE_TIMER_BASE_IDX 1 +#define regUVD_JPEG_DEC_SCRATCH0 0x05c9 +#define regUVD_JPEG_DEC_SCRATCH0_BASE_IDX 1 +#define regUVD_JPEG_INT_EN 0x05ca +#define regUVD_JPEG_INT_EN_BASE_IDX 1 +#define regUVD_JPEG_INT_STAT 0x05cb +#define regUVD_JPEG_INT_STAT_BASE_IDX 1 +#define regUVD_JPEG_TIER_CNTL0 0x05cc +#define regUVD_JPEG_TIER_CNTL0_BASE_IDX 1 +#define regUVD_JPEG_TIER_CNTL1 0x05cd +#define regUVD_JPEG_TIER_CNTL1_BASE_IDX 1 +#define regUVD_JPEG_TIER_CNTL2 0x05ce +#define regUVD_JPEG_TIER_CNTL2_BASE_IDX 1 +#define regUVD_JPEG_TIER_STATUS 0x05cf +#define regUVD_JPEG_TIER_STATUS_BASE_IDX 1 + + +// addressBlock: aid_uvd0_uvd_jpeg_sclk0_jpegnpsclkdec +// base address: 0x21000 +#define regUVD_JPEG_OUTBUF_CNTL 0x0600 +#define regUVD_JPEG_OUTBUF_CNTL_BASE_IDX 1 +#define regUVD_JPEG_OUTBUF_WPTR 0x0601 +#define regUVD_JPEG_OUTBUF_WPTR_BASE_IDX 1 +#define regUVD_JPEG_OUTBUF_RPTR 0x0602 +#define regUVD_JPEG_OUTBUF_RPTR_BASE_IDX 1 +#define regUVD_JPEG_PITCH 0x0603 +#define regUVD_JPEG_PITCH_BASE_IDX 1 +#define regUVD_JPEG_UV_PITCH 0x0604 +#define regUVD_JPEG_UV_PITCH_BASE_IDX 1 +#define regJPEG_DEC_Y_GFX8_TILING_SURFACE 0x0605 +#define regJPEG_DEC_Y_GFX8_TILING_SURFACE_BASE_IDX 1 +#define regJPEG_DEC_UV_GFX8_TILING_SURFACE 0x0606 +#define regJPEG_DEC_UV_GFX8_TILING_SURFACE_BASE_IDX 1 +#define regJPEG_DEC_GFX8_ADDR_CONFIG 0x0607 +#define regJPEG_DEC_GFX8_ADDR_CONFIG_BASE_IDX 1 +#define regJPEG_DEC_Y_GFX10_TILING_SURFACE 0x0608 +#define regJPEG_DEC_Y_GFX10_TILING_SURFACE_BASE_IDX 1 +#define regJPEG_DEC_UV_GFX10_TILING_SURFACE 0x0609 +#define regJPEG_DEC_UV_GFX10_TILING_SURFACE_BASE_IDX 1 +#define regJPEG_DEC_GFX10_ADDR_CONFIG 0x060a +#define regJPEG_DEC_GFX10_ADDR_CONFIG_BASE_IDX 1 +#define regJPEG_DEC_ADDR_MODE 0x060b +#define regJPEG_DEC_ADDR_MODE_BASE_IDX 1 +#define regUVD_JPEG_OUTPUT_XY 0x060c +#define regUVD_JPEG_OUTPUT_XY_BASE_IDX 1 +#define regUVD_JPEG_GPCOM_CMD 0x060d +#define regUVD_JPEG_GPCOM_CMD_BASE_IDX 1 +#define regUVD_JPEG_GPCOM_DATA0 0x060e +#define regUVD_JPEG_GPCOM_DATA0_BASE_IDX 1 +#define regUVD_JPEG_GPCOM_DATA1 0x060f +#define regUVD_JPEG_GPCOM_DATA1_BASE_IDX 1 +#define regUVD_JPEG_SCRATCH1 0x0610 +#define regUVD_JPEG_SCRATCH1_BASE_IDX 1 +#define regUVD_JPEG_DEC_SOFT_RST 0x0611 +#define regUVD_JPEG_DEC_SOFT_RST_BASE_IDX 1 + + +// addressBlock: aid_uvd0_uvd_jrbc0_uvd_jrbc_dec +// base address: 0x21100 +#define regUVD_JRBC0_UVD_JRBC_RB_WPTR 0x0640 +#define regUVD_JRBC0_UVD_JRBC_RB_WPTR_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JRBC_RB_CNTL 0x0641 +#define regUVD_JRBC0_UVD_JRBC_RB_CNTL_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JRBC_IB_SIZE 0x0642 +#define regUVD_JRBC0_UVD_JRBC_IB_SIZE_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JRBC_URGENT_CNTL 0x0643 +#define regUVD_JRBC0_UVD_JRBC_URGENT_CNTL_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JRBC_RB_REF_DATA 0x0644 +#define regUVD_JRBC0_UVD_JRBC_RB_REF_DATA_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER 0x0645 +#define regUVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JRBC_SOFT_RESET 0x0648 +#define regUVD_JRBC0_UVD_JRBC_SOFT_RESET_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JRBC_STATUS 0x0649 +#define regUVD_JRBC0_UVD_JRBC_STATUS_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JRBC_RB_RPTR 0x064a +#define regUVD_JRBC0_UVD_JRBC_RB_RPTR_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JRBC_RB_BUF_STATUS 0x064b +#define regUVD_JRBC0_UVD_JRBC_RB_BUF_STATUS_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JRBC_IB_BUF_STATUS 0x064c +#define regUVD_JRBC0_UVD_JRBC_IB_BUF_STATUS_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JRBC_IB_SIZE_UPDATE 0x064d +#define regUVD_JRBC0_UVD_JRBC_IB_SIZE_UPDATE_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER 0x064e +#define regUVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JRBC_IB_REF_DATA 0x064f +#define regUVD_JRBC0_UVD_JRBC_IB_REF_DATA_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JPEG_PREEMPT_CMD 0x0650 +#define regUVD_JRBC0_UVD_JPEG_PREEMPT_CMD_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA0 0x0651 +#define regUVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA0_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA1 0x0652 +#define regUVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA1_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JRBC_RB_SIZE 0x0653 +#define regUVD_JRBC0_UVD_JRBC_RB_SIZE_BASE_IDX 1 +#define regUVD_JRBC0_UVD_JRBC_SCRATCH0 0x0654 +#define regUVD_JRBC0_UVD_JRBC_SCRATCH0_BASE_IDX 1 + + +// addressBlock: aid_uvd0_uvd_jmi0_uvd_jmi_dec +// base address: 0x21180 +#define regUVD_JMI0_UVD_JPEG_DEC_PF_CTRL 0x0660 +#define regUVD_JMI0_UVD_JPEG_DEC_PF_CTRL_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JRBC_CTRL 0x0661 +#define regUVD_JMI0_UVD_LMI_JRBC_CTRL_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JPEG_CTRL 0x0662 +#define regUVD_JMI0_UVD_LMI_JPEG_CTRL_BASE_IDX 1 +#define regUVD_JMI0_JPEG_LMI_DROP 0x0663 +#define regUVD_JMI0_JPEG_LMI_DROP_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JRBC_IB_VMID 0x0664 +#define regUVD_JMI0_UVD_LMI_JRBC_IB_VMID_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JRBC_RB_VMID 0x0665 +#define regUVD_JMI0_UVD_LMI_JRBC_RB_VMID_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JPEG_VMID 0x0666 +#define regUVD_JMI0_UVD_LMI_JPEG_VMID_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW 0x0667 +#define regUVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH 0x0668 +#define regUVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW 0x0669 +#define regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH 0x066a +#define regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW 0x066b +#define regUVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH 0x066c +#define regUVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JPEG_PREEMPT_VMID 0x066d +#define regUVD_JMI0_UVD_LMI_JPEG_PREEMPT_VMID_BASE_IDX 1 +#define regUVD_JMI0_UVD_JMI_DEC_SWAP_CNTL 0x066e +#define regUVD_JMI0_UVD_JMI_DEC_SWAP_CNTL_BASE_IDX 1 +#define regUVD_JMI0_UVD_JMI_ATOMIC_CNTL 0x066f +#define regUVD_JMI0_UVD_JMI_ATOMIC_CNTL_BASE_IDX 1 +#define regUVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW 0x0670 +#define regUVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH 0x0671 +#define regUVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_LOW 0x0672 +#define regUVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH 0x0673 +#define regUVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW 0x0674 +#define regUVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH 0x0675 +#define regUVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_LOW 0x0676 +#define regUVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH 0x0677 +#define regUVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW 0x0678 +#define regUVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH 0x0679 +#define regUVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_JMI0_UVD_JMI_ATOMIC_CNTL2 0x067d +#define regUVD_JMI0_UVD_JMI_ATOMIC_CNTL2_BASE_IDX 1 + + +// addressBlock: aid_uvd0_uvd_jmi_common_dec +// base address: 0x21300 +#define regUVD_JADP_MCIF_URGENT_CTRL 0x06c1 +#define regUVD_JADP_MCIF_URGENT_CTRL_BASE_IDX 1 +#define regUVD_JMI_URGENT_CTRL 0x06c2 +#define regUVD_JMI_URGENT_CTRL_BASE_IDX 1 +#define regUVD_JMI_CTRL 0x06c3 +#define regUVD_JMI_CTRL_BASE_IDX 1 +#define regJPEG_MEMCHECK_CLAMPING_CNTL 0x06c4 +#define regJPEG_MEMCHECK_CLAMPING_CNTL_BASE_IDX 1 +#define regJPEG_MEMCHECK_SAFE_ADDR 0x06c5 +#define regJPEG_MEMCHECK_SAFE_ADDR_BASE_IDX 1 +#define regJPEG_MEMCHECK_SAFE_ADDR_64BIT 0x06c6 +#define regJPEG_MEMCHECK_SAFE_ADDR_64BIT_BASE_IDX 1 +#define regUVD_JMI_LAT_CTRL 0x06c7 +#define regUVD_JMI_LAT_CTRL_BASE_IDX 1 +#define regUVD_JMI_LAT_CNTR 0x06c8 +#define regUVD_JMI_LAT_CNTR_BASE_IDX 1 +#define regUVD_JMI_AVG_LAT_CNTR 0x06c9 +#define regUVD_JMI_AVG_LAT_CNTR_BASE_IDX 1 +#define regUVD_JMI_PERFMON_CTRL 0x06ca +#define regUVD_JMI_PERFMON_CTRL_BASE_IDX 1 +#define regUVD_JMI_PERFMON_COUNT_LO 0x06cb +#define regUVD_JMI_PERFMON_COUNT_LO_BASE_IDX 1 +#define regUVD_JMI_PERFMON_COUNT_HI 0x06cc +#define regUVD_JMI_PERFMON_COUNT_HI_BASE_IDX 1 +#define regUVD_JMI_CLEAN_STATUS 0x06cd +#define regUVD_JMI_CLEAN_STATUS_BASE_IDX 1 +#define regUVD_JMI_CNTL 0x06ce +#define regUVD_JMI_CNTL_BASE_IDX 1 + + +// addressBlock: aid_uvd0_uvd_jpeg_common_dec +// base address: 0x21400 +#define regJPEG_SOFT_RESET_STATUS 0x0700 +#define regJPEG_SOFT_RESET_STATUS_BASE_IDX 1 +#define regJPEG_SYS_INT_EN 0x0701 +#define regJPEG_SYS_INT_EN_BASE_IDX 1 +#define regJPEG_SYS_INT_EN1 0x0702 +#define regJPEG_SYS_INT_EN1_BASE_IDX 1 +#define regJPEG_SYS_INT_STATUS 0x0703 +#define regJPEG_SYS_INT_STATUS_BASE_IDX 1 +#define regJPEG_SYS_INT_STATUS1 0x0704 +#define regJPEG_SYS_INT_STATUS1_BASE_IDX 1 +#define regJPEG_SYS_INT_ACK 0x0705 +#define regJPEG_SYS_INT_ACK_BASE_IDX 1 +#define regJPEG_SYS_INT_ACK1 0x0706 +#define regJPEG_SYS_INT_ACK1_BASE_IDX 1 +#define regJPEG_MEMCHECK_SYS_INT_EN 0x0707 +#define regJPEG_MEMCHECK_SYS_INT_EN_BASE_IDX 1 +#define regJPEG_MEMCHECK_SYS_INT_EN1 0x0708 +#define regJPEG_MEMCHECK_SYS_INT_EN1_BASE_IDX 1 +#define regJPEG_MEMCHECK_SYS_INT_STAT 0x0709 +#define regJPEG_MEMCHECK_SYS_INT_STAT_BASE_IDX 1 +#define regJPEG_MEMCHECK_SYS_INT_STAT1 0x070a +#define regJPEG_MEMCHECK_SYS_INT_STAT1_BASE_IDX 1 +#define regJPEG_MEMCHECK_SYS_INT_STAT2 0x070b +#define regJPEG_MEMCHECK_SYS_INT_STAT2_BASE_IDX 1 +#define regJPEG_MEMCHECK_SYS_INT_ACK 0x070c +#define regJPEG_MEMCHECK_SYS_INT_ACK_BASE_IDX 1 +#define regJPEG_MEMCHECK_SYS_INT_ACK1 0x070d +#define regJPEG_MEMCHECK_SYS_INT_ACK1_BASE_IDX 1 +#define regJPEG_MEMCHECK_SYS_INT_ACK2 0x070e +#define regJPEG_MEMCHECK_SYS_INT_ACK2_BASE_IDX 1 +#define regJPEG_MASTINT_EN 0x0710 +#define regJPEG_MASTINT_EN_BASE_IDX 1 +#define regJPEG_IH_CTRL 0x0711 +#define regJPEG_IH_CTRL_BASE_IDX 1 +#define regJRBBM_ARB_CTRL 0x0713 +#define regJRBBM_ARB_CTRL_BASE_IDX 1 + + +// addressBlock: aid_uvd0_uvd_jpeg_common_sclk_dec +// base address: 0x21480 +#define regJPEG_CGC_GATE 0x0720 +#define regJPEG_CGC_GATE_BASE_IDX 1 +#define regJPEG_CGC_CTRL 0x0721 +#define regJPEG_CGC_CTRL_BASE_IDX 1 +#define regJPEG_CGC_STATUS 0x0722 +#define regJPEG_CGC_STATUS_BASE_IDX 1 +#define regJPEG_COMN_CGC_MEM_CTRL 0x0723 +#define regJPEG_COMN_CGC_MEM_CTRL_BASE_IDX 1 +#define regJPEG_DEC_CGC_MEM_CTRL 0x0724 +#define regJPEG_DEC_CGC_MEM_CTRL_BASE_IDX 1 +#define regJPEG_ENC_CGC_MEM_CTRL 0x0726 +#define regJPEG_ENC_CGC_MEM_CTRL_BASE_IDX 1 +#define regJPEG_PERF_BANK_CONF 0x0727 +#define regJPEG_PERF_BANK_CONF_BASE_IDX 1 +#define regJPEG_PERF_BANK_EVENT_SEL 0x0728 +#define regJPEG_PERF_BANK_EVENT_SEL_BASE_IDX 1 +#define regJPEG_PERF_BANK_COUNT0 0x0729 +#define regJPEG_PERF_BANK_COUNT0_BASE_IDX 1 +#define regJPEG_PERF_BANK_COUNT1 0x072a +#define regJPEG_PERF_BANK_COUNT1_BASE_IDX 1 +#define regJPEG_PERF_BANK_COUNT2 0x072b +#define regJPEG_PERF_BANK_COUNT2_BASE_IDX 1 +#define regJPEG_PERF_BANK_COUNT3 0x072c +#define regJPEG_PERF_BANK_COUNT3_BASE_IDX 1 + + +// addressBlock: aid_uvd0_uvd_pg_dec +// base address: 0x1f800 +#define regUVD_PGFSM_CONFIG 0x0000 +#define regUVD_PGFSM_CONFIG_BASE_IDX 1 +#define regUVD_PGFSM_STATUS 0x0001 +#define regUVD_PGFSM_STATUS_BASE_IDX 1 +#define regUVD_POWER_STATUS 0x0002 +#define regUVD_POWER_STATUS_BASE_IDX 1 +#define regUVD_JPEG_POWER_STATUS 0x0003 +#define regUVD_JPEG_POWER_STATUS_BASE_IDX 1 +#define regUVD_MC_DJPEG_RD_SPACE 0x0006 +#define regUVD_MC_DJPEG_RD_SPACE_BASE_IDX 1 +#define regUVD_MC_DJPEG_WR_SPACE 0x0007 +#define regUVD_MC_DJPEG_WR_SPACE_BASE_IDX 1 +#define regUVD_MC_EJPEG_RD_SPACE 0x0008 +#define regUVD_MC_EJPEG_RD_SPACE_BASE_IDX 1 +#define regUVD_MC_EJPEG_WR_SPACE 0x0009 +#define regUVD_MC_EJPEG_WR_SPACE_BASE_IDX 1 +#define regUVD_PG_IND_INDEX 0x000c +#define regUVD_PG_IND_INDEX_BASE_IDX 1 +#define regUVD_PG_IND_DATA 0x000e +#define regUVD_PG_IND_DATA_BASE_IDX 1 +#define regCC_UVD_HARVESTING 0x000f +#define regCC_UVD_HARVESTING_BASE_IDX 1 +#define regUVD_DPG_LMA_CTL 0x0011 +#define regUVD_DPG_LMA_CTL_BASE_IDX 1 +#define regUVD_DPG_LMA_DATA 0x0012 +#define regUVD_DPG_LMA_DATA_BASE_IDX 1 +#define regUVD_DPG_LMA_MASK 0x0013 +#define regUVD_DPG_LMA_MASK_BASE_IDX 1 +#define regUVD_DPG_PAUSE 0x0014 +#define regUVD_DPG_PAUSE_BASE_IDX 1 +#define regUVD_SCRATCH1 0x0015 +#define regUVD_SCRATCH1_BASE_IDX 1 +#define regUVD_SCRATCH2 0x0016 +#define regUVD_SCRATCH2_BASE_IDX 1 +#define regUVD_SCRATCH3 0x0017 +#define regUVD_SCRATCH3_BASE_IDX 1 +#define regUVD_SCRATCH4 0x0018 +#define regUVD_SCRATCH4_BASE_IDX 1 +#define regUVD_SCRATCH5 0x0019 +#define regUVD_SCRATCH5_BASE_IDX 1 +#define regUVD_SCRATCH6 0x001a +#define regUVD_SCRATCH6_BASE_IDX 1 +#define regUVD_SCRATCH7 0x001b +#define regUVD_SCRATCH7_BASE_IDX 1 +#define regUVD_SCRATCH8 0x001c +#define regUVD_SCRATCH8_BASE_IDX 1 +#define regUVD_SCRATCH9 0x001d +#define regUVD_SCRATCH9_BASE_IDX 1 +#define regUVD_SCRATCH10 0x001e +#define regUVD_SCRATCH10_BASE_IDX 1 +#define regUVD_SCRATCH11 0x001f +#define regUVD_SCRATCH11_BASE_IDX 1 +#define regUVD_SCRATCH12 0x0020 +#define regUVD_SCRATCH12_BASE_IDX 1 +#define regUVD_SCRATCH13 0x0021 +#define regUVD_SCRATCH13_BASE_IDX 1 +#define regUVD_SCRATCH14 0x0022 +#define regUVD_SCRATCH14_BASE_IDX 1 +#define regUVD_FREE_COUNTER_REG 0x0023 +#define regUVD_FREE_COUNTER_REG_BASE_IDX 1 +#define regUVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_LOW 0x0024 +#define regUVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_HIGH 0x0025 +#define regUVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_DPG_VCPU_CACHE_OFFSET0 0x0026 +#define regUVD_DPG_VCPU_CACHE_OFFSET0_BASE_IDX 1 +#define regUVD_DPG_LMI_VCPU_CACHE_VMID 0x0027 +#define regUVD_DPG_LMI_VCPU_CACHE_VMID_BASE_IDX 1 +#define regUVD_REG_FILTER_EN 0x0028 +#define regUVD_REG_FILTER_EN_BASE_IDX 1 +#define regUVD_SECURITY_REG_VIO_REPORT 0x0029 +#define regUVD_SECURITY_REG_VIO_REPORT_BASE_IDX 1 +#define regUVD_FW_VERSION 0x002a +#define regUVD_FW_VERSION_BASE_IDX 1 +#define regUVD_PF_STATUS 0x002c +#define regUVD_PF_STATUS_BASE_IDX 1 +#define regUVD_DPG_CLK_EN_VCPU_REPORT 0x002e +#define regUVD_DPG_CLK_EN_VCPU_REPORT_BASE_IDX 1 +#define regCC_UVD_VCPU_ERR_DETECT_BOT_LO 0x002f +#define regCC_UVD_VCPU_ERR_DETECT_BOT_LO_BASE_IDX 1 +#define regCC_UVD_VCPU_ERR_DETECT_BOT_HI 0x0030 +#define regCC_UVD_VCPU_ERR_DETECT_BOT_HI_BASE_IDX 1 +#define regCC_UVD_VCPU_ERR_DETECT_TOP_LO 0x0031 +#define regCC_UVD_VCPU_ERR_DETECT_TOP_LO_BASE_IDX 1 +#define regCC_UVD_VCPU_ERR_DETECT_TOP_HI 0x0032 +#define regCC_UVD_VCPU_ERR_DETECT_TOP_HI_BASE_IDX 1 +#define regCC_UVD_VCPU_ERR 0x0033 +#define regCC_UVD_VCPU_ERR_BASE_IDX 1 +#define regCC_UVD_VCPU_ERR_INST_ADDR_LO 0x0034 +#define regCC_UVD_VCPU_ERR_INST_ADDR_LO_BASE_IDX 1 +#define regCC_UVD_VCPU_ERR_INST_ADDR_HI 0x0035 +#define regCC_UVD_VCPU_ERR_INST_ADDR_HI_BASE_IDX 1 +#define regUVD_LMI_MMSCH_NC_SPACE 0x003d +#define regUVD_LMI_MMSCH_NC_SPACE_BASE_IDX 1 +#define regUVD_LMI_ATOMIC_SPACE 0x003e +#define regUVD_LMI_ATOMIC_SPACE_BASE_IDX 1 +#define regUVD_GFX8_ADDR_CONFIG 0x0041 +#define regUVD_GFX8_ADDR_CONFIG_BASE_IDX 1 +#define regUVD_GFX10_ADDR_CONFIG 0x0042 +#define regUVD_GFX10_ADDR_CONFIG_BASE_IDX 1 +#define regUVD_GPCNT2_CNTL 0x0043 +#define regUVD_GPCNT2_CNTL_BASE_IDX 1 +#define regUVD_GPCNT2_TARGET_LOWER 0x0044 +#define regUVD_GPCNT2_TARGET_LOWER_BASE_IDX 1 +#define regUVD_GPCNT2_STATUS_LOWER 0x0045 +#define regUVD_GPCNT2_STATUS_LOWER_BASE_IDX 1 +#define regUVD_GPCNT2_TARGET_UPPER 0x0046 +#define regUVD_GPCNT2_TARGET_UPPER_BASE_IDX 1 +#define regUVD_GPCNT2_STATUS_UPPER 0x0047 +#define regUVD_GPCNT2_STATUS_UPPER_BASE_IDX 1 +#define regUVD_GPCNT3_CNTL 0x0048 +#define regUVD_GPCNT3_CNTL_BASE_IDX 1 +#define regUVD_GPCNT3_TARGET_LOWER 0x0049 +#define regUVD_GPCNT3_TARGET_LOWER_BASE_IDX 1 +#define regUVD_GPCNT3_STATUS_LOWER 0x004a +#define regUVD_GPCNT3_STATUS_LOWER_BASE_IDX 1 +#define regUVD_GPCNT3_TARGET_UPPER 0x004b +#define regUVD_GPCNT3_TARGET_UPPER_BASE_IDX 1 +#define regUVD_GPCNT3_STATUS_UPPER 0x004c +#define regUVD_GPCNT3_STATUS_UPPER_BASE_IDX 1 +#define regUVD_VCLK_DS_CNTL 0x004d +#define regUVD_VCLK_DS_CNTL_BASE_IDX 1 +#define regUVD_DCLK_DS_CNTL 0x004e +#define regUVD_DCLK_DS_CNTL_BASE_IDX 1 +#define regUVD_TSC_LOWER 0x004f +#define regUVD_TSC_LOWER_BASE_IDX 1 +#define regUVD_TSC_UPPER 0x0050 +#define regUVD_TSC_UPPER_BASE_IDX 1 +#define regVCN_FEATURES 0x0051 +#define regVCN_FEATURES_BASE_IDX 1 +#define regUVD_GPUIOV_STATUS 0x0055 +#define regUVD_GPUIOV_STATUS_BASE_IDX 1 +#define regUVD_RAS_VCPU_VCODEC_STATUS 0x0057 +#define regUVD_RAS_VCPU_VCODEC_STATUS_BASE_IDX 1 +#define regUVD_RAS_MMSCH_FATAL_ERROR 0x0058 +#define regUVD_RAS_MMSCH_FATAL_ERROR_BASE_IDX 1 +#define regUVD_RAS_JPEG0_STATUS 0x0059 +#define regUVD_RAS_JPEG0_STATUS_BASE_IDX 1 +#define regUVD_RAS_JPEG1_STATUS 0x005a +#define regUVD_RAS_JPEG1_STATUS_BASE_IDX 1 +#define regUVD_RAS_CNTL_PMI_ARB 0x005b +#define regUVD_RAS_CNTL_PMI_ARB_BASE_IDX 1 +#define regUVD_SCRATCH15 0x005c +#define regUVD_SCRATCH15_BASE_IDX 1 +#define regVCN_JPEG_DB_CTRL1 0x005d +#define regVCN_JPEG_DB_CTRL1_BASE_IDX 1 +#define regVCN_JPEG_DB_CTRL2 0x005e +#define regVCN_JPEG_DB_CTRL2_BASE_IDX 1 +#define regVCN_JPEG_DB_CTRL3 0x005f +#define regVCN_JPEG_DB_CTRL3_BASE_IDX 1 +#define regVCN_JPEG_DB_CTRL4 0x0060 +#define regVCN_JPEG_DB_CTRL4_BASE_IDX 1 +#define regVCN_JPEG_DB_CTRL5 0x0061 +#define regVCN_JPEG_DB_CTRL5_BASE_IDX 1 +#define regVCN_JPEG_DB_CTRL6 0x0062 +#define regVCN_JPEG_DB_CTRL6_BASE_IDX 1 +#define regVCN_JPEG_DB_CTRL7 0x0063 +#define regVCN_JPEG_DB_CTRL7_BASE_IDX 1 +#define regUVD_SCRATCH32 0x006d +#define regUVD_SCRATCH32_BASE_IDX 1 +#define regUVD_VERSION 0x006e +#define regUVD_VERSION_BASE_IDX 1 +#define regVCN_RB_DB_CTRL 0x0070 +#define regVCN_RB_DB_CTRL_BASE_IDX 1 +#define regVCN_JPEG_DB_CTRL 0x0071 +#define regVCN_JPEG_DB_CTRL_BASE_IDX 1 +#define regVCN_RB1_DB_CTRL 0x0072 +#define regVCN_RB1_DB_CTRL_BASE_IDX 1 +#define regVCN_RB2_DB_CTRL 0x0073 +#define regVCN_RB2_DB_CTRL_BASE_IDX 1 +#define regVCN_RB3_DB_CTRL 0x0074 +#define regVCN_RB3_DB_CTRL_BASE_IDX 1 +#define regVCN_RB4_DB_CTRL 0x0075 +#define regVCN_RB4_DB_CTRL_BASE_IDX 1 +#define regVCN_RB_ENABLE 0x0085 +#define regVCN_RB_ENABLE_BASE_IDX 1 +#define regVCN_RB_WPTR_CTRL 0x0086 +#define regVCN_RB_WPTR_CTRL_BASE_IDX 1 +#define regUVD_RB_RPTR 0x00ac +#define regUVD_RB_RPTR_BASE_IDX 1 +#define regUVD_RB_WPTR 0x00ad +#define regUVD_RB_WPTR_BASE_IDX 1 +#define regUVD_RB_RPTR2 0x00ae +#define regUVD_RB_RPTR2_BASE_IDX 1 +#define regUVD_RB_WPTR2 0x00af +#define regUVD_RB_WPTR2_BASE_IDX 1 +#define regUVD_RB_RPTR3 0x00b0 +#define regUVD_RB_RPTR3_BASE_IDX 1 +#define regUVD_RB_WPTR3 0x00b1 +#define regUVD_RB_WPTR3_BASE_IDX 1 +#define regUVD_RB_RPTR4 0x00b2 +#define regUVD_RB_RPTR4_BASE_IDX 1 +#define regUVD_RB_WPTR4 0x00b3 +#define regUVD_RB_WPTR4_BASE_IDX 1 +#define regUVD_OUT_RB_RPTR 0x00b4 +#define regUVD_OUT_RB_RPTR_BASE_IDX 1 +#define regUVD_OUT_RB_WPTR 0x00b5 +#define regUVD_OUT_RB_WPTR_BASE_IDX 1 +#define regUVD_AUDIO_RB_RPTR 0x00b6 +#define regUVD_AUDIO_RB_RPTR_BASE_IDX 1 +#define regUVD_AUDIO_RB_WPTR 0x00b7 +#define regUVD_AUDIO_RB_WPTR_BASE_IDX 1 +#define regUVD_RBC_RB_RPTR 0x00b8 +#define regUVD_RBC_RB_RPTR_BASE_IDX 1 +#define regUVD_RBC_RB_WPTR 0x00b9 +#define regUVD_RBC_RB_WPTR_BASE_IDX 1 +#define regUVD_DPG_LMA_CTL2 0x00bb +#define regUVD_DPG_LMA_CTL2_BASE_IDX 1 + + +// addressBlock: aid_uvd0_mmsch_dec +// base address: 0x20d00 +#define regMMSCH_UCODE_ADDR 0x0540 +#define regMMSCH_UCODE_ADDR_BASE_IDX 1 +#define regMMSCH_UCODE_DATA 0x0541 +#define regMMSCH_UCODE_DATA_BASE_IDX 1 +#define regMMSCH_SRAM_ADDR 0x0542 +#define regMMSCH_SRAM_ADDR_BASE_IDX 1 +#define regMMSCH_SRAM_DATA 0x0543 +#define regMMSCH_SRAM_DATA_BASE_IDX 1 +#define regMMSCH_VF_SRAM_OFFSET 0x0544 +#define regMMSCH_VF_SRAM_OFFSET_BASE_IDX 1 +#define regMMSCH_DB_SRAM_OFFSET 0x0545 +#define regMMSCH_DB_SRAM_OFFSET_BASE_IDX 1 +#define regMMSCH_CTX_SRAM_OFFSET 0x0546 +#define regMMSCH_CTX_SRAM_OFFSET_BASE_IDX 1 +#define regMMSCH_CTL 0x0547 +#define regMMSCH_CTL_BASE_IDX 1 +#define regMMSCH_INTR 0x0548 +#define regMMSCH_INTR_BASE_IDX 1 +#define regMMSCH_INTR_ACK 0x0549 +#define regMMSCH_INTR_ACK_BASE_IDX 1 +#define regMMSCH_INTR_STATUS 0x054a +#define regMMSCH_INTR_STATUS_BASE_IDX 1 +#define regMMSCH_VF_VMID 0x054b +#define regMMSCH_VF_VMID_BASE_IDX 1 +#define regMMSCH_VF_CTX_ADDR_LO 0x054c +#define regMMSCH_VF_CTX_ADDR_LO_BASE_IDX 1 +#define regMMSCH_VF_CTX_ADDR_HI 0x054d +#define regMMSCH_VF_CTX_ADDR_HI_BASE_IDX 1 +#define regMMSCH_VF_CTX_SIZE 0x054e +#define regMMSCH_VF_CTX_SIZE_BASE_IDX 1 +#define regMMSCH_VF_GPCOM_ADDR_LO 0x054f +#define regMMSCH_VF_GPCOM_ADDR_LO_BASE_IDX 1 +#define regMMSCH_VF_GPCOM_ADDR_HI 0x0550 +#define regMMSCH_VF_GPCOM_ADDR_HI_BASE_IDX 1 +#define regMMSCH_VF_GPCOM_SIZE 0x0551 +#define regMMSCH_VF_GPCOM_SIZE_BASE_IDX 1 +#define regMMSCH_VF_MAILBOX_HOST 0x0552 +#define regMMSCH_VF_MAILBOX_HOST_BASE_IDX 1 +#define regMMSCH_VF_MAILBOX_RESP 0x0553 +#define regMMSCH_VF_MAILBOX_RESP_BASE_IDX 1 +#define regMMSCH_VF_MAILBOX_0 0x0554 +#define regMMSCH_VF_MAILBOX_0_BASE_IDX 1 +#define regMMSCH_VF_MAILBOX_0_RESP 0x0555 +#define regMMSCH_VF_MAILBOX_0_RESP_BASE_IDX 1 +#define regMMSCH_VF_MAILBOX_1 0x0556 +#define regMMSCH_VF_MAILBOX_1_BASE_IDX 1 +#define regMMSCH_VF_MAILBOX_1_RESP 0x0557 +#define regMMSCH_VF_MAILBOX_1_RESP_BASE_IDX 1 +#define regMMSCH_CNTL 0x055c +#define regMMSCH_CNTL_BASE_IDX 1 +#define regMMSCH_NONCACHE_OFFSET0 0x055d +#define regMMSCH_NONCACHE_OFFSET0_BASE_IDX 1 +#define regMMSCH_NONCACHE_SIZE0 0x055e +#define regMMSCH_NONCACHE_SIZE0_BASE_IDX 1 +#define regMMSCH_NONCACHE_OFFSET1 0x055f +#define regMMSCH_NONCACHE_OFFSET1_BASE_IDX 1 +#define regMMSCH_NONCACHE_SIZE1 0x0560 +#define regMMSCH_NONCACHE_SIZE1_BASE_IDX 1 +#define regMMSCH_PROC_STATE1 0x0566 +#define regMMSCH_PROC_STATE1_BASE_IDX 1 +#define regMMSCH_LAST_MC_ADDR 0x0567 +#define regMMSCH_LAST_MC_ADDR_BASE_IDX 1 +#define regMMSCH_LAST_MEM_ACCESS_HI 0x0568 +#define regMMSCH_LAST_MEM_ACCESS_HI_BASE_IDX 1 +#define regMMSCH_LAST_MEM_ACCESS_LO 0x0569 +#define regMMSCH_LAST_MEM_ACCESS_LO_BASE_IDX 1 +#define regMMSCH_IOV_ACTIVE_FCN_ID 0x056a +#define regMMSCH_IOV_ACTIVE_FCN_ID_BASE_IDX 1 +#define regMMSCH_SCRATCH_0 0x056b +#define regMMSCH_SCRATCH_0_BASE_IDX 1 +#define regMMSCH_SCRATCH_1 0x056c +#define regMMSCH_SCRATCH_1_BASE_IDX 1 +#define regMMSCH_GPUIOV_SCH_BLOCK_0 0x056d +#define regMMSCH_GPUIOV_SCH_BLOCK_0_BASE_IDX 1 +#define regMMSCH_GPUIOV_CMD_CONTROL_0 0x056e +#define regMMSCH_GPUIOV_CMD_CONTROL_0_BASE_IDX 1 +#define regMMSCH_GPUIOV_CMD_STATUS_0 0x056f +#define regMMSCH_GPUIOV_CMD_STATUS_0_BASE_IDX 1 +#define regMMSCH_GPUIOV_VM_BUSY_STATUS_0 0x0570 +#define regMMSCH_GPUIOV_VM_BUSY_STATUS_0_BASE_IDX 1 +#define regMMSCH_GPUIOV_ACTIVE_FCNS_0 0x0571 +#define regMMSCH_GPUIOV_ACTIVE_FCNS_0_BASE_IDX 1 +#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_0 0x0572 +#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_0_BASE_IDX 1 +#define regMMSCH_GPUIOV_DW6_0 0x0573 +#define regMMSCH_GPUIOV_DW6_0_BASE_IDX 1 +#define regMMSCH_GPUIOV_DW7_0 0x0574 +#define regMMSCH_GPUIOV_DW7_0_BASE_IDX 1 +#define regMMSCH_GPUIOV_DW8_0 0x0575 +#define regMMSCH_GPUIOV_DW8_0_BASE_IDX 1 +#define regMMSCH_GPUIOV_SCH_BLOCK_1 0x0576 +#define regMMSCH_GPUIOV_SCH_BLOCK_1_BASE_IDX 1 +#define regMMSCH_GPUIOV_CMD_CONTROL_1 0x0577 +#define regMMSCH_GPUIOV_CMD_CONTROL_1_BASE_IDX 1 +#define regMMSCH_GPUIOV_CMD_STATUS_1 0x0578 +#define regMMSCH_GPUIOV_CMD_STATUS_1_BASE_IDX 1 +#define regMMSCH_GPUIOV_VM_BUSY_STATUS_1 0x0579 +#define regMMSCH_GPUIOV_VM_BUSY_STATUS_1_BASE_IDX 1 +#define regMMSCH_GPUIOV_ACTIVE_FCNS_1 0x057a +#define regMMSCH_GPUIOV_ACTIVE_FCNS_1_BASE_IDX 1 +#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_1 0x057b +#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_1_BASE_IDX 1 +#define regMMSCH_GPUIOV_DW6_1 0x057c +#define regMMSCH_GPUIOV_DW6_1_BASE_IDX 1 +#define regMMSCH_GPUIOV_DW7_1 0x057d +#define regMMSCH_GPUIOV_DW7_1_BASE_IDX 1 +#define regMMSCH_GPUIOV_DW8_1 0x057e +#define regMMSCH_GPUIOV_DW8_1_BASE_IDX 1 +#define regMMSCH_GPUIOV_CNTXT 0x057f +#define regMMSCH_GPUIOV_CNTXT_BASE_IDX 1 +#define regMMSCH_SCRATCH_2 0x0580 +#define regMMSCH_SCRATCH_2_BASE_IDX 1 +#define regMMSCH_SCRATCH_3 0x0581 +#define regMMSCH_SCRATCH_3_BASE_IDX 1 +#define regMMSCH_SCRATCH_4 0x0582 +#define regMMSCH_SCRATCH_4_BASE_IDX 1 +#define regMMSCH_SCRATCH_5 0x0583 +#define regMMSCH_SCRATCH_5_BASE_IDX 1 +#define regMMSCH_SCRATCH_6 0x0584 +#define regMMSCH_SCRATCH_6_BASE_IDX 1 +#define regMMSCH_SCRATCH_7 0x0585 +#define regMMSCH_SCRATCH_7_BASE_IDX 1 +#define regMMSCH_VFID_FIFO_HEAD_0 0x0586 +#define regMMSCH_VFID_FIFO_HEAD_0_BASE_IDX 1 +#define regMMSCH_VFID_FIFO_TAIL_0 0x0587 +#define regMMSCH_VFID_FIFO_TAIL_0_BASE_IDX 1 +#define regMMSCH_VFID_FIFO_HEAD_1 0x0588 +#define regMMSCH_VFID_FIFO_HEAD_1_BASE_IDX 1 +#define regMMSCH_VFID_FIFO_TAIL_1 0x0589 +#define regMMSCH_VFID_FIFO_TAIL_1_BASE_IDX 1 +#define regMMSCH_NACK_STATUS 0x058a +#define regMMSCH_NACK_STATUS_BASE_IDX 1 +#define regMMSCH_VF_MAILBOX0_DATA 0x058b +#define regMMSCH_VF_MAILBOX0_DATA_BASE_IDX 1 +#define regMMSCH_VF_MAILBOX1_DATA 0x058c +#define regMMSCH_VF_MAILBOX1_DATA_BASE_IDX 1 +#define regMMSCH_GPUIOV_SCH_BLOCK_IP_0 0x058d +#define regMMSCH_GPUIOV_SCH_BLOCK_IP_0_BASE_IDX 1 +#define regMMSCH_GPUIOV_CMD_STATUS_IP_0 0x058e +#define regMMSCH_GPUIOV_CMD_STATUS_IP_0_BASE_IDX 1 +#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0 0x058f +#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0_BASE_IDX 1 +#define regMMSCH_GPUIOV_SCH_BLOCK_IP_1 0x0590 +#define regMMSCH_GPUIOV_SCH_BLOCK_IP_1_BASE_IDX 1 +#define regMMSCH_GPUIOV_CMD_STATUS_IP_1 0x0591 +#define regMMSCH_GPUIOV_CMD_STATUS_IP_1_BASE_IDX 1 +#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1 0x0592 +#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1_BASE_IDX 1 +#define regMMSCH_GPUIOV_CNTXT_IP 0x0593 +#define regMMSCH_GPUIOV_CNTXT_IP_BASE_IDX 1 +#define regMMSCH_GPUIOV_SCH_BLOCK_2 0x0594 +#define regMMSCH_GPUIOV_SCH_BLOCK_2_BASE_IDX 1 +#define regMMSCH_GPUIOV_CMD_CONTROL_2 0x0595 +#define regMMSCH_GPUIOV_CMD_CONTROL_2_BASE_IDX 1 +#define regMMSCH_GPUIOV_CMD_STATUS_2 0x0596 +#define regMMSCH_GPUIOV_CMD_STATUS_2_BASE_IDX 1 +#define regMMSCH_GPUIOV_VM_BUSY_STATUS_2 0x0597 +#define regMMSCH_GPUIOV_VM_BUSY_STATUS_2_BASE_IDX 1 +#define regMMSCH_GPUIOV_ACTIVE_FCNS_2 0x0598 +#define regMMSCH_GPUIOV_ACTIVE_FCNS_2_BASE_IDX 1 +#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_2 0x0599 +#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_2_BASE_IDX 1 +#define regMMSCH_GPUIOV_DW6_2 0x059a +#define regMMSCH_GPUIOV_DW6_2_BASE_IDX 1 +#define regMMSCH_GPUIOV_DW7_2 0x059b +#define regMMSCH_GPUIOV_DW7_2_BASE_IDX 1 +#define regMMSCH_GPUIOV_DW8_2 0x059c +#define regMMSCH_GPUIOV_DW8_2_BASE_IDX 1 +#define regMMSCH_GPUIOV_SCH_BLOCK_IP_2 0x059d +#define regMMSCH_GPUIOV_SCH_BLOCK_IP_2_BASE_IDX 1 +#define regMMSCH_GPUIOV_CMD_STATUS_IP_2 0x059e +#define regMMSCH_GPUIOV_CMD_STATUS_IP_2_BASE_IDX 1 +#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2 0x059f +#define regMMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2_BASE_IDX 1 +#define regMMSCH_VFID_FIFO_HEAD_2 0x05a0 +#define regMMSCH_VFID_FIFO_HEAD_2_BASE_IDX 1 +#define regMMSCH_VFID_FIFO_TAIL_2 0x05a1 +#define regMMSCH_VFID_FIFO_TAIL_2_BASE_IDX 1 +#define regMMSCH_VM_BUSY_STATUS_0 0x05a2 +#define regMMSCH_VM_BUSY_STATUS_0_BASE_IDX 1 +#define regMMSCH_VM_BUSY_STATUS_1 0x05a3 +#define regMMSCH_VM_BUSY_STATUS_1_BASE_IDX 1 +#define regMMSCH_VM_BUSY_STATUS_2 0x05a4 +#define regMMSCH_VM_BUSY_STATUS_2_BASE_IDX 1 + + +// addressBlock: aid_uvd0_slmi_adpdec +// base address: 0x21c00 +#define regUVD_LMI_MMSCH_NC0_64BIT_BAR_LOW 0x0900 +#define regUVD_LMI_MMSCH_NC0_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MMSCH_NC0_64BIT_BAR_HIGH 0x0901 +#define regUVD_LMI_MMSCH_NC0_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MMSCH_NC1_64BIT_BAR_LOW 0x0902 +#define regUVD_LMI_MMSCH_NC1_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MMSCH_NC1_64BIT_BAR_HIGH 0x0903 +#define regUVD_LMI_MMSCH_NC1_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MMSCH_NC2_64BIT_BAR_LOW 0x0904 +#define regUVD_LMI_MMSCH_NC2_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MMSCH_NC2_64BIT_BAR_HIGH 0x0905 +#define regUVD_LMI_MMSCH_NC2_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MMSCH_NC3_64BIT_BAR_LOW 0x0906 +#define regUVD_LMI_MMSCH_NC3_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MMSCH_NC3_64BIT_BAR_HIGH 0x0907 +#define regUVD_LMI_MMSCH_NC3_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MMSCH_NC4_64BIT_BAR_LOW 0x0908 +#define regUVD_LMI_MMSCH_NC4_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MMSCH_NC4_64BIT_BAR_HIGH 0x0909 +#define regUVD_LMI_MMSCH_NC4_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MMSCH_NC5_64BIT_BAR_LOW 0x090a +#define regUVD_LMI_MMSCH_NC5_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MMSCH_NC5_64BIT_BAR_HIGH 0x090b +#define regUVD_LMI_MMSCH_NC5_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MMSCH_NC6_64BIT_BAR_LOW 0x090c +#define regUVD_LMI_MMSCH_NC6_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MMSCH_NC6_64BIT_BAR_HIGH 0x090d +#define regUVD_LMI_MMSCH_NC6_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MMSCH_NC7_64BIT_BAR_LOW 0x090e +#define regUVD_LMI_MMSCH_NC7_64BIT_BAR_LOW_BASE_IDX 1 +#define regUVD_LMI_MMSCH_NC7_64BIT_BAR_HIGH 0x090f +#define regUVD_LMI_MMSCH_NC7_64BIT_BAR_HIGH_BASE_IDX 1 +#define regUVD_LMI_MMSCH_NC_VMID 0x0910 +#define regUVD_LMI_MMSCH_NC_VMID_BASE_IDX 1 +#define regUVD_LMI_MMSCH_CTRL 0x0911 +#define regUVD_LMI_MMSCH_CTRL_BASE_IDX 1 +#define regUVD_MMSCH_LMI_STATUS 0x0912 +#define regUVD_MMSCH_LMI_STATUS_BASE_IDX 1 +#define regVCN_RAS_CNTL_MMSCH 0x0914 +#define regVCN_RAS_CNTL_MMSCH_BASE_IDX 1 + +// addressBlock: aid_uvd0_vcn_edcc_dec +// base address: 0x21d20 +#define regVCN_UE_ERR_STATUS_LO_VIDD 0x094c +#define regVCN_UE_ERR_STATUS_LO_VIDD_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_VIDD 0x094d +#define regVCN_UE_ERR_STATUS_HI_VIDD_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_LO_VIDV 0x094e +#define regVCN_UE_ERR_STATUS_LO_VIDV_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_VIDV 0x094f +#define regVCN_UE_ERR_STATUS_HI_VIDV_BASE_IDX 1 +#define regVCN_CE_ERR_STATUS_LO_MMSCHD 0x0950 +#define regVCN_CE_ERR_STATUS_LO_MMSCHD_BASE_IDX 1 +#define regVCN_CE_ERR_STATUS_HI_MMSCHD 0x0951 +#define regVCN_CE_ERR_STATUS_HI_MMSCHD_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_LO_JPEG0S 0x0952 +#define regVCN_UE_ERR_STATUS_LO_JPEG0S_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_JPEG0S 0x0953 +#define regVCN_UE_ERR_STATUS_HI_JPEG0S_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_LO_JPEG0D 0x0954 +#define regVCN_UE_ERR_STATUS_LO_JPEG0D_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_JPEG0D 0x0955 +#define regVCN_UE_ERR_STATUS_HI_JPEG0D_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_LO_JPEG1S 0x0956 +#define regVCN_UE_ERR_STATUS_LO_JPEG1S_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_JPEG1S 0x0957 +#define regVCN_UE_ERR_STATUS_HI_JPEG1S_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_LO_JPEG1D 0x0958 +#define regVCN_UE_ERR_STATUS_LO_JPEG1D_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_JPEG1D 0x0959 +#define regVCN_UE_ERR_STATUS_HI_JPEG1D_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_LO_JPEG2S 0x095a +#define regVCN_UE_ERR_STATUS_LO_JPEG2S_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_JPEG2S 0x095b +#define regVCN_UE_ERR_STATUS_HI_JPEG2S_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_LO_JPEG2D 0x095c +#define regVCN_UE_ERR_STATUS_LO_JPEG2D_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_JPEG2D 0x095d +#define regVCN_UE_ERR_STATUS_HI_JPEG2D_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_LO_JPEG3S 0x095e +#define regVCN_UE_ERR_STATUS_LO_JPEG3S_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_JPEG3S 0x095f +#define regVCN_UE_ERR_STATUS_HI_JPEG3S_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_LO_JPEG3D 0x0960 +#define regVCN_UE_ERR_STATUS_LO_JPEG3D_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_JPEG3D 0x0961 +#define regVCN_UE_ERR_STATUS_HI_JPEG3D_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_LO_JPEG4S 0x0962 +#define regVCN_UE_ERR_STATUS_LO_JPEG4S_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_JPEG4S 0x0963 +#define regVCN_UE_ERR_STATUS_HI_JPEG4S_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_LO_JPEG4D 0x0964 +#define regVCN_UE_ERR_STATUS_LO_JPEG4D_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_JPEG4D 0x0965 +#define regVCN_UE_ERR_STATUS_HI_JPEG4D_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_LO_JPEG5S 0x0966 +#define regVCN_UE_ERR_STATUS_LO_JPEG5S_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_JPEG5S 0x0967 +#define regVCN_UE_ERR_STATUS_HI_JPEG5S_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_LO_JPEG5D 0x0968 +#define regVCN_UE_ERR_STATUS_LO_JPEG5D_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_JPEG5D 0x0969 +#define regVCN_UE_ERR_STATUS_HI_JPEG5D_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_LO_JPEG6S 0x096a +#define regVCN_UE_ERR_STATUS_LO_JPEG6S_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_JPEG6S 0x096b +#define regVCN_UE_ERR_STATUS_HI_JPEG6S_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_LO_JPEG6D 0x096c +#define regVCN_UE_ERR_STATUS_LO_JPEG6D_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_JPEG6D 0x096d +#define regVCN_UE_ERR_STATUS_HI_JPEG6D_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_LO_JPEG7S 0x096e +#define regVCN_UE_ERR_STATUS_LO_JPEG7S_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_JPEG7S 0x096f +#define regVCN_UE_ERR_STATUS_HI_JPEG7S_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_LO_JPEG7D 0x0970 +#define regVCN_UE_ERR_STATUS_LO_JPEG7D_BASE_IDX 1 +#define regVCN_UE_ERR_STATUS_HI_JPEG7D 0x0971 +#define regVCN_UE_ERR_STATUS_HI_JPEG7D_BASE_IDX 1 + +// addressBlock: aid_uvd0_uvd_jrbc1_uvd_jrbc_dec +// base address: 0x1e000 +#define regUVD_JRBC1_UVD_JRBC_RB_WPTR 0x0000 +#define regUVD_JRBC1_UVD_JRBC_RB_WPTR_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JRBC_RB_CNTL 0x0001 +#define regUVD_JRBC1_UVD_JRBC_RB_CNTL_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JRBC_IB_SIZE 0x0002 +#define regUVD_JRBC1_UVD_JRBC_IB_SIZE_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JRBC_URGENT_CNTL 0x0003 +#define regUVD_JRBC1_UVD_JRBC_URGENT_CNTL_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JRBC_RB_REF_DATA 0x0004 +#define regUVD_JRBC1_UVD_JRBC_RB_REF_DATA_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER 0x0005 +#define regUVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JRBC_SOFT_RESET 0x0008 +#define regUVD_JRBC1_UVD_JRBC_SOFT_RESET_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JRBC_STATUS 0x0009 +#define regUVD_JRBC1_UVD_JRBC_STATUS_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JRBC_RB_RPTR 0x000a +#define regUVD_JRBC1_UVD_JRBC_RB_RPTR_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JRBC_RB_BUF_STATUS 0x000b +#define regUVD_JRBC1_UVD_JRBC_RB_BUF_STATUS_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JRBC_IB_BUF_STATUS 0x000c +#define regUVD_JRBC1_UVD_JRBC_IB_BUF_STATUS_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JRBC_IB_SIZE_UPDATE 0x000d +#define regUVD_JRBC1_UVD_JRBC_IB_SIZE_UPDATE_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER 0x000e +#define regUVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JRBC_IB_REF_DATA 0x000f +#define regUVD_JRBC1_UVD_JRBC_IB_REF_DATA_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JPEG_PREEMPT_CMD 0x0010 +#define regUVD_JRBC1_UVD_JPEG_PREEMPT_CMD_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA0 0x0011 +#define regUVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA0_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA1 0x0012 +#define regUVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA1_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JRBC_RB_SIZE 0x0013 +#define regUVD_JRBC1_UVD_JRBC_RB_SIZE_BASE_IDX 0 +#define regUVD_JRBC1_UVD_JRBC_SCRATCH0 0x0014 +#define regUVD_JRBC1_UVD_JRBC_SCRATCH0_BASE_IDX 0 + + +// addressBlock: aid_uvd0_uvd_jrbc2_uvd_jrbc_dec +// base address: 0x1e100 +#define regUVD_JRBC2_UVD_JRBC_RB_WPTR 0x0040 +#define regUVD_JRBC2_UVD_JRBC_RB_WPTR_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JRBC_RB_CNTL 0x0041 +#define regUVD_JRBC2_UVD_JRBC_RB_CNTL_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JRBC_IB_SIZE 0x0042 +#define regUVD_JRBC2_UVD_JRBC_IB_SIZE_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JRBC_URGENT_CNTL 0x0043 +#define regUVD_JRBC2_UVD_JRBC_URGENT_CNTL_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JRBC_RB_REF_DATA 0x0044 +#define regUVD_JRBC2_UVD_JRBC_RB_REF_DATA_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER 0x0045 +#define regUVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JRBC_SOFT_RESET 0x0048 +#define regUVD_JRBC2_UVD_JRBC_SOFT_RESET_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JRBC_STATUS 0x0049 +#define regUVD_JRBC2_UVD_JRBC_STATUS_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JRBC_RB_RPTR 0x004a +#define regUVD_JRBC2_UVD_JRBC_RB_RPTR_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JRBC_RB_BUF_STATUS 0x004b +#define regUVD_JRBC2_UVD_JRBC_RB_BUF_STATUS_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JRBC_IB_BUF_STATUS 0x004c +#define regUVD_JRBC2_UVD_JRBC_IB_BUF_STATUS_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JRBC_IB_SIZE_UPDATE 0x004d +#define regUVD_JRBC2_UVD_JRBC_IB_SIZE_UPDATE_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER 0x004e +#define regUVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JRBC_IB_REF_DATA 0x004f +#define regUVD_JRBC2_UVD_JRBC_IB_REF_DATA_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JPEG_PREEMPT_CMD 0x0050 +#define regUVD_JRBC2_UVD_JPEG_PREEMPT_CMD_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA0 0x0051 +#define regUVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA0_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA1 0x0052 +#define regUVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA1_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JRBC_RB_SIZE 0x0053 +#define regUVD_JRBC2_UVD_JRBC_RB_SIZE_BASE_IDX 0 +#define regUVD_JRBC2_UVD_JRBC_SCRATCH0 0x0054 +#define regUVD_JRBC2_UVD_JRBC_SCRATCH0_BASE_IDX 0 + + +// addressBlock: aid_uvd0_uvd_jrbc3_uvd_jrbc_dec +// base address: 0x1e200 +#define regUVD_JRBC3_UVD_JRBC_RB_WPTR 0x0080 +#define regUVD_JRBC3_UVD_JRBC_RB_WPTR_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JRBC_RB_CNTL 0x0081 +#define regUVD_JRBC3_UVD_JRBC_RB_CNTL_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JRBC_IB_SIZE 0x0082 +#define regUVD_JRBC3_UVD_JRBC_IB_SIZE_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JRBC_URGENT_CNTL 0x0083 +#define regUVD_JRBC3_UVD_JRBC_URGENT_CNTL_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JRBC_RB_REF_DATA 0x0084 +#define regUVD_JRBC3_UVD_JRBC_RB_REF_DATA_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER 0x0085 +#define regUVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JRBC_SOFT_RESET 0x0088 +#define regUVD_JRBC3_UVD_JRBC_SOFT_RESET_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JRBC_STATUS 0x0089 +#define regUVD_JRBC3_UVD_JRBC_STATUS_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JRBC_RB_RPTR 0x008a +#define regUVD_JRBC3_UVD_JRBC_RB_RPTR_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JRBC_RB_BUF_STATUS 0x008b +#define regUVD_JRBC3_UVD_JRBC_RB_BUF_STATUS_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JRBC_IB_BUF_STATUS 0x008c +#define regUVD_JRBC3_UVD_JRBC_IB_BUF_STATUS_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JRBC_IB_SIZE_UPDATE 0x008d +#define regUVD_JRBC3_UVD_JRBC_IB_SIZE_UPDATE_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER 0x008e +#define regUVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JRBC_IB_REF_DATA 0x008f +#define regUVD_JRBC3_UVD_JRBC_IB_REF_DATA_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JPEG_PREEMPT_CMD 0x0090 +#define regUVD_JRBC3_UVD_JPEG_PREEMPT_CMD_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA0 0x0091 +#define regUVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA0_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA1 0x0092 +#define regUVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA1_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JRBC_RB_SIZE 0x0093 +#define regUVD_JRBC3_UVD_JRBC_RB_SIZE_BASE_IDX 0 +#define regUVD_JRBC3_UVD_JRBC_SCRATCH0 0x0094 +#define regUVD_JRBC3_UVD_JRBC_SCRATCH0_BASE_IDX 0 + + +// addressBlock: aid_uvd0_uvd_jrbc4_uvd_jrbc_dec +// base address: 0x1e300 +#define regUVD_JRBC4_UVD_JRBC_RB_WPTR 0x00c0 +#define regUVD_JRBC4_UVD_JRBC_RB_WPTR_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JRBC_RB_CNTL 0x00c1 +#define regUVD_JRBC4_UVD_JRBC_RB_CNTL_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JRBC_IB_SIZE 0x00c2 +#define regUVD_JRBC4_UVD_JRBC_IB_SIZE_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JRBC_URGENT_CNTL 0x00c3 +#define regUVD_JRBC4_UVD_JRBC_URGENT_CNTL_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JRBC_RB_REF_DATA 0x00c4 +#define regUVD_JRBC4_UVD_JRBC_RB_REF_DATA_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER 0x00c5 +#define regUVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JRBC_SOFT_RESET 0x00c8 +#define regUVD_JRBC4_UVD_JRBC_SOFT_RESET_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JRBC_STATUS 0x00c9 +#define regUVD_JRBC4_UVD_JRBC_STATUS_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JRBC_RB_RPTR 0x00ca +#define regUVD_JRBC4_UVD_JRBC_RB_RPTR_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JRBC_RB_BUF_STATUS 0x00cb +#define regUVD_JRBC4_UVD_JRBC_RB_BUF_STATUS_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JRBC_IB_BUF_STATUS 0x00cc +#define regUVD_JRBC4_UVD_JRBC_IB_BUF_STATUS_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JRBC_IB_SIZE_UPDATE 0x00cd +#define regUVD_JRBC4_UVD_JRBC_IB_SIZE_UPDATE_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER 0x00ce +#define regUVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JRBC_IB_REF_DATA 0x00cf +#define regUVD_JRBC4_UVD_JRBC_IB_REF_DATA_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JPEG_PREEMPT_CMD 0x00d0 +#define regUVD_JRBC4_UVD_JPEG_PREEMPT_CMD_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA0 0x00d1 +#define regUVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA0_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA1 0x00d2 +#define regUVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA1_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JRBC_RB_SIZE 0x00d3 +#define regUVD_JRBC4_UVD_JRBC_RB_SIZE_BASE_IDX 0 +#define regUVD_JRBC4_UVD_JRBC_SCRATCH0 0x00d4 +#define regUVD_JRBC4_UVD_JRBC_SCRATCH0_BASE_IDX 0 + + +// addressBlock: aid_uvd0_uvd_jrbc5_uvd_jrbc_dec +// base address: 0x1e400 +#define regUVD_JRBC5_UVD_JRBC_RB_WPTR 0x0100 +#define regUVD_JRBC5_UVD_JRBC_RB_WPTR_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JRBC_RB_CNTL 0x0101 +#define regUVD_JRBC5_UVD_JRBC_RB_CNTL_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JRBC_IB_SIZE 0x0102 +#define regUVD_JRBC5_UVD_JRBC_IB_SIZE_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JRBC_URGENT_CNTL 0x0103 +#define regUVD_JRBC5_UVD_JRBC_URGENT_CNTL_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JRBC_RB_REF_DATA 0x0104 +#define regUVD_JRBC5_UVD_JRBC_RB_REF_DATA_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER 0x0105 +#define regUVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JRBC_SOFT_RESET 0x0108 +#define regUVD_JRBC5_UVD_JRBC_SOFT_RESET_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JRBC_STATUS 0x0109 +#define regUVD_JRBC5_UVD_JRBC_STATUS_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JRBC_RB_RPTR 0x010a +#define regUVD_JRBC5_UVD_JRBC_RB_RPTR_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JRBC_RB_BUF_STATUS 0x010b +#define regUVD_JRBC5_UVD_JRBC_RB_BUF_STATUS_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JRBC_IB_BUF_STATUS 0x010c +#define regUVD_JRBC5_UVD_JRBC_IB_BUF_STATUS_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JRBC_IB_SIZE_UPDATE 0x010d +#define regUVD_JRBC5_UVD_JRBC_IB_SIZE_UPDATE_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER 0x010e +#define regUVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JRBC_IB_REF_DATA 0x010f +#define regUVD_JRBC5_UVD_JRBC_IB_REF_DATA_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JPEG_PREEMPT_CMD 0x0110 +#define regUVD_JRBC5_UVD_JPEG_PREEMPT_CMD_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA0 0x0111 +#define regUVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA0_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA1 0x0112 +#define regUVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA1_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JRBC_RB_SIZE 0x0113 +#define regUVD_JRBC5_UVD_JRBC_RB_SIZE_BASE_IDX 0 +#define regUVD_JRBC5_UVD_JRBC_SCRATCH0 0x0114 +#define regUVD_JRBC5_UVD_JRBC_SCRATCH0_BASE_IDX 0 + + +// addressBlock: aid_uvd0_uvd_jrbc6_uvd_jrbc_dec +// base address: 0x1e500 +#define regUVD_JRBC6_UVD_JRBC_RB_WPTR 0x0140 +#define regUVD_JRBC6_UVD_JRBC_RB_WPTR_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JRBC_RB_CNTL 0x0141 +#define regUVD_JRBC6_UVD_JRBC_RB_CNTL_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JRBC_IB_SIZE 0x0142 +#define regUVD_JRBC6_UVD_JRBC_IB_SIZE_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JRBC_URGENT_CNTL 0x0143 +#define regUVD_JRBC6_UVD_JRBC_URGENT_CNTL_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JRBC_RB_REF_DATA 0x0144 +#define regUVD_JRBC6_UVD_JRBC_RB_REF_DATA_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER 0x0145 +#define regUVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JRBC_SOFT_RESET 0x0148 +#define regUVD_JRBC6_UVD_JRBC_SOFT_RESET_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JRBC_STATUS 0x0149 +#define regUVD_JRBC6_UVD_JRBC_STATUS_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JRBC_RB_RPTR 0x014a +#define regUVD_JRBC6_UVD_JRBC_RB_RPTR_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JRBC_RB_BUF_STATUS 0x014b +#define regUVD_JRBC6_UVD_JRBC_RB_BUF_STATUS_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JRBC_IB_BUF_STATUS 0x014c +#define regUVD_JRBC6_UVD_JRBC_IB_BUF_STATUS_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JRBC_IB_SIZE_UPDATE 0x014d +#define regUVD_JRBC6_UVD_JRBC_IB_SIZE_UPDATE_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER 0x014e +#define regUVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JRBC_IB_REF_DATA 0x014f +#define regUVD_JRBC6_UVD_JRBC_IB_REF_DATA_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JPEG_PREEMPT_CMD 0x0150 +#define regUVD_JRBC6_UVD_JPEG_PREEMPT_CMD_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA0 0x0151 +#define regUVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA0_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA1 0x0152 +#define regUVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA1_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JRBC_RB_SIZE 0x0153 +#define regUVD_JRBC6_UVD_JRBC_RB_SIZE_BASE_IDX 0 +#define regUVD_JRBC6_UVD_JRBC_SCRATCH0 0x0154 +#define regUVD_JRBC6_UVD_JRBC_SCRATCH0_BASE_IDX 0 + + +// addressBlock: aid_uvd0_uvd_jrbc7_uvd_jrbc_dec +// base address: 0x1e600 +#define regUVD_JRBC7_UVD_JRBC_RB_WPTR 0x0180 +#define regUVD_JRBC7_UVD_JRBC_RB_WPTR_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JRBC_RB_CNTL 0x0181 +#define regUVD_JRBC7_UVD_JRBC_RB_CNTL_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JRBC_IB_SIZE 0x0182 +#define regUVD_JRBC7_UVD_JRBC_IB_SIZE_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JRBC_URGENT_CNTL 0x0183 +#define regUVD_JRBC7_UVD_JRBC_URGENT_CNTL_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JRBC_RB_REF_DATA 0x0184 +#define regUVD_JRBC7_UVD_JRBC_RB_REF_DATA_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER 0x0185 +#define regUVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JRBC_SOFT_RESET 0x0188 +#define regUVD_JRBC7_UVD_JRBC_SOFT_RESET_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JRBC_STATUS 0x0189 +#define regUVD_JRBC7_UVD_JRBC_STATUS_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JRBC_RB_RPTR 0x018a +#define regUVD_JRBC7_UVD_JRBC_RB_RPTR_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JRBC_RB_BUF_STATUS 0x018b +#define regUVD_JRBC7_UVD_JRBC_RB_BUF_STATUS_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JRBC_IB_BUF_STATUS 0x018c +#define regUVD_JRBC7_UVD_JRBC_IB_BUF_STATUS_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JRBC_IB_SIZE_UPDATE 0x018d +#define regUVD_JRBC7_UVD_JRBC_IB_SIZE_UPDATE_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER 0x018e +#define regUVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JRBC_IB_REF_DATA 0x018f +#define regUVD_JRBC7_UVD_JRBC_IB_REF_DATA_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JPEG_PREEMPT_CMD 0x0190 +#define regUVD_JRBC7_UVD_JPEG_PREEMPT_CMD_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA0 0x0191 +#define regUVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA0_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA1 0x0192 +#define regUVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA1_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JRBC_RB_SIZE 0x0193 +#define regUVD_JRBC7_UVD_JRBC_RB_SIZE_BASE_IDX 0 +#define regUVD_JRBC7_UVD_JRBC_SCRATCH0 0x0194 +#define regUVD_JRBC7_UVD_JRBC_SCRATCH0_BASE_IDX 0 + + +// addressBlock: aid_uvd0_uvd_jmi1_uvd_jmi_dec +// base address: 0x1e080 +#define regUVD_JMI1_UVD_JPEG_DEC_PF_CTRL 0x0020 +#define regUVD_JMI1_UVD_JPEG_DEC_PF_CTRL_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JRBC_CTRL 0x0021 +#define regUVD_JMI1_UVD_LMI_JRBC_CTRL_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JPEG_CTRL 0x0022 +#define regUVD_JMI1_UVD_LMI_JPEG_CTRL_BASE_IDX 0 +#define regUVD_JMI1_JPEG_LMI_DROP 0x0023 +#define regUVD_JMI1_JPEG_LMI_DROP_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JRBC_IB_VMID 0x0024 +#define regUVD_JMI1_UVD_LMI_JRBC_IB_VMID_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JRBC_RB_VMID 0x0025 +#define regUVD_JMI1_UVD_LMI_JRBC_RB_VMID_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JPEG_VMID 0x0026 +#define regUVD_JMI1_UVD_LMI_JPEG_VMID_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW 0x0027 +#define regUVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH 0x0028 +#define regUVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_LOW 0x0029 +#define regUVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH 0x002a +#define regUVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW 0x002b +#define regUVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH 0x002c +#define regUVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JPEG_PREEMPT_VMID 0x002d +#define regUVD_JMI1_UVD_LMI_JPEG_PREEMPT_VMID_BASE_IDX 0 +#define regUVD_JMI1_UVD_JMI_DEC_SWAP_CNTL 0x002e +#define regUVD_JMI1_UVD_JMI_DEC_SWAP_CNTL_BASE_IDX 0 +#define regUVD_JMI1_UVD_JMI_ATOMIC_CNTL 0x002f +#define regUVD_JMI1_UVD_JMI_ATOMIC_CNTL_BASE_IDX 0 +#define regUVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW 0x0030 +#define regUVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH 0x0031 +#define regUVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_LOW 0x0032 +#define regUVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH 0x0033 +#define regUVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW 0x0034 +#define regUVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH 0x0035 +#define regUVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_LOW 0x0036 +#define regUVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH 0x0037 +#define regUVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW 0x0038 +#define regUVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH 0x0039 +#define regUVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI1_UVD_JMI_ATOMIC_CNTL2 0x003d +#define regUVD_JMI1_UVD_JMI_ATOMIC_CNTL2_BASE_IDX 0 + + +// addressBlock: aid_uvd0_uvd_jmi2_uvd_jmi_dec +// base address: 0x1e180 +#define regUVD_JMI2_UVD_JPEG_DEC_PF_CTRL 0x0060 +#define regUVD_JMI2_UVD_JPEG_DEC_PF_CTRL_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JRBC_CTRL 0x0061 +#define regUVD_JMI2_UVD_LMI_JRBC_CTRL_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JPEG_CTRL 0x0062 +#define regUVD_JMI2_UVD_LMI_JPEG_CTRL_BASE_IDX 0 +#define regUVD_JMI2_JPEG_LMI_DROP 0x0063 +#define regUVD_JMI2_JPEG_LMI_DROP_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JRBC_IB_VMID 0x0064 +#define regUVD_JMI2_UVD_LMI_JRBC_IB_VMID_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JRBC_RB_VMID 0x0065 +#define regUVD_JMI2_UVD_LMI_JRBC_RB_VMID_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JPEG_VMID 0x0066 +#define regUVD_JMI2_UVD_LMI_JPEG_VMID_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW 0x0067 +#define regUVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH 0x0068 +#define regUVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_LOW 0x0069 +#define regUVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH 0x006a +#define regUVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW 0x006b +#define regUVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH 0x006c +#define regUVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JPEG_PREEMPT_VMID 0x006d +#define regUVD_JMI2_UVD_LMI_JPEG_PREEMPT_VMID_BASE_IDX 0 +#define regUVD_JMI2_UVD_JMI_DEC_SWAP_CNTL 0x006e +#define regUVD_JMI2_UVD_JMI_DEC_SWAP_CNTL_BASE_IDX 0 +#define regUVD_JMI2_UVD_JMI_ATOMIC_CNTL 0x006f +#define regUVD_JMI2_UVD_JMI_ATOMIC_CNTL_BASE_IDX 0 +#define regUVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW 0x0070 +#define regUVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH 0x0071 +#define regUVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_LOW 0x0072 +#define regUVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH 0x0073 +#define regUVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW 0x0074 +#define regUVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH 0x0075 +#define regUVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_LOW 0x0076 +#define regUVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH 0x0077 +#define regUVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW 0x0078 +#define regUVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH 0x0079 +#define regUVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI2_UVD_JMI_ATOMIC_CNTL2 0x007d +#define regUVD_JMI2_UVD_JMI_ATOMIC_CNTL2_BASE_IDX 0 + + +// addressBlock: aid_uvd0_uvd_jmi3_uvd_jmi_dec +// base address: 0x1e280 +#define regUVD_JMI3_UVD_JPEG_DEC_PF_CTRL 0x00a0 +#define regUVD_JMI3_UVD_JPEG_DEC_PF_CTRL_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JRBC_CTRL 0x00a1 +#define regUVD_JMI3_UVD_LMI_JRBC_CTRL_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JPEG_CTRL 0x00a2 +#define regUVD_JMI3_UVD_LMI_JPEG_CTRL_BASE_IDX 0 +#define regUVD_JMI3_JPEG_LMI_DROP 0x00a3 +#define regUVD_JMI3_JPEG_LMI_DROP_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JRBC_IB_VMID 0x00a4 +#define regUVD_JMI3_UVD_LMI_JRBC_IB_VMID_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JRBC_RB_VMID 0x00a5 +#define regUVD_JMI3_UVD_LMI_JRBC_RB_VMID_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JPEG_VMID 0x00a6 +#define regUVD_JMI3_UVD_LMI_JPEG_VMID_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW 0x00a7 +#define regUVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH 0x00a8 +#define regUVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_LOW 0x00a9 +#define regUVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH 0x00aa +#define regUVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW 0x00ab +#define regUVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH 0x00ac +#define regUVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JPEG_PREEMPT_VMID 0x00ad +#define regUVD_JMI3_UVD_LMI_JPEG_PREEMPT_VMID_BASE_IDX 0 +#define regUVD_JMI3_UVD_JMI_DEC_SWAP_CNTL 0x00ae +#define regUVD_JMI3_UVD_JMI_DEC_SWAP_CNTL_BASE_IDX 0 +#define regUVD_JMI3_UVD_JMI_ATOMIC_CNTL 0x00af +#define regUVD_JMI3_UVD_JMI_ATOMIC_CNTL_BASE_IDX 0 +#define regUVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW 0x00b0 +#define regUVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH 0x00b1 +#define regUVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_LOW 0x00b2 +#define regUVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH 0x00b3 +#define regUVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW 0x00b4 +#define regUVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH 0x00b5 +#define regUVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_LOW 0x00b6 +#define regUVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH 0x00b7 +#define regUVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW 0x00b8 +#define regUVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH 0x00b9 +#define regUVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI3_UVD_JMI_ATOMIC_CNTL2 0x00bd +#define regUVD_JMI3_UVD_JMI_ATOMIC_CNTL2_BASE_IDX 0 + + +// addressBlock: aid_uvd0_uvd_jmi4_uvd_jmi_dec +// base address: 0x1e380 +#define regUVD_JMI4_UVD_JPEG_DEC_PF_CTRL 0x00e0 +#define regUVD_JMI4_UVD_JPEG_DEC_PF_CTRL_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JRBC_CTRL 0x00e1 +#define regUVD_JMI4_UVD_LMI_JRBC_CTRL_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JPEG_CTRL 0x00e2 +#define regUVD_JMI4_UVD_LMI_JPEG_CTRL_BASE_IDX 0 +#define regUVD_JMI4_JPEG_LMI_DROP 0x00e3 +#define regUVD_JMI4_JPEG_LMI_DROP_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JRBC_IB_VMID 0x00e4 +#define regUVD_JMI4_UVD_LMI_JRBC_IB_VMID_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JRBC_RB_VMID 0x00e5 +#define regUVD_JMI4_UVD_LMI_JRBC_RB_VMID_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JPEG_VMID 0x00e6 +#define regUVD_JMI4_UVD_LMI_JPEG_VMID_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW 0x00e7 +#define regUVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH 0x00e8 +#define regUVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_LOW 0x00e9 +#define regUVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH 0x00ea +#define regUVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW 0x00eb +#define regUVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH 0x00ec +#define regUVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JPEG_PREEMPT_VMID 0x00ed +#define regUVD_JMI4_UVD_LMI_JPEG_PREEMPT_VMID_BASE_IDX 0 +#define regUVD_JMI4_UVD_JMI_DEC_SWAP_CNTL 0x00ee +#define regUVD_JMI4_UVD_JMI_DEC_SWAP_CNTL_BASE_IDX 0 +#define regUVD_JMI4_UVD_JMI_ATOMIC_CNTL 0x00ef +#define regUVD_JMI4_UVD_JMI_ATOMIC_CNTL_BASE_IDX 0 +#define regUVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW 0x00f0 +#define regUVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH 0x00f1 +#define regUVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_LOW 0x00f2 +#define regUVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH 0x00f3 +#define regUVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW 0x00f4 +#define regUVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH 0x00f5 +#define regUVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_LOW 0x00f6 +#define regUVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH 0x00f7 +#define regUVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW 0x00f8 +#define regUVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH 0x00f9 +#define regUVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI4_UVD_JMI_ATOMIC_CNTL2 0x00fd +#define regUVD_JMI4_UVD_JMI_ATOMIC_CNTL2_BASE_IDX 0 + + +// addressBlock: aid_uvd0_uvd_jmi5_uvd_jmi_dec +// base address: 0x1e480 +#define regUVD_JMI5_UVD_JPEG_DEC_PF_CTRL 0x0120 +#define regUVD_JMI5_UVD_JPEG_DEC_PF_CTRL_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JRBC_CTRL 0x0121 +#define regUVD_JMI5_UVD_LMI_JRBC_CTRL_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JPEG_CTRL 0x0122 +#define regUVD_JMI5_UVD_LMI_JPEG_CTRL_BASE_IDX 0 +#define regUVD_JMI5_JPEG_LMI_DROP 0x0123 +#define regUVD_JMI5_JPEG_LMI_DROP_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JRBC_IB_VMID 0x0124 +#define regUVD_JMI5_UVD_LMI_JRBC_IB_VMID_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JRBC_RB_VMID 0x0125 +#define regUVD_JMI5_UVD_LMI_JRBC_RB_VMID_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JPEG_VMID 0x0126 +#define regUVD_JMI5_UVD_LMI_JPEG_VMID_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW 0x0127 +#define regUVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH 0x0128 +#define regUVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_LOW 0x0129 +#define regUVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH 0x012a +#define regUVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW 0x012b +#define regUVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH 0x012c +#define regUVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JPEG_PREEMPT_VMID 0x012d +#define regUVD_JMI5_UVD_LMI_JPEG_PREEMPT_VMID_BASE_IDX 0 +#define regUVD_JMI5_UVD_JMI_DEC_SWAP_CNTL 0x012e +#define regUVD_JMI5_UVD_JMI_DEC_SWAP_CNTL_BASE_IDX 0 +#define regUVD_JMI5_UVD_JMI_ATOMIC_CNTL 0x012f +#define regUVD_JMI5_UVD_JMI_ATOMIC_CNTL_BASE_IDX 0 +#define regUVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW 0x0130 +#define regUVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH 0x0131 +#define regUVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_LOW 0x0132 +#define regUVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH 0x0133 +#define regUVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW 0x0134 +#define regUVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH 0x0135 +#define regUVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_LOW 0x0136 +#define regUVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH 0x0137 +#define regUVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW 0x0138 +#define regUVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH 0x0139 +#define regUVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI5_UVD_JMI_ATOMIC_CNTL2 0x013d +#define regUVD_JMI5_UVD_JMI_ATOMIC_CNTL2_BASE_IDX 0 + + +// addressBlock: aid_uvd0_uvd_jmi6_uvd_jmi_dec +// base address: 0x1e580 +#define regUVD_JMI6_UVD_JPEG_DEC_PF_CTRL 0x0160 +#define regUVD_JMI6_UVD_JPEG_DEC_PF_CTRL_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JRBC_CTRL 0x0161 +#define regUVD_JMI6_UVD_LMI_JRBC_CTRL_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JPEG_CTRL 0x0162 +#define regUVD_JMI6_UVD_LMI_JPEG_CTRL_BASE_IDX 0 +#define regUVD_JMI6_JPEG_LMI_DROP 0x0163 +#define regUVD_JMI6_JPEG_LMI_DROP_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JRBC_IB_VMID 0x0164 +#define regUVD_JMI6_UVD_LMI_JRBC_IB_VMID_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JRBC_RB_VMID 0x0165 +#define regUVD_JMI6_UVD_LMI_JRBC_RB_VMID_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JPEG_VMID 0x0166 +#define regUVD_JMI6_UVD_LMI_JPEG_VMID_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW 0x0167 +#define regUVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH 0x0168 +#define regUVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_LOW 0x0169 +#define regUVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH 0x016a +#define regUVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW 0x016b +#define regUVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH 0x016c +#define regUVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JPEG_PREEMPT_VMID 0x016d +#define regUVD_JMI6_UVD_LMI_JPEG_PREEMPT_VMID_BASE_IDX 0 +#define regUVD_JMI6_UVD_JMI_DEC_SWAP_CNTL 0x016e +#define regUVD_JMI6_UVD_JMI_DEC_SWAP_CNTL_BASE_IDX 0 +#define regUVD_JMI6_UVD_JMI_ATOMIC_CNTL 0x016f +#define regUVD_JMI6_UVD_JMI_ATOMIC_CNTL_BASE_IDX 0 +#define regUVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW 0x0170 +#define regUVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH 0x0171 +#define regUVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_LOW 0x0172 +#define regUVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH 0x0173 +#define regUVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW 0x0174 +#define regUVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH 0x0175 +#define regUVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_LOW 0x0176 +#define regUVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH 0x0177 +#define regUVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW 0x0178 +#define regUVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH 0x0179 +#define regUVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI6_UVD_JMI_ATOMIC_CNTL2 0x017d +#define regUVD_JMI6_UVD_JMI_ATOMIC_CNTL2_BASE_IDX 0 + + +// addressBlock: aid_uvd0_uvd_jmi7_uvd_jmi_dec +// base address: 0x1e680 +#define regUVD_JMI7_UVD_JPEG_DEC_PF_CTRL 0x01a0 +#define regUVD_JMI7_UVD_JPEG_DEC_PF_CTRL_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JRBC_CTRL 0x01a1 +#define regUVD_JMI7_UVD_LMI_JRBC_CTRL_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JPEG_CTRL 0x01a2 +#define regUVD_JMI7_UVD_LMI_JPEG_CTRL_BASE_IDX 0 +#define regUVD_JMI7_JPEG_LMI_DROP 0x01a3 +#define regUVD_JMI7_JPEG_LMI_DROP_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JRBC_IB_VMID 0x01a4 +#define regUVD_JMI7_UVD_LMI_JRBC_IB_VMID_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JRBC_RB_VMID 0x01a5 +#define regUVD_JMI7_UVD_LMI_JRBC_RB_VMID_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JPEG_VMID 0x01a6 +#define regUVD_JMI7_UVD_LMI_JPEG_VMID_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW 0x01a7 +#define regUVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH 0x01a8 +#define regUVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_LOW 0x01a9 +#define regUVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH 0x01aa +#define regUVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW 0x01ab +#define regUVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH 0x01ac +#define regUVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JPEG_PREEMPT_VMID 0x01ad +#define regUVD_JMI7_UVD_LMI_JPEG_PREEMPT_VMID_BASE_IDX 0 +#define regUVD_JMI7_UVD_JMI_DEC_SWAP_CNTL 0x01ae +#define regUVD_JMI7_UVD_JMI_DEC_SWAP_CNTL_BASE_IDX 0 +#define regUVD_JMI7_UVD_JMI_ATOMIC_CNTL 0x01af +#define regUVD_JMI7_UVD_JMI_ATOMIC_CNTL_BASE_IDX 0 +#define regUVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW 0x01b0 +#define regUVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH 0x01b1 +#define regUVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_LOW 0x01b2 +#define regUVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH 0x01b3 +#define regUVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW 0x01b4 +#define regUVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH 0x01b5 +#define regUVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_LOW 0x01b6 +#define regUVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH 0x01b7 +#define regUVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW 0x01b8 +#define regUVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW_BASE_IDX 0 +#define regUVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH 0x01b9 +#define regUVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH_BASE_IDX 0 +#define regUVD_JMI7_UVD_JMI_ATOMIC_CNTL2 0x01bd +#define regUVD_JMI7_UVD_JMI_ATOMIC_CNTL2_BASE_IDX 0 + + +// addressBlock: uvdctxind +// base address: 0x0 +#define ixUVD_CGC_MEM_CTRL 0x0000 +#define ixUVD_CGC_CTRL2 0x0001 +#define ixUVD_CGC_MEM_DS_CTRL 0x0002 +#define ixUVD_CGC_MEM_SD_CTRL 0x0003 +#define ixUVD_SW_SCRATCH_00 0x0004 +#define ixUVD_SW_SCRATCH_01 0x0005 +#define ixUVD_SW_SCRATCH_02 0x0006 +#define ixUVD_SW_SCRATCH_03 0x0007 +#define ixUVD_SW_SCRATCH_04 0x0008 +#define ixUVD_SW_SCRATCH_05 0x0009 +#define ixUVD_SW_SCRATCH_06 0x000a +#define ixUVD_SW_SCRATCH_07 0x000b +#define ixUVD_SW_SCRATCH_08 0x000c +#define ixUVD_SW_SCRATCH_09 0x000d +#define ixUVD_SW_SCRATCH_10 0x000e +#define ixUVD_SW_SCRATCH_11 0x000f +#define ixUVD_SW_SCRATCH_12 0x0010 +#define ixUVD_SW_SCRATCH_13 0x0011 +#define ixUVD_SW_SCRATCH_14 0x0012 +#define ixUVD_SW_SCRATCH_15 0x0013 +#define ixUVD_IH_SEM_CTRL 0x001e + + +// addressBlock: lmi_adp_indirect +// base address: 0x0 +#define ixUVD_LMI_CRC0 0x0000 +#define ixUVD_LMI_CRC1 0x0001 +#define ixUVD_LMI_CRC2 0x0002 +#define ixUVD_LMI_CRC3 0x0003 +#define ixUVD_LMI_CRC10 0x000a +#define ixUVD_LMI_CRC11 0x000b +#define ixUVD_LMI_CRC12 0x000c +#define ixUVD_LMI_CRC13 0x000d +#define ixUVD_LMI_CRC14 0x000e +#define ixUVD_LMI_CRC15 0x000f +#define ixUVD_LMI_SWAP_CNTL2 0x0029 +#define ixUVD_MEMCHECK_SYS_INT_EN 0x0134 +#define ixUVD_MEMCHECK_SYS_INT_STAT 0x0135 +#define ixUVD_MEMCHECK_SYS_INT_ACK 0x0136 +#define ixUVD_MEMCHECK_VCPU_INT_EN 0x0137 +#define ixUVD_MEMCHECK_VCPU_INT_STAT 0x0138 +#define ixUVD_MEMCHECK_VCPU_INT_ACK 0x0139 +#define ixUVD_MEMCHECK2_SYS_INT_STAT 0x0140 +#define ixUVD_MEMCHECK2_SYS_INT_ACK 0x0141 +#define ixUVD_MEMCHECK2_VCPU_INT_STAT 0x0142 +#define ixUVD_MEMCHECK2_VCPU_INT_ACK 0x0143 + + +#endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_sh_mask.h new file mode 100644 index 0000000000000000000000000000000000000000..5bd8111bf04adb2d00a69bf56f3ee3a12d1297cf --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/vcn/vcn_4_0_3_sh_mask.h @@ -0,0 +1,10919 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +#ifndef _vcn_4_0_3_SH_MASK_HEADER +#define _vcn_4_0_3_SH_MASK_HEADER + + +// addressBlock: aid_uvd0_uvddec +//UVD_TOP_CTRL +#define UVD_TOP_CTRL__STANDARD__SHIFT 0x0 +#define UVD_TOP_CTRL__STD_VERSION__SHIFT 0x4 +#define UVD_TOP_CTRL__STANDARD_MASK 0x0000000FL +#define UVD_TOP_CTRL__STD_VERSION_MASK 0x00000010L +//UVD_CGC_GATE +#define UVD_CGC_GATE__SYS__SHIFT 0x0 +#define UVD_CGC_GATE__UDEC__SHIFT 0x1 +#define UVD_CGC_GATE__MPEG2__SHIFT 0x2 +#define UVD_CGC_GATE__REGS__SHIFT 0x3 +#define UVD_CGC_GATE__RBC__SHIFT 0x4 +#define UVD_CGC_GATE__LMI_MC__SHIFT 0x5 +#define UVD_CGC_GATE__LMI_UMC__SHIFT 0x6 +#define UVD_CGC_GATE__IDCT__SHIFT 0x7 +#define UVD_CGC_GATE__MPRD__SHIFT 0x8 +#define UVD_CGC_GATE__MPC__SHIFT 0x9 +#define UVD_CGC_GATE__LBSI__SHIFT 0xa +#define UVD_CGC_GATE__LRBBM__SHIFT 0xb +#define UVD_CGC_GATE__UDEC_RE__SHIFT 0xc +#define UVD_CGC_GATE__UDEC_CM__SHIFT 0xd +#define UVD_CGC_GATE__UDEC_IT__SHIFT 0xe +#define UVD_CGC_GATE__UDEC_DB__SHIFT 0xf +#define UVD_CGC_GATE__UDEC_MP__SHIFT 0x10 +#define UVD_CGC_GATE__WCB__SHIFT 0x11 +#define UVD_CGC_GATE__VCPU__SHIFT 0x12 +#define UVD_CGC_GATE__MMSCH__SHIFT 0x14 +#define UVD_CGC_GATE__LCM0__SHIFT 0x15 +#define UVD_CGC_GATE__LCM1__SHIFT 0x16 +#define UVD_CGC_GATE__MIF__SHIFT 0x17 +#define UVD_CGC_GATE__VREG__SHIFT 0x18 +#define UVD_CGC_GATE__PE__SHIFT 0x19 +#define UVD_CGC_GATE__PPU__SHIFT 0x1a +#define UVD_CGC_GATE__SYS_MASK 0x00000001L +#define UVD_CGC_GATE__UDEC_MASK 0x00000002L +#define UVD_CGC_GATE__MPEG2_MASK 0x00000004L +#define UVD_CGC_GATE__REGS_MASK 0x00000008L +#define UVD_CGC_GATE__RBC_MASK 0x00000010L +#define UVD_CGC_GATE__LMI_MC_MASK 0x00000020L +#define UVD_CGC_GATE__LMI_UMC_MASK 0x00000040L +#define UVD_CGC_GATE__IDCT_MASK 0x00000080L +#define UVD_CGC_GATE__MPRD_MASK 0x00000100L +#define UVD_CGC_GATE__MPC_MASK 0x00000200L +#define UVD_CGC_GATE__LBSI_MASK 0x00000400L +#define UVD_CGC_GATE__LRBBM_MASK 0x00000800L +#define UVD_CGC_GATE__UDEC_RE_MASK 0x00001000L +#define UVD_CGC_GATE__UDEC_CM_MASK 0x00002000L +#define UVD_CGC_GATE__UDEC_IT_MASK 0x00004000L +#define UVD_CGC_GATE__UDEC_DB_MASK 0x00008000L +#define UVD_CGC_GATE__UDEC_MP_MASK 0x00010000L +#define UVD_CGC_GATE__WCB_MASK 0x00020000L +#define UVD_CGC_GATE__VCPU_MASK 0x00040000L +#define UVD_CGC_GATE__MMSCH_MASK 0x00100000L +#define UVD_CGC_GATE__LCM0_MASK 0x00200000L +#define UVD_CGC_GATE__LCM1_MASK 0x00400000L +#define UVD_CGC_GATE__MIF_MASK 0x00800000L +#define UVD_CGC_GATE__VREG_MASK 0x01000000L +#define UVD_CGC_GATE__PE_MASK 0x02000000L +#define UVD_CGC_GATE__PPU_MASK 0x04000000L +//UVD_CGC_CTRL +#define UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT 0x0 +#define UVD_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT 0x2 +#define UVD_CGC_CTRL__CLK_OFF_DELAY__SHIFT 0x6 +#define UVD_CGC_CTRL__UDEC_RE_MODE__SHIFT 0xb +#define UVD_CGC_CTRL__UDEC_CM_MODE__SHIFT 0xc +#define UVD_CGC_CTRL__UDEC_IT_MODE__SHIFT 0xd +#define UVD_CGC_CTRL__UDEC_DB_MODE__SHIFT 0xe +#define UVD_CGC_CTRL__UDEC_MP_MODE__SHIFT 0xf +#define UVD_CGC_CTRL__SYS_MODE__SHIFT 0x10 +#define UVD_CGC_CTRL__UDEC_MODE__SHIFT 0x11 +#define UVD_CGC_CTRL__MPEG2_MODE__SHIFT 0x12 +#define UVD_CGC_CTRL__REGS_MODE__SHIFT 0x13 +#define UVD_CGC_CTRL__RBC_MODE__SHIFT 0x14 +#define UVD_CGC_CTRL__LMI_MC_MODE__SHIFT 0x15 +#define UVD_CGC_CTRL__LMI_UMC_MODE__SHIFT 0x16 +#define UVD_CGC_CTRL__IDCT_MODE__SHIFT 0x17 +#define UVD_CGC_CTRL__MPRD_MODE__SHIFT 0x18 +#define UVD_CGC_CTRL__MPC_MODE__SHIFT 0x19 +#define UVD_CGC_CTRL__LBSI_MODE__SHIFT 0x1a +#define UVD_CGC_CTRL__LRBBM_MODE__SHIFT 0x1b +#define UVD_CGC_CTRL__WCB_MODE__SHIFT 0x1c +#define UVD_CGC_CTRL__VCPU_MODE__SHIFT 0x1d +#define UVD_CGC_CTRL__MMSCH_MODE__SHIFT 0x1f +#define UVD_CGC_CTRL__DYN_CLOCK_MODE_MASK 0x00000001L +#define UVD_CGC_CTRL__CLK_GATE_DLY_TIMER_MASK 0x0000003CL +#define UVD_CGC_CTRL__CLK_OFF_DELAY_MASK 0x000007C0L +#define UVD_CGC_CTRL__UDEC_RE_MODE_MASK 0x00000800L +#define UVD_CGC_CTRL__UDEC_CM_MODE_MASK 0x00001000L +#define UVD_CGC_CTRL__UDEC_IT_MODE_MASK 0x00002000L +#define UVD_CGC_CTRL__UDEC_DB_MODE_MASK 0x00004000L +#define UVD_CGC_CTRL__UDEC_MP_MODE_MASK 0x00008000L +#define UVD_CGC_CTRL__SYS_MODE_MASK 0x00010000L +#define UVD_CGC_CTRL__UDEC_MODE_MASK 0x00020000L +#define UVD_CGC_CTRL__MPEG2_MODE_MASK 0x00040000L +#define UVD_CGC_CTRL__REGS_MODE_MASK 0x00080000L +#define UVD_CGC_CTRL__RBC_MODE_MASK 0x00100000L +#define UVD_CGC_CTRL__LMI_MC_MODE_MASK 0x00200000L +#define UVD_CGC_CTRL__LMI_UMC_MODE_MASK 0x00400000L +#define UVD_CGC_CTRL__IDCT_MODE_MASK 0x00800000L +#define UVD_CGC_CTRL__MPRD_MODE_MASK 0x01000000L +#define UVD_CGC_CTRL__MPC_MODE_MASK 0x02000000L +#define UVD_CGC_CTRL__LBSI_MODE_MASK 0x04000000L +#define UVD_CGC_CTRL__LRBBM_MODE_MASK 0x08000000L +#define UVD_CGC_CTRL__WCB_MODE_MASK 0x10000000L +#define UVD_CGC_CTRL__VCPU_MODE_MASK 0x20000000L +#define UVD_CGC_CTRL__MMSCH_MODE_MASK 0x80000000L +//AVM_SUVD_CGC_GATE +#define AVM_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define AVM_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define AVM_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define AVM_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define AVM_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define AVM_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define AVM_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define AVM_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define AVM_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define AVM_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define AVM_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define AVM_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define AVM_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define AVM_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define AVM_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define AVM_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define AVM_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define AVM_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define AVM_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define AVM_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define AVM_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define AVM_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define AVM_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define AVM_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define AVM_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define AVM_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define AVM_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define AVM_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define AVM_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define AVM_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define AVM_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define AVM_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define AVM_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define AVM_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define AVM_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define AVM_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define AVM_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define AVM_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define AVM_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define AVM_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define AVM_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define AVM_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define AVM_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define AVM_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define AVM_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define AVM_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define AVM_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define AVM_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define AVM_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define AVM_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define AVM_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define AVM_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define AVM_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define AVM_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define AVM_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define AVM_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define AVM_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define AVM_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define AVM_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define AVM_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define AVM_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define AVM_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define AVM_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define AVM_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//CDEFE_SUVD_CGC_GATE +#define CDEFE_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define CDEFE_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define CDEFE_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define CDEFE_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define CDEFE_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define CDEFE_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define CDEFE_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define CDEFE_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define CDEFE_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define CDEFE_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define CDEFE_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define CDEFE_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define CDEFE_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define CDEFE_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define CDEFE_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define CDEFE_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define CDEFE_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define CDEFE_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define CDEFE_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define CDEFE_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define CDEFE_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define CDEFE_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define CDEFE_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define CDEFE_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define CDEFE_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define CDEFE_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define CDEFE_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define CDEFE_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define CDEFE_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define CDEFE_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define CDEFE_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define CDEFE_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define CDEFE_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define CDEFE_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define CDEFE_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define CDEFE_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define CDEFE_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define CDEFE_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define CDEFE_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define CDEFE_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define CDEFE_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define CDEFE_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define CDEFE_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define CDEFE_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define CDEFE_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define CDEFE_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define CDEFE_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define CDEFE_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define CDEFE_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define CDEFE_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define CDEFE_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define CDEFE_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define CDEFE_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define CDEFE_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define CDEFE_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define CDEFE_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define CDEFE_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define CDEFE_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define CDEFE_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define CDEFE_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define CDEFE_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define CDEFE_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define CDEFE_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define CDEFE_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//EFC_SUVD_CGC_GATE +#define EFC_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define EFC_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define EFC_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define EFC_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define EFC_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define EFC_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define EFC_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define EFC_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define EFC_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define EFC_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define EFC_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define EFC_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define EFC_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define EFC_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define EFC_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define EFC_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define EFC_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define EFC_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define EFC_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define EFC_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define EFC_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define EFC_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define EFC_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define EFC_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define EFC_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define EFC_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define EFC_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define EFC_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define EFC_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define EFC_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define EFC_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define EFC_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define EFC_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define EFC_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define EFC_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define EFC_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define EFC_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define EFC_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define EFC_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define EFC_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define EFC_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define EFC_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define EFC_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define EFC_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define EFC_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define EFC_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define EFC_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define EFC_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define EFC_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define EFC_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define EFC_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define EFC_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define EFC_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define EFC_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define EFC_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define EFC_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define EFC_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define EFC_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define EFC_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define EFC_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define EFC_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define EFC_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define EFC_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define EFC_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//ENT_SUVD_CGC_GATE +#define ENT_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define ENT_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define ENT_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define ENT_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define ENT_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define ENT_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define ENT_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define ENT_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define ENT_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define ENT_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define ENT_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define ENT_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define ENT_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define ENT_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define ENT_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define ENT_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define ENT_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define ENT_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define ENT_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define ENT_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define ENT_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define ENT_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define ENT_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define ENT_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define ENT_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define ENT_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define ENT_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define ENT_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define ENT_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define ENT_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define ENT_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define ENT_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define ENT_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define ENT_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define ENT_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define ENT_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define ENT_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define ENT_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define ENT_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define ENT_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define ENT_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define ENT_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define ENT_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define ENT_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define ENT_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define ENT_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define ENT_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define ENT_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define ENT_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define ENT_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define ENT_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define ENT_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define ENT_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define ENT_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define ENT_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define ENT_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define ENT_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define ENT_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define ENT_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define ENT_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define ENT_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define ENT_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define ENT_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define ENT_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//IME_SUVD_CGC_GATE +#define IME_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define IME_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define IME_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define IME_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define IME_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define IME_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define IME_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define IME_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define IME_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define IME_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define IME_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define IME_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define IME_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define IME_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define IME_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define IME_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define IME_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define IME_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define IME_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define IME_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define IME_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define IME_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define IME_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define IME_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define IME_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define IME_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define IME_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define IME_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define IME_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define IME_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define IME_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define IME_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define IME_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define IME_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define IME_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define IME_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define IME_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define IME_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define IME_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define IME_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define IME_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define IME_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define IME_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define IME_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define IME_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define IME_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define IME_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define IME_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define IME_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define IME_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define IME_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define IME_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define IME_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define IME_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define IME_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define IME_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define IME_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define IME_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define IME_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define IME_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define IME_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define IME_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define IME_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define IME_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//PPU_SUVD_CGC_GATE +#define PPU_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define PPU_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define PPU_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define PPU_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define PPU_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define PPU_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define PPU_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define PPU_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define PPU_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define PPU_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define PPU_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define PPU_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define PPU_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define PPU_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define PPU_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define PPU_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define PPU_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define PPU_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define PPU_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define PPU_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define PPU_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define PPU_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define PPU_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define PPU_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define PPU_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define PPU_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define PPU_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define PPU_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define PPU_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define PPU_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define PPU_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define PPU_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define PPU_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define PPU_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define PPU_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define PPU_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define PPU_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define PPU_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define PPU_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define PPU_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define PPU_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define PPU_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define PPU_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define PPU_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define PPU_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define PPU_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define PPU_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define PPU_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define PPU_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define PPU_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define PPU_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define PPU_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define PPU_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define PPU_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define PPU_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define PPU_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define PPU_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define PPU_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define PPU_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define PPU_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define PPU_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define PPU_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define PPU_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define PPU_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//SAOE_SUVD_CGC_GATE +#define SAOE_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define SAOE_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define SAOE_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define SAOE_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define SAOE_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define SAOE_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define SAOE_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define SAOE_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define SAOE_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define SAOE_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define SAOE_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define SAOE_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define SAOE_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define SAOE_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define SAOE_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define SAOE_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define SAOE_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define SAOE_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define SAOE_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define SAOE_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define SAOE_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define SAOE_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define SAOE_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define SAOE_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define SAOE_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define SAOE_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define SAOE_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define SAOE_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define SAOE_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define SAOE_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define SAOE_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define SAOE_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define SAOE_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define SAOE_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define SAOE_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define SAOE_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define SAOE_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define SAOE_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define SAOE_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define SAOE_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define SAOE_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define SAOE_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define SAOE_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define SAOE_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define SAOE_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define SAOE_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define SAOE_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define SAOE_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define SAOE_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define SAOE_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define SAOE_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define SAOE_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define SAOE_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define SAOE_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define SAOE_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define SAOE_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define SAOE_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define SAOE_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define SAOE_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define SAOE_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define SAOE_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define SAOE_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define SAOE_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define SAOE_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//SCM_SUVD_CGC_GATE +#define SCM_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define SCM_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define SCM_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define SCM_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define SCM_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define SCM_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define SCM_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define SCM_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define SCM_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define SCM_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define SCM_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define SCM_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define SCM_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define SCM_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define SCM_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define SCM_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define SCM_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define SCM_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define SCM_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define SCM_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define SCM_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define SCM_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define SCM_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define SCM_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define SCM_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define SCM_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define SCM_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define SCM_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define SCM_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define SCM_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define SCM_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define SCM_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define SCM_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define SCM_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define SCM_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define SCM_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define SCM_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define SCM_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define SCM_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define SCM_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define SCM_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define SCM_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define SCM_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define SCM_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define SCM_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define SCM_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define SCM_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define SCM_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define SCM_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define SCM_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define SCM_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define SCM_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define SCM_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define SCM_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define SCM_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define SCM_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define SCM_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define SCM_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define SCM_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define SCM_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define SCM_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define SCM_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define SCM_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define SCM_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//SDB_SUVD_CGC_GATE +#define SDB_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define SDB_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define SDB_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define SDB_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define SDB_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define SDB_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define SDB_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define SDB_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define SDB_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define SDB_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define SDB_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define SDB_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define SDB_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define SDB_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define SDB_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define SDB_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define SDB_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define SDB_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define SDB_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define SDB_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define SDB_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define SDB_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define SDB_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define SDB_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define SDB_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define SDB_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define SDB_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define SDB_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define SDB_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define SDB_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define SDB_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define SDB_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define SDB_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define SDB_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define SDB_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define SDB_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define SDB_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define SDB_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define SDB_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define SDB_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define SDB_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define SDB_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define SDB_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define SDB_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define SDB_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define SDB_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define SDB_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define SDB_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define SDB_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define SDB_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define SDB_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define SDB_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define SDB_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define SDB_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define SDB_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define SDB_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define SDB_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define SDB_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define SDB_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define SDB_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define SDB_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define SDB_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define SDB_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define SDB_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//SIT0_NXT_SUVD_CGC_GATE +#define SIT0_NXT_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define SIT0_NXT_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define SIT0_NXT_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define SIT0_NXT_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define SIT0_NXT_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define SIT0_NXT_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define SIT0_NXT_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define SIT0_NXT_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define SIT0_NXT_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define SIT0_NXT_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define SIT0_NXT_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define SIT0_NXT_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define SIT0_NXT_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define SIT0_NXT_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define SIT0_NXT_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define SIT0_NXT_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define SIT0_NXT_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define SIT0_NXT_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define SIT0_NXT_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define SIT0_NXT_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define SIT0_NXT_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define SIT0_NXT_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define SIT0_NXT_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define SIT0_NXT_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define SIT0_NXT_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define SIT0_NXT_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define SIT0_NXT_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define SIT0_NXT_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define SIT0_NXT_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define SIT0_NXT_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define SIT0_NXT_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define SIT0_NXT_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define SIT0_NXT_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define SIT0_NXT_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define SIT0_NXT_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define SIT0_NXT_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define SIT0_NXT_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define SIT0_NXT_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define SIT0_NXT_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define SIT0_NXT_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define SIT0_NXT_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define SIT0_NXT_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define SIT0_NXT_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define SIT0_NXT_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define SIT0_NXT_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define SIT0_NXT_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define SIT0_NXT_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define SIT0_NXT_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define SIT0_NXT_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define SIT0_NXT_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define SIT0_NXT_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define SIT0_NXT_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define SIT0_NXT_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define SIT0_NXT_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define SIT0_NXT_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define SIT0_NXT_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define SIT0_NXT_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define SIT0_NXT_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define SIT0_NXT_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define SIT0_NXT_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define SIT0_NXT_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define SIT0_NXT_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define SIT0_NXT_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define SIT0_NXT_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//SIT1_NXT_SUVD_CGC_GATE +#define SIT1_NXT_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define SIT1_NXT_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define SIT1_NXT_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define SIT1_NXT_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define SIT1_NXT_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define SIT1_NXT_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define SIT1_NXT_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define SIT1_NXT_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define SIT1_NXT_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define SIT1_NXT_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define SIT1_NXT_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define SIT1_NXT_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define SIT1_NXT_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define SIT1_NXT_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define SIT1_NXT_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define SIT1_NXT_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define SIT1_NXT_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define SIT1_NXT_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define SIT1_NXT_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define SIT1_NXT_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define SIT1_NXT_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define SIT1_NXT_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define SIT1_NXT_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define SIT1_NXT_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define SIT1_NXT_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define SIT1_NXT_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define SIT1_NXT_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define SIT1_NXT_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define SIT1_NXT_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define SIT1_NXT_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define SIT1_NXT_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define SIT1_NXT_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define SIT1_NXT_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define SIT1_NXT_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define SIT1_NXT_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define SIT1_NXT_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define SIT1_NXT_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define SIT1_NXT_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define SIT1_NXT_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define SIT1_NXT_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define SIT1_NXT_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define SIT1_NXT_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define SIT1_NXT_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define SIT1_NXT_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define SIT1_NXT_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define SIT1_NXT_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define SIT1_NXT_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define SIT1_NXT_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define SIT1_NXT_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define SIT1_NXT_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define SIT1_NXT_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define SIT1_NXT_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define SIT1_NXT_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define SIT1_NXT_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define SIT1_NXT_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define SIT1_NXT_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define SIT1_NXT_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define SIT1_NXT_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define SIT1_NXT_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define SIT1_NXT_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define SIT1_NXT_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define SIT1_NXT_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define SIT1_NXT_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define SIT1_NXT_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//SIT2_NXT_SUVD_CGC_GATE +#define SIT2_NXT_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define SIT2_NXT_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define SIT2_NXT_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define SIT2_NXT_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define SIT2_NXT_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define SIT2_NXT_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define SIT2_NXT_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define SIT2_NXT_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define SIT2_NXT_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define SIT2_NXT_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define SIT2_NXT_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define SIT2_NXT_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define SIT2_NXT_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define SIT2_NXT_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define SIT2_NXT_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define SIT2_NXT_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define SIT2_NXT_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define SIT2_NXT_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define SIT2_NXT_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define SIT2_NXT_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define SIT2_NXT_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define SIT2_NXT_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define SIT2_NXT_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define SIT2_NXT_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define SIT2_NXT_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define SIT2_NXT_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define SIT2_NXT_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define SIT2_NXT_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define SIT2_NXT_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define SIT2_NXT_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define SIT2_NXT_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define SIT2_NXT_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define SIT2_NXT_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define SIT2_NXT_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define SIT2_NXT_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define SIT2_NXT_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define SIT2_NXT_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define SIT2_NXT_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define SIT2_NXT_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define SIT2_NXT_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define SIT2_NXT_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define SIT2_NXT_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define SIT2_NXT_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define SIT2_NXT_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define SIT2_NXT_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define SIT2_NXT_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define SIT2_NXT_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define SIT2_NXT_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define SIT2_NXT_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define SIT2_NXT_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define SIT2_NXT_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define SIT2_NXT_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define SIT2_NXT_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define SIT2_NXT_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define SIT2_NXT_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define SIT2_NXT_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define SIT2_NXT_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define SIT2_NXT_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define SIT2_NXT_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define SIT2_NXT_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define SIT2_NXT_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define SIT2_NXT_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define SIT2_NXT_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define SIT2_NXT_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//SIT_SUVD_CGC_GATE +#define SIT_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define SIT_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define SIT_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define SIT_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define SIT_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define SIT_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define SIT_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define SIT_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define SIT_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define SIT_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define SIT_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define SIT_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define SIT_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define SIT_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define SIT_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define SIT_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define SIT_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define SIT_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define SIT_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define SIT_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define SIT_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define SIT_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define SIT_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define SIT_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define SIT_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define SIT_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define SIT_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define SIT_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define SIT_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define SIT_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define SIT_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define SIT_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define SIT_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define SIT_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define SIT_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define SIT_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define SIT_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define SIT_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define SIT_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define SIT_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define SIT_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define SIT_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define SIT_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define SIT_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define SIT_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define SIT_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define SIT_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define SIT_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define SIT_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define SIT_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define SIT_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define SIT_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define SIT_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define SIT_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define SIT_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define SIT_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define SIT_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define SIT_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define SIT_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define SIT_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define SIT_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define SIT_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define SIT_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define SIT_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//SMPA_SUVD_CGC_GATE +#define SMPA_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define SMPA_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define SMPA_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define SMPA_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define SMPA_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define SMPA_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define SMPA_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define SMPA_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define SMPA_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define SMPA_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define SMPA_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define SMPA_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define SMPA_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define SMPA_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define SMPA_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define SMPA_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define SMPA_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define SMPA_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define SMPA_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define SMPA_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define SMPA_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define SMPA_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define SMPA_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define SMPA_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define SMPA_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define SMPA_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define SMPA_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define SMPA_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define SMPA_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define SMPA_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define SMPA_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define SMPA_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define SMPA_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define SMPA_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define SMPA_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define SMPA_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define SMPA_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define SMPA_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define SMPA_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define SMPA_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define SMPA_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define SMPA_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define SMPA_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define SMPA_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define SMPA_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define SMPA_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define SMPA_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define SMPA_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define SMPA_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define SMPA_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define SMPA_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define SMPA_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define SMPA_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define SMPA_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define SMPA_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define SMPA_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define SMPA_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define SMPA_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define SMPA_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define SMPA_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define SMPA_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define SMPA_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define SMPA_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define SMPA_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//SMP_SUVD_CGC_GATE +#define SMP_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define SMP_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define SMP_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define SMP_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define SMP_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define SMP_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define SMP_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define SMP_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define SMP_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define SMP_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define SMP_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define SMP_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define SMP_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define SMP_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define SMP_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define SMP_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define SMP_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define SMP_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define SMP_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define SMP_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define SMP_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define SMP_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define SMP_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define SMP_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define SMP_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define SMP_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define SMP_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define SMP_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define SMP_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define SMP_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define SMP_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define SMP_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define SMP_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define SMP_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define SMP_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define SMP_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define SMP_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define SMP_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define SMP_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define SMP_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define SMP_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define SMP_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define SMP_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define SMP_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define SMP_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define SMP_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define SMP_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define SMP_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define SMP_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define SMP_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define SMP_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define SMP_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define SMP_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define SMP_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define SMP_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define SMP_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define SMP_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define SMP_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define SMP_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define SMP_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define SMP_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define SMP_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define SMP_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define SMP_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//SRE_SUVD_CGC_GATE +#define SRE_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define SRE_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define SRE_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define SRE_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define SRE_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define SRE_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define SRE_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define SRE_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define SRE_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define SRE_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define SRE_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define SRE_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define SRE_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define SRE_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define SRE_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define SRE_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define SRE_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define SRE_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define SRE_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define SRE_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define SRE_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define SRE_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define SRE_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define SRE_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define SRE_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define SRE_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define SRE_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define SRE_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define SRE_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define SRE_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define SRE_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define SRE_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define SRE_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define SRE_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define SRE_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define SRE_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define SRE_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define SRE_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define SRE_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define SRE_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define SRE_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define SRE_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define SRE_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define SRE_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define SRE_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define SRE_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define SRE_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define SRE_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define SRE_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define SRE_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define SRE_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define SRE_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define SRE_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define SRE_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define SRE_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define SRE_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define SRE_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define SRE_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define SRE_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define SRE_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define SRE_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define SRE_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define SRE_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define SRE_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//UVD_MPBE0_SUVD_CGC_GATE +#define UVD_MPBE0_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define UVD_MPBE0_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define UVD_MPBE0_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define UVD_MPBE0_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define UVD_MPBE0_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define UVD_MPBE0_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define UVD_MPBE0_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define UVD_MPBE0_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define UVD_MPBE0_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define UVD_MPBE0_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define UVD_MPBE0_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define UVD_MPBE0_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define UVD_MPBE0_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define UVD_MPBE0_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define UVD_MPBE0_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define UVD_MPBE0_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define UVD_MPBE0_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define UVD_MPBE0_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define UVD_MPBE0_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define UVD_MPBE0_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define UVD_MPBE0_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define UVD_MPBE0_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define UVD_MPBE0_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define UVD_MPBE0_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define UVD_MPBE0_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define UVD_MPBE0_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define UVD_MPBE0_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define UVD_MPBE0_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define UVD_MPBE0_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define UVD_MPBE0_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define UVD_MPBE0_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define UVD_MPBE0_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define UVD_MPBE0_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define UVD_MPBE0_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define UVD_MPBE0_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define UVD_MPBE0_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define UVD_MPBE0_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define UVD_MPBE0_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define UVD_MPBE0_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define UVD_MPBE0_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define UVD_MPBE0_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define UVD_MPBE0_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define UVD_MPBE0_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define UVD_MPBE0_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define UVD_MPBE0_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define UVD_MPBE0_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define UVD_MPBE0_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define UVD_MPBE0_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define UVD_MPBE0_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define UVD_MPBE0_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define UVD_MPBE0_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define UVD_MPBE0_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define UVD_MPBE0_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define UVD_MPBE0_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define UVD_MPBE0_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define UVD_MPBE0_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define UVD_MPBE0_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define UVD_MPBE0_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define UVD_MPBE0_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define UVD_MPBE0_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define UVD_MPBE0_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define UVD_MPBE0_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define UVD_MPBE0_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define UVD_MPBE0_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//UVD_MPBE1_SUVD_CGC_GATE +#define UVD_MPBE1_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define UVD_MPBE1_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define UVD_MPBE1_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define UVD_MPBE1_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define UVD_MPBE1_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define UVD_MPBE1_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define UVD_MPBE1_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define UVD_MPBE1_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define UVD_MPBE1_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define UVD_MPBE1_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define UVD_MPBE1_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define UVD_MPBE1_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define UVD_MPBE1_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define UVD_MPBE1_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define UVD_MPBE1_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define UVD_MPBE1_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define UVD_MPBE1_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define UVD_MPBE1_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define UVD_MPBE1_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define UVD_MPBE1_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define UVD_MPBE1_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define UVD_MPBE1_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define UVD_MPBE1_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define UVD_MPBE1_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define UVD_MPBE1_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define UVD_MPBE1_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define UVD_MPBE1_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define UVD_MPBE1_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define UVD_MPBE1_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define UVD_MPBE1_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define UVD_MPBE1_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define UVD_MPBE1_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define UVD_MPBE1_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define UVD_MPBE1_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define UVD_MPBE1_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define UVD_MPBE1_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define UVD_MPBE1_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define UVD_MPBE1_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define UVD_MPBE1_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define UVD_MPBE1_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define UVD_MPBE1_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define UVD_MPBE1_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define UVD_MPBE1_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define UVD_MPBE1_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define UVD_MPBE1_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define UVD_MPBE1_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define UVD_MPBE1_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define UVD_MPBE1_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define UVD_MPBE1_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define UVD_MPBE1_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define UVD_MPBE1_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define UVD_MPBE1_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define UVD_MPBE1_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define UVD_MPBE1_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define UVD_MPBE1_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define UVD_MPBE1_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define UVD_MPBE1_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define UVD_MPBE1_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define UVD_MPBE1_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define UVD_MPBE1_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define UVD_MPBE1_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define UVD_MPBE1_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define UVD_MPBE1_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define UVD_MPBE1_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//UVD_SUVD_CGC_GATE +#define UVD_SUVD_CGC_GATE__SRE__SHIFT 0x0 +#define UVD_SUVD_CGC_GATE__SIT__SHIFT 0x1 +#define UVD_SUVD_CGC_GATE__SMP__SHIFT 0x2 +#define UVD_SUVD_CGC_GATE__SCM__SHIFT 0x3 +#define UVD_SUVD_CGC_GATE__SDB__SHIFT 0x4 +#define UVD_SUVD_CGC_GATE__SRE_H264__SHIFT 0x5 +#define UVD_SUVD_CGC_GATE__SRE_HEVC__SHIFT 0x6 +#define UVD_SUVD_CGC_GATE__SIT_H264__SHIFT 0x7 +#define UVD_SUVD_CGC_GATE__SIT_HEVC__SHIFT 0x8 +#define UVD_SUVD_CGC_GATE__SCM_H264__SHIFT 0x9 +#define UVD_SUVD_CGC_GATE__SCM_HEVC__SHIFT 0xa +#define UVD_SUVD_CGC_GATE__SDB_H264__SHIFT 0xb +#define UVD_SUVD_CGC_GATE__SDB_HEVC__SHIFT 0xc +#define UVD_SUVD_CGC_GATE__SCLR__SHIFT 0xd +#define UVD_SUVD_CGC_GATE__UVD_SC__SHIFT 0xe +#define UVD_SUVD_CGC_GATE__ENT__SHIFT 0xf +#define UVD_SUVD_CGC_GATE__IME__SHIFT 0x10 +#define UVD_SUVD_CGC_GATE__SIT_HEVC_DEC__SHIFT 0x11 +#define UVD_SUVD_CGC_GATE__SIT_HEVC_ENC__SHIFT 0x12 +#define UVD_SUVD_CGC_GATE__SITE__SHIFT 0x13 +#define UVD_SUVD_CGC_GATE__SRE_VP9__SHIFT 0x14 +#define UVD_SUVD_CGC_GATE__SCM_VP9__SHIFT 0x15 +#define UVD_SUVD_CGC_GATE__SIT_VP9_DEC__SHIFT 0x16 +#define UVD_SUVD_CGC_GATE__SDB_VP9__SHIFT 0x17 +#define UVD_SUVD_CGC_GATE__IME_HEVC__SHIFT 0x18 +#define UVD_SUVD_CGC_GATE__EFC__SHIFT 0x19 +#define UVD_SUVD_CGC_GATE__SAOE__SHIFT 0x1a +#define UVD_SUVD_CGC_GATE__SRE_AV1__SHIFT 0x1b +#define UVD_SUVD_CGC_GATE__FBC_PCLK__SHIFT 0x1c +#define UVD_SUVD_CGC_GATE__FBC_CCLK__SHIFT 0x1d +#define UVD_SUVD_CGC_GATE__SCM_AV1__SHIFT 0x1e +#define UVD_SUVD_CGC_GATE__SMPA__SHIFT 0x1f +#define UVD_SUVD_CGC_GATE__SRE_MASK 0x00000001L +#define UVD_SUVD_CGC_GATE__SIT_MASK 0x00000002L +#define UVD_SUVD_CGC_GATE__SMP_MASK 0x00000004L +#define UVD_SUVD_CGC_GATE__SCM_MASK 0x00000008L +#define UVD_SUVD_CGC_GATE__SDB_MASK 0x00000010L +#define UVD_SUVD_CGC_GATE__SRE_H264_MASK 0x00000020L +#define UVD_SUVD_CGC_GATE__SRE_HEVC_MASK 0x00000040L +#define UVD_SUVD_CGC_GATE__SIT_H264_MASK 0x00000080L +#define UVD_SUVD_CGC_GATE__SIT_HEVC_MASK 0x00000100L +#define UVD_SUVD_CGC_GATE__SCM_H264_MASK 0x00000200L +#define UVD_SUVD_CGC_GATE__SCM_HEVC_MASK 0x00000400L +#define UVD_SUVD_CGC_GATE__SDB_H264_MASK 0x00000800L +#define UVD_SUVD_CGC_GATE__SDB_HEVC_MASK 0x00001000L +#define UVD_SUVD_CGC_GATE__SCLR_MASK 0x00002000L +#define UVD_SUVD_CGC_GATE__UVD_SC_MASK 0x00004000L +#define UVD_SUVD_CGC_GATE__ENT_MASK 0x00008000L +#define UVD_SUVD_CGC_GATE__IME_MASK 0x00010000L +#define UVD_SUVD_CGC_GATE__SIT_HEVC_DEC_MASK 0x00020000L +#define UVD_SUVD_CGC_GATE__SIT_HEVC_ENC_MASK 0x00040000L +#define UVD_SUVD_CGC_GATE__SITE_MASK 0x00080000L +#define UVD_SUVD_CGC_GATE__SRE_VP9_MASK 0x00100000L +#define UVD_SUVD_CGC_GATE__SCM_VP9_MASK 0x00200000L +#define UVD_SUVD_CGC_GATE__SIT_VP9_DEC_MASK 0x00400000L +#define UVD_SUVD_CGC_GATE__SDB_VP9_MASK 0x00800000L +#define UVD_SUVD_CGC_GATE__IME_HEVC_MASK 0x01000000L +#define UVD_SUVD_CGC_GATE__EFC_MASK 0x02000000L +#define UVD_SUVD_CGC_GATE__SAOE_MASK 0x04000000L +#define UVD_SUVD_CGC_GATE__SRE_AV1_MASK 0x08000000L +#define UVD_SUVD_CGC_GATE__FBC_PCLK_MASK 0x10000000L +#define UVD_SUVD_CGC_GATE__FBC_CCLK_MASK 0x20000000L +#define UVD_SUVD_CGC_GATE__SCM_AV1_MASK 0x40000000L +#define UVD_SUVD_CGC_GATE__SMPA_MASK 0x80000000L +//AVM_SUVD_CGC_GATE2 +#define AVM_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define AVM_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define AVM_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define AVM_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define AVM_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define AVM_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define AVM_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define AVM_SUVD_CGC_GATE2__AVM_0__SHIFT 0x7 +#define AVM_SUVD_CGC_GATE2__AVM_1__SHIFT 0x8 +#define AVM_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT 0x9 +#define AVM_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT 0xa +#define AVM_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT 0xb +#define AVM_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define AVM_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define AVM_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define AVM_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define AVM_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define AVM_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define AVM_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +#define AVM_SUVD_CGC_GATE2__AVM_0_MASK 0x00000080L +#define AVM_SUVD_CGC_GATE2__AVM_1_MASK 0x00000100L +#define AVM_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK 0x00000200L +#define AVM_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK 0x00000400L +#define AVM_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK 0x00000800L +//CDEFE_SUVD_CGC_GATE2 +#define CDEFE_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define CDEFE_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define CDEFE_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define CDEFE_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define CDEFE_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define CDEFE_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define CDEFE_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define CDEFE_SUVD_CGC_GATE2__AVM_0__SHIFT 0x7 +#define CDEFE_SUVD_CGC_GATE2__AVM_1__SHIFT 0x8 +#define CDEFE_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT 0x9 +#define CDEFE_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT 0xa +#define CDEFE_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT 0xb +#define CDEFE_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define CDEFE_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define CDEFE_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define CDEFE_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define CDEFE_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define CDEFE_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define CDEFE_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +#define CDEFE_SUVD_CGC_GATE2__AVM_0_MASK 0x00000080L +#define CDEFE_SUVD_CGC_GATE2__AVM_1_MASK 0x00000100L +#define CDEFE_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK 0x00000200L +#define CDEFE_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK 0x00000400L +#define CDEFE_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK 0x00000800L +//DBR_SUVD_CGC_GATE2 +#define DBR_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define DBR_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define DBR_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define DBR_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define DBR_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define DBR_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define DBR_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define DBR_SUVD_CGC_GATE2__AVM_0__SHIFT 0x7 +#define DBR_SUVD_CGC_GATE2__AVM_1__SHIFT 0x8 +#define DBR_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT 0x9 +#define DBR_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT 0xa +#define DBR_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT 0xb +#define DBR_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define DBR_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define DBR_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define DBR_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define DBR_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define DBR_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define DBR_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +#define DBR_SUVD_CGC_GATE2__AVM_0_MASK 0x00000080L +#define DBR_SUVD_CGC_GATE2__AVM_1_MASK 0x00000100L +#define DBR_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK 0x00000200L +#define DBR_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK 0x00000400L +#define DBR_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK 0x00000800L +//ENT_SUVD_CGC_GATE2 +#define ENT_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define ENT_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define ENT_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define ENT_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define ENT_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define ENT_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define ENT_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define ENT_SUVD_CGC_GATE2__AVM_0__SHIFT 0x7 +#define ENT_SUVD_CGC_GATE2__AVM_1__SHIFT 0x8 +#define ENT_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT 0x9 +#define ENT_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT 0xa +#define ENT_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT 0xb +#define ENT_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define ENT_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define ENT_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define ENT_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define ENT_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define ENT_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define ENT_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +#define ENT_SUVD_CGC_GATE2__AVM_0_MASK 0x00000080L +#define ENT_SUVD_CGC_GATE2__AVM_1_MASK 0x00000100L +#define ENT_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK 0x00000200L +#define ENT_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK 0x00000400L +#define ENT_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK 0x00000800L +//IME_SUVD_CGC_GATE2 +#define IME_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define IME_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define IME_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define IME_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define IME_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define IME_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define IME_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define IME_SUVD_CGC_GATE2__AVM_0__SHIFT 0x7 +#define IME_SUVD_CGC_GATE2__AVM_1__SHIFT 0x8 +#define IME_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT 0x9 +#define IME_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT 0xa +#define IME_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT 0xb +#define IME_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define IME_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define IME_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define IME_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define IME_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define IME_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define IME_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +#define IME_SUVD_CGC_GATE2__AVM_0_MASK 0x00000080L +#define IME_SUVD_CGC_GATE2__AVM_1_MASK 0x00000100L +#define IME_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK 0x00000200L +#define IME_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK 0x00000400L +#define IME_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK 0x00000800L +//MPC1_SUVD_CGC_GATE2 +#define MPC1_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define MPC1_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define MPC1_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define MPC1_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define MPC1_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define MPC1_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define MPC1_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define MPC1_SUVD_CGC_GATE2__AVM_0__SHIFT 0x7 +#define MPC1_SUVD_CGC_GATE2__AVM_1__SHIFT 0x8 +#define MPC1_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT 0x9 +#define MPC1_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT 0xa +#define MPC1_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT 0xb +#define MPC1_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define MPC1_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define MPC1_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define MPC1_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define MPC1_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define MPC1_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define MPC1_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +#define MPC1_SUVD_CGC_GATE2__AVM_0_MASK 0x00000080L +#define MPC1_SUVD_CGC_GATE2__AVM_1_MASK 0x00000100L +#define MPC1_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK 0x00000200L +#define MPC1_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK 0x00000400L +#define MPC1_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK 0x00000800L +//SAOE_SUVD_CGC_GATE2 +#define SAOE_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define SAOE_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define SAOE_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define SAOE_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define SAOE_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define SAOE_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define SAOE_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define SAOE_SUVD_CGC_GATE2__AVM_0__SHIFT 0x7 +#define SAOE_SUVD_CGC_GATE2__AVM_1__SHIFT 0x8 +#define SAOE_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT 0x9 +#define SAOE_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT 0xa +#define SAOE_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT 0xb +#define SAOE_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define SAOE_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define SAOE_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define SAOE_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define SAOE_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define SAOE_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define SAOE_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +#define SAOE_SUVD_CGC_GATE2__AVM_0_MASK 0x00000080L +#define SAOE_SUVD_CGC_GATE2__AVM_1_MASK 0x00000100L +#define SAOE_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK 0x00000200L +#define SAOE_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK 0x00000400L +#define SAOE_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK 0x00000800L +//SDB_SUVD_CGC_GATE2 +#define SDB_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define SDB_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define SDB_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define SDB_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define SDB_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define SDB_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define SDB_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define SDB_SUVD_CGC_GATE2__AVM_0__SHIFT 0x7 +#define SDB_SUVD_CGC_GATE2__AVM_1__SHIFT 0x8 +#define SDB_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT 0x9 +#define SDB_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT 0xa +#define SDB_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT 0xb +#define SDB_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define SDB_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define SDB_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define SDB_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define SDB_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define SDB_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define SDB_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +#define SDB_SUVD_CGC_GATE2__AVM_0_MASK 0x00000080L +#define SDB_SUVD_CGC_GATE2__AVM_1_MASK 0x00000100L +#define SDB_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK 0x00000200L +#define SDB_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK 0x00000400L +#define SDB_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK 0x00000800L +//SIT0_NXT_SUVD_CGC_GATE2 +#define SIT0_NXT_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define SIT0_NXT_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define SIT0_NXT_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define SIT0_NXT_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define SIT0_NXT_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define SIT0_NXT_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define SIT0_NXT_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define SIT0_NXT_SUVD_CGC_GATE2__AVM_0__SHIFT 0x7 +#define SIT0_NXT_SUVD_CGC_GATE2__AVM_1__SHIFT 0x8 +#define SIT0_NXT_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT 0x9 +#define SIT0_NXT_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT 0xa +#define SIT0_NXT_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT 0xb +#define SIT0_NXT_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define SIT0_NXT_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define SIT0_NXT_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define SIT0_NXT_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define SIT0_NXT_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define SIT0_NXT_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define SIT0_NXT_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +#define SIT0_NXT_SUVD_CGC_GATE2__AVM_0_MASK 0x00000080L +#define SIT0_NXT_SUVD_CGC_GATE2__AVM_1_MASK 0x00000100L +#define SIT0_NXT_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK 0x00000200L +#define SIT0_NXT_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK 0x00000400L +#define SIT0_NXT_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK 0x00000800L +//SIT1_NXT_SUVD_CGC_GATE2 +#define SIT1_NXT_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define SIT1_NXT_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define SIT1_NXT_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define SIT1_NXT_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define SIT1_NXT_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define SIT1_NXT_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define SIT1_NXT_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define SIT1_NXT_SUVD_CGC_GATE2__AVM_0__SHIFT 0x7 +#define SIT1_NXT_SUVD_CGC_GATE2__AVM_1__SHIFT 0x8 +#define SIT1_NXT_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT 0x9 +#define SIT1_NXT_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT 0xa +#define SIT1_NXT_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT 0xb +#define SIT1_NXT_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define SIT1_NXT_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define SIT1_NXT_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define SIT1_NXT_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define SIT1_NXT_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define SIT1_NXT_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define SIT1_NXT_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +#define SIT1_NXT_SUVD_CGC_GATE2__AVM_0_MASK 0x00000080L +#define SIT1_NXT_SUVD_CGC_GATE2__AVM_1_MASK 0x00000100L +#define SIT1_NXT_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK 0x00000200L +#define SIT1_NXT_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK 0x00000400L +#define SIT1_NXT_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK 0x00000800L +//SIT2_NXT_SUVD_CGC_GATE2 +#define SIT2_NXT_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define SIT2_NXT_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define SIT2_NXT_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define SIT2_NXT_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define SIT2_NXT_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define SIT2_NXT_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define SIT2_NXT_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define SIT2_NXT_SUVD_CGC_GATE2__AVM_0__SHIFT 0x7 +#define SIT2_NXT_SUVD_CGC_GATE2__AVM_1__SHIFT 0x8 +#define SIT2_NXT_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT 0x9 +#define SIT2_NXT_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT 0xa +#define SIT2_NXT_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT 0xb +#define SIT2_NXT_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define SIT2_NXT_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define SIT2_NXT_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define SIT2_NXT_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define SIT2_NXT_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define SIT2_NXT_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define SIT2_NXT_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +#define SIT2_NXT_SUVD_CGC_GATE2__AVM_0_MASK 0x00000080L +#define SIT2_NXT_SUVD_CGC_GATE2__AVM_1_MASK 0x00000100L +#define SIT2_NXT_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK 0x00000200L +#define SIT2_NXT_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK 0x00000400L +#define SIT2_NXT_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK 0x00000800L +//SIT_SUVD_CGC_GATE2 +#define SIT_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define SIT_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define SIT_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define SIT_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define SIT_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define SIT_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define SIT_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define SIT_SUVD_CGC_GATE2__AVM_0__SHIFT 0x7 +#define SIT_SUVD_CGC_GATE2__AVM_1__SHIFT 0x8 +#define SIT_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT 0x9 +#define SIT_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT 0xa +#define SIT_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT 0xb +#define SIT_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define SIT_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define SIT_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define SIT_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define SIT_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define SIT_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define SIT_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +#define SIT_SUVD_CGC_GATE2__AVM_0_MASK 0x00000080L +#define SIT_SUVD_CGC_GATE2__AVM_1_MASK 0x00000100L +#define SIT_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK 0x00000200L +#define SIT_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK 0x00000400L +#define SIT_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK 0x00000800L +//SMPA_SUVD_CGC_GATE2 +#define SMPA_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define SMPA_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define SMPA_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define SMPA_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define SMPA_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define SMPA_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define SMPA_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define SMPA_SUVD_CGC_GATE2__AVM_0__SHIFT 0x7 +#define SMPA_SUVD_CGC_GATE2__AVM_1__SHIFT 0x8 +#define SMPA_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT 0x9 +#define SMPA_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT 0xa +#define SMPA_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT 0xb +#define SMPA_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define SMPA_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define SMPA_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define SMPA_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define SMPA_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define SMPA_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define SMPA_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +#define SMPA_SUVD_CGC_GATE2__AVM_0_MASK 0x00000080L +#define SMPA_SUVD_CGC_GATE2__AVM_1_MASK 0x00000100L +#define SMPA_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK 0x00000200L +#define SMPA_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK 0x00000400L +#define SMPA_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK 0x00000800L +//SMP_SUVD_CGC_GATE2 +#define SMP_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define SMP_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define SMP_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define SMP_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define SMP_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define SMP_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define SMP_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define SMP_SUVD_CGC_GATE2__AVM_0__SHIFT 0x7 +#define SMP_SUVD_CGC_GATE2__AVM_1__SHIFT 0x8 +#define SMP_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT 0x9 +#define SMP_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT 0xa +#define SMP_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT 0xb +#define SMP_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define SMP_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define SMP_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define SMP_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define SMP_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define SMP_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define SMP_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +#define SMP_SUVD_CGC_GATE2__AVM_0_MASK 0x00000080L +#define SMP_SUVD_CGC_GATE2__AVM_1_MASK 0x00000100L +#define SMP_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK 0x00000200L +#define SMP_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK 0x00000400L +#define SMP_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK 0x00000800L +//SRE_SUVD_CGC_GATE2 +#define SRE_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define SRE_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define SRE_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define SRE_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define SRE_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define SRE_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define SRE_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define SRE_SUVD_CGC_GATE2__AVM_0__SHIFT 0x7 +#define SRE_SUVD_CGC_GATE2__AVM_1__SHIFT 0x8 +#define SRE_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT 0x9 +#define SRE_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT 0xa +#define SRE_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT 0xb +#define SRE_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define SRE_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define SRE_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define SRE_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define SRE_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define SRE_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define SRE_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +#define SRE_SUVD_CGC_GATE2__AVM_0_MASK 0x00000080L +#define SRE_SUVD_CGC_GATE2__AVM_1_MASK 0x00000100L +#define SRE_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK 0x00000200L +#define SRE_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK 0x00000400L +#define SRE_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK 0x00000800L +//UVD_MPBE0_SUVD_CGC_GATE2 +#define UVD_MPBE0_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define UVD_MPBE0_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define UVD_MPBE0_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define UVD_MPBE0_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define UVD_MPBE0_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define UVD_MPBE0_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define UVD_MPBE0_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define UVD_MPBE0_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define UVD_MPBE0_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define UVD_MPBE0_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define UVD_MPBE0_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define UVD_MPBE0_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define UVD_MPBE0_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define UVD_MPBE0_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +//UVD_MPBE1_SUVD_CGC_GATE2 +#define UVD_MPBE1_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define UVD_MPBE1_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define UVD_MPBE1_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define UVD_MPBE1_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define UVD_MPBE1_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define UVD_MPBE1_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define UVD_MPBE1_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define UVD_MPBE1_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define UVD_MPBE1_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define UVD_MPBE1_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define UVD_MPBE1_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define UVD_MPBE1_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define UVD_MPBE1_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define UVD_MPBE1_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +//UVD_SUVD_CGC_GATE2 +#define UVD_SUVD_CGC_GATE2__MPBE0__SHIFT 0x0 +#define UVD_SUVD_CGC_GATE2__MPBE1__SHIFT 0x1 +#define UVD_SUVD_CGC_GATE2__SIT_AV1__SHIFT 0x2 +#define UVD_SUVD_CGC_GATE2__SDB_AV1__SHIFT 0x3 +#define UVD_SUVD_CGC_GATE2__MPC1__SHIFT 0x4 +#define UVD_SUVD_CGC_GATE2__SRE_AV1_ENC__SHIFT 0x5 +#define UVD_SUVD_CGC_GATE2__CDEFE__SHIFT 0x6 +#define UVD_SUVD_CGC_GATE2__AVM_0__SHIFT 0x7 +#define UVD_SUVD_CGC_GATE2__AVM_1__SHIFT 0x8 +#define UVD_SUVD_CGC_GATE2__SIT_NXT_CMN__SHIFT 0x9 +#define UVD_SUVD_CGC_GATE2__SIT_NXT_DEC__SHIFT 0xa +#define UVD_SUVD_CGC_GATE2__SIT_NXT_ENC__SHIFT 0xb +#define UVD_SUVD_CGC_GATE2__MPBE0_MASK 0x00000001L +#define UVD_SUVD_CGC_GATE2__MPBE1_MASK 0x00000002L +#define UVD_SUVD_CGC_GATE2__SIT_AV1_MASK 0x00000004L +#define UVD_SUVD_CGC_GATE2__SDB_AV1_MASK 0x00000008L +#define UVD_SUVD_CGC_GATE2__MPC1_MASK 0x00000010L +#define UVD_SUVD_CGC_GATE2__SRE_AV1_ENC_MASK 0x00000020L +#define UVD_SUVD_CGC_GATE2__CDEFE_MASK 0x00000040L +#define UVD_SUVD_CGC_GATE2__AVM_0_MASK 0x00000080L +#define UVD_SUVD_CGC_GATE2__AVM_1_MASK 0x00000100L +#define UVD_SUVD_CGC_GATE2__SIT_NXT_CMN_MASK 0x00000200L +#define UVD_SUVD_CGC_GATE2__SIT_NXT_DEC_MASK 0x00000400L +#define UVD_SUVD_CGC_GATE2__SIT_NXT_ENC_MASK 0x00000800L +//AVM_SUVD_CGC_CTRL +#define AVM_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define AVM_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define AVM_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define AVM_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define AVM_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define AVM_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define AVM_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define AVM_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define AVM_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define AVM_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define AVM_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define AVM_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define AVM_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define AVM_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define AVM_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define AVM_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define AVM_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define AVM_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define AVM_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define AVM_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define AVM_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define AVM_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define AVM_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define AVM_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define AVM_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define AVM_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define AVM_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define AVM_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define AVM_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define AVM_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define AVM_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define AVM_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define AVM_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define AVM_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define AVM_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define AVM_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define AVM_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define AVM_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define AVM_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define AVM_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define AVM_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define AVM_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define AVM_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define AVM_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define AVM_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define AVM_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define AVM_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define AVM_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define AVM_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define AVM_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define AVM_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define AVM_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//CDEFE_SUVD_CGC_CTRL +#define CDEFE_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define CDEFE_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define CDEFE_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define CDEFE_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define CDEFE_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define CDEFE_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define CDEFE_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define CDEFE_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define CDEFE_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define CDEFE_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define CDEFE_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define CDEFE_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define CDEFE_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define CDEFE_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define CDEFE_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define CDEFE_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define CDEFE_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define CDEFE_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define CDEFE_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define CDEFE_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define CDEFE_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define CDEFE_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define CDEFE_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define CDEFE_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define CDEFE_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define CDEFE_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define CDEFE_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define CDEFE_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define CDEFE_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define CDEFE_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define CDEFE_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define CDEFE_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define CDEFE_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define CDEFE_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define CDEFE_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define CDEFE_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define CDEFE_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define CDEFE_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define CDEFE_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define CDEFE_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define CDEFE_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define CDEFE_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define CDEFE_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define CDEFE_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define CDEFE_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define CDEFE_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define CDEFE_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define CDEFE_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define CDEFE_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define CDEFE_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define CDEFE_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define CDEFE_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//DBR_SUVD_CGC_CTRL +#define DBR_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define DBR_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define DBR_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define DBR_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define DBR_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define DBR_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define DBR_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define DBR_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define DBR_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define DBR_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define DBR_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define DBR_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define DBR_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define DBR_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define DBR_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define DBR_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define DBR_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define DBR_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define DBR_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define DBR_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define DBR_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define DBR_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define DBR_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define DBR_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define DBR_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define DBR_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define DBR_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define DBR_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define DBR_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define DBR_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define DBR_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define DBR_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define DBR_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define DBR_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define DBR_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define DBR_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define DBR_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define DBR_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define DBR_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define DBR_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define DBR_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define DBR_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define DBR_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define DBR_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define DBR_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define DBR_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define DBR_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define DBR_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define DBR_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define DBR_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define DBR_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define DBR_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//EFC_SUVD_CGC_CTRL +#define EFC_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define EFC_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define EFC_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define EFC_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define EFC_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define EFC_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define EFC_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define EFC_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define EFC_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define EFC_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define EFC_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define EFC_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define EFC_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define EFC_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define EFC_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define EFC_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define EFC_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define EFC_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define EFC_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define EFC_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define EFC_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define EFC_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define EFC_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define EFC_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define EFC_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define EFC_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define EFC_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define EFC_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define EFC_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define EFC_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define EFC_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define EFC_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define EFC_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define EFC_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define EFC_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define EFC_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define EFC_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define EFC_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define EFC_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define EFC_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define EFC_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define EFC_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define EFC_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define EFC_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define EFC_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define EFC_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define EFC_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define EFC_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define EFC_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define EFC_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define EFC_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define EFC_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//ENT_SUVD_CGC_CTRL +#define ENT_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define ENT_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define ENT_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define ENT_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define ENT_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define ENT_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define ENT_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define ENT_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define ENT_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define ENT_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define ENT_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define ENT_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define ENT_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define ENT_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define ENT_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define ENT_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define ENT_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define ENT_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define ENT_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define ENT_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define ENT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define ENT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define ENT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define ENT_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define ENT_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define ENT_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define ENT_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define ENT_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define ENT_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define ENT_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define ENT_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define ENT_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define ENT_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define ENT_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define ENT_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define ENT_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define ENT_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define ENT_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define ENT_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define ENT_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define ENT_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define ENT_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define ENT_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define ENT_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define ENT_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define ENT_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define ENT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define ENT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define ENT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define ENT_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define ENT_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define ENT_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//IME_SUVD_CGC_CTRL +#define IME_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define IME_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define IME_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define IME_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define IME_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define IME_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define IME_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define IME_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define IME_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define IME_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define IME_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define IME_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define IME_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define IME_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define IME_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define IME_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define IME_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define IME_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define IME_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define IME_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define IME_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define IME_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define IME_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define IME_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define IME_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define IME_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define IME_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define IME_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define IME_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define IME_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define IME_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define IME_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define IME_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define IME_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define IME_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define IME_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define IME_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define IME_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define IME_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define IME_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define IME_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define IME_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define IME_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define IME_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define IME_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define IME_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define IME_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define IME_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define IME_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define IME_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define IME_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define IME_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//MPC1_SUVD_CGC_CTRL +#define MPC1_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define MPC1_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define MPC1_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define MPC1_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define MPC1_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define MPC1_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define MPC1_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define MPC1_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define MPC1_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define MPC1_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define MPC1_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define MPC1_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define MPC1_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define MPC1_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define MPC1_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define MPC1_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define MPC1_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define MPC1_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define MPC1_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define MPC1_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define MPC1_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define MPC1_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define MPC1_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define MPC1_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define MPC1_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define MPC1_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define MPC1_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define MPC1_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define MPC1_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define MPC1_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define MPC1_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define MPC1_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define MPC1_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define MPC1_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define MPC1_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define MPC1_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define MPC1_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define MPC1_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define MPC1_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define MPC1_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define MPC1_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define MPC1_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define MPC1_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define MPC1_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define MPC1_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define MPC1_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define MPC1_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define MPC1_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define MPC1_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define MPC1_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define MPC1_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define MPC1_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//PPU_SUVD_CGC_CTRL +#define PPU_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define PPU_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define PPU_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define PPU_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define PPU_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define PPU_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define PPU_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define PPU_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define PPU_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define PPU_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define PPU_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define PPU_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define PPU_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define PPU_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define PPU_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define PPU_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define PPU_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define PPU_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define PPU_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define PPU_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define PPU_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define PPU_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define PPU_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define PPU_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define PPU_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define PPU_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define PPU_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define PPU_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define PPU_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define PPU_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define PPU_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define PPU_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define PPU_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define PPU_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define PPU_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define PPU_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define PPU_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define PPU_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define PPU_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define PPU_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define PPU_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define PPU_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define PPU_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define PPU_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define PPU_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define PPU_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define PPU_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define PPU_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define PPU_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define PPU_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define PPU_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define PPU_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//SAOE_SUVD_CGC_CTRL +#define SAOE_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define SAOE_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define SAOE_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define SAOE_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define SAOE_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define SAOE_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define SAOE_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define SAOE_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define SAOE_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define SAOE_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define SAOE_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define SAOE_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define SAOE_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define SAOE_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define SAOE_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define SAOE_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define SAOE_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define SAOE_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define SAOE_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define SAOE_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define SAOE_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define SAOE_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define SAOE_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define SAOE_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define SAOE_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define SAOE_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define SAOE_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define SAOE_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define SAOE_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define SAOE_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define SAOE_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define SAOE_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define SAOE_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define SAOE_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define SAOE_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define SAOE_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define SAOE_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define SAOE_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define SAOE_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define SAOE_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define SAOE_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define SAOE_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define SAOE_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define SAOE_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define SAOE_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define SAOE_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define SAOE_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define SAOE_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define SAOE_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define SAOE_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define SAOE_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define SAOE_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//SCM_SUVD_CGC_CTRL +#define SCM_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define SCM_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define SCM_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define SCM_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define SCM_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define SCM_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define SCM_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define SCM_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define SCM_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define SCM_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define SCM_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define SCM_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define SCM_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define SCM_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define SCM_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define SCM_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define SCM_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define SCM_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define SCM_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define SCM_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define SCM_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define SCM_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define SCM_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define SCM_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define SCM_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define SCM_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define SCM_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define SCM_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define SCM_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define SCM_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define SCM_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define SCM_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define SCM_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define SCM_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define SCM_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define SCM_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define SCM_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define SCM_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define SCM_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define SCM_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define SCM_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define SCM_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define SCM_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define SCM_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define SCM_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define SCM_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define SCM_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define SCM_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define SCM_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define SCM_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define SCM_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define SCM_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//SDB_SUVD_CGC_CTRL +#define SDB_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define SDB_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define SDB_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define SDB_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define SDB_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define SDB_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define SDB_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define SDB_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define SDB_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define SDB_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define SDB_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define SDB_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define SDB_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define SDB_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define SDB_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define SDB_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define SDB_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define SDB_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define SDB_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define SDB_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define SDB_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define SDB_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define SDB_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define SDB_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define SDB_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define SDB_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define SDB_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define SDB_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define SDB_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define SDB_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define SDB_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define SDB_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define SDB_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define SDB_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define SDB_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define SDB_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define SDB_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define SDB_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define SDB_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define SDB_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define SDB_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define SDB_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define SDB_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define SDB_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define SDB_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define SDB_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define SDB_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define SDB_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define SDB_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define SDB_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define SDB_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define SDB_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//SIT0_NXT_SUVD_CGC_CTRL +#define SIT0_NXT_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define SIT0_NXT_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define SIT0_NXT_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define SIT0_NXT_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define SIT0_NXT_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define SIT0_NXT_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define SIT0_NXT_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define SIT0_NXT_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define SIT0_NXT_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define SIT0_NXT_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define SIT0_NXT_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define SIT0_NXT_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define SIT0_NXT_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define SIT0_NXT_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define SIT0_NXT_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define SIT0_NXT_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define SIT0_NXT_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define SIT0_NXT_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define SIT0_NXT_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define SIT0_NXT_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define SIT0_NXT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define SIT0_NXT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define SIT0_NXT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define SIT0_NXT_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define SIT0_NXT_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define SIT0_NXT_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define SIT0_NXT_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define SIT0_NXT_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define SIT0_NXT_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define SIT0_NXT_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define SIT0_NXT_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define SIT0_NXT_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define SIT0_NXT_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define SIT0_NXT_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define SIT0_NXT_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define SIT0_NXT_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define SIT0_NXT_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define SIT0_NXT_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define SIT0_NXT_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define SIT0_NXT_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define SIT0_NXT_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define SIT0_NXT_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define SIT0_NXT_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define SIT0_NXT_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define SIT0_NXT_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define SIT0_NXT_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define SIT0_NXT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define SIT0_NXT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define SIT0_NXT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define SIT0_NXT_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define SIT0_NXT_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define SIT0_NXT_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//SIT1_NXT_SUVD_CGC_CTRL +#define SIT1_NXT_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define SIT1_NXT_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define SIT1_NXT_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define SIT1_NXT_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define SIT1_NXT_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define SIT1_NXT_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define SIT1_NXT_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define SIT1_NXT_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define SIT1_NXT_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define SIT1_NXT_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define SIT1_NXT_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define SIT1_NXT_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define SIT1_NXT_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define SIT1_NXT_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define SIT1_NXT_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define SIT1_NXT_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define SIT1_NXT_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define SIT1_NXT_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define SIT1_NXT_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define SIT1_NXT_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define SIT1_NXT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define SIT1_NXT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define SIT1_NXT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define SIT1_NXT_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define SIT1_NXT_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define SIT1_NXT_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define SIT1_NXT_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define SIT1_NXT_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define SIT1_NXT_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define SIT1_NXT_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define SIT1_NXT_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define SIT1_NXT_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define SIT1_NXT_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define SIT1_NXT_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define SIT1_NXT_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define SIT1_NXT_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define SIT1_NXT_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define SIT1_NXT_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define SIT1_NXT_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define SIT1_NXT_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define SIT1_NXT_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define SIT1_NXT_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define SIT1_NXT_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define SIT1_NXT_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define SIT1_NXT_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define SIT1_NXT_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define SIT1_NXT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define SIT1_NXT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define SIT1_NXT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define SIT1_NXT_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define SIT1_NXT_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define SIT1_NXT_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//SIT2_NXT_SUVD_CGC_CTRL +#define SIT2_NXT_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define SIT2_NXT_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define SIT2_NXT_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define SIT2_NXT_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define SIT2_NXT_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define SIT2_NXT_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define SIT2_NXT_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define SIT2_NXT_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define SIT2_NXT_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define SIT2_NXT_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define SIT2_NXT_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define SIT2_NXT_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define SIT2_NXT_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define SIT2_NXT_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define SIT2_NXT_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define SIT2_NXT_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define SIT2_NXT_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define SIT2_NXT_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define SIT2_NXT_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define SIT2_NXT_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define SIT2_NXT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define SIT2_NXT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define SIT2_NXT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define SIT2_NXT_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define SIT2_NXT_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define SIT2_NXT_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define SIT2_NXT_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define SIT2_NXT_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define SIT2_NXT_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define SIT2_NXT_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define SIT2_NXT_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define SIT2_NXT_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define SIT2_NXT_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define SIT2_NXT_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define SIT2_NXT_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define SIT2_NXT_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define SIT2_NXT_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define SIT2_NXT_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define SIT2_NXT_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define SIT2_NXT_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define SIT2_NXT_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define SIT2_NXT_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define SIT2_NXT_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define SIT2_NXT_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define SIT2_NXT_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define SIT2_NXT_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define SIT2_NXT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define SIT2_NXT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define SIT2_NXT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define SIT2_NXT_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define SIT2_NXT_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define SIT2_NXT_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//SIT_SUVD_CGC_CTRL +#define SIT_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define SIT_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define SIT_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define SIT_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define SIT_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define SIT_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define SIT_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define SIT_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define SIT_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define SIT_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define SIT_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define SIT_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define SIT_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define SIT_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define SIT_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define SIT_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define SIT_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define SIT_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define SIT_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define SIT_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define SIT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define SIT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define SIT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define SIT_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define SIT_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define SIT_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define SIT_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define SIT_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define SIT_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define SIT_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define SIT_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define SIT_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define SIT_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define SIT_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define SIT_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define SIT_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define SIT_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define SIT_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define SIT_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define SIT_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define SIT_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define SIT_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define SIT_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define SIT_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define SIT_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define SIT_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define SIT_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define SIT_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define SIT_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define SIT_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define SIT_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define SIT_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//SMPA_SUVD_CGC_CTRL +#define SMPA_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define SMPA_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define SMPA_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define SMPA_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define SMPA_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define SMPA_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define SMPA_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define SMPA_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define SMPA_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define SMPA_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define SMPA_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define SMPA_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define SMPA_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define SMPA_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define SMPA_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define SMPA_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define SMPA_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define SMPA_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define SMPA_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define SMPA_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define SMPA_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define SMPA_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define SMPA_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define SMPA_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define SMPA_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define SMPA_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define SMPA_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define SMPA_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define SMPA_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define SMPA_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define SMPA_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define SMPA_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define SMPA_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define SMPA_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define SMPA_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define SMPA_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define SMPA_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define SMPA_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define SMPA_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define SMPA_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define SMPA_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define SMPA_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define SMPA_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define SMPA_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define SMPA_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define SMPA_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define SMPA_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define SMPA_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define SMPA_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define SMPA_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define SMPA_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define SMPA_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//SMP_SUVD_CGC_CTRL +#define SMP_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define SMP_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define SMP_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define SMP_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define SMP_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define SMP_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define SMP_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define SMP_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define SMP_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define SMP_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define SMP_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define SMP_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define SMP_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define SMP_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define SMP_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define SMP_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define SMP_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define SMP_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define SMP_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define SMP_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define SMP_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define SMP_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define SMP_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define SMP_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define SMP_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define SMP_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define SMP_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define SMP_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define SMP_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define SMP_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define SMP_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define SMP_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define SMP_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define SMP_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define SMP_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define SMP_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define SMP_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define SMP_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define SMP_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define SMP_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define SMP_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define SMP_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define SMP_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define SMP_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define SMP_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define SMP_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define SMP_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define SMP_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define SMP_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define SMP_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define SMP_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define SMP_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//SRE_SUVD_CGC_CTRL +#define SRE_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define SRE_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define SRE_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define SRE_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define SRE_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define SRE_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define SRE_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define SRE_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define SRE_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define SRE_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define SRE_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define SRE_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define SRE_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define SRE_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define SRE_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define SRE_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define SRE_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define SRE_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define SRE_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define SRE_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define SRE_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define SRE_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define SRE_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define SRE_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define SRE_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define SRE_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define SRE_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define SRE_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define SRE_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define SRE_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define SRE_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define SRE_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define SRE_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define SRE_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define SRE_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define SRE_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define SRE_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define SRE_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define SRE_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define SRE_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define SRE_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define SRE_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define SRE_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define SRE_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define SRE_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define SRE_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define SRE_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define SRE_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define SRE_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define SRE_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define SRE_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define SRE_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//UVD_MPBE0_SUVD_CGC_CTRL +#define UVD_MPBE0_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define UVD_MPBE0_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define UVD_MPBE0_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define UVD_MPBE0_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define UVD_MPBE0_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define UVD_MPBE0_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define UVD_MPBE0_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define UVD_MPBE0_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define UVD_MPBE0_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define UVD_MPBE0_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define UVD_MPBE0_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define UVD_MPBE0_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define UVD_MPBE0_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define UVD_MPBE0_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define UVD_MPBE0_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define UVD_MPBE0_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define UVD_MPBE0_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define UVD_MPBE0_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define UVD_MPBE0_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define UVD_MPBE0_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define UVD_MPBE0_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define UVD_MPBE0_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define UVD_MPBE0_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define UVD_MPBE0_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define UVD_MPBE0_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define UVD_MPBE0_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define UVD_MPBE0_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define UVD_MPBE0_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define UVD_MPBE0_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define UVD_MPBE0_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define UVD_MPBE0_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define UVD_MPBE0_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define UVD_MPBE0_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define UVD_MPBE0_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define UVD_MPBE0_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define UVD_MPBE0_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define UVD_MPBE0_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define UVD_MPBE0_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define UVD_MPBE0_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define UVD_MPBE0_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define UVD_MPBE0_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define UVD_MPBE0_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//UVD_MPBE1_SUVD_CGC_CTRL +#define UVD_MPBE1_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define UVD_MPBE1_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define UVD_MPBE1_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define UVD_MPBE1_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define UVD_MPBE1_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define UVD_MPBE1_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define UVD_MPBE1_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define UVD_MPBE1_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define UVD_MPBE1_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define UVD_MPBE1_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define UVD_MPBE1_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define UVD_MPBE1_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define UVD_MPBE1_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define UVD_MPBE1_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define UVD_MPBE1_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define UVD_MPBE1_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define UVD_MPBE1_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define UVD_MPBE1_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define UVD_MPBE1_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define UVD_MPBE1_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define UVD_MPBE1_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define UVD_MPBE1_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define UVD_MPBE1_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define UVD_MPBE1_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define UVD_MPBE1_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define UVD_MPBE1_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define UVD_MPBE1_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define UVD_MPBE1_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define UVD_MPBE1_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define UVD_MPBE1_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define UVD_MPBE1_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define UVD_MPBE1_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define UVD_MPBE1_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define UVD_MPBE1_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define UVD_MPBE1_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define UVD_MPBE1_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define UVD_MPBE1_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define UVD_MPBE1_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define UVD_MPBE1_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define UVD_MPBE1_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define UVD_MPBE1_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define UVD_MPBE1_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//UVD_SUVD_CGC_CTRL +#define UVD_SUVD_CGC_CTRL__SRE_MODE__SHIFT 0x0 +#define UVD_SUVD_CGC_CTRL__SIT_MODE__SHIFT 0x1 +#define UVD_SUVD_CGC_CTRL__SMP_MODE__SHIFT 0x2 +#define UVD_SUVD_CGC_CTRL__SCM_MODE__SHIFT 0x3 +#define UVD_SUVD_CGC_CTRL__SDB_MODE__SHIFT 0x4 +#define UVD_SUVD_CGC_CTRL__SCLR_MODE__SHIFT 0x5 +#define UVD_SUVD_CGC_CTRL__UVD_SC_MODE__SHIFT 0x6 +#define UVD_SUVD_CGC_CTRL__ENT_MODE__SHIFT 0x7 +#define UVD_SUVD_CGC_CTRL__IME_MODE__SHIFT 0x8 +#define UVD_SUVD_CGC_CTRL__SITE_MODE__SHIFT 0x9 +#define UVD_SUVD_CGC_CTRL__EFC_MODE__SHIFT 0xa +#define UVD_SUVD_CGC_CTRL__SAOE_MODE__SHIFT 0xb +#define UVD_SUVD_CGC_CTRL__SMPA_MODE__SHIFT 0xc +#define UVD_SUVD_CGC_CTRL__MPBE0_MODE__SHIFT 0xd +#define UVD_SUVD_CGC_CTRL__MPBE1_MODE__SHIFT 0xe +#define UVD_SUVD_CGC_CTRL__SIT_AV1_MODE__SHIFT 0xf +#define UVD_SUVD_CGC_CTRL__SDB_AV1_MODE__SHIFT 0x10 +#define UVD_SUVD_CGC_CTRL__MPC1_MODE__SHIFT 0x11 +#define UVD_SUVD_CGC_CTRL__AVM_0_MODE__SHIFT 0x12 +#define UVD_SUVD_CGC_CTRL__AVM_1_MODE__SHIFT 0x13 +#define UVD_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE__SHIFT 0x14 +#define UVD_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE__SHIFT 0x15 +#define UVD_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE__SHIFT 0x16 +#define UVD_SUVD_CGC_CTRL__FBC_PCLK__SHIFT 0x1c +#define UVD_SUVD_CGC_CTRL__FBC_CCLK__SHIFT 0x1d +#define UVD_SUVD_CGC_CTRL__CDEFE_MODE__SHIFT 0x1e +#define UVD_SUVD_CGC_CTRL__SRE_MODE_MASK 0x00000001L +#define UVD_SUVD_CGC_CTRL__SIT_MODE_MASK 0x00000002L +#define UVD_SUVD_CGC_CTRL__SMP_MODE_MASK 0x00000004L +#define UVD_SUVD_CGC_CTRL__SCM_MODE_MASK 0x00000008L +#define UVD_SUVD_CGC_CTRL__SDB_MODE_MASK 0x00000010L +#define UVD_SUVD_CGC_CTRL__SCLR_MODE_MASK 0x00000020L +#define UVD_SUVD_CGC_CTRL__UVD_SC_MODE_MASK 0x00000040L +#define UVD_SUVD_CGC_CTRL__ENT_MODE_MASK 0x00000080L +#define UVD_SUVD_CGC_CTRL__IME_MODE_MASK 0x00000100L +#define UVD_SUVD_CGC_CTRL__SITE_MODE_MASK 0x00000200L +#define UVD_SUVD_CGC_CTRL__EFC_MODE_MASK 0x00000400L +#define UVD_SUVD_CGC_CTRL__SAOE_MODE_MASK 0x00000800L +#define UVD_SUVD_CGC_CTRL__SMPA_MODE_MASK 0x00001000L +#define UVD_SUVD_CGC_CTRL__MPBE0_MODE_MASK 0x00002000L +#define UVD_SUVD_CGC_CTRL__MPBE1_MODE_MASK 0x00004000L +#define UVD_SUVD_CGC_CTRL__SIT_AV1_MODE_MASK 0x00008000L +#define UVD_SUVD_CGC_CTRL__SDB_AV1_MODE_MASK 0x00010000L +#define UVD_SUVD_CGC_CTRL__MPC1_MODE_MASK 0x00020000L +#define UVD_SUVD_CGC_CTRL__AVM_0_MODE_MASK 0x00040000L +#define UVD_SUVD_CGC_CTRL__AVM_1_MODE_MASK 0x00080000L +#define UVD_SUVD_CGC_CTRL__SIT_NXT_CMN_MODE_MASK 0x00100000L +#define UVD_SUVD_CGC_CTRL__SIT_NXT_DEC_MODE_MASK 0x00200000L +#define UVD_SUVD_CGC_CTRL__SIT_NXT_ENC_MODE_MASK 0x00400000L +#define UVD_SUVD_CGC_CTRL__FBC_PCLK_MASK 0x10000000L +#define UVD_SUVD_CGC_CTRL__FBC_CCLK_MASK 0x20000000L +#define UVD_SUVD_CGC_CTRL__CDEFE_MODE_MASK 0x40000000L +//UVD_CGC_CTRL3 +#define UVD_CGC_CTRL3__CGC_CLK_OFF_DELAY__SHIFT 0x0 +#define UVD_CGC_CTRL3__LCM0_MODE__SHIFT 0xb +#define UVD_CGC_CTRL3__LCM1_MODE__SHIFT 0xc +#define UVD_CGC_CTRL3__MIF_MODE__SHIFT 0xd +#define UVD_CGC_CTRL3__VREG_MODE__SHIFT 0xe +#define UVD_CGC_CTRL3__PE_MODE__SHIFT 0xf +#define UVD_CGC_CTRL3__PPU_MODE__SHIFT 0x10 +#define UVD_CGC_CTRL3__CGC_CLK_OFF_DELAY_MASK 0x000000FFL +#define UVD_CGC_CTRL3__LCM0_MODE_MASK 0x00000800L +#define UVD_CGC_CTRL3__LCM1_MODE_MASK 0x00001000L +#define UVD_CGC_CTRL3__MIF_MODE_MASK 0x00002000L +#define UVD_CGC_CTRL3__VREG_MODE_MASK 0x00004000L +#define UVD_CGC_CTRL3__PE_MODE_MASK 0x00008000L +#define UVD_CGC_CTRL3__PPU_MODE_MASK 0x00010000L +//UVD_GPCOM_VCPU_DATA0 +#define UVD_GPCOM_VCPU_DATA0__DATA0__SHIFT 0x0 +#define UVD_GPCOM_VCPU_DATA0__DATA0_MASK 0xFFFFFFFFL +//UVD_GPCOM_VCPU_DATA1 +#define UVD_GPCOM_VCPU_DATA1__DATA1__SHIFT 0x0 +#define UVD_GPCOM_VCPU_DATA1__DATA1_MASK 0xFFFFFFFFL +//UVD_GPCOM_SYS_CMD +#define UVD_GPCOM_SYS_CMD__CMD_SEND__SHIFT 0x0 +#define UVD_GPCOM_SYS_CMD__CMD__SHIFT 0x1 +#define UVD_GPCOM_SYS_CMD__CMD_SOURCE__SHIFT 0x1f +#define UVD_GPCOM_SYS_CMD__CMD_SEND_MASK 0x00000001L +#define UVD_GPCOM_SYS_CMD__CMD_MASK 0x7FFFFFFEL +#define UVD_GPCOM_SYS_CMD__CMD_SOURCE_MASK 0x80000000L +//UVD_GPCOM_SYS_DATA0 +#define UVD_GPCOM_SYS_DATA0__DATA0__SHIFT 0x0 +#define UVD_GPCOM_SYS_DATA0__DATA0_MASK 0xFFFFFFFFL +//UVD_GPCOM_SYS_DATA1 +#define UVD_GPCOM_SYS_DATA1__DATA1__SHIFT 0x0 +#define UVD_GPCOM_SYS_DATA1__DATA1_MASK 0xFFFFFFFFL +//UVD_VCPU_INT_EN +#define UVD_VCPU_INT_EN__PIF_ADDR_ERR_EN__SHIFT 0x0 +#define UVD_VCPU_INT_EN__SEMA_WAIT_FAULT_TIMEOUT_EN__SHIFT 0x1 +#define UVD_VCPU_INT_EN__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_EN__SHIFT 0x2 +#define UVD_VCPU_INT_EN__NJ_PF_RPT_EN__SHIFT 0x3 +#define UVD_VCPU_INT_EN__SW_RB1_INT_EN__SHIFT 0x4 +#define UVD_VCPU_INT_EN__SW_RB2_INT_EN__SHIFT 0x5 +#define UVD_VCPU_INT_EN__RBC_REG_PRIV_FAULT_EN__SHIFT 0x6 +#define UVD_VCPU_INT_EN__SW_RB3_INT_EN__SHIFT 0x7 +#define UVD_VCPU_INT_EN__SW_RB4_INT_EN__SHIFT 0x9 +#define UVD_VCPU_INT_EN__SW_RB5_INT_EN__SHIFT 0xa +#define UVD_VCPU_INT_EN__LBSI_EN__SHIFT 0xb +#define UVD_VCPU_INT_EN__UDEC_EN__SHIFT 0xc +#define UVD_VCPU_INT_EN__LMI_AXI_UNSUPPORTED_LEN_EN__SHIFT 0xd +#define UVD_VCPU_INT_EN__LMI_AXI_UNSUPPORTED_ADR_ALIGN_EN__SHIFT 0xe +#define UVD_VCPU_INT_EN__SUVD_EN__SHIFT 0xf +#define UVD_VCPU_INT_EN__RPTR_WR_EN__SHIFT 0x10 +#define UVD_VCPU_INT_EN__JOB_START_EN__SHIFT 0x11 +#define UVD_VCPU_INT_EN__NJ_PF_EN__SHIFT 0x12 +#define UVD_VCPU_INT_EN__SEMA_WAIT_FAIL_SIG_EN__SHIFT 0x17 +#define UVD_VCPU_INT_EN__IDCT_EN__SHIFT 0x18 +#define UVD_VCPU_INT_EN__MPRD_EN__SHIFT 0x19 +#define UVD_VCPU_INT_EN__AVM_INT_EN__SHIFT 0x1a +#define UVD_VCPU_INT_EN__CLK_SWT_EN__SHIFT 0x1b +#define UVD_VCPU_INT_EN__MIF_HWINT_EN__SHIFT 0x1c +#define UVD_VCPU_INT_EN__MPRD_ERR_EN__SHIFT 0x1d +#define UVD_VCPU_INT_EN__DRV_FW_REQ_EN__SHIFT 0x1e +#define UVD_VCPU_INT_EN__DRV_FW_ACK_EN__SHIFT 0x1f +#define UVD_VCPU_INT_EN__PIF_ADDR_ERR_EN_MASK 0x00000001L +#define UVD_VCPU_INT_EN__SEMA_WAIT_FAULT_TIMEOUT_EN_MASK 0x00000002L +#define UVD_VCPU_INT_EN__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_EN_MASK 0x00000004L +#define UVD_VCPU_INT_EN__NJ_PF_RPT_EN_MASK 0x00000008L +#define UVD_VCPU_INT_EN__SW_RB1_INT_EN_MASK 0x00000010L +#define UVD_VCPU_INT_EN__SW_RB2_INT_EN_MASK 0x00000020L +#define UVD_VCPU_INT_EN__RBC_REG_PRIV_FAULT_EN_MASK 0x00000040L +#define UVD_VCPU_INT_EN__SW_RB3_INT_EN_MASK 0x00000080L +#define UVD_VCPU_INT_EN__SW_RB4_INT_EN_MASK 0x00000200L +#define UVD_VCPU_INT_EN__SW_RB5_INT_EN_MASK 0x00000400L +#define UVD_VCPU_INT_EN__LBSI_EN_MASK 0x00000800L +#define UVD_VCPU_INT_EN__UDEC_EN_MASK 0x00001000L +#define UVD_VCPU_INT_EN__LMI_AXI_UNSUPPORTED_LEN_EN_MASK 0x00002000L +#define UVD_VCPU_INT_EN__LMI_AXI_UNSUPPORTED_ADR_ALIGN_EN_MASK 0x00004000L +#define UVD_VCPU_INT_EN__SUVD_EN_MASK 0x00008000L +#define UVD_VCPU_INT_EN__RPTR_WR_EN_MASK 0x00010000L +#define UVD_VCPU_INT_EN__JOB_START_EN_MASK 0x00020000L +#define UVD_VCPU_INT_EN__NJ_PF_EN_MASK 0x00040000L +#define UVD_VCPU_INT_EN__SEMA_WAIT_FAIL_SIG_EN_MASK 0x00800000L +#define UVD_VCPU_INT_EN__IDCT_EN_MASK 0x01000000L +#define UVD_VCPU_INT_EN__MPRD_EN_MASK 0x02000000L +#define UVD_VCPU_INT_EN__AVM_INT_EN_MASK 0x04000000L +#define UVD_VCPU_INT_EN__CLK_SWT_EN_MASK 0x08000000L +#define UVD_VCPU_INT_EN__MIF_HWINT_EN_MASK 0x10000000L +#define UVD_VCPU_INT_EN__MPRD_ERR_EN_MASK 0x20000000L +#define UVD_VCPU_INT_EN__DRV_FW_REQ_EN_MASK 0x40000000L +#define UVD_VCPU_INT_EN__DRV_FW_ACK_EN_MASK 0x80000000L +//UVD_VCPU_INT_STATUS +#define UVD_VCPU_INT_STATUS__PIF_ADDR_ERR_INT__SHIFT 0x0 +#define UVD_VCPU_INT_STATUS__SEMA_WAIT_FAULT_TIMEOUT_INT__SHIFT 0x1 +#define UVD_VCPU_INT_STATUS__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_INT__SHIFT 0x2 +#define UVD_VCPU_INT_STATUS__NJ_PF_RPT_INT__SHIFT 0x3 +#define UVD_VCPU_INT_STATUS__SW_RB1_INT__SHIFT 0x4 +#define UVD_VCPU_INT_STATUS__SW_RB2_INT__SHIFT 0x5 +#define UVD_VCPU_INT_STATUS__RBC_REG_PRIV_FAULT_INT__SHIFT 0x6 +#define UVD_VCPU_INT_STATUS__SW_RB3_INT__SHIFT 0x7 +#define UVD_VCPU_INT_STATUS__SW_RB4_INT__SHIFT 0x9 +#define UVD_VCPU_INT_STATUS__SW_RB5_INT__SHIFT 0xa +#define UVD_VCPU_INT_STATUS__LBSI_INT__SHIFT 0xb +#define UVD_VCPU_INT_STATUS__UDEC_INT__SHIFT 0xc +#define UVD_VCPU_INT_STATUS__LMI_AXI_UNSUPPORTED_LEN_INT__SHIFT 0xd +#define UVD_VCPU_INT_STATUS__LMI_AXI_UNSUPPORTED_ADR_ALIGN_INT__SHIFT 0xe +#define UVD_VCPU_INT_STATUS__SUVD_INT__SHIFT 0xf +#define UVD_VCPU_INT_STATUS__RPTR_WR_INT__SHIFT 0x10 +#define UVD_VCPU_INT_STATUS__JOB_START_INT__SHIFT 0x11 +#define UVD_VCPU_INT_STATUS__NJ_PF_INT__SHIFT 0x12 +#define UVD_VCPU_INT_STATUS__GPCOM_INT__SHIFT 0x14 +#define UVD_VCPU_INT_STATUS__SEMA_WAIT_FAIL_SIG_INT__SHIFT 0x17 +#define UVD_VCPU_INT_STATUS__IDCT_INT__SHIFT 0x18 +#define UVD_VCPU_INT_STATUS__MPRD_INT__SHIFT 0x19 +#define UVD_VCPU_INT_STATUS__AVM_INT__SHIFT 0x1a +#define UVD_VCPU_INT_STATUS__CLK_SWT_INT__SHIFT 0x1b +#define UVD_VCPU_INT_STATUS__MIF_HWINT__SHIFT 0x1c +#define UVD_VCPU_INT_STATUS__MPRD_ERR_INT__SHIFT 0x1d +#define UVD_VCPU_INT_STATUS__DRV_FW_REQ_INT__SHIFT 0x1e +#define UVD_VCPU_INT_STATUS__DRV_FW_ACK_INT__SHIFT 0x1f +#define UVD_VCPU_INT_STATUS__PIF_ADDR_ERR_INT_MASK 0x00000001L +#define UVD_VCPU_INT_STATUS__SEMA_WAIT_FAULT_TIMEOUT_INT_MASK 0x00000002L +#define UVD_VCPU_INT_STATUS__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_INT_MASK 0x00000004L +#define UVD_VCPU_INT_STATUS__NJ_PF_RPT_INT_MASK 0x00000008L +#define UVD_VCPU_INT_STATUS__SW_RB1_INT_MASK 0x00000010L +#define UVD_VCPU_INT_STATUS__SW_RB2_INT_MASK 0x00000020L +#define UVD_VCPU_INT_STATUS__RBC_REG_PRIV_FAULT_INT_MASK 0x00000040L +#define UVD_VCPU_INT_STATUS__SW_RB3_INT_MASK 0x00000080L +#define UVD_VCPU_INT_STATUS__SW_RB4_INT_MASK 0x00000200L +#define UVD_VCPU_INT_STATUS__SW_RB5_INT_MASK 0x00000400L +#define UVD_VCPU_INT_STATUS__LBSI_INT_MASK 0x00000800L +#define UVD_VCPU_INT_STATUS__UDEC_INT_MASK 0x00001000L +#define UVD_VCPU_INT_STATUS__LMI_AXI_UNSUPPORTED_LEN_INT_MASK 0x00002000L +#define UVD_VCPU_INT_STATUS__LMI_AXI_UNSUPPORTED_ADR_ALIGN_INT_MASK 0x00004000L +#define UVD_VCPU_INT_STATUS__SUVD_INT_MASK 0x00008000L +#define UVD_VCPU_INT_STATUS__RPTR_WR_INT_MASK 0x00010000L +#define UVD_VCPU_INT_STATUS__JOB_START_INT_MASK 0x00020000L +#define UVD_VCPU_INT_STATUS__NJ_PF_INT_MASK 0x00040000L +#define UVD_VCPU_INT_STATUS__GPCOM_INT_MASK 0x00100000L +#define UVD_VCPU_INT_STATUS__SEMA_WAIT_FAIL_SIG_INT_MASK 0x00800000L +#define UVD_VCPU_INT_STATUS__IDCT_INT_MASK 0x01000000L +#define UVD_VCPU_INT_STATUS__MPRD_INT_MASK 0x02000000L +#define UVD_VCPU_INT_STATUS__AVM_INT_MASK 0x04000000L +#define UVD_VCPU_INT_STATUS__CLK_SWT_INT_MASK 0x08000000L +#define UVD_VCPU_INT_STATUS__MIF_HWINT_MASK 0x10000000L +#define UVD_VCPU_INT_STATUS__MPRD_ERR_INT_MASK 0x20000000L +#define UVD_VCPU_INT_STATUS__DRV_FW_REQ_INT_MASK 0x40000000L +#define UVD_VCPU_INT_STATUS__DRV_FW_ACK_INT_MASK 0x80000000L +//UVD_VCPU_INT_ACK +#define UVD_VCPU_INT_ACK__PIF_ADDR_ERR_ACK__SHIFT 0x0 +#define UVD_VCPU_INT_ACK__SEMA_WAIT_FAULT_TIMEOUT_ACK__SHIFT 0x1 +#define UVD_VCPU_INT_ACK__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_ACK__SHIFT 0x2 +#define UVD_VCPU_INT_ACK__NJ_PF_RPT_ACK__SHIFT 0x3 +#define UVD_VCPU_INT_ACK__SW_RB1_INT_ACK__SHIFT 0x4 +#define UVD_VCPU_INT_ACK__SW_RB2_INT_ACK__SHIFT 0x5 +#define UVD_VCPU_INT_ACK__RBC_REG_PRIV_FAULT_ACK__SHIFT 0x6 +#define UVD_VCPU_INT_ACK__SW_RB3_INT_ACK__SHIFT 0x7 +#define UVD_VCPU_INT_ACK__SW_RB4_INT_ACK__SHIFT 0x9 +#define UVD_VCPU_INT_ACK__SW_RB5_INT_ACK__SHIFT 0xa +#define UVD_VCPU_INT_ACK__LBSI_ACK__SHIFT 0xb +#define UVD_VCPU_INT_ACK__UDEC_ACK__SHIFT 0xc +#define UVD_VCPU_INT_ACK__LMI_AXI_UNSUPPORTED_LEN_ACK__SHIFT 0xd +#define UVD_VCPU_INT_ACK__LMI_AXI_UNSUPPORTED_ADR_ALIGN_ACK__SHIFT 0xe +#define UVD_VCPU_INT_ACK__SUVD_ACK__SHIFT 0xf +#define UVD_VCPU_INT_ACK__RPTR_WR_ACK__SHIFT 0x10 +#define UVD_VCPU_INT_ACK__JOB_START_ACK__SHIFT 0x11 +#define UVD_VCPU_INT_ACK__NJ_PF_ACK__SHIFT 0x12 +#define UVD_VCPU_INT_ACK__SEMA_WAIT_FAIL_SIG_ACK__SHIFT 0x17 +#define UVD_VCPU_INT_ACK__IDCT_ACK__SHIFT 0x18 +#define UVD_VCPU_INT_ACK__MPRD_ACK__SHIFT 0x19 +#define UVD_VCPU_INT_ACK__AVM_INT_ACK__SHIFT 0x1a +#define UVD_VCPU_INT_ACK__CLK_SWT_ACK__SHIFT 0x1b +#define UVD_VCPU_INT_ACK__MIF_HWINT_ACK__SHIFT 0x1c +#define UVD_VCPU_INT_ACK__MPRD_ERR_ACK__SHIFT 0x1d +#define UVD_VCPU_INT_ACK__DRV_FW_REQ_ACK__SHIFT 0x1e +#define UVD_VCPU_INT_ACK__DRV_FW_ACK_ACK__SHIFT 0x1f +#define UVD_VCPU_INT_ACK__PIF_ADDR_ERR_ACK_MASK 0x00000001L +#define UVD_VCPU_INT_ACK__SEMA_WAIT_FAULT_TIMEOUT_ACK_MASK 0x00000002L +#define UVD_VCPU_INT_ACK__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_ACK_MASK 0x00000004L +#define UVD_VCPU_INT_ACK__NJ_PF_RPT_ACK_MASK 0x00000008L +#define UVD_VCPU_INT_ACK__SW_RB1_INT_ACK_MASK 0x00000010L +#define UVD_VCPU_INT_ACK__SW_RB2_INT_ACK_MASK 0x00000020L +#define UVD_VCPU_INT_ACK__RBC_REG_PRIV_FAULT_ACK_MASK 0x00000040L +#define UVD_VCPU_INT_ACK__SW_RB3_INT_ACK_MASK 0x00000080L +#define UVD_VCPU_INT_ACK__SW_RB4_INT_ACK_MASK 0x00000200L +#define UVD_VCPU_INT_ACK__SW_RB5_INT_ACK_MASK 0x00000400L +#define UVD_VCPU_INT_ACK__LBSI_ACK_MASK 0x00000800L +#define UVD_VCPU_INT_ACK__UDEC_ACK_MASK 0x00001000L +#define UVD_VCPU_INT_ACK__LMI_AXI_UNSUPPORTED_LEN_ACK_MASK 0x00002000L +#define UVD_VCPU_INT_ACK__LMI_AXI_UNSUPPORTED_ADR_ALIGN_ACK_MASK 0x00004000L +#define UVD_VCPU_INT_ACK__SUVD_ACK_MASK 0x00008000L +#define UVD_VCPU_INT_ACK__RPTR_WR_ACK_MASK 0x00010000L +#define UVD_VCPU_INT_ACK__JOB_START_ACK_MASK 0x00020000L +#define UVD_VCPU_INT_ACK__NJ_PF_ACK_MASK 0x00040000L +#define UVD_VCPU_INT_ACK__SEMA_WAIT_FAIL_SIG_ACK_MASK 0x00800000L +#define UVD_VCPU_INT_ACK__IDCT_ACK_MASK 0x01000000L +#define UVD_VCPU_INT_ACK__MPRD_ACK_MASK 0x02000000L +#define UVD_VCPU_INT_ACK__AVM_INT_ACK_MASK 0x04000000L +#define UVD_VCPU_INT_ACK__CLK_SWT_ACK_MASK 0x08000000L +#define UVD_VCPU_INT_ACK__MIF_HWINT_ACK_MASK 0x10000000L +#define UVD_VCPU_INT_ACK__MPRD_ERR_ACK_MASK 0x20000000L +#define UVD_VCPU_INT_ACK__DRV_FW_REQ_ACK_MASK 0x40000000L +#define UVD_VCPU_INT_ACK__DRV_FW_ACK_ACK_MASK 0x80000000L +//UVD_VCPU_INT_ROUTE +#define UVD_VCPU_INT_ROUTE__DRV_FW_MSG__SHIFT 0x0 +#define UVD_VCPU_INT_ROUTE__FW_DRV_MSG_ACK__SHIFT 0x1 +#define UVD_VCPU_INT_ROUTE__VCPU_GPCOM__SHIFT 0x2 +#define UVD_VCPU_INT_ROUTE__DRV_FW_MSG_MASK 0x00000001L +#define UVD_VCPU_INT_ROUTE__FW_DRV_MSG_ACK_MASK 0x00000002L +#define UVD_VCPU_INT_ROUTE__VCPU_GPCOM_MASK 0x00000004L +//UVD_DRV_FW_MSG +#define UVD_DRV_FW_MSG__MSG__SHIFT 0x0 +#define UVD_DRV_FW_MSG__MSG_MASK 0xFFFFFFFFL +//UVD_FW_DRV_MSG_ACK +#define UVD_FW_DRV_MSG_ACK__ACK__SHIFT 0x0 +#define UVD_FW_DRV_MSG_ACK__ACK_MASK 0x00000001L +//UVD_SUVD_INT_EN +#define UVD_SUVD_INT_EN__SRE_FUNC_INT_EN__SHIFT 0x0 +#define UVD_SUVD_INT_EN__SRE_ERR_INT_EN__SHIFT 0x5 +#define UVD_SUVD_INT_EN__SIT_FUNC_INT_EN__SHIFT 0x6 +#define UVD_SUVD_INT_EN__SIT_ERR_INT_EN__SHIFT 0xb +#define UVD_SUVD_INT_EN__SMP_FUNC_INT_EN__SHIFT 0xc +#define UVD_SUVD_INT_EN__SMP_ERR_INT_EN__SHIFT 0x11 +#define UVD_SUVD_INT_EN__SCM_FUNC_INT_EN__SHIFT 0x12 +#define UVD_SUVD_INT_EN__SCM_ERR_INT_EN__SHIFT 0x17 +#define UVD_SUVD_INT_EN__SDB_FUNC_INT_EN__SHIFT 0x18 +#define UVD_SUVD_INT_EN__SDB_ERR_INT_EN__SHIFT 0x1d +#define UVD_SUVD_INT_EN__FBC_ERR_INT_EN__SHIFT 0x1e +#define UVD_SUVD_INT_EN__SRE_FUNC_INT_EN_MASK 0x0000001FL +#define UVD_SUVD_INT_EN__SRE_ERR_INT_EN_MASK 0x00000020L +#define UVD_SUVD_INT_EN__SIT_FUNC_INT_EN_MASK 0x000007C0L +#define UVD_SUVD_INT_EN__SIT_ERR_INT_EN_MASK 0x00000800L +#define UVD_SUVD_INT_EN__SMP_FUNC_INT_EN_MASK 0x0001F000L +#define UVD_SUVD_INT_EN__SMP_ERR_INT_EN_MASK 0x00020000L +#define UVD_SUVD_INT_EN__SCM_FUNC_INT_EN_MASK 0x007C0000L +#define UVD_SUVD_INT_EN__SCM_ERR_INT_EN_MASK 0x00800000L +#define UVD_SUVD_INT_EN__SDB_FUNC_INT_EN_MASK 0x1F000000L +#define UVD_SUVD_INT_EN__SDB_ERR_INT_EN_MASK 0x20000000L +#define UVD_SUVD_INT_EN__FBC_ERR_INT_EN_MASK 0x40000000L +//UVD_SUVD_INT_STATUS +#define UVD_SUVD_INT_STATUS__SRE_FUNC_INT__SHIFT 0x0 +#define UVD_SUVD_INT_STATUS__SRE_ERR_INT__SHIFT 0x5 +#define UVD_SUVD_INT_STATUS__SIT_FUNC_INT__SHIFT 0x6 +#define UVD_SUVD_INT_STATUS__SIT_ERR_INT__SHIFT 0xb +#define UVD_SUVD_INT_STATUS__SMP_FUNC_INT__SHIFT 0xc +#define UVD_SUVD_INT_STATUS__SMP_ERR_INT__SHIFT 0x11 +#define UVD_SUVD_INT_STATUS__SCM_FUNC_INT__SHIFT 0x12 +#define UVD_SUVD_INT_STATUS__SCM_ERR_INT__SHIFT 0x17 +#define UVD_SUVD_INT_STATUS__SDB_FUNC_INT__SHIFT 0x18 +#define UVD_SUVD_INT_STATUS__SDB_ERR_INT__SHIFT 0x1d +#define UVD_SUVD_INT_STATUS__FBC_ERR_INT__SHIFT 0x1e +#define UVD_SUVD_INT_STATUS__SRE_FUNC_INT_MASK 0x0000001FL +#define UVD_SUVD_INT_STATUS__SRE_ERR_INT_MASK 0x00000020L +#define UVD_SUVD_INT_STATUS__SIT_FUNC_INT_MASK 0x000007C0L +#define UVD_SUVD_INT_STATUS__SIT_ERR_INT_MASK 0x00000800L +#define UVD_SUVD_INT_STATUS__SMP_FUNC_INT_MASK 0x0001F000L +#define UVD_SUVD_INT_STATUS__SMP_ERR_INT_MASK 0x00020000L +#define UVD_SUVD_INT_STATUS__SCM_FUNC_INT_MASK 0x007C0000L +#define UVD_SUVD_INT_STATUS__SCM_ERR_INT_MASK 0x00800000L +#define UVD_SUVD_INT_STATUS__SDB_FUNC_INT_MASK 0x1F000000L +#define UVD_SUVD_INT_STATUS__SDB_ERR_INT_MASK 0x20000000L +#define UVD_SUVD_INT_STATUS__FBC_ERR_INT_MASK 0x40000000L +//UVD_SUVD_INT_ACK +#define UVD_SUVD_INT_ACK__SRE_FUNC_INT_ACK__SHIFT 0x0 +#define UVD_SUVD_INT_ACK__SRE_ERR_INT_ACK__SHIFT 0x5 +#define UVD_SUVD_INT_ACK__SIT_FUNC_INT_ACK__SHIFT 0x6 +#define UVD_SUVD_INT_ACK__SIT_ERR_INT_ACK__SHIFT 0xb +#define UVD_SUVD_INT_ACK__SMP_FUNC_INT_ACK__SHIFT 0xc +#define UVD_SUVD_INT_ACK__SMP_ERR_INT_ACK__SHIFT 0x11 +#define UVD_SUVD_INT_ACK__SCM_FUNC_INT_ACK__SHIFT 0x12 +#define UVD_SUVD_INT_ACK__SCM_ERR_INT_ACK__SHIFT 0x17 +#define UVD_SUVD_INT_ACK__SDB_FUNC_INT_ACK__SHIFT 0x18 +#define UVD_SUVD_INT_ACK__SDB_ERR_INT_ACK__SHIFT 0x1d +#define UVD_SUVD_INT_ACK__FBC_ERR_INT_ACK__SHIFT 0x1e +#define UVD_SUVD_INT_ACK__SRE_FUNC_INT_ACK_MASK 0x0000001FL +#define UVD_SUVD_INT_ACK__SRE_ERR_INT_ACK_MASK 0x00000020L +#define UVD_SUVD_INT_ACK__SIT_FUNC_INT_ACK_MASK 0x000007C0L +#define UVD_SUVD_INT_ACK__SIT_ERR_INT_ACK_MASK 0x00000800L +#define UVD_SUVD_INT_ACK__SMP_FUNC_INT_ACK_MASK 0x0001F000L +#define UVD_SUVD_INT_ACK__SMP_ERR_INT_ACK_MASK 0x00020000L +#define UVD_SUVD_INT_ACK__SCM_FUNC_INT_ACK_MASK 0x007C0000L +#define UVD_SUVD_INT_ACK__SCM_ERR_INT_ACK_MASK 0x00800000L +#define UVD_SUVD_INT_ACK__SDB_FUNC_INT_ACK_MASK 0x1F000000L +#define UVD_SUVD_INT_ACK__SDB_ERR_INT_ACK_MASK 0x20000000L +#define UVD_SUVD_INT_ACK__FBC_ERR_INT_ACK_MASK 0x40000000L +//UVD_ENC_VCPU_INT_EN +#define UVD_ENC_VCPU_INT_EN__DCE_UVD_SCAN_IN_BUFMGR_EN__SHIFT 0x0 +#define UVD_ENC_VCPU_INT_EN__DCE_UVD_SCAN_IN_BUFMGR2_EN__SHIFT 0x1 +#define UVD_ENC_VCPU_INT_EN__DCE_UVD_SCAN_IN_BUFMGR3_EN__SHIFT 0x2 +#define UVD_ENC_VCPU_INT_EN__DCE_UVD_SCAN_IN_BUFMGR_EN_MASK 0x00000001L +#define UVD_ENC_VCPU_INT_EN__DCE_UVD_SCAN_IN_BUFMGR2_EN_MASK 0x00000002L +#define UVD_ENC_VCPU_INT_EN__DCE_UVD_SCAN_IN_BUFMGR3_EN_MASK 0x00000004L +//UVD_ENC_VCPU_INT_STATUS +#define UVD_ENC_VCPU_INT_STATUS__DCE_UVD_SCAN_IN_BUFMGR_INT__SHIFT 0x0 +#define UVD_ENC_VCPU_INT_STATUS__DCE_UVD_SCAN_IN_BUFMGR2_INT__SHIFT 0x1 +#define UVD_ENC_VCPU_INT_STATUS__DCE_UVD_SCAN_IN_BUFMGR3_INT__SHIFT 0x2 +#define UVD_ENC_VCPU_INT_STATUS__DCE_UVD_SCAN_IN_BUFMGR_INT_MASK 0x00000001L +#define UVD_ENC_VCPU_INT_STATUS__DCE_UVD_SCAN_IN_BUFMGR2_INT_MASK 0x00000002L +#define UVD_ENC_VCPU_INT_STATUS__DCE_UVD_SCAN_IN_BUFMGR3_INT_MASK 0x00000004L +//UVD_ENC_VCPU_INT_ACK +#define UVD_ENC_VCPU_INT_ACK__DCE_UVD_SCAN_IN_BUFMGR_ACK__SHIFT 0x0 +#define UVD_ENC_VCPU_INT_ACK__DCE_UVD_SCAN_IN_BUFMGR2_ACK__SHIFT 0x1 +#define UVD_ENC_VCPU_INT_ACK__DCE_UVD_SCAN_IN_BUFMGR3_ACK__SHIFT 0x2 +#define UVD_ENC_VCPU_INT_ACK__DCE_UVD_SCAN_IN_BUFMGR_ACK_MASK 0x00000001L +#define UVD_ENC_VCPU_INT_ACK__DCE_UVD_SCAN_IN_BUFMGR2_ACK_MASK 0x00000002L +#define UVD_ENC_VCPU_INT_ACK__DCE_UVD_SCAN_IN_BUFMGR3_ACK_MASK 0x00000004L +//UVD_MASTINT_EN +#define UVD_MASTINT_EN__OVERRUN_RST__SHIFT 0x0 +#define UVD_MASTINT_EN__VCPU_EN__SHIFT 0x1 +#define UVD_MASTINT_EN__SYS_EN__SHIFT 0x2 +#define UVD_MASTINT_EN__INT_OVERRUN__SHIFT 0x4 +#define UVD_MASTINT_EN__OVERRUN_RST_MASK 0x00000001L +#define UVD_MASTINT_EN__VCPU_EN_MASK 0x00000002L +#define UVD_MASTINT_EN__SYS_EN_MASK 0x00000004L +#define UVD_MASTINT_EN__INT_OVERRUN_MASK 0x00FFFFF0L +//UVD_SYS_INT_EN +#define UVD_SYS_INT_EN__PIF_ADDR_ERR_EN__SHIFT 0x0 +#define UVD_SYS_INT_EN__SEMA_WAIT_FAULT_TIMEOUT_EN__SHIFT 0x1 +#define UVD_SYS_INT_EN__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_EN__SHIFT 0x2 +#define UVD_SYS_INT_EN__CXW_WR_EN__SHIFT 0x3 +#define UVD_SYS_INT_EN__RBC_REG_PRIV_FAULT_EN__SHIFT 0x6 +#define UVD_SYS_INT_EN__LBSI_EN__SHIFT 0xb +#define UVD_SYS_INT_EN__UDEC_EN__SHIFT 0xc +#define UVD_SYS_INT_EN__LMI_AXI_UNSUPPORTED_LEN_EN__SHIFT 0xd +#define UVD_SYS_INT_EN__LMI_AXI_UNSUPPORTED_ADR_ALIGN_EN__SHIFT 0xe +#define UVD_SYS_INT_EN__SUVD_EN__SHIFT 0xf +#define UVD_SYS_INT_EN__JOB_DONE_EN__SHIFT 0x10 +#define UVD_SYS_INT_EN__SEMA_WAIT_FAIL_SIG_EN__SHIFT 0x17 +#define UVD_SYS_INT_EN__IDCT_EN__SHIFT 0x18 +#define UVD_SYS_INT_EN__MPRD_EN__SHIFT 0x19 +#define UVD_SYS_INT_EN__RASCNTL_VCPU_VCODEC_EN__SHIFT 0x1a +#define UVD_SYS_INT_EN__CLK_SWT_EN__SHIFT 0x1b +#define UVD_SYS_INT_EN__MIF_HWINT_EN__SHIFT 0x1c +#define UVD_SYS_INT_EN__MPRD_ERR_EN__SHIFT 0x1d +#define UVD_SYS_INT_EN__AVM_INT_EN__SHIFT 0x1f +#define UVD_SYS_INT_EN__PIF_ADDR_ERR_EN_MASK 0x00000001L +#define UVD_SYS_INT_EN__SEMA_WAIT_FAULT_TIMEOUT_EN_MASK 0x00000002L +#define UVD_SYS_INT_EN__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_EN_MASK 0x00000004L +#define UVD_SYS_INT_EN__CXW_WR_EN_MASK 0x00000008L +#define UVD_SYS_INT_EN__RBC_REG_PRIV_FAULT_EN_MASK 0x00000040L +#define UVD_SYS_INT_EN__LBSI_EN_MASK 0x00000800L +#define UVD_SYS_INT_EN__UDEC_EN_MASK 0x00001000L +#define UVD_SYS_INT_EN__LMI_AXI_UNSUPPORTED_LEN_EN_MASK 0x00002000L +#define UVD_SYS_INT_EN__LMI_AXI_UNSUPPORTED_ADR_ALIGN_EN_MASK 0x00004000L +#define UVD_SYS_INT_EN__SUVD_EN_MASK 0x00008000L +#define UVD_SYS_INT_EN__JOB_DONE_EN_MASK 0x00010000L +#define UVD_SYS_INT_EN__SEMA_WAIT_FAIL_SIG_EN_MASK 0x00800000L +#define UVD_SYS_INT_EN__IDCT_EN_MASK 0x01000000L +#define UVD_SYS_INT_EN__MPRD_EN_MASK 0x02000000L +#define UVD_SYS_INT_EN__RASCNTL_VCPU_VCODEC_EN_MASK 0x04000000L +#define UVD_SYS_INT_EN__CLK_SWT_EN_MASK 0x08000000L +#define UVD_SYS_INT_EN__MIF_HWINT_EN_MASK 0x10000000L +#define UVD_SYS_INT_EN__MPRD_ERR_EN_MASK 0x20000000L +#define UVD_SYS_INT_EN__AVM_INT_EN_MASK 0x80000000L +//UVD_SYS_INT_STATUS +#define UVD_SYS_INT_STATUS__PIF_ADDR_ERR_INT__SHIFT 0x0 +#define UVD_SYS_INT_STATUS__SEMA_WAIT_FAULT_TIMEOUT_INT__SHIFT 0x1 +#define UVD_SYS_INT_STATUS__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_INT__SHIFT 0x2 +#define UVD_SYS_INT_STATUS__CXW_WR_INT__SHIFT 0x3 +#define UVD_SYS_INT_STATUS__RBC_REG_PRIV_FAULT_INT__SHIFT 0x6 +#define UVD_SYS_INT_STATUS__LBSI_INT__SHIFT 0xb +#define UVD_SYS_INT_STATUS__UDEC_INT__SHIFT 0xc +#define UVD_SYS_INT_STATUS__LMI_AXI_UNSUPPORTED_LEN_INT__SHIFT 0xd +#define UVD_SYS_INT_STATUS__LMI_AXI_UNSUPPORTED_ADR_ALIGN_INT__SHIFT 0xe +#define UVD_SYS_INT_STATUS__SUVD_INT__SHIFT 0xf +#define UVD_SYS_INT_STATUS__JOB_DONE_INT__SHIFT 0x10 +#define UVD_SYS_INT_STATUS__GPCOM_INT__SHIFT 0x12 +#define UVD_SYS_INT_STATUS__SEMA_WAIT_FAIL_SIG_INT__SHIFT 0x17 +#define UVD_SYS_INT_STATUS__IDCT_INT__SHIFT 0x18 +#define UVD_SYS_INT_STATUS__MPRD_INT__SHIFT 0x19 +#define UVD_SYS_INT_STATUS__CLK_SWT_INT__SHIFT 0x1b +#define UVD_SYS_INT_STATUS__MIF_HWINT__SHIFT 0x1c +#define UVD_SYS_INT_STATUS__MPRD_ERR_INT__SHIFT 0x1d +#define UVD_SYS_INT_STATUS__RASCNTL_VCPU_VCODEC_INT__SHIFT 0x1e +#define UVD_SYS_INT_STATUS__AVM_INT__SHIFT 0x1f +#define UVD_SYS_INT_STATUS__PIF_ADDR_ERR_INT_MASK 0x00000001L +#define UVD_SYS_INT_STATUS__SEMA_WAIT_FAULT_TIMEOUT_INT_MASK 0x00000002L +#define UVD_SYS_INT_STATUS__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_INT_MASK 0x00000004L +#define UVD_SYS_INT_STATUS__CXW_WR_INT_MASK 0x00000008L +#define UVD_SYS_INT_STATUS__RBC_REG_PRIV_FAULT_INT_MASK 0x00000040L +#define UVD_SYS_INT_STATUS__LBSI_INT_MASK 0x00000800L +#define UVD_SYS_INT_STATUS__UDEC_INT_MASK 0x00001000L +#define UVD_SYS_INT_STATUS__LMI_AXI_UNSUPPORTED_LEN_INT_MASK 0x00002000L +#define UVD_SYS_INT_STATUS__LMI_AXI_UNSUPPORTED_ADR_ALIGN_INT_MASK 0x00004000L +#define UVD_SYS_INT_STATUS__SUVD_INT_MASK 0x00008000L +#define UVD_SYS_INT_STATUS__JOB_DONE_INT_MASK 0x00010000L +#define UVD_SYS_INT_STATUS__GPCOM_INT_MASK 0x00040000L +#define UVD_SYS_INT_STATUS__SEMA_WAIT_FAIL_SIG_INT_MASK 0x00800000L +#define UVD_SYS_INT_STATUS__IDCT_INT_MASK 0x01000000L +#define UVD_SYS_INT_STATUS__MPRD_INT_MASK 0x02000000L +#define UVD_SYS_INT_STATUS__CLK_SWT_INT_MASK 0x08000000L +#define UVD_SYS_INT_STATUS__MIF_HWINT_MASK 0x10000000L +#define UVD_SYS_INT_STATUS__MPRD_ERR_INT_MASK 0x20000000L +#define UVD_SYS_INT_STATUS__RASCNTL_VCPU_VCODEC_INT_MASK 0x40000000L +#define UVD_SYS_INT_STATUS__AVM_INT_MASK 0x80000000L +//UVD_SYS_INT_ACK +#define UVD_SYS_INT_ACK__PIF_ADDR_ERR_ACK__SHIFT 0x0 +#define UVD_SYS_INT_ACK__SEMA_WAIT_FAULT_TIMEOUT_ACK__SHIFT 0x1 +#define UVD_SYS_INT_ACK__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_ACK__SHIFT 0x2 +#define UVD_SYS_INT_ACK__CXW_WR_ACK__SHIFT 0x3 +#define UVD_SYS_INT_ACK__RBC_REG_PRIV_FAULT_ACK__SHIFT 0x6 +#define UVD_SYS_INT_ACK__LBSI_ACK__SHIFT 0xb +#define UVD_SYS_INT_ACK__UDEC_ACK__SHIFT 0xc +#define UVD_SYS_INT_ACK__LMI_AXI_UNSUPPORTED_LEN_ACK__SHIFT 0xd +#define UVD_SYS_INT_ACK__LMI_AXI_UNSUPPORTED_ADR_ALIGN_ACK__SHIFT 0xe +#define UVD_SYS_INT_ACK__SUVD_ACK__SHIFT 0xf +#define UVD_SYS_INT_ACK__JOB_DONE_ACK__SHIFT 0x10 +#define UVD_SYS_INT_ACK__SEMA_WAIT_FAIL_SIG_ACK__SHIFT 0x17 +#define UVD_SYS_INT_ACK__IDCT_ACK__SHIFT 0x18 +#define UVD_SYS_INT_ACK__MPRD_ACK__SHIFT 0x19 +#define UVD_SYS_INT_ACK__CLK_SWT_ACK__SHIFT 0x1b +#define UVD_SYS_INT_ACK__MIF_HWINT_ACK__SHIFT 0x1c +#define UVD_SYS_INT_ACK__MPRD_ERR_ACK__SHIFT 0x1d +#define UVD_SYS_INT_ACK__RASCNTL_VCPU_VCODEC_ACK__SHIFT 0x1e +#define UVD_SYS_INT_ACK__AVM_INT_ACK__SHIFT 0x1f +#define UVD_SYS_INT_ACK__PIF_ADDR_ERR_ACK_MASK 0x00000001L +#define UVD_SYS_INT_ACK__SEMA_WAIT_FAULT_TIMEOUT_ACK_MASK 0x00000002L +#define UVD_SYS_INT_ACK__SEMA_SIGNAL_INCOMPLETE_TIMEOUT_ACK_MASK 0x00000004L +#define UVD_SYS_INT_ACK__CXW_WR_ACK_MASK 0x00000008L +#define UVD_SYS_INT_ACK__RBC_REG_PRIV_FAULT_ACK_MASK 0x00000040L +#define UVD_SYS_INT_ACK__LBSI_ACK_MASK 0x00000800L +#define UVD_SYS_INT_ACK__UDEC_ACK_MASK 0x00001000L +#define UVD_SYS_INT_ACK__LMI_AXI_UNSUPPORTED_LEN_ACK_MASK 0x00002000L +#define UVD_SYS_INT_ACK__LMI_AXI_UNSUPPORTED_ADR_ALIGN_ACK_MASK 0x00004000L +#define UVD_SYS_INT_ACK__SUVD_ACK_MASK 0x00008000L +#define UVD_SYS_INT_ACK__JOB_DONE_ACK_MASK 0x00010000L +#define UVD_SYS_INT_ACK__SEMA_WAIT_FAIL_SIG_ACK_MASK 0x00800000L +#define UVD_SYS_INT_ACK__IDCT_ACK_MASK 0x01000000L +#define UVD_SYS_INT_ACK__MPRD_ACK_MASK 0x02000000L +#define UVD_SYS_INT_ACK__CLK_SWT_ACK_MASK 0x08000000L +#define UVD_SYS_INT_ACK__MIF_HWINT_ACK_MASK 0x10000000L +#define UVD_SYS_INT_ACK__MPRD_ERR_ACK_MASK 0x20000000L +#define UVD_SYS_INT_ACK__RASCNTL_VCPU_VCODEC_ACK_MASK 0x40000000L +#define UVD_SYS_INT_ACK__AVM_INT_ACK_MASK 0x80000000L +//UVD_JOB_DONE +#define UVD_JOB_DONE__JOB_DONE__SHIFT 0x0 +#define UVD_JOB_DONE__JOB_DONE_MASK 0x00000003L +//UVD_CBUF_ID +#define UVD_CBUF_ID__CBUF_ID__SHIFT 0x0 +#define UVD_CBUF_ID__CBUF_ID_MASK 0xFFFFFFFFL +//UVD_CONTEXT_ID +#define UVD_CONTEXT_ID__CONTEXT_ID__SHIFT 0x0 +#define UVD_CONTEXT_ID__CONTEXT_ID_MASK 0xFFFFFFFFL +//UVD_CONTEXT_ID2 +#define UVD_CONTEXT_ID2__CONTEXT_ID2__SHIFT 0x0 +#define UVD_CONTEXT_ID2__CONTEXT_ID2_MASK 0xFFFFFFFFL +//UVD_NO_OP +#define UVD_NO_OP__NO_OP__SHIFT 0x0 +#define UVD_NO_OP__NO_OP_MASK 0xFFFFFFFFL +//UVD_RB_BASE_LO +#define UVD_RB_BASE_LO__RB_BASE_LO__SHIFT 0x6 +#define UVD_RB_BASE_LO__RB_BASE_LO_MASK 0xFFFFFFC0L +//UVD_RB_BASE_HI +#define UVD_RB_BASE_HI__RB_BASE_HI__SHIFT 0x0 +#define UVD_RB_BASE_HI__RB_BASE_HI_MASK 0xFFFFFFFFL +//UVD_RB_SIZE +#define UVD_RB_SIZE__RB_SIZE__SHIFT 0x4 +#define UVD_RB_SIZE__RB_SIZE_MASK 0x007FFFF0L +//UVD_RB_BASE_LO2 +#define UVD_RB_BASE_LO2__RB_BASE_LO__SHIFT 0x6 +#define UVD_RB_BASE_LO2__RB_BASE_LO_MASK 0xFFFFFFC0L +//UVD_RB_BASE_HI2 +#define UVD_RB_BASE_HI2__RB_BASE_HI__SHIFT 0x0 +#define UVD_RB_BASE_HI2__RB_BASE_HI_MASK 0xFFFFFFFFL +//UVD_RB_SIZE2 +#define UVD_RB_SIZE2__RB_SIZE__SHIFT 0x4 +#define UVD_RB_SIZE2__RB_SIZE_MASK 0x007FFFF0L +//UVD_RB_BASE_LO3 +#define UVD_RB_BASE_LO3__RB_BASE_LO__SHIFT 0x6 +#define UVD_RB_BASE_LO3__RB_BASE_LO_MASK 0xFFFFFFC0L +//UVD_RB_BASE_HI3 +#define UVD_RB_BASE_HI3__RB_BASE_HI__SHIFT 0x0 +#define UVD_RB_BASE_HI3__RB_BASE_HI_MASK 0xFFFFFFFFL +//UVD_RB_SIZE3 +#define UVD_RB_SIZE3__RB_SIZE__SHIFT 0x4 +#define UVD_RB_SIZE3__RB_SIZE_MASK 0x007FFFF0L +//UVD_RB_BASE_LO4 +#define UVD_RB_BASE_LO4__RB_BASE_LO__SHIFT 0x6 +#define UVD_RB_BASE_LO4__RB_BASE_LO_MASK 0xFFFFFFC0L +//UVD_RB_BASE_HI4 +#define UVD_RB_BASE_HI4__RB_BASE_HI__SHIFT 0x0 +#define UVD_RB_BASE_HI4__RB_BASE_HI_MASK 0xFFFFFFFFL +//UVD_RB_SIZE4 +#define UVD_RB_SIZE4__RB_SIZE__SHIFT 0x4 +#define UVD_RB_SIZE4__RB_SIZE_MASK 0x007FFFF0L +//UVD_OUT_RB_BASE_LO +#define UVD_OUT_RB_BASE_LO__RB_BASE_LO__SHIFT 0x6 +#define UVD_OUT_RB_BASE_LO__RB_BASE_LO_MASK 0xFFFFFFC0L +//UVD_OUT_RB_BASE_HI +#define UVD_OUT_RB_BASE_HI__RB_BASE_HI__SHIFT 0x0 +#define UVD_OUT_RB_BASE_HI__RB_BASE_HI_MASK 0xFFFFFFFFL +//UVD_OUT_RB_SIZE +#define UVD_OUT_RB_SIZE__RB_SIZE__SHIFT 0x4 +#define UVD_OUT_RB_SIZE__RB_SIZE_MASK 0x007FFFF0L +//UVD_IOV_ACTIVE_FCN_ID +#define UVD_IOV_ACTIVE_FCN_ID__VF_ID__SHIFT 0x0 +#define UVD_IOV_ACTIVE_FCN_ID__PF_VF__SHIFT 0x1f +#define UVD_IOV_ACTIVE_FCN_ID__VF_ID_MASK 0x0000003FL +#define UVD_IOV_ACTIVE_FCN_ID__PF_VF_MASK 0x80000000L +//UVD_IOV_MAILBOX +#define UVD_IOV_MAILBOX__MAILBOX__SHIFT 0x0 +#define UVD_IOV_MAILBOX__MAILBOX_MASK 0xFFFFFFFFL +//UVD_IOV_MAILBOX_RESP +#define UVD_IOV_MAILBOX_RESP__RESP__SHIFT 0x0 +#define UVD_IOV_MAILBOX_RESP__RESP_MASK 0xFFFFFFFFL +//UVD_RB_ARB_CTRL +#define UVD_RB_ARB_CTRL__SRBM_DROP__SHIFT 0x0 +#define UVD_RB_ARB_CTRL__SRBM_DIS__SHIFT 0x1 +#define UVD_RB_ARB_CTRL__VCPU_DROP__SHIFT 0x2 +#define UVD_RB_ARB_CTRL__VCPU_DIS__SHIFT 0x3 +#define UVD_RB_ARB_CTRL__RBC_DROP__SHIFT 0x4 +#define UVD_RB_ARB_CTRL__RBC_DIS__SHIFT 0x5 +#define UVD_RB_ARB_CTRL__FWOFLD_DROP__SHIFT 0x6 +#define UVD_RB_ARB_CTRL__FWOFLD_DIS__SHIFT 0x7 +#define UVD_RB_ARB_CTRL__FAST_PATH_EN__SHIFT 0x8 +#define UVD_RB_ARB_CTRL__UVD_RB_DBG_EN__SHIFT 0x9 +#define UVD_RB_ARB_CTRL__SRBM_DROP_MASK 0x00000001L +#define UVD_RB_ARB_CTRL__SRBM_DIS_MASK 0x00000002L +#define UVD_RB_ARB_CTRL__VCPU_DROP_MASK 0x00000004L +#define UVD_RB_ARB_CTRL__VCPU_DIS_MASK 0x00000008L +#define UVD_RB_ARB_CTRL__RBC_DROP_MASK 0x00000010L +#define UVD_RB_ARB_CTRL__RBC_DIS_MASK 0x00000020L +#define UVD_RB_ARB_CTRL__FWOFLD_DROP_MASK 0x00000040L +#define UVD_RB_ARB_CTRL__FWOFLD_DIS_MASK 0x00000080L +#define UVD_RB_ARB_CTRL__FAST_PATH_EN_MASK 0x00000100L +#define UVD_RB_ARB_CTRL__UVD_RB_DBG_EN_MASK 0x00000200L +//UVD_CTX_INDEX +#define UVD_CTX_INDEX__INDEX__SHIFT 0x0 +#define UVD_CTX_INDEX__INDEX_MASK 0x000001FFL +//UVD_CTX_DATA +#define UVD_CTX_DATA__DATA__SHIFT 0x0 +#define UVD_CTX_DATA__DATA_MASK 0xFFFFFFFFL +//UVD_CXW_WR +#define UVD_CXW_WR__DAT__SHIFT 0x0 +#define UVD_CXW_WR__STAT__SHIFT 0x1f +#define UVD_CXW_WR__DAT_MASK 0x0FFFFFFFL +#define UVD_CXW_WR__STAT_MASK 0x80000000L +//UVD_CXW_WR_INT_ID +#define UVD_CXW_WR_INT_ID__ID__SHIFT 0x0 +#define UVD_CXW_WR_INT_ID__ID_MASK 0x000000FFL +//UVD_CXW_WR_INT_CTX_ID +#define UVD_CXW_WR_INT_CTX_ID__ID__SHIFT 0x0 +#define UVD_CXW_WR_INT_CTX_ID__ID_MASK 0x0FFFFFFFL +//UVD_CXW_INT_ID +#define UVD_CXW_INT_ID__ID__SHIFT 0x0 +#define UVD_CXW_INT_ID__ID_MASK 0x000000FFL +//UVD_MPEG2_ERROR +#define UVD_MPEG2_ERROR__STATUS__SHIFT 0x0 +#define UVD_MPEG2_ERROR__STATUS_MASK 0xFFFFFFFFL +//UVD_YBASE +#define UVD_YBASE__DUM__SHIFT 0x0 +#define UVD_YBASE__DUM_MASK 0xFFFFFFFFL +//UVD_UVBASE +#define UVD_UVBASE__DUM__SHIFT 0x0 +#define UVD_UVBASE__DUM_MASK 0xFFFFFFFFL +//UVD_PITCH +#define UVD_PITCH__DUM__SHIFT 0x0 +#define UVD_PITCH__DUM_MASK 0xFFFFFFFFL +//UVD_WIDTH +#define UVD_WIDTH__DUM__SHIFT 0x0 +#define UVD_WIDTH__DUM_MASK 0xFFFFFFFFL +//UVD_HEIGHT +#define UVD_HEIGHT__DUM__SHIFT 0x0 +#define UVD_HEIGHT__DUM_MASK 0xFFFFFFFFL +//UVD_PICCOUNT +#define UVD_PICCOUNT__DUM__SHIFT 0x0 +#define UVD_PICCOUNT__DUM_MASK 0xFFFFFFFFL +//UVD_MPRD_INITIAL_XY +#define UVD_MPRD_INITIAL_XY__MPRD_SCREEN_X__SHIFT 0x0 +#define UVD_MPRD_INITIAL_XY__MPRD_SCREEN_Y__SHIFT 0x10 +#define UVD_MPRD_INITIAL_XY__MPRD_SCREEN_X_MASK 0x00000FFFL +#define UVD_MPRD_INITIAL_XY__MPRD_SCREEN_Y_MASK 0x0FFF0000L +//UVD_MPEG2_CTRL +#define UVD_MPEG2_CTRL__EN__SHIFT 0x0 +#define UVD_MPEG2_CTRL__TRICK_MODE__SHIFT 0x1 +#define UVD_MPEG2_CTRL__NUM_MB_PER_JOB__SHIFT 0x10 +#define UVD_MPEG2_CTRL__EN_MASK 0x00000001L +#define UVD_MPEG2_CTRL__TRICK_MODE_MASK 0x00000002L +#define UVD_MPEG2_CTRL__NUM_MB_PER_JOB_MASK 0xFFFF0000L +//UVD_MB_CTL_BUF_BASE +#define UVD_MB_CTL_BUF_BASE__BASE__SHIFT 0x0 +#define UVD_MB_CTL_BUF_BASE__BASE_MASK 0xFFFFFFFFL +//UVD_PIC_CTL_BUF_BASE +#define UVD_PIC_CTL_BUF_BASE__BASE__SHIFT 0x0 +#define UVD_PIC_CTL_BUF_BASE__BASE_MASK 0xFFFFFFFFL +//UVD_DXVA_BUF_SIZE +#define UVD_DXVA_BUF_SIZE__PIC_SIZE__SHIFT 0x0 +#define UVD_DXVA_BUF_SIZE__MB_SIZE__SHIFT 0x10 +#define UVD_DXVA_BUF_SIZE__PIC_SIZE_MASK 0x0000FFFFL +#define UVD_DXVA_BUF_SIZE__MB_SIZE_MASK 0xFFFF0000L +//UVD_SCRATCH_NP +#define UVD_SCRATCH_NP__DATA__SHIFT 0x0 +#define UVD_SCRATCH_NP__DATA_MASK 0xFFFFFFFFL +//UVD_CLK_SWT_HANDSHAKE +#define UVD_CLK_SWT_HANDSHAKE__CLK_SWT_TYPE__SHIFT 0x0 +#define UVD_CLK_SWT_HANDSHAKE__CLK_DOMAIN_SWT__SHIFT 0x8 +#define UVD_CLK_SWT_HANDSHAKE__CLK_SWT_TYPE_MASK 0x00000003L +#define UVD_CLK_SWT_HANDSHAKE__CLK_DOMAIN_SWT_MASK 0x00000300L +//UVD_GP_SCRATCH0 +#define UVD_GP_SCRATCH0__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH0__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH1 +#define UVD_GP_SCRATCH1__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH1__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH2 +#define UVD_GP_SCRATCH2__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH2__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH3 +#define UVD_GP_SCRATCH3__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH3__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH4 +#define UVD_GP_SCRATCH4__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH4__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH5 +#define UVD_GP_SCRATCH5__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH5__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH6 +#define UVD_GP_SCRATCH6__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH6__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH7 +#define UVD_GP_SCRATCH7__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH7__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH8 +#define UVD_GP_SCRATCH8__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH8__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH9 +#define UVD_GP_SCRATCH9__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH9__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH10 +#define UVD_GP_SCRATCH10__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH10__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH11 +#define UVD_GP_SCRATCH11__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH11__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH12 +#define UVD_GP_SCRATCH12__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH12__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH13 +#define UVD_GP_SCRATCH13__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH13__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH14 +#define UVD_GP_SCRATCH14__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH14__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH15 +#define UVD_GP_SCRATCH15__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH15__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH16 +#define UVD_GP_SCRATCH16__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH16__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH17 +#define UVD_GP_SCRATCH17__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH17__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH18 +#define UVD_GP_SCRATCH18__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH18__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH19 +#define UVD_GP_SCRATCH19__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH19__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH20 +#define UVD_GP_SCRATCH20__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH20__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH21 +#define UVD_GP_SCRATCH21__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH21__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH22 +#define UVD_GP_SCRATCH22__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH22__DATA_MASK 0xFFFFFFFFL +//UVD_GP_SCRATCH23 +#define UVD_GP_SCRATCH23__DATA__SHIFT 0x0 +#define UVD_GP_SCRATCH23__DATA_MASK 0xFFFFFFFFL +//UVD_AUDIO_RB_BASE_LO +#define UVD_AUDIO_RB_BASE_LO__RB_BASE_LO__SHIFT 0x6 +#define UVD_AUDIO_RB_BASE_LO__RB_BASE_LO_MASK 0xFFFFFFC0L +//UVD_AUDIO_RB_BASE_HI +#define UVD_AUDIO_RB_BASE_HI__RB_BASE_HI__SHIFT 0x0 +#define UVD_AUDIO_RB_BASE_HI__RB_BASE_HI_MASK 0xFFFFFFFFL +//UVD_AUDIO_RB_SIZE +#define UVD_AUDIO_RB_SIZE__RB_SIZE__SHIFT 0x4 +#define UVD_AUDIO_RB_SIZE__RB_SIZE_MASK 0x007FFFF0L +//UVD_VCPU_INT_STATUS2 +#define UVD_VCPU_INT_STATUS2__SW_RB6_INT__SHIFT 0x0 +#define UVD_VCPU_INT_STATUS2__RASCNTL_VCPU_VCODEC_INT__SHIFT 0x15 +#define UVD_VCPU_INT_STATUS2__SW_RB6_INT_MASK 0x00000001L +#define UVD_VCPU_INT_STATUS2__RASCNTL_VCPU_VCODEC_INT_MASK 0x00200000L +//UVD_VCPU_INT_ACK2 +#define UVD_VCPU_INT_ACK2__SW_RB6_INT_ACK__SHIFT 0x0 +#define UVD_VCPU_INT_ACK2__RASCNTL_VCPU_VCODEC_ACK__SHIFT 0x16 +#define UVD_VCPU_INT_ACK2__SW_RB6_INT_ACK_MASK 0x00000001L +#define UVD_VCPU_INT_ACK2__RASCNTL_VCPU_VCODEC_ACK_MASK 0x00400000L +//UVD_VCPU_INT_EN2 +#define UVD_VCPU_INT_EN2__SW_RB6_INT_EN__SHIFT 0x0 +#define UVD_VCPU_INT_EN2__RASCNTL_VCPU_VCODEC_EN__SHIFT 0x1 +#define UVD_VCPU_INT_EN2__SW_RB6_INT_EN_MASK 0x00000001L +#define UVD_VCPU_INT_EN2__RASCNTL_VCPU_VCODEC_EN_MASK 0x00000002L +//UVD_SUVD_CGC_STATUS2 +#define UVD_SUVD_CGC_STATUS2__SMPA_VCLK__SHIFT 0x0 +#define UVD_SUVD_CGC_STATUS2__SMPA_DCLK__SHIFT 0x1 +#define UVD_SUVD_CGC_STATUS2__MPBE1_DCLK__SHIFT 0x3 +#define UVD_SUVD_CGC_STATUS2__SIT_AV1_DCLK__SHIFT 0x4 +#define UVD_SUVD_CGC_STATUS2__SDB_AV1_DCLK__SHIFT 0x5 +#define UVD_SUVD_CGC_STATUS2__MPC1_DCLK__SHIFT 0x6 +#define UVD_SUVD_CGC_STATUS2__MPC1_SCLK__SHIFT 0x7 +#define UVD_SUVD_CGC_STATUS2__MPC1_VCLK__SHIFT 0x8 +#define UVD_SUVD_CGC_STATUS2__SRE_AV1_ENC_DCLK__SHIFT 0x9 +#define UVD_SUVD_CGC_STATUS2__CDEFE_DCLK__SHIFT 0xa +#define UVD_SUVD_CGC_STATUS2__SIT0_DCLK__SHIFT 0xb +#define UVD_SUVD_CGC_STATUS2__SIT1_DCLK__SHIFT 0xc +#define UVD_SUVD_CGC_STATUS2__SIT2_DCLK__SHIFT 0xd +#define UVD_SUVD_CGC_STATUS2__FBC_PCLK__SHIFT 0x1c +#define UVD_SUVD_CGC_STATUS2__FBC_CCLK__SHIFT 0x1d +#define UVD_SUVD_CGC_STATUS2__SMPA_VCLK_MASK 0x00000001L +#define UVD_SUVD_CGC_STATUS2__SMPA_DCLK_MASK 0x00000002L +#define UVD_SUVD_CGC_STATUS2__MPBE1_DCLK_MASK 0x00000008L +#define UVD_SUVD_CGC_STATUS2__SIT_AV1_DCLK_MASK 0x00000010L +#define UVD_SUVD_CGC_STATUS2__SDB_AV1_DCLK_MASK 0x00000020L +#define UVD_SUVD_CGC_STATUS2__MPC1_DCLK_MASK 0x00000040L +#define UVD_SUVD_CGC_STATUS2__MPC1_SCLK_MASK 0x00000080L +#define UVD_SUVD_CGC_STATUS2__MPC1_VCLK_MASK 0x00000100L +#define UVD_SUVD_CGC_STATUS2__SRE_AV1_ENC_DCLK_MASK 0x00000200L +#define UVD_SUVD_CGC_STATUS2__CDEFE_DCLK_MASK 0x00000400L +#define UVD_SUVD_CGC_STATUS2__SIT0_DCLK_MASK 0x00000800L +#define UVD_SUVD_CGC_STATUS2__SIT1_DCLK_MASK 0x00001000L +#define UVD_SUVD_CGC_STATUS2__SIT2_DCLK_MASK 0x00002000L +#define UVD_SUVD_CGC_STATUS2__FBC_PCLK_MASK 0x10000000L +#define UVD_SUVD_CGC_STATUS2__FBC_CCLK_MASK 0x20000000L +//UVD_SUVD_INT_STATUS2 +#define UVD_SUVD_INT_STATUS2__SMPA_FUNC_INT__SHIFT 0x0 +#define UVD_SUVD_INT_STATUS2__SMPA_ERR_INT__SHIFT 0x5 +#define UVD_SUVD_INT_STATUS2__SDB_AV1_FUNC_INT__SHIFT 0x6 +#define UVD_SUVD_INT_STATUS2__SDB_AV1_ERR_INT__SHIFT 0xb +#define UVD_SUVD_INT_STATUS2__SMPA_FUNC_INT_MASK 0x0000001FL +#define UVD_SUVD_INT_STATUS2__SMPA_ERR_INT_MASK 0x00000020L +#define UVD_SUVD_INT_STATUS2__SDB_AV1_FUNC_INT_MASK 0x000007C0L +#define UVD_SUVD_INT_STATUS2__SDB_AV1_ERR_INT_MASK 0x00000800L +//UVD_SUVD_INT_EN2 +#define UVD_SUVD_INT_EN2__SMPA_FUNC_INT_EN__SHIFT 0x0 +#define UVD_SUVD_INT_EN2__SMPA_ERR_INT_EN__SHIFT 0x5 +#define UVD_SUVD_INT_EN2__SDB_AV1_FUNC_INT_EN__SHIFT 0x6 +#define UVD_SUVD_INT_EN2__SDB_AV1_ERR_INT_EN__SHIFT 0xb +#define UVD_SUVD_INT_EN2__SMPA_FUNC_INT_EN_MASK 0x0000001FL +#define UVD_SUVD_INT_EN2__SMPA_ERR_INT_EN_MASK 0x00000020L +#define UVD_SUVD_INT_EN2__SDB_AV1_FUNC_INT_EN_MASK 0x000007C0L +#define UVD_SUVD_INT_EN2__SDB_AV1_ERR_INT_EN_MASK 0x00000800L +//UVD_SUVD_INT_ACK2 +#define UVD_SUVD_INT_ACK2__SMPA_FUNC_INT_ACK__SHIFT 0x0 +#define UVD_SUVD_INT_ACK2__SMPA_ERR_INT_ACK__SHIFT 0x5 +#define UVD_SUVD_INT_ACK2__SDB_AV1_FUNC_INT_ACK__SHIFT 0x6 +#define UVD_SUVD_INT_ACK2__SDB_AV1_ERR_INT_ACK__SHIFT 0xb +#define UVD_SUVD_INT_ACK2__SMPA_FUNC_INT_ACK_MASK 0x0000001FL +#define UVD_SUVD_INT_ACK2__SMPA_ERR_INT_ACK_MASK 0x00000020L +#define UVD_SUVD_INT_ACK2__SDB_AV1_FUNC_INT_ACK_MASK 0x000007C0L +#define UVD_SUVD_INT_ACK2__SDB_AV1_ERR_INT_ACK_MASK 0x00000800L +//UVD_STATUS +#define UVD_STATUS__RBC_BUSY__SHIFT 0x0 +#define UVD_STATUS__VCPU_REPORT__SHIFT 0x1 +#define UVD_STATUS__FILL_0__SHIFT 0x8 +#define UVD_STATUS__RBC_ACCESS_GPCOM__SHIFT 0x10 +#define UVD_STATUS__DRM_BUSY__SHIFT 0x11 +#define UVD_STATUS__FILL_1__SHIFT 0x12 +#define UVD_STATUS__SYS_GPCOM_REQ__SHIFT 0x1f +#define UVD_STATUS__RBC_BUSY_MASK 0x00000001L +#define UVD_STATUS__VCPU_REPORT_MASK 0x000000FEL +#define UVD_STATUS__FILL_0_MASK 0x0000FF00L +#define UVD_STATUS__RBC_ACCESS_GPCOM_MASK 0x00010000L +#define UVD_STATUS__DRM_BUSY_MASK 0x00020000L +#define UVD_STATUS__FILL_1_MASK 0x7FFC0000L +#define UVD_STATUS__SYS_GPCOM_REQ_MASK 0x80000000L +//UVD_ENC_PIPE_BUSY +#define UVD_ENC_PIPE_BUSY__IME_BUSY__SHIFT 0x0 +#define UVD_ENC_PIPE_BUSY__SMP_BUSY__SHIFT 0x1 +#define UVD_ENC_PIPE_BUSY__SIT_BUSY__SHIFT 0x2 +#define UVD_ENC_PIPE_BUSY__SDB_BUSY__SHIFT 0x3 +#define UVD_ENC_PIPE_BUSY__ENT_BUSY__SHIFT 0x4 +#define UVD_ENC_PIPE_BUSY__ENT_HEADER_BUSY__SHIFT 0x5 +#define UVD_ENC_PIPE_BUSY__LCM_BUSY__SHIFT 0x6 +#define UVD_ENC_PIPE_BUSY__MDM_RD_CUR_BUSY__SHIFT 0x7 +#define UVD_ENC_PIPE_BUSY__MDM_RD_REF_BUSY__SHIFT 0x8 +#define UVD_ENC_PIPE_BUSY__MDM_RD_GEN_BUSY__SHIFT 0x9 +#define UVD_ENC_PIPE_BUSY__MDM_WR_RECON_BUSY__SHIFT 0xa +#define UVD_ENC_PIPE_BUSY__MDM_WR_GEN_BUSY__SHIFT 0xb +#define UVD_ENC_PIPE_BUSY__EFC_BUSY__SHIFT 0xc +#define UVD_ENC_PIPE_BUSY__MDM_PPU_BUSY__SHIFT 0xd +#define UVD_ENC_PIPE_BUSY__MIF_AUTODMA_BUSY__SHIFT 0xe +#define UVD_ENC_PIPE_BUSY__CDEFE_BUSY__SHIFT 0xf +#define UVD_ENC_PIPE_BUSY__MIF_RD_CUR_BUSY__SHIFT 0x10 +#define UVD_ENC_PIPE_BUSY__MIF_RD_REF0_BUSY__SHIFT 0x11 +#define UVD_ENC_PIPE_BUSY__MIF_WR_GEN0_BUSY__SHIFT 0x12 +#define UVD_ENC_PIPE_BUSY__MIF_RD_GEN0_BUSY__SHIFT 0x13 +#define UVD_ENC_PIPE_BUSY__MIF_WR_GEN1_BUSY__SHIFT 0x14 +#define UVD_ENC_PIPE_BUSY__MIF_RD_GEN1_BUSY__SHIFT 0x15 +#define UVD_ENC_PIPE_BUSY__MIF_WR_BSP0_BUSY__SHIFT 0x16 +#define UVD_ENC_PIPE_BUSY__MIF_WR_BSP1_BUSY__SHIFT 0x17 +#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD0_BUSY__SHIFT 0x18 +#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD1_BUSY__SHIFT 0x19 +#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD2_BUSY__SHIFT 0x1a +#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD3_BUSY__SHIFT 0x1b +#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD4_BUSY__SHIFT 0x1c +#define UVD_ENC_PIPE_BUSY__MIF_WR_BSP2_BUSY__SHIFT 0x1d +#define UVD_ENC_PIPE_BUSY__MIF_WR_BSP3_BUSY__SHIFT 0x1e +#define UVD_ENC_PIPE_BUSY__SAOE_BUSY__SHIFT 0x1f +#define UVD_ENC_PIPE_BUSY__IME_BUSY_MASK 0x00000001L +#define UVD_ENC_PIPE_BUSY__SMP_BUSY_MASK 0x00000002L +#define UVD_ENC_PIPE_BUSY__SIT_BUSY_MASK 0x00000004L +#define UVD_ENC_PIPE_BUSY__SDB_BUSY_MASK 0x00000008L +#define UVD_ENC_PIPE_BUSY__ENT_BUSY_MASK 0x00000010L +#define UVD_ENC_PIPE_BUSY__ENT_HEADER_BUSY_MASK 0x00000020L +#define UVD_ENC_PIPE_BUSY__LCM_BUSY_MASK 0x00000040L +#define UVD_ENC_PIPE_BUSY__MDM_RD_CUR_BUSY_MASK 0x00000080L +#define UVD_ENC_PIPE_BUSY__MDM_RD_REF_BUSY_MASK 0x00000100L +#define UVD_ENC_PIPE_BUSY__MDM_RD_GEN_BUSY_MASK 0x00000200L +#define UVD_ENC_PIPE_BUSY__MDM_WR_RECON_BUSY_MASK 0x00000400L +#define UVD_ENC_PIPE_BUSY__MDM_WR_GEN_BUSY_MASK 0x00000800L +#define UVD_ENC_PIPE_BUSY__EFC_BUSY_MASK 0x00001000L +#define UVD_ENC_PIPE_BUSY__MDM_PPU_BUSY_MASK 0x00002000L +#define UVD_ENC_PIPE_BUSY__MIF_AUTODMA_BUSY_MASK 0x00004000L +#define UVD_ENC_PIPE_BUSY__CDEFE_BUSY_MASK 0x00008000L +#define UVD_ENC_PIPE_BUSY__MIF_RD_CUR_BUSY_MASK 0x00010000L +#define UVD_ENC_PIPE_BUSY__MIF_RD_REF0_BUSY_MASK 0x00020000L +#define UVD_ENC_PIPE_BUSY__MIF_WR_GEN0_BUSY_MASK 0x00040000L +#define UVD_ENC_PIPE_BUSY__MIF_RD_GEN0_BUSY_MASK 0x00080000L +#define UVD_ENC_PIPE_BUSY__MIF_WR_GEN1_BUSY_MASK 0x00100000L +#define UVD_ENC_PIPE_BUSY__MIF_RD_GEN1_BUSY_MASK 0x00200000L +#define UVD_ENC_PIPE_BUSY__MIF_WR_BSP0_BUSY_MASK 0x00400000L +#define UVD_ENC_PIPE_BUSY__MIF_WR_BSP1_BUSY_MASK 0x00800000L +#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD0_BUSY_MASK 0x01000000L +#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD1_BUSY_MASK 0x02000000L +#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD2_BUSY_MASK 0x04000000L +#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD3_BUSY_MASK 0x08000000L +#define UVD_ENC_PIPE_BUSY__MIF_RD_BSD4_BUSY_MASK 0x10000000L +#define UVD_ENC_PIPE_BUSY__MIF_WR_BSP2_BUSY_MASK 0x20000000L +#define UVD_ENC_PIPE_BUSY__MIF_WR_BSP3_BUSY_MASK 0x40000000L +#define UVD_ENC_PIPE_BUSY__SAOE_BUSY_MASK 0x80000000L +//UVD_FW_POWER_STATUS +#define UVD_FW_POWER_STATUS__UVDF_PWR_OFF__SHIFT 0x0 +#define UVD_FW_POWER_STATUS__UVDTC_PWR_OFF__SHIFT 0x1 +#define UVD_FW_POWER_STATUS__UVDB_PWR_OFF__SHIFT 0x2 +#define UVD_FW_POWER_STATUS__UVDTA_PWR_OFF__SHIFT 0x3 +#define UVD_FW_POWER_STATUS__UVDTD_PWR_OFF__SHIFT 0x4 +#define UVD_FW_POWER_STATUS__UVDTE_PWR_OFF__SHIFT 0x5 +#define UVD_FW_POWER_STATUS__UVDE_PWR_OFF__SHIFT 0x6 +#define UVD_FW_POWER_STATUS__UVDAB_PWR_OFF__SHIFT 0x7 +#define UVD_FW_POWER_STATUS__UVDTB_PWR_OFF__SHIFT 0x8 +#define UVD_FW_POWER_STATUS__UVDNA_PWR_OFF__SHIFT 0x9 +#define UVD_FW_POWER_STATUS__UVDNB_PWR_OFF__SHIFT 0xa +#define UVD_FW_POWER_STATUS__UVDF_PWR_OFF_MASK 0x00000001L +#define UVD_FW_POWER_STATUS__UVDTC_PWR_OFF_MASK 0x00000002L +#define UVD_FW_POWER_STATUS__UVDB_PWR_OFF_MASK 0x00000004L +#define UVD_FW_POWER_STATUS__UVDTA_PWR_OFF_MASK 0x00000008L +#define UVD_FW_POWER_STATUS__UVDTD_PWR_OFF_MASK 0x00000010L +#define UVD_FW_POWER_STATUS__UVDTE_PWR_OFF_MASK 0x00000020L +#define UVD_FW_POWER_STATUS__UVDE_PWR_OFF_MASK 0x00000040L +#define UVD_FW_POWER_STATUS__UVDAB_PWR_OFF_MASK 0x00000080L +#define UVD_FW_POWER_STATUS__UVDTB_PWR_OFF_MASK 0x00000100L +#define UVD_FW_POWER_STATUS__UVDNA_PWR_OFF_MASK 0x00000200L +#define UVD_FW_POWER_STATUS__UVDNB_PWR_OFF_MASK 0x00000400L +//UVD_CNTL +#define UVD_CNTL__MIF_WR_LOW_THRESHOLD_BP__SHIFT 0x11 +#define UVD_CNTL__SUVD_EN__SHIFT 0x13 +#define UVD_CNTL__CABAC_MB_ACC__SHIFT 0x1c +#define UVD_CNTL__LRBBM_SAFE_SYNC_DIS__SHIFT 0x1f +#define UVD_CNTL__MIF_WR_LOW_THRESHOLD_BP_MASK 0x00020000L +#define UVD_CNTL__SUVD_EN_MASK 0x00080000L +#define UVD_CNTL__CABAC_MB_ACC_MASK 0x10000000L +#define UVD_CNTL__LRBBM_SAFE_SYNC_DIS_MASK 0x80000000L +//UVD_SOFT_RESET +#define UVD_SOFT_RESET__RBC_SOFT_RESET__SHIFT 0x0 +#define UVD_SOFT_RESET__LBSI_SOFT_RESET__SHIFT 0x1 +#define UVD_SOFT_RESET__LMI_SOFT_RESET__SHIFT 0x2 +#define UVD_SOFT_RESET__VCPU_SOFT_RESET__SHIFT 0x3 +#define UVD_SOFT_RESET__UDEC_SOFT_RESET__SHIFT 0x4 +#define UVD_SOFT_RESET__CXW_SOFT_RESET__SHIFT 0x6 +#define UVD_SOFT_RESET__TAP_SOFT_RESET__SHIFT 0x7 +#define UVD_SOFT_RESET__MPC_SOFT_RESET__SHIFT 0x8 +#define UVD_SOFT_RESET__EFC_SOFT_RESET__SHIFT 0x9 +#define UVD_SOFT_RESET__IH_SOFT_RESET__SHIFT 0xa +#define UVD_SOFT_RESET__MPRD_SOFT_RESET__SHIFT 0xb +#define UVD_SOFT_RESET__IDCT_SOFT_RESET__SHIFT 0xc +#define UVD_SOFT_RESET__LMI_UMC_SOFT_RESET__SHIFT 0xd +#define UVD_SOFT_RESET__SPH_SOFT_RESET__SHIFT 0xe +#define UVD_SOFT_RESET__MIF_SOFT_RESET__SHIFT 0xf +#define UVD_SOFT_RESET__LCM_SOFT_RESET__SHIFT 0x10 +#define UVD_SOFT_RESET__SUVD_SOFT_RESET__SHIFT 0x11 +#define UVD_SOFT_RESET__LBSI_VCLK_RESET_STATUS__SHIFT 0x12 +#define UVD_SOFT_RESET__VCPU_VCLK_RESET_STATUS__SHIFT 0x13 +#define UVD_SOFT_RESET__UDEC_VCLK_RESET_STATUS__SHIFT 0x14 +#define UVD_SOFT_RESET__UDEC_DCLK_RESET_STATUS__SHIFT 0x15 +#define UVD_SOFT_RESET__MPC_DCLK_RESET_STATUS__SHIFT 0x16 +#define UVD_SOFT_RESET__MPRD_VCLK_RESET_STATUS__SHIFT 0x17 +#define UVD_SOFT_RESET__MPRD_DCLK_RESET_STATUS__SHIFT 0x18 +#define UVD_SOFT_RESET__IDCT_VCLK_RESET_STATUS__SHIFT 0x19 +#define UVD_SOFT_RESET__MIF_DCLK_RESET_STATUS__SHIFT 0x1a +#define UVD_SOFT_RESET__LCM_DCLK_RESET_STATUS__SHIFT 0x1b +#define UVD_SOFT_RESET__SUVD_VCLK_RESET_STATUS__SHIFT 0x1c +#define UVD_SOFT_RESET__SUVD_DCLK_RESET_STATUS__SHIFT 0x1d +#define UVD_SOFT_RESET__RE_DCLK_RESET_STATUS__SHIFT 0x1e +#define UVD_SOFT_RESET__SRE_DCLK_RESET_STATUS__SHIFT 0x1f +#define UVD_SOFT_RESET__RBC_SOFT_RESET_MASK 0x00000001L +#define UVD_SOFT_RESET__LBSI_SOFT_RESET_MASK 0x00000002L +#define UVD_SOFT_RESET__LMI_SOFT_RESET_MASK 0x00000004L +#define UVD_SOFT_RESET__VCPU_SOFT_RESET_MASK 0x00000008L +#define UVD_SOFT_RESET__UDEC_SOFT_RESET_MASK 0x00000010L +#define UVD_SOFT_RESET__CXW_SOFT_RESET_MASK 0x00000040L +#define UVD_SOFT_RESET__TAP_SOFT_RESET_MASK 0x00000080L +#define UVD_SOFT_RESET__MPC_SOFT_RESET_MASK 0x00000100L +#define UVD_SOFT_RESET__EFC_SOFT_RESET_MASK 0x00000200L +#define UVD_SOFT_RESET__IH_SOFT_RESET_MASK 0x00000400L +#define UVD_SOFT_RESET__MPRD_SOFT_RESET_MASK 0x00000800L +#define UVD_SOFT_RESET__IDCT_SOFT_RESET_MASK 0x00001000L +#define UVD_SOFT_RESET__LMI_UMC_SOFT_RESET_MASK 0x00002000L +#define UVD_SOFT_RESET__SPH_SOFT_RESET_MASK 0x00004000L +#define UVD_SOFT_RESET__MIF_SOFT_RESET_MASK 0x00008000L +#define UVD_SOFT_RESET__LCM_SOFT_RESET_MASK 0x00010000L +#define UVD_SOFT_RESET__SUVD_SOFT_RESET_MASK 0x00020000L +#define UVD_SOFT_RESET__LBSI_VCLK_RESET_STATUS_MASK 0x00040000L +#define UVD_SOFT_RESET__VCPU_VCLK_RESET_STATUS_MASK 0x00080000L +#define UVD_SOFT_RESET__UDEC_VCLK_RESET_STATUS_MASK 0x00100000L +#define UVD_SOFT_RESET__UDEC_DCLK_RESET_STATUS_MASK 0x00200000L +#define UVD_SOFT_RESET__MPC_DCLK_RESET_STATUS_MASK 0x00400000L +#define UVD_SOFT_RESET__MPRD_VCLK_RESET_STATUS_MASK 0x00800000L +#define UVD_SOFT_RESET__MPRD_DCLK_RESET_STATUS_MASK 0x01000000L +#define UVD_SOFT_RESET__IDCT_VCLK_RESET_STATUS_MASK 0x02000000L +#define UVD_SOFT_RESET__MIF_DCLK_RESET_STATUS_MASK 0x04000000L +#define UVD_SOFT_RESET__LCM_DCLK_RESET_STATUS_MASK 0x08000000L +#define UVD_SOFT_RESET__SUVD_VCLK_RESET_STATUS_MASK 0x10000000L +#define UVD_SOFT_RESET__SUVD_DCLK_RESET_STATUS_MASK 0x20000000L +#define UVD_SOFT_RESET__RE_DCLK_RESET_STATUS_MASK 0x40000000L +#define UVD_SOFT_RESET__SRE_DCLK_RESET_STATUS_MASK 0x80000000L +//UVD_SOFT_RESET2 +#define UVD_SOFT_RESET2__ATOMIC_SOFT_RESET__SHIFT 0x0 +#define UVD_SOFT_RESET2__PPU_SOFT_RESET__SHIFT 0x1 +#define UVD_SOFT_RESET2__MMSCH_VCLK_RESET_STATUS__SHIFT 0x10 +#define UVD_SOFT_RESET2__MMSCH_SCLK_RESET_STATUS__SHIFT 0x11 +#define UVD_SOFT_RESET2__ATOMIC_SOFT_RESET_MASK 0x00000001L +#define UVD_SOFT_RESET2__PPU_SOFT_RESET_MASK 0x00000002L +#define UVD_SOFT_RESET2__MMSCH_VCLK_RESET_STATUS_MASK 0x00010000L +#define UVD_SOFT_RESET2__MMSCH_SCLK_RESET_STATUS_MASK 0x00020000L +//UVD_MMSCH_SOFT_RESET +#define UVD_MMSCH_SOFT_RESET__MMSCH_RESET__SHIFT 0x0 +#define UVD_MMSCH_SOFT_RESET__TAP_SOFT_RESET__SHIFT 0x1 +#define UVD_MMSCH_SOFT_RESET__MMSCH_LOCK__SHIFT 0x1f +#define UVD_MMSCH_SOFT_RESET__MMSCH_RESET_MASK 0x00000001L +#define UVD_MMSCH_SOFT_RESET__TAP_SOFT_RESET_MASK 0x00000002L +#define UVD_MMSCH_SOFT_RESET__MMSCH_LOCK_MASK 0x80000000L +//UVD_WIG_CTRL +#define UVD_WIG_CTRL__AVM_SOFT_RESET__SHIFT 0x0 +#define UVD_WIG_CTRL__ACAP_SOFT_RESET__SHIFT 0x1 +#define UVD_WIG_CTRL__WIG_SOFT_RESET__SHIFT 0x2 +#define UVD_WIG_CTRL__WIG_REGCLK_FORCE_ON__SHIFT 0x3 +#define UVD_WIG_CTRL__AVM_REGCLK_FORCE_ON__SHIFT 0x4 +#define UVD_WIG_CTRL__AVM_SOFT_RESET_MASK 0x00000001L +#define UVD_WIG_CTRL__ACAP_SOFT_RESET_MASK 0x00000002L +#define UVD_WIG_CTRL__WIG_SOFT_RESET_MASK 0x00000004L +#define UVD_WIG_CTRL__WIG_REGCLK_FORCE_ON_MASK 0x00000008L +#define UVD_WIG_CTRL__AVM_REGCLK_FORCE_ON_MASK 0x00000010L +//UVD_CGC_STATUS +#define UVD_CGC_STATUS__SYS_SCLK__SHIFT 0x0 +#define UVD_CGC_STATUS__SYS_DCLK__SHIFT 0x1 +#define UVD_CGC_STATUS__SYS_VCLK__SHIFT 0x2 +#define UVD_CGC_STATUS__UDEC_SCLK__SHIFT 0x3 +#define UVD_CGC_STATUS__UDEC_DCLK__SHIFT 0x4 +#define UVD_CGC_STATUS__UDEC_VCLK__SHIFT 0x5 +#define UVD_CGC_STATUS__MPEG2_SCLK__SHIFT 0x6 +#define UVD_CGC_STATUS__MPEG2_DCLK__SHIFT 0x7 +#define UVD_CGC_STATUS__MPEG2_VCLK__SHIFT 0x8 +#define UVD_CGC_STATUS__REGS_SCLK__SHIFT 0x9 +#define UVD_CGC_STATUS__REGS_VCLK__SHIFT 0xa +#define UVD_CGC_STATUS__RBC_SCLK__SHIFT 0xb +#define UVD_CGC_STATUS__LMI_MC_SCLK__SHIFT 0xc +#define UVD_CGC_STATUS__LMI_UMC_SCLK__SHIFT 0xd +#define UVD_CGC_STATUS__IDCT_SCLK__SHIFT 0xe +#define UVD_CGC_STATUS__IDCT_VCLK__SHIFT 0xf +#define UVD_CGC_STATUS__MPRD_SCLK__SHIFT 0x10 +#define UVD_CGC_STATUS__MPRD_DCLK__SHIFT 0x11 +#define UVD_CGC_STATUS__MPRD_VCLK__SHIFT 0x12 +#define UVD_CGC_STATUS__MPC_SCLK__SHIFT 0x13 +#define UVD_CGC_STATUS__MPC_DCLK__SHIFT 0x14 +#define UVD_CGC_STATUS__LBSI_SCLK__SHIFT 0x15 +#define UVD_CGC_STATUS__LBSI_VCLK__SHIFT 0x16 +#define UVD_CGC_STATUS__LRBBM_SCLK__SHIFT 0x17 +#define UVD_CGC_STATUS__WCB_SCLK__SHIFT 0x18 +#define UVD_CGC_STATUS__VCPU_SCLK__SHIFT 0x19 +#define UVD_CGC_STATUS__VCPU_VCLK__SHIFT 0x1a +#define UVD_CGC_STATUS__MMSCH_SCLK__SHIFT 0x1b +#define UVD_CGC_STATUS__MMSCH_VCLK__SHIFT 0x1c +#define UVD_CGC_STATUS__ALL_ENC_ACTIVE__SHIFT 0x1d +#define UVD_CGC_STATUS__LRBBM_DCLK__SHIFT 0x1e +#define UVD_CGC_STATUS__ALL_DEC_ACTIVE__SHIFT 0x1f +#define UVD_CGC_STATUS__SYS_SCLK_MASK 0x00000001L +#define UVD_CGC_STATUS__SYS_DCLK_MASK 0x00000002L +#define UVD_CGC_STATUS__SYS_VCLK_MASK 0x00000004L +#define UVD_CGC_STATUS__UDEC_SCLK_MASK 0x00000008L +#define UVD_CGC_STATUS__UDEC_DCLK_MASK 0x00000010L +#define UVD_CGC_STATUS__UDEC_VCLK_MASK 0x00000020L +#define UVD_CGC_STATUS__MPEG2_SCLK_MASK 0x00000040L +#define UVD_CGC_STATUS__MPEG2_DCLK_MASK 0x00000080L +#define UVD_CGC_STATUS__MPEG2_VCLK_MASK 0x00000100L +#define UVD_CGC_STATUS__REGS_SCLK_MASK 0x00000200L +#define UVD_CGC_STATUS__REGS_VCLK_MASK 0x00000400L +#define UVD_CGC_STATUS__RBC_SCLK_MASK 0x00000800L +#define UVD_CGC_STATUS__LMI_MC_SCLK_MASK 0x00001000L +#define UVD_CGC_STATUS__LMI_UMC_SCLK_MASK 0x00002000L +#define UVD_CGC_STATUS__IDCT_SCLK_MASK 0x00004000L +#define UVD_CGC_STATUS__IDCT_VCLK_MASK 0x00008000L +#define UVD_CGC_STATUS__MPRD_SCLK_MASK 0x00010000L +#define UVD_CGC_STATUS__MPRD_DCLK_MASK 0x00020000L +#define UVD_CGC_STATUS__MPRD_VCLK_MASK 0x00040000L +#define UVD_CGC_STATUS__MPC_SCLK_MASK 0x00080000L +#define UVD_CGC_STATUS__MPC_DCLK_MASK 0x00100000L +#define UVD_CGC_STATUS__LBSI_SCLK_MASK 0x00200000L +#define UVD_CGC_STATUS__LBSI_VCLK_MASK 0x00400000L +#define UVD_CGC_STATUS__LRBBM_SCLK_MASK 0x00800000L +#define UVD_CGC_STATUS__WCB_SCLK_MASK 0x01000000L +#define UVD_CGC_STATUS__VCPU_SCLK_MASK 0x02000000L +#define UVD_CGC_STATUS__VCPU_VCLK_MASK 0x04000000L +#define UVD_CGC_STATUS__MMSCH_SCLK_MASK 0x08000000L +#define UVD_CGC_STATUS__MMSCH_VCLK_MASK 0x10000000L +#define UVD_CGC_STATUS__ALL_ENC_ACTIVE_MASK 0x20000000L +#define UVD_CGC_STATUS__LRBBM_DCLK_MASK 0x40000000L +#define UVD_CGC_STATUS__ALL_DEC_ACTIVE_MASK 0x80000000L +//UVD_CGC_UDEC_STATUS +#define UVD_CGC_UDEC_STATUS__RE_SCLK__SHIFT 0x0 +#define UVD_CGC_UDEC_STATUS__RE_DCLK__SHIFT 0x1 +#define UVD_CGC_UDEC_STATUS__RE_VCLK__SHIFT 0x2 +#define UVD_CGC_UDEC_STATUS__CM_SCLK__SHIFT 0x3 +#define UVD_CGC_UDEC_STATUS__CM_DCLK__SHIFT 0x4 +#define UVD_CGC_UDEC_STATUS__CM_VCLK__SHIFT 0x5 +#define UVD_CGC_UDEC_STATUS__IT_SCLK__SHIFT 0x6 +#define UVD_CGC_UDEC_STATUS__IT_DCLK__SHIFT 0x7 +#define UVD_CGC_UDEC_STATUS__IT_VCLK__SHIFT 0x8 +#define UVD_CGC_UDEC_STATUS__DB_SCLK__SHIFT 0x9 +#define UVD_CGC_UDEC_STATUS__DB_DCLK__SHIFT 0xa +#define UVD_CGC_UDEC_STATUS__DB_VCLK__SHIFT 0xb +#define UVD_CGC_UDEC_STATUS__MP_SCLK__SHIFT 0xc +#define UVD_CGC_UDEC_STATUS__MP_DCLK__SHIFT 0xd +#define UVD_CGC_UDEC_STATUS__MP_VCLK__SHIFT 0xe +#define UVD_CGC_UDEC_STATUS__RE_SCLK_MASK 0x00000001L +#define UVD_CGC_UDEC_STATUS__RE_DCLK_MASK 0x00000002L +#define UVD_CGC_UDEC_STATUS__RE_VCLK_MASK 0x00000004L +#define UVD_CGC_UDEC_STATUS__CM_SCLK_MASK 0x00000008L +#define UVD_CGC_UDEC_STATUS__CM_DCLK_MASK 0x00000010L +#define UVD_CGC_UDEC_STATUS__CM_VCLK_MASK 0x00000020L +#define UVD_CGC_UDEC_STATUS__IT_SCLK_MASK 0x00000040L +#define UVD_CGC_UDEC_STATUS__IT_DCLK_MASK 0x00000080L +#define UVD_CGC_UDEC_STATUS__IT_VCLK_MASK 0x00000100L +#define UVD_CGC_UDEC_STATUS__DB_SCLK_MASK 0x00000200L +#define UVD_CGC_UDEC_STATUS__DB_DCLK_MASK 0x00000400L +#define UVD_CGC_UDEC_STATUS__DB_VCLK_MASK 0x00000800L +#define UVD_CGC_UDEC_STATUS__MP_SCLK_MASK 0x00001000L +#define UVD_CGC_UDEC_STATUS__MP_DCLK_MASK 0x00002000L +#define UVD_CGC_UDEC_STATUS__MP_VCLK_MASK 0x00004000L +//UVD_SUVD_CGC_STATUS +#define UVD_SUVD_CGC_STATUS__SRE_VCLK__SHIFT 0x0 +#define UVD_SUVD_CGC_STATUS__SRE_DCLK__SHIFT 0x1 +#define UVD_SUVD_CGC_STATUS__SIT_DCLK__SHIFT 0x2 +#define UVD_SUVD_CGC_STATUS__SMP_DCLK__SHIFT 0x3 +#define UVD_SUVD_CGC_STATUS__SCM_DCLK__SHIFT 0x4 +#define UVD_SUVD_CGC_STATUS__SDB_DCLK__SHIFT 0x5 +#define UVD_SUVD_CGC_STATUS__SRE_H264_VCLK__SHIFT 0x6 +#define UVD_SUVD_CGC_STATUS__SRE_HEVC_VCLK__SHIFT 0x7 +#define UVD_SUVD_CGC_STATUS__SIT_H264_DCLK__SHIFT 0x8 +#define UVD_SUVD_CGC_STATUS__SIT_HEVC_DCLK__SHIFT 0x9 +#define UVD_SUVD_CGC_STATUS__SCM_H264_DCLK__SHIFT 0xa +#define UVD_SUVD_CGC_STATUS__SCM_HEVC_DCLK__SHIFT 0xb +#define UVD_SUVD_CGC_STATUS__SDB_H264_DCLK__SHIFT 0xc +#define UVD_SUVD_CGC_STATUS__SDB_HEVC_DCLK__SHIFT 0xd +#define UVD_SUVD_CGC_STATUS__SCLR_DCLK__SHIFT 0xe +#define UVD_SUVD_CGC_STATUS__UVD_SC__SHIFT 0xf +#define UVD_SUVD_CGC_STATUS__ENT_DCLK__SHIFT 0x10 +#define UVD_SUVD_CGC_STATUS__IME_DCLK__SHIFT 0x11 +#define UVD_SUVD_CGC_STATUS__SIT_HEVC_DEC_DCLK__SHIFT 0x12 +#define UVD_SUVD_CGC_STATUS__SIT_HEVC_ENC_DCLK__SHIFT 0x13 +#define UVD_SUVD_CGC_STATUS__SITE_DCLK__SHIFT 0x14 +#define UVD_SUVD_CGC_STATUS__SITE_HEVC_DCLK__SHIFT 0x15 +#define UVD_SUVD_CGC_STATUS__SITE_HEVC_ENC_DCLK__SHIFT 0x16 +#define UVD_SUVD_CGC_STATUS__SRE_VP9_VCLK__SHIFT 0x17 +#define UVD_SUVD_CGC_STATUS__SCM_VP9_VCLK__SHIFT 0x18 +#define UVD_SUVD_CGC_STATUS__SIT_VP9_DEC_DCLK__SHIFT 0x19 +#define UVD_SUVD_CGC_STATUS__SDB_VP9_DCLK__SHIFT 0x1a +#define UVD_SUVD_CGC_STATUS__IME_HEVC_DCLK__SHIFT 0x1b +#define UVD_SUVD_CGC_STATUS__EFC_DCLK__SHIFT 0x1c +#define UVD_SUVD_CGC_STATUS__SAOE_DCLK__SHIFT 0x1d +#define UVD_SUVD_CGC_STATUS__SRE_AV1_VCLK__SHIFT 0x1e +#define UVD_SUVD_CGC_STATUS__SCM_AV1_DCLK__SHIFT 0x1f +#define UVD_SUVD_CGC_STATUS__SRE_VCLK_MASK 0x00000001L +#define UVD_SUVD_CGC_STATUS__SRE_DCLK_MASK 0x00000002L +#define UVD_SUVD_CGC_STATUS__SIT_DCLK_MASK 0x00000004L +#define UVD_SUVD_CGC_STATUS__SMP_DCLK_MASK 0x00000008L +#define UVD_SUVD_CGC_STATUS__SCM_DCLK_MASK 0x00000010L +#define UVD_SUVD_CGC_STATUS__SDB_DCLK_MASK 0x00000020L +#define UVD_SUVD_CGC_STATUS__SRE_H264_VCLK_MASK 0x00000040L +#define UVD_SUVD_CGC_STATUS__SRE_HEVC_VCLK_MASK 0x00000080L +#define UVD_SUVD_CGC_STATUS__SIT_H264_DCLK_MASK 0x00000100L +#define UVD_SUVD_CGC_STATUS__SIT_HEVC_DCLK_MASK 0x00000200L +#define UVD_SUVD_CGC_STATUS__SCM_H264_DCLK_MASK 0x00000400L +#define UVD_SUVD_CGC_STATUS__SCM_HEVC_DCLK_MASK 0x00000800L +#define UVD_SUVD_CGC_STATUS__SDB_H264_DCLK_MASK 0x00001000L +#define UVD_SUVD_CGC_STATUS__SDB_HEVC_DCLK_MASK 0x00002000L +#define UVD_SUVD_CGC_STATUS__SCLR_DCLK_MASK 0x00004000L +#define UVD_SUVD_CGC_STATUS__UVD_SC_MASK 0x00008000L +#define UVD_SUVD_CGC_STATUS__ENT_DCLK_MASK 0x00010000L +#define UVD_SUVD_CGC_STATUS__IME_DCLK_MASK 0x00020000L +#define UVD_SUVD_CGC_STATUS__SIT_HEVC_DEC_DCLK_MASK 0x00040000L +#define UVD_SUVD_CGC_STATUS__SIT_HEVC_ENC_DCLK_MASK 0x00080000L +#define UVD_SUVD_CGC_STATUS__SITE_DCLK_MASK 0x00100000L +#define UVD_SUVD_CGC_STATUS__SITE_HEVC_DCLK_MASK 0x00200000L +#define UVD_SUVD_CGC_STATUS__SITE_HEVC_ENC_DCLK_MASK 0x00400000L +#define UVD_SUVD_CGC_STATUS__SRE_VP9_VCLK_MASK 0x00800000L +#define UVD_SUVD_CGC_STATUS__SCM_VP9_VCLK_MASK 0x01000000L +#define UVD_SUVD_CGC_STATUS__SIT_VP9_DEC_DCLK_MASK 0x02000000L +#define UVD_SUVD_CGC_STATUS__SDB_VP9_DCLK_MASK 0x04000000L +#define UVD_SUVD_CGC_STATUS__IME_HEVC_DCLK_MASK 0x08000000L +#define UVD_SUVD_CGC_STATUS__EFC_DCLK_MASK 0x10000000L +#define UVD_SUVD_CGC_STATUS__SAOE_DCLK_MASK 0x20000000L +#define UVD_SUVD_CGC_STATUS__SRE_AV1_VCLK_MASK 0x40000000L +#define UVD_SUVD_CGC_STATUS__SCM_AV1_DCLK_MASK 0x80000000L +//UVD_GPCOM_VCPU_CMD +#define UVD_GPCOM_VCPU_CMD__CMD_SEND__SHIFT 0x0 +#define UVD_GPCOM_VCPU_CMD__CMD__SHIFT 0x1 +#define UVD_GPCOM_VCPU_CMD__CMD_SOURCE__SHIFT 0x1f +#define UVD_GPCOM_VCPU_CMD__CMD_SEND_MASK 0x00000001L +#define UVD_GPCOM_VCPU_CMD__CMD_MASK 0x7FFFFFFEL +#define UVD_GPCOM_VCPU_CMD__CMD_SOURCE_MASK 0x80000000L + + +// addressBlock: aid_uvd0_ecpudec +//UVD_VCPU_CACHE_OFFSET0 +#define UVD_VCPU_CACHE_OFFSET0__CACHE_OFFSET0__SHIFT 0x0 +#define UVD_VCPU_CACHE_OFFSET0__CACHE_OFFSET0_MASK 0x001FFFFFL +//UVD_VCPU_CACHE_SIZE0 +#define UVD_VCPU_CACHE_SIZE0__CACHE_SIZE0__SHIFT 0x0 +#define UVD_VCPU_CACHE_SIZE0__CACHE_SIZE0_MASK 0x001FFFFFL +//UVD_VCPU_CACHE_OFFSET1 +#define UVD_VCPU_CACHE_OFFSET1__CACHE_OFFSET1__SHIFT 0x0 +#define UVD_VCPU_CACHE_OFFSET1__CACHE_OFFSET1_MASK 0x001FFFFFL +//UVD_VCPU_CACHE_SIZE1 +#define UVD_VCPU_CACHE_SIZE1__CACHE_SIZE1__SHIFT 0x0 +#define UVD_VCPU_CACHE_SIZE1__CACHE_SIZE1_MASK 0x001FFFFFL +//UVD_VCPU_CACHE_OFFSET2 +#define UVD_VCPU_CACHE_OFFSET2__CACHE_OFFSET2__SHIFT 0x0 +#define UVD_VCPU_CACHE_OFFSET2__CACHE_OFFSET2_MASK 0x001FFFFFL +//UVD_VCPU_CACHE_SIZE2 +#define UVD_VCPU_CACHE_SIZE2__CACHE_SIZE2__SHIFT 0x0 +#define UVD_VCPU_CACHE_SIZE2__CACHE_SIZE2_MASK 0x001FFFFFL +//UVD_VCPU_CACHE_OFFSET3 +#define UVD_VCPU_CACHE_OFFSET3__CACHE_OFFSET3__SHIFT 0x0 +#define UVD_VCPU_CACHE_OFFSET3__CACHE_OFFSET3_MASK 0x001FFFFFL +//UVD_VCPU_CACHE_SIZE3 +#define UVD_VCPU_CACHE_SIZE3__CACHE_SIZE3__SHIFT 0x0 +#define UVD_VCPU_CACHE_SIZE3__CACHE_SIZE3_MASK 0x001FFFFFL +//UVD_VCPU_CACHE_OFFSET4 +#define UVD_VCPU_CACHE_OFFSET4__CACHE_OFFSET4__SHIFT 0x0 +#define UVD_VCPU_CACHE_OFFSET4__CACHE_OFFSET4_MASK 0x001FFFFFL +//UVD_VCPU_CACHE_SIZE4 +#define UVD_VCPU_CACHE_SIZE4__CACHE_SIZE4__SHIFT 0x0 +#define UVD_VCPU_CACHE_SIZE4__CACHE_SIZE4_MASK 0x001FFFFFL +//UVD_VCPU_CACHE_OFFSET5 +#define UVD_VCPU_CACHE_OFFSET5__CACHE_OFFSET5__SHIFT 0x0 +#define UVD_VCPU_CACHE_OFFSET5__CACHE_OFFSET5_MASK 0x001FFFFFL +//UVD_VCPU_CACHE_SIZE5 +#define UVD_VCPU_CACHE_SIZE5__CACHE_SIZE5__SHIFT 0x0 +#define UVD_VCPU_CACHE_SIZE5__CACHE_SIZE5_MASK 0x001FFFFFL +//UVD_VCPU_CACHE_OFFSET6 +#define UVD_VCPU_CACHE_OFFSET6__CACHE_OFFSET6__SHIFT 0x0 +#define UVD_VCPU_CACHE_OFFSET6__CACHE_OFFSET6_MASK 0x001FFFFFL +//UVD_VCPU_CACHE_SIZE6 +#define UVD_VCPU_CACHE_SIZE6__CACHE_SIZE6__SHIFT 0x0 +#define UVD_VCPU_CACHE_SIZE6__CACHE_SIZE6_MASK 0x001FFFFFL +//UVD_VCPU_CACHE_OFFSET7 +#define UVD_VCPU_CACHE_OFFSET7__CACHE_OFFSET7__SHIFT 0x0 +#define UVD_VCPU_CACHE_OFFSET7__CACHE_OFFSET7_MASK 0x001FFFFFL +//UVD_VCPU_CACHE_SIZE7 +#define UVD_VCPU_CACHE_SIZE7__CACHE_SIZE7__SHIFT 0x0 +#define UVD_VCPU_CACHE_SIZE7__CACHE_SIZE7_MASK 0x001FFFFFL +//UVD_VCPU_CACHE_OFFSET8 +#define UVD_VCPU_CACHE_OFFSET8__CACHE_OFFSET8__SHIFT 0x0 +#define UVD_VCPU_CACHE_OFFSET8__CACHE_OFFSET8_MASK 0x001FFFFFL +//UVD_VCPU_CACHE_SIZE8 +#define UVD_VCPU_CACHE_SIZE8__CACHE_SIZE8__SHIFT 0x0 +#define UVD_VCPU_CACHE_SIZE8__CACHE_SIZE8_MASK 0x001FFFFFL +//UVD_VCPU_NONCACHE_OFFSET0 +#define UVD_VCPU_NONCACHE_OFFSET0__NONCACHE_OFFSET0__SHIFT 0x0 +#define UVD_VCPU_NONCACHE_OFFSET0__NONCACHE_OFFSET0_MASK 0x01FFFFFFL +//UVD_VCPU_NONCACHE_SIZE0 +#define UVD_VCPU_NONCACHE_SIZE0__NONCACHE_SIZE0__SHIFT 0x0 +#define UVD_VCPU_NONCACHE_SIZE0__NONCACHE_SIZE0_MASK 0x001FFFFFL +//UVD_VCPU_NONCACHE_OFFSET1 +#define UVD_VCPU_NONCACHE_OFFSET1__NONCACHE_OFFSET1__SHIFT 0x0 +#define UVD_VCPU_NONCACHE_OFFSET1__NONCACHE_OFFSET1_MASK 0x01FFFFFFL +//UVD_VCPU_NONCACHE_SIZE1 +#define UVD_VCPU_NONCACHE_SIZE1__NONCACHE_SIZE1__SHIFT 0x0 +#define UVD_VCPU_NONCACHE_SIZE1__NONCACHE_SIZE1_MASK 0x001FFFFFL +//UVD_VCPU_CNTL +#define UVD_VCPU_CNTL__IRQ_ERR__SHIFT 0x0 +#define UVD_VCPU_CNTL__AXI_MAX_BRST_SIZE_IS_4__SHIFT 0x4 +#define UVD_VCPU_CNTL__PMB_ED_ENABLE__SHIFT 0x5 +#define UVD_VCPU_CNTL__PMB_SOFT_RESET__SHIFT 0x6 +#define UVD_VCPU_CNTL__RBBM_SOFT_RESET__SHIFT 0x7 +#define UVD_VCPU_CNTL__ABORT_REQ__SHIFT 0x8 +#define UVD_VCPU_CNTL__CLK_EN__SHIFT 0x9 +#define UVD_VCPU_CNTL__TRCE_EN__SHIFT 0xa +#define UVD_VCPU_CNTL__TRCE_MUX__SHIFT 0xb +#define UVD_VCPU_CNTL__DBG_MUX__SHIFT 0xd +#define UVD_VCPU_CNTL__JTAG_EN__SHIFT 0x10 +#define UVD_VCPU_CNTL__TIMEOUT_DIS__SHIFT 0x12 +#define UVD_VCPU_CNTL__PRB_TIMEOUT_VAL__SHIFT 0x14 +#define UVD_VCPU_CNTL__BLK_RST__SHIFT 0x1c +#define UVD_VCPU_CNTL__RUNSTALL__SHIFT 0x1d +#define UVD_VCPU_CNTL__SRE_CMDIF_DRST__SHIFT 0x1e +#define UVD_VCPU_CNTL__SRE_CMDIF_VRST__SHIFT 0x1f +#define UVD_VCPU_CNTL__IRQ_ERR_MASK 0x0000000FL +#define UVD_VCPU_CNTL__AXI_MAX_BRST_SIZE_IS_4_MASK 0x00000010L +#define UVD_VCPU_CNTL__PMB_ED_ENABLE_MASK 0x00000020L +#define UVD_VCPU_CNTL__PMB_SOFT_RESET_MASK 0x00000040L +#define UVD_VCPU_CNTL__RBBM_SOFT_RESET_MASK 0x00000080L +#define UVD_VCPU_CNTL__ABORT_REQ_MASK 0x00000100L +#define UVD_VCPU_CNTL__CLK_EN_MASK 0x00000200L +#define UVD_VCPU_CNTL__TRCE_EN_MASK 0x00000400L +#define UVD_VCPU_CNTL__TRCE_MUX_MASK 0x00001800L +#define UVD_VCPU_CNTL__DBG_MUX_MASK 0x0000E000L +#define UVD_VCPU_CNTL__JTAG_EN_MASK 0x00010000L +#define UVD_VCPU_CNTL__TIMEOUT_DIS_MASK 0x00040000L +#define UVD_VCPU_CNTL__PRB_TIMEOUT_VAL_MASK 0x0FF00000L +#define UVD_VCPU_CNTL__BLK_RST_MASK 0x10000000L +#define UVD_VCPU_CNTL__RUNSTALL_MASK 0x20000000L +#define UVD_VCPU_CNTL__SRE_CMDIF_DRST_MASK 0x40000000L +#define UVD_VCPU_CNTL__SRE_CMDIF_VRST_MASK 0x80000000L +//UVD_VCPU_PRID +#define UVD_VCPU_PRID__PRID__SHIFT 0x0 +#define UVD_VCPU_PRID__PRID_MASK 0x0000FFFFL +//UVD_VCPU_TRCE +#define UVD_VCPU_TRCE__PC__SHIFT 0x0 +#define UVD_VCPU_TRCE__PC_MASK 0x0FFFFFFFL +//UVD_VCPU_TRCE_RD +#define UVD_VCPU_TRCE_RD__DATA__SHIFT 0x0 +#define UVD_VCPU_TRCE_RD__DATA_MASK 0xFFFFFFFFL +//UVD_VCPU_IND_INDEX +#define UVD_VCPU_IND_INDEX__INDEX__SHIFT 0x0 +#define UVD_VCPU_IND_INDEX__INDEX_MASK 0x000001FFL +//UVD_VCPU_IND_DATA +#define UVD_VCPU_IND_DATA__DATA__SHIFT 0x0 +#define UVD_VCPU_IND_DATA__DATA_MASK 0xFFFFFFFFL + + +// addressBlock: aid_uvd0_uvd_mpcdec +//UVD_MP_SWAP_CNTL +#define UVD_MP_SWAP_CNTL__MP_REF0_MC_SWAP__SHIFT 0x0 +#define UVD_MP_SWAP_CNTL__MP_REF1_MC_SWAP__SHIFT 0x2 +#define UVD_MP_SWAP_CNTL__MP_REF2_MC_SWAP__SHIFT 0x4 +#define UVD_MP_SWAP_CNTL__MP_REF3_MC_SWAP__SHIFT 0x6 +#define UVD_MP_SWAP_CNTL__MP_REF4_MC_SWAP__SHIFT 0x8 +#define UVD_MP_SWAP_CNTL__MP_REF5_MC_SWAP__SHIFT 0xa +#define UVD_MP_SWAP_CNTL__MP_REF6_MC_SWAP__SHIFT 0xc +#define UVD_MP_SWAP_CNTL__MP_REF7_MC_SWAP__SHIFT 0xe +#define UVD_MP_SWAP_CNTL__MP_REF8_MC_SWAP__SHIFT 0x10 +#define UVD_MP_SWAP_CNTL__MP_REF9_MC_SWAP__SHIFT 0x12 +#define UVD_MP_SWAP_CNTL__MP_REF10_MC_SWAP__SHIFT 0x14 +#define UVD_MP_SWAP_CNTL__MP_REF11_MC_SWAP__SHIFT 0x16 +#define UVD_MP_SWAP_CNTL__MP_REF12_MC_SWAP__SHIFT 0x18 +#define UVD_MP_SWAP_CNTL__MP_REF13_MC_SWAP__SHIFT 0x1a +#define UVD_MP_SWAP_CNTL__MP_REF14_MC_SWAP__SHIFT 0x1c +#define UVD_MP_SWAP_CNTL__MP_REF15_MC_SWAP__SHIFT 0x1e +#define UVD_MP_SWAP_CNTL__MP_REF0_MC_SWAP_MASK 0x00000003L +#define UVD_MP_SWAP_CNTL__MP_REF1_MC_SWAP_MASK 0x0000000CL +#define UVD_MP_SWAP_CNTL__MP_REF2_MC_SWAP_MASK 0x00000030L +#define UVD_MP_SWAP_CNTL__MP_REF3_MC_SWAP_MASK 0x000000C0L +#define UVD_MP_SWAP_CNTL__MP_REF4_MC_SWAP_MASK 0x00000300L +#define UVD_MP_SWAP_CNTL__MP_REF5_MC_SWAP_MASK 0x00000C00L +#define UVD_MP_SWAP_CNTL__MP_REF6_MC_SWAP_MASK 0x00003000L +#define UVD_MP_SWAP_CNTL__MP_REF7_MC_SWAP_MASK 0x0000C000L +#define UVD_MP_SWAP_CNTL__MP_REF8_MC_SWAP_MASK 0x00030000L +#define UVD_MP_SWAP_CNTL__MP_REF9_MC_SWAP_MASK 0x000C0000L +#define UVD_MP_SWAP_CNTL__MP_REF10_MC_SWAP_MASK 0x00300000L +#define UVD_MP_SWAP_CNTL__MP_REF11_MC_SWAP_MASK 0x00C00000L +#define UVD_MP_SWAP_CNTL__MP_REF12_MC_SWAP_MASK 0x03000000L +#define UVD_MP_SWAP_CNTL__MP_REF13_MC_SWAP_MASK 0x0C000000L +#define UVD_MP_SWAP_CNTL__MP_REF14_MC_SWAP_MASK 0x30000000L +#define UVD_MP_SWAP_CNTL__MP_REF15_MC_SWAP_MASK 0xC0000000L +//UVD_MP_SWAP_CNTL2 +#define UVD_MP_SWAP_CNTL2__MP_REF16_MC_SWAP__SHIFT 0x0 +#define UVD_MP_SWAP_CNTL2__MP_REF16_MC_SWAP_MASK 0x00000003L +//UVD_MPC_LUMA_SRCH +#define UVD_MPC_LUMA_SRCH__CNTR__SHIFT 0x0 +#define UVD_MPC_LUMA_SRCH__CNTR_MASK 0xFFFFFFFFL +//UVD_MPC_LUMA_HIT +#define UVD_MPC_LUMA_HIT__CNTR__SHIFT 0x0 +#define UVD_MPC_LUMA_HIT__CNTR_MASK 0xFFFFFFFFL +//UVD_MPC_LUMA_HITPEND +#define UVD_MPC_LUMA_HITPEND__CNTR__SHIFT 0x0 +#define UVD_MPC_LUMA_HITPEND__CNTR_MASK 0xFFFFFFFFL +//UVD_MPC_CHROMA_SRCH +#define UVD_MPC_CHROMA_SRCH__CNTR__SHIFT 0x0 +#define UVD_MPC_CHROMA_SRCH__CNTR_MASK 0xFFFFFFFFL +//UVD_MPC_CHROMA_HIT +#define UVD_MPC_CHROMA_HIT__CNTR__SHIFT 0x0 +#define UVD_MPC_CHROMA_HIT__CNTR_MASK 0xFFFFFFFFL +//UVD_MPC_CHROMA_HITPEND +#define UVD_MPC_CHROMA_HITPEND__CNTR__SHIFT 0x0 +#define UVD_MPC_CHROMA_HITPEND__CNTR_MASK 0xFFFFFFFFL +//UVD_MPC_CNTL +#define UVD_MPC_CNTL__BLK_RST__SHIFT 0x0 +#define UVD_MPC_CNTL__REG_MPC1_PERF_SELECT__SHIFT 0x1 +#define UVD_MPC_CNTL__REPLACEMENT_MODE__SHIFT 0x3 +#define UVD_MPC_CNTL__PERF_RST__SHIFT 0x6 +#define UVD_MPC_CNTL__REG_MPC_CNTL_BACKWARD_COMPATIBILITY__SHIFT 0x7 +#define UVD_MPC_CNTL__DBG_MUX__SHIFT 0x8 +#define UVD_MPC_CNTL__AVE_WEIGHT__SHIFT 0x10 +#define UVD_MPC_CNTL__URGENT_EN__SHIFT 0x12 +#define UVD_MPC_CNTL__SMPAT_REQ_SPEED_UP__SHIFT 0x13 +#define UVD_MPC_CNTL__TEST_MODE_EN__SHIFT 0x14 +#define UVD_MPC_CNTL__BLK_RST_MASK 0x00000001L +#define UVD_MPC_CNTL__REG_MPC1_PERF_SELECT_MASK 0x00000002L +#define UVD_MPC_CNTL__REPLACEMENT_MODE_MASK 0x00000038L +#define UVD_MPC_CNTL__PERF_RST_MASK 0x00000040L +#define UVD_MPC_CNTL__REG_MPC_CNTL_BACKWARD_COMPATIBILITY_MASK 0x00000080L +#define UVD_MPC_CNTL__DBG_MUX_MASK 0x00000F00L +#define UVD_MPC_CNTL__AVE_WEIGHT_MASK 0x00030000L +#define UVD_MPC_CNTL__URGENT_EN_MASK 0x00040000L +#define UVD_MPC_CNTL__SMPAT_REQ_SPEED_UP_MASK 0x00080000L +#define UVD_MPC_CNTL__TEST_MODE_EN_MASK 0x00300000L +//UVD_MPC_PITCH +#define UVD_MPC_PITCH__LUMA_PITCH__SHIFT 0x0 +#define UVD_MPC_PITCH__LUMA_PITCH_MASK 0x000007FFL +//UVD_MPC_SET_MUXA0 +#define UVD_MPC_SET_MUXA0__VARA_0__SHIFT 0x0 +#define UVD_MPC_SET_MUXA0__VARA_1__SHIFT 0x6 +#define UVD_MPC_SET_MUXA0__VARA_2__SHIFT 0xc +#define UVD_MPC_SET_MUXA0__VARA_3__SHIFT 0x12 +#define UVD_MPC_SET_MUXA0__VARA_4__SHIFT 0x18 +#define UVD_MPC_SET_MUXA0__VARA_0_MASK 0x0000003FL +#define UVD_MPC_SET_MUXA0__VARA_1_MASK 0x00000FC0L +#define UVD_MPC_SET_MUXA0__VARA_2_MASK 0x0003F000L +#define UVD_MPC_SET_MUXA0__VARA_3_MASK 0x00FC0000L +#define UVD_MPC_SET_MUXA0__VARA_4_MASK 0x3F000000L +//UVD_MPC_SET_MUXA1 +#define UVD_MPC_SET_MUXA1__VARA_5__SHIFT 0x0 +#define UVD_MPC_SET_MUXA1__VARA_6__SHIFT 0x6 +#define UVD_MPC_SET_MUXA1__VARA_7__SHIFT 0xc +#define UVD_MPC_SET_MUXA1__VARA_5_MASK 0x0000003FL +#define UVD_MPC_SET_MUXA1__VARA_6_MASK 0x00000FC0L +#define UVD_MPC_SET_MUXA1__VARA_7_MASK 0x0003F000L +//UVD_MPC_SET_MUXB0 +#define UVD_MPC_SET_MUXB0__VARB_0__SHIFT 0x0 +#define UVD_MPC_SET_MUXB0__VARB_1__SHIFT 0x6 +#define UVD_MPC_SET_MUXB0__VARB_2__SHIFT 0xc +#define UVD_MPC_SET_MUXB0__VARB_3__SHIFT 0x12 +#define UVD_MPC_SET_MUXB0__VARB_4__SHIFT 0x18 +#define UVD_MPC_SET_MUXB0__VARB_0_MASK 0x0000003FL +#define UVD_MPC_SET_MUXB0__VARB_1_MASK 0x00000FC0L +#define UVD_MPC_SET_MUXB0__VARB_2_MASK 0x0003F000L +#define UVD_MPC_SET_MUXB0__VARB_3_MASK 0x00FC0000L +#define UVD_MPC_SET_MUXB0__VARB_4_MASK 0x3F000000L +//UVD_MPC_SET_MUXB1 +#define UVD_MPC_SET_MUXB1__VARB_5__SHIFT 0x0 +#define UVD_MPC_SET_MUXB1__VARB_6__SHIFT 0x6 +#define UVD_MPC_SET_MUXB1__VARB_7__SHIFT 0xc +#define UVD_MPC_SET_MUXB1__VARB_5_MASK 0x0000003FL +#define UVD_MPC_SET_MUXB1__VARB_6_MASK 0x00000FC0L +#define UVD_MPC_SET_MUXB1__VARB_7_MASK 0x0003F000L +//UVD_MPC_SET_MUX +#define UVD_MPC_SET_MUX__SET_0__SHIFT 0x0 +#define UVD_MPC_SET_MUX__SET_1__SHIFT 0x3 +#define UVD_MPC_SET_MUX__SET_2__SHIFT 0x6 +#define UVD_MPC_SET_MUX__SET_0_MASK 0x00000007L +#define UVD_MPC_SET_MUX__SET_1_MASK 0x00000038L +#define UVD_MPC_SET_MUX__SET_2_MASK 0x000001C0L +//UVD_MPC_SET_ALU +#define UVD_MPC_SET_ALU__FUNCT__SHIFT 0x0 +#define UVD_MPC_SET_ALU__OPERAND__SHIFT 0x4 +#define UVD_MPC_SET_ALU__FUNCT_MASK 0x00000007L +#define UVD_MPC_SET_ALU__OPERAND_MASK 0x00000FF0L +//UVD_MPC_PERF0 +#define UVD_MPC_PERF0__MAX_LAT__SHIFT 0x0 +#define UVD_MPC_PERF0__MAX_LAT_MASK 0x000003FFL +//UVD_MPC_PERF1 +#define UVD_MPC_PERF1__AVE_LAT__SHIFT 0x0 +#define UVD_MPC_PERF1__AVE_LAT_MASK 0x000003FFL +//UVD_MPC_IND_INDEX +#define UVD_MPC_IND_INDEX__INDEX__SHIFT 0x0 +#define UVD_MPC_IND_INDEX__INDEX_MASK 0x000001FFL +//UVD_MPC_IND_DATA +#define UVD_MPC_IND_DATA__DATA__SHIFT 0x0 +#define UVD_MPC_IND_DATA__DATA_MASK 0xFFFFFFFFL + + +// addressBlock: aid_uvd0_uvd_rbcdec +//UVD_RBC_IB_SIZE +#define UVD_RBC_IB_SIZE__IB_SIZE__SHIFT 0x4 +#define UVD_RBC_IB_SIZE__IB_SIZE_MASK 0x007FFFF0L +//UVD_RBC_IB_SIZE_UPDATE +#define UVD_RBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT 0x4 +#define UVD_RBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK 0x007FFFF0L +//UVD_RBC_RB_CNTL +#define UVD_RBC_RB_CNTL__RB_BUFSZ__SHIFT 0x0 +#define UVD_RBC_RB_CNTL__RB_BLKSZ__SHIFT 0x8 +#define UVD_RBC_RB_CNTL__RB_NO_FETCH__SHIFT 0x10 +#define UVD_RBC_RB_CNTL__RB_WPTR_POLL_EN__SHIFT 0x14 +#define UVD_RBC_RB_CNTL__RB_NO_UPDATE__SHIFT 0x18 +#define UVD_RBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT 0x1c +#define UVD_RBC_RB_CNTL__BLK_RST__SHIFT 0x1d +#define UVD_RBC_RB_CNTL__RB_BUFSZ_MASK 0x0000001FL +#define UVD_RBC_RB_CNTL__RB_BLKSZ_MASK 0x00001F00L +#define UVD_RBC_RB_CNTL__RB_NO_FETCH_MASK 0x00010000L +#define UVD_RBC_RB_CNTL__RB_WPTR_POLL_EN_MASK 0x00100000L +#define UVD_RBC_RB_CNTL__RB_NO_UPDATE_MASK 0x01000000L +#define UVD_RBC_RB_CNTL__RB_RPTR_WR_EN_MASK 0x10000000L +#define UVD_RBC_RB_CNTL__BLK_RST_MASK 0x20000000L +//UVD_RBC_RB_RPTR_ADDR +#define UVD_RBC_RB_RPTR_ADDR__RB_RPTR_ADDR__SHIFT 0x0 +#define UVD_RBC_RB_RPTR_ADDR__RB_RPTR_ADDR_MASK 0xFFFFFFFFL +//UVD_RBC_VCPU_ACCESS +#define UVD_RBC_VCPU_ACCESS__ENABLE_RBC__SHIFT 0x0 +#define UVD_RBC_VCPU_ACCESS__ENABLE_RBC_MASK 0x00000001L +//UVD_FW_SEMAPHORE_CNTL +#define UVD_FW_SEMAPHORE_CNTL__START__SHIFT 0x0 +#define UVD_FW_SEMAPHORE_CNTL__BUSY__SHIFT 0x8 +#define UVD_FW_SEMAPHORE_CNTL__PASS__SHIFT 0x9 +#define UVD_FW_SEMAPHORE_CNTL__START_MASK 0x00000001L +#define UVD_FW_SEMAPHORE_CNTL__BUSY_MASK 0x00000100L +#define UVD_FW_SEMAPHORE_CNTL__PASS_MASK 0x00000200L +//UVD_RBC_READ_REQ_URGENT_CNTL +#define UVD_RBC_READ_REQ_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT 0x0 +#define UVD_RBC_READ_REQ_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK 0x00000003L +//UVD_RBC_RB_WPTR_CNTL +#define UVD_RBC_RB_WPTR_CNTL__RB_PRE_WRITE_TIMER__SHIFT 0x0 +#define UVD_RBC_RB_WPTR_CNTL__RB_PRE_WRITE_TIMER_MASK 0x00007FFFL +//UVD_RBC_WPTR_STATUS +#define UVD_RBC_WPTR_STATUS__RB_WPTR_IN_USE__SHIFT 0x4 +#define UVD_RBC_WPTR_STATUS__RB_WPTR_IN_USE_MASK 0x007FFFF0L +//UVD_RBC_WPTR_POLL_CNTL +#define UVD_RBC_WPTR_POLL_CNTL__POLL_FREQ__SHIFT 0x0 +#define UVD_RBC_WPTR_POLL_CNTL__IDLE_POLL_COUNT__SHIFT 0x10 +#define UVD_RBC_WPTR_POLL_CNTL__POLL_FREQ_MASK 0x0000FFFFL +#define UVD_RBC_WPTR_POLL_CNTL__IDLE_POLL_COUNT_MASK 0xFFFF0000L +//UVD_RBC_WPTR_POLL_ADDR +#define UVD_RBC_WPTR_POLL_ADDR__POLL_ADDR__SHIFT 0x2 +#define UVD_RBC_WPTR_POLL_ADDR__POLL_ADDR_MASK 0xFFFFFFFCL +//UVD_SEMA_CMD +#define UVD_SEMA_CMD__REQ_CMD__SHIFT 0x0 +#define UVD_SEMA_CMD__WR_PHASE__SHIFT 0x4 +#define UVD_SEMA_CMD__MODE__SHIFT 0x6 +#define UVD_SEMA_CMD__VMID_EN__SHIFT 0x7 +#define UVD_SEMA_CMD__VMID__SHIFT 0x8 +#define UVD_SEMA_CMD__REQ_CMD_MASK 0x0000000FL +#define UVD_SEMA_CMD__WR_PHASE_MASK 0x00000030L +#define UVD_SEMA_CMD__MODE_MASK 0x00000040L +#define UVD_SEMA_CMD__VMID_EN_MASK 0x00000080L +#define UVD_SEMA_CMD__VMID_MASK 0x00000F00L +//UVD_SEMA_ADDR_LOW +#define UVD_SEMA_ADDR_LOW__ADDR_26_3__SHIFT 0x0 +#define UVD_SEMA_ADDR_LOW__ADDR_26_3_MASK 0x00FFFFFFL +//UVD_SEMA_ADDR_HIGH +#define UVD_SEMA_ADDR_HIGH__ADDR_47_27__SHIFT 0x0 +#define UVD_SEMA_ADDR_HIGH__ADDR_47_27_MASK 0x001FFFFFL +//UVD_ENGINE_CNTL +#define UVD_ENGINE_CNTL__ENGINE_START__SHIFT 0x0 +#define UVD_ENGINE_CNTL__ENGINE_START_MODE__SHIFT 0x1 +#define UVD_ENGINE_CNTL__NJ_PF_HANDLE_DISABLE__SHIFT 0x2 +#define UVD_ENGINE_CNTL__ENGINE_START_MASK 0x00000001L +#define UVD_ENGINE_CNTL__ENGINE_START_MODE_MASK 0x00000002L +#define UVD_ENGINE_CNTL__NJ_PF_HANDLE_DISABLE_MASK 0x00000004L +//UVD_SEMA_TIMEOUT_STATUS +#define UVD_SEMA_TIMEOUT_STATUS__SEMAPHORE_WAIT_INCOMPLETE_TIMEOUT_STAT__SHIFT 0x0 +#define UVD_SEMA_TIMEOUT_STATUS__SEMAPHORE_WAIT_FAULT_TIMEOUT_STAT__SHIFT 0x1 +#define UVD_SEMA_TIMEOUT_STATUS__SEMAPHORE_SIGNAL_INCOMPLETE_TIMEOUT_STAT__SHIFT 0x2 +#define UVD_SEMA_TIMEOUT_STATUS__SEMAPHORE_TIMEOUT_CLEAR__SHIFT 0x3 +#define UVD_SEMA_TIMEOUT_STATUS__SEMAPHORE_WAIT_INCOMPLETE_TIMEOUT_STAT_MASK 0x00000001L +#define UVD_SEMA_TIMEOUT_STATUS__SEMAPHORE_WAIT_FAULT_TIMEOUT_STAT_MASK 0x00000002L +#define UVD_SEMA_TIMEOUT_STATUS__SEMAPHORE_SIGNAL_INCOMPLETE_TIMEOUT_STAT_MASK 0x00000004L +#define UVD_SEMA_TIMEOUT_STATUS__SEMAPHORE_TIMEOUT_CLEAR_MASK 0x00000008L +//UVD_SEMA_CNTL +#define UVD_SEMA_CNTL__SEMAPHORE_EN__SHIFT 0x0 +#define UVD_SEMA_CNTL__ADVANCED_MODE_DIS__SHIFT 0x1 +#define UVD_SEMA_CNTL__SEMAPHORE_EN_MASK 0x00000001L +#define UVD_SEMA_CNTL__ADVANCED_MODE_DIS_MASK 0x00000002L +//UVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL +#define UVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL__SIGNAL_INCOMPLETE_EN__SHIFT 0x0 +#define UVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL__SIGNAL_INCOMPLETE_COUNT__SHIFT 0x1 +#define UVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL__RESEND_TIMER__SHIFT 0x18 +#define UVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL__SIGNAL_INCOMPLETE_EN_MASK 0x00000001L +#define UVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL__SIGNAL_INCOMPLETE_COUNT_MASK 0x001FFFFEL +#define UVD_SEMA_SIGNAL_INCOMPLETE_TIMEOUT_CNTL__RESEND_TIMER_MASK 0x07000000L +//UVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL +#define UVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL__WAIT_FAULT_EN__SHIFT 0x0 +#define UVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL__WAIT_FAULT_COUNT__SHIFT 0x1 +#define UVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL__RESEND_TIMER__SHIFT 0x18 +#define UVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL__WAIT_FAULT_EN_MASK 0x00000001L +#define UVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL__WAIT_FAULT_COUNT_MASK 0x001FFFFEL +#define UVD_SEMA_WAIT_FAULT_TIMEOUT_CNTL__RESEND_TIMER_MASK 0x07000000L +//UVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL +#define UVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL__WAIT_INCOMPLETE_EN__SHIFT 0x0 +#define UVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL__WAIT_INCOMPLETE_COUNT__SHIFT 0x1 +#define UVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL__RESEND_TIMER__SHIFT 0x18 +#define UVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL__WAIT_INCOMPLETE_EN_MASK 0x00000001L +#define UVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL__WAIT_INCOMPLETE_COUNT_MASK 0x001FFFFEL +#define UVD_SEMA_WAIT_INCOMPLETE_TIMEOUT_CNTL__RESEND_TIMER_MASK 0x07000000L +//UVD_JOB_START +#define UVD_JOB_START__JOB_START__SHIFT 0x0 +#define UVD_JOB_START__JOB_START_MASK 0x00000001L +//UVD_RBC_BUF_STATUS +#define UVD_RBC_BUF_STATUS__RB_BUF_VALID__SHIFT 0x0 +#define UVD_RBC_BUF_STATUS__IB_BUF_VALID__SHIFT 0x8 +#define UVD_RBC_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT 0x10 +#define UVD_RBC_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT 0x13 +#define UVD_RBC_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT 0x16 +#define UVD_RBC_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT 0x19 +#define UVD_RBC_BUF_STATUS__RB_BUF_VALID_MASK 0x000000FFL +#define UVD_RBC_BUF_STATUS__IB_BUF_VALID_MASK 0x0000FF00L +#define UVD_RBC_BUF_STATUS__RB_BUF_RD_ADDR_MASK 0x00070000L +#define UVD_RBC_BUF_STATUS__IB_BUF_RD_ADDR_MASK 0x00380000L +#define UVD_RBC_BUF_STATUS__RB_BUF_WR_ADDR_MASK 0x01C00000L +#define UVD_RBC_BUF_STATUS__IB_BUF_WR_ADDR_MASK 0x0E000000L +//UVD_RBC_SWAP_CNTL +#define UVD_RBC_SWAP_CNTL__RB_MC_SWAP__SHIFT 0x0 +#define UVD_RBC_SWAP_CNTL__IB_MC_SWAP__SHIFT 0x2 +#define UVD_RBC_SWAP_CNTL__RB_RPTR_MC_SWAP__SHIFT 0x4 +#define UVD_RBC_SWAP_CNTL__RB_WR_MC_SWAP__SHIFT 0x1a +#define UVD_RBC_SWAP_CNTL__RB_MC_SWAP_MASK 0x00000003L +#define UVD_RBC_SWAP_CNTL__IB_MC_SWAP_MASK 0x0000000CL +#define UVD_RBC_SWAP_CNTL__RB_RPTR_MC_SWAP_MASK 0x00000030L +#define UVD_RBC_SWAP_CNTL__RB_WR_MC_SWAP_MASK 0x0C000000L + + +// addressBlock: aid_uvd0_lmi_adpdec +//UVD_LMI_RE_64BIT_BAR_LOW +#define UVD_LMI_RE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_RE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_RE_64BIT_BAR_HIGH +#define UVD_LMI_RE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_RE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_IT_64BIT_BAR_LOW +#define UVD_LMI_IT_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_IT_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_IT_64BIT_BAR_HIGH +#define UVD_LMI_IT_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_IT_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MP_64BIT_BAR_LOW +#define UVD_LMI_MP_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MP_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MP_64BIT_BAR_HIGH +#define UVD_LMI_MP_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MP_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_CM_64BIT_BAR_LOW +#define UVD_LMI_CM_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_CM_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_CM_64BIT_BAR_HIGH +#define UVD_LMI_CM_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_CM_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_DB_64BIT_BAR_LOW +#define UVD_LMI_DB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_DB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_DB_64BIT_BAR_HIGH +#define UVD_LMI_DB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_DB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_DBW_64BIT_BAR_LOW +#define UVD_LMI_DBW_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_DBW_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_DBW_64BIT_BAR_HIGH +#define UVD_LMI_DBW_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_DBW_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_IDCT_64BIT_BAR_LOW +#define UVD_LMI_IDCT_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_IDCT_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_IDCT_64BIT_BAR_HIGH +#define UVD_LMI_IDCT_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_IDCT_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MPRD_S0_64BIT_BAR_LOW +#define UVD_LMI_MPRD_S0_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MPRD_S0_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MPRD_S0_64BIT_BAR_HIGH +#define UVD_LMI_MPRD_S0_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MPRD_S0_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MPRD_S1_64BIT_BAR_LOW +#define UVD_LMI_MPRD_S1_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MPRD_S1_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MPRD_S1_64BIT_BAR_HIGH +#define UVD_LMI_MPRD_S1_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MPRD_S1_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MPRD_DBW_64BIT_BAR_LOW +#define UVD_LMI_MPRD_DBW_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MPRD_DBW_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MPRD_DBW_64BIT_BAR_HIGH +#define UVD_LMI_MPRD_DBW_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MPRD_DBW_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MPC_64BIT_BAR_LOW +#define UVD_LMI_MPC_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MPC_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MPC_64BIT_BAR_HIGH +#define UVD_LMI_MPC_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MPC_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_RBC_RB_64BIT_BAR_LOW +#define UVD_LMI_RBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_RBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_RBC_RB_64BIT_BAR_HIGH +#define UVD_LMI_RBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_RBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_RBC_IB_64BIT_BAR_LOW +#define UVD_LMI_RBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_RBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_RBC_IB_64BIT_BAR_HIGH +#define UVD_LMI_RBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_RBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_LBSI_64BIT_BAR_LOW +#define UVD_LMI_LBSI_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_LBSI_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_LBSI_64BIT_BAR_HIGH +#define UVD_LMI_LBSI_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_LBSI_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_NC0_64BIT_BAR_LOW +#define UVD_LMI_VCPU_NC0_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_VCPU_NC0_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_NC0_64BIT_BAR_HIGH +#define UVD_LMI_VCPU_NC0_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_VCPU_NC0_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_NC1_64BIT_BAR_LOW +#define UVD_LMI_VCPU_NC1_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_VCPU_NC1_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_NC1_64BIT_BAR_HIGH +#define UVD_LMI_VCPU_NC1_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_VCPU_NC1_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE_64BIT_BAR_LOW +#define UVD_LMI_VCPU_CACHE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH +#define UVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_CENC_64BIT_BAR_LOW +#define UVD_LMI_CENC_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_CENC_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_CENC_64BIT_BAR_HIGH +#define UVD_LMI_CENC_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_CENC_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_SRE_64BIT_BAR_LOW +#define UVD_LMI_SRE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_SRE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_SRE_64BIT_BAR_HIGH +#define UVD_LMI_SRE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_SRE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_GPGPU_64BIT_BAR_LOW +#define UVD_LMI_MIF_GPGPU_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_GPGPU_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_GPGPU_64BIT_BAR_HIGH +#define UVD_LMI_MIF_GPGPU_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_GPGPU_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_CURR_LUMA_64BIT_BAR_LOW +#define UVD_LMI_MIF_CURR_LUMA_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_CURR_LUMA_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_CURR_LUMA_64BIT_BAR_HIGH +#define UVD_LMI_MIF_CURR_LUMA_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_CURR_LUMA_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_LOW +#define UVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_HIGH +#define UVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_CURR_CHROMA_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_REF_64BIT_BAR_LOW +#define UVD_LMI_MIF_REF_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_REF_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_REF_64BIT_BAR_HIGH +#define UVD_LMI_MIF_REF_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_REF_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_DBW_64BIT_BAR_LOW +#define UVD_LMI_MIF_DBW_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_DBW_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_DBW_64BIT_BAR_HIGH +#define UVD_LMI_MIF_DBW_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_DBW_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_CM_COLOC_64BIT_BAR_LOW +#define UVD_LMI_MIF_CM_COLOC_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_CM_COLOC_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_CM_COLOC_64BIT_BAR_HIGH +#define UVD_LMI_MIF_CM_COLOC_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_CM_COLOC_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSP0_64BIT_BAR_LOW +#define UVD_LMI_MIF_BSP0_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_BSP0_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSP0_64BIT_BAR_HIGH +#define UVD_LMI_MIF_BSP0_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_BSP0_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSP1_64BIT_BAR_LOW +#define UVD_LMI_MIF_BSP1_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_BSP1_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSP1_64BIT_BAR_HIGH +#define UVD_LMI_MIF_BSP1_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_BSP1_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSP2_64BIT_BAR_LOW +#define UVD_LMI_MIF_BSP2_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_BSP2_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSP2_64BIT_BAR_HIGH +#define UVD_LMI_MIF_BSP2_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_BSP2_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSP3_64BIT_BAR_LOW +#define UVD_LMI_MIF_BSP3_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_BSP3_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSP3_64BIT_BAR_HIGH +#define UVD_LMI_MIF_BSP3_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_BSP3_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSD0_64BIT_BAR_LOW +#define UVD_LMI_MIF_BSD0_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_BSD0_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSD0_64BIT_BAR_HIGH +#define UVD_LMI_MIF_BSD0_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_BSD0_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSD1_64BIT_BAR_LOW +#define UVD_LMI_MIF_BSD1_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_BSD1_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSD1_64BIT_BAR_HIGH +#define UVD_LMI_MIF_BSD1_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_BSD1_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSD2_64BIT_BAR_LOW +#define UVD_LMI_MIF_BSD2_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_BSD2_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSD2_64BIT_BAR_HIGH +#define UVD_LMI_MIF_BSD2_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_BSD2_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSD3_64BIT_BAR_LOW +#define UVD_LMI_MIF_BSD3_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_BSD3_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSD3_64BIT_BAR_HIGH +#define UVD_LMI_MIF_BSD3_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_BSD3_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSD4_64BIT_BAR_LOW +#define UVD_LMI_MIF_BSD4_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_BSD4_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_BSD4_64BIT_BAR_HIGH +#define UVD_LMI_MIF_BSD4_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_BSD4_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW +#define UVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH +#define UVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE8_64BIT_BAR_LOW +#define UVD_LMI_VCPU_CACHE8_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE8_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE8_64BIT_BAR_HIGH +#define UVD_LMI_VCPU_CACHE8_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE8_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW +#define UVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH +#define UVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE3_64BIT_BAR_LOW +#define UVD_LMI_VCPU_CACHE3_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE3_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE3_64BIT_BAR_HIGH +#define UVD_LMI_VCPU_CACHE3_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE3_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE4_64BIT_BAR_LOW +#define UVD_LMI_VCPU_CACHE4_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE4_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE4_64BIT_BAR_HIGH +#define UVD_LMI_VCPU_CACHE4_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE4_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE5_64BIT_BAR_LOW +#define UVD_LMI_VCPU_CACHE5_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE5_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE5_64BIT_BAR_HIGH +#define UVD_LMI_VCPU_CACHE5_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE5_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE6_64BIT_BAR_LOW +#define UVD_LMI_VCPU_CACHE6_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE6_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE6_64BIT_BAR_HIGH +#define UVD_LMI_VCPU_CACHE6_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE6_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE7_64BIT_BAR_LOW +#define UVD_LMI_VCPU_CACHE7_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE7_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_VCPU_CACHE7_64BIT_BAR_HIGH +#define UVD_LMI_VCPU_CACHE7_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE7_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_SCLR_64BIT_BAR_LOW +#define UVD_LMI_MIF_SCLR_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_SCLR_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_SCLR_64BIT_BAR_HIGH +#define UVD_LMI_MIF_SCLR_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_SCLR_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_SCLR2_64BIT_BAR_LOW +#define UVD_LMI_MIF_SCLR2_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_SCLR2_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_SCLR2_64BIT_BAR_HIGH +#define UVD_LMI_MIF_SCLR2_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_SCLR2_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_SPH_64BIT_BAR_HIGH +#define UVD_LMI_SPH_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_SPH_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_LOW +#define UVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_HIGH +#define UVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_IMAGEPASTE_LUMA_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_LOW +#define UVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_HIGH +#define UVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_IMAGEPASTE_CHROMA_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_LOW +#define UVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_HIGH +#define UVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_PRIVACY_LUMA_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_LOW +#define UVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_HIGH +#define UVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_PRIVACY_CHROMA_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_ADP_ATOMIC_CONFIG +#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_USER0_WR_CACHE__SHIFT 0x0 +#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_USER1_WR_CACHE__SHIFT 0x4 +#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_USER2_WR_CACHE__SHIFT 0x8 +#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_USER3_WR_CACHE__SHIFT 0xc +#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_RD_URG__SHIFT 0x10 +#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_USER0_WR_CACHE_MASK 0x0000000FL +#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_USER1_WR_CACHE_MASK 0x000000F0L +#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_USER2_WR_CACHE_MASK 0x00000F00L +#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_USER3_WR_CACHE_MASK 0x0000F000L +#define UVD_ADP_ATOMIC_CONFIG__ATOMIC_RD_URG_MASK 0x000F0000L +//UVD_LMI_ARB_CTRL2 +#define UVD_LMI_ARB_CTRL2__CENC_RD_WAIT_EN__SHIFT 0x0 +#define UVD_LMI_ARB_CTRL2__ATOMIC_WR_WAIT_EN__SHIFT 0x1 +#define UVD_LMI_ARB_CTRL2__CENC_RD_MAX_BURST__SHIFT 0x2 +#define UVD_LMI_ARB_CTRL2__ATOMIC_WR_MAX_BURST__SHIFT 0x6 +#define UVD_LMI_ARB_CTRL2__MIF_RD_REQ_RET_MAX__SHIFT 0xa +#define UVD_LMI_ARB_CTRL2__MIF_WR_REQ_RET_MAX__SHIFT 0x14 +#define UVD_LMI_ARB_CTRL2__CENC_RD_WAIT_EN_MASK 0x00000001L +#define UVD_LMI_ARB_CTRL2__ATOMIC_WR_WAIT_EN_MASK 0x00000002L +#define UVD_LMI_ARB_CTRL2__CENC_RD_MAX_BURST_MASK 0x0000003CL +#define UVD_LMI_ARB_CTRL2__ATOMIC_WR_MAX_BURST_MASK 0x000003C0L +#define UVD_LMI_ARB_CTRL2__MIF_RD_REQ_RET_MAX_MASK 0x000FFC00L +#define UVD_LMI_ARB_CTRL2__MIF_WR_REQ_RET_MAX_MASK 0xFFF00000L +//UVD_LMI_VCPU_CACHE_VMIDS_MULTI +#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE1_VMID__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE2_VMID__SHIFT 0x4 +#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE3_VMID__SHIFT 0x8 +#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE4_VMID__SHIFT 0xc +#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE5_VMID__SHIFT 0x10 +#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE6_VMID__SHIFT 0x14 +#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE7_VMID__SHIFT 0x18 +#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE8_VMID__SHIFT 0x1c +#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE1_VMID_MASK 0x0000000FL +#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE2_VMID_MASK 0x000000F0L +#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE3_VMID_MASK 0x00000F00L +#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE4_VMID_MASK 0x0000F000L +#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE5_VMID_MASK 0x000F0000L +#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE6_VMID_MASK 0x00F00000L +#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE7_VMID_MASK 0x0F000000L +#define UVD_LMI_VCPU_CACHE_VMIDS_MULTI__VCPU_CACHE8_VMID_MASK 0xF0000000L +//UVD_LMI_VCPU_NC_VMIDS_MULTI +#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC2_VMID__SHIFT 0x4 +#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC3_VMID__SHIFT 0x8 +#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC4_VMID__SHIFT 0xc +#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC5_VMID__SHIFT 0x10 +#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC6_VMID__SHIFT 0x14 +#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC7_VMID__SHIFT 0x18 +#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC2_VMID_MASK 0x000000F0L +#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC3_VMID_MASK 0x00000F00L +#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC4_VMID_MASK 0x0000F000L +#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC5_VMID_MASK 0x000F0000L +#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC6_VMID_MASK 0x00F00000L +#define UVD_LMI_VCPU_NC_VMIDS_MULTI__VCPU_NC7_VMID_MASK 0x0F000000L +//UVD_LMI_LAT_CTRL +#define UVD_LMI_LAT_CTRL__SCALE__SHIFT 0x0 +#define UVD_LMI_LAT_CTRL__MAX_START__SHIFT 0x8 +#define UVD_LMI_LAT_CTRL__MIN_START__SHIFT 0x9 +#define UVD_LMI_LAT_CTRL__AVG_START__SHIFT 0xa +#define UVD_LMI_LAT_CTRL__PERFMON_SYNC__SHIFT 0xb +#define UVD_LMI_LAT_CTRL__SKIP__SHIFT 0x10 +#define UVD_LMI_LAT_CTRL__SCALE_MASK 0x000000FFL +#define UVD_LMI_LAT_CTRL__MAX_START_MASK 0x00000100L +#define UVD_LMI_LAT_CTRL__MIN_START_MASK 0x00000200L +#define UVD_LMI_LAT_CTRL__AVG_START_MASK 0x00000400L +#define UVD_LMI_LAT_CTRL__PERFMON_SYNC_MASK 0x00000800L +#define UVD_LMI_LAT_CTRL__SKIP_MASK 0x000F0000L +//UVD_LMI_LAT_CNTR +#define UVD_LMI_LAT_CNTR__MAX_LAT__SHIFT 0x0 +#define UVD_LMI_LAT_CNTR__MIN_LAT__SHIFT 0x8 +#define UVD_LMI_LAT_CNTR__MAX_LAT_MASK 0x000000FFL +#define UVD_LMI_LAT_CNTR__MIN_LAT_MASK 0x0000FF00L +//UVD_LMI_AVG_LAT_CNTR +#define UVD_LMI_AVG_LAT_CNTR__ENV_LOW__SHIFT 0x0 +#define UVD_LMI_AVG_LAT_CNTR__ENV_HIGH__SHIFT 0x8 +#define UVD_LMI_AVG_LAT_CNTR__ENV_HIT__SHIFT 0x10 +#define UVD_LMI_AVG_LAT_CNTR__ENV_LOW_MASK 0x000000FFL +#define UVD_LMI_AVG_LAT_CNTR__ENV_HIGH_MASK 0x0000FF00L +#define UVD_LMI_AVG_LAT_CNTR__ENV_HIT_MASK 0xFFFF0000L +//UVD_LMI_SPH +#define UVD_LMI_SPH__ADDR__SHIFT 0x0 +#define UVD_LMI_SPH__STS__SHIFT 0x1c +#define UVD_LMI_SPH__STS_VALID__SHIFT 0x1e +#define UVD_LMI_SPH__STS_OVERFLOW__SHIFT 0x1f +#define UVD_LMI_SPH__ADDR_MASK 0x0FFFFFFFL +#define UVD_LMI_SPH__STS_MASK 0x30000000L +#define UVD_LMI_SPH__STS_VALID_MASK 0x40000000L +#define UVD_LMI_SPH__STS_OVERFLOW_MASK 0x80000000L +//UVD_LMI_VCPU_CACHE_VMID +#define UVD_LMI_VCPU_CACHE_VMID__VCPU_CACHE_VMID__SHIFT 0x0 +#define UVD_LMI_VCPU_CACHE_VMID__VCPU_CACHE_VMID_MASK 0x0000000FL +//UVD_LMI_CTRL2 +#define UVD_LMI_CTRL2__SPH_DIS__SHIFT 0x0 +#define UVD_LMI_CTRL2__STALL_ARB__SHIFT 0x1 +#define UVD_LMI_CTRL2__ASSERT_UMC_URGENT__SHIFT 0x2 +#define UVD_LMI_CTRL2__MASK_UMC_URGENT__SHIFT 0x3 +#define UVD_LMI_CTRL2__CRC1_RESET__SHIFT 0x4 +#define UVD_LMI_CTRL2__DRCITF_BUBBLE_FIX_DIS__SHIFT 0x7 +#define UVD_LMI_CTRL2__STALL_ARB_UMC__SHIFT 0x8 +#define UVD_LMI_CTRL2__MC_READ_ID_SEL__SHIFT 0x9 +#define UVD_LMI_CTRL2__MC_WRITE_ID_SEL__SHIFT 0xb +#define UVD_LMI_CTRL2__VCPU_NC0_EXT_EN__SHIFT 0xd +#define UVD_LMI_CTRL2__VCPU_NC1_EXT_EN__SHIFT 0xe +#define UVD_LMI_CTRL2__SPU_EXTRA_CID_EN__SHIFT 0xf +#define UVD_LMI_CTRL2__RE_OFFLOAD_EN__SHIFT 0x10 +#define UVD_LMI_CTRL2__RE_OFLD_MIF_WR_REQ_NUM__SHIFT 0x11 +#define UVD_LMI_CTRL2__CLEAR_NJ_PF_BP__SHIFT 0x19 +#define UVD_LMI_CTRL2__NJ_MIF_GATING__SHIFT 0x1a +#define UVD_LMI_CTRL2__CRC1_SEL__SHIFT 0x1b +#define UVD_LMI_CTRL2__SPH_DIS_MASK 0x00000001L +#define UVD_LMI_CTRL2__STALL_ARB_MASK 0x00000002L +#define UVD_LMI_CTRL2__ASSERT_UMC_URGENT_MASK 0x00000004L +#define UVD_LMI_CTRL2__MASK_UMC_URGENT_MASK 0x00000008L +#define UVD_LMI_CTRL2__CRC1_RESET_MASK 0x00000010L +#define UVD_LMI_CTRL2__DRCITF_BUBBLE_FIX_DIS_MASK 0x00000080L +#define UVD_LMI_CTRL2__STALL_ARB_UMC_MASK 0x00000100L +#define UVD_LMI_CTRL2__MC_READ_ID_SEL_MASK 0x00000600L +#define UVD_LMI_CTRL2__MC_WRITE_ID_SEL_MASK 0x00001800L +#define UVD_LMI_CTRL2__VCPU_NC0_EXT_EN_MASK 0x00002000L +#define UVD_LMI_CTRL2__VCPU_NC1_EXT_EN_MASK 0x00004000L +#define UVD_LMI_CTRL2__SPU_EXTRA_CID_EN_MASK 0x00008000L +#define UVD_LMI_CTRL2__RE_OFFLOAD_EN_MASK 0x00010000L +#define UVD_LMI_CTRL2__RE_OFLD_MIF_WR_REQ_NUM_MASK 0x01FE0000L +#define UVD_LMI_CTRL2__CLEAR_NJ_PF_BP_MASK 0x02000000L +#define UVD_LMI_CTRL2__NJ_MIF_GATING_MASK 0x04000000L +#define UVD_LMI_CTRL2__CRC1_SEL_MASK 0xF8000000L +//UVD_LMI_URGENT_CTRL +#define UVD_LMI_URGENT_CTRL__ENABLE_MC_RD_URGENT_STALL__SHIFT 0x0 +#define UVD_LMI_URGENT_CTRL__ASSERT_MC_RD_STALL__SHIFT 0x1 +#define UVD_LMI_URGENT_CTRL__ASSERT_MC_RD_URGENT__SHIFT 0x2 +#define UVD_LMI_URGENT_CTRL__ENABLE_MC_WR_URGENT_STALL__SHIFT 0x8 +#define UVD_LMI_URGENT_CTRL__ASSERT_MC_WR_STALL__SHIFT 0x9 +#define UVD_LMI_URGENT_CTRL__ASSERT_MC_WR_URGENT__SHIFT 0xa +#define UVD_LMI_URGENT_CTRL__ENABLE_UMC_RD_URGENT_STALL__SHIFT 0x10 +#define UVD_LMI_URGENT_CTRL__ASSERT_UMC_RD_STALL__SHIFT 0x11 +#define UVD_LMI_URGENT_CTRL__ASSERT_UMC_RD_URGENT__SHIFT 0x12 +#define UVD_LMI_URGENT_CTRL__ENABLE_UMC_WR_URGENT_STALL__SHIFT 0x18 +#define UVD_LMI_URGENT_CTRL__ASSERT_UMC_WR_STALL__SHIFT 0x19 +#define UVD_LMI_URGENT_CTRL__ASSERT_UMC_WR_URGENT__SHIFT 0x1a +#define UVD_LMI_URGENT_CTRL__ENABLE_MC_RD_URGENT_STALL_MASK 0x00000001L +#define UVD_LMI_URGENT_CTRL__ASSERT_MC_RD_STALL_MASK 0x00000002L +#define UVD_LMI_URGENT_CTRL__ASSERT_MC_RD_URGENT_MASK 0x0000003CL +#define UVD_LMI_URGENT_CTRL__ENABLE_MC_WR_URGENT_STALL_MASK 0x00000100L +#define UVD_LMI_URGENT_CTRL__ASSERT_MC_WR_STALL_MASK 0x00000200L +#define UVD_LMI_URGENT_CTRL__ASSERT_MC_WR_URGENT_MASK 0x00003C00L +#define UVD_LMI_URGENT_CTRL__ENABLE_UMC_RD_URGENT_STALL_MASK 0x00010000L +#define UVD_LMI_URGENT_CTRL__ASSERT_UMC_RD_STALL_MASK 0x00020000L +#define UVD_LMI_URGENT_CTRL__ASSERT_UMC_RD_URGENT_MASK 0x003C0000L +#define UVD_LMI_URGENT_CTRL__ENABLE_UMC_WR_URGENT_STALL_MASK 0x01000000L +#define UVD_LMI_URGENT_CTRL__ASSERT_UMC_WR_STALL_MASK 0x02000000L +#define UVD_LMI_URGENT_CTRL__ASSERT_UMC_WR_URGENT_MASK 0x3C000000L +//UVD_LMI_CTRL +#define UVD_LMI_CTRL__WRITE_CLEAN_TIMER__SHIFT 0x0 +#define UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN__SHIFT 0x8 +#define UVD_LMI_CTRL__REQ_MODE__SHIFT 0x9 +#define UVD_LMI_CTRL__ASSERT_MC_URGENT__SHIFT 0xb +#define UVD_LMI_CTRL__MASK_MC_URGENT__SHIFT 0xc +#define UVD_LMI_CTRL__DATA_COHERENCY_EN__SHIFT 0xd +#define UVD_LMI_CTRL__CRC_RESET__SHIFT 0xe +#define UVD_LMI_CTRL__CRC_SEL__SHIFT 0xf +#define UVD_LMI_CTRL__DISABLE_ON_FWV_FAIL__SHIFT 0x14 +#define UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN__SHIFT 0x15 +#define UVD_LMI_CTRL__CM_DATA_COHERENCY_EN__SHIFT 0x16 +#define UVD_LMI_CTRL__DB_DB_DATA_COHERENCY_EN__SHIFT 0x17 +#define UVD_LMI_CTRL__DB_IT_DATA_COHERENCY_EN__SHIFT 0x18 +#define UVD_LMI_CTRL__IT_IT_DATA_COHERENCY_EN__SHIFT 0x19 +#define UVD_LMI_CTRL__MIF_MIF_DATA_COHERENCY_EN__SHIFT 0x1a +#define UVD_LMI_CTRL__MIF_LESS_OUTSTANDING_RD_REQ__SHIFT 0x1b +#define UVD_LMI_CTRL__MC_BLK_RST__SHIFT 0x1c +#define UVD_LMI_CTRL__UMC_BLK_RST__SHIFT 0x1d +#define UVD_LMI_CTRL__RFU__SHIFT 0x1e +#define UVD_LMI_CTRL__WRITE_CLEAN_TIMER_MASK 0x000000FFL +#define UVD_LMI_CTRL__WRITE_CLEAN_TIMER_EN_MASK 0x00000100L +#define UVD_LMI_CTRL__REQ_MODE_MASK 0x00000200L +#define UVD_LMI_CTRL__ASSERT_MC_URGENT_MASK 0x00000800L +#define UVD_LMI_CTRL__MASK_MC_URGENT_MASK 0x00001000L +#define UVD_LMI_CTRL__DATA_COHERENCY_EN_MASK 0x00002000L +#define UVD_LMI_CTRL__CRC_RESET_MASK 0x00004000L +#define UVD_LMI_CTRL__CRC_SEL_MASK 0x000F8000L +#define UVD_LMI_CTRL__DISABLE_ON_FWV_FAIL_MASK 0x00100000L +#define UVD_LMI_CTRL__VCPU_DATA_COHERENCY_EN_MASK 0x00200000L +#define UVD_LMI_CTRL__CM_DATA_COHERENCY_EN_MASK 0x00400000L +#define UVD_LMI_CTRL__DB_DB_DATA_COHERENCY_EN_MASK 0x00800000L +#define UVD_LMI_CTRL__DB_IT_DATA_COHERENCY_EN_MASK 0x01000000L +#define UVD_LMI_CTRL__IT_IT_DATA_COHERENCY_EN_MASK 0x02000000L +#define UVD_LMI_CTRL__MIF_MIF_DATA_COHERENCY_EN_MASK 0x04000000L +#define UVD_LMI_CTRL__MIF_LESS_OUTSTANDING_RD_REQ_MASK 0x08000000L +#define UVD_LMI_CTRL__MC_BLK_RST_MASK 0x10000000L +#define UVD_LMI_CTRL__UMC_BLK_RST_MASK 0x20000000L +#define UVD_LMI_CTRL__RFU_MASK 0xC0000000L +//UVD_LMI_STATUS +#define UVD_LMI_STATUS__READ_CLEAN__SHIFT 0x0 +#define UVD_LMI_STATUS__WRITE_CLEAN__SHIFT 0x1 +#define UVD_LMI_STATUS__WRITE_CLEAN_RAW__SHIFT 0x2 +#define UVD_LMI_STATUS__VCPU_LMI_WRITE_CLEAN__SHIFT 0x3 +#define UVD_LMI_STATUS__UMC_READ_CLEAN__SHIFT 0x4 +#define UVD_LMI_STATUS__UMC_WRITE_CLEAN__SHIFT 0x5 +#define UVD_LMI_STATUS__UMC_WRITE_CLEAN_RAW__SHIFT 0x6 +#define UVD_LMI_STATUS__PENDING_UVD_MC_WRITE__SHIFT 0x7 +#define UVD_LMI_STATUS__READ_CLEAN_RAW__SHIFT 0x8 +#define UVD_LMI_STATUS__UMC_READ_CLEAN_RAW__SHIFT 0x9 +#define UVD_LMI_STATUS__UMC_UVD_IDLE__SHIFT 0xa +#define UVD_LMI_STATUS__UMC_AVP_IDLE__SHIFT 0xb +#define UVD_LMI_STATUS__ADP_MC_READ_CLEAN__SHIFT 0xc +#define UVD_LMI_STATUS__ADP_UMC_READ_CLEAN__SHIFT 0xd +#define UVD_LMI_STATUS__BSP0_WRITE_CLEAN__SHIFT 0x12 +#define UVD_LMI_STATUS__BSP1_WRITE_CLEAN__SHIFT 0x13 +#define UVD_LMI_STATUS__BSP2_WRITE_CLEAN__SHIFT 0x14 +#define UVD_LMI_STATUS__BSP3_WRITE_CLEAN__SHIFT 0x15 +#define UVD_LMI_STATUS__CENC_READ_CLEAN__SHIFT 0x16 +#define UVD_LMI_STATUS__READ_CLEAN_MASK 0x00000001L +#define UVD_LMI_STATUS__WRITE_CLEAN_MASK 0x00000002L +#define UVD_LMI_STATUS__WRITE_CLEAN_RAW_MASK 0x00000004L +#define UVD_LMI_STATUS__VCPU_LMI_WRITE_CLEAN_MASK 0x00000008L +#define UVD_LMI_STATUS__UMC_READ_CLEAN_MASK 0x00000010L +#define UVD_LMI_STATUS__UMC_WRITE_CLEAN_MASK 0x00000020L +#define UVD_LMI_STATUS__UMC_WRITE_CLEAN_RAW_MASK 0x00000040L +#define UVD_LMI_STATUS__PENDING_UVD_MC_WRITE_MASK 0x00000080L +#define UVD_LMI_STATUS__READ_CLEAN_RAW_MASK 0x00000100L +#define UVD_LMI_STATUS__UMC_READ_CLEAN_RAW_MASK 0x00000200L +#define UVD_LMI_STATUS__UMC_UVD_IDLE_MASK 0x00000400L +#define UVD_LMI_STATUS__UMC_AVP_IDLE_MASK 0x00000800L +#define UVD_LMI_STATUS__ADP_MC_READ_CLEAN_MASK 0x00001000L +#define UVD_LMI_STATUS__ADP_UMC_READ_CLEAN_MASK 0x00002000L +#define UVD_LMI_STATUS__BSP0_WRITE_CLEAN_MASK 0x00040000L +#define UVD_LMI_STATUS__BSP1_WRITE_CLEAN_MASK 0x00080000L +#define UVD_LMI_STATUS__BSP2_WRITE_CLEAN_MASK 0x00100000L +#define UVD_LMI_STATUS__BSP3_WRITE_CLEAN_MASK 0x00200000L +#define UVD_LMI_STATUS__CENC_READ_CLEAN_MASK 0x00400000L +//UVD_LMI_PERFMON_CTRL +#define UVD_LMI_PERFMON_CTRL__PERFMON_STATE__SHIFT 0x0 +#define UVD_LMI_PERFMON_CTRL__PERFMON_SEL__SHIFT 0x8 +#define UVD_LMI_PERFMON_CTRL__PERFMON_STATE_MASK 0x00000003L +#define UVD_LMI_PERFMON_CTRL__PERFMON_SEL_MASK 0x00001F00L +//UVD_LMI_PERFMON_COUNT_LO +#define UVD_LMI_PERFMON_COUNT_LO__PERFMON_COUNT__SHIFT 0x0 +#define UVD_LMI_PERFMON_COUNT_LO__PERFMON_COUNT_MASK 0xFFFFFFFFL +//UVD_LMI_PERFMON_COUNT_HI +#define UVD_LMI_PERFMON_COUNT_HI__PERFMON_COUNT__SHIFT 0x0 +#define UVD_LMI_PERFMON_COUNT_HI__PERFMON_COUNT_MASK 0x0000FFFFL +//UVD_LMI_ADP_SWAP_CNTL +#define UVD_LMI_ADP_SWAP_CNTL__VCPU_R_MC_SWAP__SHIFT 0x6 +#define UVD_LMI_ADP_SWAP_CNTL__VCPU_W_MC_SWAP__SHIFT 0x8 +#define UVD_LMI_ADP_SWAP_CNTL__CM_MC_SWAP__SHIFT 0xa +#define UVD_LMI_ADP_SWAP_CNTL__IT_MC_SWAP__SHIFT 0xc +#define UVD_LMI_ADP_SWAP_CNTL__DB_R_MC_SWAP__SHIFT 0xe +#define UVD_LMI_ADP_SWAP_CNTL__DB_W_MC_SWAP__SHIFT 0x10 +#define UVD_LMI_ADP_SWAP_CNTL__CSM_MC_SWAP__SHIFT 0x12 +#define UVD_LMI_ADP_SWAP_CNTL__PREF_MC_SWAP__SHIFT 0x14 +#define UVD_LMI_ADP_SWAP_CNTL__DBW_MC_SWAP__SHIFT 0x18 +#define UVD_LMI_ADP_SWAP_CNTL__RE_MC_SWAP__SHIFT 0x1c +#define UVD_LMI_ADP_SWAP_CNTL__MP_MC_SWAP__SHIFT 0x1e +#define UVD_LMI_ADP_SWAP_CNTL__VCPU_R_MC_SWAP_MASK 0x000000C0L +#define UVD_LMI_ADP_SWAP_CNTL__VCPU_W_MC_SWAP_MASK 0x00000300L +#define UVD_LMI_ADP_SWAP_CNTL__CM_MC_SWAP_MASK 0x00000C00L +#define UVD_LMI_ADP_SWAP_CNTL__IT_MC_SWAP_MASK 0x00003000L +#define UVD_LMI_ADP_SWAP_CNTL__DB_R_MC_SWAP_MASK 0x0000C000L +#define UVD_LMI_ADP_SWAP_CNTL__DB_W_MC_SWAP_MASK 0x00030000L +#define UVD_LMI_ADP_SWAP_CNTL__CSM_MC_SWAP_MASK 0x000C0000L +#define UVD_LMI_ADP_SWAP_CNTL__PREF_MC_SWAP_MASK 0x00300000L +#define UVD_LMI_ADP_SWAP_CNTL__DBW_MC_SWAP_MASK 0x03000000L +#define UVD_LMI_ADP_SWAP_CNTL__RE_MC_SWAP_MASK 0x30000000L +#define UVD_LMI_ADP_SWAP_CNTL__MP_MC_SWAP_MASK 0xC0000000L +//UVD_LMI_RBC_RB_VMID +#define UVD_LMI_RBC_RB_VMID__RB_VMID__SHIFT 0x0 +#define UVD_LMI_RBC_RB_VMID__RB_VMID_MASK 0x0000000FL +//UVD_LMI_RBC_IB_VMID +#define UVD_LMI_RBC_IB_VMID__IB_VMID__SHIFT 0x0 +#define UVD_LMI_RBC_IB_VMID__IB_VMID_MASK 0x0000000FL +//UVD_LMI_MC_CREDITS +#define UVD_LMI_MC_CREDITS__UVD_RD_CREDITS__SHIFT 0x0 +#define UVD_LMI_MC_CREDITS__UVD_WR_CREDITS__SHIFT 0x8 +#define UVD_LMI_MC_CREDITS__UMC_RD_CREDITS__SHIFT 0x10 +#define UVD_LMI_MC_CREDITS__UMC_WR_CREDITS__SHIFT 0x18 +#define UVD_LMI_MC_CREDITS__UVD_RD_CREDITS_MASK 0x0000003FL +#define UVD_LMI_MC_CREDITS__UVD_WR_CREDITS_MASK 0x00003F00L +#define UVD_LMI_MC_CREDITS__UMC_RD_CREDITS_MASK 0x003F0000L +#define UVD_LMI_MC_CREDITS__UMC_WR_CREDITS_MASK 0x3F000000L +//UVD_LMI_ADP_IND_INDEX +#define UVD_LMI_ADP_IND_INDEX__INDEX__SHIFT 0x0 +#define UVD_LMI_ADP_IND_INDEX__INDEX_MASK 0x00001FFFL +//UVD_LMI_ADP_IND_DATA +#define UVD_LMI_ADP_IND_DATA__DATA__SHIFT 0x0 +#define UVD_LMI_ADP_IND_DATA__DATA_MASK 0xFFFFFFFFL +//UVD_LMI_ADP_PF_EN +#define UVD_LMI_ADP_PF_EN__VCPU_CACHE0_PF_EN__SHIFT 0x0 +#define UVD_LMI_ADP_PF_EN__VCPU_CACHE1_PF_EN__SHIFT 0x1 +#define UVD_LMI_ADP_PF_EN__VCPU_CACHE2_PF_EN__SHIFT 0x2 +#define UVD_LMI_ADP_PF_EN__VCPU_CACHE0_PF_EN_MASK 0x00000001L +#define UVD_LMI_ADP_PF_EN__VCPU_CACHE1_PF_EN_MASK 0x00000002L +#define UVD_LMI_ADP_PF_EN__VCPU_CACHE2_PF_EN_MASK 0x00000004L +//UVD_LMI_PREF_CTRL +#define UVD_LMI_PREF_CTRL__PREF_RST__SHIFT 0x0 +#define UVD_LMI_PREF_CTRL__PREF_BUSY_STATUS__SHIFT 0x1 +#define UVD_LMI_PREF_CTRL__PREF_WSTRB__SHIFT 0x2 +#define UVD_LMI_PREF_CTRL__PREF_WRITE_SIZE__SHIFT 0x3 +#define UVD_LMI_PREF_CTRL__PREF_STEP_SIZE__SHIFT 0x4 +#define UVD_LMI_PREF_CTRL__PREF_SIZE__SHIFT 0x13 +#define UVD_LMI_PREF_CTRL__PREF_RST_MASK 0x00000001L +#define UVD_LMI_PREF_CTRL__PREF_BUSY_STATUS_MASK 0x00000002L +#define UVD_LMI_PREF_CTRL__PREF_WSTRB_MASK 0x00000004L +#define UVD_LMI_PREF_CTRL__PREF_WRITE_SIZE_MASK 0x00000008L +#define UVD_LMI_PREF_CTRL__PREF_STEP_SIZE_MASK 0x00000070L +#define UVD_LMI_PREF_CTRL__PREF_SIZE_MASK 0xFFF80000L +//UVD_LMI_MIF_REF_LUMA_64BIT_BAR_LOW +#define UVD_LMI_MIF_REF_LUMA_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MIF_REF_LUMA_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MIF_REF_LUMA_64BIT_BAR_HIGH +#define UVD_LMI_MIF_REF_LUMA_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MIF_REF_LUMA_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//VCN_RAS_CNTL +#define VCN_RAS_CNTL__VCPU_VCODEC_IH_EN__SHIFT 0x0 +#define VCN_RAS_CNTL__VCPU_VCODEC_PMI_EN__SHIFT 0x4 +#define VCN_RAS_CNTL__VCPU_VCODEC_REARM__SHIFT 0x8 +#define VCN_RAS_CNTL__VCPU_VCODEC_STALL_EN__SHIFT 0xc +#define VCN_RAS_CNTL__VCPU_VCODEC_READY__SHIFT 0x10 +#define VCN_RAS_CNTL__VCPU_VCODEC_IH_EN_MASK 0x00000001L +#define VCN_RAS_CNTL__VCPU_VCODEC_PMI_EN_MASK 0x00000010L +#define VCN_RAS_CNTL__VCPU_VCODEC_REARM_MASK 0x00000100L +#define VCN_RAS_CNTL__VCPU_VCODEC_STALL_EN_MASK 0x00001000L +#define VCN_RAS_CNTL__VCPU_VCODEC_READY_MASK 0x00010000L + + +// addressBlock: aid_uvd0_uvd_jpeg0_jpegnpdec +//UVD_JPEG_CNTL +#define UVD_JPEG_CNTL__REQUEST_EN__SHIFT 0x1 +#define UVD_JPEG_CNTL__ERR_RST_EN__SHIFT 0x2 +#define UVD_JPEG_CNTL__DBG_MUX_SEL__SHIFT 0x8 +#define UVD_JPEG_CNTL__FORMAT_CONV_EN__SHIFT 0x10 +#define UVD_JPEG_CNTL__VUP_MODE__SHIFT 0x11 +#define UVD_JPEG_CNTL__FC_TIMEOUT_EN__SHIFT 0x12 +#define UVD_JPEG_CNTL__ROI_CROP_EN__SHIFT 0x18 +#define UVD_JPEG_CNTL__ROI_CROP_EARLY_DECODE_STOP_DIS__SHIFT 0x19 +#define UVD_JPEG_CNTL__REQUEST_EN_MASK 0x00000002L +#define UVD_JPEG_CNTL__ERR_RST_EN_MASK 0x00000004L +#define UVD_JPEG_CNTL__DBG_MUX_SEL_MASK 0x00007F00L +#define UVD_JPEG_CNTL__FORMAT_CONV_EN_MASK 0x00010000L +#define UVD_JPEG_CNTL__VUP_MODE_MASK 0x00020000L +#define UVD_JPEG_CNTL__FC_TIMEOUT_EN_MASK 0x00040000L +#define UVD_JPEG_CNTL__ROI_CROP_EN_MASK 0x01000000L +#define UVD_JPEG_CNTL__ROI_CROP_EARLY_DECODE_STOP_DIS_MASK 0x02000000L +//UVD_JPEG_RB_BASE +#define UVD_JPEG_RB_BASE__RB_BYTE_OFF__SHIFT 0x0 +#define UVD_JPEG_RB_BASE__RB_BASE__SHIFT 0x6 +#define UVD_JPEG_RB_BASE__RB_BYTE_OFF_MASK 0x0000003FL +#define UVD_JPEG_RB_BASE__RB_BASE_MASK 0xFFFFFFC0L +//UVD_JPEG_RB_WPTR +#define UVD_JPEG_RB_WPTR__RB_WPTR__SHIFT 0x4 +#define UVD_JPEG_RB_WPTR__RB_WPTR_MASK 0x3FFFFFF0L +//UVD_JPEG_RB_RPTR +#define UVD_JPEG_RB_RPTR__RB_RPTR__SHIFT 0x4 +#define UVD_JPEG_RB_RPTR__RB_RPTR_MASK 0x3FFFFFF0L +//UVD_JPEG_RB_SIZE +#define UVD_JPEG_RB_SIZE__RB_SIZE__SHIFT 0x4 +#define UVD_JPEG_RB_SIZE__RB_SIZE_MASK 0x3FFFFFF0L +//UVD_JPEG_DEC_CNT +#define UVD_JPEG_DEC_CNT__DECODE_COUNT__SHIFT 0x0 +#define UVD_JPEG_DEC_CNT__DECODE_COUNT_MASK 0xFFFFFFFFL +//UVD_JPEG_SPS_INFO +#define UVD_JPEG_SPS_INFO__PIC_WIDTH__SHIFT 0x0 +#define UVD_JPEG_SPS_INFO__PIC_HEIGHT__SHIFT 0x10 +#define UVD_JPEG_SPS_INFO__PIC_WIDTH_MASK 0x0000FFFFL +#define UVD_JPEG_SPS_INFO__PIC_HEIGHT_MASK 0xFFFF0000L +//UVD_JPEG_SPS1_INFO +#define UVD_JPEG_SPS1_INFO__CHROMA_FORMAT_IDC__SHIFT 0x0 +#define UVD_JPEG_SPS1_INFO__YUV422_SUBFORMAT__SHIFT 0x3 +#define UVD_JPEG_SPS1_INFO__OUT_FMT_422__SHIFT 0x4 +#define UVD_JPEG_SPS1_INFO__CHROMA_FORMAT_IDC_MASK 0x00000007L +#define UVD_JPEG_SPS1_INFO__YUV422_SUBFORMAT_MASK 0x00000008L +#define UVD_JPEG_SPS1_INFO__OUT_FMT_422_MASK 0x00000010L +//UVD_JPEG_RE_TIMER +#define UVD_JPEG_RE_TIMER__TIMER_OUT__SHIFT 0x0 +#define UVD_JPEG_RE_TIMER__TIMER_OUT_EN__SHIFT 0x10 +#define UVD_JPEG_RE_TIMER__TIMER_OUT_MASK 0x000000FFL +#define UVD_JPEG_RE_TIMER__TIMER_OUT_EN_MASK 0x00010000L +//UVD_JPEG_DEC_SCRATCH0 +#define UVD_JPEG_DEC_SCRATCH0__SCRATCH0__SHIFT 0x0 +#define UVD_JPEG_DEC_SCRATCH0__SCRATCH0_MASK 0xFFFFFFFFL +//UVD_JPEG_INT_EN +#define UVD_JPEG_INT_EN__OUTBUF_WPTR_INC_EN__SHIFT 0x0 +#define UVD_JPEG_INT_EN__JOB_AVAIL_EN__SHIFT 0x1 +#define UVD_JPEG_INT_EN__FENCE_VAL_EN__SHIFT 0x2 +#define UVD_JPEG_INT_EN__FIFO_OVERFLOW_ERR_EN__SHIFT 0x6 +#define UVD_JPEG_INT_EN__BLK_CNT_OUT_OF_SYNC_ERR_EN__SHIFT 0x7 +#define UVD_JPEG_INT_EN__EOI_ERR_EN__SHIFT 0x8 +#define UVD_JPEG_INT_EN__HFM_ERR_EN__SHIFT 0x9 +#define UVD_JPEG_INT_EN__RST_ERR_EN__SHIFT 0xa +#define UVD_JPEG_INT_EN__ECS_MK_ERR_EN__SHIFT 0xb +#define UVD_JPEG_INT_EN__TIMEOUT_ERR_EN__SHIFT 0xc +#define UVD_JPEG_INT_EN__MARKER_ERR_EN__SHIFT 0xd +#define UVD_JPEG_INT_EN__FMT_ERR_EN__SHIFT 0xe +#define UVD_JPEG_INT_EN__PROFILE_ERR_EN__SHIFT 0xf +#define UVD_JPEG_INT_EN__FC_TIMEOUT_ERR_EN__SHIFT 0x10 +#define UVD_JPEG_INT_EN__FC_FMT_ERR_EN__SHIFT 0x11 +#define UVD_JPEG_INT_EN__FC_SRC_ERR_EN__SHIFT 0x12 +#define UVD_JPEG_INT_EN__CROP_SIZE_ERR_EN__SHIFT 0x13 +#define UVD_JPEG_INT_EN__OUTBUF_WPTR_INC_EN_MASK 0x00000001L +#define UVD_JPEG_INT_EN__JOB_AVAIL_EN_MASK 0x00000002L +#define UVD_JPEG_INT_EN__FENCE_VAL_EN_MASK 0x00000004L +#define UVD_JPEG_INT_EN__FIFO_OVERFLOW_ERR_EN_MASK 0x00000040L +#define UVD_JPEG_INT_EN__BLK_CNT_OUT_OF_SYNC_ERR_EN_MASK 0x00000080L +#define UVD_JPEG_INT_EN__EOI_ERR_EN_MASK 0x00000100L +#define UVD_JPEG_INT_EN__HFM_ERR_EN_MASK 0x00000200L +#define UVD_JPEG_INT_EN__RST_ERR_EN_MASK 0x00000400L +#define UVD_JPEG_INT_EN__ECS_MK_ERR_EN_MASK 0x00000800L +#define UVD_JPEG_INT_EN__TIMEOUT_ERR_EN_MASK 0x00001000L +#define UVD_JPEG_INT_EN__MARKER_ERR_EN_MASK 0x00002000L +#define UVD_JPEG_INT_EN__FMT_ERR_EN_MASK 0x00004000L +#define UVD_JPEG_INT_EN__PROFILE_ERR_EN_MASK 0x00008000L +#define UVD_JPEG_INT_EN__FC_TIMEOUT_ERR_EN_MASK 0x00010000L +#define UVD_JPEG_INT_EN__FC_FMT_ERR_EN_MASK 0x00020000L +#define UVD_JPEG_INT_EN__FC_SRC_ERR_EN_MASK 0x00040000L +#define UVD_JPEG_INT_EN__CROP_SIZE_ERR_EN_MASK 0x00080000L +//UVD_JPEG_INT_STAT +#define UVD_JPEG_INT_STAT__OUTBUF_WPTR_INC_INT__SHIFT 0x0 +#define UVD_JPEG_INT_STAT__JOB_AVAIL_INT__SHIFT 0x1 +#define UVD_JPEG_INT_STAT__FENCE_VAL_INT__SHIFT 0x2 +#define UVD_JPEG_INT_STAT__FIFO_OVERFLOW_ERR_INT__SHIFT 0x6 +#define UVD_JPEG_INT_STAT__BLK_CNT_OUT_OF_SYNC_ERR_INT__SHIFT 0x7 +#define UVD_JPEG_INT_STAT__EOI_ERR_INT__SHIFT 0x8 +#define UVD_JPEG_INT_STAT__HFM_ERR_INT__SHIFT 0x9 +#define UVD_JPEG_INT_STAT__RST_ERR_INT__SHIFT 0xa +#define UVD_JPEG_INT_STAT__ECS_MK_ERR_INT__SHIFT 0xb +#define UVD_JPEG_INT_STAT__TIMEOUT_ERR_INT__SHIFT 0xc +#define UVD_JPEG_INT_STAT__MARKER_ERR_INT__SHIFT 0xd +#define UVD_JPEG_INT_STAT__FMT_ERR_INT__SHIFT 0xe +#define UVD_JPEG_INT_STAT__PROFILE_ERR_INT__SHIFT 0xf +#define UVD_JPEG_INT_STAT__FC_TIMEOUT_ERR_INT__SHIFT 0x10 +#define UVD_JPEG_INT_STAT__FC_FMT_ERR_INT__SHIFT 0x11 +#define UVD_JPEG_INT_STAT__FC_SRC_ERR_INT__SHIFT 0x12 +#define UVD_JPEG_INT_STAT__CROP_SIZE_ERR_INT__SHIFT 0x13 +#define UVD_JPEG_INT_STAT__OUTBUF_WPTR_INC_INT_MASK 0x00000001L +#define UVD_JPEG_INT_STAT__JOB_AVAIL_INT_MASK 0x00000002L +#define UVD_JPEG_INT_STAT__FENCE_VAL_INT_MASK 0x00000004L +#define UVD_JPEG_INT_STAT__FIFO_OVERFLOW_ERR_INT_MASK 0x00000040L +#define UVD_JPEG_INT_STAT__BLK_CNT_OUT_OF_SYNC_ERR_INT_MASK 0x00000080L +#define UVD_JPEG_INT_STAT__EOI_ERR_INT_MASK 0x00000100L +#define UVD_JPEG_INT_STAT__HFM_ERR_INT_MASK 0x00000200L +#define UVD_JPEG_INT_STAT__RST_ERR_INT_MASK 0x00000400L +#define UVD_JPEG_INT_STAT__ECS_MK_ERR_INT_MASK 0x00000800L +#define UVD_JPEG_INT_STAT__TIMEOUT_ERR_INT_MASK 0x00001000L +#define UVD_JPEG_INT_STAT__MARKER_ERR_INT_MASK 0x00002000L +#define UVD_JPEG_INT_STAT__FMT_ERR_INT_MASK 0x00004000L +#define UVD_JPEG_INT_STAT__PROFILE_ERR_INT_MASK 0x00008000L +#define UVD_JPEG_INT_STAT__FC_TIMEOUT_ERR_INT_MASK 0x00010000L +#define UVD_JPEG_INT_STAT__FC_FMT_ERR_INT_MASK 0x00020000L +#define UVD_JPEG_INT_STAT__FC_SRC_ERR_INT_MASK 0x00040000L +#define UVD_JPEG_INT_STAT__CROP_SIZE_ERR_INT_MASK 0x00080000L +//UVD_JPEG_TIER_CNTL0 +#define UVD_JPEG_TIER_CNTL0__TIER_SEL__SHIFT 0x0 +#define UVD_JPEG_TIER_CNTL0__Y_COMP_ID__SHIFT 0x2 +#define UVD_JPEG_TIER_CNTL0__U_COMP_ID__SHIFT 0x4 +#define UVD_JPEG_TIER_CNTL0__V_COMP_ID__SHIFT 0x6 +#define UVD_JPEG_TIER_CNTL0__Y_H_SAMP_FAC__SHIFT 0x8 +#define UVD_JPEG_TIER_CNTL0__Y_V_SAMP_FAC__SHIFT 0xb +#define UVD_JPEG_TIER_CNTL0__U_H_SAMP_FAC__SHIFT 0xe +#define UVD_JPEG_TIER_CNTL0__U_V_SAMP_FAC__SHIFT 0x11 +#define UVD_JPEG_TIER_CNTL0__V_H_SAMP_FAC__SHIFT 0x14 +#define UVD_JPEG_TIER_CNTL0__V_V_SAMP_FAC__SHIFT 0x17 +#define UVD_JPEG_TIER_CNTL0__Y_TQ__SHIFT 0x1a +#define UVD_JPEG_TIER_CNTL0__U_TQ__SHIFT 0x1c +#define UVD_JPEG_TIER_CNTL0__V_TQ__SHIFT 0x1e +#define UVD_JPEG_TIER_CNTL0__TIER_SEL_MASK 0x00000003L +#define UVD_JPEG_TIER_CNTL0__Y_COMP_ID_MASK 0x0000000CL +#define UVD_JPEG_TIER_CNTL0__U_COMP_ID_MASK 0x00000030L +#define UVD_JPEG_TIER_CNTL0__V_COMP_ID_MASK 0x000000C0L +#define UVD_JPEG_TIER_CNTL0__Y_H_SAMP_FAC_MASK 0x00000700L +#define UVD_JPEG_TIER_CNTL0__Y_V_SAMP_FAC_MASK 0x00003800L +#define UVD_JPEG_TIER_CNTL0__U_H_SAMP_FAC_MASK 0x0001C000L +#define UVD_JPEG_TIER_CNTL0__U_V_SAMP_FAC_MASK 0x000E0000L +#define UVD_JPEG_TIER_CNTL0__V_H_SAMP_FAC_MASK 0x00700000L +#define UVD_JPEG_TIER_CNTL0__V_V_SAMP_FAC_MASK 0x03800000L +#define UVD_JPEG_TIER_CNTL0__Y_TQ_MASK 0x0C000000L +#define UVD_JPEG_TIER_CNTL0__U_TQ_MASK 0x30000000L +#define UVD_JPEG_TIER_CNTL0__V_TQ_MASK 0xC0000000L +//UVD_JPEG_TIER_CNTL1 +#define UVD_JPEG_TIER_CNTL1__SRC_WIDTH__SHIFT 0x0 +#define UVD_JPEG_TIER_CNTL1__SRC_HEIGHT__SHIFT 0x10 +#define UVD_JPEG_TIER_CNTL1__SRC_WIDTH_MASK 0x0000FFFFL +#define UVD_JPEG_TIER_CNTL1__SRC_HEIGHT_MASK 0xFFFF0000L +//UVD_JPEG_TIER_CNTL2 +#define UVD_JPEG_TIER_CNTL2__TBL_ECS_SEL__SHIFT 0x0 +#define UVD_JPEG_TIER_CNTL2__TBL_TYPE__SHIFT 0x1 +#define UVD_JPEG_TIER_CNTL2__TQ__SHIFT 0x2 +#define UVD_JPEG_TIER_CNTL2__TH__SHIFT 0x4 +#define UVD_JPEG_TIER_CNTL2__TC__SHIFT 0x6 +#define UVD_JPEG_TIER_CNTL2__TD__SHIFT 0x7 +#define UVD_JPEG_TIER_CNTL2__TA__SHIFT 0xa +#define UVD_JPEG_TIER_CNTL2__TIER2_HTBL_CNTLEN__SHIFT 0xe +#define UVD_JPEG_TIER_CNTL2__DRI_VAL__SHIFT 0x10 +#define UVD_JPEG_TIER_CNTL2__TBL_ECS_SEL_MASK 0x00000001L +#define UVD_JPEG_TIER_CNTL2__TBL_TYPE_MASK 0x00000002L +#define UVD_JPEG_TIER_CNTL2__TQ_MASK 0x0000000CL +#define UVD_JPEG_TIER_CNTL2__TH_MASK 0x00000030L +#define UVD_JPEG_TIER_CNTL2__TC_MASK 0x00000040L +#define UVD_JPEG_TIER_CNTL2__TD_MASK 0x00000380L +#define UVD_JPEG_TIER_CNTL2__TA_MASK 0x00001C00L +#define UVD_JPEG_TIER_CNTL2__TIER2_HTBL_CNTLEN_MASK 0x00004000L +#define UVD_JPEG_TIER_CNTL2__DRI_VAL_MASK 0xFFFF0000L +//UVD_JPEG_TIER_STATUS +#define UVD_JPEG_TIER_STATUS__BSI_FETCH_DONE__SHIFT 0x0 +#define UVD_JPEG_TIER_STATUS__DECODE_DONE__SHIFT 0x1 +#define UVD_JPEG_TIER_STATUS__BSI_FETCH_DONE_MASK 0x00000001L +#define UVD_JPEG_TIER_STATUS__DECODE_DONE_MASK 0x00000002L + + +// addressBlock: aid_uvd0_uvd_jpeg_sclk0_jpegnpsclkdec +//UVD_JPEG_OUTBUF_CNTL +#define UVD_JPEG_OUTBUF_CNTL__OUTBUF_CNT__SHIFT 0x0 +#define UVD_JPEG_OUTBUF_CNTL__HGT_ALIGN__SHIFT 0x2 +#define UVD_JPEG_OUTBUF_CNTL__JPEG0_DECODE_DONE_FIX__SHIFT 0x6 +#define UVD_JPEG_OUTBUF_CNTL__JPEG0_WR_COMB_MAX_CNT__SHIFT 0x7 +#define UVD_JPEG_OUTBUF_CNTL__JPEG0_WR_COMB_TIMER__SHIFT 0x9 +#define UVD_JPEG_OUTBUF_CNTL__DIS_OBUF_AVAIL_CHECK__SHIFT 0x10 +#define UVD_JPEG_OUTBUF_CNTL__OUTBUF_CNT_MASK 0x00000003L +#define UVD_JPEG_OUTBUF_CNTL__HGT_ALIGN_MASK 0x00000004L +#define UVD_JPEG_OUTBUF_CNTL__JPEG0_DECODE_DONE_FIX_MASK 0x00000040L +#define UVD_JPEG_OUTBUF_CNTL__JPEG0_WR_COMB_MAX_CNT_MASK 0x00000180L +#define UVD_JPEG_OUTBUF_CNTL__JPEG0_WR_COMB_TIMER_MASK 0x00001E00L +#define UVD_JPEG_OUTBUF_CNTL__DIS_OBUF_AVAIL_CHECK_MASK 0x00010000L +//UVD_JPEG_OUTBUF_WPTR +#define UVD_JPEG_OUTBUF_WPTR__OUTBUF_WPTR__SHIFT 0x0 +#define UVD_JPEG_OUTBUF_WPTR__OUTBUF_WPTR_MASK 0xFFFFFFFFL +//UVD_JPEG_OUTBUF_RPTR +#define UVD_JPEG_OUTBUF_RPTR__OUTBUF_RPTR__SHIFT 0x0 +#define UVD_JPEG_OUTBUF_RPTR__OUTBUF_RPTR_MASK 0xFFFFFFFFL +//UVD_JPEG_PITCH +#define UVD_JPEG_PITCH__PITCH__SHIFT 0x0 +#define UVD_JPEG_PITCH__PITCH_MASK 0xFFFFFFFFL +//UVD_JPEG_UV_PITCH +#define UVD_JPEG_UV_PITCH__UV_PITCH__SHIFT 0x0 +#define UVD_JPEG_UV_PITCH__UV_PITCH_MASK 0xFFFFFFFFL +//JPEG_DEC_Y_GFX8_TILING_SURFACE +#define JPEG_DEC_Y_GFX8_TILING_SURFACE__BANK_WIDTH__SHIFT 0x0 +#define JPEG_DEC_Y_GFX8_TILING_SURFACE__BANK_HEIGHT__SHIFT 0x2 +#define JPEG_DEC_Y_GFX8_TILING_SURFACE__MACRO_TILE_ASPECT__SHIFT 0x4 +#define JPEG_DEC_Y_GFX8_TILING_SURFACE__NUM_BANKS__SHIFT 0x6 +#define JPEG_DEC_Y_GFX8_TILING_SURFACE__PIPE_CONFIG__SHIFT 0x8 +#define JPEG_DEC_Y_GFX8_TILING_SURFACE__TILE_SPLIT__SHIFT 0xd +#define JPEG_DEC_Y_GFX8_TILING_SURFACE__ARRAY_MODE__SHIFT 0x10 +#define JPEG_DEC_Y_GFX8_TILING_SURFACE__BANK_WIDTH_MASK 0x00000003L +#define JPEG_DEC_Y_GFX8_TILING_SURFACE__BANK_HEIGHT_MASK 0x0000000CL +#define JPEG_DEC_Y_GFX8_TILING_SURFACE__MACRO_TILE_ASPECT_MASK 0x00000030L +#define JPEG_DEC_Y_GFX8_TILING_SURFACE__NUM_BANKS_MASK 0x000000C0L +#define JPEG_DEC_Y_GFX8_TILING_SURFACE__PIPE_CONFIG_MASK 0x00001F00L +#define JPEG_DEC_Y_GFX8_TILING_SURFACE__TILE_SPLIT_MASK 0x0000E000L +#define JPEG_DEC_Y_GFX8_TILING_SURFACE__ARRAY_MODE_MASK 0x000F0000L +//JPEG_DEC_UV_GFX8_TILING_SURFACE +#define JPEG_DEC_UV_GFX8_TILING_SURFACE__BANK_WIDTH__SHIFT 0x0 +#define JPEG_DEC_UV_GFX8_TILING_SURFACE__BANK_HEIGHT__SHIFT 0x2 +#define JPEG_DEC_UV_GFX8_TILING_SURFACE__MACRO_TILE_ASPECT__SHIFT 0x4 +#define JPEG_DEC_UV_GFX8_TILING_SURFACE__NUM_BANKS__SHIFT 0x6 +#define JPEG_DEC_UV_GFX8_TILING_SURFACE__PIPE_CONFIG__SHIFT 0x8 +#define JPEG_DEC_UV_GFX8_TILING_SURFACE__TILE_SPLIT__SHIFT 0xd +#define JPEG_DEC_UV_GFX8_TILING_SURFACE__ARRAY_MODE__SHIFT 0x10 +#define JPEG_DEC_UV_GFX8_TILING_SURFACE__BANK_WIDTH_MASK 0x00000003L +#define JPEG_DEC_UV_GFX8_TILING_SURFACE__BANK_HEIGHT_MASK 0x0000000CL +#define JPEG_DEC_UV_GFX8_TILING_SURFACE__MACRO_TILE_ASPECT_MASK 0x00000030L +#define JPEG_DEC_UV_GFX8_TILING_SURFACE__NUM_BANKS_MASK 0x000000C0L +#define JPEG_DEC_UV_GFX8_TILING_SURFACE__PIPE_CONFIG_MASK 0x00001F00L +#define JPEG_DEC_UV_GFX8_TILING_SURFACE__TILE_SPLIT_MASK 0x0000E000L +#define JPEG_DEC_UV_GFX8_TILING_SURFACE__ARRAY_MODE_MASK 0x000F0000L +//JPEG_DEC_GFX8_ADDR_CONFIG +#define JPEG_DEC_GFX8_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE__SHIFT 0x4 +#define JPEG_DEC_GFX8_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE_MASK 0x00000070L +//JPEG_DEC_Y_GFX10_TILING_SURFACE +#define JPEG_DEC_Y_GFX10_TILING_SURFACE__SWIZZLE_MODE__SHIFT 0x0 +#define JPEG_DEC_Y_GFX10_TILING_SURFACE__SWIZZLE_MODE_MASK 0x0000001FL +//JPEG_DEC_UV_GFX10_TILING_SURFACE +#define JPEG_DEC_UV_GFX10_TILING_SURFACE__SWIZZLE_MODE__SHIFT 0x0 +#define JPEG_DEC_UV_GFX10_TILING_SURFACE__SWIZZLE_MODE_MASK 0x0000001FL +//JPEG_DEC_GFX10_ADDR_CONFIG +#define JPEG_DEC_GFX10_ADDR_CONFIG__NUM_PIPES__SHIFT 0x0 +#define JPEG_DEC_GFX10_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE__SHIFT 0x3 +#define JPEG_DEC_GFX10_ADDR_CONFIG__NUM_PKRS__SHIFT 0x8 +#define JPEG_DEC_GFX10_ADDR_CONFIG__NUM_BANKS__SHIFT 0xc +#define JPEG_DEC_GFX10_ADDR_CONFIG__NUM_SHADER_ENGINES__SHIFT 0x13 +#define JPEG_DEC_GFX10_ADDR_CONFIG__NUM_PIPES_MASK 0x00000007L +#define JPEG_DEC_GFX10_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE_MASK 0x00000038L +#define JPEG_DEC_GFX10_ADDR_CONFIG__NUM_PKRS_MASK 0x00000700L +#define JPEG_DEC_GFX10_ADDR_CONFIG__NUM_BANKS_MASK 0x00007000L +#define JPEG_DEC_GFX10_ADDR_CONFIG__NUM_SHADER_ENGINES_MASK 0x00180000L +//JPEG_DEC_ADDR_MODE +#define JPEG_DEC_ADDR_MODE__ADDR_MODE_Y__SHIFT 0x0 +#define JPEG_DEC_ADDR_MODE__ADDR_MODE_UV__SHIFT 0x2 +#define JPEG_DEC_ADDR_MODE__ADDR_LIB_SEL__SHIFT 0xc +#define JPEG_DEC_ADDR_MODE__ADDR_MODE_Y_MASK 0x00000003L +#define JPEG_DEC_ADDR_MODE__ADDR_MODE_UV_MASK 0x0000000CL +#define JPEG_DEC_ADDR_MODE__ADDR_LIB_SEL_MASK 0x00007000L +//UVD_JPEG_OUTPUT_XY +#define UVD_JPEG_OUTPUT_XY__OUTPUT_X__SHIFT 0x0 +#define UVD_JPEG_OUTPUT_XY__OUTPUT_Y__SHIFT 0x10 +#define UVD_JPEG_OUTPUT_XY__OUTPUT_X_MASK 0x00003FFFL +#define UVD_JPEG_OUTPUT_XY__OUTPUT_Y_MASK 0x3FFF0000L +//UVD_JPEG_GPCOM_CMD +#define UVD_JPEG_GPCOM_CMD__CMD__SHIFT 0x1 +#define UVD_JPEG_GPCOM_CMD__CMD_MASK 0x0000000EL +//UVD_JPEG_GPCOM_DATA0 +#define UVD_JPEG_GPCOM_DATA0__DATA0__SHIFT 0x0 +#define UVD_JPEG_GPCOM_DATA0__DATA0_MASK 0xFFFFFFFFL +//UVD_JPEG_GPCOM_DATA1 +#define UVD_JPEG_GPCOM_DATA1__DATA1__SHIFT 0x0 +#define UVD_JPEG_GPCOM_DATA1__DATA1_MASK 0xFFFFFFFFL +//UVD_JPEG_SCRATCH1 +#define UVD_JPEG_SCRATCH1__SCRATCH1__SHIFT 0x0 +#define UVD_JPEG_SCRATCH1__SCRATCH1_MASK 0xFFFFFFFFL +//UVD_JPEG_DEC_SOFT_RST +#define UVD_JPEG_DEC_SOFT_RST__SOFT_RESET__SHIFT 0x0 +#define UVD_JPEG_DEC_SOFT_RST__RESET_STATUS__SHIFT 0x10 +#define UVD_JPEG_DEC_SOFT_RST__SOFT_RESET_MASK 0x00000001L +#define UVD_JPEG_DEC_SOFT_RST__RESET_STATUS_MASK 0x00010000L + +// addressBlock: aid_uvd0_vcn_edcc_dec +//VCN_UE_ERR_STATUS_LO_VIDD +#define VCN_UE_ERR_STATUS_LO_VIDD__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_VIDD__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_VIDD__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_VIDD__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_VIDD__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_VIDD__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_VIDD__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_VIDD__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_VIDD +#define VCN_UE_ERR_STATUS_HI_VIDD__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_VIDD__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_VIDD__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_VIDD__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_VIDD__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_VIDD__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_VIDD__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_VIDD__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_VIDD__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_VIDD__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_VIDD__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_VIDD__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_VIDD__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_VIDD__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_VIDD__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_VIDD__Err_clr_MASK 0x80000000L +//VCN_UE_ERR_STATUS_LO_VIDV +#define VCN_UE_ERR_STATUS_LO_VIDV__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_VIDV__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_VIDV__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_VIDV__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_VIDV__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_VIDV__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_VIDV__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_VIDV__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_VIDV +#define VCN_UE_ERR_STATUS_HI_VIDV__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_VIDV__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_VIDV__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_VIDV__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_VIDV__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_VIDV__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_VIDV__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_VIDV__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_VIDV__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_VIDV__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_VIDV__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_VIDV__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_VIDV__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_VIDV__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_VIDV__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_VIDV__Err_clr_MASK 0x80000000L +//VCN_CE_ERR_STATUS_LO_MMSCHD +#define VCN_CE_ERR_STATUS_LO_MMSCHD__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_CE_ERR_STATUS_LO_MMSCHD__Address_Valid_Flag__SHIFT 0x1 +#define VCN_CE_ERR_STATUS_LO_MMSCHD__Address__SHIFT 0x2 +#define VCN_CE_ERR_STATUS_LO_MMSCHD__Memory_id__SHIFT 0x18 +#define VCN_CE_ERR_STATUS_LO_MMSCHD__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_CE_ERR_STATUS_LO_MMSCHD__Address_Valid_Flag_MASK 0x00000002L +#define VCN_CE_ERR_STATUS_LO_MMSCHD__Address_MASK 0x00FFFFFCL +#define VCN_CE_ERR_STATUS_LO_MMSCHD__Memory_id_MASK 0xFF000000L +//VCN_CE_ERR_STATUS_HI_MMSCHD +#define VCN_CE_ERR_STATUS_HI_MMSCHD__ECC__SHIFT 0x0 +#define VCN_CE_ERR_STATUS_HI_MMSCHD__Other__SHIFT 0x1 +#define VCN_CE_ERR_STATUS_HI_MMSCHD__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_CE_ERR_STATUS_HI_MMSCHD__Err_Info__SHIFT 0x3 +#define VCN_CE_ERR_STATUS_HI_MMSCHD__CE_Cnt__SHIFT 0x17 +#define VCN_CE_ERR_STATUS_HI_MMSCHD__Poison__SHIFT 0x1c +#define VCN_CE_ERR_STATUS_HI_MMSCHD__RESERVED__SHIFT 0x1d +#define VCN_CE_ERR_STATUS_HI_MMSCHD__Err_clr__SHIFT 0x1f +#define VCN_CE_ERR_STATUS_HI_MMSCHD__ECC_MASK 0x00000001L +#define VCN_CE_ERR_STATUS_HI_MMSCHD__Other_MASK 0x00000002L +#define VCN_CE_ERR_STATUS_HI_MMSCHD__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_CE_ERR_STATUS_HI_MMSCHD__Err_Info_MASK 0x007FFFF8L +#define VCN_CE_ERR_STATUS_HI_MMSCHD__CE_Cnt_MASK 0x03800000L +#define VCN_CE_ERR_STATUS_HI_MMSCHD__Poison_MASK 0x10000000L +#define VCN_CE_ERR_STATUS_HI_MMSCHD__RESERVED_MASK 0x60000000L +#define VCN_CE_ERR_STATUS_HI_MMSCHD__Err_clr_MASK 0x80000000L +//VCN_UE_ERR_STATUS_LO_JPEG0S +#define VCN_UE_ERR_STATUS_LO_JPEG0S__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_JPEG0S__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_JPEG0S__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_JPEG0S__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_JPEG0S__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_JPEG0S__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_JPEG0S__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_JPEG0S__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_JPEG0S +#define VCN_UE_ERR_STATUS_HI_JPEG0S__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_JPEG0S__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_JPEG0S__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_JPEG0S__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_JPEG0S__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_JPEG0S__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_JPEG0S__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_JPEG0S__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_JPEG0S__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_JPEG0S__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_JPEG0S__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_JPEG0S__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_JPEG0S__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_JPEG0S__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_JPEG0S__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_JPEG0S__Err_clr_MASK 0x80000000L +//VCN_UE_ERR_STATUS_LO_JPEG0D +#define VCN_UE_ERR_STATUS_LO_JPEG0D__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_JPEG0D__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_JPEG0D__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_JPEG0D__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_JPEG0D__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_JPEG0D__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_JPEG0D__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_JPEG0D__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_JPEG0D +#define VCN_UE_ERR_STATUS_HI_JPEG0D__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_JPEG0D__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_JPEG0D__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_JPEG0D__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_JPEG0D__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_JPEG0D__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_JPEG0D__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_JPEG0D__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_JPEG0D__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_JPEG0D__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_JPEG0D__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_JPEG0D__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_JPEG0D__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_JPEG0D__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_JPEG0D__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_JPEG0D__Err_clr_MASK 0x80000000L +//VCN_UE_ERR_STATUS_LO_JPEG1S +#define VCN_UE_ERR_STATUS_LO_JPEG1S__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_JPEG1S__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_JPEG1S__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_JPEG1S__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_JPEG1S__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_JPEG1S__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_JPEG1S__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_JPEG1S__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_JPEG1S +#define VCN_UE_ERR_STATUS_HI_JPEG1S__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_JPEG1S__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_JPEG1S__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_JPEG1S__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_JPEG1S__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_JPEG1S__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_JPEG1S__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_JPEG1S__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_JPEG1S__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_JPEG1S__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_JPEG1S__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_JPEG1S__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_JPEG1S__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_JPEG1S__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_JPEG1S__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_JPEG1S__Err_clr_MASK 0x80000000L +//VCN_UE_ERR_STATUS_LO_JPEG1D +#define VCN_UE_ERR_STATUS_LO_JPEG1D__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_JPEG1D__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_JPEG1D__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_JPEG1D__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_JPEG1D__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_JPEG1D__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_JPEG1D__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_JPEG1D__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_JPEG1D +#define VCN_UE_ERR_STATUS_HI_JPEG1D__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_JPEG1D__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_JPEG1D__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_JPEG1D__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_JPEG1D__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_JPEG1D__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_JPEG1D__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_JPEG1D__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_JPEG1D__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_JPEG1D__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_JPEG1D__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_JPEG1D__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_JPEG1D__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_JPEG1D__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_JPEG1D__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_JPEG1D__Err_clr_MASK 0x80000000L +//VCN_UE_ERR_STATUS_LO_JPEG2S +#define VCN_UE_ERR_STATUS_LO_JPEG2S__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_JPEG2S__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_JPEG2S__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_JPEG2S__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_JPEG2S__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_JPEG2S__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_JPEG2S__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_JPEG2S__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_JPEG2S +#define VCN_UE_ERR_STATUS_HI_JPEG2S__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_JPEG2S__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_JPEG2S__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_JPEG2S__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_JPEG2S__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_JPEG2S__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_JPEG2S__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_JPEG2S__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_JPEG2S__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_JPEG2S__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_JPEG2S__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_JPEG2S__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_JPEG2S__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_JPEG2S__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_JPEG2S__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_JPEG2S__Err_clr_MASK 0x80000000L +//VCN_UE_ERR_STATUS_LO_JPEG2D +#define VCN_UE_ERR_STATUS_LO_JPEG2D__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_JPEG2D__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_JPEG2D__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_JPEG2D__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_JPEG2D__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_JPEG2D__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_JPEG2D__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_JPEG2D__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_JPEG2D +#define VCN_UE_ERR_STATUS_HI_JPEG2D__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_JPEG2D__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_JPEG2D__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_JPEG2D__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_JPEG2D__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_JPEG2D__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_JPEG2D__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_JPEG2D__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_JPEG2D__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_JPEG2D__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_JPEG2D__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_JPEG2D__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_JPEG2D__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_JPEG2D__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_JPEG2D__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_JPEG2D__Err_clr_MASK 0x80000000L +//VCN_UE_ERR_STATUS_LO_JPEG3S +#define VCN_UE_ERR_STATUS_LO_JPEG3S__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_JPEG3S__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_JPEG3S__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_JPEG3S__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_JPEG3S__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_JPEG3S__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_JPEG3S__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_JPEG3S__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_JPEG3S +#define VCN_UE_ERR_STATUS_HI_JPEG3S__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_JPEG3S__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_JPEG3S__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_JPEG3S__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_JPEG3S__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_JPEG3S__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_JPEG3S__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_JPEG3S__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_JPEG3S__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_JPEG3S__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_JPEG3S__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_JPEG3S__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_JPEG3S__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_JPEG3S__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_JPEG3S__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_JPEG3S__Err_clr_MASK 0x80000000L +//VCN_UE_ERR_STATUS_LO_JPEG3D +#define VCN_UE_ERR_STATUS_LO_JPEG3D__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_JPEG3D__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_JPEG3D__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_JPEG3D__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_JPEG3D__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_JPEG3D__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_JPEG3D__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_JPEG3D__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_JPEG3D +#define VCN_UE_ERR_STATUS_HI_JPEG3D__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_JPEG3D__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_JPEG3D__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_JPEG3D__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_JPEG3D__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_JPEG3D__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_JPEG3D__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_JPEG3D__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_JPEG3D__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_JPEG3D__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_JPEG3D__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_JPEG3D__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_JPEG3D__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_JPEG3D__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_JPEG3D__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_JPEG3D__Err_clr_MASK 0x80000000L +//VCN_UE_ERR_STATUS_LO_JPEG4S +#define VCN_UE_ERR_STATUS_LO_JPEG4S__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_JPEG4S__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_JPEG4S__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_JPEG4S__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_JPEG4S__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_JPEG4S__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_JPEG4S__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_JPEG4S__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_JPEG4S +#define VCN_UE_ERR_STATUS_HI_JPEG4S__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_JPEG4S__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_JPEG4S__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_JPEG4S__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_JPEG4S__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_JPEG4S__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_JPEG4S__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_JPEG4S__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_JPEG4S__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_JPEG4S__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_JPEG4S__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_JPEG4S__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_JPEG4S__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_JPEG4S__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_JPEG4S__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_JPEG4S__Err_clr_MASK 0x80000000L +//VCN_UE_ERR_STATUS_LO_JPEG4D +#define VCN_UE_ERR_STATUS_LO_JPEG4D__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_JPEG4D__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_JPEG4D__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_JPEG4D__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_JPEG4D__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_JPEG4D__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_JPEG4D__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_JPEG4D__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_JPEG4D +#define VCN_UE_ERR_STATUS_HI_JPEG4D__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_JPEG4D__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_JPEG4D__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_JPEG4D__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_JPEG4D__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_JPEG4D__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_JPEG4D__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_JPEG4D__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_JPEG4D__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_JPEG4D__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_JPEG4D__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_JPEG4D__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_JPEG4D__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_JPEG4D__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_JPEG4D__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_JPEG4D__Err_clr_MASK 0x80000000L +//VCN_UE_ERR_STATUS_LO_JPEG5S +#define VCN_UE_ERR_STATUS_LO_JPEG5S__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_JPEG5S__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_JPEG5S__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_JPEG5S__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_JPEG5S__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_JPEG5S__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_JPEG5S__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_JPEG5S__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_JPEG5S +#define VCN_UE_ERR_STATUS_HI_JPEG5S__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_JPEG5S__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_JPEG5S__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_JPEG5S__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_JPEG5S__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_JPEG5S__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_JPEG5S__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_JPEG5S__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_JPEG5S__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_JPEG5S__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_JPEG5S__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_JPEG5S__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_JPEG5S__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_JPEG5S__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_JPEG5S__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_JPEG5S__Err_clr_MASK 0x80000000L +//VCN_UE_ERR_STATUS_LO_JPEG5D +#define VCN_UE_ERR_STATUS_LO_JPEG5D__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_JPEG5D__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_JPEG5D__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_JPEG5D__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_JPEG5D__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_JPEG5D__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_JPEG5D__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_JPEG5D__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_JPEG5D +#define VCN_UE_ERR_STATUS_HI_JPEG5D__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_JPEG5D__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_JPEG5D__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_JPEG5D__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_JPEG5D__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_JPEG5D__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_JPEG5D__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_JPEG5D__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_JPEG5D__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_JPEG5D__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_JPEG5D__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_JPEG5D__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_JPEG5D__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_JPEG5D__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_JPEG5D__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_JPEG5D__Err_clr_MASK 0x80000000L +//VCN_UE_ERR_STATUS_LO_JPEG6S +#define VCN_UE_ERR_STATUS_LO_JPEG6S__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_JPEG6S__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_JPEG6S__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_JPEG6S__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_JPEG6S__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_JPEG6S__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_JPEG6S__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_JPEG6S__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_JPEG6S +#define VCN_UE_ERR_STATUS_HI_JPEG6S__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_JPEG6S__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_JPEG6S__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_JPEG6S__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_JPEG6S__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_JPEG6S__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_JPEG6S__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_JPEG6S__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_JPEG6S__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_JPEG6S__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_JPEG6S__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_JPEG6S__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_JPEG6S__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_JPEG6S__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_JPEG6S__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_JPEG6S__Err_clr_MASK 0x80000000L +//VCN_UE_ERR_STATUS_LO_JPEG6D +#define VCN_UE_ERR_STATUS_LO_JPEG6D__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_JPEG6D__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_JPEG6D__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_JPEG6D__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_JPEG6D__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_JPEG6D__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_JPEG6D__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_JPEG6D__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_JPEG6D +#define VCN_UE_ERR_STATUS_HI_JPEG6D__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_JPEG6D__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_JPEG6D__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_JPEG6D__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_JPEG6D__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_JPEG6D__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_JPEG6D__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_JPEG6D__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_JPEG6D__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_JPEG6D__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_JPEG6D__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_JPEG6D__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_JPEG6D__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_JPEG6D__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_JPEG6D__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_JPEG6D__Err_clr_MASK 0x80000000L +//VCN_UE_ERR_STATUS_LO_JPEG7S +#define VCN_UE_ERR_STATUS_LO_JPEG7S__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_JPEG7S__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_JPEG7S__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_JPEG7S__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_JPEG7S__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_JPEG7S__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_JPEG7S__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_JPEG7S__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_JPEG7S +#define VCN_UE_ERR_STATUS_HI_JPEG7S__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_JPEG7S__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_JPEG7S__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_JPEG7S__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_JPEG7S__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_JPEG7S__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_JPEG7S__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_JPEG7S__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_JPEG7S__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_JPEG7S__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_JPEG7S__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_JPEG7S__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_JPEG7S__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_JPEG7S__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_JPEG7S__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_JPEG7S__Err_clr_MASK 0x80000000L +//VCN_UE_ERR_STATUS_LO_JPEG7D +#define VCN_UE_ERR_STATUS_LO_JPEG7D__Err_Status_Valid_Flag__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_LO_JPEG7D__Address_Valid_Flag__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_LO_JPEG7D__Address__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_LO_JPEG7D__Memory_id__SHIFT 0x18 +#define VCN_UE_ERR_STATUS_LO_JPEG7D__Err_Status_Valid_Flag_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_LO_JPEG7D__Address_Valid_Flag_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_LO_JPEG7D__Address_MASK 0x00FFFFFCL +#define VCN_UE_ERR_STATUS_LO_JPEG7D__Memory_id_MASK 0xFF000000L +//VCN_UE_ERR_STATUS_HI_JPEG7D +#define VCN_UE_ERR_STATUS_HI_JPEG7D__ECC__SHIFT 0x0 +#define VCN_UE_ERR_STATUS_HI_JPEG7D__Parity__SHIFT 0x1 +#define VCN_UE_ERR_STATUS_HI_JPEG7D__Err_Info_Valid_Flag__SHIFT 0x2 +#define VCN_UE_ERR_STATUS_HI_JPEG7D__Err_Info__SHIFT 0x3 +#define VCN_UE_ERR_STATUS_HI_JPEG7D__UE_Cnt__SHIFT 0x17 +#define VCN_UE_ERR_STATUS_HI_JPEG7D__FED_Cnt__SHIFT 0x1a +#define VCN_UE_ERR_STATUS_HI_JPEG7D__RESERVED__SHIFT 0x1d +#define VCN_UE_ERR_STATUS_HI_JPEG7D__Err_clr__SHIFT 0x1f +#define VCN_UE_ERR_STATUS_HI_JPEG7D__ECC_MASK 0x00000001L +#define VCN_UE_ERR_STATUS_HI_JPEG7D__Parity_MASK 0x00000002L +#define VCN_UE_ERR_STATUS_HI_JPEG7D__Err_Info_Valid_Flag_MASK 0x00000004L +#define VCN_UE_ERR_STATUS_HI_JPEG7D__Err_Info_MASK 0x007FFFF8L +#define VCN_UE_ERR_STATUS_HI_JPEG7D__UE_Cnt_MASK 0x03800000L +#define VCN_UE_ERR_STATUS_HI_JPEG7D__FED_Cnt_MASK 0x1C000000L +#define VCN_UE_ERR_STATUS_HI_JPEG7D__RESERVED_MASK 0x60000000L +#define VCN_UE_ERR_STATUS_HI_JPEG7D__Err_clr_MASK 0x80000000L + +// addressBlock: aid_uvd0_uvd_jrbc0_uvd_jrbc_dec +//UVD_JRBC0_UVD_JRBC_RB_WPTR +#define UVD_JRBC0_UVD_JRBC_RB_WPTR__RB_WPTR__SHIFT 0x4 +#define UVD_JRBC0_UVD_JRBC_RB_WPTR__RB_WPTR_MASK 0x007FFFF0L +//UVD_JRBC0_UVD_JRBC_RB_CNTL +#define UVD_JRBC0_UVD_JRBC_RB_CNTL__RB_NO_FETCH__SHIFT 0x0 +#define UVD_JRBC0_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT 0x1 +#define UVD_JRBC0_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER__SHIFT 0x4 +#define UVD_JRBC0_UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK 0x00000001L +#define UVD_JRBC0_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK 0x00000002L +#define UVD_JRBC0_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER_MASK 0x0007FFF0L +//UVD_JRBC0_UVD_JRBC_IB_SIZE +#define UVD_JRBC0_UVD_JRBC_IB_SIZE__IB_SIZE__SHIFT 0x4 +#define UVD_JRBC0_UVD_JRBC_IB_SIZE__IB_SIZE_MASK 0x007FFFF0L +//UVD_JRBC0_UVD_JRBC_URGENT_CNTL +#define UVD_JRBC0_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT 0x0 +#define UVD_JRBC0_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK 0x00000003L +//UVD_JRBC0_UVD_JRBC_RB_REF_DATA +#define UVD_JRBC0_UVD_JRBC_RB_REF_DATA__REF_DATA__SHIFT 0x0 +#define UVD_JRBC0_UVD_JRBC_RB_REF_DATA__REF_DATA_MASK 0xFFFFFFFFL +//UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER +#define UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT 0x0 +#define UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT 0x10 +#define UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT 0x18 +#define UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT 0x19 +#define UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK 0x0000FFFFL +#define UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK 0x00FF0000L +#define UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK 0x01000000L +#define UVD_JRBC0_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK 0x02000000L +//UVD_JRBC0_UVD_JRBC_SOFT_RESET +#define UVD_JRBC0_UVD_JRBC_SOFT_RESET__RESET__SHIFT 0x0 +#define UVD_JRBC0_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS__SHIFT 0x11 +#define UVD_JRBC0_UVD_JRBC_SOFT_RESET__RESET_MASK 0x00000001L +#define UVD_JRBC0_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS_MASK 0x00020000L +//UVD_JRBC0_UVD_JRBC_STATUS +#define UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE__SHIFT 0x0 +#define UVD_JRBC0_UVD_JRBC_STATUS__IB_JOB_DONE__SHIFT 0x1 +#define UVD_JRBC0_UVD_JRBC_STATUS__RB_ILLEGAL_CMD__SHIFT 0x2 +#define UVD_JRBC0_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT__SHIFT 0x3 +#define UVD_JRBC0_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT__SHIFT 0x4 +#define UVD_JRBC0_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT__SHIFT 0x5 +#define UVD_JRBC0_UVD_JRBC_STATUS__IB_ILLEGAL_CMD__SHIFT 0x6 +#define UVD_JRBC0_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT__SHIFT 0x7 +#define UVD_JRBC0_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT__SHIFT 0x8 +#define UVD_JRBC0_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT__SHIFT 0x9 +#define UVD_JRBC0_UVD_JRBC_STATUS__RB_TRAP_STATUS__SHIFT 0xa +#define UVD_JRBC0_UVD_JRBC_STATUS__PREEMPT_STATUS__SHIFT 0xb +#define UVD_JRBC0_UVD_JRBC_STATUS__IB_TRAP_STATUS__SHIFT 0xc +#define UVD_JRBC0_UVD_JRBC_STATUS__INT_EN__SHIFT 0x10 +#define UVD_JRBC0_UVD_JRBC_STATUS__INT_ACK__SHIFT 0x11 +#define UVD_JRBC0_UVD_JRBC_STATUS__RB_JOB_DONE_MASK 0x00000001L +#define UVD_JRBC0_UVD_JRBC_STATUS__IB_JOB_DONE_MASK 0x00000002L +#define UVD_JRBC0_UVD_JRBC_STATUS__RB_ILLEGAL_CMD_MASK 0x00000004L +#define UVD_JRBC0_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT_MASK 0x00000008L +#define UVD_JRBC0_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT_MASK 0x00000010L +#define UVD_JRBC0_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT_MASK 0x00000020L +#define UVD_JRBC0_UVD_JRBC_STATUS__IB_ILLEGAL_CMD_MASK 0x00000040L +#define UVD_JRBC0_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT_MASK 0x00000080L +#define UVD_JRBC0_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT_MASK 0x00000100L +#define UVD_JRBC0_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT_MASK 0x00000200L +#define UVD_JRBC0_UVD_JRBC_STATUS__RB_TRAP_STATUS_MASK 0x00000400L +#define UVD_JRBC0_UVD_JRBC_STATUS__PREEMPT_STATUS_MASK 0x00000800L +#define UVD_JRBC0_UVD_JRBC_STATUS__IB_TRAP_STATUS_MASK 0x00001000L +#define UVD_JRBC0_UVD_JRBC_STATUS__INT_EN_MASK 0x00010000L +#define UVD_JRBC0_UVD_JRBC_STATUS__INT_ACK_MASK 0x00020000L +//UVD_JRBC0_UVD_JRBC_RB_RPTR +#define UVD_JRBC0_UVD_JRBC_RB_RPTR__RB_RPTR__SHIFT 0x4 +#define UVD_JRBC0_UVD_JRBC_RB_RPTR__RB_RPTR_MASK 0x007FFFF0L +//UVD_JRBC0_UVD_JRBC_RB_BUF_STATUS +#define UVD_JRBC0_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID__SHIFT 0x0 +#define UVD_JRBC0_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT 0x10 +#define UVD_JRBC0_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT 0x18 +#define UVD_JRBC0_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID_MASK 0x0000FFFFL +#define UVD_JRBC0_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR_MASK 0x000F0000L +#define UVD_JRBC0_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR_MASK 0x03000000L +//UVD_JRBC0_UVD_JRBC_IB_BUF_STATUS +#define UVD_JRBC0_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID__SHIFT 0x0 +#define UVD_JRBC0_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT 0x10 +#define UVD_JRBC0_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT 0x18 +#define UVD_JRBC0_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID_MASK 0x0000FFFFL +#define UVD_JRBC0_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR_MASK 0x000F0000L +#define UVD_JRBC0_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR_MASK 0x03000000L +//UVD_JRBC0_UVD_JRBC_IB_SIZE_UPDATE +#define UVD_JRBC0_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT 0x4 +#define UVD_JRBC0_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK 0x007FFFF0L +//UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER +#define UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT 0x0 +#define UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT 0x10 +#define UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT 0x18 +#define UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT 0x19 +#define UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK 0x0000FFFFL +#define UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK 0x00FF0000L +#define UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK 0x01000000L +#define UVD_JRBC0_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK 0x02000000L +//UVD_JRBC0_UVD_JRBC_IB_REF_DATA +#define UVD_JRBC0_UVD_JRBC_IB_REF_DATA__REF_DATA__SHIFT 0x0 +#define UVD_JRBC0_UVD_JRBC_IB_REF_DATA__REF_DATA_MASK 0xFFFFFFFFL +//UVD_JRBC0_UVD_JPEG_PREEMPT_CMD +#define UVD_JRBC0_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN__SHIFT 0x0 +#define UVD_JRBC0_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE__SHIFT 0x1 +#define UVD_JRBC0_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD__SHIFT 0x2 +#define UVD_JRBC0_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN_MASK 0x00000001L +#define UVD_JRBC0_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE_MASK 0x00000002L +#define UVD_JRBC0_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD_MASK 0x00000004L +//UVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA0 +#define UVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0__SHIFT 0x0 +#define UVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0_MASK 0xFFFFFFFFL +//UVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA1 +#define UVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1__SHIFT 0x0 +#define UVD_JRBC0_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1_MASK 0xFFFFFFFFL +//UVD_JRBC0_UVD_JRBC_RB_SIZE +#define UVD_JRBC0_UVD_JRBC_RB_SIZE__RB_SIZE__SHIFT 0x4 +#define UVD_JRBC0_UVD_JRBC_RB_SIZE__RB_SIZE_MASK 0x00FFFFF0L +//UVD_JRBC0_UVD_JRBC_SCRATCH0 +#define UVD_JRBC0_UVD_JRBC_SCRATCH0__SCRATCH0__SHIFT 0x0 +#define UVD_JRBC0_UVD_JRBC_SCRATCH0__SCRATCH0_MASK 0xFFFFFFFFL + + +// addressBlock: aid_uvd0_uvd_jmi0_uvd_jmi_dec +//UVD_JMI0_UVD_JPEG_DEC_PF_CTRL +#define UVD_JMI0_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS__SHIFT 0x0 +#define UVD_JMI0_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING__SHIFT 0x1 +#define UVD_JMI0_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS_MASK 0x00000001L +#define UVD_JMI0_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING_MASK 0x00000002L +//UVD_JMI0_UVD_LMI_JRBC_CTRL +#define UVD_JMI0_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN__SHIFT 0x1 +#define UVD_JMI0_UVD_LMI_JRBC_CTRL__RD_MAX_BURST__SHIFT 0x4 +#define UVD_JMI0_UVD_LMI_JRBC_CTRL__WR_MAX_BURST__SHIFT 0x8 +#define UVD_JMI0_UVD_LMI_JRBC_CTRL__RD_SWAP__SHIFT 0x14 +#define UVD_JMI0_UVD_LMI_JRBC_CTRL__WR_SWAP__SHIFT 0x16 +#define UVD_JMI0_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN_MASK 0x00000001L +#define UVD_JMI0_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN_MASK 0x00000002L +#define UVD_JMI0_UVD_LMI_JRBC_CTRL__RD_MAX_BURST_MASK 0x000000F0L +#define UVD_JMI0_UVD_LMI_JRBC_CTRL__WR_MAX_BURST_MASK 0x00000F00L +#define UVD_JMI0_UVD_LMI_JRBC_CTRL__RD_SWAP_MASK 0x00300000L +#define UVD_JMI0_UVD_LMI_JRBC_CTRL__WR_SWAP_MASK 0x00C00000L +//UVD_JMI0_UVD_LMI_JPEG_CTRL +#define UVD_JMI0_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN__SHIFT 0x1 +#define UVD_JMI0_UVD_LMI_JPEG_CTRL__RD_MAX_BURST__SHIFT 0x4 +#define UVD_JMI0_UVD_LMI_JPEG_CTRL__WR_MAX_BURST__SHIFT 0x8 +#define UVD_JMI0_UVD_LMI_JPEG_CTRL__RD_SWAP__SHIFT 0x14 +#define UVD_JMI0_UVD_LMI_JPEG_CTRL__WR_SWAP__SHIFT 0x16 +#define UVD_JMI0_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN_MASK 0x00000001L +#define UVD_JMI0_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN_MASK 0x00000002L +#define UVD_JMI0_UVD_LMI_JPEG_CTRL__RD_MAX_BURST_MASK 0x000000F0L +#define UVD_JMI0_UVD_LMI_JPEG_CTRL__WR_MAX_BURST_MASK 0x00000F00L +#define UVD_JMI0_UVD_LMI_JPEG_CTRL__RD_SWAP_MASK 0x00300000L +#define UVD_JMI0_UVD_LMI_JPEG_CTRL__WR_SWAP_MASK 0x00C00000L +//UVD_JMI0_JPEG_LMI_DROP +#define UVD_JMI0_JPEG_LMI_DROP__JPEG_WR_DROP__SHIFT 0x0 +#define UVD_JMI0_JPEG_LMI_DROP__JRBC_WR_DROP__SHIFT 0x1 +#define UVD_JMI0_JPEG_LMI_DROP__JPEG_RD_DROP__SHIFT 0x2 +#define UVD_JMI0_JPEG_LMI_DROP__JRBC_RD_DROP__SHIFT 0x3 +#define UVD_JMI0_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP__SHIFT 0x4 +#define UVD_JMI0_JPEG_LMI_DROP__JPEG_WR_DROP_MASK 0x00000001L +#define UVD_JMI0_JPEG_LMI_DROP__JRBC_WR_DROP_MASK 0x00000002L +#define UVD_JMI0_JPEG_LMI_DROP__JPEG_RD_DROP_MASK 0x00000004L +#define UVD_JMI0_JPEG_LMI_DROP__JRBC_RD_DROP_MASK 0x00000008L +#define UVD_JMI0_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP_MASK 0x00000010L +//UVD_JMI0_UVD_LMI_JRBC_IB_VMID +#define UVD_JMI0_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID__SHIFT 0x4 +#define UVD_JMI0_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID__SHIFT 0x8 +#define UVD_JMI0_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID_MASK 0x0000000FL +#define UVD_JMI0_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID_MASK 0x000000F0L +#define UVD_JMI0_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID_MASK 0x00000F00L +//UVD_JMI0_UVD_LMI_JRBC_RB_VMID +#define UVD_JMI0_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID__SHIFT 0x4 +#define UVD_JMI0_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID__SHIFT 0x8 +#define UVD_JMI0_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID_MASK 0x0000000FL +#define UVD_JMI0_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID_MASK 0x000000F0L +#define UVD_JMI0_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID_MASK 0x00000F00L +//UVD_JMI0_UVD_LMI_JPEG_VMID +#define UVD_JMI0_UVD_LMI_JPEG_VMID__JPEG_RD_VMID__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JPEG_VMID__JPEG_WR_VMID__SHIFT 0x4 +#define UVD_JMI0_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID__SHIFT 0x8 +#define UVD_JMI0_UVD_LMI_JPEG_VMID__JPEG_RD_VMID_MASK 0x0000000FL +#define UVD_JMI0_UVD_LMI_JPEG_VMID__JPEG_WR_VMID_MASK 0x000000F0L +#define UVD_JMI0_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID_MASK 0x00000F00L +//UVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW +#define UVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH +#define UVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW +#define UVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH +#define UVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW +#define UVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH +#define UVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI0_UVD_LMI_JPEG_PREEMPT_VMID +#define UVD_JMI0_UVD_LMI_JPEG_PREEMPT_VMID__VMID__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JPEG_PREEMPT_VMID__VMID_MASK 0x0000000FL +//UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP__SHIFT 0x0 +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP__SHIFT 0x2 +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP__SHIFT 0x4 +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP__SHIFT 0x6 +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP__SHIFT 0x8 +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP__SHIFT 0xa +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP__SHIFT 0xc +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP__SHIFT 0xe +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP__SHIFT 0x10 +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP_MASK 0x00000003L +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP_MASK 0x0000000CL +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP_MASK 0x00000030L +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP_MASK 0x000000C0L +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP_MASK 0x00000300L +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP_MASK 0x00000C00L +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP_MASK 0x00003000L +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP_MASK 0x0000C000L +#define UVD_JMI0_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP_MASK 0x00030000L +//UVD_JMI0_UVD_JMI_ATOMIC_CNTL +#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en__SHIFT 0x0 +#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__atomic_max_burst__SHIFT 0x1 +#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop__SHIFT 0x5 +#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en__SHIFT 0x6 +#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG__SHIFT 0x7 +#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE__SHIFT 0xb +#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en_MASK 0x00000001L +#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__atomic_max_burst_MASK 0x0000001EL +#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop_MASK 0x00000020L +#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en_MASK 0x00000040L +#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG_MASK 0x00000780L +#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE_MASK 0x00000800L +//UVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW +#define UVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH +#define UVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI0_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_LOW +#define UVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH +#define UVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW +#define UVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH +#define UVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_LOW +#define UVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH +#define UVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW +#define UVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH +#define UVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI0_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI0_UVD_JMI_ATOMIC_CNTL2 +#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap__SHIFT 0x10 +#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP__SHIFT 0x18 +#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap_MASK 0x00FF0000L +#define UVD_JMI0_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP_MASK 0xFF000000L + + +// addressBlock: aid_uvd0_uvd_jmi_common_dec +//UVD_JADP_MCIF_URGENT_CTRL +#define UVD_JADP_MCIF_URGENT_CTRL__WR_WATERMARK__SHIFT 0x0 +#define UVD_JADP_MCIF_URGENT_CTRL__RD_WATERMARK__SHIFT 0x6 +#define UVD_JADP_MCIF_URGENT_CTRL__WR_RD_URGENT_TIMER__SHIFT 0xb +#define UVD_JADP_MCIF_URGENT_CTRL__WR_URGENT_PROG_STEP__SHIFT 0x11 +#define UVD_JADP_MCIF_URGENT_CTRL__RD_URGENT_PROG_STEP__SHIFT 0x15 +#define UVD_JADP_MCIF_URGENT_CTRL__WR_QOS_EN__SHIFT 0x19 +#define UVD_JADP_MCIF_URGENT_CTRL__RD_QOS_EN__SHIFT 0x1a +#define UVD_JADP_MCIF_URGENT_CTRL__WR_WATERMARK_MASK 0x0000003FL +#define UVD_JADP_MCIF_URGENT_CTRL__RD_WATERMARK_MASK 0x000007C0L +#define UVD_JADP_MCIF_URGENT_CTRL__WR_RD_URGENT_TIMER_MASK 0x0001F800L +#define UVD_JADP_MCIF_URGENT_CTRL__WR_URGENT_PROG_STEP_MASK 0x001E0000L +#define UVD_JADP_MCIF_URGENT_CTRL__RD_URGENT_PROG_STEP_MASK 0x01E00000L +#define UVD_JADP_MCIF_URGENT_CTRL__WR_QOS_EN_MASK 0x02000000L +#define UVD_JADP_MCIF_URGENT_CTRL__RD_QOS_EN_MASK 0x04000000L +//UVD_JMI_URGENT_CTRL +#define UVD_JMI_URGENT_CTRL__ENABLE_MC_RD_URGENT_STALL__SHIFT 0x0 +#define UVD_JMI_URGENT_CTRL__ASSERT_MC_RD_URGENT__SHIFT 0x4 +#define UVD_JMI_URGENT_CTRL__ENABLE_MC_WR_URGENT_STALL__SHIFT 0x10 +#define UVD_JMI_URGENT_CTRL__ASSERT_MC_WR_URGENT__SHIFT 0x14 +#define UVD_JMI_URGENT_CTRL__ENABLE_MC_RD_URGENT_STALL_MASK 0x00000001L +#define UVD_JMI_URGENT_CTRL__ASSERT_MC_RD_URGENT_MASK 0x000000F0L +#define UVD_JMI_URGENT_CTRL__ENABLE_MC_WR_URGENT_STALL_MASK 0x00010000L +#define UVD_JMI_URGENT_CTRL__ASSERT_MC_WR_URGENT_MASK 0x00F00000L +//UVD_JMI_CTRL +#define UVD_JMI_CTRL__STALL_MC_ARB__SHIFT 0x0 +#define UVD_JMI_CTRL__MASK_MC_URGENT__SHIFT 0x1 +#define UVD_JMI_CTRL__ASSERT_MC_URGENT__SHIFT 0x2 +#define UVD_JMI_CTRL__MC_RD_ARB_WAIT_TIMER__SHIFT 0x8 +#define UVD_JMI_CTRL__MC_WR_ARB_WAIT_TIMER__SHIFT 0x10 +#define UVD_JMI_CTRL__STALL_MC_ARB_MASK 0x00000001L +#define UVD_JMI_CTRL__MASK_MC_URGENT_MASK 0x00000002L +#define UVD_JMI_CTRL__ASSERT_MC_URGENT_MASK 0x00000004L +#define UVD_JMI_CTRL__MC_RD_ARB_WAIT_TIMER_MASK 0x0000FF00L +#define UVD_JMI_CTRL__MC_WR_ARB_WAIT_TIMER_MASK 0x00FF0000L +//JPEG_MEMCHECK_CLAMPING_CNTL +#define JPEG_MEMCHECK_CLAMPING_CNTL__CLAMP_TO_SAFE_ADDR_EN__SHIFT 0x0 +#define JPEG_MEMCHECK_CLAMPING_CNTL__CLAMP_TO_SAFE_ADDR_EN_MASK 0x00000001L +//JPEG_MEMCHECK_SAFE_ADDR +#define JPEG_MEMCHECK_SAFE_ADDR__MEMCHECK_SAFE_ADDR__SHIFT 0x0 +#define JPEG_MEMCHECK_SAFE_ADDR__MEMCHECK_SAFE_ADDR_MASK 0xFFFFFFFFL +//JPEG_MEMCHECK_SAFE_ADDR_64BIT +#define JPEG_MEMCHECK_SAFE_ADDR_64BIT__MEMCHECK_SAFE_ADDR_64BIT__SHIFT 0x0 +#define JPEG_MEMCHECK_SAFE_ADDR_64BIT__MEMCHECK_SAFE_ADDR_64BIT_MASK 0xFFFFFFFFL +//UVD_JMI_LAT_CTRL +#define UVD_JMI_LAT_CTRL__SCALE__SHIFT 0x0 +#define UVD_JMI_LAT_CTRL__MAX_START__SHIFT 0x8 +#define UVD_JMI_LAT_CTRL__MIN_START__SHIFT 0x9 +#define UVD_JMI_LAT_CTRL__AVG_START__SHIFT 0xa +#define UVD_JMI_LAT_CTRL__PERFMON_SYNC__SHIFT 0xb +#define UVD_JMI_LAT_CTRL__SKIP__SHIFT 0x10 +#define UVD_JMI_LAT_CTRL__SCALE_MASK 0x000000FFL +#define UVD_JMI_LAT_CTRL__MAX_START_MASK 0x00000100L +#define UVD_JMI_LAT_CTRL__MIN_START_MASK 0x00000200L +#define UVD_JMI_LAT_CTRL__AVG_START_MASK 0x00000400L +#define UVD_JMI_LAT_CTRL__PERFMON_SYNC_MASK 0x00000800L +#define UVD_JMI_LAT_CTRL__SKIP_MASK 0x000F0000L +//UVD_JMI_LAT_CNTR +#define UVD_JMI_LAT_CNTR__MAX_LAT__SHIFT 0x0 +#define UVD_JMI_LAT_CNTR__MIN_LAT__SHIFT 0x8 +#define UVD_JMI_LAT_CNTR__MAX_LAT_MASK 0x000000FFL +#define UVD_JMI_LAT_CNTR__MIN_LAT_MASK 0x0000FF00L +//UVD_JMI_AVG_LAT_CNTR +#define UVD_JMI_AVG_LAT_CNTR__ENV_LOW__SHIFT 0x0 +#define UVD_JMI_AVG_LAT_CNTR__ENV_HIGH__SHIFT 0x8 +#define UVD_JMI_AVG_LAT_CNTR__ENV_HIT__SHIFT 0x10 +#define UVD_JMI_AVG_LAT_CNTR__ENV_LOW_MASK 0x000000FFL +#define UVD_JMI_AVG_LAT_CNTR__ENV_HIGH_MASK 0x0000FF00L +#define UVD_JMI_AVG_LAT_CNTR__ENV_HIT_MASK 0xFFFF0000L +//UVD_JMI_PERFMON_CTRL +#define UVD_JMI_PERFMON_CTRL__PERFMON_STATE__SHIFT 0x0 +#define UVD_JMI_PERFMON_CTRL__PERFMON_SEL__SHIFT 0x8 +#define UVD_JMI_PERFMON_CTRL__PERFMON_STATE_MASK 0x00000003L +#define UVD_JMI_PERFMON_CTRL__PERFMON_SEL_MASK 0x00001F00L +//UVD_JMI_PERFMON_COUNT_LO +#define UVD_JMI_PERFMON_COUNT_LO__PERFMON_COUNT__SHIFT 0x0 +#define UVD_JMI_PERFMON_COUNT_LO__PERFMON_COUNT_MASK 0xFFFFFFFFL +//UVD_JMI_PERFMON_COUNT_HI +#define UVD_JMI_PERFMON_COUNT_HI__PERFMON_COUNT__SHIFT 0x0 +#define UVD_JMI_PERFMON_COUNT_HI__PERFMON_COUNT_MASK 0x0000FFFFL +//UVD_JMI_CLEAN_STATUS +#define UVD_JMI_CLEAN_STATUS__LMI_READ_CLEAN__SHIFT 0x0 +#define UVD_JMI_CLEAN_STATUS__LMI_READ_CLEAN_RAW__SHIFT 0x1 +#define UVD_JMI_CLEAN_STATUS__LMI_WRITE_CLEAN__SHIFT 0x2 +#define UVD_JMI_CLEAN_STATUS__LMI_WRITE_CLEAN_RAW__SHIFT 0x3 +#define UVD_JMI_CLEAN_STATUS__MC_WRITE_PENDING__SHIFT 0x4 +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE0_READ_CLEAN__SHIFT 0x8 +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE1_READ_CLEAN__SHIFT 0x9 +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE2_READ_CLEAN__SHIFT 0xa +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE3_READ_CLEAN__SHIFT 0xb +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE4_READ_CLEAN__SHIFT 0xc +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE5_READ_CLEAN__SHIFT 0xd +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE6_READ_CLEAN__SHIFT 0xe +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE7_READ_CLEAN__SHIFT 0xf +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE0_WRITE_CLEAN__SHIFT 0x10 +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE1_WRITE_CLEAN__SHIFT 0x11 +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE2_WRITE_CLEAN__SHIFT 0x12 +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE3_WRITE_CLEAN__SHIFT 0x13 +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE4_WRITE_CLEAN__SHIFT 0x14 +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE5_WRITE_CLEAN__SHIFT 0x15 +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE6_WRITE_CLEAN__SHIFT 0x16 +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE7_WRITE_CLEAN__SHIFT 0x17 +#define UVD_JMI_CLEAN_STATUS__LMI_READ_CLEAN_MASK 0x00000001L +#define UVD_JMI_CLEAN_STATUS__LMI_READ_CLEAN_RAW_MASK 0x00000002L +#define UVD_JMI_CLEAN_STATUS__LMI_WRITE_CLEAN_MASK 0x00000004L +#define UVD_JMI_CLEAN_STATUS__LMI_WRITE_CLEAN_RAW_MASK 0x00000008L +#define UVD_JMI_CLEAN_STATUS__MC_WRITE_PENDING_MASK 0x00000010L +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE0_READ_CLEAN_MASK 0x00000100L +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE1_READ_CLEAN_MASK 0x00000200L +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE2_READ_CLEAN_MASK 0x00000400L +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE3_READ_CLEAN_MASK 0x00000800L +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE4_READ_CLEAN_MASK 0x00001000L +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE5_READ_CLEAN_MASK 0x00002000L +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE6_READ_CLEAN_MASK 0x00004000L +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE7_READ_CLEAN_MASK 0x00008000L +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE0_WRITE_CLEAN_MASK 0x00010000L +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE1_WRITE_CLEAN_MASK 0x00020000L +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE2_WRITE_CLEAN_MASK 0x00040000L +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE3_WRITE_CLEAN_MASK 0x00080000L +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE4_WRITE_CLEAN_MASK 0x00100000L +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE5_WRITE_CLEAN_MASK 0x00200000L +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE6_WRITE_CLEAN_MASK 0x00400000L +#define UVD_JMI_CLEAN_STATUS__DJPEG_CORE7_WRITE_CLEAN_MASK 0x00800000L +//UVD_JMI_CNTL +#define UVD_JMI_CNTL__SOFT_RESET__SHIFT 0x0 +#define UVD_JMI_CNTL__MC_RD_REQ_RET_MAX__SHIFT 0x8 +#define UVD_JMI_CNTL__SOFT_RESET_MASK 0x00000001L +#define UVD_JMI_CNTL__MC_RD_REQ_RET_MAX_MASK 0x0003FF00L + + +// addressBlock: aid_uvd0_uvd_jpeg_common_dec +//JPEG_SOFT_RESET_STATUS +#define JPEG_SOFT_RESET_STATUS__JPEG0_DEC_RESET_STATUS__SHIFT 0x0 +#define JPEG_SOFT_RESET_STATUS__JPEG1_DEC_RESET_STATUS__SHIFT 0x1 +#define JPEG_SOFT_RESET_STATUS__JPEG2_DEC_RESET_STATUS__SHIFT 0x2 +#define JPEG_SOFT_RESET_STATUS__JPEG3_DEC_RESET_STATUS__SHIFT 0x3 +#define JPEG_SOFT_RESET_STATUS__JPEG4_DEC_RESET_STATUS__SHIFT 0x4 +#define JPEG_SOFT_RESET_STATUS__JPEG5_DEC_RESET_STATUS__SHIFT 0x5 +#define JPEG_SOFT_RESET_STATUS__JPEG6_DEC_RESET_STATUS__SHIFT 0x6 +#define JPEG_SOFT_RESET_STATUS__JPEG7_DEC_RESET_STATUS__SHIFT 0x7 +#define JPEG_SOFT_RESET_STATUS__DJRBC0_RESET_STATUS__SHIFT 0x8 +#define JPEG_SOFT_RESET_STATUS__DJRBC1_RESET_STATUS__SHIFT 0x9 +#define JPEG_SOFT_RESET_STATUS__DJRBC2_RESET_STATUS__SHIFT 0xa +#define JPEG_SOFT_RESET_STATUS__DJRBC3_RESET_STATUS__SHIFT 0xb +#define JPEG_SOFT_RESET_STATUS__DJRBC4_RESET_STATUS__SHIFT 0xc +#define JPEG_SOFT_RESET_STATUS__DJRBC5_RESET_STATUS__SHIFT 0xd +#define JPEG_SOFT_RESET_STATUS__DJRBC6_RESET_STATUS__SHIFT 0xe +#define JPEG_SOFT_RESET_STATUS__DJRBC7_RESET_STATUS__SHIFT 0xf +#define JPEG_SOFT_RESET_STATUS__JPEG_ENC_RESET_STATUS__SHIFT 0x11 +#define JPEG_SOFT_RESET_STATUS__EJRBC_RESET_STATUS__SHIFT 0x12 +#define JPEG_SOFT_RESET_STATUS__JMCIF_RESET_STATUS__SHIFT 0x18 +#define JPEG_SOFT_RESET_STATUS__JPEG0_DEC_RESET_STATUS_MASK 0x00000001L +#define JPEG_SOFT_RESET_STATUS__JPEG1_DEC_RESET_STATUS_MASK 0x00000002L +#define JPEG_SOFT_RESET_STATUS__JPEG2_DEC_RESET_STATUS_MASK 0x00000004L +#define JPEG_SOFT_RESET_STATUS__JPEG3_DEC_RESET_STATUS_MASK 0x00000008L +#define JPEG_SOFT_RESET_STATUS__JPEG4_DEC_RESET_STATUS_MASK 0x00000010L +#define JPEG_SOFT_RESET_STATUS__JPEG5_DEC_RESET_STATUS_MASK 0x00000020L +#define JPEG_SOFT_RESET_STATUS__JPEG6_DEC_RESET_STATUS_MASK 0x00000040L +#define JPEG_SOFT_RESET_STATUS__JPEG7_DEC_RESET_STATUS_MASK 0x00000080L +#define JPEG_SOFT_RESET_STATUS__DJRBC0_RESET_STATUS_MASK 0x00000100L +#define JPEG_SOFT_RESET_STATUS__DJRBC1_RESET_STATUS_MASK 0x00000200L +#define JPEG_SOFT_RESET_STATUS__DJRBC2_RESET_STATUS_MASK 0x00000400L +#define JPEG_SOFT_RESET_STATUS__DJRBC3_RESET_STATUS_MASK 0x00000800L +#define JPEG_SOFT_RESET_STATUS__DJRBC4_RESET_STATUS_MASK 0x00001000L +#define JPEG_SOFT_RESET_STATUS__DJRBC5_RESET_STATUS_MASK 0x00002000L +#define JPEG_SOFT_RESET_STATUS__DJRBC6_RESET_STATUS_MASK 0x00004000L +#define JPEG_SOFT_RESET_STATUS__DJRBC7_RESET_STATUS_MASK 0x00008000L +#define JPEG_SOFT_RESET_STATUS__JPEG_ENC_RESET_STATUS_MASK 0x00020000L +#define JPEG_SOFT_RESET_STATUS__EJRBC_RESET_STATUS_MASK 0x00040000L +#define JPEG_SOFT_RESET_STATUS__JMCIF_RESET_STATUS_MASK 0x01000000L +//JPEG_SYS_INT_EN +#define JPEG_SYS_INT_EN__DJPEG0_CORE__SHIFT 0x0 +#define JPEG_SYS_INT_EN__DJPEG1_CORE__SHIFT 0x1 +#define JPEG_SYS_INT_EN__DJPEG2_CORE__SHIFT 0x2 +#define JPEG_SYS_INT_EN__DJPEG3_CORE__SHIFT 0x3 +#define JPEG_SYS_INT_EN__DJPEG4_CORE__SHIFT 0x4 +#define JPEG_SYS_INT_EN__DJPEG5_CORE__SHIFT 0x5 +#define JPEG_SYS_INT_EN__DJPEG6_CORE__SHIFT 0x6 +#define JPEG_SYS_INT_EN__DJPEG7_CORE__SHIFT 0x7 +#define JPEG_SYS_INT_EN__DJRBC0__SHIFT 0x8 +#define JPEG_SYS_INT_EN__DJRBC1__SHIFT 0x9 +#define JPEG_SYS_INT_EN__DJRBC2__SHIFT 0xa +#define JPEG_SYS_INT_EN__DJRBC3__SHIFT 0xb +#define JPEG_SYS_INT_EN__DJRBC4__SHIFT 0xc +#define JPEG_SYS_INT_EN__DJRBC5__SHIFT 0xd +#define JPEG_SYS_INT_EN__DJRBC6__SHIFT 0xe +#define JPEG_SYS_INT_EN__DJRBC7__SHIFT 0xf +#define JPEG_SYS_INT_EN__DJPEG0_PF_RPT__SHIFT 0x10 +#define JPEG_SYS_INT_EN__DJPEG1_PF_RPT__SHIFT 0x11 +#define JPEG_SYS_INT_EN__DJPEG2_PF_RPT__SHIFT 0x12 +#define JPEG_SYS_INT_EN__DJPEG3_PF_RPT__SHIFT 0x13 +#define JPEG_SYS_INT_EN__DJPEG4_PF_RPT__SHIFT 0x14 +#define JPEG_SYS_INT_EN__DJPEG5_PF_RPT__SHIFT 0x15 +#define JPEG_SYS_INT_EN__DJPEG6_PF_RPT__SHIFT 0x16 +#define JPEG_SYS_INT_EN__DJPEG7_PF_RPT__SHIFT 0x17 +#define JPEG_SYS_INT_EN__DJPEG0_RAS_CNTL__SHIFT 0x18 +#define JPEG_SYS_INT_EN__DJPEG1_RAS_CNTL__SHIFT 0x19 +#define JPEG_SYS_INT_EN__DJPEG0_CORE_MASK 0x00000001L +#define JPEG_SYS_INT_EN__DJPEG1_CORE_MASK 0x00000002L +#define JPEG_SYS_INT_EN__DJPEG2_CORE_MASK 0x00000004L +#define JPEG_SYS_INT_EN__DJPEG3_CORE_MASK 0x00000008L +#define JPEG_SYS_INT_EN__DJPEG4_CORE_MASK 0x00000010L +#define JPEG_SYS_INT_EN__DJPEG5_CORE_MASK 0x00000020L +#define JPEG_SYS_INT_EN__DJPEG6_CORE_MASK 0x00000040L +#define JPEG_SYS_INT_EN__DJPEG7_CORE_MASK 0x00000080L +#define JPEG_SYS_INT_EN__DJRBC0_MASK 0x00000100L +#define JPEG_SYS_INT_EN__DJRBC1_MASK 0x00000200L +#define JPEG_SYS_INT_EN__DJRBC2_MASK 0x00000400L +#define JPEG_SYS_INT_EN__DJRBC3_MASK 0x00000800L +#define JPEG_SYS_INT_EN__DJRBC4_MASK 0x00001000L +#define JPEG_SYS_INT_EN__DJRBC5_MASK 0x00002000L +#define JPEG_SYS_INT_EN__DJRBC6_MASK 0x00004000L +#define JPEG_SYS_INT_EN__DJRBC7_MASK 0x00008000L +#define JPEG_SYS_INT_EN__DJPEG0_PF_RPT_MASK 0x00010000L +#define JPEG_SYS_INT_EN__DJPEG1_PF_RPT_MASK 0x00020000L +#define JPEG_SYS_INT_EN__DJPEG2_PF_RPT_MASK 0x00040000L +#define JPEG_SYS_INT_EN__DJPEG3_PF_RPT_MASK 0x00080000L +#define JPEG_SYS_INT_EN__DJPEG4_PF_RPT_MASK 0x00100000L +#define JPEG_SYS_INT_EN__DJPEG5_PF_RPT_MASK 0x00200000L +#define JPEG_SYS_INT_EN__DJPEG6_PF_RPT_MASK 0x00400000L +#define JPEG_SYS_INT_EN__DJPEG7_PF_RPT_MASK 0x00800000L +#define JPEG_SYS_INT_EN__DJPEG0_RAS_CNTL_MASK 0x01000000L +#define JPEG_SYS_INT_EN__DJPEG1_RAS_CNTL_MASK 0x02000000L +//JPEG_SYS_INT_EN1 +#define JPEG_SYS_INT_EN1__EJPEG_PF_RPT__SHIFT 0x0 +#define JPEG_SYS_INT_EN1__EJPEG_CORE__SHIFT 0x1 +#define JPEG_SYS_INT_EN1__EJRBC__SHIFT 0x2 +#define JPEG_SYS_INT_EN1__EJPEG_RAS_CNTL__SHIFT 0x3 +#define JPEG_SYS_INT_EN1__EJPEG_PF_RPT_MASK 0x00000001L +#define JPEG_SYS_INT_EN1__EJPEG_CORE_MASK 0x00000002L +#define JPEG_SYS_INT_EN1__EJRBC_MASK 0x00000004L +#define JPEG_SYS_INT_EN1__EJPEG_RAS_CNTL_MASK 0x00000008L +//JPEG_SYS_INT_STATUS +#define JPEG_SYS_INT_STATUS__DJPEG0_CORE__SHIFT 0x0 +#define JPEG_SYS_INT_STATUS__DJPEG1_CORE__SHIFT 0x1 +#define JPEG_SYS_INT_STATUS__DJPEG2_CORE__SHIFT 0x2 +#define JPEG_SYS_INT_STATUS__DJPEG3_CORE__SHIFT 0x3 +#define JPEG_SYS_INT_STATUS__DJPEG4_CORE__SHIFT 0x4 +#define JPEG_SYS_INT_STATUS__DJPEG5_CORE__SHIFT 0x5 +#define JPEG_SYS_INT_STATUS__DJPEG6_CORE__SHIFT 0x6 +#define JPEG_SYS_INT_STATUS__DJPEG7_CORE__SHIFT 0x7 +#define JPEG_SYS_INT_STATUS__DJRBC0__SHIFT 0x8 +#define JPEG_SYS_INT_STATUS__DJRBC1__SHIFT 0x9 +#define JPEG_SYS_INT_STATUS__DJRBC2__SHIFT 0xa +#define JPEG_SYS_INT_STATUS__DJRBC3__SHIFT 0xb +#define JPEG_SYS_INT_STATUS__DJRBC4__SHIFT 0xc +#define JPEG_SYS_INT_STATUS__DJRBC5__SHIFT 0xd +#define JPEG_SYS_INT_STATUS__DJRBC6__SHIFT 0xe +#define JPEG_SYS_INT_STATUS__DJRBC7__SHIFT 0xf +#define JPEG_SYS_INT_STATUS__DJPEG0_PF_RPT__SHIFT 0x10 +#define JPEG_SYS_INT_STATUS__DJPEG1_PF_RPT__SHIFT 0x11 +#define JPEG_SYS_INT_STATUS__DJPEG2_PF_RPT__SHIFT 0x12 +#define JPEG_SYS_INT_STATUS__DJPEG3_PF_RPT__SHIFT 0x13 +#define JPEG_SYS_INT_STATUS__DJPEG4_PF_RPT__SHIFT 0x14 +#define JPEG_SYS_INT_STATUS__DJPEG5_PF_RPT__SHIFT 0x15 +#define JPEG_SYS_INT_STATUS__DJPEG6_PF_RPT__SHIFT 0x16 +#define JPEG_SYS_INT_STATUS__DJPEG7_PF_RPT__SHIFT 0x17 +#define JPEG_SYS_INT_STATUS__DJPEG0_RAS_CNTL__SHIFT 0x18 +#define JPEG_SYS_INT_STATUS__DJPEG1_RAS_CNTL__SHIFT 0x19 +#define JPEG_SYS_INT_STATUS__DJPEG0_CORE_MASK 0x00000001L +#define JPEG_SYS_INT_STATUS__DJPEG1_CORE_MASK 0x00000002L +#define JPEG_SYS_INT_STATUS__DJPEG2_CORE_MASK 0x00000004L +#define JPEG_SYS_INT_STATUS__DJPEG3_CORE_MASK 0x00000008L +#define JPEG_SYS_INT_STATUS__DJPEG4_CORE_MASK 0x00000010L +#define JPEG_SYS_INT_STATUS__DJPEG5_CORE_MASK 0x00000020L +#define JPEG_SYS_INT_STATUS__DJPEG6_CORE_MASK 0x00000040L +#define JPEG_SYS_INT_STATUS__DJPEG7_CORE_MASK 0x00000080L +#define JPEG_SYS_INT_STATUS__DJRBC0_MASK 0x00000100L +#define JPEG_SYS_INT_STATUS__DJRBC1_MASK 0x00000200L +#define JPEG_SYS_INT_STATUS__DJRBC2_MASK 0x00000400L +#define JPEG_SYS_INT_STATUS__DJRBC3_MASK 0x00000800L +#define JPEG_SYS_INT_STATUS__DJRBC4_MASK 0x00001000L +#define JPEG_SYS_INT_STATUS__DJRBC5_MASK 0x00002000L +#define JPEG_SYS_INT_STATUS__DJRBC6_MASK 0x00004000L +#define JPEG_SYS_INT_STATUS__DJRBC7_MASK 0x00008000L +#define JPEG_SYS_INT_STATUS__DJPEG0_PF_RPT_MASK 0x00010000L +#define JPEG_SYS_INT_STATUS__DJPEG1_PF_RPT_MASK 0x00020000L +#define JPEG_SYS_INT_STATUS__DJPEG2_PF_RPT_MASK 0x00040000L +#define JPEG_SYS_INT_STATUS__DJPEG3_PF_RPT_MASK 0x00080000L +#define JPEG_SYS_INT_STATUS__DJPEG4_PF_RPT_MASK 0x00100000L +#define JPEG_SYS_INT_STATUS__DJPEG5_PF_RPT_MASK 0x00200000L +#define JPEG_SYS_INT_STATUS__DJPEG6_PF_RPT_MASK 0x00400000L +#define JPEG_SYS_INT_STATUS__DJPEG7_PF_RPT_MASK 0x00800000L +#define JPEG_SYS_INT_STATUS__DJPEG0_RAS_CNTL_MASK 0x01000000L +#define JPEG_SYS_INT_STATUS__DJPEG1_RAS_CNTL_MASK 0x02000000L +//JPEG_SYS_INT_STATUS1 +#define JPEG_SYS_INT_STATUS1__EJPEG_PF_RPT__SHIFT 0x0 +#define JPEG_SYS_INT_STATUS1__EJPEG_CORE__SHIFT 0x1 +#define JPEG_SYS_INT_STATUS1__EJRBC__SHIFT 0x2 +#define JPEG_SYS_INT_STATUS1__EJPEG_RAS_CNTL__SHIFT 0x3 +#define JPEG_SYS_INT_STATUS1__EJPEG_PF_RPT_MASK 0x00000001L +#define JPEG_SYS_INT_STATUS1__EJPEG_CORE_MASK 0x00000002L +#define JPEG_SYS_INT_STATUS1__EJRBC_MASK 0x00000004L +#define JPEG_SYS_INT_STATUS1__EJPEG_RAS_CNTL_MASK 0x00000008L +//JPEG_SYS_INT_ACK +#define JPEG_SYS_INT_ACK__DJPEG0_CORE__SHIFT 0x0 +#define JPEG_SYS_INT_ACK__DJPEG1_CORE__SHIFT 0x1 +#define JPEG_SYS_INT_ACK__DJPEG2_CORE__SHIFT 0x2 +#define JPEG_SYS_INT_ACK__DJPEG3_CORE__SHIFT 0x3 +#define JPEG_SYS_INT_ACK__DJPEG4_CORE__SHIFT 0x4 +#define JPEG_SYS_INT_ACK__DJPEG5_CORE__SHIFT 0x5 +#define JPEG_SYS_INT_ACK__DJPEG6_CORE__SHIFT 0x6 +#define JPEG_SYS_INT_ACK__DJPEG7_CORE__SHIFT 0x7 +#define JPEG_SYS_INT_ACK__DJRBC0__SHIFT 0x8 +#define JPEG_SYS_INT_ACK__DJRBC1__SHIFT 0x9 +#define JPEG_SYS_INT_ACK__DJRBC2__SHIFT 0xa +#define JPEG_SYS_INT_ACK__DJRBC3__SHIFT 0xb +#define JPEG_SYS_INT_ACK__DJRBC4__SHIFT 0xc +#define JPEG_SYS_INT_ACK__DJRBC5__SHIFT 0xd +#define JPEG_SYS_INT_ACK__DJRBC6__SHIFT 0xe +#define JPEG_SYS_INT_ACK__DJRBC7__SHIFT 0xf +#define JPEG_SYS_INT_ACK__DJPEG0_PF_RPT__SHIFT 0x10 +#define JPEG_SYS_INT_ACK__DJPEG1_PF_RPT__SHIFT 0x11 +#define JPEG_SYS_INT_ACK__DJPEG2_PF_RPT__SHIFT 0x12 +#define JPEG_SYS_INT_ACK__DJPEG3_PF_RPT__SHIFT 0x13 +#define JPEG_SYS_INT_ACK__DJPEG4_PF_RPT__SHIFT 0x14 +#define JPEG_SYS_INT_ACK__DJPEG5_PF_RPT__SHIFT 0x15 +#define JPEG_SYS_INT_ACK__DJPEG6_PF_RPT__SHIFT 0x16 +#define JPEG_SYS_INT_ACK__DJPEG7_PF_RPT__SHIFT 0x17 +#define JPEG_SYS_INT_ACK__DJPEG0_RAS_CNTL__SHIFT 0x18 +#define JPEG_SYS_INT_ACK__DJPEG1_RAS_CNTL__SHIFT 0x19 +#define JPEG_SYS_INT_ACK__DJPEG0_CORE_MASK 0x00000001L +#define JPEG_SYS_INT_ACK__DJPEG1_CORE_MASK 0x00000002L +#define JPEG_SYS_INT_ACK__DJPEG2_CORE_MASK 0x00000004L +#define JPEG_SYS_INT_ACK__DJPEG3_CORE_MASK 0x00000008L +#define JPEG_SYS_INT_ACK__DJPEG4_CORE_MASK 0x00000010L +#define JPEG_SYS_INT_ACK__DJPEG5_CORE_MASK 0x00000020L +#define JPEG_SYS_INT_ACK__DJPEG6_CORE_MASK 0x00000040L +#define JPEG_SYS_INT_ACK__DJPEG7_CORE_MASK 0x00000080L +#define JPEG_SYS_INT_ACK__DJRBC0_MASK 0x00000100L +#define JPEG_SYS_INT_ACK__DJRBC1_MASK 0x00000200L +#define JPEG_SYS_INT_ACK__DJRBC2_MASK 0x00000400L +#define JPEG_SYS_INT_ACK__DJRBC3_MASK 0x00000800L +#define JPEG_SYS_INT_ACK__DJRBC4_MASK 0x00001000L +#define JPEG_SYS_INT_ACK__DJRBC5_MASK 0x00002000L +#define JPEG_SYS_INT_ACK__DJRBC6_MASK 0x00004000L +#define JPEG_SYS_INT_ACK__DJRBC7_MASK 0x00008000L +#define JPEG_SYS_INT_ACK__DJPEG0_PF_RPT_MASK 0x00010000L +#define JPEG_SYS_INT_ACK__DJPEG1_PF_RPT_MASK 0x00020000L +#define JPEG_SYS_INT_ACK__DJPEG2_PF_RPT_MASK 0x00040000L +#define JPEG_SYS_INT_ACK__DJPEG3_PF_RPT_MASK 0x00080000L +#define JPEG_SYS_INT_ACK__DJPEG4_PF_RPT_MASK 0x00100000L +#define JPEG_SYS_INT_ACK__DJPEG5_PF_RPT_MASK 0x00200000L +#define JPEG_SYS_INT_ACK__DJPEG6_PF_RPT_MASK 0x00400000L +#define JPEG_SYS_INT_ACK__DJPEG7_PF_RPT_MASK 0x00800000L +#define JPEG_SYS_INT_ACK__DJPEG0_RAS_CNTL_MASK 0x01000000L +#define JPEG_SYS_INT_ACK__DJPEG1_RAS_CNTL_MASK 0x02000000L +//JPEG_SYS_INT_ACK1 +#define JPEG_SYS_INT_ACK1__EJPEG_PF_RPT__SHIFT 0x0 +#define JPEG_SYS_INT_ACK1__EJPEG_CORE__SHIFT 0x1 +#define JPEG_SYS_INT_ACK1__EJRBC__SHIFT 0x2 +#define JPEG_SYS_INT_ACK1__EJPEG_RAS_CNTL__SHIFT 0x3 +#define JPEG_SYS_INT_ACK1__EJPEG_PF_RPT_MASK 0x00000001L +#define JPEG_SYS_INT_ACK1__EJPEG_CORE_MASK 0x00000002L +#define JPEG_SYS_INT_ACK1__EJRBC_MASK 0x00000004L +#define JPEG_SYS_INT_ACK1__EJPEG_RAS_CNTL_MASK 0x00000008L +//JPEG_MEMCHECK_SYS_INT_EN +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC0_RD_ERR_EN__SHIFT 0x0 +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC1_RD_ERR_EN__SHIFT 0x1 +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC2_RD_ERR_EN__SHIFT 0x2 +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC3_RD_ERR_EN__SHIFT 0x3 +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC4_RD_ERR_EN__SHIFT 0x4 +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC5_RD_ERR_EN__SHIFT 0x5 +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC6_RD_ERR_EN__SHIFT 0x6 +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC7_RD_ERR_EN__SHIFT 0x7 +#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH0_RD_ERR_EN__SHIFT 0x8 +#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH1_RD_ERR_EN__SHIFT 0x9 +#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH2_RD_ERR_EN__SHIFT 0xa +#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH3_RD_ERR_EN__SHIFT 0xb +#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH4_RD_ERR_EN__SHIFT 0xc +#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH5_RD_ERR_EN__SHIFT 0xd +#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH6_RD_ERR_EN__SHIFT 0xe +#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH7_RD_ERR_EN__SHIFT 0xf +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC0_WR_ERR_EN__SHIFT 0x10 +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC1_WR_ERR_EN__SHIFT 0x11 +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC2_WR_ERR_EN__SHIFT 0x12 +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC3_WR_ERR_EN__SHIFT 0x13 +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC4_WR_ERR_EN__SHIFT 0x14 +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC5_WR_ERR_EN__SHIFT 0x15 +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC6_WR_ERR_EN__SHIFT 0x16 +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC7_WR_ERR_EN__SHIFT 0x17 +#define JPEG_MEMCHECK_SYS_INT_EN__OBUF0_WR_ERR_EN__SHIFT 0x18 +#define JPEG_MEMCHECK_SYS_INT_EN__OBUF1_WR_ERR_EN__SHIFT 0x19 +#define JPEG_MEMCHECK_SYS_INT_EN__OBUF2_WR_ERR_EN__SHIFT 0x1a +#define JPEG_MEMCHECK_SYS_INT_EN__OBUF3_WR_ERR_EN__SHIFT 0x1b +#define JPEG_MEMCHECK_SYS_INT_EN__OBUF4_WR_ERR_EN__SHIFT 0x1c +#define JPEG_MEMCHECK_SYS_INT_EN__OBUF5_WR_ERR_EN__SHIFT 0x1d +#define JPEG_MEMCHECK_SYS_INT_EN__OBUF6_WR_ERR_EN__SHIFT 0x1e +#define JPEG_MEMCHECK_SYS_INT_EN__OBUF7_WR_ERR_EN__SHIFT 0x1f +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC0_RD_ERR_EN_MASK 0x00000001L +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC1_RD_ERR_EN_MASK 0x00000002L +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC2_RD_ERR_EN_MASK 0x00000004L +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC3_RD_ERR_EN_MASK 0x00000008L +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC4_RD_ERR_EN_MASK 0x00000010L +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC5_RD_ERR_EN_MASK 0x00000020L +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC6_RD_ERR_EN_MASK 0x00000040L +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC7_RD_ERR_EN_MASK 0x00000080L +#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH0_RD_ERR_EN_MASK 0x00000100L +#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH1_RD_ERR_EN_MASK 0x00000200L +#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH2_RD_ERR_EN_MASK 0x00000400L +#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH3_RD_ERR_EN_MASK 0x00000800L +#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH4_RD_ERR_EN_MASK 0x00001000L +#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH5_RD_ERR_EN_MASK 0x00002000L +#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH6_RD_ERR_EN_MASK 0x00004000L +#define JPEG_MEMCHECK_SYS_INT_EN__BSFETCH7_RD_ERR_EN_MASK 0x00008000L +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC0_WR_ERR_EN_MASK 0x00010000L +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC1_WR_ERR_EN_MASK 0x00020000L +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC2_WR_ERR_EN_MASK 0x00040000L +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC3_WR_ERR_EN_MASK 0x00080000L +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC4_WR_ERR_EN_MASK 0x00100000L +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC5_WR_ERR_EN_MASK 0x00200000L +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC6_WR_ERR_EN_MASK 0x00400000L +#define JPEG_MEMCHECK_SYS_INT_EN__DJRBC7_WR_ERR_EN_MASK 0x00800000L +#define JPEG_MEMCHECK_SYS_INT_EN__OBUF0_WR_ERR_EN_MASK 0x01000000L +#define JPEG_MEMCHECK_SYS_INT_EN__OBUF1_WR_ERR_EN_MASK 0x02000000L +#define JPEG_MEMCHECK_SYS_INT_EN__OBUF2_WR_ERR_EN_MASK 0x04000000L +#define JPEG_MEMCHECK_SYS_INT_EN__OBUF3_WR_ERR_EN_MASK 0x08000000L +#define JPEG_MEMCHECK_SYS_INT_EN__OBUF4_WR_ERR_EN_MASK 0x10000000L +#define JPEG_MEMCHECK_SYS_INT_EN__OBUF5_WR_ERR_EN_MASK 0x20000000L +#define JPEG_MEMCHECK_SYS_INT_EN__OBUF6_WR_ERR_EN_MASK 0x40000000L +#define JPEG_MEMCHECK_SYS_INT_EN__OBUF7_WR_ERR_EN_MASK 0x80000000L +//JPEG_MEMCHECK_SYS_INT_EN1 +#define JPEG_MEMCHECK_SYS_INT_EN1__EJRBC_RD_ERR_EN__SHIFT 0x0 +#define JPEG_MEMCHECK_SYS_INT_EN1__PELFETCH_RD_ERR_EN__SHIFT 0x1 +#define JPEG_MEMCHECK_SYS_INT_EN1__SCALAR_RD_ERR_EN__SHIFT 0x2 +#define JPEG_MEMCHECK_SYS_INT_EN1__EJRBC_WR_ERR_EN__SHIFT 0x3 +#define JPEG_MEMCHECK_SYS_INT_EN1__BS_WR_ERR_EN__SHIFT 0x4 +#define JPEG_MEMCHECK_SYS_INT_EN1__SCALAR_WR_ERR_EN__SHIFT 0x5 +#define JPEG_MEMCHECK_SYS_INT_EN1__EJRBC_RD_ERR_EN_MASK 0x00000001L +#define JPEG_MEMCHECK_SYS_INT_EN1__PELFETCH_RD_ERR_EN_MASK 0x00000002L +#define JPEG_MEMCHECK_SYS_INT_EN1__SCALAR_RD_ERR_EN_MASK 0x00000004L +#define JPEG_MEMCHECK_SYS_INT_EN1__EJRBC_WR_ERR_EN_MASK 0x00000008L +#define JPEG_MEMCHECK_SYS_INT_EN1__BS_WR_ERR_EN_MASK 0x00000010L +#define JPEG_MEMCHECK_SYS_INT_EN1__SCALAR_WR_ERR_EN_MASK 0x00000020L +//JPEG_MEMCHECK_SYS_INT_STAT +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH0_RD_HI_ERR__SHIFT 0x0 +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH1_RD_HI_ERR__SHIFT 0x1 +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH2_RD_HI_ERR__SHIFT 0x2 +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH3_RD_HI_ERR__SHIFT 0x3 +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH4_RD_HI_ERR__SHIFT 0x4 +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH5_RD_HI_ERR__SHIFT 0x5 +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH6_RD_HI_ERR__SHIFT 0x6 +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH7_RD_HI_ERR__SHIFT 0x7 +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH0_RD_LO_ERR__SHIFT 0x8 +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH1_RD_LO_ERR__SHIFT 0x9 +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH2_RD_LO_ERR__SHIFT 0xa +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH3_RD_LO_ERR__SHIFT 0xb +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH4_RD_LO_ERR__SHIFT 0xc +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH5_RD_LO_ERR__SHIFT 0xd +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH6_RD_LO_ERR__SHIFT 0xe +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH7_RD_LO_ERR__SHIFT 0xf +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF0_WR_HI_ERR__SHIFT 0x10 +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF1_WR_HI_ERR__SHIFT 0x11 +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF2_WR_HI_ERR__SHIFT 0x12 +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF3_WR_HI_ERR__SHIFT 0x13 +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF4_WR_HI_ERR__SHIFT 0x14 +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF5_WR_HI_ERR__SHIFT 0x15 +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF6_WR_HI_ERR__SHIFT 0x16 +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF7_WR_HI_ERR__SHIFT 0x17 +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF0_WR_LO_ERR__SHIFT 0x18 +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF1_WR_LO_ERR__SHIFT 0x19 +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF2_WR_LO_ERR__SHIFT 0x1a +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF3_WR_LO_ERR__SHIFT 0x1b +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF4_WR_LO_ERR__SHIFT 0x1c +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF5_WR_LO_ERR__SHIFT 0x1d +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF6_WR_LO_ERR__SHIFT 0x1e +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF7_WR_LO_ERR__SHIFT 0x1f +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH0_RD_HI_ERR_MASK 0x00000001L +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH1_RD_HI_ERR_MASK 0x00000002L +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH2_RD_HI_ERR_MASK 0x00000004L +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH3_RD_HI_ERR_MASK 0x00000008L +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH4_RD_HI_ERR_MASK 0x00000010L +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH5_RD_HI_ERR_MASK 0x00000020L +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH6_RD_HI_ERR_MASK 0x00000040L +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH7_RD_HI_ERR_MASK 0x00000080L +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH0_RD_LO_ERR_MASK 0x00000100L +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH1_RD_LO_ERR_MASK 0x00000200L +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH2_RD_LO_ERR_MASK 0x00000400L +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH3_RD_LO_ERR_MASK 0x00000800L +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH4_RD_LO_ERR_MASK 0x00001000L +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH5_RD_LO_ERR_MASK 0x00002000L +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH6_RD_LO_ERR_MASK 0x00004000L +#define JPEG_MEMCHECK_SYS_INT_STAT__BSFETCH7_RD_LO_ERR_MASK 0x00008000L +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF0_WR_HI_ERR_MASK 0x00010000L +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF1_WR_HI_ERR_MASK 0x00020000L +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF2_WR_HI_ERR_MASK 0x00040000L +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF3_WR_HI_ERR_MASK 0x00080000L +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF4_WR_HI_ERR_MASK 0x00100000L +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF5_WR_HI_ERR_MASK 0x00200000L +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF6_WR_HI_ERR_MASK 0x00400000L +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF7_WR_HI_ERR_MASK 0x00800000L +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF0_WR_LO_ERR_MASK 0x01000000L +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF1_WR_LO_ERR_MASK 0x02000000L +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF2_WR_LO_ERR_MASK 0x04000000L +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF3_WR_LO_ERR_MASK 0x08000000L +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF4_WR_LO_ERR_MASK 0x10000000L +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF5_WR_LO_ERR_MASK 0x20000000L +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF6_WR_LO_ERR_MASK 0x40000000L +#define JPEG_MEMCHECK_SYS_INT_STAT__OBUF7_WR_LO_ERR_MASK 0x80000000L +//JPEG_MEMCHECK_SYS_INT_STAT1 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC0_RD_HI_ERR__SHIFT 0x0 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC1_RD_HI_ERR__SHIFT 0x1 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC2_RD_HI_ERR__SHIFT 0x2 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC3_RD_HI_ERR__SHIFT 0x3 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC4_RD_HI_ERR__SHIFT 0x4 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC5_RD_HI_ERR__SHIFT 0x5 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC6_RD_HI_ERR__SHIFT 0x6 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC7_RD_HI_ERR__SHIFT 0x7 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC0_RD_LO_ERR__SHIFT 0x8 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC1_RD_LO_ERR__SHIFT 0x9 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC2_RD_LO_ERR__SHIFT 0xa +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC3_RD_LO_ERR__SHIFT 0xb +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC4_RD_LO_ERR__SHIFT 0xc +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC5_RD_LO_ERR__SHIFT 0xd +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC6_RD_LO_ERR__SHIFT 0xe +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC7_RD_LO_ERR__SHIFT 0xf +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC0_WR_HI_ERR__SHIFT 0x10 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC1_WR_HI_ERR__SHIFT 0x11 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC2_WR_HI_ERR__SHIFT 0x12 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC3_WR_HI_ERR__SHIFT 0x13 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC4_WR_HI_ERR__SHIFT 0x14 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC5_WR_HI_ERR__SHIFT 0x15 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC6_WR_HI_ERR__SHIFT 0x16 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC7_WR_HI_ERR__SHIFT 0x17 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC0_WR_LO_ERR__SHIFT 0x18 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC1_WR_LO_ERR__SHIFT 0x19 +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC2_WR_LO_ERR__SHIFT 0x1a +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC3_WR_LO_ERR__SHIFT 0x1b +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC4_WR_LO_ERR__SHIFT 0x1c +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC5_WR_LO_ERR__SHIFT 0x1d +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC6_WR_LO_ERR__SHIFT 0x1e +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC7_WR_LO_ERR__SHIFT 0x1f +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC0_RD_HI_ERR_MASK 0x00000001L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC1_RD_HI_ERR_MASK 0x00000002L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC2_RD_HI_ERR_MASK 0x00000004L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC3_RD_HI_ERR_MASK 0x00000008L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC4_RD_HI_ERR_MASK 0x00000010L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC5_RD_HI_ERR_MASK 0x00000020L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC6_RD_HI_ERR_MASK 0x00000040L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC7_RD_HI_ERR_MASK 0x00000080L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC0_RD_LO_ERR_MASK 0x00000100L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC1_RD_LO_ERR_MASK 0x00000200L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC2_RD_LO_ERR_MASK 0x00000400L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC3_RD_LO_ERR_MASK 0x00000800L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC4_RD_LO_ERR_MASK 0x00001000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC5_RD_LO_ERR_MASK 0x00002000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC6_RD_LO_ERR_MASK 0x00004000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC7_RD_LO_ERR_MASK 0x00008000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC0_WR_HI_ERR_MASK 0x00010000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC1_WR_HI_ERR_MASK 0x00020000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC2_WR_HI_ERR_MASK 0x00040000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC3_WR_HI_ERR_MASK 0x00080000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC4_WR_HI_ERR_MASK 0x00100000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC5_WR_HI_ERR_MASK 0x00200000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC6_WR_HI_ERR_MASK 0x00400000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC7_WR_HI_ERR_MASK 0x00800000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC0_WR_LO_ERR_MASK 0x01000000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC1_WR_LO_ERR_MASK 0x02000000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC2_WR_LO_ERR_MASK 0x04000000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC3_WR_LO_ERR_MASK 0x08000000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC4_WR_LO_ERR_MASK 0x10000000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC5_WR_LO_ERR_MASK 0x20000000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC6_WR_LO_ERR_MASK 0x40000000L +#define JPEG_MEMCHECK_SYS_INT_STAT1__DJRBC7_WR_LO_ERR_MASK 0x80000000L +//JPEG_MEMCHECK_SYS_INT_STAT2 +#define JPEG_MEMCHECK_SYS_INT_STAT2__EJRBC_RD_HI_ERR__SHIFT 0x0 +#define JPEG_MEMCHECK_SYS_INT_STAT2__EJRBC_RD_LO_ERR__SHIFT 0x1 +#define JPEG_MEMCHECK_SYS_INT_STAT2__PELFETCH_RD_HI_ERR__SHIFT 0x2 +#define JPEG_MEMCHECK_SYS_INT_STAT2__PELFETCH_RD_LO_ERR__SHIFT 0x3 +#define JPEG_MEMCHECK_SYS_INT_STAT2__SCALAR_RD_HI_ERR__SHIFT 0x4 +#define JPEG_MEMCHECK_SYS_INT_STAT2__SCALAR_RD_LO_ERR__SHIFT 0x5 +#define JPEG_MEMCHECK_SYS_INT_STAT2__EJRBC_WR_HI_ERR__SHIFT 0x6 +#define JPEG_MEMCHECK_SYS_INT_STAT2__EJRBC_WR_LO_ERR__SHIFT 0x7 +#define JPEG_MEMCHECK_SYS_INT_STAT2__BS_WR_HI_ERR__SHIFT 0x8 +#define JPEG_MEMCHECK_SYS_INT_STAT2__BS_WR_LO_ERR__SHIFT 0x9 +#define JPEG_MEMCHECK_SYS_INT_STAT2__SCALAR_WR_HI_ERR__SHIFT 0xa +#define JPEG_MEMCHECK_SYS_INT_STAT2__SCALAR_WR_LO_ERR__SHIFT 0xb +#define JPEG_MEMCHECK_SYS_INT_STAT2__EJRBC_RD_HI_ERR_MASK 0x00000001L +#define JPEG_MEMCHECK_SYS_INT_STAT2__EJRBC_RD_LO_ERR_MASK 0x00000002L +#define JPEG_MEMCHECK_SYS_INT_STAT2__PELFETCH_RD_HI_ERR_MASK 0x00000004L +#define JPEG_MEMCHECK_SYS_INT_STAT2__PELFETCH_RD_LO_ERR_MASK 0x00000008L +#define JPEG_MEMCHECK_SYS_INT_STAT2__SCALAR_RD_HI_ERR_MASK 0x00000010L +#define JPEG_MEMCHECK_SYS_INT_STAT2__SCALAR_RD_LO_ERR_MASK 0x00000020L +#define JPEG_MEMCHECK_SYS_INT_STAT2__EJRBC_WR_HI_ERR_MASK 0x00000040L +#define JPEG_MEMCHECK_SYS_INT_STAT2__EJRBC_WR_LO_ERR_MASK 0x00000080L +#define JPEG_MEMCHECK_SYS_INT_STAT2__BS_WR_HI_ERR_MASK 0x00000100L +#define JPEG_MEMCHECK_SYS_INT_STAT2__BS_WR_LO_ERR_MASK 0x00000200L +#define JPEG_MEMCHECK_SYS_INT_STAT2__SCALAR_WR_HI_ERR_MASK 0x00000400L +#define JPEG_MEMCHECK_SYS_INT_STAT2__SCALAR_WR_LO_ERR_MASK 0x00000800L +//JPEG_MEMCHECK_SYS_INT_ACK +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH0_RD_HI_ERR__SHIFT 0x0 +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH1_RD_HI_ERR__SHIFT 0x1 +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH2_RD_HI_ERR__SHIFT 0x2 +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH3_RD_HI_ERR__SHIFT 0x3 +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH4_RD_HI_ERR__SHIFT 0x4 +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH5_RD_HI_ERR__SHIFT 0x5 +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH6_RD_HI_ERR__SHIFT 0x6 +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH7_RD_HI_ERR__SHIFT 0x7 +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH0_RD_LO_ERR__SHIFT 0x8 +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH1_RD_LO_ERR__SHIFT 0x9 +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH2_RD_LO_ERR__SHIFT 0xa +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH3_RD_LO_ERR__SHIFT 0xb +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH4_RD_LO_ERR__SHIFT 0xc +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH5_RD_LO_ERR__SHIFT 0xd +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH6_RD_LO_ERR__SHIFT 0xe +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH7_RD_LO_ERR__SHIFT 0xf +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF0_WR_HI_ERR__SHIFT 0x10 +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF1_WR_HI_ERR__SHIFT 0x11 +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF2_WR_HI_ERR__SHIFT 0x12 +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF3_WR_HI_ERR__SHIFT 0x13 +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF4_WR_HI_ERR__SHIFT 0x14 +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF5_WR_HI_ERR__SHIFT 0x15 +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF6_WR_HI_ERR__SHIFT 0x16 +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF7_WR_HI_ERR__SHIFT 0x17 +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF0_WR_LO_ERR__SHIFT 0x18 +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF1_WR_LO_ERR__SHIFT 0x19 +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF2_WR_LO_ERR__SHIFT 0x1a +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF3_WR_LO_ERR__SHIFT 0x1b +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF4_WR_LO_ERR__SHIFT 0x1c +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF5_WR_LO_ERR__SHIFT 0x1d +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF6_WR_LO_ERR__SHIFT 0x1e +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF7_WR_LO_ERR__SHIFT 0x1f +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH0_RD_HI_ERR_MASK 0x00000001L +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH1_RD_HI_ERR_MASK 0x00000002L +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH2_RD_HI_ERR_MASK 0x00000004L +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH3_RD_HI_ERR_MASK 0x00000008L +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH4_RD_HI_ERR_MASK 0x00000010L +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH5_RD_HI_ERR_MASK 0x00000020L +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH6_RD_HI_ERR_MASK 0x00000040L +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH7_RD_HI_ERR_MASK 0x00000080L +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH0_RD_LO_ERR_MASK 0x00000100L +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH1_RD_LO_ERR_MASK 0x00000200L +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH2_RD_LO_ERR_MASK 0x00000400L +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH3_RD_LO_ERR_MASK 0x00000800L +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH4_RD_LO_ERR_MASK 0x00001000L +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH5_RD_LO_ERR_MASK 0x00002000L +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH6_RD_LO_ERR_MASK 0x00004000L +#define JPEG_MEMCHECK_SYS_INT_ACK__BSFETCH7_RD_LO_ERR_MASK 0x00008000L +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF0_WR_HI_ERR_MASK 0x00010000L +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF1_WR_HI_ERR_MASK 0x00020000L +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF2_WR_HI_ERR_MASK 0x00040000L +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF3_WR_HI_ERR_MASK 0x00080000L +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF4_WR_HI_ERR_MASK 0x00100000L +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF5_WR_HI_ERR_MASK 0x00200000L +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF6_WR_HI_ERR_MASK 0x00400000L +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF7_WR_HI_ERR_MASK 0x00800000L +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF0_WR_LO_ERR_MASK 0x01000000L +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF1_WR_LO_ERR_MASK 0x02000000L +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF2_WR_LO_ERR_MASK 0x04000000L +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF3_WR_LO_ERR_MASK 0x08000000L +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF4_WR_LO_ERR_MASK 0x10000000L +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF5_WR_LO_ERR_MASK 0x20000000L +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF6_WR_LO_ERR_MASK 0x40000000L +#define JPEG_MEMCHECK_SYS_INT_ACK__OBUF7_WR_LO_ERR_MASK 0x80000000L +//JPEG_MEMCHECK_SYS_INT_ACK1 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC0_RD_HI_ERR__SHIFT 0x0 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC1_RD_HI_ERR__SHIFT 0x1 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC2_RD_HI_ERR__SHIFT 0x2 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC3_RD_HI_ERR__SHIFT 0x3 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC4_RD_HI_ERR__SHIFT 0x4 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC5_RD_HI_ERR__SHIFT 0x5 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC6_RD_HI_ERR__SHIFT 0x6 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC7_RD_HI_ERR__SHIFT 0x7 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC0_RD_LO_ERR__SHIFT 0x8 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC1_RD_LO_ERR__SHIFT 0x9 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC2_RD_LO_ERR__SHIFT 0xa +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC3_RD_LO_ERR__SHIFT 0xb +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC4_RD_LO_ERR__SHIFT 0xc +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC5_RD_LO_ERR__SHIFT 0xd +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC6_RD_LO_ERR__SHIFT 0xe +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC7_RD_LO_ERR__SHIFT 0xf +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC0_WR_HI_ERR__SHIFT 0x10 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC1_WR_HI_ERR__SHIFT 0x11 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC2_WR_HI_ERR__SHIFT 0x12 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC3_WR_HI_ERR__SHIFT 0x13 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC4_WR_HI_ERR__SHIFT 0x14 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC5_WR_HI_ERR__SHIFT 0x15 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC6_WR_HI_ERR__SHIFT 0x16 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC7_WR_HI_ERR__SHIFT 0x17 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC0_WR_LO_ERR__SHIFT 0x18 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC1_WR_LO_ERR__SHIFT 0x19 +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC2_WR_LO_ERR__SHIFT 0x1a +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC3_WR_LO_ERR__SHIFT 0x1b +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC4_WR_LO_ERR__SHIFT 0x1c +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC5_WR_LO_ERR__SHIFT 0x1d +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC6_WR_LO_ERR__SHIFT 0x1e +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC7_WR_LO_ERR__SHIFT 0x1f +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC0_RD_HI_ERR_MASK 0x00000001L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC1_RD_HI_ERR_MASK 0x00000002L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC2_RD_HI_ERR_MASK 0x00000004L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC3_RD_HI_ERR_MASK 0x00000008L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC4_RD_HI_ERR_MASK 0x00000010L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC5_RD_HI_ERR_MASK 0x00000020L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC6_RD_HI_ERR_MASK 0x00000040L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC7_RD_HI_ERR_MASK 0x00000080L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC0_RD_LO_ERR_MASK 0x00000100L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC1_RD_LO_ERR_MASK 0x00000200L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC2_RD_LO_ERR_MASK 0x00000400L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC3_RD_LO_ERR_MASK 0x00000800L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC4_RD_LO_ERR_MASK 0x00001000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC5_RD_LO_ERR_MASK 0x00002000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC6_RD_LO_ERR_MASK 0x00004000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC7_RD_LO_ERR_MASK 0x00008000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC0_WR_HI_ERR_MASK 0x00010000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC1_WR_HI_ERR_MASK 0x00020000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC2_WR_HI_ERR_MASK 0x00040000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC3_WR_HI_ERR_MASK 0x00080000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC4_WR_HI_ERR_MASK 0x00100000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC5_WR_HI_ERR_MASK 0x00200000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC6_WR_HI_ERR_MASK 0x00400000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC7_WR_HI_ERR_MASK 0x00800000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC0_WR_LO_ERR_MASK 0x01000000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC1_WR_LO_ERR_MASK 0x02000000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC2_WR_LO_ERR_MASK 0x04000000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC3_WR_LO_ERR_MASK 0x08000000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC4_WR_LO_ERR_MASK 0x10000000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC5_WR_LO_ERR_MASK 0x20000000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC6_WR_LO_ERR_MASK 0x40000000L +#define JPEG_MEMCHECK_SYS_INT_ACK1__DJRBC7_WR_LO_ERR_MASK 0x80000000L +//JPEG_MEMCHECK_SYS_INT_ACK2 +#define JPEG_MEMCHECK_SYS_INT_ACK2__EJRBC_RD_HI_ERR__SHIFT 0x0 +#define JPEG_MEMCHECK_SYS_INT_ACK2__EJRBC_RD_LO_ERR__SHIFT 0x1 +#define JPEG_MEMCHECK_SYS_INT_ACK2__PELFETCH_RD_HI_ERR__SHIFT 0x2 +#define JPEG_MEMCHECK_SYS_INT_ACK2__PELFETCH_RD_LO_ERR__SHIFT 0x3 +#define JPEG_MEMCHECK_SYS_INT_ACK2__SCALAR_RD_HI_ERR__SHIFT 0x4 +#define JPEG_MEMCHECK_SYS_INT_ACK2__SCALAR_RD_LO_ERR__SHIFT 0x5 +#define JPEG_MEMCHECK_SYS_INT_ACK2__EJRBC_WR_HI_ERR__SHIFT 0x6 +#define JPEG_MEMCHECK_SYS_INT_ACK2__EJRBC_WR_LO_ERR__SHIFT 0x7 +#define JPEG_MEMCHECK_SYS_INT_ACK2__BS_WR_HI_ERR__SHIFT 0x8 +#define JPEG_MEMCHECK_SYS_INT_ACK2__BS_WR_LO_ERR__SHIFT 0x9 +#define JPEG_MEMCHECK_SYS_INT_ACK2__SCALAR_WR_HI_ERR__SHIFT 0xa +#define JPEG_MEMCHECK_SYS_INT_ACK2__SCALAR_WR_LO_ERR__SHIFT 0xb +#define JPEG_MEMCHECK_SYS_INT_ACK2__EJRBC_RD_HI_ERR_MASK 0x00000001L +#define JPEG_MEMCHECK_SYS_INT_ACK2__EJRBC_RD_LO_ERR_MASK 0x00000002L +#define JPEG_MEMCHECK_SYS_INT_ACK2__PELFETCH_RD_HI_ERR_MASK 0x00000004L +#define JPEG_MEMCHECK_SYS_INT_ACK2__PELFETCH_RD_LO_ERR_MASK 0x00000008L +#define JPEG_MEMCHECK_SYS_INT_ACK2__SCALAR_RD_HI_ERR_MASK 0x00000010L +#define JPEG_MEMCHECK_SYS_INT_ACK2__SCALAR_RD_LO_ERR_MASK 0x00000020L +#define JPEG_MEMCHECK_SYS_INT_ACK2__EJRBC_WR_HI_ERR_MASK 0x00000040L +#define JPEG_MEMCHECK_SYS_INT_ACK2__EJRBC_WR_LO_ERR_MASK 0x00000080L +#define JPEG_MEMCHECK_SYS_INT_ACK2__BS_WR_HI_ERR_MASK 0x00000100L +#define JPEG_MEMCHECK_SYS_INT_ACK2__BS_WR_LO_ERR_MASK 0x00000200L +#define JPEG_MEMCHECK_SYS_INT_ACK2__SCALAR_WR_HI_ERR_MASK 0x00000400L +#define JPEG_MEMCHECK_SYS_INT_ACK2__SCALAR_WR_LO_ERR_MASK 0x00000800L +//JPEG_MASTINT_EN +#define JPEG_MASTINT_EN__OVERRUN_RST__SHIFT 0x0 +#define JPEG_MASTINT_EN__INT_OVERRUN__SHIFT 0x4 +#define JPEG_MASTINT_EN__OVERRUN_RST_MASK 0x00000001L +#define JPEG_MASTINT_EN__INT_OVERRUN_MASK 0x007FFFF0L +//JPEG_IH_CTRL +#define JPEG_IH_CTRL__IH_SOFT_RESET__SHIFT 0x0 +#define JPEG_IH_CTRL__IH_STALL_EN__SHIFT 0x1 +#define JPEG_IH_CTRL__IH_STATUS_CLEAN__SHIFT 0x2 +#define JPEG_IH_CTRL__IH_VMID__SHIFT 0x3 +#define JPEG_IH_CTRL__IH_USER_DATA__SHIFT 0x7 +#define JPEG_IH_CTRL__IH_RINGID__SHIFT 0x13 +#define JPEG_IH_CTRL__IH_SOFT_RESET_MASK 0x00000001L +#define JPEG_IH_CTRL__IH_STALL_EN_MASK 0x00000002L +#define JPEG_IH_CTRL__IH_STATUS_CLEAN_MASK 0x00000004L +#define JPEG_IH_CTRL__IH_VMID_MASK 0x00000078L +#define JPEG_IH_CTRL__IH_USER_DATA_MASK 0x0007FF80L +#define JPEG_IH_CTRL__IH_RINGID_MASK 0x07F80000L +//JRBBM_ARB_CTRL +#define JRBBM_ARB_CTRL__DJRBC0_DROP__SHIFT 0x0 +#define JRBBM_ARB_CTRL__DJRBC1_DROP__SHIFT 0x1 +#define JRBBM_ARB_CTRL__DJRBC2_DROP__SHIFT 0x2 +#define JRBBM_ARB_CTRL__DJRBC3_DROP__SHIFT 0x3 +#define JRBBM_ARB_CTRL__DJRBC4_DROP__SHIFT 0x4 +#define JRBBM_ARB_CTRL__DJRBC5_DROP__SHIFT 0x5 +#define JRBBM_ARB_CTRL__DJRBC6_DROP__SHIFT 0x6 +#define JRBBM_ARB_CTRL__DJRBC7_DROP__SHIFT 0x7 +#define JRBBM_ARB_CTRL__EJRBC_DROP__SHIFT 0x8 +#define JRBBM_ARB_CTRL__SRBM_DROP__SHIFT 0x9 +#define JRBBM_ARB_CTRL__DJRBC0_DROP_MASK 0x00000001L +#define JRBBM_ARB_CTRL__DJRBC1_DROP_MASK 0x00000002L +#define JRBBM_ARB_CTRL__DJRBC2_DROP_MASK 0x00000004L +#define JRBBM_ARB_CTRL__DJRBC3_DROP_MASK 0x00000008L +#define JRBBM_ARB_CTRL__DJRBC4_DROP_MASK 0x00000010L +#define JRBBM_ARB_CTRL__DJRBC5_DROP_MASK 0x00000020L +#define JRBBM_ARB_CTRL__DJRBC6_DROP_MASK 0x00000040L +#define JRBBM_ARB_CTRL__DJRBC7_DROP_MASK 0x00000080L +#define JRBBM_ARB_CTRL__EJRBC_DROP_MASK 0x00000100L +#define JRBBM_ARB_CTRL__SRBM_DROP_MASK 0x00000200L + + +// addressBlock: aid_uvd0_uvd_jpeg_common_sclk_dec +//JPEG_CGC_GATE +#define JPEG_CGC_GATE__JPEG0_DEC__SHIFT 0x0 +#define JPEG_CGC_GATE__JPEG1_DEC__SHIFT 0x1 +#define JPEG_CGC_GATE__JPEG2_DEC__SHIFT 0x2 +#define JPEG_CGC_GATE__JPEG3_DEC__SHIFT 0x3 +#define JPEG_CGC_GATE__JPEG4_DEC__SHIFT 0x4 +#define JPEG_CGC_GATE__JPEG5_DEC__SHIFT 0x5 +#define JPEG_CGC_GATE__JPEG6_DEC__SHIFT 0x6 +#define JPEG_CGC_GATE__JPEG7_DEC__SHIFT 0x7 +#define JPEG_CGC_GATE__JPEG_ENC__SHIFT 0x8 +#define JPEG_CGC_GATE__JMCIF__SHIFT 0x9 +#define JPEG_CGC_GATE__JRBBM__SHIFT 0xa +#define JPEG_CGC_GATE__JPEG0_DEC_MASK 0x00000001L +#define JPEG_CGC_GATE__JPEG1_DEC_MASK 0x00000002L +#define JPEG_CGC_GATE__JPEG2_DEC_MASK 0x00000004L +#define JPEG_CGC_GATE__JPEG3_DEC_MASK 0x00000008L +#define JPEG_CGC_GATE__JPEG4_DEC_MASK 0x00000010L +#define JPEG_CGC_GATE__JPEG5_DEC_MASK 0x00000020L +#define JPEG_CGC_GATE__JPEG6_DEC_MASK 0x00000040L +#define JPEG_CGC_GATE__JPEG7_DEC_MASK 0x00000080L +#define JPEG_CGC_GATE__JPEG_ENC_MASK 0x00000100L +#define JPEG_CGC_GATE__JMCIF_MASK 0x00000200L +#define JPEG_CGC_GATE__JRBBM_MASK 0x00000400L +//JPEG_CGC_CTRL +#define JPEG_CGC_CTRL__DYN_CLOCK_MODE__SHIFT 0x0 +#define JPEG_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT 0x1 +#define JPEG_CGC_CTRL__CLK_OFF_DELAY__SHIFT 0x5 +#define JPEG_CGC_CTRL__JPEG0_DEC_MODE__SHIFT 0x10 +#define JPEG_CGC_CTRL__JPEG1_DEC_MODE__SHIFT 0x11 +#define JPEG_CGC_CTRL__JPEG2_DEC_MODE__SHIFT 0x12 +#define JPEG_CGC_CTRL__JPEG3_DEC_MODE__SHIFT 0x13 +#define JPEG_CGC_CTRL__JPEG4_DEC_MODE__SHIFT 0x14 +#define JPEG_CGC_CTRL__JPEG5_DEC_MODE__SHIFT 0x15 +#define JPEG_CGC_CTRL__JPEG6_DEC_MODE__SHIFT 0x16 +#define JPEG_CGC_CTRL__JPEG7_DEC_MODE__SHIFT 0x17 +#define JPEG_CGC_CTRL__JPEG_ENC_MODE__SHIFT 0x18 +#define JPEG_CGC_CTRL__JMCIF_MODE__SHIFT 0x19 +#define JPEG_CGC_CTRL__JRBBM_MODE__SHIFT 0x1a +#define JPEG_CGC_CTRL__DYN_CLOCK_MODE_MASK 0x00000001L +#define JPEG_CGC_CTRL__CLK_GATE_DLY_TIMER_MASK 0x0000001EL +#define JPEG_CGC_CTRL__CLK_OFF_DELAY_MASK 0x00001FE0L +#define JPEG_CGC_CTRL__JPEG0_DEC_MODE_MASK 0x00010000L +#define JPEG_CGC_CTRL__JPEG1_DEC_MODE_MASK 0x00020000L +#define JPEG_CGC_CTRL__JPEG2_DEC_MODE_MASK 0x00040000L +#define JPEG_CGC_CTRL__JPEG3_DEC_MODE_MASK 0x00080000L +#define JPEG_CGC_CTRL__JPEG4_DEC_MODE_MASK 0x00100000L +#define JPEG_CGC_CTRL__JPEG5_DEC_MODE_MASK 0x00200000L +#define JPEG_CGC_CTRL__JPEG6_DEC_MODE_MASK 0x00400000L +#define JPEG_CGC_CTRL__JPEG7_DEC_MODE_MASK 0x00800000L +#define JPEG_CGC_CTRL__JPEG_ENC_MODE_MASK 0x01000000L +#define JPEG_CGC_CTRL__JMCIF_MODE_MASK 0x02000000L +#define JPEG_CGC_CTRL__JRBBM_MODE_MASK 0x04000000L +//JPEG_CGC_STATUS +#define JPEG_CGC_STATUS__JPEG0_DEC_VCLK_ACTIVE__SHIFT 0x0 +#define JPEG_CGC_STATUS__JPEG0_DEC_SCLK_ACTIVE__SHIFT 0x1 +#define JPEG_CGC_STATUS__JPEG1_DEC_VCLK_ACTIVE__SHIFT 0x2 +#define JPEG_CGC_STATUS__JPEG1_DEC_SCLK_ACTIVE__SHIFT 0x3 +#define JPEG_CGC_STATUS__JPEG2_DEC_VCLK_ACTIVE__SHIFT 0x4 +#define JPEG_CGC_STATUS__JPEG2_DEC_SCLK_ACTIVE__SHIFT 0x5 +#define JPEG_CGC_STATUS__JPEG3_DEC_VCLK_ACTIVE__SHIFT 0x6 +#define JPEG_CGC_STATUS__JPEG3_DEC_SCLK_ACTIVE__SHIFT 0x7 +#define JPEG_CGC_STATUS__JPEG4_DEC_VCLK_ACTIVE__SHIFT 0x8 +#define JPEG_CGC_STATUS__JPEG4_DEC_SCLK_ACTIVE__SHIFT 0x9 +#define JPEG_CGC_STATUS__JPEG5_DEC_VCLK_ACTIVE__SHIFT 0xa +#define JPEG_CGC_STATUS__JPEG5_DEC_SCLK_ACTIVE__SHIFT 0xb +#define JPEG_CGC_STATUS__JPEG6_DEC_VCLK_ACTIVE__SHIFT 0xc +#define JPEG_CGC_STATUS__JPEG6_DEC_SCLK_ACTIVE__SHIFT 0xd +#define JPEG_CGC_STATUS__JPEG7_DEC_VCLK_ACTIVE__SHIFT 0xe +#define JPEG_CGC_STATUS__JPEG7_DEC_SCLK_ACTIVE__SHIFT 0xf +#define JPEG_CGC_STATUS__JPEG_ENC_VCLK_ACTIVE__SHIFT 0x10 +#define JPEG_CGC_STATUS__JPEG_ENC_SCLK_ACTIVE__SHIFT 0x11 +#define JPEG_CGC_STATUS__JMCIF_SCLK_ACTIVE__SHIFT 0x12 +#define JPEG_CGC_STATUS__JRBBM_VCLK_ACTIVE__SHIFT 0x13 +#define JPEG_CGC_STATUS__JRBBM_SCLK_ACTIVE__SHIFT 0x14 +#define JPEG_CGC_STATUS__JPEG0_DEC_VCLK_ACTIVE_MASK 0x00000001L +#define JPEG_CGC_STATUS__JPEG0_DEC_SCLK_ACTIVE_MASK 0x00000002L +#define JPEG_CGC_STATUS__JPEG1_DEC_VCLK_ACTIVE_MASK 0x00000004L +#define JPEG_CGC_STATUS__JPEG1_DEC_SCLK_ACTIVE_MASK 0x00000008L +#define JPEG_CGC_STATUS__JPEG2_DEC_VCLK_ACTIVE_MASK 0x00000010L +#define JPEG_CGC_STATUS__JPEG2_DEC_SCLK_ACTIVE_MASK 0x00000020L +#define JPEG_CGC_STATUS__JPEG3_DEC_VCLK_ACTIVE_MASK 0x00000040L +#define JPEG_CGC_STATUS__JPEG3_DEC_SCLK_ACTIVE_MASK 0x00000080L +#define JPEG_CGC_STATUS__JPEG4_DEC_VCLK_ACTIVE_MASK 0x00000100L +#define JPEG_CGC_STATUS__JPEG4_DEC_SCLK_ACTIVE_MASK 0x00000200L +#define JPEG_CGC_STATUS__JPEG5_DEC_VCLK_ACTIVE_MASK 0x00000400L +#define JPEG_CGC_STATUS__JPEG5_DEC_SCLK_ACTIVE_MASK 0x00000800L +#define JPEG_CGC_STATUS__JPEG6_DEC_VCLK_ACTIVE_MASK 0x00001000L +#define JPEG_CGC_STATUS__JPEG6_DEC_SCLK_ACTIVE_MASK 0x00002000L +#define JPEG_CGC_STATUS__JPEG7_DEC_VCLK_ACTIVE_MASK 0x00004000L +#define JPEG_CGC_STATUS__JPEG7_DEC_SCLK_ACTIVE_MASK 0x00008000L +#define JPEG_CGC_STATUS__JPEG_ENC_VCLK_ACTIVE_MASK 0x00010000L +#define JPEG_CGC_STATUS__JPEG_ENC_SCLK_ACTIVE_MASK 0x00020000L +#define JPEG_CGC_STATUS__JMCIF_SCLK_ACTIVE_MASK 0x00040000L +#define JPEG_CGC_STATUS__JRBBM_VCLK_ACTIVE_MASK 0x00080000L +#define JPEG_CGC_STATUS__JRBBM_SCLK_ACTIVE_MASK 0x00100000L +//JPEG_COMN_CGC_MEM_CTRL +#define JPEG_COMN_CGC_MEM_CTRL__JMCIF_LS_EN__SHIFT 0x0 +#define JPEG_COMN_CGC_MEM_CTRL__JMCIF_DS_EN__SHIFT 0x1 +#define JPEG_COMN_CGC_MEM_CTRL__JMCIF_SD_EN__SHIFT 0x2 +#define JPEG_COMN_CGC_MEM_CTRL__JMCIF_LS_SW_EN__SHIFT 0x3 +#define JPEG_COMN_CGC_MEM_CTRL__JMCIF_LS_EN_MASK 0x00000001L +#define JPEG_COMN_CGC_MEM_CTRL__JMCIF_DS_EN_MASK 0x00000002L +#define JPEG_COMN_CGC_MEM_CTRL__JMCIF_SD_EN_MASK 0x00000004L +#define JPEG_COMN_CGC_MEM_CTRL__JMCIF_LS_SW_EN_MASK 0x00000008L +//JPEG_DEC_CGC_MEM_CTRL +#define JPEG_DEC_CGC_MEM_CTRL__JPEG0_DEC_LS_EN__SHIFT 0x0 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG0_DEC_DS_EN__SHIFT 0x1 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG0_DEC_SD_EN__SHIFT 0x2 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG0_DEC_LS_SW_EN__SHIFT 0x3 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG1_DEC_LS_EN__SHIFT 0x4 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG1_DEC_DS_EN__SHIFT 0x5 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG1_DEC_SD_EN__SHIFT 0x6 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG1_DEC_LS_SW_EN__SHIFT 0x7 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG2_DEC_LS_EN__SHIFT 0x8 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG2_DEC_DS_EN__SHIFT 0x9 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG2_DEC_SD_EN__SHIFT 0xa +#define JPEG_DEC_CGC_MEM_CTRL__JPEG2_DEC_LS_SW_EN__SHIFT 0xb +#define JPEG_DEC_CGC_MEM_CTRL__JPEG3_DEC_LS_EN__SHIFT 0xc +#define JPEG_DEC_CGC_MEM_CTRL__JPEG3_DEC_DS_EN__SHIFT 0xd +#define JPEG_DEC_CGC_MEM_CTRL__JPEG3_DEC_SD_EN__SHIFT 0xe +#define JPEG_DEC_CGC_MEM_CTRL__JPEG3_DEC_LS_SW_EN__SHIFT 0xf +#define JPEG_DEC_CGC_MEM_CTRL__JPEG4_DEC_LS_EN__SHIFT 0x10 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG4_DEC_DS_EN__SHIFT 0x11 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG4_DEC_SD_EN__SHIFT 0x12 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG4_DEC_LS_SW_EN__SHIFT 0x13 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG5_DEC_LS_EN__SHIFT 0x14 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG5_DEC_DS_EN__SHIFT 0x15 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG5_DEC_SD_EN__SHIFT 0x16 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG5_DEC_LS_SW_EN__SHIFT 0x17 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG6_DEC_LS_EN__SHIFT 0x18 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG6_DEC_DS_EN__SHIFT 0x19 +#define JPEG_DEC_CGC_MEM_CTRL__JPEG6_DEC_SD_EN__SHIFT 0x1a +#define JPEG_DEC_CGC_MEM_CTRL__JPEG6_DEC_LS_SW_EN__SHIFT 0x1b +#define JPEG_DEC_CGC_MEM_CTRL__JPEG7_DEC_LS_EN__SHIFT 0x1c +#define JPEG_DEC_CGC_MEM_CTRL__JPEG7_DEC_DS_EN__SHIFT 0x1d +#define JPEG_DEC_CGC_MEM_CTRL__JPEG7_DEC_SD_EN__SHIFT 0x1e +#define JPEG_DEC_CGC_MEM_CTRL__JPEG7_DEC_LS_SW_EN__SHIFT 0x1f +#define JPEG_DEC_CGC_MEM_CTRL__JPEG0_DEC_LS_EN_MASK 0x00000001L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG0_DEC_DS_EN_MASK 0x00000002L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG0_DEC_SD_EN_MASK 0x00000004L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG0_DEC_LS_SW_EN_MASK 0x00000008L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG1_DEC_LS_EN_MASK 0x00000010L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG1_DEC_DS_EN_MASK 0x00000020L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG1_DEC_SD_EN_MASK 0x00000040L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG1_DEC_LS_SW_EN_MASK 0x00000080L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG2_DEC_LS_EN_MASK 0x00000100L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG2_DEC_DS_EN_MASK 0x00000200L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG2_DEC_SD_EN_MASK 0x00000400L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG2_DEC_LS_SW_EN_MASK 0x00000800L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG3_DEC_LS_EN_MASK 0x00001000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG3_DEC_DS_EN_MASK 0x00002000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG3_DEC_SD_EN_MASK 0x00004000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG3_DEC_LS_SW_EN_MASK 0x00008000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG4_DEC_LS_EN_MASK 0x00010000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG4_DEC_DS_EN_MASK 0x00020000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG4_DEC_SD_EN_MASK 0x00040000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG4_DEC_LS_SW_EN_MASK 0x00080000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG5_DEC_LS_EN_MASK 0x00100000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG5_DEC_DS_EN_MASK 0x00200000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG5_DEC_SD_EN_MASK 0x00400000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG5_DEC_LS_SW_EN_MASK 0x00800000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG6_DEC_LS_EN_MASK 0x01000000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG6_DEC_DS_EN_MASK 0x02000000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG6_DEC_SD_EN_MASK 0x04000000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG6_DEC_LS_SW_EN_MASK 0x08000000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG7_DEC_LS_EN_MASK 0x10000000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG7_DEC_DS_EN_MASK 0x20000000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG7_DEC_SD_EN_MASK 0x40000000L +#define JPEG_DEC_CGC_MEM_CTRL__JPEG7_DEC_LS_SW_EN_MASK 0x80000000L +//JPEG_ENC_CGC_MEM_CTRL +#define JPEG_ENC_CGC_MEM_CTRL__JPEG_ENC_LS_EN__SHIFT 0x0 +#define JPEG_ENC_CGC_MEM_CTRL__JPEG_ENC_DS_EN__SHIFT 0x1 +#define JPEG_ENC_CGC_MEM_CTRL__JPEG_ENC_SD_EN__SHIFT 0x2 +#define JPEG_ENC_CGC_MEM_CTRL__JPEG_ENC_LS_SW_EN__SHIFT 0x3 +#define JPEG_ENC_CGC_MEM_CTRL__JPEG_ENC_LS_EN_MASK 0x00000001L +#define JPEG_ENC_CGC_MEM_CTRL__JPEG_ENC_DS_EN_MASK 0x00000002L +#define JPEG_ENC_CGC_MEM_CTRL__JPEG_ENC_SD_EN_MASK 0x00000004L +#define JPEG_ENC_CGC_MEM_CTRL__JPEG_ENC_LS_SW_EN_MASK 0x00000008L +//JPEG_PERF_BANK_CONF +#define JPEG_PERF_BANK_CONF__RESET__SHIFT 0x0 +#define JPEG_PERF_BANK_CONF__PEEK__SHIFT 0x8 +#define JPEG_PERF_BANK_CONF__CONCATENATE__SHIFT 0x10 +#define JPEG_PERF_BANK_CONF__CORE_SEL__SHIFT 0x15 +#define JPEG_PERF_BANK_CONF__RESET_MASK 0x0000000FL +#define JPEG_PERF_BANK_CONF__PEEK_MASK 0x00000F00L +#define JPEG_PERF_BANK_CONF__CONCATENATE_MASK 0x00030000L +#define JPEG_PERF_BANK_CONF__CORE_SEL_MASK 0x00E00000L +//JPEG_PERF_BANK_EVENT_SEL +#define JPEG_PERF_BANK_EVENT_SEL__SEL0__SHIFT 0x0 +#define JPEG_PERF_BANK_EVENT_SEL__SEL1__SHIFT 0x8 +#define JPEG_PERF_BANK_EVENT_SEL__SEL2__SHIFT 0x10 +#define JPEG_PERF_BANK_EVENT_SEL__SEL3__SHIFT 0x18 +#define JPEG_PERF_BANK_EVENT_SEL__SEL0_MASK 0x000000FFL +#define JPEG_PERF_BANK_EVENT_SEL__SEL1_MASK 0x0000FF00L +#define JPEG_PERF_BANK_EVENT_SEL__SEL2_MASK 0x00FF0000L +#define JPEG_PERF_BANK_EVENT_SEL__SEL3_MASK 0xFF000000L +//JPEG_PERF_BANK_COUNT0 +#define JPEG_PERF_BANK_COUNT0__COUNT__SHIFT 0x0 +#define JPEG_PERF_BANK_COUNT0__COUNT_MASK 0xFFFFFFFFL +//JPEG_PERF_BANK_COUNT1 +#define JPEG_PERF_BANK_COUNT1__COUNT__SHIFT 0x0 +#define JPEG_PERF_BANK_COUNT1__COUNT_MASK 0xFFFFFFFFL +//JPEG_PERF_BANK_COUNT2 +#define JPEG_PERF_BANK_COUNT2__COUNT__SHIFT 0x0 +#define JPEG_PERF_BANK_COUNT2__COUNT_MASK 0xFFFFFFFFL +//JPEG_PERF_BANK_COUNT3 +#define JPEG_PERF_BANK_COUNT3__COUNT__SHIFT 0x0 +#define JPEG_PERF_BANK_COUNT3__COUNT_MASK 0xFFFFFFFFL + + +// addressBlock: aid_uvd0_uvd_pg_dec +//UVD_PGFSM_CONFIG +#define UVD_PGFSM_CONFIG__UVDM_PWR_CONFIG__SHIFT 0x0 +#define UVD_PGFSM_CONFIG__UVDS_PWR_CONFIG__SHIFT 0x2 +#define UVD_PGFSM_CONFIG__UVDF_PWR_CONFIG__SHIFT 0x4 +#define UVD_PGFSM_CONFIG__UVDTC_PWR_CONFIG__SHIFT 0x6 +#define UVD_PGFSM_CONFIG__UVDB_PWR_CONFIG__SHIFT 0x8 +#define UVD_PGFSM_CONFIG__UVDTA_PWR_CONFIG__SHIFT 0xa +#define UVD_PGFSM_CONFIG__UVDLM_PWR_CONFIG__SHIFT 0xc +#define UVD_PGFSM_CONFIG__UVDTD_PWR_CONFIG__SHIFT 0xe +#define UVD_PGFSM_CONFIG__UVDTE_PWR_CONFIG__SHIFT 0x10 +#define UVD_PGFSM_CONFIG__UVDE_PWR_CONFIG__SHIFT 0x12 +#define UVD_PGFSM_CONFIG__UVDAB_PWR_CONFIG__SHIFT 0x14 +#define UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG__SHIFT 0x16 +#define UVD_PGFSM_CONFIG__UVDTB_PWR_CONFIG__SHIFT 0x18 +#define UVD_PGFSM_CONFIG__UVDNA_PWR_CONFIG__SHIFT 0x1a +#define UVD_PGFSM_CONFIG__UVDNB_PWR_CONFIG__SHIFT 0x1c +#define UVD_PGFSM_CONFIG__UVDM_PWR_CONFIG_MASK 0x00000003L +#define UVD_PGFSM_CONFIG__UVDS_PWR_CONFIG_MASK 0x0000000CL +#define UVD_PGFSM_CONFIG__UVDF_PWR_CONFIG_MASK 0x00000030L +#define UVD_PGFSM_CONFIG__UVDTC_PWR_CONFIG_MASK 0x000000C0L +#define UVD_PGFSM_CONFIG__UVDB_PWR_CONFIG_MASK 0x00000300L +#define UVD_PGFSM_CONFIG__UVDTA_PWR_CONFIG_MASK 0x00000C00L +#define UVD_PGFSM_CONFIG__UVDLM_PWR_CONFIG_MASK 0x00003000L +#define UVD_PGFSM_CONFIG__UVDTD_PWR_CONFIG_MASK 0x0000C000L +#define UVD_PGFSM_CONFIG__UVDTE_PWR_CONFIG_MASK 0x00030000L +#define UVD_PGFSM_CONFIG__UVDE_PWR_CONFIG_MASK 0x000C0000L +#define UVD_PGFSM_CONFIG__UVDAB_PWR_CONFIG_MASK 0x00300000L +#define UVD_PGFSM_CONFIG__UVDJ_PWR_CONFIG_MASK 0x00C00000L +#define UVD_PGFSM_CONFIG__UVDTB_PWR_CONFIG_MASK 0x03000000L +#define UVD_PGFSM_CONFIG__UVDNA_PWR_CONFIG_MASK 0x0C000000L +#define UVD_PGFSM_CONFIG__UVDNB_PWR_CONFIG_MASK 0x30000000L +//UVD_PGFSM_STATUS +#define UVD_PGFSM_STATUS__UVDM_PWR_STATUS__SHIFT 0x0 +#define UVD_PGFSM_STATUS__UVDS_PWR_STATUS__SHIFT 0x2 +#define UVD_PGFSM_STATUS__UVDF_PWR_STATUS__SHIFT 0x4 +#define UVD_PGFSM_STATUS__UVDTC_PWR_STATUS__SHIFT 0x6 +#define UVD_PGFSM_STATUS__UVDB_PWR_STATUS__SHIFT 0x8 +#define UVD_PGFSM_STATUS__UVDTA_PWR_STATUS__SHIFT 0xa +#define UVD_PGFSM_STATUS__UVDLM_PWR_STATUS__SHIFT 0xc +#define UVD_PGFSM_STATUS__UVDTD_PWR_STATUS__SHIFT 0xe +#define UVD_PGFSM_STATUS__UVDTE_PWR_STATUS__SHIFT 0x10 +#define UVD_PGFSM_STATUS__UVDE_PWR_STATUS__SHIFT 0x12 +#define UVD_PGFSM_STATUS__UVDAB_PWR_STATUS__SHIFT 0x14 +#define UVD_PGFSM_STATUS__UVDJ_PWR_STATUS__SHIFT 0x16 +#define UVD_PGFSM_STATUS__UVDTB_PWR_STATUS__SHIFT 0x18 +#define UVD_PGFSM_STATUS__UVDNA_PWR_STATUS__SHIFT 0x1a +#define UVD_PGFSM_STATUS__UVDNB_PWR_STATUS__SHIFT 0x1c +#define UVD_PGFSM_STATUS__UVDM_PWR_STATUS_MASK 0x00000003L +#define UVD_PGFSM_STATUS__UVDS_PWR_STATUS_MASK 0x0000000CL +#define UVD_PGFSM_STATUS__UVDF_PWR_STATUS_MASK 0x00000030L +#define UVD_PGFSM_STATUS__UVDTC_PWR_STATUS_MASK 0x000000C0L +#define UVD_PGFSM_STATUS__UVDB_PWR_STATUS_MASK 0x00000300L +#define UVD_PGFSM_STATUS__UVDTA_PWR_STATUS_MASK 0x00000C00L +#define UVD_PGFSM_STATUS__UVDLM_PWR_STATUS_MASK 0x00003000L +#define UVD_PGFSM_STATUS__UVDTD_PWR_STATUS_MASK 0x0000C000L +#define UVD_PGFSM_STATUS__UVDTE_PWR_STATUS_MASK 0x00030000L +#define UVD_PGFSM_STATUS__UVDE_PWR_STATUS_MASK 0x000C0000L +#define UVD_PGFSM_STATUS__UVDAB_PWR_STATUS_MASK 0x00300000L +#define UVD_PGFSM_STATUS__UVDJ_PWR_STATUS_MASK 0x00C00000L +#define UVD_PGFSM_STATUS__UVDTB_PWR_STATUS_MASK 0x03000000L +#define UVD_PGFSM_STATUS__UVDNA_PWR_STATUS_MASK 0x0C000000L +#define UVD_PGFSM_STATUS__UVDNB_PWR_STATUS_MASK 0x30000000L +//UVD_POWER_STATUS +#define UVD_POWER_STATUS__UVD_POWER_STATUS__SHIFT 0x0 +#define UVD_POWER_STATUS__UVD_PG_MODE__SHIFT 0x2 +#define UVD_POWER_STATUS__UVD_CG_MODE__SHIFT 0x4 +#define UVD_POWER_STATUS__UVD_PG_EN__SHIFT 0x8 +#define UVD_POWER_STATUS__RBC_SNOOP_DIS__SHIFT 0x9 +#define UVD_POWER_STATUS__SW_RB_SNOOP_DIS__SHIFT 0xb +#define UVD_POWER_STATUS__STALL_DPG_POWER_UP__SHIFT 0x1f +#define UVD_POWER_STATUS__UVD_POWER_STATUS_MASK 0x00000003L +#define UVD_POWER_STATUS__UVD_PG_MODE_MASK 0x00000004L +#define UVD_POWER_STATUS__UVD_CG_MODE_MASK 0x00000030L +#define UVD_POWER_STATUS__UVD_PG_EN_MASK 0x00000100L +#define UVD_POWER_STATUS__RBC_SNOOP_DIS_MASK 0x00000200L +#define UVD_POWER_STATUS__SW_RB_SNOOP_DIS_MASK 0x00000800L +#define UVD_POWER_STATUS__STALL_DPG_POWER_UP_MASK 0x80000000L +//UVD_JPEG_POWER_STATUS +#define UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS__SHIFT 0x0 +#define UVD_JPEG_POWER_STATUS__JPEG_PG_MODE__SHIFT 0x4 +#define UVD_JPEG_POWER_STATUS__JRBC_DEC_SNOOP_DIS__SHIFT 0x8 +#define UVD_JPEG_POWER_STATUS__JRBC_ENC_SNOOP_DIS__SHIFT 0x9 +#define UVD_JPEG_POWER_STATUS__STALL_JDPG_POWER_UP__SHIFT 0x1f +#define UVD_JPEG_POWER_STATUS__JPEG_POWER_STATUS_MASK 0x00000001L +#define UVD_JPEG_POWER_STATUS__JPEG_PG_MODE_MASK 0x00000010L +#define UVD_JPEG_POWER_STATUS__JRBC_DEC_SNOOP_DIS_MASK 0x00000100L +#define UVD_JPEG_POWER_STATUS__JRBC_ENC_SNOOP_DIS_MASK 0x00000200L +#define UVD_JPEG_POWER_STATUS__STALL_JDPG_POWER_UP_MASK 0x80000000L +//UVD_MC_DJPEG_RD_SPACE +#define UVD_MC_DJPEG_RD_SPACE__DJPEG_RD_SPACE__SHIFT 0x0 +#define UVD_MC_DJPEG_RD_SPACE__DJPEG_RD_SPACE_MASK 0x0003FFFFL +//UVD_MC_DJPEG_WR_SPACE +#define UVD_MC_DJPEG_WR_SPACE__DJPEG_WR_SPACE__SHIFT 0x0 +#define UVD_MC_DJPEG_WR_SPACE__DJPEG_WR_SPACE_MASK 0x0003FFFFL +//UVD_MC_EJPEG_RD_SPACE +#define UVD_MC_EJPEG_RD_SPACE__EJPEG_RD_SPACE__SHIFT 0x0 +#define UVD_MC_EJPEG_RD_SPACE__EJPEG_RD_SPACE_MASK 0x0003FFFFL +//UVD_MC_EJPEG_WR_SPACE +#define UVD_MC_EJPEG_WR_SPACE__EJPEG_WR_SPACE__SHIFT 0x0 +#define UVD_MC_EJPEG_WR_SPACE__EJPEG_WR_SPACE_MASK 0x0003FFFFL +//UVD_PG_IND_INDEX +#define UVD_PG_IND_INDEX__INDEX__SHIFT 0x0 +#define UVD_PG_IND_INDEX__INDEX_MASK 0x0000003FL +//UVD_PG_IND_DATA +#define UVD_PG_IND_DATA__DATA__SHIFT 0x0 +#define UVD_PG_IND_DATA__DATA_MASK 0xFFFFFFFFL +//CC_UVD_HARVESTING +#define CC_UVD_HARVESTING__MMSCH_DISABLE__SHIFT 0x0 +#define CC_UVD_HARVESTING__UVD_DISABLE__SHIFT 0x1 +#define CC_UVD_HARVESTING__MMSCH_DISABLE_MASK 0x00000001L +#define CC_UVD_HARVESTING__UVD_DISABLE_MASK 0x00000002L +//UVD_DPG_LMA_CTL +#define UVD_DPG_LMA_CTL__READ_WRITE__SHIFT 0x0 +#define UVD_DPG_LMA_CTL__MASK_EN__SHIFT 0x1 +#define UVD_DPG_LMA_CTL__ADDR_AUTO_INCREMENT__SHIFT 0x2 +#define UVD_DPG_LMA_CTL__SRAM_SEL__SHIFT 0x4 +#define UVD_DPG_LMA_CTL__READ_WRITE_ADDR__SHIFT 0x10 +#define UVD_DPG_LMA_CTL__READ_WRITE_MASK 0x00000001L +#define UVD_DPG_LMA_CTL__MASK_EN_MASK 0x00000002L +#define UVD_DPG_LMA_CTL__ADDR_AUTO_INCREMENT_MASK 0x00000004L +#define UVD_DPG_LMA_CTL__SRAM_SEL_MASK 0x00000010L +#define UVD_DPG_LMA_CTL__READ_WRITE_ADDR_MASK 0xFFFF0000L +//UVD_DPG_LMA_DATA +#define UVD_DPG_LMA_DATA__LMA_DATA__SHIFT 0x0 +#define UVD_DPG_LMA_DATA__LMA_DATA_MASK 0xFFFFFFFFL +//UVD_DPG_LMA_MASK +#define UVD_DPG_LMA_MASK__LMA_MASK__SHIFT 0x0 +#define UVD_DPG_LMA_MASK__LMA_MASK_MASK 0xFFFFFFFFL +//UVD_DPG_PAUSE +#define UVD_DPG_PAUSE__JPEG_PAUSE_DPG_REQ__SHIFT 0x0 +#define UVD_DPG_PAUSE__JPEG_PAUSE_DPG_ACK__SHIFT 0x1 +#define UVD_DPG_PAUSE__NJ_PAUSE_DPG_REQ__SHIFT 0x2 +#define UVD_DPG_PAUSE__NJ_PAUSE_DPG_ACK__SHIFT 0x3 +#define UVD_DPG_PAUSE__JPEG_PAUSE_DPG_REQ_MASK 0x00000001L +#define UVD_DPG_PAUSE__JPEG_PAUSE_DPG_ACK_MASK 0x00000002L +#define UVD_DPG_PAUSE__NJ_PAUSE_DPG_REQ_MASK 0x00000004L +#define UVD_DPG_PAUSE__NJ_PAUSE_DPG_ACK_MASK 0x00000008L +//UVD_SCRATCH1 +#define UVD_SCRATCH1__SCRATCH1_DATA__SHIFT 0x0 +#define UVD_SCRATCH1__SCRATCH1_DATA_MASK 0xFFFFFFFFL +//UVD_SCRATCH2 +#define UVD_SCRATCH2__SCRATCH2_DATA__SHIFT 0x0 +#define UVD_SCRATCH2__SCRATCH2_DATA_MASK 0xFFFFFFFFL +//UVD_SCRATCH3 +#define UVD_SCRATCH3__SCRATCH3_DATA__SHIFT 0x0 +#define UVD_SCRATCH3__SCRATCH3_DATA_MASK 0xFFFFFFFFL +//UVD_SCRATCH4 +#define UVD_SCRATCH4__SCRATCH4_DATA__SHIFT 0x0 +#define UVD_SCRATCH4__SCRATCH4_DATA_MASK 0xFFFFFFFFL +//UVD_SCRATCH5 +#define UVD_SCRATCH5__SCRATCH5_DATA__SHIFT 0x0 +#define UVD_SCRATCH5__SCRATCH5_DATA_MASK 0xFFFFFFFFL +//UVD_SCRATCH6 +#define UVD_SCRATCH6__SCRATCH6_DATA__SHIFT 0x0 +#define UVD_SCRATCH6__SCRATCH6_DATA_MASK 0xFFFFFFFFL +//UVD_SCRATCH7 +#define UVD_SCRATCH7__SCRATCH7_DATA__SHIFT 0x0 +#define UVD_SCRATCH7__SCRATCH7_DATA_MASK 0xFFFFFFFFL +//UVD_SCRATCH8 +#define UVD_SCRATCH8__SCRATCH8_DATA__SHIFT 0x0 +#define UVD_SCRATCH8__SCRATCH8_DATA_MASK 0xFFFFFFFFL +//UVD_SCRATCH9 +#define UVD_SCRATCH9__SCRATCH9_DATA__SHIFT 0x0 +#define UVD_SCRATCH9__SCRATCH9_DATA_MASK 0xFFFFFFFFL +//UVD_SCRATCH10 +#define UVD_SCRATCH10__SCRATCH10_DATA__SHIFT 0x0 +#define UVD_SCRATCH10__SCRATCH10_DATA_MASK 0xFFFFFFFFL +//UVD_SCRATCH11 +#define UVD_SCRATCH11__SCRATCH11_DATA__SHIFT 0x0 +#define UVD_SCRATCH11__SCRATCH11_DATA_MASK 0xFFFFFFFFL +//UVD_SCRATCH12 +#define UVD_SCRATCH12__SCRATCH12_DATA__SHIFT 0x0 +#define UVD_SCRATCH12__SCRATCH12_DATA_MASK 0xFFFFFFFFL +//UVD_SCRATCH13 +#define UVD_SCRATCH13__SCRATCH13_DATA__SHIFT 0x0 +#define UVD_SCRATCH13__SCRATCH13_DATA_MASK 0xFFFFFFFFL +//UVD_SCRATCH14 +#define UVD_SCRATCH14__SCRATCH14_DATA__SHIFT 0x0 +#define UVD_SCRATCH14__SCRATCH14_DATA_MASK 0xFFFFFFFFL +//UVD_FREE_COUNTER_REG +#define UVD_FREE_COUNTER_REG__FREE_COUNTER__SHIFT 0x0 +#define UVD_FREE_COUNTER_REG__FREE_COUNTER_MASK 0xFFFFFFFFL +//UVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_LOW +#define UVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_HIGH +#define UVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_DPG_LMI_VCPU_CACHE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_DPG_VCPU_CACHE_OFFSET0 +#define UVD_DPG_VCPU_CACHE_OFFSET0__CACHE_OFFSET0__SHIFT 0x0 +#define UVD_DPG_VCPU_CACHE_OFFSET0__CACHE_OFFSET0_MASK 0x01FFFFFFL +//UVD_DPG_LMI_VCPU_CACHE_VMID +#define UVD_DPG_LMI_VCPU_CACHE_VMID__VCPU_CACHE_VMID__SHIFT 0x0 +#define UVD_DPG_LMI_VCPU_CACHE_VMID__VCPU_CACHE_VMID_MASK 0x0000000FL +//UVD_REG_FILTER_EN +#define UVD_REG_FILTER_EN__UVD_REG_FILTER_EN__SHIFT 0x0 +#define UVD_REG_FILTER_EN__MMSCH_HI_PRIV__SHIFT 0x1 +#define UVD_REG_FILTER_EN__VIDEO_PRIV_EN__SHIFT 0x2 +#define UVD_REG_FILTER_EN__JPEG_PRIV_EN__SHIFT 0x3 +#define UVD_REG_FILTER_EN__UVD_REG_FILTER_EN_MASK 0x00000001L +#define UVD_REG_FILTER_EN__MMSCH_HI_PRIV_MASK 0x00000002L +#define UVD_REG_FILTER_EN__VIDEO_PRIV_EN_MASK 0x00000004L +#define UVD_REG_FILTER_EN__JPEG_PRIV_EN_MASK 0x00000008L +//UVD_SECURITY_REG_VIO_REPORT +#define UVD_SECURITY_REG_VIO_REPORT__HOST_REG_VIO__SHIFT 0x0 +#define UVD_SECURITY_REG_VIO_REPORT__VCPU_REG_VIO__SHIFT 0x1 +#define UVD_SECURITY_REG_VIO_REPORT__VIDEO_REG_VIO__SHIFT 0x2 +#define UVD_SECURITY_REG_VIO_REPORT__DPG_REG_VIO__SHIFT 0x3 +#define UVD_SECURITY_REG_VIO_REPORT__JPEG_REG_VIO__SHIFT 0x4 +#define UVD_SECURITY_REG_VIO_REPORT__JDPG_REG_VIO__SHIFT 0x5 +#define UVD_SECURITY_REG_VIO_REPORT__HOST_REG_VIO_MASK 0x00000001L +#define UVD_SECURITY_REG_VIO_REPORT__VCPU_REG_VIO_MASK 0x00000002L +#define UVD_SECURITY_REG_VIO_REPORT__VIDEO_REG_VIO_MASK 0x00000004L +#define UVD_SECURITY_REG_VIO_REPORT__DPG_REG_VIO_MASK 0x00000008L +#define UVD_SECURITY_REG_VIO_REPORT__JPEG_REG_VIO_MASK 0x00000010L +#define UVD_SECURITY_REG_VIO_REPORT__JDPG_REG_VIO_MASK 0x00000020L +//UVD_FW_VERSION +#define UVD_FW_VERSION__FW_VERSION__SHIFT 0x0 +#define UVD_FW_VERSION__FW_VERSION_MASK 0xFFFFFFFFL +//UVD_PF_STATUS +#define UVD_PF_STATUS__JPEG_PF_OCCURED__SHIFT 0x0 +#define UVD_PF_STATUS__NJ_PF_OCCURED__SHIFT 0x1 +#define UVD_PF_STATUS__ENCODER0_PF_OCCURED__SHIFT 0x2 +#define UVD_PF_STATUS__ENCODER1_PF_OCCURED__SHIFT 0x3 +#define UVD_PF_STATUS__ENCODER2_PF_OCCURED__SHIFT 0x4 +#define UVD_PF_STATUS__ENCODER3_PF_OCCURED__SHIFT 0x5 +#define UVD_PF_STATUS__ENCODER4_PF_OCCURED__SHIFT 0x6 +#define UVD_PF_STATUS__EJPEG_PF_OCCURED__SHIFT 0x7 +#define UVD_PF_STATUS__JPEG_PF_CLEAR__SHIFT 0x8 +#define UVD_PF_STATUS__NJ_PF_CLEAR__SHIFT 0x9 +#define UVD_PF_STATUS__ENCODER0_PF_CLEAR__SHIFT 0xa +#define UVD_PF_STATUS__ENCODER1_PF_CLEAR__SHIFT 0xb +#define UVD_PF_STATUS__ENCODER2_PF_CLEAR__SHIFT 0xc +#define UVD_PF_STATUS__ENCODER3_PF_CLEAR__SHIFT 0xd +#define UVD_PF_STATUS__ENCODER4_PF_CLEAR__SHIFT 0xe +#define UVD_PF_STATUS__EJPEG_PF_CLEAR__SHIFT 0xf +#define UVD_PF_STATUS__NJ_ATM_PF_OCCURED__SHIFT 0x10 +#define UVD_PF_STATUS__DJ_ATM_PF_OCCURED__SHIFT 0x11 +#define UVD_PF_STATUS__EJ_ATM_PF_OCCURED__SHIFT 0x12 +#define UVD_PF_STATUS__JPEG2_PF_OCCURED__SHIFT 0x13 +#define UVD_PF_STATUS__DJ2_ATM_PF_OCCURED__SHIFT 0x14 +#define UVD_PF_STATUS__JPEG2_PF_CLEAR__SHIFT 0x15 +#define UVD_PF_STATUS__ENCODER5_PF_OCCURED__SHIFT 0x16 +#define UVD_PF_STATUS__ENCODER5_PF_CLEAR__SHIFT 0x17 +#define UVD_PF_STATUS__JPEG_PF_OCCURED_MASK 0x00000001L +#define UVD_PF_STATUS__NJ_PF_OCCURED_MASK 0x00000002L +#define UVD_PF_STATUS__ENCODER0_PF_OCCURED_MASK 0x00000004L +#define UVD_PF_STATUS__ENCODER1_PF_OCCURED_MASK 0x00000008L +#define UVD_PF_STATUS__ENCODER2_PF_OCCURED_MASK 0x00000010L +#define UVD_PF_STATUS__ENCODER3_PF_OCCURED_MASK 0x00000020L +#define UVD_PF_STATUS__ENCODER4_PF_OCCURED_MASK 0x00000040L +#define UVD_PF_STATUS__EJPEG_PF_OCCURED_MASK 0x00000080L +#define UVD_PF_STATUS__JPEG_PF_CLEAR_MASK 0x00000100L +#define UVD_PF_STATUS__NJ_PF_CLEAR_MASK 0x00000200L +#define UVD_PF_STATUS__ENCODER0_PF_CLEAR_MASK 0x00000400L +#define UVD_PF_STATUS__ENCODER1_PF_CLEAR_MASK 0x00000800L +#define UVD_PF_STATUS__ENCODER2_PF_CLEAR_MASK 0x00001000L +#define UVD_PF_STATUS__ENCODER3_PF_CLEAR_MASK 0x00002000L +#define UVD_PF_STATUS__ENCODER4_PF_CLEAR_MASK 0x00004000L +#define UVD_PF_STATUS__EJPEG_PF_CLEAR_MASK 0x00008000L +#define UVD_PF_STATUS__NJ_ATM_PF_OCCURED_MASK 0x00010000L +#define UVD_PF_STATUS__DJ_ATM_PF_OCCURED_MASK 0x00020000L +#define UVD_PF_STATUS__EJ_ATM_PF_OCCURED_MASK 0x00040000L +#define UVD_PF_STATUS__JPEG2_PF_OCCURED_MASK 0x00080000L +#define UVD_PF_STATUS__DJ2_ATM_PF_OCCURED_MASK 0x00100000L +#define UVD_PF_STATUS__JPEG2_PF_CLEAR_MASK 0x00200000L +#define UVD_PF_STATUS__ENCODER5_PF_OCCURED_MASK 0x00400000L +#define UVD_PF_STATUS__ENCODER5_PF_CLEAR_MASK 0x00800000L +//UVD_DPG_CLK_EN_VCPU_REPORT +#define UVD_DPG_CLK_EN_VCPU_REPORT__CLK_EN__SHIFT 0x0 +#define UVD_DPG_CLK_EN_VCPU_REPORT__VCPU_REPORT__SHIFT 0x1 +#define UVD_DPG_CLK_EN_VCPU_REPORT__CLK_EN_MASK 0x00000001L +#define UVD_DPG_CLK_EN_VCPU_REPORT__VCPU_REPORT_MASK 0x000000FEL +//CC_UVD_VCPU_ERR_DETECT_BOT_LO +#define CC_UVD_VCPU_ERR_DETECT_BOT_LO__UVD_VCPU_ERR_DETECT_BOT_LO__SHIFT 0xc +#define CC_UVD_VCPU_ERR_DETECT_BOT_LO__UVD_VCPU_ERR_DETECT_BOT_LO_MASK 0xFFFFF000L +//CC_UVD_VCPU_ERR_DETECT_BOT_HI +#define CC_UVD_VCPU_ERR_DETECT_BOT_HI__UVD_VCPU_ERR_DETECT_BOT_HI__SHIFT 0x0 +#define CC_UVD_VCPU_ERR_DETECT_BOT_HI__UVD_VCPU_ERR_DETECT_BOT_HI_MASK 0x0000FFFFL +//CC_UVD_VCPU_ERR_DETECT_TOP_LO +#define CC_UVD_VCPU_ERR_DETECT_TOP_LO__UVD_VCPU_ERR_DETECT_TOP_LO__SHIFT 0xc +#define CC_UVD_VCPU_ERR_DETECT_TOP_LO__UVD_VCPU_ERR_DETECT_TOP_LO_MASK 0xFFFFF000L +//CC_UVD_VCPU_ERR_DETECT_TOP_HI +#define CC_UVD_VCPU_ERR_DETECT_TOP_HI__UVD_VCPU_ERR_DETECT_TOP_HI__SHIFT 0x0 +#define CC_UVD_VCPU_ERR_DETECT_TOP_HI__UVD_VCPU_ERR_DETECT_TOP_HI_MASK 0x0000FFFFL +//CC_UVD_VCPU_ERR +#define CC_UVD_VCPU_ERR__UVD_VCPU_ERR_STATUS__SHIFT 0x0 +#define CC_UVD_VCPU_ERR__UVD_VCPU_ERR_CLEAR__SHIFT 0x1 +#define CC_UVD_VCPU_ERR__UVD_VCPU_ERR_DETECT_EN__SHIFT 0x2 +#define CC_UVD_VCPU_ERR__UVD_TMZ_DBG_DIS__SHIFT 0x3 +#define CC_UVD_VCPU_ERR__RESET_ON_FAULT__SHIFT 0x4 +#define CC_UVD_VCPU_ERR__UVD_VCPU_ERR_STATUS_MASK 0x00000001L +#define CC_UVD_VCPU_ERR__UVD_VCPU_ERR_CLEAR_MASK 0x00000002L +#define CC_UVD_VCPU_ERR__UVD_VCPU_ERR_DETECT_EN_MASK 0x00000004L +#define CC_UVD_VCPU_ERR__UVD_TMZ_DBG_DIS_MASK 0x00000008L +#define CC_UVD_VCPU_ERR__RESET_ON_FAULT_MASK 0x00000010L +//CC_UVD_VCPU_ERR_INST_ADDR_LO +#define CC_UVD_VCPU_ERR_INST_ADDR_LO__UVD_VCPU_ERR_INST_ADDR_LO__SHIFT 0x0 +#define CC_UVD_VCPU_ERR_INST_ADDR_LO__UVD_VCPU_ERR_INST_ADDR_LO_MASK 0xFFFFFFFFL +//CC_UVD_VCPU_ERR_INST_ADDR_HI +#define CC_UVD_VCPU_ERR_INST_ADDR_HI__UVD_VCPU_ERR_INST_ADDR_HI__SHIFT 0x0 +#define CC_UVD_VCPU_ERR_INST_ADDR_HI__UVD_VCPU_ERR_INST_ADDR_HI_MASK 0x0000FFFFL +//UVD_LMI_MMSCH_NC_SPACE +#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC0_SPACE__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC1_SPACE__SHIFT 0x3 +#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC2_SPACE__SHIFT 0x6 +#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC3_SPACE__SHIFT 0x9 +#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC4_SPACE__SHIFT 0xc +#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC5_SPACE__SHIFT 0xf +#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC6_SPACE__SHIFT 0x12 +#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC7_SPACE__SHIFT 0x15 +#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC0_SPACE_MASK 0x00000007L +#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC1_SPACE_MASK 0x00000038L +#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC2_SPACE_MASK 0x000001C0L +#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC3_SPACE_MASK 0x00000E00L +#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC4_SPACE_MASK 0x00007000L +#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC5_SPACE_MASK 0x00038000L +#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC6_SPACE_MASK 0x001C0000L +#define UVD_LMI_MMSCH_NC_SPACE__MMSCH_NC7_SPACE_MASK 0x00E00000L +//UVD_LMI_ATOMIC_SPACE +#define UVD_LMI_ATOMIC_SPACE__ATOMIC_USER0_SPACE__SHIFT 0x0 +#define UVD_LMI_ATOMIC_SPACE__ATOMIC_USER1_SPACE__SHIFT 0x3 +#define UVD_LMI_ATOMIC_SPACE__ATOMIC_USER2_SPACE__SHIFT 0x6 +#define UVD_LMI_ATOMIC_SPACE__ATOMIC_USER3_SPACE__SHIFT 0x9 +#define UVD_LMI_ATOMIC_SPACE__ATOMIC_USER0_SPACE_MASK 0x00000007L +#define UVD_LMI_ATOMIC_SPACE__ATOMIC_USER1_SPACE_MASK 0x00000038L +#define UVD_LMI_ATOMIC_SPACE__ATOMIC_USER2_SPACE_MASK 0x000001C0L +#define UVD_LMI_ATOMIC_SPACE__ATOMIC_USER3_SPACE_MASK 0x00000E00L +//UVD_GFX8_ADDR_CONFIG +#define UVD_GFX8_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE__SHIFT 0x4 +#define UVD_GFX8_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE_MASK 0x00000070L +//UVD_GFX10_ADDR_CONFIG +#define UVD_GFX10_ADDR_CONFIG__NUM_PIPES__SHIFT 0x0 +#define UVD_GFX10_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE__SHIFT 0x3 +#define UVD_GFX10_ADDR_CONFIG__MAX_COMPRESSED_FRAGS__SHIFT 0x6 +#define UVD_GFX10_ADDR_CONFIG__NUM_PKRS__SHIFT 0x8 +#define UVD_GFX10_ADDR_CONFIG__NUM_BANKS__SHIFT 0xc +#define UVD_GFX10_ADDR_CONFIG__NUM_SHADER_ENGINES__SHIFT 0x13 +#define UVD_GFX10_ADDR_CONFIG__NUM_PIPES_MASK 0x00000007L +#define UVD_GFX10_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE_MASK 0x00000038L +#define UVD_GFX10_ADDR_CONFIG__MAX_COMPRESSED_FRAGS_MASK 0x000000C0L +#define UVD_GFX10_ADDR_CONFIG__NUM_PKRS_MASK 0x00000700L +#define UVD_GFX10_ADDR_CONFIG__NUM_BANKS_MASK 0x00007000L +#define UVD_GFX10_ADDR_CONFIG__NUM_SHADER_ENGINES_MASK 0x00180000L +//UVD_GPCNT2_CNTL +#define UVD_GPCNT2_CNTL__CLR__SHIFT 0x0 +#define UVD_GPCNT2_CNTL__START__SHIFT 0x1 +#define UVD_GPCNT2_CNTL__COUNTUP__SHIFT 0x2 +#define UVD_GPCNT2_CNTL__CLR_MASK 0x00000001L +#define UVD_GPCNT2_CNTL__START_MASK 0x00000002L +#define UVD_GPCNT2_CNTL__COUNTUP_MASK 0x00000004L +//UVD_GPCNT2_TARGET_LOWER +#define UVD_GPCNT2_TARGET_LOWER__TARGET__SHIFT 0x0 +#define UVD_GPCNT2_TARGET_LOWER__TARGET_MASK 0xFFFFFFFFL +//UVD_GPCNT2_STATUS_LOWER +#define UVD_GPCNT2_STATUS_LOWER__COUNT__SHIFT 0x0 +#define UVD_GPCNT2_STATUS_LOWER__COUNT_MASK 0xFFFFFFFFL +//UVD_GPCNT2_TARGET_UPPER +#define UVD_GPCNT2_TARGET_UPPER__TARGET__SHIFT 0x0 +#define UVD_GPCNT2_TARGET_UPPER__TARGET_MASK 0x0000FFFFL +//UVD_GPCNT2_STATUS_UPPER +#define UVD_GPCNT2_STATUS_UPPER__COUNT__SHIFT 0x0 +#define UVD_GPCNT2_STATUS_UPPER__COUNT_MASK 0x0000FFFFL +//UVD_GPCNT3_CNTL +#define UVD_GPCNT3_CNTL__CLR__SHIFT 0x0 +#define UVD_GPCNT3_CNTL__START__SHIFT 0x1 +#define UVD_GPCNT3_CNTL__COUNTUP__SHIFT 0x2 +#define UVD_GPCNT3_CNTL__FREQ__SHIFT 0x3 +#define UVD_GPCNT3_CNTL__DIV__SHIFT 0xa +#define UVD_GPCNT3_CNTL__CLR_MASK 0x00000001L +#define UVD_GPCNT3_CNTL__START_MASK 0x00000002L +#define UVD_GPCNT3_CNTL__COUNTUP_MASK 0x00000004L +#define UVD_GPCNT3_CNTL__FREQ_MASK 0x000003F8L +#define UVD_GPCNT3_CNTL__DIV_MASK 0x0001FC00L +//UVD_GPCNT3_TARGET_LOWER +#define UVD_GPCNT3_TARGET_LOWER__TARGET__SHIFT 0x0 +#define UVD_GPCNT3_TARGET_LOWER__TARGET_MASK 0xFFFFFFFFL +//UVD_GPCNT3_STATUS_LOWER +#define UVD_GPCNT3_STATUS_LOWER__COUNT__SHIFT 0x0 +#define UVD_GPCNT3_STATUS_LOWER__COUNT_MASK 0xFFFFFFFFL +//UVD_GPCNT3_TARGET_UPPER +#define UVD_GPCNT3_TARGET_UPPER__TARGET__SHIFT 0x0 +#define UVD_GPCNT3_TARGET_UPPER__TARGET_MASK 0x0000FFFFL +//UVD_GPCNT3_STATUS_UPPER +#define UVD_GPCNT3_STATUS_UPPER__COUNT__SHIFT 0x0 +#define UVD_GPCNT3_STATUS_UPPER__COUNT_MASK 0x0000FFFFL +//UVD_VCLK_DS_CNTL +#define UVD_VCLK_DS_CNTL__VCLK_DS_EN__SHIFT 0x0 +#define UVD_VCLK_DS_CNTL__VCLK_DS_STATUS__SHIFT 0x4 +#define UVD_VCLK_DS_CNTL__VCLK_DS_HYSTERESIS_CNT__SHIFT 0x10 +#define UVD_VCLK_DS_CNTL__VCLK_DS_EN_MASK 0x00000001L +#define UVD_VCLK_DS_CNTL__VCLK_DS_STATUS_MASK 0x00000010L +#define UVD_VCLK_DS_CNTL__VCLK_DS_HYSTERESIS_CNT_MASK 0xFFFF0000L +//UVD_DCLK_DS_CNTL +#define UVD_DCLK_DS_CNTL__DCLK_DS_EN__SHIFT 0x0 +#define UVD_DCLK_DS_CNTL__DCLK_DS_STATUS__SHIFT 0x4 +#define UVD_DCLK_DS_CNTL__DCLK_DS_HYSTERESIS_CNT__SHIFT 0x10 +#define UVD_DCLK_DS_CNTL__DCLK_DS_EN_MASK 0x00000001L +#define UVD_DCLK_DS_CNTL__DCLK_DS_STATUS_MASK 0x00000010L +#define UVD_DCLK_DS_CNTL__DCLK_DS_HYSTERESIS_CNT_MASK 0xFFFF0000L +//UVD_TSC_LOWER +#define UVD_TSC_LOWER__COUNT__SHIFT 0x0 +#define UVD_TSC_LOWER__COUNT_MASK 0xFFFFFFFFL +//UVD_TSC_UPPER +#define UVD_TSC_UPPER__COUNT__SHIFT 0x0 +#define UVD_TSC_UPPER__COUNT_MASK 0x00FFFFFFL +//VCN_FEATURES +#define VCN_FEATURES__HAS_VIDEO_DEC__SHIFT 0x0 +#define VCN_FEATURES__HAS_VIDEO_ENC__SHIFT 0x1 +#define VCN_FEATURES__HAS_MJPEG_DEC__SHIFT 0x2 +#define VCN_FEATURES__HAS_MJPEG_ENC__SHIFT 0x3 +#define VCN_FEATURES__HAS_VIDEO_VIRT__SHIFT 0x4 +#define VCN_FEATURES__HAS_H264_LEGACY_DEC__SHIFT 0x5 +#define VCN_FEATURES__HAS_UDEC_DEC__SHIFT 0x6 +#define VCN_FEATURES__HAS_MJPEG2_IDCT_DEC__SHIFT 0x7 +#define VCN_FEATURES__HAS_SCLR_DEC__SHIFT 0x8 +#define VCN_FEATURES__HAS_VP9_DEC__SHIFT 0x9 +#define VCN_FEATURES__HAS_AV1_DEC__SHIFT 0xa +#define VCN_FEATURES__HAS_EFC_ENC__SHIFT 0xb +#define VCN_FEATURES__HAS_EFC_HDR2SDR_ENC__SHIFT 0xc +#define VCN_FEATURES__HAS_DUAL_MJPEG_DEC__SHIFT 0xd +#define VCN_FEATURES__HAS_AV1_ENC__SHIFT 0xe +#define VCN_FEATURES__INSTANCE_ID__SHIFT 0x1c +#define VCN_FEATURES__HAS_VIDEO_DEC_MASK 0x00000001L +#define VCN_FEATURES__HAS_VIDEO_ENC_MASK 0x00000002L +#define VCN_FEATURES__HAS_MJPEG_DEC_MASK 0x00000004L +#define VCN_FEATURES__HAS_MJPEG_ENC_MASK 0x00000008L +#define VCN_FEATURES__HAS_VIDEO_VIRT_MASK 0x00000010L +#define VCN_FEATURES__HAS_H264_LEGACY_DEC_MASK 0x00000020L +#define VCN_FEATURES__HAS_UDEC_DEC_MASK 0x00000040L +#define VCN_FEATURES__HAS_MJPEG2_IDCT_DEC_MASK 0x00000080L +#define VCN_FEATURES__HAS_SCLR_DEC_MASK 0x00000100L +#define VCN_FEATURES__HAS_VP9_DEC_MASK 0x00000200L +#define VCN_FEATURES__HAS_AV1_DEC_MASK 0x00000400L +#define VCN_FEATURES__HAS_EFC_ENC_MASK 0x00000800L +#define VCN_FEATURES__HAS_EFC_HDR2SDR_ENC_MASK 0x00001000L +#define VCN_FEATURES__HAS_DUAL_MJPEG_DEC_MASK 0x00002000L +#define VCN_FEATURES__HAS_AV1_ENC_MASK 0x00004000L +#define VCN_FEATURES__INSTANCE_ID_MASK 0xF0000000L +//UVD_GPUIOV_STATUS +#define UVD_GPUIOV_STATUS__UVD_GPUIOV_STATUS_VF_ENABLE__SHIFT 0x0 +#define UVD_GPUIOV_STATUS__UVD_GPUIOV_STATUS_VF_ENABLE_MASK 0x00000001L +//UVD_RAS_VCPU_VCODEC_STATUS +#define UVD_RAS_VCPU_VCODEC_STATUS__POISONED_VF__SHIFT 0x0 +#define UVD_RAS_VCPU_VCODEC_STATUS__POISONED_PF__SHIFT 0x1f +#define UVD_RAS_VCPU_VCODEC_STATUS__POISONED_VF_MASK 0x7FFFFFFFL +#define UVD_RAS_VCPU_VCODEC_STATUS__POISONED_PF_MASK 0x80000000L +//UVD_RAS_MMSCH_FATAL_ERROR +#define UVD_RAS_MMSCH_FATAL_ERROR__POISONED_VF__SHIFT 0x0 +#define UVD_RAS_MMSCH_FATAL_ERROR__POISONED_PF__SHIFT 0x1f +#define UVD_RAS_MMSCH_FATAL_ERROR__POISONED_VF_MASK 0x7FFFFFFFL +#define UVD_RAS_MMSCH_FATAL_ERROR__POISONED_PF_MASK 0x80000000L +//UVD_RAS_JPEG0_STATUS +#define UVD_RAS_JPEG0_STATUS__POISONED_VF__SHIFT 0x0 +#define UVD_RAS_JPEG0_STATUS__POISONED_PF__SHIFT 0x1f +#define UVD_RAS_JPEG0_STATUS__POISONED_VF_MASK 0x7FFFFFFFL +#define UVD_RAS_JPEG0_STATUS__POISONED_PF_MASK 0x80000000L +//UVD_RAS_JPEG1_STATUS +#define UVD_RAS_JPEG1_STATUS__POISONED_VF__SHIFT 0x0 +#define UVD_RAS_JPEG1_STATUS__POISONED_PF__SHIFT 0x1f +#define UVD_RAS_JPEG1_STATUS__POISONED_VF_MASK 0x7FFFFFFFL +#define UVD_RAS_JPEG1_STATUS__POISONED_PF_MASK 0x80000000L +//UVD_RAS_CNTL_PMI_ARB +#define UVD_RAS_CNTL_PMI_ARB__STAT_VCPU_VCODEC__SHIFT 0x0 +#define UVD_RAS_CNTL_PMI_ARB__ACK_VCPU_VCODEC__SHIFT 0x1 +#define UVD_RAS_CNTL_PMI_ARB__STAT_MMSCH__SHIFT 0x2 +#define UVD_RAS_CNTL_PMI_ARB__ACK_MMSCH__SHIFT 0x3 +#define UVD_RAS_CNTL_PMI_ARB__STAT_JPEG0__SHIFT 0x4 +#define UVD_RAS_CNTL_PMI_ARB__ACK_JPEG0__SHIFT 0x5 +#define UVD_RAS_CNTL_PMI_ARB__STAT_JPEG1__SHIFT 0x6 +#define UVD_RAS_CNTL_PMI_ARB__ACK_JPEG1__SHIFT 0x7 +#define UVD_RAS_CNTL_PMI_ARB__STAT_VCPU_VCODEC_MASK 0x00000001L +#define UVD_RAS_CNTL_PMI_ARB__ACK_VCPU_VCODEC_MASK 0x00000002L +#define UVD_RAS_CNTL_PMI_ARB__STAT_MMSCH_MASK 0x00000004L +#define UVD_RAS_CNTL_PMI_ARB__ACK_MMSCH_MASK 0x00000008L +#define UVD_RAS_CNTL_PMI_ARB__STAT_JPEG0_MASK 0x00000010L +#define UVD_RAS_CNTL_PMI_ARB__ACK_JPEG0_MASK 0x00000020L +#define UVD_RAS_CNTL_PMI_ARB__STAT_JPEG1_MASK 0x00000040L +#define UVD_RAS_CNTL_PMI_ARB__ACK_JPEG1_MASK 0x00000080L +//UVD_SCRATCH15 +#define UVD_SCRATCH15__SCRATCH15_DATA__SHIFT 0x0 +#define UVD_SCRATCH15__SCRATCH15_DATA_MASK 0xFFFFFFFFL +//VCN_JPEG_DB_CTRL1 +#define VCN_JPEG_DB_CTRL1__OFFSET__SHIFT 0x2 +#define VCN_JPEG_DB_CTRL1__EN__SHIFT 0x1e +#define VCN_JPEG_DB_CTRL1__HIT__SHIFT 0x1f +#define VCN_JPEG_DB_CTRL1__OFFSET_MASK 0x0FFFFFFCL +#define VCN_JPEG_DB_CTRL1__EN_MASK 0x40000000L +#define VCN_JPEG_DB_CTRL1__HIT_MASK 0x80000000L +//VCN_JPEG_DB_CTRL2 +#define VCN_JPEG_DB_CTRL2__OFFSET__SHIFT 0x2 +#define VCN_JPEG_DB_CTRL2__EN__SHIFT 0x1e +#define VCN_JPEG_DB_CTRL2__HIT__SHIFT 0x1f +#define VCN_JPEG_DB_CTRL2__OFFSET_MASK 0x0FFFFFFCL +#define VCN_JPEG_DB_CTRL2__EN_MASK 0x40000000L +#define VCN_JPEG_DB_CTRL2__HIT_MASK 0x80000000L +//VCN_JPEG_DB_CTRL3 +#define VCN_JPEG_DB_CTRL3__OFFSET__SHIFT 0x2 +#define VCN_JPEG_DB_CTRL3__EN__SHIFT 0x1e +#define VCN_JPEG_DB_CTRL3__HIT__SHIFT 0x1f +#define VCN_JPEG_DB_CTRL3__OFFSET_MASK 0x0FFFFFFCL +#define VCN_JPEG_DB_CTRL3__EN_MASK 0x40000000L +#define VCN_JPEG_DB_CTRL3__HIT_MASK 0x80000000L +//VCN_JPEG_DB_CTRL4 +#define VCN_JPEG_DB_CTRL4__OFFSET__SHIFT 0x2 +#define VCN_JPEG_DB_CTRL4__EN__SHIFT 0x1e +#define VCN_JPEG_DB_CTRL4__HIT__SHIFT 0x1f +#define VCN_JPEG_DB_CTRL4__OFFSET_MASK 0x0FFFFFFCL +#define VCN_JPEG_DB_CTRL4__EN_MASK 0x40000000L +#define VCN_JPEG_DB_CTRL4__HIT_MASK 0x80000000L +//VCN_JPEG_DB_CTRL5 +#define VCN_JPEG_DB_CTRL5__OFFSET__SHIFT 0x2 +#define VCN_JPEG_DB_CTRL5__EN__SHIFT 0x1e +#define VCN_JPEG_DB_CTRL5__HIT__SHIFT 0x1f +#define VCN_JPEG_DB_CTRL5__OFFSET_MASK 0x0FFFFFFCL +#define VCN_JPEG_DB_CTRL5__EN_MASK 0x40000000L +#define VCN_JPEG_DB_CTRL5__HIT_MASK 0x80000000L +//VCN_JPEG_DB_CTRL6 +#define VCN_JPEG_DB_CTRL6__OFFSET__SHIFT 0x2 +#define VCN_JPEG_DB_CTRL6__EN__SHIFT 0x1e +#define VCN_JPEG_DB_CTRL6__HIT__SHIFT 0x1f +#define VCN_JPEG_DB_CTRL6__OFFSET_MASK 0x0FFFFFFCL +#define VCN_JPEG_DB_CTRL6__EN_MASK 0x40000000L +#define VCN_JPEG_DB_CTRL6__HIT_MASK 0x80000000L +//VCN_JPEG_DB_CTRL7 +#define VCN_JPEG_DB_CTRL7__OFFSET__SHIFT 0x2 +#define VCN_JPEG_DB_CTRL7__EN__SHIFT 0x1e +#define VCN_JPEG_DB_CTRL7__HIT__SHIFT 0x1f +#define VCN_JPEG_DB_CTRL7__OFFSET_MASK 0x0FFFFFFCL +#define VCN_JPEG_DB_CTRL7__EN_MASK 0x40000000L +#define VCN_JPEG_DB_CTRL7__HIT_MASK 0x80000000L +//UVD_SCRATCH32 +#define UVD_SCRATCH32__SCRATCH32_DATA__SHIFT 0x0 +#define UVD_SCRATCH32__SCRATCH32_DATA_MASK 0xFFFFFFFFL +//UVD_VERSION +#define UVD_VERSION__VARIANT_TYPE__SHIFT 0x0 +#define UVD_VERSION__MINOR_VERSION__SHIFT 0x8 +#define UVD_VERSION__MAJOR_VERSION__SHIFT 0x10 +#define UVD_VERSION__INSTANCE_ID__SHIFT 0x1c +#define UVD_VERSION__VARIANT_TYPE_MASK 0x000000FFL +#define UVD_VERSION__MINOR_VERSION_MASK 0x0000FF00L +#define UVD_VERSION__MAJOR_VERSION_MASK 0x0FFF0000L +#define UVD_VERSION__INSTANCE_ID_MASK 0xF0000000L +//VCN_RB_DB_CTRL +#define VCN_RB_DB_CTRL__OFFSET__SHIFT 0x2 +#define VCN_RB_DB_CTRL__EN__SHIFT 0x1e +#define VCN_RB_DB_CTRL__HIT__SHIFT 0x1f +#define VCN_RB_DB_CTRL__OFFSET_MASK 0x0FFFFFFCL +#define VCN_RB_DB_CTRL__EN_MASK 0x40000000L +#define VCN_RB_DB_CTRL__HIT_MASK 0x80000000L +//VCN_JPEG_DB_CTRL +#define VCN_JPEG_DB_CTRL__OFFSET__SHIFT 0x2 +#define VCN_JPEG_DB_CTRL__EN__SHIFT 0x1e +#define VCN_JPEG_DB_CTRL__HIT__SHIFT 0x1f +#define VCN_JPEG_DB_CTRL__OFFSET_MASK 0x0FFFFFFCL +#define VCN_JPEG_DB_CTRL__EN_MASK 0x40000000L +#define VCN_JPEG_DB_CTRL__HIT_MASK 0x80000000L +//VCN_RB1_DB_CTRL +#define VCN_RB1_DB_CTRL__OFFSET__SHIFT 0x2 +#define VCN_RB1_DB_CTRL__EN__SHIFT 0x1e +#define VCN_RB1_DB_CTRL__HIT__SHIFT 0x1f +#define VCN_RB1_DB_CTRL__OFFSET_MASK 0x0FFFFFFCL +#define VCN_RB1_DB_CTRL__EN_MASK 0x40000000L +#define VCN_RB1_DB_CTRL__HIT_MASK 0x80000000L +//VCN_RB2_DB_CTRL +#define VCN_RB2_DB_CTRL__OFFSET__SHIFT 0x2 +#define VCN_RB2_DB_CTRL__EN__SHIFT 0x1e +#define VCN_RB2_DB_CTRL__HIT__SHIFT 0x1f +#define VCN_RB2_DB_CTRL__OFFSET_MASK 0x0FFFFFFCL +#define VCN_RB2_DB_CTRL__EN_MASK 0x40000000L +#define VCN_RB2_DB_CTRL__HIT_MASK 0x80000000L +//VCN_RB3_DB_CTRL +#define VCN_RB3_DB_CTRL__OFFSET__SHIFT 0x2 +#define VCN_RB3_DB_CTRL__EN__SHIFT 0x1e +#define VCN_RB3_DB_CTRL__HIT__SHIFT 0x1f +#define VCN_RB3_DB_CTRL__OFFSET_MASK 0x0FFFFFFCL +#define VCN_RB3_DB_CTRL__EN_MASK 0x40000000L +#define VCN_RB3_DB_CTRL__HIT_MASK 0x80000000L +//VCN_RB4_DB_CTRL +#define VCN_RB4_DB_CTRL__OFFSET__SHIFT 0x2 +#define VCN_RB4_DB_CTRL__EN__SHIFT 0x1e +#define VCN_RB4_DB_CTRL__HIT__SHIFT 0x1f +#define VCN_RB4_DB_CTRL__OFFSET_MASK 0x0FFFFFFCL +#define VCN_RB4_DB_CTRL__EN_MASK 0x40000000L +#define VCN_RB4_DB_CTRL__HIT_MASK 0x80000000L +//VCN_RB_ENABLE +#define VCN_RB_ENABLE__RB_EN__SHIFT 0x0 +#define VCN_RB_ENABLE__JPEG_RB_EN__SHIFT 0x1 +#define VCN_RB_ENABLE__RB1_EN__SHIFT 0x2 +#define VCN_RB_ENABLE__RB2_EN__SHIFT 0x3 +#define VCN_RB_ENABLE__RB3_EN__SHIFT 0x4 +#define VCN_RB_ENABLE__RB4_EN__SHIFT 0x5 +#define VCN_RB_ENABLE__UMSCH_RB_EN__SHIFT 0x6 +#define VCN_RB_ENABLE__EJPEG_RB_EN__SHIFT 0x7 +#define VCN_RB_ENABLE__AUDIO_RB_EN__SHIFT 0x8 +#define VCN_RB_ENABLE__RB_EN_MASK 0x00000001L +#define VCN_RB_ENABLE__JPEG_RB_EN_MASK 0x00000002L +#define VCN_RB_ENABLE__RB1_EN_MASK 0x00000004L +#define VCN_RB_ENABLE__RB2_EN_MASK 0x00000008L +#define VCN_RB_ENABLE__RB3_EN_MASK 0x00000010L +#define VCN_RB_ENABLE__RB4_EN_MASK 0x00000020L +#define VCN_RB_ENABLE__UMSCH_RB_EN_MASK 0x00000040L +#define VCN_RB_ENABLE__EJPEG_RB_EN_MASK 0x00000080L +#define VCN_RB_ENABLE__AUDIO_RB_EN_MASK 0x00000100L +//VCN_RB_WPTR_CTRL +#define VCN_RB_WPTR_CTRL__RB_CS_EN__SHIFT 0x0 +#define VCN_RB_WPTR_CTRL__JPEG_CS_EN__SHIFT 0x1 +#define VCN_RB_WPTR_CTRL__RB1_CS_EN__SHIFT 0x2 +#define VCN_RB_WPTR_CTRL__RB2_CS_EN__SHIFT 0x3 +#define VCN_RB_WPTR_CTRL__RB3_CS_EN__SHIFT 0x4 +#define VCN_RB_WPTR_CTRL__RB4_CS_EN__SHIFT 0x5 +#define VCN_RB_WPTR_CTRL__UMSCH_RB_CS_EN__SHIFT 0x6 +#define VCN_RB_WPTR_CTRL__EJPEG_RB_CS_EN__SHIFT 0x7 +#define VCN_RB_WPTR_CTRL__AUDIO_RB_CS_EN__SHIFT 0x8 +#define VCN_RB_WPTR_CTRL__RB_CS_EN_MASK 0x00000001L +#define VCN_RB_WPTR_CTRL__JPEG_CS_EN_MASK 0x00000002L +#define VCN_RB_WPTR_CTRL__RB1_CS_EN_MASK 0x00000004L +#define VCN_RB_WPTR_CTRL__RB2_CS_EN_MASK 0x00000008L +#define VCN_RB_WPTR_CTRL__RB3_CS_EN_MASK 0x00000010L +#define VCN_RB_WPTR_CTRL__RB4_CS_EN_MASK 0x00000020L +#define VCN_RB_WPTR_CTRL__UMSCH_RB_CS_EN_MASK 0x00000040L +#define VCN_RB_WPTR_CTRL__EJPEG_RB_CS_EN_MASK 0x00000080L +#define VCN_RB_WPTR_CTRL__AUDIO_RB_CS_EN_MASK 0x00000100L +//UVD_RB_RPTR +#define UVD_RB_RPTR__RB_RPTR__SHIFT 0x4 +#define UVD_RB_RPTR__RB_RPTR_MASK 0x007FFFF0L +//UVD_RB_WPTR +#define UVD_RB_WPTR__RB_WPTR__SHIFT 0x4 +#define UVD_RB_WPTR__RB_WPTR_MASK 0x007FFFF0L +//UVD_RB_RPTR2 +#define UVD_RB_RPTR2__RB_RPTR__SHIFT 0x4 +#define UVD_RB_RPTR2__RB_RPTR_MASK 0x007FFFF0L +//UVD_RB_WPTR2 +#define UVD_RB_WPTR2__RB_WPTR__SHIFT 0x4 +#define UVD_RB_WPTR2__RB_WPTR_MASK 0x007FFFF0L +//UVD_RB_RPTR3 +#define UVD_RB_RPTR3__RB_RPTR__SHIFT 0x4 +#define UVD_RB_RPTR3__RB_RPTR_MASK 0x007FFFF0L +//UVD_RB_WPTR3 +#define UVD_RB_WPTR3__RB_WPTR__SHIFT 0x4 +#define UVD_RB_WPTR3__RB_WPTR_MASK 0x007FFFF0L +//UVD_RB_RPTR4 +#define UVD_RB_RPTR4__RB_RPTR__SHIFT 0x4 +#define UVD_RB_RPTR4__RB_RPTR_MASK 0x007FFFF0L +//UVD_RB_WPTR4 +#define UVD_RB_WPTR4__RB_WPTR__SHIFT 0x4 +#define UVD_RB_WPTR4__RB_WPTR_MASK 0x007FFFF0L +//UVD_OUT_RB_RPTR +#define UVD_OUT_RB_RPTR__RB_RPTR__SHIFT 0x4 +#define UVD_OUT_RB_RPTR__RB_RPTR_MASK 0x007FFFF0L +//UVD_OUT_RB_WPTR +#define UVD_OUT_RB_WPTR__RB_WPTR__SHIFT 0x4 +#define UVD_OUT_RB_WPTR__RB_WPTR_MASK 0x007FFFF0L +//UVD_AUDIO_RB_RPTR +#define UVD_AUDIO_RB_RPTR__RB_RPTR__SHIFT 0x4 +#define UVD_AUDIO_RB_RPTR__RB_RPTR_MASK 0x007FFFF0L +//UVD_AUDIO_RB_WPTR +#define UVD_AUDIO_RB_WPTR__RB_WPTR__SHIFT 0x4 +#define UVD_AUDIO_RB_WPTR__RB_WPTR_MASK 0x007FFFF0L +//UVD_RBC_RB_RPTR +#define UVD_RBC_RB_RPTR__RB_RPTR__SHIFT 0x4 +#define UVD_RBC_RB_RPTR__RB_RPTR_MASK 0x007FFFF0L +//UVD_RBC_RB_WPTR +#define UVD_RBC_RB_WPTR__RB_WPTR__SHIFT 0x4 +#define UVD_RBC_RB_WPTR__RB_WPTR_MASK 0x007FFFF0L +//UVD_DPG_LMA_CTL2 +#define UVD_DPG_LMA_CTL2__DIRECT_ACCESS_SRAM_SEL__SHIFT 0x0 +#define UVD_DPG_LMA_CTL2__FIFO_DIRECT_ACCESS_EN__SHIFT 0x1 +#define UVD_DPG_LMA_CTL2__VID_WRITE_PTR__SHIFT 0x2 +#define UVD_DPG_LMA_CTL2__JPEG_WRITE_PTR__SHIFT 0x9 +#define UVD_DPG_LMA_CTL2__DIRECT_ACCESS_SRAM_SEL_MASK 0x00000001L +#define UVD_DPG_LMA_CTL2__FIFO_DIRECT_ACCESS_EN_MASK 0x00000002L +#define UVD_DPG_LMA_CTL2__VID_WRITE_PTR_MASK 0x000001FCL +#define UVD_DPG_LMA_CTL2__JPEG_WRITE_PTR_MASK 0x0000FE00L + + +// addressBlock: aid_uvd0_mmsch_dec +//MMSCH_UCODE_ADDR +#define MMSCH_UCODE_ADDR__UCODE_ADDR__SHIFT 0x2 +#define MMSCH_UCODE_ADDR__UCODE_LOCK__SHIFT 0x1f +#define MMSCH_UCODE_ADDR__UCODE_ADDR_MASK 0x00003FFCL +#define MMSCH_UCODE_ADDR__UCODE_LOCK_MASK 0x80000000L +//MMSCH_UCODE_DATA +#define MMSCH_UCODE_DATA__UCODE_DATA__SHIFT 0x0 +#define MMSCH_UCODE_DATA__UCODE_DATA_MASK 0xFFFFFFFFL +//MMSCH_SRAM_ADDR +#define MMSCH_SRAM_ADDR__SRAM_ADDR__SHIFT 0x2 +#define MMSCH_SRAM_ADDR__SRAM_LOCK__SHIFT 0x1f +#define MMSCH_SRAM_ADDR__SRAM_ADDR_MASK 0x00001FFCL +#define MMSCH_SRAM_ADDR__SRAM_LOCK_MASK 0x80000000L +//MMSCH_SRAM_DATA +#define MMSCH_SRAM_DATA__SRAM_DATA__SHIFT 0x0 +#define MMSCH_SRAM_DATA__SRAM_DATA_MASK 0xFFFFFFFFL +//MMSCH_VF_SRAM_OFFSET +#define MMSCH_VF_SRAM_OFFSET__VF_SRAM_OFFSET__SHIFT 0x2 +#define MMSCH_VF_SRAM_OFFSET__VF_SRAM_NUM_DW_PER_VF__SHIFT 0x10 +#define MMSCH_VF_SRAM_OFFSET__VF_SRAM_OFFSET_MASK 0x00001FFCL +#define MMSCH_VF_SRAM_OFFSET__VF_SRAM_NUM_DW_PER_VF_MASK 0x00FF0000L +//MMSCH_DB_SRAM_OFFSET +#define MMSCH_DB_SRAM_OFFSET__DB_SRAM_OFFSET__SHIFT 0x2 +#define MMSCH_DB_SRAM_OFFSET__DB_SRAM_NUM_ENG__SHIFT 0x10 +#define MMSCH_DB_SRAM_OFFSET__DB_SRAM_NUM_RING_PER_ENG__SHIFT 0x18 +#define MMSCH_DB_SRAM_OFFSET__DB_SRAM_OFFSET_MASK 0x00001FFCL +#define MMSCH_DB_SRAM_OFFSET__DB_SRAM_NUM_ENG_MASK 0x00FF0000L +#define MMSCH_DB_SRAM_OFFSET__DB_SRAM_NUM_RING_PER_ENG_MASK 0xFF000000L +//MMSCH_CTX_SRAM_OFFSET +#define MMSCH_CTX_SRAM_OFFSET__CTX_SRAM_OFFSET__SHIFT 0x2 +#define MMSCH_CTX_SRAM_OFFSET__CTX_SRAM_SIZE__SHIFT 0x10 +#define MMSCH_CTX_SRAM_OFFSET__CTX_SRAM_OFFSET_MASK 0x00001FFCL +#define MMSCH_CTX_SRAM_OFFSET__CTX_SRAM_SIZE_MASK 0xFFFF0000L +//MMSCH_CTL +#define MMSCH_CTL__P_RUNSTALL__SHIFT 0x0 +#define MMSCH_CTL__P_RESET__SHIFT 0x1 +#define MMSCH_CTL__VFID_FIFO_EN__SHIFT 0x4 +#define MMSCH_CTL__P_LOCK__SHIFT 0x1f +#define MMSCH_CTL__P_RUNSTALL_MASK 0x00000001L +#define MMSCH_CTL__P_RESET_MASK 0x00000002L +#define MMSCH_CTL__VFID_FIFO_EN_MASK 0x00000010L +#define MMSCH_CTL__P_LOCK_MASK 0x80000000L +//MMSCH_INTR +#define MMSCH_INTR__INTR__SHIFT 0x0 +#define MMSCH_INTR__INTR_MASK 0x00001FFFL +//MMSCH_INTR_ACK +#define MMSCH_INTR_ACK__INTR__SHIFT 0x0 +#define MMSCH_INTR_ACK__INTR_MASK 0x00001FFFL +//MMSCH_INTR_STATUS +#define MMSCH_INTR_STATUS__INTR__SHIFT 0x0 +#define MMSCH_INTR_STATUS__INTR_MASK 0x00001FFFL +//MMSCH_VF_VMID +#define MMSCH_VF_VMID__VF_CTX_VMID__SHIFT 0x0 +#define MMSCH_VF_VMID__VF_GPCOM_VMID__SHIFT 0x5 +#define MMSCH_VF_VMID__VF_CTX_VMID_MASK 0x0000001FL +#define MMSCH_VF_VMID__VF_GPCOM_VMID_MASK 0x000003E0L +//MMSCH_VF_CTX_ADDR_LO +#define MMSCH_VF_CTX_ADDR_LO__VF_CTX_ADDR_LO__SHIFT 0x6 +#define MMSCH_VF_CTX_ADDR_LO__VF_CTX_ADDR_LO_MASK 0xFFFFFFC0L +//MMSCH_VF_CTX_ADDR_HI +#define MMSCH_VF_CTX_ADDR_HI__VF_CTX_ADDR_HI__SHIFT 0x0 +#define MMSCH_VF_CTX_ADDR_HI__VF_CTX_ADDR_HI_MASK 0xFFFFFFFFL +//MMSCH_VF_CTX_SIZE +#define MMSCH_VF_CTX_SIZE__VF_CTX_SIZE__SHIFT 0x0 +#define MMSCH_VF_CTX_SIZE__VF_CTX_SIZE_MASK 0xFFFFFFFFL +//MMSCH_VF_GPCOM_ADDR_LO +#define MMSCH_VF_GPCOM_ADDR_LO__VF_GPCOM_ADDR_LO__SHIFT 0x6 +#define MMSCH_VF_GPCOM_ADDR_LO__VF_GPCOM_ADDR_LO_MASK 0xFFFFFFC0L +//MMSCH_VF_GPCOM_ADDR_HI +#define MMSCH_VF_GPCOM_ADDR_HI__VF_GPCOM_ADDR_HI__SHIFT 0x0 +#define MMSCH_VF_GPCOM_ADDR_HI__VF_GPCOM_ADDR_HI_MASK 0xFFFFFFFFL +//MMSCH_VF_GPCOM_SIZE +#define MMSCH_VF_GPCOM_SIZE__VF_GPCOM_SIZE__SHIFT 0x0 +#define MMSCH_VF_GPCOM_SIZE__VF_GPCOM_SIZE_MASK 0xFFFFFFFFL +//MMSCH_VF_MAILBOX_HOST +#define MMSCH_VF_MAILBOX_HOST__DATA__SHIFT 0x0 +#define MMSCH_VF_MAILBOX_HOST__DATA_MASK 0xFFFFFFFFL +//MMSCH_VF_MAILBOX_RESP +#define MMSCH_VF_MAILBOX_RESP__RESP__SHIFT 0x0 +#define MMSCH_VF_MAILBOX_RESP__RESP_MASK 0xFFFFFFFFL +//MMSCH_VF_MAILBOX_0 +#define MMSCH_VF_MAILBOX_0__DATA__SHIFT 0x0 +#define MMSCH_VF_MAILBOX_0__DATA_MASK 0xFFFFFFFFL +//MMSCH_VF_MAILBOX_0_RESP +#define MMSCH_VF_MAILBOX_0_RESP__RESP__SHIFT 0x0 +#define MMSCH_VF_MAILBOX_0_RESP__RESP_MASK 0xFFFFFFFFL +//MMSCH_VF_MAILBOX_1 +#define MMSCH_VF_MAILBOX_1__DATA__SHIFT 0x0 +#define MMSCH_VF_MAILBOX_1__DATA_MASK 0xFFFFFFFFL +//MMSCH_VF_MAILBOX_1_RESP +#define MMSCH_VF_MAILBOX_1_RESP__RESP__SHIFT 0x0 +#define MMSCH_VF_MAILBOX_1_RESP__RESP_MASK 0xFFFFFFFFL +//MMSCH_CNTL +#define MMSCH_CNTL__CLK_EN__SHIFT 0x0 +#define MMSCH_CNTL__ED_ENABLE__SHIFT 0x1 +#define MMSCH_CNTL__AXI_MAX_BRST_SIZE_IS_4__SHIFT 0x2 +#define MMSCH_CNTL__AXI_40BIT_PIF_ADDR_FIX_EN__SHIFT 0x3 +#define MMSCH_CNTL__PDEBUG_ENABLE__SHIFT 0x4 +#define MMSCH_CNTL__MMSCH_IRQ_ERR__SHIFT 0x5 +#define MMSCH_CNTL__MMSCH_NACK_INTR_EN__SHIFT 0x9 +#define MMSCH_CNTL__MMSCH_DB_BUSY_INTR_EN__SHIFT 0xa +#define MMSCH_CNTL__PRB_TIMEOUT_VAL__SHIFT 0x14 +#define MMSCH_CNTL__TIMEOUT_DIS__SHIFT 0x1c +#define MMSCH_CNTL__MMSCH_IDLE__SHIFT 0x1d +#define MMSCH_CNTL__CLK_EN_MASK 0x00000001L +#define MMSCH_CNTL__ED_ENABLE_MASK 0x00000002L +#define MMSCH_CNTL__AXI_MAX_BRST_SIZE_IS_4_MASK 0x00000004L +#define MMSCH_CNTL__AXI_40BIT_PIF_ADDR_FIX_EN_MASK 0x00000008L +#define MMSCH_CNTL__PDEBUG_ENABLE_MASK 0x00000010L +#define MMSCH_CNTL__MMSCH_IRQ_ERR_MASK 0x000001E0L +#define MMSCH_CNTL__MMSCH_NACK_INTR_EN_MASK 0x00000200L +#define MMSCH_CNTL__MMSCH_DB_BUSY_INTR_EN_MASK 0x00000400L +#define MMSCH_CNTL__PRB_TIMEOUT_VAL_MASK 0x0FF00000L +#define MMSCH_CNTL__TIMEOUT_DIS_MASK 0x10000000L +#define MMSCH_CNTL__MMSCH_IDLE_MASK 0x20000000L +//MMSCH_NONCACHE_OFFSET0 +#define MMSCH_NONCACHE_OFFSET0__OFFSET__SHIFT 0x0 +#define MMSCH_NONCACHE_OFFSET0__OFFSET_MASK 0x0FFFFFFFL +//MMSCH_NONCACHE_SIZE0 +#define MMSCH_NONCACHE_SIZE0__SIZE__SHIFT 0x0 +#define MMSCH_NONCACHE_SIZE0__SIZE_MASK 0x00FFFFFFL +//MMSCH_NONCACHE_OFFSET1 +#define MMSCH_NONCACHE_OFFSET1__OFFSET__SHIFT 0x0 +#define MMSCH_NONCACHE_OFFSET1__OFFSET_MASK 0x0FFFFFFFL +//MMSCH_NONCACHE_SIZE1 +#define MMSCH_NONCACHE_SIZE1__SIZE__SHIFT 0x0 +#define MMSCH_NONCACHE_SIZE1__SIZE_MASK 0x00FFFFFFL +//MMSCH_PROC_STATE1 +#define MMSCH_PROC_STATE1__PC__SHIFT 0x0 +#define MMSCH_PROC_STATE1__PC_MASK 0xFFFFFFFFL +//MMSCH_LAST_MC_ADDR +#define MMSCH_LAST_MC_ADDR__MC_ADDR__SHIFT 0x0 +#define MMSCH_LAST_MC_ADDR__RW__SHIFT 0x1f +#define MMSCH_LAST_MC_ADDR__MC_ADDR_MASK 0x0FFFFFFFL +#define MMSCH_LAST_MC_ADDR__RW_MASK 0x80000000L +//MMSCH_LAST_MEM_ACCESS_HI +#define MMSCH_LAST_MEM_ACCESS_HI__PROC_CMD__SHIFT 0x0 +#define MMSCH_LAST_MEM_ACCESS_HI__FIFO_RPTR__SHIFT 0x8 +#define MMSCH_LAST_MEM_ACCESS_HI__FIFO_WPTR__SHIFT 0xc +#define MMSCH_LAST_MEM_ACCESS_HI__PROC_CMD_MASK 0x00000007L +#define MMSCH_LAST_MEM_ACCESS_HI__FIFO_RPTR_MASK 0x00000700L +#define MMSCH_LAST_MEM_ACCESS_HI__FIFO_WPTR_MASK 0x00007000L +//MMSCH_LAST_MEM_ACCESS_LO +#define MMSCH_LAST_MEM_ACCESS_LO__PROC_ADDR__SHIFT 0x0 +#define MMSCH_LAST_MEM_ACCESS_LO__PROC_ADDR_MASK 0xFFFFFFFFL +//MMSCH_IOV_ACTIVE_FCN_ID +#define MMSCH_IOV_ACTIVE_FCN_ID__ACTIVE_VF_ID__SHIFT 0x0 +#define MMSCH_IOV_ACTIVE_FCN_ID__ACTIVE_PF_VF__SHIFT 0x1f +#define MMSCH_IOV_ACTIVE_FCN_ID__ACTIVE_VF_ID_MASK 0x0000001FL +#define MMSCH_IOV_ACTIVE_FCN_ID__ACTIVE_PF_VF_MASK 0x80000000L +//MMSCH_SCRATCH_0 +#define MMSCH_SCRATCH_0__SCRATCH_0__SHIFT 0x0 +#define MMSCH_SCRATCH_0__SCRATCH_0_MASK 0xFFFFFFFFL +//MMSCH_SCRATCH_1 +#define MMSCH_SCRATCH_1__SCRATCH_1__SHIFT 0x0 +#define MMSCH_SCRATCH_1__SCRATCH_1_MASK 0xFFFFFFFFL +//MMSCH_GPUIOV_SCH_BLOCK_0 +#define MMSCH_GPUIOV_SCH_BLOCK_0__ID__SHIFT 0x0 +#define MMSCH_GPUIOV_SCH_BLOCK_0__VERSION__SHIFT 0x4 +#define MMSCH_GPUIOV_SCH_BLOCK_0__SIZE__SHIFT 0x8 +#define MMSCH_GPUIOV_SCH_BLOCK_0__ID_MASK 0x0000000FL +#define MMSCH_GPUIOV_SCH_BLOCK_0__VERSION_MASK 0x000000F0L +#define MMSCH_GPUIOV_SCH_BLOCK_0__SIZE_MASK 0x0000FF00L +//MMSCH_GPUIOV_CMD_CONTROL_0 +#define MMSCH_GPUIOV_CMD_CONTROL_0__CMD_TYPE__SHIFT 0x0 +#define MMSCH_GPUIOV_CMD_CONTROL_0__CMD_EXECUTE__SHIFT 0x4 +#define MMSCH_GPUIOV_CMD_CONTROL_0__CMD_EXECUTE_INTR_EN__SHIFT 0x5 +#define MMSCH_GPUIOV_CMD_CONTROL_0__VM_BUSY_INTR_EN__SHIFT 0x6 +#define MMSCH_GPUIOV_CMD_CONTROL_0__FUNCTINO_ID__SHIFT 0x8 +#define MMSCH_GPUIOV_CMD_CONTROL_0__NEXT_FUNCTINO_ID__SHIFT 0x10 +#define MMSCH_GPUIOV_CMD_CONTROL_0__CMD_TYPE_MASK 0x0000000FL +#define MMSCH_GPUIOV_CMD_CONTROL_0__CMD_EXECUTE_MASK 0x00000010L +#define MMSCH_GPUIOV_CMD_CONTROL_0__CMD_EXECUTE_INTR_EN_MASK 0x00000020L +#define MMSCH_GPUIOV_CMD_CONTROL_0__VM_BUSY_INTR_EN_MASK 0x00000040L +#define MMSCH_GPUIOV_CMD_CONTROL_0__FUNCTINO_ID_MASK 0x0000FF00L +#define MMSCH_GPUIOV_CMD_CONTROL_0__NEXT_FUNCTINO_ID_MASK 0x00FF0000L +//MMSCH_GPUIOV_CMD_STATUS_0 +#define MMSCH_GPUIOV_CMD_STATUS_0__CMD_STATUS__SHIFT 0x0 +#define MMSCH_GPUIOV_CMD_STATUS_0__CMD_STATUS_MASK 0x0000000FL +//MMSCH_GPUIOV_VM_BUSY_STATUS_0 +#define MMSCH_GPUIOV_VM_BUSY_STATUS_0__BUSY__SHIFT 0x0 +#define MMSCH_GPUIOV_VM_BUSY_STATUS_0__BUSY_MASK 0xFFFFFFFFL +//MMSCH_GPUIOV_ACTIVE_FCNS_0 +#define MMSCH_GPUIOV_ACTIVE_FCNS_0__ACTIVE_FCNS__SHIFT 0x0 +#define MMSCH_GPUIOV_ACTIVE_FCNS_0__ACTIVE_FCNS_MASK 0xFFFFFFFFL +//MMSCH_GPUIOV_ACTIVE_FCN_ID_0 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_0__ID__SHIFT 0x0 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_0__ID_STATUS__SHIFT 0x8 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_0__ID_MASK 0x000000FFL +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_0__ID_STATUS_MASK 0x00000F00L +//MMSCH_GPUIOV_DW6_0 +#define MMSCH_GPUIOV_DW6_0__DATA__SHIFT 0x0 +#define MMSCH_GPUIOV_DW6_0__DATA_MASK 0xFFFFFFFFL +//MMSCH_GPUIOV_DW7_0 +#define MMSCH_GPUIOV_DW7_0__DATA__SHIFT 0x0 +#define MMSCH_GPUIOV_DW7_0__DATA_MASK 0xFFFFFFFFL +//MMSCH_GPUIOV_DW8_0 +#define MMSCH_GPUIOV_DW8_0__DATA__SHIFT 0x0 +#define MMSCH_GPUIOV_DW8_0__DATA_MASK 0xFFFFFFFFL +//MMSCH_GPUIOV_SCH_BLOCK_1 +#define MMSCH_GPUIOV_SCH_BLOCK_1__ID__SHIFT 0x0 +#define MMSCH_GPUIOV_SCH_BLOCK_1__VERSION__SHIFT 0x4 +#define MMSCH_GPUIOV_SCH_BLOCK_1__SIZE__SHIFT 0x8 +#define MMSCH_GPUIOV_SCH_BLOCK_1__ID_MASK 0x0000000FL +#define MMSCH_GPUIOV_SCH_BLOCK_1__VERSION_MASK 0x000000F0L +#define MMSCH_GPUIOV_SCH_BLOCK_1__SIZE_MASK 0x0000FF00L +//MMSCH_GPUIOV_CMD_CONTROL_1 +#define MMSCH_GPUIOV_CMD_CONTROL_1__CMD_TYPE__SHIFT 0x0 +#define MMSCH_GPUIOV_CMD_CONTROL_1__CMD_EXECUTE__SHIFT 0x4 +#define MMSCH_GPUIOV_CMD_CONTROL_1__CMD_EXECUTE_INTR_EN__SHIFT 0x5 +#define MMSCH_GPUIOV_CMD_CONTROL_1__VM_BUSY_INTR_EN__SHIFT 0x6 +#define MMSCH_GPUIOV_CMD_CONTROL_1__FUNCTINO_ID__SHIFT 0x8 +#define MMSCH_GPUIOV_CMD_CONTROL_1__NEXT_FUNCTINO_ID__SHIFT 0x10 +#define MMSCH_GPUIOV_CMD_CONTROL_1__CMD_TYPE_MASK 0x0000000FL +#define MMSCH_GPUIOV_CMD_CONTROL_1__CMD_EXECUTE_MASK 0x00000010L +#define MMSCH_GPUIOV_CMD_CONTROL_1__CMD_EXECUTE_INTR_EN_MASK 0x00000020L +#define MMSCH_GPUIOV_CMD_CONTROL_1__VM_BUSY_INTR_EN_MASK 0x00000040L +#define MMSCH_GPUIOV_CMD_CONTROL_1__FUNCTINO_ID_MASK 0x0000FF00L +#define MMSCH_GPUIOV_CMD_CONTROL_1__NEXT_FUNCTINO_ID_MASK 0x00FF0000L +//MMSCH_GPUIOV_CMD_STATUS_1 +#define MMSCH_GPUIOV_CMD_STATUS_1__CMD_STATUS__SHIFT 0x0 +#define MMSCH_GPUIOV_CMD_STATUS_1__CMD_STATUS_MASK 0x0000000FL +//MMSCH_GPUIOV_VM_BUSY_STATUS_1 +#define MMSCH_GPUIOV_VM_BUSY_STATUS_1__BUSY__SHIFT 0x0 +#define MMSCH_GPUIOV_VM_BUSY_STATUS_1__BUSY_MASK 0xFFFFFFFFL +//MMSCH_GPUIOV_ACTIVE_FCNS_1 +#define MMSCH_GPUIOV_ACTIVE_FCNS_1__ACTIVE_FCNS__SHIFT 0x0 +#define MMSCH_GPUIOV_ACTIVE_FCNS_1__ACTIVE_FCNS_MASK 0xFFFFFFFFL +//MMSCH_GPUIOV_ACTIVE_FCN_ID_1 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_1__ID__SHIFT 0x0 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_1__ID_STATUS__SHIFT 0x8 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_1__ID_MASK 0x000000FFL +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_1__ID_STATUS_MASK 0x00000F00L +//MMSCH_GPUIOV_DW6_1 +#define MMSCH_GPUIOV_DW6_1__DATA__SHIFT 0x0 +#define MMSCH_GPUIOV_DW6_1__DATA_MASK 0xFFFFFFFFL +//MMSCH_GPUIOV_DW7_1 +#define MMSCH_GPUIOV_DW7_1__DATA__SHIFT 0x0 +#define MMSCH_GPUIOV_DW7_1__DATA_MASK 0xFFFFFFFFL +//MMSCH_GPUIOV_DW8_1 +#define MMSCH_GPUIOV_DW8_1__DATA__SHIFT 0x0 +#define MMSCH_GPUIOV_DW8_1__DATA_MASK 0xFFFFFFFFL +//MMSCH_GPUIOV_CNTXT +#define MMSCH_GPUIOV_CNTXT__CNTXT_SIZE__SHIFT 0x0 +#define MMSCH_GPUIOV_CNTXT__CNTXT_LOCATION__SHIFT 0x7 +#define MMSCH_GPUIOV_CNTXT__CNTXT_OFFSET__SHIFT 0xa +#define MMSCH_GPUIOV_CNTXT__CNTXT_SIZE_MASK 0x0000007FL +#define MMSCH_GPUIOV_CNTXT__CNTXT_LOCATION_MASK 0x00000080L +#define MMSCH_GPUIOV_CNTXT__CNTXT_OFFSET_MASK 0xFFFFFC00L +//MMSCH_SCRATCH_2 +#define MMSCH_SCRATCH_2__SCRATCH_2__SHIFT 0x0 +#define MMSCH_SCRATCH_2__SCRATCH_2_MASK 0xFFFFFFFFL +//MMSCH_SCRATCH_3 +#define MMSCH_SCRATCH_3__SCRATCH_3__SHIFT 0x0 +#define MMSCH_SCRATCH_3__SCRATCH_3_MASK 0xFFFFFFFFL +//MMSCH_SCRATCH_4 +#define MMSCH_SCRATCH_4__SCRATCH_4__SHIFT 0x0 +#define MMSCH_SCRATCH_4__SCRATCH_4_MASK 0xFFFFFFFFL +//MMSCH_SCRATCH_5 +#define MMSCH_SCRATCH_5__SCRATCH_5__SHIFT 0x0 +#define MMSCH_SCRATCH_5__SCRATCH_5_MASK 0xFFFFFFFFL +//MMSCH_SCRATCH_6 +#define MMSCH_SCRATCH_6__SCRATCH_6__SHIFT 0x0 +#define MMSCH_SCRATCH_6__SCRATCH_6_MASK 0xFFFFFFFFL +//MMSCH_SCRATCH_7 +#define MMSCH_SCRATCH_7__SCRATCH_7__SHIFT 0x0 +#define MMSCH_SCRATCH_7__SCRATCH_7_MASK 0xFFFFFFFFL +//MMSCH_VFID_FIFO_HEAD_0 +#define MMSCH_VFID_FIFO_HEAD_0__HEAD__SHIFT 0x0 +#define MMSCH_VFID_FIFO_HEAD_0__HEAD_MASK 0x0000003FL +//MMSCH_VFID_FIFO_TAIL_0 +#define MMSCH_VFID_FIFO_TAIL_0__TAIL__SHIFT 0x0 +#define MMSCH_VFID_FIFO_TAIL_0__TAIL_MASK 0x0000003FL +//MMSCH_VFID_FIFO_HEAD_1 +#define MMSCH_VFID_FIFO_HEAD_1__HEAD__SHIFT 0x0 +#define MMSCH_VFID_FIFO_HEAD_1__HEAD_MASK 0x0000003FL +//MMSCH_VFID_FIFO_TAIL_1 +#define MMSCH_VFID_FIFO_TAIL_1__TAIL__SHIFT 0x0 +#define MMSCH_VFID_FIFO_TAIL_1__TAIL_MASK 0x0000003FL +//MMSCH_NACK_STATUS +#define MMSCH_NACK_STATUS__WR_NACK_STATUS__SHIFT 0x0 +#define MMSCH_NACK_STATUS__RD_NACK_STATUS__SHIFT 0x2 +#define MMSCH_NACK_STATUS__WR_NACK_STATUS_MASK 0x00000003L +#define MMSCH_NACK_STATUS__RD_NACK_STATUS_MASK 0x0000000CL +//MMSCH_VF_MAILBOX0_DATA +#define MMSCH_VF_MAILBOX0_DATA__DATA__SHIFT 0x0 +#define MMSCH_VF_MAILBOX0_DATA__DATA_MASK 0xFFFFFFFFL +//MMSCH_VF_MAILBOX1_DATA +#define MMSCH_VF_MAILBOX1_DATA__DATA__SHIFT 0x0 +#define MMSCH_VF_MAILBOX1_DATA__DATA_MASK 0xFFFFFFFFL +//MMSCH_GPUIOV_SCH_BLOCK_IP_0 +#define MMSCH_GPUIOV_SCH_BLOCK_IP_0__ID__SHIFT 0x0 +#define MMSCH_GPUIOV_SCH_BLOCK_IP_0__VERSION__SHIFT 0x4 +#define MMSCH_GPUIOV_SCH_BLOCK_IP_0__SIZE__SHIFT 0x8 +#define MMSCH_GPUIOV_SCH_BLOCK_IP_0__ID_MASK 0x0000000FL +#define MMSCH_GPUIOV_SCH_BLOCK_IP_0__VERSION_MASK 0x000000F0L +#define MMSCH_GPUIOV_SCH_BLOCK_IP_0__SIZE_MASK 0x0000FF00L +//MMSCH_GPUIOV_CMD_STATUS_IP_0 +#define MMSCH_GPUIOV_CMD_STATUS_IP_0__CMD_STATUS__SHIFT 0x0 +#define MMSCH_GPUIOV_CMD_STATUS_IP_0__CMD_STATUS_MASK 0x0000000FL +//MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0__ID__SHIFT 0x0 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0__ID_STATUS__SHIFT 0x8 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0__ID_MASK 0x000000FFL +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_0__ID_STATUS_MASK 0x00000F00L +//MMSCH_GPUIOV_SCH_BLOCK_IP_1 +#define MMSCH_GPUIOV_SCH_BLOCK_IP_1__ID__SHIFT 0x0 +#define MMSCH_GPUIOV_SCH_BLOCK_IP_1__VERSION__SHIFT 0x4 +#define MMSCH_GPUIOV_SCH_BLOCK_IP_1__SIZE__SHIFT 0x8 +#define MMSCH_GPUIOV_SCH_BLOCK_IP_1__ID_MASK 0x0000000FL +#define MMSCH_GPUIOV_SCH_BLOCK_IP_1__VERSION_MASK 0x000000F0L +#define MMSCH_GPUIOV_SCH_BLOCK_IP_1__SIZE_MASK 0x0000FF00L +//MMSCH_GPUIOV_CMD_STATUS_IP_1 +#define MMSCH_GPUIOV_CMD_STATUS_IP_1__CMD_STATUS__SHIFT 0x0 +#define MMSCH_GPUIOV_CMD_STATUS_IP_1__CMD_STATUS_MASK 0x0000000FL +//MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1__ID__SHIFT 0x0 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1__ID_STATUS__SHIFT 0x8 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1__ID_MASK 0x000000FFL +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_1__ID_STATUS_MASK 0x00000F00L +//MMSCH_GPUIOV_CNTXT_IP +#define MMSCH_GPUIOV_CNTXT_IP__CNTXT_SIZE__SHIFT 0x0 +#define MMSCH_GPUIOV_CNTXT_IP__CNTXT_LOCATION__SHIFT 0x7 +#define MMSCH_GPUIOV_CNTXT_IP__CNTXT_SIZE_MASK 0x0000007FL +#define MMSCH_GPUIOV_CNTXT_IP__CNTXT_LOCATION_MASK 0x00000080L +//MMSCH_GPUIOV_SCH_BLOCK_2 +#define MMSCH_GPUIOV_SCH_BLOCK_2__ID__SHIFT 0x0 +#define MMSCH_GPUIOV_SCH_BLOCK_2__VERSION__SHIFT 0x4 +#define MMSCH_GPUIOV_SCH_BLOCK_2__SIZE__SHIFT 0x8 +#define MMSCH_GPUIOV_SCH_BLOCK_2__ID_MASK 0x0000000FL +#define MMSCH_GPUIOV_SCH_BLOCK_2__VERSION_MASK 0x000000F0L +#define MMSCH_GPUIOV_SCH_BLOCK_2__SIZE_MASK 0x0000FF00L +//MMSCH_GPUIOV_CMD_CONTROL_2 +#define MMSCH_GPUIOV_CMD_CONTROL_2__CMD_TYPE__SHIFT 0x0 +#define MMSCH_GPUIOV_CMD_CONTROL_2__CMD_EXECUTE__SHIFT 0x4 +#define MMSCH_GPUIOV_CMD_CONTROL_2__CMD_EXECUTE_INTR_EN__SHIFT 0x5 +#define MMSCH_GPUIOV_CMD_CONTROL_2__VM_BUSY_INTR_EN__SHIFT 0x6 +#define MMSCH_GPUIOV_CMD_CONTROL_2__FUNCTINO_ID__SHIFT 0x8 +#define MMSCH_GPUIOV_CMD_CONTROL_2__NEXT_FUNCTINO_ID__SHIFT 0x10 +#define MMSCH_GPUIOV_CMD_CONTROL_2__CMD_TYPE_MASK 0x0000000FL +#define MMSCH_GPUIOV_CMD_CONTROL_2__CMD_EXECUTE_MASK 0x00000010L +#define MMSCH_GPUIOV_CMD_CONTROL_2__CMD_EXECUTE_INTR_EN_MASK 0x00000020L +#define MMSCH_GPUIOV_CMD_CONTROL_2__VM_BUSY_INTR_EN_MASK 0x00000040L +#define MMSCH_GPUIOV_CMD_CONTROL_2__FUNCTINO_ID_MASK 0x0000FF00L +#define MMSCH_GPUIOV_CMD_CONTROL_2__NEXT_FUNCTINO_ID_MASK 0x00FF0000L +//MMSCH_GPUIOV_CMD_STATUS_2 +#define MMSCH_GPUIOV_CMD_STATUS_2__CMD_STATUS__SHIFT 0x0 +#define MMSCH_GPUIOV_CMD_STATUS_2__CMD_STATUS_MASK 0x0000000FL +//MMSCH_GPUIOV_VM_BUSY_STATUS_2 +#define MMSCH_GPUIOV_VM_BUSY_STATUS_2__BUSY__SHIFT 0x0 +#define MMSCH_GPUIOV_VM_BUSY_STATUS_2__BUSY_MASK 0xFFFFFFFFL +//MMSCH_GPUIOV_ACTIVE_FCNS_2 +#define MMSCH_GPUIOV_ACTIVE_FCNS_2__ACTIVE_FCNS__SHIFT 0x0 +#define MMSCH_GPUIOV_ACTIVE_FCNS_2__ACTIVE_FCNS_MASK 0xFFFFFFFFL +//MMSCH_GPUIOV_ACTIVE_FCN_ID_2 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_2__ID__SHIFT 0x0 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_2__ID_STATUS__SHIFT 0x8 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_2__ID_MASK 0x000000FFL +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_2__ID_STATUS_MASK 0x00000F00L +//MMSCH_GPUIOV_DW6_2 +#define MMSCH_GPUIOV_DW6_2__DATA__SHIFT 0x0 +#define MMSCH_GPUIOV_DW6_2__DATA_MASK 0xFFFFFFFFL +//MMSCH_GPUIOV_DW7_2 +#define MMSCH_GPUIOV_DW7_2__DATA__SHIFT 0x0 +#define MMSCH_GPUIOV_DW7_2__DATA_MASK 0xFFFFFFFFL +//MMSCH_GPUIOV_DW8_2 +#define MMSCH_GPUIOV_DW8_2__DATA__SHIFT 0x0 +#define MMSCH_GPUIOV_DW8_2__DATA_MASK 0xFFFFFFFFL +//MMSCH_GPUIOV_SCH_BLOCK_IP_2 +#define MMSCH_GPUIOV_SCH_BLOCK_IP_2__ID__SHIFT 0x0 +#define MMSCH_GPUIOV_SCH_BLOCK_IP_2__VERSION__SHIFT 0x4 +#define MMSCH_GPUIOV_SCH_BLOCK_IP_2__SIZE__SHIFT 0x8 +#define MMSCH_GPUIOV_SCH_BLOCK_IP_2__ID_MASK 0x0000000FL +#define MMSCH_GPUIOV_SCH_BLOCK_IP_2__VERSION_MASK 0x000000F0L +#define MMSCH_GPUIOV_SCH_BLOCK_IP_2__SIZE_MASK 0x0000FF00L +//MMSCH_GPUIOV_CMD_STATUS_IP_2 +#define MMSCH_GPUIOV_CMD_STATUS_IP_2__CMD_STATUS__SHIFT 0x0 +#define MMSCH_GPUIOV_CMD_STATUS_IP_2__CMD_STATUS_MASK 0x0000000FL +//MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2__ID__SHIFT 0x0 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2__ID_STATUS__SHIFT 0x8 +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2__ID_MASK 0x000000FFL +#define MMSCH_GPUIOV_ACTIVE_FCN_ID_IP_2__ID_STATUS_MASK 0x00000F00L +//MMSCH_VFID_FIFO_HEAD_2 +#define MMSCH_VFID_FIFO_HEAD_2__HEAD__SHIFT 0x0 +#define MMSCH_VFID_FIFO_HEAD_2__HEAD_MASK 0x0000003FL +//MMSCH_VFID_FIFO_TAIL_2 +#define MMSCH_VFID_FIFO_TAIL_2__TAIL__SHIFT 0x0 +#define MMSCH_VFID_FIFO_TAIL_2__TAIL_MASK 0x0000003FL +//MMSCH_VM_BUSY_STATUS_0 +#define MMSCH_VM_BUSY_STATUS_0__BUSY__SHIFT 0x0 +#define MMSCH_VM_BUSY_STATUS_0__BUSY_MASK 0xFFFFFFFFL +//MMSCH_VM_BUSY_STATUS_1 +#define MMSCH_VM_BUSY_STATUS_1__BUSY__SHIFT 0x0 +#define MMSCH_VM_BUSY_STATUS_1__BUSY_MASK 0xFFFFFFFFL +//MMSCH_VM_BUSY_STATUS_2 +#define MMSCH_VM_BUSY_STATUS_2__BUSY__SHIFT 0x0 +#define MMSCH_VM_BUSY_STATUS_2__BUSY_MASK 0xFFFFFFFFL + + +// addressBlock: aid_uvd0_slmi_adpdec +//UVD_LMI_MMSCH_NC0_64BIT_BAR_LOW +#define UVD_LMI_MMSCH_NC0_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC0_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MMSCH_NC0_64BIT_BAR_HIGH +#define UVD_LMI_MMSCH_NC0_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC0_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MMSCH_NC1_64BIT_BAR_LOW +#define UVD_LMI_MMSCH_NC1_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC1_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MMSCH_NC1_64BIT_BAR_HIGH +#define UVD_LMI_MMSCH_NC1_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC1_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MMSCH_NC2_64BIT_BAR_LOW +#define UVD_LMI_MMSCH_NC2_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC2_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MMSCH_NC2_64BIT_BAR_HIGH +#define UVD_LMI_MMSCH_NC2_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC2_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MMSCH_NC3_64BIT_BAR_LOW +#define UVD_LMI_MMSCH_NC3_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC3_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MMSCH_NC3_64BIT_BAR_HIGH +#define UVD_LMI_MMSCH_NC3_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC3_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MMSCH_NC4_64BIT_BAR_LOW +#define UVD_LMI_MMSCH_NC4_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC4_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MMSCH_NC4_64BIT_BAR_HIGH +#define UVD_LMI_MMSCH_NC4_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC4_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MMSCH_NC5_64BIT_BAR_LOW +#define UVD_LMI_MMSCH_NC5_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC5_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MMSCH_NC5_64BIT_BAR_HIGH +#define UVD_LMI_MMSCH_NC5_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC5_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MMSCH_NC6_64BIT_BAR_LOW +#define UVD_LMI_MMSCH_NC6_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC6_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MMSCH_NC6_64BIT_BAR_HIGH +#define UVD_LMI_MMSCH_NC6_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC6_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MMSCH_NC7_64BIT_BAR_LOW +#define UVD_LMI_MMSCH_NC7_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC7_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_LMI_MMSCH_NC7_64BIT_BAR_HIGH +#define UVD_LMI_MMSCH_NC7_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC7_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_LMI_MMSCH_NC_VMID +#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC0_VMID__SHIFT 0x0 +#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC1_VMID__SHIFT 0x4 +#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC2_VMID__SHIFT 0x8 +#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC3_VMID__SHIFT 0xc +#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC4_VMID__SHIFT 0x10 +#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC5_VMID__SHIFT 0x14 +#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC6_VMID__SHIFT 0x18 +#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC7_VMID__SHIFT 0x1c +#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC0_VMID_MASK 0x0000000FL +#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC1_VMID_MASK 0x000000F0L +#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC2_VMID_MASK 0x00000F00L +#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC3_VMID_MASK 0x0000F000L +#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC4_VMID_MASK 0x000F0000L +#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC5_VMID_MASK 0x00F00000L +#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC6_VMID_MASK 0x0F000000L +#define UVD_LMI_MMSCH_NC_VMID__MMSCH_NC7_VMID_MASK 0xF0000000L +//UVD_LMI_MMSCH_CTRL +#define UVD_LMI_MMSCH_CTRL__MMSCH_DATA_COHERENCY_EN__SHIFT 0x0 +#define UVD_LMI_MMSCH_CTRL__MMSCH_VM__SHIFT 0x1 +#define UVD_LMI_MMSCH_CTRL__PRIV_CLIENT_MMSCH__SHIFT 0x2 +#define UVD_LMI_MMSCH_CTRL__MMSCH_R_MC_SWAP__SHIFT 0x3 +#define UVD_LMI_MMSCH_CTRL__MMSCH_W_MC_SWAP__SHIFT 0x5 +#define UVD_LMI_MMSCH_CTRL__MMSCH_RD__SHIFT 0x7 +#define UVD_LMI_MMSCH_CTRL__MMSCH_WR__SHIFT 0x9 +#define UVD_LMI_MMSCH_CTRL__MMSCH_RD_DROP__SHIFT 0xb +#define UVD_LMI_MMSCH_CTRL__MMSCH_WR_DROP__SHIFT 0xc +#define UVD_LMI_MMSCH_CTRL__MMSCH_DATA_COHERENCY_EN_MASK 0x00000001L +#define UVD_LMI_MMSCH_CTRL__MMSCH_VM_MASK 0x00000002L +#define UVD_LMI_MMSCH_CTRL__PRIV_CLIENT_MMSCH_MASK 0x00000004L +#define UVD_LMI_MMSCH_CTRL__MMSCH_R_MC_SWAP_MASK 0x00000018L +#define UVD_LMI_MMSCH_CTRL__MMSCH_W_MC_SWAP_MASK 0x00000060L +#define UVD_LMI_MMSCH_CTRL__MMSCH_RD_MASK 0x00000180L +#define UVD_LMI_MMSCH_CTRL__MMSCH_WR_MASK 0x00000600L +#define UVD_LMI_MMSCH_CTRL__MMSCH_RD_DROP_MASK 0x00000800L +#define UVD_LMI_MMSCH_CTRL__MMSCH_WR_DROP_MASK 0x00001000L +//UVD_MMSCH_LMI_STATUS +#define UVD_MMSCH_LMI_STATUS__LMI_AXI_MMSCH_UNSUPPORTED_LEN_INT__SHIFT 0x0 +#define UVD_MMSCH_LMI_STATUS__LMI_AXI_MMSCH_UNSUPPORTED_ADR_ALIGN_INT__SHIFT 0x1 +#define UVD_MMSCH_LMI_STATUS__MMSCH_LMI_WRITE_CLEAN__SHIFT 0x2 +#define UVD_MMSCH_LMI_STATUS__AXI_MMSCH_ERR_LEN__SHIFT 0x4 +#define UVD_MMSCH_LMI_STATUS__AXI_MMSCH_ERR_ADDR_LSBS__SHIFT 0x8 +#define UVD_MMSCH_LMI_STATUS__AXI_MMSCH_ERR_AWRITE__SHIFT 0xc +#define UVD_MMSCH_LMI_STATUS__MMSCH_RD_CLEAN__SHIFT 0xd +#define UVD_MMSCH_LMI_STATUS__MMSCH_WR_CLEAN__SHIFT 0xe +#define UVD_MMSCH_LMI_STATUS__LMI_AXI_MMSCH_UNSUPPORTED_LEN_INT_MASK 0x00000001L +#define UVD_MMSCH_LMI_STATUS__LMI_AXI_MMSCH_UNSUPPORTED_ADR_ALIGN_INT_MASK 0x00000002L +#define UVD_MMSCH_LMI_STATUS__MMSCH_LMI_WRITE_CLEAN_MASK 0x00000004L +#define UVD_MMSCH_LMI_STATUS__AXI_MMSCH_ERR_LEN_MASK 0x000000F0L +#define UVD_MMSCH_LMI_STATUS__AXI_MMSCH_ERR_ADDR_LSBS_MASK 0x00000700L +#define UVD_MMSCH_LMI_STATUS__AXI_MMSCH_ERR_AWRITE_MASK 0x00001000L +#define UVD_MMSCH_LMI_STATUS__MMSCH_RD_CLEAN_MASK 0x00002000L +#define UVD_MMSCH_LMI_STATUS__MMSCH_WR_CLEAN_MASK 0x00004000L +//VCN_RAS_CNTL_MMSCH +#define VCN_RAS_CNTL_MMSCH__MMSCH_FATAL_ERROR_EN__SHIFT 0x1 +#define VCN_RAS_CNTL_MMSCH__MMSCH_PMI_EN__SHIFT 0x5 +#define VCN_RAS_CNTL_MMSCH__MMSCH_REARM__SHIFT 0x9 +#define VCN_RAS_CNTL_MMSCH__MMSCH_READY__SHIFT 0x11 +#define VCN_RAS_CNTL_MMSCH__MMSCH_FATAL_ERROR_EN_MASK 0x00000002L +#define VCN_RAS_CNTL_MMSCH__MMSCH_PMI_EN_MASK 0x00000020L +#define VCN_RAS_CNTL_MMSCH__MMSCH_REARM_MASK 0x00000200L +#define VCN_RAS_CNTL_MMSCH__MMSCH_READY_MASK 0x00020000L + + +// addressBlock: aid_uvd0_uvd_jrbc1_uvd_jrbc_dec +//UVD_JRBC1_UVD_JRBC_RB_WPTR +#define UVD_JRBC1_UVD_JRBC_RB_WPTR__RB_WPTR__SHIFT 0x4 +#define UVD_JRBC1_UVD_JRBC_RB_WPTR__RB_WPTR_MASK 0x007FFFF0L +//UVD_JRBC1_UVD_JRBC_RB_CNTL +#define UVD_JRBC1_UVD_JRBC_RB_CNTL__RB_NO_FETCH__SHIFT 0x0 +#define UVD_JRBC1_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT 0x1 +#define UVD_JRBC1_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER__SHIFT 0x4 +#define UVD_JRBC1_UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK 0x00000001L +#define UVD_JRBC1_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK 0x00000002L +#define UVD_JRBC1_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER_MASK 0x0007FFF0L +//UVD_JRBC1_UVD_JRBC_IB_SIZE +#define UVD_JRBC1_UVD_JRBC_IB_SIZE__IB_SIZE__SHIFT 0x4 +#define UVD_JRBC1_UVD_JRBC_IB_SIZE__IB_SIZE_MASK 0x007FFFF0L +//UVD_JRBC1_UVD_JRBC_URGENT_CNTL +#define UVD_JRBC1_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT 0x0 +#define UVD_JRBC1_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK 0x00000003L +//UVD_JRBC1_UVD_JRBC_RB_REF_DATA +#define UVD_JRBC1_UVD_JRBC_RB_REF_DATA__REF_DATA__SHIFT 0x0 +#define UVD_JRBC1_UVD_JRBC_RB_REF_DATA__REF_DATA_MASK 0xFFFFFFFFL +//UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER +#define UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT 0x0 +#define UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT 0x10 +#define UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT 0x18 +#define UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT 0x19 +#define UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK 0x0000FFFFL +#define UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK 0x00FF0000L +#define UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK 0x01000000L +#define UVD_JRBC1_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK 0x02000000L +//UVD_JRBC1_UVD_JRBC_SOFT_RESET +#define UVD_JRBC1_UVD_JRBC_SOFT_RESET__RESET__SHIFT 0x0 +#define UVD_JRBC1_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS__SHIFT 0x11 +#define UVD_JRBC1_UVD_JRBC_SOFT_RESET__RESET_MASK 0x00000001L +#define UVD_JRBC1_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS_MASK 0x00020000L +//UVD_JRBC1_UVD_JRBC_STATUS +#define UVD_JRBC1_UVD_JRBC_STATUS__RB_JOB_DONE__SHIFT 0x0 +#define UVD_JRBC1_UVD_JRBC_STATUS__IB_JOB_DONE__SHIFT 0x1 +#define UVD_JRBC1_UVD_JRBC_STATUS__RB_ILLEGAL_CMD__SHIFT 0x2 +#define UVD_JRBC1_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT__SHIFT 0x3 +#define UVD_JRBC1_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT__SHIFT 0x4 +#define UVD_JRBC1_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT__SHIFT 0x5 +#define UVD_JRBC1_UVD_JRBC_STATUS__IB_ILLEGAL_CMD__SHIFT 0x6 +#define UVD_JRBC1_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT__SHIFT 0x7 +#define UVD_JRBC1_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT__SHIFT 0x8 +#define UVD_JRBC1_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT__SHIFT 0x9 +#define UVD_JRBC1_UVD_JRBC_STATUS__RB_TRAP_STATUS__SHIFT 0xa +#define UVD_JRBC1_UVD_JRBC_STATUS__PREEMPT_STATUS__SHIFT 0xb +#define UVD_JRBC1_UVD_JRBC_STATUS__IB_TRAP_STATUS__SHIFT 0xc +#define UVD_JRBC1_UVD_JRBC_STATUS__INT_EN__SHIFT 0x10 +#define UVD_JRBC1_UVD_JRBC_STATUS__INT_ACK__SHIFT 0x11 +#define UVD_JRBC1_UVD_JRBC_STATUS__RB_JOB_DONE_MASK 0x00000001L +#define UVD_JRBC1_UVD_JRBC_STATUS__IB_JOB_DONE_MASK 0x00000002L +#define UVD_JRBC1_UVD_JRBC_STATUS__RB_ILLEGAL_CMD_MASK 0x00000004L +#define UVD_JRBC1_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT_MASK 0x00000008L +#define UVD_JRBC1_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT_MASK 0x00000010L +#define UVD_JRBC1_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT_MASK 0x00000020L +#define UVD_JRBC1_UVD_JRBC_STATUS__IB_ILLEGAL_CMD_MASK 0x00000040L +#define UVD_JRBC1_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT_MASK 0x00000080L +#define UVD_JRBC1_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT_MASK 0x00000100L +#define UVD_JRBC1_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT_MASK 0x00000200L +#define UVD_JRBC1_UVD_JRBC_STATUS__RB_TRAP_STATUS_MASK 0x00000400L +#define UVD_JRBC1_UVD_JRBC_STATUS__PREEMPT_STATUS_MASK 0x00000800L +#define UVD_JRBC1_UVD_JRBC_STATUS__IB_TRAP_STATUS_MASK 0x00001000L +#define UVD_JRBC1_UVD_JRBC_STATUS__INT_EN_MASK 0x00010000L +#define UVD_JRBC1_UVD_JRBC_STATUS__INT_ACK_MASK 0x00020000L +//UVD_JRBC1_UVD_JRBC_RB_RPTR +#define UVD_JRBC1_UVD_JRBC_RB_RPTR__RB_RPTR__SHIFT 0x4 +#define UVD_JRBC1_UVD_JRBC_RB_RPTR__RB_RPTR_MASK 0x007FFFF0L +//UVD_JRBC1_UVD_JRBC_RB_BUF_STATUS +#define UVD_JRBC1_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID__SHIFT 0x0 +#define UVD_JRBC1_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT 0x10 +#define UVD_JRBC1_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT 0x18 +#define UVD_JRBC1_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID_MASK 0x0000FFFFL +#define UVD_JRBC1_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR_MASK 0x000F0000L +#define UVD_JRBC1_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR_MASK 0x03000000L +//UVD_JRBC1_UVD_JRBC_IB_BUF_STATUS +#define UVD_JRBC1_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID__SHIFT 0x0 +#define UVD_JRBC1_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT 0x10 +#define UVD_JRBC1_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT 0x18 +#define UVD_JRBC1_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID_MASK 0x0000FFFFL +#define UVD_JRBC1_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR_MASK 0x000F0000L +#define UVD_JRBC1_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR_MASK 0x03000000L +//UVD_JRBC1_UVD_JRBC_IB_SIZE_UPDATE +#define UVD_JRBC1_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT 0x4 +#define UVD_JRBC1_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK 0x007FFFF0L +//UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER +#define UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT 0x0 +#define UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT 0x10 +#define UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT 0x18 +#define UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT 0x19 +#define UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK 0x0000FFFFL +#define UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK 0x00FF0000L +#define UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK 0x01000000L +#define UVD_JRBC1_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK 0x02000000L +//UVD_JRBC1_UVD_JRBC_IB_REF_DATA +#define UVD_JRBC1_UVD_JRBC_IB_REF_DATA__REF_DATA__SHIFT 0x0 +#define UVD_JRBC1_UVD_JRBC_IB_REF_DATA__REF_DATA_MASK 0xFFFFFFFFL +//UVD_JRBC1_UVD_JPEG_PREEMPT_CMD +#define UVD_JRBC1_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN__SHIFT 0x0 +#define UVD_JRBC1_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE__SHIFT 0x1 +#define UVD_JRBC1_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD__SHIFT 0x2 +#define UVD_JRBC1_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN_MASK 0x00000001L +#define UVD_JRBC1_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE_MASK 0x00000002L +#define UVD_JRBC1_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD_MASK 0x00000004L +//UVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA0 +#define UVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0__SHIFT 0x0 +#define UVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0_MASK 0xFFFFFFFFL +//UVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA1 +#define UVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1__SHIFT 0x0 +#define UVD_JRBC1_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1_MASK 0xFFFFFFFFL +//UVD_JRBC1_UVD_JRBC_RB_SIZE +#define UVD_JRBC1_UVD_JRBC_RB_SIZE__RB_SIZE__SHIFT 0x4 +#define UVD_JRBC1_UVD_JRBC_RB_SIZE__RB_SIZE_MASK 0x00FFFFF0L +//UVD_JRBC1_UVD_JRBC_SCRATCH0 +#define UVD_JRBC1_UVD_JRBC_SCRATCH0__SCRATCH0__SHIFT 0x0 +#define UVD_JRBC1_UVD_JRBC_SCRATCH0__SCRATCH0_MASK 0xFFFFFFFFL + + +// addressBlock: aid_uvd0_uvd_jrbc2_uvd_jrbc_dec +//UVD_JRBC2_UVD_JRBC_RB_WPTR +#define UVD_JRBC2_UVD_JRBC_RB_WPTR__RB_WPTR__SHIFT 0x4 +#define UVD_JRBC2_UVD_JRBC_RB_WPTR__RB_WPTR_MASK 0x007FFFF0L +//UVD_JRBC2_UVD_JRBC_RB_CNTL +#define UVD_JRBC2_UVD_JRBC_RB_CNTL__RB_NO_FETCH__SHIFT 0x0 +#define UVD_JRBC2_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT 0x1 +#define UVD_JRBC2_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER__SHIFT 0x4 +#define UVD_JRBC2_UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK 0x00000001L +#define UVD_JRBC2_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK 0x00000002L +#define UVD_JRBC2_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER_MASK 0x0007FFF0L +//UVD_JRBC2_UVD_JRBC_IB_SIZE +#define UVD_JRBC2_UVD_JRBC_IB_SIZE__IB_SIZE__SHIFT 0x4 +#define UVD_JRBC2_UVD_JRBC_IB_SIZE__IB_SIZE_MASK 0x007FFFF0L +//UVD_JRBC2_UVD_JRBC_URGENT_CNTL +#define UVD_JRBC2_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT 0x0 +#define UVD_JRBC2_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK 0x00000003L +//UVD_JRBC2_UVD_JRBC_RB_REF_DATA +#define UVD_JRBC2_UVD_JRBC_RB_REF_DATA__REF_DATA__SHIFT 0x0 +#define UVD_JRBC2_UVD_JRBC_RB_REF_DATA__REF_DATA_MASK 0xFFFFFFFFL +//UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER +#define UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT 0x0 +#define UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT 0x10 +#define UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT 0x18 +#define UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT 0x19 +#define UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK 0x0000FFFFL +#define UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK 0x00FF0000L +#define UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK 0x01000000L +#define UVD_JRBC2_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK 0x02000000L +//UVD_JRBC2_UVD_JRBC_SOFT_RESET +#define UVD_JRBC2_UVD_JRBC_SOFT_RESET__RESET__SHIFT 0x0 +#define UVD_JRBC2_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS__SHIFT 0x11 +#define UVD_JRBC2_UVD_JRBC_SOFT_RESET__RESET_MASK 0x00000001L +#define UVD_JRBC2_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS_MASK 0x00020000L +//UVD_JRBC2_UVD_JRBC_STATUS +#define UVD_JRBC2_UVD_JRBC_STATUS__RB_JOB_DONE__SHIFT 0x0 +#define UVD_JRBC2_UVD_JRBC_STATUS__IB_JOB_DONE__SHIFT 0x1 +#define UVD_JRBC2_UVD_JRBC_STATUS__RB_ILLEGAL_CMD__SHIFT 0x2 +#define UVD_JRBC2_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT__SHIFT 0x3 +#define UVD_JRBC2_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT__SHIFT 0x4 +#define UVD_JRBC2_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT__SHIFT 0x5 +#define UVD_JRBC2_UVD_JRBC_STATUS__IB_ILLEGAL_CMD__SHIFT 0x6 +#define UVD_JRBC2_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT__SHIFT 0x7 +#define UVD_JRBC2_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT__SHIFT 0x8 +#define UVD_JRBC2_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT__SHIFT 0x9 +#define UVD_JRBC2_UVD_JRBC_STATUS__RB_TRAP_STATUS__SHIFT 0xa +#define UVD_JRBC2_UVD_JRBC_STATUS__PREEMPT_STATUS__SHIFT 0xb +#define UVD_JRBC2_UVD_JRBC_STATUS__IB_TRAP_STATUS__SHIFT 0xc +#define UVD_JRBC2_UVD_JRBC_STATUS__INT_EN__SHIFT 0x10 +#define UVD_JRBC2_UVD_JRBC_STATUS__INT_ACK__SHIFT 0x11 +#define UVD_JRBC2_UVD_JRBC_STATUS__RB_JOB_DONE_MASK 0x00000001L +#define UVD_JRBC2_UVD_JRBC_STATUS__IB_JOB_DONE_MASK 0x00000002L +#define UVD_JRBC2_UVD_JRBC_STATUS__RB_ILLEGAL_CMD_MASK 0x00000004L +#define UVD_JRBC2_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT_MASK 0x00000008L +#define UVD_JRBC2_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT_MASK 0x00000010L +#define UVD_JRBC2_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT_MASK 0x00000020L +#define UVD_JRBC2_UVD_JRBC_STATUS__IB_ILLEGAL_CMD_MASK 0x00000040L +#define UVD_JRBC2_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT_MASK 0x00000080L +#define UVD_JRBC2_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT_MASK 0x00000100L +#define UVD_JRBC2_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT_MASK 0x00000200L +#define UVD_JRBC2_UVD_JRBC_STATUS__RB_TRAP_STATUS_MASK 0x00000400L +#define UVD_JRBC2_UVD_JRBC_STATUS__PREEMPT_STATUS_MASK 0x00000800L +#define UVD_JRBC2_UVD_JRBC_STATUS__IB_TRAP_STATUS_MASK 0x00001000L +#define UVD_JRBC2_UVD_JRBC_STATUS__INT_EN_MASK 0x00010000L +#define UVD_JRBC2_UVD_JRBC_STATUS__INT_ACK_MASK 0x00020000L +//UVD_JRBC2_UVD_JRBC_RB_RPTR +#define UVD_JRBC2_UVD_JRBC_RB_RPTR__RB_RPTR__SHIFT 0x4 +#define UVD_JRBC2_UVD_JRBC_RB_RPTR__RB_RPTR_MASK 0x007FFFF0L +//UVD_JRBC2_UVD_JRBC_RB_BUF_STATUS +#define UVD_JRBC2_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID__SHIFT 0x0 +#define UVD_JRBC2_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT 0x10 +#define UVD_JRBC2_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT 0x18 +#define UVD_JRBC2_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID_MASK 0x0000FFFFL +#define UVD_JRBC2_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR_MASK 0x000F0000L +#define UVD_JRBC2_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR_MASK 0x03000000L +//UVD_JRBC2_UVD_JRBC_IB_BUF_STATUS +#define UVD_JRBC2_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID__SHIFT 0x0 +#define UVD_JRBC2_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT 0x10 +#define UVD_JRBC2_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT 0x18 +#define UVD_JRBC2_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID_MASK 0x0000FFFFL +#define UVD_JRBC2_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR_MASK 0x000F0000L +#define UVD_JRBC2_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR_MASK 0x03000000L +//UVD_JRBC2_UVD_JRBC_IB_SIZE_UPDATE +#define UVD_JRBC2_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT 0x4 +#define UVD_JRBC2_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK 0x007FFFF0L +//UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER +#define UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT 0x0 +#define UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT 0x10 +#define UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT 0x18 +#define UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT 0x19 +#define UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK 0x0000FFFFL +#define UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK 0x00FF0000L +#define UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK 0x01000000L +#define UVD_JRBC2_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK 0x02000000L +//UVD_JRBC2_UVD_JRBC_IB_REF_DATA +#define UVD_JRBC2_UVD_JRBC_IB_REF_DATA__REF_DATA__SHIFT 0x0 +#define UVD_JRBC2_UVD_JRBC_IB_REF_DATA__REF_DATA_MASK 0xFFFFFFFFL +//UVD_JRBC2_UVD_JPEG_PREEMPT_CMD +#define UVD_JRBC2_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN__SHIFT 0x0 +#define UVD_JRBC2_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE__SHIFT 0x1 +#define UVD_JRBC2_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD__SHIFT 0x2 +#define UVD_JRBC2_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN_MASK 0x00000001L +#define UVD_JRBC2_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE_MASK 0x00000002L +#define UVD_JRBC2_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD_MASK 0x00000004L +//UVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA0 +#define UVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0__SHIFT 0x0 +#define UVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0_MASK 0xFFFFFFFFL +//UVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA1 +#define UVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1__SHIFT 0x0 +#define UVD_JRBC2_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1_MASK 0xFFFFFFFFL +//UVD_JRBC2_UVD_JRBC_RB_SIZE +#define UVD_JRBC2_UVD_JRBC_RB_SIZE__RB_SIZE__SHIFT 0x4 +#define UVD_JRBC2_UVD_JRBC_RB_SIZE__RB_SIZE_MASK 0x00FFFFF0L +//UVD_JRBC2_UVD_JRBC_SCRATCH0 +#define UVD_JRBC2_UVD_JRBC_SCRATCH0__SCRATCH0__SHIFT 0x0 +#define UVD_JRBC2_UVD_JRBC_SCRATCH0__SCRATCH0_MASK 0xFFFFFFFFL + + +// addressBlock: aid_uvd0_uvd_jrbc3_uvd_jrbc_dec +//UVD_JRBC3_UVD_JRBC_RB_WPTR +#define UVD_JRBC3_UVD_JRBC_RB_WPTR__RB_WPTR__SHIFT 0x4 +#define UVD_JRBC3_UVD_JRBC_RB_WPTR__RB_WPTR_MASK 0x007FFFF0L +//UVD_JRBC3_UVD_JRBC_RB_CNTL +#define UVD_JRBC3_UVD_JRBC_RB_CNTL__RB_NO_FETCH__SHIFT 0x0 +#define UVD_JRBC3_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT 0x1 +#define UVD_JRBC3_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER__SHIFT 0x4 +#define UVD_JRBC3_UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK 0x00000001L +#define UVD_JRBC3_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK 0x00000002L +#define UVD_JRBC3_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER_MASK 0x0007FFF0L +//UVD_JRBC3_UVD_JRBC_IB_SIZE +#define UVD_JRBC3_UVD_JRBC_IB_SIZE__IB_SIZE__SHIFT 0x4 +#define UVD_JRBC3_UVD_JRBC_IB_SIZE__IB_SIZE_MASK 0x007FFFF0L +//UVD_JRBC3_UVD_JRBC_URGENT_CNTL +#define UVD_JRBC3_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT 0x0 +#define UVD_JRBC3_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK 0x00000003L +//UVD_JRBC3_UVD_JRBC_RB_REF_DATA +#define UVD_JRBC3_UVD_JRBC_RB_REF_DATA__REF_DATA__SHIFT 0x0 +#define UVD_JRBC3_UVD_JRBC_RB_REF_DATA__REF_DATA_MASK 0xFFFFFFFFL +//UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER +#define UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT 0x0 +#define UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT 0x10 +#define UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT 0x18 +#define UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT 0x19 +#define UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK 0x0000FFFFL +#define UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK 0x00FF0000L +#define UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK 0x01000000L +#define UVD_JRBC3_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK 0x02000000L +//UVD_JRBC3_UVD_JRBC_SOFT_RESET +#define UVD_JRBC3_UVD_JRBC_SOFT_RESET__RESET__SHIFT 0x0 +#define UVD_JRBC3_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS__SHIFT 0x11 +#define UVD_JRBC3_UVD_JRBC_SOFT_RESET__RESET_MASK 0x00000001L +#define UVD_JRBC3_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS_MASK 0x00020000L +//UVD_JRBC3_UVD_JRBC_STATUS +#define UVD_JRBC3_UVD_JRBC_STATUS__RB_JOB_DONE__SHIFT 0x0 +#define UVD_JRBC3_UVD_JRBC_STATUS__IB_JOB_DONE__SHIFT 0x1 +#define UVD_JRBC3_UVD_JRBC_STATUS__RB_ILLEGAL_CMD__SHIFT 0x2 +#define UVD_JRBC3_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT__SHIFT 0x3 +#define UVD_JRBC3_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT__SHIFT 0x4 +#define UVD_JRBC3_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT__SHIFT 0x5 +#define UVD_JRBC3_UVD_JRBC_STATUS__IB_ILLEGAL_CMD__SHIFT 0x6 +#define UVD_JRBC3_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT__SHIFT 0x7 +#define UVD_JRBC3_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT__SHIFT 0x8 +#define UVD_JRBC3_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT__SHIFT 0x9 +#define UVD_JRBC3_UVD_JRBC_STATUS__RB_TRAP_STATUS__SHIFT 0xa +#define UVD_JRBC3_UVD_JRBC_STATUS__PREEMPT_STATUS__SHIFT 0xb +#define UVD_JRBC3_UVD_JRBC_STATUS__IB_TRAP_STATUS__SHIFT 0xc +#define UVD_JRBC3_UVD_JRBC_STATUS__INT_EN__SHIFT 0x10 +#define UVD_JRBC3_UVD_JRBC_STATUS__INT_ACK__SHIFT 0x11 +#define UVD_JRBC3_UVD_JRBC_STATUS__RB_JOB_DONE_MASK 0x00000001L +#define UVD_JRBC3_UVD_JRBC_STATUS__IB_JOB_DONE_MASK 0x00000002L +#define UVD_JRBC3_UVD_JRBC_STATUS__RB_ILLEGAL_CMD_MASK 0x00000004L +#define UVD_JRBC3_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT_MASK 0x00000008L +#define UVD_JRBC3_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT_MASK 0x00000010L +#define UVD_JRBC3_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT_MASK 0x00000020L +#define UVD_JRBC3_UVD_JRBC_STATUS__IB_ILLEGAL_CMD_MASK 0x00000040L +#define UVD_JRBC3_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT_MASK 0x00000080L +#define UVD_JRBC3_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT_MASK 0x00000100L +#define UVD_JRBC3_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT_MASK 0x00000200L +#define UVD_JRBC3_UVD_JRBC_STATUS__RB_TRAP_STATUS_MASK 0x00000400L +#define UVD_JRBC3_UVD_JRBC_STATUS__PREEMPT_STATUS_MASK 0x00000800L +#define UVD_JRBC3_UVD_JRBC_STATUS__IB_TRAP_STATUS_MASK 0x00001000L +#define UVD_JRBC3_UVD_JRBC_STATUS__INT_EN_MASK 0x00010000L +#define UVD_JRBC3_UVD_JRBC_STATUS__INT_ACK_MASK 0x00020000L +//UVD_JRBC3_UVD_JRBC_RB_RPTR +#define UVD_JRBC3_UVD_JRBC_RB_RPTR__RB_RPTR__SHIFT 0x4 +#define UVD_JRBC3_UVD_JRBC_RB_RPTR__RB_RPTR_MASK 0x007FFFF0L +//UVD_JRBC3_UVD_JRBC_RB_BUF_STATUS +#define UVD_JRBC3_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID__SHIFT 0x0 +#define UVD_JRBC3_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT 0x10 +#define UVD_JRBC3_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT 0x18 +#define UVD_JRBC3_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID_MASK 0x0000FFFFL +#define UVD_JRBC3_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR_MASK 0x000F0000L +#define UVD_JRBC3_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR_MASK 0x03000000L +//UVD_JRBC3_UVD_JRBC_IB_BUF_STATUS +#define UVD_JRBC3_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID__SHIFT 0x0 +#define UVD_JRBC3_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT 0x10 +#define UVD_JRBC3_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT 0x18 +#define UVD_JRBC3_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID_MASK 0x0000FFFFL +#define UVD_JRBC3_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR_MASK 0x000F0000L +#define UVD_JRBC3_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR_MASK 0x03000000L +//UVD_JRBC3_UVD_JRBC_IB_SIZE_UPDATE +#define UVD_JRBC3_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT 0x4 +#define UVD_JRBC3_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK 0x007FFFF0L +//UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER +#define UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT 0x0 +#define UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT 0x10 +#define UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT 0x18 +#define UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT 0x19 +#define UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK 0x0000FFFFL +#define UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK 0x00FF0000L +#define UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK 0x01000000L +#define UVD_JRBC3_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK 0x02000000L +//UVD_JRBC3_UVD_JRBC_IB_REF_DATA +#define UVD_JRBC3_UVD_JRBC_IB_REF_DATA__REF_DATA__SHIFT 0x0 +#define UVD_JRBC3_UVD_JRBC_IB_REF_DATA__REF_DATA_MASK 0xFFFFFFFFL +//UVD_JRBC3_UVD_JPEG_PREEMPT_CMD +#define UVD_JRBC3_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN__SHIFT 0x0 +#define UVD_JRBC3_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE__SHIFT 0x1 +#define UVD_JRBC3_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD__SHIFT 0x2 +#define UVD_JRBC3_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN_MASK 0x00000001L +#define UVD_JRBC3_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE_MASK 0x00000002L +#define UVD_JRBC3_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD_MASK 0x00000004L +//UVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA0 +#define UVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0__SHIFT 0x0 +#define UVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0_MASK 0xFFFFFFFFL +//UVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA1 +#define UVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1__SHIFT 0x0 +#define UVD_JRBC3_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1_MASK 0xFFFFFFFFL +//UVD_JRBC3_UVD_JRBC_RB_SIZE +#define UVD_JRBC3_UVD_JRBC_RB_SIZE__RB_SIZE__SHIFT 0x4 +#define UVD_JRBC3_UVD_JRBC_RB_SIZE__RB_SIZE_MASK 0x00FFFFF0L +//UVD_JRBC3_UVD_JRBC_SCRATCH0 +#define UVD_JRBC3_UVD_JRBC_SCRATCH0__SCRATCH0__SHIFT 0x0 +#define UVD_JRBC3_UVD_JRBC_SCRATCH0__SCRATCH0_MASK 0xFFFFFFFFL + + +// addressBlock: aid_uvd0_uvd_jrbc4_uvd_jrbc_dec +//UVD_JRBC4_UVD_JRBC_RB_WPTR +#define UVD_JRBC4_UVD_JRBC_RB_WPTR__RB_WPTR__SHIFT 0x4 +#define UVD_JRBC4_UVD_JRBC_RB_WPTR__RB_WPTR_MASK 0x007FFFF0L +//UVD_JRBC4_UVD_JRBC_RB_CNTL +#define UVD_JRBC4_UVD_JRBC_RB_CNTL__RB_NO_FETCH__SHIFT 0x0 +#define UVD_JRBC4_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT 0x1 +#define UVD_JRBC4_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER__SHIFT 0x4 +#define UVD_JRBC4_UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK 0x00000001L +#define UVD_JRBC4_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK 0x00000002L +#define UVD_JRBC4_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER_MASK 0x0007FFF0L +//UVD_JRBC4_UVD_JRBC_IB_SIZE +#define UVD_JRBC4_UVD_JRBC_IB_SIZE__IB_SIZE__SHIFT 0x4 +#define UVD_JRBC4_UVD_JRBC_IB_SIZE__IB_SIZE_MASK 0x007FFFF0L +//UVD_JRBC4_UVD_JRBC_URGENT_CNTL +#define UVD_JRBC4_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT 0x0 +#define UVD_JRBC4_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK 0x00000003L +//UVD_JRBC4_UVD_JRBC_RB_REF_DATA +#define UVD_JRBC4_UVD_JRBC_RB_REF_DATA__REF_DATA__SHIFT 0x0 +#define UVD_JRBC4_UVD_JRBC_RB_REF_DATA__REF_DATA_MASK 0xFFFFFFFFL +//UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER +#define UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT 0x0 +#define UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT 0x10 +#define UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT 0x18 +#define UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT 0x19 +#define UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK 0x0000FFFFL +#define UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK 0x00FF0000L +#define UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK 0x01000000L +#define UVD_JRBC4_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK 0x02000000L +//UVD_JRBC4_UVD_JRBC_SOFT_RESET +#define UVD_JRBC4_UVD_JRBC_SOFT_RESET__RESET__SHIFT 0x0 +#define UVD_JRBC4_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS__SHIFT 0x11 +#define UVD_JRBC4_UVD_JRBC_SOFT_RESET__RESET_MASK 0x00000001L +#define UVD_JRBC4_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS_MASK 0x00020000L +//UVD_JRBC4_UVD_JRBC_STATUS +#define UVD_JRBC4_UVD_JRBC_STATUS__RB_JOB_DONE__SHIFT 0x0 +#define UVD_JRBC4_UVD_JRBC_STATUS__IB_JOB_DONE__SHIFT 0x1 +#define UVD_JRBC4_UVD_JRBC_STATUS__RB_ILLEGAL_CMD__SHIFT 0x2 +#define UVD_JRBC4_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT__SHIFT 0x3 +#define UVD_JRBC4_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT__SHIFT 0x4 +#define UVD_JRBC4_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT__SHIFT 0x5 +#define UVD_JRBC4_UVD_JRBC_STATUS__IB_ILLEGAL_CMD__SHIFT 0x6 +#define UVD_JRBC4_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT__SHIFT 0x7 +#define UVD_JRBC4_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT__SHIFT 0x8 +#define UVD_JRBC4_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT__SHIFT 0x9 +#define UVD_JRBC4_UVD_JRBC_STATUS__RB_TRAP_STATUS__SHIFT 0xa +#define UVD_JRBC4_UVD_JRBC_STATUS__PREEMPT_STATUS__SHIFT 0xb +#define UVD_JRBC4_UVD_JRBC_STATUS__IB_TRAP_STATUS__SHIFT 0xc +#define UVD_JRBC4_UVD_JRBC_STATUS__INT_EN__SHIFT 0x10 +#define UVD_JRBC4_UVD_JRBC_STATUS__INT_ACK__SHIFT 0x11 +#define UVD_JRBC4_UVD_JRBC_STATUS__RB_JOB_DONE_MASK 0x00000001L +#define UVD_JRBC4_UVD_JRBC_STATUS__IB_JOB_DONE_MASK 0x00000002L +#define UVD_JRBC4_UVD_JRBC_STATUS__RB_ILLEGAL_CMD_MASK 0x00000004L +#define UVD_JRBC4_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT_MASK 0x00000008L +#define UVD_JRBC4_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT_MASK 0x00000010L +#define UVD_JRBC4_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT_MASK 0x00000020L +#define UVD_JRBC4_UVD_JRBC_STATUS__IB_ILLEGAL_CMD_MASK 0x00000040L +#define UVD_JRBC4_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT_MASK 0x00000080L +#define UVD_JRBC4_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT_MASK 0x00000100L +#define UVD_JRBC4_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT_MASK 0x00000200L +#define UVD_JRBC4_UVD_JRBC_STATUS__RB_TRAP_STATUS_MASK 0x00000400L +#define UVD_JRBC4_UVD_JRBC_STATUS__PREEMPT_STATUS_MASK 0x00000800L +#define UVD_JRBC4_UVD_JRBC_STATUS__IB_TRAP_STATUS_MASK 0x00001000L +#define UVD_JRBC4_UVD_JRBC_STATUS__INT_EN_MASK 0x00010000L +#define UVD_JRBC4_UVD_JRBC_STATUS__INT_ACK_MASK 0x00020000L +//UVD_JRBC4_UVD_JRBC_RB_RPTR +#define UVD_JRBC4_UVD_JRBC_RB_RPTR__RB_RPTR__SHIFT 0x4 +#define UVD_JRBC4_UVD_JRBC_RB_RPTR__RB_RPTR_MASK 0x007FFFF0L +//UVD_JRBC4_UVD_JRBC_RB_BUF_STATUS +#define UVD_JRBC4_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID__SHIFT 0x0 +#define UVD_JRBC4_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT 0x10 +#define UVD_JRBC4_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT 0x18 +#define UVD_JRBC4_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID_MASK 0x0000FFFFL +#define UVD_JRBC4_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR_MASK 0x000F0000L +#define UVD_JRBC4_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR_MASK 0x03000000L +//UVD_JRBC4_UVD_JRBC_IB_BUF_STATUS +#define UVD_JRBC4_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID__SHIFT 0x0 +#define UVD_JRBC4_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT 0x10 +#define UVD_JRBC4_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT 0x18 +#define UVD_JRBC4_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID_MASK 0x0000FFFFL +#define UVD_JRBC4_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR_MASK 0x000F0000L +#define UVD_JRBC4_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR_MASK 0x03000000L +//UVD_JRBC4_UVD_JRBC_IB_SIZE_UPDATE +#define UVD_JRBC4_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT 0x4 +#define UVD_JRBC4_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK 0x007FFFF0L +//UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER +#define UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT 0x0 +#define UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT 0x10 +#define UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT 0x18 +#define UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT 0x19 +#define UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK 0x0000FFFFL +#define UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK 0x00FF0000L +#define UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK 0x01000000L +#define UVD_JRBC4_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK 0x02000000L +//UVD_JRBC4_UVD_JRBC_IB_REF_DATA +#define UVD_JRBC4_UVD_JRBC_IB_REF_DATA__REF_DATA__SHIFT 0x0 +#define UVD_JRBC4_UVD_JRBC_IB_REF_DATA__REF_DATA_MASK 0xFFFFFFFFL +//UVD_JRBC4_UVD_JPEG_PREEMPT_CMD +#define UVD_JRBC4_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN__SHIFT 0x0 +#define UVD_JRBC4_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE__SHIFT 0x1 +#define UVD_JRBC4_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD__SHIFT 0x2 +#define UVD_JRBC4_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN_MASK 0x00000001L +#define UVD_JRBC4_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE_MASK 0x00000002L +#define UVD_JRBC4_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD_MASK 0x00000004L +//UVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA0 +#define UVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0__SHIFT 0x0 +#define UVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0_MASK 0xFFFFFFFFL +//UVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA1 +#define UVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1__SHIFT 0x0 +#define UVD_JRBC4_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1_MASK 0xFFFFFFFFL +//UVD_JRBC4_UVD_JRBC_RB_SIZE +#define UVD_JRBC4_UVD_JRBC_RB_SIZE__RB_SIZE__SHIFT 0x4 +#define UVD_JRBC4_UVD_JRBC_RB_SIZE__RB_SIZE_MASK 0x00FFFFF0L +//UVD_JRBC4_UVD_JRBC_SCRATCH0 +#define UVD_JRBC4_UVD_JRBC_SCRATCH0__SCRATCH0__SHIFT 0x0 +#define UVD_JRBC4_UVD_JRBC_SCRATCH0__SCRATCH0_MASK 0xFFFFFFFFL + + +// addressBlock: aid_uvd0_uvd_jrbc5_uvd_jrbc_dec +//UVD_JRBC5_UVD_JRBC_RB_WPTR +#define UVD_JRBC5_UVD_JRBC_RB_WPTR__RB_WPTR__SHIFT 0x4 +#define UVD_JRBC5_UVD_JRBC_RB_WPTR__RB_WPTR_MASK 0x007FFFF0L +//UVD_JRBC5_UVD_JRBC_RB_CNTL +#define UVD_JRBC5_UVD_JRBC_RB_CNTL__RB_NO_FETCH__SHIFT 0x0 +#define UVD_JRBC5_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT 0x1 +#define UVD_JRBC5_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER__SHIFT 0x4 +#define UVD_JRBC5_UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK 0x00000001L +#define UVD_JRBC5_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK 0x00000002L +#define UVD_JRBC5_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER_MASK 0x0007FFF0L +//UVD_JRBC5_UVD_JRBC_IB_SIZE +#define UVD_JRBC5_UVD_JRBC_IB_SIZE__IB_SIZE__SHIFT 0x4 +#define UVD_JRBC5_UVD_JRBC_IB_SIZE__IB_SIZE_MASK 0x007FFFF0L +//UVD_JRBC5_UVD_JRBC_URGENT_CNTL +#define UVD_JRBC5_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT 0x0 +#define UVD_JRBC5_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK 0x00000003L +//UVD_JRBC5_UVD_JRBC_RB_REF_DATA +#define UVD_JRBC5_UVD_JRBC_RB_REF_DATA__REF_DATA__SHIFT 0x0 +#define UVD_JRBC5_UVD_JRBC_RB_REF_DATA__REF_DATA_MASK 0xFFFFFFFFL +//UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER +#define UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT 0x0 +#define UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT 0x10 +#define UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT 0x18 +#define UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT 0x19 +#define UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK 0x0000FFFFL +#define UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK 0x00FF0000L +#define UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK 0x01000000L +#define UVD_JRBC5_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK 0x02000000L +//UVD_JRBC5_UVD_JRBC_SOFT_RESET +#define UVD_JRBC5_UVD_JRBC_SOFT_RESET__RESET__SHIFT 0x0 +#define UVD_JRBC5_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS__SHIFT 0x11 +#define UVD_JRBC5_UVD_JRBC_SOFT_RESET__RESET_MASK 0x00000001L +#define UVD_JRBC5_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS_MASK 0x00020000L +//UVD_JRBC5_UVD_JRBC_STATUS +#define UVD_JRBC5_UVD_JRBC_STATUS__RB_JOB_DONE__SHIFT 0x0 +#define UVD_JRBC5_UVD_JRBC_STATUS__IB_JOB_DONE__SHIFT 0x1 +#define UVD_JRBC5_UVD_JRBC_STATUS__RB_ILLEGAL_CMD__SHIFT 0x2 +#define UVD_JRBC5_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT__SHIFT 0x3 +#define UVD_JRBC5_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT__SHIFT 0x4 +#define UVD_JRBC5_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT__SHIFT 0x5 +#define UVD_JRBC5_UVD_JRBC_STATUS__IB_ILLEGAL_CMD__SHIFT 0x6 +#define UVD_JRBC5_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT__SHIFT 0x7 +#define UVD_JRBC5_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT__SHIFT 0x8 +#define UVD_JRBC5_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT__SHIFT 0x9 +#define UVD_JRBC5_UVD_JRBC_STATUS__RB_TRAP_STATUS__SHIFT 0xa +#define UVD_JRBC5_UVD_JRBC_STATUS__PREEMPT_STATUS__SHIFT 0xb +#define UVD_JRBC5_UVD_JRBC_STATUS__IB_TRAP_STATUS__SHIFT 0xc +#define UVD_JRBC5_UVD_JRBC_STATUS__INT_EN__SHIFT 0x10 +#define UVD_JRBC5_UVD_JRBC_STATUS__INT_ACK__SHIFT 0x11 +#define UVD_JRBC5_UVD_JRBC_STATUS__RB_JOB_DONE_MASK 0x00000001L +#define UVD_JRBC5_UVD_JRBC_STATUS__IB_JOB_DONE_MASK 0x00000002L +#define UVD_JRBC5_UVD_JRBC_STATUS__RB_ILLEGAL_CMD_MASK 0x00000004L +#define UVD_JRBC5_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT_MASK 0x00000008L +#define UVD_JRBC5_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT_MASK 0x00000010L +#define UVD_JRBC5_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT_MASK 0x00000020L +#define UVD_JRBC5_UVD_JRBC_STATUS__IB_ILLEGAL_CMD_MASK 0x00000040L +#define UVD_JRBC5_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT_MASK 0x00000080L +#define UVD_JRBC5_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT_MASK 0x00000100L +#define UVD_JRBC5_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT_MASK 0x00000200L +#define UVD_JRBC5_UVD_JRBC_STATUS__RB_TRAP_STATUS_MASK 0x00000400L +#define UVD_JRBC5_UVD_JRBC_STATUS__PREEMPT_STATUS_MASK 0x00000800L +#define UVD_JRBC5_UVD_JRBC_STATUS__IB_TRAP_STATUS_MASK 0x00001000L +#define UVD_JRBC5_UVD_JRBC_STATUS__INT_EN_MASK 0x00010000L +#define UVD_JRBC5_UVD_JRBC_STATUS__INT_ACK_MASK 0x00020000L +//UVD_JRBC5_UVD_JRBC_RB_RPTR +#define UVD_JRBC5_UVD_JRBC_RB_RPTR__RB_RPTR__SHIFT 0x4 +#define UVD_JRBC5_UVD_JRBC_RB_RPTR__RB_RPTR_MASK 0x007FFFF0L +//UVD_JRBC5_UVD_JRBC_RB_BUF_STATUS +#define UVD_JRBC5_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID__SHIFT 0x0 +#define UVD_JRBC5_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT 0x10 +#define UVD_JRBC5_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT 0x18 +#define UVD_JRBC5_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID_MASK 0x0000FFFFL +#define UVD_JRBC5_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR_MASK 0x000F0000L +#define UVD_JRBC5_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR_MASK 0x03000000L +//UVD_JRBC5_UVD_JRBC_IB_BUF_STATUS +#define UVD_JRBC5_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID__SHIFT 0x0 +#define UVD_JRBC5_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT 0x10 +#define UVD_JRBC5_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT 0x18 +#define UVD_JRBC5_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID_MASK 0x0000FFFFL +#define UVD_JRBC5_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR_MASK 0x000F0000L +#define UVD_JRBC5_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR_MASK 0x03000000L +//UVD_JRBC5_UVD_JRBC_IB_SIZE_UPDATE +#define UVD_JRBC5_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT 0x4 +#define UVD_JRBC5_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK 0x007FFFF0L +//UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER +#define UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT 0x0 +#define UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT 0x10 +#define UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT 0x18 +#define UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT 0x19 +#define UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK 0x0000FFFFL +#define UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK 0x00FF0000L +#define UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK 0x01000000L +#define UVD_JRBC5_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK 0x02000000L +//UVD_JRBC5_UVD_JRBC_IB_REF_DATA +#define UVD_JRBC5_UVD_JRBC_IB_REF_DATA__REF_DATA__SHIFT 0x0 +#define UVD_JRBC5_UVD_JRBC_IB_REF_DATA__REF_DATA_MASK 0xFFFFFFFFL +//UVD_JRBC5_UVD_JPEG_PREEMPT_CMD +#define UVD_JRBC5_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN__SHIFT 0x0 +#define UVD_JRBC5_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE__SHIFT 0x1 +#define UVD_JRBC5_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD__SHIFT 0x2 +#define UVD_JRBC5_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN_MASK 0x00000001L +#define UVD_JRBC5_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE_MASK 0x00000002L +#define UVD_JRBC5_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD_MASK 0x00000004L +//UVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA0 +#define UVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0__SHIFT 0x0 +#define UVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0_MASK 0xFFFFFFFFL +//UVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA1 +#define UVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1__SHIFT 0x0 +#define UVD_JRBC5_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1_MASK 0xFFFFFFFFL +//UVD_JRBC5_UVD_JRBC_RB_SIZE +#define UVD_JRBC5_UVD_JRBC_RB_SIZE__RB_SIZE__SHIFT 0x4 +#define UVD_JRBC5_UVD_JRBC_RB_SIZE__RB_SIZE_MASK 0x00FFFFF0L +//UVD_JRBC5_UVD_JRBC_SCRATCH0 +#define UVD_JRBC5_UVD_JRBC_SCRATCH0__SCRATCH0__SHIFT 0x0 +#define UVD_JRBC5_UVD_JRBC_SCRATCH0__SCRATCH0_MASK 0xFFFFFFFFL + + +// addressBlock: aid_uvd0_uvd_jrbc6_uvd_jrbc_dec +//UVD_JRBC6_UVD_JRBC_RB_WPTR +#define UVD_JRBC6_UVD_JRBC_RB_WPTR__RB_WPTR__SHIFT 0x4 +#define UVD_JRBC6_UVD_JRBC_RB_WPTR__RB_WPTR_MASK 0x007FFFF0L +//UVD_JRBC6_UVD_JRBC_RB_CNTL +#define UVD_JRBC6_UVD_JRBC_RB_CNTL__RB_NO_FETCH__SHIFT 0x0 +#define UVD_JRBC6_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT 0x1 +#define UVD_JRBC6_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER__SHIFT 0x4 +#define UVD_JRBC6_UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK 0x00000001L +#define UVD_JRBC6_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK 0x00000002L +#define UVD_JRBC6_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER_MASK 0x0007FFF0L +//UVD_JRBC6_UVD_JRBC_IB_SIZE +#define UVD_JRBC6_UVD_JRBC_IB_SIZE__IB_SIZE__SHIFT 0x4 +#define UVD_JRBC6_UVD_JRBC_IB_SIZE__IB_SIZE_MASK 0x007FFFF0L +//UVD_JRBC6_UVD_JRBC_URGENT_CNTL +#define UVD_JRBC6_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT 0x0 +#define UVD_JRBC6_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK 0x00000003L +//UVD_JRBC6_UVD_JRBC_RB_REF_DATA +#define UVD_JRBC6_UVD_JRBC_RB_REF_DATA__REF_DATA__SHIFT 0x0 +#define UVD_JRBC6_UVD_JRBC_RB_REF_DATA__REF_DATA_MASK 0xFFFFFFFFL +//UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER +#define UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT 0x0 +#define UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT 0x10 +#define UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT 0x18 +#define UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT 0x19 +#define UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK 0x0000FFFFL +#define UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK 0x00FF0000L +#define UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK 0x01000000L +#define UVD_JRBC6_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK 0x02000000L +//UVD_JRBC6_UVD_JRBC_SOFT_RESET +#define UVD_JRBC6_UVD_JRBC_SOFT_RESET__RESET__SHIFT 0x0 +#define UVD_JRBC6_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS__SHIFT 0x11 +#define UVD_JRBC6_UVD_JRBC_SOFT_RESET__RESET_MASK 0x00000001L +#define UVD_JRBC6_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS_MASK 0x00020000L +//UVD_JRBC6_UVD_JRBC_STATUS +#define UVD_JRBC6_UVD_JRBC_STATUS__RB_JOB_DONE__SHIFT 0x0 +#define UVD_JRBC6_UVD_JRBC_STATUS__IB_JOB_DONE__SHIFT 0x1 +#define UVD_JRBC6_UVD_JRBC_STATUS__RB_ILLEGAL_CMD__SHIFT 0x2 +#define UVD_JRBC6_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT__SHIFT 0x3 +#define UVD_JRBC6_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT__SHIFT 0x4 +#define UVD_JRBC6_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT__SHIFT 0x5 +#define UVD_JRBC6_UVD_JRBC_STATUS__IB_ILLEGAL_CMD__SHIFT 0x6 +#define UVD_JRBC6_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT__SHIFT 0x7 +#define UVD_JRBC6_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT__SHIFT 0x8 +#define UVD_JRBC6_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT__SHIFT 0x9 +#define UVD_JRBC6_UVD_JRBC_STATUS__RB_TRAP_STATUS__SHIFT 0xa +#define UVD_JRBC6_UVD_JRBC_STATUS__PREEMPT_STATUS__SHIFT 0xb +#define UVD_JRBC6_UVD_JRBC_STATUS__IB_TRAP_STATUS__SHIFT 0xc +#define UVD_JRBC6_UVD_JRBC_STATUS__INT_EN__SHIFT 0x10 +#define UVD_JRBC6_UVD_JRBC_STATUS__INT_ACK__SHIFT 0x11 +#define UVD_JRBC6_UVD_JRBC_STATUS__RB_JOB_DONE_MASK 0x00000001L +#define UVD_JRBC6_UVD_JRBC_STATUS__IB_JOB_DONE_MASK 0x00000002L +#define UVD_JRBC6_UVD_JRBC_STATUS__RB_ILLEGAL_CMD_MASK 0x00000004L +#define UVD_JRBC6_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT_MASK 0x00000008L +#define UVD_JRBC6_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT_MASK 0x00000010L +#define UVD_JRBC6_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT_MASK 0x00000020L +#define UVD_JRBC6_UVD_JRBC_STATUS__IB_ILLEGAL_CMD_MASK 0x00000040L +#define UVD_JRBC6_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT_MASK 0x00000080L +#define UVD_JRBC6_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT_MASK 0x00000100L +#define UVD_JRBC6_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT_MASK 0x00000200L +#define UVD_JRBC6_UVD_JRBC_STATUS__RB_TRAP_STATUS_MASK 0x00000400L +#define UVD_JRBC6_UVD_JRBC_STATUS__PREEMPT_STATUS_MASK 0x00000800L +#define UVD_JRBC6_UVD_JRBC_STATUS__IB_TRAP_STATUS_MASK 0x00001000L +#define UVD_JRBC6_UVD_JRBC_STATUS__INT_EN_MASK 0x00010000L +#define UVD_JRBC6_UVD_JRBC_STATUS__INT_ACK_MASK 0x00020000L +//UVD_JRBC6_UVD_JRBC_RB_RPTR +#define UVD_JRBC6_UVD_JRBC_RB_RPTR__RB_RPTR__SHIFT 0x4 +#define UVD_JRBC6_UVD_JRBC_RB_RPTR__RB_RPTR_MASK 0x007FFFF0L +//UVD_JRBC6_UVD_JRBC_RB_BUF_STATUS +#define UVD_JRBC6_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID__SHIFT 0x0 +#define UVD_JRBC6_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT 0x10 +#define UVD_JRBC6_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT 0x18 +#define UVD_JRBC6_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID_MASK 0x0000FFFFL +#define UVD_JRBC6_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR_MASK 0x000F0000L +#define UVD_JRBC6_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR_MASK 0x03000000L +//UVD_JRBC6_UVD_JRBC_IB_BUF_STATUS +#define UVD_JRBC6_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID__SHIFT 0x0 +#define UVD_JRBC6_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT 0x10 +#define UVD_JRBC6_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT 0x18 +#define UVD_JRBC6_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID_MASK 0x0000FFFFL +#define UVD_JRBC6_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR_MASK 0x000F0000L +#define UVD_JRBC6_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR_MASK 0x03000000L +//UVD_JRBC6_UVD_JRBC_IB_SIZE_UPDATE +#define UVD_JRBC6_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT 0x4 +#define UVD_JRBC6_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK 0x007FFFF0L +//UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER +#define UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT 0x0 +#define UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT 0x10 +#define UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT 0x18 +#define UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT 0x19 +#define UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK 0x0000FFFFL +#define UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK 0x00FF0000L +#define UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK 0x01000000L +#define UVD_JRBC6_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK 0x02000000L +//UVD_JRBC6_UVD_JRBC_IB_REF_DATA +#define UVD_JRBC6_UVD_JRBC_IB_REF_DATA__REF_DATA__SHIFT 0x0 +#define UVD_JRBC6_UVD_JRBC_IB_REF_DATA__REF_DATA_MASK 0xFFFFFFFFL +//UVD_JRBC6_UVD_JPEG_PREEMPT_CMD +#define UVD_JRBC6_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN__SHIFT 0x0 +#define UVD_JRBC6_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE__SHIFT 0x1 +#define UVD_JRBC6_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD__SHIFT 0x2 +#define UVD_JRBC6_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN_MASK 0x00000001L +#define UVD_JRBC6_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE_MASK 0x00000002L +#define UVD_JRBC6_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD_MASK 0x00000004L +//UVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA0 +#define UVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0__SHIFT 0x0 +#define UVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0_MASK 0xFFFFFFFFL +//UVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA1 +#define UVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1__SHIFT 0x0 +#define UVD_JRBC6_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1_MASK 0xFFFFFFFFL +//UVD_JRBC6_UVD_JRBC_RB_SIZE +#define UVD_JRBC6_UVD_JRBC_RB_SIZE__RB_SIZE__SHIFT 0x4 +#define UVD_JRBC6_UVD_JRBC_RB_SIZE__RB_SIZE_MASK 0x00FFFFF0L +//UVD_JRBC6_UVD_JRBC_SCRATCH0 +#define UVD_JRBC6_UVD_JRBC_SCRATCH0__SCRATCH0__SHIFT 0x0 +#define UVD_JRBC6_UVD_JRBC_SCRATCH0__SCRATCH0_MASK 0xFFFFFFFFL + + +// addressBlock: aid_uvd0_uvd_jrbc7_uvd_jrbc_dec +//UVD_JRBC7_UVD_JRBC_RB_WPTR +#define UVD_JRBC7_UVD_JRBC_RB_WPTR__RB_WPTR__SHIFT 0x4 +#define UVD_JRBC7_UVD_JRBC_RB_WPTR__RB_WPTR_MASK 0x007FFFF0L +//UVD_JRBC7_UVD_JRBC_RB_CNTL +#define UVD_JRBC7_UVD_JRBC_RB_CNTL__RB_NO_FETCH__SHIFT 0x0 +#define UVD_JRBC7_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN__SHIFT 0x1 +#define UVD_JRBC7_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER__SHIFT 0x4 +#define UVD_JRBC7_UVD_JRBC_RB_CNTL__RB_NO_FETCH_MASK 0x00000001L +#define UVD_JRBC7_UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK 0x00000002L +#define UVD_JRBC7_UVD_JRBC_RB_CNTL__RB_PRE_WRITE_TIMER_MASK 0x0007FFF0L +//UVD_JRBC7_UVD_JRBC_IB_SIZE +#define UVD_JRBC7_UVD_JRBC_IB_SIZE__IB_SIZE__SHIFT 0x4 +#define UVD_JRBC7_UVD_JRBC_IB_SIZE__IB_SIZE_MASK 0x007FFFF0L +//UVD_JRBC7_UVD_JRBC_URGENT_CNTL +#define UVD_JRBC7_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK__SHIFT 0x0 +#define UVD_JRBC7_UVD_JRBC_URGENT_CNTL__CMD_READ_REQ_PRIORITY_MARK_MASK 0x00000003L +//UVD_JRBC7_UVD_JRBC_RB_REF_DATA +#define UVD_JRBC7_UVD_JRBC_RB_REF_DATA__REF_DATA__SHIFT 0x0 +#define UVD_JRBC7_UVD_JRBC_RB_REF_DATA__REF_DATA_MASK 0xFFFFFFFFL +//UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER +#define UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT 0x0 +#define UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT 0x10 +#define UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT 0x18 +#define UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT 0x19 +#define UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK 0x0000FFFFL +#define UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK 0x00FF0000L +#define UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK 0x01000000L +#define UVD_JRBC7_UVD_JRBC_RB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK 0x02000000L +//UVD_JRBC7_UVD_JRBC_SOFT_RESET +#define UVD_JRBC7_UVD_JRBC_SOFT_RESET__RESET__SHIFT 0x0 +#define UVD_JRBC7_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS__SHIFT 0x11 +#define UVD_JRBC7_UVD_JRBC_SOFT_RESET__RESET_MASK 0x00000001L +#define UVD_JRBC7_UVD_JRBC_SOFT_RESET__SCLK_RESET_STATUS_MASK 0x00020000L +//UVD_JRBC7_UVD_JRBC_STATUS +#define UVD_JRBC7_UVD_JRBC_STATUS__RB_JOB_DONE__SHIFT 0x0 +#define UVD_JRBC7_UVD_JRBC_STATUS__IB_JOB_DONE__SHIFT 0x1 +#define UVD_JRBC7_UVD_JRBC_STATUS__RB_ILLEGAL_CMD__SHIFT 0x2 +#define UVD_JRBC7_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT__SHIFT 0x3 +#define UVD_JRBC7_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT__SHIFT 0x4 +#define UVD_JRBC7_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT__SHIFT 0x5 +#define UVD_JRBC7_UVD_JRBC_STATUS__IB_ILLEGAL_CMD__SHIFT 0x6 +#define UVD_JRBC7_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT__SHIFT 0x7 +#define UVD_JRBC7_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT__SHIFT 0x8 +#define UVD_JRBC7_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT__SHIFT 0x9 +#define UVD_JRBC7_UVD_JRBC_STATUS__RB_TRAP_STATUS__SHIFT 0xa +#define UVD_JRBC7_UVD_JRBC_STATUS__PREEMPT_STATUS__SHIFT 0xb +#define UVD_JRBC7_UVD_JRBC_STATUS__IB_TRAP_STATUS__SHIFT 0xc +#define UVD_JRBC7_UVD_JRBC_STATUS__INT_EN__SHIFT 0x10 +#define UVD_JRBC7_UVD_JRBC_STATUS__INT_ACK__SHIFT 0x11 +#define UVD_JRBC7_UVD_JRBC_STATUS__RB_JOB_DONE_MASK 0x00000001L +#define UVD_JRBC7_UVD_JRBC_STATUS__IB_JOB_DONE_MASK 0x00000002L +#define UVD_JRBC7_UVD_JRBC_STATUS__RB_ILLEGAL_CMD_MASK 0x00000004L +#define UVD_JRBC7_UVD_JRBC_STATUS__RB_COND_REG_RD_TIMEOUT_MASK 0x00000008L +#define UVD_JRBC7_UVD_JRBC_STATUS__RB_MEM_WR_TIMEOUT_MASK 0x00000010L +#define UVD_JRBC7_UVD_JRBC_STATUS__RB_MEM_RD_TIMEOUT_MASK 0x00000020L +#define UVD_JRBC7_UVD_JRBC_STATUS__IB_ILLEGAL_CMD_MASK 0x00000040L +#define UVD_JRBC7_UVD_JRBC_STATUS__IB_COND_REG_RD_TIMEOUT_MASK 0x00000080L +#define UVD_JRBC7_UVD_JRBC_STATUS__IB_MEM_WR_TIMEOUT_MASK 0x00000100L +#define UVD_JRBC7_UVD_JRBC_STATUS__IB_MEM_RD_TIMEOUT_MASK 0x00000200L +#define UVD_JRBC7_UVD_JRBC_STATUS__RB_TRAP_STATUS_MASK 0x00000400L +#define UVD_JRBC7_UVD_JRBC_STATUS__PREEMPT_STATUS_MASK 0x00000800L +#define UVD_JRBC7_UVD_JRBC_STATUS__IB_TRAP_STATUS_MASK 0x00001000L +#define UVD_JRBC7_UVD_JRBC_STATUS__INT_EN_MASK 0x00010000L +#define UVD_JRBC7_UVD_JRBC_STATUS__INT_ACK_MASK 0x00020000L +//UVD_JRBC7_UVD_JRBC_RB_RPTR +#define UVD_JRBC7_UVD_JRBC_RB_RPTR__RB_RPTR__SHIFT 0x4 +#define UVD_JRBC7_UVD_JRBC_RB_RPTR__RB_RPTR_MASK 0x007FFFF0L +//UVD_JRBC7_UVD_JRBC_RB_BUF_STATUS +#define UVD_JRBC7_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID__SHIFT 0x0 +#define UVD_JRBC7_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR__SHIFT 0x10 +#define UVD_JRBC7_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR__SHIFT 0x18 +#define UVD_JRBC7_UVD_JRBC_RB_BUF_STATUS__RB_BUF_VALID_MASK 0x0000FFFFL +#define UVD_JRBC7_UVD_JRBC_RB_BUF_STATUS__RB_BUF_RD_ADDR_MASK 0x000F0000L +#define UVD_JRBC7_UVD_JRBC_RB_BUF_STATUS__RB_BUF_WR_ADDR_MASK 0x03000000L +//UVD_JRBC7_UVD_JRBC_IB_BUF_STATUS +#define UVD_JRBC7_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID__SHIFT 0x0 +#define UVD_JRBC7_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR__SHIFT 0x10 +#define UVD_JRBC7_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR__SHIFT 0x18 +#define UVD_JRBC7_UVD_JRBC_IB_BUF_STATUS__IB_BUF_VALID_MASK 0x0000FFFFL +#define UVD_JRBC7_UVD_JRBC_IB_BUF_STATUS__IB_BUF_RD_ADDR_MASK 0x000F0000L +#define UVD_JRBC7_UVD_JRBC_IB_BUF_STATUS__IB_BUF_WR_ADDR_MASK 0x03000000L +//UVD_JRBC7_UVD_JRBC_IB_SIZE_UPDATE +#define UVD_JRBC7_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE__SHIFT 0x4 +#define UVD_JRBC7_UVD_JRBC_IB_SIZE_UPDATE__REMAIN_IB_SIZE_MASK 0x007FFFF0L +//UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER +#define UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT__SHIFT 0x0 +#define UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT__SHIFT 0x10 +#define UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN__SHIFT 0x18 +#define UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN__SHIFT 0x19 +#define UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER__RETRY_TIMER_CNT_MASK 0x0000FFFFL +#define UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER__RETRY_INTERVAL_CNT_MASK 0x00FF0000L +#define UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER__CONTINUOUS_POLL_EN_MASK 0x01000000L +#define UVD_JRBC7_UVD_JRBC_IB_COND_RD_TIMER__MEM_TIMEOUT_EN_MASK 0x02000000L +//UVD_JRBC7_UVD_JRBC_IB_REF_DATA +#define UVD_JRBC7_UVD_JRBC_IB_REF_DATA__REF_DATA__SHIFT 0x0 +#define UVD_JRBC7_UVD_JRBC_IB_REF_DATA__REF_DATA_MASK 0xFFFFFFFFL +//UVD_JRBC7_UVD_JPEG_PREEMPT_CMD +#define UVD_JRBC7_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN__SHIFT 0x0 +#define UVD_JRBC7_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE__SHIFT 0x1 +#define UVD_JRBC7_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD__SHIFT 0x2 +#define UVD_JRBC7_UVD_JPEG_PREEMPT_CMD__PREEMPT_EN_MASK 0x00000001L +#define UVD_JRBC7_UVD_JPEG_PREEMPT_CMD__WAIT_JPEG_JOB_DONE_MASK 0x00000002L +#define UVD_JRBC7_UVD_JPEG_PREEMPT_CMD__PREEMPT_FENCE_CMD_MASK 0x00000004L +//UVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA0 +#define UVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0__SHIFT 0x0 +#define UVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA0__PREEMPT_FENCE_DATA0_MASK 0xFFFFFFFFL +//UVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA1 +#define UVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1__SHIFT 0x0 +#define UVD_JRBC7_UVD_JPEG_PREEMPT_FENCE_DATA1__PREEMPT_FENCE_DATA1_MASK 0xFFFFFFFFL +//UVD_JRBC7_UVD_JRBC_RB_SIZE +#define UVD_JRBC7_UVD_JRBC_RB_SIZE__RB_SIZE__SHIFT 0x4 +#define UVD_JRBC7_UVD_JRBC_RB_SIZE__RB_SIZE_MASK 0x00FFFFF0L +//UVD_JRBC7_UVD_JRBC_SCRATCH0 +#define UVD_JRBC7_UVD_JRBC_SCRATCH0__SCRATCH0__SHIFT 0x0 +#define UVD_JRBC7_UVD_JRBC_SCRATCH0__SCRATCH0_MASK 0xFFFFFFFFL + + +// addressBlock: aid_uvd0_uvd_jmi1_uvd_jmi_dec +//UVD_JMI1_UVD_JPEG_DEC_PF_CTRL +#define UVD_JMI1_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS__SHIFT 0x0 +#define UVD_JMI1_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING__SHIFT 0x1 +#define UVD_JMI1_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS_MASK 0x00000001L +#define UVD_JMI1_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING_MASK 0x00000002L +//UVD_JMI1_UVD_LMI_JRBC_CTRL +#define UVD_JMI1_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN__SHIFT 0x1 +#define UVD_JMI1_UVD_LMI_JRBC_CTRL__RD_MAX_BURST__SHIFT 0x4 +#define UVD_JMI1_UVD_LMI_JRBC_CTRL__WR_MAX_BURST__SHIFT 0x8 +#define UVD_JMI1_UVD_LMI_JRBC_CTRL__RD_SWAP__SHIFT 0x14 +#define UVD_JMI1_UVD_LMI_JRBC_CTRL__WR_SWAP__SHIFT 0x16 +#define UVD_JMI1_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN_MASK 0x00000001L +#define UVD_JMI1_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN_MASK 0x00000002L +#define UVD_JMI1_UVD_LMI_JRBC_CTRL__RD_MAX_BURST_MASK 0x000000F0L +#define UVD_JMI1_UVD_LMI_JRBC_CTRL__WR_MAX_BURST_MASK 0x00000F00L +#define UVD_JMI1_UVD_LMI_JRBC_CTRL__RD_SWAP_MASK 0x00300000L +#define UVD_JMI1_UVD_LMI_JRBC_CTRL__WR_SWAP_MASK 0x00C00000L +//UVD_JMI1_UVD_LMI_JPEG_CTRL +#define UVD_JMI1_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN__SHIFT 0x1 +#define UVD_JMI1_UVD_LMI_JPEG_CTRL__RD_MAX_BURST__SHIFT 0x4 +#define UVD_JMI1_UVD_LMI_JPEG_CTRL__WR_MAX_BURST__SHIFT 0x8 +#define UVD_JMI1_UVD_LMI_JPEG_CTRL__RD_SWAP__SHIFT 0x14 +#define UVD_JMI1_UVD_LMI_JPEG_CTRL__WR_SWAP__SHIFT 0x16 +#define UVD_JMI1_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN_MASK 0x00000001L +#define UVD_JMI1_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN_MASK 0x00000002L +#define UVD_JMI1_UVD_LMI_JPEG_CTRL__RD_MAX_BURST_MASK 0x000000F0L +#define UVD_JMI1_UVD_LMI_JPEG_CTRL__WR_MAX_BURST_MASK 0x00000F00L +#define UVD_JMI1_UVD_LMI_JPEG_CTRL__RD_SWAP_MASK 0x00300000L +#define UVD_JMI1_UVD_LMI_JPEG_CTRL__WR_SWAP_MASK 0x00C00000L +//UVD_JMI1_JPEG_LMI_DROP +#define UVD_JMI1_JPEG_LMI_DROP__JPEG_WR_DROP__SHIFT 0x0 +#define UVD_JMI1_JPEG_LMI_DROP__JRBC_WR_DROP__SHIFT 0x1 +#define UVD_JMI1_JPEG_LMI_DROP__JPEG_RD_DROP__SHIFT 0x2 +#define UVD_JMI1_JPEG_LMI_DROP__JRBC_RD_DROP__SHIFT 0x3 +#define UVD_JMI1_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP__SHIFT 0x4 +#define UVD_JMI1_JPEG_LMI_DROP__JPEG_WR_DROP_MASK 0x00000001L +#define UVD_JMI1_JPEG_LMI_DROP__JRBC_WR_DROP_MASK 0x00000002L +#define UVD_JMI1_JPEG_LMI_DROP__JPEG_RD_DROP_MASK 0x00000004L +#define UVD_JMI1_JPEG_LMI_DROP__JRBC_RD_DROP_MASK 0x00000008L +#define UVD_JMI1_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP_MASK 0x00000010L +//UVD_JMI1_UVD_LMI_JRBC_IB_VMID +#define UVD_JMI1_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID__SHIFT 0x4 +#define UVD_JMI1_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID__SHIFT 0x8 +#define UVD_JMI1_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID_MASK 0x0000000FL +#define UVD_JMI1_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID_MASK 0x000000F0L +#define UVD_JMI1_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID_MASK 0x00000F00L +//UVD_JMI1_UVD_LMI_JRBC_RB_VMID +#define UVD_JMI1_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID__SHIFT 0x4 +#define UVD_JMI1_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID__SHIFT 0x8 +#define UVD_JMI1_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID_MASK 0x0000000FL +#define UVD_JMI1_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID_MASK 0x000000F0L +#define UVD_JMI1_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID_MASK 0x00000F00L +//UVD_JMI1_UVD_LMI_JPEG_VMID +#define UVD_JMI1_UVD_LMI_JPEG_VMID__JPEG_RD_VMID__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JPEG_VMID__JPEG_WR_VMID__SHIFT 0x4 +#define UVD_JMI1_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID__SHIFT 0x8 +#define UVD_JMI1_UVD_LMI_JPEG_VMID__JPEG_RD_VMID_MASK 0x0000000FL +#define UVD_JMI1_UVD_LMI_JPEG_VMID__JPEG_WR_VMID_MASK 0x000000F0L +#define UVD_JMI1_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID_MASK 0x00000F00L +//UVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW +#define UVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH +#define UVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_LOW +#define UVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH +#define UVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW +#define UVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH +#define UVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI1_UVD_LMI_JPEG_PREEMPT_VMID +#define UVD_JMI1_UVD_LMI_JPEG_PREEMPT_VMID__VMID__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JPEG_PREEMPT_VMID__VMID_MASK 0x0000000FL +//UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP__SHIFT 0x0 +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP__SHIFT 0x2 +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP__SHIFT 0x4 +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP__SHIFT 0x6 +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP__SHIFT 0x8 +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP__SHIFT 0xa +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP__SHIFT 0xc +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP__SHIFT 0xe +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP__SHIFT 0x10 +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP_MASK 0x00000003L +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP_MASK 0x0000000CL +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP_MASK 0x00000030L +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP_MASK 0x000000C0L +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP_MASK 0x00000300L +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP_MASK 0x00000C00L +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP_MASK 0x00003000L +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP_MASK 0x0000C000L +#define UVD_JMI1_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP_MASK 0x00030000L +//UVD_JMI1_UVD_JMI_ATOMIC_CNTL +#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en__SHIFT 0x0 +#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__atomic_max_burst__SHIFT 0x1 +#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop__SHIFT 0x5 +#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en__SHIFT 0x6 +#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG__SHIFT 0x7 +#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE__SHIFT 0xb +#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en_MASK 0x00000001L +#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__atomic_max_burst_MASK 0x0000001EL +#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop_MASK 0x00000020L +#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en_MASK 0x00000040L +#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG_MASK 0x00000780L +#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE_MASK 0x00000800L +//UVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW +#define UVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH +#define UVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI1_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_LOW +#define UVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH +#define UVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW +#define UVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH +#define UVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_LOW +#define UVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH +#define UVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW +#define UVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH +#define UVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI1_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI1_UVD_JMI_ATOMIC_CNTL2 +#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap__SHIFT 0x10 +#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP__SHIFT 0x18 +#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap_MASK 0x00FF0000L +#define UVD_JMI1_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP_MASK 0xFF000000L + + +// addressBlock: aid_uvd0_uvd_jmi2_uvd_jmi_dec +//UVD_JMI2_UVD_JPEG_DEC_PF_CTRL +#define UVD_JMI2_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS__SHIFT 0x0 +#define UVD_JMI2_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING__SHIFT 0x1 +#define UVD_JMI2_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS_MASK 0x00000001L +#define UVD_JMI2_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING_MASK 0x00000002L +//UVD_JMI2_UVD_LMI_JRBC_CTRL +#define UVD_JMI2_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN__SHIFT 0x1 +#define UVD_JMI2_UVD_LMI_JRBC_CTRL__RD_MAX_BURST__SHIFT 0x4 +#define UVD_JMI2_UVD_LMI_JRBC_CTRL__WR_MAX_BURST__SHIFT 0x8 +#define UVD_JMI2_UVD_LMI_JRBC_CTRL__RD_SWAP__SHIFT 0x14 +#define UVD_JMI2_UVD_LMI_JRBC_CTRL__WR_SWAP__SHIFT 0x16 +#define UVD_JMI2_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN_MASK 0x00000001L +#define UVD_JMI2_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN_MASK 0x00000002L +#define UVD_JMI2_UVD_LMI_JRBC_CTRL__RD_MAX_BURST_MASK 0x000000F0L +#define UVD_JMI2_UVD_LMI_JRBC_CTRL__WR_MAX_BURST_MASK 0x00000F00L +#define UVD_JMI2_UVD_LMI_JRBC_CTRL__RD_SWAP_MASK 0x00300000L +#define UVD_JMI2_UVD_LMI_JRBC_CTRL__WR_SWAP_MASK 0x00C00000L +//UVD_JMI2_UVD_LMI_JPEG_CTRL +#define UVD_JMI2_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN__SHIFT 0x1 +#define UVD_JMI2_UVD_LMI_JPEG_CTRL__RD_MAX_BURST__SHIFT 0x4 +#define UVD_JMI2_UVD_LMI_JPEG_CTRL__WR_MAX_BURST__SHIFT 0x8 +#define UVD_JMI2_UVD_LMI_JPEG_CTRL__RD_SWAP__SHIFT 0x14 +#define UVD_JMI2_UVD_LMI_JPEG_CTRL__WR_SWAP__SHIFT 0x16 +#define UVD_JMI2_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN_MASK 0x00000001L +#define UVD_JMI2_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN_MASK 0x00000002L +#define UVD_JMI2_UVD_LMI_JPEG_CTRL__RD_MAX_BURST_MASK 0x000000F0L +#define UVD_JMI2_UVD_LMI_JPEG_CTRL__WR_MAX_BURST_MASK 0x00000F00L +#define UVD_JMI2_UVD_LMI_JPEG_CTRL__RD_SWAP_MASK 0x00300000L +#define UVD_JMI2_UVD_LMI_JPEG_CTRL__WR_SWAP_MASK 0x00C00000L +//UVD_JMI2_JPEG_LMI_DROP +#define UVD_JMI2_JPEG_LMI_DROP__JPEG_WR_DROP__SHIFT 0x0 +#define UVD_JMI2_JPEG_LMI_DROP__JRBC_WR_DROP__SHIFT 0x1 +#define UVD_JMI2_JPEG_LMI_DROP__JPEG_RD_DROP__SHIFT 0x2 +#define UVD_JMI2_JPEG_LMI_DROP__JRBC_RD_DROP__SHIFT 0x3 +#define UVD_JMI2_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP__SHIFT 0x4 +#define UVD_JMI2_JPEG_LMI_DROP__JPEG_WR_DROP_MASK 0x00000001L +#define UVD_JMI2_JPEG_LMI_DROP__JRBC_WR_DROP_MASK 0x00000002L +#define UVD_JMI2_JPEG_LMI_DROP__JPEG_RD_DROP_MASK 0x00000004L +#define UVD_JMI2_JPEG_LMI_DROP__JRBC_RD_DROP_MASK 0x00000008L +#define UVD_JMI2_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP_MASK 0x00000010L +//UVD_JMI2_UVD_LMI_JRBC_IB_VMID +#define UVD_JMI2_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID__SHIFT 0x4 +#define UVD_JMI2_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID__SHIFT 0x8 +#define UVD_JMI2_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID_MASK 0x0000000FL +#define UVD_JMI2_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID_MASK 0x000000F0L +#define UVD_JMI2_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID_MASK 0x00000F00L +//UVD_JMI2_UVD_LMI_JRBC_RB_VMID +#define UVD_JMI2_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID__SHIFT 0x4 +#define UVD_JMI2_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID__SHIFT 0x8 +#define UVD_JMI2_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID_MASK 0x0000000FL +#define UVD_JMI2_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID_MASK 0x000000F0L +#define UVD_JMI2_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID_MASK 0x00000F00L +//UVD_JMI2_UVD_LMI_JPEG_VMID +#define UVD_JMI2_UVD_LMI_JPEG_VMID__JPEG_RD_VMID__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JPEG_VMID__JPEG_WR_VMID__SHIFT 0x4 +#define UVD_JMI2_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID__SHIFT 0x8 +#define UVD_JMI2_UVD_LMI_JPEG_VMID__JPEG_RD_VMID_MASK 0x0000000FL +#define UVD_JMI2_UVD_LMI_JPEG_VMID__JPEG_WR_VMID_MASK 0x000000F0L +#define UVD_JMI2_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID_MASK 0x00000F00L +//UVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW +#define UVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH +#define UVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_LOW +#define UVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH +#define UVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW +#define UVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH +#define UVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI2_UVD_LMI_JPEG_PREEMPT_VMID +#define UVD_JMI2_UVD_LMI_JPEG_PREEMPT_VMID__VMID__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JPEG_PREEMPT_VMID__VMID_MASK 0x0000000FL +//UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP__SHIFT 0x0 +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP__SHIFT 0x2 +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP__SHIFT 0x4 +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP__SHIFT 0x6 +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP__SHIFT 0x8 +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP__SHIFT 0xa +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP__SHIFT 0xc +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP__SHIFT 0xe +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP__SHIFT 0x10 +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP_MASK 0x00000003L +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP_MASK 0x0000000CL +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP_MASK 0x00000030L +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP_MASK 0x000000C0L +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP_MASK 0x00000300L +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP_MASK 0x00000C00L +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP_MASK 0x00003000L +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP_MASK 0x0000C000L +#define UVD_JMI2_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP_MASK 0x00030000L +//UVD_JMI2_UVD_JMI_ATOMIC_CNTL +#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en__SHIFT 0x0 +#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__atomic_max_burst__SHIFT 0x1 +#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop__SHIFT 0x5 +#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en__SHIFT 0x6 +#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG__SHIFT 0x7 +#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE__SHIFT 0xb +#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en_MASK 0x00000001L +#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__atomic_max_burst_MASK 0x0000001EL +#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop_MASK 0x00000020L +#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en_MASK 0x00000040L +#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG_MASK 0x00000780L +#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE_MASK 0x00000800L +//UVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW +#define UVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH +#define UVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI2_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_LOW +#define UVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH +#define UVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW +#define UVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH +#define UVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_LOW +#define UVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH +#define UVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW +#define UVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH +#define UVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI2_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI2_UVD_JMI_ATOMIC_CNTL2 +#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap__SHIFT 0x10 +#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP__SHIFT 0x18 +#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap_MASK 0x00FF0000L +#define UVD_JMI2_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP_MASK 0xFF000000L + + +// addressBlock: aid_uvd0_uvd_jmi3_uvd_jmi_dec +//UVD_JMI3_UVD_JPEG_DEC_PF_CTRL +#define UVD_JMI3_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS__SHIFT 0x0 +#define UVD_JMI3_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING__SHIFT 0x1 +#define UVD_JMI3_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS_MASK 0x00000001L +#define UVD_JMI3_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING_MASK 0x00000002L +//UVD_JMI3_UVD_LMI_JRBC_CTRL +#define UVD_JMI3_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN__SHIFT 0x1 +#define UVD_JMI3_UVD_LMI_JRBC_CTRL__RD_MAX_BURST__SHIFT 0x4 +#define UVD_JMI3_UVD_LMI_JRBC_CTRL__WR_MAX_BURST__SHIFT 0x8 +#define UVD_JMI3_UVD_LMI_JRBC_CTRL__RD_SWAP__SHIFT 0x14 +#define UVD_JMI3_UVD_LMI_JRBC_CTRL__WR_SWAP__SHIFT 0x16 +#define UVD_JMI3_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN_MASK 0x00000001L +#define UVD_JMI3_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN_MASK 0x00000002L +#define UVD_JMI3_UVD_LMI_JRBC_CTRL__RD_MAX_BURST_MASK 0x000000F0L +#define UVD_JMI3_UVD_LMI_JRBC_CTRL__WR_MAX_BURST_MASK 0x00000F00L +#define UVD_JMI3_UVD_LMI_JRBC_CTRL__RD_SWAP_MASK 0x00300000L +#define UVD_JMI3_UVD_LMI_JRBC_CTRL__WR_SWAP_MASK 0x00C00000L +//UVD_JMI3_UVD_LMI_JPEG_CTRL +#define UVD_JMI3_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN__SHIFT 0x1 +#define UVD_JMI3_UVD_LMI_JPEG_CTRL__RD_MAX_BURST__SHIFT 0x4 +#define UVD_JMI3_UVD_LMI_JPEG_CTRL__WR_MAX_BURST__SHIFT 0x8 +#define UVD_JMI3_UVD_LMI_JPEG_CTRL__RD_SWAP__SHIFT 0x14 +#define UVD_JMI3_UVD_LMI_JPEG_CTRL__WR_SWAP__SHIFT 0x16 +#define UVD_JMI3_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN_MASK 0x00000001L +#define UVD_JMI3_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN_MASK 0x00000002L +#define UVD_JMI3_UVD_LMI_JPEG_CTRL__RD_MAX_BURST_MASK 0x000000F0L +#define UVD_JMI3_UVD_LMI_JPEG_CTRL__WR_MAX_BURST_MASK 0x00000F00L +#define UVD_JMI3_UVD_LMI_JPEG_CTRL__RD_SWAP_MASK 0x00300000L +#define UVD_JMI3_UVD_LMI_JPEG_CTRL__WR_SWAP_MASK 0x00C00000L +//UVD_JMI3_JPEG_LMI_DROP +#define UVD_JMI3_JPEG_LMI_DROP__JPEG_WR_DROP__SHIFT 0x0 +#define UVD_JMI3_JPEG_LMI_DROP__JRBC_WR_DROP__SHIFT 0x1 +#define UVD_JMI3_JPEG_LMI_DROP__JPEG_RD_DROP__SHIFT 0x2 +#define UVD_JMI3_JPEG_LMI_DROP__JRBC_RD_DROP__SHIFT 0x3 +#define UVD_JMI3_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP__SHIFT 0x4 +#define UVD_JMI3_JPEG_LMI_DROP__JPEG_WR_DROP_MASK 0x00000001L +#define UVD_JMI3_JPEG_LMI_DROP__JRBC_WR_DROP_MASK 0x00000002L +#define UVD_JMI3_JPEG_LMI_DROP__JPEG_RD_DROP_MASK 0x00000004L +#define UVD_JMI3_JPEG_LMI_DROP__JRBC_RD_DROP_MASK 0x00000008L +#define UVD_JMI3_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP_MASK 0x00000010L +//UVD_JMI3_UVD_LMI_JRBC_IB_VMID +#define UVD_JMI3_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID__SHIFT 0x4 +#define UVD_JMI3_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID__SHIFT 0x8 +#define UVD_JMI3_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID_MASK 0x0000000FL +#define UVD_JMI3_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID_MASK 0x000000F0L +#define UVD_JMI3_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID_MASK 0x00000F00L +//UVD_JMI3_UVD_LMI_JRBC_RB_VMID +#define UVD_JMI3_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID__SHIFT 0x4 +#define UVD_JMI3_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID__SHIFT 0x8 +#define UVD_JMI3_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID_MASK 0x0000000FL +#define UVD_JMI3_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID_MASK 0x000000F0L +#define UVD_JMI3_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID_MASK 0x00000F00L +//UVD_JMI3_UVD_LMI_JPEG_VMID +#define UVD_JMI3_UVD_LMI_JPEG_VMID__JPEG_RD_VMID__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JPEG_VMID__JPEG_WR_VMID__SHIFT 0x4 +#define UVD_JMI3_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID__SHIFT 0x8 +#define UVD_JMI3_UVD_LMI_JPEG_VMID__JPEG_RD_VMID_MASK 0x0000000FL +#define UVD_JMI3_UVD_LMI_JPEG_VMID__JPEG_WR_VMID_MASK 0x000000F0L +#define UVD_JMI3_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID_MASK 0x00000F00L +//UVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW +#define UVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH +#define UVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_LOW +#define UVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH +#define UVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW +#define UVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH +#define UVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI3_UVD_LMI_JPEG_PREEMPT_VMID +#define UVD_JMI3_UVD_LMI_JPEG_PREEMPT_VMID__VMID__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JPEG_PREEMPT_VMID__VMID_MASK 0x0000000FL +//UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP__SHIFT 0x0 +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP__SHIFT 0x2 +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP__SHIFT 0x4 +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP__SHIFT 0x6 +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP__SHIFT 0x8 +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP__SHIFT 0xa +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP__SHIFT 0xc +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP__SHIFT 0xe +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP__SHIFT 0x10 +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP_MASK 0x00000003L +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP_MASK 0x0000000CL +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP_MASK 0x00000030L +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP_MASK 0x000000C0L +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP_MASK 0x00000300L +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP_MASK 0x00000C00L +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP_MASK 0x00003000L +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP_MASK 0x0000C000L +#define UVD_JMI3_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP_MASK 0x00030000L +//UVD_JMI3_UVD_JMI_ATOMIC_CNTL +#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en__SHIFT 0x0 +#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__atomic_max_burst__SHIFT 0x1 +#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop__SHIFT 0x5 +#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en__SHIFT 0x6 +#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG__SHIFT 0x7 +#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE__SHIFT 0xb +#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en_MASK 0x00000001L +#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__atomic_max_burst_MASK 0x0000001EL +#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop_MASK 0x00000020L +#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en_MASK 0x00000040L +#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG_MASK 0x00000780L +#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE_MASK 0x00000800L +//UVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW +#define UVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH +#define UVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI3_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_LOW +#define UVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH +#define UVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW +#define UVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH +#define UVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_LOW +#define UVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH +#define UVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW +#define UVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH +#define UVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI3_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI3_UVD_JMI_ATOMIC_CNTL2 +#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap__SHIFT 0x10 +#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP__SHIFT 0x18 +#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap_MASK 0x00FF0000L +#define UVD_JMI3_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP_MASK 0xFF000000L + + +// addressBlock: aid_uvd0_uvd_jmi4_uvd_jmi_dec +//UVD_JMI4_UVD_JPEG_DEC_PF_CTRL +#define UVD_JMI4_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS__SHIFT 0x0 +#define UVD_JMI4_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING__SHIFT 0x1 +#define UVD_JMI4_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS_MASK 0x00000001L +#define UVD_JMI4_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING_MASK 0x00000002L +//UVD_JMI4_UVD_LMI_JRBC_CTRL +#define UVD_JMI4_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN__SHIFT 0x1 +#define UVD_JMI4_UVD_LMI_JRBC_CTRL__RD_MAX_BURST__SHIFT 0x4 +#define UVD_JMI4_UVD_LMI_JRBC_CTRL__WR_MAX_BURST__SHIFT 0x8 +#define UVD_JMI4_UVD_LMI_JRBC_CTRL__RD_SWAP__SHIFT 0x14 +#define UVD_JMI4_UVD_LMI_JRBC_CTRL__WR_SWAP__SHIFT 0x16 +#define UVD_JMI4_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN_MASK 0x00000001L +#define UVD_JMI4_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN_MASK 0x00000002L +#define UVD_JMI4_UVD_LMI_JRBC_CTRL__RD_MAX_BURST_MASK 0x000000F0L +#define UVD_JMI4_UVD_LMI_JRBC_CTRL__WR_MAX_BURST_MASK 0x00000F00L +#define UVD_JMI4_UVD_LMI_JRBC_CTRL__RD_SWAP_MASK 0x00300000L +#define UVD_JMI4_UVD_LMI_JRBC_CTRL__WR_SWAP_MASK 0x00C00000L +//UVD_JMI4_UVD_LMI_JPEG_CTRL +#define UVD_JMI4_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN__SHIFT 0x1 +#define UVD_JMI4_UVD_LMI_JPEG_CTRL__RD_MAX_BURST__SHIFT 0x4 +#define UVD_JMI4_UVD_LMI_JPEG_CTRL__WR_MAX_BURST__SHIFT 0x8 +#define UVD_JMI4_UVD_LMI_JPEG_CTRL__RD_SWAP__SHIFT 0x14 +#define UVD_JMI4_UVD_LMI_JPEG_CTRL__WR_SWAP__SHIFT 0x16 +#define UVD_JMI4_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN_MASK 0x00000001L +#define UVD_JMI4_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN_MASK 0x00000002L +#define UVD_JMI4_UVD_LMI_JPEG_CTRL__RD_MAX_BURST_MASK 0x000000F0L +#define UVD_JMI4_UVD_LMI_JPEG_CTRL__WR_MAX_BURST_MASK 0x00000F00L +#define UVD_JMI4_UVD_LMI_JPEG_CTRL__RD_SWAP_MASK 0x00300000L +#define UVD_JMI4_UVD_LMI_JPEG_CTRL__WR_SWAP_MASK 0x00C00000L +//UVD_JMI4_JPEG_LMI_DROP +#define UVD_JMI4_JPEG_LMI_DROP__JPEG_WR_DROP__SHIFT 0x0 +#define UVD_JMI4_JPEG_LMI_DROP__JRBC_WR_DROP__SHIFT 0x1 +#define UVD_JMI4_JPEG_LMI_DROP__JPEG_RD_DROP__SHIFT 0x2 +#define UVD_JMI4_JPEG_LMI_DROP__JRBC_RD_DROP__SHIFT 0x3 +#define UVD_JMI4_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP__SHIFT 0x4 +#define UVD_JMI4_JPEG_LMI_DROP__JPEG_WR_DROP_MASK 0x00000001L +#define UVD_JMI4_JPEG_LMI_DROP__JRBC_WR_DROP_MASK 0x00000002L +#define UVD_JMI4_JPEG_LMI_DROP__JPEG_RD_DROP_MASK 0x00000004L +#define UVD_JMI4_JPEG_LMI_DROP__JRBC_RD_DROP_MASK 0x00000008L +#define UVD_JMI4_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP_MASK 0x00000010L +//UVD_JMI4_UVD_LMI_JRBC_IB_VMID +#define UVD_JMI4_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID__SHIFT 0x4 +#define UVD_JMI4_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID__SHIFT 0x8 +#define UVD_JMI4_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID_MASK 0x0000000FL +#define UVD_JMI4_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID_MASK 0x000000F0L +#define UVD_JMI4_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID_MASK 0x00000F00L +//UVD_JMI4_UVD_LMI_JRBC_RB_VMID +#define UVD_JMI4_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID__SHIFT 0x4 +#define UVD_JMI4_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID__SHIFT 0x8 +#define UVD_JMI4_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID_MASK 0x0000000FL +#define UVD_JMI4_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID_MASK 0x000000F0L +#define UVD_JMI4_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID_MASK 0x00000F00L +//UVD_JMI4_UVD_LMI_JPEG_VMID +#define UVD_JMI4_UVD_LMI_JPEG_VMID__JPEG_RD_VMID__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JPEG_VMID__JPEG_WR_VMID__SHIFT 0x4 +#define UVD_JMI4_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID__SHIFT 0x8 +#define UVD_JMI4_UVD_LMI_JPEG_VMID__JPEG_RD_VMID_MASK 0x0000000FL +#define UVD_JMI4_UVD_LMI_JPEG_VMID__JPEG_WR_VMID_MASK 0x000000F0L +#define UVD_JMI4_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID_MASK 0x00000F00L +//UVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW +#define UVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH +#define UVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_LOW +#define UVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH +#define UVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW +#define UVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH +#define UVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI4_UVD_LMI_JPEG_PREEMPT_VMID +#define UVD_JMI4_UVD_LMI_JPEG_PREEMPT_VMID__VMID__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JPEG_PREEMPT_VMID__VMID_MASK 0x0000000FL +//UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP__SHIFT 0x0 +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP__SHIFT 0x2 +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP__SHIFT 0x4 +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP__SHIFT 0x6 +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP__SHIFT 0x8 +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP__SHIFT 0xa +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP__SHIFT 0xc +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP__SHIFT 0xe +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP__SHIFT 0x10 +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP_MASK 0x00000003L +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP_MASK 0x0000000CL +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP_MASK 0x00000030L +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP_MASK 0x000000C0L +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP_MASK 0x00000300L +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP_MASK 0x00000C00L +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP_MASK 0x00003000L +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP_MASK 0x0000C000L +#define UVD_JMI4_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP_MASK 0x00030000L +//UVD_JMI4_UVD_JMI_ATOMIC_CNTL +#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en__SHIFT 0x0 +#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__atomic_max_burst__SHIFT 0x1 +#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop__SHIFT 0x5 +#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en__SHIFT 0x6 +#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG__SHIFT 0x7 +#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE__SHIFT 0xb +#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en_MASK 0x00000001L +#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__atomic_max_burst_MASK 0x0000001EL +#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop_MASK 0x00000020L +#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en_MASK 0x00000040L +#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG_MASK 0x00000780L +#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE_MASK 0x00000800L +//UVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW +#define UVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH +#define UVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI4_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_LOW +#define UVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH +#define UVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW +#define UVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH +#define UVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_LOW +#define UVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH +#define UVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW +#define UVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH +#define UVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI4_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI4_UVD_JMI_ATOMIC_CNTL2 +#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap__SHIFT 0x10 +#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP__SHIFT 0x18 +#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap_MASK 0x00FF0000L +#define UVD_JMI4_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP_MASK 0xFF000000L + + +// addressBlock: aid_uvd0_uvd_jmi5_uvd_jmi_dec +//UVD_JMI5_UVD_JPEG_DEC_PF_CTRL +#define UVD_JMI5_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS__SHIFT 0x0 +#define UVD_JMI5_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING__SHIFT 0x1 +#define UVD_JMI5_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS_MASK 0x00000001L +#define UVD_JMI5_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING_MASK 0x00000002L +//UVD_JMI5_UVD_LMI_JRBC_CTRL +#define UVD_JMI5_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN__SHIFT 0x1 +#define UVD_JMI5_UVD_LMI_JRBC_CTRL__RD_MAX_BURST__SHIFT 0x4 +#define UVD_JMI5_UVD_LMI_JRBC_CTRL__WR_MAX_BURST__SHIFT 0x8 +#define UVD_JMI5_UVD_LMI_JRBC_CTRL__RD_SWAP__SHIFT 0x14 +#define UVD_JMI5_UVD_LMI_JRBC_CTRL__WR_SWAP__SHIFT 0x16 +#define UVD_JMI5_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN_MASK 0x00000001L +#define UVD_JMI5_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN_MASK 0x00000002L +#define UVD_JMI5_UVD_LMI_JRBC_CTRL__RD_MAX_BURST_MASK 0x000000F0L +#define UVD_JMI5_UVD_LMI_JRBC_CTRL__WR_MAX_BURST_MASK 0x00000F00L +#define UVD_JMI5_UVD_LMI_JRBC_CTRL__RD_SWAP_MASK 0x00300000L +#define UVD_JMI5_UVD_LMI_JRBC_CTRL__WR_SWAP_MASK 0x00C00000L +//UVD_JMI5_UVD_LMI_JPEG_CTRL +#define UVD_JMI5_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN__SHIFT 0x1 +#define UVD_JMI5_UVD_LMI_JPEG_CTRL__RD_MAX_BURST__SHIFT 0x4 +#define UVD_JMI5_UVD_LMI_JPEG_CTRL__WR_MAX_BURST__SHIFT 0x8 +#define UVD_JMI5_UVD_LMI_JPEG_CTRL__RD_SWAP__SHIFT 0x14 +#define UVD_JMI5_UVD_LMI_JPEG_CTRL__WR_SWAP__SHIFT 0x16 +#define UVD_JMI5_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN_MASK 0x00000001L +#define UVD_JMI5_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN_MASK 0x00000002L +#define UVD_JMI5_UVD_LMI_JPEG_CTRL__RD_MAX_BURST_MASK 0x000000F0L +#define UVD_JMI5_UVD_LMI_JPEG_CTRL__WR_MAX_BURST_MASK 0x00000F00L +#define UVD_JMI5_UVD_LMI_JPEG_CTRL__RD_SWAP_MASK 0x00300000L +#define UVD_JMI5_UVD_LMI_JPEG_CTRL__WR_SWAP_MASK 0x00C00000L +//UVD_JMI5_JPEG_LMI_DROP +#define UVD_JMI5_JPEG_LMI_DROP__JPEG_WR_DROP__SHIFT 0x0 +#define UVD_JMI5_JPEG_LMI_DROP__JRBC_WR_DROP__SHIFT 0x1 +#define UVD_JMI5_JPEG_LMI_DROP__JPEG_RD_DROP__SHIFT 0x2 +#define UVD_JMI5_JPEG_LMI_DROP__JRBC_RD_DROP__SHIFT 0x3 +#define UVD_JMI5_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP__SHIFT 0x4 +#define UVD_JMI5_JPEG_LMI_DROP__JPEG_WR_DROP_MASK 0x00000001L +#define UVD_JMI5_JPEG_LMI_DROP__JRBC_WR_DROP_MASK 0x00000002L +#define UVD_JMI5_JPEG_LMI_DROP__JPEG_RD_DROP_MASK 0x00000004L +#define UVD_JMI5_JPEG_LMI_DROP__JRBC_RD_DROP_MASK 0x00000008L +#define UVD_JMI5_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP_MASK 0x00000010L +//UVD_JMI5_UVD_LMI_JRBC_IB_VMID +#define UVD_JMI5_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID__SHIFT 0x4 +#define UVD_JMI5_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID__SHIFT 0x8 +#define UVD_JMI5_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID_MASK 0x0000000FL +#define UVD_JMI5_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID_MASK 0x000000F0L +#define UVD_JMI5_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID_MASK 0x00000F00L +//UVD_JMI5_UVD_LMI_JRBC_RB_VMID +#define UVD_JMI5_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID__SHIFT 0x4 +#define UVD_JMI5_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID__SHIFT 0x8 +#define UVD_JMI5_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID_MASK 0x0000000FL +#define UVD_JMI5_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID_MASK 0x000000F0L +#define UVD_JMI5_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID_MASK 0x00000F00L +//UVD_JMI5_UVD_LMI_JPEG_VMID +#define UVD_JMI5_UVD_LMI_JPEG_VMID__JPEG_RD_VMID__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JPEG_VMID__JPEG_WR_VMID__SHIFT 0x4 +#define UVD_JMI5_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID__SHIFT 0x8 +#define UVD_JMI5_UVD_LMI_JPEG_VMID__JPEG_RD_VMID_MASK 0x0000000FL +#define UVD_JMI5_UVD_LMI_JPEG_VMID__JPEG_WR_VMID_MASK 0x000000F0L +#define UVD_JMI5_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID_MASK 0x00000F00L +//UVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW +#define UVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH +#define UVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_LOW +#define UVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH +#define UVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW +#define UVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH +#define UVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI5_UVD_LMI_JPEG_PREEMPT_VMID +#define UVD_JMI5_UVD_LMI_JPEG_PREEMPT_VMID__VMID__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JPEG_PREEMPT_VMID__VMID_MASK 0x0000000FL +//UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP__SHIFT 0x0 +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP__SHIFT 0x2 +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP__SHIFT 0x4 +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP__SHIFT 0x6 +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP__SHIFT 0x8 +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP__SHIFT 0xa +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP__SHIFT 0xc +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP__SHIFT 0xe +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP__SHIFT 0x10 +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP_MASK 0x00000003L +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP_MASK 0x0000000CL +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP_MASK 0x00000030L +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP_MASK 0x000000C0L +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP_MASK 0x00000300L +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP_MASK 0x00000C00L +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP_MASK 0x00003000L +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP_MASK 0x0000C000L +#define UVD_JMI5_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP_MASK 0x00030000L +//UVD_JMI5_UVD_JMI_ATOMIC_CNTL +#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en__SHIFT 0x0 +#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__atomic_max_burst__SHIFT 0x1 +#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop__SHIFT 0x5 +#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en__SHIFT 0x6 +#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG__SHIFT 0x7 +#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE__SHIFT 0xb +#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en_MASK 0x00000001L +#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__atomic_max_burst_MASK 0x0000001EL +#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop_MASK 0x00000020L +#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en_MASK 0x00000040L +#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG_MASK 0x00000780L +#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE_MASK 0x00000800L +//UVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW +#define UVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH +#define UVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI5_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_LOW +#define UVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH +#define UVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW +#define UVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH +#define UVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_LOW +#define UVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH +#define UVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW +#define UVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH +#define UVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI5_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI5_UVD_JMI_ATOMIC_CNTL2 +#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap__SHIFT 0x10 +#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP__SHIFT 0x18 +#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap_MASK 0x00FF0000L +#define UVD_JMI5_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP_MASK 0xFF000000L + + +// addressBlock: aid_uvd0_uvd_jmi6_uvd_jmi_dec +//UVD_JMI6_UVD_JPEG_DEC_PF_CTRL +#define UVD_JMI6_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS__SHIFT 0x0 +#define UVD_JMI6_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING__SHIFT 0x1 +#define UVD_JMI6_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS_MASK 0x00000001L +#define UVD_JMI6_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING_MASK 0x00000002L +//UVD_JMI6_UVD_LMI_JRBC_CTRL +#define UVD_JMI6_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN__SHIFT 0x1 +#define UVD_JMI6_UVD_LMI_JRBC_CTRL__RD_MAX_BURST__SHIFT 0x4 +#define UVD_JMI6_UVD_LMI_JRBC_CTRL__WR_MAX_BURST__SHIFT 0x8 +#define UVD_JMI6_UVD_LMI_JRBC_CTRL__RD_SWAP__SHIFT 0x14 +#define UVD_JMI6_UVD_LMI_JRBC_CTRL__WR_SWAP__SHIFT 0x16 +#define UVD_JMI6_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN_MASK 0x00000001L +#define UVD_JMI6_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN_MASK 0x00000002L +#define UVD_JMI6_UVD_LMI_JRBC_CTRL__RD_MAX_BURST_MASK 0x000000F0L +#define UVD_JMI6_UVD_LMI_JRBC_CTRL__WR_MAX_BURST_MASK 0x00000F00L +#define UVD_JMI6_UVD_LMI_JRBC_CTRL__RD_SWAP_MASK 0x00300000L +#define UVD_JMI6_UVD_LMI_JRBC_CTRL__WR_SWAP_MASK 0x00C00000L +//UVD_JMI6_UVD_LMI_JPEG_CTRL +#define UVD_JMI6_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN__SHIFT 0x1 +#define UVD_JMI6_UVD_LMI_JPEG_CTRL__RD_MAX_BURST__SHIFT 0x4 +#define UVD_JMI6_UVD_LMI_JPEG_CTRL__WR_MAX_BURST__SHIFT 0x8 +#define UVD_JMI6_UVD_LMI_JPEG_CTRL__RD_SWAP__SHIFT 0x14 +#define UVD_JMI6_UVD_LMI_JPEG_CTRL__WR_SWAP__SHIFT 0x16 +#define UVD_JMI6_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN_MASK 0x00000001L +#define UVD_JMI6_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN_MASK 0x00000002L +#define UVD_JMI6_UVD_LMI_JPEG_CTRL__RD_MAX_BURST_MASK 0x000000F0L +#define UVD_JMI6_UVD_LMI_JPEG_CTRL__WR_MAX_BURST_MASK 0x00000F00L +#define UVD_JMI6_UVD_LMI_JPEG_CTRL__RD_SWAP_MASK 0x00300000L +#define UVD_JMI6_UVD_LMI_JPEG_CTRL__WR_SWAP_MASK 0x00C00000L +//UVD_JMI6_JPEG_LMI_DROP +#define UVD_JMI6_JPEG_LMI_DROP__JPEG_WR_DROP__SHIFT 0x0 +#define UVD_JMI6_JPEG_LMI_DROP__JRBC_WR_DROP__SHIFT 0x1 +#define UVD_JMI6_JPEG_LMI_DROP__JPEG_RD_DROP__SHIFT 0x2 +#define UVD_JMI6_JPEG_LMI_DROP__JRBC_RD_DROP__SHIFT 0x3 +#define UVD_JMI6_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP__SHIFT 0x4 +#define UVD_JMI6_JPEG_LMI_DROP__JPEG_WR_DROP_MASK 0x00000001L +#define UVD_JMI6_JPEG_LMI_DROP__JRBC_WR_DROP_MASK 0x00000002L +#define UVD_JMI6_JPEG_LMI_DROP__JPEG_RD_DROP_MASK 0x00000004L +#define UVD_JMI6_JPEG_LMI_DROP__JRBC_RD_DROP_MASK 0x00000008L +#define UVD_JMI6_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP_MASK 0x00000010L +//UVD_JMI6_UVD_LMI_JRBC_IB_VMID +#define UVD_JMI6_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID__SHIFT 0x4 +#define UVD_JMI6_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID__SHIFT 0x8 +#define UVD_JMI6_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID_MASK 0x0000000FL +#define UVD_JMI6_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID_MASK 0x000000F0L +#define UVD_JMI6_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID_MASK 0x00000F00L +//UVD_JMI6_UVD_LMI_JRBC_RB_VMID +#define UVD_JMI6_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID__SHIFT 0x4 +#define UVD_JMI6_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID__SHIFT 0x8 +#define UVD_JMI6_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID_MASK 0x0000000FL +#define UVD_JMI6_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID_MASK 0x000000F0L +#define UVD_JMI6_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID_MASK 0x00000F00L +//UVD_JMI6_UVD_LMI_JPEG_VMID +#define UVD_JMI6_UVD_LMI_JPEG_VMID__JPEG_RD_VMID__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JPEG_VMID__JPEG_WR_VMID__SHIFT 0x4 +#define UVD_JMI6_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID__SHIFT 0x8 +#define UVD_JMI6_UVD_LMI_JPEG_VMID__JPEG_RD_VMID_MASK 0x0000000FL +#define UVD_JMI6_UVD_LMI_JPEG_VMID__JPEG_WR_VMID_MASK 0x000000F0L +#define UVD_JMI6_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID_MASK 0x00000F00L +//UVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW +#define UVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH +#define UVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_LOW +#define UVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH +#define UVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW +#define UVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH +#define UVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI6_UVD_LMI_JPEG_PREEMPT_VMID +#define UVD_JMI6_UVD_LMI_JPEG_PREEMPT_VMID__VMID__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JPEG_PREEMPT_VMID__VMID_MASK 0x0000000FL +//UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP__SHIFT 0x0 +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP__SHIFT 0x2 +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP__SHIFT 0x4 +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP__SHIFT 0x6 +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP__SHIFT 0x8 +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP__SHIFT 0xa +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP__SHIFT 0xc +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP__SHIFT 0xe +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP__SHIFT 0x10 +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP_MASK 0x00000003L +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP_MASK 0x0000000CL +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP_MASK 0x00000030L +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP_MASK 0x000000C0L +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP_MASK 0x00000300L +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP_MASK 0x00000C00L +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP_MASK 0x00003000L +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP_MASK 0x0000C000L +#define UVD_JMI6_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP_MASK 0x00030000L +//UVD_JMI6_UVD_JMI_ATOMIC_CNTL +#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en__SHIFT 0x0 +#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__atomic_max_burst__SHIFT 0x1 +#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop__SHIFT 0x5 +#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en__SHIFT 0x6 +#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG__SHIFT 0x7 +#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE__SHIFT 0xb +#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en_MASK 0x00000001L +#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__atomic_max_burst_MASK 0x0000001EL +#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop_MASK 0x00000020L +#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en_MASK 0x00000040L +#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG_MASK 0x00000780L +#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE_MASK 0x00000800L +//UVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW +#define UVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH +#define UVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI6_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_LOW +#define UVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH +#define UVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW +#define UVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH +#define UVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_LOW +#define UVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH +#define UVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW +#define UVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH +#define UVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI6_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI6_UVD_JMI_ATOMIC_CNTL2 +#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap__SHIFT 0x10 +#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP__SHIFT 0x18 +#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap_MASK 0x00FF0000L +#define UVD_JMI6_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP_MASK 0xFF000000L + + +// addressBlock: aid_uvd0_uvd_jmi7_uvd_jmi_dec +//UVD_JMI7_UVD_JPEG_DEC_PF_CTRL +#define UVD_JMI7_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS__SHIFT 0x0 +#define UVD_JMI7_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING__SHIFT 0x1 +#define UVD_JMI7_UVD_JPEG_DEC_PF_CTRL__DEC_PF_HANDLING_DIS_MASK 0x00000001L +#define UVD_JMI7_UVD_JPEG_DEC_PF_CTRL__DEC_PF_SW_GATING_MASK 0x00000002L +//UVD_JMI7_UVD_LMI_JRBC_CTRL +#define UVD_JMI7_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN__SHIFT 0x1 +#define UVD_JMI7_UVD_LMI_JRBC_CTRL__RD_MAX_BURST__SHIFT 0x4 +#define UVD_JMI7_UVD_LMI_JRBC_CTRL__WR_MAX_BURST__SHIFT 0x8 +#define UVD_JMI7_UVD_LMI_JRBC_CTRL__RD_SWAP__SHIFT 0x14 +#define UVD_JMI7_UVD_LMI_JRBC_CTRL__WR_SWAP__SHIFT 0x16 +#define UVD_JMI7_UVD_LMI_JRBC_CTRL__ARB_RD_WAIT_EN_MASK 0x00000001L +#define UVD_JMI7_UVD_LMI_JRBC_CTRL__ARB_WR_WAIT_EN_MASK 0x00000002L +#define UVD_JMI7_UVD_LMI_JRBC_CTRL__RD_MAX_BURST_MASK 0x000000F0L +#define UVD_JMI7_UVD_LMI_JRBC_CTRL__WR_MAX_BURST_MASK 0x00000F00L +#define UVD_JMI7_UVD_LMI_JRBC_CTRL__RD_SWAP_MASK 0x00300000L +#define UVD_JMI7_UVD_LMI_JRBC_CTRL__WR_SWAP_MASK 0x00C00000L +//UVD_JMI7_UVD_LMI_JPEG_CTRL +#define UVD_JMI7_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN__SHIFT 0x1 +#define UVD_JMI7_UVD_LMI_JPEG_CTRL__RD_MAX_BURST__SHIFT 0x4 +#define UVD_JMI7_UVD_LMI_JPEG_CTRL__WR_MAX_BURST__SHIFT 0x8 +#define UVD_JMI7_UVD_LMI_JPEG_CTRL__RD_SWAP__SHIFT 0x14 +#define UVD_JMI7_UVD_LMI_JPEG_CTRL__WR_SWAP__SHIFT 0x16 +#define UVD_JMI7_UVD_LMI_JPEG_CTRL__ARB_RD_WAIT_EN_MASK 0x00000001L +#define UVD_JMI7_UVD_LMI_JPEG_CTRL__ARB_WR_WAIT_EN_MASK 0x00000002L +#define UVD_JMI7_UVD_LMI_JPEG_CTRL__RD_MAX_BURST_MASK 0x000000F0L +#define UVD_JMI7_UVD_LMI_JPEG_CTRL__WR_MAX_BURST_MASK 0x00000F00L +#define UVD_JMI7_UVD_LMI_JPEG_CTRL__RD_SWAP_MASK 0x00300000L +#define UVD_JMI7_UVD_LMI_JPEG_CTRL__WR_SWAP_MASK 0x00C00000L +//UVD_JMI7_JPEG_LMI_DROP +#define UVD_JMI7_JPEG_LMI_DROP__JPEG_WR_DROP__SHIFT 0x0 +#define UVD_JMI7_JPEG_LMI_DROP__JRBC_WR_DROP__SHIFT 0x1 +#define UVD_JMI7_JPEG_LMI_DROP__JPEG_RD_DROP__SHIFT 0x2 +#define UVD_JMI7_JPEG_LMI_DROP__JRBC_RD_DROP__SHIFT 0x3 +#define UVD_JMI7_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP__SHIFT 0x4 +#define UVD_JMI7_JPEG_LMI_DROP__JPEG_WR_DROP_MASK 0x00000001L +#define UVD_JMI7_JPEG_LMI_DROP__JRBC_WR_DROP_MASK 0x00000002L +#define UVD_JMI7_JPEG_LMI_DROP__JPEG_RD_DROP_MASK 0x00000004L +#define UVD_JMI7_JPEG_LMI_DROP__JRBC_RD_DROP_MASK 0x00000008L +#define UVD_JMI7_JPEG_LMI_DROP__JPEG_ATOMIC_WR_DROP_MASK 0x00000010L +//UVD_JMI7_UVD_LMI_JRBC_IB_VMID +#define UVD_JMI7_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID__SHIFT 0x4 +#define UVD_JMI7_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID__SHIFT 0x8 +#define UVD_JMI7_UVD_LMI_JRBC_IB_VMID__IB_WR_VMID_MASK 0x0000000FL +#define UVD_JMI7_UVD_LMI_JRBC_IB_VMID__IB_RD_VMID_MASK 0x000000F0L +#define UVD_JMI7_UVD_LMI_JRBC_IB_VMID__MEM_RD_VMID_MASK 0x00000F00L +//UVD_JMI7_UVD_LMI_JRBC_RB_VMID +#define UVD_JMI7_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID__SHIFT 0x4 +#define UVD_JMI7_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID__SHIFT 0x8 +#define UVD_JMI7_UVD_LMI_JRBC_RB_VMID__RB_WR_VMID_MASK 0x0000000FL +#define UVD_JMI7_UVD_LMI_JRBC_RB_VMID__RB_RD_VMID_MASK 0x000000F0L +#define UVD_JMI7_UVD_LMI_JRBC_RB_VMID__MEM_RD_VMID_MASK 0x00000F00L +//UVD_JMI7_UVD_LMI_JPEG_VMID +#define UVD_JMI7_UVD_LMI_JPEG_VMID__JPEG_RD_VMID__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JPEG_VMID__JPEG_WR_VMID__SHIFT 0x4 +#define UVD_JMI7_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID__SHIFT 0x8 +#define UVD_JMI7_UVD_LMI_JPEG_VMID__JPEG_RD_VMID_MASK 0x0000000FL +#define UVD_JMI7_UVD_LMI_JPEG_VMID__JPEG_WR_VMID_MASK 0x000000F0L +#define UVD_JMI7_UVD_LMI_JPEG_VMID__ATOMIC_USER0_WR_VMID_MASK 0x00000F00L +//UVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW +#define UVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH +#define UVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JPEG_PREEMPT_FENCE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_LOW +#define UVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH +#define UVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW +#define UVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH +#define UVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JRBC_RB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI7_UVD_LMI_JPEG_PREEMPT_VMID +#define UVD_JMI7_UVD_LMI_JPEG_PREEMPT_VMID__VMID__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JPEG_PREEMPT_VMID__VMID_MASK 0x0000000FL +//UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP__SHIFT 0x0 +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP__SHIFT 0x2 +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP__SHIFT 0x4 +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP__SHIFT 0x6 +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP__SHIFT 0x8 +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP__SHIFT 0xa +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP__SHIFT 0xc +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP__SHIFT 0xe +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP__SHIFT 0x10 +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__RB_MC_SWAP_MASK 0x00000003L +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__IB_MC_SWAP_MASK 0x0000000CL +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_WR_MC_SWAP_MASK 0x00000030L +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_WR_MC_SWAP_MASK 0x000000C0L +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__RB_MEM_RD_MC_SWAP_MASK 0x00000300L +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__IB_MEM_RD_MC_SWAP_MASK 0x00000C00L +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__PREEMPT_WR_MC_SWAP_MASK 0x00003000L +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__JPEG_RD_MC_SWAP_MASK 0x0000C000L +#define UVD_JMI7_UVD_JMI_DEC_SWAP_CNTL__JPEG_WR_MC_SWAP_MASK 0x00030000L +//UVD_JMI7_UVD_JMI_ATOMIC_CNTL +#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en__SHIFT 0x0 +#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__atomic_max_burst__SHIFT 0x1 +#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop__SHIFT 0x5 +#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en__SHIFT 0x6 +#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG__SHIFT 0x7 +#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE__SHIFT 0xb +#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__atomic_arb_wait_en_MASK 0x00000001L +#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__atomic_max_burst_MASK 0x0000001EL +#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__atomic_wr_drop_MASK 0x00000020L +#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__atomic_wr_clamping_en_MASK 0x00000040L +#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__ATOMIC_WR_URG_MASK 0x00000780L +#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL__ATOMIC_SW_GATE_MASK 0x00000800L +//UVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW +#define UVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH +#define UVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI7_UVD_JMI_ATOMIC_USER0_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_LOW +#define UVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH +#define UVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JPEG_READ_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW +#define UVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH +#define UVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JPEG_WRITE_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_LOW +#define UVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH +#define UVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JRBC_IB_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW +#define UVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_LOW__BITS_31_0_MASK 0xFFFFFFFFL +//UVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH +#define UVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32__SHIFT 0x0 +#define UVD_JMI7_UVD_LMI_JRBC_IB_MEM_WR_64BIT_BAR_HIGH__BITS_63_32_MASK 0xFFFFFFFFL +//UVD_JMI7_UVD_JMI_ATOMIC_CNTL2 +#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap__SHIFT 0x10 +#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP__SHIFT 0x18 +#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL2__atomic_uvd_swap_MASK 0x00FF0000L +#define UVD_JMI7_UVD_JMI_ATOMIC_CNTL2__ATOMIC_MC_SWAP_MASK 0xFF000000L + + +// addressBlock: uvdctxind +//UVD_CGC_MEM_CTRL +#define UVD_CGC_MEM_CTRL__LMI_MC_LS_EN__SHIFT 0x0 +#define UVD_CGC_MEM_CTRL__MPC_LS_EN__SHIFT 0x1 +#define UVD_CGC_MEM_CTRL__MPRD_LS_EN__SHIFT 0x2 +#define UVD_CGC_MEM_CTRL__WCB_LS_EN__SHIFT 0x3 +#define UVD_CGC_MEM_CTRL__UDEC_RE_LS_EN__SHIFT 0x4 +#define UVD_CGC_MEM_CTRL__UDEC_CM_LS_EN__SHIFT 0x5 +#define UVD_CGC_MEM_CTRL__UDEC_IT_LS_EN__SHIFT 0x6 +#define UVD_CGC_MEM_CTRL__UDEC_DB_LS_EN__SHIFT 0x7 +#define UVD_CGC_MEM_CTRL__UDEC_MP_LS_EN__SHIFT 0x8 +#define UVD_CGC_MEM_CTRL__SYS_LS_EN__SHIFT 0x9 +#define UVD_CGC_MEM_CTRL__VCPU_LS_EN__SHIFT 0xa +#define UVD_CGC_MEM_CTRL__MIF_LS_EN__SHIFT 0xc +#define UVD_CGC_MEM_CTRL__LCM_LS_EN__SHIFT 0xd +#define UVD_CGC_MEM_CTRL__MMSCH_LS_EN__SHIFT 0xe +#define UVD_CGC_MEM_CTRL__MPC1_LS_EN__SHIFT 0xf +#define UVD_CGC_MEM_CTRL__LS_SET_DELAY__SHIFT 0x10 +#define UVD_CGC_MEM_CTRL__LS_CLEAR_DELAY__SHIFT 0x14 +#define UVD_CGC_MEM_CTRL__LMI_MC_LS_EN_MASK 0x00000001L +#define UVD_CGC_MEM_CTRL__MPC_LS_EN_MASK 0x00000002L +#define UVD_CGC_MEM_CTRL__MPRD_LS_EN_MASK 0x00000004L +#define UVD_CGC_MEM_CTRL__WCB_LS_EN_MASK 0x00000008L +#define UVD_CGC_MEM_CTRL__UDEC_RE_LS_EN_MASK 0x00000010L +#define UVD_CGC_MEM_CTRL__UDEC_CM_LS_EN_MASK 0x00000020L +#define UVD_CGC_MEM_CTRL__UDEC_IT_LS_EN_MASK 0x00000040L +#define UVD_CGC_MEM_CTRL__UDEC_DB_LS_EN_MASK 0x00000080L +#define UVD_CGC_MEM_CTRL__UDEC_MP_LS_EN_MASK 0x00000100L +#define UVD_CGC_MEM_CTRL__SYS_LS_EN_MASK 0x00000200L +#define UVD_CGC_MEM_CTRL__VCPU_LS_EN_MASK 0x00000400L +#define UVD_CGC_MEM_CTRL__MIF_LS_EN_MASK 0x00001000L +#define UVD_CGC_MEM_CTRL__LCM_LS_EN_MASK 0x00002000L +#define UVD_CGC_MEM_CTRL__MMSCH_LS_EN_MASK 0x00004000L +#define UVD_CGC_MEM_CTRL__MPC1_LS_EN_MASK 0x00008000L +#define UVD_CGC_MEM_CTRL__LS_SET_DELAY_MASK 0x000F0000L +#define UVD_CGC_MEM_CTRL__LS_CLEAR_DELAY_MASK 0x00F00000L +//UVD_CGC_CTRL2 +#define UVD_CGC_CTRL2__DYN_OCLK_RAMP_EN__SHIFT 0x0 +#define UVD_CGC_CTRL2__DYN_RCLK_RAMP_EN__SHIFT 0x1 +#define UVD_CGC_CTRL2__GATER_DIV_ID__SHIFT 0x2 +#define UVD_CGC_CTRL2__DYN_OCLK_RAMP_EN_MASK 0x00000001L +#define UVD_CGC_CTRL2__DYN_RCLK_RAMP_EN_MASK 0x00000002L +#define UVD_CGC_CTRL2__GATER_DIV_ID_MASK 0x0000001CL +//UVD_CGC_MEM_DS_CTRL +#define UVD_CGC_MEM_DS_CTRL__LMI_MC_DS_EN__SHIFT 0x0 +#define UVD_CGC_MEM_DS_CTRL__MPC_DS_EN__SHIFT 0x1 +#define UVD_CGC_MEM_DS_CTRL__MPRD_DS_EN__SHIFT 0x2 +#define UVD_CGC_MEM_DS_CTRL__WCB_DS_EN__SHIFT 0x3 +#define UVD_CGC_MEM_DS_CTRL__UDEC_RE_DS_EN__SHIFT 0x4 +#define UVD_CGC_MEM_DS_CTRL__UDEC_CM_DS_EN__SHIFT 0x5 +#define UVD_CGC_MEM_DS_CTRL__UDEC_IT_DS_EN__SHIFT 0x6 +#define UVD_CGC_MEM_DS_CTRL__UDEC_DB_DS_EN__SHIFT 0x7 +#define UVD_CGC_MEM_DS_CTRL__UDEC_MP_DS_EN__SHIFT 0x8 +#define UVD_CGC_MEM_DS_CTRL__SYS_DS_EN__SHIFT 0x9 +#define UVD_CGC_MEM_DS_CTRL__VCPU_DS_EN__SHIFT 0xa +#define UVD_CGC_MEM_DS_CTRL__MIF_DS_EN__SHIFT 0xc +#define UVD_CGC_MEM_DS_CTRL__LCM_DS_EN__SHIFT 0xd +#define UVD_CGC_MEM_DS_CTRL__MMSCH_DS_EN__SHIFT 0xe +#define UVD_CGC_MEM_DS_CTRL__MPC1_DS_EN__SHIFT 0xf +#define UVD_CGC_MEM_DS_CTRL__LMI_MC_DS_EN_MASK 0x00000001L +#define UVD_CGC_MEM_DS_CTRL__MPC_DS_EN_MASK 0x00000002L +#define UVD_CGC_MEM_DS_CTRL__MPRD_DS_EN_MASK 0x00000004L +#define UVD_CGC_MEM_DS_CTRL__WCB_DS_EN_MASK 0x00000008L +#define UVD_CGC_MEM_DS_CTRL__UDEC_RE_DS_EN_MASK 0x00000010L +#define UVD_CGC_MEM_DS_CTRL__UDEC_CM_DS_EN_MASK 0x00000020L +#define UVD_CGC_MEM_DS_CTRL__UDEC_IT_DS_EN_MASK 0x00000040L +#define UVD_CGC_MEM_DS_CTRL__UDEC_DB_DS_EN_MASK 0x00000080L +#define UVD_CGC_MEM_DS_CTRL__UDEC_MP_DS_EN_MASK 0x00000100L +#define UVD_CGC_MEM_DS_CTRL__SYS_DS_EN_MASK 0x00000200L +#define UVD_CGC_MEM_DS_CTRL__VCPU_DS_EN_MASK 0x00000400L +#define UVD_CGC_MEM_DS_CTRL__MIF_DS_EN_MASK 0x00001000L +#define UVD_CGC_MEM_DS_CTRL__LCM_DS_EN_MASK 0x00002000L +#define UVD_CGC_MEM_DS_CTRL__MMSCH_DS_EN_MASK 0x00004000L +#define UVD_CGC_MEM_DS_CTRL__MPC1_DS_EN_MASK 0x00008000L +//UVD_CGC_MEM_SD_CTRL +#define UVD_CGC_MEM_SD_CTRL__LMI_MC_SD_EN__SHIFT 0x0 +#define UVD_CGC_MEM_SD_CTRL__MPC_SD_EN__SHIFT 0x1 +#define UVD_CGC_MEM_SD_CTRL__MPRD_SD_EN__SHIFT 0x2 +#define UVD_CGC_MEM_SD_CTRL__WCB_SD_EN__SHIFT 0x3 +#define UVD_CGC_MEM_SD_CTRL__UDEC_RE_SD_EN__SHIFT 0x4 +#define UVD_CGC_MEM_SD_CTRL__UDEC_CM_SD_EN__SHIFT 0x5 +#define UVD_CGC_MEM_SD_CTRL__UDEC_IT_SD_EN__SHIFT 0x6 +#define UVD_CGC_MEM_SD_CTRL__UDEC_DB_SD_EN__SHIFT 0x7 +#define UVD_CGC_MEM_SD_CTRL__UDEC_MP_SD_EN__SHIFT 0x8 +#define UVD_CGC_MEM_SD_CTRL__SYS_SD_EN__SHIFT 0x9 +#define UVD_CGC_MEM_SD_CTRL__VCPU_SD_EN__SHIFT 0xa +#define UVD_CGC_MEM_SD_CTRL__MIF_SD_EN__SHIFT 0xc +#define UVD_CGC_MEM_SD_CTRL__LCM_SD_EN__SHIFT 0xd +#define UVD_CGC_MEM_SD_CTRL__MMSCH_SD_EN__SHIFT 0xe +#define UVD_CGC_MEM_SD_CTRL__MPC1_SD_EN__SHIFT 0xf +#define UVD_CGC_MEM_SD_CTRL__LMI_MC_SD_EN_MASK 0x00000001L +#define UVD_CGC_MEM_SD_CTRL__MPC_SD_EN_MASK 0x00000002L +#define UVD_CGC_MEM_SD_CTRL__MPRD_SD_EN_MASK 0x00000004L +#define UVD_CGC_MEM_SD_CTRL__WCB_SD_EN_MASK 0x00000008L +#define UVD_CGC_MEM_SD_CTRL__UDEC_RE_SD_EN_MASK 0x00000010L +#define UVD_CGC_MEM_SD_CTRL__UDEC_CM_SD_EN_MASK 0x00000020L +#define UVD_CGC_MEM_SD_CTRL__UDEC_IT_SD_EN_MASK 0x00000040L +#define UVD_CGC_MEM_SD_CTRL__UDEC_DB_SD_EN_MASK 0x00000080L +#define UVD_CGC_MEM_SD_CTRL__UDEC_MP_SD_EN_MASK 0x00000100L +#define UVD_CGC_MEM_SD_CTRL__SYS_SD_EN_MASK 0x00000200L +#define UVD_CGC_MEM_SD_CTRL__VCPU_SD_EN_MASK 0x00000400L +#define UVD_CGC_MEM_SD_CTRL__MIF_SD_EN_MASK 0x00001000L +#define UVD_CGC_MEM_SD_CTRL__LCM_SD_EN_MASK 0x00002000L +#define UVD_CGC_MEM_SD_CTRL__MMSCH_SD_EN_MASK 0x00004000L +#define UVD_CGC_MEM_SD_CTRL__MPC1_SD_EN_MASK 0x00008000L +//UVD_SW_SCRATCH_00 +#define UVD_SW_SCRATCH_00__DATA__SHIFT 0x0 +#define UVD_SW_SCRATCH_00__DATA_MASK 0xFFFFFFFFL +//UVD_SW_SCRATCH_01 +#define UVD_SW_SCRATCH_01__DATA__SHIFT 0x0 +#define UVD_SW_SCRATCH_01__DATA_MASK 0xFFFFFFFFL +//UVD_SW_SCRATCH_02 +#define UVD_SW_SCRATCH_02__DATA__SHIFT 0x0 +#define UVD_SW_SCRATCH_02__DATA_MASK 0xFFFFFFFFL +//UVD_SW_SCRATCH_03 +#define UVD_SW_SCRATCH_03__DATA__SHIFT 0x0 +#define UVD_SW_SCRATCH_03__DATA_MASK 0xFFFFFFFFL +//UVD_SW_SCRATCH_04 +#define UVD_SW_SCRATCH_04__DATA__SHIFT 0x0 +#define UVD_SW_SCRATCH_04__DATA_MASK 0xFFFFFFFFL +//UVD_SW_SCRATCH_05 +#define UVD_SW_SCRATCH_05__DATA__SHIFT 0x0 +#define UVD_SW_SCRATCH_05__DATA_MASK 0xFFFFFFFFL +//UVD_SW_SCRATCH_06 +#define UVD_SW_SCRATCH_06__DATA__SHIFT 0x0 +#define UVD_SW_SCRATCH_06__DATA_MASK 0xFFFFFFFFL +//UVD_SW_SCRATCH_07 +#define UVD_SW_SCRATCH_07__DATA__SHIFT 0x0 +#define UVD_SW_SCRATCH_07__DATA_MASK 0xFFFFFFFFL +//UVD_SW_SCRATCH_08 +#define UVD_SW_SCRATCH_08__DATA__SHIFT 0x0 +#define UVD_SW_SCRATCH_08__DATA_MASK 0xFFFFFFFFL +//UVD_SW_SCRATCH_09 +#define UVD_SW_SCRATCH_09__DATA__SHIFT 0x0 +#define UVD_SW_SCRATCH_09__DATA_MASK 0xFFFFFFFFL +//UVD_SW_SCRATCH_10 +#define UVD_SW_SCRATCH_10__DATA__SHIFT 0x0 +#define UVD_SW_SCRATCH_10__DATA_MASK 0xFFFFFFFFL +//UVD_SW_SCRATCH_11 +#define UVD_SW_SCRATCH_11__DATA__SHIFT 0x0 +#define UVD_SW_SCRATCH_11__DATA_MASK 0xFFFFFFFFL +//UVD_SW_SCRATCH_12 +#define UVD_SW_SCRATCH_12__DATA__SHIFT 0x0 +#define UVD_SW_SCRATCH_12__DATA_MASK 0xFFFFFFFFL +//UVD_SW_SCRATCH_13 +#define UVD_SW_SCRATCH_13__DATA__SHIFT 0x0 +#define UVD_SW_SCRATCH_13__DATA_MASK 0xFFFFFFFFL +//UVD_SW_SCRATCH_14 +#define UVD_SW_SCRATCH_14__DATA__SHIFT 0x0 +#define UVD_SW_SCRATCH_14__DATA_MASK 0xFFFFFFFFL +//UVD_SW_SCRATCH_15 +#define UVD_SW_SCRATCH_15__DATA__SHIFT 0x0 +#define UVD_SW_SCRATCH_15__DATA_MASK 0xFFFFFFFFL +//UVD_IH_SEM_CTRL +#define UVD_IH_SEM_CTRL__IH_STALL_EN__SHIFT 0x0 +#define UVD_IH_SEM_CTRL__SEM_STALL_EN__SHIFT 0x1 +#define UVD_IH_SEM_CTRL__IH_STATUS_CLEAN__SHIFT 0x2 +#define UVD_IH_SEM_CTRL__SEM_STATUS_CLEAN__SHIFT 0x3 +#define UVD_IH_SEM_CTRL__IH_VMID__SHIFT 0x4 +#define UVD_IH_SEM_CTRL__IH_USER_DATA__SHIFT 0x8 +#define UVD_IH_SEM_CTRL__IH_RINGID__SHIFT 0x14 +#define UVD_IH_SEM_CTRL__IH_STALL_EN_MASK 0x00000001L +#define UVD_IH_SEM_CTRL__SEM_STALL_EN_MASK 0x00000002L +#define UVD_IH_SEM_CTRL__IH_STATUS_CLEAN_MASK 0x00000004L +#define UVD_IH_SEM_CTRL__SEM_STATUS_CLEAN_MASK 0x00000008L +#define UVD_IH_SEM_CTRL__IH_VMID_MASK 0x000000F0L +#define UVD_IH_SEM_CTRL__IH_USER_DATA_MASK 0x000FFF00L +#define UVD_IH_SEM_CTRL__IH_RINGID_MASK 0x0FF00000L + + +// addressBlock: lmi_adp_indirect +//UVD_LMI_CRC0 +#define UVD_LMI_CRC0__CRC32__SHIFT 0x0 +#define UVD_LMI_CRC0__CRC32_MASK 0xFFFFFFFFL +//UVD_LMI_CRC1 +#define UVD_LMI_CRC1__CRC32__SHIFT 0x0 +#define UVD_LMI_CRC1__CRC32_MASK 0xFFFFFFFFL +//UVD_LMI_CRC2 +#define UVD_LMI_CRC2__CRC32__SHIFT 0x0 +#define UVD_LMI_CRC2__CRC32_MASK 0xFFFFFFFFL +//UVD_LMI_CRC3 +#define UVD_LMI_CRC3__CRC32__SHIFT 0x0 +#define UVD_LMI_CRC3__CRC32_MASK 0xFFFFFFFFL +//UVD_LMI_CRC10 +#define UVD_LMI_CRC10__CRC32__SHIFT 0x0 +#define UVD_LMI_CRC10__CRC32_MASK 0xFFFFFFFFL +//UVD_LMI_CRC11 +#define UVD_LMI_CRC11__CRC32__SHIFT 0x0 +#define UVD_LMI_CRC11__CRC32_MASK 0xFFFFFFFFL +//UVD_LMI_CRC12 +#define UVD_LMI_CRC12__CRC32__SHIFT 0x0 +#define UVD_LMI_CRC12__CRC32_MASK 0xFFFFFFFFL +//UVD_LMI_CRC13 +#define UVD_LMI_CRC13__CRC32__SHIFT 0x0 +#define UVD_LMI_CRC13__CRC32_MASK 0xFFFFFFFFL +//UVD_LMI_CRC14 +#define UVD_LMI_CRC14__CRC32__SHIFT 0x0 +#define UVD_LMI_CRC14__CRC32_MASK 0xFFFFFFFFL +//UVD_LMI_CRC15 +#define UVD_LMI_CRC15__CRC32__SHIFT 0x0 +#define UVD_LMI_CRC15__CRC32_MASK 0xFFFFFFFFL +//UVD_LMI_SWAP_CNTL2 +#define UVD_LMI_SWAP_CNTL2__SCPU_R_MC_SWAP__SHIFT 0x0 +#define UVD_LMI_SWAP_CNTL2__SCPU_W_MC_SWAP__SHIFT 0x2 +#define UVD_LMI_SWAP_CNTL2__ATOMIC_MC_SWAP__SHIFT 0x4 +#define UVD_LMI_SWAP_CNTL2__CENC_MC_SWAP__SHIFT 0xc +#define UVD_LMI_SWAP_CNTL2__FBC_KEY_MC_SWAP__SHIFT 0xe +#define UVD_LMI_SWAP_CNTL2__SCPU_R_MC_SWAP_MASK 0x00000003L +#define UVD_LMI_SWAP_CNTL2__SCPU_W_MC_SWAP_MASK 0x0000000CL +#define UVD_LMI_SWAP_CNTL2__ATOMIC_MC_SWAP_MASK 0x00000FF0L +#define UVD_LMI_SWAP_CNTL2__CENC_MC_SWAP_MASK 0x00003000L +#define UVD_LMI_SWAP_CNTL2__FBC_KEY_MC_SWAP_MASK 0x0000C000L +//UVD_MEMCHECK_SYS_INT_EN +#define UVD_MEMCHECK_SYS_INT_EN__RE_ERR_EN__SHIFT 0x0 +#define UVD_MEMCHECK_SYS_INT_EN__IT_ERR_EN__SHIFT 0x1 +#define UVD_MEMCHECK_SYS_INT_EN__MP_ERR_EN__SHIFT 0x2 +#define UVD_MEMCHECK_SYS_INT_EN__DB_ERR_EN__SHIFT 0x3 +#define UVD_MEMCHECK_SYS_INT_EN__DBW_ERR_EN__SHIFT 0x4 +#define UVD_MEMCHECK_SYS_INT_EN__CM_ERR_EN__SHIFT 0x5 +#define UVD_MEMCHECK_SYS_INT_EN__MIF_REF_ERR_EN__SHIFT 0x6 +#define UVD_MEMCHECK_SYS_INT_EN__VCPU_ERR_EN__SHIFT 0x7 +#define UVD_MEMCHECK_SYS_INT_EN__MIF_DBW_ERR_EN__SHIFT 0x8 +#define UVD_MEMCHECK_SYS_INT_EN__MIF_CM_COLOC_ERR_EN__SHIFT 0x9 +#define UVD_MEMCHECK_SYS_INT_EN__MIF_BSP0_ERR_EN__SHIFT 0xa +#define UVD_MEMCHECK_SYS_INT_EN__MIF_BSP1_ERR_EN__SHIFT 0xb +#define UVD_MEMCHECK_SYS_INT_EN__SRE_ERR_EN__SHIFT 0xc +#define UVD_MEMCHECK_SYS_INT_EN__IT_RD_ERR_EN__SHIFT 0xf +#define UVD_MEMCHECK_SYS_INT_EN__CM_RD_ERR_EN__SHIFT 0x10 +#define UVD_MEMCHECK_SYS_INT_EN__DB_RD_ERR_EN__SHIFT 0x11 +#define UVD_MEMCHECK_SYS_INT_EN__MIF_RD_ERR_EN__SHIFT 0x12 +#define UVD_MEMCHECK_SYS_INT_EN__IDCT_RD_ERR_EN__SHIFT 0x13 +#define UVD_MEMCHECK_SYS_INT_EN__MPC_RD_ERR_EN__SHIFT 0x14 +#define UVD_MEMCHECK_SYS_INT_EN__LBSI_RD_ERR_EN__SHIFT 0x15 +#define UVD_MEMCHECK_SYS_INT_EN__RBC_RD_ERR_EN__SHIFT 0x18 +#define UVD_MEMCHECK_SYS_INT_EN__MIF_BSP2_ERR_EN__SHIFT 0x1b +#define UVD_MEMCHECK_SYS_INT_EN__MIF_BSP3_ERR_EN__SHIFT 0x1c +#define UVD_MEMCHECK_SYS_INT_EN__MIF_SCLR_ERR_EN__SHIFT 0x1d +#define UVD_MEMCHECK_SYS_INT_EN__MIF_SCLR2_ERR_EN__SHIFT 0x1e +#define UVD_MEMCHECK_SYS_INT_EN__PREF_ERR_EN__SHIFT 0x1f +#define UVD_MEMCHECK_SYS_INT_EN__RE_ERR_EN_MASK 0x00000001L +#define UVD_MEMCHECK_SYS_INT_EN__IT_ERR_EN_MASK 0x00000002L +#define UVD_MEMCHECK_SYS_INT_EN__MP_ERR_EN_MASK 0x00000004L +#define UVD_MEMCHECK_SYS_INT_EN__DB_ERR_EN_MASK 0x00000008L +#define UVD_MEMCHECK_SYS_INT_EN__DBW_ERR_EN_MASK 0x00000010L +#define UVD_MEMCHECK_SYS_INT_EN__CM_ERR_EN_MASK 0x00000020L +#define UVD_MEMCHECK_SYS_INT_EN__MIF_REF_ERR_EN_MASK 0x00000040L +#define UVD_MEMCHECK_SYS_INT_EN__VCPU_ERR_EN_MASK 0x00000080L +#define UVD_MEMCHECK_SYS_INT_EN__MIF_DBW_ERR_EN_MASK 0x00000100L +#define UVD_MEMCHECK_SYS_INT_EN__MIF_CM_COLOC_ERR_EN_MASK 0x00000200L +#define UVD_MEMCHECK_SYS_INT_EN__MIF_BSP0_ERR_EN_MASK 0x00000400L +#define UVD_MEMCHECK_SYS_INT_EN__MIF_BSP1_ERR_EN_MASK 0x00000800L +#define UVD_MEMCHECK_SYS_INT_EN__SRE_ERR_EN_MASK 0x00001000L +#define UVD_MEMCHECK_SYS_INT_EN__IT_RD_ERR_EN_MASK 0x00008000L +#define UVD_MEMCHECK_SYS_INT_EN__CM_RD_ERR_EN_MASK 0x00010000L +#define UVD_MEMCHECK_SYS_INT_EN__DB_RD_ERR_EN_MASK 0x00020000L +#define UVD_MEMCHECK_SYS_INT_EN__MIF_RD_ERR_EN_MASK 0x00040000L +#define UVD_MEMCHECK_SYS_INT_EN__IDCT_RD_ERR_EN_MASK 0x00080000L +#define UVD_MEMCHECK_SYS_INT_EN__MPC_RD_ERR_EN_MASK 0x00100000L +#define UVD_MEMCHECK_SYS_INT_EN__LBSI_RD_ERR_EN_MASK 0x00200000L +#define UVD_MEMCHECK_SYS_INT_EN__RBC_RD_ERR_EN_MASK 0x01000000L +#define UVD_MEMCHECK_SYS_INT_EN__MIF_BSP2_ERR_EN_MASK 0x08000000L +#define UVD_MEMCHECK_SYS_INT_EN__MIF_BSP3_ERR_EN_MASK 0x10000000L +#define UVD_MEMCHECK_SYS_INT_EN__MIF_SCLR_ERR_EN_MASK 0x20000000L +#define UVD_MEMCHECK_SYS_INT_EN__MIF_SCLR2_ERR_EN_MASK 0x40000000L +#define UVD_MEMCHECK_SYS_INT_EN__PREF_ERR_EN_MASK 0x80000000L +//UVD_MEMCHECK_SYS_INT_STAT +#define UVD_MEMCHECK_SYS_INT_STAT__RE_LO_ERR__SHIFT 0x0 +#define UVD_MEMCHECK_SYS_INT_STAT__RE_HI_ERR__SHIFT 0x1 +#define UVD_MEMCHECK_SYS_INT_STAT__IT_LO_ERR__SHIFT 0x2 +#define UVD_MEMCHECK_SYS_INT_STAT__IT_HI_ERR__SHIFT 0x3 +#define UVD_MEMCHECK_SYS_INT_STAT__MP_LO_ERR__SHIFT 0x4 +#define UVD_MEMCHECK_SYS_INT_STAT__MP_HI_ERR__SHIFT 0x5 +#define UVD_MEMCHECK_SYS_INT_STAT__DB_LO_ERR__SHIFT 0x6 +#define UVD_MEMCHECK_SYS_INT_STAT__DB_HI_ERR__SHIFT 0x7 +#define UVD_MEMCHECK_SYS_INT_STAT__DBW_LO_ERR__SHIFT 0x8 +#define UVD_MEMCHECK_SYS_INT_STAT__DBW_HI_ERR__SHIFT 0x9 +#define UVD_MEMCHECK_SYS_INT_STAT__CM_LO_ERR__SHIFT 0xa +#define UVD_MEMCHECK_SYS_INT_STAT__CM_HI_ERR__SHIFT 0xb +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_REF_LO_ERR__SHIFT 0xc +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_REF_HI_ERR__SHIFT 0xd +#define UVD_MEMCHECK_SYS_INT_STAT__VCPU_LO_ERR__SHIFT 0xe +#define UVD_MEMCHECK_SYS_INT_STAT__VCPU_HI_ERR__SHIFT 0xf +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_DBW_LO_ERR__SHIFT 0x10 +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_DBW_HI_ERR__SHIFT 0x11 +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_CM_COLOC_LO_ERR__SHIFT 0x12 +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_CM_COLOC_HI_ERR__SHIFT 0x13 +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_BSP0_LO_ERR__SHIFT 0x14 +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_BSP0_HI_ERR__SHIFT 0x15 +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_BSP1_LO_ERR__SHIFT 0x16 +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_BSP1_HI_ERR__SHIFT 0x17 +#define UVD_MEMCHECK_SYS_INT_STAT__SRE_LO_ERR__SHIFT 0x18 +#define UVD_MEMCHECK_SYS_INT_STAT__SRE_HI_ERR__SHIFT 0x19 +#define UVD_MEMCHECK_SYS_INT_STAT__IT_RD_LO_ERR__SHIFT 0x1e +#define UVD_MEMCHECK_SYS_INT_STAT__IT_RD_HI_ERR__SHIFT 0x1f +#define UVD_MEMCHECK_SYS_INT_STAT__RE_LO_ERR_MASK 0x00000001L +#define UVD_MEMCHECK_SYS_INT_STAT__RE_HI_ERR_MASK 0x00000002L +#define UVD_MEMCHECK_SYS_INT_STAT__IT_LO_ERR_MASK 0x00000004L +#define UVD_MEMCHECK_SYS_INT_STAT__IT_HI_ERR_MASK 0x00000008L +#define UVD_MEMCHECK_SYS_INT_STAT__MP_LO_ERR_MASK 0x00000010L +#define UVD_MEMCHECK_SYS_INT_STAT__MP_HI_ERR_MASK 0x00000020L +#define UVD_MEMCHECK_SYS_INT_STAT__DB_LO_ERR_MASK 0x00000040L +#define UVD_MEMCHECK_SYS_INT_STAT__DB_HI_ERR_MASK 0x00000080L +#define UVD_MEMCHECK_SYS_INT_STAT__DBW_LO_ERR_MASK 0x00000100L +#define UVD_MEMCHECK_SYS_INT_STAT__DBW_HI_ERR_MASK 0x00000200L +#define UVD_MEMCHECK_SYS_INT_STAT__CM_LO_ERR_MASK 0x00000400L +#define UVD_MEMCHECK_SYS_INT_STAT__CM_HI_ERR_MASK 0x00000800L +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_REF_LO_ERR_MASK 0x00001000L +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_REF_HI_ERR_MASK 0x00002000L +#define UVD_MEMCHECK_SYS_INT_STAT__VCPU_LO_ERR_MASK 0x00004000L +#define UVD_MEMCHECK_SYS_INT_STAT__VCPU_HI_ERR_MASK 0x00008000L +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_DBW_LO_ERR_MASK 0x00010000L +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_DBW_HI_ERR_MASK 0x00020000L +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_CM_COLOC_LO_ERR_MASK 0x00040000L +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_CM_COLOC_HI_ERR_MASK 0x00080000L +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_BSP0_LO_ERR_MASK 0x00100000L +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_BSP0_HI_ERR_MASK 0x00200000L +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_BSP1_LO_ERR_MASK 0x00400000L +#define UVD_MEMCHECK_SYS_INT_STAT__MIF_BSP1_HI_ERR_MASK 0x00800000L +#define UVD_MEMCHECK_SYS_INT_STAT__SRE_LO_ERR_MASK 0x01000000L +#define UVD_MEMCHECK_SYS_INT_STAT__SRE_HI_ERR_MASK 0x02000000L +#define UVD_MEMCHECK_SYS_INT_STAT__IT_RD_LO_ERR_MASK 0x40000000L +#define UVD_MEMCHECK_SYS_INT_STAT__IT_RD_HI_ERR_MASK 0x80000000L +//UVD_MEMCHECK_SYS_INT_ACK +#define UVD_MEMCHECK_SYS_INT_ACK__RE_LO_ACK__SHIFT 0x0 +#define UVD_MEMCHECK_SYS_INT_ACK__RE_HI_ACK__SHIFT 0x1 +#define UVD_MEMCHECK_SYS_INT_ACK__IT_LO_ACK__SHIFT 0x2 +#define UVD_MEMCHECK_SYS_INT_ACK__IT_HI_ACK__SHIFT 0x3 +#define UVD_MEMCHECK_SYS_INT_ACK__MP_LO_ACK__SHIFT 0x4 +#define UVD_MEMCHECK_SYS_INT_ACK__MP_HI_ACK__SHIFT 0x5 +#define UVD_MEMCHECK_SYS_INT_ACK__DB_LO_ACK__SHIFT 0x6 +#define UVD_MEMCHECK_SYS_INT_ACK__DB_HI_ACK__SHIFT 0x7 +#define UVD_MEMCHECK_SYS_INT_ACK__DBW_LO_ACK__SHIFT 0x8 +#define UVD_MEMCHECK_SYS_INT_ACK__DBW_HI_ACK__SHIFT 0x9 +#define UVD_MEMCHECK_SYS_INT_ACK__CM_LO_ACK__SHIFT 0xa +#define UVD_MEMCHECK_SYS_INT_ACK__CM_HI_ACK__SHIFT 0xb +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_REF_LO_ACK__SHIFT 0xc +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_REF_HI_ACK__SHIFT 0xd +#define UVD_MEMCHECK_SYS_INT_ACK__VCPU_LO_ACK__SHIFT 0xe +#define UVD_MEMCHECK_SYS_INT_ACK__VCPU_HI_ACK__SHIFT 0xf +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_DBW_LO_ACK__SHIFT 0x10 +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_DBW_HI_ACK__SHIFT 0x11 +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_CM_COLOC_LO_ACK__SHIFT 0x12 +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_CM_COLOC_HI_ACK__SHIFT 0x13 +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_BSP0_LO_ACK__SHIFT 0x14 +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_BSP0_HI_ACK__SHIFT 0x15 +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_BSP1_LO_ACK__SHIFT 0x16 +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_BSP1_HI_ACK__SHIFT 0x17 +#define UVD_MEMCHECK_SYS_INT_ACK__SRE_LO_ACK__SHIFT 0x18 +#define UVD_MEMCHECK_SYS_INT_ACK__SRE_HI_ACK__SHIFT 0x19 +#define UVD_MEMCHECK_SYS_INT_ACK__IT_RD_LO_ACK__SHIFT 0x1e +#define UVD_MEMCHECK_SYS_INT_ACK__IT_RD_HI_ACK__SHIFT 0x1f +#define UVD_MEMCHECK_SYS_INT_ACK__RE_LO_ACK_MASK 0x00000001L +#define UVD_MEMCHECK_SYS_INT_ACK__RE_HI_ACK_MASK 0x00000002L +#define UVD_MEMCHECK_SYS_INT_ACK__IT_LO_ACK_MASK 0x00000004L +#define UVD_MEMCHECK_SYS_INT_ACK__IT_HI_ACK_MASK 0x00000008L +#define UVD_MEMCHECK_SYS_INT_ACK__MP_LO_ACK_MASK 0x00000010L +#define UVD_MEMCHECK_SYS_INT_ACK__MP_HI_ACK_MASK 0x00000020L +#define UVD_MEMCHECK_SYS_INT_ACK__DB_LO_ACK_MASK 0x00000040L +#define UVD_MEMCHECK_SYS_INT_ACK__DB_HI_ACK_MASK 0x00000080L +#define UVD_MEMCHECK_SYS_INT_ACK__DBW_LO_ACK_MASK 0x00000100L +#define UVD_MEMCHECK_SYS_INT_ACK__DBW_HI_ACK_MASK 0x00000200L +#define UVD_MEMCHECK_SYS_INT_ACK__CM_LO_ACK_MASK 0x00000400L +#define UVD_MEMCHECK_SYS_INT_ACK__CM_HI_ACK_MASK 0x00000800L +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_REF_LO_ACK_MASK 0x00001000L +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_REF_HI_ACK_MASK 0x00002000L +#define UVD_MEMCHECK_SYS_INT_ACK__VCPU_LO_ACK_MASK 0x00004000L +#define UVD_MEMCHECK_SYS_INT_ACK__VCPU_HI_ACK_MASK 0x00008000L +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_DBW_LO_ACK_MASK 0x00010000L +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_DBW_HI_ACK_MASK 0x00020000L +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_CM_COLOC_LO_ACK_MASK 0x00040000L +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_CM_COLOC_HI_ACK_MASK 0x00080000L +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_BSP0_LO_ACK_MASK 0x00100000L +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_BSP0_HI_ACK_MASK 0x00200000L +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_BSP1_LO_ACK_MASK 0x00400000L +#define UVD_MEMCHECK_SYS_INT_ACK__MIF_BSP1_HI_ACK_MASK 0x00800000L +#define UVD_MEMCHECK_SYS_INT_ACK__SRE_LO_ACK_MASK 0x01000000L +#define UVD_MEMCHECK_SYS_INT_ACK__SRE_HI_ACK_MASK 0x02000000L +#define UVD_MEMCHECK_SYS_INT_ACK__IT_RD_LO_ACK_MASK 0x40000000L +#define UVD_MEMCHECK_SYS_INT_ACK__IT_RD_HI_ACK_MASK 0x80000000L +//UVD_MEMCHECK_VCPU_INT_EN +#define UVD_MEMCHECK_VCPU_INT_EN__RE_ERR_EN__SHIFT 0x0 +#define UVD_MEMCHECK_VCPU_INT_EN__IT_ERR_EN__SHIFT 0x1 +#define UVD_MEMCHECK_VCPU_INT_EN__MP_ERR_EN__SHIFT 0x2 +#define UVD_MEMCHECK_VCPU_INT_EN__DB_ERR_EN__SHIFT 0x3 +#define UVD_MEMCHECK_VCPU_INT_EN__DBW_ERR_EN__SHIFT 0x4 +#define UVD_MEMCHECK_VCPU_INT_EN__CM_ERR_EN__SHIFT 0x5 +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_REF_ERR_EN__SHIFT 0x6 +#define UVD_MEMCHECK_VCPU_INT_EN__VCPU_ERR_EN__SHIFT 0x7 +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_DBW_ERR_EN__SHIFT 0x8 +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_CM_COLOC_ERR_EN__SHIFT 0x9 +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_BSP0_ERR_EN__SHIFT 0xa +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_BSP1_ERR_EN__SHIFT 0xb +#define UVD_MEMCHECK_VCPU_INT_EN__SRE_ERR_EN__SHIFT 0xc +#define UVD_MEMCHECK_VCPU_INT_EN__IT_RD_ERR_EN__SHIFT 0xf +#define UVD_MEMCHECK_VCPU_INT_EN__CM_RD_ERR_EN__SHIFT 0x10 +#define UVD_MEMCHECK_VCPU_INT_EN__DB_RD_ERR_EN__SHIFT 0x11 +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_RD_ERR_EN__SHIFT 0x12 +#define UVD_MEMCHECK_VCPU_INT_EN__IDCT_RD_ERR_EN__SHIFT 0x13 +#define UVD_MEMCHECK_VCPU_INT_EN__MPC_RD_ERR_EN__SHIFT 0x14 +#define UVD_MEMCHECK_VCPU_INT_EN__LBSI_RD_ERR_EN__SHIFT 0x15 +#define UVD_MEMCHECK_VCPU_INT_EN__RBC_RD_ERR_EN__SHIFT 0x18 +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_BSP2_ERR_EN__SHIFT 0x19 +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_BSP3_ERR_EN__SHIFT 0x1a +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_SCLR_ERR_EN__SHIFT 0x1b +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_SCLR2_ERR_EN__SHIFT 0x1c +#define UVD_MEMCHECK_VCPU_INT_EN__PREF_ERR_EN__SHIFT 0x1d +#define UVD_MEMCHECK_VCPU_INT_EN__RE_ERR_EN_MASK 0x00000001L +#define UVD_MEMCHECK_VCPU_INT_EN__IT_ERR_EN_MASK 0x00000002L +#define UVD_MEMCHECK_VCPU_INT_EN__MP_ERR_EN_MASK 0x00000004L +#define UVD_MEMCHECK_VCPU_INT_EN__DB_ERR_EN_MASK 0x00000008L +#define UVD_MEMCHECK_VCPU_INT_EN__DBW_ERR_EN_MASK 0x00000010L +#define UVD_MEMCHECK_VCPU_INT_EN__CM_ERR_EN_MASK 0x00000020L +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_REF_ERR_EN_MASK 0x00000040L +#define UVD_MEMCHECK_VCPU_INT_EN__VCPU_ERR_EN_MASK 0x00000080L +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_DBW_ERR_EN_MASK 0x00000100L +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_CM_COLOC_ERR_EN_MASK 0x00000200L +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_BSP0_ERR_EN_MASK 0x00000400L +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_BSP1_ERR_EN_MASK 0x00000800L +#define UVD_MEMCHECK_VCPU_INT_EN__SRE_ERR_EN_MASK 0x00001000L +#define UVD_MEMCHECK_VCPU_INT_EN__IT_RD_ERR_EN_MASK 0x00008000L +#define UVD_MEMCHECK_VCPU_INT_EN__CM_RD_ERR_EN_MASK 0x00010000L +#define UVD_MEMCHECK_VCPU_INT_EN__DB_RD_ERR_EN_MASK 0x00020000L +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_RD_ERR_EN_MASK 0x00040000L +#define UVD_MEMCHECK_VCPU_INT_EN__IDCT_RD_ERR_EN_MASK 0x00080000L +#define UVD_MEMCHECK_VCPU_INT_EN__MPC_RD_ERR_EN_MASK 0x00100000L +#define UVD_MEMCHECK_VCPU_INT_EN__LBSI_RD_ERR_EN_MASK 0x00200000L +#define UVD_MEMCHECK_VCPU_INT_EN__RBC_RD_ERR_EN_MASK 0x01000000L +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_BSP2_ERR_EN_MASK 0x02000000L +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_BSP3_ERR_EN_MASK 0x04000000L +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_SCLR_ERR_EN_MASK 0x08000000L +#define UVD_MEMCHECK_VCPU_INT_EN__MIF_SCLR2_ERR_EN_MASK 0x10000000L +#define UVD_MEMCHECK_VCPU_INT_EN__PREF_ERR_EN_MASK 0x20000000L +//UVD_MEMCHECK_VCPU_INT_STAT +#define UVD_MEMCHECK_VCPU_INT_STAT__RE_LO_ERR__SHIFT 0x0 +#define UVD_MEMCHECK_VCPU_INT_STAT__RE_HI_ERR__SHIFT 0x1 +#define UVD_MEMCHECK_VCPU_INT_STAT__IT_LO_ERR__SHIFT 0x2 +#define UVD_MEMCHECK_VCPU_INT_STAT__IT_HI_ERR__SHIFT 0x3 +#define UVD_MEMCHECK_VCPU_INT_STAT__MP_LO_ERR__SHIFT 0x4 +#define UVD_MEMCHECK_VCPU_INT_STAT__MP_HI_ERR__SHIFT 0x5 +#define UVD_MEMCHECK_VCPU_INT_STAT__DB_LO_ERR__SHIFT 0x6 +#define UVD_MEMCHECK_VCPU_INT_STAT__DB_HI_ERR__SHIFT 0x7 +#define UVD_MEMCHECK_VCPU_INT_STAT__DBW_LO_ERR__SHIFT 0x8 +#define UVD_MEMCHECK_VCPU_INT_STAT__DBW_HI_ERR__SHIFT 0x9 +#define UVD_MEMCHECK_VCPU_INT_STAT__CM_LO_ERR__SHIFT 0xa +#define UVD_MEMCHECK_VCPU_INT_STAT__CM_HI_ERR__SHIFT 0xb +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_REF_LO_ERR__SHIFT 0xc +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_REF_HI_ERR__SHIFT 0xd +#define UVD_MEMCHECK_VCPU_INT_STAT__VCPU_LO_ERR__SHIFT 0xe +#define UVD_MEMCHECK_VCPU_INT_STAT__VCPU_HI_ERR__SHIFT 0xf +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_DBW_LO_ERR__SHIFT 0x10 +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_DBW_HI_ERR__SHIFT 0x11 +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_CM_COLOC_LO_ERR__SHIFT 0x12 +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_CM_COLOC_HI_ERR__SHIFT 0x13 +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_BSP0_LO_ERR__SHIFT 0x14 +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_BSP0_HI_ERR__SHIFT 0x15 +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_BSP1_LO_ERR__SHIFT 0x16 +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_BSP1_HI_ERR__SHIFT 0x17 +#define UVD_MEMCHECK_VCPU_INT_STAT__SRE_LO_ERR__SHIFT 0x18 +#define UVD_MEMCHECK_VCPU_INT_STAT__SRE_HI_ERR__SHIFT 0x19 +#define UVD_MEMCHECK_VCPU_INT_STAT__IT_RD_LO_ERR__SHIFT 0x1e +#define UVD_MEMCHECK_VCPU_INT_STAT__IT_RD_HI_ERR__SHIFT 0x1f +#define UVD_MEMCHECK_VCPU_INT_STAT__RE_LO_ERR_MASK 0x00000001L +#define UVD_MEMCHECK_VCPU_INT_STAT__RE_HI_ERR_MASK 0x00000002L +#define UVD_MEMCHECK_VCPU_INT_STAT__IT_LO_ERR_MASK 0x00000004L +#define UVD_MEMCHECK_VCPU_INT_STAT__IT_HI_ERR_MASK 0x00000008L +#define UVD_MEMCHECK_VCPU_INT_STAT__MP_LO_ERR_MASK 0x00000010L +#define UVD_MEMCHECK_VCPU_INT_STAT__MP_HI_ERR_MASK 0x00000020L +#define UVD_MEMCHECK_VCPU_INT_STAT__DB_LO_ERR_MASK 0x00000040L +#define UVD_MEMCHECK_VCPU_INT_STAT__DB_HI_ERR_MASK 0x00000080L +#define UVD_MEMCHECK_VCPU_INT_STAT__DBW_LO_ERR_MASK 0x00000100L +#define UVD_MEMCHECK_VCPU_INT_STAT__DBW_HI_ERR_MASK 0x00000200L +#define UVD_MEMCHECK_VCPU_INT_STAT__CM_LO_ERR_MASK 0x00000400L +#define UVD_MEMCHECK_VCPU_INT_STAT__CM_HI_ERR_MASK 0x00000800L +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_REF_LO_ERR_MASK 0x00001000L +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_REF_HI_ERR_MASK 0x00002000L +#define UVD_MEMCHECK_VCPU_INT_STAT__VCPU_LO_ERR_MASK 0x00004000L +#define UVD_MEMCHECK_VCPU_INT_STAT__VCPU_HI_ERR_MASK 0x00008000L +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_DBW_LO_ERR_MASK 0x00010000L +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_DBW_HI_ERR_MASK 0x00020000L +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_CM_COLOC_LO_ERR_MASK 0x00040000L +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_CM_COLOC_HI_ERR_MASK 0x00080000L +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_BSP0_LO_ERR_MASK 0x00100000L +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_BSP0_HI_ERR_MASK 0x00200000L +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_BSP1_LO_ERR_MASK 0x00400000L +#define UVD_MEMCHECK_VCPU_INT_STAT__MIF_BSP1_HI_ERR_MASK 0x00800000L +#define UVD_MEMCHECK_VCPU_INT_STAT__SRE_LO_ERR_MASK 0x01000000L +#define UVD_MEMCHECK_VCPU_INT_STAT__SRE_HI_ERR_MASK 0x02000000L +#define UVD_MEMCHECK_VCPU_INT_STAT__IT_RD_LO_ERR_MASK 0x40000000L +#define UVD_MEMCHECK_VCPU_INT_STAT__IT_RD_HI_ERR_MASK 0x80000000L +//UVD_MEMCHECK_VCPU_INT_ACK +#define UVD_MEMCHECK_VCPU_INT_ACK__RE_LO_ACK__SHIFT 0x0 +#define UVD_MEMCHECK_VCPU_INT_ACK__RE_HI_ACK__SHIFT 0x1 +#define UVD_MEMCHECK_VCPU_INT_ACK__IT_LO_ACK__SHIFT 0x2 +#define UVD_MEMCHECK_VCPU_INT_ACK__IT_HI_ACK__SHIFT 0x3 +#define UVD_MEMCHECK_VCPU_INT_ACK__MP_LO_ACK__SHIFT 0x4 +#define UVD_MEMCHECK_VCPU_INT_ACK__MP_HI_ACK__SHIFT 0x5 +#define UVD_MEMCHECK_VCPU_INT_ACK__DB_LO_ACK__SHIFT 0x6 +#define UVD_MEMCHECK_VCPU_INT_ACK__DB_HI_ACK__SHIFT 0x7 +#define UVD_MEMCHECK_VCPU_INT_ACK__DBW_LO_ACK__SHIFT 0x8 +#define UVD_MEMCHECK_VCPU_INT_ACK__DBW_HI_ACK__SHIFT 0x9 +#define UVD_MEMCHECK_VCPU_INT_ACK__CM_LO_ACK__SHIFT 0xa +#define UVD_MEMCHECK_VCPU_INT_ACK__CM_HI_ACK__SHIFT 0xb +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_REF_LO_ACK__SHIFT 0xc +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_REF_HI_ACK__SHIFT 0xd +#define UVD_MEMCHECK_VCPU_INT_ACK__VCPU_LO_ACK__SHIFT 0xe +#define UVD_MEMCHECK_VCPU_INT_ACK__VCPU_HI_ACK__SHIFT 0xf +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_DBW_LO_ACK__SHIFT 0x10 +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_DBW_HI_ACK__SHIFT 0x11 +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_CM_COLOC_LO_ACK__SHIFT 0x12 +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_CM_COLOC_HI_ACK__SHIFT 0x13 +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_BSP0_LO_ACK__SHIFT 0x14 +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_BSP0_HI_ACK__SHIFT 0x15 +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_BSP1_LO_ACK__SHIFT 0x16 +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_BSP1_HI_ACK__SHIFT 0x17 +#define UVD_MEMCHECK_VCPU_INT_ACK__SRE_LO_ACK__SHIFT 0x18 +#define UVD_MEMCHECK_VCPU_INT_ACK__SRE_HI_ACK__SHIFT 0x19 +#define UVD_MEMCHECK_VCPU_INT_ACK__IT_RD_LO_ACK__SHIFT 0x1e +#define UVD_MEMCHECK_VCPU_INT_ACK__IT_RD_HI_ACK__SHIFT 0x1f +#define UVD_MEMCHECK_VCPU_INT_ACK__RE_LO_ACK_MASK 0x00000001L +#define UVD_MEMCHECK_VCPU_INT_ACK__RE_HI_ACK_MASK 0x00000002L +#define UVD_MEMCHECK_VCPU_INT_ACK__IT_LO_ACK_MASK 0x00000004L +#define UVD_MEMCHECK_VCPU_INT_ACK__IT_HI_ACK_MASK 0x00000008L +#define UVD_MEMCHECK_VCPU_INT_ACK__MP_LO_ACK_MASK 0x00000010L +#define UVD_MEMCHECK_VCPU_INT_ACK__MP_HI_ACK_MASK 0x00000020L +#define UVD_MEMCHECK_VCPU_INT_ACK__DB_LO_ACK_MASK 0x00000040L +#define UVD_MEMCHECK_VCPU_INT_ACK__DB_HI_ACK_MASK 0x00000080L +#define UVD_MEMCHECK_VCPU_INT_ACK__DBW_LO_ACK_MASK 0x00000100L +#define UVD_MEMCHECK_VCPU_INT_ACK__DBW_HI_ACK_MASK 0x00000200L +#define UVD_MEMCHECK_VCPU_INT_ACK__CM_LO_ACK_MASK 0x00000400L +#define UVD_MEMCHECK_VCPU_INT_ACK__CM_HI_ACK_MASK 0x00000800L +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_REF_LO_ACK_MASK 0x00001000L +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_REF_HI_ACK_MASK 0x00002000L +#define UVD_MEMCHECK_VCPU_INT_ACK__VCPU_LO_ACK_MASK 0x00004000L +#define UVD_MEMCHECK_VCPU_INT_ACK__VCPU_HI_ACK_MASK 0x00008000L +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_DBW_LO_ACK_MASK 0x00010000L +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_DBW_HI_ACK_MASK 0x00020000L +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_CM_COLOC_LO_ACK_MASK 0x00040000L +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_CM_COLOC_HI_ACK_MASK 0x00080000L +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_BSP0_LO_ACK_MASK 0x00100000L +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_BSP0_HI_ACK_MASK 0x00200000L +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_BSP1_LO_ACK_MASK 0x00400000L +#define UVD_MEMCHECK_VCPU_INT_ACK__MIF_BSP1_HI_ACK_MASK 0x00800000L +#define UVD_MEMCHECK_VCPU_INT_ACK__SRE_LO_ACK_MASK 0x01000000L +#define UVD_MEMCHECK_VCPU_INT_ACK__SRE_HI_ACK_MASK 0x02000000L +#define UVD_MEMCHECK_VCPU_INT_ACK__IT_RD_LO_ACK_MASK 0x40000000L +#define UVD_MEMCHECK_VCPU_INT_ACK__IT_RD_HI_ACK_MASK 0x80000000L +//UVD_MEMCHECK2_SYS_INT_STAT +#define UVD_MEMCHECK2_SYS_INT_STAT__CM_RD_LO_ERR__SHIFT 0x0 +#define UVD_MEMCHECK2_SYS_INT_STAT__CM_RD_HI_ERR__SHIFT 0x1 +#define UVD_MEMCHECK2_SYS_INT_STAT__DB_RD_LO_ERR__SHIFT 0x2 +#define UVD_MEMCHECK2_SYS_INT_STAT__DB_RD_HI_ERR__SHIFT 0x3 +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_RD_LO_ERR__SHIFT 0x4 +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_RD_HI_ERR__SHIFT 0x5 +#define UVD_MEMCHECK2_SYS_INT_STAT__IDCT_RD_LO_ERR__SHIFT 0x6 +#define UVD_MEMCHECK2_SYS_INT_STAT__IDCT_RD_HI_ERR__SHIFT 0x7 +#define UVD_MEMCHECK2_SYS_INT_STAT__MPC_RD_LO_ERR__SHIFT 0x8 +#define UVD_MEMCHECK2_SYS_INT_STAT__MPC_RD_HI_ERR__SHIFT 0x9 +#define UVD_MEMCHECK2_SYS_INT_STAT__LBSI_RD_LO_ERR__SHIFT 0xa +#define UVD_MEMCHECK2_SYS_INT_STAT__LBSI_RD_HI_ERR__SHIFT 0xb +#define UVD_MEMCHECK2_SYS_INT_STAT__RBC_RD_LO_ERR__SHIFT 0x10 +#define UVD_MEMCHECK2_SYS_INT_STAT__RBC_RD_HI_ERR__SHIFT 0x11 +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_BSP2_LO_ERR__SHIFT 0x16 +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_BSP2_HI_ERR__SHIFT 0x17 +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_BSP3_LO_ERR__SHIFT 0x18 +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_BSP3_HI_ERR__SHIFT 0x19 +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_SCLR_LO_ERR__SHIFT 0x1a +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_SCLR_HI_ERR__SHIFT 0x1b +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_SCLR2_LO_ERR__SHIFT 0x1c +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_SCLR2_HI_ERR__SHIFT 0x1d +#define UVD_MEMCHECK2_SYS_INT_STAT__PREF_LO_ERR__SHIFT 0x1e +#define UVD_MEMCHECK2_SYS_INT_STAT__PREF_HI_ERR__SHIFT 0x1f +#define UVD_MEMCHECK2_SYS_INT_STAT__CM_RD_LO_ERR_MASK 0x00000001L +#define UVD_MEMCHECK2_SYS_INT_STAT__CM_RD_HI_ERR_MASK 0x00000002L +#define UVD_MEMCHECK2_SYS_INT_STAT__DB_RD_LO_ERR_MASK 0x00000004L +#define UVD_MEMCHECK2_SYS_INT_STAT__DB_RD_HI_ERR_MASK 0x00000008L +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_RD_LO_ERR_MASK 0x00000010L +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_RD_HI_ERR_MASK 0x00000020L +#define UVD_MEMCHECK2_SYS_INT_STAT__IDCT_RD_LO_ERR_MASK 0x00000040L +#define UVD_MEMCHECK2_SYS_INT_STAT__IDCT_RD_HI_ERR_MASK 0x00000080L +#define UVD_MEMCHECK2_SYS_INT_STAT__MPC_RD_LO_ERR_MASK 0x00000100L +#define UVD_MEMCHECK2_SYS_INT_STAT__MPC_RD_HI_ERR_MASK 0x00000200L +#define UVD_MEMCHECK2_SYS_INT_STAT__LBSI_RD_LO_ERR_MASK 0x00000400L +#define UVD_MEMCHECK2_SYS_INT_STAT__LBSI_RD_HI_ERR_MASK 0x00000800L +#define UVD_MEMCHECK2_SYS_INT_STAT__RBC_RD_LO_ERR_MASK 0x00010000L +#define UVD_MEMCHECK2_SYS_INT_STAT__RBC_RD_HI_ERR_MASK 0x00020000L +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_BSP2_LO_ERR_MASK 0x00400000L +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_BSP2_HI_ERR_MASK 0x00800000L +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_BSP3_LO_ERR_MASK 0x01000000L +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_BSP3_HI_ERR_MASK 0x02000000L +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_SCLR_LO_ERR_MASK 0x04000000L +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_SCLR_HI_ERR_MASK 0x08000000L +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_SCLR2_LO_ERR_MASK 0x10000000L +#define UVD_MEMCHECK2_SYS_INT_STAT__MIF_SCLR2_HI_ERR_MASK 0x20000000L +#define UVD_MEMCHECK2_SYS_INT_STAT__PREF_LO_ERR_MASK 0x40000000L +#define UVD_MEMCHECK2_SYS_INT_STAT__PREF_HI_ERR_MASK 0x80000000L +//UVD_MEMCHECK2_SYS_INT_ACK +#define UVD_MEMCHECK2_SYS_INT_ACK__CM_RD_LO_ACK__SHIFT 0x0 +#define UVD_MEMCHECK2_SYS_INT_ACK__CM_RD_HI_ACK__SHIFT 0x1 +#define UVD_MEMCHECK2_SYS_INT_ACK__DB_RD_LO_ACK__SHIFT 0x2 +#define UVD_MEMCHECK2_SYS_INT_ACK__DB_RD_HI_ACK__SHIFT 0x3 +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_RD_LO_ACK__SHIFT 0x4 +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_RD_HI_ACK__SHIFT 0x5 +#define UVD_MEMCHECK2_SYS_INT_ACK__IDCT_RD_LO_ACK__SHIFT 0x6 +#define UVD_MEMCHECK2_SYS_INT_ACK__IDCT_RD_HI_ACK__SHIFT 0x7 +#define UVD_MEMCHECK2_SYS_INT_ACK__MPC_RD_LO_ACK__SHIFT 0x8 +#define UVD_MEMCHECK2_SYS_INT_ACK__MPC_RD_HI_ACK__SHIFT 0x9 +#define UVD_MEMCHECK2_SYS_INT_ACK__LBSI_RD_LO_ACK__SHIFT 0xa +#define UVD_MEMCHECK2_SYS_INT_ACK__LBSI_RD_HI_ACK__SHIFT 0xb +#define UVD_MEMCHECK2_SYS_INT_ACK__RBC_RD_LO_ACK__SHIFT 0x10 +#define UVD_MEMCHECK2_SYS_INT_ACK__RBC_RD_HI_ACK__SHIFT 0x11 +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_BSP2_LO_ACK__SHIFT 0x16 +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_BSP2_HI_ACK__SHIFT 0x17 +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_BSP3_LO_ACK__SHIFT 0x18 +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_BSP3_HI_ACK__SHIFT 0x19 +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_SCLR_LO_ACK__SHIFT 0x1a +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_SCLR_HI_ACK__SHIFT 0x1b +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_SCLR2_LO_ACK__SHIFT 0x1c +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_SCLR2_HI_ACK__SHIFT 0x1d +#define UVD_MEMCHECK2_SYS_INT_ACK__PREF_LO_ACK__SHIFT 0x1e +#define UVD_MEMCHECK2_SYS_INT_ACK__PREF_HI_ACK__SHIFT 0x1f +#define UVD_MEMCHECK2_SYS_INT_ACK__CM_RD_LO_ACK_MASK 0x00000001L +#define UVD_MEMCHECK2_SYS_INT_ACK__CM_RD_HI_ACK_MASK 0x00000002L +#define UVD_MEMCHECK2_SYS_INT_ACK__DB_RD_LO_ACK_MASK 0x00000004L +#define UVD_MEMCHECK2_SYS_INT_ACK__DB_RD_HI_ACK_MASK 0x00000008L +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_RD_LO_ACK_MASK 0x00000010L +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_RD_HI_ACK_MASK 0x00000020L +#define UVD_MEMCHECK2_SYS_INT_ACK__IDCT_RD_LO_ACK_MASK 0x00000040L +#define UVD_MEMCHECK2_SYS_INT_ACK__IDCT_RD_HI_ACK_MASK 0x00000080L +#define UVD_MEMCHECK2_SYS_INT_ACK__MPC_RD_LO_ACK_MASK 0x00000100L +#define UVD_MEMCHECK2_SYS_INT_ACK__MPC_RD_HI_ACK_MASK 0x00000200L +#define UVD_MEMCHECK2_SYS_INT_ACK__LBSI_RD_LO_ACK_MASK 0x00000400L +#define UVD_MEMCHECK2_SYS_INT_ACK__LBSI_RD_HI_ACK_MASK 0x00000800L +#define UVD_MEMCHECK2_SYS_INT_ACK__RBC_RD_LO_ACK_MASK 0x00010000L +#define UVD_MEMCHECK2_SYS_INT_ACK__RBC_RD_HI_ACK_MASK 0x00020000L +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_BSP2_LO_ACK_MASK 0x00400000L +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_BSP2_HI_ACK_MASK 0x00800000L +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_BSP3_LO_ACK_MASK 0x01000000L +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_BSP3_HI_ACK_MASK 0x02000000L +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_SCLR_LO_ACK_MASK 0x04000000L +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_SCLR_HI_ACK_MASK 0x08000000L +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_SCLR2_LO_ACK_MASK 0x10000000L +#define UVD_MEMCHECK2_SYS_INT_ACK__MIF_SCLR2_HI_ACK_MASK 0x20000000L +#define UVD_MEMCHECK2_SYS_INT_ACK__PREF_LO_ACK_MASK 0x40000000L +#define UVD_MEMCHECK2_SYS_INT_ACK__PREF_HI_ACK_MASK 0x80000000L +//UVD_MEMCHECK2_VCPU_INT_STAT +#define UVD_MEMCHECK2_VCPU_INT_STAT__CM_RD_LO_ERR__SHIFT 0x0 +#define UVD_MEMCHECK2_VCPU_INT_STAT__CM_RD_HI_ERR__SHIFT 0x1 +#define UVD_MEMCHECK2_VCPU_INT_STAT__DB_RD_LO_ERR__SHIFT 0x2 +#define UVD_MEMCHECK2_VCPU_INT_STAT__DB_RD_HI_ERR__SHIFT 0x3 +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_RD_LO_ERR__SHIFT 0x4 +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_RD_HI_ERR__SHIFT 0x5 +#define UVD_MEMCHECK2_VCPU_INT_STAT__IDCT_RD_LO_ERR__SHIFT 0x6 +#define UVD_MEMCHECK2_VCPU_INT_STAT__IDCT_RD_HI_ERR__SHIFT 0x7 +#define UVD_MEMCHECK2_VCPU_INT_STAT__MPC_RD_LO_ERR__SHIFT 0x8 +#define UVD_MEMCHECK2_VCPU_INT_STAT__MPC_RD_HI_ERR__SHIFT 0x9 +#define UVD_MEMCHECK2_VCPU_INT_STAT__LBSI_RD_LO_ERR__SHIFT 0xa +#define UVD_MEMCHECK2_VCPU_INT_STAT__LBSI_RD_HI_ERR__SHIFT 0xb +#define UVD_MEMCHECK2_VCPU_INT_STAT__RBC_RD_LO_ERR__SHIFT 0x10 +#define UVD_MEMCHECK2_VCPU_INT_STAT__RBC_RD_HI_ERR__SHIFT 0x11 +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_BSP2_LO_ERR__SHIFT 0x12 +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_BSP2_HI_ERR__SHIFT 0x13 +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_BSP3_LO_ERR__SHIFT 0x14 +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_BSP3_HI_ERR__SHIFT 0x15 +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_SCLR_LO_ERR__SHIFT 0x16 +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_SCLR_HI_ERR__SHIFT 0x17 +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_SCLR2_LO_ERR__SHIFT 0x18 +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_SCLR2_HI_ERR__SHIFT 0x19 +#define UVD_MEMCHECK2_VCPU_INT_STAT__PREF_LO_ERR__SHIFT 0x1a +#define UVD_MEMCHECK2_VCPU_INT_STAT__PREF_HI_ERR__SHIFT 0x1b +#define UVD_MEMCHECK2_VCPU_INT_STAT__CM_RD_LO_ERR_MASK 0x00000001L +#define UVD_MEMCHECK2_VCPU_INT_STAT__CM_RD_HI_ERR_MASK 0x00000002L +#define UVD_MEMCHECK2_VCPU_INT_STAT__DB_RD_LO_ERR_MASK 0x00000004L +#define UVD_MEMCHECK2_VCPU_INT_STAT__DB_RD_HI_ERR_MASK 0x00000008L +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_RD_LO_ERR_MASK 0x00000010L +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_RD_HI_ERR_MASK 0x00000020L +#define UVD_MEMCHECK2_VCPU_INT_STAT__IDCT_RD_LO_ERR_MASK 0x00000040L +#define UVD_MEMCHECK2_VCPU_INT_STAT__IDCT_RD_HI_ERR_MASK 0x00000080L +#define UVD_MEMCHECK2_VCPU_INT_STAT__MPC_RD_LO_ERR_MASK 0x00000100L +#define UVD_MEMCHECK2_VCPU_INT_STAT__MPC_RD_HI_ERR_MASK 0x00000200L +#define UVD_MEMCHECK2_VCPU_INT_STAT__LBSI_RD_LO_ERR_MASK 0x00000400L +#define UVD_MEMCHECK2_VCPU_INT_STAT__LBSI_RD_HI_ERR_MASK 0x00000800L +#define UVD_MEMCHECK2_VCPU_INT_STAT__RBC_RD_LO_ERR_MASK 0x00010000L +#define UVD_MEMCHECK2_VCPU_INT_STAT__RBC_RD_HI_ERR_MASK 0x00020000L +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_BSP2_LO_ERR_MASK 0x00040000L +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_BSP2_HI_ERR_MASK 0x00080000L +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_BSP3_LO_ERR_MASK 0x00100000L +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_BSP3_HI_ERR_MASK 0x00200000L +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_SCLR_LO_ERR_MASK 0x00400000L +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_SCLR_HI_ERR_MASK 0x00800000L +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_SCLR2_LO_ERR_MASK 0x01000000L +#define UVD_MEMCHECK2_VCPU_INT_STAT__MIF_SCLR2_HI_ERR_MASK 0x02000000L +#define UVD_MEMCHECK2_VCPU_INT_STAT__PREF_LO_ERR_MASK 0x04000000L +#define UVD_MEMCHECK2_VCPU_INT_STAT__PREF_HI_ERR_MASK 0x08000000L +//UVD_MEMCHECK2_VCPU_INT_ACK +#define UVD_MEMCHECK2_VCPU_INT_ACK__CM_RD_LO_ACK__SHIFT 0x0 +#define UVD_MEMCHECK2_VCPU_INT_ACK__CM_RD_HI_ACK__SHIFT 0x1 +#define UVD_MEMCHECK2_VCPU_INT_ACK__DB_RD_LO_ACK__SHIFT 0x2 +#define UVD_MEMCHECK2_VCPU_INT_ACK__DB_RD_HI_ACK__SHIFT 0x3 +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_RD_LO_ACK__SHIFT 0x4 +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_RD_HI_ACK__SHIFT 0x5 +#define UVD_MEMCHECK2_VCPU_INT_ACK__IDCT_RD_LO_ACK__SHIFT 0x6 +#define UVD_MEMCHECK2_VCPU_INT_ACK__IDCT_RD_HI_ACK__SHIFT 0x7 +#define UVD_MEMCHECK2_VCPU_INT_ACK__MPC_RD_LO_ACK__SHIFT 0x8 +#define UVD_MEMCHECK2_VCPU_INT_ACK__MPC_RD_HI_ACK__SHIFT 0x9 +#define UVD_MEMCHECK2_VCPU_INT_ACK__LBSI_RD_LO_ACK__SHIFT 0xa +#define UVD_MEMCHECK2_VCPU_INT_ACK__LBSI_RD_HI_ACK__SHIFT 0xb +#define UVD_MEMCHECK2_VCPU_INT_ACK__RBC_RD_LO_ACK__SHIFT 0x10 +#define UVD_MEMCHECK2_VCPU_INT_ACK__RBC_RD_HI_ACK__SHIFT 0x11 +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_BSP2_LO_ACK__SHIFT 0x12 +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_BSP2_HI_ACK__SHIFT 0x13 +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_BSP3_LO_ACK__SHIFT 0x14 +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_BSP3_HI_ACK__SHIFT 0x15 +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_SCLR_LO_ACK__SHIFT 0x16 +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_SCLR_HI_ACK__SHIFT 0x17 +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_SCLR2_LO_ACK__SHIFT 0x18 +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_SCLR2_HI_ACK__SHIFT 0x19 +#define UVD_MEMCHECK2_VCPU_INT_ACK__PREF_LO_ACK__SHIFT 0x1a +#define UVD_MEMCHECK2_VCPU_INT_ACK__PREF_HI_ACK__SHIFT 0x1b +#define UVD_MEMCHECK2_VCPU_INT_ACK__CM_RD_LO_ACK_MASK 0x00000001L +#define UVD_MEMCHECK2_VCPU_INT_ACK__CM_RD_HI_ACK_MASK 0x00000002L +#define UVD_MEMCHECK2_VCPU_INT_ACK__DB_RD_LO_ACK_MASK 0x00000004L +#define UVD_MEMCHECK2_VCPU_INT_ACK__DB_RD_HI_ACK_MASK 0x00000008L +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_RD_LO_ACK_MASK 0x00000010L +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_RD_HI_ACK_MASK 0x00000020L +#define UVD_MEMCHECK2_VCPU_INT_ACK__IDCT_RD_LO_ACK_MASK 0x00000040L +#define UVD_MEMCHECK2_VCPU_INT_ACK__IDCT_RD_HI_ACK_MASK 0x00000080L +#define UVD_MEMCHECK2_VCPU_INT_ACK__MPC_RD_LO_ACK_MASK 0x00000100L +#define UVD_MEMCHECK2_VCPU_INT_ACK__MPC_RD_HI_ACK_MASK 0x00000200L +#define UVD_MEMCHECK2_VCPU_INT_ACK__LBSI_RD_LO_ACK_MASK 0x00000400L +#define UVD_MEMCHECK2_VCPU_INT_ACK__LBSI_RD_HI_ACK_MASK 0x00000800L +#define UVD_MEMCHECK2_VCPU_INT_ACK__RBC_RD_LO_ACK_MASK 0x00010000L +#define UVD_MEMCHECK2_VCPU_INT_ACK__RBC_RD_HI_ACK_MASK 0x00020000L +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_BSP2_LO_ACK_MASK 0x00040000L +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_BSP2_HI_ACK_MASK 0x00080000L +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_BSP3_LO_ACK_MASK 0x00100000L +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_BSP3_HI_ACK_MASK 0x00200000L +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_SCLR_LO_ACK_MASK 0x00400000L +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_SCLR_HI_ACK_MASK 0x00800000L +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_SCLR2_LO_ACK_MASK 0x01000000L +#define UVD_MEMCHECK2_VCPU_INT_ACK__MIF_SCLR2_HI_ACK_MASK 0x02000000L +#define UVD_MEMCHECK2_VCPU_INT_ACK__PREF_LO_ACK_MASK 0x04000000L +#define UVD_MEMCHECK2_VCPU_INT_ACK__PREF_HI_ACK_MASK 0x08000000L + + +#endif diff --git a/drivers/gpu/drm/amd/include/atomfirmware.h b/drivers/gpu/drm/amd/include/atomfirmware.h index bbe1337a8cee3ce80f01d9021a20bcb61d303903..e68c1e280322ace66fd825d6350cf463bf7e6f22 100644 --- a/drivers/gpu/drm/amd/include/atomfirmware.h +++ b/drivers/gpu/drm/amd/include/atomfirmware.h @@ -182,6 +182,7 @@ enum atom_dgpu_vram_type { ATOM_DGPU_VRAM_TYPE_HBM2 = 0x60, ATOM_DGPU_VRAM_TYPE_HBM2E = 0x61, ATOM_DGPU_VRAM_TYPE_GDDR6 = 0x70, + ATOM_DGPU_VRAM_TYPE_HBM3 = 0x80, }; enum atom_dp_vs_preemph_def{ diff --git a/drivers/gpu/drm/amd/include/discovery.h b/drivers/gpu/drm/amd/include/discovery.h index f150404ffc68207b82b68e51bf4b6f90bd38100a..f43e29722ef7104b1ef2a1370ac64f02a508d764 100644 --- a/drivers/gpu/drm/amd/include/discovery.h +++ b/drivers/gpu/drm/amd/include/discovery.h @@ -79,7 +79,14 @@ typedef struct ip_discovery_header uint32_t id; /* Table ID */ uint16_t num_dies; /* Number of Dies */ die_info die_info[16]; /* list die information for up to 16 dies */ - uint16_t padding[1]; /* padding */ + union { + uint16_t padding[1]; /* version <= 3 */ + struct { /* version == 4 */ + uint8_t base_addr_64_bit : 1; /* ip structures are using 64 bit base address */ + uint8_t reserved : 7; + uint8_t reserved2; + }; + }; } ip_discovery_header; typedef struct ip @@ -115,9 +122,29 @@ typedef struct ip_v3 uint8_t sub_revision : 4; /* HCID Sub-Revision */ uint8_t variant : 4; /* HW variant */ #endif - uint32_t base_address[1]; /* Base Address list. Corresponds to the num_base_address field*/ + uint32_t base_address[]; /* Base Address list. Corresponds to the num_base_address field*/ } ip_v3; +typedef struct ip_v4 { + uint16_t hw_id; /* Hardware ID */ + uint8_t instance_number; /* Instance number for the IP */ + uint8_t num_base_address; /* Number of base addresses*/ + uint8_t major; /* Hardware ID.major version */ + uint8_t minor; /* Hardware ID.minor version */ + uint8_t revision; /* Hardware ID.revision version */ +#if defined(LITTLEENDIAN_CPU) + uint8_t sub_revision : 4; /* HCID Sub-Revision */ + uint8_t variant : 4; /* HW variant */ +#elif defined(BIGENDIAN_CPU) + uint8_t variant : 4; /* HW variant */ + uint8_t sub_revision : 4; /* HCID Sub-Revision */ +#endif + union { + DECLARE_FLEX_ARRAY(uint32_t, base_address); /* 32-bit Base Address list. Corresponds to the num_base_address field*/ + DECLARE_FLEX_ARRAY(uint64_t, base_address_64); /* 64-bit Base Address list. Corresponds to the num_base_address field*/ + } __packed; +} ip_v4; + typedef struct die_header { uint16_t die_id; @@ -134,6 +161,7 @@ typedef struct ip_structure { ip *ip_list; ip_v3 *ip_v3_list; + ip_v4 *ip_v4_list; }; /* IP list. Variable size*/ } die; } ip_structure; diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h index 5cb3e8634739dc2d475ad025041ec62db65044cf..d0df3381539f0a6452189c74697ede5bfd1d55b5 100644 --- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h @@ -230,28 +230,30 @@ struct kfd2kgd_calls { /* Register access functions */ void (*program_sh_mem_settings)(struct amdgpu_device *adev, uint32_t vmid, uint32_t sh_mem_config, uint32_t sh_mem_ape1_base, - uint32_t sh_mem_ape1_limit, uint32_t sh_mem_bases); + uint32_t sh_mem_ape1_limit, uint32_t sh_mem_bases, + uint32_t inst); int (*set_pasid_vmid_mapping)(struct amdgpu_device *adev, u32 pasid, - unsigned int vmid); + unsigned int vmid, uint32_t inst); - int (*init_interrupts)(struct amdgpu_device *adev, uint32_t pipe_id); + int (*init_interrupts)(struct amdgpu_device *adev, uint32_t pipe_id, + uint32_t inst); int (*hqd_load)(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, uint32_t queue_id, uint32_t __user *wptr, uint32_t wptr_shift, uint32_t wptr_mask, - struct mm_struct *mm); + struct mm_struct *mm, uint32_t inst); int (*hiq_mqd_load)(struct amdgpu_device *adev, void *mqd, uint32_t pipe_id, uint32_t queue_id, - uint32_t doorbell_off); + uint32_t doorbell_off, uint32_t inst); int (*hqd_sdma_load)(struct amdgpu_device *adev, void *mqd, uint32_t __user *wptr, struct mm_struct *mm); int (*hqd_dump)(struct amdgpu_device *adev, uint32_t pipe_id, uint32_t queue_id, - uint32_t (**dump)[2], uint32_t *n_regs); + uint32_t (**dump)[2], uint32_t *n_regs, uint32_t inst); int (*hqd_sdma_dump)(struct amdgpu_device *adev, uint32_t engine_id, uint32_t queue_id, @@ -259,12 +261,12 @@ struct kfd2kgd_calls { bool (*hqd_is_occupied)(struct amdgpu_device *adev, uint64_t queue_address, uint32_t pipe_id, - uint32_t queue_id); + uint32_t queue_id, uint32_t inst); int (*hqd_destroy)(struct amdgpu_device *adev, void *mqd, enum kfd_preempt_type reset_type, unsigned int timeout, uint32_t pipe_id, - uint32_t queue_id); + uint32_t queue_id, uint32_t inst); bool (*hqd_sdma_is_occupied)(struct amdgpu_device *adev, void *mqd); @@ -273,7 +275,7 @@ struct kfd2kgd_calls { int (*wave_control_execute)(struct amdgpu_device *adev, uint32_t gfx_index_val, - uint32_t sq_cmd); + uint32_t sq_cmd, uint32_t inst); bool (*get_atc_vmid_pasid_mapping_info)(struct amdgpu_device *adev, uint8_t vmid, uint16_t *p_pasid); @@ -289,10 +291,45 @@ struct kfd2kgd_calls { uint32_t vmid, uint64_t page_table_base); uint32_t (*read_vmid_from_vmfault_reg)(struct amdgpu_device *adev); + uint32_t (*enable_debug_trap)(struct amdgpu_device *adev, + bool restore_dbg_registers, + uint32_t vmid); + uint32_t (*disable_debug_trap)(struct amdgpu_device *adev, + bool keep_trap_enabled, + uint32_t vmid); + int (*validate_trap_override_request)(struct amdgpu_device *adev, + uint32_t trap_override, + uint32_t *trap_mask_supported); + uint32_t (*set_wave_launch_trap_override)(struct amdgpu_device *adev, + uint32_t vmid, + uint32_t trap_override, + uint32_t trap_mask_bits, + uint32_t trap_mask_request, + uint32_t *trap_mask_prev, + uint32_t kfd_dbg_trap_cntl_prev); + uint32_t (*set_wave_launch_mode)(struct amdgpu_device *adev, + uint8_t wave_launch_mode, + uint32_t vmid); + uint32_t (*set_address_watch)(struct amdgpu_device *adev, + uint64_t watch_address, + uint32_t watch_address_mask, + uint32_t watch_id, + uint32_t watch_mode, + uint32_t debug_vmid); + uint32_t (*clear_address_watch)(struct amdgpu_device *adev, + uint32_t watch_id); + void (*get_iq_wait_times)(struct amdgpu_device *adev, + uint32_t *wait_times); + void (*build_grace_period_packet_info)(struct amdgpu_device *adev, + uint32_t wait_times, + uint32_t grace_period, + uint32_t *reg_offset, + uint32_t *reg_data); void (*get_cu_occupancy)(struct amdgpu_device *adev, int pasid, - int *wave_cnt, int *max_waves_per_cu); + int *wave_cnt, int *max_waves_per_cu, uint32_t inst); void (*program_trap_handler_settings)(struct amdgpu_device *adev, - uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr); + uint32_t vmid, uint64_t tba_addr, uint64_t tma_addr, + uint32_t inst); }; #endif /* KGD_KFD_INTERFACE_H_INCLUDED */ diff --git a/drivers/gpu/drm/amd/include/mes_v11_api_def.h b/drivers/gpu/drm/amd/include/mes_v11_api_def.h index dc694cb246d9f457a4e82f6446f5879b3487ada8..0997e999416a48ec905379083fb4b2b0ba7b7f76 100644 --- a/drivers/gpu/drm/amd/include/mes_v11_api_def.h +++ b/drivers/gpu/drm/amd/include/mes_v11_api_def.h @@ -274,7 +274,8 @@ union MESAPI__ADD_QUEUE { uint32_t is_kfd_process : 1; uint32_t trap_en : 1; uint32_t is_aql_queue : 1; - uint32_t reserved : 20; + uint32_t skip_process_ctx_clear : 1; + uint32_t reserved : 19; }; struct MES_API_STATUS api_status; uint64_t tma_addr; @@ -523,6 +524,7 @@ enum MESAPI_MISC_OPCODE { MESAPI_MISC__QUERY_STATUS, MESAPI_MISC__READ_REG, MESAPI_MISC__WAIT_REG_MEM, + MESAPI_MISC__SET_SHADER_DEBUGGER, MESAPI_MISC__MAX, }; @@ -561,6 +563,21 @@ struct QUERY_STATUS { uint32_t context_id; }; +struct SET_SHADER_DEBUGGER { + uint64_t process_context_addr; + union { + struct { + uint32_t single_memop : 1; /* SQ_DEBUG.single_memop */ + uint32_t single_alu_op : 1; /* SQ_DEBUG.single_alu_op */ + uint32_t reserved : 30; + }; + uint32_t u32all; + } flags; + uint32_t spi_gdbg_per_vmid_cntl; + uint32_t tcp_watch_cntl[4]; /* TCP_WATCHx_CNTL */ + uint32_t trap_en; +}; + union MESAPI__MISC { struct { union MES_API_HEADER header; @@ -573,6 +590,9 @@ union MESAPI__MISC { struct QUERY_STATUS query_status; struct READ_REG read_reg; struct WAIT_REG_MEM wait_reg_mem; + struct SET_SHADER_DEBUGGER set_shader_debugger; + enum MES_AMD_PRIORITY_LEVEL queue_sch_level; + uint32_t data[MISC_DATA_MAX_SIZE_IN_DWORDS]; }; }; diff --git a/drivers/gpu/drm/amd/include/v9_structs.h b/drivers/gpu/drm/amd/include/v9_structs.h index a0c672889fe43b0e93443db8fe29f39e2c2602c8..a2f81b9c38afcebbd501b60b17c762afd8584e78 100644 --- a/drivers/gpu/drm/amd/include/v9_structs.h +++ b/drivers/gpu/drm/amd/include/v9_structs.h @@ -196,10 +196,20 @@ struct v9_mqd { uint32_t compute_wave_restore_addr_lo; uint32_t compute_wave_restore_addr_hi; uint32_t compute_wave_restore_control; - uint32_t compute_static_thread_mgmt_se4; - uint32_t compute_static_thread_mgmt_se5; - uint32_t compute_static_thread_mgmt_se6; - uint32_t compute_static_thread_mgmt_se7; + union { + struct { + uint32_t compute_static_thread_mgmt_se4; + uint32_t compute_static_thread_mgmt_se5; + uint32_t compute_static_thread_mgmt_se6; + uint32_t compute_static_thread_mgmt_se7; + }; + struct { + uint32_t compute_current_logic_xcc_id; // offset: 39 (0x27) + uint32_t compute_restart_cg_tg_id; // offset: 40 (0x28) + uint32_t compute_tg_chunk_size; // offset: 41 (0x29) + uint32_t compute_restore_tg_chunk_size; // offset: 42 (0x2A) + }; + }; uint32_t reserved_43; uint32_t reserved_44; uint32_t reserved_45; @@ -382,8 +392,16 @@ struct v9_mqd { uint32_t iqtimer_pkt_dw29; uint32_t iqtimer_pkt_dw30; uint32_t iqtimer_pkt_dw31; - uint32_t reserved_225; - uint32_t reserved_226; + union { + struct { + uint32_t reserved_225; + uint32_t reserved_226; + }; + struct { + uint32_t pm4_target_xcc_in_xcp; // offset: 225 (0xE1) + uint32_t cp_mqd_stride_size; // offset: 226 (0xE2) + }; + }; uint32_t reserved_227; uint32_t set_resources_header; uint32_t set_resources_dw1; diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c index 58c2246918fdabd1de4d9f91262c0ac20b2bc587..a57952b93e73fa727e45f1d6e6a8afd9346f1c1a 100644 --- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c @@ -678,7 +678,12 @@ static ssize_t amdgpu_set_pp_table(struct device *dev, * clock labeled OD_MCLK * * - three points labeled OD_VDDC_CURVE. - * They can be used to calibrate the sclk voltage curve. + * They can be used to calibrate the sclk voltage curve. This is + * available for Vega20 and NV1X. + * + * - voltage offset for the six anchor points of the v/f curve labeled + * OD_VDDC_CURVE. They can be used to calibrate the v/f curve. This + * is only availabe for some SMU13 ASICs. * * - voltage offset(in mV) applied on target voltage calculation. * This is available for Sienna Cichlid, Navy Flounder and Dimgrey @@ -719,12 +724,19 @@ static ssize_t amdgpu_set_pp_table(struct device *dev, * E.g., "p 2 0 800" would set the minimum core clock on core * 2 to 800Mhz. * - * For sclk voltage curve, enter the new values by writing a - * string that contains "vc point clock voltage" to the file. The - * points are indexed by 0, 1 and 2. E.g., "vc 0 300 600" will - * update point1 with clock set as 300Mhz and voltage as - * 600mV. "vc 2 1000 1000" will update point3 with clock set - * as 1000Mhz and voltage 1000mV. + * For sclk voltage curve, + * - For NV1X, enter the new values by writing a string that + * contains "vc point clock voltage" to the file. The points + * are indexed by 0, 1 and 2. E.g., "vc 0 300 600" will update + * point1 with clock set as 300Mhz and voltage as 600mV. "vc 2 + * 1000 1000" will update point3 with clock set as 1000Mhz and + * voltage 1000mV. + * - For SMU13 ASICs, enter the new values by writing a string that + * contains "vc anchor_point_index voltage_offset" to the file. + * There are total six anchor points defined on the v/f curve with + * index as 0 - 5. + * - "vc 0 10" will update the voltage offset for point1 as 10mv. + * - "vc 5 -10" will update the voltage offset for point6 as -10mv. * * To update the voltage offset applied for gfxclk/voltage calculation, * enter the new value by writing a string that contains "vo offset". @@ -871,13 +883,11 @@ static ssize_t amdgpu_get_pp_od_clk_voltage(struct device *dev, } if (ret == -ENOENT) { size = amdgpu_dpm_print_clock_levels(adev, OD_SCLK, buf); - if (size > 0) { - size += amdgpu_dpm_print_clock_levels(adev, OD_MCLK, buf + size); - size += amdgpu_dpm_print_clock_levels(adev, OD_VDDC_CURVE, buf + size); - size += amdgpu_dpm_print_clock_levels(adev, OD_VDDGFX_OFFSET, buf + size); - size += amdgpu_dpm_print_clock_levels(adev, OD_RANGE, buf + size); - size += amdgpu_dpm_print_clock_levels(adev, OD_CCLK, buf + size); - } + size += amdgpu_dpm_print_clock_levels(adev, OD_MCLK, buf + size); + size += amdgpu_dpm_print_clock_levels(adev, OD_VDDC_CURVE, buf + size); + size += amdgpu_dpm_print_clock_levels(adev, OD_VDDGFX_OFFSET, buf + size); + size += amdgpu_dpm_print_clock_levels(adev, OD_RANGE, buf + size); + size += amdgpu_dpm_print_clock_levels(adev, OD_CCLK, buf + size); } if (size == 0) @@ -3362,7 +3372,8 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj, return 0; /* Skip crit temp on APU */ - if ((adev->flags & AMD_IS_APU) && (adev->family >= AMDGPU_FAMILY_CZ) && + if ((((adev->flags & AMD_IS_APU) && (adev->family >= AMDGPU_FAMILY_CZ)) || + (gc_ver == IP_VERSION(9, 4, 3))) && (attr == &sensor_dev_attr_temp1_crit.dev_attr.attr || attr == &sensor_dev_attr_temp1_crit_hyst.dev_attr.attr)) return 0; @@ -3395,9 +3406,10 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj, attr == &sensor_dev_attr_pwm1_enable.dev_attr.attr)) /* can't manage state */ effective_mode &= ~S_IWUSR; - /* In the case of APUs, this is only implemented on Vangogh */ + /* not implemented yet for APUs other than GC 10.3.1 (vangogh) and 9.4.3 */ if (((adev->family == AMDGPU_FAMILY_SI) || - ((adev->flags & AMD_IS_APU) && (gc_ver != IP_VERSION(10, 3, 1)))) && + ((adev->flags & AMD_IS_APU) && (gc_ver != IP_VERSION(10, 3, 1)) && + (gc_ver != IP_VERSION(9, 4, 3)))) && (attr == &sensor_dev_attr_power1_cap_max.dev_attr.attr || attr == &sensor_dev_attr_power1_cap_min.dev_attr.attr || attr == &sensor_dev_attr_power1_cap.dev_attr.attr || @@ -3426,36 +3438,48 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj, return 0; if ((adev->family == AMDGPU_FAMILY_SI || /* not implemented yet */ - adev->family == AMDGPU_FAMILY_KV) && /* not implemented yet */ + adev->family == AMDGPU_FAMILY_KV || /* not implemented yet */ + (gc_ver == IP_VERSION(9, 4, 3))) && (attr == &sensor_dev_attr_in0_input.dev_attr.attr || attr == &sensor_dev_attr_in0_label.dev_attr.attr)) return 0; - /* only APUs have vddnb */ - if (!(adev->flags & AMD_IS_APU) && + /* only APUs other than gc 9,4,3 have vddnb */ + if ((!(adev->flags & AMD_IS_APU) || (gc_ver == IP_VERSION(9, 4, 3))) && (attr == &sensor_dev_attr_in1_input.dev_attr.attr || attr == &sensor_dev_attr_in1_label.dev_attr.attr)) return 0; - /* no mclk on APUs */ - if ((adev->flags & AMD_IS_APU) && + /* no mclk on APUs other than gc 9,4,3*/ + if (((adev->flags & AMD_IS_APU) && (gc_ver != IP_VERSION(9, 4, 3))) && (attr == &sensor_dev_attr_freq2_input.dev_attr.attr || attr == &sensor_dev_attr_freq2_label.dev_attr.attr)) return 0; - /* only SOC15 dGPUs support hotspot and mem temperatures */ if (((adev->flags & AMD_IS_APU) || gc_ver < IP_VERSION(9, 0, 0)) && + (gc_ver != IP_VERSION(9, 4, 3)) && + (attr == &sensor_dev_attr_temp2_input.dev_attr.attr || + attr == &sensor_dev_attr_temp2_label.dev_attr.attr || + attr == &sensor_dev_attr_temp3_input.dev_attr.attr || + attr == &sensor_dev_attr_temp3_label.dev_attr.attr)) + return 0; + + /* hotspot temperature for gc 9,4,3*/ + if ((gc_ver == IP_VERSION(9, 4, 3)) && + (attr == &sensor_dev_attr_temp1_input.dev_attr.attr || + attr == &sensor_dev_attr_temp1_label.dev_attr.attr)) + return 0; + + /* only SOC15 dGPUs support hotspot and mem temperatures */ + if (((adev->flags & AMD_IS_APU) || gc_ver < IP_VERSION(9, 0, 0) || + (gc_ver == IP_VERSION(9, 4, 3))) && (attr == &sensor_dev_attr_temp2_crit.dev_attr.attr || attr == &sensor_dev_attr_temp2_crit_hyst.dev_attr.attr || attr == &sensor_dev_attr_temp3_crit.dev_attr.attr || attr == &sensor_dev_attr_temp3_crit_hyst.dev_attr.attr || attr == &sensor_dev_attr_temp1_emergency.dev_attr.attr || attr == &sensor_dev_attr_temp2_emergency.dev_attr.attr || - attr == &sensor_dev_attr_temp3_emergency.dev_attr.attr || - attr == &sensor_dev_attr_temp2_input.dev_attr.attr || - attr == &sensor_dev_attr_temp3_input.dev_attr.attr || - attr == &sensor_dev_attr_temp2_label.dev_attr.attr || - attr == &sensor_dev_attr_temp3_label.dev_attr.attr)) + attr == &sensor_dev_attr_temp3_emergency.dev_attr.attr)) return 0; /* only Vangogh has fast PPT limit and power labels */ diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v13_0_0_pptable.h b/drivers/gpu/drm/amd/pm/inc/smu_v13_0_0_pptable.h index 566a0da59e530d6e7e1a0dfa69be54fc65a2cc88..1dc7a065a6d484b37cf285c9b44d5d82b8dc83aa 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_v13_0_0_pptable.h +++ b/drivers/gpu/drm/amd/pm/inc/smu_v13_0_0_pptable.h @@ -38,13 +38,12 @@ #define SMU_13_0_0_PP_THERMALCONTROLLER_NONE 0 #define SMU_13_0_0_PP_THERMALCONTROLLER_NAVI21 28 -#define SMU_13_0_0_PP_OVERDRIVE_VERSION 0x81 // OverDrive 8 Table Version 0.2 +#define SMU_13_0_0_PP_OVERDRIVE_VERSION 0x83 // OverDrive 8 Table Version 0.2 #define SMU_13_0_0_PP_POWERSAVINGCLOCK_VERSION 0x01 // Power Saving Clock Table Version 1.00 enum SMU_13_0_0_ODFEATURE_CAP { SMU_13_0_0_ODCAP_GFXCLK_LIMITS = 0, - SMU_13_0_0_ODCAP_GFXCLK_CURVE, SMU_13_0_0_ODCAP_UCLK_LIMITS, SMU_13_0_0_ODCAP_POWER_LIMIT, SMU_13_0_0_ODCAP_FAN_ACOUSTIC_LIMIT, @@ -59,13 +58,13 @@ enum SMU_13_0_0_ODFEATURE_CAP SMU_13_0_0_ODCAP_FAN_CURVE, SMU_13_0_0_ODCAP_AUTO_FAN_ACOUSTIC_LIMIT, SMU_13_0_0_ODCAP_POWER_MODE, + SMU_13_0_0_ODCAP_PER_ZONE_GFX_VOLTAGE_OFFSET, SMU_13_0_0_ODCAP_COUNT, }; enum SMU_13_0_0_ODFEATURE_ID { SMU_13_0_0_ODFEATURE_GFXCLK_LIMITS = 1 << SMU_13_0_0_ODCAP_GFXCLK_LIMITS, //GFXCLK Limit feature - SMU_13_0_0_ODFEATURE_GFXCLK_CURVE = 1 << SMU_13_0_0_ODCAP_GFXCLK_CURVE, //GFXCLK Curve feature SMU_13_0_0_ODFEATURE_UCLK_LIMITS = 1 << SMU_13_0_0_ODCAP_UCLK_LIMITS, //UCLK Limit feature SMU_13_0_0_ODFEATURE_POWER_LIMIT = 1 << SMU_13_0_0_ODCAP_POWER_LIMIT, //Power Limit feature SMU_13_0_0_ODFEATURE_FAN_ACOUSTIC_LIMIT = 1 << SMU_13_0_0_ODCAP_FAN_ACOUSTIC_LIMIT, //Fan Acoustic RPM feature @@ -80,6 +79,7 @@ enum SMU_13_0_0_ODFEATURE_ID SMU_13_0_0_ODFEATURE_FAN_CURVE = 1 << SMU_13_0_0_ODCAP_FAN_CURVE, //Fan Curve feature SMU_13_0_0_ODFEATURE_AUTO_FAN_ACOUSTIC_LIMIT = 1 << SMU_13_0_0_ODCAP_AUTO_FAN_ACOUSTIC_LIMIT, //Auto Fan Acoustic RPM feature SMU_13_0_0_ODFEATURE_POWER_MODE = 1 << SMU_13_0_0_ODCAP_POWER_MODE, //Optimized GPU Power Mode feature + SMU_13_0_0_ODFEATURE_PER_ZONE_GFX_VOLTAGE_OFFSET = 1 << SMU_13_0_0_ODCAP_PER_ZONE_GFX_VOLTAGE_OFFSET, //Perzone voltage offset feature SMU_13_0_0_ODFEATURE_COUNT = 16, }; @@ -89,10 +89,6 @@ enum SMU_13_0_0_ODSETTING_ID { SMU_13_0_0_ODSETTING_GFXCLKFMAX = 0, SMU_13_0_0_ODSETTING_GFXCLKFMIN, - SMU_13_0_0_ODSETTING_CUSTOM_GFX_VF_CURVE_A, - SMU_13_0_0_ODSETTING_CUSTOM_GFX_VF_CURVE_B, - SMU_13_0_0_ODSETTING_CUSTOM_GFX_VF_CURVE_C, - SMU_13_0_0_ODSETTING_CUSTOM_CURVE_VFT_FMIN, SMU_13_0_0_ODSETTING_UCLKFMIN, SMU_13_0_0_ODSETTING_UCLKFMAX, SMU_13_0_0_ODSETTING_POWERPERCENTAGE, @@ -117,6 +113,12 @@ enum SMU_13_0_0_ODSETTING_ID SMU_13_0_0_ODSETTING_FAN_CURVE_SPEED_5, SMU_13_0_0_ODSETTING_AUTO_FAN_ACOUSTIC_LIMIT, SMU_13_0_0_ODSETTING_POWER_MODE, + SMU_13_0_0_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_1, + SMU_13_0_0_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_2, + SMU_13_0_0_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_3, + SMU_13_0_0_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_4, + SMU_13_0_0_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_5, + SMU_13_0_0_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_6, SMU_13_0_0_ODSETTING_COUNT, }; #define SMU_13_0_0_MAX_ODSETTING 64 //Maximum Number of ODSettings diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c index f5e08b60f66ef9fa2d84bd2a86d712a839928e36..36c831b280ed9f63210e66c56de892b578693ddb 100644 --- a/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c @@ -508,19 +508,19 @@ static int kv_enable_didt(struct amdgpu_device *adev, bool enable) pi->caps_db_ramping || pi->caps_td_ramping || pi->caps_tcp_ramping) { - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); if (enable) { ret = kv_program_pt_config_registers(adev, didt_config_kv); if (ret) { - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); return ret; } } kv_do_enable_didt(adev, enable); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); } return 0; diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c index d6d9e3b1b2c0e4511707680dbb1286d13188977a..02e69ccff3bac47cefb29b4369e196bd09dfae9d 100644 --- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c @@ -6925,23 +6925,6 @@ static int si_dpm_enable(struct amdgpu_device *adev) return 0; } -static int si_set_temperature_range(struct amdgpu_device *adev) -{ - int ret; - - ret = si_thermal_enable_alert(adev, false); - if (ret) - return ret; - ret = si_thermal_set_temperature_range(adev, R600_TEMP_RANGE_MIN, R600_TEMP_RANGE_MAX); - if (ret) - return ret; - ret = si_thermal_enable_alert(adev, true); - if (ret) - return ret; - - return ret; -} - static void si_dpm_disable(struct amdgpu_device *adev) { struct rv7xx_power_info *pi = rv770_get_pi(adev); @@ -7626,18 +7609,6 @@ static int si_dpm_process_interrupt(struct amdgpu_device *adev, static int si_dpm_late_init(void *handle) { - int ret; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - if (!adev->pm.dpm_enabled) - return 0; - - ret = si_set_temperature_range(adev); - if (ret) - return ret; -#if 0 //TODO ? - si_dpm_powergate_uvd(adev, true); -#endif return 0; } diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_powertune.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_powertune.c index 32a5a00fd8ae7a963bf7b5d18b70bb86ad0ceb68..21be23ec3c79e2f446aa4139fdf965ac252c2019 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_powertune.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_powertune.c @@ -973,7 +973,7 @@ int smu7_enable_didt_config(struct pp_hwmgr *hwmgr) PP_CAP(PHM_PlatformCaps_TDRamping) || PP_CAP(PHM_PlatformCaps_TCPRamping)) { - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); mutex_lock(&adev->grbm_idx_mutex); value = 0; value2 = cgs_read_register(hwmgr->device, mmGRBM_GFX_INDEX); @@ -1048,13 +1048,13 @@ int smu7_enable_didt_config(struct pp_hwmgr *hwmgr) } mutex_unlock(&adev->grbm_idx_mutex); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); } return 0; error: mutex_unlock(&adev->grbm_idx_mutex); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); return result; } @@ -1068,7 +1068,7 @@ int smu7_disable_didt_config(struct pp_hwmgr *hwmgr) PP_CAP(PHM_PlatformCaps_TDRamping) || PP_CAP(PHM_PlatformCaps_TCPRamping)) { - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); result = smu7_enable_didt(hwmgr, false); PP_ASSERT_WITH_CODE((result == 0), @@ -1081,12 +1081,12 @@ int smu7_disable_didt_config(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE((0 == result), "Failed to disable DPM DIDT.", goto error); } - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); } return 0; error: - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); return result; } diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_powertune.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_powertune.c index 9757d47dd6b815bfe208676b435551b830fce5ff..309a9d3bc1b790f7c1564afac43c7b0688fed2b6 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_powertune.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_powertune.c @@ -915,7 +915,7 @@ static int vega10_enable_cac_driving_se_didt_config(struct pp_hwmgr *hwmgr) num_se = adev->gfx.config.max_shader_engines; - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); mutex_lock(&adev->grbm_idx_mutex); for (count = 0; count < num_se; count++) { @@ -940,7 +940,7 @@ static int vega10_enable_cac_driving_se_didt_config(struct pp_hwmgr *hwmgr) vega10_didt_set_mask(hwmgr, true); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); return 0; } @@ -949,11 +949,11 @@ static int vega10_disable_cac_driving_se_didt_config(struct pp_hwmgr *hwmgr) { struct amdgpu_device *adev = hwmgr->adev; - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); vega10_didt_set_mask(hwmgr, false); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); return 0; } @@ -966,7 +966,7 @@ static int vega10_enable_psm_gc_didt_config(struct pp_hwmgr *hwmgr) num_se = adev->gfx.config.max_shader_engines; - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); mutex_lock(&adev->grbm_idx_mutex); for (count = 0; count < num_se; count++) { @@ -985,7 +985,7 @@ static int vega10_enable_psm_gc_didt_config(struct pp_hwmgr *hwmgr) vega10_didt_set_mask(hwmgr, true); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); vega10_program_gc_didt_config_registers(hwmgr, GCDiDtDroopCtrlConfig_vega10); if (PP_CAP(PHM_PlatformCaps_GCEDC)) @@ -1002,11 +1002,11 @@ static int vega10_disable_psm_gc_didt_config(struct pp_hwmgr *hwmgr) struct amdgpu_device *adev = hwmgr->adev; uint32_t data; - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); vega10_didt_set_mask(hwmgr, false); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); if (PP_CAP(PHM_PlatformCaps_GCEDC)) { data = 0x00000000; @@ -1027,7 +1027,7 @@ static int vega10_enable_se_edc_config(struct pp_hwmgr *hwmgr) num_se = adev->gfx.config.max_shader_engines; - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); mutex_lock(&adev->grbm_idx_mutex); for (count = 0; count < num_se; count++) { @@ -1048,7 +1048,7 @@ static int vega10_enable_se_edc_config(struct pp_hwmgr *hwmgr) vega10_didt_set_mask(hwmgr, true); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); return 0; } @@ -1057,11 +1057,11 @@ static int vega10_disable_se_edc_config(struct pp_hwmgr *hwmgr) { struct amdgpu_device *adev = hwmgr->adev; - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); vega10_didt_set_mask(hwmgr, false); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); return 0; } @@ -1075,7 +1075,7 @@ static int vega10_enable_psm_gc_edc_config(struct pp_hwmgr *hwmgr) num_se = adev->gfx.config.max_shader_engines; - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); vega10_program_gc_didt_config_registers(hwmgr, AvfsPSMResetConfig_vega10); @@ -1096,7 +1096,7 @@ static int vega10_enable_psm_gc_edc_config(struct pp_hwmgr *hwmgr) vega10_didt_set_mask(hwmgr, true); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); vega10_program_gc_didt_config_registers(hwmgr, PSMGCEDCDroopCtrlConfig_vega10); @@ -1116,11 +1116,11 @@ static int vega10_disable_psm_gc_edc_config(struct pp_hwmgr *hwmgr) struct amdgpu_device *adev = hwmgr->adev; uint32_t data; - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); vega10_didt_set_mask(hwmgr, false); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); if (PP_CAP(PHM_PlatformCaps_GCEDC)) { data = 0x00000000; @@ -1138,7 +1138,7 @@ static int vega10_enable_se_edc_force_stall_config(struct pp_hwmgr *hwmgr) struct amdgpu_device *adev = hwmgr->adev; int result; - amdgpu_gfx_rlc_enter_safe_mode(adev); + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); mutex_lock(&adev->grbm_idx_mutex); WREG32_SOC15(GC, 0, mmGRBM_GFX_INDEX, 0xE0000000); @@ -1151,7 +1151,7 @@ static int vega10_enable_se_edc_force_stall_config(struct pp_hwmgr *hwmgr) vega10_didt_set_mask(hwmgr, false); - amdgpu_gfx_rlc_exit_safe_mode(adev); + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); return 0; } diff --git a/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h b/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h index 5ce433e2c16a569b89e7a3a2bb17c223d3254105..f1580a26a85061039b380e1ae0b7627a6629c9c5 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h @@ -359,7 +359,7 @@ struct pp_hwmgr_func { int (*set_ppfeature_status)(struct pp_hwmgr *hwmgr, uint64_t ppfeature_masks); int (*set_mp1_state)(struct pp_hwmgr *hwmgr, enum pp_mp1_state mp1_state); int (*asic_reset)(struct pp_hwmgr *hwmgr, enum SMU_ASIC_RESET_MODE mode); - int (*smu_i2c_bus_access)(struct pp_hwmgr *hwmgr, bool aquire); + int (*smu_i2c_bus_access)(struct pp_hwmgr *hwmgr, bool acquire); int (*set_df_cstate)(struct pp_hwmgr *hwmgr, enum pp_df_cstate state); int (*set_xgmi_pstate)(struct pp_hwmgr *hwmgr, uint32_t pstate); int (*disable_power_features_for_compute_performance)(struct pp_hwmgr *hwmgr, diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c index 5633c5797e85a4c7651301e06cd1c06eea90d6da..4dea79a0c5b5e182f178cf058406d83e5f778685 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c +++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c @@ -733,6 +733,24 @@ static int smu_late_init(void *handle) return ret; } + /* + * Explicitly notify PMFW the power mode the system in. Since + * the PMFW may boot the ASIC with a different mode. + * For those supporting ACDC switch via gpio, PMFW will + * handle the switch automatically. Driver involvement + * is unnecessary. + */ + if (!smu->dc_controlled_by_gpio) { + ret = smu_set_power_source(smu, + adev->pm.ac_power ? SMU_POWER_SOURCE_AC : + SMU_POWER_SOURCE_DC); + if (ret) { + dev_err(adev->dev, "Failed to switch to %s mode!\n", + adev->pm.ac_power ? "AC" : "DC"); + return ret; + } + } + if ((adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 1)) || (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 3))) return 0; @@ -804,11 +822,20 @@ static int smu_init_fb_allocations(struct smu_context *smu) } } + driver_table->domain = AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT; /* VRAM allocation for driver table */ for (i = 0; i < SMU_TABLE_COUNT; i++) { if (tables[i].size == 0) continue; + /* If one of the tables has VRAM domain restriction, keep it in + * VRAM + */ + if ((tables[i].domain & + (AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT)) == + AMDGPU_GEM_DOMAIN_VRAM) + driver_table->domain = AMDGPU_GEM_DOMAIN_VRAM; + if (i == SMU_TABLE_PMSTATUSLOG) continue; @@ -818,7 +845,6 @@ static int smu_init_fb_allocations(struct smu_context *smu) driver_table->size = max_table_size; driver_table->align = PAGE_SIZE; - driver_table->domain = AMDGPU_GEM_DOMAIN_VRAM; ret = amdgpu_bo_create_kernel(adev, driver_table->size, diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_aldebaran.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_aldebaran.h index 90200f31ff5269e4947f473ddbcf9d5af4adb866..cddf45eebee8df3e10100b2f2a5a04cd1adac2c5 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_aldebaran.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_aldebaran.h @@ -24,6 +24,8 @@ #ifndef SMU13_DRIVER_IF_ALDEBARAN_H #define SMU13_DRIVER_IF_ALDEBARAN_H +#define SMU13_DRIVER_IF_VERSION_ALDE 0x08 + #define NUM_VCLK_DPM_LEVELS 8 #define NUM_DCLK_DPM_LEVELS 8 #define NUM_SOCCLK_DPM_LEVELS 8 diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h index b686fb68a6e765f2c904673c1df1c38fb7aada25..9dd1ed5b8940670bc72b8263157e05bd3c3999ea 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h @@ -24,8 +24,10 @@ #ifndef SMU13_DRIVER_IF_V13_0_0_H #define SMU13_DRIVER_IF_V13_0_0_H +#define SMU13_0_0_DRIVER_IF_VERSION 0x3D + //Increment this version if SkuTable_t or BoardTable_t change -#define PPTABLE_VERSION 0x26 +#define PPTABLE_VERSION 0x2B #define NUM_GFXCLK_DPM_LEVELS 16 #define NUM_SOCCLK_DPM_LEVELS 8 @@ -94,7 +96,7 @@ #define FEATURE_ATHUB_MMHUB_PG_BIT 48 #define FEATURE_SOC_PCC_BIT 49 #define FEATURE_EDC_PWRBRK_BIT 50 -#define FEATURE_SPARE_51_BIT 51 +#define FEATURE_BOMXCO_SVI3_PROG_BIT 51 #define FEATURE_SPARE_52_BIT 52 #define FEATURE_SPARE_53_BIT 53 #define FEATURE_SPARE_54_BIT 54 @@ -310,6 +312,7 @@ typedef enum { I2C_CONTROLLER_PROTOCOL_VR_IR35217, I2C_CONTROLLER_PROTOCOL_TMP_MAX31875, I2C_CONTROLLER_PROTOCOL_INA3221, + I2C_CONTROLLER_PROTOCOL_TMP_MAX6604, I2C_CONTROLLER_PROTOCOL_COUNT, } I2cControllerProtocol_e; @@ -568,6 +571,7 @@ typedef enum { } POWER_SOURCE_e; typedef enum { + MEM_VENDOR_PLACEHOLDER0, MEM_VENDOR_SAMSUNG, MEM_VENDOR_INFINEON, MEM_VENDOR_ELPIDA, @@ -577,7 +581,6 @@ typedef enum { MEM_VENDOR_MOSEL, MEM_VENDOR_WINBOND, MEM_VENDOR_ESMT, - MEM_VENDOR_PLACEHOLDER0, MEM_VENDOR_PLACEHOLDER1, MEM_VENDOR_PLACEHOLDER2, MEM_VENDOR_PLACEHOLDER3, @@ -665,7 +668,14 @@ typedef enum { #define PP_NUM_RTAVFS_PWL_ZONES 5 - +#define PP_OD_FEATURE_GFX_VF_CURVE_BIT 0 +#define PP_OD_FEATURE_PPT_BIT 2 +#define PP_OD_FEATURE_FAN_CURVE_BIT 3 +#define PP_OD_FEATURE_GFXCLK_BIT 7 +#define PP_OD_FEATURE_UCLK_BIT 8 +#define PP_OD_FEATURE_ZERO_FAN_BIT 9 +#define PP_OD_FEATURE_TEMPERATURE_BIT 10 +#define PP_OD_FEATURE_COUNT 13 // VBIOS or PPLIB configures telemetry slope and offset. Only slope expected to be set for SVI3 // Slope Q1.7, Offset Q1.2 @@ -687,10 +697,8 @@ typedef struct { //Voltage control int16_t VoltageOffsetPerZoneBoundary[PP_NUM_OD_VF_CURVE_POINTS]; - uint16_t VddGfxVmax; // in mV - uint8_t IdlePwrSavingFeaturesCtrl; - uint8_t RuntimePwrSavingFeaturesCtrl; + uint32_t Reserved; //Frequency changes int16_t GfxclkFmin; // MHz @@ -727,10 +735,9 @@ typedef struct { uint32_t FeatureCtrlMask; int16_t VoltageOffsetPerZoneBoundary; - uint16_t VddGfxVmax; // in mV + uint16_t Reserved1; - uint8_t IdlePwrSavingFeaturesCtrl; - uint8_t RuntimePwrSavingFeaturesCtrl; + uint16_t Reserved2; int16_t GfxclkFmin; // MHz int16_t GfxclkFmax; // MHz @@ -806,6 +813,9 @@ typedef enum { #define INVALID_BOARD_GPIO 0xFF +#define MARKETING_BASE_CLOCKS 0 +#define MARKETING_GAME_CLOCKS 1 +#define MARKETING_BOOST_CLOCKS 2 typedef struct { //PLL 0 @@ -1096,10 +1106,15 @@ typedef struct { uint16_t DcsExitHysteresis; //The min amount of time power credit accumulator should have a value > 0 before SMU exits the DCS throttling phase. uint16_t DcsTimeout; //This is the amount of time SMU FW waits for RLC to put GFX into GFXOFF before reverting to the fallback mechanism of throttling GFXCLK to Fmin. + uint8_t FoptEnabled; + uint8_t DcsSpare2[3]; + uint32_t DcsFoptM; //Tuning paramters to shift Fopt calculation + uint32_t DcsFoptB; //Tuning paramters to shift Fopt calculation - uint32_t DcsSpare[16]; + uint32_t DcsSpare[11]; // UCLK section + uint16_t ShadowFreqTableUclk[NUM_UCLK_DPM_LEVELS]; // In MHz uint8_t UseStrobeModeOptimizations; //Set to indicate that FW should use strobe mode optimizations uint8_t PaddingMem[3]; @@ -1245,8 +1260,13 @@ typedef struct { QuadraticInt_t qFeffCoeffBaseClock[POWER_SOURCE_COUNT]; QuadraticInt_t qFeffCoeffBoostClock[POWER_SOURCE_COUNT]; + uint16_t TemperatureLimit_Hynix; // In degrees Celsius. Memory temperature limit associated with Hynix + uint16_t TemperatureLimit_Micron; // In degrees Celsius. Memory temperature limit associated with Micron + uint16_t TemperatureFwCtfLimit_Hynix; + uint16_t TemperatureFwCtfLimit_Micron; + // SECTION: Sku Reserved - uint32_t Spare[43]; + uint32_t Spare[41]; // Padding for MMHUB - do not modify this uint32_t MmHubPadding[8]; @@ -1318,8 +1338,9 @@ typedef struct { // UCLK Spread Spectrum uint8_t UclkSpreadPercent[MEM_VENDOR_COUNT]; + uint8_t GfxclkSpreadEnable; + // FCLK Spread Spectrum - uint8_t FclkSpreadPadding; uint8_t FclkSpreadPercent; // Q4.4 uint16_t FclkSpreadFreq; // kHz @@ -1444,6 +1465,8 @@ typedef struct { uint8_t ThrottlingPercentage[THROTTLER_COUNT]; + uint8_t VmaxThrottlingPercentage; + uint8_t Padding1[3]; //metrics for D3hot entry/exit and driver ARM msgs uint32_t D3HotEntryCountPerMode[D3HOT_SEQUENCE_COUNT]; @@ -1463,7 +1486,7 @@ typedef struct { typedef struct { SmuMetrics_t SmuMetrics; - uint32_t Spare[30]; + uint32_t Spare[29]; // Padding - ignore uint32_t MmHubPadding[8]; // SMU internal use diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_4.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_4.h index 2162ecd1057d1c40286644ac4774c0bfb30dcbf9..fee9293b3f9700b842f41bc68eba55ee72076352 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_4.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_4.h @@ -27,7 +27,7 @@ // *** IMPORTANT *** // SMU TEAM: Always increment the interface version if // any structure is changed in this file -#define PMFW_DRIVER_IF_VERSION 8 +#define SMU13_0_4_DRIVER_IF_VERSION 8 typedef struct { int32_t value; diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_5.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_5.h index aa971412b4347f556e23785cda63c30c7750eedd..7589faa0232d8c965231ff60085a97dbc2560649 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_5.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_5.h @@ -23,7 +23,7 @@ #ifndef __SMU13_DRIVER_IF_V13_0_5_H__ #define __SMU13_DRIVER_IF_V13_0_5_H__ -#define PMFW_DRIVER_IF_VERSION 4 +#define SMU13_0_5_DRIVER_IF_VERSION 4 // Throttler Status Bitmask #define THROTTLER_STATUS_BIT_SPL 0 diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h index be596777cd2caaddb9e05729a02e2933d87b1c89..ca4a5e99ccd1f69f721adc34263183dfc290c7c0 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_6.h @@ -26,7 +26,7 @@ // *** IMPORTANT *** // PMFW TEAM: Always increment the interface version if // anything is changed in this file -#define SMU13_0_6_DRIVER_IF_VERSION 0x08042022 +#define SMU13_0_6_DRIVER_IF_VERSION 0x08042024 //I2C Interface #define NUM_I2C_CONTROLLERS 8 @@ -106,7 +106,7 @@ typedef enum { } UCLK_DPM_MODE_e; typedef struct { - //0-26 SOC, 27-29 SOCIO + //0-23 SOC, 24-26 SOCIO, 27-29 SOC uint16_t avgPsmCount[30]; uint16_t minPsmCount[30]; float avgPsmVoltage[30]; @@ -121,6 +121,17 @@ typedef struct { float minPsmVoltage[30]; } AvfsDebugTableXcd_t; +// Defines used for IH-based thermal interrupts to GFX driver - A/X only +#define IH_INTERRUPT_ID_TO_DRIVER 0xFE +#define IH_INTERRUPT_CONTEXT_ID_THERMAL_THROTTLING 0x7 + +//thermal over-temp mask defines for IH interrupt to host +#define THROTTLER_PROCHOT_BIT 0 +#define THROTTLER_PPT_BIT 1 +#define THROTTLER_THERMAL_SOCKET_BIT 2//AID, XCD, CCD throttling +#define THROTTLER_THERMAL_VR_BIT 3//VRHOT +#define THROTTLER_THERMAL_HBM_BIT 4 + // These defines are used with the following messages: // SMC_MSG_TransferTableDram2Smu // SMC_MSG_TransferTableSmu2Dram diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_7.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_7.h index 4c46a039245122d3673603089c6f38ef997d1488..62b7c0daff6873e062b27ea181c1990db78cb099 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_7.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_7.h @@ -25,7 +25,7 @@ // *** IMPORTANT *** // PMFW TEAM: Always increment the interface version on any change to this file -#define SMU13_DRIVER_IF_VERSION 0x35 +#define SMU13_0_7_DRIVER_IF_VERSION 0x35 //Increment this version if SkuTable_t or BoardTable_t change #define PPTABLE_VERSION 0x27 @@ -683,18 +683,12 @@ typedef struct { #define PP_OD_FEATURE_GFX_VF_CURVE_BIT 0 -#define PP_OD_FEATURE_VMAX_BIT 1 #define PP_OD_FEATURE_PPT_BIT 2 #define PP_OD_FEATURE_FAN_CURVE_BIT 3 -#define PP_OD_FEATURE_FREQ_DETER_BIT 4 -#define PP_OD_FEATURE_FULL_CTRL_BIT 5 -#define PP_OD_FEATURE_TDC_BIT 6 #define PP_OD_FEATURE_GFXCLK_BIT 7 #define PP_OD_FEATURE_UCLK_BIT 8 #define PP_OD_FEATURE_ZERO_FAN_BIT 9 #define PP_OD_FEATURE_TEMPERATURE_BIT 10 -#define PP_OD_FEATURE_POWER_FEATURE_CTRL_BIT 11 -#define PP_OD_FEATURE_ASIC_TDC_BIT 12 #define PP_OD_FEATURE_COUNT 13 typedef enum { @@ -713,10 +707,8 @@ typedef struct { //Voltage control int16_t VoltageOffsetPerZoneBoundary[PP_NUM_OD_VF_CURVE_POINTS]; - uint16_t VddGfxVmax; // in mV - uint8_t IdlePwrSavingFeaturesCtrl; - uint8_t RuntimePwrSavingFeaturesCtrl; + uint32_t Reserved; //Frequency changes int16_t GfxclkFmin; // MHz @@ -741,12 +733,7 @@ typedef struct { uint8_t MaxOpTemp; uint8_t Padding[4]; - uint16_t GfxVoltageFullCtrlMode; - uint16_t GfxclkFullCtrlMode; - uint16_t UclkFullCtrlMode; - int16_t AsicTdc; - - uint32_t Spare[10]; + uint32_t Spare[12]; uint32_t MmHubPadding[8]; // SMU internal use. Adding here instead of external as a workaround } OverDriveTable_t; @@ -759,10 +746,9 @@ typedef struct { uint32_t FeatureCtrlMask; int16_t VoltageOffsetPerZoneBoundary; - uint16_t VddGfxVmax; // in mV + uint16_t Reserved1; - uint8_t IdlePwrSavingFeaturesCtrl; - uint8_t RuntimePwrSavingFeaturesCtrl; + uint16_t Reserved2; int16_t GfxclkFmin; // MHz int16_t GfxclkFmax; // MHz @@ -785,12 +771,7 @@ typedef struct { uint8_t MaxOpTemp; uint8_t Padding[4]; - uint16_t GfxVoltageFullCtrlMode; - uint16_t GfxclkFullCtrlMode; - uint16_t UclkFullCtrlMode; - int16_t AsicTdc; - - uint32_t Spare[10]; + uint32_t Spare[12]; } OverDriveLimits_t; diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_yellow_carp.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_yellow_carp.h index 25540cb282088fb077c287a06aed951fa561d40b..7417634827ad65f09ce5523ca0f8abc60ecbf341 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_yellow_carp.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_yellow_carp.h @@ -26,7 +26,7 @@ // *** IMPORTANT *** // SMU TEAM: Always increment the interface version if // any structure is changed in this file -#define SMU13_DRIVER_IF_VERSION 4 +#define SMU13_YELLOW_CARP_DRIVER_IF_VERSION 4 typedef struct { int32_t value; diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h index bdccbb4a62763fea4493d9f7f60e6927a76fc4aa..252aef190c5c2cf15232accaad6ed94900726de6 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h @@ -123,9 +123,9 @@ typedef enum { VOLTAGE_GUARDBAND_COUNT } GFX_GUARDBAND_e; -#define SMU_METRICS_TABLE_VERSION 0x1 +#define SMU_METRICS_TABLE_VERSION 0x5 -typedef struct { +typedef struct __attribute__((packed, aligned(4))) { uint32_t AccumulationCounter; //TEMPERATURE @@ -198,11 +198,20 @@ typedef struct { uint32_t SocketThmResidencyAcc; uint32_t VrThmResidencyAcc; uint32_t HbmThmResidencyAcc; + uint32_t spare; + + // New Items at end to maintain driver compatibility + uint32_t GfxclkFrequency[8]; + + //PSNs + uint64_t PublicSerialNumber_AID[4]; + uint64_t PublicSerialNumber_XCD[8]; + uint64_t PublicSerialNumber_CCD[12]; } MetricsTable_t; -#define SMU_VF_METRICS_TABLE_VERSION 0x1 +#define SMU_VF_METRICS_TABLE_VERSION 0x3 -typedef struct { +typedef struct __attribute__((packed, aligned(4))) { uint32_t AccumulationCounter; uint32_t InstGfxclk_TargFreq; uint64_t AccGfxclk_TargFreq; diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h index b838e8db395acb068af402d5bd8724331b0bfc09..ae4f44c4b87714b54c4b738dfa45fed7b43846ce 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h @@ -82,7 +82,8 @@ #define PPSMC_MSG_SetSoftMaxGfxClk 0x31 #define PPSMC_MSG_GetMinGfxDpmFreq 0x32 #define PPSMC_MSG_GetMaxGfxDpmFreq 0x33 -#define PPSMC_Message_Count 0x34 +#define PPSMC_MSG_PrepareForDriverUnload 0x34 +#define PPSMC_Message_Count 0x35 //PPSMC Reset Types for driver msg argument #define PPSMC_RESET_TYPE_DRIVER_MODE_1_RESET 0x1 diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h index df3baaab0037813b248991b100404d1a072a8c91..6a0ac0bbaace995764351cd19add1b25367cfa7d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h @@ -25,17 +25,6 @@ #include "amdgpu_smu.h" -#define SMU13_DRIVER_IF_VERSION_INV 0xFFFFFFFF -#define SMU13_DRIVER_IF_VERSION_YELLOW_CARP 0x04 -#define SMU13_DRIVER_IF_VERSION_ALDE 0x08 -#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_0_0 0x37 -#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_4 0x08 -#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_5 0x04 -#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_0_10 0x32 -#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_7 0x37 -#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_10 0x1D -#define SMU13_DRIVER_IF_VERSION_SMU_V13_0_6 0x0 - #define SMU13_MODE1_RESET_WAIT_TIME_IN_MS 500 //500ms /* MP Apertures */ @@ -62,6 +51,8 @@ #define CTF_OFFSET_HOTSPOT 5 #define CTF_OFFSET_MEM 5 +#define SMU_13_VCLK_SHIFT 16 + extern const int pmfw_decoded_link_speed[5]; extern const int pmfw_decoded_link_width[7]; @@ -130,6 +121,7 @@ struct smu_13_0_power_context { uint32_t power_source; uint8_t in_power_limit_boost_mode; enum smu_13_0_power_state power_state; + atomic_t throttle_status; }; #if defined(SWSMU_CODE_LAYER_L2) || defined(SWSMU_CODE_LAYER_L3) diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0_7_pptable.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0_7_pptable.h index 478862ded0bdebbf9ac3b422e93202bcaffd9062..eadbe0149cae558be12588a8528d6f5a351a7673 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0_7_pptable.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0_7_pptable.h @@ -38,13 +38,12 @@ #define SMU_13_0_7_PP_THERMALCONTROLLER_NONE 0 #define SMU_13_0_7_PP_THERMALCONTROLLER_NAVI21 28 -#define SMU_13_0_7_PP_OVERDRIVE_VERSION 0x81 // OverDrive 8 Table Version 0.2 +#define SMU_13_0_7_PP_OVERDRIVE_VERSION 0x83 // OverDrive 8 Table Version 0.2 #define SMU_13_0_7_PP_POWERSAVINGCLOCK_VERSION 0x01 // Power Saving Clock Table Version 1.00 enum SMU_13_0_7_ODFEATURE_CAP { SMU_13_0_7_ODCAP_GFXCLK_LIMITS = 0, - SMU_13_0_7_ODCAP_GFXCLK_CURVE, SMU_13_0_7_ODCAP_UCLK_LIMITS, SMU_13_0_7_ODCAP_POWER_LIMIT, SMU_13_0_7_ODCAP_FAN_ACOUSTIC_LIMIT, @@ -59,13 +58,13 @@ enum SMU_13_0_7_ODFEATURE_CAP SMU_13_0_7_ODCAP_FAN_CURVE, SMU_13_0_7_ODCAP_AUTO_FAN_ACOUSTIC_LIMIT, SMU_13_0_7_ODCAP_POWER_MODE, + SMU_13_0_7_ODCAP_PER_ZONE_GFX_VOLTAGE_OFFSET, SMU_13_0_7_ODCAP_COUNT, }; enum SMU_13_0_7_ODFEATURE_ID { SMU_13_0_7_ODFEATURE_GFXCLK_LIMITS = 1 << SMU_13_0_7_ODCAP_GFXCLK_LIMITS, //GFXCLK Limit feature - SMU_13_0_7_ODFEATURE_GFXCLK_CURVE = 1 << SMU_13_0_7_ODCAP_GFXCLK_CURVE, //GFXCLK Curve feature SMU_13_0_7_ODFEATURE_UCLK_LIMITS = 1 << SMU_13_0_7_ODCAP_UCLK_LIMITS, //UCLK Limit feature SMU_13_0_7_ODFEATURE_POWER_LIMIT = 1 << SMU_13_0_7_ODCAP_POWER_LIMIT, //Power Limit feature SMU_13_0_7_ODFEATURE_FAN_ACOUSTIC_LIMIT = 1 << SMU_13_0_7_ODCAP_FAN_ACOUSTIC_LIMIT, //Fan Acoustic RPM feature @@ -80,6 +79,7 @@ enum SMU_13_0_7_ODFEATURE_ID SMU_13_0_7_ODFEATURE_FAN_CURVE = 1 << SMU_13_0_7_ODCAP_FAN_CURVE, //Fan Curve feature SMU_13_0_7_ODFEATURE_AUTO_FAN_ACOUSTIC_LIMIT = 1 << SMU_13_0_7_ODCAP_AUTO_FAN_ACOUSTIC_LIMIT, //Auto Fan Acoustic RPM feature SMU_13_0_7_ODFEATURE_POWER_MODE = 1 << SMU_13_0_7_ODCAP_POWER_MODE, //Optimized GPU Power Mode feature + SMU_13_0_7_ODFEATURE_PER_ZONE_GFX_VOLTAGE_OFFSET = 1 << SMU_13_0_7_ODCAP_PER_ZONE_GFX_VOLTAGE_OFFSET, //Perzone voltage offset feature SMU_13_0_7_ODFEATURE_COUNT = 16, }; @@ -89,10 +89,6 @@ enum SMU_13_0_7_ODSETTING_ID { SMU_13_0_7_ODSETTING_GFXCLKFMAX = 0, SMU_13_0_7_ODSETTING_GFXCLKFMIN, - SMU_13_0_7_ODSETTING_CUSTOM_GFX_VF_CURVE_A, - SMU_13_0_7_ODSETTING_CUSTOM_GFX_VF_CURVE_B, - SMU_13_0_7_ODSETTING_CUSTOM_GFX_VF_CURVE_C, - SMU_13_0_7_ODSETTING_CUSTOM_CURVE_VFT_FMIN, SMU_13_0_7_ODSETTING_UCLKFMIN, SMU_13_0_7_ODSETTING_UCLKFMAX, SMU_13_0_7_ODSETTING_POWERPERCENTAGE, @@ -117,6 +113,12 @@ enum SMU_13_0_7_ODSETTING_ID SMU_13_0_7_ODSETTING_FAN_CURVE_SPEED_5, SMU_13_0_7_ODSETTING_AUTO_FAN_ACOUSTIC_LIMIT, SMU_13_0_7_ODSETTING_POWER_MODE, + SMU_13_0_7_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_1, + SMU_13_0_7_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_2, + SMU_13_0_7_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_3, + SMU_13_0_7_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_4, + SMU_13_0_7_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_5, + SMU_13_0_7_ODSETTING_PER_ZONE_GFX_VOLTAGE_OFFSET_POINT_6, SMU_13_0_7_ODSETTING_COUNT, }; #define SMU_13_0_7_MAX_ODSETTING 64 //Maximum Number of ODSettings diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c index c4000518dc56d84da730a439eb192671d18f503f..275f708db63626183e2a45b14782a7b54ac12328 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c @@ -3413,26 +3413,8 @@ static int navi10_post_smu_init(struct smu_context *smu) return 0; ret = navi10_run_umc_cdr_workaround(smu); - if (ret) { + if (ret) dev_err(adev->dev, "Failed to apply umc cdr workaround!\n"); - return ret; - } - - if (!smu->dc_controlled_by_gpio) { - /* - * For Navi1X, manually switch it to AC mode as PMFW - * may boot it with DC mode. - */ - ret = smu_v11_0_set_power_source(smu, - adev->pm.ac_power ? - SMU_POWER_SOURCE_AC : - SMU_POWER_SOURCE_DC); - if (ret) { - dev_err(adev->dev, "Failed to switch to %s mode!\n", - adev->pm.ac_power ? "AC" : "DC"); - return ret; - } - } return ret; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c index 75f18681e984c33a5c7cf83d7a22aa92b794f3c9..f7ed3e655e39796861ec9e34f342d8eed6f147de 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c @@ -431,7 +431,13 @@ static int sienna_cichlid_append_powerplay_table(struct smu_context *smu) { struct atom_smc_dpm_info_v4_9 *smc_dpm_table; int index, ret; - I2cControllerConfig_t *table_member; + PPTable_beige_goby_t *ppt_beige_goby; + PPTable_t *ppt; + + if (smu->adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 13)) + ppt_beige_goby = smu->smu_table.driver_pptable; + else + ppt = smu->smu_table.driver_pptable; index = get_index_into_master_table(atom_master_list_of_data_tables_v2_1, smc_dpm_info); @@ -440,9 +446,13 @@ static int sienna_cichlid_append_powerplay_table(struct smu_context *smu) (uint8_t **)&smc_dpm_table); if (ret) return ret; - GET_PPTABLE_MEMBER(I2cControllers, &table_member); - memcpy(table_member, smc_dpm_table->I2cControllers, - sizeof(*smc_dpm_table) - sizeof(smc_dpm_table->table_header)); + + if (smu->adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 13)) + smu_memcpy_trailing(ppt_beige_goby, I2cControllers, BoardReserved, + smc_dpm_table, I2cControllers); + else + smu_memcpy_trailing(ppt, I2cControllers, BoardReserved, + smc_dpm_table, I2cControllers); return 0; } @@ -2067,33 +2077,94 @@ static int sienna_cichlid_display_disable_memory_clock_switch(struct smu_context return ret; } +static void sienna_cichlid_get_override_pcie_settings(struct smu_context *smu, + uint32_t *gen_speed_override, + uint32_t *lane_width_override) +{ + struct amdgpu_device *adev = smu->adev; + + *gen_speed_override = 0xff; + *lane_width_override = 0xff; + + switch (adev->pdev->device) { + case 0x73A0: + case 0x73A1: + case 0x73A2: + case 0x73A3: + case 0x73AB: + case 0x73AE: + /* Bit 7:0: PCIE lane width, 1 to 7 corresponds is x1 to x32 */ + *lane_width_override = 6; + break; + case 0x73E0: + case 0x73E1: + case 0x73E3: + *lane_width_override = 4; + break; + case 0x7420: + case 0x7421: + case 0x7422: + case 0x7423: + case 0x7424: + *lane_width_override = 3; + break; + default: + break; + } +} + +#define MAX(a, b) ((a) > (b) ? (a) : (b)) + static int sienna_cichlid_update_pcie_parameters(struct smu_context *smu, uint32_t pcie_gen_cap, uint32_t pcie_width_cap) { struct smu_11_0_dpm_context *dpm_context = smu->smu_dpm.dpm_context; - - uint32_t smu_pcie_arg; + struct smu_11_0_pcie_table *pcie_table = &dpm_context->dpm_tables.pcie_table; + uint32_t gen_speed_override, lane_width_override; uint8_t *table_member1, *table_member2; + uint32_t min_gen_speed, max_gen_speed; + uint32_t min_lane_width, max_lane_width; + uint32_t smu_pcie_arg; int ret, i; GET_PPTABLE_MEMBER(PcieGenSpeed, &table_member1); GET_PPTABLE_MEMBER(PcieLaneCount, &table_member2); - /* lclk dpm table setup */ - for (i = 0; i < MAX_PCIE_CONF; i++) { - dpm_context->dpm_tables.pcie_table.pcie_gen[i] = table_member1[i]; - dpm_context->dpm_tables.pcie_table.pcie_lane[i] = table_member2[i]; + sienna_cichlid_get_override_pcie_settings(smu, + &gen_speed_override, + &lane_width_override); + + /* PCIE gen speed override */ + if (gen_speed_override != 0xff) { + min_gen_speed = MIN(pcie_gen_cap, gen_speed_override); + max_gen_speed = MIN(pcie_gen_cap, gen_speed_override); + } else { + min_gen_speed = MAX(0, table_member1[0]); + max_gen_speed = MIN(pcie_gen_cap, table_member1[1]); + min_gen_speed = min_gen_speed > max_gen_speed ? + max_gen_speed : min_gen_speed; + } + pcie_table->pcie_gen[0] = min_gen_speed; + pcie_table->pcie_gen[1] = max_gen_speed; + + /* PCIE lane width override */ + if (lane_width_override != 0xff) { + min_lane_width = MIN(pcie_width_cap, lane_width_override); + max_lane_width = MIN(pcie_width_cap, lane_width_override); + } else { + min_lane_width = MAX(1, table_member2[0]); + max_lane_width = MIN(pcie_width_cap, table_member2[1]); + min_lane_width = min_lane_width > max_lane_width ? + max_lane_width : min_lane_width; } + pcie_table->pcie_lane[0] = min_lane_width; + pcie_table->pcie_lane[1] = max_lane_width; for (i = 0; i < NUM_LINK_LEVELS; i++) { - smu_pcie_arg = (i << 16) | - ((table_member1[i] <= pcie_gen_cap) ? - (table_member1[i] << 8) : - (pcie_gen_cap << 8)) | - ((table_member2[i] <= pcie_width_cap) ? - table_member2[i] : - pcie_width_cap); + smu_pcie_arg = (i << 16 | + pcie_table->pcie_gen[i] << 8 | + pcie_table->pcie_lane[i]); ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_OverridePcieParameters, @@ -2101,11 +2172,6 @@ static int sienna_cichlid_update_pcie_parameters(struct smu_context *smu, NULL); if (ret) return ret; - - if (table_member1[i] > pcie_gen_cap) - dpm_context->dpm_tables.pcie_table.pcie_gen[i] = pcie_gen_cap; - if (table_member2[i] > pcie_width_cap) - dpm_context->dpm_tables.pcie_table.pcie_lane[i] = pcie_width_cap; } return 0; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c index 7433dcaa16e046af9c311276c0a04f5f7214928d..067b4e0b026c0b9ce76391723bc45a6492b7afcb 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c @@ -582,7 +582,7 @@ static int vangogh_print_legacy_clk_levels(struct smu_context *smu, DpmClocks_t *clk_table = smu->smu_table.clocks_table; SmuMetrics_legacy_t metrics; struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm); - int i, size = 0, ret = 0; + int i, idx, size = 0, ret = 0; uint32_t cur_value = 0, value = 0, count = 0; bool cur_value_match_level = false; @@ -656,7 +656,8 @@ static int vangogh_print_legacy_clk_levels(struct smu_context *smu, case SMU_MCLK: case SMU_FCLK: for (i = 0; i < count; i++) { - ret = vangogh_get_dpm_clk_limited(smu, clk_type, i, &value); + idx = (clk_type == SMU_FCLK || clk_type == SMU_MCLK) ? (count - i - 1) : i; + ret = vangogh_get_dpm_clk_limited(smu, clk_type, idx, &value); if (ret) return ret; if (!value) @@ -683,7 +684,7 @@ static int vangogh_print_clk_levels(struct smu_context *smu, DpmClocks_t *clk_table = smu->smu_table.clocks_table; SmuMetrics_t metrics; struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm); - int i, size = 0, ret = 0; + int i, idx, size = 0, ret = 0; uint32_t cur_value = 0, value = 0, count = 0; bool cur_value_match_level = false; uint32_t min, max; @@ -765,7 +766,8 @@ static int vangogh_print_clk_levels(struct smu_context *smu, case SMU_MCLK: case SMU_FCLK: for (i = 0; i < count; i++) { - ret = vangogh_get_dpm_clk_limited(smu, clk_type, i, &value); + idx = (clk_type == SMU_FCLK || clk_type == SMU_MCLK) ? (count - i - 1) : i; + ret = vangogh_get_dpm_clk_limited(smu, clk_type, idx, &value); if (ret) return ret; if (!value) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c index 5cdc07165480b5a3be51ec46c722699d34d52b84..8a8ba25c9ad7cc29437103b22d58555cd8afbe31 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c @@ -494,7 +494,7 @@ static int renoir_set_fine_grain_gfx_freq_parameters(struct smu_context *smu) static int renoir_print_clk_levels(struct smu_context *smu, enum smu_clk_type clk_type, char *buf) { - int i, size = 0, ret = 0; + int i, idx, size = 0, ret = 0; uint32_t cur_value = 0, value = 0, count = 0, min = 0, max = 0; SmuMetrics_t metrics; struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm); @@ -594,7 +594,8 @@ static int renoir_print_clk_levels(struct smu_context *smu, case SMU_VCLK: case SMU_DCLK: for (i = 0; i < count; i++) { - ret = renoir_get_dpm_clk_limited(smu, clk_type, i, &value); + idx = (clk_type == SMU_FCLK || clk_type == SMU_MCLK) ? (count - i - 1) : i; + ret = renoir_get_dpm_clk_limited(smu, clk_type, idx, &value); if (ret) return ret; if (!value) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c index d30ec3005ea191ad693a473d20921f1a46bdaf9d..e80f122d8aec5bab03ad03b53f71dade1fe5e377 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c @@ -2147,5 +2147,6 @@ void aldebaran_set_ppt_funcs(struct smu_context *smu) smu->clock_map = aldebaran_clk_map; smu->feature_map = aldebaran_feature_mask_map; smu->table_map = aldebaran_table_map; + smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_ALDE; smu_v13_0_set_smu_mailbox_registers(smu); } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c index 393c6a7b960962603a756c4e4651f52ce98735d5..e52c563f0dacc50a4aeed6dbe0e6c0809cec5f8b 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c @@ -269,45 +269,10 @@ int smu_v13_0_check_fw_version(struct smu_context *smu) smu_major = (smu_version >> 16) & 0xff; smu_minor = (smu_version >> 8) & 0xff; smu_debug = (smu_version >> 0) & 0xff; - if (smu->is_apu) + if (smu->is_apu || + adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 6)) adev->pm.fw_version = smu_version; - switch (adev->ip_versions[MP1_HWIP][0]) { - case IP_VERSION(13, 0, 2): - smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_ALDE; - break; - case IP_VERSION(13, 0, 0): - smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_SMU_V13_0_0_0; - break; - case IP_VERSION(13, 0, 10): - smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_SMU_V13_0_0_10; - break; - case IP_VERSION(13, 0, 7): - smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_SMU_V13_0_7; - break; - case IP_VERSION(13, 0, 1): - case IP_VERSION(13, 0, 3): - case IP_VERSION(13, 0, 8): - smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_YELLOW_CARP; - break; - case IP_VERSION(13, 0, 4): - case IP_VERSION(13, 0, 11): - smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_SMU_V13_0_4; - break; - case IP_VERSION(13, 0, 5): - smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_SMU_V13_0_5; - break; - case IP_VERSION(13, 0, 6): - smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_SMU_V13_0_6; - adev->pm.fw_version = smu_version; - break; - default: - dev_err(adev->dev, "smu unsupported IP version: 0x%x.\n", - adev->ip_versions[MP1_HWIP][0]); - smu->smc_driver_if_version = SMU13_DRIVER_IF_VERSION_INV; - break; - } - /* only for dGPU w/ SMU13*/ if (adev->pm.fw) dev_dbg(smu->adev->dev, "smu fw reported program %d, version = 0x%08x (%d.%d.%d)\n", @@ -502,17 +467,26 @@ int smu_v13_0_init_smc_tables(struct smu_context *smu) ret = -ENOMEM; goto err3_out; } + + smu_table->user_overdrive_table = + kzalloc(tables[SMU_TABLE_OVERDRIVE].size, GFP_KERNEL); + if (!smu_table->user_overdrive_table) { + ret = -ENOMEM; + goto err4_out; + } } smu_table->combo_pptable = kzalloc(tables[SMU_TABLE_COMBO_PPTABLE].size, GFP_KERNEL); if (!smu_table->combo_pptable) { ret = -ENOMEM; - goto err4_out; + goto err5_out; } return 0; +err5_out: + kfree(smu_table->user_overdrive_table); err4_out: kfree(smu_table->boot_overdrive_table); err3_out: @@ -532,12 +506,14 @@ int smu_v13_0_fini_smc_tables(struct smu_context *smu) kfree(smu_table->gpu_metrics_table); kfree(smu_table->combo_pptable); + kfree(smu_table->user_overdrive_table); kfree(smu_table->boot_overdrive_table); kfree(smu_table->overdrive_table); kfree(smu_table->max_sustainable_clocks); kfree(smu_table->driver_pptable); smu_table->gpu_metrics_table = NULL; smu_table->combo_pptable = NULL; + smu_table->user_overdrive_table = NULL; smu_table->boot_overdrive_table = NULL; smu_table->overdrive_table = NULL; smu_table->max_sustainable_clocks = NULL; @@ -573,11 +549,11 @@ int smu_v13_0_init_power(struct smu_context *smu) if (smu_power->power_context || smu_power->power_context_size != 0) return -EINVAL; - smu_power->power_context = kzalloc(sizeof(struct smu_13_0_dpm_context), + smu_power->power_context = kzalloc(sizeof(struct smu_13_0_power_context), GFP_KERNEL); if (!smu_power->power_context) return -ENOMEM; - smu_power->power_context_size = sizeof(struct smu_13_0_dpm_context); + smu_power->power_context_size = sizeof(struct smu_13_0_power_context); return 0; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c index 09405ef1e3c83a8f7bf0263cc8e1c0794a6454dd..a6083957ae51b531896bc582b7b77269e3cb8cb7 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c @@ -46,7 +46,6 @@ #include "asic_reg/mp/mp_13_0_0_sh_mask.h" #include "smu_cmn.h" #include "amdgpu_ras.h" -#include "umc_v8_10.h" /* * DO NOT use these for err/warn/info/debug messages. @@ -237,6 +236,7 @@ static struct cmn2asic_mapping smu_v13_0_0_table_map[SMU_TABLE_COUNT] = { [SMU_TABLE_COMBO_PPTABLE] = {1, TABLE_COMBO_PPTABLE}, TAB_MAP(I2C_COMMANDS), TAB_MAP(ECCINFO), + TAB_MAP(OVERDRIVE), }; static struct cmn2asic_mapping smu_v13_0_0_pwr_src_map[SMU_POWER_SOURCE_COUNT] = { @@ -331,6 +331,11 @@ static int smu_v13_0_0_check_powerplay_table(struct smu_context *smu) struct smu_13_0_0_powerplay_table *powerplay_table = table_context->power_play_table; struct smu_baco_context *smu_baco = &smu->smu_baco; + PPTable_t *pptable = smu->smu_table.driver_pptable; + const OverDriveLimits_t * const overdrive_upperlimits = + &pptable->SkuTable.OverDriveLimitsBasicMax; + const OverDriveLimits_t * const overdrive_lowerlimits = + &pptable->SkuTable.OverDriveLimitsMin; if (powerplay_table->platform_caps & SMU_13_0_0_PP_PLATFORM_CAP_HARDWAREDC) smu->dc_controlled_by_gpio = true; @@ -342,6 +347,10 @@ static int smu_v13_0_0_check_powerplay_table(struct smu_context *smu) if (powerplay_table->platform_caps & SMU_13_0_0_PP_PLATFORM_CAP_MACO) smu_baco->maco_support = true; + if (!overdrive_lowerlimits->FeatureCtrlMask || + !overdrive_upperlimits->FeatureCtrlMask) + smu->od_enabled = false; + table_context->thermal_controller_type = powerplay_table->thermal_controller_type; @@ -461,7 +470,7 @@ static int smu_v13_0_0_tables_init(struct smu_context *smu) PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); SMU_TABLE_INIT(tables, SMU_TABLE_I2C_COMMANDS, sizeof(SwI2cRequest_t), PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); - SMU_TABLE_INIT(tables, SMU_TABLE_OVERDRIVE, sizeof(OverDriveTable_t), + SMU_TABLE_INIT(tables, SMU_TABLE_OVERDRIVE, sizeof(OverDriveTableExternal_t), PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); SMU_TABLE_INIT(tables, SMU_TABLE_PMSTATUSLOG, SMU13_TOOL_SIZE, PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); @@ -1022,17 +1031,119 @@ static int smu_v13_0_0_get_current_clk_freq_by_table(struct smu_context *smu, value); } +static bool smu_v13_0_0_is_od_feature_supported(struct smu_context *smu, + int od_feature_bit) +{ + PPTable_t *pptable = smu->smu_table.driver_pptable; + const OverDriveLimits_t * const overdrive_upperlimits = + &pptable->SkuTable.OverDriveLimitsBasicMax; + + return overdrive_upperlimits->FeatureCtrlMask & (1U << od_feature_bit); +} + +static void smu_v13_0_0_get_od_setting_limits(struct smu_context *smu, + int od_feature_bit, + bool lower_boundary, + int32_t *min, + int32_t *max) +{ + PPTable_t *pptable = smu->smu_table.driver_pptable; + const OverDriveLimits_t * const overdrive_upperlimits = + &pptable->SkuTable.OverDriveLimitsBasicMax; + const OverDriveLimits_t * const overdrive_lowerlimits = + &pptable->SkuTable.OverDriveLimitsMin; + int32_t od_min_setting, od_max_setting; + + switch (od_feature_bit) { + case PP_OD_FEATURE_GFXCLK_BIT: + if (lower_boundary) { + od_min_setting = overdrive_lowerlimits->GfxclkFmin; + od_max_setting = overdrive_upperlimits->GfxclkFmin; + } else { + od_min_setting = overdrive_lowerlimits->GfxclkFmax; + od_max_setting = overdrive_upperlimits->GfxclkFmax; + } + break; + case PP_OD_FEATURE_UCLK_BIT: + if (lower_boundary) { + od_min_setting = overdrive_lowerlimits->UclkFmin; + od_max_setting = overdrive_upperlimits->UclkFmin; + } else { + od_min_setting = overdrive_lowerlimits->UclkFmax; + od_max_setting = overdrive_upperlimits->UclkFmax; + } + break; + case PP_OD_FEATURE_GFX_VF_CURVE_BIT: + od_min_setting = overdrive_lowerlimits->VoltageOffsetPerZoneBoundary; + od_max_setting = overdrive_upperlimits->VoltageOffsetPerZoneBoundary; + break; + default: + break; + } + + if (min) + *min = od_min_setting; + if (max) + *max = od_max_setting; +} + +static void smu_v13_0_0_dump_od_table(struct smu_context *smu, + OverDriveTableExternal_t *od_table) +{ + struct amdgpu_device *adev = smu->adev; + + dev_dbg(adev->dev, "OD: Gfxclk: (%d, %d)\n", od_table->OverDriveTable.GfxclkFmin, + od_table->OverDriveTable.GfxclkFmax); + dev_dbg(adev->dev, "OD: Uclk: (%d, %d)\n", od_table->OverDriveTable.UclkFmin, + od_table->OverDriveTable.UclkFmax); +} + +static int smu_v13_0_0_get_overdrive_table(struct smu_context *smu, + OverDriveTableExternal_t *od_table) +{ + int ret = 0; + + ret = smu_cmn_update_table(smu, + SMU_TABLE_OVERDRIVE, + 0, + (void *)od_table, + false); + if (ret) + dev_err(smu->adev->dev, "Failed to get overdrive table!\n"); + + return ret; +} + +static int smu_v13_0_0_upload_overdrive_table(struct smu_context *smu, + OverDriveTableExternal_t *od_table) +{ + int ret = 0; + + ret = smu_cmn_update_table(smu, + SMU_TABLE_OVERDRIVE, + 0, + (void *)od_table, + true); + if (ret) + dev_err(smu->adev->dev, "Failed to upload overdrive table!\n"); + + return ret; +} + static int smu_v13_0_0_print_clk_levels(struct smu_context *smu, enum smu_clk_type clk_type, char *buf) { struct smu_dpm_context *smu_dpm = &smu->smu_dpm; struct smu_13_0_dpm_context *dpm_context = smu_dpm->dpm_context; + OverDriveTableExternal_t *od_table = + (OverDriveTableExternal_t *)smu->smu_table.overdrive_table; struct smu_13_0_dpm_table *single_dpm_table; struct smu_13_0_pcie_table *pcie_table; const int link_width[] = {0, 1, 2, 4, 8, 12, 16}; uint32_t gen_speed, lane_width; int i, curr_freq, size = 0; + int32_t min_value, max_value; int ret = 0; smu_cmn_get_sysfs_buf(&buf, &size); @@ -1149,6 +1260,89 @@ static int smu_v13_0_0_print_clk_levels(struct smu_context *smu, "*" : ""); break; + case SMU_OD_SCLK: + if (!smu_v13_0_0_is_od_feature_supported(smu, + PP_OD_FEATURE_GFXCLK_BIT)) + break; + + size += sysfs_emit_at(buf, size, "OD_SCLK:\n"); + size += sysfs_emit_at(buf, size, "0: %uMhz\n1: %uMhz\n", + od_table->OverDriveTable.GfxclkFmin, + od_table->OverDriveTable.GfxclkFmax); + break; + + case SMU_OD_MCLK: + if (!smu_v13_0_0_is_od_feature_supported(smu, + PP_OD_FEATURE_UCLK_BIT)) + break; + + size += sysfs_emit_at(buf, size, "OD_MCLK:\n"); + size += sysfs_emit_at(buf, size, "0: %uMhz\n1: %uMHz\n", + od_table->OverDriveTable.UclkFmin, + od_table->OverDriveTable.UclkFmax); + break; + + case SMU_OD_VDDC_CURVE: + if (!smu_v13_0_0_is_od_feature_supported(smu, + PP_OD_FEATURE_GFX_VF_CURVE_BIT)) + break; + + size += sysfs_emit_at(buf, size, "OD_VDDC_CURVE:\n"); + for (i = 0; i < PP_NUM_OD_VF_CURVE_POINTS; i++) + size += sysfs_emit_at(buf, size, "%d: %dmv\n", + i, + od_table->OverDriveTable.VoltageOffsetPerZoneBoundary[i]); + break; + + case SMU_OD_RANGE: + if (!smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_GFXCLK_BIT) && + !smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_UCLK_BIT) && + !smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_GFX_VF_CURVE_BIT)) + break; + + size += sysfs_emit_at(buf, size, "%s:\n", "OD_RANGE"); + + if (smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_GFXCLK_BIT)) { + smu_v13_0_0_get_od_setting_limits(smu, + PP_OD_FEATURE_GFXCLK_BIT, + true, + &min_value, + NULL); + smu_v13_0_0_get_od_setting_limits(smu, + PP_OD_FEATURE_GFXCLK_BIT, + false, + NULL, + &max_value); + size += sysfs_emit_at(buf, size, "SCLK: %7uMhz %10uMhz\n", + min_value, max_value); + } + + if (smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_UCLK_BIT)) { + smu_v13_0_0_get_od_setting_limits(smu, + PP_OD_FEATURE_UCLK_BIT, + true, + &min_value, + NULL); + smu_v13_0_0_get_od_setting_limits(smu, + PP_OD_FEATURE_UCLK_BIT, + false, + NULL, + &max_value); + size += sysfs_emit_at(buf, size, "MCLK: %7uMhz %10uMhz\n", + min_value, max_value); + } + + if (smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_GFX_VF_CURVE_BIT)) { + smu_v13_0_0_get_od_setting_limits(smu, + PP_OD_FEATURE_GFX_VF_CURVE_BIT, + true, + &min_value, + &max_value); + size += sysfs_emit_at(buf, size, "VDDC_CURVE: %7dmv %10dmv\n", + min_value, max_value); + } + break; + default: break; } @@ -1156,6 +1350,222 @@ static int smu_v13_0_0_print_clk_levels(struct smu_context *smu, return size; } +static int smu_v13_0_0_od_edit_dpm_table(struct smu_context *smu, + enum PP_OD_DPM_TABLE_COMMAND type, + long input[], + uint32_t size) +{ + struct smu_table_context *table_context = &smu->smu_table; + OverDriveTableExternal_t *od_table = + (OverDriveTableExternal_t *)table_context->overdrive_table; + struct amdgpu_device *adev = smu->adev; + uint32_t offset_of_featurectrlmask; + int32_t minimum, maximum; + uint32_t feature_ctrlmask; + int i, ret = 0; + + switch (type) { + case PP_OD_EDIT_SCLK_VDDC_TABLE: + if (!smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_GFXCLK_BIT)) { + dev_warn(adev->dev, "GFXCLK_LIMITS setting not supported!\n"); + return -ENOTSUPP; + } + + for (i = 0; i < size; i += 2) { + if (i + 2 > size) { + dev_info(adev->dev, "invalid number of input parameters %d\n", size); + return -EINVAL; + } + + switch (input[i]) { + case 0: + smu_v13_0_0_get_od_setting_limits(smu, + PP_OD_FEATURE_GFXCLK_BIT, + true, + &minimum, + &maximum); + if (input[i + 1] < minimum || + input[i + 1] > maximum) { + dev_info(adev->dev, "GfxclkFmin (%ld) must be within [%u, %u]!\n", + input[i + 1], minimum, maximum); + return -EINVAL; + } + + od_table->OverDriveTable.GfxclkFmin = input[i + 1]; + od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_GFXCLK_BIT; + break; + + case 1: + smu_v13_0_0_get_od_setting_limits(smu, + PP_OD_FEATURE_GFXCLK_BIT, + false, + &minimum, + &maximum); + if (input[i + 1] < minimum || + input[i + 1] > maximum) { + dev_info(adev->dev, "GfxclkFmax (%ld) must be within [%u, %u]!\n", + input[i + 1], minimum, maximum); + return -EINVAL; + } + + od_table->OverDriveTable.GfxclkFmax = input[i + 1]; + od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_GFXCLK_BIT; + break; + + default: + dev_info(adev->dev, "Invalid SCLK_VDDC_TABLE index: %ld\n", input[i]); + dev_info(adev->dev, "Supported indices: [0:min,1:max]\n"); + return -EINVAL; + } + } + + if (od_table->OverDriveTable.GfxclkFmin > od_table->OverDriveTable.GfxclkFmax) { + dev_err(adev->dev, + "Invalid setting: GfxclkFmin(%u) is bigger than GfxclkFmax(%u)\n", + (uint32_t)od_table->OverDriveTable.GfxclkFmin, + (uint32_t)od_table->OverDriveTable.GfxclkFmax); + return -EINVAL; + } + break; + + case PP_OD_EDIT_MCLK_VDDC_TABLE: + if (!smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_UCLK_BIT)) { + dev_warn(adev->dev, "UCLK_LIMITS setting not supported!\n"); + return -ENOTSUPP; + } + + for (i = 0; i < size; i += 2) { + if (i + 2 > size) { + dev_info(adev->dev, "invalid number of input parameters %d\n", size); + return -EINVAL; + } + + switch (input[i]) { + case 0: + smu_v13_0_0_get_od_setting_limits(smu, + PP_OD_FEATURE_UCLK_BIT, + true, + &minimum, + &maximum); + if (input[i + 1] < minimum || + input[i + 1] > maximum) { + dev_info(adev->dev, "UclkFmin (%ld) must be within [%u, %u]!\n", + input[i + 1], minimum, maximum); + return -EINVAL; + } + + od_table->OverDriveTable.UclkFmin = input[i + 1]; + od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_UCLK_BIT; + break; + + case 1: + smu_v13_0_0_get_od_setting_limits(smu, + PP_OD_FEATURE_UCLK_BIT, + false, + &minimum, + &maximum); + if (input[i + 1] < minimum || + input[i + 1] > maximum) { + dev_info(adev->dev, "UclkFmax (%ld) must be within [%u, %u]!\n", + input[i + 1], minimum, maximum); + return -EINVAL; + } + + od_table->OverDriveTable.UclkFmax = input[i + 1]; + od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_UCLK_BIT; + break; + + default: + dev_info(adev->dev, "Invalid MCLK_VDDC_TABLE index: %ld\n", input[i]); + dev_info(adev->dev, "Supported indices: [0:min,1:max]\n"); + return -EINVAL; + } + } + + if (od_table->OverDriveTable.UclkFmin > od_table->OverDriveTable.UclkFmax) { + dev_err(adev->dev, + "Invalid setting: UclkFmin(%u) is bigger than UclkFmax(%u)\n", + (uint32_t)od_table->OverDriveTable.UclkFmin, + (uint32_t)od_table->OverDriveTable.UclkFmax); + return -EINVAL; + } + break; + + case PP_OD_EDIT_VDDC_CURVE: + if (!smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_GFX_VF_CURVE_BIT)) { + dev_warn(adev->dev, "VF curve setting not supported!\n"); + return -ENOTSUPP; + } + + if (input[0] >= PP_NUM_OD_VF_CURVE_POINTS || + input[0] < 0) + return -EINVAL; + + smu_v13_0_0_get_od_setting_limits(smu, + PP_OD_FEATURE_GFX_VF_CURVE_BIT, + true, + &minimum, + &maximum); + if (input[1] < minimum || + input[1] > maximum) { + dev_info(adev->dev, "Voltage offset (%ld) must be within [%d, %d]!\n", + input[1], minimum, maximum); + return -EINVAL; + } + + od_table->OverDriveTable.VoltageOffsetPerZoneBoundary[input[0]] = input[1]; + od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_GFX_VF_CURVE_BIT; + break; + + case PP_OD_RESTORE_DEFAULT_TABLE: + feature_ctrlmask = od_table->OverDriveTable.FeatureCtrlMask; + memcpy(od_table, + table_context->boot_overdrive_table, + sizeof(OverDriveTableExternal_t)); + od_table->OverDriveTable.FeatureCtrlMask = feature_ctrlmask; + fallthrough; + + case PP_OD_COMMIT_DPM_TABLE: + /* + * The member below instructs PMFW the settings focused in + * this single operation. + * `uint32_t FeatureCtrlMask;` + * It does not contain actual informations about user's custom + * settings. Thus we do not cache it. + */ + offset_of_featurectrlmask = offsetof(OverDriveTable_t, FeatureCtrlMask); + if (memcmp((u8 *)od_table + offset_of_featurectrlmask, + table_context->user_overdrive_table + offset_of_featurectrlmask, + sizeof(OverDriveTableExternal_t) - offset_of_featurectrlmask)) { + smu_v13_0_0_dump_od_table(smu, od_table); + + ret = smu_v13_0_0_upload_overdrive_table(smu, od_table); + if (ret) { + dev_err(adev->dev, "Failed to upload overdrive table!\n"); + return ret; + } + + od_table->OverDriveTable.FeatureCtrlMask = 0; + memcpy(table_context->user_overdrive_table + offset_of_featurectrlmask, + (u8 *)od_table + offset_of_featurectrlmask, + sizeof(OverDriveTableExternal_t) - offset_of_featurectrlmask); + + if (!memcmp(table_context->user_overdrive_table, + table_context->boot_overdrive_table, + sizeof(OverDriveTableExternal_t))) + smu->user_dpm_profile.user_od = false; + else + smu->user_dpm_profile.user_od = true; + } + break; + + default: + return -ENOSYS; + } + + return ret; +} + static int smu_v13_0_0_force_clk_levels(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t mask) @@ -1384,6 +1794,78 @@ static ssize_t smu_v13_0_0_get_gpu_metrics(struct smu_context *smu, return sizeof(struct gpu_metrics_v1_3); } +static int smu_v13_0_0_set_default_od_settings(struct smu_context *smu) +{ + OverDriveTableExternal_t *od_table = + (OverDriveTableExternal_t *)smu->smu_table.overdrive_table; + OverDriveTableExternal_t *boot_od_table = + (OverDriveTableExternal_t *)smu->smu_table.boot_overdrive_table; + OverDriveTableExternal_t *user_od_table = + (OverDriveTableExternal_t *)smu->smu_table.user_overdrive_table; + OverDriveTableExternal_t user_od_table_bak; + int ret = 0; + int i; + + ret = smu_v13_0_0_get_overdrive_table(smu, boot_od_table); + if (ret) + return ret; + + smu_v13_0_0_dump_od_table(smu, boot_od_table); + + memcpy(od_table, + boot_od_table, + sizeof(OverDriveTableExternal_t)); + + /* + * For S3/S4/Runpm resume, we need to setup those overdrive tables again, + * but we have to preserve user defined values in "user_od_table". + */ + if (!smu->adev->in_suspend) { + memcpy(user_od_table, + boot_od_table, + sizeof(OverDriveTableExternal_t)); + smu->user_dpm_profile.user_od = false; + } else if (smu->user_dpm_profile.user_od) { + memcpy(&user_od_table_bak, + user_od_table, + sizeof(OverDriveTableExternal_t)); + memcpy(user_od_table, + boot_od_table, + sizeof(OverDriveTableExternal_t)); + user_od_table->OverDriveTable.GfxclkFmin = + user_od_table_bak.OverDriveTable.GfxclkFmin; + user_od_table->OverDriveTable.GfxclkFmax = + user_od_table_bak.OverDriveTable.GfxclkFmax; + user_od_table->OverDriveTable.UclkFmin = + user_od_table_bak.OverDriveTable.UclkFmin; + user_od_table->OverDriveTable.UclkFmax = + user_od_table_bak.OverDriveTable.UclkFmax; + for (i = 0; i < PP_NUM_OD_VF_CURVE_POINTS; i++) + user_od_table->OverDriveTable.VoltageOffsetPerZoneBoundary[i] = + user_od_table_bak.OverDriveTable.VoltageOffsetPerZoneBoundary[i]; + } + + return 0; +} + +static int smu_v13_0_0_restore_user_od_settings(struct smu_context *smu) +{ + struct smu_table_context *table_context = &smu->smu_table; + OverDriveTableExternal_t *od_table = table_context->overdrive_table; + OverDriveTableExternal_t *user_od_table = table_context->user_overdrive_table; + int res; + + user_od_table->OverDriveTable.FeatureCtrlMask = 1U << PP_OD_FEATURE_GFXCLK_BIT | + 1U << PP_OD_FEATURE_UCLK_BIT | + 1U << PP_OD_FEATURE_GFX_VF_CURVE_BIT; + res = smu_v13_0_0_upload_overdrive_table(smu, user_od_table); + user_od_table->OverDriveTable.FeatureCtrlMask = 0; + if (res == 0) + memcpy(od_table, user_od_table, sizeof(OverDriveTableExternal_t)); + + return res; +} + static int smu_v13_0_0_populate_umd_state_clk(struct smu_context *smu) { struct smu_13_0_dpm_context *dpm_context = @@ -1696,10 +2178,39 @@ static int smu_v13_0_0_set_power_profile_mode(struct smu_context *smu, } } - /* conv PP_SMC_POWER_PROFILE* to WORKLOAD_PPLIB_*_BIT */ - workload_type = smu_cmn_to_asic_specific_index(smu, + if (smu->power_profile_mode == PP_SMC_POWER_PROFILE_COMPUTE && + (((smu->adev->pdev->device == 0x744C) && (smu->adev->pdev->revision == 0xC8)) || + ((smu->adev->pdev->device == 0x744C) && (smu->adev->pdev->revision == 0xCC)))) { + ret = smu_cmn_update_table(smu, + SMU_TABLE_ACTIVITY_MONITOR_COEFF, + WORKLOAD_PPLIB_COMPUTE_BIT, + (void *)(&activity_monitor_external), + false); + if (ret) { + dev_err(smu->adev->dev, "[%s] Failed to get activity monitor!", __func__); + return ret; + } + + ret = smu_cmn_update_table(smu, + SMU_TABLE_ACTIVITY_MONITOR_COEFF, + WORKLOAD_PPLIB_CUSTOM_BIT, + (void *)(&activity_monitor_external), + true); + if (ret) { + dev_err(smu->adev->dev, "[%s] Failed to set activity monitor!", __func__); + return ret; + } + + workload_type = smu_cmn_to_asic_specific_index(smu, + CMN2ASIC_MAPPING_WORKLOAD, + PP_SMC_POWER_PROFILE_CUSTOM); + } else { + /* conv PP_SMC_POWER_PROFILE* to WORKLOAD_PPLIB_*_BIT */ + workload_type = smu_cmn_to_asic_specific_index(smu, CMN2ASIC_MAPPING_WORKLOAD, smu->power_profile_mode); + } + if (workload_type < 0) return -EINVAL; @@ -2097,7 +2608,7 @@ static ssize_t smu_v13_0_0_get_ecc_info(struct smu_context *smu, ecc_table = (EccInfoTable_t *)smu_table->ecc_table; - for (i = 0; i < UMC_V8_10_TOTAL_CHANNEL_NUM(adev); i++) { + for (i = 0; i < ARRAY_SIZE(ecc_table->EccInfo); i++) { ecc_info_per_channel = &(eccinfo->ecc[i]); ecc_info_per_channel->ce_count_lo_chip = ecc_table->EccInfo[i].ce_count_lo_chip; @@ -2150,6 +2661,9 @@ static const struct pptable_funcs smu_v13_0_0_ppt_funcs = { .notify_memory_pool_location = smu_v13_0_notify_memory_pool_location, .get_gpu_metrics = smu_v13_0_0_get_gpu_metrics, .set_soft_freq_limited_range = smu_v13_0_set_soft_freq_limited_range, + .set_default_od_settings = smu_v13_0_0_set_default_od_settings, + .restore_user_od_settings = smu_v13_0_0_restore_user_od_settings, + .od_edit_dpm_table = smu_v13_0_0_od_edit_dpm_table, .init_pptable_microcode = smu_v13_0_init_pptable_microcode, .populate_umd_state_clk = smu_v13_0_0_populate_umd_state_clk, .set_performance_level = smu_v13_0_set_performance_level, @@ -2199,5 +2713,6 @@ void smu_v13_0_0_set_ppt_funcs(struct smu_context *smu) smu->table_map = smu_v13_0_0_table_map; smu->pwr_src_map = smu_v13_0_0_pwr_src_map; smu->workload_map = smu_v13_0_0_workload_map; + smu->smc_driver_if_version = SMU13_0_0_DRIVER_IF_VERSION; smu_v13_0_0_set_smu_mailbox_registers(smu); } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c index 8fa9a36c38b64e5b0281561308c70a22887adb5c..ef37dda9908fa7bee413bbe2241869e60785ff3d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c @@ -54,6 +54,10 @@ #define FEATURE_MASK(feature) (1ULL << feature) +#define SMU_13_0_4_UMD_PSTATE_GFXCLK 938 +#define SMU_13_0_4_UMD_PSTATE_SOCCLK 938 +#define SMU_13_0_4_UMD_PSTATE_FCLK 1875 + #define SMC_DPM_FEATURE ( \ FEATURE_MASK(FEATURE_CCLK_DPM_BIT) | \ FEATURE_MASK(FEATURE_VCN_DPM_BIT) | \ @@ -478,7 +482,7 @@ static int smu_v13_0_4_get_dpm_level_count(struct smu_context *smu, static int smu_v13_0_4_print_clk_levels(struct smu_context *smu, enum smu_clk_type clk_type, char *buf) { - int i, size = 0, ret = 0; + int i, idx, size = 0, ret = 0; uint32_t cur_value = 0, value = 0, count = 0; uint32_t min, max; @@ -512,7 +516,8 @@ static int smu_v13_0_4_print_clk_levels(struct smu_context *smu, break; for (i = 0; i < count; i++) { - ret = smu_v13_0_4_get_dpm_freq_by_index(smu, clk_type, i, &value); + idx = (clk_type == SMU_FCLK || clk_type == SMU_MCLK) ? (count - i - 1) : i; + ret = smu_v13_0_4_get_dpm_freq_by_index(smu, clk_type, idx, &value); if (ret) break; @@ -830,6 +835,8 @@ static int smu_v13_0_4_set_soft_freq_limited_range(struct smu_context *smu, uint32_t max) { enum smu_message_type msg_set_min, msg_set_max; + uint32_t min_clk = min; + uint32_t max_clk = max; int ret = 0; if (!smu_v13_0_4_clk_dpm_is_enabled(smu, clk_type)) @@ -858,12 +865,17 @@ static int smu_v13_0_4_set_soft_freq_limited_range(struct smu_context *smu, return -EINVAL; } - ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_min, min, NULL); + if (clk_type == SMU_VCLK) { + min_clk = min << SMU_13_VCLK_SHIFT; + max_clk = max << SMU_13_VCLK_SHIFT; + } + + ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_min, min_clk, NULL); if (ret) return ret; return smu_cmn_send_smc_msg_with_param(smu, msg_set_max, - max, NULL); + max_clk, NULL); } static int smu_v13_0_4_force_clk_levels(struct smu_context *smu, @@ -900,6 +912,50 @@ static int smu_v13_0_4_force_clk_levels(struct smu_context *smu, return ret; } +static int smu_v13_0_4_get_dpm_profile_freq(struct smu_context *smu, + enum amd_dpm_forced_level level, + enum smu_clk_type clk_type, + uint32_t *min_clk, + uint32_t *max_clk) +{ + int ret = 0; + uint32_t clk_limit = 0; + + switch (clk_type) { + case SMU_GFXCLK: + case SMU_SCLK: + clk_limit = SMU_13_0_4_UMD_PSTATE_GFXCLK; + if (level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK) + smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SCLK, NULL, &clk_limit); + else if (level == AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK) + smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SCLK, &clk_limit, NULL); + break; + case SMU_SOCCLK: + clk_limit = SMU_13_0_4_UMD_PSTATE_SOCCLK; + if (level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK) + smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SOCCLK, NULL, &clk_limit); + break; + case SMU_FCLK: + clk_limit = SMU_13_0_4_UMD_PSTATE_FCLK; + if (level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK) + smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_FCLK, NULL, &clk_limit); + else if (level == AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK) + smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_FCLK, &clk_limit, NULL); + break; + case SMU_VCLK: + smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_VCLK, NULL, &clk_limit); + break; + case SMU_DCLK: + smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_DCLK, NULL, &clk_limit); + break; + default: + ret = -EINVAL; + break; + } + *min_clk = *max_clk = clk_limit; + return ret; +} + static int smu_v13_0_4_set_performance_level(struct smu_context *smu, enum amd_dpm_forced_level level) { @@ -907,6 +963,8 @@ static int smu_v13_0_4_set_performance_level(struct smu_context *smu, uint32_t sclk_min = 0, sclk_max = 0; uint32_t fclk_min = 0, fclk_max = 0; uint32_t socclk_min = 0, socclk_max = 0; + uint32_t vclk_min = 0, vclk_max = 0; + uint32_t dclk_min = 0, dclk_max = 0; int ret = 0; switch (level) { @@ -914,28 +972,42 @@ static int smu_v13_0_4_set_performance_level(struct smu_context *smu, smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SCLK, NULL, &sclk_max); smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_FCLK, NULL, &fclk_max); smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SOCCLK, NULL, &socclk_max); + smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_VCLK, NULL, &vclk_max); + smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_DCLK, NULL, &dclk_max); sclk_min = sclk_max; fclk_min = fclk_max; socclk_min = socclk_max; + vclk_min = vclk_max; + dclk_min = dclk_max; break; case AMD_DPM_FORCED_LEVEL_LOW: smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SCLK, &sclk_min, NULL); smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_FCLK, &fclk_min, NULL); smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SOCCLK, &socclk_min, NULL); + smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_VCLK, &vclk_min, NULL); + smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_DCLK, &dclk_min, NULL); sclk_max = sclk_min; fclk_max = fclk_min; socclk_max = socclk_min; + vclk_max = vclk_min; + dclk_max = dclk_min; break; case AMD_DPM_FORCED_LEVEL_AUTO: smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SCLK, &sclk_min, &sclk_max); smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_FCLK, &fclk_min, &fclk_max); smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_SOCCLK, &socclk_min, &socclk_max); + smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_VCLK, &vclk_min, &vclk_max); + smu_v13_0_4_get_dpm_ultimate_freq(smu, SMU_DCLK, &dclk_min, &dclk_max); break; case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD: case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK: case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK: case AMD_DPM_FORCED_LEVEL_PROFILE_PEAK: - /* Temporarily do nothing since the optimal clocks haven't been provided yet */ + smu_v13_0_4_get_dpm_profile_freq(smu, level, SMU_SCLK, &sclk_min, &sclk_max); + smu_v13_0_4_get_dpm_profile_freq(smu, level, SMU_FCLK, &fclk_min, &fclk_max); + smu_v13_0_4_get_dpm_profile_freq(smu, level, SMU_SOCCLK, &socclk_min, &socclk_max); + smu_v13_0_4_get_dpm_profile_freq(smu, level, SMU_VCLK, &vclk_min, &vclk_max); + smu_v13_0_4_get_dpm_profile_freq(smu, level, SMU_DCLK, &dclk_min, &dclk_max); break; case AMD_DPM_FORCED_LEVEL_MANUAL: case AMD_DPM_FORCED_LEVEL_PROFILE_EXIT: @@ -975,6 +1047,23 @@ static int smu_v13_0_4_set_performance_level(struct smu_context *smu, return ret; } + if (vclk_min && vclk_max) { + ret = smu_v13_0_4_set_soft_freq_limited_range(smu, + SMU_VCLK, + vclk_min, + vclk_max); + if (ret) + return ret; + } + + if (dclk_min && dclk_max) { + ret = smu_v13_0_4_set_soft_freq_limited_range(smu, + SMU_DCLK, + dclk_min, + dclk_max); + if (ret) + return ret; + } return ret; } @@ -1043,6 +1132,7 @@ void smu_v13_0_4_set_ppt_funcs(struct smu_context *smu) smu->message_map = smu_v13_0_4_message_map; smu->feature_map = smu_v13_0_4_feature_mask_map; smu->table_map = smu_v13_0_4_table_map; + smu->smc_driver_if_version = SMU13_0_4_DRIVER_IF_VERSION; smu->is_apu = true; if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 4)) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c index 66445964efbd1e5a94c7cfa3d2bbfca7b76e8c89..42f110602eb1a99cdbdc7b9f86b499b8e96df3d4 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c @@ -831,6 +831,8 @@ static int smu_v13_0_5_set_soft_freq_limited_range(struct smu_context *smu, uint32_t max) { enum smu_message_type msg_set_min, msg_set_max; + uint32_t min_clk = min; + uint32_t max_clk = max; int ret = 0; if (!smu_v13_0_5_clk_dpm_is_enabled(smu, clk_type)) @@ -851,11 +853,16 @@ static int smu_v13_0_5_set_soft_freq_limited_range(struct smu_context *smu, return -EINVAL; } - ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_min, min, NULL); + if (clk_type == SMU_VCLK) { + min_clk = min << SMU_13_VCLK_SHIFT; + max_clk = max << SMU_13_VCLK_SHIFT; + } + + ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_min, min_clk, NULL); if (ret) goto out; - ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_max, max, NULL); + ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_max, max_clk, NULL); if (ret) goto out; @@ -866,7 +873,7 @@ static int smu_v13_0_5_set_soft_freq_limited_range(struct smu_context *smu, static int smu_v13_0_5_print_clk_levels(struct smu_context *smu, enum smu_clk_type clk_type, char *buf) { - int i, size = 0, ret = 0; + int i, idx, size = 0, ret = 0; uint32_t cur_value = 0, value = 0, count = 0; uint32_t min = 0, max = 0; @@ -898,7 +905,8 @@ static int smu_v13_0_5_print_clk_levels(struct smu_context *smu, goto print_clk_out; for (i = 0; i < count; i++) { - ret = smu_v13_0_5_get_dpm_freq_by_index(smu, clk_type, i, &value); + idx = (clk_type == SMU_MCLK) ? (count - i - 1) : i; + ret = smu_v13_0_5_get_dpm_freq_by_index(smu, clk_type, idx, &value); if (ret) goto print_clk_out; @@ -970,31 +978,79 @@ static int smu_v13_0_5_force_clk_levels(struct smu_context *smu, return ret; } +static int smu_v13_0_5_get_dpm_profile_freq(struct smu_context *smu, + enum amd_dpm_forced_level level, + enum smu_clk_type clk_type, + uint32_t *min_clk, + uint32_t *max_clk) +{ + int ret = 0; + uint32_t clk_limit = 0; + + switch (clk_type) { + case SMU_GFXCLK: + case SMU_SCLK: + clk_limit = SMU_13_0_5_UMD_PSTATE_GFXCLK; + if (level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK) + smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_SCLK, NULL, &clk_limit); + else if (level == AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK) + smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_SCLK, &clk_limit, NULL); + break; + case SMU_VCLK: + smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_VCLK, NULL, &clk_limit); + break; + case SMU_DCLK: + smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_DCLK, NULL, &clk_limit); + break; + default: + ret = -EINVAL; + break; + } + *min_clk = *max_clk = clk_limit; + return ret; +} + static int smu_v13_0_5_set_performance_level(struct smu_context *smu, enum amd_dpm_forced_level level) { struct amdgpu_device *adev = smu->adev; uint32_t sclk_min = 0, sclk_max = 0; + uint32_t vclk_min = 0, vclk_max = 0; + uint32_t dclk_min = 0, dclk_max = 0; int ret = 0; switch (level) { case AMD_DPM_FORCED_LEVEL_HIGH: smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_SCLK, NULL, &sclk_max); + smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_VCLK, NULL, &vclk_max); + smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_DCLK, NULL, &dclk_max); sclk_min = sclk_max; + vclk_min = vclk_max; + dclk_min = dclk_max; break; case AMD_DPM_FORCED_LEVEL_LOW: smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_SCLK, &sclk_min, NULL); + smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_VCLK, &vclk_min, NULL); + smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_DCLK, &dclk_min, NULL); sclk_max = sclk_min; + vclk_max = vclk_min; + dclk_max = dclk_min; break; case AMD_DPM_FORCED_LEVEL_AUTO: smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_SCLK, &sclk_min, &sclk_max); + smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_VCLK, &vclk_min, &vclk_max); + smu_v13_0_5_get_dpm_ultimate_freq(smu, SMU_DCLK, &dclk_min, &dclk_max); break; case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD: case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK: - case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK: case AMD_DPM_FORCED_LEVEL_PROFILE_PEAK: - /* Temporarily do nothing since the optimal clocks haven't been provided yet */ + smu_v13_0_5_get_dpm_profile_freq(smu, level, SMU_SCLK, &sclk_min, &sclk_max); + smu_v13_0_5_get_dpm_profile_freq(smu, level, SMU_VCLK, &vclk_min, &vclk_max); + smu_v13_0_5_get_dpm_profile_freq(smu, level, SMU_DCLK, &dclk_min, &dclk_max); break; + case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK: + dev_err(adev->dev, "The performance level profile_min_mclk is not supported."); + return -EOPNOTSUPP; case AMD_DPM_FORCED_LEVEL_MANUAL: case AMD_DPM_FORCED_LEVEL_PROFILE_EXIT: return 0; @@ -1015,6 +1071,23 @@ static int smu_v13_0_5_set_performance_level(struct smu_context *smu, smu->gfx_actual_soft_max_freq = sclk_max; } + if (vclk_min && vclk_max) { + ret = smu_v13_0_5_set_soft_freq_limited_range(smu, + SMU_VCLK, + vclk_min, + vclk_max); + if (ret) + return ret; + } + + if (dclk_min && dclk_max) { + ret = smu_v13_0_5_set_soft_freq_limited_range(smu, + SMU_DCLK, + dclk_min, + dclk_max); + if (ret) + return ret; + } return ret; } @@ -1068,6 +1141,7 @@ void smu_v13_0_5_set_ppt_funcs(struct smu_context *smu) smu->feature_map = smu_v13_0_5_feature_mask_map; smu->table_map = smu_v13_0_5_table_map; smu->is_apu = true; + smu->smc_driver_if_version = SMU13_0_5_DRIVER_IF_VERSION; smu->param_reg = SOC15_REG_OFFSET(MP1, 0, mmMP1_C2PMSG_34); smu->msg_reg = SOC15_REG_OFFSET(MP1, 0, mmMP1_C2PMSG_2); smu->resp_reg = SOC15_REG_OFFSET(MP1, 0, mmMP1_C2PMSG_33); diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.h index 40bc0f8e6d610b95140de65a1977866cda64a436..263cd651855eacbb0c0b1e0c6a05189564c7a515 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.h +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.h @@ -24,6 +24,6 @@ #define __SMU_V13_0_5_PPT_H__ extern void smu_v13_0_5_set_ppt_funcs(struct smu_context *smu); -#define SMU_13_0_5_UMD_PSTATE_GFXCLK 1100 +#define SMU_13_0_5_UMD_PSTATE_GFXCLK 700 #endif diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c index ea8f3d6fb98b3fdbd5614b34a94082a4eea9bfea..a92ea4601ea4740d4fb14223f2d605316ccc4017 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c @@ -80,7 +80,10 @@ /* possible frequency drift (1Mhz) */ #define EPSILON 1 -#define smnPCIE_ESM_CTRL 0x111003D0 +#define smnPCIE_ESM_CTRL 0x193D0 +#define smnPCIE_LC_LINK_WIDTH_CNTL 0x1ab40288 +#define PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD_MASK 0x00000070L +#define PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD__SHIFT 0x4 static const struct cmn2asic_msg_mapping smu_v13_0_6_message_map[SMU_MSG_MAX_COUNT] = { MSG_MAP(TestMessage, PPSMC_MSG_TestMessage, 0), @@ -122,6 +125,7 @@ static const struct cmn2asic_msg_mapping smu_v13_0_6_message_map[SMU_MSG_MAX_COU MSG_MAP(GetMaxGfxclkFrequency, PPSMC_MSG_GetMaxGfxDpmFreq, 0), MSG_MAP(SetSoftMinGfxclk, PPSMC_MSG_SetSoftMinGfxClk, 0), MSG_MAP(SetSoftMaxGfxClk, PPSMC_MSG_SetSoftMaxGfxClk, 0), + MSG_MAP(PrepareMp1ForUnload, PPSMC_MSG_PrepareForDriverUnload, 0), }; static const struct cmn2asic_mapping smu_v13_0_6_clk_map[SMU_CLK_COUNT] = { @@ -171,18 +175,12 @@ static const struct cmn2asic_mapping smu_v13_0_6_table_map[SMU_TABLE_COUNT] = { TAB_MAP(I2C_COMMANDS), }; -#define THROTTLER_PROCHOT_GFX_BIT 0 -#define THROTTLER_PPT_BIT 1 -#define THROTTLER_TEMP_SOC_BIT 2 -#define THROTTLER_TEMP_VR_GFX_BIT 3 -#define THROTTLER_TEMP_HBM_BIT 4 - static const uint8_t smu_v13_0_6_throttler_map[] = { [THROTTLER_PPT_BIT] = (SMU_THROTTLER_PPT0_BIT), - [THROTTLER_TEMP_SOC_BIT] = (SMU_THROTTLER_TEMP_GPU_BIT), - [THROTTLER_TEMP_HBM_BIT] = (SMU_THROTTLER_TEMP_MEM_BIT), - [THROTTLER_TEMP_VR_GFX_BIT] = (SMU_THROTTLER_TEMP_VR_GFX_BIT), - [THROTTLER_PROCHOT_GFX_BIT] = (SMU_THROTTLER_PROCHOT_GFX_BIT), + [THROTTLER_THERMAL_SOCKET_BIT] = (SMU_THROTTLER_TEMP_GPU_BIT), + [THROTTLER_THERMAL_HBM_BIT] = (SMU_THROTTLER_TEMP_MEM_BIT), + [THROTTLER_THERMAL_VR_BIT] = (SMU_THROTTLER_TEMP_VR_GFX_BIT), + [THROTTLER_PROCHOT_BIT] = (SMU_THROTTLER_PROCHOT_GFX_BIT), }; struct PPTable_t { @@ -197,10 +195,12 @@ struct PPTable_t { uint32_t LclkFrequencyTable[4]; uint32_t MaxLclkDpmRange; uint32_t MinLclkDpmRange; + uint64_t PublicSerialNumber_AID; bool Init; }; #define SMUQ10_TO_UINT(x) ((x) >> 10) +#define SMUQ16_TO_UINT(x) ((x) >> 16) struct smu_v13_0_6_dpm_map { enum smu_clk_type clk_type; @@ -220,10 +220,12 @@ static int smu_v13_0_6_tables_init(struct smu_context *smu) PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); SMU_TABLE_INIT(tables, SMU_TABLE_SMU_METRICS, sizeof(MetricsTable_t), - PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); + PAGE_SIZE, + AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT); SMU_TABLE_INIT(tables, SMU_TABLE_I2C_COMMANDS, sizeof(SwI2cRequest_t), - PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); + PAGE_SIZE, + AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT); smu_table->metrics_table = kzalloc(sizeof(MetricsTable_t), GFP_KERNEL); if (!smu_table->metrics_table) @@ -355,6 +357,9 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu) SMUQ10_TO_UINT(metrics->LclkFrequencyTable[i]); } + /* use AID0 serial number by default */ + pptable->PublicSerialNumber_AID = metrics->PublicSerialNumber_AID[0]; + pptable->Init = true; } @@ -385,7 +390,7 @@ static int smu_v13_0_6_get_dpm_ultimate_freq(struct smu_context *smu, break; case SMU_SOCCLK: if (pptable->Init) - clock_limit = pptable->UclkFrequencyTable[0]; + clock_limit = pptable->SocclkFrequencyTable[0]; break; case SMU_FCLK: if (pptable->Init) @@ -638,16 +643,14 @@ static int smu_v13_0_6_freqs_in_same_level(int32_t frequency1, return (abs(frequency1 - frequency2) <= EPSILON); } -static uint32_t smu_v13_0_6_get_throttler_status(struct smu_context *smu, - MetricsTable_t *metrics) +static uint32_t smu_v13_0_6_get_throttler_status(struct smu_context *smu) { + struct smu_power_context *smu_power = &smu->smu_power; + struct smu_13_0_power_context *power_context = smu_power->power_context; uint32_t throttler_status = 0; - throttler_status |= metrics->ProchotResidencyAcc > 0 ? 1U << THROTTLER_PROCHOT_GFX_BIT : 0; - throttler_status |= metrics->PptResidencyAcc > 0 ? 1U << THROTTLER_PPT_BIT : 0; - throttler_status |= metrics->SocketThmResidencyAcc > 0 ? 1U << THROTTLER_TEMP_SOC_BIT : 0; - throttler_status |= metrics->VrThmResidencyAcc > 0 ? 1U << THROTTLER_TEMP_VR_GFX_BIT : 0; - throttler_status |= metrics->HbmThmResidencyAcc > 0 ? 1U << THROTTLER_TEMP_HBM_BIT : 0; + throttler_status = atomic_read(&power_context->throttle_status); + dev_dbg(smu->adev->dev, "SMU Throttler status: %u", throttler_status); return throttler_status; } @@ -658,7 +661,10 @@ static int smu_v13_0_6_get_smu_metrics_data(struct smu_context *smu, { struct smu_table_context *smu_table = &smu->smu_table; MetricsTable_t *metrics = (MetricsTable_t *)smu_table->metrics_table; + struct amdgpu_device *adev = smu->adev; + uint32_t smu_version; int ret = 0; + int xcc_id; ret = smu_v13_0_6_get_metrics_table(smu, NULL, false); if (ret) @@ -668,7 +674,13 @@ static int smu_v13_0_6_get_smu_metrics_data(struct smu_context *smu, switch (member) { case METRICS_CURR_GFXCLK: case METRICS_AVERAGE_GFXCLK: - *value = 0; + smu_cmn_get_smc_version(smu, NULL, &smu_version); + if (smu_version >= 0x552F00) { + xcc_id = GET_INST(GC, 0); + *value = SMUQ10_TO_UINT(metrics->GfxclkFrequency[xcc_id]); + } else { + *value = 0; + } break; case METRICS_CURR_SOCCLK: case METRICS_AVERAGE_SOCCLK: @@ -708,9 +720,6 @@ static int smu_v13_0_6_get_smu_metrics_data(struct smu_context *smu, case METRICS_TEMPERATURE_VRSOC: *value = SMUQ10_TO_UINT(metrics->MaxVrTemperature); break; - case METRICS_THROTTLER_STATUS: - *value = smu_v13_0_6_get_throttler_status(smu, metrics); - break; default: *value = UINT_MAX; break; @@ -1246,21 +1255,6 @@ static int smu_v13_0_6_get_power_limit(struct smu_context *smu, uint32_t power_limit = 0; int ret; - if (!smu_cmn_feature_is_enabled(smu, SMU_FEATURE_PPT_BIT)) { - if (current_power_limit) - *current_power_limit = 0; - if (default_power_limit) - *default_power_limit = 0; - if (max_power_limit) - *max_power_limit = 0; - - dev_warn( - smu->adev->dev, - "PPT feature is not enabled, power values can't be fetched."); - - return 0; - } - ret = smu_cmn_send_smc_msg(smu, SMU_MSG_GetPptLimit, &power_limit); if (ret) { @@ -1287,16 +1281,147 @@ static int smu_v13_0_6_set_power_limit(struct smu_context *smu, return smu_v13_0_set_power_limit(smu, limit_type, limit); } -static int smu_v13_0_6_system_features_control(struct smu_context *smu, - bool enable) +static int smu_v13_0_6_irq_process(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + struct amdgpu_iv_entry *entry) { - int ret; + struct smu_context *smu = adev->powerplay.pp_handle; + struct smu_power_context *smu_power = &smu->smu_power; + struct smu_13_0_power_context *power_context = smu_power->power_context; + uint32_t client_id = entry->client_id; + uint32_t ctxid = entry->src_data[0]; + uint32_t src_id = entry->src_id; + uint32_t data; + + if (client_id == SOC15_IH_CLIENTID_MP1) { + if (src_id == IH_INTERRUPT_ID_TO_DRIVER) { + /* ACK SMUToHost interrupt */ + data = RREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL); + data = REG_SET_FIELD(data, MP1_SMN_IH_SW_INT_CTRL, INT_ACK, 1); + WREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL, data); + /* + * ctxid is used to distinguish different events for SMCToHost + * interrupt. + */ + switch (ctxid) { + case IH_INTERRUPT_CONTEXT_ID_THERMAL_THROTTLING: + /* + * Increment the throttle interrupt counter + */ + atomic64_inc(&smu->throttle_int_counter); + + if (!atomic_read(&adev->throttling_logging_enabled)) + return 0; + + /* This uses the new method which fixes the + * incorrect throttling status reporting + * through metrics table. For older FWs, + * it will be ignored. + */ + if (__ratelimit(&adev->throttling_logging_rs)) { + atomic_set( + &power_context->throttle_status, + entry->src_data[1]); + schedule_work(&smu->throttling_logging_work); + } + + break; + } + } + } + + return 0; +} + +static int smu_v13_0_6_set_irq_state(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + unsigned tyep, + enum amdgpu_interrupt_state state) +{ + uint32_t val = 0; + + switch (state) { + case AMDGPU_IRQ_STATE_DISABLE: + /* For MP1 SW irqs */ + val = RREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL); + val = REG_SET_FIELD(val, MP1_SMN_IH_SW_INT_CTRL, INT_MASK, 1); + WREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL, val); - /* Nothing to be done for APU */ - if (smu->adev->flags & AMD_IS_APU) + break; + case AMDGPU_IRQ_STATE_ENABLE: + /* For MP1 SW irqs */ + val = RREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT); + val = REG_SET_FIELD(val, MP1_SMN_IH_SW_INT, ID, 0xFE); + val = REG_SET_FIELD(val, MP1_SMN_IH_SW_INT, VALID, 0); + WREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT, val); + + val = RREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL); + val = REG_SET_FIELD(val, MP1_SMN_IH_SW_INT_CTRL, INT_MASK, 0); + WREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL, val); + + break; + default: + break; + } + + return 0; +} + +static const struct amdgpu_irq_src_funcs smu_v13_0_6_irq_funcs = +{ + .set = smu_v13_0_6_set_irq_state, + .process = smu_v13_0_6_irq_process, +}; + +static int smu_v13_0_6_register_irq_handler(struct smu_context *smu) +{ + struct amdgpu_device *adev = smu->adev; + struct amdgpu_irq_src *irq_src = &smu->irq_source; + int ret = 0; + + if (amdgpu_sriov_vf(adev)) return 0; - ret = smu_v13_0_system_features_control(smu, enable); + irq_src->num_types = 1; + irq_src->funcs = &smu_v13_0_6_irq_funcs; + + ret = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_MP1, + IH_INTERRUPT_ID_TO_DRIVER, + irq_src); + if (ret) + return ret; + + return ret; +} + +static int smu_v13_0_6_notify_unload(struct smu_context *smu) +{ + uint32_t smu_version; + + smu_cmn_get_smc_version(smu, NULL, &smu_version); + if (smu_version <= 0x553500) + return 0; + + dev_dbg(smu->adev->dev, "Notify PMFW about driver unload"); + /* Ignore return, just intimate FW that driver is not going to be there */ + smu_cmn_send_smc_msg(smu, SMU_MSG_PrepareMp1ForUnload, NULL); + + return 0; +} + +static int smu_v13_0_6_system_features_control(struct smu_context *smu, + bool enable) +{ + struct amdgpu_device *adev = smu->adev; + int ret = 0; + + if (enable) { + if (!(adev->flags & AMD_IS_APU)) + ret = smu_v13_0_system_features_control(smu, enable); + } else { + /* Notify FW that the device is no longer driver managed */ + smu_v13_0_6_notify_unload(smu); + } return ret; } @@ -1737,19 +1862,11 @@ static void smu_v13_0_6_i2c_control_fini(struct smu_context *smu) static void smu_v13_0_6_get_unique_id(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; - //SmuMetrics_t *metrics = smu->smu_table.metrics_table; - uint32_t upper32 = 0, lower32 = 0; - int ret; - - ret = smu_cmn_get_metrics_table(smu, NULL, false); - if (ret) - goto out; - - //upper32 = metrics->PublicSerialNumUpper32; - //lower32 = metrics->PublicSerialNumLower32; + struct smu_table_context *smu_table = &smu->smu_table; + struct PPTable_t *pptable = + (struct PPTable_t *)smu_table->driver_pptable; -out: - adev->unique_id = ((uint64_t)upper32 << 32) | lower32; + adev->unique_id = pptable->PublicSerialNumber_AID; if (adev->serial[0] == '\0') sprintf(adev->serial, "%016llx", adev->unique_id); } @@ -1774,37 +1891,35 @@ static int smu_v13_0_6_allow_xgmi_power_down(struct smu_context *smu, bool en) en ? 0 : 1, NULL); } -static const struct throttling_logging_label { - uint32_t feature_mask; - const char *label; -} logging_label[] = { - { (1U << THROTTLER_TEMP_HBM_BIT), "HBM" }, - { (1U << THROTTLER_TEMP_SOC_BIT), "SOC" }, - { (1U << THROTTLER_TEMP_VR_GFX_BIT), "VR limit" }, +static const char *const throttling_logging_label[] = { + [THROTTLER_PROCHOT_BIT] = "Prochot", + [THROTTLER_PPT_BIT] = "PPT", + [THROTTLER_THERMAL_SOCKET_BIT] = "SOC", + [THROTTLER_THERMAL_VR_BIT] = "VR", + [THROTTLER_THERMAL_HBM_BIT] = "HBM" }; + static void smu_v13_0_6_log_thermal_throttling_event(struct smu_context *smu) { - int ret; int throttler_idx, throtting_events = 0, buf_idx = 0; struct amdgpu_device *adev = smu->adev; uint32_t throttler_status; char log_buf[256]; - ret = smu_v13_0_6_get_smu_metrics_data(smu, METRICS_THROTTLER_STATUS, - &throttler_status); - if (ret) + throttler_status = smu_v13_0_6_get_throttler_status(smu); + if (!throttler_status) return; memset(log_buf, 0, sizeof(log_buf)); - for (throttler_idx = 0; throttler_idx < ARRAY_SIZE(logging_label); + for (throttler_idx = 0; + throttler_idx < ARRAY_SIZE(throttling_logging_label); throttler_idx++) { - if (throttler_status & - logging_label[throttler_idx].feature_mask) { + if (throttler_status & (1U << throttler_idx)) { throtting_events++; - buf_idx += snprintf(log_buf + buf_idx, - sizeof(log_buf) - buf_idx, "%s%s", - throtting_events > 1 ? " and " : "", - logging_label[throttler_idx].label); + buf_idx += snprintf( + log_buf + buf_idx, sizeof(log_buf) - buf_idx, + "%s%s", throtting_events > 1 ? " and " : "", + throttling_logging_label[throttler_idx]); if (buf_idx >= sizeof(log_buf)) { dev_err(adev->dev, "buffer overflow!\n"); log_buf[sizeof(log_buf) - 1] = '\0'; @@ -1813,16 +1928,24 @@ static void smu_v13_0_6_log_thermal_throttling_event(struct smu_context *smu) } } - dev_warn( - adev->dev, - "WARN: GPU thermal throttling temperature reached, expect performance decrease. %s.\n", - log_buf); + dev_warn(adev->dev, + "WARN: GPU is throttled, expect performance decrease. %s.\n", + log_buf); kgd2kfd_smi_event_throttle( smu->adev->kfd.dev, smu_cmn_get_indep_throttler_status(throttler_status, smu_v13_0_6_throttler_map)); } +static int +smu_v13_0_6_get_current_pcie_link_width_level(struct smu_context *smu) +{ + struct amdgpu_device *adev = smu->adev; + + return REG_GET_FIELD(RREG32_PCIE(smnPCIE_LC_LINK_WIDTH_CNTL), + PCIE_LC_LINK_WIDTH_CNTL, LC_LINK_WIDTH_RD); +} + static int smu_v13_0_6_get_current_pcie_link_speed(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; @@ -1841,8 +1964,12 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table struct smu_table_context *smu_table = &smu->smu_table; struct gpu_metrics_v1_3 *gpu_metrics = (struct gpu_metrics_v1_3 *)smu_table->gpu_metrics_table; + struct amdgpu_device *adev = smu->adev; + int ret = 0, inst0, xcc0; MetricsTable_t *metrics; - int i, ret = 0; + + inst0 = adev->sdma.instance[0].aid_id; + xcc0 = GET_INST(GC, 0); metrics = kzalloc(sizeof(MetricsTable_t), GFP_KERNEL); ret = smu_v13_0_6_get_metrics_table(smu, metrics, true); @@ -1851,51 +1978,59 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table smu_cmn_init_soft_gpu_metrics(gpu_metrics, 1, 3); - /* TODO: Decide on how to fill in zero value fields */ - gpu_metrics->temperature_edge = 0; - gpu_metrics->temperature_hotspot = 0; - gpu_metrics->temperature_mem = 0; - gpu_metrics->temperature_vrgfx = 0; - gpu_metrics->temperature_vrsoc = 0; - gpu_metrics->temperature_vrmem = 0; - - gpu_metrics->average_gfx_activity = 0; - gpu_metrics->average_umc_activity = 0; - gpu_metrics->average_mm_activity = 0; - - gpu_metrics->average_socket_power = 0; - gpu_metrics->energy_accumulator = 0; - - gpu_metrics->average_gfxclk_frequency = 0; - gpu_metrics->average_socclk_frequency = 0; - gpu_metrics->average_uclk_frequency = 0; - gpu_metrics->average_vclk0_frequency = 0; - gpu_metrics->average_dclk0_frequency = 0; - - gpu_metrics->current_gfxclk = 0; - gpu_metrics->current_socclk = 0; - gpu_metrics->current_uclk = 0; - gpu_metrics->current_vclk0 = 0; - gpu_metrics->current_dclk0 = 0; - + gpu_metrics->temperature_hotspot = + SMUQ10_TO_UINT(metrics->MaxSocketTemperature); + /* Individual HBM stack temperature is not reported */ + gpu_metrics->temperature_mem = + SMUQ10_TO_UINT(metrics->MaxHbmTemperature); + /* Reports max temperature of all voltage rails */ + gpu_metrics->temperature_vrsoc = + SMUQ10_TO_UINT(metrics->MaxVrTemperature); + + gpu_metrics->average_gfx_activity = + SMUQ10_TO_UINT(metrics->SocketGfxBusy); + gpu_metrics->average_umc_activity = + SMUQ10_TO_UINT(metrics->DramBandwidthUtilization); + + gpu_metrics->average_socket_power = + SMUQ10_TO_UINT(metrics->SocketPower); + gpu_metrics->energy_accumulator = + SMUQ16_TO_UINT(metrics->SocketEnergyAcc); + + gpu_metrics->current_gfxclk = + SMUQ10_TO_UINT(metrics->GfxclkFrequency[xcc0]); + gpu_metrics->current_socclk = + SMUQ10_TO_UINT(metrics->SocclkFrequency[inst0]); + gpu_metrics->current_uclk = SMUQ10_TO_UINT(metrics->UclkFrequency); + gpu_metrics->current_vclk0 = + SMUQ10_TO_UINT(metrics->VclkFrequency[inst0]); + gpu_metrics->current_dclk0 = + SMUQ10_TO_UINT(metrics->DclkFrequency[inst0]); + + gpu_metrics->average_gfxclk_frequency = gpu_metrics->current_gfxclk; + gpu_metrics->average_socclk_frequency = gpu_metrics->current_socclk; + gpu_metrics->average_uclk_frequency = gpu_metrics->current_uclk; + gpu_metrics->average_vclk0_frequency = gpu_metrics->current_vclk0; + gpu_metrics->average_dclk0_frequency = gpu_metrics->current_dclk0; + + /* Throttle status is not reported through metrics now */ gpu_metrics->throttle_status = 0; - gpu_metrics->indep_throttle_status = smu_cmn_get_indep_throttler_status( - gpu_metrics->throttle_status, smu_v13_0_6_throttler_map); - gpu_metrics->current_fan_speed = 0; - - gpu_metrics->pcie_link_width = 0; - gpu_metrics->pcie_link_speed = smu_v13_0_6_get_current_pcie_link_speed(smu); + if (!(adev->flags & AMD_IS_APU)) { + gpu_metrics->pcie_link_width = + smu_v13_0_6_get_current_pcie_link_width_level(smu); + gpu_metrics->pcie_link_speed = + smu_v13_0_6_get_current_pcie_link_speed(smu); + } gpu_metrics->system_clock_counter = ktime_get_boottime_ns(); - gpu_metrics->gfx_activity_acc = 0; - gpu_metrics->mem_activity_acc = 0; + gpu_metrics->gfx_activity_acc = + SMUQ10_TO_UINT(metrics->SocketGfxBusyAcc); + gpu_metrics->mem_activity_acc = + SMUQ10_TO_UINT(metrics->DramBandwidthUtilizationAcc); - for (i = 0; i < NUM_HBM_INSTANCES; i++) - gpu_metrics->temperature_hbm[i] = 0; - - gpu_metrics->firmware_timestamp = 0; + gpu_metrics->firmware_timestamp = metrics->Timestamp; *table = (void *)gpu_metrics; kfree(metrics); @@ -1905,27 +2040,27 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table static int smu_v13_0_6_mode2_reset(struct smu_context *smu) { - u32 smu_version; int ret = 0, index; struct amdgpu_device *adev = smu->adev; int timeout = 10; - smu_cmn_get_smc_version(smu, NULL, &smu_version); - index = smu_cmn_to_asic_specific_index(smu, CMN2ASIC_MAPPING_MSG, SMU_MSG_GfxDeviceDriverReset); mutex_lock(&smu->message_lock); + ret = smu_cmn_send_msg_without_waiting(smu, (uint16_t)index, SMU_RESET_MODE_2); + /* This is similar to FLR, wait till max FLR timeout */ msleep(100); + dev_dbg(smu->adev->dev, "restore config space...\n"); /* Restore the config space saved during init */ amdgpu_device_load_pci_state(adev->pdev); dev_dbg(smu->adev->dev, "wait for reset ack\n"); - while (ret == -ETIME && timeout) { + do { ret = smu_cmn_wait_for_response(smu); /* Wait a bit more time for getting ACK */ if (ret == -ETIME) { @@ -1934,16 +2069,14 @@ static int smu_v13_0_6_mode2_reset(struct smu_context *smu) continue; } - if (ret != 1) { + if (ret) { dev_err(adev->dev, - "failed to send mode2 message \tparam: 0x%08x response %#x\n", + "failed to send mode2 message \tparam: 0x%08x error code %d\n", SMU_RESET_MODE_2, ret); goto out; } - } + } while (ret == -ETIME && timeout); - if (ret == 1) - ret = 0; out: mutex_unlock(&smu->message_lock); @@ -2032,11 +2165,9 @@ static const struct pptable_funcs smu_v13_0_6_ppt_funcs = { .feature_is_enabled = smu_cmn_feature_is_enabled, .set_power_limit = smu_v13_0_6_set_power_limit, .set_xgmi_pstate = smu_v13_0_set_xgmi_pstate, - /* TODO: Thermal limits unknown, skip these for now - .register_irq_handler = smu_v13_0_register_irq_handler, + .register_irq_handler = smu_v13_0_6_register_irq_handler, .enable_thermal_alert = smu_v13_0_enable_thermal_alert, .disable_thermal_alert = smu_v13_0_disable_thermal_alert, - */ .setup_pptable = smu_v13_0_6_setup_pptable, .baco_is_support = smu_v13_0_6_is_baco_supported, .get_dpm_ultimate_freq = smu_v13_0_6_get_dpm_ultimate_freq, @@ -2065,5 +2196,6 @@ void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu) smu->clock_map = smu_v13_0_6_clk_map; smu->feature_map = smu_v13_0_6_feature_mask_map; smu->table_map = smu_v13_0_6_table_map; + smu->smc_driver_if_version = SMU13_0_6_DRIVER_IF_VERSION; smu_v13_0_set_smu_mailbox_registers(smu); } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c index 3d9ff46706fb704855615f353165a353b43adf96..cda4e818aab7e9381273f5ee4739d88f54a41305 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c @@ -125,6 +125,7 @@ static struct cmn2asic_msg_mapping smu_v13_0_7_message_map[SMU_MSG_MAX_COUNT] = MSG_MAP(ArmD3, PPSMC_MSG_ArmD3, 0), MSG_MAP(AllowGpo, PPSMC_MSG_SetGpoAllow, 0), MSG_MAP(GetPptLimit, PPSMC_MSG_GetPptLimit, 0), + MSG_MAP(NotifyPowerSource, PPSMC_MSG_NotifyPowerSource, 0), }; static struct cmn2asic_mapping smu_v13_0_7_clk_map[SMU_CLK_COUNT] = { @@ -205,6 +206,7 @@ static struct cmn2asic_mapping smu_v13_0_7_table_map[SMU_TABLE_COUNT] = { TAB_MAP(DRIVER_SMU_CONFIG), TAB_MAP(ACTIVITY_MONITOR_COEFF), [SMU_TABLE_COMBO_PPTABLE] = {1, TABLE_COMBO_PPTABLE}, + TAB_MAP(OVERDRIVE), }; static struct cmn2asic_mapping smu_v13_0_7_pwr_src_map[SMU_POWER_SOURCE_COUNT] = { @@ -321,6 +323,10 @@ static int smu_v13_0_7_check_powerplay_table(struct smu_context *smu) struct smu_baco_context *smu_baco = &smu->smu_baco; PPTable_t *smc_pptable = table_context->driver_pptable; BoardTable_t *BoardTable = &smc_pptable->BoardTable; + const OverDriveLimits_t * const overdrive_upperlimits = + &smc_pptable->SkuTable.OverDriveLimitsBasicMax; + const OverDriveLimits_t * const overdrive_lowerlimits = + &smc_pptable->SkuTable.OverDriveLimitsMin; if (powerplay_table->platform_caps & SMU_13_0_7_PP_PLATFORM_CAP_HARDWAREDC) smu->dc_controlled_by_gpio = true; @@ -332,6 +338,10 @@ static int smu_v13_0_7_check_powerplay_table(struct smu_context *smu) if (smu_baco->platform_support && (BoardTable->HsrEnabled || BoardTable->VddqOffEnabled)) smu_baco->maco_support = true; + if (!overdrive_lowerlimits->FeatureCtrlMask || + !overdrive_upperlimits->FeatureCtrlMask) + smu->od_enabled = false; + table_context->thermal_controller_type = powerplay_table->thermal_controller_type; @@ -478,7 +488,7 @@ static int smu_v13_0_7_tables_init(struct smu_context *smu) PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); SMU_TABLE_INIT(tables, SMU_TABLE_I2C_COMMANDS, sizeof(SwI2cRequest_t), PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); - SMU_TABLE_INIT(tables, SMU_TABLE_OVERDRIVE, sizeof(OverDriveTable_t), + SMU_TABLE_INIT(tables, SMU_TABLE_OVERDRIVE, sizeof(OverDriveTableExternal_t), PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); SMU_TABLE_INIT(tables, SMU_TABLE_PMSTATUSLOG, SMU13_TOOL_SIZE, PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); @@ -1012,16 +1022,118 @@ static int smu_v13_0_7_get_current_clk_freq_by_table(struct smu_context *smu, value); } +static bool smu_v13_0_7_is_od_feature_supported(struct smu_context *smu, + int od_feature_bit) +{ + PPTable_t *pptable = smu->smu_table.driver_pptable; + const OverDriveLimits_t * const overdrive_upperlimits = + &pptable->SkuTable.OverDriveLimitsBasicMax; + + return overdrive_upperlimits->FeatureCtrlMask & (1U << od_feature_bit); +} + +static void smu_v13_0_7_get_od_setting_limits(struct smu_context *smu, + int od_feature_bit, + bool lower_boundary, + int32_t *min, + int32_t *max) +{ + PPTable_t *pptable = smu->smu_table.driver_pptable; + const OverDriveLimits_t * const overdrive_upperlimits = + &pptable->SkuTable.OverDriveLimitsBasicMax; + const OverDriveLimits_t * const overdrive_lowerlimits = + &pptable->SkuTable.OverDriveLimitsMin; + int32_t od_min_setting, od_max_setting; + + switch (od_feature_bit) { + case PP_OD_FEATURE_GFXCLK_BIT: + if (lower_boundary) { + od_min_setting = overdrive_lowerlimits->GfxclkFmin; + od_max_setting = overdrive_upperlimits->GfxclkFmin; + } else { + od_min_setting = overdrive_lowerlimits->GfxclkFmax; + od_max_setting = overdrive_upperlimits->GfxclkFmax; + } + break; + case PP_OD_FEATURE_UCLK_BIT: + if (lower_boundary) { + od_min_setting = overdrive_lowerlimits->UclkFmin; + od_max_setting = overdrive_upperlimits->UclkFmin; + } else { + od_min_setting = overdrive_lowerlimits->UclkFmax; + od_max_setting = overdrive_upperlimits->UclkFmax; + } + break; + case PP_OD_FEATURE_GFX_VF_CURVE_BIT: + od_min_setting = overdrive_lowerlimits->VoltageOffsetPerZoneBoundary; + od_max_setting = overdrive_upperlimits->VoltageOffsetPerZoneBoundary; + break; + default: + break; + } + + if (min) + *min = od_min_setting; + if (max) + *max = od_max_setting; +} + +static void smu_v13_0_7_dump_od_table(struct smu_context *smu, + OverDriveTableExternal_t *od_table) +{ + struct amdgpu_device *adev = smu->adev; + + dev_dbg(adev->dev, "OD: Gfxclk: (%d, %d)\n", od_table->OverDriveTable.GfxclkFmin, + od_table->OverDriveTable.GfxclkFmax); + dev_dbg(adev->dev, "OD: Uclk: (%d, %d)\n", od_table->OverDriveTable.UclkFmin, + od_table->OverDriveTable.UclkFmax); +} + +static int smu_v13_0_7_get_overdrive_table(struct smu_context *smu, + OverDriveTableExternal_t *od_table) +{ + int ret = 0; + + ret = smu_cmn_update_table(smu, + SMU_TABLE_OVERDRIVE, + 0, + (void *)od_table, + false); + if (ret) + dev_err(smu->adev->dev, "Failed to get overdrive table!\n"); + + return ret; +} + +static int smu_v13_0_7_upload_overdrive_table(struct smu_context *smu, + OverDriveTableExternal_t *od_table) +{ + int ret = 0; + + ret = smu_cmn_update_table(smu, + SMU_TABLE_OVERDRIVE, + 0, + (void *)od_table, + true); + if (ret) + dev_err(smu->adev->dev, "Failed to upload overdrive table!\n"); + + return ret; +} + static int smu_v13_0_7_print_clk_levels(struct smu_context *smu, enum smu_clk_type clk_type, char *buf) { struct smu_dpm_context *smu_dpm = &smu->smu_dpm; struct smu_13_0_dpm_context *dpm_context = smu_dpm->dpm_context; + OverDriveTableExternal_t *od_table = + (OverDriveTableExternal_t *)smu->smu_table.overdrive_table; struct smu_13_0_dpm_table *single_dpm_table; struct smu_13_0_pcie_table *pcie_table; uint32_t gen_speed, lane_width; int i, curr_freq, size = 0; + int32_t min_value, max_value; int ret = 0; smu_cmn_get_sysfs_buf(&buf, &size); @@ -1138,6 +1250,89 @@ static int smu_v13_0_7_print_clk_levels(struct smu_context *smu, "*" : ""); break; + case SMU_OD_SCLK: + if (!smu_v13_0_7_is_od_feature_supported(smu, + PP_OD_FEATURE_GFXCLK_BIT)) + break; + + size += sysfs_emit_at(buf, size, "OD_SCLK:\n"); + size += sysfs_emit_at(buf, size, "0: %uMhz\n1: %uMhz\n", + od_table->OverDriveTable.GfxclkFmin, + od_table->OverDriveTable.GfxclkFmax); + break; + + case SMU_OD_MCLK: + if (!smu_v13_0_7_is_od_feature_supported(smu, + PP_OD_FEATURE_UCLK_BIT)) + break; + + size += sysfs_emit_at(buf, size, "OD_MCLK:\n"); + size += sysfs_emit_at(buf, size, "0: %uMhz\n1: %uMHz\n", + od_table->OverDriveTable.UclkFmin, + od_table->OverDriveTable.UclkFmax); + break; + + case SMU_OD_VDDC_CURVE: + if (!smu_v13_0_7_is_od_feature_supported(smu, + PP_OD_FEATURE_GFX_VF_CURVE_BIT)) + break; + + size += sysfs_emit_at(buf, size, "OD_VDDC_CURVE:\n"); + for (i = 0; i < PP_NUM_OD_VF_CURVE_POINTS; i++) + size += sysfs_emit_at(buf, size, "%d: %dmv\n", + i, + od_table->OverDriveTable.VoltageOffsetPerZoneBoundary[i]); + break; + + case SMU_OD_RANGE: + if (!smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_GFXCLK_BIT) && + !smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_UCLK_BIT) && + !smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_GFX_VF_CURVE_BIT)) + break; + + size += sysfs_emit_at(buf, size, "%s:\n", "OD_RANGE"); + + if (smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_GFXCLK_BIT)) { + smu_v13_0_7_get_od_setting_limits(smu, + PP_OD_FEATURE_GFXCLK_BIT, + true, + &min_value, + NULL); + smu_v13_0_7_get_od_setting_limits(smu, + PP_OD_FEATURE_GFXCLK_BIT, + false, + NULL, + &max_value); + size += sysfs_emit_at(buf, size, "SCLK: %7uMhz %10uMhz\n", + min_value, max_value); + } + + if (smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_UCLK_BIT)) { + smu_v13_0_7_get_od_setting_limits(smu, + PP_OD_FEATURE_UCLK_BIT, + true, + &min_value, + NULL); + smu_v13_0_7_get_od_setting_limits(smu, + PP_OD_FEATURE_UCLK_BIT, + false, + NULL, + &max_value); + size += sysfs_emit_at(buf, size, "MCLK: %7uMhz %10uMhz\n", + min_value, max_value); + } + + if (smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_GFX_VF_CURVE_BIT)) { + smu_v13_0_7_get_od_setting_limits(smu, + PP_OD_FEATURE_GFX_VF_CURVE_BIT, + true, + &min_value, + &max_value); + size += sysfs_emit_at(buf, size, "VDDC_CURVE: %7dmv %10dmv\n", + min_value, max_value); + } + break; + default: break; } @@ -1145,6 +1340,222 @@ static int smu_v13_0_7_print_clk_levels(struct smu_context *smu, return size; } +static int smu_v13_0_7_od_edit_dpm_table(struct smu_context *smu, + enum PP_OD_DPM_TABLE_COMMAND type, + long input[], + uint32_t size) +{ + struct smu_table_context *table_context = &smu->smu_table; + OverDriveTableExternal_t *od_table = + (OverDriveTableExternal_t *)table_context->overdrive_table; + struct amdgpu_device *adev = smu->adev; + uint32_t offset_of_featurectrlmask; + int32_t minimum, maximum; + uint32_t feature_ctrlmask; + int i, ret = 0; + + switch (type) { + case PP_OD_EDIT_SCLK_VDDC_TABLE: + if (!smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_GFXCLK_BIT)) { + dev_warn(adev->dev, "GFXCLK_LIMITS setting not supported!\n"); + return -ENOTSUPP; + } + + for (i = 0; i < size; i += 2) { + if (i + 2 > size) { + dev_info(adev->dev, "invalid number of input parameters %d\n", size); + return -EINVAL; + } + + switch (input[i]) { + case 0: + smu_v13_0_7_get_od_setting_limits(smu, + PP_OD_FEATURE_GFXCLK_BIT, + true, + &minimum, + &maximum); + if (input[i + 1] < minimum || + input[i + 1] > maximum) { + dev_info(adev->dev, "GfxclkFmin (%ld) must be within [%u, %u]!\n", + input[i + 1], minimum, maximum); + return -EINVAL; + } + + od_table->OverDriveTable.GfxclkFmin = input[i + 1]; + od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_GFXCLK_BIT; + break; + + case 1: + smu_v13_0_7_get_od_setting_limits(smu, + PP_OD_FEATURE_GFXCLK_BIT, + false, + &minimum, + &maximum); + if (input[i + 1] < minimum || + input[i + 1] > maximum) { + dev_info(adev->dev, "GfxclkFmax (%ld) must be within [%u, %u]!\n", + input[i + 1], minimum, maximum); + return -EINVAL; + } + + od_table->OverDriveTable.GfxclkFmax = input[i + 1]; + od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_GFXCLK_BIT; + break; + + default: + dev_info(adev->dev, "Invalid SCLK_VDDC_TABLE index: %ld\n", input[i]); + dev_info(adev->dev, "Supported indices: [0:min,1:max]\n"); + return -EINVAL; + } + } + + if (od_table->OverDriveTable.GfxclkFmin > od_table->OverDriveTable.GfxclkFmax) { + dev_err(adev->dev, + "Invalid setting: GfxclkFmin(%u) is bigger than GfxclkFmax(%u)\n", + (uint32_t)od_table->OverDriveTable.GfxclkFmin, + (uint32_t)od_table->OverDriveTable.GfxclkFmax); + return -EINVAL; + } + break; + + case PP_OD_EDIT_MCLK_VDDC_TABLE: + if (!smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_UCLK_BIT)) { + dev_warn(adev->dev, "UCLK_LIMITS setting not supported!\n"); + return -ENOTSUPP; + } + + for (i = 0; i < size; i += 2) { + if (i + 2 > size) { + dev_info(adev->dev, "invalid number of input parameters %d\n", size); + return -EINVAL; + } + + switch (input[i]) { + case 0: + smu_v13_0_7_get_od_setting_limits(smu, + PP_OD_FEATURE_UCLK_BIT, + true, + &minimum, + &maximum); + if (input[i + 1] < minimum || + input[i + 1] > maximum) { + dev_info(adev->dev, "UclkFmin (%ld) must be within [%u, %u]!\n", + input[i + 1], minimum, maximum); + return -EINVAL; + } + + od_table->OverDriveTable.UclkFmin = input[i + 1]; + od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_UCLK_BIT; + break; + + case 1: + smu_v13_0_7_get_od_setting_limits(smu, + PP_OD_FEATURE_UCLK_BIT, + false, + &minimum, + &maximum); + if (input[i + 1] < minimum || + input[i + 1] > maximum) { + dev_info(adev->dev, "UclkFmax (%ld) must be within [%u, %u]!\n", + input[i + 1], minimum, maximum); + return -EINVAL; + } + + od_table->OverDriveTable.UclkFmax = input[i + 1]; + od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_UCLK_BIT; + break; + + default: + dev_info(adev->dev, "Invalid MCLK_VDDC_TABLE index: %ld\n", input[i]); + dev_info(adev->dev, "Supported indices: [0:min,1:max]\n"); + return -EINVAL; + } + } + + if (od_table->OverDriveTable.UclkFmin > od_table->OverDriveTable.UclkFmax) { + dev_err(adev->dev, + "Invalid setting: UclkFmin(%u) is bigger than UclkFmax(%u)\n", + (uint32_t)od_table->OverDriveTable.UclkFmin, + (uint32_t)od_table->OverDriveTable.UclkFmax); + return -EINVAL; + } + break; + + case PP_OD_EDIT_VDDC_CURVE: + if (!smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_GFX_VF_CURVE_BIT)) { + dev_warn(adev->dev, "VF curve setting not supported!\n"); + return -ENOTSUPP; + } + + if (input[0] >= PP_NUM_OD_VF_CURVE_POINTS || + input[0] < 0) + return -EINVAL; + + smu_v13_0_7_get_od_setting_limits(smu, + PP_OD_FEATURE_GFX_VF_CURVE_BIT, + true, + &minimum, + &maximum); + if (input[1] < minimum || + input[1] > maximum) { + dev_info(adev->dev, "Voltage offset (%ld) must be within [%d, %d]!\n", + input[1], minimum, maximum); + return -EINVAL; + } + + od_table->OverDriveTable.VoltageOffsetPerZoneBoundary[input[0]] = input[1]; + od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_GFX_VF_CURVE_BIT; + break; + + case PP_OD_RESTORE_DEFAULT_TABLE: + feature_ctrlmask = od_table->OverDriveTable.FeatureCtrlMask; + memcpy(od_table, + table_context->boot_overdrive_table, + sizeof(OverDriveTableExternal_t)); + od_table->OverDriveTable.FeatureCtrlMask = feature_ctrlmask; + fallthrough; + + case PP_OD_COMMIT_DPM_TABLE: + /* + * The member below instructs PMFW the settings focused in + * this single operation. + * `uint32_t FeatureCtrlMask;` + * It does not contain actual informations about user's custom + * settings. Thus we do not cache it. + */ + offset_of_featurectrlmask = offsetof(OverDriveTable_t, FeatureCtrlMask); + if (memcmp((u8 *)od_table + offset_of_featurectrlmask, + table_context->user_overdrive_table + offset_of_featurectrlmask, + sizeof(OverDriveTableExternal_t) - offset_of_featurectrlmask)) { + smu_v13_0_7_dump_od_table(smu, od_table); + + ret = smu_v13_0_7_upload_overdrive_table(smu, od_table); + if (ret) { + dev_err(adev->dev, "Failed to upload overdrive table!\n"); + return ret; + } + + od_table->OverDriveTable.FeatureCtrlMask = 0; + memcpy(table_context->user_overdrive_table + offset_of_featurectrlmask, + (u8 *)od_table + offset_of_featurectrlmask, + sizeof(OverDriveTableExternal_t) - offset_of_featurectrlmask); + + if (!memcmp(table_context->user_overdrive_table, + table_context->boot_overdrive_table, + sizeof(OverDriveTableExternal_t))) + smu->user_dpm_profile.user_od = false; + else + smu->user_dpm_profile.user_od = true; + } + break; + + default: + return -ENOSYS; + } + + return ret; +} + static int smu_v13_0_7_force_clk_levels(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t mask) @@ -1370,6 +1781,78 @@ static ssize_t smu_v13_0_7_get_gpu_metrics(struct smu_context *smu, return sizeof(struct gpu_metrics_v1_3); } +static int smu_v13_0_7_set_default_od_settings(struct smu_context *smu) +{ + OverDriveTableExternal_t *od_table = + (OverDriveTableExternal_t *)smu->smu_table.overdrive_table; + OverDriveTableExternal_t *boot_od_table = + (OverDriveTableExternal_t *)smu->smu_table.boot_overdrive_table; + OverDriveTableExternal_t *user_od_table = + (OverDriveTableExternal_t *)smu->smu_table.user_overdrive_table; + OverDriveTableExternal_t user_od_table_bak; + int ret = 0; + int i; + + ret = smu_v13_0_7_get_overdrive_table(smu, boot_od_table); + if (ret) + return ret; + + smu_v13_0_7_dump_od_table(smu, boot_od_table); + + memcpy(od_table, + boot_od_table, + sizeof(OverDriveTableExternal_t)); + + /* + * For S3/S4/Runpm resume, we need to setup those overdrive tables again, + * but we have to preserve user defined values in "user_od_table". + */ + if (!smu->adev->in_suspend) { + memcpy(user_od_table, + boot_od_table, + sizeof(OverDriveTableExternal_t)); + smu->user_dpm_profile.user_od = false; + } else if (smu->user_dpm_profile.user_od) { + memcpy(&user_od_table_bak, + user_od_table, + sizeof(OverDriveTableExternal_t)); + memcpy(user_od_table, + boot_od_table, + sizeof(OverDriveTableExternal_t)); + user_od_table->OverDriveTable.GfxclkFmin = + user_od_table_bak.OverDriveTable.GfxclkFmin; + user_od_table->OverDriveTable.GfxclkFmax = + user_od_table_bak.OverDriveTable.GfxclkFmax; + user_od_table->OverDriveTable.UclkFmin = + user_od_table_bak.OverDriveTable.UclkFmin; + user_od_table->OverDriveTable.UclkFmax = + user_od_table_bak.OverDriveTable.UclkFmax; + for (i = 0; i < PP_NUM_OD_VF_CURVE_POINTS; i++) + user_od_table->OverDriveTable.VoltageOffsetPerZoneBoundary[i] = + user_od_table_bak.OverDriveTable.VoltageOffsetPerZoneBoundary[i]; + } + + return 0; +} + +static int smu_v13_0_7_restore_user_od_settings(struct smu_context *smu) +{ + struct smu_table_context *table_context = &smu->smu_table; + OverDriveTableExternal_t *od_table = table_context->overdrive_table; + OverDriveTableExternal_t *user_od_table = table_context->user_overdrive_table; + int res; + + user_od_table->OverDriveTable.FeatureCtrlMask = 1U << PP_OD_FEATURE_GFXCLK_BIT | + 1U << PP_OD_FEATURE_UCLK_BIT | + 1U << PP_OD_FEATURE_GFX_VF_CURVE_BIT; + res = smu_v13_0_7_upload_overdrive_table(smu, user_od_table); + user_od_table->OverDriveTable.FeatureCtrlMask = 0; + if (res == 0) + memcpy(od_table, user_od_table, sizeof(OverDriveTableExternal_t)); + + return res; +} + static int smu_v13_0_7_populate_umd_state_clk(struct smu_context *smu) { struct smu_13_0_dpm_context *dpm_context = @@ -1759,6 +2242,9 @@ static const struct pptable_funcs smu_v13_0_7_ppt_funcs = { .notify_memory_pool_location = smu_v13_0_notify_memory_pool_location, .get_gpu_metrics = smu_v13_0_7_get_gpu_metrics, .set_soft_freq_limited_range = smu_v13_0_set_soft_freq_limited_range, + .set_default_od_settings = smu_v13_0_7_set_default_od_settings, + .restore_user_od_settings = smu_v13_0_7_restore_user_od_settings, + .od_edit_dpm_table = smu_v13_0_7_od_edit_dpm_table, .set_performance_level = smu_v13_0_set_performance_level, .gfx_off_control = smu_v13_0_gfx_off_control, .get_fan_speed_pwm = smu_v13_0_7_get_fan_speed_pwm, @@ -1770,6 +2256,7 @@ static const struct pptable_funcs smu_v13_0_7_ppt_funcs = { .enable_mgpu_fan_boost = smu_v13_0_7_enable_mgpu_fan_boost, .get_power_limit = smu_v13_0_7_get_power_limit, .set_power_limit = smu_v13_0_set_power_limit, + .set_power_source = smu_v13_0_set_power_source, .get_power_profile_mode = smu_v13_0_7_get_power_profile_mode, .set_power_profile_mode = smu_v13_0_7_set_power_profile_mode, .set_tool_table_location = smu_v13_0_set_tool_table_location, @@ -1796,5 +2283,6 @@ void smu_v13_0_7_set_ppt_funcs(struct smu_context *smu) smu->table_map = smu_v13_0_7_table_map; smu->pwr_src_map = smu_v13_0_7_pwr_src_map; smu->workload_map = smu_v13_0_7_workload_map; + smu->smc_driver_if_version = SMU13_0_7_DRIVER_IF_VERSION; smu_v13_0_set_smu_mailbox_registers(smu); } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c index 04e56b0b3033eaef681e9b3c9d2b14a18853cb3e..a1be2029ba4ab9e44d9a9a4e884806e2024a8fd0 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c @@ -47,6 +47,14 @@ #define SMUIO_GFX_MISC_CNTL__PWR_GFXOFF_STATUS_MASK 0x00000006L #define SMUIO_GFX_MISC_CNTL__PWR_GFXOFF_STATUS__SHIFT 0x1L +#define SMU_13_0_8_UMD_PSTATE_GFXCLK 533 +#define SMU_13_0_8_UMD_PSTATE_SOCCLK 533 +#define SMU_13_0_8_UMD_PSTATE_FCLK 800 + +#define SMU_13_0_1_UMD_PSTATE_GFXCLK 700 +#define SMU_13_0_1_UMD_PSTATE_SOCCLK 678 +#define SMU_13_0_1_UMD_PSTATE_FCLK 1800 + #define FEATURE_MASK(feature) (1ULL << feature) #define SMC_DPM_FEATURE ( \ FEATURE_MASK(FEATURE_CCLK_DPM_BIT) | \ @@ -957,6 +965,9 @@ static int yellow_carp_set_soft_freq_limited_range(struct smu_context *smu, uint32_t max) { enum smu_message_type msg_set_min, msg_set_max; + uint32_t min_clk = min; + uint32_t max_clk = max; + int ret = 0; if (!yellow_carp_clk_dpm_is_enabled(smu, clk_type)) @@ -985,11 +996,17 @@ static int yellow_carp_set_soft_freq_limited_range(struct smu_context *smu, return -EINVAL; } - ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_min, min, NULL); + if (clk_type == SMU_VCLK) { + min_clk = min << SMU_13_VCLK_SHIFT; + max_clk = max << SMU_13_VCLK_SHIFT; + } + + ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_min, min_clk, NULL); + if (ret) goto out; - ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_max, max, NULL); + ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_max, max_clk, NULL); if (ret) goto out; @@ -997,12 +1014,49 @@ static int yellow_carp_set_soft_freq_limited_range(struct smu_context *smu, return ret; } +static uint32_t yellow_carp_get_umd_pstate_clk_default(struct smu_context *smu, + enum smu_clk_type clk_type) +{ + uint32_t clk_limit = 0; + struct amdgpu_device *adev = smu->adev; + + switch (clk_type) { + case SMU_GFXCLK: + case SMU_SCLK: + if ((adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 8)) + clk_limit = SMU_13_0_8_UMD_PSTATE_GFXCLK; + if ((adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 1) || + (adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 3)) + clk_limit = SMU_13_0_1_UMD_PSTATE_GFXCLK; + break; + case SMU_SOCCLK: + if ((adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 8)) + clk_limit = SMU_13_0_8_UMD_PSTATE_SOCCLK; + if ((adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 1) || + (adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 3)) + clk_limit = SMU_13_0_1_UMD_PSTATE_SOCCLK; + break; + case SMU_FCLK: + if ((adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 8)) + clk_limit = SMU_13_0_8_UMD_PSTATE_FCLK; + if ((adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 1) || + (adev->ip_versions[MP1_HWIP][0]) == IP_VERSION(13, 0, 3)) + clk_limit = SMU_13_0_1_UMD_PSTATE_FCLK; + break; + default: + break; + } + + return clk_limit; +} + static int yellow_carp_print_clk_levels(struct smu_context *smu, enum smu_clk_type clk_type, char *buf) { - int i, size = 0, ret = 0; + int i, idx, size = 0, ret = 0; uint32_t cur_value = 0, value = 0, count = 0; uint32_t min, max; + uint32_t clk_limit = 0; smu_cmn_get_sysfs_buf(&buf, &size); @@ -1033,7 +1087,8 @@ static int yellow_carp_print_clk_levels(struct smu_context *smu, goto print_clk_out; for (i = 0; i < count; i++) { - ret = yellow_carp_get_dpm_freq_by_index(smu, clk_type, i, &value); + idx = (clk_type == SMU_FCLK || clk_type == SMU_MCLK) ? (count - i - 1) : i; + ret = yellow_carp_get_dpm_freq_by_index(smu, clk_type, idx, &value); if (ret) goto print_clk_out; @@ -1043,6 +1098,7 @@ static int yellow_carp_print_clk_levels(struct smu_context *smu, break; case SMU_GFXCLK: case SMU_SCLK: + clk_limit = yellow_carp_get_umd_pstate_clk_default(smu, clk_type); ret = yellow_carp_get_current_clk_freq(smu, clk_type, &cur_value); if (ret) goto print_clk_out; @@ -1057,7 +1113,7 @@ static int yellow_carp_print_clk_levels(struct smu_context *smu, size += sysfs_emit_at(buf, size, "0: %uMhz %s\n", min, i == 0 ? "*" : ""); size += sysfs_emit_at(buf, size, "1: %uMhz %s\n", - i == 1 ? cur_value : YELLOW_CARP_UMD_PSTATE_GFXCLK, + i == 1 ? cur_value : clk_limit, i == 1 ? "*" : ""); size += sysfs_emit_at(buf, size, "2: %uMhz %s\n", max, i == 2 ? "*" : ""); @@ -1106,6 +1162,49 @@ static int yellow_carp_force_clk_levels(struct smu_context *smu, return ret; } +static int yellow_carp_get_dpm_profile_freq(struct smu_context *smu, + enum amd_dpm_forced_level level, + enum smu_clk_type clk_type, + uint32_t *min_clk, + uint32_t *max_clk) +{ + int ret = 0; + uint32_t clk_limit = 0; + + clk_limit = yellow_carp_get_umd_pstate_clk_default(smu, clk_type); + + switch (clk_type) { + case SMU_GFXCLK: + case SMU_SCLK: + if (level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK) + yellow_carp_get_dpm_ultimate_freq(smu, SMU_SCLK, NULL, &clk_limit); + else if (level == AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK) + yellow_carp_get_dpm_ultimate_freq(smu, SMU_SCLK, &clk_limit, NULL); + break; + case SMU_SOCCLK: + if (level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK) + yellow_carp_get_dpm_ultimate_freq(smu, SMU_SOCCLK, NULL, &clk_limit); + break; + case SMU_FCLK: + if (level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK) + yellow_carp_get_dpm_ultimate_freq(smu, SMU_FCLK, NULL, &clk_limit); + else if (level == AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK) + yellow_carp_get_dpm_ultimate_freq(smu, SMU_FCLK, &clk_limit, NULL); + break; + case SMU_VCLK: + yellow_carp_get_dpm_ultimate_freq(smu, SMU_VCLK, NULL, &clk_limit); + break; + case SMU_DCLK: + yellow_carp_get_dpm_ultimate_freq(smu, SMU_DCLK, NULL, &clk_limit); + break; + default: + ret = -EINVAL; + break; + } + *min_clk = *max_clk = clk_limit; + return ret; +} + static int yellow_carp_set_performance_level(struct smu_context *smu, enum amd_dpm_forced_level level) { @@ -1113,6 +1212,9 @@ static int yellow_carp_set_performance_level(struct smu_context *smu, uint32_t sclk_min = 0, sclk_max = 0; uint32_t fclk_min = 0, fclk_max = 0; uint32_t socclk_min = 0, socclk_max = 0; + uint32_t vclk_min = 0, vclk_max = 0; + uint32_t dclk_min = 0, dclk_max = 0; + int ret = 0; switch (level) { @@ -1120,28 +1222,42 @@ static int yellow_carp_set_performance_level(struct smu_context *smu, yellow_carp_get_dpm_ultimate_freq(smu, SMU_SCLK, NULL, &sclk_max); yellow_carp_get_dpm_ultimate_freq(smu, SMU_FCLK, NULL, &fclk_max); yellow_carp_get_dpm_ultimate_freq(smu, SMU_SOCCLK, NULL, &socclk_max); + yellow_carp_get_dpm_ultimate_freq(smu, SMU_VCLK, NULL, &vclk_max); + yellow_carp_get_dpm_ultimate_freq(smu, SMU_DCLK, NULL, &dclk_max); sclk_min = sclk_max; fclk_min = fclk_max; socclk_min = socclk_max; + vclk_min = vclk_max; + dclk_min = dclk_max; break; case AMD_DPM_FORCED_LEVEL_LOW: yellow_carp_get_dpm_ultimate_freq(smu, SMU_SCLK, &sclk_min, NULL); yellow_carp_get_dpm_ultimate_freq(smu, SMU_FCLK, &fclk_min, NULL); yellow_carp_get_dpm_ultimate_freq(smu, SMU_SOCCLK, &socclk_min, NULL); + yellow_carp_get_dpm_ultimate_freq(smu, SMU_VCLK, &vclk_min, NULL); + yellow_carp_get_dpm_ultimate_freq(smu, SMU_DCLK, &dclk_min, NULL); sclk_max = sclk_min; fclk_max = fclk_min; socclk_max = socclk_min; + vclk_max = vclk_min; + dclk_max = dclk_min; break; case AMD_DPM_FORCED_LEVEL_AUTO: yellow_carp_get_dpm_ultimate_freq(smu, SMU_SCLK, &sclk_min, &sclk_max); yellow_carp_get_dpm_ultimate_freq(smu, SMU_FCLK, &fclk_min, &fclk_max); yellow_carp_get_dpm_ultimate_freq(smu, SMU_SOCCLK, &socclk_min, &socclk_max); + yellow_carp_get_dpm_ultimate_freq(smu, SMU_VCLK, &vclk_min, &vclk_max); + yellow_carp_get_dpm_ultimate_freq(smu, SMU_DCLK, &dclk_min, &dclk_max); break; case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD: case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK: case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK: case AMD_DPM_FORCED_LEVEL_PROFILE_PEAK: - /* Temporarily do nothing since the optimal clocks haven't been provided yet */ + yellow_carp_get_dpm_profile_freq(smu, level, SMU_SCLK, &sclk_min, &sclk_max); + yellow_carp_get_dpm_profile_freq(smu, level, SMU_FCLK, &fclk_min, &fclk_max); + yellow_carp_get_dpm_profile_freq(smu, level, SMU_SOCCLK, &socclk_min, &socclk_max); + yellow_carp_get_dpm_profile_freq(smu, level, SMU_VCLK, &vclk_min, &vclk_max); + yellow_carp_get_dpm_profile_freq(smu, level, SMU_DCLK, &dclk_min, &dclk_max); break; case AMD_DPM_FORCED_LEVEL_MANUAL: case AMD_DPM_FORCED_LEVEL_PROFILE_EXIT: @@ -1181,6 +1297,24 @@ static int yellow_carp_set_performance_level(struct smu_context *smu, return ret; } + if (vclk_min && vclk_max) { + ret = yellow_carp_set_soft_freq_limited_range(smu, + SMU_VCLK, + vclk_min, + vclk_max); + if (ret) + return ret; + } + + if (dclk_min && dclk_max) { + ret = yellow_carp_set_soft_freq_limited_range(smu, + SMU_DCLK, + dclk_min, + dclk_max); + if (ret) + return ret; + } + return ret; } @@ -1234,5 +1368,6 @@ void yellow_carp_set_ppt_funcs(struct smu_context *smu) smu->feature_map = yellow_carp_feature_mask_map; smu->table_map = yellow_carp_table_map; smu->is_apu = true; + smu->smc_driver_if_version = SMU13_YELLOW_CARP_DRIVER_IF_VERSION; smu_v13_0_set_smu_mailbox_registers(smu); } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.h index a9205a8ea3ad2cc7410d14a5a58b291ff7c2f62c..b3ad8352c68ae1445b34cff2ee91b9f55489b47d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.h +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.h @@ -24,6 +24,5 @@ #define __YELLOW_CARP_PPT_H__ extern void yellow_carp_set_ppt_funcs(struct smu_context *smu); -#define YELLOW_CARP_UMD_PSTATE_GFXCLK 1100 #endif diff --git a/drivers/gpu/drm/ast/ast_dp.c b/drivers/gpu/drm/ast/ast_dp.c index fbb070f63e36b8656bedfadea9b629dbad310f20..6dc1a09504e13ec2bad4bbf095c28a00b710d9e4 100644 --- a/drivers/gpu/drm/ast/ast_dp.c +++ b/drivers/gpu/drm/ast/ast_dp.c @@ -119,53 +119,32 @@ int ast_astdp_read_edid(struct drm_device *dev, u8 *ediddata) /* * Launch Aspeed DP */ -void ast_dp_launch(struct drm_device *dev, u8 bPower) +void ast_dp_launch(struct drm_device *dev) { - u32 i = 0, j = 0, WaitCount = 1; - u8 bDPTX = 0; + u32 i = 0; u8 bDPExecute = 1; - struct ast_device *ast = to_ast_device(dev); - // S3 come back, need more time to wait BMC ready. - if (bPower) - WaitCount = 300; - - - // Wait total count by different condition. - for (j = 0; j < WaitCount; j++) { - bDPTX = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xD1, TX_TYPE_MASK); - - if (bDPTX) - break; + // Wait one second then timeout. + while (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xD1, ASTDP_MCU_FW_EXECUTING) != + ASTDP_MCU_FW_EXECUTING) { + i++; + // wait 100 ms msleep(100); - } - // 0xE : ASTDP with DPMCU FW handling - if (bDPTX == ASTDP_DPMCU_TX) { - // Wait one second then timeout. - i = 0; - - while (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xD1, COPROCESSOR_LAUNCH) != - COPROCESSOR_LAUNCH) { - i++; - // wait 100 ms - msleep(100); - - if (i >= 10) { - // DP would not be ready. - bDPExecute = 0; - break; - } + if (i >= 10) { + // DP would not be ready. + bDPExecute = 0; + break; } + } - if (bDPExecute) - ast->tx_chip_types |= BIT(AST_TX_ASTDP); + if (!bDPExecute) + drm_err(dev, "Wait DPMCU executing timeout\n"); - ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xE5, - (u8) ~ASTDP_HOST_EDID_READ_DONE_MASK, - ASTDP_HOST_EDID_READ_DONE); - } + ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xE5, + (u8) ~ASTDP_HOST_EDID_READ_DONE_MASK, + ASTDP_HOST_EDID_READ_DONE); } diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h index a501169cddad890c0c845cf9217d8b517e16af94..5498a6676f2e8f2eede492b53c34e875ccf81635 100644 --- a/drivers/gpu/drm/ast/ast_drv.h +++ b/drivers/gpu/drm/ast/ast_drv.h @@ -350,9 +350,6 @@ int ast_mode_config_init(struct ast_device *ast); #define AST_DP501_LINKRATE 0xf014 #define AST_DP501_EDID_DATA 0xf020 -/* Define for Soc scratched reg */ -#define COPROCESSOR_LAUNCH BIT(5) - /* * Display Transmitter Type: */ @@ -480,7 +477,7 @@ struct ast_i2c_chan *ast_i2c_create(struct drm_device *dev); /* aspeed DP */ int ast_astdp_read_edid(struct drm_device *dev, u8 *ediddata); -void ast_dp_launch(struct drm_device *dev, u8 bPower); +void ast_dp_launch(struct drm_device *dev); void ast_dp_power_on_off(struct drm_device *dev, bool no); void ast_dp_set_on_off(struct drm_device *dev, bool no); void ast_dp_set_mode(struct drm_crtc *crtc, struct ast_vbios_mode_info *vbios_mode); diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index f32ce29edba726a903d7b082f770ae896bfce3cc..1f35438f614a7ba441af81e1e564da3ec1dd5d52 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -254,8 +254,13 @@ static int ast_detect_chip(struct drm_device *dev, bool *need_post) case 0x0c: ast->tx_chip_types = AST_TX_DP501_BIT; } - } else if (ast->chip == AST2600) - ast_dp_launch(&ast->base, 0); + } else if (ast->chip == AST2600) { + if (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xD1, TX_TYPE_MASK) == + ASTDP_DPMCU_TX) { + ast->tx_chip_types = AST_TX_ASTDP_BIT; + ast_dp_launch(&ast->base); + } + } /* Print stuff for diagnostic purposes */ if (ast->tx_chip_types & AST_TX_NONE_BIT) @@ -264,6 +269,8 @@ static int ast_detect_chip(struct drm_device *dev, bool *need_post) drm_info(dev, "Using Sil164 TMDS transmitter\n"); if (ast->tx_chip_types & AST_TX_DP501_BIT) drm_info(dev, "Using DP501 DisplayPort transmitter\n"); + if (ast->tx_chip_types & AST_TX_ASTDP_BIT) + drm_info(dev, "Using ASPEED DisplayPort transmitter\n"); return 0; } diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index 36374828f6c8fa790d89028a5a26e53722048baf..b3c670af6ef2bc309051457d292f6919200e3f5c 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -1647,6 +1647,8 @@ static int ast_dp501_output_init(struct ast_device *ast) static int ast_astdp_connector_helper_get_modes(struct drm_connector *connector) { void *edid; + struct drm_device *dev = connector->dev; + struct ast_device *ast = to_ast_device(dev); int succ; int count; @@ -1655,9 +1657,17 @@ static int ast_astdp_connector_helper_get_modes(struct drm_connector *connector) if (!edid) goto err_drm_connector_update_edid_property; + /* + * Protect access to I/O registers from concurrent modesetting + * by acquiring the I/O-register lock. + */ + mutex_lock(&ast->ioregs_lock); + succ = ast_astdp_read_edid(connector->dev, edid); if (succ < 0) - goto err_kfree; + goto err_mutex_unlock; + + mutex_unlock(&ast->ioregs_lock); drm_connector_update_edid_property(connector, edid); count = drm_add_edid_modes(connector, edid); @@ -1665,7 +1675,8 @@ static int ast_astdp_connector_helper_get_modes(struct drm_connector *connector) return count; -err_kfree: +err_mutex_unlock: + mutex_unlock(&ast->ioregs_lock); kfree(edid); err_drm_connector_update_edid_property: drm_connector_update_edid_property(connector, NULL); diff --git a/drivers/gpu/drm/ast/ast_post.c b/drivers/gpu/drm/ast/ast_post.c index 71bb36b865fdac627ac846ac210d7d1f3c0da5e4..a005aec18a0209a9bdb6e0dec252a750d3c99409 100644 --- a/drivers/gpu/drm/ast/ast_post.c +++ b/drivers/gpu/drm/ast/ast_post.c @@ -380,7 +380,8 @@ void ast_post_gpu(struct drm_device *dev) ast_set_def_ext_reg(dev); if (ast->chip == AST2600) { - ast_dp_launch(dev, 1); + if (ast->tx_chip_types & AST_TX_ASTDP_BIT) + ast_dp_launch(dev); } else if (ast->config_mode == ast_use_p2a) { if (ast->chip == AST2500) ast_post_chip_2500(dev); diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi86.c b/drivers/gpu/drm/bridge/ti-sn65dsi86.c index 597ceb7024e0ea9baff44791e47a07ca32230a13..c499a14d0b980a12333d6ea809d198ec568c16f3 100644 --- a/drivers/gpu/drm/bridge/ti-sn65dsi86.c +++ b/drivers/gpu/drm/bridge/ti-sn65dsi86.c @@ -298,6 +298,10 @@ static void ti_sn_bridge_set_refclk_freq(struct ti_sn65dsi86 *pdata) if (refclk_lut[i] == refclk_rate) break; + /* avoid buffer overflow and "1" is the default rate in the datasheet. */ + if (i >= refclk_lut_size) + i = 1; + regmap_update_bits(pdata->regmap, SN_DPPLL_SRC_REG, REFCLK_FREQ_MASK, REFCLK_FREQ(i)); diff --git a/drivers/gpu/drm/display/drm_dp_mst_topology.c b/drivers/gpu/drm/display/drm_dp_mst_topology.c index 38dab76ae69ea6232062be9b431d52a57ed4015c..8fe7b635e5bb261d2f3499845fcfdafdb9b405f8 100644 --- a/drivers/gpu/drm/display/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/display/drm_dp_mst_topology.c @@ -1823,7 +1823,7 @@ static void drm_dp_destroy_port(struct kref *kref) return; } - kfree(port->cached_edid); + drm_edid_free(port->cached_edid); /* * we can't destroy the connector here, as we might be holding the @@ -2272,8 +2272,8 @@ drm_dp_mst_port_add_connector(struct drm_dp_mst_branch *mstb, if (port->pdt != DP_PEER_DEVICE_NONE && drm_dp_mst_is_end_device(port->pdt, port->mcs) && port->port_num >= DP_MST_LOGICAL_PORT_0) - port->cached_edid = drm_get_edid(port->connector, - &port->aux.ddc); + port->cached_edid = drm_edid_read_ddc(port->connector, + &port->aux.ddc); drm_connector_register(port->connector); return; @@ -4053,17 +4053,28 @@ static int drm_dp_mst_handle_up_req(struct drm_dp_mst_topology_mgr *mgr) } /** - * drm_dp_mst_hpd_irq() - MST hotplug IRQ notify + * drm_dp_mst_hpd_irq_handle_event() - MST hotplug IRQ handle MST event * @mgr: manager to notify irq for. * @esi: 4 bytes from SINK_COUNT_ESI + * @ack: 4 bytes used to ack events starting from SINK_COUNT_ESI * @handled: whether the hpd interrupt was consumed or not * - * This should be called from the driver when it detects a short IRQ, + * This should be called from the driver when it detects a HPD IRQ, * along with the value of the DEVICE_SERVICE_IRQ_VECTOR_ESI0. The - * topology manager will process the sideband messages received as a result - * of this. + * topology manager will process the sideband messages received + * as indicated in the DEVICE_SERVICE_IRQ_VECTOR_ESI0 and set the + * corresponding flags that Driver has to ack the DP receiver later. + * + * Note that driver shall also call + * drm_dp_mst_hpd_irq_send_new_request() if the 'handled' is set + * after calling this function, to try to kick off a new request in + * the queue if the previous message transaction is completed. + * + * See also: + * drm_dp_mst_hpd_irq_send_new_request() */ -int drm_dp_mst_hpd_irq(struct drm_dp_mst_topology_mgr *mgr, u8 *esi, bool *handled) +int drm_dp_mst_hpd_irq_handle_event(struct drm_dp_mst_topology_mgr *mgr, const u8 *esi, + u8 *ack, bool *handled) { int ret = 0; int sc; @@ -4078,18 +4089,47 @@ int drm_dp_mst_hpd_irq(struct drm_dp_mst_topology_mgr *mgr, u8 *esi, bool *handl if (esi[1] & DP_DOWN_REP_MSG_RDY) { ret = drm_dp_mst_handle_down_rep(mgr); *handled = true; + ack[1] |= DP_DOWN_REP_MSG_RDY; } if (esi[1] & DP_UP_REQ_MSG_RDY) { ret |= drm_dp_mst_handle_up_req(mgr); *handled = true; + ack[1] |= DP_UP_REQ_MSG_RDY; } - drm_dp_mst_kick_tx(mgr); return ret; } -EXPORT_SYMBOL(drm_dp_mst_hpd_irq); +EXPORT_SYMBOL(drm_dp_mst_hpd_irq_handle_event); +/** + * drm_dp_mst_hpd_irq_send_new_request() - MST hotplug IRQ kick off new request + * @mgr: manager to notify irq for. + * + * This should be called from the driver when mst irq event is handled + * and acked. Note that new down request should only be sent when + * previous message transaction is completed. Source is not supposed to generate + * interleaved message transactions. + */ +void drm_dp_mst_hpd_irq_send_new_request(struct drm_dp_mst_topology_mgr *mgr) +{ + struct drm_dp_sideband_msg_tx *txmsg; + bool kick = true; + + mutex_lock(&mgr->qlock); + txmsg = list_first_entry_or_null(&mgr->tx_msg_downq, + struct drm_dp_sideband_msg_tx, next); + /* If last transaction is not completed yet*/ + if (!txmsg || + txmsg->state == DRM_DP_SIDEBAND_TX_START_SEND || + txmsg->state == DRM_DP_SIDEBAND_TX_SENT) + kick = false; + mutex_unlock(&mgr->qlock); + + if (kick) + drm_dp_mst_kick_tx(mgr); +} +EXPORT_SYMBOL(drm_dp_mst_hpd_irq_send_new_request); /** * drm_dp_mst_detect_port() - get connection status for an MST port * @connector: DRM connector for this port @@ -4133,7 +4173,7 @@ drm_dp_mst_detect_port(struct drm_connector *connector, ret = connector_status_connected; /* for logical ports - cache the EDID */ if (port->port_num >= DP_MST_LOGICAL_PORT_0 && !port->cached_edid) - port->cached_edid = drm_get_edid(connector, &port->aux.ddc); + port->cached_edid = drm_edid_read_ddc(connector, &port->aux.ddc); break; case DP_PEER_DEVICE_DP_LEGACY_CONV: if (port->ldps) @@ -4147,7 +4187,7 @@ drm_dp_mst_detect_port(struct drm_connector *connector, EXPORT_SYMBOL(drm_dp_mst_detect_port); /** - * drm_dp_mst_get_edid() - get EDID for an MST port + * drm_dp_mst_edid_read() - get EDID for an MST port * @connector: toplevel connector to get EDID for * @mgr: manager for this port * @port: unverified pointer to a port. @@ -4156,9 +4196,11 @@ EXPORT_SYMBOL(drm_dp_mst_detect_port); * It validates the pointer still exists so the caller doesn't require a * reference. */ -struct edid *drm_dp_mst_get_edid(struct drm_connector *connector, struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_port *port) +const struct drm_edid *drm_dp_mst_edid_read(struct drm_connector *connector, + struct drm_dp_mst_topology_mgr *mgr, + struct drm_dp_mst_port *port) { - struct edid *edid = NULL; + const struct drm_edid *drm_edid; /* we need to search for the port in the mgr in case it's gone */ port = drm_dp_mst_topology_get_port_validated(mgr, port); @@ -4166,12 +4208,41 @@ struct edid *drm_dp_mst_get_edid(struct drm_connector *connector, struct drm_dp_ return NULL; if (port->cached_edid) - edid = drm_edid_duplicate(port->cached_edid); - else { - edid = drm_get_edid(connector, &port->aux.ddc); - } - port->has_audio = drm_detect_monitor_audio(edid); + drm_edid = drm_edid_dup(port->cached_edid); + else + drm_edid = drm_edid_read_ddc(connector, &port->aux.ddc); + drm_dp_mst_topology_put_port(port); + + return drm_edid; +} +EXPORT_SYMBOL(drm_dp_mst_edid_read); + +/** + * drm_dp_mst_get_edid() - get EDID for an MST port + * @connector: toplevel connector to get EDID for + * @mgr: manager for this port + * @port: unverified pointer to a port. + * + * This function is deprecated; please use drm_dp_mst_edid_read() instead. + * + * This returns an EDID for the port connected to a connector, + * It validates the pointer still exists so the caller doesn't require a + * reference. + */ +struct edid *drm_dp_mst_get_edid(struct drm_connector *connector, + struct drm_dp_mst_topology_mgr *mgr, + struct drm_dp_mst_port *port) +{ + const struct drm_edid *drm_edid; + struct edid *edid; + + drm_edid = drm_dp_mst_edid_read(connector, mgr, port); + + edid = drm_edid_duplicate(drm_edid_raw(drm_edid)); + + drm_edid_free(drm_edid); + return edid; } EXPORT_SYMBOL(drm_dp_mst_get_edid); diff --git a/drivers/gpu/drm/display/drm_dsc_helper.c b/drivers/gpu/drm/display/drm_dsc_helper.c index c869c6e51e2b7aecd14ccad9e8abbb5c4e28c769..4424380c6cb63120a004b76f4c93acfe28dd5ce0 100644 --- a/drivers/gpu/drm/display/drm_dsc_helper.c +++ b/drivers/gpu/drm/display/drm_dsc_helper.c @@ -270,6 +270,1034 @@ void drm_dsc_pps_payload_pack(struct drm_dsc_picture_parameter_set *pps_payload, } EXPORT_SYMBOL(drm_dsc_pps_payload_pack); +/** + * drm_dsc_set_const_params() - Set DSC parameters considered typically + * constant across operation modes + * + * @vdsc_cfg: + * DSC Configuration data partially filled by driver + */ +void drm_dsc_set_const_params(struct drm_dsc_config *vdsc_cfg) +{ + if (!vdsc_cfg->rc_model_size) + vdsc_cfg->rc_model_size = DSC_RC_MODEL_SIZE_CONST; + vdsc_cfg->rc_edge_factor = DSC_RC_EDGE_FACTOR_CONST; + vdsc_cfg->rc_tgt_offset_high = DSC_RC_TGT_OFFSET_HI_CONST; + vdsc_cfg->rc_tgt_offset_low = DSC_RC_TGT_OFFSET_LO_CONST; + + if (vdsc_cfg->bits_per_component <= 10) + vdsc_cfg->mux_word_size = DSC_MUX_WORD_SIZE_8_10_BPC; + else + vdsc_cfg->mux_word_size = DSC_MUX_WORD_SIZE_12_BPC; +} +EXPORT_SYMBOL(drm_dsc_set_const_params); + +/* From DSC_v1.11 spec, rc_parameter_Set syntax element typically constant */ +static const u16 drm_dsc_rc_buf_thresh[] = { + 896, 1792, 2688, 3584, 4480, 5376, 6272, 6720, 7168, 7616, + 7744, 7872, 8000, 8064 +}; + +/** + * drm_dsc_set_rc_buf_thresh() - Set thresholds for the RC model + * in accordance with the DSC 1.2 specification. + * + * @vdsc_cfg: DSC Configuration data partially filled by driver + */ +void drm_dsc_set_rc_buf_thresh(struct drm_dsc_config *vdsc_cfg) +{ + int i; + + BUILD_BUG_ON(ARRAY_SIZE(drm_dsc_rc_buf_thresh) != + DSC_NUM_BUF_RANGES - 1); + BUILD_BUG_ON(ARRAY_SIZE(drm_dsc_rc_buf_thresh) != + ARRAY_SIZE(vdsc_cfg->rc_buf_thresh)); + + for (i = 0; i < ARRAY_SIZE(drm_dsc_rc_buf_thresh); i++) + vdsc_cfg->rc_buf_thresh[i] = drm_dsc_rc_buf_thresh[i] >> 6; + + /* + * For 6bpp, RC Buffer threshold 12 and 13 need a different value + * as per C Model + */ + if (vdsc_cfg->bits_per_pixel == 6 << 4) { + vdsc_cfg->rc_buf_thresh[12] = 7936 >> 6; + vdsc_cfg->rc_buf_thresh[13] = 8000 >> 6; + } +} +EXPORT_SYMBOL(drm_dsc_set_rc_buf_thresh); + +struct rc_parameters { + u16 initial_xmit_delay; + u8 first_line_bpg_offset; + u16 initial_offset; + u8 flatness_min_qp; + u8 flatness_max_qp; + u8 rc_quant_incr_limit0; + u8 rc_quant_incr_limit1; + struct drm_dsc_rc_range_parameters rc_range_params[DSC_NUM_BUF_RANGES]; +}; + +struct rc_parameters_data { + u8 bpp; + u8 bpc; + struct rc_parameters params; +}; + +#define DSC_BPP(bpp) ((bpp) << 4) + +/* + * Rate Control Related Parameter Recommended Values from DSC_v1.1 spec prior + * to DSC 1.1 fractional bpp underflow SCR (DSC_v1.1_E1.pdf) + * + * Cross-checked against C Model releases: DSC_model_20161212 and 20210623 + */ +static const struct rc_parameters_data rc_parameters_pre_scr[] = { + { + .bpp = DSC_BPP(6), .bpc = 8, + { 683, 15, 6144, 3, 13, 11, 11, { + { 0, 2, 0 }, { 1, 4, -2 }, { 3, 6, -2 }, { 4, 6, -4 }, + { 5, 7, -6 }, { 5, 7, -6 }, { 6, 7, -6 }, { 6, 8, -8 }, + { 7, 9, -8 }, { 8, 10, -10 }, { 9, 11, -10 }, { 10, 12, -12 }, + { 10, 13, -12 }, { 12, 14, -12 }, { 15, 15, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 8, + { 512, 12, 6144, 3, 12, 11, 11, { + { 0, 4, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 }, + { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, + { 3, 9, -8 }, { 3, 10, -10 }, { 5, 11, -10 }, { 5, 12, -12 }, + { 5, 13, -12 }, { 7, 13, -12 }, { 13, 15, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 10, + { 512, 12, 6144, 7, 16, 15, 15, { + /* + * DSC model/pre-SCR-cfg has 8 for range_max_qp[0], however + * VESA DSC 1.1 Table E-5 sets it to 4. + */ + { 0, 4, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 5, 10, -2 }, + { 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, + { 7, 13, -8 }, { 7, 14, -10 }, { 9, 15, -10 }, { 9, 16, -12 }, + { 9, 17, -12 }, { 11, 17, -12 }, { 17, 19, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 12, + { 512, 12, 6144, 11, 20, 19, 19, { + { 0, 12, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 9, 14, -2 }, + { 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, + { 11, 17, -8 }, { 11, 18, -10 }, { 13, 19, -10 }, + { 13, 20, -12 }, { 13, 21, -12 }, { 15, 21, -12 }, + { 21, 23, -12 } + } + } + }, + { + .bpp = DSC_BPP(10), .bpc = 8, + { 410, 12, 5632, 3, 12, 11, 11, { + { 0, 3, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 2, 6, -2 }, + { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, + { 3, 9, -8 }, { 3, 9, -10 }, { 5, 10, -10 }, { 5, 11, -10 }, + { 5, 12, -12 }, { 7, 13, -12 }, { 13, 15, -12 } + } + } + }, + { + .bpp = DSC_BPP(10), .bpc = 10, + { 410, 12, 5632, 7, 16, 15, 15, { + { 0, 7, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 6, 10, -2 }, + { 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, + { 7, 13, -8 }, { 7, 13, -10 }, { 9, 14, -10 }, { 9, 15, -10 }, + { 9, 16, -12 }, { 11, 17, -12 }, { 17, 19, -12 } + } + } + }, + { + .bpp = DSC_BPP(10), .bpc = 12, + { 410, 12, 5632, 11, 20, 19, 19, { + { 0, 11, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 10, 14, -2 }, + { 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, + { 11, 17, -8 }, { 11, 17, -10 }, { 13, 18, -10 }, + { 13, 19, -10 }, { 13, 20, -12 }, { 15, 21, -12 }, + { 21, 23, -12 } + } + } + }, + { + .bpp = DSC_BPP(12), .bpc = 8, + { 341, 15, 2048, 3, 12, 11, 11, { + { 0, 2, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 }, + { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, + { 3, 9, -8 }, { 3, 10, -10 }, { 5, 11, -10 }, + { 5, 12, -12 }, { 5, 13, -12 }, { 7, 13, -12 }, { 13, 15, -12 } + } + } + }, + { + .bpp = DSC_BPP(12), .bpc = 10, + { 341, 15, 2048, 7, 16, 15, 15, { + { 0, 2, 2 }, { 2, 5, 0 }, { 3, 7, 0 }, { 4, 8, -2 }, + { 6, 9, -4 }, { 7, 10, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, + { 7, 13, -8 }, { 7, 14, -10 }, { 9, 15, -10 }, { 9, 16, -12 }, + { 9, 17, -12 }, { 11, 17, -12 }, { 17, 19, -12 } + } + } + }, + { + .bpp = DSC_BPP(12), .bpc = 12, + { 341, 15, 2048, 11, 20, 19, 19, { + { 0, 6, 2 }, { 4, 9, 0 }, { 7, 11, 0 }, { 8, 12, -2 }, + { 10, 13, -4 }, { 11, 14, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, + { 11, 17, -8 }, { 11, 18, -10 }, { 13, 19, -10 }, + { 13, 20, -12 }, { 13, 21, -12 }, { 15, 21, -12 }, + { 21, 23, -12 } + } + } + }, + { + .bpp = DSC_BPP(15), .bpc = 8, + { 273, 15, 2048, 3, 12, 11, 11, { + { 0, 0, 10 }, { 0, 1, 8 }, { 0, 1, 6 }, { 0, 2, 4 }, + { 1, 2, 2 }, { 1, 3, 0 }, { 1, 4, -2 }, { 2, 4, -4 }, + { 3, 4, -6 }, { 3, 5, -8 }, { 4, 6, -10 }, { 5, 7, -10 }, + { 5, 8, -12 }, { 7, 13, -12 }, { 13, 15, -12 } + } + } + }, + { + .bpp = DSC_BPP(15), .bpc = 10, + { 273, 15, 2048, 7, 16, 15, 15, { + { 0, 2, 10 }, { 2, 5, 8 }, { 3, 5, 6 }, { 4, 6, 4 }, + { 5, 6, 2 }, { 5, 7, 0 }, { 5, 8, -2 }, { 6, 8, -4 }, + { 7, 8, -6 }, { 7, 9, -8 }, { 8, 10, -10 }, { 9, 11, -10 }, + { 9, 12, -12 }, { 11, 17, -12 }, { 17, 19, -12 } + } + } + }, + { + .bpp = DSC_BPP(15), .bpc = 12, + { 273, 15, 2048, 11, 20, 19, 19, { + { 0, 4, 10 }, { 2, 7, 8 }, { 4, 9, 6 }, { 6, 11, 4 }, + { 9, 11, 2 }, { 9, 11, 0 }, { 9, 12, -2 }, { 10, 12, -4 }, + { 11, 12, -6 }, { 11, 13, -8 }, { 12, 14, -10 }, + { 13, 15, -10 }, { 13, 16, -12 }, { 15, 21, -12 }, + { 21, 23, -12 } + } + } + }, + { /* sentinel */ } +}; + +/* + * Selected Rate Control Related Parameter Recommended Values from DSC v1.2, v1.2a, v1.2b and + * DSC_v1.1_E1 specs. + * + * Cross-checked against C Model releases: DSC_model_20161212 and 20210623 + */ +static const struct rc_parameters_data rc_parameters_1_2_444[] = { + { + .bpp = DSC_BPP(6), .bpc = 8, + { 768, 15, 6144, 3, 13, 11, 11, { + { 0, 4, 0 }, { 1, 6, -2 }, { 3, 8, -2 }, { 4, 8, -4 }, + { 5, 9, -6 }, { 5, 9, -6 }, { 6, 9, -6 }, { 6, 10, -8 }, + { 7, 11, -8 }, { 8, 12, -10 }, { 9, 12, -10 }, { 10, 12, -12 }, + { 10, 12, -12 }, { 11, 12, -12 }, { 13, 14, -12 } + } + } + }, + { + .bpp = DSC_BPP(6), .bpc = 10, + { 768, 15, 6144, 7, 17, 15, 15, { + { 0, 8, 0 }, { 3, 10, -2 }, { 7, 12, -2 }, { 8, 12, -4 }, + { 9, 13, -6 }, { 9, 13, -6 }, { 10, 13, -6 }, { 10, 14, -8 }, + { 11, 15, -8 }, { 12, 16, -10 }, { 13, 16, -10 }, + { 14, 16, -12 }, { 14, 16, -12 }, { 15, 16, -12 }, + { 17, 18, -12 } + } + } + }, + { + .bpp = DSC_BPP(6), .bpc = 12, + { 768, 15, 6144, 11, 21, 19, 19, { + { 0, 12, 0 }, { 5, 14, -2 }, { 11, 16, -2 }, { 12, 16, -4 }, + { 13, 17, -6 }, { 13, 17, -6 }, { 14, 17, -6 }, { 14, 18, -8 }, + { 15, 19, -8 }, { 16, 20, -10 }, { 17, 20, -10 }, + { 18, 20, -12 }, { 18, 20, -12 }, { 19, 20, -12 }, + { 21, 22, -12 } + } + } + }, + { + .bpp = DSC_BPP(6), .bpc = 14, + { 768, 15, 6144, 15, 25, 23, 23, { + { 0, 16, 0 }, { 7, 18, -2 }, { 15, 20, -2 }, { 16, 20, -4 }, + { 17, 21, -6 }, { 17, 21, -6 }, { 18, 21, -6 }, { 18, 22, -8 }, + { 19, 23, -8 }, { 20, 24, -10 }, { 21, 24, -10 }, + { 22, 24, -12 }, { 22, 24, -12 }, { 23, 24, -12 }, + { 25, 26, -12 } + } + } + }, + { + .bpp = DSC_BPP(6), .bpc = 16, + { 768, 15, 6144, 19, 29, 27, 27, { + { 0, 20, 0 }, { 9, 22, -2 }, { 19, 24, -2 }, { 20, 24, -4 }, + { 21, 25, -6 }, { 21, 25, -6 }, { 22, 25, -6 }, { 22, 26, -8 }, + { 23, 27, -8 }, { 24, 28, -10 }, { 25, 28, -10 }, + { 26, 28, -12 }, { 26, 28, -12 }, { 27, 28, -12 }, + { 29, 30, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 8, + { 512, 12, 6144, 3, 12, 11, 11, { + { 0, 4, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 }, + { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, + { 3, 9, -8 }, { 3, 10, -10 }, { 5, 10, -10 }, { 5, 11, -12 }, + { 5, 11, -12 }, { 9, 12, -12 }, { 12, 13, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 10, + { 512, 12, 6144, 7, 16, 15, 15, { + { 0, 8, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 5, 10, -2 }, + { 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, + { 7, 13, -8 }, { 7, 14, -10 }, { 9, 14, -10 }, { 9, 15, -12 }, + { 9, 15, -12 }, { 13, 16, -12 }, { 16, 17, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 12, + { 512, 12, 6144, 11, 20, 19, 19, { + { 0, 12, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 9, 14, -2 }, + { 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, + { 11, 17, -8 }, { 11, 18, -10 }, { 13, 18, -10 }, + { 13, 19, -12 }, { 13, 19, -12 }, { 17, 20, -12 }, + { 20, 21, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 14, + { 512, 12, 6144, 15, 24, 23, 23, { + { 0, 12, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 12, 17, -2 }, + { 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 }, + { 15, 21, -8 }, { 15, 22, -10 }, { 17, 22, -10 }, + { 17, 23, -12 }, { 17, 23, -12 }, { 21, 24, -12 }, + { 24, 25, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 16, + { 512, 12, 6144, 19, 28, 27, 27, { + { 0, 12, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 15, 20, -2 }, + { 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 }, + { 19, 25, -8 }, { 19, 26, -10 }, { 21, 26, -10 }, + { 21, 27, -12 }, { 21, 27, -12 }, { 25, 28, -12 }, + { 28, 29, -12 } + } + } + }, + { + .bpp = DSC_BPP(10), .bpc = 8, + { 410, 15, 5632, 3, 12, 11, 11, { + { 0, 3, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 2, 6, -2 }, + { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, + { 3, 9, -8 }, { 3, 9, -10 }, { 5, 10, -10 }, { 5, 10, -10 }, + { 5, 11, -12 }, { 7, 11, -12 }, { 11, 12, -12 } + } + } + }, + { + .bpp = DSC_BPP(10), .bpc = 10, + { 410, 15, 5632, 7, 16, 15, 15, { + { 0, 7, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 6, 10, -2 }, + { 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, + { 7, 13, -8 }, { 7, 13, -10 }, { 9, 14, -10 }, { 9, 14, -10 }, + { 9, 15, -12 }, { 11, 15, -12 }, { 15, 16, -12 } + } + } + }, + { + .bpp = DSC_BPP(10), .bpc = 12, + { 410, 15, 5632, 11, 20, 19, 19, { + { 0, 11, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 10, 14, -2 }, + { 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, + { 11, 17, -8 }, { 11, 17, -10 }, { 13, 18, -10 }, + { 13, 18, -10 }, { 13, 19, -12 }, { 15, 19, -12 }, + { 19, 20, -12 } + } + } + }, + { + .bpp = DSC_BPP(10), .bpc = 14, + { 410, 15, 5632, 15, 24, 23, 23, { + { 0, 11, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 13, 18, -2 }, + { 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 }, + { 15, 21, -8 }, { 15, 21, -10 }, { 17, 22, -10 }, + { 17, 22, -10 }, { 17, 23, -12 }, { 19, 23, -12 }, + { 23, 24, -12 } + } + } + }, + { + .bpp = DSC_BPP(10), .bpc = 16, + { 410, 15, 5632, 19, 28, 27, 27, { + { 0, 11, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 16, 20, -2 }, + { 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 }, + { 19, 25, -8 }, { 19, 25, -10 }, { 21, 26, -10 }, + { 21, 26, -10 }, { 21, 27, -12 }, { 23, 27, -12 }, + { 27, 28, -12 } + } + } + }, + { + .bpp = DSC_BPP(12), .bpc = 8, + { 341, 15, 2048, 3, 12, 11, 11, { + { 0, 2, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 }, + { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, + { 3, 8, -8 }, { 3, 9, -10 }, { 5, 9, -10 }, { 5, 9, -12 }, + { 5, 9, -12 }, { 7, 10, -12 }, { 10, 11, -12 } + } + } + }, + { + .bpp = DSC_BPP(12), .bpc = 10, + { 341, 15, 2048, 7, 16, 15, 15, { + { 0, 2, 2 }, { 2, 5, 0 }, { 3, 7, 0 }, { 4, 8, -2 }, + { 6, 9, -4 }, { 7, 10, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, + { 7, 12, -8 }, { 7, 13, -10 }, { 9, 13, -10 }, { 9, 13, -12 }, + { 9, 13, -12 }, { 11, 14, -12 }, { 14, 15, -12 } + } + } + }, + { + .bpp = DSC_BPP(12), .bpc = 12, + { 341, 15, 2048, 11, 20, 19, 19, { + { 0, 6, 2 }, { 4, 9, 0 }, { 7, 11, 0 }, { 8, 12, -2 }, + { 10, 13, -4 }, { 11, 14, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, + { 11, 16, -8 }, { 11, 17, -10 }, { 13, 17, -10 }, + { 13, 17, -12 }, { 13, 17, -12 }, { 15, 18, -12 }, + { 18, 19, -12 } + } + } + }, + { + .bpp = DSC_BPP(12), .bpc = 14, + { 341, 15, 2048, 15, 24, 23, 23, { + { 0, 6, 2 }, { 7, 10, 0 }, { 9, 13, 0 }, { 11, 16, -2 }, + { 14, 17, -4 }, { 15, 18, -6 }, { 15, 19, -8 }, { 15, 20, -8 }, + { 15, 20, -8 }, { 15, 21, -10 }, { 17, 21, -10 }, + { 17, 21, -12 }, { 17, 21, -12 }, { 19, 22, -12 }, + { 22, 23, -12 } + } + } + }, + { + .bpp = DSC_BPP(12), .bpc = 16, + { 341, 15, 2048, 19, 28, 27, 27, { + { 0, 6, 2 }, { 6, 11, 0 }, { 11, 15, 0 }, { 14, 18, -2 }, + { 18, 21, -4 }, { 19, 22, -6 }, { 19, 23, -8 }, { 19, 24, -8 }, + { 19, 24, -8 }, { 19, 25, -10 }, { 21, 25, -10 }, + { 21, 25, -12 }, { 21, 25, -12 }, { 23, 26, -12 }, + { 26, 27, -12 } + } + } + }, + { + .bpp = DSC_BPP(15), .bpc = 8, + { 273, 15, 2048, 3, 12, 11, 11, { + { 0, 0, 10 }, { 0, 1, 8 }, { 0, 1, 6 }, { 0, 2, 4 }, + { 1, 2, 2 }, { 1, 3, 0 }, { 1, 3, -2 }, { 2, 4, -4 }, + { 2, 5, -6 }, { 3, 5, -8 }, { 4, 6, -10 }, { 4, 7, -10 }, + { 5, 7, -12 }, { 7, 8, -12 }, { 8, 9, -12 } + } + } + }, + { + .bpp = DSC_BPP(15), .bpc = 10, + { 273, 15, 2048, 7, 16, 15, 15, { + { 0, 2, 10 }, { 2, 5, 8 }, { 3, 5, 6 }, { 4, 6, 4 }, + { 5, 6, 2 }, { 5, 7, 0 }, { 5, 7, -2 }, { 6, 8, -4 }, + { 6, 9, -6 }, { 7, 9, -8 }, { 8, 10, -10 }, { 8, 11, -10 }, + { 9, 11, -12 }, { 11, 12, -12 }, { 12, 13, -12 } + } + } + }, + { + .bpp = DSC_BPP(15), .bpc = 12, + { 273, 15, 2048, 11, 20, 19, 19, { + { 0, 4, 10 }, { 2, 7, 8 }, { 4, 9, 6 }, { 6, 11, 4 }, + { 9, 11, 2 }, { 9, 11, 0 }, { 9, 12, -2 }, { 10, 12, -4 }, + { 11, 13, -6 }, { 11, 13, -8 }, { 12, 14, -10 }, + { 13, 15, -10 }, { 13, 15, -12 }, { 15, 16, -12 }, + { 16, 17, -12 } + } + } + }, + { + .bpp = DSC_BPP(15), .bpc = 14, + { 273, 15, 2048, 15, 24, 23, 23, { + { 0, 4, 10 }, { 3, 8, 8 }, { 6, 11, 6 }, { 9, 14, 4 }, + { 13, 15, 2 }, { 13, 15, 0 }, { 13, 16, -2 }, { 14, 16, -4 }, + { 15, 17, -6 }, { 15, 17, -8 }, { 16, 18, -10 }, + { 17, 19, -10 }, { 17, 19, -12 }, { 19, 20, -12 }, + { 20, 21, -12 } + } + } + }, + { + .bpp = DSC_BPP(15), .bpc = 16, + { 273, 15, 2048, 19, 28, 27, 27, { + { 0, 4, 10 }, { 4, 9, 8 }, { 8, 13, 6 }, { 12, 17, 4 }, + { 17, 19, 2 }, { 17, 20, 0 }, { 17, 20, -2 }, { 18, 20, -4 }, + { 19, 21, -6 }, { 19, 21, -8 }, { 20, 22, -10 }, + { 21, 23, -10 }, { 21, 23, -12 }, { 23, 24, -12 }, + { 24, 25, -12 } + } + } + }, + { /* sentinel */ } +}; + +/* + * Selected Rate Control Related Parameter Recommended Values for 4:2:2 from + * DSC v1.2, v1.2a, v1.2b + * + * Cross-checked against C Model releases: DSC_model_20161212 and 20210623 + */ +static const struct rc_parameters_data rc_parameters_1_2_422[] = { + { + .bpp = DSC_BPP(6), .bpc = 8, + { 512, 15, 6144, 3, 12, 11, 11, { + { 0, 4, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 }, + { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, + { 3, 9, -8 }, { 3, 10, -10 }, { 5, 10, -10 }, { 5, 11, -12 }, + { 5, 11, -12 }, { 9, 12, -12 }, { 12, 13, -12 } + } + } + }, + { + .bpp = DSC_BPP(6), .bpc = 10, + { 512, 15, 6144, 7, 16, 15, 15, { + { 0, 8, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 5, 10, -2 }, + { 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, + { 7, 13, -8 }, { 7, 14, -10 }, { 9, 14, -10 }, { 9, 15, -12 }, + { 9, 15, -12 }, { 13, 16, -12 }, { 16, 17, -12 } + } + } + }, + { + .bpp = DSC_BPP(6), .bpc = 12, + { 512, 15, 6144, 11, 20, 19, 19, { + { 0, 12, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 9, 14, -2 }, + { 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, + { 11, 17, -8 }, { 11, 18, -10 }, { 13, 18, -10 }, + { 13, 19, -12 }, { 13, 19, -12 }, { 17, 20, -12 }, + { 20, 21, -12 } + } + } + }, + { + .bpp = DSC_BPP(6), .bpc = 14, + { 512, 15, 6144, 15, 24, 23, 23, { + { 0, 12, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 12, 17, -2 }, + { 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 }, + { 15, 21, -8 }, { 15, 22, -10 }, { 17, 22, -10 }, + { 17, 23, -12 }, { 17, 23, -12 }, { 21, 24, -12 }, + { 24, 25, -12 } + } + } + }, + { + .bpp = DSC_BPP(6), .bpc = 16, + { 512, 15, 6144, 19, 28, 27, 27, { + { 0, 12, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 15, 20, -2 }, + { 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 }, + { 19, 25, -8 }, { 19, 26, -10 }, { 21, 26, -10 }, + { 21, 27, -12 }, { 21, 27, -12 }, { 25, 28, -12 }, + { 28, 29, -12 } + } + } + }, + { + .bpp = DSC_BPP(7), .bpc = 8, + { 410, 15, 5632, 3, 12, 11, 11, { + { 0, 3, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 2, 6, -2 }, + { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, + { 3, 9, -8 }, { 3, 9, -10 }, { 5, 10, -10 }, { 5, 10, -10 }, + { 5, 11, -12 }, { 7, 11, -12 }, { 11, 12, -12 } + } + } + }, + { + .bpp = DSC_BPP(7), .bpc = 10, + { 410, 15, 5632, 7, 16, 15, 15, { + { 0, 7, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 6, 10, -2 }, + { 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, + { 7, 13, -8 }, { 7, 13, -10 }, { 9, 14, -10 }, { 9, 14, -10 }, + { 9, 15, -12 }, { 11, 15, -12 }, { 15, 16, -12 } + } + } + }, + { + .bpp = DSC_BPP(7), .bpc = 12, + { 410, 15, 5632, 11, 20, 19, 19, { + { 0, 11, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 10, 14, -2 }, + { 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, + { 11, 17, -8 }, { 11, 17, -10 }, { 13, 18, -10 }, + { 13, 18, -10 }, { 13, 19, -12 }, { 15, 19, -12 }, + { 19, 20, -12 } + } + } + }, + { + .bpp = DSC_BPP(7), .bpc = 14, + { 410, 15, 5632, 15, 24, 23, 23, { + { 0, 11, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 13, 18, -2 }, + { 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 }, + { 15, 21, -8 }, { 15, 21, -10 }, { 17, 22, -10 }, + { 17, 22, -10 }, { 17, 23, -12 }, { 19, 23, -12 }, + { 23, 24, -12 } + } + } + }, + { + .bpp = DSC_BPP(7), .bpc = 16, + { 410, 15, 5632, 19, 28, 27, 27, { + { 0, 11, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 16, 20, -2 }, + { 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 }, + { 19, 25, -8 }, { 19, 25, -10 }, { 21, 26, -10 }, + { 21, 26, -10 }, { 21, 27, -12 }, { 23, 27, -12 }, + { 27, 28, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 8, + { 341, 15, 2048, 3, 12, 11, 11, { + { 0, 2, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 }, + { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, + { 3, 8, -8 }, { 3, 9, -10 }, { 5, 9, -10 }, { 5, 9, -12 }, + { 5, 9, -12 }, { 7, 10, -12 }, { 10, 11, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 10, + { 341, 15, 2048, 7, 16, 15, 15, { + { 0, 2, 2 }, { 2, 5, 0 }, { 3, 7, 0 }, { 4, 8, -2 }, + { 6, 9, -4 }, { 7, 10, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, + { 7, 12, -8 }, { 7, 13, -10 }, { 9, 13, -10 }, { 9, 13, -12 }, + { 9, 13, -12 }, { 11, 14, -12 }, { 14, 15, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 12, + { 341, 15, 2048, 11, 20, 19, 19, { + { 0, 6, 2 }, { 4, 9, 0 }, { 7, 11, 0 }, { 8, 12, -2 }, + { 10, 13, -4 }, { 11, 14, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, + { 11, 16, -8 }, { 11, 17, -10 }, { 13, 17, -10 }, + { 13, 17, -12 }, { 13, 17, -12 }, { 15, 18, -12 }, + { 18, 19, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 14, + { 341, 15, 2048, 15, 24, 23, 23, { + { 0, 6, 2 }, { 7, 10, 0 }, { 9, 13, 0 }, { 11, 16, -2 }, + { 14, 17, -4 }, { 15, 18, -6 }, { 15, 19, -8 }, { 15, 20, -8 }, + { 15, 20, -8 }, { 15, 21, -10 }, { 17, 21, -10 }, + { 17, 21, -12 }, { 17, 21, -12 }, { 19, 22, -12 }, + { 22, 23, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 16, + { 341, 15, 2048, 19, 28, 27, 27, { + { 0, 6, 2 }, { 6, 11, 0 }, { 11, 15, 0 }, { 14, 18, -2 }, + { 18, 21, -4 }, { 19, 22, -6 }, { 19, 23, -8 }, { 19, 24, -8 }, + { 19, 24, -8 }, { 19, 25, -10 }, { 21, 25, -10 }, + { 21, 25, -12 }, { 21, 25, -12 }, { 23, 26, -12 }, + { 26, 27, -12 } + } + } + }, + { + .bpp = DSC_BPP(10), .bpc = 8, + { 273, 15, 2048, 3, 12, 11, 11, { + { 0, 0, 10 }, { 0, 1, 8 }, { 0, 1, 6 }, { 0, 2, 4 }, + { 1, 2, 2 }, { 1, 3, 0 }, { 1, 3, -2 }, { 2, 4, -4 }, + { 2, 5, -6 }, { 3, 5, -8 }, { 4, 6, -10 }, { 4, 7, -10 }, + { 5, 7, -12 }, { 7, 8, -12 }, { 8, 9, -12 } + } + } + }, + { + .bpp = DSC_BPP(10), .bpc = 10, + { 273, 15, 2048, 7, 16, 15, 15, { + { 0, 2, 10 }, { 2, 5, 8 }, { 3, 5, 6 }, { 4, 6, 4 }, + { 5, 6, 2 }, { 5, 7, 0 }, { 5, 7, -2 }, { 6, 8, -4 }, + { 6, 9, -6 }, { 7, 9, -8 }, { 8, 10, -10 }, { 8, 11, -10 }, + { 9, 11, -12 }, { 11, 12, -12 }, { 12, 13, -12 } + } + } + }, + { + .bpp = DSC_BPP(10), .bpc = 12, + { 273, 15, 2048, 11, 20, 19, 19, { + { 0, 4, 10 }, { 2, 7, 8 }, { 4, 9, 6 }, { 6, 11, 4 }, + { 9, 11, 2 }, { 9, 11, 0 }, { 9, 12, -2 }, { 10, 12, -4 }, + { 11, 13, -6 }, { 11, 13, -8 }, { 12, 14, -10 }, + { 13, 15, -10 }, { 13, 15, -12 }, { 15, 16, -12 }, + { 16, 17, -12 } + } + } + }, + { + .bpp = DSC_BPP(10), .bpc = 14, + { 273, 15, 2048, 15, 24, 23, 23, { + { 0, 4, 10 }, { 3, 8, 8 }, { 6, 11, 6 }, { 9, 14, 4 }, + { 13, 15, 2 }, { 13, 15, 0 }, { 13, 16, -2 }, { 14, 16, -4 }, + { 15, 17, -6 }, { 15, 17, -8 }, { 16, 18, -10 }, + { 17, 19, -10 }, { 17, 19, -12 }, { 19, 20, -12 }, + { 20, 21, -12 } + } + } + }, + { + .bpp = DSC_BPP(10), .bpc = 16, + { 273, 15, 2048, 19, 28, 27, 27, { + { 0, 4, 10 }, { 4, 9, 8 }, { 8, 13, 6 }, { 12, 17, 4 }, + { 17, 19, 2 }, { 17, 20, 0 }, { 17, 20, -2 }, { 18, 20, -4 }, + { 19, 21, -6 }, { 19, 21, -8 }, { 20, 22, -10 }, + { 21, 23, -10 }, { 21, 23, -12 }, { 23, 24, -12 }, + { 24, 25, -12 } + } + } + }, + { /* sentinel */ } +}; + +/* + * Selected Rate Control Related Parameter Recommended Values for 4:2:2 from + * DSC v1.2, v1.2a, v1.2b + * + * Cross-checked against C Model releases: DSC_model_20161212 and 20210623 + */ +static const struct rc_parameters_data rc_parameters_1_2_420[] = { + { + .bpp = DSC_BPP(4), .bpc = 8, + { 512, 12, 6144, 3, 12, 11, 11, { + { 0, 4, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 }, + { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, + { 3, 9, -8 }, { 3, 10, -10 }, { 5, 10, -10 }, { 5, 11, -12 }, + { 5, 11, -12 }, { 9, 12, -12 }, { 12, 13, -12 } + } + } + }, + { + .bpp = DSC_BPP(4), .bpc = 10, + { 512, 12, 6144, 7, 16, 15, 15, { + { 0, 8, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 5, 10, -2 }, + { 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, + { 7, 13, -8 }, { 7, 14, -10 }, { 9, 14, -10 }, { 9, 15, -12 }, + { 9, 15, -12 }, { 13, 16, -12 }, { 16, 17, -12 } + } + } + }, + { + .bpp = DSC_BPP(4), .bpc = 12, + { 512, 12, 6144, 11, 20, 19, 19, { + { 0, 12, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 9, 14, -2 }, + { 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, + { 11, 17, -8 }, { 11, 18, -10 }, { 13, 18, -10 }, + { 13, 19, -12 }, { 13, 19, -12 }, { 17, 20, -12 }, + { 20, 21, -12 } + } + } + }, + { + .bpp = DSC_BPP(4), .bpc = 14, + { 512, 12, 6144, 15, 24, 23, 23, { + { 0, 12, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 12, 17, -2 }, + { 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 }, + { 15, 21, -8 }, { 15, 22, -10 }, { 17, 22, -10 }, + { 17, 23, -12 }, { 17, 23, -12 }, { 21, 24, -12 }, + { 24, 25, -12 } + } + } + }, + { + .bpp = DSC_BPP(4), .bpc = 16, + { 512, 12, 6144, 19, 28, 27, 27, { + { 0, 12, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 15, 20, -2 }, + { 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 }, + { 19, 25, -8 }, { 19, 26, -10 }, { 21, 26, -10 }, + { 21, 27, -12 }, { 21, 27, -12 }, { 25, 28, -12 }, + { 28, 29, -12 } + } + } + }, + { + .bpp = DSC_BPP(5), .bpc = 8, + { 410, 15, 5632, 3, 12, 11, 11, { + { 0, 3, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 2, 6, -2 }, + { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, + { 3, 9, -8 }, { 3, 9, -10 }, { 5, 10, -10 }, { 5, 10, -10 }, + { 5, 11, -12 }, { 7, 11, -12 }, { 11, 12, -12 } + } + } + }, + { + .bpp = DSC_BPP(5), .bpc = 10, + { 410, 15, 5632, 7, 16, 15, 15, { + { 0, 7, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 6, 10, -2 }, + { 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, + { 7, 13, -8 }, { 7, 13, -10 }, { 9, 14, -10 }, { 9, 14, -10 }, + { 9, 15, -12 }, { 11, 15, -12 }, { 15, 16, -12 } + } + } + }, + { + .bpp = DSC_BPP(5), .bpc = 12, + { 410, 15, 5632, 11, 20, 19, 19, { + { 0, 11, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 10, 14, -2 }, + { 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, + { 11, 17, -8 }, { 11, 17, -10 }, { 13, 18, -10 }, + { 13, 18, -10 }, { 13, 19, -12 }, { 15, 19, -12 }, + { 19, 20, -12 } + } + } + }, + { + .bpp = DSC_BPP(5), .bpc = 14, + { 410, 15, 5632, 15, 24, 23, 23, { + { 0, 11, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 13, 18, -2 }, + { 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 }, + { 15, 21, -8 }, { 15, 21, -10 }, { 17, 22, -10 }, + { 17, 22, -10 }, { 17, 23, -12 }, { 19, 23, -12 }, + { 23, 24, -12 } + } + } + }, + { + .bpp = DSC_BPP(5), .bpc = 16, + { 410, 15, 5632, 19, 28, 27, 27, { + { 0, 11, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 16, 20, -2 }, + { 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 }, + { 19, 25, -8 }, { 19, 25, -10 }, { 21, 26, -10 }, + { 21, 26, -10 }, { 21, 27, -12 }, { 23, 27, -12 }, + { 27, 28, -12 } + } + } + }, + { + .bpp = DSC_BPP(6), .bpc = 8, + { 341, 15, 2048, 3, 12, 11, 11, { + { 0, 2, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 }, + { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, + { 3, 8, -8 }, { 3, 9, -10 }, { 5, 9, -10 }, { 5, 9, -12 }, + { 5, 9, -12 }, { 7, 10, -12 }, { 10, 12, -12 } + } + } + }, + { + .bpp = DSC_BPP(6), .bpc = 10, + { 341, 15, 2048, 7, 16, 15, 15, { + { 0, 2, 2 }, { 2, 5, 0 }, { 3, 7, 0 }, { 4, 8, -2 }, + { 6, 9, -4 }, { 7, 10, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, + { 7, 12, -8 }, { 7, 13, -10 }, { 9, 13, -10 }, { 9, 13, -12 }, + { 9, 13, -12 }, { 11, 14, -12 }, { 14, 15, -12 } + } + } + }, + { + .bpp = DSC_BPP(6), .bpc = 12, + { 341, 15, 2048, 11, 20, 19, 19, { + { 0, 6, 2 }, { 4, 9, 0 }, { 7, 11, 0 }, { 8, 12, -2 }, + { 10, 13, -4 }, { 11, 14, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, + { 11, 16, -8 }, { 11, 17, -10 }, { 13, 17, -10 }, + { 13, 17, -12 }, { 13, 17, -12 }, { 15, 18, -12 }, + { 18, 19, -12 } + } + } + }, + { + .bpp = DSC_BPP(6), .bpc = 14, + { 341, 15, 2048, 15, 24, 23, 23, { + { 0, 6, 2 }, { 7, 10, 0 }, { 9, 13, 0 }, { 11, 16, -2 }, + { 14, 17, -4 }, { 15, 18, -6 }, { 15, 19, -8 }, { 15, 20, -8 }, + { 15, 20, -8 }, { 15, 21, -10 }, { 17, 21, -10 }, + { 17, 21, -12 }, { 17, 21, -12 }, { 19, 22, -12 }, + { 22, 23, -12 } + } + } + }, + { + .bpp = DSC_BPP(6), .bpc = 16, + { 341, 15, 2048, 19, 28, 27, 27, { + { 0, 6, 2 }, { 6, 11, 0 }, { 11, 15, 0 }, { 14, 18, -2 }, + { 18, 21, -4 }, { 19, 22, -6 }, { 19, 23, -8 }, { 19, 24, -8 }, + { 19, 24, -8 }, { 19, 25, -10 }, { 21, 25, -10 }, + { 21, 25, -12 }, { 21, 25, -12 }, { 23, 26, -12 }, + { 26, 27, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 8, + { 256, 15, 2048, 3, 12, 11, 11, { + { 0, 0, 10 }, { 0, 1, 8 }, { 0, 1, 6 }, { 0, 2, 4 }, + { 1, 2, 2 }, { 1, 3, 0 }, { 1, 3, -2 }, { 2, 4, -4 }, + { 2, 5, -6 }, { 3, 5, -8 }, { 4, 6, -10 }, { 4, 7, -10 }, + { 5, 7, -12 }, { 7, 8, -12 }, { 8, 9, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 10, + { 256, 15, 2048, 7, 16, 15, 15, { + { 0, 2, 10 }, { 2, 5, 8 }, { 3, 5, 6 }, { 4, 6, 4 }, + { 5, 6, 2 }, { 5, 7, 0 }, { 5, 7, -2 }, { 6, 8, -4 }, + { 6, 9, -6 }, { 7, 9, -8 }, { 8, 10, -10 }, { 8, 11, -10 }, + { 9, 11, -12 }, { 11, 12, -12 }, { 12, 13, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 12, + { 256, 15, 2048, 11, 20, 19, 19, { + { 0, 4, 10 }, { 2, 7, 8 }, { 4, 9, 6 }, { 6, 11, 4 }, + { 9, 11, 2 }, { 9, 11, 0 }, { 9, 12, -2 }, { 10, 12, -4 }, + { 11, 13, -6 }, { 11, 13, -8 }, { 12, 14, -10 }, + { 13, 15, -10 }, { 13, 15, -12 }, { 15, 16, -12 }, + { 16, 17, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 14, + { 256, 15, 2048, 15, 24, 23, 23, { + { 0, 4, 10 }, { 3, 8, 8 }, { 6, 11, 6 }, { 9, 14, 4 }, + { 13, 15, 2 }, { 13, 15, 0 }, { 13, 16, -2 }, { 14, 16, -4 }, + { 15, 17, -6 }, { 15, 17, -8 }, { 16, 18, -10 }, + { 17, 19, -10 }, { 17, 19, -12 }, { 19, 20, -12 }, + { 20, 21, -12 } + } + } + }, + { + .bpp = DSC_BPP(8), .bpc = 16, + { 256, 15, 2048, 19, 28, 27, 27, { + { 0, 4, 10 }, { 4, 9, 8 }, { 8, 13, 6 }, { 12, 17, 4 }, + { 17, 19, 2 }, { 17, 20, 0 }, { 17, 20, -2 }, { 18, 20, -4 }, + { 19, 21, -6 }, { 19, 21, -8 }, { 20, 22, -10 }, + { 21, 23, -10 }, { 21, 23, -12 }, { 23, 24, -12 }, + { 24, 25, -12 } + } + } + }, + { /* sentinel */ } +}; + +static const struct rc_parameters *get_rc_params(const struct rc_parameters_data *rc_parameters, + u16 dsc_bpp, + u8 bits_per_component) +{ + int i; + + for (i = 0; rc_parameters[i].bpp; i++) + if (rc_parameters[i].bpp == dsc_bpp && + rc_parameters[i].bpc == bits_per_component) + return &rc_parameters[i].params; + + return NULL; +} + +/** + * drm_dsc_setup_rc_params() - Set parameters and limits for RC model in + * accordance with the DSC 1.1 or 1.2 specification and DSC C Model + * Required bits_per_pixel and bits_per_component to be set before calling this + * function. + * + * @vdsc_cfg: DSC Configuration data partially filled by driver + * @type: operating mode and standard to follow + * + * Return: 0 or -error code in case of an error + */ +int drm_dsc_setup_rc_params(struct drm_dsc_config *vdsc_cfg, enum drm_dsc_params_type type) +{ + const struct rc_parameters_data *data; + const struct rc_parameters *rc_params; + int i; + + if (WARN_ON_ONCE(!vdsc_cfg->bits_per_pixel || + !vdsc_cfg->bits_per_component)) + return -EINVAL; + + switch (type) { + case DRM_DSC_1_2_444: + data = rc_parameters_1_2_444; + break; + case DRM_DSC_1_1_PRE_SCR: + data = rc_parameters_pre_scr; + break; + case DRM_DSC_1_2_422: + data = rc_parameters_1_2_422; + break; + case DRM_DSC_1_2_420: + data = rc_parameters_1_2_420; + break; + default: + return -EINVAL; + } + + rc_params = get_rc_params(data, + vdsc_cfg->bits_per_pixel, + vdsc_cfg->bits_per_component); + if (!rc_params) + return -EINVAL; + + vdsc_cfg->first_line_bpg_offset = rc_params->first_line_bpg_offset; + vdsc_cfg->initial_xmit_delay = rc_params->initial_xmit_delay; + vdsc_cfg->initial_offset = rc_params->initial_offset; + vdsc_cfg->flatness_min_qp = rc_params->flatness_min_qp; + vdsc_cfg->flatness_max_qp = rc_params->flatness_max_qp; + vdsc_cfg->rc_quant_incr_limit0 = rc_params->rc_quant_incr_limit0; + vdsc_cfg->rc_quant_incr_limit1 = rc_params->rc_quant_incr_limit1; + + for (i = 0; i < DSC_NUM_BUF_RANGES; i++) { + vdsc_cfg->rc_range_params[i].range_min_qp = + rc_params->rc_range_params[i].range_min_qp; + vdsc_cfg->rc_range_params[i].range_max_qp = + rc_params->rc_range_params[i].range_max_qp; + /* + * Range BPG Offset uses 2's complement and is only a 6 bits. So + * mask it to get only 6 bits. + */ + vdsc_cfg->rc_range_params[i].range_bpg_offset = + rc_params->rc_range_params[i].range_bpg_offset & + DSC_RANGE_BPG_OFFSET_MASK; + } + + return 0; +} +EXPORT_SYMBOL(drm_dsc_setup_rc_params); + /** * drm_dsc_compute_rc_parameters() - Write rate control * parameters to the dsc configuration defined in @@ -407,3 +1435,40 @@ int drm_dsc_compute_rc_parameters(struct drm_dsc_config *vdsc_cfg) return 0; } EXPORT_SYMBOL(drm_dsc_compute_rc_parameters); + +/** + * drm_dsc_get_bpp_int() - Get integer bits per pixel value for the given DRM DSC config + * @vdsc_cfg: Pointer to DRM DSC config struct + * + * Return: Integer BPP value + */ +u32 drm_dsc_get_bpp_int(const struct drm_dsc_config *vdsc_cfg) +{ + WARN_ON_ONCE(vdsc_cfg->bits_per_pixel & 0xf); + return vdsc_cfg->bits_per_pixel >> 4; +} +EXPORT_SYMBOL(drm_dsc_get_bpp_int); + +/** + * drm_dsc_initial_scale_value() - Calculate the initial scale value for the given DSC config + * @dsc: Pointer to DRM DSC config struct + * + * Return: Calculated initial scale value + */ +u8 drm_dsc_initial_scale_value(const struct drm_dsc_config *dsc) +{ + return 8 * dsc->rc_model_size / (dsc->rc_model_size - dsc->initial_offset); +} +EXPORT_SYMBOL(drm_dsc_initial_scale_value); + +/** + * drm_dsc_flatness_det_thresh() - Calculate the flatness_det_thresh for the given DSC config + * @dsc: Pointer to DRM DSC config struct + * + * Return: Calculated flatness det thresh value + */ +u32 drm_dsc_flatness_det_thresh(const struct drm_dsc_config *dsc) +{ + return 2 << (dsc->bits_per_component - 8); +} +EXPORT_SYMBOL(drm_dsc_flatness_det_thresh); diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c index b4c6ffc438da2fe4eb67a0d190b72cde95734e58..2c454568a607c367fd5ed9c3eb8ba3eaaf60529d 100644 --- a/drivers/gpu/drm/drm_atomic.c +++ b/drivers/gpu/drm/drm_atomic.c @@ -1131,6 +1131,7 @@ static void drm_atomic_connector_print_state(struct drm_printer *p, drm_printf(p, "\tcrtc=%s\n", state->crtc ? state->crtc->name : "(null)"); drm_printf(p, "\tself_refresh_aware=%d\n", state->self_refresh_aware); drm_printf(p, "\tmax_requested_bpc=%d\n", state->max_requested_bpc); + drm_printf(p, "\tcolorspace=%s\n", drm_get_colorspace_name(state->colorspace)); if (connector->connector_type == DRM_MODE_CONNECTOR_WRITEBACK) if (state->writeback_job && state->writeback_job->fb) diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c index 48df7a5ea503fbce0033f1daa2dcab46c716dd4b..3ed4cfcb350c4db19e0db135842cb65f4ddf2a08 100644 --- a/drivers/gpu/drm/drm_connector.c +++ b/drivers/gpu/drm/drm_connector.c @@ -1055,64 +1055,85 @@ static const struct drm_prop_enum_list drm_dp_subconnector_enum_list[] = { DRM_ENUM_NAME_FN(drm_get_dp_subconnector_name, drm_dp_subconnector_enum_list) -static const struct drm_prop_enum_list hdmi_colorspaces[] = { + +static const char * const colorspace_names[] = { /* For Default case, driver will set the colorspace */ - { DRM_MODE_COLORIMETRY_DEFAULT, "Default" }, + [DRM_MODE_COLORIMETRY_DEFAULT] = "Default", /* Standard Definition Colorimetry based on CEA 861 */ - { DRM_MODE_COLORIMETRY_SMPTE_170M_YCC, "SMPTE_170M_YCC" }, - { DRM_MODE_COLORIMETRY_BT709_YCC, "BT709_YCC" }, + [DRM_MODE_COLORIMETRY_SMPTE_170M_YCC] = "SMPTE_170M_YCC", + [DRM_MODE_COLORIMETRY_BT709_YCC] = "BT709_YCC", /* Standard Definition Colorimetry based on IEC 61966-2-4 */ - { DRM_MODE_COLORIMETRY_XVYCC_601, "XVYCC_601" }, + [DRM_MODE_COLORIMETRY_XVYCC_601] = "XVYCC_601", /* High Definition Colorimetry based on IEC 61966-2-4 */ - { DRM_MODE_COLORIMETRY_XVYCC_709, "XVYCC_709" }, + [DRM_MODE_COLORIMETRY_XVYCC_709] = "XVYCC_709", /* Colorimetry based on IEC 61966-2-1/Amendment 1 */ - { DRM_MODE_COLORIMETRY_SYCC_601, "SYCC_601" }, + [DRM_MODE_COLORIMETRY_SYCC_601] = "SYCC_601", /* Colorimetry based on IEC 61966-2-5 [33] */ - { DRM_MODE_COLORIMETRY_OPYCC_601, "opYCC_601" }, + [DRM_MODE_COLORIMETRY_OPYCC_601] = "opYCC_601", /* Colorimetry based on IEC 61966-2-5 */ - { DRM_MODE_COLORIMETRY_OPRGB, "opRGB" }, + [DRM_MODE_COLORIMETRY_OPRGB] = "opRGB", /* Colorimetry based on ITU-R BT.2020 */ - { DRM_MODE_COLORIMETRY_BT2020_CYCC, "BT2020_CYCC" }, + [DRM_MODE_COLORIMETRY_BT2020_CYCC] = "BT2020_CYCC", /* Colorimetry based on ITU-R BT.2020 */ - { DRM_MODE_COLORIMETRY_BT2020_RGB, "BT2020_RGB" }, + [DRM_MODE_COLORIMETRY_BT2020_RGB] = "BT2020_RGB", /* Colorimetry based on ITU-R BT.2020 */ - { DRM_MODE_COLORIMETRY_BT2020_YCC, "BT2020_YCC" }, + [DRM_MODE_COLORIMETRY_BT2020_YCC] = "BT2020_YCC", /* Added as part of Additional Colorimetry Extension in 861.G */ - { DRM_MODE_COLORIMETRY_DCI_P3_RGB_D65, "DCI-P3_RGB_D65" }, - { DRM_MODE_COLORIMETRY_DCI_P3_RGB_THEATER, "DCI-P3_RGB_Theater" }, + [DRM_MODE_COLORIMETRY_DCI_P3_RGB_D65] = "DCI-P3_RGB_D65", + [DRM_MODE_COLORIMETRY_DCI_P3_RGB_THEATER] = "DCI-P3_RGB_Theater", + [DRM_MODE_COLORIMETRY_RGB_WIDE_FIXED] = "RGB_WIDE_FIXED", + /* Colorimetry based on scRGB (IEC 61966-2-2) */ + [DRM_MODE_COLORIMETRY_RGB_WIDE_FLOAT] = "RGB_WIDE_FLOAT", + [DRM_MODE_COLORIMETRY_BT601_YCC] = "BT601_YCC", }; +/** + * drm_get_colorspace_name - return a string for color encoding + * @colorspace: color space to compute name of + * + * In contrast to the other drm_get_*_name functions this one here returns a + * const pointer and hence is threadsafe. + */ +const char *drm_get_colorspace_name(enum drm_colorspace colorspace) +{ + if (colorspace < ARRAY_SIZE(colorspace_names) && colorspace_names[colorspace]) + return colorspace_names[colorspace]; + else + return "(null)"; +} + +static const u32 hdmi_colorspaces = + BIT(DRM_MODE_COLORIMETRY_SMPTE_170M_YCC) | + BIT(DRM_MODE_COLORIMETRY_BT709_YCC) | + BIT(DRM_MODE_COLORIMETRY_XVYCC_601) | + BIT(DRM_MODE_COLORIMETRY_XVYCC_709) | + BIT(DRM_MODE_COLORIMETRY_SYCC_601) | + BIT(DRM_MODE_COLORIMETRY_OPYCC_601) | + BIT(DRM_MODE_COLORIMETRY_OPRGB) | + BIT(DRM_MODE_COLORIMETRY_BT2020_CYCC) | + BIT(DRM_MODE_COLORIMETRY_BT2020_RGB) | + BIT(DRM_MODE_COLORIMETRY_BT2020_YCC) | + BIT(DRM_MODE_COLORIMETRY_DCI_P3_RGB_D65) | + BIT(DRM_MODE_COLORIMETRY_DCI_P3_RGB_THEATER); + /* * As per DP 1.4a spec, 2.2.5.7.5 VSC SDP Payload for Pixel Encoding/Colorimetry * Format Table 2-120 */ -static const struct drm_prop_enum_list dp_colorspaces[] = { - /* For Default case, driver will set the colorspace */ - { DRM_MODE_COLORIMETRY_DEFAULT, "Default" }, - { DRM_MODE_COLORIMETRY_RGB_WIDE_FIXED, "RGB_Wide_Gamut_Fixed_Point" }, - /* Colorimetry based on scRGB (IEC 61966-2-2) */ - { DRM_MODE_COLORIMETRY_RGB_WIDE_FLOAT, "RGB_Wide_Gamut_Floating_Point" }, - /* Colorimetry based on IEC 61966-2-5 */ - { DRM_MODE_COLORIMETRY_OPRGB, "opRGB" }, - /* Colorimetry based on SMPTE RP 431-2 */ - { DRM_MODE_COLORIMETRY_DCI_P3_RGB_D65, "DCI-P3_RGB_D65" }, - /* Colorimetry based on ITU-R BT.2020 */ - { DRM_MODE_COLORIMETRY_BT2020_RGB, "BT2020_RGB" }, - { DRM_MODE_COLORIMETRY_BT601_YCC, "BT601_YCC" }, - { DRM_MODE_COLORIMETRY_BT709_YCC, "BT709_YCC" }, - /* Standard Definition Colorimetry based on IEC 61966-2-4 */ - { DRM_MODE_COLORIMETRY_XVYCC_601, "XVYCC_601" }, - /* High Definition Colorimetry based on IEC 61966-2-4 */ - { DRM_MODE_COLORIMETRY_XVYCC_709, "XVYCC_709" }, - /* Colorimetry based on IEC 61966-2-1/Amendment 1 */ - { DRM_MODE_COLORIMETRY_SYCC_601, "SYCC_601" }, - /* Colorimetry based on IEC 61966-2-5 [33] */ - { DRM_MODE_COLORIMETRY_OPYCC_601, "opYCC_601" }, - /* Colorimetry based on ITU-R BT.2020 */ - { DRM_MODE_COLORIMETRY_BT2020_CYCC, "BT2020_CYCC" }, - /* Colorimetry based on ITU-R BT.2020 */ - { DRM_MODE_COLORIMETRY_BT2020_YCC, "BT2020_YCC" }, -}; +static const u32 dp_colorspaces = + BIT(DRM_MODE_COLORIMETRY_RGB_WIDE_FIXED) | + BIT(DRM_MODE_COLORIMETRY_RGB_WIDE_FLOAT) | + BIT(DRM_MODE_COLORIMETRY_OPRGB) | + BIT(DRM_MODE_COLORIMETRY_DCI_P3_RGB_D65) | + BIT(DRM_MODE_COLORIMETRY_BT2020_RGB) | + BIT(DRM_MODE_COLORIMETRY_BT601_YCC) | + BIT(DRM_MODE_COLORIMETRY_BT709_YCC) | + BIT(DRM_MODE_COLORIMETRY_XVYCC_601) | + BIT(DRM_MODE_COLORIMETRY_XVYCC_709) | + BIT(DRM_MODE_COLORIMETRY_SYCC_601) | + BIT(DRM_MODE_COLORIMETRY_OPYCC_601) | + BIT(DRM_MODE_COLORIMETRY_BT2020_CYCC) | + BIT(DRM_MODE_COLORIMETRY_BT2020_YCC); /** * DOC: standard connector properties @@ -2135,33 +2156,72 @@ EXPORT_SYMBOL(drm_mode_create_aspect_ratio_property); * drm_mode_create_dp_colorspace_property() is used for DP connector. */ -/** - * drm_mode_create_hdmi_colorspace_property - create hdmi colorspace property - * @connector: connector to create the Colorspace property on. - * - * Called by a driver the first time it's needed, must be attached to desired - * HDMI connectors. - * - * Returns: - * Zero on success, negative errno on failure. - */ -int drm_mode_create_hdmi_colorspace_property(struct drm_connector *connector) +static int drm_mode_create_colorspace_property(struct drm_connector *connector, + u32 supported_colorspaces) { struct drm_device *dev = connector->dev; + u32 colorspaces = supported_colorspaces | BIT(DRM_MODE_COLORIMETRY_DEFAULT); + struct drm_prop_enum_list enum_list[DRM_MODE_COLORIMETRY_COUNT]; + int i, len; if (connector->colorspace_property) return 0; + if (!supported_colorspaces) { + drm_err(dev, "No supported colorspaces provded on [CONNECTOR:%d:%s]\n", + connector->base.id, connector->name); + return -EINVAL; + } + + if ((supported_colorspaces & -BIT(DRM_MODE_COLORIMETRY_COUNT)) != 0) { + drm_err(dev, "Unknown colorspace provded on [CONNECTOR:%d:%s]\n", + connector->base.id, connector->name); + return -EINVAL; + } + + len = 0; + for (i = 0; i < DRM_MODE_COLORIMETRY_COUNT; i++) { + if ((colorspaces & BIT(i)) == 0) + continue; + + enum_list[len].type = i; + enum_list[len].name = colorspace_names[i]; + len++; + } + connector->colorspace_property = drm_property_create_enum(dev, DRM_MODE_PROP_ENUM, "Colorspace", - hdmi_colorspaces, - ARRAY_SIZE(hdmi_colorspaces)); + enum_list, + len); if (!connector->colorspace_property) return -ENOMEM; return 0; } + +/** + * drm_mode_create_hdmi_colorspace_property - create hdmi colorspace property + * @connector: connector to create the Colorspace property on. + * + * Called by a driver the first time it's needed, must be attached to desired + * HDMI connectors. + * + * Returns: + * Zero on success, negative errno on failure. + */ +int drm_mode_create_hdmi_colorspace_property(struct drm_connector *connector, + u32 supported_colorspaces) +{ + u32 colorspaces; + + if (supported_colorspaces) + colorspaces = supported_colorspaces & hdmi_colorspaces; + else + colorspaces = hdmi_colorspaces; + + return drm_mode_create_colorspace_property(connector, colorspaces); +} EXPORT_SYMBOL(drm_mode_create_hdmi_colorspace_property); /** @@ -2174,22 +2234,17 @@ EXPORT_SYMBOL(drm_mode_create_hdmi_colorspace_property); * Returns: * Zero on success, negative errno on failure. */ -int drm_mode_create_dp_colorspace_property(struct drm_connector *connector) +int drm_mode_create_dp_colorspace_property(struct drm_connector *connector, + u32 supported_colorspaces) { - struct drm_device *dev = connector->dev; + u32 colorspaces; - if (connector->colorspace_property) - return 0; - - connector->colorspace_property = - drm_property_create_enum(dev, DRM_MODE_PROP_ENUM, "Colorspace", - dp_colorspaces, - ARRAY_SIZE(dp_colorspaces)); - - if (!connector->colorspace_property) - return -ENOMEM; + if (supported_colorspaces) + colorspaces = supported_colorspaces & dp_colorspaces; + else + colorspaces = dp_colorspaces; - return 0; + return drm_mode_create_colorspace_property(connector, colorspaces); } EXPORT_SYMBOL(drm_mode_create_dp_colorspace_property); diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c index 0454da505687b7ab7f51e2ee61d95b3669a89a4f..e0dbd9140726b336669db2db854b2556a463ebd1 100644 --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -2844,6 +2844,35 @@ struct edid *drm_get_edid_switcheroo(struct drm_connector *connector, } EXPORT_SYMBOL(drm_get_edid_switcheroo); +/** + * drm_edid_read_switcheroo - get EDID data for a vga_switcheroo output + * @connector: connector we're probing + * @adapter: I2C adapter to use for DDC + * + * Wrapper around drm_edid_read_ddc() for laptops with dual GPUs using one set + * of outputs. The wrapper adds the requisite vga_switcheroo calls to + * temporarily switch DDC to the GPU which is retrieving EDID. + * + * Return: Pointer to valid EDID or %NULL if we couldn't find any. + */ +const struct drm_edid *drm_edid_read_switcheroo(struct drm_connector *connector, + struct i2c_adapter *adapter) +{ + struct drm_device *dev = connector->dev; + struct pci_dev *pdev = to_pci_dev(dev->dev); + const struct drm_edid *drm_edid; + + if (drm_WARN_ON_ONCE(dev, !dev_is_pci(dev->dev))) + return NULL; + + vga_switcheroo_lock_ddc(pdev); + drm_edid = drm_edid_read_ddc(connector, adapter); + vga_switcheroo_unlock_ddc(pdev); + + return drm_edid; +} +EXPORT_SYMBOL(drm_edid_read_switcheroo); + /** * drm_edid_duplicate - duplicate an EDID and the extensions * @edid: EDID to duplicate @@ -2852,6 +2881,9 @@ EXPORT_SYMBOL(drm_get_edid_switcheroo); */ struct edid *drm_edid_duplicate(const struct edid *edid) { + if (!edid) + return NULL; + return kmemdup(edid, edid_size(edid), GFP_KERNEL); } EXPORT_SYMBOL(drm_edid_duplicate); @@ -6243,6 +6275,9 @@ static void drm_parse_cea_ext(struct drm_connector *connector, info->color_formats |= DRM_COLOR_FORMAT_YCBCR444; if (edid_ext[3] & EDID_CEA_YCRCB422) info->color_formats |= DRM_COLOR_FORMAT_YCBCR422; + if (edid_ext[3] & EDID_BASIC_AUDIO) + info->has_audio = true; + } drm_edid_iter_end(&edid_iter); @@ -6268,6 +6303,8 @@ static void drm_parse_cea_ext(struct drm_connector *connector, drm_parse_hdr_metadata_block(connector, data); else if (cea_db_tag(db) == CTA_DB_VIDEO) parse_cta_vdb(connector, db); + else if (cea_db_tag(db) == CTA_DB_AUDIO) + info->has_audio = true; } cea_db_iter_end(&iter); @@ -6424,6 +6461,7 @@ static void drm_reset_display_info(struct drm_connector *connector) info->max_tmds_clock = 0; info->dvi_dual = false; info->is_hdmi = false; + info->has_audio = false; info->has_hdmi_infoframe = false; info->rgb_quant_range_selectable = false; memset(&info->hdmi, 0, sizeof(info->hdmi)); diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c index 9978147bbc8ab52866116a3d3696e41ba7fccec4..61a5d450cc20ef0a40400357498b21eeaa8d092d 100644 --- a/drivers/gpu/drm/drm_fb_helper.c +++ b/drivers/gpu/drm/drm_fb_helper.c @@ -1187,17 +1187,19 @@ static void drm_fb_helper_fill_pixel_fmt(struct fb_var_screeninfo *var, } } -static void __fill_var(struct fb_var_screeninfo *var, +static void __fill_var(struct fb_var_screeninfo *var, struct fb_info *info, struct drm_framebuffer *fb) { int i; var->xres_virtual = fb->width; var->yres_virtual = fb->height; - var->accel_flags = FB_ACCELF_TEXT; + var->accel_flags = 0; var->bits_per_pixel = drm_format_info_bpp(fb->format, 0); - var->height = var->width = 0; + var->height = info->var.height; + var->width = info->var.width; + var->left_margin = var->right_margin = 0; var->upper_margin = var->lower_margin = 0; var->hsync_len = var->vsync_len = 0; @@ -1260,7 +1262,7 @@ int drm_fb_helper_check_var(struct fb_var_screeninfo *var, return -EINVAL; } - __fill_var(var, fb); + __fill_var(var, info, fb); /* * fb_pan_display() validates this, but fb_set_par() doesn't and just @@ -1716,7 +1718,7 @@ static void drm_fb_helper_fill_var(struct fb_info *info, info->pseudo_palette = fb_helper->pseudo_palette; info->var.xoffset = 0; info->var.yoffset = 0; - __fill_var(&info->var, fb); + __fill_var(&info->var, info, fb); info->var.activate = FB_ACTIVATE_NOW; drm_fb_helper_fill_pixel_fmt(&info->var, format); diff --git a/drivers/gpu/drm/drm_managed.c b/drivers/gpu/drm/drm_managed.c index 4cf214de50c4037ab659fda9097f2e6e34da7d06..c21c3f6230335f4a93a2c9cf0c4d2768138a72bb 100644 --- a/drivers/gpu/drm/drm_managed.c +++ b/drivers/gpu/drm/drm_managed.c @@ -264,28 +264,10 @@ void drmm_kfree(struct drm_device *dev, void *data) } EXPORT_SYMBOL(drmm_kfree); -static void drmm_mutex_release(struct drm_device *dev, void *res) +void __drmm_mutex_release(struct drm_device *dev, void *res) { struct mutex *lock = res; mutex_destroy(lock); } - -/** - * drmm_mutex_init - &drm_device-managed mutex_init() - * @dev: DRM device - * @lock: lock to be initialized - * - * Returns: - * 0 on success, or a negative errno code otherwise. - * - * This is a &drm_device-managed version of mutex_init(). The initialized - * lock is automatically destroyed on the final drm_dev_put(). - */ -int drmm_mutex_init(struct drm_device *dev, struct mutex *lock) -{ - mutex_init(lock); - - return drmm_add_action_or_reset(dev, drmm_mutex_release, lock); -} -EXPORT_SYMBOL(drmm_mutex_init); +EXPORT_SYMBOL(__drmm_mutex_release); diff --git a/drivers/gpu/drm/drm_panel_orientation_quirks.c b/drivers/gpu/drm/drm_panel_orientation_quirks.c index b1a38e6ce2f8fad28bbef491a02c29b7257beaa1..0cb646cb04ee1ae5028aa03439df18f06925760a 100644 --- a/drivers/gpu/drm/drm_panel_orientation_quirks.c +++ b/drivers/gpu/drm/drm_panel_orientation_quirks.c @@ -179,7 +179,7 @@ static const struct dmi_system_id orientation_data[] = { }, { /* AYA NEO AIR */ .matches = { DMI_EXACT_MATCH(DMI_SYS_VENDOR, "AYANEO"), - DMI_MATCH(DMI_BOARD_NAME, "AIR"), + DMI_MATCH(DMI_PRODUCT_NAME, "AIR"), }, .driver_data = (void *)&lcd1080x1920_leftside_up, }, { /* AYA NEO NEXT */ diff --git a/drivers/gpu/drm/exynos/exynos_drm_g2d.c b/drivers/gpu/drm/exynos/exynos_drm_g2d.c index ec784e58da5c1cd612059c5f80daea4316410309..414e585ec7dd0b0561f338da1b24f623fe4462b0 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_g2d.c +++ b/drivers/gpu/drm/exynos/exynos_drm_g2d.c @@ -1335,7 +1335,7 @@ int exynos_g2d_exec_ioctl(struct drm_device *drm_dev, void *data, /* Let the runqueue know that there is work to do. */ queue_work(g2d->g2d_workq, &g2d->runqueue_work); - if (runqueue_node->async) + if (req->async) goto out; wait_for_completion(&runqueue_node->complete); diff --git a/drivers/gpu/drm/exynos/exynos_drm_g2d.h b/drivers/gpu/drm/exynos/exynos_drm_g2d.h index 74ea3c26deadcebabc7cae82b4e0d261269a2761..1a5ae781b56c60e41af91a5725e8cffa3e6f7b46 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_g2d.h +++ b/drivers/gpu/drm/exynos/exynos_drm_g2d.h @@ -34,11 +34,11 @@ static inline int exynos_g2d_exec_ioctl(struct drm_device *dev, void *data, return -ENODEV; } -int g2d_open(struct drm_device *drm_dev, struct drm_file *file) +static inline int g2d_open(struct drm_device *drm_dev, struct drm_file *file) { return 0; } -void g2d_close(struct drm_device *drm_dev, struct drm_file *file) +static inline void g2d_close(struct drm_device *drm_dev, struct drm_file *file) { } #endif diff --git a/drivers/gpu/drm/exynos/exynos_drm_vidi.c b/drivers/gpu/drm/exynos/exynos_drm_vidi.c index 4d56c8c799c5a9d5a7faa2133a5e21cd6c18536f..f5e1adfcaa514ee0a84b00b1c7357e7d43a64ecf 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_vidi.c +++ b/drivers/gpu/drm/exynos/exynos_drm_vidi.c @@ -469,8 +469,6 @@ static int vidi_remove(struct platform_device *pdev) if (ctx->raw_edid != (struct edid *)fake_edid_info) { kfree(ctx->raw_edid); ctx->raw_edid = NULL; - - return -EINVAL; } component_del(&pdev->dev, &vidi_component_ops); diff --git a/drivers/gpu/drm/i915/Kconfig.debug b/drivers/gpu/drm/i915/Kconfig.debug index 47e845353ffad82547ec6cf864a2ff4bf09f0e13..2d21930d5501580a1ecb83c91f44c3fcf05d4a00 100644 --- a/drivers/gpu/drm/i915/Kconfig.debug +++ b/drivers/gpu/drm/i915/Kconfig.debug @@ -157,6 +157,7 @@ config DRM_I915_SW_FENCE_CHECK_DAG config DRM_I915_DEBUG_GUC bool "Enable additional driver debugging for GuC" depends on DRM_I915 + select STACKDEPOT default n help Choose this option to turn on extra driver debugging that may affect diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index c58d7b19366442256e3d9ed04c34d8d8f8ffe1b2..789dce9e26088c25b2cd58d76de865ad9378295f 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -19,14 +19,10 @@ subdir-ccflags-y += -Wno-type-limits subdir-ccflags-y += -Wno-missing-field-initializers subdir-ccflags-y += -Wno-sign-compare subdir-ccflags-y += -Wno-shift-negative-value -subdir-ccflags-y += $(call cc-disable-warning, unused-but-set-variable) +subdir-ccflags-y += $(call cc-option, -Wunused-but-set-variable) subdir-ccflags-y += $(call cc-disable-warning, frame-address) subdir-ccflags-$(CONFIG_DRM_I915_WERROR) += -Werror -# Fine grained warnings disable -CFLAGS_i915_pci.o = $(call cc-disable-warning, override-init) -CFLAGS_display/intel_fbdev.o = $(call cc-disable-warning, override-init) - subdir-ccflags-y += -I$(srctree)/$(src) # Please keep these build lists sorted! @@ -238,9 +234,12 @@ i915-y += \ display/intel_crtc_state_dump.o \ display/intel_cursor.o \ display/intel_display.o \ + display/intel_display_driver.o \ + display/intel_display_irq.o \ display/intel_display_power.o \ display/intel_display_power_map.o \ display/intel_display_power_well.o \ + display/intel_display_reset.o \ display/intel_display_rps.o \ display/intel_dmc.o \ display/intel_dpio_phy.o \ @@ -259,14 +258,18 @@ i915-y += \ display/intel_hdcp.o \ display/intel_hdcp_gsc.o \ display/intel_hotplug.o \ + display/intel_hotplug_irq.o \ display/intel_hti.o \ + display/intel_load_detect.o \ display/intel_lpe_audio.o \ + display/intel_modeset_lock.o \ display/intel_modeset_verify.o \ display/intel_modeset_setup.o \ display/intel_overlay.o \ display/intel_pch_display.o \ display/intel_pch_refclk.o \ display/intel_plane_initial.o \ + display/intel_pmdemand.o \ display/intel_psr.o \ display/intel_quirks.o \ display/intel_sprite.o \ @@ -299,8 +302,10 @@ i915-y += \ display/icl_dsi.o \ display/intel_backlight.o \ display/intel_crt.o \ + display/intel_cx0_phy.o \ display/intel_ddi.o \ display/intel_ddi_buf_trans.o \ + display/intel_display_device.o \ display/intel_display_trace.o \ display/intel_dkl_phy.o \ display/intel_dp.o \ diff --git a/drivers/gpu/drm/i915/display/g4x_dp.c b/drivers/gpu/drm/i915/display/g4x_dp.c index 920d570f759433e3467d949b7c1994653fb46f49..112d91d81fdc4c3585d9472a9c02f97a0bfb172f 100644 --- a/drivers/gpu/drm/i915/display/g4x_dp.c +++ b/drivers/gpu/drm/i915/display/g4x_dp.c @@ -169,7 +169,7 @@ static void assert_dp_port(struct intel_dp *intel_dp, bool state) struct drm_i915_private *dev_priv = to_i915(dig_port->base.base.dev); bool cur_state = intel_de_read(dev_priv, intel_dp->output_reg) & DP_PORT_EN; - I915_STATE_WARN(cur_state != state, + I915_STATE_WARN(dev_priv, cur_state != state, "[ENCODER:%d:%s] state assertion failure (expected %s, current %s)\n", dig_port->base.base.base.id, dig_port->base.base.name, str_on_off(state), str_on_off(cur_state)); @@ -180,7 +180,7 @@ static void assert_edp_pll(struct drm_i915_private *dev_priv, bool state) { bool cur_state = intel_de_read(dev_priv, DP_A) & DP_PLL_ENABLE; - I915_STATE_WARN(cur_state != state, + I915_STATE_WARN(dev_priv, cur_state != state, "eDP PLL state assertion failure (expected %s, current %s)\n", str_on_off(state), str_on_off(cur_state)); } diff --git a/drivers/gpu/drm/i915/display/g4x_hdmi.c b/drivers/gpu/drm/i915/display/g4x_hdmi.c index 448ea26786e0f27f5d497f9a7da9efd3ead85fe0..5c187e6e0472ae4584649acb6b0ef33c772fc1df 100644 --- a/drivers/gpu/drm/i915/display/g4x_hdmi.c +++ b/drivers/gpu/drm/i915/display/g4x_hdmi.c @@ -7,6 +7,7 @@ #include "g4x_hdmi.h" #include "i915_reg.h" +#include "intel_atomic.h" #include "intel_audio.h" #include "intel_connector.h" #include "intel_crtc.h" @@ -80,15 +81,67 @@ static bool intel_hdmi_get_hw_state(struct intel_encoder *encoder, return ret; } +static bool connector_is_hdmi(struct drm_connector *connector) +{ + struct intel_encoder *encoder = + intel_attached_encoder(to_intel_connector(connector)); + + return encoder && encoder->type == INTEL_OUTPUT_HDMI; +} + +static bool g4x_compute_has_hdmi_sink(struct intel_atomic_state *state, + struct intel_crtc *this_crtc) +{ + const struct drm_connector_state *conn_state; + struct drm_connector *connector; + int i; + + /* + * On g4x only one HDMI port can transmit infoframes/audio at + * any given time. Select the first suitable port for this duty. + * + * See also g4x_hdmi_connector_atomic_check(). + */ + for_each_new_connector_in_state(&state->base, connector, conn_state, i) { + struct intel_encoder *encoder = to_intel_encoder(conn_state->best_encoder); + const struct intel_crtc_state *crtc_state; + struct intel_crtc *crtc; + + if (!connector_is_hdmi(connector)) + continue; + + crtc = to_intel_crtc(conn_state->crtc); + if (!crtc) + continue; + + crtc_state = intel_atomic_get_new_crtc_state(state, crtc); + + if (!intel_hdmi_compute_has_hdmi_sink(encoder, crtc_state, conn_state)) + continue; + + return crtc == this_crtc; + } + + return false; +} + static int g4x_hdmi_compute_config(struct intel_encoder *encoder, struct intel_crtc_state *crtc_state, struct drm_connector_state *conn_state) { + struct intel_atomic_state *state = to_intel_atomic_state(crtc_state->uapi.state); + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct drm_i915_private *i915 = to_i915(encoder->base.dev); if (HAS_PCH_SPLIT(i915)) crtc_state->has_pch_encoder = true; + if (IS_G4X(i915)) + crtc_state->has_hdmi_sink = g4x_compute_has_hdmi_sink(state, crtc); + else + crtc_state->has_hdmi_sink = + intel_hdmi_compute_has_hdmi_sink(encoder, crtc_state, conn_state); + return intel_hdmi_compute_config(encoder, crtc_state, conn_state); } @@ -546,6 +599,66 @@ intel_hdmi_hotplug(struct intel_encoder *encoder, return state; } +int g4x_hdmi_connector_atomic_check(struct drm_connector *connector, + struct drm_atomic_state *state) +{ + struct drm_i915_private *i915 = to_i915(state->dev); + struct drm_connector_list_iter conn_iter; + struct drm_connector *conn; + int ret; + + ret = intel_digital_connector_atomic_check(connector, state); + if (ret) + return ret; + + if (!IS_G4X(i915)) + return 0; + + if (!intel_connector_needs_modeset(to_intel_atomic_state(state), connector)) + return 0; + + /* + * On g4x only one HDMI port can transmit infoframes/audio + * at any given time. Make sure all enabled HDMI ports are + * included in the state so that it's possible to select + * one of them for this duty. + * + * See also g4x_compute_has_hdmi_sink(). + */ + drm_connector_list_iter_begin(&i915->drm, &conn_iter); + drm_for_each_connector_iter(conn, &conn_iter) { + struct drm_connector_state *conn_state; + struct drm_crtc_state *crtc_state; + struct drm_crtc *crtc; + + if (!connector_is_hdmi(conn)) + continue; + + drm_dbg_kms(&i915->drm, "Adding [CONNECTOR:%d:%s]\n", + conn->base.id, conn->name); + + conn_state = drm_atomic_get_connector_state(state, conn); + if (IS_ERR(conn_state)) { + ret = PTR_ERR(conn_state); + break; + } + + crtc = conn_state->crtc; + if (!crtc) + continue; + + crtc_state = drm_atomic_get_new_crtc_state(state, crtc); + crtc_state->mode_changed = true; + + ret = drm_atomic_add_affected_planes(state, crtc); + if (ret) + break; + } + drm_connector_list_iter_end(&conn_iter); + + return ret; +} + void g4x_hdmi_init(struct drm_i915_private *dev_priv, i915_reg_t hdmi_reg, enum port port) { diff --git a/drivers/gpu/drm/i915/display/g4x_hdmi.h b/drivers/gpu/drm/i915/display/g4x_hdmi.h index db9a93bc9321efedb320d8eb966e8281cadc4f9c..1e3ea7f3c846fd4b7427e650ccbc50f7128eb970 100644 --- a/drivers/gpu/drm/i915/display/g4x_hdmi.h +++ b/drivers/gpu/drm/i915/display/g4x_hdmi.h @@ -11,9 +11,13 @@ #include "i915_reg_defs.h" enum port; +struct drm_atomic_state; +struct drm_connector; struct drm_i915_private; void g4x_hdmi_init(struct drm_i915_private *dev_priv, i915_reg_t hdmi_reg, enum port port); +int g4x_hdmi_connector_atomic_check(struct drm_connector *connector, + struct drm_atomic_state *state); #endif diff --git a/drivers/gpu/drm/i915/display/i9xx_plane.c b/drivers/gpu/drm/i915/display/i9xx_plane.c index ecaeb7dc196bcc8cee193d57a664e4c4dd765385..b104883244576b1420c3c3f0de71b0336684b47c 100644 --- a/drivers/gpu/drm/i915/display/i9xx_plane.c +++ b/drivers/gpu/drm/i915/display/i9xx_plane.c @@ -8,12 +8,12 @@ #include #include -#include "i915_irq.h" #include "i915_reg.h" #include "i9xx_plane.h" #include "intel_atomic.h" #include "intel_atomic_plane.h" #include "intel_de.h" +#include "intel_display_irq.h" #include "intel_display_types.h" #include "intel_fb.h" #include "intel_fbc.h" @@ -1033,10 +1033,13 @@ i9xx_get_initial_plane_config(struct intel_crtc *crtc, DSPLINOFF(i9xx_plane)); base = intel_de_read(dev_priv, DSPSURF(i9xx_plane)) & DISP_ADDR_MASK; } else { + offset = 0; base = intel_de_read(dev_priv, DSPADDR(i9xx_plane)); } plane_config->base = base; + drm_WARN_ON(&dev_priv->drm, offset != 0); + val = intel_de_read(dev_priv, PIPESRC(pipe)); fb->width = REG_FIELD_GET(PIPESRC_WIDTH_MASK, val) + 1; fb->height = REG_FIELD_GET(PIPESRC_HEIGHT_MASK, val) + 1; diff --git a/drivers/gpu/drm/i915/display/i9xx_wm.c b/drivers/gpu/drm/i915/display/i9xx_wm.c index caef72d387981d7b2dcf6d7324bf6fcaea058687..af0c79a4c9a4f50d1e41f8e438ad475f6dedfcd4 100644 --- a/drivers/gpu/drm/i915/display/i9xx_wm.c +++ b/drivers/gpu/drm/i915/display/i9xx_wm.c @@ -4,6 +4,7 @@ */ #include "i915_drv.h" +#include "i915_reg.h" #include "i9xx_wm.h" #include "intel_atomic.h" #include "intel_display.h" @@ -3447,9 +3448,10 @@ void ilk_wm_sanitize(struct drm_i915_private *dev_priv) drm_modeset_acquire_init(&ctx, 0); -retry: state->acquire_ctx = &ctx; + to_intel_atomic_state(state)->internal = true; +retry: /* * Hardware readout is the only time we don't want to calculate * intermediate watermarks (since we don't trust the current diff --git a/drivers/gpu/drm/i915/display/i9xx_wm.h b/drivers/gpu/drm/i915/display/i9xx_wm.h index a7875cbcd05aae82d6833cd969651f0892e50653..b87ae369685a90e935a94687f92c7059578a9058 100644 --- a/drivers/gpu/drm/i915/display/i9xx_wm.h +++ b/drivers/gpu/drm/i915/display/i9xx_wm.h @@ -12,7 +12,6 @@ struct drm_i915_private; struct intel_crtc_state; struct intel_plane_state; -int ilk_wm_max_level(const struct drm_i915_private *i915); bool ilk_disable_lp_wm(struct drm_i915_private *i915); void ilk_wm_sanitize(struct drm_i915_private *i915); bool intel_set_memory_cxsr(struct drm_i915_private *i915, bool enable); diff --git a/drivers/gpu/drm/i915/display/icl_dsi.c b/drivers/gpu/drm/i915/display/icl_dsi.c index c9aeba0ecf911fbd6dfb50167443f1bc7297dd92..c133928a0655c72d435270236fd7b839cbf949b8 100644 --- a/drivers/gpu/drm/i915/display/icl_dsi.c +++ b/drivers/gpu/drm/i915/display/icl_dsi.c @@ -1591,6 +1591,7 @@ static int gen11_dsi_compute_config(struct intel_encoder *encoder, &pipe_config->hw.adjusted_mode; int ret; + pipe_config->sink_format = INTEL_OUTPUT_FORMAT_RGB; pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB; ret = intel_panel_compute_config(intel_connector, adjusted_mode); diff --git a/drivers/gpu/drm/i915/display/intel_atomic.c b/drivers/gpu/drm/i915/display/intel_atomic.c index a9a3f3715279ddc5e875045461c7f4d0bfcf36cc..7cf51dd8c0567065116f9187e3e5d80258d05686 100644 --- a/drivers/gpu/drm/i915/display/intel_atomic.c +++ b/drivers/gpu/drm/i915/display/intel_atomic.c @@ -265,7 +265,6 @@ intel_crtc_duplicate_state(struct drm_crtc *crtc) crtc_state->update_wm_post = false; crtc_state->fifo_changed = false; crtc_state->preload_luts = false; - crtc_state->inherited = false; crtc_state->wm.need_postvbl_update = false; crtc_state->do_async_flip = false; crtc_state->fb_bits = 0; @@ -311,262 +310,6 @@ intel_crtc_destroy_state(struct drm_crtc *crtc, kfree(crtc_state); } -static int intel_atomic_setup_scaler(struct intel_crtc_scaler_state *scaler_state, - int num_scalers_need, struct intel_crtc *intel_crtc, - const char *name, int idx, - struct intel_plane_state *plane_state, - int *scaler_id) -{ - struct drm_i915_private *dev_priv = to_i915(intel_crtc->base.dev); - int j; - u32 mode; - - if (*scaler_id < 0) { - /* find a free scaler */ - for (j = 0; j < intel_crtc->num_scalers; j++) { - if (scaler_state->scalers[j].in_use) - continue; - - *scaler_id = j; - scaler_state->scalers[*scaler_id].in_use = 1; - break; - } - } - - if (drm_WARN(&dev_priv->drm, *scaler_id < 0, - "Cannot find scaler for %s:%d\n", name, idx)) - return -EINVAL; - - /* set scaler mode */ - if (plane_state && plane_state->hw.fb && - plane_state->hw.fb->format->is_yuv && - plane_state->hw.fb->format->num_planes > 1) { - struct intel_plane *plane = to_intel_plane(plane_state->uapi.plane); - if (DISPLAY_VER(dev_priv) == 9) { - mode = SKL_PS_SCALER_MODE_NV12; - } else if (icl_is_hdr_plane(dev_priv, plane->id)) { - /* - * On gen11+'s HDR planes we only use the scaler for - * scaling. They have a dedicated chroma upsampler, so - * we don't need the scaler to upsample the UV plane. - */ - mode = PS_SCALER_MODE_NORMAL; - } else { - struct intel_plane *linked = - plane_state->planar_linked_plane; - - mode = PS_SCALER_MODE_PLANAR; - - if (linked) - mode |= PS_PLANE_Y_SEL(linked->id); - } - } else if (DISPLAY_VER(dev_priv) >= 10) { - mode = PS_SCALER_MODE_NORMAL; - } else if (num_scalers_need == 1 && intel_crtc->num_scalers > 1) { - /* - * when only 1 scaler is in use on a pipe with 2 scalers - * scaler 0 operates in high quality (HQ) mode. - * In this case use scaler 0 to take advantage of HQ mode - */ - scaler_state->scalers[*scaler_id].in_use = 0; - *scaler_id = 0; - scaler_state->scalers[0].in_use = 1; - mode = SKL_PS_SCALER_MODE_HQ; - } else { - mode = SKL_PS_SCALER_MODE_DYN; - } - - /* - * FIXME: we should also check the scaler factors for pfit, so - * this shouldn't be tied directly to planes. - */ - if (plane_state && plane_state->hw.fb) { - const struct drm_framebuffer *fb = plane_state->hw.fb; - const struct drm_rect *src = &plane_state->uapi.src; - const struct drm_rect *dst = &plane_state->uapi.dst; - int hscale, vscale, max_vscale, max_hscale; - - /* - * FIXME: When two scalers are needed, but only one of - * them needs to downscale, we should make sure that - * the one that needs downscaling support is assigned - * as the first scaler, so we don't reject downscaling - * unnecessarily. - */ - - if (DISPLAY_VER(dev_priv) >= 14) { - /* - * On versions 14 and up, only the first - * scaler supports a vertical scaling factor - * of more than 1.0, while a horizontal - * scaling factor of 3.0 is supported. - */ - max_hscale = 0x30000 - 1; - if (*scaler_id == 0) - max_vscale = 0x30000 - 1; - else - max_vscale = 0x10000; - - } else if (DISPLAY_VER(dev_priv) >= 10 || - !intel_format_info_is_yuv_semiplanar(fb->format, fb->modifier)) { - max_hscale = 0x30000 - 1; - max_vscale = 0x30000 - 1; - } else { - max_hscale = 0x20000 - 1; - max_vscale = 0x20000 - 1; - } - - /* - * FIXME: We should change the if-else block above to - * support HQ vs dynamic scaler properly. - */ - - /* Check if required scaling is within limits */ - hscale = drm_rect_calc_hscale(src, dst, 1, max_hscale); - vscale = drm_rect_calc_vscale(src, dst, 1, max_vscale); - - if (hscale < 0 || vscale < 0) { - drm_dbg_kms(&dev_priv->drm, - "Scaler %d doesn't support required plane scaling\n", - *scaler_id); - drm_rect_debug_print("src: ", src, true); - drm_rect_debug_print("dst: ", dst, false); - - return -EINVAL; - } - } - - drm_dbg_kms(&dev_priv->drm, "Attached scaler id %u.%u to %s:%d\n", - intel_crtc->pipe, *scaler_id, name, idx); - scaler_state->scalers[*scaler_id].mode = mode; - - return 0; -} - -/** - * intel_atomic_setup_scalers() - setup scalers for crtc per staged requests - * @dev_priv: i915 device - * @intel_crtc: intel crtc - * @crtc_state: incoming crtc_state to validate and setup scalers - * - * This function sets up scalers based on staged scaling requests for - * a @crtc and its planes. It is called from crtc level check path. If request - * is a supportable request, it attaches scalers to requested planes and crtc. - * - * This function takes into account the current scaler(s) in use by any planes - * not being part of this atomic state - * - * Returns: - * 0 - scalers were setup succesfully - * error code - otherwise - */ -int intel_atomic_setup_scalers(struct drm_i915_private *dev_priv, - struct intel_crtc *intel_crtc, - struct intel_crtc_state *crtc_state) -{ - struct drm_plane *plane = NULL; - struct intel_plane *intel_plane; - struct intel_plane_state *plane_state = NULL; - struct intel_crtc_scaler_state *scaler_state = - &crtc_state->scaler_state; - struct drm_atomic_state *drm_state = crtc_state->uapi.state; - struct intel_atomic_state *intel_state = to_intel_atomic_state(drm_state); - int num_scalers_need; - int i; - - num_scalers_need = hweight32(scaler_state->scaler_users); - - /* - * High level flow: - * - staged scaler requests are already in scaler_state->scaler_users - * - check whether staged scaling requests can be supported - * - add planes using scalers that aren't in current transaction - * - assign scalers to requested users - * - as part of plane commit, scalers will be committed - * (i.e., either attached or detached) to respective planes in hw - * - as part of crtc_commit, scaler will be either attached or detached - * to crtc in hw - */ - - /* fail if required scalers > available scalers */ - if (num_scalers_need > intel_crtc->num_scalers){ - drm_dbg_kms(&dev_priv->drm, - "Too many scaling requests %d > %d\n", - num_scalers_need, intel_crtc->num_scalers); - return -EINVAL; - } - - /* walkthrough scaler_users bits and start assigning scalers */ - for (i = 0; i < sizeof(scaler_state->scaler_users) * 8; i++) { - int *scaler_id; - const char *name; - int idx, ret; - - /* skip if scaler not required */ - if (!(scaler_state->scaler_users & (1 << i))) - continue; - - if (i == SKL_CRTC_INDEX) { - name = "CRTC"; - idx = intel_crtc->base.base.id; - - /* panel fitter case: assign as a crtc scaler */ - scaler_id = &scaler_state->scaler_id; - } else { - name = "PLANE"; - - /* plane scaler case: assign as a plane scaler */ - /* find the plane that set the bit as scaler_user */ - plane = drm_state->planes[i].ptr; - - /* - * to enable/disable hq mode, add planes that are using scaler - * into this transaction - */ - if (!plane) { - struct drm_plane_state *state; - - /* - * GLK+ scalers don't have a HQ mode so it - * isn't necessary to change between HQ and dyn mode - * on those platforms. - */ - if (DISPLAY_VER(dev_priv) >= 10) - continue; - - plane = drm_plane_from_index(&dev_priv->drm, i); - state = drm_atomic_get_plane_state(drm_state, plane); - if (IS_ERR(state)) { - drm_dbg_kms(&dev_priv->drm, - "Failed to add [PLANE:%d] to drm_state\n", - plane->base.id); - return PTR_ERR(state); - } - } - - intel_plane = to_intel_plane(plane); - idx = plane->base.id; - - /* plane on different crtc cannot be a scaler user of this crtc */ - if (drm_WARN_ON(&dev_priv->drm, - intel_plane->pipe != intel_crtc->pipe)) - continue; - - plane_state = intel_atomic_get_new_plane_state(intel_state, - intel_plane); - scaler_id = &plane_state->scaler_id; - } - - ret = intel_atomic_setup_scaler(scaler_state, num_scalers_need, - intel_crtc, name, idx, - plane_state, scaler_id); - if (ret < 0) - return ret; - } - - return 0; -} - struct drm_atomic_state * intel_atomic_state_alloc(struct drm_device *dev) { @@ -599,6 +342,8 @@ void intel_atomic_state_clear(struct drm_atomic_state *s) drm_atomic_state_default_clear(&state->base); intel_atomic_clear_global_state(state); + /* state->internal not reset on purpose */ + state->dpll_set = state->modeset = false; } diff --git a/drivers/gpu/drm/i915/display/intel_atomic.h b/drivers/gpu/drm/i915/display/intel_atomic.h index 1dc439983dd943092664f1d63f1ca336f9b5052f..e506f6a87344773484ffa9a47e9e35123278c9d5 100644 --- a/drivers/gpu/drm/i915/display/intel_atomic.h +++ b/drivers/gpu/drm/i915/display/intel_atomic.h @@ -52,8 +52,4 @@ struct intel_crtc_state * intel_atomic_get_crtc_state(struct drm_atomic_state *state, struct intel_crtc *crtc); -int intel_atomic_setup_scalers(struct drm_i915_private *dev_priv, - struct intel_crtc *intel_crtc, - struct intel_crtc_state *crtc_state); - #endif /* __INTEL_ATOMIC_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_atomic_plane.c b/drivers/gpu/drm/i915/display/intel_atomic_plane.c index f33164b10292c5c9e37a191870b8ac7828a6a2bc..7d9578ebae556665aa798b8d09d586574ce622db 100644 --- a/drivers/gpu/drm/i915/display/intel_atomic_plane.c +++ b/drivers/gpu/drm/i915/display/intel_atomic_plane.c @@ -36,6 +36,7 @@ #include #include "i915_config.h" +#include "i915_reg.h" #include "intel_atomic_plane.h" #include "intel_cdclk.h" #include "intel_display_rps.h" @@ -721,7 +722,7 @@ skl_next_plane_to_commit(struct intel_atomic_state *state, { struct intel_crtc_state *crtc_state = intel_atomic_get_new_crtc_state(state, crtc); - struct intel_plane_state *plane_state; + struct intel_plane_state __maybe_unused *plane_state; struct intel_plane *plane; int i; diff --git a/drivers/gpu/drm/i915/display/intel_audio.c b/drivers/gpu/drm/i915/display/intel_audio.c index 3d5a9bbc6fde370fcb40a73b72a55766d677716c..3d9c9b4f27f80268b16bb53b2f82764d9ca3c9e6 100644 --- a/drivers/gpu/drm/i915/display/intel_audio.c +++ b/drivers/gpu/drm/i915/display/intel_audio.c @@ -1039,6 +1039,7 @@ static void glk_force_audio_cdclk(struct drm_i915_private *i915, return; state->acquire_ctx = &ctx; + to_intel_atomic_state(state)->internal = true; retry: ret = glk_force_audio_cdclk_commit(to_intel_atomic_state(state), crtc, diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c index 75e69dffc5e90ace3397a3c4b0e78b1168e2c89a..34a397adbd6b05529d6182d27c782081a3b68e43 100644 --- a/drivers/gpu/drm/i915/display/intel_bios.c +++ b/drivers/gpu/drm/i915/display/intel_bios.c @@ -2141,58 +2141,58 @@ static u8 translate_iboost(u8 val) static const u8 cnp_ddc_pin_map[] = { [0] = 0, /* N/A */ - [DDC_BUS_DDI_B] = GMBUS_PIN_1_BXT, - [DDC_BUS_DDI_C] = GMBUS_PIN_2_BXT, - [DDC_BUS_DDI_D] = GMBUS_PIN_4_CNP, /* sic */ - [DDC_BUS_DDI_F] = GMBUS_PIN_3_BXT, /* sic */ + [GMBUS_PIN_1_BXT] = DDC_BUS_DDI_B, + [GMBUS_PIN_2_BXT] = DDC_BUS_DDI_C, + [GMBUS_PIN_4_CNP] = DDC_BUS_DDI_D, /* sic */ + [GMBUS_PIN_3_BXT] = DDC_BUS_DDI_F, /* sic */ }; static const u8 icp_ddc_pin_map[] = { - [ICL_DDC_BUS_DDI_A] = GMBUS_PIN_1_BXT, - [ICL_DDC_BUS_DDI_B] = GMBUS_PIN_2_BXT, - [TGL_DDC_BUS_DDI_C] = GMBUS_PIN_3_BXT, - [ICL_DDC_BUS_PORT_1] = GMBUS_PIN_9_TC1_ICP, - [ICL_DDC_BUS_PORT_2] = GMBUS_PIN_10_TC2_ICP, - [ICL_DDC_BUS_PORT_3] = GMBUS_PIN_11_TC3_ICP, - [ICL_DDC_BUS_PORT_4] = GMBUS_PIN_12_TC4_ICP, - [TGL_DDC_BUS_PORT_5] = GMBUS_PIN_13_TC5_TGP, - [TGL_DDC_BUS_PORT_6] = GMBUS_PIN_14_TC6_TGP, + [GMBUS_PIN_1_BXT] = ICL_DDC_BUS_DDI_A, + [GMBUS_PIN_2_BXT] = ICL_DDC_BUS_DDI_B, + [GMBUS_PIN_3_BXT] = TGL_DDC_BUS_DDI_C, + [GMBUS_PIN_9_TC1_ICP] = ICL_DDC_BUS_PORT_1, + [GMBUS_PIN_10_TC2_ICP] = ICL_DDC_BUS_PORT_2, + [GMBUS_PIN_11_TC3_ICP] = ICL_DDC_BUS_PORT_3, + [GMBUS_PIN_12_TC4_ICP] = ICL_DDC_BUS_PORT_4, + [GMBUS_PIN_13_TC5_TGP] = TGL_DDC_BUS_PORT_5, + [GMBUS_PIN_14_TC6_TGP] = TGL_DDC_BUS_PORT_6, }; static const u8 rkl_pch_tgp_ddc_pin_map[] = { - [ICL_DDC_BUS_DDI_A] = GMBUS_PIN_1_BXT, - [ICL_DDC_BUS_DDI_B] = GMBUS_PIN_2_BXT, - [RKL_DDC_BUS_DDI_D] = GMBUS_PIN_9_TC1_ICP, - [RKL_DDC_BUS_DDI_E] = GMBUS_PIN_10_TC2_ICP, + [GMBUS_PIN_1_BXT] = ICL_DDC_BUS_DDI_A, + [GMBUS_PIN_2_BXT] = ICL_DDC_BUS_DDI_B, + [GMBUS_PIN_9_TC1_ICP] = RKL_DDC_BUS_DDI_D, + [GMBUS_PIN_10_TC2_ICP] = RKL_DDC_BUS_DDI_E, }; static const u8 adls_ddc_pin_map[] = { - [ICL_DDC_BUS_DDI_A] = GMBUS_PIN_1_BXT, - [ADLS_DDC_BUS_PORT_TC1] = GMBUS_PIN_9_TC1_ICP, - [ADLS_DDC_BUS_PORT_TC2] = GMBUS_PIN_10_TC2_ICP, - [ADLS_DDC_BUS_PORT_TC3] = GMBUS_PIN_11_TC3_ICP, - [ADLS_DDC_BUS_PORT_TC4] = GMBUS_PIN_12_TC4_ICP, + [GMBUS_PIN_1_BXT] = ICL_DDC_BUS_DDI_A, + [GMBUS_PIN_9_TC1_ICP] = ADLS_DDC_BUS_PORT_TC1, + [GMBUS_PIN_10_TC2_ICP] = ADLS_DDC_BUS_PORT_TC2, + [GMBUS_PIN_11_TC3_ICP] = ADLS_DDC_BUS_PORT_TC3, + [GMBUS_PIN_12_TC4_ICP] = ADLS_DDC_BUS_PORT_TC4, }; static const u8 gen9bc_tgp_ddc_pin_map[] = { - [DDC_BUS_DDI_B] = GMBUS_PIN_2_BXT, - [DDC_BUS_DDI_C] = GMBUS_PIN_9_TC1_ICP, - [DDC_BUS_DDI_D] = GMBUS_PIN_10_TC2_ICP, + [GMBUS_PIN_2_BXT] = DDC_BUS_DDI_B, + [GMBUS_PIN_9_TC1_ICP] = DDC_BUS_DDI_C, + [GMBUS_PIN_10_TC2_ICP] = DDC_BUS_DDI_D, }; static const u8 adlp_ddc_pin_map[] = { - [ICL_DDC_BUS_DDI_A] = GMBUS_PIN_1_BXT, - [ICL_DDC_BUS_DDI_B] = GMBUS_PIN_2_BXT, - [ADLP_DDC_BUS_PORT_TC1] = GMBUS_PIN_9_TC1_ICP, - [ADLP_DDC_BUS_PORT_TC2] = GMBUS_PIN_10_TC2_ICP, - [ADLP_DDC_BUS_PORT_TC3] = GMBUS_PIN_11_TC3_ICP, - [ADLP_DDC_BUS_PORT_TC4] = GMBUS_PIN_12_TC4_ICP, + [GMBUS_PIN_1_BXT] = ICL_DDC_BUS_DDI_A, + [GMBUS_PIN_2_BXT] = ICL_DDC_BUS_DDI_B, + [GMBUS_PIN_9_TC1_ICP] = ADLP_DDC_BUS_PORT_TC1, + [GMBUS_PIN_10_TC2_ICP] = ADLP_DDC_BUS_PORT_TC2, + [GMBUS_PIN_11_TC3_ICP] = ADLP_DDC_BUS_PORT_TC3, + [GMBUS_PIN_12_TC4_ICP] = ADLP_DDC_BUS_PORT_TC4, }; static u8 map_ddc_pin(struct drm_i915_private *i915, u8 vbt_pin) { const u8 *ddc_pin_map; - int n_entries; + int i, n_entries; if (HAS_PCH_MTP(i915) || IS_ALDERLAKE_P(i915)) { ddc_pin_map = adlp_ddc_pin_map; @@ -2219,8 +2219,10 @@ static u8 map_ddc_pin(struct drm_i915_private *i915, u8 vbt_pin) return vbt_pin; } - if (vbt_pin < n_entries && ddc_pin_map[vbt_pin] != 0) - return ddc_pin_map[vbt_pin]; + for (i = 0; i < n_entries; i++) { + if (ddc_pin_map[i] == vbt_pin) + return i; + } drm_dbg_kms(&i915->drm, "Ignoring alternate pin: VBT claims DDC pin %d, which is not valid for this platform\n", @@ -2675,8 +2677,9 @@ static void print_ddi_port(const struct intel_bios_encoder_data *devdata, supports_tbt = intel_bios_encoder_supports_tbt(devdata); drm_dbg_kms(&i915->drm, - "Port %c VBT info: CRT:%d DVI:%d HDMI:%d DP:%d eDP:%d DSI:%d LSPCON:%d USB-Type-C:%d TBT:%d DSC:%d\n", + "Port %c VBT info: CRT:%d DVI:%d HDMI:%d DP:%d eDP:%d DSI:%d DP++:%d LSPCON:%d USB-Type-C:%d TBT:%d DSC:%d\n", port_name(port), is_crt, is_dvi, is_hdmi, is_dp, is_edp, is_dsi, + intel_bios_encoder_supports_dp_dual_mode(devdata), intel_bios_encoder_is_lspcon(devdata), supports_typec_usb, supports_tbt, devdata->dsc != NULL); @@ -3030,6 +3033,13 @@ bool intel_bios_is_valid_vbt(const void *buf, size_t size) return vbt; } +static u32 intel_spi_read(struct intel_uncore *uncore, u32 offset) +{ + intel_uncore_write(uncore, PRIMARY_SPI_ADDRESS, offset); + + return intel_uncore_read(uncore, PRIMARY_SPI_TRIGGER); +} + static struct vbt_header *spi_oprom_get_vbt(struct drm_i915_private *i915) { u32 count, data, found, store = 0; @@ -3046,9 +3056,7 @@ static struct vbt_header *spi_oprom_get_vbt(struct drm_i915_private *i915) oprom_offset &= OROM_OFFSET_MASK; for (count = 0; count < oprom_size; count += 4) { - intel_uncore_write(&i915->uncore, PRIMARY_SPI_ADDRESS, oprom_offset + count); - data = intel_uncore_read(&i915->uncore, PRIMARY_SPI_TRIGGER); - + data = intel_spi_read(&i915->uncore, oprom_offset + count); if (data == *((const u32 *)"$VBT")) { found = oprom_offset + count; break; @@ -3059,20 +3067,16 @@ static struct vbt_header *spi_oprom_get_vbt(struct drm_i915_private *i915) goto err_not_found; /* Get VBT size and allocate space for the VBT */ - intel_uncore_write(&i915->uncore, PRIMARY_SPI_ADDRESS, found + - offsetof(struct vbt_header, vbt_size)); - vbt_size = intel_uncore_read(&i915->uncore, PRIMARY_SPI_TRIGGER); + vbt_size = intel_spi_read(&i915->uncore, + found + offsetof(struct vbt_header, vbt_size)); vbt_size &= 0xffff; vbt = kzalloc(round_up(vbt_size, 4), GFP_KERNEL); if (!vbt) goto err_not_found; - for (count = 0; count < vbt_size; count += 4) { - intel_uncore_write(&i915->uncore, PRIMARY_SPI_ADDRESS, found + count); - data = intel_uncore_read(&i915->uncore, PRIMARY_SPI_TRIGGER); - *(vbt + store++) = data; - } + for (count = 0; count < vbt_size; count += 4) + *(vbt + store++) = intel_spi_read(&i915->uncore, found + count); if (!intel_bios_is_valid_vbt(vbt, vbt_size)) goto err_free_vbt; @@ -3424,7 +3428,7 @@ bool intel_bios_is_port_present(struct drm_i915_private *i915, enum port port) return false; } -static bool intel_bios_encoder_supports_dp_dual_mode(const struct intel_bios_encoder_data *devdata) +bool intel_bios_encoder_supports_dp_dual_mode(const struct intel_bios_encoder_data *devdata) { const struct child_device_config *child = &devdata->child; @@ -3443,15 +3447,6 @@ static bool intel_bios_encoder_supports_dp_dual_mode(const struct intel_bios_enc return false; } -bool intel_bios_is_port_dp_dual_mode(struct drm_i915_private *i915, - enum port port) -{ - const struct intel_bios_encoder_data *devdata = - intel_bios_encoder_data_lookup(i915, port); - - return devdata && intel_bios_encoder_supports_dp_dual_mode(devdata); -} - /** * intel_bios_is_dsi_present - is DSI present in VBT * @i915: i915 device instance @@ -3578,84 +3573,82 @@ bool intel_bios_get_dsc_params(struct intel_encoder *encoder, return false; } +static const u8 adlp_aux_ch_map[] = { + [AUX_CH_A] = DP_AUX_A, + [AUX_CH_B] = DP_AUX_B, + [AUX_CH_C] = DP_AUX_C, + [AUX_CH_D_XELPD] = DP_AUX_D, + [AUX_CH_E_XELPD] = DP_AUX_E, + [AUX_CH_USBC1] = DP_AUX_F, + [AUX_CH_USBC2] = DP_AUX_G, + [AUX_CH_USBC3] = DP_AUX_H, + [AUX_CH_USBC4] = DP_AUX_I, +}; + +/* + * ADL-S VBT uses PHY based mapping. Combo PHYs A,B,C,D,E + * map to DDI A,TC1,TC2,TC3,TC4 respectively. + */ +static const u8 adls_aux_ch_map[] = { + [AUX_CH_A] = DP_AUX_A, + [AUX_CH_USBC1] = DP_AUX_B, + [AUX_CH_USBC2] = DP_AUX_C, + [AUX_CH_USBC3] = DP_AUX_D, + [AUX_CH_USBC4] = DP_AUX_E, +}; + +/* + * RKL/DG1 VBT uses PHY based mapping. Combo PHYs A,B,C,D + * map to DDI A,B,TC1,TC2 respectively. + */ +static const u8 rkl_aux_ch_map[] = { + [AUX_CH_A] = DP_AUX_A, + [AUX_CH_B] = DP_AUX_B, + [AUX_CH_USBC1] = DP_AUX_C, + [AUX_CH_USBC2] = DP_AUX_D, +}; + +static const u8 direct_aux_ch_map[] = { + [AUX_CH_A] = DP_AUX_A, + [AUX_CH_B] = DP_AUX_B, + [AUX_CH_C] = DP_AUX_C, + [AUX_CH_D] = DP_AUX_D, /* aka AUX_CH_USBC1 */ + [AUX_CH_E] = DP_AUX_E, /* aka AUX_CH_USBC2 */ + [AUX_CH_F] = DP_AUX_F, /* aka AUX_CH_USBC3 */ + [AUX_CH_G] = DP_AUX_G, /* aka AUX_CH_USBC4 */ + [AUX_CH_H] = DP_AUX_H, /* aka AUX_CH_USBC5 */ + [AUX_CH_I] = DP_AUX_I, /* aka AUX_CH_USBC6 */ +}; + static enum aux_ch map_aux_ch(struct drm_i915_private *i915, u8 aux_channel) { - enum aux_ch aux_ch; + const u8 *aux_ch_map; + int i, n_entries; - /* - * RKL/DG1 VBT uses PHY based mapping. Combo PHYs A,B,C,D - * map to DDI A,B,TC1,TC2 respectively. - * - * ADL-S VBT uses PHY based mapping. Combo PHYs A,B,C,D,E - * map to DDI A,TC1,TC2,TC3,TC4 respectively. - */ - switch (aux_channel) { - case DP_AUX_A: - aux_ch = AUX_CH_A; - break; - case DP_AUX_B: - if (IS_ALDERLAKE_S(i915)) - aux_ch = AUX_CH_USBC1; - else - aux_ch = AUX_CH_B; - break; - case DP_AUX_C: - if (IS_ALDERLAKE_S(i915)) - aux_ch = AUX_CH_USBC2; - else if (IS_DG1(i915) || IS_ROCKETLAKE(i915)) - aux_ch = AUX_CH_USBC1; - else - aux_ch = AUX_CH_C; - break; - case DP_AUX_D: - if (DISPLAY_VER(i915) >= 13) - aux_ch = AUX_CH_D_XELPD; - else if (IS_ALDERLAKE_S(i915)) - aux_ch = AUX_CH_USBC3; - else if (IS_DG1(i915) || IS_ROCKETLAKE(i915)) - aux_ch = AUX_CH_USBC2; - else - aux_ch = AUX_CH_D; - break; - case DP_AUX_E: - if (DISPLAY_VER(i915) >= 13) - aux_ch = AUX_CH_E_XELPD; - else if (IS_ALDERLAKE_S(i915)) - aux_ch = AUX_CH_USBC4; - else - aux_ch = AUX_CH_E; - break; - case DP_AUX_F: - if (DISPLAY_VER(i915) >= 13) - aux_ch = AUX_CH_USBC1; - else - aux_ch = AUX_CH_F; - break; - case DP_AUX_G: - if (DISPLAY_VER(i915) >= 13) - aux_ch = AUX_CH_USBC2; - else - aux_ch = AUX_CH_G; - break; - case DP_AUX_H: - if (DISPLAY_VER(i915) >= 13) - aux_ch = AUX_CH_USBC3; - else - aux_ch = AUX_CH_H; - break; - case DP_AUX_I: - if (DISPLAY_VER(i915) >= 13) - aux_ch = AUX_CH_USBC4; - else - aux_ch = AUX_CH_I; - break; - default: - MISSING_CASE(aux_channel); - aux_ch = AUX_CH_A; - break; + if (DISPLAY_VER(i915) >= 13) { + aux_ch_map = adlp_aux_ch_map; + n_entries = ARRAY_SIZE(adlp_aux_ch_map); + } else if (IS_ALDERLAKE_S(i915)) { + aux_ch_map = adls_aux_ch_map; + n_entries = ARRAY_SIZE(adls_aux_ch_map); + } else if (IS_DG1(i915) || IS_ROCKETLAKE(i915)) { + aux_ch_map = rkl_aux_ch_map; + n_entries = ARRAY_SIZE(rkl_aux_ch_map); + } else { + aux_ch_map = direct_aux_ch_map; + n_entries = ARRAY_SIZE(direct_aux_ch_map); + } + + for (i = 0; i < n_entries; i++) { + if (aux_ch_map[i] == aux_channel) + return i; } - return aux_ch; + drm_dbg_kms(&i915->drm, + "Ignoring alternate AUX CH: VBT claims AUX 0x%x, which is not valid for this platform\n", + aux_channel); + + return AUX_CH_NONE; } enum aux_ch intel_bios_dp_aux_ch(const struct intel_bios_encoder_data *devdata) diff --git a/drivers/gpu/drm/i915/display/intel_bios.h b/drivers/gpu/drm/i915/display/intel_bios.h index 8a0730c9b48cf430743e81db6febabe2d33901dd..45fae97d97192aa4710f251a8a1aa339602391bb 100644 --- a/drivers/gpu/drm/i915/display/intel_bios.h +++ b/drivers/gpu/drm/i915/display/intel_bios.h @@ -247,7 +247,6 @@ bool intel_bios_is_tv_present(struct drm_i915_private *dev_priv); bool intel_bios_is_lvds_present(struct drm_i915_private *dev_priv, u8 *i2c_pin); bool intel_bios_is_port_present(struct drm_i915_private *dev_priv, enum port port); bool intel_bios_is_port_edp(struct drm_i915_private *dev_priv, enum port port); -bool intel_bios_is_port_dp_dual_mode(struct drm_i915_private *dev_priv, enum port port); bool intel_bios_is_dsi_present(struct drm_i915_private *dev_priv, enum port *port); bool intel_bios_get_dsc_params(struct intel_encoder *encoder, struct intel_crtc_state *crtc_state, @@ -264,6 +263,7 @@ bool intel_bios_encoder_supports_dp(const struct intel_bios_encoder_data *devdat bool intel_bios_encoder_supports_edp(const struct intel_bios_encoder_data *devdata); bool intel_bios_encoder_supports_typec_usb(const struct intel_bios_encoder_data *devdata); bool intel_bios_encoder_supports_tbt(const struct intel_bios_encoder_data *devdata); +bool intel_bios_encoder_supports_dp_dual_mode(const struct intel_bios_encoder_data *devdata); bool intel_bios_encoder_is_lspcon(const struct intel_bios_encoder_data *devdata); bool intel_bios_encoder_lane_reversal(const struct intel_bios_encoder_data *devdata); bool intel_bios_encoder_hpd_invert(const struct intel_bios_encoder_data *devdata); diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c index 202321ffbe2ac21214c3d4d24b88663fd8f73184..bef96db62c8075cd033dcbb6692e572cdd442d96 100644 --- a/drivers/gpu/drm/i915/display/intel_bw.c +++ b/drivers/gpu/drm/i915/display/intel_bw.c @@ -150,6 +150,9 @@ int icl_pcode_restrict_qgv_points(struct drm_i915_private *dev_priv, { int ret; + if (DISPLAY_VER(dev_priv) >= 14) + return 0; + /* bspec says to keep retrying for at least 1 ms */ ret = skl_pcode_request(&dev_priv->uncore, ICL_PCODE_SAGV_DE_MEM_SS_CONFIG, points_mask, @@ -179,7 +182,7 @@ static int mtl_read_qgv_point_info(struct drm_i915_private *dev_priv, val2 = intel_uncore_read(&dev_priv->uncore, MTL_MEM_SS_INFO_QGV_POINT_HIGH(point)); dclk = REG_FIELD_GET(MTL_DCLK_MASK, val); - sp->dclk = DIV_ROUND_UP((16667 * dclk), 1000); + sp->dclk = DIV_ROUND_CLOSEST(16667 * dclk, 1000); sp->t_rp = REG_FIELD_GET(MTL_TRP_MASK, val); sp->t_rcd = REG_FIELD_GET(MTL_TRCD_MASK, val); @@ -376,7 +379,7 @@ static const struct intel_sa_info mtl_sa_info = { .deburst = 32, .deprogbwlimit = 38, /* GB/s */ .displayrtids = 256, - .derating = 20, + .derating = 10, }; static int icl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel_sa_info *sa) @@ -531,10 +534,14 @@ static int tgl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel bi->deratedbw[j] = min(maxdebw, bw * (100 - sa->derating) / 100); + bi->peakbw[j] = DIV_ROUND_CLOSEST(sp->dclk * + num_channels * + qi.channel_width, 8); drm_dbg_kms(&dev_priv->drm, - "BW%d / QGV %d: num_planes=%d deratedbw=%u\n", - i, j, bi->num_planes, bi->deratedbw[j]); + "BW%d / QGV %d: num_planes=%d deratedbw=%u peakbw: %u\n", + i, j, bi->num_planes, bi->deratedbw[j], + bi->peakbw[j]); } for (j = 0; j < qi.num_psf_points; j++) { @@ -586,8 +593,8 @@ static void dg2_get_bw_info(struct drm_i915_private *i915) i915->display.sagv.status = I915_SAGV_NOT_CONTROLLED; } -static unsigned int icl_max_bw(struct drm_i915_private *dev_priv, - int num_planes, int qgv_point) +static unsigned int icl_max_bw_index(struct drm_i915_private *dev_priv, + int num_planes, int qgv_point) { int i; @@ -608,14 +615,14 @@ static unsigned int icl_max_bw(struct drm_i915_private *dev_priv, return UINT_MAX; if (num_planes >= bi->num_planes) - return bi->deratedbw[qgv_point]; + return i; } - return 0; + return UINT_MAX; } -static unsigned int tgl_max_bw(struct drm_i915_private *dev_priv, - int num_planes, int qgv_point) +static unsigned int tgl_max_bw_index(struct drm_i915_private *dev_priv, + int num_planes, int qgv_point) { int i; @@ -636,10 +643,10 @@ static unsigned int tgl_max_bw(struct drm_i915_private *dev_priv, return UINT_MAX; if (num_planes <= bi->num_planes) - return bi->deratedbw[qgv_point]; + return i; } - return dev_priv->display.bw.max[0].deratedbw[qgv_point]; + return 0; } static unsigned int adl_psf_bw(struct drm_i915_private *dev_priv, @@ -796,6 +803,210 @@ intel_atomic_get_bw_state(struct intel_atomic_state *state) return to_intel_bw_state(bw_state); } +static int mtl_find_qgv_points(struct drm_i915_private *i915, + unsigned int data_rate, + unsigned int num_active_planes, + struct intel_bw_state *new_bw_state) +{ + unsigned int best_rate = UINT_MAX; + unsigned int num_qgv_points = i915->display.bw.max[0].num_qgv_points; + unsigned int qgv_peak_bw = 0; + int i; + int ret; + + ret = intel_atomic_lock_global_state(&new_bw_state->base); + if (ret) + return ret; + + /* + * If SAGV cannot be enabled, disable the pcode SAGV by passing all 1's + * for qgv peak bw in PM Demand request. So assign UINT_MAX if SAGV is + * not enabled. PM Demand code will clamp the value for the register + */ + if (!intel_can_enable_sagv(i915, new_bw_state)) { + new_bw_state->qgv_point_peakbw = U16_MAX; + drm_dbg_kms(&i915->drm, "No SAGV, use UINT_MAX as peak bw."); + return 0; + } + + /* + * Find the best QGV point by comparing the data_rate with max data rate + * offered per plane group + */ + for (i = 0; i < num_qgv_points; i++) { + unsigned int bw_index = + tgl_max_bw_index(i915, num_active_planes, i); + unsigned int max_data_rate; + + if (bw_index >= ARRAY_SIZE(i915->display.bw.max)) + continue; + + max_data_rate = i915->display.bw.max[bw_index].deratedbw[i]; + + if (max_data_rate < data_rate) + continue; + + if (max_data_rate - data_rate < best_rate) { + best_rate = max_data_rate - data_rate; + qgv_peak_bw = i915->display.bw.max[bw_index].peakbw[i]; + } + + drm_dbg_kms(&i915->drm, "QGV point %d: max bw %d required %d qgv_peak_bw: %d\n", + i, max_data_rate, data_rate, qgv_peak_bw); + } + + drm_dbg_kms(&i915->drm, "Matching peaks QGV bw: %d for required data rate: %d\n", + qgv_peak_bw, data_rate); + + /* + * The display configuration cannot be supported if no QGV point + * satisfying the required data rate is found + */ + if (qgv_peak_bw == 0) { + drm_dbg_kms(&i915->drm, "No QGV points for bw %d for display configuration(%d active planes).\n", + data_rate, num_active_planes); + return -EINVAL; + } + + /* MTL PM DEMAND expects QGV BW parameter in multiples of 100 mbps */ + new_bw_state->qgv_point_peakbw = DIV_ROUND_CLOSEST(qgv_peak_bw, 100); + + return 0; +} + +static int icl_find_qgv_points(struct drm_i915_private *i915, + unsigned int data_rate, + unsigned int num_active_planes, + const struct intel_bw_state *old_bw_state, + struct intel_bw_state *new_bw_state) +{ + unsigned int max_bw_point = 0; + unsigned int max_bw = 0; + unsigned int num_psf_gv_points = i915->display.bw.max[0].num_psf_gv_points; + unsigned int num_qgv_points = i915->display.bw.max[0].num_qgv_points; + u16 psf_points = 0; + u16 qgv_points = 0; + int i; + int ret; + + ret = intel_atomic_lock_global_state(&new_bw_state->base); + if (ret) + return ret; + + for (i = 0; i < num_qgv_points; i++) { + unsigned int idx; + unsigned int max_data_rate; + + if (DISPLAY_VER(i915) > 11) + idx = tgl_max_bw_index(i915, num_active_planes, i); + else + idx = icl_max_bw_index(i915, num_active_planes, i); + + if (idx >= ARRAY_SIZE(i915->display.bw.max)) + continue; + + max_data_rate = i915->display.bw.max[idx].deratedbw[i]; + + /* + * We need to know which qgv point gives us + * maximum bandwidth in order to disable SAGV + * if we find that we exceed SAGV block time + * with watermarks. By that moment we already + * have those, as it is calculated earlier in + * intel_atomic_check, + */ + if (max_data_rate > max_bw) { + max_bw_point = i; + max_bw = max_data_rate; + } + if (max_data_rate >= data_rate) + qgv_points |= BIT(i); + + drm_dbg_kms(&i915->drm, "QGV point %d: max bw %d required %d\n", + i, max_data_rate, data_rate); + } + + for (i = 0; i < num_psf_gv_points; i++) { + unsigned int max_data_rate = adl_psf_bw(i915, i); + + if (max_data_rate >= data_rate) + psf_points |= BIT(i); + + drm_dbg_kms(&i915->drm, "PSF GV point %d: max bw %d" + " required %d\n", + i, max_data_rate, data_rate); + } + + /* + * BSpec states that we always should have at least one allowed point + * left, so if we couldn't - simply reject the configuration for obvious + * reasons. + */ + if (qgv_points == 0) { + drm_dbg_kms(&i915->drm, "No QGV points provide sufficient memory" + " bandwidth %d for display configuration(%d active planes).\n", + data_rate, num_active_planes); + return -EINVAL; + } + + if (num_psf_gv_points > 0 && psf_points == 0) { + drm_dbg_kms(&i915->drm, "No PSF GV points provide sufficient memory" + " bandwidth %d for display configuration(%d active planes).\n", + data_rate, num_active_planes); + return -EINVAL; + } + + /* + * Leave only single point with highest bandwidth, if + * we can't enable SAGV due to the increased memory latency it may + * cause. + */ + if (!intel_can_enable_sagv(i915, new_bw_state)) { + qgv_points = BIT(max_bw_point); + drm_dbg_kms(&i915->drm, "No SAGV, using single QGV point %d\n", + max_bw_point); + } + + /* + * We store the ones which need to be masked as that is what PCode + * actually accepts as a parameter. + */ + new_bw_state->qgv_points_mask = + ~(ICL_PCODE_REQ_QGV_PT(qgv_points) | + ADLS_PCODE_REQ_PSF_PT(psf_points)) & + icl_qgv_points_mask(i915); + + /* + * If the actual mask had changed we need to make sure that + * the commits are serialized(in case this is a nomodeset, nonblocking) + */ + if (new_bw_state->qgv_points_mask != old_bw_state->qgv_points_mask) { + ret = intel_atomic_serialize_global_state(&new_bw_state->base); + if (ret) + return ret; + } + + return 0; +} + +static int intel_bw_check_qgv_points(struct drm_i915_private *i915, + const struct intel_bw_state *old_bw_state, + struct intel_bw_state *new_bw_state) +{ + unsigned int data_rate = intel_bw_data_rate(i915, new_bw_state); + unsigned int num_active_planes = + intel_bw_num_active_planes(i915, new_bw_state); + + data_rate = DIV_ROUND_UP(data_rate, 1000); + + if (DISPLAY_VER(i915) >= 14) + return mtl_find_qgv_points(i915, data_rate, num_active_planes, + new_bw_state); + else + return icl_find_qgv_points(i915, data_rate, num_active_planes, + old_bw_state, new_bw_state); +} + static bool intel_bw_state_changed(struct drm_i915_private *i915, const struct intel_bw_state *old_bw_state, const struct intel_bw_state *new_bw_state) @@ -1042,20 +1253,14 @@ static int intel_bw_check_data_rate(struct intel_atomic_state *state, bool *chan int intel_bw_atomic_check(struct intel_atomic_state *state) { - struct drm_i915_private *dev_priv = to_i915(state->base.dev); - const struct intel_bw_state *old_bw_state; - struct intel_bw_state *new_bw_state; - unsigned int data_rate; - unsigned int num_active_planes; - int i, ret; - u16 qgv_points = 0, psf_points = 0; - unsigned int max_bw_point = 0, max_bw = 0; - unsigned int num_qgv_points = dev_priv->display.bw.max[0].num_qgv_points; - unsigned int num_psf_gv_points = dev_priv->display.bw.max[0].num_psf_gv_points; bool changed = false; + struct drm_i915_private *i915 = to_i915(state->base.dev); + struct intel_bw_state *new_bw_state; + const struct intel_bw_state *old_bw_state; + int ret; /* FIXME earlier gens need some checks too */ - if (DISPLAY_VER(dev_priv) < 11) + if (DISPLAY_VER(i915) < 11) return 0; ret = intel_bw_check_data_rate(state, &changed); @@ -1066,8 +1271,8 @@ int intel_bw_atomic_check(struct intel_atomic_state *state) new_bw_state = intel_atomic_get_new_bw_state(state); if (new_bw_state && - intel_can_enable_sagv(dev_priv, old_bw_state) != - intel_can_enable_sagv(dev_priv, new_bw_state)) + intel_can_enable_sagv(i915, old_bw_state) != + intel_can_enable_sagv(i915, new_bw_state)) changed = true; /* @@ -1077,101 +1282,10 @@ int intel_bw_atomic_check(struct intel_atomic_state *state) if (!changed) return 0; - ret = intel_atomic_lock_global_state(&new_bw_state->base); + ret = intel_bw_check_qgv_points(i915, old_bw_state, new_bw_state); if (ret) return ret; - data_rate = intel_bw_data_rate(dev_priv, new_bw_state); - data_rate = DIV_ROUND_UP(data_rate, 1000); - - num_active_planes = intel_bw_num_active_planes(dev_priv, new_bw_state); - - for (i = 0; i < num_qgv_points; i++) { - unsigned int max_data_rate; - - if (DISPLAY_VER(dev_priv) > 11) - max_data_rate = tgl_max_bw(dev_priv, num_active_planes, i); - else - max_data_rate = icl_max_bw(dev_priv, num_active_planes, i); - /* - * We need to know which qgv point gives us - * maximum bandwidth in order to disable SAGV - * if we find that we exceed SAGV block time - * with watermarks. By that moment we already - * have those, as it is calculated earlier in - * intel_atomic_check, - */ - if (max_data_rate > max_bw) { - max_bw_point = i; - max_bw = max_data_rate; - } - if (max_data_rate >= data_rate) - qgv_points |= BIT(i); - - drm_dbg_kms(&dev_priv->drm, "QGV point %d: max bw %d required %d\n", - i, max_data_rate, data_rate); - } - - for (i = 0; i < num_psf_gv_points; i++) { - unsigned int max_data_rate = adl_psf_bw(dev_priv, i); - - if (max_data_rate >= data_rate) - psf_points |= BIT(i); - - drm_dbg_kms(&dev_priv->drm, "PSF GV point %d: max bw %d" - " required %d\n", - i, max_data_rate, data_rate); - } - - /* - * BSpec states that we always should have at least one allowed point - * left, so if we couldn't - simply reject the configuration for obvious - * reasons. - */ - if (qgv_points == 0) { - drm_dbg_kms(&dev_priv->drm, "No QGV points provide sufficient memory" - " bandwidth %d for display configuration(%d active planes).\n", - data_rate, num_active_planes); - return -EINVAL; - } - - if (num_psf_gv_points > 0 && psf_points == 0) { - drm_dbg_kms(&dev_priv->drm, "No PSF GV points provide sufficient memory" - " bandwidth %d for display configuration(%d active planes).\n", - data_rate, num_active_planes); - return -EINVAL; - } - - /* - * Leave only single point with highest bandwidth, if - * we can't enable SAGV due to the increased memory latency it may - * cause. - */ - if (!intel_can_enable_sagv(dev_priv, new_bw_state)) { - qgv_points = BIT(max_bw_point); - drm_dbg_kms(&dev_priv->drm, "No SAGV, using single QGV point %d\n", - max_bw_point); - } - - /* - * We store the ones which need to be masked as that is what PCode - * actually accepts as a parameter. - */ - new_bw_state->qgv_points_mask = - ~(ICL_PCODE_REQ_QGV_PT(qgv_points) | - ADLS_PCODE_REQ_PSF_PT(psf_points)) & - icl_qgv_points_mask(dev_priv); - - /* - * If the actual mask had changed we need to make sure that - * the commits are serialized(in case this is a nomodeset, nonblocking) - */ - if (new_bw_state->qgv_points_mask != old_bw_state->qgv_points_mask) { - ret = intel_atomic_serialize_global_state(&new_bw_state->base); - if (ret) - return ret; - } - return 0; } diff --git a/drivers/gpu/drm/i915/display/intel_bw.h b/drivers/gpu/drm/i915/display/intel_bw.h index f20292143745502c48e05078052644531dc2c864..59cb4fc5db76c37a633024b6267c8d0accabadf4 100644 --- a/drivers/gpu/drm/i915/display/intel_bw.h +++ b/drivers/gpu/drm/i915/display/intel_bw.h @@ -34,6 +34,12 @@ struct intel_bw_state { /* bitmask of active pipes */ u8 active_pipes; + /* + * From MTL onwards, to lock a QGV point, punit expects the peak BW of + * the selected QGV point as the parameter in multiples of 100MB/s + */ + u16 qgv_point_peakbw; + /* * Current QGV points mask, which restricts * some particular SAGV states, not to confuse diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c index 084a483f97766165472c2f6b4fb71f57e2b07c00..4207863b7b2ae5d4ad29d83deac0d875417984d1 100644 --- a/drivers/gpu/drm/i915/display/intel_cdclk.c +++ b/drivers/gpu/drm/i915/display/intel_cdclk.c @@ -1453,6 +1453,18 @@ static u8 tgl_calc_voltage_level(int cdclk) return 0; } +static u8 rplu_calc_voltage_level(int cdclk) +{ + if (cdclk > 556800) + return 3; + else if (cdclk > 480000) + return 2; + else if (cdclk > 312000) + return 1; + else + return 0; +} + static void icl_readout_refclk(struct drm_i915_private *dev_priv, struct intel_cdclk_config *cdclk_config) { @@ -1896,7 +1908,7 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv, * mailbox communication, skip * this step. */ - if (DISPLAY_VER(dev_priv) >= 14) + if (DISPLAY_VER(dev_priv) >= 14 || IS_DG2(dev_priv)) /* NOOP */; else if (DISPLAY_VER(dev_priv) >= 11) ret = skl_pcode_request(&dev_priv->uncore, SKL_PCODE_CDCLK_CONTROL, @@ -1932,10 +1944,10 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv, * NOOP - No Pcode communication needed for * Display versions 14 and beyond */; - else if (DISPLAY_VER(dev_priv) >= 11) + else if (DISPLAY_VER(dev_priv) >= 11 && !IS_DG2(dev_priv)) ret = snb_pcode_write(&dev_priv->uncore, SKL_PCODE_CDCLK_CONTROL, cdclk_config->voltage_level); - else + if (DISPLAY_VER(dev_priv) < 11) { /* * The timeout isn't specified, the 2ms used here is based on * experiment. @@ -1946,7 +1958,7 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv, HSW_PCODE_DE_WRITE_FREQ_REQ, cdclk_config->voltage_level, 150, 2); - + } if (ret) { drm_err(&dev_priv->drm, "PCode CDCLK freq set failed, (err %d, freq %d)\n", @@ -2242,6 +2254,38 @@ void intel_cdclk_dump_config(struct drm_i915_private *i915, cdclk_config->voltage_level); } +static void intel_pcode_notify(struct drm_i915_private *i915, + u8 voltage_level, + u8 active_pipe_count, + u16 cdclk, + bool cdclk_update_valid, + bool pipe_count_update_valid) +{ + int ret; + u32 update_mask = 0; + + if (!IS_DG2(i915)) + return; + + update_mask = DISPLAY_TO_PCODE_UPDATE_MASK(cdclk, active_pipe_count, voltage_level); + + if (cdclk_update_valid) + update_mask |= DISPLAY_TO_PCODE_CDCLK_VALID; + + if (pipe_count_update_valid) + update_mask |= DISPLAY_TO_PCODE_PIPE_COUNT_VALID; + + ret = skl_pcode_request(&i915->uncore, SKL_PCODE_CDCLK_CONTROL, + SKL_CDCLK_PREPARE_FOR_CHANGE | + update_mask, + SKL_CDCLK_READY_FOR_CHANGE, + SKL_CDCLK_READY_FOR_CHANGE, 3); + if (ret) + drm_err(&i915->drm, + "Failed to inform PCU about display config (err %d)\n", + ret); +} + /** * intel_set_cdclk - Push the CDCLK configuration to the hardware * @dev_priv: i915 device @@ -2311,6 +2355,88 @@ static void intel_set_cdclk(struct drm_i915_private *dev_priv, } } +static void intel_cdclk_pcode_pre_notify(struct intel_atomic_state *state) +{ + struct drm_i915_private *i915 = to_i915(state->base.dev); + const struct intel_cdclk_state *old_cdclk_state = + intel_atomic_get_old_cdclk_state(state); + const struct intel_cdclk_state *new_cdclk_state = + intel_atomic_get_new_cdclk_state(state); + unsigned int cdclk = 0; u8 voltage_level, num_active_pipes = 0; + bool change_cdclk, update_pipe_count; + + if (!intel_cdclk_changed(&old_cdclk_state->actual, + &new_cdclk_state->actual) && + new_cdclk_state->active_pipes == + old_cdclk_state->active_pipes) + return; + + /* According to "Sequence Before Frequency Change", voltage level set to 0x3 */ + voltage_level = DISPLAY_TO_PCODE_VOLTAGE_MAX; + + change_cdclk = new_cdclk_state->actual.cdclk != old_cdclk_state->actual.cdclk; + update_pipe_count = hweight8(new_cdclk_state->active_pipes) > + hweight8(old_cdclk_state->active_pipes); + + /* + * According to "Sequence Before Frequency Change", + * if CDCLK is increasing, set bits 25:16 to upcoming CDCLK, + * if CDCLK is decreasing or not changing, set bits 25:16 to current CDCLK, + * which basically means we choose the maximum of old and new CDCLK, if we know both + */ + if (change_cdclk) + cdclk = max(new_cdclk_state->actual.cdclk, old_cdclk_state->actual.cdclk); + + /* + * According to "Sequence For Pipe Count Change", + * if pipe count is increasing, set bits 25:16 to upcoming pipe count + * (power well is enabled) + * no action if it is decreasing, before the change + */ + if (update_pipe_count) + num_active_pipes = hweight8(new_cdclk_state->active_pipes); + + intel_pcode_notify(i915, voltage_level, num_active_pipes, cdclk, + change_cdclk, update_pipe_count); +} + +static void intel_cdclk_pcode_post_notify(struct intel_atomic_state *state) +{ + struct drm_i915_private *i915 = to_i915(state->base.dev); + const struct intel_cdclk_state *new_cdclk_state = + intel_atomic_get_new_cdclk_state(state); + const struct intel_cdclk_state *old_cdclk_state = + intel_atomic_get_old_cdclk_state(state); + unsigned int cdclk = 0; u8 voltage_level, num_active_pipes = 0; + bool update_cdclk, update_pipe_count; + + /* According to "Sequence After Frequency Change", set voltage to used level */ + voltage_level = new_cdclk_state->actual.voltage_level; + + update_cdclk = new_cdclk_state->actual.cdclk != old_cdclk_state->actual.cdclk; + update_pipe_count = hweight8(new_cdclk_state->active_pipes) < + hweight8(old_cdclk_state->active_pipes); + + /* + * According to "Sequence After Frequency Change", + * set bits 25:16 to current CDCLK + */ + if (update_cdclk) + cdclk = new_cdclk_state->actual.cdclk; + + /* + * According to "Sequence For Pipe Count Change", + * if pipe count is decreasing, set bits 25:16 to current pipe count, + * after the change(power well is disabled) + * no action if it is increasing, after the change + */ + if (update_pipe_count) + num_active_pipes = hweight8(new_cdclk_state->active_pipes); + + intel_pcode_notify(i915, voltage_level, num_active_pipes, cdclk, + update_cdclk, update_pipe_count); +} + /** * intel_set_cdclk_pre_plane_update - Push the CDCLK state to the hardware * @state: intel atomic state @@ -2321,7 +2447,7 @@ static void intel_set_cdclk(struct drm_i915_private *dev_priv, void intel_set_cdclk_pre_plane_update(struct intel_atomic_state *state) { - struct drm_i915_private *dev_priv = to_i915(state->base.dev); + struct drm_i915_private *i915 = to_i915(state->base.dev); const struct intel_cdclk_state *old_cdclk_state = intel_atomic_get_old_cdclk_state(state); const struct intel_cdclk_state *new_cdclk_state = @@ -2332,11 +2458,14 @@ intel_set_cdclk_pre_plane_update(struct intel_atomic_state *state) &new_cdclk_state->actual)) return; + if (IS_DG2(i915)) + intel_cdclk_pcode_pre_notify(state); + if (pipe == INVALID_PIPE || old_cdclk_state->actual.cdclk <= new_cdclk_state->actual.cdclk) { - drm_WARN_ON(&dev_priv->drm, !new_cdclk_state->base.changed); + drm_WARN_ON(&i915->drm, !new_cdclk_state->base.changed); - intel_set_cdclk(dev_priv, &new_cdclk_state->actual, pipe); + intel_set_cdclk(i915, &new_cdclk_state->actual, pipe); } } @@ -2350,7 +2479,7 @@ intel_set_cdclk_pre_plane_update(struct intel_atomic_state *state) void intel_set_cdclk_post_plane_update(struct intel_atomic_state *state) { - struct drm_i915_private *dev_priv = to_i915(state->base.dev); + struct drm_i915_private *i915 = to_i915(state->base.dev); const struct intel_cdclk_state *old_cdclk_state = intel_atomic_get_old_cdclk_state(state); const struct intel_cdclk_state *new_cdclk_state = @@ -2361,11 +2490,14 @@ intel_set_cdclk_post_plane_update(struct intel_atomic_state *state) &new_cdclk_state->actual)) return; + if (IS_DG2(i915)) + intel_cdclk_pcode_post_notify(state); + if (pipe != INVALID_PIPE && old_cdclk_state->actual.cdclk > new_cdclk_state->actual.cdclk) { - drm_WARN_ON(&dev_priv->drm, !new_cdclk_state->base.changed); + drm_WARN_ON(&i915->drm, !new_cdclk_state->base.changed); - intel_set_cdclk(dev_priv, &new_cdclk_state->actual, pipe); + intel_set_cdclk(i915, &new_cdclk_state->actual, pipe); } } @@ -2827,7 +2959,7 @@ int intel_cdclk_atomic_check(struct intel_atomic_state *state, { const struct intel_cdclk_state *old_cdclk_state; const struct intel_cdclk_state *new_cdclk_state; - struct intel_plane_state *plane_state; + struct intel_plane_state __maybe_unused *plane_state; struct intel_plane *plane; int ret; int i; @@ -2871,6 +3003,21 @@ int intel_cdclk_init(struct drm_i915_private *dev_priv) return 0; } +static bool intel_cdclk_need_serialize(struct drm_i915_private *i915, + const struct intel_cdclk_state *old_cdclk_state, + const struct intel_cdclk_state *new_cdclk_state) +{ + bool power_well_cnt_changed = hweight8(old_cdclk_state->active_pipes) != + hweight8(new_cdclk_state->active_pipes); + bool cdclk_changed = intel_cdclk_changed(&old_cdclk_state->actual, + &new_cdclk_state->actual); + /* + * We need to poke hw for gen >= 12, because we notify PCode if + * pipe power well count changes. + */ + return cdclk_changed || (IS_DG2(i915) && power_well_cnt_changed); +} + int intel_modeset_calc_cdclk(struct intel_atomic_state *state) { struct drm_i915_private *dev_priv = to_i915(state->base.dev); @@ -2892,8 +3039,7 @@ int intel_modeset_calc_cdclk(struct intel_atomic_state *state) if (ret) return ret; - if (intel_cdclk_changed(&old_cdclk_state->actual, - &new_cdclk_state->actual)) { + if (intel_cdclk_need_serialize(dev_priv, old_cdclk_state, new_cdclk_state)) { /* * Also serialize commits across all crtcs * if the actual hw needs to be poked. @@ -3235,6 +3381,27 @@ u32 intel_read_rawclk(struct drm_i915_private *dev_priv) return freq; } +static int i915_cdclk_info_show(struct seq_file *m, void *unused) +{ + struct drm_i915_private *i915 = m->private; + + seq_printf(m, "Current CD clock frequency: %d kHz\n", i915->display.cdclk.hw.cdclk); + seq_printf(m, "Max CD clock frequency: %d kHz\n", i915->display.cdclk.max_cdclk_freq); + seq_printf(m, "Max pixel clock frequency: %d kHz\n", i915->max_dotclk_freq); + + return 0; +} + +DEFINE_SHOW_ATTRIBUTE(i915_cdclk_info); + +void intel_cdclk_debugfs_register(struct drm_i915_private *i915) +{ + struct drm_minor *minor = i915->drm.primary; + + debugfs_create_file("i915_cdclk_info", 0444, minor->debugfs_root, + i915, &i915_cdclk_info_fops); +} + static const struct intel_cdclk_funcs mtl_cdclk_funcs = { .get_cdclk = bxt_get_cdclk, .set_cdclk = bxt_set_cdclk, @@ -3242,6 +3409,13 @@ static const struct intel_cdclk_funcs mtl_cdclk_funcs = { .calc_voltage_level = tgl_calc_voltage_level, }; +static const struct intel_cdclk_funcs rplu_cdclk_funcs = { + .get_cdclk = bxt_get_cdclk, + .set_cdclk = bxt_set_cdclk, + .modeset_calc_cdclk = bxt_modeset_calc_cdclk, + .calc_voltage_level = rplu_calc_voltage_level, +}; + static const struct intel_cdclk_funcs tgl_cdclk_funcs = { .get_cdclk = bxt_get_cdclk, .set_cdclk = bxt_set_cdclk, @@ -3384,14 +3558,17 @@ void intel_init_cdclk_hooks(struct drm_i915_private *dev_priv) dev_priv->display.funcs.cdclk = &tgl_cdclk_funcs; dev_priv->display.cdclk.table = dg2_cdclk_table; } else if (IS_ALDERLAKE_P(dev_priv)) { - dev_priv->display.funcs.cdclk = &tgl_cdclk_funcs; /* Wa_22011320316:adl-p[a0] */ - if (IS_ADLP_DISPLAY_STEP(dev_priv, STEP_A0, STEP_B0)) + if (IS_ADLP_DISPLAY_STEP(dev_priv, STEP_A0, STEP_B0)) { dev_priv->display.cdclk.table = adlp_a_step_cdclk_table; - else if (IS_ADLP_RPLU(dev_priv)) + dev_priv->display.funcs.cdclk = &tgl_cdclk_funcs; + } else if (IS_ADLP_RPLU(dev_priv)) { dev_priv->display.cdclk.table = rplu_cdclk_table; - else + dev_priv->display.funcs.cdclk = &rplu_cdclk_funcs; + } else { dev_priv->display.cdclk.table = adlp_cdclk_table; + dev_priv->display.funcs.cdclk = &tgl_cdclk_funcs; + } } else if (IS_ROCKETLAKE(dev_priv)) { dev_priv->display.funcs.cdclk = &tgl_cdclk_funcs; dev_priv->display.cdclk.table = rkl_cdclk_table; diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.h b/drivers/gpu/drm/i915/display/intel_cdclk.h index 51e2f6a11ce4135a9b5b359736c67f3e67436874..48fd7d39e0cd9c4f6d57970f35531b91aa6ab055 100644 --- a/drivers/gpu/drm/i915/display/intel_cdclk.h +++ b/drivers/gpu/drm/i915/display/intel_cdclk.h @@ -82,5 +82,6 @@ intel_atomic_get_cdclk_state(struct intel_atomic_state *state); to_intel_cdclk_state(intel_atomic_get_new_global_obj_state(state, &to_i915(state->base.dev)->display.cdclk.obj)) int intel_cdclk_init(struct drm_i915_private *dev_priv); +void intel_cdclk_debugfs_register(struct drm_i915_private *i915); #endif /* __INTEL_CDCLK_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c index 36aac88143ac1936fe469d7424841d3606729a66..8966e65605168fb1798fa42634726596b72154b0 100644 --- a/drivers/gpu/drm/i915/display/intel_color.c +++ b/drivers/gpu/drm/i915/display/intel_color.c @@ -70,6 +70,11 @@ struct intel_color_funcs { const struct drm_property_blob *blob1, const struct drm_property_blob *blob2, bool is_pre_csc_lut); + /* + * Read out the CSCs (if any) from the hardware into the + * software state. Used by eg. the hardware state checker. + */ + void (*read_csc)(struct intel_crtc_state *crtc_state); }; #define CTM_COEFF_SIGN (1ULL << 63) @@ -116,46 +121,52 @@ struct intel_color_funcs { #define ILK_CSC_COEFF_FP(coeff, fbits) \ (clamp_val(((coeff) >> (32 - (fbits) - 3)) + 4, 0, 0xfff) & 0xff8) -#define ILK_CSC_COEFF_LIMITED_RANGE 0x0dc0 #define ILK_CSC_COEFF_1_0 0x7800 - -#define ILK_CSC_POSTOFF_LIMITED_RANGE (16 * (1 << 12) / 255) - -/* Nop pre/post offsets */ -static const u16 ilk_csc_off_zero[3] = {}; - -/* Identity matrix */ -static const u16 ilk_csc_coeff_identity[9] = { - ILK_CSC_COEFF_1_0, 0, 0, - 0, ILK_CSC_COEFF_1_0, 0, - 0, 0, ILK_CSC_COEFF_1_0, -}; - -/* Limited range RGB post offsets */ -static const u16 ilk_csc_postoff_limited_range[3] = { - ILK_CSC_POSTOFF_LIMITED_RANGE, - ILK_CSC_POSTOFF_LIMITED_RANGE, - ILK_CSC_POSTOFF_LIMITED_RANGE, +#define ILK_CSC_COEFF_LIMITED_RANGE ((235 - 16) << (12 - 8)) /* exponent 0 */ +#define ILK_CSC_POSTOFF_LIMITED_RANGE (16 << (12 - 8)) + +static const struct intel_csc_matrix ilk_csc_matrix_identity = { + .preoff = {}, + .coeff = { + ILK_CSC_COEFF_1_0, 0, 0, + 0, ILK_CSC_COEFF_1_0, 0, + 0, 0, ILK_CSC_COEFF_1_0, + }, + .postoff = {}, }; /* Full range RGB -> limited range RGB matrix */ -static const u16 ilk_csc_coeff_limited_range[9] = { - ILK_CSC_COEFF_LIMITED_RANGE, 0, 0, - 0, ILK_CSC_COEFF_LIMITED_RANGE, 0, - 0, 0, ILK_CSC_COEFF_LIMITED_RANGE, +static const struct intel_csc_matrix ilk_csc_matrix_limited_range = { + .preoff = {}, + .coeff = { + ILK_CSC_COEFF_LIMITED_RANGE, 0, 0, + 0, ILK_CSC_COEFF_LIMITED_RANGE, 0, + 0, 0, ILK_CSC_COEFF_LIMITED_RANGE, + }, + .postoff = { + ILK_CSC_POSTOFF_LIMITED_RANGE, + ILK_CSC_POSTOFF_LIMITED_RANGE, + ILK_CSC_POSTOFF_LIMITED_RANGE, + }, }; /* BT.709 full range RGB -> limited range YCbCr matrix */ -static const u16 ilk_csc_coeff_rgb_to_ycbcr[9] = { - 0x1e08, 0x9cc0, 0xb528, - 0x2ba8, 0x09d8, 0x37e8, - 0xbce8, 0x9ad8, 0x1e08, +static const struct intel_csc_matrix ilk_csc_matrix_rgb_to_ycbcr = { + .preoff = {}, + .coeff = { + 0x1e08, 0x9cc0, 0xb528, + 0x2ba8, 0x09d8, 0x37e8, + 0xbce8, 0x9ad8, 0x1e08, + }, + .postoff = { + 0x0800, 0x0100, 0x0800, + }, }; -/* Limited range YCbCr post offsets */ -static const u16 ilk_csc_postoff_rgb_to_ycbcr[3] = { - 0x0800, 0x0100, 0x0800, -}; +static void intel_csc_clear(struct intel_csc_matrix *csc) +{ + memset(csc, 0, sizeof(*csc)); +} static bool lut_is_legacy(const struct drm_property_blob *lut) { @@ -189,69 +200,182 @@ static u64 *ctm_mult_by_limited(u64 *result, const u64 *input) } static void ilk_update_pipe_csc(struct intel_crtc *crtc, - const u16 preoff[3], - const u16 coeff[9], - const u16 postoff[3]) + const struct intel_csc_matrix *csc) { struct drm_i915_private *i915 = to_i915(crtc->base.dev); enum pipe pipe = crtc->pipe; - intel_de_write_fw(i915, PIPE_CSC_PREOFF_HI(pipe), preoff[0]); - intel_de_write_fw(i915, PIPE_CSC_PREOFF_ME(pipe), preoff[1]); - intel_de_write_fw(i915, PIPE_CSC_PREOFF_LO(pipe), preoff[2]); + intel_de_write_fw(i915, PIPE_CSC_PREOFF_HI(pipe), csc->preoff[0]); + intel_de_write_fw(i915, PIPE_CSC_PREOFF_ME(pipe), csc->preoff[1]); + intel_de_write_fw(i915, PIPE_CSC_PREOFF_LO(pipe), csc->preoff[2]); intel_de_write_fw(i915, PIPE_CSC_COEFF_RY_GY(pipe), - coeff[0] << 16 | coeff[1]); - intel_de_write_fw(i915, PIPE_CSC_COEFF_BY(pipe), coeff[2] << 16); + csc->coeff[0] << 16 | csc->coeff[1]); + intel_de_write_fw(i915, PIPE_CSC_COEFF_BY(pipe), + csc->coeff[2] << 16); intel_de_write_fw(i915, PIPE_CSC_COEFF_RU_GU(pipe), - coeff[3] << 16 | coeff[4]); - intel_de_write_fw(i915, PIPE_CSC_COEFF_BU(pipe), coeff[5] << 16); + csc->coeff[3] << 16 | csc->coeff[4]); + intel_de_write_fw(i915, PIPE_CSC_COEFF_BU(pipe), + csc->coeff[5] << 16); intel_de_write_fw(i915, PIPE_CSC_COEFF_RV_GV(pipe), - coeff[6] << 16 | coeff[7]); - intel_de_write_fw(i915, PIPE_CSC_COEFF_BV(pipe), coeff[8] << 16); + csc->coeff[6] << 16 | csc->coeff[7]); + intel_de_write_fw(i915, PIPE_CSC_COEFF_BV(pipe), + csc->coeff[8] << 16); - if (DISPLAY_VER(i915) >= 7) { - intel_de_write_fw(i915, PIPE_CSC_POSTOFF_HI(pipe), - postoff[0]); - intel_de_write_fw(i915, PIPE_CSC_POSTOFF_ME(pipe), - postoff[1]); - intel_de_write_fw(i915, PIPE_CSC_POSTOFF_LO(pipe), - postoff[2]); - } + if (DISPLAY_VER(i915) < 7) + return; + + intel_de_write_fw(i915, PIPE_CSC_POSTOFF_HI(pipe), csc->postoff[0]); + intel_de_write_fw(i915, PIPE_CSC_POSTOFF_ME(pipe), csc->postoff[1]); + intel_de_write_fw(i915, PIPE_CSC_POSTOFF_LO(pipe), csc->postoff[2]); +} + +static void ilk_read_pipe_csc(struct intel_crtc *crtc, + struct intel_csc_matrix *csc) +{ + struct drm_i915_private *i915 = to_i915(crtc->base.dev); + enum pipe pipe = crtc->pipe; + u32 tmp; + + csc->preoff[0] = intel_de_read_fw(i915, PIPE_CSC_PREOFF_HI(pipe)); + csc->preoff[1] = intel_de_read_fw(i915, PIPE_CSC_PREOFF_ME(pipe)); + csc->preoff[2] = intel_de_read_fw(i915, PIPE_CSC_PREOFF_LO(pipe)); + + tmp = intel_de_read_fw(i915, PIPE_CSC_COEFF_RY_GY(pipe)); + csc->coeff[0] = tmp >> 16; + csc->coeff[1] = tmp & 0xffff; + tmp = intel_de_read_fw(i915, PIPE_CSC_COEFF_BY(pipe)); + csc->coeff[2] = tmp >> 16; + + tmp = intel_de_read_fw(i915, PIPE_CSC_COEFF_RU_GU(pipe)); + csc->coeff[3] = tmp >> 16; + csc->coeff[4] = tmp & 0xffff; + tmp = intel_de_read_fw(i915, PIPE_CSC_COEFF_BU(pipe)); + csc->coeff[5] = tmp >> 16; + + tmp = intel_de_read_fw(i915, PIPE_CSC_COEFF_RV_GV(pipe)); + csc->coeff[6] = tmp >> 16; + csc->coeff[7] = tmp & 0xffff; + tmp = intel_de_read_fw(i915, PIPE_CSC_COEFF_BV(pipe)); + csc->coeff[8] = tmp >> 16; + + if (DISPLAY_VER(i915) < 7) + return; + + csc->postoff[0] = intel_de_read_fw(i915, PIPE_CSC_POSTOFF_HI(pipe)); + csc->postoff[1] = intel_de_read_fw(i915, PIPE_CSC_POSTOFF_ME(pipe)); + csc->postoff[2] = intel_de_read_fw(i915, PIPE_CSC_POSTOFF_LO(pipe)); +} + +static void ilk_read_csc(struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + + if (crtc_state->csc_enable) + ilk_read_pipe_csc(crtc, &crtc_state->csc); +} + +static void skl_read_csc(struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + + /* + * Display WA #1184: skl,glk + * Wa_1406463849: icl + * + * Danger! On SKL-ICL *reads* from the CSC coeff/offset registers + * will disarm an already armed CSC double buffer update. + * So this must not be called while armed. Fortunately the state checker + * readout happens only after the update has been already been latched. + * + * On earlier and later platforms only writes to said registers will + * disarm the update. This is considered normal behavior and also + * happens with various other hardware units. + */ + if (crtc_state->csc_enable) + ilk_read_pipe_csc(crtc, &crtc_state->csc); } static void icl_update_output_csc(struct intel_crtc *crtc, - const u16 preoff[3], - const u16 coeff[9], - const u16 postoff[3]) + const struct intel_csc_matrix *csc) { struct drm_i915_private *i915 = to_i915(crtc->base.dev); enum pipe pipe = crtc->pipe; - intel_de_write_fw(i915, PIPE_CSC_OUTPUT_PREOFF_HI(pipe), preoff[0]); - intel_de_write_fw(i915, PIPE_CSC_OUTPUT_PREOFF_ME(pipe), preoff[1]); - intel_de_write_fw(i915, PIPE_CSC_OUTPUT_PREOFF_LO(pipe), preoff[2]); + intel_de_write_fw(i915, PIPE_CSC_OUTPUT_PREOFF_HI(pipe), csc->preoff[0]); + intel_de_write_fw(i915, PIPE_CSC_OUTPUT_PREOFF_ME(pipe), csc->preoff[1]); + intel_de_write_fw(i915, PIPE_CSC_OUTPUT_PREOFF_LO(pipe), csc->preoff[2]); intel_de_write_fw(i915, PIPE_CSC_OUTPUT_COEFF_RY_GY(pipe), - coeff[0] << 16 | coeff[1]); + csc->coeff[0] << 16 | csc->coeff[1]); intel_de_write_fw(i915, PIPE_CSC_OUTPUT_COEFF_BY(pipe), - coeff[2] << 16); + csc->coeff[2] << 16); intel_de_write_fw(i915, PIPE_CSC_OUTPUT_COEFF_RU_GU(pipe), - coeff[3] << 16 | coeff[4]); + csc->coeff[3] << 16 | csc->coeff[4]); intel_de_write_fw(i915, PIPE_CSC_OUTPUT_COEFF_BU(pipe), - coeff[5] << 16); + csc->coeff[5] << 16); intel_de_write_fw(i915, PIPE_CSC_OUTPUT_COEFF_RV_GV(pipe), - coeff[6] << 16 | coeff[7]); + csc->coeff[6] << 16 | csc->coeff[7]); intel_de_write_fw(i915, PIPE_CSC_OUTPUT_COEFF_BV(pipe), - coeff[8] << 16); + csc->coeff[8] << 16); + + intel_de_write_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_HI(pipe), csc->postoff[0]); + intel_de_write_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_ME(pipe), csc->postoff[1]); + intel_de_write_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_LO(pipe), csc->postoff[2]); +} + +static void icl_read_output_csc(struct intel_crtc *crtc, + struct intel_csc_matrix *csc) +{ + struct drm_i915_private *i915 = to_i915(crtc->base.dev); + enum pipe pipe = crtc->pipe; + u32 tmp; + + csc->preoff[0] = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_PREOFF_HI(pipe)); + csc->preoff[1] = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_PREOFF_ME(pipe)); + csc->preoff[2] = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_PREOFF_LO(pipe)); + + tmp = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_COEFF_RY_GY(pipe)); + csc->coeff[0] = tmp >> 16; + csc->coeff[1] = tmp & 0xffff; + tmp = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_COEFF_BY(pipe)); + csc->coeff[2] = tmp >> 16; + + tmp = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_COEFF_RU_GU(pipe)); + csc->coeff[3] = tmp >> 16; + csc->coeff[4] = tmp & 0xffff; + tmp = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_COEFF_BU(pipe)); + csc->coeff[5] = tmp >> 16; + + tmp = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_COEFF_RV_GV(pipe)); + csc->coeff[6] = tmp >> 16; + csc->coeff[7] = tmp & 0xffff; + tmp = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_COEFF_BV(pipe)); + csc->coeff[8] = tmp >> 16; - intel_de_write_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_HI(pipe), postoff[0]); - intel_de_write_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_ME(pipe), postoff[1]); - intel_de_write_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_LO(pipe), postoff[2]); + csc->postoff[0] = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_HI(pipe)); + csc->postoff[1] = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_ME(pipe)); + csc->postoff[2] = intel_de_read_fw(i915, PIPE_CSC_OUTPUT_POSTOFF_LO(pipe)); +} + +static void icl_read_csc(struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + + /* + * Wa_1406463849: icl + * + * See skl_read_csc() + */ + if (crtc_state->csc_mode & ICL_CSC_ENABLE) + ilk_read_pipe_csc(crtc, &crtc_state->csc); + + if (crtc_state->csc_mode & ICL_OUTPUT_CSC_ENABLE) + icl_read_output_csc(crtc, &crtc_state->output_csc); } static bool ilk_limited_range(const struct intel_crtc_state *crtc_state) @@ -294,14 +418,32 @@ static bool ilk_csc_limited_range(const struct intel_crtc_state *crtc_state) return !ilk_lut_limited_range(crtc_state); } +static void ilk_csc_copy(struct drm_i915_private *i915, + struct intel_csc_matrix *dst, + const struct intel_csc_matrix *src) +{ + *dst = *src; + + if (DISPLAY_VER(i915) < 7) + memset(dst->postoff, 0, sizeof(dst->postoff)); +} + static void ilk_csc_convert_ctm(const struct intel_crtc_state *crtc_state, - u16 coeffs[9], bool limited_color_range) + struct intel_csc_matrix *csc, + bool limited_color_range) { + struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev); const struct drm_color_ctm *ctm = crtc_state->hw.ctm->data; const u64 *input; u64 temp[9]; int i; + /* for preoff/postoff */ + if (limited_color_range) + ilk_csc_copy(i915, csc, &ilk_csc_matrix_limited_range); + else + ilk_csc_copy(i915, csc, &ilk_csc_matrix_identity); + if (limited_color_range) input = ctm_mult_by_limited(temp, ctm->matrix); else @@ -320,54 +462,49 @@ static void ilk_csc_convert_ctm(const struct intel_crtc_state *crtc_state, */ abs_coeff = clamp_val(abs_coeff, 0, CTM_COEFF_4_0 - 1); - coeffs[i] = 0; + csc->coeff[i] = 0; /* sign bit */ if (CTM_COEFF_NEGATIVE(input[i])) - coeffs[i] |= 1 << 15; + csc->coeff[i] |= 1 << 15; if (abs_coeff < CTM_COEFF_0_125) - coeffs[i] |= (3 << 12) | + csc->coeff[i] |= (3 << 12) | ILK_CSC_COEFF_FP(abs_coeff, 12); else if (abs_coeff < CTM_COEFF_0_25) - coeffs[i] |= (2 << 12) | + csc->coeff[i] |= (2 << 12) | ILK_CSC_COEFF_FP(abs_coeff, 11); else if (abs_coeff < CTM_COEFF_0_5) - coeffs[i] |= (1 << 12) | + csc->coeff[i] |= (1 << 12) | ILK_CSC_COEFF_FP(abs_coeff, 10); else if (abs_coeff < CTM_COEFF_1_0) - coeffs[i] |= ILK_CSC_COEFF_FP(abs_coeff, 9); + csc->coeff[i] |= ILK_CSC_COEFF_FP(abs_coeff, 9); else if (abs_coeff < CTM_COEFF_2_0) - coeffs[i] |= (7 << 12) | + csc->coeff[i] |= (7 << 12) | ILK_CSC_COEFF_FP(abs_coeff, 8); else - coeffs[i] |= (6 << 12) | + csc->coeff[i] |= (6 << 12) | ILK_CSC_COEFF_FP(abs_coeff, 7); } } -static void ilk_load_csc_matrix(const struct intel_crtc_state *crtc_state) +static void ilk_assign_csc(struct intel_crtc_state *crtc_state) { - struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); - struct drm_i915_private *i915 = to_i915(crtc->base.dev); + struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev); bool limited_color_range = ilk_csc_limited_range(crtc_state); if (crtc_state->hw.ctm) { - u16 coeff[9]; + drm_WARN_ON(&i915->drm, !crtc_state->csc_enable); - ilk_csc_convert_ctm(crtc_state, coeff, limited_color_range); - ilk_update_pipe_csc(crtc, ilk_csc_off_zero, coeff, - limited_color_range ? - ilk_csc_postoff_limited_range : - ilk_csc_off_zero); + ilk_csc_convert_ctm(crtc_state, &crtc_state->csc, limited_color_range); } else if (crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB) { - ilk_update_pipe_csc(crtc, ilk_csc_off_zero, - ilk_csc_coeff_rgb_to_ycbcr, - ilk_csc_postoff_rgb_to_ycbcr); + drm_WARN_ON(&i915->drm, !crtc_state->csc_enable); + + ilk_csc_copy(i915, &crtc_state->csc, &ilk_csc_matrix_rgb_to_ycbcr); } else if (limited_color_range) { - ilk_update_pipe_csc(crtc, ilk_csc_off_zero, - ilk_csc_coeff_limited_range, - ilk_csc_postoff_limited_range); + drm_WARN_ON(&i915->drm, !crtc_state->csc_enable); + + ilk_csc_copy(i915, &crtc_state->csc, &ilk_csc_matrix_limited_range); } else if (crtc_state->csc_enable) { /* * On GLK both pipe CSC and degamma LUT are controlled @@ -377,72 +514,267 @@ static void ilk_load_csc_matrix(const struct intel_crtc_state *crtc_state) */ drm_WARN_ON(&i915->drm, !IS_GEMINILAKE(i915)); - ilk_update_pipe_csc(crtc, ilk_csc_off_zero, - ilk_csc_coeff_identity, - ilk_csc_off_zero); + ilk_csc_copy(i915, &crtc_state->csc, &ilk_csc_matrix_identity); + } else { + intel_csc_clear(&crtc_state->csc); } } -static void icl_load_csc_matrix(const struct intel_crtc_state *crtc_state) +static void ilk_load_csc_matrix(const struct intel_crtc_state *crtc_state) { struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + if (crtc_state->csc_enable) + ilk_update_pipe_csc(crtc, &crtc_state->csc); +} + +static void icl_assign_csc(struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev); + if (crtc_state->hw.ctm) { - u16 coeff[9]; + drm_WARN_ON(&i915->drm, (crtc_state->csc_mode & ICL_CSC_ENABLE) == 0); + + ilk_csc_convert_ctm(crtc_state, &crtc_state->csc, false); + } else { + drm_WARN_ON(&i915->drm, (crtc_state->csc_mode & ICL_CSC_ENABLE) != 0); - ilk_csc_convert_ctm(crtc_state, coeff, false); - ilk_update_pipe_csc(crtc, ilk_csc_off_zero, - coeff, ilk_csc_off_zero); + intel_csc_clear(&crtc_state->csc); } if (crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB) { - icl_update_output_csc(crtc, ilk_csc_off_zero, - ilk_csc_coeff_rgb_to_ycbcr, - ilk_csc_postoff_rgb_to_ycbcr); + drm_WARN_ON(&i915->drm, (crtc_state->csc_mode & ICL_OUTPUT_CSC_ENABLE) == 0); + + ilk_csc_copy(i915, &crtc_state->output_csc, &ilk_csc_matrix_rgb_to_ycbcr); } else if (crtc_state->limited_color_range) { - icl_update_output_csc(crtc, ilk_csc_off_zero, - ilk_csc_coeff_limited_range, - ilk_csc_postoff_limited_range); + drm_WARN_ON(&i915->drm, (crtc_state->csc_mode & ICL_OUTPUT_CSC_ENABLE) == 0); + + ilk_csc_copy(i915, &crtc_state->output_csc, &ilk_csc_matrix_limited_range); + } else { + drm_WARN_ON(&i915->drm, (crtc_state->csc_mode & ICL_OUTPUT_CSC_ENABLE) != 0); + + intel_csc_clear(&crtc_state->output_csc); } } -static void chv_load_cgm_csc(struct intel_crtc *crtc, - const struct drm_property_blob *blob) +static void icl_load_csc_matrix(const struct intel_crtc_state *crtc_state) { - struct drm_i915_private *i915 = to_i915(crtc->base.dev); - const struct drm_color_ctm *ctm = blob->data; - enum pipe pipe = crtc->pipe; - u16 coeffs[9]; + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + + if (crtc_state->csc_mode & ICL_CSC_ENABLE) + ilk_update_pipe_csc(crtc, &crtc_state->csc); + + if (crtc_state->csc_mode & ICL_OUTPUT_CSC_ENABLE) + icl_update_output_csc(crtc, &crtc_state->output_csc); +} + +static u16 ctm_to_twos_complement(u64 coeff, int int_bits, int frac_bits) +{ + s64 c = CTM_COEFF_ABS(coeff); + + /* leave an extra bit for rounding */ + c >>= 32 - frac_bits - 1; + + /* round and drop the extra bit */ + c = (c + 1) >> 1; + + if (CTM_COEFF_NEGATIVE(coeff)) + c = -c; + + c = clamp(c, -(s64)BIT(int_bits + frac_bits - 1), + (s64)(BIT(int_bits + frac_bits - 1) - 1)); + + return c & (BIT(int_bits + frac_bits) - 1); +} + +/* + * VLV/CHV Wide Gamut Color Correction (WGC) CSC + * |r| | c0 c1 c2 | |r| + * |g| = | c3 c4 c5 | x |g| + * |b| | c6 c7 c8 | |b| + * + * Coefficients are two's complement s2.10. + */ +static void vlv_wgc_csc_convert_ctm(const struct intel_crtc_state *crtc_state, + struct intel_csc_matrix *csc) +{ + const struct drm_color_ctm *ctm = crtc_state->hw.ctm->data; int i; - for (i = 0; i < ARRAY_SIZE(coeffs); i++) { - u64 abs_coeff = ((1ULL << 63) - 1) & ctm->matrix[i]; + for (i = 0; i < 9; i++) + csc->coeff[i] = ctm_to_twos_complement(ctm->matrix[i], 2, 10); +} + +static void vlv_load_wgc_csc(struct intel_crtc *crtc, + const struct intel_csc_matrix *csc) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + enum pipe pipe = crtc->pipe; + + intel_de_write_fw(dev_priv, PIPE_WGC_C01_C00(pipe), + csc->coeff[1] << 16 | csc->coeff[0]); + intel_de_write_fw(dev_priv, PIPE_WGC_C02(pipe), + csc->coeff[2]); + + intel_de_write_fw(dev_priv, PIPE_WGC_C11_C10(pipe), + csc->coeff[4] << 16 | csc->coeff[3]); + intel_de_write_fw(dev_priv, PIPE_WGC_C12(pipe), + csc->coeff[5]); + + intel_de_write_fw(dev_priv, PIPE_WGC_C21_C20(pipe), + csc->coeff[7] << 16 | csc->coeff[6]); + intel_de_write_fw(dev_priv, PIPE_WGC_C22(pipe), + csc->coeff[8]); +} + +static void vlv_read_wgc_csc(struct intel_crtc *crtc, + struct intel_csc_matrix *csc) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + enum pipe pipe = crtc->pipe; + u32 tmp; + + tmp = intel_de_read_fw(dev_priv, PIPE_WGC_C01_C00(pipe)); + csc->coeff[0] = tmp & 0xffff; + csc->coeff[1] = tmp >> 16; - /* Round coefficient. */ - abs_coeff += 1 << (32 - 13); - /* Clamp to hardware limits. */ - abs_coeff = clamp_val(abs_coeff, 0, CTM_COEFF_8_0 - 1); + tmp = intel_de_read_fw(dev_priv, PIPE_WGC_C02(pipe)); + csc->coeff[2] = tmp & 0xffff; - coeffs[i] = 0; + tmp = intel_de_read_fw(dev_priv, PIPE_WGC_C11_C10(pipe)); + csc->coeff[3] = tmp & 0xffff; + csc->coeff[4] = tmp >> 16; - /* Write coefficients in S3.12 format. */ - if (ctm->matrix[i] & (1ULL << 63)) - coeffs[i] |= 1 << 15; + tmp = intel_de_read_fw(dev_priv, PIPE_WGC_C12(pipe)); + csc->coeff[5] = tmp & 0xffff; - coeffs[i] |= ((abs_coeff >> 32) & 7) << 12; - coeffs[i] |= (abs_coeff >> 20) & 0xfff; + tmp = intel_de_read_fw(dev_priv, PIPE_WGC_C21_C20(pipe)); + csc->coeff[6] = tmp & 0xffff; + csc->coeff[7] = tmp >> 16; + + tmp = intel_de_read_fw(dev_priv, PIPE_WGC_C22(pipe)); + csc->coeff[8] = tmp & 0xffff; +} + +static void vlv_read_csc(struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + + if (crtc_state->wgc_enable) + vlv_read_wgc_csc(crtc, &crtc_state->csc); +} + +static void vlv_assign_csc(struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev); + + if (crtc_state->hw.ctm) { + drm_WARN_ON(&i915->drm, !crtc_state->wgc_enable); + + vlv_wgc_csc_convert_ctm(crtc_state, &crtc_state->csc); + } else { + drm_WARN_ON(&i915->drm, crtc_state->wgc_enable); + + intel_csc_clear(&crtc_state->csc); } +} + +/* + * CHV Color Gamut Mapping (CGM) CSC + * |r| | c0 c1 c2 | |r| + * |g| = | c3 c4 c5 | x |g| + * |b| | c6 c7 c8 | |b| + * + * Coefficients are two's complement s4.12. + */ +static void chv_cgm_csc_convert_ctm(const struct intel_crtc_state *crtc_state, + struct intel_csc_matrix *csc) +{ + const struct drm_color_ctm *ctm = crtc_state->hw.ctm->data; + int i; + + for (i = 0; i < 9; i++) + csc->coeff[i] = ctm_to_twos_complement(ctm->matrix[i], 4, 12); +} + +#define CHV_CGM_CSC_COEFF_1_0 (1 << 12) + +static const struct intel_csc_matrix chv_cgm_csc_matrix_identity = { + .coeff = { + CHV_CGM_CSC_COEFF_1_0, 0, 0, + 0, CHV_CGM_CSC_COEFF_1_0, 0, + 0, 0, CHV_CGM_CSC_COEFF_1_0, + }, +}; + +static void chv_load_cgm_csc(struct intel_crtc *crtc, + const struct intel_csc_matrix *csc) +{ + struct drm_i915_private *i915 = to_i915(crtc->base.dev); + enum pipe pipe = crtc->pipe; intel_de_write_fw(i915, CGM_PIPE_CSC_COEFF01(pipe), - coeffs[1] << 16 | coeffs[0]); + csc->coeff[1] << 16 | csc->coeff[0]); intel_de_write_fw(i915, CGM_PIPE_CSC_COEFF23(pipe), - coeffs[3] << 16 | coeffs[2]); + csc->coeff[3] << 16 | csc->coeff[2]); intel_de_write_fw(i915, CGM_PIPE_CSC_COEFF45(pipe), - coeffs[5] << 16 | coeffs[4]); + csc->coeff[5] << 16 | csc->coeff[4]); intel_de_write_fw(i915, CGM_PIPE_CSC_COEFF67(pipe), - coeffs[7] << 16 | coeffs[6]); + csc->coeff[7] << 16 | csc->coeff[6]); intel_de_write_fw(i915, CGM_PIPE_CSC_COEFF8(pipe), - coeffs[8]); + csc->coeff[8]); +} + +static void chv_read_cgm_csc(struct intel_crtc *crtc, + struct intel_csc_matrix *csc) +{ + struct drm_i915_private *i915 = to_i915(crtc->base.dev); + enum pipe pipe = crtc->pipe; + u32 tmp; + + tmp = intel_de_read_fw(i915, CGM_PIPE_CSC_COEFF01(pipe)); + csc->coeff[0] = tmp & 0xffff; + csc->coeff[1] = tmp >> 16; + + tmp = intel_de_read_fw(i915, CGM_PIPE_CSC_COEFF23(pipe)); + csc->coeff[2] = tmp & 0xffff; + csc->coeff[3] = tmp >> 16; + + tmp = intel_de_read_fw(i915, CGM_PIPE_CSC_COEFF45(pipe)); + csc->coeff[4] = tmp & 0xffff; + csc->coeff[5] = tmp >> 16; + + tmp = intel_de_read_fw(i915, CGM_PIPE_CSC_COEFF67(pipe)); + csc->coeff[6] = tmp & 0xffff; + csc->coeff[7] = tmp >> 16; + + tmp = intel_de_read_fw(i915, CGM_PIPE_CSC_COEFF8(pipe)); + csc->coeff[8] = tmp & 0xffff; +} + +static void chv_read_csc(struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + + if (crtc_state->cgm_mode & CGM_PIPE_MODE_CSC) + chv_read_cgm_csc(crtc, &crtc_state->csc); +} + +static void chv_assign_csc(struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev); + + drm_WARN_ON(&i915->drm, crtc_state->wgc_enable); + + if (crtc_state->hw.ctm) { + drm_WARN_ON(&i915->drm, (crtc_state->cgm_mode & CGM_PIPE_MODE_CSC) == 0); + + chv_cgm_csc_convert_ctm(crtc_state, &crtc_state->csc); + } else { + drm_WARN_ON(&i915->drm, (crtc_state->cgm_mode & CGM_PIPE_MODE_CSC) == 0); + + crtc_state->csc = chv_cgm_csc_matrix_identity; + } } /* convert hw value with given bit_precision to lut property val */ @@ -1336,6 +1668,16 @@ static void icl_load_luts(const struct intel_crtc_state *crtc_state) } } +static void vlv_load_luts(const struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + + if (crtc_state->wgc_enable) + vlv_load_wgc_csc(crtc, &crtc_state->csc); + + i965_load_luts(crtc_state); +} + static u32 chv_cgm_degamma_ldw(const struct drm_color_lut *color) { return REG_FIELD_PREP(CGM_PIPE_DEGAMMA_GREEN_LDW_MASK, drm_color_lut_extract(color->green, 14)) | @@ -1410,10 +1752,9 @@ static void chv_load_luts(const struct intel_crtc_state *crtc_state) struct drm_i915_private *i915 = to_i915(crtc->base.dev); const struct drm_property_blob *pre_csc_lut = crtc_state->pre_csc_lut; const struct drm_property_blob *post_csc_lut = crtc_state->post_csc_lut; - const struct drm_property_blob *ctm = crtc_state->hw.ctm; if (crtc_state->cgm_mode & CGM_PIPE_MODE_CSC) - chv_load_cgm_csc(crtc, ctm); + chv_load_cgm_csc(crtc, &crtc_state->csc); if (crtc_state->cgm_mode & CGM_PIPE_MODE_DEGAMMA) chv_load_cgm_degamma(crtc, pre_csc_lut); @@ -1491,6 +1832,18 @@ static bool intel_can_preload_luts(const struct intel_crtc_state *new_crtc_state !old_crtc_state->pre_csc_lut; } +static bool vlv_can_preload_luts(const struct intel_crtc_state *new_crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(new_crtc_state->uapi.crtc); + struct intel_atomic_state *state = + to_intel_atomic_state(new_crtc_state->uapi.state); + const struct intel_crtc_state *old_crtc_state = + intel_atomic_get_old_crtc_state(state, crtc); + + return !old_crtc_state->wgc_enable && + !old_crtc_state->post_csc_lut; +} + static bool chv_can_preload_luts(const struct intel_crtc_state *new_crtc_state) { struct intel_crtc *crtc = to_intel_crtc(new_crtc_state->uapi.crtc); @@ -1507,7 +1860,7 @@ static bool chv_can_preload_luts(const struct intel_crtc_state *new_crtc_state) if (old_crtc_state->cgm_mode || new_crtc_state->cgm_mode) return false; - return !old_crtc_state->post_csc_lut; + return vlv_can_preload_luts(new_crtc_state); } int intel_color_check(struct intel_crtc_state *crtc_state) @@ -1522,6 +1875,9 @@ void intel_color_get_config(struct intel_crtc_state *crtc_state) struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev); i915->display.funcs.color->read_luts(crtc_state); + + if (i915->display.funcs.color->read_csc) + i915->display.funcs.color->read_csc(crtc_state); } bool intel_color_lut_equal(const struct intel_crtc_state *crtc_state, @@ -1606,14 +1962,14 @@ static u32 intel_gamma_lut_tests(const struct intel_crtc_state *crtc_state) if (lut_is_legacy(gamma_lut)) return 0; - return INTEL_INFO(i915)->display.color.gamma_lut_tests; + return DISPLAY_INFO(i915)->color.gamma_lut_tests; } static u32 intel_degamma_lut_tests(const struct intel_crtc_state *crtc_state) { struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev); - return INTEL_INFO(i915)->display.color.degamma_lut_tests; + return DISPLAY_INFO(i915)->color.degamma_lut_tests; } static int intel_gamma_lut_size(const struct intel_crtc_state *crtc_state) @@ -1624,14 +1980,14 @@ static int intel_gamma_lut_size(const struct intel_crtc_state *crtc_state) if (lut_is_legacy(gamma_lut)) return LEGACY_LUT_LENGTH; - return INTEL_INFO(i915)->display.color.gamma_lut_size; + return DISPLAY_INFO(i915)->color.gamma_lut_size; } static u32 intel_degamma_lut_size(const struct intel_crtc_state *crtc_state) { struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev); - return INTEL_INFO(i915)->display.color.degamma_lut_size; + return DISPLAY_INFO(i915)->color.degamma_lut_size; } static int check_lut_size(const struct drm_property_blob *lut, int expected) @@ -1791,6 +2147,39 @@ static int i9xx_color_check(struct intel_crtc_state *crtc_state) return 0; } +/* + * VLV color pipeline: + * u0.10 -> WGC csc -> u0.10 -> pipe gamma -> u0.10 + */ +static int vlv_color_check(struct intel_crtc_state *crtc_state) +{ + int ret; + + ret = check_luts(crtc_state); + if (ret) + return ret; + + crtc_state->gamma_enable = + crtc_state->hw.gamma_lut && + !crtc_state->c8_planes; + + crtc_state->gamma_mode = i9xx_gamma_mode(crtc_state); + + crtc_state->wgc_enable = crtc_state->hw.ctm; + + ret = intel_color_add_affected_planes(crtc_state); + if (ret) + return ret; + + intel_assign_luts(crtc_state); + + vlv_assign_csc(crtc_state); + + crtc_state->preload_luts = vlv_can_preload_luts(crtc_state); + + return 0; +} + static u32 chv_cgm_mode(const struct intel_crtc_state *crtc_state) { u32 cgm_mode = 0; @@ -1803,6 +2192,13 @@ static u32 chv_cgm_mode(const struct intel_crtc_state *crtc_state) !lut_is_legacy(crtc_state->hw.gamma_lut)) cgm_mode |= CGM_PIPE_MODE_GAMMA; + /* + * Toggling the CGM CSC on/off outside of the tiny window + * between start of vblank and frame start causes underruns. + * Always enable the CGM CSC as a workaround. + */ + cgm_mode |= CGM_PIPE_MODE_CSC; + return cgm_mode; } @@ -1834,12 +2230,20 @@ static int chv_color_check(struct intel_crtc_state *crtc_state) crtc_state->cgm_mode = chv_cgm_mode(crtc_state); + /* + * We always bypass the WGC CSC and use the CGM CSC + * instead since it has degamma and better precision. + */ + crtc_state->wgc_enable = false; + ret = intel_color_add_affected_planes(crtc_state); if (ret) return ret; intel_assign_luts(crtc_state); + chv_assign_csc(crtc_state); + crtc_state->preload_luts = chv_can_preload_luts(crtc_state); return 0; @@ -1962,6 +2366,8 @@ static int ilk_color_check(struct intel_crtc_state *crtc_state) if (ret) return ret; + ilk_assign_csc(crtc_state); + crtc_state->preload_luts = intel_can_preload_luts(crtc_state); return 0; @@ -2068,6 +2474,8 @@ static int ivb_color_check(struct intel_crtc_state *crtc_state) if (ret) return ret; + ilk_assign_csc(crtc_state); + crtc_state->preload_luts = intel_can_preload_luts(crtc_state); return 0; @@ -2097,7 +2505,7 @@ static int glk_assign_luts(struct intel_crtc_state *crtc_state) struct drm_property_blob *gamma_lut; gamma_lut = create_resized_lut(i915, crtc_state->hw.gamma_lut, - INTEL_INFO(i915)->display.color.degamma_lut_size, + DISPLAY_INFO(i915)->color.degamma_lut_size, false); if (IS_ERR(gamma_lut)) return PTR_ERR(gamma_lut); @@ -2199,6 +2607,8 @@ static int glk_color_check(struct intel_crtc_state *crtc_state) if (ret) return ret; + ilk_assign_csc(crtc_state); + crtc_state->preload_luts = intel_can_preload_luts(crtc_state); return 0; @@ -2261,6 +2671,8 @@ static int icl_color_check(struct intel_crtc_state *crtc_state) intel_assign_luts(crtc_state); + icl_assign_csc(crtc_state); + crtc_state->preload_luts = intel_can_preload_luts(crtc_state); return 0; @@ -2627,7 +3039,7 @@ static struct drm_property_blob *i9xx_read_lut_8(struct intel_crtc *crtc) static struct drm_property_blob *i9xx_read_lut_10(struct intel_crtc *crtc) { struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - u32 lut_size = INTEL_INFO(dev_priv)->display.color.gamma_lut_size; + u32 lut_size = DISPLAY_INFO(dev_priv)->color.gamma_lut_size; enum pipe pipe = crtc->pipe; struct drm_property_blob *blob; struct drm_color_lut *lut; @@ -2676,7 +3088,7 @@ static void i9xx_read_luts(struct intel_crtc_state *crtc_state) static struct drm_property_blob *i965_read_lut_10p6(struct intel_crtc *crtc) { struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - int i, lut_size = INTEL_INFO(dev_priv)->display.color.gamma_lut_size; + int i, lut_size = DISPLAY_INFO(dev_priv)->color.gamma_lut_size; enum pipe pipe = crtc->pipe; struct drm_property_blob *blob; struct drm_color_lut *lut; @@ -2726,7 +3138,7 @@ static void i965_read_luts(struct intel_crtc_state *crtc_state) static struct drm_property_blob *chv_read_cgm_degamma(struct intel_crtc *crtc) { struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - int i, lut_size = INTEL_INFO(dev_priv)->display.color.degamma_lut_size; + int i, lut_size = DISPLAY_INFO(dev_priv)->color.degamma_lut_size; enum pipe pipe = crtc->pipe; struct drm_property_blob *blob; struct drm_color_lut *lut; @@ -2752,7 +3164,7 @@ static struct drm_property_blob *chv_read_cgm_degamma(struct intel_crtc *crtc) static struct drm_property_blob *chv_read_cgm_gamma(struct intel_crtc *crtc) { struct drm_i915_private *i915 = to_i915(crtc->base.dev); - int i, lut_size = INTEL_INFO(i915)->display.color.gamma_lut_size; + int i, lut_size = DISPLAY_INFO(i915)->color.gamma_lut_size; enum pipe pipe = crtc->pipe; struct drm_property_blob *blob; struct drm_color_lut *lut; @@ -2816,7 +3228,7 @@ static struct drm_property_blob *ilk_read_lut_8(struct intel_crtc *crtc) static struct drm_property_blob *ilk_read_lut_10(struct intel_crtc *crtc) { struct drm_i915_private *i915 = to_i915(crtc->base.dev); - int i, lut_size = INTEL_INFO(i915)->display.color.gamma_lut_size; + int i, lut_size = DISPLAY_INFO(i915)->color.gamma_lut_size; enum pipe pipe = crtc->pipe; struct drm_property_blob *blob; struct drm_color_lut *lut; @@ -3000,7 +3412,7 @@ static void bdw_read_luts(struct intel_crtc_state *crtc_state) static struct drm_property_blob *glk_read_degamma_lut(struct intel_crtc *crtc) { struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - int i, lut_size = INTEL_INFO(dev_priv)->display.color.degamma_lut_size; + int i, lut_size = DISPLAY_INFO(dev_priv)->color.degamma_lut_size; enum pipe pipe = crtc->pipe; struct drm_property_blob *blob; struct drm_color_lut *lut; @@ -3065,7 +3477,7 @@ static struct drm_property_blob * icl_read_lut_multi_segment(struct intel_crtc *crtc) { struct drm_i915_private *i915 = to_i915(crtc->base.dev); - int i, lut_size = INTEL_INFO(i915)->display.color.gamma_lut_size; + int i, lut_size = DISPLAY_INFO(i915)->color.gamma_lut_size; enum pipe pipe = crtc->pipe; struct drm_property_blob *blob; struct drm_color_lut *lut; @@ -3135,6 +3547,16 @@ static const struct intel_color_funcs chv_color_funcs = { .load_luts = chv_load_luts, .read_luts = chv_read_luts, .lut_equal = chv_lut_equal, + .read_csc = chv_read_csc, +}; + +static const struct intel_color_funcs vlv_color_funcs = { + .color_check = vlv_color_check, + .color_commit_arm = i9xx_color_commit_arm, + .load_luts = vlv_load_luts, + .read_luts = i965_read_luts, + .lut_equal = i965_lut_equal, + .read_csc = vlv_read_csc, }; static const struct intel_color_funcs i965_color_funcs = { @@ -3160,6 +3582,7 @@ static const struct intel_color_funcs tgl_color_funcs = { .load_luts = icl_load_luts, .read_luts = icl_read_luts, .lut_equal = icl_lut_equal, + .read_csc = icl_read_csc, }; static const struct intel_color_funcs icl_color_funcs = { @@ -3170,6 +3593,7 @@ static const struct intel_color_funcs icl_color_funcs = { .load_luts = icl_load_luts, .read_luts = icl_read_luts, .lut_equal = icl_lut_equal, + .read_csc = icl_read_csc, }; static const struct intel_color_funcs glk_color_funcs = { @@ -3179,6 +3603,7 @@ static const struct intel_color_funcs glk_color_funcs = { .load_luts = glk_load_luts, .read_luts = glk_read_luts, .lut_equal = glk_lut_equal, + .read_csc = skl_read_csc, }; static const struct intel_color_funcs skl_color_funcs = { @@ -3188,6 +3613,7 @@ static const struct intel_color_funcs skl_color_funcs = { .load_luts = bdw_load_luts, .read_luts = bdw_read_luts, .lut_equal = ivb_lut_equal, + .read_csc = skl_read_csc, }; static const struct intel_color_funcs bdw_color_funcs = { @@ -3197,6 +3623,7 @@ static const struct intel_color_funcs bdw_color_funcs = { .load_luts = bdw_load_luts, .read_luts = bdw_read_luts, .lut_equal = ivb_lut_equal, + .read_csc = ilk_read_csc, }; static const struct intel_color_funcs hsw_color_funcs = { @@ -3206,6 +3633,7 @@ static const struct intel_color_funcs hsw_color_funcs = { .load_luts = ivb_load_luts, .read_luts = ivb_read_luts, .lut_equal = ivb_lut_equal, + .read_csc = ilk_read_csc, }; static const struct intel_color_funcs ivb_color_funcs = { @@ -3215,6 +3643,7 @@ static const struct intel_color_funcs ivb_color_funcs = { .load_luts = ivb_load_luts, .read_luts = ivb_read_luts, .lut_equal = ivb_lut_equal, + .read_csc = ilk_read_csc, }; static const struct intel_color_funcs ilk_color_funcs = { @@ -3224,6 +3653,7 @@ static const struct intel_color_funcs ilk_color_funcs = { .load_luts = ilk_load_luts, .read_luts = ilk_read_luts, .lut_equal = ilk_lut_equal, + .read_csc = ilk_read_csc, }; void intel_color_crtc_init(struct intel_crtc *crtc) @@ -3234,9 +3664,9 @@ void intel_color_crtc_init(struct intel_crtc *crtc) drm_mode_crtc_set_gamma_size(&crtc->base, 256); - gamma_lut_size = INTEL_INFO(i915)->display.color.gamma_lut_size; - degamma_lut_size = INTEL_INFO(i915)->display.color.degamma_lut_size; - has_ctm = degamma_lut_size != 0; + gamma_lut_size = DISPLAY_INFO(i915)->color.gamma_lut_size; + degamma_lut_size = DISPLAY_INFO(i915)->color.degamma_lut_size; + has_ctm = DISPLAY_VER(i915) >= 5; /* * "DPALETTE_A: NOTE: The 8-bit (non-10-bit) mode is the @@ -3260,7 +3690,8 @@ int intel_color_init(struct drm_i915_private *i915) if (DISPLAY_VER(i915) != 10) return 0; - blob = create_linear_lut(i915, INTEL_INFO(i915)->display.color.degamma_lut_size); + blob = create_linear_lut(i915, + DISPLAY_INFO(i915)->color.degamma_lut_size); if (IS_ERR(blob)) return PTR_ERR(blob); @@ -3274,6 +3705,8 @@ void intel_color_init_hooks(struct drm_i915_private *i915) if (HAS_GMCH(i915)) { if (IS_CHERRYVIEW(i915)) i915->display.funcs.color = &chv_color_funcs; + else if (IS_VALLEYVIEW(i915)) + i915->display.funcs.color = &vlv_color_funcs; else if (DISPLAY_VER(i915) >= 4) i915->display.funcs.color = &i965_color_funcs; else diff --git a/drivers/gpu/drm/i915/display/intel_connector.c b/drivers/gpu/drm/i915/display/intel_connector.c index 257afac348397dcaf794480aa0f149cbf4e46f25..ff3bcadebe59b0b7a5af246495b86452581a1cda 100644 --- a/drivers/gpu/drm/i915/display/intel_connector.c +++ b/drivers/gpu/drm/i915/display/intel_connector.c @@ -176,15 +176,15 @@ enum pipe intel_connector_get_pipe(struct intel_connector *connector) /** * intel_connector_update_modes - update connector from edid * @connector: DRM connector device to use - * @edid: previously read EDID information + * @drm_edid: previously read EDID information */ int intel_connector_update_modes(struct drm_connector *connector, - struct edid *edid) + const struct drm_edid *drm_edid) { int ret; - drm_connector_update_edid_property(connector, edid); - ret = drm_add_edid_modes(connector, edid); + drm_edid_connector_update(connector, drm_edid); + ret = drm_edid_connector_add_modes(connector); return ret; } @@ -199,15 +199,15 @@ int intel_connector_update_modes(struct drm_connector *connector, int intel_ddc_get_modes(struct drm_connector *connector, struct i2c_adapter *adapter) { - struct edid *edid; + const struct drm_edid *drm_edid; int ret; - edid = drm_get_edid(connector, adapter); - if (!edid) + drm_edid = drm_edid_read_ddc(connector, adapter); + if (!drm_edid) return 0; - ret = intel_connector_update_modes(connector, edid); - kfree(edid); + ret = intel_connector_update_modes(connector, drm_edid); + drm_edid_free(drm_edid); return ret; } @@ -280,14 +280,14 @@ intel_attach_aspect_ratio_property(struct drm_connector *connector) void intel_attach_hdmi_colorspace_property(struct drm_connector *connector) { - if (!drm_mode_create_hdmi_colorspace_property(connector)) + if (!drm_mode_create_hdmi_colorspace_property(connector, 0)) drm_connector_attach_colorspace_property(connector); } void intel_attach_dp_colorspace_property(struct drm_connector *connector) { - if (!drm_mode_create_dp_colorspace_property(connector)) + if (!drm_mode_create_dp_colorspace_property(connector, 0)) drm_connector_attach_colorspace_property(connector); } diff --git a/drivers/gpu/drm/i915/display/intel_connector.h b/drivers/gpu/drm/i915/display/intel_connector.h index 9d2bc261b2042951933edd75ba5542bece50c682..aaf7281462dcdedc0657a55499a4829aef888037 100644 --- a/drivers/gpu/drm/i915/display/intel_connector.h +++ b/drivers/gpu/drm/i915/display/intel_connector.h @@ -9,7 +9,7 @@ #include struct drm_connector; -struct edid; +struct drm_edid; struct i2c_adapter; struct intel_connector; struct intel_encoder; @@ -25,7 +25,7 @@ void intel_connector_attach_encoder(struct intel_connector *connector, bool intel_connector_get_hw_state(struct intel_connector *connector); enum pipe intel_connector_get_pipe(struct intel_connector *connector); int intel_connector_update_modes(struct drm_connector *connector, - struct edid *edid); + const struct drm_edid *drm_edid); int intel_ddc_get_modes(struct drm_connector *c, struct i2c_adapter *adapter); void intel_attach_force_audio_property(struct drm_connector *connector); void intel_attach_broadcast_rgb_property(struct drm_connector *connector); diff --git a/drivers/gpu/drm/i915/display/intel_crt.c b/drivers/gpu/drm/i915/display/intel_crt.c index 38e9c61c2344ee49a1b32292c949d844a7fb02d8..ab7cd5e60a0a403f19e02a2f90f3d51990b03b94 100644 --- a/drivers/gpu/drm/i915/display/intel_crt.c +++ b/drivers/gpu/drm/i915/display/intel_crt.c @@ -48,6 +48,8 @@ #include "intel_fifo_underrun.h" #include "intel_gmbus.h" #include "intel_hotplug.h" +#include "intel_hotplug_irq.h" +#include "intel_load_detect.h" #include "intel_pch_display.h" #include "intel_pch_refclk.h" @@ -394,6 +396,7 @@ static int intel_crt_compute_config(struct intel_encoder *encoder, if (adjusted_mode->flags & DRM_MODE_FLAG_DBLSCAN) return -EINVAL; + pipe_config->sink_format = INTEL_OUTPUT_FORMAT_RGB; pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB; return 0; @@ -606,37 +609,38 @@ static bool intel_crt_detect_hotplug(struct drm_connector *connector) return ret; } -static struct edid *intel_crt_get_edid(struct drm_connector *connector, - struct i2c_adapter *i2c) +static const struct drm_edid *intel_crt_get_edid(struct drm_connector *connector, + struct i2c_adapter *i2c) { - struct edid *edid; + const struct drm_edid *drm_edid; - edid = drm_get_edid(connector, i2c); + drm_edid = drm_edid_read_ddc(connector, i2c); - if (!edid && !intel_gmbus_is_forced_bit(i2c)) { + if (!drm_edid && !intel_gmbus_is_forced_bit(i2c)) { drm_dbg_kms(connector->dev, "CRT GMBUS EDID read failed, retry using GPIO bit-banging\n"); intel_gmbus_force_bit(i2c, true); - edid = drm_get_edid(connector, i2c); + drm_edid = drm_edid_read_ddc(connector, i2c); intel_gmbus_force_bit(i2c, false); } - return edid; + return drm_edid; } /* local version of intel_ddc_get_modes() to use intel_crt_get_edid() */ static int intel_crt_ddc_get_modes(struct drm_connector *connector, struct i2c_adapter *adapter) { - struct edid *edid; + const struct drm_edid *drm_edid; int ret; - edid = intel_crt_get_edid(connector, adapter); - if (!edid) + drm_edid = intel_crt_get_edid(connector, adapter); + if (!drm_edid) return 0; - ret = intel_connector_update_modes(connector, edid); - kfree(edid); + ret = intel_connector_update_modes(connector, drm_edid); + + drm_edid_free(drm_edid); return ret; } @@ -645,14 +649,15 @@ static bool intel_crt_detect_ddc(struct drm_connector *connector) { struct intel_crt *crt = intel_attached_crt(to_intel_connector(connector)); struct drm_i915_private *dev_priv = to_i915(crt->base.base.dev); - struct edid *edid; + const struct drm_edid *drm_edid; struct i2c_adapter *i2c; bool ret = false; i2c = intel_gmbus_get_adapter(dev_priv, dev_priv->display.vbt.crt_ddc_pin); - edid = intel_crt_get_edid(connector, i2c); + drm_edid = intel_crt_get_edid(connector, i2c); - if (edid) { + if (drm_edid) { + const struct edid *edid = drm_edid_raw(drm_edid); bool is_digital = edid->input & DRM_EDID_INPUT_DIGITAL; /* @@ -673,7 +678,7 @@ static bool intel_crt_detect_ddc(struct drm_connector *connector) "CRT not detected via DDC:0x50 [no valid EDID found]\n"); } - kfree(edid); + drm_edid_free(drm_edid); return ret; } @@ -821,9 +826,9 @@ intel_crt_detect(struct drm_connector *connector, struct drm_i915_private *dev_priv = to_i915(connector->dev); struct intel_crt *crt = intel_attached_crt(to_intel_connector(connector)); struct intel_encoder *intel_encoder = &crt->base; + struct drm_atomic_state *state; intel_wakeref_t wakeref; - int status, ret; - struct intel_load_detect_pipe tmp; + int status; drm_dbg_kms(&dev_priv->drm, "[CONNECTOR:%d:%s] force=%d\n", connector->base.id, connector->name, @@ -881,8 +886,12 @@ intel_crt_detect(struct drm_connector *connector, } /* for pre-945g platforms use load detect */ - ret = intel_get_load_detect_pipe(connector, &tmp, ctx); - if (ret > 0) { + state = intel_load_detect_get_pipe(connector, ctx); + if (IS_ERR(state)) { + status = PTR_ERR(state); + } else if (!state) { + status = connector_status_unknown; + } else { if (intel_crt_detect_ddc(connector)) status = connector_status_connected; else if (DISPLAY_VER(dev_priv) < 4) @@ -892,11 +901,7 @@ intel_crt_detect(struct drm_connector *connector, status = connector_status_disconnected; else status = connector_status_unknown; - intel_release_load_detect_pipe(connector, &tmp, ctx); - } else if (ret == 0) { - status = connector_status_unknown; - } else { - status = ret; + intel_load_detect_release_pipe(connector, state, ctx); } out: diff --git a/drivers/gpu/drm/i915/display/intel_crtc.c b/drivers/gpu/drm/i915/display/intel_crtc.c index ed45a693485480e63c6541f6d492ed5caf7c040c..182c6dd64f47cf0153512a5d3d1b92fedb41c494 100644 --- a/drivers/gpu/drm/i915/display/intel_crtc.c +++ b/drivers/gpu/drm/i915/display/intel_crtc.c @@ -11,7 +11,6 @@ #include #include -#include "i915_irq.h" #include "i915_vgpu.h" #include "i9xx_plane.h" #include "icl_dsi.h" @@ -21,6 +20,7 @@ #include "intel_crtc.h" #include "intel_cursor.h" #include "intel_display_debugfs.h" +#include "intel_display_irq.h" #include "intel_display_trace.h" #include "intel_display_types.h" #include "intel_drrs.h" @@ -35,7 +35,11 @@ static void assert_vblank_disabled(struct drm_crtc *crtc) { - if (I915_STATE_WARN_ON(drm_crtc_vblank_get(crtc) == 0)) + struct drm_i915_private *i915 = to_i915(crtc->dev); + + if (I915_STATE_WARN(i915, drm_crtc_vblank_get(crtc) == 0, + "[CRTC:%d:%s] vblank assertion failure (expected off, current on)\n", + crtc->base.id, crtc->name)) drm_crtc_vblank_put(crtc); } @@ -302,7 +306,7 @@ int intel_crtc_init(struct drm_i915_private *dev_priv, enum pipe pipe) return PTR_ERR(crtc); crtc->pipe = pipe; - crtc->num_scalers = RUNTIME_INFO(dev_priv)->num_scalers[pipe]; + crtc->num_scalers = DISPLAY_RUNTIME_INFO(dev_priv)->num_scalers[pipe]; if (DISPLAY_VER(dev_priv) >= 9) primary = skl_universal_plane_create(dev_priv, pipe, @@ -510,6 +514,13 @@ void intel_pipe_update_start(struct intel_crtc_state *new_crtc_state) VBLANK_EVASION_TIME_US); max = vblank_start - 1; + /* + * M/N is double buffered on the transcoder's undelayed vblank, + * so with seamless M/N we must evade both vblanks. + */ + if (new_crtc_state->seamless_m_n && intel_crtc_needs_fastset(new_crtc_state)) + min -= adjusted_mode->crtc_vblank_start - adjusted_mode->crtc_vdisplay; + if (min <= 0 || max <= 0) goto irq_disable; @@ -692,7 +703,8 @@ void intel_pipe_update_end(struct intel_crtc_state *new_crtc_state) * FIXME Should be synchronized with the start of vblank somehow... */ if (new_crtc_state->seamless_m_n && intel_crtc_needs_fastset(new_crtc_state)) - intel_crtc_update_active_timings(new_crtc_state); + intel_crtc_update_active_timings(new_crtc_state, + new_crtc_state->vrr.enable); local_irq_enable(); diff --git a/drivers/gpu/drm/i915/display/intel_crtc.h b/drivers/gpu/drm/i915/display/intel_crtc.h index 73077137fb990ee2a689b1157432cff1b30552ba..51a4c8df9e6574cd55dfcd575d78a173d3012312 100644 --- a/drivers/gpu/drm/i915/display/intel_crtc.h +++ b/drivers/gpu/drm/i915/display/intel_crtc.h @@ -16,6 +16,16 @@ struct intel_atomic_state; struct intel_crtc; struct intel_crtc_state; +/* + * FIXME: We should instead only take spinlocks once for the entire update + * instead of once per mmio. + */ +#if IS_ENABLED(CONFIG_PROVE_LOCKING) +#define VBLANK_EVASION_TIME_US 250 +#else +#define VBLANK_EVASION_TIME_US 100 +#endif + int intel_usecs_to_scanlines(const struct drm_display_mode *adjusted_mode, int usecs); u32 intel_crtc_max_vblank_count(const struct intel_crtc_state *crtc_state); diff --git a/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c b/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c index 54c8adc0702eeba1dd9ecc417c43ef8224bea8e7..8d4640d0fd346b51fdb599ff79759a7ceebdf49c 100644 --- a/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c +++ b/drivers/gpu/drm/i915/display/intel_crtc_state_dump.c @@ -158,6 +158,45 @@ static void intel_dump_plane_state(const struct intel_plane_state *plane_state) DRM_RECT_ARG(&plane_state->uapi.dst)); } +static void +ilk_dump_csc(struct drm_i915_private *i915, const char *name, + const struct intel_csc_matrix *csc) +{ + int i; + + drm_dbg_kms(&i915->drm, + "%s: pre offsets: 0x%04x 0x%04x 0x%04x\n", name, + csc->preoff[0], csc->preoff[1], csc->preoff[2]); + + for (i = 0; i < 3; i++) + drm_dbg_kms(&i915->drm, + "%s: coefficients: 0x%04x 0x%04x 0x%04x\n", name, + csc->coeff[3 * i + 0], + csc->coeff[3 * i + 1], + csc->coeff[3 * i + 2]); + + if (DISPLAY_VER(i915) < 7) + return; + + drm_dbg_kms(&i915->drm, + "%s: post offsets: 0x%04x 0x%04x 0x%04x\n", name, + csc->postoff[0], csc->postoff[1], csc->postoff[2]); +} + +static void +vlv_dump_csc(struct drm_i915_private *i915, const char *name, + const struct intel_csc_matrix *csc) +{ + int i; + + for (i = 0; i < 3; i++) + drm_dbg_kms(&i915->drm, + "%s: coefficients: 0x%04x 0x%04x 0x%04x\n", name, + csc->coeff[3 * i + 0], + csc->coeff[3 * i + 1], + csc->coeff[3 * i + 2]); +} + void intel_crtc_state_dump(const struct intel_crtc_state *pipe_config, struct intel_atomic_state *state, const char *context) @@ -178,10 +217,11 @@ void intel_crtc_state_dump(const struct intel_crtc_state *pipe_config, snprintf_output_types(buf, sizeof(buf), pipe_config->output_types); drm_dbg_kms(&i915->drm, - "active: %s, output_types: %s (0x%x), output format: %s\n", + "active: %s, output_types: %s (0x%x), output format: %s, sink format: %s\n", str_yes_no(pipe_config->hw.active), buf, pipe_config->output_types, - intel_output_format_name(pipe_config->output_format)); + intel_output_format_name(pipe_config->output_format), + intel_output_format_name(pipe_config->sink_format)); drm_dbg_kms(&i915->drm, "cpu_transcoder: %s, pipe bpp: %i, dithering: %i\n", @@ -325,6 +365,16 @@ void intel_crtc_state_dump(const struct intel_crtc_state *pipe_config, pipe_config->post_csc_lut ? drm_color_lut_size(pipe_config->post_csc_lut) : 0); + if (DISPLAY_VER(i915) >= 11) + ilk_dump_csc(i915, "output csc", &pipe_config->output_csc); + + if (!HAS_GMCH(i915)) + ilk_dump_csc(i915, "pipe csc", &pipe_config->csc); + else if (IS_CHERRYVIEW(i915)) + vlv_dump_csc(i915, "cgm csc", &pipe_config->csc); + else if (IS_VALLEYVIEW(i915)) + vlv_dump_csc(i915, "wgc csc", &pipe_config->csc); + dump_planes: if (!state) return; diff --git a/drivers/gpu/drm/i915/display/intel_cursor.c b/drivers/gpu/drm/i915/display/intel_cursor.c index 31bef04273779e0aa9b5811f8e0569227a6c9972..b342fad180ca5be59e8edfee5fd0f45b9e68c6ec 100644 --- a/drivers/gpu/drm/i915/display/intel_cursor.c +++ b/drivers/gpu/drm/i915/display/intel_cursor.c @@ -36,7 +36,7 @@ static u32 intel_cursor_base(const struct intel_plane_state *plane_state) const struct drm_i915_gem_object *obj = intel_fb_obj(fb); u32 base; - if (INTEL_INFO(dev_priv)->display.cursor_needs_physical) + if (DISPLAY_INFO(dev_priv)->cursor_needs_physical) base = sg_dma_address(obj->mm.pages->sgl); else base = intel_plane_ggtt_offset(plane_state); @@ -814,7 +814,7 @@ intel_cursor_plane_create(struct drm_i915_private *dev_priv, DRM_MODE_ROTATE_0 | DRM_MODE_ROTATE_180); - zpos = RUNTIME_INFO(dev_priv)->num_sprites[pipe] + 1; + zpos = DISPLAY_RUNTIME_INFO(dev_priv)->num_sprites[pipe] + 1; drm_plane_create_zpos_immutable_property(&cursor->base, zpos); if (DISPLAY_VER(dev_priv) >= 12) diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c new file mode 100644 index 0000000000000000000000000000000000000000..0600fdcd06ef2a4bd76deec7869e379a302c22cc --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c @@ -0,0 +1,3046 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2023 Intel Corporation + */ + +#include +#include +#include "i915_reg.h" +#include "intel_cx0_phy.h" +#include "intel_cx0_phy_regs.h" +#include "intel_ddi.h" +#include "intel_ddi_buf_trans.h" +#include "intel_de.h" +#include "intel_display_types.h" +#include "intel_dp.h" +#include "intel_hdmi.h" +#include "intel_panel.h" +#include "intel_psr.h" +#include "intel_tc.h" + +#define MB_WRITE_COMMITTED true +#define MB_WRITE_UNCOMMITTED false + +#define for_each_cx0_lane_in_mask(__lane_mask, __lane) \ + for ((__lane) = 0; (__lane) < 2; (__lane)++) \ + for_each_if((__lane_mask) & BIT(__lane)) + +#define INTEL_CX0_LANE0 BIT(0) +#define INTEL_CX0_LANE1 BIT(1) +#define INTEL_CX0_BOTH_LANES (INTEL_CX0_LANE1 | INTEL_CX0_LANE0) + +bool intel_is_c10phy(struct drm_i915_private *i915, enum phy phy) +{ + if (IS_METEORLAKE(i915) && (phy < PHY_C)) + return true; + + return false; +} + +static int lane_mask_to_lane(u8 lane_mask) +{ + if (WARN_ON((lane_mask & ~INTEL_CX0_BOTH_LANES) || + hweight8(lane_mask) != 1)) + return 0; + + return ilog2(lane_mask); +} + +static void +assert_dc_off(struct drm_i915_private *i915) +{ + bool enabled; + + enabled = intel_display_power_is_enabled(i915, POWER_DOMAIN_DC_OFF); + drm_WARN_ON(&i915->drm, !enabled); +} + +/* + * Prepare HW for CX0 phy transactions. + * + * It is required that PSR and DC5/6 are disabled before any CX0 message + * bus transaction is executed. + */ +static intel_wakeref_t intel_cx0_phy_transaction_begin(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + struct intel_dp *intel_dp = enc_to_intel_dp(encoder); + + intel_psr_pause(intel_dp); + return intel_display_power_get(i915, POWER_DOMAIN_DC_OFF); +} + +static void intel_cx0_phy_transaction_end(struct intel_encoder *encoder, intel_wakeref_t wakeref) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + struct intel_dp *intel_dp = enc_to_intel_dp(encoder); + + intel_psr_resume(intel_dp); + intel_display_power_put(i915, POWER_DOMAIN_DC_OFF, wakeref); +} + +static void intel_clear_response_ready_flag(struct drm_i915_private *i915, + enum port port, int lane) +{ + intel_de_rmw(i915, XELPDP_PORT_P2M_MSGBUS_STATUS(port, lane), + 0, XELPDP_PORT_P2M_RESPONSE_READY | XELPDP_PORT_P2M_ERROR_SET); +} + +static void intel_cx0_bus_reset(struct drm_i915_private *i915, enum port port, int lane) +{ + enum phy phy = intel_port_to_phy(i915, port); + + intel_de_write(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane), + XELPDP_PORT_M2P_TRANSACTION_RESET); + + if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane), + XELPDP_PORT_M2P_TRANSACTION_RESET, + XELPDP_MSGBUS_TIMEOUT_SLOW)) { + drm_err_once(&i915->drm, "Failed to bring PHY %c to idle.\n", phy_name(phy)); + return; + } + + intel_clear_response_ready_flag(i915, port, lane); +} + +static int intel_cx0_wait_for_ack(struct drm_i915_private *i915, enum port port, + int command, int lane, u32 *val) +{ + enum phy phy = intel_port_to_phy(i915, port); + + if (__intel_de_wait_for_register(i915, + XELPDP_PORT_P2M_MSGBUS_STATUS(port, lane), + XELPDP_PORT_P2M_RESPONSE_READY, + XELPDP_PORT_P2M_RESPONSE_READY, + XELPDP_MSGBUS_TIMEOUT_FAST_US, + XELPDP_MSGBUS_TIMEOUT_SLOW, val)) { + drm_dbg_kms(&i915->drm, "PHY %c Timeout waiting for message ACK. Status: 0x%x\n", + phy_name(phy), *val); + return -ETIMEDOUT; + } + + if (*val & XELPDP_PORT_P2M_ERROR_SET) { + drm_dbg_kms(&i915->drm, "PHY %c Error occurred during %s command. Status: 0x%x\n", phy_name(phy), + command == XELPDP_PORT_P2M_COMMAND_READ_ACK ? "read" : "write", *val); + intel_cx0_bus_reset(i915, port, lane); + return -EINVAL; + } + + if (REG_FIELD_GET(XELPDP_PORT_P2M_COMMAND_TYPE_MASK, *val) != command) { + drm_dbg_kms(&i915->drm, "PHY %c Not a %s response. MSGBUS Status: 0x%x.\n", phy_name(phy), + command == XELPDP_PORT_P2M_COMMAND_READ_ACK ? "read" : "write", *val); + intel_cx0_bus_reset(i915, port, lane); + return -EINVAL; + } + + return 0; +} + +static int __intel_cx0_read_once(struct drm_i915_private *i915, enum port port, + int lane, u16 addr) +{ + enum phy phy = intel_port_to_phy(i915, port); + int ack; + u32 val; + + if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane), + XELPDP_PORT_M2P_TRANSACTION_PENDING, + XELPDP_MSGBUS_TIMEOUT_SLOW)) { + drm_dbg_kms(&i915->drm, + "PHY %c Timeout waiting for previous transaction to complete. Reset the bus and retry.\n", phy_name(phy)); + intel_cx0_bus_reset(i915, port, lane); + return -ETIMEDOUT; + } + + intel_de_write(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane), + XELPDP_PORT_M2P_TRANSACTION_PENDING | + XELPDP_PORT_M2P_COMMAND_READ | + XELPDP_PORT_M2P_ADDRESS(addr)); + + ack = intel_cx0_wait_for_ack(i915, port, XELPDP_PORT_P2M_COMMAND_READ_ACK, lane, &val); + if (ack < 0) { + intel_cx0_bus_reset(i915, port, lane); + return ack; + } + + intel_clear_response_ready_flag(i915, port, lane); + + return REG_FIELD_GET(XELPDP_PORT_P2M_DATA_MASK, val); +} + +static u8 __intel_cx0_read(struct drm_i915_private *i915, enum port port, + int lane, u16 addr) +{ + enum phy phy = intel_port_to_phy(i915, port); + int i, status; + + assert_dc_off(i915); + + /* 3 tries is assumed to be enough to read successfully */ + for (i = 0; i < 3; i++) { + status = __intel_cx0_read_once(i915, port, lane, addr); + + if (status >= 0) + return status; + } + + drm_err_once(&i915->drm, "PHY %c Read %04x failed after %d retries.\n", + phy_name(phy), addr, i); + + return 0; +} + +static u8 intel_cx0_read(struct drm_i915_private *i915, enum port port, + u8 lane_mask, u16 addr) +{ + int lane = lane_mask_to_lane(lane_mask); + + return __intel_cx0_read(i915, port, lane, addr); +} + +static int __intel_cx0_write_once(struct drm_i915_private *i915, enum port port, + int lane, u16 addr, u8 data, bool committed) +{ + enum phy phy = intel_port_to_phy(i915, port); + u32 val; + + if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane), + XELPDP_PORT_M2P_TRANSACTION_PENDING, + XELPDP_MSGBUS_TIMEOUT_SLOW)) { + drm_dbg_kms(&i915->drm, + "PHY %c Timeout waiting for previous transaction to complete. Resetting the bus.\n", phy_name(phy)); + intel_cx0_bus_reset(i915, port, lane); + return -ETIMEDOUT; + } + + intel_de_write(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane), + XELPDP_PORT_M2P_TRANSACTION_PENDING | + (committed ? XELPDP_PORT_M2P_COMMAND_WRITE_COMMITTED : + XELPDP_PORT_M2P_COMMAND_WRITE_UNCOMMITTED) | + XELPDP_PORT_M2P_DATA(data) | + XELPDP_PORT_M2P_ADDRESS(addr)); + + if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane), + XELPDP_PORT_M2P_TRANSACTION_PENDING, + XELPDP_MSGBUS_TIMEOUT_SLOW)) { + drm_dbg_kms(&i915->drm, + "PHY %c Timeout waiting for write to complete. Resetting the bus.\n", phy_name(phy)); + intel_cx0_bus_reset(i915, port, lane); + return -ETIMEDOUT; + } + + if (committed) { + if (intel_cx0_wait_for_ack(i915, port, XELPDP_PORT_P2M_COMMAND_WRITE_ACK, lane, &val) < 0) { + intel_cx0_bus_reset(i915, port, lane); + return -EINVAL; + } + } else if ((intel_de_read(i915, XELPDP_PORT_P2M_MSGBUS_STATUS(port, lane)) & + XELPDP_PORT_P2M_ERROR_SET)) { + drm_dbg_kms(&i915->drm, + "PHY %c Error occurred during write command.\n", phy_name(phy)); + intel_cx0_bus_reset(i915, port, lane); + return -EINVAL; + } + + intel_clear_response_ready_flag(i915, port, lane); + + return 0; +} + +static void __intel_cx0_write(struct drm_i915_private *i915, enum port port, + int lane, u16 addr, u8 data, bool committed) +{ + enum phy phy = intel_port_to_phy(i915, port); + int i, status; + + assert_dc_off(i915); + + /* 3 tries is assumed to be enough to write successfully */ + for (i = 0; i < 3; i++) { + status = __intel_cx0_write_once(i915, port, lane, addr, data, committed); + + if (status == 0) + return; + } + + drm_err_once(&i915->drm, + "PHY %c Write %04x failed after %d retries.\n", phy_name(phy), addr, i); +} + +static void intel_cx0_write(struct drm_i915_private *i915, enum port port, + u8 lane_mask, u16 addr, u8 data, bool committed) +{ + int lane; + + for_each_cx0_lane_in_mask(lane_mask, lane) + __intel_cx0_write(i915, port, lane, addr, data, committed); +} + +static void intel_c20_sram_write(struct drm_i915_private *i915, enum port port, + int lane, u16 addr, u16 data) +{ + assert_dc_off(i915); + + intel_cx0_write(i915, port, lane, PHY_C20_WR_ADDRESS_H, addr >> 8, 0); + intel_cx0_write(i915, port, lane, PHY_C20_WR_ADDRESS_L, addr & 0xff, 0); + + intel_cx0_write(i915, port, lane, PHY_C20_WR_DATA_H, data >> 8, 0); + intel_cx0_write(i915, port, lane, PHY_C20_WR_DATA_L, data & 0xff, 1); +} + +static u16 intel_c20_sram_read(struct drm_i915_private *i915, enum port port, + int lane, u16 addr) +{ + u16 val; + + assert_dc_off(i915); + + intel_cx0_write(i915, port, lane, PHY_C20_RD_ADDRESS_H, addr >> 8, 0); + intel_cx0_write(i915, port, lane, PHY_C20_RD_ADDRESS_L, addr & 0xff, 1); + + val = intel_cx0_read(i915, port, lane, PHY_C20_RD_DATA_H); + val <<= 8; + val |= intel_cx0_read(i915, port, lane, PHY_C20_RD_DATA_L); + + return val; +} + +static void __intel_cx0_rmw(struct drm_i915_private *i915, enum port port, + int lane, u16 addr, u8 clear, u8 set, bool committed) +{ + u8 old, val; + + old = __intel_cx0_read(i915, port, lane, addr); + val = (old & ~clear) | set; + + if (val != old) + __intel_cx0_write(i915, port, lane, addr, val, committed); +} + +static void intel_cx0_rmw(struct drm_i915_private *i915, enum port port, + u8 lane_mask, u16 addr, u8 clear, u8 set, bool committed) +{ + u8 lane; + + for_each_cx0_lane_in_mask(lane_mask, lane) + __intel_cx0_rmw(i915, port, lane, addr, clear, set, committed); +} + +static u8 intel_c10_get_tx_vboost_lvl(const struct intel_crtc_state *crtc_state) +{ + if (intel_crtc_has_dp_encoder(crtc_state)) { + if (!intel_crtc_has_type(crtc_state, INTEL_OUTPUT_EDP) && + (crtc_state->port_clock == 540000 || + crtc_state->port_clock == 810000)) + return 5; + else + return 4; + } else { + return 5; + } +} + +static u8 intel_c10_get_tx_term_ctl(const struct intel_crtc_state *crtc_state) +{ + if (intel_crtc_has_dp_encoder(crtc_state)) { + if (!intel_crtc_has_type(crtc_state, INTEL_OUTPUT_EDP) && + (crtc_state->port_clock == 540000 || + crtc_state->port_clock == 810000)) + return 5; + else + return 2; + } else { + return 6; + } +} + +void intel_cx0_phy_set_signal_levels(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + const struct intel_ddi_buf_trans *trans; + enum phy phy = intel_port_to_phy(i915, encoder->port); + intel_wakeref_t wakeref; + int n_entries, ln; + + wakeref = intel_cx0_phy_transaction_begin(encoder); + + trans = encoder->get_buf_trans(encoder, crtc_state, &n_entries); + if (drm_WARN_ON_ONCE(&i915->drm, !trans)) { + intel_cx0_phy_transaction_end(encoder, wakeref); + return; + } + + if (intel_is_c10phy(i915, phy)) { + intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CONTROL(1), + 0, C10_VDR_CTRL_MSGBUS_ACCESS, MB_WRITE_COMMITTED); + intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CMN(3), + C10_CMN3_TXVBOOST_MASK, + C10_CMN3_TXVBOOST(intel_c10_get_tx_vboost_lvl(crtc_state)), + MB_WRITE_UNCOMMITTED); + intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_TX(1), + C10_TX1_TERMCTL_MASK, + C10_TX1_TERMCTL(intel_c10_get_tx_term_ctl(crtc_state)), + MB_WRITE_COMMITTED); + } + + for (ln = 0; ln < crtc_state->lane_count; ln++) { + int level = intel_ddi_level(encoder, crtc_state, ln); + int lane, tx; + + lane = ln / 2; + tx = ln % 2; + + intel_cx0_rmw(i915, encoder->port, BIT(lane), PHY_CX0_VDROVRD_CTL(lane, tx, 0), + C10_PHY_OVRD_LEVEL_MASK, + C10_PHY_OVRD_LEVEL(trans->entries[level].snps.pre_cursor), + MB_WRITE_COMMITTED); + intel_cx0_rmw(i915, encoder->port, BIT(lane), PHY_CX0_VDROVRD_CTL(lane, tx, 1), + C10_PHY_OVRD_LEVEL_MASK, + C10_PHY_OVRD_LEVEL(trans->entries[level].snps.vswing), + MB_WRITE_COMMITTED); + intel_cx0_rmw(i915, encoder->port, BIT(lane), PHY_CX0_VDROVRD_CTL(lane, tx, 2), + C10_PHY_OVRD_LEVEL_MASK, + C10_PHY_OVRD_LEVEL(trans->entries[level].snps.post_cursor), + MB_WRITE_COMMITTED); + } + + /* Write Override enables in 0xD71 */ + intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_OVRD, + 0, PHY_C10_VDR_OVRD_TX1 | PHY_C10_VDR_OVRD_TX2, + MB_WRITE_COMMITTED); + + if (intel_is_c10phy(i915, phy)) + intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CONTROL(1), + 0, C10_VDR_CTRL_UPDATE_CFG, MB_WRITE_COMMITTED); + + intel_cx0_phy_transaction_end(encoder, wakeref); +} + +/* + * Basic DP link rates with 38.4 MHz reference clock. + * Note: The tables below are with SSC. In non-ssc + * registers 0xC04 to 0xC08(pll[4] to pll[8]) will be + * programmed 0. + */ + +static const struct intel_c10pll_state mtl_c10_dp_rbr = { + .clock = 162000, + .tx = 0x10, + .cmn = 0x21, + .pll[0] = 0xB4, + .pll[1] = 0, + .pll[2] = 0x30, + .pll[3] = 0x1, + .pll[4] = 0x26, + .pll[5] = 0x0C, + .pll[6] = 0x98, + .pll[7] = 0x46, + .pll[8] = 0x1, + .pll[9] = 0x1, + .pll[10] = 0, + .pll[11] = 0, + .pll[12] = 0xC0, + .pll[13] = 0, + .pll[14] = 0, + .pll[15] = 0x2, + .pll[16] = 0x84, + .pll[17] = 0x4F, + .pll[18] = 0xE5, + .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_edp_r216 = { + .clock = 216000, + .tx = 0x10, + .cmn = 0x21, + .pll[0] = 0x4, + .pll[1] = 0, + .pll[2] = 0xA2, + .pll[3] = 0x1, + .pll[4] = 0x33, + .pll[5] = 0x10, + .pll[6] = 0x75, + .pll[7] = 0xB3, + .pll[8] = 0x1, + .pll[9] = 0x1, + .pll[10] = 0, + .pll[11] = 0, + .pll[12] = 0, + .pll[13] = 0, + .pll[14] = 0, + .pll[15] = 0x2, + .pll[16] = 0x85, + .pll[17] = 0x0F, + .pll[18] = 0xE6, + .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_edp_r243 = { + .clock = 243000, + .tx = 0x10, + .cmn = 0x21, + .pll[0] = 0x34, + .pll[1] = 0, + .pll[2] = 0xDA, + .pll[3] = 0x1, + .pll[4] = 0x39, + .pll[5] = 0x12, + .pll[6] = 0xE3, + .pll[7] = 0xE9, + .pll[8] = 0x1, + .pll[9] = 0x1, + .pll[10] = 0, + .pll[11] = 0, + .pll[12] = 0x20, + .pll[13] = 0, + .pll[14] = 0, + .pll[15] = 0x2, + .pll[16] = 0x85, + .pll[17] = 0x8F, + .pll[18] = 0xE6, + .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_dp_hbr1 = { + .clock = 270000, + .tx = 0x10, + .cmn = 0x21, + .pll[0] = 0xF4, + .pll[1] = 0, + .pll[2] = 0xF8, + .pll[3] = 0x0, + .pll[4] = 0x20, + .pll[5] = 0x0A, + .pll[6] = 0x29, + .pll[7] = 0x10, + .pll[8] = 0x1, /* Verify */ + .pll[9] = 0x1, + .pll[10] = 0, + .pll[11] = 0, + .pll[12] = 0xA0, + .pll[13] = 0, + .pll[14] = 0, + .pll[15] = 0x1, + .pll[16] = 0x84, + .pll[17] = 0x4F, + .pll[18] = 0xE5, + .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_edp_r324 = { + .clock = 324000, + .tx = 0x10, + .cmn = 0x21, + .pll[0] = 0xB4, + .pll[1] = 0, + .pll[2] = 0x30, + .pll[3] = 0x1, + .pll[4] = 0x26, + .pll[5] = 0x0C, + .pll[6] = 0x98, + .pll[7] = 0x46, + .pll[8] = 0x1, + .pll[9] = 0x1, + .pll[10] = 0, + .pll[11] = 0, + .pll[12] = 0xC0, + .pll[13] = 0, + .pll[14] = 0, + .pll[15] = 0x1, + .pll[16] = 0x85, + .pll[17] = 0x4F, + .pll[18] = 0xE6, + .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_edp_r432 = { + .clock = 432000, + .tx = 0x10, + .cmn = 0x21, + .pll[0] = 0x4, + .pll[1] = 0, + .pll[2] = 0xA2, + .pll[3] = 0x1, + .pll[4] = 0x33, + .pll[5] = 0x10, + .pll[6] = 0x75, + .pll[7] = 0xB3, + .pll[8] = 0x1, + .pll[9] = 0x1, + .pll[10] = 0, + .pll[11] = 0, + .pll[12] = 0, + .pll[13] = 0, + .pll[14] = 0, + .pll[15] = 0x1, + .pll[16] = 0x85, + .pll[17] = 0x0F, + .pll[18] = 0xE6, + .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_dp_hbr2 = { + .clock = 540000, + .tx = 0x10, + .cmn = 0x21, + .pll[0] = 0xF4, + .pll[1] = 0, + .pll[2] = 0xF8, + .pll[3] = 0, + .pll[4] = 0x20, + .pll[5] = 0x0A, + .pll[6] = 0x29, + .pll[7] = 0x10, + .pll[8] = 0x1, + .pll[9] = 0x1, + .pll[10] = 0, + .pll[11] = 0, + .pll[12] = 0xA0, + .pll[13] = 0, + .pll[14] = 0, + .pll[15] = 0, + .pll[16] = 0x84, + .pll[17] = 0x4F, + .pll[18] = 0xE5, + .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_edp_r675 = { + .clock = 675000, + .tx = 0x10, + .cmn = 0x21, + .pll[0] = 0xB4, + .pll[1] = 0, + .pll[2] = 0x3E, + .pll[3] = 0x1, + .pll[4] = 0xA8, + .pll[5] = 0x0C, + .pll[6] = 0x33, + .pll[7] = 0x54, + .pll[8] = 0x1, + .pll[9] = 0x1, + .pll[10] = 0, + .pll[11] = 0, + .pll[12] = 0xC8, + .pll[13] = 0, + .pll[14] = 0, + .pll[15] = 0, + .pll[16] = 0x85, + .pll[17] = 0x8F, + .pll[18] = 0xE6, + .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_dp_hbr3 = { + .clock = 810000, + .tx = 0x10, + .cmn = 0x21, + .pll[0] = 0x34, + .pll[1] = 0, + .pll[2] = 0x84, + .pll[3] = 0x1, + .pll[4] = 0x30, + .pll[5] = 0x0F, + .pll[6] = 0x3D, + .pll[7] = 0x98, + .pll[8] = 0x1, + .pll[9] = 0x1, + .pll[10] = 0, + .pll[11] = 0, + .pll[12] = 0xF0, + .pll[13] = 0, + .pll[14] = 0, + .pll[15] = 0, + .pll[16] = 0x84, + .pll[17] = 0x0F, + .pll[18] = 0xE5, + .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state * const mtl_c10_dp_tables[] = { + &mtl_c10_dp_rbr, + &mtl_c10_dp_hbr1, + &mtl_c10_dp_hbr2, + &mtl_c10_dp_hbr3, + NULL, +}; + +static const struct intel_c10pll_state * const mtl_c10_edp_tables[] = { + &mtl_c10_dp_rbr, + &mtl_c10_edp_r216, + &mtl_c10_edp_r243, + &mtl_c10_dp_hbr1, + &mtl_c10_edp_r324, + &mtl_c10_edp_r432, + &mtl_c10_dp_hbr2, + &mtl_c10_edp_r675, + &mtl_c10_dp_hbr3, + NULL, +}; + +/* C20 basic DP 1.4 tables */ +static const struct intel_c20pll_state mtl_c20_dp_rbr = { + .link_bit_rate = 162000, + .clock = 162000, + .tx = { 0xbe88, /* tx cfg0 */ + 0x5800, /* tx cfg1 */ + 0x0000, /* tx cfg2 */ + }, + .cmn = {0x0500, /* cmn cfg0*/ + 0x0005, /* cmn cfg1 */ + 0x0000, /* cmn cfg2 */ + 0x0000, /* cmn cfg3 */ + }, + .mpllb = { 0x50a8, /* mpllb cfg0 */ + 0x2120, /* mpllb cfg1 */ + 0xcd9a, /* mpllb cfg2 */ + 0xbfc1, /* mpllb cfg3 */ + 0x5ab8, /* mpllb cfg4 */ + 0x4c34, /* mpllb cfg5 */ + 0x2000, /* mpllb cfg6 */ + 0x0001, /* mpllb cfg7 */ + 0x6000, /* mpllb cfg8 */ + 0x0000, /* mpllb cfg9 */ + 0x0000, /* mpllb cfg10 */ + }, +}; + +static const struct intel_c20pll_state mtl_c20_dp_hbr1 = { + .link_bit_rate = 270000, + .clock = 270000, + .tx = { 0xbe88, /* tx cfg0 */ + 0x4800, /* tx cfg1 */ + 0x0000, /* tx cfg2 */ + }, + .cmn = {0x0500, /* cmn cfg0*/ + 0x0005, /* cmn cfg1 */ + 0x0000, /* cmn cfg2 */ + 0x0000, /* cmn cfg3 */ + }, + .mpllb = { 0x308c, /* mpllb cfg0 */ + 0x2110, /* mpllb cfg1 */ + 0xcc9c, /* mpllb cfg2 */ + 0xbfc1, /* mpllb cfg3 */ + 0x4b9a, /* mpllb cfg4 */ + 0x3f81, /* mpllb cfg5 */ + 0x2000, /* mpllb cfg6 */ + 0x0001, /* mpllb cfg7 */ + 0x5000, /* mpllb cfg8 */ + 0x0000, /* mpllb cfg9 */ + 0x0000, /* mpllb cfg10 */ + }, +}; + +static const struct intel_c20pll_state mtl_c20_dp_hbr2 = { + .link_bit_rate = 540000, + .clock = 540000, + .tx = { 0xbe88, /* tx cfg0 */ + 0x4800, /* tx cfg1 */ + 0x0000, /* tx cfg2 */ + }, + .cmn = {0x0500, /* cmn cfg0*/ + 0x0005, /* cmn cfg1 */ + 0x0000, /* cmn cfg2 */ + 0x0000, /* cmn cfg3 */ + }, + .mpllb = { 0x108c, /* mpllb cfg0 */ + 0x2108, /* mpllb cfg1 */ + 0xcc9c, /* mpllb cfg2 */ + 0xbfc1, /* mpllb cfg3 */ + 0x4b9a, /* mpllb cfg4 */ + 0x3f81, /* mpllb cfg5 */ + 0x2000, /* mpllb cfg6 */ + 0x0001, /* mpllb cfg7 */ + 0x5000, /* mpllb cfg8 */ + 0x0000, /* mpllb cfg9 */ + 0x0000, /* mpllb cfg10 */ + }, +}; + +static const struct intel_c20pll_state mtl_c20_dp_hbr3 = { + .link_bit_rate = 810000, + .clock = 810000, + .tx = { 0xbe88, /* tx cfg0 */ + 0x4800, /* tx cfg1 */ + 0x0000, /* tx cfg2 */ + }, + .cmn = {0x0500, /* cmn cfg0*/ + 0x0005, /* cmn cfg1 */ + 0x0000, /* cmn cfg2 */ + 0x0000, /* cmn cfg3 */ + }, + .mpllb = { 0x10d2, /* mpllb cfg0 */ + 0x2108, /* mpllb cfg1 */ + 0x8d98, /* mpllb cfg2 */ + 0xbfc1, /* mpllb cfg3 */ + 0x7166, /* mpllb cfg4 */ + 0x5f42, /* mpllb cfg5 */ + 0x2000, /* mpllb cfg6 */ + 0x0001, /* mpllb cfg7 */ + 0x7800, /* mpllb cfg8 */ + 0x0000, /* mpllb cfg9 */ + 0x0000, /* mpllb cfg10 */ + }, +}; + +/* C20 basic DP 2.0 tables */ +static const struct intel_c20pll_state mtl_c20_dp_uhbr10 = { + .link_bit_rate = 1000000, /* 10 Gbps */ + .clock = 312500, + .tx = { 0xbe21, /* tx cfg0 */ + 0x4800, /* tx cfg1 */ + 0x0000, /* tx cfg2 */ + }, + .cmn = {0x0500, /* cmn cfg0*/ + 0x0005, /* cmn cfg1 */ + 0x0000, /* cmn cfg2 */ + 0x0000, /* cmn cfg3 */ + }, + .mplla = { 0x3104, /* mplla cfg0 */ + 0xd105, /* mplla cfg1 */ + 0xc025, /* mplla cfg2 */ + 0xc025, /* mplla cfg3 */ + 0x8c00, /* mplla cfg4 */ + 0x759a, /* mplla cfg5 */ + 0x4000, /* mplla cfg6 */ + 0x0003, /* mplla cfg7 */ + 0x3555, /* mplla cfg8 */ + 0x0001, /* mplla cfg9 */ + }, +}; + +static const struct intel_c20pll_state mtl_c20_dp_uhbr13_5 = { + .link_bit_rate = 1350000, /* 13.5 Gbps */ + .clock = 421875, + .tx = { 0xbea0, /* tx cfg0 */ + 0x4800, /* tx cfg1 */ + 0x0000, /* tx cfg2 */ + }, + .cmn = {0x0500, /* cmn cfg0*/ + 0x0005, /* cmn cfg1 */ + 0x0000, /* cmn cfg2 */ + 0x0000, /* cmn cfg3 */ + }, + .mpllb = { 0x015f, /* mpllb cfg0 */ + 0x2205, /* mpllb cfg1 */ + 0x1b17, /* mpllb cfg2 */ + 0xffc1, /* mpllb cfg3 */ + 0xe100, /* mpllb cfg4 */ + 0xbd00, /* mpllb cfg5 */ + 0x2000, /* mpllb cfg6 */ + 0x0001, /* mpllb cfg7 */ + 0x4800, /* mpllb cfg8 */ + 0x0000, /* mpllb cfg9 */ + 0x0000, /* mpllb cfg10 */ + }, +}; + +static const struct intel_c20pll_state mtl_c20_dp_uhbr20 = { + .link_bit_rate = 2000000, /* 20 Gbps */ + .clock = 625000, + .tx = { 0xbe20, /* tx cfg0 */ + 0x4800, /* tx cfg1 */ + 0x0000, /* tx cfg2 */ + }, + .cmn = {0x0500, /* cmn cfg0*/ + 0x0005, /* cmn cfg1 */ + 0x0000, /* cmn cfg2 */ + 0x0000, /* cmn cfg3 */ + }, + .mplla = { 0x3104, /* mplla cfg0 */ + 0xd105, /* mplla cfg1 */ + 0xc025, /* mplla cfg2 */ + 0xc025, /* mplla cfg3 */ + 0xa6ab, /* mplla cfg4 */ + 0x8c00, /* mplla cfg5 */ + 0x4000, /* mplla cfg6 */ + 0x0003, /* mplla cfg7 */ + 0x3555, /* mplla cfg8 */ + 0x0001, /* mplla cfg9 */ + }, +}; + +static const struct intel_c20pll_state * const mtl_c20_dp_tables[] = { + &mtl_c20_dp_rbr, + &mtl_c20_dp_hbr1, + &mtl_c20_dp_hbr2, + &mtl_c20_dp_hbr3, + &mtl_c20_dp_uhbr10, + &mtl_c20_dp_uhbr13_5, + &mtl_c20_dp_uhbr20, + NULL, +}; + +/* + * HDMI link rates with 38.4 MHz reference clock. + */ + +static const struct intel_c10pll_state mtl_c10_hdmi_25_2 = { + .clock = 25200, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x4, + .pll[1] = 0, + .pll[2] = 0xB2, + .pll[3] = 0, + .pll[4] = 0, + .pll[5] = 0, + .pll[6] = 0, + .pll[7] = 0, + .pll[8] = 0x20, + .pll[9] = 0x1, + .pll[10] = 0, + .pll[11] = 0, + .pll[12] = 0, + .pll[13] = 0, + .pll[14] = 0, + .pll[15] = 0xD, + .pll[16] = 0x6, + .pll[17] = 0x8F, + .pll[18] = 0x84, + .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_27_0 = { + .clock = 27000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x34, + .pll[1] = 0, + .pll[2] = 0xC0, + .pll[3] = 0, + .pll[4] = 0, + .pll[5] = 0, + .pll[6] = 0, + .pll[7] = 0, + .pll[8] = 0x20, + .pll[9] = 0x1, + .pll[10] = 0, + .pll[11] = 0, + .pll[12] = 0x80, + .pll[13] = 0, + .pll[14] = 0, + .pll[15] = 0xD, + .pll[16] = 0x6, + .pll[17] = 0xCF, + .pll[18] = 0x84, + .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_74_25 = { + .clock = 74250, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, + .pll[1] = 0, + .pll[2] = 0x7A, + .pll[3] = 0, + .pll[4] = 0, + .pll[5] = 0, + .pll[6] = 0, + .pll[7] = 0, + .pll[8] = 0x20, + .pll[9] = 0x1, + .pll[10] = 0, + .pll[11] = 0, + .pll[12] = 0x58, + .pll[13] = 0, + .pll[14] = 0, + .pll[15] = 0xB, + .pll[16] = 0x6, + .pll[17] = 0xF, + .pll[18] = 0x85, + .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_148_5 = { + .clock = 148500, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, + .pll[1] = 0, + .pll[2] = 0x7A, + .pll[3] = 0, + .pll[4] = 0, + .pll[5] = 0, + .pll[6] = 0, + .pll[7] = 0, + .pll[8] = 0x20, + .pll[9] = 0x1, + .pll[10] = 0, + .pll[11] = 0, + .pll[12] = 0x58, + .pll[13] = 0, + .pll[14] = 0, + .pll[15] = 0xA, + .pll[16] = 0x6, + .pll[17] = 0xF, + .pll[18] = 0x85, + .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_594 = { + .clock = 594000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, + .pll[1] = 0, + .pll[2] = 0x7A, + .pll[3] = 0, + .pll[4] = 0, + .pll[5] = 0, + .pll[6] = 0, + .pll[7] = 0, + .pll[8] = 0x20, + .pll[9] = 0x1, + .pll[10] = 0, + .pll[11] = 0, + .pll[12] = 0x58, + .pll[13] = 0, + .pll[14] = 0, + .pll[15] = 0x8, + .pll[16] = 0x6, + .pll[17] = 0xF, + .pll[18] = 0x85, + .pll[19] = 0x23, +}; + +/* Precomputed C10 HDMI PLL tables */ +static const struct intel_c10pll_state mtl_c10_hdmi_27027 = { + .clock = 27027, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xC0, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0xCC, .pll[12] = 0x9C, .pll[13] = 0xCB, .pll[14] = 0xCC, + .pll[15] = 0x0D, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_28320 = { + .clock = 28320, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x04, .pll[1] = 0x00, .pll[2] = 0xCC, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x00, .pll[13] = 0x00, .pll[14] = 0x00, + .pll[15] = 0x0D, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_30240 = { + .clock = 30240, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x04, .pll[1] = 0x00, .pll[2] = 0xDC, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x00, .pll[13] = 0x00, .pll[14] = 0x00, + .pll[15] = 0x0D, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_31500 = { + .clock = 31500, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x62, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0xA0, .pll[13] = 0x00, .pll[14] = 0x00, + .pll[15] = 0x0C, .pll[16] = 0x09, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_36000 = { + .clock = 36000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xC4, .pll[1] = 0x00, .pll[2] = 0x76, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x00, .pll[13] = 0x00, .pll[14] = 0x00, + .pll[15] = 0x0C, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_40000 = { + .clock = 40000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xB4, .pll[1] = 0x00, .pll[2] = 0x86, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0x55, .pll[13] = 0x55, .pll[14] = 0x55, + .pll[15] = 0x0C, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_49500 = { + .clock = 49500, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x74, .pll[1] = 0x00, .pll[2] = 0xAE, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x20, .pll[13] = 0x00, .pll[14] = 0x00, + .pll[15] = 0x0C, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_50000 = { + .clock = 50000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x74, .pll[1] = 0x00, .pll[2] = 0xB0, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0xAA, .pll[12] = 0x2A, .pll[13] = 0xA9, .pll[14] = 0xAA, + .pll[15] = 0x0C, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_57284 = { + .clock = 57284, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xCE, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x77, .pll[12] = 0x57, .pll[13] = 0x77, .pll[14] = 0x77, + .pll[15] = 0x0C, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_58000 = { + .clock = 58000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xD0, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0xD5, .pll[13] = 0x55, .pll[14] = 0x55, + .pll[15] = 0x0C, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_65000 = { + .clock = 65000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x66, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0xB5, .pll[13] = 0x55, .pll[14] = 0x55, + .pll[15] = 0x0B, .pll[16] = 0x09, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_71000 = { + .clock = 71000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x72, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0xF5, .pll[13] = 0x55, .pll[14] = 0x55, + .pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_74176 = { + .clock = 74176, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x7A, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x44, .pll[12] = 0x44, .pll[13] = 0x44, .pll[14] = 0x44, + .pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_75000 = { + .clock = 75000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x7C, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x20, .pll[13] = 0x00, .pll[14] = 0x00, + .pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_78750 = { + .clock = 78750, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xB4, .pll[1] = 0x00, .pll[2] = 0x84, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x08, .pll[13] = 0x00, .pll[14] = 0x00, + .pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_85500 = { + .clock = 85500, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xB4, .pll[1] = 0x00, .pll[2] = 0x92, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x10, .pll[13] = 0x00, .pll[14] = 0x00, + .pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_88750 = { + .clock = 88750, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x74, .pll[1] = 0x00, .pll[2] = 0x98, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0xAA, .pll[12] = 0x72, .pll[13] = 0xA9, .pll[14] = 0xAA, + .pll[15] = 0x0B, .pll[16] = 0x09, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_106500 = { + .clock = 106500, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xBC, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0xF0, .pll[13] = 0x00, .pll[14] = 0x00, + .pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_108000 = { + .clock = 108000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xC0, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x80, .pll[13] = 0x00, .pll[14] = 0x00, + .pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_115500 = { + .clock = 115500, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xD0, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x50, .pll[13] = 0x00, .pll[14] = 0x00, + .pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_119000 = { + .clock = 119000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xD6, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0xF5, .pll[13] = 0x55, .pll[14] = 0x55, + .pll[15] = 0x0B, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_135000 = { + .clock = 135000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x6C, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x50, .pll[13] = 0x00, .pll[14] = 0x00, + .pll[15] = 0x0A, .pll[16] = 0x09, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_138500 = { + .clock = 138500, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x70, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0xAA, .pll[12] = 0x22, .pll[13] = 0xA9, .pll[14] = 0xAA, + .pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_147160 = { + .clock = 147160, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x78, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0xA5, .pll[13] = 0x55, .pll[14] = 0x55, + .pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_148352 = { + .clock = 148352, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x7A, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x44, .pll[12] = 0x44, .pll[13] = 0x44, .pll[14] = 0x44, + .pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_154000 = { + .clock = 154000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xB4, .pll[1] = 0x00, .pll[2] = 0x80, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0x35, .pll[13] = 0x55, .pll[14] = 0x55, + .pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_162000 = { + .clock = 162000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xB4, .pll[1] = 0x00, .pll[2] = 0x88, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x60, .pll[13] = 0x00, .pll[14] = 0x00, + .pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_167000 = { + .clock = 167000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xB4, .pll[1] = 0x00, .pll[2] = 0x8C, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0xAA, .pll[12] = 0xFA, .pll[13] = 0xA9, .pll[14] = 0xAA, + .pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_197802 = { + .clock = 197802, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x74, .pll[1] = 0x00, .pll[2] = 0xAE, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x99, .pll[12] = 0x05, .pll[13] = 0x98, .pll[14] = 0x99, + .pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_198000 = { + .clock = 198000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x74, .pll[1] = 0x00, .pll[2] = 0xAE, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x20, .pll[13] = 0x00, .pll[14] = 0x00, + .pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_209800 = { + .clock = 209800, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xBA, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0x45, .pll[13] = 0x55, .pll[14] = 0x55, + .pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_241500 = { + .clock = 241500, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xDA, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0xC8, .pll[13] = 0x00, .pll[14] = 0x00, + .pll[15] = 0x0A, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_262750 = { + .clock = 262750, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x68, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0xAA, .pll[12] = 0x6C, .pll[13] = 0xA9, .pll[14] = 0xAA, + .pll[15] = 0x09, .pll[16] = 0x09, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_268500 = { + .clock = 268500, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x6A, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0xEC, .pll[13] = 0x00, .pll[14] = 0x00, + .pll[15] = 0x09, .pll[16] = 0x09, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_296703 = { + .clock = 296703, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x7A, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x33, .pll[12] = 0x44, .pll[13] = 0x33, .pll[14] = 0x33, + .pll[15] = 0x09, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_297000 = { + .clock = 297000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x7A, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x00, .pll[12] = 0x58, .pll[13] = 0x00, .pll[14] = 0x00, + .pll[15] = 0x09, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_319750 = { + .clock = 319750, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xB4, .pll[1] = 0x00, .pll[2] = 0x86, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0xAA, .pll[12] = 0x44, .pll[13] = 0xA9, .pll[14] = 0xAA, + .pll[15] = 0x09, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_497750 = { + .clock = 497750, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0x34, .pll[1] = 0x00, .pll[2] = 0xE2, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0x9F, .pll[13] = 0x55, .pll[14] = 0x55, + .pll[15] = 0x09, .pll[16] = 0x08, .pll[17] = 0xCF, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_592000 = { + .clock = 592000, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x7A, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x55, .pll[12] = 0x15, .pll[13] = 0x55, .pll[14] = 0x55, + .pll[15] = 0x08, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state mtl_c10_hdmi_593407 = { + .clock = 593407, + .tx = 0x10, + .cmn = 0x1, + .pll[0] = 0xF4, .pll[1] = 0x00, .pll[2] = 0x7A, .pll[3] = 0x00, .pll[4] = 0x00, + .pll[5] = 0x00, .pll[6] = 0x00, .pll[7] = 0x00, .pll[8] = 0x20, .pll[9] = 0xFF, + .pll[10] = 0xFF, .pll[11] = 0x3B, .pll[12] = 0x44, .pll[13] = 0xBA, .pll[14] = 0xBB, + .pll[15] = 0x08, .pll[16] = 0x08, .pll[17] = 0x8F, .pll[18] = 0x84, .pll[19] = 0x23, +}; + +static const struct intel_c10pll_state * const mtl_c10_hdmi_tables[] = { + &mtl_c10_hdmi_25_2, /* Consolidated Table */ + &mtl_c10_hdmi_27_0, /* Consolidated Table */ + &mtl_c10_hdmi_27027, + &mtl_c10_hdmi_28320, + &mtl_c10_hdmi_30240, + &mtl_c10_hdmi_31500, + &mtl_c10_hdmi_36000, + &mtl_c10_hdmi_40000, + &mtl_c10_hdmi_49500, + &mtl_c10_hdmi_50000, + &mtl_c10_hdmi_57284, + &mtl_c10_hdmi_58000, + &mtl_c10_hdmi_65000, + &mtl_c10_hdmi_71000, + &mtl_c10_hdmi_74176, + &mtl_c10_hdmi_74_25, /* Consolidated Table */ + &mtl_c10_hdmi_75000, + &mtl_c10_hdmi_78750, + &mtl_c10_hdmi_85500, + &mtl_c10_hdmi_88750, + &mtl_c10_hdmi_106500, + &mtl_c10_hdmi_108000, + &mtl_c10_hdmi_115500, + &mtl_c10_hdmi_119000, + &mtl_c10_hdmi_135000, + &mtl_c10_hdmi_138500, + &mtl_c10_hdmi_147160, + &mtl_c10_hdmi_148352, + &mtl_c10_hdmi_148_5, /* Consolidated Table */ + &mtl_c10_hdmi_154000, + &mtl_c10_hdmi_162000, + &mtl_c10_hdmi_167000, + &mtl_c10_hdmi_197802, + &mtl_c10_hdmi_198000, + &mtl_c10_hdmi_209800, + &mtl_c10_hdmi_241500, + &mtl_c10_hdmi_262750, + &mtl_c10_hdmi_268500, + &mtl_c10_hdmi_296703, + &mtl_c10_hdmi_297000, + &mtl_c10_hdmi_319750, + &mtl_c10_hdmi_497750, + &mtl_c10_hdmi_592000, + &mtl_c10_hdmi_593407, + &mtl_c10_hdmi_594, /* Consolidated Table */ + NULL, +}; + +static const struct intel_c20pll_state mtl_c20_hdmi_25_175 = { + .link_bit_rate = 25175, + .clock = 25175, + .tx = { 0xbe88, /* tx cfg0 */ + 0x9800, /* tx cfg1 */ + 0x0000, /* tx cfg2 */ + }, + .cmn = { 0x0500, /* cmn cfg0*/ + 0x0005, /* cmn cfg1 */ + 0x0000, /* cmn cfg2 */ + 0x0000, /* cmn cfg3 */ + }, + .mpllb = { 0xa0d2, /* mpllb cfg0 */ + 0x7d80, /* mpllb cfg1 */ + 0x0906, /* mpllb cfg2 */ + 0xbe40, /* mpllb cfg3 */ + 0x0000, /* mpllb cfg4 */ + 0x0000, /* mpllb cfg5 */ + 0x0200, /* mpllb cfg6 */ + 0x0001, /* mpllb cfg7 */ + 0x0000, /* mpllb cfg8 */ + 0x0000, /* mpllb cfg9 */ + 0x0001, /* mpllb cfg10 */ + }, +}; + +static const struct intel_c20pll_state mtl_c20_hdmi_27_0 = { + .link_bit_rate = 27000, + .clock = 27000, + .tx = { 0xbe88, /* tx cfg0 */ + 0x9800, /* tx cfg1 */ + 0x0000, /* tx cfg2 */ + }, + .cmn = { 0x0500, /* cmn cfg0*/ + 0x0005, /* cmn cfg1 */ + 0x0000, /* cmn cfg2 */ + 0x0000, /* cmn cfg3 */ + }, + .mpllb = { 0xa0e0, /* mpllb cfg0 */ + 0x7d80, /* mpllb cfg1 */ + 0x0906, /* mpllb cfg2 */ + 0xbe40, /* mpllb cfg3 */ + 0x0000, /* mpllb cfg4 */ + 0x0000, /* mpllb cfg5 */ + 0x2200, /* mpllb cfg6 */ + 0x0001, /* mpllb cfg7 */ + 0x8000, /* mpllb cfg8 */ + 0x0000, /* mpllb cfg9 */ + 0x0001, /* mpllb cfg10 */ + }, +}; + +static const struct intel_c20pll_state mtl_c20_hdmi_74_25 = { + .link_bit_rate = 74250, + .clock = 74250, + .tx = { 0xbe88, /* tx cfg0 */ + 0x9800, /* tx cfg1 */ + 0x0000, /* tx cfg2 */ + }, + .cmn = { 0x0500, /* cmn cfg0*/ + 0x0005, /* cmn cfg1 */ + 0x0000, /* cmn cfg2 */ + 0x0000, /* cmn cfg3 */ + }, + .mpllb = { 0x609a, /* mpllb cfg0 */ + 0x7d40, /* mpllb cfg1 */ + 0xca06, /* mpllb cfg2 */ + 0xbe40, /* mpllb cfg3 */ + 0x0000, /* mpllb cfg4 */ + 0x0000, /* mpllb cfg5 */ + 0x2200, /* mpllb cfg6 */ + 0x0001, /* mpllb cfg7 */ + 0x5800, /* mpllb cfg8 */ + 0x0000, /* mpllb cfg9 */ + 0x0001, /* mpllb cfg10 */ + }, +}; + +static const struct intel_c20pll_state mtl_c20_hdmi_148_5 = { + .link_bit_rate = 148500, + .clock = 148500, + .tx = { 0xbe88, /* tx cfg0 */ + 0x9800, /* tx cfg1 */ + 0x0000, /* tx cfg2 */ + }, + .cmn = { 0x0500, /* cmn cfg0*/ + 0x0005, /* cmn cfg1 */ + 0x0000, /* cmn cfg2 */ + 0x0000, /* cmn cfg3 */ + }, + .mpllb = { 0x409a, /* mpllb cfg0 */ + 0x7d20, /* mpllb cfg1 */ + 0xca06, /* mpllb cfg2 */ + 0xbe40, /* mpllb cfg3 */ + 0x0000, /* mpllb cfg4 */ + 0x0000, /* mpllb cfg5 */ + 0x2200, /* mpllb cfg6 */ + 0x0001, /* mpllb cfg7 */ + 0x5800, /* mpllb cfg8 */ + 0x0000, /* mpllb cfg9 */ + 0x0001, /* mpllb cfg10 */ + }, +}; + +static const struct intel_c20pll_state mtl_c20_hdmi_594 = { + .link_bit_rate = 594000, + .clock = 594000, + .tx = { 0xbe88, /* tx cfg0 */ + 0x9800, /* tx cfg1 */ + 0x0000, /* tx cfg2 */ + }, + .cmn = { 0x0500, /* cmn cfg0*/ + 0x0005, /* cmn cfg1 */ + 0x0000, /* cmn cfg2 */ + 0x0000, /* cmn cfg3 */ + }, + .mpllb = { 0x009a, /* mpllb cfg0 */ + 0x7d08, /* mpllb cfg1 */ + 0xca06, /* mpllb cfg2 */ + 0xbe40, /* mpllb cfg3 */ + 0x0000, /* mpllb cfg4 */ + 0x0000, /* mpllb cfg5 */ + 0x2200, /* mpllb cfg6 */ + 0x0001, /* mpllb cfg7 */ + 0x5800, /* mpllb cfg8 */ + 0x0000, /* mpllb cfg9 */ + 0x0001, /* mpllb cfg10 */ + }, +}; + +static const struct intel_c20pll_state mtl_c20_hdmi_300 = { + .link_bit_rate = 3000000, + .clock = 166670, + .tx = { 0xbe98, /* tx cfg0 */ + 0x9800, /* tx cfg1 */ + 0x0000, /* tx cfg2 */ + }, + .cmn = { 0x0500, /* cmn cfg0*/ + 0x0005, /* cmn cfg1 */ + 0x0000, /* cmn cfg2 */ + 0x0000, /* cmn cfg3 */ + }, + .mpllb = { 0x209c, /* mpllb cfg0 */ + 0x7d10, /* mpllb cfg1 */ + 0xca06, /* mpllb cfg2 */ + 0xbe40, /* mpllb cfg3 */ + 0x0000, /* mpllb cfg4 */ + 0x0000, /* mpllb cfg5 */ + 0x2200, /* mpllb cfg6 */ + 0x0001, /* mpllb cfg7 */ + 0x2000, /* mpllb cfg8 */ + 0x0000, /* mpllb cfg9 */ + 0x0004, /* mpllb cfg10 */ + }, +}; + +static const struct intel_c20pll_state mtl_c20_hdmi_600 = { + .link_bit_rate = 6000000, + .clock = 333330, + .tx = { 0xbe98, /* tx cfg0 */ + 0x9800, /* tx cfg1 */ + 0x0000, /* tx cfg2 */ + }, + .cmn = { 0x0500, /* cmn cfg0*/ + 0x0005, /* cmn cfg1 */ + 0x0000, /* cmn cfg2 */ + 0x0000, /* cmn cfg3 */ + }, + .mpllb = { 0x009c, /* mpllb cfg0 */ + 0x7d08, /* mpllb cfg1 */ + 0xca06, /* mpllb cfg2 */ + 0xbe40, /* mpllb cfg3 */ + 0x0000, /* mpllb cfg4 */ + 0x0000, /* mpllb cfg5 */ + 0x2200, /* mpllb cfg6 */ + 0x0001, /* mpllb cfg7 */ + 0x2000, /* mpllb cfg8 */ + 0x0000, /* mpllb cfg9 */ + 0x0004, /* mpllb cfg10 */ + }, +}; + +static const struct intel_c20pll_state mtl_c20_hdmi_800 = { + .link_bit_rate = 8000000, + .clock = 444440, + .tx = { 0xbe98, /* tx cfg0 */ + 0x9800, /* tx cfg1 */ + 0x0000, /* tx cfg2 */ + }, + .cmn = { 0x0500, /* cmn cfg0*/ + 0x0005, /* cmn cfg1 */ + 0x0000, /* cmn cfg2 */ + 0x0000, /* cmn cfg3 */ + }, + .mpllb = { 0x00d0, /* mpllb cfg0 */ + 0x7d08, /* mpllb cfg1 */ + 0x4a06, /* mpllb cfg2 */ + 0xbe40, /* mpllb cfg3 */ + 0x0000, /* mpllb cfg4 */ + 0x0000, /* mpllb cfg5 */ + 0x2200, /* mpllb cfg6 */ + 0x0003, /* mpllb cfg7 */ + 0x2aaa, /* mpllb cfg8 */ + 0x0002, /* mpllb cfg9 */ + 0x0004, /* mpllb cfg10 */ + }, +}; + +static const struct intel_c20pll_state mtl_c20_hdmi_1000 = { + .link_bit_rate = 10000000, + .clock = 555560, + .tx = { 0xbe98, /* tx cfg0 */ + 0x9800, /* tx cfg1 */ + 0x0000, /* tx cfg2 */ + }, + .cmn = { 0x0500, /* cmn cfg0*/ + 0x0005, /* cmn cfg1 */ + 0x0000, /* cmn cfg2 */ + 0x0000, /* cmn cfg3 */ + }, + .mpllb = { 0x1104, /* mpllb cfg0 */ + 0x7d08, /* mpllb cfg1 */ + 0x0a06, /* mpllb cfg2 */ + 0xbe40, /* mpllb cfg3 */ + 0x0000, /* mpllb cfg4 */ + 0x0000, /* mpllb cfg5 */ + 0x2200, /* mpllb cfg6 */ + 0x0003, /* mpllb cfg7 */ + 0x3555, /* mpllb cfg8 */ + 0x0001, /* mpllb cfg9 */ + 0x0004, /* mpllb cfg10 */ + }, +}; + +static const struct intel_c20pll_state mtl_c20_hdmi_1200 = { + .link_bit_rate = 12000000, + .clock = 666670, + .tx = { 0xbe98, /* tx cfg0 */ + 0x9800, /* tx cfg1 */ + 0x0000, /* tx cfg2 */ + }, + .cmn = { 0x0500, /* cmn cfg0*/ + 0x0005, /* cmn cfg1 */ + 0x0000, /* cmn cfg2 */ + 0x0000, /* cmn cfg3 */ + }, + .mpllb = { 0x0138, /* mpllb cfg0 */ + 0x7d08, /* mpllb cfg1 */ + 0x5486, /* mpllb cfg2 */ + 0xfe40, /* mpllb cfg3 */ + 0x0000, /* mpllb cfg4 */ + 0x0000, /* mpllb cfg5 */ + 0x2200, /* mpllb cfg6 */ + 0x0001, /* mpllb cfg7 */ + 0x4000, /* mpllb cfg8 */ + 0x0000, /* mpllb cfg9 */ + 0x0004, /* mpllb cfg10 */ + }, +}; + +static const struct intel_c20pll_state * const mtl_c20_hdmi_tables[] = { + &mtl_c20_hdmi_25_175, + &mtl_c20_hdmi_27_0, + &mtl_c20_hdmi_74_25, + &mtl_c20_hdmi_148_5, + &mtl_c20_hdmi_594, + &mtl_c20_hdmi_300, + &mtl_c20_hdmi_600, + &mtl_c20_hdmi_800, + &mtl_c20_hdmi_1000, + &mtl_c20_hdmi_1200, + NULL, +}; + +static int intel_c10_phy_check_hdmi_link_rate(int clock) +{ + const struct intel_c10pll_state * const *tables = mtl_c10_hdmi_tables; + int i; + + for (i = 0; tables[i]; i++) { + if (clock == tables[i]->clock) + return MODE_OK; + } + + return MODE_CLOCK_RANGE; +} + +static const struct intel_c10pll_state * const * +intel_c10pll_tables_get(struct intel_crtc_state *crtc_state, + struct intel_encoder *encoder) +{ + if (intel_crtc_has_dp_encoder(crtc_state)) { + if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_EDP)) + return mtl_c10_edp_tables; + else + return mtl_c10_dp_tables; + } else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) { + return mtl_c10_hdmi_tables; + } + + MISSING_CASE(encoder->type); + return NULL; +} + +static void intel_c10pll_update_pll(struct intel_crtc_state *crtc_state, + struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + struct intel_cx0pll_state *pll_state = &crtc_state->cx0pll_state; + int i; + + if (intel_crtc_has_dp_encoder(crtc_state)) { + if (intel_panel_use_ssc(i915)) { + struct intel_dp *intel_dp = enc_to_intel_dp(encoder); + + pll_state->ssc_enabled = + (intel_dp->dpcd[DP_MAX_DOWNSPREAD] & DP_MAX_DOWNSPREAD_0_5); + } + } + + if (pll_state->ssc_enabled) + return; + + drm_WARN_ON(&i915->drm, ARRAY_SIZE(pll_state->c10.pll) < 9); + for (i = 4; i < 9; i++) + pll_state->c10.pll[i] = 0; +} + +static int intel_c10pll_calc_state(struct intel_crtc_state *crtc_state, + struct intel_encoder *encoder) +{ + const struct intel_c10pll_state * const *tables; + int i; + + tables = intel_c10pll_tables_get(crtc_state, encoder); + if (!tables) + return -EINVAL; + + for (i = 0; tables[i]; i++) { + if (crtc_state->port_clock == tables[i]->clock) { + crtc_state->cx0pll_state.c10 = *tables[i]; + intel_c10pll_update_pll(crtc_state, encoder); + + return 0; + } + } + + return -EINVAL; +} + +void intel_c10pll_readout_hw_state(struct intel_encoder *encoder, + struct intel_c10pll_state *pll_state) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + u8 lane = INTEL_CX0_LANE0; + intel_wakeref_t wakeref; + int i; + + wakeref = intel_cx0_phy_transaction_begin(encoder); + + /* + * According to C10 VDR Register programming Sequence we need + * to do this to read PHY internal registers from MsgBus. + */ + intel_cx0_rmw(i915, encoder->port, lane, PHY_C10_VDR_CONTROL(1), + 0, C10_VDR_CTRL_MSGBUS_ACCESS, + MB_WRITE_COMMITTED); + + for (i = 0; i < ARRAY_SIZE(pll_state->pll); i++) + pll_state->pll[i] = intel_cx0_read(i915, encoder->port, lane, + PHY_C10_VDR_PLL(i)); + + pll_state->cmn = intel_cx0_read(i915, encoder->port, lane, PHY_C10_VDR_CMN(0)); + pll_state->tx = intel_cx0_read(i915, encoder->port, lane, PHY_C10_VDR_TX(0)); + + intel_cx0_phy_transaction_end(encoder, wakeref); +} + +static void intel_c10_pll_program(struct drm_i915_private *i915, + const struct intel_crtc_state *crtc_state, + struct intel_encoder *encoder) +{ + const struct intel_c10pll_state *pll_state = &crtc_state->cx0pll_state.c10; + int i; + + intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CONTROL(1), + 0, C10_VDR_CTRL_MSGBUS_ACCESS, + MB_WRITE_COMMITTED); + + /* Custom width needs to be programmed to 0 for both the phy lanes */ + intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CUSTOM_WIDTH, + C10_VDR_CUSTOM_WIDTH_MASK, C10_VDR_CUSTOM_WIDTH_8_10, + MB_WRITE_COMMITTED); + intel_cx0_rmw(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C10_VDR_CONTROL(1), + 0, C10_VDR_CTRL_UPDATE_CFG, + MB_WRITE_COMMITTED); + + /* Program the pll values only for the master lane */ + for (i = 0; i < ARRAY_SIZE(pll_state->pll); i++) + intel_cx0_write(i915, encoder->port, INTEL_CX0_LANE0, PHY_C10_VDR_PLL(i), + pll_state->pll[i], + (i % 4) ? MB_WRITE_UNCOMMITTED : MB_WRITE_COMMITTED); + + intel_cx0_write(i915, encoder->port, INTEL_CX0_LANE0, PHY_C10_VDR_CMN(0), pll_state->cmn, MB_WRITE_COMMITTED); + intel_cx0_write(i915, encoder->port, INTEL_CX0_LANE0, PHY_C10_VDR_TX(0), pll_state->tx, MB_WRITE_COMMITTED); + + intel_cx0_rmw(i915, encoder->port, INTEL_CX0_LANE0, PHY_C10_VDR_CONTROL(1), + 0, C10_VDR_CTRL_MASTER_LANE | C10_VDR_CTRL_UPDATE_CFG, + MB_WRITE_COMMITTED); +} + +void intel_c10pll_dump_hw_state(struct drm_i915_private *i915, + const struct intel_c10pll_state *hw_state) +{ + bool fracen; + int i; + unsigned int frac_quot = 0, frac_rem = 0, frac_den = 1; + unsigned int multiplier, tx_clk_div; + + fracen = hw_state->pll[0] & C10_PLL0_FRACEN; + drm_dbg_kms(&i915->drm, "c10pll_hw_state: fracen: %s, ", + str_yes_no(fracen)); + + if (fracen) { + frac_quot = hw_state->pll[12] << 8 | hw_state->pll[11]; + frac_rem = hw_state->pll[14] << 8 | hw_state->pll[13]; + frac_den = hw_state->pll[10] << 8 | hw_state->pll[9]; + drm_dbg_kms(&i915->drm, "quot: %u, rem: %u, den: %u,\n", + frac_quot, frac_rem, frac_den); + } + + multiplier = (REG_FIELD_GET8(C10_PLL3_MULTIPLIERH_MASK, hw_state->pll[3]) << 8 | + hw_state->pll[2]) / 2 + 16; + tx_clk_div = REG_FIELD_GET8(C10_PLL15_TXCLKDIV_MASK, hw_state->pll[15]); + drm_dbg_kms(&i915->drm, + "multiplier: %u, tx_clk_div: %u.\n", multiplier, tx_clk_div); + + drm_dbg_kms(&i915->drm, "c10pll_rawhw_state:"); + drm_dbg_kms(&i915->drm, "tx: 0x%x, cmn: 0x%x\n", hw_state->tx, hw_state->cmn); + + BUILD_BUG_ON(ARRAY_SIZE(hw_state->pll) % 4); + for (i = 0; i < ARRAY_SIZE(hw_state->pll); i = i + 4) + drm_dbg_kms(&i915->drm, "pll[%d] = 0x%x, pll[%d] = 0x%x, pll[%d] = 0x%x, pll[%d] = 0x%x\n", + i, hw_state->pll[i], i + 1, hw_state->pll[i + 1], + i + 2, hw_state->pll[i + 2], i + 3, hw_state->pll[i + 3]); +} + +static int intel_c20_compute_hdmi_tmds_pll(u64 pixel_clock, struct intel_c20pll_state *pll_state) +{ + u64 datarate; + u64 mpll_tx_clk_div; + u64 vco_freq_shift; + u64 vco_freq; + u64 multiplier; + u64 mpll_multiplier; + u64 mpll_fracn_quot; + u64 mpll_fracn_rem; + u8 mpllb_ana_freq_vco; + u8 mpll_div_multiplier; + + if (pixel_clock < 25175 || pixel_clock > 600000) + return -EINVAL; + + datarate = ((u64)pixel_clock * 1000) * 10; + mpll_tx_clk_div = ilog2(div64_u64((u64)CLOCK_9999MHZ, (u64)datarate)); + vco_freq_shift = ilog2(div64_u64((u64)CLOCK_4999MHZ * (u64)256, (u64)datarate)); + vco_freq = (datarate << vco_freq_shift) >> 8; + multiplier = div64_u64((vco_freq << 28), (REFCLK_38_4_MHZ >> 4)); + mpll_multiplier = 2 * (multiplier >> 32); + + mpll_fracn_quot = (multiplier >> 16) & 0xFFFF; + mpll_fracn_rem = multiplier & 0xFFFF; + + mpll_div_multiplier = min_t(u8, div64_u64((vco_freq * 16 + (datarate >> 1)), + datarate), 255); + + if (vco_freq <= DATARATE_3000000000) + mpllb_ana_freq_vco = MPLLB_ANA_FREQ_VCO_3; + else if (vco_freq <= DATARATE_3500000000) + mpllb_ana_freq_vco = MPLLB_ANA_FREQ_VCO_2; + else if (vco_freq <= DATARATE_4000000000) + mpllb_ana_freq_vco = MPLLB_ANA_FREQ_VCO_1; + else + mpllb_ana_freq_vco = MPLLB_ANA_FREQ_VCO_0; + + pll_state->link_bit_rate = pixel_clock; + pll_state->clock = pixel_clock; + pll_state->tx[0] = 0xbe88; + pll_state->tx[1] = 0x9800; + pll_state->tx[2] = 0x0000; + pll_state->cmn[0] = 0x0500; + pll_state->cmn[1] = 0x0005; + pll_state->cmn[2] = 0x0000; + pll_state->cmn[3] = 0x0000; + pll_state->mpllb[0] = (MPLL_TX_CLK_DIV(mpll_tx_clk_div) | + MPLL_MULTIPLIER(mpll_multiplier)); + pll_state->mpllb[1] = (CAL_DAC_CODE(CAL_DAC_CODE_31) | + WORD_CLK_DIV | + MPLL_DIV_MULTIPLIER(mpll_div_multiplier)); + pll_state->mpllb[2] = (MPLLB_ANA_FREQ_VCO(mpllb_ana_freq_vco) | + CP_PROP(CP_PROP_20) | + CP_INT(CP_INT_6)); + pll_state->mpllb[3] = (V2I(V2I_2) | + CP_PROP_GS(CP_PROP_GS_30) | + CP_INT_GS(CP_INT_GS_28)); + pll_state->mpllb[4] = 0x0000; + pll_state->mpllb[5] = 0x0000; + pll_state->mpllb[6] = (C20_MPLLB_FRACEN | SSC_UP_SPREAD); + pll_state->mpllb[7] = MPLL_FRACN_DEN; + pll_state->mpllb[8] = mpll_fracn_quot; + pll_state->mpllb[9] = mpll_fracn_rem; + pll_state->mpllb[10] = HDMI_DIV(HDMI_DIV_1); + + return 0; +} + +static int intel_c20_phy_check_hdmi_link_rate(int clock) +{ + const struct intel_c20pll_state * const *tables = mtl_c20_hdmi_tables; + int i; + + for (i = 0; tables[i]; i++) { + if (clock == tables[i]->link_bit_rate) + return MODE_OK; + } + + if (clock >= 25175 && clock <= 594000) + return MODE_OK; + + return MODE_CLOCK_RANGE; +} + +int intel_cx0_phy_check_hdmi_link_rate(struct intel_hdmi *hdmi, int clock) +{ + struct intel_digital_port *dig_port = hdmi_to_dig_port(hdmi); + struct drm_i915_private *i915 = intel_hdmi_to_i915(hdmi); + enum phy phy = intel_port_to_phy(i915, dig_port->base.port); + + if (intel_is_c10phy(i915, phy)) + return intel_c10_phy_check_hdmi_link_rate(clock); + return intel_c20_phy_check_hdmi_link_rate(clock); +} + +static const struct intel_c20pll_state * const * +intel_c20_pll_tables_get(struct intel_crtc_state *crtc_state, + struct intel_encoder *encoder) +{ + if (intel_crtc_has_dp_encoder(crtc_state)) + return mtl_c20_dp_tables; + else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) + return mtl_c20_hdmi_tables; + + MISSING_CASE(encoder->type); + return NULL; +} + +static int intel_c20pll_calc_state(struct intel_crtc_state *crtc_state, + struct intel_encoder *encoder) +{ + const struct intel_c20pll_state * const *tables; + int i; + + /* try computed C20 HDMI tables before using consolidated tables */ + if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) { + if (intel_c20_compute_hdmi_tmds_pll(crtc_state->port_clock, + &crtc_state->cx0pll_state.c20) == 0) + return 0; + } + + tables = intel_c20_pll_tables_get(crtc_state, encoder); + if (!tables) + return -EINVAL; + + for (i = 0; tables[i]; i++) { + if (crtc_state->port_clock == tables[i]->link_bit_rate) { + crtc_state->cx0pll_state.c20 = *tables[i]; + return 0; + } + } + + return -EINVAL; +} + +int intel_cx0pll_calc_state(struct intel_crtc_state *crtc_state, + struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + enum phy phy = intel_port_to_phy(i915, encoder->port); + + if (intel_is_c10phy(i915, phy)) + return intel_c10pll_calc_state(crtc_state, encoder); + return intel_c20pll_calc_state(crtc_state, encoder); +} + +static bool intel_c20_use_mplla(u32 clock) +{ + /* 10G and 20G rates use MPLLA */ + if (clock == 312500 || clock == 625000) + return true; + + return false; +} + +void intel_c20pll_readout_hw_state(struct intel_encoder *encoder, + struct intel_c20pll_state *pll_state) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + bool cntx; + intel_wakeref_t wakeref; + int i; + + wakeref = intel_cx0_phy_transaction_begin(encoder); + + /* 1. Read current context selection */ + cntx = intel_cx0_read(i915, encoder->port, INTEL_CX0_LANE0, PHY_C20_VDR_CUSTOM_SERDES_RATE) & PHY_C20_CONTEXT_TOGGLE; + + /* Read Tx configuration */ + for (i = 0; i < ARRAY_SIZE(pll_state->tx); i++) { + if (cntx) + pll_state->tx[i] = intel_c20_sram_read(i915, encoder->port, INTEL_CX0_LANE0, + PHY_C20_B_TX_CNTX_CFG(i)); + else + pll_state->tx[i] = intel_c20_sram_read(i915, encoder->port, INTEL_CX0_LANE0, + PHY_C20_A_TX_CNTX_CFG(i)); + } + + /* Read common configuration */ + for (i = 0; i < ARRAY_SIZE(pll_state->cmn); i++) { + if (cntx) + pll_state->cmn[i] = intel_c20_sram_read(i915, encoder->port, INTEL_CX0_LANE0, + PHY_C20_B_CMN_CNTX_CFG(i)); + else + pll_state->cmn[i] = intel_c20_sram_read(i915, encoder->port, INTEL_CX0_LANE0, + PHY_C20_A_CMN_CNTX_CFG(i)); + } + + if (pll_state->tx[0] & C20_PHY_USE_MPLLB) { + /* MPLLB configuration */ + for (i = 0; i < ARRAY_SIZE(pll_state->mpllb); i++) { + if (cntx) + pll_state->mpllb[i] = intel_c20_sram_read(i915, encoder->port, INTEL_CX0_LANE0, + PHY_C20_B_MPLLB_CNTX_CFG(i)); + else + pll_state->mpllb[i] = intel_c20_sram_read(i915, encoder->port, INTEL_CX0_LANE0, + PHY_C20_A_MPLLB_CNTX_CFG(i)); + } + } else { + /* MPLLA configuration */ + for (i = 0; i < ARRAY_SIZE(pll_state->mplla); i++) { + if (cntx) + pll_state->mplla[i] = intel_c20_sram_read(i915, encoder->port, INTEL_CX0_LANE0, + PHY_C20_B_MPLLA_CNTX_CFG(i)); + else + pll_state->mplla[i] = intel_c20_sram_read(i915, encoder->port, INTEL_CX0_LANE0, + PHY_C20_A_MPLLA_CNTX_CFG(i)); + } + } + + intel_cx0_phy_transaction_end(encoder, wakeref); +} + +void intel_c20pll_dump_hw_state(struct drm_i915_private *i915, + const struct intel_c20pll_state *hw_state) +{ + int i; + + drm_dbg_kms(&i915->drm, "c20pll_hw_state:\n"); + drm_dbg_kms(&i915->drm, "tx[0] = 0x%.4x, tx[1] = 0x%.4x, tx[2] = 0x%.4x\n", + hw_state->tx[0], hw_state->tx[1], hw_state->tx[2]); + drm_dbg_kms(&i915->drm, "cmn[0] = 0x%.4x, cmn[1] = 0x%.4x, cmn[2] = 0x%.4x, cmn[3] = 0x%.4x\n", + hw_state->cmn[0], hw_state->cmn[1], hw_state->cmn[2], hw_state->cmn[3]); + + if (intel_c20_use_mplla(hw_state->clock)) { + for (i = 0; i < ARRAY_SIZE(hw_state->mplla); i++) + drm_dbg_kms(&i915->drm, "mplla[%d] = 0x%.4x\n", i, hw_state->mplla[i]); + } else { + for (i = 0; i < ARRAY_SIZE(hw_state->mpllb); i++) + drm_dbg_kms(&i915->drm, "mpllb[%d] = 0x%.4x\n", i, hw_state->mpllb[i]); + } +} + +static u8 intel_c20_get_dp_rate(u32 clock) +{ + switch (clock) { + case 162000: /* 1.62 Gbps DP1.4 */ + return 0; + case 270000: /* 2.7 Gbps DP1.4 */ + return 1; + case 540000: /* 5.4 Gbps DP 1.4 */ + return 2; + case 810000: /* 8.1 Gbps DP1.4 */ + return 3; + case 216000: /* 2.16 Gbps eDP */ + return 4; + case 243000: /* 2.43 Gbps eDP */ + return 5; + case 324000: /* 3.24 Gbps eDP */ + return 6; + case 432000: /* 4.32 Gbps eDP */ + return 7; + case 312500: /* 10 Gbps DP2.0 */ + return 8; + case 421875: /* 13.5 Gbps DP2.0 */ + return 9; + case 625000: /* 20 Gbps DP2.0*/ + return 10; + case 648000: /* 6.48 Gbps eDP*/ + return 11; + case 675000: /* 6.75 Gbps eDP*/ + return 12; + default: + MISSING_CASE(clock); + return 0; + } +} + +static u8 intel_c20_get_hdmi_rate(u32 clock) +{ + if (clock >= 25175 && clock <= 600000) + return 0; + + switch (clock) { + case 166670: /* 3 Gbps */ + case 333330: /* 6 Gbps */ + case 666670: /* 12 Gbps */ + return 1; + case 444440: /* 8 Gbps */ + return 2; + case 555560: /* 10 Gbps */ + return 3; + default: + MISSING_CASE(clock); + return 0; + } +} + +static bool is_dp2(u32 clock) +{ + /* DP2.0 clock rates */ + if (clock == 312500 || clock == 421875 || clock == 625000) + return true; + + return false; +} + +static bool is_hdmi_frl(u32 clock) +{ + switch (clock) { + case 166670: /* 3 Gbps */ + case 333330: /* 6 Gbps */ + case 444440: /* 8 Gbps */ + case 555560: /* 10 Gbps */ + case 666670: /* 12 Gbps */ + return true; + default: + return false; + } +} + +static bool intel_c20_protocol_switch_valid(struct intel_encoder *encoder) +{ + struct intel_digital_port *intel_dig_port = enc_to_dig_port(encoder); + + /* banks should not be cleared for DPALT/USB4/TBT modes */ + /* TODO: optimize re-calibration in legacy mode */ + return intel_tc_port_in_legacy_mode(intel_dig_port); +} + +static int intel_get_c20_custom_width(u32 clock, bool dp) +{ + if (dp && is_dp2(clock)) + return 2; + else if (is_hdmi_frl(clock)) + return 1; + else + return 0; +} + +static void intel_c20_pll_program(struct drm_i915_private *i915, + const struct intel_crtc_state *crtc_state, + struct intel_encoder *encoder) +{ + const struct intel_c20pll_state *pll_state = &crtc_state->cx0pll_state.c20; + bool dp = false; + int lane = crtc_state->lane_count > 2 ? INTEL_CX0_BOTH_LANES : INTEL_CX0_LANE0; + bool cntx; + int i; + + if (intel_crtc_has_dp_encoder(crtc_state)) + dp = true; + + /* 1. Read current context selection */ + cntx = intel_cx0_read(i915, encoder->port, INTEL_CX0_LANE0, PHY_C20_VDR_CUSTOM_SERDES_RATE) & BIT(0); + + /* + * 2. If there is a protocol switch from HDMI to DP or vice versa, clear + * the lane #0 MPLLB CAL_DONE_BANK DP2.0 10G and 20G rates enable MPLLA. + * Protocol switch is only applicable for MPLLA + */ + if (intel_c20_protocol_switch_valid(encoder)) { + for (i = 0; i < 4; i++) + intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, RAWLANEAONX_DIG_TX_MPLLB_CAL_DONE_BANK(i), 0); + usleep_range(4000, 4100); + } + + /* 3. Write SRAM configuration context. If A in use, write configuration to B context */ + /* 3.1 Tx configuration */ + for (i = 0; i < ARRAY_SIZE(pll_state->tx); i++) { + if (cntx) + intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, PHY_C20_A_TX_CNTX_CFG(i), pll_state->tx[i]); + else + intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, PHY_C20_B_TX_CNTX_CFG(i), pll_state->tx[i]); + } + + /* 3.2 common configuration */ + for (i = 0; i < ARRAY_SIZE(pll_state->cmn); i++) { + if (cntx) + intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, PHY_C20_A_CMN_CNTX_CFG(i), pll_state->cmn[i]); + else + intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, PHY_C20_B_CMN_CNTX_CFG(i), pll_state->cmn[i]); + } + + /* 3.3 mpllb or mplla configuration */ + if (intel_c20_use_mplla(pll_state->clock)) { + for (i = 0; i < ARRAY_SIZE(pll_state->mplla); i++) { + if (cntx) + intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, + PHY_C20_A_MPLLA_CNTX_CFG(i), + pll_state->mplla[i]); + else + intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, + PHY_C20_B_MPLLA_CNTX_CFG(i), + pll_state->mplla[i]); + } + } else { + for (i = 0; i < ARRAY_SIZE(pll_state->mpllb); i++) { + if (cntx) + intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, + PHY_C20_A_MPLLB_CNTX_CFG(i), + pll_state->mpllb[i]); + else + intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, + PHY_C20_B_MPLLB_CNTX_CFG(i), + pll_state->mpllb[i]); + } + } + + /* 4. Program custom width to match the link protocol */ + intel_cx0_rmw(i915, encoder->port, lane, PHY_C20_VDR_CUSTOM_WIDTH, + PHY_C20_CUSTOM_WIDTH_MASK, + PHY_C20_CUSTOM_WIDTH(intel_get_c20_custom_width(pll_state->clock, dp)), + MB_WRITE_COMMITTED); + + /* 5. For DP or 6. For HDMI */ + if (dp) { + intel_cx0_rmw(i915, encoder->port, lane, PHY_C20_VDR_CUSTOM_SERDES_RATE, + BIT(6) | PHY_C20_CUSTOM_SERDES_MASK, + BIT(6) | PHY_C20_CUSTOM_SERDES(intel_c20_get_dp_rate(pll_state->clock)), + MB_WRITE_COMMITTED); + } else { + intel_cx0_rmw(i915, encoder->port, lane, PHY_C20_VDR_CUSTOM_SERDES_RATE, + BIT(7) | PHY_C20_CUSTOM_SERDES_MASK, + is_hdmi_frl(pll_state->clock) ? BIT(7) : 0, + MB_WRITE_COMMITTED); + + intel_cx0_write(i915, encoder->port, INTEL_CX0_BOTH_LANES, PHY_C20_VDR_HDMI_RATE, + intel_c20_get_hdmi_rate(pll_state->clock), + MB_WRITE_COMMITTED); + } + + /* + * 7. Write Vendor specific registers to toggle context setting to load + * the updated programming toggle context bit + */ + intel_cx0_rmw(i915, encoder->port, lane, PHY_C20_VDR_CUSTOM_SERDES_RATE, + BIT(0), cntx ? 0 : 1, MB_WRITE_COMMITTED); +} + +int intel_c10pll_calc_port_clock(struct intel_encoder *encoder, + const struct intel_c10pll_state *pll_state) +{ + unsigned int frac_quot = 0, frac_rem = 0, frac_den = 1; + unsigned int multiplier, tx_clk_div, hdmi_div, refclk = 38400; + int tmpclk = 0; + + if (pll_state->pll[0] & C10_PLL0_FRACEN) { + frac_quot = pll_state->pll[12] << 8 | pll_state->pll[11]; + frac_rem = pll_state->pll[14] << 8 | pll_state->pll[13]; + frac_den = pll_state->pll[10] << 8 | pll_state->pll[9]; + } + + multiplier = (REG_FIELD_GET8(C10_PLL3_MULTIPLIERH_MASK, pll_state->pll[3]) << 8 | + pll_state->pll[2]) / 2 + 16; + + tx_clk_div = REG_FIELD_GET8(C10_PLL15_TXCLKDIV_MASK, pll_state->pll[15]); + hdmi_div = REG_FIELD_GET8(C10_PLL15_HDMIDIV_MASK, pll_state->pll[15]); + + tmpclk = DIV_ROUND_CLOSEST_ULL(mul_u32_u32(refclk, (multiplier << 16) + frac_quot) + + DIV_ROUND_CLOSEST(refclk * frac_rem, frac_den), + 10 << (tx_clk_div + 16)); + tmpclk *= (hdmi_div ? 2 : 1); + + return tmpclk; +} + +int intel_c20pll_calc_port_clock(struct intel_encoder *encoder, + const struct intel_c20pll_state *pll_state) +{ + unsigned int frac, frac_en, frac_quot, frac_rem, frac_den; + unsigned int multiplier, refclk = 38400; + unsigned int tx_clk_div; + unsigned int ref_clk_mpllb_div; + unsigned int fb_clk_div4_en; + unsigned int ref, vco; + unsigned int tx_rate_mult; + unsigned int tx_rate = REG_FIELD_GET(C20_PHY_TX_RATE, pll_state->tx[0]); + + if (pll_state->tx[0] & C20_PHY_USE_MPLLB) { + tx_rate_mult = 1; + frac_en = REG_FIELD_GET(C20_MPLLB_FRACEN, pll_state->mpllb[6]); + frac_quot = pll_state->mpllb[8]; + frac_rem = pll_state->mpllb[9]; + frac_den = pll_state->mpllb[7]; + multiplier = REG_FIELD_GET(C20_MULTIPLIER_MASK, pll_state->mpllb[0]); + tx_clk_div = REG_FIELD_GET(C20_MPLLB_TX_CLK_DIV_MASK, pll_state->mpllb[0]); + ref_clk_mpllb_div = REG_FIELD_GET(C20_REF_CLK_MPLLB_DIV_MASK, pll_state->mpllb[6]); + fb_clk_div4_en = 0; + } else { + tx_rate_mult = 2; + frac_en = REG_FIELD_GET(C20_MPLLA_FRACEN, pll_state->mplla[6]); + frac_quot = pll_state->mplla[8]; + frac_rem = pll_state->mplla[9]; + frac_den = pll_state->mplla[7]; + multiplier = REG_FIELD_GET(C20_MULTIPLIER_MASK, pll_state->mplla[0]); + tx_clk_div = REG_FIELD_GET(C20_MPLLA_TX_CLK_DIV_MASK, pll_state->mplla[1]); + ref_clk_mpllb_div = REG_FIELD_GET(C20_REF_CLK_MPLLB_DIV_MASK, pll_state->mplla[6]); + fb_clk_div4_en = REG_FIELD_GET(C20_FB_CLK_DIV4_EN, pll_state->mplla[0]); + } + + if (frac_en) + frac = frac_quot + DIV_ROUND_CLOSEST(frac_rem, frac_den); + else + frac = 0; + + ref = DIV_ROUND_CLOSEST(refclk * (1 << (1 + fb_clk_div4_en)), 1 << ref_clk_mpllb_div); + vco = DIV_ROUND_CLOSEST_ULL(mul_u32_u32(ref, (multiplier << (17 - 2)) + frac) >> 17, 10); + + return vco << tx_rate_mult >> tx_clk_div >> tx_rate; +} + +static void intel_program_port_clock_ctl(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state, + bool lane_reversal) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + u32 val = 0; + + intel_de_rmw(i915, XELPDP_PORT_BUF_CTL1(encoder->port), XELPDP_PORT_REVERSAL, + lane_reversal ? XELPDP_PORT_REVERSAL : 0); + + if (lane_reversal) + val |= XELPDP_LANE1_PHY_CLOCK_SELECT; + + val |= XELPDP_FORWARD_CLOCK_UNGATE; + + if (is_hdmi_frl(crtc_state->port_clock)) + val |= XELPDP_DDI_CLOCK_SELECT(XELPDP_DDI_CLOCK_SELECT_DIV18CLK); + else + val |= XELPDP_DDI_CLOCK_SELECT(XELPDP_DDI_CLOCK_SELECT_MAXPCLK); + + /* TODO: HDMI FRL */ + /* DP2.0 10G and 20G rates enable MPLLA*/ + if (crtc_state->port_clock == 1000000 || crtc_state->port_clock == 2000000) + val |= crtc_state->cx0pll_state.ssc_enabled ? XELPDP_SSC_ENABLE_PLLA : 0; + else + val |= crtc_state->cx0pll_state.ssc_enabled ? XELPDP_SSC_ENABLE_PLLB : 0; + + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + XELPDP_LANE1_PHY_CLOCK_SELECT | XELPDP_FORWARD_CLOCK_UNGATE | + XELPDP_DDI_CLOCK_SELECT_MASK | XELPDP_SSC_ENABLE_PLLB, val); +} + +static u32 intel_cx0_get_powerdown_update(u8 lane_mask) +{ + u32 val = 0; + int lane = 0; + + for_each_cx0_lane_in_mask(lane_mask, lane) + val |= XELPDP_LANE_POWERDOWN_UPDATE(lane); + + return val; +} + +static u32 intel_cx0_get_powerdown_state(u8 lane_mask, u8 state) +{ + u32 val = 0; + int lane = 0; + + for_each_cx0_lane_in_mask(lane_mask, lane) + val |= XELPDP_LANE_POWERDOWN_NEW_STATE(lane, state); + + return val; +} + +static void intel_cx0_powerdown_change_sequence(struct drm_i915_private *i915, + enum port port, + u8 lane_mask, u8 state) +{ + enum phy phy = intel_port_to_phy(i915, port); + int lane; + + intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port), + intel_cx0_get_powerdown_state(INTEL_CX0_BOTH_LANES, XELPDP_LANE_POWERDOWN_NEW_STATE_MASK), + intel_cx0_get_powerdown_state(lane_mask, state)); + + /* Wait for pending transactions.*/ + for_each_cx0_lane_in_mask(lane_mask, lane) + if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane), + XELPDP_PORT_M2P_TRANSACTION_PENDING, + XELPDP_MSGBUS_TIMEOUT_SLOW)) { + drm_dbg_kms(&i915->drm, + "PHY %c Timeout waiting for previous transaction to complete. Reset the bus.\n", + phy_name(phy)); + intel_cx0_bus_reset(i915, port, lane); + } + + intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port), + intel_cx0_get_powerdown_update(INTEL_CX0_BOTH_LANES), + intel_cx0_get_powerdown_update(lane_mask)); + + /* Update Timeout Value */ + if (__intel_de_wait_for_register(i915, XELPDP_PORT_BUF_CTL2(port), + intel_cx0_get_powerdown_update(lane_mask), 0, + XELPDP_PORT_POWERDOWN_UPDATE_TIMEOUT_US, 0, NULL)) + drm_warn(&i915->drm, "PHY %c failed to bring out of Lane reset after %dus.\n", + phy_name(phy), XELPDP_PORT_RESET_START_TIMEOUT_US); +} + +static void intel_cx0_setup_powerdown(struct drm_i915_private *i915, enum port port) +{ + intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port), + XELPDP_POWER_STATE_READY_MASK, + XELPDP_POWER_STATE_READY(CX0_P2_STATE_READY)); + intel_de_rmw(i915, XELPDP_PORT_BUF_CTL3(port), + XELPDP_POWER_STATE_ACTIVE_MASK | + XELPDP_PLL_LANE_STAGGERING_DELAY_MASK, + XELPDP_POWER_STATE_ACTIVE(CX0_P0_STATE_ACTIVE) | + XELPDP_PLL_LANE_STAGGERING_DELAY(0)); +} + +static u32 intel_cx0_get_pclk_refclk_request(u8 lane_mask) +{ + u32 val = 0; + int lane = 0; + + for_each_cx0_lane_in_mask(lane_mask, lane) + val |= XELPDP_LANE_PCLK_REFCLK_REQUEST(lane); + + return val; +} + +static u32 intel_cx0_get_pclk_refclk_ack(u8 lane_mask) +{ + u32 val = 0; + int lane = 0; + + for_each_cx0_lane_in_mask(lane_mask, lane) + val |= XELPDP_LANE_PCLK_REFCLK_ACK(lane); + + return val; +} + +static void intel_cx0_phy_lane_reset(struct drm_i915_private *i915, + struct intel_encoder *encoder, + bool lane_reversal) +{ + enum port port = encoder->port; + enum phy phy = intel_port_to_phy(i915, port); + bool both_lanes = intel_tc_port_fia_max_lane_count(enc_to_dig_port(encoder)) > 2; + u8 lane_mask = lane_reversal ? INTEL_CX0_LANE1 : + INTEL_CX0_LANE0; + u32 lane_pipe_reset = both_lanes ? + XELPDP_LANE_PIPE_RESET(0) | + XELPDP_LANE_PIPE_RESET(1) : + XELPDP_LANE_PIPE_RESET(0); + u32 lane_phy_current_status = both_lanes ? + XELPDP_LANE_PHY_CURRENT_STATUS(0) | + XELPDP_LANE_PHY_CURRENT_STATUS(1) : + XELPDP_LANE_PHY_CURRENT_STATUS(0); + + if (__intel_de_wait_for_register(i915, XELPDP_PORT_BUF_CTL1(port), + XELPDP_PORT_BUF_SOC_PHY_READY, + XELPDP_PORT_BUF_SOC_PHY_READY, + XELPDP_PORT_BUF_SOC_READY_TIMEOUT_US, 0, NULL)) + drm_warn(&i915->drm, "PHY %c failed to bring out of SOC reset after %dus.\n", + phy_name(phy), XELPDP_PORT_BUF_SOC_READY_TIMEOUT_US); + + intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port), + XELPDP_LANE_PIPE_RESET(0) | XELPDP_LANE_PIPE_RESET(1), + lane_pipe_reset); + + if (__intel_de_wait_for_register(i915, XELPDP_PORT_BUF_CTL2(port), + lane_phy_current_status, lane_phy_current_status, + XELPDP_PORT_RESET_START_TIMEOUT_US, 0, NULL)) + drm_warn(&i915->drm, "PHY %c failed to bring out of Lane reset after %dus.\n", + phy_name(phy), XELPDP_PORT_RESET_START_TIMEOUT_US); + + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(port), + intel_cx0_get_pclk_refclk_request(both_lanes ? + INTEL_CX0_BOTH_LANES : + INTEL_CX0_LANE0), + intel_cx0_get_pclk_refclk_request(lane_mask)); + + if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(port), + intel_cx0_get_pclk_refclk_ack(both_lanes ? + INTEL_CX0_BOTH_LANES : + INTEL_CX0_LANE0), + intel_cx0_get_pclk_refclk_ack(lane_mask), + XELPDP_REFCLK_ENABLE_TIMEOUT_US, 0, NULL)) + drm_warn(&i915->drm, "PHY %c failed to request refclk after %dus.\n", + phy_name(phy), XELPDP_REFCLK_ENABLE_TIMEOUT_US); + + intel_cx0_powerdown_change_sequence(i915, port, INTEL_CX0_BOTH_LANES, + CX0_P2_STATE_RESET); + intel_cx0_setup_powerdown(i915, port); + + intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port), lane_pipe_reset, 0); + + if (intel_de_wait_for_clear(i915, XELPDP_PORT_BUF_CTL2(port), lane_phy_current_status, + XELPDP_PORT_RESET_END_TIMEOUT)) + drm_warn(&i915->drm, "PHY %c failed to bring out of Lane reset after %dms.\n", + phy_name(phy), XELPDP_PORT_RESET_END_TIMEOUT); +} + +static void intel_cx0_program_phy_lane(struct drm_i915_private *i915, + struct intel_encoder *encoder, int lane_count, + bool lane_reversal) +{ + u8 l0t1, l0t2, l1t1, l1t2; + bool dp_alt_mode = intel_tc_port_in_dp_alt_mode(enc_to_dig_port(encoder)); + enum port port = encoder->port; + + if (intel_is_c10phy(i915, intel_port_to_phy(i915, port))) + intel_cx0_rmw(i915, port, INTEL_CX0_BOTH_LANES, + PHY_C10_VDR_CONTROL(1), 0, + C10_VDR_CTRL_MSGBUS_ACCESS, + MB_WRITE_COMMITTED); + + /* TODO: DP-alt MFD case where only one PHY lane should be programmed. */ + l0t1 = intel_cx0_read(i915, port, INTEL_CX0_LANE0, PHY_CX0_TX_CONTROL(1, 2)); + l0t2 = intel_cx0_read(i915, port, INTEL_CX0_LANE0, PHY_CX0_TX_CONTROL(2, 2)); + l1t1 = intel_cx0_read(i915, port, INTEL_CX0_LANE1, PHY_CX0_TX_CONTROL(1, 2)); + l1t2 = intel_cx0_read(i915, port, INTEL_CX0_LANE1, PHY_CX0_TX_CONTROL(2, 2)); + + l0t1 |= CONTROL2_DISABLE_SINGLE_TX; + l0t2 |= CONTROL2_DISABLE_SINGLE_TX; + l1t1 |= CONTROL2_DISABLE_SINGLE_TX; + l1t2 |= CONTROL2_DISABLE_SINGLE_TX; + + if (lane_reversal) { + switch (lane_count) { + case 4: + l0t1 &= ~CONTROL2_DISABLE_SINGLE_TX; + fallthrough; + case 3: + l0t2 &= ~CONTROL2_DISABLE_SINGLE_TX; + fallthrough; + case 2: + l1t1 &= ~CONTROL2_DISABLE_SINGLE_TX; + fallthrough; + case 1: + l1t2 &= ~CONTROL2_DISABLE_SINGLE_TX; + break; + default: + MISSING_CASE(lane_count); + } + } else { + switch (lane_count) { + case 4: + l1t2 &= ~CONTROL2_DISABLE_SINGLE_TX; + fallthrough; + case 3: + l1t1 &= ~CONTROL2_DISABLE_SINGLE_TX; + fallthrough; + case 2: + l0t2 &= ~CONTROL2_DISABLE_SINGLE_TX; + l0t1 &= ~CONTROL2_DISABLE_SINGLE_TX; + break; + case 1: + if (dp_alt_mode) + l0t2 &= ~CONTROL2_DISABLE_SINGLE_TX; + else + l0t1 &= ~CONTROL2_DISABLE_SINGLE_TX; + break; + default: + MISSING_CASE(lane_count); + } + } + + /* disable MLs */ + intel_cx0_write(i915, port, INTEL_CX0_LANE0, PHY_CX0_TX_CONTROL(1, 2), + l0t1, MB_WRITE_COMMITTED); + intel_cx0_write(i915, port, INTEL_CX0_LANE0, PHY_CX0_TX_CONTROL(2, 2), + l0t2, MB_WRITE_COMMITTED); + intel_cx0_write(i915, port, INTEL_CX0_LANE1, PHY_CX0_TX_CONTROL(1, 2), + l1t1, MB_WRITE_COMMITTED); + intel_cx0_write(i915, port, INTEL_CX0_LANE1, PHY_CX0_TX_CONTROL(2, 2), + l1t2, MB_WRITE_COMMITTED); + + if (intel_is_c10phy(i915, intel_port_to_phy(i915, port))) + intel_cx0_rmw(i915, port, INTEL_CX0_BOTH_LANES, + PHY_C10_VDR_CONTROL(1), 0, + C10_VDR_CTRL_UPDATE_CFG, + MB_WRITE_COMMITTED); +} + +static u32 intel_cx0_get_pclk_pll_request(u8 lane_mask) +{ + u32 val = 0; + int lane = 0; + + for_each_cx0_lane_in_mask(lane_mask, lane) + val |= XELPDP_LANE_PCLK_PLL_REQUEST(lane); + + return val; +} + +static u32 intel_cx0_get_pclk_pll_ack(u8 lane_mask) +{ + u32 val = 0; + int lane = 0; + + for_each_cx0_lane_in_mask(lane_mask, lane) + val |= XELPDP_LANE_PCLK_PLL_ACK(lane); + + return val; +} + +static void intel_cx0pll_enable(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + enum phy phy = intel_port_to_phy(i915, encoder->port); + struct intel_digital_port *dig_port = enc_to_dig_port(encoder); + bool lane_reversal = dig_port->saved_port_bits & DDI_BUF_PORT_REVERSAL; + u8 maxpclk_lane = lane_reversal ? INTEL_CX0_LANE1 : + INTEL_CX0_LANE0; + intel_wakeref_t wakeref = intel_cx0_phy_transaction_begin(encoder); + + /* + * 1. Program PORT_CLOCK_CTL REGISTER to configure + * clock muxes, gating and SSC + */ + intel_program_port_clock_ctl(encoder, crtc_state, lane_reversal); + + /* 2. Bring PHY out of reset. */ + intel_cx0_phy_lane_reset(i915, encoder, lane_reversal); + + /* + * 3. Change Phy power state to Ready. + * TODO: For DP alt mode use only one lane. + */ + intel_cx0_powerdown_change_sequence(i915, encoder->port, INTEL_CX0_BOTH_LANES, + CX0_P2_STATE_READY); + + /* 4. Program PHY internal PLL internal registers. */ + if (intel_is_c10phy(i915, phy)) + intel_c10_pll_program(i915, crtc_state, encoder); + else + intel_c20_pll_program(i915, crtc_state, encoder); + + /* + * 5. Program the enabled and disabled owned PHY lane + * transmitters over message bus + */ + intel_cx0_program_phy_lane(i915, encoder, crtc_state->lane_count, lane_reversal); + + /* + * 6. Follow the Display Voltage Frequency Switching - Sequence + * Before Frequency Change. We handle this step in bxt_set_cdclk(). + */ + + /* + * 7. Program DDI_CLK_VALFREQ to match intended DDI + * clock frequency. + */ + intel_de_write(i915, DDI_CLK_VALFREQ(encoder->port), + crtc_state->port_clock); + + /* + * 8. Set PORT_CLOCK_CTL register PCLK PLL Request + * LN to "1" to enable PLL. + */ + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + intel_cx0_get_pclk_pll_request(INTEL_CX0_BOTH_LANES), + intel_cx0_get_pclk_pll_request(maxpclk_lane)); + + /* 9. Poll on PORT_CLOCK_CTL PCLK PLL Ack LN == "1". */ + if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + intel_cx0_get_pclk_pll_ack(INTEL_CX0_BOTH_LANES), + intel_cx0_get_pclk_pll_ack(maxpclk_lane), + XELPDP_PCLK_PLL_ENABLE_TIMEOUT_US, 0, NULL)) + drm_warn(&i915->drm, "Port %c PLL not locked after %dus.\n", + phy_name(phy), XELPDP_PCLK_PLL_ENABLE_TIMEOUT_US); + + /* + * 10. Follow the Display Voltage Frequency Switching Sequence After + * Frequency Change. We handle this step in bxt_set_cdclk(). + */ + + /* TODO: enable TBT-ALT mode */ + intel_cx0_phy_transaction_end(encoder, wakeref); +} + +int intel_mtl_tbt_calc_port_clock(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + u32 clock; + u32 val = intel_de_read(i915, XELPDP_PORT_CLOCK_CTL(encoder->port)); + + clock = REG_FIELD_GET(XELPDP_DDI_CLOCK_SELECT_MASK, val); + + drm_WARN_ON(&i915->drm, !(val & XELPDP_FORWARD_CLOCK_UNGATE)); + drm_WARN_ON(&i915->drm, !(val & XELPDP_TBT_CLOCK_REQUEST)); + drm_WARN_ON(&i915->drm, !(val & XELPDP_TBT_CLOCK_ACK)); + + switch (clock) { + case XELPDP_DDI_CLOCK_SELECT_TBT_162: + return 162000; + case XELPDP_DDI_CLOCK_SELECT_TBT_270: + return 270000; + case XELPDP_DDI_CLOCK_SELECT_TBT_540: + return 540000; + case XELPDP_DDI_CLOCK_SELECT_TBT_810: + return 810000; + default: + MISSING_CASE(clock); + return 162000; + } +} + +static int intel_mtl_tbt_clock_select(struct drm_i915_private *i915, int clock) +{ + switch (clock) { + case 162000: + return XELPDP_DDI_CLOCK_SELECT_TBT_162; + case 270000: + return XELPDP_DDI_CLOCK_SELECT_TBT_270; + case 540000: + return XELPDP_DDI_CLOCK_SELECT_TBT_540; + case 810000: + return XELPDP_DDI_CLOCK_SELECT_TBT_810; + default: + MISSING_CASE(clock); + return XELPDP_DDI_CLOCK_SELECT_TBT_162; + } +} + +static void intel_mtl_tbt_pll_enable(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + enum phy phy = intel_port_to_phy(i915, encoder->port); + u32 val = 0; + + /* + * 1. Program PORT_CLOCK_CTL REGISTER to configure + * clock muxes, gating and SSC + */ + val |= XELPDP_DDI_CLOCK_SELECT(intel_mtl_tbt_clock_select(i915, crtc_state->port_clock)); + val |= XELPDP_FORWARD_CLOCK_UNGATE; + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + XELPDP_DDI_CLOCK_SELECT_MASK | XELPDP_FORWARD_CLOCK_UNGATE, val); + + /* 2. Read back PORT_CLOCK_CTL REGISTER */ + val = intel_de_read(i915, XELPDP_PORT_CLOCK_CTL(encoder->port)); + + /* + * 3. Follow the Display Voltage Frequency Switching - Sequence + * Before Frequency Change. We handle this step in bxt_set_cdclk(). + */ + + /* + * 4. Set PORT_CLOCK_CTL register TBT CLOCK Request to "1" to enable PLL. + */ + val |= XELPDP_TBT_CLOCK_REQUEST; + intel_de_write(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), val); + + /* 5. Poll on PORT_CLOCK_CTL TBT CLOCK Ack == "1". */ + if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + XELPDP_TBT_CLOCK_ACK, + XELPDP_TBT_CLOCK_ACK, + 100, 0, NULL)) + drm_warn(&i915->drm, "[ENCODER:%d:%s][%c] PHY PLL not locked after 100us.\n", + encoder->base.base.id, encoder->base.name, phy_name(phy)); + + /* + * 6. Follow the Display Voltage Frequency Switching Sequence After + * Frequency Change. We handle this step in bxt_set_cdclk(). + */ + + /* + * 7. Program DDI_CLK_VALFREQ to match intended DDI + * clock frequency. + */ + intel_de_write(i915, DDI_CLK_VALFREQ(encoder->port), + crtc_state->port_clock); +} + +void intel_mtl_pll_enable(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) +{ + struct intel_digital_port *dig_port = enc_to_dig_port(encoder); + + if (intel_tc_port_in_tbt_alt_mode(dig_port)) + intel_mtl_tbt_pll_enable(encoder, crtc_state); + else + intel_cx0pll_enable(encoder, crtc_state); +} + +static void intel_cx0pll_disable(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + enum phy phy = intel_port_to_phy(i915, encoder->port); + bool is_c10 = intel_is_c10phy(i915, phy); + intel_wakeref_t wakeref = intel_cx0_phy_transaction_begin(encoder); + + /* 1. Change owned PHY lane power to Disable state. */ + intel_cx0_powerdown_change_sequence(i915, encoder->port, INTEL_CX0_BOTH_LANES, + is_c10 ? CX0_P2PG_STATE_DISABLE : + CX0_P4PG_STATE_DISABLE); + + /* + * 2. Follow the Display Voltage Frequency Switching Sequence Before + * Frequency Change. We handle this step in bxt_set_cdclk(). + */ + + /* + * 3. Set PORT_CLOCK_CTL register PCLK PLL Request LN + * to "0" to disable PLL. + */ + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + intel_cx0_get_pclk_pll_request(INTEL_CX0_BOTH_LANES) | + intel_cx0_get_pclk_refclk_request(INTEL_CX0_BOTH_LANES), 0); + + /* 4. Program DDI_CLK_VALFREQ to 0. */ + intel_de_write(i915, DDI_CLK_VALFREQ(encoder->port), 0); + + /* + * 5. Poll on PORT_CLOCK_CTL PCLK PLL Ack LN == "0". + */ + if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + intel_cx0_get_pclk_pll_ack(INTEL_CX0_BOTH_LANES) | + intel_cx0_get_pclk_refclk_ack(INTEL_CX0_BOTH_LANES), 0, + XELPDP_PCLK_PLL_DISABLE_TIMEOUT_US, 0, NULL)) + drm_warn(&i915->drm, "Port %c PLL not unlocked after %dus.\n", + phy_name(phy), XELPDP_PCLK_PLL_DISABLE_TIMEOUT_US); + + /* + * 6. Follow the Display Voltage Frequency Switching Sequence After + * Frequency Change. We handle this step in bxt_set_cdclk(). + */ + + /* 7. Program PORT_CLOCK_CTL register to disable and gate clocks. */ + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + XELPDP_DDI_CLOCK_SELECT_MASK, 0); + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + XELPDP_FORWARD_CLOCK_UNGATE, 0); + + intel_cx0_phy_transaction_end(encoder, wakeref); +} + +static void intel_mtl_tbt_pll_disable(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + enum phy phy = intel_port_to_phy(i915, encoder->port); + + /* + * 1. Follow the Display Voltage Frequency Switching Sequence Before + * Frequency Change. We handle this step in bxt_set_cdclk(). + */ + + /* + * 2. Set PORT_CLOCK_CTL register TBT CLOCK Request to "0" to disable PLL. + */ + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + XELPDP_TBT_CLOCK_REQUEST, 0); + + /* 3. Poll on PORT_CLOCK_CTL TBT CLOCK Ack == "0". */ + if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + XELPDP_TBT_CLOCK_ACK, 0, 10, 0, NULL)) + drm_warn(&i915->drm, "[ENCODER:%d:%s][%c] PHY PLL not unlocked after 10us.\n", + encoder->base.base.id, encoder->base.name, phy_name(phy)); + + /* + * 4. Follow the Display Voltage Frequency Switching Sequence After + * Frequency Change. We handle this step in bxt_set_cdclk(). + */ + + /* + * 5. Program PORT CLOCK CTRL register to disable and gate clocks + */ + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + XELPDP_DDI_CLOCK_SELECT_MASK | + XELPDP_FORWARD_CLOCK_UNGATE, 0); + + /* 6. Program DDI_CLK_VALFREQ to 0. */ + intel_de_write(i915, DDI_CLK_VALFREQ(encoder->port), 0); +} + +void intel_mtl_pll_disable(struct intel_encoder *encoder) +{ + struct intel_digital_port *dig_port = enc_to_dig_port(encoder); + + if (intel_tc_port_in_tbt_alt_mode(dig_port)) + intel_mtl_tbt_pll_disable(encoder); + else + intel_cx0pll_disable(encoder); +} + +enum icl_port_dpll_id +intel_mtl_port_pll_type(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + /* + * TODO: Determine the PLL type from the SW state, once MTL PLL + * handling is done via the standard shared DPLL framework. + */ + u32 val = intel_de_read(i915, XELPDP_PORT_CLOCK_CTL(encoder->port)); + u32 clock = REG_FIELD_GET(XELPDP_DDI_CLOCK_SELECT_MASK, val); + + if (clock == XELPDP_DDI_CLOCK_SELECT_MAXPCLK || + clock == XELPDP_DDI_CLOCK_SELECT_DIV18CLK) + return ICL_PORT_DPLL_MG_PHY; + else + return ICL_PORT_DPLL_DEFAULT; +} + +void intel_c10pll_state_verify(struct intel_atomic_state *state, + struct intel_crtc_state *new_crtc_state) +{ + struct drm_i915_private *i915 = to_i915(state->base.dev); + struct intel_c10pll_state mpllb_hw_state = { 0 }; + struct intel_c10pll_state *mpllb_sw_state = &new_crtc_state->cx0pll_state.c10; + struct intel_crtc *crtc = to_intel_crtc(new_crtc_state->uapi.crtc); + struct intel_encoder *encoder; + enum phy phy; + int i; + + if (DISPLAY_VER(i915) < 14) + return; + + if (!new_crtc_state->hw.active) + return; + + /* intel_get_crtc_new_encoder() only works for modeset/fastset commits */ + if (!intel_crtc_needs_modeset(new_crtc_state) && + !intel_crtc_needs_fastset(new_crtc_state)) + return; + + encoder = intel_get_crtc_new_encoder(state, new_crtc_state); + phy = intel_port_to_phy(i915, encoder->port); + + if (!intel_is_c10phy(i915, phy)) + return; + + intel_c10pll_readout_hw_state(encoder, &mpllb_hw_state); + + for (i = 0; i < ARRAY_SIZE(mpllb_sw_state->pll); i++) { + u8 expected = mpllb_sw_state->pll[i]; + + I915_STATE_WARN(i915, mpllb_hw_state.pll[i] != expected, + "[CRTC:%d:%s] mismatch in C10MPLLB: Register[%d] (expected 0x%02x, found 0x%02x)", + crtc->base.base.id, crtc->base.name, i, + expected, mpllb_hw_state.pll[i]); + } + + I915_STATE_WARN(i915, mpllb_hw_state.tx != mpllb_sw_state->tx, + "[CRTC:%d:%s] mismatch in C10MPLLB: Register TX0 (expected 0x%02x, found 0x%02x)", + crtc->base.base.id, crtc->base.name, + mpllb_sw_state->tx, mpllb_hw_state.tx); + + I915_STATE_WARN(i915, mpllb_hw_state.cmn != mpllb_sw_state->cmn, + "[CRTC:%d:%s] mismatch in C10MPLLB: Register CMN0 (expected 0x%02x, found 0x%02x)", + crtc->base.base.id, crtc->base.name, + mpllb_sw_state->cmn, mpllb_hw_state.cmn); +} diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.h b/drivers/gpu/drm/i915/display/intel_cx0_phy.h new file mode 100644 index 0000000000000000000000000000000000000000..f99809af257dc385b525ffa4f3a5b51040232eb7 --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.h @@ -0,0 +1,50 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2023 Intel Corporation + */ + +#ifndef __INTEL_CX0_PHY_H__ +#define __INTEL_CX0_PHY_H__ + +#include +#include +#include + +#include "i915_drv.h" +#include "intel_display_types.h" + +struct drm_i915_private; +struct intel_encoder; +struct intel_crtc_state; +enum icl_port_dpll_id; +enum phy; + +bool intel_is_c10phy(struct drm_i915_private *dev_priv, enum phy phy); +void intel_mtl_pll_enable(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state); +void intel_mtl_pll_disable(struct intel_encoder *encoder); +enum icl_port_dpll_id +intel_mtl_port_pll_type(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state); +void intel_c10pll_readout_hw_state(struct intel_encoder *encoder, struct intel_c10pll_state *pll_state); +int intel_cx0pll_calc_state(struct intel_crtc_state *crtc_state, struct intel_encoder *encoder); +void intel_c10pll_dump_hw_state(struct drm_i915_private *dev_priv, + const struct intel_c10pll_state *hw_state); +int intel_c10pll_calc_port_clock(struct intel_encoder *encoder, + const struct intel_c10pll_state *pll_state); +void intel_c10pll_state_verify(struct intel_atomic_state *state, + struct intel_crtc_state *new_crtc_state); +void intel_c20pll_readout_hw_state(struct intel_encoder *encoder, + struct intel_c20pll_state *pll_state); +void intel_c20pll_dump_hw_state(struct drm_i915_private *i915, + const struct intel_c20pll_state *hw_state); +int intel_c20pll_calc_port_clock(struct intel_encoder *encoder, + const struct intel_c20pll_state *pll_state); +void intel_cx0_phy_set_signal_levels(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state); +int intel_cx0_phy_check_hdmi_link_rate(struct intel_hdmi *hdmi, int clock); +void intel_cx0_phy_ddi_vswing_sequence(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state, + u32 level); +int intel_mtl_tbt_calc_port_clock(struct intel_encoder *encoder); +#endif /* __INTEL_CX0_PHY_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h new file mode 100644 index 0000000000000000000000000000000000000000..cb5d1be2ba191e9629c02682cc77897d9381294f --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h @@ -0,0 +1,274 @@ +/* SPDX-License-Identifier: MIT + * + * Copyright © 2023 Intel Corporation + */ + +#ifndef __INTEL_CX0_PHY_REGS_H__ +#define __INTEL_CX0_PHY_REGS_H__ + +#include "i915_reg_defs.h" + +#define _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_A 0x64040 +#define _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_B 0x64140 +#define _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC1 0x16F240 +#define _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC2 0x16F440 +#define XELPDP_PORT_M2P_MSGBUS_CTL(port, lane) _MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \ + _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_A, \ + _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_B, \ + _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC1, \ + _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC2) + (lane) * 4) +#define XELPDP_PORT_M2P_TRANSACTION_PENDING REG_BIT(31) +#define XELPDP_PORT_M2P_COMMAND_TYPE_MASK REG_GENMASK(30, 27) +#define XELPDP_PORT_M2P_COMMAND_WRITE_UNCOMMITTED REG_FIELD_PREP(XELPDP_PORT_M2P_COMMAND_TYPE_MASK, 0x1) +#define XELPDP_PORT_M2P_COMMAND_WRITE_COMMITTED REG_FIELD_PREP(XELPDP_PORT_M2P_COMMAND_TYPE_MASK, 0x2) +#define XELPDP_PORT_M2P_COMMAND_READ REG_FIELD_PREP(XELPDP_PORT_M2P_COMMAND_TYPE_MASK, 0x3) +#define XELPDP_PORT_M2P_DATA_MASK REG_GENMASK(23, 16) +#define XELPDP_PORT_M2P_DATA(val) REG_FIELD_PREP(XELPDP_PORT_M2P_DATA_MASK, val) +#define XELPDP_PORT_M2P_TRANSACTION_RESET REG_BIT(15) +#define XELPDP_PORT_M2P_ADDRESS_MASK REG_GENMASK(11, 0) +#define XELPDP_PORT_M2P_ADDRESS(val) REG_FIELD_PREP(XELPDP_PORT_M2P_ADDRESS_MASK, val) +#define XELPDP_PORT_P2M_MSGBUS_STATUS(port, lane) _MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \ + _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_A, \ + _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_B, \ + _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC1, \ + _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC2) + (lane) * 4 + 8) +#define XELPDP_PORT_P2M_RESPONSE_READY REG_BIT(31) +#define XELPDP_PORT_P2M_COMMAND_TYPE_MASK REG_GENMASK(30, 27) +#define XELPDP_PORT_P2M_COMMAND_READ_ACK 0x4 +#define XELPDP_PORT_P2M_COMMAND_WRITE_ACK 0x5 +#define XELPDP_PORT_P2M_DATA_MASK REG_GENMASK(23, 16) +#define XELPDP_PORT_P2M_DATA(val) REG_FIELD_PREP(XELPDP_PORT_P2M_DATA_MASK, val) +#define XELPDP_PORT_P2M_ERROR_SET REG_BIT(15) + +#define XELPDP_MSGBUS_TIMEOUT_SLOW 1 +#define XELPDP_MSGBUS_TIMEOUT_FAST_US 2 +#define XELPDP_PCLK_PLL_ENABLE_TIMEOUT_US 3200 +#define XELPDP_PCLK_PLL_DISABLE_TIMEOUT_US 20 +#define XELPDP_PORT_BUF_SOC_READY_TIMEOUT_US 100 +#define XELPDP_PORT_RESET_START_TIMEOUT_US 5 +#define XELPDP_PORT_POWERDOWN_UPDATE_TIMEOUT_US 100 +#define XELPDP_PORT_RESET_END_TIMEOUT 15 +#define XELPDP_REFCLK_ENABLE_TIMEOUT_US 1 + +#define _XELPDP_PORT_BUF_CTL1_LN0_A 0x64004 +#define _XELPDP_PORT_BUF_CTL1_LN0_B 0x64104 +#define _XELPDP_PORT_BUF_CTL1_LN0_USBC1 0x16F200 +#define _XELPDP_PORT_BUF_CTL1_LN0_USBC2 0x16F400 +#define XELPDP_PORT_BUF_CTL1(port) _MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \ + _XELPDP_PORT_BUF_CTL1_LN0_A, \ + _XELPDP_PORT_BUF_CTL1_LN0_B, \ + _XELPDP_PORT_BUF_CTL1_LN0_USBC1, \ + _XELPDP_PORT_BUF_CTL1_LN0_USBC2)) +#define XELPDP_PORT_BUF_D2D_LINK_ENABLE REG_BIT(29) +#define XELPDP_PORT_BUF_D2D_LINK_STATE REG_BIT(28) +#define XELPDP_PORT_BUF_SOC_PHY_READY REG_BIT(24) +#define XELPDP_PORT_BUF_PORT_DATA_WIDTH_MASK REG_GENMASK(19, 18) +#define XELPDP_PORT_BUF_PORT_DATA_10BIT REG_FIELD_PREP(XELPDP_PORT_BUF_PORT_DATA_WIDTH_MASK, 0) +#define XELPDP_PORT_BUF_PORT_DATA_20BIT REG_FIELD_PREP(XELPDP_PORT_BUF_PORT_DATA_WIDTH_MASK, 1) +#define XELPDP_PORT_BUF_PORT_DATA_40BIT REG_FIELD_PREP(XELPDP_PORT_BUF_PORT_DATA_WIDTH_MASK, 2) +#define XELPDP_PORT_REVERSAL REG_BIT(16) +#define XELPDP_PORT_BUF_IO_SELECT_TBT REG_BIT(11) +#define XELPDP_PORT_BUF_PHY_IDLE REG_BIT(7) +#define XELPDP_TC_PHY_OWNERSHIP REG_BIT(6) +#define XELPDP_TCSS_POWER_REQUEST REG_BIT(5) +#define XELPDP_TCSS_POWER_STATE REG_BIT(4) +#define XELPDP_PORT_WIDTH_MASK REG_GENMASK(3, 1) +#define XELPDP_PORT_WIDTH(val) REG_FIELD_PREP(XELPDP_PORT_WIDTH_MASK, val) + +#define XELPDP_PORT_BUF_CTL2(port) _MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \ + _XELPDP_PORT_BUF_CTL1_LN0_A, \ + _XELPDP_PORT_BUF_CTL1_LN0_B, \ + _XELPDP_PORT_BUF_CTL1_LN0_USBC1, \ + _XELPDP_PORT_BUF_CTL1_LN0_USBC2) + 4) + +#define XELPDP_LANE_PIPE_RESET(lane) _PICK(lane, REG_BIT(31), REG_BIT(30)) +#define XELPDP_LANE_PHY_CURRENT_STATUS(lane) _PICK(lane, REG_BIT(29), REG_BIT(28)) +#define XELPDP_LANE_POWERDOWN_UPDATE(lane) _PICK(lane, REG_BIT(25), REG_BIT(24)) +#define _XELPDP_LANE0_POWERDOWN_NEW_STATE_MASK REG_GENMASK(23, 20) +#define _XELPDP_LANE0_POWERDOWN_NEW_STATE(val) REG_FIELD_PREP(_XELPDP_LANE0_POWERDOWN_NEW_STATE_MASK, val) +#define _XELPDP_LANE1_POWERDOWN_NEW_STATE_MASK REG_GENMASK(19, 16) +#define _XELPDP_LANE1_POWERDOWN_NEW_STATE(val) REG_FIELD_PREP(_XELPDP_LANE1_POWERDOWN_NEW_STATE_MASK, val) +#define XELPDP_LANE_POWERDOWN_NEW_STATE(lane, val) _PICK(lane, \ + _XELPDP_LANE0_POWERDOWN_NEW_STATE(val), \ + _XELPDP_LANE1_POWERDOWN_NEW_STATE(val)) +#define XELPDP_LANE_POWERDOWN_NEW_STATE_MASK REG_GENMASK(3, 0) +#define XELPDP_POWER_STATE_READY_MASK REG_GENMASK(7, 4) +#define XELPDP_POWER_STATE_READY(val) REG_FIELD_PREP(XELPDP_POWER_STATE_READY_MASK, val) + +#define XELPDP_PORT_BUF_CTL3(port) _MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \ + _XELPDP_PORT_BUF_CTL1_LN0_A, \ + _XELPDP_PORT_BUF_CTL1_LN0_B, \ + _XELPDP_PORT_BUF_CTL1_LN0_USBC1, \ + _XELPDP_PORT_BUF_CTL1_LN0_USBC2) + 8) +#define XELPDP_PLL_LANE_STAGGERING_DELAY_MASK REG_GENMASK(15, 8) +#define XELPDP_PLL_LANE_STAGGERING_DELAY(val) REG_FIELD_PREP(XELPDP_PLL_LANE_STAGGERING_DELAY_MASK, val) +#define XELPDP_POWER_STATE_ACTIVE_MASK REG_GENMASK(3, 0) +#define XELPDP_POWER_STATE_ACTIVE(val) REG_FIELD_PREP(XELPDP_POWER_STATE_ACTIVE_MASK, val) +#define CX0_P0_STATE_ACTIVE 0x0 +#define CX0_P2_STATE_READY 0x2 +#define CX0_P2PG_STATE_DISABLE 0x9 +#define CX0_P4PG_STATE_DISABLE 0xC +#define CX0_P2_STATE_RESET 0x2 + +#define _XELPDP_PORT_CLOCK_CTL_A 0x640E0 +#define _XELPDP_PORT_CLOCK_CTL_B 0x641E0 +#define _XELPDP_PORT_CLOCK_CTL_USBC1 0x16F260 +#define _XELPDP_PORT_CLOCK_CTL_USBC2 0x16F460 +#define XELPDP_PORT_CLOCK_CTL(port) _MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \ + _XELPDP_PORT_CLOCK_CTL_A, \ + _XELPDP_PORT_CLOCK_CTL_B, \ + _XELPDP_PORT_CLOCK_CTL_USBC1, \ + _XELPDP_PORT_CLOCK_CTL_USBC2)) +#define XELPDP_LANE_PCLK_PLL_REQUEST(lane) REG_BIT(31 - ((lane) * 4)) +#define XELPDP_LANE_PCLK_PLL_ACK(lane) REG_BIT(30 - ((lane) * 4)) +#define XELPDP_LANE_PCLK_REFCLK_REQUEST(lane) REG_BIT(29 - ((lane) * 4)) +#define XELPDP_LANE_PCLK_REFCLK_ACK(lane) REG_BIT(28 - ((lane) * 4)) + +#define XELPDP_TBT_CLOCK_REQUEST REG_BIT(19) +#define XELPDP_TBT_CLOCK_ACK REG_BIT(18) +#define XELPDP_DDI_CLOCK_SELECT_MASK REG_GENMASK(15, 12) +#define XELPDP_DDI_CLOCK_SELECT(val) REG_FIELD_PREP(XELPDP_DDI_CLOCK_SELECT_MASK, val) +#define XELPDP_DDI_CLOCK_SELECT_NONE 0x0 +#define XELPDP_DDI_CLOCK_SELECT_MAXPCLK 0x8 +#define XELPDP_DDI_CLOCK_SELECT_DIV18CLK 0x9 +#define XELPDP_DDI_CLOCK_SELECT_TBT_162 0xc +#define XELPDP_DDI_CLOCK_SELECT_TBT_270 0xd +#define XELPDP_DDI_CLOCK_SELECT_TBT_540 0xe +#define XELPDP_DDI_CLOCK_SELECT_TBT_810 0xf +#define XELPDP_FORWARD_CLOCK_UNGATE REG_BIT(10) +#define XELPDP_LANE1_PHY_CLOCK_SELECT REG_BIT(8) +#define XELPDP_SSC_ENABLE_PLLA REG_BIT(1) +#define XELPDP_SSC_ENABLE_PLLB REG_BIT(0) + +/* C10 Vendor Registers */ +#define PHY_C10_VDR_PLL(idx) (0xC00 + (idx)) +#define C10_PLL0_FRACEN REG_BIT8(4) +#define C10_PLL3_MULTIPLIERH_MASK REG_GENMASK8(3, 0) +#define C10_PLL15_TXCLKDIV_MASK REG_GENMASK8(2, 0) +#define C10_PLL15_HDMIDIV_MASK REG_GENMASK8(5, 3) + +#define PHY_C10_VDR_CMN(idx) (0xC20 + (idx)) +#define C10_CMN0_REF_RANGE REG_FIELD_PREP(REG_GENMASK(4, 0), 1) +#define C10_CMN0_REF_CLK_MPLLB_DIV REG_FIELD_PREP(REG_GENMASK(7, 5), 1) +#define C10_CMN3_TXVBOOST_MASK REG_GENMASK8(7, 5) +#define C10_CMN3_TXVBOOST(val) REG_FIELD_PREP8(C10_CMN3_TXVBOOST_MASK, val) +#define PHY_C10_VDR_TX(idx) (0xC30 + (idx)) +#define C10_TX0_TX_MPLLB_SEL REG_BIT(4) +#define C10_TX1_TERMCTL_MASK REG_GENMASK8(7, 5) +#define C10_TX1_TERMCTL(val) REG_FIELD_PREP8(C10_TX1_TERMCTL_MASK, val) +#define PHY_C10_VDR_CONTROL(idx) (0xC70 + (idx) - 1) +#define C10_VDR_CTRL_MSGBUS_ACCESS REG_BIT8(2) +#define C10_VDR_CTRL_MASTER_LANE REG_BIT8(1) +#define C10_VDR_CTRL_UPDATE_CFG REG_BIT8(0) +#define PHY_C10_VDR_CUSTOM_WIDTH 0xD02 +#define C10_VDR_CUSTOM_WIDTH_MASK REG_GENMASK(1, 0) +#define C10_VDR_CUSTOM_WIDTH_8_10 REG_FIELD_PREP(C10_VDR_CUSTOM_WIDTH_MASK, 0) +#define PHY_C10_VDR_OVRD 0xD71 +#define PHY_C10_VDR_OVRD_TX1 REG_BIT8(0) +#define PHY_C10_VDR_OVRD_TX2 REG_BIT8(2) +#define PHY_C10_VDR_PRE_OVRD_TX1 0xD80 +#define C10_PHY_OVRD_LEVEL_MASK REG_GENMASK8(5, 0) +#define C10_PHY_OVRD_LEVEL(val) REG_FIELD_PREP8(C10_PHY_OVRD_LEVEL_MASK, val) +#define PHY_CX0_VDROVRD_CTL(lane, tx, control) \ + (PHY_C10_VDR_PRE_OVRD_TX1 + \ + ((lane) ^ (tx)) * 0x10 + (control)) + +/* PIPE SPEC Defined Registers */ +#define PHY_CX0_TX_CONTROL(tx, control) (0x400 + ((tx) - 1) * 0x200 + (control)) +#define CONTROL2_DISABLE_SINGLE_TX REG_BIT(6) + +/* C20 Registers */ +#define PHY_C20_WR_ADDRESS_L 0xC02 +#define PHY_C20_WR_ADDRESS_H 0xC03 +#define PHY_C20_WR_DATA_L 0xC04 +#define PHY_C20_WR_DATA_H 0xC05 +#define PHY_C20_RD_ADDRESS_L 0xC06 +#define PHY_C20_RD_ADDRESS_H 0xC07 +#define PHY_C20_RD_DATA_L 0xC08 +#define PHY_C20_RD_DATA_H 0xC09 +#define PHY_C20_VDR_CUSTOM_SERDES_RATE 0xD00 +#define PHY_C20_VDR_HDMI_RATE 0xD01 +#define PHY_C20_CONTEXT_TOGGLE REG_BIT8(0) +#define PHY_C20_CUSTOM_SERDES_MASK REG_GENMASK8(4, 1) +#define PHY_C20_CUSTOM_SERDES(val) REG_FIELD_PREP8(PHY_C20_CUSTOM_SERDES_MASK, val) +#define PHY_C20_VDR_CUSTOM_WIDTH 0xD02 +#define PHY_C20_CUSTOM_WIDTH_MASK REG_GENMASK(1, 0) +#define PHY_C20_CUSTOM_WIDTH(val) REG_FIELD_PREP8(PHY_C20_CUSTOM_WIDTH_MASK, val) +#define PHY_C20_A_TX_CNTX_CFG(idx) (0xCF2E - (idx)) +#define PHY_C20_B_TX_CNTX_CFG(idx) (0xCF2A - (idx)) +#define C20_PHY_TX_RATE REG_GENMASK(2, 0) +#define PHY_C20_A_CMN_CNTX_CFG(idx) (0xCDAA - (idx)) +#define PHY_C20_B_CMN_CNTX_CFG(idx) (0xCDA5 - (idx)) +#define PHY_C20_A_MPLLA_CNTX_CFG(idx) (0xCCF0 - (idx)) +#define PHY_C20_B_MPLLA_CNTX_CFG(idx) (0xCCE5 - (idx)) +#define C20_MPLLA_FRACEN REG_BIT(14) +#define C20_FB_CLK_DIV4_EN REG_BIT(13) +#define C20_MPLLA_TX_CLK_DIV_MASK REG_GENMASK(10, 8) +#define PHY_C20_A_MPLLB_CNTX_CFG(idx) (0xCB5A - (idx)) +#define PHY_C20_B_MPLLB_CNTX_CFG(idx) (0xCB4E - (idx)) +#define C20_MPLLB_TX_CLK_DIV_MASK REG_GENMASK(15, 13) +#define C20_MPLLB_FRACEN REG_BIT(13) +#define C20_REF_CLK_MPLLB_DIV_MASK REG_GENMASK(12, 10) +#define C20_MULTIPLIER_MASK REG_GENMASK(11, 0) +#define C20_PHY_USE_MPLLB REG_BIT(7) + +/* C20 Phy VSwing Masks */ +#define C20_PHY_VSWING_PREEMPH_MASK REG_GENMASK8(5, 0) +#define C20_PHY_VSWING_PREEMPH(val) REG_FIELD_PREP8(C20_PHY_VSWING_PREEMPH_MASK, val) + +#define RAWLANEAONX_DIG_TX_MPLLB_CAL_DONE_BANK(idx) (0x303D + (idx)) + +/* C20 HDMI computed pll definitions */ +#define REFCLK_38_4_MHZ 38400000 +#define CLOCK_4999MHZ 4999999999 +#define CLOCK_9999MHZ 9999999999 +#define DATARATE_3000000000 3000000000 +#define DATARATE_3500000000 3500000000 +#define DATARATE_4000000000 4000000000 +#define MPLL_FRACN_DEN 0xFFFF + +#define SSC_UP_SPREAD REG_BIT16(9) +#define WORD_CLK_DIV REG_BIT16(8) + +#define MPLL_TX_CLK_DIV(val) REG_FIELD_PREP16(C20_MPLLB_TX_CLK_DIV_MASK, val) +#define MPLL_MULTIPLIER(val) REG_FIELD_PREP16(C20_MULTIPLIER_MASK, val) + +#define MPLLB_ANA_FREQ_VCO_0 0 +#define MPLLB_ANA_FREQ_VCO_1 1 +#define MPLLB_ANA_FREQ_VCO_2 2 +#define MPLLB_ANA_FREQ_VCO_3 3 +#define MPLLB_ANA_FREQ_VCO_MASK REG_GENMASK16(15, 14) +#define MPLLB_ANA_FREQ_VCO(val) REG_FIELD_PREP16(MPLLB_ANA_FREQ_VCO_MASK, val) + +#define MPLL_DIV_MULTIPLIER_MASK REG_GENMASK16(7, 0) +#define MPLL_DIV_MULTIPLIER(val) REG_FIELD_PREP16(MPLL_DIV_MULTIPLIER_MASK, val) + +#define CAL_DAC_CODE_31 31 +#define CAL_DAC_CODE_MASK REG_GENMASK16(14, 10) +#define CAL_DAC_CODE(val) REG_FIELD_PREP16(CAL_DAC_CODE_MASK, val) + +#define CP_INT_GS_28 28 +#define CP_INT_GS_MASK REG_GENMASK16(6, 0) +#define CP_INT_GS(val) REG_FIELD_PREP16(CP_INT_GS_MASK, val) + +#define CP_PROP_GS_30 30 +#define CP_PROP_GS_MASK REG_GENMASK16(13, 7) +#define CP_PROP_GS(val) REG_FIELD_PREP16(CP_PROP_GS_MASK, val) + +#define CP_INT_6 6 +#define CP_INT_MASK REG_GENMASK16(6, 0) +#define CP_INT(val) REG_FIELD_PREP16(CP_INT_MASK, val) + +#define CP_PROP_20 20 +#define CP_PROP_MASK REG_GENMASK16(13, 7) +#define CP_PROP(val) REG_FIELD_PREP16(CP_PROP_MASK, val) + +#define V2I_2 2 +#define V2I_MASK REG_GENMASK16(15, 14) +#define V2I(val) REG_FIELD_PREP16(V2I_MASK, val) + +#define HDMI_DIV_1 1 +#define HDMI_DIV_MASK REG_GENMASK16(2, 0) +#define HDMI_DIV(val) REG_FIELD_PREP16(HDMI_DIV_MASK, val) + +#endif /* __INTEL_CX0_REG_DEFS_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c index 3a7b988375164ff8e8567a4f159a7ecb2e29840e..090f242e610c54c257e74d0715c7796680314d24 100644 --- a/drivers/gpu/drm/i915/display/intel_ddi.c +++ b/drivers/gpu/drm/i915/display/intel_ddi.c @@ -39,6 +39,8 @@ #include "intel_combo_phy_regs.h" #include "intel_connector.h" #include "intel_crtc.h" +#include "intel_cx0_phy.h" +#include "intel_cx0_phy_regs.h" #include "intel_ddi.h" #include "intel_ddi_buf_trans.h" #include "intel_de.h" @@ -61,6 +63,7 @@ #include "intel_hti.h" #include "intel_lspcon.h" #include "intel_mg_phy_regs.h" +#include "intel_modeset_lock.h" #include "intel_pps.h" #include "intel_psr.h" #include "intel_quirks.h" @@ -68,7 +71,6 @@ #include "intel_tc.h" #include "intel_vdsc.h" #include "intel_vdsc_regs.h" -#include "intel_vrr.h" #include "skl_scaler.h" #include "skl_universal_plane.h" @@ -169,6 +171,18 @@ static void hsw_prepare_hdmi_ddi_buffers(struct intel_encoder *encoder, trans->entries[level].hsw.trans2); } +static void mtl_wait_ddi_buf_idle(struct drm_i915_private *i915, enum port port) +{ + int ret; + + /* FIXME: find out why Bspec's 100us timeout is too short */ + ret = wait_for_us((intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port)) & + XELPDP_PORT_BUF_PHY_IDLE), 10000); + if (ret) + drm_err(&i915->drm, "Timeout waiting for DDI BUF %c to get idle\n", + port_name(port)); +} + void intel_wait_ddi_buf_idle(struct drm_i915_private *dev_priv, enum port port) { @@ -196,7 +210,9 @@ static void intel_wait_ddi_buf_active(struct drm_i915_private *dev_priv, return; } - if (IS_DG2(dev_priv)) { + if (DISPLAY_VER(dev_priv) >= 14) { + timeout_us = 10000; + } else if (IS_DG2(dev_priv)) { timeout_us = 1200; } else if (DISPLAY_VER(dev_priv) >= 12) { if (intel_phy_is_tc(dev_priv, phy)) @@ -207,8 +223,12 @@ static void intel_wait_ddi_buf_active(struct drm_i915_private *dev_priv, timeout_us = 500; } - ret = _wait_for(!(intel_de_read(dev_priv, DDI_BUF_CTL(port)) & - DDI_BUF_IS_IDLE), timeout_us, 10, 10); + if (DISPLAY_VER(dev_priv) >= 14) + ret = _wait_for(!(intel_de_read(dev_priv, XELPDP_PORT_BUF_CTL1(port)) & XELPDP_PORT_BUF_PHY_IDLE), + timeout_us, 10, 10); + else + ret = _wait_for(!(intel_de_read(dev_priv, DDI_BUF_CTL(port)) & DDI_BUF_IS_IDLE), + timeout_us, 10, 10); if (ret) drm_err(&dev_priv->drm, "Timeout waiting for DDI BUF %c to get active\n", @@ -313,6 +333,13 @@ static void intel_ddi_init_dp_buf_reg(struct intel_encoder *encoder, DDI_PORT_WIDTH(crtc_state->lane_count) | DDI_BUF_TRANS_SELECT(0); + if (DISPLAY_VER(i915) >= 14) { + if (intel_dp_is_uhbr(crtc_state)) + intel_dp->DP |= DDI_BUF_PORT_DATA_40BIT; + else + intel_dp->DP |= DDI_BUF_PORT_DATA_10BIT; + } + if (IS_ALDERLAKE_P(i915) && intel_phy_is_tc(i915, phy)) { intel_dp->DP |= ddi_buf_phy_link_rate(crtc_state->port_clock); if (!intel_tc_port_in_tbt_alt_mode(dig_port)) @@ -515,6 +542,8 @@ intel_ddi_transcoder_func_reg_val_get(struct intel_encoder *encoder, temp |= TRANS_DDI_HDMI_SCRAMBLING; if (crtc_state->hdmi_high_tmds_clock_ratio) temp |= TRANS_DDI_HIGH_TMDS_CHAR_RATE; + if (DISPLAY_VER(dev_priv) >= 14) + temp |= TRANS_DDI_PORT_WIDTH(crtc_state->lane_count); } else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_ANALOG)) { temp |= TRANS_DDI_MODE_SELECT_FDI_OR_128B132B; temp |= (crtc_state->fdi_lanes - 1) << 1; @@ -2196,12 +2225,10 @@ static void intel_ddi_enable_fec(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state) { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); - struct intel_dp *intel_dp; if (!crtc_state->fec_enable) return; - intel_dp = enc_to_intel_dp(encoder); intel_de_rmw(dev_priv, dp_tp_ctl_reg(encoder, crtc_state), 0, DP_TP_CTL_FEC_ENABLE); } @@ -2210,12 +2237,10 @@ static void intel_ddi_disable_fec_state(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state) { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); - struct intel_dp *intel_dp; if (!crtc_state->fec_enable) return; - intel_dp = enc_to_intel_dp(encoder); intel_de_rmw(dev_priv, dp_tp_ctl_reg(encoder, crtc_state), DP_TP_CTL_FEC_ENABLE, 0); intel_de_posting_read(dev_priv, dp_tp_ctl_reg(encoder, crtc_state)); @@ -2309,6 +2334,179 @@ static void intel_ddi_mso_configure(const struct intel_crtc_state *crtc_state) OVERLAP_PIXELS_MASK, dss1); } +static u8 mtl_get_port_width(u8 lane_count) +{ + switch (lane_count) { + case 1: + return 0; + case 2: + return 1; + case 3: + return 4; + case 4: + return 3; + default: + MISSING_CASE(lane_count); + return 4; + } +} + +static void +mtl_ddi_enable_d2d(struct intel_encoder *encoder) +{ + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + enum port port = encoder->port; + + intel_de_rmw(dev_priv, XELPDP_PORT_BUF_CTL1(port), 0, + XELPDP_PORT_BUF_D2D_LINK_ENABLE); + + if (wait_for_us((intel_de_read(dev_priv, XELPDP_PORT_BUF_CTL1(port)) & + XELPDP_PORT_BUF_D2D_LINK_STATE), 100)) { + drm_err(&dev_priv->drm, "Timeout waiting for D2D Link enable for PORT_BUF_CTL %c\n", + port_name(port)); + } +} + +static void mtl_port_buf_ctl_program(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + struct intel_digital_port *dig_port = enc_to_dig_port(encoder); + enum port port = encoder->port; + u32 val; + + val = intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port)); + val &= ~XELPDP_PORT_WIDTH_MASK; + val |= XELPDP_PORT_WIDTH(mtl_get_port_width(crtc_state->lane_count)); + + val &= ~XELPDP_PORT_BUF_PORT_DATA_WIDTH_MASK; + if (intel_dp_is_uhbr(crtc_state)) + val |= XELPDP_PORT_BUF_PORT_DATA_40BIT; + else + val |= XELPDP_PORT_BUF_PORT_DATA_10BIT; + + if (dig_port->saved_port_bits & DDI_BUF_PORT_REVERSAL) + val |= XELPDP_PORT_REVERSAL; + + intel_de_write(i915, XELPDP_PORT_BUF_CTL1(port), val); +} + +static void mtl_port_buf_ctl_io_selection(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + struct intel_digital_port *dig_port = enc_to_dig_port(encoder); + u32 val; + + val = intel_tc_port_in_tbt_alt_mode(dig_port) ? + XELPDP_PORT_BUF_IO_SELECT_TBT : 0; + intel_de_rmw(i915, XELPDP_PORT_BUF_CTL1(encoder->port), + XELPDP_PORT_BUF_IO_SELECT_TBT, val); +} + +static void mtl_ddi_pre_enable_dp(struct intel_atomic_state *state, + struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state, + const struct drm_connector_state *conn_state) +{ + struct intel_dp *intel_dp = enc_to_intel_dp(encoder); + bool is_mst = intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DP_MST); + + intel_dp_set_link_params(intel_dp, + crtc_state->port_clock, + crtc_state->lane_count); + + /* + * We only configure what the register value will be here. Actual + * enabling happens during link training farther down. + */ + intel_ddi_init_dp_buf_reg(encoder, crtc_state); + + /* + * 1. Enable Power Wells + * + * This was handled at the beginning of intel_atomic_commit_tail(), + * before we called down into this function. + */ + + /* 2. PMdemand was already set */ + + /* 3. Select Thunderbolt */ + mtl_port_buf_ctl_io_selection(encoder); + + /* 4. Enable Panel Power if PPS is required */ + intel_pps_on(intel_dp); + + /* 5. Enable the port PLL */ + intel_ddi_enable_clock(encoder, crtc_state); + + /* + * 6.a Configure Transcoder Clock Select to direct the Port clock to the + * Transcoder. + */ + intel_ddi_enable_transcoder_clock(encoder, crtc_state); + + /* + * 6.b If DP v2.0/128b mode - Configure TRANS_DP2_CTL register settings. + */ + intel_ddi_config_transcoder_dp2(encoder, crtc_state); + + /* + * 6.c Configure TRANS_DDI_FUNC_CTL DDI Select, DDI Mode Select & MST + * Transport Select + */ + intel_ddi_config_transcoder_func(encoder, crtc_state); + + /* + * 6.e Program CoG/MSO configuration bits in DSS_CTL1 if selected. + */ + intel_ddi_mso_configure(crtc_state); + + if (!is_mst) + intel_dp_set_power(intel_dp, DP_SET_POWER_D0); + + intel_dp_configure_protocol_converter(intel_dp, crtc_state); + intel_dp_sink_set_decompression_state(intel_dp, crtc_state, true); + /* + * DDI FEC: "anticipates enabling FEC encoding sets the FEC_READY bit + * in the FEC_CONFIGURATION register to 1 before initiating link + * training + */ + intel_dp_sink_set_fec_ready(intel_dp, crtc_state); + + intel_dp_check_frl_training(intel_dp); + intel_dp_pcon_dsc_configure(intel_dp, crtc_state); + + /* + * 6. The rest of the below are substeps under the bspec's "Enable and + * Train Display Port" step. Note that steps that are specific to + * MST will be handled by intel_mst_pre_enable_dp() before/after it + * calls into this function. Also intel_mst_pre_enable_dp() only calls + * us when active_mst_links==0, so any steps designated for "single + * stream or multi-stream master transcoder" can just be performed + * unconditionally here. + * + * mtl_ddi_prepare_link_retrain() that is called by + * intel_dp_start_link_train() will execute steps: 6.d, 6.f, 6.g, 6.h, + * 6.i and 6.j + * + * 6.k Follow DisplayPort specification training sequence (see notes for + * failure handling) + * 6.m If DisplayPort multi-stream - Set DP_TP_CTL link training to Idle + * Pattern, wait for 5 idle patterns (DP_TP_STATUS Min_Idles_Sent) + * (timeout after 800 us) + */ + intel_dp_start_link_train(intel_dp, crtc_state); + + /* 6.n Set DP_TP_CTL link training to Normal */ + if (!is_trans_port_sync_mode(crtc_state)) + intel_dp_stop_link_train(intel_dp, crtc_state); + + /* 6.o Configure and enable FEC if needed */ + intel_ddi_enable_fec(encoder, crtc_state); + + intel_dsc_dp_pps_write(encoder, crtc_state); +} + static void tgl_ddi_pre_enable_dp(struct intel_atomic_state *state, struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state, @@ -2523,7 +2721,9 @@ static void intel_ddi_pre_enable_dp(struct intel_atomic_state *state, intel_dp_128b132b_sdp_crc16(enc_to_intel_dp(encoder), crtc_state); - if (DISPLAY_VER(dev_priv) >= 12) + if (DISPLAY_VER(dev_priv) >= 14) + mtl_ddi_pre_enable_dp(state, encoder, crtc_state, conn_state); + else if (DISPLAY_VER(dev_priv) >= 12) tgl_ddi_pre_enable_dp(state, encoder, crtc_state, conn_state); else hsw_ddi_pre_enable_dp(state, encoder, crtc_state, conn_state); @@ -2597,15 +2797,57 @@ static void intel_ddi_pre_enable(struct intel_atomic_state *state, /* FIXME precompute everything properly */ /* FIXME how do we turn infoframes off again? */ - if (dig_port->lspcon.active && dig_port->dp.has_hdmi_sink) + if (dig_port->lspcon.active && intel_dp_has_hdmi_sink(&dig_port->dp)) dig_port->set_infoframes(encoder, crtc_state->has_infoframe, crtc_state, conn_state); } } -static void intel_disable_ddi_buf(struct intel_encoder *encoder, - const struct intel_crtc_state *crtc_state) +static void +mtl_ddi_disable_d2d_link(struct intel_encoder *encoder) +{ + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + enum port port = encoder->port; + + intel_de_rmw(dev_priv, XELPDP_PORT_BUF_CTL1(port), + XELPDP_PORT_BUF_D2D_LINK_ENABLE, 0); + + if (wait_for_us(!(intel_de_read(dev_priv, XELPDP_PORT_BUF_CTL1(port)) & + XELPDP_PORT_BUF_D2D_LINK_STATE), 100)) + drm_err(&dev_priv->drm, "Timeout waiting for D2D Link disable for PORT_BUF_CTL %c\n", + port_name(port)); +} + +static void mtl_disable_ddi_buf(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + enum port port = encoder->port; + u32 val; + + /* 3.b Clear DDI_CTL_DE Enable to 0. */ + val = intel_de_read(dev_priv, DDI_BUF_CTL(port)); + if (val & DDI_BUF_CTL_ENABLE) { + val &= ~DDI_BUF_CTL_ENABLE; + intel_de_write(dev_priv, DDI_BUF_CTL(port), val); + + /* 3.c Poll for PORT_BUF_CTL Idle Status == 1, timeout after 100us */ + mtl_wait_ddi_buf_idle(dev_priv, port); + } + + /* 3.d Disable D2D Link */ + mtl_ddi_disable_d2d_link(encoder); + + /* 3.e Disable DP_TP_CTL */ + if (intel_crtc_has_dp_encoder(crtc_state)) { + intel_de_rmw(dev_priv, dp_tp_ctl_reg(encoder, crtc_state), + DP_TP_CTL_ENABLE, 0); + } +} + +static void disable_ddi_buf(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); enum port port = encoder->port; @@ -2630,6 +2872,21 @@ static void intel_disable_ddi_buf(struct intel_encoder *encoder, intel_wait_ddi_buf_idle(dev_priv, port); } +static void intel_disable_ddi_buf(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + + if (DISPLAY_VER(dev_priv) >= 14) { + mtl_disable_ddi_buf(encoder, crtc_state); + + /* 3.f Disable DP_TP_CTL FEC Enable if it is needed */ + intel_ddi_disable_fec_state(encoder, crtc_state); + } else { + disable_ddi_buf(encoder, crtc_state); + } +} + static void intel_ddi_post_disable_dp(struct intel_atomic_state *state, struct intel_encoder *encoder, const struct intel_crtc_state *old_crtc_state, @@ -2638,6 +2895,7 @@ static void intel_ddi_post_disable_dp(struct intel_atomic_state *state, struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); struct intel_digital_port *dig_port = enc_to_dig_port(encoder); struct intel_dp *intel_dp = &dig_port->dp; + intel_wakeref_t wakeref; bool is_mst = intel_crtc_has_type(old_crtc_state, INTEL_OUTPUT_DP_MST); @@ -2677,12 +2935,19 @@ static void intel_ddi_post_disable_dp(struct intel_atomic_state *state, intel_pps_vdd_on(intel_dp); intel_pps_off(intel_dp); - if (!intel_tc_port_in_tbt_alt_mode(dig_port)) + wakeref = fetch_and_zero(&dig_port->ddi_io_wakeref); + + if (wakeref) intel_display_power_put(dev_priv, dig_port->ddi_io_power_domain, - fetch_and_zero(&dig_port->ddi_io_wakeref)); + wakeref); intel_ddi_disable_clock(encoder); + + /* De-select Thunderbolt */ + if (DISPLAY_VER(dev_priv) >= 14) + intel_de_rmw(dev_priv, XELPDP_PORT_BUF_CTL1(encoder->port), + XELPDP_PORT_BUF_IO_SELECT_TBT, 0); } static void intel_ddi_post_disable_hdmi(struct intel_atomic_state *state, @@ -2693,6 +2958,7 @@ static void intel_ddi_post_disable_hdmi(struct intel_atomic_state *state, struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); struct intel_digital_port *dig_port = enc_to_dig_port(encoder); struct intel_hdmi *intel_hdmi = &dig_port->hdmi; + intel_wakeref_t wakeref; dig_port->set_infoframes(encoder, false, old_crtc_state, old_conn_state); @@ -2705,9 +2971,11 @@ static void intel_ddi_post_disable_hdmi(struct intel_atomic_state *state, if (DISPLAY_VER(dev_priv) >= 12) intel_ddi_disable_transcoder_clock(old_crtc_state); - intel_display_power_put(dev_priv, - dig_port->ddi_io_power_domain, - fetch_and_zero(&dig_port->ddi_io_wakeref)); + wakeref = fetch_and_zero(&dig_port->ddi_io_wakeref); + if (wakeref) + intel_display_power_put(dev_priv, + dig_port->ddi_io_power_domain, + wakeref); intel_ddi_disable_clock(encoder); @@ -2725,8 +2993,6 @@ static void intel_ddi_post_disable(struct intel_atomic_state *state, if (!intel_crtc_has_type(old_crtc_state, INTEL_OUTPUT_DP_MST)) { intel_crtc_vblank_off(old_crtc_state); - intel_vrr_disable(old_crtc_state); - intel_disable_transcoder(old_crtc_state); intel_ddi_disable_transcoder_func(old_crtc_state); @@ -2840,7 +3106,7 @@ static void intel_enable_ddi_dp(struct intel_atomic_state *state, drm_connector_update_privacy_screen(conn_state); intel_edp_backlight_on(crtc_state, conn_state); - if (!dig_port->lspcon.active || dig_port->dp.has_hdmi_sink) + if (!dig_port->lspcon.active || intel_dp_has_hdmi_sink(&dig_port->dp)) intel_dp_set_infoframes(encoder, true, crtc_state, conn_state); intel_audio_codec_enable(encoder, crtc_state, conn_state); @@ -2890,6 +3156,10 @@ static void intel_enable_ddi_hdmi(struct intel_atomic_state *state, if (has_buf_trans_select(dev_priv)) hsw_prepare_hdmi_ddi_buffers(encoder, crtc_state); + /* e. Enable D2D Link for C10/C20 Phy */ + if (DISPLAY_VER(dev_priv) >= 14) + mtl_ddi_enable_d2d(encoder); + encoder->set_signal_levels(encoder, crtc_state); /* Display WA #1143: skl,kbl,cfl */ @@ -2935,12 +3205,30 @@ static void intel_enable_ddi_hdmi(struct intel_atomic_state *state, * * On ADL_P the PHY link rate and lane count must be programmed but * these are both 0 for HDMI. + * + * But MTL onwards HDMI2.1 is supported and in TMDS mode this + * is filled with lane count, already set in the crtc_state. + * The same is required to be filled in PORT_BUF_CTL for C10/20 Phy. */ buf_ctl = dig_port->saved_port_bits | DDI_BUF_CTL_ENABLE; - if (IS_ALDERLAKE_P(dev_priv) && intel_phy_is_tc(dev_priv, phy)) { + if (DISPLAY_VER(dev_priv) >= 14) { + u8 lane_count = mtl_get_port_width(crtc_state->lane_count); + u32 port_buf = 0; + + port_buf |= XELPDP_PORT_WIDTH(lane_count); + + if (dig_port->saved_port_bits & DDI_BUF_PORT_REVERSAL) + port_buf |= XELPDP_PORT_REVERSAL; + + intel_de_rmw(dev_priv, XELPDP_PORT_BUF_CTL1(port), + XELPDP_PORT_WIDTH_MASK | XELPDP_PORT_REVERSAL, port_buf); + + buf_ctl |= DDI_PORT_WIDTH(lane_count); + } else if (IS_ALDERLAKE_P(dev_priv) && intel_phy_is_tc(dev_priv, phy)) { drm_WARN_ON(&dev_priv->drm, !intel_tc_port_in_legacy_mode(dig_port)); buf_ctl |= DDI_BUF_CTL_TC_PHY_OWNERSHIP; } + intel_de_write(dev_priv, DDI_BUF_CTL(port), buf_ctl); intel_wait_ddi_buf_active(dev_priv, port); @@ -2963,8 +3251,6 @@ static void intel_enable_ddi(struct intel_atomic_state *state, intel_enable_transcoder(crtc_state); - intel_vrr_enable(encoder, crtc_state); - intel_crtc_vblank_on(crtc_state); if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) @@ -2975,9 +3261,7 @@ static void intel_enable_ddi(struct intel_atomic_state *state, /* Enable hdcp if it's desired */ if (conn_state->content_protection == DRM_MODE_CONTENT_PROTECTION_DESIRED) - intel_hdcp_enable(to_intel_connector(conn_state->connector), - crtc_state, - (u8)conn_state->hdcp_content_type); + intel_hdcp_enable(state, encoder, crtc_state, conn_state); } static void intel_disable_ddi_dp(struct intel_atomic_state *state, @@ -3023,6 +3307,8 @@ static void intel_disable_ddi(struct intel_atomic_state *state, const struct intel_crtc_state *old_crtc_state, const struct drm_connector_state *old_conn_state) { + intel_tc_port_link_cancel_reset_work(enc_to_dig_port(encoder)); + intel_hdcp_disable(to_intel_connector(old_conn_state->connector)); if (intel_crtc_has_type(old_crtc_state, INTEL_OUTPUT_HDMI)) @@ -3070,7 +3356,8 @@ void intel_ddi_update_active_dpll(struct intel_atomic_state *state, struct intel_crtc *slave_crtc; enum phy phy = intel_port_to_phy(i915, encoder->port); - if (!intel_phy_is_tc(i915, phy)) + /* FIXME: Add MTL pll_mgr */ + if (DISPLAY_VER(i915) >= 14 || !intel_phy_is_tc(i915, phy)) return; intel_update_active_dpll(state, crtc, encoder); @@ -3121,6 +3408,53 @@ static void adlp_tbt_to_dp_alt_switch_wa(struct intel_encoder *encoder) intel_dkl_phy_rmw(i915, DKL_PCS_DW5(tc_port, ln), DKL_PCS_DW5_CORE_SOFTRESET, 0); } +static void mtl_ddi_prepare_link_retrain(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state) +{ + struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); + struct intel_encoder *encoder = &dig_port->base; + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + enum port port = encoder->port; + u32 dp_tp_ctl; + + /* + * TODO: To train with only a different voltage swing entry is not + * necessary disable and enable port + */ + dp_tp_ctl = intel_de_read(dev_priv, dp_tp_ctl_reg(encoder, crtc_state)); + if (dp_tp_ctl & DP_TP_CTL_ENABLE) + mtl_disable_ddi_buf(encoder, crtc_state); + + /* 6.d Configure and enable DP_TP_CTL with link training pattern 1 selected */ + dp_tp_ctl = DP_TP_CTL_ENABLE | DP_TP_CTL_LINK_TRAIN_PAT1; + if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DP_MST)) { + dp_tp_ctl |= DP_TP_CTL_MODE_MST; + } else { + dp_tp_ctl |= DP_TP_CTL_MODE_SST; + if (drm_dp_enhanced_frame_cap(intel_dp->dpcd)) + dp_tp_ctl |= DP_TP_CTL_ENHANCED_FRAME_ENABLE; + } + intel_de_write(dev_priv, dp_tp_ctl_reg(encoder, crtc_state), dp_tp_ctl); + intel_de_posting_read(dev_priv, dp_tp_ctl_reg(encoder, crtc_state)); + + /* 6.f Enable D2D Link */ + mtl_ddi_enable_d2d(encoder); + + /* 6.g Configure voltage swing and related IO settings */ + encoder->set_signal_levels(encoder, crtc_state); + + /* 6.h Configure PORT_BUF_CTL1 */ + mtl_port_buf_ctl_program(encoder, crtc_state); + + /* 6.i Configure and enable DDI_CTL_DE to start sending valid data to port slice */ + intel_dp->DP |= DDI_BUF_CTL_ENABLE; + intel_de_write(dev_priv, DDI_BUF_CTL(port), intel_dp->DP); + intel_de_posting_read(dev_priv, DDI_BUF_CTL(port)); + + /* 6.j Poll for PORT_BUF_CTL Idle Status == 0, timeout after 100 us */ + intel_wait_ddi_buf_active(dev_priv, port); +} + static void intel_ddi_prepare_link_retrain(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state) { @@ -3369,7 +3703,11 @@ static void intel_ddi_read_func_ctl(struct intel_encoder *encoder, fallthrough; case TRANS_DDI_MODE_SELECT_DVI: pipe_config->output_types |= BIT(INTEL_OUTPUT_HDMI); - pipe_config->lane_count = 4; + if (DISPLAY_VER(dev_priv) >= 14) + pipe_config->lane_count = + ((temp & DDI_PORT_WIDTH_MASK) >> DDI_PORT_WIDTH_SHIFT) + 1; + else + pipe_config->lane_count = 4; break; case TRANS_DDI_MODE_SELECT_DP_SST: if (encoder->type == INTEL_OUTPUT_EDP) @@ -3396,7 +3734,7 @@ static void intel_ddi_read_func_ctl(struct intel_encoder *encoder, pipe_config->fec_enable); } - if (dig_port->lspcon.active && dig_port->dp.has_hdmi_sink) + if (dig_port->lspcon.active && intel_dp_has_hdmi_sink(&dig_port->dp)) pipe_config->infoframes.enable |= intel_lspcon_infoframes_enabled(encoder, pipe_config); else @@ -3506,6 +3844,28 @@ void intel_ddi_get_clock(struct intel_encoder *encoder, &crtc_state->dpll_hw_state); } +static void mtl_ddi_get_config(struct intel_encoder *encoder, + struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + enum phy phy = intel_port_to_phy(i915, encoder->port); + struct intel_digital_port *dig_port = enc_to_dig_port(encoder); + + if (intel_tc_port_in_tbt_alt_mode(dig_port)) { + crtc_state->port_clock = intel_mtl_tbt_calc_port_clock(encoder); + } else if (intel_is_c10phy(i915, phy)) { + intel_c10pll_readout_hw_state(encoder, &crtc_state->cx0pll_state.c10); + intel_c10pll_dump_hw_state(i915, &crtc_state->cx0pll_state.c10); + crtc_state->port_clock = intel_c10pll_calc_port_clock(encoder, &crtc_state->cx0pll_state.c10); + } else { + intel_c20pll_readout_hw_state(encoder, &crtc_state->cx0pll_state.c20); + intel_c20pll_dump_hw_state(i915, &crtc_state->cx0pll_state.c20); + crtc_state->port_clock = intel_c20pll_calc_port_clock(encoder, &crtc_state->cx0pll_state.c20); + } + + intel_ddi_get_config(encoder, crtc_state); +} + static void dg2_ddi_get_config(struct intel_encoder *encoder, struct intel_crtc_state *crtc_state) { @@ -3700,6 +4060,9 @@ static int intel_ddi_compute_config(struct intel_encoder *encoder, pipe_config->cpu_transcoder = TRANSCODER_EDP; if (intel_crtc_has_type(pipe_config, INTEL_OUTPUT_HDMI)) { + pipe_config->has_hdmi_sink = + intel_hdmi_compute_has_hdmi_sink(encoder, pipe_config, conn_state); + ret = intel_hdmi_compute_config(encoder, pipe_config, conn_state); } else { ret = intel_dp_compute_config(encoder, pipe_config, conn_state); @@ -3865,14 +4228,25 @@ static void intel_ddi_encoder_reset(struct drm_encoder *encoder) intel_tc_port_init_mode(dig_port); } +static int intel_ddi_encoder_late_register(struct drm_encoder *_encoder) +{ + struct intel_encoder *encoder = to_intel_encoder(_encoder); + + intel_tc_port_link_reset(enc_to_dig_port(encoder)); + + return 0; +} + static const struct drm_encoder_funcs intel_ddi_funcs = { .reset = intel_ddi_encoder_reset, .destroy = intel_ddi_encoder_destroy, + .late_register = intel_ddi_encoder_late_register, }; static struct intel_connector * intel_ddi_init_dp_connector(struct intel_digital_port *dig_port) { + struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev); struct intel_connector *connector; enum port port = dig_port->base.port; @@ -3881,7 +4255,10 @@ intel_ddi_init_dp_connector(struct intel_digital_port *dig_port) return NULL; dig_port->dp.output_reg = DDI_BUF_CTL(port); - dig_port->dp.prepare_link_retrain = intel_ddi_prepare_link_retrain; + if (DISPLAY_VER(i915) >= 14) + dig_port->dp.prepare_link_retrain = mtl_ddi_prepare_link_retrain; + else + dig_port->dp.prepare_link_retrain = intel_ddi_prepare_link_retrain; dig_port->dp.set_link_train = intel_ddi_set_link_train; dig_port->dp.set_idle_link_train = intel_ddi_set_idle_link_train; @@ -3921,6 +4298,7 @@ static int modeset_pipe(struct drm_crtc *crtc, return -ENOMEM; state->acquire_ctx = ctx; + to_intel_atomic_state(state)->internal = true; crtc_state = drm_atomic_get_crtc_state(state, crtc); if (IS_ERR(crtc_state)) { @@ -4032,27 +4410,17 @@ intel_ddi_hotplug(struct intel_encoder *encoder, state = intel_encoder_hotplug(encoder, connector); - drm_modeset_acquire_init(&ctx, 0); - - for (;;) { - if (connector->base.connector_type == DRM_MODE_CONNECTOR_HDMIA) - ret = intel_hdmi_reset_link(encoder, &ctx); - else - ret = intel_dp_retrain_link(encoder, &ctx); - - if (ret == -EDEADLK) { - drm_modeset_backoff(&ctx); - continue; + if (!intel_tc_port_link_reset(dig_port)) { + intel_modeset_lock_ctx_retry(&ctx, NULL, 0, ret) { + if (connector->base.connector_type == DRM_MODE_CONNECTOR_HDMIA) + ret = intel_hdmi_reset_link(encoder, &ctx); + else + ret = intel_dp_retrain_link(encoder, &ctx); } - break; + drm_WARN_ON(encoder->base.dev, ret); } - drm_modeset_drop_locks(&ctx); - drm_modeset_acquire_fini(&ctx); - drm_WARN(encoder->base.dev, ret, - "Acquiring modeset locks failed with %i\n", ret); - /* * Unpowered type-c dongles can take some time to boot and be * responsible, so here giving some time to those dongles to power up @@ -4257,31 +4625,27 @@ static bool intel_ddi_is_tc(struct drm_i915_private *i915, enum port port) static void intel_ddi_encoder_suspend(struct intel_encoder *encoder) { - struct intel_dp *intel_dp = enc_to_intel_dp(encoder); - struct drm_i915_private *i915 = dp_to_i915(intel_dp); - struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); - enum phy phy = intel_port_to_phy(i915, encoder->port); - intel_dp_encoder_suspend(encoder); - - if (!intel_phy_is_tc(i915, phy)) - return; - - intel_tc_port_flush_work(dig_port); } -static void intel_ddi_encoder_shutdown(struct intel_encoder *encoder) +static void intel_ddi_tc_encoder_suspend_complete(struct intel_encoder *encoder) { struct intel_dp *intel_dp = enc_to_intel_dp(encoder); - struct drm_i915_private *i915 = dp_to_i915(intel_dp); struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); - enum phy phy = intel_port_to_phy(i915, encoder->port); + intel_tc_port_suspend(dig_port); +} + +static void intel_ddi_encoder_shutdown(struct intel_encoder *encoder) +{ intel_dp_encoder_shutdown(encoder); intel_hdmi_encoder_shutdown(encoder); +} - if (!intel_phy_is_tc(i915, phy)) - return; +static void intel_ddi_tc_encoder_shutdown_complete(struct intel_encoder *encoder) +{ + struct intel_dp *intel_dp = enc_to_intel_dp(encoder); + struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); intel_tc_port_cleanup(dig_port); } @@ -4413,7 +4777,12 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port) encoder->cloneable = 0; encoder->pipe_mask = ~0; - if (IS_DG2(dev_priv)) { + if (DISPLAY_VER(dev_priv) >= 14) { + encoder->enable_clock = intel_mtl_pll_enable; + encoder->disable_clock = intel_mtl_pll_disable; + encoder->port_pll_type = intel_mtl_port_pll_type; + encoder->get_config = mtl_ddi_get_config; + } else if (IS_DG2(dev_priv)) { encoder->enable_clock = intel_mpllb_enable; encoder->disable_clock = intel_mpllb_disable; encoder->get_config = dg2_ddi_get_config; @@ -4473,7 +4842,9 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port) encoder->get_config = hsw_ddi_get_config; } - if (IS_DG2(dev_priv)) { + if (DISPLAY_VER(dev_priv) >= 14) { + encoder->set_signal_levels = intel_cx0_phy_set_signal_levels; + } else if (IS_DG2(dev_priv)) { encoder->set_signal_levels = intel_snps_phy_set_signal_levels; } else if (DISPLAY_VER(dev_priv) >= 12) { if (intel_phy_is_combo(dev_priv, phy)) @@ -4541,6 +4912,9 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port) is_legacy ? "legacy" : "non-legacy"); } + encoder->suspend_complete = intel_ddi_tc_encoder_suspend_complete; + encoder->shutdown_complete = intel_ddi_tc_encoder_shutdown_complete; + if (intel_tc_port_init(dig_port, is_legacy) < 0) goto err; } diff --git a/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c b/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c index 006a2e979000370d3b9768e914e60978833f5b41..b7d20485bde50175f290046136c430687ed4e607 100644 --- a/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c +++ b/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c @@ -9,6 +9,7 @@ #include "intel_de.h" #include "intel_display_types.h" #include "intel_dp.h" +#include "intel_cx0_phy.h" /* HDMI/DVI modes ignore everything but the last 2 items. So we share * them for both DP and FDI transports, allowing those ports to @@ -1035,6 +1036,65 @@ static const struct intel_ddi_buf_trans dg2_snps_trans_uhbr = { .num_entries = ARRAY_SIZE(_dg2_snps_trans_uhbr), }; +static const union intel_ddi_buf_trans_entry _mtl_c10_trans_dp14[] = { + { .snps = { 26, 0, 0 } }, /* preset 0 */ + { .snps = { 33, 0, 6 } }, /* preset 1 */ + { .snps = { 38, 0, 11 } }, /* preset 2 */ + { .snps = { 43, 0, 19 } }, /* preset 3 */ + { .snps = { 39, 0, 0 } }, /* preset 4 */ + { .snps = { 45, 0, 7 } }, /* preset 5 */ + { .snps = { 46, 0, 13 } }, /* preset 6 */ + { .snps = { 46, 0, 0 } }, /* preset 7 */ + { .snps = { 55, 0, 7 } }, /* preset 8 */ + { .snps = { 62, 0, 0 } }, /* preset 9 */ +}; + +static const struct intel_ddi_buf_trans mtl_cx0_trans = { + .entries = _mtl_c10_trans_dp14, + .num_entries = ARRAY_SIZE(_mtl_c10_trans_dp14), + .hdmi_default_entry = ARRAY_SIZE(_mtl_c10_trans_dp14) - 1, +}; + +/* DP2.0 */ +static const union intel_ddi_buf_trans_entry _mtl_c20_trans_uhbr[] = { + { .snps = { 48, 0, 0 } }, /* preset 0 */ + { .snps = { 43, 0, 5 } }, /* preset 1 */ + { .snps = { 40, 0, 8 } }, /* preset 2 */ + { .snps = { 37, 0, 11 } }, /* preset 3 */ + { .snps = { 33, 0, 15 } }, /* preset 4 */ + { .snps = { 46, 2, 0 } }, /* preset 5 */ + { .snps = { 42, 2, 4 } }, /* preset 6 */ + { .snps = { 38, 2, 8 } }, /* preset 7 */ + { .snps = { 35, 2, 11 } }, /* preset 8 */ + { .snps = { 33, 2, 13 } }, /* preset 9 */ + { .snps = { 44, 4, 0 } }, /* preset 10 */ + { .snps = { 40, 4, 4 } }, /* preset 11 */ + { .snps = { 37, 4, 7 } }, /* preset 12 */ + { .snps = { 33, 4, 11 } }, /* preset 13 */ + { .snps = { 40, 8, 0 } }, /* preset 14 */ + { .snps = { 28, 2, 2 } }, /* preset 15 */ +}; + +/* HDMI2.0 */ +static const union intel_ddi_buf_trans_entry _mtl_c20_trans_hdmi[] = { + { .snps = { 48, 0, 0 } }, /* preset 0 */ + { .snps = { 38, 4, 6 } }, /* preset 1 */ + { .snps = { 36, 4, 8 } }, /* preset 2 */ + { .snps = { 34, 4, 10 } }, /* preset 3 */ + { .snps = { 32, 4, 12 } }, /* preset 4 */ +}; + +static const struct intel_ddi_buf_trans mtl_c20_trans_hdmi = { + .entries = _mtl_c20_trans_hdmi, + .num_entries = ARRAY_SIZE(_mtl_c20_trans_hdmi), + .hdmi_default_entry = 0, +}; + +static const struct intel_ddi_buf_trans mtl_c20_trans_uhbr = { + .entries = _mtl_c20_trans_uhbr, + .num_entries = ARRAY_SIZE(_mtl_c20_trans_uhbr), +}; + bool is_hobl_buf_trans(const struct intel_ddi_buf_trans *table) { return table == &tgl_combo_phy_trans_edp_hbr2_hobl; @@ -1606,12 +1666,30 @@ dg2_get_snps_buf_trans(struct intel_encoder *encoder, return intel_get_buf_trans(&dg2_snps_trans, n_entries); } +static const struct intel_ddi_buf_trans * +mtl_get_cx0_buf_trans(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state, + int *n_entries) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + enum phy phy = intel_port_to_phy(i915, encoder->port); + + if (intel_crtc_has_dp_encoder(crtc_state) && crtc_state->port_clock >= 1000000) + return intel_get_buf_trans(&mtl_c20_trans_uhbr, n_entries); + else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI) && !(intel_is_c10phy(i915, phy))) + return intel_get_buf_trans(&mtl_c20_trans_hdmi, n_entries); + else + return intel_get_buf_trans(&mtl_cx0_trans, n_entries); +} + void intel_ddi_buf_trans_init(struct intel_encoder *encoder) { struct drm_i915_private *i915 = to_i915(encoder->base.dev); enum phy phy = intel_port_to_phy(i915, encoder->port); - if (IS_DG2(i915)) { + if (DISPLAY_VER(i915) >= 14) { + encoder->get_buf_trans = mtl_get_cx0_buf_trans; + } else if (IS_DG2(i915)) { encoder->get_buf_trans = dg2_get_snps_buf_trans; } else if (IS_ALDERLAKE_P(i915)) { if (intel_phy_is_combo(i915, phy)) diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c index 3c29792137a56d06e2c287b6c37a1e6671bfaf91..d8533603ad0523e1b621a4f6a95877df461ac433 100644 --- a/drivers/gpu/drm/i915/display/intel_display.c +++ b/drivers/gpu/drm/i915/display/intel_display.c @@ -31,8 +31,6 @@ #include #include #include -#include -#include #include #include @@ -41,7 +39,6 @@ #include #include #include -#include #include #include @@ -57,7 +54,6 @@ #include "i9xx_plane.h" #include "i9xx_wm.h" #include "icl_dsi.h" -#include "intel_acpi.h" #include "intel_atomic.h" #include "intel_atomic_plane.h" #include "intel_audio.h" @@ -70,7 +66,7 @@ #include "intel_crtc_state_dump.h" #include "intel_ddi.h" #include "intel_de.h" -#include "intel_display_debugfs.h" +#include "intel_display_driver.h" #include "intel_display_power.h" #include "intel_display_types.h" #include "intel_dmc.h" @@ -90,11 +86,8 @@ #include "intel_fdi.h" #include "intel_fifo_underrun.h" #include "intel_frontbuffer.h" -#include "intel_gmbus.h" -#include "intel_hdcp.h" #include "intel_hdmi.h" #include "intel_hotplug.h" -#include "intel_hti.h" #include "intel_lvds.h" #include "intel_lvds_regs.h" #include "intel_modeset_setup.h" @@ -106,9 +99,9 @@ #include "intel_pcode.h" #include "intel_pipe_crc.h" #include "intel_plane_initial.h" +#include "intel_pmdemand.h" #include "intel_pps.h" #include "intel_psr.h" -#include "intel_quirks.h" #include "intel_sdvo.h" #include "intel_snps_phy.h" #include "intel_tc.h" @@ -131,7 +124,6 @@ static void intel_set_transcoder_timings(const struct intel_crtc_state *crtc_sta static void intel_set_pipe_src_size(const struct intel_crtc_state *crtc_state); static void hsw_set_transconf(const struct intel_crtc_state *crtc_state); static void bdw_set_pipe_misc(const struct intel_crtc_state *crtc_state); -static void ilk_pfit_enable(const struct intel_crtc_state *crtc_state); /* returns HPLL frequency in kHz */ int vlv_get_hpll_vco(struct drm_i915_private *dev_priv) @@ -178,7 +170,7 @@ int vlv_get_cck_clock_hpll(struct drm_i915_private *dev_priv, return hpll; } -static void intel_update_czclk(struct drm_i915_private *dev_priv) +void intel_update_czclk(struct drm_i915_private *dev_priv) { if (!(IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))) return; @@ -234,7 +226,7 @@ is_trans_port_sync_slave(const struct intel_crtc_state *crtc_state) return crtc_state->master_transcoder != INVALID_TRANSCODER; } -static bool +bool is_trans_port_sync_master(const struct intel_crtc_state *crtc_state) { return crtc_state->sync_mode_slaves_mask != 0; @@ -331,20 +323,21 @@ void assert_transcoder(struct drm_i915_private *dev_priv, cur_state = false; } - I915_STATE_WARN(cur_state != state, + I915_STATE_WARN(dev_priv, cur_state != state, "transcoder %s assertion failure (expected %s, current %s)\n", - transcoder_name(cpu_transcoder), - str_on_off(state), str_on_off(cur_state)); + transcoder_name(cpu_transcoder), str_on_off(state), + str_on_off(cur_state)); } static void assert_plane(struct intel_plane *plane, bool state) { + struct drm_i915_private *i915 = to_i915(plane->base.dev); enum pipe pipe; bool cur_state; cur_state = plane->get_hw_state(plane, &pipe); - I915_STATE_WARN(cur_state != state, + I915_STATE_WARN(i915, cur_state != state, "%s assertion failure (expected %s, current %s)\n", plane->base.name, str_on_off(state), str_on_off(cur_state)); @@ -701,175 +694,6 @@ intel_plane_fence_y_offset(const struct intel_plane_state *plane_state) return y; } -static int -intel_display_commit_duplicated_state(struct intel_atomic_state *state, - struct drm_modeset_acquire_ctx *ctx) -{ - struct drm_i915_private *i915 = to_i915(state->base.dev); - int ret; - - ret = drm_atomic_helper_commit_duplicated_state(&state->base, ctx); - - drm_WARN_ON(&i915->drm, ret == -EDEADLK); - - return ret; -} - -static int -__intel_display_resume(struct drm_i915_private *i915, - struct drm_atomic_state *state, - struct drm_modeset_acquire_ctx *ctx) -{ - struct drm_crtc_state *crtc_state; - struct drm_crtc *crtc; - int i; - - intel_modeset_setup_hw_state(i915, ctx); - intel_vga_redisable(i915); - - if (!state) - return 0; - - /* - * We've duplicated the state, pointers to the old state are invalid. - * - * Don't attempt to use the old state until we commit the duplicated state. - */ - for_each_new_crtc_in_state(state, crtc, crtc_state, i) { - /* - * Force recalculation even if we restore - * current state. With fast modeset this may not result - * in a modeset when the state is compatible. - */ - crtc_state->mode_changed = true; - } - - /* ignore any reset values/BIOS leftovers in the WM registers */ - if (!HAS_GMCH(i915)) - to_intel_atomic_state(state)->skip_intermediate_wm = true; - - return intel_display_commit_duplicated_state(to_intel_atomic_state(state), ctx); -} - -static bool gpu_reset_clobbers_display(struct drm_i915_private *dev_priv) -{ - return (INTEL_INFO(dev_priv)->gpu_reset_clobbers_display && - intel_has_gpu_reset(to_gt(dev_priv))); -} - -void intel_display_prepare_reset(struct drm_i915_private *dev_priv) -{ - struct drm_modeset_acquire_ctx *ctx = &dev_priv->display.restore.reset_ctx; - struct drm_atomic_state *state; - int ret; - - if (!HAS_DISPLAY(dev_priv)) - return; - - /* reset doesn't touch the display */ - if (!dev_priv->params.force_reset_modeset_test && - !gpu_reset_clobbers_display(dev_priv)) - return; - - /* We have a modeset vs reset deadlock, defensively unbreak it. */ - set_bit(I915_RESET_MODESET, &to_gt(dev_priv)->reset.flags); - smp_mb__after_atomic(); - wake_up_bit(&to_gt(dev_priv)->reset.flags, I915_RESET_MODESET); - - if (atomic_read(&dev_priv->gpu_error.pending_fb_pin)) { - drm_dbg_kms(&dev_priv->drm, - "Modeset potentially stuck, unbreaking through wedging\n"); - intel_gt_set_wedged(to_gt(dev_priv)); - } - - /* - * Need mode_config.mutex so that we don't - * trample ongoing ->detect() and whatnot. - */ - mutex_lock(&dev_priv->drm.mode_config.mutex); - drm_modeset_acquire_init(ctx, 0); - while (1) { - ret = drm_modeset_lock_all_ctx(&dev_priv->drm, ctx); - if (ret != -EDEADLK) - break; - - drm_modeset_backoff(ctx); - } - /* - * Disabling the crtcs gracefully seems nicer. Also the - * g33 docs say we should at least disable all the planes. - */ - state = drm_atomic_helper_duplicate_state(&dev_priv->drm, ctx); - if (IS_ERR(state)) { - ret = PTR_ERR(state); - drm_err(&dev_priv->drm, "Duplicating state failed with %i\n", - ret); - return; - } - - ret = drm_atomic_helper_disable_all(&dev_priv->drm, ctx); - if (ret) { - drm_err(&dev_priv->drm, "Suspending crtc's failed with %i\n", - ret); - drm_atomic_state_put(state); - return; - } - - dev_priv->display.restore.modeset_state = state; - state->acquire_ctx = ctx; -} - -void intel_display_finish_reset(struct drm_i915_private *i915) -{ - struct drm_modeset_acquire_ctx *ctx = &i915->display.restore.reset_ctx; - struct drm_atomic_state *state; - int ret; - - if (!HAS_DISPLAY(i915)) - return; - - /* reset doesn't touch the display */ - if (!test_bit(I915_RESET_MODESET, &to_gt(i915)->reset.flags)) - return; - - state = fetch_and_zero(&i915->display.restore.modeset_state); - if (!state) - goto unlock; - - /* reset doesn't touch the display */ - if (!gpu_reset_clobbers_display(i915)) { - /* for testing only restore the display */ - ret = intel_display_commit_duplicated_state(to_intel_atomic_state(state), ctx); - if (ret) - drm_err(&i915->drm, - "Restoring old state failed with %i\n", ret); - } else { - /* - * The display has been reset as well, - * so need a full re-initialization. - */ - intel_pps_unlock_regs_wa(i915); - intel_modeset_init_hw(i915); - intel_clock_gating_init(i915); - intel_hpd_init(i915); - - ret = __intel_display_resume(i915, state, ctx); - if (ret) - drm_err(&i915->drm, - "Restoring old state failed with %i\n", ret); - - intel_hpd_poll_disable(i915); - } - - drm_atomic_state_put(state); -unlock: - drm_modeset_drop_locks(ctx); - drm_modeset_acquire_fini(ctx); - mutex_unlock(&i915->drm.mode_config.mutex); - - clear_bit_unlock(I915_RESET_MODESET, &to_gt(i915)->reset.flags); -} - static void icl_set_pipe_chicken(const struct intel_crtc_state *crtc_state) { struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); @@ -990,8 +814,10 @@ static void ilk_pfit_enable(const struct intel_crtc_state *crtc_state) else intel_de_write_fw(dev_priv, PF_CTL(pipe), PF_ENABLE | PF_FILTER_MED_3x3); - intel_de_write_fw(dev_priv, PF_WIN_POS(pipe), x << 16 | y); - intel_de_write_fw(dev_priv, PF_WIN_SZ(pipe), width << 16 | height); + intel_de_write_fw(dev_priv, PF_WIN_POS(pipe), + PF_WIN_XPOS(x) | PF_WIN_YPOS(y)); + intel_de_write_fw(dev_priv, PF_WIN_SZ(pipe), + PF_WIN_XSIZE(width) | PF_WIN_YSIZE(height)); } static void intel_crtc_dpms_overlay_disable(struct intel_crtc *crtc) @@ -1069,20 +895,40 @@ static bool needs_async_flip_vtd_wa(const struct intel_crtc_state *crtc_state) (DISPLAY_VER(i915) == 9 || IS_BROADWELL(i915) || IS_HASWELL(i915)); } +#define is_enabling(feature, old_crtc_state, new_crtc_state) \ + ((!(old_crtc_state)->feature || intel_crtc_needs_modeset(new_crtc_state)) && \ + (new_crtc_state)->feature) +#define is_disabling(feature, old_crtc_state, new_crtc_state) \ + ((old_crtc_state)->feature && \ + (!(new_crtc_state)->feature || intel_crtc_needs_modeset(new_crtc_state))) + static bool planes_enabling(const struct intel_crtc_state *old_crtc_state, const struct intel_crtc_state *new_crtc_state) { - return (!old_crtc_state->active_planes || intel_crtc_needs_modeset(new_crtc_state)) && - new_crtc_state->active_planes; + return is_enabling(active_planes, old_crtc_state, new_crtc_state); } static bool planes_disabling(const struct intel_crtc_state *old_crtc_state, const struct intel_crtc_state *new_crtc_state) { - return old_crtc_state->active_planes && - (!new_crtc_state->active_planes || intel_crtc_needs_modeset(new_crtc_state)); + return is_disabling(active_planes, old_crtc_state, new_crtc_state); +} + +static bool vrr_enabling(const struct intel_crtc_state *old_crtc_state, + const struct intel_crtc_state *new_crtc_state) +{ + return is_enabling(vrr.enable, old_crtc_state, new_crtc_state); +} + +static bool vrr_disabling(const struct intel_crtc_state *old_crtc_state, + const struct intel_crtc_state *new_crtc_state) +{ + return is_disabling(vrr.enable, old_crtc_state, new_crtc_state); } +#undef is_disabling +#undef is_enabling + static void intel_post_plane_update(struct intel_atomic_state *state, struct intel_crtc *crtc) { @@ -1126,7 +972,7 @@ static void intel_crtc_enable_flip_done(struct intel_atomic_state *state, const struct intel_crtc_state *crtc_state = intel_atomic_get_new_crtc_state(state, crtc); u8 update_planes = crtc_state->update_planes; - const struct intel_plane_state *plane_state; + const struct intel_plane_state __maybe_unused *plane_state; struct intel_plane *plane; int i; @@ -1143,7 +989,7 @@ static void intel_crtc_disable_flip_done(struct intel_atomic_state *state, const struct intel_crtc_state *crtc_state = intel_atomic_get_new_crtc_state(state, crtc); u8 update_planes = crtc_state->update_planes; - const struct intel_plane_state *plane_state; + const struct intel_plane_state __maybe_unused *plane_state; struct intel_plane *plane; int i; @@ -1196,6 +1042,11 @@ static void intel_pre_plane_update(struct intel_atomic_state *state, intel_atomic_get_new_crtc_state(state, crtc); enum pipe pipe = crtc->pipe; + if (vrr_disabling(old_crtc_state, new_crtc_state)) { + intel_vrr_disable(old_crtc_state); + intel_crtc_update_active_timings(old_crtc_state, false); + } + intel_drrs_deactivate(old_crtc_state); intel_psr_pre_plane_update(state, crtc); @@ -1676,6 +1527,8 @@ static void hsw_configure_cpu_transcoder(const struct intel_crtc_state *crtc_sta } intel_set_transcoder_timings(crtc_state); + if (HAS_VRR(dev_priv)) + intel_vrr_set_transcoder_timings(crtc_state); if (cpu_transcoder != TRANSCODER_EDP) intel_de_write(dev_priv, TRANS_MULT(cpu_transcoder), @@ -1851,9 +1704,17 @@ static void hsw_crtc_disable(struct intel_atomic_state *state, intel_disable_shared_dpll(old_crtc_state); - intel_encoders_post_pll_disable(state, crtc); + if (!intel_crtc_is_bigjoiner_slave(old_crtc_state)) { + struct intel_crtc *slave_crtc; + + intel_encoders_post_pll_disable(state, crtc); - intel_dmc_disable_pipe(i915, crtc->pipe); + intel_dmc_disable_pipe(i915, crtc->pipe); + + for_each_intel_crtc_in_pipe_mask(&i915->drm, slave_crtc, + intel_crtc_bigjoiner_slave_pipes(old_crtc_state)) + intel_dmc_disable_pipe(i915, slave_crtc->pipe); + } } static void i9xx_pfit_enable(const struct intel_crtc_state *crtc_state) @@ -1907,7 +1768,7 @@ bool intel_phy_is_tc(struct drm_i915_private *dev_priv, enum phy phy) if (IS_DG2(dev_priv)) /* DG2's "TC1" output uses a SNPS PHY */ return false; - else if (IS_ALDERLAKE_P(dev_priv)) + else if (IS_ALDERLAKE_P(dev_priv) || IS_METEORLAKE(dev_priv)) return phy >= PHY_F && phy <= PHY_I; else if (IS_TIGERLAKE(dev_priv)) return phy >= PHY_D && phy <= PHY_I; @@ -2214,30 +2075,6 @@ static void i9xx_crtc_disable(struct intel_atomic_state *state, i830_enable_pipe(dev_priv, pipe); } - -/* - * turn all crtc's off, but do not adjust state - * This has to be paired with a call to intel_modeset_setup_hw_state. - */ -int intel_display_suspend(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = to_i915(dev); - struct drm_atomic_state *state; - int ret; - - if (!HAS_DISPLAY(dev_priv)) - return 0; - - state = drm_atomic_helper_suspend(dev); - ret = PTR_ERR_OR_ZERO(state); - if (ret) - drm_err(&dev_priv->drm, "Suspending crtc's failed with %i\n", - ret); - else - dev_priv->display.restore.modeset_state = state; - return ret; -} - void intel_encoder_destroy(struct drm_encoder *encoder) { struct intel_encoder *intel_encoder = to_intel_encoder(encoder); @@ -2567,7 +2404,7 @@ intel_link_compute_m_n(u16 bits_per_pixel, int nlanes, 0x80000); } -static void intel_panel_sanitize_ssc(struct drm_i915_private *dev_priv) +void intel_panel_sanitize_ssc(struct drm_i915_private *dev_priv) { /* * There may be no VBT; and if the BIOS enabled SSC we can @@ -2904,6 +2741,9 @@ void i9xx_set_pipeconf(const struct intel_crtc_state *crtc_state) val |= TRANSCONF_GAMMA_MODE(crtc_state->gamma_mode); + if (crtc_state->wgc_enable) + val |= TRANSCONF_WGC_ENABLE; + val |= TRANSCONF_FRAME_START_DELAY(crtc_state->framestart_delay - 1); intel_de_write(dev_priv, TRANSCONF(cpu_transcoder), val); @@ -2923,6 +2763,7 @@ static void i9xx_get_pfit_config(struct intel_crtc_state *crtc_state) { struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + enum pipe pipe; u32 tmp; if (!i9xx_has_pfit(dev_priv)) @@ -2933,13 +2774,13 @@ static void i9xx_get_pfit_config(struct intel_crtc_state *crtc_state) return; /* Check whether the pfit is attached to our pipe. */ - if (DISPLAY_VER(dev_priv) < 4) { - if (crtc->pipe != PIPE_B) - return; - } else { - if ((tmp & PFIT_PIPE_MASK) != (crtc->pipe << PFIT_PIPE_SHIFT)) - return; - } + if (DISPLAY_VER(dev_priv) >= 4) + pipe = REG_FIELD_GET(PFIT_PIPE_MASK, tmp); + else + pipe = PIPE_B; + + if (pipe != crtc->pipe) + return; crtc_state->gmch_pfit.control = tmp; crtc_state->gmch_pfit.pgm_ratios = @@ -3061,6 +2902,7 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc, return false; pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB; + pipe_config->sink_format = pipe_config->output_format; pipe_config->cpu_transcoder = (enum transcoder) crtc->pipe; pipe_config->shared_dpll = NULL; @@ -3096,6 +2938,10 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc, pipe_config->framestart_delay = REG_FIELD_GET(TRANSCONF_FRAME_START_DELAY_MASK, tmp) + 1; + if ((IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) && + (tmp & TRANSCONF_WGC_ENABLE)) + pipe_config->wgc_enable = true; + if (IS_CHERRYVIEW(dev_priv)) pipe_config->cgm_mode = intel_de_read(dev_priv, CGM_PIPE_MODE(crtc->pipe)); @@ -3398,73 +3244,39 @@ void intel_cpu_transcoder_get_m2_n2(struct intel_crtc *crtc, PIPE_LINK_M2(transcoder), PIPE_LINK_N2(transcoder)); } -static void ilk_get_pfit_pos_size(struct intel_crtc_state *crtc_state, - u32 pos, u32 size) -{ - drm_rect_init(&crtc_state->pch_pfit.dst, - pos >> 16, pos & 0xffff, - size >> 16, size & 0xffff); -} - -static void skl_get_pfit_config(struct intel_crtc_state *crtc_state) -{ - struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); - struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - struct intel_crtc_scaler_state *scaler_state = &crtc_state->scaler_state; - int id = -1; - int i; - - /* find scaler attached to this pipe */ - for (i = 0; i < crtc->num_scalers; i++) { - u32 ctl, pos, size; - - ctl = intel_de_read(dev_priv, SKL_PS_CTRL(crtc->pipe, i)); - if ((ctl & (PS_SCALER_EN | PS_PLANE_SEL_MASK)) != PS_SCALER_EN) - continue; - - id = i; - crtc_state->pch_pfit.enabled = true; - - pos = intel_de_read(dev_priv, SKL_PS_WIN_POS(crtc->pipe, i)); - size = intel_de_read(dev_priv, SKL_PS_WIN_SZ(crtc->pipe, i)); - - ilk_get_pfit_pos_size(crtc_state, pos, size); - - scaler_state->scalers[i].in_use = true; - break; - } - - scaler_state->scaler_id = id; - if (id >= 0) - scaler_state->scaler_users |= (1 << SKL_CRTC_INDEX); - else - scaler_state->scaler_users &= ~(1 << SKL_CRTC_INDEX); -} - static void ilk_get_pfit_config(struct intel_crtc_state *crtc_state) { struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); u32 ctl, pos, size; + enum pipe pipe; ctl = intel_de_read(dev_priv, PF_CTL(crtc->pipe)); if ((ctl & PF_ENABLE) == 0) return; + if (IS_IVYBRIDGE(dev_priv) || IS_HASWELL(dev_priv)) + pipe = REG_FIELD_GET(PF_PIPE_SEL_MASK_IVB, ctl); + else + pipe = crtc->pipe; + crtc_state->pch_pfit.enabled = true; pos = intel_de_read(dev_priv, PF_WIN_POS(crtc->pipe)); size = intel_de_read(dev_priv, PF_WIN_SZ(crtc->pipe)); - ilk_get_pfit_pos_size(crtc_state, pos, size); + drm_rect_init(&crtc_state->pch_pfit.dst, + REG_FIELD_GET(PF_WIN_XPOS_MASK, pos), + REG_FIELD_GET(PF_WIN_YPOS_MASK, pos), + REG_FIELD_GET(PF_WIN_XSIZE_MASK, size), + REG_FIELD_GET(PF_WIN_YSIZE_MASK, size)); /* * We currently do not free assignements of panel fitters on * ivb/hsw (since we don't use the higher upscaling modes which * differentiates them) so just WARN about this case for now. */ - drm_WARN_ON(&dev_priv->drm, DISPLAY_VER(dev_priv) == 7 && - (ctl & PF_PIPE_SEL_MASK_IVB) != PF_PIPE_SEL_IVB(crtc->pipe)); + drm_WARN_ON(&dev_priv->drm, pipe != crtc->pipe); } static bool ilk_get_pipe_config(struct intel_crtc *crtc, @@ -3520,6 +3332,8 @@ static bool ilk_get_pipe_config(struct intel_crtc *crtc, break; } + pipe_config->sink_format = pipe_config->output_format; + pipe_config->gamma_mode = REG_FIELD_GET(TRANSCONF_GAMMA_MODE_MASK_ILK, tmp); pipe_config->framestart_delay = REG_FIELD_GET(TRANSCONF_FRAME_START_DELAY_MASK, tmp) + 1; @@ -3560,7 +3374,7 @@ static u8 bigjoiner_pipes(struct drm_i915_private *i915) else pipes = 0; - return pipes & RUNTIME_INFO(i915)->pipe_mask; + return pipes & DISPLAY_RUNTIME_INFO(i915)->pipe_mask; } static bool transcoder_ddi_func_is_enabled(struct drm_i915_private *dev_priv, @@ -3901,7 +3715,7 @@ static bool hsw_get_pipe_config(struct intel_crtc *crtc, intel_get_transcoder_timings(crtc, pipe_config); if (HAS_VRR(dev_priv) && !transcoder_is_dsi(pipe_config->cpu_transcoder)) - intel_vrr_get_config(crtc, pipe_config); + intel_vrr_get_config(pipe_config); intel_get_pipe_src_size(crtc, pipe_config); @@ -3918,6 +3732,8 @@ static bool hsw_get_pipe_config(struct intel_crtc *crtc, bdw_get_pipe_misc_output_format(crtc); } + pipe_config->sink_format = pipe_config->output_format; + pipe_config->gamma_mode = intel_de_read(dev_priv, GAMMA_MODE(crtc->pipe)); @@ -3947,7 +3763,7 @@ static bool hsw_get_pipe_config(struct intel_crtc *crtc, if (intel_display_power_get_in_set_if_enabled(dev_priv, &crtc->hw_readout_power_domains, POWER_DOMAIN_PIPE_PANEL_FITTER(crtc->pipe))) { if (DISPLAY_VER(dev_priv) >= 9) - skl_get_pfit_config(pipe_config); + skl_scaler_get_config(pipe_config); else ilk_get_pfit_config(pipe_config); } @@ -3995,290 +3811,78 @@ bool intel_crtc_get_pipe_config(struct intel_crtc_state *crtc_state) return true; } -/* VESA 640x480x72Hz mode to set on the pipe */ -static const struct drm_display_mode load_detect_mode = { - DRM_MODE("640x480", DRM_MODE_TYPE_DEFAULT, 31500, 640, 664, - 704, 832, 0, 480, 489, 491, 520, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC), -}; - -static int intel_modeset_disable_planes(struct drm_atomic_state *state, - struct drm_crtc *crtc) +static int i9xx_pll_refclk(struct drm_device *dev, + const struct intel_crtc_state *pipe_config) { - struct drm_plane *plane; - struct drm_plane_state *plane_state; - int ret, i; - - ret = drm_atomic_add_affected_planes(state, crtc); - if (ret) - return ret; - - for_each_new_plane_in_state(state, plane, plane_state, i) { - if (plane_state->crtc != crtc) - continue; - - ret = drm_atomic_set_crtc_for_plane(plane_state, NULL); - if (ret) - return ret; - - drm_atomic_set_fb_for_plane(plane_state, NULL); - } + struct drm_i915_private *dev_priv = to_i915(dev); + u32 dpll = pipe_config->dpll_hw_state.dpll; - return 0; + if ((dpll & PLL_REF_INPUT_MASK) == PLLB_REF_INPUT_SPREADSPECTRUMIN) + return dev_priv->display.vbt.lvds_ssc_freq; + else if (HAS_PCH_SPLIT(dev_priv)) + return 120000; + else if (DISPLAY_VER(dev_priv) != 2) + return 96000; + else + return 48000; } -int intel_get_load_detect_pipe(struct drm_connector *connector, - struct intel_load_detect_pipe *old, - struct drm_modeset_acquire_ctx *ctx) +/* Returns the clock of the currently programmed mode of the given pipe. */ +void i9xx_crtc_clock_get(struct intel_crtc *crtc, + struct intel_crtc_state *pipe_config) { - struct intel_encoder *encoder = - intel_attached_encoder(to_intel_connector(connector)); - struct intel_crtc *possible_crtc; - struct intel_crtc *crtc = NULL; - struct drm_device *dev = encoder->base.dev; + struct drm_device *dev = crtc->base.dev; struct drm_i915_private *dev_priv = to_i915(dev); - struct drm_mode_config *config = &dev->mode_config; - struct drm_atomic_state *state = NULL, *restore_state = NULL; - struct drm_connector_state *connector_state; - struct intel_crtc_state *crtc_state; - int ret; - - drm_dbg_kms(&dev_priv->drm, "[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n", - connector->base.id, connector->name, - encoder->base.base.id, encoder->base.name); - - old->restore_state = NULL; - - drm_WARN_ON(dev, !drm_modeset_is_locked(&config->connection_mutex)); - - /* - * Algorithm gets a little messy: - * - * - if the connector already has an assigned crtc, use it (but make - * sure it's on first) - * - * - try to find the first unused crtc that can drive this connector, - * and use that if we find one - */ - - /* See if we already have a CRTC for this connector */ - if (connector->state->crtc) { - crtc = to_intel_crtc(connector->state->crtc); + u32 dpll = pipe_config->dpll_hw_state.dpll; + u32 fp; + struct dpll clock; + int port_clock; + int refclk = i9xx_pll_refclk(dev, pipe_config); - ret = drm_modeset_lock(&crtc->base.mutex, ctx); - if (ret) - goto fail; + if ((dpll & DISPLAY_RATE_SELECT_FPA1) == 0) + fp = pipe_config->dpll_hw_state.fp0; + else + fp = pipe_config->dpll_hw_state.fp1; - /* Make sure the crtc and connector are running */ - goto found; + clock.m1 = (fp & FP_M1_DIV_MASK) >> FP_M1_DIV_SHIFT; + if (IS_PINEVIEW(dev_priv)) { + clock.n = ffs((fp & FP_N_PINEVIEW_DIV_MASK) >> FP_N_DIV_SHIFT) - 1; + clock.m2 = (fp & FP_M2_PINEVIEW_DIV_MASK) >> FP_M2_DIV_SHIFT; + } else { + clock.n = (fp & FP_N_DIV_MASK) >> FP_N_DIV_SHIFT; + clock.m2 = (fp & FP_M2_DIV_MASK) >> FP_M2_DIV_SHIFT; } - /* Find an unused one (if possible) */ - for_each_intel_crtc(dev, possible_crtc) { - if (!(encoder->base.possible_crtcs & - drm_crtc_mask(&possible_crtc->base))) - continue; - - ret = drm_modeset_lock(&possible_crtc->base.mutex, ctx); - if (ret) - goto fail; + if (DISPLAY_VER(dev_priv) != 2) { + if (IS_PINEVIEW(dev_priv)) + clock.p1 = ffs((dpll & DPLL_FPA01_P1_POST_DIV_MASK_PINEVIEW) >> + DPLL_FPA01_P1_POST_DIV_SHIFT_PINEVIEW); + else + clock.p1 = ffs((dpll & DPLL_FPA01_P1_POST_DIV_MASK) >> + DPLL_FPA01_P1_POST_DIV_SHIFT); - if (possible_crtc->base.state->enable) { - drm_modeset_unlock(&possible_crtc->base.mutex); - continue; + switch (dpll & DPLL_MODE_MASK) { + case DPLLB_MODE_DAC_SERIAL: + clock.p2 = dpll & DPLL_DAC_SERIAL_P2_CLOCK_DIV_5 ? + 5 : 10; + break; + case DPLLB_MODE_LVDS: + clock.p2 = dpll & DPLLB_LVDS_P2_CLOCK_DIV_7 ? + 7 : 14; + break; + default: + drm_dbg_kms(&dev_priv->drm, + "Unknown DPLL mode %08x in programmed " + "mode\n", (int)(dpll & DPLL_MODE_MASK)); + return; } - crtc = possible_crtc; - break; - } - - /* - * If we didn't find an unused CRTC, don't use any. - */ - if (!crtc) { - drm_dbg_kms(&dev_priv->drm, - "no pipe available for load-detect\n"); - ret = -ENODEV; - goto fail; - } - -found: - state = drm_atomic_state_alloc(dev); - restore_state = drm_atomic_state_alloc(dev); - if (!state || !restore_state) { - ret = -ENOMEM; - goto fail; - } - - state->acquire_ctx = ctx; - restore_state->acquire_ctx = ctx; - - connector_state = drm_atomic_get_connector_state(state, connector); - if (IS_ERR(connector_state)) { - ret = PTR_ERR(connector_state); - goto fail; - } - - ret = drm_atomic_set_crtc_for_connector(connector_state, &crtc->base); - if (ret) - goto fail; - - crtc_state = intel_atomic_get_crtc_state(state, crtc); - if (IS_ERR(crtc_state)) { - ret = PTR_ERR(crtc_state); - goto fail; - } - - crtc_state->uapi.active = true; - - ret = drm_atomic_set_mode_for_crtc(&crtc_state->uapi, - &load_detect_mode); - if (ret) - goto fail; - - ret = intel_modeset_disable_planes(state, &crtc->base); - if (ret) - goto fail; - - ret = PTR_ERR_OR_ZERO(drm_atomic_get_connector_state(restore_state, connector)); - if (!ret) - ret = PTR_ERR_OR_ZERO(drm_atomic_get_crtc_state(restore_state, &crtc->base)); - if (!ret) - ret = drm_atomic_add_affected_planes(restore_state, &crtc->base); - if (ret) { - drm_dbg_kms(&dev_priv->drm, - "Failed to create a copy of old state to restore: %i\n", - ret); - goto fail; - } - - ret = drm_atomic_commit(state); - if (ret) { - drm_dbg_kms(&dev_priv->drm, - "failed to set mode on load-detect pipe\n"); - goto fail; - } - - old->restore_state = restore_state; - drm_atomic_state_put(state); - - /* let the connector get through one full cycle before testing */ - intel_crtc_wait_for_next_vblank(crtc); - - return true; - -fail: - if (state) { - drm_atomic_state_put(state); - state = NULL; - } - if (restore_state) { - drm_atomic_state_put(restore_state); - restore_state = NULL; - } - - if (ret == -EDEADLK) - return ret; - - return false; -} - -void intel_release_load_detect_pipe(struct drm_connector *connector, - struct intel_load_detect_pipe *old, - struct drm_modeset_acquire_ctx *ctx) -{ - struct intel_encoder *intel_encoder = - intel_attached_encoder(to_intel_connector(connector)); - struct drm_i915_private *i915 = to_i915(intel_encoder->base.dev); - struct drm_encoder *encoder = &intel_encoder->base; - struct drm_atomic_state *state = old->restore_state; - int ret; - - drm_dbg_kms(&i915->drm, "[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n", - connector->base.id, connector->name, - encoder->base.id, encoder->name); - - if (!state) - return; - - ret = drm_atomic_helper_commit_duplicated_state(state, ctx); - if (ret) - drm_dbg_kms(&i915->drm, - "Couldn't release load detect pipe: %i\n", ret); - drm_atomic_state_put(state); -} - -static int i9xx_pll_refclk(struct drm_device *dev, - const struct intel_crtc_state *pipe_config) -{ - struct drm_i915_private *dev_priv = to_i915(dev); - u32 dpll = pipe_config->dpll_hw_state.dpll; - - if ((dpll & PLL_REF_INPUT_MASK) == PLLB_REF_INPUT_SPREADSPECTRUMIN) - return dev_priv->display.vbt.lvds_ssc_freq; - else if (HAS_PCH_SPLIT(dev_priv)) - return 120000; - else if (DISPLAY_VER(dev_priv) != 2) - return 96000; - else - return 48000; -} - -/* Returns the clock of the currently programmed mode of the given pipe. */ -void i9xx_crtc_clock_get(struct intel_crtc *crtc, - struct intel_crtc_state *pipe_config) -{ - struct drm_device *dev = crtc->base.dev; - struct drm_i915_private *dev_priv = to_i915(dev); - u32 dpll = pipe_config->dpll_hw_state.dpll; - u32 fp; - struct dpll clock; - int port_clock; - int refclk = i9xx_pll_refclk(dev, pipe_config); - - if ((dpll & DISPLAY_RATE_SELECT_FPA1) == 0) - fp = pipe_config->dpll_hw_state.fp0; - else - fp = pipe_config->dpll_hw_state.fp1; - - clock.m1 = (fp & FP_M1_DIV_MASK) >> FP_M1_DIV_SHIFT; - if (IS_PINEVIEW(dev_priv)) { - clock.n = ffs((fp & FP_N_PINEVIEW_DIV_MASK) >> FP_N_DIV_SHIFT) - 1; - clock.m2 = (fp & FP_M2_PINEVIEW_DIV_MASK) >> FP_M2_DIV_SHIFT; - } else { - clock.n = (fp & FP_N_DIV_MASK) >> FP_N_DIV_SHIFT; - clock.m2 = (fp & FP_M2_DIV_MASK) >> FP_M2_DIV_SHIFT; - } - - if (DISPLAY_VER(dev_priv) != 2) { - if (IS_PINEVIEW(dev_priv)) - clock.p1 = ffs((dpll & DPLL_FPA01_P1_POST_DIV_MASK_PINEVIEW) >> - DPLL_FPA01_P1_POST_DIV_SHIFT_PINEVIEW); - else - clock.p1 = ffs((dpll & DPLL_FPA01_P1_POST_DIV_MASK) >> - DPLL_FPA01_P1_POST_DIV_SHIFT); - - switch (dpll & DPLL_MODE_MASK) { - case DPLLB_MODE_DAC_SERIAL: - clock.p2 = dpll & DPLL_DAC_SERIAL_P2_CLOCK_DIV_5 ? - 5 : 10; - break; - case DPLLB_MODE_LVDS: - clock.p2 = dpll & DPLLB_LVDS_P2_CLOCK_DIV_7 ? - 7 : 14; - break; - default: - drm_dbg_kms(&dev_priv->drm, - "Unknown DPLL mode %08x in programmed " - "mode\n", (int)(dpll & DPLL_MODE_MASK)); - return; - } - - if (IS_PINEVIEW(dev_priv)) - port_clock = pnv_calc_dpll_params(refclk, &clock); - else - port_clock = i9xx_calc_dpll_params(refclk, &clock); - } else { - enum pipe lvds_pipe; + if (IS_PINEVIEW(dev_priv)) + port_clock = pnv_calc_dpll_params(refclk, &clock); + else + port_clock = i9xx_calc_dpll_params(refclk, &clock); + } else { + enum pipe lvds_pipe; if (IS_I85X(dev_priv) && intel_lvds_port_enabled(dev_priv, LVDS, &lvds_pipe) && @@ -5269,7 +4873,7 @@ pipe_config_infoframe_mismatch(struct drm_i915_private *dev_priv, return; drm_dbg_kms(&dev_priv->drm, - "fastset mismatch in %s infoframe\n", name); + "fastset requirement not met in %s infoframe\n", name); drm_dbg_kms(&dev_priv->drm, "expected:\n"); hdmi_infoframe_log(KERN_DEBUG, dev_priv->drm.dev, a); drm_dbg_kms(&dev_priv->drm, "found:\n"); @@ -5294,7 +4898,7 @@ pipe_config_dp_vsc_sdp_mismatch(struct drm_i915_private *dev_priv, return; drm_dbg_kms(&dev_priv->drm, - "fastset mismatch in %s dp sdp\n", name); + "fastset requirement not met in %s dp sdp\n", name); drm_dbg_kms(&dev_priv->drm, "expected:\n"); drm_dp_vsc_sdp_log(KERN_DEBUG, dev_priv->drm.dev, a); drm_dbg_kms(&dev_priv->drm, "found:\n"); @@ -5335,7 +4939,7 @@ pipe_config_buffer_mismatch(struct drm_i915_private *dev_priv, len = memcmp_diff_len(a, b, len); drm_dbg_kms(&dev_priv->drm, - "fastset mismatch in %s buffer\n", name); + "fastset requirement not met in %s buffer\n", name); print_hex_dump(KERN_DEBUG, "expected: ", DUMP_PREFIX_NONE, 16, 0, a, len, false); print_hex_dump(KERN_DEBUG, "found: ", DUMP_PREFIX_NONE, @@ -5366,7 +4970,7 @@ pipe_config_mismatch(bool fastset, const struct intel_crtc *crtc, if (fastset) drm_dbg_kms(&i915->drm, - "[CRTC:%d:%s] fastset mismatch in %s %pV\n", + "[CRTC:%d:%s] fastset requirement not met in %s %pV\n", crtc->base.base.id, crtc->base.name, name, &vaf); else drm_err(&i915->drm, "[CRTC:%d:%s] mismatch in %s %pV\n", @@ -5573,6 +5177,24 @@ intel_pipe_config_compare(const struct intel_crtc_state *current_config, } \ } while (0) +#define PIPE_CONF_CHECK_CSC(name) do { \ + PIPE_CONF_CHECK_X(name.preoff[0]); \ + PIPE_CONF_CHECK_X(name.preoff[1]); \ + PIPE_CONF_CHECK_X(name.preoff[2]); \ + PIPE_CONF_CHECK_X(name.coeff[0]); \ + PIPE_CONF_CHECK_X(name.coeff[1]); \ + PIPE_CONF_CHECK_X(name.coeff[2]); \ + PIPE_CONF_CHECK_X(name.coeff[3]); \ + PIPE_CONF_CHECK_X(name.coeff[4]); \ + PIPE_CONF_CHECK_X(name.coeff[5]); \ + PIPE_CONF_CHECK_X(name.coeff[6]); \ + PIPE_CONF_CHECK_X(name.coeff[7]); \ + PIPE_CONF_CHECK_X(name.coeff[8]); \ + PIPE_CONF_CHECK_X(name.postoff[0]); \ + PIPE_CONF_CHECK_X(name.postoff[1]); \ + PIPE_CONF_CHECK_X(name.postoff[2]); \ +} while (0) + #define PIPE_CONF_QUIRK(quirk) \ ((current_config->quirks | pipe_config->quirks) & (quirk)) @@ -5663,6 +5285,7 @@ intel_pipe_config_compare(const struct intel_crtc_state *current_config, PIPE_CONF_CHECK_X(csc_mode); PIPE_CONF_CHECK_BOOL(gamma_enable); PIPE_CONF_CHECK_BOOL(csc_enable); + PIPE_CONF_CHECK_BOOL(wgc_enable); PIPE_CONF_CHECK_I(linetime); PIPE_CONF_CHECK_I(ips_linetime); @@ -5670,6 +5293,9 @@ intel_pipe_config_compare(const struct intel_crtc_state *current_config, PIPE_CONF_CHECK_COLOR_LUT(pre_csc_lut, true); PIPE_CONF_CHECK_COLOR_LUT(post_csc_lut, false); + PIPE_CONF_CHECK_CSC(csc); + PIPE_CONF_CHECK_CSC(output_csc); + if (current_config->active_planes) { PIPE_CONF_CHECK_BOOL(has_psr); PIPE_CONF_CHECK_BOOL(has_psr2); @@ -5756,7 +5382,8 @@ intel_pipe_config_compare(const struct intel_crtc_state *current_config, PIPE_CONF_CHECK_I(splitter.link_count); PIPE_CONF_CHECK_I(splitter.pixel_overlap); - PIPE_CONF_CHECK_BOOL(vrr.enable); + if (!fastset) + PIPE_CONF_CHECK_BOOL(vrr.enable); PIPE_CONF_CHECK_I(vrr.vmin); PIPE_CONF_CHECK_I(vrr.vmax); PIPE_CONF_CHECK_I(vrr.flipline); @@ -5932,8 +5559,13 @@ static int intel_modeset_checks(struct intel_atomic_state *state) static void intel_crtc_check_fastset(const struct intel_crtc_state *old_crtc_state, struct intel_crtc_state *new_crtc_state) { - if (!intel_pipe_config_compare(old_crtc_state, new_crtc_state, true)) + struct drm_i915_private *i915 = to_i915(old_crtc_state->uapi.crtc->dev); + + if (!intel_pipe_config_compare(old_crtc_state, new_crtc_state, true)) { + drm_dbg_kms(&i915->drm, "fastset requirement not met, forcing full modeset\n"); + return; + } new_crtc_state->uapi.mode_changed = false; if (!intel_crtc_needs_modeset(new_crtc_state)) @@ -5986,7 +5618,7 @@ static int intel_crtc_add_bigjoiner_planes(struct intel_atomic_state *state, struct intel_crtc *crtc, struct intel_crtc *other) { - const struct intel_plane_state *plane_state; + const struct intel_plane_state __maybe_unused *plane_state; struct intel_plane *plane; u8 plane_ids = 0; int i; @@ -6029,7 +5661,7 @@ static int intel_atomic_check_planes(struct intel_atomic_state *state) { struct drm_i915_private *dev_priv = to_i915(state->base.dev); struct intel_crtc_state *old_crtc_state, *new_crtc_state; - struct intel_plane_state *plane_state; + struct intel_plane_state __maybe_unused *plane_state; struct intel_plane *plane; struct intel_crtc *crtc; int i, ret; @@ -6084,7 +5716,7 @@ static int intel_atomic_check_planes(struct intel_atomic_state *state) static int intel_atomic_check_crtcs(struct intel_atomic_state *state) { - struct intel_crtc_state *crtc_state; + struct intel_crtc_state __maybe_unused *crtc_state; struct intel_crtc *crtc; int i; @@ -6372,6 +6004,22 @@ static int intel_async_flip_check_hw(struct intel_atomic_state *state, struct in * this selectively if required. */ switch (new_plane_state->hw.fb->modifier) { + case DRM_FORMAT_MOD_LINEAR: + /* + * FIXME: Async on Linear buffer is supported on ICL as + * but with additional alignment and fbc restrictions + * need to be taken care of. These aren't applicable for + * gen12+. + */ + if (DISPLAY_VER(i915) < 12) { + drm_dbg_kms(&i915->drm, + "[PLANE:%d:%s] Modifier 0x%llx does not support async flip on display ver %d\n", + plane->base.base.id, plane->base.name, + new_plane_state->hw.fb->modifier, DISPLAY_VER(i915)); + return -EINVAL; + } + break; + case I915_FORMAT_MOD_X_TILED: case I915_FORMAT_MOD_Y_TILED: case I915_FORMAT_MOD_Yf_TILED: @@ -6379,8 +6027,9 @@ static int intel_async_flip_check_hw(struct intel_atomic_state *state, struct in break; default: drm_dbg_kms(&i915->drm, - "[PLANE:%d:%s] Modifier does not support async flips\n", - plane->base.base.id, plane->base.name); + "[PLANE:%d:%s] Modifier 0x%llx does not support async flip\n", + plane->base.base.id, plane->base.name, + new_plane_state->hw.fb->modifier); return -EINVAL; } @@ -6536,6 +6185,13 @@ int intel_atomic_check(struct drm_device *dev, for_each_oldnew_intel_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { + /* + * crtc's state no longer considered to be inherited + * after the first userspace/client initiated commit. + */ + if (!state->internal) + new_crtc_state->inherited = false; + if (new_crtc_state->inherited != old_crtc_state->inherited) new_crtc_state->uapi.mode_changed = true; @@ -6699,6 +6355,10 @@ int intel_atomic_check(struct drm_device *dev, return ret; } + ret = intel_pmdemand_atomic_check(state); + if (ret) + goto fail; + ret = intel_atomic_check_crtcs(state); if (ret) goto fail; @@ -6881,7 +6541,8 @@ static void intel_enable_crtc(struct intel_atomic_state *state, if (!intel_crtc_needs_modeset(new_crtc_state)) return; - intel_crtc_update_active_timings(new_crtc_state); + /* VRR will be enable later, if required */ + intel_crtc_update_active_timings(new_crtc_state, false); dev_priv->display.funcs.display->crtc_enable(state, crtc); @@ -6908,6 +6569,12 @@ static void intel_update_crtc(struct intel_atomic_state *state, intel_dpt_configure(crtc); } + if (vrr_enabling(old_crtc_state, new_crtc_state)) { + intel_vrr_enable(new_crtc_state); + intel_crtc_update_active_timings(new_crtc_state, + new_crtc_state->vrr.enable); + } + if (!modeset) { if (new_crtc_state->preload_luts && intel_crtc_needs_color_update(new_crtc_state)) @@ -7181,7 +6848,7 @@ static void intel_atomic_helper_free_state(struct drm_i915_private *dev_priv) drm_atomic_state_put(&state->base); } -static void intel_atomic_helper_free_state_worker(struct work_struct *work) +void intel_atomic_helper_free_state_worker(struct work_struct *work) { struct drm_i915_private *dev_priv = container_of(work, typeof(*dev_priv), display.atomic_helper.free_work); @@ -7313,6 +6980,12 @@ static void intel_atomic_commit_tail(struct intel_atomic_state *state) * 7. New _arm() registers are finally written * 8. Hardware finally latches a complete set of new * register values, and subsequent frames will be OK again + * + * Also note that due to the pipe CSC hardware issues on + * SKL/GLK DC states must remain off until the pipe CSC + * state readout has happened. Otherwise we risk corrupting + * the CSC latched register values with the readout (see + * skl_read_csc() and skl_color_commit_noarm()). */ wakeref = intel_display_power_get(dev_priv, POWER_DOMAIN_DC_OFF); @@ -7331,6 +7004,14 @@ static void intel_atomic_commit_tail(struct intel_atomic_state *state) for_each_new_intel_crtc_in_state(state, crtc, new_crtc_state, i) crtc->config = new_crtc_state; + /* + * In XE_LPD+ Pmdemand combines many parameters such as voltage index, + * plls, cdclk frequency, QGV point selection parameter etc. Voltage + * index, cdclk/ddiclk frequencies are supposed to be configured before + * the cdclk config is set. + */ + intel_pmdemand_pre_plane_update(state); + if (state->modeset) { drm_atomic_helper_update_legacy_modeset_state(dev, &state->base); @@ -7450,6 +7131,7 @@ static void intel_atomic_commit_tail(struct intel_atomic_state *state) intel_verify_planes(state); intel_sagv_post_plane_update(state); + intel_pmdemand_post_plane_update(state); drm_atomic_helper_commit_hw_done(&state->base); @@ -7498,11 +7180,12 @@ intel_atomic_commit_ready(struct i915_sw_fence *fence, break; case FENCE_FREE: { + struct drm_i915_private *i915 = to_i915(state->base.dev); struct intel_atomic_helper *helper = - &to_i915(state->base.dev)->display.atomic_helper; + &i915->display.atomic_helper; if (llist_add(&state->freed, &helper->free_list)) - schedule_work(&helper->free_work); + queue_work(i915->unordered_wq, &helper->free_work); break; } } @@ -7523,9 +7206,8 @@ static void intel_atomic_track_fbs(struct intel_atomic_state *state) plane->frontbuffer_bit); } -static int intel_atomic_commit(struct drm_device *dev, - struct drm_atomic_state *_state, - bool nonblock) +int intel_atomic_commit(struct drm_device *dev, struct drm_atomic_state *_state, + bool nonblock) { struct intel_atomic_state *state = to_intel_atomic_state(_state); struct drm_i915_private *dev_priv = to_i915(dev); @@ -7627,19 +7309,6 @@ void intel_plane_destroy(struct drm_plane *plane) kfree(to_intel_plane(plane)); } -static void intel_plane_possible_crtcs_init(struct drm_i915_private *dev_priv) -{ - struct intel_plane *plane; - - for_each_intel_plane(&dev_priv->drm, plane) { - struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv, - plane->pipe); - - plane->base.possible_crtcs = drm_crtc_mask(&crtc->base); - } -} - - int intel_get_pipe_from_crtc_id_ioctl(struct drm_device *dev, void *data, struct drm_file *file) { @@ -7719,7 +7388,7 @@ static bool intel_ddi_crt_present(struct drm_i915_private *dev_priv) return true; } -static void intel_setup_outputs(struct drm_i915_private *dev_priv) +void intel_setup_outputs(struct drm_i915_private *dev_priv) { struct intel_encoder *encoder; bool dpd_is_edp = false; @@ -7729,7 +7398,14 @@ static void intel_setup_outputs(struct drm_i915_private *dev_priv) if (!HAS_DISPLAY(dev_priv)) return; - if (IS_DG2(dev_priv)) { + if (IS_METEORLAKE(dev_priv)) { + intel_ddi_init(dev_priv, PORT_A); + intel_ddi_init(dev_priv, PORT_B); + intel_ddi_init(dev_priv, PORT_TC1); + intel_ddi_init(dev_priv, PORT_TC2); + intel_ddi_init(dev_priv, PORT_TC3); + intel_ddi_init(dev_priv, PORT_TC4); + } else if (IS_DG2(dev_priv)) { intel_ddi_init(dev_priv, PORT_A); intel_ddi_init(dev_priv, PORT_B); intel_ddi_init(dev_priv, PORT_C); @@ -7971,9 +7647,8 @@ static int max_dotclock(struct drm_i915_private *i915) return max_dotclock; } -static enum drm_mode_status -intel_mode_valid(struct drm_device *dev, - const struct drm_display_mode *mode) +enum drm_mode_status intel_mode_valid(struct drm_device *dev, + const struct drm_display_mode *mode) { struct drm_i915_private *dev_priv = to_i915(dev); int hdisplay_max, htotal_max; @@ -8113,18 +7788,6 @@ intel_mode_valid_max_plane_size(struct drm_i915_private *dev_priv, return MODE_OK; } -static const struct drm_mode_config_funcs intel_mode_funcs = { - .fb_create = intel_user_framebuffer_create, - .get_format_info = intel_fb_get_format_info, - .output_poll_changed = intel_fbdev_output_poll_changed, - .mode_valid = intel_mode_valid, - .atomic_check = intel_atomic_check, - .atomic_commit = intel_atomic_commit, - .atomic_state_alloc = intel_atomic_state_alloc, - .atomic_state_clear = intel_atomic_state_clear, - .atomic_state_free = intel_atomic_state_free, -}; - static const struct intel_display_funcs skl_display_funcs = { .get_pipe_config = hsw_get_pipe_config, .crtc_enable = hsw_crtc_enable, @@ -8171,15 +7834,6 @@ static const struct intel_display_funcs i9xx_display_funcs = { */ void intel_init_display_hooks(struct drm_i915_private *dev_priv) { - if (!HAS_DISPLAY(dev_priv)) - return; - - intel_color_init_hooks(dev_priv); - intel_init_cdclk_hooks(dev_priv); - intel_audio_hooks_init(dev_priv); - - intel_dpll_init_clock_hook(dev_priv); - if (DISPLAY_VER(dev_priv) >= 9) { dev_priv->display.funcs.display = &skl_display_funcs; } else if (HAS_DDI(dev_priv)) { @@ -8192,25 +7846,9 @@ void intel_init_display_hooks(struct drm_i915_private *dev_priv) } else { dev_priv->display.funcs.display = &i9xx_display_funcs; } - - intel_fdi_init_hook(dev_priv); } -void intel_modeset_init_hw(struct drm_i915_private *i915) -{ - struct intel_cdclk_state *cdclk_state; - - if (!HAS_DISPLAY(i915)) - return; - - cdclk_state = to_intel_cdclk_state(i915->display.cdclk.obj.state); - - intel_update_cdclk(i915); - intel_cdclk_dump_config(i915, &i915->display.cdclk.hw, "Current CDCLK"); - cdclk_state->logical = cdclk_state->actual = i915->display.cdclk.hw; -} - -static int intel_initial_commit(struct drm_device *dev) +int intel_initial_commit(struct drm_device *dev) { struct drm_atomic_state *state = NULL; struct drm_modeset_acquire_ctx ctx; @@ -8223,9 +7861,10 @@ static int intel_initial_commit(struct drm_device *dev) drm_modeset_acquire_init(&ctx, 0); -retry: state->acquire_ctx = &ctx; + to_intel_atomic_state(state)->internal = true; +retry: for_each_intel_crtc(dev, crtc) { struct intel_crtc_state *crtc_state = intel_atomic_get_crtc_state(state, crtc); @@ -8238,15 +7877,6 @@ static int intel_initial_commit(struct drm_device *dev) if (crtc_state->hw.active) { struct intel_encoder *encoder; - /* - * We've not yet detected sink capabilities - * (audio,infoframes,etc.) and thus we don't want to - * force a full state recomputation yet. We want that to - * happen only for the first real commit from userspace. - * So preserve the inherited flag for the time being. - */ - crtc_state->inherited = true; - ret = drm_atomic_add_affected_planes(state, &crtc->base); if (ret) goto out; @@ -8289,246 +7919,6 @@ static int intel_initial_commit(struct drm_device *dev) return ret; } -static const struct drm_mode_config_helper_funcs intel_mode_config_funcs = { - .atomic_commit_setup = drm_dp_mst_atomic_setup_commit, -}; - -static void intel_mode_config_init(struct drm_i915_private *i915) -{ - struct drm_mode_config *mode_config = &i915->drm.mode_config; - - drm_mode_config_init(&i915->drm); - INIT_LIST_HEAD(&i915->display.global.obj_list); - - mode_config->min_width = 0; - mode_config->min_height = 0; - - mode_config->preferred_depth = 24; - mode_config->prefer_shadow = 1; - - mode_config->funcs = &intel_mode_funcs; - mode_config->helper_private = &intel_mode_config_funcs; - - mode_config->async_page_flip = HAS_ASYNC_FLIPS(i915); - - /* - * Maximum framebuffer dimensions, chosen to match - * the maximum render engine surface size on gen4+. - */ - if (DISPLAY_VER(i915) >= 7) { - mode_config->max_width = 16384; - mode_config->max_height = 16384; - } else if (DISPLAY_VER(i915) >= 4) { - mode_config->max_width = 8192; - mode_config->max_height = 8192; - } else if (DISPLAY_VER(i915) == 3) { - mode_config->max_width = 4096; - mode_config->max_height = 4096; - } else { - mode_config->max_width = 2048; - mode_config->max_height = 2048; - } - - if (IS_I845G(i915) || IS_I865G(i915)) { - mode_config->cursor_width = IS_I845G(i915) ? 64 : 512; - mode_config->cursor_height = 1023; - } else if (IS_I830(i915) || IS_I85X(i915) || - IS_I915G(i915) || IS_I915GM(i915)) { - mode_config->cursor_width = 64; - mode_config->cursor_height = 64; - } else { - mode_config->cursor_width = 256; - mode_config->cursor_height = 256; - } -} - -static void intel_mode_config_cleanup(struct drm_i915_private *i915) -{ - intel_atomic_global_obj_cleanup(i915); - drm_mode_config_cleanup(&i915->drm); -} - -/* part #1: call before irq install */ -int intel_modeset_init_noirq(struct drm_i915_private *i915) -{ - int ret; - - if (i915_inject_probe_failure(i915)) - return -ENODEV; - - if (HAS_DISPLAY(i915)) { - ret = drm_vblank_init(&i915->drm, - INTEL_NUM_PIPES(i915)); - if (ret) - return ret; - } - - intel_bios_init(i915); - - ret = intel_vga_register(i915); - if (ret) - goto cleanup_bios; - - /* FIXME: completely on the wrong abstraction layer */ - ret = intel_power_domains_init(i915); - if (ret < 0) - goto cleanup_vga; - - intel_power_domains_init_hw(i915, false); - - if (!HAS_DISPLAY(i915)) - return 0; - - intel_dmc_init(i915); - - i915->display.wq.modeset = alloc_ordered_workqueue("i915_modeset", 0); - i915->display.wq.flip = alloc_workqueue("i915_flip", WQ_HIGHPRI | - WQ_UNBOUND, WQ_UNBOUND_MAX_ACTIVE); - - intel_mode_config_init(i915); - - ret = intel_cdclk_init(i915); - if (ret) - goto cleanup_vga_client_pw_domain_dmc; - - ret = intel_color_init(i915); - if (ret) - goto cleanup_vga_client_pw_domain_dmc; - - ret = intel_dbuf_init(i915); - if (ret) - goto cleanup_vga_client_pw_domain_dmc; - - ret = intel_bw_init(i915); - if (ret) - goto cleanup_vga_client_pw_domain_dmc; - - init_llist_head(&i915->display.atomic_helper.free_list); - INIT_WORK(&i915->display.atomic_helper.free_work, - intel_atomic_helper_free_state_worker); - - intel_init_quirks(i915); - - intel_fbc_init(i915); - - return 0; - -cleanup_vga_client_pw_domain_dmc: - intel_dmc_fini(i915); - intel_power_domains_driver_remove(i915); -cleanup_vga: - intel_vga_unregister(i915); -cleanup_bios: - intel_bios_driver_remove(i915); - - return ret; -} - -/* part #2: call after irq install, but before gem init */ -int intel_modeset_init_nogem(struct drm_i915_private *i915) -{ - struct drm_device *dev = &i915->drm; - enum pipe pipe; - struct intel_crtc *crtc; - int ret; - - if (!HAS_DISPLAY(i915)) - return 0; - - intel_wm_init(i915); - - intel_panel_sanitize_ssc(i915); - - intel_pps_setup(i915); - - intel_gmbus_setup(i915); - - drm_dbg_kms(&i915->drm, "%d display pipe%s available.\n", - INTEL_NUM_PIPES(i915), - INTEL_NUM_PIPES(i915) > 1 ? "s" : ""); - - for_each_pipe(i915, pipe) { - ret = intel_crtc_init(i915, pipe); - if (ret) { - intel_mode_config_cleanup(i915); - return ret; - } - } - - intel_plane_possible_crtcs_init(i915); - intel_shared_dpll_init(i915); - intel_fdi_pll_freq_update(i915); - - intel_update_czclk(i915); - intel_modeset_init_hw(i915); - intel_dpll_update_ref_clks(i915); - - intel_hdcp_component_init(i915); - - if (i915->display.cdclk.max_cdclk_freq == 0) - intel_update_max_cdclk(i915); - - intel_hti_init(i915); - - /* Just disable it once at startup */ - intel_vga_disable(i915); - intel_setup_outputs(i915); - - drm_modeset_lock_all(dev); - intel_modeset_setup_hw_state(i915, dev->mode_config.acquire_ctx); - intel_acpi_assign_connector_fwnodes(i915); - drm_modeset_unlock_all(dev); - - for_each_intel_crtc(dev, crtc) { - if (!to_intel_crtc_state(crtc->base.state)->uapi.active) - continue; - intel_crtc_initial_plane_config(crtc); - } - - /* - * Make sure hardware watermarks really match the state we read out. - * Note that we need to do this after reconstructing the BIOS fb's - * since the watermark calculation done here will use pstate->fb. - */ - if (!HAS_GMCH(i915)) - ilk_wm_sanitize(i915); - - return 0; -} - -/* part #3: call after gem init */ -int intel_modeset_init(struct drm_i915_private *i915) -{ - int ret; - - if (!HAS_DISPLAY(i915)) - return 0; - - /* - * Force all active planes to recompute their states. So that on - * mode_setcrtc after probe, all the intel_plane_state variables - * are already calculated and there is no assert_plane warnings - * during bootup. - */ - ret = intel_initial_commit(&i915->drm); - if (ret) - drm_dbg_kms(&i915->drm, "Initial modeset failed, %d\n", ret); - - intel_overlay_setup(i915); - - ret = intel_fbdev_init(&i915->drm); - if (ret) - return ret; - - /* Only enable hotplug handling once the fbdev is fully set up. */ - intel_hpd_init(i915); - intel_hpd_poll_disable(i915); - - skl_watermark_ipc_init(i915); - - return 0; -} - void i830_enable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe) { struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv, pipe); @@ -8636,45 +8026,7 @@ void i830_disable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe) intel_de_posting_read(dev_priv, DPLL(pipe)); } -void intel_display_resume(struct drm_device *dev) -{ - struct drm_i915_private *i915 = to_i915(dev); - struct drm_atomic_state *state = i915->display.restore.modeset_state; - struct drm_modeset_acquire_ctx ctx; - int ret; - - if (!HAS_DISPLAY(i915)) - return; - - i915->display.restore.modeset_state = NULL; - if (state) - state->acquire_ctx = &ctx; - - drm_modeset_acquire_init(&ctx, 0); - - while (1) { - ret = drm_modeset_lock_all_ctx(dev, &ctx); - if (ret != -EDEADLK) - break; - - drm_modeset_backoff(&ctx); - } - - if (!ret) - ret = __intel_display_resume(i915, state, &ctx); - - skl_watermark_ipc_update(i915); - drm_modeset_drop_locks(&ctx); - drm_modeset_acquire_fini(&ctx); - - if (ret) - drm_err(&i915->drm, - "Restoring old state failed with %i\n", ret); - if (state) - drm_atomic_state_put(state); -} - -static void intel_hpd_poll_fini(struct drm_i915_private *i915) +void intel_hpd_poll_fini(struct drm_i915_private *i915) { struct intel_connector *connector; struct drm_connector_list_iter conn_iter; @@ -8692,144 +8044,6 @@ static void intel_hpd_poll_fini(struct drm_i915_private *i915) drm_connector_list_iter_end(&conn_iter); } -/* part #1: call before irq uninstall */ -void intel_modeset_driver_remove(struct drm_i915_private *i915) -{ - if (!HAS_DISPLAY(i915)) - return; - - flush_workqueue(i915->display.wq.flip); - flush_workqueue(i915->display.wq.modeset); - - flush_work(&i915->display.atomic_helper.free_work); - drm_WARN_ON(&i915->drm, !llist_empty(&i915->display.atomic_helper.free_list)); - - /* - * MST topology needs to be suspended so we don't have any calls to - * fbdev after it's finalized. MST will be destroyed later as part of - * drm_mode_config_cleanup() - */ - intel_dp_mst_suspend(i915); -} - -/* part #2: call after irq uninstall */ -void intel_modeset_driver_remove_noirq(struct drm_i915_private *i915) -{ - if (!HAS_DISPLAY(i915)) - return; - - /* - * Due to the hpd irq storm handling the hotplug work can re-arm the - * poll handlers. Hence disable polling after hpd handling is shut down. - */ - intel_hpd_poll_fini(i915); - - /* poll work can call into fbdev, hence clean that up afterwards */ - intel_fbdev_fini(i915); - - intel_unregister_dsm_handler(); - - /* flush any delayed tasks or pending work */ - flush_scheduled_work(); - - intel_hdcp_component_fini(i915); - - intel_mode_config_cleanup(i915); - - intel_overlay_cleanup(i915); - - intel_gmbus_teardown(i915); - - destroy_workqueue(i915->display.wq.flip); - destroy_workqueue(i915->display.wq.modeset); - - intel_fbc_cleanup(i915); -} - -/* part #3: call after gem init */ -void intel_modeset_driver_remove_nogem(struct drm_i915_private *i915) -{ - intel_dmc_fini(i915); - - intel_power_domains_driver_remove(i915); - - intel_vga_unregister(i915); - - intel_bios_driver_remove(i915); -} - -bool intel_modeset_probe_defer(struct pci_dev *pdev) -{ - struct drm_privacy_screen *privacy_screen; - - /* - * apple-gmux is needed on dual GPU MacBook Pro - * to probe the panel if we're the inactive GPU. - */ - if (vga_switcheroo_client_probe_defer(pdev)) - return true; - - /* If the LCD panel has a privacy-screen, wait for it */ - privacy_screen = drm_privacy_screen_get(&pdev->dev, NULL); - if (IS_ERR(privacy_screen) && PTR_ERR(privacy_screen) == -EPROBE_DEFER) - return true; - - drm_privacy_screen_put(privacy_screen); - - return false; -} - -void intel_display_driver_register(struct drm_i915_private *i915) -{ - if (!HAS_DISPLAY(i915)) - return; - - /* Must be done after probing outputs */ - intel_opregion_register(i915); - intel_acpi_video_register(i915); - - intel_audio_init(i915); - - intel_display_debugfs_register(i915); - - /* - * Some ports require correctly set-up hpd registers for - * detection to work properly (leading to ghost connected - * connector status), e.g. VGA on gm45. Hence we can only set - * up the initial fbdev config after hpd irqs are fully - * enabled. We do it last so that the async config cannot run - * before the connectors are registered. - */ - intel_fbdev_initial_config_async(i915); - - /* - * We need to coordinate the hotplugs with the asynchronous - * fbdev configuration, for which we use the - * fbdev->async_cookie. - */ - drm_kms_helper_poll_init(&i915->drm); -} - -void intel_display_driver_unregister(struct drm_i915_private *i915) -{ - if (!HAS_DISPLAY(i915)) - return; - - intel_fbdev_unregister(i915); - intel_audio_deinit(i915); - - /* - * After flushing the fbdev (incl. a late async config which - * will have delayed queuing of a hotplug event), then flush - * the hotplug events. - */ - drm_kms_helper_poll_fini(&i915->drm); - drm_atomic_helper_shutdown(&i915->drm); - - acpi_video_unregister(); - intel_opregion_unregister(i915); -} - bool intel_scanout_needs_vtd_wa(struct drm_i915_private *i915) { return DISPLAY_VER(i915) >= 6 && i915_vtd_active(i915); diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h index 287159bdeb0d1c685eb42c3b079cc6f3f46c9b75..c744c021af23689926d949faac7f9c8026b42a05 100644 --- a/drivers/gpu/drm/i915/display/intel_display.h +++ b/drivers/gpu/drm/i915/display/intel_display.h @@ -56,13 +56,13 @@ struct intel_dp; struct intel_encoder; struct intel_initial_plane_config; struct intel_link_m_n; -struct intel_load_detect_pipe; struct intel_plane; struct intel_plane_state; struct intel_power_domain_mask; struct intel_remapped_info; struct intel_rotation_info; struct pci_dev; +struct work_struct; #define pipe_name(p) ((p) + 'A') @@ -105,7 +105,7 @@ enum i9xx_plane_id { }; #define plane_name(p) ((p) + 'A') -#define sprite_name(p, s) ((p) * RUNTIME_INFO(dev_priv)->num_sprites[(p)] + (s) + 'A') +#define sprite_name(p, s) ((p) * DISPLAY_RUNTIME_INFO(dev_priv)->num_sprites[(p)] + (s) + 'A') #define for_each_plane_id_on_crtc(__crtc, __p) \ for ((__p) = PLANE_PRIMARY; (__p) < I915_MAX_PLANES; (__p)++) \ @@ -113,7 +113,7 @@ enum i9xx_plane_id { #define for_each_dbuf_slice(__dev_priv, __slice) \ for ((__slice) = DBUF_S1; (__slice) < I915_MAX_DBUF_SLICES; (__slice)++) \ - for_each_if(INTEL_INFO(__dev_priv)->display.dbuf.slice_mask & BIT(__slice)) + for_each_if(INTEL_INFO(__dev_priv)->display->dbuf.slice_mask & BIT(__slice)) #define for_each_dbuf_slice_in_mask(__dev_priv, __slice, __mask) \ for_each_dbuf_slice((__dev_priv), (__slice)) \ @@ -221,7 +221,7 @@ enum phy_fia { #define for_each_pipe(__dev_priv, __p) \ for ((__p) = 0; (__p) < I915_MAX_PIPES; (__p)++) \ - for_each_if(RUNTIME_INFO(__dev_priv)->pipe_mask & BIT(__p)) + for_each_if(DISPLAY_RUNTIME_INFO(__dev_priv)->pipe_mask & BIT(__p)) #define for_each_pipe_masked(__dev_priv, __p, __mask) \ for_each_pipe(__dev_priv, __p) \ @@ -229,7 +229,7 @@ enum phy_fia { #define for_each_cpu_transcoder(__dev_priv, __t) \ for ((__t) = 0; (__t) < I915_MAX_TRANSCODERS; (__t)++) \ - for_each_if (RUNTIME_INFO(__dev_priv)->cpu_transcoder_mask & BIT(__t)) + for_each_if (DISPLAY_RUNTIME_INFO(__dev_priv)->cpu_transcoder_mask & BIT(__t)) #define for_each_cpu_transcoder_masked(__dev_priv, __t, __mask) \ for_each_cpu_transcoder(__dev_priv, __t) \ @@ -237,7 +237,7 @@ enum phy_fia { #define for_each_sprite(__dev_priv, __p, __s) \ for ((__s) = 0; \ - (__s) < RUNTIME_INFO(__dev_priv)->num_sprites[(__p)]; \ + (__s) < DISPLAY_RUNTIME_INFO(__dev_priv)->num_sprites[(__p)]; \ (__s)++) #define for_each_port(__port) \ @@ -407,6 +407,7 @@ intel_mode_valid_max_plane_size(struct drm_i915_private *dev_priv, bool bigjoiner); enum phy intel_port_to_phy(struct drm_i915_private *i915, enum port port); bool is_trans_port_sync_mode(const struct intel_crtc_state *state); +bool is_trans_port_sync_master(const struct intel_crtc_state *state); bool intel_crtc_is_bigjoiner_slave(const struct intel_crtc_state *crtc_state); bool intel_crtc_is_bigjoiner_master(const struct intel_crtc_state *crtc_state); u8 intel_crtc_bigjoiner_slave_pipes(const struct intel_crtc_state *crtc_state); @@ -437,7 +438,6 @@ void intel_add_fb_offsets(int *x, int *y, unsigned int intel_rotation_info_size(const struct intel_rotation_info *rot_info); unsigned int intel_remapped_info_size(const struct intel_remapped_info *rem_info); bool intel_has_pending_fb_unpin(struct drm_i915_private *dev_priv); -int intel_display_suspend(struct drm_device *dev); void intel_encoder_destroy(struct drm_encoder *encoder); struct drm_display_mode * intel_encoder_current_mode(struct intel_encoder *encoder); @@ -455,20 +455,12 @@ int ilk_get_lanes_required(int target_clock, int link_bw, int bpp); void vlv_wait_port_ready(struct drm_i915_private *dev_priv, struct intel_digital_port *dig_port, unsigned int expected_mask); -int intel_get_load_detect_pipe(struct drm_connector *connector, - struct intel_load_detect_pipe *old, - struct drm_modeset_acquire_ctx *ctx); -void intel_release_load_detect_pipe(struct drm_connector *connector, - struct intel_load_detect_pipe *old, - struct drm_modeset_acquire_ctx *ctx); struct drm_framebuffer * intel_framebuffer_create(struct drm_i915_gem_object *obj, struct drm_mode_fb_cmd2 *mode_cmd); bool intel_fuzzy_clock_check(int clock1, int clock2); -void intel_display_prepare_reset(struct drm_i915_private *dev_priv); -void intel_display_finish_reset(struct drm_i915_private *dev_priv); void intel_zero_m_n(struct intel_link_m_n *m_n); void intel_set_m_n(struct drm_i915_private *i915, const struct intel_link_m_n *m_n, @@ -518,21 +510,9 @@ void intel_set_plane_visible(struct intel_crtc_state *crtc_state, bool visible); void intel_plane_fixup_bitmasks(struct intel_crtc_state *crtc_state); -void intel_display_driver_register(struct drm_i915_private *i915); -void intel_display_driver_unregister(struct drm_i915_private *i915); - void intel_update_watermarks(struct drm_i915_private *i915); /* modesetting */ -bool intel_modeset_probe_defer(struct pci_dev *pdev); -void intel_modeset_init_hw(struct drm_i915_private *i915); -int intel_modeset_init_noirq(struct drm_i915_private *i915); -int intel_modeset_init_nogem(struct drm_i915_private *i915); -int intel_modeset_init(struct drm_i915_private *i915); -void intel_modeset_driver_remove(struct drm_i915_private *i915); -void intel_modeset_driver_remove_noirq(struct drm_i915_private *i915); -void intel_modeset_driver_remove_nogem(struct drm_i915_private *i915); -void intel_display_resume(struct drm_device *dev); int intel_modeset_all_pipes(struct intel_atomic_state *state, const char *reason); void intel_modeset_get_crtc_power_domains(struct intel_crtc_state *crtc_state, @@ -540,30 +520,41 @@ void intel_modeset_get_crtc_power_domains(struct intel_crtc_state *crtc_state, void intel_modeset_put_crtc_power_domains(struct intel_crtc *crtc, struct intel_power_domain_mask *domains); +/* interface for intel_display_driver.c */ +void intel_setup_outputs(struct drm_i915_private *i915); +int intel_initial_commit(struct drm_device *dev); +void intel_panel_sanitize_ssc(struct drm_i915_private *i915); +void intel_update_czclk(struct drm_i915_private *i915); +void intel_atomic_helper_free_state_worker(struct work_struct *work); +enum drm_mode_status intel_mode_valid(struct drm_device *dev, + const struct drm_display_mode *mode); +int intel_atomic_commit(struct drm_device *dev, struct drm_atomic_state *_state, + bool nonblock); + +void intel_hpd_poll_fini(struct drm_i915_private *i915); + /* modesetting asserts */ void assert_transcoder(struct drm_i915_private *dev_priv, enum transcoder cpu_transcoder, bool state); #define assert_transcoder_enabled(d, t) assert_transcoder(d, t, true) #define assert_transcoder_disabled(d, t) assert_transcoder(d, t, false) -/* Use I915_STATE_WARN(x) and I915_STATE_WARN_ON() (rather than WARN() and - * WARN_ON()) for hw state sanity checks to check for unexpected conditions - * which may not necessarily be a user visible problem. This will either - * WARN() or DRM_ERROR() depending on the verbose_checks moduleparam, to - * enable distros and users to tailor their preferred amount of i915 abrt - * spam. +/* + * Use I915_STATE_WARN(x) (rather than WARN() and WARN_ON()) for hw state sanity + * checks to check for unexpected conditions which may not necessarily be a user + * visible problem. This will either WARN() or DRM_ERROR() depending on the + * verbose_state_checks module param, to enable distros and users to tailor + * their preferred amount of i915 abrt spam. */ -#define I915_STATE_WARN(condition, format...) ({ \ +#define I915_STATE_WARN(__i915, condition, format...) ({ \ + struct drm_device *drm = &(__i915)->drm; \ int __ret_warn_on = !!(condition); \ if (unlikely(__ret_warn_on)) \ - if (!WARN(i915_modparams.verbose_state_checks, format)) \ - DRM_ERROR(format); \ + if (!drm_WARN(drm, i915_modparams.verbose_state_checks, format)) \ + drm_err(drm, format); \ unlikely(__ret_warn_on); \ }) -#define I915_STATE_WARN_ON(x) \ - I915_STATE_WARN((x), "%s", "WARN_ON(" __stringify(x) ")") - bool intel_scanout_needs_vtd_wa(struct drm_i915_private *i915); #endif diff --git a/drivers/gpu/drm/i915/display/intel_display_core.h b/drivers/gpu/drm/i915/display/intel_display_core.h index e36f88a39b86349413f0f4c9fc7115170c24d484..8d2243c71dd85f91948c3fb778da2bd2640957dc 100644 --- a/drivers/gpu/drm/i915/display/intel_display_core.h +++ b/drivers/gpu/drm/i915/display/intel_display_core.h @@ -30,7 +30,7 @@ struct drm_i915_private; struct drm_property; struct drm_property_blob; struct i915_audio_component; -struct i915_hdcp_comp_master; +struct i915_hdcp_arbiter; struct intel_atomic_state; struct intel_audio_funcs; struct intel_bios_encoder_data; @@ -314,6 +314,8 @@ struct intel_display { unsigned int deratedbw[I915_NUM_QGV_POINTS]; /* for each PSF GV point */ unsigned int psf_bw[I915_NUM_PSF_GV_POINTS]; + /* Peak BW for each QGV point */ + unsigned int peakbw[I915_NUM_QGV_POINTS]; u8 num_qgv_points; u8 num_psf_gv_points; u8 num_planes; @@ -343,6 +345,15 @@ struct intel_display { struct intel_global_obj obj; } dbuf; + struct { + wait_queue_head_t waitqueue; + + /* mutex to protect pmdemand programming sequence */ + struct mutex lock; + + struct intel_global_obj obj; + } pmdemand; + struct { /* * dkl.phy_lock protects against concurrent access of the @@ -395,7 +406,7 @@ struct intel_display { } gmbus; struct { - struct i915_hdcp_master *master; + struct i915_hdcp_arbiter *arbiter; bool comp_added; /* @@ -404,8 +415,8 @@ struct intel_display { * this is only populated post Meteorlake */ struct intel_hdcp_gsc_message *hdcp_message; - /* Mutex to protect the above hdcp component related values. */ - struct mutex comp_mutex; + /* Mutex to protect the above hdcp related values. */ + struct mutex hdcp_mutex; } hdcp; struct { diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c b/drivers/gpu/drm/i915/display/intel_display_debugfs.c index 45113ae107ba8b8df658c47fb00d9cd20dc75b9a..165e2c7e3126a20893dbd93e2da14bba3e1b15c1 100644 --- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c +++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c @@ -6,12 +6,14 @@ #include #include +#include #include #include "hsw_ips.h" #include "i915_debugfs.h" #include "i915_irq.h" #include "i915_reg.h" +#include "intel_crtc.h" #include "intel_de.h" #include "intel_crtc_state_dump.h" #include "intel_display_debugfs.h" @@ -30,7 +32,6 @@ #include "intel_panel.h" #include "intel_psr.h" #include "intel_psr_regs.h" -#include "intel_sprite.h" #include "intel_wm.h" static inline struct drm_i915_private *node_to_i915(struct drm_info_node *node) @@ -228,37 +229,34 @@ static void intel_hdcp_info(struct seq_file *m, seq_puts(m, "\n"); } -static void intel_dp_info(struct seq_file *m, - struct intel_connector *intel_connector) +static void intel_dp_info(struct seq_file *m, struct intel_connector *connector) { - struct intel_encoder *intel_encoder = intel_attached_encoder(intel_connector); + struct intel_encoder *intel_encoder = intel_attached_encoder(connector); struct intel_dp *intel_dp = enc_to_intel_dp(intel_encoder); - const struct drm_property_blob *edid = intel_connector->base.edid_blob_ptr; + const struct edid *edid = drm_edid_raw(connector->detect_edid); seq_printf(m, "\tDPCD rev: %x\n", intel_dp->dpcd[DP_DPCD_REV]); seq_printf(m, "\taudio support: %s\n", - str_yes_no(intel_dp->has_audio)); + str_yes_no(connector->base.display_info.has_audio)); drm_dp_downstream_debug(m, intel_dp->dpcd, intel_dp->downstream_ports, - edid ? edid->data : NULL, &intel_dp->aux); + edid, &intel_dp->aux); } static void intel_dp_mst_info(struct seq_file *m, - struct intel_connector *intel_connector) + struct intel_connector *connector) { - bool has_audio = intel_connector->port->has_audio; + bool has_audio = connector->base.display_info.has_audio; seq_printf(m, "\taudio support: %s\n", str_yes_no(has_audio)); } static void intel_hdmi_info(struct seq_file *m, - struct intel_connector *intel_connector) + struct intel_connector *connector) { - struct intel_encoder *intel_encoder = intel_attached_encoder(intel_connector); - struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(intel_encoder); + bool has_audio = connector->base.display_info.has_audio; - seq_printf(m, "\taudio support: %s\n", - str_yes_no(intel_hdmi->has_audio)); + seq_printf(m, "\taudio support: %s\n", str_yes_no(has_audio)); } static void intel_connector_info(struct seq_file *m, @@ -1094,6 +1092,7 @@ void intel_display_debugfs_register(struct drm_i915_private *i915) ARRAY_SIZE(intel_display_debugfs_list), minor->debugfs_root, minor); + intel_cdclk_debugfs_register(i915); intel_dmc_debugfs_register(i915); intel_fbc_debugfs_register(i915); intel_hpd_debugfs_register(i915); diff --git a/drivers/gpu/drm/i915/display/intel_display_device.c b/drivers/gpu/drm/i915/display/intel_display_device.c new file mode 100644 index 0000000000000000000000000000000000000000..3fd30e7f006297981534aa4b3d273f27e14e747b --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_display_device.c @@ -0,0 +1,917 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2023 Intel Corporation + */ + +#include +#include +#include + +#include "i915_drv.h" +#include "i915_reg.h" +#include "intel_de.h" +#include "intel_display.h" +#include "intel_display_device.h" +#include "intel_display_power.h" +#include "intel_display_reg_defs.h" +#include "intel_fbc.h" + +__diag_push(); +__diag_ignore_all("-Woverride-init", "Allow overriding inherited members"); + +static const struct intel_display_device_info no_display = {}; + +#define PIPE_A_OFFSET 0x70000 +#define PIPE_B_OFFSET 0x71000 +#define PIPE_C_OFFSET 0x72000 +#define PIPE_D_OFFSET 0x73000 +#define CHV_PIPE_C_OFFSET 0x74000 +/* + * There's actually no pipe EDP. Some pipe registers have + * simply shifted from the pipe to the transcoder, while + * keeping their original offset. Thus we need PIPE_EDP_OFFSET + * to access such registers in transcoder EDP. + */ +#define PIPE_EDP_OFFSET 0x7f000 + +/* ICL DSI 0 and 1 */ +#define PIPE_DSI0_OFFSET 0x7b000 +#define PIPE_DSI1_OFFSET 0x7b800 + +#define TRANSCODER_A_OFFSET 0x60000 +#define TRANSCODER_B_OFFSET 0x61000 +#define TRANSCODER_C_OFFSET 0x62000 +#define CHV_TRANSCODER_C_OFFSET 0x63000 +#define TRANSCODER_D_OFFSET 0x63000 +#define TRANSCODER_EDP_OFFSET 0x6f000 +#define TRANSCODER_DSI0_OFFSET 0x6b000 +#define TRANSCODER_DSI1_OFFSET 0x6b800 + +#define CURSOR_A_OFFSET 0x70080 +#define CURSOR_B_OFFSET 0x700c0 +#define CHV_CURSOR_C_OFFSET 0x700e0 +#define IVB_CURSOR_B_OFFSET 0x71080 +#define IVB_CURSOR_C_OFFSET 0x72080 +#define TGL_CURSOR_D_OFFSET 0x73080 + +#define I845_PIPE_OFFSETS \ + .pipe_offsets = { \ + [TRANSCODER_A] = PIPE_A_OFFSET, \ + }, \ + .trans_offsets = { \ + [TRANSCODER_A] = TRANSCODER_A_OFFSET, \ + } + +#define I9XX_PIPE_OFFSETS \ + .pipe_offsets = { \ + [TRANSCODER_A] = PIPE_A_OFFSET, \ + [TRANSCODER_B] = PIPE_B_OFFSET, \ + }, \ + .trans_offsets = { \ + [TRANSCODER_A] = TRANSCODER_A_OFFSET, \ + [TRANSCODER_B] = TRANSCODER_B_OFFSET, \ + } + +#define IVB_PIPE_OFFSETS \ + .pipe_offsets = { \ + [TRANSCODER_A] = PIPE_A_OFFSET, \ + [TRANSCODER_B] = PIPE_B_OFFSET, \ + [TRANSCODER_C] = PIPE_C_OFFSET, \ + }, \ + .trans_offsets = { \ + [TRANSCODER_A] = TRANSCODER_A_OFFSET, \ + [TRANSCODER_B] = TRANSCODER_B_OFFSET, \ + [TRANSCODER_C] = TRANSCODER_C_OFFSET, \ + } + +#define HSW_PIPE_OFFSETS \ + .pipe_offsets = { \ + [TRANSCODER_A] = PIPE_A_OFFSET, \ + [TRANSCODER_B] = PIPE_B_OFFSET, \ + [TRANSCODER_C] = PIPE_C_OFFSET, \ + [TRANSCODER_EDP] = PIPE_EDP_OFFSET, \ + }, \ + .trans_offsets = { \ + [TRANSCODER_A] = TRANSCODER_A_OFFSET, \ + [TRANSCODER_B] = TRANSCODER_B_OFFSET, \ + [TRANSCODER_C] = TRANSCODER_C_OFFSET, \ + [TRANSCODER_EDP] = TRANSCODER_EDP_OFFSET, \ + } + +#define CHV_PIPE_OFFSETS \ + .pipe_offsets = { \ + [TRANSCODER_A] = PIPE_A_OFFSET, \ + [TRANSCODER_B] = PIPE_B_OFFSET, \ + [TRANSCODER_C] = CHV_PIPE_C_OFFSET, \ + }, \ + .trans_offsets = { \ + [TRANSCODER_A] = TRANSCODER_A_OFFSET, \ + [TRANSCODER_B] = TRANSCODER_B_OFFSET, \ + [TRANSCODER_C] = CHV_TRANSCODER_C_OFFSET, \ + } + +#define I845_CURSOR_OFFSETS \ + .cursor_offsets = { \ + [PIPE_A] = CURSOR_A_OFFSET, \ + } + +#define I9XX_CURSOR_OFFSETS \ + .cursor_offsets = { \ + [PIPE_A] = CURSOR_A_OFFSET, \ + [PIPE_B] = CURSOR_B_OFFSET, \ + } + +#define CHV_CURSOR_OFFSETS \ + .cursor_offsets = { \ + [PIPE_A] = CURSOR_A_OFFSET, \ + [PIPE_B] = CURSOR_B_OFFSET, \ + [PIPE_C] = CHV_CURSOR_C_OFFSET, \ + } + +#define IVB_CURSOR_OFFSETS \ + .cursor_offsets = { \ + [PIPE_A] = CURSOR_A_OFFSET, \ + [PIPE_B] = IVB_CURSOR_B_OFFSET, \ + [PIPE_C] = IVB_CURSOR_C_OFFSET, \ + } + +#define TGL_CURSOR_OFFSETS \ + .cursor_offsets = { \ + [PIPE_A] = CURSOR_A_OFFSET, \ + [PIPE_B] = IVB_CURSOR_B_OFFSET, \ + [PIPE_C] = IVB_CURSOR_C_OFFSET, \ + [PIPE_D] = TGL_CURSOR_D_OFFSET, \ + } + +#define I845_COLORS \ + .color = { .gamma_lut_size = 256 } +#define I9XX_COLORS \ + .color = { .gamma_lut_size = 129, \ + .gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \ + } +#define ILK_COLORS \ + .color = { .gamma_lut_size = 1024 } +#define IVB_COLORS \ + .color = { .degamma_lut_size = 1024, .gamma_lut_size = 1024 } +#define CHV_COLORS \ + .color = { \ + .degamma_lut_size = 65, .gamma_lut_size = 257, \ + .degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \ + .gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \ + } +#define GLK_COLORS \ + .color = { \ + .degamma_lut_size = 33, .gamma_lut_size = 1024, \ + .degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING | \ + DRM_COLOR_LUT_EQUAL_CHANNELS, \ + } +#define ICL_COLORS \ + .color = { \ + .degamma_lut_size = 33, .gamma_lut_size = 262145, \ + .degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING | \ + DRM_COLOR_LUT_EQUAL_CHANNELS, \ + .gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \ + } + +#define I830_DISPLAY \ + .has_overlay = 1, \ + .cursor_needs_physical = 1, \ + .overlay_needs_physical = 1, \ + .has_gmch = 1, \ + I9XX_PIPE_OFFSETS, \ + I9XX_CURSOR_OFFSETS, \ + I9XX_COLORS, \ + \ + .__runtime_defaults.ip.ver = 2, \ + .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \ + .__runtime_defaults.cpu_transcoder_mask = \ + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) + +static const struct intel_display_device_info i830_display = { + I830_DISPLAY, +}; + +#define I845_DISPLAY \ + .has_overlay = 1, \ + .overlay_needs_physical = 1, \ + .has_gmch = 1, \ + I845_PIPE_OFFSETS, \ + I845_CURSOR_OFFSETS, \ + I845_COLORS, \ + \ + .__runtime_defaults.ip.ver = 2, \ + .__runtime_defaults.pipe_mask = BIT(PIPE_A), \ + .__runtime_defaults.cpu_transcoder_mask = BIT(TRANSCODER_A) + +static const struct intel_display_device_info i845_display = { + I845_DISPLAY, +}; + +static const struct intel_display_device_info i85x_display = { + I830_DISPLAY, + + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), +}; + +static const struct intel_display_device_info i865g_display = { + I845_DISPLAY, + + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), +}; + +#define GEN3_DISPLAY \ + .has_gmch = 1, \ + .has_overlay = 1, \ + I9XX_PIPE_OFFSETS, \ + I9XX_CURSOR_OFFSETS, \ + \ + .__runtime_defaults.ip.ver = 3, \ + .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \ + .__runtime_defaults.cpu_transcoder_mask = \ + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) + +static const struct intel_display_device_info i915g_display = { + GEN3_DISPLAY, + I845_COLORS, + .cursor_needs_physical = 1, + .overlay_needs_physical = 1, +}; + +static const struct intel_display_device_info i915gm_display = { + GEN3_DISPLAY, + I9XX_COLORS, + .cursor_needs_physical = 1, + .overlay_needs_physical = 1, + .supports_tv = 1, + + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), +}; + +static const struct intel_display_device_info i945g_display = { + GEN3_DISPLAY, + I845_COLORS, + .has_hotplug = 1, + .cursor_needs_physical = 1, + .overlay_needs_physical = 1, +}; + +static const struct intel_display_device_info i945gm_display = { + GEN3_DISPLAY, + I9XX_COLORS, + .has_hotplug = 1, + .cursor_needs_physical = 1, + .overlay_needs_physical = 1, + .supports_tv = 1, + + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), +}; + +static const struct intel_display_device_info g33_display = { + GEN3_DISPLAY, + I845_COLORS, + .has_hotplug = 1, +}; + +static const struct intel_display_device_info pnv_display = { + GEN3_DISPLAY, + I9XX_COLORS, + .has_hotplug = 1, +}; + +#define GEN4_DISPLAY \ + .has_hotplug = 1, \ + .has_gmch = 1, \ + I9XX_PIPE_OFFSETS, \ + I9XX_CURSOR_OFFSETS, \ + I9XX_COLORS, \ + \ + .__runtime_defaults.ip.ver = 4, \ + .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \ + .__runtime_defaults.cpu_transcoder_mask = \ + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) + +static const struct intel_display_device_info i965g_display = { + GEN4_DISPLAY, + .has_overlay = 1, +}; + +static const struct intel_display_device_info i965gm_display = { + GEN4_DISPLAY, + .has_overlay = 1, + .supports_tv = 1, + + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), +}; + +static const struct intel_display_device_info g45_display = { + GEN4_DISPLAY, +}; + +static const struct intel_display_device_info gm45_display = { + GEN4_DISPLAY, + .supports_tv = 1, + + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), +}; + +#define ILK_DISPLAY \ + .has_hotplug = 1, \ + I9XX_PIPE_OFFSETS, \ + I9XX_CURSOR_OFFSETS, \ + ILK_COLORS, \ + \ + .__runtime_defaults.ip.ver = 5, \ + .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \ + .__runtime_defaults.cpu_transcoder_mask = \ + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) + +static const struct intel_display_device_info ilk_d_display = { + ILK_DISPLAY, +}; + +static const struct intel_display_device_info ilk_m_display = { + ILK_DISPLAY, + + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), +}; + +static const struct intel_display_device_info snb_display = { + .has_hotplug = 1, + I9XX_PIPE_OFFSETS, + I9XX_CURSOR_OFFSETS, + ILK_COLORS, + + .__runtime_defaults.ip.ver = 6, + .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), + .__runtime_defaults.cpu_transcoder_mask = + BIT(TRANSCODER_A) | BIT(TRANSCODER_B), + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), +}; + +static const struct intel_display_device_info ivb_display = { + .has_hotplug = 1, + IVB_PIPE_OFFSETS, + IVB_CURSOR_OFFSETS, + IVB_COLORS, + + .__runtime_defaults.ip.ver = 7, + .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), + .__runtime_defaults.cpu_transcoder_mask = + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C), + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), +}; + +static const struct intel_display_device_info vlv_display = { + .has_gmch = 1, + .has_hotplug = 1, + .mmio_offset = VLV_DISPLAY_BASE, + I9XX_PIPE_OFFSETS, + I9XX_CURSOR_OFFSETS, + I9XX_COLORS, + + .__runtime_defaults.ip.ver = 7, + .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), + .__runtime_defaults.cpu_transcoder_mask = + BIT(TRANSCODER_A) | BIT(TRANSCODER_B), +}; + +static const struct intel_display_device_info hsw_display = { + .has_ddi = 1, + .has_dp_mst = 1, + .has_fpga_dbg = 1, + .has_hotplug = 1, + HSW_PIPE_OFFSETS, + IVB_CURSOR_OFFSETS, + IVB_COLORS, + + .__runtime_defaults.ip.ver = 7, + .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), + .__runtime_defaults.cpu_transcoder_mask = + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | + BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP), + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), +}; + +static const struct intel_display_device_info bdw_display = { + .has_ddi = 1, + .has_dp_mst = 1, + .has_fpga_dbg = 1, + .has_hotplug = 1, + HSW_PIPE_OFFSETS, + IVB_CURSOR_OFFSETS, + IVB_COLORS, + + .__runtime_defaults.ip.ver = 8, + .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), + .__runtime_defaults.cpu_transcoder_mask = + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | + BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP), + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), +}; + +static const struct intel_display_device_info chv_display = { + .has_hotplug = 1, + .has_gmch = 1, + .mmio_offset = VLV_DISPLAY_BASE, + CHV_PIPE_OFFSETS, + CHV_CURSOR_OFFSETS, + CHV_COLORS, + + .__runtime_defaults.ip.ver = 8, + .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), + .__runtime_defaults.cpu_transcoder_mask = + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C), +}; + +static const struct intel_display_device_info skl_display = { + .dbuf.size = 896 - 4, /* 4 blocks for bypass path allocation */ + .dbuf.slice_mask = BIT(DBUF_S1), + .has_ddi = 1, + .has_dp_mst = 1, + .has_fpga_dbg = 1, + .has_hotplug = 1, + .has_ipc = 1, + .has_psr = 1, + .has_psr_hw_tracking = 1, + HSW_PIPE_OFFSETS, + IVB_CURSOR_OFFSETS, + IVB_COLORS, + + .__runtime_defaults.ip.ver = 9, + .__runtime_defaults.has_dmc = 1, + .__runtime_defaults.has_hdcp = 1, + .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), + .__runtime_defaults.cpu_transcoder_mask = + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | + BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP), + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), +}; + +#define GEN9_LP_DISPLAY \ + .dbuf.slice_mask = BIT(DBUF_S1), \ + .has_dp_mst = 1, \ + .has_ddi = 1, \ + .has_fpga_dbg = 1, \ + .has_hotplug = 1, \ + .has_ipc = 1, \ + .has_psr = 1, \ + .has_psr_hw_tracking = 1, \ + HSW_PIPE_OFFSETS, \ + IVB_CURSOR_OFFSETS, \ + IVB_COLORS, \ + \ + .__runtime_defaults.has_dmc = 1, \ + .__runtime_defaults.has_hdcp = 1, \ + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), \ + .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), \ + .__runtime_defaults.cpu_transcoder_mask = \ + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \ + BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP) | \ + BIT(TRANSCODER_DSI_A) | BIT(TRANSCODER_DSI_C) + +static const struct intel_display_device_info bxt_display = { + GEN9_LP_DISPLAY, + .dbuf.size = 512 - 4, /* 4 blocks for bypass path allocation */ + + .__runtime_defaults.ip.ver = 9, +}; + +static const struct intel_display_device_info glk_display = { + GEN9_LP_DISPLAY, + .dbuf.size = 1024 - 4, /* 4 blocks for bypass path allocation */ + GLK_COLORS, + + .__runtime_defaults.ip.ver = 10, +}; + +static const struct intel_display_device_info gen11_display = { + .abox_mask = BIT(0), + .dbuf.size = 2048, + .dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2), + .has_ddi = 1, + .has_dp_mst = 1, + .has_fpga_dbg = 1, + .has_hotplug = 1, + .has_ipc = 1, + .has_psr = 1, + .has_psr_hw_tracking = 1, + .pipe_offsets = { + [TRANSCODER_A] = PIPE_A_OFFSET, + [TRANSCODER_B] = PIPE_B_OFFSET, + [TRANSCODER_C] = PIPE_C_OFFSET, + [TRANSCODER_EDP] = PIPE_EDP_OFFSET, + [TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET, + [TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET, + }, + .trans_offsets = { + [TRANSCODER_A] = TRANSCODER_A_OFFSET, + [TRANSCODER_B] = TRANSCODER_B_OFFSET, + [TRANSCODER_C] = TRANSCODER_C_OFFSET, + [TRANSCODER_EDP] = TRANSCODER_EDP_OFFSET, + [TRANSCODER_DSI_0] = TRANSCODER_DSI0_OFFSET, + [TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET, + }, + IVB_CURSOR_OFFSETS, + ICL_COLORS, + + .__runtime_defaults.ip.ver = 11, + .__runtime_defaults.has_dmc = 1, + .__runtime_defaults.has_dsc = 1, + .__runtime_defaults.has_hdcp = 1, + .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), + .__runtime_defaults.cpu_transcoder_mask = + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | + BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP) | + BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1), + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), +}; + +#define XE_D_DISPLAY \ + .abox_mask = GENMASK(2, 1), \ + .dbuf.size = 2048, \ + .dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2), \ + .has_ddi = 1, \ + .has_dp_mst = 1, \ + .has_dsb = 1, \ + .has_fpga_dbg = 1, \ + .has_hotplug = 1, \ + .has_ipc = 1, \ + .has_psr = 1, \ + .has_psr_hw_tracking = 1, \ + .pipe_offsets = { \ + [TRANSCODER_A] = PIPE_A_OFFSET, \ + [TRANSCODER_B] = PIPE_B_OFFSET, \ + [TRANSCODER_C] = PIPE_C_OFFSET, \ + [TRANSCODER_D] = PIPE_D_OFFSET, \ + [TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET, \ + [TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET, \ + }, \ + .trans_offsets = { \ + [TRANSCODER_A] = TRANSCODER_A_OFFSET, \ + [TRANSCODER_B] = TRANSCODER_B_OFFSET, \ + [TRANSCODER_C] = TRANSCODER_C_OFFSET, \ + [TRANSCODER_D] = TRANSCODER_D_OFFSET, \ + [TRANSCODER_DSI_0] = TRANSCODER_DSI0_OFFSET, \ + [TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET, \ + }, \ + TGL_CURSOR_OFFSETS, \ + ICL_COLORS, \ + \ + .__runtime_defaults.ip.ver = 12, \ + .__runtime_defaults.has_dmc = 1, \ + .__runtime_defaults.has_dsc = 1, \ + .__runtime_defaults.has_hdcp = 1, \ + .__runtime_defaults.pipe_mask = \ + BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D), \ + .__runtime_defaults.cpu_transcoder_mask = \ + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \ + BIT(TRANSCODER_C) | BIT(TRANSCODER_D) | \ + BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1), \ + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A) + +static const struct intel_display_device_info tgl_display = { + XE_D_DISPLAY, +}; + +static const struct intel_display_device_info rkl_display = { + XE_D_DISPLAY, + .abox_mask = BIT(0), + .has_hti = 1, + .has_psr_hw_tracking = 0, + + .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), + .__runtime_defaults.cpu_transcoder_mask = + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C), +}; + +static const struct intel_display_device_info adl_s_display = { + XE_D_DISPLAY, + .has_hti = 1, + .has_psr_hw_tracking = 0, +}; + +#define XE_LPD_FEATURES \ + .abox_mask = GENMASK(1, 0), \ + .color = { \ + .degamma_lut_size = 129, .gamma_lut_size = 1024, \ + .degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING | \ + DRM_COLOR_LUT_EQUAL_CHANNELS, \ + }, \ + .dbuf.size = 4096, \ + .dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2) | BIT(DBUF_S3) | \ + BIT(DBUF_S4), \ + .has_ddi = 1, \ + .has_dp_mst = 1, \ + .has_dsb = 1, \ + .has_fpga_dbg = 1, \ + .has_hotplug = 1, \ + .has_ipc = 1, \ + .has_psr = 1, \ + .pipe_offsets = { \ + [TRANSCODER_A] = PIPE_A_OFFSET, \ + [TRANSCODER_B] = PIPE_B_OFFSET, \ + [TRANSCODER_C] = PIPE_C_OFFSET, \ + [TRANSCODER_D] = PIPE_D_OFFSET, \ + [TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET, \ + [TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET, \ + }, \ + .trans_offsets = { \ + [TRANSCODER_A] = TRANSCODER_A_OFFSET, \ + [TRANSCODER_B] = TRANSCODER_B_OFFSET, \ + [TRANSCODER_C] = TRANSCODER_C_OFFSET, \ + [TRANSCODER_D] = TRANSCODER_D_OFFSET, \ + [TRANSCODER_DSI_0] = TRANSCODER_DSI0_OFFSET, \ + [TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET, \ + }, \ + TGL_CURSOR_OFFSETS, \ + \ + .__runtime_defaults.ip.ver = 13, \ + .__runtime_defaults.has_dmc = 1, \ + .__runtime_defaults.has_dsc = 1, \ + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), \ + .__runtime_defaults.has_hdcp = 1, \ + .__runtime_defaults.pipe_mask = \ + BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D) + +static const struct intel_display_device_info xe_lpd_display = { + XE_LPD_FEATURES, + .has_cdclk_crawl = 1, + .has_psr_hw_tracking = 0, + + .__runtime_defaults.cpu_transcoder_mask = + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | + BIT(TRANSCODER_C) | BIT(TRANSCODER_D) | + BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1), +}; + +static const struct intel_display_device_info xe_hpd_display = { + XE_LPD_FEATURES, + .has_cdclk_squash = 1, + + .__runtime_defaults.cpu_transcoder_mask = + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | + BIT(TRANSCODER_C) | BIT(TRANSCODER_D), +}; + +static const struct intel_display_device_info xe_lpdp_display = { + XE_LPD_FEATURES, + .has_cdclk_crawl = 1, + .has_cdclk_squash = 1, + + .__runtime_defaults.ip.ver = 14, + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A) | BIT(INTEL_FBC_B), + .__runtime_defaults.cpu_transcoder_mask = + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | + BIT(TRANSCODER_C) | BIT(TRANSCODER_D), +}; + +__diag_pop(); + +#undef INTEL_VGA_DEVICE +#undef INTEL_QUANTA_VGA_DEVICE +#define INTEL_VGA_DEVICE(id, info) { id, info } +#define INTEL_QUANTA_VGA_DEVICE(info) { 0x16a, info } + +static const struct { + u32 devid; + const struct intel_display_device_info *info; +} intel_display_ids[] = { + INTEL_I830_IDS(&i830_display), + INTEL_I845G_IDS(&i845_display), + INTEL_I85X_IDS(&i85x_display), + INTEL_I865G_IDS(&i865g_display), + INTEL_I915G_IDS(&i915g_display), + INTEL_I915GM_IDS(&i915gm_display), + INTEL_I945G_IDS(&i945g_display), + INTEL_I945GM_IDS(&i945gm_display), + INTEL_I965G_IDS(&i965g_display), + INTEL_G33_IDS(&g33_display), + INTEL_I965GM_IDS(&i965gm_display), + INTEL_GM45_IDS(&gm45_display), + INTEL_G45_IDS(&g45_display), + INTEL_PINEVIEW_G_IDS(&pnv_display), + INTEL_PINEVIEW_M_IDS(&pnv_display), + INTEL_IRONLAKE_D_IDS(&ilk_d_display), + INTEL_IRONLAKE_M_IDS(&ilk_m_display), + INTEL_SNB_D_IDS(&snb_display), + INTEL_SNB_M_IDS(&snb_display), + INTEL_IVB_Q_IDS(NULL), /* must be first IVB in list */ + INTEL_IVB_M_IDS(&ivb_display), + INTEL_IVB_D_IDS(&ivb_display), + INTEL_HSW_IDS(&hsw_display), + INTEL_VLV_IDS(&vlv_display), + INTEL_BDW_IDS(&bdw_display), + INTEL_CHV_IDS(&chv_display), + INTEL_SKL_IDS(&skl_display), + INTEL_BXT_IDS(&bxt_display), + INTEL_GLK_IDS(&glk_display), + INTEL_KBL_IDS(&skl_display), + INTEL_CFL_IDS(&skl_display), + INTEL_ICL_11_IDS(&gen11_display), + INTEL_EHL_IDS(&gen11_display), + INTEL_JSL_IDS(&gen11_display), + INTEL_TGL_12_IDS(&tgl_display), + INTEL_DG1_IDS(&tgl_display), + INTEL_RKL_IDS(&rkl_display), + INTEL_ADLS_IDS(&adl_s_display), + INTEL_RPLS_IDS(&adl_s_display), + INTEL_ADLP_IDS(&xe_lpd_display), + INTEL_ADLN_IDS(&xe_lpd_display), + INTEL_RPLP_IDS(&xe_lpd_display), + INTEL_DG2_IDS(&xe_hpd_display), + + /* + * Do not add any GMD_ID-based platforms to this list. They will + * be probed automatically based on the IP version reported by + * the hardware. + */ +}; + +static const struct { + u16 ver; + u16 rel; + const struct intel_display_device_info *display; +} gmdid_display_map[] = { + { 14, 0, &xe_lpdp_display }, +}; + +static const struct intel_display_device_info * +probe_gmdid_display(struct drm_i915_private *i915, u16 *ver, u16 *rel, u16 *step) +{ + struct pci_dev *pdev = to_pci_dev(i915->drm.dev); + void __iomem *addr; + u32 val; + int i; + + addr = pci_iomap_range(pdev, 0, i915_mmio_reg_offset(GMD_ID_DISPLAY), sizeof(u32)); + if (!addr) { + drm_err(&i915->drm, "Cannot map MMIO BAR to read display GMD_ID\n"); + return &no_display; + } + + val = ioread32(addr); + pci_iounmap(pdev, addr); + + if (val == 0) + /* Platform doesn't have display */ + return &no_display; + + *ver = REG_FIELD_GET(GMD_ID_ARCH_MASK, val); + *rel = REG_FIELD_GET(GMD_ID_RELEASE_MASK, val); + *step = REG_FIELD_GET(GMD_ID_STEP, val); + + for (i = 0; i < ARRAY_SIZE(gmdid_display_map); i++) + if (*ver == gmdid_display_map[i].ver && + *rel == gmdid_display_map[i].rel) + return gmdid_display_map[i].display; + + drm_err(&i915->drm, "Unrecognized display IP version %d.%02d; disabling display.\n", + *ver, *rel); + return &no_display; +} + +const struct intel_display_device_info * +intel_display_device_probe(struct drm_i915_private *i915, bool has_gmdid, + u16 *gmdid_ver, u16 *gmdid_rel, u16 *gmdid_step) +{ + struct pci_dev *pdev = to_pci_dev(i915->drm.dev); + int i; + + if (has_gmdid) + return probe_gmdid_display(i915, gmdid_ver, gmdid_rel, gmdid_step); + + for (i = 0; i < ARRAY_SIZE(intel_display_ids); i++) { + if (intel_display_ids[i].devid == pdev->device) + return intel_display_ids[i].info; + } + + drm_dbg(&i915->drm, "No display ID found for device ID %04x; disabling display.\n", + pdev->device); + + return &no_display; +} + +void intel_display_device_info_runtime_init(struct drm_i915_private *i915) +{ + struct intel_display_runtime_info *display_runtime = DISPLAY_RUNTIME_INFO(i915); + enum pipe pipe; + + /* Wa_14011765242: adl-s A0,A1 */ + if (IS_ADLS_DISPLAY_STEP(i915, STEP_A0, STEP_A2)) + for_each_pipe(i915, pipe) + display_runtime->num_scalers[pipe] = 0; + else if (DISPLAY_VER(i915) >= 11) { + for_each_pipe(i915, pipe) + display_runtime->num_scalers[pipe] = 2; + } else if (DISPLAY_VER(i915) >= 9) { + display_runtime->num_scalers[PIPE_A] = 2; + display_runtime->num_scalers[PIPE_B] = 2; + display_runtime->num_scalers[PIPE_C] = 1; + } + + if (DISPLAY_VER(i915) >= 13 || HAS_D12_PLANE_MINIMIZATION(i915)) + for_each_pipe(i915, pipe) + display_runtime->num_sprites[pipe] = 4; + else if (DISPLAY_VER(i915) >= 11) + for_each_pipe(i915, pipe) + display_runtime->num_sprites[pipe] = 6; + else if (DISPLAY_VER(i915) == 10) + for_each_pipe(i915, pipe) + display_runtime->num_sprites[pipe] = 3; + else if (IS_BROXTON(i915)) { + /* + * Skylake and Broxton currently don't expose the topmost plane as its + * use is exclusive with the legacy cursor and we only want to expose + * one of those, not both. Until we can safely expose the topmost plane + * as a DRM_PLANE_TYPE_CURSOR with all the features exposed/supported, + * we don't expose the topmost plane at all to prevent ABI breakage + * down the line. + */ + + display_runtime->num_sprites[PIPE_A] = 2; + display_runtime->num_sprites[PIPE_B] = 2; + display_runtime->num_sprites[PIPE_C] = 1; + } else if (IS_VALLEYVIEW(i915) || IS_CHERRYVIEW(i915)) { + for_each_pipe(i915, pipe) + display_runtime->num_sprites[pipe] = 2; + } else if (DISPLAY_VER(i915) >= 5 || IS_G4X(i915)) { + for_each_pipe(i915, pipe) + display_runtime->num_sprites[pipe] = 1; + } + + if ((IS_DGFX(i915) || DISPLAY_VER(i915) >= 14) && + !(intel_de_read(i915, GU_CNTL_PROTECTED) & DEPRESENT)) { + drm_info(&i915->drm, "Display not present, disabling\n"); + goto display_fused_off; + } + + if (IS_GRAPHICS_VER(i915, 7, 8) && HAS_PCH_SPLIT(i915)) { + u32 fuse_strap = intel_de_read(i915, FUSE_STRAP); + u32 sfuse_strap = intel_de_read(i915, SFUSE_STRAP); + + /* + * SFUSE_STRAP is supposed to have a bit signalling the display + * is fused off. Unfortunately it seems that, at least in + * certain cases, fused off display means that PCH display + * reads don't land anywhere. In that case, we read 0s. + * + * On CPT/PPT, we can detect this case as SFUSE_STRAP_FUSE_LOCK + * should be set when taking over after the firmware. + */ + if (fuse_strap & ILK_INTERNAL_DISPLAY_DISABLE || + sfuse_strap & SFUSE_STRAP_DISPLAY_DISABLED || + (HAS_PCH_CPT(i915) && + !(sfuse_strap & SFUSE_STRAP_FUSE_LOCK))) { + drm_info(&i915->drm, + "Display fused off, disabling\n"); + goto display_fused_off; + } else if (fuse_strap & IVB_PIPE_C_DISABLE) { + drm_info(&i915->drm, "PipeC fused off\n"); + display_runtime->pipe_mask &= ~BIT(PIPE_C); + display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_C); + } + } else if (DISPLAY_VER(i915) >= 9) { + u32 dfsm = intel_de_read(i915, SKL_DFSM); + + if (dfsm & SKL_DFSM_PIPE_A_DISABLE) { + display_runtime->pipe_mask &= ~BIT(PIPE_A); + display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_A); + display_runtime->fbc_mask &= ~BIT(INTEL_FBC_A); + } + if (dfsm & SKL_DFSM_PIPE_B_DISABLE) { + display_runtime->pipe_mask &= ~BIT(PIPE_B); + display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_B); + } + if (dfsm & SKL_DFSM_PIPE_C_DISABLE) { + display_runtime->pipe_mask &= ~BIT(PIPE_C); + display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_C); + } + + if (DISPLAY_VER(i915) >= 12 && + (dfsm & TGL_DFSM_PIPE_D_DISABLE)) { + display_runtime->pipe_mask &= ~BIT(PIPE_D); + display_runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_D); + } + + if (!display_runtime->pipe_mask) + goto display_fused_off; + + if (dfsm & SKL_DFSM_DISPLAY_HDCP_DISABLE) + display_runtime->has_hdcp = 0; + + if (dfsm & SKL_DFSM_DISPLAY_PM_DISABLE) + display_runtime->fbc_mask = 0; + + if (DISPLAY_VER(i915) >= 11 && (dfsm & ICL_DFSM_DMC_DISABLE)) + display_runtime->has_dmc = 0; + + if (IS_DISPLAY_VER(i915, 10, 12) && + (dfsm & GLK_DFSM_DISPLAY_DSC_DISABLE)) + display_runtime->has_dsc = 0; + } + + return; + +display_fused_off: + memset(display_runtime, 0, sizeof(*display_runtime)); +} diff --git a/drivers/gpu/drm/i915/display/intel_display_device.h b/drivers/gpu/drm/i915/display/intel_display_device.h new file mode 100644 index 0000000000000000000000000000000000000000..706ff2aa1f55edd966c5a3086a0ef0a65595ba98 --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_display_device.h @@ -0,0 +1,129 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2023 Intel Corporation + */ + +#ifndef __INTEL_DISPLAY_DEVICE_H__ +#define __INTEL_DISPLAY_DEVICE_H__ + +#include + +#include "intel_display_limits.h" + +struct drm_i915_private; + +#define DEV_INFO_DISPLAY_FOR_EACH_FLAG(func) \ + /* Keep in alphabetical order */ \ + func(cursor_needs_physical); \ + func(has_cdclk_crawl); \ + func(has_cdclk_squash); \ + func(has_ddi); \ + func(has_dp_mst); \ + func(has_dsb); \ + func(has_fpga_dbg); \ + func(has_gmch); \ + func(has_hotplug); \ + func(has_hti); \ + func(has_ipc); \ + func(has_overlay); \ + func(has_psr); \ + func(has_psr_hw_tracking); \ + func(overlay_needs_physical); \ + func(supports_tv); + +#define HAS_ASYNC_FLIPS(i915) (DISPLAY_VER(i915) >= 5) +#define HAS_CDCLK_CRAWL(i915) (DISPLAY_INFO(i915)->has_cdclk_crawl) +#define HAS_CDCLK_SQUASH(i915) (DISPLAY_INFO(i915)->has_cdclk_squash) +#define HAS_CUR_FBC(i915) (!HAS_GMCH(i915) && DISPLAY_VER(i915) >= 7) +#define HAS_D12_PLANE_MINIMIZATION(i915) (IS_ROCKETLAKE(i915) || IS_ALDERLAKE_S(i915)) +#define HAS_DDI(i915) (DISPLAY_INFO(i915)->has_ddi) +#define HAS_DISPLAY(i915) (DISPLAY_RUNTIME_INFO(i915)->pipe_mask != 0) +#define HAS_DMC(i915) (DISPLAY_RUNTIME_INFO(i915)->has_dmc) +#define HAS_DOUBLE_BUFFERED_M_N(i915) (DISPLAY_VER(i915) >= 9 || IS_BROADWELL(i915)) +#define HAS_DP_MST(i915) (DISPLAY_INFO(i915)->has_dp_mst) +#define HAS_DP20(i915) (IS_DG2(i915) || DISPLAY_VER(i915) >= 14) +#define HAS_DPT(i915) (DISPLAY_VER(i915) >= 13) +#define HAS_DSB(i915) (DISPLAY_INFO(i915)->has_dsb) +#define HAS_DSC(__i915) (DISPLAY_RUNTIME_INFO(__i915)->has_dsc) +#define HAS_FBC(i915) (DISPLAY_RUNTIME_INFO(i915)->fbc_mask != 0) +#define HAS_FPGA_DBG_UNCLAIMED(i915) (DISPLAY_INFO(i915)->has_fpga_dbg) +#define HAS_FW_BLC(i915) (DISPLAY_VER(i915) > 2) +#define HAS_GMBUS_IRQ(i915) (DISPLAY_VER(i915) >= 4) +#define HAS_GMBUS_BURST_READ(i915) (DISPLAY_VER(i915) >= 10 || IS_KABYLAKE(i915)) +#define HAS_GMCH(i915) (DISPLAY_INFO(i915)->has_gmch) +#define HAS_HW_SAGV_WM(i915) (DISPLAY_VER(i915) >= 13 && !IS_DGFX(i915)) +#define HAS_IPC(i915) (DISPLAY_INFO(i915)->has_ipc) +#define HAS_IPS(i915) (IS_HSW_ULT(i915) || IS_BROADWELL(i915)) +#define HAS_LSPCON(i915) (IS_DISPLAY_VER(i915, 9, 10)) +#define HAS_MBUS_JOINING(i915) (IS_ALDERLAKE_P(i915) || DISPLAY_VER(i915) >= 14) +#define HAS_MSO(i915) (DISPLAY_VER(i915) >= 12) +#define HAS_OVERLAY(i915) (DISPLAY_INFO(i915)->has_overlay) +#define HAS_PSR(i915) (DISPLAY_INFO(i915)->has_psr) +#define HAS_PSR_HW_TRACKING(i915) (DISPLAY_INFO(i915)->has_psr_hw_tracking) +#define HAS_PSR2_SEL_FETCH(i915) (DISPLAY_VER(i915) >= 12) +#define HAS_SAGV(i915) (DISPLAY_VER(i915) >= 9 && !IS_LP(i915)) +#define HAS_TRANSCODER(i915, trans) ((DISPLAY_RUNTIME_INFO(i915)->cpu_transcoder_mask & \ + BIT(trans)) != 0) +#define HAS_VRR(i915) (DISPLAY_VER(i915) >= 11) +#define INTEL_NUM_PIPES(i915) (hweight8(DISPLAY_RUNTIME_INFO(i915)->pipe_mask)) +#define I915_HAS_HOTPLUG(i915) (DISPLAY_INFO(i915)->has_hotplug) +#define OVERLAY_NEEDS_PHYSICAL(i915) (DISPLAY_INFO(i915)->overlay_needs_physical) +#define SUPPORTS_TV(i915) (DISPLAY_INFO(i915)->supports_tv) + +struct intel_display_runtime_info { + struct { + u16 ver; + u16 rel; + u16 step; + } ip; + + u8 pipe_mask; + u8 cpu_transcoder_mask; + + u8 num_sprites[I915_MAX_PIPES]; + u8 num_scalers[I915_MAX_PIPES]; + + u8 fbc_mask; + + bool has_hdcp; + bool has_dmc; + bool has_dsc; +}; + +struct intel_display_device_info { + /* Initial runtime info. */ + const struct intel_display_runtime_info __runtime_defaults; + + u8 abox_mask; + + struct { + u16 size; /* in blocks */ + u8 slice_mask; + } dbuf; + +#define DEFINE_FLAG(name) u8 name:1 + DEV_INFO_DISPLAY_FOR_EACH_FLAG(DEFINE_FLAG); +#undef DEFINE_FLAG + + /* Global register offset for the display engine */ + u32 mmio_offset; + + /* Register offsets for the various display pipes and transcoders */ + u32 pipe_offsets[I915_MAX_TRANSCODERS]; + u32 trans_offsets[I915_MAX_TRANSCODERS]; + u32 cursor_offsets[I915_MAX_PIPES]; + + struct { + u32 degamma_lut_size; + u32 gamma_lut_size; + u32 degamma_lut_tests; + u32 gamma_lut_tests; + } color; +}; + +const struct intel_display_device_info * +intel_display_device_probe(struct drm_i915_private *i915, bool has_gmdid, + u16 *ver, u16 *rel, u16 *step); +void intel_display_device_info_runtime_init(struct drm_i915_private *i915); + +#endif diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c new file mode 100644 index 0000000000000000000000000000000000000000..b909814ae02b2717bef32016d6cc14747ceb7ed1 --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_display_driver.c @@ -0,0 +1,590 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2022-2023 Intel Corporation + * + * High level display driver entry points. This is a layer between top level + * driver code and low level display functionality; no low level display code or + * details here. + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#include "i915_drv.h" +#include "i9xx_wm.h" +#include "intel_acpi.h" +#include "intel_atomic.h" +#include "intel_audio.h" +#include "intel_bios.h" +#include "intel_bw.h" +#include "intel_cdclk.h" +#include "intel_color.h" +#include "intel_crtc.h" +#include "intel_display_debugfs.h" +#include "intel_display_driver.h" +#include "intel_display_power.h" +#include "intel_display_types.h" +#include "intel_dkl_phy.h" +#include "intel_dmc.h" +#include "intel_dp.h" +#include "intel_dpll.h" +#include "intel_dpll_mgr.h" +#include "intel_fb.h" +#include "intel_fbc.h" +#include "intel_fbdev.h" +#include "intel_fdi.h" +#include "intel_gmbus.h" +#include "intel_hdcp.h" +#include "intel_hotplug.h" +#include "intel_hti.h" +#include "intel_modeset_setup.h" +#include "intel_opregion.h" +#include "intel_overlay.h" +#include "intel_plane_initial.h" +#include "intel_pmdemand.h" +#include "intel_pps.h" +#include "intel_quirks.h" +#include "intel_vga.h" +#include "intel_wm.h" +#include "skl_watermark.h" + +bool intel_display_driver_probe_defer(struct pci_dev *pdev) +{ + struct drm_privacy_screen *privacy_screen; + + /* + * apple-gmux is needed on dual GPU MacBook Pro + * to probe the panel if we're the inactive GPU. + */ + if (vga_switcheroo_client_probe_defer(pdev)) + return true; + + /* If the LCD panel has a privacy-screen, wait for it */ + privacy_screen = drm_privacy_screen_get(&pdev->dev, NULL); + if (IS_ERR(privacy_screen) && PTR_ERR(privacy_screen) == -EPROBE_DEFER) + return true; + + drm_privacy_screen_put(privacy_screen); + + return false; +} + +void intel_display_driver_init_hw(struct drm_i915_private *i915) +{ + struct intel_cdclk_state *cdclk_state; + + if (!HAS_DISPLAY(i915)) + return; + + cdclk_state = to_intel_cdclk_state(i915->display.cdclk.obj.state); + + intel_update_cdclk(i915); + intel_cdclk_dump_config(i915, &i915->display.cdclk.hw, "Current CDCLK"); + cdclk_state->logical = cdclk_state->actual = i915->display.cdclk.hw; +} + +static const struct drm_mode_config_funcs intel_mode_funcs = { + .fb_create = intel_user_framebuffer_create, + .get_format_info = intel_fb_get_format_info, + .output_poll_changed = intel_fbdev_output_poll_changed, + .mode_valid = intel_mode_valid, + .atomic_check = intel_atomic_check, + .atomic_commit = intel_atomic_commit, + .atomic_state_alloc = intel_atomic_state_alloc, + .atomic_state_clear = intel_atomic_state_clear, + .atomic_state_free = intel_atomic_state_free, +}; + +static const struct drm_mode_config_helper_funcs intel_mode_config_funcs = { + .atomic_commit_setup = drm_dp_mst_atomic_setup_commit, +}; + +static void intel_mode_config_init(struct drm_i915_private *i915) +{ + struct drm_mode_config *mode_config = &i915->drm.mode_config; + + drm_mode_config_init(&i915->drm); + INIT_LIST_HEAD(&i915->display.global.obj_list); + + mode_config->min_width = 0; + mode_config->min_height = 0; + + mode_config->preferred_depth = 24; + mode_config->prefer_shadow = 1; + + mode_config->funcs = &intel_mode_funcs; + mode_config->helper_private = &intel_mode_config_funcs; + + mode_config->async_page_flip = HAS_ASYNC_FLIPS(i915); + + /* + * Maximum framebuffer dimensions, chosen to match + * the maximum render engine surface size on gen4+. + */ + if (DISPLAY_VER(i915) >= 7) { + mode_config->max_width = 16384; + mode_config->max_height = 16384; + } else if (DISPLAY_VER(i915) >= 4) { + mode_config->max_width = 8192; + mode_config->max_height = 8192; + } else if (DISPLAY_VER(i915) == 3) { + mode_config->max_width = 4096; + mode_config->max_height = 4096; + } else { + mode_config->max_width = 2048; + mode_config->max_height = 2048; + } + + if (IS_I845G(i915) || IS_I865G(i915)) { + mode_config->cursor_width = IS_I845G(i915) ? 64 : 512; + mode_config->cursor_height = 1023; + } else if (IS_I830(i915) || IS_I85X(i915) || + IS_I915G(i915) || IS_I915GM(i915)) { + mode_config->cursor_width = 64; + mode_config->cursor_height = 64; + } else { + mode_config->cursor_width = 256; + mode_config->cursor_height = 256; + } +} + +static void intel_mode_config_cleanup(struct drm_i915_private *i915) +{ + intel_atomic_global_obj_cleanup(i915); + drm_mode_config_cleanup(&i915->drm); +} + +static void intel_plane_possible_crtcs_init(struct drm_i915_private *dev_priv) +{ + struct intel_plane *plane; + + for_each_intel_plane(&dev_priv->drm, plane) { + struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv, + plane->pipe); + + plane->base.possible_crtcs = drm_crtc_mask(&crtc->base); + } +} + +void intel_display_driver_early_probe(struct drm_i915_private *i915) +{ + if (!HAS_DISPLAY(i915)) + return; + + intel_dkl_phy_init(i915); + intel_color_init_hooks(i915); + intel_init_cdclk_hooks(i915); + intel_audio_hooks_init(i915); + intel_dpll_init_clock_hook(i915); + intel_init_display_hooks(i915); + intel_fdi_init_hook(i915); +} + +/* part #1: call before irq install */ +int intel_display_driver_probe_noirq(struct drm_i915_private *i915) +{ + int ret; + + if (i915_inject_probe_failure(i915)) + return -ENODEV; + + if (HAS_DISPLAY(i915)) { + ret = drm_vblank_init(&i915->drm, + INTEL_NUM_PIPES(i915)); + if (ret) + return ret; + } + + intel_bios_init(i915); + + ret = intel_vga_register(i915); + if (ret) + goto cleanup_bios; + + /* FIXME: completely on the wrong abstraction layer */ + ret = intel_power_domains_init(i915); + if (ret < 0) + goto cleanup_vga; + + intel_pmdemand_init_early(i915); + + intel_power_domains_init_hw(i915, false); + + if (!HAS_DISPLAY(i915)) + return 0; + + intel_dmc_init(i915); + + i915->display.wq.modeset = alloc_ordered_workqueue("i915_modeset", 0); + i915->display.wq.flip = alloc_workqueue("i915_flip", WQ_HIGHPRI | + WQ_UNBOUND, WQ_UNBOUND_MAX_ACTIVE); + + intel_mode_config_init(i915); + + ret = intel_cdclk_init(i915); + if (ret) + goto cleanup_vga_client_pw_domain_dmc; + + ret = intel_color_init(i915); + if (ret) + goto cleanup_vga_client_pw_domain_dmc; + + ret = intel_dbuf_init(i915); + if (ret) + goto cleanup_vga_client_pw_domain_dmc; + + ret = intel_bw_init(i915); + if (ret) + goto cleanup_vga_client_pw_domain_dmc; + + ret = intel_pmdemand_init(i915); + if (ret) + goto cleanup_vga_client_pw_domain_dmc; + + init_llist_head(&i915->display.atomic_helper.free_list); + INIT_WORK(&i915->display.atomic_helper.free_work, + intel_atomic_helper_free_state_worker); + + intel_init_quirks(i915); + + intel_fbc_init(i915); + + return 0; + +cleanup_vga_client_pw_domain_dmc: + intel_dmc_fini(i915); + intel_power_domains_driver_remove(i915); +cleanup_vga: + intel_vga_unregister(i915); +cleanup_bios: + intel_bios_driver_remove(i915); + + return ret; +} + +/* part #2: call after irq install, but before gem init */ +int intel_display_driver_probe_nogem(struct drm_i915_private *i915) +{ + struct drm_device *dev = &i915->drm; + enum pipe pipe; + struct intel_crtc *crtc; + int ret; + + if (!HAS_DISPLAY(i915)) + return 0; + + intel_wm_init(i915); + + intel_panel_sanitize_ssc(i915); + + intel_pps_setup(i915); + + intel_gmbus_setup(i915); + + drm_dbg_kms(&i915->drm, "%d display pipe%s available.\n", + INTEL_NUM_PIPES(i915), + INTEL_NUM_PIPES(i915) > 1 ? "s" : ""); + + for_each_pipe(i915, pipe) { + ret = intel_crtc_init(i915, pipe); + if (ret) { + intel_mode_config_cleanup(i915); + return ret; + } + } + + intel_plane_possible_crtcs_init(i915); + intel_shared_dpll_init(i915); + intel_fdi_pll_freq_update(i915); + + intel_update_czclk(i915); + intel_display_driver_init_hw(i915); + intel_dpll_update_ref_clks(i915); + + intel_hdcp_component_init(i915); + + if (i915->display.cdclk.max_cdclk_freq == 0) + intel_update_max_cdclk(i915); + + intel_hti_init(i915); + + /* Just disable it once at startup */ + intel_vga_disable(i915); + intel_setup_outputs(i915); + + drm_modeset_lock_all(dev); + intel_modeset_setup_hw_state(i915, dev->mode_config.acquire_ctx); + intel_acpi_assign_connector_fwnodes(i915); + drm_modeset_unlock_all(dev); + + for_each_intel_crtc(dev, crtc) { + if (!to_intel_crtc_state(crtc->base.state)->uapi.active) + continue; + intel_crtc_initial_plane_config(crtc); + } + + /* + * Make sure hardware watermarks really match the state we read out. + * Note that we need to do this after reconstructing the BIOS fb's + * since the watermark calculation done here will use pstate->fb. + */ + if (!HAS_GMCH(i915)) + ilk_wm_sanitize(i915); + + return 0; +} + +/* part #3: call after gem init */ +int intel_display_driver_probe(struct drm_i915_private *i915) +{ + int ret; + + if (!HAS_DISPLAY(i915)) + return 0; + + /* + * Force all active planes to recompute their states. So that on + * mode_setcrtc after probe, all the intel_plane_state variables + * are already calculated and there is no assert_plane warnings + * during bootup. + */ + ret = intel_initial_commit(&i915->drm); + if (ret) + drm_dbg_kms(&i915->drm, "Initial modeset failed, %d\n", ret); + + intel_overlay_setup(i915); + + ret = intel_fbdev_init(&i915->drm); + if (ret) + return ret; + + /* Only enable hotplug handling once the fbdev is fully set up. */ + intel_hpd_init(i915); + intel_hpd_poll_disable(i915); + + skl_watermark_ipc_init(i915); + + return 0; +} + +void intel_display_driver_register(struct drm_i915_private *i915) +{ + if (!HAS_DISPLAY(i915)) + return; + + /* Must be done after probing outputs */ + intel_opregion_register(i915); + intel_acpi_video_register(i915); + + intel_audio_init(i915); + + intel_display_debugfs_register(i915); + + /* + * Some ports require correctly set-up hpd registers for + * detection to work properly (leading to ghost connected + * connector status), e.g. VGA on gm45. Hence we can only set + * up the initial fbdev config after hpd irqs are fully + * enabled. We do it last so that the async config cannot run + * before the connectors are registered. + */ + intel_fbdev_initial_config_async(i915); + + /* + * We need to coordinate the hotplugs with the asynchronous + * fbdev configuration, for which we use the + * fbdev->async_cookie. + */ + drm_kms_helper_poll_init(&i915->drm); +} + +/* part #1: call before irq uninstall */ +void intel_display_driver_remove(struct drm_i915_private *i915) +{ + if (!HAS_DISPLAY(i915)) + return; + + flush_workqueue(i915->display.wq.flip); + flush_workqueue(i915->display.wq.modeset); + + flush_work(&i915->display.atomic_helper.free_work); + drm_WARN_ON(&i915->drm, !llist_empty(&i915->display.atomic_helper.free_list)); + + /* + * MST topology needs to be suspended so we don't have any calls to + * fbdev after it's finalized. MST will be destroyed later as part of + * drm_mode_config_cleanup() + */ + intel_dp_mst_suspend(i915); +} + +/* part #2: call after irq uninstall */ +void intel_display_driver_remove_noirq(struct drm_i915_private *i915) +{ + if (!HAS_DISPLAY(i915)) + return; + + /* + * Due to the hpd irq storm handling the hotplug work can re-arm the + * poll handlers. Hence disable polling after hpd handling is shut down. + */ + intel_hpd_poll_fini(i915); + + /* poll work can call into fbdev, hence clean that up afterwards */ + intel_fbdev_fini(i915); + + intel_unregister_dsm_handler(); + + /* flush any delayed tasks or pending work */ + flush_workqueue(i915->unordered_wq); + + intel_hdcp_component_fini(i915); + + intel_mode_config_cleanup(i915); + + intel_overlay_cleanup(i915); + + intel_gmbus_teardown(i915); + + destroy_workqueue(i915->display.wq.flip); + destroy_workqueue(i915->display.wq.modeset); + + intel_fbc_cleanup(i915); +} + +/* part #3: call after gem init */ +void intel_display_driver_remove_nogem(struct drm_i915_private *i915) +{ + intel_dmc_fini(i915); + + intel_power_domains_driver_remove(i915); + + intel_vga_unregister(i915); + + intel_bios_driver_remove(i915); +} + +void intel_display_driver_unregister(struct drm_i915_private *i915) +{ + if (!HAS_DISPLAY(i915)) + return; + + intel_fbdev_unregister(i915); + intel_audio_deinit(i915); + + /* + * After flushing the fbdev (incl. a late async config which + * will have delayed queuing of a hotplug event), then flush + * the hotplug events. + */ + drm_kms_helper_poll_fini(&i915->drm); + drm_atomic_helper_shutdown(&i915->drm); + + acpi_video_unregister(); + intel_opregion_unregister(i915); +} + +/* + * turn all crtc's off, but do not adjust state + * This has to be paired with a call to intel_modeset_setup_hw_state. + */ +int intel_display_driver_suspend(struct drm_i915_private *i915) +{ + struct drm_atomic_state *state; + int ret; + + if (!HAS_DISPLAY(i915)) + return 0; + + state = drm_atomic_helper_suspend(&i915->drm); + ret = PTR_ERR_OR_ZERO(state); + if (ret) + drm_err(&i915->drm, "Suspending crtc's failed with %i\n", + ret); + else + i915->display.restore.modeset_state = state; + return ret; +} + +int +__intel_display_driver_resume(struct drm_i915_private *i915, + struct drm_atomic_state *state, + struct drm_modeset_acquire_ctx *ctx) +{ + struct drm_crtc_state *crtc_state; + struct drm_crtc *crtc; + int ret, i; + + intel_modeset_setup_hw_state(i915, ctx); + intel_vga_redisable(i915); + + if (!state) + return 0; + + /* + * We've duplicated the state, pointers to the old state are invalid. + * + * Don't attempt to use the old state until we commit the duplicated state. + */ + for_each_new_crtc_in_state(state, crtc, crtc_state, i) { + /* + * Force recalculation even if we restore + * current state. With fast modeset this may not result + * in a modeset when the state is compatible. + */ + crtc_state->mode_changed = true; + } + + /* ignore any reset values/BIOS leftovers in the WM registers */ + if (!HAS_GMCH(i915)) + to_intel_atomic_state(state)->skip_intermediate_wm = true; + + ret = drm_atomic_helper_commit_duplicated_state(state, ctx); + + drm_WARN_ON(&i915->drm, ret == -EDEADLK); + + return ret; +} + +void intel_display_driver_resume(struct drm_i915_private *i915) +{ + struct drm_atomic_state *state = i915->display.restore.modeset_state; + struct drm_modeset_acquire_ctx ctx; + int ret; + + if (!HAS_DISPLAY(i915)) + return; + + i915->display.restore.modeset_state = NULL; + if (state) + state->acquire_ctx = &ctx; + + drm_modeset_acquire_init(&ctx, 0); + + while (1) { + ret = drm_modeset_lock_all_ctx(&i915->drm, &ctx); + if (ret != -EDEADLK) + break; + + drm_modeset_backoff(&ctx); + } + + if (!ret) + ret = __intel_display_driver_resume(i915, state, &ctx); + + skl_watermark_ipc_update(i915); + drm_modeset_drop_locks(&ctx); + drm_modeset_acquire_fini(&ctx); + + if (ret) + drm_err(&i915->drm, + "Restoring old state failed with %i\n", ret); + if (state) + drm_atomic_state_put(state); +} diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.h b/drivers/gpu/drm/i915/display/intel_display_driver.h new file mode 100644 index 0000000000000000000000000000000000000000..c276a58ee3293f8f0e1761154a28d0ce46e046ba --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_display_driver.h @@ -0,0 +1,36 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2022-2023 Intel Corporation + */ + +#ifndef __INTEL_DISPLAY_DRIVER_H__ +#define __INTEL_DISPLAY_DRIVER_H__ + +#include + +struct drm_atomic_state; +struct drm_i915_private; +struct drm_modeset_acquire_ctx; +struct pci_dev; + +bool intel_display_driver_probe_defer(struct pci_dev *pdev); +void intel_display_driver_init_hw(struct drm_i915_private *i915); +void intel_display_driver_early_probe(struct drm_i915_private *i915); +int intel_display_driver_probe_noirq(struct drm_i915_private *i915); +int intel_display_driver_probe_nogem(struct drm_i915_private *i915); +int intel_display_driver_probe(struct drm_i915_private *i915); +void intel_display_driver_register(struct drm_i915_private *i915); +void intel_display_driver_remove(struct drm_i915_private *i915); +void intel_display_driver_remove_noirq(struct drm_i915_private *i915); +void intel_display_driver_remove_nogem(struct drm_i915_private *i915); +void intel_display_driver_unregister(struct drm_i915_private *i915); +int intel_display_driver_suspend(struct drm_i915_private *i915); +void intel_display_driver_resume(struct drm_i915_private *i915); + +/* interface for intel_display_reset.c */ +int __intel_display_driver_resume(struct drm_i915_private *i915, + struct drm_atomic_state *state, + struct drm_modeset_acquire_ctx *ctx); + +#endif /* __INTEL_DISPLAY_DRIVER_H__ */ + diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.c b/drivers/gpu/drm/i915/display/intel_display_irq.c new file mode 100644 index 0000000000000000000000000000000000000000..ae2578741dfe896458dd225a03107c2a60fb592c --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_display_irq.c @@ -0,0 +1,1687 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2023 Intel Corporation + */ + +#include "gt/intel_rps.h" +#include "i915_drv.h" +#include "i915_irq.h" +#include "i915_reg.h" +#include "icl_dsi_regs.h" +#include "intel_crtc.h" +#include "intel_de.h" +#include "intel_display_irq.h" +#include "intel_display_trace.h" +#include "intel_display_types.h" +#include "intel_dp_aux.h" +#include "intel_fdi_regs.h" +#include "intel_fifo_underrun.h" +#include "intel_gmbus.h" +#include "intel_hotplug_irq.h" +#include "intel_pmdemand.h" +#include "intel_psr.h" +#include "intel_psr_regs.h" + +static void +intel_handle_vblank(struct drm_i915_private *dev_priv, enum pipe pipe) +{ + struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv, pipe); + + drm_crtc_handle_vblank(&crtc->base); +} + +/** + * ilk_update_display_irq - update DEIMR + * @dev_priv: driver private + * @interrupt_mask: mask of interrupt bits to update + * @enabled_irq_mask: mask of interrupt bits to enable + */ +void ilk_update_display_irq(struct drm_i915_private *dev_priv, + u32 interrupt_mask, u32 enabled_irq_mask) +{ + u32 new_val; + + lockdep_assert_held(&dev_priv->irq_lock); + drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask); + + new_val = dev_priv->irq_mask; + new_val &= ~interrupt_mask; + new_val |= (~enabled_irq_mask & interrupt_mask); + + if (new_val != dev_priv->irq_mask && + !drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv))) { + dev_priv->irq_mask = new_val; + intel_uncore_write(&dev_priv->uncore, DEIMR, dev_priv->irq_mask); + intel_uncore_posting_read(&dev_priv->uncore, DEIMR); + } +} + +void ilk_enable_display_irq(struct drm_i915_private *i915, u32 bits) +{ + ilk_update_display_irq(i915, bits, bits); +} + +void ilk_disable_display_irq(struct drm_i915_private *i915, u32 bits) +{ + ilk_update_display_irq(i915, bits, 0); +} + +/** + * bdw_update_port_irq - update DE port interrupt + * @dev_priv: driver private + * @interrupt_mask: mask of interrupt bits to update + * @enabled_irq_mask: mask of interrupt bits to enable + */ +void bdw_update_port_irq(struct drm_i915_private *dev_priv, + u32 interrupt_mask, u32 enabled_irq_mask) +{ + u32 new_val; + u32 old_val; + + lockdep_assert_held(&dev_priv->irq_lock); + + drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask); + + if (drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv))) + return; + + old_val = intel_uncore_read(&dev_priv->uncore, GEN8_DE_PORT_IMR); + + new_val = old_val; + new_val &= ~interrupt_mask; + new_val |= (~enabled_irq_mask & interrupt_mask); + + if (new_val != old_val) { + intel_uncore_write(&dev_priv->uncore, GEN8_DE_PORT_IMR, new_val); + intel_uncore_posting_read(&dev_priv->uncore, GEN8_DE_PORT_IMR); + } +} + +/** + * bdw_update_pipe_irq - update DE pipe interrupt + * @dev_priv: driver private + * @pipe: pipe whose interrupt to update + * @interrupt_mask: mask of interrupt bits to update + * @enabled_irq_mask: mask of interrupt bits to enable + */ +static void bdw_update_pipe_irq(struct drm_i915_private *dev_priv, + enum pipe pipe, u32 interrupt_mask, + u32 enabled_irq_mask) +{ + u32 new_val; + + lockdep_assert_held(&dev_priv->irq_lock); + + drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask); + + if (drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv))) + return; + + new_val = dev_priv->de_irq_mask[pipe]; + new_val &= ~interrupt_mask; + new_val |= (~enabled_irq_mask & interrupt_mask); + + if (new_val != dev_priv->de_irq_mask[pipe]) { + dev_priv->de_irq_mask[pipe] = new_val; + intel_uncore_write(&dev_priv->uncore, GEN8_DE_PIPE_IMR(pipe), dev_priv->de_irq_mask[pipe]); + intel_uncore_posting_read(&dev_priv->uncore, GEN8_DE_PIPE_IMR(pipe)); + } +} + +void bdw_enable_pipe_irq(struct drm_i915_private *i915, + enum pipe pipe, u32 bits) +{ + bdw_update_pipe_irq(i915, pipe, bits, bits); +} + +void bdw_disable_pipe_irq(struct drm_i915_private *i915, + enum pipe pipe, u32 bits) +{ + bdw_update_pipe_irq(i915, pipe, bits, 0); +} + +/** + * ibx_display_interrupt_update - update SDEIMR + * @dev_priv: driver private + * @interrupt_mask: mask of interrupt bits to update + * @enabled_irq_mask: mask of interrupt bits to enable + */ +void ibx_display_interrupt_update(struct drm_i915_private *dev_priv, + u32 interrupt_mask, + u32 enabled_irq_mask) +{ + u32 sdeimr = intel_uncore_read(&dev_priv->uncore, SDEIMR); + + sdeimr &= ~interrupt_mask; + sdeimr |= (~enabled_irq_mask & interrupt_mask); + + drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask); + + lockdep_assert_held(&dev_priv->irq_lock); + + if (drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv))) + return; + + intel_uncore_write(&dev_priv->uncore, SDEIMR, sdeimr); + intel_uncore_posting_read(&dev_priv->uncore, SDEIMR); +} + +void ibx_enable_display_interrupt(struct drm_i915_private *i915, u32 bits) +{ + ibx_display_interrupt_update(i915, bits, bits); +} + +void ibx_disable_display_interrupt(struct drm_i915_private *i915, u32 bits) +{ + ibx_display_interrupt_update(i915, bits, 0); +} + +u32 i915_pipestat_enable_mask(struct drm_i915_private *dev_priv, + enum pipe pipe) +{ + u32 status_mask = dev_priv->pipestat_irq_mask[pipe]; + u32 enable_mask = status_mask << 16; + + lockdep_assert_held(&dev_priv->irq_lock); + + if (DISPLAY_VER(dev_priv) < 5) + goto out; + + /* + * On pipe A we don't support the PSR interrupt yet, + * on pipe B and C the same bit MBZ. + */ + if (drm_WARN_ON_ONCE(&dev_priv->drm, + status_mask & PIPE_A_PSR_STATUS_VLV)) + return 0; + /* + * On pipe B and C we don't support the PSR interrupt yet, on pipe + * A the same bit is for perf counters which we don't use either. + */ + if (drm_WARN_ON_ONCE(&dev_priv->drm, + status_mask & PIPE_B_PSR_STATUS_VLV)) + return 0; + + enable_mask &= ~(PIPE_FIFO_UNDERRUN_STATUS | + SPRITE0_FLIP_DONE_INT_EN_VLV | + SPRITE1_FLIP_DONE_INT_EN_VLV); + if (status_mask & SPRITE0_FLIP_DONE_INT_STATUS_VLV) + enable_mask |= SPRITE0_FLIP_DONE_INT_EN_VLV; + if (status_mask & SPRITE1_FLIP_DONE_INT_STATUS_VLV) + enable_mask |= SPRITE1_FLIP_DONE_INT_EN_VLV; + +out: + drm_WARN_ONCE(&dev_priv->drm, + enable_mask & ~PIPESTAT_INT_ENABLE_MASK || + status_mask & ~PIPESTAT_INT_STATUS_MASK, + "pipe %c: enable_mask=0x%x, status_mask=0x%x\n", + pipe_name(pipe), enable_mask, status_mask); + + return enable_mask; +} + +void i915_enable_pipestat(struct drm_i915_private *dev_priv, + enum pipe pipe, u32 status_mask) +{ + i915_reg_t reg = PIPESTAT(pipe); + u32 enable_mask; + + drm_WARN_ONCE(&dev_priv->drm, status_mask & ~PIPESTAT_INT_STATUS_MASK, + "pipe %c: status_mask=0x%x\n", + pipe_name(pipe), status_mask); + + lockdep_assert_held(&dev_priv->irq_lock); + drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv)); + + if ((dev_priv->pipestat_irq_mask[pipe] & status_mask) == status_mask) + return; + + dev_priv->pipestat_irq_mask[pipe] |= status_mask; + enable_mask = i915_pipestat_enable_mask(dev_priv, pipe); + + intel_uncore_write(&dev_priv->uncore, reg, enable_mask | status_mask); + intel_uncore_posting_read(&dev_priv->uncore, reg); +} + +void i915_disable_pipestat(struct drm_i915_private *dev_priv, + enum pipe pipe, u32 status_mask) +{ + i915_reg_t reg = PIPESTAT(pipe); + u32 enable_mask; + + drm_WARN_ONCE(&dev_priv->drm, status_mask & ~PIPESTAT_INT_STATUS_MASK, + "pipe %c: status_mask=0x%x\n", + pipe_name(pipe), status_mask); + + lockdep_assert_held(&dev_priv->irq_lock); + drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv)); + + if ((dev_priv->pipestat_irq_mask[pipe] & status_mask) == 0) + return; + + dev_priv->pipestat_irq_mask[pipe] &= ~status_mask; + enable_mask = i915_pipestat_enable_mask(dev_priv, pipe); + + intel_uncore_write(&dev_priv->uncore, reg, enable_mask | status_mask); + intel_uncore_posting_read(&dev_priv->uncore, reg); +} + +static bool i915_has_asle(struct drm_i915_private *dev_priv) +{ + if (!dev_priv->display.opregion.asle) + return false; + + return IS_PINEVIEW(dev_priv) || IS_MOBILE(dev_priv); +} + +/** + * i915_enable_asle_pipestat - enable ASLE pipestat for OpRegion + * @dev_priv: i915 device private + */ +void i915_enable_asle_pipestat(struct drm_i915_private *dev_priv) +{ + if (!i915_has_asle(dev_priv)) + return; + + spin_lock_irq(&dev_priv->irq_lock); + + i915_enable_pipestat(dev_priv, PIPE_B, PIPE_LEGACY_BLC_EVENT_STATUS); + if (DISPLAY_VER(dev_priv) >= 4) + i915_enable_pipestat(dev_priv, PIPE_A, + PIPE_LEGACY_BLC_EVENT_STATUS); + + spin_unlock_irq(&dev_priv->irq_lock); +} + +#if defined(CONFIG_DEBUG_FS) +static void display_pipe_crc_irq_handler(struct drm_i915_private *dev_priv, + enum pipe pipe, + u32 crc0, u32 crc1, + u32 crc2, u32 crc3, + u32 crc4) +{ + struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv, pipe); + struct intel_pipe_crc *pipe_crc = &crtc->pipe_crc; + u32 crcs[5] = { crc0, crc1, crc2, crc3, crc4 }; + + trace_intel_pipe_crc(crtc, crcs); + + spin_lock(&pipe_crc->lock); + /* + * For some not yet identified reason, the first CRC is + * bonkers. So let's just wait for the next vblank and read + * out the buggy result. + * + * On GEN8+ sometimes the second CRC is bonkers as well, so + * don't trust that one either. + */ + if (pipe_crc->skipped <= 0 || + (DISPLAY_VER(dev_priv) >= 8 && pipe_crc->skipped == 1)) { + pipe_crc->skipped++; + spin_unlock(&pipe_crc->lock); + return; + } + spin_unlock(&pipe_crc->lock); + + drm_crtc_add_crc_entry(&crtc->base, true, + drm_crtc_accurate_vblank_count(&crtc->base), + crcs); +} +#else +static inline void +display_pipe_crc_irq_handler(struct drm_i915_private *dev_priv, + enum pipe pipe, + u32 crc0, u32 crc1, + u32 crc2, u32 crc3, + u32 crc4) {} +#endif + +static void flip_done_handler(struct drm_i915_private *i915, + enum pipe pipe) +{ + struct intel_crtc *crtc = intel_crtc_for_pipe(i915, pipe); + struct drm_crtc_state *crtc_state = crtc->base.state; + struct drm_pending_vblank_event *e = crtc_state->event; + struct drm_device *dev = &i915->drm; + unsigned long irqflags; + + spin_lock_irqsave(&dev->event_lock, irqflags); + + crtc_state->event = NULL; + + drm_crtc_send_vblank_event(&crtc->base, e); + + spin_unlock_irqrestore(&dev->event_lock, irqflags); +} + +static void hsw_pipe_crc_irq_handler(struct drm_i915_private *dev_priv, + enum pipe pipe) +{ + display_pipe_crc_irq_handler(dev_priv, pipe, + intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_1_IVB(pipe)), + 0, 0, 0, 0); +} + +static void ivb_pipe_crc_irq_handler(struct drm_i915_private *dev_priv, + enum pipe pipe) +{ + display_pipe_crc_irq_handler(dev_priv, pipe, + intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_1_IVB(pipe)), + intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_2_IVB(pipe)), + intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_3_IVB(pipe)), + intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_4_IVB(pipe)), + intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_5_IVB(pipe))); +} + +static void i9xx_pipe_crc_irq_handler(struct drm_i915_private *dev_priv, + enum pipe pipe) +{ + u32 res1, res2; + + if (DISPLAY_VER(dev_priv) >= 3) + res1 = intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_RES1_I915(pipe)); + else + res1 = 0; + + if (DISPLAY_VER(dev_priv) >= 5 || IS_G4X(dev_priv)) + res2 = intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_RES2_G4X(pipe)); + else + res2 = 0; + + display_pipe_crc_irq_handler(dev_priv, pipe, + intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_RED(pipe)), + intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_GREEN(pipe)), + intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_BLUE(pipe)), + res1, res2); +} + +void i9xx_pipestat_irq_reset(struct drm_i915_private *dev_priv) +{ + enum pipe pipe; + + for_each_pipe(dev_priv, pipe) { + intel_uncore_write(&dev_priv->uncore, PIPESTAT(pipe), + PIPESTAT_INT_STATUS_MASK | + PIPE_FIFO_UNDERRUN_STATUS); + + dev_priv->pipestat_irq_mask[pipe] = 0; + } +} + +void i9xx_pipestat_irq_ack(struct drm_i915_private *dev_priv, + u32 iir, u32 pipe_stats[I915_MAX_PIPES]) +{ + enum pipe pipe; + + spin_lock(&dev_priv->irq_lock); + + if (!dev_priv->display_irqs_enabled) { + spin_unlock(&dev_priv->irq_lock); + return; + } + + for_each_pipe(dev_priv, pipe) { + i915_reg_t reg; + u32 status_mask, enable_mask, iir_bit = 0; + + /* + * PIPESTAT bits get signalled even when the interrupt is + * disabled with the mask bits, and some of the status bits do + * not generate interrupts at all (like the underrun bit). Hence + * we need to be careful that we only handle what we want to + * handle. + */ + + /* fifo underruns are filterered in the underrun handler. */ + status_mask = PIPE_FIFO_UNDERRUN_STATUS; + + switch (pipe) { + default: + case PIPE_A: + iir_bit = I915_DISPLAY_PIPE_A_EVENT_INTERRUPT; + break; + case PIPE_B: + iir_bit = I915_DISPLAY_PIPE_B_EVENT_INTERRUPT; + break; + case PIPE_C: + iir_bit = I915_DISPLAY_PIPE_C_EVENT_INTERRUPT; + break; + } + if (iir & iir_bit) + status_mask |= dev_priv->pipestat_irq_mask[pipe]; + + if (!status_mask) + continue; + + reg = PIPESTAT(pipe); + pipe_stats[pipe] = intel_uncore_read(&dev_priv->uncore, reg) & status_mask; + enable_mask = i915_pipestat_enable_mask(dev_priv, pipe); + + /* + * Clear the PIPE*STAT regs before the IIR + * + * Toggle the enable bits to make sure we get an + * edge in the ISR pipe event bit if we don't clear + * all the enabled status bits. Otherwise the edge + * triggered IIR on i965/g4x wouldn't notice that + * an interrupt is still pending. + */ + if (pipe_stats[pipe]) { + intel_uncore_write(&dev_priv->uncore, reg, pipe_stats[pipe]); + intel_uncore_write(&dev_priv->uncore, reg, enable_mask); + } + } + spin_unlock(&dev_priv->irq_lock); +} + +void i8xx_pipestat_irq_handler(struct drm_i915_private *dev_priv, + u16 iir, u32 pipe_stats[I915_MAX_PIPES]) +{ + enum pipe pipe; + + for_each_pipe(dev_priv, pipe) { + if (pipe_stats[pipe] & PIPE_VBLANK_INTERRUPT_STATUS) + intel_handle_vblank(dev_priv, pipe); + + if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS) + i9xx_pipe_crc_irq_handler(dev_priv, pipe); + + if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS) + intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe); + } +} + +void i915_pipestat_irq_handler(struct drm_i915_private *dev_priv, + u32 iir, u32 pipe_stats[I915_MAX_PIPES]) +{ + bool blc_event = false; + enum pipe pipe; + + for_each_pipe(dev_priv, pipe) { + if (pipe_stats[pipe] & PIPE_VBLANK_INTERRUPT_STATUS) + intel_handle_vblank(dev_priv, pipe); + + if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS) + blc_event = true; + + if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS) + i9xx_pipe_crc_irq_handler(dev_priv, pipe); + + if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS) + intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe); + } + + if (blc_event || (iir & I915_ASLE_INTERRUPT)) + intel_opregion_asle_intr(dev_priv); +} + +void i965_pipestat_irq_handler(struct drm_i915_private *dev_priv, + u32 iir, u32 pipe_stats[I915_MAX_PIPES]) +{ + bool blc_event = false; + enum pipe pipe; + + for_each_pipe(dev_priv, pipe) { + if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS) + intel_handle_vblank(dev_priv, pipe); + + if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS) + blc_event = true; + + if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS) + i9xx_pipe_crc_irq_handler(dev_priv, pipe); + + if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS) + intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe); + } + + if (blc_event || (iir & I915_ASLE_INTERRUPT)) + intel_opregion_asle_intr(dev_priv); + + if (pipe_stats[0] & PIPE_GMBUS_INTERRUPT_STATUS) + intel_gmbus_irq_handler(dev_priv); +} + +void valleyview_pipestat_irq_handler(struct drm_i915_private *dev_priv, + u32 pipe_stats[I915_MAX_PIPES]) +{ + enum pipe pipe; + + for_each_pipe(dev_priv, pipe) { + if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS) + intel_handle_vblank(dev_priv, pipe); + + if (pipe_stats[pipe] & PLANE_FLIP_DONE_INT_STATUS_VLV) + flip_done_handler(dev_priv, pipe); + + if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS) + i9xx_pipe_crc_irq_handler(dev_priv, pipe); + + if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS) + intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe); + } + + if (pipe_stats[0] & PIPE_GMBUS_INTERRUPT_STATUS) + intel_gmbus_irq_handler(dev_priv); +} + +static void ibx_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir) +{ + enum pipe pipe; + u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK; + + ibx_hpd_irq_handler(dev_priv, hotplug_trigger); + + if (pch_iir & SDE_AUDIO_POWER_MASK) { + int port = ffs((pch_iir & SDE_AUDIO_POWER_MASK) >> + SDE_AUDIO_POWER_SHIFT); + drm_dbg(&dev_priv->drm, "PCH audio power change on port %d\n", + port_name(port)); + } + + if (pch_iir & SDE_AUX_MASK) + intel_dp_aux_irq_handler(dev_priv); + + if (pch_iir & SDE_GMBUS) + intel_gmbus_irq_handler(dev_priv); + + if (pch_iir & SDE_AUDIO_HDCP_MASK) + drm_dbg(&dev_priv->drm, "PCH HDCP audio interrupt\n"); + + if (pch_iir & SDE_AUDIO_TRANS_MASK) + drm_dbg(&dev_priv->drm, "PCH transcoder audio interrupt\n"); + + if (pch_iir & SDE_POISON) + drm_err(&dev_priv->drm, "PCH poison interrupt\n"); + + if (pch_iir & SDE_FDI_MASK) { + for_each_pipe(dev_priv, pipe) + drm_dbg(&dev_priv->drm, " pipe %c FDI IIR: 0x%08x\n", + pipe_name(pipe), + intel_uncore_read(&dev_priv->uncore, FDI_RX_IIR(pipe))); + } + + if (pch_iir & (SDE_TRANSB_CRC_DONE | SDE_TRANSA_CRC_DONE)) + drm_dbg(&dev_priv->drm, "PCH transcoder CRC done interrupt\n"); + + if (pch_iir & (SDE_TRANSB_CRC_ERR | SDE_TRANSA_CRC_ERR)) + drm_dbg(&dev_priv->drm, + "PCH transcoder CRC error interrupt\n"); + + if (pch_iir & SDE_TRANSA_FIFO_UNDER) + intel_pch_fifo_underrun_irq_handler(dev_priv, PIPE_A); + + if (pch_iir & SDE_TRANSB_FIFO_UNDER) + intel_pch_fifo_underrun_irq_handler(dev_priv, PIPE_B); +} + +static void ivb_err_int_handler(struct drm_i915_private *dev_priv) +{ + u32 err_int = intel_uncore_read(&dev_priv->uncore, GEN7_ERR_INT); + enum pipe pipe; + + if (err_int & ERR_INT_POISON) + drm_err(&dev_priv->drm, "Poison interrupt\n"); + + for_each_pipe(dev_priv, pipe) { + if (err_int & ERR_INT_FIFO_UNDERRUN(pipe)) + intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe); + + if (err_int & ERR_INT_PIPE_CRC_DONE(pipe)) { + if (IS_IVYBRIDGE(dev_priv)) + ivb_pipe_crc_irq_handler(dev_priv, pipe); + else + hsw_pipe_crc_irq_handler(dev_priv, pipe); + } + } + + intel_uncore_write(&dev_priv->uncore, GEN7_ERR_INT, err_int); +} + +static void cpt_serr_int_handler(struct drm_i915_private *dev_priv) +{ + u32 serr_int = intel_uncore_read(&dev_priv->uncore, SERR_INT); + enum pipe pipe; + + if (serr_int & SERR_INT_POISON) + drm_err(&dev_priv->drm, "PCH poison interrupt\n"); + + for_each_pipe(dev_priv, pipe) + if (serr_int & SERR_INT_TRANS_FIFO_UNDERRUN(pipe)) + intel_pch_fifo_underrun_irq_handler(dev_priv, pipe); + + intel_uncore_write(&dev_priv->uncore, SERR_INT, serr_int); +} + +static void cpt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir) +{ + enum pipe pipe; + u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK_CPT; + + ibx_hpd_irq_handler(dev_priv, hotplug_trigger); + + if (pch_iir & SDE_AUDIO_POWER_MASK_CPT) { + int port = ffs((pch_iir & SDE_AUDIO_POWER_MASK_CPT) >> + SDE_AUDIO_POWER_SHIFT_CPT); + drm_dbg(&dev_priv->drm, "PCH audio power change on port %c\n", + port_name(port)); + } + + if (pch_iir & SDE_AUX_MASK_CPT) + intel_dp_aux_irq_handler(dev_priv); + + if (pch_iir & SDE_GMBUS_CPT) + intel_gmbus_irq_handler(dev_priv); + + if (pch_iir & SDE_AUDIO_CP_REQ_CPT) + drm_dbg(&dev_priv->drm, "Audio CP request interrupt\n"); + + if (pch_iir & SDE_AUDIO_CP_CHG_CPT) + drm_dbg(&dev_priv->drm, "Audio CP change interrupt\n"); + + if (pch_iir & SDE_FDI_MASK_CPT) { + for_each_pipe(dev_priv, pipe) + drm_dbg(&dev_priv->drm, " pipe %c FDI IIR: 0x%08x\n", + pipe_name(pipe), + intel_uncore_read(&dev_priv->uncore, FDI_RX_IIR(pipe))); + } + + if (pch_iir & SDE_ERROR_CPT) + cpt_serr_int_handler(dev_priv); +} + +void ilk_display_irq_handler(struct drm_i915_private *dev_priv, u32 de_iir) +{ + enum pipe pipe; + u32 hotplug_trigger = de_iir & DE_DP_A_HOTPLUG; + + if (hotplug_trigger) + ilk_hpd_irq_handler(dev_priv, hotplug_trigger); + + if (de_iir & DE_AUX_CHANNEL_A) + intel_dp_aux_irq_handler(dev_priv); + + if (de_iir & DE_GSE) + intel_opregion_asle_intr(dev_priv); + + if (de_iir & DE_POISON) + drm_err(&dev_priv->drm, "Poison interrupt\n"); + + for_each_pipe(dev_priv, pipe) { + if (de_iir & DE_PIPE_VBLANK(pipe)) + intel_handle_vblank(dev_priv, pipe); + + if (de_iir & DE_PLANE_FLIP_DONE(pipe)) + flip_done_handler(dev_priv, pipe); + + if (de_iir & DE_PIPE_FIFO_UNDERRUN(pipe)) + intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe); + + if (de_iir & DE_PIPE_CRC_DONE(pipe)) + i9xx_pipe_crc_irq_handler(dev_priv, pipe); + } + + /* check event from PCH */ + if (de_iir & DE_PCH_EVENT) { + u32 pch_iir = intel_uncore_read(&dev_priv->uncore, SDEIIR); + + if (HAS_PCH_CPT(dev_priv)) + cpt_irq_handler(dev_priv, pch_iir); + else + ibx_irq_handler(dev_priv, pch_iir); + + /* should clear PCH hotplug event before clear CPU irq */ + intel_uncore_write(&dev_priv->uncore, SDEIIR, pch_iir); + } + + if (DISPLAY_VER(dev_priv) == 5 && de_iir & DE_PCU_EVENT) + gen5_rps_irq_handler(&to_gt(dev_priv)->rps); +} + +void ivb_display_irq_handler(struct drm_i915_private *dev_priv, u32 de_iir) +{ + enum pipe pipe; + u32 hotplug_trigger = de_iir & DE_DP_A_HOTPLUG_IVB; + + if (hotplug_trigger) + ilk_hpd_irq_handler(dev_priv, hotplug_trigger); + + if (de_iir & DE_ERR_INT_IVB) + ivb_err_int_handler(dev_priv); + + if (de_iir & DE_AUX_CHANNEL_A_IVB) + intel_dp_aux_irq_handler(dev_priv); + + if (de_iir & DE_GSE_IVB) + intel_opregion_asle_intr(dev_priv); + + for_each_pipe(dev_priv, pipe) { + if (de_iir & DE_PIPE_VBLANK_IVB(pipe)) + intel_handle_vblank(dev_priv, pipe); + + if (de_iir & DE_PLANE_FLIP_DONE_IVB(pipe)) + flip_done_handler(dev_priv, pipe); + } + + /* check event from PCH */ + if (!HAS_PCH_NOP(dev_priv) && (de_iir & DE_PCH_EVENT_IVB)) { + u32 pch_iir = intel_uncore_read(&dev_priv->uncore, SDEIIR); + + cpt_irq_handler(dev_priv, pch_iir); + + /* clear PCH hotplug event before clear CPU irq */ + intel_uncore_write(&dev_priv->uncore, SDEIIR, pch_iir); + } +} + +static u32 gen8_de_port_aux_mask(struct drm_i915_private *dev_priv) +{ + u32 mask; + + if (DISPLAY_VER(dev_priv) >= 14) + return TGL_DE_PORT_AUX_DDIA | + TGL_DE_PORT_AUX_DDIB; + else if (DISPLAY_VER(dev_priv) >= 13) + return TGL_DE_PORT_AUX_DDIA | + TGL_DE_PORT_AUX_DDIB | + TGL_DE_PORT_AUX_DDIC | + XELPD_DE_PORT_AUX_DDID | + XELPD_DE_PORT_AUX_DDIE | + TGL_DE_PORT_AUX_USBC1 | + TGL_DE_PORT_AUX_USBC2 | + TGL_DE_PORT_AUX_USBC3 | + TGL_DE_PORT_AUX_USBC4; + else if (DISPLAY_VER(dev_priv) >= 12) + return TGL_DE_PORT_AUX_DDIA | + TGL_DE_PORT_AUX_DDIB | + TGL_DE_PORT_AUX_DDIC | + TGL_DE_PORT_AUX_USBC1 | + TGL_DE_PORT_AUX_USBC2 | + TGL_DE_PORT_AUX_USBC3 | + TGL_DE_PORT_AUX_USBC4 | + TGL_DE_PORT_AUX_USBC5 | + TGL_DE_PORT_AUX_USBC6; + + mask = GEN8_AUX_CHANNEL_A; + if (DISPLAY_VER(dev_priv) >= 9) + mask |= GEN9_AUX_CHANNEL_B | + GEN9_AUX_CHANNEL_C | + GEN9_AUX_CHANNEL_D; + + if (DISPLAY_VER(dev_priv) == 11) { + mask |= ICL_AUX_CHANNEL_F; + mask |= ICL_AUX_CHANNEL_E; + } + + return mask; +} + +static u32 gen8_de_pipe_fault_mask(struct drm_i915_private *dev_priv) +{ + if (DISPLAY_VER(dev_priv) >= 13 || HAS_D12_PLANE_MINIMIZATION(dev_priv)) + return RKL_DE_PIPE_IRQ_FAULT_ERRORS; + else if (DISPLAY_VER(dev_priv) >= 11) + return GEN11_DE_PIPE_IRQ_FAULT_ERRORS; + else if (DISPLAY_VER(dev_priv) >= 9) + return GEN9_DE_PIPE_IRQ_FAULT_ERRORS; + else + return GEN8_DE_PIPE_IRQ_FAULT_ERRORS; +} + +static void intel_pmdemand_irq_handler(struct drm_i915_private *dev_priv) +{ + wake_up_all(&dev_priv->display.pmdemand.waitqueue); +} + +static void +gen8_de_misc_irq_handler(struct drm_i915_private *dev_priv, u32 iir) +{ + bool found = false; + + if (DISPLAY_VER(dev_priv) >= 14) { + if (iir & (XELPDP_PMDEMAND_RSP | + XELPDP_PMDEMAND_RSPTOUT_ERR)) { + if (iir & XELPDP_PMDEMAND_RSPTOUT_ERR) + drm_dbg(&dev_priv->drm, + "Error waiting for Punit PM Demand Response\n"); + + intel_pmdemand_irq_handler(dev_priv); + found = true; + } + } else if (iir & GEN8_DE_MISC_GSE) { + intel_opregion_asle_intr(dev_priv); + found = true; + } + + if (iir & GEN8_DE_EDP_PSR) { + struct intel_encoder *encoder; + u32 psr_iir; + i915_reg_t iir_reg; + + for_each_intel_encoder_with_psr(&dev_priv->drm, encoder) { + struct intel_dp *intel_dp = enc_to_intel_dp(encoder); + + if (DISPLAY_VER(dev_priv) >= 12) + iir_reg = TRANS_PSR_IIR(intel_dp->psr.transcoder); + else + iir_reg = EDP_PSR_IIR; + + psr_iir = intel_uncore_rmw(&dev_priv->uncore, iir_reg, 0, 0); + + if (psr_iir) + found = true; + + intel_psr_irq_handler(intel_dp, psr_iir); + + /* prior GEN12 only have one EDP PSR */ + if (DISPLAY_VER(dev_priv) < 12) + break; + } + } + + if (!found) + drm_err(&dev_priv->drm, "Unexpected DE Misc interrupt\n"); +} + +static void gen11_dsi_te_interrupt_handler(struct drm_i915_private *dev_priv, + u32 te_trigger) +{ + enum pipe pipe = INVALID_PIPE; + enum transcoder dsi_trans; + enum port port; + u32 val; + + /* + * Incase of dual link, TE comes from DSI_1 + * this is to check if dual link is enabled + */ + val = intel_uncore_read(&dev_priv->uncore, TRANS_DDI_FUNC_CTL2(TRANSCODER_DSI_0)); + val &= PORT_SYNC_MODE_ENABLE; + + /* + * if dual link is enabled, then read DSI_0 + * transcoder registers + */ + port = ((te_trigger & DSI1_TE && val) || (te_trigger & DSI0_TE)) ? + PORT_A : PORT_B; + dsi_trans = (port == PORT_A) ? TRANSCODER_DSI_0 : TRANSCODER_DSI_1; + + /* Check if DSI configured in command mode */ + val = intel_uncore_read(&dev_priv->uncore, DSI_TRANS_FUNC_CONF(dsi_trans)); + val = val & OP_MODE_MASK; + + if (val != CMD_MODE_NO_GATE && val != CMD_MODE_TE_GATE) { + drm_err(&dev_priv->drm, "DSI trancoder not configured in command mode\n"); + return; + } + + /* Get PIPE for handling VBLANK event */ + val = intel_uncore_read(&dev_priv->uncore, TRANS_DDI_FUNC_CTL(dsi_trans)); + switch (val & TRANS_DDI_EDP_INPUT_MASK) { + case TRANS_DDI_EDP_INPUT_A_ON: + pipe = PIPE_A; + break; + case TRANS_DDI_EDP_INPUT_B_ONOFF: + pipe = PIPE_B; + break; + case TRANS_DDI_EDP_INPUT_C_ONOFF: + pipe = PIPE_C; + break; + default: + drm_err(&dev_priv->drm, "Invalid PIPE\n"); + return; + } + + intel_handle_vblank(dev_priv, pipe); + + /* clear TE in dsi IIR */ + port = (te_trigger & DSI1_TE) ? PORT_B : PORT_A; + intel_uncore_rmw(&dev_priv->uncore, DSI_INTR_IDENT_REG(port), 0, 0); +} + +static u32 gen8_de_pipe_flip_done_mask(struct drm_i915_private *i915) +{ + if (DISPLAY_VER(i915) >= 9) + return GEN9_PIPE_PLANE1_FLIP_DONE; + else + return GEN8_PIPE_PRIMARY_FLIP_DONE; +} + +u32 gen8_de_pipe_underrun_mask(struct drm_i915_private *dev_priv) +{ + u32 mask = GEN8_PIPE_FIFO_UNDERRUN; + + if (DISPLAY_VER(dev_priv) >= 13) + mask |= XELPD_PIPE_SOFT_UNDERRUN | + XELPD_PIPE_HARD_UNDERRUN; + + return mask; +} + +static void gen8_read_and_ack_pch_irqs(struct drm_i915_private *i915, u32 *pch_iir, u32 *pica_iir) +{ + u32 pica_ier = 0; + + *pica_iir = 0; + *pch_iir = intel_de_read(i915, SDEIIR); + if (!*pch_iir) + return; + + /** + * PICA IER must be disabled/re-enabled around clearing PICA IIR and + * SDEIIR, to avoid losing PICA IRQs and to ensure that such IRQs set + * their flags both in the PICA and SDE IIR. + */ + if (*pch_iir & SDE_PICAINTERRUPT) { + drm_WARN_ON(&i915->drm, INTEL_PCH_TYPE(i915) < PCH_MTP); + + pica_ier = intel_de_rmw(i915, PICAINTERRUPT_IER, ~0, 0); + *pica_iir = intel_de_read(i915, PICAINTERRUPT_IIR); + intel_de_write(i915, PICAINTERRUPT_IIR, *pica_iir); + } + + intel_de_write(i915, SDEIIR, *pch_iir); + + if (pica_ier) + intel_de_write(i915, PICAINTERRUPT_IER, pica_ier); +} + +void gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl) +{ + u32 iir; + enum pipe pipe; + + drm_WARN_ON_ONCE(&dev_priv->drm, !HAS_DISPLAY(dev_priv)); + + if (master_ctl & GEN8_DE_MISC_IRQ) { + iir = intel_uncore_read(&dev_priv->uncore, GEN8_DE_MISC_IIR); + if (iir) { + intel_uncore_write(&dev_priv->uncore, GEN8_DE_MISC_IIR, iir); + gen8_de_misc_irq_handler(dev_priv, iir); + } else { + drm_err_ratelimited(&dev_priv->drm, + "The master control interrupt lied (DE MISC)!\n"); + } + } + + if (DISPLAY_VER(dev_priv) >= 11 && (master_ctl & GEN11_DE_HPD_IRQ)) { + iir = intel_uncore_read(&dev_priv->uncore, GEN11_DE_HPD_IIR); + if (iir) { + intel_uncore_write(&dev_priv->uncore, GEN11_DE_HPD_IIR, iir); + gen11_hpd_irq_handler(dev_priv, iir); + } else { + drm_err_ratelimited(&dev_priv->drm, + "The master control interrupt lied, (DE HPD)!\n"); + } + } + + if (master_ctl & GEN8_DE_PORT_IRQ) { + iir = intel_uncore_read(&dev_priv->uncore, GEN8_DE_PORT_IIR); + if (iir) { + bool found = false; + + intel_uncore_write(&dev_priv->uncore, GEN8_DE_PORT_IIR, iir); + + if (iir & gen8_de_port_aux_mask(dev_priv)) { + intel_dp_aux_irq_handler(dev_priv); + found = true; + } + + if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) { + u32 hotplug_trigger = iir & BXT_DE_PORT_HOTPLUG_MASK; + + if (hotplug_trigger) { + bxt_hpd_irq_handler(dev_priv, hotplug_trigger); + found = true; + } + } else if (IS_BROADWELL(dev_priv)) { + u32 hotplug_trigger = iir & BDW_DE_PORT_HOTPLUG_MASK; + + if (hotplug_trigger) { + ilk_hpd_irq_handler(dev_priv, hotplug_trigger); + found = true; + } + } + + if ((IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) && + (iir & BXT_DE_PORT_GMBUS)) { + intel_gmbus_irq_handler(dev_priv); + found = true; + } + + if (DISPLAY_VER(dev_priv) >= 11) { + u32 te_trigger = iir & (DSI0_TE | DSI1_TE); + + if (te_trigger) { + gen11_dsi_te_interrupt_handler(dev_priv, te_trigger); + found = true; + } + } + + if (!found) + drm_err_ratelimited(&dev_priv->drm, + "Unexpected DE Port interrupt\n"); + } else { + drm_err_ratelimited(&dev_priv->drm, + "The master control interrupt lied (DE PORT)!\n"); + } + } + + for_each_pipe(dev_priv, pipe) { + u32 fault_errors; + + if (!(master_ctl & GEN8_DE_PIPE_IRQ(pipe))) + continue; + + iir = intel_uncore_read(&dev_priv->uncore, GEN8_DE_PIPE_IIR(pipe)); + if (!iir) { + drm_err_ratelimited(&dev_priv->drm, + "The master control interrupt lied (DE PIPE)!\n"); + continue; + } + + intel_uncore_write(&dev_priv->uncore, GEN8_DE_PIPE_IIR(pipe), iir); + + if (iir & GEN8_PIPE_VBLANK) + intel_handle_vblank(dev_priv, pipe); + + if (iir & gen8_de_pipe_flip_done_mask(dev_priv)) + flip_done_handler(dev_priv, pipe); + + if (iir & GEN8_PIPE_CDCLK_CRC_DONE) + hsw_pipe_crc_irq_handler(dev_priv, pipe); + + if (iir & gen8_de_pipe_underrun_mask(dev_priv)) + intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe); + + fault_errors = iir & gen8_de_pipe_fault_mask(dev_priv); + if (fault_errors) + drm_err_ratelimited(&dev_priv->drm, + "Fault errors on pipe %c: 0x%08x\n", + pipe_name(pipe), + fault_errors); + } + + if (HAS_PCH_SPLIT(dev_priv) && !HAS_PCH_NOP(dev_priv) && + master_ctl & GEN8_DE_PCH_IRQ) { + u32 pica_iir; + + /* + * FIXME(BDW): Assume for now that the new interrupt handling + * scheme also closed the SDE interrupt handling race we've seen + * on older pch-split platforms. But this needs testing. + */ + gen8_read_and_ack_pch_irqs(dev_priv, &iir, &pica_iir); + if (iir) { + if (pica_iir) + xelpdp_pica_irq_handler(dev_priv, pica_iir); + + if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP) + icp_irq_handler(dev_priv, iir); + else if (INTEL_PCH_TYPE(dev_priv) >= PCH_SPT) + spt_irq_handler(dev_priv, iir); + else + cpt_irq_handler(dev_priv, iir); + } else { + /* + * Like on previous PCH there seems to be something + * fishy going on with forwarding PCH interrupts. + */ + drm_dbg(&dev_priv->drm, + "The master control interrupt lied (SDE)!\n"); + } + } +} + +u32 gen11_gu_misc_irq_ack(struct drm_i915_private *i915, const u32 master_ctl) +{ + void __iomem * const regs = i915->uncore.regs; + u32 iir; + + if (!(master_ctl & GEN11_GU_MISC_IRQ)) + return 0; + + iir = raw_reg_read(regs, GEN11_GU_MISC_IIR); + if (likely(iir)) + raw_reg_write(regs, GEN11_GU_MISC_IIR, iir); + + return iir; +} + +void gen11_gu_misc_irq_handler(struct drm_i915_private *i915, const u32 iir) +{ + if (iir & GEN11_GU_MISC_GSE) + intel_opregion_asle_intr(i915); +} + +void gen11_display_irq_handler(struct drm_i915_private *i915) +{ + void __iomem * const regs = i915->uncore.regs; + const u32 disp_ctl = raw_reg_read(regs, GEN11_DISPLAY_INT_CTL); + + disable_rpm_wakeref_asserts(&i915->runtime_pm); + /* + * GEN11_DISPLAY_INT_CTL has same format as GEN8_MASTER_IRQ + * for the display related bits. + */ + raw_reg_write(regs, GEN11_DISPLAY_INT_CTL, 0x0); + gen8_de_irq_handler(i915, disp_ctl); + raw_reg_write(regs, GEN11_DISPLAY_INT_CTL, + GEN11_DISPLAY_IRQ_ENABLE); + + enable_rpm_wakeref_asserts(&i915->runtime_pm); +} + +/* Called from drm generic code, passed 'crtc' which + * we use as a pipe index + */ +int i8xx_enable_vblank(struct drm_crtc *crtc) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->dev); + enum pipe pipe = to_intel_crtc(crtc)->pipe; + unsigned long irqflags; + + spin_lock_irqsave(&dev_priv->irq_lock, irqflags); + i915_enable_pipestat(dev_priv, pipe, PIPE_VBLANK_INTERRUPT_STATUS); + spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); + + return 0; +} + +int i915gm_enable_vblank(struct drm_crtc *crtc) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->dev); + + /* + * Vblank interrupts fail to wake the device up from C2+. + * Disabling render clock gating during C-states avoids + * the problem. There is a small power cost so we do this + * only when vblank interrupts are actually enabled. + */ + if (dev_priv->vblank_enabled++ == 0) + intel_uncore_write(&dev_priv->uncore, SCPD0, _MASKED_BIT_ENABLE(CSTATE_RENDER_CLOCK_GATE_DISABLE)); + + return i8xx_enable_vblank(crtc); +} + +int i965_enable_vblank(struct drm_crtc *crtc) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->dev); + enum pipe pipe = to_intel_crtc(crtc)->pipe; + unsigned long irqflags; + + spin_lock_irqsave(&dev_priv->irq_lock, irqflags); + i915_enable_pipestat(dev_priv, pipe, + PIPE_START_VBLANK_INTERRUPT_STATUS); + spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); + + return 0; +} + +int ilk_enable_vblank(struct drm_crtc *crtc) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->dev); + enum pipe pipe = to_intel_crtc(crtc)->pipe; + unsigned long irqflags; + u32 bit = DISPLAY_VER(dev_priv) >= 7 ? + DE_PIPE_VBLANK_IVB(pipe) : DE_PIPE_VBLANK(pipe); + + spin_lock_irqsave(&dev_priv->irq_lock, irqflags); + ilk_enable_display_irq(dev_priv, bit); + spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); + + /* Even though there is no DMC, frame counter can get stuck when + * PSR is active as no frames are generated. + */ + if (HAS_PSR(dev_priv)) + drm_crtc_vblank_restore(crtc); + + return 0; +} + +static bool gen11_dsi_configure_te(struct intel_crtc *intel_crtc, + bool enable) +{ + struct drm_i915_private *dev_priv = to_i915(intel_crtc->base.dev); + enum port port; + + if (!(intel_crtc->mode_flags & + (I915_MODE_FLAG_DSI_USE_TE1 | I915_MODE_FLAG_DSI_USE_TE0))) + return false; + + /* for dual link cases we consider TE from slave */ + if (intel_crtc->mode_flags & I915_MODE_FLAG_DSI_USE_TE1) + port = PORT_B; + else + port = PORT_A; + + intel_uncore_rmw(&dev_priv->uncore, DSI_INTR_MASK_REG(port), DSI_TE_EVENT, + enable ? 0 : DSI_TE_EVENT); + + intel_uncore_rmw(&dev_priv->uncore, DSI_INTR_IDENT_REG(port), 0, 0); + + return true; +} + +int bdw_enable_vblank(struct drm_crtc *_crtc) +{ + struct intel_crtc *crtc = to_intel_crtc(_crtc); + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + enum pipe pipe = crtc->pipe; + unsigned long irqflags; + + if (gen11_dsi_configure_te(crtc, true)) + return 0; + + spin_lock_irqsave(&dev_priv->irq_lock, irqflags); + bdw_enable_pipe_irq(dev_priv, pipe, GEN8_PIPE_VBLANK); + spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); + + /* Even if there is no DMC, frame counter can get stuck when + * PSR is active as no frames are generated, so check only for PSR. + */ + if (HAS_PSR(dev_priv)) + drm_crtc_vblank_restore(&crtc->base); + + return 0; +} + +/* Called from drm generic code, passed 'crtc' which + * we use as a pipe index + */ +void i8xx_disable_vblank(struct drm_crtc *crtc) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->dev); + enum pipe pipe = to_intel_crtc(crtc)->pipe; + unsigned long irqflags; + + spin_lock_irqsave(&dev_priv->irq_lock, irqflags); + i915_disable_pipestat(dev_priv, pipe, PIPE_VBLANK_INTERRUPT_STATUS); + spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); +} + +void i915gm_disable_vblank(struct drm_crtc *crtc) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->dev); + + i8xx_disable_vblank(crtc); + + if (--dev_priv->vblank_enabled == 0) + intel_uncore_write(&dev_priv->uncore, SCPD0, _MASKED_BIT_DISABLE(CSTATE_RENDER_CLOCK_GATE_DISABLE)); +} + +void i965_disable_vblank(struct drm_crtc *crtc) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->dev); + enum pipe pipe = to_intel_crtc(crtc)->pipe; + unsigned long irqflags; + + spin_lock_irqsave(&dev_priv->irq_lock, irqflags); + i915_disable_pipestat(dev_priv, pipe, + PIPE_START_VBLANK_INTERRUPT_STATUS); + spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); +} + +void ilk_disable_vblank(struct drm_crtc *crtc) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->dev); + enum pipe pipe = to_intel_crtc(crtc)->pipe; + unsigned long irqflags; + u32 bit = DISPLAY_VER(dev_priv) >= 7 ? + DE_PIPE_VBLANK_IVB(pipe) : DE_PIPE_VBLANK(pipe); + + spin_lock_irqsave(&dev_priv->irq_lock, irqflags); + ilk_disable_display_irq(dev_priv, bit); + spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); +} + +void bdw_disable_vblank(struct drm_crtc *_crtc) +{ + struct intel_crtc *crtc = to_intel_crtc(_crtc); + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + enum pipe pipe = crtc->pipe; + unsigned long irqflags; + + if (gen11_dsi_configure_te(crtc, false)) + return; + + spin_lock_irqsave(&dev_priv->irq_lock, irqflags); + bdw_disable_pipe_irq(dev_priv, pipe, GEN8_PIPE_VBLANK); + spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); +} + +void vlv_display_irq_reset(struct drm_i915_private *dev_priv) +{ + struct intel_uncore *uncore = &dev_priv->uncore; + + if (IS_CHERRYVIEW(dev_priv)) + intel_uncore_write(uncore, DPINVGTT, DPINVGTT_STATUS_MASK_CHV); + else + intel_uncore_write(uncore, DPINVGTT, DPINVGTT_STATUS_MASK_VLV); + + i915_hotplug_interrupt_update_locked(dev_priv, 0xffffffff, 0); + intel_uncore_rmw(uncore, PORT_HOTPLUG_STAT, 0, 0); + + i9xx_pipestat_irq_reset(dev_priv); + + GEN3_IRQ_RESET(uncore, VLV_); + dev_priv->irq_mask = ~0u; +} + +void vlv_display_irq_postinstall(struct drm_i915_private *dev_priv) +{ + struct intel_uncore *uncore = &dev_priv->uncore; + + u32 pipestat_mask; + u32 enable_mask; + enum pipe pipe; + + pipestat_mask = PIPE_CRC_DONE_INTERRUPT_STATUS; + + i915_enable_pipestat(dev_priv, PIPE_A, PIPE_GMBUS_INTERRUPT_STATUS); + for_each_pipe(dev_priv, pipe) + i915_enable_pipestat(dev_priv, pipe, pipestat_mask); + + enable_mask = I915_DISPLAY_PORT_INTERRUPT | + I915_DISPLAY_PIPE_A_EVENT_INTERRUPT | + I915_DISPLAY_PIPE_B_EVENT_INTERRUPT | + I915_LPE_PIPE_A_INTERRUPT | + I915_LPE_PIPE_B_INTERRUPT; + + if (IS_CHERRYVIEW(dev_priv)) + enable_mask |= I915_DISPLAY_PIPE_C_EVENT_INTERRUPT | + I915_LPE_PIPE_C_INTERRUPT; + + drm_WARN_ON(&dev_priv->drm, dev_priv->irq_mask != ~0u); + + dev_priv->irq_mask = ~enable_mask; + + GEN3_IRQ_INIT(uncore, VLV_, dev_priv->irq_mask, enable_mask); +} + +void gen8_display_irq_reset(struct drm_i915_private *dev_priv) +{ + struct intel_uncore *uncore = &dev_priv->uncore; + enum pipe pipe; + + if (!HAS_DISPLAY(dev_priv)) + return; + + intel_uncore_write(uncore, EDP_PSR_IMR, 0xffffffff); + intel_uncore_write(uncore, EDP_PSR_IIR, 0xffffffff); + + for_each_pipe(dev_priv, pipe) + if (intel_display_power_is_enabled(dev_priv, + POWER_DOMAIN_PIPE(pipe))) + GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe); + + GEN3_IRQ_RESET(uncore, GEN8_DE_PORT_); + GEN3_IRQ_RESET(uncore, GEN8_DE_MISC_); +} + +void gen11_display_irq_reset(struct drm_i915_private *dev_priv) +{ + struct intel_uncore *uncore = &dev_priv->uncore; + enum pipe pipe; + u32 trans_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | + BIT(TRANSCODER_C) | BIT(TRANSCODER_D); + + if (!HAS_DISPLAY(dev_priv)) + return; + + intel_uncore_write(uncore, GEN11_DISPLAY_INT_CTL, 0); + + if (DISPLAY_VER(dev_priv) >= 12) { + enum transcoder trans; + + for_each_cpu_transcoder_masked(dev_priv, trans, trans_mask) { + enum intel_display_power_domain domain; + + domain = POWER_DOMAIN_TRANSCODER(trans); + if (!intel_display_power_is_enabled(dev_priv, domain)) + continue; + + intel_uncore_write(uncore, TRANS_PSR_IMR(trans), 0xffffffff); + intel_uncore_write(uncore, TRANS_PSR_IIR(trans), 0xffffffff); + } + } else { + intel_uncore_write(uncore, EDP_PSR_IMR, 0xffffffff); + intel_uncore_write(uncore, EDP_PSR_IIR, 0xffffffff); + } + + for_each_pipe(dev_priv, pipe) + if (intel_display_power_is_enabled(dev_priv, + POWER_DOMAIN_PIPE(pipe))) + GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe); + + GEN3_IRQ_RESET(uncore, GEN8_DE_PORT_); + GEN3_IRQ_RESET(uncore, GEN8_DE_MISC_); + + if (DISPLAY_VER(dev_priv) >= 14) + GEN3_IRQ_RESET(uncore, PICAINTERRUPT_); + else + GEN3_IRQ_RESET(uncore, GEN11_DE_HPD_); + + if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP) + GEN3_IRQ_RESET(uncore, SDE); +} + +void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv, + u8 pipe_mask) +{ + struct intel_uncore *uncore = &dev_priv->uncore; + u32 extra_ier = GEN8_PIPE_VBLANK | + gen8_de_pipe_underrun_mask(dev_priv) | + gen8_de_pipe_flip_done_mask(dev_priv); + enum pipe pipe; + + spin_lock_irq(&dev_priv->irq_lock); + + if (!intel_irqs_enabled(dev_priv)) { + spin_unlock_irq(&dev_priv->irq_lock); + return; + } + + for_each_pipe_masked(dev_priv, pipe, pipe_mask) + GEN8_IRQ_INIT_NDX(uncore, DE_PIPE, pipe, + dev_priv->de_irq_mask[pipe], + ~dev_priv->de_irq_mask[pipe] | extra_ier); + + spin_unlock_irq(&dev_priv->irq_lock); +} + +void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv, + u8 pipe_mask) +{ + struct intel_uncore *uncore = &dev_priv->uncore; + enum pipe pipe; + + spin_lock_irq(&dev_priv->irq_lock); + + if (!intel_irqs_enabled(dev_priv)) { + spin_unlock_irq(&dev_priv->irq_lock); + return; + } + + for_each_pipe_masked(dev_priv, pipe, pipe_mask) + GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe); + + spin_unlock_irq(&dev_priv->irq_lock); + + /* make sure we're done processing display irqs */ + intel_synchronize_irq(dev_priv); +} + +/* + * SDEIER is also touched by the interrupt handler to work around missed PCH + * interrupts. Hence we can't update it after the interrupt handler is enabled - + * instead we unconditionally enable all PCH interrupt sources here, but then + * only unmask them as needed with SDEIMR. + * + * Note that we currently do this after installing the interrupt handler, + * but before we enable the master interrupt. That should be sufficient + * to avoid races with the irq handler, assuming we have MSI. Shared legacy + * interrupts could still race. + */ +void ibx_irq_postinstall(struct drm_i915_private *dev_priv) +{ + struct intel_uncore *uncore = &dev_priv->uncore; + u32 mask; + + if (HAS_PCH_NOP(dev_priv)) + return; + + if (HAS_PCH_IBX(dev_priv)) + mask = SDE_GMBUS | SDE_AUX_MASK | SDE_POISON; + else if (HAS_PCH_CPT(dev_priv) || HAS_PCH_LPT(dev_priv)) + mask = SDE_GMBUS_CPT | SDE_AUX_MASK_CPT; + else + mask = SDE_GMBUS_CPT; + + GEN3_IRQ_INIT(uncore, SDE, ~mask, 0xffffffff); +} + +void valleyview_enable_display_irqs(struct drm_i915_private *dev_priv) +{ + lockdep_assert_held(&dev_priv->irq_lock); + + if (dev_priv->display_irqs_enabled) + return; + + dev_priv->display_irqs_enabled = true; + + if (intel_irqs_enabled(dev_priv)) { + vlv_display_irq_reset(dev_priv); + vlv_display_irq_postinstall(dev_priv); + } +} + +void valleyview_disable_display_irqs(struct drm_i915_private *dev_priv) +{ + lockdep_assert_held(&dev_priv->irq_lock); + + if (!dev_priv->display_irqs_enabled) + return; + + dev_priv->display_irqs_enabled = false; + + if (intel_irqs_enabled(dev_priv)) + vlv_display_irq_reset(dev_priv); +} + +void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv) +{ + struct intel_uncore *uncore = &dev_priv->uncore; + + u32 de_pipe_masked = gen8_de_pipe_fault_mask(dev_priv) | + GEN8_PIPE_CDCLK_CRC_DONE; + u32 de_pipe_enables; + u32 de_port_masked = gen8_de_port_aux_mask(dev_priv); + u32 de_port_enables; + u32 de_misc_masked = GEN8_DE_EDP_PSR; + u32 trans_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | + BIT(TRANSCODER_C) | BIT(TRANSCODER_D); + enum pipe pipe; + + if (!HAS_DISPLAY(dev_priv)) + return; + + if (DISPLAY_VER(dev_priv) <= 10) + de_misc_masked |= GEN8_DE_MISC_GSE; + + if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) + de_port_masked |= BXT_DE_PORT_GMBUS; + + if (DISPLAY_VER(dev_priv) >= 14) { + de_misc_masked |= XELPDP_PMDEMAND_RSPTOUT_ERR | + XELPDP_PMDEMAND_RSP; + } else if (DISPLAY_VER(dev_priv) >= 11) { + enum port port; + + if (intel_bios_is_dsi_present(dev_priv, &port)) + de_port_masked |= DSI0_TE | DSI1_TE; + } + + de_pipe_enables = de_pipe_masked | + GEN8_PIPE_VBLANK | + gen8_de_pipe_underrun_mask(dev_priv) | + gen8_de_pipe_flip_done_mask(dev_priv); + + de_port_enables = de_port_masked; + if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) + de_port_enables |= BXT_DE_PORT_HOTPLUG_MASK; + else if (IS_BROADWELL(dev_priv)) + de_port_enables |= BDW_DE_PORT_HOTPLUG_MASK; + + if (DISPLAY_VER(dev_priv) >= 12) { + enum transcoder trans; + + for_each_cpu_transcoder_masked(dev_priv, trans, trans_mask) { + enum intel_display_power_domain domain; + + domain = POWER_DOMAIN_TRANSCODER(trans); + if (!intel_display_power_is_enabled(dev_priv, domain)) + continue; + + gen3_assert_iir_is_zero(uncore, TRANS_PSR_IIR(trans)); + } + } else { + gen3_assert_iir_is_zero(uncore, EDP_PSR_IIR); + } + + for_each_pipe(dev_priv, pipe) { + dev_priv->de_irq_mask[pipe] = ~de_pipe_masked; + + if (intel_display_power_is_enabled(dev_priv, + POWER_DOMAIN_PIPE(pipe))) + GEN8_IRQ_INIT_NDX(uncore, DE_PIPE, pipe, + dev_priv->de_irq_mask[pipe], + de_pipe_enables); + } + + GEN3_IRQ_INIT(uncore, GEN8_DE_PORT_, ~de_port_masked, de_port_enables); + GEN3_IRQ_INIT(uncore, GEN8_DE_MISC_, ~de_misc_masked, de_misc_masked); + + if (IS_DISPLAY_VER(dev_priv, 11, 13)) { + u32 de_hpd_masked = 0; + u32 de_hpd_enables = GEN11_DE_TC_HOTPLUG_MASK | + GEN11_DE_TBT_HOTPLUG_MASK; + + GEN3_IRQ_INIT(uncore, GEN11_DE_HPD_, ~de_hpd_masked, + de_hpd_enables); + } +} + +void mtp_irq_postinstall(struct drm_i915_private *i915) +{ + struct intel_uncore *uncore = &i915->uncore; + u32 sde_mask = SDE_GMBUS_ICP | SDE_PICAINTERRUPT; + u32 de_hpd_mask = XELPDP_AUX_TC_MASK; + u32 de_hpd_enables = de_hpd_mask | XELPDP_DP_ALT_HOTPLUG_MASK | + XELPDP_TBT_HOTPLUG_MASK; + + GEN3_IRQ_INIT(uncore, PICAINTERRUPT_, ~de_hpd_mask, + de_hpd_enables); + + GEN3_IRQ_INIT(uncore, SDE, ~sde_mask, 0xffffffff); +} + +void icp_irq_postinstall(struct drm_i915_private *dev_priv) +{ + struct intel_uncore *uncore = &dev_priv->uncore; + u32 mask = SDE_GMBUS_ICP; + + GEN3_IRQ_INIT(uncore, SDE, ~mask, 0xffffffff); +} + +void gen11_de_irq_postinstall(struct drm_i915_private *dev_priv) +{ + if (!HAS_DISPLAY(dev_priv)) + return; + + gen8_de_irq_postinstall(dev_priv); + + intel_uncore_write(&dev_priv->uncore, GEN11_DISPLAY_INT_CTL, + GEN11_DISPLAY_IRQ_ENABLE); +} + diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.h b/drivers/gpu/drm/i915/display/intel_display_irq.h new file mode 100644 index 0000000000000000000000000000000000000000..874893f4f16dbdf6f17af5b32969e9a5b7a7a6fa --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_display_irq.h @@ -0,0 +1,81 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2023 Intel Corporation + */ + +#ifndef __INTEL_DISPLAY_IRQ_H__ +#define __INTEL_DISPLAY_IRQ_H__ + +#include + +#include "intel_display_limits.h" + +enum pipe; +struct drm_i915_private; +struct drm_crtc; + +void valleyview_enable_display_irqs(struct drm_i915_private *i915); +void valleyview_disable_display_irqs(struct drm_i915_private *i915); + +void ilk_update_display_irq(struct drm_i915_private *i915, + u32 interrupt_mask, u32 enabled_irq_mask); +void ilk_enable_display_irq(struct drm_i915_private *i915, u32 bits); +void ilk_disable_display_irq(struct drm_i915_private *i915, u32 bits); + +void bdw_update_port_irq(struct drm_i915_private *i915, u32 interrupt_mask, u32 enabled_irq_mask); +void bdw_enable_pipe_irq(struct drm_i915_private *i915, enum pipe pipe, u32 bits); +void bdw_disable_pipe_irq(struct drm_i915_private *i915, enum pipe pipe, u32 bits); + +void ibx_display_interrupt_update(struct drm_i915_private *i915, + u32 interrupt_mask, u32 enabled_irq_mask); +void ibx_enable_display_interrupt(struct drm_i915_private *i915, u32 bits); +void ibx_disable_display_interrupt(struct drm_i915_private *i915, u32 bits); + +void gen8_irq_power_well_post_enable(struct drm_i915_private *i915, u8 pipe_mask); +void gen8_irq_power_well_pre_disable(struct drm_i915_private *i915, u8 pipe_mask); +u32 gen8_de_pipe_underrun_mask(struct drm_i915_private *i915); + +int i8xx_enable_vblank(struct drm_crtc *crtc); +int i915gm_enable_vblank(struct drm_crtc *crtc); +int i965_enable_vblank(struct drm_crtc *crtc); +int ilk_enable_vblank(struct drm_crtc *crtc); +int bdw_enable_vblank(struct drm_crtc *crtc); +void i8xx_disable_vblank(struct drm_crtc *crtc); +void i915gm_disable_vblank(struct drm_crtc *crtc); +void i965_disable_vblank(struct drm_crtc *crtc); +void ilk_disable_vblank(struct drm_crtc *crtc); +void bdw_disable_vblank(struct drm_crtc *crtc); + +void ivb_display_irq_handler(struct drm_i915_private *i915, u32 de_iir); +void ilk_display_irq_handler(struct drm_i915_private *i915, u32 de_iir); +void gen8_de_irq_handler(struct drm_i915_private *i915, u32 master_ctl); +void gen11_display_irq_handler(struct drm_i915_private *i915); + +u32 gen11_gu_misc_irq_ack(struct drm_i915_private *i915, const u32 master_ctl); +void gen11_gu_misc_irq_handler(struct drm_i915_private *i915, const u32 iir); + +void vlv_display_irq_reset(struct drm_i915_private *i915); +void gen8_display_irq_reset(struct drm_i915_private *i915); +void gen11_display_irq_reset(struct drm_i915_private *i915); + +void ibx_irq_postinstall(struct drm_i915_private *i915); +void vlv_display_irq_postinstall(struct drm_i915_private *i915); +void icp_irq_postinstall(struct drm_i915_private *i915); +void gen8_de_irq_postinstall(struct drm_i915_private *i915); +void mtp_irq_postinstall(struct drm_i915_private *i915); +void gen11_de_irq_postinstall(struct drm_i915_private *i915); + +u32 i915_pipestat_enable_mask(struct drm_i915_private *i915, enum pipe pipe); +void i915_enable_pipestat(struct drm_i915_private *i915, enum pipe pipe, u32 status_mask); +void i915_disable_pipestat(struct drm_i915_private *i915, enum pipe pipe, u32 status_mask); +void i915_enable_asle_pipestat(struct drm_i915_private *i915); +void i9xx_pipestat_irq_reset(struct drm_i915_private *i915); + +void i9xx_pipestat_irq_ack(struct drm_i915_private *i915, u32 iir, u32 pipe_stats[I915_MAX_PIPES]); + +void i915_pipestat_irq_handler(struct drm_i915_private *i915, u32 iir, u32 pipe_stats[I915_MAX_PIPES]); +void i965_pipestat_irq_handler(struct drm_i915_private *i915, u32 iir, u32 pipe_stats[I915_MAX_PIPES]); +void valleyview_pipestat_irq_handler(struct drm_i915_private *i915, u32 pipe_stats[I915_MAX_PIPES]); +void i8xx_pipestat_irq_handler(struct drm_i915_private *i915, u16 iir, u32 pipe_stats[I915_MAX_PIPES]); + +#endif /* __INTEL_DISPLAY_IRQ_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c index 7c9f4288329ede68b644038b4585b14485f7f985..db54370439042b24de78ec2365ad1f6a4c88932e 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power.c +++ b/drivers/gpu/drm/i915/display/intel_display_power.c @@ -7,6 +7,7 @@ #include "i915_drv.h" #include "i915_irq.h" +#include "i915_reg.h" #include "intel_backlight_regs.h" #include "intel_cdclk.h" #include "intel_combo_phy.h" @@ -19,6 +20,7 @@ #include "intel_mchbar_regs.h" #include "intel_pch_refclk.h" #include "intel_pcode.h" +#include "intel_pmdemand.h" #include "intel_pps_regs.h" #include "intel_snps_phy.h" #include "skl_watermark.h" @@ -1052,7 +1054,7 @@ void gen9_dbuf_slices_update(struct drm_i915_private *dev_priv, u8 req_slices) { struct i915_power_domains *power_domains = &dev_priv->display.power.domains; - u8 slice_mask = INTEL_INFO(dev_priv)->display.dbuf.slice_mask; + u8 slice_mask = DISPLAY_INFO(dev_priv)->dbuf.slice_mask; enum dbuf_slice slice; drm_WARN(&dev_priv->drm, req_slices & ~slice_mask, @@ -1081,20 +1083,29 @@ void gen9_dbuf_slices_update(struct drm_i915_private *dev_priv, static void gen9_dbuf_enable(struct drm_i915_private *dev_priv) { + u8 slices_mask; + dev_priv->display.dbuf.enabled_slices = intel_enabled_dbuf_slices_mask(dev_priv); + slices_mask = BIT(DBUF_S1) | dev_priv->display.dbuf.enabled_slices; + + if (DISPLAY_VER(dev_priv) >= 14) + intel_pmdemand_program_dbuf(dev_priv, slices_mask); + /* * Just power up at least 1 slice, we will * figure out later which slices we have and what we need. */ - gen9_dbuf_slices_update(dev_priv, BIT(DBUF_S1) | - dev_priv->display.dbuf.enabled_slices); + gen9_dbuf_slices_update(dev_priv, slices_mask); } static void gen9_dbuf_disable(struct drm_i915_private *dev_priv) { gen9_dbuf_slices_update(dev_priv, 0); + + if (DISPLAY_VER(dev_priv) >= 14) + intel_pmdemand_program_dbuf(dev_priv, 0); } static void gen12_dbuf_slices_config(struct drm_i915_private *dev_priv) @@ -1112,7 +1123,7 @@ static void gen12_dbuf_slices_config(struct drm_i915_private *dev_priv) static void icl_mbus_init(struct drm_i915_private *dev_priv) { - unsigned long abox_regs = INTEL_INFO(dev_priv)->display.abox_mask; + unsigned long abox_regs = DISPLAY_INFO(dev_priv)->abox_mask; u32 mask, val, i; if (IS_ALDERLAKE_P(dev_priv) || DISPLAY_VER(dev_priv) >= 14) @@ -1164,31 +1175,39 @@ static void assert_can_disable_lcpll(struct drm_i915_private *dev_priv) struct intel_crtc *crtc; for_each_intel_crtc(&dev_priv->drm, crtc) - I915_STATE_WARN(crtc->active, "CRTC for pipe %c enabled\n", + I915_STATE_WARN(dev_priv, crtc->active, + "CRTC for pipe %c enabled\n", pipe_name(crtc->pipe)); - I915_STATE_WARN(intel_de_read(dev_priv, HSW_PWR_WELL_CTL2), + I915_STATE_WARN(dev_priv, intel_de_read(dev_priv, HSW_PWR_WELL_CTL2), "Display power well on\n"); - I915_STATE_WARN(intel_de_read(dev_priv, SPLL_CTL) & SPLL_PLL_ENABLE, + I915_STATE_WARN(dev_priv, + intel_de_read(dev_priv, SPLL_CTL) & SPLL_PLL_ENABLE, "SPLL enabled\n"); - I915_STATE_WARN(intel_de_read(dev_priv, WRPLL_CTL(0)) & WRPLL_PLL_ENABLE, + I915_STATE_WARN(dev_priv, + intel_de_read(dev_priv, WRPLL_CTL(0)) & WRPLL_PLL_ENABLE, "WRPLL1 enabled\n"); - I915_STATE_WARN(intel_de_read(dev_priv, WRPLL_CTL(1)) & WRPLL_PLL_ENABLE, + I915_STATE_WARN(dev_priv, + intel_de_read(dev_priv, WRPLL_CTL(1)) & WRPLL_PLL_ENABLE, "WRPLL2 enabled\n"); - I915_STATE_WARN(intel_de_read(dev_priv, PP_STATUS(0)) & PP_ON, + I915_STATE_WARN(dev_priv, + intel_de_read(dev_priv, PP_STATUS(0)) & PP_ON, "Panel power on\n"); - I915_STATE_WARN(intel_de_read(dev_priv, BLC_PWM_CPU_CTL2) & BLM_PWM_ENABLE, + I915_STATE_WARN(dev_priv, + intel_de_read(dev_priv, BLC_PWM_CPU_CTL2) & BLM_PWM_ENABLE, "CPU PWM1 enabled\n"); if (IS_HASWELL(dev_priv)) - I915_STATE_WARN(intel_de_read(dev_priv, HSW_BLC_PWM2_CTL) & BLM_PWM_ENABLE, + I915_STATE_WARN(dev_priv, + intel_de_read(dev_priv, HSW_BLC_PWM2_CTL) & BLM_PWM_ENABLE, "CPU PWM2 enabled\n"); - I915_STATE_WARN(intel_de_read(dev_priv, BLC_PWM_PCH_CTL1) & BLM_PCH_PWM_ENABLE, + I915_STATE_WARN(dev_priv, + intel_de_read(dev_priv, BLC_PWM_PCH_CTL1) & BLM_PCH_PWM_ENABLE, "PCH PWM1 enabled\n"); - I915_STATE_WARN((intel_de_read(dev_priv, UTIL_PIN_CTL) & - (UTIL_PIN_ENABLE | UTIL_PIN_MODE_MASK)) == - (UTIL_PIN_ENABLE | UTIL_PIN_MODE_PWM), + I915_STATE_WARN(dev_priv, + (intel_de_read(dev_priv, UTIL_PIN_CTL) & (UTIL_PIN_ENABLE | UTIL_PIN_MODE_MASK)) == (UTIL_PIN_ENABLE | UTIL_PIN_MODE_PWM), "Utility pin enabled in PWM mode\n"); - I915_STATE_WARN(intel_de_read(dev_priv, PCH_GTC_CTL) & PCH_GTC_ENABLE, + I915_STATE_WARN(dev_priv, + intel_de_read(dev_priv, PCH_GTC_CTL) & PCH_GTC_ENABLE, "PCH GTC enabled\n"); /* @@ -1197,7 +1216,8 @@ static void assert_can_disable_lcpll(struct drm_i915_private *dev_priv) * gen-specific and since we only disable LCPLL after we fully disable * the interrupts, the check below should be enough. */ - I915_STATE_WARN(intel_irqs_enabled(dev_priv), "IRQs enabled\n"); + I915_STATE_WARN(dev_priv, intel_irqs_enabled(dev_priv), + "IRQs enabled\n"); } static u32 hsw_read_dcomp(struct drm_i915_private *dev_priv) @@ -1558,7 +1578,7 @@ static void tgl_bw_buddy_init(struct drm_i915_private *dev_priv) enum intel_dram_type type = dev_priv->dram_info.type; u8 num_channels = dev_priv->dram_info.num_channels; const struct buddy_page_mask *table; - unsigned long abox_mask = INTEL_INFO(dev_priv)->display.abox_mask; + unsigned long abox_mask = DISPLAY_INFO(dev_priv)->abox_mask; int config, i; /* BW_BUDDY registers are not used on dgpu's beyond DG1 */ @@ -2021,7 +2041,7 @@ void intel_power_domains_disable(struct drm_i915_private *i915) /** * intel_power_domains_suspend - suspend power domain state * @i915: i915 device instance - * @suspend_mode: specifies the target suspend state (idle, mem, hibernation) + * @s2idle: specifies whether we go to idle, or deeper sleep * * This function prepares the hardware power domain state before entering * system suspend. @@ -2029,8 +2049,7 @@ void intel_power_domains_disable(struct drm_i915_private *i915) * It must be called with power domains already disabled (after a call to * intel_power_domains_disable()) and paired with intel_power_domains_resume(). */ -void intel_power_domains_suspend(struct drm_i915_private *i915, - enum i915_drm_suspend_mode suspend_mode) +void intel_power_domains_suspend(struct drm_i915_private *i915, bool s2idle) { struct i915_power_domains *power_domains = &i915->display.power.domains; intel_wakeref_t wakeref __maybe_unused = @@ -2045,8 +2064,7 @@ void intel_power_domains_suspend(struct drm_i915_private *i915, * resources as required and also enable deeper system power states * that would be blocked if the firmware was inactive. */ - if (!(power_domains->allowed_dc_mask & DC_STATE_EN_DC9) && - suspend_mode == I915_DRM_SUSPEND_IDLE && + if (!(power_domains->allowed_dc_mask & DC_STATE_EN_DC9) && s2idle && intel_dmc_has_payload(i915)) { intel_display_power_flush_work(i915); intel_power_domains_verify_state(i915); diff --git a/drivers/gpu/drm/i915/display/intel_display_power.h b/drivers/gpu/drm/i915/display/intel_display_power.h index 8e96be8e6330ac6ac3bd2238b23ca4728242ce07..be1a87bde0c9dec69f27b39405ef43cb66435540 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power.h +++ b/drivers/gpu/drm/i915/display/intel_display_power.h @@ -171,8 +171,7 @@ void intel_power_domains_init_hw(struct drm_i915_private *dev_priv, bool resume) void intel_power_domains_driver_remove(struct drm_i915_private *dev_priv); void intel_power_domains_enable(struct drm_i915_private *dev_priv); void intel_power_domains_disable(struct drm_i915_private *dev_priv); -void intel_power_domains_suspend(struct drm_i915_private *dev_priv, - enum i915_drm_suspend_mode); +void intel_power_domains_suspend(struct drm_i915_private *dev_priv, bool s2idle); void intel_power_domains_resume(struct drm_i915_private *dev_priv); void intel_power_domains_sanitize_state(struct drm_i915_private *dev_priv); diff --git a/drivers/gpu/drm/i915/display/intel_display_power_map.c b/drivers/gpu/drm/i915/display/intel_display_power_map.c index 6645eb1911d8507ea161c93d3a6629180e9e7bc1..1118ee9d224cabe25be352d0ce55410dd2b0086f 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power_map.c +++ b/drivers/gpu/drm/i915/display/intel_display_power_map.c @@ -1251,22 +1251,11 @@ I915_DECL_PW_DOMAINS(xelpd_pwdoms_pw_a, POWER_DOMAIN_PIPE_PANEL_FITTER_A, POWER_DOMAIN_INIT); -#define XELPD_PW_2_POWER_DOMAINS \ - XELPD_PW_B_POWER_DOMAINS, \ - XELPD_PW_C_POWER_DOMAINS, \ - XELPD_PW_D_POWER_DOMAINS, \ - POWER_DOMAIN_PORT_DDI_LANES_C, \ - POWER_DOMAIN_PORT_DDI_LANES_D, \ - POWER_DOMAIN_PORT_DDI_LANES_E, \ +#define XELPD_DC_OFF_PORT_POWER_DOMAINS \ POWER_DOMAIN_PORT_DDI_LANES_TC1, \ POWER_DOMAIN_PORT_DDI_LANES_TC2, \ POWER_DOMAIN_PORT_DDI_LANES_TC3, \ POWER_DOMAIN_PORT_DDI_LANES_TC4, \ - POWER_DOMAIN_VGA, \ - POWER_DOMAIN_AUDIO_PLAYBACK, \ - POWER_DOMAIN_AUX_IO_C, \ - POWER_DOMAIN_AUX_IO_D, \ - POWER_DOMAIN_AUX_IO_E, \ POWER_DOMAIN_AUX_C, \ POWER_DOMAIN_AUX_D, \ POWER_DOMAIN_AUX_E, \ @@ -1279,6 +1268,20 @@ I915_DECL_PW_DOMAINS(xelpd_pwdoms_pw_a, POWER_DOMAIN_AUX_TBT3, \ POWER_DOMAIN_AUX_TBT4 +#define XELPD_PW_2_POWER_DOMAINS \ + XELPD_PW_B_POWER_DOMAINS, \ + XELPD_PW_C_POWER_DOMAINS, \ + XELPD_PW_D_POWER_DOMAINS, \ + POWER_DOMAIN_PORT_DDI_LANES_C, \ + POWER_DOMAIN_PORT_DDI_LANES_D, \ + POWER_DOMAIN_PORT_DDI_LANES_E, \ + POWER_DOMAIN_VGA, \ + POWER_DOMAIN_AUDIO_PLAYBACK, \ + POWER_DOMAIN_AUX_IO_C, \ + POWER_DOMAIN_AUX_IO_D, \ + POWER_DOMAIN_AUX_IO_E, \ + XELPD_DC_OFF_PORT_POWER_DOMAINS + I915_DECL_PW_DOMAINS(xelpd_pwdoms_pw_2, XELPD_PW_2_POWER_DOMAINS, POWER_DOMAIN_INIT); @@ -1301,7 +1304,9 @@ I915_DECL_PW_DOMAINS(xelpd_pwdoms_pw_2, */ I915_DECL_PW_DOMAINS(xelpd_pwdoms_dc_off, - XELPD_PW_2_POWER_DOMAINS, + XELPD_DC_OFF_PORT_POWER_DOMAINS, + XELPD_PW_C_POWER_DOMAINS, + XELPD_PW_D_POWER_DOMAINS, POWER_DOMAIN_PORT_DSI, POWER_DOMAIN_AUDIO_MMIO, POWER_DOMAIN_AUX_A, @@ -1310,14 +1315,18 @@ I915_DECL_PW_DOMAINS(xelpd_pwdoms_dc_off, POWER_DOMAIN_DC_OFF, POWER_DOMAIN_INIT); -static const struct i915_power_well_desc xelpd_power_wells_main[] = { +static const struct i915_power_well_desc xelpd_power_wells_dc_off[] = { { .instances = &I915_PW_INSTANCES( I915_PW("DC_off", &xelpd_pwdoms_dc_off, .id = SKL_DISP_DC_OFF), ), .ops = &gen9_dc_off_power_well_ops, - }, { + } +}; + +static const struct i915_power_well_desc xelpd_power_wells_main[] = { + { .instances = &I915_PW_INSTANCES( I915_PW("PW_2", &xelpd_pwdoms_pw_2, .hsw.idx = ICL_PW_CTL_IDX_PW_2, @@ -1378,6 +1387,11 @@ static const struct i915_power_well_desc xelpd_power_wells_main[] = { I915_PW("AUX_C", &icl_pwdoms_aux_c, .hsw.idx = ICL_PW_CTL_IDX_AUX_C), I915_PW("AUX_D", &icl_pwdoms_aux_d, .hsw.idx = XELPD_PW_CTL_IDX_AUX_D), I915_PW("AUX_E", &icl_pwdoms_aux_e, .hsw.idx = XELPD_PW_CTL_IDX_AUX_E), + ), + .ops = &icl_aux_power_well_ops, + .fixed_enable_delay = true, + }, { + .instances = &I915_PW_INSTANCES( I915_PW("AUX_USBC1", &tgl_pwdoms_aux_usbc1, .hsw.idx = TGL_PW_CTL_IDX_AUX_TC1), I915_PW("AUX_USBC2", &tgl_pwdoms_aux_usbc2, .hsw.idx = TGL_PW_CTL_IDX_AUX_TC2), I915_PW("AUX_USBC3", &tgl_pwdoms_aux_usbc3, .hsw.idx = TGL_PW_CTL_IDX_AUX_TC3), @@ -1385,6 +1399,8 @@ static const struct i915_power_well_desc xelpd_power_wells_main[] = { ), .ops = &icl_aux_power_well_ops, .fixed_enable_delay = true, + /* WA_14017248603: adlp */ + .enable_timeout = 500, }, { .instances = &I915_PW_INSTANCES( I915_PW("AUX_TBT1", &icl_pwdoms_aux_tbt1, .hsw.idx = TGL_PW_CTL_IDX_AUX_TBT1), @@ -1400,6 +1416,34 @@ static const struct i915_power_well_desc xelpd_power_wells_main[] = { static const struct i915_power_well_desc_list xelpd_power_wells[] = { I915_PW_DESCRIPTORS(i9xx_power_wells_always_on), I915_PW_DESCRIPTORS(icl_power_wells_pw_1), + I915_PW_DESCRIPTORS(xelpd_power_wells_dc_off), + I915_PW_DESCRIPTORS(xelpd_power_wells_main), +}; + +I915_DECL_PW_DOMAINS(xehpd_pwdoms_dc_off, + XELPD_PW_2_POWER_DOMAINS, + POWER_DOMAIN_PORT_DSI, + POWER_DOMAIN_AUDIO_MMIO, + POWER_DOMAIN_AUX_A, + POWER_DOMAIN_AUX_B, + POWER_DOMAIN_MODESET, + POWER_DOMAIN_DC_OFF, + POWER_DOMAIN_INIT); + +static const struct i915_power_well_desc xehpd_power_wells_dc_off[] = { + { + .instances = &I915_PW_INSTANCES( + I915_PW("DC_off", &xehpd_pwdoms_dc_off, + .id = SKL_DISP_DC_OFF), + ), + .ops = &gen9_dc_off_power_well_ops, + } +}; + +static const struct i915_power_well_desc_list xehpd_power_wells[] = { + I915_PW_DESCRIPTORS(i9xx_power_wells_always_on), + I915_PW_DESCRIPTORS(icl_power_wells_pw_1), + I915_PW_DESCRIPTORS(xehpd_power_wells_dc_off), I915_PW_DESCRIPTORS(xelpd_power_wells_main), }; @@ -1423,15 +1467,6 @@ I915_DECL_PW_DOMAINS(xelpdp_pwdoms_pw_2, XELPDP_PW_2_POWER_DOMAINS, POWER_DOMAIN_INIT); -I915_DECL_PW_DOMAINS(xelpdp_pwdoms_dc_off, - XELPDP_PW_2_POWER_DOMAINS, - POWER_DOMAIN_AUDIO_MMIO, - POWER_DOMAIN_MODESET, - POWER_DOMAIN_AUX_A, - POWER_DOMAIN_AUX_B, - POWER_DOMAIN_DC_OFF, - POWER_DOMAIN_INIT); - I915_DECL_PW_DOMAINS(xelpdp_pwdoms_aux_tc1, POWER_DOMAIN_AUX_USBC1, POWER_DOMAIN_AUX_TBT1); @@ -1450,12 +1485,6 @@ I915_DECL_PW_DOMAINS(xelpdp_pwdoms_aux_tc4, static const struct i915_power_well_desc xelpdp_power_wells_main[] = { { - .instances = &I915_PW_INSTANCES( - I915_PW("DC_off", &xelpdp_pwdoms_dc_off, - .id = SKL_DISP_DC_OFF), - ), - .ops = &gen9_dc_off_power_well_ops, - }, { .instances = &I915_PW_INSTANCES( I915_PW("PW_2", &xelpdp_pwdoms_pw_2, .hsw.idx = ICL_PW_CTL_IDX_PW_2, @@ -1512,6 +1541,7 @@ static const struct i915_power_well_desc xelpdp_power_wells_main[] = { static const struct i915_power_well_desc_list xelpdp_power_wells[] = { I915_PW_DESCRIPTORS(i9xx_power_wells_always_on), I915_PW_DESCRIPTORS(icl_power_wells_pw_1), + I915_PW_DESCRIPTORS(xelpd_power_wells_dc_off), I915_PW_DESCRIPTORS(xelpdp_power_wells_main), }; @@ -1624,6 +1654,8 @@ int intel_display_power_map_init(struct i915_power_domains *power_domains) if (DISPLAY_VER(i915) >= 14) return set_power_wells(power_domains, xelpdp_power_wells); + else if (IS_DG2(i915)) + return set_power_wells(power_domains, xehpd_power_wells); else if (DISPLAY_VER(i915) >= 13) return set_power_wells(power_domains, xelpd_power_wells); else if (IS_DG1(i915)) diff --git a/drivers/gpu/drm/i915/display/intel_display_power_well.c b/drivers/gpu/drm/i915/display/intel_display_power_well.c index 62bafcbc7937cf759912f6c082d2214df838618f..916009894d89c7fe20c1489736d036a1f7a9b7a7 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power_well.c +++ b/drivers/gpu/drm/i915/display/intel_display_power_well.c @@ -5,11 +5,13 @@ #include "i915_drv.h" #include "i915_irq.h" +#include "i915_reg.h" #include "intel_backlight_regs.h" #include "intel_combo_phy.h" #include "intel_combo_phy_regs.h" #include "intel_crt.h" #include "intel_de.h" +#include "intel_display_irq.h" #include "intel_display_power_well.h" #include "intel_display_types.h" #include "intel_dkl_phy.h" @@ -253,6 +255,7 @@ static void hsw_wait_for_power_well_enable(struct drm_i915_private *dev_priv, { const struct i915_power_well_regs *regs = power_well->desc->ops->regs; int pw_idx = i915_power_well_instance(power_well)->hsw.idx; + int timeout = power_well->desc->enable_timeout ? : 1; /* * For some power wells we're not supposed to watch the status bit for @@ -266,7 +269,7 @@ static void hsw_wait_for_power_well_enable(struct drm_i915_private *dev_priv, /* Timeout for PW1:10 us, AUX:not specified, other PWs:20 us. */ if (intel_de_wait_for_set(dev_priv, regs->driver, - HSW_PWR_WELL_CTL_STATE(pw_idx), 1)) { + HSW_PWR_WELL_CTL_STATE(pw_idx), timeout)) { drm_dbg_kms(&dev_priv->drm, "%s power well enable timeout\n", intel_power_well_name(power_well)); diff --git a/drivers/gpu/drm/i915/display/intel_display_power_well.h b/drivers/gpu/drm/i915/display/intel_display_power_well.h index ba7cb977e7c7f7e51fdad42bb99a46b07ca009a8..e494df379e6c23fd52a29dd07a8994b357b74103 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power_well.h +++ b/drivers/gpu/drm/i915/display/intel_display_power_well.h @@ -110,6 +110,8 @@ struct i915_power_well_desc { * Thunderbolt mode. */ u16 is_tc_tbt:1; + /* Enable timeout if greater than the default 1ms */ + u16 enable_timeout; }; struct i915_power_well { diff --git a/drivers/gpu/drm/i915/display/intel_display_reg_defs.h b/drivers/gpu/drm/i915/display/intel_display_reg_defs.h index 755c1ea8225c5a6147c76de31c5bbe452fa7c944..2f07b7afa3bfea5631d40a4a62291d96d31def49 100644 --- a/drivers/gpu/drm/i915/display/intel_display_reg_defs.h +++ b/drivers/gpu/drm/i915/display/intel_display_reg_defs.h @@ -8,7 +8,7 @@ #include "i915_reg_defs.h" -#define DISPLAY_MMIO_BASE(dev_priv) (INTEL_INFO(dev_priv)->display.mmio_offset) +#define DISPLAY_MMIO_BASE(dev_priv) (DISPLAY_INFO(dev_priv)->mmio_offset) #define VLV_DISPLAY_BASE 0x180000 @@ -36,14 +36,14 @@ * Device info offset array based helpers for groups of registers with unevenly * spaced base offsets. */ -#define _MMIO_PIPE2(pipe, reg) _MMIO(INTEL_INFO(dev_priv)->display.pipe_offsets[(pipe)] - \ - INTEL_INFO(dev_priv)->display.pipe_offsets[PIPE_A] + \ +#define _MMIO_PIPE2(pipe, reg) _MMIO(DISPLAY_INFO(dev_priv)->pipe_offsets[(pipe)] - \ + DISPLAY_INFO(dev_priv)->pipe_offsets[PIPE_A] + \ DISPLAY_MMIO_BASE(dev_priv) + (reg)) -#define _MMIO_TRANS2(tran, reg) _MMIO(INTEL_INFO(dev_priv)->display.trans_offsets[(tran)] - \ - INTEL_INFO(dev_priv)->display.trans_offsets[TRANSCODER_A] + \ +#define _MMIO_TRANS2(tran, reg) _MMIO(DISPLAY_INFO(dev_priv)->trans_offsets[(tran)] - \ + DISPLAY_INFO(dev_priv)->trans_offsets[TRANSCODER_A] + \ DISPLAY_MMIO_BASE(dev_priv) + (reg)) -#define _MMIO_CURSOR2(pipe, reg) _MMIO(INTEL_INFO(dev_priv)->display.cursor_offsets[(pipe)] - \ - INTEL_INFO(dev_priv)->display.cursor_offsets[PIPE_A] + \ +#define _MMIO_CURSOR2(pipe, reg) _MMIO(DISPLAY_INFO(dev_priv)->cursor_offsets[(pipe)] - \ + DISPLAY_INFO(dev_priv)->cursor_offsets[PIPE_A] + \ DISPLAY_MMIO_BASE(dev_priv) + (reg)) #endif /* __INTEL_DISPLAY_REG_DEFS_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_display_reset.c b/drivers/gpu/drm/i915/display/intel_display_reset.c new file mode 100644 index 0000000000000000000000000000000000000000..17178d5d7788a97fd0be7d215c0358e81bb4daa0 --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_display_reset.c @@ -0,0 +1,135 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2023 Intel Corporation + */ + +#include + +#include "i915_drv.h" +#include "intel_clock_gating.h" +#include "intel_display_driver.h" +#include "intel_display_reset.h" +#include "intel_display_types.h" +#include "intel_hotplug.h" +#include "intel_pps.h" + +static bool gpu_reset_clobbers_display(struct drm_i915_private *dev_priv) +{ + return (INTEL_INFO(dev_priv)->gpu_reset_clobbers_display && + intel_has_gpu_reset(to_gt(dev_priv))); +} + +void intel_display_reset_prepare(struct drm_i915_private *dev_priv) +{ + struct drm_modeset_acquire_ctx *ctx = &dev_priv->display.restore.reset_ctx; + struct drm_atomic_state *state; + int ret; + + if (!HAS_DISPLAY(dev_priv)) + return; + + /* reset doesn't touch the display */ + if (!dev_priv->params.force_reset_modeset_test && + !gpu_reset_clobbers_display(dev_priv)) + return; + + /* We have a modeset vs reset deadlock, defensively unbreak it. */ + set_bit(I915_RESET_MODESET, &to_gt(dev_priv)->reset.flags); + smp_mb__after_atomic(); + wake_up_bit(&to_gt(dev_priv)->reset.flags, I915_RESET_MODESET); + + if (atomic_read(&dev_priv->gpu_error.pending_fb_pin)) { + drm_dbg_kms(&dev_priv->drm, + "Modeset potentially stuck, unbreaking through wedging\n"); + intel_gt_set_wedged(to_gt(dev_priv)); + } + + /* + * Need mode_config.mutex so that we don't + * trample ongoing ->detect() and whatnot. + */ + mutex_lock(&dev_priv->drm.mode_config.mutex); + drm_modeset_acquire_init(ctx, 0); + while (1) { + ret = drm_modeset_lock_all_ctx(&dev_priv->drm, ctx); + if (ret != -EDEADLK) + break; + + drm_modeset_backoff(ctx); + } + /* + * Disabling the crtcs gracefully seems nicer. Also the + * g33 docs say we should at least disable all the planes. + */ + state = drm_atomic_helper_duplicate_state(&dev_priv->drm, ctx); + if (IS_ERR(state)) { + ret = PTR_ERR(state); + drm_err(&dev_priv->drm, "Duplicating state failed with %i\n", + ret); + return; + } + + ret = drm_atomic_helper_disable_all(&dev_priv->drm, ctx); + if (ret) { + drm_err(&dev_priv->drm, "Suspending crtc's failed with %i\n", + ret); + drm_atomic_state_put(state); + return; + } + + dev_priv->display.restore.modeset_state = state; + state->acquire_ctx = ctx; +} + +void intel_display_reset_finish(struct drm_i915_private *i915) +{ + struct drm_modeset_acquire_ctx *ctx = &i915->display.restore.reset_ctx; + struct drm_atomic_state *state; + int ret; + + if (!HAS_DISPLAY(i915)) + return; + + /* reset doesn't touch the display */ + if (!test_bit(I915_RESET_MODESET, &to_gt(i915)->reset.flags)) + return; + + state = fetch_and_zero(&i915->display.restore.modeset_state); + if (!state) + goto unlock; + + /* reset doesn't touch the display */ + if (!gpu_reset_clobbers_display(i915)) { + /* for testing only restore the display */ + ret = drm_atomic_helper_commit_duplicated_state(state, ctx); + if (ret) { + drm_WARN_ON(&i915->drm, ret == -EDEADLK); + drm_err(&i915->drm, + "Restoring old state failed with %i\n", ret); + } + } else { + /* + * The display has been reset as well, + * so need a full re-initialization. + */ + intel_pps_unlock_regs_wa(i915); + intel_display_driver_init_hw(i915); + intel_clock_gating_init(i915); + intel_hpd_init(i915); + + ret = __intel_display_driver_resume(i915, state, ctx); + if (ret) + drm_err(&i915->drm, + "Restoring old state failed with %i\n", ret); + + intel_hpd_poll_disable(i915); + } + + drm_atomic_state_put(state); +unlock: + drm_modeset_drop_locks(ctx); + drm_modeset_acquire_fini(ctx); + mutex_unlock(&i915->drm.mode_config.mutex); + + clear_bit_unlock(I915_RESET_MODESET, &to_gt(i915)->reset.flags); +} diff --git a/drivers/gpu/drm/i915/display/intel_display_reset.h b/drivers/gpu/drm/i915/display/intel_display_reset.h new file mode 100644 index 0000000000000000000000000000000000000000..f06d0d35b86b12670839789484e2581194aac1cb --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_display_reset.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2023 Intel Corporation + */ + +#ifndef __INTEL_RESET_H__ +#define __INTEL_RESET_H__ + +struct drm_i915_private; + +void intel_display_reset_prepare(struct drm_i915_private *i915); +void intel_display_reset_finish(struct drm_i915_private *i915); + +#endif /* __INTEL_RESET_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_display_trace.h b/drivers/gpu/drm/i915/display/intel_display_trace.h index 651ea8564e1b7d379abef7cf4028f93f4a14b471..99bdb833591ce1a44e14ae1df964d7166cfcf9e0 100644 --- a/drivers/gpu/drm/i915/display/intel_display_trace.h +++ b/drivers/gpu/drm/i915/display/intel_display_trace.h @@ -14,7 +14,6 @@ #include #include "i915_drv.h" -#include "i915_irq.h" #include "intel_crtc.h" #include "intel_display_types.h" #include "intel_vblank.h" diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h index 47395b39c8f49b2cf65e5189a1e3d39ac69a2645..731f2ec04d5cdac4e1041aed33297a063dc3c96a 100644 --- a/drivers/gpu/drm/i915/display/intel_display_types.h +++ b/drivers/gpu/drm/i915/display/intel_display_types.h @@ -233,13 +233,25 @@ struct intel_encoder { * Called during system suspend after all pending requests for the * encoder are flushed (for example for DP AUX transactions) and * device interrupts are disabled. + * All modeset locks are held while the hook is called. */ void (*suspend)(struct intel_encoder *); + /* + * Called without the modeset locks held after the suspend() hook for + * all encoders have been called. + */ + void (*suspend_complete)(struct intel_encoder *encoder); /* * Called during system reboot/shutdown after all the * encoders have been disabled and suspended. + * All modeset locks are held while the hook is called. */ void (*shutdown)(struct intel_encoder *encoder); + /* + * Called without the modeset locks held after the shutdown() hook for + * all encoders have been called. + */ + void (*shutdown_complete)(struct intel_encoder *encoder); /* * Enable/disable the clock to the port. */ @@ -643,6 +655,9 @@ struct intel_atomic_state { struct __intel_global_objs_state *global_objs; int num_global_objs; + /* Internal commit, as opposed to userspace/client initiated one */ + bool internal; + bool dpll_set, modeset; struct intel_shared_dpll_state shared_dpll[I915_NUM_PLLS]; @@ -980,6 +995,38 @@ struct intel_link_m_n { u32 link_n; }; +struct intel_csc_matrix { + u16 coeff[9]; + u16 preoff[3]; + u16 postoff[3]; +}; + +struct intel_c10pll_state { + u32 clock; /* in KHz */ + u8 tx; + u8 cmn; + u8 pll[20]; +}; + +struct intel_c20pll_state { + u32 link_bit_rate; + u32 clock; /* in kHz */ + u16 tx[3]; + u16 cmn[4]; + union { + u16 mplla[10]; + u16 mpllb[11]; + }; +}; + +struct intel_cx0pll_state { + union { + struct intel_c10pll_state c10; + struct intel_c20pll_state c20; + }; + bool ssc_enabled; +}; + struct intel_crtc_state { /* * uapi (drm) state. This is the software state shown to userspace. @@ -1021,6 +1068,8 @@ struct intel_crtc_state { /* actual state of LUTs */ struct drm_property_blob *pre_csc_lut, *post_csc_lut; + struct intel_csc_matrix csc, output_csc; + /** * quirks - bitfield with hw state readout quirks * @@ -1123,6 +1172,7 @@ struct intel_crtc_state { union { struct intel_dpll_hw_state dpll_hw_state; struct intel_mpllb_state mpllb_state; + struct intel_cx0pll_state cx0pll_state; }; /* @@ -1275,15 +1325,27 @@ struct intel_crtc_state { /* HDMI High TMDS char rate ratio */ bool hdmi_high_tmds_clock_ratio; - /* Output format RGB/YCBCR etc */ + /* + * Output format RGB/YCBCR etc., that is coming out + * at the end of the pipe. + */ enum intel_output_format output_format; + /* + * Sink output format RGB/YCBCR etc., that is going + * into the sink. + */ + enum intel_output_format sink_format; + /* enable pipe gamma? */ bool gamma_enable; /* enable pipe csc? */ bool csc_enable; + /* enable vlv/chv wgc csc? */ + bool wgc_enable; + /* big joiner pipe bitmask */ u8 bigjoiner_pipes; @@ -1524,8 +1586,6 @@ struct intel_hdmi { enum drm_dp_dual_mode_type type; int max_tmds_clock; } dp_dual_mode; - bool has_hdmi_sink; - bool has_audio; struct intel_connector *attached_connector; struct cec_notifier *cec_notifier; }; @@ -1645,8 +1705,6 @@ struct intel_dp { u8 lane_count; u8 sink_count; bool link_trained; - bool has_hdmi_sink; - bool has_audio; bool reset_link_params; bool use_max_params; u8 dpcd[DP_RECEIVER_CAP_SIZE]; @@ -1728,6 +1786,7 @@ struct intel_dp { int pcon_max_frl_bw; u8 max_bpc; bool ycbcr_444_to_420; + bool ycbcr420_passthrough; bool rgb_to_ycbcr; } dfp; @@ -1814,10 +1873,6 @@ struct intel_dp_mst_encoder { struct intel_connector *connector; }; -struct intel_load_detect_pipe { - struct drm_atomic_state *restore_state; -}; - static inline struct intel_encoder * intel_attached_encoder(struct intel_connector *connector) { diff --git a/drivers/gpu/drm/i915/display/intel_dkl_phy.c b/drivers/gpu/drm/i915/display/intel_dkl_phy.c index 57cc3edba0163e2aee6b8e022cc3f297d4a45d24..a001232ad445ea08befcd238cf77f412670835d8 100644 --- a/drivers/gpu/drm/i915/display/intel_dkl_phy.c +++ b/drivers/gpu/drm/i915/display/intel_dkl_phy.c @@ -11,6 +11,15 @@ #include "intel_dkl_phy.h" #include "intel_dkl_phy_regs.h" +/** + * intel_dkl_phy_init - initialize Dekel PHY + * @i915: i915 device instance + */ +void intel_dkl_phy_init(struct drm_i915_private *i915) +{ + spin_lock_init(&i915->display.dkl.phy_lock); +} + static void dkl_phy_set_hip_idx(struct drm_i915_private *i915, struct intel_dkl_phy_reg reg) { diff --git a/drivers/gpu/drm/i915/display/intel_dkl_phy.h b/drivers/gpu/drm/i915/display/intel_dkl_phy.h index 570ee36f9386f0df881edb4b6a208d25f4d7d100..5956ec3e940b1b846d6fa85cfbc70be14fdfb096 100644 --- a/drivers/gpu/drm/i915/display/intel_dkl_phy.h +++ b/drivers/gpu/drm/i915/display/intel_dkl_phy.h @@ -12,6 +12,7 @@ struct drm_i915_private; +void intel_dkl_phy_init(struct drm_i915_private *i915); u32 intel_dkl_phy_read(struct drm_i915_private *i915, struct intel_dkl_phy_reg reg); void diff --git a/drivers/gpu/drm/i915/display/intel_dmc.c b/drivers/gpu/drm/i915/display/intel_dmc.c index 8a88de67ff0a98f5b1350213d4471d987628f856..5f479f3828bbee25f7578627cb352f40cd01d794 100644 --- a/drivers/gpu/drm/i915/display/intel_dmc.c +++ b/drivers/gpu/drm/i915/display/intel_dmc.c @@ -1057,7 +1057,7 @@ void intel_dmc_init(struct drm_i915_private *i915) i915->display.dmc.dmc = dmc; drm_dbg_kms(&i915->drm, "Loading %s\n", dmc->fw_path); - schedule_work(&dmc->work); + queue_work(i915->unordered_wq, &dmc->work); return; diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index 529ee22be872e7aea648d85aa54b0269095bb278..9f40da20e88d28e6dca7fa3286f3b4d970a34002 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -44,8 +44,8 @@ #include #include "g4x_dp.h" -#include "i915_debugfs.h" #include "i915_drv.h" +#include "i915_irq.h" #include "i915_reg.h" #include "intel_atomic.h" #include "intel_audio.h" @@ -53,6 +53,7 @@ #include "intel_combo_phy_regs.h" #include "intel_connector.h" #include "intel_crtc.h" +#include "intel_cx0_phy.h" #include "intel_ddi.h" #include "intel_de.h" #include "intel_display_types.h" @@ -67,6 +68,7 @@ #include "intel_hdcp.h" #include "intel_hdmi.h" #include "intel_hotplug.h" +#include "intel_hotplug_irq.h" #include "intel_lspcon.h" #include "intel_lvds.h" #include "intel_panel.h" @@ -421,6 +423,18 @@ static int ehl_max_source_rate(struct intel_dp *intel_dp) return 810000; } +static int mtl_max_source_rate(struct intel_dp *intel_dp) +{ + struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); + struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev); + enum phy phy = intel_port_to_phy(i915, dig_port->base.port); + + if (intel_is_c10phy(i915, phy)) + return intel_dp_is_edp(intel_dp) ? 675000 : 810000; + + return 2000000; +} + static int vbt_max_link_rate(struct intel_dp *intel_dp) { struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; @@ -445,6 +459,10 @@ static void intel_dp_set_source_rates(struct intel_dp *intel_dp) { /* The values must be in increasing order */ + static const int mtl_rates[] = { + 162000, 216000, 243000, 270000, 324000, 432000, 540000, 675000, + 810000, 1000000, 1350000, 2000000, + }; static const int icl_rates[] = { 162000, 216000, 270000, 324000, 432000, 540000, 648000, 810000, 1000000, 1350000, @@ -470,7 +488,11 @@ intel_dp_set_source_rates(struct intel_dp *intel_dp) drm_WARN_ON(&dev_priv->drm, intel_dp->source_rates || intel_dp->num_source_rates); - if (DISPLAY_VER(dev_priv) >= 11) { + if (DISPLAY_VER(dev_priv) >= 14) { + source_rates = mtl_rates; + size = ARRAY_SIZE(mtl_rates); + max_rate = mtl_max_source_rate(intel_dp); + } else if (DISPLAY_VER(dev_priv) >= 11) { source_rates = icl_rates; size = ARRAY_SIZE(icl_rates); if (IS_DG2(dev_priv)) @@ -828,26 +850,88 @@ u8 intel_dp_dsc_get_slice_count(struct intel_dp *intel_dp, return 0; } +static bool source_can_output(struct intel_dp *intel_dp, + enum intel_output_format format) +{ + struct drm_i915_private *i915 = dp_to_i915(intel_dp); + + switch (format) { + case INTEL_OUTPUT_FORMAT_RGB: + return true; + + case INTEL_OUTPUT_FORMAT_YCBCR444: + /* + * No YCbCr output support on gmch platforms. + * Also, ILK doesn't seem capable of DP YCbCr output. + * The displayed image is severly corrupted. SNB+ is fine. + */ + return !HAS_GMCH(i915) && !IS_IRONLAKE(i915); + + case INTEL_OUTPUT_FORMAT_YCBCR420: + /* Platform < Gen 11 cannot output YCbCr420 format */ + return DISPLAY_VER(i915) >= 11; + + default: + MISSING_CASE(format); + return false; + } +} + +static bool +dfp_can_convert_from_rgb(struct intel_dp *intel_dp, + enum intel_output_format sink_format) +{ + if (!drm_dp_is_branch(intel_dp->dpcd)) + return false; + + if (sink_format == INTEL_OUTPUT_FORMAT_YCBCR444) + return intel_dp->dfp.rgb_to_ycbcr; + + if (sink_format == INTEL_OUTPUT_FORMAT_YCBCR420) + return intel_dp->dfp.rgb_to_ycbcr && + intel_dp->dfp.ycbcr_444_to_420; + + return false; +} + +static bool +dfp_can_convert_from_ycbcr444(struct intel_dp *intel_dp, + enum intel_output_format sink_format) +{ + if (!drm_dp_is_branch(intel_dp->dpcd)) + return false; + + if (sink_format == INTEL_OUTPUT_FORMAT_YCBCR420) + return intel_dp->dfp.ycbcr_444_to_420; + + return false; +} + static enum intel_output_format intel_dp_output_format(struct intel_connector *connector, - bool ycbcr_420_output) + enum intel_output_format sink_format) { struct intel_dp *intel_dp = intel_attached_dp(connector); + struct drm_i915_private *i915 = dp_to_i915(intel_dp); + enum intel_output_format output_format; if (intel_dp->force_dsc_output_format) return intel_dp->force_dsc_output_format; - if (!connector->base.ycbcr_420_allowed || !ycbcr_420_output) - return INTEL_OUTPUT_FORMAT_RGB; + if (sink_format == INTEL_OUTPUT_FORMAT_RGB || + dfp_can_convert_from_rgb(intel_dp, sink_format)) + output_format = INTEL_OUTPUT_FORMAT_RGB; - if (intel_dp->dfp.rgb_to_ycbcr && - intel_dp->dfp.ycbcr_444_to_420) - return INTEL_OUTPUT_FORMAT_RGB; + else if (sink_format == INTEL_OUTPUT_FORMAT_YCBCR444 || + dfp_can_convert_from_ycbcr444(intel_dp, sink_format)) + output_format = INTEL_OUTPUT_FORMAT_YCBCR444; - if (intel_dp->dfp.ycbcr_444_to_420) - return INTEL_OUTPUT_FORMAT_YCBCR444; else - return INTEL_OUTPUT_FORMAT_YCBCR420; + output_format = INTEL_OUTPUT_FORMAT_YCBCR420; + + drm_WARN_ON(&i915->drm, !source_can_output(intel_dp, output_format)); + + return output_format; } int intel_dp_min_bpp(enum intel_output_format output_format) @@ -871,13 +955,27 @@ static int intel_dp_output_bpp(enum intel_output_format output_format, int bpp) return bpp; } +static enum intel_output_format +intel_dp_sink_format(struct intel_connector *connector, + const struct drm_display_mode *mode) +{ + const struct drm_display_info *info = &connector->base.display_info; + + if (drm_mode_is_420_only(info, mode)) + return INTEL_OUTPUT_FORMAT_YCBCR420; + + return INTEL_OUTPUT_FORMAT_RGB; +} + static int intel_dp_mode_min_output_bpp(struct intel_connector *connector, const struct drm_display_mode *mode) { - const struct drm_display_info *info = &connector->base.display_info; - enum intel_output_format output_format = - intel_dp_output_format(connector, drm_mode_is_420_only(info, mode)); + enum intel_output_format output_format, sink_format; + + sink_format = intel_dp_sink_format(connector, mode); + + output_format = intel_dp_output_format(connector, sink_format); return intel_dp_output_bpp(output_format, intel_dp_min_bpp(output_format)); } @@ -916,7 +1014,8 @@ static int intel_dp_max_tmds_clock(struct intel_dp *intel_dp) static enum drm_mode_status intel_dp_tmds_clock_valid(struct intel_dp *intel_dp, - int clock, int bpc, bool ycbcr420_output, + int clock, int bpc, + enum intel_output_format sink_format, bool respect_downstream_limits) { int tmds_clock, min_tmds_clock, max_tmds_clock; @@ -924,7 +1023,7 @@ intel_dp_tmds_clock_valid(struct intel_dp *intel_dp, if (!respect_downstream_limits) return MODE_OK; - tmds_clock = intel_hdmi_tmds_clock(clock, bpc, ycbcr420_output); + tmds_clock = intel_hdmi_tmds_clock(clock, bpc, sink_format); min_tmds_clock = intel_dp->dfp.min_tmds_clock; max_tmds_clock = intel_dp_max_tmds_clock(intel_dp); @@ -946,7 +1045,7 @@ intel_dp_mode_valid_downstream(struct intel_connector *connector, struct intel_dp *intel_dp = intel_attached_dp(connector); const struct drm_display_info *info = &connector->base.display_info; enum drm_mode_status status; - bool ycbcr_420_only; + enum intel_output_format sink_format; /* If PCON supports FRL MODE, check FRL bandwidth constraints */ if (intel_dp->dfp.pcon_max_frl_bw) { @@ -971,20 +1070,20 @@ intel_dp_mode_valid_downstream(struct intel_connector *connector, target_clock > intel_dp->dfp.max_dotclock) return MODE_CLOCK_HIGH; - ycbcr_420_only = drm_mode_is_420_only(info, mode); + sink_format = intel_dp_sink_format(connector, mode); /* Assume 8bpc for the DP++/HDMI/DVI TMDS clock check */ status = intel_dp_tmds_clock_valid(intel_dp, target_clock, - 8, ycbcr_420_only, true); + 8, sink_format, true); if (status != MODE_OK) { - if (ycbcr_420_only || + if (sink_format == INTEL_OUTPUT_FORMAT_YCBCR420 || !connector->base.ycbcr_420_allowed || !drm_mode_is_420_also(info, mode)) return status; - + sink_format = INTEL_OUTPUT_FORMAT_YCBCR420; status = intel_dp_tmds_clock_valid(intel_dp, target_clock, - 8, true, true); + 8, sink_format, true); if (status != MODE_OK) return status; } @@ -1041,6 +1140,9 @@ intel_dp_mode_valid(struct drm_connector *_connector, if (target_clock > max_dotclk) return MODE_CLOCK_HIGH; + if (intel_dp_hdisplay_bad(dev_priv, mode->hdisplay)) + return MODE_H_ILLEGAL; + max_link_clock = intel_dp_max_link_rate(intel_dp); max_lanes = intel_dp_max_lane_count(intel_dp); @@ -1048,13 +1150,6 @@ intel_dp_mode_valid(struct drm_connector *_connector, mode_rate = intel_dp_link_required(target_clock, intel_dp_mode_min_output_bpp(connector, mode)); - if (intel_dp_hdisplay_bad(dev_priv, mode->hdisplay)) - return MODE_H_ILLEGAL; - - /* - * Output bpp is stored in 6.4 format so right shift by 4 to get the - * integer value since we support only integer values of bpp. - */ if (HAS_DSC(dev_priv) && drm_dp_sink_supports_dsc(intel_dp->dsc_dpcd)) { /* @@ -1063,6 +1158,10 @@ intel_dp_mode_valid(struct drm_connector *_connector, */ int pipe_bpp = intel_dp_dsc_compute_bpp(intel_dp, U8_MAX); + /* + * Output bpp is stored in 6.4 format so right shift by 4 to get the + * integer value since we support only integer values of bpp. + */ if (intel_dp_is_edp(intel_dp)) { dsc_max_output_bpp = drm_edp_dsc_sink_output_bpp(intel_dp->dsc_dpcd) >> 4; @@ -1188,6 +1287,13 @@ void intel_dp_compute_rate(struct intel_dp *intel_dp, int port_clock, } } +bool intel_dp_has_hdmi_sink(struct intel_dp *intel_dp) +{ + struct intel_connector *connector = intel_dp->attached_connector; + + return connector->base.display_info.is_hdmi; +} + static bool intel_dp_source_supports_fec(struct intel_dp *intel_dp, const struct intel_crtc_state *pipe_config) { @@ -1220,19 +1326,10 @@ static bool intel_dp_supports_dsc(struct intel_dp *intel_dp, drm_dp_sink_supports_dsc(intel_dp->dsc_dpcd); } -static bool intel_dp_is_ycbcr420(struct intel_dp *intel_dp, - const struct intel_crtc_state *crtc_state) -{ - return crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR420 || - (crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR444 && - intel_dp->dfp.ycbcr_444_to_420); -} - static int intel_dp_hdmi_compute_bpc(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state, int bpc, bool respect_downstream_limits) { - bool ycbcr420_output = intel_dp_is_ycbcr420(intel_dp, crtc_state); int clock = crtc_state->hw.adjusted_mode.crtc_clock; /* @@ -1252,8 +1349,8 @@ static int intel_dp_hdmi_compute_bpc(struct intel_dp *intel_dp, for (; bpc >= 8; bpc -= 2) { if (intel_hdmi_bpc_possible(crtc_state, bpc, - intel_dp->has_hdmi_sink, ycbcr420_output) && - intel_dp_tmds_clock_valid(intel_dp, clock, bpc, ycbcr420_output, + intel_dp_has_hdmi_sink(intel_dp)) && + intel_dp_tmds_clock_valid(intel_dp, clock, bpc, crtc_state->sink_format, respect_downstream_limits) == MODE_OK) return bpc; } @@ -2069,6 +2166,7 @@ static bool intel_dp_has_audio(struct intel_encoder *encoder, { struct drm_i915_private *i915 = to_i915(encoder->base.dev); struct intel_dp *intel_dp = enc_to_intel_dp(encoder); + struct intel_connector *connector = intel_dp->attached_connector; const struct intel_digital_connector_state *intel_conn_state = to_intel_digital_connector_state(conn_state); @@ -2076,7 +2174,7 @@ static bool intel_dp_has_audio(struct intel_encoder *encoder, return false; if (intel_conn_state->force_audio == HDMI_AUDIO_AUTO) - return intel_dp->has_audio; + return connector->base.display_info.has_audio; else return intel_conn_state->force_audio == HDMI_AUDIO_ON; } @@ -2097,23 +2195,27 @@ intel_dp_compute_output_format(struct intel_encoder *encoder, ycbcr_420_only = drm_mode_is_420_only(info, adjusted_mode); - crtc_state->output_format = intel_dp_output_format(connector, ycbcr_420_only); - - if (ycbcr_420_only && !intel_dp_is_ycbcr420(intel_dp, crtc_state)) { + if (ycbcr_420_only && !connector->base.ycbcr_420_allowed) { drm_dbg_kms(&i915->drm, "YCbCr 4:2:0 mode but YCbCr 4:2:0 output not possible. Falling back to RGB.\n"); - crtc_state->output_format = INTEL_OUTPUT_FORMAT_RGB; + crtc_state->sink_format = INTEL_OUTPUT_FORMAT_RGB; + } else { + crtc_state->sink_format = intel_dp_sink_format(connector, adjusted_mode); } + crtc_state->output_format = intel_dp_output_format(connector, crtc_state->sink_format); + ret = intel_dp_compute_link_config(encoder, crtc_state, conn_state, respect_downstream_limits); if (ret) { - if (intel_dp_is_ycbcr420(intel_dp, crtc_state) || + if (crtc_state->sink_format == INTEL_OUTPUT_FORMAT_YCBCR420 || !connector->base.ycbcr_420_allowed || !drm_mode_is_420_also(info, adjusted_mode)) return ret; - crtc_state->output_format = intel_dp_output_format(connector, true); + crtc_state->sink_format = INTEL_OUTPUT_FORMAT_YCBCR420; + crtc_state->output_format = intel_dp_output_format(connector, + crtc_state->sink_format); ret = intel_dp_compute_link_config(encoder, crtc_state, conn_state, respect_downstream_limits); } @@ -2637,7 +2739,7 @@ static int intel_dp_pcon_start_frl_training(struct intel_dp *intel_dp) static bool intel_dp_is_hdmi_2_1_sink(struct intel_dp *intel_dp) { if (drm_dp_is_branch(intel_dp->dpcd) && - intel_dp->has_hdmi_sink && + intel_dp_has_hdmi_sink(intel_dp) && intel_dp_hdmi_sink_max_frl(intel_dp) > 0) return true; @@ -2795,6 +2897,8 @@ void intel_dp_configure_protocol_converter(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state) { struct drm_i915_private *i915 = dp_to_i915(intel_dp); + bool ycbcr444_to_420 = false; + bool rgb_to_ycbcr = false; u8 tmp; if (intel_dp->dpcd[DP_DPCD_REV] < 0x13) @@ -2803,16 +2907,42 @@ void intel_dp_configure_protocol_converter(struct intel_dp *intel_dp, if (!drm_dp_is_branch(intel_dp->dpcd)) return; - tmp = intel_dp->has_hdmi_sink ? - DP_HDMI_DVI_OUTPUT_CONFIG : 0; + tmp = intel_dp_has_hdmi_sink(intel_dp) ? DP_HDMI_DVI_OUTPUT_CONFIG : 0; if (drm_dp_dpcd_writeb(&intel_dp->aux, DP_PROTOCOL_CONVERTER_CONTROL_0, tmp) != 1) drm_dbg_kms(&i915->drm, "Failed to %s protocol converter HDMI mode\n", - str_enable_disable(intel_dp->has_hdmi_sink)); + str_enable_disable(intel_dp_has_hdmi_sink(intel_dp))); - tmp = crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR444 && - intel_dp->dfp.ycbcr_444_to_420 ? DP_CONVERSION_TO_YCBCR420_ENABLE : 0; + if (crtc_state->sink_format == INTEL_OUTPUT_FORMAT_YCBCR420) { + switch (crtc_state->output_format) { + case INTEL_OUTPUT_FORMAT_YCBCR420: + break; + case INTEL_OUTPUT_FORMAT_YCBCR444: + ycbcr444_to_420 = true; + break; + case INTEL_OUTPUT_FORMAT_RGB: + rgb_to_ycbcr = true; + ycbcr444_to_420 = true; + break; + default: + MISSING_CASE(crtc_state->output_format); + break; + } + } else if (crtc_state->sink_format == INTEL_OUTPUT_FORMAT_YCBCR444) { + switch (crtc_state->output_format) { + case INTEL_OUTPUT_FORMAT_YCBCR444: + break; + case INTEL_OUTPUT_FORMAT_RGB: + rgb_to_ycbcr = true; + break; + default: + MISSING_CASE(crtc_state->output_format); + break; + } + } + + tmp = ycbcr444_to_420 ? DP_CONVERSION_TO_YCBCR420_ENABLE : 0; if (drm_dp_dpcd_writeb(&intel_dp->aux, DP_PROTOCOL_CONVERTER_CONTROL_1, tmp) != 1) @@ -2820,13 +2950,12 @@ void intel_dp_configure_protocol_converter(struct intel_dp *intel_dp, "Failed to %s protocol converter YCbCr 4:2:0 conversion mode\n", str_enable_disable(intel_dp->dfp.ycbcr_444_to_420)); - tmp = intel_dp->dfp.rgb_to_ycbcr ? - DP_CONVERSION_BT709_RGB_YCBCR_ENABLE : 0; + tmp = rgb_to_ycbcr ? DP_CONVERSION_BT709_RGB_YCBCR_ENABLE : 0; if (drm_dp_pcon_convert_rgb_to_ycbcr(&intel_dp->aux, tmp) < 0) drm_dbg_kms(&i915->drm, - "Failed to %s protocol converter RGB->YCbCr conversion mode\n", - str_enable_disable(tmp)); + "Failed to %s protocol converter RGB->YCbCr conversion mode\n", + str_enable_disable(tmp)); } bool intel_dp_get_colorimetry_status(struct intel_dp *intel_dp) @@ -3940,9 +4069,7 @@ intel_dp_mst_hpd_irq(struct intel_dp *intel_dp, u8 *esi, u8 *ack) { bool handled = false; - drm_dp_mst_hpd_irq(&intel_dp->mst_mgr, esi, &handled); - if (handled) - ack[1] |= esi[1] & (DP_DOWN_REP_MSG_RDY | DP_UP_REQ_MSG_RDY); + drm_dp_mst_hpd_irq_handle_event(&intel_dp->mst_mgr, esi, ack, &handled); if (esi[1] & DP_CP_IRQ) { intel_hdcp_handle_cp_irq(intel_dp->attached_connector); @@ -4017,6 +4144,9 @@ intel_dp_check_mst_status(struct intel_dp *intel_dp) if (!intel_dp_ack_sink_irq_esi(intel_dp, ack)) drm_dbg_kms(&i915->drm, "Failed to ack ESI\n"); + + if (ack[1] & (DP_DOWN_REP_MSG_RDY | DP_UP_REQ_MSG_RDY)) + drm_dp_mst_hpd_irq_send_new_request(&intel_dp->mst_mgr); } return link_ok; @@ -4110,9 +4240,9 @@ static bool intel_dp_has_connector(struct intel_dp *intel_dp, return false; } -static int intel_dp_prep_link_retrain(struct intel_dp *intel_dp, - struct drm_modeset_acquire_ctx *ctx, - u8 *pipe_mask) +int intel_dp_get_active_pipes(struct intel_dp *intel_dp, + struct drm_modeset_acquire_ctx *ctx, + u8 *pipe_mask) { struct drm_i915_private *i915 = dp_to_i915(intel_dp); struct drm_connector_list_iter conn_iter; @@ -4121,9 +4251,6 @@ static int intel_dp_prep_link_retrain(struct intel_dp *intel_dp, *pipe_mask = 0; - if (!intel_dp_needs_link_retrain(intel_dp)) - return 0; - drm_connector_list_iter_begin(&i915->drm, &conn_iter); for_each_intel_connector_iter(connector, &conn_iter) { struct drm_connector_state *conn_state = @@ -4157,9 +4284,6 @@ static int intel_dp_prep_link_retrain(struct intel_dp *intel_dp, } drm_connector_list_iter_end(&conn_iter); - if (!intel_dp_needs_link_retrain(intel_dp)) - *pipe_mask = 0; - return ret; } @@ -4188,13 +4312,19 @@ int intel_dp_retrain_link(struct intel_encoder *encoder, if (ret) return ret; - ret = intel_dp_prep_link_retrain(intel_dp, ctx, &pipe_mask); + if (!intel_dp_needs_link_retrain(intel_dp)) + return 0; + + ret = intel_dp_get_active_pipes(intel_dp, ctx, &pipe_mask); if (ret) return ret; if (pipe_mask == 0) return 0; + if (!intel_dp_needs_link_retrain(intel_dp)) + return 0; + drm_dbg_kms(&dev_priv->drm, "[ENCODER:%d:%s] retraining link\n", encoder->base.base.id, encoder->base.name); @@ -4616,57 +4746,44 @@ intel_dp_update_dfp(struct intel_dp *intel_dp, intel_dp_get_pcon_dsc_cap(intel_dp); } +static bool +intel_dp_can_ycbcr420(struct intel_dp *intel_dp) +{ + if (source_can_output(intel_dp, INTEL_OUTPUT_FORMAT_YCBCR420) && + (!drm_dp_is_branch(intel_dp->dpcd) || intel_dp->dfp.ycbcr420_passthrough)) + return true; + + if (source_can_output(intel_dp, INTEL_OUTPUT_FORMAT_RGB) && + dfp_can_convert_from_rgb(intel_dp, INTEL_OUTPUT_FORMAT_YCBCR420)) + return true; + + if (source_can_output(intel_dp, INTEL_OUTPUT_FORMAT_YCBCR444) && + dfp_can_convert_from_ycbcr444(intel_dp, INTEL_OUTPUT_FORMAT_YCBCR420)) + return true; + + return false; +} + static void intel_dp_update_420(struct intel_dp *intel_dp) { struct drm_i915_private *i915 = dp_to_i915(intel_dp); struct intel_connector *connector = intel_dp->attached_connector; - bool is_branch, ycbcr_420_passthrough, ycbcr_444_to_420, rgb_to_ycbcr; - - /* No YCbCr output support on gmch platforms */ - if (HAS_GMCH(i915)) - return; - - /* - * ILK doesn't seem capable of DP YCbCr output. The - * displayed image is severly corrupted. SNB+ is fine. - */ - if (IS_IRONLAKE(i915)) - return; - is_branch = drm_dp_is_branch(intel_dp->dpcd); - ycbcr_420_passthrough = + intel_dp->dfp.ycbcr420_passthrough = drm_dp_downstream_420_passthrough(intel_dp->dpcd, intel_dp->downstream_ports); /* on-board LSPCON always assumed to support 4:4:4->4:2:0 conversion */ - ycbcr_444_to_420 = + intel_dp->dfp.ycbcr_444_to_420 = dp_to_dig_port(intel_dp)->lspcon.active || drm_dp_downstream_444_to_420_conversion(intel_dp->dpcd, intel_dp->downstream_ports); - rgb_to_ycbcr = drm_dp_downstream_rgb_to_ycbcr_conversion(intel_dp->dpcd, - intel_dp->downstream_ports, - DP_DS_HDMI_BT709_RGB_YCBCR_CONV); - - if (DISPLAY_VER(i915) >= 11) { - /* Let PCON convert from RGB->YCbCr if possible */ - if (is_branch && rgb_to_ycbcr && ycbcr_444_to_420) { - intel_dp->dfp.rgb_to_ycbcr = true; - intel_dp->dfp.ycbcr_444_to_420 = true; - connector->base.ycbcr_420_allowed = true; - } else { - /* Prefer 4:2:0 passthrough over 4:4:4->4:2:0 conversion */ - intel_dp->dfp.ycbcr_444_to_420 = - ycbcr_444_to_420 && !ycbcr_420_passthrough; - - connector->base.ycbcr_420_allowed = - !is_branch || ycbcr_444_to_420 || ycbcr_420_passthrough; - } - } else { - /* 4:4:4->4:2:0 conversion is the only way */ - intel_dp->dfp.ycbcr_444_to_420 = ycbcr_444_to_420; + intel_dp->dfp.rgb_to_ycbcr = + drm_dp_downstream_rgb_to_ycbcr_conversion(intel_dp->dpcd, + intel_dp->downstream_ports, + DP_DS_HDMI_BT709_RGB_YCBCR_CONV); - connector->base.ycbcr_420_allowed = ycbcr_444_to_420; - } + connector->base.ycbcr_420_allowed = intel_dp_can_ycbcr420(intel_dp); drm_dbg_kms(&i915->drm, "[CONNECTOR:%d:%s] RGB->YcbCr conversion? %s, YCbCr 4:2:0 allowed? %s, YCbCr 4:4:4->4:2:0 conversion? %s\n", @@ -4702,10 +4819,6 @@ intel_dp_set_edid(struct intel_dp *intel_dp) /* FIXME: Get rid of drm_edid_raw() */ edid = drm_edid_raw(drm_edid); - if (edid && edid->input & DRM_EDID_INPUT_DIGITAL) { - intel_dp->has_hdmi_sink = drm_detect_hdmi_monitor(edid); - intel_dp->has_audio = drm_detect_monitor_audio(edid); - } drm_dp_cec_set_edid(&intel_dp->aux, edid); } @@ -4719,9 +4832,6 @@ intel_dp_unset_edid(struct intel_dp *intel_dp) drm_edid_free(connector->detect_edid); connector->detect_edid = NULL; - intel_dp->has_hdmi_sink = false; - intel_dp->has_audio = false; - intel_dp->dfp.max_bpc = 0; intel_dp->dfp.max_dotclock = 0; intel_dp->dfp.min_tmds_clock = 0; @@ -5142,7 +5252,7 @@ static void intel_dp_oob_hotplug_event(struct drm_connector *connector) spin_lock_irq(&i915->irq_lock); i915->display.hotplug.event_bits |= BIT(encoder->hpd_pin); spin_unlock_irq(&i915->irq_lock); - queue_delayed_work(system_wq, &i915->display.hotplug.hotplug_work, 0); + queue_delayed_work(i915->unordered_wq, &i915->display.hotplug.hotplug_work, 0); } static const struct drm_connector_funcs intel_dp_connector_funcs = { @@ -5370,6 +5480,15 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp, goto out_vdd_off; } + /* + * Enable HPD sense for live status check. + * intel_hpd_irq_setup() will turn it off again + * if it's no longer needed later. + * + * The DPCD probe below will make sure VDD is on. + */ + intel_hpd_enable_detection(encoder); + /* Cache DPCD and EDID for edp. */ has_dpcd = intel_edp_init_dpcd(intel_dp); @@ -5381,6 +5500,24 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp, goto out_vdd_off; } + /* + * VBT and straps are liars. Also check HPD as that seems + * to be the most reliable piece of information available. + */ + if (!intel_digital_port_connected(encoder)) { + /* + * If this fails, presume the DPCD answer came + * from some other port using the same AUX CH. + * + * FIXME maybe cleaner to check this before the + * DPCD read? Would need sort out the VDD handling... + */ + drm_info(&dev_priv->drm, + "[ENCODER:%d:%s] HPD is down, disabling eDP\n", + encoder->base.base.id, encoder->base.name); + goto out_vdd_off; + } + mutex_lock(&dev_priv->drm.mode_config.mutex); drm_edid = drm_edid_read_ddc(connector, &intel_dp->aux.ddc); if (!drm_edid) { @@ -5583,6 +5720,7 @@ intel_dp_init_connector(struct intel_digital_port *dig_port, return true; fail: + intel_display_power_flush_work(dev_priv); drm_connector_cleanup(connector); return false; diff --git a/drivers/gpu/drm/i915/display/intel_dp.h b/drivers/gpu/drm/i915/display/intel_dp.h index ef39e4f7a329e1c337524621510af39a17a3fd77..22099de3ca458b0bfd931b243666a25cbd4a0b9e 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.h +++ b/drivers/gpu/drm/i915/display/intel_dp.h @@ -42,6 +42,9 @@ void intel_dp_set_link_params(struct intel_dp *intel_dp, int link_rate, int lane_count); int intel_dp_get_link_train_fallback_values(struct intel_dp *intel_dp, int link_rate, u8 lane_count); +int intel_dp_get_active_pipes(struct intel_dp *intel_dp, + struct drm_modeset_acquire_ctx *ctx, + u8 *pipe_mask); int intel_dp_retrain_link(struct intel_encoder *encoder, struct drm_modeset_acquire_ctx *ctx); void intel_dp_set_power(struct intel_dp *intel_dp, u8 mode); @@ -62,6 +65,7 @@ int intel_dp_dsc_compute_config(struct intel_dp *intel_dp, struct link_config_limits *limits, int timeslots, bool recompute_pipe_bpp); +bool intel_dp_has_hdmi_sink(struct intel_dp *intel_dp); bool intel_dp_is_edp(struct intel_dp *intel_dp); bool intel_dp_is_uhbr(const struct intel_crtc_state *crtc_state); bool intel_dp_is_port_edp(struct drm_i915_private *dev_priv, enum port port); diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux.c b/drivers/gpu/drm/i915/display/intel_dp_aux.c index 705915d505652553179a832a11427fd6c06ec08f..197c6e81db147e23cadb3f34cb0820cd0c185d4a 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_aux.c +++ b/drivers/gpu/drm/i915/display/intel_dp_aux.c @@ -129,7 +129,7 @@ static int intel_dp_aux_sync_len(void) static int intel_dp_aux_fw_sync_len(void) { - int precharge = 16; /* 10-16 */ + int precharge = 10; /* 10-16 */ int preamble = 8; return precharge + preamble; @@ -161,14 +161,14 @@ static u32 g4x_get_aux_send_ctl(struct intel_dp *intel_dp, timeout = DP_AUX_CH_CTL_TIME_OUT_400us; return DP_AUX_CH_CTL_SEND_BUSY | - DP_AUX_CH_CTL_DONE | - DP_AUX_CH_CTL_INTERRUPT | - DP_AUX_CH_CTL_TIME_OUT_ERROR | - timeout | - DP_AUX_CH_CTL_RECEIVE_ERROR | - (send_bytes << DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT) | - (g4x_dp_aux_precharge_len() << DP_AUX_CH_CTL_PRECHARGE_2US_SHIFT) | - (aux_clock_divider << DP_AUX_CH_CTL_BIT_CLOCK_2X_SHIFT); + DP_AUX_CH_CTL_DONE | + DP_AUX_CH_CTL_INTERRUPT | + DP_AUX_CH_CTL_TIME_OUT_ERROR | + timeout | + DP_AUX_CH_CTL_RECEIVE_ERROR | + DP_AUX_CH_CTL_MESSAGE_SIZE(send_bytes) | + DP_AUX_CH_CTL_PRECHARGE_2US(g4x_dp_aux_precharge_len()) | + DP_AUX_CH_CTL_BIT_CLOCK_2X(aux_clock_divider); } static u32 skl_get_aux_send_ctl(struct intel_dp *intel_dp, @@ -185,14 +185,14 @@ static u32 skl_get_aux_send_ctl(struct intel_dp *intel_dp, * ICL+: 4ms */ ret = DP_AUX_CH_CTL_SEND_BUSY | - DP_AUX_CH_CTL_DONE | - DP_AUX_CH_CTL_INTERRUPT | - DP_AUX_CH_CTL_TIME_OUT_ERROR | - DP_AUX_CH_CTL_TIME_OUT_MAX | - DP_AUX_CH_CTL_RECEIVE_ERROR | - (send_bytes << DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT) | - DP_AUX_CH_CTL_FW_SYNC_PULSE_SKL(intel_dp_aux_fw_sync_len()) | - DP_AUX_CH_CTL_SYNC_PULSE_SKL(intel_dp_aux_sync_len()); + DP_AUX_CH_CTL_DONE | + DP_AUX_CH_CTL_INTERRUPT | + DP_AUX_CH_CTL_TIME_OUT_ERROR | + DP_AUX_CH_CTL_TIME_OUT_MAX | + DP_AUX_CH_CTL_RECEIVE_ERROR | + DP_AUX_CH_CTL_MESSAGE_SIZE(send_bytes) | + DP_AUX_CH_CTL_FW_SYNC_PULSE_SKL(intel_dp_aux_fw_sync_len()) | + DP_AUX_CH_CTL_SYNC_PULSE_SKL(intel_dp_aux_sync_len()); if (intel_tc_port_in_tbt_alt_mode(dig_port)) ret |= DP_AUX_CH_CTL_TBT_IO; @@ -268,6 +268,11 @@ intel_dp_aux_xfer(struct intel_dp *intel_dp, intel_pps_check_power_unlocked(intel_dp); + /* + * FIXME PSR should be disabled here to prevent + * it using the same AUX CH simultaneously + */ + /* Try to wait for any previous AUX channel activity */ for (try = 0; try < 3; try++) { status = intel_de_read_notrace(i915, ch_ctl); @@ -373,8 +378,7 @@ intel_dp_aux_xfer(struct intel_dp *intel_dp, } /* Unload any bytes sent back from the other side */ - recv_bytes = ((status & DP_AUX_CH_CTL_MESSAGE_SIZE_MASK) >> - DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT); + recv_bytes = REG_FIELD_GET(DP_AUX_CH_CTL_MESSAGE_SIZE_MASK, status); /* * By BSpec: "Message sizes of 0 or >20 are not allowed." @@ -810,3 +814,8 @@ enum aux_ch intel_dp_aux_ch(struct intel_encoder *encoder) return aux_ch; } + +void intel_dp_aux_irq_handler(struct drm_i915_private *i915) +{ + wake_up_all(&i915->display.gmbus.wait_queue); +} diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux.h b/drivers/gpu/drm/i915/display/intel_dp_aux.h index 138e340f94ee06a64aa07e8ca882cf32b6bd0d25..5b608f9d34993be06b991eb7772dd68112dd6393 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_aux.h +++ b/drivers/gpu/drm/i915/display/intel_dp_aux.h @@ -7,6 +7,7 @@ #define __INTEL_DP_AUX_H__ enum aux_ch; +struct drm_i915_private; struct intel_dp; struct intel_encoder; @@ -15,4 +16,6 @@ void intel_dp_aux_init(struct intel_dp *intel_dp); enum aux_ch intel_dp_aux_ch(struct intel_encoder *encoder); +void intel_dp_aux_irq_handler(struct drm_i915_private *i915); + #endif /* __INTEL_DP_AUX_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux_regs.h b/drivers/gpu/drm/i915/display/intel_dp_aux_regs.h index 5702f318d5378f3b555cc7be12bff6bca936f976..5185345277c7c2e67c017d3d91e5dec74db5bb07 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_aux_regs.h +++ b/drivers/gpu/drm/i915/display/intel_dp_aux_regs.h @@ -50,35 +50,37 @@ _XELPDP_USBC3_AUX_CH_DATA1, \ _XELPDP_USBC4_AUX_CH_DATA1) + (i) * 4) -#define DP_AUX_CH_CTL_SEND_BUSY (1 << 31) -#define DP_AUX_CH_CTL_DONE (1 << 30) -#define DP_AUX_CH_CTL_INTERRUPT (1 << 29) -#define DP_AUX_CH_CTL_TIME_OUT_ERROR (1 << 28) -#define DP_AUX_CH_CTL_TIME_OUT_400us (0 << 26) -#define DP_AUX_CH_CTL_TIME_OUT_600us (1 << 26) -#define DP_AUX_CH_CTL_TIME_OUT_800us (2 << 26) -#define DP_AUX_CH_CTL_TIME_OUT_MAX (3 << 26) /* Varies per platform */ -#define DP_AUX_CH_CTL_TIME_OUT_MASK (3 << 26) -#define DP_AUX_CH_CTL_RECEIVE_ERROR (1 << 25) -#define DP_AUX_CH_CTL_MESSAGE_SIZE_MASK (0x1f << 20) -#define DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT 20 -#define XELPDP_DP_AUX_CH_CTL_POWER_REQUEST REG_BIT(19) -#define XELPDP_DP_AUX_CH_CTL_POWER_STATUS REG_BIT(18) -#define DP_AUX_CH_CTL_PRECHARGE_2US_MASK (0xf << 16) -#define DP_AUX_CH_CTL_PRECHARGE_2US_SHIFT 16 -#define DP_AUX_CH_CTL_AUX_AKSV_SELECT (1 << 15) -#define DP_AUX_CH_CTL_MANCHESTER_TEST (1 << 14) -#define DP_AUX_CH_CTL_SYNC_TEST (1 << 13) -#define DP_AUX_CH_CTL_DEGLITCH_TEST (1 << 12) -#define DP_AUX_CH_CTL_PRECHARGE_TEST (1 << 11) -#define DP_AUX_CH_CTL_BIT_CLOCK_2X_MASK (0x7ff) -#define DP_AUX_CH_CTL_BIT_CLOCK_2X_SHIFT 0 -#define DP_AUX_CH_CTL_PSR_DATA_AUX_REG_SKL (1 << 14) -#define DP_AUX_CH_CTL_FS_DATA_AUX_REG_SKL (1 << 13) -#define DP_AUX_CH_CTL_GTC_DATA_AUX_REG_SKL (1 << 12) -#define DP_AUX_CH_CTL_TBT_IO (1 << 11) -#define DP_AUX_CH_CTL_FW_SYNC_PULSE_SKL_MASK (0x1f << 5) -#define DP_AUX_CH_CTL_FW_SYNC_PULSE_SKL(c) (((c) - 1) << 5) -#define DP_AUX_CH_CTL_SYNC_PULSE_SKL(c) ((c) - 1) +#define DP_AUX_CH_CTL_SEND_BUSY REG_BIT(31) +#define DP_AUX_CH_CTL_DONE REG_BIT(30) +#define DP_AUX_CH_CTL_INTERRUPT REG_BIT(29) +#define DP_AUX_CH_CTL_TIME_OUT_ERROR REG_BIT(28) + +#define DP_AUX_CH_CTL_TIME_OUT_MASK REG_GENMASK(27, 26) +#define DP_AUX_CH_CTL_TIME_OUT_400us REG_FIELD_PREP(DP_AUX_CH_CTL_TIME_OUT_MASK, 0) +#define DP_AUX_CH_CTL_TIME_OUT_600us REG_FIELD_PREP(DP_AUX_CH_CTL_TIME_OUT_MASK, 1) +#define DP_AUX_CH_CTL_TIME_OUT_800us REG_FIELD_PREP(DP_AUX_CH_CTL_TIME_OUT_MASK, 2) +#define DP_AUX_CH_CTL_TIME_OUT_MAX REG_FIELD_PREP(DP_AUX_CH_CTL_TIME_OUT_MASK, 3) /* Varies per platform */ +#define DP_AUX_CH_CTL_RECEIVE_ERROR REG_BIT(25) +#define DP_AUX_CH_CTL_MESSAGE_SIZE_MASK REG_GENMASK(24, 20) +#define DP_AUX_CH_CTL_MESSAGE_SIZE(x) REG_FIELD_PREP(DP_AUX_CH_CTL_MESSAGE_SIZE_MASK, (x)) +#define DP_AUX_CH_CTL_PRECHARGE_2US_MASK REG_GENMASK(19, 16) /* pre-skl */ +#define DP_AUX_CH_CTL_PRECHARGE_2US(x) REG_FIELD_PREP(DP_AUX_CH_CTL_PRECHARGE_2US_MASK, (x)) +#define XELPDP_DP_AUX_CH_CTL_POWER_REQUEST REG_BIT(19) /* mtl+ */ +#define XELPDP_DP_AUX_CH_CTL_POWER_STATUS REG_BIT(18) /* mtl+ */ +#define DP_AUX_CH_CTL_AUX_AKSV_SELECT REG_BIT(15) +#define DP_AUX_CH_CTL_MANCHESTER_TEST REG_BIT(14) /* pre-hsw */ +#define DP_AUX_CH_CTL_PSR_DATA_AUX_REG_SKL REG_BIT(14) /* skl+ */ +#define DP_AUX_CH_CTL_SYNC_TEST REG_BIT(13) /* pre-hsw */ +#define DP_AUX_CH_CTL_FS_DATA_AUX_REG_SKL REG_BIT(13) /* skl+ */ +#define DP_AUX_CH_CTL_DEGLITCH_TEST REG_BIT(12) /* pre-hsw */ +#define DP_AUX_CH_CTL_GTC_DATA_AUX_REG_SKL REG_BIT(12) /* skl+ */ +#define DP_AUX_CH_CTL_PRECHARGE_TEST REG_BIT(11) /* pre-hsw */ +#define DP_AUX_CH_CTL_TBT_IO REG_BIT(11) /* icl+ */ +#define DP_AUX_CH_CTL_BIT_CLOCK_2X_MASK REG_GENMASK(10, 0) /* pre-skl */ +#define DP_AUX_CH_CTL_BIT_CLOCK_2X(x) REG_FIELD_PREP(DP_AUX_CH_CTL_BIT_CLOCK_2X_MASK, (x)) +#define DP_AUX_CH_CTL_FW_SYNC_PULSE_SKL_MASK REG_GENMASK(9, 5) /* skl+ */ +#define DP_AUX_CH_CTL_FW_SYNC_PULSE_SKL(c) REG_FIELD_PREP(DP_AUX_CH_CTL_FW_SYNC_PULSE_SKL_MASK, (c) - 1) +#define DP_AUX_CH_CTL_SYNC_PULSE_SKL_MASK REG_GENMASK(4, 0) /* skl+ */ +#define DP_AUX_CH_CTL_SYNC_PULSE_SKL(c) REG_FIELD_PREP(DP_AUX_CH_CTL_SYNC_PULSE_SKL_MASK, (c) - 1) #endif /* __INTEL_DP_AUX_REGS_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c index d638054c74ac182cf9d05414a197f9b2857175cc..a263773f4d68a4564fcef94afd3108768a7e9008 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c +++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c @@ -26,6 +26,27 @@ #include "intel_dp.h" #include "intel_dp_link_training.h" +#define LT_MSG_PREFIX "[CONNECTOR:%d:%s][ENCODER:%d:%s][%s] " +#define LT_MSG_ARGS(_intel_dp, _dp_phy) (_intel_dp)->attached_connector->base.base.id, \ + (_intel_dp)->attached_connector->base.name, \ + dp_to_dig_port(_intel_dp)->base.base.base.id, \ + dp_to_dig_port(_intel_dp)->base.base.name, \ + drm_dp_phy_name(_dp_phy) + +#define lt_dbg(_intel_dp, _dp_phy, _format, ...) \ + drm_dbg_kms(&dp_to_i915(_intel_dp)->drm, \ + LT_MSG_PREFIX _format, \ + LT_MSG_ARGS(_intel_dp, _dp_phy), ## __VA_ARGS__) + +#define lt_err(_intel_dp, _dp_phy, _format, ...) do { \ + if (intel_digital_port_connected(&dp_to_dig_port(_intel_dp)->base)) \ + drm_err(&dp_to_i915(_intel_dp)->drm, \ + LT_MSG_PREFIX _format, \ + LT_MSG_ARGS(_intel_dp, _dp_phy), ## __VA_ARGS__); \ + else \ + lt_dbg(_intel_dp, _dp_phy, "Sink disconnected: " _format, ## __VA_ARGS__); \ +} while (0) + static void intel_dp_reset_lttpr_common_caps(struct intel_dp *intel_dp) { memset(intel_dp->lttpr_common_caps, 0, sizeof(intel_dp->lttpr_common_caps)); @@ -47,29 +68,21 @@ static void intel_dp_read_lttpr_phy_caps(struct intel_dp *intel_dp, const u8 dpcd[DP_RECEIVER_CAP_SIZE], enum drm_dp_phy dp_phy) { - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; u8 *phy_caps = intel_dp_lttpr_phy_caps(intel_dp, dp_phy); if (drm_dp_read_lttpr_phy_caps(&intel_dp->aux, dpcd, dp_phy, phy_caps) < 0) { - drm_dbg_kms(&dp_to_i915(intel_dp)->drm, - "[ENCODER:%d:%s][%s] failed to read the PHY caps\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy)); + lt_dbg(intel_dp, dp_phy, "failed to read the PHY caps\n"); return; } - drm_dbg_kms(&dp_to_i915(intel_dp)->drm, - "[ENCODER:%d:%s][%s] PHY capabilities: %*ph\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy), - (int)sizeof(intel_dp->lttpr_phy_caps[0]), - phy_caps); + lt_dbg(intel_dp, dp_phy, "PHY capabilities: %*ph\n", + (int)sizeof(intel_dp->lttpr_phy_caps[0]), + phy_caps); } static bool intel_dp_read_lttpr_common_caps(struct intel_dp *intel_dp, const u8 dpcd[DP_RECEIVER_CAP_SIZE]) { - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; int ret; ret = drm_dp_read_lttpr_common_caps(&intel_dp->aux, dpcd, @@ -77,11 +90,9 @@ static bool intel_dp_read_lttpr_common_caps(struct intel_dp *intel_dp, if (ret < 0) goto reset_caps; - drm_dbg_kms(&dp_to_i915(intel_dp)->drm, - "[ENCODER:%d:%s] LTTPR common capabilities: %*ph\n", - encoder->base.base.id, encoder->base.name, - (int)sizeof(intel_dp->lttpr_common_caps), - intel_dp->lttpr_common_caps); + lt_dbg(intel_dp, DP_PHY_DPRX, "LTTPR common capabilities: %*ph\n", + (int)sizeof(intel_dp->lttpr_common_caps), + intel_dp->lttpr_common_caps); /* The minimum value of LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV is 1.4 */ if (intel_dp->lttpr_common_caps[0] < 0x14) @@ -105,8 +116,6 @@ intel_dp_set_lttpr_transparent_mode(struct intel_dp *intel_dp, bool enable) static int intel_dp_init_lttpr(struct intel_dp *intel_dp, const u8 dpcd[DP_RECEIVER_CAP_SIZE]) { - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; - struct drm_i915_private *i915 = to_i915(encoder->base.dev); int lttpr_count; int i; @@ -138,9 +147,8 @@ static int intel_dp_init_lttpr(struct intel_dp *intel_dp, const u8 dpcd[DP_RECEI return 0; if (!intel_dp_set_lttpr_transparent_mode(intel_dp, false)) { - drm_dbg_kms(&i915->drm, - "[ENCODER:%d:%s] Switching to LTTPR non-transparent LT mode failed, fall-back to transparent mode\n", - encoder->base.base.id, encoder->base.name); + lt_dbg(intel_dp, DP_PHY_DPRX, + "Switching to LTTPR non-transparent LT mode failed, fall-back to transparent mode\n"); intel_dp_set_lttpr_transparent_mode(intel_dp, true); intel_dp_reset_lttpr_count(intel_dp); @@ -409,26 +417,22 @@ intel_dp_get_adjust_train(struct intel_dp *intel_dp, enum drm_dp_phy dp_phy, const u8 link_status[DP_LINK_STATUS_SIZE]) { - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; - struct drm_i915_private *i915 = to_i915(encoder->base.dev); int lane; if (intel_dp_is_uhbr(crtc_state)) { - drm_dbg_kms(&i915->drm, "[ENCODER:%d:%s][%s] 128b/132b, lanes: %d, " - "TX FFE request: " TRAIN_REQ_FMT "\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy), - crtc_state->lane_count, - TRAIN_REQ_TX_FFE_ARGS(link_status)); + lt_dbg(intel_dp, dp_phy, + "128b/132b, lanes: %d, " + "TX FFE request: " TRAIN_REQ_FMT "\n", + crtc_state->lane_count, + TRAIN_REQ_TX_FFE_ARGS(link_status)); } else { - drm_dbg_kms(&i915->drm, "[ENCODER:%d:%s][%s] 8b/10b, lanes: %d, " - "vswing request: " TRAIN_REQ_FMT ", " - "pre-emphasis request: " TRAIN_REQ_FMT "\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy), - crtc_state->lane_count, - TRAIN_REQ_VSWING_ARGS(link_status), - TRAIN_REQ_PREEMPH_ARGS(link_status)); + lt_dbg(intel_dp, dp_phy, + "8b/10b, lanes: %d, " + "vswing request: " TRAIN_REQ_FMT ", " + "pre-emphasis request: " TRAIN_REQ_FMT "\n", + crtc_state->lane_count, + TRAIN_REQ_VSWING_ARGS(link_status), + TRAIN_REQ_PREEMPH_ARGS(link_status)); } for (lane = 0; lane < 4; lane++) @@ -487,16 +491,11 @@ intel_dp_program_link_training_pattern(struct intel_dp *intel_dp, enum drm_dp_phy dp_phy, u8 dp_train_pat) { - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; - struct drm_i915_private *i915 = to_i915(encoder->base.dev); u8 train_pat = intel_dp_training_pattern_symbol(dp_train_pat); if (train_pat != DP_TRAINING_PATTERN_DISABLE) - drm_dbg_kms(&i915->drm, - "[ENCODER:%d:%s][%s] Using DP training pattern TPS%c\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy), - dp_training_pattern_name(train_pat)); + lt_dbg(intel_dp, dp_phy, "Using DP training pattern TPS%c\n", + dp_training_pattern_name(train_pat)); intel_dp->set_link_train(intel_dp, crtc_state, dp_train_pat); } @@ -531,24 +530,21 @@ void intel_dp_set_signal_levels(struct intel_dp *intel_dp, enum drm_dp_phy dp_phy) { struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; - struct drm_i915_private *i915 = to_i915(encoder->base.dev); if (intel_dp_is_uhbr(crtc_state)) { - drm_dbg_kms(&i915->drm, "[ENCODER:%d:%s][%s] 128b/132b, lanes: %d, " - "TX FFE presets: " TRAIN_SET_FMT "\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy), - crtc_state->lane_count, - TRAIN_SET_TX_FFE_ARGS(intel_dp->train_set)); + lt_dbg(intel_dp, dp_phy, + "128b/132b, lanes: %d, " + "TX FFE presets: " TRAIN_SET_FMT "\n", + crtc_state->lane_count, + TRAIN_SET_TX_FFE_ARGS(intel_dp->train_set)); } else { - drm_dbg_kms(&i915->drm, "[ENCODER:%d:%s][%s] 8b/10b, lanes: %d, " - "vswing levels: " TRAIN_SET_FMT ", " - "pre-emphasis levels: " TRAIN_SET_FMT "\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy), - crtc_state->lane_count, - TRAIN_SET_VSWING_ARGS(intel_dp->train_set), - TRAIN_SET_PREEMPH_ARGS(intel_dp->train_set)); + lt_dbg(intel_dp, dp_phy, + "8b/10b, lanes: %d, " + "vswing levels: " TRAIN_SET_FMT ", " + "pre-emphasis levels: " TRAIN_SET_FMT "\n", + crtc_state->lane_count, + TRAIN_SET_VSWING_ARGS(intel_dp->train_set), + TRAIN_SET_PREEMPH_ARGS(intel_dp->train_set)); } if (intel_dp_phy_is_downstream_of_source(intel_dp, dp_phy)) @@ -637,6 +633,38 @@ static bool intel_dp_link_max_vswing_reached(struct intel_dp *intel_dp, return true; } +static void +intel_dp_update_downspread_ctrl(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state) +{ + u8 link_config[2]; + + link_config[0] = crtc_state->vrr.flipline ? DP_MSA_TIMING_PAR_IGNORE_EN : 0; + link_config[1] = intel_dp_is_uhbr(crtc_state) ? + DP_SET_ANSI_128B132B : DP_SET_ANSI_8B10B; + drm_dp_dpcd_write(&intel_dp->aux, DP_DOWNSPREAD_CTRL, link_config, 2); +} + +static void +intel_dp_update_link_bw_set(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state, + u8 link_bw, u8 rate_select) +{ + u8 link_config[2]; + + /* Write the link configuration data */ + link_config[0] = link_bw; + link_config[1] = crtc_state->lane_count; + if (drm_dp_enhanced_frame_cap(intel_dp->dpcd)) + link_config[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN; + drm_dp_dpcd_write(&intel_dp->aux, DP_LINK_BW_SET, link_config, 2); + + /* eDP 1.4 rate select method. */ + if (!link_bw) + drm_dp_dpcd_write(&intel_dp->aux, DP_LINK_RATE_SET, + &rate_select, 1); +} + /* * Prepare link training by configuring the link parameters. On DDI platforms * also enable the port here. @@ -645,9 +673,6 @@ static bool intel_dp_prepare_link_train(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state) { - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; - struct drm_i915_private *i915 = to_i915(encoder->base.dev); - u8 link_config[2]; u8 link_bw, rate_select; if (intel_dp->prepare_link_retrain) @@ -668,41 +693,28 @@ intel_dp_prepare_link_train(struct intel_dp *intel_dp, * link rates are not stable. */ if (!link_bw) { - struct intel_connector *connector = intel_dp->attached_connector; __le16 sink_rates[DP_MAX_SUPPORTED_RATES]; - drm_dbg_kms(&i915->drm, "[CONNECTOR:%d:%s] Reloading eDP link rates\n", - connector->base.base.id, connector->base.name); + lt_dbg(intel_dp, DP_PHY_DPRX, "Reloading eDP link rates\n"); drm_dp_dpcd_read(&intel_dp->aux, DP_SUPPORTED_LINK_RATES, sink_rates, sizeof(sink_rates)); } if (link_bw) - drm_dbg_kms(&i915->drm, - "[ENCODER:%d:%s] Using LINK_BW_SET value %02x\n", - encoder->base.base.id, encoder->base.name, link_bw); + lt_dbg(intel_dp, DP_PHY_DPRX, "Using LINK_BW_SET value %02x\n", + link_bw); else - drm_dbg_kms(&i915->drm, - "[ENCODER:%d:%s] Using LINK_RATE_SET value %02x\n", - encoder->base.base.id, encoder->base.name, rate_select); - - /* Write the link configuration data */ - link_config[0] = link_bw; - link_config[1] = crtc_state->lane_count; - if (drm_dp_enhanced_frame_cap(intel_dp->dpcd)) - link_config[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN; - drm_dp_dpcd_write(&intel_dp->aux, DP_LINK_BW_SET, link_config, 2); - - /* eDP 1.4 rate select method. */ - if (!link_bw) - drm_dp_dpcd_write(&intel_dp->aux, DP_LINK_RATE_SET, - &rate_select, 1); - - link_config[0] = crtc_state->vrr.enable ? DP_MSA_TIMING_PAR_IGNORE_EN : 0; - link_config[1] = intel_dp_is_uhbr(crtc_state) ? - DP_SET_ANSI_128B132B : DP_SET_ANSI_8B10B; - drm_dp_dpcd_write(&intel_dp->aux, DP_DOWNSPREAD_CTRL, link_config, 2); + lt_dbg(intel_dp, DP_PHY_DPRX, + "Using LINK_RATE_SET value %02x\n", + rate_select); + /* + * Spec DP2.1 Section 3.5.2.16 + * Prior to LT DPTX should set 128b/132b DP Channel coding and then set link rate + */ + intel_dp_update_downspread_ctrl(intel_dp, crtc_state); + intel_dp_update_link_bw_set(intel_dp, crtc_state, link_bw, + rate_select); return true; } @@ -737,15 +749,10 @@ void intel_dp_dump_link_status(struct intel_dp *intel_dp, enum drm_dp_phy dp_phy, const u8 link_status[DP_LINK_STATUS_SIZE]) { - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; - struct drm_i915_private *i915 = to_i915(encoder->base.dev); - - drm_dbg_kms(&i915->drm, - "[ENCODER:%d:%s][%s] ln0_1:0x%x ln2_3:0x%x align:0x%x sink:0x%x adj_req0_1:0x%x adj_req2_3:0x%x\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy), - link_status[0], link_status[1], link_status[2], - link_status[3], link_status[4], link_status[5]); + lt_dbg(intel_dp, dp_phy, + "ln0_1:0x%x ln2_3:0x%x align:0x%x sink:0x%x adj_req0_1:0x%x adj_req2_3:0x%x\n", + link_status[0], link_status[1], link_status[2], + link_status[3], link_status[4], link_status[5]); } /* @@ -757,8 +764,6 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state, enum drm_dp_phy dp_phy) { - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; - struct drm_i915_private *i915 = to_i915(encoder->base.dev); u8 old_link_status[DP_LINK_STATUS_SIZE] = {}; int voltage_tries, cr_tries, max_cr_tries; u8 link_status[DP_LINK_STATUS_SIZE]; @@ -773,9 +778,7 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp, if (!intel_dp_reset_link_train(intel_dp, crtc_state, dp_phy, DP_TRAINING_PATTERN_1 | DP_LINK_SCRAMBLING_DISABLE)) { - drm_err(&i915->drm, "[ENCODER:%d:%s][%s] Failed to enable link training\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy)); + lt_err(intel_dp, dp_phy, "Failed to enable link training\n"); return false; } @@ -798,35 +801,24 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp, if (drm_dp_dpcd_read_phy_link_status(&intel_dp->aux, dp_phy, link_status) < 0) { - drm_err(&i915->drm, "[ENCODER:%d:%s][%s] Failed to get link status\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy)); + lt_err(intel_dp, dp_phy, "Failed to get link status\n"); return false; } if (drm_dp_clock_recovery_ok(link_status, crtc_state->lane_count)) { - drm_dbg_kms(&i915->drm, - "[ENCODER:%d:%s][%s] Clock recovery OK\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy)); + lt_dbg(intel_dp, dp_phy, "Clock recovery OK\n"); return true; } if (voltage_tries == 5) { intel_dp_dump_link_status(intel_dp, dp_phy, link_status); - drm_dbg_kms(&i915->drm, - "[ENCODER:%d:%s][%s] Same voltage tried 5 times\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy)); + lt_dbg(intel_dp, dp_phy, "Same voltage tried 5 times\n"); return false; } if (max_vswing_reached) { intel_dp_dump_link_status(intel_dp, dp_phy, link_status); - drm_dbg_kms(&i915->drm, - "[ENCODER:%d:%s][%s] Max Voltage Swing reached\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy)); + lt_dbg(intel_dp, dp_phy, "Max Voltage Swing reached\n"); return false; } @@ -834,10 +826,7 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp, intel_dp_get_adjust_train(intel_dp, crtc_state, dp_phy, link_status); if (!intel_dp_update_link_train(intel_dp, crtc_state, dp_phy)) { - drm_err(&i915->drm, - "[ENCODER:%d:%s][%s] Failed to update link training\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy)); + lt_err(intel_dp, dp_phy, "Failed to update link training\n"); return false; } @@ -853,10 +842,8 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp, } intel_dp_dump_link_status(intel_dp, dp_phy, link_status); - drm_err(&i915->drm, - "[ENCODER:%d:%s][%s] Failed clock recovery %d times, giving up!\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy), max_cr_tries); + lt_err(intel_dp, dp_phy, "Failed clock recovery %d times, giving up!\n", + max_cr_tries); return false; } @@ -890,11 +877,11 @@ static u32 intel_dp_training_pattern(struct intel_dp *intel_dp, return DP_TRAINING_PATTERN_4; } else if (crtc_state->port_clock == 810000) { if (!source_tps4) - drm_dbg_kms(&i915->drm, - "8.1 Gbps link rate without source TPS4 support\n"); + lt_dbg(intel_dp, dp_phy, + "8.1 Gbps link rate without source TPS4 support\n"); if (!sink_tps4) - drm_dbg_kms(&i915->drm, - "8.1 Gbps link rate without sink TPS4 support\n"); + lt_dbg(intel_dp, dp_phy, + "8.1 Gbps link rate without sink TPS4 support\n"); } /* @@ -908,11 +895,11 @@ static u32 intel_dp_training_pattern(struct intel_dp *intel_dp, return DP_TRAINING_PATTERN_3; } else if (crtc_state->port_clock >= 540000) { if (!source_tps3) - drm_dbg_kms(&i915->drm, - ">=5.4/6.48 Gbps link rate without source TPS3 support\n"); + lt_dbg(intel_dp, dp_phy, + ">=5.4/6.48 Gbps link rate without source TPS3 support\n"); if (!sink_tps3) - drm_dbg_kms(&i915->drm, - ">=5.4/6.48 Gbps link rate without sink TPS3 support\n"); + lt_dbg(intel_dp, dp_phy, + ">=5.4/6.48 Gbps link rate without sink TPS3 support\n"); } return DP_TRAINING_PATTERN_2; @@ -928,8 +915,6 @@ intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state, enum drm_dp_phy dp_phy) { - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; - struct drm_i915_private *i915 = to_i915(encoder->base.dev); int tries; u32 training_pattern; u8 link_status[DP_LINK_STATUS_SIZE]; @@ -948,10 +933,7 @@ intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp, /* channel equalization */ if (!intel_dp_set_link_train(intel_dp, crtc_state, dp_phy, training_pattern)) { - drm_err(&i915->drm, - "[ENCODER:%d:%s][%s] Failed to start channel equalization\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy)); + lt_err(intel_dp, dp_phy, "Failed to start channel equalization\n"); return false; } @@ -960,10 +942,7 @@ intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp, if (drm_dp_dpcd_read_phy_link_status(&intel_dp->aux, dp_phy, link_status) < 0) { - drm_err(&i915->drm, - "[ENCODER:%d:%s][%s] Failed to get link status\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy)); + lt_err(intel_dp, dp_phy, "Failed to get link status\n"); break; } @@ -971,21 +950,15 @@ intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp, if (!drm_dp_clock_recovery_ok(link_status, crtc_state->lane_count)) { intel_dp_dump_link_status(intel_dp, dp_phy, link_status); - drm_dbg_kms(&i915->drm, - "[ENCODER:%d:%s][%s] Clock recovery check failed, cannot " - "continue channel equalization\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy)); + lt_dbg(intel_dp, dp_phy, + "Clock recovery check failed, cannot continue channel equalization\n"); break; } if (drm_dp_channel_eq_ok(link_status, crtc_state->lane_count)) { channel_eq = true; - drm_dbg_kms(&i915->drm, - "[ENCODER:%d:%s][%s] Channel EQ done. DP Training successful\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy)); + lt_dbg(intel_dp, dp_phy, "Channel EQ done. DP Training successful\n"); break; } @@ -993,10 +966,7 @@ intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp, intel_dp_get_adjust_train(intel_dp, crtc_state, dp_phy, link_status); if (!intel_dp_update_link_train(intel_dp, crtc_state, dp_phy)) { - drm_err(&i915->drm, - "[ENCODER:%d:%s][%s] Failed to update link training\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy)); + lt_err(intel_dp, dp_phy, "Failed to update link training\n"); break; } } @@ -1004,10 +974,7 @@ intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp, /* Try 5 times, else fail and try at lower BW */ if (tries == 5) { intel_dp_dump_link_status(intel_dp, dp_phy, link_status); - drm_dbg_kms(&i915->drm, - "[ENCODER:%d:%s][%s] Channel equalization failed 5 times\n", - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy)); + lt_dbg(intel_dp, dp_phy, "Channel equalization failed 5 times\n"); } return channel_eq; @@ -1026,13 +993,12 @@ static int intel_dp_128b132b_intra_hop(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state) { - struct drm_i915_private *i915 = dp_to_i915(intel_dp); u8 sink_status; int ret; ret = drm_dp_dpcd_readb(&intel_dp->aux, DP_SINK_STATUS, &sink_status); if (ret != 1) { - drm_dbg_kms(&i915->drm, "Failed to read sink status\n"); + lt_dbg(intel_dp, DP_PHY_DPRX, "Failed to read sink status\n"); return ret < 0 ? ret : -EIO; } @@ -1058,9 +1024,6 @@ intel_dp_128b132b_intra_hop(struct intel_dp *intel_dp, void intel_dp_stop_link_train(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state) { - struct drm_i915_private *i915 = dp_to_i915(intel_dp); - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; - intel_dp->link_trained = true; intel_dp_disable_dpcd_training_pattern(intel_dp, DP_PHY_DPRX); @@ -1069,9 +1032,7 @@ void intel_dp_stop_link_train(struct intel_dp *intel_dp, if (intel_dp_is_uhbr(crtc_state) && wait_for(intel_dp_128b132b_intra_hop(intel_dp, crtc_state) == 0, 500)) { - drm_dbg_kms(&i915->drm, - "[ENCODER:%d:%s] 128b/132b intra-hop not clearing\n", - encoder->base.base.id, encoder->base.name); + lt_dbg(intel_dp, DP_PHY_DPRX, "128b/132b intra-hop not clearing\n"); } } @@ -1080,8 +1041,6 @@ intel_dp_link_train_phy(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state, enum drm_dp_phy dp_phy) { - struct intel_connector *connector = intel_dp->attached_connector; - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; bool ret = false; if (!intel_dp_link_training_clock_recovery(intel_dp, crtc_state, dp_phy)) @@ -1093,13 +1052,10 @@ intel_dp_link_train_phy(struct intel_dp *intel_dp, ret = true; out: - drm_dbg_kms(&dp_to_i915(intel_dp)->drm, - "[CONNECTOR:%d:%s][ENCODER:%d:%s][%s] Link Training %s at link rate = %d, lane count = %d\n", - connector->base.base.id, connector->base.name, - encoder->base.base.id, encoder->base.name, - drm_dp_phy_name(dp_phy), - ret ? "passed" : "failed", - crtc_state->port_clock, crtc_state->lane_count); + lt_dbg(intel_dp, dp_phy, + "Link Training %s at link rate = %d, lane count = %d\n", + ret ? "passed" : "failed", + crtc_state->port_clock, crtc_state->lane_count); return ret; } @@ -1108,13 +1064,16 @@ static void intel_dp_schedule_fallback_link_training(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state) { struct intel_connector *intel_connector = intel_dp->attached_connector; - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; + struct drm_i915_private *i915 = dp_to_i915(intel_dp); + + if (!intel_digital_port_connected(&dp_to_dig_port(intel_dp)->base)) { + lt_dbg(intel_dp, DP_PHY_DPRX, "Link Training failed on disconnected sink.\n"); + return; + } if (intel_dp->hobl_active) { - drm_dbg_kms(&dp_to_i915(intel_dp)->drm, - "[ENCODER:%d:%s] Link Training failed with HOBL active, " - "not enabling it from now on", - encoder->base.base.id, encoder->base.name); + lt_dbg(intel_dp, DP_PHY_DPRX, + "Link Training failed with HOBL active, not enabling it from now on\n"); intel_dp->hobl_failed = true; } else if (intel_dp_get_link_train_fallback_values(intel_dp, crtc_state->port_clock, @@ -1123,7 +1082,7 @@ static void intel_dp_schedule_fallback_link_training(struct intel_dp *intel_dp, } /* Schedule a Hotplug Uevent to userspace to start modeset */ - schedule_work(&intel_connector->modeset_retry_work); + queue_work(i915->unordered_wq, &intel_connector->modeset_retry_work); } /* Perform the link training on all LTTPRs and the DPRX on a link. */ @@ -1161,8 +1120,6 @@ static bool intel_dp_128b132b_lane_eq(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state) { - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; - struct drm_i915_private *i915 = to_i915(encoder->base.dev); u8 link_status[DP_LINK_STATUS_SIZE]; int delay_us; int try, max_tries = 20; @@ -1177,9 +1134,7 @@ intel_dp_128b132b_lane_eq(struct intel_dp *intel_dp, */ if (!intel_dp_reset_link_train(intel_dp, crtc_state, DP_PHY_DPRX, DP_TRAINING_PATTERN_1)) { - drm_err(&i915->drm, - "[ENCODER:%d:%s] Failed to start 128b/132b TPS1\n", - encoder->base.base.id, encoder->base.name); + lt_err(intel_dp, DP_PHY_DPRX, "Failed to start 128b/132b TPS1\n"); return false; } @@ -1187,27 +1142,21 @@ intel_dp_128b132b_lane_eq(struct intel_dp *intel_dp, /* Read the initial TX FFE settings. */ if (drm_dp_dpcd_read_link_status(&intel_dp->aux, link_status) < 0) { - drm_err(&i915->drm, - "[ENCODER:%d:%s] Failed to read TX FFE presets\n", - encoder->base.base.id, encoder->base.name); + lt_err(intel_dp, DP_PHY_DPRX, "Failed to read TX FFE presets\n"); return false; } /* Update signal levels and training set as requested. */ intel_dp_get_adjust_train(intel_dp, crtc_state, DP_PHY_DPRX, link_status); if (!intel_dp_update_link_train(intel_dp, crtc_state, DP_PHY_DPRX)) { - drm_err(&i915->drm, - "[ENCODER:%d:%s] Failed to set initial TX FFE settings\n", - encoder->base.base.id, encoder->base.name); + lt_err(intel_dp, DP_PHY_DPRX, "Failed to set initial TX FFE settings\n"); return false; } /* Start transmitting 128b/132b TPS2. */ if (!intel_dp_set_link_train(intel_dp, crtc_state, DP_PHY_DPRX, DP_TRAINING_PATTERN_2)) { - drm_err(&i915->drm, - "[ENCODER:%d:%s] Failed to start 128b/132b TPS2\n", - encoder->base.base.id, encoder->base.name); + lt_err(intel_dp, DP_PHY_DPRX, "Failed to start 128b/132b TPS2\n"); return false; } @@ -1224,32 +1173,25 @@ intel_dp_128b132b_lane_eq(struct intel_dp *intel_dp, delay_us = drm_dp_128b132b_read_aux_rd_interval(&intel_dp->aux); if (drm_dp_dpcd_read_link_status(&intel_dp->aux, link_status) < 0) { - drm_err(&i915->drm, - "[ENCODER:%d:%s] Failed to read link status\n", - encoder->base.base.id, encoder->base.name); + lt_err(intel_dp, DP_PHY_DPRX, "Failed to read link status\n"); return false; } if (drm_dp_128b132b_link_training_failed(link_status)) { intel_dp_dump_link_status(intel_dp, DP_PHY_DPRX, link_status); - drm_err(&i915->drm, - "[ENCODER:%d:%s] Downstream link training failure\n", - encoder->base.base.id, encoder->base.name); + lt_err(intel_dp, DP_PHY_DPRX, + "Downstream link training failure\n"); return false; } if (drm_dp_128b132b_lane_channel_eq_done(link_status, crtc_state->lane_count)) { - drm_dbg_kms(&i915->drm, - "[ENCODER:%d:%s] Lane channel eq done\n", - encoder->base.base.id, encoder->base.name); + lt_dbg(intel_dp, DP_PHY_DPRX, "Lane channel eq done\n"); break; } if (timeout) { intel_dp_dump_link_status(intel_dp, DP_PHY_DPRX, link_status); - drm_err(&i915->drm, - "[ENCODER:%d:%s] Lane channel eq timeout\n", - encoder->base.base.id, encoder->base.name); + lt_err(intel_dp, DP_PHY_DPRX, "Lane channel eq timeout\n"); return false; } @@ -1259,18 +1201,14 @@ intel_dp_128b132b_lane_eq(struct intel_dp *intel_dp, /* Update signal levels and training set as requested. */ intel_dp_get_adjust_train(intel_dp, crtc_state, DP_PHY_DPRX, link_status); if (!intel_dp_update_link_train(intel_dp, crtc_state, DP_PHY_DPRX)) { - drm_err(&i915->drm, - "[ENCODER:%d:%s] Failed to update TX FFE settings\n", - encoder->base.base.id, encoder->base.name); + lt_err(intel_dp, DP_PHY_DPRX, "Failed to update TX FFE settings\n"); return false; } } if (try == max_tries) { intel_dp_dump_link_status(intel_dp, DP_PHY_DPRX, link_status); - drm_err(&i915->drm, - "[ENCODER:%d:%s] Max loop count reached\n", - encoder->base.base.id, encoder->base.name); + lt_err(intel_dp, DP_PHY_DPRX, "Max loop count reached\n"); return false; } @@ -1279,32 +1217,24 @@ intel_dp_128b132b_lane_eq(struct intel_dp *intel_dp, timeout = true; /* try one last time after deadline */ if (drm_dp_dpcd_read_link_status(&intel_dp->aux, link_status) < 0) { - drm_err(&i915->drm, - "[ENCODER:%d:%s] Failed to read link status\n", - encoder->base.base.id, encoder->base.name); + lt_err(intel_dp, DP_PHY_DPRX, "Failed to read link status\n"); return false; } if (drm_dp_128b132b_link_training_failed(link_status)) { intel_dp_dump_link_status(intel_dp, DP_PHY_DPRX, link_status); - drm_err(&i915->drm, - "[ENCODER:%d:%s] Downstream link training failure\n", - encoder->base.base.id, encoder->base.name); + lt_err(intel_dp, DP_PHY_DPRX, "Downstream link training failure\n"); return false; } if (drm_dp_128b132b_eq_interlane_align_done(link_status)) { - drm_dbg_kms(&i915->drm, - "[ENCODER:%d:%s] Interlane align done\n", - encoder->base.base.id, encoder->base.name); + lt_dbg(intel_dp, DP_PHY_DPRX, "Interlane align done\n"); break; } if (timeout) { intel_dp_dump_link_status(intel_dp, DP_PHY_DPRX, link_status); - drm_err(&i915->drm, - "[ENCODER:%d:%s] Interlane align timeout\n", - encoder->base.base.id, encoder->base.name); + lt_err(intel_dp, DP_PHY_DPRX, "Interlane align timeout\n"); return false; } @@ -1322,16 +1252,12 @@ intel_dp_128b132b_lane_cds(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state, int lttpr_count) { - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; - struct drm_i915_private *i915 = to_i915(encoder->base.dev); u8 link_status[DP_LINK_STATUS_SIZE]; unsigned long deadline; if (drm_dp_dpcd_writeb(&intel_dp->aux, DP_TRAINING_PATTERN_SET, DP_TRAINING_PATTERN_2_CDS) != 1) { - drm_err(&i915->drm, - "[ENCODER:%d:%s] Failed to start 128b/132b TPS2 CDS\n", - encoder->base.base.id, encoder->base.name); + lt_err(intel_dp, DP_PHY_DPRX, "Failed to start 128b/132b TPS2 CDS\n"); return false; } @@ -1347,34 +1273,26 @@ intel_dp_128b132b_lane_cds(struct intel_dp *intel_dp, usleep_range(2000, 3000); if (drm_dp_dpcd_read_link_status(&intel_dp->aux, link_status) < 0) { - drm_err(&i915->drm, - "[ENCODER:%d:%s] Failed to read link status\n", - encoder->base.base.id, encoder->base.name); + lt_err(intel_dp, DP_PHY_DPRX, "Failed to read link status\n"); return false; } if (drm_dp_128b132b_eq_interlane_align_done(link_status) && drm_dp_128b132b_cds_interlane_align_done(link_status) && drm_dp_128b132b_lane_symbol_locked(link_status, crtc_state->lane_count)) { - drm_dbg_kms(&i915->drm, - "[ENCODER:%d:%s] CDS interlane align done\n", - encoder->base.base.id, encoder->base.name); + lt_dbg(intel_dp, DP_PHY_DPRX, "CDS interlane align done\n"); break; } if (drm_dp_128b132b_link_training_failed(link_status)) { intel_dp_dump_link_status(intel_dp, DP_PHY_DPRX, link_status); - drm_err(&i915->drm, - "[ENCODER:%d:%s] Downstream link training failure\n", - encoder->base.base.id, encoder->base.name); + lt_err(intel_dp, DP_PHY_DPRX, "Downstream link training failure\n"); return false; } if (timeout) { intel_dp_dump_link_status(intel_dp, DP_PHY_DPRX, link_status); - drm_err(&i915->drm, - "[ENCODER:%d:%s] CDS timeout\n", - encoder->base.base.id, encoder->base.name); + lt_err(intel_dp, DP_PHY_DPRX, "CDS timeout\n"); return false; } } @@ -1390,15 +1308,10 @@ intel_dp_128b132b_link_train(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state, int lttpr_count) { - struct drm_i915_private *i915 = dp_to_i915(intel_dp); - struct intel_connector *connector = intel_dp->attached_connector; - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; bool passed = false; if (wait_for(intel_dp_128b132b_intra_hop(intel_dp, crtc_state) == 0, 500)) { - drm_err(&i915->drm, - "[ENCODER:%d:%s] 128b/132b intra-hop not clear\n", - encoder->base.base.id, encoder->base.name); + lt_err(intel_dp, DP_PHY_DPRX, "128b/132b intra-hop not clear\n"); return false; } @@ -1406,12 +1319,10 @@ intel_dp_128b132b_link_train(struct intel_dp *intel_dp, intel_dp_128b132b_lane_cds(intel_dp, crtc_state, lttpr_count)) passed = true; - drm_dbg_kms(&i915->drm, - "[CONNECTOR:%d:%s][ENCODER:%d:%s] 128b/132b Link Training %s at link rate = %d, lane count = %d\n", - connector->base.base.id, connector->base.name, - encoder->base.base.id, encoder->base.name, - passed ? "passed" : "failed", - crtc_state->port_clock, crtc_state->lane_count); + lt_dbg(intel_dp, DP_PHY_DPRX, + "128b/132b Link Training %s at link rate = %d, lane count = %d\n", + passed ? "passed" : "failed", + crtc_state->port_clock, crtc_state->lane_count); return passed; } @@ -1430,8 +1341,6 @@ void intel_dp_start_link_train(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state) { struct drm_i915_private *i915 = dp_to_i915(intel_dp); - struct intel_connector *connector = intel_dp->attached_connector; - struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; bool passed; /* @@ -1464,10 +1373,7 @@ void intel_dp_start_link_train(struct intel_dp *intel_dp, * ignore_long_hpd flag can unset from the testcase. */ if (!passed && i915->display.hotplug.ignore_long_hpd) { - drm_dbg_kms(&i915->drm, - "[CONNECTOR:%d:%s][ENCODER:%d:%s] Ignore the link failure\n", - connector->base.base.id, connector->base.name, - encoder->base.base.id, encoder->base.name); + lt_dbg(intel_dp, DP_PHY_DPRX, "Ignore the link failure\n"); return; } @@ -1478,8 +1384,6 @@ void intel_dp_start_link_train(struct intel_dp *intel_dp, void intel_dp_128b132b_sdp_crc16(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state) { - struct drm_i915_private *i915 = dp_to_i915(intel_dp); - /* * VIDEO_DIP_CTL register bit 31 should be set to '0' to not * disable SDP CRC. This is applicable for Display version 13. @@ -1492,5 +1396,5 @@ void intel_dp_128b132b_sdp_crc16(struct intel_dp *intel_dp, DP_SDP_ERROR_DETECTION_CONFIGURATION, DP_SDP_CRC16_128B132B_EN); - drm_dbg_kms(&i915->drm, "DP2.0 SDP CRC16 for 128b/132b enabled\n"); + lt_dbg(intel_dp, DP_PHY_DPRX, "DP2.0 SDP CRC16 for 128b/132b enabled\n"); } diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c index 2c49d9ab86a2a58bf8711a44052e83143fc9ff70..e3f176a093d2f34e315ace4f861198eae71952a2 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c @@ -298,7 +298,7 @@ static bool intel_dp_mst_has_audio(const struct drm_connector_state *conn_state) to_intel_connector(conn_state->connector); if (intel_conn_state->force_audio == HDMI_AUDIO_AUTO) - return connector->port->has_audio; + return connector->base.display_info.has_audio; else return intel_conn_state->force_audio == HDMI_AUDIO_ON; } @@ -318,6 +318,7 @@ static int intel_dp_mst_compute_config(struct intel_encoder *encoder, if (adjusted_mode->flags & DRM_MODE_FLAG_DBLSCAN) return -EINVAL; + pipe_config->sink_format = INTEL_OUTPUT_FORMAT_RGB; pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB; pipe_config->has_pch_encoder = false; @@ -800,9 +801,7 @@ static void intel_mst_enable_dp(struct intel_atomic_state *state, /* Enable hdcp if it's desired */ if (conn_state->content_protection == DRM_MODE_CONTENT_PROTECTION_DESIRED) - intel_hdcp_enable(to_intel_connector(conn_state->connector), - pipe_config, - (u8)conn_state->hdcp_content_type); + intel_hdcp_enable(state, encoder, pipe_config, conn_state); } static bool intel_dp_mst_enc_get_hw_state(struct intel_encoder *encoder, @@ -837,15 +836,17 @@ static int intel_dp_mst_get_ddc_modes(struct drm_connector *connector) { struct intel_connector *intel_connector = to_intel_connector(connector); struct intel_dp *intel_dp = intel_connector->mst_port; - struct edid *edid; + const struct drm_edid *drm_edid; int ret; if (drm_connector_is_unregistered(connector)) return intel_connector_update_modes(connector, NULL); - edid = drm_dp_mst_get_edid(connector, &intel_dp->mst_mgr, intel_connector->port); - ret = intel_connector_update_modes(connector, edid); - kfree(edid); + drm_edid = drm_dp_mst_edid_read(connector, &intel_dp->mst_mgr, intel_connector->port); + + ret = intel_connector_update_modes(connector, drm_edid); + + drm_edid_free(drm_edid); return ret; } diff --git a/drivers/gpu/drm/i915/display/intel_dpll.c b/drivers/gpu/drm/i915/display/intel_dpll.c index 4e9c18be7e1fd1f0b8d20d95424b8e200020c25a..999badfe2906df9aef0801453446adc2f646ad8d 100644 --- a/drivers/gpu/drm/i915/display/intel_dpll.c +++ b/drivers/gpu/drm/i915/display/intel_dpll.c @@ -8,6 +8,7 @@ #include "i915_reg.h" #include "intel_crtc.h" +#include "intel_cx0_phy.h" #include "intel_de.h" #include "intel_display.h" #include "intel_display_types.h" @@ -995,6 +996,32 @@ static int dg2_crtc_compute_clock(struct intel_atomic_state *state, return 0; } +static int mtl_crtc_compute_clock(struct intel_atomic_state *state, + struct intel_crtc *crtc) +{ + struct drm_i915_private *i915 = to_i915(state->base.dev); + struct intel_crtc_state *crtc_state = + intel_atomic_get_new_crtc_state(state, crtc); + struct intel_encoder *encoder = + intel_get_crtc_new_encoder(state, crtc_state); + enum phy phy = intel_port_to_phy(i915, encoder->port); + int ret; + + ret = intel_cx0pll_calc_state(crtc_state, encoder); + if (ret) + return ret; + + /* TODO: Do the readback via intel_compute_shared_dplls() */ + if (intel_is_c10phy(i915, phy)) + crtc_state->port_clock = intel_c10pll_calc_port_clock(encoder, &crtc_state->cx0pll_state.c10); + else + crtc_state->port_clock = intel_c20pll_calc_port_clock(encoder, &crtc_state->cx0pll_state.c20); + + crtc_state->hw.adjusted_mode.crtc_clock = intel_crtc_dotclock(crtc_state); + + return 0; +} + static bool ilk_needs_fb_cb_tune(const struct dpll *dpll, int factor) { return dpll->m < factor * dpll->n; @@ -1423,6 +1450,10 @@ static int i8xx_crtc_compute_clock(struct intel_atomic_state *state, return 0; } +static const struct intel_dpll_funcs mtl_dpll_funcs = { + .crtc_compute_clock = mtl_crtc_compute_clock, +}; + static const struct intel_dpll_funcs dg2_dpll_funcs = { .crtc_compute_clock = dg2_crtc_compute_clock, }; @@ -1517,7 +1548,9 @@ int intel_dpll_crtc_get_shared_dpll(struct intel_atomic_state *state, void intel_dpll_init_clock_hook(struct drm_i915_private *dev_priv) { - if (IS_DG2(dev_priv)) + if (DISPLAY_VER(dev_priv) >= 14) + dev_priv->display.funcs.dpll = &mtl_dpll_funcs; + else if (IS_DG2(dev_priv)) dev_priv->display.funcs.dpll = &dg2_dpll_funcs; else if (DISPLAY_VER(dev_priv) >= 9 || HAS_DDI(dev_priv)) dev_priv->display.funcs.dpll = &hsw_dpll_funcs; @@ -1760,13 +1793,11 @@ static void chv_prepare_pll(const struct intel_crtc_state *crtc_state) enum pipe pipe = crtc->pipe; enum dpio_channel port = vlv_pipe_to_channel(pipe); u32 loopfilter, tribuf_calcntr; - u32 bestn, bestm1, bestm2, bestp1, bestp2, bestm2_frac; + u32 bestm2, bestp1, bestp2, bestm2_frac; u32 dpio_val; int vco; - bestn = crtc_state->dpll.n; bestm2_frac = crtc_state->dpll.m2 & 0x3fffff; - bestm1 = crtc_state->dpll.m1; bestm2 = crtc_state->dpll.m2 >> 22; bestp1 = crtc_state->dpll.p1; bestp2 = crtc_state->dpll.p2; @@ -2047,7 +2078,7 @@ static void assert_pll(struct drm_i915_private *dev_priv, bool cur_state; cur_state = intel_de_read(dev_priv, DPLL(pipe)) & DPLL_VCO_ENABLE; - I915_STATE_WARN(cur_state != state, + I915_STATE_WARN(dev_priv, cur_state != state, "PLL state assertion failure (expected %s, current %s)\n", str_on_off(state), str_on_off(cur_state)); } diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c index 22fc908b7e5d6d306810e768884ada6a65cc00d2..6b2d8a1e2aa90bc598d44b25a798dd1e6632484a 100644 --- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c +++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c @@ -169,8 +169,8 @@ void assert_shared_dpll(struct drm_i915_private *dev_priv, return; cur_state = intel_dpll_get_hw_state(dev_priv, pll, &hw_state); - I915_STATE_WARN(cur_state != state, - "%s assertion failure (expected %s, current %s)\n", + I915_STATE_WARN(dev_priv, cur_state != state, + "%s assertion failure (expected %s, current %s)\n", pll->info->name, str_on_off(state), str_on_off(cur_state)); } @@ -351,13 +351,35 @@ intel_find_shared_dpll(struct intel_atomic_state *state, return NULL; } +/** + * intel_reference_shared_dpll_crtc - Get a DPLL reference for a CRTC + * @crtc: CRTC on which behalf the reference is taken + * @pll: DPLL for which the reference is taken + * @shared_dpll_state: the DPLL atomic state in which the reference is tracked + * + * Take a reference for @pll tracking the use of it by @crtc. + */ +static void +intel_reference_shared_dpll_crtc(const struct intel_crtc *crtc, + const struct intel_shared_dpll *pll, + struct intel_shared_dpll_state *shared_dpll_state) +{ + struct drm_i915_private *i915 = to_i915(crtc->base.dev); + + drm_WARN_ON(&i915->drm, (shared_dpll_state->pipe_mask & BIT(crtc->pipe)) != 0); + + shared_dpll_state->pipe_mask |= BIT(crtc->pipe); + + drm_dbg_kms(&i915->drm, "[CRTC:%d:%s] reserving %s\n", + crtc->base.base.id, crtc->base.name, pll->info->name); +} + static void intel_reference_shared_dpll(struct intel_atomic_state *state, const struct intel_crtc *crtc, const struct intel_shared_dpll *pll, const struct intel_dpll_hw_state *pll_state) { - struct drm_i915_private *i915 = to_i915(state->base.dev); struct intel_shared_dpll_state *shared_dpll; const enum intel_dpll_id id = pll->info->id; @@ -366,11 +388,29 @@ intel_reference_shared_dpll(struct intel_atomic_state *state, if (shared_dpll[id].pipe_mask == 0) shared_dpll[id].hw_state = *pll_state; - drm_WARN_ON(&i915->drm, (shared_dpll[id].pipe_mask & BIT(crtc->pipe)) != 0); + intel_reference_shared_dpll_crtc(crtc, pll, &shared_dpll[id]); +} + +/** + * intel_unreference_shared_dpll_crtc - Drop a DPLL reference for a CRTC + * @crtc: CRTC on which behalf the reference is dropped + * @pll: DPLL for which the reference is dropped + * @shared_dpll_state: the DPLL atomic state in which the reference is tracked + * + * Drop a reference for @pll tracking the end of use of it by @crtc. + */ +void +intel_unreference_shared_dpll_crtc(const struct intel_crtc *crtc, + const struct intel_shared_dpll *pll, + struct intel_shared_dpll_state *shared_dpll_state) +{ + struct drm_i915_private *i915 = to_i915(crtc->base.dev); + + drm_WARN_ON(&i915->drm, (shared_dpll_state->pipe_mask & BIT(crtc->pipe)) == 0); - shared_dpll[id].pipe_mask |= BIT(crtc->pipe); + shared_dpll_state->pipe_mask &= ~BIT(crtc->pipe); - drm_dbg_kms(&i915->drm, "[CRTC:%d:%s] reserving %s\n", + drm_dbg_kms(&i915->drm, "[CRTC:%d:%s] releasing %s\n", crtc->base.base.id, crtc->base.name, pll->info->name); } @@ -378,18 +418,12 @@ static void intel_unreference_shared_dpll(struct intel_atomic_state *state, const struct intel_crtc *crtc, const struct intel_shared_dpll *pll) { - struct drm_i915_private *i915 = to_i915(state->base.dev); struct intel_shared_dpll_state *shared_dpll; const enum intel_dpll_id id = pll->info->id; shared_dpll = intel_atomic_get_shared_dpll_state(&state->base); - drm_WARN_ON(&i915->drm, (shared_dpll[id].pipe_mask & BIT(crtc->pipe)) == 0); - - shared_dpll[id].pipe_mask &= ~BIT(crtc->pipe); - - drm_dbg_kms(&i915->drm, "[CRTC:%d:%s] releasing %s\n", - crtc->base.base.id, crtc->base.name, pll->info->name); + intel_unreference_shared_dpll_crtc(crtc, pll, &shared_dpll[id]); } static void intel_put_dpll(struct intel_atomic_state *state, @@ -464,12 +498,11 @@ static void ibx_assert_pch_refclk_enabled(struct drm_i915_private *dev_priv) u32 val; bool enabled; - I915_STATE_WARN_ON(!(HAS_PCH_IBX(dev_priv) || HAS_PCH_CPT(dev_priv))); - val = intel_de_read(dev_priv, PCH_DREF_CONTROL); enabled = !!(val & (DREF_SSC_SOURCE_MASK | DREF_NONSPREAD_SOURCE_MASK | DREF_SUPERSPREAD_SOURCE_MASK)); - I915_STATE_WARN(!enabled, "PCH refclk assertion failure, should be active but is disabled\n"); + I915_STATE_WARN(dev_priv, !enabled, + "PCH refclk assertion failure, should be active but is disabled\n"); } static void ibx_pch_dpll_enable(struct drm_i915_private *dev_priv, @@ -4104,7 +4137,7 @@ void intel_shared_dpll_init(struct drm_i915_private *dev_priv) mutex_init(&dev_priv->display.dpll.lock); - if (IS_DG2(dev_priv)) + if (DISPLAY_VER(dev_priv) >= 14 || IS_DG2(dev_priv)) /* No shared DPLLs on DG2; port PLLs are part of the PHY */ dpll_mgr = NULL; else if (IS_ALDERLAKE_P(dev_priv)) @@ -4314,7 +4347,7 @@ static void readout_dpll_hw_state(struct drm_i915_private *i915, to_intel_crtc_state(crtc->base.state); if (crtc_state->hw.active && crtc_state->shared_dpll == pll) - pll->state.pipe_mask |= BIT(crtc->pipe); + intel_reference_shared_dpll_crtc(crtc, pll, &pll->state); } pll->active_mask = pll->state.pipe_mask; @@ -4407,17 +4440,18 @@ verify_single_dpll_state(struct drm_i915_private *dev_priv, active = intel_dpll_get_hw_state(dev_priv, pll, &dpll_hw_state); if (!(pll->info->flags & INTEL_DPLL_ALWAYS_ON)) { - I915_STATE_WARN(!pll->on && pll->active_mask, + I915_STATE_WARN(dev_priv, !pll->on && pll->active_mask, "pll in active use but not on in sw tracking\n"); - I915_STATE_WARN(pll->on && !pll->active_mask, + I915_STATE_WARN(dev_priv, pll->on && !pll->active_mask, "pll is on but not used by any active pipe\n"); - I915_STATE_WARN(pll->on != active, + I915_STATE_WARN(dev_priv, pll->on != active, "pll on state mismatch (expected %i, found %i)\n", pll->on, active); } if (!crtc) { - I915_STATE_WARN(pll->active_mask & ~pll->state.pipe_mask, + I915_STATE_WARN(dev_priv, + pll->active_mask & ~pll->state.pipe_mask, "more active pll users than references: 0x%x vs 0x%x\n", pll->active_mask, pll->state.pipe_mask); @@ -4427,20 +4461,20 @@ verify_single_dpll_state(struct drm_i915_private *dev_priv, pipe_mask = BIT(crtc->pipe); if (new_crtc_state->hw.active) - I915_STATE_WARN(!(pll->active_mask & pipe_mask), + I915_STATE_WARN(dev_priv, !(pll->active_mask & pipe_mask), "pll active mismatch (expected pipe %c in active mask 0x%x)\n", pipe_name(crtc->pipe), pll->active_mask); else - I915_STATE_WARN(pll->active_mask & pipe_mask, + I915_STATE_WARN(dev_priv, pll->active_mask & pipe_mask, "pll active mismatch (didn't expect pipe %c in active mask 0x%x)\n", pipe_name(crtc->pipe), pll->active_mask); - I915_STATE_WARN(!(pll->state.pipe_mask & pipe_mask), + I915_STATE_WARN(dev_priv, !(pll->state.pipe_mask & pipe_mask), "pll enabled crtcs mismatch (expected 0x%x in 0x%x)\n", pipe_mask, pll->state.pipe_mask); - I915_STATE_WARN(pll->on && memcmp(&pll->state.hw_state, - &dpll_hw_state, + I915_STATE_WARN(dev_priv, + pll->on && memcmp(&pll->state.hw_state, &dpll_hw_state, sizeof(dpll_hw_state)), "pll hw state mismatch\n"); } @@ -4460,10 +4494,10 @@ void intel_shared_dpll_state_verify(struct intel_crtc *crtc, u8 pipe_mask = BIT(crtc->pipe); struct intel_shared_dpll *pll = old_crtc_state->shared_dpll; - I915_STATE_WARN(pll->active_mask & pipe_mask, + I915_STATE_WARN(dev_priv, pll->active_mask & pipe_mask, "pll active mismatch (didn't expect pipe %c in active mask (0x%x))\n", pipe_name(crtc->pipe), pll->active_mask); - I915_STATE_WARN(pll->state.pipe_mask & pipe_mask, + I915_STATE_WARN(dev_priv, pll->state.pipe_mask & pipe_mask, "pll enabled crtcs mismatch (found %x in enabled mask (0x%x))\n", pipe_name(crtc->pipe), pll->state.pipe_mask); } diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.h b/drivers/gpu/drm/i915/display/intel_dpll_mgr.h index 3854f1b4299ac3e06f9a681c418cf55b8f667d3a..ba62eb5d7c5170b79d95344f4a21beb644e8872e 100644 --- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.h +++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.h @@ -341,6 +341,9 @@ int intel_reserve_shared_dplls(struct intel_atomic_state *state, struct intel_encoder *encoder); void intel_release_shared_dplls(struct intel_atomic_state *state, struct intel_crtc *crtc); +void intel_unreference_shared_dpll_crtc(const struct intel_crtc *crtc, + const struct intel_shared_dpll *pll, + struct intel_shared_dpll_state *shared_dpll_state); void icl_set_active_port_dpll(struct intel_crtc_state *crtc_state, enum icl_port_dpll_id port_dpll_id); void intel_update_active_dpll(struct intel_atomic_state *state, diff --git a/drivers/gpu/drm/i915/display/intel_drrs.c b/drivers/gpu/drm/i915/display/intel_drrs.c index 760e63cdc0c843c8507b409df8fa9b59357e3e2f..0d35b6be5b6aa10723ae4b6abe3580da6d3b6dee 100644 --- a/drivers/gpu/drm/i915/display/intel_drrs.c +++ b/drivers/gpu/drm/i915/display/intel_drrs.c @@ -111,7 +111,9 @@ static void intel_drrs_set_state(struct intel_crtc *crtc, static void intel_drrs_schedule_work(struct intel_crtc *crtc) { - mod_delayed_work(system_wq, &crtc->drrs.work, msecs_to_jiffies(1000)); + struct drm_i915_private *i915 = to_i915(crtc->base.dev); + + mod_delayed_work(i915->unordered_wq, &crtc->drrs.work, msecs_to_jiffies(1000)); } static unsigned int intel_drrs_frontbuffer_bits(const struct intel_crtc_state *crtc_state) diff --git a/drivers/gpu/drm/i915/display/intel_dvo.c b/drivers/gpu/drm/i915/display/intel_dvo.c index eb2dcd866cc871df573dfc778d7ef29cf7601142..9884678743b66589ba1d3e7f503bbd2a18e1a8bb 100644 --- a/drivers/gpu/drm/i915/display/intel_dvo.c +++ b/drivers/gpu/drm/i915/display/intel_dvo.c @@ -271,6 +271,7 @@ static int intel_dvo_compute_config(struct intel_encoder *encoder, if (adjusted_mode->flags & DRM_MODE_FLAG_DBLSCAN) return -EINVAL; + pipe_config->sink_format = INTEL_OUTPUT_FORMAT_RGB; pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB; return 0; diff --git a/drivers/gpu/drm/i915/display/intel_fb.c b/drivers/gpu/drm/i915/display/intel_fb.c index c004f08fcfe1550404662a82d4c56e249153ba36..446bbf7986b6f7e71448c5e390af875911f01b8b 100644 --- a/drivers/gpu/drm/i915/display/intel_fb.c +++ b/drivers/gpu/drm/i915/display/intel_fb.c @@ -157,6 +157,32 @@ struct intel_modifier_desc { static const struct intel_modifier_desc intel_modifiers[] = { { + .modifier = I915_FORMAT_MOD_4_TILED_MTL_MC_CCS, + .display_ver = { 14, 14 }, + .plane_caps = INTEL_PLANE_CAP_TILING_4 | INTEL_PLANE_CAP_CCS_MC, + + .ccs.packed_aux_planes = BIT(1), + .ccs.planar_aux_planes = BIT(2) | BIT(3), + + FORMAT_OVERRIDE(gen12_ccs_formats), + }, { + .modifier = I915_FORMAT_MOD_4_TILED_MTL_RC_CCS, + .display_ver = { 14, 14 }, + .plane_caps = INTEL_PLANE_CAP_TILING_4 | INTEL_PLANE_CAP_CCS_RC, + + .ccs.packed_aux_planes = BIT(1), + + FORMAT_OVERRIDE(gen12_ccs_formats), + }, { + .modifier = I915_FORMAT_MOD_4_TILED_MTL_RC_CCS_CC, + .display_ver = { 14, 14 }, + .plane_caps = INTEL_PLANE_CAP_TILING_4 | INTEL_PLANE_CAP_CCS_RC_CC, + + .ccs.cc_planes = BIT(2), + .ccs.packed_aux_planes = BIT(1), + + FORMAT_OVERRIDE(gen12_ccs_cc_formats), + }, { .modifier = I915_FORMAT_MOD_4_TILED_DG2_MC_CCS, .display_ver = { 13, 13 }, .plane_caps = INTEL_PLANE_CAP_TILING_4 | INTEL_PLANE_CAP_CCS_MC, @@ -370,6 +396,14 @@ static bool plane_has_modifier(struct drm_i915_private *i915, if (!plane_caps_contain_all(plane_caps, md->plane_caps)) return false; + /* + * Separate AuxCCS and Flat CCS modifiers to be run only on platforms + * where supported. + */ + if (intel_fb_is_ccs_modifier(md->modifier) && + HAS_FLAT_CCS(i915) != !md->ccs.packed_aux_planes) + return false; + return true; } @@ -489,7 +523,7 @@ static bool intel_fb_is_gen12_ccs_aux_plane(const struct drm_framebuffer *fb, in { const struct intel_modifier_desc *md = lookup_modifier(fb->modifier); - return check_modifier_display_ver_range(md, 12, 13) && + return check_modifier_display_ver_range(md, 12, 14) && ccs_aux_plane_mask(md, fb->format) & BIT(color_plane); } @@ -605,6 +639,9 @@ intel_tile_width_bytes(const struct drm_framebuffer *fb, int color_plane) if (intel_fb_is_ccs_aux_plane(fb, color_plane)) return 128; fallthrough; + case I915_FORMAT_MOD_4_TILED_MTL_RC_CCS: + case I915_FORMAT_MOD_4_TILED_MTL_RC_CCS_CC: + case I915_FORMAT_MOD_4_TILED_MTL_MC_CCS: case I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS: case I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS_CC: case I915_FORMAT_MOD_Y_TILED_GEN12_MC_CCS: @@ -791,6 +828,9 @@ unsigned int intel_surf_alignment(const struct drm_framebuffer *fb, case I915_FORMAT_MOD_Y_TILED_GEN12_MC_CCS: case I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS: case I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS_CC: + case I915_FORMAT_MOD_4_TILED_MTL_MC_CCS: + case I915_FORMAT_MOD_4_TILED_MTL_RC_CCS: + case I915_FORMAT_MOD_4_TILED_MTL_RC_CCS_CC: return 16 * 1024; case I915_FORMAT_MOD_Y_TILED_CCS: case I915_FORMAT_MOD_Yf_TILED_CCS: @@ -1561,7 +1601,7 @@ int intel_fill_fb_info(struct drm_i915_private *i915, struct intel_framebuffer * for (i = 0; i < num_planes; i++) { struct fb_plane_view_dims view_dims; unsigned int width, height; - unsigned int cpp, size; + unsigned int size; u32 offset; int x, y; int ret; @@ -1578,7 +1618,6 @@ int intel_fill_fb_info(struct drm_i915_private *i915, struct intel_framebuffer * return -EINVAL; } - cpp = fb->base.format->cpp[i]; intel_fb_plane_dims(fb, i, &width, &height); ret = convert_plane_offset_to_xy(fb, i, width, &x, &y); diff --git a/drivers/gpu/drm/i915/display/intel_fb_pin.c b/drivers/gpu/drm/i915/display/intel_fb_pin.c index 1aca7552a85d03bde806c340cf44b8800ef437c6..fffd568070d4141c5c3f7b7f0e1d220f1e0481c7 100644 --- a/drivers/gpu/drm/i915/display/intel_fb_pin.c +++ b/drivers/gpu/drm/i915/display/intel_fb_pin.c @@ -243,7 +243,7 @@ int intel_plane_pin_fb(struct intel_plane_state *plane_state) struct i915_vma *vma; bool phys_cursor = plane->id == PLANE_CURSOR && - INTEL_INFO(dev_priv)->display.cursor_needs_physical; + DISPLAY_INFO(dev_priv)->cursor_needs_physical; if (!intel_fb_uses_dpt(fb)) { vma = intel_pin_and_fence_fb_obj(fb, phys_cursor, diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c index b507ff944864ebd7a5e62818650c10338001f68d..7f8b2d7713c7a3bfbafd9b82c5f1f1552357284f 100644 --- a/drivers/gpu/drm/i915/display/intel_fbc.c +++ b/drivers/gpu/drm/i915/display/intel_fbc.c @@ -44,6 +44,7 @@ #include #include "i915_drv.h" +#include "i915_reg.h" #include "i915_utils.h" #include "i915_vgpu.h" #include "intel_cdclk.h" @@ -55,7 +56,7 @@ #define for_each_fbc_id(__dev_priv, __fbc_id) \ for ((__fbc_id) = INTEL_FBC_A; (__fbc_id) < I915_MAX_FBCS; (__fbc_id)++) \ - for_each_if(RUNTIME_INFO(__dev_priv)->fbc_mask & BIT(__fbc_id)) + for_each_if(DISPLAY_RUNTIME_INFO(__dev_priv)->fbc_mask & BIT(__fbc_id)) #define for_each_intel_fbc(__dev_priv, __fbc, __fbc_id) \ for_each_fbc_id((__dev_priv), (__fbc_id)) \ @@ -1253,7 +1254,7 @@ static bool __intel_fbc_pre_update(struct intel_atomic_state *state, bool intel_fbc_pre_update(struct intel_atomic_state *state, struct intel_crtc *crtc) { - const struct intel_plane_state *plane_state; + const struct intel_plane_state __maybe_unused *plane_state; bool need_vblank_wait = false; struct intel_plane *plane; int i; @@ -1308,7 +1309,7 @@ static void __intel_fbc_post_update(struct intel_fbc *fbc) void intel_fbc_post_update(struct intel_atomic_state *state, struct intel_crtc *crtc) { - const struct intel_plane_state *plane_state; + const struct intel_plane_state __maybe_unused *plane_state; struct intel_plane *plane; int i; @@ -1407,7 +1408,7 @@ void intel_fbc_flush(struct drm_i915_private *i915, int intel_fbc_atomic_check(struct intel_atomic_state *state) { - struct intel_plane_state *plane_state; + struct intel_plane_state __maybe_unused *plane_state; struct intel_plane *plane; int i; @@ -1599,7 +1600,7 @@ static void __intel_fbc_handle_fifo_underrun_irq(struct intel_fbc *fbc) if (READ_ONCE(fbc->underrun_detected)) return; - schedule_work(&fbc->underrun_work); + queue_work(fbc->i915->unordered_wq, &fbc->underrun_work); } /** @@ -1707,10 +1708,10 @@ void intel_fbc_init(struct drm_i915_private *i915) enum intel_fbc_id fbc_id; if (!drm_mm_initialized(&i915->mm.stolen)) - RUNTIME_INFO(i915)->fbc_mask = 0; + DISPLAY_RUNTIME_INFO(i915)->fbc_mask = 0; if (need_fbc_vtd_wa(i915)) - RUNTIME_INFO(i915)->fbc_mask = 0; + DISPLAY_RUNTIME_INFO(i915)->fbc_mask = 0; i915->params.enable_fbc = intel_sanitize_fbc_option(i915); drm_dbg_kms(&i915->drm, "Sanitized enable_fbc value: %d\n", diff --git a/drivers/gpu/drm/i915/display/intel_fbdev.c b/drivers/gpu/drm/i915/display/intel_fbdev.c index eccaceaf8b9d17fc210e31edb0dc1708e04c7ced..1cc0ddc6a310ed64275d350ad2e85ebce7a44226 100644 --- a/drivers/gpu/drm/i915/display/intel_fbdev.c +++ b/drivers/gpu/drm/i915/display/intel_fbdev.c @@ -135,6 +135,9 @@ static int intel_fbdev_mmap(struct fb_info *info, struct vm_area_struct *vma) return i915_gem_fb_mmap(obj, vma); } +__diag_push(); +__diag_ignore_all("-Woverride-init", "Allow overriding the default ops"); + static const struct fb_ops intelfb_ops = { .owner = THIS_MODULE, __FB_DEFAULT_DEFERRED_OPS_RDWR(intel_fbdev), @@ -146,6 +149,8 @@ static const struct fb_ops intelfb_ops = { .fb_mmap = intel_fbdev_mmap, }; +__diag_pop(); + static int intelfb_alloc(struct drm_fb_helper *helper, struct drm_fb_helper_surface_size *sizes) { @@ -691,7 +696,8 @@ void intel_fbdev_set_suspend(struct drm_device *dev, int state, bool synchronous /* Don't block our own workqueue as this can * be run in parallel with other i915.ko tasks. */ - schedule_work(&dev_priv->display.fbdev.suspend_work); + queue_work(dev_priv->unordered_wq, + &dev_priv->display.fbdev.suspend_work); return; } } diff --git a/drivers/gpu/drm/i915/display/intel_fdi.c b/drivers/gpu/drm/i915/display/intel_fdi.c index 55283677c45a4c4b0e410f376cef672eac582104..e12b46a84fa11dc2e923aa1dec585f03f00b5040 100644 --- a/drivers/gpu/drm/i915/display/intel_fdi.c +++ b/drivers/gpu/drm/i915/display/intel_fdi.c @@ -36,7 +36,7 @@ static void assert_fdi_tx(struct drm_i915_private *dev_priv, } else { cur_state = intel_de_read(dev_priv, FDI_TX_CTL(pipe)) & FDI_TX_ENABLE; } - I915_STATE_WARN(cur_state != state, + I915_STATE_WARN(dev_priv, cur_state != state, "FDI TX state assertion failure (expected %s, current %s)\n", str_on_off(state), str_on_off(cur_state)); } @@ -57,7 +57,7 @@ static void assert_fdi_rx(struct drm_i915_private *dev_priv, bool cur_state; cur_state = intel_de_read(dev_priv, FDI_RX_CTL(pipe)) & FDI_RX_ENABLE; - I915_STATE_WARN(cur_state != state, + I915_STATE_WARN(dev_priv, cur_state != state, "FDI RX state assertion failure (expected %s, current %s)\n", str_on_off(state), str_on_off(cur_state)); } @@ -86,7 +86,8 @@ void assert_fdi_tx_pll_enabled(struct drm_i915_private *i915, return; cur_state = intel_de_read(i915, FDI_TX_CTL(pipe)) & FDI_TX_PLL_ENABLE; - I915_STATE_WARN(!cur_state, "FDI TX PLL assertion failure, should be active but is disabled\n"); + I915_STATE_WARN(i915, !cur_state, + "FDI TX PLL assertion failure, should be active but is disabled\n"); } static void assert_fdi_rx_pll(struct drm_i915_private *i915, @@ -95,7 +96,7 @@ static void assert_fdi_rx_pll(struct drm_i915_private *i915, bool cur_state; cur_state = intel_de_read(i915, FDI_RX_CTL(pipe)) & FDI_RX_PLL_ENABLE; - I915_STATE_WARN(cur_state != state, + I915_STATE_WARN(i915, cur_state != state, "FDI RX PLL assertion failure (expected %s, current %s)\n", str_on_off(state), str_on_off(cur_state)); } diff --git a/drivers/gpu/drm/i915/display/intel_fifo_underrun.c b/drivers/gpu/drm/i915/display/intel_fifo_underrun.c index b708a62e509a0a71c53b58db1142523208796e78..09a7fa6c0c37550cf688a5ce0bd991a7ce7a9c9d 100644 --- a/drivers/gpu/drm/i915/display/intel_fifo_underrun.c +++ b/drivers/gpu/drm/i915/display/intel_fifo_underrun.c @@ -26,7 +26,9 @@ */ #include "i915_drv.h" +#include "i915_reg.h" #include "intel_de.h" +#include "intel_display_irq.h" #include "intel_display_trace.h" #include "intel_display_types.h" #include "intel_fbc.h" diff --git a/drivers/gpu/drm/i915/display/intel_global_state.c b/drivers/gpu/drm/i915/display/intel_global_state.c index 02b593b1e2ea15d65209566b149aea83b7c5e90c..e8e8be54143b431393bfedcac3b399429defd3ce 100644 --- a/drivers/gpu/drm/i915/display/intel_global_state.c +++ b/drivers/gpu/drm/i915/display/intel_global_state.c @@ -255,3 +255,15 @@ int intel_atomic_serialize_global_state(struct intel_global_state *obj_state) return 0; } + +bool +intel_atomic_global_state_is_serialized(struct intel_atomic_state *state) +{ + struct drm_i915_private *i915 = to_i915(state->base.dev); + struct intel_crtc *crtc; + + for_each_intel_crtc(&i915->drm, crtc) + if (!intel_atomic_get_new_crtc_state(state, crtc)) + return false; + return true; +} diff --git a/drivers/gpu/drm/i915/display/intel_global_state.h b/drivers/gpu/drm/i915/display/intel_global_state.h index f01ee0bb3e5a34917dc3dcfa72df3e7ad48bf123..5477de8f0b30ce4ec0edfc9337814365810cd5eb 100644 --- a/drivers/gpu/drm/i915/display/intel_global_state.h +++ b/drivers/gpu/drm/i915/display/intel_global_state.h @@ -87,4 +87,6 @@ void intel_atomic_clear_global_state(struct intel_atomic_state *state); int intel_atomic_lock_global_state(struct intel_global_state *obj_state); int intel_atomic_serialize_global_state(struct intel_global_state *obj_state); +bool intel_atomic_global_state_is_serialized(struct intel_atomic_state *state); + #endif diff --git a/drivers/gpu/drm/i915/display/intel_gmbus.c b/drivers/gpu/drm/i915/display/intel_gmbus.c index 3ddfc8080ee890620388b12b6a4ea699cb6876d8..e95ddb580ef6cceed52c9f8235c55c968dd6dc5d 100644 --- a/drivers/gpu/drm/i915/display/intel_gmbus.c +++ b/drivers/gpu/drm/i915/display/intel_gmbus.c @@ -991,3 +991,8 @@ void intel_gmbus_teardown(struct drm_i915_private *i915) i915->display.gmbus.bus[pin] = NULL; } } + +void intel_gmbus_irq_handler(struct drm_i915_private *i915) +{ + wake_up_all(&i915->display.gmbus.wait_queue); +} diff --git a/drivers/gpu/drm/i915/display/intel_gmbus.h b/drivers/gpu/drm/i915/display/intel_gmbus.h index 20f704bd4e7004fab13e6aa097f4a09a99355e93..8111eb23e2af207c04cff3169835318deb49e173 100644 --- a/drivers/gpu/drm/i915/display/intel_gmbus.h +++ b/drivers/gpu/drm/i915/display/intel_gmbus.h @@ -46,4 +46,6 @@ void intel_gmbus_force_bit(struct i2c_adapter *adapter, bool force_bit); bool intel_gmbus_is_forced_bit(struct i2c_adapter *adapter); void intel_gmbus_reset(struct drm_i915_private *dev_priv); +void intel_gmbus_irq_handler(struct drm_i915_private *i915); + #endif /* __INTEL_GMBUS_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c index 650232c4892b183800fa86adbe518e964cc0c29b..5ed450111f77c4387503351ed2c3a79f071937b0 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp.c +++ b/drivers/gpu/drm/i915/display/intel_hdcp.c @@ -30,7 +30,8 @@ #define KEY_LOAD_TRIES 5 #define HDCP2_LC_RETRY_CNT 3 -static int intel_conn_to_vcpi(struct intel_connector *connector) +static int intel_conn_to_vcpi(struct drm_atomic_state *state, + struct intel_connector *connector) { struct drm_dp_mst_topology_mgr *mgr; struct drm_dp_mst_atomic_payload *payload; @@ -42,7 +43,7 @@ static int intel_conn_to_vcpi(struct intel_connector *connector) return 0; mgr = connector->port->mgr; - drm_modeset_lock(&mgr->base.lock, NULL); + drm_modeset_lock(&mgr->base.lock, state->acquire_ctx); mst_state = to_drm_dp_mst_topology_state(mgr->base.state); payload = drm_atomic_get_mst_payload_state(mst_state, connector->port); if (drm_WARN_ON(mgr->dev, !payload)) @@ -54,7 +55,6 @@ static int intel_conn_to_vcpi(struct intel_connector *connector) goto out; } out: - drm_modeset_unlock(&mgr->base.lock); return vcpi; } @@ -68,48 +68,18 @@ static int intel_conn_to_vcpi(struct intel_connector *connector) * DP MST topology. Though it is not compulsory, security fw should change its * policy to mark different content_types for different streams. */ -static int +static void intel_hdcp_required_content_stream(struct intel_digital_port *dig_port) { - struct drm_connector_list_iter conn_iter; - struct intel_digital_port *conn_dig_port; - struct intel_connector *connector; - struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev); struct hdcp_port_data *data = &dig_port->hdcp_port_data; bool enforce_type0 = false; int k; - data->k = 0; - if (dig_port->hdcp_auth_status) - return 0; - - drm_connector_list_iter_begin(&i915->drm, &conn_iter); - for_each_intel_connector_iter(connector, &conn_iter) { - if (connector->base.status == connector_status_disconnected) - continue; - - if (!intel_encoder_is_mst(intel_attached_encoder(connector))) - continue; - - conn_dig_port = intel_attached_dig_port(connector); - if (conn_dig_port != dig_port) - continue; - - if (!enforce_type0 && !dig_port->hdcp_mst_type1_capable) - enforce_type0 = true; - - data->streams[data->k].stream_id = intel_conn_to_vcpi(connector); - data->k++; - - /* if there is only one active stream */ - if (dig_port->dp.active_mst_links <= 1) - break; - } - drm_connector_list_iter_end(&conn_iter); + return; - if (drm_WARN_ON(&i915->drm, data->k > INTEL_NUM_PIPES(i915) || data->k == 0)) - return -EINVAL; + if (!dig_port->hdcp_mst_type1_capable) + enforce_type0 = true; /* * Apply common protection level across all streams in DP MST Topology. @@ -118,27 +88,19 @@ intel_hdcp_required_content_stream(struct intel_digital_port *dig_port) for (k = 0; k < data->k; k++) data->streams[k].stream_type = enforce_type0 ? DRM_MODE_HDCP_CONTENT_TYPE0 : DRM_MODE_HDCP_CONTENT_TYPE1; - - return 0; } -static int intel_hdcp_prepare_streams(struct intel_connector *connector) +static void intel_hdcp_prepare_streams(struct intel_connector *connector) { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); struct hdcp_port_data *data = &dig_port->hdcp_port_data; struct intel_hdcp *hdcp = &connector->hdcp; - int ret; if (!intel_encoder_is_mst(intel_attached_encoder(connector))) { - data->k = 1; data->streams[0].stream_type = hdcp->content_type; } else { - ret = intel_hdcp_required_content_stream(dig_port); - if (ret) - return ret; + intel_hdcp_required_content_stream(dig_port); } - - return 0; } static @@ -202,10 +164,8 @@ bool intel_hdcp_capable(struct intel_connector *connector) bool intel_hdcp2_capable(struct intel_connector *connector) { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct drm_i915_private *i915 = to_i915(connector->base.dev); struct intel_hdcp *hdcp = &connector->hdcp; - struct intel_gt *gt = dev_priv->media_gt; - struct intel_gsc_uc *gsc = >->uc.gsc; bool capable = false; /* I915 support for HDCP2.2 */ @@ -213,17 +173,21 @@ bool intel_hdcp2_capable(struct intel_connector *connector) return false; /* If MTL+ make sure gsc is loaded and proxy is setup */ - if (intel_hdcp_gsc_cs_required(dev_priv)) - if (!intel_uc_fw_is_running(&gsc->fw)) + if (intel_hdcp_gsc_cs_required(i915)) { + struct intel_gt *gt = i915->media_gt; + struct intel_gsc_uc *gsc = gt ? >->uc.gsc : NULL; + + if (!gsc || !intel_uc_fw_is_running(&gsc->fw)) return false; + } /* MEI/GSC interface is solid depending on which is used */ - mutex_lock(&dev_priv->display.hdcp.comp_mutex); - if (!dev_priv->display.hdcp.comp_added || !dev_priv->display.hdcp.master) { - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_lock(&i915->display.hdcp.hdcp_mutex); + if (!i915->display.hdcp.comp_added || !i915->display.hdcp.arbiter) { + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return false; } - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); /* Sink's capability for HDCP2.2 */ hdcp->shim->hdcp_2_2_capable(dig_port, &capable); @@ -231,20 +195,20 @@ bool intel_hdcp2_capable(struct intel_connector *connector) return capable; } -static bool intel_hdcp_in_use(struct drm_i915_private *dev_priv, +static bool intel_hdcp_in_use(struct drm_i915_private *i915, enum transcoder cpu_transcoder, enum port port) { - return intel_de_read(dev_priv, - HDCP_STATUS(dev_priv, cpu_transcoder, port)) & - HDCP_STATUS_ENC; + return intel_de_read(i915, + HDCP_STATUS(i915, cpu_transcoder, port)) & + HDCP_STATUS_ENC; } -static bool intel_hdcp2_in_use(struct drm_i915_private *dev_priv, +static bool intel_hdcp2_in_use(struct drm_i915_private *i915, enum transcoder cpu_transcoder, enum port port) { - return intel_de_read(dev_priv, - HDCP2_STATUS(dev_priv, cpu_transcoder, port)) & - LINK_ENCRYPTION_STATUS; + return intel_de_read(i915, + HDCP2_STATUS(i915, cpu_transcoder, port)) & + LINK_ENCRYPTION_STATUS; } static int intel_hdcp_poll_ksv_fifo(struct intel_digital_port *dig_port, @@ -268,7 +232,7 @@ static int intel_hdcp_poll_ksv_fifo(struct intel_digital_port *dig_port, return 0; } -static bool hdcp_key_loadable(struct drm_i915_private *dev_priv) +static bool hdcp_key_loadable(struct drm_i915_private *i915) { enum i915_power_well_id id; intel_wakeref_t wakeref; @@ -278,14 +242,14 @@ static bool hdcp_key_loadable(struct drm_i915_private *dev_priv) * On HSW and BDW, Display HW loads the Key as soon as Display resumes. * On all BXT+, SW can load the keys only when the PW#1 is turned on. */ - if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) + if (IS_HASWELL(i915) || IS_BROADWELL(i915)) id = HSW_DISP_PW_GLOBAL; else id = SKL_DISP_PW_1; /* PG1 (power well #1) needs to be enabled */ - with_intel_runtime_pm(&dev_priv->runtime_pm, wakeref) - enabled = intel_display_power_well_is_enabled(dev_priv, id); + with_intel_runtime_pm(&i915->runtime_pm, wakeref) + enabled = intel_display_power_well_is_enabled(i915, id); /* * Another req for hdcp key loadability is enabled state of pll for @@ -296,19 +260,19 @@ static bool hdcp_key_loadable(struct drm_i915_private *dev_priv) return enabled; } -static void intel_hdcp_clear_keys(struct drm_i915_private *dev_priv) +static void intel_hdcp_clear_keys(struct drm_i915_private *i915) { - intel_de_write(dev_priv, HDCP_KEY_CONF, HDCP_CLEAR_KEYS_TRIGGER); - intel_de_write(dev_priv, HDCP_KEY_STATUS, + intel_de_write(i915, HDCP_KEY_CONF, HDCP_CLEAR_KEYS_TRIGGER); + intel_de_write(i915, HDCP_KEY_STATUS, HDCP_KEY_LOAD_DONE | HDCP_KEY_LOAD_STATUS | HDCP_FUSE_IN_PROGRESS | HDCP_FUSE_ERROR | HDCP_FUSE_DONE); } -static int intel_hdcp_load_keys(struct drm_i915_private *dev_priv) +static int intel_hdcp_load_keys(struct drm_i915_private *i915) { int ret; u32 val; - val = intel_de_read(dev_priv, HDCP_KEY_STATUS); + val = intel_de_read(i915, HDCP_KEY_STATUS); if ((val & HDCP_KEY_LOAD_DONE) && (val & HDCP_KEY_LOAD_STATUS)) return 0; @@ -316,8 +280,8 @@ static int intel_hdcp_load_keys(struct drm_i915_private *dev_priv) * On HSW and BDW HW loads the HDCP1.4 Key when Display comes * out of reset. So if Key is not already loaded, its an error state. */ - if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) - if (!(intel_de_read(dev_priv, HDCP_KEY_STATUS) & HDCP_KEY_LOAD_DONE)) + if (IS_HASWELL(i915) || IS_BROADWELL(i915)) + if (!(intel_de_read(i915, HDCP_KEY_STATUS) & HDCP_KEY_LOAD_DONE)) return -ENXIO; /* @@ -328,20 +292,20 @@ static int intel_hdcp_load_keys(struct drm_i915_private *dev_priv) * process from other platforms. These platforms use the GT Driver * Mailbox interface. */ - if (DISPLAY_VER(dev_priv) == 9 && !IS_BROXTON(dev_priv)) { - ret = snb_pcode_write(&dev_priv->uncore, SKL_PCODE_LOAD_HDCP_KEYS, 1); + if (DISPLAY_VER(i915) == 9 && !IS_BROXTON(i915)) { + ret = snb_pcode_write(&i915->uncore, SKL_PCODE_LOAD_HDCP_KEYS, 1); if (ret) { - drm_err(&dev_priv->drm, + drm_err(&i915->drm, "Failed to initiate HDCP key load (%d)\n", ret); return ret; } } else { - intel_de_write(dev_priv, HDCP_KEY_CONF, HDCP_KEY_LOAD_TRIGGER); + intel_de_write(i915, HDCP_KEY_CONF, HDCP_KEY_LOAD_TRIGGER); } /* Wait for the keys to load (500us) */ - ret = __intel_wait_for_register(&dev_priv->uncore, HDCP_KEY_STATUS, + ret = __intel_wait_for_register(&i915->uncore, HDCP_KEY_STATUS, HDCP_KEY_LOAD_DONE, HDCP_KEY_LOAD_DONE, 10, 1, &val); if (ret) @@ -350,27 +314,27 @@ static int intel_hdcp_load_keys(struct drm_i915_private *dev_priv) return -ENXIO; /* Send Aksv over to PCH display for use in authentication */ - intel_de_write(dev_priv, HDCP_KEY_CONF, HDCP_AKSV_SEND_TRIGGER); + intel_de_write(i915, HDCP_KEY_CONF, HDCP_AKSV_SEND_TRIGGER); return 0; } /* Returns updated SHA-1 index */ -static int intel_write_sha_text(struct drm_i915_private *dev_priv, u32 sha_text) +static int intel_write_sha_text(struct drm_i915_private *i915, u32 sha_text) { - intel_de_write(dev_priv, HDCP_SHA_TEXT, sha_text); - if (intel_de_wait_for_set(dev_priv, HDCP_REP_CTL, HDCP_SHA1_READY, 1)) { - drm_err(&dev_priv->drm, "Timed out waiting for SHA1 ready\n"); + intel_de_write(i915, HDCP_SHA_TEXT, sha_text); + if (intel_de_wait_for_set(i915, HDCP_REP_CTL, HDCP_SHA1_READY, 1)) { + drm_err(&i915->drm, "Timed out waiting for SHA1 ready\n"); return -ETIMEDOUT; } return 0; } static -u32 intel_hdcp_get_repeater_ctl(struct drm_i915_private *dev_priv, +u32 intel_hdcp_get_repeater_ctl(struct drm_i915_private *i915, enum transcoder cpu_transcoder, enum port port) { - if (DISPLAY_VER(dev_priv) >= 12) { + if (DISPLAY_VER(i915) >= 12) { switch (cpu_transcoder) { case TRANSCODER_A: return HDCP_TRANSA_REP_PRESENT | @@ -385,7 +349,7 @@ u32 intel_hdcp_get_repeater_ctl(struct drm_i915_private *dev_priv, return HDCP_TRANSD_REP_PRESENT | HDCP_TRANSD_SHA1_M0; default: - drm_err(&dev_priv->drm, "Unknown transcoder %d\n", + drm_err(&i915->drm, "Unknown transcoder %d\n", cpu_transcoder); return -EINVAL; } @@ -403,7 +367,7 @@ u32 intel_hdcp_get_repeater_ctl(struct drm_i915_private *dev_priv, case PORT_E: return HDCP_DDIE_REP_PRESENT | HDCP_DDIE_SHA1_M0; default: - drm_err(&dev_priv->drm, "Unknown port %d\n", port); + drm_err(&i915->drm, "Unknown port %d\n", port); return -EINVAL; } } @@ -414,7 +378,7 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector, u8 *ksv_fifo, u8 num_downstream, u8 *bstatus) { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct drm_i915_private *i915 = to_i915(connector->base.dev); enum transcoder cpu_transcoder = connector->hdcp.cpu_transcoder; enum port port = dig_port->base.port; u32 vprime, sha_text, sha_leftovers, rep_ctl; @@ -425,7 +389,7 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector, ret = shim->read_v_prime_part(dig_port, i, &vprime); if (ret) return ret; - intel_de_write(dev_priv, HDCP_SHA_V_PRIME(i), vprime); + intel_de_write(i915, HDCP_SHA_V_PRIME(i), vprime); } /* @@ -441,8 +405,8 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector, sha_idx = 0; sha_text = 0; sha_leftovers = 0; - rep_ctl = intel_hdcp_get_repeater_ctl(dev_priv, cpu_transcoder, port); - intel_de_write(dev_priv, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_32); + rep_ctl = intel_hdcp_get_repeater_ctl(i915, cpu_transcoder, port); + intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_32); for (i = 0; i < num_downstream; i++) { unsigned int sha_empty; u8 *ksv = &ksv_fifo[i * DRM_HDCP_KSV_LEN]; @@ -454,14 +418,14 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector, sha_text |= ksv[j] << off; } - ret = intel_write_sha_text(dev_priv, sha_text); + ret = intel_write_sha_text(i915, sha_text); if (ret < 0) return ret; /* Programming guide writes this every 64 bytes */ sha_idx += sizeof(sha_text); if (!(sha_idx % 64)) - intel_de_write(dev_priv, HDCP_REP_CTL, + intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_32); /* Store the leftover bytes from the ksv in sha_text */ @@ -478,7 +442,7 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector, if (sizeof(sha_text) > sha_leftovers) continue; - ret = intel_write_sha_text(dev_priv, sha_text); + ret = intel_write_sha_text(i915, sha_text); if (ret < 0) return ret; sha_leftovers = 0; @@ -494,73 +458,73 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector, */ if (sha_leftovers == 0) { /* Write 16 bits of text, 16 bits of M0 */ - intel_de_write(dev_priv, HDCP_REP_CTL, + intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_16); - ret = intel_write_sha_text(dev_priv, + ret = intel_write_sha_text(i915, bstatus[0] << 8 | bstatus[1]); if (ret < 0) return ret; sha_idx += sizeof(sha_text); /* Write 32 bits of M0 */ - intel_de_write(dev_priv, HDCP_REP_CTL, + intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_0); - ret = intel_write_sha_text(dev_priv, 0); + ret = intel_write_sha_text(i915, 0); if (ret < 0) return ret; sha_idx += sizeof(sha_text); /* Write 16 bits of M0 */ - intel_de_write(dev_priv, HDCP_REP_CTL, + intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_16); - ret = intel_write_sha_text(dev_priv, 0); + ret = intel_write_sha_text(i915, 0); if (ret < 0) return ret; sha_idx += sizeof(sha_text); } else if (sha_leftovers == 1) { /* Write 24 bits of text, 8 bits of M0 */ - intel_de_write(dev_priv, HDCP_REP_CTL, + intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_24); sha_text |= bstatus[0] << 16 | bstatus[1] << 8; /* Only 24-bits of data, must be in the LSB */ sha_text = (sha_text & 0xffffff00) >> 8; - ret = intel_write_sha_text(dev_priv, sha_text); + ret = intel_write_sha_text(i915, sha_text); if (ret < 0) return ret; sha_idx += sizeof(sha_text); /* Write 32 bits of M0 */ - intel_de_write(dev_priv, HDCP_REP_CTL, + intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_0); - ret = intel_write_sha_text(dev_priv, 0); + ret = intel_write_sha_text(i915, 0); if (ret < 0) return ret; sha_idx += sizeof(sha_text); /* Write 24 bits of M0 */ - intel_de_write(dev_priv, HDCP_REP_CTL, + intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_8); - ret = intel_write_sha_text(dev_priv, 0); + ret = intel_write_sha_text(i915, 0); if (ret < 0) return ret; sha_idx += sizeof(sha_text); } else if (sha_leftovers == 2) { /* Write 32 bits of text */ - intel_de_write(dev_priv, HDCP_REP_CTL, + intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_32); sha_text |= bstatus[0] << 8 | bstatus[1]; - ret = intel_write_sha_text(dev_priv, sha_text); + ret = intel_write_sha_text(i915, sha_text); if (ret < 0) return ret; sha_idx += sizeof(sha_text); /* Write 64 bits of M0 */ - intel_de_write(dev_priv, HDCP_REP_CTL, + intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_0); for (i = 0; i < 2; i++) { - ret = intel_write_sha_text(dev_priv, 0); + ret = intel_write_sha_text(i915, 0); if (ret < 0) return ret; sha_idx += sizeof(sha_text); @@ -570,56 +534,56 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector, * Terminate the SHA-1 stream by hand. For the other leftover * cases this is appended by the hardware. */ - intel_de_write(dev_priv, HDCP_REP_CTL, + intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_32); sha_text = DRM_HDCP_SHA1_TERMINATOR << 24; - ret = intel_write_sha_text(dev_priv, sha_text); + ret = intel_write_sha_text(i915, sha_text); if (ret < 0) return ret; sha_idx += sizeof(sha_text); } else if (sha_leftovers == 3) { /* Write 32 bits of text (filled from LSB) */ - intel_de_write(dev_priv, HDCP_REP_CTL, + intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_32); sha_text |= bstatus[0]; - ret = intel_write_sha_text(dev_priv, sha_text); + ret = intel_write_sha_text(i915, sha_text); if (ret < 0) return ret; sha_idx += sizeof(sha_text); /* Write 8 bits of text (filled from LSB), 24 bits of M0 */ - intel_de_write(dev_priv, HDCP_REP_CTL, + intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_8); - ret = intel_write_sha_text(dev_priv, bstatus[1]); + ret = intel_write_sha_text(i915, bstatus[1]); if (ret < 0) return ret; sha_idx += sizeof(sha_text); /* Write 32 bits of M0 */ - intel_de_write(dev_priv, HDCP_REP_CTL, + intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_0); - ret = intel_write_sha_text(dev_priv, 0); + ret = intel_write_sha_text(i915, 0); if (ret < 0) return ret; sha_idx += sizeof(sha_text); /* Write 8 bits of M0 */ - intel_de_write(dev_priv, HDCP_REP_CTL, + intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_24); - ret = intel_write_sha_text(dev_priv, 0); + ret = intel_write_sha_text(i915, 0); if (ret < 0) return ret; sha_idx += sizeof(sha_text); } else { - drm_dbg_kms(&dev_priv->drm, "Invalid number of leftovers %d\n", + drm_dbg_kms(&i915->drm, "Invalid number of leftovers %d\n", sha_leftovers); return -EINVAL; } - intel_de_write(dev_priv, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_32); + intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_TEXT_32); /* Fill up to 64-4 bytes with zeros (leave the last write for length) */ while ((sha_idx % 64) < (64 - sizeof(sha_text))) { - ret = intel_write_sha_text(dev_priv, 0); + ret = intel_write_sha_text(i915, 0); if (ret < 0) return ret; sha_idx += sizeof(sha_text); @@ -631,20 +595,20 @@ int intel_hdcp_validate_v_prime(struct intel_connector *connector, * - 10 bytes for BINFO/BSTATUS(2), M0(8) */ sha_text = (num_downstream * 5 + 10) * 8; - ret = intel_write_sha_text(dev_priv, sha_text); + ret = intel_write_sha_text(i915, sha_text); if (ret < 0) return ret; /* Tell the HW we're done with the hash and wait for it to ACK */ - intel_de_write(dev_priv, HDCP_REP_CTL, + intel_de_write(i915, HDCP_REP_CTL, rep_ctl | HDCP_SHA1_COMPLETE_HASH); - if (intel_de_wait_for_set(dev_priv, HDCP_REP_CTL, + if (intel_de_wait_for_set(i915, HDCP_REP_CTL, HDCP_SHA1_COMPLETE, 1)) { - drm_err(&dev_priv->drm, "Timed out waiting for SHA1 complete\n"); + drm_err(&i915->drm, "Timed out waiting for SHA1 complete\n"); return -ETIMEDOUT; } - if (!(intel_de_read(dev_priv, HDCP_REP_CTL) & HDCP_SHA1_V_MATCH)) { - drm_dbg_kms(&dev_priv->drm, "SHA-1 mismatch, HDCP failed\n"); + if (!(intel_de_read(i915, HDCP_REP_CTL) & HDCP_SHA1_V_MATCH)) { + drm_dbg_kms(&i915->drm, "SHA-1 mismatch, HDCP failed\n"); return -ENXIO; } @@ -656,14 +620,14 @@ static int intel_hdcp_auth_downstream(struct intel_connector *connector) { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct drm_i915_private *i915 = to_i915(connector->base.dev); const struct intel_hdcp_shim *shim = connector->hdcp.shim; u8 bstatus[2], num_downstream, *ksv_fifo; int ret, i, tries = 3; ret = intel_hdcp_poll_ksv_fifo(dig_port, shim); if (ret) { - drm_dbg_kms(&dev_priv->drm, + drm_dbg_kms(&i915->drm, "KSV list failed to become ready (%d)\n", ret); return ret; } @@ -674,7 +638,7 @@ int intel_hdcp_auth_downstream(struct intel_connector *connector) if (DRM_HDCP_MAX_DEVICE_EXCEEDED(bstatus[0]) || DRM_HDCP_MAX_CASCADE_EXCEEDED(bstatus[1])) { - drm_dbg_kms(&dev_priv->drm, "Max Topology Limit Exceeded\n"); + drm_dbg_kms(&i915->drm, "Max Topology Limit Exceeded\n"); return -EPERM; } @@ -687,14 +651,14 @@ int intel_hdcp_auth_downstream(struct intel_connector *connector) */ num_downstream = DRM_HDCP_NUM_DOWNSTREAM(bstatus[0]); if (num_downstream == 0) { - drm_dbg_kms(&dev_priv->drm, + drm_dbg_kms(&i915->drm, "Repeater with zero downstream devices\n"); return -EINVAL; } ksv_fifo = kcalloc(DRM_HDCP_KSV_LEN, num_downstream, GFP_KERNEL); if (!ksv_fifo) { - drm_dbg_kms(&dev_priv->drm, "Out of mem: ksv_fifo\n"); + drm_dbg_kms(&i915->drm, "Out of mem: ksv_fifo\n"); return -ENOMEM; } @@ -702,9 +666,9 @@ int intel_hdcp_auth_downstream(struct intel_connector *connector) if (ret) goto err; - if (drm_hdcp_check_ksvs_revoked(&dev_priv->drm, ksv_fifo, + if (drm_hdcp_check_ksvs_revoked(&i915->drm, ksv_fifo, num_downstream) > 0) { - drm_err(&dev_priv->drm, "Revoked Ksv(s) in ksv_fifo\n"); + drm_err(&i915->drm, "Revoked Ksv(s) in ksv_fifo\n"); ret = -EPERM; goto err; } @@ -722,12 +686,12 @@ int intel_hdcp_auth_downstream(struct intel_connector *connector) } if (i == tries) { - drm_dbg_kms(&dev_priv->drm, + drm_dbg_kms(&i915->drm, "V Prime validation failed.(%d)\n", ret); goto err; } - drm_dbg_kms(&dev_priv->drm, "HDCP is enabled (%d downstream devices)\n", + drm_dbg_kms(&i915->drm, "HDCP is enabled (%d downstream devices)\n", num_downstream); ret = 0; err: @@ -739,7 +703,7 @@ int intel_hdcp_auth_downstream(struct intel_connector *connector) static int intel_hdcp_auth(struct intel_connector *connector) { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct drm_i915_private *i915 = to_i915(connector->base.dev); struct intel_hdcp *hdcp = &connector->hdcp; const struct intel_hdcp_shim *shim = hdcp->shim; enum transcoder cpu_transcoder = connector->hdcp.cpu_transcoder; @@ -771,7 +735,7 @@ static int intel_hdcp_auth(struct intel_connector *connector) if (ret) return ret; if (!hdcp_capable) { - drm_dbg_kms(&dev_priv->drm, + drm_dbg_kms(&i915->drm, "Panel is not HDCP capable\n"); return -EINVAL; } @@ -779,24 +743,24 @@ static int intel_hdcp_auth(struct intel_connector *connector) /* Initialize An with 2 random values and acquire it */ for (i = 0; i < 2; i++) - intel_de_write(dev_priv, - HDCP_ANINIT(dev_priv, cpu_transcoder, port), + intel_de_write(i915, + HDCP_ANINIT(i915, cpu_transcoder, port), get_random_u32()); - intel_de_write(dev_priv, HDCP_CONF(dev_priv, cpu_transcoder, port), + intel_de_write(i915, HDCP_CONF(i915, cpu_transcoder, port), HDCP_CONF_CAPTURE_AN); /* Wait for An to be acquired */ - if (intel_de_wait_for_set(dev_priv, - HDCP_STATUS(dev_priv, cpu_transcoder, port), + if (intel_de_wait_for_set(i915, + HDCP_STATUS(i915, cpu_transcoder, port), HDCP_STATUS_AN_READY, 1)) { - drm_err(&dev_priv->drm, "Timed out waiting for An\n"); + drm_err(&i915->drm, "Timed out waiting for An\n"); return -ETIMEDOUT; } - an.reg[0] = intel_de_read(dev_priv, - HDCP_ANLO(dev_priv, cpu_transcoder, port)); - an.reg[1] = intel_de_read(dev_priv, - HDCP_ANHI(dev_priv, cpu_transcoder, port)); + an.reg[0] = intel_de_read(i915, + HDCP_ANLO(i915, cpu_transcoder, port)); + an.reg[1] = intel_de_read(i915, + HDCP_ANHI(i915, cpu_transcoder, port)); ret = shim->write_an_aksv(dig_port, an.shim); if (ret) return ret; @@ -809,34 +773,34 @@ static int intel_hdcp_auth(struct intel_connector *connector) if (ret < 0) return ret; - if (drm_hdcp_check_ksvs_revoked(&dev_priv->drm, bksv.shim, 1) > 0) { - drm_err(&dev_priv->drm, "BKSV is revoked\n"); + if (drm_hdcp_check_ksvs_revoked(&i915->drm, bksv.shim, 1) > 0) { + drm_err(&i915->drm, "BKSV is revoked\n"); return -EPERM; } - intel_de_write(dev_priv, HDCP_BKSVLO(dev_priv, cpu_transcoder, port), + intel_de_write(i915, HDCP_BKSVLO(i915, cpu_transcoder, port), bksv.reg[0]); - intel_de_write(dev_priv, HDCP_BKSVHI(dev_priv, cpu_transcoder, port), + intel_de_write(i915, HDCP_BKSVHI(i915, cpu_transcoder, port), bksv.reg[1]); ret = shim->repeater_present(dig_port, &repeater_present); if (ret) return ret; if (repeater_present) - intel_de_write(dev_priv, HDCP_REP_CTL, - intel_hdcp_get_repeater_ctl(dev_priv, cpu_transcoder, port)); + intel_de_write(i915, HDCP_REP_CTL, + intel_hdcp_get_repeater_ctl(i915, cpu_transcoder, port)); ret = shim->toggle_signalling(dig_port, cpu_transcoder, true); if (ret) return ret; - intel_de_write(dev_priv, HDCP_CONF(dev_priv, cpu_transcoder, port), + intel_de_write(i915, HDCP_CONF(i915, cpu_transcoder, port), HDCP_CONF_AUTH_AND_ENC); /* Wait for R0 ready */ - if (wait_for(intel_de_read(dev_priv, HDCP_STATUS(dev_priv, cpu_transcoder, port)) & + if (wait_for(intel_de_read(i915, HDCP_STATUS(i915, cpu_transcoder, port)) & (HDCP_STATUS_R0_READY | HDCP_STATUS_ENC), 1)) { - drm_err(&dev_priv->drm, "Timed out waiting for R0 ready\n"); + drm_err(&i915->drm, "Timed out waiting for R0 ready\n"); return -ETIMEDOUT; } @@ -862,30 +826,30 @@ static int intel_hdcp_auth(struct intel_connector *connector) ret = shim->read_ri_prime(dig_port, ri.shim); if (ret) return ret; - intel_de_write(dev_priv, - HDCP_RPRIME(dev_priv, cpu_transcoder, port), + intel_de_write(i915, + HDCP_RPRIME(i915, cpu_transcoder, port), ri.reg); /* Wait for Ri prime match */ - if (!wait_for(intel_de_read(dev_priv, HDCP_STATUS(dev_priv, cpu_transcoder, port)) & + if (!wait_for(intel_de_read(i915, HDCP_STATUS(i915, cpu_transcoder, port)) & (HDCP_STATUS_RI_MATCH | HDCP_STATUS_ENC), 1)) break; } if (i == tries) { - drm_dbg_kms(&dev_priv->drm, + drm_dbg_kms(&i915->drm, "Timed out waiting for Ri prime match (%x)\n", - intel_de_read(dev_priv, HDCP_STATUS(dev_priv, - cpu_transcoder, port))); + intel_de_read(i915, + HDCP_STATUS(i915, cpu_transcoder, port))); return -ETIMEDOUT; } /* Wait for encryption confirmation */ - if (intel_de_wait_for_set(dev_priv, - HDCP_STATUS(dev_priv, cpu_transcoder, port), + if (intel_de_wait_for_set(i915, + HDCP_STATUS(i915, cpu_transcoder, port), HDCP_STATUS_ENC, HDCP_ENCRYPT_STATUS_CHANGE_TIMEOUT_MS)) { - drm_err(&dev_priv->drm, "Timed out waiting for encryption\n"); + drm_err(&i915->drm, "Timed out waiting for encryption\n"); return -ETIMEDOUT; } @@ -893,42 +857,42 @@ static int intel_hdcp_auth(struct intel_connector *connector) if (shim->stream_encryption) { ret = shim->stream_encryption(connector, true); if (ret) { - drm_err(&dev_priv->drm, "[%s:%d] Failed to enable HDCP 1.4 stream enc\n", + drm_err(&i915->drm, "[%s:%d] Failed to enable HDCP 1.4 stream enc\n", connector->base.name, connector->base.base.id); return ret; } - drm_dbg_kms(&dev_priv->drm, "HDCP 1.4 transcoder: %s stream encrypted\n", + drm_dbg_kms(&i915->drm, "HDCP 1.4 transcoder: %s stream encrypted\n", transcoder_name(hdcp->stream_transcoder)); } if (repeater_present) return intel_hdcp_auth_downstream(connector); - drm_dbg_kms(&dev_priv->drm, "HDCP is enabled (no repeater present)\n"); + drm_dbg_kms(&i915->drm, "HDCP is enabled (no repeater present)\n"); return 0; } static int _intel_hdcp_disable(struct intel_connector *connector) { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct drm_i915_private *i915 = to_i915(connector->base.dev); struct intel_hdcp *hdcp = &connector->hdcp; enum port port = dig_port->base.port; enum transcoder cpu_transcoder = hdcp->cpu_transcoder; u32 repeater_ctl; int ret; - drm_dbg_kms(&dev_priv->drm, "[%s:%d] HDCP is being disabled...\n", + drm_dbg_kms(&i915->drm, "[%s:%d] HDCP is being disabled...\n", connector->base.name, connector->base.base.id); if (hdcp->shim->stream_encryption) { ret = hdcp->shim->stream_encryption(connector, false); if (ret) { - drm_err(&dev_priv->drm, "[%s:%d] Failed to disable HDCP 1.4 stream enc\n", + drm_err(&i915->drm, "[%s:%d] Failed to disable HDCP 1.4 stream enc\n", connector->base.name, connector->base.base.id); return ret; } - drm_dbg_kms(&dev_priv->drm, "HDCP 1.4 transcoder: %s stream encryption disabled\n", + drm_dbg_kms(&i915->drm, "HDCP 1.4 transcoder: %s stream encryption disabled\n", transcoder_name(hdcp->stream_transcoder)); /* * If there are other connectors on this port using HDCP, @@ -940,51 +904,51 @@ static int _intel_hdcp_disable(struct intel_connector *connector) } hdcp->hdcp_encrypted = false; - intel_de_write(dev_priv, HDCP_CONF(dev_priv, cpu_transcoder, port), 0); - if (intel_de_wait_for_clear(dev_priv, - HDCP_STATUS(dev_priv, cpu_transcoder, port), + intel_de_write(i915, HDCP_CONF(i915, cpu_transcoder, port), 0); + if (intel_de_wait_for_clear(i915, + HDCP_STATUS(i915, cpu_transcoder, port), ~0, HDCP_ENCRYPT_STATUS_CHANGE_TIMEOUT_MS)) { - drm_err(&dev_priv->drm, + drm_err(&i915->drm, "Failed to disable HDCP, timeout clearing status\n"); return -ETIMEDOUT; } - repeater_ctl = intel_hdcp_get_repeater_ctl(dev_priv, cpu_transcoder, + repeater_ctl = intel_hdcp_get_repeater_ctl(i915, cpu_transcoder, port); - intel_de_rmw(dev_priv, HDCP_REP_CTL, repeater_ctl, 0); + intel_de_rmw(i915, HDCP_REP_CTL, repeater_ctl, 0); ret = hdcp->shim->toggle_signalling(dig_port, cpu_transcoder, false); if (ret) { - drm_err(&dev_priv->drm, "Failed to disable HDCP signalling\n"); + drm_err(&i915->drm, "Failed to disable HDCP signalling\n"); return ret; } - drm_dbg_kms(&dev_priv->drm, "HDCP is disabled\n"); + drm_dbg_kms(&i915->drm, "HDCP is disabled\n"); return 0; } static int _intel_hdcp_enable(struct intel_connector *connector) { - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct drm_i915_private *i915 = to_i915(connector->base.dev); struct intel_hdcp *hdcp = &connector->hdcp; int i, ret, tries = 3; - drm_dbg_kms(&dev_priv->drm, "[%s:%d] HDCP is being enabled...\n", + drm_dbg_kms(&i915->drm, "[%s:%d] HDCP is being enabled...\n", connector->base.name, connector->base.base.id); - if (!hdcp_key_loadable(dev_priv)) { - drm_err(&dev_priv->drm, "HDCP key Load is not possible\n"); + if (!hdcp_key_loadable(i915)) { + drm_err(&i915->drm, "HDCP key Load is not possible\n"); return -ENXIO; } for (i = 0; i < KEY_LOAD_TRIES; i++) { - ret = intel_hdcp_load_keys(dev_priv); + ret = intel_hdcp_load_keys(i915); if (!ret) break; - intel_hdcp_clear_keys(dev_priv); + intel_hdcp_clear_keys(i915); } if (ret) { - drm_err(&dev_priv->drm, "Could not load HDCP keys, (%d)\n", + drm_err(&i915->drm, "Could not load HDCP keys, (%d)\n", ret); return ret; } @@ -997,13 +961,13 @@ static int _intel_hdcp_enable(struct intel_connector *connector) return 0; } - drm_dbg_kms(&dev_priv->drm, "HDCP Auth failure (%d)\n", ret); + drm_dbg_kms(&i915->drm, "HDCP Auth failure (%d)\n", ret); /* Ensuring HDCP encryption and signalling are stopped. */ _intel_hdcp_disable(connector); } - drm_dbg_kms(&dev_priv->drm, + drm_dbg_kms(&i915->drm, "HDCP authentication failed (%d tries/%d)\n", tries, ret); return ret; } @@ -1019,6 +983,7 @@ static void intel_hdcp_update_value(struct intel_connector *connector, struct drm_device *dev = connector->base.dev; struct intel_digital_port *dig_port = intel_attached_dig_port(connector); struct intel_hdcp *hdcp = &connector->hdcp; + struct drm_i915_private *i915 = to_i915(connector->base.dev); drm_WARN_ON(connector->base.dev, !mutex_is_locked(&hdcp->mutex)); @@ -1037,7 +1002,7 @@ static void intel_hdcp_update_value(struct intel_connector *connector, hdcp->value = value; if (update_property) { drm_connector_get(&connector->base); - schedule_work(&hdcp->prop_work); + queue_work(i915->unordered_wq, &hdcp->prop_work); } } @@ -1045,7 +1010,7 @@ static void intel_hdcp_update_value(struct intel_connector *connector, static int intel_hdcp_check_link(struct intel_connector *connector) { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct drm_i915_private *i915 = to_i915(connector->base.dev); struct intel_hdcp *hdcp = &connector->hdcp; enum port port = dig_port->base.port; enum transcoder cpu_transcoder; @@ -1063,12 +1028,12 @@ static int intel_hdcp_check_link(struct intel_connector *connector) goto out; } - if (drm_WARN_ON(&dev_priv->drm, - !intel_hdcp_in_use(dev_priv, cpu_transcoder, port))) { - drm_err(&dev_priv->drm, + if (drm_WARN_ON(&i915->drm, + !intel_hdcp_in_use(i915, cpu_transcoder, port))) { + drm_err(&i915->drm, "%s:%d HDCP link stopped encryption,%x\n", connector->base.name, connector->base.base.id, - intel_de_read(dev_priv, HDCP_STATUS(dev_priv, cpu_transcoder, port))); + intel_de_read(i915, HDCP_STATUS(i915, cpu_transcoder, port))); ret = -ENXIO; intel_hdcp_update_value(connector, DRM_MODE_CONTENT_PROTECTION_DESIRED, @@ -1084,13 +1049,13 @@ static int intel_hdcp_check_link(struct intel_connector *connector) goto out; } - drm_dbg_kms(&dev_priv->drm, + drm_dbg_kms(&i915->drm, "[%s:%d] HDCP link failed, retrying authentication\n", connector->base.name, connector->base.base.id); ret = _intel_hdcp_disable(connector); if (ret) { - drm_err(&dev_priv->drm, "Failed to disable hdcp (%d)\n", ret); + drm_err(&i915->drm, "Failed to disable hdcp (%d)\n", ret); intel_hdcp_update_value(connector, DRM_MODE_CONTENT_PROTECTION_DESIRED, true); @@ -1099,7 +1064,7 @@ static int intel_hdcp_check_link(struct intel_connector *connector) ret = _intel_hdcp_enable(connector); if (ret) { - drm_err(&dev_priv->drm, "Failed to enable hdcp (%d)\n", ret); + drm_err(&i915->drm, "Failed to enable hdcp (%d)\n", ret); intel_hdcp_update_value(connector, DRM_MODE_CONTENT_PROTECTION_DESIRED, true); @@ -1117,9 +1082,9 @@ static void intel_hdcp_prop_work(struct work_struct *work) struct intel_hdcp *hdcp = container_of(work, struct intel_hdcp, prop_work); struct intel_connector *connector = intel_hdcp_to_connector(hdcp); - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct drm_i915_private *i915 = to_i915(connector->base.dev); - drm_modeset_lock(&dev_priv->drm.mode_config.connection_mutex, NULL); + drm_modeset_lock(&i915->drm.mode_config.connection_mutex, NULL); mutex_lock(&hdcp->mutex); /* @@ -1132,15 +1097,15 @@ static void intel_hdcp_prop_work(struct work_struct *work) hdcp->value); mutex_unlock(&hdcp->mutex); - drm_modeset_unlock(&dev_priv->drm.mode_config.connection_mutex); + drm_modeset_unlock(&i915->drm.mode_config.connection_mutex); drm_connector_put(&connector->base); } -bool is_hdcp_supported(struct drm_i915_private *dev_priv, enum port port) +bool is_hdcp_supported(struct drm_i915_private *i915, enum port port) { - return RUNTIME_INFO(dev_priv)->has_hdcp && - (DISPLAY_VER(dev_priv) >= 12 || port < PORT_E); + return DISPLAY_RUNTIME_INFO(i915)->has_hdcp && + (DISPLAY_VER(i915) >= 12 || port < PORT_E); } static int @@ -1149,23 +1114,23 @@ hdcp2_prepare_ake_init(struct intel_connector *connector, { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); struct hdcp_port_data *data = &dig_port->hdcp_port_data; - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); - struct i915_hdcp_master *arbiter; + struct drm_i915_private *i915 = to_i915(connector->base.dev); + struct i915_hdcp_arbiter *arbiter; int ret; - mutex_lock(&dev_priv->display.hdcp.comp_mutex); - arbiter = dev_priv->display.hdcp.master; + mutex_lock(&i915->display.hdcp.hdcp_mutex); + arbiter = i915->display.hdcp.arbiter; if (!arbiter || !arbiter->ops) { - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return -EINVAL; } ret = arbiter->ops->initiate_hdcp2_session(arbiter->hdcp_dev, data, ake_data); if (ret) - drm_dbg_kms(&dev_priv->drm, "Prepare_ake_init failed. %d\n", + drm_dbg_kms(&i915->drm, "Prepare_ake_init failed. %d\n", ret); - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return ret; } @@ -1179,15 +1144,15 @@ hdcp2_verify_rx_cert_prepare_km(struct intel_connector *connector, { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); struct hdcp_port_data *data = &dig_port->hdcp_port_data; - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); - struct i915_hdcp_master *arbiter; + struct drm_i915_private *i915 = to_i915(connector->base.dev); + struct i915_hdcp_arbiter *arbiter; int ret; - mutex_lock(&dev_priv->display.hdcp.comp_mutex); - arbiter = dev_priv->display.hdcp.master; + mutex_lock(&i915->display.hdcp.hdcp_mutex); + arbiter = i915->display.hdcp.arbiter; if (!arbiter || !arbiter->ops) { - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return -EINVAL; } @@ -1195,9 +1160,9 @@ hdcp2_verify_rx_cert_prepare_km(struct intel_connector *connector, rx_cert, paired, ek_pub_km, msg_sz); if (ret < 0) - drm_dbg_kms(&dev_priv->drm, "Verify rx_cert failed. %d\n", + drm_dbg_kms(&i915->drm, "Verify rx_cert failed. %d\n", ret); - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return ret; } @@ -1207,22 +1172,22 @@ static int hdcp2_verify_hprime(struct intel_connector *connector, { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); struct hdcp_port_data *data = &dig_port->hdcp_port_data; - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); - struct i915_hdcp_master *arbiter; + struct drm_i915_private *i915 = to_i915(connector->base.dev); + struct i915_hdcp_arbiter *arbiter; int ret; - mutex_lock(&dev_priv->display.hdcp.comp_mutex); - arbiter = dev_priv->display.hdcp.master; + mutex_lock(&i915->display.hdcp.hdcp_mutex); + arbiter = i915->display.hdcp.arbiter; if (!arbiter || !arbiter->ops) { - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return -EINVAL; } ret = arbiter->ops->verify_hprime(arbiter->hdcp_dev, data, rx_hprime); if (ret < 0) - drm_dbg_kms(&dev_priv->drm, "Verify hprime failed. %d\n", ret); - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + drm_dbg_kms(&i915->drm, "Verify hprime failed. %d\n", ret); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return ret; } @@ -1233,23 +1198,23 @@ hdcp2_store_pairing_info(struct intel_connector *connector, { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); struct hdcp_port_data *data = &dig_port->hdcp_port_data; - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); - struct i915_hdcp_master *arbiter; + struct drm_i915_private *i915 = to_i915(connector->base.dev); + struct i915_hdcp_arbiter *arbiter; int ret; - mutex_lock(&dev_priv->display.hdcp.comp_mutex); - arbiter = dev_priv->display.hdcp.master; + mutex_lock(&i915->display.hdcp.hdcp_mutex); + arbiter = i915->display.hdcp.arbiter; if (!arbiter || !arbiter->ops) { - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return -EINVAL; } ret = arbiter->ops->store_pairing_info(arbiter->hdcp_dev, data, pairing_info); if (ret < 0) - drm_dbg_kms(&dev_priv->drm, "Store pairing info failed. %d\n", + drm_dbg_kms(&i915->drm, "Store pairing info failed. %d\n", ret); - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return ret; } @@ -1260,23 +1225,23 @@ hdcp2_prepare_lc_init(struct intel_connector *connector, { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); struct hdcp_port_data *data = &dig_port->hdcp_port_data; - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); - struct i915_hdcp_master *arbiter; + struct drm_i915_private *i915 = to_i915(connector->base.dev); + struct i915_hdcp_arbiter *arbiter; int ret; - mutex_lock(&dev_priv->display.hdcp.comp_mutex); - arbiter = dev_priv->display.hdcp.master; + mutex_lock(&i915->display.hdcp.hdcp_mutex); + arbiter = i915->display.hdcp.arbiter; if (!arbiter || !arbiter->ops) { - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return -EINVAL; } ret = arbiter->ops->initiate_locality_check(arbiter->hdcp_dev, data, lc_init); if (ret < 0) - drm_dbg_kms(&dev_priv->drm, "Prepare lc_init failed. %d\n", + drm_dbg_kms(&i915->drm, "Prepare lc_init failed. %d\n", ret); - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return ret; } @@ -1287,23 +1252,23 @@ hdcp2_verify_lprime(struct intel_connector *connector, { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); struct hdcp_port_data *data = &dig_port->hdcp_port_data; - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); - struct i915_hdcp_master *arbiter; + struct drm_i915_private *i915 = to_i915(connector->base.dev); + struct i915_hdcp_arbiter *arbiter; int ret; - mutex_lock(&dev_priv->display.hdcp.comp_mutex); - arbiter = dev_priv->display.hdcp.master; + mutex_lock(&i915->display.hdcp.hdcp_mutex); + arbiter = i915->display.hdcp.arbiter; if (!arbiter || !arbiter->ops) { - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return -EINVAL; } ret = arbiter->ops->verify_lprime(arbiter->hdcp_dev, data, rx_lprime); if (ret < 0) - drm_dbg_kms(&dev_priv->drm, "Verify L_Prime failed. %d\n", + drm_dbg_kms(&i915->drm, "Verify L_Prime failed. %d\n", ret); - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return ret; } @@ -1313,23 +1278,23 @@ static int hdcp2_prepare_skey(struct intel_connector *connector, { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); struct hdcp_port_data *data = &dig_port->hdcp_port_data; - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); - struct i915_hdcp_master *arbiter; + struct drm_i915_private *i915 = to_i915(connector->base.dev); + struct i915_hdcp_arbiter *arbiter; int ret; - mutex_lock(&dev_priv->display.hdcp.comp_mutex); - arbiter = dev_priv->display.hdcp.master; + mutex_lock(&i915->display.hdcp.hdcp_mutex); + arbiter = i915->display.hdcp.arbiter; if (!arbiter || !arbiter->ops) { - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return -EINVAL; } ret = arbiter->ops->get_session_key(arbiter->hdcp_dev, data, ske_data); if (ret < 0) - drm_dbg_kms(&dev_priv->drm, "Get session key failed. %d\n", + drm_dbg_kms(&i915->drm, "Get session key failed. %d\n", ret); - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return ret; } @@ -1342,15 +1307,15 @@ hdcp2_verify_rep_topology_prepare_ack(struct intel_connector *connector, { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); struct hdcp_port_data *data = &dig_port->hdcp_port_data; - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); - struct i915_hdcp_master *arbiter; + struct drm_i915_private *i915 = to_i915(connector->base.dev); + struct i915_hdcp_arbiter *arbiter; int ret; - mutex_lock(&dev_priv->display.hdcp.comp_mutex); - arbiter = dev_priv->display.hdcp.master; + mutex_lock(&i915->display.hdcp.hdcp_mutex); + arbiter = i915->display.hdcp.arbiter; if (!arbiter || !arbiter->ops) { - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return -EINVAL; } @@ -1359,9 +1324,9 @@ hdcp2_verify_rep_topology_prepare_ack(struct intel_connector *connector, rep_topology, rep_send_ack); if (ret < 0) - drm_dbg_kms(&dev_priv->drm, + drm_dbg_kms(&i915->drm, "Verify rep topology failed. %d\n", ret); - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return ret; } @@ -1372,22 +1337,22 @@ hdcp2_verify_mprime(struct intel_connector *connector, { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); struct hdcp_port_data *data = &dig_port->hdcp_port_data; - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); - struct i915_hdcp_master *arbiter; + struct drm_i915_private *i915 = to_i915(connector->base.dev); + struct i915_hdcp_arbiter *arbiter; int ret; - mutex_lock(&dev_priv->display.hdcp.comp_mutex); - arbiter = dev_priv->display.hdcp.master; + mutex_lock(&i915->display.hdcp.hdcp_mutex); + arbiter = i915->display.hdcp.arbiter; if (!arbiter || !arbiter->ops) { - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return -EINVAL; } ret = arbiter->ops->verify_mprime(arbiter->hdcp_dev, data, stream_ready); if (ret < 0) - drm_dbg_kms(&dev_priv->drm, "Verify mprime failed. %d\n", ret); - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + drm_dbg_kms(&i915->drm, "Verify mprime failed. %d\n", ret); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return ret; } @@ -1396,23 +1361,23 @@ static int hdcp2_authenticate_port(struct intel_connector *connector) { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); struct hdcp_port_data *data = &dig_port->hdcp_port_data; - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); - struct i915_hdcp_master *arbiter; + struct drm_i915_private *i915 = to_i915(connector->base.dev); + struct i915_hdcp_arbiter *arbiter; int ret; - mutex_lock(&dev_priv->display.hdcp.comp_mutex); - arbiter = dev_priv->display.hdcp.master; + mutex_lock(&i915->display.hdcp.hdcp_mutex); + arbiter = i915->display.hdcp.arbiter; if (!arbiter || !arbiter->ops) { - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return -EINVAL; } ret = arbiter->ops->enable_hdcp_authentication(arbiter->hdcp_dev, data); if (ret < 0) - drm_dbg_kms(&dev_priv->drm, "Enable hdcp auth failed. %d\n", + drm_dbg_kms(&i915->drm, "Enable hdcp auth failed. %d\n", ret); - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return ret; } @@ -1420,21 +1385,21 @@ static int hdcp2_authenticate_port(struct intel_connector *connector) static int hdcp2_close_session(struct intel_connector *connector) { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); - struct i915_hdcp_master *arbiter; + struct drm_i915_private *i915 = to_i915(connector->base.dev); + struct i915_hdcp_arbiter *arbiter; int ret; - mutex_lock(&dev_priv->display.hdcp.comp_mutex); - arbiter = dev_priv->display.hdcp.master; + mutex_lock(&i915->display.hdcp.hdcp_mutex); + arbiter = i915->display.hdcp.arbiter; if (!arbiter || !arbiter->ops) { - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return -EINVAL; } ret = arbiter->ops->close_hdcp_session(arbiter->hdcp_dev, &dig_port->hdcp_port_data); - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return ret; } @@ -1448,7 +1413,7 @@ static int hdcp2_deauthenticate_port(struct intel_connector *connector) static int hdcp2_authentication_key_exchange(struct intel_connector *connector) { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct drm_i915_private *i915 = to_i915(connector->base.dev); struct intel_hdcp *hdcp = &connector->hdcp; union { struct hdcp2_ake_init ake_init; @@ -1480,16 +1445,16 @@ static int hdcp2_authentication_key_exchange(struct intel_connector *connector) return ret; if (msgs.send_cert.rx_caps[0] != HDCP_2_2_RX_CAPS_VERSION_VAL) { - drm_dbg_kms(&dev_priv->drm, "cert.rx_caps dont claim HDCP2.2\n"); + drm_dbg_kms(&i915->drm, "cert.rx_caps dont claim HDCP2.2\n"); return -EINVAL; } hdcp->is_repeater = HDCP_2_2_RX_REPEATER(msgs.send_cert.rx_caps[2]); - if (drm_hdcp_check_ksvs_revoked(&dev_priv->drm, + if (drm_hdcp_check_ksvs_revoked(&i915->drm, msgs.send_cert.cert_rx.receiver_id, 1) > 0) { - drm_err(&dev_priv->drm, "Receiver ID is revoked\n"); + drm_err(&i915->drm, "Receiver ID is revoked\n"); return -EPERM; } @@ -1643,7 +1608,7 @@ static int hdcp2_authenticate_repeater_topology(struct intel_connector *connector) { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct drm_i915_private *i915 = to_i915(connector->base.dev); struct intel_hdcp *hdcp = &connector->hdcp; union { struct hdcp2_rep_send_receiverid_list recvid_list; @@ -1663,7 +1628,7 @@ int hdcp2_authenticate_repeater_topology(struct intel_connector *connector) if (HDCP_2_2_MAX_CASCADE_EXCEEDED(rx_info[1]) || HDCP_2_2_MAX_DEVS_EXCEEDED(rx_info[1])) { - drm_dbg_kms(&dev_priv->drm, "Topology Max Size Exceeded\n"); + drm_dbg_kms(&i915->drm, "Topology Max Size Exceeded\n"); return -EINVAL; } @@ -1680,23 +1645,23 @@ int hdcp2_authenticate_repeater_topology(struct intel_connector *connector) drm_hdcp_be24_to_cpu((const u8 *)msgs.recvid_list.seq_num_v); if (!hdcp->hdcp2_encrypted && seq_num_v) { - drm_dbg_kms(&dev_priv->drm, + drm_dbg_kms(&i915->drm, "Non zero Seq_num_v at first RecvId_List msg\n"); return -EINVAL; } if (seq_num_v < hdcp->seq_num_v) { /* Roll over of the seq_num_v from repeater. Reauthenticate. */ - drm_dbg_kms(&dev_priv->drm, "Seq_num_v roll over.\n"); + drm_dbg_kms(&i915->drm, "Seq_num_v roll over.\n"); return -EINVAL; } device_cnt = (HDCP_2_2_DEV_COUNT_HI(rx_info[0]) << 4 | HDCP_2_2_DEV_COUNT_LO(rx_info[1])); - if (drm_hdcp_check_ksvs_revoked(&dev_priv->drm, + if (drm_hdcp_check_ksvs_revoked(&i915->drm, msgs.recvid_list.receiver_ids, device_cnt) > 0) { - drm_err(&dev_priv->drm, "Revoked receiver ID(s) is in list\n"); + drm_err(&i915->drm, "Revoked receiver ID(s) is in list\n"); return -EPERM; } @@ -1765,16 +1730,16 @@ static int hdcp2_authenticate_sink(struct intel_connector *connector) static int hdcp2_enable_stream_encryption(struct intel_connector *connector) { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct drm_i915_private *i915 = to_i915(connector->base.dev); struct hdcp_port_data *data = &dig_port->hdcp_port_data; struct intel_hdcp *hdcp = &connector->hdcp; enum transcoder cpu_transcoder = hdcp->cpu_transcoder; enum port port = dig_port->base.port; int ret = 0; - if (!(intel_de_read(dev_priv, HDCP2_STATUS(dev_priv, cpu_transcoder, port)) & + if (!(intel_de_read(i915, HDCP2_STATUS(i915, cpu_transcoder, port)) & LINK_ENCRYPTION_STATUS)) { - drm_err(&dev_priv->drm, "[%s:%d] HDCP 2.2 Link is not encrypted\n", + drm_err(&i915->drm, "[%s:%d] HDCP 2.2 Link is not encrypted\n", connector->base.name, connector->base.base.id); ret = -EPERM; goto link_recover; @@ -1783,11 +1748,11 @@ static int hdcp2_enable_stream_encryption(struct intel_connector *connector) if (hdcp->shim->stream_2_2_encryption) { ret = hdcp->shim->stream_2_2_encryption(connector, true); if (ret) { - drm_err(&dev_priv->drm, "[%s:%d] Failed to enable HDCP 2.2 stream enc\n", + drm_err(&i915->drm, "[%s:%d] Failed to enable HDCP 2.2 stream enc\n", connector->base.name, connector->base.base.id); return ret; } - drm_dbg_kms(&dev_priv->drm, "HDCP 2.2 transcoder: %s stream encrypted\n", + drm_dbg_kms(&i915->drm, "HDCP 2.2 transcoder: %s stream encrypted\n", transcoder_name(hdcp->stream_transcoder)); } @@ -1795,7 +1760,7 @@ static int hdcp2_enable_stream_encryption(struct intel_connector *connector) link_recover: if (hdcp2_deauthenticate_port(connector) < 0) - drm_dbg_kms(&dev_priv->drm, "Port deauth failed.\n"); + drm_dbg_kms(&i915->drm, "Port deauth failed.\n"); dig_port->hdcp_auth_status = false; data->k = 0; @@ -1806,34 +1771,34 @@ static int hdcp2_enable_stream_encryption(struct intel_connector *connector) static int hdcp2_enable_encryption(struct intel_connector *connector) { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct drm_i915_private *i915 = to_i915(connector->base.dev); struct intel_hdcp *hdcp = &connector->hdcp; enum port port = dig_port->base.port; enum transcoder cpu_transcoder = hdcp->cpu_transcoder; int ret; - drm_WARN_ON(&dev_priv->drm, - intel_de_read(dev_priv, HDCP2_STATUS(dev_priv, cpu_transcoder, port)) & + drm_WARN_ON(&i915->drm, + intel_de_read(i915, HDCP2_STATUS(i915, cpu_transcoder, port)) & LINK_ENCRYPTION_STATUS); if (hdcp->shim->toggle_signalling) { ret = hdcp->shim->toggle_signalling(dig_port, cpu_transcoder, true); if (ret) { - drm_err(&dev_priv->drm, + drm_err(&i915->drm, "Failed to enable HDCP signalling. %d\n", ret); return ret; } } - if (intel_de_read(dev_priv, HDCP2_STATUS(dev_priv, cpu_transcoder, port)) & + if (intel_de_read(i915, HDCP2_STATUS(i915, cpu_transcoder, port)) & LINK_AUTH_STATUS) /* Link is Authenticated. Now set for Encryption */ - intel_de_rmw(dev_priv, HDCP2_CTL(dev_priv, cpu_transcoder, port), + intel_de_rmw(i915, HDCP2_CTL(i915, cpu_transcoder, port), 0, CTL_LINK_ENCRYPTION_REQ); - ret = intel_de_wait_for_set(dev_priv, - HDCP2_STATUS(dev_priv, cpu_transcoder, + ret = intel_de_wait_for_set(i915, + HDCP2_STATUS(i915, cpu_transcoder, port), LINK_ENCRYPTION_STATUS, HDCP_ENCRYPT_STATUS_CHANGE_TIMEOUT_MS); @@ -1845,31 +1810,31 @@ static int hdcp2_enable_encryption(struct intel_connector *connector) static int hdcp2_disable_encryption(struct intel_connector *connector) { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct drm_i915_private *i915 = to_i915(connector->base.dev); struct intel_hdcp *hdcp = &connector->hdcp; enum port port = dig_port->base.port; enum transcoder cpu_transcoder = hdcp->cpu_transcoder; int ret; - drm_WARN_ON(&dev_priv->drm, !(intel_de_read(dev_priv, HDCP2_STATUS(dev_priv, cpu_transcoder, port)) & + drm_WARN_ON(&i915->drm, !(intel_de_read(i915, HDCP2_STATUS(i915, cpu_transcoder, port)) & LINK_ENCRYPTION_STATUS)); - intel_de_rmw(dev_priv, HDCP2_CTL(dev_priv, cpu_transcoder, port), + intel_de_rmw(i915, HDCP2_CTL(i915, cpu_transcoder, port), CTL_LINK_ENCRYPTION_REQ, 0); - ret = intel_de_wait_for_clear(dev_priv, - HDCP2_STATUS(dev_priv, cpu_transcoder, + ret = intel_de_wait_for_clear(i915, + HDCP2_STATUS(i915, cpu_transcoder, port), LINK_ENCRYPTION_STATUS, HDCP_ENCRYPT_STATUS_CHANGE_TIMEOUT_MS); if (ret == -ETIMEDOUT) - drm_dbg_kms(&dev_priv->drm, "Disable Encryption Timedout"); + drm_dbg_kms(&i915->drm, "Disable Encryption Timedout"); if (hdcp->shim->toggle_signalling) { ret = hdcp->shim->toggle_signalling(dig_port, cpu_transcoder, false); if (ret) { - drm_err(&dev_priv->drm, + drm_err(&i915->drm, "Failed to disable HDCP signalling. %d\n", ret); return ret; @@ -1917,13 +1882,7 @@ static int hdcp2_authenticate_and_encrypt(struct intel_connector *connector) for (i = 0; i < tries && !dig_port->hdcp_auth_status; i++) { ret = hdcp2_authenticate_sink(connector); if (!ret) { - ret = intel_hdcp_prepare_streams(connector); - if (ret) { - drm_dbg_kms(&i915->drm, - "Prepare streams failed.(%d)\n", - ret); - break; - } + intel_hdcp_prepare_streams(connector); ret = hdcp2_propagate_stream_management_info(connector); if (ret) { @@ -2035,7 +1994,7 @@ _intel_hdcp2_disable(struct intel_connector *connector, bool hdcp2_link_recovery static int intel_hdcp2_check_link(struct intel_connector *connector) { struct intel_digital_port *dig_port = intel_attached_dig_port(connector); - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct drm_i915_private *i915 = to_i915(connector->base.dev); struct intel_hdcp *hdcp = &connector->hdcp; enum port port = dig_port->base.port; enum transcoder cpu_transcoder; @@ -2052,11 +2011,11 @@ static int intel_hdcp2_check_link(struct intel_connector *connector) goto out; } - if (drm_WARN_ON(&dev_priv->drm, - !intel_hdcp2_in_use(dev_priv, cpu_transcoder, port))) { - drm_err(&dev_priv->drm, + if (drm_WARN_ON(&i915->drm, + !intel_hdcp2_in_use(i915, cpu_transcoder, port))) { + drm_err(&i915->drm, "HDCP2.2 link stopped the encryption, %x\n", - intel_de_read(dev_priv, HDCP2_STATUS(dev_priv, cpu_transcoder, port))); + intel_de_read(i915, HDCP2_STATUS(i915, cpu_transcoder, port))); ret = -ENXIO; _intel_hdcp2_disable(connector, true); intel_hdcp_update_value(connector, @@ -2079,7 +2038,7 @@ static int intel_hdcp2_check_link(struct intel_connector *connector) if (hdcp->value == DRM_MODE_CONTENT_PROTECTION_UNDESIRED) goto out; - drm_dbg_kms(&dev_priv->drm, + drm_dbg_kms(&i915->drm, "HDCP2.2 Downstream topology change\n"); ret = hdcp2_authenticate_repeater_topology(connector); if (!ret) { @@ -2088,19 +2047,19 @@ static int intel_hdcp2_check_link(struct intel_connector *connector) true); goto out; } - drm_dbg_kms(&dev_priv->drm, + drm_dbg_kms(&i915->drm, "[%s:%d] Repeater topology auth failed.(%d)\n", connector->base.name, connector->base.base.id, ret); } else { - drm_dbg_kms(&dev_priv->drm, + drm_dbg_kms(&i915->drm, "[%s:%d] HDCP2.2 link failed, retrying auth\n", connector->base.name, connector->base.base.id); } ret = _intel_hdcp2_disable(connector, true); if (ret) { - drm_err(&dev_priv->drm, + drm_err(&i915->drm, "[%s:%d] Failed to disable hdcp2.2 (%d)\n", connector->base.name, connector->base.base.id, ret); intel_hdcp_update_value(connector, @@ -2110,7 +2069,7 @@ static int intel_hdcp2_check_link(struct intel_connector *connector) ret = _intel_hdcp2_enable(connector); if (ret) { - drm_dbg_kms(&dev_priv->drm, + drm_dbg_kms(&i915->drm, "[%s:%d] Failed to enable hdcp2.2 (%d)\n", connector->base.name, connector->base.base.id, ret); @@ -2132,28 +2091,29 @@ static void intel_hdcp_check_work(struct work_struct *work) struct intel_hdcp, check_work); struct intel_connector *connector = intel_hdcp_to_connector(hdcp); + struct drm_i915_private *i915 = to_i915(connector->base.dev); if (drm_connector_is_unregistered(&connector->base)) return; if (!intel_hdcp2_check_link(connector)) - schedule_delayed_work(&hdcp->check_work, - DRM_HDCP2_CHECK_PERIOD_MS); + queue_delayed_work(i915->unordered_wq, &hdcp->check_work, + DRM_HDCP2_CHECK_PERIOD_MS); else if (!intel_hdcp_check_link(connector)) - schedule_delayed_work(&hdcp->check_work, - DRM_HDCP_CHECK_PERIOD_MS); + queue_delayed_work(i915->unordered_wq, &hdcp->check_work, + DRM_HDCP_CHECK_PERIOD_MS); } static int i915_hdcp_component_bind(struct device *i915_kdev, struct device *mei_kdev, void *data) { - struct drm_i915_private *dev_priv = kdev_to_i915(i915_kdev); + struct drm_i915_private *i915 = kdev_to_i915(i915_kdev); - drm_dbg(&dev_priv->drm, "I915 HDCP comp bind\n"); - mutex_lock(&dev_priv->display.hdcp.comp_mutex); - dev_priv->display.hdcp.master = (struct i915_hdcp_master *)data; - dev_priv->display.hdcp.master->hdcp_dev = mei_kdev; - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + drm_dbg(&i915->drm, "I915 HDCP comp bind\n"); + mutex_lock(&i915->display.hdcp.hdcp_mutex); + i915->display.hdcp.arbiter = (struct i915_hdcp_arbiter *)data; + i915->display.hdcp.arbiter->hdcp_dev = mei_kdev; + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return 0; } @@ -2161,12 +2121,12 @@ static int i915_hdcp_component_bind(struct device *i915_kdev, static void i915_hdcp_component_unbind(struct device *i915_kdev, struct device *mei_kdev, void *data) { - struct drm_i915_private *dev_priv = kdev_to_i915(i915_kdev); + struct drm_i915_private *i915 = kdev_to_i915(i915_kdev); - drm_dbg(&dev_priv->drm, "I915 HDCP comp unbind\n"); - mutex_lock(&dev_priv->display.hdcp.comp_mutex); - dev_priv->display.hdcp.master = NULL; - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + drm_dbg(&i915->drm, "I915 HDCP comp unbind\n"); + mutex_lock(&i915->display.hdcp.hdcp_mutex); + i915->display.hdcp.arbiter = NULL; + mutex_unlock(&i915->display.hdcp.hdcp_mutex); } static const struct component_ops i915_hdcp_ops = { @@ -2200,12 +2160,11 @@ static int initialize_hdcp_port_data(struct intel_connector *connector, struct intel_digital_port *dig_port, const struct intel_hdcp_shim *shim) { - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct drm_i915_private *i915 = to_i915(connector->base.dev); struct hdcp_port_data *data = &dig_port->hdcp_port_data; - struct intel_hdcp *hdcp = &connector->hdcp; enum port port = dig_port->base.port; - if (DISPLAY_VER(dev_priv) < 12) + if (DISPLAY_VER(i915) < 12) data->hdcp_ddi = intel_get_hdcp_ddi_index(port); else /* @@ -2225,58 +2184,55 @@ static int initialize_hdcp_port_data(struct intel_connector *connector, data->protocol = (u8)shim->protocol; if (!data->streams) - data->streams = kcalloc(INTEL_NUM_PIPES(dev_priv), + data->streams = kcalloc(INTEL_NUM_PIPES(i915), sizeof(struct hdcp2_streamid_type), GFP_KERNEL); if (!data->streams) { - drm_err(&dev_priv->drm, "Out of Memory\n"); + drm_err(&i915->drm, "Out of Memory\n"); return -ENOMEM; } - /* For SST */ - data->streams[0].stream_id = 0; - data->streams[0].stream_type = hdcp->content_type; return 0; } -static bool is_hdcp2_supported(struct drm_i915_private *dev_priv) +static bool is_hdcp2_supported(struct drm_i915_private *i915) { - if (intel_hdcp_gsc_cs_required(dev_priv)) + if (intel_hdcp_gsc_cs_required(i915)) return true; if (!IS_ENABLED(CONFIG_INTEL_MEI_HDCP)) return false; - return (DISPLAY_VER(dev_priv) >= 10 || - IS_KABYLAKE(dev_priv) || - IS_COFFEELAKE(dev_priv) || - IS_COMETLAKE(dev_priv)); + return (DISPLAY_VER(i915) >= 10 || + IS_KABYLAKE(i915) || + IS_COFFEELAKE(i915) || + IS_COMETLAKE(i915)); } -void intel_hdcp_component_init(struct drm_i915_private *dev_priv) +void intel_hdcp_component_init(struct drm_i915_private *i915) { int ret; - if (!is_hdcp2_supported(dev_priv)) + if (!is_hdcp2_supported(i915)) return; - mutex_lock(&dev_priv->display.hdcp.comp_mutex); - drm_WARN_ON(&dev_priv->drm, dev_priv->display.hdcp.comp_added); + mutex_lock(&i915->display.hdcp.hdcp_mutex); + drm_WARN_ON(&i915->drm, i915->display.hdcp.comp_added); - dev_priv->display.hdcp.comp_added = true; - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); - if (intel_hdcp_gsc_cs_required(dev_priv)) - ret = intel_hdcp_gsc_init(dev_priv); + i915->display.hdcp.comp_added = true; + mutex_unlock(&i915->display.hdcp.hdcp_mutex); + if (intel_hdcp_gsc_cs_required(i915)) + ret = intel_hdcp_gsc_init(i915); else - ret = component_add_typed(dev_priv->drm.dev, &i915_hdcp_ops, + ret = component_add_typed(i915->drm.dev, &i915_hdcp_ops, I915_COMPONENT_HDCP); if (ret < 0) { - drm_dbg_kms(&dev_priv->drm, "Failed at fw component add(%d)\n", + drm_dbg_kms(&i915->drm, "Failed at fw component add(%d)\n", ret); - mutex_lock(&dev_priv->display.hdcp.comp_mutex); - dev_priv->display.hdcp.comp_added = false; - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_lock(&i915->display.hdcp.hdcp_mutex); + i915->display.hdcp.comp_added = false; + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return; } } @@ -2302,14 +2258,14 @@ int intel_hdcp_init(struct intel_connector *connector, struct intel_digital_port *dig_port, const struct intel_hdcp_shim *shim) { - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct drm_i915_private *i915 = to_i915(connector->base.dev); struct intel_hdcp *hdcp = &connector->hdcp; int ret; if (!shim) return -EINVAL; - if (is_hdcp2_supported(dev_priv)) + if (is_hdcp2_supported(i915)) intel_hdcp2_init(connector, dig_port, shim); ret = @@ -2330,10 +2286,60 @@ int intel_hdcp_init(struct intel_connector *connector, return 0; } -int intel_hdcp_enable(struct intel_connector *connector, - const struct intel_crtc_state *pipe_config, u8 content_type) +static int +intel_hdcp_set_streams(struct intel_digital_port *dig_port, + struct intel_atomic_state *state) { - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct drm_connector_list_iter conn_iter; + struct intel_digital_port *conn_dig_port; + struct intel_connector *connector; + struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev); + struct hdcp_port_data *data = &dig_port->hdcp_port_data; + + if (!intel_encoder_is_mst(&dig_port->base)) { + data->k = 1; + data->streams[0].stream_id = 0; + return 0; + } + + data->k = 0; + + drm_connector_list_iter_begin(&i915->drm, &conn_iter); + for_each_intel_connector_iter(connector, &conn_iter) { + if (connector->base.status == connector_status_disconnected) + continue; + + if (!intel_encoder_is_mst(intel_attached_encoder(connector))) + continue; + + conn_dig_port = intel_attached_dig_port(connector); + if (conn_dig_port != dig_port) + continue; + + data->streams[data->k].stream_id = + intel_conn_to_vcpi(&state->base, connector); + data->k++; + + /* if there is only one active stream */ + if (dig_port->dp.active_mst_links <= 1) + break; + } + drm_connector_list_iter_end(&conn_iter); + + if (drm_WARN_ON(&i915->drm, data->k > INTEL_NUM_PIPES(i915) || data->k == 0)) + return -EINVAL; + + return 0; +} + +int intel_hdcp_enable(struct intel_atomic_state *state, + struct intel_encoder *encoder, + const struct intel_crtc_state *pipe_config, + const struct drm_connector_state *conn_state) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + struct intel_connector *connector = + to_intel_connector(conn_state->connector); struct intel_digital_port *dig_port = intel_attached_dig_port(connector); struct intel_hdcp *hdcp = &connector->hdcp; unsigned long check_link_interval = DRM_HDCP_CHECK_PERIOD_MS; @@ -2343,16 +2349,16 @@ int intel_hdcp_enable(struct intel_connector *connector, return -ENOENT; if (!connector->encoder) { - drm_err(&dev_priv->drm, "[%s:%d] encoder is not initialized\n", + drm_err(&i915->drm, "[%s:%d] encoder is not initialized\n", connector->base.name, connector->base.base.id); return -ENODEV; } mutex_lock(&hdcp->mutex); mutex_lock(&dig_port->hdcp_mutex); - drm_WARN_ON(&dev_priv->drm, + drm_WARN_ON(&i915->drm, hdcp->value == DRM_MODE_CONTENT_PROTECTION_ENABLED); - hdcp->content_type = content_type; + hdcp->content_type = (u8)conn_state->content_type; if (intel_crtc_has_type(pipe_config, INTEL_OUTPUT_DP_MST)) { hdcp->cpu_transcoder = pipe_config->mst_master_transcoder; @@ -2362,7 +2368,7 @@ int intel_hdcp_enable(struct intel_connector *connector, hdcp->stream_transcoder = INVALID_TRANSCODER; } - if (DISPLAY_VER(dev_priv) >= 12) + if (DISPLAY_VER(i915) >= 12) dig_port->hdcp_port_data.hdcp_transcoder = intel_get_hdcp_transcoder(hdcp->cpu_transcoder); @@ -2371,9 +2377,17 @@ int intel_hdcp_enable(struct intel_connector *connector, * is capable of HDCP2.2, it is preferred to use HDCP2.2. */ if (intel_hdcp2_capable(connector)) { - ret = _intel_hdcp2_enable(connector); - if (!ret) - check_link_interval = DRM_HDCP2_CHECK_PERIOD_MS; + ret = intel_hdcp_set_streams(dig_port, state); + if (!ret) { + ret = _intel_hdcp2_enable(connector); + if (!ret) + check_link_interval = + DRM_HDCP2_CHECK_PERIOD_MS; + } else { + drm_dbg_kms(&i915->drm, + "Set content streams failed: (%d)\n", + ret); + } } /* @@ -2386,7 +2400,8 @@ int intel_hdcp_enable(struct intel_connector *connector, } if (!ret) { - schedule_delayed_work(&hdcp->check_work, check_link_interval); + queue_delayed_work(i915->unordered_wq, &hdcp->check_work, + check_link_interval); intel_hdcp_update_value(connector, DRM_MODE_CONTENT_PROTECTION_ENABLED, true); @@ -2435,6 +2450,7 @@ void intel_hdcp_update_pipe(struct intel_atomic_state *state, to_intel_connector(conn_state->connector); struct intel_hdcp *hdcp = &connector->hdcp; bool content_protection_type_changed, desired_and_not_enabled = false; + struct drm_i915_private *i915 = to_i915(connector->base.dev); if (!connector->hdcp.shim) return; @@ -2461,7 +2477,7 @@ void intel_hdcp_update_pipe(struct intel_atomic_state *state, mutex_lock(&hdcp->mutex); hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED; drm_connector_get(&connector->base); - schedule_work(&hdcp->prop_work); + queue_work(i915->unordered_wq, &hdcp->prop_work); mutex_unlock(&hdcp->mutex); } @@ -2478,31 +2494,29 @@ void intel_hdcp_update_pipe(struct intel_atomic_state *state, */ if (!desired_and_not_enabled && !content_protection_type_changed) { drm_connector_get(&connector->base); - schedule_work(&hdcp->prop_work); + queue_work(i915->unordered_wq, &hdcp->prop_work); } } if (desired_and_not_enabled || content_protection_type_changed) - intel_hdcp_enable(connector, - crtc_state, - (u8)conn_state->hdcp_content_type); + intel_hdcp_enable(state, encoder, crtc_state, conn_state); } -void intel_hdcp_component_fini(struct drm_i915_private *dev_priv) +void intel_hdcp_component_fini(struct drm_i915_private *i915) { - mutex_lock(&dev_priv->display.hdcp.comp_mutex); - if (!dev_priv->display.hdcp.comp_added) { - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + mutex_lock(&i915->display.hdcp.hdcp_mutex); + if (!i915->display.hdcp.comp_added) { + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return; } - dev_priv->display.hdcp.comp_added = false; - mutex_unlock(&dev_priv->display.hdcp.comp_mutex); + i915->display.hdcp.comp_added = false; + mutex_unlock(&i915->display.hdcp.hdcp_mutex); - if (intel_hdcp_gsc_cs_required(dev_priv)) - intel_hdcp_gsc_fini(dev_priv); + if (intel_hdcp_gsc_cs_required(i915)) + intel_hdcp_gsc_fini(i915); else - component_del(dev_priv->drm.dev, &i915_hdcp_ops); + component_del(i915->drm.dev, &i915_hdcp_ops); } void intel_hdcp_cleanup(struct intel_connector *connector) @@ -2592,6 +2606,7 @@ void intel_hdcp_atomic_check(struct drm_connector *connector, void intel_hdcp_handle_cp_irq(struct intel_connector *connector) { struct intel_hdcp *hdcp = &connector->hdcp; + struct drm_i915_private *i915 = to_i915(connector->base.dev); if (!hdcp->shim) return; @@ -2599,5 +2614,5 @@ void intel_hdcp_handle_cp_irq(struct intel_connector *connector) atomic_inc(&connector->hdcp.cp_irq_count); wake_up_all(&connector->hdcp.cp_irq_queue); - schedule_delayed_work(&hdcp->check_work, 0); + queue_delayed_work(i915->unordered_wq, &hdcp->check_work, 0); } diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.h b/drivers/gpu/drm/i915/display/intel_hdcp.h index 8f53b0c7fe5cfb37dadd4d7511c2db666b48af8e..5997c52a0958975931b5e851106f585e0b33ec6d 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp.h +++ b/drivers/gpu/drm/i915/display/intel_hdcp.h @@ -28,18 +28,20 @@ void intel_hdcp_atomic_check(struct drm_connector *connector, int intel_hdcp_init(struct intel_connector *connector, struct intel_digital_port *dig_port, const struct intel_hdcp_shim *hdcp_shim); -int intel_hdcp_enable(struct intel_connector *connector, - const struct intel_crtc_state *pipe_config, u8 content_type); +int intel_hdcp_enable(struct intel_atomic_state *state, + struct intel_encoder *encoder, + const struct intel_crtc_state *pipe_config, + const struct drm_connector_state *conn_state); int intel_hdcp_disable(struct intel_connector *connector); void intel_hdcp_update_pipe(struct intel_atomic_state *state, struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state, const struct drm_connector_state *conn_state); -bool is_hdcp_supported(struct drm_i915_private *dev_priv, enum port port); +bool is_hdcp_supported(struct drm_i915_private *i915, enum port port); bool intel_hdcp_capable(struct intel_connector *connector); bool intel_hdcp2_capable(struct intel_connector *connector); -void intel_hdcp_component_init(struct drm_i915_private *dev_priv); -void intel_hdcp_component_fini(struct drm_i915_private *dev_priv); +void intel_hdcp_component_init(struct drm_i915_private *i915); +void intel_hdcp_component_fini(struct drm_i915_private *i915); void intel_hdcp_cleanup(struct intel_connector *connector); void intel_hdcp_handle_cp_irq(struct intel_connector *connector); diff --git a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c index 7e52aea6aa177fbe37b23b77a05c6ff5adca69b3..72573ce1d0e983908a7708ae9c4b81ccec72ef31 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c +++ b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c @@ -5,11 +5,11 @@ #include -#include "display/intel_hdcp_gsc.h" #include "gem/i915_gem_region.h" #include "gt/uc/intel_gsc_uc_heci_cmd_submit.h" #include "i915_drv.h" #include "i915_utils.h" +#include "intel_hdcp_gsc.h" bool intel_hdcp_gsc_cs_required(struct drm_i915_private *i915) { @@ -697,19 +697,19 @@ static void intel_hdcp_gsc_free_message(struct drm_i915_private *i915) int intel_hdcp_gsc_init(struct drm_i915_private *i915) { - struct i915_hdcp_master *data; + struct i915_hdcp_arbiter *data; int ret; - data = kzalloc(sizeof(struct i915_hdcp_master), GFP_KERNEL); + data = kzalloc(sizeof(struct i915_hdcp_arbiter), GFP_KERNEL); if (!data) return -ENOMEM; - mutex_lock(&i915->display.hdcp.comp_mutex); - i915->display.hdcp.master = data; - i915->display.hdcp.master->hdcp_dev = i915->drm.dev; - i915->display.hdcp.master->ops = &gsc_hdcp_ops; + mutex_lock(&i915->display.hdcp.hdcp_mutex); + i915->display.hdcp.arbiter = data; + i915->display.hdcp.arbiter->hdcp_dev = i915->drm.dev; + i915->display.hdcp.arbiter->ops = &gsc_hdcp_ops; ret = intel_hdcp_gsc_hdcp2_init(i915); - mutex_unlock(&i915->display.hdcp.comp_mutex); + mutex_unlock(&i915->display.hdcp.hdcp_mutex); return ret; } @@ -717,7 +717,7 @@ int intel_hdcp_gsc_init(struct drm_i915_private *i915) void intel_hdcp_gsc_fini(struct drm_i915_private *i915) { intel_hdcp_gsc_free_message(i915); - kfree(i915->display.hdcp.master); + kfree(i915->display.hdcp.arbiter); } static int intel_gsc_send_sync(struct drm_i915_private *i915, diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c index a690a5616506b218072a8c12d086215c4c3e9b88..7ac5e6c5e00dc2ce6024a3dfc52ae39ce31c8e9a 100644 --- a/drivers/gpu/drm/i915/display/intel_hdmi.c +++ b/drivers/gpu/drm/i915/display/intel_hdmi.c @@ -40,12 +40,13 @@ #include #include -#include "i915_debugfs.h" +#include "g4x_hdmi.h" #include "i915_drv.h" #include "i915_reg.h" #include "intel_atomic.h" #include "intel_audio.h" #include "intel_connector.h" +#include "intel_cx0_phy.h" #include "intel_ddi.h" #include "intel_de.h" #include "intel_display_types.h" @@ -58,7 +59,7 @@ #include "intel_panel.h" #include "intel_snps_phy.h" -static struct drm_i915_private *intel_hdmi_to_i915(struct intel_hdmi *intel_hdmi) +inline struct drm_i915_private *intel_hdmi_to_i915(struct intel_hdmi *intel_hdmi) { return to_i915(hdmi_to_dig_port(intel_hdmi)->base.base.dev); } @@ -1789,7 +1790,9 @@ static int intel_hdmi_source_max_tmds_clock(struct intel_encoder *encoder) static bool intel_has_hdmi_sink(struct intel_hdmi *hdmi, const struct drm_connector_state *conn_state) { - return hdmi->has_hdmi_sink && + struct intel_connector *connector = hdmi->attached_connector; + + return connector->base.display_info.is_hdmi && READ_ONCE(to_intel_digital_connector_state(conn_state)->force_audio) != HDMI_AUDIO_OFF_DVI; } @@ -1865,16 +1868,19 @@ hdmi_port_clock_valid(struct intel_hdmi *hdmi, * FIXME: We will hopefully get an algorithmic way of programming * the MPLLB for HDMI in the future. */ - if (IS_DG2(dev_priv)) + if (DISPLAY_VER(dev_priv) >= 14) + return intel_cx0_phy_check_hdmi_link_rate(hdmi, clock); + else if (IS_DG2(dev_priv)) return intel_snps_phy_check_hdmi_link_rate(clock); return MODE_OK; } -int intel_hdmi_tmds_clock(int clock, int bpc, bool ycbcr420_output) +int intel_hdmi_tmds_clock(int clock, int bpc, + enum intel_output_format sink_format) { /* YCBCR420 TMDS rate requirement is half the pixel clock */ - if (ycbcr420_output) + if (sink_format == INTEL_OUTPUT_FORMAT_YCBCR420) clock /= 2; /* @@ -1901,7 +1907,8 @@ static bool intel_hdmi_source_bpc_possible(struct drm_i915_private *i915, int bp } static bool intel_hdmi_sink_bpc_possible(struct drm_connector *connector, - int bpc, bool has_hdmi_sink, bool ycbcr420_output) + int bpc, bool has_hdmi_sink, + enum intel_output_format sink_format) { const struct drm_display_info *info = &connector->display_info; const struct drm_hdmi_info *hdmi = &info->hdmi; @@ -1911,7 +1918,7 @@ static bool intel_hdmi_sink_bpc_possible(struct drm_connector *connector, if (!has_hdmi_sink) return false; - if (ycbcr420_output) + if (sink_format == INTEL_OUTPUT_FORMAT_YCBCR420) return hdmi->y420_dc_modes & DRM_EDID_YCBCR420_DC_36; else return info->edid_hdmi_rgb444_dc_modes & DRM_EDID_HDMI_DC_36; @@ -1919,7 +1926,7 @@ static bool intel_hdmi_sink_bpc_possible(struct drm_connector *connector, if (!has_hdmi_sink) return false; - if (ycbcr420_output) + if (sink_format == INTEL_OUTPUT_FORMAT_YCBCR420) return hdmi->y420_dc_modes & DRM_EDID_YCBCR420_DC_30; else return info->edid_hdmi_rgb444_dc_modes & DRM_EDID_HDMI_DC_30; @@ -1933,7 +1940,8 @@ static bool intel_hdmi_sink_bpc_possible(struct drm_connector *connector, static enum drm_mode_status intel_hdmi_mode_clock_valid(struct drm_connector *connector, int clock, - bool has_hdmi_sink, bool ycbcr420_output) + bool has_hdmi_sink, + enum intel_output_format sink_format) { struct drm_i915_private *i915 = to_i915(connector->dev); struct intel_hdmi *hdmi = intel_attached_hdmi(to_intel_connector(connector)); @@ -1946,12 +1954,12 @@ intel_hdmi_mode_clock_valid(struct drm_connector *connector, int clock, * least one color depth is accepted. */ for (bpc = 12; bpc >= 8; bpc -= 2) { - int tmds_clock = intel_hdmi_tmds_clock(clock, bpc, ycbcr420_output); + int tmds_clock = intel_hdmi_tmds_clock(clock, bpc, sink_format); if (!intel_hdmi_source_bpc_possible(i915, bpc)) continue; - if (!intel_hdmi_sink_bpc_possible(connector, bpc, has_hdmi_sink, ycbcr420_output)) + if (!intel_hdmi_sink_bpc_possible(connector, bpc, has_hdmi_sink, sink_format)) continue; status = hdmi_port_clock_valid(hdmi, tmds_clock, true, has_hdmi_sink); @@ -1976,6 +1984,7 @@ intel_hdmi_mode_valid(struct drm_connector *connector, int max_dotclk = to_i915(connector->dev)->max_dotclk_freq; bool has_hdmi_sink = intel_has_hdmi_sink(hdmi, connector->state); bool ycbcr_420_only; + enum intel_output_format sink_format; if ((mode->flags & DRM_MODE_FLAG_3D_MASK) == DRM_MODE_FLAG_3D_FRAME_PACKING) clock *= 2; @@ -2000,14 +2009,20 @@ intel_hdmi_mode_valid(struct drm_connector *connector, ycbcr_420_only = drm_mode_is_420_only(&connector->display_info, mode); - status = intel_hdmi_mode_clock_valid(connector, clock, has_hdmi_sink, ycbcr_420_only); + if (ycbcr_420_only) + sink_format = INTEL_OUTPUT_FORMAT_YCBCR420; + else + sink_format = INTEL_OUTPUT_FORMAT_RGB; + + status = intel_hdmi_mode_clock_valid(connector, clock, has_hdmi_sink, sink_format); if (status != MODE_OK) { if (ycbcr_420_only || !connector->ycbcr_420_allowed || !drm_mode_is_420_also(&connector->display_info, mode)) return status; - status = intel_hdmi_mode_clock_valid(connector, clock, has_hdmi_sink, true); + sink_format = INTEL_OUTPUT_FORMAT_YCBCR420; + status = intel_hdmi_mode_clock_valid(connector, clock, has_hdmi_sink, sink_format); if (status != MODE_OK) return status; } @@ -2016,7 +2031,7 @@ intel_hdmi_mode_valid(struct drm_connector *connector, } bool intel_hdmi_bpc_possible(const struct intel_crtc_state *crtc_state, - int bpc, bool has_hdmi_sink, bool ycbcr420_output) + int bpc, bool has_hdmi_sink) { struct drm_atomic_state *state = crtc_state->uapi.state; struct drm_connector_state *connector_state; @@ -2027,7 +2042,8 @@ bool intel_hdmi_bpc_possible(const struct intel_crtc_state *crtc_state, if (connector_state->crtc != crtc_state->uapi.crtc) continue; - if (!intel_hdmi_sink_bpc_possible(connector, bpc, has_hdmi_sink, ycbcr420_output)) + if (!intel_hdmi_sink_bpc_possible(connector, bpc, has_hdmi_sink, + crtc_state->sink_format)) return false; } @@ -2051,8 +2067,7 @@ static bool hdmi_bpc_possible(const struct intel_crtc_state *crtc_state, int bpc adjusted_mode->crtc_hblank_start) % 8 == 2) return false; - return intel_hdmi_bpc_possible(crtc_state, bpc, crtc_state->has_hdmi_sink, - intel_hdmi_is_ycbcr420(crtc_state)); + return intel_hdmi_bpc_possible(crtc_state, bpc, crtc_state->has_hdmi_sink); } static int intel_hdmi_compute_bpc(struct intel_encoder *encoder, @@ -2060,7 +2075,6 @@ static int intel_hdmi_compute_bpc(struct intel_encoder *encoder, int clock, bool respect_downstream_limits) { struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); - bool ycbcr420_output = intel_hdmi_is_ycbcr420(crtc_state); int bpc; /* @@ -2078,7 +2092,8 @@ static int intel_hdmi_compute_bpc(struct intel_encoder *encoder, bpc = 8; for (; bpc >= 8; bpc -= 2) { - int tmds_clock = intel_hdmi_tmds_clock(clock, bpc, ycbcr420_output); + int tmds_clock = intel_hdmi_tmds_clock(clock, bpc, + crtc_state->sink_format); if (hdmi_bpc_possible(crtc_state, bpc) && hdmi_port_clock_valid(intel_hdmi, tmds_clock, @@ -2108,7 +2123,7 @@ static int intel_hdmi_compute_clock(struct intel_encoder *encoder, return bpc; crtc_state->port_clock = - intel_hdmi_tmds_clock(clock, bpc, intel_hdmi_is_ycbcr420(crtc_state)); + intel_hdmi_tmds_clock(clock, bpc, crtc_state->sink_format); /* * pipe_bpp could already be below 8bpc due to @@ -2156,7 +2171,7 @@ static bool intel_hdmi_has_audio(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state, const struct drm_connector_state *conn_state) { - struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); + struct drm_connector *connector = conn_state->connector; const struct intel_digital_connector_state *intel_conn_state = to_intel_digital_connector_state(conn_state); @@ -2164,15 +2179,15 @@ static bool intel_hdmi_has_audio(struct intel_encoder *encoder, return false; if (intel_conn_state->force_audio == HDMI_AUDIO_AUTO) - return intel_hdmi->has_audio; + return connector->display_info.has_audio; else return intel_conn_state->force_audio == HDMI_AUDIO_ON; } static enum intel_output_format -intel_hdmi_output_format(const struct intel_crtc_state *crtc_state, - struct intel_connector *connector, - bool ycbcr_420_output) +intel_hdmi_sink_format(const struct intel_crtc_state *crtc_state, + struct intel_connector *connector, + bool ycbcr_420_output) { if (!crtc_state->has_hdmi_sink) return INTEL_OUTPUT_FORMAT_RGB; @@ -2183,6 +2198,12 @@ intel_hdmi_output_format(const struct intel_crtc_state *crtc_state, return INTEL_OUTPUT_FORMAT_RGB; } +static enum intel_output_format +intel_hdmi_output_format(const struct intel_crtc_state *crtc_state) +{ + return crtc_state->sink_format; +} + static int intel_hdmi_compute_output_format(struct intel_encoder *encoder, struct intel_crtc_state *crtc_state, const struct drm_connector_state *conn_state, @@ -2195,23 +2216,26 @@ static int intel_hdmi_compute_output_format(struct intel_encoder *encoder, bool ycbcr_420_only = drm_mode_is_420_only(info, adjusted_mode); int ret; - crtc_state->output_format = - intel_hdmi_output_format(crtc_state, connector, ycbcr_420_only); + crtc_state->sink_format = + intel_hdmi_sink_format(crtc_state, connector, ycbcr_420_only); - if (ycbcr_420_only && !intel_hdmi_is_ycbcr420(crtc_state)) { + if (ycbcr_420_only && crtc_state->sink_format != INTEL_OUTPUT_FORMAT_YCBCR420) { drm_dbg_kms(&i915->drm, "YCbCr 4:2:0 mode but YCbCr 4:2:0 output not possible. Falling back to RGB.\n"); - crtc_state->output_format = INTEL_OUTPUT_FORMAT_RGB; + crtc_state->sink_format = INTEL_OUTPUT_FORMAT_RGB; } + crtc_state->output_format = intel_hdmi_output_format(crtc_state); ret = intel_hdmi_compute_clock(encoder, crtc_state, respect_downstream_limits); if (ret) { - if (intel_hdmi_is_ycbcr420(crtc_state) || + if (crtc_state->sink_format == INTEL_OUTPUT_FORMAT_YCBCR420 || + !crtc_state->has_hdmi_sink || !connector->base.ycbcr_420_allowed || !drm_mode_is_420_also(info, adjusted_mode)) return ret; - crtc_state->output_format = intel_hdmi_output_format(crtc_state, connector, true); + crtc_state->sink_format = INTEL_OUTPUT_FORMAT_YCBCR420; + crtc_state->output_format = intel_hdmi_output_format(crtc_state); ret = intel_hdmi_compute_clock(encoder, crtc_state, respect_downstream_limits); } @@ -2243,11 +2267,20 @@ static bool source_supports_scrambling(struct intel_encoder *encoder) return intel_hdmi_source_max_tmds_clock(encoder) > 340000; } +bool intel_hdmi_compute_has_hdmi_sink(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state, + const struct drm_connector_state *conn_state) +{ + struct intel_hdmi *hdmi = enc_to_intel_hdmi(encoder); + + return intel_has_hdmi_sink(hdmi, conn_state) && + !intel_hdmi_is_cloned(crtc_state); +} + int intel_hdmi_compute_config(struct intel_encoder *encoder, struct intel_crtc_state *pipe_config, struct drm_connector_state *conn_state) { - struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); struct drm_display_mode *adjusted_mode = &pipe_config->hw.adjusted_mode; struct drm_connector *connector = conn_state->connector; @@ -2262,9 +2295,6 @@ int intel_hdmi_compute_config(struct intel_encoder *encoder, return -EINVAL; pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB; - pipe_config->has_hdmi_sink = - intel_has_hdmi_sink(intel_hdmi, conn_state) && - !intel_hdmi_is_cloned(pipe_config); if (pipe_config->has_hdmi_sink) pipe_config->has_infoframe = true; @@ -2357,9 +2387,6 @@ intel_hdmi_unset_edid(struct drm_connector *connector) { struct intel_hdmi *intel_hdmi = intel_attached_hdmi(to_intel_connector(connector)); - intel_hdmi->has_hdmi_sink = false; - intel_hdmi->has_audio = false; - intel_hdmi->dp_dual_mode.type = DRM_DP_DUAL_MODE_NONE; intel_hdmi->dp_dual_mode.max_tmds_clock = 0; @@ -2372,7 +2399,7 @@ intel_hdmi_dp_dual_mode_detect(struct drm_connector *connector) { struct drm_i915_private *dev_priv = to_i915(connector->dev); struct intel_hdmi *hdmi = intel_attached_hdmi(to_intel_connector(connector)); - enum port port = hdmi_to_dig_port(hdmi)->base.port; + struct intel_encoder *encoder = &hdmi_to_dig_port(hdmi)->base; struct i2c_adapter *adapter = intel_gmbus_get_adapter(dev_priv, hdmi->ddc_bus); enum drm_dp_dual_mode_type type = drm_dp_dual_mode_detect(&dev_priv->drm, adapter); @@ -2388,7 +2415,7 @@ intel_hdmi_dp_dual_mode_detect(struct drm_connector *connector) */ if (type == DRM_DP_DUAL_MODE_UNKNOWN) { if (!connector->force && - intel_bios_is_port_dp_dual_mode(dev_priv, port)) { + intel_bios_encoder_supports_dp_dual_mode(encoder->devdata)) { drm_dbg_kms(&dev_priv->drm, "Assuming DP dual mode adaptor presence based on VBT\n"); type = DRM_DP_DUAL_MODE_TYPE1_DVI; @@ -2411,7 +2438,7 @@ intel_hdmi_dp_dual_mode_detect(struct drm_connector *connector) /* Older VBTs are often buggy and can't be trusted :( Play it safe. */ if ((DISPLAY_VER(dev_priv) >= 8 || IS_HASWELL(dev_priv)) && - !intel_bios_is_port_dp_dual_mode(dev_priv, port)) { + !intel_bios_encoder_supports_dp_dual_mode(encoder->devdata)) { drm_dbg_kms(&dev_priv->drm, "Ignoring DP dual mode adaptor max TMDS clock for native HDMI port\n"); hdmi->dp_dual_mode.max_tmds_clock = 0; @@ -2451,9 +2478,6 @@ intel_hdmi_set_edid(struct drm_connector *connector) /* FIXME: Get rid of drm_edid_raw() */ edid = drm_edid_raw(drm_edid); if (edid && edid->input & DRM_EDID_INPUT_DIGITAL) { - intel_hdmi->has_audio = drm_detect_monitor_audio(edid); - intel_hdmi->has_hdmi_sink = drm_detect_hdmi_monitor(edid); - intel_hdmi_dp_dual_mode_detect(connector); connected = true; @@ -2597,10 +2621,21 @@ static const struct drm_connector_funcs intel_hdmi_connector_funcs = { .atomic_duplicate_state = intel_digital_connector_duplicate_state, }; +static int intel_hdmi_connector_atomic_check(struct drm_connector *connector, + struct drm_atomic_state *state) +{ + struct drm_i915_private *i915 = to_i915(state->dev); + + if (HAS_DDI(i915)) + return intel_digital_connector_atomic_check(connector, state); + else + return g4x_hdmi_connector_atomic_check(connector, state); +} + static const struct drm_connector_helper_funcs intel_hdmi_connector_helper_funcs = { .get_modes = intel_hdmi_get_modes, .mode_valid = intel_hdmi_mode_valid, - .atomic_check = intel_digital_connector_atomic_check, + .atomic_check = intel_hdmi_connector_atomic_check, }; static void diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.h b/drivers/gpu/drm/i915/display/intel_hdmi.h index 774dda2376ed20ed7afe23783beaf5abe8c2e5ae..6b39df38d57ab6eb1946da1cb4b5507b2a386871 100644 --- a/drivers/gpu/drm/i915/display/intel_hdmi.h +++ b/drivers/gpu/drm/i915/display/intel_hdmi.h @@ -9,6 +9,7 @@ #include enum hdmi_infoframe_type; +enum intel_output_format; enum port; struct drm_connector; struct drm_connector_state; @@ -23,6 +24,9 @@ union hdmi_infoframe; void intel_hdmi_init_connector(struct intel_digital_port *dig_port, struct intel_connector *intel_connector); +bool intel_hdmi_compute_has_hdmi_sink(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state, + const struct drm_connector_state *conn_state); int intel_hdmi_compute_config(struct intel_encoder *encoder, struct intel_crtc_state *pipe_config, struct drm_connector_state *conn_state); @@ -45,8 +49,8 @@ void intel_read_infoframe(struct intel_encoder *encoder, bool intel_hdmi_limited_color_range(const struct intel_crtc_state *crtc_state, const struct drm_connector_state *conn_state); bool intel_hdmi_bpc_possible(const struct intel_crtc_state *crtc_state, - int bpc, bool has_hdmi_sink, bool ycbcr420_output); -int intel_hdmi_tmds_clock(int clock, int bpc, bool ycbcr420_output); + int bpc, bool has_hdmi_sink); +int intel_hdmi_tmds_clock(int clock, int bpc, enum intel_output_format sink_format); int intel_hdmi_dsc_get_bpp(int src_fractional_bpp, int slice_width, int num_slices, int output_format, bool hdmi_all_bpp, int hdmi_max_chunk_bytes); @@ -54,5 +58,6 @@ int intel_hdmi_dsc_get_num_slices(const struct intel_crtc_state *crtc_state, int src_max_slices, int src_max_slice_width, int hdmi_max_slices, int hdmi_throughput); int intel_hdmi_dsc_get_slice_height(int vactive); +struct drm_i915_private *intel_hdmi_to_i915(struct intel_hdmi *intel_hdmi); #endif /* __INTEL_HDMI_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_hotplug.c b/drivers/gpu/drm/i915/display/intel_hotplug.c index b12900446828acf5871d044e6f56095394bb1f5e..1160fa20433bd65a649728d92d8f19fd6d706082 100644 --- a/drivers/gpu/drm/i915/display/intel_hotplug.c +++ b/drivers/gpu/drm/i915/display/intel_hotplug.c @@ -27,6 +27,7 @@ #include "i915_irq.h" #include "intel_display_types.h" #include "intel_hotplug.h" +#include "intel_hotplug_irq.h" /** * DOC: Hotplug @@ -211,7 +212,8 @@ intel_hpd_irq_storm_switch_to_polling(struct drm_i915_private *dev_priv) /* Enable polling and queue hotplug re-enabling. */ if (hpd_disabled) { drm_kms_helper_poll_enable(&dev_priv->drm); - mod_delayed_work(system_wq, &dev_priv->display.hotplug.reenable_work, + mod_delayed_work(dev_priv->unordered_wq, + &dev_priv->display.hotplug.reenable_work, msecs_to_jiffies(HPD_STORM_REENABLE_DELAY)); } } @@ -338,7 +340,8 @@ static void i915_digport_work_func(struct work_struct *work) spin_lock_irq(&dev_priv->irq_lock); dev_priv->display.hotplug.event_bits |= old_bits; spin_unlock_irq(&dev_priv->irq_lock); - queue_delayed_work(system_wq, &dev_priv->display.hotplug.hotplug_work, 0); + queue_delayed_work(dev_priv->unordered_wq, + &dev_priv->display.hotplug.hotplug_work, 0); } } @@ -445,7 +448,8 @@ static void i915_hotplug_work_func(struct work_struct *work) dev_priv->display.hotplug.retry_bits |= retry; spin_unlock_irq(&dev_priv->irq_lock); - mod_delayed_work(system_wq, &dev_priv->display.hotplug.hotplug_work, + mod_delayed_work(dev_priv->unordered_wq, + &dev_priv->display.hotplug.hotplug_work, msecs_to_jiffies(HPD_RETRY_DELAY)); } } @@ -576,7 +580,8 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv, if (queue_dig) queue_work(dev_priv->display.hotplug.dp_wq, &dev_priv->display.hotplug.dig_port_work); if (queue_hp) - queue_delayed_work(system_wq, &dev_priv->display.hotplug.hotplug_work, 0); + queue_delayed_work(dev_priv->unordered_wq, + &dev_priv->display.hotplug.hotplug_work, 0); } /** @@ -686,7 +691,8 @@ void intel_hpd_poll_enable(struct drm_i915_private *dev_priv) * As well, there's no issue if we race here since we always reschedule * this worker anyway */ - schedule_work(&dev_priv->display.hotplug.poll_init_work); + queue_work(dev_priv->unordered_wq, + &dev_priv->display.hotplug.poll_init_work); } /** @@ -714,7 +720,8 @@ void intel_hpd_poll_disable(struct drm_i915_private *dev_priv) return; WRITE_ONCE(dev_priv->display.hotplug.poll_enabled, false); - schedule_work(&dev_priv->display.hotplug.poll_init_work); + queue_work(dev_priv->unordered_wq, + &dev_priv->display.hotplug.poll_init_work); } void intel_hpd_init_early(struct drm_i915_private *i915) diff --git a/drivers/gpu/drm/i915/display/intel_hotplug_irq.c b/drivers/gpu/drm/i915/display/intel_hotplug_irq.c new file mode 100644 index 0000000000000000000000000000000000000000..f95fa793fabba4f1812d779ad510b12c4ccf91c1 --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_hotplug_irq.c @@ -0,0 +1,1442 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2023 Intel Corporation + */ + +#include "i915_drv.h" +#include "i915_reg.h" +#include "intel_de.h" +#include "intel_display_irq.h" +#include "intel_display_types.h" +#include "intel_dp_aux.h" +#include "intel_gmbus.h" +#include "intel_hotplug.h" +#include "intel_hotplug_irq.h" + +typedef bool (*long_pulse_detect_func)(enum hpd_pin pin, u32 val); +typedef u32 (*hotplug_enables_func)(struct intel_encoder *encoder); +typedef u32 (*hotplug_mask_func)(enum hpd_pin pin); + +static const u32 hpd_ilk[HPD_NUM_PINS] = { + [HPD_PORT_A] = DE_DP_A_HOTPLUG, +}; + +static const u32 hpd_ivb[HPD_NUM_PINS] = { + [HPD_PORT_A] = DE_DP_A_HOTPLUG_IVB, +}; + +static const u32 hpd_bdw[HPD_NUM_PINS] = { + [HPD_PORT_A] = GEN8_DE_PORT_HOTPLUG(HPD_PORT_A), +}; + +static const u32 hpd_ibx[HPD_NUM_PINS] = { + [HPD_CRT] = SDE_CRT_HOTPLUG, + [HPD_SDVO_B] = SDE_SDVOB_HOTPLUG, + [HPD_PORT_B] = SDE_PORTB_HOTPLUG, + [HPD_PORT_C] = SDE_PORTC_HOTPLUG, + [HPD_PORT_D] = SDE_PORTD_HOTPLUG, +}; + +static const u32 hpd_cpt[HPD_NUM_PINS] = { + [HPD_CRT] = SDE_CRT_HOTPLUG_CPT, + [HPD_SDVO_B] = SDE_SDVOB_HOTPLUG_CPT, + [HPD_PORT_B] = SDE_PORTB_HOTPLUG_CPT, + [HPD_PORT_C] = SDE_PORTC_HOTPLUG_CPT, + [HPD_PORT_D] = SDE_PORTD_HOTPLUG_CPT, +}; + +static const u32 hpd_spt[HPD_NUM_PINS] = { + [HPD_PORT_A] = SDE_PORTA_HOTPLUG_SPT, + [HPD_PORT_B] = SDE_PORTB_HOTPLUG_CPT, + [HPD_PORT_C] = SDE_PORTC_HOTPLUG_CPT, + [HPD_PORT_D] = SDE_PORTD_HOTPLUG_CPT, + [HPD_PORT_E] = SDE_PORTE_HOTPLUG_SPT, +}; + +static const u32 hpd_mask_i915[HPD_NUM_PINS] = { + [HPD_CRT] = CRT_HOTPLUG_INT_EN, + [HPD_SDVO_B] = SDVOB_HOTPLUG_INT_EN, + [HPD_SDVO_C] = SDVOC_HOTPLUG_INT_EN, + [HPD_PORT_B] = PORTB_HOTPLUG_INT_EN, + [HPD_PORT_C] = PORTC_HOTPLUG_INT_EN, + [HPD_PORT_D] = PORTD_HOTPLUG_INT_EN, +}; + +static const u32 hpd_status_g4x[HPD_NUM_PINS] = { + [HPD_CRT] = CRT_HOTPLUG_INT_STATUS, + [HPD_SDVO_B] = SDVOB_HOTPLUG_INT_STATUS_G4X, + [HPD_SDVO_C] = SDVOC_HOTPLUG_INT_STATUS_G4X, + [HPD_PORT_B] = PORTB_HOTPLUG_INT_STATUS, + [HPD_PORT_C] = PORTC_HOTPLUG_INT_STATUS, + [HPD_PORT_D] = PORTD_HOTPLUG_INT_STATUS, +}; + +static const u32 hpd_status_i915[HPD_NUM_PINS] = { + [HPD_CRT] = CRT_HOTPLUG_INT_STATUS, + [HPD_SDVO_B] = SDVOB_HOTPLUG_INT_STATUS_I915, + [HPD_SDVO_C] = SDVOC_HOTPLUG_INT_STATUS_I915, + [HPD_PORT_B] = PORTB_HOTPLUG_INT_STATUS, + [HPD_PORT_C] = PORTC_HOTPLUG_INT_STATUS, + [HPD_PORT_D] = PORTD_HOTPLUG_INT_STATUS, +}; + +static const u32 hpd_bxt[HPD_NUM_PINS] = { + [HPD_PORT_A] = GEN8_DE_PORT_HOTPLUG(HPD_PORT_A), + [HPD_PORT_B] = GEN8_DE_PORT_HOTPLUG(HPD_PORT_B), + [HPD_PORT_C] = GEN8_DE_PORT_HOTPLUG(HPD_PORT_C), +}; + +static const u32 hpd_gen11[HPD_NUM_PINS] = { + [HPD_PORT_TC1] = GEN11_TC_HOTPLUG(HPD_PORT_TC1) | GEN11_TBT_HOTPLUG(HPD_PORT_TC1), + [HPD_PORT_TC2] = GEN11_TC_HOTPLUG(HPD_PORT_TC2) | GEN11_TBT_HOTPLUG(HPD_PORT_TC2), + [HPD_PORT_TC3] = GEN11_TC_HOTPLUG(HPD_PORT_TC3) | GEN11_TBT_HOTPLUG(HPD_PORT_TC3), + [HPD_PORT_TC4] = GEN11_TC_HOTPLUG(HPD_PORT_TC4) | GEN11_TBT_HOTPLUG(HPD_PORT_TC4), + [HPD_PORT_TC5] = GEN11_TC_HOTPLUG(HPD_PORT_TC5) | GEN11_TBT_HOTPLUG(HPD_PORT_TC5), + [HPD_PORT_TC6] = GEN11_TC_HOTPLUG(HPD_PORT_TC6) | GEN11_TBT_HOTPLUG(HPD_PORT_TC6), +}; + +static const u32 hpd_xelpdp[HPD_NUM_PINS] = { + [HPD_PORT_TC1] = XELPDP_TBT_HOTPLUG(HPD_PORT_TC1) | XELPDP_DP_ALT_HOTPLUG(HPD_PORT_TC1), + [HPD_PORT_TC2] = XELPDP_TBT_HOTPLUG(HPD_PORT_TC2) | XELPDP_DP_ALT_HOTPLUG(HPD_PORT_TC2), + [HPD_PORT_TC3] = XELPDP_TBT_HOTPLUG(HPD_PORT_TC3) | XELPDP_DP_ALT_HOTPLUG(HPD_PORT_TC3), + [HPD_PORT_TC4] = XELPDP_TBT_HOTPLUG(HPD_PORT_TC4) | XELPDP_DP_ALT_HOTPLUG(HPD_PORT_TC4), +}; + +static const u32 hpd_icp[HPD_NUM_PINS] = { + [HPD_PORT_A] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_A), + [HPD_PORT_B] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_B), + [HPD_PORT_C] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_C), + [HPD_PORT_TC1] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC1), + [HPD_PORT_TC2] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC2), + [HPD_PORT_TC3] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC3), + [HPD_PORT_TC4] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC4), + [HPD_PORT_TC5] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC5), + [HPD_PORT_TC6] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC6), +}; + +static const u32 hpd_sde_dg1[HPD_NUM_PINS] = { + [HPD_PORT_A] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_A), + [HPD_PORT_B] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_B), + [HPD_PORT_C] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_C), + [HPD_PORT_D] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_D), + [HPD_PORT_TC1] = SDE_TC_HOTPLUG_DG2(HPD_PORT_TC1), +}; + +static const u32 hpd_mtp[HPD_NUM_PINS] = { + [HPD_PORT_A] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_A), + [HPD_PORT_B] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_B), + [HPD_PORT_TC1] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC1), + [HPD_PORT_TC2] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC2), + [HPD_PORT_TC3] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC3), + [HPD_PORT_TC4] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC4), +}; + +static void intel_hpd_init_pins(struct drm_i915_private *dev_priv) +{ + struct intel_hotplug *hpd = &dev_priv->display.hotplug; + + if (HAS_GMCH(dev_priv)) { + if (IS_G4X(dev_priv) || IS_VALLEYVIEW(dev_priv) || + IS_CHERRYVIEW(dev_priv)) + hpd->hpd = hpd_status_g4x; + else + hpd->hpd = hpd_status_i915; + return; + } + + if (DISPLAY_VER(dev_priv) >= 14) + hpd->hpd = hpd_xelpdp; + else if (DISPLAY_VER(dev_priv) >= 11) + hpd->hpd = hpd_gen11; + else if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) + hpd->hpd = hpd_bxt; + else if (DISPLAY_VER(dev_priv) == 9) + hpd->hpd = NULL; /* no north HPD on SKL */ + else if (DISPLAY_VER(dev_priv) >= 8) + hpd->hpd = hpd_bdw; + else if (DISPLAY_VER(dev_priv) >= 7) + hpd->hpd = hpd_ivb; + else + hpd->hpd = hpd_ilk; + + if ((INTEL_PCH_TYPE(dev_priv) < PCH_DG1) && + (!HAS_PCH_SPLIT(dev_priv) || HAS_PCH_NOP(dev_priv))) + return; + + if (INTEL_PCH_TYPE(dev_priv) >= PCH_DG1) + hpd->pch_hpd = hpd_sde_dg1; + else if (INTEL_PCH_TYPE(dev_priv) >= PCH_MTP) + hpd->pch_hpd = hpd_mtp; + else if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP) + hpd->pch_hpd = hpd_icp; + else if (HAS_PCH_CNP(dev_priv) || HAS_PCH_SPT(dev_priv)) + hpd->pch_hpd = hpd_spt; + else if (HAS_PCH_LPT(dev_priv) || HAS_PCH_CPT(dev_priv)) + hpd->pch_hpd = hpd_cpt; + else if (HAS_PCH_IBX(dev_priv)) + hpd->pch_hpd = hpd_ibx; + else + MISSING_CASE(INTEL_PCH_TYPE(dev_priv)); +} + +/* For display hotplug interrupt */ +void i915_hotplug_interrupt_update_locked(struct drm_i915_private *dev_priv, + u32 mask, u32 bits) +{ + lockdep_assert_held(&dev_priv->irq_lock); + drm_WARN_ON(&dev_priv->drm, bits & ~mask); + + intel_uncore_rmw(&dev_priv->uncore, PORT_HOTPLUG_EN, mask, bits); +} + +/** + * i915_hotplug_interrupt_update - update hotplug interrupt enable + * @dev_priv: driver private + * @mask: bits to update + * @bits: bits to enable + * NOTE: the HPD enable bits are modified both inside and outside + * of an interrupt context. To avoid that read-modify-write cycles + * interfer, these bits are protected by a spinlock. Since this + * function is usually not called from a context where the lock is + * held already, this function acquires the lock itself. A non-locking + * version is also available. + */ +void i915_hotplug_interrupt_update(struct drm_i915_private *dev_priv, + u32 mask, + u32 bits) +{ + spin_lock_irq(&dev_priv->irq_lock); + i915_hotplug_interrupt_update_locked(dev_priv, mask, bits); + spin_unlock_irq(&dev_priv->irq_lock); +} + +static bool gen11_port_hotplug_long_detect(enum hpd_pin pin, u32 val) +{ + switch (pin) { + case HPD_PORT_TC1: + case HPD_PORT_TC2: + case HPD_PORT_TC3: + case HPD_PORT_TC4: + case HPD_PORT_TC5: + case HPD_PORT_TC6: + return val & GEN11_HOTPLUG_CTL_LONG_DETECT(pin); + default: + return false; + } +} + +static bool bxt_port_hotplug_long_detect(enum hpd_pin pin, u32 val) +{ + switch (pin) { + case HPD_PORT_A: + return val & PORTA_HOTPLUG_LONG_DETECT; + case HPD_PORT_B: + return val & PORTB_HOTPLUG_LONG_DETECT; + case HPD_PORT_C: + return val & PORTC_HOTPLUG_LONG_DETECT; + default: + return false; + } +} + +static bool icp_ddi_port_hotplug_long_detect(enum hpd_pin pin, u32 val) +{ + switch (pin) { + case HPD_PORT_A: + case HPD_PORT_B: + case HPD_PORT_C: + case HPD_PORT_D: + return val & SHOTPLUG_CTL_DDI_HPD_LONG_DETECT(pin); + default: + return false; + } +} + +static bool icp_tc_port_hotplug_long_detect(enum hpd_pin pin, u32 val) +{ + switch (pin) { + case HPD_PORT_TC1: + case HPD_PORT_TC2: + case HPD_PORT_TC3: + case HPD_PORT_TC4: + case HPD_PORT_TC5: + case HPD_PORT_TC6: + return val & ICP_TC_HPD_LONG_DETECT(pin); + default: + return false; + } +} + +static bool spt_port_hotplug2_long_detect(enum hpd_pin pin, u32 val) +{ + switch (pin) { + case HPD_PORT_E: + return val & PORTE_HOTPLUG_LONG_DETECT; + default: + return false; + } +} + +static bool spt_port_hotplug_long_detect(enum hpd_pin pin, u32 val) +{ + switch (pin) { + case HPD_PORT_A: + return val & PORTA_HOTPLUG_LONG_DETECT; + case HPD_PORT_B: + return val & PORTB_HOTPLUG_LONG_DETECT; + case HPD_PORT_C: + return val & PORTC_HOTPLUG_LONG_DETECT; + case HPD_PORT_D: + return val & PORTD_HOTPLUG_LONG_DETECT; + default: + return false; + } +} + +static bool ilk_port_hotplug_long_detect(enum hpd_pin pin, u32 val) +{ + switch (pin) { + case HPD_PORT_A: + return val & DIGITAL_PORTA_HOTPLUG_LONG_DETECT; + default: + return false; + } +} + +static bool pch_port_hotplug_long_detect(enum hpd_pin pin, u32 val) +{ + switch (pin) { + case HPD_PORT_B: + return val & PORTB_HOTPLUG_LONG_DETECT; + case HPD_PORT_C: + return val & PORTC_HOTPLUG_LONG_DETECT; + case HPD_PORT_D: + return val & PORTD_HOTPLUG_LONG_DETECT; + default: + return false; + } +} + +static bool i9xx_port_hotplug_long_detect(enum hpd_pin pin, u32 val) +{ + switch (pin) { + case HPD_PORT_B: + return val & PORTB_HOTPLUG_INT_LONG_PULSE; + case HPD_PORT_C: + return val & PORTC_HOTPLUG_INT_LONG_PULSE; + case HPD_PORT_D: + return val & PORTD_HOTPLUG_INT_LONG_PULSE; + default: + return false; + } +} + +/* + * Get a bit mask of pins that have triggered, and which ones may be long. + * This can be called multiple times with the same masks to accumulate + * hotplug detection results from several registers. + * + * Note that the caller is expected to zero out the masks initially. + */ +static void intel_get_hpd_pins(struct drm_i915_private *dev_priv, + u32 *pin_mask, u32 *long_mask, + u32 hotplug_trigger, u32 dig_hotplug_reg, + const u32 hpd[HPD_NUM_PINS], + bool long_pulse_detect(enum hpd_pin pin, u32 val)) +{ + enum hpd_pin pin; + + BUILD_BUG_ON(BITS_PER_TYPE(*pin_mask) < HPD_NUM_PINS); + + for_each_hpd_pin(pin) { + if ((hpd[pin] & hotplug_trigger) == 0) + continue; + + *pin_mask |= BIT(pin); + + if (long_pulse_detect(pin, dig_hotplug_reg)) + *long_mask |= BIT(pin); + } + + drm_dbg(&dev_priv->drm, + "hotplug event received, stat 0x%08x, dig 0x%08x, pins 0x%08x, long 0x%08x\n", + hotplug_trigger, dig_hotplug_reg, *pin_mask, *long_mask); +} + +static u32 intel_hpd_enabled_irqs(struct drm_i915_private *dev_priv, + const u32 hpd[HPD_NUM_PINS]) +{ + struct intel_encoder *encoder; + u32 enabled_irqs = 0; + + for_each_intel_encoder(&dev_priv->drm, encoder) + if (dev_priv->display.hotplug.stats[encoder->hpd_pin].state == HPD_ENABLED) + enabled_irqs |= hpd[encoder->hpd_pin]; + + return enabled_irqs; +} + +static u32 intel_hpd_hotplug_irqs(struct drm_i915_private *dev_priv, + const u32 hpd[HPD_NUM_PINS]) +{ + struct intel_encoder *encoder; + u32 hotplug_irqs = 0; + + for_each_intel_encoder(&dev_priv->drm, encoder) + hotplug_irqs |= hpd[encoder->hpd_pin]; + + return hotplug_irqs; +} + +static u32 intel_hpd_hotplug_mask(struct drm_i915_private *i915, + hotplug_mask_func hotplug_mask) +{ + enum hpd_pin pin; + u32 hotplug = 0; + + for_each_hpd_pin(pin) + hotplug |= hotplug_mask(pin); + + return hotplug; +} + +static u32 intel_hpd_hotplug_enables(struct drm_i915_private *i915, + hotplug_enables_func hotplug_enables) +{ + struct intel_encoder *encoder; + u32 hotplug = 0; + + for_each_intel_encoder(&i915->drm, encoder) + hotplug |= hotplug_enables(encoder); + + return hotplug; +} + +u32 i9xx_hpd_irq_ack(struct drm_i915_private *dev_priv) +{ + u32 hotplug_status = 0, hotplug_status_mask; + int i; + + if (IS_G4X(dev_priv) || + IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) + hotplug_status_mask = HOTPLUG_INT_STATUS_G4X | + DP_AUX_CHANNEL_MASK_INT_STATUS_G4X; + else + hotplug_status_mask = HOTPLUG_INT_STATUS_I915; + + /* + * We absolutely have to clear all the pending interrupt + * bits in PORT_HOTPLUG_STAT. Otherwise the ISR port + * interrupt bit won't have an edge, and the i965/g4x + * edge triggered IIR will not notice that an interrupt + * is still pending. We can't use PORT_HOTPLUG_EN to + * guarantee the edge as the act of toggling the enable + * bits can itself generate a new hotplug interrupt :( + */ + for (i = 0; i < 10; i++) { + u32 tmp = intel_uncore_read(&dev_priv->uncore, PORT_HOTPLUG_STAT) & hotplug_status_mask; + + if (tmp == 0) + return hotplug_status; + + hotplug_status |= tmp; + intel_uncore_write(&dev_priv->uncore, PORT_HOTPLUG_STAT, hotplug_status); + } + + drm_WARN_ONCE(&dev_priv->drm, 1, + "PORT_HOTPLUG_STAT did not clear (0x%08x)\n", + intel_uncore_read(&dev_priv->uncore, PORT_HOTPLUG_STAT)); + + return hotplug_status; +} + +void i9xx_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 hotplug_status) +{ + u32 pin_mask = 0, long_mask = 0; + u32 hotplug_trigger; + + if (IS_G4X(dev_priv) || + IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) + hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_G4X; + else + hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_I915; + + if (hotplug_trigger) { + intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, + hotplug_trigger, hotplug_trigger, + dev_priv->display.hotplug.hpd, + i9xx_port_hotplug_long_detect); + + intel_hpd_irq_handler(dev_priv, pin_mask, long_mask); + } + + if ((IS_G4X(dev_priv) || + IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) && + hotplug_status & DP_AUX_CHANNEL_MASK_INT_STATUS_G4X) + intel_dp_aux_irq_handler(dev_priv); +} + +void ibx_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 hotplug_trigger) +{ + u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0; + + /* + * Somehow the PCH doesn't seem to really ack the interrupt to the CPU + * unless we touch the hotplug register, even if hotplug_trigger is + * zero. Not acking leads to "The master control interrupt lied (SDE)!" + * errors. + */ + dig_hotplug_reg = intel_uncore_read(&dev_priv->uncore, PCH_PORT_HOTPLUG); + if (!hotplug_trigger) { + u32 mask = PORTA_HOTPLUG_STATUS_MASK | + PORTD_HOTPLUG_STATUS_MASK | + PORTC_HOTPLUG_STATUS_MASK | + PORTB_HOTPLUG_STATUS_MASK; + dig_hotplug_reg &= ~mask; + } + + intel_uncore_write(&dev_priv->uncore, PCH_PORT_HOTPLUG, dig_hotplug_reg); + if (!hotplug_trigger) + return; + + intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, + hotplug_trigger, dig_hotplug_reg, + dev_priv->display.hotplug.pch_hpd, + pch_port_hotplug_long_detect); + + intel_hpd_irq_handler(dev_priv, pin_mask, long_mask); +} + +void xelpdp_pica_irq_handler(struct drm_i915_private *i915, u32 iir) +{ + enum hpd_pin pin; + u32 hotplug_trigger = iir & (XELPDP_DP_ALT_HOTPLUG_MASK | XELPDP_TBT_HOTPLUG_MASK); + u32 trigger_aux = iir & XELPDP_AUX_TC_MASK; + u32 pin_mask = 0, long_mask = 0; + + for (pin = HPD_PORT_TC1; pin <= HPD_PORT_TC4; pin++) { + u32 val; + + if (!(i915->display.hotplug.hpd[pin] & hotplug_trigger)) + continue; + + pin_mask |= BIT(pin); + + val = intel_de_read(i915, XELPDP_PORT_HOTPLUG_CTL(pin)); + intel_de_write(i915, XELPDP_PORT_HOTPLUG_CTL(pin), val); + + if (val & (XELPDP_DP_ALT_HPD_LONG_DETECT | XELPDP_TBT_HPD_LONG_DETECT)) + long_mask |= BIT(pin); + } + + if (pin_mask) { + drm_dbg(&i915->drm, + "pica hotplug event received, stat 0x%08x, pins 0x%08x, long 0x%08x\n", + hotplug_trigger, pin_mask, long_mask); + + intel_hpd_irq_handler(i915, pin_mask, long_mask); + } + + if (trigger_aux) + intel_dp_aux_irq_handler(i915); + + if (!pin_mask && !trigger_aux) + drm_err(&i915->drm, + "Unexpected DE HPD/AUX interrupt 0x%08x\n", iir); +} + +void icp_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir) +{ + u32 ddi_hotplug_trigger = pch_iir & SDE_DDI_HOTPLUG_MASK_ICP; + u32 tc_hotplug_trigger = pch_iir & SDE_TC_HOTPLUG_MASK_ICP; + u32 pin_mask = 0, long_mask = 0; + + if (ddi_hotplug_trigger) { + u32 dig_hotplug_reg; + + /* Locking due to DSI native GPIO sequences */ + spin_lock(&dev_priv->irq_lock); + dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_DDI, 0, 0); + spin_unlock(&dev_priv->irq_lock); + + intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, + ddi_hotplug_trigger, dig_hotplug_reg, + dev_priv->display.hotplug.pch_hpd, + icp_ddi_port_hotplug_long_detect); + } + + if (tc_hotplug_trigger) { + u32 dig_hotplug_reg; + + dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_TC, 0, 0); + + intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, + tc_hotplug_trigger, dig_hotplug_reg, + dev_priv->display.hotplug.pch_hpd, + icp_tc_port_hotplug_long_detect); + } + + if (pin_mask) + intel_hpd_irq_handler(dev_priv, pin_mask, long_mask); + + if (pch_iir & SDE_GMBUS_ICP) + intel_gmbus_irq_handler(dev_priv); +} + +void spt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir) +{ + u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK_SPT & + ~SDE_PORTE_HOTPLUG_SPT; + u32 hotplug2_trigger = pch_iir & SDE_PORTE_HOTPLUG_SPT; + u32 pin_mask = 0, long_mask = 0; + + if (hotplug_trigger) { + u32 dig_hotplug_reg; + + dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG, 0, 0); + + intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, + hotplug_trigger, dig_hotplug_reg, + dev_priv->display.hotplug.pch_hpd, + spt_port_hotplug_long_detect); + } + + if (hotplug2_trigger) { + u32 dig_hotplug_reg; + + dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG2, 0, 0); + + intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, + hotplug2_trigger, dig_hotplug_reg, + dev_priv->display.hotplug.pch_hpd, + spt_port_hotplug2_long_detect); + } + + if (pin_mask) + intel_hpd_irq_handler(dev_priv, pin_mask, long_mask); + + if (pch_iir & SDE_GMBUS_CPT) + intel_gmbus_irq_handler(dev_priv); +} + +void ilk_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 hotplug_trigger) +{ + u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0; + + dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, DIGITAL_PORT_HOTPLUG_CNTRL, 0, 0); + + intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, + hotplug_trigger, dig_hotplug_reg, + dev_priv->display.hotplug.hpd, + ilk_port_hotplug_long_detect); + + intel_hpd_irq_handler(dev_priv, pin_mask, long_mask); +} + +void bxt_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 hotplug_trigger) +{ + u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0; + + dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG, 0, 0); + + intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, + hotplug_trigger, dig_hotplug_reg, + dev_priv->display.hotplug.hpd, + bxt_port_hotplug_long_detect); + + intel_hpd_irq_handler(dev_priv, pin_mask, long_mask); +} + +void gen11_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 iir) +{ + u32 pin_mask = 0, long_mask = 0; + u32 trigger_tc = iir & GEN11_DE_TC_HOTPLUG_MASK; + u32 trigger_tbt = iir & GEN11_DE_TBT_HOTPLUG_MASK; + + if (trigger_tc) { + u32 dig_hotplug_reg; + + dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, GEN11_TC_HOTPLUG_CTL, 0, 0); + + intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, + trigger_tc, dig_hotplug_reg, + dev_priv->display.hotplug.hpd, + gen11_port_hotplug_long_detect); + } + + if (trigger_tbt) { + u32 dig_hotplug_reg; + + dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, GEN11_TBT_HOTPLUG_CTL, 0, 0); + + intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, + trigger_tbt, dig_hotplug_reg, + dev_priv->display.hotplug.hpd, + gen11_port_hotplug_long_detect); + } + + if (pin_mask) + intel_hpd_irq_handler(dev_priv, pin_mask, long_mask); + else + drm_err(&dev_priv->drm, + "Unexpected DE HPD interrupt 0x%08x\n", iir); +} + +static u32 ibx_hotplug_mask(enum hpd_pin hpd_pin) +{ + switch (hpd_pin) { + case HPD_PORT_A: + return PORTA_HOTPLUG_ENABLE; + case HPD_PORT_B: + return PORTB_HOTPLUG_ENABLE | PORTB_PULSE_DURATION_MASK; + case HPD_PORT_C: + return PORTC_HOTPLUG_ENABLE | PORTC_PULSE_DURATION_MASK; + case HPD_PORT_D: + return PORTD_HOTPLUG_ENABLE | PORTD_PULSE_DURATION_MASK; + default: + return 0; + } +} + +static u32 ibx_hotplug_enables(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + + switch (encoder->hpd_pin) { + case HPD_PORT_A: + /* + * When CPU and PCH are on the same package, port A + * HPD must be enabled in both north and south. + */ + return HAS_PCH_LPT_LP(i915) ? + PORTA_HOTPLUG_ENABLE : 0; + case HPD_PORT_B: + return PORTB_HOTPLUG_ENABLE | + PORTB_PULSE_DURATION_2ms; + case HPD_PORT_C: + return PORTC_HOTPLUG_ENABLE | + PORTC_PULSE_DURATION_2ms; + case HPD_PORT_D: + return PORTD_HOTPLUG_ENABLE | + PORTD_PULSE_DURATION_2ms; + default: + return 0; + } +} + +static void ibx_hpd_detection_setup(struct drm_i915_private *dev_priv) +{ + /* + * Enable digital hotplug on the PCH, and configure the DP short pulse + * duration to 2ms (which is the minimum in the Display Port spec). + * The pulse duration bits are reserved on LPT+. + */ + intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG, + intel_hpd_hotplug_mask(dev_priv, ibx_hotplug_mask), + intel_hpd_hotplug_enables(dev_priv, ibx_hotplug_enables)); +} + +static void ibx_hpd_enable_detection(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + + intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG, + ibx_hotplug_mask(encoder->hpd_pin), + ibx_hotplug_enables(encoder)); +} + +static void ibx_hpd_irq_setup(struct drm_i915_private *dev_priv) +{ + u32 hotplug_irqs, enabled_irqs; + + enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd); + hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd); + + ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs); + + ibx_hpd_detection_setup(dev_priv); +} + +static u32 icp_ddi_hotplug_mask(enum hpd_pin hpd_pin) +{ + switch (hpd_pin) { + case HPD_PORT_A: + case HPD_PORT_B: + case HPD_PORT_C: + case HPD_PORT_D: + return SHOTPLUG_CTL_DDI_HPD_ENABLE(hpd_pin); + default: + return 0; + } +} + +static u32 icp_ddi_hotplug_enables(struct intel_encoder *encoder) +{ + return icp_ddi_hotplug_mask(encoder->hpd_pin); +} + +static u32 icp_tc_hotplug_mask(enum hpd_pin hpd_pin) +{ + switch (hpd_pin) { + case HPD_PORT_TC1: + case HPD_PORT_TC2: + case HPD_PORT_TC3: + case HPD_PORT_TC4: + case HPD_PORT_TC5: + case HPD_PORT_TC6: + return ICP_TC_HPD_ENABLE(hpd_pin); + default: + return 0; + } +} + +static u32 icp_tc_hotplug_enables(struct intel_encoder *encoder) +{ + return icp_tc_hotplug_mask(encoder->hpd_pin); +} + +static void icp_ddi_hpd_detection_setup(struct drm_i915_private *dev_priv) +{ + intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_DDI, + intel_hpd_hotplug_mask(dev_priv, icp_ddi_hotplug_mask), + intel_hpd_hotplug_enables(dev_priv, icp_ddi_hotplug_enables)); +} + +static void icp_ddi_hpd_enable_detection(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + + intel_uncore_rmw(&i915->uncore, SHOTPLUG_CTL_DDI, + icp_ddi_hotplug_mask(encoder->hpd_pin), + icp_ddi_hotplug_enables(encoder)); +} + +static void icp_tc_hpd_detection_setup(struct drm_i915_private *dev_priv) +{ + intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_TC, + intel_hpd_hotplug_mask(dev_priv, icp_tc_hotplug_mask), + intel_hpd_hotplug_enables(dev_priv, icp_tc_hotplug_enables)); +} + +static void icp_tc_hpd_enable_detection(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + + intel_uncore_rmw(&i915->uncore, SHOTPLUG_CTL_TC, + icp_tc_hotplug_mask(encoder->hpd_pin), + icp_tc_hotplug_enables(encoder)); +} + +static void icp_hpd_enable_detection(struct intel_encoder *encoder) +{ + icp_ddi_hpd_enable_detection(encoder); + icp_tc_hpd_enable_detection(encoder); +} + +static void icp_hpd_irq_setup(struct drm_i915_private *dev_priv) +{ + u32 hotplug_irqs, enabled_irqs; + + enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd); + hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd); + + if (INTEL_PCH_TYPE(dev_priv) <= PCH_TGP) + intel_uncore_write(&dev_priv->uncore, SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ); + + ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs); + + icp_ddi_hpd_detection_setup(dev_priv); + icp_tc_hpd_detection_setup(dev_priv); +} + +static u32 gen11_hotplug_mask(enum hpd_pin hpd_pin) +{ + switch (hpd_pin) { + case HPD_PORT_TC1: + case HPD_PORT_TC2: + case HPD_PORT_TC3: + case HPD_PORT_TC4: + case HPD_PORT_TC5: + case HPD_PORT_TC6: + return GEN11_HOTPLUG_CTL_ENABLE(hpd_pin); + default: + return 0; + } +} + +static u32 gen11_hotplug_enables(struct intel_encoder *encoder) +{ + return gen11_hotplug_mask(encoder->hpd_pin); +} + +static void dg1_hpd_invert(struct drm_i915_private *i915) +{ + u32 val = (INVERT_DDIA_HPD | + INVERT_DDIB_HPD | + INVERT_DDIC_HPD | + INVERT_DDID_HPD); + intel_uncore_rmw(&i915->uncore, SOUTH_CHICKEN1, 0, val); +} + +static void dg1_hpd_enable_detection(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + + dg1_hpd_invert(i915); + icp_hpd_enable_detection(encoder); +} + +static void dg1_hpd_irq_setup(struct drm_i915_private *dev_priv) +{ + dg1_hpd_invert(dev_priv); + icp_hpd_irq_setup(dev_priv); +} + +static void gen11_tc_hpd_detection_setup(struct drm_i915_private *dev_priv) +{ + intel_uncore_rmw(&dev_priv->uncore, GEN11_TC_HOTPLUG_CTL, + intel_hpd_hotplug_mask(dev_priv, gen11_hotplug_mask), + intel_hpd_hotplug_enables(dev_priv, gen11_hotplug_enables)); +} + +static void gen11_tc_hpd_enable_detection(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + + intel_uncore_rmw(&i915->uncore, GEN11_TC_HOTPLUG_CTL, + gen11_hotplug_mask(encoder->hpd_pin), + gen11_hotplug_enables(encoder)); +} + +static void gen11_tbt_hpd_detection_setup(struct drm_i915_private *dev_priv) +{ + intel_uncore_rmw(&dev_priv->uncore, GEN11_TBT_HOTPLUG_CTL, + intel_hpd_hotplug_mask(dev_priv, gen11_hotplug_mask), + intel_hpd_hotplug_enables(dev_priv, gen11_hotplug_enables)); +} + +static void gen11_tbt_hpd_enable_detection(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + + intel_uncore_rmw(&i915->uncore, GEN11_TBT_HOTPLUG_CTL, + gen11_hotplug_mask(encoder->hpd_pin), + gen11_hotplug_enables(encoder)); +} + +static void gen11_hpd_enable_detection(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + + gen11_tc_hpd_enable_detection(encoder); + gen11_tbt_hpd_enable_detection(encoder); + + if (INTEL_PCH_TYPE(i915) >= PCH_ICP) + icp_hpd_enable_detection(encoder); +} + +static void gen11_hpd_irq_setup(struct drm_i915_private *dev_priv) +{ + u32 hotplug_irqs, enabled_irqs; + + enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.hpd); + hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.hpd); + + intel_uncore_rmw(&dev_priv->uncore, GEN11_DE_HPD_IMR, hotplug_irqs, + ~enabled_irqs & hotplug_irqs); + intel_uncore_posting_read(&dev_priv->uncore, GEN11_DE_HPD_IMR); + + gen11_tc_hpd_detection_setup(dev_priv); + gen11_tbt_hpd_detection_setup(dev_priv); + + if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP) + icp_hpd_irq_setup(dev_priv); +} + +static u32 mtp_ddi_hotplug_mask(enum hpd_pin hpd_pin) +{ + switch (hpd_pin) { + case HPD_PORT_A: + case HPD_PORT_B: + return SHOTPLUG_CTL_DDI_HPD_ENABLE(hpd_pin); + default: + return 0; + } +} + +static u32 mtp_ddi_hotplug_enables(struct intel_encoder *encoder) +{ + return mtp_ddi_hotplug_mask(encoder->hpd_pin); +} + +static u32 mtp_tc_hotplug_mask(enum hpd_pin hpd_pin) +{ + switch (hpd_pin) { + case HPD_PORT_TC1: + case HPD_PORT_TC2: + case HPD_PORT_TC3: + case HPD_PORT_TC4: + return ICP_TC_HPD_ENABLE(hpd_pin); + default: + return 0; + } +} + +static u32 mtp_tc_hotplug_enables(struct intel_encoder *encoder) +{ + return mtp_tc_hotplug_mask(encoder->hpd_pin); +} + +static void mtp_ddi_hpd_detection_setup(struct drm_i915_private *i915) +{ + intel_de_rmw(i915, SHOTPLUG_CTL_DDI, + intel_hpd_hotplug_mask(i915, mtp_ddi_hotplug_mask), + intel_hpd_hotplug_enables(i915, mtp_ddi_hotplug_enables)); +} + +static void mtp_ddi_hpd_enable_detection(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + + intel_de_rmw(i915, SHOTPLUG_CTL_DDI, + mtp_ddi_hotplug_mask(encoder->hpd_pin), + mtp_ddi_hotplug_enables(encoder)); +} + +static void mtp_tc_hpd_detection_setup(struct drm_i915_private *i915) +{ + intel_de_rmw(i915, SHOTPLUG_CTL_TC, + intel_hpd_hotplug_mask(i915, mtp_tc_hotplug_mask), + intel_hpd_hotplug_enables(i915, mtp_tc_hotplug_enables)); +} + +static void mtp_tc_hpd_enable_detection(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + + intel_de_rmw(i915, SHOTPLUG_CTL_DDI, + mtp_tc_hotplug_mask(encoder->hpd_pin), + mtp_tc_hotplug_enables(encoder)); +} + +static void mtp_hpd_invert(struct drm_i915_private *i915) +{ + u32 val = (INVERT_DDIA_HPD | + INVERT_DDIB_HPD | + INVERT_DDIC_HPD | + INVERT_TC1_HPD | + INVERT_TC2_HPD | + INVERT_TC3_HPD | + INVERT_TC4_HPD | + INVERT_DDID_HPD_MTP | + INVERT_DDIE_HPD); + intel_de_rmw(i915, SOUTH_CHICKEN1, 0, val); +} + +static void mtp_hpd_enable_detection(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + + mtp_hpd_invert(i915); + mtp_ddi_hpd_enable_detection(encoder); + mtp_tc_hpd_enable_detection(encoder); +} + +static void mtp_hpd_irq_setup(struct drm_i915_private *i915) +{ + u32 hotplug_irqs, enabled_irqs; + + enabled_irqs = intel_hpd_enabled_irqs(i915, i915->display.hotplug.pch_hpd); + hotplug_irqs = intel_hpd_hotplug_irqs(i915, i915->display.hotplug.pch_hpd); + + intel_de_write(i915, SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ); + + mtp_hpd_invert(i915); + ibx_display_interrupt_update(i915, hotplug_irqs, enabled_irqs); + + mtp_ddi_hpd_detection_setup(i915); + mtp_tc_hpd_detection_setup(i915); +} + +static bool is_xelpdp_pica_hpd_pin(enum hpd_pin hpd_pin) +{ + return hpd_pin >= HPD_PORT_TC1 && hpd_pin <= HPD_PORT_TC4; +} + +static void _xelpdp_pica_hpd_detection_setup(struct drm_i915_private *i915, + enum hpd_pin hpd_pin, bool enable) +{ + u32 mask = XELPDP_TBT_HOTPLUG_ENABLE | + XELPDP_DP_ALT_HOTPLUG_ENABLE; + + if (!is_xelpdp_pica_hpd_pin(hpd_pin)) + return; + + intel_de_rmw(i915, XELPDP_PORT_HOTPLUG_CTL(hpd_pin), + mask, enable ? mask : 0); +} + +static void xelpdp_pica_hpd_enable_detection(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + + _xelpdp_pica_hpd_detection_setup(i915, encoder->hpd_pin, true); +} + +static void xelpdp_pica_hpd_detection_setup(struct drm_i915_private *i915) +{ + struct intel_encoder *encoder; + u32 available_pins = 0; + enum hpd_pin pin; + + BUILD_BUG_ON(BITS_PER_TYPE(available_pins) < HPD_NUM_PINS); + + for_each_intel_encoder(&i915->drm, encoder) + available_pins |= BIT(encoder->hpd_pin); + + for_each_hpd_pin(pin) + _xelpdp_pica_hpd_detection_setup(i915, pin, available_pins & BIT(pin)); +} + +static void xelpdp_hpd_enable_detection(struct intel_encoder *encoder) +{ + xelpdp_pica_hpd_enable_detection(encoder); + mtp_hpd_enable_detection(encoder); +} + +static void xelpdp_hpd_irq_setup(struct drm_i915_private *i915) +{ + u32 hotplug_irqs, enabled_irqs; + + enabled_irqs = intel_hpd_enabled_irqs(i915, i915->display.hotplug.hpd); + hotplug_irqs = intel_hpd_hotplug_irqs(i915, i915->display.hotplug.hpd); + + intel_de_rmw(i915, PICAINTERRUPT_IMR, hotplug_irqs, + ~enabled_irqs & hotplug_irqs); + intel_uncore_posting_read(&i915->uncore, PICAINTERRUPT_IMR); + + xelpdp_pica_hpd_detection_setup(i915); + + if (INTEL_PCH_TYPE(i915) >= PCH_MTP) + mtp_hpd_irq_setup(i915); +} + +static u32 spt_hotplug_mask(enum hpd_pin hpd_pin) +{ + switch (hpd_pin) { + case HPD_PORT_A: + return PORTA_HOTPLUG_ENABLE; + case HPD_PORT_B: + return PORTB_HOTPLUG_ENABLE; + case HPD_PORT_C: + return PORTC_HOTPLUG_ENABLE; + case HPD_PORT_D: + return PORTD_HOTPLUG_ENABLE; + default: + return 0; + } +} + +static u32 spt_hotplug_enables(struct intel_encoder *encoder) +{ + return spt_hotplug_mask(encoder->hpd_pin); +} + +static u32 spt_hotplug2_mask(enum hpd_pin hpd_pin) +{ + switch (hpd_pin) { + case HPD_PORT_E: + return PORTE_HOTPLUG_ENABLE; + default: + return 0; + } +} + +static u32 spt_hotplug2_enables(struct intel_encoder *encoder) +{ + return spt_hotplug2_mask(encoder->hpd_pin); +} + +static void spt_hpd_detection_setup(struct drm_i915_private *dev_priv) +{ + /* Display WA #1179 WaHardHangonHotPlug: cnp */ + if (HAS_PCH_CNP(dev_priv)) { + intel_uncore_rmw(&dev_priv->uncore, SOUTH_CHICKEN1, CHASSIS_CLK_REQ_DURATION_MASK, + CHASSIS_CLK_REQ_DURATION(0xf)); + } + + /* Enable digital hotplug on the PCH */ + intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG, + intel_hpd_hotplug_mask(dev_priv, spt_hotplug_mask), + intel_hpd_hotplug_enables(dev_priv, spt_hotplug_enables)); + + intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG2, + intel_hpd_hotplug_mask(dev_priv, spt_hotplug2_mask), + intel_hpd_hotplug_enables(dev_priv, spt_hotplug2_enables)); +} + +static void spt_hpd_enable_detection(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + + /* Display WA #1179 WaHardHangonHotPlug: cnp */ + if (HAS_PCH_CNP(i915)) { + intel_uncore_rmw(&i915->uncore, SOUTH_CHICKEN1, + CHASSIS_CLK_REQ_DURATION_MASK, + CHASSIS_CLK_REQ_DURATION(0xf)); + } + + intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG, + spt_hotplug_mask(encoder->hpd_pin), + spt_hotplug_enables(encoder)); + + intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG2, + spt_hotplug2_mask(encoder->hpd_pin), + spt_hotplug2_enables(encoder)); +} + +static void spt_hpd_irq_setup(struct drm_i915_private *dev_priv) +{ + u32 hotplug_irqs, enabled_irqs; + + if (INTEL_PCH_TYPE(dev_priv) >= PCH_CNP) + intel_uncore_write(&dev_priv->uncore, SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ); + + enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd); + hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd); + + ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs); + + spt_hpd_detection_setup(dev_priv); +} + +static u32 ilk_hotplug_mask(enum hpd_pin hpd_pin) +{ + switch (hpd_pin) { + case HPD_PORT_A: + return DIGITAL_PORTA_HOTPLUG_ENABLE | + DIGITAL_PORTA_PULSE_DURATION_MASK; + default: + return 0; + } +} + +static u32 ilk_hotplug_enables(struct intel_encoder *encoder) +{ + switch (encoder->hpd_pin) { + case HPD_PORT_A: + return DIGITAL_PORTA_HOTPLUG_ENABLE | + DIGITAL_PORTA_PULSE_DURATION_2ms; + default: + return 0; + } +} + +static void ilk_hpd_detection_setup(struct drm_i915_private *dev_priv) +{ + /* + * Enable digital hotplug on the CPU, and configure the DP short pulse + * duration to 2ms (which is the minimum in the Display Port spec) + * The pulse duration bits are reserved on HSW+. + */ + intel_uncore_rmw(&dev_priv->uncore, DIGITAL_PORT_HOTPLUG_CNTRL, + intel_hpd_hotplug_mask(dev_priv, ilk_hotplug_mask), + intel_hpd_hotplug_enables(dev_priv, ilk_hotplug_enables)); +} + +static void ilk_hpd_enable_detection(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + + intel_uncore_rmw(&i915->uncore, DIGITAL_PORT_HOTPLUG_CNTRL, + ilk_hotplug_mask(encoder->hpd_pin), + ilk_hotplug_enables(encoder)); + + ibx_hpd_enable_detection(encoder); +} + +static void ilk_hpd_irq_setup(struct drm_i915_private *dev_priv) +{ + u32 hotplug_irqs, enabled_irqs; + + enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.hpd); + hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.hpd); + + if (DISPLAY_VER(dev_priv) >= 8) + bdw_update_port_irq(dev_priv, hotplug_irqs, enabled_irqs); + else + ilk_update_display_irq(dev_priv, hotplug_irqs, enabled_irqs); + + ilk_hpd_detection_setup(dev_priv); + + ibx_hpd_irq_setup(dev_priv); +} + +static u32 bxt_hotplug_mask(enum hpd_pin hpd_pin) +{ + switch (hpd_pin) { + case HPD_PORT_A: + return PORTA_HOTPLUG_ENABLE | BXT_DDIA_HPD_INVERT; + case HPD_PORT_B: + return PORTB_HOTPLUG_ENABLE | BXT_DDIB_HPD_INVERT; + case HPD_PORT_C: + return PORTC_HOTPLUG_ENABLE | BXT_DDIC_HPD_INVERT; + default: + return 0; + } +} + +static u32 bxt_hotplug_enables(struct intel_encoder *encoder) +{ + u32 hotplug; + + switch (encoder->hpd_pin) { + case HPD_PORT_A: + hotplug = PORTA_HOTPLUG_ENABLE; + if (intel_bios_encoder_hpd_invert(encoder->devdata)) + hotplug |= BXT_DDIA_HPD_INVERT; + return hotplug; + case HPD_PORT_B: + hotplug = PORTB_HOTPLUG_ENABLE; + if (intel_bios_encoder_hpd_invert(encoder->devdata)) + hotplug |= BXT_DDIB_HPD_INVERT; + return hotplug; + case HPD_PORT_C: + hotplug = PORTC_HOTPLUG_ENABLE; + if (intel_bios_encoder_hpd_invert(encoder->devdata)) + hotplug |= BXT_DDIC_HPD_INVERT; + return hotplug; + default: + return 0; + } +} + +static void bxt_hpd_detection_setup(struct drm_i915_private *dev_priv) +{ + intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG, + intel_hpd_hotplug_mask(dev_priv, bxt_hotplug_mask), + intel_hpd_hotplug_enables(dev_priv, bxt_hotplug_enables)); +} + +static void bxt_hpd_enable_detection(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + + intel_uncore_rmw(&i915->uncore, PCH_PORT_HOTPLUG, + bxt_hotplug_mask(encoder->hpd_pin), + bxt_hotplug_enables(encoder)); +} + +static void bxt_hpd_irq_setup(struct drm_i915_private *dev_priv) +{ + u32 hotplug_irqs, enabled_irqs; + + enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.hpd); + hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.hpd); + + bdw_update_port_irq(dev_priv, hotplug_irqs, enabled_irqs); + + bxt_hpd_detection_setup(dev_priv); +} + +static void i915_hpd_enable_detection(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + u32 hotplug_en = hpd_mask_i915[encoder->hpd_pin]; + + /* HPD sense and interrupt enable are one and the same */ + i915_hotplug_interrupt_update(i915, hotplug_en, hotplug_en); +} + +static void i915_hpd_irq_setup(struct drm_i915_private *dev_priv) +{ + u32 hotplug_en; + + lockdep_assert_held(&dev_priv->irq_lock); + + /* + * Note HDMI and DP share hotplug bits. Enable bits are the same for all + * generations. + */ + hotplug_en = intel_hpd_enabled_irqs(dev_priv, hpd_mask_i915); + /* + * Programming the CRT detection parameters tends to generate a spurious + * hotplug event about three seconds later. So just do it once. + */ + if (IS_G4X(dev_priv)) + hotplug_en |= CRT_HOTPLUG_ACTIVATION_PERIOD_64; + hotplug_en |= CRT_HOTPLUG_VOLTAGE_COMPARE_50; + + /* Ignore TV since it's buggy */ + i915_hotplug_interrupt_update_locked(dev_priv, + HOTPLUG_INT_EN_MASK | + CRT_HOTPLUG_VOLTAGE_COMPARE_MASK | + CRT_HOTPLUG_ACTIVATION_PERIOD_64, + hotplug_en); +} + +struct intel_hotplug_funcs { + /* Enable HPD sense and interrupts for all present encoders */ + void (*hpd_irq_setup)(struct drm_i915_private *i915); + /* Enable HPD sense for a single encoder */ + void (*hpd_enable_detection)(struct intel_encoder *encoder); +}; + +#define HPD_FUNCS(platform) \ +static const struct intel_hotplug_funcs platform##_hpd_funcs = { \ + .hpd_irq_setup = platform##_hpd_irq_setup, \ + .hpd_enable_detection = platform##_hpd_enable_detection, \ +} + +HPD_FUNCS(i915); +HPD_FUNCS(xelpdp); +HPD_FUNCS(dg1); +HPD_FUNCS(gen11); +HPD_FUNCS(bxt); +HPD_FUNCS(icp); +HPD_FUNCS(spt); +HPD_FUNCS(ilk); +#undef HPD_FUNCS + +void intel_hpd_enable_detection(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + + if (i915->display.funcs.hotplug) + i915->display.funcs.hotplug->hpd_enable_detection(encoder); +} + +void intel_hpd_irq_setup(struct drm_i915_private *i915) +{ + if (i915->display_irqs_enabled && i915->display.funcs.hotplug) + i915->display.funcs.hotplug->hpd_irq_setup(i915); +} + +void intel_hotplug_irq_init(struct drm_i915_private *i915) +{ + intel_hpd_init_pins(i915); + + intel_hpd_init_early(i915); + + if (HAS_GMCH(i915)) { + if (I915_HAS_HOTPLUG(i915)) + i915->display.funcs.hotplug = &i915_hpd_funcs; + } else { + if (HAS_PCH_DG2(i915)) + i915->display.funcs.hotplug = &icp_hpd_funcs; + else if (HAS_PCH_DG1(i915)) + i915->display.funcs.hotplug = &dg1_hpd_funcs; + else if (DISPLAY_VER(i915) >= 14) + i915->display.funcs.hotplug = &xelpdp_hpd_funcs; + else if (DISPLAY_VER(i915) >= 11) + i915->display.funcs.hotplug = &gen11_hpd_funcs; + else if (IS_GEMINILAKE(i915) || IS_BROXTON(i915)) + i915->display.funcs.hotplug = &bxt_hpd_funcs; + else if (INTEL_PCH_TYPE(i915) >= PCH_ICP) + i915->display.funcs.hotplug = &icp_hpd_funcs; + else if (INTEL_PCH_TYPE(i915) >= PCH_SPT) + i915->display.funcs.hotplug = &spt_hpd_funcs; + else + i915->display.funcs.hotplug = &ilk_hpd_funcs; + } +} diff --git a/drivers/gpu/drm/i915/display/intel_hotplug_irq.h b/drivers/gpu/drm/i915/display/intel_hotplug_irq.h new file mode 100644 index 0000000000000000000000000000000000000000..e4db752df096758e14f731a53ba1623b1c0d4b06 --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_hotplug_irq.h @@ -0,0 +1,35 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2023 Intel Corporation + */ + +#ifndef __INTEL_HOTPLUG_IRQ_H__ +#define __INTEL_HOTPLUG_IRQ_H__ + +#include + +struct drm_i915_private; +struct intel_encoder; + +u32 i9xx_hpd_irq_ack(struct drm_i915_private *i915); + +void i9xx_hpd_irq_handler(struct drm_i915_private *i915, u32 hotplug_status); +void ibx_hpd_irq_handler(struct drm_i915_private *i915, u32 hotplug_trigger); +void ilk_hpd_irq_handler(struct drm_i915_private *i915, u32 hotplug_trigger); +void gen11_hpd_irq_handler(struct drm_i915_private *i915, u32 iir); +void bxt_hpd_irq_handler(struct drm_i915_private *i915, u32 hotplug_trigger); +void xelpdp_pica_irq_handler(struct drm_i915_private *i915, u32 iir); +void icp_irq_handler(struct drm_i915_private *i915, u32 pch_iir); +void spt_irq_handler(struct drm_i915_private *i915, u32 pch_iir); + +void i915_hotplug_interrupt_update_locked(struct drm_i915_private *i915, + u32 mask, u32 bits); +void i915_hotplug_interrupt_update(struct drm_i915_private *i915, + u32 mask, u32 bits); + +void intel_hpd_enable_detection(struct intel_encoder *encoder); +void intel_hpd_irq_setup(struct drm_i915_private *i915); + +void intel_hotplug_irq_init(struct drm_i915_private *i915); + +#endif /* __INTEL_HOTPLUG_IRQ_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_hti.c b/drivers/gpu/drm/i915/display/intel_hti.c index c518efebdf779fe76741b091007cd2a7920aa873..a92d008d4e6e593366b5e5a8d15152e4dc0def24 100644 --- a/drivers/gpu/drm/i915/display/intel_hti.c +++ b/drivers/gpu/drm/i915/display/intel_hti.c @@ -15,7 +15,7 @@ void intel_hti_init(struct drm_i915_private *i915) * If the platform has HTI, we need to find out whether it has reserved * any display resources before we create our display outputs. */ - if (INTEL_INFO(i915)->display.has_hti) + if (DISPLAY_INFO(i915)->has_hti) i915->display.hti.state = intel_de_read(i915, HDPORT_STATE); } diff --git a/drivers/gpu/drm/i915/display/intel_load_detect.c b/drivers/gpu/drm/i915/display/intel_load_detect.c new file mode 100644 index 0000000000000000000000000000000000000000..d5a0aecf3e8ffc744003b42a79c71ec59bd8e946 --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_load_detect.c @@ -0,0 +1,225 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2023 Intel Corporation + */ + +#include +#include +#include + +#include "i915_drv.h" +#include "intel_atomic.h" +#include "intel_crtc.h" +#include "intel_display_types.h" +#include "intel_load_detect.h" + +/* VESA 640x480x72Hz mode to set on the pipe */ +static const struct drm_display_mode load_detect_mode = { + DRM_MODE("640x480", DRM_MODE_TYPE_DEFAULT, 31500, 640, 664, + 704, 832, 0, 480, 489, 491, 520, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC), +}; + +static int intel_modeset_disable_planes(struct drm_atomic_state *state, + struct drm_crtc *crtc) +{ + struct drm_plane *plane; + struct drm_plane_state *plane_state; + int ret, i; + + ret = drm_atomic_add_affected_planes(state, crtc); + if (ret) + return ret; + + for_each_new_plane_in_state(state, plane, plane_state, i) { + if (plane_state->crtc != crtc) + continue; + + ret = drm_atomic_set_crtc_for_plane(plane_state, NULL); + if (ret) + return ret; + + drm_atomic_set_fb_for_plane(plane_state, NULL); + } + + return 0; +} + +struct drm_atomic_state * +intel_load_detect_get_pipe(struct drm_connector *connector, + struct drm_modeset_acquire_ctx *ctx) +{ + struct intel_encoder *encoder = + intel_attached_encoder(to_intel_connector(connector)); + struct intel_crtc *possible_crtc; + struct intel_crtc *crtc = NULL; + struct drm_device *dev = encoder->base.dev; + struct drm_i915_private *dev_priv = to_i915(dev); + struct drm_mode_config *config = &dev->mode_config; + struct drm_atomic_state *state = NULL, *restore_state = NULL; + struct drm_connector_state *connector_state; + struct intel_crtc_state *crtc_state; + int ret; + + drm_dbg_kms(&dev_priv->drm, "[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n", + connector->base.id, connector->name, + encoder->base.base.id, encoder->base.name); + + drm_WARN_ON(dev, !drm_modeset_is_locked(&config->connection_mutex)); + + /* + * Algorithm gets a little messy: + * + * - if the connector already has an assigned crtc, use it (but make + * sure it's on first) + * + * - try to find the first unused crtc that can drive this connector, + * and use that if we find one + */ + + /* See if we already have a CRTC for this connector */ + if (connector->state->crtc) { + crtc = to_intel_crtc(connector->state->crtc); + + ret = drm_modeset_lock(&crtc->base.mutex, ctx); + if (ret) + goto fail; + + /* Make sure the crtc and connector are running */ + goto found; + } + + /* Find an unused one (if possible) */ + for_each_intel_crtc(dev, possible_crtc) { + if (!(encoder->base.possible_crtcs & + drm_crtc_mask(&possible_crtc->base))) + continue; + + ret = drm_modeset_lock(&possible_crtc->base.mutex, ctx); + if (ret) + goto fail; + + if (possible_crtc->base.state->enable) { + drm_modeset_unlock(&possible_crtc->base.mutex); + continue; + } + + crtc = possible_crtc; + break; + } + + /* + * If we didn't find an unused CRTC, don't use any. + */ + if (!crtc) { + drm_dbg_kms(&dev_priv->drm, + "no pipe available for load-detect\n"); + ret = -ENODEV; + goto fail; + } + +found: + state = drm_atomic_state_alloc(dev); + restore_state = drm_atomic_state_alloc(dev); + if (!state || !restore_state) { + ret = -ENOMEM; + goto fail; + } + + state->acquire_ctx = ctx; + to_intel_atomic_state(state)->internal = true; + + restore_state->acquire_ctx = ctx; + to_intel_atomic_state(restore_state)->internal = true; + + connector_state = drm_atomic_get_connector_state(state, connector); + if (IS_ERR(connector_state)) { + ret = PTR_ERR(connector_state); + goto fail; + } + + ret = drm_atomic_set_crtc_for_connector(connector_state, &crtc->base); + if (ret) + goto fail; + + crtc_state = intel_atomic_get_crtc_state(state, crtc); + if (IS_ERR(crtc_state)) { + ret = PTR_ERR(crtc_state); + goto fail; + } + + crtc_state->uapi.active = true; + + ret = drm_atomic_set_mode_for_crtc(&crtc_state->uapi, + &load_detect_mode); + if (ret) + goto fail; + + ret = intel_modeset_disable_planes(state, &crtc->base); + if (ret) + goto fail; + + ret = PTR_ERR_OR_ZERO(drm_atomic_get_connector_state(restore_state, connector)); + if (!ret) + ret = PTR_ERR_OR_ZERO(drm_atomic_get_crtc_state(restore_state, &crtc->base)); + if (!ret) + ret = drm_atomic_add_affected_planes(restore_state, &crtc->base); + if (ret) { + drm_dbg_kms(&dev_priv->drm, + "Failed to create a copy of old state to restore: %i\n", + ret); + goto fail; + } + + ret = drm_atomic_commit(state); + if (ret) { + drm_dbg_kms(&dev_priv->drm, + "failed to set mode on load-detect pipe\n"); + goto fail; + } + + drm_atomic_state_put(state); + + /* let the connector get through one full cycle before testing */ + intel_crtc_wait_for_next_vblank(crtc); + + return restore_state; + +fail: + if (state) { + drm_atomic_state_put(state); + state = NULL; + } + if (restore_state) { + drm_atomic_state_put(restore_state); + restore_state = NULL; + } + + if (ret == -EDEADLK) + return ERR_PTR(ret); + + return NULL; +} + +void intel_load_detect_release_pipe(struct drm_connector *connector, + struct drm_atomic_state *state, + struct drm_modeset_acquire_ctx *ctx) +{ + struct intel_encoder *intel_encoder = + intel_attached_encoder(to_intel_connector(connector)); + struct drm_i915_private *i915 = to_i915(intel_encoder->base.dev); + struct drm_encoder *encoder = &intel_encoder->base; + int ret; + + drm_dbg_kms(&i915->drm, "[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n", + connector->base.id, connector->name, + encoder->base.id, encoder->name); + + if (IS_ERR_OR_NULL(state)) + return; + + ret = drm_atomic_helper_commit_duplicated_state(state, ctx); + if (ret) + drm_dbg_kms(&i915->drm, + "Couldn't release load detect pipe: %i\n", ret); + drm_atomic_state_put(state); +} diff --git a/drivers/gpu/drm/i915/display/intel_load_detect.h b/drivers/gpu/drm/i915/display/intel_load_detect.h new file mode 100644 index 0000000000000000000000000000000000000000..aed51901b9ba8951e5ab48882a85e60d00dddbc0 --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_load_detect.h @@ -0,0 +1,20 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2023 Intel Corporation + */ + +#ifndef __INTEL_LOAD_DETECT_H__ +#define __INTEL_LOAD_DETECT_H__ + +struct drm_atomic_state; +struct drm_connector; +struct drm_modeset_acquire_ctx; + +struct drm_atomic_state * +intel_load_detect_get_pipe(struct drm_connector *connector, + struct drm_modeset_acquire_ctx *ctx); +void intel_load_detect_release_pipe(struct drm_connector *connector, + struct drm_atomic_state *old, + struct drm_modeset_acquire_ctx *ctx); + +#endif /* __INTEL_LOAD_DETECT_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_lvds.c b/drivers/gpu/drm/i915/display/intel_lvds.c index 0de44b3631cde7242a12e60ac078dc2230b1bd94..3ace56979b70e0bed3bc442624a23a603a8d7036 100644 --- a/drivers/gpu/drm/i915/display/intel_lvds.c +++ b/drivers/gpu/drm/i915/display/intel_lvds.c @@ -150,7 +150,7 @@ static void intel_lvds_get_config(struct intel_encoder *encoder, if (DISPLAY_VER(dev_priv) < 4) { tmp = intel_de_read(dev_priv, PFIT_CONTROL); - crtc_state->gmch_pfit.control |= tmp & PANEL_8TO6_DITHER_ENABLE; + crtc_state->gmch_pfit.control |= tmp & PFIT_PANEL_8TO6_DITHER_ENABLE; } crtc_state->hw.adjusted_mode.crtc_clock = crtc_state->port_clock; @@ -437,6 +437,7 @@ static int intel_lvds_compute_config(struct intel_encoder *encoder, crtc_state->pipe_bpp = lvds_bpp; } + crtc_state->sink_format = INTEL_OUTPUT_FORMAT_RGB; crtc_state->output_format = INTEL_OUTPUT_FORMAT_RGB; /* @@ -943,17 +944,8 @@ void intel_lvds_init(struct drm_i915_private *i915) */ mutex_lock(&i915->drm.mode_config.mutex); if (vga_switcheroo_handler_flags() & VGA_SWITCHEROO_CAN_SWITCH_DDC) { - const struct edid *edid; - - /* FIXME: Make drm_get_edid_switcheroo() return drm_edid */ - edid = drm_get_edid_switcheroo(&connector->base, - intel_gmbus_get_adapter(i915, pin)); - if (edid) { - drm_edid = drm_edid_alloc(edid, (edid->extensions + 1) * EDID_LENGTH); - kfree(edid); - } else { - drm_edid = NULL; - } + drm_edid = drm_edid_read_switcheroo(&connector->base, + intel_gmbus_get_adapter(i915, pin)); } else { drm_edid = drm_edid_read_ddc(&connector->base, intel_gmbus_get_adapter(i915, pin)); diff --git a/drivers/gpu/drm/i915/display/intel_modeset_lock.c b/drivers/gpu/drm/i915/display/intel_modeset_lock.c new file mode 100644 index 0000000000000000000000000000000000000000..8fb6fd849a75db33fe0c0c2ff59ecf000f877598 --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_modeset_lock.c @@ -0,0 +1,50 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2023 Intel Corporation + */ + +#include + +#include "intel_display_types.h" +#include "intel_modeset_lock.h" + +void _intel_modeset_lock_begin(struct drm_modeset_acquire_ctx *ctx, + struct intel_atomic_state *state, + unsigned int flags, int *ret) +{ + drm_modeset_acquire_init(ctx, flags); + + if (state) + state->base.acquire_ctx = ctx; + + *ret = -EDEADLK; +} + +bool _intel_modeset_lock_loop(int *ret) +{ + if (*ret == -EDEADLK) { + *ret = 0; + return true; + } + + return false; +} + +void _intel_modeset_lock_end(struct drm_modeset_acquire_ctx *ctx, + struct intel_atomic_state *state, + int *ret) +{ + if (*ret == -EDEADLK) { + if (state) + drm_atomic_state_clear(&state->base); + + *ret = drm_modeset_backoff(ctx); + if (*ret == 0) { + *ret = -EDEADLK; + return; + } + } + + drm_modeset_drop_locks(ctx); + drm_modeset_acquire_fini(ctx); +} diff --git a/drivers/gpu/drm/i915/display/intel_modeset_lock.h b/drivers/gpu/drm/i915/display/intel_modeset_lock.h new file mode 100644 index 0000000000000000000000000000000000000000..edb5099bcd99c0a31bc3e48dfd6e7a54ef8f6732 --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_modeset_lock.h @@ -0,0 +1,33 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2023 Intel Corporation + */ + +#ifndef __INTEL_MODESET_LOCK_H__ +#define __INTEL_MODESET_LOCK_H__ + +#include + +struct drm_modeset_acquire_ctx; +struct intel_atomic_state; + +void _intel_modeset_lock_begin(struct drm_modeset_acquire_ctx *ctx, + struct intel_atomic_state *state, + unsigned int flags, + int *ret); +bool _intel_modeset_lock_loop(int *ret); +void _intel_modeset_lock_end(struct drm_modeset_acquire_ctx *ctx, + struct intel_atomic_state *state, + int *ret); + +/* + * Note that one must always use "continue" rather than + * "break" or "return" to handle errors within the + * intel_modeset_lock_ctx_retry() block. + */ +#define intel_modeset_lock_ctx_retry(ctx, state, flags, ret) \ + for (_intel_modeset_lock_begin((ctx), (state), (flags), &(ret)); \ + _intel_modeset_lock_loop(&(ret)); \ + _intel_modeset_lock_end((ctx), (state), &(ret))) + +#endif /* __INTEL_MODESET_LOCK_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_modeset_setup.c b/drivers/gpu/drm/i915/display/intel_modeset_setup.c index 134b943f195333867e291183ec1729b49ae2c1e6..b8f43efb0ab5a2b71abde08d46cf76a8cfcbc8c7 100644 --- a/drivers/gpu/drm/i915/display/intel_modeset_setup.c +++ b/drivers/gpu/drm/i915/display/intel_modeset_setup.c @@ -26,28 +26,22 @@ #include "intel_fifo_underrun.h" #include "intel_modeset_setup.h" #include "intel_pch_display.h" +#include "intel_pmdemand.h" +#include "intel_tc.h" #include "intel_vblank.h" #include "intel_wm.h" #include "skl_watermark.h" -static void intel_crtc_disable_noatomic(struct intel_crtc *crtc, - struct drm_modeset_acquire_ctx *ctx) +static void intel_crtc_disable_noatomic_begin(struct intel_crtc *crtc, + struct drm_modeset_acquire_ctx *ctx) { - struct intel_encoder *encoder; struct drm_i915_private *i915 = to_i915(crtc->base.dev); - struct intel_bw_state *bw_state = - to_intel_bw_state(i915->display.bw.obj.state); - struct intel_cdclk_state *cdclk_state = - to_intel_cdclk_state(i915->display.cdclk.obj.state); - struct intel_dbuf_state *dbuf_state = - to_intel_dbuf_state(i915->display.dbuf.obj.state); struct intel_crtc_state *crtc_state = to_intel_crtc_state(crtc->base.state); struct intel_plane *plane; struct drm_atomic_state *state; - struct intel_crtc_state *temp_crtc_state; + struct intel_crtc *temp_crtc; enum pipe pipe = crtc->pipe; - int ret; if (!crtc_state->hw.active) return; @@ -69,12 +63,20 @@ static void intel_crtc_disable_noatomic(struct intel_crtc *crtc, } state->acquire_ctx = ctx; + to_intel_atomic_state(state)->internal = true; /* Everything's already locked, -EDEADLK can't happen. */ - temp_crtc_state = intel_atomic_get_crtc_state(state, crtc); - ret = drm_atomic_add_affected_connectors(state, &crtc->base); + for_each_intel_crtc_in_pipe_mask(&i915->drm, temp_crtc, + BIT(pipe) | + intel_crtc_bigjoiner_slave_pipes(crtc_state)) { + struct intel_crtc_state *temp_crtc_state = + intel_atomic_get_crtc_state(state, temp_crtc); + int ret; + + ret = drm_atomic_add_affected_connectors(state, &temp_crtc->base); - drm_WARN_ON(&i915->drm, IS_ERR(temp_crtc_state) || ret); + drm_WARN_ON(&i915->drm, IS_ERR(temp_crtc_state) || ret); + } i915->display.funcs.display->crtc_disable(to_intel_atomic_state(state), crtc); @@ -87,16 +89,86 @@ static void intel_crtc_disable_noatomic(struct intel_crtc *crtc, crtc->active = false; crtc->base.enabled = false; - drm_WARN_ON(&i915->drm, - drm_atomic_set_mode_for_crtc(&crtc_state->uapi, NULL) < 0); - crtc_state->uapi.active = false; - crtc_state->uapi.connector_mask = 0; - crtc_state->uapi.encoder_mask = 0; - intel_crtc_free_hw_state(crtc_state); - memset(&crtc_state->hw, 0, sizeof(crtc_state->hw)); + if (crtc_state->shared_dpll) + intel_unreference_shared_dpll_crtc(crtc, + crtc_state->shared_dpll, + &crtc_state->shared_dpll->state); +} + +static void set_encoder_for_connector(struct intel_connector *connector, + struct intel_encoder *encoder) +{ + struct drm_connector_state *conn_state = connector->base.state; + + if (conn_state->crtc) + drm_connector_put(&connector->base); + + if (encoder) { + conn_state->best_encoder = &encoder->base; + conn_state->crtc = encoder->base.crtc; + drm_connector_get(&connector->base); + } else { + conn_state->best_encoder = NULL; + conn_state->crtc = NULL; + } +} + +static void reset_encoder_connector_state(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + struct intel_pmdemand_state *pmdemand_state = + to_intel_pmdemand_state(i915->display.pmdemand.obj.state); + struct intel_connector *connector; + struct drm_connector_list_iter conn_iter; + + drm_connector_list_iter_begin(&i915->drm, &conn_iter); + for_each_intel_connector_iter(connector, &conn_iter) { + if (connector->base.encoder != &encoder->base) + continue; + + /* Clear the corresponding bit in pmdemand active phys mask */ + intel_pmdemand_update_phys_mask(i915, encoder, + pmdemand_state, false); + + set_encoder_for_connector(connector, NULL); + + connector->base.dpms = DRM_MODE_DPMS_OFF; + connector->base.encoder = NULL; + } + drm_connector_list_iter_end(&conn_iter); +} - for_each_encoder_on_crtc(&i915->drm, &crtc->base, encoder) +static void reset_crtc_encoder_state(struct intel_crtc *crtc) +{ + struct drm_i915_private *i915 = to_i915(crtc->base.dev); + struct intel_encoder *encoder; + + for_each_encoder_on_crtc(&i915->drm, &crtc->base, encoder) { + reset_encoder_connector_state(encoder); encoder->base.crtc = NULL; + } +} + +static void intel_crtc_disable_noatomic_complete(struct intel_crtc *crtc) +{ + struct drm_i915_private *i915 = to_i915(crtc->base.dev); + struct intel_bw_state *bw_state = + to_intel_bw_state(i915->display.bw.obj.state); + struct intel_cdclk_state *cdclk_state = + to_intel_cdclk_state(i915->display.cdclk.obj.state); + struct intel_dbuf_state *dbuf_state = + to_intel_dbuf_state(i915->display.dbuf.obj.state); + struct intel_pmdemand_state *pmdemand_state = + to_intel_pmdemand_state(i915->display.pmdemand.obj.state); + struct intel_crtc_state *crtc_state = + to_intel_crtc_state(crtc->base.state); + enum pipe pipe = crtc->pipe; + + __drm_atomic_helper_crtc_destroy_state(&crtc_state->uapi); + intel_crtc_free_hw_state(crtc_state); + intel_crtc_state_reset(crtc_state, crtc); + + reset_crtc_encoder_state(crtc); intel_fbc_disable(crtc); intel_update_watermarks(i915); @@ -111,6 +183,120 @@ static void intel_crtc_disable_noatomic(struct intel_crtc *crtc, bw_state->data_rate[pipe] = 0; bw_state->num_active_planes[pipe] = 0; + + intel_pmdemand_update_port_clock(i915, pmdemand_state, pipe, 0); +} + +/* + * Return all the pipes using a transcoder in @transcoder_mask. + * For bigjoiner configs return only the bigjoiner master. + */ +static u8 get_transcoder_pipes(struct drm_i915_private *i915, + u8 transcoder_mask) +{ + struct intel_crtc *temp_crtc; + u8 pipes = 0; + + for_each_intel_crtc(&i915->drm, temp_crtc) { + struct intel_crtc_state *temp_crtc_state = + to_intel_crtc_state(temp_crtc->base.state); + + if (temp_crtc_state->cpu_transcoder == INVALID_TRANSCODER) + continue; + + if (intel_crtc_is_bigjoiner_slave(temp_crtc_state)) + continue; + + if (transcoder_mask & BIT(temp_crtc_state->cpu_transcoder)) + pipes |= BIT(temp_crtc->pipe); + } + + return pipes; +} + +/* + * Return the port sync master and slave pipes linked to @crtc. + * For bigjoiner configs return only the bigjoiner master pipes. + */ +static void get_portsync_pipes(struct intel_crtc *crtc, + u8 *master_pipe_mask, u8 *slave_pipes_mask) +{ + struct drm_i915_private *i915 = to_i915(crtc->base.dev); + struct intel_crtc_state *crtc_state = + to_intel_crtc_state(crtc->base.state); + struct intel_crtc *master_crtc; + struct intel_crtc_state *master_crtc_state; + enum transcoder master_transcoder; + + if (!is_trans_port_sync_mode(crtc_state)) { + *master_pipe_mask = BIT(crtc->pipe); + *slave_pipes_mask = 0; + + return; + } + + if (is_trans_port_sync_master(crtc_state)) + master_transcoder = crtc_state->cpu_transcoder; + else + master_transcoder = crtc_state->master_transcoder; + + *master_pipe_mask = get_transcoder_pipes(i915, BIT(master_transcoder)); + drm_WARN_ON(&i915->drm, !is_power_of_2(*master_pipe_mask)); + + master_crtc = intel_crtc_for_pipe(i915, ffs(*master_pipe_mask) - 1); + master_crtc_state = to_intel_crtc_state(master_crtc->base.state); + *slave_pipes_mask = get_transcoder_pipes(i915, master_crtc_state->sync_mode_slaves_mask); +} + +static u8 get_bigjoiner_slave_pipes(struct drm_i915_private *i915, u8 master_pipes_mask) +{ + struct intel_crtc *master_crtc; + u8 pipes = 0; + + for_each_intel_crtc_in_pipe_mask(&i915->drm, master_crtc, master_pipes_mask) { + struct intel_crtc_state *master_crtc_state = + to_intel_crtc_state(master_crtc->base.state); + + pipes |= intel_crtc_bigjoiner_slave_pipes(master_crtc_state); + } + + return pipes; +} + +static void intel_crtc_disable_noatomic(struct intel_crtc *crtc, + struct drm_modeset_acquire_ctx *ctx) +{ + struct drm_i915_private *i915 = to_i915(crtc->base.dev); + u8 portsync_master_mask; + u8 portsync_slaves_mask; + u8 bigjoiner_slaves_mask; + struct intel_crtc *temp_crtc; + + /* TODO: Add support for MST */ + get_portsync_pipes(crtc, &portsync_master_mask, &portsync_slaves_mask); + bigjoiner_slaves_mask = get_bigjoiner_slave_pipes(i915, + portsync_master_mask | + portsync_slaves_mask); + + drm_WARN_ON(&i915->drm, + portsync_master_mask & portsync_slaves_mask || + portsync_master_mask & bigjoiner_slaves_mask || + portsync_slaves_mask & bigjoiner_slaves_mask); + + for_each_intel_crtc_in_pipe_mask(&i915->drm, temp_crtc, bigjoiner_slaves_mask) + intel_crtc_disable_noatomic_begin(temp_crtc, ctx); + + for_each_intel_crtc_in_pipe_mask(&i915->drm, temp_crtc, portsync_slaves_mask) + intel_crtc_disable_noatomic_begin(temp_crtc, ctx); + + for_each_intel_crtc_in_pipe_mask(&i915->drm, temp_crtc, portsync_master_mask) + intel_crtc_disable_noatomic_begin(temp_crtc, ctx); + + for_each_intel_crtc_in_pipe_mask(&i915->drm, temp_crtc, + bigjoiner_slaves_mask | + portsync_slaves_mask | + portsync_master_mask) + intel_crtc_disable_noatomic_complete(temp_crtc); } static void intel_modeset_update_connector_atomic_state(struct drm_i915_private *i915) @@ -124,8 +310,7 @@ static void intel_modeset_update_connector_atomic_state(struct drm_i915_private struct intel_encoder *encoder = to_intel_encoder(connector->base.encoder); - if (conn_state->crtc) - drm_connector_put(&connector->base); + set_encoder_for_connector(connector, encoder); if (encoder) { struct intel_crtc *crtc = @@ -133,14 +318,7 @@ static void intel_modeset_update_connector_atomic_state(struct drm_i915_private const struct intel_crtc_state *crtc_state = to_intel_crtc_state(crtc->base.state); - conn_state->best_encoder = &encoder->base; - conn_state->crtc = &crtc->base; conn_state->max_bpc = (crtc_state->pipe_bpp ?: 24) / 3; - - drm_connector_get(&connector->base); - } else { - conn_state->best_encoder = NULL; - conn_state->crtc = NULL; } } drm_connector_list_iter_end(&conn_iter); @@ -213,6 +391,21 @@ static bool intel_crtc_has_encoders(struct intel_crtc *crtc) return false; } +static bool intel_crtc_needs_link_reset(struct intel_crtc *crtc) +{ + struct drm_device *dev = crtc->base.dev; + struct intel_encoder *encoder; + + for_each_encoder_on_crtc(dev, &crtc->base, encoder) { + struct intel_digital_port *dig_port = enc_to_dig_port(encoder); + + if (dig_port && intel_tc_port_link_needs_reset(dig_port)) + return true; + } + + return false; +} + static struct intel_connector *intel_encoder_find_connector(struct intel_encoder *encoder) { struct drm_i915_private *i915 = to_i915(encoder->base.dev); @@ -255,11 +448,12 @@ static void intel_sanitize_fifo_underrun_reporting(const struct intel_crtc_state !HAS_GMCH(i915)); } -static void intel_sanitize_crtc(struct intel_crtc *crtc, +static bool intel_sanitize_crtc(struct intel_crtc *crtc, struct drm_modeset_acquire_ctx *ctx) { struct drm_i915_private *i915 = to_i915(crtc->base.dev); struct intel_crtc_state *crtc_state = to_intel_crtc_state(crtc->base.state); + bool needs_link_reset; if (crtc_state->hw.active) { struct intel_plane *plane; @@ -279,13 +473,67 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc, intel_color_commit_arm(crtc_state); } + if (!crtc_state->hw.active || + intel_crtc_is_bigjoiner_slave(crtc_state)) + return false; + + needs_link_reset = intel_crtc_needs_link_reset(crtc); + /* * Adjust the state of the output pipe according to whether we have * active connectors/encoders. */ - if (crtc_state->hw.active && !intel_crtc_has_encoders(crtc) && - !intel_crtc_is_bigjoiner_slave(crtc_state)) - intel_crtc_disable_noatomic(crtc, ctx); + if (!needs_link_reset && intel_crtc_has_encoders(crtc)) + return false; + + intel_crtc_disable_noatomic(crtc, ctx); + + /* + * The HPD state on other active/disconnected TC ports may be stuck in + * the connected state until this port is disabled and a ~10ms delay has + * passed, wait here for that so that sanitizing other CRTCs will see the + * up-to-date HPD state. + */ + if (needs_link_reset) + msleep(20); + + return true; +} + +static void intel_sanitize_all_crtcs(struct drm_i915_private *i915, + struct drm_modeset_acquire_ctx *ctx) +{ + struct intel_crtc *crtc; + u32 crtcs_forced_off = 0; + + /* + * An active and disconnected TypeC port prevents the HPD live state + * to get updated on other active/disconnected TypeC ports, so after + * a port gets disabled the CRTCs using other TypeC ports must be + * rechecked wrt. their link status. + */ + for (;;) { + u32 old_mask = crtcs_forced_off; + + for_each_intel_crtc(&i915->drm, crtc) { + u32 crtc_mask = drm_crtc_mask(&crtc->base); + + if (crtcs_forced_off & crtc_mask) + continue; + + if (intel_sanitize_crtc(crtc, ctx)) + crtcs_forced_off |= crtc_mask; + } + if (crtcs_forced_off == old_mask) + break; + } + + for_each_intel_crtc(&i915->drm, crtc) { + struct intel_crtc_state *crtc_state = + to_intel_crtc_state(crtc->base.state); + + intel_crtc_state_dump(crtc_state, NULL, "setup_hw_state"); + } } static bool has_bogus_dpll_config(const struct intel_crtc_state *crtc_state) @@ -315,6 +563,8 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder) struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc); struct intel_crtc_state *crtc_state = crtc ? to_intel_crtc_state(crtc->base.state) : NULL; + struct intel_pmdemand_state *pmdemand_state = + to_intel_pmdemand_state(i915->display.pmdemand.obj.state); /* * We need to check both for a crtc link (meaning that the encoder is @@ -338,6 +588,10 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder) encoder->base.base.id, encoder->base.name); + /* Clear the corresponding bit in pmdemand active phys mask */ + intel_pmdemand_update_phys_mask(i915, encoder, + pmdemand_state, false); + /* * Connector is active, but has no active pipe. This is fallout * from our resume register restoring. Disable the encoder @@ -424,6 +678,8 @@ static void intel_modeset_readout_hw_state(struct drm_i915_private *i915) to_intel_cdclk_state(i915->display.cdclk.obj.state); struct intel_dbuf_state *dbuf_state = to_intel_dbuf_state(i915->display.dbuf.obj.state); + struct intel_pmdemand_state *pmdemand_state = + to_intel_pmdemand_state(i915->display.pmdemand.obj.state); enum pipe pipe; struct intel_crtc *crtc; struct intel_encoder *encoder; @@ -487,7 +743,15 @@ static void intel_modeset_readout_hw_state(struct drm_i915_private *i915) intel_encoder_get_config(encoder, slave_crtc_state); } } + + intel_pmdemand_update_phys_mask(i915, encoder, + pmdemand_state, + true); } else { + intel_pmdemand_update_phys_mask(i915, encoder, + pmdemand_state, + false); + encoder->base.crtc = NULL; } @@ -559,7 +823,8 @@ static void intel_modeset_readout_hw_state(struct drm_i915_private *i915) */ crtc_state->inherited = true; - intel_crtc_update_active_timings(crtc_state); + intel_crtc_update_active_timings(crtc_state, + crtc_state->vrr.enable); intel_crtc_copy_hw_to_uapi_state(crtc_state); } @@ -603,8 +868,13 @@ static void intel_modeset_readout_hw_state(struct drm_i915_private *i915) cdclk_state->min_voltage_level[crtc->pipe] = crtc_state->min_voltage_level; + intel_pmdemand_update_port_clock(i915, pmdemand_state, pipe, + crtc_state->port_clock); + intel_bw_crtc_update(bw_state, crtc_state); } + + intel_pmdemand_init_pmdemand_params(i915, pmdemand_state); } static void @@ -698,16 +968,14 @@ void intel_modeset_setup_hw_state(struct drm_i915_private *i915, for_each_intel_encoder(&i915->drm, encoder) intel_sanitize_encoder(encoder); - for_each_intel_crtc(&i915->drm, crtc) { - struct intel_crtc_state *crtc_state = - to_intel_crtc_state(crtc->base.state); - - intel_sanitize_crtc(crtc, ctx); - intel_crtc_state_dump(crtc_state, NULL, "setup_hw_state"); - } - + /* + * Sanitizing CRTCs needs their connector atomic state to be + * up-to-date, so ensure that already here. + */ intel_modeset_update_connector_atomic_state(i915); + intel_sanitize_all_crtcs(i915, ctx); + intel_dpll_sanitize_state(i915); intel_wm_get_hw_state(i915); diff --git a/drivers/gpu/drm/i915/display/intel_modeset_verify.c b/drivers/gpu/drm/i915/display/intel_modeset_verify.c index 842d70f0dfd2a12b2b847ee1333b0e39c42e1ebf..138144a65a457d8104d6779ca672df0cdbcd34e6 100644 --- a/drivers/gpu/drm/i915/display/intel_modeset_verify.c +++ b/drivers/gpu/drm/i915/display/intel_modeset_verify.c @@ -11,6 +11,7 @@ #include "intel_atomic.h" #include "intel_crtc.h" #include "intel_crtc_state_dump.h" +#include "intel_cx0_phy.h" #include "intel_display.h" #include "intel_display_types.h" #include "intel_fdi.h" @@ -34,27 +35,28 @@ static void intel_connector_verify_state(struct intel_crtc_state *crtc_state, if (connector->get_hw_state(connector)) { struct intel_encoder *encoder = intel_attached_encoder(connector); - I915_STATE_WARN(!crtc_state, + I915_STATE_WARN(i915, !crtc_state, "connector enabled without attached crtc\n"); if (!crtc_state) return; - I915_STATE_WARN(!crtc_state->hw.active, + I915_STATE_WARN(i915, !crtc_state->hw.active, "connector is active, but attached crtc isn't\n"); if (!encoder || encoder->type == INTEL_OUTPUT_DP_MST) return; - I915_STATE_WARN(conn_state->best_encoder != &encoder->base, + I915_STATE_WARN(i915, + conn_state->best_encoder != &encoder->base, "atomic encoder doesn't match attached encoder\n"); - I915_STATE_WARN(conn_state->crtc != encoder->base.crtc, + I915_STATE_WARN(i915, conn_state->crtc != encoder->base.crtc, "attached encoder crtc differs from connector crtc\n"); } else { - I915_STATE_WARN(crtc_state && crtc_state->hw.active, + I915_STATE_WARN(i915, crtc_state && crtc_state->hw.active, "attached crtc is active, but connector isn't\n"); - I915_STATE_WARN(!crtc_state && conn_state->best_encoder, + I915_STATE_WARN(i915, !crtc_state && conn_state->best_encoder, "best encoder set without crtc!\n"); } } @@ -79,7 +81,7 @@ verify_connector_state(struct intel_atomic_state *state, intel_connector_verify_state(crtc_state, new_conn_state); - I915_STATE_WARN(new_conn_state->best_encoder != encoder, + I915_STATE_WARN(to_i915(connector->dev), new_conn_state->best_encoder != encoder, "connector's atomic encoder doesn't match legacy encoder\n"); } } @@ -130,15 +132,15 @@ verify_encoder_state(struct drm_i915_private *dev_priv, struct intel_atomic_stat found = true; enabled = true; - I915_STATE_WARN(new_conn_state->crtc != - encoder->base.crtc, + I915_STATE_WARN(dev_priv, + new_conn_state->crtc != encoder->base.crtc, "connector's crtc doesn't match encoder crtc\n"); } if (!found) continue; - I915_STATE_WARN(!!encoder->base.crtc != enabled, + I915_STATE_WARN(dev_priv, !!encoder->base.crtc != enabled, "encoder's enabled state mismatch (expected %i, found %i)\n", !!encoder->base.crtc, enabled); @@ -146,7 +148,7 @@ verify_encoder_state(struct drm_i915_private *dev_priv, struct intel_atomic_stat bool active; active = encoder->get_hw_state(encoder, &pipe); - I915_STATE_WARN(active, + I915_STATE_WARN(dev_priv, active, "encoder detached but still enabled on pipe %c.\n", pipe_name(pipe)); } @@ -181,11 +183,12 @@ verify_crtc_state(struct intel_crtc *crtc, if (IS_I830(dev_priv) && pipe_config->hw.active) pipe_config->hw.active = new_crtc_state->hw.active; - I915_STATE_WARN(new_crtc_state->hw.active != pipe_config->hw.active, + I915_STATE_WARN(dev_priv, + new_crtc_state->hw.active != pipe_config->hw.active, "crtc active state doesn't match with hw state (expected %i, found %i)\n", new_crtc_state->hw.active, pipe_config->hw.active); - I915_STATE_WARN(crtc->active != new_crtc_state->hw.active, + I915_STATE_WARN(dev_priv, crtc->active != new_crtc_state->hw.active, "transitional active state does not match atomic hw state (expected %i, found %i)\n", new_crtc_state->hw.active, crtc->active); @@ -196,12 +199,12 @@ verify_crtc_state(struct intel_crtc *crtc, bool active; active = encoder->get_hw_state(encoder, &pipe); - I915_STATE_WARN(active != new_crtc_state->hw.active, + I915_STATE_WARN(dev_priv, active != new_crtc_state->hw.active, "[ENCODER:%i] active %i with crtc active %i\n", encoder->base.base.id, active, new_crtc_state->hw.active); - I915_STATE_WARN(active && master_crtc->pipe != pipe, + I915_STATE_WARN(dev_priv, active && master_crtc->pipe != pipe, "Encoder connected to wrong pipe %c\n", pipe_name(pipe)); @@ -216,7 +219,7 @@ verify_crtc_state(struct intel_crtc *crtc, if (!intel_pipe_config_compare(new_crtc_state, pipe_config, false)) { - I915_STATE_WARN(1, "pipe state doesn't match!\n"); + I915_STATE_WARN(dev_priv, 1, "pipe state doesn't match!\n"); intel_crtc_state_dump(pipe_config, NULL, "hw state"); intel_crtc_state_dump(new_crtc_state, NULL, "sw state"); } @@ -236,6 +239,7 @@ void intel_modeset_verify_crtc(struct intel_crtc *crtc, verify_crtc_state(crtc, old_crtc_state, new_crtc_state); intel_shared_dpll_state_verify(crtc, old_crtc_state, new_crtc_state); intel_mpllb_state_verify(state, new_crtc_state); + intel_c10pll_state_verify(state, new_crtc_state); } void intel_modeset_verify_disabled(struct drm_i915_private *dev_priv, diff --git a/drivers/gpu/drm/i915/display/intel_opregion.c b/drivers/gpu/drm/i915/display/intel_opregion.c index b7973a05d022d741f8d3286a2e120283360d89ca..84078fb82b2f74ad94191fca86f609e3847901dc 100644 --- a/drivers/gpu/drm/i915/display/intel_opregion.c +++ b/drivers/gpu/drm/i915/display/intel_opregion.c @@ -635,7 +635,8 @@ static void asle_work(struct work_struct *work) void intel_opregion_asle_intr(struct drm_i915_private *dev_priv) { if (dev_priv->display.opregion.asle) - schedule_work(&dev_priv->display.opregion.asle_work); + queue_work(dev_priv->unordered_wq, + &dev_priv->display.opregion.asle_work); } #define ACPI_EV_DISPLAY_SWITCH (1<<0) diff --git a/drivers/gpu/drm/i915/display/intel_overlay.c b/drivers/gpu/drm/i915/display/intel_overlay.c index c12bdca8da9ba60b0ccc55ce30e49ad1e7a43102..d6fe2bbabe559a16ed3a22a43fedd26bf831244f 100644 --- a/drivers/gpu/drm/i915/display/intel_overlay.c +++ b/drivers/gpu/drm/i915/display/intel_overlay.c @@ -935,21 +935,25 @@ static int check_overlay_possible_on_crtc(struct intel_overlay *overlay, static void update_pfit_vscale_ratio(struct intel_overlay *overlay) { struct drm_i915_private *dev_priv = overlay->i915; - u32 pfit_control = intel_de_read(dev_priv, PFIT_CONTROL); u32 ratio; /* XXX: This is not the same logic as in the xorg driver, but more in * line with the intel documentation for the i965 */ if (DISPLAY_VER(dev_priv) >= 4) { + u32 tmp = intel_de_read(dev_priv, PFIT_PGM_RATIOS); + /* on i965 use the PGM reg to read out the autoscaler values */ - ratio = intel_de_read(dev_priv, PFIT_PGM_RATIOS) >> PFIT_VERT_SCALE_SHIFT_965; + ratio = REG_FIELD_GET(PFIT_VERT_SCALE_MASK_965, tmp); } else { - if (pfit_control & VERT_AUTO_SCALE) - ratio = intel_de_read(dev_priv, PFIT_AUTO_RATIOS); + u32 tmp; + + if (intel_de_read(dev_priv, PFIT_CONTROL) & PFIT_VERT_AUTO_SCALE) + tmp = intel_de_read(dev_priv, PFIT_AUTO_RATIOS); else - ratio = intel_de_read(dev_priv, PFIT_PGM_RATIOS); - ratio >>= PFIT_VERT_SCALE_SHIFT; + tmp = intel_de_read(dev_priv, PFIT_PGM_RATIOS); + + ratio = REG_FIELD_GET(PFIT_VERT_SCALE_MASK, tmp); } overlay->pfit_vscale_ratio = ratio; diff --git a/drivers/gpu/drm/i915/display/intel_panel.c b/drivers/gpu/drm/i915/display/intel_panel.c index ce2a34a25211789001622d7f94b980ae96f654a4..9232a305b1e6f612f1d871400499a82c8221a872 100644 --- a/drivers/gpu/drm/i915/display/intel_panel.c +++ b/drivers/gpu/drm/i915/display/intel_panel.c @@ -42,6 +42,7 @@ #include "intel_lvds_regs.h" #include "intel_panel.h" #include "intel_quirks.h" +#include "intel_vrr.h" bool intel_panel_use_ssc(struct drm_i915_private *i915) { @@ -58,6 +59,38 @@ intel_panel_preferred_fixed_mode(struct intel_connector *connector) struct drm_display_mode, head); } +static bool is_in_vrr_range(struct intel_connector *connector, int vrefresh) +{ + const struct drm_display_info *info = &connector->base.display_info; + + return intel_vrr_is_capable(connector) && + vrefresh >= info->monitor_range.min_vfreq && + vrefresh <= info->monitor_range.max_vfreq; +} + +static bool is_best_fixed_mode(struct intel_connector *connector, + int vrefresh, int fixed_mode_vrefresh, + const struct drm_display_mode *best_mode) +{ + /* we want to always return something */ + if (!best_mode) + return true; + + /* + * With VRR always pick a mode with equal/higher than requested + * vrefresh, which we can then reduce to match the requested + * vrefresh by extending the vblank length. + */ + if (is_in_vrr_range(connector, vrefresh) && + is_in_vrr_range(connector, fixed_mode_vrefresh) && + fixed_mode_vrefresh < vrefresh) + return false; + + /* pick the fixed_mode that is closest in terms of vrefresh */ + return abs(fixed_mode_vrefresh - vrefresh) < + abs(drm_mode_vrefresh(best_mode) - vrefresh); +} + const struct drm_display_mode * intel_panel_fixed_mode(struct intel_connector *connector, const struct drm_display_mode *mode) @@ -65,11 +98,11 @@ intel_panel_fixed_mode(struct intel_connector *connector, const struct drm_display_mode *fixed_mode, *best_mode = NULL; int vrefresh = drm_mode_vrefresh(mode); - /* pick the fixed_mode that is closest in terms of vrefresh */ list_for_each_entry(fixed_mode, &connector->panel.fixed_modes, head) { - if (!best_mode || - abs(drm_mode_vrefresh(fixed_mode) - vrefresh) < - abs(drm_mode_vrefresh(best_mode) - vrefresh)) + int fixed_mode_vrefresh = drm_mode_vrefresh(fixed_mode); + + if (is_best_fixed_mode(connector, vrefresh, + fixed_mode_vrefresh, best_mode)) best_mode = fixed_mode; } @@ -178,27 +211,46 @@ int intel_panel_compute_config(struct intel_connector *connector, { const struct drm_display_mode *fixed_mode = intel_panel_fixed_mode(connector, adjusted_mode); + int vrefresh, fixed_mode_vrefresh; + bool is_vrr; if (!fixed_mode) return 0; + vrefresh = drm_mode_vrefresh(adjusted_mode); + fixed_mode_vrefresh = drm_mode_vrefresh(fixed_mode); + /* - * We don't want to lie too much to the user about the refresh - * rate they're going to get. But we have to allow a bit of latitude - * for Xorg since it likes to automagically cook up modes with slightly - * off refresh rates. + * Assume that we shouldn't muck about with the + * timings if they don't land in the VRR range. */ - if (abs(drm_mode_vrefresh(adjusted_mode) - drm_mode_vrefresh(fixed_mode)) > 1) { - drm_dbg_kms(connector->base.dev, - "[CONNECTOR:%d:%s] Requested mode vrefresh (%d Hz) does not match fixed mode vrefresh (%d Hz)\n", - connector->base.base.id, connector->base.name, - drm_mode_vrefresh(adjusted_mode), drm_mode_vrefresh(fixed_mode)); + is_vrr = is_in_vrr_range(connector, vrefresh) && + is_in_vrr_range(connector, fixed_mode_vrefresh); - return -EINVAL; + if (!is_vrr) { + /* + * We don't want to lie too much to the user about the refresh + * rate they're going to get. But we have to allow a bit of latitude + * for Xorg since it likes to automagically cook up modes with slightly + * off refresh rates. + */ + if (abs(vrefresh - fixed_mode_vrefresh) > 1) { + drm_dbg_kms(connector->base.dev, + "[CONNECTOR:%d:%s] Requested mode vrefresh (%d Hz) does not match fixed mode vrefresh (%d Hz)\n", + connector->base.base.id, connector->base.name, + vrefresh, fixed_mode_vrefresh); + + return -EINVAL; + } } drm_mode_copy(adjusted_mode, fixed_mode); + if (is_vrr && fixed_mode_vrefresh != vrefresh) + adjusted_mode->vtotal = + DIV_ROUND_CLOSEST(adjusted_mode->clock * 1000, + adjusted_mode->htotal * vrefresh); + drm_mode_set_crtcinfo(adjusted_mode, 0); return 0; @@ -512,11 +564,11 @@ static void i9xx_scale_aspect(struct intel_crtc_state *crtc_state, bits = panel_fitter_scaling(pipe_src_h, adjusted_mode->crtc_vdisplay); - *pfit_pgm_ratios |= (bits << PFIT_HORIZ_SCALE_SHIFT | - bits << PFIT_VERT_SCALE_SHIFT); + *pfit_pgm_ratios |= (PFIT_HORIZ_SCALE(bits) | + PFIT_VERT_SCALE(bits)); *pfit_control |= (PFIT_ENABLE | - VERT_INTERP_BILINEAR | - HORIZ_INTERP_BILINEAR); + PFIT_VERT_INTERP_BILINEAR | + PFIT_HORIZ_INTERP_BILINEAR); } } else if (scaled_width < scaled_height) { /* letter */ centre_vertically(adjusted_mode, @@ -527,18 +579,19 @@ static void i9xx_scale_aspect(struct intel_crtc_state *crtc_state, bits = panel_fitter_scaling(pipe_src_w, adjusted_mode->crtc_hdisplay); - *pfit_pgm_ratios |= (bits << PFIT_HORIZ_SCALE_SHIFT | - bits << PFIT_VERT_SCALE_SHIFT); + *pfit_pgm_ratios |= (PFIT_HORIZ_SCALE(bits) | + PFIT_VERT_SCALE(bits)); *pfit_control |= (PFIT_ENABLE | - VERT_INTERP_BILINEAR | - HORIZ_INTERP_BILINEAR); + PFIT_VERT_INTERP_BILINEAR | + PFIT_HORIZ_INTERP_BILINEAR); } } else { /* Aspects match, Let hw scale both directions */ *pfit_control |= (PFIT_ENABLE | - VERT_AUTO_SCALE | HORIZ_AUTO_SCALE | - VERT_INTERP_BILINEAR | - HORIZ_INTERP_BILINEAR); + PFIT_VERT_AUTO_SCALE | + PFIT_HORIZ_AUTO_SCALE | + PFIT_VERT_INTERP_BILINEAR | + PFIT_HORIZ_INTERP_BILINEAR); } } @@ -586,10 +639,10 @@ static int gmch_panel_fitting(struct intel_crtc_state *crtc_state, if (DISPLAY_VER(dev_priv) >= 4) pfit_control |= PFIT_SCALING_AUTO; else - pfit_control |= (VERT_AUTO_SCALE | - VERT_INTERP_BILINEAR | - HORIZ_AUTO_SCALE | - HORIZ_INTERP_BILINEAR); + pfit_control |= (PFIT_VERT_AUTO_SCALE | + PFIT_VERT_INTERP_BILINEAR | + PFIT_HORIZ_AUTO_SCALE | + PFIT_HORIZ_INTERP_BILINEAR); } break; default: @@ -610,7 +663,7 @@ static int gmch_panel_fitting(struct intel_crtc_state *crtc_state, /* Make sure pre-965 set dither correctly for 18bpp panels. */ if (DISPLAY_VER(dev_priv) < 4 && crtc_state->pipe_bpp == 18) - pfit_control |= PANEL_8TO6_DITHER_ENABLE; + pfit_control |= PFIT_PANEL_8TO6_DITHER_ENABLE; crtc_state->gmch_pfit.control = pfit_control; crtc_state->gmch_pfit.pgm_ratios = pfit_pgm_ratios; diff --git a/drivers/gpu/drm/i915/display/intel_pch_display.c b/drivers/gpu/drm/i915/display/intel_pch_display.c index 2411fe4dee8b2e1abee98411aceca08784926085..866786e6b32f62cefad81b2da73648d1a2b1a97e 100644 --- a/drivers/gpu/drm/i915/display/intel_pch_display.c +++ b/drivers/gpu/drm/i915/display/intel_pch_display.c @@ -43,11 +43,12 @@ static void assert_pch_dp_disabled(struct drm_i915_private *dev_priv, state = g4x_dp_port_enabled(dev_priv, dp_reg, port, &port_pipe); - I915_STATE_WARN(state && port_pipe == pipe, + I915_STATE_WARN(dev_priv, state && port_pipe == pipe, "PCH DP %c enabled on transcoder %c, should be disabled\n", port_name(port), pipe_name(pipe)); - I915_STATE_WARN(HAS_PCH_IBX(dev_priv) && !state && port_pipe == PIPE_B, + I915_STATE_WARN(dev_priv, + HAS_PCH_IBX(dev_priv) && !state && port_pipe == PIPE_B, "IBX PCH DP %c still using transcoder B\n", port_name(port)); } @@ -61,11 +62,12 @@ static void assert_pch_hdmi_disabled(struct drm_i915_private *dev_priv, state = intel_sdvo_port_enabled(dev_priv, hdmi_reg, &port_pipe); - I915_STATE_WARN(state && port_pipe == pipe, + I915_STATE_WARN(dev_priv, state && port_pipe == pipe, "PCH HDMI %c enabled on transcoder %c, should be disabled\n", port_name(port), pipe_name(pipe)); - I915_STATE_WARN(HAS_PCH_IBX(dev_priv) && !state && port_pipe == PIPE_B, + I915_STATE_WARN(dev_priv, + HAS_PCH_IBX(dev_priv) && !state && port_pipe == PIPE_B, "IBX PCH HDMI %c still using transcoder B\n", port_name(port)); } @@ -79,13 +81,13 @@ static void assert_pch_ports_disabled(struct drm_i915_private *dev_priv, assert_pch_dp_disabled(dev_priv, pipe, PORT_C, PCH_DP_C); assert_pch_dp_disabled(dev_priv, pipe, PORT_D, PCH_DP_D); - I915_STATE_WARN(intel_crt_port_enabled(dev_priv, PCH_ADPA, &port_pipe) && - port_pipe == pipe, + I915_STATE_WARN(dev_priv, + intel_crt_port_enabled(dev_priv, PCH_ADPA, &port_pipe) && port_pipe == pipe, "PCH VGA enabled on transcoder %c, should be disabled\n", pipe_name(pipe)); - I915_STATE_WARN(intel_lvds_port_enabled(dev_priv, PCH_LVDS, &port_pipe) && - port_pipe == pipe, + I915_STATE_WARN(dev_priv, + intel_lvds_port_enabled(dev_priv, PCH_LVDS, &port_pipe) && port_pipe == pipe, "PCH LVDS enabled on transcoder %c, should be disabled\n", pipe_name(pipe)); @@ -103,7 +105,7 @@ static void assert_pch_transcoder_disabled(struct drm_i915_private *dev_priv, val = intel_de_read(dev_priv, PCH_TRANSCONF(pipe)); enabled = !!(val & TRANS_ENABLE); - I915_STATE_WARN(enabled, + I915_STATE_WARN(dev_priv, enabled, "transcoder assertion failed, should be off on pipe %c but is still active\n", pipe_name(pipe)); } diff --git a/drivers/gpu/drm/i915/display/intel_pipe_crc.c b/drivers/gpu/drm/i915/display/intel_pipe_crc.c index 8d3ea8d7b737202af307143672d6050409374500..5a468ed6e26c71e862729724e3cba38aff4bf736 100644 --- a/drivers/gpu/drm/i915/display/intel_pipe_crc.c +++ b/drivers/gpu/drm/i915/display/intel_pipe_crc.c @@ -293,6 +293,7 @@ intel_crtc_crc_setup_workarounds(struct intel_crtc *crtc, bool enable) } state->acquire_ctx = &ctx; + to_intel_atomic_state(state)->internal = true; retry: pipe_config = intel_atomic_get_crtc_state(state, crtc); diff --git a/drivers/gpu/drm/i915/display/intel_pmdemand.c b/drivers/gpu/drm/i915/display/intel_pmdemand.c new file mode 100644 index 0000000000000000000000000000000000000000..f7608d36363469bcbe2ee41268f6386ce6ee2924 --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_pmdemand.c @@ -0,0 +1,620 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2023 Intel Corporation + */ + +#include + +#include "i915_drv.h" +#include "i915_reg.h" +#include "intel_atomic.h" +#include "intel_bw.h" +#include "intel_cdclk.h" +#include "intel_de.h" +#include "intel_display_trace.h" +#include "intel_pmdemand.h" +#include "skl_watermark.h" + +static struct intel_global_state * +intel_pmdemand_duplicate_state(struct intel_global_obj *obj) +{ + struct intel_pmdemand_state *pmdemand_state; + + pmdemand_state = kmemdup(obj->state, sizeof(*pmdemand_state), GFP_KERNEL); + if (!pmdemand_state) + return NULL; + + return &pmdemand_state->base; +} + +static void intel_pmdemand_destroy_state(struct intel_global_obj *obj, + struct intel_global_state *state) +{ + kfree(state); +} + +static const struct intel_global_state_funcs intel_pmdemand_funcs = { + .atomic_duplicate_state = intel_pmdemand_duplicate_state, + .atomic_destroy_state = intel_pmdemand_destroy_state, +}; + +static struct intel_pmdemand_state * +intel_atomic_get_pmdemand_state(struct intel_atomic_state *state) +{ + struct drm_i915_private *i915 = to_i915(state->base.dev); + struct intel_global_state *pmdemand_state = + intel_atomic_get_global_obj_state(state, + &i915->display.pmdemand.obj); + + if (IS_ERR(pmdemand_state)) + return ERR_CAST(pmdemand_state); + + return to_intel_pmdemand_state(pmdemand_state); +} + +static struct intel_pmdemand_state * +intel_atomic_get_old_pmdemand_state(struct intel_atomic_state *state) +{ + struct drm_i915_private *i915 = to_i915(state->base.dev); + struct intel_global_state *pmdemand_state = + intel_atomic_get_old_global_obj_state(state, + &i915->display.pmdemand.obj); + + if (!pmdemand_state) + return NULL; + + return to_intel_pmdemand_state(pmdemand_state); +} + +static struct intel_pmdemand_state * +intel_atomic_get_new_pmdemand_state(struct intel_atomic_state *state) +{ + struct drm_i915_private *i915 = to_i915(state->base.dev); + struct intel_global_state *pmdemand_state = + intel_atomic_get_new_global_obj_state(state, + &i915->display.pmdemand.obj); + + if (!pmdemand_state) + return NULL; + + return to_intel_pmdemand_state(pmdemand_state); +} + +int intel_pmdemand_init(struct drm_i915_private *i915) +{ + struct intel_pmdemand_state *pmdemand_state; + + pmdemand_state = kzalloc(sizeof(*pmdemand_state), GFP_KERNEL); + if (!pmdemand_state) + return -ENOMEM; + + intel_atomic_global_obj_init(i915, &i915->display.pmdemand.obj, + &pmdemand_state->base, + &intel_pmdemand_funcs); + + if (IS_MTL_DISPLAY_STEP(i915, STEP_A0, STEP_C0)) + /* Wa_14016740474 */ + intel_de_rmw(i915, XELPD_CHICKEN_DCPR_3, 0, DMD_RSP_TIMEOUT_DISABLE); + + return 0; +} + +void intel_pmdemand_init_early(struct drm_i915_private *i915) +{ + mutex_init(&i915->display.pmdemand.lock); + init_waitqueue_head(&i915->display.pmdemand.waitqueue); +} + +void +intel_pmdemand_update_phys_mask(struct drm_i915_private *i915, + struct intel_encoder *encoder, + struct intel_pmdemand_state *pmdemand_state, + bool set_bit) +{ + enum phy phy; + + if (DISPLAY_VER(i915) < 14) + return; + + if (!encoder) + return; + + phy = intel_port_to_phy(i915, encoder->port); + if (intel_phy_is_tc(i915, phy)) + return; + + if (set_bit) + pmdemand_state->active_combo_phys_mask |= BIT(phy); + else + pmdemand_state->active_combo_phys_mask &= ~BIT(phy); +} + +void +intel_pmdemand_update_port_clock(struct drm_i915_private *i915, + struct intel_pmdemand_state *pmdemand_state, + enum pipe pipe, int port_clock) +{ + if (DISPLAY_VER(i915) < 14) + return; + + pmdemand_state->ddi_clocks[pipe] = port_clock; +} + +static void +intel_pmdemand_update_max_ddiclk(struct drm_i915_private *i915, + struct intel_atomic_state *state, + struct intel_pmdemand_state *pmdemand_state) +{ + int max_ddiclk = 0; + const struct intel_crtc_state *new_crtc_state; + struct intel_crtc *crtc; + int i; + + for_each_new_intel_crtc_in_state(state, crtc, new_crtc_state, i) + intel_pmdemand_update_port_clock(i915, pmdemand_state, + crtc->pipe, + new_crtc_state->port_clock); + + for (i = 0; i < ARRAY_SIZE(pmdemand_state->ddi_clocks); i++) + max_ddiclk = max(pmdemand_state->ddi_clocks[i], max_ddiclk); + + pmdemand_state->params.ddiclk_max = DIV_ROUND_UP(max_ddiclk, 1000); +} + +static void +intel_pmdemand_update_connector_phys(struct drm_i915_private *i915, + struct intel_atomic_state *state, + struct drm_connector_state *conn_state, + bool set_bit, + struct intel_pmdemand_state *pmdemand_state) +{ + struct intel_encoder *encoder = to_intel_encoder(conn_state->best_encoder); + struct intel_crtc *crtc = to_intel_crtc(conn_state->crtc); + struct intel_crtc_state *crtc_state; + + if (!crtc) + return; + + if (set_bit) + crtc_state = intel_atomic_get_new_crtc_state(state, crtc); + else + crtc_state = intel_atomic_get_old_crtc_state(state, crtc); + + if (!crtc_state->hw.active) + return; + + intel_pmdemand_update_phys_mask(i915, encoder, pmdemand_state, + set_bit); +} + +static void +intel_pmdemand_update_active_non_tc_phys(struct drm_i915_private *i915, + struct intel_atomic_state *state, + struct intel_pmdemand_state *pmdemand_state) +{ + struct drm_connector_state *old_conn_state; + struct drm_connector_state *new_conn_state; + struct drm_connector *connector; + int i; + + for_each_oldnew_connector_in_state(&state->base, connector, + old_conn_state, new_conn_state, i) { + if (!intel_connector_needs_modeset(state, connector)) + continue; + + /* First clear the active phys in the old connector state */ + intel_pmdemand_update_connector_phys(i915, state, + old_conn_state, false, + pmdemand_state); + + /* Then set the active phys in new connector state */ + intel_pmdemand_update_connector_phys(i915, state, + new_conn_state, true, + pmdemand_state); + } + + pmdemand_state->params.active_phys = + min_t(u16, hweight16(pmdemand_state->active_combo_phys_mask), + 7); +} + +static bool +intel_pmdemand_encoder_has_tc_phy(struct drm_i915_private *i915, + struct intel_encoder *encoder) +{ + enum phy phy; + + if (!encoder) + return false; + + phy = intel_port_to_phy(i915, encoder->port); + + return intel_phy_is_tc(i915, phy); +} + +static bool +intel_pmdemand_connector_needs_update(struct intel_atomic_state *state) +{ + struct drm_i915_private *i915 = to_i915(state->base.dev); + struct drm_connector_state *old_conn_state; + struct drm_connector_state *new_conn_state; + struct drm_connector *connector; + int i; + + for_each_oldnew_connector_in_state(&state->base, connector, + old_conn_state, new_conn_state, i) { + struct intel_encoder *old_encoder = + to_intel_encoder(old_conn_state->best_encoder); + struct intel_encoder *new_encoder = + to_intel_encoder(new_conn_state->best_encoder); + + if (!intel_connector_needs_modeset(state, connector)) + continue; + + if (old_encoder == new_encoder || + (intel_pmdemand_encoder_has_tc_phy(i915, old_encoder) && + intel_pmdemand_encoder_has_tc_phy(i915, new_encoder))) + continue; + + return true; + } + + return false; +} + +static bool intel_pmdemand_needs_update(struct intel_atomic_state *state) +{ + const struct intel_bw_state *new_bw_state, *old_bw_state; + const struct intel_cdclk_state *new_cdclk_state, *old_cdclk_state; + const struct intel_crtc_state *new_crtc_state, *old_crtc_state; + const struct intel_dbuf_state *new_dbuf_state, *old_dbuf_state; + struct intel_crtc *crtc; + int i; + + new_bw_state = intel_atomic_get_new_bw_state(state); + old_bw_state = intel_atomic_get_old_bw_state(state); + if (new_bw_state && new_bw_state->qgv_point_peakbw != + old_bw_state->qgv_point_peakbw) + return true; + + new_dbuf_state = intel_atomic_get_new_dbuf_state(state); + old_dbuf_state = intel_atomic_get_old_dbuf_state(state); + if (new_dbuf_state && + (new_dbuf_state->active_pipes != + old_dbuf_state->active_pipes || + new_dbuf_state->enabled_slices != + old_dbuf_state->enabled_slices)) + return true; + + new_cdclk_state = intel_atomic_get_new_cdclk_state(state); + old_cdclk_state = intel_atomic_get_old_cdclk_state(state); + if (new_cdclk_state && + (new_cdclk_state->actual.cdclk != + old_cdclk_state->actual.cdclk || + new_cdclk_state->actual.voltage_level != + old_cdclk_state->actual.voltage_level)) + return true; + + for_each_oldnew_intel_crtc_in_state(state, crtc, old_crtc_state, + new_crtc_state, i) + if (new_crtc_state->port_clock != old_crtc_state->port_clock) + return true; + + return intel_pmdemand_connector_needs_update(state); +} + +int intel_pmdemand_atomic_check(struct intel_atomic_state *state) +{ + struct drm_i915_private *i915 = to_i915(state->base.dev); + const struct intel_bw_state *new_bw_state; + const struct intel_cdclk_state *new_cdclk_state; + const struct intel_dbuf_state *new_dbuf_state; + struct intel_pmdemand_state *new_pmdemand_state; + + if (DISPLAY_VER(i915) < 14) + return 0; + + if (!intel_pmdemand_needs_update(state)) + return 0; + + new_pmdemand_state = intel_atomic_get_pmdemand_state(state); + if (IS_ERR(new_pmdemand_state)) + return PTR_ERR(new_pmdemand_state); + + new_bw_state = intel_atomic_get_bw_state(state); + if (IS_ERR(new_bw_state)) + return PTR_ERR(new_bw_state); + + /* firmware will calculate the qclk_gv_index, requirement is set to 0 */ + new_pmdemand_state->params.qclk_gv_index = 0; + new_pmdemand_state->params.qclk_gv_bw = new_bw_state->qgv_point_peakbw; + + new_dbuf_state = intel_atomic_get_dbuf_state(state); + if (IS_ERR(new_dbuf_state)) + return PTR_ERR(new_dbuf_state); + + new_pmdemand_state->params.active_pipes = + min_t(u8, hweight8(new_dbuf_state->active_pipes), 3); + new_pmdemand_state->params.active_dbufs = + min_t(u8, hweight8(new_dbuf_state->enabled_slices), 3); + + new_cdclk_state = intel_atomic_get_cdclk_state(state); + if (IS_ERR(new_cdclk_state)) + return PTR_ERR(new_cdclk_state); + + new_pmdemand_state->params.voltage_index = + new_cdclk_state->actual.voltage_level; + new_pmdemand_state->params.cdclk_freq_mhz = + DIV_ROUND_UP(new_cdclk_state->actual.cdclk, 1000); + + intel_pmdemand_update_max_ddiclk(i915, state, new_pmdemand_state); + + intel_pmdemand_update_active_non_tc_phys(i915, state, new_pmdemand_state); + + /* + * Active_PLLs starts with 1 because of CDCLK PLL. + * TODO: Missing to account genlock filter when it gets used. + */ + new_pmdemand_state->params.plls = + min_t(u16, new_pmdemand_state->params.active_phys + 1, 7); + + /* + * Setting scalers to max as it can not be calculated during flips and + * fastsets without taking global states locks. + */ + new_pmdemand_state->params.scalers = 7; + + if (state->base.allow_modeset) + return intel_atomic_serialize_global_state(&new_pmdemand_state->base); + else + return intel_atomic_lock_global_state(&new_pmdemand_state->base); +} + +static bool intel_pmdemand_check_prev_transaction(struct drm_i915_private *i915) +{ + return !(intel_de_wait_for_clear(i915, + XELPDP_INITIATE_PMDEMAND_REQUEST(1), + XELPDP_PMDEMAND_REQ_ENABLE, 10) || + intel_de_wait_for_clear(i915, + GEN12_DCPR_STATUS_1, + XELPDP_PMDEMAND_INFLIGHT_STATUS, 10)); +} + +void +intel_pmdemand_init_pmdemand_params(struct drm_i915_private *i915, + struct intel_pmdemand_state *pmdemand_state) +{ + u32 reg1, reg2; + + if (DISPLAY_VER(i915) < 14) + return; + + mutex_lock(&i915->display.pmdemand.lock); + if (drm_WARN_ON(&i915->drm, + !intel_pmdemand_check_prev_transaction(i915))) { + memset(&pmdemand_state->params, 0, + sizeof(pmdemand_state->params)); + goto unlock; + } + + reg1 = intel_de_read(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(0)); + + reg2 = intel_de_read(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(1)); + + /* Set 1*/ + pmdemand_state->params.qclk_gv_bw = + REG_FIELD_GET(XELPDP_PMDEMAND_QCLK_GV_BW_MASK, reg1); + pmdemand_state->params.voltage_index = + REG_FIELD_GET(XELPDP_PMDEMAND_VOLTAGE_INDEX_MASK, reg1); + pmdemand_state->params.qclk_gv_index = + REG_FIELD_GET(XELPDP_PMDEMAND_QCLK_GV_INDEX_MASK, reg1); + pmdemand_state->params.active_pipes = + REG_FIELD_GET(XELPDP_PMDEMAND_PIPES_MASK, reg1); + pmdemand_state->params.active_dbufs = + REG_FIELD_GET(XELPDP_PMDEMAND_DBUFS_MASK, reg1); + pmdemand_state->params.active_phys = + REG_FIELD_GET(XELPDP_PMDEMAND_PHYS_MASK, reg1); + + /* Set 2*/ + pmdemand_state->params.cdclk_freq_mhz = + REG_FIELD_GET(XELPDP_PMDEMAND_CDCLK_FREQ_MASK, reg2); + pmdemand_state->params.ddiclk_max = + REG_FIELD_GET(XELPDP_PMDEMAND_DDICLK_FREQ_MASK, reg2); + pmdemand_state->params.scalers = + REG_FIELD_GET(XELPDP_PMDEMAND_SCALERS_MASK, reg2); + +unlock: + mutex_unlock(&i915->display.pmdemand.lock); +} + +static bool intel_pmdemand_req_complete(struct drm_i915_private *i915) +{ + return !(intel_de_read(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(1)) & + XELPDP_PMDEMAND_REQ_ENABLE); +} + +static void intel_pmdemand_wait(struct drm_i915_private *i915) +{ + if (!wait_event_timeout(i915->display.pmdemand.waitqueue, + intel_pmdemand_req_complete(i915), + msecs_to_jiffies_timeout(10))) + drm_err(&i915->drm, + "timed out waiting for Punit PM Demand Response\n"); +} + +/* Required to be programmed during Display Init Sequences. */ +void intel_pmdemand_program_dbuf(struct drm_i915_private *i915, + u8 dbuf_slices) +{ + u32 dbufs = min_t(u32, hweight8(dbuf_slices), 3); + + mutex_lock(&i915->display.pmdemand.lock); + if (drm_WARN_ON(&i915->drm, + !intel_pmdemand_check_prev_transaction(i915))) + goto unlock; + + intel_de_rmw(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(0), + XELPDP_PMDEMAND_DBUFS_MASK, + REG_FIELD_PREP(XELPDP_PMDEMAND_DBUFS_MASK, dbufs)); + intel_de_rmw(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(1), 0, + XELPDP_PMDEMAND_REQ_ENABLE); + + intel_pmdemand_wait(i915); + +unlock: + mutex_unlock(&i915->display.pmdemand.lock); +} + +static void +intel_pmdemand_update_params(const struct intel_pmdemand_state *new, + const struct intel_pmdemand_state *old, + u32 *reg1, u32 *reg2, bool serialized) +{ + /* + * The pmdemand parameter updates happens in two steps. Pre plane and + * post plane updates. During the pre plane, as DE might still be + * handling with some old operations, to avoid unexpected performance + * issues, program the pmdemand parameters with higher of old and new + * values. And then after once settled, use the new parameter values + * as part of the post plane update. + * + * If the pmdemand params update happens without modeset allowed, this + * means we can't serialize the updates. So that implies possibility of + * some parallel atomic commits affecting the pmdemand parameters. In + * that case, we need to consider the current values from the register + * as well. So in pre-plane case, we need to check the max of old, new + * and current register value if not serialized. In post plane update + * we need to consider max of new and current register value if not + * serialized + */ + +#define update_reg(reg, field, mask) do { \ + u32 current_val = serialized ? 0 : REG_FIELD_GET((mask), *(reg)); \ + u32 old_val = old ? old->params.field : 0; \ + u32 new_val = new->params.field; \ +\ + *(reg) &= ~(mask); \ + *(reg) |= REG_FIELD_PREP((mask), max3(old_val, new_val, current_val)); \ +} while (0) + + /* Set 1*/ + update_reg(reg1, qclk_gv_bw, XELPDP_PMDEMAND_QCLK_GV_BW_MASK); + update_reg(reg1, voltage_index, XELPDP_PMDEMAND_VOLTAGE_INDEX_MASK); + update_reg(reg1, qclk_gv_index, XELPDP_PMDEMAND_QCLK_GV_INDEX_MASK); + update_reg(reg1, active_pipes, XELPDP_PMDEMAND_PIPES_MASK); + update_reg(reg1, active_dbufs, XELPDP_PMDEMAND_DBUFS_MASK); + update_reg(reg1, active_phys, XELPDP_PMDEMAND_PHYS_MASK); + + /* Set 2*/ + update_reg(reg2, cdclk_freq_mhz, XELPDP_PMDEMAND_CDCLK_FREQ_MASK); + update_reg(reg2, ddiclk_max, XELPDP_PMDEMAND_DDICLK_FREQ_MASK); + update_reg(reg2, scalers, XELPDP_PMDEMAND_SCALERS_MASK); + update_reg(reg2, plls, XELPDP_PMDEMAND_PLLS_MASK); + +#undef update_reg +} + +static void +intel_pmdemand_program_params(struct drm_i915_private *i915, + const struct intel_pmdemand_state *new, + const struct intel_pmdemand_state *old, + bool serialized) +{ + bool changed = false; + u32 reg1, mod_reg1; + u32 reg2, mod_reg2; + + mutex_lock(&i915->display.pmdemand.lock); + if (drm_WARN_ON(&i915->drm, + !intel_pmdemand_check_prev_transaction(i915))) + goto unlock; + + reg1 = intel_de_read(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(0)); + mod_reg1 = reg1; + + reg2 = intel_de_read(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(1)); + mod_reg2 = reg2; + + intel_pmdemand_update_params(new, old, &mod_reg1, &mod_reg2, + serialized); + + if (reg1 != mod_reg1) { + intel_de_write(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(0), + mod_reg1); + changed = true; + } + + if (reg2 != mod_reg2) { + intel_de_write(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(1), + mod_reg2); + changed = true; + } + + /* Initiate pm demand request only if register values are changed */ + if (!changed) + goto unlock; + + drm_dbg_kms(&i915->drm, + "initate pmdemand request values: (0x%x 0x%x)\n", + mod_reg1, mod_reg2); + + intel_de_rmw(i915, XELPDP_INITIATE_PMDEMAND_REQUEST(1), 0, + XELPDP_PMDEMAND_REQ_ENABLE); + + intel_pmdemand_wait(i915); + +unlock: + mutex_unlock(&i915->display.pmdemand.lock); +} + +static bool +intel_pmdemand_state_changed(const struct intel_pmdemand_state *new, + const struct intel_pmdemand_state *old) +{ + return memcmp(&new->params, &old->params, sizeof(new->params)) != 0; +} + +void intel_pmdemand_pre_plane_update(struct intel_atomic_state *state) +{ + struct drm_i915_private *i915 = to_i915(state->base.dev); + const struct intel_pmdemand_state *new_pmdemand_state = + intel_atomic_get_new_pmdemand_state(state); + const struct intel_pmdemand_state *old_pmdemand_state = + intel_atomic_get_old_pmdemand_state(state); + + if (DISPLAY_VER(i915) < 14) + return; + + if (!new_pmdemand_state || + !intel_pmdemand_state_changed(new_pmdemand_state, + old_pmdemand_state)) + return; + + WARN_ON(!new_pmdemand_state->base.changed); + + intel_pmdemand_program_params(i915, new_pmdemand_state, + old_pmdemand_state, + intel_atomic_global_state_is_serialized(state)); +} + +void intel_pmdemand_post_plane_update(struct intel_atomic_state *state) +{ + struct drm_i915_private *i915 = to_i915(state->base.dev); + const struct intel_pmdemand_state *new_pmdemand_state = + intel_atomic_get_new_pmdemand_state(state); + const struct intel_pmdemand_state *old_pmdemand_state = + intel_atomic_get_old_pmdemand_state(state); + + if (DISPLAY_VER(i915) < 14) + return; + + if (!new_pmdemand_state || + !intel_pmdemand_state_changed(new_pmdemand_state, + old_pmdemand_state)) + return; + + WARN_ON(!new_pmdemand_state->base.changed); + + intel_pmdemand_program_params(i915, new_pmdemand_state, NULL, + intel_atomic_global_state_is_serialized(state)); +} diff --git a/drivers/gpu/drm/i915/display/intel_pmdemand.h b/drivers/gpu/drm/i915/display/intel_pmdemand.h new file mode 100644 index 0000000000000000000000000000000000000000..2941a1a18b72b9137570f43c90022e8ae9b4a264 --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_pmdemand.h @@ -0,0 +1,67 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2023 Intel Corporation + */ + +#ifndef __INTEL_PMDEMAND_H__ +#define __INTEL_PMDEMAND_H__ + +#include "intel_display_limits.h" +#include "intel_global_state.h" + +struct drm_i915_private; +struct intel_atomic_state; +struct intel_crtc_state; +struct intel_encoder; +struct intel_plane_state; + +struct pmdemand_params { + u16 qclk_gv_bw; + u8 voltage_index; + u8 qclk_gv_index; + u8 active_pipes; + u8 active_dbufs; + /* Total number of non type C active phys from active_phys_mask */ + u8 active_phys; + u8 plls; + u16 cdclk_freq_mhz; + /* max from ddi_clocks[] */ + u16 ddiclk_max; + u8 scalers; +}; + +struct intel_pmdemand_state { + struct intel_global_state base; + + /* Maintain a persistent list of port clocks across all crtcs */ + int ddi_clocks[I915_MAX_PIPES]; + + /* Maintain a persistent list of non type C phys mask */ + u16 active_combo_phys_mask; + + /* Parameters to be configured in the pmdemand registers */ + struct pmdemand_params params; +}; + +#define to_intel_pmdemand_state(x) container_of((x), \ + struct intel_pmdemand_state, \ + base) + +void intel_pmdemand_init_early(struct drm_i915_private *i915); +int intel_pmdemand_init(struct drm_i915_private *i915); +void intel_pmdemand_init_pmdemand_params(struct drm_i915_private *i915, + struct intel_pmdemand_state *pmdemand_state); +void intel_pmdemand_update_port_clock(struct drm_i915_private *i915, + struct intel_pmdemand_state *pmdemand_state, + enum pipe pipe, int port_clock); +void intel_pmdemand_update_phys_mask(struct drm_i915_private *i915, + struct intel_encoder *encoder, + struct intel_pmdemand_state *pmdemand_state, + bool clear_bit); +void intel_pmdemand_program_dbuf(struct drm_i915_private *i915, + u8 dbuf_slices); +void intel_pmdemand_pre_plane_update(struct intel_atomic_state *state); +void intel_pmdemand_post_plane_update(struct intel_atomic_state *state); +int intel_pmdemand_atomic_check(struct intel_atomic_state *state); + +#endif /* __INTEL_PMDEMAND_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_pps.c b/drivers/gpu/drm/i915/display/intel_pps.c index 7f9926672a6a58f89d58410aab12b6f5df7cbca6..73f0f1714b3778e61d41bff68eb26f098a709a99 100644 --- a/drivers/gpu/drm/i915/display/intel_pps.c +++ b/drivers/gpu/drm/i915/display/intel_pps.c @@ -787,7 +787,7 @@ void intel_pps_vdd_on(struct intel_dp *intel_dp) vdd = false; with_intel_pps_lock(intel_dp, wakeref) vdd = intel_pps_vdd_on_unlocked(intel_dp); - I915_STATE_WARN(!vdd, "[ENCODER:%d:%s] %s VDD already requested on\n", + I915_STATE_WARN(i915, !vdd, "[ENCODER:%d:%s] %s VDD already requested on\n", dp_to_dig_port(intel_dp)->base.base.base.id, dp_to_dig_port(intel_dp)->base.base.name, pps_name(i915, &intel_dp->pps)); @@ -867,6 +867,7 @@ static void edp_panel_vdd_work(struct work_struct *__work) static void edp_panel_vdd_schedule_off(struct intel_dp *intel_dp) { + struct drm_i915_private *i915 = dp_to_i915(intel_dp); unsigned long delay; /* @@ -882,7 +883,8 @@ static void edp_panel_vdd_schedule_off(struct intel_dp *intel_dp) * operations. */ delay = msecs_to_jiffies(intel_dp->pps.panel_power_cycle_delay * 5); - schedule_delayed_work(&intel_dp->pps.panel_vdd_work, delay); + queue_delayed_work(i915->unordered_wq, + &intel_dp->pps.panel_vdd_work, delay); } /* @@ -899,7 +901,8 @@ void intel_pps_vdd_off_unlocked(struct intel_dp *intel_dp, bool sync) if (!intel_dp_is_edp(intel_dp)) return; - I915_STATE_WARN(!intel_dp->pps.want_panel_vdd, "[ENCODER:%d:%s] %s VDD not forced on", + I915_STATE_WARN(dev_priv, !intel_dp->pps.want_panel_vdd, + "[ENCODER:%d:%s] %s VDD not forced on", dp_to_dig_port(intel_dp)->base.base.base.id, dp_to_dig_port(intel_dp)->base.base.name, pps_name(dev_priv, &intel_dp->pps)); @@ -1653,12 +1656,9 @@ void intel_pps_unlock_regs_wa(struct drm_i915_private *dev_priv) */ pps_num = intel_num_pps(dev_priv); - for (pps_idx = 0; pps_idx < pps_num; pps_idx++) { - u32 val = intel_de_read(dev_priv, PP_CONTROL(pps_idx)); - - val = (val & ~PANEL_UNLOCK_MASK) | PANEL_UNLOCK_REGS; - intel_de_write(dev_priv, PP_CONTROL(pps_idx), val); - } + for (pps_idx = 0; pps_idx < pps_num; pps_idx++) + intel_de_rmw(dev_priv, PP_CONTROL(pps_idx), + PANEL_UNLOCK_MASK, PANEL_UNLOCK_REGS); } void intel_pps_setup(struct drm_i915_private *i915) @@ -1724,7 +1724,7 @@ void assert_pps_unlocked(struct drm_i915_private *dev_priv, enum pipe pipe) ((val & PANEL_UNLOCK_MASK) == PANEL_UNLOCK_REGS)) locked = false; - I915_STATE_WARN(panel_pipe == pipe && locked, + I915_STATE_WARN(dev_priv, panel_pipe == pipe && locked, "panel assertion failure, pipe %c regs locked\n", pipe_name(pipe)); } diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c index 6badfff2b4a28a2c89fd68255079a96d5b9c509f..d58ed9b62e6775df4b27ed9cfc433b01cceae11b 100644 --- a/drivers/gpu/drm/i915/display/intel_psr.c +++ b/drivers/gpu/drm/i915/display/intel_psr.c @@ -85,6 +85,91 @@ * use page flips. */ +/* + * Description of PSR mask bits: + * + * EDP_PSR_DEBUG[16]/EDP_PSR_DEBUG_MASK_DISP_REG_WRITE (hsw-skl): + * + * When unmasked (nearly) all display register writes (eg. even + * SWF) trigger a PSR exit. Some registers are excluded from this + * and they have a more specific mask (described below). On icl+ + * this bit no longer exists and is effectively always set. + * + * PIPE_MISC[21]/PIPE_MISC_PSR_MASK_PIPE_REG_WRITE (skl+): + * + * When unmasked (nearly) all pipe/plane register writes + * trigger a PSR exit. Some plane registers are excluded from this + * and they have a more specific mask (described below). + * + * CHICKEN_PIPESL_1[11]/SKL_PSR_MASK_PLANE_FLIP (skl+): + * PIPE_MISC[23]/PIPE_MISC_PSR_MASK_PRIMARY_FLIP (bdw): + * EDP_PSR_DEBUG[23]/EDP_PSR_DEBUG_MASK_PRIMARY_FLIP (hsw): + * + * When unmasked PRI_SURF/PLANE_SURF writes trigger a PSR exit. + * SPR_SURF/CURBASE are not included in this and instead are + * controlled by PIPE_MISC_PSR_MASK_PIPE_REG_WRITE (skl+) or + * EDP_PSR_DEBUG_MASK_DISP_REG_WRITE (hsw/bdw). + * + * PIPE_MISC[22]/PIPE_MISC_PSR_MASK_SPRITE_ENABLE (bdw): + * EDP_PSR_DEBUG[21]/EDP_PSR_DEBUG_MASK_SPRITE_ENABLE (hsw): + * + * When unmasked PSR is blocked as long as the sprite + * plane is enabled. skl+ with their universal planes no + * longer have a mask bit like this, and no plane being + * enabledb blocks PSR. + * + * PIPE_MISC[21]/PIPE_MISC_PSR_MASK_CURSOR_MOVE (bdw): + * EDP_PSR_DEBUG[20]/EDP_PSR_DEBUG_MASK_CURSOR_MOVE (hsw): + * + * When umasked CURPOS writes trigger a PSR exit. On skl+ + * this doesn't exit but CURPOS is included in the + * PIPE_MISC_PSR_MASK_PIPE_REG_WRITE mask. + * + * PIPE_MISC[20]/PIPE_MISC_PSR_MASK_VBLANK_VSYNC_INT (bdw+): + * EDP_PSR_DEBUG[19]/EDP_PSR_DEBUG_MASK_VBLANK_VSYNC_INT (hsw): + * + * When unmasked PSR is blocked as long as vblank and/or vsync + * interrupt is unmasked in IMR *and* enabled in IER. + * + * CHICKEN_TRANS[30]/SKL_UNMASK_VBL_TO_PIPE_IN_SRD (skl+): + * CHICKEN_PAR1_1[15]/HSW_MASK_VBL_TO_PIPE_IN_SRD (hsw/bdw): + * + * Selectcs whether PSR exit generates an extra vblank before + * the first frame is transmitted. Also note the opposite polarity + * if the bit on hsw/bdw vs. skl+ (masked==generate the extra vblank, + * unmasked==do not generate the extra vblank). + * + * With DC states enabled the extra vblank happens after link training, + * with DC states disabled it happens immediately upuon PSR exit trigger. + * No idea as of now why there is a difference. HSW/BDW (which don't + * even have DMC) always generate it after link training. Go figure. + * + * Unfortunately CHICKEN_TRANS itself seems to be double buffered + * and thus won't latch until the first vblank. So with DC states + * enabled the register effctively uses the reset value during DC5 + * exit+PSR exit sequence, and thus the bit does nothing until + * latched by the vblank that it was trying to prevent from being + * generated in the first place. So we should probably call this + * one a chicken/egg bit instead on skl+. + * + * In standby mode (as opposed to link-off) this makes no difference + * as the timing generator keeps running the whole time generating + * normal periodic vblanks. + * + * WaPsrDPAMaskVBlankInSRD asks us to set the bit on hsw/bdw, + * and doing so makes the behaviour match the skl+ reset value. + * + * CHICKEN_PIPESL_1[0]/BDW_UNMASK_VBL_TO_REGS_IN_SRD (bdw): + * CHICKEN_PIPESL_1[15]/HSW_UNMASK_VBL_TO_REGS_IN_SRD (hsw): + * + * On BDW without this bit is no vblanks whatsoever are + * generated after PSR exit. On HSW this has no apparant effect. + * WaPsrDPRSUnmaskVBlankInSRD says to set this. + * + * The rest of the bits are more self-explanatory and/or + * irrelevant for normal operation. + */ + static bool psr_global_enabled(struct intel_dp *intel_dp) { struct intel_connector *connector = intel_dp->attached_connector; @@ -208,13 +293,13 @@ static void psr_event_print(struct drm_i915_private *i915, void intel_psr_irq_handler(struct intel_dp *intel_dp, u32 psr_iir) { - enum transcoder cpu_transcoder = intel_dp->psr.transcoder; struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; ktime_t time_ns = ktime_get(); i915_reg_t imr_reg; if (DISPLAY_VER(dev_priv) >= 12) - imr_reg = TRANS_PSR_IMR(intel_dp->psr.transcoder); + imr_reg = TRANS_PSR_IMR(cpu_transcoder); else imr_reg = EDP_PSR_IMR; @@ -232,13 +317,11 @@ void intel_psr_irq_handler(struct intel_dp *intel_dp, u32 psr_iir) transcoder_name(cpu_transcoder)); if (DISPLAY_VER(dev_priv) >= 9) { - u32 val = intel_de_read(dev_priv, - PSR_EVENT(cpu_transcoder)); - bool psr2_enabled = intel_dp->psr.psr2_enabled; + u32 val; + + val = intel_de_rmw(dev_priv, PSR_EVENT(cpu_transcoder), 0, 0); - intel_de_write(dev_priv, PSR_EVENT(cpu_transcoder), - val); - psr_event_print(dev_priv, val, psr2_enabled); + psr_event_print(dev_priv, val, intel_dp->psr.psr2_enabled); } } @@ -258,7 +341,7 @@ void intel_psr_irq_handler(struct intel_dp *intel_dp, u32 psr_iir) */ intel_de_rmw(dev_priv, imr_reg, 0, psr_irq_psr_error_bit_get(intel_dp)); - schedule_work(&intel_dp->psr.work); + queue_work(dev_priv->unordered_wq, &intel_dp->psr.work); } } @@ -419,7 +502,7 @@ static u32 intel_psr1_get_tp_time(struct intel_dp *intel_dp) u32 val = 0; if (DISPLAY_VER(dev_priv) >= 11) - val |= EDP_PSR_TP4_TIME_0US; + val |= EDP_PSR_TP4_TIME_0us; if (dev_priv->params.psr_safest_params) { val |= EDP_PSR_TP1_TIME_2500us; @@ -448,9 +531,9 @@ static u32 intel_psr1_get_tp_time(struct intel_dp *intel_dp) check_tp3_sel: if (intel_dp_source_supports_tps3(dev_priv) && drm_dp_tps3_supported(intel_dp->dpcd)) - val |= EDP_PSR_TP1_TP3_SEL; + val |= EDP_PSR_TP_TP1_TP3; else - val |= EDP_PSR_TP1_TP2_SEL; + val |= EDP_PSR_TP_TP1_TP2; return val; } @@ -476,12 +559,13 @@ static u8 psr_compute_idle_frames(struct intel_dp *intel_dp) static void hsw_activate_psr1(struct intel_dp *intel_dp) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; u32 max_sleep_time = 0x1f; u32 val = EDP_PSR_ENABLE; - val |= psr_compute_idle_frames(intel_dp) << EDP_PSR_IDLE_FRAME_SHIFT; + val |= EDP_PSR_IDLE_FRAMES(psr_compute_idle_frames(intel_dp)); - val |= max_sleep_time << EDP_PSR_MAX_SLEEP_TIME_SHIFT; + val |= EDP_PSR_MAX_SLEEP_TIME(max_sleep_time); if (IS_HASWELL(dev_priv)) val |= EDP_PSR_MIN_LINK_ENTRY_TIME_8_LINES; @@ -493,9 +577,8 @@ static void hsw_activate_psr1(struct intel_dp *intel_dp) if (DISPLAY_VER(dev_priv) >= 8) val |= EDP_PSR_CRC_ENABLE; - val |= (intel_de_read(dev_priv, EDP_PSR_CTL(intel_dp->psr.transcoder)) & - EDP_PSR_RESTORE_PSR_ACTIVE_CTX_MASK); - intel_de_write(dev_priv, EDP_PSR_CTL(intel_dp->psr.transcoder), val); + intel_de_rmw(dev_priv, EDP_PSR_CTL(cpu_transcoder), + ~EDP_PSR_RESTORE_PSR_ACTIVE_CTX_MASK, val); } static u32 intel_psr2_get_tp_time(struct intel_dp *intel_dp) @@ -534,9 +617,10 @@ static int psr2_block_count(struct intel_dp *intel_dp) static void hsw_activate_psr2(struct intel_dp *intel_dp) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; u32 val = EDP_PSR2_ENABLE; - val |= psr_compute_idle_frames(intel_dp) << EDP_PSR2_IDLE_FRAME_SHIFT; + val |= EDP_PSR2_IDLE_FRAMES(psr_compute_idle_frames(intel_dp)); if (DISPLAY_VER(dev_priv) <= 13 && !IS_ALDERLAKE_P(dev_priv)) val |= EDP_SU_TRACK_ENABLE; @@ -570,15 +654,13 @@ static void hsw_activate_psr2(struct intel_dp *intel_dp) * Still using the default IO_BUFFER_WAKE and FAST_WAKE, see * comments bellow for more information */ - u32 tmp; + int tmp; tmp = map[intel_dp->psr.io_wake_lines - TGL_EDP_PSR2_IO_BUFFER_WAKE_MIN_LINES]; - tmp = tmp << TGL_EDP_PSR2_IO_BUFFER_WAKE_SHIFT; - val |= tmp; + val |= TGL_EDP_PSR2_IO_BUFFER_WAKE(tmp + TGL_EDP_PSR2_IO_BUFFER_WAKE_MIN_LINES); tmp = map[intel_dp->psr.fast_wake_lines - TGL_EDP_PSR2_FAST_WAKE_MIN_LINES]; - tmp = tmp << TGL_EDP_PSR2_FAST_WAKE_MIN_SHIFT; - val |= tmp; + val |= TGL_EDP_PSR2_FAST_WAKE(tmp + TGL_EDP_PSR2_FAST_WAKE_MIN_LINES); } else if (DISPLAY_VER(dev_priv) >= 12) { val |= TGL_EDP_PSR2_IO_BUFFER_WAKE(intel_dp->psr.io_wake_lines); val |= TGL_EDP_PSR2_FAST_WAKE(intel_dp->psr.fast_wake_lines); @@ -593,31 +675,30 @@ static void hsw_activate_psr2(struct intel_dp *intel_dp) if (intel_dp->psr.psr2_sel_fetch_enabled) { u32 tmp; - tmp = intel_de_read(dev_priv, PSR2_MAN_TRK_CTL(intel_dp->psr.transcoder)); + tmp = intel_de_read(dev_priv, PSR2_MAN_TRK_CTL(cpu_transcoder)); drm_WARN_ON(&dev_priv->drm, !(tmp & PSR2_MAN_TRK_CTL_ENABLE)); } else if (HAS_PSR2_SEL_FETCH(dev_priv)) { - intel_de_write(dev_priv, - PSR2_MAN_TRK_CTL(intel_dp->psr.transcoder), 0); + intel_de_write(dev_priv, PSR2_MAN_TRK_CTL(cpu_transcoder), 0); } /* * PSR2 HW is incorrectly using EDP_PSR_TP1_TP3_SEL and BSpec is * recommending keep this bit unset while PSR2 is enabled. */ - intel_de_write(dev_priv, EDP_PSR_CTL(intel_dp->psr.transcoder), 0); + intel_de_write(dev_priv, EDP_PSR_CTL(cpu_transcoder), 0); - intel_de_write(dev_priv, EDP_PSR2_CTL(intel_dp->psr.transcoder), val); + intel_de_write(dev_priv, EDP_PSR2_CTL(cpu_transcoder), val); } static bool -transcoder_has_psr2(struct drm_i915_private *dev_priv, enum transcoder trans) +transcoder_has_psr2(struct drm_i915_private *dev_priv, enum transcoder cpu_transcoder) { if (IS_ALDERLAKE_P(dev_priv) || DISPLAY_VER(dev_priv) >= 14) - return trans == TRANSCODER_A || trans == TRANSCODER_B; + return cpu_transcoder == TRANSCODER_A || cpu_transcoder == TRANSCODER_B; else if (DISPLAY_VER(dev_priv) >= 12) - return trans == TRANSCODER_A; + return cpu_transcoder == TRANSCODER_A; else - return trans == TRANSCODER_EDP; + return cpu_transcoder == TRANSCODER_EDP; } static u32 intel_get_frame_time_us(const struct intel_crtc_state *cstate) @@ -633,10 +714,11 @@ static void psr2_program_idle_frames(struct intel_dp *intel_dp, u32 idle_frames) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; - idle_frames <<= EDP_PSR2_IDLE_FRAME_SHIFT; - intel_de_rmw(dev_priv, EDP_PSR2_CTL(intel_dp->psr.transcoder), - EDP_PSR2_IDLE_FRAME_MASK, idle_frames); + intel_de_rmw(dev_priv, EDP_PSR2_CTL(cpu_transcoder), + EDP_PSR2_IDLE_FRAMES_MASK, + EDP_PSR2_IDLE_FRAMES(idle_frames)); } static void tgl_psr2_enable_dc3co(struct intel_dp *intel_dp) @@ -1074,6 +1156,7 @@ void intel_psr_get_config(struct intel_encoder *encoder, { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); struct intel_digital_port *dig_port = enc_to_dig_port(encoder); + enum transcoder cpu_transcoder = pipe_config->cpu_transcoder; struct intel_dp *intel_dp; u32 val; @@ -1100,15 +1183,14 @@ void intel_psr_get_config(struct intel_encoder *encoder, goto unlock; if (HAS_PSR2_SEL_FETCH(dev_priv)) { - val = intel_de_read(dev_priv, PSR2_MAN_TRK_CTL(intel_dp->psr.transcoder)); + val = intel_de_read(dev_priv, PSR2_MAN_TRK_CTL(cpu_transcoder)); if (val & PSR2_MAN_TRK_CTL_ENABLE) pipe_config->enable_psr2_sel_fetch = true; } if (DISPLAY_VER(dev_priv) >= 12) { - val = intel_de_read(dev_priv, TRANS_EXITLINE(intel_dp->psr.transcoder)); - val &= EXITLINE_MASK; - pipe_config->dc3co_exitline = val; + val = intel_de_read(dev_priv, TRANS_EXITLINE(cpu_transcoder)); + pipe_config->dc3co_exitline = REG_FIELD_GET(EXITLINE_MASK, val); } unlock: mutex_unlock(&intel_dp->psr.lock); @@ -1117,14 +1199,14 @@ void intel_psr_get_config(struct intel_encoder *encoder, static void intel_psr_activate(struct intel_dp *intel_dp) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); - enum transcoder transcoder = intel_dp->psr.transcoder; + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; - if (transcoder_has_psr2(dev_priv, transcoder)) + if (transcoder_has_psr2(dev_priv, cpu_transcoder)) drm_WARN_ON(&dev_priv->drm, - intel_de_read(dev_priv, EDP_PSR2_CTL(transcoder)) & EDP_PSR2_ENABLE); + intel_de_read(dev_priv, EDP_PSR2_CTL(cpu_transcoder)) & EDP_PSR2_ENABLE); drm_WARN_ON(&dev_priv->drm, - intel_de_read(dev_priv, EDP_PSR_CTL(transcoder)) & EDP_PSR_ENABLE); + intel_de_read(dev_priv, EDP_PSR_CTL(cpu_transcoder)) & EDP_PSR_ENABLE); drm_WARN_ON(&dev_priv->drm, intel_dp->psr.active); lockdep_assert_held(&intel_dp->psr.lock); @@ -1203,7 +1285,7 @@ static void intel_psr_enable_source(struct intel_dp *intel_dp, if (DISPLAY_VER(dev_priv) < 11) mask |= EDP_PSR_DEBUG_MASK_DISP_REG_WRITE; - intel_de_write(dev_priv, EDP_PSR_DEBUG(intel_dp->psr.transcoder), + intel_de_write(dev_priv, EDP_PSR_DEBUG(cpu_transcoder), mask); psr_irq_control(intel_dp); @@ -1259,6 +1341,7 @@ static void intel_psr_enable_source(struct intel_dp *intel_dp, static bool psr_interrupt_error_check(struct intel_dp *intel_dp) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; u32 val; /* @@ -1270,8 +1353,7 @@ static bool psr_interrupt_error_check(struct intel_dp *intel_dp) * to avoid any rendering problems. */ if (DISPLAY_VER(dev_priv) >= 12) - val = intel_de_read(dev_priv, - TRANS_PSR_IIR(intel_dp->psr.transcoder)); + val = intel_de_read(dev_priv, TRANS_PSR_IIR(cpu_transcoder)); else val = intel_de_read(dev_priv, EDP_PSR_IIR); val &= psr_irq_psr_error_bit_get(intel_dp); @@ -1327,17 +1409,16 @@ static void intel_psr_enable_locked(struct intel_dp *intel_dp, static void intel_psr_exit(struct intel_dp *intel_dp) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; u32 val; if (!intel_dp->psr.active) { - if (transcoder_has_psr2(dev_priv, intel_dp->psr.transcoder)) { - val = intel_de_read(dev_priv, - EDP_PSR2_CTL(intel_dp->psr.transcoder)); + if (transcoder_has_psr2(dev_priv, cpu_transcoder)) { + val = intel_de_read(dev_priv, EDP_PSR2_CTL(cpu_transcoder)); drm_WARN_ON(&dev_priv->drm, val & EDP_PSR2_ENABLE); } - val = intel_de_read(dev_priv, - EDP_PSR_CTL(intel_dp->psr.transcoder)); + val = intel_de_read(dev_priv, EDP_PSR_CTL(cpu_transcoder)); drm_WARN_ON(&dev_priv->drm, val & EDP_PSR_ENABLE); return; @@ -1345,19 +1426,16 @@ static void intel_psr_exit(struct intel_dp *intel_dp) if (intel_dp->psr.psr2_enabled) { tgl_disallow_dc3co_on_psr2_exit(intel_dp); - val = intel_de_read(dev_priv, - EDP_PSR2_CTL(intel_dp->psr.transcoder)); + + val = intel_de_rmw(dev_priv, EDP_PSR2_CTL(cpu_transcoder), + EDP_PSR2_ENABLE, 0); + drm_WARN_ON(&dev_priv->drm, !(val & EDP_PSR2_ENABLE)); - val &= ~EDP_PSR2_ENABLE; - intel_de_write(dev_priv, - EDP_PSR2_CTL(intel_dp->psr.transcoder), val); } else { - val = intel_de_read(dev_priv, - EDP_PSR_CTL(intel_dp->psr.transcoder)); + val = intel_de_rmw(dev_priv, EDP_PSR_CTL(cpu_transcoder), + EDP_PSR_ENABLE, 0); + drm_WARN_ON(&dev_priv->drm, !(val & EDP_PSR_ENABLE)); - val &= ~EDP_PSR_ENABLE; - intel_de_write(dev_priv, - EDP_PSR_CTL(intel_dp->psr.transcoder), val); } intel_dp->psr.active = false; } @@ -1365,14 +1443,15 @@ static void intel_psr_exit(struct intel_dp *intel_dp) static void intel_psr_wait_exit_locked(struct intel_dp *intel_dp) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; i915_reg_t psr_status; u32 psr_status_mask; if (intel_dp->psr.psr2_enabled) { - psr_status = EDP_PSR2_STATUS(intel_dp->psr.transcoder); + psr_status = EDP_PSR2_STATUS(cpu_transcoder); psr_status_mask = EDP_PSR2_STATUS_STATE_MASK; } else { - psr_status = EDP_PSR_STATUS(intel_dp->psr.transcoder); + psr_status = EDP_PSR_STATUS(cpu_transcoder); psr_status_mask = EDP_PSR_STATUS_STATE_MASK; } @@ -1385,6 +1464,7 @@ static void intel_psr_wait_exit_locked(struct intel_dp *intel_dp) static void intel_psr_disable_locked(struct intel_dp *intel_dp) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; enum phy phy = intel_port_to_phy(dev_priv, dp_to_dig_port(intel_dp)->base.port); @@ -1411,7 +1491,7 @@ static void intel_psr_disable_locked(struct intel_dp *intel_dp) /* Wa_16012604467:adlp,mtl[a0,b0] */ if (IS_MTL_DISPLAY_STEP(dev_priv, STEP_A0, STEP_B0)) intel_de_rmw(dev_priv, - MTL_CLKGATE_DIS_TRANS(intel_dp->psr.transcoder), + MTL_CLKGATE_DIS_TRANS(cpu_transcoder), MTL_CLKGATE_DIS_TRANS_DMASC_GATING_DIS, 0); else if (IS_ALDERLAKE_P(dev_priv)) intel_de_rmw(dev_priv, CLKGATE_DIS_MISC, @@ -1548,10 +1628,11 @@ static u32 man_trk_ctl_continuos_full_frame(struct drm_i915_private *dev_priv) static void psr_force_hw_tracking_exit(struct intel_dp *intel_dp) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; if (intel_dp->psr.psr2_sel_fetch_enabled) intel_de_write(dev_priv, - PSR2_MAN_TRK_CTL(intel_dp->psr.transcoder), + PSR2_MAN_TRK_CTL(cpu_transcoder), man_trk_ctl_enable_bit_get(dev_priv) | man_trk_ctl_partial_frame_bit_get(dev_priv) | man_trk_ctl_single_full_frame_bit_get(dev_priv) | @@ -1651,6 +1732,7 @@ void intel_psr2_program_plane_sel_fetch_noarm(struct intel_plane *plane, void intel_psr2_program_trans_man_trk_ctl(const struct intel_crtc_state *crtc_state) { struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev); + enum transcoder cpu_transcoder = crtc_state->cpu_transcoder; struct intel_encoder *encoder; if (!crtc_state->enable_psr2_sel_fetch) @@ -1666,7 +1748,7 @@ void intel_psr2_program_trans_man_trk_ctl(const struct intel_crtc_state *crtc_st break; } - intel_de_write(dev_priv, PSR2_MAN_TRK_CTL(crtc_state->cpu_transcoder), + intel_de_write(dev_priv, PSR2_MAN_TRK_CTL(cpu_transcoder), crtc_state->psr2_man_track_ctl); } @@ -2045,6 +2127,7 @@ void intel_psr_post_plane_update(const struct intel_atomic_state *state) static int _psr2_ready_for_pipe_update_locked(struct intel_dp *intel_dp) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; /* * Any state lower than EDP_PSR2_STATUS_STATE_DEEP_SLEEP is enough. @@ -2052,13 +2135,14 @@ static int _psr2_ready_for_pipe_update_locked(struct intel_dp *intel_dp) * EDP_PSR2_STATUS_STATE_DEEP_SLEEP to be cleared. */ return intel_de_wait_for_clear(dev_priv, - EDP_PSR2_STATUS(intel_dp->psr.transcoder), + EDP_PSR2_STATUS(cpu_transcoder), EDP_PSR2_STATUS_STATE_DEEP_SLEEP, 50); } static int _psr1_ready_for_pipe_update_locked(struct intel_dp *intel_dp) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; /* * From bspec: Panel Self Refresh (BDW+) @@ -2067,7 +2151,7 @@ static int _psr1_ready_for_pipe_update_locked(struct intel_dp *intel_dp) * defensive enough to cover everything. */ return intel_de_wait_for_clear(dev_priv, - EDP_PSR_STATUS(intel_dp->psr.transcoder), + EDP_PSR_STATUS(cpu_transcoder), EDP_PSR_STATUS_STATE_MASK, 50); } @@ -2109,6 +2193,7 @@ void intel_psr_wait_for_idle_locked(const struct intel_crtc_state *new_crtc_stat static bool __psr_wait_for_idle_locked(struct intel_dp *intel_dp) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; i915_reg_t reg; u32 mask; int err; @@ -2117,10 +2202,10 @@ static bool __psr_wait_for_idle_locked(struct intel_dp *intel_dp) return false; if (intel_dp->psr.psr2_enabled) { - reg = EDP_PSR2_STATUS(intel_dp->psr.transcoder); + reg = EDP_PSR2_STATUS(cpu_transcoder); mask = EDP_PSR2_STATUS_STATE_MASK; } else { - reg = EDP_PSR_STATUS(intel_dp->psr.transcoder); + reg = EDP_PSR_STATUS(cpu_transcoder); mask = EDP_PSR_STATUS_STATE_MASK; } @@ -2149,10 +2234,11 @@ static int intel_psr_fastset_force(struct drm_i915_private *dev_priv) return -ENOMEM; drm_modeset_acquire_init(&ctx, DRM_MODESET_ACQUIRE_INTERRUPTIBLE); + state->acquire_ctx = &ctx; + to_intel_atomic_state(state)->internal = true; retry: - drm_connector_list_iter_begin(&dev_priv->drm, &conn_iter); drm_for_each_connector_iter(conn, &conn_iter) { struct drm_connector_state *conn_state; @@ -2281,6 +2367,7 @@ static void intel_psr_work(struct work_struct *work) static void _psr_invalidate_handle(struct intel_dp *intel_dp) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; if (intel_dp->psr.psr2_sel_fetch_enabled) { u32 val; @@ -2294,7 +2381,7 @@ static void _psr_invalidate_handle(struct intel_dp *intel_dp) val = man_trk_ctl_enable_bit_get(dev_priv) | man_trk_ctl_partial_frame_bit_get(dev_priv) | man_trk_ctl_continuos_full_frame(dev_priv); - intel_de_write(dev_priv, PSR2_MAN_TRK_CTL(intel_dp->psr.transcoder), val); + intel_de_write(dev_priv, PSR2_MAN_TRK_CTL(cpu_transcoder), val); intel_de_write(dev_priv, CURSURFLIVE(intel_dp->psr.pipe), 0); intel_dp->psr.psr2_sel_fetch_cff_enabled = true; } else { @@ -2353,6 +2440,8 @@ static void tgl_dc3co_flush_locked(struct intel_dp *intel_dp, unsigned int frontbuffer_bits, enum fb_op_origin origin) { + struct drm_i915_private *i915 = dp_to_i915(intel_dp); + if (!intel_dp->psr.dc3co_exitline || !intel_dp->psr.psr2_enabled || !intel_dp->psr.active) return; @@ -2366,13 +2455,14 @@ tgl_dc3co_flush_locked(struct intel_dp *intel_dp, unsigned int frontbuffer_bits, return; tgl_psr2_enable_dc3co(intel_dp); - mod_delayed_work(system_wq, &intel_dp->psr.dc3co_work, + mod_delayed_work(i915->unordered_wq, &intel_dp->psr.dc3co_work, intel_dp->psr.dc3co_exit_delay); } static void _psr_flush_handle(struct intel_dp *intel_dp) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; if (intel_dp->psr.psr2_sel_fetch_enabled) { if (intel_dp->psr.psr2_sel_fetch_cff_enabled) { @@ -2389,7 +2479,7 @@ static void _psr_flush_handle(struct intel_dp *intel_dp) * SU configuration in case update is sent for any reason after * sff bit gets cleared by the HW on next vblank. */ - intel_de_write(dev_priv, PSR2_MAN_TRK_CTL(intel_dp->psr.transcoder), + intel_de_write(dev_priv, PSR2_MAN_TRK_CTL(cpu_transcoder), val); intel_de_write(dev_priv, CURSURFLIVE(intel_dp->psr.pipe), 0); intel_dp->psr.psr2_sel_fetch_cff_enabled = false; @@ -2405,7 +2495,7 @@ static void _psr_flush_handle(struct intel_dp *intel_dp) psr_force_hw_tracking_exit(intel_dp); if (!intel_dp->psr.active && !intel_dp->psr.busy_frontbuffer_bits) - schedule_work(&intel_dp->psr.work); + queue_work(dev_priv->unordered_wq, &intel_dp->psr.work); } } @@ -2702,6 +2792,7 @@ static void psr_source_status(struct intel_dp *intel_dp, struct seq_file *m) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; const char *status = "unknown"; u32 val, status_val; @@ -2719,8 +2810,7 @@ psr_source_status(struct intel_dp *intel_dp, struct seq_file *m) "BUF_ON", "TG_ON" }; - val = intel_de_read(dev_priv, - EDP_PSR2_STATUS(intel_dp->psr.transcoder)); + val = intel_de_read(dev_priv, EDP_PSR2_STATUS(cpu_transcoder)); status_val = REG_FIELD_GET(EDP_PSR2_STATUS_STATE_MASK, val); if (status_val < ARRAY_SIZE(live_status)) status = live_status[status_val]; @@ -2735,10 +2825,8 @@ psr_source_status(struct intel_dp *intel_dp, struct seq_file *m) "SRDOFFACK", "SRDENT_ON", }; - val = intel_de_read(dev_priv, - EDP_PSR_STATUS(intel_dp->psr.transcoder)); - status_val = (val & EDP_PSR_STATUS_STATE_MASK) >> - EDP_PSR_STATUS_STATE_SHIFT; + val = intel_de_read(dev_priv, EDP_PSR_STATUS(cpu_transcoder)); + status_val = REG_FIELD_GET(EDP_PSR_STATUS_STATE_MASK, val); if (status_val < ARRAY_SIZE(live_status)) status = live_status[status_val]; } @@ -2749,6 +2837,7 @@ psr_source_status(struct intel_dp *intel_dp, struct seq_file *m) static int intel_psr_status(struct seq_file *m, struct intel_dp *intel_dp) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; struct intel_psr *psr = &intel_dp->psr; intel_wakeref_t wakeref; const char *status; @@ -2780,12 +2869,10 @@ static int intel_psr_status(struct seq_file *m, struct intel_dp *intel_dp) } if (psr->psr2_enabled) { - val = intel_de_read(dev_priv, - EDP_PSR2_CTL(intel_dp->psr.transcoder)); + val = intel_de_read(dev_priv, EDP_PSR2_CTL(cpu_transcoder)); enabled = val & EDP_PSR2_ENABLE; } else { - val = intel_de_read(dev_priv, - EDP_PSR_CTL(intel_dp->psr.transcoder)); + val = intel_de_read(dev_priv, EDP_PSR_CTL(cpu_transcoder)); enabled = val & EDP_PSR_ENABLE; } seq_printf(m, "Source PSR ctl: %s [0x%08x]\n", @@ -2797,12 +2884,9 @@ static int intel_psr_status(struct seq_file *m, struct intel_dp *intel_dp) /* * SKL+ Perf counter is reset to 0 everytime DC state is entered */ - if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) { - val = intel_de_read(dev_priv, - EDP_PSR_PERF_CNT(intel_dp->psr.transcoder)); - val &= EDP_PSR_PERF_CNT_MASK; - seq_printf(m, "Performance counter: %u\n", val); - } + val = intel_de_read(dev_priv, EDP_PSR_PERF_CNT(cpu_transcoder)); + seq_printf(m, "Performance counter: %u\n", + REG_FIELD_GET(EDP_PSR_PERF_CNT_MASK, val)); if (psr->debug & I915_PSR_DEBUG_IRQ) { seq_printf(m, "Last attempted entry at: %lld\n", @@ -2819,8 +2903,7 @@ static int intel_psr_status(struct seq_file *m, struct intel_dp *intel_dp) * frame boundary between register reads */ for (frame = 0; frame < PSR2_SU_STATUS_FRAMES; frame += 3) { - val = intel_de_read(dev_priv, - PSR2_SU_STATUS(intel_dp->psr.transcoder, frame)); + val = intel_de_read(dev_priv, PSR2_SU_STATUS(cpu_transcoder, frame)); su_frames_val[frame / 3] = val; } diff --git a/drivers/gpu/drm/i915/display/intel_psr_regs.h b/drivers/gpu/drm/i915/display/intel_psr_regs.h index 958d8cabc44b519f0257d21ecfb8526aa0e82070..0f7db617425a357dc361ff3bbeac5cc5b63ff115 100644 --- a/drivers/gpu/drm/i915/display/intel_psr_regs.h +++ b/drivers/gpu/drm/i915/display/intel_psr_regs.h @@ -22,30 +22,36 @@ #define _SRD_CTL_A 0x60800 #define _SRD_CTL_EDP 0x6f800 #define EDP_PSR_CTL(tran) _MMIO_TRANS2(tran, _SRD_CTL_A) -#define EDP_PSR_ENABLE (1 << 31) -#define BDW_PSR_SINGLE_FRAME (1 << 30) -#define EDP_PSR_RESTORE_PSR_ACTIVE_CTX_MASK (1 << 29) /* SW can't modify */ -#define EDP_PSR_LINK_STANDBY (1 << 27) -#define EDP_PSR_MIN_LINK_ENTRY_TIME_MASK (3 << 25) -#define EDP_PSR_MIN_LINK_ENTRY_TIME_8_LINES (0 << 25) -#define EDP_PSR_MIN_LINK_ENTRY_TIME_4_LINES (1 << 25) -#define EDP_PSR_MIN_LINK_ENTRY_TIME_2_LINES (2 << 25) -#define EDP_PSR_MIN_LINK_ENTRY_TIME_0_LINES (3 << 25) -#define EDP_PSR_MAX_SLEEP_TIME_SHIFT 20 -#define EDP_PSR_SKIP_AUX_EXIT (1 << 12) -#define EDP_PSR_TP1_TP2_SEL (0 << 11) -#define EDP_PSR_TP1_TP3_SEL (1 << 11) -#define EDP_PSR_CRC_ENABLE (1 << 10) /* BDW+ */ -#define EDP_PSR_TP2_TP3_TIME_500us (0 << 8) -#define EDP_PSR_TP2_TP3_TIME_100us (1 << 8) -#define EDP_PSR_TP2_TP3_TIME_2500us (2 << 8) -#define EDP_PSR_TP2_TP3_TIME_0us (3 << 8) -#define EDP_PSR_TP4_TIME_0US (3 << 6) /* ICL+ */ -#define EDP_PSR_TP1_TIME_500us (0 << 4) -#define EDP_PSR_TP1_TIME_100us (1 << 4) -#define EDP_PSR_TP1_TIME_2500us (2 << 4) -#define EDP_PSR_TP1_TIME_0us (3 << 4) -#define EDP_PSR_IDLE_FRAME_SHIFT 0 +#define EDP_PSR_ENABLE REG_BIT(31) +#define BDW_PSR_SINGLE_FRAME REG_BIT(30) +#define EDP_PSR_RESTORE_PSR_ACTIVE_CTX_MASK REG_BIT(29) /* SW can't modify */ +#define EDP_PSR_LINK_STANDBY REG_BIT(27) +#define EDP_PSR_MIN_LINK_ENTRY_TIME_MASK REG_GENMASK(26, 25) +#define EDP_PSR_MIN_LINK_ENTRY_TIME_8_LINES REG_FIELD_PREP(EDP_PSR_MIN_LINK_ENTRY_TIME_MASK, 0) +#define EDP_PSR_MIN_LINK_ENTRY_TIME_4_LINES REG_FIELD_PREP(EDP_PSR_MIN_LINK_ENTRY_TIME_MASK, 1) +#define EDP_PSR_MIN_LINK_ENTRY_TIME_2_LINES REG_FIELD_PREP(EDP_PSR_MIN_LINK_ENTRY_TIME_MASK, 2) +#define EDP_PSR_MIN_LINK_ENTRY_TIME_0_LINES REG_FIELD_PREP(EDP_PSR_MIN_LINK_ENTRY_TIME_MASK, 3) +#define EDP_PSR_MAX_SLEEP_TIME_MASK REG_GENMASK(24, 20) +#define EDP_PSR_MAX_SLEEP_TIME(x) REG_FIELD_PREP(EDP_PSR_MAX_SLEEP_TIME_MASK, (x)) +#define EDP_PSR_SKIP_AUX_EXIT REG_BIT(12) +#define EDP_PSR_TP_MASK REG_BIT(11) +#define EDP_PSR_TP_TP1_TP2 REG_FIELD_PREP(EDP_PSR_TP_MASK, 0) +#define EDP_PSR_TP_TP1_TP3 REG_FIELD_PREP(EDP_PSR_TP_MASK, 1) +#define EDP_PSR_CRC_ENABLE REG_BIT(10) /* BDW+ */ +#define EDP_PSR_TP2_TP3_TIME_MASK REG_GENMASK(9, 8) +#define EDP_PSR_TP2_TP3_TIME_500us REG_FIELD_PREP(EDP_PSR_TP2_TP3_TIME_MASK, 0) +#define EDP_PSR_TP2_TP3_TIME_100us REG_FIELD_PREP(EDP_PSR_TP2_TP3_TIME_MASK, 1) +#define EDP_PSR_TP2_TP3_TIME_2500us REG_FIELD_PREP(EDP_PSR_TP2_TP3_TIME_MASK, 2) +#define EDP_PSR_TP2_TP3_TIME_0us REG_FIELD_PREP(EDP_PSR_TP2_TP3_TIME_MASK, 3) +#define EDP_PSR_TP4_TIME_MASK REG_GENMASK(7, 6) +#define EDP_PSR_TP4_TIME_0us REG_FIELD_PREP(EDP_PSR_TP4_TIME_MASK, 3) /* ICL+ */ +#define EDP_PSR_TP1_TIME_MASK REG_GENMASK(5, 4) +#define EDP_PSR_TP1_TIME_500us REG_FIELD_PREP(EDP_PSR_TP1_TIME_MASK, 0) +#define EDP_PSR_TP1_TIME_100us REG_FIELD_PREP(EDP_PSR_TP1_TIME_MASK, 1) +#define EDP_PSR_TP1_TIME_2500us REG_FIELD_PREP(EDP_PSR_TP1_TIME_MASK, 2) +#define EDP_PSR_TP1_TIME_0us REG_FIELD_PREP(EDP_PSR_TP1_TIME_MASK, 3) +#define EDP_PSR_IDLE_FRAMES_MASK REG_GENMASK(3, 0) +#define EDP_PSR_IDLE_FRAMES(x) REG_FIELD_PREP(EDP_PSR_IDLE_FRAMES_MASK, (x)) /* * Until TGL, IMR/IIR are fixed at 0x648xx. On TGL+ those registers are relative @@ -80,81 +86,90 @@ #define _SRD_STATUS_A 0x60840 #define _SRD_STATUS_EDP 0x6f840 #define EDP_PSR_STATUS(tran) _MMIO_TRANS2(tran, _SRD_STATUS_A) -#define EDP_PSR_STATUS_STATE_MASK (7 << 29) -#define EDP_PSR_STATUS_STATE_SHIFT 29 -#define EDP_PSR_STATUS_STATE_IDLE (0 << 29) -#define EDP_PSR_STATUS_STATE_SRDONACK (1 << 29) -#define EDP_PSR_STATUS_STATE_SRDENT (2 << 29) -#define EDP_PSR_STATUS_STATE_BUFOFF (3 << 29) -#define EDP_PSR_STATUS_STATE_BUFON (4 << 29) -#define EDP_PSR_STATUS_STATE_AUXACK (5 << 29) -#define EDP_PSR_STATUS_STATE_SRDOFFACK (6 << 29) -#define EDP_PSR_STATUS_LINK_MASK (3 << 26) -#define EDP_PSR_STATUS_LINK_FULL_OFF (0 << 26) -#define EDP_PSR_STATUS_LINK_FULL_ON (1 << 26) -#define EDP_PSR_STATUS_LINK_STANDBY (2 << 26) -#define EDP_PSR_STATUS_MAX_SLEEP_TIMER_SHIFT 20 -#define EDP_PSR_STATUS_MAX_SLEEP_TIMER_MASK 0x1f -#define EDP_PSR_STATUS_COUNT_SHIFT 16 -#define EDP_PSR_STATUS_COUNT_MASK 0xf -#define EDP_PSR_STATUS_AUX_ERROR (1 << 15) -#define EDP_PSR_STATUS_AUX_SENDING (1 << 12) -#define EDP_PSR_STATUS_SENDING_IDLE (1 << 9) -#define EDP_PSR_STATUS_SENDING_TP2_TP3 (1 << 8) -#define EDP_PSR_STATUS_SENDING_TP1 (1 << 4) -#define EDP_PSR_STATUS_IDLE_MASK 0xf +#define EDP_PSR_STATUS_STATE_MASK REG_GENMASK(31, 29) +#define EDP_PSR_STATUS_STATE_IDLE REG_FIELD_PREP(EDP_PSR_STATUS_STATE_MASK, 0) +#define EDP_PSR_STATUS_STATE_SRDONACK REG_FIELD_PREP(EDP_PSR_STATUS_STATE_MASK, 1) +#define EDP_PSR_STATUS_STATE_SRDENT REG_FIELD_PREP(EDP_PSR_STATUS_STATE_MASK, 2) +#define EDP_PSR_STATUS_STATE_BUFOFF REG_FIELD_PREP(EDP_PSR_STATUS_STATE_MASK, 3) +#define EDP_PSR_STATUS_STATE_BUFON REG_FIELD_PREP(EDP_PSR_STATUS_STATE_MASK, 4) +#define EDP_PSR_STATUS_STATE_AUXACK REG_FIELD_PREP(EDP_PSR_STATUS_STATE_MASK, 5) +#define EDP_PSR_STATUS_STATE_SRDOFFACK REG_FIELD_PREP(EDP_PSR_STATUS_STATE_MASK, 6) +#define EDP_PSR_STATUS_LINK_MASK REG_GENMASK(27, 26) +#define EDP_PSR_STATUS_LINK_FULL_OFF REG_FIELD_PREP(EDP_PSR_STATUS_LINK_MASK, 0) +#define EDP_PSR_STATUS_LINK_FULL_ON REG_FIELD_PREP(EDP_PSR_STATUS_LINK_MASK, 1) +#define EDP_PSR_STATUS_LINK_STANDBY REG_FIELD_PREP(EDP_PSR_STATUS_LINK_MASK, 2) +#define EDP_PSR_STATUS_MAX_SLEEP_TIMER_MASK REG_GENMASK(24, 20) +#define EDP_PSR_STATUS_COUNT_MASK REG_GENMASK(19, 16) +#define EDP_PSR_STATUS_AUX_ERROR REG_BIT(15) +#define EDP_PSR_STATUS_AUX_SENDING REG_BIT(12) +#define EDP_PSR_STATUS_SENDING_IDLE REG_BIT(9) +#define EDP_PSR_STATUS_SENDING_TP2_TP3 REG_BIT(8) +#define EDP_PSR_STATUS_SENDING_TP1 REG_BIT(4) +#define EDP_PSR_STATUS_IDLE_MASK REG_GENMASK(3, 0) #define _SRD_PERF_CNT_A 0x60844 #define _SRD_PERF_CNT_EDP 0x6f844 #define EDP_PSR_PERF_CNT(tran) _MMIO_TRANS2(tran, _SRD_PERF_CNT_A) -#define EDP_PSR_PERF_CNT_MASK 0xffffff +#define EDP_PSR_PERF_CNT_MASK REG_GENMASK(23, 0) /* PSR_MASK on SKL+ */ #define _SRD_DEBUG_A 0x60860 #define _SRD_DEBUG_EDP 0x6f860 #define EDP_PSR_DEBUG(tran) _MMIO_TRANS2(tran, _SRD_DEBUG_A) -#define EDP_PSR_DEBUG_MASK_MAX_SLEEP (1 << 28) -#define EDP_PSR_DEBUG_MASK_LPSP (1 << 27) -#define EDP_PSR_DEBUG_MASK_MEMUP (1 << 26) -#define EDP_PSR_DEBUG_MASK_HPD (1 << 25) -#define EDP_PSR_DEBUG_MASK_DISP_REG_WRITE (1 << 16) /* Reserved in ICL+ */ -#define EDP_PSR_DEBUG_EXIT_ON_PIXEL_UNDERRUN (1 << 15) /* SKL+ */ +#define EDP_PSR_DEBUG_MASK_MAX_SLEEP REG_BIT(28) +#define EDP_PSR_DEBUG_MASK_LPSP REG_BIT(27) +#define EDP_PSR_DEBUG_MASK_MEMUP REG_BIT(26) +#define EDP_PSR_DEBUG_MASK_HPD REG_BIT(25) +#define EDP_PSR_DEBUG_MASK_FBC_MODIFY REG_BIT(24) +#define EDP_PSR_DEBUG_MASK_PRIMARY_FLIP REG_BIT(23) /* hsw */ +#define EDP_PSR_DEBUG_MASK_HDCP_ENABLE REG_BIT(22) /* hsw/bdw */ +#define EDP_PSR_DEBUG_MASK_SPRITE_ENABLE REG_BIT(21) /* hsw */ +#define EDP_PSR_DEBUG_MASK_CURSOR_MOVE REG_BIT(20) /* hsw */ +#define EDP_PSR_DEBUG_MASK_VBLANK_VSYNC_INT REG_BIT(19) /* hsw */ +#define EDP_PSR_DEBUG_MASK_DPST_PHASE_IN REG_BIT(18) /* hsw */ +#define EDP_PSR_DEBUG_MASK_KVMR_SESSION_EN REG_BIT(17) +#define EDP_PSR_DEBUG_MASK_DISP_REG_WRITE REG_BIT(16) /* hsw-skl */ +#define EDP_PSR_DEBUG_EXIT_ON_PIXEL_UNDERRUN REG_BIT(15) /* skl+ */ +#define EDP_PSR_DEBUG_RFB_UPDATE_SENT REG_BIT(2) /* bdw */ +#define EDP_PSR_DEBUG_ENTRY_COMPLETION REG_BIT(1) /* hsw/bdw */ #define _PSR2_CTL_A 0x60900 #define _PSR2_CTL_EDP 0x6f900 #define EDP_PSR2_CTL(tran) _MMIO_TRANS2(tran, _PSR2_CTL_A) -#define EDP_PSR2_ENABLE (1 << 31) -#define EDP_SU_TRACK_ENABLE (1 << 30) /* up to adl-p */ -#define TGL_EDP_PSR2_BLOCK_COUNT_NUM_2 (0 << 28) -#define TGL_EDP_PSR2_BLOCK_COUNT_NUM_3 (1 << 28) +#define EDP_PSR2_ENABLE REG_BIT(31) +#define EDP_SU_TRACK_ENABLE REG_BIT(30) /* up to adl-p */ +#define TGL_EDP_PSR2_BLOCK_COUNT_MASK REG_BIT(28) +#define TGL_EDP_PSR2_BLOCK_COUNT_NUM_2 REG_FIELD_PREP(TGL_EDP_PSR2_BLOCK_COUNT_MASK, 0) +#define TGL_EDP_PSR2_BLOCK_COUNT_NUM_3 REG_FIELD_PREP(TGL_EDP_PSR2_BLOCK_COUNT_MASK, 1) #define EDP_Y_COORDINATE_ENABLE REG_BIT(25) /* display 10, 11 and 12 */ #define EDP_PSR2_SU_SDP_SCANLINE REG_BIT(25) /* display 13+ */ -#define EDP_MAX_SU_DISABLE_TIME(t) ((t) << 20) -#define EDP_MAX_SU_DISABLE_TIME_MASK (0x1f << 20) +#define EDP_MAX_SU_DISABLE_TIME_MASK REG_GENMASK(24, 20) +#define EDP_MAX_SU_DISABLE_TIME(t) REG_FIELD_PREP(EDP_MAX_SU_DISABLE_TIME, (t)) +#define EDP_PSR2_IO_BUFFER_WAKE_MASK REG_GENMASK(14, 13) #define EDP_PSR2_IO_BUFFER_WAKE_MAX_LINES 8 -#define EDP_PSR2_IO_BUFFER_WAKE(lines) ((EDP_PSR2_IO_BUFFER_WAKE_MAX_LINES - (lines)) << 13) -#define EDP_PSR2_IO_BUFFER_WAKE_MASK (3 << 13) +#define EDP_PSR2_IO_BUFFER_WAKE(lines) REG_FIELD_PREP(EDP_PSR2_IO_BUFFER_WAKE_MASK, \ + EDP_PSR2_IO_BUFFER_WAKE_MAX_LINES - (lines)) +#define TGL_EDP_PSR2_IO_BUFFER_WAKE_MASK REG_GENMASK(15, 13) #define TGL_EDP_PSR2_IO_BUFFER_WAKE_MIN_LINES 5 -#define TGL_EDP_PSR2_IO_BUFFER_WAKE_SHIFT 13 -#define TGL_EDP_PSR2_IO_BUFFER_WAKE(lines) (((lines) - TGL_EDP_PSR2_IO_BUFFER_WAKE_MIN_LINES) << TGL_EDP_PSR2_IO_BUFFER_WAKE_SHIFT) -#define TGL_EDP_PSR2_IO_BUFFER_WAKE_MASK (7 << 13) +#define TGL_EDP_PSR2_IO_BUFFER_WAKE(lines) REG_FIELD_PREP(TGL_EDP_PSR2_IO_BUFFER_WAKE_MASK, \ + (lines) - TGL_EDP_PSR2_IO_BUFFER_WAKE_MIN_LINES) +#define EDP_PSR2_FAST_WAKE_MASK REG_GENMASK(12, 11) #define EDP_PSR2_FAST_WAKE_MAX_LINES 8 -#define EDP_PSR2_FAST_WAKE(lines) ((EDP_PSR2_FAST_WAKE_MAX_LINES - (lines)) << 11) -#define EDP_PSR2_FAST_WAKE_MASK (3 << 11) +#define EDP_PSR2_FAST_WAKE(lines) REG_FIELD_PREP(EDP_PSR2_FAST_WAKE_MASK, \ + EDP_PSR2_FAST_WAKE_MAX_LINES - (lines)) +#define TGL_EDP_PSR2_FAST_WAKE_MASK REG_GENMASK(12, 10) #define TGL_EDP_PSR2_FAST_WAKE_MIN_LINES 5 -#define TGL_EDP_PSR2_FAST_WAKE_MIN_SHIFT 10 -#define TGL_EDP_PSR2_FAST_WAKE(lines) (((lines) - TGL_EDP_PSR2_FAST_WAKE_MIN_LINES) << TGL_EDP_PSR2_FAST_WAKE_MIN_SHIFT) -#define TGL_EDP_PSR2_FAST_WAKE_MASK (7 << 10) -#define EDP_PSR2_TP2_TIME_500us (0 << 8) -#define EDP_PSR2_TP2_TIME_100us (1 << 8) -#define EDP_PSR2_TP2_TIME_2500us (2 << 8) -#define EDP_PSR2_TP2_TIME_50us (3 << 8) -#define EDP_PSR2_TP2_TIME_MASK (3 << 8) -#define EDP_PSR2_FRAME_BEFORE_SU_SHIFT 4 -#define EDP_PSR2_FRAME_BEFORE_SU_MASK (0xf << 4) -#define EDP_PSR2_FRAME_BEFORE_SU(a) ((a) << 4) -#define EDP_PSR2_IDLE_FRAME_MASK 0xf -#define EDP_PSR2_IDLE_FRAME_SHIFT 0 +#define TGL_EDP_PSR2_FAST_WAKE(lines) REG_FIELD_PREP(TGL_EDP_PSR2_FAST_WAKE_MASK, \ + (lines) - TGL_EDP_PSR2_FAST_WAKE_MIN_LINES) +#define EDP_PSR2_TP2_TIME_MASK REG_GENMASK(9, 8) +#define EDP_PSR2_TP2_TIME_500us REG_FIELD_PREP(EDP_PSR2_TP2_TIME_MASK, 0) +#define EDP_PSR2_TP2_TIME_100us REG_FIELD_PREP(EDP_PSR2_TP2_TIME_MASK, 1) +#define EDP_PSR2_TP2_TIME_2500us REG_FIELD_PREP(EDP_PSR2_TP2_TIME_MASK, 2) +#define EDP_PSR2_TP2_TIME_50us REG_FIELD_PREP(EDP_PSR2_TP2_TIME_MASK, 3) +#define EDP_PSR2_FRAME_BEFORE_SU_MASK REG_GENMASK(7, 4) +#define EDP_PSR2_FRAME_BEFORE_SU(a) REG_FIELD_PREP(EDP_PSR2_FRAME_BEFORE_SU_MASK, (a)) +#define EDP_PSR2_IDLE_FRAMES_MASK REG_GENMASK(3, 0) +#define EDP_PSR2_IDLE_FRAMES(x) REG_FIELD_PREP(EDP_PSR2_IDLE_FRAMES_MASK, (x)) #define _PSR_EVENT_TRANS_A 0x60848 #define _PSR_EVENT_TRANS_B 0x61848 @@ -162,22 +177,22 @@ #define _PSR_EVENT_TRANS_D 0x63848 #define _PSR_EVENT_TRANS_EDP 0x6f848 #define PSR_EVENT(tran) _MMIO_TRANS2(tran, _PSR_EVENT_TRANS_A) -#define PSR_EVENT_PSR2_WD_TIMER_EXPIRE (1 << 17) -#define PSR_EVENT_PSR2_DISABLED (1 << 16) -#define PSR_EVENT_SU_DIRTY_FIFO_UNDERRUN (1 << 15) -#define PSR_EVENT_SU_CRC_FIFO_UNDERRUN (1 << 14) -#define PSR_EVENT_GRAPHICS_RESET (1 << 12) -#define PSR_EVENT_PCH_INTERRUPT (1 << 11) -#define PSR_EVENT_MEMORY_UP (1 << 10) -#define PSR_EVENT_FRONT_BUFFER_MODIFY (1 << 9) -#define PSR_EVENT_WD_TIMER_EXPIRE (1 << 8) -#define PSR_EVENT_PIPE_REGISTERS_UPDATE (1 << 6) -#define PSR_EVENT_REGISTER_UPDATE (1 << 5) /* Reserved in ICL+ */ -#define PSR_EVENT_HDCP_ENABLE (1 << 4) -#define PSR_EVENT_KVMR_SESSION_ENABLE (1 << 3) -#define PSR_EVENT_VBI_ENABLE (1 << 2) -#define PSR_EVENT_LPSP_MODE_EXIT (1 << 1) -#define PSR_EVENT_PSR_DISABLE (1 << 0) +#define PSR_EVENT_PSR2_WD_TIMER_EXPIRE REG_BIT(17) +#define PSR_EVENT_PSR2_DISABLED REG_BIT(16) +#define PSR_EVENT_SU_DIRTY_FIFO_UNDERRUN REG_BIT(15) +#define PSR_EVENT_SU_CRC_FIFO_UNDERRUN REG_BIT(14) +#define PSR_EVENT_GRAPHICS_RESET REG_BIT(12) +#define PSR_EVENT_PCH_INTERRUPT REG_BIT(11) +#define PSR_EVENT_MEMORY_UP REG_BIT(10) +#define PSR_EVENT_FRONT_BUFFER_MODIFY REG_BIT(9) +#define PSR_EVENT_WD_TIMER_EXPIRE REG_BIT(8) +#define PSR_EVENT_PIPE_REGISTERS_UPDATE REG_BIT(6) +#define PSR_EVENT_REGISTER_UPDATE REG_BIT(5) /* Reserved in ICL+ */ +#define PSR_EVENT_HDCP_ENABLE REG_BIT(4) +#define PSR_EVENT_KVMR_SESSION_ENABLE REG_BIT(3) +#define PSR_EVENT_VBI_ENABLE REG_BIT(2) +#define PSR_EVENT_LPSP_MODE_EXIT REG_BIT(1) +#define PSR_EVENT_PSR_DISABLE REG_BIT(0) #define _PSR2_STATUS_A 0x60940 #define _PSR2_STATUS_EDP 0x6f940 diff --git a/drivers/gpu/drm/i915/display/intel_sdvo.c b/drivers/gpu/drm/i915/display/intel_sdvo.c index e12ba458636c1e3cf4267e44cb44c6e21450dfe9..21f92123c844645069311deb630650cccc04b6e0 100644 --- a/drivers/gpu/drm/i915/display/intel_sdvo.c +++ b/drivers/gpu/drm/i915/display/intel_sdvo.c @@ -117,9 +117,6 @@ struct intel_sdvo { enum port port; - bool has_hdmi_monitor; - bool has_hdmi_audio; - /* DDC bus used by this SDVO encoder */ u8 ddc_bus; @@ -1303,10 +1300,13 @@ static void i9xx_adjust_sdvo_tv_clock(struct intel_crtc_state *pipe_config) pipe_config->clock_set = true; } -static bool intel_has_hdmi_sink(struct intel_sdvo *sdvo, +static bool intel_has_hdmi_sink(struct intel_sdvo_connector *intel_sdvo_connector, const struct drm_connector_state *conn_state) { - return sdvo->has_hdmi_monitor && + struct drm_connector *connector = conn_state->connector; + + return intel_sdvo_connector->is_hdmi && + connector->display_info.is_hdmi && READ_ONCE(to_intel_digital_connector_state(conn_state)->force_audio) != HDMI_AUDIO_OFF_DVI; } @@ -1326,7 +1326,9 @@ static bool intel_sdvo_has_audio(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state, const struct drm_connector_state *conn_state) { - struct intel_sdvo *intel_sdvo = to_sdvo(encoder); + struct drm_connector *connector = conn_state->connector; + struct intel_sdvo_connector *intel_sdvo_connector = + to_intel_sdvo_connector(connector); const struct intel_digital_connector_state *intel_conn_state = to_intel_digital_connector_state(conn_state); @@ -1334,7 +1336,8 @@ static bool intel_sdvo_has_audio(struct intel_encoder *encoder, return false; if (intel_conn_state->force_audio == HDMI_AUDIO_AUTO) - return intel_sdvo->has_hdmi_audio; + return intel_sdvo_connector->is_hdmi && + connector->display_info.has_audio; else return intel_conn_state->force_audio == HDMI_AUDIO_ON; } @@ -1351,6 +1354,7 @@ static int intel_sdvo_compute_config(struct intel_encoder *encoder, DRM_DEBUG_KMS("forcing bpc to 8 for SDVO\n"); pipe_config->pipe_bpp = 8*3; + pipe_config->sink_format = INTEL_OUTPUT_FORMAT_RGB; pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB; if (HAS_PCH_SPLIT(to_i915(encoder->base.dev))) @@ -1400,7 +1404,7 @@ static int intel_sdvo_compute_config(struct intel_encoder *encoder, pipe_config->pixel_multiplier = intel_sdvo_get_pixel_multiplier(adjusted_mode); - pipe_config->has_hdmi_sink = intel_has_hdmi_sink(intel_sdvo, conn_state); + pipe_config->has_hdmi_sink = intel_has_hdmi_sink(intel_sdvo_connector, conn_state); pipe_config->has_audio = intel_sdvo_has_audio(encoder, pipe_config, conn_state) && @@ -1906,7 +1910,7 @@ intel_sdvo_mode_valid(struct drm_connector *connector, struct intel_sdvo_connector *intel_sdvo_connector = to_intel_sdvo_connector(connector); int max_dotclk = to_i915(connector->dev)->max_dotclk_freq; - bool has_hdmi_sink = intel_has_hdmi_sink(intel_sdvo, connector->state); + bool has_hdmi_sink = intel_has_hdmi_sink(intel_sdvo_connector, connector->state); int clock = mode->clock; if (mode->flags & DRM_MODE_FLAG_DBLSCAN) @@ -2032,36 +2036,35 @@ intel_sdvo_multifunc_encoder(struct intel_sdvo *intel_sdvo) return hweight16(intel_sdvo->caps.output_flags) > 1; } -static struct edid * +static const struct drm_edid * intel_sdvo_get_edid(struct drm_connector *connector) { struct intel_sdvo *sdvo = intel_attached_sdvo(to_intel_connector(connector)); - return drm_get_edid(connector, &sdvo->ddc); + return drm_edid_read_ddc(connector, &sdvo->ddc); } /* Mac mini hack -- use the same DDC as the analog connector */ -static struct edid * +static const struct drm_edid * intel_sdvo_get_analog_edid(struct drm_connector *connector) { - struct drm_i915_private *dev_priv = to_i915(connector->dev); + struct drm_i915_private *i915 = to_i915(connector->dev); + struct i2c_adapter *i2c; - return drm_get_edid(connector, - intel_gmbus_get_adapter(dev_priv, - dev_priv->display.vbt.crt_ddc_pin)); + i2c = intel_gmbus_get_adapter(i915, i915->display.vbt.crt_ddc_pin); + + return drm_edid_read_ddc(connector, i2c); } static enum drm_connector_status intel_sdvo_tmds_sink_detect(struct drm_connector *connector) { struct intel_sdvo *intel_sdvo = intel_attached_sdvo(to_intel_connector(connector)); - struct intel_sdvo_connector *intel_sdvo_connector = - to_intel_sdvo_connector(connector); enum drm_connector_status status; - struct edid *edid; + const struct drm_edid *drm_edid; - edid = intel_sdvo_get_edid(connector); + drm_edid = intel_sdvo_get_edid(connector); - if (edid == NULL && intel_sdvo_multifunc_encoder(intel_sdvo)) { + if (!drm_edid && intel_sdvo_multifunc_encoder(intel_sdvo)) { u8 ddc, saved_ddc = intel_sdvo->ddc_bus; /* @@ -2070,15 +2073,15 @@ intel_sdvo_tmds_sink_detect(struct drm_connector *connector) */ for (ddc = intel_sdvo->ddc_bus >> 1; ddc > 1; ddc >>= 1) { intel_sdvo->ddc_bus = ddc; - edid = intel_sdvo_get_edid(connector); - if (edid) + drm_edid = intel_sdvo_get_edid(connector); + if (drm_edid) break; } /* * If we found the EDID on the other bus, * assume that is the correct DDC bus. */ - if (edid == NULL) + if (!drm_edid) intel_sdvo->ddc_bus = saved_ddc; } @@ -2086,21 +2089,19 @@ intel_sdvo_tmds_sink_detect(struct drm_connector *connector) * When there is no edid and no monitor is connected with VGA * port, try to use the CRT ddc to read the EDID for DVI-connector. */ - if (edid == NULL) - edid = intel_sdvo_get_analog_edid(connector); + if (!drm_edid) + drm_edid = intel_sdvo_get_analog_edid(connector); status = connector_status_unknown; - if (edid != NULL) { + if (drm_edid) { + const struct edid *edid = drm_edid_raw(drm_edid); + /* DDC bus is shared, match EDID to connector type */ - if (edid->input & DRM_EDID_INPUT_DIGITAL) { + if (edid->input & DRM_EDID_INPUT_DIGITAL) status = connector_status_connected; - if (intel_sdvo_connector->is_hdmi) { - intel_sdvo->has_hdmi_monitor = drm_detect_hdmi_monitor(edid); - intel_sdvo->has_hdmi_audio = drm_detect_monitor_audio(edid); - } - } else + else status = connector_status_disconnected; - kfree(edid); + drm_edid_free(drm_edid); } return status; @@ -2108,8 +2109,9 @@ intel_sdvo_tmds_sink_detect(struct drm_connector *connector) static bool intel_sdvo_connector_matches_edid(struct intel_sdvo_connector *sdvo, - struct edid *edid) + const struct drm_edid *drm_edid) { + const struct edid *edid = drm_edid_raw(drm_edid); bool monitor_is_digital = !!(edid->input & DRM_EDID_INPUT_DIGITAL); bool connector_is_digital = !!IS_DIGITAL(sdvo); @@ -2147,30 +2149,28 @@ intel_sdvo_detect(struct drm_connector *connector, bool force) intel_sdvo->attached_output = response; - intel_sdvo->has_hdmi_monitor = false; - intel_sdvo->has_hdmi_audio = false; - if ((intel_sdvo_connector->output_flag & response) == 0) ret = connector_status_disconnected; else if (IS_TMDS(intel_sdvo_connector)) ret = intel_sdvo_tmds_sink_detect(connector); else { - struct edid *edid; + const struct drm_edid *drm_edid; /* if we have an edid check it matches the connection */ - edid = intel_sdvo_get_edid(connector); - if (edid == NULL) - edid = intel_sdvo_get_analog_edid(connector); - if (edid != NULL) { + drm_edid = intel_sdvo_get_edid(connector); + if (!drm_edid) + drm_edid = intel_sdvo_get_analog_edid(connector); + if (drm_edid) { if (intel_sdvo_connector_matches_edid(intel_sdvo_connector, - edid)) + drm_edid)) ret = connector_status_connected; else ret = connector_status_disconnected; - kfree(edid); - } else + drm_edid_free(drm_edid); + } else { ret = connector_status_connected; + } } return ret; @@ -2179,13 +2179,13 @@ intel_sdvo_detect(struct drm_connector *connector, bool force) static int intel_sdvo_get_ddc_modes(struct drm_connector *connector) { int num_modes = 0; - struct edid *edid; + const struct drm_edid *drm_edid; DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n", connector->base.id, connector->name); /* set the bus switch and get the modes */ - edid = intel_sdvo_get_edid(connector); + drm_edid = intel_sdvo_get_edid(connector); /* * Mac mini hack. On this device, the DVI-I connector shares one DDC @@ -2193,17 +2193,17 @@ static int intel_sdvo_get_ddc_modes(struct drm_connector *connector) * DDC fails, check to see if the analog output is disconnected, in * which case we'll look there for the digital DDC data. */ - if (!edid) - edid = intel_sdvo_get_analog_edid(connector); + if (!drm_edid) + drm_edid = intel_sdvo_get_analog_edid(connector); - if (!edid) + if (!drm_edid) return 0; if (intel_sdvo_connector_matches_edid(to_intel_sdvo_connector(connector), - edid)) - num_modes += intel_connector_update_modes(connector, edid); + drm_edid)) + num_modes += intel_connector_update_modes(connector, drm_edid); - kfree(edid); + drm_edid_free(drm_edid); return num_modes; } diff --git a/drivers/gpu/drm/i915/display/intel_snps_phy.c b/drivers/gpu/drm/i915/display/intel_snps_phy.c index 1cfb94b5cedbdfa757f0a16e3f25f73e5d28e37c..88ef56b6e0fd0b7e9d37b672c271a80914c6a03b 100644 --- a/drivers/gpu/drm/i915/display/intel_snps_phy.c +++ b/drivers/gpu/drm/i915/display/intel_snps_phy.c @@ -2007,11 +2007,16 @@ void intel_mpllb_state_verify(struct intel_atomic_state *state, if (!new_crtc_state->hw.active) return; + /* intel_get_crtc_new_encoder() only works for modeset/fastset commits */ + if (!intel_crtc_needs_modeset(new_crtc_state) && + !intel_crtc_needs_fastset(new_crtc_state)) + return; + encoder = intel_get_crtc_new_encoder(state, new_crtc_state); intel_mpllb_readout_hw_state(encoder, &mpllb_hw_state); #define MPLLB_CHECK(__name) \ - I915_STATE_WARN(mpllb_sw_state->__name != mpllb_hw_state.__name, \ + I915_STATE_WARN(i915, mpllb_sw_state->__name != mpllb_hw_state.__name, \ "[CRTC:%d:%s] mismatch in MPLLB: %s (expected 0x%08x, found 0x%08x)", \ crtc->base.base.id, crtc->base.name, \ __stringify(__name), \ diff --git a/drivers/gpu/drm/i915/display/intel_sprite.h b/drivers/gpu/drm/i915/display/intel_sprite.h index 4635c7ad23f95a4606998eb27e8304b3755edcc8..91c6dca342b2d647b56fbb332685e7c1730374bc 100644 --- a/drivers/gpu/drm/i915/display/intel_sprite.h +++ b/drivers/gpu/drm/i915/display/intel_sprite.h @@ -16,16 +16,6 @@ struct intel_crtc_state; struct intel_plane_state; enum pipe; -/* - * FIXME: We should instead only take spinlocks once for the entire update - * instead of once per mmio. - */ -#if IS_ENABLED(CONFIG_PROVE_LOCKING) -#define VBLANK_EVASION_TIME_US 250 -#else -#define VBLANK_EVASION_TIME_US 100 -#endif - struct intel_plane *intel_sprite_plane_create(struct drm_i915_private *dev_priv, enum pipe pipe, int plane); int intel_sprite_set_colorkey_ioctl(struct drm_device *dev, void *data, diff --git a/drivers/gpu/drm/i915/display/intel_sprite_uapi.c b/drivers/gpu/drm/i915/display/intel_sprite_uapi.c index 70a3910837519b6dc56ea64ef447a1581486566b..a76b48ebc2d31e437ff73869456681f9ffe6f8be 100644 --- a/drivers/gpu/drm/i915/display/intel_sprite_uapi.c +++ b/drivers/gpu/drm/i915/display/intel_sprite_uapi.c @@ -86,6 +86,7 @@ int intel_sprite_set_colorkey_ioctl(struct drm_device *dev, void *data, goto out; } state->acquire_ctx = &ctx; + to_intel_atomic_state(state)->internal = true; while (1) { plane_state = drm_atomic_get_plane_state(state, plane); diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c index 3b60995e9dfb3f943faefd7d33f0c60517757a22..3ebf41859043e9c83b57d2a7f100beac9887a5d7 100644 --- a/drivers/gpu/drm/i915/display/intel_tc.c +++ b/drivers/gpu/drm/i915/display/intel_tc.c @@ -5,16 +5,25 @@ #include "i915_drv.h" #include "i915_reg.h" +#include "intel_atomic.h" +#include "intel_cx0_phy_regs.h" #include "intel_ddi.h" #include "intel_de.h" #include "intel_display.h" +#include "intel_display_driver.h" #include "intel_display_power_map.h" #include "intel_display_types.h" #include "intel_dkl_phy_regs.h" +#include "intel_dp.h" #include "intel_dp_mst.h" #include "intel_mg_phy_regs.h" +#include "intel_modeset_lock.h" #include "intel_tc.h" +#define DP_PIN_ASSIGNMENT_C 0x3 +#define DP_PIN_ASSIGNMENT_D 0x4 +#define DP_PIN_ASSIGNMENT_E 0x5 + enum tc_port_mode { TC_PORT_DISCONNECTED, TC_PORT_TBT_ALT, @@ -46,6 +55,7 @@ struct intel_tc_port { enum intel_display_power_domain lock_power_domain; #endif struct delayed_work disconnect_phy_work; + struct delayed_work link_reset_work; int link_refcount; bool legacy_port:1; char port_name[8]; @@ -59,6 +69,7 @@ static enum intel_display_power_domain tc_phy_cold_off_domain(struct intel_tc_port *); static u32 tc_phy_hpd_live_status(struct intel_tc_port *tc); static bool tc_phy_is_ready(struct intel_tc_port *tc); +static bool tc_phy_wait_for_ready(struct intel_tc_port *tc); static enum tc_port_mode tc_phy_get_current_mode(struct intel_tc_port *tc); static const char *tc_port_mode_name(enum tc_port_mode mode) @@ -141,15 +152,23 @@ bool intel_tc_port_in_legacy_mode(struct intel_digital_port *dig_port) * * POWER_DOMAIN_TC_COLD_OFF: * ------------------------- - * TGL/legacy, DP-alt modes: + * ICL/DP-alt, TBT mode: + * - TCSS/TBT: block TC-cold power state for using the (direct or + * TBT DP-IN) AUX and main lanes. + * + * TGL/all modes: * - TCSS/IOM,FIA access for PHY ready, owned and HPD live state - * - TCSS/PHY: block TC-cold power state for using the PHY AUX and - * main lanes. + * - TCSS/PHY: block TC-cold power state for using the (direct or + * TBT DP-IN) AUX and main lanes. * - * ICL, TGL, ADLP/TBT mode: - * - TCSS/IOM,FIA access for HPD live state + * ADLP/TBT mode: * - TCSS/TBT: block TC-cold power state for using the (TBT DP-IN) * AUX and main lanes. + * + * XELPDP+/all modes: + * - TCSS/IOM,FIA access for PHY ready, owned state + * - TCSS/PHY: block TC-cold power state for using the (direct or + * TBT DP-IN) AUX and main lanes. */ bool intel_tc_cold_requires_aux_pw(struct intel_digital_port *dig_port) { @@ -271,6 +290,27 @@ u32 intel_tc_port_get_pin_assignment_mask(struct intel_digital_port *dig_port) DP_PIN_ASSIGNMENT_SHIFT(tc->phy_fia_idx); } +static int mtl_tc_port_get_pin_assignment_mask(struct intel_digital_port *dig_port) +{ + struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev); + intel_wakeref_t wakeref; + u32 pin_mask; + + with_intel_display_power(i915, POWER_DOMAIN_DISPLAY_CORE, wakeref) + pin_mask = intel_tc_port_get_pin_assignment_mask(dig_port); + + switch (pin_mask) { + default: + MISSING_CASE(pin_mask); + fallthrough; + case DP_PIN_ASSIGNMENT_D: + return 2; + case DP_PIN_ASSIGNMENT_C: + case DP_PIN_ASSIGNMENT_E: + return 4; + } +} + int intel_tc_port_fia_max_lane_count(struct intel_digital_port *dig_port) { struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev); @@ -284,6 +324,9 @@ int intel_tc_port_fia_max_lane_count(struct intel_digital_port *dig_port) assert_tc_cold_blocked(tc); + if (DISPLAY_VER(i915) >= 14) + return mtl_tc_port_get_pin_assignment_mask(dig_port); + lane_mask = 0; with_intel_display_power(i915, POWER_DOMAIN_DISPLAY_CORE, wakeref) lane_mask = intel_tc_port_get_lane_mask(dig_port); @@ -872,6 +915,200 @@ static const struct intel_tc_phy_ops adlp_tc_phy_ops = { .init = adlp_tc_phy_init, }; +/* + * XELPDP TC PHY handlers + * ---------------------- + */ +static u32 xelpdp_tc_phy_hpd_live_status(struct intel_tc_port *tc) +{ + struct drm_i915_private *i915 = tc_to_i915(tc); + struct intel_digital_port *dig_port = tc->dig_port; + enum hpd_pin hpd_pin = dig_port->base.hpd_pin; + u32 pica_isr_bits = i915->display.hotplug.hpd[hpd_pin]; + u32 pch_isr_bit = i915->display.hotplug.pch_hpd[hpd_pin]; + intel_wakeref_t wakeref; + u32 pica_isr; + u32 pch_isr; + u32 mask = 0; + + with_intel_display_power(i915, POWER_DOMAIN_DISPLAY_CORE, wakeref) { + pica_isr = intel_de_read(i915, PICAINTERRUPT_ISR); + pch_isr = intel_de_read(i915, SDEISR); + } + + if (pica_isr & (pica_isr_bits & XELPDP_DP_ALT_HOTPLUG_MASK)) + mask |= BIT(TC_PORT_DP_ALT); + if (pica_isr & (pica_isr_bits & XELPDP_TBT_HOTPLUG_MASK)) + mask |= BIT(TC_PORT_TBT_ALT); + + if (tc->legacy_port && (pch_isr & pch_isr_bit)) + mask |= BIT(TC_PORT_LEGACY); + + return mask; +} + +static bool +xelpdp_tc_phy_tcss_power_is_enabled(struct intel_tc_port *tc) +{ + struct drm_i915_private *i915 = tc_to_i915(tc); + enum port port = tc->dig_port->base.port; + + assert_tc_cold_blocked(tc); + + return intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port)) & XELPDP_TCSS_POWER_STATE; +} + +static bool +xelpdp_tc_phy_wait_for_tcss_power(struct intel_tc_port *tc, bool enabled) +{ + struct drm_i915_private *i915 = tc_to_i915(tc); + + if (wait_for(xelpdp_tc_phy_tcss_power_is_enabled(tc) == enabled, 5)) { + drm_dbg_kms(&i915->drm, + "Port %s: timeout waiting for TCSS power to get %s\n", + enabled ? "enabled" : "disabled", + tc->port_name); + return false; + } + + return true; +} + +static void __xelpdp_tc_phy_enable_tcss_power(struct intel_tc_port *tc, bool enable) +{ + struct drm_i915_private *i915 = tc_to_i915(tc); + enum port port = tc->dig_port->base.port; + u32 val; + + assert_tc_cold_blocked(tc); + + val = intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port)); + if (enable) + val |= XELPDP_TCSS_POWER_REQUEST; + else + val &= ~XELPDP_TCSS_POWER_REQUEST; + intel_de_write(i915, XELPDP_PORT_BUF_CTL1(port), val); +} + +static bool xelpdp_tc_phy_enable_tcss_power(struct intel_tc_port *tc, bool enable) +{ + struct drm_i915_private *i915 = tc_to_i915(tc); + + __xelpdp_tc_phy_enable_tcss_power(tc, enable); + + if ((!tc_phy_wait_for_ready(tc) || + !xelpdp_tc_phy_wait_for_tcss_power(tc, enable)) && + !drm_WARN_ON(&i915->drm, tc->mode == TC_PORT_LEGACY)) { + if (enable) { + __xelpdp_tc_phy_enable_tcss_power(tc, false); + xelpdp_tc_phy_wait_for_tcss_power(tc, false); + } + + return false; + } + + return true; +} + +static void xelpdp_tc_phy_take_ownership(struct intel_tc_port *tc, bool take) +{ + struct drm_i915_private *i915 = tc_to_i915(tc); + enum port port = tc->dig_port->base.port; + u32 val; + + assert_tc_cold_blocked(tc); + + val = intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port)); + if (take) + val |= XELPDP_TC_PHY_OWNERSHIP; + else + val &= ~XELPDP_TC_PHY_OWNERSHIP; + intel_de_write(i915, XELPDP_PORT_BUF_CTL1(port), val); +} + +static bool xelpdp_tc_phy_is_owned(struct intel_tc_port *tc) +{ + struct drm_i915_private *i915 = tc_to_i915(tc); + enum port port = tc->dig_port->base.port; + + assert_tc_cold_blocked(tc); + + return intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port)) & XELPDP_TC_PHY_OWNERSHIP; +} + +static void xelpdp_tc_phy_get_hw_state(struct intel_tc_port *tc) +{ + struct drm_i915_private *i915 = tc_to_i915(tc); + intel_wakeref_t tc_cold_wref; + enum intel_display_power_domain domain; + + tc_cold_wref = __tc_cold_block(tc, &domain); + + tc->mode = tc_phy_get_current_mode(tc); + if (tc->mode != TC_PORT_DISCONNECTED) + tc->lock_wakeref = tc_cold_block(tc); + + drm_WARN_ON(&i915->drm, + (tc->mode == TC_PORT_DP_ALT || tc->mode == TC_PORT_LEGACY) && + !xelpdp_tc_phy_tcss_power_is_enabled(tc)); + + __tc_cold_unblock(tc, domain, tc_cold_wref); +} + +static bool xelpdp_tc_phy_connect(struct intel_tc_port *tc, int required_lanes) +{ + tc->lock_wakeref = tc_cold_block(tc); + + if (tc->mode == TC_PORT_TBT_ALT) + return true; + + if (!xelpdp_tc_phy_enable_tcss_power(tc, true)) + goto out_unblock_tccold; + + xelpdp_tc_phy_take_ownership(tc, true); + + if (!tc_phy_verify_legacy_or_dp_alt_mode(tc, required_lanes)) + goto out_release_phy; + + return true; + +out_release_phy: + xelpdp_tc_phy_take_ownership(tc, false); + xelpdp_tc_phy_wait_for_tcss_power(tc, false); + +out_unblock_tccold: + tc_cold_unblock(tc, fetch_and_zero(&tc->lock_wakeref)); + + return false; +} + +static void xelpdp_tc_phy_disconnect(struct intel_tc_port *tc) +{ + switch (tc->mode) { + case TC_PORT_LEGACY: + case TC_PORT_DP_ALT: + xelpdp_tc_phy_take_ownership(tc, false); + xelpdp_tc_phy_enable_tcss_power(tc, false); + fallthrough; + case TC_PORT_TBT_ALT: + tc_cold_unblock(tc, fetch_and_zero(&tc->lock_wakeref)); + break; + default: + MISSING_CASE(tc->mode); + } +} + +static const struct intel_tc_phy_ops xelpdp_tc_phy_ops = { + .cold_off_domain = tgl_tc_phy_cold_off_domain, + .hpd_live_status = xelpdp_tc_phy_hpd_live_status, + .is_ready = adlp_tc_phy_is_ready, + .is_owned = xelpdp_tc_phy_is_owned, + .get_hw_state = xelpdp_tc_phy_get_hw_state, + .connect = xelpdp_tc_phy_connect, + .disconnect = xelpdp_tc_phy_disconnect, + .init = adlp_tc_phy_init, +}; + /* * Generic TC PHY handlers * ----------------------- @@ -945,13 +1182,18 @@ static bool tc_phy_is_connected(struct intel_tc_port *tc, return is_connected; } -static void tc_phy_wait_for_ready(struct intel_tc_port *tc) +static bool tc_phy_wait_for_ready(struct intel_tc_port *tc) { struct drm_i915_private *i915 = tc_to_i915(tc); - if (wait_for(tc_phy_is_ready(tc), 100)) + if (wait_for(tc_phy_is_ready(tc), 500)) { drm_err(&i915->drm, "Port %s: timeout waiting for PHY ready\n", tc->port_name); + + return false; + } + + return true; } static enum tc_port_mode @@ -1335,6 +1577,138 @@ bool intel_tc_port_connected(struct intel_encoder *encoder) return is_connected; } +static bool __intel_tc_port_link_needs_reset(struct intel_tc_port *tc) +{ + bool ret; + + mutex_lock(&tc->lock); + + ret = tc->link_refcount && + tc->mode == TC_PORT_DP_ALT && + intel_tc_port_needs_reset(tc); + + mutex_unlock(&tc->lock); + + return ret; +} + +bool intel_tc_port_link_needs_reset(struct intel_digital_port *dig_port) +{ + struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev); + enum phy phy = intel_port_to_phy(i915, dig_port->base.port); + + if (!intel_phy_is_tc(i915, phy)) + return false; + + return __intel_tc_port_link_needs_reset(to_tc_port(dig_port)); +} + +static int reset_link_commit(struct intel_tc_port *tc, + struct intel_atomic_state *state, + struct drm_modeset_acquire_ctx *ctx) +{ + struct drm_i915_private *i915 = tc_to_i915(tc); + struct intel_digital_port *dig_port = tc->dig_port; + struct intel_dp *intel_dp = enc_to_intel_dp(&dig_port->base); + struct intel_crtc *crtc; + u8 pipe_mask; + int ret; + + ret = drm_modeset_lock(&i915->drm.mode_config.connection_mutex, ctx); + if (ret) + return ret; + + ret = intel_dp_get_active_pipes(intel_dp, ctx, &pipe_mask); + if (ret) + return ret; + + if (!pipe_mask) + return 0; + + for_each_intel_crtc_in_pipe_mask(&i915->drm, crtc, pipe_mask) { + struct intel_crtc_state *crtc_state; + + crtc_state = intel_atomic_get_crtc_state(&state->base, crtc); + if (IS_ERR(crtc_state)) + return PTR_ERR(crtc_state); + + crtc_state->uapi.connectors_changed = true; + } + + if (!__intel_tc_port_link_needs_reset(tc)) + return 0; + + return drm_atomic_commit(&state->base); +} + +static int reset_link(struct intel_tc_port *tc) +{ + struct drm_i915_private *i915 = tc_to_i915(tc); + struct drm_modeset_acquire_ctx ctx; + struct drm_atomic_state *_state; + struct intel_atomic_state *state; + int ret; + + _state = drm_atomic_state_alloc(&i915->drm); + if (!_state) + return -ENOMEM; + + state = to_intel_atomic_state(_state); + state->internal = true; + + intel_modeset_lock_ctx_retry(&ctx, state, 0, ret) + ret = reset_link_commit(tc, state, &ctx); + + drm_atomic_state_put(&state->base); + + return ret; +} + +static void intel_tc_port_link_reset_work(struct work_struct *work) +{ + struct intel_tc_port *tc = + container_of(work, struct intel_tc_port, link_reset_work.work); + struct drm_i915_private *i915 = tc_to_i915(tc); + int ret; + + if (!__intel_tc_port_link_needs_reset(tc)) + return; + + mutex_lock(&i915->drm.mode_config.mutex); + + drm_dbg_kms(&i915->drm, + "Port %s: TypeC DP-alt sink disconnected, resetting link\n", + tc->port_name); + ret = reset_link(tc); + drm_WARN_ON(&i915->drm, ret); + + mutex_unlock(&i915->drm.mode_config.mutex); +} + +bool intel_tc_port_link_reset(struct intel_digital_port *dig_port) +{ + if (!intel_tc_port_link_needs_reset(dig_port)) + return false; + + queue_delayed_work(system_unbound_wq, + &to_tc_port(dig_port)->link_reset_work, + msecs_to_jiffies(2000)); + + return true; +} + +void intel_tc_port_link_cancel_reset_work(struct intel_digital_port *dig_port) +{ + struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev); + enum phy phy = intel_port_to_phy(i915, dig_port->base.port); + struct intel_tc_port *tc = to_tc_port(dig_port); + + if (!intel_phy_is_tc(i915, phy)) + return; + + cancel_delayed_work(&tc->link_reset_work); +} + static void __intel_tc_port_lock(struct intel_tc_port *tc, int required_lanes) { @@ -1382,11 +1756,19 @@ static void intel_tc_port_disconnect_phy_work(struct work_struct *work) * * Flush the delayed work disconnecting an idle PHY. */ -void intel_tc_port_flush_work(struct intel_digital_port *dig_port) +static void intel_tc_port_flush_work(struct intel_digital_port *dig_port) { flush_delayed_work(&to_tc_port(dig_port)->disconnect_phy_work); } +void intel_tc_port_suspend(struct intel_digital_port *dig_port) +{ + struct intel_tc_port *tc = to_tc_port(dig_port); + + cancel_delayed_work_sync(&tc->link_reset_work); + intel_tc_port_flush_work(dig_port); +} + void intel_tc_port_unlock(struct intel_digital_port *dig_port) { struct intel_tc_port *tc = to_tc_port(dig_port); @@ -1423,6 +1805,14 @@ void intel_tc_port_put_link(struct intel_digital_port *dig_port) intel_tc_port_lock(dig_port); __intel_tc_port_put_link(tc); intel_tc_port_unlock(dig_port); + + /* + * The firmware will not update the HPD status of other TypeC ports + * that are active in DP-alt mode with their sink disconnected, until + * this port is disabled and its PHY gets disconnected. Make sure this + * happens in a timely manner by disconnecting the PHY synchronously. + */ + intel_tc_port_flush_work(dig_port); } int intel_tc_port_init(struct intel_digital_port *dig_port, bool is_legacy) @@ -1442,7 +1832,9 @@ int intel_tc_port_init(struct intel_digital_port *dig_port, bool is_legacy) dig_port->tc = tc; tc->dig_port = dig_port; - if (DISPLAY_VER(i915) >= 13) + if (DISPLAY_VER(i915) >= 14) + tc->phy_ops = &xelpdp_tc_phy_ops; + else if (DISPLAY_VER(i915) >= 13) tc->phy_ops = &adlp_tc_phy_ops; else if (DISPLAY_VER(i915) >= 12) tc->phy_ops = &tgl_tc_phy_ops; @@ -1453,7 +1845,9 @@ int intel_tc_port_init(struct intel_digital_port *dig_port, bool is_legacy) "%c/TC#%d", port_name(port), tc_port + 1); mutex_init(&tc->lock); + /* TODO: Combine the two works */ INIT_DELAYED_WORK(&tc->disconnect_phy_work, intel_tc_port_disconnect_phy_work); + INIT_DELAYED_WORK(&tc->link_reset_work, intel_tc_port_link_reset_work); tc->legacy_port = is_legacy; tc->mode = TC_PORT_DISCONNECTED; tc->link_refcount = 0; @@ -1467,7 +1861,7 @@ int intel_tc_port_init(struct intel_digital_port *dig_port, bool is_legacy) void intel_tc_port_cleanup(struct intel_digital_port *dig_port) { - intel_tc_port_flush_work(dig_port); + intel_tc_port_suspend(dig_port); kfree(dig_port->tc); dig_port->tc = NULL; diff --git a/drivers/gpu/drm/i915/display/intel_tc.h b/drivers/gpu/drm/i915/display/intel_tc.h index dd0810f9ea95e7dfdf37a69ad5c229657dfa3e03..3b16491925fa9242f505ee349e49e27eb9a53fb7 100644 --- a/drivers/gpu/drm/i915/display/intel_tc.h +++ b/drivers/gpu/drm/i915/display/intel_tc.h @@ -30,11 +30,14 @@ void intel_tc_port_sanitize_mode(struct intel_digital_port *dig_port, const struct intel_crtc_state *crtc_state); void intel_tc_port_lock(struct intel_digital_port *dig_port); void intel_tc_port_unlock(struct intel_digital_port *dig_port); -void intel_tc_port_flush_work(struct intel_digital_port *dig_port); +void intel_tc_port_suspend(struct intel_digital_port *dig_port); void intel_tc_port_get_link(struct intel_digital_port *dig_port, int required_lanes); void intel_tc_port_put_link(struct intel_digital_port *dig_port); bool intel_tc_port_ref_held(struct intel_digital_port *dig_port); +bool intel_tc_port_link_needs_reset(struct intel_digital_port *dig_port); +bool intel_tc_port_link_reset(struct intel_digital_port *dig_port); +void intel_tc_port_link_cancel_reset_work(struct intel_digital_port *dig_port); int intel_tc_port_init(struct intel_digital_port *dig_port, bool is_legacy); void intel_tc_port_cleanup(struct intel_digital_port *dig_port); diff --git a/drivers/gpu/drm/i915/display/intel_tv.c b/drivers/gpu/drm/i915/display/intel_tv.c index 557ec5b62afa25c176f8c55a37e30d08a74ea4cd..36b479b46b6004992bad22c3f3423e506f73424d 100644 --- a/drivers/gpu/drm/i915/display/intel_tv.c +++ b/drivers/gpu/drm/i915/display/intel_tv.c @@ -35,14 +35,15 @@ #include #include "i915_drv.h" -#include "i915_irq.h" #include "i915_reg.h" #include "intel_connector.h" #include "intel_crtc.h" #include "intel_de.h" +#include "intel_display_irq.h" #include "intel_display_types.h" #include "intel_dpll.h" #include "intel_hotplug.h" +#include "intel_load_detect.h" #include "intel_tv.h" #include "intel_tv_regs.h" @@ -1205,6 +1206,7 @@ intel_tv_compute_config(struct intel_encoder *encoder, if (adjusted_mode->flags & DRM_MODE_FLAG_DBLSCAN) return -EINVAL; + pipe_config->sink_format = INTEL_OUTPUT_FORMAT_RGB; pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB; drm_dbg_kms(&dev_priv->drm, "forcing bpc to 8 for TV\n"); @@ -1722,21 +1724,21 @@ intel_tv_detect(struct drm_connector *connector, return connector_status_disconnected; if (force) { - struct intel_load_detect_pipe tmp; - int ret; + struct drm_atomic_state *state; - ret = intel_get_load_detect_pipe(connector, &tmp, ctx); - if (ret < 0) - return ret; + state = intel_load_detect_get_pipe(connector, ctx); + if (IS_ERR(state)) + return PTR_ERR(state); - if (ret > 0) { + if (state) { type = intel_tv_detect_type(intel_tv, connector); - intel_release_load_detect_pipe(connector, &tmp, ctx); + intel_load_detect_release_pipe(connector, state, ctx); status = type < 0 ? connector_status_disconnected : connector_status_connected; - } else + } else { status = connector_status_unknown; + } if (status == connector_status_connected) { intel_tv->type = type; diff --git a/drivers/gpu/drm/i915/display/intel_vblank.c b/drivers/gpu/drm/i915/display/intel_vblank.c index f8bf9810527de8a1e10d542c8f6459233cc8807b..f5659ebd08eb291eeb705383bd8d115a8acfd623 100644 --- a/drivers/gpu/drm/i915/display/intel_vblank.c +++ b/drivers/gpu/drm/i915/display/intel_vblank.c @@ -340,8 +340,7 @@ static bool i915_get_crtc_scanoutpos(struct drm_crtc *_crtc, * matches how the scanline counter based position works since * the scanline counter doesn't count the two half lines. */ - if (position >= vtotal) - position = vtotal - 1; + position = min(position, vtotal - 1); /* * Start of vblank interrupt is triggered at start of hsync, @@ -488,21 +487,27 @@ static int intel_crtc_scanline_offset(const struct intel_crtc_state *crtc_state) } } -void intel_crtc_update_active_timings(const struct intel_crtc_state *crtc_state) +void intel_crtc_update_active_timings(const struct intel_crtc_state *crtc_state, + bool vrr_enable) { struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct drm_i915_private *i915 = to_i915(crtc->base.dev); + u8 mode_flags = crtc_state->mode_flags; struct drm_display_mode adjusted_mode; int vmax_vblank_start = 0; unsigned long irqflags; drm_mode_init(&adjusted_mode, &crtc_state->hw.adjusted_mode); - if (crtc_state->vrr.enable) { + if (vrr_enable) { + drm_WARN_ON(&i915->drm, (mode_flags & I915_MODE_FLAG_VRR) == 0); + adjusted_mode.crtc_vtotal = crtc_state->vrr.vmax; adjusted_mode.crtc_vblank_end = crtc_state->vrr.vmax; adjusted_mode.crtc_vblank_start = intel_vrr_vmin_vblank_start(crtc_state); vmax_vblank_start = intel_vrr_vmax_vblank_start(crtc_state); + } else { + mode_flags &= ~I915_MODE_FLAG_VRR; } /* @@ -524,7 +529,7 @@ void intel_crtc_update_active_timings(const struct intel_crtc_state *crtc_state) crtc->vmax_vblank_start = vmax_vblank_start; - crtc->mode_flags = crtc_state->mode_flags; + crtc->mode_flags = mode_flags; crtc->scanline_offset = intel_crtc_scanline_offset(crtc_state); diff --git a/drivers/gpu/drm/i915/display/intel_vblank.h b/drivers/gpu/drm/i915/display/intel_vblank.h index 0884db7e76ae04966151205b3fb4153a8100f4eb..08e706b29149518ed122ec644a2ebb2196f11dc2 100644 --- a/drivers/gpu/drm/i915/display/intel_vblank.h +++ b/drivers/gpu/drm/i915/display/intel_vblank.h @@ -20,6 +20,7 @@ bool intel_crtc_get_vblank_timestamp(struct drm_crtc *crtc, int *max_error, int intel_get_crtc_scanline(struct intel_crtc *crtc); void intel_wait_for_pipe_scanline_stopped(struct intel_crtc *crtc); void intel_wait_for_pipe_scanline_moving(struct intel_crtc *crtc); -void intel_crtc_update_active_timings(const struct intel_crtc_state *crtc_state); +void intel_crtc_update_active_timings(const struct intel_crtc_state *crtc_state, + bool vrr_enable); #endif /* __INTEL_VBLANK_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_vdsc.c b/drivers/gpu/drm/i915/display/intel_vdsc.c index 8e787c13d26d06a1bee123ff8c4ba5f25b5a6cb6..bd9116d2cd76cf2f6b2b35b6dd2a21834cc59f69 100644 --- a/drivers/gpu/drm/i915/display/intel_vdsc.c +++ b/drivers/gpu/drm/i915/display/intel_vdsc.c @@ -19,327 +19,6 @@ #include "intel_vdsc.h" #include "intel_vdsc_regs.h" -enum ROW_INDEX_BPP { - ROW_INDEX_6BPP = 0, - ROW_INDEX_8BPP, - ROW_INDEX_10BPP, - ROW_INDEX_12BPP, - ROW_INDEX_15BPP, - MAX_ROW_INDEX -}; - -enum COLUMN_INDEX_BPC { - COLUMN_INDEX_8BPC = 0, - COLUMN_INDEX_10BPC, - COLUMN_INDEX_12BPC, - COLUMN_INDEX_14BPC, - COLUMN_INDEX_16BPC, - MAX_COLUMN_INDEX -}; - -/* From DSC_v1.11 spec, rc_parameter_Set syntax element typically constant */ -static const u16 rc_buf_thresh[] = { - 896, 1792, 2688, 3584, 4480, 5376, 6272, 6720, 7168, 7616, - 7744, 7872, 8000, 8064 -}; - -struct rc_parameters { - u16 initial_xmit_delay; - u8 first_line_bpg_offset; - u16 initial_offset; - u8 flatness_min_qp; - u8 flatness_max_qp; - u8 rc_quant_incr_limit0; - u8 rc_quant_incr_limit1; - struct drm_dsc_rc_range_parameters rc_range_params[DSC_NUM_BUF_RANGES]; -}; - -/* - * Selected Rate Control Related Parameter Recommended Values - * from DSC_v1.11 spec & C Model release: DSC_model_20161212 - */ -static const struct rc_parameters rc_parameters[][MAX_COLUMN_INDEX] = { -{ - /* 6BPP/8BPC */ - { 768, 15, 6144, 3, 13, 11, 11, { - { 0, 4, 0 }, { 1, 6, -2 }, { 3, 8, -2 }, { 4, 8, -4 }, - { 5, 9, -6 }, { 5, 9, -6 }, { 6, 9, -6 }, { 6, 10, -8 }, - { 7, 11, -8 }, { 8, 12, -10 }, { 9, 12, -10 }, { 10, 12, -12 }, - { 10, 12, -12 }, { 11, 12, -12 }, { 13, 14, -12 } - } - }, - /* 6BPP/10BPC */ - { 768, 15, 6144, 7, 17, 15, 15, { - { 0, 8, 0 }, { 3, 10, -2 }, { 7, 12, -2 }, { 8, 12, -4 }, - { 9, 13, -6 }, { 9, 13, -6 }, { 10, 13, -6 }, { 10, 14, -8 }, - { 11, 15, -8 }, { 12, 16, -10 }, { 13, 16, -10 }, - { 14, 16, -12 }, { 14, 16, -12 }, { 15, 16, -12 }, - { 17, 18, -12 } - } - }, - /* 6BPP/12BPC */ - { 768, 15, 6144, 11, 21, 19, 19, { - { 0, 12, 0 }, { 5, 14, -2 }, { 11, 16, -2 }, { 12, 16, -4 }, - { 13, 17, -6 }, { 13, 17, -6 }, { 14, 17, -6 }, { 14, 18, -8 }, - { 15, 19, -8 }, { 16, 20, -10 }, { 17, 20, -10 }, - { 18, 20, -12 }, { 18, 20, -12 }, { 19, 20, -12 }, - { 21, 22, -12 } - } - }, - /* 6BPP/14BPC */ - { 768, 15, 6144, 15, 25, 23, 27, { - { 0, 16, 0 }, { 7, 18, -2 }, { 15, 20, -2 }, { 16, 20, -4 }, - { 17, 21, -6 }, { 17, 21, -6 }, { 18, 21, -6 }, { 18, 22, -8 }, - { 19, 23, -8 }, { 20, 24, -10 }, { 21, 24, -10 }, - { 22, 24, -12 }, { 22, 24, -12 }, { 23, 24, -12 }, - { 25, 26, -12 } - } - }, - /* 6BPP/16BPC */ - { 768, 15, 6144, 19, 29, 27, 27, { - { 0, 20, 0 }, { 9, 22, -2 }, { 19, 24, -2 }, { 20, 24, -4 }, - { 21, 25, -6 }, { 21, 25, -6 }, { 22, 25, -6 }, { 22, 26, -8 }, - { 23, 27, -8 }, { 24, 28, -10 }, { 25, 28, -10 }, - { 26, 28, -12 }, { 26, 28, -12 }, { 27, 28, -12 }, - { 29, 30, -12 } - } - }, -}, -{ - /* 8BPP/8BPC */ - { 512, 12, 6144, 3, 12, 11, 11, { - { 0, 4, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 }, - { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, - { 3, 9, -8 }, { 3, 10, -10 }, { 5, 11, -10 }, { 5, 12, -12 }, - { 5, 13, -12 }, { 7, 13, -12 }, { 13, 15, -12 } - } - }, - /* 8BPP/10BPC */ - { 512, 12, 6144, 7, 16, 15, 15, { - { 0, 4, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 5, 10, -2 }, - { 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, - { 7, 13, -8 }, { 7, 14, -10 }, { 9, 15, -10 }, { 9, 16, -12 }, - { 9, 17, -12 }, { 11, 17, -12 }, { 17, 19, -12 } - } - }, - /* 8BPP/12BPC */ - { 512, 12, 6144, 11, 20, 19, 19, { - { 0, 12, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 9, 14, -2 }, - { 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, - { 11, 17, -8 }, { 11, 18, -10 }, { 13, 19, -10 }, - { 13, 20, -12 }, { 13, 21, -12 }, { 15, 21, -12 }, - { 21, 23, -12 } - } - }, - /* 8BPP/14BPC */ - { 512, 12, 6144, 15, 24, 23, 23, { - { 0, 12, 0 }, { 5, 13, 0 }, { 11, 15, 0 }, { 12, 17, -2 }, - { 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 }, - { 15, 21, -8 }, { 15, 22, -10 }, { 17, 22, -10 }, - { 17, 23, -12 }, { 17, 23, -12 }, { 21, 24, -12 }, - { 24, 25, -12 } - } - }, - /* 8BPP/16BPC */ - { 512, 12, 6144, 19, 28, 27, 27, { - { 0, 12, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 15, 20, -2 }, - { 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 }, - { 19, 25, -8 }, { 19, 26, -10 }, { 21, 26, -10 }, - { 21, 27, -12 }, { 21, 27, -12 }, { 25, 28, -12 }, - { 28, 29, -12 } - } - }, -}, -{ - /* 10BPP/8BPC */ - { 410, 15, 5632, 3, 12, 11, 11, { - { 0, 3, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 2, 6, -2 }, - { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, - { 3, 9, -8 }, { 3, 9, -10 }, { 5, 10, -10 }, { 5, 10, -10 }, - { 5, 11, -12 }, { 7, 11, -12 }, { 11, 12, -12 } - } - }, - /* 10BPP/10BPC */ - { 410, 15, 5632, 7, 16, 15, 15, { - { 0, 7, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 6, 10, -2 }, - { 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, - { 7, 13, -8 }, { 7, 13, -10 }, { 9, 14, -10 }, { 9, 14, -10 }, - { 9, 15, -12 }, { 11, 15, -12 }, { 15, 16, -12 } - } - }, - /* 10BPP/12BPC */ - { 410, 15, 5632, 11, 20, 19, 19, { - { 0, 11, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 10, 14, -2 }, - { 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, - { 11, 17, -8 }, { 11, 17, -10 }, { 13, 18, -10 }, - { 13, 18, -10 }, { 13, 19, -12 }, { 15, 19, -12 }, - { 19, 20, -12 } - } - }, - /* 10BPP/14BPC */ - { 410, 15, 5632, 15, 24, 23, 23, { - { 0, 11, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 13, 18, -2 }, - { 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 }, - { 15, 21, -8 }, { 15, 21, -10 }, { 17, 22, -10 }, - { 17, 22, -10 }, { 17, 23, -12 }, { 19, 23, -12 }, - { 23, 24, -12 } - } - }, - /* 10BPP/16BPC */ - { 410, 15, 5632, 19, 28, 27, 27, { - { 0, 11, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 16, 20, -2 }, - { 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 }, - { 19, 25, -8 }, { 19, 25, -10 }, { 21, 26, -10 }, - { 21, 26, -10 }, { 21, 27, -12 }, { 23, 27, -12 }, - { 27, 28, -12 } - } - }, -}, -{ - /* 12BPP/8BPC */ - { 341, 15, 2048, 3, 12, 11, 11, { - { 0, 2, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 }, - { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, - { 3, 9, -8 }, { 3, 10, -10 }, { 5, 11, -10 }, - { 5, 12, -12 }, { 5, 13, -12 }, { 7, 13, -12 }, { 13, 15, -12 } - } - }, - /* 12BPP/10BPC */ - { 341, 15, 2048, 7, 16, 15, 15, { - { 0, 2, 2 }, { 2, 5, 0 }, { 3, 7, 0 }, { 4, 8, -2 }, - { 6, 9, -4 }, { 7, 10, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, - { 7, 13, -8 }, { 7, 14, -10 }, { 9, 15, -10 }, { 9, 16, -12 }, - { 9, 17, -12 }, { 11, 17, -12 }, { 17, 19, -12 } - } - }, - /* 12BPP/12BPC */ - { 341, 15, 2048, 11, 20, 19, 19, { - { 0, 6, 2 }, { 4, 9, 0 }, { 7, 11, 0 }, { 8, 12, -2 }, - { 10, 13, -4 }, { 11, 14, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, - { 11, 17, -8 }, { 11, 18, -10 }, { 13, 19, -10 }, - { 13, 20, -12 }, { 13, 21, -12 }, { 15, 21, -12 }, - { 21, 23, -12 } - } - }, - /* 12BPP/14BPC */ - { 341, 15, 2048, 15, 24, 23, 23, { - { 0, 6, 2 }, { 7, 10, 0 }, { 9, 13, 0 }, { 11, 16, -2 }, - { 14, 17, -4 }, { 15, 18, -6 }, { 15, 19, -8 }, { 15, 20, -8 }, - { 15, 20, -8 }, { 15, 21, -10 }, { 17, 21, -10 }, - { 17, 21, -12 }, { 17, 21, -12 }, { 19, 22, -12 }, - { 22, 23, -12 } - } - }, - /* 12BPP/16BPC */ - { 341, 15, 2048, 19, 28, 27, 27, { - { 0, 6, 2 }, { 6, 11, 0 }, { 11, 15, 0 }, { 14, 18, -2 }, - { 18, 21, -4 }, { 19, 22, -6 }, { 19, 23, -8 }, { 19, 24, -8 }, - { 19, 24, -8 }, { 19, 25, -10 }, { 21, 25, -10 }, - { 21, 25, -12 }, { 21, 25, -12 }, { 23, 26, -12 }, - { 26, 27, -12 } - } - }, -}, -{ - /* 15BPP/8BPC */ - { 273, 15, 2048, 3, 12, 11, 11, { - { 0, 0, 10 }, { 0, 1, 8 }, { 0, 1, 6 }, { 0, 2, 4 }, - { 1, 2, 2 }, { 1, 3, 0 }, { 1, 3, -2 }, { 2, 4, -4 }, - { 2, 5, -6 }, { 3, 5, -8 }, { 4, 6, -10 }, { 4, 7, -10 }, - { 5, 7, -12 }, { 7, 8, -12 }, { 8, 9, -12 } - } - }, - /* 15BPP/10BPC */ - { 273, 15, 2048, 7, 16, 15, 15, { - { 0, 2, 10 }, { 2, 5, 8 }, { 3, 5, 6 }, { 4, 6, 4 }, - { 5, 6, 2 }, { 5, 7, 0 }, { 5, 7, -2 }, { 6, 8, -4 }, - { 6, 9, -6 }, { 7, 9, -8 }, { 8, 10, -10 }, { 8, 11, -10 }, - { 9, 11, -12 }, { 11, 12, -12 }, { 12, 13, -12 } - } - }, - /* 15BPP/12BPC */ - { 273, 15, 2048, 11, 20, 19, 19, { - { 0, 4, 10 }, { 2, 7, 8 }, { 4, 9, 6 }, { 6, 11, 4 }, - { 9, 11, 2 }, { 9, 11, 0 }, { 9, 12, -2 }, { 10, 12, -4 }, - { 11, 13, -6 }, { 11, 13, -8 }, { 12, 14, -10 }, - { 13, 15, -10 }, { 13, 15, -12 }, { 15, 16, -12 }, - { 16, 17, -12 } - } - }, - /* 15BPP/14BPC */ - { 273, 15, 2048, 15, 24, 23, 23, { - { 0, 4, 10 }, { 3, 8, 8 }, { 6, 11, 6 }, { 9, 14, 4 }, - { 13, 15, 2 }, { 13, 15, 0 }, { 13, 16, -2 }, { 14, 16, -4 }, - { 15, 17, -6 }, { 15, 17, -8 }, { 16, 18, -10 }, - { 17, 19, -10 }, { 17, 19, -12 }, { 19, 20, -12 }, - { 20, 21, -12 } - } - }, - /* 15BPP/16BPC */ - { 273, 15, 2048, 19, 28, 27, 27, { - { 0, 4, 10 }, { 4, 9, 8 }, { 8, 13, 6 }, { 12, 17, 4 }, - { 17, 19, 2 }, { 17, 20, 0 }, { 17, 20, -2 }, { 18, 20, -4 }, - { 19, 21, -6 }, { 19, 21, -8 }, { 20, 22, -10 }, - { 21, 23, -10 }, { 21, 23, -12 }, { 23, 24, -12 }, - { 24, 25, -12 } - } - } -} - -}; - -static int get_row_index_for_rc_params(u16 compressed_bpp) -{ - switch (compressed_bpp) { - case 6: - return ROW_INDEX_6BPP; - case 8: - return ROW_INDEX_8BPP; - case 10: - return ROW_INDEX_10BPP; - case 12: - return ROW_INDEX_12BPP; - case 15: - return ROW_INDEX_15BPP; - default: - return -EINVAL; - } -} - -static int get_column_index_for_rc_params(u8 bits_per_component) -{ - switch (bits_per_component) { - case 8: - return COLUMN_INDEX_8BPC; - case 10: - return COLUMN_INDEX_10BPC; - case 12: - return COLUMN_INDEX_12BPC; - case 14: - return COLUMN_INDEX_14BPC; - case 16: - return COLUMN_INDEX_16BPC; - default: - return -EINVAL; - } -} - -static const struct rc_parameters *get_rc_params(u16 compressed_bpp, - u8 bits_per_component) -{ - int row_index, column_index; - - row_index = get_row_index_for_rc_params(compressed_bpp); - if (row_index < 0) - return NULL; - - column_index = get_column_index_for_rc_params(bits_per_component); - if (column_index < 0) - return NULL; - - return &rc_parameters[row_index][column_index]; -} - bool intel_dsc_source_support(const struct intel_crtc_state *crtc_state) { const struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); @@ -374,8 +53,7 @@ static bool is_pipe_dsc(struct intel_crtc *crtc, enum transcoder cpu_transcoder) } static void -calculate_rc_params(struct rc_parameters *rc, - struct drm_dsc_config *vdsc_cfg) +calculate_rc_params(struct drm_dsc_config *vdsc_cfg) { int bpc = vdsc_cfg->bits_per_component; int bpp = vdsc_cfg->bits_per_pixel >> 4; @@ -395,56 +73,57 @@ calculate_rc_params(struct rc_parameters *rc, u32 res, buf_i, bpp_i; if (vdsc_cfg->slice_height >= 8) - rc->first_line_bpg_offset = + vdsc_cfg->first_line_bpg_offset = 12 + DIV_ROUND_UP((9 * min(34, vdsc_cfg->slice_height - 8)), 100); else - rc->first_line_bpg_offset = 2 * (vdsc_cfg->slice_height - 1); + vdsc_cfg->first_line_bpg_offset = 2 * (vdsc_cfg->slice_height - 1); /* Our hw supports only 444 modes as of today */ if (bpp >= 12) - rc->initial_offset = 2048; + vdsc_cfg->initial_offset = 2048; else if (bpp >= 10) - rc->initial_offset = 5632 - DIV_ROUND_UP(((bpp - 10) * 3584), 2); + vdsc_cfg->initial_offset = 5632 - DIV_ROUND_UP(((bpp - 10) * 3584), 2); else if (bpp >= 8) - rc->initial_offset = 6144 - DIV_ROUND_UP(((bpp - 8) * 512), 2); + vdsc_cfg->initial_offset = 6144 - DIV_ROUND_UP(((bpp - 8) * 512), 2); else - rc->initial_offset = 6144; + vdsc_cfg->initial_offset = 6144; /* initial_xmit_delay = rc_model_size/2/compression_bpp */ - rc->initial_xmit_delay = DIV_ROUND_UP(DSC_RC_MODEL_SIZE_CONST, 2 * bpp); + vdsc_cfg->initial_xmit_delay = DIV_ROUND_UP(DSC_RC_MODEL_SIZE_CONST, 2 * bpp); - rc->flatness_min_qp = 3 + qp_bpc_modifier; - rc->flatness_max_qp = 12 + qp_bpc_modifier; + vdsc_cfg->flatness_min_qp = 3 + qp_bpc_modifier; + vdsc_cfg->flatness_max_qp = 12 + qp_bpc_modifier; - rc->rc_quant_incr_limit0 = 11 + qp_bpc_modifier; - rc->rc_quant_incr_limit1 = 11 + qp_bpc_modifier; + vdsc_cfg->rc_quant_incr_limit0 = 11 + qp_bpc_modifier; + vdsc_cfg->rc_quant_incr_limit1 = 11 + qp_bpc_modifier; bpp_i = (2 * (bpp - 6)); for (buf_i = 0; buf_i < DSC_NUM_BUF_RANGES; buf_i++) { + u8 range_bpg_offset; + /* Read range_minqp and range_max_qp from qp tables */ - rc->rc_range_params[buf_i].range_min_qp = + vdsc_cfg->rc_range_params[buf_i].range_min_qp = intel_lookup_range_min_qp(bpc, buf_i, bpp_i, vdsc_cfg->native_420); - rc->rc_range_params[buf_i].range_max_qp = + vdsc_cfg->rc_range_params[buf_i].range_max_qp = intel_lookup_range_max_qp(bpc, buf_i, bpp_i, vdsc_cfg->native_420); - /* Calculate range_bgp_offset */ + /* Calculate range_bpg_offset */ if (bpp <= 6) { - rc->rc_range_params[buf_i].range_bpg_offset = ofs_und6[buf_i]; + range_bpg_offset = ofs_und6[buf_i]; } else if (bpp <= 8) { res = DIV_ROUND_UP(((bpp - 6) * (ofs_und8[buf_i] - ofs_und6[buf_i])), 2); - rc->rc_range_params[buf_i].range_bpg_offset = - ofs_und6[buf_i] + res; + range_bpg_offset = ofs_und6[buf_i] + res; } else if (bpp <= 12) { - rc->rc_range_params[buf_i].range_bpg_offset = - ofs_und8[buf_i]; + range_bpg_offset = ofs_und8[buf_i]; } else if (bpp <= 15) { res = DIV_ROUND_UP(((bpp - 12) * (ofs_und15[buf_i] - ofs_und12[buf_i])), 3); - rc->rc_range_params[buf_i].range_bpg_offset = - ofs_und12[buf_i] + res; + range_bpg_offset = ofs_und12[buf_i] + res; } else { - rc->rc_range_params[buf_i].range_bpg_offset = - ofs_und15[buf_i]; + range_bpg_offset = ofs_und15[buf_i]; } + + vdsc_cfg->rc_range_params[buf_i].range_bpg_offset = + range_bpg_offset & DSC_RANGE_BPG_OFFSET_MASK; } } @@ -477,10 +156,8 @@ int intel_dsc_compute_params(struct intel_crtc_state *pipe_config) struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); struct drm_dsc_config *vdsc_cfg = &pipe_config->dsc.config; u16 compressed_bpp = pipe_config->dsc.compressed_bpp; - const struct rc_parameters *rc_params; - struct rc_parameters *rc = NULL; int err; - u8 i = 0; + int ret; vdsc_cfg->pic_width = pipe_config->hw.adjusted_mode.crtc_hdisplay; vdsc_cfg->slice_width = DIV_ROUND_UP(vdsc_cfg->pic_width, @@ -539,23 +216,7 @@ int intel_dsc_compute_params(struct intel_crtc_state *pipe_config) vdsc_cfg->bits_per_component = pipe_config->pipe_bpp / 3; - for (i = 0; i < DSC_NUM_BUF_RANGES - 1; i++) { - /* - * six 0s are appended to the lsb of each threshold value - * internally in h/w. - * Only 8 bits are allowed for programming RcBufThreshold - */ - vdsc_cfg->rc_buf_thresh[i] = rc_buf_thresh[i] >> 6; - } - - /* - * For 6bpp, RC Buffer threshold 12 and 13 need a different value - * as per C Model - */ - if (compressed_bpp == 6) { - vdsc_cfg->rc_buf_thresh[12] = 0x7C; - vdsc_cfg->rc_buf_thresh[13] = 0x7D; - } + drm_dsc_set_rc_buf_thresh(vdsc_cfg); /* * From XE_LPD onwards we supports compression bpps in steps of 1 @@ -563,39 +224,31 @@ int intel_dsc_compute_params(struct intel_crtc_state *pipe_config) * parameters */ if (DISPLAY_VER(dev_priv) >= 13) { - rc = kmalloc(sizeof(*rc), GFP_KERNEL); - if (!rc) - return -ENOMEM; - - calculate_rc_params(rc, vdsc_cfg); - rc_params = rc; + calculate_rc_params(vdsc_cfg); } else { - rc_params = get_rc_params(compressed_bpp, - vdsc_cfg->bits_per_component); - if (!rc_params) - return -EINVAL; - } + if ((compressed_bpp == 8 || + compressed_bpp == 12) && + (vdsc_cfg->bits_per_component == 8 || + vdsc_cfg->bits_per_component == 10 || + vdsc_cfg->bits_per_component == 12)) + ret = drm_dsc_setup_rc_params(vdsc_cfg, DRM_DSC_1_1_PRE_SCR); + else + ret = drm_dsc_setup_rc_params(vdsc_cfg, DRM_DSC_1_2_444); - vdsc_cfg->first_line_bpg_offset = rc_params->first_line_bpg_offset; - vdsc_cfg->initial_xmit_delay = rc_params->initial_xmit_delay; - vdsc_cfg->initial_offset = rc_params->initial_offset; - vdsc_cfg->flatness_min_qp = rc_params->flatness_min_qp; - vdsc_cfg->flatness_max_qp = rc_params->flatness_max_qp; - vdsc_cfg->rc_quant_incr_limit0 = rc_params->rc_quant_incr_limit0; - vdsc_cfg->rc_quant_incr_limit1 = rc_params->rc_quant_incr_limit1; + if (ret) + return ret; - for (i = 0; i < DSC_NUM_BUF_RANGES; i++) { - vdsc_cfg->rc_range_params[i].range_min_qp = - rc_params->rc_range_params[i].range_min_qp; - vdsc_cfg->rc_range_params[i].range_max_qp = - rc_params->rc_range_params[i].range_max_qp; /* - * Range BPG Offset uses 2's complement and is only a 6 bits. So - * mask it to get only 6 bits. + * FIXME: verify that the hardware actually needs these + * modifications rather than them being simple typos. */ - vdsc_cfg->rc_range_params[i].range_bpg_offset = - rc_params->rc_range_params[i].range_bpg_offset & - DSC_RANGE_BPG_OFFSET_MASK; + if (compressed_bpp == 6 && + vdsc_cfg->bits_per_component == 8) + vdsc_cfg->rc_quant_incr_limit1 = 23; + + if (compressed_bpp == 8 && + vdsc_cfg->bits_per_component == 14) + vdsc_cfg->rc_range_params[0].range_bpg_offset = 0; } /* @@ -612,8 +265,6 @@ int intel_dsc_compute_params(struct intel_crtc_state *pipe_config) vdsc_cfg->initial_scale_value = (vdsc_cfg->rc_model_size << 3) / (vdsc_cfg->rc_model_size - vdsc_cfg->initial_offset); - kfree(rc); - return 0; } diff --git a/drivers/gpu/drm/i915/display/intel_vrr.c b/drivers/gpu/drm/i915/display/intel_vrr.c index 4228f26b4c11dd002e7eebecc9e9107f997abe96..88e4759b538b64b1ad3a572a7be933a1fda32864 100644 --- a/drivers/gpu/drm/i915/display/intel_vrr.c +++ b/drivers/gpu/drm/i915/display/intel_vrr.c @@ -114,9 +114,6 @@ intel_vrr_compute_config(struct intel_crtc_state *crtc_state, if (adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE) return; - if (!crtc_state->uapi.vrr_enabled) - return; - vmin = DIV_ROUND_UP(adjusted_mode->crtc_clock * 1000, adjusted_mode->crtc_htotal * info->monitor_range.max_vfreq); vmax = adjusted_mode->crtc_clock * 1000 / @@ -135,7 +132,6 @@ intel_vrr_compute_config(struct intel_crtc_state *crtc_state, */ crtc_state->vrr.vmin = vmin - 1; crtc_state->vrr.vmax = vmax; - crtc_state->vrr.enable = true; crtc_state->vrr.flipline = crtc_state->vrr.vmin + 1; @@ -152,7 +148,10 @@ intel_vrr_compute_config(struct intel_crtc_state *crtc_state, crtc_state->framestart_delay - 1); } - crtc_state->mode_flags |= I915_MODE_FLAG_VRR; + if (crtc_state->uapi.vrr_enabled) { + crtc_state->vrr.enable = true; + crtc_state->mode_flags |= I915_MODE_FLAG_VRR; + } } static u32 trans_vrr_ctl(const struct intel_crtc_state *crtc_state) @@ -168,23 +167,28 @@ static u32 trans_vrr_ctl(const struct intel_crtc_state *crtc_state) VRR_CTL_PIPELINE_FULL_OVERRIDE; } -void intel_vrr_enable(struct intel_encoder *encoder, - const struct intel_crtc_state *crtc_state) +void intel_vrr_set_transcoder_timings(const struct intel_crtc_state *crtc_state) { - struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev); enum transcoder cpu_transcoder = crtc_state->cpu_transcoder; - if (!crtc_state->vrr.enable) + /* + * TRANS_SET_CONTEXT_LATENCY with VRR enabled + * requires this chicken bit on ADL/DG2. + */ + if (DISPLAY_VER(dev_priv) == 13) + intel_de_rmw(dev_priv, CHICKEN_TRANS(cpu_transcoder), + 0, PIPE_VBLANK_WITH_DELAY); + + if (!crtc_state->vrr.flipline) { + intel_de_write(dev_priv, TRANS_VRR_CTL(cpu_transcoder), 0); return; + } intel_de_write(dev_priv, TRANS_VRR_VMIN(cpu_transcoder), crtc_state->vrr.vmin - 1); intel_de_write(dev_priv, TRANS_VRR_VMAX(cpu_transcoder), crtc_state->vrr.vmax - 1); intel_de_write(dev_priv, TRANS_VRR_CTL(cpu_transcoder), trans_vrr_ctl(crtc_state)); intel_de_write(dev_priv, TRANS_VRR_FLIPLINE(cpu_transcoder), crtc_state->vrr.flipline - 1); - intel_de_write(dev_priv, TRANS_PUSH(cpu_transcoder), TRANS_PUSH_EN); - - intel_de_write(dev_priv, TRANS_VRR_CTL(cpu_transcoder), - VRR_CTL_VRR_ENABLE | trans_vrr_ctl(crtc_state)); } void intel_vrr_send_push(const struct intel_crtc_state *crtc_state) @@ -212,6 +216,19 @@ bool intel_vrr_is_push_sent(const struct intel_crtc_state *crtc_state) return intel_de_read(dev_priv, TRANS_PUSH(cpu_transcoder)) & TRANS_PUSH_SEND; } +void intel_vrr_enable(const struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev); + enum transcoder cpu_transcoder = crtc_state->cpu_transcoder; + + if (!crtc_state->vrr.enable) + return; + + intel_de_write(dev_priv, TRANS_PUSH(cpu_transcoder), TRANS_PUSH_EN); + intel_de_write(dev_priv, TRANS_VRR_CTL(cpu_transcoder), + VRR_CTL_VRR_ENABLE | trans_vrr_ctl(crtc_state)); +} + void intel_vrr_disable(const struct intel_crtc_state *old_crtc_state) { struct intel_crtc *crtc = to_intel_crtc(old_crtc_state->uapi.crtc); @@ -225,22 +242,18 @@ void intel_vrr_disable(const struct intel_crtc_state *old_crtc_state) trans_vrr_ctl(old_crtc_state)); intel_de_wait_for_clear(dev_priv, TRANS_VRR_STATUS(cpu_transcoder), VRR_STATUS_VRR_EN_LIVE, 1000); - intel_de_write(dev_priv, TRANS_PUSH(cpu_transcoder), 0); - intel_de_write(dev_priv, TRANS_VRR_CTL(cpu_transcoder), 0); } -void intel_vrr_get_config(struct intel_crtc *crtc, - struct intel_crtc_state *crtc_state) +void intel_vrr_get_config(struct intel_crtc_state *crtc_state) { - struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev); enum transcoder cpu_transcoder = crtc_state->cpu_transcoder; u32 trans_vrr_ctl; trans_vrr_ctl = intel_de_read(dev_priv, TRANS_VRR_CTL(cpu_transcoder)); + crtc_state->vrr.enable = trans_vrr_ctl & VRR_CTL_VRR_ENABLE; - if (!crtc_state->vrr.enable) - return; if (DISPLAY_VER(dev_priv) >= 13) crtc_state->vrr.guardband = @@ -249,10 +262,13 @@ void intel_vrr_get_config(struct intel_crtc *crtc, if (trans_vrr_ctl & VRR_CTL_PIPELINE_FULL_OVERRIDE) crtc_state->vrr.pipeline_full = REG_FIELD_GET(VRR_CTL_PIPELINE_FULL_MASK, trans_vrr_ctl); - if (trans_vrr_ctl & VRR_CTL_FLIP_LINE_EN) + + if (trans_vrr_ctl & VRR_CTL_FLIP_LINE_EN) { crtc_state->vrr.flipline = intel_de_read(dev_priv, TRANS_VRR_FLIPLINE(cpu_transcoder)) + 1; - crtc_state->vrr.vmax = intel_de_read(dev_priv, TRANS_VRR_VMAX(cpu_transcoder)) + 1; - crtc_state->vrr.vmin = intel_de_read(dev_priv, TRANS_VRR_VMIN(cpu_transcoder)) + 1; + crtc_state->vrr.vmax = intel_de_read(dev_priv, TRANS_VRR_VMAX(cpu_transcoder)) + 1; + crtc_state->vrr.vmin = intel_de_read(dev_priv, TRANS_VRR_VMIN(cpu_transcoder)) + 1; + } - crtc_state->mode_flags |= I915_MODE_FLAG_VRR; + if (crtc_state->vrr.enable) + crtc_state->mode_flags |= I915_MODE_FLAG_VRR; } diff --git a/drivers/gpu/drm/i915/display/intel_vrr.h b/drivers/gpu/drm/i915/display/intel_vrr.h index 9fda1135b0dd53a1cd0cb5a712add6ce6b625e8c..de16960c4929f69eee618f71fb183a2bcf6e9894 100644 --- a/drivers/gpu/drm/i915/display/intel_vrr.h +++ b/drivers/gpu/drm/i915/display/intel_vrr.h @@ -11,22 +11,18 @@ struct drm_connector_state; struct intel_atomic_state; struct intel_connector; -struct intel_crtc; struct intel_crtc_state; -struct intel_dp; -struct intel_encoder; bool intel_vrr_is_capable(struct intel_connector *connector); void intel_vrr_check_modeset(struct intel_atomic_state *state); void intel_vrr_compute_config(struct intel_crtc_state *crtc_state, struct drm_connector_state *conn_state); -void intel_vrr_enable(struct intel_encoder *encoder, - const struct intel_crtc_state *crtc_state); +void intel_vrr_set_transcoder_timings(const struct intel_crtc_state *crtc_state); +void intel_vrr_enable(const struct intel_crtc_state *crtc_state); void intel_vrr_send_push(const struct intel_crtc_state *crtc_state); bool intel_vrr_is_push_sent(const struct intel_crtc_state *crtc_state); void intel_vrr_disable(const struct intel_crtc_state *old_crtc_state); -void intel_vrr_get_config(struct intel_crtc *crtc, - struct intel_crtc_state *crtc_state); +void intel_vrr_get_config(struct intel_crtc_state *crtc_state); int intel_vrr_vmax_vblank_start(const struct intel_crtc_state *crtc_state); int intel_vrr_vmin_vblank_start(const struct intel_crtc_state *crtc_state); diff --git a/drivers/gpu/drm/i915/display/skl_scaler.c b/drivers/gpu/drm/i915/display/skl_scaler.c index 0e7e014fcc717c502795cd1b78eb3bd6ccb3cdfd..1e7c97243fcf558611b338ae50d81a33bc552954 100644 --- a/drivers/gpu/drm/i915/display/skl_scaler.c +++ b/drivers/gpu/drm/i915/display/skl_scaler.c @@ -348,6 +348,263 @@ int skl_update_scaler_plane(struct intel_crtc_state *crtc_state, return 0; } +static int intel_atomic_setup_scaler(struct intel_crtc_scaler_state *scaler_state, + int num_scalers_need, struct intel_crtc *intel_crtc, + const char *name, int idx, + struct intel_plane_state *plane_state, + int *scaler_id) +{ + struct drm_i915_private *dev_priv = to_i915(intel_crtc->base.dev); + int j; + u32 mode; + + if (*scaler_id < 0) { + /* find a free scaler */ + for (j = 0; j < intel_crtc->num_scalers; j++) { + if (scaler_state->scalers[j].in_use) + continue; + + *scaler_id = j; + scaler_state->scalers[*scaler_id].in_use = 1; + break; + } + } + + if (drm_WARN(&dev_priv->drm, *scaler_id < 0, + "Cannot find scaler for %s:%d\n", name, idx)) + return -EINVAL; + + /* set scaler mode */ + if (plane_state && plane_state->hw.fb && + plane_state->hw.fb->format->is_yuv && + plane_state->hw.fb->format->num_planes > 1) { + struct intel_plane *plane = to_intel_plane(plane_state->uapi.plane); + + if (DISPLAY_VER(dev_priv) == 9) { + mode = SKL_PS_SCALER_MODE_NV12; + } else if (icl_is_hdr_plane(dev_priv, plane->id)) { + /* + * On gen11+'s HDR planes we only use the scaler for + * scaling. They have a dedicated chroma upsampler, so + * we don't need the scaler to upsample the UV plane. + */ + mode = PS_SCALER_MODE_NORMAL; + } else { + struct intel_plane *linked = + plane_state->planar_linked_plane; + + mode = PS_SCALER_MODE_PLANAR; + + if (linked) + mode |= PS_BINDING_Y_PLANE(linked->id); + } + } else if (DISPLAY_VER(dev_priv) >= 10) { + mode = PS_SCALER_MODE_NORMAL; + } else if (num_scalers_need == 1 && intel_crtc->num_scalers > 1) { + /* + * when only 1 scaler is in use on a pipe with 2 scalers + * scaler 0 operates in high quality (HQ) mode. + * In this case use scaler 0 to take advantage of HQ mode + */ + scaler_state->scalers[*scaler_id].in_use = 0; + *scaler_id = 0; + scaler_state->scalers[0].in_use = 1; + mode = SKL_PS_SCALER_MODE_HQ; + } else { + mode = SKL_PS_SCALER_MODE_DYN; + } + + /* + * FIXME: we should also check the scaler factors for pfit, so + * this shouldn't be tied directly to planes. + */ + if (plane_state && plane_state->hw.fb) { + const struct drm_framebuffer *fb = plane_state->hw.fb; + const struct drm_rect *src = &plane_state->uapi.src; + const struct drm_rect *dst = &plane_state->uapi.dst; + int hscale, vscale, max_vscale, max_hscale; + + /* + * FIXME: When two scalers are needed, but only one of + * them needs to downscale, we should make sure that + * the one that needs downscaling support is assigned + * as the first scaler, so we don't reject downscaling + * unnecessarily. + */ + + if (DISPLAY_VER(dev_priv) >= 14) { + /* + * On versions 14 and up, only the first + * scaler supports a vertical scaling factor + * of more than 1.0, while a horizontal + * scaling factor of 3.0 is supported. + */ + max_hscale = 0x30000 - 1; + if (*scaler_id == 0) + max_vscale = 0x30000 - 1; + else + max_vscale = 0x10000; + + } else if (DISPLAY_VER(dev_priv) >= 10 || + !intel_format_info_is_yuv_semiplanar(fb->format, fb->modifier)) { + max_hscale = 0x30000 - 1; + max_vscale = 0x30000 - 1; + } else { + max_hscale = 0x20000 - 1; + max_vscale = 0x20000 - 1; + } + + /* + * FIXME: We should change the if-else block above to + * support HQ vs dynamic scaler properly. + */ + + /* Check if required scaling is within limits */ + hscale = drm_rect_calc_hscale(src, dst, 1, max_hscale); + vscale = drm_rect_calc_vscale(src, dst, 1, max_vscale); + + if (hscale < 0 || vscale < 0) { + drm_dbg_kms(&dev_priv->drm, + "Scaler %d doesn't support required plane scaling\n", + *scaler_id); + drm_rect_debug_print("src: ", src, true); + drm_rect_debug_print("dst: ", dst, false); + + return -EINVAL; + } + } + + drm_dbg_kms(&dev_priv->drm, "Attached scaler id %u.%u to %s:%d\n", + intel_crtc->pipe, *scaler_id, name, idx); + scaler_state->scalers[*scaler_id].mode = mode; + + return 0; +} + +/** + * intel_atomic_setup_scalers() - setup scalers for crtc per staged requests + * @dev_priv: i915 device + * @intel_crtc: intel crtc + * @crtc_state: incoming crtc_state to validate and setup scalers + * + * This function sets up scalers based on staged scaling requests for + * a @crtc and its planes. It is called from crtc level check path. If request + * is a supportable request, it attaches scalers to requested planes and crtc. + * + * This function takes into account the current scaler(s) in use by any planes + * not being part of this atomic state + * + * Returns: + * 0 - scalers were setup successfully + * error code - otherwise + */ +int intel_atomic_setup_scalers(struct drm_i915_private *dev_priv, + struct intel_crtc *intel_crtc, + struct intel_crtc_state *crtc_state) +{ + struct drm_plane *plane = NULL; + struct intel_plane *intel_plane; + struct intel_plane_state *plane_state = NULL; + struct intel_crtc_scaler_state *scaler_state = + &crtc_state->scaler_state; + struct drm_atomic_state *drm_state = crtc_state->uapi.state; + struct intel_atomic_state *intel_state = to_intel_atomic_state(drm_state); + int num_scalers_need; + int i; + + num_scalers_need = hweight32(scaler_state->scaler_users); + + /* + * High level flow: + * - staged scaler requests are already in scaler_state->scaler_users + * - check whether staged scaling requests can be supported + * - add planes using scalers that aren't in current transaction + * - assign scalers to requested users + * - as part of plane commit, scalers will be committed + * (i.e., either attached or detached) to respective planes in hw + * - as part of crtc_commit, scaler will be either attached or detached + * to crtc in hw + */ + + /* fail if required scalers > available scalers */ + if (num_scalers_need > intel_crtc->num_scalers) { + drm_dbg_kms(&dev_priv->drm, + "Too many scaling requests %d > %d\n", + num_scalers_need, intel_crtc->num_scalers); + return -EINVAL; + } + + /* walkthrough scaler_users bits and start assigning scalers */ + for (i = 0; i < sizeof(scaler_state->scaler_users) * 8; i++) { + int *scaler_id; + const char *name; + int idx, ret; + + /* skip if scaler not required */ + if (!(scaler_state->scaler_users & (1 << i))) + continue; + + if (i == SKL_CRTC_INDEX) { + name = "CRTC"; + idx = intel_crtc->base.base.id; + + /* panel fitter case: assign as a crtc scaler */ + scaler_id = &scaler_state->scaler_id; + } else { + name = "PLANE"; + + /* plane scaler case: assign as a plane scaler */ + /* find the plane that set the bit as scaler_user */ + plane = drm_state->planes[i].ptr; + + /* + * to enable/disable hq mode, add planes that are using scaler + * into this transaction + */ + if (!plane) { + struct drm_plane_state *state; + + /* + * GLK+ scalers don't have a HQ mode so it + * isn't necessary to change between HQ and dyn mode + * on those platforms. + */ + if (DISPLAY_VER(dev_priv) >= 10) + continue; + + plane = drm_plane_from_index(&dev_priv->drm, i); + state = drm_atomic_get_plane_state(drm_state, plane); + if (IS_ERR(state)) { + drm_dbg_kms(&dev_priv->drm, + "Failed to add [PLANE:%d] to drm_state\n", + plane->base.id); + return PTR_ERR(state); + } + } + + intel_plane = to_intel_plane(plane); + idx = plane->base.id; + + /* plane on different crtc cannot be a scaler user of this crtc */ + if (drm_WARN_ON(&dev_priv->drm, + intel_plane->pipe != intel_crtc->pipe)) + continue; + + plane_state = intel_atomic_get_new_plane_state(intel_state, + intel_plane); + scaler_id = &plane_state->scaler_id; + } + + ret = intel_atomic_setup_scaler(scaler_state, num_scalers_need, + intel_crtc, name, idx, + plane_state, scaler_id); + if (ret < 0) + return ret; + } + + return 0; +} + static int glk_coef_tap(int i) { return i % 7; @@ -401,7 +658,7 @@ static void glk_program_nearest_filter_coefs(struct drm_i915_private *dev_priv, int i; intel_de_write_fw(dev_priv, GLK_PS_COEF_INDEX_SET(pipe, id, set), - PS_COEE_INDEX_AUTO_INC); + PS_COEF_INDEX_AUTO_INC); for (i = 0; i < 17 * 7; i += 2) { u32 tmp; @@ -484,8 +741,8 @@ void skl_pfit_enable(const struct intel_crtc_state *crtc_state) id = scaler_state->scaler_id; - ps_ctrl = skl_scaler_get_filter_select(crtc_state->hw.scaling_filter, 0); - ps_ctrl |= PS_SCALER_EN | scaler_state->scalers[id].mode; + ps_ctrl = PS_SCALER_EN | PS_BINDING_PIPE | scaler_state->scalers[id].mode | + skl_scaler_get_filter_select(crtc_state->hw.scaling_filter, 0); skl_scaler_setup_filter(dev_priv, pipe, id, 0, crtc_state->hw.scaling_filter); @@ -497,9 +754,9 @@ void skl_pfit_enable(const struct intel_crtc_state *crtc_state) intel_de_write_fw(dev_priv, SKL_PS_HPHASE(pipe, id), PS_Y_PHASE(0) | PS_UV_RGB_PHASE(uv_rgb_hphase)); intel_de_write_fw(dev_priv, SKL_PS_WIN_POS(pipe, id), - x << 16 | y); + PS_WIN_XPOS(x) | PS_WIN_YPOS(y)); intel_de_write_fw(dev_priv, SKL_PS_WIN_SZ(pipe, id), - width << 16 | height); + PS_WIN_XSIZE(width) | PS_WIN_YSIZE(height)); } void @@ -547,8 +804,8 @@ skl_program_plane_scaler(struct intel_plane *plane, uv_rgb_vphase = skl_scaler_calc_phase(1, vscale, false); } - ps_ctrl = skl_scaler_get_filter_select(plane_state->hw.scaling_filter, 0); - ps_ctrl |= PS_SCALER_EN | PS_PLANE_SEL(plane->id) | scaler->mode; + ps_ctrl = PS_SCALER_EN | PS_BINDING_PLANE(plane->id) | scaler->mode | + skl_scaler_get_filter_select(plane_state->hw.scaling_filter, 0); skl_scaler_setup_filter(dev_priv, pipe, scaler_id, 0, plane_state->hw.scaling_filter); @@ -559,9 +816,9 @@ skl_program_plane_scaler(struct intel_plane *plane, intel_de_write_fw(dev_priv, SKL_PS_HPHASE(pipe, scaler_id), PS_Y_PHASE(y_hphase) | PS_UV_RGB_PHASE(uv_rgb_hphase)); intel_de_write_fw(dev_priv, SKL_PS_WIN_POS(pipe, scaler_id), - (crtc_x << 16) | crtc_y); + PS_WIN_XPOS(crtc_x) | PS_WIN_YPOS(crtc_y)); intel_de_write_fw(dev_priv, SKL_PS_WIN_SZ(pipe, scaler_id), - (crtc_w << 16) | crtc_h); + PS_WIN_XSIZE(crtc_w) | PS_WIN_YSIZE(crtc_h)); } static void skl_detach_scaler(struct intel_crtc *crtc, int id) @@ -599,3 +856,42 @@ void skl_scaler_disable(const struct intel_crtc_state *old_crtc_state) for (i = 0; i < crtc->num_scalers; i++) skl_detach_scaler(crtc, i); } + +void skl_scaler_get_config(struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + struct intel_crtc_scaler_state *scaler_state = &crtc_state->scaler_state; + int id = -1; + int i; + + /* find scaler attached to this pipe */ + for (i = 0; i < crtc->num_scalers; i++) { + u32 ctl, pos, size; + + ctl = intel_de_read(dev_priv, SKL_PS_CTRL(crtc->pipe, i)); + if ((ctl & (PS_SCALER_EN | PS_BINDING_MASK)) != (PS_SCALER_EN | PS_BINDING_PIPE)) + continue; + + id = i; + crtc_state->pch_pfit.enabled = true; + + pos = intel_de_read(dev_priv, SKL_PS_WIN_POS(crtc->pipe, i)); + size = intel_de_read(dev_priv, SKL_PS_WIN_SZ(crtc->pipe, i)); + + drm_rect_init(&crtc_state->pch_pfit.dst, + REG_FIELD_GET(PS_WIN_XPOS_MASK, pos), + REG_FIELD_GET(PS_WIN_YPOS_MASK, pos), + REG_FIELD_GET(PS_WIN_XSIZE_MASK, size), + REG_FIELD_GET(PS_WIN_YSIZE_MASK, size)); + + scaler_state->scalers[i].in_use = true; + break; + } + + scaler_state->scaler_id = id; + if (id >= 0) + scaler_state->scaler_users |= (1 << SKL_CRTC_INDEX); + else + scaler_state->scaler_users &= ~(1 << SKL_CRTC_INDEX); +} diff --git a/drivers/gpu/drm/i915/display/skl_scaler.h b/drivers/gpu/drm/i915/display/skl_scaler.h index 0097d5d08e102a642c70d07fd9ca750fe1f67078..63f93ca03c89ed2821f218c236a222fda5017d01 100644 --- a/drivers/gpu/drm/i915/display/skl_scaler.h +++ b/drivers/gpu/drm/i915/display/skl_scaler.h @@ -8,17 +8,22 @@ #include enum drm_scaling_filter; +enum pipe; struct drm_i915_private; +struct intel_crtc; struct intel_crtc_state; -struct intel_plane_state; struct intel_plane; -enum pipe; +struct intel_plane_state; int skl_update_scaler_crtc(struct intel_crtc_state *crtc_state); int skl_update_scaler_plane(struct intel_crtc_state *crtc_state, struct intel_plane_state *plane_state); +int intel_atomic_setup_scalers(struct drm_i915_private *dev_priv, + struct intel_crtc *intel_crtc, + struct intel_crtc_state *crtc_state); + void skl_pfit_enable(const struct intel_crtc_state *crtc_state); void skl_program_plane_scaler(struct intel_plane *plane, @@ -26,4 +31,7 @@ void skl_program_plane_scaler(struct intel_plane *plane, const struct intel_plane_state *plane_state); void skl_detach_scalers(const struct intel_crtc_state *crtc_state); void skl_scaler_disable(const struct intel_crtc_state *old_crtc_state); + +void skl_scaler_get_config(struct intel_crtc_state *crtc_state); + #endif diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c index 8ea0598a5a07ed87320331485dea2c74182fc8ea..6b01a0b68b97fe8f5011c5e6f3c689eb508f6f91 100644 --- a/drivers/gpu/drm/i915/display/skl_universal_plane.c +++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c @@ -9,10 +9,10 @@ #include #include "i915_drv.h" -#include "i915_irq.h" #include "i915_reg.h" #include "intel_atomic_plane.h" #include "intel_de.h" +#include "intel_display_irq.h" #include "intel_display_types.h" #include "intel_fb.h" #include "intel_fbc.h" @@ -789,6 +789,14 @@ static u32 skl_plane_ctl_tiling(u64 fb_modifier) PLANE_CTL_CLEAR_COLOR_DISABLE; case I915_FORMAT_MOD_4_TILED_DG2_RC_CCS_CC: return PLANE_CTL_TILED_4 | PLANE_CTL_RENDER_DECOMPRESSION_ENABLE; + case I915_FORMAT_MOD_4_TILED_MTL_RC_CCS: + return PLANE_CTL_TILED_4 | + PLANE_CTL_RENDER_DECOMPRESSION_ENABLE | + PLANE_CTL_CLEAR_COLOR_DISABLE; + case I915_FORMAT_MOD_4_TILED_MTL_RC_CCS_CC: + return PLANE_CTL_TILED_4 | PLANE_CTL_RENDER_DECOMPRESSION_ENABLE; + case I915_FORMAT_MOD_4_TILED_MTL_MC_CCS: + return PLANE_CTL_TILED_4 | PLANE_CTL_MEDIA_DECOMPRESSION_ENABLE; case I915_FORMAT_MOD_Y_TILED_CCS: case I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS_CC: return PLANE_CTL_TILED_Y | PLANE_CTL_RENDER_DECOMPRESSION_ENABLE; @@ -1936,7 +1944,7 @@ static enum intel_fbc_id skl_fbc_id_for_pipe(enum pipe pipe) static bool skl_plane_has_fbc(struct drm_i915_private *dev_priv, enum intel_fbc_id fbc_id, enum plane_id plane_id) { - if ((RUNTIME_INFO(dev_priv)->fbc_mask & BIT(fbc_id)) == 0) + if ((DISPLAY_RUNTIME_INFO(dev_priv)->fbc_mask & BIT(fbc_id)) == 0) return false; return plane_id == PLANE_PRIMARY; @@ -2160,6 +2168,11 @@ skl_plane_disable_flip_done(struct intel_plane *plane) static bool skl_plane_has_rc_ccs(struct drm_i915_private *i915, enum pipe pipe, enum plane_id plane_id) { + /* Wa_14017240301 */ + if (IS_MTL_GRAPHICS_STEP(i915, M, STEP_A0, STEP_B0) || + IS_MTL_GRAPHICS_STEP(i915, P, STEP_A0, STEP_B0)) + return false; + /* Wa_22011186057 */ if (IS_ADLP_DISPLAY_STEP(i915, STEP_A0, STEP_B0)) return false; @@ -2441,12 +2454,17 @@ skl_get_initial_plane_config(struct intel_crtc *crtc, case PLANE_CTL_TILED_Y: plane_config->tiling = I915_TILING_Y; if (val & PLANE_CTL_RENDER_DECOMPRESSION_ENABLE) - if (DISPLAY_VER(dev_priv) >= 12) + if (DISPLAY_VER(dev_priv) >= 14) + fb->modifier = I915_FORMAT_MOD_4_TILED_MTL_RC_CCS; + else if (DISPLAY_VER(dev_priv) >= 12) fb->modifier = I915_FORMAT_MOD_Y_TILED_GEN12_RC_CCS; else fb->modifier = I915_FORMAT_MOD_Y_TILED_CCS; else if (val & PLANE_CTL_MEDIA_DECOMPRESSION_ENABLE) - fb->modifier = I915_FORMAT_MOD_Y_TILED_GEN12_MC_CCS; + if (DISPLAY_VER(dev_priv) >= 14) + fb->modifier = I915_FORMAT_MOD_4_TILED_MTL_MC_CCS; + else + fb->modifier = I915_FORMAT_MOD_Y_TILED_GEN12_MC_CCS; else fb->modifier = I915_FORMAT_MOD_Y_TILED; break; @@ -2511,6 +2529,7 @@ skl_get_initial_plane_config(struct intel_crtc *crtc, plane_config->base = base; offset = intel_de_read(dev_priv, PLANE_OFFSET(pipe, plane_id)); + drm_WARN_ON(&dev_priv->drm, offset != 0); val = intel_de_read(dev_priv, PLANE_SIZE(pipe, plane_id)); fb->height = REG_FIELD_GET(PLANE_HEIGHT_MASK, val) + 1; diff --git a/drivers/gpu/drm/i915/display/skl_watermark.c b/drivers/gpu/drm/i915/display/skl_watermark.c index 1c7e6468f3e34a06d3c400f6dd2c512d81582642..063929a42a42f543decb97f6385070eb79f06365 100644 --- a/drivers/gpu/drm/i915/display/skl_watermark.c +++ b/drivers/gpu/drm/i915/display/skl_watermark.c @@ -507,8 +507,8 @@ static u16 skl_ddb_entry_init(struct skl_ddb_entry *entry, static int intel_dbuf_slice_size(struct drm_i915_private *i915) { - return INTEL_INFO(i915)->display.dbuf.size / - hweight8(INTEL_INFO(i915)->display.dbuf.slice_mask); + return DISPLAY_INFO(i915)->dbuf.size / + hweight8(DISPLAY_INFO(i915)->dbuf.slice_mask); } static void @@ -527,7 +527,7 @@ skl_ddb_entry_for_slices(struct drm_i915_private *i915, u8 slice_mask, ddb->end = fls(slice_mask) * slice_size; WARN_ON(ddb->start >= ddb->end); - WARN_ON(ddb->end > INTEL_INFO(i915)->display.dbuf.size); + WARN_ON(ddb->end > DISPLAY_INFO(i915)->dbuf.size); } static unsigned int mbus_ddb_offset(struct drm_i915_private *i915, u8 slice_mask) @@ -2625,7 +2625,7 @@ skl_compute_ddb(struct intel_atomic_state *state) "Enabled dbuf slices 0x%x -> 0x%x (total dbuf slices 0x%x), mbus joined? %s->%s\n", old_dbuf_state->enabled_slices, new_dbuf_state->enabled_slices, - INTEL_INFO(i915)->display.dbuf.slice_mask, + DISPLAY_INFO(i915)->dbuf.slice_mask, str_yes_no(old_dbuf_state->joined_mbus), str_yes_no(new_dbuf_state->joined_mbus)); } @@ -2900,7 +2900,7 @@ static int skl_compute_wm(struct intel_atomic_state *state) { struct intel_crtc *crtc; - struct intel_crtc_state *new_crtc_state; + struct intel_crtc_state __maybe_unused *new_crtc_state; int ret, i; for_each_new_intel_crtc_in_state(state, crtc, new_crtc_state, i) { diff --git a/drivers/gpu/drm/i915/display/vlv_dsi.c b/drivers/gpu/drm/i915/display/vlv_dsi.c index 61d008d4e5f1f4e54022ffedd8b1c16d71f24ad6..ae2f3ab3e73de82dabcb3c1d728ac1b37d13f58f 100644 --- a/drivers/gpu/drm/i915/display/vlv_dsi.c +++ b/drivers/gpu/drm/i915/display/vlv_dsi.c @@ -136,7 +136,7 @@ static ssize_t intel_dsi_host_transfer(struct mipi_dsi_host *host, enum port port = intel_dsi_host->port; struct mipi_dsi_packet packet; ssize_t ret; - const u8 *header, *data; + const u8 *header; i915_reg_t data_reg, ctrl_reg; u32 data_mask, ctrl_mask; @@ -145,7 +145,6 @@ static ssize_t intel_dsi_host_transfer(struct mipi_dsi_host *host, return ret; header = packet.header; - data = packet.payload; if (msg->flags & MIPI_DSI_MSG_USE_LPM) { data_reg = MIPI_LP_GEN_DATA(port); @@ -280,6 +279,7 @@ static int intel_dsi_compute_config(struct intel_encoder *encoder, int ret; drm_dbg_kms(&dev_priv->drm, "\n"); + pipe_config->sink_format = INTEL_OUTPUT_FORMAT_RGB; pipe_config->output_format = INTEL_OUTPUT_FORMAT_RGB; ret = intel_panel_compute_config(intel_connector, adjusted_mode); @@ -1039,7 +1039,7 @@ static void bxt_dsi_get_pipe_config(struct intel_encoder *encoder, unsigned int lane_count = intel_dsi->lane_count; unsigned int bpp, fmt; enum port port; - u16 hactive, hfp, hsync, hbp, vfp, vsync, vbp; + u16 hactive, hfp, hsync, hbp, vfp, vsync; u16 hfp_sw, hsync_sw, hbp_sw; u16 crtc_htotal_sw, crtc_hsync_start_sw, crtc_hsync_end_sw, crtc_hblank_start_sw, crtc_hblank_end_sw; @@ -1104,7 +1104,6 @@ static void bxt_dsi_get_pipe_config(struct intel_encoder *encoder, /* vertical values are in terms of lines */ vfp = intel_de_read(dev_priv, MIPI_VFP_COUNT(port)); vsync = intel_de_read(dev_priv, MIPI_VSYNC_PADDING_COUNT(port)); - vbp = intel_de_read(dev_priv, MIPI_VBP_COUNT(port)); adjusted_mode->crtc_htotal = hactive + hfp + hsync + hbp; adjusted_mode->crtc_hsync_start = hfp + adjusted_mode->crtc_hdisplay; diff --git a/drivers/gpu/drm/i915/display/vlv_dsi_pll.c b/drivers/gpu/drm/i915/display/vlv_dsi_pll.c index b697badbbe7110cf8841a18cc6384c26fff8739c..ae0a0b11bae3b590fa5513f10e16dc5d55f752b8 100644 --- a/drivers/gpu/drm/i915/display/vlv_dsi_pll.c +++ b/drivers/gpu/drm/i915/display/vlv_dsi_pll.c @@ -598,7 +598,7 @@ static void assert_dsi_pll(struct drm_i915_private *i915, bool state) cur_state = vlv_cck_read(i915, CCK_REG_DSI_PLL_CONTROL) & DSI_PLL_VCO_EN; vlv_cck_put(i915); - I915_STATE_WARN(cur_state != state, + I915_STATE_WARN(i915, cur_state != state, "DSI PLL state assertion failure (expected %s, current %s)\n", str_on_off(state), str_on_off(cur_state)); } diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context.c b/drivers/gpu/drm/i915/gem/i915_gem_context.c index 5402a7bbcb1d14e1c7c976b8f228d5cafa8ea9f8..9a9ff84c90d7e69d9f6c3b8a5f892f639c960e98 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_context.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_context.c @@ -964,7 +964,11 @@ static int intel_context_set_gem(struct intel_context *ce, RCU_INIT_POINTER(ce->gem_context, ctx); GEM_BUG_ON(intel_context_is_pinned(ce)); - ce->ring_size = SZ_16K; + + if (ce->engine->class == COMPUTE_CLASS) + ce->ring_size = SZ_512K; + else + ce->ring_size = SZ_16K; i915_vm_put(ce->vm); ce->vm = i915_gem_context_get_eb_vm(ctx); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_create.c b/drivers/gpu/drm/i915/gem/i915_gem_create.c index bfe1dbda4cb75afe55fb3080d7cc1881679f7d56..d24c0ce8805c7056444d66374ed2c86cf14d93bc 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_create.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_create.c @@ -245,6 +245,7 @@ struct create_ext { unsigned int n_placements; unsigned int placement_mask; unsigned long flags; + unsigned int pat_index; }; static void repr_placements(char *buf, size_t size, @@ -394,11 +395,43 @@ static int ext_set_protected(struct i915_user_extension __user *base, void *data return 0; } +static int ext_set_pat(struct i915_user_extension __user *base, void *data) +{ + struct create_ext *ext_data = data; + struct drm_i915_private *i915 = ext_data->i915; + struct drm_i915_gem_create_ext_set_pat ext; + unsigned int max_pat_index; + + BUILD_BUG_ON(sizeof(struct drm_i915_gem_create_ext_set_pat) != + offsetofend(struct drm_i915_gem_create_ext_set_pat, rsvd)); + + /* Limiting the extension only to Meteor Lake */ + if (!IS_METEORLAKE(i915)) + return -ENODEV; + + if (copy_from_user(&ext, base, sizeof(ext))) + return -EFAULT; + + max_pat_index = INTEL_INFO(i915)->max_pat_index; + + if (ext.pat_index > max_pat_index) { + drm_dbg(&i915->drm, "PAT index is invalid: %u\n", + ext.pat_index); + return -EINVAL; + } + + ext_data->pat_index = ext.pat_index; + + return 0; +} + static const i915_user_extension_fn create_extensions[] = { [I915_GEM_CREATE_EXT_MEMORY_REGIONS] = ext_set_placements, [I915_GEM_CREATE_EXT_PROTECTED_CONTENT] = ext_set_protected, + [I915_GEM_CREATE_EXT_SET_PAT] = ext_set_pat, }; +#define PAT_INDEX_NOT_SET 0xffff /** * i915_gem_create_ext_ioctl - Creates a new mm object and returns a handle to it. * @dev: drm device pointer @@ -418,6 +451,7 @@ i915_gem_create_ext_ioctl(struct drm_device *dev, void *data, if (args->flags & ~I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS) return -EINVAL; + ext_data.pat_index = PAT_INDEX_NOT_SET; ret = i915_user_extensions(u64_to_user_ptr(args->extensions), create_extensions, ARRAY_SIZE(create_extensions), @@ -454,5 +488,11 @@ i915_gem_create_ext_ioctl(struct drm_device *dev, void *data, if (IS_ERR(obj)) return PTR_ERR(obj); + if (ext_data.pat_index != PAT_INDEX_NOT_SET) { + i915_gem_object_set_pat_index(obj, ext_data.pat_index); + /* Mark pat_index is set by UMD */ + obj->pat_set_by_user = true; + } + return i915_gem_publish(obj, file, &args->size, &args->handle); } diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c index 5fb459ea42945a5b1792158c8c1b344930f2c327..cfd7929587d8fa7709806c13c2291bc3071e0e87 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c @@ -736,7 +736,6 @@ static int eb_reserve(struct i915_execbuffer *eb) struct eb_vma *ev; unsigned int pass; int err = 0; - bool unpinned; /* * We have one more buffers that we couldn't bind, which could be due to @@ -776,7 +775,7 @@ static int eb_reserve(struct i915_execbuffer *eb) pin_flags |= PIN_NONBLOCK; if (pass >= 1) - unpinned = eb_unbind(eb, pass >= 2); + eb_unbind(eb, pass >= 2); if (pass == 2) { err = mutex_lock_interruptible(&eb->context->vm->mutex); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c index 46a19b099ec88c3c26ce20680abc61589dd653f2..97ac6fb37958f4a44e8ee8b5c826219be908eda8 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c @@ -208,6 +208,12 @@ bool i915_gem_object_can_bypass_llc(struct drm_i915_gem_object *obj) if (!(obj->flags & I915_BO_ALLOC_USER)) return false; + /* + * Always flush cache for UMD objects at creation time. + */ + if (obj->pat_set_by_user) + return true; + /* * EHL and JSL add the 'Bypass LLC' MOCS entry, which should make it * possible for userspace to bypass the GTT caching bits set by the diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c index cad4a6017f4b52e5ce5abc8379e1c07c7f6cf520..33d5d5178103b78f8d6a6cd3287f65ef59480bf5 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c @@ -455,7 +455,7 @@ shmem_pwrite(struct drm_i915_gem_object *obj, struct page *page; void *data, *vaddr; int err; - char c; + char __maybe_unused c; len = PAGE_SIZE - pg; if (len > remain) diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c index ad6a3b2fb387bb039f69b83861ed12a4df54a5b4..7021b6e9b219ef3502af7beaa3770a5a9a251409 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c @@ -348,8 +348,10 @@ static int live_parallel_switch(void *arg) continue; ce = intel_context_create(data[m].ce[0]->engine); - if (IS_ERR(ce)) + if (IS_ERR(ce)) { + err = PTR_ERR(ce); goto out; + } err = intel_context_pin(ce); if (err) { @@ -369,8 +371,10 @@ static int live_parallel_switch(void *arg) worker = kthread_create_worker(0, "igt/parallel:%s", data[n].ce[0]->engine->name); - if (IS_ERR(worker)) + if (IS_ERR(worker)) { + err = PTR_ERR(worker); goto out; + } data[n].worker = worker; } @@ -399,8 +403,10 @@ static int live_parallel_switch(void *arg) } } - if (igt_live_test_end(&t)) - err = -EIO; + if (igt_live_test_end(&t)) { + err = err ?: -EIO; + break; + } } out: diff --git a/drivers/gpu/drm/i915/gt/gen8_engine_cs.c b/drivers/gpu/drm/i915/gt/gen8_engine_cs.c index e1c76e5bfa82796432b74fd7c6637b8342a4ffcf..23857cc08eca1fbfec37a6ab0c96987798706a49 100644 --- a/drivers/gpu/drm/i915/gt/gen8_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/gen8_engine_cs.c @@ -177,14 +177,40 @@ u32 *gen12_emit_aux_table_inv(struct intel_gt *gt, u32 *cs, const i915_reg_t inv return cs; } +static int mtl_dummy_pipe_control(struct i915_request *rq) +{ + /* Wa_14016712196 */ + if (IS_MTL_GRAPHICS_STEP(rq->engine->i915, M, STEP_A0, STEP_B0) || + IS_MTL_GRAPHICS_STEP(rq->engine->i915, P, STEP_A0, STEP_B0)) { + u32 *cs; + + /* dummy PIPE_CONTROL + depth flush */ + cs = intel_ring_begin(rq, 6); + if (IS_ERR(cs)) + return PTR_ERR(cs); + cs = gen12_emit_pipe_control(cs, + 0, + PIPE_CONTROL_DEPTH_CACHE_FLUSH, + LRC_PPHWSP_SCRATCH_ADDR); + intel_ring_advance(rq, cs); + } + + return 0; +} + int gen12_emit_flush_rcs(struct i915_request *rq, u32 mode) { struct intel_engine_cs *engine = rq->engine; if (mode & EMIT_FLUSH) { u32 flags = 0; + int err; u32 *cs; + err = mtl_dummy_pipe_control(rq); + if (err) + return err; + flags |= PIPE_CONTROL_TILE_CACHE_FLUSH; flags |= PIPE_CONTROL_FLUSH_L3; flags |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH; @@ -217,6 +243,11 @@ int gen12_emit_flush_rcs(struct i915_request *rq, u32 mode) if (mode & EMIT_INVALIDATE) { u32 flags = 0; u32 *cs, count; + int err; + + err = mtl_dummy_pipe_control(rq); + if (err) + return err; flags |= PIPE_CONTROL_COMMAND_CACHE_INVALIDATE; flags |= PIPE_CONTROL_TLB_INVALIDATE; @@ -733,6 +764,13 @@ u32 *gen12_emit_fini_breadcrumb_rcs(struct i915_request *rq, u32 *cs) PIPE_CONTROL_DC_FLUSH_ENABLE | PIPE_CONTROL_FLUSH_ENABLE); + /* Wa_14016712196 */ + if (IS_MTL_GRAPHICS_STEP(i915, M, STEP_A0, STEP_B0) || + IS_MTL_GRAPHICS_STEP(i915, P, STEP_A0, STEP_B0)) + /* dummy PIPE_CONTROL + depth flush */ + cs = gen12_emit_pipe_control(cs, 0, + PIPE_CONTROL_DEPTH_CACHE_FLUSH, 0); + if (GRAPHICS_VER(i915) == 12 && GRAPHICS_VER_FULL(i915) < IP_VER(12, 50)) /* Wa_1409600907 */ flags |= PIPE_CONTROL_DEPTH_STALL; diff --git a/drivers/gpu/drm/i915/gt/intel_engine_pm.c b/drivers/gpu/drm/i915/gt/intel_engine_pm.c index ee531a5c142c775f24e350e123770f628ce5c853..21af0ec52223421115011d30cad0c88ff020c126 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_pm.c +++ b/drivers/gpu/drm/i915/gt/intel_engine_pm.c @@ -296,9 +296,7 @@ static const struct intel_wakeref_ops wf_ops = { void intel_engine_init__pm(struct intel_engine_cs *engine) { - struct intel_runtime_pm *rpm = engine->uncore->rpm; - - intel_wakeref_init(&engine->wakeref, rpm, &wf_ops); + intel_wakeref_init(&engine->wakeref, engine->i915, &wf_ops); intel_engine_init_heartbeat(engine); intel_gsc_idle_msg_enable(engine); diff --git a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c index 750326434677f1703f1c21b8e015a1436fa05f40..2ebd937f3b4cb61907c234fc94746d127077143d 100644 --- a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c +++ b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c @@ -2327,6 +2327,7 @@ static u32 active_ccid(struct intel_engine_cs *engine) static void execlists_capture(struct intel_engine_cs *engine) { + struct drm_i915_private *i915 = engine->i915; struct execlists_capture *cap; if (!IS_ENABLED(CONFIG_DRM_I915_CAPTURE_ERROR)) @@ -2375,7 +2376,7 @@ static void execlists_capture(struct intel_engine_cs *engine) goto err_rq; INIT_WORK(&cap->work, execlists_capture_work); - schedule_work(&cap->work); + queue_work(i915->unordered_wq, &cap->work); return; err_rq: @@ -3680,7 +3681,7 @@ static void virtual_context_destroy(struct kref *kref) * lock, we can delegate the free of the engine to an RCU worker. */ INIT_RCU_WORK(&ve->rcu, rcu_virtual_context_destroy); - queue_rcu_work(system_wq, &ve->rcu); + queue_rcu_work(ve->context.engine->i915->unordered_wq, &ve->rcu); } static void virtual_engine_initial_hint(struct virtual_engine *ve) diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt.c b/drivers/gpu/drm/i915/gt/intel_ggtt.c index 2a7942fac7980598f327d13c8752a5a5bfd85c6c..dd0ed941441aab84be728ee5f333f44d682a8c66 100644 --- a/drivers/gpu/drm/i915/gt/intel_ggtt.c +++ b/drivers/gpu/drm/i915/gt/intel_ggtt.c @@ -1015,16 +1015,16 @@ static int gen8_gmch_probe(struct i915_ggtt *ggtt) /* * For pre-gen8 platforms pat_index is the same as enum i915_cache_level, - * so these PTE encode functions are left with using cache_level. + * so the switch-case statements in these PTE encode functions are still valid. * See translation table LEGACY_CACHELEVEL. */ static u64 snb_pte_encode(dma_addr_t addr, - enum i915_cache_level level, + unsigned int pat_index, u32 flags) { gen6_pte_t pte = GEN6_PTE_ADDR_ENCODE(addr) | GEN6_PTE_VALID; - switch (level) { + switch (pat_index) { case I915_CACHE_L3_LLC: case I915_CACHE_LLC: pte |= GEN6_PTE_CACHE_LLC; @@ -1033,19 +1033,19 @@ static u64 snb_pte_encode(dma_addr_t addr, pte |= GEN6_PTE_UNCACHED; break; default: - MISSING_CASE(level); + MISSING_CASE(pat_index); } return pte; } static u64 ivb_pte_encode(dma_addr_t addr, - enum i915_cache_level level, + unsigned int pat_index, u32 flags) { gen6_pte_t pte = GEN6_PTE_ADDR_ENCODE(addr) | GEN6_PTE_VALID; - switch (level) { + switch (pat_index) { case I915_CACHE_L3_LLC: pte |= GEN7_PTE_CACHE_L3_LLC; break; @@ -1056,14 +1056,14 @@ static u64 ivb_pte_encode(dma_addr_t addr, pte |= GEN6_PTE_UNCACHED; break; default: - MISSING_CASE(level); + MISSING_CASE(pat_index); } return pte; } static u64 byt_pte_encode(dma_addr_t addr, - enum i915_cache_level level, + unsigned int pat_index, u32 flags) { gen6_pte_t pte = GEN6_PTE_ADDR_ENCODE(addr) | GEN6_PTE_VALID; @@ -1071,31 +1071,31 @@ static u64 byt_pte_encode(dma_addr_t addr, if (!(flags & PTE_READ_ONLY)) pte |= BYT_PTE_WRITEABLE; - if (level != I915_CACHE_NONE) + if (pat_index != I915_CACHE_NONE) pte |= BYT_PTE_SNOOPED_BY_CPU_CACHES; return pte; } static u64 hsw_pte_encode(dma_addr_t addr, - enum i915_cache_level level, + unsigned int pat_index, u32 flags) { gen6_pte_t pte = HSW_PTE_ADDR_ENCODE(addr) | GEN6_PTE_VALID; - if (level != I915_CACHE_NONE) + if (pat_index != I915_CACHE_NONE) pte |= HSW_WB_LLC_AGE3; return pte; } static u64 iris_pte_encode(dma_addr_t addr, - enum i915_cache_level level, + unsigned int pat_index, u32 flags) { gen6_pte_t pte = HSW_PTE_ADDR_ENCODE(addr) | GEN6_PTE_VALID; - switch (level) { + switch (pat_index) { case I915_CACHE_NONE: break; case I915_CACHE_WT: @@ -1326,6 +1326,9 @@ void i915_ggtt_resume(struct i915_ggtt *ggtt) ggtt->vm.scratch_range(&ggtt->vm, ggtt->error_capture.start, ggtt->error_capture.size); + list_for_each_entry(gt, &ggtt->gt_list, ggtt_link) + intel_uc_resume_mappings(>->uc); + ggtt->invalidate(ggtt); if (flush) diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt_gmch.c b/drivers/gpu/drm/i915/gt/intel_ggtt_gmch.c index d6a74ae2527bd9c6bba8d9ff5176570e2077b2a0..866c416afb7348226543157c46b646834336d10b 100644 --- a/drivers/gpu/drm/i915/gt/intel_ggtt_gmch.c +++ b/drivers/gpu/drm/i915/gt/intel_ggtt_gmch.c @@ -18,10 +18,10 @@ static void gmch_ggtt_insert_page(struct i915_address_space *vm, dma_addr_t addr, u64 offset, - enum i915_cache_level cache_level, + unsigned int pat_index, u32 unused) { - unsigned int flags = (cache_level == I915_CACHE_NONE) ? + unsigned int flags = (pat_index == I915_CACHE_NONE) ? AGP_USER_MEMORY : AGP_USER_CACHED_MEMORY; intel_gmch_gtt_insert_page(addr, offset >> PAGE_SHIFT, flags); @@ -29,10 +29,10 @@ static void gmch_ggtt_insert_page(struct i915_address_space *vm, static void gmch_ggtt_insert_entries(struct i915_address_space *vm, struct i915_vma_resource *vma_res, - enum i915_cache_level cache_level, + unsigned int pat_index, u32 unused) { - unsigned int flags = (cache_level == I915_CACHE_NONE) ? + unsigned int flags = (pat_index == I915_CACHE_NONE) ? AGP_USER_MEMORY : AGP_USER_CACHED_MEMORY; intel_gmch_gtt_insert_sg_entries(vma_res->bi.pages, vma_res->start >> PAGE_SHIFT, diff --git a/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c b/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c index cadfd85785b12870430c16fedcd7d5ddc0e8f71b..86b5a9ba323d55ea0061f94fa95971f7339136dd 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c @@ -88,10 +88,11 @@ static void pool_free_work(struct work_struct *wrk) { struct intel_gt_buffer_pool *pool = container_of(wrk, typeof(*pool), work.work); + struct intel_gt *gt = container_of(pool, struct intel_gt, buffer_pool); if (pool_free_older_than(pool, HZ)) - schedule_delayed_work(&pool->work, - round_jiffies_up_relative(HZ)); + queue_delayed_work(gt->i915->unordered_wq, &pool->work, + round_jiffies_up_relative(HZ)); } static void pool_retire(struct i915_active *ref) @@ -99,6 +100,7 @@ static void pool_retire(struct i915_active *ref) struct intel_gt_buffer_pool_node *node = container_of(ref, typeof(*node), active); struct intel_gt_buffer_pool *pool = node->pool; + struct intel_gt *gt = container_of(pool, struct intel_gt, buffer_pool); struct list_head *list = bucket_for_size(pool, node->obj->base.size); unsigned long flags; @@ -116,8 +118,8 @@ static void pool_retire(struct i915_active *ref) WRITE_ONCE(node->age, jiffies ?: 1); /* 0 reserved for active nodes */ spin_unlock_irqrestore(&pool->lock, flags); - schedule_delayed_work(&pool->work, - round_jiffies_up_relative(HZ)); + queue_delayed_work(gt->i915->unordered_wq, &pool->work, + round_jiffies_up_relative(HZ)); } void intel_gt_buffer_pool_mark_used(struct intel_gt_buffer_pool_node *node) diff --git a/drivers/gpu/drm/i915/gt/intel_gt_irq.c b/drivers/gpu/drm/i915/gt/intel_gt_irq.c index acec6566b914592fb43b537fc8e648f38f54a7be..62fd00c9e5192a0e8da971119eb9fe836384592d 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_irq.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_irq.c @@ -7,6 +7,7 @@ #include "i915_drv.h" #include "i915_irq.h" +#include "i915_reg.h" #include "intel_breadcrumbs.h" #include "intel_gt.h" #include "intel_gt_irq.h" @@ -375,7 +376,7 @@ static void gen7_parity_error_irq_handler(struct intel_gt *gt, u32 iir) if (iir & GT_RENDER_L3_PARITY_ERROR_INTERRUPT) gt->i915->l3_parity.which_slice |= 1 << 0; - schedule_work(>->i915->l3_parity.error_work); + queue_work(gt->i915->unordered_wq, >->i915->l3_parity.error_work); } void gen6_gt_irq_handler(struct intel_gt *gt, u32 gt_iir) diff --git a/drivers/gpu/drm/i915/gt/intel_gt_pm.c b/drivers/gpu/drm/i915/gt/intel_gt_pm.c index c2e69bafd02b066e4e6695a61b9b10ef33361f6e..5a942af0a14e63f8201abaf238e6cf6da592fead 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_pm.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_pm.c @@ -137,7 +137,7 @@ void intel_gt_pm_init_early(struct intel_gt *gt) * runtime_pm is per-device rather than per-tile, so this is still the * correct structure. */ - intel_wakeref_init(>->wakeref, >->i915->runtime_pm, &wf_ops); + intel_wakeref_init(>->wakeref, gt->i915, &wf_ops); seqcount_mutex_init(>->stats.lock, >->wakeref.mutex); } diff --git a/drivers/gpu/drm/i915/gt/intel_gt_regs.h b/drivers/gpu/drm/i915/gt/intel_gt_regs.h index b8a39c219b60eec3cd8bcedfc31e15ab6b05b332..718cb2c80f79ee2d4c61e2792deeda526d99d64b 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_regs.h +++ b/drivers/gpu/drm/i915/gt/intel_gt_regs.h @@ -7,7 +7,8 @@ #define __INTEL_GT_REGS__ #include "i915_reg_defs.h" -#include "display/intel_display_reg_defs.h" /* VLV_DISPLAY_BASE */ + +#define VLV_GUNIT_BASE 0x180000 /* * The perf control registers are technically multicast registers, but the @@ -1469,7 +1470,7 @@ #define GEN12_RCU_MODE _MMIO(0x14800) #define GEN12_RCU_MODE_CCS_ENABLE REG_BIT(0) -#define CHV_FUSE_GT _MMIO(VLV_DISPLAY_BASE + 0x2168) +#define CHV_FUSE_GT _MMIO(VLV_GUNIT_BASE + 0x2168) #define CHV_FGT_DISABLE_SS0 (1 << 10) #define CHV_FGT_DISABLE_SS1 (1 << 11) #define CHV_FGT_EU_DIS_SS0_R0_SHIFT 16 diff --git a/drivers/gpu/drm/i915/gt/intel_gt_requests.c b/drivers/gpu/drm/i915/gt/intel_gt_requests.c index 1dfd01668c79c638aa525d9f6a224b904bf381f4..d1a382dfaa1dfc6fec971e12118ea2f7401523c3 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_requests.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_requests.c @@ -116,7 +116,7 @@ void intel_engine_add_retire(struct intel_engine_cs *engine, GEM_BUG_ON(intel_engine_is_virtual(engine)); if (add_retire(engine, tl)) - schedule_work(&engine->retire_work); + queue_work(engine->i915->unordered_wq, &engine->retire_work); } void intel_engine_init_retire(struct intel_engine_cs *engine) @@ -207,8 +207,8 @@ static void retire_work_handler(struct work_struct *work) struct intel_gt *gt = container_of(work, typeof(*gt), requests.retire_work.work); - schedule_delayed_work(>->requests.retire_work, - round_jiffies_up_relative(HZ)); + queue_delayed_work(gt->i915->unordered_wq, >->requests.retire_work, + round_jiffies_up_relative(HZ)); intel_gt_retire_requests(gt); } @@ -224,8 +224,8 @@ void intel_gt_park_requests(struct intel_gt *gt) void intel_gt_unpark_requests(struct intel_gt *gt) { - schedule_delayed_work(>->requests.retire_work, - round_jiffies_up_relative(HZ)); + queue_delayed_work(gt->i915->unordered_wq, >->requests.retire_work, + round_jiffies_up_relative(HZ)); } void intel_gt_fini_requests(struct intel_gt *gt) diff --git a/drivers/gpu/drm/i915/gt/intel_reset.c b/drivers/gpu/drm/i915/gt/intel_reset.c index 797ea8340467ee5edc4cc1944d756c8f915b5729..e2152f75ba2effd1a67c8664a86119a2eec482df 100644 --- a/drivers/gpu/drm/i915/gt/intel_reset.c +++ b/drivers/gpu/drm/i915/gt/intel_reset.c @@ -7,7 +7,7 @@ #include #include -#include "display/intel_display.h" +#include "display/intel_display_reset.h" #include "display/intel_overlay.h" #include "gem/i915_gem_context.h" @@ -20,6 +20,7 @@ #include "i915_file_private.h" #include "i915_gpu_error.h" #include "i915_irq.h" +#include "i915_reg.h" #include "intel_breadcrumbs.h" #include "intel_engine_pm.h" #include "intel_engine_regs.h" @@ -1370,11 +1371,11 @@ static void intel_gt_reset_global(struct intel_gt *gt, /* Use a watchdog to ensure that our reset completes */ intel_wedge_on_timeout(&w, gt, 60 * HZ) { - intel_display_prepare_reset(gt->i915); + intel_display_reset_prepare(gt->i915); intel_gt_reset(gt, engine_mask, reason); - intel_display_finish_reset(gt->i915); + intel_display_reset_finish(gt->i915); } if (!test_bit(I915_WEDGED, >->reset.flags)) @@ -1624,7 +1625,7 @@ void __intel_init_wedge(struct intel_wedge_me *w, w->name = name; INIT_DELAYED_WORK_ONSTACK(&w->work, intel_wedge_me); - schedule_delayed_work(&w->work, timeout); + queue_delayed_work(gt->i915->unordered_wq, &w->work, timeout); } void __intel_fini_wedge(struct intel_wedge_me *w) diff --git a/drivers/gpu/drm/i915/gt/intel_rps.c b/drivers/gpu/drm/i915/gt/intel_rps.c index b2671ac59dc04a7b2851ce1e370ef6f3eae4e55d..e92e626d4994e8d023e7108a5a23546d72ee713f 100644 --- a/drivers/gpu/drm/i915/gt/intel_rps.c +++ b/drivers/gpu/drm/i915/gt/intel_rps.c @@ -8,8 +8,10 @@ #include #include "display/intel_display.h" +#include "display/intel_display_irq.h" #include "i915_drv.h" #include "i915_irq.h" +#include "i915_reg.h" #include "intel_breadcrumbs.h" #include "intel_gt.h" #include "intel_gt_clock_utils.h" @@ -71,13 +73,14 @@ static void set(struct intel_uncore *uncore, i915_reg_t reg, u32 val) static void rps_timer(struct timer_list *t) { struct intel_rps *rps = from_timer(rps, t, timer); + struct intel_gt *gt = rps_to_gt(rps); struct intel_engine_cs *engine; ktime_t dt, last, timestamp; enum intel_engine_id id; s64 max_busy[3] = {}; timestamp = 0; - for_each_engine(engine, rps_to_gt(rps), id) { + for_each_engine(engine, gt, id) { s64 busy; int i; @@ -121,7 +124,7 @@ static void rps_timer(struct timer_list *t) busy += div_u64(max_busy[i], 1 << i); } - GT_TRACE(rps_to_gt(rps), + GT_TRACE(gt, "busy:%lld [%d%%], max:[%lld, %lld, %lld], interval:%d\n", busy, (int)div64_u64(100 * busy, dt), max_busy[0], max_busy[1], max_busy[2], @@ -131,12 +134,12 @@ static void rps_timer(struct timer_list *t) rps->cur_freq < rps->max_freq_softlimit) { rps->pm_iir |= GEN6_PM_RP_UP_THRESHOLD; rps->pm_interval = 1; - schedule_work(&rps->work); + queue_work(gt->i915->unordered_wq, &rps->work); } else if (100 * busy < rps->power.down_threshold * dt && rps->cur_freq > rps->min_freq_softlimit) { rps->pm_iir |= GEN6_PM_RP_DOWN_THRESHOLD; rps->pm_interval = 1; - schedule_work(&rps->work); + queue_work(gt->i915->unordered_wq, &rps->work); } else { rps->last_adj = 0; } @@ -971,7 +974,7 @@ static int rps_set_boost_freq(struct intel_rps *rps, u32 val) } mutex_unlock(&rps->lock); if (boost) - schedule_work(&rps->work); + queue_work(rps_to_gt(rps)->i915->unordered_wq, &rps->work); return 0; } @@ -1023,7 +1026,8 @@ void intel_rps_boost(struct i915_request *rq) if (!atomic_fetch_inc(&slpc->num_waiters)) { GT_TRACE(rps_to_gt(rps), "boost fence:%llx:%llx\n", rq->fence.context, rq->fence.seqno); - schedule_work(&slpc->boost_work); + queue_work(rps_to_gt(rps)->i915->unordered_wq, + &slpc->boost_work); } return; @@ -1039,7 +1043,7 @@ void intel_rps_boost(struct i915_request *rq) rq->fence.context, rq->fence.seqno); if (READ_ONCE(rps->cur_freq) < rps->boost_freq) - schedule_work(&rps->work); + queue_work(rps_to_gt(rps)->i915->unordered_wq, &rps->work); WRITE_ONCE(rps->boosts, rps->boosts + 1); /* debug only */ } @@ -1898,7 +1902,7 @@ void gen11_rps_irq_handler(struct intel_rps *rps, u32 pm_iir) gen6_gt_pm_mask_irq(gt, events); rps->pm_iir |= events; - schedule_work(&rps->work); + queue_work(gt->i915->unordered_wq, &rps->work); } void gen6_rps_irq_handler(struct intel_rps *rps, u32 pm_iir) @@ -1915,7 +1919,7 @@ void gen6_rps_irq_handler(struct intel_rps *rps, u32 pm_iir) gen6_gt_pm_mask_irq(gt, events); rps->pm_iir |= events; - schedule_work(&rps->work); + queue_work(gt->i915->unordered_wq, &rps->work); spin_unlock(gt->irq_lock); } diff --git a/drivers/gpu/drm/i915/gt/selftest_engine_cs.c b/drivers/gpu/drm/i915/gt/selftest_engine_cs.c index 542ce6d2de1922386c9096fe217bfb38686981e4..78cdfc6f315f2a9cb20105f911972d99794ca71e 100644 --- a/drivers/gpu/drm/i915/gt/selftest_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/selftest_engine_cs.c @@ -27,7 +27,7 @@ static void perf_begin(struct intel_gt *gt) /* Boost gpufreq to max [waitboost] and keep it fixed */ atomic_inc(>->rps.num_waiters); - schedule_work(>->rps.work); + queue_work(gt->i915->unordered_wq, >->rps.work); flush_work(>->rps.work); } diff --git a/drivers/gpu/drm/i915/gt/selftest_execlists.c b/drivers/gpu/drm/i915/gt/selftest_execlists.c index 736b89a8ecf54fa5dffdb1e349d87c748adfe20e..4202df5b8c122f9a2edb9203bfe7f5eb882cea3b 100644 --- a/drivers/gpu/drm/i915/gt/selftest_execlists.c +++ b/drivers/gpu/drm/i915/gt/selftest_execlists.c @@ -1530,8 +1530,8 @@ static int live_busywait_preempt(void *arg) struct drm_i915_gem_object *obj; struct i915_vma *vma; enum intel_engine_id id; - int err = -ENOMEM; u32 *map; + int err; /* * Verify that even without HAS_LOGICAL_RING_PREEMPTION, we can @@ -1539,13 +1539,17 @@ static int live_busywait_preempt(void *arg) */ ctx_hi = kernel_context(gt->i915, NULL); - if (!ctx_hi) - return -ENOMEM; + if (IS_ERR(ctx_hi)) + return PTR_ERR(ctx_hi); + ctx_hi->sched.priority = I915_CONTEXT_MAX_USER_PRIORITY; ctx_lo = kernel_context(gt->i915, NULL); - if (!ctx_lo) + if (IS_ERR(ctx_lo)) { + err = PTR_ERR(ctx_lo); goto err_ctx_hi; + } + ctx_lo->sched.priority = I915_CONTEXT_MIN_USER_PRIORITY; obj = i915_gem_object_create_internal(gt->i915, PAGE_SIZE); diff --git a/drivers/gpu/drm/i915/gt/selftest_rps.c b/drivers/gpu/drm/i915/gt/selftest_rps.c index 84e77e8dbba19c5e543bcf8f63c3b282530b38af..fb30f733b03665eae52a5cdb24d4e2802b02ecc4 100644 --- a/drivers/gpu/drm/i915/gt/selftest_rps.c +++ b/drivers/gpu/drm/i915/gt/selftest_rps.c @@ -8,6 +8,7 @@ #include "gem/i915_gem_internal.h" +#include "i915_reg.h" #include "intel_engine_heartbeat.h" #include "intel_engine_pm.h" #include "intel_engine_regs.h" diff --git a/drivers/gpu/drm/i915/gt/selftest_tlb.c b/drivers/gpu/drm/i915/gt/selftest_tlb.c index 4493c8518e91b214ba9460b5ac2ff88b92a20e1a..3bd6b540257b46c89b65de46a992e5eb9c3a0d06 100644 --- a/drivers/gpu/drm/i915/gt/selftest_tlb.c +++ b/drivers/gpu/drm/i915/gt/selftest_tlb.c @@ -190,11 +190,18 @@ pte_tlbinv(struct intel_context *ce, static struct drm_i915_gem_object *create_lmem(struct intel_gt *gt) { + struct intel_memory_region *mr = gt->i915->mm.regions[INTEL_REGION_LMEM_0]; + resource_size_t size = SZ_1G; + /* * Allocation of largest possible page size allows to test all types - * of pages. + * of pages. To succeed with both allocations, especially in case of Small + * BAR, try to allocate no more than quarter of mappable memory. */ - return i915_gem_object_create_lmem(gt->i915, SZ_1G, I915_BO_ALLOC_CONTIGUOUS); + if (mr && size > mr->io_size / 4) + size = mr->io_size / 4; + + return i915_gem_object_create_lmem(gt->i915, size, I915_BO_ALLOC_CONTIGUOUS); } static struct drm_i915_gem_object *create_smem(struct intel_gt *gt) diff --git a/drivers/gpu/drm/i915/gt/uc/abi/guc_communication_ctb_abi.h b/drivers/gpu/drm/i915/gt/uc/abi/guc_communication_ctb_abi.h index 28b8387f97b77f129f757728a5a7842f9fa93d13..f7d70db16d76027a6b65c5974daeef6eccbb2fdf 100644 --- a/drivers/gpu/drm/i915/gt/uc/abi/guc_communication_ctb_abi.h +++ b/drivers/gpu/drm/i915/gt/uc/abi/guc_communication_ctb_abi.h @@ -167,25 +167,4 @@ static_assert(sizeof(struct guc_ct_buffer_desc) == 64); * - **flags**, holds various bits to control message handling */ -/* - * Definition of the command transport message header (DW0) - * - * bit[4..0] message len (in dwords) - * bit[7..5] reserved - * bit[8] response (G2H only) - * bit[8] write fence to desc (H2G only) - * bit[9] write status to H2G buff (H2G only) - * bit[10] send status back via G2H (H2G only) - * bit[15..11] reserved - * bit[31..16] action code - */ -#define GUC_CT_MSG_LEN_SHIFT 0 -#define GUC_CT_MSG_LEN_MASK 0x1F -#define GUC_CT_MSG_IS_RESPONSE (1 << 8) -#define GUC_CT_MSG_WRITE_FENCE_TO_DESC (1 << 8) -#define GUC_CT_MSG_WRITE_STATUS_TO_BUFF (1 << 9) -#define GUC_CT_MSG_SEND_STATUS (1 << 10) -#define GUC_CT_MSG_ACTION_SHIFT 16 -#define GUC_CT_MSG_ACTION_MASK 0xFFFF - #endif /* _ABI_GUC_COMMUNICATION_CTB_ABI_H */ diff --git a/drivers/gpu/drm/i915/gt/uc/abi/guc_messages_abi.h b/drivers/gpu/drm/i915/gt/uc/abi/guc_messages_abi.h index 7d5ba4d97d708a8f186dbeeef3231f1c1f8907a7..98eb4f46572b940f671c358a3b5f5d55d934f42b 100644 --- a/drivers/gpu/drm/i915/gt/uc/abi/guc_messages_abi.h +++ b/drivers/gpu/drm/i915/gt/uc/abi/guc_messages_abi.h @@ -24,6 +24,7 @@ * | | 30:28 | **TYPE** - message type | * | | | - _`GUC_HXG_TYPE_REQUEST` = 0 | * | | | - _`GUC_HXG_TYPE_EVENT` = 1 | + * | | | - _`GUC_HXG_TYPE_FAST_REQUEST` = 2 | * | | | - _`GUC_HXG_TYPE_NO_RESPONSE_BUSY` = 3 | * | | | - _`GUC_HXG_TYPE_NO_RESPONSE_RETRY` = 5 | * | | | - _`GUC_HXG_TYPE_RESPONSE_FAILURE` = 6 | @@ -46,6 +47,7 @@ #define GUC_HXG_MSG_0_TYPE (0x7 << 28) #define GUC_HXG_TYPE_REQUEST 0u #define GUC_HXG_TYPE_EVENT 1u +#define GUC_HXG_TYPE_FAST_REQUEST 2u #define GUC_HXG_TYPE_NO_RESPONSE_BUSY 3u #define GUC_HXG_TYPE_NO_RESPONSE_RETRY 5u #define GUC_HXG_TYPE_RESPONSE_FAILURE 6u @@ -89,6 +91,34 @@ #define GUC_HXG_REQUEST_MSG_0_ACTION (0xffff << 0) #define GUC_HXG_REQUEST_MSG_n_DATAn GUC_HXG_MSG_n_PAYLOAD +/** + * DOC: HXG Fast Request + * + * The `HXG Request`_ message should be used to initiate asynchronous activity + * for which confirmation or return data is not expected. + * + * If confirmation is required then `HXG Request`_ shall be used instead. + * + * The recipient of this message may only use `HXG Failure`_ message if it was + * unable to accept this request (like invalid data). + * + * Format of `HXG Fast Request`_ message is same as `HXG Request`_ except @TYPE. + * + * +---+-------+--------------------------------------------------------------+ + * | | Bits | Description | + * +===+=======+==============================================================+ + * | 0 | 31 | ORIGIN - see `HXG Message`_ | + * | +-------+--------------------------------------------------------------+ + * | | 30:28 | TYPE = `GUC_HXG_TYPE_FAST_REQUEST`_ | + * | +-------+--------------------------------------------------------------+ + * | | 27:16 | DATA0 - see `HXG Request`_ | + * | +-------+--------------------------------------------------------------+ + * | | 15:0 | ACTION - see `HXG Request`_ | + * +---+-------+--------------------------------------------------------------+ + * |...| | DATAn - see `HXG Request`_ | + * +---+-------+--------------------------------------------------------------+ + */ + /** * DOC: HXG Event * diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_binary_headers.h b/drivers/gpu/drm/i915/gt/uc/intel_gsc_binary_headers.h new file mode 100644 index 0000000000000000000000000000000000000000..714f0c256118c3923f2e34aa7f7d1328757466c5 --- /dev/null +++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_binary_headers.h @@ -0,0 +1,74 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2023 Intel Corporation + */ + +#ifndef _INTEL_GSC_BINARY_HEADERS_H_ +#define _INTEL_GSC_BINARY_HEADERS_H_ + +#include + +/* Code partition directory (CPD) structures */ +struct intel_gsc_cpd_header_v2 { + u32 header_marker; +#define INTEL_GSC_CPD_HEADER_MARKER 0x44504324 + + u32 num_of_entries; + u8 header_version; + u8 entry_version; + u8 header_length; /* in bytes */ + u8 flags; + u32 partition_name; + u32 crc32; +} __packed; + +struct intel_gsc_cpd_entry { + u8 name[12]; + + /* + * Bits 0-24: offset from the beginning of the code partition + * Bit 25: huffman compressed + * Bits 26-31: reserved + */ + u32 offset; +#define INTEL_GSC_CPD_ENTRY_OFFSET_MASK GENMASK(24, 0) +#define INTEL_GSC_CPD_ENTRY_HUFFMAN_COMP BIT(25) + + /* + * Module/Item length, in bytes. For Huffman-compressed modules, this + * refers to the uncompressed size. For software-compressed modules, + * this refers to the compressed size. + */ + u32 length; + + u8 reserved[4]; +} __packed; + +struct intel_gsc_version { + u16 major; + u16 minor; + u16 hotfix; + u16 build; +} __packed; + +struct intel_gsc_manifest_header { + u32 header_type; /* 0x4 for manifest type */ + u32 header_length; /* in dwords */ + u32 header_version; + u32 flags; + u32 vendor; + u32 date; + u32 size; /* In dwords, size of entire manifest (header + extensions) */ + u32 header_id; + u32 internal_data; + struct intel_gsc_version fw_version; + u32 security_version; + struct intel_gsc_version meu_kit_version; + u32 meu_manifest_version; + u8 general_data[4]; + u8 reserved3[56]; + u32 modulus_size; /* in dwords */ + u32 exponent_size; /* in dwords */ +} __packed; + +#endif diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c index ebee0b5a2c1dd0fa300001c0b363ecf463ea7ea5..5f138de3c14f93c2d9ee89079714e2c22c7b0269 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c @@ -5,8 +5,8 @@ #include -#include "drm/i915_component.h" -#include "drm/i915_gsc_proxy_mei_interface.h" +#include +#include #include "gt/intel_gt.h" #include "gt/intel_gt_print.h" diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c index fb0984f875f9e24913760496ecaeafbff40c55a0..c659cc01f32f643302d0284593795358b220ea3c 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c @@ -29,13 +29,32 @@ static void gsc_work(struct work_struct *work) if (actions & GSC_ACTION_FW_LOAD) { ret = intel_gsc_uc_fw_upload(gsc); - if (ret == -EEXIST) /* skip proxy if not a new load */ - actions &= ~GSC_ACTION_FW_LOAD; - else if (ret) + if (!ret) + /* setup proxy on a new load */ + actions |= GSC_ACTION_SW_PROXY; + else if (ret != -EEXIST) goto out_put; + + /* + * The HuC auth can be done both before or after the proxy init; + * if done after, a proxy request will be issued and must be + * serviced before the authentication can complete. + * Since this worker also handles proxy requests, we can't + * perform an action that requires the proxy from within it and + * then stall waiting for it, because we'd be blocking the + * service path. Therefore, it is easier for us to load HuC + * first and do proxy later. The GSC will ack the HuC auth and + * then send the HuC proxy request as part of the proxy init + * flow. + * Note that we can only do the GSC auth if the GuC auth was + * successful. + */ + if (intel_uc_uses_huc(>->uc) && + intel_huc_is_authenticated(>->uc.huc, INTEL_HUC_AUTH_BY_GUC)) + intel_huc_auth(>->uc.huc, INTEL_HUC_AUTH_BY_GSC); } - if (actions & (GSC_ACTION_FW_LOAD | GSC_ACTION_SW_PROXY)) { + if (actions & GSC_ACTION_SW_PROXY) { if (!intel_gsc_uc_fw_init_done(gsc)) { gt_err(gt, "Proxy request received with GSC not loaded!\n"); goto out_put; @@ -90,7 +109,12 @@ void intel_gsc_uc_init_early(struct intel_gsc_uc *gsc) { struct intel_gt *gt = gsc_uc_to_gt(gsc); - intel_uc_fw_init_early(&gsc->fw, INTEL_UC_FW_TYPE_GSC); + /* + * GSC FW needs to be copied to a dedicated memory allocations for + * loading (see gsc->local), so we don't need to GGTT map the FW image + * itself into GGTT. + */ + intel_uc_fw_init_early(&gsc->fw, INTEL_UC_FW_TYPE_GSC, false); INIT_WORK(&gsc->work, gsc_work); /* we can arrive here from i915_driver_early_probe for primary diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c index 579c0f5a14384e03844cbdc7bf22a4bd0c9a18cb..89ed5ee9cdedca4187e318bd937992e26201275c 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c @@ -99,7 +99,7 @@ void intel_gsc_uc_heci_cmd_emit_mtl_header(struct intel_gsc_mtl_header *header, u64 host_session_id) { host_session_id &= ~HOST_SESSION_MASK; - if (heci_client_id == HECI_MEADDRESS_PXP) + if (host_session_id && heci_client_id == HECI_MEADDRESS_PXP) host_session_id |= HOST_SESSION_PXP_SINGLE; header->validity_marker = GSC_HECI_VALIDITY_MARKER; @@ -202,7 +202,7 @@ intel_gsc_uc_heci_cmd_submit_nonpriv(struct intel_gsc_uc *gsc, if (++trials < 10) goto retry; else - err = EAGAIN; + err = -EAGAIN; } } i915_gem_ww_ctx_fini(&ww); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc.c index f9bddaa876d9b8eac738cb4d3c9e5782f4663c47..2eb891b270aecaca84ee84a917f97e09e26f8e41 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc.c @@ -16,6 +16,7 @@ #include "intel_guc_submission.h" #include "i915_drv.h" #include "i915_irq.h" +#include "i915_reg.h" /** * DOC: GuC @@ -163,7 +164,7 @@ void intel_guc_init_early(struct intel_guc *guc) struct intel_gt *gt = guc_to_gt(guc); struct drm_i915_private *i915 = gt->i915; - intel_uc_fw_init_early(&guc->fw, INTEL_UC_FW_TYPE_GUC); + intel_uc_fw_init_early(&guc->fw, INTEL_UC_FW_TYPE_GUC, true); intel_guc_ct_init_early(&guc->ct); intel_guc_log_init_early(&guc->log); intel_guc_submission_init_early(guc); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c index 0ff864da92dfe16461328b859349330e5b2b6c6f..331cec07c1253e5e6ec6e9728333b80cb8d9413d 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c @@ -301,7 +301,6 @@ guc_capture_alloc_steered_lists(struct intel_guc *guc, const struct __guc_mmio_reg_descr_group *list; struct __guc_mmio_reg_descr_group *extlists; struct __guc_mmio_reg_descr *extarray; - struct sseu_dev_info *sseu; bool has_xehpg_extregs; /* steered registers currently only exist for the render-class */ @@ -318,7 +317,6 @@ guc_capture_alloc_steered_lists(struct intel_guc *guc, if (has_xehpg_extregs) num_steer_regs += ARRAY_SIZE(xehpg_extregs); - sseu = >->info.sseu; for_each_ss_steering(iter, gt, slice, subslice) num_tot_regs += num_steer_regs; diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c index a22e33f37cae6cff125fd283478c45656612c1fc..f28a3a83742dc45073c5207003e7ead9629de5d3 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c @@ -376,6 +376,24 @@ void intel_guc_ct_disable(struct intel_guc_ct *ct) } } +#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM) +static void ct_track_lost_and_found(struct intel_guc_ct *ct, u32 fence, u32 action) +{ + unsigned int lost = fence % ARRAY_SIZE(ct->requests.lost_and_found); +#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC) + unsigned long entries[SZ_32]; + unsigned int n; + + n = stack_trace_save(entries, ARRAY_SIZE(entries), 1); + + /* May be called under spinlock, so avoid sleeping */ + ct->requests.lost_and_found[lost].stack = stack_depot_save(entries, n, GFP_NOWAIT); +#endif + ct->requests.lost_and_found[lost].fence = fence; + ct->requests.lost_and_found[lost].action = action; +} +#endif + static u32 ct_get_next_fence(struct intel_guc_ct *ct) { /* For now it's trivial */ @@ -426,11 +444,11 @@ static int ct_write(struct intel_guc_ct *ct, FIELD_PREP(GUC_CTB_MSG_0_NUM_DWORDS, len) | FIELD_PREP(GUC_CTB_MSG_0_FENCE, fence); - type = (flags & INTEL_GUC_CT_SEND_NB) ? GUC_HXG_TYPE_EVENT : + type = (flags & INTEL_GUC_CT_SEND_NB) ? GUC_HXG_TYPE_FAST_REQUEST : GUC_HXG_TYPE_REQUEST; hxg = FIELD_PREP(GUC_HXG_MSG_0_TYPE, type) | - FIELD_PREP(GUC_HXG_EVENT_MSG_0_ACTION | - GUC_HXG_EVENT_MSG_0_DATA0, action[0]); + FIELD_PREP(GUC_HXG_REQUEST_MSG_0_ACTION | + GUC_HXG_REQUEST_MSG_0_DATA0, action[0]); CT_DEBUG(ct, "writing (tail %u) %*ph %*ph %*ph\n", tail, 4, &header, 4, &hxg, 4 * (len - 1), &action[1]); @@ -447,6 +465,11 @@ static int ct_write(struct intel_guc_ct *ct, } GEM_BUG_ON(tail > size); +#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM) + ct_track_lost_and_found(ct, fence, + FIELD_GET(GUC_HXG_EVENT_MSG_0_ACTION, action[0])); +#endif + /* * make sure H2G buffer update and LRC tail update (if this triggering a * submission) are visible before updating the descriptor tail @@ -675,7 +698,7 @@ static int ct_send(struct intel_guc_ct *ct, GEM_BUG_ON(!ct->enabled); GEM_BUG_ON(!len); - GEM_BUG_ON(len & ~GUC_CT_MSG_LEN_MASK); + GEM_BUG_ON(len > GUC_CTB_HXG_MSG_MAX_LEN - GUC_CTB_HDR_LEN); GEM_BUG_ON(!response_buf && response_buf_size); might_sleep(); @@ -953,6 +976,43 @@ static int ct_read(struct intel_guc_ct *ct, struct ct_incoming_msg **msg) return -EPIPE; } +#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM) +static bool ct_check_lost_and_found(struct intel_guc_ct *ct, u32 fence) +{ + unsigned int n; + char *buf = NULL; + bool found = false; + + lockdep_assert_held(&ct->requests.lock); + + for (n = 0; n < ARRAY_SIZE(ct->requests.lost_and_found); n++) { + if (ct->requests.lost_and_found[n].fence != fence) + continue; + found = true; + +#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC) + buf = kmalloc(SZ_4K, GFP_NOWAIT); + if (buf && stack_depot_snprint(ct->requests.lost_and_found[n].stack, + buf, SZ_4K, 0)) { + CT_ERROR(ct, "Fence %u was used by action %#04x sent at\n%s", + fence, ct->requests.lost_and_found[n].action, buf); + break; + } +#endif + CT_ERROR(ct, "Fence %u was used by action %#04x\n", + fence, ct->requests.lost_and_found[n].action); + break; + } + kfree(buf); + return found; +} +#else +static bool ct_check_lost_and_found(struct intel_guc_ct *ct, u32 fence) +{ + return false; +} +#endif + static int ct_handle_response(struct intel_guc_ct *ct, struct ct_incoming_msg *response) { u32 len = FIELD_GET(GUC_CTB_MSG_0_NUM_DWORDS, response->msg[0]); @@ -994,12 +1054,13 @@ static int ct_handle_response(struct intel_guc_ct *ct, struct ct_incoming_msg *r break; } if (!found) { - CT_ERROR(ct, "Unsolicited response (fence %u)\n", fence); - CT_ERROR(ct, "Could not find fence=%u, last_fence=%u\n", fence, - ct->requests.last_fence); - list_for_each_entry(req, &ct->requests.pending, link) - CT_ERROR(ct, "request %u awaits response\n", - req->fence); + CT_ERROR(ct, "Unsolicited response message: len %u, data %#x (fence %u, last %u)\n", + len, hxg[0], fence, ct->requests.last_fence); + if (!ct_check_lost_and_found(ct, fence)) { + list_for_each_entry(req, &ct->requests.pending, link) + CT_ERROR(ct, "request %u awaits response\n", + req->fence); + } err = -ENOKEY; } spin_unlock_irqrestore(&ct->requests.lock, flags); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h index 818415b64f4d1bc1097483e43717764707f4d9f1..58e42901ff498fae1715a3e274eaaf4a12c5091d 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h @@ -8,6 +8,7 @@ #include #include +#include #include #include #include @@ -81,6 +82,16 @@ struct intel_guc_ct { struct list_head incoming; /* incoming requests */ struct work_struct worker; /* handler for incoming requests */ + +#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM) + struct { + u16 fence; + u16 action; +#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC) + depot_stack_handle_t stack; +#endif + } lost_and_found[SZ_16]; +#endif } requests; /** @stall_time: time of first time a CTB submission is stalled */ diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h index 4e57bd09d50d9803eea7bd8d448daf9e9a72065d..b4d56eccfb1f0aff724ba93a98e57bf5cbe0449a 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_fwif.h @@ -35,13 +35,6 @@ #define GUC_MAX_CONTEXT_ID 65535 #define GUC_INVALID_CONTEXT_ID GUC_MAX_CONTEXT_ID -#define GUC_RENDER_ENGINE 0 -#define GUC_VIDEO_ENGINE 1 -#define GUC_BLITTER_ENGINE 2 -#define GUC_VIDEOENHANCE_ENGINE 3 -#define GUC_VIDEO_ENGINE2 4 -#define GUC_MAX_ENGINES_NUM (GUC_VIDEO_ENGINE2 + 1) - #define GUC_RENDER_CLASS 0 #define GUC_VIDEO_CLASS 1 #define GUC_VIDEOENHANCE_CLASS 2 @@ -499,32 +492,6 @@ struct guc_log_buffer_state { u32 version; } __packed; -struct guc_ctx_report { - u32 report_return_status; - u32 reserved1[64]; - u32 affected_count; - u32 reserved2[2]; -} __packed; - -/* GuC Shared Context Data Struct */ -struct guc_shared_ctx_data { - u32 addr_of_last_preempted_data_low; - u32 addr_of_last_preempted_data_high; - u32 addr_of_last_preempted_data_high_tmp; - u32 padding; - u32 is_mapped_to_proxy; - u32 proxy_ctx_id; - u32 engine_reset_ctx_id; - u32 media_reset_count; - u32 reserved1[8]; - u32 uk_last_ctx_switch_reason; - u32 was_reset; - u32 lrca_gpu_addr; - u64 execlist_ctx; - u32 reserved2[66]; - struct guc_ctx_report preempt_ctx_report[GUC_MAX_ENGINES_NUM]; -} __packed; - /* This action will be programmed in C1BC - SOFT_SCRATCH_15_REG */ enum intel_guc_recv_message { INTEL_GUC_RECV_MSG_CRASH_DUMP_POSTED = BIT(1), diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.c b/drivers/gpu/drm/i915/gt/uc/intel_huc.c index 04724ff56ded3faa05b403856e43b01671e76d86..e0afd8f895026cc3dae26e3e1230d234fc1cac60 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_huc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.c @@ -6,23 +6,16 @@ #include #include "gt/intel_gt.h" -#include "gt/intel_gt_print.h" #include "intel_guc_reg.h" #include "intel_huc.h" +#include "intel_huc_print.h" #include "i915_drv.h" +#include "i915_reg.h" +#include "pxp/intel_pxp_cmd_interface_43.h" #include #include -#define huc_printk(_huc, _level, _fmt, ...) \ - gt_##_level(huc_to_gt(_huc), "HuC: " _fmt, ##__VA_ARGS__) -#define huc_err(_huc, _fmt, ...) huc_printk((_huc), err, _fmt, ##__VA_ARGS__) -#define huc_warn(_huc, _fmt, ...) huc_printk((_huc), warn, _fmt, ##__VA_ARGS__) -#define huc_notice(_huc, _fmt, ...) huc_printk((_huc), notice, _fmt, ##__VA_ARGS__) -#define huc_info(_huc, _fmt, ...) huc_printk((_huc), info, _fmt, ##__VA_ARGS__) -#define huc_dbg(_huc, _fmt, ...) huc_printk((_huc), dbg, _fmt, ##__VA_ARGS__) -#define huc_probe_error(_huc, _fmt, ...) huc_printk((_huc), probe_error, _fmt, ##__VA_ARGS__) - /** * DOC: HuC * @@ -31,15 +24,23 @@ * capabilities by adding HuC specific commands to batch buffers. * * The kernel driver is only responsible for loading the HuC firmware and - * triggering its security authentication, which is performed by the GuC on - * older platforms and by the GSC on newer ones. For the GuC to correctly - * perform the authentication, the HuC binary must be loaded before the GuC one. + * triggering its security authentication. This is done differently depending + * on the platform: + * - older platforms (from Gen9 to most Gen12s): the load is performed via DMA + * and the authentication via GuC + * - DG2: load and authentication are both performed via GSC. + * - MTL and newer platforms: the load is performed via DMA (same as with + * not-DG2 older platforms), while the authentication is done in 2-steps, + * a first auth for clear-media workloads via GuC and a second one for all + * workloads via GSC. + * On platforms where the GuC does the authentication, to correctly do so the + * HuC binary must be loaded before the GuC one. * Loading the HuC is optional; however, not using the HuC might negatively * impact power usage and/or performance of media workloads, depending on the * use-cases. * HuC must be reloaded on events that cause the WOPCM to lose its contents - * (S3/S4, FLR); GuC-authenticated HuC must also be reloaded on GuC/GT reset, - * while GSC-managed HuC will survive that. + * (S3/S4, FLR); on older platforms the HuC must also be reloaded on GuC/GT + * reset, while on newer ones it will survive that. * * See https://github.com/intel/media-driver for the latest details on HuC * functionality. @@ -115,7 +116,7 @@ static enum hrtimer_restart huc_delayed_load_timer_callback(struct hrtimer *hrti { struct intel_huc *huc = container_of(hrtimer, struct intel_huc, delayed_load.timer); - if (!intel_huc_is_authenticated(huc)) { + if (!intel_huc_is_authenticated(huc, INTEL_HUC_AUTH_BY_GSC)) { if (huc->delayed_load.status == INTEL_HUC_WAITING_ON_GSC) huc_notice(huc, "timed out waiting for MEI GSC\n"); else if (huc->delayed_load.status == INTEL_HUC_WAITING_ON_PXP) @@ -133,7 +134,7 @@ static void huc_delayed_load_start(struct intel_huc *huc) { ktime_t delay; - GEM_BUG_ON(intel_huc_is_authenticated(huc)); + GEM_BUG_ON(intel_huc_is_authenticated(huc, INTEL_HUC_AUTH_BY_GSC)); /* * On resume we don't have to wait for MEI-GSC to be re-probed, but we @@ -276,7 +277,7 @@ void intel_huc_init_early(struct intel_huc *huc) struct drm_i915_private *i915 = huc_to_gt(huc)->i915; struct intel_gt *gt = huc_to_gt(huc); - intel_uc_fw_init_early(&huc->fw, INTEL_UC_FW_TYPE_HUC); + intel_uc_fw_init_early(&huc->fw, INTEL_UC_FW_TYPE_HUC, true); /* * we always init the fence as already completed, even if HuC is not @@ -293,13 +294,23 @@ void intel_huc_init_early(struct intel_huc *huc) } if (GRAPHICS_VER(i915) >= 11) { - huc->status.reg = GEN11_HUC_KERNEL_LOAD_INFO; - huc->status.mask = HUC_LOAD_SUCCESSFUL; - huc->status.value = HUC_LOAD_SUCCESSFUL; + huc->status[INTEL_HUC_AUTH_BY_GUC].reg = GEN11_HUC_KERNEL_LOAD_INFO; + huc->status[INTEL_HUC_AUTH_BY_GUC].mask = HUC_LOAD_SUCCESSFUL; + huc->status[INTEL_HUC_AUTH_BY_GUC].value = HUC_LOAD_SUCCESSFUL; + } else { + huc->status[INTEL_HUC_AUTH_BY_GUC].reg = HUC_STATUS2; + huc->status[INTEL_HUC_AUTH_BY_GUC].mask = HUC_FW_VERIFIED; + huc->status[INTEL_HUC_AUTH_BY_GUC].value = HUC_FW_VERIFIED; + } + + if (IS_DG2(i915)) { + huc->status[INTEL_HUC_AUTH_BY_GSC].reg = GEN11_HUC_KERNEL_LOAD_INFO; + huc->status[INTEL_HUC_AUTH_BY_GSC].mask = HUC_LOAD_SUCCESSFUL; + huc->status[INTEL_HUC_AUTH_BY_GSC].value = HUC_LOAD_SUCCESSFUL; } else { - huc->status.reg = HUC_STATUS2; - huc->status.mask = HUC_FW_VERIFIED; - huc->status.value = HUC_FW_VERIFIED; + huc->status[INTEL_HUC_AUTH_BY_GSC].reg = HECI_FWSTS5(MTL_GSC_HECI1_BASE); + huc->status[INTEL_HUC_AUTH_BY_GSC].mask = HECI_FWSTS5_HUC_AUTH_DONE; + huc->status[INTEL_HUC_AUTH_BY_GSC].value = HECI_FWSTS5_HUC_AUTH_DONE; } } @@ -307,51 +318,90 @@ void intel_huc_init_early(struct intel_huc *huc) static int check_huc_loading_mode(struct intel_huc *huc) { struct intel_gt *gt = huc_to_gt(huc); - bool fw_needs_gsc = intel_huc_is_loaded_by_gsc(huc); - bool hw_uses_gsc = false; + bool gsc_enabled = huc->fw.has_gsc_headers; /* * The fuse for HuC load via GSC is only valid on platforms that have * GuC deprivilege. */ if (HAS_GUC_DEPRIVILEGE(gt->i915)) - hw_uses_gsc = intel_uncore_read(gt->uncore, GUC_SHIM_CONTROL2) & - GSC_LOADS_HUC; + huc->loaded_via_gsc = intel_uncore_read(gt->uncore, GUC_SHIM_CONTROL2) & + GSC_LOADS_HUC; - if (fw_needs_gsc != hw_uses_gsc) { - huc_err(huc, "mismatch between FW (%s) and HW (%s) load modes\n", - HUC_LOAD_MODE_STRING(fw_needs_gsc), HUC_LOAD_MODE_STRING(hw_uses_gsc)); + if (huc->loaded_via_gsc && !gsc_enabled) { + huc_err(huc, "HW requires a GSC-enabled blob, but we found a legacy one\n"); return -ENOEXEC; } - /* make sure we can access the GSC via the mei driver if we need it */ - if (!(IS_ENABLED(CONFIG_INTEL_MEI_PXP) && IS_ENABLED(CONFIG_INTEL_MEI_GSC)) && - fw_needs_gsc) { - huc_info(huc, "can't load due to missing MEI modules\n"); - return -EIO; + /* + * On newer platforms we have GSC-enabled binaries but we load the HuC + * via DMA. To do so we need to find the location of the legacy-style + * binary inside the GSC-enabled one, which we do at fetch time. Make + * sure that we were able to do so if the fuse says we need to load via + * DMA and the binary is GSC-enabled. + */ + if (!huc->loaded_via_gsc && gsc_enabled && !huc->fw.dma_start_offset) { + huc_err(huc, "HW in DMA mode, but we have an incompatible GSC-enabled blob\n"); + return -ENOEXEC; + } + + /* + * If the HuC is loaded via GSC, we need to be able to access the GSC. + * On DG2 this is done via the mei components, while on newer platforms + * it is done via the GSCCS, + */ + if (huc->loaded_via_gsc) { + if (IS_DG2(gt->i915)) { + if (!IS_ENABLED(CONFIG_INTEL_MEI_PXP) || + !IS_ENABLED(CONFIG_INTEL_MEI_GSC)) { + huc_info(huc, "can't load due to missing mei modules\n"); + return -EIO; + } + } else { + if (!HAS_ENGINE(gt, GSC0)) { + huc_info(huc, "can't load due to missing GSCCS\n"); + return -EIO; + } + } } - huc_dbg(huc, "loaded by GSC = %s\n", str_yes_no(fw_needs_gsc)); + huc_dbg(huc, "loaded by GSC = %s\n", str_yes_no(huc->loaded_via_gsc)); return 0; } int intel_huc_init(struct intel_huc *huc) { + struct intel_gt *gt = huc_to_gt(huc); int err; err = check_huc_loading_mode(huc); if (err) goto out; + if (HAS_ENGINE(gt, GSC0)) { + struct i915_vma *vma; + + vma = intel_guc_allocate_vma(>->uc.guc, PXP43_HUC_AUTH_INOUT_SIZE * 2); + if (IS_ERR(vma)) { + huc_info(huc, "Failed to allocate heci pkt\n"); + goto out; + } + + huc->heci_pkt = vma; + } + err = intel_uc_fw_init(&huc->fw); if (err) - goto out; + goto out_pkt; intel_uc_fw_change_status(&huc->fw, INTEL_UC_FIRMWARE_LOADABLE); return 0; +out_pkt: + if (huc->heci_pkt) + i915_vma_unpin_and_release(&huc->heci_pkt, 0); out: intel_uc_fw_change_status(&huc->fw, INTEL_UC_FIRMWARE_INIT_FAIL); huc_info(huc, "initialization failed %pe\n", ERR_PTR(err)); @@ -366,6 +416,9 @@ void intel_huc_fini(struct intel_huc *huc) */ delayed_huc_load_fini(huc); + if (huc->heci_pkt) + i915_vma_unpin_and_release(&huc->heci_pkt, 0); + if (intel_uc_fw_is_loadable(&huc->fw)) intel_uc_fw_fini(&huc->fw); } @@ -383,34 +436,45 @@ void intel_huc_suspend(struct intel_huc *huc) delayed_huc_load_complete(huc); } -int intel_huc_wait_for_auth_complete(struct intel_huc *huc) +static const char *auth_mode_string(struct intel_huc *huc, + enum intel_huc_authentication_type type) +{ + bool partial = huc->fw.has_gsc_headers && type == INTEL_HUC_AUTH_BY_GUC; + + return partial ? "clear media" : "all workloads"; +} + +int intel_huc_wait_for_auth_complete(struct intel_huc *huc, + enum intel_huc_authentication_type type) { struct intel_gt *gt = huc_to_gt(huc); int ret; ret = __intel_wait_for_register(gt->uncore, - huc->status.reg, - huc->status.mask, - huc->status.value, + huc->status[type].reg, + huc->status[type].mask, + huc->status[type].value, 2, 50, NULL); /* mark the load process as complete even if the wait failed */ delayed_huc_load_complete(huc); if (ret) { - huc_err(huc, "firmware not verified %pe\n", ERR_PTR(ret)); + huc_err(huc, "firmware not verified for %s: %pe\n", + auth_mode_string(huc, type), ERR_PTR(ret)); intel_uc_fw_change_status(&huc->fw, INTEL_UC_FIRMWARE_LOAD_FAIL); return ret; } intel_uc_fw_change_status(&huc->fw, INTEL_UC_FIRMWARE_RUNNING); - huc_info(huc, "authenticated!\n"); + huc_info(huc, "authenticated for %s\n", auth_mode_string(huc, type)); return 0; } /** * intel_huc_auth() - Authenticate HuC uCode * @huc: intel_huc structure + * @type: authentication type (via GuC or via GSC) * * Called after HuC and GuC firmware loading during intel_uc_init_hw(). * @@ -418,7 +482,7 @@ int intel_huc_wait_for_auth_complete(struct intel_huc *huc) * passing the offset of the RSA signature to intel_guc_auth_huc(). It then * waits for up to 50ms for firmware verification ACK. */ -int intel_huc_auth(struct intel_huc *huc) +int intel_huc_auth(struct intel_huc *huc, enum intel_huc_authentication_type type) { struct intel_gt *gt = huc_to_gt(huc); struct intel_guc *guc = >->uc.guc; @@ -427,44 +491,67 @@ int intel_huc_auth(struct intel_huc *huc) if (!intel_uc_fw_is_loaded(&huc->fw)) return -ENOEXEC; - /* GSC will do the auth */ + /* GSC will do the auth with the load */ if (intel_huc_is_loaded_by_gsc(huc)) return -ENODEV; + if (intel_huc_is_authenticated(huc, type)) + return -EEXIST; + ret = i915_inject_probe_error(gt->i915, -ENXIO); if (ret) goto fail; - GEM_BUG_ON(intel_uc_fw_is_running(&huc->fw)); - - ret = intel_guc_auth_huc(guc, intel_guc_ggtt_offset(guc, huc->fw.rsa_data)); - if (ret) { - huc_err(huc, "authentication by GuC failed %pe\n", ERR_PTR(ret)); - goto fail; + switch (type) { + case INTEL_HUC_AUTH_BY_GUC: + ret = intel_guc_auth_huc(guc, intel_guc_ggtt_offset(guc, huc->fw.rsa_data)); + break; + case INTEL_HUC_AUTH_BY_GSC: + ret = intel_huc_fw_auth_via_gsccs(huc); + break; + default: + MISSING_CASE(type); + ret = -EINVAL; } + if (ret) + goto fail; /* Check authentication status, it should be done by now */ - ret = intel_huc_wait_for_auth_complete(huc); + ret = intel_huc_wait_for_auth_complete(huc, type); if (ret) goto fail; return 0; fail: - huc_probe_error(huc, "authentication failed %pe\n", ERR_PTR(ret)); + huc_probe_error(huc, "%s authentication failed %pe\n", + auth_mode_string(huc, type), ERR_PTR(ret)); return ret; } -bool intel_huc_is_authenticated(struct intel_huc *huc) +bool intel_huc_is_authenticated(struct intel_huc *huc, + enum intel_huc_authentication_type type) { struct intel_gt *gt = huc_to_gt(huc); intel_wakeref_t wakeref; u32 status = 0; with_intel_runtime_pm(gt->uncore->rpm, wakeref) - status = intel_uncore_read(gt->uncore, huc->status.reg); + status = intel_uncore_read(gt->uncore, huc->status[type].reg); + + return (status & huc->status[type].mask) == huc->status[type].value; +} + +static bool huc_is_fully_authenticated(struct intel_huc *huc) +{ + struct intel_uc_fw *huc_fw = &huc->fw; - return (status & huc->status.mask) == huc->status.value; + if (!huc_fw->has_gsc_headers) + return intel_huc_is_authenticated(huc, INTEL_HUC_AUTH_BY_GUC); + else if (intel_huc_is_loaded_by_gsc(huc) || HAS_ENGINE(huc_to_gt(huc), GSC0)) + return intel_huc_is_authenticated(huc, INTEL_HUC_AUTH_BY_GSC); + else + return false; } /** @@ -479,7 +566,9 @@ bool intel_huc_is_authenticated(struct intel_huc *huc) */ int intel_huc_check_status(struct intel_huc *huc) { - switch (__intel_uc_fw_status(&huc->fw)) { + struct intel_uc_fw *huc_fw = &huc->fw; + + switch (__intel_uc_fw_status(huc_fw)) { case INTEL_UC_FIRMWARE_NOT_SUPPORTED: return -ENODEV; case INTEL_UC_FIRMWARE_DISABLED: @@ -496,7 +585,17 @@ int intel_huc_check_status(struct intel_huc *huc) break; } - return intel_huc_is_authenticated(huc); + /* + * GSC-enabled binaries loaded via DMA are first partially + * authenticated by GuC and then fully authenticated by GSC + */ + if (huc_is_fully_authenticated(huc)) + return 1; /* full auth */ + else if (huc_fw->has_gsc_headers && !intel_huc_is_loaded_by_gsc(huc) && + intel_huc_is_authenticated(huc, INTEL_HUC_AUTH_BY_GUC)) + return 2; /* clear media only */ + else + return 0; } static bool huc_has_delayed_load(struct intel_huc *huc) @@ -510,7 +609,10 @@ void intel_huc_update_auth_status(struct intel_huc *huc) if (!intel_uc_fw_is_loadable(&huc->fw)) return; - if (intel_huc_is_authenticated(huc)) + if (!huc->fw.has_gsc_headers) + return; + + if (huc_is_fully_authenticated(huc)) intel_uc_fw_change_status(&huc->fw, INTEL_UC_FIRMWARE_RUNNING); else if (huc_has_delayed_load(huc)) @@ -543,5 +645,5 @@ void intel_huc_load_status(struct intel_huc *huc, struct drm_printer *p) with_intel_runtime_pm(gt->uncore->rpm, wakeref) drm_printf(p, "HuC status: 0x%08x\n", - intel_uncore_read(gt->uncore, huc->status.reg)); + intel_uncore_read(gt->uncore, huc->status[INTEL_HUC_AUTH_BY_GUC].reg)); } diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.h b/drivers/gpu/drm/i915/gt/uc/intel_huc.h index 0789184d81a2a53d9b4cf567316bb00d3b260f35..ba5cb08e9e7bf1edb27dfa310e91234f84241097 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_huc.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.h @@ -15,6 +15,7 @@ #include struct bus_type; +struct i915_vma; enum intel_huc_delayed_load_status { INTEL_HUC_WAITING_ON_GSC = 0, @@ -22,6 +23,12 @@ enum intel_huc_delayed_load_status { INTEL_HUC_DELAYED_LOAD_ERROR, }; +enum intel_huc_authentication_type { + INTEL_HUC_AUTH_BY_GUC = 0, + INTEL_HUC_AUTH_BY_GSC, + INTEL_HUC_AUTH_MAX_MODES +}; + struct intel_huc { /* Generic uC firmware management */ struct intel_uc_fw fw; @@ -31,7 +38,7 @@ struct intel_huc { i915_reg_t reg; u32 mask; u32 value; - } status; + } status[INTEL_HUC_AUTH_MAX_MODES]; struct { struct i915_sw_fence fence; @@ -39,6 +46,11 @@ struct intel_huc { struct notifier_block nb; enum intel_huc_delayed_load_status status; } delayed_load; + + /* for load via GSCCS */ + struct i915_vma *heci_pkt; + + bool loaded_via_gsc; }; int intel_huc_sanitize(struct intel_huc *huc); @@ -46,11 +58,13 @@ void intel_huc_init_early(struct intel_huc *huc); int intel_huc_init(struct intel_huc *huc); void intel_huc_fini(struct intel_huc *huc); void intel_huc_suspend(struct intel_huc *huc); -int intel_huc_auth(struct intel_huc *huc); -int intel_huc_wait_for_auth_complete(struct intel_huc *huc); +int intel_huc_auth(struct intel_huc *huc, enum intel_huc_authentication_type type); +int intel_huc_wait_for_auth_complete(struct intel_huc *huc, + enum intel_huc_authentication_type type); +bool intel_huc_is_authenticated(struct intel_huc *huc, + enum intel_huc_authentication_type type); int intel_huc_check_status(struct intel_huc *huc); void intel_huc_update_auth_status(struct intel_huc *huc); -bool intel_huc_is_authenticated(struct intel_huc *huc); void intel_huc_register_gsc_notifier(struct intel_huc *huc, const struct bus_type *bus); void intel_huc_unregister_gsc_notifier(struct intel_huc *huc, const struct bus_type *bus); @@ -73,13 +87,13 @@ static inline bool intel_huc_is_used(struct intel_huc *huc) static inline bool intel_huc_is_loaded_by_gsc(const struct intel_huc *huc) { - return huc->fw.loaded_via_gsc; + return huc->loaded_via_gsc; } static inline bool intel_huc_wait_required(struct intel_huc *huc) { return intel_huc_is_used(huc) && intel_huc_is_loaded_by_gsc(huc) && - !intel_huc_is_authenticated(huc); + !intel_huc_is_authenticated(huc, INTEL_HUC_AUTH_BY_GSC); } void intel_huc_load_status(struct intel_huc *huc, struct drm_printer *p); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c index 534b0aa433166bc5e8bf4137f2cbf0024a6dc3e2..e608152fecfcc55ae285c22344ec7d0557d1be3c 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c @@ -5,10 +5,241 @@ #include "gt/intel_gsc.h" #include "gt/intel_gt.h" +#include "intel_gsc_binary_headers.h" +#include "intel_gsc_uc_heci_cmd_submit.h" #include "intel_huc.h" #include "intel_huc_fw.h" +#include "intel_huc_print.h" #include "i915_drv.h" #include "pxp/intel_pxp_huc.h" +#include "pxp/intel_pxp_cmd_interface_43.h" + +struct mtl_huc_auth_msg_in { + struct intel_gsc_mtl_header header; + struct pxp43_new_huc_auth_in huc_in; +} __packed; + +struct mtl_huc_auth_msg_out { + struct intel_gsc_mtl_header header; + struct pxp43_huc_auth_out huc_out; +} __packed; + +int intel_huc_fw_auth_via_gsccs(struct intel_huc *huc) +{ + struct intel_gt *gt = huc_to_gt(huc); + struct drm_i915_private *i915 = gt->i915; + struct drm_i915_gem_object *obj; + struct mtl_huc_auth_msg_in *msg_in; + struct mtl_huc_auth_msg_out *msg_out; + void *pkt_vaddr; + u64 pkt_offset; + int retry = 5; + int err = 0; + + if (!huc->heci_pkt) + return -ENODEV; + + obj = huc->heci_pkt->obj; + pkt_offset = i915_ggtt_offset(huc->heci_pkt); + + pkt_vaddr = i915_gem_object_pin_map_unlocked(obj, + i915_coherent_map_type(i915, obj, true)); + if (IS_ERR(pkt_vaddr)) + return PTR_ERR(pkt_vaddr); + + msg_in = pkt_vaddr; + msg_out = pkt_vaddr + PXP43_HUC_AUTH_INOUT_SIZE; + + intel_gsc_uc_heci_cmd_emit_mtl_header(&msg_in->header, + HECI_MEADDRESS_PXP, + sizeof(*msg_in), 0); + + msg_in->huc_in.header.api_version = PXP_APIVER(4, 3); + msg_in->huc_in.header.command_id = PXP43_CMDID_NEW_HUC_AUTH; + msg_in->huc_in.header.status = 0; + msg_in->huc_in.header.buffer_len = sizeof(msg_in->huc_in) - + sizeof(msg_in->huc_in.header); + msg_in->huc_in.huc_base_address = huc->fw.vma_res.start; + msg_in->huc_in.huc_size = huc->fw.obj->base.size; + + do { + err = intel_gsc_uc_heci_cmd_submit_packet(>->uc.gsc, + pkt_offset, sizeof(*msg_in), + pkt_offset + PXP43_HUC_AUTH_INOUT_SIZE, + PXP43_HUC_AUTH_INOUT_SIZE); + if (err) { + huc_err(huc, "failed to submit GSC request to auth: %d\n", err); + goto out_unpin; + } + + if (msg_out->header.flags & GSC_OUTFLAG_MSG_PENDING) { + msg_in->header.gsc_message_handle = msg_out->header.gsc_message_handle; + err = -EBUSY; + msleep(50); + } + } while (--retry && err == -EBUSY); + + if (err) + goto out_unpin; + + if (msg_out->header.message_size != sizeof(*msg_out)) { + huc_err(huc, "invalid GSC reply length %u [expected %zu]\n", + msg_out->header.message_size, sizeof(*msg_out)); + err = -EPROTO; + goto out_unpin; + } + + /* + * The GSC will return PXP_STATUS_OP_NOT_PERMITTED if the HuC is already + * loaded. If the same error is ever returned with HuC not loaded we'll + * still catch it when we check the authentication bit later. + */ + if (msg_out->huc_out.header.status != PXP_STATUS_SUCCESS && + msg_out->huc_out.header.status != PXP_STATUS_OP_NOT_PERMITTED) { + huc_err(huc, "auth failed with GSC error = 0x%x\n", + msg_out->huc_out.header.status); + err = -EIO; + goto out_unpin; + } + +out_unpin: + i915_gem_object_unpin_map(obj); + return err; +} + +static void get_version_from_gsc_manifest(struct intel_uc_fw_ver *ver, const void *data) +{ + const struct intel_gsc_manifest_header *manifest = data; + + ver->major = manifest->fw_version.major; + ver->minor = manifest->fw_version.minor; + ver->patch = manifest->fw_version.hotfix; +} + +static bool css_valid(const void *data, size_t size) +{ + const struct uc_css_header *css = data; + + if (unlikely(size < sizeof(struct uc_css_header))) + return false; + + if (css->module_type != 0x6) + return false; + + if (css->module_vendor != PCI_VENDOR_ID_INTEL) + return false; + + return true; +} + +static inline u32 entry_offset(const struct intel_gsc_cpd_entry *entry) +{ + return entry->offset & INTEL_GSC_CPD_ENTRY_OFFSET_MASK; +} + +int intel_huc_fw_get_binary_info(struct intel_uc_fw *huc_fw, const void *data, size_t size) +{ + struct intel_huc *huc = container_of(huc_fw, struct intel_huc, fw); + const struct intel_gsc_cpd_header_v2 *header = data; + const struct intel_gsc_cpd_entry *entry; + size_t min_size = sizeof(*header); + int i; + + if (!huc_fw->has_gsc_headers) { + huc_err(huc, "Invalid FW type for GSC header parsing!\n"); + return -EINVAL; + } + + if (size < sizeof(*header)) { + huc_err(huc, "FW too small! %zu < %zu\n", size, min_size); + return -ENODATA; + } + + /* + * The GSC-enabled HuC binary starts with a directory header, followed + * by a series of entries. Each entry is identified by a name and + * points to a specific section of the binary containing the relevant + * data. The entries we're interested in are: + * - "HUCP.man": points to the GSC manifest header for the HuC, which + * contains the version info. + * - "huc_fw": points to the legacy-style binary that can be used for + * load via the DMA. This entry only contains a valid CSS + * on binaries for platforms that support 2-step HuC load + * via dma and auth via GSC (like MTL). + * + * -------------------------------------------------- + * [ intel_gsc_cpd_header_v2 ] + * -------------------------------------------------- + * [ intel_gsc_cpd_entry[] ] + * [ entry1 ] + * [ ... ] + * [ entryX ] + * [ "HUCP.man" ] + * [ ... ] + * [ offset >----------------------------]------o + * [ ... ] | + * [ entryY ] | + * [ "huc_fw" ] | + * [ ... ] | + * [ offset >----------------------------]----------o + * -------------------------------------------------- | | + * | | + * -------------------------------------------------- | | + * [ intel_gsc_manifest_header ]<-----o | + * [ ... ] | + * [ intel_gsc_version fw_version ] | + * [ ... ] | + * -------------------------------------------------- | + * | + * -------------------------------------------------- | + * [ data[] ]<---------o + * [ ... ] + * [ ... ] + * -------------------------------------------------- + */ + + if (header->header_marker != INTEL_GSC_CPD_HEADER_MARKER) { + huc_err(huc, "invalid marker for CPD header: 0x%08x!\n", + header->header_marker); + return -EINVAL; + } + + /* we only have binaries with header v2 and entry v1 for now */ + if (header->header_version != 2 || header->entry_version != 1) { + huc_err(huc, "invalid CPD header/entry version %u:%u!\n", + header->header_version, header->entry_version); + return -EINVAL; + } + + if (header->header_length < sizeof(struct intel_gsc_cpd_header_v2)) { + huc_err(huc, "invalid CPD header length %u!\n", + header->header_length); + return -EINVAL; + } + + min_size = header->header_length + sizeof(*entry) * header->num_of_entries; + if (size < min_size) { + huc_err(huc, "FW too small! %zu < %zu\n", size, min_size); + return -ENODATA; + } + + entry = data + header->header_length; + + for (i = 0; i < header->num_of_entries; i++, entry++) { + if (strcmp(entry->name, "HUCP.man") == 0) + get_version_from_gsc_manifest(&huc_fw->file_selected.ver, + data + entry_offset(entry)); + + if (strcmp(entry->name, "huc_fw") == 0) { + u32 offset = entry_offset(entry); + + if (offset < size && css_valid(data + offset, size - offset)) + huc_fw->dma_start_offset = offset; + } + } + + return 0; +} int intel_huc_fw_load_and_auth_via_gsc(struct intel_huc *huc) { @@ -25,7 +256,7 @@ int intel_huc_fw_load_and_auth_via_gsc(struct intel_huc *huc) * component gets re-bound and this function called again. If so, just * mark the HuC as loaded. */ - if (intel_huc_is_authenticated(huc)) { + if (intel_huc_is_authenticated(huc, INTEL_HUC_AUTH_BY_GSC)) { intel_uc_fw_change_status(&huc->fw, INTEL_UC_FIRMWARE_RUNNING); return 0; } @@ -38,7 +269,7 @@ int intel_huc_fw_load_and_auth_via_gsc(struct intel_huc *huc) intel_uc_fw_change_status(&huc->fw, INTEL_UC_FIRMWARE_TRANSFERRED); - return intel_huc_wait_for_auth_complete(huc); + return intel_huc_wait_for_auth_complete(huc, INTEL_HUC_AUTH_BY_GSC); } /** diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.h b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.h index db42e238b45f21d34307ed156a9eb960023118cf..307ab45e6b090a0499ccf036f956232b9c01c8f1 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.h @@ -7,8 +7,12 @@ #define _INTEL_HUC_FW_H_ struct intel_huc; +struct intel_uc_fw; + +#include int intel_huc_fw_load_and_auth_via_gsc(struct intel_huc *huc); +int intel_huc_fw_auth_via_gsccs(struct intel_huc *huc); int intel_huc_fw_upload(struct intel_huc *huc); - +int intel_huc_fw_get_binary_info(struct intel_uc_fw *huc_fw, const void *data, size_t size); #endif diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc_print.h b/drivers/gpu/drm/i915/gt/uc/intel_huc_print.h new file mode 100644 index 0000000000000000000000000000000000000000..915d310ee1df81c01a2b89db2fa53a2b11e5ce54 --- /dev/null +++ b/drivers/gpu/drm/i915/gt/uc/intel_huc_print.h @@ -0,0 +1,21 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2023 Intel Corporation + */ + +#ifndef __INTEL_HUC_PRINT__ +#define __INTEL_HUC_PRINT__ + +#include "gt/intel_gt.h" +#include "gt/intel_gt_print.h" + +#define huc_printk(_huc, _level, _fmt, ...) \ + gt_##_level(huc_to_gt(_huc), "HuC: " _fmt, ##__VA_ARGS__) +#define huc_err(_huc, _fmt, ...) huc_printk((_huc), err, _fmt, ##__VA_ARGS__) +#define huc_warn(_huc, _fmt, ...) huc_printk((_huc), warn, _fmt, ##__VA_ARGS__) +#define huc_notice(_huc, _fmt, ...) huc_printk((_huc), notice, _fmt, ##__VA_ARGS__) +#define huc_info(_huc, _fmt, ...) huc_printk((_huc), info, _fmt, ##__VA_ARGS__) +#define huc_dbg(_huc, _fmt, ...) huc_printk((_huc), dbg, _fmt, ##__VA_ARGS__) +#define huc_probe_error(_huc, _fmt, ...) huc_printk((_huc), probe_error, _fmt, ##__VA_ARGS__) + +#endif /* __INTEL_HUC_PRINT__ */ diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_uc.c index c8b9cbb7ba3a9a7bcdc079e3fb901ac788e30862..18250fb64bd881f09f0c1db0ff5a49cc9344f7c0 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc.c @@ -538,7 +538,7 @@ static int __uc_init_hw(struct intel_uc *uc) if (intel_huc_is_loaded_by_gsc(huc)) intel_huc_update_auth_status(huc); else - intel_huc_auth(huc); + intel_huc_auth(huc, INTEL_HUC_AUTH_BY_GUC); if (intel_uc_uses_guc_submission(uc)) { ret = intel_guc_submission_enable(guc); @@ -700,6 +700,12 @@ void intel_uc_suspend(struct intel_uc *uc) } } +static void __uc_resume_mappings(struct intel_uc *uc) +{ + intel_uc_fw_resume_mapping(&uc->guc.fw); + intel_uc_fw_resume_mapping(&uc->huc.fw); +} + static int __uc_resume(struct intel_uc *uc, bool enable_communication) { struct intel_guc *guc = &uc->guc; @@ -767,4 +773,6 @@ static const struct intel_uc_ops uc_ops_on = { .init_hw = __uc_init_hw, .fini_hw = __uc_fini_hw, + + .resume_mappings = __uc_resume_mappings, }; diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc.h b/drivers/gpu/drm/i915/gt/uc/intel_uc.h index d585524d94deb9a6ff733c9b070e34a0663229a8..014bb7d83689a9e1d9391f41348704eda16fc9dd 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc.h @@ -24,6 +24,7 @@ struct intel_uc_ops { void (*fini)(struct intel_uc *uc); int (*init_hw)(struct intel_uc *uc); void (*fini_hw)(struct intel_uc *uc); + void (*resume_mappings)(struct intel_uc *uc); }; struct intel_uc { @@ -114,6 +115,7 @@ intel_uc_ops_function(init, init, int, 0); intel_uc_ops_function(fini, fini, void, ); intel_uc_ops_function(init_hw, init_hw, int, 0); intel_uc_ops_function(fini_hw, fini_hw, void, ); +intel_uc_ops_function(resume_mappings, resume_mappings, void, ); #undef intel_uc_ops_function #endif diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c index dc5c96c503a9299aaa5fc3d6a54b9f5b0b5cc5c9..944725e624149695569e33dd97a3312e698573a7 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c @@ -108,6 +108,7 @@ void intel_uc_fw_change_status(struct intel_uc_fw *uc_fw, fw_def(SKYLAKE, 0, guc_mmp(skl, 70, 1, 1)) #define INTEL_HUC_FIRMWARE_DEFS(fw_def, huc_raw, huc_mmp, huc_gsc) \ + fw_def(METEORLAKE, 0, huc_gsc(mtl)) \ fw_def(DG2, 0, huc_gsc(dg2)) \ fw_def(ALDERLAKE_P, 0, huc_raw(tgl)) \ fw_def(ALDERLAKE_P, 0, huc_mmp(tgl, 7, 9, 3)) \ @@ -186,7 +187,7 @@ struct __packed uc_fw_blob { u8 major; u8 minor; u8 patch; - bool loaded_via_gsc; + bool has_gsc_headers; }; #define UC_FW_BLOB_BASE(major_, minor_, patch_, path_) \ @@ -197,7 +198,7 @@ struct __packed uc_fw_blob { #define UC_FW_BLOB_NEW(major_, minor_, patch_, gsc_, path_) \ { UC_FW_BLOB_BASE(major_, minor_, patch_, path_) \ - .legacy = false, .loaded_via_gsc = gsc_ } + .legacy = false, .has_gsc_headers = gsc_ } #define UC_FW_BLOB_OLD(major_, minor_, patch_, path_) \ { UC_FW_BLOB_BASE(major_, minor_, patch_, path_) \ @@ -310,7 +311,7 @@ __uc_fw_auto_select(struct drm_i915_private *i915, struct intel_uc_fw *uc_fw) uc_fw->file_wanted.ver.major = blob->major; uc_fw->file_wanted.ver.minor = blob->minor; uc_fw->file_wanted.ver.patch = blob->patch; - uc_fw->loaded_via_gsc = blob->loaded_via_gsc; + uc_fw->has_gsc_headers = blob->has_gsc_headers; found = true; break; } @@ -471,12 +472,14 @@ static void __uc_fw_user_override(struct drm_i915_private *i915, struct intel_uc * intel_uc_fw_init_early - initialize the uC object and select the firmware * @uc_fw: uC firmware * @type: type of uC + * @needs_ggtt_mapping: whether the FW needs to be GGTT mapped for loading * * Initialize the state of our uC object and relevant tracking and select the * firmware to fetch and load. */ void intel_uc_fw_init_early(struct intel_uc_fw *uc_fw, - enum intel_uc_fw_type type) + enum intel_uc_fw_type type, + bool needs_ggtt_mapping) { struct intel_gt *gt = ____uc_fw_to_gt(uc_fw, type); struct drm_i915_private *i915 = gt->i915; @@ -490,6 +493,7 @@ void intel_uc_fw_init_early(struct intel_uc_fw *uc_fw, GEM_BUG_ON(uc_fw->file_selected.path); uc_fw->type = type; + uc_fw->needs_ggtt_mapping = needs_ggtt_mapping; if (HAS_GT_UC(i915)) { if (!validate_fw_table_type(i915, type)) { @@ -545,33 +549,6 @@ static void __force_fw_fetch_failures(struct intel_uc_fw *uc_fw, int e) } } -static int check_gsc_manifest(struct intel_gt *gt, - const struct firmware *fw, - struct intel_uc_fw *uc_fw) -{ - u32 *dw = (u32 *)fw->data; - u32 version_hi, version_lo; - size_t min_size; - - /* Check the size of the blob before examining buffer contents */ - min_size = sizeof(u32) * (HUC_GSC_VERSION_LO_DW + 1); - if (unlikely(fw->size < min_size)) { - gt_warn(gt, "%s firmware %s: invalid size: %zu < %zu\n", - intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path, - fw->size, min_size); - return -ENODATA; - } - - version_hi = dw[HUC_GSC_VERSION_HI_DW]; - version_lo = dw[HUC_GSC_VERSION_LO_DW]; - - uc_fw->file_selected.ver.major = FIELD_GET(HUC_GSC_MAJOR_VER_HI_MASK, version_hi); - uc_fw->file_selected.ver.minor = FIELD_GET(HUC_GSC_MINOR_VER_HI_MASK, version_hi); - uc_fw->file_selected.ver.patch = FIELD_GET(HUC_GSC_PATCH_VER_LO_MASK, version_lo); - - return 0; -} - static void uc_unpack_css_version(struct intel_uc_fw_ver *ver, u32 css_value) { /* Get version numbers from the CSS header */ @@ -628,22 +605,22 @@ static void guc_read_css_info(struct intel_uc_fw *uc_fw, struct uc_css_header *c uc_fw->private_data_size = css->private_data_size; } -static int check_ccs_header(struct intel_gt *gt, - const struct firmware *fw, - struct intel_uc_fw *uc_fw) +static int __check_ccs_header(struct intel_gt *gt, + const void *fw_data, size_t fw_size, + struct intel_uc_fw *uc_fw) { struct uc_css_header *css; size_t size; /* Check the size of the blob before examining buffer contents */ - if (unlikely(fw->size < sizeof(struct uc_css_header))) { + if (unlikely(fw_size < sizeof(struct uc_css_header))) { gt_warn(gt, "%s firmware %s: invalid size: %zu < %zu\n", intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path, - fw->size, sizeof(struct uc_css_header)); + fw_size, sizeof(struct uc_css_header)); return -ENODATA; } - css = (struct uc_css_header *)fw->data; + css = (struct uc_css_header *)fw_data; /* Check integrity of size values inside CSS header */ size = (css->header_size_dw - css->key_size_dw - css->modulus_size_dw - @@ -651,7 +628,7 @@ static int check_ccs_header(struct intel_gt *gt, if (unlikely(size != sizeof(struct uc_css_header))) { gt_warn(gt, "%s firmware %s: unexpected header size: %zu != %zu\n", intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path, - fw->size, sizeof(struct uc_css_header)); + fw_size, sizeof(struct uc_css_header)); return -EPROTO; } @@ -663,10 +640,10 @@ static int check_ccs_header(struct intel_gt *gt, /* At least, it should have header, uCode and RSA. Size of all three. */ size = sizeof(struct uc_css_header) + uc_fw->ucode_size + uc_fw->rsa_size; - if (unlikely(fw->size < size)) { + if (unlikely(fw_size < size)) { gt_warn(gt, "%s firmware %s: invalid size: %zu < %zu\n", intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path, - fw->size, size); + fw_size, size); return -ENOEXEC; } @@ -687,6 +664,33 @@ static int check_ccs_header(struct intel_gt *gt, return 0; } +static int check_gsc_manifest(struct intel_gt *gt, + const struct firmware *fw, + struct intel_uc_fw *uc_fw) +{ + if (uc_fw->type != INTEL_UC_FW_TYPE_HUC) { + gt_err(gt, "trying to GSC-parse a non-HuC binary"); + return -EINVAL; + } + + intel_huc_fw_get_binary_info(uc_fw, fw->data, fw->size); + + if (uc_fw->dma_start_offset) { + u32 delta = uc_fw->dma_start_offset; + + __check_ccs_header(gt, fw->data + delta, fw->size - delta, uc_fw); + } + + return 0; +} + +static int check_ccs_header(struct intel_gt *gt, + const struct firmware *fw, + struct intel_uc_fw *uc_fw) +{ + return __check_ccs_header(gt, fw->data, fw->size, uc_fw); +} + static bool is_ver_8bit(struct intel_uc_fw_ver *ver) { return ver->major < 0xFF && ver->minor < 0xFF && ver->patch < 0xFF; @@ -734,7 +738,7 @@ static int check_fw_header(struct intel_gt *gt, if (uc_fw->type == INTEL_UC_FW_TYPE_GSC) return 0; - if (uc_fw->loaded_via_gsc) + if (uc_fw->has_gsc_headers) err = check_gsc_manifest(gt, fw, uc_fw); else err = check_ccs_header(gt, fw, uc_fw); @@ -755,7 +759,7 @@ static int try_firmware_load(struct intel_uc_fw *uc_fw, const struct firmware ** if (err) return err; - if ((*fw)->size > INTEL_UC_RSVD_GGTT_PER_FW) { + if (uc_fw->needs_ggtt_mapping && (*fw)->size > INTEL_UC_RSVD_GGTT_PER_FW) { gt_err(gt, "%s firmware %s: size (%zuKB) exceeds max supported size (%uKB)\n", intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path, (*fw)->size / SZ_1K, INTEL_UC_RSVD_GGTT_PER_FW / SZ_1K); @@ -940,29 +944,32 @@ static void uc_fw_bind_ggtt(struct intel_uc_fw *uc_fw) { struct drm_i915_gem_object *obj = uc_fw->obj; struct i915_ggtt *ggtt = __uc_fw_to_gt(uc_fw)->ggtt; - struct i915_vma_resource *dummy = &uc_fw->dummy; + struct i915_vma_resource *vma_res = &uc_fw->vma_res; u32 pte_flags = 0; - dummy->start = uc_fw_ggtt_offset(uc_fw); - dummy->node_size = obj->base.size; - dummy->bi.pages = obj->mm.pages; + if (!uc_fw->needs_ggtt_mapping) + return; + + vma_res->start = uc_fw_ggtt_offset(uc_fw); + vma_res->node_size = obj->base.size; + vma_res->bi.pages = obj->mm.pages; GEM_BUG_ON(!i915_gem_object_has_pinned_pages(obj)); /* uc_fw->obj cache domains were not controlled across suspend */ if (i915_gem_object_has_struct_page(obj)) - drm_clflush_sg(dummy->bi.pages); + drm_clflush_sg(vma_res->bi.pages); if (i915_gem_object_is_lmem(obj)) pte_flags |= PTE_LM; if (ggtt->vm.raw_insert_entries) - ggtt->vm.raw_insert_entries(&ggtt->vm, dummy, + ggtt->vm.raw_insert_entries(&ggtt->vm, vma_res, i915_gem_get_pat_index(ggtt->vm.i915, I915_CACHE_NONE), pte_flags); else - ggtt->vm.insert_entries(&ggtt->vm, dummy, + ggtt->vm.insert_entries(&ggtt->vm, vma_res, i915_gem_get_pat_index(ggtt->vm.i915, I915_CACHE_NONE), pte_flags); @@ -970,11 +977,13 @@ static void uc_fw_bind_ggtt(struct intel_uc_fw *uc_fw) static void uc_fw_unbind_ggtt(struct intel_uc_fw *uc_fw) { - struct drm_i915_gem_object *obj = uc_fw->obj; struct i915_ggtt *ggtt = __uc_fw_to_gt(uc_fw)->ggtt; - u64 start = uc_fw_ggtt_offset(uc_fw); + struct i915_vma_resource *vma_res = &uc_fw->vma_res; + + if (!vma_res->node_size) + return; - ggtt->vm.clear_range(&ggtt->vm, start, obj->base.size); + ggtt->vm.clear_range(&ggtt->vm, vma_res->start, vma_res->node_size); } static int uc_fw_xfer(struct intel_uc_fw *uc_fw, u32 dst_offset, u32 dma_flags) @@ -991,7 +1000,7 @@ static int uc_fw_xfer(struct intel_uc_fw *uc_fw, u32 dst_offset, u32 dma_flags) intel_uncore_forcewake_get(uncore, FORCEWAKE_ALL); /* Set the source address for the uCode */ - offset = uc_fw_ggtt_offset(uc_fw); + offset = uc_fw->vma_res.start + uc_fw->dma_start_offset; GEM_BUG_ON(upper_32_bits(offset) & 0xFFFF0000); intel_uncore_write_fw(uncore, DMA_ADDR_0_LOW, lower_32_bits(offset)); intel_uncore_write_fw(uncore, DMA_ADDR_0_HIGH, upper_32_bits(offset)); @@ -1065,9 +1074,7 @@ int intel_uc_fw_upload(struct intel_uc_fw *uc_fw, u32 dst_offset, u32 dma_flags) return -ENOEXEC; /* Call custom loader */ - uc_fw_bind_ggtt(uc_fw); err = uc_fw_xfer(uc_fw, dst_offset, dma_flags); - uc_fw_unbind_ggtt(uc_fw); if (err) goto fail; @@ -1171,6 +1178,8 @@ int intel_uc_fw_init(struct intel_uc_fw *uc_fw) goto out_unpin; } + uc_fw_bind_ggtt(uc_fw); + return 0; out_unpin: @@ -1181,6 +1190,7 @@ int intel_uc_fw_init(struct intel_uc_fw *uc_fw) void intel_uc_fw_fini(struct intel_uc_fw *uc_fw) { + uc_fw_unbind_ggtt(uc_fw); uc_fw_rsa_data_destroy(uc_fw); if (i915_gem_object_has_pinned_pages(uc_fw->obj)) @@ -1189,6 +1199,17 @@ void intel_uc_fw_fini(struct intel_uc_fw *uc_fw) intel_uc_fw_change_status(uc_fw, INTEL_UC_FIRMWARE_AVAILABLE); } +void intel_uc_fw_resume_mapping(struct intel_uc_fw *uc_fw) +{ + if (!intel_uc_fw_is_available(uc_fw)) + return; + + if (!i915_gem_object_has_pinned_pages(uc_fw->obj)) + return; + + uc_fw_bind_ggtt(uc_fw); +} + /** * intel_uc_fw_cleanup_fetch - cleanup uC firmware * @uc_fw: uC firmware @@ -1218,7 +1239,7 @@ size_t intel_uc_fw_copy_rsa(struct intel_uc_fw *uc_fw, void *dst, u32 max_len) { struct intel_memory_region *mr = uc_fw->obj->mm.region; u32 size = min_t(u32, uc_fw->rsa_size, max_len); - u32 offset = sizeof(struct uc_css_header) + uc_fw->ucode_size; + u32 offset = uc_fw->dma_start_offset + sizeof(struct uc_css_header) + uc_fw->ucode_size; struct sgt_iter iter; size_t count = 0; int idx; diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h index 6ba00e6b39753b18cd7bce1493fe2c47291fbf1a..054f0281197185049f0f3901fed203d8ac298887 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h @@ -99,20 +99,28 @@ struct intel_uc_fw { struct drm_i915_gem_object *obj; /** - * @dummy: A vma used in binding the uc fw to ggtt. We can't define this - * vma on the stack as it can lead to a stack overflow, so we define it - * here. Safe to have 1 copy per uc fw because the binding is single - * threaded as it done during driver load (inherently single threaded) - * or during a GT reset (mutex guarantees single threaded). + * @needs_ggtt_mapping: indicates whether the fw object needs to be + * pinned to ggtt. If true, the fw is pinned at init time and unpinned + * during driver unload. */ - struct i915_vma_resource dummy; + bool needs_ggtt_mapping; + + /** + * @vma_res: A vma resource used in binding the uc fw to ggtt. The fw is + * pinned in a reserved area of the ggtt (above the maximum address + * usable by GuC); therefore, we can't use the normal vma functions to + * do the pinning and we instead use this resource to do so. + */ + struct i915_vma_resource vma_res; struct i915_vma *rsa_data; u32 rsa_size; u32 ucode_size; u32 private_data_size; - bool loaded_via_gsc; + u32 dma_start_offset; + + bool has_gsc_headers; }; /* @@ -282,12 +290,14 @@ static inline u32 intel_uc_fw_get_upload_size(struct intel_uc_fw *uc_fw) } void intel_uc_fw_init_early(struct intel_uc_fw *uc_fw, - enum intel_uc_fw_type type); + enum intel_uc_fw_type type, + bool needs_ggtt_mapping); int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw); void intel_uc_fw_cleanup_fetch(struct intel_uc_fw *uc_fw); int intel_uc_fw_upload(struct intel_uc_fw *uc_fw, u32 offset, u32 dma_flags); int intel_uc_fw_init(struct intel_uc_fw *uc_fw); void intel_uc_fw_fini(struct intel_uc_fw *uc_fw); +void intel_uc_fw_resume_mapping(struct intel_uc_fw *uc_fw); size_t intel_uc_fw_copy_rsa(struct intel_uc_fw *uc_fw, void *dst, u32 max_len); int intel_uc_fw_mark_load_failed(struct intel_uc_fw *uc_fw, int err); void intel_uc_fw_dump(const struct intel_uc_fw *uc_fw, struct drm_printer *p); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw_abi.h b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw_abi.h index 646fa8aa6cf194762b4cd2c382f5334dc70b731f..7fe4051262497bafe2bf25e82a9a06315cc52bdc 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw_abi.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw_abi.h @@ -84,10 +84,4 @@ struct uc_css_header { } __packed; static_assert(sizeof(struct uc_css_header) == 128); -#define HUC_GSC_VERSION_HI_DW 44 -#define HUC_GSC_MAJOR_VER_HI_MASK (0xFF << 0) -#define HUC_GSC_MINOR_VER_HI_MASK (0xFF << 16) -#define HUC_GSC_VERSION_LO_DW 45 -#define HUC_GSC_PATCH_VER_LO_MASK (0xFF << 0) - #endif /* _INTEL_UC_FW_ABI_H */ diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c index 3c4ae1da0d41621935f640f6a25c0f31e2c7aa7f..05f9348b7a9d8ee7a11a2108e2f2439ddc7435ee 100644 --- a/drivers/gpu/drm/i915/gvt/cmd_parser.c +++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c @@ -2833,7 +2833,7 @@ static int command_scan(struct parser_exec_state *s, static int scan_workload(struct intel_vgpu_workload *workload) { - unsigned long gma_head, gma_tail, gma_bottom; + unsigned long gma_head, gma_tail; struct parser_exec_state s; int ret = 0; @@ -2843,7 +2843,6 @@ static int scan_workload(struct intel_vgpu_workload *workload) gma_head = workload->rb_start + workload->rb_head; gma_tail = workload->rb_start + workload->rb_tail; - gma_bottom = workload->rb_start + _RING_CTL_BUF_SIZE(workload->rb_ctl); s.buf_type = RING_BUFFER_INSTRUCTION; s.buf_addr_type = GTT_BUFFER; @@ -2874,7 +2873,7 @@ static int scan_workload(struct intel_vgpu_workload *workload) static int scan_wa_ctx(struct intel_shadow_wa_ctx *wa_ctx) { - unsigned long gma_head, gma_tail, gma_bottom, ring_size, ring_tail; + unsigned long gma_head, gma_tail, ring_size, ring_tail; struct parser_exec_state s; int ret = 0; struct intel_vgpu_workload *workload = container_of(wa_ctx, @@ -2891,7 +2890,6 @@ static int scan_wa_ctx(struct intel_shadow_wa_ctx *wa_ctx) PAGE_SIZE); gma_head = wa_ctx->indirect_ctx.guest_gma; gma_tail = wa_ctx->indirect_ctx.guest_gma + ring_tail; - gma_bottom = wa_ctx->indirect_ctx.guest_gma + ring_size; s.buf_type = RING_BUFFER_INSTRUCTION; s.buf_addr_type = GTT_BUFFER; diff --git a/drivers/gpu/drm/i915/gvt/edid.c b/drivers/gpu/drm/i915/gvt/edid.c index 7c49a3d673a547e82259746de292a5f9fc362efb..2a0438f12a1499b46227066d6fee3f65d3142673 100644 --- a/drivers/gpu/drm/i915/gvt/edid.c +++ b/drivers/gpu/drm/i915/gvt/edid.c @@ -463,10 +463,6 @@ static inline int get_aux_ch_reg(unsigned int offset) return reg; } -#define AUX_CTL_MSG_LENGTH(reg) \ - ((reg & DP_AUX_CH_CTL_MESSAGE_SIZE_MASK) >> \ - DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT) - /** * intel_gvt_i2c_handle_aux_ch_write - emulate AUX channel register write * @vgpu: a vGPU @@ -495,7 +491,8 @@ void intel_gvt_i2c_handle_aux_ch_write(struct intel_vgpu *vgpu, return; } - msg_length = AUX_CTL_MSG_LENGTH(value); + msg_length = REG_FIELD_GET(DP_AUX_CH_CTL_MESSAGE_SIZE_MASK, reg); + // check the msg in DATA register. msg = vgpu_vreg(vgpu, offset + 4); addr = (msg >> 8) & 0xffff; @@ -510,8 +507,7 @@ void intel_gvt_i2c_handle_aux_ch_write(struct intel_vgpu *vgpu, ret_msg_size = (((op & 0x1) == GVT_AUX_I2C_READ) ? 2 : 1); vgpu_vreg(vgpu, offset) = DP_AUX_CH_CTL_DONE | - ((ret_msg_size << DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT) & - DP_AUX_CH_CTL_MESSAGE_SIZE_MASK); + DP_AUX_CH_CTL_MESSAGE_SIZE(ret_msg_size); if (msg_length == 3) { if (!(op & GVT_AUX_I2C_MOT)) { diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c index 4b45a041ac5c9eb2a9221482134edbb90dd201c8..a9f7fa9b90bdadc7585e8f48908cbefecab2a487 100644 --- a/drivers/gpu/drm/i915/gvt/handlers.c +++ b/drivers/gpu/drm/i915/gvt/handlers.c @@ -1562,7 +1562,7 @@ static int pf_write(struct intel_vgpu *vgpu, if ((offset == _PS_1A_CTRL || offset == _PS_2A_CTRL || offset == _PS_1B_CTRL || offset == _PS_2B_CTRL || - offset == _PS_1C_CTRL) && (val & PS_PLANE_SEL_MASK) != 0) { + offset == _PS_1C_CTRL) && (val & PS_BINDING_MASK) != PS_BINDING_PIPE) { drm_WARN_ONCE(&i915->drm, true, "VM(%d): guest is trying to scaling a plane\n", vgpu->id); diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index e39937c82d730bdec67254cdf192e20b95e32a81..76ccd4e03e3128f7ae968ba436f8aaec5e421e8c 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -50,6 +50,7 @@ #include "i915_debugfs_params.h" #include "i915_driver.h" #include "i915_irq.h" +#include "i915_reg.h" #include "i915_scheduler.h" #include "intel_mchbar_regs.h" diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c index 93fdc40d724fa7fe70da0c1fe6c3b6320013327d..75cbc43b326dd00f4217695b5d6ebf03aa33b237 100644 --- a/drivers/gpu/drm/i915/i915_driver.c +++ b/drivers/gpu/drm/i915/i915_driver.c @@ -48,6 +48,7 @@ #include "display/intel_acpi.h" #include "display/intel_bw.h" #include "display/intel_cdclk.h" +#include "display/intel_display_driver.h" #include "display/intel_display_types.h" #include "display/intel_dmc.h" #include "display/intel_dp.h" @@ -131,8 +132,20 @@ static int i915_workqueues_init(struct drm_i915_private *dev_priv) if (dev_priv->display.hotplug.dp_wq == NULL) goto out_free_wq; + /* + * The unordered i915 workqueue should be used for all work + * scheduling that do not require running in order, which used + * to be scheduled on the system_wq before moving to a driver + * instance due deprecation of flush_scheduled_work(). + */ + dev_priv->unordered_wq = alloc_workqueue("i915-unordered", 0, 0); + if (dev_priv->unordered_wq == NULL) + goto out_free_dp_wq; + return 0; +out_free_dp_wq: + destroy_workqueue(dev_priv->display.hotplug.dp_wq); out_free_wq: destroy_workqueue(dev_priv->wq); out_err: @@ -143,6 +156,7 @@ static int i915_workqueues_init(struct drm_i915_private *dev_priv) static void i915_workqueues_cleanup(struct drm_i915_private *dev_priv) { + destroy_workqueue(dev_priv->unordered_wq); destroy_workqueue(dev_priv->display.hotplug.dp_wq); destroy_workqueue(dev_priv->wq); } @@ -221,8 +235,7 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv) mutex_init(&dev_priv->display.audio.mutex); mutex_init(&dev_priv->display.wm.wm_mutex); mutex_init(&dev_priv->display.pps.mutex); - mutex_init(&dev_priv->display.hdcp.comp_mutex); - spin_lock_init(&dev_priv->display.dkl.phy_lock); + mutex_init(&dev_priv->display.hdcp.hdcp_mutex); i915_memcpy_init_early(dev_priv); intel_runtime_pm_init_early(&dev_priv->runtime_pm); @@ -243,15 +256,13 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv) if (ret < 0) goto err_rootgt; - i915_drm_clients_init(&dev_priv->clients, dev_priv); - i915_gem_init_early(dev_priv); /* This must be called before any calls to HAS_PCH_* */ intel_detect_pch(dev_priv); intel_irq_init(dev_priv); - intel_init_display_hooks(dev_priv); + intel_display_driver_early_probe(dev_priv); intel_clock_gating_hooks_init(dev_priv); intel_detect_preproduction_hw(dev_priv); @@ -278,7 +289,6 @@ static void i915_driver_late_release(struct drm_i915_private *dev_priv) intel_power_domains_cleanup(dev_priv); i915_gem_cleanup_early(dev_priv); intel_gt_driver_late_release_all(dev_priv); - i915_drm_clients_fini(&dev_priv->clients); intel_region_ttm_device_fini(dev_priv); vlv_suspend_cleanup(dev_priv); i915_workqueues_cleanup(dev_priv); @@ -720,8 +730,6 @@ i915_driver_create(struct pci_dev *pdev, const struct pci_device_id *ent) { const struct intel_device_info *match_info = (struct intel_device_info *)ent->driver_data; - struct intel_device_info *device_info; - struct intel_runtime_info *runtime; struct drm_i915_private *i915; i915 = devm_drm_dev_alloc(&pdev->dev, &i915_drm_driver, @@ -734,14 +742,8 @@ i915_driver_create(struct pci_dev *pdev, const struct pci_device_id *ent) /* Device parameters start as a copy of module parameters. */ i915_params_copy(&i915->params, &i915_modparams); - /* Setup the write-once "constant" device info */ - device_info = mkwrite_device_info(i915); - memcpy(device_info, match_info, sizeof(*device_info)); - - /* Initialize initial runtime info from static const data and pdev. */ - runtime = RUNTIME_INFO(i915); - memcpy(runtime, &INTEL_INFO(i915)->__runtime, sizeof(*runtime)); - runtime->device_id = pdev->device; + /* Set up device info and initial runtime info. */ + intel_device_info_driver_create(i915, pdev->device, match_info); return i915; } @@ -752,7 +754,7 @@ i915_driver_create(struct pci_dev *pdev, const struct pci_device_id *ent) * @ent: matching PCI ID entry * * The driver probe routine has to do several things: - * - drive output discovery via intel_modeset_init() + * - drive output discovery via intel_display_driver_probe() * - initialize the memory manager * - allocate initial config memory * - setup the DRM framebuffer with the allocated memory @@ -762,13 +764,17 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent) struct drm_i915_private *i915; int ret; + ret = pci_enable_device(pdev); + if (ret) { + pr_err("Failed to enable graphics device: %pe\n", ERR_PTR(ret)); + return ret; + } + i915 = i915_driver_create(pdev, ent); - if (IS_ERR(i915)) + if (IS_ERR(i915)) { + pci_disable_device(pdev); return PTR_ERR(i915); - - ret = pci_enable_device(pdev); - if (ret) - goto out_fini; + } ret = i915_driver_early_probe(i915); if (ret < 0) @@ -790,7 +796,7 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (ret < 0) goto out_cleanup_mmio; - ret = intel_modeset_init_noirq(i915); + ret = intel_display_driver_probe_noirq(i915); if (ret < 0) goto out_cleanup_hw; @@ -798,7 +804,7 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (ret) goto out_cleanup_modeset; - ret = intel_modeset_init_nogem(i915); + ret = intel_display_driver_probe_nogem(i915); if (ret) goto out_cleanup_irq; @@ -808,7 +814,7 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent) intel_pxp_init(i915); - ret = intel_modeset_init(i915); + ret = intel_display_driver_probe(i915); if (ret) goto out_cleanup_gem; @@ -828,14 +834,14 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent) i915_gem_driver_release(i915); out_cleanup_modeset2: /* FIXME clean up the error path */ - intel_modeset_driver_remove(i915); + intel_display_driver_remove(i915); intel_irq_uninstall(i915); - intel_modeset_driver_remove_noirq(i915); + intel_display_driver_remove_noirq(i915); goto out_cleanup_modeset; out_cleanup_irq: intel_irq_uninstall(i915); out_cleanup_modeset: - intel_modeset_driver_remove_nogem(i915); + intel_display_driver_remove_nogem(i915); out_cleanup_hw: i915_driver_hw_remove(i915); intel_memory_regions_driver_release(i915); @@ -851,7 +857,6 @@ int i915_driver_probe(struct pci_dev *pdev, const struct pci_device_id *ent) i915_driver_late_release(i915); out_pci_disable: pci_disable_device(pdev); -out_fini: i915_probe_error(i915, "Device initialization failed (%d)\n", ret); return ret; } @@ -871,16 +876,16 @@ void i915_driver_remove(struct drm_i915_private *i915) intel_gvt_driver_remove(i915); - intel_modeset_driver_remove(i915); + intel_display_driver_remove(i915); intel_irq_uninstall(i915); - intel_modeset_driver_remove_noirq(i915); + intel_display_driver_remove_noirq(i915); i915_reset_error_state(i915); i915_gem_driver_remove(i915); - intel_modeset_driver_remove_nogem(i915); + intel_display_driver_remove_nogem(i915); i915_driver_hw_remove(i915); @@ -967,11 +972,19 @@ static void intel_suspend_encoders(struct drm_i915_private *dev_priv) if (!HAS_DISPLAY(dev_priv)) return; + /* + * TODO: check and remove holding the modeset locks if none of + * the encoders depends on this. + */ drm_modeset_lock_all(&dev_priv->drm); for_each_intel_encoder(&dev_priv->drm, encoder) if (encoder->suspend) encoder->suspend(encoder); drm_modeset_unlock_all(&dev_priv->drm); + + for_each_intel_encoder(&dev_priv->drm, encoder) + if (encoder->suspend_complete) + encoder->suspend_complete(encoder); } static void intel_shutdown_encoders(struct drm_i915_private *dev_priv) @@ -981,11 +994,19 @@ static void intel_shutdown_encoders(struct drm_i915_private *dev_priv) if (!HAS_DISPLAY(dev_priv)) return; + /* + * TODO: check and remove holding the modeset locks if none of + * the encoders depends on this. + */ drm_modeset_lock_all(&dev_priv->drm); for_each_intel_encoder(&dev_priv->drm, encoder) if (encoder->shutdown) encoder->shutdown(encoder); drm_modeset_unlock_all(&dev_priv->drm); + + for_each_intel_encoder(&dev_priv->drm, encoder) + if (encoder->shutdown_complete) + encoder->shutdown_complete(encoder); } void i915_driver_shutdown(struct drm_i915_private *i915) @@ -1052,7 +1073,7 @@ static int i915_drm_prepare(struct drm_device *dev) intel_pxp_suspend_prepare(i915->pxp); /* - * NB intel_display_suspend() may issue new requests after we've + * NB intel_display_driver_suspend() may issue new requests after we've * ostensibly marked the GPU as ready-to-sleep here. We need to * split out that work and pull it forward so that after point, * the GPU is not woken again. @@ -1076,7 +1097,7 @@ static int i915_drm_suspend(struct drm_device *dev) pci_save_state(pdev); - intel_display_suspend(dev); + intel_display_driver_suspend(dev_priv); intel_dp_mst_suspend(dev_priv); @@ -1107,18 +1128,6 @@ static int i915_drm_suspend(struct drm_device *dev) return 0; } -static enum i915_drm_suspend_mode -get_suspend_mode(struct drm_i915_private *dev_priv, bool hibernate) -{ - if (hibernate) - return I915_DRM_SUSPEND_HIBERNATE; - - if (suspend_to_idle(dev_priv)) - return I915_DRM_SUSPEND_IDLE; - - return I915_DRM_SUSPEND_MEM; -} - static int i915_drm_suspend_late(struct drm_device *dev, bool hibernation) { struct drm_i915_private *dev_priv = to_i915(dev); @@ -1126,6 +1135,7 @@ static int i915_drm_suspend_late(struct drm_device *dev, bool hibernation) struct intel_runtime_pm *rpm = &dev_priv->runtime_pm; struct intel_gt *gt; int ret, i; + bool s2idle = !hibernation && suspend_to_idle(dev_priv); disable_rpm_wakeref_asserts(rpm); @@ -1136,8 +1146,7 @@ static int i915_drm_suspend_late(struct drm_device *dev, bool hibernation) for_each_gt(gt, dev_priv, i) intel_uncore_suspend(gt->uncore); - intel_power_domains_suspend(dev_priv, - get_suspend_mode(dev_priv, hibernation)); + intel_power_domains_suspend(dev_priv, s2idle); intel_display_power_suspend_late(dev_priv); @@ -1233,7 +1242,7 @@ static int i915_drm_resume(struct drm_device *dev) * * drm_mode_config_reset() needs AUX interrupts. * - * Modeset enabling in intel_modeset_init_hw() also needs working + * Modeset enabling in intel_display_driver_init_hw() also needs working * interrupts. */ intel_runtime_pm_enable_interrupts(dev_priv); @@ -1243,13 +1252,14 @@ static int i915_drm_resume(struct drm_device *dev) i915_gem_resume(dev_priv); - intel_modeset_init_hw(dev_priv); + intel_display_driver_init_hw(dev_priv); + intel_clock_gating_init(dev_priv); intel_hpd_init(dev_priv); /* MST sideband requires HPD interrupts enabled */ intel_dp_mst_resume(dev_priv); - intel_display_resume(dev); + intel_display_driver_resume(dev_priv); intel_hpd_poll_disable(dev_priv); if (HAS_DISPLAY(dev_priv)) @@ -1706,7 +1716,7 @@ static const struct file_operations i915_driver_fops = { .compat_ioctl = i915_ioc32_compat_ioctl, .llseek = noop_llseek, #ifdef CONFIG_PROC_FS - .show_fdinfo = i915_drm_client_fdinfo, + .show_fdinfo = drm_show_fdinfo, #endif }; @@ -1806,6 +1816,7 @@ static const struct drm_driver i915_drm_driver = { .open = i915_driver_open, .lastclose = i915_driver_lastclose, .postclose = i915_driver_postclose, + .show_fdinfo = i915_drm_client_fdinfo, .prime_handle_to_fd = drm_gem_prime_handle_to_fd, .prime_fd_to_handle = drm_gem_prime_fd_to_handle, diff --git a/drivers/gpu/drm/i915/i915_drm_client.c b/drivers/gpu/drm/i915/i915_drm_client.c index d18d0a3ed905bf4f79674c182f18a9e52ab10eaf..2a44b3876cb510a825d38fe04b7f1d5259622998 100644 --- a/drivers/gpu/drm/i915/i915_drm_client.c +++ b/drivers/gpu/drm/i915/i915_drm_client.c @@ -17,64 +17,29 @@ #include "i915_gem.h" #include "i915_utils.h" -void i915_drm_clients_init(struct i915_drm_clients *clients, - struct drm_i915_private *i915) -{ - clients->i915 = i915; - clients->next_id = 0; - - xa_init_flags(&clients->xarray, XA_FLAGS_ALLOC | XA_FLAGS_LOCK_IRQ); -} - -struct i915_drm_client *i915_drm_client_add(struct i915_drm_clients *clients) +struct i915_drm_client *i915_drm_client_alloc(void) { struct i915_drm_client *client; - struct xarray *xa = &clients->xarray; - int ret; client = kzalloc(sizeof(*client), GFP_KERNEL); if (!client) - return ERR_PTR(-ENOMEM); - - xa_lock_irq(xa); - ret = __xa_alloc_cyclic(xa, &client->id, client, xa_limit_32b, - &clients->next_id, GFP_KERNEL); - xa_unlock_irq(xa); - if (ret < 0) - goto err; + return NULL; kref_init(&client->kref); spin_lock_init(&client->ctx_lock); INIT_LIST_HEAD(&client->ctx_list); - client->clients = clients; return client; - -err: - kfree(client); - - return ERR_PTR(ret); } void __i915_drm_client_free(struct kref *kref) { struct i915_drm_client *client = container_of(kref, typeof(*client), kref); - struct xarray *xa = &client->clients->xarray; - unsigned long flags; - xa_lock_irqsave(xa, flags); - __xa_erase(xa, client->id); - xa_unlock_irqrestore(xa, flags); kfree(client); } -void i915_drm_clients_fini(struct i915_drm_clients *clients) -{ - GEM_BUG_ON(!xa_empty(&clients->xarray)); - xa_destroy(&clients->xarray); -} - #ifdef CONFIG_PROC_FS static const char * const uabi_class_names[] = { [I915_ENGINE_CLASS_RENDER] = "render", @@ -101,38 +66,34 @@ static u64 busy_add(struct i915_gem_context *ctx, unsigned int class) } static void -show_client_class(struct seq_file *m, +show_client_class(struct drm_printer *p, + struct drm_i915_private *i915, struct i915_drm_client *client, unsigned int class) { - const struct list_head *list = &client->ctx_list; + const unsigned int capacity = i915->engine_uabi_class_count[class]; u64 total = atomic64_read(&client->past_runtime[class]); - const unsigned int capacity = - client->clients->i915->engine_uabi_class_count[class]; struct i915_gem_context *ctx; rcu_read_lock(); - list_for_each_entry_rcu(ctx, list, client_link) + list_for_each_entry_rcu(ctx, &client->ctx_list, client_link) total += busy_add(ctx, class); rcu_read_unlock(); if (capacity) - seq_printf(m, "drm-engine-%s:\t%llu ns\n", + drm_printf(p, "drm-engine-%s:\t%llu ns\n", uabi_class_names[class], total); if (capacity > 1) - seq_printf(m, "drm-engine-capacity-%s:\t%u\n", + drm_printf(p, "drm-engine-capacity-%s:\t%u\n", uabi_class_names[class], capacity); } -void i915_drm_client_fdinfo(struct seq_file *m, struct file *f) +void i915_drm_client_fdinfo(struct drm_printer *p, struct drm_file *file) { - struct drm_file *file = f->private_data; struct drm_i915_file_private *file_priv = file->driver_priv; struct drm_i915_private *i915 = file_priv->i915; - struct i915_drm_client *client = file_priv->client; - struct pci_dev *pdev = to_pci_dev(i915->drm.dev); unsigned int i; /* @@ -141,16 +102,10 @@ void i915_drm_client_fdinfo(struct seq_file *m, struct file *f) * ****************************************************************** */ - seq_printf(m, "drm-driver:\t%s\n", i915->drm.driver->name); - seq_printf(m, "drm-pdev:\t%04x:%02x:%02x.%d\n", - pci_domain_nr(pdev->bus), pdev->bus->number, - PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn)); - seq_printf(m, "drm-client-id:\t%u\n", client->id); - if (GRAPHICS_VER(i915) < 8) return; for (i = 0; i < ARRAY_SIZE(uabi_class_names); i++) - show_client_class(m, client, i); + show_client_class(p, i915, file_priv->client, i); } #endif diff --git a/drivers/gpu/drm/i915/i915_drm_client.h b/drivers/gpu/drm/i915/i915_drm_client.h index 69496af996d9e94ac1903bd834563d4ef80a2151..4c18b99e10a4e2ea947731310cce55d838da9204 100644 --- a/drivers/gpu/drm/i915/i915_drm_client.h +++ b/drivers/gpu/drm/i915/i915_drm_client.h @@ -9,20 +9,13 @@ #include #include #include -#include #include #define I915_LAST_UABI_ENGINE_CLASS I915_ENGINE_CLASS_COMPUTE -struct drm_i915_private; - -struct i915_drm_clients { - struct drm_i915_private *i915; - - struct xarray xarray; - u32 next_id; -}; +struct drm_file; +struct drm_printer; struct i915_drm_client { struct kref kref; @@ -32,17 +25,12 @@ struct i915_drm_client { spinlock_t ctx_lock; /* For add/remove from ctx_list. */ struct list_head ctx_list; /* List of contexts belonging to client. */ - struct i915_drm_clients *clients; - /** * @past_runtime: Accumulation of pphwsp runtimes from closed contexts. */ atomic64_t past_runtime[I915_LAST_UABI_ENGINE_CLASS + 1]; }; -void i915_drm_clients_init(struct i915_drm_clients *clients, - struct drm_i915_private *i915); - static inline struct i915_drm_client * i915_drm_client_get(struct i915_drm_client *client) { @@ -57,12 +45,10 @@ static inline void i915_drm_client_put(struct i915_drm_client *client) kref_put(&client->kref, __i915_drm_client_free); } -struct i915_drm_client *i915_drm_client_add(struct i915_drm_clients *clients); +struct i915_drm_client *i915_drm_client_alloc(void); #ifdef CONFIG_PROC_FS -void i915_drm_client_fdinfo(struct seq_file *m, struct file *f); +void i915_drm_client_fdinfo(struct drm_printer *p, struct drm_file *file); #endif -void i915_drm_clients_fini(struct i915_drm_clients *clients); - #endif /* !__I915_DRM_CLIENT_H__ */ diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index f23b030aaf09f3a0b016ed06a4157f5971163daa..b4cf6f0f636d6e5172b491cd07510766dd8f6919 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -205,6 +205,7 @@ struct drm_i915_private { const struct intel_device_info __info; /* Use INTEL_INFO() to access. */ struct intel_runtime_info __runtime; /* Use RUNTIME_INFO() to access. */ + struct intel_display_runtime_info __display_runtime; /* Access with DISPLAY_RUNTIME_INFO() */ struct intel_driver_caps caps; struct i915_dsm dsm; @@ -259,6 +260,16 @@ struct drm_i915_private { */ struct workqueue_struct *wq; + /** + * unordered_wq - internal workqueue for unordered work + * + * This workqueue should be used for all unordered work + * scheduling within i915, which used to be scheduled on the + * system_wq before moving to a driver instance due + * deprecation of flush_scheduled_work(). + */ + struct workqueue_struct *unordered_wq; + /* pm private clock gating functions */ const struct drm_i915_clock_gating_funcs *clock_gating_funcs; @@ -313,7 +324,7 @@ struct drm_i915_private { /* * i915->gt[0] == &i915->gt0 */ -#define I915_MAX_GT 4 +#define I915_MAX_GT 2 struct intel_gt *gt[I915_MAX_GT]; struct kobject *sysfs_gt; @@ -347,8 +358,6 @@ struct drm_i915_private { struct i915_pmu pmu; - struct i915_drm_clients clients; - /* The TTM device structure. */ struct ttm_device bdev; @@ -408,7 +417,9 @@ static inline struct intel_gt *to_gt(struct drm_i915_private *i915) (engine__) = rb_to_uabi_engine(rb_next(&(engine__)->uabi_node))) #define INTEL_INFO(i915) (&(i915)->__info) +#define DISPLAY_INFO(i915) (INTEL_INFO(i915)->display) #define RUNTIME_INFO(i915) (&(i915)->__runtime) +#define DISPLAY_RUNTIME_INFO(i915) (&(i915)->__display_runtime) #define DRIVER_CAPS(i915) (&(i915)->caps) #define INTEL_DEVID(i915) (RUNTIME_INFO(i915)->device_id) @@ -427,7 +438,7 @@ static inline struct intel_gt *to_gt(struct drm_i915_private *i915) #define IS_MEDIA_VER(i915, from, until) \ (MEDIA_VER(i915) >= (from) && MEDIA_VER(i915) <= (until)) -#define DISPLAY_VER(i915) (RUNTIME_INFO(i915)->display.ip.ver) +#define DISPLAY_VER(i915) (DISPLAY_RUNTIME_INFO(i915)->ip.ver) #define IS_DISPLAY_VER(i915, from, until) \ (DISPLAY_VER(i915) >= (from) && DISPLAY_VER(i915) <= (until)) @@ -782,10 +793,6 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, ((sizes) & ~RUNTIME_INFO(i915)->page_sizes) == 0; \ }) -#define HAS_OVERLAY(i915) (INTEL_INFO(i915)->display.has_overlay) -#define OVERLAY_NEEDS_PHYSICAL(i915) \ - (INTEL_INFO(i915)->display.overlay_needs_physical) - /* Early gen2 have a totally busted CS tlb and require pinned batches. */ #define HAS_BROKEN_CS_TLB(i915) (IS_I830(i915) || IS_I845G(i915)) @@ -796,41 +803,11 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define NEEDS_WaRsDisableCoarsePowerGating(i915) \ (IS_SKL_GT3(i915) || IS_SKL_GT4(i915)) -#define HAS_GMBUS_IRQ(i915) (DISPLAY_VER(i915) >= 4) -#define HAS_GMBUS_BURST_READ(i915) (DISPLAY_VER(i915) >= 11 || \ - IS_GEMINILAKE(i915) || \ - IS_KABYLAKE(i915)) - /* With the 945 and later, Y tiling got adjusted so that it was 32 128-byte * rows, which changed the alignment requirements and fence programming. */ #define HAS_128_BYTE_Y_TILING(i915) (GRAPHICS_VER(i915) != 2 && \ !(IS_I915G(i915) || IS_I915GM(i915))) -#define SUPPORTS_TV(i915) (INTEL_INFO(i915)->display.supports_tv) -#define I915_HAS_HOTPLUG(i915) (INTEL_INFO(i915)->display.has_hotplug) - -#define HAS_FW_BLC(i915) (DISPLAY_VER(i915) > 2) -#define HAS_FBC(i915) (RUNTIME_INFO(i915)->fbc_mask != 0) -#define HAS_CUR_FBC(i915) (!HAS_GMCH(i915) && DISPLAY_VER(i915) >= 7) - -#define HAS_DPT(i915) (DISPLAY_VER(i915) >= 13) - -#define HAS_IPS(i915) (IS_HSW_ULT(i915) || IS_BROADWELL(i915)) - -#define HAS_DP_MST(i915) (INTEL_INFO(i915)->display.has_dp_mst) -#define HAS_DP20(i915) (IS_DG2(i915) || DISPLAY_VER(i915) >= 14) - -#define HAS_DOUBLE_BUFFERED_M_N(i915) (DISPLAY_VER(i915) >= 9 || IS_BROADWELL(i915)) - -#define HAS_CDCLK_CRAWL(i915) (INTEL_INFO(i915)->display.has_cdclk_crawl) -#define HAS_CDCLK_SQUASH(i915) (INTEL_INFO(i915)->display.has_cdclk_squash) -#define HAS_DDI(i915) (INTEL_INFO(i915)->display.has_ddi) -#define HAS_FPGA_DBG_UNCLAIMED(i915) (INTEL_INFO(i915)->display.has_fpga_dbg) -#define HAS_PSR(i915) (INTEL_INFO(i915)->display.has_psr) -#define HAS_PSR_HW_TRACKING(i915) \ - (INTEL_INFO(i915)->display.has_psr_hw_tracking) -#define HAS_PSR2_SEL_FETCH(i915) (DISPLAY_VER(i915) >= 12) -#define HAS_TRANSCODER(i915, trans) ((RUNTIME_INFO(i915)->cpu_transcoder_mask & BIT(trans)) != 0) #define HAS_RC6(i915) (INTEL_INFO(i915)->has_rc6) #define HAS_RC6p(i915) (INTEL_INFO(i915)->has_rc6p) @@ -838,11 +815,6 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define HAS_RPS(i915) (INTEL_INFO(i915)->has_rps) -#define HAS_DMC(i915) (RUNTIME_INFO(i915)->has_dmc) -#define HAS_DSB(i915) (INTEL_INFO(i915)->display.has_dsb) -#define HAS_DSC(__i915) (RUNTIME_INFO(__i915)->has_dsc) -#define HAS_HW_SAGV_WM(i915) (DISPLAY_VER(i915) >= 13 && !IS_DGFX(i915)) - #define HAS_HECI_PXP(i915) \ (INTEL_INFO(i915)->has_heci_pxp) @@ -851,8 +823,6 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define HAS_HECI_GSC(i915) (HAS_HECI_PXP(i915) || HAS_HECI_GSCFI(i915)) -#define HAS_MSO(i915) (DISPLAY_VER(i915) >= 12) - #define HAS_RUNTIME_PM(i915) (INTEL_INFO(i915)->has_runtime_pm) #define HAS_64BIT_RELOC(i915) (INTEL_INFO(i915)->has_64bit_reloc) @@ -869,9 +839,6 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, */ #define HAS_64K_PAGES(i915) (INTEL_INFO(i915)->has_64k_pages) -#define HAS_IPC(i915) (INTEL_INFO(i915)->display.has_ipc) -#define HAS_SAGV(i915) (DISPLAY_VER(i915) >= 9 && !IS_LP(i915)) - #define HAS_REGION(i915, i) (RUNTIME_INFO(i915)->memory_regions & (i)) #define HAS_LMEM(i915) HAS_REGION(i915, REGION_LMEM) @@ -889,12 +856,8 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define HAS_GLOBAL_MOCS_REGISTERS(i915) (INTEL_INFO(i915)->has_global_mocs) -#define HAS_GMCH(i915) (INTEL_INFO(i915)->display.has_gmch) - #define HAS_GMD_ID(i915) (INTEL_INFO(i915)->has_gmd_id) -#define HAS_LSPCON(i915) (IS_DISPLAY_VER(i915, 9, 10)) - #define HAS_L3_CCS_READ(i915) (INTEL_INFO(i915)->has_l3_ccs_read) /* DPF == dynamic parity feature */ @@ -902,14 +865,6 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define NUM_L3_SLICES(i915) (IS_HSW_GT3(i915) ? \ 2 : HAS_L3_DPF(i915)) -#define INTEL_NUM_PIPES(i915) (hweight8(RUNTIME_INFO(i915)->pipe_mask)) - -#define HAS_DISPLAY(i915) (RUNTIME_INFO(i915)->pipe_mask != 0) - -#define HAS_VRR(i915) (DISPLAY_VER(i915) >= 11) - -#define HAS_ASYNC_FLIPS(i915) (DISPLAY_VER(i915) >= 5) - /* Only valid when HAS_DISPLAY() is true */ #define INTEL_DISPLAY_ENABLED(i915) \ (drm_WARN_ON(&(i915)->drm, !HAS_DISPLAY(i915)), \ @@ -919,11 +874,6 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define HAS_GUC_DEPRIVILEGE(i915) \ (INTEL_INFO(i915)->has_guc_deprivilege) -#define HAS_D12_PLANE_MINIMIZATION(i915) (IS_ROCKETLAKE(i915) || \ - IS_ALDERLAKE_S(i915)) - -#define HAS_MBUS_JOINING(i915) (IS_ALDERLAKE_P(i915) || DISPLAY_VER(i915) >= 14) - #define HAS_3D_PIPELINE(i915) (INTEL_INFO(i915)->has_3d_pipeline) #define HAS_ONE_EU_PER_FUSE_BIT(i915) (INTEL_INFO(i915)->has_one_eu_per_fuse_bit) @@ -931,11 +881,4 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define HAS_LMEMBAR_SMEM_STOLEN(i915) (!HAS_LMEM(i915) && \ GRAPHICS_VER_FULL(i915) >= IP_VER(12, 70)) -/* intel_device_info.c */ -static inline struct intel_device_info * -mkwrite_device_info(struct drm_i915_private *dev_priv) -{ - return (struct intel_device_info *)INTEL_INFO(dev_priv); -} - #endif diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index e70b762f0b032c75250cf16abc4780d3a54e621d..1f65bb33dd212f2c44895f2317c8738a2d05b64c 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -1325,11 +1325,9 @@ int i915_gem_open(struct drm_i915_private *i915, struct drm_file *file) if (!file_priv) goto err_alloc; - client = i915_drm_client_add(&i915->clients); - if (IS_ERR(client)) { - ret = PTR_ERR(client); + client = i915_drm_client_alloc(); + if (!client) goto err_client; - } file->driver_priv = file_priv; file_priv->i915 = i915; diff --git a/drivers/gpu/drm/i915/i915_getparam.c b/drivers/gpu/drm/i915/i915_getparam.c index 6f11d7eaa91a99e29efbfa5bc63bad917dfca918..890f2b382bee36d2a6b6abf64730bcf65280903b 100644 --- a/drivers/gpu/drm/i915/i915_getparam.c +++ b/drivers/gpu/drm/i915/i915_getparam.c @@ -100,7 +100,11 @@ int i915_getparam_ioctl(struct drm_device *dev, void *data, value = sseu->min_eu_in_pool; break; case I915_PARAM_HUC_STATUS: - value = intel_huc_check_status(&to_gt(i915)->uc.huc); + /* On platform with a media GT, the HuC is on that GT */ + if (i915->media_gt) + value = intel_huc_check_status(&i915->media_gt->uc.huc); + else + value = intel_huc_check_status(&to_gt(i915)->uc.huc); if (value < 0) return value; break; diff --git a/drivers/gpu/drm/i915/i915_hwmon.c b/drivers/gpu/drm/i915/i915_hwmon.c index a3bdd9f68a458287c70de6d048d0527da3eae177..975da8e7f2a9f8138c80b03c731ce03856ef7a89 100644 --- a/drivers/gpu/drm/i915/i915_hwmon.c +++ b/drivers/gpu/drm/i915/i915_hwmon.c @@ -269,7 +269,7 @@ static const struct attribute_group *hwm_groups[] = { NULL }; -static const struct hwmon_channel_info *hwm_info[] = { +static const struct hwmon_channel_info * const hwm_info[] = { HWMON_CHANNEL_INFO(in, HWMON_I_INPUT), HWMON_CHANNEL_INFO(power, HWMON_P_MAX | HWMON_P_RATED_MAX | HWMON_P_CRIT), HWMON_CHANNEL_INFO(energy, HWMON_E_INPUT), @@ -277,7 +277,7 @@ static const struct hwmon_channel_info *hwm_info[] = { NULL }; -static const struct hwmon_channel_info *hwm_gt_info[] = { +static const struct hwmon_channel_info * const hwm_gt_info[] = { HWMON_CHANNEL_INFO(energy, HWMON_E_INPUT), NULL }; diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 53c83e2570553fa156f4804f05c6bdc33f089a1b..82fbabcdd7a594f97d4c8a3526ca5b67979610cf 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -33,15 +33,11 @@ #include -#include "display/icl_dsi_regs.h" -#include "display/intel_de.h" -#include "display/intel_display_trace.h" +#include "display/intel_display_irq.h" #include "display/intel_display_types.h" -#include "display/intel_fdi_regs.h" -#include "display/intel_fifo_underrun.h" #include "display/intel_hotplug.h" +#include "display/intel_hotplug_irq.h" #include "display/intel_lpe_audio.h" -#include "display/intel_psr.h" #include "display/intel_psr_regs.h" #include "gt/intel_breadcrumbs.h" @@ -54,6 +50,7 @@ #include "i915_driver.h" #include "i915_drv.h" #include "i915_irq.h" +#include "i915_reg.h" /** * DOC: interrupt handling @@ -81,159 +78,6 @@ static inline void pmu_irq_stats(struct drm_i915_private *i915, WRITE_ONCE(i915->pmu.irq_count, i915->pmu.irq_count + 1); } -typedef bool (*long_pulse_detect_func)(enum hpd_pin pin, u32 val); -typedef u32 (*hotplug_enables_func)(struct intel_encoder *encoder); - -static const u32 hpd_ilk[HPD_NUM_PINS] = { - [HPD_PORT_A] = DE_DP_A_HOTPLUG, -}; - -static const u32 hpd_ivb[HPD_NUM_PINS] = { - [HPD_PORT_A] = DE_DP_A_HOTPLUG_IVB, -}; - -static const u32 hpd_bdw[HPD_NUM_PINS] = { - [HPD_PORT_A] = GEN8_DE_PORT_HOTPLUG(HPD_PORT_A), -}; - -static const u32 hpd_ibx[HPD_NUM_PINS] = { - [HPD_CRT] = SDE_CRT_HOTPLUG, - [HPD_SDVO_B] = SDE_SDVOB_HOTPLUG, - [HPD_PORT_B] = SDE_PORTB_HOTPLUG, - [HPD_PORT_C] = SDE_PORTC_HOTPLUG, - [HPD_PORT_D] = SDE_PORTD_HOTPLUG, -}; - -static const u32 hpd_cpt[HPD_NUM_PINS] = { - [HPD_CRT] = SDE_CRT_HOTPLUG_CPT, - [HPD_SDVO_B] = SDE_SDVOB_HOTPLUG_CPT, - [HPD_PORT_B] = SDE_PORTB_HOTPLUG_CPT, - [HPD_PORT_C] = SDE_PORTC_HOTPLUG_CPT, - [HPD_PORT_D] = SDE_PORTD_HOTPLUG_CPT, -}; - -static const u32 hpd_spt[HPD_NUM_PINS] = { - [HPD_PORT_A] = SDE_PORTA_HOTPLUG_SPT, - [HPD_PORT_B] = SDE_PORTB_HOTPLUG_CPT, - [HPD_PORT_C] = SDE_PORTC_HOTPLUG_CPT, - [HPD_PORT_D] = SDE_PORTD_HOTPLUG_CPT, - [HPD_PORT_E] = SDE_PORTE_HOTPLUG_SPT, -}; - -static const u32 hpd_mask_i915[HPD_NUM_PINS] = { - [HPD_CRT] = CRT_HOTPLUG_INT_EN, - [HPD_SDVO_B] = SDVOB_HOTPLUG_INT_EN, - [HPD_SDVO_C] = SDVOC_HOTPLUG_INT_EN, - [HPD_PORT_B] = PORTB_HOTPLUG_INT_EN, - [HPD_PORT_C] = PORTC_HOTPLUG_INT_EN, - [HPD_PORT_D] = PORTD_HOTPLUG_INT_EN, -}; - -static const u32 hpd_status_g4x[HPD_NUM_PINS] = { - [HPD_CRT] = CRT_HOTPLUG_INT_STATUS, - [HPD_SDVO_B] = SDVOB_HOTPLUG_INT_STATUS_G4X, - [HPD_SDVO_C] = SDVOC_HOTPLUG_INT_STATUS_G4X, - [HPD_PORT_B] = PORTB_HOTPLUG_INT_STATUS, - [HPD_PORT_C] = PORTC_HOTPLUG_INT_STATUS, - [HPD_PORT_D] = PORTD_HOTPLUG_INT_STATUS, -}; - -static const u32 hpd_status_i915[HPD_NUM_PINS] = { - [HPD_CRT] = CRT_HOTPLUG_INT_STATUS, - [HPD_SDVO_B] = SDVOB_HOTPLUG_INT_STATUS_I915, - [HPD_SDVO_C] = SDVOC_HOTPLUG_INT_STATUS_I915, - [HPD_PORT_B] = PORTB_HOTPLUG_INT_STATUS, - [HPD_PORT_C] = PORTC_HOTPLUG_INT_STATUS, - [HPD_PORT_D] = PORTD_HOTPLUG_INT_STATUS, -}; - -static const u32 hpd_bxt[HPD_NUM_PINS] = { - [HPD_PORT_A] = GEN8_DE_PORT_HOTPLUG(HPD_PORT_A), - [HPD_PORT_B] = GEN8_DE_PORT_HOTPLUG(HPD_PORT_B), - [HPD_PORT_C] = GEN8_DE_PORT_HOTPLUG(HPD_PORT_C), -}; - -static const u32 hpd_gen11[HPD_NUM_PINS] = { - [HPD_PORT_TC1] = GEN11_TC_HOTPLUG(HPD_PORT_TC1) | GEN11_TBT_HOTPLUG(HPD_PORT_TC1), - [HPD_PORT_TC2] = GEN11_TC_HOTPLUG(HPD_PORT_TC2) | GEN11_TBT_HOTPLUG(HPD_PORT_TC2), - [HPD_PORT_TC3] = GEN11_TC_HOTPLUG(HPD_PORT_TC3) | GEN11_TBT_HOTPLUG(HPD_PORT_TC3), - [HPD_PORT_TC4] = GEN11_TC_HOTPLUG(HPD_PORT_TC4) | GEN11_TBT_HOTPLUG(HPD_PORT_TC4), - [HPD_PORT_TC5] = GEN11_TC_HOTPLUG(HPD_PORT_TC5) | GEN11_TBT_HOTPLUG(HPD_PORT_TC5), - [HPD_PORT_TC6] = GEN11_TC_HOTPLUG(HPD_PORT_TC6) | GEN11_TBT_HOTPLUG(HPD_PORT_TC6), -}; - -static const u32 hpd_icp[HPD_NUM_PINS] = { - [HPD_PORT_A] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_A), - [HPD_PORT_B] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_B), - [HPD_PORT_C] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_C), - [HPD_PORT_TC1] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC1), - [HPD_PORT_TC2] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC2), - [HPD_PORT_TC3] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC3), - [HPD_PORT_TC4] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC4), - [HPD_PORT_TC5] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC5), - [HPD_PORT_TC6] = SDE_TC_HOTPLUG_ICP(HPD_PORT_TC6), -}; - -static const u32 hpd_sde_dg1[HPD_NUM_PINS] = { - [HPD_PORT_A] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_A), - [HPD_PORT_B] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_B), - [HPD_PORT_C] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_C), - [HPD_PORT_D] = SDE_DDI_HOTPLUG_ICP(HPD_PORT_D), - [HPD_PORT_TC1] = SDE_TC_HOTPLUG_DG2(HPD_PORT_TC1), -}; - -static void intel_hpd_init_pins(struct drm_i915_private *dev_priv) -{ - struct intel_hotplug *hpd = &dev_priv->display.hotplug; - - if (HAS_GMCH(dev_priv)) { - if (IS_G4X(dev_priv) || IS_VALLEYVIEW(dev_priv) || - IS_CHERRYVIEW(dev_priv)) - hpd->hpd = hpd_status_g4x; - else - hpd->hpd = hpd_status_i915; - return; - } - - if (DISPLAY_VER(dev_priv) >= 11) - hpd->hpd = hpd_gen11; - else if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) - hpd->hpd = hpd_bxt; - else if (DISPLAY_VER(dev_priv) == 9) - hpd->hpd = NULL; /* no north HPD on SKL */ - else if (DISPLAY_VER(dev_priv) >= 8) - hpd->hpd = hpd_bdw; - else if (DISPLAY_VER(dev_priv) >= 7) - hpd->hpd = hpd_ivb; - else - hpd->hpd = hpd_ilk; - - if ((INTEL_PCH_TYPE(dev_priv) < PCH_DG1) && - (!HAS_PCH_SPLIT(dev_priv) || HAS_PCH_NOP(dev_priv))) - return; - - if (INTEL_PCH_TYPE(dev_priv) >= PCH_DG1) - hpd->pch_hpd = hpd_sde_dg1; - else if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP) - hpd->pch_hpd = hpd_icp; - else if (HAS_PCH_CNP(dev_priv) || HAS_PCH_SPT(dev_priv)) - hpd->pch_hpd = hpd_spt; - else if (HAS_PCH_LPT(dev_priv) || HAS_PCH_CPT(dev_priv)) - hpd->pch_hpd = hpd_cpt; - else if (HAS_PCH_IBX(dev_priv)) - hpd->pch_hpd = hpd_ibx; - else - MISSING_CASE(INTEL_PCH_TYPE(dev_priv)); -} - -static void -intel_handle_vblank(struct drm_i915_private *dev_priv, enum pipe pipe) -{ - struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv, pipe); - - drm_crtc_handle_vblank(&crtc->base); -} - void gen3_irq_reset(struct intel_uncore *uncore, i915_reg_t imr, i915_reg_t iir, i915_reg_t ier) { @@ -266,7 +110,7 @@ static void gen2_irq_reset(struct intel_uncore *uncore) /* * We should clear IMR at preinstall/uninstall, and just check at postinstall. */ -static void gen3_assert_iir_is_zero(struct intel_uncore *uncore, i915_reg_t reg) +void gen3_assert_iir_is_zero(struct intel_uncore *uncore, i915_reg_t reg) { u32 val = intel_uncore_read(uncore, reg); @@ -320,302 +164,6 @@ static void gen2_irq_init(struct intel_uncore *uncore, intel_uncore_posting_read16(uncore, GEN2_IMR); } -/* For display hotplug interrupt */ -static inline void -i915_hotplug_interrupt_update_locked(struct drm_i915_private *dev_priv, - u32 mask, - u32 bits) -{ - lockdep_assert_held(&dev_priv->irq_lock); - drm_WARN_ON(&dev_priv->drm, bits & ~mask); - - intel_uncore_rmw(&dev_priv->uncore, PORT_HOTPLUG_EN, mask, bits); -} - -/** - * i915_hotplug_interrupt_update - update hotplug interrupt enable - * @dev_priv: driver private - * @mask: bits to update - * @bits: bits to enable - * NOTE: the HPD enable bits are modified both inside and outside - * of an interrupt context. To avoid that read-modify-write cycles - * interfer, these bits are protected by a spinlock. Since this - * function is usually not called from a context where the lock is - * held already, this function acquires the lock itself. A non-locking - * version is also available. - */ -void i915_hotplug_interrupt_update(struct drm_i915_private *dev_priv, - u32 mask, - u32 bits) -{ - spin_lock_irq(&dev_priv->irq_lock); - i915_hotplug_interrupt_update_locked(dev_priv, mask, bits); - spin_unlock_irq(&dev_priv->irq_lock); -} - -/** - * ilk_update_display_irq - update DEIMR - * @dev_priv: driver private - * @interrupt_mask: mask of interrupt bits to update - * @enabled_irq_mask: mask of interrupt bits to enable - */ -static void ilk_update_display_irq(struct drm_i915_private *dev_priv, - u32 interrupt_mask, u32 enabled_irq_mask) -{ - u32 new_val; - - lockdep_assert_held(&dev_priv->irq_lock); - drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask); - - new_val = dev_priv->irq_mask; - new_val &= ~interrupt_mask; - new_val |= (~enabled_irq_mask & interrupt_mask); - - if (new_val != dev_priv->irq_mask && - !drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv))) { - dev_priv->irq_mask = new_val; - intel_uncore_write(&dev_priv->uncore, DEIMR, dev_priv->irq_mask); - intel_uncore_posting_read(&dev_priv->uncore, DEIMR); - } -} - -void ilk_enable_display_irq(struct drm_i915_private *i915, u32 bits) -{ - ilk_update_display_irq(i915, bits, bits); -} - -void ilk_disable_display_irq(struct drm_i915_private *i915, u32 bits) -{ - ilk_update_display_irq(i915, bits, 0); -} - -/** - * bdw_update_port_irq - update DE port interrupt - * @dev_priv: driver private - * @interrupt_mask: mask of interrupt bits to update - * @enabled_irq_mask: mask of interrupt bits to enable - */ -static void bdw_update_port_irq(struct drm_i915_private *dev_priv, - u32 interrupt_mask, - u32 enabled_irq_mask) -{ - u32 new_val; - u32 old_val; - - lockdep_assert_held(&dev_priv->irq_lock); - - drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask); - - if (drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv))) - return; - - old_val = intel_uncore_read(&dev_priv->uncore, GEN8_DE_PORT_IMR); - - new_val = old_val; - new_val &= ~interrupt_mask; - new_val |= (~enabled_irq_mask & interrupt_mask); - - if (new_val != old_val) { - intel_uncore_write(&dev_priv->uncore, GEN8_DE_PORT_IMR, new_val); - intel_uncore_posting_read(&dev_priv->uncore, GEN8_DE_PORT_IMR); - } -} - -/** - * bdw_update_pipe_irq - update DE pipe interrupt - * @dev_priv: driver private - * @pipe: pipe whose interrupt to update - * @interrupt_mask: mask of interrupt bits to update - * @enabled_irq_mask: mask of interrupt bits to enable - */ -static void bdw_update_pipe_irq(struct drm_i915_private *dev_priv, - enum pipe pipe, u32 interrupt_mask, - u32 enabled_irq_mask) -{ - u32 new_val; - - lockdep_assert_held(&dev_priv->irq_lock); - - drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask); - - if (drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv))) - return; - - new_val = dev_priv->de_irq_mask[pipe]; - new_val &= ~interrupt_mask; - new_val |= (~enabled_irq_mask & interrupt_mask); - - if (new_val != dev_priv->de_irq_mask[pipe]) { - dev_priv->de_irq_mask[pipe] = new_val; - intel_uncore_write(&dev_priv->uncore, GEN8_DE_PIPE_IMR(pipe), dev_priv->de_irq_mask[pipe]); - intel_uncore_posting_read(&dev_priv->uncore, GEN8_DE_PIPE_IMR(pipe)); - } -} - -void bdw_enable_pipe_irq(struct drm_i915_private *i915, - enum pipe pipe, u32 bits) -{ - bdw_update_pipe_irq(i915, pipe, bits, bits); -} - -void bdw_disable_pipe_irq(struct drm_i915_private *i915, - enum pipe pipe, u32 bits) -{ - bdw_update_pipe_irq(i915, pipe, bits, 0); -} - -/** - * ibx_display_interrupt_update - update SDEIMR - * @dev_priv: driver private - * @interrupt_mask: mask of interrupt bits to update - * @enabled_irq_mask: mask of interrupt bits to enable - */ -static void ibx_display_interrupt_update(struct drm_i915_private *dev_priv, - u32 interrupt_mask, - u32 enabled_irq_mask) -{ - u32 sdeimr = intel_uncore_read(&dev_priv->uncore, SDEIMR); - sdeimr &= ~interrupt_mask; - sdeimr |= (~enabled_irq_mask & interrupt_mask); - - drm_WARN_ON(&dev_priv->drm, enabled_irq_mask & ~interrupt_mask); - - lockdep_assert_held(&dev_priv->irq_lock); - - if (drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv))) - return; - - intel_uncore_write(&dev_priv->uncore, SDEIMR, sdeimr); - intel_uncore_posting_read(&dev_priv->uncore, SDEIMR); -} - -void ibx_enable_display_interrupt(struct drm_i915_private *i915, u32 bits) -{ - ibx_display_interrupt_update(i915, bits, bits); -} - -void ibx_disable_display_interrupt(struct drm_i915_private *i915, u32 bits) -{ - ibx_display_interrupt_update(i915, bits, 0); -} - -u32 i915_pipestat_enable_mask(struct drm_i915_private *dev_priv, - enum pipe pipe) -{ - u32 status_mask = dev_priv->pipestat_irq_mask[pipe]; - u32 enable_mask = status_mask << 16; - - lockdep_assert_held(&dev_priv->irq_lock); - - if (DISPLAY_VER(dev_priv) < 5) - goto out; - - /* - * On pipe A we don't support the PSR interrupt yet, - * on pipe B and C the same bit MBZ. - */ - if (drm_WARN_ON_ONCE(&dev_priv->drm, - status_mask & PIPE_A_PSR_STATUS_VLV)) - return 0; - /* - * On pipe B and C we don't support the PSR interrupt yet, on pipe - * A the same bit is for perf counters which we don't use either. - */ - if (drm_WARN_ON_ONCE(&dev_priv->drm, - status_mask & PIPE_B_PSR_STATUS_VLV)) - return 0; - - enable_mask &= ~(PIPE_FIFO_UNDERRUN_STATUS | - SPRITE0_FLIP_DONE_INT_EN_VLV | - SPRITE1_FLIP_DONE_INT_EN_VLV); - if (status_mask & SPRITE0_FLIP_DONE_INT_STATUS_VLV) - enable_mask |= SPRITE0_FLIP_DONE_INT_EN_VLV; - if (status_mask & SPRITE1_FLIP_DONE_INT_STATUS_VLV) - enable_mask |= SPRITE1_FLIP_DONE_INT_EN_VLV; - -out: - drm_WARN_ONCE(&dev_priv->drm, - enable_mask & ~PIPESTAT_INT_ENABLE_MASK || - status_mask & ~PIPESTAT_INT_STATUS_MASK, - "pipe %c: enable_mask=0x%x, status_mask=0x%x\n", - pipe_name(pipe), enable_mask, status_mask); - - return enable_mask; -} - -void i915_enable_pipestat(struct drm_i915_private *dev_priv, - enum pipe pipe, u32 status_mask) -{ - i915_reg_t reg = PIPESTAT(pipe); - u32 enable_mask; - - drm_WARN_ONCE(&dev_priv->drm, status_mask & ~PIPESTAT_INT_STATUS_MASK, - "pipe %c: status_mask=0x%x\n", - pipe_name(pipe), status_mask); - - lockdep_assert_held(&dev_priv->irq_lock); - drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv)); - - if ((dev_priv->pipestat_irq_mask[pipe] & status_mask) == status_mask) - return; - - dev_priv->pipestat_irq_mask[pipe] |= status_mask; - enable_mask = i915_pipestat_enable_mask(dev_priv, pipe); - - intel_uncore_write(&dev_priv->uncore, reg, enable_mask | status_mask); - intel_uncore_posting_read(&dev_priv->uncore, reg); -} - -void i915_disable_pipestat(struct drm_i915_private *dev_priv, - enum pipe pipe, u32 status_mask) -{ - i915_reg_t reg = PIPESTAT(pipe); - u32 enable_mask; - - drm_WARN_ONCE(&dev_priv->drm, status_mask & ~PIPESTAT_INT_STATUS_MASK, - "pipe %c: status_mask=0x%x\n", - pipe_name(pipe), status_mask); - - lockdep_assert_held(&dev_priv->irq_lock); - drm_WARN_ON(&dev_priv->drm, !intel_irqs_enabled(dev_priv)); - - if ((dev_priv->pipestat_irq_mask[pipe] & status_mask) == 0) - return; - - dev_priv->pipestat_irq_mask[pipe] &= ~status_mask; - enable_mask = i915_pipestat_enable_mask(dev_priv, pipe); - - intel_uncore_write(&dev_priv->uncore, reg, enable_mask | status_mask); - intel_uncore_posting_read(&dev_priv->uncore, reg); -} - -static bool i915_has_asle(struct drm_i915_private *dev_priv) -{ - if (!dev_priv->display.opregion.asle) - return false; - - return IS_PINEVIEW(dev_priv) || IS_MOBILE(dev_priv); -} - -/** - * i915_enable_asle_pipestat - enable ASLE pipestat for OpRegion - * @dev_priv: i915 device private - */ -static void i915_enable_asle_pipestat(struct drm_i915_private *dev_priv) -{ - if (!i915_has_asle(dev_priv)) - return; - - spin_lock_irq(&dev_priv->irq_lock); - - i915_enable_pipestat(dev_priv, PIPE_B, PIPE_LEGACY_BLC_EVENT_STATUS); - if (DISPLAY_VER(dev_priv) >= 4) - i915_enable_pipestat(dev_priv, PIPE_A, - PIPE_LEGACY_BLC_EVENT_STATUS); - - spin_unlock_irq(&dev_priv->irq_lock); -} - /** * ivb_parity_work - Workqueue called when a parity error interrupt * occurred. @@ -700,640 +248,102 @@ static void ivb_parity_work(struct work_struct *work) mutex_unlock(&dev_priv->drm.struct_mutex); } -static bool gen11_port_hotplug_long_detect(enum hpd_pin pin, u32 val) -{ - switch (pin) { - case HPD_PORT_TC1: - case HPD_PORT_TC2: - case HPD_PORT_TC3: - case HPD_PORT_TC4: - case HPD_PORT_TC5: - case HPD_PORT_TC6: - return val & GEN11_HOTPLUG_CTL_LONG_DETECT(pin); - default: - return false; - } -} - -static bool bxt_port_hotplug_long_detect(enum hpd_pin pin, u32 val) -{ - switch (pin) { - case HPD_PORT_A: - return val & PORTA_HOTPLUG_LONG_DETECT; - case HPD_PORT_B: - return val & PORTB_HOTPLUG_LONG_DETECT; - case HPD_PORT_C: - return val & PORTC_HOTPLUG_LONG_DETECT; - default: - return false; - } -} - -static bool icp_ddi_port_hotplug_long_detect(enum hpd_pin pin, u32 val) +static irqreturn_t valleyview_irq_handler(int irq, void *arg) { - switch (pin) { - case HPD_PORT_A: - case HPD_PORT_B: - case HPD_PORT_C: - case HPD_PORT_D: - return val & SHOTPLUG_CTL_DDI_HPD_LONG_DETECT(pin); - default: - return false; - } -} + struct drm_i915_private *dev_priv = arg; + irqreturn_t ret = IRQ_NONE; -static bool icp_tc_port_hotplug_long_detect(enum hpd_pin pin, u32 val) -{ - switch (pin) { - case HPD_PORT_TC1: - case HPD_PORT_TC2: - case HPD_PORT_TC3: - case HPD_PORT_TC4: - case HPD_PORT_TC5: - case HPD_PORT_TC6: - return val & ICP_TC_HPD_LONG_DETECT(pin); - default: - return false; - } -} + if (!intel_irqs_enabled(dev_priv)) + return IRQ_NONE; -static bool spt_port_hotplug2_long_detect(enum hpd_pin pin, u32 val) -{ - switch (pin) { - case HPD_PORT_E: - return val & PORTE_HOTPLUG_LONG_DETECT; - default: - return false; - } -} + /* IRQs are synced during runtime_suspend, we don't require a wakeref */ + disable_rpm_wakeref_asserts(&dev_priv->runtime_pm); -static bool spt_port_hotplug_long_detect(enum hpd_pin pin, u32 val) -{ - switch (pin) { - case HPD_PORT_A: - return val & PORTA_HOTPLUG_LONG_DETECT; - case HPD_PORT_B: - return val & PORTB_HOTPLUG_LONG_DETECT; - case HPD_PORT_C: - return val & PORTC_HOTPLUG_LONG_DETECT; - case HPD_PORT_D: - return val & PORTD_HOTPLUG_LONG_DETECT; - default: - return false; - } -} + do { + u32 iir, gt_iir, pm_iir; + u32 pipe_stats[I915_MAX_PIPES] = {}; + u32 hotplug_status = 0; + u32 ier = 0; -static bool ilk_port_hotplug_long_detect(enum hpd_pin pin, u32 val) -{ - switch (pin) { - case HPD_PORT_A: - return val & DIGITAL_PORTA_HOTPLUG_LONG_DETECT; - default: - return false; - } -} + gt_iir = intel_uncore_read(&dev_priv->uncore, GTIIR); + pm_iir = intel_uncore_read(&dev_priv->uncore, GEN6_PMIIR); + iir = intel_uncore_read(&dev_priv->uncore, VLV_IIR); -static bool pch_port_hotplug_long_detect(enum hpd_pin pin, u32 val) -{ - switch (pin) { - case HPD_PORT_B: - return val & PORTB_HOTPLUG_LONG_DETECT; - case HPD_PORT_C: - return val & PORTC_HOTPLUG_LONG_DETECT; - case HPD_PORT_D: - return val & PORTD_HOTPLUG_LONG_DETECT; - default: - return false; - } -} + if (gt_iir == 0 && pm_iir == 0 && iir == 0) + break; -static bool i9xx_port_hotplug_long_detect(enum hpd_pin pin, u32 val) -{ - switch (pin) { - case HPD_PORT_B: - return val & PORTB_HOTPLUG_INT_LONG_PULSE; - case HPD_PORT_C: - return val & PORTC_HOTPLUG_INT_LONG_PULSE; - case HPD_PORT_D: - return val & PORTD_HOTPLUG_INT_LONG_PULSE; - default: - return false; - } -} + ret = IRQ_HANDLED; -/* - * Get a bit mask of pins that have triggered, and which ones may be long. - * This can be called multiple times with the same masks to accumulate - * hotplug detection results from several registers. - * - * Note that the caller is expected to zero out the masks initially. - */ -static void intel_get_hpd_pins(struct drm_i915_private *dev_priv, - u32 *pin_mask, u32 *long_mask, - u32 hotplug_trigger, u32 dig_hotplug_reg, - const u32 hpd[HPD_NUM_PINS], - bool long_pulse_detect(enum hpd_pin pin, u32 val)) -{ - enum hpd_pin pin; + /* + * Theory on interrupt generation, based on empirical evidence: + * + * x = ((VLV_IIR & VLV_IER) || + * (((GT_IIR & GT_IER) || (GEN6_PMIIR & GEN6_PMIER)) && + * (VLV_MASTER_IER & MASTER_INTERRUPT_ENABLE))); + * + * A CPU interrupt will only be raised when 'x' has a 0->1 edge. + * Hence we clear MASTER_INTERRUPT_ENABLE and VLV_IER to + * guarantee the CPU interrupt will be raised again even if we + * don't end up clearing all the VLV_IIR, GT_IIR, GEN6_PMIIR + * bits this time around. + */ + intel_uncore_write(&dev_priv->uncore, VLV_MASTER_IER, 0); + ier = intel_uncore_rmw(&dev_priv->uncore, VLV_IER, ~0, 0); - BUILD_BUG_ON(BITS_PER_TYPE(*pin_mask) < HPD_NUM_PINS); + if (gt_iir) + intel_uncore_write(&dev_priv->uncore, GTIIR, gt_iir); + if (pm_iir) + intel_uncore_write(&dev_priv->uncore, GEN6_PMIIR, pm_iir); - for_each_hpd_pin(pin) { - if ((hpd[pin] & hotplug_trigger) == 0) - continue; + if (iir & I915_DISPLAY_PORT_INTERRUPT) + hotplug_status = i9xx_hpd_irq_ack(dev_priv); - *pin_mask |= BIT(pin); + /* Call regardless, as some status bits might not be + * signalled in iir */ + i9xx_pipestat_irq_ack(dev_priv, iir, pipe_stats); - if (long_pulse_detect(pin, dig_hotplug_reg)) - *long_mask |= BIT(pin); - } + if (iir & (I915_LPE_PIPE_A_INTERRUPT | + I915_LPE_PIPE_B_INTERRUPT)) + intel_lpe_audio_irq_handler(dev_priv); - drm_dbg(&dev_priv->drm, - "hotplug event received, stat 0x%08x, dig 0x%08x, pins 0x%08x, long 0x%08x\n", - hotplug_trigger, dig_hotplug_reg, *pin_mask, *long_mask); + /* + * VLV_IIR is single buffered, and reflects the level + * from PIPESTAT/PORT_HOTPLUG_STAT, hence clear it last. + */ + if (iir) + intel_uncore_write(&dev_priv->uncore, VLV_IIR, iir); -} + intel_uncore_write(&dev_priv->uncore, VLV_IER, ier); + intel_uncore_write(&dev_priv->uncore, VLV_MASTER_IER, MASTER_INTERRUPT_ENABLE); -static u32 intel_hpd_enabled_irqs(struct drm_i915_private *dev_priv, - const u32 hpd[HPD_NUM_PINS]) -{ - struct intel_encoder *encoder; - u32 enabled_irqs = 0; + if (gt_iir) + gen6_gt_irq_handler(to_gt(dev_priv), gt_iir); + if (pm_iir) + gen6_rps_irq_handler(&to_gt(dev_priv)->rps, pm_iir); - for_each_intel_encoder(&dev_priv->drm, encoder) - if (dev_priv->display.hotplug.stats[encoder->hpd_pin].state == HPD_ENABLED) - enabled_irqs |= hpd[encoder->hpd_pin]; + if (hotplug_status) + i9xx_hpd_irq_handler(dev_priv, hotplug_status); - return enabled_irqs; -} + valleyview_pipestat_irq_handler(dev_priv, pipe_stats); + } while (0); -static u32 intel_hpd_hotplug_irqs(struct drm_i915_private *dev_priv, - const u32 hpd[HPD_NUM_PINS]) -{ - struct intel_encoder *encoder; - u32 hotplug_irqs = 0; + pmu_irq_stats(dev_priv, ret); - for_each_intel_encoder(&dev_priv->drm, encoder) - hotplug_irqs |= hpd[encoder->hpd_pin]; + enable_rpm_wakeref_asserts(&dev_priv->runtime_pm); - return hotplug_irqs; + return ret; } -static u32 intel_hpd_hotplug_enables(struct drm_i915_private *i915, - hotplug_enables_func hotplug_enables) +static irqreturn_t cherryview_irq_handler(int irq, void *arg) { - struct intel_encoder *encoder; - u32 hotplug = 0; - - for_each_intel_encoder(&i915->drm, encoder) - hotplug |= hotplug_enables(encoder); + struct drm_i915_private *dev_priv = arg; + irqreturn_t ret = IRQ_NONE; - return hotplug; -} + if (!intel_irqs_enabled(dev_priv)) + return IRQ_NONE; -static void gmbus_irq_handler(struct drm_i915_private *dev_priv) -{ - wake_up_all(&dev_priv->display.gmbus.wait_queue); -} - -static void dp_aux_irq_handler(struct drm_i915_private *dev_priv) -{ - wake_up_all(&dev_priv->display.gmbus.wait_queue); -} - -#if defined(CONFIG_DEBUG_FS) -static void display_pipe_crc_irq_handler(struct drm_i915_private *dev_priv, - enum pipe pipe, - u32 crc0, u32 crc1, - u32 crc2, u32 crc3, - u32 crc4) -{ - struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv, pipe); - struct intel_pipe_crc *pipe_crc = &crtc->pipe_crc; - u32 crcs[5] = { crc0, crc1, crc2, crc3, crc4 }; - - trace_intel_pipe_crc(crtc, crcs); - - spin_lock(&pipe_crc->lock); - /* - * For some not yet identified reason, the first CRC is - * bonkers. So let's just wait for the next vblank and read - * out the buggy result. - * - * On GEN8+ sometimes the second CRC is bonkers as well, so - * don't trust that one either. - */ - if (pipe_crc->skipped <= 0 || - (DISPLAY_VER(dev_priv) >= 8 && pipe_crc->skipped == 1)) { - pipe_crc->skipped++; - spin_unlock(&pipe_crc->lock); - return; - } - spin_unlock(&pipe_crc->lock); - - drm_crtc_add_crc_entry(&crtc->base, true, - drm_crtc_accurate_vblank_count(&crtc->base), - crcs); -} -#else -static inline void -display_pipe_crc_irq_handler(struct drm_i915_private *dev_priv, - enum pipe pipe, - u32 crc0, u32 crc1, - u32 crc2, u32 crc3, - u32 crc4) {} -#endif - -static void flip_done_handler(struct drm_i915_private *i915, - enum pipe pipe) -{ - struct intel_crtc *crtc = intel_crtc_for_pipe(i915, pipe); - struct drm_crtc_state *crtc_state = crtc->base.state; - struct drm_pending_vblank_event *e = crtc_state->event; - struct drm_device *dev = &i915->drm; - unsigned long irqflags; - - spin_lock_irqsave(&dev->event_lock, irqflags); - - crtc_state->event = NULL; - - drm_crtc_send_vblank_event(&crtc->base, e); - - spin_unlock_irqrestore(&dev->event_lock, irqflags); -} - -static void hsw_pipe_crc_irq_handler(struct drm_i915_private *dev_priv, - enum pipe pipe) -{ - display_pipe_crc_irq_handler(dev_priv, pipe, - intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_1_IVB(pipe)), - 0, 0, 0, 0); -} - -static void ivb_pipe_crc_irq_handler(struct drm_i915_private *dev_priv, - enum pipe pipe) -{ - display_pipe_crc_irq_handler(dev_priv, pipe, - intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_1_IVB(pipe)), - intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_2_IVB(pipe)), - intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_3_IVB(pipe)), - intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_4_IVB(pipe)), - intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_5_IVB(pipe))); -} - -static void i9xx_pipe_crc_irq_handler(struct drm_i915_private *dev_priv, - enum pipe pipe) -{ - u32 res1, res2; - - if (DISPLAY_VER(dev_priv) >= 3) - res1 = intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_RES1_I915(pipe)); - else - res1 = 0; - - if (DISPLAY_VER(dev_priv) >= 5 || IS_G4X(dev_priv)) - res2 = intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_RES2_G4X(pipe)); - else - res2 = 0; - - display_pipe_crc_irq_handler(dev_priv, pipe, - intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_RED(pipe)), - intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_GREEN(pipe)), - intel_uncore_read(&dev_priv->uncore, PIPE_CRC_RES_BLUE(pipe)), - res1, res2); -} - -static void i9xx_pipestat_irq_reset(struct drm_i915_private *dev_priv) -{ - enum pipe pipe; - - for_each_pipe(dev_priv, pipe) { - intel_uncore_write(&dev_priv->uncore, PIPESTAT(pipe), - PIPESTAT_INT_STATUS_MASK | - PIPE_FIFO_UNDERRUN_STATUS); - - dev_priv->pipestat_irq_mask[pipe] = 0; - } -} - -static void i9xx_pipestat_irq_ack(struct drm_i915_private *dev_priv, - u32 iir, u32 pipe_stats[I915_MAX_PIPES]) -{ - enum pipe pipe; - - spin_lock(&dev_priv->irq_lock); - - if (!dev_priv->display_irqs_enabled) { - spin_unlock(&dev_priv->irq_lock); - return; - } - - for_each_pipe(dev_priv, pipe) { - i915_reg_t reg; - u32 status_mask, enable_mask, iir_bit = 0; - - /* - * PIPESTAT bits get signalled even when the interrupt is - * disabled with the mask bits, and some of the status bits do - * not generate interrupts at all (like the underrun bit). Hence - * we need to be careful that we only handle what we want to - * handle. - */ - - /* fifo underruns are filterered in the underrun handler. */ - status_mask = PIPE_FIFO_UNDERRUN_STATUS; - - switch (pipe) { - default: - case PIPE_A: - iir_bit = I915_DISPLAY_PIPE_A_EVENT_INTERRUPT; - break; - case PIPE_B: - iir_bit = I915_DISPLAY_PIPE_B_EVENT_INTERRUPT; - break; - case PIPE_C: - iir_bit = I915_DISPLAY_PIPE_C_EVENT_INTERRUPT; - break; - } - if (iir & iir_bit) - status_mask |= dev_priv->pipestat_irq_mask[pipe]; - - if (!status_mask) - continue; - - reg = PIPESTAT(pipe); - pipe_stats[pipe] = intel_uncore_read(&dev_priv->uncore, reg) & status_mask; - enable_mask = i915_pipestat_enable_mask(dev_priv, pipe); - - /* - * Clear the PIPE*STAT regs before the IIR - * - * Toggle the enable bits to make sure we get an - * edge in the ISR pipe event bit if we don't clear - * all the enabled status bits. Otherwise the edge - * triggered IIR on i965/g4x wouldn't notice that - * an interrupt is still pending. - */ - if (pipe_stats[pipe]) { - intel_uncore_write(&dev_priv->uncore, reg, pipe_stats[pipe]); - intel_uncore_write(&dev_priv->uncore, reg, enable_mask); - } - } - spin_unlock(&dev_priv->irq_lock); -} - -static void i8xx_pipestat_irq_handler(struct drm_i915_private *dev_priv, - u16 iir, u32 pipe_stats[I915_MAX_PIPES]) -{ - enum pipe pipe; - - for_each_pipe(dev_priv, pipe) { - if (pipe_stats[pipe] & PIPE_VBLANK_INTERRUPT_STATUS) - intel_handle_vblank(dev_priv, pipe); - - if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS) - i9xx_pipe_crc_irq_handler(dev_priv, pipe); - - if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS) - intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe); - } -} - -static void i915_pipestat_irq_handler(struct drm_i915_private *dev_priv, - u32 iir, u32 pipe_stats[I915_MAX_PIPES]) -{ - bool blc_event = false; - enum pipe pipe; - - for_each_pipe(dev_priv, pipe) { - if (pipe_stats[pipe] & PIPE_VBLANK_INTERRUPT_STATUS) - intel_handle_vblank(dev_priv, pipe); - - if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS) - blc_event = true; - - if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS) - i9xx_pipe_crc_irq_handler(dev_priv, pipe); - - if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS) - intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe); - } - - if (blc_event || (iir & I915_ASLE_INTERRUPT)) - intel_opregion_asle_intr(dev_priv); -} - -static void i965_pipestat_irq_handler(struct drm_i915_private *dev_priv, - u32 iir, u32 pipe_stats[I915_MAX_PIPES]) -{ - bool blc_event = false; - enum pipe pipe; - - for_each_pipe(dev_priv, pipe) { - if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS) - intel_handle_vblank(dev_priv, pipe); - - if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS) - blc_event = true; - - if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS) - i9xx_pipe_crc_irq_handler(dev_priv, pipe); - - if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS) - intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe); - } - - if (blc_event || (iir & I915_ASLE_INTERRUPT)) - intel_opregion_asle_intr(dev_priv); - - if (pipe_stats[0] & PIPE_GMBUS_INTERRUPT_STATUS) - gmbus_irq_handler(dev_priv); -} - -static void valleyview_pipestat_irq_handler(struct drm_i915_private *dev_priv, - u32 pipe_stats[I915_MAX_PIPES]) -{ - enum pipe pipe; - - for_each_pipe(dev_priv, pipe) { - if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS) - intel_handle_vblank(dev_priv, pipe); - - if (pipe_stats[pipe] & PLANE_FLIP_DONE_INT_STATUS_VLV) - flip_done_handler(dev_priv, pipe); - - if (pipe_stats[pipe] & PIPE_CRC_DONE_INTERRUPT_STATUS) - i9xx_pipe_crc_irq_handler(dev_priv, pipe); - - if (pipe_stats[pipe] & PIPE_FIFO_UNDERRUN_STATUS) - intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe); - } - - if (pipe_stats[0] & PIPE_GMBUS_INTERRUPT_STATUS) - gmbus_irq_handler(dev_priv); -} - -static u32 i9xx_hpd_irq_ack(struct drm_i915_private *dev_priv) -{ - u32 hotplug_status = 0, hotplug_status_mask; - int i; - - if (IS_G4X(dev_priv) || - IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) - hotplug_status_mask = HOTPLUG_INT_STATUS_G4X | - DP_AUX_CHANNEL_MASK_INT_STATUS_G4X; - else - hotplug_status_mask = HOTPLUG_INT_STATUS_I915; - - /* - * We absolutely have to clear all the pending interrupt - * bits in PORT_HOTPLUG_STAT. Otherwise the ISR port - * interrupt bit won't have an edge, and the i965/g4x - * edge triggered IIR will not notice that an interrupt - * is still pending. We can't use PORT_HOTPLUG_EN to - * guarantee the edge as the act of toggling the enable - * bits can itself generate a new hotplug interrupt :( - */ - for (i = 0; i < 10; i++) { - u32 tmp = intel_uncore_read(&dev_priv->uncore, PORT_HOTPLUG_STAT) & hotplug_status_mask; - - if (tmp == 0) - return hotplug_status; - - hotplug_status |= tmp; - intel_uncore_write(&dev_priv->uncore, PORT_HOTPLUG_STAT, hotplug_status); - } - - drm_WARN_ONCE(&dev_priv->drm, 1, - "PORT_HOTPLUG_STAT did not clear (0x%08x)\n", - intel_uncore_read(&dev_priv->uncore, PORT_HOTPLUG_STAT)); - - return hotplug_status; -} - -static void i9xx_hpd_irq_handler(struct drm_i915_private *dev_priv, - u32 hotplug_status) -{ - u32 pin_mask = 0, long_mask = 0; - u32 hotplug_trigger; - - if (IS_G4X(dev_priv) || - IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) - hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_G4X; - else - hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_I915; - - if (hotplug_trigger) { - intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, - hotplug_trigger, hotplug_trigger, - dev_priv->display.hotplug.hpd, - i9xx_port_hotplug_long_detect); - - intel_hpd_irq_handler(dev_priv, pin_mask, long_mask); - } - - if ((IS_G4X(dev_priv) || - IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) && - hotplug_status & DP_AUX_CHANNEL_MASK_INT_STATUS_G4X) - dp_aux_irq_handler(dev_priv); -} - -static irqreturn_t valleyview_irq_handler(int irq, void *arg) -{ - struct drm_i915_private *dev_priv = arg; - irqreturn_t ret = IRQ_NONE; - - if (!intel_irqs_enabled(dev_priv)) - return IRQ_NONE; - - /* IRQs are synced during runtime_suspend, we don't require a wakeref */ - disable_rpm_wakeref_asserts(&dev_priv->runtime_pm); - - do { - u32 iir, gt_iir, pm_iir; - u32 pipe_stats[I915_MAX_PIPES] = {}; - u32 hotplug_status = 0; - u32 ier = 0; - - gt_iir = intel_uncore_read(&dev_priv->uncore, GTIIR); - pm_iir = intel_uncore_read(&dev_priv->uncore, GEN6_PMIIR); - iir = intel_uncore_read(&dev_priv->uncore, VLV_IIR); - - if (gt_iir == 0 && pm_iir == 0 && iir == 0) - break; - - ret = IRQ_HANDLED; - - /* - * Theory on interrupt generation, based on empirical evidence: - * - * x = ((VLV_IIR & VLV_IER) || - * (((GT_IIR & GT_IER) || (GEN6_PMIIR & GEN6_PMIER)) && - * (VLV_MASTER_IER & MASTER_INTERRUPT_ENABLE))); - * - * A CPU interrupt will only be raised when 'x' has a 0->1 edge. - * Hence we clear MASTER_INTERRUPT_ENABLE and VLV_IER to - * guarantee the CPU interrupt will be raised again even if we - * don't end up clearing all the VLV_IIR, GT_IIR, GEN6_PMIIR - * bits this time around. - */ - intel_uncore_write(&dev_priv->uncore, VLV_MASTER_IER, 0); - ier = intel_uncore_rmw(&dev_priv->uncore, VLV_IER, ~0, 0); - - if (gt_iir) - intel_uncore_write(&dev_priv->uncore, GTIIR, gt_iir); - if (pm_iir) - intel_uncore_write(&dev_priv->uncore, GEN6_PMIIR, pm_iir); - - if (iir & I915_DISPLAY_PORT_INTERRUPT) - hotplug_status = i9xx_hpd_irq_ack(dev_priv); - - /* Call regardless, as some status bits might not be - * signalled in iir */ - i9xx_pipestat_irq_ack(dev_priv, iir, pipe_stats); - - if (iir & (I915_LPE_PIPE_A_INTERRUPT | - I915_LPE_PIPE_B_INTERRUPT)) - intel_lpe_audio_irq_handler(dev_priv); - - /* - * VLV_IIR is single buffered, and reflects the level - * from PIPESTAT/PORT_HOTPLUG_STAT, hence clear it last. - */ - if (iir) - intel_uncore_write(&dev_priv->uncore, VLV_IIR, iir); - - intel_uncore_write(&dev_priv->uncore, VLV_IER, ier); - intel_uncore_write(&dev_priv->uncore, VLV_MASTER_IER, MASTER_INTERRUPT_ENABLE); - - if (gt_iir) - gen6_gt_irq_handler(to_gt(dev_priv), gt_iir); - if (pm_iir) - gen6_rps_irq_handler(&to_gt(dev_priv)->rps, pm_iir); - - if (hotplug_status) - i9xx_hpd_irq_handler(dev_priv, hotplug_status); - - valleyview_pipestat_irq_handler(dev_priv, pipe_stats); - } while (0); - - pmu_irq_stats(dev_priv, ret); - - enable_rpm_wakeref_asserts(&dev_priv->runtime_pm); - - return ret; -} - -static irqreturn_t cherryview_irq_handler(int irq, void *arg) -{ - struct drm_i915_private *dev_priv = arg; - irqreturn_t ret = IRQ_NONE; - - if (!intel_irqs_enabled(dev_priv)) - return IRQ_NONE; - - /* IRQs are synced during runtime_suspend, we don't require a wakeref */ - disable_rpm_wakeref_asserts(&dev_priv->runtime_pm); + /* IRQs are synced during runtime_suspend, we don't require a wakeref */ + disable_rpm_wakeref_asserts(&dev_priv->runtime_pm); do { u32 master_ctl, iir; @@ -1402,374 +412,42 @@ static irqreturn_t cherryview_irq_handler(int irq, void *arg) return ret; } -static void ibx_hpd_irq_handler(struct drm_i915_private *dev_priv, - u32 hotplug_trigger) +/* + * To handle irqs with the minimum potential races with fresh interrupts, we: + * 1 - Disable Master Interrupt Control. + * 2 - Find the source(s) of the interrupt. + * 3 - Clear the Interrupt Identity bits (IIR). + * 4 - Process the interrupt(s) that had bits set in the IIRs. + * 5 - Re-enable Master Interrupt Control. + */ +static irqreturn_t ilk_irq_handler(int irq, void *arg) { - u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0; + struct drm_i915_private *i915 = arg; + void __iomem * const regs = i915->uncore.regs; + u32 de_iir, gt_iir, de_ier, sde_ier = 0; + irqreturn_t ret = IRQ_NONE; - /* - * Somehow the PCH doesn't seem to really ack the interrupt to the CPU - * unless we touch the hotplug register, even if hotplug_trigger is - * zero. Not acking leads to "The master control interrupt lied (SDE)!" - * errors. - */ - dig_hotplug_reg = intel_uncore_read(&dev_priv->uncore, PCH_PORT_HOTPLUG); - if (!hotplug_trigger) { - u32 mask = PORTA_HOTPLUG_STATUS_MASK | - PORTD_HOTPLUG_STATUS_MASK | - PORTC_HOTPLUG_STATUS_MASK | - PORTB_HOTPLUG_STATUS_MASK; - dig_hotplug_reg &= ~mask; - } + if (unlikely(!intel_irqs_enabled(i915))) + return IRQ_NONE; - intel_uncore_write(&dev_priv->uncore, PCH_PORT_HOTPLUG, dig_hotplug_reg); - if (!hotplug_trigger) - return; + /* IRQs are synced during runtime_suspend, we don't require a wakeref */ + disable_rpm_wakeref_asserts(&i915->runtime_pm); - intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, - hotplug_trigger, dig_hotplug_reg, - dev_priv->display.hotplug.pch_hpd, - pch_port_hotplug_long_detect); + /* disable master interrupt before clearing iir */ + de_ier = raw_reg_read(regs, DEIER); + raw_reg_write(regs, DEIER, de_ier & ~DE_MASTER_IRQ_CONTROL); - intel_hpd_irq_handler(dev_priv, pin_mask, long_mask); -} + /* Disable south interrupts. We'll only write to SDEIIR once, so further + * interrupts will will be stored on its back queue, and then we'll be + * able to process them after we restore SDEIER (as soon as we restore + * it, we'll get an interrupt if SDEIIR still has something to process + * due to its back queue). */ + if (!HAS_PCH_NOP(i915)) { + sde_ier = raw_reg_read(regs, SDEIER); + raw_reg_write(regs, SDEIER, 0); + } -static void ibx_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir) -{ - enum pipe pipe; - u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK; - - ibx_hpd_irq_handler(dev_priv, hotplug_trigger); - - if (pch_iir & SDE_AUDIO_POWER_MASK) { - int port = ffs((pch_iir & SDE_AUDIO_POWER_MASK) >> - SDE_AUDIO_POWER_SHIFT); - drm_dbg(&dev_priv->drm, "PCH audio power change on port %d\n", - port_name(port)); - } - - if (pch_iir & SDE_AUX_MASK) - dp_aux_irq_handler(dev_priv); - - if (pch_iir & SDE_GMBUS) - gmbus_irq_handler(dev_priv); - - if (pch_iir & SDE_AUDIO_HDCP_MASK) - drm_dbg(&dev_priv->drm, "PCH HDCP audio interrupt\n"); - - if (pch_iir & SDE_AUDIO_TRANS_MASK) - drm_dbg(&dev_priv->drm, "PCH transcoder audio interrupt\n"); - - if (pch_iir & SDE_POISON) - drm_err(&dev_priv->drm, "PCH poison interrupt\n"); - - if (pch_iir & SDE_FDI_MASK) { - for_each_pipe(dev_priv, pipe) - drm_dbg(&dev_priv->drm, " pipe %c FDI IIR: 0x%08x\n", - pipe_name(pipe), - intel_uncore_read(&dev_priv->uncore, FDI_RX_IIR(pipe))); - } - - if (pch_iir & (SDE_TRANSB_CRC_DONE | SDE_TRANSA_CRC_DONE)) - drm_dbg(&dev_priv->drm, "PCH transcoder CRC done interrupt\n"); - - if (pch_iir & (SDE_TRANSB_CRC_ERR | SDE_TRANSA_CRC_ERR)) - drm_dbg(&dev_priv->drm, - "PCH transcoder CRC error interrupt\n"); - - if (pch_iir & SDE_TRANSA_FIFO_UNDER) - intel_pch_fifo_underrun_irq_handler(dev_priv, PIPE_A); - - if (pch_iir & SDE_TRANSB_FIFO_UNDER) - intel_pch_fifo_underrun_irq_handler(dev_priv, PIPE_B); -} - -static void ivb_err_int_handler(struct drm_i915_private *dev_priv) -{ - u32 err_int = intel_uncore_read(&dev_priv->uncore, GEN7_ERR_INT); - enum pipe pipe; - - if (err_int & ERR_INT_POISON) - drm_err(&dev_priv->drm, "Poison interrupt\n"); - - for_each_pipe(dev_priv, pipe) { - if (err_int & ERR_INT_FIFO_UNDERRUN(pipe)) - intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe); - - if (err_int & ERR_INT_PIPE_CRC_DONE(pipe)) { - if (IS_IVYBRIDGE(dev_priv)) - ivb_pipe_crc_irq_handler(dev_priv, pipe); - else - hsw_pipe_crc_irq_handler(dev_priv, pipe); - } - } - - intel_uncore_write(&dev_priv->uncore, GEN7_ERR_INT, err_int); -} - -static void cpt_serr_int_handler(struct drm_i915_private *dev_priv) -{ - u32 serr_int = intel_uncore_read(&dev_priv->uncore, SERR_INT); - enum pipe pipe; - - if (serr_int & SERR_INT_POISON) - drm_err(&dev_priv->drm, "PCH poison interrupt\n"); - - for_each_pipe(dev_priv, pipe) - if (serr_int & SERR_INT_TRANS_FIFO_UNDERRUN(pipe)) - intel_pch_fifo_underrun_irq_handler(dev_priv, pipe); - - intel_uncore_write(&dev_priv->uncore, SERR_INT, serr_int); -} - -static void cpt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir) -{ - enum pipe pipe; - u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK_CPT; - - ibx_hpd_irq_handler(dev_priv, hotplug_trigger); - - if (pch_iir & SDE_AUDIO_POWER_MASK_CPT) { - int port = ffs((pch_iir & SDE_AUDIO_POWER_MASK_CPT) >> - SDE_AUDIO_POWER_SHIFT_CPT); - drm_dbg(&dev_priv->drm, "PCH audio power change on port %c\n", - port_name(port)); - } - - if (pch_iir & SDE_AUX_MASK_CPT) - dp_aux_irq_handler(dev_priv); - - if (pch_iir & SDE_GMBUS_CPT) - gmbus_irq_handler(dev_priv); - - if (pch_iir & SDE_AUDIO_CP_REQ_CPT) - drm_dbg(&dev_priv->drm, "Audio CP request interrupt\n"); - - if (pch_iir & SDE_AUDIO_CP_CHG_CPT) - drm_dbg(&dev_priv->drm, "Audio CP change interrupt\n"); - - if (pch_iir & SDE_FDI_MASK_CPT) { - for_each_pipe(dev_priv, pipe) - drm_dbg(&dev_priv->drm, " pipe %c FDI IIR: 0x%08x\n", - pipe_name(pipe), - intel_uncore_read(&dev_priv->uncore, FDI_RX_IIR(pipe))); - } - - if (pch_iir & SDE_ERROR_CPT) - cpt_serr_int_handler(dev_priv); -} - -static void icp_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir) -{ - u32 ddi_hotplug_trigger = pch_iir & SDE_DDI_HOTPLUG_MASK_ICP; - u32 tc_hotplug_trigger = pch_iir & SDE_TC_HOTPLUG_MASK_ICP; - u32 pin_mask = 0, long_mask = 0; - - if (ddi_hotplug_trigger) { - u32 dig_hotplug_reg; - - /* Locking due to DSI native GPIO sequences */ - spin_lock(&dev_priv->irq_lock); - dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_DDI, 0, 0); - spin_unlock(&dev_priv->irq_lock); - - intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, - ddi_hotplug_trigger, dig_hotplug_reg, - dev_priv->display.hotplug.pch_hpd, - icp_ddi_port_hotplug_long_detect); - } - - if (tc_hotplug_trigger) { - u32 dig_hotplug_reg; - - dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_TC, 0, 0); - - intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, - tc_hotplug_trigger, dig_hotplug_reg, - dev_priv->display.hotplug.pch_hpd, - icp_tc_port_hotplug_long_detect); - } - - if (pin_mask) - intel_hpd_irq_handler(dev_priv, pin_mask, long_mask); - - if (pch_iir & SDE_GMBUS_ICP) - gmbus_irq_handler(dev_priv); -} - -static void spt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir) -{ - u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK_SPT & - ~SDE_PORTE_HOTPLUG_SPT; - u32 hotplug2_trigger = pch_iir & SDE_PORTE_HOTPLUG_SPT; - u32 pin_mask = 0, long_mask = 0; - - if (hotplug_trigger) { - u32 dig_hotplug_reg; - - dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG, 0, 0); - - intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, - hotplug_trigger, dig_hotplug_reg, - dev_priv->display.hotplug.pch_hpd, - spt_port_hotplug_long_detect); - } - - if (hotplug2_trigger) { - u32 dig_hotplug_reg; - - dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG2, 0, 0); - - intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, - hotplug2_trigger, dig_hotplug_reg, - dev_priv->display.hotplug.pch_hpd, - spt_port_hotplug2_long_detect); - } - - if (pin_mask) - intel_hpd_irq_handler(dev_priv, pin_mask, long_mask); - - if (pch_iir & SDE_GMBUS_CPT) - gmbus_irq_handler(dev_priv); -} - -static void ilk_hpd_irq_handler(struct drm_i915_private *dev_priv, - u32 hotplug_trigger) -{ - u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0; - - dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, DIGITAL_PORT_HOTPLUG_CNTRL, 0, 0); - - intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, - hotplug_trigger, dig_hotplug_reg, - dev_priv->display.hotplug.hpd, - ilk_port_hotplug_long_detect); - - intel_hpd_irq_handler(dev_priv, pin_mask, long_mask); -} - -static void ilk_display_irq_handler(struct drm_i915_private *dev_priv, - u32 de_iir) -{ - enum pipe pipe; - u32 hotplug_trigger = de_iir & DE_DP_A_HOTPLUG; - - if (hotplug_trigger) - ilk_hpd_irq_handler(dev_priv, hotplug_trigger); - - if (de_iir & DE_AUX_CHANNEL_A) - dp_aux_irq_handler(dev_priv); - - if (de_iir & DE_GSE) - intel_opregion_asle_intr(dev_priv); - - if (de_iir & DE_POISON) - drm_err(&dev_priv->drm, "Poison interrupt\n"); - - for_each_pipe(dev_priv, pipe) { - if (de_iir & DE_PIPE_VBLANK(pipe)) - intel_handle_vblank(dev_priv, pipe); - - if (de_iir & DE_PLANE_FLIP_DONE(pipe)) - flip_done_handler(dev_priv, pipe); - - if (de_iir & DE_PIPE_FIFO_UNDERRUN(pipe)) - intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe); - - if (de_iir & DE_PIPE_CRC_DONE(pipe)) - i9xx_pipe_crc_irq_handler(dev_priv, pipe); - } - - /* check event from PCH */ - if (de_iir & DE_PCH_EVENT) { - u32 pch_iir = intel_uncore_read(&dev_priv->uncore, SDEIIR); - - if (HAS_PCH_CPT(dev_priv)) - cpt_irq_handler(dev_priv, pch_iir); - else - ibx_irq_handler(dev_priv, pch_iir); - - /* should clear PCH hotplug event before clear CPU irq */ - intel_uncore_write(&dev_priv->uncore, SDEIIR, pch_iir); - } - - if (DISPLAY_VER(dev_priv) == 5 && de_iir & DE_PCU_EVENT) - gen5_rps_irq_handler(&to_gt(dev_priv)->rps); -} - -static void ivb_display_irq_handler(struct drm_i915_private *dev_priv, - u32 de_iir) -{ - enum pipe pipe; - u32 hotplug_trigger = de_iir & DE_DP_A_HOTPLUG_IVB; - - if (hotplug_trigger) - ilk_hpd_irq_handler(dev_priv, hotplug_trigger); - - if (de_iir & DE_ERR_INT_IVB) - ivb_err_int_handler(dev_priv); - - if (de_iir & DE_AUX_CHANNEL_A_IVB) - dp_aux_irq_handler(dev_priv); - - if (de_iir & DE_GSE_IVB) - intel_opregion_asle_intr(dev_priv); - - for_each_pipe(dev_priv, pipe) { - if (de_iir & DE_PIPE_VBLANK_IVB(pipe)) - intel_handle_vblank(dev_priv, pipe); - - if (de_iir & DE_PLANE_FLIP_DONE_IVB(pipe)) - flip_done_handler(dev_priv, pipe); - } - - /* check event from PCH */ - if (!HAS_PCH_NOP(dev_priv) && (de_iir & DE_PCH_EVENT_IVB)) { - u32 pch_iir = intel_uncore_read(&dev_priv->uncore, SDEIIR); - - cpt_irq_handler(dev_priv, pch_iir); - - /* clear PCH hotplug event before clear CPU irq */ - intel_uncore_write(&dev_priv->uncore, SDEIIR, pch_iir); - } -} - -/* - * To handle irqs with the minimum potential races with fresh interrupts, we: - * 1 - Disable Master Interrupt Control. - * 2 - Find the source(s) of the interrupt. - * 3 - Clear the Interrupt Identity bits (IIR). - * 4 - Process the interrupt(s) that had bits set in the IIRs. - * 5 - Re-enable Master Interrupt Control. - */ -static irqreturn_t ilk_irq_handler(int irq, void *arg) -{ - struct drm_i915_private *i915 = arg; - void __iomem * const regs = i915->uncore.regs; - u32 de_iir, gt_iir, de_ier, sde_ier = 0; - irqreturn_t ret = IRQ_NONE; - - if (unlikely(!intel_irqs_enabled(i915))) - return IRQ_NONE; - - /* IRQs are synced during runtime_suspend, we don't require a wakeref */ - disable_rpm_wakeref_asserts(&i915->runtime_pm); - - /* disable master interrupt before clearing iir */ - de_ier = raw_reg_read(regs, DEIER); - raw_reg_write(regs, DEIER, de_ier & ~DE_MASTER_IRQ_CONTROL); - - /* Disable south interrupts. We'll only write to SDEIIR once, so further - * interrupts will will be stored on its back queue, and then we'll be - * able to process them after we restore SDEIER (as soon as we restore - * it, we'll get an interrupt if SDEIIR still has something to process - * due to its back queue). */ - if (!HAS_PCH_NOP(i915)) { - sde_ier = raw_reg_read(regs, SDEIER); - raw_reg_write(regs, SDEIER, 0); - } - - /* Find, clear, then process each source of interrupt */ + /* Find, clear, then process each source of interrupt */ gt_iir = raw_reg_read(regs, GTIIR); if (gt_iir) { @@ -1812,376 +490,6 @@ static irqreturn_t ilk_irq_handler(int irq, void *arg) return ret; } -static void bxt_hpd_irq_handler(struct drm_i915_private *dev_priv, - u32 hotplug_trigger) -{ - u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0; - - dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG, 0, 0); - - intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, - hotplug_trigger, dig_hotplug_reg, - dev_priv->display.hotplug.hpd, - bxt_port_hotplug_long_detect); - - intel_hpd_irq_handler(dev_priv, pin_mask, long_mask); -} - -static void gen11_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 iir) -{ - u32 pin_mask = 0, long_mask = 0; - u32 trigger_tc = iir & GEN11_DE_TC_HOTPLUG_MASK; - u32 trigger_tbt = iir & GEN11_DE_TBT_HOTPLUG_MASK; - - if (trigger_tc) { - u32 dig_hotplug_reg; - - dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, GEN11_TC_HOTPLUG_CTL, 0, 0); - - intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, - trigger_tc, dig_hotplug_reg, - dev_priv->display.hotplug.hpd, - gen11_port_hotplug_long_detect); - } - - if (trigger_tbt) { - u32 dig_hotplug_reg; - - dig_hotplug_reg = intel_uncore_rmw(&dev_priv->uncore, GEN11_TBT_HOTPLUG_CTL, 0, 0); - - intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, - trigger_tbt, dig_hotplug_reg, - dev_priv->display.hotplug.hpd, - gen11_port_hotplug_long_detect); - } - - if (pin_mask) - intel_hpd_irq_handler(dev_priv, pin_mask, long_mask); - else - drm_err(&dev_priv->drm, - "Unexpected DE HPD interrupt 0x%08x\n", iir); -} - -static u32 gen8_de_port_aux_mask(struct drm_i915_private *dev_priv) -{ - u32 mask; - - if (DISPLAY_VER(dev_priv) >= 13) - return TGL_DE_PORT_AUX_DDIA | - TGL_DE_PORT_AUX_DDIB | - TGL_DE_PORT_AUX_DDIC | - XELPD_DE_PORT_AUX_DDID | - XELPD_DE_PORT_AUX_DDIE | - TGL_DE_PORT_AUX_USBC1 | - TGL_DE_PORT_AUX_USBC2 | - TGL_DE_PORT_AUX_USBC3 | - TGL_DE_PORT_AUX_USBC4; - else if (DISPLAY_VER(dev_priv) >= 12) - return TGL_DE_PORT_AUX_DDIA | - TGL_DE_PORT_AUX_DDIB | - TGL_DE_PORT_AUX_DDIC | - TGL_DE_PORT_AUX_USBC1 | - TGL_DE_PORT_AUX_USBC2 | - TGL_DE_PORT_AUX_USBC3 | - TGL_DE_PORT_AUX_USBC4 | - TGL_DE_PORT_AUX_USBC5 | - TGL_DE_PORT_AUX_USBC6; - - - mask = GEN8_AUX_CHANNEL_A; - if (DISPLAY_VER(dev_priv) >= 9) - mask |= GEN9_AUX_CHANNEL_B | - GEN9_AUX_CHANNEL_C | - GEN9_AUX_CHANNEL_D; - - if (DISPLAY_VER(dev_priv) == 11) { - mask |= ICL_AUX_CHANNEL_F; - mask |= ICL_AUX_CHANNEL_E; - } - - return mask; -} - -static u32 gen8_de_pipe_fault_mask(struct drm_i915_private *dev_priv) -{ - if (DISPLAY_VER(dev_priv) >= 13 || HAS_D12_PLANE_MINIMIZATION(dev_priv)) - return RKL_DE_PIPE_IRQ_FAULT_ERRORS; - else if (DISPLAY_VER(dev_priv) >= 11) - return GEN11_DE_PIPE_IRQ_FAULT_ERRORS; - else if (DISPLAY_VER(dev_priv) >= 9) - return GEN9_DE_PIPE_IRQ_FAULT_ERRORS; - else - return GEN8_DE_PIPE_IRQ_FAULT_ERRORS; -} - -static void -gen8_de_misc_irq_handler(struct drm_i915_private *dev_priv, u32 iir) -{ - bool found = false; - - if (iir & GEN8_DE_MISC_GSE) { - intel_opregion_asle_intr(dev_priv); - found = true; - } - - if (iir & GEN8_DE_EDP_PSR) { - struct intel_encoder *encoder; - u32 psr_iir; - i915_reg_t iir_reg; - - for_each_intel_encoder_with_psr(&dev_priv->drm, encoder) { - struct intel_dp *intel_dp = enc_to_intel_dp(encoder); - - if (DISPLAY_VER(dev_priv) >= 12) - iir_reg = TRANS_PSR_IIR(intel_dp->psr.transcoder); - else - iir_reg = EDP_PSR_IIR; - - psr_iir = intel_uncore_rmw(&dev_priv->uncore, iir_reg, 0, 0); - - if (psr_iir) - found = true; - - intel_psr_irq_handler(intel_dp, psr_iir); - - /* prior GEN12 only have one EDP PSR */ - if (DISPLAY_VER(dev_priv) < 12) - break; - } - } - - if (!found) - drm_err(&dev_priv->drm, "Unexpected DE Misc interrupt\n"); -} - -static void gen11_dsi_te_interrupt_handler(struct drm_i915_private *dev_priv, - u32 te_trigger) -{ - enum pipe pipe = INVALID_PIPE; - enum transcoder dsi_trans; - enum port port; - u32 val, tmp; - - /* - * Incase of dual link, TE comes from DSI_1 - * this is to check if dual link is enabled - */ - val = intel_uncore_read(&dev_priv->uncore, TRANS_DDI_FUNC_CTL2(TRANSCODER_DSI_0)); - val &= PORT_SYNC_MODE_ENABLE; - - /* - * if dual link is enabled, then read DSI_0 - * transcoder registers - */ - port = ((te_trigger & DSI1_TE && val) || (te_trigger & DSI0_TE)) ? - PORT_A : PORT_B; - dsi_trans = (port == PORT_A) ? TRANSCODER_DSI_0 : TRANSCODER_DSI_1; - - /* Check if DSI configured in command mode */ - val = intel_uncore_read(&dev_priv->uncore, DSI_TRANS_FUNC_CONF(dsi_trans)); - val = val & OP_MODE_MASK; - - if (val != CMD_MODE_NO_GATE && val != CMD_MODE_TE_GATE) { - drm_err(&dev_priv->drm, "DSI trancoder not configured in command mode\n"); - return; - } - - /* Get PIPE for handling VBLANK event */ - val = intel_uncore_read(&dev_priv->uncore, TRANS_DDI_FUNC_CTL(dsi_trans)); - switch (val & TRANS_DDI_EDP_INPUT_MASK) { - case TRANS_DDI_EDP_INPUT_A_ON: - pipe = PIPE_A; - break; - case TRANS_DDI_EDP_INPUT_B_ONOFF: - pipe = PIPE_B; - break; - case TRANS_DDI_EDP_INPUT_C_ONOFF: - pipe = PIPE_C; - break; - default: - drm_err(&dev_priv->drm, "Invalid PIPE\n"); - return; - } - - intel_handle_vblank(dev_priv, pipe); - - /* clear TE in dsi IIR */ - port = (te_trigger & DSI1_TE) ? PORT_B : PORT_A; - tmp = intel_uncore_rmw(&dev_priv->uncore, DSI_INTR_IDENT_REG(port), 0, 0); -} - -static u32 gen8_de_pipe_flip_done_mask(struct drm_i915_private *i915) -{ - if (DISPLAY_VER(i915) >= 9) - return GEN9_PIPE_PLANE1_FLIP_DONE; - else - return GEN8_PIPE_PRIMARY_FLIP_DONE; -} - -u32 gen8_de_pipe_underrun_mask(struct drm_i915_private *dev_priv) -{ - u32 mask = GEN8_PIPE_FIFO_UNDERRUN; - - if (DISPLAY_VER(dev_priv) >= 13) - mask |= XELPD_PIPE_SOFT_UNDERRUN | - XELPD_PIPE_HARD_UNDERRUN; - - return mask; -} - -static irqreturn_t -gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl) -{ - irqreturn_t ret = IRQ_NONE; - u32 iir; - enum pipe pipe; - - drm_WARN_ON_ONCE(&dev_priv->drm, !HAS_DISPLAY(dev_priv)); - - if (master_ctl & GEN8_DE_MISC_IRQ) { - iir = intel_uncore_read(&dev_priv->uncore, GEN8_DE_MISC_IIR); - if (iir) { - intel_uncore_write(&dev_priv->uncore, GEN8_DE_MISC_IIR, iir); - ret = IRQ_HANDLED; - gen8_de_misc_irq_handler(dev_priv, iir); - } else { - drm_err_ratelimited(&dev_priv->drm, - "The master control interrupt lied (DE MISC)!\n"); - } - } - - if (DISPLAY_VER(dev_priv) >= 11 && (master_ctl & GEN11_DE_HPD_IRQ)) { - iir = intel_uncore_read(&dev_priv->uncore, GEN11_DE_HPD_IIR); - if (iir) { - intel_uncore_write(&dev_priv->uncore, GEN11_DE_HPD_IIR, iir); - ret = IRQ_HANDLED; - gen11_hpd_irq_handler(dev_priv, iir); - } else { - drm_err_ratelimited(&dev_priv->drm, - "The master control interrupt lied, (DE HPD)!\n"); - } - } - - if (master_ctl & GEN8_DE_PORT_IRQ) { - iir = intel_uncore_read(&dev_priv->uncore, GEN8_DE_PORT_IIR); - if (iir) { - bool found = false; - - intel_uncore_write(&dev_priv->uncore, GEN8_DE_PORT_IIR, iir); - ret = IRQ_HANDLED; - - if (iir & gen8_de_port_aux_mask(dev_priv)) { - dp_aux_irq_handler(dev_priv); - found = true; - } - - if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) { - u32 hotplug_trigger = iir & BXT_DE_PORT_HOTPLUG_MASK; - - if (hotplug_trigger) { - bxt_hpd_irq_handler(dev_priv, hotplug_trigger); - found = true; - } - } else if (IS_BROADWELL(dev_priv)) { - u32 hotplug_trigger = iir & BDW_DE_PORT_HOTPLUG_MASK; - - if (hotplug_trigger) { - ilk_hpd_irq_handler(dev_priv, hotplug_trigger); - found = true; - } - } - - if ((IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) && - (iir & BXT_DE_PORT_GMBUS)) { - gmbus_irq_handler(dev_priv); - found = true; - } - - if (DISPLAY_VER(dev_priv) >= 11) { - u32 te_trigger = iir & (DSI0_TE | DSI1_TE); - - if (te_trigger) { - gen11_dsi_te_interrupt_handler(dev_priv, te_trigger); - found = true; - } - } - - if (!found) - drm_err_ratelimited(&dev_priv->drm, - "Unexpected DE Port interrupt\n"); - } - else - drm_err_ratelimited(&dev_priv->drm, - "The master control interrupt lied (DE PORT)!\n"); - } - - for_each_pipe(dev_priv, pipe) { - u32 fault_errors; - - if (!(master_ctl & GEN8_DE_PIPE_IRQ(pipe))) - continue; - - iir = intel_uncore_read(&dev_priv->uncore, GEN8_DE_PIPE_IIR(pipe)); - if (!iir) { - drm_err_ratelimited(&dev_priv->drm, - "The master control interrupt lied (DE PIPE)!\n"); - continue; - } - - ret = IRQ_HANDLED; - intel_uncore_write(&dev_priv->uncore, GEN8_DE_PIPE_IIR(pipe), iir); - - if (iir & GEN8_PIPE_VBLANK) - intel_handle_vblank(dev_priv, pipe); - - if (iir & gen8_de_pipe_flip_done_mask(dev_priv)) - flip_done_handler(dev_priv, pipe); - - if (iir & GEN8_PIPE_CDCLK_CRC_DONE) - hsw_pipe_crc_irq_handler(dev_priv, pipe); - - if (iir & gen8_de_pipe_underrun_mask(dev_priv)) - intel_cpu_fifo_underrun_irq_handler(dev_priv, pipe); - - fault_errors = iir & gen8_de_pipe_fault_mask(dev_priv); - if (fault_errors) - drm_err_ratelimited(&dev_priv->drm, - "Fault errors on pipe %c: 0x%08x\n", - pipe_name(pipe), - fault_errors); - } - - if (HAS_PCH_SPLIT(dev_priv) && !HAS_PCH_NOP(dev_priv) && - master_ctl & GEN8_DE_PCH_IRQ) { - /* - * FIXME(BDW): Assume for now that the new interrupt handling - * scheme also closed the SDE interrupt handling race we've seen - * on older pch-split platforms. But this needs testing. - */ - iir = intel_uncore_read(&dev_priv->uncore, SDEIIR); - if (iir) { - intel_uncore_write(&dev_priv->uncore, SDEIIR, iir); - ret = IRQ_HANDLED; - - if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP) - icp_irq_handler(dev_priv, iir); - else if (INTEL_PCH_TYPE(dev_priv) >= PCH_SPT) - spt_irq_handler(dev_priv, iir); - else - cpt_irq_handler(dev_priv, iir); - } else { - /* - * Like on previous PCH there seems to be something - * fishy going on with forwarding PCH interrupts. - */ - drm_dbg(&dev_priv->drm, - "The master control interrupt lied (SDE)!\n"); - } - } - - return ret; -} - static inline u32 gen8_master_intr_disable(void __iomem * const regs) { raw_reg_write(regs, GEN8_MASTER_IRQ, 0); @@ -2232,32 +540,9 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg) return IRQ_HANDLED; } -static u32 -gen11_gu_misc_irq_ack(struct drm_i915_private *i915, const u32 master_ctl) +static inline u32 gen11_master_intr_disable(void __iomem * const regs) { - void __iomem * const regs = i915->uncore.regs; - u32 iir; - - if (!(master_ctl & GEN11_GU_MISC_IRQ)) - return 0; - - iir = raw_reg_read(regs, GEN11_GU_MISC_IIR); - if (likely(iir)) - raw_reg_write(regs, GEN11_GU_MISC_IIR, iir); - - return iir; -} - -static void -gen11_gu_misc_irq_handler(struct drm_i915_private *i915, const u32 iir) -{ - if (iir & GEN11_GU_MISC_GSE) - intel_opregion_asle_intr(i915); -} - -static inline u32 gen11_master_intr_disable(void __iomem * const regs) -{ - raw_reg_write(regs, GEN11_GFX_MSTR_IRQ, 0); + raw_reg_write(regs, GEN11_GFX_MSTR_IRQ, 0); /* * Now with master disabled, get a sample of level indications @@ -2273,25 +558,6 @@ static inline void gen11_master_intr_enable(void __iomem * const regs) raw_reg_write(regs, GEN11_GFX_MSTR_IRQ, GEN11_MASTER_IRQ); } -static void -gen11_display_irq_handler(struct drm_i915_private *i915) -{ - void __iomem * const regs = i915->uncore.regs; - const u32 disp_ctl = raw_reg_read(regs, GEN11_DISPLAY_INT_CTL); - - disable_rpm_wakeref_asserts(&i915->runtime_pm); - /* - * GEN11_DISPLAY_INT_CTL has same format as GEN8_MASTER_IRQ - * for the display related bits. - */ - raw_reg_write(regs, GEN11_DISPLAY_INT_CTL, 0x0); - gen8_de_irq_handler(i915, disp_ctl); - raw_reg_write(regs, GEN11_DISPLAY_INT_CTL, - GEN11_DISPLAY_IRQ_ENABLE); - - enable_rpm_wakeref_asserts(&i915->runtime_pm); -} - static irqreturn_t gen11_irq_handler(int irq, void *arg) { struct drm_i915_private *i915 = arg; @@ -2313,902 +579,195 @@ static irqreturn_t gen11_irq_handler(int irq, void *arg) gen11_gt_irq_handler(gt, master_ctl); /* IRQs are synced during runtime_suspend, we don't require a wakeref */ - if (master_ctl & GEN11_DISPLAY_IRQ) - gen11_display_irq_handler(i915); - - gu_misc_iir = gen11_gu_misc_irq_ack(i915, master_ctl); - - gen11_master_intr_enable(regs); - - gen11_gu_misc_irq_handler(i915, gu_misc_iir); - - pmu_irq_stats(i915, IRQ_HANDLED); - - return IRQ_HANDLED; -} - -static inline u32 dg1_master_intr_disable(void __iomem * const regs) -{ - u32 val; - - /* First disable interrupts */ - raw_reg_write(regs, DG1_MSTR_TILE_INTR, 0); - - /* Get the indication levels and ack the master unit */ - val = raw_reg_read(regs, DG1_MSTR_TILE_INTR); - if (unlikely(!val)) - return 0; - - raw_reg_write(regs, DG1_MSTR_TILE_INTR, val); - - return val; -} - -static inline void dg1_master_intr_enable(void __iomem * const regs) -{ - raw_reg_write(regs, DG1_MSTR_TILE_INTR, DG1_MSTR_IRQ); -} - -static irqreturn_t dg1_irq_handler(int irq, void *arg) -{ - struct drm_i915_private * const i915 = arg; - struct intel_gt *gt = to_gt(i915); - void __iomem * const regs = gt->uncore->regs; - u32 master_tile_ctl, master_ctl; - u32 gu_misc_iir; - - if (!intel_irqs_enabled(i915)) - return IRQ_NONE; - - master_tile_ctl = dg1_master_intr_disable(regs); - if (!master_tile_ctl) { - dg1_master_intr_enable(regs); - return IRQ_NONE; - } - - /* FIXME: we only support tile 0 for now. */ - if (master_tile_ctl & DG1_MSTR_TILE(0)) { - master_ctl = raw_reg_read(regs, GEN11_GFX_MSTR_IRQ); - raw_reg_write(regs, GEN11_GFX_MSTR_IRQ, master_ctl); - } else { - drm_err(&i915->drm, "Tile not supported: 0x%08x\n", - master_tile_ctl); - dg1_master_intr_enable(regs); - return IRQ_NONE; - } - - gen11_gt_irq_handler(gt, master_ctl); - - if (master_ctl & GEN11_DISPLAY_IRQ) - gen11_display_irq_handler(i915); - - gu_misc_iir = gen11_gu_misc_irq_ack(i915, master_ctl); - - dg1_master_intr_enable(regs); - - gen11_gu_misc_irq_handler(i915, gu_misc_iir); - - pmu_irq_stats(i915, IRQ_HANDLED); - - return IRQ_HANDLED; -} - -/* Called from drm generic code, passed 'crtc' which - * we use as a pipe index - */ -int i8xx_enable_vblank(struct drm_crtc *crtc) -{ - struct drm_i915_private *dev_priv = to_i915(crtc->dev); - enum pipe pipe = to_intel_crtc(crtc)->pipe; - unsigned long irqflags; - - spin_lock_irqsave(&dev_priv->irq_lock, irqflags); - i915_enable_pipestat(dev_priv, pipe, PIPE_VBLANK_INTERRUPT_STATUS); - spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); - - return 0; -} - -int i915gm_enable_vblank(struct drm_crtc *crtc) -{ - struct drm_i915_private *dev_priv = to_i915(crtc->dev); - - /* - * Vblank interrupts fail to wake the device up from C2+. - * Disabling render clock gating during C-states avoids - * the problem. There is a small power cost so we do this - * only when vblank interrupts are actually enabled. - */ - if (dev_priv->vblank_enabled++ == 0) - intel_uncore_write(&dev_priv->uncore, SCPD0, _MASKED_BIT_ENABLE(CSTATE_RENDER_CLOCK_GATE_DISABLE)); - - return i8xx_enable_vblank(crtc); -} - -int i965_enable_vblank(struct drm_crtc *crtc) -{ - struct drm_i915_private *dev_priv = to_i915(crtc->dev); - enum pipe pipe = to_intel_crtc(crtc)->pipe; - unsigned long irqflags; - - spin_lock_irqsave(&dev_priv->irq_lock, irqflags); - i915_enable_pipestat(dev_priv, pipe, - PIPE_START_VBLANK_INTERRUPT_STATUS); - spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); - - return 0; -} - -int ilk_enable_vblank(struct drm_crtc *crtc) -{ - struct drm_i915_private *dev_priv = to_i915(crtc->dev); - enum pipe pipe = to_intel_crtc(crtc)->pipe; - unsigned long irqflags; - u32 bit = DISPLAY_VER(dev_priv) >= 7 ? - DE_PIPE_VBLANK_IVB(pipe) : DE_PIPE_VBLANK(pipe); - - spin_lock_irqsave(&dev_priv->irq_lock, irqflags); - ilk_enable_display_irq(dev_priv, bit); - spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); - - /* Even though there is no DMC, frame counter can get stuck when - * PSR is active as no frames are generated. - */ - if (HAS_PSR(dev_priv)) - drm_crtc_vblank_restore(crtc); - - return 0; -} - -static bool gen11_dsi_configure_te(struct intel_crtc *intel_crtc, - bool enable) -{ - struct drm_i915_private *dev_priv = to_i915(intel_crtc->base.dev); - enum port port; - - if (!(intel_crtc->mode_flags & - (I915_MODE_FLAG_DSI_USE_TE1 | I915_MODE_FLAG_DSI_USE_TE0))) - return false; - - /* for dual link cases we consider TE from slave */ - if (intel_crtc->mode_flags & I915_MODE_FLAG_DSI_USE_TE1) - port = PORT_B; - else - port = PORT_A; - - intel_uncore_rmw(&dev_priv->uncore, DSI_INTR_MASK_REG(port), DSI_TE_EVENT, - enable ? 0 : DSI_TE_EVENT); - - intel_uncore_rmw(&dev_priv->uncore, DSI_INTR_IDENT_REG(port), 0, 0); - - return true; -} - -int bdw_enable_vblank(struct drm_crtc *_crtc) -{ - struct intel_crtc *crtc = to_intel_crtc(_crtc); - struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - enum pipe pipe = crtc->pipe; - unsigned long irqflags; - - if (gen11_dsi_configure_te(crtc, true)) - return 0; - - spin_lock_irqsave(&dev_priv->irq_lock, irqflags); - bdw_enable_pipe_irq(dev_priv, pipe, GEN8_PIPE_VBLANK); - spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); - - /* Even if there is no DMC, frame counter can get stuck when - * PSR is active as no frames are generated, so check only for PSR. - */ - if (HAS_PSR(dev_priv)) - drm_crtc_vblank_restore(&crtc->base); - - return 0; -} - -/* Called from drm generic code, passed 'crtc' which - * we use as a pipe index - */ -void i8xx_disable_vblank(struct drm_crtc *crtc) -{ - struct drm_i915_private *dev_priv = to_i915(crtc->dev); - enum pipe pipe = to_intel_crtc(crtc)->pipe; - unsigned long irqflags; - - spin_lock_irqsave(&dev_priv->irq_lock, irqflags); - i915_disable_pipestat(dev_priv, pipe, PIPE_VBLANK_INTERRUPT_STATUS); - spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); -} - -void i915gm_disable_vblank(struct drm_crtc *crtc) -{ - struct drm_i915_private *dev_priv = to_i915(crtc->dev); - - i8xx_disable_vblank(crtc); - - if (--dev_priv->vblank_enabled == 0) - intel_uncore_write(&dev_priv->uncore, SCPD0, _MASKED_BIT_DISABLE(CSTATE_RENDER_CLOCK_GATE_DISABLE)); -} - -void i965_disable_vblank(struct drm_crtc *crtc) -{ - struct drm_i915_private *dev_priv = to_i915(crtc->dev); - enum pipe pipe = to_intel_crtc(crtc)->pipe; - unsigned long irqflags; - - spin_lock_irqsave(&dev_priv->irq_lock, irqflags); - i915_disable_pipestat(dev_priv, pipe, - PIPE_START_VBLANK_INTERRUPT_STATUS); - spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); -} - -void ilk_disable_vblank(struct drm_crtc *crtc) -{ - struct drm_i915_private *dev_priv = to_i915(crtc->dev); - enum pipe pipe = to_intel_crtc(crtc)->pipe; - unsigned long irqflags; - u32 bit = DISPLAY_VER(dev_priv) >= 7 ? - DE_PIPE_VBLANK_IVB(pipe) : DE_PIPE_VBLANK(pipe); - - spin_lock_irqsave(&dev_priv->irq_lock, irqflags); - ilk_disable_display_irq(dev_priv, bit); - spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); -} - -void bdw_disable_vblank(struct drm_crtc *_crtc) -{ - struct intel_crtc *crtc = to_intel_crtc(_crtc); - struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - enum pipe pipe = crtc->pipe; - unsigned long irqflags; - - if (gen11_dsi_configure_te(crtc, false)) - return; - - spin_lock_irqsave(&dev_priv->irq_lock, irqflags); - bdw_disable_pipe_irq(dev_priv, pipe, GEN8_PIPE_VBLANK); - spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); -} - -static void ibx_irq_reset(struct drm_i915_private *dev_priv) -{ - struct intel_uncore *uncore = &dev_priv->uncore; - - if (HAS_PCH_NOP(dev_priv)) - return; - - GEN3_IRQ_RESET(uncore, SDE); - - if (HAS_PCH_CPT(dev_priv) || HAS_PCH_LPT(dev_priv)) - intel_uncore_write(&dev_priv->uncore, SERR_INT, 0xffffffff); -} - -static void vlv_display_irq_reset(struct drm_i915_private *dev_priv) -{ - struct intel_uncore *uncore = &dev_priv->uncore; - - if (IS_CHERRYVIEW(dev_priv)) - intel_uncore_write(uncore, DPINVGTT, DPINVGTT_STATUS_MASK_CHV); - else - intel_uncore_write(uncore, DPINVGTT, DPINVGTT_STATUS_MASK_VLV); - - i915_hotplug_interrupt_update_locked(dev_priv, 0xffffffff, 0); - intel_uncore_rmw(uncore, PORT_HOTPLUG_STAT, 0, 0); - - i9xx_pipestat_irq_reset(dev_priv); - - GEN3_IRQ_RESET(uncore, VLV_); - dev_priv->irq_mask = ~0u; -} - -static void vlv_display_irq_postinstall(struct drm_i915_private *dev_priv) -{ - struct intel_uncore *uncore = &dev_priv->uncore; - - u32 pipestat_mask; - u32 enable_mask; - enum pipe pipe; - - pipestat_mask = PIPE_CRC_DONE_INTERRUPT_STATUS; - - i915_enable_pipestat(dev_priv, PIPE_A, PIPE_GMBUS_INTERRUPT_STATUS); - for_each_pipe(dev_priv, pipe) - i915_enable_pipestat(dev_priv, pipe, pipestat_mask); - - enable_mask = I915_DISPLAY_PORT_INTERRUPT | - I915_DISPLAY_PIPE_A_EVENT_INTERRUPT | - I915_DISPLAY_PIPE_B_EVENT_INTERRUPT | - I915_LPE_PIPE_A_INTERRUPT | - I915_LPE_PIPE_B_INTERRUPT; - - if (IS_CHERRYVIEW(dev_priv)) - enable_mask |= I915_DISPLAY_PIPE_C_EVENT_INTERRUPT | - I915_LPE_PIPE_C_INTERRUPT; - - drm_WARN_ON(&dev_priv->drm, dev_priv->irq_mask != ~0u); - - dev_priv->irq_mask = ~enable_mask; - - GEN3_IRQ_INIT(uncore, VLV_, dev_priv->irq_mask, enable_mask); -} - -/* drm_dma.h hooks -*/ -static void ilk_irq_reset(struct drm_i915_private *dev_priv) -{ - struct intel_uncore *uncore = &dev_priv->uncore; - - GEN3_IRQ_RESET(uncore, DE); - dev_priv->irq_mask = ~0u; - - if (GRAPHICS_VER(dev_priv) == 7) - intel_uncore_write(uncore, GEN7_ERR_INT, 0xffffffff); - - if (IS_HASWELL(dev_priv)) { - intel_uncore_write(uncore, EDP_PSR_IMR, 0xffffffff); - intel_uncore_write(uncore, EDP_PSR_IIR, 0xffffffff); - } - - gen5_gt_irq_reset(to_gt(dev_priv)); - - ibx_irq_reset(dev_priv); -} - -static void valleyview_irq_reset(struct drm_i915_private *dev_priv) -{ - intel_uncore_write(&dev_priv->uncore, VLV_MASTER_IER, 0); - intel_uncore_posting_read(&dev_priv->uncore, VLV_MASTER_IER); - - gen5_gt_irq_reset(to_gt(dev_priv)); - - spin_lock_irq(&dev_priv->irq_lock); - if (dev_priv->display_irqs_enabled) - vlv_display_irq_reset(dev_priv); - spin_unlock_irq(&dev_priv->irq_lock); -} - -static void gen8_display_irq_reset(struct drm_i915_private *dev_priv) -{ - struct intel_uncore *uncore = &dev_priv->uncore; - enum pipe pipe; - - if (!HAS_DISPLAY(dev_priv)) - return; - - intel_uncore_write(uncore, EDP_PSR_IMR, 0xffffffff); - intel_uncore_write(uncore, EDP_PSR_IIR, 0xffffffff); - - for_each_pipe(dev_priv, pipe) - if (intel_display_power_is_enabled(dev_priv, - POWER_DOMAIN_PIPE(pipe))) - GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe); - - GEN3_IRQ_RESET(uncore, GEN8_DE_PORT_); - GEN3_IRQ_RESET(uncore, GEN8_DE_MISC_); -} - -static void gen8_irq_reset(struct drm_i915_private *dev_priv) -{ - struct intel_uncore *uncore = &dev_priv->uncore; - - gen8_master_intr_disable(uncore->regs); - - gen8_gt_irq_reset(to_gt(dev_priv)); - gen8_display_irq_reset(dev_priv); - GEN3_IRQ_RESET(uncore, GEN8_PCU_); - - if (HAS_PCH_SPLIT(dev_priv)) - ibx_irq_reset(dev_priv); - -} - -static void gen11_display_irq_reset(struct drm_i915_private *dev_priv) -{ - struct intel_uncore *uncore = &dev_priv->uncore; - enum pipe pipe; - u32 trans_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | - BIT(TRANSCODER_C) | BIT(TRANSCODER_D); - - if (!HAS_DISPLAY(dev_priv)) - return; - - intel_uncore_write(uncore, GEN11_DISPLAY_INT_CTL, 0); - - if (DISPLAY_VER(dev_priv) >= 12) { - enum transcoder trans; - - for_each_cpu_transcoder_masked(dev_priv, trans, trans_mask) { - enum intel_display_power_domain domain; - - domain = POWER_DOMAIN_TRANSCODER(trans); - if (!intel_display_power_is_enabled(dev_priv, domain)) - continue; - - intel_uncore_write(uncore, TRANS_PSR_IMR(trans), 0xffffffff); - intel_uncore_write(uncore, TRANS_PSR_IIR(trans), 0xffffffff); - } - } else { - intel_uncore_write(uncore, EDP_PSR_IMR, 0xffffffff); - intel_uncore_write(uncore, EDP_PSR_IIR, 0xffffffff); - } - - for_each_pipe(dev_priv, pipe) - if (intel_display_power_is_enabled(dev_priv, - POWER_DOMAIN_PIPE(pipe))) - GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe); - - GEN3_IRQ_RESET(uncore, GEN8_DE_PORT_); - GEN3_IRQ_RESET(uncore, GEN8_DE_MISC_); - GEN3_IRQ_RESET(uncore, GEN11_DE_HPD_); - - if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP) - GEN3_IRQ_RESET(uncore, SDE); -} - -static void gen11_irq_reset(struct drm_i915_private *dev_priv) -{ - struct intel_gt *gt = to_gt(dev_priv); - struct intel_uncore *uncore = gt->uncore; - - gen11_master_intr_disable(dev_priv->uncore.regs); - - gen11_gt_irq_reset(gt); - gen11_display_irq_reset(dev_priv); - - GEN3_IRQ_RESET(uncore, GEN11_GU_MISC_); - GEN3_IRQ_RESET(uncore, GEN8_PCU_); -} - -static void dg1_irq_reset(struct drm_i915_private *dev_priv) -{ - struct intel_uncore *uncore = &dev_priv->uncore; - struct intel_gt *gt; - unsigned int i; - - dg1_master_intr_disable(dev_priv->uncore.regs); - - for_each_gt(gt, dev_priv, i) - gen11_gt_irq_reset(gt); - - gen11_display_irq_reset(dev_priv); - - GEN3_IRQ_RESET(uncore, GEN11_GU_MISC_); - GEN3_IRQ_RESET(uncore, GEN8_PCU_); -} - -void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv, - u8 pipe_mask) -{ - struct intel_uncore *uncore = &dev_priv->uncore; - u32 extra_ier = GEN8_PIPE_VBLANK | - gen8_de_pipe_underrun_mask(dev_priv) | - gen8_de_pipe_flip_done_mask(dev_priv); - enum pipe pipe; - - spin_lock_irq(&dev_priv->irq_lock); - - if (!intel_irqs_enabled(dev_priv)) { - spin_unlock_irq(&dev_priv->irq_lock); - return; - } - - for_each_pipe_masked(dev_priv, pipe, pipe_mask) - GEN8_IRQ_INIT_NDX(uncore, DE_PIPE, pipe, - dev_priv->de_irq_mask[pipe], - ~dev_priv->de_irq_mask[pipe] | extra_ier); - - spin_unlock_irq(&dev_priv->irq_lock); -} - -void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv, - u8 pipe_mask) -{ - struct intel_uncore *uncore = &dev_priv->uncore; - enum pipe pipe; - - spin_lock_irq(&dev_priv->irq_lock); - - if (!intel_irqs_enabled(dev_priv)) { - spin_unlock_irq(&dev_priv->irq_lock); - return; - } - - for_each_pipe_masked(dev_priv, pipe, pipe_mask) - GEN8_IRQ_RESET_NDX(uncore, DE_PIPE, pipe); - - spin_unlock_irq(&dev_priv->irq_lock); - - /* make sure we're done processing display irqs */ - intel_synchronize_irq(dev_priv); -} - -static void cherryview_irq_reset(struct drm_i915_private *dev_priv) -{ - struct intel_uncore *uncore = &dev_priv->uncore; - - intel_uncore_write(uncore, GEN8_MASTER_IRQ, 0); - intel_uncore_posting_read(&dev_priv->uncore, GEN8_MASTER_IRQ); - - gen8_gt_irq_reset(to_gt(dev_priv)); - - GEN3_IRQ_RESET(uncore, GEN8_PCU_); - - spin_lock_irq(&dev_priv->irq_lock); - if (dev_priv->display_irqs_enabled) - vlv_display_irq_reset(dev_priv); - spin_unlock_irq(&dev_priv->irq_lock); -} - -static u32 ibx_hotplug_enables(struct intel_encoder *encoder) -{ - struct drm_i915_private *i915 = to_i915(encoder->base.dev); - - switch (encoder->hpd_pin) { - case HPD_PORT_A: - /* - * When CPU and PCH are on the same package, port A - * HPD must be enabled in both north and south. - */ - return HAS_PCH_LPT_LP(i915) ? - PORTA_HOTPLUG_ENABLE : 0; - case HPD_PORT_B: - return PORTB_HOTPLUG_ENABLE | - PORTB_PULSE_DURATION_2ms; - case HPD_PORT_C: - return PORTC_HOTPLUG_ENABLE | - PORTC_PULSE_DURATION_2ms; - case HPD_PORT_D: - return PORTD_HOTPLUG_ENABLE | - PORTD_PULSE_DURATION_2ms; - default: - return 0; - } -} - -static void ibx_hpd_detection_setup(struct drm_i915_private *dev_priv) -{ - /* - * Enable digital hotplug on the PCH, and configure the DP short pulse - * duration to 2ms (which is the minimum in the Display Port spec). - * The pulse duration bits are reserved on LPT+. - */ - intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG, - PORTA_HOTPLUG_ENABLE | - PORTB_HOTPLUG_ENABLE | - PORTC_HOTPLUG_ENABLE | - PORTD_HOTPLUG_ENABLE | - PORTB_PULSE_DURATION_MASK | - PORTC_PULSE_DURATION_MASK | - PORTD_PULSE_DURATION_MASK, - intel_hpd_hotplug_enables(dev_priv, ibx_hotplug_enables)); -} - -static void ibx_hpd_irq_setup(struct drm_i915_private *dev_priv) -{ - u32 hotplug_irqs, enabled_irqs; - - enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd); - hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd); - - ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs); + if (master_ctl & GEN11_DISPLAY_IRQ) + gen11_display_irq_handler(i915); - ibx_hpd_detection_setup(dev_priv); -} + gu_misc_iir = gen11_gu_misc_irq_ack(i915, master_ctl); -static u32 icp_ddi_hotplug_enables(struct intel_encoder *encoder) -{ - switch (encoder->hpd_pin) { - case HPD_PORT_A: - case HPD_PORT_B: - case HPD_PORT_C: - case HPD_PORT_D: - return SHOTPLUG_CTL_DDI_HPD_ENABLE(encoder->hpd_pin); - default: - return 0; - } -} + gen11_master_intr_enable(regs); -static u32 icp_tc_hotplug_enables(struct intel_encoder *encoder) -{ - switch (encoder->hpd_pin) { - case HPD_PORT_TC1: - case HPD_PORT_TC2: - case HPD_PORT_TC3: - case HPD_PORT_TC4: - case HPD_PORT_TC5: - case HPD_PORT_TC6: - return ICP_TC_HPD_ENABLE(encoder->hpd_pin); - default: - return 0; - } -} + gen11_gu_misc_irq_handler(i915, gu_misc_iir); -static void icp_ddi_hpd_detection_setup(struct drm_i915_private *dev_priv) -{ - intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_DDI, - SHOTPLUG_CTL_DDI_HPD_ENABLE(HPD_PORT_A) | - SHOTPLUG_CTL_DDI_HPD_ENABLE(HPD_PORT_B) | - SHOTPLUG_CTL_DDI_HPD_ENABLE(HPD_PORT_C) | - SHOTPLUG_CTL_DDI_HPD_ENABLE(HPD_PORT_D), - intel_hpd_hotplug_enables(dev_priv, icp_ddi_hotplug_enables)); -} + pmu_irq_stats(i915, IRQ_HANDLED); -static void icp_tc_hpd_detection_setup(struct drm_i915_private *dev_priv) -{ - intel_uncore_rmw(&dev_priv->uncore, SHOTPLUG_CTL_TC, - ICP_TC_HPD_ENABLE(HPD_PORT_TC1) | - ICP_TC_HPD_ENABLE(HPD_PORT_TC2) | - ICP_TC_HPD_ENABLE(HPD_PORT_TC3) | - ICP_TC_HPD_ENABLE(HPD_PORT_TC4) | - ICP_TC_HPD_ENABLE(HPD_PORT_TC5) | - ICP_TC_HPD_ENABLE(HPD_PORT_TC6), - intel_hpd_hotplug_enables(dev_priv, icp_tc_hotplug_enables)); + return IRQ_HANDLED; } -static void icp_hpd_irq_setup(struct drm_i915_private *dev_priv) +static inline u32 dg1_master_intr_disable(void __iomem * const regs) { - u32 hotplug_irqs, enabled_irqs; + u32 val; - enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd); - hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd); + /* First disable interrupts */ + raw_reg_write(regs, DG1_MSTR_TILE_INTR, 0); - if (INTEL_PCH_TYPE(dev_priv) <= PCH_TGP) - intel_uncore_write(&dev_priv->uncore, SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ); + /* Get the indication levels and ack the master unit */ + val = raw_reg_read(regs, DG1_MSTR_TILE_INTR); + if (unlikely(!val)) + return 0; - ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs); + raw_reg_write(regs, DG1_MSTR_TILE_INTR, val); - icp_ddi_hpd_detection_setup(dev_priv); - icp_tc_hpd_detection_setup(dev_priv); + return val; } -static u32 gen11_hotplug_enables(struct intel_encoder *encoder) +static inline void dg1_master_intr_enable(void __iomem * const regs) { - switch (encoder->hpd_pin) { - case HPD_PORT_TC1: - case HPD_PORT_TC2: - case HPD_PORT_TC3: - case HPD_PORT_TC4: - case HPD_PORT_TC5: - case HPD_PORT_TC6: - return GEN11_HOTPLUG_CTL_ENABLE(encoder->hpd_pin); - default: - return 0; - } + raw_reg_write(regs, DG1_MSTR_TILE_INTR, DG1_MSTR_IRQ); } -static void dg1_hpd_invert(struct drm_i915_private *i915) +static irqreturn_t dg1_irq_handler(int irq, void *arg) { - u32 val = (INVERT_DDIA_HPD | - INVERT_DDIB_HPD | - INVERT_DDIC_HPD | - INVERT_DDID_HPD); - intel_uncore_rmw(&i915->uncore, SOUTH_CHICKEN1, 0, val); -} + struct drm_i915_private * const i915 = arg; + struct intel_gt *gt = to_gt(i915); + void __iomem * const regs = gt->uncore->regs; + u32 master_tile_ctl, master_ctl; + u32 gu_misc_iir; -static void dg1_hpd_irq_setup(struct drm_i915_private *dev_priv) -{ - dg1_hpd_invert(dev_priv); - icp_hpd_irq_setup(dev_priv); -} + if (!intel_irqs_enabled(i915)) + return IRQ_NONE; -static void gen11_tc_hpd_detection_setup(struct drm_i915_private *dev_priv) -{ - intel_uncore_rmw(&dev_priv->uncore, GEN11_TC_HOTPLUG_CTL, - GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC1) | - GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC2) | - GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC3) | - GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC4) | - GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC5) | - GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC6), - intel_hpd_hotplug_enables(dev_priv, gen11_hotplug_enables)); -} + master_tile_ctl = dg1_master_intr_disable(regs); + if (!master_tile_ctl) { + dg1_master_intr_enable(regs); + return IRQ_NONE; + } -static void gen11_tbt_hpd_detection_setup(struct drm_i915_private *dev_priv) -{ - intel_uncore_rmw(&dev_priv->uncore, GEN11_TBT_HOTPLUG_CTL, - GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC1) | - GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC2) | - GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC3) | - GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC4) | - GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC5) | - GEN11_HOTPLUG_CTL_ENABLE(HPD_PORT_TC6), - intel_hpd_hotplug_enables(dev_priv, gen11_hotplug_enables)); -} + /* FIXME: we only support tile 0 for now. */ + if (master_tile_ctl & DG1_MSTR_TILE(0)) { + master_ctl = raw_reg_read(regs, GEN11_GFX_MSTR_IRQ); + raw_reg_write(regs, GEN11_GFX_MSTR_IRQ, master_ctl); + } else { + drm_err(&i915->drm, "Tile not supported: 0x%08x\n", + master_tile_ctl); + dg1_master_intr_enable(regs); + return IRQ_NONE; + } -static void gen11_hpd_irq_setup(struct drm_i915_private *dev_priv) -{ - u32 hotplug_irqs, enabled_irqs; + gen11_gt_irq_handler(gt, master_ctl); - enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.hpd); - hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.hpd); + if (master_ctl & GEN11_DISPLAY_IRQ) + gen11_display_irq_handler(i915); - intel_uncore_rmw(&dev_priv->uncore, GEN11_DE_HPD_IMR, hotplug_irqs, - ~enabled_irqs & hotplug_irqs); - intel_uncore_posting_read(&dev_priv->uncore, GEN11_DE_HPD_IMR); + gu_misc_iir = gen11_gu_misc_irq_ack(i915, master_ctl); - gen11_tc_hpd_detection_setup(dev_priv); - gen11_tbt_hpd_detection_setup(dev_priv); + dg1_master_intr_enable(regs); - if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP) - icp_hpd_irq_setup(dev_priv); -} + gen11_gu_misc_irq_handler(i915, gu_misc_iir); -static u32 spt_hotplug_enables(struct intel_encoder *encoder) -{ - switch (encoder->hpd_pin) { - case HPD_PORT_A: - return PORTA_HOTPLUG_ENABLE; - case HPD_PORT_B: - return PORTB_HOTPLUG_ENABLE; - case HPD_PORT_C: - return PORTC_HOTPLUG_ENABLE; - case HPD_PORT_D: - return PORTD_HOTPLUG_ENABLE; - default: - return 0; - } -} + pmu_irq_stats(i915, IRQ_HANDLED); -static u32 spt_hotplug2_enables(struct intel_encoder *encoder) -{ - switch (encoder->hpd_pin) { - case HPD_PORT_E: - return PORTE_HOTPLUG_ENABLE; - default: - return 0; - } + return IRQ_HANDLED; } -static void spt_hpd_detection_setup(struct drm_i915_private *dev_priv) +static void ibx_irq_reset(struct drm_i915_private *dev_priv) { - /* Display WA #1179 WaHardHangonHotPlug: cnp */ - if (HAS_PCH_CNP(dev_priv)) { - intel_uncore_rmw(&dev_priv->uncore, SOUTH_CHICKEN1, CHASSIS_CLK_REQ_DURATION_MASK, - CHASSIS_CLK_REQ_DURATION(0xf)); - } + struct intel_uncore *uncore = &dev_priv->uncore; - /* Enable digital hotplug on the PCH */ - intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG, - PORTA_HOTPLUG_ENABLE | - PORTB_HOTPLUG_ENABLE | - PORTC_HOTPLUG_ENABLE | - PORTD_HOTPLUG_ENABLE, - intel_hpd_hotplug_enables(dev_priv, spt_hotplug_enables)); + if (HAS_PCH_NOP(dev_priv)) + return; + + GEN3_IRQ_RESET(uncore, SDE); - intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG2, PORTE_HOTPLUG_ENABLE, - intel_hpd_hotplug_enables(dev_priv, spt_hotplug2_enables)); + if (HAS_PCH_CPT(dev_priv) || HAS_PCH_LPT(dev_priv)) + intel_uncore_write(&dev_priv->uncore, SERR_INT, 0xffffffff); } -static void spt_hpd_irq_setup(struct drm_i915_private *dev_priv) +/* drm_dma.h hooks +*/ +static void ilk_irq_reset(struct drm_i915_private *dev_priv) { - u32 hotplug_irqs, enabled_irqs; + struct intel_uncore *uncore = &dev_priv->uncore; + + GEN3_IRQ_RESET(uncore, DE); + dev_priv->irq_mask = ~0u; - if (INTEL_PCH_TYPE(dev_priv) >= PCH_CNP) - intel_uncore_write(&dev_priv->uncore, SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ); + if (GRAPHICS_VER(dev_priv) == 7) + intel_uncore_write(uncore, GEN7_ERR_INT, 0xffffffff); - enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd); - hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.pch_hpd); + if (IS_HASWELL(dev_priv)) { + intel_uncore_write(uncore, EDP_PSR_IMR, 0xffffffff); + intel_uncore_write(uncore, EDP_PSR_IIR, 0xffffffff); + } - ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs); + gen5_gt_irq_reset(to_gt(dev_priv)); - spt_hpd_detection_setup(dev_priv); + ibx_irq_reset(dev_priv); } -static u32 ilk_hotplug_enables(struct intel_encoder *encoder) +static void valleyview_irq_reset(struct drm_i915_private *dev_priv) { - switch (encoder->hpd_pin) { - case HPD_PORT_A: - return DIGITAL_PORTA_HOTPLUG_ENABLE | - DIGITAL_PORTA_PULSE_DURATION_2ms; - default: - return 0; - } -} + intel_uncore_write(&dev_priv->uncore, VLV_MASTER_IER, 0); + intel_uncore_posting_read(&dev_priv->uncore, VLV_MASTER_IER); -static void ilk_hpd_detection_setup(struct drm_i915_private *dev_priv) -{ - /* - * Enable digital hotplug on the CPU, and configure the DP short pulse - * duration to 2ms (which is the minimum in the Display Port spec) - * The pulse duration bits are reserved on HSW+. - */ - intel_uncore_rmw(&dev_priv->uncore, DIGITAL_PORT_HOTPLUG_CNTRL, - DIGITAL_PORTA_HOTPLUG_ENABLE | DIGITAL_PORTA_PULSE_DURATION_MASK, - intel_hpd_hotplug_enables(dev_priv, ilk_hotplug_enables)); + gen5_gt_irq_reset(to_gt(dev_priv)); + + spin_lock_irq(&dev_priv->irq_lock); + if (dev_priv->display_irqs_enabled) + vlv_display_irq_reset(dev_priv); + spin_unlock_irq(&dev_priv->irq_lock); } -static void ilk_hpd_irq_setup(struct drm_i915_private *dev_priv) +static void gen8_irq_reset(struct drm_i915_private *dev_priv) { - u32 hotplug_irqs, enabled_irqs; + struct intel_uncore *uncore = &dev_priv->uncore; - enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.hpd); - hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.hpd); + gen8_master_intr_disable(uncore->regs); - if (DISPLAY_VER(dev_priv) >= 8) - bdw_update_port_irq(dev_priv, hotplug_irqs, enabled_irqs); - else - ilk_update_display_irq(dev_priv, hotplug_irqs, enabled_irqs); + gen8_gt_irq_reset(to_gt(dev_priv)); + gen8_display_irq_reset(dev_priv); + GEN3_IRQ_RESET(uncore, GEN8_PCU_); - ilk_hpd_detection_setup(dev_priv); + if (HAS_PCH_SPLIT(dev_priv)) + ibx_irq_reset(dev_priv); - ibx_hpd_irq_setup(dev_priv); } -static u32 bxt_hotplug_enables(struct intel_encoder *encoder) +static void gen11_irq_reset(struct drm_i915_private *dev_priv) { - u32 hotplug; - - switch (encoder->hpd_pin) { - case HPD_PORT_A: - hotplug = PORTA_HOTPLUG_ENABLE; - if (intel_bios_encoder_hpd_invert(encoder->devdata)) - hotplug |= BXT_DDIA_HPD_INVERT; - return hotplug; - case HPD_PORT_B: - hotplug = PORTB_HOTPLUG_ENABLE; - if (intel_bios_encoder_hpd_invert(encoder->devdata)) - hotplug |= BXT_DDIB_HPD_INVERT; - return hotplug; - case HPD_PORT_C: - hotplug = PORTC_HOTPLUG_ENABLE; - if (intel_bios_encoder_hpd_invert(encoder->devdata)) - hotplug |= BXT_DDIC_HPD_INVERT; - return hotplug; - default: - return 0; - } -} + struct intel_gt *gt = to_gt(dev_priv); + struct intel_uncore *uncore = gt->uncore; -static void bxt_hpd_detection_setup(struct drm_i915_private *dev_priv) -{ - intel_uncore_rmw(&dev_priv->uncore, PCH_PORT_HOTPLUG, - PORTA_HOTPLUG_ENABLE | - PORTB_HOTPLUG_ENABLE | - PORTC_HOTPLUG_ENABLE | - BXT_DDI_HPD_INVERT_MASK, - intel_hpd_hotplug_enables(dev_priv, bxt_hotplug_enables)); + gen11_master_intr_disable(dev_priv->uncore.regs); + + gen11_gt_irq_reset(gt); + gen11_display_irq_reset(dev_priv); + + GEN3_IRQ_RESET(uncore, GEN11_GU_MISC_); + GEN3_IRQ_RESET(uncore, GEN8_PCU_); } -static void bxt_hpd_irq_setup(struct drm_i915_private *dev_priv) +static void dg1_irq_reset(struct drm_i915_private *dev_priv) { - u32 hotplug_irqs, enabled_irqs; + struct intel_uncore *uncore = &dev_priv->uncore; + struct intel_gt *gt; + unsigned int i; + + dg1_master_intr_disable(dev_priv->uncore.regs); - enabled_irqs = intel_hpd_enabled_irqs(dev_priv, dev_priv->display.hotplug.hpd); - hotplug_irqs = intel_hpd_hotplug_irqs(dev_priv, dev_priv->display.hotplug.hpd); + for_each_gt(gt, dev_priv, i) + gen11_gt_irq_reset(gt); - bdw_update_port_irq(dev_priv, hotplug_irqs, enabled_irqs); + gen11_display_irq_reset(dev_priv); - bxt_hpd_detection_setup(dev_priv); + GEN3_IRQ_RESET(uncore, GEN11_GU_MISC_); + GEN3_IRQ_RESET(uncore, GEN8_PCU_); } -/* - * SDEIER is also touched by the interrupt handler to work around missed PCH - * interrupts. Hence we can't update it after the interrupt handler is enabled - - * instead we unconditionally enable all PCH interrupt sources here, but then - * only unmask them as needed with SDEIMR. - * - * Note that we currently do this after installing the interrupt handler, - * but before we enable the master interrupt. That should be sufficient - * to avoid races with the irq handler, assuming we have MSI. Shared legacy - * interrupts could still race. - */ -static void ibx_irq_postinstall(struct drm_i915_private *dev_priv) +static void cherryview_irq_reset(struct drm_i915_private *dev_priv) { struct intel_uncore *uncore = &dev_priv->uncore; - u32 mask; - if (HAS_PCH_NOP(dev_priv)) - return; + intel_uncore_write(uncore, GEN8_MASTER_IRQ, 0); + intel_uncore_posting_read(&dev_priv->uncore, GEN8_MASTER_IRQ); - if (HAS_PCH_IBX(dev_priv)) - mask = SDE_GMBUS | SDE_AUX_MASK | SDE_POISON; - else if (HAS_PCH_CPT(dev_priv) || HAS_PCH_LPT(dev_priv)) - mask = SDE_GMBUS_CPT | SDE_AUX_MASK_CPT; - else - mask = SDE_GMBUS_CPT; + gen8_gt_irq_reset(to_gt(dev_priv)); - GEN3_IRQ_INIT(uncore, SDE, ~mask, 0xffffffff); + GEN3_IRQ_RESET(uncore, GEN8_PCU_); + + spin_lock_irq(&dev_priv->irq_lock); + if (dev_priv->display_irqs_enabled) + vlv_display_irq_reset(dev_priv); + spin_unlock_irq(&dev_priv->irq_lock); } static void ilk_irq_postinstall(struct drm_i915_private *dev_priv) @@ -3254,35 +813,6 @@ static void ilk_irq_postinstall(struct drm_i915_private *dev_priv) display_mask | extra_mask); } -void valleyview_enable_display_irqs(struct drm_i915_private *dev_priv) -{ - lockdep_assert_held(&dev_priv->irq_lock); - - if (dev_priv->display_irqs_enabled) - return; - - dev_priv->display_irqs_enabled = true; - - if (intel_irqs_enabled(dev_priv)) { - vlv_display_irq_reset(dev_priv); - vlv_display_irq_postinstall(dev_priv); - } -} - -void valleyview_disable_display_irqs(struct drm_i915_private *dev_priv) -{ - lockdep_assert_held(&dev_priv->irq_lock); - - if (!dev_priv->display_irqs_enabled) - return; - - dev_priv->display_irqs_enabled = false; - - if (intel_irqs_enabled(dev_priv)) - vlv_display_irq_reset(dev_priv); -} - - static void valleyview_irq_postinstall(struct drm_i915_private *dev_priv) { gen5_gt_irq_postinstall(to_gt(dev_priv)); @@ -3296,94 +826,6 @@ static void valleyview_irq_postinstall(struct drm_i915_private *dev_priv) intel_uncore_posting_read(&dev_priv->uncore, VLV_MASTER_IER); } -static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv) -{ - struct intel_uncore *uncore = &dev_priv->uncore; - - u32 de_pipe_masked = gen8_de_pipe_fault_mask(dev_priv) | - GEN8_PIPE_CDCLK_CRC_DONE; - u32 de_pipe_enables; - u32 de_port_masked = gen8_de_port_aux_mask(dev_priv); - u32 de_port_enables; - u32 de_misc_masked = GEN8_DE_EDP_PSR; - u32 trans_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | - BIT(TRANSCODER_C) | BIT(TRANSCODER_D); - enum pipe pipe; - - if (!HAS_DISPLAY(dev_priv)) - return; - - if (DISPLAY_VER(dev_priv) <= 10) - de_misc_masked |= GEN8_DE_MISC_GSE; - - if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) - de_port_masked |= BXT_DE_PORT_GMBUS; - - if (DISPLAY_VER(dev_priv) >= 11) { - enum port port; - - if (intel_bios_is_dsi_present(dev_priv, &port)) - de_port_masked |= DSI0_TE | DSI1_TE; - } - - de_pipe_enables = de_pipe_masked | - GEN8_PIPE_VBLANK | - gen8_de_pipe_underrun_mask(dev_priv) | - gen8_de_pipe_flip_done_mask(dev_priv); - - de_port_enables = de_port_masked; - if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) - de_port_enables |= BXT_DE_PORT_HOTPLUG_MASK; - else if (IS_BROADWELL(dev_priv)) - de_port_enables |= BDW_DE_PORT_HOTPLUG_MASK; - - if (DISPLAY_VER(dev_priv) >= 12) { - enum transcoder trans; - - for_each_cpu_transcoder_masked(dev_priv, trans, trans_mask) { - enum intel_display_power_domain domain; - - domain = POWER_DOMAIN_TRANSCODER(trans); - if (!intel_display_power_is_enabled(dev_priv, domain)) - continue; - - gen3_assert_iir_is_zero(uncore, TRANS_PSR_IIR(trans)); - } - } else { - gen3_assert_iir_is_zero(uncore, EDP_PSR_IIR); - } - - for_each_pipe(dev_priv, pipe) { - dev_priv->de_irq_mask[pipe] = ~de_pipe_masked; - - if (intel_display_power_is_enabled(dev_priv, - POWER_DOMAIN_PIPE(pipe))) - GEN8_IRQ_INIT_NDX(uncore, DE_PIPE, pipe, - dev_priv->de_irq_mask[pipe], - de_pipe_enables); - } - - GEN3_IRQ_INIT(uncore, GEN8_DE_PORT_, ~de_port_masked, de_port_enables); - GEN3_IRQ_INIT(uncore, GEN8_DE_MISC_, ~de_misc_masked, de_misc_masked); - - if (DISPLAY_VER(dev_priv) >= 11) { - u32 de_hpd_masked = 0; - u32 de_hpd_enables = GEN11_DE_TC_HOTPLUG_MASK | - GEN11_DE_TBT_HOTPLUG_MASK; - - GEN3_IRQ_INIT(uncore, GEN11_DE_HPD_, ~de_hpd_masked, - de_hpd_enables); - } -} - -static void icp_irq_postinstall(struct drm_i915_private *dev_priv) -{ - struct intel_uncore *uncore = &dev_priv->uncore; - u32 mask = SDE_GMBUS_ICP; - - GEN3_IRQ_INIT(uncore, SDE, ~mask, 0xffffffff); -} - static void gen8_irq_postinstall(struct drm_i915_private *dev_priv) { if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP) @@ -3397,17 +839,6 @@ static void gen8_irq_postinstall(struct drm_i915_private *dev_priv) gen8_master_intr_enable(dev_priv->uncore.regs); } -static void gen11_de_irq_postinstall(struct drm_i915_private *dev_priv) -{ - if (!HAS_DISPLAY(dev_priv)) - return; - - gen8_de_irq_postinstall(dev_priv); - - intel_uncore_write(&dev_priv->uncore, GEN11_DISPLAY_INT_CTL, - GEN11_DISPLAY_IRQ_ENABLE); -} - static void gen11_irq_postinstall(struct drm_i915_private *dev_priv) { struct intel_gt *gt = to_gt(dev_priv); @@ -3439,7 +870,11 @@ static void dg1_irq_postinstall(struct drm_i915_private *dev_priv) GEN3_IRQ_INIT(uncore, GEN11_GU_MISC_, ~gu_misc_masked, gu_misc_masked); if (HAS_DISPLAY(dev_priv)) { - icp_irq_postinstall(dev_priv); + if (DISPLAY_VER(dev_priv) >= 14) + mtp_irq_postinstall(dev_priv); + else + icp_irq_postinstall(dev_priv); + gen8_de_irq_postinstall(dev_priv); intel_uncore_write(&dev_priv->uncore, GEN11_DISPLAY_INT_CTL, GEN11_DISPLAY_IRQ_ENABLE); @@ -3831,31 +1266,6 @@ static void i965_irq_postinstall(struct drm_i915_private *dev_priv) i915_enable_asle_pipestat(dev_priv); } -static void i915_hpd_irq_setup(struct drm_i915_private *dev_priv) -{ - u32 hotplug_en; - - lockdep_assert_held(&dev_priv->irq_lock); - - /* Note HDMI and DP share hotplug bits */ - /* enable bits are the same for all generations */ - hotplug_en = intel_hpd_enabled_irqs(dev_priv, hpd_mask_i915); - /* Programming the CRT detection parameters tends - to generate a spurious hotplug event about three - seconds later. So just do it once. - */ - if (IS_G4X(dev_priv)) - hotplug_en |= CRT_HOTPLUG_ACTIVATION_PERIOD_64; - hotplug_en |= CRT_HOTPLUG_VOLTAGE_COMPARE_50; - - /* Ignore TV since it's buggy */ - i915_hotplug_interrupt_update_locked(dev_priv, - HOTPLUG_INT_EN_MASK | - CRT_HOTPLUG_VOLTAGE_COMPARE_MASK | - CRT_HOTPLUG_ACTIVATION_PERIOD_64, - hotplug_en); -} - static irqreturn_t i965_irq_handler(int irq, void *arg) { struct drm_i915_private *dev_priv = arg; @@ -3915,30 +1325,6 @@ static irqreturn_t i965_irq_handler(int irq, void *arg) return ret; } -struct intel_hotplug_funcs { - void (*hpd_irq_setup)(struct drm_i915_private *i915); -}; - -#define HPD_FUNCS(platform) \ -static const struct intel_hotplug_funcs platform##_hpd_funcs = { \ - .hpd_irq_setup = platform##_hpd_irq_setup, \ -} - -HPD_FUNCS(i915); -HPD_FUNCS(dg1); -HPD_FUNCS(gen11); -HPD_FUNCS(bxt); -HPD_FUNCS(icp); -HPD_FUNCS(spt); -HPD_FUNCS(ilk); -#undef HPD_FUNCS - -void intel_hpd_irq_setup(struct drm_i915_private *i915) -{ - if (i915->display_irqs_enabled && i915->display.funcs.hotplug) - i915->display.funcs.hotplug->hpd_irq_setup(i915); -} - /** * intel_irq_init - initializes irq support * @dev_priv: i915 device instance @@ -3961,10 +1347,6 @@ void intel_irq_init(struct drm_i915_private *dev_priv) if (!HAS_DISPLAY(dev_priv)) return; - intel_hpd_init_pins(dev_priv); - - intel_hpd_init_early(dev_priv); - dev_priv->drm.vblank_disable_immediate = true; /* Most platforms treat the display irq block as an always-on @@ -3977,25 +1359,7 @@ void intel_irq_init(struct drm_i915_private *dev_priv) if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) dev_priv->display_irqs_enabled = false; - if (HAS_GMCH(dev_priv)) { - if (I915_HAS_HOTPLUG(dev_priv)) - dev_priv->display.funcs.hotplug = &i915_hpd_funcs; - } else { - if (HAS_PCH_DG2(dev_priv)) - dev_priv->display.funcs.hotplug = &icp_hpd_funcs; - else if (HAS_PCH_DG1(dev_priv)) - dev_priv->display.funcs.hotplug = &dg1_hpd_funcs; - else if (DISPLAY_VER(dev_priv) >= 11) - dev_priv->display.funcs.hotplug = &gen11_hpd_funcs; - else if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) - dev_priv->display.funcs.hotplug = &bxt_hpd_funcs; - else if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP) - dev_priv->display.funcs.hotplug = &icp_hpd_funcs; - else if (INTEL_PCH_TYPE(dev_priv) >= PCH_SPT) - dev_priv->display.funcs.hotplug = &spt_hpd_funcs; - else - dev_priv->display.funcs.hotplug = &ilk_hpd_funcs; - } + intel_hotplug_irq_init(dev_priv); } /** @@ -4140,7 +1504,7 @@ void intel_irq_uninstall(struct drm_i915_private *dev_priv) /* * FIXME we can get called twice during driver probe * error handling as well as during driver remove due to - * intel_modeset_driver_remove() calling us out of sequence. + * intel_display_driver_remove() calling us out of sequence. * Would be nice if it didn't do that... */ if (!dev_priv->irq_enabled) diff --git a/drivers/gpu/drm/i915/i915_irq.h b/drivers/gpu/drm/i915/i915_irq.h index 03ee4c8b1ed38a6c0ac0bd8a898e2a4b5be5c509..e665a1b007dccc9324eb673dcccefeeb9cca96b9 100644 --- a/drivers/gpu/drm/i915/i915_irq.h +++ b/drivers/gpu/drm/i915/i915_irq.h @@ -9,7 +9,7 @@ #include #include -#include "i915_reg.h" +#include "i915_reg_defs.h" enum pipe; struct drm_crtc; @@ -17,6 +17,7 @@ struct drm_device; struct drm_display_mode; struct drm_i915_private; struct intel_crtc; +struct intel_encoder; struct intel_uncore; void intel_irq_init(struct drm_i915_private *dev_priv); @@ -24,33 +25,6 @@ void intel_irq_fini(struct drm_i915_private *dev_priv); int intel_irq_install(struct drm_i915_private *dev_priv); void intel_irq_uninstall(struct drm_i915_private *dev_priv); -u32 i915_pipestat_enable_mask(struct drm_i915_private *dev_priv, - enum pipe pipe); -void -i915_enable_pipestat(struct drm_i915_private *dev_priv, enum pipe pipe, - u32 status_mask); - -void -i915_disable_pipestat(struct drm_i915_private *dev_priv, enum pipe pipe, - u32 status_mask); - -void valleyview_enable_display_irqs(struct drm_i915_private *dev_priv); -void valleyview_disable_display_irqs(struct drm_i915_private *dev_priv); - -void intel_hpd_irq_setup(struct drm_i915_private *i915); -void i915_hotplug_interrupt_update(struct drm_i915_private *dev_priv, - u32 mask, - u32 bits); - -void ilk_enable_display_irq(struct drm_i915_private *i915, u32 bits); -void ilk_disable_display_irq(struct drm_i915_private *i915, u32 bits); - -void bdw_enable_pipe_irq(struct drm_i915_private *i915, enum pipe pipe, u32 bits); -void bdw_disable_pipe_irq(struct drm_i915_private *i915, enum pipe pipe, u32 bits); - -void ibx_enable_display_interrupt(struct drm_i915_private *i915, u32 bits); -void ibx_disable_display_interrupt(struct drm_i915_private *i915, u32 bits); - void gen5_enable_gt_irq(struct drm_i915_private *dev_priv, u32 mask); void gen5_disable_gt_irq(struct drm_i915_private *dev_priv, u32 mask); void gen11_reset_rps_interrupts(struct drm_i915_private *dev_priv); @@ -66,23 +40,7 @@ bool intel_irqs_enabled(struct drm_i915_private *dev_priv); void intel_synchronize_irq(struct drm_i915_private *i915); void intel_synchronize_hardirq(struct drm_i915_private *i915); -void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv, - u8 pipe_mask); -void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv, - u8 pipe_mask); -u32 gen8_de_pipe_underrun_mask(struct drm_i915_private *dev_priv); - - -int i8xx_enable_vblank(struct drm_crtc *crtc); -int i915gm_enable_vblank(struct drm_crtc *crtc); -int i965_enable_vblank(struct drm_crtc *crtc); -int ilk_enable_vblank(struct drm_crtc *crtc); -int bdw_enable_vblank(struct drm_crtc *crtc); -void i8xx_disable_vblank(struct drm_crtc *crtc); -void i915gm_disable_vblank(struct drm_crtc *crtc); -void i965_disable_vblank(struct drm_crtc *crtc); -void ilk_disable_vblank(struct drm_crtc *crtc); -void bdw_disable_vblank(struct drm_crtc *crtc); +void gen3_assert_iir_is_zero(struct intel_uncore *uncore, i915_reg_t reg); void gen3_irq_reset(struct intel_uncore *uncore, i915_reg_t imr, i915_reg_t iir, i915_reg_t ier); diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c index 75cbccd1a441b7775a0dd2e150745d7a204e0d84..3d7a5db9833b731e02e17d37dec97e07e0e11a99 100644 --- a/drivers/gpu/drm/i915/i915_pci.c +++ b/drivers/gpu/drm/i915/i915_pci.c @@ -27,6 +27,7 @@ #include #include "display/intel_display.h" +#include "display/intel_display_driver.h" #include "gt/intel_gt_regs.h" #include "gt/intel_sa_media.h" #include "gem/i915_gem_object_types.h" @@ -37,132 +38,13 @@ #include "i915_reg.h" #include "intel_pci_config.h" +__diag_push(); +__diag_ignore_all("-Woverride-init", "Allow overriding inherited members"); + #define PLATFORM(x) .platform = (x) #define GEN(x) \ .__runtime.graphics.ip.ver = (x), \ - .__runtime.media.ip.ver = (x), \ - .__runtime.display.ip.ver = (x) - -#define NO_DISPLAY .__runtime.pipe_mask = 0 - -#define I845_PIPE_OFFSETS \ - .display.pipe_offsets = { \ - [TRANSCODER_A] = PIPE_A_OFFSET, \ - }, \ - .display.trans_offsets = { \ - [TRANSCODER_A] = TRANSCODER_A_OFFSET, \ - } - -#define I9XX_PIPE_OFFSETS \ - .display.pipe_offsets = { \ - [TRANSCODER_A] = PIPE_A_OFFSET, \ - [TRANSCODER_B] = PIPE_B_OFFSET, \ - }, \ - .display.trans_offsets = { \ - [TRANSCODER_A] = TRANSCODER_A_OFFSET, \ - [TRANSCODER_B] = TRANSCODER_B_OFFSET, \ - } - -#define IVB_PIPE_OFFSETS \ - .display.pipe_offsets = { \ - [TRANSCODER_A] = PIPE_A_OFFSET, \ - [TRANSCODER_B] = PIPE_B_OFFSET, \ - [TRANSCODER_C] = PIPE_C_OFFSET, \ - }, \ - .display.trans_offsets = { \ - [TRANSCODER_A] = TRANSCODER_A_OFFSET, \ - [TRANSCODER_B] = TRANSCODER_B_OFFSET, \ - [TRANSCODER_C] = TRANSCODER_C_OFFSET, \ - } - -#define HSW_PIPE_OFFSETS \ - .display.pipe_offsets = { \ - [TRANSCODER_A] = PIPE_A_OFFSET, \ - [TRANSCODER_B] = PIPE_B_OFFSET, \ - [TRANSCODER_C] = PIPE_C_OFFSET, \ - [TRANSCODER_EDP] = PIPE_EDP_OFFSET, \ - }, \ - .display.trans_offsets = { \ - [TRANSCODER_A] = TRANSCODER_A_OFFSET, \ - [TRANSCODER_B] = TRANSCODER_B_OFFSET, \ - [TRANSCODER_C] = TRANSCODER_C_OFFSET, \ - [TRANSCODER_EDP] = TRANSCODER_EDP_OFFSET, \ - } - -#define CHV_PIPE_OFFSETS \ - .display.pipe_offsets = { \ - [TRANSCODER_A] = PIPE_A_OFFSET, \ - [TRANSCODER_B] = PIPE_B_OFFSET, \ - [TRANSCODER_C] = CHV_PIPE_C_OFFSET, \ - }, \ - .display.trans_offsets = { \ - [TRANSCODER_A] = TRANSCODER_A_OFFSET, \ - [TRANSCODER_B] = TRANSCODER_B_OFFSET, \ - [TRANSCODER_C] = CHV_TRANSCODER_C_OFFSET, \ - } - -#define I845_CURSOR_OFFSETS \ - .display.cursor_offsets = { \ - [PIPE_A] = CURSOR_A_OFFSET, \ - } - -#define I9XX_CURSOR_OFFSETS \ - .display.cursor_offsets = { \ - [PIPE_A] = CURSOR_A_OFFSET, \ - [PIPE_B] = CURSOR_B_OFFSET, \ - } - -#define CHV_CURSOR_OFFSETS \ - .display.cursor_offsets = { \ - [PIPE_A] = CURSOR_A_OFFSET, \ - [PIPE_B] = CURSOR_B_OFFSET, \ - [PIPE_C] = CHV_CURSOR_C_OFFSET, \ - } - -#define IVB_CURSOR_OFFSETS \ - .display.cursor_offsets = { \ - [PIPE_A] = CURSOR_A_OFFSET, \ - [PIPE_B] = IVB_CURSOR_B_OFFSET, \ - [PIPE_C] = IVB_CURSOR_C_OFFSET, \ - } - -#define TGL_CURSOR_OFFSETS \ - .display.cursor_offsets = { \ - [PIPE_A] = CURSOR_A_OFFSET, \ - [PIPE_B] = IVB_CURSOR_B_OFFSET, \ - [PIPE_C] = IVB_CURSOR_C_OFFSET, \ - [PIPE_D] = TGL_CURSOR_D_OFFSET, \ - } - -#define I845_COLORS \ - .display.color = { .gamma_lut_size = 256 } -#define I9XX_COLORS \ - .display.color = { .gamma_lut_size = 129, \ - .gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \ - } -#define ILK_COLORS \ - .display.color = { .gamma_lut_size = 1024 } -#define IVB_COLORS \ - .display.color = { .degamma_lut_size = 1024, .gamma_lut_size = 1024 } -#define CHV_COLORS \ - .display.color = { \ - .degamma_lut_size = 65, .gamma_lut_size = 257, \ - .degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \ - .gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \ - } -#define GLK_COLORS \ - .display.color = { \ - .degamma_lut_size = 33, .gamma_lut_size = 1024, \ - .degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING | \ - DRM_COLOR_LUT_EQUAL_CHANNELS, \ - } -#define ICL_COLORS \ - .display.color = { \ - .degamma_lut_size = 33, .gamma_lut_size = 262145, \ - .degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING | \ - DRM_COLOR_LUT_EQUAL_CHANNELS, \ - .gamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING, \ - } + .__runtime.media.ip.ver = (x) #define LEGACY_CACHELEVEL \ .cachelevel_to_pat = { \ @@ -207,12 +89,6 @@ #define I830_FEATURES \ GEN(2), \ .is_mobile = 1, \ - .__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \ - .__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \ - .display.has_overlay = 1, \ - .display.cursor_needs_physical = 1, \ - .display.overlay_needs_physical = 1, \ - .display.has_gmch = 1, \ .gpu_reset_clobbers_display = true, \ .has_3d_pipeline = 1, \ .hws_needs_physical = 1, \ @@ -222,20 +98,12 @@ .has_coherent_ggtt = false, \ .dma_mask_size = 32, \ .max_pat_index = 3, \ - I9XX_PIPE_OFFSETS, \ - I9XX_CURSOR_OFFSETS, \ - I9XX_COLORS, \ GEN_DEFAULT_PAGE_SIZES, \ GEN_DEFAULT_REGIONS, \ LEGACY_CACHELEVEL #define I845_FEATURES \ GEN(2), \ - .__runtime.pipe_mask = BIT(PIPE_A), \ - .__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A), \ - .display.has_overlay = 1, \ - .display.overlay_needs_physical = 1, \ - .display.has_gmch = 1, \ .has_3d_pipeline = 1, \ .gpu_reset_clobbers_display = true, \ .hws_needs_physical = 1, \ @@ -245,9 +113,6 @@ .has_coherent_ggtt = false, \ .dma_mask_size = 32, \ .max_pat_index = 3, \ - I845_PIPE_OFFSETS, \ - I845_CURSOR_OFFSETS, \ - I845_COLORS, \ GEN_DEFAULT_PAGE_SIZES, \ GEN_DEFAULT_REGIONS, \ LEGACY_CACHELEVEL @@ -265,20 +130,15 @@ static const struct intel_device_info i845g_info = { static const struct intel_device_info i85x_info = { I830_FEATURES, PLATFORM(INTEL_I85X), - .__runtime.fbc_mask = BIT(INTEL_FBC_A), }; static const struct intel_device_info i865g_info = { I845_FEATURES, PLATFORM(INTEL_I865G), - .__runtime.fbc_mask = BIT(INTEL_FBC_A), }; #define GEN3_FEATURES \ GEN(3), \ - .__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \ - .__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \ - .display.has_gmch = 1, \ .gpu_reset_clobbers_display = true, \ .__runtime.platform_engine_mask = BIT(RCS0), \ .has_3d_pipeline = 1, \ @@ -286,9 +146,6 @@ static const struct intel_device_info i865g_info = { .has_coherent_ggtt = true, \ .dma_mask_size = 32, \ .max_pat_index = 3, \ - I9XX_PIPE_OFFSETS, \ - I9XX_CURSOR_OFFSETS, \ - I9XX_COLORS, \ GEN_DEFAULT_PAGE_SIZES, \ GEN_DEFAULT_REGIONS, \ LEGACY_CACHELEVEL @@ -297,9 +154,6 @@ static const struct intel_device_info i915g_info = { GEN3_FEATURES, PLATFORM(INTEL_I915G), .has_coherent_ggtt = false, - .display.cursor_needs_physical = 1, - .display.has_overlay = 1, - .display.overlay_needs_physical = 1, .hws_needs_physical = 1, .unfenced_needs_alignment = 1, }; @@ -308,11 +162,6 @@ static const struct intel_device_info i915gm_info = { GEN3_FEATURES, PLATFORM(INTEL_I915GM), .is_mobile = 1, - .display.cursor_needs_physical = 1, - .display.has_overlay = 1, - .display.overlay_needs_physical = 1, - .display.supports_tv = 1, - .__runtime.fbc_mask = BIT(INTEL_FBC_A), .hws_needs_physical = 1, .unfenced_needs_alignment = 1, }; @@ -320,10 +169,6 @@ static const struct intel_device_info i915gm_info = { static const struct intel_device_info i945g_info = { GEN3_FEATURES, PLATFORM(INTEL_I945G), - .display.has_hotplug = 1, - .display.cursor_needs_physical = 1, - .display.has_overlay = 1, - .display.overlay_needs_physical = 1, .hws_needs_physical = 1, .unfenced_needs_alignment = 1, }; @@ -332,12 +177,6 @@ static const struct intel_device_info i945gm_info = { GEN3_FEATURES, PLATFORM(INTEL_I945GM), .is_mobile = 1, - .display.has_hotplug = 1, - .display.cursor_needs_physical = 1, - .display.has_overlay = 1, - .display.overlay_needs_physical = 1, - .display.supports_tv = 1, - .__runtime.fbc_mask = BIT(INTEL_FBC_A), .hws_needs_physical = 1, .unfenced_needs_alignment = 1, }; @@ -345,16 +184,12 @@ static const struct intel_device_info i945gm_info = { static const struct intel_device_info g33_info = { GEN3_FEATURES, PLATFORM(INTEL_G33), - .display.has_hotplug = 1, - .display.has_overlay = 1, .dma_mask_size = 36, }; static const struct intel_device_info pnv_g_info = { GEN3_FEATURES, PLATFORM(INTEL_PINEVIEW), - .display.has_hotplug = 1, - .display.has_overlay = 1, .dma_mask_size = 36, }; @@ -362,17 +197,11 @@ static const struct intel_device_info pnv_m_info = { GEN3_FEATURES, PLATFORM(INTEL_PINEVIEW), .is_mobile = 1, - .display.has_hotplug = 1, - .display.has_overlay = 1, .dma_mask_size = 36, }; #define GEN4_FEATURES \ GEN(4), \ - .__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \ - .__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \ - .display.has_hotplug = 1, \ - .display.has_gmch = 1, \ .gpu_reset_clobbers_display = true, \ .__runtime.platform_engine_mask = BIT(RCS0), \ .has_3d_pipeline = 1, \ @@ -380,9 +209,6 @@ static const struct intel_device_info pnv_m_info = { .has_coherent_ggtt = true, \ .dma_mask_size = 36, \ .max_pat_index = 3, \ - I9XX_PIPE_OFFSETS, \ - I9XX_CURSOR_OFFSETS, \ - I9XX_COLORS, \ GEN_DEFAULT_PAGE_SIZES, \ GEN_DEFAULT_REGIONS, \ LEGACY_CACHELEVEL @@ -390,7 +216,6 @@ static const struct intel_device_info pnv_m_info = { static const struct intel_device_info i965g_info = { GEN4_FEATURES, PLATFORM(INTEL_I965G), - .display.has_overlay = 1, .hws_needs_physical = 1, .has_snoop = false, }; @@ -399,9 +224,6 @@ static const struct intel_device_info i965gm_info = { GEN4_FEATURES, PLATFORM(INTEL_I965GM), .is_mobile = 1, - .__runtime.fbc_mask = BIT(INTEL_FBC_A), - .display.has_overlay = 1, - .display.supports_tv = 1, .hws_needs_physical = 1, .has_snoop = false, }; @@ -417,17 +239,12 @@ static const struct intel_device_info gm45_info = { GEN4_FEATURES, PLATFORM(INTEL_GM45), .is_mobile = 1, - .__runtime.fbc_mask = BIT(INTEL_FBC_A), - .display.supports_tv = 1, .__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0), .gpu_reset_clobbers_display = false, }; #define GEN5_FEATURES \ GEN(5), \ - .__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \ - .__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \ - .display.has_hotplug = 1, \ .__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0), \ .has_3d_pipeline = 1, \ .has_snoop = true, \ @@ -436,9 +253,6 @@ static const struct intel_device_info gm45_info = { .has_rc6 = 0, \ .dma_mask_size = 36, \ .max_pat_index = 3, \ - I9XX_PIPE_OFFSETS, \ - I9XX_CURSOR_OFFSETS, \ - ILK_COLORS, \ GEN_DEFAULT_PAGE_SIZES, \ GEN_DEFAULT_REGIONS, \ LEGACY_CACHELEVEL @@ -453,15 +267,10 @@ static const struct intel_device_info ilk_m_info = { PLATFORM(INTEL_IRONLAKE), .is_mobile = 1, .has_rps = true, - .__runtime.fbc_mask = BIT(INTEL_FBC_A), }; #define GEN6_FEATURES \ GEN(6), \ - .__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \ - .__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \ - .display.has_hotplug = 1, \ - .__runtime.fbc_mask = BIT(INTEL_FBC_A), \ .__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0), \ .has_3d_pipeline = 1, \ .has_coherent_ggtt = true, \ @@ -474,9 +283,6 @@ static const struct intel_device_info ilk_m_info = { .max_pat_index = 3, \ .__runtime.ppgtt_type = INTEL_PPGTT_ALIASING, \ .__runtime.ppgtt_size = 31, \ - I9XX_PIPE_OFFSETS, \ - I9XX_CURSOR_OFFSETS, \ - ILK_COLORS, \ GEN_DEFAULT_PAGE_SIZES, \ GEN_DEFAULT_REGIONS, \ LEGACY_CACHELEVEL @@ -513,10 +319,6 @@ static const struct intel_device_info snb_m_gt2_info = { #define GEN7_FEATURES \ GEN(7), \ - .__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), \ - .__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C), \ - .display.has_hotplug = 1, \ - .__runtime.fbc_mask = BIT(INTEL_FBC_A), \ .__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0), \ .has_3d_pipeline = 1, \ .has_coherent_ggtt = true, \ @@ -529,9 +331,6 @@ static const struct intel_device_info snb_m_gt2_info = { .max_pat_index = 3, \ .__runtime.ppgtt_type = INTEL_PPGTT_ALIASING, \ .__runtime.ppgtt_size = 31, \ - IVB_PIPE_OFFSETS, \ - IVB_CURSOR_OFFSETS, \ - IVB_COLORS, \ GEN_DEFAULT_PAGE_SIZES, \ GEN_DEFAULT_REGIONS, \ LEGACY_CACHELEVEL @@ -570,7 +369,6 @@ static const struct intel_device_info ivb_m_gt2_info = { static const struct intel_device_info ivb_q_info = { GEN7_FEATURES, PLATFORM(INTEL_IVYBRIDGE), - NO_DISPLAY, .gt = 2, .has_l3_dpf = 1, }; @@ -579,14 +377,10 @@ static const struct intel_device_info vlv_info = { PLATFORM(INTEL_VALLEYVIEW), GEN(7), .is_lp = 1, - .__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), - .__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), .has_runtime_pm = 1, .has_rc6 = 1, .has_reset_engine = true, .has_rps = true, - .display.has_gmch = 1, - .display.has_hotplug = 1, .dma_mask_size = 40, .max_pat_index = 3, .__runtime.ppgtt_type = INTEL_PPGTT_ALIASING, @@ -594,10 +388,6 @@ static const struct intel_device_info vlv_info = { .has_snoop = true, .has_coherent_ggtt = false, .__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0), - .display.mmio_offset = VLV_DISPLAY_BASE, - I9XX_PIPE_OFFSETS, - I9XX_CURSOR_OFFSETS, - I9XX_COLORS, GEN_DEFAULT_PAGE_SIZES, GEN_DEFAULT_REGIONS, LEGACY_CACHELEVEL, @@ -606,13 +396,7 @@ static const struct intel_device_info vlv_info = { #define G75_FEATURES \ GEN7_FEATURES, \ .__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0), \ - .__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \ - BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP), \ - .display.has_ddi = 1, \ - .display.has_fpga_dbg = 1, \ - .display.has_dp_mst = 1, \ .has_rc6p = 0 /* RC6p removed-by HSW */, \ - HSW_PIPE_OFFSETS, \ .has_runtime_pm = 1 #define HSW_PLATFORM \ @@ -676,9 +460,6 @@ static const struct intel_device_info bdw_gt3_info = { static const struct intel_device_info chv_info = { PLATFORM(INTEL_CHERRYVIEW), GEN(8), - .__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), - .__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C), - .display.has_hotplug = 1, .is_lp = 1, .__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0), .has_64bit_reloc = 1, @@ -686,7 +467,6 @@ static const struct intel_device_info chv_info = { .has_rc6 = 1, .has_rps = true, .has_logical_ring_contexts = 1, - .display.has_gmch = 1, .dma_mask_size = 39, .max_pat_index = 3, .__runtime.ppgtt_type = INTEL_PPGTT_FULL, @@ -694,10 +474,6 @@ static const struct intel_device_info chv_info = { .has_reset_engine = 1, .has_snoop = true, .has_coherent_ggtt = false, - .display.mmio_offset = VLV_DISPLAY_BASE, - CHV_PIPE_OFFSETS, - CHV_CURSOR_OFFSETS, - CHV_COLORS, GEN_DEFAULT_PAGE_SIZES, GEN_DEFAULT_REGIONS, LEGACY_CACHELEVEL, @@ -711,14 +487,7 @@ static const struct intel_device_info chv_info = { GEN8_FEATURES, \ GEN(9), \ GEN9_DEFAULT_PAGE_SIZES, \ - .__runtime.has_dmc = 1, \ - .has_gt_uc = 1, \ - .__runtime.has_hdcp = 1, \ - .display.has_ipc = 1, \ - .display.has_psr = 1, \ - .display.has_psr_hw_tracking = 1, \ - .display.dbuf.size = 896 - 4, /* 4 blocks for bypass path allocation */ \ - .display.dbuf.slice_mask = BIT(DBUF_S1) + .has_gt_uc = 1 #define SKL_PLATFORM \ GEN9_FEATURES, \ @@ -753,26 +522,12 @@ static const struct intel_device_info skl_gt4_info = { #define GEN9_LP_FEATURES \ GEN(9), \ .is_lp = 1, \ - .display.dbuf.slice_mask = BIT(DBUF_S1), \ - .display.has_hotplug = 1, \ .__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0), \ - .__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), \ - .__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \ - BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP) | \ - BIT(TRANSCODER_DSI_A) | BIT(TRANSCODER_DSI_C), \ .has_3d_pipeline = 1, \ .has_64bit_reloc = 1, \ - .display.has_ddi = 1, \ - .display.has_fpga_dbg = 1, \ - .__runtime.fbc_mask = BIT(INTEL_FBC_A), \ - .__runtime.has_hdcp = 1, \ - .display.has_psr = 1, \ - .display.has_psr_hw_tracking = 1, \ .has_runtime_pm = 1, \ - .__runtime.has_dmc = 1, \ .has_rc6 = 1, \ .has_rps = true, \ - .display.has_dp_mst = 1, \ .has_logical_ring_contexts = 1, \ .has_gt_uc = 1, \ .dma_mask_size = 39, \ @@ -781,11 +536,7 @@ static const struct intel_device_info skl_gt4_info = { .has_reset_engine = 1, \ .has_snoop = true, \ .has_coherent_ggtt = false, \ - .display.has_ipc = 1, \ .max_pat_index = 3, \ - HSW_PIPE_OFFSETS, \ - IVB_CURSOR_OFFSETS, \ - IVB_COLORS, \ GEN9_DEFAULT_PAGE_SIZES, \ GEN_DEFAULT_REGIONS, \ LEGACY_CACHELEVEL @@ -793,15 +544,11 @@ static const struct intel_device_info skl_gt4_info = { static const struct intel_device_info bxt_info = { GEN9_LP_FEATURES, PLATFORM(INTEL_BROXTON), - .display.dbuf.size = 512 - 4, /* 4 blocks for bypass path allocation */ }; static const struct intel_device_info glk_info = { GEN9_LP_FEATURES, PLATFORM(INTEL_GEMINILAKE), - .__runtime.display.ip.ver = 10, - .display.dbuf.size = 1024 - 4, /* 4 blocks for bypass path allocation */ - GLK_COLORS, }; #define KBL_PLATFORM \ @@ -868,31 +615,7 @@ static const struct intel_device_info cml_gt2_info = { #define GEN11_FEATURES \ GEN9_FEATURES, \ GEN11_DEFAULT_PAGE_SIZES, \ - .display.abox_mask = BIT(0), \ - .__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \ - BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP) | \ - BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1), \ - .display.pipe_offsets = { \ - [TRANSCODER_A] = PIPE_A_OFFSET, \ - [TRANSCODER_B] = PIPE_B_OFFSET, \ - [TRANSCODER_C] = PIPE_C_OFFSET, \ - [TRANSCODER_EDP] = PIPE_EDP_OFFSET, \ - [TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET, \ - [TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET, \ - }, \ - .display.trans_offsets = { \ - [TRANSCODER_A] = TRANSCODER_A_OFFSET, \ - [TRANSCODER_B] = TRANSCODER_B_OFFSET, \ - [TRANSCODER_C] = TRANSCODER_C_OFFSET, \ - [TRANSCODER_EDP] = TRANSCODER_EDP_OFFSET, \ - [TRANSCODER_DSI_0] = TRANSCODER_DSI0_OFFSET, \ - [TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET, \ - }, \ GEN(11), \ - ICL_COLORS, \ - .display.dbuf.size = 2048, \ - .display.dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2), \ - .__runtime.has_dsc = 1, \ .has_coherent_ggtt = false, \ .has_logical_ring_elsq = 1 @@ -920,32 +643,9 @@ static const struct intel_device_info jsl_info = { #define GEN12_FEATURES \ GEN11_FEATURES, \ GEN(12), \ - .display.abox_mask = GENMASK(2, 1), \ - .__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D), \ - .__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \ - BIT(TRANSCODER_C) | BIT(TRANSCODER_D) | \ - BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1), \ - .display.pipe_offsets = { \ - [TRANSCODER_A] = PIPE_A_OFFSET, \ - [TRANSCODER_B] = PIPE_B_OFFSET, \ - [TRANSCODER_C] = PIPE_C_OFFSET, \ - [TRANSCODER_D] = PIPE_D_OFFSET, \ - [TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET, \ - [TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET, \ - }, \ - .display.trans_offsets = { \ - [TRANSCODER_A] = TRANSCODER_A_OFFSET, \ - [TRANSCODER_B] = TRANSCODER_B_OFFSET, \ - [TRANSCODER_C] = TRANSCODER_C_OFFSET, \ - [TRANSCODER_D] = TRANSCODER_D_OFFSET, \ - [TRANSCODER_DSI_0] = TRANSCODER_DSI0_OFFSET, \ - [TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET, \ - }, \ - TGL_CURSOR_OFFSETS, \ TGL_CACHELEVEL, \ .has_global_mocs = 1, \ .has_pxp = 1, \ - .display.has_dsb = 1, \ .max_pat_index = 3 static const struct intel_device_info tgl_info = { @@ -958,12 +658,6 @@ static const struct intel_device_info tgl_info = { static const struct intel_device_info rkl_info = { GEN12_FEATURES, PLATFORM(INTEL_ROCKETLAKE), - .display.abox_mask = BIT(0), - .__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), - .__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | - BIT(TRANSCODER_C), - .display.has_hti = 1, - .display.has_psr_hw_tracking = 0, .__runtime.platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0), }; @@ -981,7 +675,6 @@ static const struct intel_device_info dg1_info = { DGFX_FEATURES, .__runtime.graphics.ip.rel = 10, PLATFORM(INTEL_DG1), - .__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D), .require_force_probe = 1, .__runtime.platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | @@ -993,64 +686,14 @@ static const struct intel_device_info dg1_info = { static const struct intel_device_info adl_s_info = { GEN12_FEATURES, PLATFORM(INTEL_ALDERLAKE_S), - .__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D), - .display.has_hti = 1, - .display.has_psr_hw_tracking = 0, .__runtime.platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2), .dma_mask_size = 39, }; -#define XE_LPD_FEATURES \ - .display.abox_mask = GENMASK(1, 0), \ - .display.color = { \ - .degamma_lut_size = 129, .gamma_lut_size = 1024, \ - .degamma_lut_tests = DRM_COLOR_LUT_NON_DECREASING | \ - DRM_COLOR_LUT_EQUAL_CHANNELS, \ - }, \ - .display.dbuf.size = 4096, \ - .display.dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2) | BIT(DBUF_S3) | \ - BIT(DBUF_S4), \ - .display.has_ddi = 1, \ - .__runtime.has_dmc = 1, \ - .display.has_dp_mst = 1, \ - .display.has_dsb = 1, \ - .__runtime.has_dsc = 1, \ - .__runtime.fbc_mask = BIT(INTEL_FBC_A), \ - .display.has_fpga_dbg = 1, \ - .__runtime.has_hdcp = 1, \ - .display.has_hotplug = 1, \ - .display.has_ipc = 1, \ - .display.has_psr = 1, \ - .__runtime.display.ip.ver = 13, \ - .__runtime.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C) | BIT(PIPE_D), \ - .display.pipe_offsets = { \ - [TRANSCODER_A] = PIPE_A_OFFSET, \ - [TRANSCODER_B] = PIPE_B_OFFSET, \ - [TRANSCODER_C] = PIPE_C_OFFSET, \ - [TRANSCODER_D] = PIPE_D_OFFSET, \ - [TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET, \ - [TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET, \ - }, \ - .display.trans_offsets = { \ - [TRANSCODER_A] = TRANSCODER_A_OFFSET, \ - [TRANSCODER_B] = TRANSCODER_B_OFFSET, \ - [TRANSCODER_C] = TRANSCODER_C_OFFSET, \ - [TRANSCODER_D] = TRANSCODER_D_OFFSET, \ - [TRANSCODER_DSI_0] = TRANSCODER_DSI0_OFFSET, \ - [TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET, \ - }, \ - TGL_CURSOR_OFFSETS - static const struct intel_device_info adl_p_info = { GEN12_FEATURES, - XE_LPD_FEATURES, PLATFORM(INTEL_ALDERLAKE_P), - .__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | - BIT(TRANSCODER_C) | BIT(TRANSCODER_D) | - BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1), - .display.has_cdclk_crawl = 1, - .display.has_psr_hw_tracking = 0, .__runtime.platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2), .__runtime.ppgtt_size = 48, @@ -1101,7 +744,6 @@ static const struct intel_device_info xehpsdv_info = { XE_HPM_FEATURES, DGFX_FEATURES, PLATFORM(INTEL_XEHPSDV), - NO_DISPLAY, .has_64k_pages = 1, .has_media_ratio_mode = 1, .__runtime.platform_engine_mask = @@ -1124,7 +766,6 @@ static const struct intel_device_info xehpsdv_info = { .has_guc_deprivilege = 1, \ .has_heci_pxp = 1, \ .has_media_ratio_mode = 1, \ - .display.has_cdclk_squash = 1, \ .__runtime.platform_engine_mask = \ BIT(RCS0) | BIT(BCS0) | \ BIT(VECS0) | BIT(VECS1) | \ @@ -1133,14 +774,10 @@ static const struct intel_device_info xehpsdv_info = { static const struct intel_device_info dg2_info = { DG2_FEATURES, - XE_LPD_FEATURES, - .__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | - BIT(TRANSCODER_C) | BIT(TRANSCODER_D), }; static const struct intel_device_info ats_m_info = { DG2_FEATURES, - NO_DISPLAY, .require_force_probe = 1, .tuning_thread_rr_after_dep = 1, }; @@ -1162,7 +799,6 @@ static const struct intel_device_info pvc_info = { .__runtime.graphics.ip.rel = 60, .__runtime.media.ip.rel = 60, PLATFORM(INTEL_PONTEVECCHIO), - NO_DISPLAY, .has_flat_ccs = 0, .max_pat_index = 7, .__runtime.platform_engine_mask = @@ -1173,13 +809,6 @@ static const struct intel_device_info pvc_info = { PVC_CACHELEVEL, }; -#define XE_LPDP_FEATURES \ - XE_LPD_FEATURES, \ - .__runtime.display.ip.ver = 14, \ - .display.has_cdclk_crawl = 1, \ - .display.has_cdclk_squash = 1, \ - .__runtime.fbc_mask = BIT(INTEL_FBC_A) | BIT(INTEL_FBC_B) - static const struct intel_gt_definition xelpmp_extra_gt[] = { { .type = GT_MEDIA, @@ -1192,9 +821,6 @@ static const struct intel_gt_definition xelpmp_extra_gt[] = { static const struct intel_device_info mtl_info = { XE_HP_FEATURES, - XE_LPDP_FEATURES, - .__runtime.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | - BIT(TRANSCODER_C) | BIT(TRANSCODER_D), /* * Real graphics IP version will be obtained from hardware GMD_ID * register. Value provided here is just for sanity checking. @@ -1220,6 +846,8 @@ static const struct intel_device_info mtl_info = { #undef PLATFORM +__diag_pop(); + /* * Make sure any device matches here are from most specific to most * general. For example, since the Quanta match is based on the subsystem @@ -1425,7 +1053,7 @@ static int i915_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) return -ENXIO; /* Detect if we need to wait for other drivers early on */ - if (intel_modeset_probe_defer(pdev)) + if (intel_display_driver_probe_defer(pdev)) return -EPROBE_DEFER; err = i915_driver_probe(pdev, ent); diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c index 19d5652300eebadf75540dcfaf0790d6e13973a0..0a111b281578ba0fc95b7902923e44ed2ae894bf 100644 --- a/drivers/gpu/drm/i915/i915_perf.c +++ b/drivers/gpu/drm/i915/i915_perf.c @@ -531,8 +531,7 @@ static void oa_context_id_squash(struct i915_perf_stream *stream, u32 *report) * (See description of OA_TAIL_MARGIN_NSEC above for further details.) * * Besides returning true when there is data available to read() this function - * also updates the tail, aging_tail and aging_timestamp in the oa_buffer - * object. + * also updates the tail in the oa_buffer object. * * Note: It's safe to read OA config state here unlocked, assuming that this is * only called while the stream is enabled, while the global OA configuration @@ -544,10 +543,10 @@ static bool oa_buffer_check_unlocked(struct i915_perf_stream *stream) { u32 gtt_offset = i915_ggtt_offset(stream->oa_buffer.vma); int report_size = stream->oa_buffer.format->size; + u32 head, tail, read_tail; unsigned long flags; bool pollin; u32 hw_tail; - u64 now; u32 partial_report_size; /* We have to consider the (unlikely) possibility that read() errors @@ -566,64 +565,48 @@ static bool oa_buffer_check_unlocked(struct i915_perf_stream *stream) partial_report_size %= report_size; /* Subtract partial amount off the tail */ - hw_tail = gtt_offset + OA_TAKEN(hw_tail, partial_report_size); + hw_tail = OA_TAKEN(hw_tail, partial_report_size); - now = ktime_get_mono_fast_ns(); + /* NB: The head we observe here might effectively be a little + * out of date. If a read() is in progress, the head could be + * anywhere between this head and stream->oa_buffer.tail. + */ + head = stream->oa_buffer.head - gtt_offset; + read_tail = stream->oa_buffer.tail - gtt_offset; - if (hw_tail == stream->oa_buffer.aging_tail && - (now - stream->oa_buffer.aging_timestamp) > OA_TAIL_MARGIN_NSEC) { - /* If the HW tail hasn't move since the last check and the HW - * tail has been aging for long enough, declare it the new - * tail. - */ - stream->oa_buffer.tail = stream->oa_buffer.aging_tail; - } else { - u32 head, tail, aged_tail; + tail = hw_tail; - /* NB: The head we observe here might effectively be a little - * out of date. If a read() is in progress, the head could be - * anywhere between this head and stream->oa_buffer.tail. - */ - head = stream->oa_buffer.head - gtt_offset; - aged_tail = stream->oa_buffer.tail - gtt_offset; - - hw_tail -= gtt_offset; - tail = hw_tail; - - /* Walk the stream backward until we find a report with report - * id and timestmap not at 0. Since the circular buffer pointers - * progress by increments of 64 bytes and that reports can be up - * to 256 bytes long, we can't tell whether a report has fully - * landed in memory before the report id and timestamp of the - * following report have effectively landed. - * - * This is assuming that the writes of the OA unit land in - * memory in the order they were written to. - * If not : (╯°□°）╯︵ ┻━┻ - */ - while (OA_TAKEN(tail, aged_tail) >= report_size) { - void *report = stream->oa_buffer.vaddr + tail; + /* Walk the stream backward until we find a report with report + * id and timestmap not at 0. Since the circular buffer pointers + * progress by increments of 64 bytes and that reports can be up + * to 256 bytes long, we can't tell whether a report has fully + * landed in memory before the report id and timestamp of the + * following report have effectively landed. + * + * This is assuming that the writes of the OA unit land in + * memory in the order they were written to. + * If not : (╯°□°）╯︵ ┻━┻ + */ + while (OA_TAKEN(tail, read_tail) >= report_size) { + void *report = stream->oa_buffer.vaddr + tail; - if (oa_report_id(stream, report) || - oa_timestamp(stream, report)) - break; + if (oa_report_id(stream, report) || + oa_timestamp(stream, report)) + break; - tail = (tail - report_size) & (OA_BUFFER_SIZE - 1); - } + tail = (tail - report_size) & (OA_BUFFER_SIZE - 1); + } - if (OA_TAKEN(hw_tail, tail) > report_size && - __ratelimit(&stream->perf->tail_pointer_race)) - drm_notice(&stream->uncore->i915->drm, - "unlanded report(s) head=0x%x tail=0x%x hw_tail=0x%x\n", - head, tail, hw_tail); + if (OA_TAKEN(hw_tail, tail) > report_size && + __ratelimit(&stream->perf->tail_pointer_race)) + drm_notice(&stream->uncore->i915->drm, + "unlanded report(s) head=0x%x tail=0x%x hw_tail=0x%x\n", + head, tail, hw_tail); - stream->oa_buffer.tail = gtt_offset + tail; - stream->oa_buffer.aging_tail = gtt_offset + hw_tail; - stream->oa_buffer.aging_timestamp = now; - } + stream->oa_buffer.tail = gtt_offset + tail; - pollin = OA_TAKEN(stream->oa_buffer.tail - gtt_offset, - stream->oa_buffer.head - gtt_offset) >= report_size; + pollin = OA_TAKEN(stream->oa_buffer.tail, + stream->oa_buffer.head) >= report_size; spin_unlock_irqrestore(&stream->oa_buffer.ptr_lock, flags); @@ -877,12 +860,17 @@ static int gen8_append_oa_reports(struct i915_perf_stream *stream, stream->oa_buffer.last_ctx_id = ctx_id; } - /* - * Clear out the report id and timestamp as a means to detect unlanded - * reports. - */ - oa_report_id_clear(stream, report32); - oa_timestamp_clear(stream, report32); + if (is_power_of_2(report_size)) { + /* + * Clear out the report id and timestamp as a means + * to detect unlanded reports. + */ + oa_report_id_clear(stream, report32); + oa_timestamp_clear(stream, report32); + } else { + /* Zero out the entire report */ + memset(report32, 0, report_size); + } } if (start_offset != *offset) { @@ -1722,7 +1710,6 @@ static void gen7_init_oa_buffer(struct i915_perf_stream *stream) gtt_offset | OABUFFER_SIZE_16M); /* Mark that we need updated tail pointers to read from... */ - stream->oa_buffer.aging_tail = INVALID_TAIL_PTR; stream->oa_buffer.tail = gtt_offset; spin_unlock_irqrestore(&stream->oa_buffer.ptr_lock, flags); @@ -1774,7 +1761,6 @@ static void gen8_init_oa_buffer(struct i915_perf_stream *stream) intel_uncore_write(uncore, GEN8_OATAILPTR, gtt_offset & GEN8_OATAILPTR_MASK); /* Mark that we need updated tail pointers to read from... */ - stream->oa_buffer.aging_tail = INVALID_TAIL_PTR; stream->oa_buffer.tail = gtt_offset; /* @@ -1828,7 +1814,6 @@ static void gen12_init_oa_buffer(struct i915_perf_stream *stream) gtt_offset & GEN12_OAG_OATAILPTR_MASK); /* Mark that we need updated tail pointers to read from... */ - stream->oa_buffer.aging_tail = INVALID_TAIL_PTR; stream->oa_buffer.tail = gtt_offset; /* diff --git a/drivers/gpu/drm/i915/i915_perf_types.h b/drivers/gpu/drm/i915/i915_perf_types.h index 66dd5f74de056c2456beae503c8205ea4d3b3231..fe3a5dae8c22809f80d3d0aebb7b9d02c218bbca 100644 --- a/drivers/gpu/drm/i915/i915_perf_types.h +++ b/drivers/gpu/drm/i915/i915_perf_types.h @@ -312,18 +312,6 @@ struct i915_perf_stream { */ spinlock_t ptr_lock; - /** - * @aging_tail: The last HW tail reported by HW. The data - * might not have made it to memory yet though. - */ - u32 aging_tail; - - /** - * @aging_timestamp: A monotonic timestamp for when the current aging tail pointer - * was read; used to determine when it is old enough to trust. - */ - u64 aging_timestamp; - /** * @head: Although we can always read back the head pointer register, * we prefer to avoid trusting the HW state, just to avoid any diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c index a814583e19fd726684b5f79f293ae6d3939b821d..d35973b411863b0de152024f09a01893f2fd8d49 100644 --- a/drivers/gpu/drm/i915/i915_pmu.c +++ b/drivers/gpu/drm/i915/i915_pmu.c @@ -132,14 +132,14 @@ static u32 frequency_enabled_mask(void) unsigned int i; u32 mask = 0; - for (i = 0; i < I915_PMU_MAX_GTS; i++) + for (i = 0; i < I915_PMU_MAX_GT; i++) mask |= config_mask(__I915_PMU_ACTUAL_FREQUENCY(i)) | config_mask(__I915_PMU_REQUESTED_FREQUENCY(i)); return mask; } -static bool pmu_needs_timer(struct i915_pmu *pmu, bool gpu_active) +static bool pmu_needs_timer(struct i915_pmu *pmu) { struct drm_i915_private *i915 = container_of(pmu, typeof(*i915), pmu); u32 enable; @@ -157,17 +157,11 @@ static bool pmu_needs_timer(struct i915_pmu *pmu, bool gpu_active) */ enable &= frequency_enabled_mask() | ENGINE_SAMPLE_MASK; - /* - * When the GPU is idle per-engine counters do not need to be - * running so clear those bits out. - */ - if (!gpu_active) - enable &= ~ENGINE_SAMPLE_MASK; /* * Also there is software busyness tracking available we do not * need the timer for I915_SAMPLE_BUSY counter. */ - else if (i915->caps.scheduler & I915_SCHEDULER_CAP_ENGINE_BUSY_STATS) + if (i915->caps.scheduler & I915_SCHEDULER_CAP_ENGINE_BUSY_STATS) enable &= ~BIT(I915_SAMPLE_BUSY); /* @@ -197,31 +191,21 @@ static inline s64 ktime_since_raw(const ktime_t kt) return ktime_to_ns(ktime_sub(ktime_get_raw(), kt)); } -static unsigned int -__sample_idx(struct i915_pmu *pmu, unsigned int gt_id, int sample) -{ - unsigned int idx = gt_id * __I915_NUM_PMU_SAMPLERS + sample; - - GEM_BUG_ON(idx >= ARRAY_SIZE(pmu->sample)); - - return idx; -} - static u64 read_sample(struct i915_pmu *pmu, unsigned int gt_id, int sample) { - return pmu->sample[__sample_idx(pmu, gt_id, sample)].cur; + return pmu->sample[gt_id][sample].cur; } static void store_sample(struct i915_pmu *pmu, unsigned int gt_id, int sample, u64 val) { - pmu->sample[__sample_idx(pmu, gt_id, sample)].cur = val; + pmu->sample[gt_id][sample].cur = val; } static void add_sample_mult(struct i915_pmu *pmu, unsigned int gt_id, int sample, u32 val, u32 mul) { - pmu->sample[__sample_idx(pmu, gt_id, sample)].cur += mul_u32_u32(val, mul); + pmu->sample[gt_id][sample].cur += mul_u32_u32(val, mul); } static u64 get_rc6(struct intel_gt *gt) @@ -295,7 +279,7 @@ static void park_rc6(struct intel_gt *gt) static void __i915_pmu_maybe_start_timer(struct i915_pmu *pmu) { - if (!pmu->timer_enabled && pmu_needs_timer(pmu, true)) { + if (!pmu->timer_enabled && pmu_needs_timer(pmu)) { pmu->timer_enabled = true; pmu->timer_last = ktime_get(); hrtimer_start_range_ns(&pmu->timer, @@ -321,7 +305,7 @@ void i915_pmu_gt_parked(struct intel_gt *gt) */ pmu->unparked &= ~BIT(gt->info.id); if (pmu->unparked == 0) - pmu->timer_enabled = pmu_needs_timer(pmu, false); + pmu->timer_enabled = false; spin_unlock_irq(&pmu->lock); } @@ -827,7 +811,7 @@ static void i915_pmu_disable(struct perf_event *event) */ if (--pmu->enable_count[bit] == 0) { pmu->enable &= ~BIT(bit); - pmu->timer_enabled &= pmu_needs_timer(pmu, true); + pmu->timer_enabled &= pmu_needs_timer(pmu); } spin_unlock_irqrestore(&pmu->lock, flags); diff --git a/drivers/gpu/drm/i915/i915_pmu.h b/drivers/gpu/drm/i915/i915_pmu.h index 33d80fbaab8bc07a29aa8ecb8df26f3243741893..41af038c373882c785e3d8793acc2b73543fdff5 100644 --- a/drivers/gpu/drm/i915/i915_pmu.h +++ b/drivers/gpu/drm/i915/i915_pmu.h @@ -38,7 +38,7 @@ enum { __I915_NUM_PMU_SAMPLERS }; -#define I915_PMU_MAX_GTS 2 +#define I915_PMU_MAX_GT 2 /* * How many different events we track in the global PMU mask. @@ -47,7 +47,7 @@ enum { */ #define I915_PMU_MASK_BITS \ (I915_ENGINE_SAMPLE_COUNT + \ - I915_PMU_MAX_GTS * __I915_PMU_TRACKED_EVENT_COUNT) + I915_PMU_MAX_GT * __I915_PMU_TRACKED_EVENT_COUNT) #define I915_ENGINE_SAMPLE_COUNT (I915_SAMPLE_SEMA + 1) @@ -127,11 +127,11 @@ struct i915_pmu { * Only global counters are held here, while the per-engine ones are in * struct intel_engine_cs. */ - struct i915_pmu_sample sample[I915_PMU_MAX_GTS * __I915_NUM_PMU_SAMPLERS]; + struct i915_pmu_sample sample[I915_PMU_MAX_GT][__I915_NUM_PMU_SAMPLERS]; /** * @sleep_last: Last time GT parked for RC6 estimation. */ - ktime_t sleep_last[I915_PMU_MAX_GTS]; + ktime_t sleep_last[I915_PMU_MAX_GT]; /** * @irq_count: Number of interrupts * diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index c4197e31962e177e9dabedb40353cc6ca61de245..7a4f462e8b70bc115cbf25cb2378b31e995965ab 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -941,6 +941,9 @@ #define HECI_H_GS1(base) _MMIO((base) + 0xc4c) #define HECI_H_GS1_ER_PREP REG_BIT(0) +#define HECI_FWSTS5(base) _MMIO((base) + 0xc68) +#define HECI_FWSTS5_HUC_AUTH_DONE (1 << 19) + #define HSW_GTT_CACHE_EN _MMIO(0x4024) #define GTT_CACHE_EN_ALL 0xF0007FFF #define GEN7_WR_WATERMARK _MMIO(0x4028) @@ -1343,9 +1346,9 @@ #define SNB_FBC_FRONT_BUFFER REG_BIT(1) #define ILK_DISPLAY_CHICKEN1 _MMIO(0x42000) -#define ILK_FBCQ_DIS (1 << 22) -#define ILK_PABSTRETCH_DIS REG_BIT(21) -#define ILK_SABSTRETCH_DIS REG_BIT(20) +#define ILK_FBCQ_DIS REG_BIT(22) +#define ILK_PABSTRETCH_DIS REG_BIT(21) +#define ILK_SABSTRETCH_DIS REG_BIT(20) #define IVB_PRI_STRETCH_MAX_MASK REG_GENMASK(21, 20) #define IVB_PRI_STRETCH_MAX_X8 REG_FIELD_PREP(IVB_PRI_STRETCH_MAX_MASK, 0) #define IVB_PRI_STRETCH_MAX_X4 REG_FIELD_PREP(IVB_PRI_STRETCH_MAX_MASK, 1) @@ -1809,6 +1812,11 @@ #define CLKGATE_DIS_PSL_EXT(pipe) \ _MMIO_PIPE(pipe, _CLKGATE_DIS_PSL_EXT_A, _CLKGATE_DIS_PSL_EXT_B) +/* DDI Buffer Control */ +#define _DDI_CLK_VALFREQ_A 0x64030 +#define _DDI_CLK_VALFREQ_B 0x64130 +#define DDI_CLK_VALFREQ(port) _MMIO_PORT(port, _DDI_CLK_VALFREQ_A, _DDI_CLK_VALFREQ_B) + /* * Display engine regs */ @@ -1961,15 +1969,6 @@ #define _TRANS_VSYNC_DSI1 0x6b814 #define _TRANS_VSYNCSHIFT_DSI1 0x6b828 -#define TRANSCODER_A_OFFSET 0x60000 -#define TRANSCODER_B_OFFSET 0x61000 -#define TRANSCODER_C_OFFSET 0x62000 -#define CHV_TRANSCODER_C_OFFSET 0x63000 -#define TRANSCODER_D_OFFSET 0x63000 -#define TRANSCODER_EDP_OFFSET 0x6f000 -#define TRANSCODER_DSI0_OFFSET 0x6b000 -#define TRANSCODER_DSI1_OFFSET 0x6b800 - #define TRANS_HTOTAL(trans) _MMIO_TRANS2((trans), _TRANS_HTOTAL_A) #define TRANS_HBLANK(trans) _MMIO_TRANS2((trans), _TRANS_HBLANK_A) #define TRANS_HSYNC(trans) _MMIO_TRANS2((trans), _TRANS_HSYNC_A) @@ -2332,35 +2331,33 @@ /* Panel fitting */ #define PFIT_CONTROL _MMIO(DISPLAY_MMIO_BASE(dev_priv) + 0x61230) -#define PFIT_ENABLE (1 << 31) -#define PFIT_PIPE_MASK (3 << 29) -#define PFIT_PIPE_SHIFT 29 -#define PFIT_PIPE(pipe) ((pipe) << 29) -#define VERT_INTERP_DISABLE (0 << 10) -#define VERT_INTERP_BILINEAR (1 << 10) -#define VERT_INTERP_MASK (3 << 10) -#define VERT_AUTO_SCALE (1 << 9) -#define HORIZ_INTERP_DISABLE (0 << 6) -#define HORIZ_INTERP_BILINEAR (1 << 6) -#define HORIZ_INTERP_MASK (3 << 6) -#define HORIZ_AUTO_SCALE (1 << 5) -#define PANEL_8TO6_DITHER_ENABLE (1 << 3) -#define PFIT_FILTER_FUZZY (0 << 24) -#define PFIT_SCALING_AUTO (0 << 26) -#define PFIT_SCALING_PROGRAMMED (1 << 26) -#define PFIT_SCALING_PILLAR (2 << 26) -#define PFIT_SCALING_LETTER (3 << 26) +#define PFIT_ENABLE REG_BIT(31) +#define PFIT_PIPE_MASK REG_GENMASK(30, 29) /* 965+ */ +#define PFIT_PIPE(pipe) REG_FIELD_PREP(PFIT_PIPE_MASK, (pipe)) +#define PFIT_SCALING_MASK REG_GENMASK(28, 26) /* 965+ */ +#define PFIT_SCALING_AUTO REG_FIELD_PREP(PFIT_SCALING_MASK, 0) +#define PFIT_SCALING_PROGRAMMED REG_FIELD_PREP(PFIT_SCALING_MASK, 1) +#define PFIT_SCALING_PILLAR REG_FIELD_PREP(PFIT_SCALING_MASK, 2) +#define PFIT_SCALING_LETTER REG_FIELD_PREP(PFIT_SCALING_MASK, 3) +#define PFIT_FILTER_MASK REG_GENMASK(25, 24) /* 965+ */ +#define PFIT_FILTER_FUZZY REG_FIELD_PREP(PFIT_FILTER_MASK, 0) +#define PFIT_FILTER_CRISP REG_FIELD_PREP(PFIT_FILTER_MASK, 1) +#define PFIT_FILTER_MEDIAN REG_FIELD_PREP(PFIT_FILTER_MASK, 2) +#define PFIT_VERT_INTERP_MASK REG_GENMASK(11, 10) /* pre-965 */ +#define PFIT_VERT_INTERP_BILINEAR REG_FIELD_PREP(PFIT_VERT_INTERP_MASK, 1) +#define PFIT_VERT_AUTO_SCALE REG_BIT(9) /* pre-965 */ +#define PFIT_HORIZ_INTERP_MASK REG_GENMASK(7, 6) /* pre-965 */ +#define PFIT_HORIZ_INTERP_BILINEAR REG_FIELD_PREP(PFIT_HORIZ_INTERP_MASK, 1) +#define PFIT_HORIZ_AUTO_SCALE REG_BIT(5) /* pre-965 */ +#define PFIT_PANEL_8TO6_DITHER_ENABLE REG_BIT(3) /* pre-965 */ + #define PFIT_PGM_RATIOS _MMIO(DISPLAY_MMIO_BASE(dev_priv) + 0x61234) -/* Pre-965 */ -#define PFIT_VERT_SCALE_SHIFT 20 -#define PFIT_VERT_SCALE_MASK 0xfff00000 -#define PFIT_HORIZ_SCALE_SHIFT 4 -#define PFIT_HORIZ_SCALE_MASK 0x0000fff0 -/* 965+ */ -#define PFIT_VERT_SCALE_SHIFT_965 16 -#define PFIT_VERT_SCALE_MASK_965 0x1fff0000 -#define PFIT_HORIZ_SCALE_SHIFT_965 0 -#define PFIT_HORIZ_SCALE_MASK_965 0x00001fff +#define PFIT_VERT_SCALE_MASK REG_GENMASK(31, 20) /* pre-965 */ +#define PFIT_VERT_SCALE(x) REG_FIELD_PREP(PFIT_VERT_SCALE_MASK, (x)) +#define PFIT_HORIZ_SCALE_MASK REG_GENMASK(15, 4) /* pre-965 */ +#define PFIT_HORIZ_SCALE(x) REG_FIELD_PREP(PFIT_HORIZ_SCALE_MASK, (x)) +#define PFIT_VERT_SCALE_MASK_965 REG_GENMASK(28, 16) /* 965+ */ +#define PFIT_HORIZ_SCALE_MASK_965 REG_GENMASK(12, 0) /* 965+ */ #define PFIT_AUTO_RATIOS _MMIO(DISPLAY_MMIO_BASE(dev_priv) + 0x61238) @@ -2550,6 +2547,7 @@ #define TRANSCONF_MSA_TIMING_DELAY_MASK REG_GENMASK(19, 18) /* ilk/snb/ivb */ #define TRANSCONF_MSA_TIMING_DELAY(x) REG_FIELD_PREP(TRANSCONF_MSA_TIMING_DELAY_MASK, (x)) #define TRANSCONF_CXSR_DOWNCLOCK REG_BIT(16) +#define TRANSCONF_WGC_ENABLE REG_BIT(15) /* vlv/chv only */ #define TRANSCONF_REFRESH_RATE_ALT_VLV REG_BIT(14) #define TRANSCONF_COLOR_RANGE_SELECT REG_BIT(13) #define TRANSCONF_OUTPUT_COLORSPACE_MASK REG_GENMASK(12, 11) /* ilk-ivb */ @@ -2619,23 +2617,6 @@ #define PIPESTAT_INT_ENABLE_MASK 0x7fff0000 #define PIPESTAT_INT_STATUS_MASK 0x0000ffff -#define PIPE_A_OFFSET 0x70000 -#define PIPE_B_OFFSET 0x71000 -#define PIPE_C_OFFSET 0x72000 -#define PIPE_D_OFFSET 0x73000 -#define CHV_PIPE_C_OFFSET 0x74000 -/* - * There's actually no pipe EDP. Some pipe registers have - * simply shifted from the pipe to the transcoder, while - * keeping their original offset. Thus we need PIPE_EDP_OFFSET - * to access such registers in transcoder EDP. - */ -#define PIPE_EDP_OFFSET 0x7f000 - -/* ICL DSI 0 and 1 */ -#define PIPE_DSI0_OFFSET 0x7b000 -#define PIPE_DSI1_OFFSET 0x7b800 - #define TRANSCONF(trans) _MMIO_PIPE2((trans), _TRANSACONF) #define PIPEDSL(pipe) _MMIO_PIPE2(pipe, _PIPEADSL) #define PIPEFRAME(pipe) _MMIO_PIPE2(pipe, _PIPEAFRAMEHIGH) @@ -2655,8 +2636,13 @@ #define PIPE_MISC_YUV420_ENABLE REG_BIT(27) /* glk+ */ #define PIPE_MISC_YUV420_MODE_FULL_BLEND REG_BIT(26) /* glk+ */ #define PIPE_MISC_HDR_MODE_PRECISION REG_BIT(23) /* icl+ */ +#define PIPE_MISC_PSR_MASK_PRIMARY_FLIP REG_BIT(23) /* bdw */ +#define PIPE_MISC_PSR_MASK_SPRITE_ENABLE REG_BIT(22) /* bdw */ +#define PIPE_MISC_PSR_MASK_PIPE_REG_WRITE REG_BIT(21) /* skl+ */ +#define PIPE_MISC_PSR_MASK_CURSOR_MOVE REG_BIT(21) /* bdw */ +#define PIPE_MISC_PSR_MASK_VBLANK_VSYNC_INT REG_BIT(20) #define PIPE_MISC_OUTPUT_COLORSPACE_YUV REG_BIT(11) -#define PIPE_MISC_PIXEL_ROUNDING_TRUNC REG_BIT(8) /* tgl+ */ +#define PIPE_MISC_PIXEL_ROUNDING_TRUNC REG_BIT(8) /* tgl+ */ /* * For Display < 13, Bits 5-7 of PIPE MISC represent DITHER BPC with * valid values of: 6, 8, 10 BPC. @@ -3091,13 +3077,6 @@ #define CUR_CHICKEN(pipe) _MMIO_CURSOR2(pipe, _CUR_CHICKEN_A) #define CURSURFLIVE(pipe) _MMIO_CURSOR2(pipe, _CURASURFLIVE) -#define CURSOR_A_OFFSET 0x70080 -#define CURSOR_B_OFFSET 0x700c0 -#define CHV_CURSOR_C_OFFSET 0x700e0 -#define IVB_CURSOR_B_OFFSET 0x71080 -#define IVB_CURSOR_C_OFFSET 0x72080 -#define TGL_CURSOR_D_OFFSET 0x73080 - /* Display A control */ #define _DSPAADDR_VLV 0x7017C /* vlv/chv */ #define _DSPACNTR 0x70180 @@ -4005,20 +3984,28 @@ /* CPU panel fitter */ /* IVB+ has 3 fitters, 0 is 7x5 capable, the other two only 3x3 */ -#define _PFA_CTL_1 0x68080 -#define _PFB_CTL_1 0x68880 -#define PF_ENABLE (1 << 31) -#define PF_PIPE_SEL_MASK_IVB (3 << 29) -#define PF_PIPE_SEL_IVB(pipe) ((pipe) << 29) -#define PF_FILTER_MASK (3 << 23) -#define PF_FILTER_PROGRAMMED (0 << 23) -#define PF_FILTER_MED_3x3 (1 << 23) -#define PF_FILTER_EDGE_ENHANCE (2 << 23) -#define PF_FILTER_EDGE_SOFTEN (3 << 23) +#define _PFA_CTL_1 0x68080 +#define _PFB_CTL_1 0x68880 +#define PF_ENABLE REG_BIT(31) +#define PF_PIPE_SEL_MASK_IVB REG_GENMASK(30, 29) /* ivb/hsw */ +#define PF_PIPE_SEL_IVB(pipe) REG_FIELD_PREP(PF_PIPE_SEL_MASK_IVB, (pipe)) +#define PF_FILTER_MASK REG_GENMASK(24, 23) +#define PF_FILTER_PROGRAMMED REG_FIELD_PREP(PF_FILTER_MASK, 0) +#define PF_FILTER_MED_3x3 REG_FIELD_PREP(PF_FILTER_MASK, 1) +#define PF_FILTER_EDGE_ENHANCE REG_FIELD_PREP(PF_FILTER_EDGE_MASK, 2) +#define PF_FILTER_EDGE_SOFTEN REG_FIELD_PREP(PF_FILTER_EDGE_MASK, 3) #define _PFA_WIN_SZ 0x68074 #define _PFB_WIN_SZ 0x68874 +#define PF_WIN_XSIZE_MASK REG_GENMASK(31, 16) +#define PF_WIN_XSIZE(w) REG_FIELD_PREP(PF_WIN_XSIZE_MASK, (w)) +#define PF_WIN_YSIZE_MASK REG_GENMASK(15, 0) +#define PF_WIN_YSIZE(h) REG_FIELD_PREP(PF_WIN_YSIZE_MASK, (h)) #define _PFA_WIN_POS 0x68070 #define _PFB_WIN_POS 0x68870 +#define PF_WIN_XPOS_MASK REG_GENMASK(31, 16) +#define PF_WIN_XPOS(x) REG_FIELD_PREP(PF_WIN_XPOS_MASK, (x)) +#define PF_WIN_YPOS_MASK REG_GENMASK(15, 0) +#define PF_WIN_YPOS(y) REG_FIELD_PREP(PF_WIN_YPOS_MASK, (y)) #define _PFA_VSCALE 0x68084 #define _PFB_VSCALE 0x68884 #define _PFA_HSCALE 0x68090 @@ -4030,18 +4017,6 @@ #define PF_VSCALE(pipe) _MMIO_PIPE(pipe, _PFA_VSCALE, _PFB_VSCALE) #define PF_HSCALE(pipe) _MMIO_PIPE(pipe, _PFA_HSCALE, _PFB_HSCALE) -#define _PSA_CTL 0x68180 -#define _PSB_CTL 0x68980 -#define PS_ENABLE (1 << 31) -#define _PSA_WIN_SZ 0x68174 -#define _PSB_WIN_SZ 0x68974 -#define _PSA_WIN_POS 0x68170 -#define _PSB_WIN_POS 0x68970 - -#define PS_CTL(pipe) _MMIO_PIPE(pipe, _PSA_CTL, _PSB_CTL) -#define PS_WIN_SZ(pipe) _MMIO_PIPE(pipe, _PSA_WIN_SZ, _PSB_WIN_SZ) -#define PS_WIN_POS(pipe) _MMIO_PIPE(pipe, _PSA_WIN_POS, _PSB_WIN_POS) - /* * Skylake scalers */ @@ -4050,63 +4025,89 @@ #define _PS_1B_CTRL 0x68980 #define _PS_2B_CTRL 0x68A80 #define _PS_1C_CTRL 0x69180 -#define PS_SCALER_EN (1 << 31) -#define SKL_PS_SCALER_MODE_MASK (3 << 28) -#define SKL_PS_SCALER_MODE_DYN (0 << 28) -#define SKL_PS_SCALER_MODE_HQ (1 << 28) -#define SKL_PS_SCALER_MODE_NV12 (2 << 28) -#define PS_SCALER_MODE_PLANAR (1 << 29) -#define PS_SCALER_MODE_NORMAL (0 << 29) -#define PS_PLANE_SEL_MASK (7 << 25) -#define PS_PLANE_SEL(plane) (((plane) + 1) << 25) -#define PS_FILTER_MASK (3 << 23) -#define PS_FILTER_MEDIUM (0 << 23) -#define PS_FILTER_PROGRAMMED (1 << 23) -#define PS_FILTER_EDGE_ENHANCE (2 << 23) -#define PS_FILTER_BILINEAR (3 << 23) -#define PS_VERT3TAP (1 << 21) -#define PS_VERT_INT_INVERT_FIELD1 (0 << 20) -#define PS_VERT_INT_INVERT_FIELD0 (1 << 20) -#define PS_PWRUP_PROGRESS (1 << 17) -#define PS_V_FILTER_BYPASS (1 << 8) -#define PS_VADAPT_EN (1 << 7) -#define PS_VADAPT_MODE_MASK (3 << 5) -#define PS_VADAPT_MODE_LEAST_ADAPT (0 << 5) -#define PS_VADAPT_MODE_MOD_ADAPT (1 << 5) -#define PS_VADAPT_MODE_MOST_ADAPT (3 << 5) -#define PS_PLANE_Y_SEL_MASK (7 << 5) -#define PS_PLANE_Y_SEL(plane) (((plane) + 1) << 5) -#define PS_Y_VERT_FILTER_SELECT(set) ((set) << 4) -#define PS_Y_HORZ_FILTER_SELECT(set) ((set) << 3) -#define PS_UV_VERT_FILTER_SELECT(set) ((set) << 2) -#define PS_UV_HORZ_FILTER_SELECT(set) ((set) << 1) +#define PS_SCALER_EN REG_BIT(31) +#define PS_SCALER_TYPE_MASK REG_BIT(30) /* icl+ */ +#define PS_SCALER_TYPE_NON_LINEAR REG_FIELD_PREP(PS_SCALER_TYPE_MASK, 0) +#define PS_SCALER_TYPE_LINEAR REG_FIELD_PREP(PS_SCALER_TYPE_MASK, 1) +#define SKL_PS_SCALER_MODE_MASK REG_GENMASK(29, 28) /* skl/bxt */ +#define SKL_PS_SCALER_MODE_DYN REG_FIELD_PREP(SKL_PS_SCALER_MODE_MASK, 0) +#define SKL_PS_SCALER_MODE_HQ REG_FIELD_PREP(SKL_PS_SCALER_MODE_MASK, 1) +#define SKL_PS_SCALER_MODE_NV12 REG_FIELD_PREP(SKL_PS_SCALER_MODE_MASK, 2) +#define PS_SCALER_MODE_MASK REG_BIT(29) /* glk-tgl */ +#define PS_SCALER_MODE_NORMAL REG_FIELD_PREP(PS_SCALER_MODE_MASK, 0) +#define PS_SCALER_MODE_PLANAR REG_FIELD_PREP(PS_SCALER_MODE_MASK, 1) +#define PS_ADAPTIVE_FILTERING_EN REG_BIT(28) /* icl+ */ +#define PS_BINDING_MASK REG_GENMASK(27, 25) +#define PS_BINDING_PIPE REG_FIELD_PREP(PS_BINDING_MASK, 0) +#define PS_BINDING_PLANE(plane_id) REG_FIELD_PREP(PS_BINDING_MASK, (plane_id) + 1) +#define PS_FILTER_MASK REG_GENMASK(24, 23) +#define PS_FILTER_MEDIUM REG_FIELD_PREP(PS_FILTER_MASK, 0) +#define PS_FILTER_PROGRAMMED REG_FIELD_PREP(PS_FILTER_MASK, 1) +#define PS_FILTER_EDGE_ENHANCE REG_FIELD_PREP(PS_FILTER_MASK, 2) +#define PS_FILTER_BILINEAR REG_FIELD_PREP(PS_FILTER_MASK, 3) +#define PS_ADAPTIVE_FILTER_MASK REG_BIT(22) /* icl+ */ +#define PS_ADAPTIVE_FILTER_MEDIUM REG_FIELD_PREP(PS_ADAPTIVE_FILTER_MASK, 0) +#define PS_ADAPTIVE_FILTER_EDGE_ENHANCE REG_FIELD_PREP(PS_ADAPTIVE_FILTER_MASK, 1) +#define PS_PIPE_SCALER_LOC_MASK REG_BIT(21) /* icl+ */ +#define PS_PIPE_SCALER_LOC_AFTER_OUTPUT_CSC REG_FIELD_PREP(PS_SCALER_LOCATION_MASK, 0) /* non-linear */ +#define PS_PIPE_SCALER_LOC_AFTER_CSC REG_FIELD_PREP(PS_SCALER_LOCATION_MASK, 1) /* linear */ +#define PS_VERT3TAP REG_BIT(21) /* skl/bxt */ +#define PS_VERT_INT_INVERT_FIELD REG_BIT(20) +#define PS_PROG_SCALE_FACTOR REG_BIT(19) /* tgl+ */ +#define PS_PWRUP_PROGRESS REG_BIT(17) +#define PS_V_FILTER_BYPASS REG_BIT(8) +#define PS_VADAPT_EN REG_BIT(7) /* skl/bxt */ +#define PS_VADAPT_MODE_MASK REG_GENMASK(6, 5) /* skl/bxt */ +#define PS_VADAPT_MODE_LEAST_ADAPT REG_FIELD_PREP(PS_VADAPT_MODE_MASK, 0) +#define PS_VADAPT_MODE_MOD_ADAPT REG_FIELD_PREP(PS_VADAPT_MODE_MASK, 1) +#define PS_VADAPT_MODE_MOST_ADAPT REG_FIELD_PREP(PS_VADAPT_MODE_MASK, 3) +#define PS_BINDING_Y_MASK REG_GENMASK(7, 5) /* icl-tgl */ +#define PS_BINDING_Y_PLANE(plane_id) REG_FIELD_PREP(PS_BINDING_Y_MASK, (plane_id) + 1) +#define PS_Y_VERT_FILTER_SELECT_MASK REG_BIT(4) /* glk+ */ +#define PS_Y_VERT_FILTER_SELECT(set) REG_FIELD_PREP(PS_Y_VERT_FILTER_SELECT_MASK, (set)) +#define PS_Y_HORZ_FILTER_SELECT_MASK REG_BIT(3) /* glk+ */ +#define PS_Y_HORZ_FILTER_SELECT(set) REG_FIELD_PREP(PS_Y_HORZ_FILTER_SELECT_MASK, (set)) +#define PS_UV_VERT_FILTER_SELECT_MASK REG_BIT(2) /* glk+ */ +#define PS_UV_VERT_FILTER_SELECT(set) REG_FIELD_PREP(PS_UV_VERT_FILTER_SELECT_MASK, (set)) +#define PS_UV_HORZ_FILTER_SELECT_MASK REG_BIT(1) /* glk+ */ +#define PS_UV_HORZ_FILTER_SELECT(set) REG_FIELD_PREP(PS_UV_HORZ_FILTER_SELECT_MASK, (set)) #define _PS_PWR_GATE_1A 0x68160 #define _PS_PWR_GATE_2A 0x68260 #define _PS_PWR_GATE_1B 0x68960 #define _PS_PWR_GATE_2B 0x68A60 #define _PS_PWR_GATE_1C 0x69160 -#define PS_PWR_GATE_DIS_OVERRIDE (1 << 31) -#define PS_PWR_GATE_SETTLING_TIME_32 (0 << 3) -#define PS_PWR_GATE_SETTLING_TIME_64 (1 << 3) -#define PS_PWR_GATE_SETTLING_TIME_96 (2 << 3) -#define PS_PWR_GATE_SETTLING_TIME_128 (3 << 3) -#define PS_PWR_GATE_SLPEN_8 0 -#define PS_PWR_GATE_SLPEN_16 1 -#define PS_PWR_GATE_SLPEN_24 2 -#define PS_PWR_GATE_SLPEN_32 3 +#define PS_PWR_GATE_DIS_OVERRIDE REG_BIT(31) +#define PS_PWR_GATE_SETTLING_TIME_MASK REG_GENMASK(4, 3) +#define PS_PWR_GATE_SETTLING_TIME_32 REG_FIELD_PREP(PS_PWR_GATE_SETTLING_TIME_MASK, 0) +#define PS_PWR_GATE_SETTLING_TIME_64 REG_FIELD_PREP(PS_PWR_GATE_SETTLING_TIME_MASK, 1) +#define PS_PWR_GATE_SETTLING_TIME_96 REG_FIELD_PREP(PS_PWR_GATE_SETTLING_TIME_MASK, 2) +#define PS_PWR_GATE_SETTLING_TIME_128 REG_FIELD_PREP(PS_PWR_GATE_SETTLING_TIME_MASK, 3) +#define PS_PWR_GATE_SLPEN_MASK REG_GENMASK(1, 0) +#define PS_PWR_GATE_SLPEN_8 REG_FIELD_PREP(PS_PWR_GATE_SLPEN_MASK, 0) +#define PS_PWR_GATE_SLPEN_16 REG_FIELD_PREP(PS_PWR_GATE_SLPEN_MASK, 1) +#define PS_PWR_GATE_SLPEN_24 REG_FIELD_PREP(PS_PWR_GATE_SLPEN_MASK, 2) +#define PS_PWR_GATE_SLPEN_32 REG_FIELD_PREP(PS_PWR_GATE_SLPEN_MASK, 3) #define _PS_WIN_POS_1A 0x68170 #define _PS_WIN_POS_2A 0x68270 #define _PS_WIN_POS_1B 0x68970 #define _PS_WIN_POS_2B 0x68A70 #define _PS_WIN_POS_1C 0x69170 +#define PS_WIN_XPOS_MASK REG_GENMASK(31, 16) +#define PS_WIN_XPOS(x) REG_FIELD_PREP(PS_WIN_XPOS_MASK, (x)) +#define PS_WIN_YPOS_MASK REG_GENMASK(15, 0) +#define PS_WIN_YPOS(y) REG_FIELD_PREP(PS_WIN_YPOS_MASK, (y)) #define _PS_WIN_SZ_1A 0x68174 #define _PS_WIN_SZ_2A 0x68274 #define _PS_WIN_SZ_1B 0x68974 #define _PS_WIN_SZ_2B 0x68A74 #define _PS_WIN_SZ_1C 0x69174 +#define PS_WIN_XSIZE_MASK REG_GENMASK(31, 16) +#define PS_WIN_XSIZE(w) REG_FIELD_PREP(PS_WIN_XSIZE_MASK, (w)) +#define PS_WIN_YSIZE_MASK REG_GENMASK(15, 0) +#define PS_WIN_YSIZE(h) REG_FIELD_PREP(PS_WIN_YSIZE_MASK, (h)) #define _PS_VSCALE_1A 0x68184 #define _PS_VSCALE_2A 0x68284 @@ -4125,10 +4126,12 @@ #define _PS_VPHASE_1B 0x68988 #define _PS_VPHASE_2B 0x68A88 #define _PS_VPHASE_1C 0x69188 -#define PS_Y_PHASE(x) ((x) << 16) -#define PS_UV_RGB_PHASE(x) ((x) << 0) -#define PS_PHASE_MASK (0x7fff << 1) /* u2.13 */ -#define PS_PHASE_TRIP (1 << 0) +#define PS_Y_PHASE_MASK REG_GENMASK(31, 16) +#define PS_Y_PHASE(x) REG_FIELD_PREP(PS_Y_PHASE_MASK, (x)) +#define PS_UV_RGB_PHASE_MASK REG_GENMASK(15, 0) +#define PS_UV_RGB_PHASE(x) REG_FIELD_PREP(PS_UV_RGB_PHASE_MASK, (x)) +#define PS_PHASE_MASK (0x7fff << 1) /* u2.13 */ +#define PS_PHASE_TRIP (1 << 0) #define _PS_HPHASE_1A 0x68194 #define _PS_HPHASE_2A 0x68294 @@ -4146,7 +4149,7 @@ #define _PS_COEF_SET0_INDEX_2A 0x68298 #define _PS_COEF_SET0_INDEX_1B 0x68998 #define _PS_COEF_SET0_INDEX_2B 0x68A98 -#define PS_COEE_INDEX_AUTO_INC (1 << 10) +#define PS_COEF_INDEX_AUTO_INC REG_BIT(10) #define _PS_COEF_SET0_DATA_1A 0x6819C #define _PS_COEF_SET0_DATA_2A 0x6829C @@ -4418,8 +4421,10 @@ #define GEN8_DE_MISC_IMR _MMIO(0x44464) #define GEN8_DE_MISC_IIR _MMIO(0x44468) #define GEN8_DE_MISC_IER _MMIO(0x4446c) -#define GEN8_DE_MISC_GSE (1 << 27) -#define GEN8_DE_EDP_PSR (1 << 19) +#define XELPDP_PMDEMAND_RSPTOUT_ERR REG_BIT(27) +#define GEN8_DE_MISC_GSE REG_BIT(27) +#define GEN8_DE_EDP_PSR REG_BIT(19) +#define XELPDP_PMDEMAND_RSP REG_BIT(3) #define GEN8_PCU_ISR _MMIO(0x444e0) #define GEN8_PCU_IMR _MMIO(0x444e4) @@ -4482,56 +4487,96 @@ #define GEN11_HOTPLUG_CTL_SHORT_DETECT(hpd_pin) (1 << (_HPD_PIN_TC(hpd_pin) * 4)) #define GEN11_HOTPLUG_CTL_NO_DETECT(hpd_pin) (0 << (_HPD_PIN_TC(hpd_pin) * 4)) +#define PICAINTERRUPT_ISR _MMIO(0x16FE50) +#define PICAINTERRUPT_IMR _MMIO(0x16FE54) +#define PICAINTERRUPT_IIR _MMIO(0x16FE58) +#define PICAINTERRUPT_IER _MMIO(0x16FE5C) + +#define XELPDP_DP_ALT_HOTPLUG(hpd_pin) REG_BIT(16 + _HPD_PIN_TC(hpd_pin)) +#define XELPDP_DP_ALT_HOTPLUG_MASK REG_GENMASK(19, 16) + +#define XELPDP_AUX_TC(hpd_pin) REG_BIT(8 + _HPD_PIN_TC(hpd_pin)) +#define XELPDP_AUX_TC_MASK REG_GENMASK(11, 8) + +#define XELPDP_TBT_HOTPLUG(hpd_pin) REG_BIT(_HPD_PIN_TC(hpd_pin)) +#define XELPDP_TBT_HOTPLUG_MASK REG_GENMASK(3, 0) + +#define XELPDP_PORT_HOTPLUG_CTL(hpd_pin) _MMIO(0x16F270 + (_HPD_PIN_TC(hpd_pin) * 0x200)) +#define XELPDP_TBT_HOTPLUG_ENABLE REG_BIT(6) +#define XELPDP_TBT_HPD_LONG_DETECT REG_BIT(5) +#define XELPDP_TBT_HPD_SHORT_DETECT REG_BIT(4) +#define XELPDP_DP_ALT_HOTPLUG_ENABLE REG_BIT(2) +#define XELPDP_DP_ALT_HPD_LONG_DETECT REG_BIT(1) +#define XELPDP_DP_ALT_HPD_SHORT_DETECT REG_BIT(0) + +#define XELPDP_INITIATE_PMDEMAND_REQUEST(dword) _MMIO(0x45230 + 4 * (dword)) +#define XELPDP_PMDEMAND_QCLK_GV_BW_MASK REG_GENMASK(31, 16) +#define XELPDP_PMDEMAND_VOLTAGE_INDEX_MASK REG_GENMASK(14, 12) +#define XELPDP_PMDEMAND_QCLK_GV_INDEX_MASK REG_GENMASK(11, 8) +#define XELPDP_PMDEMAND_PIPES_MASK REG_GENMASK(7, 6) +#define XELPDP_PMDEMAND_DBUFS_MASK REG_GENMASK(5, 4) +#define XELPDP_PMDEMAND_PHYS_MASK REG_GENMASK(2, 0) + +#define XELPDP_PMDEMAND_REQ_ENABLE REG_BIT(31) +#define XELPDP_PMDEMAND_CDCLK_FREQ_MASK REG_GENMASK(30, 20) +#define XELPDP_PMDEMAND_DDICLK_FREQ_MASK REG_GENMASK(18, 8) +#define XELPDP_PMDEMAND_SCALERS_MASK REG_GENMASK(6, 4) +#define XELPDP_PMDEMAND_PLLS_MASK REG_GENMASK(2, 0) + +#define GEN12_DCPR_STATUS_1 _MMIO(0x46440) +#define XELPDP_PMDEMAND_INFLIGHT_STATUS REG_BIT(26) + #define ILK_DISPLAY_CHICKEN2 _MMIO(0x42004) /* Required on all Ironlake and Sandybridge according to the B-Spec. */ -#define ILK_ELPIN_409_SELECT (1 << 25) -#define ILK_DPARB_GATE (1 << 22) -#define ILK_VSDPFD_FULL (1 << 21) -#define FUSE_STRAP _MMIO(0x42014) -#define ILK_INTERNAL_GRAPHICS_DISABLE (1 << 31) -#define ILK_INTERNAL_DISPLAY_DISABLE (1 << 30) -#define ILK_DISPLAY_DEBUG_DISABLE (1 << 29) -#define IVB_PIPE_C_DISABLE (1 << 28) -#define ILK_HDCP_DISABLE (1 << 25) -#define ILK_eDP_A_DISABLE (1 << 24) -#define HSW_CDCLK_LIMIT (1 << 24) -#define ILK_DESKTOP (1 << 23) -#define HSW_CPU_SSC_ENABLE (1 << 21) - -#define FUSE_STRAP3 _MMIO(0x42020) -#define HSW_REF_CLK_SELECT (1 << 1) - -#define ILK_DSPCLK_GATE_D _MMIO(0x42020) -#define ILK_VRHUNIT_CLOCK_GATE_DISABLE (1 << 28) -#define ILK_DPFCUNIT_CLOCK_GATE_DISABLE (1 << 9) -#define ILK_DPFCRUNIT_CLOCK_GATE_DISABLE (1 << 8) -#define ILK_DPFDUNIT_CLOCK_GATE_ENABLE (1 << 7) -#define ILK_DPARBUNIT_CLOCK_GATE_ENABLE (1 << 5) - -#define IVB_CHICKEN3 _MMIO(0x4200c) -# define CHICKEN3_DGMG_REQ_OUT_FIX_DISABLE (1 << 5) -# define CHICKEN3_DGMG_DONE_FIX_DISABLE (1 << 2) - -#define CHICKEN_PAR1_1 _MMIO(0x42080) -#define IGNORE_KVMR_PIPE_A REG_BIT(23) -#define KBL_ARB_FILL_SPARE_22 REG_BIT(22) -#define DIS_RAM_BYPASS_PSR2_MAN_TRACK (1 << 16) -#define SKL_DE_COMPRESSED_HASH_MODE (1 << 15) -#define DPA_MASK_VBLANK_SRD (1 << 15) -#define FORCE_ARB_IDLE_PLANES (1 << 14) -#define SKL_EDP_PSR_FIX_RDWRAP (1 << 3) -#define IGNORE_PSR2_HW_TRACKING (1 << 1) +#define ILK_ELPIN_409_SELECT REG_BIT(25) +#define ILK_DPARB_GATE REG_BIT(22) +#define ILK_VSDPFD_FULL REG_BIT(21) + +#define FUSE_STRAP _MMIO(0x42014) +#define ILK_INTERNAL_GRAPHICS_DISABLE REG_BIT(31) +#define ILK_INTERNAL_DISPLAY_DISABLE REG_BIT(30) +#define ILK_DISPLAY_DEBUG_DISABLE REG_BIT(29) +#define IVB_PIPE_C_DISABLE REG_BIT(28) +#define ILK_HDCP_DISABLE REG_BIT(25) +#define ILK_eDP_A_DISABLE REG_BIT(24) +#define HSW_CDCLK_LIMIT REG_BIT(24) +#define ILK_DESKTOP REG_BIT(23) +#define HSW_CPU_SSC_ENABLE REG_BIT(21) + +#define FUSE_STRAP3 _MMIO(0x42020) +#define HSW_REF_CLK_SELECT REG_BIT(1) + +#define ILK_DSPCLK_GATE_D _MMIO(0x42020) +#define ILK_VRHUNIT_CLOCK_GATE_DISABLE REG_BIT(28) +#define ILK_DPFCUNIT_CLOCK_GATE_DISABLE REG_BIT(9) +#define ILK_DPFCRUNIT_CLOCK_GATE_DISABLE REG_BIT(8) +#define ILK_DPFDUNIT_CLOCK_GATE_ENABLE REG_BIT(7) +#define ILK_DPARBUNIT_CLOCK_GATE_ENABLE REG_BIT(5) + +#define IVB_CHICKEN3 _MMIO(0x4200c) +#define CHICKEN3_DGMG_REQ_OUT_FIX_DISABLE REG_BIT(5) +#define CHICKEN3_DGMG_DONE_FIX_DISABLE REG_BIT(2) + +#define CHICKEN_PAR1_1 _MMIO(0x42080) +#define IGNORE_KVMR_PIPE_A REG_BIT(23) +#define KBL_ARB_FILL_SPARE_22 REG_BIT(22) +#define DIS_RAM_BYPASS_PSR2_MAN_TRACK REG_BIT(16) +#define SKL_DE_COMPRESSED_HASH_MODE REG_BIT(15) +#define HSW_MASK_VBL_TO_PIPE_IN_SRD REG_BIT(15) /* hsw/bdw */ +#define FORCE_ARB_IDLE_PLANES REG_BIT(14) +#define SKL_EDP_PSR_FIX_RDWRAP REG_BIT(3) +#define IGNORE_PSR2_HW_TRACKING REG_BIT(1) #define CHICKEN_PAR2_1 _MMIO(0x42090) -#define KVM_CONFIG_CHANGE_NOTIFICATION_SELECT (1 << 14) +#define KVM_CONFIG_CHANGE_NOTIFICATION_SELECT REG_BIT(14) #define CHICKEN_MISC_2 _MMIO(0x42084) -#define CHICKEN_MISC_DISABLE_DPT REG_BIT(30) /* adl,dg2 */ -#define KBL_ARB_FILL_SPARE_14 REG_BIT(14) -#define KBL_ARB_FILL_SPARE_13 REG_BIT(13) -#define GLK_CL2_PWR_DOWN (1 << 12) -#define GLK_CL1_PWR_DOWN (1 << 11) -#define GLK_CL0_PWR_DOWN (1 << 10) +#define CHICKEN_MISC_DISABLE_DPT REG_BIT(30) /* adl,dg2 */ +#define KBL_ARB_FILL_SPARE_14 REG_BIT(14) +#define KBL_ARB_FILL_SPARE_13 REG_BIT(13) +#define GLK_CL2_PWR_DOWN REG_BIT(12) +#define GLK_CL1_PWR_DOWN REG_BIT(11) +#define GLK_CL0_PWR_DOWN REG_BIT(10) #define CHICKEN_MISC_4 _MMIO(0x4208c) #define CHICKEN_FBC_STRIDE_OVERRIDE REG_BIT(13) @@ -4540,24 +4585,26 @@ #define _CHICKEN_PIPESL_1_A 0x420b0 #define _CHICKEN_PIPESL_1_B 0x420b4 -#define HSW_PRI_STRETCH_MAX_MASK REG_GENMASK(28, 27) -#define HSW_PRI_STRETCH_MAX_X8 REG_FIELD_PREP(HSW_PRI_STRETCH_MAX_MASK, 0) -#define HSW_PRI_STRETCH_MAX_X4 REG_FIELD_PREP(HSW_PRI_STRETCH_MAX_MASK, 1) -#define HSW_PRI_STRETCH_MAX_X2 REG_FIELD_PREP(HSW_PRI_STRETCH_MAX_MASK, 2) -#define HSW_PRI_STRETCH_MAX_X1 REG_FIELD_PREP(HSW_PRI_STRETCH_MAX_MASK, 3) -#define HSW_SPR_STRETCH_MAX_MASK REG_GENMASK(26, 25) -#define HSW_SPR_STRETCH_MAX_X8 REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 0) -#define HSW_SPR_STRETCH_MAX_X4 REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 1) -#define HSW_SPR_STRETCH_MAX_X2 REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 2) -#define HSW_SPR_STRETCH_MAX_X1 REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 3) -#define HSW_FBCQ_DIS (1 << 22) -#define BDW_DPRS_MASK_VBLANK_SRD (1 << 0) -#define SKL_PLANE1_STRETCH_MAX_MASK REG_GENMASK(1, 0) -#define SKL_PLANE1_STRETCH_MAX_X8 REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 0) -#define SKL_PLANE1_STRETCH_MAX_X4 REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 1) -#define SKL_PLANE1_STRETCH_MAX_X2 REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 2) -#define SKL_PLANE1_STRETCH_MAX_X1 REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 3) -#define CHICKEN_PIPESL_1(pipe) _MMIO_PIPE(pipe, _CHICKEN_PIPESL_1_A, _CHICKEN_PIPESL_1_B) +#define CHICKEN_PIPESL_1(pipe) _MMIO_PIPE(pipe, _CHICKEN_PIPESL_1_A, _CHICKEN_PIPESL_1_B) +#define HSW_PRI_STRETCH_MAX_MASK REG_GENMASK(28, 27) +#define HSW_PRI_STRETCH_MAX_X8 REG_FIELD_PREP(HSW_PRI_STRETCH_MAX_MASK, 0) +#define HSW_PRI_STRETCH_MAX_X4 REG_FIELD_PREP(HSW_PRI_STRETCH_MAX_MASK, 1) +#define HSW_PRI_STRETCH_MAX_X2 REG_FIELD_PREP(HSW_PRI_STRETCH_MAX_MASK, 2) +#define HSW_PRI_STRETCH_MAX_X1 REG_FIELD_PREP(HSW_PRI_STRETCH_MAX_MASK, 3) +#define HSW_SPR_STRETCH_MAX_MASK REG_GENMASK(26, 25) +#define HSW_SPR_STRETCH_MAX_X8 REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 0) +#define HSW_SPR_STRETCH_MAX_X4 REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 1) +#define HSW_SPR_STRETCH_MAX_X2 REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 2) +#define HSW_SPR_STRETCH_MAX_X1 REG_FIELD_PREP(HSW_SPR_STRETCH_MAX_MASK, 3) +#define HSW_FBCQ_DIS REG_BIT(22) +#define HSW_UNMASK_VBL_TO_REGS_IN_SRD REG_BIT(15) /* hsw */ +#define SKL_PSR_MASK_PLANE_FLIP REG_BIT(11) /* skl+ */ +#define SKL_PLANE1_STRETCH_MAX_MASK REG_GENMASK(1, 0) +#define SKL_PLANE1_STRETCH_MAX_X8 REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 0) +#define SKL_PLANE1_STRETCH_MAX_X4 REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 1) +#define SKL_PLANE1_STRETCH_MAX_X2 REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 2) +#define SKL_PLANE1_STRETCH_MAX_X1 REG_FIELD_PREP(SKL_PLANE1_STRETCH_MAX_MASK, 3) +#define BDW_UNMASK_VBL_TO_REGS_IN_SRD REG_BIT(0) /* bdw */ #define _CHICKEN_TRANS_A 0x420c0 #define _CHICKEN_TRANS_B 0x420c4 @@ -4570,32 +4617,33 @@ [TRANSCODER_B] = _CHICKEN_TRANS_B, \ [TRANSCODER_C] = _CHICKEN_TRANS_C, \ [TRANSCODER_D] = _CHICKEN_TRANS_D)) - #define _MTL_CHICKEN_TRANS_A 0x604e0 #define _MTL_CHICKEN_TRANS_B 0x614e0 #define MTL_CHICKEN_TRANS(trans) _MMIO_TRANS((trans), \ _MTL_CHICKEN_TRANS_A, \ _MTL_CHICKEN_TRANS_B) - -#define HSW_FRAME_START_DELAY_MASK REG_GENMASK(28, 27) -#define HSW_FRAME_START_DELAY(x) REG_FIELD_PREP(HSW_FRAME_START_DELAY_MASK, x) -#define VSC_DATA_SEL_SOFTWARE_CONTROL REG_BIT(25) /* GLK */ -#define FECSTALL_DIS_DPTSTREAM_DPTTG REG_BIT(23) -#define DDI_TRAINING_OVERRIDE_ENABLE REG_BIT(19) -#define ADLP_1_BASED_X_GRANULARITY REG_BIT(18) -#define DDI_TRAINING_OVERRIDE_VALUE REG_BIT(18) -#define DDIE_TRAINING_OVERRIDE_ENABLE REG_BIT(17) /* CHICKEN_TRANS_A only */ -#define DDIE_TRAINING_OVERRIDE_VALUE REG_BIT(16) /* CHICKEN_TRANS_A only */ -#define PSR2_ADD_VERTICAL_LINE_COUNT REG_BIT(15) -#define PSR2_VSC_ENABLE_PROG_HEADER REG_BIT(12) +#define PIPE_VBLANK_WITH_DELAY REG_BIT(31) /* ADL/DG2 */ +#define SKL_UNMASK_VBL_TO_PIPE_IN_SRD REG_BIT(30) /* skl+ */ +#define HSW_FRAME_START_DELAY_MASK REG_GENMASK(28, 27) +#define HSW_FRAME_START_DELAY(x) REG_FIELD_PREP(HSW_FRAME_START_DELAY_MASK, x) +#define VSC_DATA_SEL_SOFTWARE_CONTROL REG_BIT(25) /* GLK */ +#define FECSTALL_DIS_DPTSTREAM_DPTTG REG_BIT(23) +#define DDI_TRAINING_OVERRIDE_ENABLE REG_BIT(19) +#define ADLP_1_BASED_X_GRANULARITY REG_BIT(18) +#define DDI_TRAINING_OVERRIDE_VALUE REG_BIT(18) +#define DDIE_TRAINING_OVERRIDE_ENABLE REG_BIT(17) /* CHICKEN_TRANS_A only */ +#define DDIE_TRAINING_OVERRIDE_VALUE REG_BIT(16) /* CHICKEN_TRANS_A only */ +#define PSR2_ADD_VERTICAL_LINE_COUNT REG_BIT(15) +#define PSR2_VSC_ENABLE_PROG_HEADER REG_BIT(12) #define DISP_ARB_CTL _MMIO(0x45000) -#define DISP_FBC_MEMORY_WAKE (1 << 31) -#define DISP_TILE_SURFACE_SWIZZLING (1 << 13) -#define DISP_FBC_WM_DIS (1 << 15) +#define DISP_FBC_MEMORY_WAKE REG_BIT(31) +#define DISP_TILE_SURFACE_SWIZZLING REG_BIT(13) +#define DISP_FBC_WM_DIS REG_BIT(15) + #define DISP_ARB_CTL2 _MMIO(0x45004) -#define DISP_DATA_PARTITION_5_6 (1 << 6) -#define DISP_IPC_ENABLE (1 << 3) +#define DISP_DATA_PARTITION_5_6 REG_BIT(6) +#define DISP_IPC_ENABLE REG_BIT(3) #define GEN7_MSG_CTL _MMIO(0x45010) #define WAIT_FOR_PCH_RESET_ACK (1 << 1) @@ -4637,6 +4685,9 @@ #define DCPR_SEND_RESP_IMM REG_BIT(25) #define DCPR_CLEAR_MEMSTAT_DIS REG_BIT(24) +#define XELPD_CHICKEN_DCPR_3 _MMIO(0x46438) +#define DMD_RSP_TIMEOUT_DISABLE REG_BIT(19) + #define SKL_DFSM _MMIO(0x51000) #define SKL_DFSM_DISPLAY_PM_DISABLE (1 << 27) #define SKL_DFSM_DISPLAY_HDCP_DISABLE (1 << 25) @@ -4768,7 +4819,8 @@ SDE_FDI_RXB_CPT | \ SDE_FDI_RXA_CPT) -/* south display engine interrupt: ICP/TGP */ +/* south display engine interrupt: ICP/TGP/MTP */ +#define SDE_PICAINTERRUPT REG_BIT(31) #define SDE_GMBUS_ICP (1 << 23) #define SDE_TC_HOTPLUG_ICP(hpd_pin) REG_BIT(24 + _HPD_PIN_TC(hpd_pin)) #define SDE_TC_HOTPLUG_DG2(hpd_pin) REG_BIT(25 + _HPD_PIN_TC(hpd_pin)) /* sigh */ @@ -5104,24 +5156,32 @@ #define TRANS_BPC_10 REG_FIELD_PREP(TRANS_BPC_MASK, 1) #define TRANS_BPC_6 REG_FIELD_PREP(TRANS_BPC_MASK, 2) #define TRANS_BPC_12 REG_FIELD_PREP(TRANS_BPC_MASK, 3) + #define _TRANSA_CHICKEN1 0xf0060 #define _TRANSB_CHICKEN1 0xf1060 #define TRANS_CHICKEN1(pipe) _MMIO_PIPE(pipe, _TRANSA_CHICKEN1, _TRANSB_CHICKEN1) -#define TRANS_CHICKEN1_HDMIUNIT_GC_DISABLE (1 << 10) -#define TRANS_CHICKEN1_DP0UNIT_GC_DISABLE (1 << 4) +#define TRANS_CHICKEN1_HDMIUNIT_GC_DISABLE REG_BIT(10) +#define TRANS_CHICKEN1_DP0UNIT_GC_DISABLE REG_BIT(4) + #define _TRANSA_CHICKEN2 0xf0064 #define _TRANSB_CHICKEN2 0xf1064 #define TRANS_CHICKEN2(pipe) _MMIO_PIPE(pipe, _TRANSA_CHICKEN2, _TRANSB_CHICKEN2) -#define TRANS_CHICKEN2_TIMING_OVERRIDE (1 << 31) -#define TRANS_CHICKEN2_FDI_POLARITY_REVERSED (1 << 29) -#define TRANS_CHICKEN2_FRAME_START_DELAY_MASK (3 << 27) -#define TRANS_CHICKEN2_FRAME_START_DELAY(x) ((x) << 27) /* 0-3 */ -#define TRANS_CHICKEN2_DISABLE_DEEP_COLOR_COUNTER (1 << 26) -#define TRANS_CHICKEN2_DISABLE_DEEP_COLOR_MODESWITCH (1 << 25) +#define TRANS_CHICKEN2_TIMING_OVERRIDE REG_BIT(31) +#define TRANS_CHICKEN2_FDI_POLARITY_REVERSED REG_BIT(29) +#define TRANS_CHICKEN2_FRAME_START_DELAY_MASK REG_GENMASK(28, 27) +#define TRANS_CHICKEN2_FRAME_START_DELAY(x) REG_FIELD_PREP(TRANS_CHICKEN2_FRAME_START_DELAY_MASK, (x)) /* 0-3 */ +#define TRANS_CHICKEN2_DISABLE_DEEP_COLOR_COUNTER REG_BIT(26) +#define TRANS_CHICKEN2_DISABLE_DEEP_COLOR_MODESWITCH REG_BIT(25) #define SOUTH_CHICKEN1 _MMIO(0xc2000) #define FDIA_PHASE_SYNC_SHIFT_OVR 19 #define FDIA_PHASE_SYNC_SHIFT_EN 18 +#define INVERT_DDIE_HPD REG_BIT(28) +#define INVERT_DDID_HPD_MTP REG_BIT(27) +#define INVERT_TC4_HPD REG_BIT(26) +#define INVERT_TC3_HPD REG_BIT(25) +#define INVERT_TC2_HPD REG_BIT(24) +#define INVERT_TC1_HPD REG_BIT(23) #define INVERT_DDID_HPD (1 << 18) #define INVERT_DDIC_HPD (1 << 17) #define INVERT_DDIB_HPD (1 << 16) @@ -5303,6 +5363,20 @@ #define ICL_PCODE_MEM_SS_READ_GLOBAL_INFO (0x0 << 8) #define ICL_PCODE_MEM_SS_READ_QGV_POINT_INFO(point) (((point) << 16) | (0x1 << 8)) #define ADL_PCODE_MEM_SS_READ_PSF_GV_INFO ((0) | (0x2 << 8)) +#define DISPLAY_TO_PCODE_CDCLK_MAX 0x28D +#define DISPLAY_TO_PCODE_VOLTAGE_MASK REG_GENMASK(1, 0) +#define DISPLAY_TO_PCODE_VOLTAGE_MAX DISPLAY_TO_PCODE_VOLTAGE_MASK +#define DISPLAY_TO_PCODE_CDCLK_VALID REG_BIT(27) +#define DISPLAY_TO_PCODE_PIPE_COUNT_VALID REG_BIT(31) +#define DISPLAY_TO_PCODE_CDCLK_MASK REG_GENMASK(25, 16) +#define DISPLAY_TO_PCODE_PIPE_COUNT_MASK REG_GENMASK(30, 28) +#define DISPLAY_TO_PCODE_CDCLK(x) REG_FIELD_PREP(DISPLAY_TO_PCODE_CDCLK_MASK, (x)) +#define DISPLAY_TO_PCODE_PIPE_COUNT(x) REG_FIELD_PREP(DISPLAY_TO_PCODE_PIPE_COUNT_MASK, (x)) +#define DISPLAY_TO_PCODE_VOLTAGE(x) REG_FIELD_PREP(DISPLAY_TO_PCODE_VOLTAGE_MASK, (x)) +#define DISPLAY_TO_PCODE_UPDATE_MASK(cdclk, num_pipes, voltage_level) \ + ((DISPLAY_TO_PCODE_CDCLK(cdclk)) | \ + (DISPLAY_TO_PCODE_PIPE_COUNT(num_pipes)) | \ + (DISPLAY_TO_PCODE_VOLTAGE(voltage_level))) #define ICL_PCODE_SAGV_DE_MEM_SS_CONFIG 0xe #define ICL_PCODE_REP_QGV_MASK REG_GENMASK(1, 0) #define ICL_PCODE_REP_QGV_SAFE REG_FIELD_PREP(ICL_PCODE_REP_QGV_MASK, 0) @@ -5563,6 +5637,8 @@ enum skl_power_gate { #define TRANS_DDI_HDCP_SELECT REG_BIT(5) #define TRANS_DDI_BFI_ENABLE (1 << 4) #define TRANS_DDI_HIGH_TMDS_CHAR_RATE (1 << 4) +#define TRANS_DDI_PORT_WIDTH_MASK REG_GENMASK(3, 1) +#define TRANS_DDI_PORT_WIDTH(width) REG_FIELD_PREP(TRANS_DDI_PORT_WIDTH_MASK, (width) - 1) #define TRANS_DDI_HDMI_SCRAMBLING (1 << 0) #define TRANS_DDI_HDMI_SCRAMBLING_MASK (TRANS_DDI_HDMI_SCRAMBLER_CTS_ENABLE \ | TRANS_DDI_HDMI_SCRAMBLER_RESET_FREQ \ @@ -5622,11 +5698,16 @@ enum skl_power_gate { /* DDI Buffer Control */ #define _DDI_BUF_CTL_A 0x64000 #define _DDI_BUF_CTL_B 0x64100 +/* Known as DDI_CTL_DE in MTL+ */ #define DDI_BUF_CTL(port) _MMIO_PORT(port, _DDI_BUF_CTL_A, _DDI_BUF_CTL_B) #define DDI_BUF_CTL_ENABLE (1 << 31) #define DDI_BUF_TRANS_SELECT(n) ((n) << 24) #define DDI_BUF_EMP_MASK (0xf << 24) #define DDI_BUF_PHY_LINK_RATE(r) ((r) << 20) +#define DDI_BUF_PORT_DATA_MASK REG_GENMASK(19, 18) +#define DDI_BUF_PORT_DATA_10BIT REG_FIELD_PREP(DDI_BUF_PORT_DATA_MASK, 0) +#define DDI_BUF_PORT_DATA_20BIT REG_FIELD_PREP(DDI_BUF_PORT_DATA_MASK, 1) +#define DDI_BUF_PORT_DATA_40BIT REG_FIELD_PREP(DDI_BUF_PORT_DATA_MASK, 2) #define DDI_BUF_PORT_REVERSAL (1 << 16) #define DDI_BUF_IS_IDLE (1 << 7) #define DDI_BUF_CTL_TC_PHY_OWNERSHIP REG_BIT(6) @@ -6383,6 +6464,20 @@ enum skl_power_gate { (index) * 4, _PLANE_CSC_POSTOFF_HI_2(pipe) + \ (index) * 4) +#define _PIPE_A_WGC_C01_C00 0x600B0 /* s2.10 */ +#define _PIPE_A_WGC_C02 0x600B4 /* s2.10 */ +#define _PIPE_A_WGC_C11_C10 0x600B8 /* s2.10 */ +#define _PIPE_A_WGC_C12 0x600BC /* s2.10 */ +#define _PIPE_A_WGC_C21_C20 0x600C0 /* s2.10 */ +#define _PIPE_A_WGC_C22 0x600C4 /* s2.10 */ + +#define PIPE_WGC_C01_C00(pipe) _MMIO_TRANS2(pipe, _PIPE_A_WGC_C01_C00) +#define PIPE_WGC_C02(pipe) _MMIO_TRANS2(pipe, _PIPE_A_WGC_C02) +#define PIPE_WGC_C11_C10(pipe) _MMIO_TRANS2(pipe, _PIPE_A_WGC_C11_C10) +#define PIPE_WGC_C12(pipe) _MMIO_TRANS2(pipe, _PIPE_A_WGC_C12) +#define PIPE_WGC_C21_C20(pipe) _MMIO_TRANS2(pipe, _PIPE_A_WGC_C21_C20) +#define PIPE_WGC_C22(pipe) _MMIO_TRANS2(pipe, _PIPE_A_WGC_C22) + /* pipe CSC & degamma/gamma LUTs on CHV */ #define _CGM_PIPE_A_CSC_COEFF01 (VLV_DISPLAY_BASE + 0x67900) #define _CGM_PIPE_A_CSC_COEFF23 (VLV_DISPLAY_BASE + 0x67904) diff --git a/drivers/gpu/drm/i915/i915_reg_defs.h b/drivers/gpu/drm/i915/i915_reg_defs.h index db26de6b57bc8d967f7c1a1645f683280527c195..a685db1e815d804543be36ca826d4ab24d008667 100644 --- a/drivers/gpu/drm/i915/i915_reg_defs.h +++ b/drivers/gpu/drm/i915/i915_reg_defs.h @@ -22,6 +22,19 @@ BUILD_BUG_ON_ZERO(__is_constexpr(__n) && \ ((__n) < 0 || (__n) > 31)))) +/** + * REG_BIT8() - Prepare a u8 bit value + * @__n: 0-based bit number + * + * Local wrapper for BIT() to force u8, with compile time checks. + * + * @return: Value with bit @__n set. + */ +#define REG_BIT8(__n) \ + ((u8)(BIT(__n) + \ + BUILD_BUG_ON_ZERO(__is_constexpr(__n) && \ + ((__n) < 0 || (__n) > 7)))) + /** * REG_GENMASK() - Prepare a continuous u32 bitmask * @__high: 0-based high bit @@ -52,6 +65,21 @@ __is_constexpr(__low) && \ ((__low) < 0 || (__high) > 63 || (__low) > (__high))))) +/** + * REG_GENMASK8() - Prepare a continuous u8 bitmask + * @__high: 0-based high bit + * @__low: 0-based low bit + * + * Local wrapper for GENMASK() to force u8, with compile time checks. + * + * @return: Continuous bitmask from @__high to @__low, inclusive. + */ +#define REG_GENMASK8(__high, __low) \ + ((u8)(GENMASK(__high, __low) + \ + BUILD_BUG_ON_ZERO(__is_constexpr(__high) && \ + __is_constexpr(__low) && \ + ((__low) < 0 || (__high) > 7 || (__low) > (__high))))) + /* * Local integer constant expression version of is_power_of_2(). */ @@ -74,6 +102,23 @@ BUILD_BUG_ON_ZERO(!IS_POWER_OF_2((__mask) + (1ULL << __bf_shf(__mask)))) + \ BUILD_BUG_ON_ZERO(__builtin_choose_expr(__is_constexpr(__val), (~((__mask) >> __bf_shf(__mask)) & (__val)), 0)))) +/** + * REG_FIELD_PREP8() - Prepare a u8 bitfield value + * @__mask: shifted mask defining the field's length and position + * @__val: value to put in the field + * + * Local copy of FIELD_PREP() to generate an integer constant expression, force + * u8 and for consistency with REG_FIELD_GET8(), REG_BIT8() and REG_GENMASK8(). + * + * @return: @__val masked and shifted into the field defined by @__mask. + */ +#define REG_FIELD_PREP8(__mask, __val) \ + ((u8)((((typeof(__mask))(__val) << __bf_shf(__mask)) & (__mask)) + \ + BUILD_BUG_ON_ZERO(!__is_constexpr(__mask)) + \ + BUILD_BUG_ON_ZERO((__mask) == 0 || (__mask) > U8_MAX) + \ + BUILD_BUG_ON_ZERO(!IS_POWER_OF_2((__mask) + (1ULL << __bf_shf(__mask)))) + \ + BUILD_BUG_ON_ZERO(__builtin_choose_expr(__is_constexpr(__val), (~((__mask) >> __bf_shf(__mask)) & (__val)), 0)))) + /** * REG_FIELD_GET() - Extract a u32 bitfield value * @__mask: shifted mask defining the field's length and position @@ -98,6 +143,54 @@ */ #define REG_FIELD_GET64(__mask, __val) ((u64)FIELD_GET(__mask, __val)) +/** + * REG_BIT16() - Prepare a u16 bit value + * @__n: 0-based bit number + * + * Local wrapper for BIT() to force u16, with compile time + * checks. + * + * @return: Value with bit @__n set. + */ +#define REG_BIT16(__n) \ + ((u16)(BIT(__n) + \ + BUILD_BUG_ON_ZERO(__is_constexpr(__n) && \ + ((__n) < 0 || (__n) > 15)))) + +/** + * REG_GENMASK16() - Prepare a continuous u8 bitmask + * @__high: 0-based high bit + * @__low: 0-based low bit + * + * Local wrapper for GENMASK() to force u16, with compile time + * checks. + * + * @return: Continuous bitmask from @__high to @__low, inclusive. + */ +#define REG_GENMASK16(__high, __low) \ + ((u16)(GENMASK(__high, __low) + \ + BUILD_BUG_ON_ZERO(__is_constexpr(__high) && \ + __is_constexpr(__low) && \ + ((__low) < 0 || (__high) > 15 || (__low) > (__high))))) + +/** + * REG_FIELD_PREP16() - Prepare a u16 bitfield value + * @__mask: shifted mask defining the field's length and position + * @__val: value to put in the field + * + * Local copy of FIELD_PREP16() to generate an integer constant + * expression, force u8 and for consistency with + * REG_FIELD_GET16(), REG_BIT16() and REG_GENMASK16(). + * + * @return: @__val masked and shifted into the field defined by @__mask. + */ +#define REG_FIELD_PREP16(__mask, __val) \ + ((u16)((((typeof(__mask))(__val) << __bf_shf(__mask)) & (__mask)) + \ + BUILD_BUG_ON_ZERO(!__is_constexpr(__mask)) + \ + BUILD_BUG_ON_ZERO((__mask) == 0 || (__mask) > U16_MAX) + \ + BUILD_BUG_ON_ZERO(!IS_POWER_OF_2((__mask) + (1ULL << __bf_shf(__mask)))) + \ + BUILD_BUG_ON_ZERO(__builtin_choose_expr(__is_constexpr(__val), (~((__mask) >> __bf_shf(__mask)) & (__val)), 0)))) + #define __MASKED_FIELD(mask, value) ((mask) << 16 | (value)) #define _MASKED_FIELD(mask, value) ({ \ if (__builtin_constant_p(mask)) \ @@ -155,6 +248,18 @@ */ #define _PICK(__index, ...) (((const u32 []){ __VA_ARGS__ })[__index]) +/** + * REG_FIELD_GET8() - Extract a u8 bitfield value + * @__mask: shifted mask defining the field's length and position + * @__val: value to extract the bitfield value from + * + * Local wrapper for FIELD_GET() to force u8 and for consistency with + * REG_FIELD_PREP(), REG_BIT() and REG_GENMASK(). + * + * @return: Masked and shifted value of the field defined by @__mask in @__val. + */ +#define REG_FIELD_GET8(__mask, __val) ((u8)FIELD_GET(__mask, __val)) + typedef struct { u32 reg; } i915_reg_t; diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c index 630a732aaecca8fb70ebe4f54d5f9695a7a8fcee..894068bb37b6f1b647c7b7bb55bba08f4a379e47 100644 --- a/drivers/gpu/drm/i915/i915_request.c +++ b/drivers/gpu/drm/i915/i915_request.c @@ -290,7 +290,7 @@ static enum hrtimer_restart __rq_watchdog_expired(struct hrtimer *hrtimer) if (!i915_request_completed(rq)) { if (llist_add(&rq->watchdog.link, >->watchdog.list)) - schedule_work(>->watchdog.work); + queue_work(gt->i915->unordered_wq, >->watchdog.work); } else { i915_request_put(rq); } diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c index a814775a363df416c73fdef5c8783a386d55312d..ffb425ba591c7935649d80da8d6387090aab9e22 100644 --- a/drivers/gpu/drm/i915/i915_vma.c +++ b/drivers/gpu/drm/i915/i915_vma.c @@ -1710,6 +1710,8 @@ static void release_references(struct i915_vma *vma, struct intel_gt *gt, if (vm_ddestroy) i915_vm_resv_put(vma->vm); + /* Wait for async active retire */ + i915_active_wait(&vma->active); i915_active_fini(&vma->active); GEM_WARN_ON(vma->resource); i915_vma_free(vma); diff --git a/drivers/gpu/drm/i915/intel_clock_gating.c b/drivers/gpu/drm/i915/intel_clock_gating.c index 2c5302bcba19c310153b38b28e179dcc897cb343..a27600bc597663b06ee3d549371de37c71896080 100644 --- a/drivers/gpu/drm/i915/intel_clock_gating.c +++ b/drivers/gpu/drm/i915/intel_clock_gating.c @@ -36,6 +36,7 @@ #include "gt/intel_gt_regs.h" #include "i915_drv.h" +#include "i915_reg.h" #include "intel_clock_gating.h" #include "intel_mchbar_regs.h" #include "vlv_sideband.h" @@ -520,12 +521,12 @@ static void bdw_init_clock_gating(struct drm_i915_private *i915) intel_uncore_rmw(&i915->uncore, GAM_ECOCHK, 0, HSW_ECOCHK_ARB_PRIO_SOL); /* WaPsrDPAMaskVBlankInSRD:bdw */ - intel_uncore_rmw(&i915->uncore, CHICKEN_PAR1_1, 0, DPA_MASK_VBLANK_SRD); + intel_uncore_rmw(&i915->uncore, CHICKEN_PAR1_1, 0, HSW_MASK_VBL_TO_PIPE_IN_SRD); for_each_pipe(i915, pipe) { /* WaPsrDPRSUnmaskVBlankInSRD:bdw */ intel_uncore_rmw(&i915->uncore, CHICKEN_PIPESL_1(pipe), - 0, BDW_DPRS_MASK_VBLANK_SRD); + 0, BDW_UNMASK_VBL_TO_REGS_IN_SRD); } /* WaVSRefCountFullforceMissDisable:bdw */ diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c index fc5cd14adfccbc7c3393bd1ebfbe005e84c8566a..6e49caf241a5f54e1bdf7f46669d12c7e3274c6d 100644 --- a/drivers/gpu/drm/i915/intel_device_info.c +++ b/drivers/gpu/drm/i915/intel_device_info.c @@ -27,9 +27,7 @@ #include #include -#include "display/intel_cdclk.h" -#include "display/intel_de.h" -#include "display/intel_display.h" +#include "display/intel_display_device.h" #include "gt/intel_gt_regs.h" #include "i915_drv.h" #include "i915_reg.h" @@ -95,6 +93,9 @@ void intel_device_info_print(const struct intel_device_info *info, const struct intel_runtime_info *runtime, struct drm_printer *p) { + const struct intel_display_runtime_info *display_runtime = + &info->display->__runtime_defaults; + if (runtime->graphics.ip.rel) drm_printf(p, "graphics version: %u.%02u\n", runtime->graphics.ip.ver, @@ -111,13 +112,13 @@ void intel_device_info_print(const struct intel_device_info *info, drm_printf(p, "media version: %u\n", runtime->media.ip.ver); - if (runtime->display.ip.rel) + if (display_runtime->ip.rel) drm_printf(p, "display version: %u.%02u\n", - runtime->display.ip.ver, - runtime->display.ip.rel); + display_runtime->ip.ver, + display_runtime->ip.rel); else drm_printf(p, "display version: %u\n", - runtime->display.ip.ver); + display_runtime->ip.ver); drm_printf(p, "graphics stepping: %s\n", intel_step_name(runtime->step.graphics_step)); drm_printf(p, "media stepping: %s\n", intel_step_name(runtime->step.media_step)); @@ -138,13 +139,13 @@ void intel_device_info_print(const struct intel_device_info *info, drm_printf(p, "has_pooled_eu: %s\n", str_yes_no(runtime->has_pooled_eu)); -#define PRINT_FLAG(name) drm_printf(p, "%s: %s\n", #name, str_yes_no(info->display.name)) +#define PRINT_FLAG(name) drm_printf(p, "%s: %s\n", #name, str_yes_no(info->display->name)) DEV_INFO_DISPLAY_FOR_EACH_FLAG(PRINT_FLAG); #undef PRINT_FLAG - drm_printf(p, "has_hdcp: %s\n", str_yes_no(runtime->has_hdcp)); - drm_printf(p, "has_dmc: %s\n", str_yes_no(runtime->has_dmc)); - drm_printf(p, "has_dsc: %s\n", str_yes_no(runtime->has_dsc)); + drm_printf(p, "has_hdcp: %s\n", str_yes_no(display_runtime->has_hdcp)); + drm_printf(p, "has_dmc: %s\n", str_yes_no(display_runtime->has_dmc)); + drm_printf(p, "has_dsc: %s\n", str_yes_no(display_runtime->has_dsc)); drm_printf(p, "rawclk rate: %u kHz\n", runtime->rawclk_freq); } @@ -362,8 +363,6 @@ static void intel_ipver_early_init(struct drm_i915_private *i915) RUNTIME_INFO(i915)->graphics.ip.ver = 12; RUNTIME_INFO(i915)->graphics.ip.rel = 70; } - ip_ver_read(i915, i915_mmio_reg_offset(GMD_ID_DISPLAY), - &runtime->display.ip); ip_ver_read(i915, i915_mmio_reg_offset(GMD_ID_MEDIA), &runtime->media.ip); } @@ -381,6 +380,15 @@ void intel_device_info_runtime_init_early(struct drm_i915_private *i915) intel_device_info_subplatform_init(i915); } +/* FIXME: Remove this, and make device info a const pointer to rodata. */ +static struct intel_device_info * +mkwrite_device_info(struct drm_i915_private *i915) +{ + return (struct intel_device_info *)INTEL_INFO(i915); +} + +static const struct intel_display_device_info no_display = {}; + /** * intel_device_info_runtime_init - initialize runtime info * @dev_priv: the i915 device @@ -401,123 +409,23 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv) { struct intel_device_info *info = mkwrite_device_info(dev_priv); struct intel_runtime_info *runtime = RUNTIME_INFO(dev_priv); - enum pipe pipe; - - /* Wa_14011765242: adl-s A0,A1 */ - if (IS_ADLS_DISPLAY_STEP(dev_priv, STEP_A0, STEP_A2)) - for_each_pipe(dev_priv, pipe) - runtime->num_scalers[pipe] = 0; - else if (DISPLAY_VER(dev_priv) >= 11) { - for_each_pipe(dev_priv, pipe) - runtime->num_scalers[pipe] = 2; - } else if (DISPLAY_VER(dev_priv) >= 9) { - runtime->num_scalers[PIPE_A] = 2; - runtime->num_scalers[PIPE_B] = 2; - runtime->num_scalers[PIPE_C] = 1; - } - - BUILD_BUG_ON(BITS_PER_TYPE(intel_engine_mask_t) < I915_NUM_ENGINES); - - if (DISPLAY_VER(dev_priv) >= 13 || HAS_D12_PLANE_MINIMIZATION(dev_priv)) - for_each_pipe(dev_priv, pipe) - runtime->num_sprites[pipe] = 4; - else if (DISPLAY_VER(dev_priv) >= 11) - for_each_pipe(dev_priv, pipe) - runtime->num_sprites[pipe] = 6; - else if (DISPLAY_VER(dev_priv) == 10) - for_each_pipe(dev_priv, pipe) - runtime->num_sprites[pipe] = 3; - else if (IS_BROXTON(dev_priv)) { - /* - * Skylake and Broxton currently don't expose the topmost plane as its - * use is exclusive with the legacy cursor and we only want to expose - * one of those, not both. Until we can safely expose the topmost plane - * as a DRM_PLANE_TYPE_CURSOR with all the features exposed/supported, - * we don't expose the topmost plane at all to prevent ABI breakage - * down the line. - */ - - runtime->num_sprites[PIPE_A] = 2; - runtime->num_sprites[PIPE_B] = 2; - runtime->num_sprites[PIPE_C] = 1; - } else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) { - for_each_pipe(dev_priv, pipe) - runtime->num_sprites[pipe] = 2; - } else if (DISPLAY_VER(dev_priv) >= 5 || IS_G4X(dev_priv)) { - for_each_pipe(dev_priv, pipe) - runtime->num_sprites[pipe] = 1; - } - if (HAS_DISPLAY(dev_priv) && - (IS_DGFX(dev_priv) || DISPLAY_VER(dev_priv) >= 14) && - !(intel_de_read(dev_priv, GU_CNTL_PROTECTED) & DEPRESENT)) { - drm_info(&dev_priv->drm, "Display not present, disabling\n"); + if (HAS_DISPLAY(dev_priv)) + intel_display_device_info_runtime_init(dev_priv); - runtime->pipe_mask = 0; + /* Display may have been disabled by runtime init */ + if (!HAS_DISPLAY(dev_priv)) { + dev_priv->drm.driver_features &= ~(DRIVER_MODESET | + DRIVER_ATOMIC); + info->display = &no_display; } - if (HAS_DISPLAY(dev_priv) && IS_GRAPHICS_VER(dev_priv, 7, 8) && - HAS_PCH_SPLIT(dev_priv)) { - u32 fuse_strap = intel_de_read(dev_priv, FUSE_STRAP); - u32 sfuse_strap = intel_de_read(dev_priv, SFUSE_STRAP); - - /* - * SFUSE_STRAP is supposed to have a bit signalling the display - * is fused off. Unfortunately it seems that, at least in - * certain cases, fused off display means that PCH display - * reads don't land anywhere. In that case, we read 0s. - * - * On CPT/PPT, we can detect this case as SFUSE_STRAP_FUSE_LOCK - * should be set when taking over after the firmware. - */ - if (fuse_strap & ILK_INTERNAL_DISPLAY_DISABLE || - sfuse_strap & SFUSE_STRAP_DISPLAY_DISABLED || - (HAS_PCH_CPT(dev_priv) && - !(sfuse_strap & SFUSE_STRAP_FUSE_LOCK))) { - drm_info(&dev_priv->drm, - "Display fused off, disabling\n"); - runtime->pipe_mask = 0; - } else if (fuse_strap & IVB_PIPE_C_DISABLE) { - drm_info(&dev_priv->drm, "PipeC fused off\n"); - runtime->pipe_mask &= ~BIT(PIPE_C); - runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_C); - } - } else if (HAS_DISPLAY(dev_priv) && DISPLAY_VER(dev_priv) >= 9) { - u32 dfsm = intel_de_read(dev_priv, SKL_DFSM); - - if (dfsm & SKL_DFSM_PIPE_A_DISABLE) { - runtime->pipe_mask &= ~BIT(PIPE_A); - runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_A); - runtime->fbc_mask &= ~BIT(INTEL_FBC_A); - } - if (dfsm & SKL_DFSM_PIPE_B_DISABLE) { - runtime->pipe_mask &= ~BIT(PIPE_B); - runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_B); - } - if (dfsm & SKL_DFSM_PIPE_C_DISABLE) { - runtime->pipe_mask &= ~BIT(PIPE_C); - runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_C); - } - - if (DISPLAY_VER(dev_priv) >= 12 && - (dfsm & TGL_DFSM_PIPE_D_DISABLE)) { - runtime->pipe_mask &= ~BIT(PIPE_D); - runtime->cpu_transcoder_mask &= ~BIT(TRANSCODER_D); - } - - if (dfsm & SKL_DFSM_DISPLAY_HDCP_DISABLE) - runtime->has_hdcp = 0; - - if (dfsm & SKL_DFSM_DISPLAY_PM_DISABLE) - runtime->fbc_mask = 0; - - if (DISPLAY_VER(dev_priv) >= 11 && (dfsm & ICL_DFSM_DMC_DISABLE)) - runtime->has_dmc = 0; + /* Disable nuclear pageflip by default on pre-g4x */ + if (!dev_priv->params.nuclear_pageflip && + DISPLAY_VER(dev_priv) < 5 && !IS_G4X(dev_priv)) + dev_priv->drm.driver_features &= ~DRIVER_ATOMIC; - if (IS_DISPLAY_VER(dev_priv, 10, 12) && - (dfsm & GLK_DFSM_DISPLAY_DSC_DISABLE)) - runtime->has_dsc = 0; - } + BUILD_BUG_ON(BITS_PER_TYPE(intel_engine_mask_t) < I915_NUM_ENGINES); if (GRAPHICS_VER(dev_priv) == 6 && i915_vtd_active(dev_priv)) { drm_info(&dev_priv->drm, @@ -528,24 +436,43 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv) runtime->rawclk_freq = intel_read_rawclk(dev_priv); drm_dbg(&dev_priv->drm, "rawclk rate: %d kHz\n", runtime->rawclk_freq); - if (!HAS_DISPLAY(dev_priv)) { - dev_priv->drm.driver_features &= ~(DRIVER_MODESET | - DRIVER_ATOMIC); - memset(&info->display, 0, sizeof(info->display)); - - runtime->cpu_transcoder_mask = 0; - memset(runtime->num_sprites, 0, sizeof(runtime->num_sprites)); - memset(runtime->num_scalers, 0, sizeof(runtime->num_scalers)); - runtime->fbc_mask = 0; - runtime->has_hdcp = false; - runtime->has_dmc = false; - runtime->has_dsc = false; +} + +/* + * Set up device info and initial runtime info at driver create. + * + * Note: i915 is only an allocated blob of memory at this point. + */ +void intel_device_info_driver_create(struct drm_i915_private *i915, + u16 device_id, + const struct intel_device_info *match_info) +{ + struct intel_device_info *info; + struct intel_runtime_info *runtime; + u16 ver, rel, step; + + /* Setup the write-once "constant" device info */ + info = mkwrite_device_info(i915); + memcpy(info, match_info, sizeof(*info)); + + /* Initialize initial runtime info from static const data and pdev. */ + runtime = RUNTIME_INFO(i915); + memcpy(runtime, &INTEL_INFO(i915)->__runtime, sizeof(*runtime)); + + /* Probe display support */ + info->display = intel_display_device_probe(i915, info->has_gmd_id, + &ver, &rel, &step); + memcpy(DISPLAY_RUNTIME_INFO(i915), + &DISPLAY_INFO(i915)->__runtime_defaults, + sizeof(*DISPLAY_RUNTIME_INFO(i915))); + + if (info->has_gmd_id) { + DISPLAY_RUNTIME_INFO(i915)->ip.ver = ver; + DISPLAY_RUNTIME_INFO(i915)->ip.rel = rel; + DISPLAY_RUNTIME_INFO(i915)->ip.step = step; } - /* Disable nuclear pageflip by default on pre-g4x */ - if (!dev_priv->params.nuclear_pageflip && - DISPLAY_VER(dev_priv) < 5 && !IS_G4X(dev_priv)) - dev_priv->drm.driver_features &= ~DRIVER_ATOMIC; + runtime->device_id = device_id; } void intel_driver_caps_print(const struct intel_driver_caps *caps, diff --git a/drivers/gpu/drm/i915/intel_device_info.h b/drivers/gpu/drm/i915/intel_device_info.h index d737dd601a6e3bb0fe78785c04a7b94048ebe65f..069291b3bd37017110456efbd63607f166f028eb 100644 --- a/drivers/gpu/drm/i915/intel_device_info.h +++ b/drivers/gpu/drm/i915/intel_device_info.h @@ -29,7 +29,7 @@ #include "intel_step.h" -#include "display/intel_display_limits.h" +#include "display/intel_display_device.h" #include "gt/intel_engine_types.h" #include "gt/intel_context_types.h" @@ -182,25 +182,6 @@ enum intel_ppgtt_type { func(unfenced_needs_alignment); \ func(hws_needs_physical); -#define DEV_INFO_DISPLAY_FOR_EACH_FLAG(func) \ - /* Keep in alphabetical order */ \ - func(cursor_needs_physical); \ - func(has_cdclk_crawl); \ - func(has_cdclk_squash); \ - func(has_ddi); \ - func(has_dp_mst); \ - func(has_dsb); \ - func(has_fpga_dbg); \ - func(has_gmch); \ - func(has_hotplug); \ - func(has_hti); \ - func(has_ipc); \ - func(has_overlay); \ - func(has_psr); \ - func(has_psr_hw_tracking); \ - func(overlay_needs_physical); \ - func(supports_tv); - struct intel_ip_version { u8 ver; u8 rel; @@ -218,9 +199,6 @@ struct intel_runtime_info { struct { struct intel_ip_version ip; } media; - struct { - struct intel_ip_version ip; - } display; /* * Platform mask is used for optimizing or-ed IS_PLATFORM calls into @@ -248,21 +226,6 @@ struct intel_runtime_info { u32 memory_regions; /* regions supported by the HW */ bool has_pooled_eu; - - /* display */ - struct { - u8 pipe_mask; - u8 cpu_transcoder_mask; - - u8 num_sprites[I915_MAX_PIPES]; - u8 num_scalers[I915_MAX_PIPES]; - - u8 fbc_mask; - - bool has_hdcp; - bool has_dmc; - bool has_dsc; - }; }; struct intel_device_info { @@ -278,33 +241,7 @@ struct intel_device_info { DEV_INFO_FOR_EACH_FLAG(DEFINE_FLAG); #undef DEFINE_FLAG - struct { - u8 abox_mask; - - struct { - u16 size; /* in blocks */ - u8 slice_mask; - } dbuf; - -#define DEFINE_FLAG(name) u8 name:1 - DEV_INFO_DISPLAY_FOR_EACH_FLAG(DEFINE_FLAG); -#undef DEFINE_FLAG - - /* Global register offset for the display engine */ - u32 mmio_offset; - - /* Register offsets for the various display pipes and transcoders */ - u32 pipe_offsets[I915_MAX_TRANSCODERS]; - u32 trans_offsets[I915_MAX_TRANSCODERS]; - u32 cursor_offsets[I915_MAX_PIPES]; - - struct { - u32 degamma_lut_size; - u32 gamma_lut_size; - u32 degamma_lut_tests; - u32 gamma_lut_tests; - } color; - } display; + const struct intel_display_device_info *display; /* * Initial runtime info. Do not access outside of i915_driver_create(). @@ -322,6 +259,8 @@ struct intel_driver_caps { const char *intel_platform_name(enum intel_platform platform); +void intel_device_info_driver_create(struct drm_i915_private *i915, u16 device_id, + const struct intel_device_info *match_info); void intel_device_info_runtime_init_early(struct drm_i915_private *dev_priv); void intel_device_info_runtime_init(struct drm_i915_private *dev_priv); diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c index cf5122299b6b8c36a85bee5ab5af3dd6e85a1fb6..6d8e5e5c0cba28b9ed23a212128b367e7111b370 100644 --- a/drivers/gpu/drm/i915/intel_runtime_pm.c +++ b/drivers/gpu/drm/i915/intel_runtime_pm.c @@ -658,5 +658,5 @@ void intel_runtime_pm_init_early(struct intel_runtime_pm *rpm) init_intel_runtime_pm_wakeref(rpm); INIT_LIST_HEAD(&rpm->lmem_userfault_list); spin_lock_init(&rpm->lmem_userfault_lock); - intel_wakeref_auto_init(&rpm->userfault_wakeref, rpm); + intel_wakeref_auto_init(&rpm->userfault_wakeref, i915); } diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.h b/drivers/gpu/drm/i915/intel_runtime_pm.h index e592e8d6499a1fac71714877a771ec9483d88324..764b183ae45293b363b0874cd00abc9ffb73f118 100644 --- a/drivers/gpu/drm/i915/intel_runtime_pm.h +++ b/drivers/gpu/drm/i915/intel_runtime_pm.h @@ -16,12 +16,6 @@ struct device; struct drm_i915_private; struct drm_printer; -enum i915_drm_suspend_mode { - I915_DRM_SUSPEND_IDLE, - I915_DRM_SUSPEND_MEM, - I915_DRM_SUSPEND_HIBERNATE, -}; - /* * This struct helps tracking the state needed for runtime PM, which puts the * device in PCI D3 state. Notice that when this happens, nothing on the diff --git a/drivers/gpu/drm/i915/intel_step.c b/drivers/gpu/drm/i915/intel_step.c index 84a6fe736a3b527185cbe3df54eb35e8f797f4b9..8a9ff6227e5367105d2ad50f66d31455e883a532 100644 --- a/drivers/gpu/drm/i915/intel_step.c +++ b/drivers/gpu/drm/i915/intel_step.c @@ -166,8 +166,12 @@ void intel_step_init(struct drm_i915_private *i915) &RUNTIME_INFO(i915)->graphics.ip); step.media_step = gmd_to_intel_step(i915, &RUNTIME_INFO(i915)->media.ip); - step.display_step = gmd_to_intel_step(i915, - &RUNTIME_INFO(i915)->display.ip); + step.display_step = STEP_A0 + DISPLAY_RUNTIME_INFO(i915)->ip.step; + if (step.display_step >= STEP_FUTURE) { + drm_dbg(&i915->drm, "Using future display steppings\n"); + step.display_step = STEP_FUTURE; + } + RUNTIME_INFO(i915)->step = step; return; diff --git a/drivers/gpu/drm/i915/intel_wakeref.c b/drivers/gpu/drm/i915/intel_wakeref.c index dfd87d082218078cb98ae368e812dec337f43cb3..718f2f1b6174d1ba3e943ca6fb3fdf4ad7f9611f 100644 --- a/drivers/gpu/drm/i915/intel_wakeref.c +++ b/drivers/gpu/drm/i915/intel_wakeref.c @@ -8,17 +8,18 @@ #include "intel_runtime_pm.h" #include "intel_wakeref.h" +#include "i915_drv.h" static void rpm_get(struct intel_wakeref *wf) { - wf->wakeref = intel_runtime_pm_get(wf->rpm); + wf->wakeref = intel_runtime_pm_get(&wf->i915->runtime_pm); } static void rpm_put(struct intel_wakeref *wf) { intel_wakeref_t wakeref = fetch_and_zero(&wf->wakeref); - intel_runtime_pm_put(wf->rpm, wakeref); + intel_runtime_pm_put(&wf->i915->runtime_pm, wakeref); INTEL_WAKEREF_BUG_ON(!wakeref); } @@ -74,7 +75,7 @@ void __intel_wakeref_put_last(struct intel_wakeref *wf, unsigned long flags) /* Assume we are not in process context and so cannot sleep. */ if (flags & INTEL_WAKEREF_PUT_ASYNC || !mutex_trylock(&wf->mutex)) { - mod_delayed_work(system_wq, &wf->work, + mod_delayed_work(wf->i915->unordered_wq, &wf->work, FIELD_GET(INTEL_WAKEREF_PUT_DELAY, flags)); return; } @@ -94,11 +95,11 @@ static void __intel_wakeref_put_work(struct work_struct *wrk) } void __intel_wakeref_init(struct intel_wakeref *wf, - struct intel_runtime_pm *rpm, + struct drm_i915_private *i915, const struct intel_wakeref_ops *ops, struct intel_wakeref_lockclass *key) { - wf->rpm = rpm; + wf->i915 = i915; wf->ops = ops; __mutex_init(&wf->mutex, "wakeref.mutex", &key->mutex); @@ -137,17 +138,17 @@ static void wakeref_auto_timeout(struct timer_list *t) wakeref = fetch_and_zero(&wf->wakeref); spin_unlock_irqrestore(&wf->lock, flags); - intel_runtime_pm_put(wf->rpm, wakeref); + intel_runtime_pm_put(&wf->i915->runtime_pm, wakeref); } void intel_wakeref_auto_init(struct intel_wakeref_auto *wf, - struct intel_runtime_pm *rpm) + struct drm_i915_private *i915) { spin_lock_init(&wf->lock); timer_setup(&wf->timer, wakeref_auto_timeout, 0); refcount_set(&wf->count, 0); wf->wakeref = 0; - wf->rpm = rpm; + wf->i915 = i915; } void intel_wakeref_auto(struct intel_wakeref_auto *wf, unsigned long timeout) @@ -161,13 +162,14 @@ void intel_wakeref_auto(struct intel_wakeref_auto *wf, unsigned long timeout) } /* Our mission is that we only extend an already active wakeref */ - assert_rpm_wakelock_held(wf->rpm); + assert_rpm_wakelock_held(&wf->i915->runtime_pm); if (!refcount_inc_not_zero(&wf->count)) { spin_lock_irqsave(&wf->lock, flags); if (!refcount_inc_not_zero(&wf->count)) { INTEL_WAKEREF_BUG_ON(wf->wakeref); - wf->wakeref = intel_runtime_pm_get_if_in_use(wf->rpm); + wf->wakeref = + intel_runtime_pm_get_if_in_use(&wf->i915->runtime_pm); refcount_set(&wf->count, 1); } spin_unlock_irqrestore(&wf->lock, flags); diff --git a/drivers/gpu/drm/i915/intel_wakeref.h b/drivers/gpu/drm/i915/intel_wakeref.h index 0b6b4852ab236a80381a647099890a6653aa5caf..ec881b097368962d6afa3f5b8ec5846678205bc7 100644 --- a/drivers/gpu/drm/i915/intel_wakeref.h +++ b/drivers/gpu/drm/i915/intel_wakeref.h @@ -39,7 +39,7 @@ struct intel_wakeref { intel_wakeref_t wakeref; - struct intel_runtime_pm *rpm; + struct drm_i915_private *i915; const struct intel_wakeref_ops *ops; struct delayed_work work; @@ -51,13 +51,13 @@ struct intel_wakeref_lockclass { }; void __intel_wakeref_init(struct intel_wakeref *wf, - struct intel_runtime_pm *rpm, + struct drm_i915_private *i915, const struct intel_wakeref_ops *ops, struct intel_wakeref_lockclass *key); -#define intel_wakeref_init(wf, rpm, ops) do { \ +#define intel_wakeref_init(wf, i915, ops) do { \ static struct intel_wakeref_lockclass __key; \ \ - __intel_wakeref_init((wf), (rpm), (ops), &__key); \ + __intel_wakeref_init((wf), (i915), (ops), &__key); \ } while (0) int __intel_wakeref_get_first(struct intel_wakeref *wf); @@ -262,7 +262,7 @@ __intel_wakeref_defer_park(struct intel_wakeref *wf) int intel_wakeref_wait_for_idle(struct intel_wakeref *wf); struct intel_wakeref_auto { - struct intel_runtime_pm *rpm; + struct drm_i915_private *i915; struct timer_list timer; intel_wakeref_t wakeref; spinlock_t lock; @@ -287,7 +287,7 @@ struct intel_wakeref_auto { void intel_wakeref_auto(struct intel_wakeref_auto *wf, unsigned long timeout); void intel_wakeref_auto_init(struct intel_wakeref_auto *wf, - struct intel_runtime_pm *rpm); + struct drm_i915_private *i915); void intel_wakeref_auto_fini(struct intel_wakeref_auto *wf); #endif /* INTEL_WAKEREF_H */ diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_cmd_interface_43.h b/drivers/gpu/drm/i915/pxp/intel_pxp_cmd_interface_43.h index 09777719cd8476fb36077b639890b942e4b9a705..0165d38fbead189f8012abccaffb14ccb5164a7d 100644 --- a/drivers/gpu/drm/i915/pxp/intel_pxp_cmd_interface_43.h +++ b/drivers/gpu/drm/i915/pxp/intel_pxp_cmd_interface_43.h @@ -11,19 +11,30 @@ /* PXP-Cmd-Op definitions */ #define PXP43_CMDID_START_HUC_AUTH 0x0000003A +#define PXP43_CMDID_NEW_HUC_AUTH 0x0000003F /* MTL+ */ #define PXP43_CMDID_INIT_SESSION 0x00000036 /* PXP-Packet sizes for MTL's GSCCS-HECI instruction */ #define PXP43_MAX_HECI_INOUT_SIZE (SZ_32K) -/* PXP-Input-Packet: HUC-Authentication */ +/* PXP-Packet size for MTL's NEW_HUC_AUTH instruction */ +#define PXP43_HUC_AUTH_INOUT_SIZE (SZ_4K) + +/* PXP-Input-Packet: HUC Load and Authentication */ struct pxp43_start_huc_auth_in { struct pxp_cmd_header header; __le64 huc_base_address; } __packed; -/* PXP-Output-Packet: HUC-Authentication */ -struct pxp43_start_huc_auth_out { +/* PXP-Input-Packet: HUC Auth-only */ +struct pxp43_new_huc_auth_in { + struct pxp_cmd_header header; + u64 huc_base_address; + u32 huc_size; +} __packed; + +/* PXP-Output-Packet: HUC Load and Authentication or Auth-only */ +struct pxp43_huc_auth_out { struct pxp_cmd_header header; } __packed; diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c index 8dc41de3f6f747702dcd819a52868d9b0e360d52..f13890ec7db15a17f02be4d870bbb39a9aac2b87 100644 --- a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c +++ b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c @@ -143,7 +143,7 @@ gsccs_send_message(struct intel_pxp *pxp, reply_size = header->message_size - sizeof(*header); if (reply_size > msg_out_size_max) { - drm_warn(&i915->drm, "caller with insufficient PXP reply size %u (%ld)\n", + drm_warn(&i915->drm, "caller with insufficient PXP reply size %u (%zu)\n", reply_size, msg_out_size_max); reply_size = msg_out_size_max; } @@ -196,7 +196,7 @@ bool intel_pxp_gsccs_is_ready_for_sessions(struct intel_pxp *pxp) * gsc-proxy init flow (the last set of dependencies that * are out of order) will suffice. */ - if (intel_huc_is_authenticated(&pxp->ctrl_gt->uc.huc) && + if (intel_huc_is_authenticated(&pxp->ctrl_gt->uc.huc, INTEL_HUC_AUTH_BY_GSC) && intel_gsc_uc_fw_proxy_init_done(&pxp->ctrl_gt->uc.gsc)) return true; diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_huc.c b/drivers/gpu/drm/i915/pxp/intel_pxp_huc.c index 23431c36b60b732e5aae66c61555a1c6c5d8797b..5eedce9169421bb0318400a82a5da33d91842bcd 100644 --- a/drivers/gpu/drm/i915/pxp/intel_pxp_huc.c +++ b/drivers/gpu/drm/i915/pxp/intel_pxp_huc.c @@ -19,7 +19,7 @@ int intel_pxp_huc_load_and_auth(struct intel_pxp *pxp) struct intel_gt *gt; struct intel_huc *huc; struct pxp43_start_huc_auth_in huc_in = {0}; - struct pxp43_start_huc_auth_out huc_out = {0}; + struct pxp43_huc_auth_out huc_out = {0}; dma_addr_t huc_phys_addr; u8 client_id = 0; u8 fence_id = 0; diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c index 36940ef10108dca8c3a09484159bc40217bf9e99..5c397a2df70e28ce2e816b77b9cc7ba697f549c5 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c @@ -391,7 +391,7 @@ static void close_object_list(struct list_head *objects, struct i915_address_space *vm) { struct drm_i915_gem_object *obj, *on; - int ignored; + int __maybe_unused ignored; list_for_each_entry_safe(obj, on, objects, st_link) { struct i915_vma *vma; diff --git a/drivers/gpu/drm/i915/selftests/i915_perf.c b/drivers/gpu/drm/i915/selftests/i915_perf.c index 24dde55314230f2c721b907b2479b66b86212dca..d4608b220123ce623789527e568a5b85c45e6197 100644 --- a/drivers/gpu/drm/i915/selftests/i915_perf.c +++ b/drivers/gpu/drm/i915/selftests/i915_perf.c @@ -28,7 +28,7 @@ alloc_empty_config(struct i915_perf *perf) oa_config->perf = perf; kref_init(&oa_config->ref); - strlcpy(oa_config->uuid, TEST_OA_CONFIG_UUID, sizeof(oa_config->uuid)); + strscpy(oa_config->uuid, TEST_OA_CONFIG_UUID, sizeof(oa_config->uuid)); mutex_lock(&perf->metrics_lock); diff --git a/drivers/gpu/drm/i915/selftests/i915_sw_fence.c b/drivers/gpu/drm/i915/selftests/i915_sw_fence.c index daa985e5a19bc1b10b64a4942928f05ae509c4c0..8f5ce71fa453f4b4a5dc82119f19a845c4d9b9ac 100644 --- a/drivers/gpu/drm/i915/selftests/i915_sw_fence.c +++ b/drivers/gpu/drm/i915/selftests/i915_sw_fence.c @@ -523,12 +523,19 @@ static void task_ipc(struct work_struct *work) static int test_ipc(void *arg) { struct task_ipc ipc; + struct workqueue_struct *wq; int ret = 0; + wq = alloc_workqueue("i1915-selftest", 0, 0); + if (wq == NULL) + return -ENOMEM; + /* Test use of i915_sw_fence as an interprocess signaling mechanism */ ipc.in = alloc_fence(); - if (!ipc.in) - return -ENOMEM; + if (!ipc.in) { + ret = -ENOMEM; + goto err_work; + } ipc.out = alloc_fence(); if (!ipc.out) { ret = -ENOMEM; @@ -540,7 +547,7 @@ static int test_ipc(void *arg) ipc.value = 0; INIT_WORK_ONSTACK(&ipc.work, task_ipc); - schedule_work(&ipc.work); + queue_work(wq, &ipc.work); wait_for_completion(&ipc.started); @@ -563,6 +570,9 @@ static int test_ipc(void *arg) free_fence(ipc.out); err_in: free_fence(ipc.in); +err_work: + destroy_workqueue(wq); + return ret; } diff --git a/drivers/gpu/drm/i915/selftests/mock_gem_device.c b/drivers/gpu/drm/i915/selftests/mock_gem_device.c index 0eda8b4ee17f5bf14f0192bc1ad47670294ce615..09d4bbcdcdbfec57695bd40dc9637de67d143228 100644 --- a/drivers/gpu/drm/i915/selftests/mock_gem_device.c +++ b/drivers/gpu/drm/i915/selftests/mock_gem_device.c @@ -69,6 +69,7 @@ static void mock_device_release(struct drm_device *dev) i915_gem_drain_workqueue(i915); mock_fini_ggtt(to_gt(i915)->ggtt); + destroy_workqueue(i915->unordered_wq); destroy_workqueue(i915->wq); intel_region_ttm_device_fini(i915); @@ -208,6 +209,10 @@ struct drm_i915_private *mock_gem_device(void) if (!i915->wq) goto err_drv; + i915->unordered_wq = alloc_workqueue("mock-unordered", 0, 0); + if (!i915->unordered_wq) + goto err_wq; + mock_init_contexts(i915); /* allocate the ggtt */ @@ -239,6 +244,8 @@ struct drm_i915_private *mock_gem_device(void) err_context: intel_gt_driver_remove(to_gt(i915)); err_unlock: + destroy_workqueue(i915->unordered_wq); +err_wq: destroy_workqueue(i915->wq); err_drv: intel_region_ttm_device_fini(i915); diff --git a/drivers/gpu/drm/lima/lima_sched.c b/drivers/gpu/drm/lima/lima_sched.c index ff003403fbbc7ffd04cd2986b5dbe833a32e7db6..ffd91a5ee29901e934eae58e2e6b42e9a5959457 100644 --- a/drivers/gpu/drm/lima/lima_sched.c +++ b/drivers/gpu/drm/lima/lima_sched.c @@ -165,7 +165,7 @@ int lima_sched_context_init(struct lima_sched_pipe *pipe, void lima_sched_context_fini(struct lima_sched_pipe *pipe, struct lima_sched_context *context) { - drm_sched_entity_fini(&context->base); + drm_sched_entity_destroy(&context->base); } struct dma_fence *lima_sched_context_queue_task(struct lima_sched_task *task) diff --git a/drivers/gpu/drm/mgag200/mgag200_mode.c b/drivers/gpu/drm/mgag200/mgag200_mode.c index 0f2dd26755df98e9052484a5fafd208b645eefa6..af3ce5a6a636ac7bceb83539b66cbfc0a66d45b4 100644 --- a/drivers/gpu/drm/mgag200/mgag200_mode.c +++ b/drivers/gpu/drm/mgag200/mgag200_mode.c @@ -642,6 +642,11 @@ void mgag200_crtc_helper_atomic_enable(struct drm_crtc *crtc, struct drm_atomic_ if (funcs->pixpllc_atomic_update) funcs->pixpllc_atomic_update(crtc, old_state); + if (crtc_state->gamma_lut) + mgag200_crtc_set_gamma(mdev, format, crtc_state->gamma_lut->data); + else + mgag200_crtc_set_gamma_linear(mdev, format); + mgag200_enable_display(mdev); if (funcs->enable_vidrst) diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile index 7274c41228ed981bb8809340827e8d8c3ad4300e..8d02d8c33069509a8806cadbc2867d6a51f96b14 100644 --- a/drivers/gpu/drm/msm/Makefile +++ b/drivers/gpu/drm/msm/Makefile @@ -65,6 +65,7 @@ msm-$(CONFIG_DRM_MSM_DPU) += \ disp/dpu1/dpu_hw_catalog.o \ disp/dpu1/dpu_hw_ctl.o \ disp/dpu1/dpu_hw_dsc.o \ + disp/dpu1/dpu_hw_dsc_1_2.o \ disp/dpu1/dpu_hw_interrupts.o \ disp/dpu1/dpu_hw_intf.o \ disp/dpu1/dpu_hw_lm.o \ @@ -122,7 +123,6 @@ msm-$(CONFIG_DRM_MSM_DP)+= dp/dp_aux.o \ dp/dp_ctrl.o \ dp/dp_display.o \ dp/dp_drm.o \ - dp/dp_hpd.o \ dp/dp_link.o \ dp/dp_panel.o \ dp/dp_parser.o \ diff --git a/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c b/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c index 6bd397a85834f40105f17fc840149db92f4a0e13..169b8fe688f896dc08ad59e1169c25965a4bd066 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c @@ -69,7 +69,7 @@ static void roq_print(struct msm_gpu *gpu, struct drm_printer *p) static int show(struct seq_file *m, void *arg) { - struct drm_info_node *node = (struct drm_info_node *) m->private; + struct drm_info_node *node = m->private; struct drm_device *dev = node->minor->dev; struct msm_drm_private *priv = dev->dev_private; struct drm_printer p = drm_seq_file_printer(m); diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c index 1e8d2982d603cd17b0e2f95c772db43419d68245..a99310b68793205a07d0d088c9de251c00edff9a 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c @@ -1743,6 +1743,7 @@ struct msm_gpu *a5xx_gpu_init(struct drm_device *dev) { struct msm_drm_private *priv = dev->dev_private; struct platform_device *pdev = priv->gpu_pdev; + struct adreno_platform_config *config = pdev->dev.platform_data; struct a5xx_gpu *a5xx_gpu = NULL; struct adreno_gpu *adreno_gpu; struct msm_gpu *gpu; @@ -1769,7 +1770,7 @@ struct msm_gpu *a5xx_gpu_init(struct drm_device *dev) nr_rings = 4; - if (adreno_is_a510(adreno_gpu)) + if (adreno_cmp_rev(ADRENO_REV(5, 1, 0, ANY_ID), config->rev)) nr_rings = 1; ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, nr_rings); diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c index e16b4b3f8535f718e8a6a0ba221a47189e613962..5deb79924897afded45a9399e1e3e29e5e534921 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c @@ -354,7 +354,7 @@ void a6xx_gmu_clear_oob(struct a6xx_gmu *gmu, enum a6xx_gmu_oob_state state) } /* Enable CPU control of SPTP power power collapse */ -static int a6xx_sptprac_enable(struct a6xx_gmu *gmu) +int a6xx_sptprac_enable(struct a6xx_gmu *gmu) { int ret; u32 val; @@ -376,7 +376,7 @@ static int a6xx_sptprac_enable(struct a6xx_gmu *gmu) } /* Disable CPU control of SPTP power power collapse */ -static void a6xx_sptprac_disable(struct a6xx_gmu *gmu) +void a6xx_sptprac_disable(struct a6xx_gmu *gmu) { u32 val; int ret; @@ -479,12 +479,6 @@ static int a6xx_rpmh_start(struct a6xx_gmu *gmu) gmu_write(gmu, REG_A6XX_GMU_RSCC_CONTROL_REQ, 0); - /* Set up CX GMU counter 0 to count busy ticks */ - gmu_write(gmu, REG_A6XX_GPU_GMU_AO_GPU_CX_BUSY_MASK, 0xff000000); - gmu_rmw(gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_SELECT_0, 0xff, 0x20); - - /* Enable the power counter */ - gmu_write(gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_ENABLE, 1); return 0; } @@ -868,43 +862,6 @@ static void a6xx_gmu_rpmh_off(struct a6xx_gmu *gmu) (val & 1), 100, 1000); } -#define GBIF_CLIENT_HALT_MASK BIT(0) -#define GBIF_ARB_HALT_MASK BIT(1) - -static void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu, - bool gx_off) -{ - struct msm_gpu *gpu = &adreno_gpu->base; - - if (!a6xx_has_gbif(adreno_gpu)) { - gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0xf); - spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) & - 0xf) == 0xf); - gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0); - - return; - } - - if (gx_off) { - /* Halt the gx side of GBIF */ - gpu_write(gpu, REG_A6XX_RBBM_GBIF_HALT, 1); - spin_until(gpu_read(gpu, REG_A6XX_RBBM_GBIF_HALT_ACK) & 1); - } - - /* Halt new client requests on GBIF */ - gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_CLIENT_HALT_MASK); - spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) & - (GBIF_CLIENT_HALT_MASK)) == GBIF_CLIENT_HALT_MASK); - - /* Halt all AXI requests on GBIF */ - gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_ARB_HALT_MASK); - spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) & - (GBIF_ARB_HALT_MASK)) == GBIF_ARB_HALT_MASK); - - /* The GBIF halt needs to be explicitly cleared */ - gpu_write(gpu, REG_A6XX_GBIF_HALT, 0x0); -} - /* Force the GMU off in case it isn't responsive */ static void a6xx_gmu_force_off(struct a6xx_gmu *gmu) { @@ -912,6 +869,12 @@ static void a6xx_gmu_force_off(struct a6xx_gmu *gmu) struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; struct msm_gpu *gpu = &adreno_gpu->base; + /* + * Turn off keep alive that might have been enabled by the hang + * interrupt + */ + gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_GMU_PWR_COL_KEEPALIVE, 0); + /* Flush all the queues */ a6xx_hfi_stop(gmu); @@ -930,8 +893,7 @@ static void a6xx_gmu_force_off(struct a6xx_gmu *gmu) a6xx_bus_clear_pending_transactions(adreno_gpu, true); /* Reset GPU core blocks */ - gpu_write(gpu, REG_A6XX_RBBM_SW_RESET_CMD, 1); - udelay(100); + a6xx_gpu_sw_reset(gpu, true); } static void a6xx_gmu_set_initial_freq(struct msm_gpu *gpu, struct a6xx_gmu *gmu) @@ -1469,6 +1431,7 @@ static int a6xx_gmu_get_irq(struct a6xx_gmu *gmu, struct platform_device *pdev, void a6xx_gmu_remove(struct a6xx_gpu *a6xx_gpu) { + struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; struct a6xx_gmu *gmu = &a6xx_gpu->gmu; struct platform_device *pdev = to_platform_device(gmu->dev); @@ -1494,10 +1457,12 @@ void a6xx_gmu_remove(struct a6xx_gpu *a6xx_gpu) gmu->mmio = NULL; gmu->rscc = NULL; - a6xx_gmu_memory_free(gmu); + if (!adreno_has_gmu_wrapper(adreno_gpu)) { + a6xx_gmu_memory_free(gmu); - free_irq(gmu->gmu_irq, gmu); - free_irq(gmu->hfi_irq, gmu); + free_irq(gmu->gmu_irq, gmu); + free_irq(gmu->hfi_irq, gmu); + } /* Drop reference taken in of_find_device_by_node */ put_device(gmu->dev); @@ -1516,6 +1481,69 @@ static int cxpd_notifier_cb(struct notifier_block *nb, return 0; } +int a6xx_gmu_wrapper_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node) +{ + struct platform_device *pdev = of_find_device_by_node(node); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + int ret; + + if (!pdev) + return -ENODEV; + + gmu->dev = &pdev->dev; + + of_dma_configure(gmu->dev, node, true); + + pm_runtime_enable(gmu->dev); + + /* Mark legacy for manual SPTPRAC control */ + gmu->legacy = true; + + /* Map the GMU registers */ + gmu->mmio = a6xx_gmu_get_mmio(pdev, "gmu"); + if (IS_ERR(gmu->mmio)) { + ret = PTR_ERR(gmu->mmio); + goto err_mmio; + } + + gmu->cxpd = dev_pm_domain_attach_by_name(gmu->dev, "cx"); + if (IS_ERR(gmu->cxpd)) { + ret = PTR_ERR(gmu->cxpd); + goto err_mmio; + } + + if (!device_link_add(gmu->dev, gmu->cxpd, DL_FLAG_PM_RUNTIME)) { + ret = -ENODEV; + goto detach_cxpd; + } + + init_completion(&gmu->pd_gate); + complete_all(&gmu->pd_gate); + gmu->pd_nb.notifier_call = cxpd_notifier_cb; + + /* Get a link to the GX power domain to reset the GPU */ + gmu->gxpd = dev_pm_domain_attach_by_name(gmu->dev, "gx"); + if (IS_ERR(gmu->gxpd)) { + ret = PTR_ERR(gmu->gxpd); + goto err_mmio; + } + + gmu->initialized = true; + + return 0; + +detach_cxpd: + dev_pm_domain_detach(gmu->cxpd, false); + +err_mmio: + iounmap(gmu->mmio); + + /* Drop reference taken in of_find_device_by_node */ + put_device(gmu->dev); + + return ret; +} + int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node) { struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; @@ -1526,8 +1554,6 @@ int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node) if (!pdev) return -ENODEV; - mutex_init(&gmu->lock); - gmu->dev = &pdev->dev; of_dma_configure(gmu->dev, node, true); diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h index 0bc3eb443fec80842fc5335b32bd4aff6481ebdb..236f81a43caa62b8574739e7723aeb9b8ada20d3 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h @@ -51,8 +51,8 @@ struct a6xx_gmu { struct msm_gem_address_space *aspace; - void * __iomem mmio; - void * __iomem rscc; + void __iomem *mmio; + void __iomem *rscc; int hfi_irq; int gmu_irq; @@ -193,5 +193,7 @@ int a6xx_hfi_set_freq(struct a6xx_gmu *gmu, int index); bool a6xx_gmu_gx_is_on(struct a6xx_gmu *gmu); bool a6xx_gmu_sptprac_is_on(struct a6xx_gmu *gmu); +void a6xx_sptprac_disable(struct a6xx_gmu *gmu); +int a6xx_sptprac_enable(struct a6xx_gmu *gmu); #endif diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c index 9fb214f150dd47d98ba9dfb0786a366a3e45d5ce..b3ada1e7b598baf5dd63a89e39139d52462cc340 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c @@ -21,7 +21,7 @@ static inline bool _a6xx_check_idle(struct msm_gpu *gpu) struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); /* Check that the GMU is idle */ - if (!a6xx_gmu_isidle(&a6xx_gpu->gmu)) + if (!adreno_has_gmu_wrapper(adreno_gpu) && !a6xx_gmu_isidle(&a6xx_gpu->gmu)) return false; /* Check tha the CX master is idle */ @@ -252,6 +252,56 @@ static void a6xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) a6xx_flush(gpu, ring); } +const struct adreno_reglist a612_hwcg[] = { + {REG_A6XX_RBBM_CLOCK_CNTL_SP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_SP0, 0x02222220}, + {REG_A6XX_RBBM_CLOCK_DELAY_SP0, 0x00000081}, + {REG_A6XX_RBBM_CLOCK_HYST_SP0, 0x0000f3cf}, + {REG_A6XX_RBBM_CLOCK_CNTL_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL3_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL4_TP0, 0x00022222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY3_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY4_TP0, 0x00011111}, + {REG_A6XX_RBBM_CLOCK_HYST_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST2_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST3_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST4_TP0, 0x00077777}, + {REG_A6XX_RBBM_CLOCK_CNTL_RB0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RB0, 0x01202222}, + {REG_A6XX_RBBM_CLOCK_CNTL_CCU0, 0x00002220}, + {REG_A6XX_RBBM_CLOCK_HYST_RB_CCU0, 0x00040f00}, + {REG_A6XX_RBBM_CLOCK_CNTL_RAC, 0x05522022}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RAC, 0x00005555}, + {REG_A6XX_RBBM_CLOCK_DELAY_RAC, 0x00000011}, + {REG_A6XX_RBBM_CLOCK_HYST_RAC, 0x00445044}, + {REG_A6XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222}, + {REG_A6XX_RBBM_CLOCK_MODE_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_MODE_GPC, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ_2, 0x00000002}, + {REG_A6XX_RBBM_CLOCK_MODE_HLSQ, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00004000}, + {REG_A6XX_RBBM_CLOCK_DELAY_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GPC, 0x00000200}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_VFD, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_GPC, 0x04104004}, + {REG_A6XX_RBBM_CLOCK_HYST_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_HYST_UCHE, 0x00000004}, + {REG_A6XX_RBBM_CLOCK_DELAY_UCHE, 0x00000002}, + {REG_A6XX_RBBM_ISDB_CNT, 0x00000182}, + {REG_A6XX_RBBM_RAC_THRESHOLD_CNT, 0x00000000}, + {REG_A6XX_RBBM_SP_HYST_CNT, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_GMU_GX, 0x00000222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GMU_GX, 0x00000111}, + {REG_A6XX_RBBM_CLOCK_HYST_GMU_GX, 0x00000555}, + {}, +}; + /* For a615 family (a615, a616, a618 and a619) */ const struct adreno_reglist a615_hwcg[] = { {REG_A6XX_RBBM_CLOCK_CNTL_SP0, 0x02222222}, @@ -588,6 +638,63 @@ const struct adreno_reglist a660_hwcg[] = { {}, }; +const struct adreno_reglist a690_hwcg[] = { + {REG_A6XX_RBBM_CLOCK_CNTL_SP0, 0x02222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_SP0, 0x02222220}, + {REG_A6XX_RBBM_CLOCK_DELAY_SP0, 0x00000080}, + {REG_A6XX_RBBM_CLOCK_HYST_SP0, 0x0000F3CF}, + {REG_A6XX_RBBM_CLOCK_CNTL_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL3_TP0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL4_TP0, 0x00022222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY2_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY3_TP0, 0x11111111}, + {REG_A6XX_RBBM_CLOCK_DELAY4_TP0, 0x00011111}, + {REG_A6XX_RBBM_CLOCK_HYST_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST2_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST3_TP0, 0x77777777}, + {REG_A6XX_RBBM_CLOCK_HYST4_TP0, 0x00077777}, + {REG_A6XX_RBBM_CLOCK_CNTL_RB0, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RB0, 0x01002222}, + {REG_A6XX_RBBM_CLOCK_CNTL_CCU0, 0x00002220}, + {REG_A6XX_RBBM_CLOCK_HYST_RB_CCU0, 0x00040F00}, + {REG_A6XX_RBBM_CLOCK_CNTL_RAC, 0x25222022}, + {REG_A6XX_RBBM_CLOCK_CNTL2_RAC, 0x00005555}, + {REG_A6XX_RBBM_CLOCK_DELAY_RAC, 0x00000011}, + {REG_A6XX_RBBM_CLOCK_HYST_RAC, 0x00445044}, + {REG_A6XX_RBBM_CLOCK_CNTL_TSE_RAS_RBBM, 0x04222222}, + {REG_A6XX_RBBM_CLOCK_MODE_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_MODE_GPC, 0x00222222}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ_2, 0x00000002}, + {REG_A6XX_RBBM_CLOCK_MODE_HLSQ, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00004000}, + {REG_A6XX_RBBM_CLOCK_DELAY_VFD, 0x00002222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GPC, 0x00000200}, + {REG_A6XX_RBBM_CLOCK_DELAY_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_TSE_RAS_RBBM, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_VFD, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_HYST_GPC, 0x04104004}, + {REG_A6XX_RBBM_CLOCK_HYST_HLSQ, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_TEX_FCHE, 0x00000222}, + {REG_A6XX_RBBM_CLOCK_DELAY_TEX_FCHE, 0x00000111}, + {REG_A6XX_RBBM_CLOCK_HYST_TEX_FCHE, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_UCHE, 0x22222222}, + {REG_A6XX_RBBM_CLOCK_HYST_UCHE, 0x00000004}, + {REG_A6XX_RBBM_CLOCK_DELAY_UCHE, 0x00000002}, + {REG_A6XX_RBBM_CLOCK_CNTL, 0x8AA8AA82}, + {REG_A6XX_RBBM_ISDB_CNT, 0x00000182}, + {REG_A6XX_RBBM_RAC_THRESHOLD_CNT, 0x00000000}, + {REG_A6XX_RBBM_SP_HYST_CNT, 0x00000000}, + {REG_A6XX_RBBM_CLOCK_CNTL_GMU_GX, 0x00000222}, + {REG_A6XX_RBBM_CLOCK_DELAY_GMU_GX, 0x00000111}, + {REG_A6XX_RBBM_CLOCK_HYST_GMU_GX, 0x00000555}, + {REG_A6XX_GPU_GMU_AO_GMU_CGC_MODE_CNTL, 0x20200}, + {REG_A6XX_GPU_GMU_AO_GMU_CGC_DELAY_CNTL, 0x10111}, + {REG_A6XX_GPU_GMU_AO_GMU_CGC_HYST_CNTL, 0x5555}, + {} +}; + static void a6xx_set_hwcg(struct msm_gpu *gpu, bool state) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); @@ -602,6 +709,8 @@ static void a6xx_set_hwcg(struct msm_gpu *gpu, bool state) if (adreno_is_a630(adreno_gpu)) clock_cntl_on = 0x8aa8aa02; + else if (adreno_is_a610(adreno_gpu)) + clock_cntl_on = 0xaaa8aa82; else clock_cntl_on = 0x8aa8aa82; @@ -612,13 +721,15 @@ static void a6xx_set_hwcg(struct msm_gpu *gpu, bool state) return; /* Disable SP clock before programming HWCG registers */ - gmu_rmw(gmu, REG_A6XX_GPU_GMU_GX_SPTPRAC_CLOCK_CONTROL, 1, 0); + if (!adreno_is_a610(adreno_gpu)) + gmu_rmw(gmu, REG_A6XX_GPU_GMU_GX_SPTPRAC_CLOCK_CONTROL, 1, 0); for (i = 0; (reg = &adreno_gpu->info->hwcg[i], reg->offset); i++) gpu_write(gpu, reg->offset, state ? reg->value : 0); /* Enable SP clock */ - gmu_rmw(gmu, REG_A6XX_GPU_GMU_GX_SPTPRAC_CLOCK_CONTROL, 0, 1); + if (!adreno_is_a610(adreno_gpu)) + gmu_rmw(gmu, REG_A6XX_GPU_GMU_GX_SPTPRAC_CLOCK_CONTROL, 0, 1); gpu_write(gpu, REG_A6XX_RBBM_CLOCK_CNTL, state ? clock_cntl_on : 0); } @@ -747,6 +858,45 @@ static const u32 a660_protect[] = { A6XX_PROTECT_NORDWR(0x1f8c0, 0x0000), /* note: infinite range */ }; +/* These are for a690 */ +static const u32 a690_protect[] = { + A6XX_PROTECT_RDONLY(0x00000, 0x004ff), + A6XX_PROTECT_RDONLY(0x00501, 0x00001), + A6XX_PROTECT_RDONLY(0x0050b, 0x002f4), + A6XX_PROTECT_NORDWR(0x0050e, 0x00000), + A6XX_PROTECT_NORDWR(0x00510, 0x00000), + A6XX_PROTECT_NORDWR(0x00534, 0x00000), + A6XX_PROTECT_NORDWR(0x00800, 0x00082), + A6XX_PROTECT_NORDWR(0x008a0, 0x00008), + A6XX_PROTECT_NORDWR(0x008ab, 0x00024), + A6XX_PROTECT_RDONLY(0x008de, 0x000ae), + A6XX_PROTECT_NORDWR(0x00900, 0x0004d), + A6XX_PROTECT_NORDWR(0x0098d, 0x00272), + A6XX_PROTECT_NORDWR(0x00e00, 0x00001), + A6XX_PROTECT_NORDWR(0x00e03, 0x0000c), + A6XX_PROTECT_NORDWR(0x03c00, 0x000c3), + A6XX_PROTECT_RDONLY(0x03cc4, 0x01fff), + A6XX_PROTECT_NORDWR(0x08630, 0x001cf), + A6XX_PROTECT_NORDWR(0x08e00, 0x00000), + A6XX_PROTECT_NORDWR(0x08e08, 0x00007), + A6XX_PROTECT_NORDWR(0x08e50, 0x0001f), + A6XX_PROTECT_NORDWR(0x08e80, 0x0027f), + A6XX_PROTECT_NORDWR(0x09624, 0x001db), + A6XX_PROTECT_NORDWR(0x09e60, 0x00011), + A6XX_PROTECT_NORDWR(0x09e78, 0x00187), + A6XX_PROTECT_NORDWR(0x0a630, 0x001cf), + A6XX_PROTECT_NORDWR(0x0ae02, 0x00000), + A6XX_PROTECT_NORDWR(0x0ae50, 0x0012f), + A6XX_PROTECT_NORDWR(0x0b604, 0x00000), + A6XX_PROTECT_NORDWR(0x0b608, 0x00006), + A6XX_PROTECT_NORDWR(0x0be02, 0x00001), + A6XX_PROTECT_NORDWR(0x0be20, 0x0015f), + A6XX_PROTECT_NORDWR(0x0d000, 0x005ff), + A6XX_PROTECT_NORDWR(0x0f000, 0x00bff), + A6XX_PROTECT_RDONLY(0x0fc00, 0x01fff), + A6XX_PROTECT_NORDWR(0x11c00, 0x00000), /*note: infiite range */ +}; + static void a6xx_set_cp_protect(struct msm_gpu *gpu) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); @@ -758,6 +908,11 @@ static void a6xx_set_cp_protect(struct msm_gpu *gpu) count = ARRAY_SIZE(a650_protect); count_max = 48; BUILD_BUG_ON(ARRAY_SIZE(a650_protect) > 48); + } else if (adreno_is_a690(adreno_gpu)) { + regs = a690_protect; + count = ARRAY_SIZE(a690_protect); + count_max = 48; + BUILD_BUG_ON(ARRAY_SIZE(a690_protect) > 48); } else if (adreno_is_a660_family(adreno_gpu)) { regs = a660_protect; count = ARRAY_SIZE(a660_protect); @@ -786,39 +941,77 @@ static void a6xx_set_cp_protect(struct msm_gpu *gpu) static void a6xx_set_ubwc_config(struct msm_gpu *gpu) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); - u32 lower_bit = 2; - u32 amsbc = 0; + /* Unknown, introduced with A650 family, related to UBWC mode/ver 4 */ u32 rgb565_predicator = 0; + /* Unknown, introduced with A650 family */ u32 uavflagprd_inv = 0; + /* Whether the minimum access length is 64 bits */ + u32 min_acc_len = 0; + /* Entirely magic, per-GPU-gen value */ + u32 ubwc_mode = 0; + /* + * The Highest Bank Bit value represents the bit of the highest DDR bank. + * We then subtract 13 from it (13 is the minimum value allowed by hw) and + * write the lowest two bits of the remaining value as hbb_lo and the + * one above it as hbb_hi to the hardware. This should ideally use DRAM + * type detection. + */ + u32 hbb_hi = 0; + u32 hbb_lo = 2; + /* Unknown, introduced with A640/680 */ + u32 amsbc = 0; + + if (adreno_is_a610(adreno_gpu)) { + /* HBB = 14 */ + hbb_lo = 1; + min_acc_len = 1; + ubwc_mode = 1; + } /* a618 is using the hw default values */ if (adreno_is_a618(adreno_gpu)) return; + if (adreno_is_a619_holi(adreno_gpu)) + hbb_lo = 0; + if (adreno_is_a640_family(adreno_gpu)) amsbc = 1; if (adreno_is_a650(adreno_gpu) || adreno_is_a660(adreno_gpu)) { /* TODO: get ddr type from bootloader and use 2 for LPDDR4 */ - lower_bit = 3; + hbb_lo = 3; + amsbc = 1; + rgb565_predicator = 1; + uavflagprd_inv = 2; + } + + if (adreno_is_a690(adreno_gpu)) { + hbb_lo = 2; amsbc = 1; rgb565_predicator = 1; uavflagprd_inv = 2; } if (adreno_is_7c3(adreno_gpu)) { - lower_bit = 1; + hbb_lo = 1; amsbc = 1; rgb565_predicator = 1; uavflagprd_inv = 2; } gpu_write(gpu, REG_A6XX_RB_NC_MODE_CNTL, - rgb565_predicator << 11 | amsbc << 4 | lower_bit << 1); - gpu_write(gpu, REG_A6XX_TPL1_NC_MODE_CNTL, lower_bit << 1); - gpu_write(gpu, REG_A6XX_SP_NC_MODE_CNTL, - uavflagprd_inv << 4 | lower_bit << 1); - gpu_write(gpu, REG_A6XX_UCHE_MODE_CNTL, lower_bit << 21); + rgb565_predicator << 11 | hbb_hi << 10 | amsbc << 4 | + min_acc_len << 3 | hbb_lo << 1 | ubwc_mode); + + gpu_write(gpu, REG_A6XX_TPL1_NC_MODE_CNTL, hbb_hi << 4 | + min_acc_len << 3 | hbb_lo << 1 | ubwc_mode); + + gpu_write(gpu, REG_A6XX_SP_NC_MODE_CNTL, hbb_hi << 10 | + uavflagprd_inv << 4 | min_acc_len << 3 | + hbb_lo << 1 | ubwc_mode); + + gpu_write(gpu, REG_A6XX_UCHE_MODE_CNTL, min_acc_len << 23 | hbb_lo << 21); } static int a6xx_cp_init(struct msm_gpu *gpu) @@ -997,17 +1190,32 @@ static int hw_init(struct msm_gpu *gpu) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; int ret; - /* Make sure the GMU keeps the GPU on while we set it up */ - a6xx_gmu_set_oob(&a6xx_gpu->gmu, GMU_OOB_GPU_SET); + if (!adreno_has_gmu_wrapper(adreno_gpu)) { + /* Make sure the GMU keeps the GPU on while we set it up */ + a6xx_gmu_set_oob(&a6xx_gpu->gmu, GMU_OOB_GPU_SET); + } /* Clear GBIF halt in case GX domain was not collapsed */ - if (a6xx_has_gbif(adreno_gpu)) + if (adreno_is_a619_holi(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_GBIF_HALT, 0); + gpu_write(gpu, REG_A6XX_RBBM_GPR0_CNTL, 0); + /* Let's make extra sure that the GPU can access the memory.. */ + mb(); + } else if (a6xx_has_gbif(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_GBIF_HALT, 0); gpu_write(gpu, REG_A6XX_RBBM_GBIF_HALT, 0); + /* Let's make extra sure that the GPU can access the memory.. */ + mb(); + } gpu_write(gpu, REG_A6XX_RBBM_SECVID_TSB_CNTL, 0); + if (adreno_is_a619_holi(adreno_gpu)) + a6xx_sptprac_enable(gmu); + /* * Disable the trusted memory range - we don't actually supported secure * memory rendering at this point in time and we don't want to block off @@ -1034,13 +1242,13 @@ static int hw_init(struct msm_gpu *gpu) a6xx_set_hwcg(gpu, true); /* VBIF/GBIF start*/ - if (adreno_is_a640_family(adreno_gpu) || + if (adreno_is_a610(adreno_gpu) || + adreno_is_a640_family(adreno_gpu) || adreno_is_a650_family(adreno_gpu)) { gpu_write(gpu, REG_A6XX_GBIF_QSB_SIDE0, 0x00071620); gpu_write(gpu, REG_A6XX_GBIF_QSB_SIDE1, 0x00071620); gpu_write(gpu, REG_A6XX_GBIF_QSB_SIDE2, 0x00071620); gpu_write(gpu, REG_A6XX_GBIF_QSB_SIDE3, 0x00071620); - gpu_write(gpu, REG_A6XX_GBIF_QSB_SIDE3, 0x00071620); gpu_write(gpu, REG_A6XX_RBBM_GBIF_CLIENT_QOS_CNTL, 0x3); } else { gpu_write(gpu, REG_A6XX_RBBM_VBIF_CLIENT_QOS_CNTL, 0x3); @@ -1068,28 +1276,40 @@ static int hw_init(struct msm_gpu *gpu) gpu_write(gpu, REG_A6XX_UCHE_FILTER_CNTL, 0x804); gpu_write(gpu, REG_A6XX_UCHE_CACHE_WAYS, 0x4); - if (adreno_is_a640_family(adreno_gpu) || - adreno_is_a650_family(adreno_gpu)) + if (adreno_is_a640_family(adreno_gpu) || adreno_is_a650_family(adreno_gpu)) { gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_2, 0x02000140); - else + gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_1, 0x8040362c); + } else if (adreno_is_a610(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_2, 0x00800060); + gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_1, 0x40201b16); + } else { gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_2, 0x010000c0); - gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_1, 0x8040362c); + gpu_write(gpu, REG_A6XX_CP_ROQ_THRESHOLDS_1, 0x8040362c); + } if (adreno_is_a660_family(adreno_gpu)) gpu_write(gpu, REG_A6XX_CP_LPAC_PROG_FIFO_SIZE, 0x00000020); /* Setting the mem pool size */ - gpu_write(gpu, REG_A6XX_CP_MEM_POOL_SIZE, 128); + if (adreno_is_a610(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_CP_MEM_POOL_SIZE, 48); + gpu_write(gpu, REG_A6XX_CP_MEM_POOL_DBG_ADDR, 47); + } else + gpu_write(gpu, REG_A6XX_CP_MEM_POOL_SIZE, 128); /* Setting the primFifo thresholds default values, * and vccCacheSkipDis=1 bit (0x200) for A640 and newer */ - if (adreno_is_a650(adreno_gpu) || adreno_is_a660(adreno_gpu)) + if (adreno_is_a650(adreno_gpu) || adreno_is_a660(adreno_gpu) || adreno_is_a690(adreno_gpu)) gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00300200); else if (adreno_is_a640_family(adreno_gpu) || adreno_is_7c3(adreno_gpu)) gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00200200); else if (adreno_is_a650(adreno_gpu) || adreno_is_a660(adreno_gpu)) gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00300200); + else if (adreno_is_a619(adreno_gpu)) + gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00018000); + else if (adreno_is_a610(adreno_gpu)) + gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00080000); else gpu_write(gpu, REG_A6XX_PC_DBG_ECO_CNTL, 0x00180000); @@ -1105,8 +1325,12 @@ static int hw_init(struct msm_gpu *gpu) a6xx_set_ubwc_config(gpu); /* Enable fault detection */ - gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL, - (1 << 30) | 0x1fffff); + if (adreno_is_a619(adreno_gpu)) + gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL, (1 << 30) | 0x3fffff); + else if (adreno_is_a610(adreno_gpu)) + gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL, (1 << 30) | 0x3ffff); + else + gpu_write(gpu, REG_A6XX_RBBM_INTERFACE_HANG_INT_CNTL, (1 << 30) | 0x1fffff); gpu_write(gpu, REG_A6XX_UCHE_CLIENT_PF, 1); @@ -1123,6 +1347,13 @@ static int hw_init(struct msm_gpu *gpu) 0x3f0243f0); } + /* Set up the CX GMU counter 0 to count busy ticks */ + gmu_write(gmu, REG_A6XX_GPU_GMU_AO_GPU_CX_BUSY_MASK, 0xff000000); + + /* Enable the power counter */ + gmu_rmw(gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_SELECT_0, 0xff, BIT(5)); + gmu_write(gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_ENABLE, 1); + /* Protect registers from the CP */ a6xx_set_cp_protect(gpu); @@ -1212,6 +1443,8 @@ static int hw_init(struct msm_gpu *gpu) } out: + if (adreno_has_gmu_wrapper(adreno_gpu)) + return ret; /* * Tell the GMU that we are done touching the GPU and it can start power * management @@ -1246,9 +1479,6 @@ static void a6xx_dump(struct msm_gpu *gpu) adreno_dump(gpu); } -#define VBIF_RESET_ACK_TIMEOUT 100 -#define VBIF_RESET_ACK_MASK 0x00f0 - static void a6xx_recover(struct msm_gpu *gpu) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); @@ -1274,12 +1504,6 @@ static void a6xx_recover(struct msm_gpu *gpu) /* Halt SQE first */ gpu_write(gpu, REG_A6XX_CP_SQE_CNTL, 3); - /* - * Turn off keep alive that might have been enabled by the hang - * interrupt - */ - gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_GMU_PWR_COL_KEEPALIVE, 0); - pm_runtime_dont_use_autosuspend(&gpu->pdev->dev); /* active_submit won't change until we make a submission */ @@ -1292,6 +1516,15 @@ static void a6xx_recover(struct msm_gpu *gpu) */ gpu->active_submits = 0; + if (adreno_has_gmu_wrapper(adreno_gpu)) { + /* Drain the outstanding traffic on memory buses */ + a6xx_bus_clear_pending_transactions(adreno_gpu, true); + + /* Reset the GPU to a clean state */ + a6xx_gpu_sw_reset(gpu, true); + a6xx_gpu_sw_reset(gpu, false); + } + reinit_completion(&gmu->pd_gate); dev_pm_genpd_add_notifier(gmu->cxpd, &gmu->pd_nb); dev_pm_genpd_synced_poweroff(gmu->cxpd); @@ -1442,7 +1675,8 @@ static void a6xx_fault_detect_irq(struct msm_gpu *gpu) * Force the GPU to stay on until after we finish * collecting information */ - gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_GMU_PWR_COL_KEEPALIVE, 1); + if (!adreno_has_gmu_wrapper(adreno_gpu)) + gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_GMU_PWR_COL_KEEPALIVE, 1); DRM_DEV_ERROR(&gpu->pdev->dev, "gpu fault ring %d fence %x status %8.8X rb %4.4x/%4.4x ib1 %16.16llX/%4.4x ib2 %16.16llX/%4.4x\n", @@ -1573,6 +1807,10 @@ static void a6xx_llc_activate(struct a6xx_gpu *a6xx_gpu) static void a6xx_llc_slices_destroy(struct a6xx_gpu *a6xx_gpu) { + /* No LLCC on non-RPMh (and by extension, non-GMU) SoCs */ + if (adreno_has_gmu_wrapper(&a6xx_gpu->base)) + return; + llcc_slice_putd(a6xx_gpu->llc_slice); llcc_slice_putd(a6xx_gpu->htw_llc_slice); } @@ -1582,6 +1820,10 @@ static void a6xx_llc_slices_init(struct platform_device *pdev, { struct device_node *phandle; + /* No LLCC on non-RPMh (and by extension, non-GMU) SoCs */ + if (adreno_has_gmu_wrapper(&a6xx_gpu->base)) + return; + /* * There is a different programming path for targets with an mmu500 * attached, so detect if that is the case @@ -1603,7 +1845,66 @@ static void a6xx_llc_slices_init(struct platform_device *pdev, a6xx_gpu->llc_mmio = ERR_PTR(-EINVAL); } -static int a6xx_pm_resume(struct msm_gpu *gpu) +#define GBIF_CLIENT_HALT_MASK BIT(0) +#define GBIF_ARB_HALT_MASK BIT(1) +#define VBIF_XIN_HALT_CTRL0_MASK GENMASK(3, 0) +#define VBIF_RESET_ACK_MASK 0xF0 +#define GPR0_GBIF_HALT_REQUEST 0x1E0 + +void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu, bool gx_off) +{ + struct msm_gpu *gpu = &adreno_gpu->base; + + if (adreno_is_a619_holi(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_RBBM_GPR0_CNTL, GPR0_GBIF_HALT_REQUEST); + spin_until((gpu_read(gpu, REG_A6XX_RBBM_VBIF_GX_RESET_STATUS) & + (VBIF_RESET_ACK_MASK)) == VBIF_RESET_ACK_MASK); + } else if (!a6xx_has_gbif(adreno_gpu)) { + gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, VBIF_XIN_HALT_CTRL0_MASK); + spin_until((gpu_read(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL1) & + (VBIF_XIN_HALT_CTRL0_MASK)) == VBIF_XIN_HALT_CTRL0_MASK); + gpu_write(gpu, REG_A6XX_VBIF_XIN_HALT_CTRL0, 0); + + return; + } + + if (gx_off) { + /* Halt the gx side of GBIF */ + gpu_write(gpu, REG_A6XX_RBBM_GBIF_HALT, 1); + spin_until(gpu_read(gpu, REG_A6XX_RBBM_GBIF_HALT_ACK) & 1); + } + + /* Halt new client requests on GBIF */ + gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_CLIENT_HALT_MASK); + spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) & + (GBIF_CLIENT_HALT_MASK)) == GBIF_CLIENT_HALT_MASK); + + /* Halt all AXI requests on GBIF */ + gpu_write(gpu, REG_A6XX_GBIF_HALT, GBIF_ARB_HALT_MASK); + spin_until((gpu_read(gpu, REG_A6XX_GBIF_HALT_ACK) & + (GBIF_ARB_HALT_MASK)) == GBIF_ARB_HALT_MASK); + + /* The GBIF halt needs to be explicitly cleared */ + gpu_write(gpu, REG_A6XX_GBIF_HALT, 0x0); +} + +void a6xx_gpu_sw_reset(struct msm_gpu *gpu, bool assert) +{ + /* 11nm chips (e.g. ones with A610) have hw issues with the reset line! */ + if (adreno_is_a610(to_adreno_gpu(gpu))) + return; + + gpu_write(gpu, REG_A6XX_RBBM_SW_RESET_CMD, assert); + /* Perform a bogus read and add a brief delay to ensure ordering. */ + gpu_read(gpu, REG_A6XX_RBBM_SW_RESET_CMD); + udelay(1); + + /* The reset line needs to be asserted for at least 100 us */ + if (assert) + udelay(100); +} + +static int a6xx_gmu_pm_resume(struct msm_gpu *gpu) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); @@ -1623,10 +1924,61 @@ static int a6xx_pm_resume(struct msm_gpu *gpu) a6xx_llc_activate(a6xx_gpu); - return 0; + return ret; } -static int a6xx_pm_suspend(struct msm_gpu *gpu) +static int a6xx_pm_resume(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + unsigned long freq = gpu->fast_rate; + struct dev_pm_opp *opp; + int ret; + + gpu->needs_hw_init = true; + + trace_msm_gpu_resume(0); + + mutex_lock(&a6xx_gpu->gmu.lock); + + opp = dev_pm_opp_find_freq_ceil(&gpu->pdev->dev, &freq); + if (IS_ERR(opp)) { + ret = PTR_ERR(opp); + goto err_set_opp; + } + dev_pm_opp_put(opp); + + /* Set the core clock and bus bw, having VDD scaling in mind */ + dev_pm_opp_set_opp(&gpu->pdev->dev, opp); + + pm_runtime_resume_and_get(gmu->dev); + pm_runtime_resume_and_get(gmu->gxpd); + + ret = clk_bulk_prepare_enable(gpu->nr_clocks, gpu->grp_clks); + if (ret) + goto err_bulk_clk; + + if (adreno_is_a619_holi(adreno_gpu)) + a6xx_sptprac_enable(gmu); + + /* If anything goes south, tear the GPU down piece by piece.. */ + if (ret) { +err_bulk_clk: + pm_runtime_put(gmu->gxpd); + pm_runtime_put(gmu->dev); + dev_pm_opp_set_opp(&gpu->pdev->dev, NULL); + } +err_set_opp: + mutex_unlock(&a6xx_gpu->gmu.lock); + + if (!ret) + msm_devfreq_resume(gpu); + + return ret; +} + +static int a6xx_gmu_pm_suspend(struct msm_gpu *gpu) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); @@ -1653,7 +2005,43 @@ static int a6xx_pm_suspend(struct msm_gpu *gpu) return 0; } -static int a6xx_get_timestamp(struct msm_gpu *gpu, uint64_t *value) +static int a6xx_pm_suspend(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + int i; + + trace_msm_gpu_suspend(0); + + msm_devfreq_suspend(gpu); + + mutex_lock(&a6xx_gpu->gmu.lock); + + /* Drain the outstanding traffic on memory buses */ + a6xx_bus_clear_pending_transactions(adreno_gpu, true); + + if (adreno_is_a619_holi(adreno_gpu)) + a6xx_sptprac_disable(gmu); + + clk_bulk_disable_unprepare(gpu->nr_clocks, gpu->grp_clks); + + pm_runtime_put_sync(gmu->gxpd); + dev_pm_opp_set_opp(&gpu->pdev->dev, NULL); + pm_runtime_put_sync(gmu->dev); + + mutex_unlock(&a6xx_gpu->gmu.lock); + + if (a6xx_gpu->shadow_bo) + for (i = 0; i < gpu->nr_rings; i++) + a6xx_gpu->shadow[i] = 0; + + gpu->suspend_count++; + + return 0; +} + +static int a6xx_gmu_get_timestamp(struct msm_gpu *gpu, uint64_t *value) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); @@ -1672,6 +2060,12 @@ static int a6xx_get_timestamp(struct msm_gpu *gpu, uint64_t *value) return 0; } +static int a6xx_get_timestamp(struct msm_gpu *gpu, uint64_t *value) +{ + *value = gpu_read64(gpu, REG_A6XX_CP_ALWAYS_ON_COUNTER); + return 0; +} + static struct msm_ringbuffer *a6xx_active_ring(struct msm_gpu *gpu) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); @@ -1744,7 +2138,8 @@ a6xx_create_address_space(struct msm_gpu *gpu, struct platform_device *pdev) * This allows GPU to set the bus attributes required to use system * cache on behalf of the iommu page table walker. */ - if (!IS_ERR_OR_NULL(a6xx_gpu->htw_llc_slice)) + if (!IS_ERR_OR_NULL(a6xx_gpu->htw_llc_slice) && + !device_iommu_capable(&pdev->dev, IOMMU_CAP_CACHE_COHERENCY)) quirks |= IO_PGTABLE_QUIRK_ARM_OUTER_WBWA; return adreno_iommu_create_address_space(gpu, pdev, quirks); @@ -1809,6 +2204,30 @@ static bool a6xx_progress(struct msm_gpu *gpu, struct msm_ringbuffer *ring) return progress; } +static u32 a610_get_speed_bin(u32 fuse) +{ + /* + * There are (at least) three SoCs implementing A610: SM6125 (trinket), + * SM6115 (bengal) and SM6225 (khaje). Trinket does not have speedbinning, + * as only a single SKU exists and we don't support khaje upstream yet. + * Hence, this matching table is only valid for bengal and can be easily + * expanded if need be. + */ + + if (fuse == 0) + return 0; + else if (fuse == 206) + return 1; + else if (fuse == 200) + return 2; + else if (fuse == 157) + return 3; + else if (fuse == 127) + return 4; + + return UINT_MAX; +} + static u32 a618_get_speed_bin(u32 fuse) { if (fuse == 0) @@ -1821,6 +2240,34 @@ static u32 a618_get_speed_bin(u32 fuse) return UINT_MAX; } +static u32 a619_holi_get_speed_bin(u32 fuse) +{ + /* + * There are (at least) two SoCs implementing A619_holi: SM4350 (holi) + * and SM6375 (blair). Limit the fuse matching to the corresponding + * SoC to prevent bogus frequency setting (as improbable as it may be, + * given unexpected fuse values are.. unexpected! But still possible.) + */ + + if (fuse == 0) + return 0; + + if (of_machine_is_compatible("qcom,sm4350")) { + if (fuse == 138) + return 1; + else if (fuse == 92) + return 2; + } else if (of_machine_is_compatible("qcom,sm6375")) { + if (fuse == 190) + return 1; + else if (fuse == 177) + return 2; + } else + pr_warn("Unknown SoC implementing A619_holi!\n"); + + return UINT_MAX; +} + static u32 a619_get_speed_bin(u32 fuse) { if (fuse == 0) @@ -1874,23 +2321,29 @@ static u32 adreno_7c3_get_speed_bin(u32 fuse) return UINT_MAX; } -static u32 fuse_to_supp_hw(struct device *dev, struct adreno_rev rev, u32 fuse) +static u32 fuse_to_supp_hw(struct device *dev, struct adreno_gpu *adreno_gpu, u32 fuse) { u32 val = UINT_MAX; - if (adreno_cmp_rev(ADRENO_REV(6, 1, 8, ANY_ID), rev)) + if (adreno_is_a610(adreno_gpu)) + val = a610_get_speed_bin(fuse); + + if (adreno_is_a618(adreno_gpu)) val = a618_get_speed_bin(fuse); - if (adreno_cmp_rev(ADRENO_REV(6, 1, 9, ANY_ID), rev)) + else if (adreno_is_a619_holi(adreno_gpu)) + val = a619_holi_get_speed_bin(fuse); + + else if (adreno_is_a619(adreno_gpu)) val = a619_get_speed_bin(fuse); - if (adreno_cmp_rev(ADRENO_REV(6, 3, 5, ANY_ID), rev)) + else if (adreno_is_7c3(adreno_gpu)) val = adreno_7c3_get_speed_bin(fuse); - if (adreno_cmp_rev(ADRENO_REV(6, 4, 0, ANY_ID), rev)) + else if (adreno_is_a640(adreno_gpu)) val = a640_get_speed_bin(fuse); - if (adreno_cmp_rev(ADRENO_REV(6, 5, 0, ANY_ID), rev)) + else if (adreno_is_a650(adreno_gpu)) val = a650_get_speed_bin(fuse); if (val == UINT_MAX) { @@ -1903,7 +2356,7 @@ static u32 fuse_to_supp_hw(struct device *dev, struct adreno_rev rev, u32 fuse) return (1 << val); } -static int a6xx_set_supported_hw(struct device *dev, struct adreno_rev rev) +static int a6xx_set_supported_hw(struct device *dev, struct adreno_gpu *adreno_gpu) { u32 supp_hw; u32 speedbin; @@ -1922,7 +2375,7 @@ static int a6xx_set_supported_hw(struct device *dev, struct adreno_rev rev) return ret; } - supp_hw = fuse_to_supp_hw(dev, rev, speedbin); + supp_hw = fuse_to_supp_hw(dev, adreno_gpu, speedbin); ret = devm_pm_opp_set_supported_hw(dev, &supp_hw, 1); if (ret) @@ -1937,8 +2390,8 @@ static const struct adreno_gpu_funcs funcs = { .set_param = adreno_set_param, .hw_init = a6xx_hw_init, .ucode_load = a6xx_ucode_load, - .pm_suspend = a6xx_pm_suspend, - .pm_resume = a6xx_pm_resume, + .pm_suspend = a6xx_gmu_pm_suspend, + .pm_resume = a6xx_gmu_pm_resume, .recover = a6xx_recover, .submit = a6xx_submit, .active_ring = a6xx_active_ring, @@ -1953,6 +2406,35 @@ static const struct adreno_gpu_funcs funcs = { #if defined(CONFIG_DRM_MSM_GPU_STATE) .gpu_state_get = a6xx_gpu_state_get, .gpu_state_put = a6xx_gpu_state_put, +#endif + .create_address_space = a6xx_create_address_space, + .create_private_address_space = a6xx_create_private_address_space, + .get_rptr = a6xx_get_rptr, + .progress = a6xx_progress, + }, + .get_timestamp = a6xx_gmu_get_timestamp, +}; + +static const struct adreno_gpu_funcs funcs_gmuwrapper = { + .base = { + .get_param = adreno_get_param, + .set_param = adreno_set_param, + .hw_init = a6xx_hw_init, + .ucode_load = a6xx_ucode_load, + .pm_suspend = a6xx_pm_suspend, + .pm_resume = a6xx_pm_resume, + .recover = a6xx_recover, + .submit = a6xx_submit, + .active_ring = a6xx_active_ring, + .irq = a6xx_irq, + .destroy = a6xx_destroy, +#if defined(CONFIG_DRM_MSM_GPU_STATE) + .show = a6xx_show, +#endif + .gpu_busy = a6xx_gpu_busy, +#if defined(CONFIG_DRM_MSM_GPU_STATE) + .gpu_state_get = a6xx_gpu_state_get, + .gpu_state_put = a6xx_gpu_state_put, #endif .create_address_space = a6xx_create_address_space, .create_private_address_space = a6xx_create_private_address_space, @@ -1981,28 +2463,49 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev) adreno_gpu = &a6xx_gpu->base; gpu = &adreno_gpu->base; + mutex_init(&a6xx_gpu->gmu.lock); + adreno_gpu->registers = NULL; + /* Check if there is a GMU phandle and set it up */ + node = of_parse_phandle(pdev->dev.of_node, "qcom,gmu", 0); + /* FIXME: How do we gracefully handle this? */ + BUG_ON(!node); + + adreno_gpu->gmu_is_wrapper = of_device_is_compatible(node, "qcom,adreno-gmu-wrapper"); + /* * We need to know the platform type before calling into adreno_gpu_init * so that the hw_apriv flag can be correctly set. Snoop into the info * and grab the revision number */ info = adreno_info(config->rev); - - if (info && (info->revn == 650 || info->revn == 660 || - adreno_cmp_rev(ADRENO_REV(6, 3, 5, ANY_ID), info->rev))) + if (!info) + return ERR_PTR(-EINVAL); + + /* Assign these early so that we can use the is_aXYZ helpers */ + /* Numeric revision IDs (e.g. 630) */ + adreno_gpu->revn = info->revn; + /* New-style ADRENO_REV()-only */ + adreno_gpu->rev = info->rev; + /* Quirk data */ + adreno_gpu->info = info; + + if (adreno_is_a650(adreno_gpu) || adreno_is_a660_family(adreno_gpu)) adreno_gpu->base.hw_apriv = true; a6xx_llc_slices_init(pdev, a6xx_gpu); - ret = a6xx_set_supported_hw(&pdev->dev, config->rev); + ret = a6xx_set_supported_hw(&pdev->dev, adreno_gpu); if (ret) { a6xx_destroy(&(a6xx_gpu->base.base)); return ERR_PTR(ret); } - ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 1); + if (adreno_has_gmu_wrapper(adreno_gpu)) + ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs_gmuwrapper, 1); + else + ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 1); if (ret) { a6xx_destroy(&(a6xx_gpu->base.base)); return ERR_PTR(ret); @@ -2015,13 +2518,10 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev) if (adreno_is_a618(adreno_gpu) || adreno_is_7c3(adreno_gpu)) priv->gpu_clamp_to_idle = true; - /* Check if there is a GMU phandle and set it up */ - node = of_parse_phandle(pdev->dev.of_node, "qcom,gmu", 0); - - /* FIXME: How do we gracefully handle this? */ - BUG_ON(!node); - - ret = a6xx_gmu_init(a6xx_gpu, node); + if (adreno_has_gmu_wrapper(adreno_gpu)) + ret = a6xx_gmu_wrapper_init(a6xx_gpu, node); + else + ret = a6xx_gmu_init(a6xx_gpu, node); of_node_put(node); if (ret) { a6xx_destroy(&(a6xx_gpu->base.base)); diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h index eea2e60ce3b7bb3baf94bffbfd40a34a90e4b669..c788b06e72da84e4db635705cf8a2af869a221e0 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h @@ -76,6 +76,7 @@ int a6xx_gmu_set_oob(struct a6xx_gmu *gmu, enum a6xx_gmu_oob_state state); void a6xx_gmu_clear_oob(struct a6xx_gmu *gmu, enum a6xx_gmu_oob_state state); int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node); +int a6xx_gmu_wrapper_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node); void a6xx_gmu_remove(struct a6xx_gpu *a6xx_gpu); void a6xx_gmu_set_freq(struct msm_gpu *gpu, struct dev_pm_opp *opp, @@ -88,4 +89,7 @@ void a6xx_show(struct msm_gpu *gpu, struct msm_gpu_state *state, struct msm_gpu_state *a6xx_gpu_state_get(struct msm_gpu *gpu); int a6xx_gpu_state_put(struct msm_gpu_state *state); +void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu, bool gx_off); +void a6xx_gpu_sw_reset(struct msm_gpu *gpu, bool assert); + #endif /* __A6XX_GPU_H__ */ diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c index 30ecdff363e7feee544e70629579481884c8475d..4e5d650578c60439f7485c2aefa98bb5937fa249 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c @@ -1041,16 +1041,18 @@ struct msm_gpu_state *a6xx_gpu_state_get(struct msm_gpu *gpu) /* Get the generic state from the adreno core */ adreno_gpu_state_get(gpu, &a6xx_state->base); - a6xx_get_gmu_registers(gpu, a6xx_state); + if (!adreno_has_gmu_wrapper(adreno_gpu)) { + a6xx_get_gmu_registers(gpu, a6xx_state); - a6xx_state->gmu_log = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.log); - a6xx_state->gmu_hfi = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.hfi); - a6xx_state->gmu_debug = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.debug); + a6xx_state->gmu_log = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.log); + a6xx_state->gmu_hfi = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.hfi); + a6xx_state->gmu_debug = a6xx_snapshot_gmu_bo(a6xx_state, &a6xx_gpu->gmu.debug); - a6xx_snapshot_gmu_hfi_history(gpu, a6xx_state); + a6xx_snapshot_gmu_hfi_history(gpu, a6xx_state); + } /* If GX isn't on the rest of the data isn't going to be accessible */ - if (!a6xx_gmu_gx_is_on(&a6xx_gpu->gmu)) + if (!adreno_has_gmu_wrapper(adreno_gpu) && !a6xx_gmu_gx_is_on(&a6xx_gpu->gmu)) return &a6xx_state->base; /* Get the banks of indexed registers */ diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c index 2cc83e049613308371eab877c26d60469fc0d3d2..25b235b49ebc212f1b4d35db1a1f1cd2040a5bbe 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c @@ -414,6 +414,37 @@ static void a650_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) msg->cnoc_cmds_data[1][0] = 0x60000001; } +static void a690_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) +{ + /* + * Send a single "off" entry just to get things running + * TODO: bus scaling + */ + msg->bw_level_num = 1; + + msg->ddr_cmds_num = 3; + msg->ddr_wait_bitmask = 0x01; + + msg->ddr_cmds_addrs[0] = 0x50004; + msg->ddr_cmds_addrs[1] = 0x50000; + msg->ddr_cmds_addrs[2] = 0x500ac; + + msg->ddr_cmds_data[0][0] = 0x40000000; + msg->ddr_cmds_data[0][1] = 0x40000000; + msg->ddr_cmds_data[0][2] = 0x40000000; + + /* + * These are the CX (CNOC) votes - these are used by the GMU but the + * votes are known and fixed for the target + */ + msg->cnoc_cmds_num = 1; + msg->cnoc_wait_bitmask = 0x01; + + msg->cnoc_cmds_addrs[0] = 0x5003c; + msg->cnoc_cmds_data[0][0] = 0x40000000; + msg->cnoc_cmds_data[1][0] = 0x60000001; +} + static void a660_build_bw_table(struct a6xx_hfi_msg_bw_table *msg) { /* @@ -531,6 +562,8 @@ static int a6xx_hfi_send_bw_table(struct a6xx_gmu *gmu) adreno_7c3_build_bw_table(&msg); else if (adreno_is_a660(adreno_gpu)) a660_build_bw_table(&msg); + else if (adreno_is_a690(adreno_gpu)) + a690_build_bw_table(&msg); else a6xx_build_bw_table(&msg); diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c index 8cff86e9d35c57cd1b04c5e4894dba8877dd72f8..cb94cfd137a81808c91e5cacde0d659894f807d9 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_device.c +++ b/drivers/gpu/drm/msm/adreno/adreno_device.c @@ -253,6 +253,18 @@ static const struct adreno_info gpulist[] = { .quirks = ADRENO_QUIRK_LMLOADKILL_DISABLE, .init = a5xx_gpu_init, .zapfw = "a540_zap.mdt", + }, { + .rev = ADRENO_REV(6, 1, 0, ANY_ID), + .revn = 610, + .name = "A610", + .fw = { + [ADRENO_FW_SQE] = "a630_sqe.fw", + }, + .gmem = (SZ_128K + SZ_4K), + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .init = a6xx_gpu_init, + .zapfw = "a610_zap.mdt", + .hwcg = a612_hwcg, }, { .rev = ADRENO_REV(6, 1, 8, ANY_ID), .revn = 618, @@ -355,6 +367,20 @@ static const struct adreno_info gpulist[] = { .init = a6xx_gpu_init, .zapfw = "a640_zap.mdt", .hwcg = a640_hwcg, + }, { + .rev = ADRENO_REV(6, 9, 0, ANY_ID), + .revn = 690, + .name = "A690", + .fw = { + [ADRENO_FW_SQE] = "a660_sqe.fw", + [ADRENO_FW_GMU] = "a690_gmu.bin", + }, + .gmem = SZ_4M, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .init = a6xx_gpu_init, + .zapfw = "a690_zap.mdt", + .hwcg = a690_hwcg, + .address_space_size = SZ_16G, }, }; @@ -551,7 +577,6 @@ static int adreno_bind(struct device *dev, struct device *master, void *data) config.rev.minor, config.rev.patchid); priv->is_a2xx = config.rev.core == 2; - priv->has_cached_coherent = config.rev.core >= 6; gpu = info->init(drm); if (IS_ERR(gpu)) { @@ -563,6 +588,10 @@ static int adreno_bind(struct device *dev, struct device *master, void *data) if (ret) return ret; + if (config.rev.core >= 6) + if (!adreno_has_gmu_wrapper(to_adreno_gpu(gpu))) + priv->has_cached_coherent = true; + return 0; } diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index 6934cee07d42d078b3195c08a94cba8c6c49584c..5c5901d6595073586daa4a98d724bbff527e76ff 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -528,6 +528,10 @@ int adreno_load_fw(struct adreno_gpu *adreno_gpu) if (!adreno_gpu->info->fw[i]) continue; + /* Skip loading GMU firwmare with GMU Wrapper */ + if (adreno_has_gmu_wrapper(adreno_gpu) && i == ADRENO_FW_GMU) + continue; + /* Skip if the firmware has already been loaded */ if (adreno_gpu->fw[i]) continue; @@ -1074,8 +1078,8 @@ int adreno_gpu_init(struct drm_device *drm, struct platform_device *pdev, u32 speedbin; int ret; - /* Only handle the core clock when GMU is not in use */ - if (config->rev.core < 6) { + /* Only handle the core clock when GMU is not in use (or is absent). */ + if (adreno_has_gmu_wrapper(adreno_gpu) || config->rev.core < 6) { /* * This can only be done before devm_pm_opp_of_add_table(), or * dev_pm_opp_set_config() will WARN_ON() diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h index f62612a5c70f35194a690ee058b62780fbafb020..d8c9e8cc375355cd40b7858e711184341d52b0b8 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h @@ -55,7 +55,8 @@ struct adreno_reglist { u32 value; }; -extern const struct adreno_reglist a615_hwcg[], a630_hwcg[], a640_hwcg[], a650_hwcg[], a660_hwcg[]; +extern const struct adreno_reglist a612_hwcg[], a615_hwcg[], a630_hwcg[], a640_hwcg[], a650_hwcg[]; +extern const struct adreno_reglist a660_hwcg[], a690_hwcg[]; struct adreno_info { struct adreno_rev rev; @@ -115,6 +116,7 @@ struct adreno_gpu { * code (a3xx_gpu.c) and stored in this common location. */ const unsigned int *reg_offsets; + bool gmu_is_wrapper; }; #define to_adreno_gpu(x) container_of(x, struct adreno_gpu, base) @@ -145,148 +147,194 @@ struct adreno_platform_config { bool adreno_cmp_rev(struct adreno_rev rev1, struct adreno_rev rev2); -static inline bool adreno_is_a2xx(struct adreno_gpu *gpu) +static inline bool adreno_is_revn(const struct adreno_gpu *gpu, uint32_t revn) { + WARN_ON_ONCE(!gpu->revn); + + return gpu->revn == revn; +} + +static inline bool adreno_has_gmu_wrapper(const struct adreno_gpu *gpu) +{ + return gpu->gmu_is_wrapper; +} + +static inline bool adreno_is_a2xx(const struct adreno_gpu *gpu) +{ + WARN_ON_ONCE(!gpu->revn); + return (gpu->revn < 300); } -static inline bool adreno_is_a20x(struct adreno_gpu *gpu) +static inline bool adreno_is_a20x(const struct adreno_gpu *gpu) { + WARN_ON_ONCE(!gpu->revn); + return (gpu->revn < 210); } -static inline bool adreno_is_a225(struct adreno_gpu *gpu) +static inline bool adreno_is_a225(const struct adreno_gpu *gpu) { - return gpu->revn == 225; + return adreno_is_revn(gpu, 225); } -static inline bool adreno_is_a305(struct adreno_gpu *gpu) +static inline bool adreno_is_a305(const struct adreno_gpu *gpu) { - return gpu->revn == 305; + return adreno_is_revn(gpu, 305); } -static inline bool adreno_is_a306(struct adreno_gpu *gpu) +static inline bool adreno_is_a306(const struct adreno_gpu *gpu) { /* yes, 307, because a305c is 306 */ - return gpu->revn == 307; + return adreno_is_revn(gpu, 307); } -static inline bool adreno_is_a320(struct adreno_gpu *gpu) +static inline bool adreno_is_a320(const struct adreno_gpu *gpu) { - return gpu->revn == 320; + return adreno_is_revn(gpu, 320); } -static inline bool adreno_is_a330(struct adreno_gpu *gpu) +static inline bool adreno_is_a330(const struct adreno_gpu *gpu) { - return gpu->revn == 330; + return adreno_is_revn(gpu, 330); } -static inline bool adreno_is_a330v2(struct adreno_gpu *gpu) +static inline bool adreno_is_a330v2(const struct adreno_gpu *gpu) { return adreno_is_a330(gpu) && (gpu->rev.patchid > 0); } -static inline int adreno_is_a405(struct adreno_gpu *gpu) +static inline int adreno_is_a405(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 405); +} + +static inline int adreno_is_a420(const struct adreno_gpu *gpu) { - return gpu->revn == 405; + return adreno_is_revn(gpu, 420); } -static inline int adreno_is_a420(struct adreno_gpu *gpu) +static inline int adreno_is_a430(const struct adreno_gpu *gpu) { - return gpu->revn == 420; + return adreno_is_revn(gpu, 430); } -static inline int adreno_is_a430(struct adreno_gpu *gpu) +static inline int adreno_is_a506(const struct adreno_gpu *gpu) { - return gpu->revn == 430; + return adreno_is_revn(gpu, 506); } -static inline int adreno_is_a506(struct adreno_gpu *gpu) +static inline int adreno_is_a508(const struct adreno_gpu *gpu) { - return gpu->revn == 506; + return adreno_is_revn(gpu, 508); } -static inline int adreno_is_a508(struct adreno_gpu *gpu) +static inline int adreno_is_a509(const struct adreno_gpu *gpu) { - return gpu->revn == 508; + return adreno_is_revn(gpu, 509); } -static inline int adreno_is_a509(struct adreno_gpu *gpu) +static inline int adreno_is_a510(const struct adreno_gpu *gpu) { - return gpu->revn == 509; + return adreno_is_revn(gpu, 510); } -static inline int adreno_is_a510(struct adreno_gpu *gpu) +static inline int adreno_is_a512(const struct adreno_gpu *gpu) { - return gpu->revn == 510; + return adreno_is_revn(gpu, 512); } -static inline int adreno_is_a512(struct adreno_gpu *gpu) +static inline int adreno_is_a530(const struct adreno_gpu *gpu) { - return gpu->revn == 512; + return adreno_is_revn(gpu, 530); } -static inline int adreno_is_a530(struct adreno_gpu *gpu) +static inline int adreno_is_a540(const struct adreno_gpu *gpu) { - return gpu->revn == 530; + return adreno_is_revn(gpu, 540); } -static inline int adreno_is_a540(struct adreno_gpu *gpu) +static inline int adreno_is_a610(const struct adreno_gpu *gpu) { - return gpu->revn == 540; + return adreno_is_revn(gpu, 610); } -static inline int adreno_is_a618(struct adreno_gpu *gpu) +static inline int adreno_is_a618(const struct adreno_gpu *gpu) { - return gpu->revn == 618; + return adreno_is_revn(gpu, 618); } -static inline int adreno_is_a619(struct adreno_gpu *gpu) +static inline int adreno_is_a619(const struct adreno_gpu *gpu) { - return gpu->revn == 619; + return adreno_is_revn(gpu, 619); } -static inline int adreno_is_a630(struct adreno_gpu *gpu) +static inline int adreno_is_a619_holi(const struct adreno_gpu *gpu) { - return gpu->revn == 630; + return adreno_is_a619(gpu) && adreno_has_gmu_wrapper(gpu); } -static inline int adreno_is_a640_family(struct adreno_gpu *gpu) +static inline int adreno_is_a630(const struct adreno_gpu *gpu) { - return (gpu->revn == 640) || (gpu->revn == 680); + return adreno_is_revn(gpu, 630); } -static inline int adreno_is_a650(struct adreno_gpu *gpu) +static inline int adreno_is_a640(const struct adreno_gpu *gpu) { - return gpu->revn == 650; + return adreno_is_revn(gpu, 640); } -static inline int adreno_is_7c3(struct adreno_gpu *gpu) +static inline int adreno_is_a650(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 650); +} + +static inline int adreno_is_7c3(const struct adreno_gpu *gpu) { /* The order of args is important here to handle ANY_ID correctly */ return adreno_cmp_rev(ADRENO_REV(6, 3, 5, ANY_ID), gpu->rev); } -static inline int adreno_is_a660(struct adreno_gpu *gpu) +static inline int adreno_is_a660(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 660); +} + +static inline int adreno_is_a680(const struct adreno_gpu *gpu) { - return gpu->revn == 660; + return adreno_is_revn(gpu, 680); } +static inline int adreno_is_a690(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 690); +}; + /* check for a615, a616, a618, a619 or any derivatives */ -static inline int adreno_is_a615_family(struct adreno_gpu *gpu) +static inline int adreno_is_a615_family(const struct adreno_gpu *gpu) { - return gpu->revn == 615 || gpu->revn == 616 || gpu->revn == 618 || gpu->revn == 619; + return adreno_is_revn(gpu, 615) || + adreno_is_revn(gpu, 616) || + adreno_is_revn(gpu, 618) || + adreno_is_revn(gpu, 619); } -static inline int adreno_is_a660_family(struct adreno_gpu *gpu) +static inline int adreno_is_a660_family(const struct adreno_gpu *gpu) { - return adreno_is_a660(gpu) || adreno_is_7c3(gpu); + return adreno_is_a660(gpu) || adreno_is_a690(gpu) || adreno_is_7c3(gpu); } /* check for a650, a660, or any derivatives */ -static inline int adreno_is_a650_family(struct adreno_gpu *gpu) +static inline int adreno_is_a650_family(const struct adreno_gpu *gpu) +{ + return adreno_is_revn(gpu, 650) || + adreno_is_revn(gpu, 620) || + adreno_is_a660_family(gpu); +} + +static inline int adreno_is_a640_family(const struct adreno_gpu *gpu) { - return gpu->revn == 650 || gpu->revn == 620 || adreno_is_a660_family(gpu); + return adreno_is_a640(gpu) || adreno_is_a680(gpu); } u64 adreno_private_address_space_size(struct msm_gpu *gpu); diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h index 2b3ae84057dfe0fdc87ef73b40da378dd06979e3..7d0d0e74c3b0807bac2516015647b731d275635a 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h @@ -30,7 +30,7 @@ static const struct dpu_mdp_cfg msm8998_mdp[] = { { .name = "top_0", .id = MDP_TOP, .base = 0x0, .len = 0x458, - .features = 0, + .features = BIT(DPU_MDP_VSYNC_SEL), .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, .clk_ctrls[DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, .clk_ctrls[DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, @@ -39,8 +39,8 @@ static const struct dpu_mdp_cfg msm8998_mdp[] = { .clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, .clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 }, .clk_ctrls[DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 12 }, - .clk_ctrls[DPU_CLK_CTRL_CURSOR0] = { .reg_off = 0x3a8, .bit_off = 15 }, - .clk_ctrls[DPU_CLK_CTRL_CURSOR1] = { .reg_off = 0x3b0, .bit_off = 15 }, + .clk_ctrls[DPU_CLK_CTRL_CURSOR0] = { .reg_off = 0x3a8, .bit_off = 16 }, + .clk_ctrls[DPU_CLK_CTRL_CURSOR1] = { .reg_off = 0x3b0, .bit_off = 16 }, }, }; @@ -98,46 +98,59 @@ static const struct dpu_sspp_cfg msm8998_sspp[] = { static const struct dpu_lm_cfg msm8998_lm[] = { LM_BLK("lm_0", LM_0, 0x44000, MIXER_MSM8998_MASK, - &msm8998_lm_sblk, PINGPONG_0, LM_2, DSPP_0), + &msm8998_lm_sblk, PINGPONG_0, LM_1, DSPP_0), LM_BLK("lm_1", LM_1, 0x45000, MIXER_MSM8998_MASK, - &msm8998_lm_sblk, PINGPONG_1, LM_5, DSPP_1), + &msm8998_lm_sblk, PINGPONG_1, LM_0, DSPP_1), LM_BLK("lm_2", LM_2, 0x46000, MIXER_MSM8998_MASK, - &msm8998_lm_sblk, PINGPONG_2, LM_0, 0), + &msm8998_lm_sblk, PINGPONG_2, LM_5, 0), LM_BLK("lm_3", LM_3, 0x47000, MIXER_MSM8998_MASK, - &msm8998_lm_sblk, PINGPONG_MAX, 0, 0), + &msm8998_lm_sblk, PINGPONG_NONE, 0, 0), LM_BLK("lm_4", LM_4, 0x48000, MIXER_MSM8998_MASK, - &msm8998_lm_sblk, PINGPONG_MAX, 0, 0), + &msm8998_lm_sblk, PINGPONG_NONE, 0, 0), LM_BLK("lm_5", LM_5, 0x49000, MIXER_MSM8998_MASK, - &msm8998_lm_sblk, PINGPONG_3, LM_1, 0), + &msm8998_lm_sblk, PINGPONG_3, LM_2, 0), }; static const struct dpu_pingpong_cfg msm8998_pp[] = { - PP_BLK_TE("pingpong_0", PINGPONG_0, 0x70000, 0, sdm845_pp_sblk_te, + PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SDM845_TE2_MASK, 0, sdm845_pp_sblk_te, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)), - PP_BLK_TE("pingpong_1", PINGPONG_1, 0x70800, 0, sdm845_pp_sblk_te, + PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SDM845_TE2_MASK, 0, sdm845_pp_sblk_te, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)), - PP_BLK("pingpong_2", PINGPONG_2, 0x71000, 0, sdm845_pp_sblk, + PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SDM845_MASK, 0, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)), - PP_BLK("pingpong_3", PINGPONG_3, 0x71800, 0, sdm845_pp_sblk, + PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SDM845_MASK, 0, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)), }; +static const struct dpu_dsc_cfg msm8998_dsc[] = { + DSC_BLK("dsc_0", DSC_0, 0x80000, 0), + DSC_BLK("dsc_1", DSC_1, 0x80400, 0), +}; + static const struct dpu_dspp_cfg msm8998_dspp[] = { - DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_MSM8998_MASK, + DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, &msm8998_dspp_sblk), - DSPP_BLK("dspp_1", DSPP_1, 0x56000, DSPP_MSM8998_MASK, + DSPP_BLK("dspp_1", DSPP_1, 0x56000, DSPP_SC7180_MASK, &msm8998_dspp_sblk), }; static const struct dpu_intf_cfg msm8998_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 25, INTF_SDM845_MASK, MDP_SSPP_TOP0_INTR, 24, 25), - INTF_BLK("intf_1", INTF_1, 0x6a800, 0x280, INTF_DSI, 0, 25, INTF_SDM845_MASK, MDP_SSPP_TOP0_INTR, 26, 27), - INTF_BLK("intf_2", INTF_2, 0x6b000, 0x280, INTF_DSI, 1, 25, INTF_SDM845_MASK, MDP_SSPP_TOP0_INTR, 28, 29), - INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_HDMI, 0, 25, INTF_SDM845_MASK, MDP_SSPP_TOP0_INTR, 30, 31), + INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 21, INTF_SDM845_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), + INTF_BLK("intf_1", INTF_1, 0x6a800, 0x280, INTF_DSI, 0, 21, INTF_SDM845_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)), + INTF_BLK("intf_2", INTF_2, 0x6b000, 0x280, INTF_DSI, 1, 21, INTF_SDM845_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)), + INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_HDMI, 0, 21, INTF_SDM845_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), }; static const struct dpu_perf_cfg msm8998_perf_data = { @@ -191,11 +204,12 @@ const struct dpu_mdss_cfg dpu_msm8998_cfg = { .dspp = msm8998_dspp, .pingpong_count = ARRAY_SIZE(msm8998_pp), .pingpong = msm8998_pp, + .dsc_count = ARRAY_SIZE(msm8998_dsc), + .dsc = msm8998_dsc, .intf_count = ARRAY_SIZE(msm8998_intf), .intf = msm8998_intf, .vbif_count = ARRAY_SIZE(msm8998_vbif), .vbif = msm8998_vbif, - .reg_dma_count = 0, .perf = &msm8998_perf_data, .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ BIT(MDP_SSPP_TOP0_INTR2) | \ diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h index ceca741e93c9b812af8a52fe6bffa6685ad8b6d6..b6098141bb9bbfa2266305b6fe704724347431a2 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h @@ -30,7 +30,7 @@ static const struct dpu_mdp_cfg sdm845_mdp[] = { { .name = "top_0", .id = MDP_TOP, .base = 0x0, .len = 0x45c, - .features = BIT(DPU_MDP_AUDIO_SELECT), + .features = BIT(DPU_MDP_AUDIO_SELECT) | BIT(DPU_MDP_VSYNC_SEL), .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, .clk_ctrls[DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, .clk_ctrls[DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, @@ -96,30 +96,41 @@ static const struct dpu_sspp_cfg sdm845_sspp[] = { static const struct dpu_lm_cfg sdm845_lm[] = { LM_BLK("lm_0", LM_0, 0x44000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_0, LM_1, 0), + &sdm845_lm_sblk, PINGPONG_0, LM_1, DSPP_0), LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_1, LM_0, 0), + &sdm845_lm_sblk, PINGPONG_1, LM_0, DSPP_1), LM_BLK("lm_2", LM_2, 0x46000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_2, LM_5, 0), + &sdm845_lm_sblk, PINGPONG_2, LM_5, DSPP_2), LM_BLK("lm_3", LM_3, 0x0, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_MAX, 0, 0), + &sdm845_lm_sblk, PINGPONG_NONE, 0, DSPP_3), LM_BLK("lm_4", LM_4, 0x0, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_MAX, 0, 0), + &sdm845_lm_sblk, PINGPONG_NONE, 0, 0), LM_BLK("lm_5", LM_5, 0x49000, MIXER_SDM845_MASK, &sdm845_lm_sblk, PINGPONG_3, LM_2, 0), }; +static const struct dpu_dspp_cfg sdm845_dspp[] = { + DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, + &sm8150_dspp_sblk), + DSPP_BLK("dspp_1", DSPP_1, 0x56000, DSPP_SC7180_MASK, + &sm8150_dspp_sblk), + DSPP_BLK("dspp_2", DSPP_2, 0x58000, DSPP_SC7180_MASK, + &sm8150_dspp_sblk), + DSPP_BLK("dspp_3", DSPP_3, 0x5a000, DSPP_SC7180_MASK, + &sm8150_dspp_sblk), +}; + static const struct dpu_pingpong_cfg sdm845_pp[] = { - PP_BLK_TE("pingpong_0", PINGPONG_0, 0x70000, 0, sdm845_pp_sblk_te, + PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SDM845_TE2_MASK, 0, sdm845_pp_sblk_te, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)), - PP_BLK_TE("pingpong_1", PINGPONG_1, 0x70800, 0, sdm845_pp_sblk_te, + PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SDM845_TE2_MASK, 0, sdm845_pp_sblk_te, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)), - PP_BLK("pingpong_2", PINGPONG_2, 0x71000, 0, sdm845_pp_sblk, + PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SDM845_MASK, 0, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)), - PP_BLK("pingpong_3", PINGPONG_3, 0x71800, 0, sdm845_pp_sblk, + PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SDM845_MASK, 0, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)), }; @@ -132,10 +143,18 @@ static const struct dpu_dsc_cfg sdm845_dsc[] = { }; static const struct dpu_intf_cfg sdm845_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 24, INTF_SDM845_MASK, MDP_SSPP_TOP0_INTR, 24, 25), - INTF_BLK("intf_1", INTF_1, 0x6a800, 0x280, INTF_DSI, 0, 24, INTF_SDM845_MASK, MDP_SSPP_TOP0_INTR, 26, 27), - INTF_BLK("intf_2", INTF_2, 0x6b000, 0x280, INTF_DSI, 1, 24, INTF_SDM845_MASK, MDP_SSPP_TOP0_INTR, 28, 29), - INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 1, 24, INTF_SDM845_MASK, MDP_SSPP_TOP0_INTR, 30, 31), + INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 24, INTF_SDM845_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), + INTF_BLK("intf_1", INTF_1, 0x6a800, 0x280, INTF_DSI, 0, 24, INTF_SDM845_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)), + INTF_BLK("intf_2", INTF_2, 0x6b000, 0x280, INTF_DSI, 1, 24, INTF_SDM845_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)), + INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 1, 24, INTF_SDM845_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), }; static const struct dpu_perf_cfg sdm845_perf_data = { @@ -185,6 +204,8 @@ const struct dpu_mdss_cfg dpu_sdm845_cfg = { .sspp = sdm845_sspp, .mixer_count = ARRAY_SIZE(sdm845_lm), .mixer = sdm845_lm, + .dspp_count = ARRAY_SIZE(sdm845_dspp), + .dspp = sdm845_dspp, .pingpong_count = ARRAY_SIZE(sdm845_pp), .pingpong = sdm845_pp, .dsc_count = ARRAY_SIZE(sdm845_dsc), @@ -193,8 +214,6 @@ const struct dpu_mdss_cfg dpu_sdm845_cfg = { .intf = sdm845_intf, .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, - .reg_dma_count = 1, - .dma_cfg = &sdm845_regdma, .perf = &sdm845_perf_data, .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ BIT(MDP_SSPP_TOP0_INTR2) | \ diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h index 282d410269ff62b55108794a0c8479e665c855e6..b5f75135426781fc13e7839707b8f3ae00bf8a57 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h @@ -128,22 +128,22 @@ static const struct dpu_dspp_cfg sm8150_dspp[] = { }; static const struct dpu_pingpong_cfg sm8150_pp[] = { - PP_BLK_TE("pingpong_0", PINGPONG_0, 0x70000, MERGE_3D_0, sdm845_pp_sblk_te, + PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)), - PP_BLK_TE("pingpong_1", PINGPONG_1, 0x70800, MERGE_3D_0, sdm845_pp_sblk_te, + -1), + PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)), - PP_BLK("pingpong_2", PINGPONG_2, 0x71000, MERGE_3D_1, sdm845_pp_sblk, + -1), + PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)), - PP_BLK("pingpong_3", PINGPONG_3, 0x71800, MERGE_3D_1, sdm845_pp_sblk, + -1), + PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)), - PP_BLK("pingpong_4", PINGPONG_4, 0x72000, MERGE_3D_2, sdm845_pp_sblk, + -1), + PP_BLK("pingpong_4", PINGPONG_4, 0x72000, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), -1), - PP_BLK("pingpong_5", PINGPONG_5, 0x72800, MERGE_3D_2, sdm845_pp_sblk, + PP_BLK("pingpong_5", PINGPONG_5, 0x72800, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), -1), }; @@ -162,10 +162,20 @@ static const struct dpu_dsc_cfg sm8150_dsc[] = { }; static const struct dpu_intf_cfg sm8150_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 24, 25), - INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2bc, INTF_DSI, 0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 26, 27), - INTF_BLK("intf_2", INTF_2, 0x6b000, 0x2bc, INTF_DSI, 1, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 28, 29), - INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 1, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 30, 31), + INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), + INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2bc, INTF_DSI, 0, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)), + INTF_BLK_DSI_TE("intf_2", INTF_2, 0x6b000, 0x2bc, INTF_DSI, 1, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2)), + INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 1, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), }; static const struct dpu_perf_cfg sm8150_perf_data = { @@ -220,15 +230,15 @@ const struct dpu_mdss_cfg dpu_sm8150_cfg = { .intf = sm8150_intf, .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, - .reg_dma_count = 1, - .dma_cfg = &sm8150_regdma, .perf = &sm8150_perf_data, .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ BIT(MDP_SSPP_TOP0_INTR2) | \ BIT(MDP_SSPP_TOP0_HIST_INTR) | \ BIT(MDP_INTF0_INTR) | \ BIT(MDP_INTF1_INTR) | \ + BIT(MDP_INTF1_TEAR_INTR) | \ BIT(MDP_INTF2_INTR) | \ + BIT(MDP_INTF2_TEAR_INTR) | \ BIT(MDP_INTF3_INTR) | \ BIT(MDP_AD4_0_INTR) | \ BIT(MDP_AD4_1_INTR), diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h index c57400265f288403f39b11d4f87fd489a248eca6..8ed2b263c5ea3d01cdc4deb472f33612fab483f4 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h @@ -102,9 +102,9 @@ static const struct dpu_sspp_cfg sc8180x_sspp[] = { static const struct dpu_lm_cfg sc8180x_lm[] = { LM_BLK("lm_0", LM_0, 0x44000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_0, LM_1, 0), + &sdm845_lm_sblk, PINGPONG_0, LM_1, DSPP_0), LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_1, LM_0, 0), + &sdm845_lm_sblk, PINGPONG_1, LM_0, DSPP_1), LM_BLK("lm_2", LM_2, 0x46000, MIXER_SDM845_MASK, &sdm845_lm_sblk, PINGPONG_2, LM_3, 0), LM_BLK("lm_3", LM_3, 0x47000, MIXER_SDM845_MASK, @@ -115,23 +115,34 @@ static const struct dpu_lm_cfg sc8180x_lm[] = { &sdm845_lm_sblk, PINGPONG_5, LM_4, 0), }; +static const struct dpu_dspp_cfg sc8180x_dspp[] = { + DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, + &sm8150_dspp_sblk), + DSPP_BLK("dspp_1", DSPP_1, 0x56000, DSPP_SC7180_MASK, + &sm8150_dspp_sblk), + DSPP_BLK("dspp_2", DSPP_2, 0x58000, DSPP_SC7180_MASK, + &sm8150_dspp_sblk), + DSPP_BLK("dspp_3", DSPP_3, 0x5a000, DSPP_SC7180_MASK, + &sm8150_dspp_sblk), +}; + static const struct dpu_pingpong_cfg sc8180x_pp[] = { - PP_BLK_TE("pingpong_0", PINGPONG_0, 0x70000, MERGE_3D_0, sdm845_pp_sblk_te, + PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)), - PP_BLK_TE("pingpong_1", PINGPONG_1, 0x70800, MERGE_3D_0, sdm845_pp_sblk_te, + -1), + PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)), - PP_BLK("pingpong_2", PINGPONG_2, 0x71000, MERGE_3D_1, sdm845_pp_sblk, + -1), + PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)), - PP_BLK("pingpong_3", PINGPONG_3, 0x71800, MERGE_3D_1, sdm845_pp_sblk, + -1), + PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)), - PP_BLK("pingpong_4", PINGPONG_4, 0x72000, MERGE_3D_2, sdm845_pp_sblk, + -1), + PP_BLK("pingpong_4", PINGPONG_4, 0x72000, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), -1), - PP_BLK("pingpong_5", PINGPONG_5, 0x72800, MERGE_3D_2, sdm845_pp_sblk, + PP_BLK("pingpong_5", PINGPONG_5, 0x72800, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), -1), }; @@ -142,14 +153,37 @@ static const struct dpu_merge_3d_cfg sc8180x_merge_3d[] = { MERGE_3D_BLK("merge_3d_2", MERGE_3D_2, 0x83200), }; +static const struct dpu_dsc_cfg sc8180x_dsc[] = { + DSC_BLK("dsc_0", DSC_0, 0x80000, BIT(DPU_DSC_OUTPUT_CTRL)), + DSC_BLK("dsc_1", DSC_1, 0x80400, BIT(DPU_DSC_OUTPUT_CTRL)), + DSC_BLK("dsc_2", DSC_2, 0x80800, BIT(DPU_DSC_OUTPUT_CTRL)), + DSC_BLK("dsc_3", DSC_3, 0x80c00, BIT(DPU_DSC_OUTPUT_CTRL)), + DSC_BLK("dsc_4", DSC_4, 0x81000, BIT(DPU_DSC_OUTPUT_CTRL)), + DSC_BLK("dsc_5", DSC_5, 0x81400, BIT(DPU_DSC_OUTPUT_CTRL)), +}; + static const struct dpu_intf_cfg sc8180x_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 24, 25), - INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2bc, INTF_DSI, 0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 26, 27), - INTF_BLK("intf_2", INTF_2, 0x6b000, 0x2bc, INTF_DSI, 1, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 28, 29), + INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), + INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2bc, INTF_DSI, 0, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)), + INTF_BLK_DSI_TE("intf_2", INTF_2, 0x6b000, 0x2bc, INTF_DSI, 1, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2)), /* INTF_3 is for MST, wired to INTF_DP 0 and 1, use dummy index until this is supported */ - INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 999, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 30, 31), - INTF_BLK("intf_4", INTF_4, 0x6c000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 20, 21), - INTF_BLK("intf_5", INTF_5, 0x6c800, 0x280, INTF_DP, MSM_DP_CONTROLLER_2, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 22, 23), + INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 999, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), + INTF_BLK("intf_4", INTF_4, 0x6c000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 20), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 21)), + INTF_BLK("intf_5", INTF_5, 0x6c800, 0x280, INTF_DP, MSM_DP_CONTROLLER_2, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23)), }; static const struct dpu_perf_cfg sc8180x_perf_data = { @@ -190,6 +224,10 @@ const struct dpu_mdss_cfg dpu_sc8180x_cfg = { .sspp = sc8180x_sspp, .mixer_count = ARRAY_SIZE(sc8180x_lm), .mixer = sc8180x_lm, + .dspp_count = ARRAY_SIZE(sc8180x_dspp), + .dspp = sc8180x_dspp, + .dsc_count = ARRAY_SIZE(sc8180x_dsc), + .dsc = sc8180x_dsc, .pingpong_count = ARRAY_SIZE(sc8180x_pp), .pingpong = sc8180x_pp, .merge_3d_count = ARRAY_SIZE(sc8180x_merge_3d), @@ -198,15 +236,15 @@ const struct dpu_mdss_cfg dpu_sc8180x_cfg = { .intf = sc8180x_intf, .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, - .reg_dma_count = 1, - .dma_cfg = &sm8150_regdma, .perf = &sc8180x_perf_data, .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ BIT(MDP_SSPP_TOP0_INTR2) | \ BIT(MDP_SSPP_TOP0_HIST_INTR) | \ BIT(MDP_INTF0_INTR) | \ BIT(MDP_INTF1_INTR) | \ + BIT(MDP_INTF1_TEAR_INTR) | \ BIT(MDP_INTF2_INTR) | \ + BIT(MDP_INTF2_TEAR_INTR) | \ BIT(MDP_INTF3_INTR) | \ BIT(MDP_INTF4_INTR) | \ BIT(MDP_INTF5_INTR) | \ diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h index 2c40229ea51594439a02acb7c16f04d05cfccb49..daebd217004135ba6f3cda0b4b2f07d4593fb526 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h @@ -129,22 +129,22 @@ static const struct dpu_dspp_cfg sm8250_dspp[] = { }; static const struct dpu_pingpong_cfg sm8250_pp[] = { - PP_BLK_TE("pingpong_0", PINGPONG_0, 0x70000, MERGE_3D_0, sdm845_pp_sblk_te, + PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)), - PP_BLK_TE("pingpong_1", PINGPONG_1, 0x70800, MERGE_3D_0, sdm845_pp_sblk_te, + -1), + PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)), - PP_BLK("pingpong_2", PINGPONG_2, 0x71000, MERGE_3D_1, sdm845_pp_sblk, + -1), + PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)), - PP_BLK("pingpong_3", PINGPONG_3, 0x71800, MERGE_3D_1, sdm845_pp_sblk, + -1), + PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)), - PP_BLK("pingpong_4", PINGPONG_4, 0x72000, MERGE_3D_2, sdm845_pp_sblk, + -1), + PP_BLK("pingpong_4", PINGPONG_4, 0x72000, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), -1), - PP_BLK("pingpong_5", PINGPONG_5, 0x72800, MERGE_3D_2, sdm845_pp_sblk, + PP_BLK("pingpong_5", PINGPONG_5, 0x72800, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), -1), }; @@ -163,10 +163,20 @@ static const struct dpu_dsc_cfg sm8250_dsc[] = { }; static const struct dpu_intf_cfg sm8250_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 24, 25), - INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 26, 27), - INTF_BLK("intf_2", INTF_2, 0x6b000, 0x2c0, INTF_DSI, 1, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 28, 29), - INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 1, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 30, 31), + INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), + INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)), + INTF_BLK_DSI_TE("intf_2", INTF_2, 0x6b000, 0x2c0, INTF_DSI, 1, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2)), + INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 1, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), }; static const struct dpu_wb_cfg sm8250_wb[] = { @@ -228,15 +238,15 @@ const struct dpu_mdss_cfg dpu_sm8250_cfg = { .vbif = sdm845_vbif, .wb_count = ARRAY_SIZE(sm8250_wb), .wb = sm8250_wb, - .reg_dma_count = 1, - .dma_cfg = &sm8250_regdma, .perf = &sm8250_perf_data, .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ BIT(MDP_SSPP_TOP0_INTR2) | \ BIT(MDP_SSPP_TOP0_HIST_INTR) | \ BIT(MDP_INTF0_INTR) | \ BIT(MDP_INTF1_INTR) | \ + BIT(MDP_INTF1_TEAR_INTR) | \ BIT(MDP_INTF2_INTR) | \ + BIT(MDP_INTF2_TEAR_INTR) | \ BIT(MDP_INTF3_INTR) | \ BIT(MDP_INTF4_INTR), }; diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h index 8799ed7571190fb7a4726d72a04a31b6dcf26458..67566b07195a2457f94d607e64d8fd14f3706d15 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h @@ -76,17 +76,26 @@ static const struct dpu_lm_cfg sc7180_lm[] = { static const struct dpu_dspp_cfg sc7180_dspp[] = { DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, - &sc7180_dspp_sblk), + &sm8150_dspp_sblk), }; static const struct dpu_pingpong_cfg sc7180_pp[] = { - PP_BLK_TE("pingpong_0", PINGPONG_0, 0x70000, 0, sdm845_pp_sblk_te, -1, -1), - PP_BLK_TE("pingpong_1", PINGPONG_1, 0x70800, 0, sdm845_pp_sblk_te, -1, -1), + PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + -1), + PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + -1), }; static const struct dpu_intf_cfg sc7180_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 24, 25), - INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 26, 27), + INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), + INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)), }; static const struct dpu_wb_cfg sc7180_wb[] = { @@ -143,14 +152,13 @@ const struct dpu_mdss_cfg dpu_sc7180_cfg = { .wb = sc7180_wb, .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, - .reg_dma_count = 1, - .dma_cfg = &sdm845_regdma, .perf = &sc7180_perf_data, .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ BIT(MDP_SSPP_TOP0_INTR2) | \ BIT(MDP_SSPP_TOP0_HIST_INTR) | \ BIT(MDP_INTF0_INTR) | \ - BIT(MDP_INTF1_INTR), + BIT(MDP_INTF1_INTR) | \ + BIT(MDP_INTF1_TEAR_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h index 6f04d8f85c9250ef65aa7b2bc3e628d497a0c275..031fc8dae3c69852b7a9f26831dabec1a6994caa 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h @@ -60,14 +60,16 @@ static const struct dpu_dspp_cfg sm6115_dspp[] = { }; static const struct dpu_pingpong_cfg sm6115_pp[] = { - PP_BLK("pingpong_0", PINGPONG_0, 0x70000, 0, sdm845_pp_sblk, + PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)), + -1), }; static const struct dpu_intf_cfg sm6115_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x00000, 0x280, INTF_NONE, 0, 0, 0, 0, 0, 0), - INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 26, 27), + INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)), }; static const struct dpu_perf_cfg sm6115_perf_data = { @@ -122,8 +124,8 @@ const struct dpu_mdss_cfg dpu_sm6115_cfg = { .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ BIT(MDP_SSPP_TOP0_INTR2) | \ BIT(MDP_SSPP_TOP0_HIST_INTR) | \ - BIT(MDP_INTF0_INTR) | \ - BIT(MDP_INTF1_INTR), + BIT(MDP_INTF1_INTR) | \ + BIT(MDP_INTF1_TEAR_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h new file mode 100644 index 0000000000000000000000000000000000000000..06eba23b02364b6a46394f972ea13d99c6c2aaab --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h @@ -0,0 +1,173 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + * Copyright (c) 2023, Linaro Limited + */ + +#ifndef _DPU_6_4_SM6350_H +#define _DPU_6_4_SM6350_H + +static const struct dpu_caps sm6350_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0x7, + .qseed_type = DPU_SSPP_SCALER_QSEED4, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .max_linewidth = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_ubwc_cfg sm6350_ubwc_cfg = { + .ubwc_version = DPU_HW_UBWC_VER_20, + .ubwc_swizzle = 6, + .highest_bank_bit = 1, +}; + +static const struct dpu_mdp_cfg sm6350_mdp[] = { + { + .name = "top_0", .id = MDP_TOP, + .base = 0x0, .len = 0x494, + .features = 0, + .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + .clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + .clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 }, + .clk_ctrls[DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + }, +}; + +static const struct dpu_ctl_cfg sm6350_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1dc, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, + { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x1dc, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, + { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x1dc, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, + { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x1dc, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, +}; + +static const struct dpu_sspp_cfg sm6350_sspp[] = { + SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x1f8, VIG_SC7180_MASK, + sc7180_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0), + SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x1f8, DMA_SDM845_MASK, + sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0), + SSPP_BLK("sspp_9", SSPP_DMA1, 0x26000, 0x1f8, DMA_CURSOR_SDM845_MASK, + sdm845_dma_sblk_1, 5, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA1), + SSPP_BLK("sspp_10", SSPP_DMA2, 0x28000, 0x1f8, DMA_CURSOR_SDM845_MASK, + sdm845_dma_sblk_2, 9, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA2), +}; + +static const struct dpu_lm_cfg sm6350_lm[] = { + LM_BLK("lm_0", LM_0, 0x44000, MIXER_SDM845_MASK, + &sc7180_lm_sblk, PINGPONG_0, LM_1, DSPP_0), + LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK, + &sc7180_lm_sblk, PINGPONG_1, LM_0, 0), +}; + +static const struct dpu_dspp_cfg sm6350_dspp[] = { + DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, + &sm8150_dspp_sblk), +}; + +static struct dpu_pingpong_cfg sm6350_pp[] = { + PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + -1), + PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + -1), +}; + +static const struct dpu_dsc_cfg sm6350_dsc[] = { + DSC_BLK("dsc_0", DSC_0, 0x80000, BIT(DPU_DSC_OUTPUT_CTRL)), +}; + +static const struct dpu_intf_cfg sm6350_intf[] = { + INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 35, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), + INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 35, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)), +}; + +static const struct dpu_perf_cfg sm6350_perf_data = { + .max_bw_low = 4200000, + .max_bw_high = 5100000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, + .min_dram_ib = 1600000, + .min_prefill_lines = 35, + /* TODO: confirm danger_lut_tbl */ + .danger_lut_tbl = {0xffff, 0xffff, 0x0}, + .safe_lut_tbl = {0xff00, 0xff00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sm6350_qos_linear_macrotile), + .entries = sm6350_qos_linear_macrotile + }, + {.nentry = ARRAY_SIZE(sm6350_qos_linear_macrotile), + .entries = sm6350_qos_linear_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +const struct dpu_mdss_cfg dpu_sm6350_cfg = { + .caps = &sm6350_dpu_caps, + .ubwc = &sm6350_ubwc_cfg, + .mdp_count = ARRAY_SIZE(sm6350_mdp), + .mdp = sm6350_mdp, + .ctl_count = ARRAY_SIZE(sm6350_ctl), + .ctl = sm6350_ctl, + .sspp_count = ARRAY_SIZE(sm6350_sspp), + .sspp = sm6350_sspp, + .mixer_count = ARRAY_SIZE(sm6350_lm), + .mixer = sm6350_lm, + .dspp_count = ARRAY_SIZE(sm6350_dspp), + .dspp = sm6350_dspp, + .dsc_count = ARRAY_SIZE(sm6350_dsc), + .dsc = sm6350_dsc, + .pingpong_count = ARRAY_SIZE(sm6350_pp), + .pingpong = sm6350_pp, + .intf_count = ARRAY_SIZE(sm6350_intf), + .intf = sm6350_intf, + .vbif_count = ARRAY_SIZE(sdm845_vbif), + .vbif = sdm845_vbif, + .perf = &sm6350_perf_data, + .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ + BIT(MDP_SSPP_TOP0_INTR2) | \ + BIT(MDP_SSPP_TOP0_HIST_INTR) | \ + BIT(MDP_INTF0_INTR) | \ + BIT(MDP_INTF1_INTR) | \ + BIT(MDP_INTF1_TEAR_INTR), +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h index 303492d62a5caaa38e39921289f35b672909a9c8..f2808098af39936aab90931e427060c3765b2c68 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h @@ -57,14 +57,16 @@ static const struct dpu_dspp_cfg qcm2290_dspp[] = { }; static const struct dpu_pingpong_cfg qcm2290_pp[] = { - PP_BLK("pingpong_0", PINGPONG_0, 0x70000, 0, sdm845_pp_sblk, + PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)), + -1), }; static const struct dpu_intf_cfg qcm2290_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x00000, 0x280, INTF_NONE, 0, 0, 0, 0, 0, 0), - INTF_BLK("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, MDP_SSPP_TOP0_INTR, 26, 27), + INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)), }; static const struct dpu_perf_cfg qcm2290_perf_data = { @@ -112,8 +114,8 @@ const struct dpu_mdss_cfg dpu_qcm2290_cfg = { .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ BIT(MDP_SSPP_TOP0_INTR2) | \ BIT(MDP_SSPP_TOP0_HIST_INTR) | \ - BIT(MDP_INTF0_INTR) | \ - BIT(MDP_INTF1_INTR), + BIT(MDP_INTF1_INTR) | \ + BIT(MDP_INTF1_TEAR_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h new file mode 100644 index 0000000000000000000000000000000000000000..241fa6746674d914c7f01a905c4b09bc3c6daedd --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h @@ -0,0 +1,138 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + * Copyright (c) 2023, Linaro Limited + */ + +#ifndef _DPU_6_9_SM6375_H +#define _DPU_6_9_SM6375_H + +static const struct dpu_caps sm6375_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_LINE_WIDTH, + .max_mixer_blendstages = 0x4, + .qseed_type = DPU_SSPP_SCALER_QSEED4, + .has_dim_layer = true, + .has_idle_pc = true, + .max_linewidth = 2160, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_ubwc_cfg sm6375_ubwc_cfg = { + .ubwc_version = DPU_HW_UBWC_VER_20, + .ubwc_swizzle = 6, + .highest_bank_bit = 1, +}; + +static const struct dpu_mdp_cfg sm6375_mdp[] = { + { + .name = "top_0", .id = MDP_TOP, + .base = 0x0, .len = 0x494, + .features = 0, + .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + }, +}; + +static const struct dpu_ctl_cfg sm6375_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1dc, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, +}; + +static const struct dpu_sspp_cfg sm6375_sspp[] = { + SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x1f8, VIG_SC7180_MASK, + sm6115_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0), + SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x1f8, DMA_SDM845_MASK, + sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0), +}; + +static const struct dpu_lm_cfg sm6375_lm[] = { + LM_BLK("lm_0", LM_0, 0x44000, MIXER_QCM2290_MASK, + &qcm2290_lm_sblk, PINGPONG_0, 0, DSPP_0), +}; + +static const struct dpu_dspp_cfg sm6375_dspp[] = { + DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, + &sm8150_dspp_sblk), +}; + +static const struct dpu_pingpong_cfg sm6375_pp[] = { + PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + -1), +}; + +static const struct dpu_dsc_cfg sm6375_dsc[] = { + DSC_BLK("dsc_0", DSC_0, 0x80000, BIT(DPU_DSC_OUTPUT_CTRL)), +}; + +static const struct dpu_intf_cfg sm6375_intf[] = { + INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)), +}; + +static const struct dpu_perf_cfg sm6375_perf_data = { + .max_bw_low = 5200000, + .max_bw_high = 6200000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, /* No LLCC on this SoC */ + .min_dram_ib = 1600000, + .min_prefill_lines = 24, + /* TODO: confirm danger_lut_tbl */ + .danger_lut_tbl = {0xffff, 0xffff, 0x0}, + .safe_lut_tbl = {0xfe00, 0xfe00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sm6350_qos_linear_macrotile), + .entries = sm6350_qos_linear_macrotile + }, + {.nentry = ARRAY_SIZE(sm6350_qos_linear_macrotile), + .entries = sm6350_qos_linear_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +const struct dpu_mdss_cfg dpu_sm6375_cfg = { + .caps = &sm6375_dpu_caps, + .ubwc = &sm6375_ubwc_cfg, + .mdp_count = ARRAY_SIZE(sm6375_mdp), + .mdp = sm6375_mdp, + .ctl_count = ARRAY_SIZE(sm6375_ctl), + .ctl = sm6375_ctl, + .sspp_count = ARRAY_SIZE(sm6375_sspp), + .sspp = sm6375_sspp, + .mixer_count = ARRAY_SIZE(sm6375_lm), + .mixer = sm6375_lm, + .dspp_count = ARRAY_SIZE(sm6375_dspp), + .dspp = sm6375_dspp, + .dsc_count = ARRAY_SIZE(sm6375_dsc), + .dsc = sm6375_dsc, + .pingpong_count = ARRAY_SIZE(sm6375_pp), + .pingpong = sm6375_pp, + .intf_count = ARRAY_SIZE(sm6375_intf), + .intf = sm6375_intf, + .vbif_count = ARRAY_SIZE(sdm845_vbif), + .vbif = sdm845_vbif, + .perf = &sm6375_perf_data, + .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ + BIT(MDP_SSPP_TOP0_INTR2) | \ + BIT(MDP_SSPP_TOP0_HIST_INTR) | \ + BIT(MDP_INTF1_INTR) | \ + BIT(MDP_INTF1_TEAR_INTR), +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h index ca107ca8de462482278c3799cf01b953504293ac..8da424eaee6a2db53802b303e131ef7348ab0021 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h @@ -127,22 +127,22 @@ static const struct dpu_dspp_cfg sm8350_dspp[] = { }; static const struct dpu_pingpong_cfg sm8350_pp[] = { - PP_BLK_TE("pingpong_0", PINGPONG_0, 0x69000, MERGE_3D_0, sdm845_pp_sblk_te, + PP_BLK_DITHER("pingpong_0", PINGPONG_0, 0x69000, MERGE_3D_0, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)), - PP_BLK_TE("pingpong_1", PINGPONG_1, 0x6a000, MERGE_3D_0, sdm845_pp_sblk_te, + -1), + PP_BLK_DITHER("pingpong_1", PINGPONG_1, 0x6a000, MERGE_3D_0, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)), - PP_BLK("pingpong_2", PINGPONG_2, 0x6b000, MERGE_3D_1, sdm845_pp_sblk, + -1), + PP_BLK_DITHER("pingpong_2", PINGPONG_2, 0x6b000, MERGE_3D_1, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)), - PP_BLK("pingpong_3", PINGPONG_3, 0x6c000, MERGE_3D_1, sdm845_pp_sblk, + -1), + PP_BLK_DITHER("pingpong_3", PINGPONG_3, 0x6c000, MERGE_3D_1, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)), - PP_BLK("pingpong_4", PINGPONG_4, 0x6d000, MERGE_3D_2, sdm845_pp_sblk, + -1), + PP_BLK_DITHER("pingpong_4", PINGPONG_4, 0x6d000, MERGE_3D_2, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), -1), - PP_BLK("pingpong_5", PINGPONG_5, 0x6e000, MERGE_3D_2, sdm845_pp_sblk, + PP_BLK_DITHER("pingpong_5", PINGPONG_5, 0x6e000, MERGE_3D_2, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), -1), }; @@ -153,11 +153,33 @@ static const struct dpu_merge_3d_cfg sm8350_merge_3d[] = { MERGE_3D_BLK("merge_3d_2", MERGE_3D_2, 0x50000), }; +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg sm8350_dsc[] = { + DSC_BLK_1_2("dce_0_0", DSC_0, 0x80000, 0x29c, 0, dsc_sblk_0), + DSC_BLK_1_2("dce_0_1", DSC_1, 0x80000, 0x29c, 0, dsc_sblk_1), + DSC_BLK_1_2("dce_1_0", DSC_2, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_0), + DSC_BLK_1_2("dce_1_1", DSC_3, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_1), +}; + static const struct dpu_intf_cfg sm8350_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 24, 25), - INTF_BLK("intf_1", INTF_1, 0x35000, 0x2c4, INTF_DSI, 0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 26, 27), - INTF_BLK("intf_2", INTF_2, 0x36000, 0x2c4, INTF_DSI, 1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 28, 29), - INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 30, 31), + INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), + INTF_BLK_DSI_TE("intf_1", INTF_1, 0x35000, 0x2c4, INTF_DSI, 0, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + DPU_IRQ_IDX(MDP_INTF1_7xxx_TEAR_INTR, 2)), + INTF_BLK_DSI_TE("intf_2", INTF_2, 0x36000, 0x2c4, INTF_DSI, 1, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + DPU_IRQ_IDX(MDP_INTF2_7xxx_TEAR_INTR, 2)), + INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), }; static const struct dpu_perf_cfg sm8350_perf_data = { @@ -205,21 +227,23 @@ const struct dpu_mdss_cfg dpu_sm8350_cfg = { .dspp = sm8350_dspp, .pingpong_count = ARRAY_SIZE(sm8350_pp), .pingpong = sm8350_pp, + .dsc_count = ARRAY_SIZE(sm8350_dsc), + .dsc = sm8350_dsc, .merge_3d_count = ARRAY_SIZE(sm8350_merge_3d), .merge_3d = sm8350_merge_3d, .intf_count = ARRAY_SIZE(sm8350_intf), .intf = sm8350_intf, .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, - .reg_dma_count = 1, - .dma_cfg = &sm8350_regdma, .perf = &sm8350_perf_data, .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ BIT(MDP_SSPP_TOP0_INTR2) | \ BIT(MDP_SSPP_TOP0_HIST_INTR) | \ BIT(MDP_INTF0_7xxx_INTR) | \ BIT(MDP_INTF1_7xxx_INTR) | \ + BIT(MDP_INTF1_7xxx_TEAR_INTR) | \ BIT(MDP_INTF2_7xxx_INTR) | \ + BIT(MDP_INTF2_7xxx_TEAR_INTR) | \ BIT(MDP_INTF3_7xxx_INTR), }; diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h index 5957de1859844f633cc414d3e9059ab3203f78be..900fee410e113adf4fb4a6eb9c25eb512ec879fc 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h @@ -31,6 +31,7 @@ static const struct dpu_mdp_cfg sc7280_mdp[] = { .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, .clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, .clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 }, + .clk_ctrls[DPU_CLK_CTRL_WB2] = { .reg_off = 0x3b8, .bit_off = 24 }, }, }; @@ -83,20 +84,45 @@ static const struct dpu_lm_cfg sc7280_lm[] = { static const struct dpu_dspp_cfg sc7280_dspp[] = { DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, - &sc7180_dspp_sblk), + &sm8150_dspp_sblk), }; static const struct dpu_pingpong_cfg sc7280_pp[] = { - PP_BLK("pingpong_0", PINGPONG_0, 0x69000, 0, sc7280_pp_sblk, -1, -1), - PP_BLK("pingpong_1", PINGPONG_1, 0x6a000, 0, sc7280_pp_sblk, -1, -1), - PP_BLK("pingpong_2", PINGPONG_2, 0x6b000, 0, sc7280_pp_sblk, -1, -1), - PP_BLK("pingpong_3", PINGPONG_3, 0x6c000, 0, sc7280_pp_sblk, -1, -1), + PP_BLK_DITHER("pingpong_0", PINGPONG_0, 0x69000, 0, sc7280_pp_sblk, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + -1), + PP_BLK_DITHER("pingpong_1", PINGPONG_1, 0x6a000, 0, sc7280_pp_sblk, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + -1), + PP_BLK_DITHER("pingpong_2", PINGPONG_2, 0x6b000, 0, sc7280_pp_sblk, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + -1), + PP_BLK_DITHER("pingpong_3", PINGPONG_3, 0x6c000, 0, sc7280_pp_sblk, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + -1), +}; + +/* NOTE: sc7280 only has one DSC hard slice encoder */ +static const struct dpu_dsc_cfg sc7280_dsc[] = { + DSC_BLK_1_2("dce_0_0", DSC_0, 0x80000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_0), +}; + +static const struct dpu_wb_cfg sc7280_wb[] = { + WB_BLK("wb_2", WB_2, 0x65000, WB_SM8250_MASK, DPU_CLK_CTRL_WB2, 6, + VBIF_RT, MDP_SSPP_TOP0_INTR, 4096, 4), }; static const struct dpu_intf_cfg sc7280_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 24, 25), - INTF_BLK("intf_1", INTF_1, 0x35000, 0x2c4, INTF_DSI, 0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 26, 27), - INTF_BLK("intf_5", INTF_5, 0x39000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 22, 23), + INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), + INTF_BLK_DSI_TE("intf_1", INTF_1, 0x35000, 0x2c4, INTF_DSI, 0, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + DPU_IRQ_IDX(MDP_INTF1_7xxx_TEAR_INTR, 2)), + INTF_BLK("intf_5", INTF_5, 0x39000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23)), }; static const struct dpu_perf_cfg sc7280_perf_data = { @@ -142,6 +168,10 @@ const struct dpu_mdss_cfg dpu_sc7280_cfg = { .mixer = sc7280_lm, .pingpong_count = ARRAY_SIZE(sc7280_pp), .pingpong = sc7280_pp, + .dsc_count = ARRAY_SIZE(sc7280_dsc), + .dsc = sc7280_dsc, + .wb_count = ARRAY_SIZE(sc7280_wb), + .wb = sc7280_wb, .intf_count = ARRAY_SIZE(sc7280_intf), .intf = sc7280_intf, .vbif_count = ARRAY_SIZE(sdm845_vbif), @@ -152,6 +182,7 @@ const struct dpu_mdss_cfg dpu_sc7280_cfg = { BIT(MDP_SSPP_TOP0_HIST_INTR) | \ BIT(MDP_INTF0_7xxx_INTR) | \ BIT(MDP_INTF1_7xxx_INTR) | \ + BIT(MDP_INTF1_7xxx_TEAR_INTR) | \ BIT(MDP_INTF5_7xxx_INTR), }; diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h index 9aab110b8c44d67458eb507a0bea99b54ed73553..f6ce6b090f7188632c638ee9d177fe65dd0aad34 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h @@ -42,17 +42,18 @@ static const struct dpu_mdp_cfg sc8280xp_mdp[] = { }, }; +/* FIXME: get rid of DPU_CTL_SPLIT_DISPLAY in favour of proper ACTIVE_CTL support */ static const struct dpu_ctl_cfg sc8280xp_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x15000, .len = 0x204, - .features = CTL_SC7280_MASK, + .features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x16000, .len = 0x204, - .features = CTL_SC7280_MASK, + .features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { @@ -121,18 +122,18 @@ static const struct dpu_dspp_cfg sc8280xp_dspp[] = { }; static const struct dpu_pingpong_cfg sc8280xp_pp[] = { - PP_BLK_TE("pingpong_0", PINGPONG_0, 0x69000, MERGE_3D_0, sdm845_pp_sblk_te, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), -1), - PP_BLK_TE("pingpong_1", PINGPONG_1, 0x6a000, MERGE_3D_0, sdm845_pp_sblk_te, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), -1), - PP_BLK_TE("pingpong_2", PINGPONG_2, 0x6b000, MERGE_3D_1, sdm845_pp_sblk_te, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), -1), - PP_BLK_TE("pingpong_3", PINGPONG_3, 0x6c000, MERGE_3D_1, sdm845_pp_sblk_te, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), -1), - PP_BLK_TE("pingpong_4", PINGPONG_4, 0x6d000, MERGE_3D_2, sdm845_pp_sblk_te, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), -1), - PP_BLK_TE("pingpong_5", PINGPONG_5, 0x6e000, MERGE_3D_2, sdm845_pp_sblk_te, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), -1), + PP_BLK_DITHER("pingpong_0", PINGPONG_0, 0x69000, MERGE_3D_0, sc7280_pp_sblk, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), -1), + PP_BLK_DITHER("pingpong_1", PINGPONG_1, 0x6a000, MERGE_3D_0, sc7280_pp_sblk, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), -1), + PP_BLK_DITHER("pingpong_2", PINGPONG_2, 0x6b000, MERGE_3D_1, sc7280_pp_sblk, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), -1), + PP_BLK_DITHER("pingpong_3", PINGPONG_3, 0x6c000, MERGE_3D_1, sc7280_pp_sblk, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), -1), + PP_BLK_DITHER("pingpong_4", PINGPONG_4, 0x6d000, MERGE_3D_2, sc7280_pp_sblk, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), -1), + PP_BLK_DITHER("pingpong_5", PINGPONG_5, 0x6e000, MERGE_3D_2, sc7280_pp_sblk, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), -1), }; static const struct dpu_merge_3d_cfg sc8280xp_merge_3d[] = { @@ -141,17 +142,51 @@ static const struct dpu_merge_3d_cfg sc8280xp_merge_3d[] = { MERGE_3D_BLK("merge_3d_2", MERGE_3D_2, 0x50000), }; +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg sc8280xp_dsc[] = { + DSC_BLK_1_2("dce_0_0", DSC_0, 0x80000, 0x29c, 0, dsc_sblk_0), + DSC_BLK_1_2("dce_0_1", DSC_1, 0x80000, 0x29c, 0, dsc_sblk_1), + DSC_BLK_1_2("dce_1_0", DSC_2, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_0), + DSC_BLK_1_2("dce_1_1", DSC_3, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_1), + DSC_BLK_1_2("dce_2_0", DSC_4, 0x82000, 0x29c, 0, dsc_sblk_0), + DSC_BLK_1_2("dce_2_1", DSC_5, 0x82000, 0x29c, 0, dsc_sblk_1), +}; + /* TODO: INTF 3, 8 and 7 are used for MST, marked as INTF_NONE for now */ static const struct dpu_intf_cfg sc8280xp_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 24, 25), - INTF_BLK("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 26, 27), - INTF_BLK("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 28, 29), - INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_NONE, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 30, 31), - INTF_BLK("intf_4", INTF_4, 0x38000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 20, 21), - INTF_BLK("intf_5", INTF_5, 0x39000, 0x280, INTF_DP, MSM_DP_CONTROLLER_3, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 22, 23), - INTF_BLK("intf_6", INTF_6, 0x3a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_2, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 16, 17), - INTF_BLK("intf_7", INTF_7, 0x3b000, 0x280, INTF_NONE, MSM_DP_CONTROLLER_2, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 18, 19), - INTF_BLK("intf_8", INTF_8, 0x3c000, 0x280, INTF_NONE, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 12, 13), + INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), + INTF_BLK_DSI_TE("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + DPU_IRQ_IDX(MDP_INTF1_7xxx_TEAR_INTR, 2)), + INTF_BLK_DSI_TE("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + DPU_IRQ_IDX(MDP_INTF2_7xxx_TEAR_INTR, 2)), + INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_NONE, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), + INTF_BLK("intf_4", INTF_4, 0x38000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 20), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 21)), + INTF_BLK("intf_5", INTF_5, 0x39000, 0x280, INTF_DP, MSM_DP_CONTROLLER_3, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23)), + INTF_BLK("intf_6", INTF_6, 0x3a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_2, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 16), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 17)), + INTF_BLK("intf_7", INTF_7, 0x3b000, 0x280, INTF_NONE, MSM_DP_CONTROLLER_2, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 18), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 19)), + INTF_BLK("intf_8", INTF_8, 0x3c000, 0x280, INTF_NONE, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)), }; static const struct dpu_perf_cfg sc8280xp_perf_data = { @@ -196,21 +231,23 @@ const struct dpu_mdss_cfg dpu_sc8280xp_cfg = { .dspp = sc8280xp_dspp, .pingpong_count = ARRAY_SIZE(sc8280xp_pp), .pingpong = sc8280xp_pp, + .dsc_count = ARRAY_SIZE(sc8280xp_dsc), + .dsc = sc8280xp_dsc, .merge_3d_count = ARRAY_SIZE(sc8280xp_merge_3d), .merge_3d = sc8280xp_merge_3d, .intf_count = ARRAY_SIZE(sc8280xp_intf), .intf = sc8280xp_intf, .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, - .reg_dma_count = 1, - .dma_cfg = &sc8280xp_regdma, .perf = &sc8280xp_perf_data, .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ BIT(MDP_SSPP_TOP0_INTR2) | \ BIT(MDP_SSPP_TOP0_HIST_INTR) | \ BIT(MDP_INTF0_7xxx_INTR) | \ BIT(MDP_INTF1_7xxx_INTR) | \ + BIT(MDP_INTF1_7xxx_TEAR_INTR) | \ BIT(MDP_INTF2_7xxx_INTR) | \ + BIT(MDP_INTF2_7xxx_TEAR_INTR) | \ BIT(MDP_INTF3_7xxx_INTR) | \ BIT(MDP_INTF4_7xxx_INTR) | \ BIT(MDP_INTF5_7xxx_INTR) | \ diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h index 02a259b6b4268e8ac521e7b5069b5e0f024a01f7..8d13c369213c071638ecda18c0b623cfbdffae03 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h @@ -47,7 +47,7 @@ static const struct dpu_ctl_cfg sm8450_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x15000, .len = 0x204, - .features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY) | BIT(DPU_CTL_FETCH_ACTIVE), + .features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { @@ -107,9 +107,9 @@ static const struct dpu_lm_cfg sm8450_lm[] = { LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK, &sdm845_lm_sblk, PINGPONG_1, LM_0, DSPP_1), LM_BLK("lm_2", LM_2, 0x46000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_2, LM_3, 0), + &sdm845_lm_sblk, PINGPONG_2, LM_3, DSPP_2), LM_BLK("lm_3", LM_3, 0x47000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_3, LM_2, 0), + &sdm845_lm_sblk, PINGPONG_3, LM_2, DSPP_3), LM_BLK("lm_4", LM_4, 0x48000, MIXER_SDM845_MASK, &sdm845_lm_sblk, PINGPONG_4, LM_5, 0), LM_BLK("lm_5", LM_5, 0x49000, MIXER_SDM845_MASK, @@ -126,30 +126,30 @@ static const struct dpu_dspp_cfg sm8450_dspp[] = { DSPP_BLK("dspp_3", DSPP_3, 0x5a000, DSPP_SC7180_MASK, &sm8150_dspp_sblk), }; -/* FIXME: interrupts */ + static const struct dpu_pingpong_cfg sm8450_pp[] = { - PP_BLK_TE("pingpong_0", PINGPONG_0, 0x69000, MERGE_3D_0, sdm845_pp_sblk_te, + PP_BLK_DITHER("pingpong_0", PINGPONG_0, 0x69000, MERGE_3D_0, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)), - PP_BLK_TE("pingpong_1", PINGPONG_1, 0x6a000, MERGE_3D_0, sdm845_pp_sblk_te, + -1), + PP_BLK_DITHER("pingpong_1", PINGPONG_1, 0x6a000, MERGE_3D_0, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)), - PP_BLK("pingpong_2", PINGPONG_2, 0x6b000, MERGE_3D_1, sdm845_pp_sblk, + -1), + PP_BLK_DITHER("pingpong_2", PINGPONG_2, 0x6b000, MERGE_3D_1, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)), - PP_BLK("pingpong_3", PINGPONG_3, 0x6c000, MERGE_3D_1, sdm845_pp_sblk, + -1), + PP_BLK_DITHER("pingpong_3", PINGPONG_3, 0x6c000, MERGE_3D_1, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)), - PP_BLK("pingpong_4", PINGPONG_4, 0x6d000, MERGE_3D_2, sdm845_pp_sblk, + -1), + PP_BLK_DITHER("pingpong_4", PINGPONG_4, 0x6d000, MERGE_3D_2, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), -1), - PP_BLK("pingpong_5", PINGPONG_5, 0x6e000, MERGE_3D_2, sdm845_pp_sblk, + PP_BLK_DITHER("pingpong_5", PINGPONG_5, 0x6e000, MERGE_3D_2, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), -1), - PP_BLK("pingpong_6", PINGPONG_6, 0x65800, MERGE_3D_3, sdm845_pp_sblk, + PP_BLK_DITHER("pingpong_6", PINGPONG_6, 0x65800, MERGE_3D_3, sc7280_pp_sblk, -1, -1), - PP_BLK("pingpong_7", PINGPONG_7, 0x65c00, MERGE_3D_3, sdm845_pp_sblk, + PP_BLK_DITHER("pingpong_7", PINGPONG_7, 0x65c00, MERGE_3D_3, sc7280_pp_sblk, -1, -1), }; @@ -161,11 +161,33 @@ static const struct dpu_merge_3d_cfg sm8450_merge_3d[] = { MERGE_3D_BLK("merge_3d_3", MERGE_3D_3, 0x65f00), }; +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg sm8450_dsc[] = { + DSC_BLK_1_2("dce_0_0", DSC_0, 0x80000, 0x29c, 0, dsc_sblk_0), + DSC_BLK_1_2("dce_0_1", DSC_1, 0x80000, 0x29c, 0, dsc_sblk_1), + DSC_BLK_1_2("dce_1_0", DSC_2, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_0), + DSC_BLK_1_2("dce_1_1", DSC_3, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_1), +}; + static const struct dpu_intf_cfg sm8450_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 24, 25), - INTF_BLK("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 26, 27), - INTF_BLK("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 28, 29), - INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 30, 31), + INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), + INTF_BLK_DSI_TE("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + DPU_IRQ_IDX(MDP_INTF1_7xxx_TEAR_INTR, 2)), + INTF_BLK_DSI_TE("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + DPU_IRQ_IDX(MDP_INTF2_7xxx_TEAR_INTR, 2)), + INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), }; static const struct dpu_perf_cfg sm8450_perf_data = { @@ -213,21 +235,23 @@ const struct dpu_mdss_cfg dpu_sm8450_cfg = { .dspp = sm8450_dspp, .pingpong_count = ARRAY_SIZE(sm8450_pp), .pingpong = sm8450_pp, + .dsc_count = ARRAY_SIZE(sm8450_dsc), + .dsc = sm8450_dsc, .merge_3d_count = ARRAY_SIZE(sm8450_merge_3d), .merge_3d = sm8450_merge_3d, .intf_count = ARRAY_SIZE(sm8450_intf), .intf = sm8450_intf, .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, - .reg_dma_count = 1, - .dma_cfg = &sm8450_regdma, .perf = &sm8450_perf_data, .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ BIT(MDP_SSPP_TOP0_INTR2) | \ BIT(MDP_SSPP_TOP0_HIST_INTR) | \ BIT(MDP_INTF0_7xxx_INTR) | \ BIT(MDP_INTF1_7xxx_INTR) | \ + BIT(MDP_INTF1_7xxx_TEAR_INTR) | \ BIT(MDP_INTF2_7xxx_INTR) | \ + BIT(MDP_INTF2_7xxx_TEAR_INTR) | \ BIT(MDP_INTF3_7xxx_INTR), }; diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h index 9e403034093fdc81cf04c90bce561afe14d3bca0..f17b9a7fee851ff2a40ab616913d9f3b145f0352 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h @@ -132,28 +132,28 @@ static const struct dpu_dspp_cfg sm8550_dspp[] = { &sm8150_dspp_sblk), }; static const struct dpu_pingpong_cfg sm8550_pp[] = { - PP_BLK_DIPHER("pingpong_0", PINGPONG_0, 0x69000, MERGE_3D_0, sc7280_pp_sblk, + PP_BLK_DITHER("pingpong_0", PINGPONG_0, 0x69000, MERGE_3D_0, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), -1), - PP_BLK_DIPHER("pingpong_1", PINGPONG_1, 0x6a000, MERGE_3D_0, sc7280_pp_sblk, + PP_BLK_DITHER("pingpong_1", PINGPONG_1, 0x6a000, MERGE_3D_0, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), -1), - PP_BLK_DIPHER("pingpong_2", PINGPONG_2, 0x6b000, MERGE_3D_1, sc7280_pp_sblk, + PP_BLK_DITHER("pingpong_2", PINGPONG_2, 0x6b000, MERGE_3D_1, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), -1), - PP_BLK_DIPHER("pingpong_3", PINGPONG_3, 0x6c000, MERGE_3D_1, sc7280_pp_sblk, + PP_BLK_DITHER("pingpong_3", PINGPONG_3, 0x6c000, MERGE_3D_1, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), -1), - PP_BLK_DIPHER("pingpong_4", PINGPONG_4, 0x6d000, MERGE_3D_2, sc7280_pp_sblk, + PP_BLK_DITHER("pingpong_4", PINGPONG_4, 0x6d000, MERGE_3D_2, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), -1), - PP_BLK_DIPHER("pingpong_5", PINGPONG_5, 0x6e000, MERGE_3D_2, sc7280_pp_sblk, + PP_BLK_DITHER("pingpong_5", PINGPONG_5, 0x6e000, MERGE_3D_2, sc7280_pp_sblk, DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), -1), - PP_BLK_DIPHER("pingpong_6", PINGPONG_6, 0x66000, MERGE_3D_3, sc7280_pp_sblk, + PP_BLK_DITHER("pingpong_6", PINGPONG_6, 0x66000, MERGE_3D_3, sc7280_pp_sblk, -1, -1), - PP_BLK_DIPHER("pingpong_7", PINGPONG_7, 0x66400, MERGE_3D_3, sc7280_pp_sblk, + PP_BLK_DITHER("pingpong_7", PINGPONG_7, 0x66400, MERGE_3D_3, sc7280_pp_sblk, -1, -1), }; @@ -165,12 +165,33 @@ static const struct dpu_merge_3d_cfg sm8550_merge_3d[] = { MERGE_3D_BLK("merge_3d_3", MERGE_3D_3, 0x66700), }; +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg sm8550_dsc[] = { + DSC_BLK_1_2("dce_0_0", DSC_0, 0x80000, 0x29c, 0, dsc_sblk_0), + DSC_BLK_1_2("dce_0_1", DSC_1, 0x80000, 0x29c, 0, dsc_sblk_1), + DSC_BLK_1_2("dce_1_0", DSC_2, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_0), + DSC_BLK_1_2("dce_1_1", DSC_3, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_1), +}; + static const struct dpu_intf_cfg sm8550_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 24, 25), - /* TODO TE sub-blocks for intf1 & intf2 */ - INTF_BLK("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 26, 27), - INTF_BLK("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 28, 29), - INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, MDP_SSPP_TOP0_INTR, 30, 31), + INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), + INTF_BLK_DSI_TE("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + DPU_IRQ_IDX(MDP_INTF1_7xxx_TEAR_INTR, 2)), + INTF_BLK_DSI_TE("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + DPU_IRQ_IDX(MDP_INTF2_7xxx_TEAR_INTR, 2)), + INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), }; static const struct dpu_perf_cfg sm8550_perf_data = { @@ -218,21 +239,23 @@ const struct dpu_mdss_cfg dpu_sm8550_cfg = { .dspp = sm8550_dspp, .pingpong_count = ARRAY_SIZE(sm8550_pp), .pingpong = sm8550_pp, + .dsc_count = ARRAY_SIZE(sm8550_dsc), + .dsc = sm8550_dsc, .merge_3d_count = ARRAY_SIZE(sm8550_merge_3d), .merge_3d = sm8550_merge_3d, .intf_count = ARRAY_SIZE(sm8550_intf), .intf = sm8550_intf, .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, - .reg_dma_count = 1, - .dma_cfg = &sm8450_regdma, .perf = &sm8550_perf_data, .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ BIT(MDP_SSPP_TOP0_INTR2) | \ BIT(MDP_SSPP_TOP0_HIST_INTR) | \ BIT(MDP_INTF0_7xxx_INTR) | \ BIT(MDP_INTF1_7xxx_INTR) | \ + BIT(MDP_INTF1_7xxx_TEAR_INTR) | \ BIT(MDP_INTF2_7xxx_INTR) | \ + BIT(MDP_INTF2_7xxx_TEAR_INTR) | \ BIT(MDP_INTF3_7xxx_INTR), }; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c index cc66ddffe6723b748134fd2a3def11096ff033e0..1edf2b6b0a262e9f0599c3825a4b4ff02ee01dfd 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c @@ -1392,7 +1392,7 @@ DEFINE_SHOW_ATTRIBUTE(_dpu_debugfs_status); static int dpu_crtc_debugfs_state_show(struct seq_file *s, void *v) { - struct drm_crtc *crtc = (struct drm_crtc *) s->private; + struct drm_crtc *crtc = s->private; struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); seq_printf(s, "client type: %d\n", dpu_crtc_get_client_type(crtc)); @@ -1463,6 +1463,8 @@ static const struct drm_crtc_helper_funcs dpu_crtc_helper_funcs = { struct drm_crtc *dpu_crtc_init(struct drm_device *dev, struct drm_plane *plane, struct drm_plane *cursor) { + struct msm_drm_private *priv = dev->dev_private; + struct dpu_kms *dpu_kms = to_dpu_kms(priv->kms); struct drm_crtc *crtc = NULL; struct dpu_crtc *dpu_crtc = NULL; int i, ret; @@ -1494,7 +1496,8 @@ struct drm_crtc *dpu_crtc_init(struct drm_device *dev, struct drm_plane *plane, drm_crtc_helper_add(crtc, &dpu_crtc_helper_funcs); - drm_crtc_enable_color_mgmt(crtc, 0, true, 0); + if (dpu_kms->catalog->dspp_count) + drm_crtc_enable_color_mgmt(crtc, 0, true, 0); /* save user friendly CRTC name for later */ snprintf(dpu_crtc->name, DPU_CRTC_NAME_SIZE, "crtc%u", crtc->base.id); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c index 1dc5dbe5857233a8ba89dad7a189c49f58d56e81..493905a5b63a6075102f11582bb4ca59853d920e 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c @@ -339,7 +339,8 @@ void dpu_encoder_helper_report_irq_timeout(struct dpu_encoder_phys *phys_enc, DRM_ERROR("irq timeout id=%u, intf_mode=%s intf=%d wb=%d, pp=%d, intr=%d\n", DRMID(phys_enc->parent), dpu_encoder_helper_get_intf_type(phys_enc->intf_mode), - phys_enc->intf_idx - INTF_0, phys_enc->wb_idx - WB_0, + phys_enc->hw_intf ? phys_enc->hw_intf->idx - INTF_0 : -1, + phys_enc->hw_wb ? phys_enc->hw_wb->idx - WB_0 : -1, phys_enc->hw_pp->idx - PINGPONG_0, intr_idx); dpu_encoder_frame_done_callback(phys_enc->parent, phys_enc, @@ -495,7 +496,7 @@ void dpu_encoder_helper_split_config( hw_mdptop = phys_enc->hw_mdptop; disp_info = &dpu_enc->disp_info; - if (disp_info->intf_type != DRM_MODE_ENCODER_DSI) + if (disp_info->intf_type != INTF_DSI) return; /** @@ -666,6 +667,7 @@ static void _dpu_encoder_update_vsync_source(struct dpu_encoder_virt *dpu_enc, struct dpu_kms *dpu_kms; struct dpu_hw_mdp *hw_mdptop; struct drm_encoder *drm_enc; + struct dpu_encoder_phys *phys_enc; int i; if (!dpu_enc || !disp_info) { @@ -696,12 +698,22 @@ static void _dpu_encoder_update_vsync_source(struct dpu_encoder_virt *dpu_enc, vsync_cfg.ppnumber[i] = dpu_enc->hw_pp[i]->idx; vsync_cfg.pp_count = dpu_enc->num_phys_encs; + vsync_cfg.frame_rate = drm_mode_vrefresh(&dpu_enc->base.crtc->state->adjusted_mode); + if (disp_info->is_te_using_watchdog_timer) vsync_cfg.vsync_source = DPU_VSYNC_SOURCE_WD_TIMER_0; else vsync_cfg.vsync_source = DPU_VSYNC0_SOURCE_GPIO; hw_mdptop->ops.setup_vsync_source(hw_mdptop, &vsync_cfg); + + for (i = 0; i < dpu_enc->num_phys_encs; i++) { + phys_enc = dpu_enc->phys_encs[i]; + + if (phys_enc->has_intf_te && phys_enc->hw_intf->ops.vsync_sel) + phys_enc->hw_intf->ops.vsync_sel(phys_enc->hw_intf, + vsync_cfg.vsync_source); + } } } @@ -1127,7 +1139,7 @@ static void _dpu_encoder_virt_enable_helper(struct drm_encoder *drm_enc) } - if (dpu_enc->disp_info.intf_type == DRM_MODE_ENCODER_TMDS && + if (dpu_enc->disp_info.intf_type == INTF_DP && dpu_enc->cur_master->hw_mdptop && dpu_enc->cur_master->hw_mdptop->ops.intf_audio_select) dpu_enc->cur_master->hw_mdptop->ops.intf_audio_select( @@ -1135,7 +1147,7 @@ static void _dpu_encoder_virt_enable_helper(struct drm_encoder *drm_enc) _dpu_encoder_update_vsync_source(dpu_enc, &dpu_enc->disp_info); - if (dpu_enc->disp_info.intf_type == DRM_MODE_ENCODER_DSI && + if (dpu_enc->disp_info.intf_type == INTF_DSI && !WARN_ON(dpu_enc->num_phys_encs == 0)) { unsigned bpc = dpu_enc->connector->display_info.bpc; for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) { @@ -1258,38 +1270,23 @@ static void dpu_encoder_virt_atomic_disable(struct drm_encoder *drm_enc, mutex_unlock(&dpu_enc->enc_lock); } -static enum dpu_intf dpu_encoder_get_intf(const struct dpu_mdss_cfg *catalog, +static struct dpu_hw_intf *dpu_encoder_get_intf(const struct dpu_mdss_cfg *catalog, + struct dpu_rm *dpu_rm, enum dpu_intf_type type, u32 controller_id) { int i = 0; if (type == INTF_WB) - return INTF_MAX; + return NULL; for (i = 0; i < catalog->intf_count; i++) { if (catalog->intf[i].type == type && catalog->intf[i].controller_id == controller_id) { - return catalog->intf[i].id; + return dpu_rm_get_intf(dpu_rm, catalog->intf[i].id); } } - return INTF_MAX; -} - -static enum dpu_wb dpu_encoder_get_wb(const struct dpu_mdss_cfg *catalog, - enum dpu_intf_type type, u32 controller_id) -{ - int i = 0; - - if (type != INTF_WB) - return WB_MAX; - - for (i = 0; i < catalog->wb_count; i++) { - if (catalog->wb[i].id == controller_id) - return catalog->wb[i].id; - } - - return WB_MAX; + return NULL; } void dpu_encoder_vblank_callback(struct drm_encoder *drm_enc, @@ -1408,7 +1405,8 @@ void dpu_encoder_frame_done_callback( */ trace_dpu_enc_frame_done_cb_not_busy(DRMID(drm_enc), event, dpu_encoder_helper_get_intf_type(ready_phys->intf_mode), - ready_phys->intf_idx, ready_phys->wb_idx); + ready_phys->hw_intf ? ready_phys->hw_intf->idx : -1, + ready_phys->hw_wb ? ready_phys->hw_wb->idx : -1); return; } @@ -1488,7 +1486,8 @@ static void _dpu_encoder_trigger_flush(struct drm_encoder *drm_enc, trace_dpu_enc_trigger_flush(DRMID(drm_enc), dpu_encoder_helper_get_intf_type(phys->intf_mode), - phys->intf_idx, phys->wb_idx, + phys->hw_intf ? phys->hw_intf->idx : -1, + phys->hw_wb ? phys->hw_wb->idx : -1, pending_kickoff_cnt, ctl->idx, extra_flush_bits, ret); } @@ -1823,7 +1822,8 @@ dpu_encoder_dsc_initial_line_calc(struct drm_dsc_config *dsc, return DIV_ROUND_UP(total_pixels, dsc->slice_width); } -static void dpu_encoder_dsc_pipe_cfg(struct dpu_hw_dsc *hw_dsc, +static void dpu_encoder_dsc_pipe_cfg(struct dpu_hw_ctl *ctl, + struct dpu_hw_dsc *hw_dsc, struct dpu_hw_pingpong *hw_pp, struct drm_dsc_config *dsc, u32 common_mode, @@ -1839,10 +1839,13 @@ static void dpu_encoder_dsc_pipe_cfg(struct dpu_hw_dsc *hw_dsc, hw_pp->ops.setup_dsc(hw_pp); if (hw_dsc->ops.dsc_bind_pingpong_blk) - hw_dsc->ops.dsc_bind_pingpong_blk(hw_dsc, true, hw_pp->idx); + hw_dsc->ops.dsc_bind_pingpong_blk(hw_dsc, hw_pp->idx); if (hw_pp->ops.enable_dsc) hw_pp->ops.enable_dsc(hw_pp); + + if (ctl->ops.update_pending_flush_dsc) + ctl->ops.update_pending_flush_dsc(ctl, hw_dsc->idx); } static void dpu_encoder_prep_dsc(struct dpu_encoder_virt *dpu_enc, @@ -1850,6 +1853,7 @@ static void dpu_encoder_prep_dsc(struct dpu_encoder_virt *dpu_enc, { /* coding only for 2LM, 2enc, 1 dsc config */ struct dpu_encoder_phys *enc_master = dpu_enc->cur_master; + struct dpu_hw_ctl *ctl = enc_master->hw_ctl; struct dpu_hw_dsc *hw_dsc[MAX_CHANNELS_PER_ENC]; struct dpu_hw_pingpong *hw_pp[MAX_CHANNELS_PER_ENC]; int this_frame_slices; @@ -1887,7 +1891,8 @@ static void dpu_encoder_prep_dsc(struct dpu_encoder_virt *dpu_enc, initial_lines = dpu_encoder_dsc_initial_line_calc(dsc, enc_ip_w); for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) - dpu_encoder_dsc_pipe_cfg(hw_dsc[i], hw_pp[i], dsc, dsc_common_mode, initial_lines); + dpu_encoder_dsc_pipe_cfg(ctl, hw_dsc[i], hw_pp[i], + dsc, dsc_common_mode, initial_lines); } void dpu_encoder_prepare_for_kickoff(struct drm_encoder *drm_enc) @@ -1977,7 +1982,7 @@ void dpu_encoder_kickoff(struct drm_encoder *drm_enc) phys->ops.handle_post_kickoff(phys); } - if (dpu_enc->disp_info.intf_type == DRM_MODE_ENCODER_DSI && + if (dpu_enc->disp_info.intf_type == INTF_DSI && !dpu_encoder_vsync_time(drm_enc, &wakeup_time)) { trace_dpu_enc_early_kickoff(DRMID(drm_enc), ktime_to_ms(wakeup_time)); @@ -2019,6 +2024,41 @@ static void dpu_encoder_helper_reset_mixers(struct dpu_encoder_phys *phys_enc) } } +static void dpu_encoder_dsc_pipe_clr(struct dpu_hw_ctl *ctl, + struct dpu_hw_dsc *hw_dsc, + struct dpu_hw_pingpong *hw_pp) +{ + if (hw_dsc->ops.dsc_disable) + hw_dsc->ops.dsc_disable(hw_dsc); + + if (hw_pp->ops.disable_dsc) + hw_pp->ops.disable_dsc(hw_pp); + + if (hw_dsc->ops.dsc_bind_pingpong_blk) + hw_dsc->ops.dsc_bind_pingpong_blk(hw_dsc, PINGPONG_NONE); + + if (ctl->ops.update_pending_flush_dsc) + ctl->ops.update_pending_flush_dsc(ctl, hw_dsc->idx); +} + +static void dpu_encoder_unprep_dsc(struct dpu_encoder_virt *dpu_enc) +{ + /* coding only for 2LM, 2enc, 1 dsc config */ + struct dpu_encoder_phys *enc_master = dpu_enc->cur_master; + struct dpu_hw_ctl *ctl = enc_master->hw_ctl; + struct dpu_hw_dsc *hw_dsc[MAX_CHANNELS_PER_ENC]; + struct dpu_hw_pingpong *hw_pp[MAX_CHANNELS_PER_ENC]; + int i; + + for (i = 0; i < MAX_CHANNELS_PER_ENC; i++) { + hw_pp[i] = dpu_enc->hw_pp[i]; + hw_dsc[i] = dpu_enc->hw_dsc[i]; + + if (hw_pp[i] && hw_dsc[i]) + dpu_encoder_dsc_pipe_clr(ctl, hw_dsc[i], hw_pp[i]); + } +} + void dpu_encoder_helper_phys_cleanup(struct dpu_encoder_phys *phys_enc) { struct dpu_hw_ctl *ctl = phys_enc->hw_ctl; @@ -2040,8 +2080,7 @@ void dpu_encoder_helper_phys_cleanup(struct dpu_encoder_phys *phys_enc) if (phys_enc->hw_wb) { /* disable the PP block */ if (phys_enc->hw_wb->ops.bind_pingpong_blk) - phys_enc->hw_wb->ops.bind_pingpong_blk(phys_enc->hw_wb, false, - phys_enc->hw_pp->idx); + phys_enc->hw_wb->ops.bind_pingpong_blk(phys_enc->hw_wb, PINGPONG_NONE); /* mark WB flush as pending */ if (phys_enc->hw_ctl->ops.update_pending_flush_wb) @@ -2050,8 +2089,8 @@ void dpu_encoder_helper_phys_cleanup(struct dpu_encoder_phys *phys_enc) for (i = 0; i < dpu_enc->num_phys_encs; i++) { if (dpu_enc->phys_encs[i] && phys_enc->hw_intf->ops.bind_pingpong_blk) phys_enc->hw_intf->ops.bind_pingpong_blk( - dpu_enc->phys_encs[i]->hw_intf, false, - dpu_enc->phys_encs[i]->hw_pp->idx); + dpu_enc->phys_encs[i]->hw_intf, + PINGPONG_NONE); /* mark INTF flush as pending */ if (phys_enc->hw_ctl->ops.update_pending_flush_intf) @@ -2069,8 +2108,12 @@ void dpu_encoder_helper_phys_cleanup(struct dpu_encoder_phys *phys_enc) phys_enc->hw_pp->merge_3d->idx); } + if (dpu_enc->dsc) + dpu_encoder_unprep_dsc(dpu_enc); + intf_cfg.stream_sel = 0; /* Don't care value for video mode */ intf_cfg.mode_3d = dpu_encoder_helper_get_3d_blend_mode(phys_enc); + intf_cfg.dsc = dpu_encoder_helper_get_dsc(phys_enc); if (phys_enc->hw_intf) intf_cfg.intf = phys_enc->hw_intf->idx; @@ -2099,7 +2142,8 @@ static int _dpu_encoder_status_show(struct seq_file *s, void *data) struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; seq_printf(s, "intf:%d wb:%d vsync:%8d underrun:%8d ", - phys->intf_idx - INTF_0, phys->wb_idx - WB_0, + phys->hw_intf ? phys->hw_intf->idx - INTF_0 : -1, + phys->hw_wb ? phys->hw_wb->idx - WB_0 : -1, atomic_read(&phys->vsync_cnt), atomic_read(&phys->underrun_cnt)); @@ -2115,16 +2159,15 @@ DEFINE_SHOW_ATTRIBUTE(_dpu_encoder_status); static int _dpu_encoder_init_debugfs(struct drm_encoder *drm_enc) { struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); - int i; - char name[DPU_NAME_SIZE]; + char name[12]; if (!drm_enc->dev) { DPU_ERROR("invalid encoder or kms\n"); return -EINVAL; } - snprintf(name, DPU_NAME_SIZE, "encoder%u", drm_enc->base.id); + snprintf(name, sizeof(name), "encoder%u", drm_enc->base.id); /* create overall sub-directory for the encoder */ dpu_enc->debugfs_root = debugfs_create_dir(name, @@ -2134,12 +2177,6 @@ static int _dpu_encoder_init_debugfs(struct drm_encoder *drm_enc) debugfs_create_file("status", 0600, dpu_enc->debugfs_root, dpu_enc, &_dpu_encoder_status_fops); - for (i = 0; i < dpu_enc->num_phys_encs; i++) - if (dpu_enc->phys_encs[i]->ops.late_register) - dpu_enc->phys_encs[i]->ops.late_register( - dpu_enc->phys_encs[i], - dpu_enc->debugfs_root); - return 0; } #else @@ -2182,7 +2219,7 @@ static int dpu_encoder_virt_add_phys_encs( } - if (disp_info->intf_type == DRM_MODE_ENCODER_VIRTUAL) { + if (disp_info->intf_type == INTF_WB) { enc = dpu_encoder_phys_wb_init(params); if (IS_ERR(enc)) { @@ -2231,7 +2268,6 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc, { int ret = 0; int i = 0; - enum dpu_intf_type intf_type = INTF_NONE; struct dpu_enc_phys_init_params phys_params; if (!dpu_enc) { @@ -2246,23 +2282,11 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc, phys_params.parent = &dpu_enc->base; phys_params.enc_spinlock = &dpu_enc->enc_spinlock; - switch (disp_info->intf_type) { - case DRM_MODE_ENCODER_DSI: - intf_type = INTF_DSI; - break; - case DRM_MODE_ENCODER_TMDS: - intf_type = INTF_DP; - break; - case DRM_MODE_ENCODER_VIRTUAL: - intf_type = INTF_WB; - break; - } - WARN_ON(disp_info->num_of_h_tiles < 1); DPU_DEBUG("dsi_info->num_of_h_tiles %d\n", disp_info->num_of_h_tiles); - if (disp_info->intf_type != DRM_MODE_ENCODER_VIRTUAL) + if (disp_info->intf_type != INTF_WB) dpu_enc->idle_pc_supported = dpu_kms->catalog->caps->has_idle_pc; @@ -2289,58 +2313,31 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc, DPU_DEBUG("h_tile_instance %d = %d, split_role %d\n", i, controller_id, phys_params.split_role); - phys_params.intf_idx = dpu_encoder_get_intf(dpu_kms->catalog, - intf_type, - controller_id); - - phys_params.wb_idx = dpu_encoder_get_wb(dpu_kms->catalog, - intf_type, controller_id); - /* - * The phys_params might represent either an INTF or a WB unit, but not - * both of them at the same time. - */ - if ((phys_params.intf_idx == INTF_MAX) && - (phys_params.wb_idx == WB_MAX)) { - DPU_ERROR_ENC(dpu_enc, "could not get intf or wb: type %d, id %d\n", - intf_type, controller_id); - ret = -EINVAL; - } - - if ((phys_params.intf_idx != INTF_MAX) && - (phys_params.wb_idx != WB_MAX)) { - DPU_ERROR_ENC(dpu_enc, "both intf and wb present: type %d, id %d\n", - intf_type, controller_id); - ret = -EINVAL; - } - - if (!ret) { - ret = dpu_encoder_virt_add_phys_encs(disp_info, - dpu_enc, &phys_params); - if (ret) - DPU_ERROR_ENC(dpu_enc, "failed to add phys encs\n"); - } - } - - for (i = 0; i < dpu_enc->num_phys_encs; i++) { - struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; - atomic_set(&phys->vsync_cnt, 0); - atomic_set(&phys->underrun_cnt, 0); - - if (phys->intf_idx >= INTF_0 && phys->intf_idx < INTF_MAX) - phys->hw_intf = dpu_rm_get_intf(&dpu_kms->rm, phys->intf_idx); + phys_params.hw_intf = dpu_encoder_get_intf(dpu_kms->catalog, &dpu_kms->rm, + disp_info->intf_type, + controller_id); - if (phys->wb_idx >= WB_0 && phys->wb_idx < WB_MAX) - phys->hw_wb = dpu_rm_get_wb(&dpu_kms->rm, phys->wb_idx); + if (disp_info->intf_type == INTF_WB && controller_id < WB_MAX) + phys_params.hw_wb = dpu_rm_get_wb(&dpu_kms->rm, controller_id); - if (!phys->hw_intf && !phys->hw_wb) { + if (!phys_params.hw_intf && !phys_params.hw_wb) { DPU_ERROR_ENC(dpu_enc, "no intf or wb block assigned at idx: %d\n", i); ret = -EINVAL; + break; } - if (phys->hw_intf && phys->hw_wb) { + if (phys_params.hw_intf && phys_params.hw_wb) { DPU_ERROR_ENC(dpu_enc, "invalid phys both intf and wb block at idx: %d\n", i); ret = -EINVAL; + break; + } + + ret = dpu_encoder_virt_add_phys_encs(disp_info, + dpu_enc, &phys_params); + if (ret) { + DPU_ERROR_ENC(dpu_enc, "failed to add phys encs\n"); + break; } } @@ -2390,7 +2387,8 @@ static const struct drm_encoder_funcs dpu_encoder_funcs = { .early_unregister = dpu_encoder_early_unregister, }; -int dpu_encoder_setup(struct drm_device *dev, struct drm_encoder *enc, +struct drm_encoder *dpu_encoder_init(struct drm_device *dev, + int drm_enc_mode, struct msm_display_info *disp_info) { struct msm_drm_private *priv = dev->dev_private; @@ -2399,7 +2397,23 @@ int dpu_encoder_setup(struct drm_device *dev, struct drm_encoder *enc, struct dpu_encoder_virt *dpu_enc = NULL; int ret = 0; - dpu_enc = to_dpu_encoder_virt(enc); + dpu_enc = devm_kzalloc(dev->dev, sizeof(*dpu_enc), GFP_KERNEL); + if (!dpu_enc) + return ERR_PTR(-ENOMEM); + + ret = drm_encoder_init(dev, &dpu_enc->base, &dpu_encoder_funcs, + drm_enc_mode, NULL); + if (ret) { + devm_kfree(dev->dev, dpu_enc); + return ERR_PTR(ret); + } + + drm_encoder_helper_add(&dpu_enc->base, &dpu_encoder_helper_funcs); + + spin_lock_init(&dpu_enc->enc_spinlock); + dpu_enc->enabled = false; + mutex_init(&dpu_enc->enc_lock); + mutex_init(&dpu_enc->rc_lock); ret = dpu_encoder_setup_display(dpu_enc, dpu_kms, disp_info); if (ret) @@ -2409,11 +2423,11 @@ int dpu_encoder_setup(struct drm_device *dev, struct drm_encoder *enc, timer_setup(&dpu_enc->frame_done_timer, dpu_encoder_frame_done_timeout, 0); - if (disp_info->intf_type == DRM_MODE_ENCODER_DSI) + if (disp_info->intf_type == INTF_DSI) timer_setup(&dpu_enc->vsync_event_timer, dpu_encoder_vsync_event_handler, 0); - else if (disp_info->intf_type == DRM_MODE_ENCODER_TMDS) + else if (disp_info->intf_type == INTF_DP) dpu_enc->wide_bus_en = msm_dp_wide_bus_available( priv->dp[disp_info->h_tile_instance[0]]); @@ -2428,44 +2442,14 @@ int dpu_encoder_setup(struct drm_device *dev, struct drm_encoder *enc, DPU_DEBUG_ENC(dpu_enc, "created\n"); - return ret; + return &dpu_enc->base; fail: DPU_ERROR("failed to create encoder\n"); if (drm_enc) dpu_encoder_destroy(drm_enc); - return ret; - - -} - -struct drm_encoder *dpu_encoder_init(struct drm_device *dev, - int drm_enc_mode) -{ - struct dpu_encoder_virt *dpu_enc = NULL; - int rc = 0; - - dpu_enc = devm_kzalloc(dev->dev, sizeof(*dpu_enc), GFP_KERNEL); - if (!dpu_enc) - return ERR_PTR(-ENOMEM); - - - rc = drm_encoder_init(dev, &dpu_enc->base, &dpu_encoder_funcs, - drm_enc_mode, NULL); - if (rc) { - devm_kfree(dev->dev, dpu_enc); - return ERR_PTR(rc); - } - - drm_encoder_helper_add(&dpu_enc->base, &dpu_encoder_helper_funcs); - - spin_lock_init(&dpu_enc->enc_spinlock); - dpu_enc->enabled = false; - mutex_init(&dpu_enc->enc_lock); - mutex_init(&dpu_enc->rc_lock); - - return &dpu_enc->base; + return ERR_PTR(ret); } int dpu_encoder_wait_for_event(struct drm_encoder *drm_enc, @@ -2539,3 +2523,30 @@ unsigned int dpu_encoder_helper_get_dsc(struct dpu_encoder_phys *phys_enc) return dpu_enc->dsc_mask; } + +void dpu_encoder_phys_init(struct dpu_encoder_phys *phys_enc, + struct dpu_enc_phys_init_params *p) +{ + int i; + + phys_enc->hw_mdptop = p->dpu_kms->hw_mdp; + phys_enc->hw_intf = p->hw_intf; + phys_enc->hw_wb = p->hw_wb; + phys_enc->parent = p->parent; + phys_enc->dpu_kms = p->dpu_kms; + phys_enc->split_role = p->split_role; + phys_enc->enc_spinlock = p->enc_spinlock; + phys_enc->enable_state = DPU_ENC_DISABLED; + + for (i = 0; i < ARRAY_SIZE(phys_enc->irq); i++) + phys_enc->irq[i] = -EINVAL; + + atomic_set(&phys_enc->vblank_refcount, 0); + atomic_set(&phys_enc->pending_kickoff_cnt, 0); + atomic_set(&phys_enc->pending_ctlstart_cnt, 0); + + atomic_set(&phys_enc->vsync_cnt, 0); + atomic_set(&phys_enc->underrun_cnt, 0); + + init_waitqueue_head(&phys_enc->pending_kickoff_wq); +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h index 2c9ef8d1b87799802c8acc586b054cd9a65f919f..90e1925d7770f9590e9d5492ff1e9fc6cefe7d13 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h @@ -21,7 +21,7 @@ /** * struct msm_display_info - defines display properties - * @intf_type: DRM_MODE_ENCODER_ type + * @intf_type: INTF_ type * @num_of_h_tiles: Number of horizontal tiles in case of split interface * @h_tile_instance: Controller instance used per tile. Number of elements is * based on num_of_h_tiles @@ -31,7 +31,7 @@ * @dsc: DSC configuration data for DSC-enabled displays */ struct msm_display_info { - int intf_type; + enum dpu_intf_type intf_type; uint32_t num_of_h_tiles; uint32_t h_tile_instance[MAX_H_TILES_PER_DISPLAY]; bool is_cmd_mode; @@ -130,20 +130,12 @@ void dpu_encoder_virt_runtime_resume(struct drm_encoder *encoder); /** * dpu_encoder_init - initialize virtual encoder object * @dev: Pointer to drm device structure + * @drm_enc_mode: corresponding DRM_MODE_ENCODER_* constant * @disp_info: Pointer to display information structure * Returns: Pointer to newly created drm encoder */ -struct drm_encoder *dpu_encoder_init( - struct drm_device *dev, - int drm_enc_mode); - -/** - * dpu_encoder_setup - setup dpu_encoder for the display probed - * @dev: Pointer to drm device structure - * @enc: Pointer to the drm_encoder - * @disp_info: Pointer to the display info - */ -int dpu_encoder_setup(struct drm_device *dev, struct drm_encoder *enc, +struct drm_encoder *dpu_encoder_init(struct drm_device *dev, + int drm_enc_mode, struct msm_display_info *disp_info); /** diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h index 1d434b22180d9e152f5d88457c5884a0d279912b..d48558ede488d51e5b7bf36d58aabeb346f36103 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h @@ -63,7 +63,6 @@ struct dpu_encoder_phys; /** * struct dpu_encoder_phys_ops - Interface the physical encoders provide to * the containing virtual encoder. - * @late_register: DRM Call. Add Userspace interfaces, debugfs. * @prepare_commit: MSM Atomic Call, start of atomic commit sequence * @is_master: Whether this phys_enc is the current master * encoder. Can be switched at enable time. Based @@ -93,8 +92,6 @@ struct dpu_encoder_phys; */ struct dpu_encoder_phys_ops { - int (*late_register)(struct dpu_encoder_phys *encoder, - struct dentry *debugfs_root); void (*prepare_commit)(struct dpu_encoder_phys *encoder); bool (*is_master)(struct dpu_encoder_phys *encoder); void (*atomic_mode_set)(struct dpu_encoder_phys *encoder, @@ -129,10 +126,10 @@ struct dpu_encoder_phys_ops { /** * enum dpu_intr_idx - dpu encoder interrupt index * @INTR_IDX_VSYNC: Vsync interrupt for video mode panel - * @INTR_IDX_PINGPONG: Pingpong done unterrupt for cmd mode panel - * @INTR_IDX_UNDERRUN: Underrun unterrupt for video and cmd mode panel - * @INTR_IDX_RDPTR: Readpointer done unterrupt for cmd mode panel - * @INTR_IDX_WB_DONE: Writeback fone interrupt for virtual connector + * @INTR_IDX_PINGPONG: Pingpong done interrupt for cmd mode panel + * @INTR_IDX_UNDERRUN: Underrun interrupt for video and cmd mode panel + * @INTR_IDX_RDPTR: Readpointer done interrupt for cmd mode panel + * @INTR_IDX_WB_DONE: Writeback done interrupt for virtual connector */ enum dpu_intr_idx { INTR_IDX_VSYNC, @@ -161,8 +158,6 @@ enum dpu_intr_idx { * @enabled: Whether the encoder has enabled and running a mode * @split_role: Role to play in a split-panel configuration * @intf_mode: Interface mode - * @intf_idx: Interface index on dpu hardware - * @wb_idx: Writeback index on dpu hardware * @enc_spinlock: Virtual-Encoder-Wide Spin Lock for IRQ purposes * @enable_state: Enable state tracking * @vblank_refcount: Reference count of vblank request @@ -176,6 +171,7 @@ enum dpu_intr_idx { * pending. * @pending_kickoff_wq: Wait queue for blocking until kickoff completes * @irq: IRQ indices + * @has_intf_te: Interface TE configuration support */ struct dpu_encoder_phys { struct drm_encoder *parent; @@ -189,8 +185,6 @@ struct dpu_encoder_phys { struct drm_display_mode cached_mode; enum dpu_enc_split_role split_role; enum dpu_intf_mode intf_mode; - enum dpu_intf intf_idx; - enum dpu_wb wb_idx; spinlock_t *enc_spinlock; enum dpu_enc_enable_state enable_state; atomic_t vblank_refcount; @@ -200,6 +194,7 @@ struct dpu_encoder_phys { atomic_t pending_kickoff_cnt; wait_queue_head_t pending_kickoff_wq; int irq[INTR_IDX_MAX]; + bool has_intf_te; }; static inline int dpu_encoder_phys_inc_pending(struct dpu_encoder_phys *phys) @@ -256,16 +251,16 @@ struct dpu_encoder_phys_cmd { * @parent: Pointer to the containing virtual encoder * @parent_ops: Callbacks exposed by the parent to the phys_enc * @split_role: Role to play in a split-panel configuration - * @intf_idx: Interface index this phys_enc will control - * @wb_idx: Writeback index this phys_enc will control + * @hw_intf: Hardware interface to the intf registers + * @hw_wb: Hardware interface to the wb registers * @enc_spinlock: Virtual-Encoder-Wide Spin Lock for IRQ purposes */ struct dpu_enc_phys_init_params { struct dpu_kms *dpu_kms; struct drm_encoder *parent; enum dpu_enc_split_role split_role; - enum dpu_intf intf_idx; - enum dpu_wb wb_idx; + struct dpu_hw_intf *hw_intf; + struct dpu_hw_wb *hw_wb; spinlock_t *enc_spinlock; }; @@ -405,4 +400,7 @@ void dpu_encoder_frame_done_callback( struct drm_encoder *drm_enc, struct dpu_encoder_phys *ready_phys, u32 event); +void dpu_encoder_phys_init(struct dpu_encoder_phys *phys, + struct dpu_enc_phys_init_params *p); + #endif /* __dpu_encoder_phys_H__ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c index 74470d068622e0d42f0c022a590ed3b8d7b85bf5..b856c6286c85acc15393134366634e03d52689c5 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c @@ -16,12 +16,12 @@ #define DPU_DEBUG_CMDENC(e, fmt, ...) DPU_DEBUG("enc%d intf%d " fmt, \ (e) && (e)->base.parent ? \ (e)->base.parent->base.id : -1, \ - (e) ? (e)->base.intf_idx - INTF_0 : -1, ##__VA_ARGS__) + (e) ? (e)->base.hw_intf->idx - INTF_0 : -1, ##__VA_ARGS__) #define DPU_ERROR_CMDENC(e, fmt, ...) DPU_ERROR("enc%d intf%d " fmt, \ (e) && (e)->base.parent ? \ (e)->base.parent->base.id : -1, \ - (e) ? (e)->base.intf_idx - INTF_0 : -1, ##__VA_ARGS__) + (e) ? (e)->base.hw_intf->idx - INTF_0 : -1, ##__VA_ARGS__) #define to_dpu_encoder_phys_cmd(x) \ container_of(x, struct dpu_encoder_phys_cmd, base) @@ -36,10 +36,6 @@ #define DEFAULT_TEARCHECK_SYNC_THRESH_START 4 #define DEFAULT_TEARCHECK_SYNC_THRESH_CONTINUE 4 -#define DPU_ENC_WR_PTR_START_TIMEOUT_US 20000 - -#define DPU_ENC_MAX_POLL_TIMEOUT_US 2000 - static void dpu_encoder_phys_cmd_enable_te(struct dpu_encoder_phys *phys_enc); static bool dpu_encoder_phys_cmd_is_master(struct dpu_encoder_phys *phys_enc) @@ -59,7 +55,7 @@ static void _dpu_encoder_phys_cmd_update_intf_cfg( if (!ctl->ops.setup_intf_cfg) return; - intf_cfg.intf = phys_enc->intf_idx; + intf_cfg.intf = phys_enc->hw_intf->idx; intf_cfg.intf_mode_sel = DPU_CTL_MODE_SEL_CMD; intf_cfg.stream_sel = cmd_enc->stream_sel; intf_cfg.mode_3d = dpu_encoder_helper_get_3d_blend_mode(phys_enc); @@ -70,8 +66,10 @@ static void _dpu_encoder_phys_cmd_update_intf_cfg( if (test_bit(DPU_CTL_ACTIVE_CFG, &ctl->caps->features) && phys_enc->hw_intf->ops.bind_pingpong_blk) phys_enc->hw_intf->ops.bind_pingpong_blk( phys_enc->hw_intf, - true, phys_enc->hw_pp->idx); + + if (intf_cfg.dsc != 0 && phys_enc->hw_intf->ops.enable_compression) + phys_enc->hw_intf->ops.enable_compression(phys_enc->hw_intf); } static void dpu_encoder_phys_cmd_pp_tx_done_irq(void *arg, int irq_idx) @@ -101,13 +99,18 @@ static void dpu_encoder_phys_cmd_pp_tx_done_irq(void *arg, int irq_idx) DPU_ATRACE_END("pp_done_irq"); } -static void dpu_encoder_phys_cmd_pp_rd_ptr_irq(void *arg, int irq_idx) +static void dpu_encoder_phys_cmd_te_rd_ptr_irq(void *arg, int irq_idx) { struct dpu_encoder_phys *phys_enc = arg; struct dpu_encoder_phys_cmd *cmd_enc; - if (!phys_enc->hw_pp) - return; + if (phys_enc->has_intf_te) { + if (!phys_enc->hw_intf) + return; + } else { + if (!phys_enc->hw_pp) + return; + } DPU_ATRACE_BEGIN("rd_ptr_irq"); cmd_enc = to_dpu_encoder_phys_cmd(phys_enc); @@ -148,7 +151,10 @@ static void dpu_encoder_phys_cmd_atomic_mode_set( phys_enc->irq[INTR_IDX_PINGPONG] = phys_enc->hw_pp->caps->intr_done; - phys_enc->irq[INTR_IDX_RDPTR] = phys_enc->hw_pp->caps->intr_rdptr; + if (phys_enc->has_intf_te) + phys_enc->irq[INTR_IDX_RDPTR] = phys_enc->hw_intf->cap->intr_tear_rd_ptr; + else + phys_enc->irq[INTR_IDX_RDPTR] = phys_enc->hw_pp->caps->intr_rdptr; phys_enc->irq[INTR_IDX_UNDERRUN] = phys_enc->hw_intf->cap->intr_underrun; } @@ -259,7 +265,7 @@ static int dpu_encoder_phys_cmd_control_vblank_irq( if (enable && atomic_inc_return(&phys_enc->vblank_refcount) == 1) ret = dpu_core_irq_register_callback(phys_enc->dpu_kms, phys_enc->irq[INTR_IDX_RDPTR], - dpu_encoder_phys_cmd_pp_rd_ptr_irq, + dpu_encoder_phys_cmd_te_rd_ptr_irq, phys_enc); else if (!enable && atomic_dec_return(&phys_enc->vblank_refcount) == 0) ret = dpu_core_irq_unregister_callback(phys_enc->dpu_kms, @@ -320,23 +326,29 @@ static void dpu_encoder_phys_cmd_tearcheck_config( struct dpu_hw_tear_check tc_cfg = { 0 }; struct drm_display_mode *mode; bool tc_enable = true; - u32 vsync_hz; + unsigned long vsync_hz; struct dpu_kms *dpu_kms; - if (!phys_enc->hw_pp) { - DPU_ERROR("invalid encoder\n"); - return; - } - mode = &phys_enc->cached_mode; + if (phys_enc->has_intf_te) { + if (!phys_enc->hw_intf || + !phys_enc->hw_intf->ops.enable_tearcheck) { + DPU_DEBUG_CMDENC(cmd_enc, "tearcheck not supported\n"); + return; + } - DPU_DEBUG_CMDENC(cmd_enc, "pp %d\n", phys_enc->hw_pp->idx - PINGPONG_0); + DPU_DEBUG_CMDENC(cmd_enc, ""); + } else { + if (!phys_enc->hw_pp || + !phys_enc->hw_pp->ops.enable_tearcheck) { + DPU_DEBUG_CMDENC(cmd_enc, "tearcheck not supported\n"); + return; + } - if (!phys_enc->hw_pp->ops.setup_tearcheck || - !phys_enc->hw_pp->ops.enable_tearcheck) { - DPU_DEBUG_CMDENC(cmd_enc, "tearcheck not supported\n"); - return; + DPU_DEBUG_CMDENC(cmd_enc, "pp %d\n", phys_enc->hw_pp->idx - PINGPONG_0); } + mode = &phys_enc->cached_mode; + dpu_kms = phys_enc->dpu_kms; /* @@ -349,9 +361,8 @@ static void dpu_encoder_phys_cmd_tearcheck_config( * frequency divided by the no. of rows (lines) in the LCDpanel. */ vsync_hz = dpu_kms_get_clk_rate(dpu_kms, "vsync"); - if (vsync_hz <= 0) { - DPU_DEBUG_CMDENC(cmd_enc, "invalid - vsync_hz %u\n", - vsync_hz); + if (!vsync_hz) { + DPU_DEBUG_CMDENC(cmd_enc, "invalid - no vsync clock\n"); return; } @@ -371,24 +382,24 @@ static void dpu_encoder_phys_cmd_tearcheck_config( tc_cfg.rd_ptr_irq = mode->vdisplay + 1; DPU_DEBUG_CMDENC(cmd_enc, - "tc %d vsync_clk_speed_hz %u vtotal %u vrefresh %u\n", - phys_enc->hw_pp->idx - PINGPONG_0, vsync_hz, - mode->vtotal, drm_mode_vrefresh(mode)); + "tc vsync_clk_speed_hz %lu vtotal %u vrefresh %u\n", + vsync_hz, mode->vtotal, drm_mode_vrefresh(mode)); DPU_DEBUG_CMDENC(cmd_enc, - "tc %d enable %u start_pos %u rd_ptr_irq %u\n", - phys_enc->hw_pp->idx - PINGPONG_0, tc_enable, tc_cfg.start_pos, - tc_cfg.rd_ptr_irq); + "tc enable %u start_pos %u rd_ptr_irq %u\n", + tc_enable, tc_cfg.start_pos, tc_cfg.rd_ptr_irq); DPU_DEBUG_CMDENC(cmd_enc, - "tc %d hw_vsync_mode %u vsync_count %u vsync_init_val %u\n", - phys_enc->hw_pp->idx - PINGPONG_0, tc_cfg.hw_vsync_mode, - tc_cfg.vsync_count, tc_cfg.vsync_init_val); + "tc hw_vsync_mode %u vsync_count %u vsync_init_val %u\n", + tc_cfg.hw_vsync_mode, tc_cfg.vsync_count, + tc_cfg.vsync_init_val); DPU_DEBUG_CMDENC(cmd_enc, - "tc %d cfgheight %u thresh_start %u thresh_cont %u\n", - phys_enc->hw_pp->idx - PINGPONG_0, tc_cfg.sync_cfg_height, - tc_cfg.sync_threshold_start, tc_cfg.sync_threshold_continue); + "tc cfgheight %u thresh_start %u thresh_cont %u\n", + tc_cfg.sync_cfg_height, tc_cfg.sync_threshold_start, + tc_cfg.sync_threshold_continue); - phys_enc->hw_pp->ops.setup_tearcheck(phys_enc->hw_pp, &tc_cfg); - phys_enc->hw_pp->ops.enable_tearcheck(phys_enc->hw_pp, tc_enable); + if (phys_enc->has_intf_te) + phys_enc->hw_intf->ops.enable_tearcheck(phys_enc->hw_intf, &tc_cfg); + else + phys_enc->hw_pp->ops.enable_tearcheck(phys_enc->hw_pp, &tc_cfg); } static void _dpu_encoder_phys_cmd_pingpong_config( @@ -430,7 +441,7 @@ static void dpu_encoder_phys_cmd_enable_helper( return; } - dpu_encoder_helper_split_config(phys_enc, phys_enc->intf_idx); + dpu_encoder_helper_split_config(phys_enc, phys_enc->hw_intf->idx); _dpu_encoder_phys_cmd_pingpong_config(phys_enc); @@ -438,7 +449,7 @@ static void dpu_encoder_phys_cmd_enable_helper( return; ctl = phys_enc->hw_ctl; - ctl->ops.update_pending_flush_intf(ctl, phys_enc->intf_idx); + ctl->ops.update_pending_flush_intf(ctl, phys_enc->hw_intf->idx); } static void dpu_encoder_phys_cmd_enable(struct dpu_encoder_phys *phys_enc) @@ -465,11 +476,19 @@ static void dpu_encoder_phys_cmd_enable(struct dpu_encoder_phys *phys_enc) static void _dpu_encoder_phys_cmd_connect_te( struct dpu_encoder_phys *phys_enc, bool enable) { - if (!phys_enc->hw_pp || !phys_enc->hw_pp->ops.connect_external_te) - return; + if (phys_enc->has_intf_te) { + if (!phys_enc->hw_intf || !phys_enc->hw_intf->ops.connect_external_te) + return; - trace_dpu_enc_phys_cmd_connect_te(DRMID(phys_enc->parent), enable); - phys_enc->hw_pp->ops.connect_external_te(phys_enc->hw_pp, enable); + trace_dpu_enc_phys_cmd_connect_te(DRMID(phys_enc->parent), enable); + phys_enc->hw_intf->ops.connect_external_te(phys_enc->hw_intf, enable); + } else { + if (!phys_enc->hw_pp || !phys_enc->hw_pp->ops.connect_external_te) + return; + + trace_dpu_enc_phys_cmd_connect_te(DRMID(phys_enc->parent), enable); + phys_enc->hw_pp->ops.connect_external_te(phys_enc->hw_pp, enable); + } } static void dpu_encoder_phys_cmd_prepare_idle_pc( @@ -482,17 +501,21 @@ static int dpu_encoder_phys_cmd_get_line_count( struct dpu_encoder_phys *phys_enc) { struct dpu_hw_pingpong *hw_pp; - - if (!phys_enc->hw_pp) - return -EINVAL; + struct dpu_hw_intf *hw_intf; if (!dpu_encoder_phys_cmd_is_master(phys_enc)) return -EINVAL; + if (phys_enc->has_intf_te) { + hw_intf = phys_enc->hw_intf; + if (!hw_intf || !hw_intf->ops.get_line_count) + return -EINVAL; + return hw_intf->ops.get_line_count(hw_intf); + } + hw_pp = phys_enc->hw_pp; - if (!hw_pp->ops.get_line_count) + if (!hw_pp || !hw_pp->ops.get_line_count) return -EINVAL; - return hw_pp->ops.get_line_count(hw_pp); } @@ -502,30 +525,39 @@ static void dpu_encoder_phys_cmd_disable(struct dpu_encoder_phys *phys_enc) to_dpu_encoder_phys_cmd(phys_enc); struct dpu_hw_ctl *ctl; - if (!phys_enc->hw_pp) { - DPU_ERROR("invalid encoder\n"); - return; - } - DRM_DEBUG_KMS("id:%u pp:%d state:%d\n", DRMID(phys_enc->parent), - phys_enc->hw_pp->idx - PINGPONG_0, - phys_enc->enable_state); - if (phys_enc->enable_state == DPU_ENC_DISABLED) { DPU_ERROR_CMDENC(cmd_enc, "already disabled\n"); return; } - if (phys_enc->hw_pp->ops.enable_tearcheck) - phys_enc->hw_pp->ops.enable_tearcheck(phys_enc->hw_pp, false); + if (phys_enc->has_intf_te) { + DRM_DEBUG_KMS("id:%u intf:%d state:%d\n", DRMID(phys_enc->parent), + phys_enc->hw_intf->idx - INTF_0, + phys_enc->enable_state); + + if (phys_enc->hw_intf->ops.disable_tearcheck) + phys_enc->hw_intf->ops.disable_tearcheck(phys_enc->hw_intf); + } else { + if (!phys_enc->hw_pp) { + DPU_ERROR("invalid encoder\n"); + return; + } + + DRM_DEBUG_KMS("id:%u pp:%d state:%d\n", DRMID(phys_enc->parent), + phys_enc->hw_pp->idx - PINGPONG_0, + phys_enc->enable_state); + + if (phys_enc->hw_pp->ops.disable_tearcheck) + phys_enc->hw_pp->ops.disable_tearcheck(phys_enc->hw_pp); + } if (phys_enc->hw_intf->ops.bind_pingpong_blk) { phys_enc->hw_intf->ops.bind_pingpong_blk( phys_enc->hw_intf, - false, - phys_enc->hw_pp->idx); + PINGPONG_NONE); ctl = phys_enc->hw_ctl; - ctl->ops.update_pending_flush_intf(ctl, phys_enc->intf_idx); + ctl->ops.update_pending_flush_intf(ctl, phys_enc->hw_intf->idx); } phys_enc->enable_state = DPU_ENC_DISABLED; @@ -574,66 +606,31 @@ static void dpu_encoder_phys_cmd_prepare_for_kickoff( atomic_read(&phys_enc->pending_kickoff_cnt)); } -static bool dpu_encoder_phys_cmd_is_ongoing_pptx( - struct dpu_encoder_phys *phys_enc) -{ - struct dpu_hw_pp_vsync_info info; - - if (!phys_enc) - return false; - - phys_enc->hw_pp->ops.get_vsync_info(phys_enc->hw_pp, &info); - if (info.wr_ptr_line_count > 0 && - info.wr_ptr_line_count < phys_enc->cached_mode.vdisplay) - return true; - - return false; -} - static void dpu_encoder_phys_cmd_enable_te(struct dpu_encoder_phys *phys_enc) { - struct dpu_encoder_phys_cmd *cmd_enc = - to_dpu_encoder_phys_cmd(phys_enc); - int trial = 0; - if (!phys_enc) return; - if (!phys_enc->hw_pp) - return; if (!dpu_encoder_phys_cmd_is_master(phys_enc)) return; - /* If autorefresh is already disabled, we have nothing to do */ - if (!phys_enc->hw_pp->ops.get_autorefresh(phys_enc->hw_pp, NULL)) - return; - - /* - * If autorefresh is enabled, disable it and make sure it is safe to - * proceed with current frame commit/push. Sequence fallowed is, - * 1. Disable TE - * 2. Disable autorefresh config - * 4. Poll for frame transfer ongoing to be false - * 5. Enable TE back - */ - _dpu_encoder_phys_cmd_connect_te(phys_enc, false); - phys_enc->hw_pp->ops.setup_autorefresh(phys_enc->hw_pp, 0, false); - - do { - udelay(DPU_ENC_MAX_POLL_TIMEOUT_US); - if ((trial * DPU_ENC_MAX_POLL_TIMEOUT_US) - > (KICKOFF_TIMEOUT_MS * USEC_PER_MSEC)) { - DPU_ERROR_CMDENC(cmd_enc, - "disable autorefresh failed\n"); - break; - } - - trial++; - } while (dpu_encoder_phys_cmd_is_ongoing_pptx(phys_enc)); + if (phys_enc->has_intf_te) { + if (!phys_enc->hw_intf->ops.disable_autorefresh) + return; - _dpu_encoder_phys_cmd_connect_te(phys_enc, true); - - DPU_DEBUG_CMDENC(to_dpu_encoder_phys_cmd(phys_enc), - "disabled autorefresh\n"); + phys_enc->hw_intf->ops.disable_autorefresh( + phys_enc->hw_intf, + DRMID(phys_enc->parent), + phys_enc->cached_mode.vdisplay); + } else { + if (!phys_enc->hw_pp || + !phys_enc->hw_pp->ops.disable_autorefresh) + return; + + phys_enc->hw_pp->ops.disable_autorefresh( + phys_enc->hw_pp, + DRMID(phys_enc->parent), + phys_enc->cached_mode.vdisplay); + } } static int _dpu_encoder_phys_cmd_wait_for_ctl_start( @@ -670,7 +667,7 @@ static int dpu_encoder_phys_cmd_wait_for_tx_complete( if (rc) { DRM_ERROR("failed wait_for_idle: id:%u ret:%d intf:%d\n", DRMID(phys_enc->parent), rc, - phys_enc->intf_idx - INTF_0); + phys_enc->hw_intf->idx - INTF_0); } return rc; @@ -710,7 +707,7 @@ static int dpu_encoder_phys_cmd_wait_for_vblank( rc = dpu_encoder_helper_wait_for_irq(phys_enc, phys_enc->irq[INTR_IDX_RDPTR], - dpu_encoder_phys_cmd_pp_rd_ptr_irq, + dpu_encoder_phys_cmd_te_rd_ptr_irq, &wait_info); return rc; @@ -759,36 +756,26 @@ struct dpu_encoder_phys *dpu_encoder_phys_cmd_init( { struct dpu_encoder_phys *phys_enc = NULL; struct dpu_encoder_phys_cmd *cmd_enc = NULL; - int i, ret = 0; - DPU_DEBUG("intf %d\n", p->intf_idx - INTF_0); + DPU_DEBUG("intf\n"); cmd_enc = kzalloc(sizeof(*cmd_enc), GFP_KERNEL); if (!cmd_enc) { - ret = -ENOMEM; DPU_ERROR("failed to allocate\n"); - return ERR_PTR(ret); + return ERR_PTR(-ENOMEM); } phys_enc = &cmd_enc->base; - phys_enc->hw_mdptop = p->dpu_kms->hw_mdp; - phys_enc->intf_idx = p->intf_idx; + + dpu_encoder_phys_init(phys_enc, p); dpu_encoder_phys_cmd_init_ops(&phys_enc->ops); - phys_enc->parent = p->parent; - phys_enc->dpu_kms = p->dpu_kms; - phys_enc->split_role = p->split_role; phys_enc->intf_mode = INTF_MODE_CMD; - phys_enc->enc_spinlock = p->enc_spinlock; cmd_enc->stream_sel = 0; - phys_enc->enable_state = DPU_ENC_DISABLED; - for (i = 0; i < ARRAY_SIZE(phys_enc->irq); i++) - phys_enc->irq[i] = -EINVAL; - atomic_set(&phys_enc->vblank_refcount, 0); - atomic_set(&phys_enc->pending_kickoff_cnt, 0); - atomic_set(&phys_enc->pending_ctlstart_cnt, 0); + phys_enc->has_intf_te = test_bit(DPU_INTF_TE, + &phys_enc->hw_intf->cap->features); + atomic_set(&cmd_enc->pending_vblank_cnt, 0); - init_waitqueue_head(&phys_enc->pending_kickoff_wq); init_waitqueue_head(&cmd_enc->pending_vblank_wq); DPU_DEBUG_CMDENC(cmd_enc, "created\n"); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c index 3a374292f31167e3268657b50d7fa7962b96e0f3..662d74ded1b94207b22446ff565535448f04338b 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c @@ -287,7 +287,6 @@ static void dpu_encoder_phys_vid_setup_timing_engine( if (phys_enc->hw_intf->ops.bind_pingpong_blk) phys_enc->hw_intf->ops.bind_pingpong_blk( phys_enc->hw_intf, - true, phys_enc->hw_pp->idx); if (phys_enc->hw_pp->merge_3d) @@ -699,7 +698,6 @@ struct dpu_encoder_phys *dpu_encoder_phys_vid_init( struct dpu_enc_phys_init_params *p) { struct dpu_encoder_phys *phys_enc = NULL; - int i; if (!p) { DPU_ERROR("failed to create encoder due to invalid parameter\n"); @@ -712,26 +710,14 @@ struct dpu_encoder_phys *dpu_encoder_phys_vid_init( return ERR_PTR(-ENOMEM); } - phys_enc->hw_mdptop = p->dpu_kms->hw_mdp; - phys_enc->intf_idx = p->intf_idx; - DPU_DEBUG_VIDENC(phys_enc, "\n"); + dpu_encoder_phys_init(phys_enc, p); + dpu_encoder_phys_vid_init_ops(&phys_enc->ops); - phys_enc->parent = p->parent; - phys_enc->dpu_kms = p->dpu_kms; - phys_enc->split_role = p->split_role; phys_enc->intf_mode = INTF_MODE_VIDEO; - phys_enc->enc_spinlock = p->enc_spinlock; - for (i = 0; i < ARRAY_SIZE(phys_enc->irq); i++) - phys_enc->irq[i] = -EINVAL; - - atomic_set(&phys_enc->vblank_refcount, 0); - atomic_set(&phys_enc->pending_kickoff_cnt, 0); - init_waitqueue_head(&phys_enc->pending_kickoff_wq); - phys_enc->enable_state = DPU_ENC_DISABLED; - DPU_DEBUG_VIDENC(phys_enc, "created intf idx:%d\n", p->intf_idx); + DPU_DEBUG_VIDENC(phys_enc, "created intf idx:%d\n", p->hw_intf->idx); return phys_enc; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c index bac4aa807b4bc00991a104f55b4dcfd7b0145334..a466ff70a4d62beabee0117ea902b97f7673221e 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c @@ -102,7 +102,7 @@ static void dpu_encoder_phys_wb_set_qos_remap( static void dpu_encoder_phys_wb_set_qos(struct dpu_encoder_phys *phys_enc) { struct dpu_hw_wb *hw_wb; - struct dpu_hw_wb_qos_cfg qos_cfg; + struct dpu_hw_qos_cfg qos_cfg; const struct dpu_mdss_cfg *catalog; const struct dpu_qos_lut_tbl *qos_lut_tb; @@ -115,7 +115,7 @@ static void dpu_encoder_phys_wb_set_qos(struct dpu_encoder_phys *phys_enc) hw_wb = phys_enc->hw_wb; - memset(&qos_cfg, 0, sizeof(struct dpu_hw_wb_qos_cfg)); + memset(&qos_cfg, 0, sizeof(struct dpu_hw_qos_cfg)); qos_cfg.danger_safe_en = true; qos_cfg.danger_lut = catalog->perf->danger_lut_tbl[DPU_QOS_LUT_USAGE_NRT]; @@ -140,7 +140,6 @@ static void dpu_encoder_phys_wb_setup_fb(struct dpu_encoder_phys *phys_enc, struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys_enc); struct dpu_hw_wb *hw_wb; struct dpu_hw_wb_cfg *wb_cfg; - struct dpu_hw_cdp_cfg cdp_cfg; if (!phys_enc || !phys_enc->dpu_kms || !phys_enc->dpu_kms->catalog) { DPU_ERROR("invalid encoder\n"); @@ -163,18 +162,10 @@ static void dpu_encoder_phys_wb_setup_fb(struct dpu_encoder_phys *phys_enc, hw_wb->ops.setup_outformat(hw_wb, wb_cfg); if (hw_wb->ops.setup_cdp) { - memset(&cdp_cfg, 0, sizeof(struct dpu_hw_cdp_cfg)); - - cdp_cfg.enable = phys_enc->dpu_kms->catalog->perf->cdp_cfg - [DPU_PERF_CDP_USAGE_NRT].wr_enable; - cdp_cfg.ubwc_meta_enable = - DPU_FORMAT_IS_UBWC(wb_cfg->dest.format); - cdp_cfg.tile_amortize_enable = - DPU_FORMAT_IS_UBWC(wb_cfg->dest.format) || - DPU_FORMAT_IS_TILE(wb_cfg->dest.format); - cdp_cfg.preload_ahead = DPU_WB_CDP_PRELOAD_AHEAD_64; - - hw_wb->ops.setup_cdp(hw_wb, &cdp_cfg); + const struct dpu_perf_cfg *perf = phys_enc->dpu_kms->catalog->perf; + + hw_wb->ops.setup_cdp(hw_wb, wb_cfg->dest.format, + perf->cdp_cfg[DPU_PERF_CDP_USAGE_NRT].wr_enable); } if (hw_wb->ops.setup_outaddress) @@ -219,7 +210,7 @@ static void dpu_encoder_phys_wb_setup_cdp(struct dpu_encoder_phys *phys_enc) /* setup which pp blk will connect to this wb */ if (hw_pp && phys_enc->hw_wb->ops.bind_pingpong_blk) - phys_enc->hw_wb->ops.bind_pingpong_blk(phys_enc->hw_wb, true, + phys_enc->hw_wb->ops.bind_pingpong_blk(phys_enc->hw_wb, phys_enc->hw_pp->idx); phys_enc->hw_ctl->ops.setup_intf_cfg(phys_enc->hw_ctl, &intf_cfg); @@ -249,7 +240,7 @@ static int dpu_encoder_phys_wb_atomic_check( const struct drm_display_mode *mode = &crtc_state->mode; DPU_DEBUG("[atomic_check:%d, \"%s\",%d,%d]\n", - phys_enc->wb_idx, mode->name, mode->hdisplay, mode->vdisplay); + phys_enc->hw_wb->idx, mode->name, mode->hdisplay, mode->vdisplay); if (!conn_state || !conn_state->connector) { DPU_ERROR("invalid connector state\n"); @@ -570,7 +561,7 @@ static void dpu_encoder_phys_wb_destroy(struct dpu_encoder_phys *phys_enc) if (!phys_enc) return; - DPU_DEBUG("[wb:%d]\n", phys_enc->wb_idx - WB_0); + DPU_DEBUG("[wb:%d]\n", phys_enc->hw_wb->idx - WB_0); kfree(phys_enc); } @@ -693,53 +684,32 @@ struct dpu_encoder_phys *dpu_encoder_phys_wb_init( { struct dpu_encoder_phys *phys_enc = NULL; struct dpu_encoder_phys_wb *wb_enc = NULL; - int ret = 0; - int i; DPU_DEBUG("\n"); if (!p || !p->parent) { DPU_ERROR("invalid params\n"); - ret = -EINVAL; - goto fail_alloc; + return ERR_PTR(-EINVAL); } wb_enc = kzalloc(sizeof(*wb_enc), GFP_KERNEL); if (!wb_enc) { DPU_ERROR("failed to allocate wb phys_enc enc\n"); - ret = -ENOMEM; - goto fail_alloc; + return ERR_PTR(-ENOMEM); } phys_enc = &wb_enc->base; - phys_enc->hw_mdptop = p->dpu_kms->hw_mdp; - phys_enc->wb_idx = p->wb_idx; + + dpu_encoder_phys_init(phys_enc, p); dpu_encoder_phys_wb_init_ops(&phys_enc->ops); - phys_enc->parent = p->parent; - phys_enc->dpu_kms = p->dpu_kms; - phys_enc->split_role = p->split_role; phys_enc->intf_mode = INTF_MODE_WB_LINE; - phys_enc->wb_idx = p->wb_idx; - phys_enc->enc_spinlock = p->enc_spinlock; atomic_set(&wb_enc->wbirq_refcount, 0); - for (i = 0; i < ARRAY_SIZE(phys_enc->irq); i++) - phys_enc->irq[i] = -EINVAL; - - atomic_set(&phys_enc->pending_kickoff_cnt, 0); - atomic_set(&phys_enc->vblank_refcount, 0); wb_enc->wb_done_timeout_cnt = 0; - init_waitqueue_head(&phys_enc->pending_kickoff_wq); - phys_enc->enable_state = DPU_ENC_DISABLED; - - DPU_DEBUG("Created dpu_encoder_phys for wb %d\n", - phys_enc->wb_idx); + DPU_DEBUG("Created dpu_encoder_phys for wb %d\n", phys_enc->hw_wb->idx); return phys_enc; - -fail_alloc: - return ERR_PTR(ret); } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c index 03f162af1a50bcaa996576f99dc6c3ac23bbc6d5..0de507d4d7b7a26870044501b2af0efce630bdfb 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0-only /* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. - * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2022-2023, Qualcomm Innovation Center, Inc. All rights reserved. */ #define pr_fmt(fmt) "[drm:%s:%d] " fmt, __func__, __LINE__ @@ -13,7 +13,7 @@ #include "dpu_kms.h" #define VIG_BASE_MASK \ - (BIT(DPU_SSPP_SRC) | BIT(DPU_SSPP_QOS) |\ + (BIT(DPU_SSPP_QOS) |\ BIT(DPU_SSPP_CDP) |\ BIT(DPU_SSPP_TS_PREFILL) | BIT(DPU_SSPP_EXCL_RECT)) @@ -39,7 +39,7 @@ #define VIG_QCM2290_MASK (VIG_BASE_MASK | BIT(DPU_SSPP_QOS_8LVL)) #define DMA_MSM8998_MASK \ - (BIT(DPU_SSPP_SRC) | BIT(DPU_SSPP_QOS) |\ + (BIT(DPU_SSPP_QOS) |\ BIT(DPU_SSPP_TS_PREFILL) | BIT(DPU_SSPP_TS_PREFILL_REC1) |\ BIT(DPU_SSPP_CDP) | BIT(DPU_SSPP_EXCL_RECT)) @@ -50,7 +50,7 @@ (VIG_SC7280_MASK | BIT(DPU_SSPP_SMART_DMA_V2)) #define DMA_SDM845_MASK \ - (BIT(DPU_SSPP_SRC) | BIT(DPU_SSPP_QOS) | BIT(DPU_SSPP_QOS_8LVL) |\ + (BIT(DPU_SSPP_QOS) | BIT(DPU_SSPP_QOS_8LVL) |\ BIT(DPU_SSPP_TS_PREFILL) | BIT(DPU_SSPP_TS_PREFILL_REC1) |\ BIT(DPU_SSPP_CDP) | BIT(DPU_SSPP_EXCL_RECT)) @@ -75,11 +75,15 @@ #define MIXER_QCM2290_MASK \ (BIT(DPU_DIM_LAYER) | BIT(DPU_MIXER_COMBINED_ALPHA)) -#define PINGPONG_SDM845_MASK BIT(DPU_PINGPONG_DITHER) +#define PINGPONG_SDM845_MASK \ + (BIT(DPU_PINGPONG_DITHER) | BIT(DPU_PINGPONG_TE) | BIT(DPU_PINGPONG_DSC)) -#define PINGPONG_SDM845_SPLIT_MASK \ +#define PINGPONG_SDM845_TE2_MASK \ (PINGPONG_SDM845_MASK | BIT(DPU_PINGPONG_TE2)) +#define PINGPONG_SM8150_MASK \ + (BIT(DPU_PINGPONG_DITHER) | BIT(DPU_PINGPONG_DSC)) + #define CTL_SC7280_MASK \ (BIT(DPU_CTL_ACTIVE_CFG) | \ BIT(DPU_CTL_FETCH_ACTIVE) | \ @@ -91,16 +95,17 @@ #define MERGE_3D_SM8150_MASK (0) -#define DSPP_MSM8998_MASK BIT(DPU_DSPP_PCC) | BIT(DPU_DSPP_GC) - #define DSPP_SC7180_MASK BIT(DPU_DSPP_PCC) #define INTF_SDM845_MASK (0) #define INTF_SC7180_MASK \ - (BIT(DPU_INTF_INPUT_CTRL) | BIT(DPU_INTF_TE) | BIT(DPU_INTF_STATUS_SUPPORTED)) + (BIT(DPU_INTF_INPUT_CTRL) | \ + BIT(DPU_INTF_TE) | \ + BIT(DPU_INTF_STATUS_SUPPORTED) | \ + BIT(DPU_DATA_HCTL_EN)) -#define INTF_SC7280_MASK INTF_SC7180_MASK | BIT(DPU_DATA_HCTL_EN) +#define INTF_SC7280_MASK (INTF_SC7180_MASK | BIT(DPU_INTF_DATA_COMPRESS)) #define WB_SM8250_MASK (BIT(DPU_WB_LINE_MODE) | \ BIT(DPU_WB_UBWC) | \ @@ -252,8 +257,6 @@ static const uint32_t wb2_formats[] = { .maxdwnscale = MAX_DOWNSCALE_RATIO, \ .maxupscale = MAX_UPSCALE_RATIO, \ .smart_dma_priority = sdma_pri, \ - .src_blk = {.name = STRCAT("sspp_src_", num), \ - .id = DPU_SSPP_SRC, .base = 0x00, .len = 0x150,}, \ .scaler_blk = {.name = STRCAT("sspp_scaler", num), \ .id = qseed_ver, \ .base = 0xa00, .len = 0xa0,}, \ @@ -272,8 +275,6 @@ static const uint32_t wb2_formats[] = { .maxdwnscale = MAX_DOWNSCALE_RATIO, \ .maxupscale = MAX_UPSCALE_RATIO, \ .smart_dma_priority = sdma_pri, \ - .src_blk = {.name = STRCAT("sspp_src_", num), \ - .id = DPU_SSPP_SRC, .base = 0x00, .len = 0x150,}, \ .scaler_blk = {.name = STRCAT("sspp_scaler", num), \ .id = qseed_ver, \ .base = 0xa00, .len = 0xa0,}, \ @@ -292,8 +293,6 @@ static const uint32_t wb2_formats[] = { .maxdwnscale = SSPP_UNITY_SCALE, \ .maxupscale = SSPP_UNITY_SCALE, \ .smart_dma_priority = sdma_pri, \ - .src_blk = {.name = STRCAT("sspp_src_", num), \ - .id = DPU_SSPP_SRC, .base = 0x00, .len = 0x150,}, \ .format_list = plane_formats, \ .num_formats = ARRAY_SIZE(plane_formats), \ .virt_format_list = plane_formats, \ @@ -375,8 +374,6 @@ static const struct dpu_sspp_sub_blks sm8550_dma_sblk_5 = _DMA_SBLK("13", 6); .maxdwnscale = SSPP_UNITY_SCALE, \ .maxupscale = SSPP_UNITY_SCALE, \ .smart_dma_priority = sdma_pri, \ - .src_blk = {.name = STRCAT("sspp_src_", num), \ - .id = DPU_SSPP_SRC, .base = 0x00, .len = 0x150,}, \ .format_list = plane_formats_yuv, \ .num_formats = ARRAY_SIZE(plane_formats_yuv), \ .virt_format_list = plane_formats, \ @@ -449,13 +446,6 @@ static const struct dpu_lm_sub_blks qcm2290_lm_sblk = { static const struct dpu_dspp_sub_blks msm8998_dspp_sblk = { .pcc = {.id = DPU_DSPP_PCC, .base = 0x1700, .len = 0x90, .version = 0x10007}, - .gc = { .id = DPU_DSPP_GC, .base = 0x17c0, - .len = 0x90, .version = 0x10007}, -}; - -static const struct dpu_dspp_sub_blks sc7180_dspp_sblk = { - .pcc = {.id = DPU_DSPP_PCC, .base = 0x1700, - .len = 0x90, .version = 0x10000}, }; static const struct dpu_dspp_sub_blks sm8150_dspp_sblk = { @@ -491,7 +481,7 @@ static const struct dpu_pingpong_sub_blks sc7280_pp_sblk = { .len = 0x20, .version = 0x20000}, }; -#define PP_BLK_DIPHER(_name, _id, _base, _merge_3d, _sblk, _done, _rdptr) \ +#define PP_BLK_DITHER(_name, _id, _base, _merge_3d, _sblk, _done, _rdptr) \ {\ .name = _name, .id = _id, \ .base = _base, .len = 0, \ @@ -501,21 +491,11 @@ static const struct dpu_pingpong_sub_blks sc7280_pp_sblk = { .intr_done = _done, \ .intr_rdptr = _rdptr, \ } -#define PP_BLK_TE(_name, _id, _base, _merge_3d, _sblk, _done, _rdptr) \ - {\ - .name = _name, .id = _id, \ - .base = _base, .len = 0xd4, \ - .features = PINGPONG_SDM845_SPLIT_MASK, \ - .merge_3d = _merge_3d, \ - .sblk = &_sblk, \ - .intr_done = _done, \ - .intr_rdptr = _rdptr, \ - } -#define PP_BLK(_name, _id, _base, _merge_3d, _sblk, _done, _rdptr) \ +#define PP_BLK(_name, _id, _base, _features, _merge_3d, _sblk, _done, _rdptr) \ {\ .name = _name, .id = _id, \ .base = _base, .len = 0xd4, \ - .features = PINGPONG_SDM845_MASK, \ + .features = _features, \ .merge_3d = _merge_3d, \ .sblk = &_sblk, \ .intr_done = _done, \ @@ -528,7 +508,7 @@ static const struct dpu_pingpong_sub_blks sc7280_pp_sblk = { #define MERGE_3D_BLK(_name, _id, _base) \ {\ .name = _name, .id = _id, \ - .base = _base, .len = 0x100, \ + .base = _base, .len = 0x8, \ .features = MERGE_3D_SM8150_MASK, \ .sblk = NULL \ } @@ -536,6 +516,16 @@ static const struct dpu_pingpong_sub_blks sc7280_pp_sblk = { /************************************************************* * DSC sub blocks config *************************************************************/ +static const struct dpu_dsc_sub_blks dsc_sblk_0 = { + .enc = {.base = 0x100, .len = 0x100}, + .ctl = {.base = 0xF00, .len = 0x10}, +}; + +static const struct dpu_dsc_sub_blks dsc_sblk_1 = { + .enc = {.base = 0x200, .len = 0x100}, + .ctl = {.base = 0xF80, .len = 0x10}, +}; + #define DSC_BLK(_name, _id, _base, _features) \ {\ .name = _name, .id = _id, \ @@ -543,10 +533,32 @@ static const struct dpu_pingpong_sub_blks sc7280_pp_sblk = { .features = _features, \ } +#define DSC_BLK_1_2(_name, _id, _base, _len, _features, _sblk) \ + {\ + .name = _name, .id = _id, \ + .base = _base, .len = _len, \ + .features = BIT(DPU_DSC_HW_REV_1_2) | _features, \ + .sblk = &_sblk, \ + } + /************************************************************* * INTF sub blocks config *************************************************************/ -#define INTF_BLK(_name, _id, _base, _len, _type, _ctrl_id, _progfetch, _features, _reg, _underrun_bit, _vsync_bit) \ +#define INTF_BLK(_name, _id, _base, _len, _type, _ctrl_id, _progfetch, _features, _underrun, _vsync) \ + {\ + .name = _name, .id = _id, \ + .base = _base, .len = _len, \ + .features = _features, \ + .type = _type, \ + .controller_id = _ctrl_id, \ + .prog_fetch_lines_worst_case = _progfetch, \ + .intr_underrun = _underrun, \ + .intr_vsync = _vsync, \ + .intr_tear_rd_ptr = -1, \ + } + +/* DSI Interface sub-block with TEAR registers (since DPU 5.0.0) */ +#define INTF_BLK_DSI_TE(_name, _id, _base, _len, _type, _ctrl_id, _progfetch, _features, _underrun, _vsync, _tear_rd_ptr) \ {\ .name = _name, .id = _id, \ .base = _base, .len = _len, \ @@ -554,8 +566,9 @@ static const struct dpu_pingpong_sub_blks sc7280_pp_sblk = { .type = _type, \ .controller_id = _ctrl_id, \ .prog_fetch_lines_worst_case = _progfetch, \ - .intr_underrun = DPU_IRQ_IDX(_reg, _underrun_bit), \ - .intr_vsync = DPU_IRQ_IDX(_reg, _vsync_bit), \ + .intr_underrun = _underrun, \ + .intr_vsync = _vsync, \ + .intr_tear_rd_ptr = _tear_rd_ptr, \ } /************************************************************* @@ -587,12 +600,12 @@ static const u32 sdm845_nrt_pri_lvl[] = {3, 3, 3, 3, 3, 3, 3, 3}; static const struct dpu_vbif_dynamic_ot_cfg msm8998_ot_rdwr_cfg[] = { { - .pps = 1088 * 1920 * 30, + .pps = 1920 * 1080 * 30, .ot_limit = 2, }, { - .pps = 1088 * 1920 * 60, - .ot_limit = 6, + .pps = 1920 * 1080 * 60, + .ot_limit = 4, }, { .pps = 3840 * 2160 * 30, @@ -650,46 +663,6 @@ static const struct dpu_vbif_cfg sdm845_vbif[] = { }, }; -static const struct dpu_reg_dma_cfg sc8280xp_regdma = { - .base = 0x0, - .version = 0x00020000, - .trigger_sel_off = 0x119c, - .xin_id = 7, - .clk_ctrl = DPU_CLK_CTRL_REG_DMA, -}; - -static const struct dpu_reg_dma_cfg sdm845_regdma = { - .base = 0x0, .version = 0x1, .trigger_sel_off = 0x119c -}; - -static const struct dpu_reg_dma_cfg sm8150_regdma = { - .base = 0x0, .version = 0x00010001, .trigger_sel_off = 0x119c -}; - -static const struct dpu_reg_dma_cfg sm8250_regdma = { - .base = 0x0, - .version = 0x00010002, - .trigger_sel_off = 0x119c, - .xin_id = 7, - .clk_ctrl = DPU_CLK_CTRL_REG_DMA, -}; - -static const struct dpu_reg_dma_cfg sm8350_regdma = { - .base = 0x400, - .version = 0x00020000, - .trigger_sel_off = 0x119c, - .xin_id = 7, - .clk_ctrl = DPU_CLK_CTRL_REG_DMA, -}; - -static const struct dpu_reg_dma_cfg sm8450_regdma = { - .base = 0x0, - .version = 0x00020000, - .trigger_sel_off = 0x119c, - .xin_id = 7, - .clk_ctrl = DPU_CLK_CTRL_REG_DMA, -}; - /************************************************************* * PERF data config *************************************************************/ @@ -705,10 +678,7 @@ static const struct dpu_qos_lut_entry msm8998_qos_linear[] = { {.fl = 10, .lut = 0x1555b}, {.fl = 11, .lut = 0x5555b}, {.fl = 12, .lut = 0x15555b}, - {.fl = 13, .lut = 0x55555b}, - {.fl = 14, .lut = 0}, - {.fl = 1, .lut = 0x1b}, - {.fl = 0, .lut = 0} + {.fl = 0, .lut = 0x55555b} }; static const struct dpu_qos_lut_entry sdm845_qos_linear[] = { @@ -730,15 +700,17 @@ static const struct dpu_qos_lut_entry msm8998_qos_macrotile[] = { {.fl = 10, .lut = 0x1aaff}, {.fl = 11, .lut = 0x5aaff}, {.fl = 12, .lut = 0x15aaff}, - {.fl = 13, .lut = 0x55aaff}, - {.fl = 1, .lut = 0x1aaff}, - {.fl = 0, .lut = 0}, + {.fl = 0, .lut = 0x55aaff}, }; static const struct dpu_qos_lut_entry sc7180_qos_linear[] = { {.fl = 0, .lut = 0x0011222222335777}, }; +static const struct dpu_qos_lut_entry sm6350_qos_linear_macrotile[] = { + {.fl = 0, .lut = 0x0011223445566777 }, +}; + static const struct dpu_qos_lut_entry sm8150_qos_linear[] = { {.fl = 0, .lut = 0x0011222222223357 }, }; @@ -794,7 +766,9 @@ static const struct dpu_qos_lut_entry sc7180_qos_nrt[] = { #include "catalog/dpu_6_0_sm8250.h" #include "catalog/dpu_6_2_sc7180.h" #include "catalog/dpu_6_3_sm6115.h" +#include "catalog/dpu_6_4_sm6350.h" #include "catalog/dpu_6_5_qcm2290.h" +#include "catalog/dpu_6_9_sm6375.h" #include "catalog/dpu_7_0_sm8350.h" #include "catalog/dpu_7_2_sc7280.h" diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h index 71584cd56fd759c2a2a6a792f7d76c42f2babb2e..b860784ade726aca28663aad55a3ac24360a1369 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0-only */ /* - * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2022-2023, Qualcomm Innovation Center, Inc. All rights reserved. * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. */ @@ -48,6 +48,8 @@ enum { * @DPU_MDP_UBWC_1_5, Universal Bandwidth compression version 1.5 * @DPU_MDP_PERIPH_0_REMOVED Indicates that access to periph top0 block results * in a failure + * @DPU_MDP_VSYNC_SEL Enables vsync source selection via MDP_VSYNC_SEL register + * (moved into INTF block since DPU 5.0.0) * @DPU_MDP_MAX Maximum value */ @@ -59,12 +61,12 @@ enum { DPU_MDP_UBWC_1_5, DPU_MDP_AUDIO_SELECT, DPU_MDP_PERIPH_0_REMOVED, + DPU_MDP_VSYNC_SEL, DPU_MDP_MAX }; /** * SSPP sub-blocks/features - * @DPU_SSPP_SRC Src and fetch part of the pipes, * @DPU_SSPP_SCALER_QSEED2, QSEED2 algorithm support * @DPU_SSPP_SCALER_QSEED3, QSEED3 alogorithm support * @DPU_SSPP_SCALER_QSEED3LITE, QSEED3 Lite alogorithm support @@ -85,8 +87,7 @@ enum { * @DPU_SSPP_MAX maximum value */ enum { - DPU_SSPP_SRC = 0x1, - DPU_SSPP_SCALER_QSEED2, + DPU_SSPP_SCALER_QSEED2 = 0x1, DPU_SSPP_SCALER_QSEED3, DPU_SSPP_SCALER_QSEED3LITE, DPU_SSPP_SCALER_QSEED4, @@ -127,13 +128,9 @@ enum { /** * DSPP sub-blocks * @DPU_DSPP_PCC Panel color correction block - * @DPU_DSPP_GC Gamma correction block - * @DPU_DSPP_IGC Inverse gamma correction block */ enum { DPU_DSPP_PCC = 0x1, - DPU_DSPP_GC, - DPU_DSPP_IGC, DPU_DSPP_MAX }; @@ -143,7 +140,8 @@ enum { * @DPU_PINGPONG_TE2 Additional tear check block for split pipes * @DPU_PINGPONG_SPLIT PP block supports split fifo * @DPU_PINGPONG_SLAVE PP block is a suitable slave for split fifo - * @DPU_PINGPONG_DITHER, Dither blocks + * @DPU_PINGPONG_DITHER Dither blocks + * @DPU_PINGPONG_DSC PP block supports DSC * @DPU_PINGPONG_MAX */ enum { @@ -152,6 +150,7 @@ enum { DPU_PINGPONG_SPLIT, DPU_PINGPONG_SLAVE, DPU_PINGPONG_DITHER, + DPU_PINGPONG_DSC, DPU_PINGPONG_MAX }; @@ -182,6 +181,7 @@ enum { * @DPU_DATA_HCTL_EN Allows data to be transferred at different rate * than video timing * @DPU_INTF_STATUS_SUPPORTED INTF block has INTF_STATUS register + * @DPU_INTF_DATA_COMPRESS INTF block has DATA_COMPRESS register * @DPU_INTF_MAX */ enum { @@ -189,6 +189,7 @@ enum { DPU_INTF_TE, DPU_DATA_HCTL_EN, DPU_INTF_STATUS_SUPPORTED, + DPU_INTF_DATA_COMPRESS, DPU_INTF_MAX }; @@ -241,12 +242,18 @@ enum { }; /** - * DSC features + * DSC sub-blocks/features * @DPU_DSC_OUTPUT_CTRL Configure which PINGPONG block gets * the pixel output from this DSC. + * @DPU_DSC_HW_REV_1_2 DSC block supports DSC 1.1 and 1.2 + * @DPU_DSC_NATIVE_42x_EN Supports NATIVE_422_EN and NATIVE_420_EN encoding + * @DPU_DSC_MAX */ enum { DPU_DSC_OUTPUT_CTRL = 0x1, + DPU_DSC_HW_REV_1_2, + DPU_DSC_NATIVE_42x_EN, + DPU_DSC_MAX }; /** @@ -278,14 +285,6 @@ enum { u32 base; \ u32 len -/** - * struct dpu_src_blk: SSPP part of the source pipes - * @info: HW register and features supported by this sub-blk - */ -struct dpu_src_blk { - DPU_HW_SUBBLK_INFO; -}; - /** * struct dpu_scaler_blk: Scaler information * @info: HW register and features supported by this sub-blk @@ -310,6 +309,14 @@ struct dpu_pp_blk { u32 version; }; +/** + * struct dpu_dsc_blk - DSC Encoder sub-blk information + * @info: HW register and features supported by this sub-blk + */ +struct dpu_dsc_blk { + DPU_HW_SUBBLK_INFO; +}; + /** * enum dpu_qos_lut_usage - define QoS LUT use cases */ @@ -385,20 +392,13 @@ struct dpu_caps { /** * struct dpu_sspp_sub_blks : SSPP sub-blocks * common: Pointer to common configurations shared by sub blocks - * @creq_vblank: creq priority during vertical blanking - * @danger_vblank: danger priority during vertical blanking * @maxdwnscale: max downscale ratio supported(without DECIMATION) * @maxupscale: maxupscale ratio supported * @smart_dma_priority: hw priority of rect1 of multirect pipe * @max_per_pipe_bw: maximum allowable bandwidth of this pipe in kBps * @qseed_ver: qseed version - * @src_blk: * @scaler_blk: * @csc_blk: - * @hsic: - * @memcolor: - * @pcc_blk: - * @igc_blk: * @format_list: Pointer to list of supported formats * @num_formats: Number of supported formats * @virt_format_list: Pointer to list of supported formats for virtual planes @@ -406,20 +406,13 @@ struct dpu_caps { * @dpu_rotation_cfg: inline rotation configuration */ struct dpu_sspp_sub_blks { - u32 creq_vblank; - u32 danger_vblank; u32 maxdwnscale; u32 maxupscale; u32 smart_dma_priority; u32 max_per_pipe_bw; u32 qseed_ver; - struct dpu_src_blk src_blk; struct dpu_scaler_blk scaler_blk; struct dpu_pp_blk csc_blk; - struct dpu_pp_blk hsic_blk; - struct dpu_pp_blk memcolor_blk; - struct dpu_pp_blk pcc_blk; - struct dpu_pp_blk igc_blk; const u32 *format_list; u32 num_formats; @@ -433,22 +426,18 @@ struct dpu_sspp_sub_blks { * @maxwidth: Max pixel width supported by this mixer * @maxblendstages: Max number of blend-stages supported * @blendstage_base: Blend-stage register base offset - * @gc: gamma correction block */ struct dpu_lm_sub_blks { u32 maxwidth; u32 maxblendstages; u32 blendstage_base[MAX_BLOCKS]; - struct dpu_pp_blk gc; }; /** * struct dpu_dspp_sub_blks: Information of DSPP block - * @gc : gamma correction block * @pcc: pixel color correction block */ struct dpu_dspp_sub_blks { - struct dpu_pp_blk gc; struct dpu_pp_blk pcc; }; @@ -458,6 +447,16 @@ struct dpu_pingpong_sub_blks { struct dpu_pp_blk dither; }; +/** + * struct dpu_dsc_sub_blks - DSC sub-blks + * @enc: DSC encoder sub-block + * @ctl: DSC controller sub-block + */ +struct dpu_dsc_sub_blks { + struct dpu_dsc_blk enc; + struct dpu_dsc_blk ctl; +}; + /** * dpu_clk_ctrl_type - Defines top level clock control signals */ @@ -554,7 +553,7 @@ struct dpu_sspp_cfg { * @base register offset of this block * @features bit mask identifying sub-blocks/features * @sblk: LM Sub-blocks information - * @pingpong: ID of connected PingPong, PINGPONG_MAX if unsupported + * @pingpong: ID of connected PingPong, PINGPONG_NONE if unsupported * @lm_pair_mask: Bitmask of LMs that can be controlled by same CTL */ struct dpu_lm_cfg { @@ -612,10 +611,13 @@ struct dpu_merge_3d_cfg { * struct dpu_dsc_cfg - information of DSC blocks * @id enum identifying this block * @base register offset of this block + * @len: length of hardware block * @features bit mask identifying sub-blocks/features + * @sblk: sub-blocks information */ struct dpu_dsc_cfg { DPU_HW_BLK_INFO; + const struct dpu_dsc_sub_blks *sblk; }; /** @@ -628,6 +630,7 @@ struct dpu_dsc_cfg { * @prog_fetch_lines_worst_case Worst case latency num lines needed to prefetch * @intr_underrun: index for INTF underrun interrupt * @intr_vsync: index for INTF VSYNC interrupt + * @intr_tear_rd_ptr: Index for INTF TEAR_RD_PTR interrupt */ struct dpu_intf_cfg { DPU_HW_BLK_INFO; @@ -636,6 +639,7 @@ struct dpu_intf_cfg { u32 prog_fetch_lines_worst_case; s32 intr_underrun; s32 intr_vsync; + s32 intr_tear_rd_ptr; }; /** @@ -720,21 +724,6 @@ struct dpu_vbif_cfg { u32 memtype_count; u32 memtype[MAX_XIN_COUNT]; }; -/** - * struct dpu_reg_dma_cfg - information of lut dma blocks - * @id enum identifying this block - * @base register offset of this block - * @features bit mask identifying sub-blocks/features - * @version version of lutdma hw block - * @trigger_sel_off offset to trigger select registers of lutdma - */ -struct dpu_reg_dma_cfg { - DPU_HW_BLK_INFO; - u32 version; - u32 trigger_sel_off; - u32 xin_id; - enum dpu_clk_ctrl_type clk_ctrl; -}; /** * Define CDP use cases @@ -850,9 +839,6 @@ struct dpu_mdss_cfg { u32 wb_count; const struct dpu_wb_cfg *wb; - u32 reg_dma_count; - const struct dpu_reg_dma_cfg *dma_cfg; - u32 ad_count; u32 dspp_count; @@ -875,7 +861,9 @@ extern const struct dpu_mdss_cfg dpu_sc8180x_cfg; extern const struct dpu_mdss_cfg dpu_sm8250_cfg; extern const struct dpu_mdss_cfg dpu_sc7180_cfg; extern const struct dpu_mdss_cfg dpu_sm6115_cfg; +extern const struct dpu_mdss_cfg dpu_sm6350_cfg; extern const struct dpu_mdss_cfg dpu_qcm2290_cfg; +extern const struct dpu_mdss_cfg dpu_sm6375_cfg; extern const struct dpu_mdss_cfg dpu_sm8350_cfg; extern const struct dpu_mdss_cfg dpu_sc7280_cfg; extern const struct dpu_mdss_cfg dpu_sc8280xp_cfg; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c index bbdc95ce374a754cfa64c4d8603bb2ddd3bde0c1..c278fb9d2b5b5bc1b69f13fe9a03c74119ef4128 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c @@ -53,23 +53,6 @@ static const u32 fetch_tbl[SSPP_MAX] = {CTL_INVALID_BIT, 16, 17, 18, 19, CTL_INVALID_BIT, CTL_INVALID_BIT, CTL_INVALID_BIT, CTL_INVALID_BIT, 0, 1, 2, 3, CTL_INVALID_BIT, CTL_INVALID_BIT}; -static const struct dpu_ctl_cfg *_ctl_offset(enum dpu_ctl ctl, - const struct dpu_mdss_cfg *m, - void __iomem *addr, - struct dpu_hw_blk_reg_map *b) -{ - int i; - - for (i = 0; i < m->ctl_count; i++) { - if (ctl == m->ctl[i].id) { - b->blk_addr = addr + m->ctl[i].base; - b->log_mask = DPU_DBG_MASK_CTL; - return &m->ctl[i]; - } - } - return ERR_PTR(-ENOMEM); -} - static int _mixer_stages(const struct dpu_lm_cfg *mixer, int count, enum dpu_lm lm) { @@ -117,6 +100,10 @@ static inline void dpu_hw_ctl_clear_pending_flush(struct dpu_hw_ctl *ctx) trace_dpu_hw_ctl_clear_pending_flush(ctx->pending_flush_mask, dpu_hw_ctl_get_flush_register(ctx)); ctx->pending_flush_mask = 0x0; + ctx->pending_intf_flush_mask = 0; + ctx->pending_wb_flush_mask = 0; + ctx->pending_merge_3d_flush_mask = 0; + ctx->pending_dsc_flush_mask = 0; memset(ctx->pending_dspp_flush_mask, 0, sizeof(ctx->pending_dspp_flush_mask)); @@ -156,6 +143,11 @@ static inline void dpu_hw_ctl_trigger_flush_v1(struct dpu_hw_ctl *ctx) CTL_DSPP_n_FLUSH(dspp - DSPP_0), ctx->pending_dspp_flush_mask[dspp - DSPP_0]); } + + if (ctx->pending_flush_mask & BIT(DSC_IDX)) + DPU_REG_WRITE(&ctx->hw, CTL_DSC_FLUSH, + ctx->pending_dsc_flush_mask); + DPU_REG_WRITE(&ctx->hw, CTL_FLUSH, ctx->pending_flush_mask); } @@ -302,6 +294,13 @@ static void dpu_hw_ctl_update_pending_flush_merge_3d_v1(struct dpu_hw_ctl *ctx, ctx->pending_flush_mask |= BIT(MERGE_3D_IDX); } +static void dpu_hw_ctl_update_pending_flush_dsc_v1(struct dpu_hw_ctl *ctx, + enum dpu_dsc dsc_num) +{ + ctx->pending_dsc_flush_mask |= BIT(dsc_num - DSC_0); + ctx->pending_flush_mask |= BIT(DSC_IDX); +} + static void dpu_hw_ctl_update_pending_flush_dspp(struct dpu_hw_ctl *ctx, enum dpu_dspp dspp, u32 dspp_sub_blk) { @@ -330,15 +329,9 @@ static void dpu_hw_ctl_update_pending_flush_dspp_sub_blocks( return; switch (dspp_sub_blk) { - case DPU_DSPP_IGC: - ctx->pending_dspp_flush_mask[dspp - DSPP_0] |= BIT(2); - break; case DPU_DSPP_PCC: ctx->pending_dspp_flush_mask[dspp - DSPP_0] |= BIT(4); break; - case DPU_DSPP_GC: - ctx->pending_dspp_flush_mask[dspp - DSPP_0] |= BIT(5); - break; default: return; } @@ -519,9 +512,6 @@ static void dpu_hw_ctl_intf_cfg_v1(struct dpu_hw_ctl *ctx, if ((test_bit(DPU_CTL_VM_CFG, &ctx->caps->features))) mode_sel = CTL_DEFAULT_GROUP_ID << 28; - if (cfg->dsc) - DPU_REG_WRITE(&ctx->hw, CTL_DSC_FLUSH, cfg->dsc); - if (cfg->intf_mode_sel == DPU_CTL_MODE_SEL_CMD) mode_sel |= BIT(17); @@ -541,10 +531,9 @@ static void dpu_hw_ctl_intf_cfg_v1(struct dpu_hw_ctl *ctx, if (cfg->merge_3d) DPU_REG_WRITE(c, CTL_MERGE_3D_ACTIVE, BIT(cfg->merge_3d - MERGE_3D_0)); - if (cfg->dsc) { - DPU_REG_WRITE(&ctx->hw, CTL_FLUSH, DSC_IDX); + + if (cfg->dsc) DPU_REG_WRITE(c, CTL_DSC_ACTIVE, cfg->dsc); - } } static void dpu_hw_ctl_intf_cfg(struct dpu_hw_ctl *ctx, @@ -587,6 +576,7 @@ static void dpu_hw_ctl_reset_intf_cfg_v1(struct dpu_hw_ctl *ctx, u32 intf_active = 0; u32 wb_active = 0; u32 merge3d_active = 0; + u32 dsc_active; /* * This API resets each portion of the CTL path namely, @@ -616,6 +606,12 @@ static void dpu_hw_ctl_reset_intf_cfg_v1(struct dpu_hw_ctl *ctx, wb_active &= ~BIT(cfg->wb - WB_0); DPU_REG_WRITE(c, CTL_WB_ACTIVE, wb_active); } + + if (cfg->dsc) { + dsc_active = DPU_REG_READ(c, CTL_DSC_ACTIVE); + dsc_active &= ~cfg->dsc; + DPU_REG_WRITE(c, CTL_DSC_ACTIVE, dsc_active); + } } static void dpu_hw_ctl_set_fetch_pipe_active(struct dpu_hw_ctl *ctx, @@ -647,6 +643,8 @@ static void _setup_ctl_ops(struct dpu_hw_ctl_ops *ops, ops->update_pending_flush_merge_3d = dpu_hw_ctl_update_pending_flush_merge_3d_v1; ops->update_pending_flush_wb = dpu_hw_ctl_update_pending_flush_wb_v1; + ops->update_pending_flush_dsc = + dpu_hw_ctl_update_pending_flush_dsc_v1; } else { ops->trigger_flush = dpu_hw_ctl_trigger_flush; ops->setup_intf_cfg = dpu_hw_ctl_intf_cfg; @@ -676,29 +674,25 @@ static void _setup_ctl_ops(struct dpu_hw_ctl_ops *ops, ops->set_active_pipes = dpu_hw_ctl_set_fetch_pipe_active; }; -struct dpu_hw_ctl *dpu_hw_ctl_init(enum dpu_ctl idx, +struct dpu_hw_ctl *dpu_hw_ctl_init(const struct dpu_ctl_cfg *cfg, void __iomem *addr, - const struct dpu_mdss_cfg *m) + u32 mixer_count, + const struct dpu_lm_cfg *mixer) { struct dpu_hw_ctl *c; - const struct dpu_ctl_cfg *cfg; c = kzalloc(sizeof(*c), GFP_KERNEL); if (!c) return ERR_PTR(-ENOMEM); - cfg = _ctl_offset(idx, m, addr, &c->hw); - if (IS_ERR_OR_NULL(cfg)) { - kfree(c); - pr_err("failed to create dpu_hw_ctl %d\n", idx); - return ERR_PTR(-EINVAL); - } + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_CTL; c->caps = cfg; _setup_ctl_ops(&c->ops, c->caps->features); - c->idx = idx; - c->mixer_count = m->mixer_count; - c->mixer_hw_caps = m->mixer; + c->idx = cfg->id; + c->mixer_count = mixer_count; + c->mixer_hw_caps = mixer; return c; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h index 78611a8316974262756ec281d83ba0ebb36a5813..1c242298ff2ee0d47e2bc90f2c73259f79f7b746 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h @@ -157,6 +157,15 @@ struct dpu_hw_ctl_ops { void (*update_pending_flush_dspp)(struct dpu_hw_ctl *ctx, enum dpu_dspp blk, u32 dspp_sub_blk); + /** + * OR in the given flushbits to the cached pending_(dsc_)flush_mask + * No effect on hardware + * @ctx: ctl path ctx pointer + * @blk: interface block index + */ + void (*update_pending_flush_dsc)(struct dpu_hw_ctl *ctx, + enum dpu_dsc blk); + /** * Write the value of the pending_flush_mask to hardware * @ctx : ctl path ctx pointer @@ -229,6 +238,7 @@ struct dpu_hw_ctl_ops { * @pending_flush_mask: storage for pending ctl_flush managed via ops * @pending_intf_flush_mask: pending INTF flush * @pending_wb_flush_mask: pending WB flush + * @pending_dsc_flush_mask: pending DSC flush * @ops: operation list */ struct dpu_hw_ctl { @@ -245,6 +255,7 @@ struct dpu_hw_ctl { u32 pending_wb_flush_mask; u32 pending_merge_3d_flush_mask; u32 pending_dspp_flush_mask[DSPP_MAX - DSPP_0]; + u32 pending_dsc_flush_mask; /* ops */ struct dpu_hw_ctl_ops ops; @@ -261,15 +272,17 @@ static inline struct dpu_hw_ctl *to_dpu_hw_ctl(struct dpu_hw_blk *hw) } /** - * dpu_hw_ctl_init(): Initializes the ctl_path hw driver object. - * should be called before accessing every ctl path registers. - * @idx: ctl_path index for which driver object is required + * dpu_hw_ctl_init() - Initializes the ctl_path hw driver object. + * Should be called before accessing any ctl_path register. + * @cfg: ctl_path catalog entry for which driver object is required * @addr: mapped register io address of MDP - * @m : pointer to mdss catalog data + * @mixer_count: Number of mixers in @mixer + * @mixer: Pointer to an array of Layer Mixers defined in the catalog */ -struct dpu_hw_ctl *dpu_hw_ctl_init(enum dpu_ctl idx, +struct dpu_hw_ctl *dpu_hw_ctl_init(const struct dpu_ctl_cfg *cfg, void __iomem *addr, - const struct dpu_mdss_cfg *m); + u32 mixer_count, + const struct dpu_lm_cfg *mixer); /** * dpu_hw_ctl_destroy(): Destroys ctl driver context diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c index 4e1396575e6aaaed2d0a25644f1d5d5c399bb504..509dbaa51d87864c05adf1ec026b0327bd7dafb6 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c @@ -3,6 +3,8 @@ * Copyright (c) 2020-2022, Linaro Limited */ +#include + #include "dpu_kms.h" #include "dpu_hw_catalog.h" #include "dpu_hwio.h" @@ -54,9 +56,10 @@ static void dpu_hw_dsc_config(struct dpu_hw_dsc *hw_dsc, if (is_cmd_mode) initial_lines += 1; - slice_last_group_size = 3 - (dsc->slice_width % 3); + slice_last_group_size = (dsc->slice_width + 2) % 3; + data = (initial_lines << 20); - data |= ((slice_last_group_size - 1) << 18); + data |= (slice_last_group_size << 18); /* bpp is 6.4 format, 4 LSBs bits are for fractional part */ data |= (dsc->bits_per_pixel << 8); data |= (dsc->block_pred_enable << 7); @@ -102,7 +105,7 @@ static void dpu_hw_dsc_config(struct dpu_hw_dsc *hw_dsc, data |= dsc->final_offset; DPU_REG_WRITE(c, DSC_DSC_OFFSET, data); - det_thresh_flatness = 7 + 2 * (dsc->bits_per_component - 8); + det_thresh_flatness = drm_dsc_flatness_det_thresh(dsc); data = det_thresh_flatness << 10; data |= dsc->flatness_max_qp << 5; data |= dsc->flatness_min_qp; @@ -154,7 +157,6 @@ static void dpu_hw_dsc_config_thresh(struct dpu_hw_dsc *hw_dsc, static void dpu_hw_dsc_bind_pingpong_blk( struct dpu_hw_dsc *hw_dsc, - bool enable, const enum dpu_pingpong pp) { struct dpu_hw_blk_reg_map *c = &hw_dsc->hw; @@ -163,36 +165,19 @@ static void dpu_hw_dsc_bind_pingpong_blk( dsc_ctl_offset = DSC_CTL(hw_dsc->idx); - if (enable) + if (pp) mux_cfg = (pp - PINGPONG_0) & 0x7; - DRM_DEBUG_KMS("%s dsc:%d %s pp:%d\n", - enable ? "Binding" : "Unbinding", - hw_dsc->idx - DSC_0, - enable ? "to" : "from", - pp - PINGPONG_0); + if (pp) + DRM_DEBUG_KMS("Binding dsc:%d to pp:%d\n", + hw_dsc->idx - DSC_0, pp - PINGPONG_0); + else + DRM_DEBUG_KMS("Unbinding dsc:%d from any pp\n", + hw_dsc->idx - DSC_0); DPU_REG_WRITE(c, dsc_ctl_offset, mux_cfg); } -static const struct dpu_dsc_cfg *_dsc_offset(enum dpu_dsc dsc, - const struct dpu_mdss_cfg *m, - void __iomem *addr, - struct dpu_hw_blk_reg_map *b) -{ - int i; - - for (i = 0; i < m->dsc_count; i++) { - if (dsc == m->dsc[i].id) { - b->blk_addr = addr + m->dsc[i].base; - b->log_mask = DPU_DBG_MASK_DSC; - return &m->dsc[i]; - } - } - - return NULL; -} - static void _setup_dsc_ops(struct dpu_hw_dsc_ops *ops, unsigned long cap) { @@ -203,23 +188,19 @@ static void _setup_dsc_ops(struct dpu_hw_dsc_ops *ops, ops->dsc_bind_pingpong_blk = dpu_hw_dsc_bind_pingpong_blk; }; -struct dpu_hw_dsc *dpu_hw_dsc_init(enum dpu_dsc idx, void __iomem *addr, - const struct dpu_mdss_cfg *m) +struct dpu_hw_dsc *dpu_hw_dsc_init(const struct dpu_dsc_cfg *cfg, + void __iomem *addr) { struct dpu_hw_dsc *c; - const struct dpu_dsc_cfg *cfg; c = kzalloc(sizeof(*c), GFP_KERNEL); if (!c) return ERR_PTR(-ENOMEM); - cfg = _dsc_offset(idx, m, addr, &c->hw); - if (IS_ERR_OR_NULL(cfg)) { - kfree(c); - return ERR_PTR(-EINVAL); - } + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_DSC; - c->idx = idx; + c->idx = cfg->id; c->caps = cfg; _setup_dsc_ops(&c->ops, c->caps->features); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h index ae9b5db53d7f648056a58e17dc1b240dc65d197a..d5b597ab8c5ca199cf3d9e19b26be4a2fc546057 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h @@ -1,5 +1,8 @@ /* SPDX-License-Identifier: GPL-2.0-only */ -/* Copyright (c) 2020-2022, Linaro Limited */ +/* + * Copyright (c) 2020-2022, Linaro Limited + * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved + */ #ifndef _DPU_HW_DSC_H #define _DPU_HW_DSC_H @@ -44,7 +47,6 @@ struct dpu_hw_dsc_ops { struct drm_dsc_config *dsc); void (*dsc_bind_pingpong_blk)(struct dpu_hw_dsc *hw_dsc, - bool enable, enum dpu_pingpong pp); }; @@ -61,14 +63,22 @@ struct dpu_hw_dsc { }; /** - * dpu_hw_dsc_init - initializes the dsc block for the passed dsc idx. - * @idx: DSC index for which driver object is required + * dpu_hw_dsc_init() - Initializes the DSC hw driver object. + * @cfg: DSC catalog entry for which driver object is required + * @addr: Mapped register io address of MDP + * Return: Error code or allocated dpu_hw_dsc context + */ +struct dpu_hw_dsc *dpu_hw_dsc_init(const struct dpu_dsc_cfg *cfg, + void __iomem *addr); + +/** + * dpu_hw_dsc_init_1_2() - initializes the v1.2 DSC hw driver object + * @cfg: DSC catalog entry for which driver object is required * @addr: Mapped register io address of MDP - * @m: Pointer to mdss catalog data * Returns: Error code or allocated dpu_hw_dsc context */ -struct dpu_hw_dsc *dpu_hw_dsc_init(enum dpu_dsc idx, void __iomem *addr, - const struct dpu_mdss_cfg *m); +struct dpu_hw_dsc *dpu_hw_dsc_init_1_2(const struct dpu_dsc_cfg *cfg, + void __iomem *addr); /** * dpu_hw_dsc_destroy - destroys dsc driver context diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc_1_2.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc_1_2.c new file mode 100644 index 0000000000000000000000000000000000000000..24fe1d98eb86aed2a922b21967adaaee19161b6a --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc_1_2.c @@ -0,0 +1,387 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2020-2021, The Linux Foundation. All rights reserved. + * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved + */ + +#include + +#include "dpu_kms.h" +#include "dpu_hw_catalog.h" +#include "dpu_hwio.h" +#include "dpu_hw_mdss.h" +#include "dpu_hw_dsc.h" + +#define DSC_CMN_MAIN_CNF 0x00 + +/* DPU_DSC_ENC register offsets */ +#define ENC_DF_CTRL 0x00 +#define ENC_GENERAL_STATUS 0x04 +#define ENC_HSLICE_STATUS 0x08 +#define ENC_OUT_STATUS 0x0C +#define ENC_INT_STAT 0x10 +#define ENC_INT_CLR 0x14 +#define ENC_INT_MASK 0x18 +#define DSC_MAIN_CONF 0x30 +#define DSC_PICTURE_SIZE 0x34 +#define DSC_SLICE_SIZE 0x38 +#define DSC_MISC_SIZE 0x3C +#define DSC_HRD_DELAYS 0x40 +#define DSC_RC_SCALE 0x44 +#define DSC_RC_SCALE_INC_DEC 0x48 +#define DSC_RC_OFFSETS_1 0x4C +#define DSC_RC_OFFSETS_2 0x50 +#define DSC_RC_OFFSETS_3 0x54 +#define DSC_RC_OFFSETS_4 0x58 +#define DSC_FLATNESS_QP 0x5C +#define DSC_RC_MODEL_SIZE 0x60 +#define DSC_RC_CONFIG 0x64 +#define DSC_RC_BUF_THRESH_0 0x68 +#define DSC_RC_BUF_THRESH_1 0x6C +#define DSC_RC_BUF_THRESH_2 0x70 +#define DSC_RC_BUF_THRESH_3 0x74 +#define DSC_RC_MIN_QP_0 0x78 +#define DSC_RC_MIN_QP_1 0x7C +#define DSC_RC_MIN_QP_2 0x80 +#define DSC_RC_MAX_QP_0 0x84 +#define DSC_RC_MAX_QP_1 0x88 +#define DSC_RC_MAX_QP_2 0x8C +#define DSC_RC_RANGE_BPG_OFFSETS_0 0x90 +#define DSC_RC_RANGE_BPG_OFFSETS_1 0x94 +#define DSC_RC_RANGE_BPG_OFFSETS_2 0x98 + +/* DPU_DSC_CTL register offsets */ +#define DSC_CTL 0x00 +#define DSC_CFG 0x04 +#define DSC_DATA_IN_SWAP 0x08 +#define DSC_CLK_CTRL 0x0C + +static int _dsc_calc_output_buf_max_addr(struct dpu_hw_dsc *hw_dsc, int num_softslice) +{ + int max_addr = 2400 / num_softslice; + + if (hw_dsc->caps->features & BIT(DPU_DSC_NATIVE_42x_EN)) + max_addr /= 2; + + return max_addr - 1; +}; + +static void dpu_hw_dsc_disable_1_2(struct dpu_hw_dsc *hw_dsc) +{ + struct dpu_hw_blk_reg_map *hw; + const struct dpu_dsc_sub_blks *sblk; + + if (!hw_dsc) + return; + + hw = &hw_dsc->hw; + sblk = hw_dsc->caps->sblk; + DPU_REG_WRITE(hw, sblk->ctl.base + DSC_CFG, 0); + + DPU_REG_WRITE(hw, sblk->enc.base + ENC_DF_CTRL, 0); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_MAIN_CONF, 0); +} + +static void dpu_hw_dsc_config_1_2(struct dpu_hw_dsc *hw_dsc, + struct drm_dsc_config *dsc, + u32 mode, + u32 initial_lines) +{ + struct dpu_hw_blk_reg_map *hw; + const struct dpu_dsc_sub_blks *sblk; + u32 data = 0; + u32 det_thresh_flatness; + u32 num_active_slice_per_enc; + u32 bpp; + + if (!hw_dsc || !dsc) + return; + + hw = &hw_dsc->hw; + + sblk = hw_dsc->caps->sblk; + + if (mode & DSC_MODE_SPLIT_PANEL) + data |= BIT(0); + + if (mode & DSC_MODE_MULTIPLEX) + data |= BIT(1); + + num_active_slice_per_enc = dsc->slice_count; + if (mode & DSC_MODE_MULTIPLEX) + num_active_slice_per_enc = dsc->slice_count / 2; + + data |= (num_active_slice_per_enc & 0x3) << 7; + + DPU_REG_WRITE(hw, DSC_CMN_MAIN_CNF, data); + + data = (initial_lines & 0xff); + + if (mode & DSC_MODE_VIDEO) + data |= BIT(9); + + data |= (_dsc_calc_output_buf_max_addr(hw_dsc, num_active_slice_per_enc) << 18); + + DPU_REG_WRITE(hw, sblk->enc.base + ENC_DF_CTRL, data); + + data = (dsc->dsc_version_minor & 0xf) << 28; + if (dsc->dsc_version_minor == 0x2) { + if (dsc->native_422) + data |= BIT(22); + if (dsc->native_420) + data |= BIT(21); + } + + bpp = dsc->bits_per_pixel; + /* as per hw requirement bpp should be programmed + * twice the actual value in case of 420 or 422 encoding + */ + if (dsc->native_422 || dsc->native_420) + bpp = 2 * bpp; + + data |= bpp << 10; + + if (dsc->block_pred_enable) + data |= BIT(20); + + if (dsc->convert_rgb) + data |= BIT(4); + + data |= (dsc->line_buf_depth & 0xf) << 6; + data |= dsc->bits_per_component & 0xf; + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_MAIN_CONF, data); + + data = (dsc->pic_width & 0xffff) | + ((dsc->pic_height & 0xffff) << 16); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_PICTURE_SIZE, data); + + data = (dsc->slice_width & 0xffff) | + ((dsc->slice_height & 0xffff) << 16); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_SLICE_SIZE, data); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_MISC_SIZE, + (dsc->slice_chunk_size) & 0xffff); + + data = (dsc->initial_xmit_delay & 0xffff) | + ((dsc->initial_dec_delay & 0xffff) << 16); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_HRD_DELAYS, data); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_SCALE, + dsc->initial_scale_value & 0x3f); + + data = (dsc->scale_increment_interval & 0xffff) | + ((dsc->scale_decrement_interval & 0x7ff) << 16); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_SCALE_INC_DEC, data); + + data = (dsc->first_line_bpg_offset & 0x1f) | + ((dsc->second_line_bpg_offset & 0x1f) << 5); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_OFFSETS_1, data); + + data = (dsc->nfl_bpg_offset & 0xffff) | + ((dsc->slice_bpg_offset & 0xffff) << 16); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_OFFSETS_2, data); + + data = (dsc->initial_offset & 0xffff) | + ((dsc->final_offset & 0xffff) << 16); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_OFFSETS_3, data); + + data = (dsc->nsl_bpg_offset & 0xffff) | + ((dsc->second_line_offset_adj & 0xffff) << 16); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_OFFSETS_4, data); + + det_thresh_flatness = drm_dsc_flatness_det_thresh(dsc); + data = (dsc->flatness_min_qp & 0x1f) | + ((dsc->flatness_max_qp & 0x1f) << 5) | + ((det_thresh_flatness & 0xff) << 10); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_FLATNESS_QP, data); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MODEL_SIZE, + (dsc->rc_model_size) & 0xffff); + + data = dsc->rc_edge_factor & 0xf; + data |= (dsc->rc_quant_incr_limit0 & 0x1f) << 8; + data |= (dsc->rc_quant_incr_limit1 & 0x1f) << 13; + data |= (dsc->rc_tgt_offset_high & 0xf) << 20; + data |= (dsc->rc_tgt_offset_low & 0xf) << 24; + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_CONFIG, data); + + /* program the dsc wrapper */ + data = BIT(0); /* encoder enable */ + if (dsc->native_422) + data |= BIT(8); + else if (dsc->native_420) + data |= BIT(9); + if (!dsc->convert_rgb) + data |= BIT(10); + if (dsc->bits_per_component == 8) + data |= BIT(11); + if (mode & DSC_MODE_SPLIT_PANEL) + data |= BIT(12); + if (mode & DSC_MODE_MULTIPLEX) + data |= BIT(13); + if (!(mode & DSC_MODE_VIDEO)) + data |= BIT(17); + + DPU_REG_WRITE(hw, sblk->ctl.base + DSC_CFG, data); +} + +static void dpu_hw_dsc_config_thresh_1_2(struct dpu_hw_dsc *hw_dsc, + struct drm_dsc_config *dsc) +{ + struct dpu_hw_blk_reg_map *hw; + const struct dpu_dsc_sub_blks *sblk; + struct drm_dsc_rc_range_parameters *rc; + + if (!hw_dsc || !dsc) + return; + + hw = &hw_dsc->hw; + + sblk = hw_dsc->caps->sblk; + + rc = dsc->rc_range_params; + + /* + * With BUF_THRESH -- 14 in total + * each register contains 4 thresh values with the last register + * containing only 2 thresh values + */ + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_BUF_THRESH_0, + (dsc->rc_buf_thresh[0] << 0) | + (dsc->rc_buf_thresh[1] << 8) | + (dsc->rc_buf_thresh[2] << 16) | + (dsc->rc_buf_thresh[3] << 24)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_BUF_THRESH_1, + (dsc->rc_buf_thresh[4] << 0) | + (dsc->rc_buf_thresh[5] << 8) | + (dsc->rc_buf_thresh[6] << 16) | + (dsc->rc_buf_thresh[7] << 24)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_BUF_THRESH_2, + (dsc->rc_buf_thresh[8] << 0) | + (dsc->rc_buf_thresh[9] << 8) | + (dsc->rc_buf_thresh[10] << 16) | + (dsc->rc_buf_thresh[11] << 24)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_BUF_THRESH_3, + (dsc->rc_buf_thresh[12] << 0) | + (dsc->rc_buf_thresh[13] << 8)); + + /* + * with min/max_QP -- 5 bits + * each register contains 5 min_qp or max_qp for total of 15 + * + * With BPG_OFFSET -- 6 bits + * each register contains 5 BPG_offset for total of 15 + */ + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MIN_QP_0, + (rc[0].range_min_qp << 0) | + (rc[1].range_min_qp << 5) | + (rc[2].range_min_qp << 10) | + (rc[3].range_min_qp << 15) | + (rc[4].range_min_qp << 20)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MAX_QP_0, + (rc[0].range_max_qp << 0) | + (rc[1].range_max_qp << 5) | + (rc[2].range_max_qp << 10) | + (rc[3].range_max_qp << 15) | + (rc[4].range_max_qp << 20)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_RANGE_BPG_OFFSETS_0, + (rc[0].range_bpg_offset << 0) | + (rc[1].range_bpg_offset << 6) | + (rc[2].range_bpg_offset << 12) | + (rc[3].range_bpg_offset << 18) | + (rc[4].range_bpg_offset << 24)); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MIN_QP_1, + (rc[5].range_min_qp << 0) | + (rc[6].range_min_qp << 5) | + (rc[7].range_min_qp << 10) | + (rc[8].range_min_qp << 15) | + (rc[9].range_min_qp << 20)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MAX_QP_1, + (rc[5].range_max_qp << 0) | + (rc[6].range_max_qp << 5) | + (rc[7].range_max_qp << 10) | + (rc[8].range_max_qp << 15) | + (rc[9].range_max_qp << 20)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_RANGE_BPG_OFFSETS_1, + (rc[5].range_bpg_offset << 0) | + (rc[6].range_bpg_offset << 6) | + (rc[7].range_bpg_offset << 12) | + (rc[8].range_bpg_offset << 18) | + (rc[9].range_bpg_offset << 24)); + + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MIN_QP_2, + (rc[10].range_min_qp << 0) | + (rc[11].range_min_qp << 5) | + (rc[12].range_min_qp << 10) | + (rc[13].range_min_qp << 15) | + (rc[14].range_min_qp << 20)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_MAX_QP_2, + (rc[10].range_max_qp << 0) | + (rc[11].range_max_qp << 5) | + (rc[12].range_max_qp << 10) | + (rc[13].range_max_qp << 15) | + (rc[14].range_max_qp << 20)); + DPU_REG_WRITE(hw, sblk->enc.base + DSC_RC_RANGE_BPG_OFFSETS_2, + (rc[10].range_bpg_offset << 0) | + (rc[11].range_bpg_offset << 6) | + (rc[12].range_bpg_offset << 12) | + (rc[13].range_bpg_offset << 18) | + (rc[14].range_bpg_offset << 24)); +} + +static void dpu_hw_dsc_bind_pingpong_blk_1_2(struct dpu_hw_dsc *hw_dsc, + const enum dpu_pingpong pp) +{ + struct dpu_hw_blk_reg_map *hw; + const struct dpu_dsc_sub_blks *sblk; + int mux_cfg = 0xf; /* Disabled */ + + hw = &hw_dsc->hw; + + sblk = hw_dsc->caps->sblk; + + if (pp) + mux_cfg = (pp - PINGPONG_0) & 0x7; + + DPU_REG_WRITE(hw, sblk->ctl.base + DSC_CTL, mux_cfg); +} + +static void _setup_dcs_ops_1_2(struct dpu_hw_dsc_ops *ops, + const unsigned long features) +{ + ops->dsc_disable = dpu_hw_dsc_disable_1_2; + ops->dsc_config = dpu_hw_dsc_config_1_2; + ops->dsc_config_thresh = dpu_hw_dsc_config_thresh_1_2; + ops->dsc_bind_pingpong_blk = dpu_hw_dsc_bind_pingpong_blk_1_2; +} + +struct dpu_hw_dsc *dpu_hw_dsc_init_1_2(const struct dpu_dsc_cfg *cfg, + void __iomem *addr) +{ + struct dpu_hw_dsc *c; + + c = kzalloc(sizeof(*c), GFP_KERNEL); + if (!c) + return ERR_PTR(-ENOMEM); + + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_DSC; + + c->idx = cfg->id; + c->caps = cfg; + _setup_dcs_ops_1_2(&c->ops, c->caps->features); + + return c; +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c index 8ab5ace34a2d5b7d0b372e7bfc46b7cd680b0e18..9419b2209af88ef500ae68552b5ca53f50778c5a 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c @@ -68,49 +68,23 @@ static void _setup_dspp_ops(struct dpu_hw_dspp *c, c->ops.setup_pcc = dpu_setup_dspp_pcc; } -static const struct dpu_dspp_cfg *_dspp_offset(enum dpu_dspp dspp, - const struct dpu_mdss_cfg *m, - void __iomem *addr, - struct dpu_hw_blk_reg_map *b) -{ - int i; - - if (!m || !addr || !b) - return ERR_PTR(-EINVAL); - - for (i = 0; i < m->dspp_count; i++) { - if (dspp == m->dspp[i].id) { - b->blk_addr = addr + m->dspp[i].base; - b->log_mask = DPU_DBG_MASK_DSPP; - return &m->dspp[i]; - } - } - - return ERR_PTR(-EINVAL); -} - -struct dpu_hw_dspp *dpu_hw_dspp_init(enum dpu_dspp idx, - void __iomem *addr, - const struct dpu_mdss_cfg *m) +struct dpu_hw_dspp *dpu_hw_dspp_init(const struct dpu_dspp_cfg *cfg, + void __iomem *addr) { struct dpu_hw_dspp *c; - const struct dpu_dspp_cfg *cfg; - if (!addr || !m) + if (!addr) return ERR_PTR(-EINVAL); c = kzalloc(sizeof(*c), GFP_KERNEL); if (!c) return ERR_PTR(-ENOMEM); - cfg = _dspp_offset(idx, m, addr, &c->hw); - if (IS_ERR_OR_NULL(cfg)) { - kfree(c); - return ERR_PTR(-EINVAL); - } + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_DSPP; /* Assign ops */ - c->idx = idx; + c->idx = cfg->id; c->cap = cfg; _setup_dspp_ops(c, c->cap->features); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.h index 05ecfdfac93b7d146aeef1b84653cfa3be5ab773..bea96568133011109ab6d5bf3dd1ec4216e62ab1 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.h @@ -79,14 +79,14 @@ static inline struct dpu_hw_dspp *to_dpu_hw_dspp(struct dpu_hw_blk *hw) } /** - * dpu_hw_dspp_init - initializes the dspp hw driver object. - * should be called once before accessing every dspp. - * @idx: DSPP index for which driver object is required + * dpu_hw_dspp_init() - Initializes the DSPP hw driver object. + * should be called once before accessing every DSPP. + * @cfg: DSPP catalog entry for which driver object is required * @addr: Mapped register io address of MDP - * @Return: pointer to structure or ERR_PTR + * Return: pointer to structure or ERR_PTR */ -struct dpu_hw_dspp *dpu_hw_dspp_init(enum dpu_dspp idx, - void __iomem *addr, const struct dpu_mdss_cfg *m); +struct dpu_hw_dspp *dpu_hw_dspp_init(const struct dpu_dspp_cfg *cfg, + void __iomem *addr); /** * dpu_hw_dspp_destroy(): Destroys DSPP driver context diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c index 53326f25e40ef10b22bb09da70b829d4defa0553..5e2d68ebb113e06f6eb4cc164d5bf904ccdb16c1 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c @@ -15,29 +15,28 @@ /* * Register offsets in MDSS register file for the interrupt registers - * w.r.t. to the MDP base + * w.r.t. the MDP base */ -#define MDP_SSPP_TOP0_OFF 0x0 -#define MDP_INTF_0_OFF 0x6A000 -#define MDP_INTF_1_OFF 0x6A800 -#define MDP_INTF_2_OFF 0x6B000 -#define MDP_INTF_3_OFF 0x6B800 -#define MDP_INTF_4_OFF 0x6C000 -#define MDP_INTF_5_OFF 0x6C800 -#define MDP_AD4_0_OFF 0x7C000 -#define MDP_AD4_1_OFF 0x7D000 -#define MDP_AD4_INTR_EN_OFF 0x41c -#define MDP_AD4_INTR_CLEAR_OFF 0x424 -#define MDP_AD4_INTR_STATUS_OFF 0x420 -#define MDP_INTF_0_OFF_REV_7xxx 0x34000 -#define MDP_INTF_1_OFF_REV_7xxx 0x35000 -#define MDP_INTF_2_OFF_REV_7xxx 0x36000 -#define MDP_INTF_3_OFF_REV_7xxx 0x37000 -#define MDP_INTF_4_OFF_REV_7xxx 0x38000 -#define MDP_INTF_5_OFF_REV_7xxx 0x39000 -#define MDP_INTF_6_OFF_REV_7xxx 0x3a000 -#define MDP_INTF_7_OFF_REV_7xxx 0x3b000 -#define MDP_INTF_8_OFF_REV_7xxx 0x3c000 +#define MDP_INTF_OFF(intf) (0x6A000 + 0x800 * (intf)) +#define MDP_INTF_INTR_EN(intf) (MDP_INTF_OFF(intf) + 0x1c0) +#define MDP_INTF_INTR_STATUS(intf) (MDP_INTF_OFF(intf) + 0x1c4) +#define MDP_INTF_INTR_CLEAR(intf) (MDP_INTF_OFF(intf) + 0x1c8) +#define MDP_INTF_TEAR_OFF(intf) (0x6D700 + 0x100 * (intf)) +#define MDP_INTF_INTR_TEAR_EN(intf) (MDP_INTF_TEAR_OFF(intf) + 0x000) +#define MDP_INTF_INTR_TEAR_STATUS(intf) (MDP_INTF_TEAR_OFF(intf) + 0x004) +#define MDP_INTF_INTR_TEAR_CLEAR(intf) (MDP_INTF_TEAR_OFF(intf) + 0x008) +#define MDP_AD4_OFF(ad4) (0x7C000 + 0x1000 * (ad4)) +#define MDP_AD4_INTR_EN_OFF(ad4) (MDP_AD4_OFF(ad4) + 0x41c) +#define MDP_AD4_INTR_CLEAR_OFF(ad4) (MDP_AD4_OFF(ad4) + 0x424) +#define MDP_AD4_INTR_STATUS_OFF(ad4) (MDP_AD4_OFF(ad4) + 0x420) +#define MDP_INTF_REV_7xxx_OFF(intf) (0x34000 + 0x1000 * (intf)) +#define MDP_INTF_REV_7xxx_INTR_EN(intf) (MDP_INTF_REV_7xxx_OFF(intf) + 0x1c0) +#define MDP_INTF_REV_7xxx_INTR_STATUS(intf) (MDP_INTF_REV_7xxx_OFF(intf) + 0x1c4) +#define MDP_INTF_REV_7xxx_INTR_CLEAR(intf) (MDP_INTF_REV_7xxx_OFF(intf) + 0x1c8) +#define MDP_INTF_REV_7xxx_TEAR_OFF(intf) (0x34800 + 0x1000 * (intf)) +#define MDP_INTF_REV_7xxx_INTR_TEAR_EN(intf) (MDP_INTF_REV_7xxx_TEAR_OFF(intf) + 0x000) +#define MDP_INTF_REV_7xxx_INTR_TEAR_STATUS(intf) (MDP_INTF_REV_7xxx_TEAR_OFF(intf) + 0x004) +#define MDP_INTF_REV_7xxx_INTR_TEAR_CLEAR(intf) (MDP_INTF_REV_7xxx_TEAR_OFF(intf) + 0x008) /** * struct dpu_intr_reg - array of DPU register sets @@ -58,104 +57,124 @@ struct dpu_intr_reg { */ static const struct dpu_intr_reg dpu_intr_set[] = { [MDP_SSPP_TOP0_INTR] = { - MDP_SSPP_TOP0_OFF+INTR_CLEAR, - MDP_SSPP_TOP0_OFF+INTR_EN, - MDP_SSPP_TOP0_OFF+INTR_STATUS + INTR_CLEAR, + INTR_EN, + INTR_STATUS }, [MDP_SSPP_TOP0_INTR2] = { - MDP_SSPP_TOP0_OFF+INTR2_CLEAR, - MDP_SSPP_TOP0_OFF+INTR2_EN, - MDP_SSPP_TOP0_OFF+INTR2_STATUS + INTR2_CLEAR, + INTR2_EN, + INTR2_STATUS }, [MDP_SSPP_TOP0_HIST_INTR] = { - MDP_SSPP_TOP0_OFF+HIST_INTR_CLEAR, - MDP_SSPP_TOP0_OFF+HIST_INTR_EN, - MDP_SSPP_TOP0_OFF+HIST_INTR_STATUS + HIST_INTR_CLEAR, + HIST_INTR_EN, + HIST_INTR_STATUS }, [MDP_INTF0_INTR] = { - MDP_INTF_0_OFF+INTF_INTR_CLEAR, - MDP_INTF_0_OFF+INTF_INTR_EN, - MDP_INTF_0_OFF+INTF_INTR_STATUS + MDP_INTF_INTR_CLEAR(0), + MDP_INTF_INTR_EN(0), + MDP_INTF_INTR_STATUS(0) }, [MDP_INTF1_INTR] = { - MDP_INTF_1_OFF+INTF_INTR_CLEAR, - MDP_INTF_1_OFF+INTF_INTR_EN, - MDP_INTF_1_OFF+INTF_INTR_STATUS + MDP_INTF_INTR_CLEAR(1), + MDP_INTF_INTR_EN(1), + MDP_INTF_INTR_STATUS(1) }, [MDP_INTF2_INTR] = { - MDP_INTF_2_OFF+INTF_INTR_CLEAR, - MDP_INTF_2_OFF+INTF_INTR_EN, - MDP_INTF_2_OFF+INTF_INTR_STATUS + MDP_INTF_INTR_CLEAR(2), + MDP_INTF_INTR_EN(2), + MDP_INTF_INTR_STATUS(2) }, [MDP_INTF3_INTR] = { - MDP_INTF_3_OFF+INTF_INTR_CLEAR, - MDP_INTF_3_OFF+INTF_INTR_EN, - MDP_INTF_3_OFF+INTF_INTR_STATUS + MDP_INTF_INTR_CLEAR(3), + MDP_INTF_INTR_EN(3), + MDP_INTF_INTR_STATUS(3) }, [MDP_INTF4_INTR] = { - MDP_INTF_4_OFF+INTF_INTR_CLEAR, - MDP_INTF_4_OFF+INTF_INTR_EN, - MDP_INTF_4_OFF+INTF_INTR_STATUS + MDP_INTF_INTR_CLEAR(4), + MDP_INTF_INTR_EN(4), + MDP_INTF_INTR_STATUS(4) }, [MDP_INTF5_INTR] = { - MDP_INTF_5_OFF+INTF_INTR_CLEAR, - MDP_INTF_5_OFF+INTF_INTR_EN, - MDP_INTF_5_OFF+INTF_INTR_STATUS + MDP_INTF_INTR_CLEAR(5), + MDP_INTF_INTR_EN(5), + MDP_INTF_INTR_STATUS(5) + }, + [MDP_INTF1_TEAR_INTR] = { + MDP_INTF_INTR_TEAR_CLEAR(1), + MDP_INTF_INTR_TEAR_EN(1), + MDP_INTF_INTR_TEAR_STATUS(1) + }, + [MDP_INTF2_TEAR_INTR] = { + MDP_INTF_INTR_TEAR_CLEAR(2), + MDP_INTF_INTR_TEAR_EN(2), + MDP_INTF_INTR_TEAR_STATUS(2) }, [MDP_AD4_0_INTR] = { - MDP_AD4_0_OFF + MDP_AD4_INTR_CLEAR_OFF, - MDP_AD4_0_OFF + MDP_AD4_INTR_EN_OFF, - MDP_AD4_0_OFF + MDP_AD4_INTR_STATUS_OFF, + MDP_AD4_INTR_CLEAR_OFF(0), + MDP_AD4_INTR_EN_OFF(0), + MDP_AD4_INTR_STATUS_OFF(0), }, [MDP_AD4_1_INTR] = { - MDP_AD4_1_OFF + MDP_AD4_INTR_CLEAR_OFF, - MDP_AD4_1_OFF + MDP_AD4_INTR_EN_OFF, - MDP_AD4_1_OFF + MDP_AD4_INTR_STATUS_OFF, + MDP_AD4_INTR_CLEAR_OFF(1), + MDP_AD4_INTR_EN_OFF(1), + MDP_AD4_INTR_STATUS_OFF(1), }, [MDP_INTF0_7xxx_INTR] = { - MDP_INTF_0_OFF_REV_7xxx+INTF_INTR_CLEAR, - MDP_INTF_0_OFF_REV_7xxx+INTF_INTR_EN, - MDP_INTF_0_OFF_REV_7xxx+INTF_INTR_STATUS + MDP_INTF_REV_7xxx_INTR_CLEAR(0), + MDP_INTF_REV_7xxx_INTR_EN(0), + MDP_INTF_REV_7xxx_INTR_STATUS(0) }, [MDP_INTF1_7xxx_INTR] = { - MDP_INTF_1_OFF_REV_7xxx+INTF_INTR_CLEAR, - MDP_INTF_1_OFF_REV_7xxx+INTF_INTR_EN, - MDP_INTF_1_OFF_REV_7xxx+INTF_INTR_STATUS + MDP_INTF_REV_7xxx_INTR_CLEAR(1), + MDP_INTF_REV_7xxx_INTR_EN(1), + MDP_INTF_REV_7xxx_INTR_STATUS(1) + }, + [MDP_INTF1_7xxx_TEAR_INTR] = { + MDP_INTF_REV_7xxx_INTR_TEAR_CLEAR(1), + MDP_INTF_REV_7xxx_INTR_TEAR_EN(1), + MDP_INTF_REV_7xxx_INTR_TEAR_STATUS(1) }, [MDP_INTF2_7xxx_INTR] = { - MDP_INTF_2_OFF_REV_7xxx+INTF_INTR_CLEAR, - MDP_INTF_2_OFF_REV_7xxx+INTF_INTR_EN, - MDP_INTF_2_OFF_REV_7xxx+INTF_INTR_STATUS + MDP_INTF_REV_7xxx_INTR_CLEAR(2), + MDP_INTF_REV_7xxx_INTR_EN(2), + MDP_INTF_REV_7xxx_INTR_STATUS(2) + }, + [MDP_INTF2_7xxx_TEAR_INTR] = { + MDP_INTF_REV_7xxx_INTR_TEAR_CLEAR(2), + MDP_INTF_REV_7xxx_INTR_TEAR_EN(2), + MDP_INTF_REV_7xxx_INTR_TEAR_STATUS(2) }, [MDP_INTF3_7xxx_INTR] = { - MDP_INTF_3_OFF_REV_7xxx+INTF_INTR_CLEAR, - MDP_INTF_3_OFF_REV_7xxx+INTF_INTR_EN, - MDP_INTF_3_OFF_REV_7xxx+INTF_INTR_STATUS + MDP_INTF_REV_7xxx_INTR_CLEAR(3), + MDP_INTF_REV_7xxx_INTR_EN(3), + MDP_INTF_REV_7xxx_INTR_STATUS(3) }, [MDP_INTF4_7xxx_INTR] = { - MDP_INTF_4_OFF_REV_7xxx+INTF_INTR_CLEAR, - MDP_INTF_4_OFF_REV_7xxx+INTF_INTR_EN, - MDP_INTF_4_OFF_REV_7xxx+INTF_INTR_STATUS + MDP_INTF_REV_7xxx_INTR_CLEAR(4), + MDP_INTF_REV_7xxx_INTR_EN(4), + MDP_INTF_REV_7xxx_INTR_STATUS(4) }, [MDP_INTF5_7xxx_INTR] = { - MDP_INTF_5_OFF_REV_7xxx+INTF_INTR_CLEAR, - MDP_INTF_5_OFF_REV_7xxx+INTF_INTR_EN, - MDP_INTF_5_OFF_REV_7xxx+INTF_INTR_STATUS + MDP_INTF_REV_7xxx_INTR_CLEAR(5), + MDP_INTF_REV_7xxx_INTR_EN(5), + MDP_INTF_REV_7xxx_INTR_STATUS(5) }, [MDP_INTF6_7xxx_INTR] = { - MDP_INTF_6_OFF_REV_7xxx+INTF_INTR_CLEAR, - MDP_INTF_6_OFF_REV_7xxx+INTF_INTR_EN, - MDP_INTF_6_OFF_REV_7xxx+INTF_INTR_STATUS + MDP_INTF_REV_7xxx_INTR_CLEAR(6), + MDP_INTF_REV_7xxx_INTR_EN(6), + MDP_INTF_REV_7xxx_INTR_STATUS(6) }, [MDP_INTF7_7xxx_INTR] = { - MDP_INTF_7_OFF_REV_7xxx+INTF_INTR_CLEAR, - MDP_INTF_7_OFF_REV_7xxx+INTF_INTR_EN, - MDP_INTF_7_OFF_REV_7xxx+INTF_INTR_STATUS + MDP_INTF_REV_7xxx_INTR_CLEAR(7), + MDP_INTF_REV_7xxx_INTR_EN(7), + MDP_INTF_REV_7xxx_INTR_STATUS(7) }, [MDP_INTF8_7xxx_INTR] = { - MDP_INTF_8_OFF_REV_7xxx+INTF_INTR_CLEAR, - MDP_INTF_8_OFF_REV_7xxx+INTF_INTR_EN, - MDP_INTF_8_OFF_REV_7xxx+INTF_INTR_STATUS + MDP_INTF_REV_7xxx_INTR_CLEAR(8), + MDP_INTF_REV_7xxx_INTR_EN(8), + MDP_INTF_REV_7xxx_INTR_STATUS(8) }, }; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h index 425465011c807f1caebc51428833166de3e47220..1f2dabc54c2299a7b982ff6b14ccffb55315e008 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h @@ -23,11 +23,15 @@ enum dpu_hw_intr_reg { MDP_INTF3_INTR, MDP_INTF4_INTR, MDP_INTF5_INTR, + MDP_INTF1_TEAR_INTR, + MDP_INTF2_TEAR_INTR, MDP_AD4_0_INTR, MDP_AD4_1_INTR, MDP_INTF0_7xxx_INTR, MDP_INTF1_7xxx_INTR, + MDP_INTF1_7xxx_TEAR_INTR, MDP_INTF2_7xxx_INTR, + MDP_INTF2_7xxx_TEAR_INTR, MDP_INTF3_7xxx_INTR, MDP_INTF4_7xxx_INTR, MDP_INTF5_7xxx_INTR, @@ -67,7 +71,7 @@ struct dpu_hw_intr { /** * dpu_hw_intr_init(): Initializes the interrupts hw object * @addr: mapped register io address of MDP - * @m : pointer to mdss catalog data + * @m: pointer to MDSS catalog data */ struct dpu_hw_intr *dpu_hw_intr_init(void __iomem *addr, const struct dpu_mdss_cfg *m); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c index 84ee2efa9c664e04c89e3d898cd11914c99cc275..5b0f6627e29b856d47314d7d7cf7574e0607cfba 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c @@ -8,6 +8,9 @@ #include "dpu_hw_catalog.h" #include "dpu_hw_intf.h" #include "dpu_kms.h" +#include "dpu_trace.h" + +#include #define INTF_TIMING_ENGINE_EN 0x000 #define INTF_CONFIG 0x004 @@ -36,61 +39,60 @@ #define INTF_CONFIG2 0x060 #define INTF_DISPLAY_DATA_HCTL 0x064 #define INTF_ACTIVE_DATA_HCTL 0x068 + +#define INTF_DSI_CMD_MODE_TRIGGER_EN 0x084 +#define INTF_PANEL_FORMAT 0x090 + #define INTF_FRAME_LINE_COUNT_EN 0x0A8 #define INTF_FRAME_COUNT 0x0AC -#define INTF_LINE_COUNT 0x0B0 - -#define INTF_DEFLICKER_CONFIG 0x0F0 -#define INTF_DEFLICKER_STRNG_COEFF 0x0F4 -#define INTF_DEFLICKER_WEAK_COEFF 0x0F8 - -#define INTF_DSI_CMD_MODE_TRIGGER_EN 0x084 -#define INTF_PANEL_FORMAT 0x090 -#define INTF_TPG_ENABLE 0x100 -#define INTF_TPG_MAIN_CONTROL 0x104 -#define INTF_TPG_VIDEO_CONFIG 0x108 -#define INTF_TPG_COMPONENT_LIMITS 0x10C -#define INTF_TPG_RECTANGLE 0x110 -#define INTF_TPG_INITIAL_VALUE 0x114 -#define INTF_TPG_BLK_WHITE_PATTERN_FRAMES 0x118 -#define INTF_TPG_RGB_MAPPING 0x11C -#define INTF_PROG_FETCH_START 0x170 -#define INTF_PROG_ROT_START 0x174 - -#define INTF_FRAME_LINE_COUNT_EN 0x0A8 -#define INTF_FRAME_COUNT 0x0AC -#define INTF_LINE_COUNT 0x0B0 - -#define INTF_MUX 0x25C -#define INTF_STATUS 0x26C +#define INTF_LINE_COUNT 0x0B0 + +#define INTF_DEFLICKER_CONFIG 0x0F0 +#define INTF_DEFLICKER_STRNG_COEFF 0x0F4 +#define INTF_DEFLICKER_WEAK_COEFF 0x0F8 + +#define INTF_TPG_ENABLE 0x100 +#define INTF_TPG_MAIN_CONTROL 0x104 +#define INTF_TPG_VIDEO_CONFIG 0x108 +#define INTF_TPG_COMPONENT_LIMITS 0x10C +#define INTF_TPG_RECTANGLE 0x110 +#define INTF_TPG_INITIAL_VALUE 0x114 +#define INTF_TPG_BLK_WHITE_PATTERN_FRAMES 0x118 +#define INTF_TPG_RGB_MAPPING 0x11C +#define INTF_PROG_FETCH_START 0x170 +#define INTF_PROG_ROT_START 0x174 + +#define INTF_MISR_CTRL 0x180 +#define INTF_MISR_SIGNATURE 0x184 + +#define INTF_MUX 0x25C +#define INTF_STATUS 0x26C +#define INTF_AVR_CONTROL 0x270 +#define INTF_AVR_MODE 0x274 +#define INTF_AVR_TRIGGER 0x278 +#define INTF_AVR_VTOTAL 0x27C +#define INTF_TEAR_MDP_VSYNC_SEL 0x280 +#define INTF_TEAR_TEAR_CHECK_EN 0x284 +#define INTF_TEAR_SYNC_CONFIG_VSYNC 0x288 +#define INTF_TEAR_SYNC_CONFIG_HEIGHT 0x28C +#define INTF_TEAR_SYNC_WRCOUNT 0x290 +#define INTF_TEAR_VSYNC_INIT_VAL 0x294 +#define INTF_TEAR_INT_COUNT_VAL 0x298 +#define INTF_TEAR_SYNC_THRESH 0x29C +#define INTF_TEAR_START_POS 0x2A0 +#define INTF_TEAR_RD_PTR_IRQ 0x2A4 +#define INTF_TEAR_WR_PTR_IRQ 0x2A8 +#define INTF_TEAR_OUT_LINE_COUNT 0x2AC +#define INTF_TEAR_LINE_COUNT 0x2B0 +#define INTF_TEAR_AUTOREFRESH_CONFIG 0x2B4 #define INTF_CFG_ACTIVE_H_EN BIT(29) #define INTF_CFG_ACTIVE_V_EN BIT(30) #define INTF_CFG2_DATABUS_WIDEN BIT(0) #define INTF_CFG2_DATA_HCTL_EN BIT(4) +#define INTF_CFG2_DCE_DATA_COMPRESS BIT(12) -#define INTF_MISR_CTRL 0x180 -#define INTF_MISR_SIGNATURE 0x184 - -static const struct dpu_intf_cfg *_intf_offset(enum dpu_intf intf, - const struct dpu_mdss_cfg *m, - void __iomem *addr, - struct dpu_hw_blk_reg_map *b) -{ - int i; - - for (i = 0; i < m->intf_count; i++) { - if ((intf == m->intf[i].id) && - (m->intf[i].type != INTF_NONE)) { - b->blk_addr = addr + m->intf[i].base; - b->log_mask = DPU_DBG_MASK_INTF; - return &m->intf[i]; - } - } - - return ERR_PTR(-EINVAL); -} static void dpu_hw_intf_setup_timing_engine(struct dpu_hw_intf *ctx, const struct intf_timing_params *p, @@ -104,7 +106,7 @@ static void dpu_hw_intf_setup_timing_engine(struct dpu_hw_intf *ctx, u32 active_h_start, active_h_end; u32 active_v_start, active_v_end; u32 active_hctl, display_hctl, hsync_ctl; - u32 polarity_ctl, den_polarity, hsync_polarity, vsync_polarity; + u32 polarity_ctl, den_polarity; u32 panel_format; u32 intf_cfg, intf_cfg2 = 0; u32 display_data_hctl = 0, active_data_hctl = 0; @@ -191,19 +193,9 @@ static void dpu_hw_intf_setup_timing_engine(struct dpu_hw_intf *ctx, } den_polarity = 0; - if (ctx->cap->type == INTF_HDMI) { - hsync_polarity = p->yres >= 720 ? 0 : 1; - vsync_polarity = p->yres >= 720 ? 0 : 1; - } else if (ctx->cap->type == INTF_DP) { - hsync_polarity = p->hsync_polarity; - vsync_polarity = p->vsync_polarity; - } else { - hsync_polarity = 0; - vsync_polarity = 0; - } polarity_ctl = (den_polarity << 2) | /* DEN Polarity */ - (vsync_polarity << 1) | /* VSYNC Polarity */ - (hsync_polarity << 0); /* HSYNC Polarity */ + (p->vsync_polarity << 1) | /* VSYNC Polarity */ + (p->hsync_polarity << 0); /* HSYNC Polarity */ if (!DPU_FORMAT_IS_YUV(fmt)) panel_format = (fmt->bits[C0_G_Y] | @@ -276,7 +268,6 @@ static void dpu_hw_intf_setup_prg_fetch( static void dpu_hw_intf_bind_pingpong_blk( struct dpu_hw_intf *intf, - bool enable, const enum dpu_pingpong pp) { struct dpu_hw_blk_reg_map *c = &intf->hw; @@ -285,7 +276,7 @@ static void dpu_hw_intf_bind_pingpong_blk( mux_cfg = DPU_REG_READ(c, INTF_MUX); mux_cfg &= ~0xf; - if (enable) + if (pp) mux_cfg |= (pp - PINGPONG_0) & 0x7; else mux_cfg |= 0xf; @@ -337,6 +328,200 @@ static int dpu_hw_intf_collect_misr(struct dpu_hw_intf *intf, u32 *misr_value) return dpu_hw_collect_misr(&intf->hw, INTF_MISR_CTRL, INTF_MISR_SIGNATURE, misr_value); } +static int dpu_hw_intf_enable_te(struct dpu_hw_intf *intf, + struct dpu_hw_tear_check *te) +{ + struct dpu_hw_blk_reg_map *c; + int cfg; + + if (!intf) + return -EINVAL; + + c = &intf->hw; + + cfg = BIT(19); /* VSYNC_COUNTER_EN */ + if (te->hw_vsync_mode) + cfg |= BIT(20); + + cfg |= te->vsync_count; + + DPU_REG_WRITE(c, INTF_TEAR_SYNC_CONFIG_VSYNC, cfg); + DPU_REG_WRITE(c, INTF_TEAR_SYNC_CONFIG_HEIGHT, te->sync_cfg_height); + DPU_REG_WRITE(c, INTF_TEAR_VSYNC_INIT_VAL, te->vsync_init_val); + DPU_REG_WRITE(c, INTF_TEAR_RD_PTR_IRQ, te->rd_ptr_irq); + DPU_REG_WRITE(c, INTF_TEAR_START_POS, te->start_pos); + DPU_REG_WRITE(c, INTF_TEAR_SYNC_THRESH, + ((te->sync_threshold_continue << 16) | + te->sync_threshold_start)); + DPU_REG_WRITE(c, INTF_TEAR_SYNC_WRCOUNT, + (te->start_pos + te->sync_threshold_start + 1)); + + DPU_REG_WRITE(c, INTF_TEAR_TEAR_CHECK_EN, 1); + + return 0; +} + +static void dpu_hw_intf_setup_autorefresh_config(struct dpu_hw_intf *intf, + u32 frame_count, bool enable) +{ + struct dpu_hw_blk_reg_map *c; + u32 refresh_cfg; + + c = &intf->hw; + refresh_cfg = DPU_REG_READ(c, INTF_TEAR_AUTOREFRESH_CONFIG); + if (enable) + refresh_cfg = BIT(31) | frame_count; + else + refresh_cfg &= ~BIT(31); + + DPU_REG_WRITE(c, INTF_TEAR_AUTOREFRESH_CONFIG, refresh_cfg); +} + +/* + * dpu_hw_intf_get_autorefresh_config - Get autorefresh config from HW + * @intf: DPU intf structure + * @frame_count: Used to return the current frame count from hw + * + * Returns: True if autorefresh enabled, false if disabled. + */ +static bool dpu_hw_intf_get_autorefresh_config(struct dpu_hw_intf *intf, + u32 *frame_count) +{ + u32 val = DPU_REG_READ(&intf->hw, INTF_TEAR_AUTOREFRESH_CONFIG); + + if (frame_count != NULL) + *frame_count = val & 0xffff; + return !!((val & BIT(31)) >> 31); +} + +static int dpu_hw_intf_disable_te(struct dpu_hw_intf *intf) +{ + struct dpu_hw_blk_reg_map *c; + + if (!intf) + return -EINVAL; + + c = &intf->hw; + DPU_REG_WRITE(c, INTF_TEAR_TEAR_CHECK_EN, 0); + return 0; +} + +static int dpu_hw_intf_connect_external_te(struct dpu_hw_intf *intf, + bool enable_external_te) +{ + struct dpu_hw_blk_reg_map *c = &intf->hw; + u32 cfg; + int orig; + + if (!intf) + return -EINVAL; + + c = &intf->hw; + cfg = DPU_REG_READ(c, INTF_TEAR_SYNC_CONFIG_VSYNC); + orig = (bool)(cfg & BIT(20)); + if (enable_external_te) + cfg |= BIT(20); + else + cfg &= ~BIT(20); + DPU_REG_WRITE(c, INTF_TEAR_SYNC_CONFIG_VSYNC, cfg); + trace_dpu_intf_connect_ext_te(intf->idx - INTF_0, cfg); + + return orig; +} + +static int dpu_hw_intf_get_vsync_info(struct dpu_hw_intf *intf, + struct dpu_hw_pp_vsync_info *info) +{ + struct dpu_hw_blk_reg_map *c = &intf->hw; + u32 val; + + if (!intf || !info) + return -EINVAL; + + c = &intf->hw; + + val = DPU_REG_READ(c, INTF_TEAR_VSYNC_INIT_VAL); + info->rd_ptr_init_val = val & 0xffff; + + val = DPU_REG_READ(c, INTF_TEAR_INT_COUNT_VAL); + info->rd_ptr_frame_count = (val & 0xffff0000) >> 16; + info->rd_ptr_line_count = val & 0xffff; + + val = DPU_REG_READ(c, INTF_TEAR_LINE_COUNT); + info->wr_ptr_line_count = val & 0xffff; + + val = DPU_REG_READ(c, INTF_FRAME_COUNT); + info->intf_frame_count = val; + + return 0; +} + +static void dpu_hw_intf_vsync_sel(struct dpu_hw_intf *intf, + u32 vsync_source) +{ + struct dpu_hw_blk_reg_map *c; + + if (!intf) + return; + + c = &intf->hw; + + DPU_REG_WRITE(c, INTF_TEAR_MDP_VSYNC_SEL, (vsync_source & 0xf)); +} + +static void dpu_hw_intf_disable_autorefresh(struct dpu_hw_intf *intf, + uint32_t encoder_id, u16 vdisplay) +{ + struct dpu_hw_pp_vsync_info info; + int trial = 0; + + /* If autorefresh is already disabled, we have nothing to do */ + if (!dpu_hw_intf_get_autorefresh_config(intf, NULL)) + return; + + /* + * If autorefresh is enabled, disable it and make sure it is safe to + * proceed with current frame commit/push. Sequence followed is, + * 1. Disable TE + * 2. Disable autorefresh config + * 4. Poll for frame transfer ongoing to be false + * 5. Enable TE back + */ + + dpu_hw_intf_connect_external_te(intf, false); + dpu_hw_intf_setup_autorefresh_config(intf, 0, false); + + do { + udelay(DPU_ENC_MAX_POLL_TIMEOUT_US); + if ((trial * DPU_ENC_MAX_POLL_TIMEOUT_US) + > (KICKOFF_TIMEOUT_MS * USEC_PER_MSEC)) { + DPU_ERROR("enc%d intf%d disable autorefresh failed\n", + encoder_id, intf->idx - INTF_0); + break; + } + + trial++; + + dpu_hw_intf_get_vsync_info(intf, &info); + } while (info.wr_ptr_line_count > 0 && + info.wr_ptr_line_count < vdisplay); + + dpu_hw_intf_connect_external_te(intf, true); + + DPU_DEBUG("enc%d intf%d disabled autorefresh\n", + encoder_id, intf->idx - INTF_0); + +} + +static void dpu_hw_intf_enable_compression(struct dpu_hw_intf *ctx) +{ + u32 intf_cfg2 = DPU_REG_READ(&ctx->hw, INTF_CONFIG2); + + intf_cfg2 |= INTF_CFG2_DCE_DATA_COMPRESS; + + DPU_REG_WRITE(&ctx->hw, INTF_CONFIG2, intf_cfg2); +} + static void _setup_intf_ops(struct dpu_hw_intf_ops *ops, unsigned long cap) { @@ -349,32 +534,41 @@ static void _setup_intf_ops(struct dpu_hw_intf_ops *ops, ops->bind_pingpong_blk = dpu_hw_intf_bind_pingpong_blk; ops->setup_misr = dpu_hw_intf_setup_misr; ops->collect_misr = dpu_hw_intf_collect_misr; + + if (cap & BIT(DPU_INTF_TE)) { + ops->enable_tearcheck = dpu_hw_intf_enable_te; + ops->disable_tearcheck = dpu_hw_intf_disable_te; + ops->connect_external_te = dpu_hw_intf_connect_external_te; + ops->vsync_sel = dpu_hw_intf_vsync_sel; + ops->disable_autorefresh = dpu_hw_intf_disable_autorefresh; + } + + if (cap & BIT(DPU_INTF_DATA_COMPRESS)) + ops->enable_compression = dpu_hw_intf_enable_compression; } -struct dpu_hw_intf *dpu_hw_intf_init(enum dpu_intf idx, - void __iomem *addr, - const struct dpu_mdss_cfg *m) +struct dpu_hw_intf *dpu_hw_intf_init(const struct dpu_intf_cfg *cfg, + void __iomem *addr) { struct dpu_hw_intf *c; - const struct dpu_intf_cfg *cfg; + + if (cfg->type == INTF_NONE) { + DPU_DEBUG("Skip intf %d with type NONE\n", cfg->id - INTF_0); + return NULL; + } c = kzalloc(sizeof(*c), GFP_KERNEL); if (!c) return ERR_PTR(-ENOMEM); - cfg = _intf_offset(idx, m, addr, &c->hw); - if (IS_ERR_OR_NULL(cfg)) { - kfree(c); - pr_err("failed to create dpu_hw_intf %d\n", idx); - return ERR_PTR(-EINVAL); - } + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_INTF; /* * Assign ops */ - c->idx = idx; + c->idx = cfg->id; c->cap = cfg; - c->mdss = m; _setup_intf_ops(&c->ops, c->cap->features); return c; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h index 643dd10bc0306a02277d0dc01316fa0c2f44b8a8..99e21c4137f91b6cad785fe5069cf9da7c88f712 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h @@ -60,6 +60,17 @@ struct intf_status { * feed pixels to this interface * @setup_misr: enable/disable MISR * @collect_misr: read MISR signature + * @enable_tearcheck: Enables vsync generation and sets up init value of read + * pointer and programs the tear check configuration + * @disable_tearcheck: Disables tearcheck block + * @connect_external_te: Read, modify, write to either set or clear listening to external TE + * Return: 1 if TE was originally connected, 0 if not, or -ERROR + * @get_vsync_info: Provides the programmed and current line_count + * @setup_autorefresh: Configure and enable the autorefresh config + * @get_autorefresh: Retrieve autorefresh config from hardware + * Return: 0 on success, -ETIMEDOUT on timeout + * @vsync_sel: Select vsync signal for tear-effect configuration + * @enable_compression: Enable data compression */ struct dpu_hw_intf_ops { void (*setup_timing_gen)(struct dpu_hw_intf *intf, @@ -78,10 +89,26 @@ struct dpu_hw_intf_ops { u32 (*get_line_count)(struct dpu_hw_intf *intf); void (*bind_pingpong_blk)(struct dpu_hw_intf *intf, - bool enable, const enum dpu_pingpong pp); void (*setup_misr)(struct dpu_hw_intf *intf, bool enable, u32 frame_count); int (*collect_misr)(struct dpu_hw_intf *intf, u32 *misr_value); + + // Tearcheck on INTF since DPU 5.0.0 + + int (*enable_tearcheck)(struct dpu_hw_intf *intf, struct dpu_hw_tear_check *cfg); + + int (*disable_tearcheck)(struct dpu_hw_intf *intf); + + int (*connect_external_te)(struct dpu_hw_intf *intf, bool enable_external_te); + + void (*vsync_sel)(struct dpu_hw_intf *intf, u32 vsync_source); + + /** + * Disable autorefresh if enabled + */ + void (*disable_autorefresh)(struct dpu_hw_intf *intf, uint32_t encoder_id, u16 vdisplay); + + void (*enable_compression)(struct dpu_hw_intf *intf); }; struct dpu_hw_intf { @@ -90,22 +117,19 @@ struct dpu_hw_intf { /* intf */ enum dpu_intf idx; const struct dpu_intf_cfg *cap; - const struct dpu_mdss_cfg *mdss; /* ops */ struct dpu_hw_intf_ops ops; }; /** - * dpu_hw_intf_init(): Initializes the intf driver for the passed - * interface idx. - * @idx: interface index for which driver object is required + * dpu_hw_intf_init() - Initializes the INTF driver for the passed + * interface catalog entry. + * @cfg: interface catalog entry for which driver object is required * @addr: mapped register io address of MDP - * @m : pointer to mdss catalog data */ -struct dpu_hw_intf *dpu_hw_intf_init(enum dpu_intf idx, - void __iomem *addr, - const struct dpu_mdss_cfg *m); +struct dpu_hw_intf *dpu_hw_intf_init(const struct dpu_intf_cfg *cfg, + void __iomem *addr); /** * dpu_hw_intf_destroy(): Destroys INTF driver context diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c index f5120ea91edee8d7f269e7bc2e4df386c7c82179..d1c3bd8379ea947280d84516c607d7d70786c40b 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c @@ -30,24 +30,6 @@ #define LM_MISR_SIGNATURE 0x314 -static const struct dpu_lm_cfg *_lm_offset(enum dpu_lm mixer, - const struct dpu_mdss_cfg *m, - void __iomem *addr, - struct dpu_hw_blk_reg_map *b) -{ - int i; - - for (i = 0; i < m->mixer_count; i++) { - if (mixer == m->mixer[i].id) { - b->blk_addr = addr + m->mixer[i].base; - b->log_mask = DPU_DBG_MASK_LM; - return &m->mixer[i]; - } - } - - return ERR_PTR(-ENOMEM); -} - /** * _stage_offset(): returns the relative offset of the blend registers * for the stage to be setup @@ -160,8 +142,7 @@ static void dpu_hw_lm_setup_color3(struct dpu_hw_mixer *ctx, DPU_REG_WRITE(c, LM_OP_MODE, op_mode); } -static void _setup_mixer_ops(const struct dpu_mdss_cfg *m, - struct dpu_hw_lm_ops *ops, +static void _setup_mixer_ops(struct dpu_hw_lm_ops *ops, unsigned long features) { ops->setup_mixer_out = dpu_hw_lm_setup_out; @@ -175,27 +156,27 @@ static void _setup_mixer_ops(const struct dpu_mdss_cfg *m, ops->collect_misr = dpu_hw_lm_collect_misr; } -struct dpu_hw_mixer *dpu_hw_lm_init(enum dpu_lm idx, - void __iomem *addr, - const struct dpu_mdss_cfg *m) +struct dpu_hw_mixer *dpu_hw_lm_init(const struct dpu_lm_cfg *cfg, + void __iomem *addr) { struct dpu_hw_mixer *c; - const struct dpu_lm_cfg *cfg; + + if (cfg->pingpong == PINGPONG_NONE) { + DPU_DEBUG("skip mixer %d without pingpong\n", cfg->id); + return NULL; + } c = kzalloc(sizeof(*c), GFP_KERNEL); if (!c) return ERR_PTR(-ENOMEM); - cfg = _lm_offset(idx, m, addr, &c->hw); - if (IS_ERR_OR_NULL(cfg)) { - kfree(c); - return ERR_PTR(-EINVAL); - } + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_LM; /* Assign ops */ - c->idx = idx; + c->idx = cfg->id; c->cap = cfg; - _setup_mixer_ops(m, &c->ops, c->cap->features); + _setup_mixer_ops(&c->ops, c->cap->features); return c; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h index 652ddfdedec375b04f09fafd4e62a93360e70763..36992d046a533b979ab674808fe044a2a2a0ce85 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h @@ -93,15 +93,13 @@ static inline struct dpu_hw_mixer *to_dpu_hw_mixer(struct dpu_hw_blk *hw) } /** - * dpu_hw_lm_init(): Initializes the mixer hw driver object. + * dpu_hw_lm_init() - Initializes the mixer hw driver object. * should be called once before accessing every mixer. - * @idx: mixer index for which driver object is required + * @cfg: mixer catalog entry for which driver object is required * @addr: mapped register io address of MDP - * @m : pointer to mdss catalog data */ -struct dpu_hw_mixer *dpu_hw_lm_init(enum dpu_lm idx, - void __iomem *addr, - const struct dpu_mdss_cfg *m); +struct dpu_hw_mixer *dpu_hw_lm_init(const struct dpu_lm_cfg *cfg, + void __iomem *addr); /** * dpu_hw_lm_destroy(): Destroys layer mixer driver context diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h index 2d9192a6ce006a1a9e37095aa593ae6053cf35b5..02a0f48aac94a5492f344bd3c64eb723297ec49c 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h @@ -191,7 +191,8 @@ enum dpu_dsc { }; enum dpu_pingpong { - PINGPONG_0 = 1, + PINGPONG_NONE, + PINGPONG_0, PINGPONG_1, PINGPONG_2, PINGPONG_3, @@ -463,4 +464,52 @@ struct dpu_mdss_color { #define DPU_DBG_MASK_DSPP (1 << 10) #define DPU_DBG_MASK_DSC (1 << 11) +/** + * struct dpu_hw_tear_check - Struct contains parameters to configure + * tear-effect module. This structure is used to configure tear-check + * logic present either in ping-pong or in interface module. + * @vsync_count: Ratio of MDP VSYNC clk freq(Hz) to refresh rate divided + * by no of lines + * @sync_cfg_height: Total vertical lines (display height - 1) + * @vsync_init_val: Init value to which the read pointer gets loaded at + * vsync edge + * @sync_threshold_start: Read pointer threshold start ROI for write operation + * @sync_threshold_continue: The minimum number of lines the write pointer + * needs to be above the read pointer + * @start_pos: The position from which the start_threshold value is added + * @rd_ptr_irq: The read pointer line at which interrupt has to be generated + * @hw_vsync_mode: Sync with external frame sync input + */ +struct dpu_hw_tear_check { + /* + * This is ratio of MDP VSYNC clk freq(Hz) to + * refresh rate divided by no of lines + */ + u32 vsync_count; + u32 sync_cfg_height; + u32 vsync_init_val; + u32 sync_threshold_start; + u32 sync_threshold_continue; + u32 start_pos; + u32 rd_ptr_irq; + u8 hw_vsync_mode; +}; + +/** + * struct dpu_hw_pp_vsync_info - Struct contains parameters to configure + * read and write pointers for command mode panels + * @rd_ptr_init_val: Value of rd pointer at vsync edge + * @rd_ptr_frame_count: Num frames sent since enabling interface + * @rd_ptr_line_count: Current line on panel (rd ptr) + * @wr_ptr_line_count: Current line within pp fifo (wr ptr) + * @intf_frame_count: Frames read from intf + */ +struct dpu_hw_pp_vsync_info { + u32 rd_ptr_init_val; + u32 rd_ptr_frame_count; + u32 rd_ptr_line_count; + u32 wr_ptr_line_count; + u32 intf_frame_count; +}; + #endif /* _DPU_HW_MDSS_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c index def0a87fdba56711b6544d9ac3a64e362b106d0e..90e0e05eff8d312e85e8bb10e5fa78c15f26c639 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c @@ -14,24 +14,6 @@ #define MERGE_3D_MUX 0x000 #define MERGE_3D_MODE 0x004 -static const struct dpu_merge_3d_cfg *_merge_3d_offset(enum dpu_merge_3d idx, - const struct dpu_mdss_cfg *m, - void __iomem *addr, - struct dpu_hw_blk_reg_map *b) -{ - int i; - - for (i = 0; i < m->merge_3d_count; i++) { - if (idx == m->merge_3d[i].id) { - b->blk_addr = addr + m->merge_3d[i].base; - b->log_mask = DPU_DBG_MASK_PINGPONG; - return &m->merge_3d[i]; - } - } - - return ERR_PTR(-EINVAL); -} - static void dpu_hw_merge_3d_setup_3d_mode(struct dpu_hw_merge_3d *merge_3d, enum dpu_3d_blend_mode mode_3d) { @@ -55,24 +37,19 @@ static void _setup_merge_3d_ops(struct dpu_hw_merge_3d *c, c->ops.setup_3d_mode = dpu_hw_merge_3d_setup_3d_mode; }; -struct dpu_hw_merge_3d *dpu_hw_merge_3d_init(enum dpu_merge_3d idx, - void __iomem *addr, - const struct dpu_mdss_cfg *m) +struct dpu_hw_merge_3d *dpu_hw_merge_3d_init(const struct dpu_merge_3d_cfg *cfg, + void __iomem *addr) { struct dpu_hw_merge_3d *c; - const struct dpu_merge_3d_cfg *cfg; c = kzalloc(sizeof(*c), GFP_KERNEL); if (!c) return ERR_PTR(-ENOMEM); - cfg = _merge_3d_offset(idx, m, addr, &c->hw); - if (IS_ERR_OR_NULL(cfg)) { - kfree(c); - return ERR_PTR(-EINVAL); - } + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_PINGPONG; - c->idx = idx; + c->idx = cfg->id; c->caps = cfg; _setup_merge_3d_ops(c, c->caps->features); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.h index 81fd1d5f718e7862391ef8d71102a849edbb36be..19cec5e8872219ebbc2a1649c27d26c369715247 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.h @@ -46,16 +46,14 @@ static inline struct dpu_hw_merge_3d *to_dpu_hw_merge_3d(struct dpu_hw_blk *hw) } /** - * dpu_hw_merge_3d_init - initializes the merge_3d driver for the passed - * merge_3d idx. - * @idx: Pingpong index for which driver object is required + * dpu_hw_merge_3d_init() - Initializes the merge_3d driver for the passed + * merge3d catalog entry. + * @cfg: Pingpong catalog entry for which driver object is required * @addr: Mapped register io address of MDP - * @m: Pointer to mdss catalog data - * Returns: Error code or allocated dpu_hw_merge_3d context + * Return: Error code or allocated dpu_hw_merge_3d context */ -struct dpu_hw_merge_3d *dpu_hw_merge_3d_init(enum dpu_merge_3d idx, - void __iomem *addr, - const struct dpu_mdss_cfg *m); +struct dpu_hw_merge_3d *dpu_hw_merge_3d_init(const struct dpu_merge_3d_cfg *cfg, + void __iomem *addr); /** * dpu_hw_merge_3d_destroy - destroys merge_3d driver context diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c index 0fcad9760b6fcde2770dbe2100c5cee0d4da6865..437d9e62a841b2421dfb0bdb78a49be8b506a798 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c @@ -42,24 +42,6 @@ static u32 dither_depth_map[DITHER_DEPTH_MAP_INDEX] = { 0, 0, 0, 0, 0, 0, 0, 1, 2 }; -static const struct dpu_pingpong_cfg *_pingpong_offset(enum dpu_pingpong pp, - const struct dpu_mdss_cfg *m, - void __iomem *addr, - struct dpu_hw_blk_reg_map *b) -{ - int i; - - for (i = 0; i < m->pingpong_count; i++) { - if (pp == m->pingpong[i].id) { - b->blk_addr = addr + m->pingpong[i].base; - b->log_mask = DPU_DBG_MASK_PINGPONG; - return &m->pingpong[i]; - } - } - - return ERR_PTR(-EINVAL); -} - static void dpu_hw_pp_setup_dither(struct dpu_hw_pingpong *pp, struct dpu_hw_dither_cfg *cfg) { @@ -91,7 +73,7 @@ static void dpu_hw_pp_setup_dither(struct dpu_hw_pingpong *pp, DPU_REG_WRITE(c, base + PP_DITHER_EN, 1); } -static int dpu_hw_pp_setup_te_config(struct dpu_hw_pingpong *pp, +static int dpu_hw_pp_enable_te(struct dpu_hw_pingpong *pp, struct dpu_hw_tear_check *te) { struct dpu_hw_blk_reg_map *c; @@ -118,6 +100,8 @@ static int dpu_hw_pp_setup_te_config(struct dpu_hw_pingpong *pp, DPU_REG_WRITE(c, PP_SYNC_WRCOUNT, (te->start_pos + te->sync_threshold_start + 1)); + DPU_REG_WRITE(c, PP_TEAR_CHECK_EN, 1); + return 0; } @@ -144,24 +128,7 @@ static bool dpu_hw_pp_get_autorefresh_config(struct dpu_hw_pingpong *pp, return !!((val & BIT(31)) >> 31); } -static int dpu_hw_pp_poll_timeout_wr_ptr(struct dpu_hw_pingpong *pp, - u32 timeout_us) -{ - struct dpu_hw_blk_reg_map *c; - u32 val; - int rc; - - if (!pp) - return -EINVAL; - - c = &pp->hw; - rc = readl_poll_timeout(c->blk_addr + PP_LINE_COUNT, - val, (val & 0xffff) >= 1, 10, timeout_us); - - return rc; -} - -static int dpu_hw_pp_enable_te(struct dpu_hw_pingpong *pp, bool enable) +static int dpu_hw_pp_disable_te(struct dpu_hw_pingpong *pp) { struct dpu_hw_blk_reg_map *c; @@ -169,7 +136,7 @@ static int dpu_hw_pp_enable_te(struct dpu_hw_pingpong *pp, bool enable) return -EINVAL; c = &pp->hw; - DPU_REG_WRITE(c, PP_TEAR_CHECK_EN, enable); + DPU_REG_WRITE(c, PP_TEAR_CHECK_EN, 0); return 0; } @@ -245,6 +212,49 @@ static u32 dpu_hw_pp_get_line_count(struct dpu_hw_pingpong *pp) return line; } +static void dpu_hw_pp_disable_autorefresh(struct dpu_hw_pingpong *pp, + uint32_t encoder_id, u16 vdisplay) +{ + struct dpu_hw_pp_vsync_info info; + int trial = 0; + + /* If autorefresh is already disabled, we have nothing to do */ + if (!dpu_hw_pp_get_autorefresh_config(pp, NULL)) + return; + + /* + * If autorefresh is enabled, disable it and make sure it is safe to + * proceed with current frame commit/push. Sequence followed is, + * 1. Disable TE + * 2. Disable autorefresh config + * 4. Poll for frame transfer ongoing to be false + * 5. Enable TE back + */ + + dpu_hw_pp_connect_external_te(pp, false); + dpu_hw_pp_setup_autorefresh_config(pp, 0, false); + + do { + udelay(DPU_ENC_MAX_POLL_TIMEOUT_US); + if ((trial * DPU_ENC_MAX_POLL_TIMEOUT_US) + > (KICKOFF_TIMEOUT_MS * USEC_PER_MSEC)) { + DPU_ERROR("enc%d pp%d disable autorefresh failed\n", + encoder_id, pp->idx - PINGPONG_0); + break; + } + + trial++; + + dpu_hw_pp_get_vsync_info(pp, &info); + } while (info.wr_ptr_line_count > 0 && + info.wr_ptr_line_count < vdisplay); + + dpu_hw_pp_connect_external_te(pp, true); + + DPU_DEBUG("enc%d pp%d disabled autorefresh\n", + encoder_id, pp->idx - PINGPONG_0); +} + static int dpu_hw_pp_dsc_enable(struct dpu_hw_pingpong *pp) { struct dpu_hw_blk_reg_map *c = &pp->hw; @@ -274,40 +284,37 @@ static int dpu_hw_pp_setup_dsc(struct dpu_hw_pingpong *pp) static void _setup_pingpong_ops(struct dpu_hw_pingpong *c, unsigned long features) { - c->ops.setup_tearcheck = dpu_hw_pp_setup_te_config; - c->ops.enable_tearcheck = dpu_hw_pp_enable_te; - c->ops.connect_external_te = dpu_hw_pp_connect_external_te; - c->ops.get_vsync_info = dpu_hw_pp_get_vsync_info; - c->ops.setup_autorefresh = dpu_hw_pp_setup_autorefresh_config; - c->ops.get_autorefresh = dpu_hw_pp_get_autorefresh_config; - c->ops.poll_timeout_wr_ptr = dpu_hw_pp_poll_timeout_wr_ptr; - c->ops.get_line_count = dpu_hw_pp_get_line_count; - c->ops.setup_dsc = dpu_hw_pp_setup_dsc; - c->ops.enable_dsc = dpu_hw_pp_dsc_enable; - c->ops.disable_dsc = dpu_hw_pp_dsc_disable; + if (test_bit(DPU_PINGPONG_TE, &features)) { + c->ops.enable_tearcheck = dpu_hw_pp_enable_te; + c->ops.disable_tearcheck = dpu_hw_pp_disable_te; + c->ops.connect_external_te = dpu_hw_pp_connect_external_te; + c->ops.get_line_count = dpu_hw_pp_get_line_count; + c->ops.disable_autorefresh = dpu_hw_pp_disable_autorefresh; + } + + if (test_bit(DPU_PINGPONG_DSC, &features)) { + c->ops.setup_dsc = dpu_hw_pp_setup_dsc; + c->ops.enable_dsc = dpu_hw_pp_dsc_enable; + c->ops.disable_dsc = dpu_hw_pp_dsc_disable; + } if (test_bit(DPU_PINGPONG_DITHER, &features)) c->ops.setup_dither = dpu_hw_pp_setup_dither; }; -struct dpu_hw_pingpong *dpu_hw_pingpong_init(enum dpu_pingpong idx, - void __iomem *addr, - const struct dpu_mdss_cfg *m) +struct dpu_hw_pingpong *dpu_hw_pingpong_init(const struct dpu_pingpong_cfg *cfg, + void __iomem *addr) { struct dpu_hw_pingpong *c; - const struct dpu_pingpong_cfg *cfg; c = kzalloc(sizeof(*c), GFP_KERNEL); if (!c) return ERR_PTR(-ENOMEM); - cfg = _pingpong_offset(idx, m, addr, &c->hw); - if (IS_ERR_OR_NULL(cfg)) { - kfree(c); - return ERR_PTR(-EINVAL); - } + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_PINGPONG; - c->idx = idx; + c->idx = cfg->id; c->caps = cfg; _setup_pingpong_ops(c, c->caps->features); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h index c00223441d99020ae85bd30c2174a5c7904a22e0..d3246a9a5808fa62e49763be95abfc2d3059a71e 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h @@ -13,28 +13,6 @@ struct dpu_hw_pingpong; -struct dpu_hw_tear_check { - /* - * This is ratio of MDP VSYNC clk freq(Hz) to - * refresh rate divided by no of lines - */ - u32 vsync_count; - u32 sync_cfg_height; - u32 vsync_init_val; - u32 sync_threshold_start; - u32 sync_threshold_continue; - u32 start_pos; - u32 rd_ptr_irq; - u8 hw_vsync_mode; -}; - -struct dpu_hw_pp_vsync_info { - u32 rd_ptr_init_val; /* value of rd pointer at vsync edge */ - u32 rd_ptr_frame_count; /* num frames sent since enabling interface */ - u32 rd_ptr_line_count; /* current line on panel (rd ptr) */ - u32 wr_ptr_line_count; /* current line within pp fifo (wr ptr) */ -}; - /** * struct dpu_hw_dither_cfg - dither feature structure * @flags: for customizing operations @@ -59,11 +37,8 @@ struct dpu_hw_dither_cfg { * * struct dpu_hw_pingpong_ops : Interface to the pingpong Hw driver functions * Assumption is these functions will be called after clocks are enabled - * @setup_tearcheck : program tear check values - * @enable_tearcheck : enables tear check - * @get_vsync_info : retries timing info of the panel - * @setup_autorefresh : configure and enable the autorefresh config - * @get_autorefresh : retrieve autorefresh config from hardware + * @enable_tearcheck: program and enable tear check block + * @disable_tearcheck: disable able tear check block * @setup_dither : function to program the dither hw block * @get_line_count: obtain current vertical line counter */ @@ -72,14 +47,13 @@ struct dpu_hw_pingpong_ops { * enables vysnc generation and sets up init value of * read pointer and programs the tear check cofiguration */ - int (*setup_tearcheck)(struct dpu_hw_pingpong *pp, + int (*enable_tearcheck)(struct dpu_hw_pingpong *pp, struct dpu_hw_tear_check *cfg); /** - * enables tear check block + * disables tear check block */ - int (*enable_tearcheck)(struct dpu_hw_pingpong *pp, - bool enable); + int (*disable_tearcheck)(struct dpu_hw_pingpong *pp); /** * read, modify, write to either set or clear listening to external TE @@ -89,34 +63,14 @@ struct dpu_hw_pingpong_ops { bool enable_external_te); /** - * provides the programmed and current - * line_count - */ - int (*get_vsync_info)(struct dpu_hw_pingpong *pp, - struct dpu_hw_pp_vsync_info *info); - - /** - * configure and enable the autorefresh config - */ - void (*setup_autorefresh)(struct dpu_hw_pingpong *pp, - u32 frame_count, bool enable); - - /** - * retrieve autorefresh config from hardware - */ - bool (*get_autorefresh)(struct dpu_hw_pingpong *pp, - u32 *frame_count); - - /** - * poll until write pointer transmission starts - * @Return: 0 on success, -ETIMEDOUT on timeout + * Obtain current vertical line counter */ - int (*poll_timeout_wr_ptr)(struct dpu_hw_pingpong *pp, u32 timeout_us); + u32 (*get_line_count)(struct dpu_hw_pingpong *pp); /** - * Obtain current vertical line counter + * Disable autorefresh if enabled */ - u32 (*get_line_count)(struct dpu_hw_pingpong *pp); + void (*disable_autorefresh)(struct dpu_hw_pingpong *pp, uint32_t encoder_id, u16 vdisplay); /** * Setup dither matix for pingpong block @@ -165,16 +119,14 @@ static inline struct dpu_hw_pingpong *to_dpu_hw_pingpong(struct dpu_hw_blk *hw) } /** - * dpu_hw_pingpong_init - initializes the pingpong driver for the passed - * pingpong idx. - * @idx: Pingpong index for which driver object is required + * dpu_hw_pingpong_init() - initializes the pingpong driver for the passed + * pingpong catalog entry. + * @cfg: Pingpong catalog entry for which driver object is required * @addr: Mapped register io address of MDP - * @m: Pointer to mdss catalog data - * Returns: Error code or allocated dpu_hw_pingpong context + * Return: Error code or allocated dpu_hw_pingpong context */ -struct dpu_hw_pingpong *dpu_hw_pingpong_init(enum dpu_pingpong idx, - void __iomem *addr, - const struct dpu_mdss_cfg *m); +struct dpu_hw_pingpong *dpu_hw_pingpong_init(const struct dpu_pingpong_cfg *cfg, + void __iomem *addr); /** * dpu_hw_pingpong_destroy - destroys pingpong driver context diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c index cf70a9bd1034f08654e9e27afcce752136db0ca7..b364cf75bb3f1059b867a341dcde8dd80f08eed2 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c @@ -12,7 +12,7 @@ #define DPU_FETCH_CONFIG_RESET_VALUE 0x00000087 -/* DPU_SSPP_SRC */ +/* SSPP registers */ #define SSPP_SRC_SIZE 0x00 #define SSPP_SRC_XY 0x08 #define SSPP_OUT_SIZE 0x0c @@ -26,45 +26,18 @@ #define SSPP_SRC_FORMAT 0x30 #define SSPP_SRC_UNPACK_PATTERN 0x34 #define SSPP_SRC_OP_MODE 0x38 - -/* SSPP_MULTIRECT*/ -#define SSPP_SRC_SIZE_REC1 0x16C -#define SSPP_SRC_XY_REC1 0x168 -#define SSPP_OUT_SIZE_REC1 0x160 -#define SSPP_OUT_XY_REC1 0x164 -#define SSPP_SRC_FORMAT_REC1 0x174 -#define SSPP_SRC_UNPACK_PATTERN_REC1 0x178 -#define SSPP_SRC_OP_MODE_REC1 0x17C -#define SSPP_MULTIRECT_OPMODE 0x170 -#define SSPP_SRC_CONSTANT_COLOR_REC1 0x180 -#define SSPP_EXCL_REC_SIZE_REC1 0x184 -#define SSPP_EXCL_REC_XY_REC1 0x188 - -#define MDSS_MDP_OP_DEINTERLACE BIT(22) -#define MDSS_MDP_OP_DEINTERLACE_ODD BIT(23) -#define MDSS_MDP_OP_IGC_ROM_1 BIT(18) -#define MDSS_MDP_OP_IGC_ROM_0 BIT(17) -#define MDSS_MDP_OP_IGC_EN BIT(16) -#define MDSS_MDP_OP_FLIP_UD BIT(14) -#define MDSS_MDP_OP_FLIP_LR BIT(13) -#define MDSS_MDP_OP_BWC_EN BIT(0) -#define MDSS_MDP_OP_PE_OVERRIDE BIT(31) -#define MDSS_MDP_OP_BWC_LOSSLESS (0 << 1) -#define MDSS_MDP_OP_BWC_Q_HIGH (1 << 1) -#define MDSS_MDP_OP_BWC_Q_MED (2 << 1) - #define SSPP_SRC_CONSTANT_COLOR 0x3c #define SSPP_EXCL_REC_CTL 0x40 #define SSPP_UBWC_STATIC_CTRL 0x44 -#define SSPP_FETCH_CONFIG 0x048 +#define SSPP_FETCH_CONFIG 0x48 #define SSPP_DANGER_LUT 0x60 #define SSPP_SAFE_LUT 0x64 #define SSPP_CREQ_LUT 0x68 #define SSPP_QOS_CTRL 0x6C -#define SSPP_DECIMATION_CONFIG 0xB4 #define SSPP_SRC_ADDR_SW_STATUS 0x70 #define SSPP_CREQ_LUT_0 0x74 #define SSPP_CREQ_LUT_1 0x78 +#define SSPP_DECIMATION_CONFIG 0xB4 #define SSPP_SW_PIX_EXT_C0_LR 0x100 #define SSPP_SW_PIX_EXT_C0_TB 0x104 #define SSPP_SW_PIX_EXT_C0_REQ_PIXELS 0x108 @@ -81,11 +54,33 @@ #define SSPP_TRAFFIC_SHAPER_PREFILL 0x150 #define SSPP_TRAFFIC_SHAPER_REC1_PREFILL 0x154 #define SSPP_TRAFFIC_SHAPER_REC1 0x158 +#define SSPP_OUT_SIZE_REC1 0x160 +#define SSPP_OUT_XY_REC1 0x164 +#define SSPP_SRC_XY_REC1 0x168 +#define SSPP_SRC_SIZE_REC1 0x16C +#define SSPP_MULTIRECT_OPMODE 0x170 +#define SSPP_SRC_FORMAT_REC1 0x174 +#define SSPP_SRC_UNPACK_PATTERN_REC1 0x178 +#define SSPP_SRC_OP_MODE_REC1 0x17C +#define SSPP_SRC_CONSTANT_COLOR_REC1 0x180 +#define SSPP_EXCL_REC_SIZE_REC1 0x184 +#define SSPP_EXCL_REC_XY_REC1 0x188 #define SSPP_EXCL_REC_SIZE 0x1B4 #define SSPP_EXCL_REC_XY 0x1B8 -#define SSPP_VIG_OP_MODE 0x0 -#define SSPP_VIG_CSC_10_OP_MODE 0x0 -#define SSPP_TRAFFIC_SHAPER_BPC_MAX 0xFF + +/* SSPP_SRC_OP_MODE & OP_MODE_REC1 */ +#define MDSS_MDP_OP_DEINTERLACE BIT(22) +#define MDSS_MDP_OP_DEINTERLACE_ODD BIT(23) +#define MDSS_MDP_OP_IGC_ROM_1 BIT(18) +#define MDSS_MDP_OP_IGC_ROM_0 BIT(17) +#define MDSS_MDP_OP_IGC_EN BIT(16) +#define MDSS_MDP_OP_FLIP_UD BIT(14) +#define MDSS_MDP_OP_FLIP_LR BIT(13) +#define MDSS_MDP_OP_BWC_EN BIT(0) +#define MDSS_MDP_OP_PE_OVERRIDE BIT(31) +#define MDSS_MDP_OP_BWC_LOSSLESS (0 << 1) +#define MDSS_MDP_OP_BWC_Q_HIGH (1 << 1) +#define MDSS_MDP_OP_BWC_Q_MED (2 << 1) /* SSPP_QOS_CTRL */ #define SSPP_QOS_CTRL_VBLANK_EN BIT(16) @@ -96,6 +91,7 @@ #define SSPP_QOS_CTRL_CREQ_VBLANK_OFF 20 /* DPU_SSPP_SCALER_QSEED2 */ +#define SSPP_VIG_OP_MODE 0x0 #define SCALE_CONFIG 0x04 #define COMP0_3_PHASE_STEP_X 0x10 #define COMP0_3_PHASE_STEP_Y 0x14 @@ -107,6 +103,9 @@ #define COMP1_2_INIT_PHASE_Y 0x2C #define VIG_0_QSEED2_SHARP 0x30 +/* SSPP_TRAFFIC_SHAPER and _REC1 */ +#define SSPP_TRAFFIC_SHAPER_BPC_MAX 0xFF + /* * Definitions for ViG op modes */ @@ -128,6 +127,7 @@ /* * Definitions for CSC 10 op modes */ +#define SSPP_VIG_CSC_10_OP_MODE 0x0 #define VIG_CSC_10_SRC_DATAFMT BIT(1) #define VIG_CSC_10_EN BIT(0) #define CSC_10BIT_OFFSET 4 @@ -136,45 +136,12 @@ #define TS_CLK 19200000 -static int _sspp_subblk_offset(struct dpu_hw_sspp *ctx, - int s_id, - u32 *idx) -{ - int rc = 0; - const struct dpu_sspp_sub_blks *sblk; - - if (!ctx || !ctx->cap || !ctx->cap->sblk) - return -EINVAL; - - sblk = ctx->cap->sblk; - - switch (s_id) { - case DPU_SSPP_SRC: - *idx = sblk->src_blk.base; - break; - case DPU_SSPP_SCALER_QSEED2: - case DPU_SSPP_SCALER_QSEED3: - case DPU_SSPP_SCALER_RGB: - *idx = sblk->scaler_blk.base; - break; - case DPU_SSPP_CSC: - case DPU_SSPP_CSC_10BIT: - *idx = sblk->csc_blk.base; - break; - default: - rc = -EINVAL; - } - - return rc; -} - static void dpu_hw_sspp_setup_multirect(struct dpu_sw_pipe *pipe) { struct dpu_hw_sspp *ctx = pipe->sspp; u32 mode_mask; - u32 idx; - if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx)) + if (!ctx) return; if (pipe->multirect_index == DPU_SSPP_RECT_SOLO) { @@ -185,7 +152,7 @@ static void dpu_hw_sspp_setup_multirect(struct dpu_sw_pipe *pipe) */ mode_mask = 0; } else { - mode_mask = DPU_REG_READ(&ctx->hw, SSPP_MULTIRECT_OPMODE + idx); + mode_mask = DPU_REG_READ(&ctx->hw, SSPP_MULTIRECT_OPMODE); mode_mask |= pipe->multirect_index; if (pipe->multirect_mode == DPU_SSPP_MULTIRECT_TIME_MX) mode_mask |= BIT(2); @@ -193,46 +160,42 @@ static void dpu_hw_sspp_setup_multirect(struct dpu_sw_pipe *pipe) mode_mask &= ~BIT(2); } - DPU_REG_WRITE(&ctx->hw, SSPP_MULTIRECT_OPMODE + idx, mode_mask); + DPU_REG_WRITE(&ctx->hw, SSPP_MULTIRECT_OPMODE, mode_mask); } static void _sspp_setup_opmode(struct dpu_hw_sspp *ctx, u32 mask, u8 en) { - u32 idx; + const struct dpu_sspp_sub_blks *sblk = ctx->cap->sblk; u32 opmode; if (!test_bit(DPU_SSPP_SCALER_QSEED2, &ctx->cap->features) || - _sspp_subblk_offset(ctx, DPU_SSPP_SCALER_QSEED2, &idx) || !test_bit(DPU_SSPP_CSC, &ctx->cap->features)) return; - opmode = DPU_REG_READ(&ctx->hw, SSPP_VIG_OP_MODE + idx); + opmode = DPU_REG_READ(&ctx->hw, sblk->scaler_blk.base + SSPP_VIG_OP_MODE); if (en) opmode |= mask; else opmode &= ~mask; - DPU_REG_WRITE(&ctx->hw, SSPP_VIG_OP_MODE + idx, opmode); + DPU_REG_WRITE(&ctx->hw, sblk->scaler_blk.base + SSPP_VIG_OP_MODE, opmode); } static void _sspp_setup_csc10_opmode(struct dpu_hw_sspp *ctx, u32 mask, u8 en) { - u32 idx; + const struct dpu_sspp_sub_blks *sblk = ctx->cap->sblk; u32 opmode; - if (_sspp_subblk_offset(ctx, DPU_SSPP_CSC_10BIT, &idx)) - return; - - opmode = DPU_REG_READ(&ctx->hw, SSPP_VIG_CSC_10_OP_MODE + idx); + opmode = DPU_REG_READ(&ctx->hw, sblk->csc_blk.base + SSPP_VIG_CSC_10_OP_MODE); if (en) opmode |= mask; else opmode &= ~mask; - DPU_REG_WRITE(&ctx->hw, SSPP_VIG_CSC_10_OP_MODE + idx, opmode); + DPU_REG_WRITE(&ctx->hw, sblk->csc_blk.base + SSPP_VIG_CSC_10_OP_MODE, opmode); } /* @@ -247,9 +210,8 @@ static void dpu_hw_sspp_setup_format(struct dpu_sw_pipe *pipe, u32 opmode = 0; u32 fast_clear = 0; u32 op_mode_off, unpack_pat_off, format_off; - u32 idx; - if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx) || !fmt) + if (!ctx || !fmt) return; if (pipe->multirect_index == DPU_SSPP_RECT_SOLO || @@ -264,7 +226,7 @@ static void dpu_hw_sspp_setup_format(struct dpu_sw_pipe *pipe, } c = &ctx->hw; - opmode = DPU_REG_READ(c, op_mode_off + idx); + opmode = DPU_REG_READ(c, op_mode_off); opmode &= ~(MDSS_MDP_OP_FLIP_LR | MDSS_MDP_OP_FLIP_UD | MDSS_MDP_OP_BWC_EN | MDSS_MDP_OP_PE_OVERRIDE); @@ -352,12 +314,12 @@ static void dpu_hw_sspp_setup_format(struct dpu_sw_pipe *pipe, VIG_CSC_10_EN | VIG_CSC_10_SRC_DATAFMT, DPU_FORMAT_IS_YUV(fmt)); - DPU_REG_WRITE(c, format_off + idx, src_format); - DPU_REG_WRITE(c, unpack_pat_off + idx, unpack); - DPU_REG_WRITE(c, op_mode_off + idx, opmode); + DPU_REG_WRITE(c, format_off, src_format); + DPU_REG_WRITE(c, unpack_pat_off, unpack); + DPU_REG_WRITE(c, op_mode_off, opmode); /* clear previous UBWC error */ - DPU_REG_WRITE(c, SSPP_UBWC_ERROR_STATUS + idx, BIT(31)); + DPU_REG_WRITE(c, SSPP_UBWC_ERROR_STATUS, BIT(31)); } static void dpu_hw_sspp_setup_pe_config(struct dpu_hw_sspp *ctx, @@ -368,9 +330,8 @@ static void dpu_hw_sspp_setup_pe_config(struct dpu_hw_sspp *ctx, u32 lr_pe[4], tb_pe[4], tot_req_pixels[4]; const u32 bytemask = 0xff; const u32 shortmask = 0xffff; - u32 idx; - if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx) || !pe_ext) + if (!ctx || !pe_ext) return; c = &ctx->hw; @@ -400,21 +361,21 @@ static void dpu_hw_sspp_setup_pe_config(struct dpu_hw_sspp *ctx, } /* color 0 */ - DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C0_LR + idx, lr_pe[0]); - DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C0_TB + idx, tb_pe[0]); - DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C0_REQ_PIXELS + idx, + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C0_LR, lr_pe[0]); + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C0_TB, tb_pe[0]); + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C0_REQ_PIXELS, tot_req_pixels[0]); /* color 1 and color 2 */ - DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C1C2_LR + idx, lr_pe[1]); - DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C1C2_TB + idx, tb_pe[1]); - DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C1C2_REQ_PIXELS + idx, + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C1C2_LR, lr_pe[1]); + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C1C2_TB, tb_pe[1]); + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C1C2_REQ_PIXELS, tot_req_pixels[1]); /* color 3 */ - DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C3_LR + idx, lr_pe[3]); - DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C3_TB + idx, lr_pe[3]); - DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C3_REQ_PIXELS + idx, + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C3_LR, lr_pe[3]); + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C3_TB, lr_pe[3]); + DPU_REG_WRITE(c, SSPP_SW_PIX_EXT_C3_REQ_PIXELS, tot_req_pixels[3]); } @@ -422,25 +383,22 @@ static void _dpu_hw_sspp_setup_scaler3(struct dpu_hw_sspp *ctx, struct dpu_hw_scaler3_cfg *scaler3_cfg, const struct dpu_format *format) { - u32 idx; - - if (_sspp_subblk_offset(ctx, DPU_SSPP_SCALER_QSEED3, &idx) - || !scaler3_cfg) + if (!ctx || !scaler3_cfg) return; - dpu_hw_setup_scaler3(&ctx->hw, scaler3_cfg, idx, + dpu_hw_setup_scaler3(&ctx->hw, scaler3_cfg, + ctx->cap->sblk->scaler_blk.base, ctx->cap->sblk->scaler_blk.version, format); } static u32 _dpu_hw_sspp_get_scaler3_ver(struct dpu_hw_sspp *ctx) { - u32 idx; - - if (!ctx || _sspp_subblk_offset(ctx, DPU_SSPP_SCALER_QSEED3, &idx)) + if (!ctx) return 0; - return dpu_hw_get_scaler3_ver(&ctx->hw, idx); + return dpu_hw_get_scaler3_ver(&ctx->hw, + ctx->cap->sblk->scaler_blk.base); } /* @@ -453,9 +411,8 @@ static void dpu_hw_sspp_setup_rects(struct dpu_sw_pipe *pipe, struct dpu_hw_blk_reg_map *c; u32 src_size, src_xy, dst_size, dst_xy; u32 src_size_off, src_xy_off, out_size_off, out_xy_off; - u32 idx; - if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx) || !cfg) + if (!ctx || !cfg) return; c = &ctx->hw; @@ -483,10 +440,10 @@ static void dpu_hw_sspp_setup_rects(struct dpu_sw_pipe *pipe, drm_rect_width(&cfg->dst_rect); /* rectangle register programming */ - DPU_REG_WRITE(c, src_size_off + idx, src_size); - DPU_REG_WRITE(c, src_xy_off + idx, src_xy); - DPU_REG_WRITE(c, out_size_off + idx, dst_size); - DPU_REG_WRITE(c, out_xy_off + idx, dst_xy); + DPU_REG_WRITE(c, src_size_off, src_size); + DPU_REG_WRITE(c, src_xy_off, src_xy); + DPU_REG_WRITE(c, out_size_off, dst_size); + DPU_REG_WRITE(c, out_xy_off, dst_xy); } static void dpu_hw_sspp_setup_sourceaddress(struct dpu_sw_pipe *pipe, @@ -495,24 +452,23 @@ static void dpu_hw_sspp_setup_sourceaddress(struct dpu_sw_pipe *pipe, struct dpu_hw_sspp *ctx = pipe->sspp; u32 ystride0, ystride1; int i; - u32 idx; - if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx)) + if (!ctx) return; if (pipe->multirect_index == DPU_SSPP_RECT_SOLO) { for (i = 0; i < ARRAY_SIZE(layout->plane_addr); i++) - DPU_REG_WRITE(&ctx->hw, SSPP_SRC0_ADDR + idx + i * 0x4, + DPU_REG_WRITE(&ctx->hw, SSPP_SRC0_ADDR + i * 0x4, layout->plane_addr[i]); } else if (pipe->multirect_index == DPU_SSPP_RECT_0) { - DPU_REG_WRITE(&ctx->hw, SSPP_SRC0_ADDR + idx, + DPU_REG_WRITE(&ctx->hw, SSPP_SRC0_ADDR, layout->plane_addr[0]); - DPU_REG_WRITE(&ctx->hw, SSPP_SRC2_ADDR + idx, + DPU_REG_WRITE(&ctx->hw, SSPP_SRC2_ADDR, layout->plane_addr[2]); } else { - DPU_REG_WRITE(&ctx->hw, SSPP_SRC1_ADDR + idx, + DPU_REG_WRITE(&ctx->hw, SSPP_SRC1_ADDR, layout->plane_addr[0]); - DPU_REG_WRITE(&ctx->hw, SSPP_SRC3_ADDR + idx, + DPU_REG_WRITE(&ctx->hw, SSPP_SRC3_ADDR, layout->plane_addr[2]); } @@ -522,8 +478,8 @@ static void dpu_hw_sspp_setup_sourceaddress(struct dpu_sw_pipe *pipe, ystride1 = (layout->plane_pitch[2]) | (layout->plane_pitch[3] << 16); } else { - ystride0 = DPU_REG_READ(&ctx->hw, SSPP_SRC_YSTRIDE0 + idx); - ystride1 = DPU_REG_READ(&ctx->hw, SSPP_SRC_YSTRIDE1 + idx); + ystride0 = DPU_REG_READ(&ctx->hw, SSPP_SRC_YSTRIDE0); + ystride1 = DPU_REG_READ(&ctx->hw, SSPP_SRC_YSTRIDE1); if (pipe->multirect_index == DPU_SSPP_RECT_0) { ystride0 = (ystride0 & 0xFFFF0000) | @@ -540,34 +496,35 @@ static void dpu_hw_sspp_setup_sourceaddress(struct dpu_sw_pipe *pipe, } } - DPU_REG_WRITE(&ctx->hw, SSPP_SRC_YSTRIDE0 + idx, ystride0); - DPU_REG_WRITE(&ctx->hw, SSPP_SRC_YSTRIDE1 + idx, ystride1); + DPU_REG_WRITE(&ctx->hw, SSPP_SRC_YSTRIDE0, ystride0); + DPU_REG_WRITE(&ctx->hw, SSPP_SRC_YSTRIDE1, ystride1); } static void dpu_hw_sspp_setup_csc(struct dpu_hw_sspp *ctx, const struct dpu_csc_cfg *data) { - u32 idx; + u32 offset; bool csc10 = false; - if (_sspp_subblk_offset(ctx, DPU_SSPP_CSC, &idx) || !data) + if (!ctx || !data) return; + offset = ctx->cap->sblk->csc_blk.base; + if (test_bit(DPU_SSPP_CSC_10BIT, &ctx->cap->features)) { - idx += CSC_10BIT_OFFSET; + offset += CSC_10BIT_OFFSET; csc10 = true; } - dpu_hw_csc_setup(&ctx->hw, idx, data, csc10); + dpu_hw_csc_setup(&ctx->hw, offset, data, csc10); } static void dpu_hw_sspp_setup_solidfill(struct dpu_sw_pipe *pipe, u32 color) { struct dpu_hw_sspp *ctx = pipe->sspp; struct dpu_hw_fmt_layout cfg; - u32 idx; - if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx)) + if (!ctx) return; /* cleanup source addresses */ @@ -576,79 +533,41 @@ static void dpu_hw_sspp_setup_solidfill(struct dpu_sw_pipe *pipe, u32 color) if (pipe->multirect_index == DPU_SSPP_RECT_SOLO || pipe->multirect_index == DPU_SSPP_RECT_0) - DPU_REG_WRITE(&ctx->hw, SSPP_SRC_CONSTANT_COLOR + idx, color); + DPU_REG_WRITE(&ctx->hw, SSPP_SRC_CONSTANT_COLOR, color); else - DPU_REG_WRITE(&ctx->hw, SSPP_SRC_CONSTANT_COLOR_REC1 + idx, + DPU_REG_WRITE(&ctx->hw, SSPP_SRC_CONSTANT_COLOR_REC1, color); } -static void dpu_hw_sspp_setup_danger_safe_lut(struct dpu_hw_sspp *ctx, - u32 danger_lut, - u32 safe_lut) +static void dpu_hw_sspp_setup_qos_lut(struct dpu_hw_sspp *ctx, + struct dpu_hw_qos_cfg *cfg) { - u32 idx; - - if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx)) - return; - - DPU_REG_WRITE(&ctx->hw, SSPP_DANGER_LUT + idx, danger_lut); - DPU_REG_WRITE(&ctx->hw, SSPP_SAFE_LUT + idx, safe_lut); -} - -static void dpu_hw_sspp_setup_creq_lut(struct dpu_hw_sspp *ctx, - u64 creq_lut) -{ - u32 idx; - - if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx)) + if (!ctx || !cfg) return; - if (ctx->cap && test_bit(DPU_SSPP_QOS_8LVL, &ctx->cap->features)) { - DPU_REG_WRITE(&ctx->hw, SSPP_CREQ_LUT_0 + idx, creq_lut); - DPU_REG_WRITE(&ctx->hw, SSPP_CREQ_LUT_1 + idx, - creq_lut >> 32); - } else { - DPU_REG_WRITE(&ctx->hw, SSPP_CREQ_LUT + idx, creq_lut); - } + _dpu_hw_setup_qos_lut(&ctx->hw, SSPP_DANGER_LUT, + test_bit(DPU_SSPP_QOS_8LVL, &ctx->cap->features), + cfg); } static void dpu_hw_sspp_setup_qos_ctrl(struct dpu_hw_sspp *ctx, - struct dpu_hw_pipe_qos_cfg *cfg) + bool danger_safe_en) { - u32 idx; - u32 qos_ctrl = 0; - - if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx)) + if (!ctx) return; - if (cfg->vblank_en) { - qos_ctrl |= ((cfg->creq_vblank & - SSPP_QOS_CTRL_CREQ_VBLANK_MASK) << - SSPP_QOS_CTRL_CREQ_VBLANK_OFF); - qos_ctrl |= ((cfg->danger_vblank & - SSPP_QOS_CTRL_DANGER_VBLANK_MASK) << - SSPP_QOS_CTRL_DANGER_VBLANK_OFF); - qos_ctrl |= SSPP_QOS_CTRL_VBLANK_EN; - } - - if (cfg->danger_safe_en) - qos_ctrl |= SSPP_QOS_CTRL_DANGER_SAFE_EN; - - DPU_REG_WRITE(&ctx->hw, SSPP_QOS_CTRL + idx, qos_ctrl); + DPU_REG_WRITE(&ctx->hw, SSPP_QOS_CTRL, + danger_safe_en ? SSPP_QOS_CTRL_DANGER_SAFE_EN : 0); } static void dpu_hw_sspp_setup_cdp(struct dpu_sw_pipe *pipe, - struct dpu_hw_cdp_cfg *cfg) + const struct dpu_format *fmt, + bool enable) { struct dpu_hw_sspp *ctx = pipe->sspp; - u32 idx; - u32 cdp_cntl = 0; u32 cdp_cntl_offset = 0; - if (!ctx || !cfg) - return; - - if (_sspp_subblk_offset(ctx, DPU_SSPP_SRC, &idx)) + if (!ctx) return; if (pipe->multirect_index == DPU_SSPP_RECT_SOLO || @@ -657,33 +576,20 @@ static void dpu_hw_sspp_setup_cdp(struct dpu_sw_pipe *pipe, else cdp_cntl_offset = SSPP_CDP_CNTL_REC1; - if (cfg->enable) - cdp_cntl |= BIT(0); - if (cfg->ubwc_meta_enable) - cdp_cntl |= BIT(1); - if (cfg->tile_amortize_enable) - cdp_cntl |= BIT(2); - if (cfg->preload_ahead == DPU_SSPP_CDP_PRELOAD_AHEAD_64) - cdp_cntl |= BIT(3); - - DPU_REG_WRITE(&ctx->hw, cdp_cntl_offset, cdp_cntl); + dpu_setup_cdp(&ctx->hw, cdp_cntl_offset, fmt, enable); } static void _setup_layer_ops(struct dpu_hw_sspp *c, unsigned long features) { - if (test_bit(DPU_SSPP_SRC, &features)) { - c->ops.setup_format = dpu_hw_sspp_setup_format; - c->ops.setup_rects = dpu_hw_sspp_setup_rects; - c->ops.setup_sourceaddress = dpu_hw_sspp_setup_sourceaddress; - c->ops.setup_solidfill = dpu_hw_sspp_setup_solidfill; - c->ops.setup_pe = dpu_hw_sspp_setup_pe_config; - } + c->ops.setup_format = dpu_hw_sspp_setup_format; + c->ops.setup_rects = dpu_hw_sspp_setup_rects; + c->ops.setup_sourceaddress = dpu_hw_sspp_setup_sourceaddress; + c->ops.setup_solidfill = dpu_hw_sspp_setup_solidfill; + c->ops.setup_pe = dpu_hw_sspp_setup_pe_config; if (test_bit(DPU_SSPP_QOS, &features)) { - c->ops.setup_danger_safe_lut = - dpu_hw_sspp_setup_danger_safe_lut; - c->ops.setup_creq_lut = dpu_hw_sspp_setup_creq_lut; + c->ops.setup_qos_lut = dpu_hw_sspp_setup_qos_lut; c->ops.setup_qos_ctrl = dpu_hw_sspp_setup_qos_ctrl; } @@ -728,8 +634,8 @@ int _dpu_hw_sspp_init_debugfs(struct dpu_hw_sspp *hw_pipe, struct dpu_kms *kms, /* add register dump support */ dpu_debugfs_create_regset32("src_blk", 0400, debugfs_root, - sblk->src_blk.base + cfg->base, - sblk->src_blk.len, + cfg->base, + cfg->len, kms); if (cfg->features & BIT(DPU_SSPP_SCALER_QSEED3) || @@ -758,63 +664,29 @@ int _dpu_hw_sspp_init_debugfs(struct dpu_hw_sspp *hw_pipe, struct dpu_kms *kms, 0400, debugfs_root, (u32 *) &cfg->clk_ctrl); - debugfs_create_x32("creq_vblank", - 0600, - debugfs_root, - (u32 *) &sblk->creq_vblank); - debugfs_create_x32("danger_vblank", - 0600, - debugfs_root, - (u32 *) &sblk->danger_vblank); return 0; } #endif - -static const struct dpu_sspp_cfg *_sspp_offset(enum dpu_sspp sspp, - void __iomem *addr, - const struct dpu_mdss_cfg *catalog, - struct dpu_hw_blk_reg_map *b) -{ - int i; - - if ((sspp < SSPP_MAX) && catalog && addr && b) { - for (i = 0; i < catalog->sspp_count; i++) { - if (sspp == catalog->sspp[i].id) { - b->blk_addr = addr + catalog->sspp[i].base; - b->log_mask = DPU_DBG_MASK_SSPP; - return &catalog->sspp[i]; - } - } - } - - return ERR_PTR(-ENOMEM); -} - -struct dpu_hw_sspp *dpu_hw_sspp_init(enum dpu_sspp idx, - void __iomem *addr, const struct dpu_mdss_cfg *catalog) +struct dpu_hw_sspp *dpu_hw_sspp_init(const struct dpu_sspp_cfg *cfg, + void __iomem *addr, const struct dpu_ubwc_cfg *ubwc) { struct dpu_hw_sspp *hw_pipe; - const struct dpu_sspp_cfg *cfg; - if (!addr || !catalog) + if (!addr || !ubwc) return ERR_PTR(-EINVAL); hw_pipe = kzalloc(sizeof(*hw_pipe), GFP_KERNEL); if (!hw_pipe) return ERR_PTR(-ENOMEM); - cfg = _sspp_offset(idx, addr, catalog, &hw_pipe->hw); - if (IS_ERR_OR_NULL(cfg)) { - kfree(hw_pipe); - return ERR_PTR(-EINVAL); - } + hw_pipe->hw.blk_addr = addr + cfg->base; + hw_pipe->hw.log_mask = DPU_DBG_MASK_SSPP; /* Assign ops */ - hw_pipe->catalog = catalog; - hw_pipe->ubwc = catalog->ubwc; - hw_pipe->idx = idx; + hw_pipe->ubwc = ubwc; + hw_pipe->idx = cfg->id; hw_pipe->cap = cfg; _setup_layer_ops(hw_pipe, hw_pipe->cap->features); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h index 74b98b6b3bc3a19151ae0834079d0237847686d9..085f34bc6b88d430c8176c1e9d4850bc069f90c7 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h @@ -163,28 +163,6 @@ struct dpu_sw_pipe_cfg { struct drm_rect dst_rect; }; -/** - * struct dpu_hw_pipe_qos_cfg : Source pipe QoS configuration - * @creq_vblank: creq value generated to vbif during vertical blanking - * @danger_vblank: danger value generated during vertical blanking - * @vblank_en: enable creq_vblank and danger_vblank during vblank - * @danger_safe_en: enable danger safe generation - */ -struct dpu_hw_pipe_qos_cfg { - u32 creq_vblank; - u32 danger_vblank; - bool vblank_en; - bool danger_safe_en; -}; - -/** - * enum CDP preload ahead address size - */ -enum { - DPU_SSPP_CDP_PRELOAD_AHEAD_32, - DPU_SSPP_CDP_PRELOAD_AHEAD_64 -}; - /** * struct dpu_hw_pipe_ts_cfg - traffic shaper configuration * @size: size to prefill in bytes, or zero to disable @@ -276,34 +254,22 @@ struct dpu_hw_sspp_ops { void (*setup_sharpening)(struct dpu_hw_sspp *ctx, struct dpu_hw_sharp_cfg *cfg); - /** - * setup_danger_safe_lut - setup danger safe LUTs - * @ctx: Pointer to pipe context - * @danger_lut: LUT for generate danger level based on fill level - * @safe_lut: LUT for generate safe level based on fill level - * - */ - void (*setup_danger_safe_lut)(struct dpu_hw_sspp *ctx, - u32 danger_lut, - u32 safe_lut); /** - * setup_creq_lut - setup CREQ LUT + * setup_qos_lut - setup QoS LUTs * @ctx: Pointer to pipe context - * @creq_lut: LUT for generate creq level based on fill level - * + * @cfg: LUT configuration */ - void (*setup_creq_lut)(struct dpu_hw_sspp *ctx, - u64 creq_lut); + void (*setup_qos_lut)(struct dpu_hw_sspp *ctx, + struct dpu_hw_qos_cfg *cfg); /** * setup_qos_ctrl - setup QoS control * @ctx: Pointer to pipe context - * @cfg: Pointer to pipe QoS configuration - * + * @danger_safe_en: flags controlling enabling of danger/safe QoS/LUT */ void (*setup_qos_ctrl)(struct dpu_hw_sspp *ctx, - struct dpu_hw_pipe_qos_cfg *cfg); + bool danger_safe_en); /** * setup_histogram - setup histograms @@ -331,18 +297,19 @@ struct dpu_hw_sspp_ops { /** * setup_cdp - setup client driven prefetch * @pipe: Pointer to software pipe context - * @cfg: Pointer to cdp configuration + * @fmt: format used by the sw pipe + * @enable: whether the CDP should be enabled for this pipe */ void (*setup_cdp)(struct dpu_sw_pipe *pipe, - struct dpu_hw_cdp_cfg *cfg); + const struct dpu_format *fmt, + bool enable); }; /** * struct dpu_hw_sspp - pipe description * @base: hardware block base structure * @hw: block hardware details - * @catalog: back pointer to catalog - * @ubwc: ubwc configuration data + * @ubwc: UBWC configuration data * @idx: pipe index * @cap: pointer to layer_cfg * @ops: pointer to operations possible for this pipe @@ -350,7 +317,6 @@ struct dpu_hw_sspp_ops { struct dpu_hw_sspp { struct dpu_hw_blk base; struct dpu_hw_blk_reg_map hw; - const struct dpu_mdss_cfg *catalog; const struct dpu_ubwc_cfg *ubwc; /* Pipe */ @@ -363,14 +329,14 @@ struct dpu_hw_sspp { struct dpu_kms; /** - * dpu_hw_sspp_init - initializes the sspp hw driver object. + * dpu_hw_sspp_init() - Initializes the sspp hw driver object. * Should be called once before accessing every pipe. - * @idx: Pipe index for which driver object is required + * @cfg: Pipe catalog entry for which driver object is required * @addr: Mapped register io address of MDP - * @catalog : Pointer to mdss catalog data + * @ubwc: UBWC configuration data */ -struct dpu_hw_sspp *dpu_hw_sspp_init(enum dpu_sspp idx, - void __iomem *addr, const struct dpu_mdss_cfg *catalog); +struct dpu_hw_sspp *dpu_hw_sspp_init(const struct dpu_sspp_cfg *cfg, + void __iomem *addr, const struct dpu_ubwc_cfg *ubwc); /** * dpu_hw_sspp_destroy(): Destroys SSPP driver context diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c index 2bb02e17ee52c4d0664f1e83c7df90e038a9a026..963bdb5e025219649bff4833ad7d6d22740990a3 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c @@ -130,24 +130,12 @@ static void dpu_hw_setup_vsync_source(struct dpu_hw_mdp *mdp, struct dpu_vsync_source_cfg *cfg) { struct dpu_hw_blk_reg_map *c; - u32 reg, wd_load_value, wd_ctl, wd_ctl2, i; - static const u32 pp_offset[PINGPONG_MAX] = {0xC, 0x8, 0x4, 0x13, 0x18}; + u32 reg, wd_load_value, wd_ctl, wd_ctl2; - if (!mdp || !cfg || (cfg->pp_count > ARRAY_SIZE(cfg->ppnumber))) + if (!mdp || !cfg) return; c = &mdp->hw; - reg = DPU_REG_READ(c, MDP_VSYNC_SEL); - for (i = 0; i < cfg->pp_count; i++) { - int pp_idx = cfg->ppnumber[i] - PINGPONG_0; - - if (pp_idx >= ARRAY_SIZE(pp_offset)) - continue; - - reg &= ~(0xf << pp_offset[pp_idx]); - reg |= (cfg->vsync_source & 0xf) << pp_offset[pp_idx]; - } - DPU_REG_WRITE(c, MDP_VSYNC_SEL, reg); if (cfg->vsync_source >= DPU_VSYNC_SOURCE_WD_TIMER_4 && cfg->vsync_source <= DPU_VSYNC_SOURCE_WD_TIMER_0) { @@ -194,6 +182,33 @@ static void dpu_hw_setup_vsync_source(struct dpu_hw_mdp *mdp, } } +static void dpu_hw_setup_vsync_source_and_vsync_sel(struct dpu_hw_mdp *mdp, + struct dpu_vsync_source_cfg *cfg) +{ + struct dpu_hw_blk_reg_map *c; + u32 reg, i; + static const u32 pp_offset[PINGPONG_MAX] = {0xC, 0x8, 0x4, 0x13, 0x18}; + + if (!mdp || !cfg || (cfg->pp_count > ARRAY_SIZE(cfg->ppnumber))) + return; + + c = &mdp->hw; + + reg = DPU_REG_READ(c, MDP_VSYNC_SEL); + for (i = 0; i < cfg->pp_count; i++) { + int pp_idx = cfg->ppnumber[i] - PINGPONG_0; + + if (pp_idx >= ARRAY_SIZE(pp_offset)) + continue; + + reg &= ~(0xf << pp_offset[pp_idx]); + reg |= (cfg->vsync_source & 0xf) << pp_offset[pp_idx]; + } + DPU_REG_WRITE(c, MDP_VSYNC_SEL, reg); + + dpu_hw_setup_vsync_source(mdp, cfg); +} + static void dpu_hw_get_safe_status(struct dpu_hw_mdp *mdp, struct dpu_danger_safe_status *status) { @@ -241,7 +256,12 @@ static void _setup_mdp_ops(struct dpu_hw_mdp_ops *ops, ops->setup_split_pipe = dpu_hw_setup_split_pipe; ops->setup_clk_force_ctrl = dpu_hw_setup_clk_force_ctrl; ops->get_danger_status = dpu_hw_get_danger_status; - ops->setup_vsync_source = dpu_hw_setup_vsync_source; + + if (cap & BIT(DPU_MDP_VSYNC_SEL)) + ops->setup_vsync_source = dpu_hw_setup_vsync_source_and_vsync_sel; + else + ops->setup_vsync_source = dpu_hw_setup_vsync_source; + ops->get_safe_status = dpu_hw_get_safe_status; if (cap & BIT(DPU_MDP_AUDIO_SELECT)) diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c index 8062228eada687dcfb074b292eb97d6359e0730d..9d2273fd2fed58e5a58965d78958bfea49272b8d 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.c @@ -73,6 +73,19 @@ static u32 dpu_hw_util_log_mask = DPU_DBG_MASK_NONE; #define QSEED3LITE_SEP_LUT_SIZE \ (QSEED3LITE_LUT_SIZE * QSEED3LITE_SEPARABLE_LUTS * sizeof(u32)) +/* QOS_LUT */ +#define QOS_DANGER_LUT 0x00 +#define QOS_SAFE_LUT 0x04 +#define QOS_CREQ_LUT 0x08 +#define QOS_QOS_CTRL 0x0C +#define QOS_CREQ_LUT_0 0x14 +#define QOS_CREQ_LUT_1 0x18 + +/* QOS_QOS_CTRL */ +#define QOS_QOS_CTRL_DANGER_SAFE_EN BIT(0) +#define QOS_QOS_CTRL_DANGER_VBLANK_MASK GENMASK(5, 4) +#define QOS_QOS_CTRL_VBLANK_EN BIT(16) +#define QOS_QOS_CTRL_CREQ_VBLANK_MASK GENMASK(21, 20) void dpu_reg_write(struct dpu_hw_blk_reg_map *c, u32 reg_off, @@ -450,6 +463,24 @@ u64 _dpu_hw_get_qos_lut(const struct dpu_qos_lut_tbl *tbl, return 0; } +void _dpu_hw_setup_qos_lut(struct dpu_hw_blk_reg_map *c, u32 offset, + bool qos_8lvl, + const struct dpu_hw_qos_cfg *cfg) +{ + DPU_REG_WRITE(c, offset + QOS_DANGER_LUT, cfg->danger_lut); + DPU_REG_WRITE(c, offset + QOS_SAFE_LUT, cfg->safe_lut); + + if (qos_8lvl) { + DPU_REG_WRITE(c, offset + QOS_CREQ_LUT_0, cfg->creq_lut); + DPU_REG_WRITE(c, offset + QOS_CREQ_LUT_1, cfg->creq_lut >> 32); + } else { + DPU_REG_WRITE(c, offset + QOS_CREQ_LUT, cfg->creq_lut); + } + + DPU_REG_WRITE(c, offset + QOS_QOS_CTRL, + cfg->danger_safe_en ? QOS_QOS_CTRL_DANGER_SAFE_EN : 0); +} + void dpu_hw_setup_misr(struct dpu_hw_blk_reg_map *c, u32 misr_ctrl_offset, bool enable, u32 frame_count) @@ -494,3 +525,24 @@ int dpu_hw_collect_misr(struct dpu_hw_blk_reg_map *c, return 0; } + +#define CDP_ENABLE BIT(0) +#define CDP_UBWC_META_ENABLE BIT(1) +#define CDP_TILE_AMORTIZE_ENABLE BIT(2) +#define CDP_PRELOAD_AHEAD_64 BIT(3) + +void dpu_setup_cdp(struct dpu_hw_blk_reg_map *c, u32 offset, + const struct dpu_format *fmt, bool enable) +{ + u32 cdp_cntl = CDP_PRELOAD_AHEAD_64; + + if (enable) + cdp_cntl |= CDP_ENABLE; + if (DPU_FORMAT_IS_UBWC(fmt)) + cdp_cntl |= CDP_UBWC_META_ENABLE; + if (DPU_FORMAT_IS_UBWC(fmt) || + DPU_FORMAT_IS_TILE(fmt)) + cdp_cntl |= CDP_TILE_AMORTIZE_ENABLE; + + DPU_REG_WRITE(c, offset, cdp_cntl); +} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h index 27f4c39e35ab3b70a83c33235835692c05182b05..1f6079f47071095774cd75d826038a93b57cac6c 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_util.h @@ -306,19 +306,20 @@ struct dpu_drm_scaler_v2 { }; /** - * struct dpu_hw_cdp_cfg : CDP configuration - * @enable: true to enable CDP - * @ubwc_meta_enable: true to enable ubwc metadata preload - * @tile_amortize_enable: true to enable amortization control for tile format - * @preload_ahead: number of request to preload ahead - * DPU_*_CDP_PRELOAD_AHEAD_32, - * DPU_*_CDP_PRELOAD_AHEAD_64 + * struct dpu_hw_qos_cfg: pipe QoS configuration + * @danger_lut: LUT for generate danger level based on fill level + * @safe_lut: LUT for generate safe level based on fill level + * @creq_lut: LUT for generate creq level based on fill level + * @creq_vblank: creq value generated to vbif during vertical blanking + * @danger_vblank: danger value generated during vertical blanking + * @vblank_en: enable creq_vblank and danger_vblank during vblank + * @danger_safe_en: enable danger safe generation */ -struct dpu_hw_cdp_cfg { - bool enable; - bool ubwc_meta_enable; - bool tile_amortize_enable; - u32 preload_ahead; +struct dpu_hw_qos_cfg { + u32 danger_lut; + u32 safe_lut; + u64 creq_lut; + bool danger_safe_en; }; u32 *dpu_hw_util_get_log_mask_ptr(void); @@ -346,9 +347,16 @@ void dpu_hw_csc_setup(struct dpu_hw_blk_reg_map *c, u32 csc_reg_off, const struct dpu_csc_cfg *data, bool csc10); +void dpu_setup_cdp(struct dpu_hw_blk_reg_map *c, u32 offset, + const struct dpu_format *fmt, bool enable); + u64 _dpu_hw_get_qos_lut(const struct dpu_qos_lut_tbl *tbl, u32 total_fl); +void _dpu_hw_setup_qos_lut(struct dpu_hw_blk_reg_map *c, u32 offset, + bool qos_8lvl, + const struct dpu_hw_qos_cfg *cfg); + void dpu_hw_setup_misr(struct dpu_hw_blk_reg_map *c, u32 misr_ctrl_offset, bool enable, diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c index 16c56e2407064bd1c2e7611982ace5fd234a5849..a5121a50b2bbf8cc4b6cc7becbfd8c48510cd465 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c @@ -211,45 +211,22 @@ static void _setup_vbif_ops(struct dpu_hw_vbif_ops *ops, ops->set_write_gather_en = dpu_hw_set_write_gather_en; } -static const struct dpu_vbif_cfg *_top_offset(enum dpu_vbif vbif, - const struct dpu_mdss_cfg *m, - void __iomem *addr, - struct dpu_hw_blk_reg_map *b) -{ - int i; - - for (i = 0; i < m->vbif_count; i++) { - if (vbif == m->vbif[i].id) { - b->blk_addr = addr + m->vbif[i].base; - b->log_mask = DPU_DBG_MASK_VBIF; - return &m->vbif[i]; - } - } - - return ERR_PTR(-EINVAL); -} - -struct dpu_hw_vbif *dpu_hw_vbif_init(enum dpu_vbif idx, - void __iomem *addr, - const struct dpu_mdss_cfg *m) +struct dpu_hw_vbif *dpu_hw_vbif_init(const struct dpu_vbif_cfg *cfg, + void __iomem *addr) { struct dpu_hw_vbif *c; - const struct dpu_vbif_cfg *cfg; c = kzalloc(sizeof(*c), GFP_KERNEL); if (!c) return ERR_PTR(-ENOMEM); - cfg = _top_offset(idx, m, addr, &c->hw); - if (IS_ERR_OR_NULL(cfg)) { - kfree(c); - return ERR_PTR(-EINVAL); - } + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_VBIF; /* * Assign ops */ - c->idx = idx; + c->idx = cfg->id; c->cap = cfg; _setup_vbif_ops(&c->ops, c->cap->features); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.h index 6417aa28d32c336287e2d6f425113aa36c03619f..7e10d2a172b45aef00bc53a6686ae753741d6780 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.h @@ -106,14 +106,13 @@ struct dpu_hw_vbif { }; /** - * dpu_hw_vbif_init - initializes the vbif driver for the passed interface idx - * @idx: Interface index for which driver object is required + * dpu_hw_vbif_init() - Initializes the VBIF driver for the passed + * VBIF catalog entry. + * @cfg: VBIF catalog entry for which driver object is required * @addr: Mapped register io address of MDSS - * @m: Pointer to mdss catalog data */ -struct dpu_hw_vbif *dpu_hw_vbif_init(enum dpu_vbif idx, - void __iomem *addr, - const struct dpu_mdss_cfg *m); +struct dpu_hw_vbif *dpu_hw_vbif_init(const struct dpu_vbif_cfg *cfg, + void __iomem *addr); void dpu_hw_vbif_destroy(struct dpu_hw_vbif *vbif); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c index 2d28afdf860efc43709e1c648be6607763ef4603..ebc416400382202194550ed789f7f4c0bb3ecb64 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c @@ -49,24 +49,6 @@ #define WB_OUT_IMAGE_SIZE 0x2C0 #define WB_OUT_XY 0x2C4 -/* WB_QOS_CTRL */ -#define WB_QOS_CTRL_DANGER_SAFE_EN BIT(0) - -static const struct dpu_wb_cfg *_wb_offset(enum dpu_wb wb, - const struct dpu_mdss_cfg *m, void __iomem *addr, - struct dpu_hw_blk_reg_map *b) -{ - int i; - - for (i = 0; i < m->wb_count; i++) { - if (wb == m->wb[i].id) { - b->blk_addr = addr + m->wb[i].base; - return &m->wb[i]; - } - } - return ERR_PTR(-EINVAL); -} - static void dpu_hw_wb_setup_outaddress(struct dpu_hw_wb *ctx, struct dpu_hw_wb_cfg *data) { @@ -150,58 +132,29 @@ static void dpu_hw_wb_roi(struct dpu_hw_wb *ctx, struct dpu_hw_wb_cfg *wb) } static void dpu_hw_wb_setup_qos_lut(struct dpu_hw_wb *ctx, - struct dpu_hw_wb_qos_cfg *cfg) + struct dpu_hw_qos_cfg *cfg) { - struct dpu_hw_blk_reg_map *c = &ctx->hw; - u32 qos_ctrl = 0; - if (!ctx || !cfg) return; - DPU_REG_WRITE(c, WB_DANGER_LUT, cfg->danger_lut); - DPU_REG_WRITE(c, WB_SAFE_LUT, cfg->safe_lut); - - /* - * for chipsets not using DPU_WB_QOS_8LVL but still using DPU - * driver such as msm8998, the reset value of WB_CREQ_LUT is - * sufficient for writeback to work. SW doesn't need to explicitly - * program a value. - */ - if (ctx->caps && test_bit(DPU_WB_QOS_8LVL, &ctx->caps->features)) { - DPU_REG_WRITE(c, WB_CREQ_LUT_0, cfg->creq_lut); - DPU_REG_WRITE(c, WB_CREQ_LUT_1, cfg->creq_lut >> 32); - } - - if (cfg->danger_safe_en) - qos_ctrl |= WB_QOS_CTRL_DANGER_SAFE_EN; - - DPU_REG_WRITE(c, WB_QOS_CTRL, qos_ctrl); + _dpu_hw_setup_qos_lut(&ctx->hw, WB_DANGER_LUT, + test_bit(DPU_WB_QOS_8LVL, &ctx->caps->features), + cfg); } static void dpu_hw_wb_setup_cdp(struct dpu_hw_wb *ctx, - struct dpu_hw_cdp_cfg *cfg) + const struct dpu_format *fmt, + bool enable) { - struct dpu_hw_blk_reg_map *c; - u32 cdp_cntl = 0; - - if (!ctx || !cfg) + if (!ctx) return; - c = &ctx->hw; - - if (cfg->enable) - cdp_cntl |= BIT(0); - if (cfg->ubwc_meta_enable) - cdp_cntl |= BIT(1); - if (cfg->preload_ahead == DPU_WB_CDP_PRELOAD_AHEAD_64) - cdp_cntl |= BIT(3); - - DPU_REG_WRITE(c, WB_CDP_CNTL, cdp_cntl); + dpu_setup_cdp(&ctx->hw, WB_CDP_CNTL, fmt, enable); } static void dpu_hw_wb_bind_pingpong_blk( struct dpu_hw_wb *ctx, - bool enable, const enum dpu_pingpong pp) + const enum dpu_pingpong pp) { struct dpu_hw_blk_reg_map *c; int mux_cfg; @@ -214,7 +167,7 @@ static void dpu_hw_wb_bind_pingpong_blk( mux_cfg = DPU_REG_READ(c, WB_MUX); mux_cfg &= ~0xf; - if (enable) + if (pp) mux_cfg |= (pp - PINGPONG_0) & 0x7; else mux_cfg |= 0xf; @@ -241,29 +194,23 @@ static void _setup_wb_ops(struct dpu_hw_wb_ops *ops, ops->bind_pingpong_blk = dpu_hw_wb_bind_pingpong_blk; } -struct dpu_hw_wb *dpu_hw_wb_init(enum dpu_wb idx, - void __iomem *addr, const struct dpu_mdss_cfg *m) +struct dpu_hw_wb *dpu_hw_wb_init(const struct dpu_wb_cfg *cfg, + void __iomem *addr) { struct dpu_hw_wb *c; - const struct dpu_wb_cfg *cfg; - if (!addr || !m) + if (!addr) return ERR_PTR(-EINVAL); c = kzalloc(sizeof(*c), GFP_KERNEL); if (!c) return ERR_PTR(-ENOMEM); - cfg = _wb_offset(idx, m, addr, &c->hw); - if (IS_ERR(cfg)) { - WARN(1, "Unable to find wb idx=%d\n", idx); - kfree(c); - return ERR_PTR(-EINVAL); - } + c->hw.blk_addr = addr + cfg->base; + c->hw.log_mask = DPU_DBG_MASK_WB; /* Assign ops */ - c->mdp = &m->mdp[0]; - c->idx = idx; + c->idx = cfg->id; c->caps = cfg; _setup_wb_ops(&c->ops, c->caps->features); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h index 3ff5a48541e2ecfeab5915b2e9e2a562e5f7ab8b..2d7db2efa3d0181fa6ba5247d55ce2ac566ed9b3 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.h @@ -21,28 +21,6 @@ struct dpu_hw_wb_cfg { struct drm_rect crop; }; -/** - * enum CDP preload ahead address size - */ -enum { - DPU_WB_CDP_PRELOAD_AHEAD_32, - DPU_WB_CDP_PRELOAD_AHEAD_64 -}; - -/** - * struct dpu_hw_wb_qos_cfg : Writeback pipe QoS configuration - * @danger_lut: LUT for generate danger level based on fill level - * @safe_lut: LUT for generate safe level based on fill level - * @creq_lut: LUT for generate creq level based on fill level - * @danger_safe_en: enable danger safe generation - */ -struct dpu_hw_wb_qos_cfg { - u32 danger_lut; - u32 safe_lut; - u64 creq_lut; - bool danger_safe_en; -}; - /** * * struct dpu_hw_wb_ops : Interface to the wb hw driver functions @@ -64,27 +42,25 @@ struct dpu_hw_wb_ops { struct dpu_hw_wb_cfg *wb); void (*setup_qos_lut)(struct dpu_hw_wb *ctx, - struct dpu_hw_wb_qos_cfg *cfg); + struct dpu_hw_qos_cfg *cfg); void (*setup_cdp)(struct dpu_hw_wb *ctx, - struct dpu_hw_cdp_cfg *cfg); + const struct dpu_format *fmt, + bool enable); void (*bind_pingpong_blk)(struct dpu_hw_wb *ctx, - bool enable, const enum dpu_pingpong pp); + const enum dpu_pingpong pp); }; /** * struct dpu_hw_wb : WB driver object * @hw: block hardware details - * @mdp: pointer to associated mdp portion of the catalog * @idx: hardware index number within type * @wb_hw_caps: hardware capabilities * @ops: function pointers - * @hw_mdp: MDP top level hardware block */ struct dpu_hw_wb { struct dpu_hw_blk_reg_map hw; - const struct dpu_mdp_cfg *mdp; /* wb path */ int idx; @@ -92,19 +68,16 @@ struct dpu_hw_wb { /* ops */ struct dpu_hw_wb_ops ops; - - struct dpu_hw_mdp *hw_mdp; }; /** - * dpu_hw_wb_init(): Initializes and return writeback hw driver object. - * @idx: wb_path index for which driver object is required + * dpu_hw_wb_init() - Initializes the writeback hw driver object. + * @cfg: wb_path catalog entry for which driver object is required * @addr: mapped register io address of MDP - * @m : pointer to mdss catalog data + * Return: Error code or allocated dpu_hw_wb context */ -struct dpu_hw_wb *dpu_hw_wb_init(enum dpu_wb idx, - void __iomem *addr, - const struct dpu_mdss_cfg *m); +struct dpu_hw_wb *dpu_hw_wb_init(const struct dpu_wb_cfg *cfg, + void __iomem *addr); /** * dpu_hw_wb_destroy(): Destroy writeback hw driver object. diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hwio.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hwio.h index feb9a729844a3c1488419e360382e48ac3f16e1b..5acd5683d25a4021950a80d341d732ac6d7fc856 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hwio.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hwio.h @@ -21,9 +21,6 @@ #define HIST_INTR_EN 0x01c #define HIST_INTR_STATUS 0x020 #define HIST_INTR_CLEAR 0x024 -#define INTF_INTR_EN 0x1C0 -#define INTF_INTR_STATUS 0x1C4 -#define INTF_INTR_CLEAR 0x1C8 #define SPLIT_DISPLAY_EN 0x2F4 #define SPLIT_DISPLAY_UPPER_PIPE_CTRL 0x2F8 #define DSPP_IGC_COLOR0_RAM_LUTN 0x300 diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c index 0e7a68714e9e16f158c8c84eebd99a113afc3228..aa8499de1b9f3179027e96775f940794059fe181 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c @@ -57,8 +57,8 @@ static void _dpu_kms_mmu_destroy(struct dpu_kms *dpu_kms); static int _dpu_danger_signal_status(struct seq_file *s, bool danger_status) { - struct dpu_kms *kms = (struct dpu_kms *)s->private; struct dpu_danger_safe_status status; + struct dpu_kms *kms = s->private; int i; if (!kms->hw_mdp) { @@ -535,15 +535,23 @@ static int _dpu_kms_initialize_dsi(struct drm_device *dev, !msm_dsi_is_master_dsi(priv->dsi[i])) continue; - encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_DSI); + memset(&info, 0, sizeof(info)); + info.intf_type = INTF_DSI; + + info.h_tile_instance[info.num_of_h_tiles++] = i; + if (msm_dsi_is_bonded_dsi(priv->dsi[i])) + info.h_tile_instance[info.num_of_h_tiles++] = other; + + info.is_cmd_mode = msm_dsi_is_cmd_mode(priv->dsi[i]); + + info.dsc = msm_dsi_get_dsc_config(priv->dsi[i]); + + encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_DSI, &info); if (IS_ERR(encoder)) { DPU_ERROR("encoder init failed for dsi display\n"); return PTR_ERR(encoder); } - memset(&info, 0, sizeof(info)); - info.intf_type = encoder->encoder_type; - rc = msm_dsi_modeset_init(priv->dsi[i], dev, encoder); if (rc) { DPU_ERROR("modeset_init failed for dsi[%d], rc = %d\n", @@ -551,11 +559,6 @@ static int _dpu_kms_initialize_dsi(struct drm_device *dev, break; } - info.h_tile_instance[info.num_of_h_tiles++] = i; - info.is_cmd_mode = msm_dsi_is_cmd_mode(priv->dsi[i]); - - info.dsc = msm_dsi_get_dsc_config(priv->dsi[i]); - if (msm_dsi_is_bonded_dsi(priv->dsi[i]) && priv->dsi[other]) { rc = msm_dsi_modeset_init(priv->dsi[other], dev, encoder); if (rc) { @@ -563,14 +566,7 @@ static int _dpu_kms_initialize_dsi(struct drm_device *dev, other, rc); break; } - - info.h_tile_instance[info.num_of_h_tiles++] = other; } - - rc = dpu_encoder_setup(dev, encoder, &info); - if (rc) - DPU_ERROR("failed to setup DPU encoder %d: rc:%d\n", - encoder->base.id, rc); } return rc; @@ -589,67 +585,86 @@ static int _dpu_kms_initialize_displayport(struct drm_device *dev, if (!priv->dp[i]) continue; - encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_TMDS); + memset(&info, 0, sizeof(info)); + info.num_of_h_tiles = 1; + info.h_tile_instance[0] = i; + info.intf_type = INTF_DP; + + encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_TMDS, &info); if (IS_ERR(encoder)) { DPU_ERROR("encoder init failed for dsi display\n"); return PTR_ERR(encoder); } - memset(&info, 0, sizeof(info)); rc = msm_dp_modeset_init(priv->dp[i], dev, encoder); if (rc) { DPU_ERROR("modeset_init failed for DP, rc = %d\n", rc); drm_encoder_cleanup(encoder); return rc; } - - info.num_of_h_tiles = 1; - info.h_tile_instance[0] = i; - info.intf_type = encoder->encoder_type; - rc = dpu_encoder_setup(dev, encoder, &info); - if (rc) { - DPU_ERROR("failed to setup DPU encoder %d: rc:%d\n", - encoder->base.id, rc); - return rc; - } } return 0; } -static int _dpu_kms_initialize_writeback(struct drm_device *dev, - struct msm_drm_private *priv, struct dpu_kms *dpu_kms, - const u32 *wb_formats, int n_formats) +static int _dpu_kms_initialize_hdmi(struct drm_device *dev, + struct msm_drm_private *priv, + struct dpu_kms *dpu_kms) { struct drm_encoder *encoder = NULL; struct msm_display_info info; int rc; - encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_VIRTUAL); + if (!priv->hdmi) + return 0; + + memset(&info, 0, sizeof(info)); + info.num_of_h_tiles = 1; + info.h_tile_instance[0] = 0; + info.intf_type = INTF_HDMI; + + encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_TMDS, &info); if (IS_ERR(encoder)) { - DPU_ERROR("encoder init failed for dsi display\n"); + DPU_ERROR("encoder init failed for HDMI display\n"); return PTR_ERR(encoder); } - memset(&info, 0, sizeof(info)); - - rc = dpu_writeback_init(dev, encoder, wb_formats, - n_formats); + rc = msm_hdmi_modeset_init(priv->hdmi, dev, encoder); if (rc) { - DPU_ERROR("dpu_writeback_init, rc = %d\n", rc); + DPU_ERROR("modeset_init failed for DP, rc = %d\n", rc); drm_encoder_cleanup(encoder); return rc; } + return 0; +} + +static int _dpu_kms_initialize_writeback(struct drm_device *dev, + struct msm_drm_private *priv, struct dpu_kms *dpu_kms, + const u32 *wb_formats, int n_formats) +{ + struct drm_encoder *encoder = NULL; + struct msm_display_info info; + int rc; + + memset(&info, 0, sizeof(info)); + info.num_of_h_tiles = 1; /* use only WB idx 2 instance for DPU */ info.h_tile_instance[0] = WB_2; - info.intf_type = encoder->encoder_type; + info.intf_type = INTF_WB; - rc = dpu_encoder_setup(dev, encoder, &info); + encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_VIRTUAL, &info); + if (IS_ERR(encoder)) { + DPU_ERROR("encoder init failed for dsi display\n"); + return PTR_ERR(encoder); + } + + rc = dpu_writeback_init(dev, encoder, wb_formats, + n_formats); if (rc) { - DPU_ERROR("failed to setup DPU encoder %d: rc:%d\n", - encoder->base.id, rc); + DPU_ERROR("dpu_writeback_init, rc = %d\n", rc); + drm_encoder_cleanup(encoder); return rc; } @@ -683,6 +698,12 @@ static int _dpu_kms_setup_displays(struct drm_device *dev, return rc; } + rc = _dpu_kms_initialize_hdmi(dev, priv, dpu_kms); + if (rc) { + DPU_ERROR("initialize HDMI failed, rc = %d\n", rc); + return rc; + } + /* Since WB isn't a driver check the catalog before initializing */ if (dpu_kms->catalog->wb_count) { for (i = 0; i < dpu_kms->catalog->wb_count; i++) { @@ -979,13 +1000,13 @@ static int _dpu_kms_mmu_init(struct dpu_kms *dpu_kms) return 0; } -u64 dpu_kms_get_clk_rate(struct dpu_kms *dpu_kms, char *clock_name) +unsigned long dpu_kms_get_clk_rate(struct dpu_kms *dpu_kms, char *clock_name) { struct clk *clk; clk = msm_clk_bulk_get_clock(dpu_kms->clocks, dpu_kms->num_clocks, clock_name); if (!clk) - return -EINVAL; + return 0; return clk_get_rate(clk); } @@ -1005,6 +1026,9 @@ static int dpu_kms_hw_init(struct msm_kms *kms) dpu_kms = to_dpu_kms(kms); dev = dpu_kms->dev; + dev->mode_config.cursor_width = 512; + dev->mode_config.cursor_height = 512; + rc = dpu_kms_global_obj_init(dpu_kms); if (rc) return rc; @@ -1033,12 +1057,6 @@ static int dpu_kms_hw_init(struct msm_kms *kms) DPU_DEBUG("VBIF NRT is not defined"); } - dpu_kms->reg_dma = msm_ioremap_quiet(dpu_kms->pdev, "regdma"); - if (IS_ERR(dpu_kms->reg_dma)) { - dpu_kms->reg_dma = NULL; - DPU_DEBUG("REG_DMA is not defined"); - } - dpu_kms_parse_data_bus_icc_path(dpu_kms); rc = pm_runtime_resume_and_get(&dpu_kms->pdev->dev); @@ -1084,16 +1102,17 @@ static int dpu_kms_hw_init(struct msm_kms *kms) } for (i = 0; i < dpu_kms->catalog->vbif_count; i++) { - u32 vbif_idx = dpu_kms->catalog->vbif[i].id; - - dpu_kms->hw_vbif[vbif_idx] = dpu_hw_vbif_init(vbif_idx, - dpu_kms->vbif[vbif_idx], dpu_kms->catalog); - if (IS_ERR(dpu_kms->hw_vbif[vbif_idx])) { - rc = PTR_ERR(dpu_kms->hw_vbif[vbif_idx]); - DPU_ERROR("failed to init vbif %d: %d\n", vbif_idx, rc); - dpu_kms->hw_vbif[vbif_idx] = NULL; + struct dpu_hw_vbif *hw; + const struct dpu_vbif_cfg *vbif = &dpu_kms->catalog->vbif[i]; + + hw = dpu_hw_vbif_init(vbif, dpu_kms->vbif[vbif->id]); + if (IS_ERR(hw)) { + rc = PTR_ERR(hw); + DPU_ERROR("failed to init vbif %d: %d\n", vbif->id, rc); goto power_error; } + + dpu_kms->hw_vbif[vbif->id] = hw; } rc = dpu_core_perf_init(&dpu_kms->perf, dev, dpu_kms->catalog, @@ -1286,6 +1305,8 @@ static const struct of_device_id dpu_dt_match[] = { { .compatible = "qcom,sc8180x-dpu", .data = &dpu_sc8180x_cfg, }, { .compatible = "qcom,sc8280xp-dpu", .data = &dpu_sc8280xp_cfg, }, { .compatible = "qcom,sm6115-dpu", .data = &dpu_sm6115_cfg, }, + { .compatible = "qcom,sm6350-dpu", .data = &dpu_sm6350_cfg, }, + { .compatible = "qcom,sm6375-dpu", .data = &dpu_sm6375_cfg, }, { .compatible = "qcom,sm8150-dpu", .data = &dpu_sm8150_cfg, }, { .compatible = "qcom,sm8250-dpu", .data = &dpu_sm8250_cfg, }, { .compatible = "qcom,sm8350-dpu", .data = &dpu_sm8350_cfg, }, diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h index aca39a4689f489e1a7d603e7d2f646cd7ae5664a..f3bdd4f11108a1752991554319bdb7f87ffe6bf4 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h @@ -63,15 +63,13 @@ #define ktime_compare_safe(A, B) \ ktime_compare(ktime_sub((A), (B)), ktime_set(0, 0)) -#define DPU_NAME_SIZE 12 - struct dpu_kms { struct msm_kms base; struct drm_device *dev; const struct dpu_mdss_cfg *catalog; /* io/register spaces: */ - void __iomem *mmio, *vbif[VBIF_MAX], *reg_dma; + void __iomem *mmio, *vbif[VBIF_MAX]; struct regulator *vdd; struct regulator *mmagic; @@ -118,6 +116,10 @@ struct vsync_info { u32 line_count; }; +#define DPU_ENC_WR_PTR_START_TIMEOUT_US 20000 + +#define DPU_ENC_MAX_POLL_TIMEOUT_US 2000 + #define to_dpu_kms(x) container_of(x, struct dpu_kms, base) #define to_dpu_global_state(x) container_of(x, struct dpu_global_state, base) @@ -201,6 +203,6 @@ void dpu_disable_vblank(struct msm_kms *kms, struct drm_crtc *crtc); * * Return: current clock rate */ -u64 dpu_kms_get_clk_rate(struct dpu_kms *dpu_kms, char *clock_name); +unsigned long dpu_kms_get_clk_rate(struct dpu_kms *dpu_kms, char *clock_name); #endif /* __dpu_kms_H__ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c index 14b5cfe30611396d5ceee44e722b676d05697c33..c2aaaded07ed6a9808eb4b93fe945715203f0cf1 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c @@ -42,8 +42,6 @@ #define SHARP_SMOOTH_THR_DEFAULT 8 #define SHARP_NOISE_THR_DEFAULT 2 -#define DPU_NAME_SIZE 12 - #define DPU_PLANE_COLOR_FILL_FLAG BIT(31) #define DPU_ZPOS_MAX 255 @@ -70,20 +68,6 @@ static const uint32_t qcom_compressed_supported_formats[] = { DRM_FORMAT_P010, }; -/** - * enum dpu_plane_qos - Different qos configurations for each pipe - * - * @DPU_PLANE_QOS_VBLANK_CTRL: Setup VBLANK qos for the pipe. - * @DPU_PLANE_QOS_VBLANK_AMORTIZE: Enables Amortization within pipe. - * this configuration is mutually exclusive from VBLANK_CTRL. - * @DPU_PLANE_QOS_PANIC_CTRL: Setup panic for the pipe. - */ -enum dpu_plane_qos { - DPU_PLANE_QOS_VBLANK_CTRL = BIT(0), - DPU_PLANE_QOS_VBLANK_AMORTIZE = BIT(1), - DPU_PLANE_QOS_PANIC_CTRL = BIT(2), -}; - /* * struct dpu_plane - local dpu plane structure * @aspace: address space pointer @@ -204,12 +188,14 @@ static u64 _dpu_plane_calc_clk(const struct drm_display_mode *mode, * _dpu_plane_calc_fill_level - calculate fill level of the given source format * @plane: Pointer to drm plane * @pipe: Pointer to software pipe + * @lut_usage: LUT usecase * @fmt: Pointer to source buffer format * @src_width: width of source buffer * Return: fill level corresponding to the source buffer/format or 0 if error */ static int _dpu_plane_calc_fill_level(struct drm_plane *plane, struct dpu_sw_pipe *pipe, + enum dpu_qos_lut_usage lut_usage, const struct dpu_format *fmt, u32 src_width) { struct dpu_plane *pdpu; @@ -221,6 +207,9 @@ static int _dpu_plane_calc_fill_level(struct drm_plane *plane, return 0; } + if (lut_usage == DPU_QOS_LUT_USAGE_NRT) + return 0; + pdpu = to_dpu_plane(plane); fixed_buff_size = pdpu->catalog->caps->pixel_ram_size; @@ -266,83 +255,58 @@ static void _dpu_plane_set_qos_lut(struct drm_plane *plane, const struct dpu_format *fmt, struct dpu_sw_pipe_cfg *pipe_cfg) { struct dpu_plane *pdpu = to_dpu_plane(plane); - u64 qos_lut; - u32 total_fl = 0, lut_usage; + struct dpu_hw_qos_cfg cfg; + u32 total_fl, lut_usage; if (!pdpu->is_rt_pipe) { lut_usage = DPU_QOS_LUT_USAGE_NRT; } else { - total_fl = _dpu_plane_calc_fill_level(plane, pipe, fmt, - drm_rect_width(&pipe_cfg->src_rect)); - if (fmt && DPU_FORMAT_IS_LINEAR(fmt)) lut_usage = DPU_QOS_LUT_USAGE_LINEAR; else lut_usage = DPU_QOS_LUT_USAGE_MACROTILE; } - qos_lut = _dpu_hw_get_qos_lut( - &pdpu->catalog->perf->qos_lut_tbl[lut_usage], total_fl); + total_fl = _dpu_plane_calc_fill_level(plane, pipe, lut_usage, fmt, + drm_rect_width(&pipe_cfg->src_rect)); + + cfg.creq_lut = _dpu_hw_get_qos_lut(&pdpu->catalog->perf->qos_lut_tbl[lut_usage], total_fl); + cfg.danger_lut = pdpu->catalog->perf->danger_lut_tbl[lut_usage]; + cfg.safe_lut = pdpu->catalog->perf->safe_lut_tbl[lut_usage]; + + if (pipe->sspp->idx != SSPP_CURSOR0 && + pipe->sspp->idx != SSPP_CURSOR1 && + pdpu->is_rt_pipe) + cfg.danger_safe_en = true; + + DPU_DEBUG_PLANE(pdpu, "pnum:%d ds:%d is_rt:%d\n", + pdpu->pipe - SSPP_VIG0, + cfg.danger_safe_en, + pdpu->is_rt_pipe); trace_dpu_perf_set_qos_luts(pipe->sspp->idx - SSPP_VIG0, (fmt) ? fmt->base.pixel_format : 0, - pdpu->is_rt_pipe, total_fl, qos_lut, lut_usage); + pdpu->is_rt_pipe, total_fl, cfg.creq_lut, lut_usage); DPU_DEBUG_PLANE(pdpu, "pnum:%d fmt: %4.4s rt:%d fl:%u lut:0x%llx\n", pdpu->pipe - SSPP_VIG0, fmt ? (char *)&fmt->base.pixel_format : NULL, - pdpu->is_rt_pipe, total_fl, qos_lut); - - pipe->sspp->ops.setup_creq_lut(pipe->sspp, qos_lut); -} - -/** - * _dpu_plane_set_danger_lut - set danger/safe LUT of the given plane - * @plane: Pointer to drm plane - * @pipe: Pointer to software pipe - * @fmt: Pointer to source buffer format - */ -static void _dpu_plane_set_danger_lut(struct drm_plane *plane, - struct dpu_sw_pipe *pipe, - const struct dpu_format *fmt) -{ - struct dpu_plane *pdpu = to_dpu_plane(plane); - u32 danger_lut, safe_lut; - - if (!pdpu->is_rt_pipe) { - danger_lut = pdpu->catalog->perf->danger_lut_tbl - [DPU_QOS_LUT_USAGE_NRT]; - safe_lut = pdpu->catalog->perf->safe_lut_tbl - [DPU_QOS_LUT_USAGE_NRT]; - } else { - if (fmt && DPU_FORMAT_IS_LINEAR(fmt)) { - danger_lut = pdpu->catalog->perf->danger_lut_tbl - [DPU_QOS_LUT_USAGE_LINEAR]; - safe_lut = pdpu->catalog->perf->safe_lut_tbl - [DPU_QOS_LUT_USAGE_LINEAR]; - } else { - danger_lut = pdpu->catalog->perf->danger_lut_tbl - [DPU_QOS_LUT_USAGE_MACROTILE]; - safe_lut = pdpu->catalog->perf->safe_lut_tbl - [DPU_QOS_LUT_USAGE_MACROTILE]; - } - } + pdpu->is_rt_pipe, total_fl, cfg.creq_lut); trace_dpu_perf_set_danger_luts(pdpu->pipe - SSPP_VIG0, (fmt) ? fmt->base.pixel_format : 0, (fmt) ? fmt->fetch_mode : 0, - danger_lut, - safe_lut); + cfg.danger_lut, + cfg.safe_lut); DPU_DEBUG_PLANE(pdpu, "pnum:%d fmt: %4.4s mode:%d luts[0x%x, 0x%x]\n", pdpu->pipe - SSPP_VIG0, fmt ? (char *)&fmt->base.pixel_format : NULL, fmt ? fmt->fetch_mode : -1, - danger_lut, - safe_lut); + cfg.danger_lut, + cfg.safe_lut); - pipe->sspp->ops.setup_danger_safe_lut(pipe->sspp, - danger_lut, safe_lut); + pipe->sspp->ops.setup_qos_lut(pipe->sspp, &cfg); } /** @@ -350,48 +314,23 @@ static void _dpu_plane_set_danger_lut(struct drm_plane *plane, * @plane: Pointer to drm plane * @pipe: Pointer to software pipe * @enable: true to enable QoS control - * @flags: QoS control mode (enum dpu_plane_qos) */ static void _dpu_plane_set_qos_ctrl(struct drm_plane *plane, struct dpu_sw_pipe *pipe, - bool enable, u32 flags) + bool enable) { struct dpu_plane *pdpu = to_dpu_plane(plane); - struct dpu_hw_pipe_qos_cfg pipe_qos_cfg; - - memset(&pipe_qos_cfg, 0, sizeof(pipe_qos_cfg)); - - if (flags & DPU_PLANE_QOS_VBLANK_CTRL) { - pipe_qos_cfg.creq_vblank = pipe->sspp->cap->sblk->creq_vblank; - pipe_qos_cfg.danger_vblank = - pipe->sspp->cap->sblk->danger_vblank; - pipe_qos_cfg.vblank_en = enable; - } - if (flags & DPU_PLANE_QOS_VBLANK_AMORTIZE) { - /* this feature overrules previous VBLANK_CTRL */ - pipe_qos_cfg.vblank_en = false; - pipe_qos_cfg.creq_vblank = 0; /* clear vblank bits */ - } - - if (flags & DPU_PLANE_QOS_PANIC_CTRL) - pipe_qos_cfg.danger_safe_en = enable; - - if (!pdpu->is_rt_pipe) { - pipe_qos_cfg.vblank_en = false; - pipe_qos_cfg.danger_safe_en = false; - } + if (!pdpu->is_rt_pipe) + enable = false; - DPU_DEBUG_PLANE(pdpu, "pnum:%d ds:%d vb:%d pri[0x%x, 0x%x] is_rt:%d\n", + DPU_DEBUG_PLANE(pdpu, "pnum:%d ds:%d is_rt:%d\n", pdpu->pipe - SSPP_VIG0, - pipe_qos_cfg.danger_safe_en, - pipe_qos_cfg.vblank_en, - pipe_qos_cfg.creq_vblank, - pipe_qos_cfg.danger_vblank, + enable, pdpu->is_rt_pipe); pipe->sspp->ops.setup_qos_ctrl(pipe->sspp, - &pipe_qos_cfg); + enable); } /** @@ -1079,10 +1018,10 @@ static void dpu_plane_sspp_update_pipe(struct drm_plane *plane, pipe->sspp->ops.setup_sourceaddress(pipe, layout); } - _dpu_plane_set_qos_ctrl(plane, pipe, false, DPU_PLANE_QOS_PANIC_CTRL); - /* override for color fill */ if (pdpu->color_fill & DPU_PLANE_COLOR_FILL_FLAG) { + _dpu_plane_set_qos_ctrl(plane, pipe, false); + /* skip remaining processing on color fill */ return; } @@ -1116,30 +1055,18 @@ static void dpu_plane_sspp_update_pipe(struct drm_plane *plane, pipe->sspp->ops.setup_format(pipe, fmt, src_flags); if (pipe->sspp->ops.setup_cdp) { - struct dpu_hw_cdp_cfg cdp_cfg; - - memset(&cdp_cfg, 0, sizeof(struct dpu_hw_cdp_cfg)); + const struct dpu_perf_cfg *perf = pdpu->catalog->perf; - cdp_cfg.enable = pdpu->catalog->perf->cdp_cfg - [DPU_PERF_CDP_USAGE_RT].rd_enable; - cdp_cfg.ubwc_meta_enable = - DPU_FORMAT_IS_UBWC(fmt); - cdp_cfg.tile_amortize_enable = - DPU_FORMAT_IS_UBWC(fmt) || - DPU_FORMAT_IS_TILE(fmt); - cdp_cfg.preload_ahead = DPU_SSPP_CDP_PRELOAD_AHEAD_64; - - pipe->sspp->ops.setup_cdp(pipe, &cdp_cfg); + pipe->sspp->ops.setup_cdp(pipe, fmt, + perf->cdp_cfg[DPU_PERF_CDP_USAGE_RT].rd_enable); } } _dpu_plane_set_qos_lut(plane, pipe, fmt, pipe_cfg); - _dpu_plane_set_danger_lut(plane, pipe, fmt); - if (plane->type != DRM_PLANE_TYPE_CURSOR) { - _dpu_plane_set_qos_ctrl(plane, pipe, true, DPU_PLANE_QOS_PANIC_CTRL); + if (pipe->sspp->idx != SSPP_CURSOR0 && + pipe->sspp->idx != SSPP_CURSOR1) _dpu_plane_set_ot_limit(plane, pipe, pipe_cfg, frame_rate); - } if (pstate->needs_qos_remap) _dpu_plane_set_qos_remap(plane, pipe); @@ -1254,10 +1181,10 @@ static void dpu_plane_destroy(struct drm_plane *plane) if (pdpu) { pstate = to_dpu_plane_state(plane->state); - _dpu_plane_set_qos_ctrl(plane, &pstate->pipe, false, DPU_PLANE_QOS_PANIC_CTRL); + _dpu_plane_set_qos_ctrl(plane, &pstate->pipe, false); if (pstate->r_pipe.sspp) - _dpu_plane_set_qos_ctrl(plane, &pstate->r_pipe, false, DPU_PLANE_QOS_PANIC_CTRL); + _dpu_plane_set_qos_ctrl(plane, &pstate->r_pipe, false); mutex_destroy(&pdpu->lock); @@ -1414,9 +1341,9 @@ void dpu_plane_danger_signal_ctrl(struct drm_plane *plane, bool enable) return; pm_runtime_get_sync(&dpu_kms->pdev->dev); - _dpu_plane_set_qos_ctrl(plane, &pstate->pipe, enable, DPU_PLANE_QOS_PANIC_CTRL); + _dpu_plane_set_qos_ctrl(plane, &pstate->pipe, enable); if (pstate->r_pipe.sspp) - _dpu_plane_set_qos_ctrl(plane, &pstate->r_pipe, enable, DPU_PLANE_QOS_PANIC_CTRL); + _dpu_plane_set_qos_ctrl(plane, &pstate->r_pipe, enable); pm_runtime_put_sync(&dpu_kms->pdev->dev); } #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c index f4dda88a73f7d376f0be010a2e30978eff56c5bd..471842bbb95077d2c70bb138c50cf37cc306da33 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c @@ -1,6 +1,7 @@ // SPDX-License-Identifier: GPL-2.0-only /* * Copyright (c) 2016-2018, The Linux Foundation. All rights reserved. + * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved. */ #define pr_fmt(fmt) "[drm:%s] " fmt, __func__ @@ -117,16 +118,7 @@ int dpu_rm_init(struct dpu_rm *rm, struct dpu_hw_mixer *hw; const struct dpu_lm_cfg *lm = &cat->mixer[i]; - if (lm->pingpong == PINGPONG_MAX) { - DPU_DEBUG("skip mixer %d without pingpong\n", lm->id); - continue; - } - - if (lm->id < LM_0 || lm->id >= LM_MAX) { - DPU_ERROR("skip mixer %d with invalid id\n", lm->id); - continue; - } - hw = dpu_hw_lm_init(lm->id, mmio, cat); + hw = dpu_hw_lm_init(lm, mmio); if (IS_ERR(hw)) { rc = PTR_ERR(hw); DPU_ERROR("failed lm object creation: err %d\n", rc); @@ -139,11 +131,7 @@ int dpu_rm_init(struct dpu_rm *rm, struct dpu_hw_merge_3d *hw; const struct dpu_merge_3d_cfg *merge_3d = &cat->merge_3d[i]; - if (merge_3d->id < MERGE_3D_0 || merge_3d->id >= MERGE_3D_MAX) { - DPU_ERROR("skip merge_3d %d with invalid id\n", merge_3d->id); - continue; - } - hw = dpu_hw_merge_3d_init(merge_3d->id, mmio, cat); + hw = dpu_hw_merge_3d_init(merge_3d, mmio); if (IS_ERR(hw)) { rc = PTR_ERR(hw); DPU_ERROR("failed merge_3d object creation: err %d\n", @@ -157,11 +145,7 @@ int dpu_rm_init(struct dpu_rm *rm, struct dpu_hw_pingpong *hw; const struct dpu_pingpong_cfg *pp = &cat->pingpong[i]; - if (pp->id < PINGPONG_0 || pp->id >= PINGPONG_MAX) { - DPU_ERROR("skip pingpong %d with invalid id\n", pp->id); - continue; - } - hw = dpu_hw_pingpong_init(pp->id, mmio, cat); + hw = dpu_hw_pingpong_init(pp, mmio); if (IS_ERR(hw)) { rc = PTR_ERR(hw); DPU_ERROR("failed pingpong object creation: err %d\n", @@ -177,15 +161,7 @@ int dpu_rm_init(struct dpu_rm *rm, struct dpu_hw_intf *hw; const struct dpu_intf_cfg *intf = &cat->intf[i]; - if (intf->type == INTF_NONE) { - DPU_DEBUG("skip intf %d with type none\n", i); - continue; - } - if (intf->id < INTF_0 || intf->id >= INTF_MAX) { - DPU_ERROR("skip intf %d with invalid id\n", intf->id); - continue; - } - hw = dpu_hw_intf_init(intf->id, mmio, cat); + hw = dpu_hw_intf_init(intf, mmio); if (IS_ERR(hw)) { rc = PTR_ERR(hw); DPU_ERROR("failed intf object creation: err %d\n", rc); @@ -198,12 +174,7 @@ int dpu_rm_init(struct dpu_rm *rm, struct dpu_hw_wb *hw; const struct dpu_wb_cfg *wb = &cat->wb[i]; - if (wb->id < WB_0 || wb->id >= WB_MAX) { - DPU_ERROR("skip intf %d with invalid id\n", wb->id); - continue; - } - - hw = dpu_hw_wb_init(wb->id, mmio, cat); + hw = dpu_hw_wb_init(wb, mmio); if (IS_ERR(hw)) { rc = PTR_ERR(hw); DPU_ERROR("failed wb object creation: err %d\n", rc); @@ -216,11 +187,7 @@ int dpu_rm_init(struct dpu_rm *rm, struct dpu_hw_ctl *hw; const struct dpu_ctl_cfg *ctl = &cat->ctl[i]; - if (ctl->id < CTL_0 || ctl->id >= CTL_MAX) { - DPU_ERROR("skip ctl %d with invalid id\n", ctl->id); - continue; - } - hw = dpu_hw_ctl_init(ctl->id, mmio, cat); + hw = dpu_hw_ctl_init(ctl, mmio, cat->mixer_count, cat->mixer); if (IS_ERR(hw)) { rc = PTR_ERR(hw); DPU_ERROR("failed ctl object creation: err %d\n", rc); @@ -233,11 +200,7 @@ int dpu_rm_init(struct dpu_rm *rm, struct dpu_hw_dspp *hw; const struct dpu_dspp_cfg *dspp = &cat->dspp[i]; - if (dspp->id < DSPP_0 || dspp->id >= DSPP_MAX) { - DPU_ERROR("skip dspp %d with invalid id\n", dspp->id); - continue; - } - hw = dpu_hw_dspp_init(dspp->id, mmio, cat); + hw = dpu_hw_dspp_init(dspp, mmio); if (IS_ERR(hw)) { rc = PTR_ERR(hw); DPU_ERROR("failed dspp object creation: err %d\n", rc); @@ -250,8 +213,12 @@ int dpu_rm_init(struct dpu_rm *rm, struct dpu_hw_dsc *hw; const struct dpu_dsc_cfg *dsc = &cat->dsc[i]; - hw = dpu_hw_dsc_init(dsc->id, mmio, cat); - if (IS_ERR_OR_NULL(hw)) { + if (test_bit(DPU_DSC_HW_REV_1_2, &dsc->features)) + hw = dpu_hw_dsc_init_1_2(dsc, mmio); + else + hw = dpu_hw_dsc_init(dsc, mmio); + + if (IS_ERR(hw)) { rc = PTR_ERR(hw); DPU_ERROR("failed dsc object creation: err %d\n", rc); goto fail; @@ -263,12 +230,7 @@ int dpu_rm_init(struct dpu_rm *rm, struct dpu_hw_sspp *hw; const struct dpu_sspp_cfg *sspp = &cat->sspp[i]; - if (sspp->id < SSPP_NONE || sspp->id >= SSPP_MAX) { - DPU_ERROR("skip intf %d with invalid id\n", sspp->id); - continue; - } - - hw = dpu_hw_sspp_init(sspp->id, mmio, cat); + hw = dpu_hw_sspp_init(sspp, mmio, cat->ubwc); if (IS_ERR(hw)) { rc = PTR_ERR(hw); DPU_ERROR("failed sspp object creation: err %d\n", rc); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h index 0ad148cc2fb8e90bb6a9dc2f40142be091038105..1a92d21094f4188edaa3d5a3aa68378781c94409 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h @@ -871,6 +871,20 @@ TRACE_EVENT(dpu_pp_connect_ext_te, TP_printk("pp:%d cfg:%u", __entry->pp, __entry->cfg) ); +TRACE_EVENT(dpu_intf_connect_ext_te, + TP_PROTO(enum dpu_intf intf, u32 cfg), + TP_ARGS(intf, cfg), + TP_STRUCT__entry( + __field( enum dpu_intf, intf ) + __field( u32, cfg ) + ), + TP_fast_assign( + __entry->intf = intf; + __entry->cfg = cfg; + ), + TP_printk("intf:%d cfg:%u", __entry->intf, __entry->cfg) +); + TRACE_EVENT(dpu_core_irq_register_callback, TP_PROTO(int irq_idx, void *callback), TP_ARGS(irq_idx, callback), diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c index 2eec2d78f32a01e22c146af9aafcee10da14bd5d..694d5434133741cfd1f30ceec20ec79ca74177a8 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c @@ -103,6 +103,87 @@ static const struct mdp5_cfg_hw msm8x74v1_config = { .max_clk = 200000000, }; +static const struct mdp5_cfg_hw msm8x26_config = { + .name = "msm8x26", + .mdp = { + .count = 1, + .caps = MDP_CAP_SMP | + 0, + }, + .smp = { + .mmb_count = 7, + .mmb_size = 4096, + .clients = { + [SSPP_VIG0] = 1, + [SSPP_DMA0] = 4, + [SSPP_RGB0] = 7, + }, + }, + .ctl = { + .count = 2, + .base = { 0x00500, 0x00600 }, + .flush_hw_mask = 0x0003ffff, + }, + .pipe_vig = { + .count = 1, + .base = { 0x01100 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | + MDP_PIPE_CAP_CSC | + 0, + }, + .pipe_rgb = { + .count = 1, + .base = { 0x01d00 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | + 0, + }, + .pipe_dma = { + .count = 1, + .base = { 0x02900 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + 0, + }, + .lm = { + .count = 2, + .base = { 0x03100, 0x03d00 }, + .instances = { + { .id = 0, .pp = 0, .dspp = 0, + .caps = MDP_LM_CAP_DISPLAY, }, + { .id = 1, .pp = -1, .dspp = -1, + .caps = MDP_LM_CAP_WB }, + }, + .nb_stages = 2, + .max_width = 2048, + .max_height = 0xFFFF, + }, + .dspp = { + .count = 1, + .base = { 0x04500 }, + }, + .pp = { + .count = 1, + .base = { 0x21a00 }, + }, + .intf = { + .base = { 0x00000, 0x21200 }, + .connect = { + [0] = INTF_DISABLED, + [1] = INTF_DSI, + }, + }, + .perf = { + .ab_inefficiency = 100, + .ib_inefficiency = 200, + .clk_inefficiency = 125 + }, + .max_clk = 200000000, +}; + static const struct mdp5_cfg_hw msm8x74v2_config = { .name = "msm8x74", .mdp = { @@ -1236,6 +1317,7 @@ static const struct mdp5_cfg_hw sdm660_config = { static const struct mdp5_cfg_handler cfg_handlers_v1[] = { { .revision = 0, .config = { .hw = &msm8x74v1_config } }, + { .revision = 1, .config = { .hw = &msm8x26_config } }, { .revision = 2, .config = { .hw = &msm8x74v2_config } }, { .revision = 3, .config = { .hw = &apq8084_config } }, { .revision = 6, .config = { .hw = &msm8x16_config } }, diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c index 29ae5c9613f36602877bacc5edb12bbce9bb4f6b..323079cfd698b3400908baa6fc6b54b84c355547 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c @@ -229,7 +229,7 @@ static void mdp5_kms_destroy(struct msm_kms *kms) #ifdef CONFIG_DEBUG_FS static int smp_show(struct seq_file *m, void *arg) { - struct drm_info_node *node = (struct drm_info_node *) m->private; + struct drm_info_node *node = m->private; struct drm_device *dev = node->minor->dev; struct msm_drm_private *priv = dev->dev_private; struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(priv->kms)); diff --git a/drivers/gpu/drm/msm/dp/dp_audio.c b/drivers/gpu/drm/msm/dp/dp_audio.c index 6666783e1468e33f3ddaefcfd51789de7588649f..1245c7aa49df844ce293207c6a74374eb0bee805 100644 --- a/drivers/gpu/drm/msm/dp/dp_audio.c +++ b/drivers/gpu/drm/msm/dp/dp_audio.c @@ -593,6 +593,18 @@ static struct hdmi_codec_pdata codec_data = { .i2s = 1, }; +void dp_unregister_audio_driver(struct device *dev, struct dp_audio *dp_audio) +{ + struct dp_audio_private *audio_priv; + + audio_priv = container_of(dp_audio, struct dp_audio_private, dp_audio); + + if (audio_priv->audio_pdev) { + platform_device_unregister(audio_priv->audio_pdev); + audio_priv->audio_pdev = NULL; + } +} + int dp_register_audio_driver(struct device *dev, struct dp_audio *dp_audio) { diff --git a/drivers/gpu/drm/msm/dp/dp_audio.h b/drivers/gpu/drm/msm/dp/dp_audio.h index 84e5f4a5d26badb0bfc8d5f8858681ab5e310ec1..4ab78880af8296efd50654967e297eccf83adc17 100644 --- a/drivers/gpu/drm/msm/dp/dp_audio.h +++ b/drivers/gpu/drm/msm/dp/dp_audio.h @@ -53,6 +53,8 @@ struct dp_audio *dp_audio_get(struct platform_device *pdev, int dp_register_audio_driver(struct device *dev, struct dp_audio *dp_audio); +void dp_unregister_audio_driver(struct device *dev, struct dp_audio *dp_audio); + /** * dp_audio_put() * diff --git a/drivers/gpu/drm/msm/dp/dp_catalog.c b/drivers/gpu/drm/msm/dp/dp_catalog.c index 7a8cf1c8233d07b64594485ee87ca778e4dbfb02..5142aeb705a44dd0891d875ba401d3ea735da71d 100644 --- a/drivers/gpu/drm/msm/dp/dp_catalog.c +++ b/drivers/gpu/drm/msm/dp/dp_catalog.c @@ -620,7 +620,7 @@ void dp_catalog_hpd_config_intr(struct dp_catalog *dp_catalog, config & DP_DP_HPD_INT_MASK); } -void dp_catalog_ctrl_hpd_config(struct dp_catalog *dp_catalog) +void dp_catalog_ctrl_hpd_enable(struct dp_catalog *dp_catalog) { struct dp_catalog_private *catalog = container_of(dp_catalog, struct dp_catalog_private, dp_catalog); @@ -635,6 +635,19 @@ void dp_catalog_ctrl_hpd_config(struct dp_catalog *dp_catalog) dp_write_aux(catalog, REG_DP_DP_HPD_CTRL, DP_DP_HPD_CTRL_HPD_EN); } +void dp_catalog_ctrl_hpd_disable(struct dp_catalog *dp_catalog) +{ + struct dp_catalog_private *catalog = container_of(dp_catalog, + struct dp_catalog_private, dp_catalog); + + u32 reftimer = dp_read_aux(catalog, REG_DP_DP_HPD_REFTIMER); + + reftimer &= ~DP_DP_HPD_REFTIMER_ENABLE; + dp_write_aux(catalog, REG_DP_DP_HPD_REFTIMER, reftimer); + + dp_write_aux(catalog, REG_DP_DP_HPD_CTRL, 0); +} + static void dp_catalog_enable_sdp(struct dp_catalog_private *catalog) { /* trigger sdp */ diff --git a/drivers/gpu/drm/msm/dp/dp_catalog.h b/drivers/gpu/drm/msm/dp/dp_catalog.h index 82376a2697eff4e14f7b8d9315f19f8cfd5bf35e..38786e855b51a0439445f1571fb0979bfb7f4556 100644 --- a/drivers/gpu/drm/msm/dp/dp_catalog.h +++ b/drivers/gpu/drm/msm/dp/dp_catalog.h @@ -104,7 +104,8 @@ bool dp_catalog_ctrl_mainlink_ready(struct dp_catalog *dp_catalog); void dp_catalog_ctrl_enable_irq(struct dp_catalog *dp_catalog, bool enable); void dp_catalog_hpd_config_intr(struct dp_catalog *dp_catalog, u32 intr_mask, bool en); -void dp_catalog_ctrl_hpd_config(struct dp_catalog *dp_catalog); +void dp_catalog_ctrl_hpd_enable(struct dp_catalog *dp_catalog); +void dp_catalog_ctrl_hpd_disable(struct dp_catalog *dp_catalog); void dp_catalog_ctrl_config_psr(struct dp_catalog *dp_catalog); void dp_catalog_ctrl_set_psr(struct dp_catalog *dp_catalog, bool enter); u32 dp_catalog_link_is_connected(struct dp_catalog *dp_catalog); diff --git a/drivers/gpu/drm/msm/dp/dp_ctrl.h b/drivers/gpu/drm/msm/dp/dp_ctrl.h index f712780149fd6823bf9146a70c0d93d474dd70f5..b2c27d3532bf57a42689254ed77629593c7ed8a0 100644 --- a/drivers/gpu/drm/msm/dp/dp_ctrl.h +++ b/drivers/gpu/drm/msm/dp/dp_ctrl.h @@ -14,7 +14,6 @@ #include "dp_catalog.h" struct dp_ctrl { - bool orientation; atomic_t aborted; bool wide_bus_en; }; diff --git a/drivers/gpu/drm/msm/dp/dp_debug.c b/drivers/gpu/drm/msm/dp/dp_debug.c index 5e35033ba3e43ea1fb421ff0b1d60ed202e52973..3bba901afe3354e65ad8fe4f035f15d93a762eda 100644 --- a/drivers/gpu/drm/msm/dp/dp_debug.c +++ b/drivers/gpu/drm/msm/dp/dp_debug.c @@ -21,7 +21,6 @@ struct dp_debug_private { struct dentry *root; - struct dp_usbpd *usbpd; struct dp_link *link; struct dp_panel *panel; struct drm_connector *connector; @@ -232,14 +231,14 @@ static void dp_debug_init(struct dp_debug *dp_debug, struct drm_minor *minor) } struct dp_debug *dp_debug_get(struct device *dev, struct dp_panel *panel, - struct dp_usbpd *usbpd, struct dp_link *link, + struct dp_link *link, struct drm_connector *connector, struct drm_minor *minor) { struct dp_debug_private *debug; struct dp_debug *dp_debug; int rc; - if (!dev || !panel || !usbpd || !link) { + if (!dev || !panel || !link) { DRM_ERROR("invalid input\n"); rc = -EINVAL; goto error; @@ -252,7 +251,6 @@ struct dp_debug *dp_debug_get(struct device *dev, struct dp_panel *panel, } debug->dp_debug.debug_en = false; - debug->usbpd = usbpd; debug->link = link; debug->panel = panel; debug->dev = dev; diff --git a/drivers/gpu/drm/msm/dp/dp_debug.h b/drivers/gpu/drm/msm/dp/dp_debug.h index 8c0d0b5178fdf0617708b285d010dad4d08a6b52..124227873d58cb929e08ae5b13eca60fbda34726 100644 --- a/drivers/gpu/drm/msm/dp/dp_debug.h +++ b/drivers/gpu/drm/msm/dp/dp_debug.h @@ -32,7 +32,6 @@ struct dp_debug { * * @dev: device instance of the caller * @panel: instance of panel module - * @usbpd: instance of usbpd module * @link: instance of link module * @connector: double pointer to display connector * @minor: pointer to drm minor number after device registration @@ -42,7 +41,7 @@ struct dp_debug { * for debugfs input to be communicated with existing modules */ struct dp_debug *dp_debug_get(struct device *dev, struct dp_panel *panel, - struct dp_usbpd *usbpd, struct dp_link *link, + struct dp_link *link, struct drm_connector *connector, struct drm_minor *minor); @@ -59,7 +58,7 @@ void dp_debug_put(struct dp_debug *dp_debug); static inline struct dp_debug *dp_debug_get(struct device *dev, struct dp_panel *panel, - struct dp_usbpd *usbpd, struct dp_link *link, + struct dp_link *link, struct drm_connector *connector, struct drm_minor *minor) { return ERR_PTR(-EINVAL); diff --git a/drivers/gpu/drm/msm/dp/dp_display.c b/drivers/gpu/drm/msm/dp/dp_display.c index 3e13acdfa7e533825f5991ee8470a595d32d1327..76f13954015b18d3df297826ea0672b9a4f10064 100644 --- a/drivers/gpu/drm/msm/dp/dp_display.c +++ b/drivers/gpu/drm/msm/dp/dp_display.c @@ -14,7 +14,6 @@ #include "msm_drv.h" #include "msm_kms.h" -#include "dp_hpd.h" #include "dp_parser.h" #include "dp_power.h" #include "dp_catalog.h" @@ -28,6 +27,10 @@ #include "dp_audio.h" #include "dp_debug.h" +static bool psr_enabled = false; +module_param(psr_enabled, bool, 0); +MODULE_PARM_DESC(psr_enabled, "enable PSR for eDP and DP displays"); + #define HPD_STRING_SIZE 30 enum { @@ -88,7 +91,6 @@ struct dp_display_private { struct platform_device *pdev; struct dentry *root; - struct dp_usbpd *usbpd; struct dp_parser *parser; struct dp_power *power; struct dp_catalog *catalog; @@ -98,7 +100,6 @@ struct dp_display_private { struct dp_ctrl *ctrl; struct dp_debug *debug; - struct dp_usbpd_cb usbpd_cb; struct dp_display_mode dp_mode; struct msm_dp dp_display; @@ -325,7 +326,10 @@ static void dp_display_unbind(struct device *dev, struct device *master, kthread_stop(dp->ev_tsk); + of_dp_aux_depopulate_bus(dp->aux); + dp_power_client_deinit(dp->power); + dp_unregister_audio_driver(dev, dp->audio); dp_aux_unregister(dp->aux); dp->drm_dev = NULL; dp->aux->drm_dev = NULL; @@ -406,7 +410,7 @@ static int dp_display_process_hpd_high(struct dp_display_private *dp) edid = dp->panel->edid; - dp->dp_display.psr_supported = dp->panel->psr_cap.version; + dp->dp_display.psr_supported = dp->panel->psr_cap.version && psr_enabled; dp->audio_supported = drm_detect_monitor_audio(edid); dp_panel_handle_sink_request(dp->panel); @@ -462,7 +466,7 @@ static void dp_display_host_init(struct dp_display_private *dp) dp->dp_display.connector_type, dp->core_initialized, dp->phy_initialized); - dp_power_init(dp->power, false); + dp_power_init(dp->power); dp_ctrl_reset_irq_ctrl(dp->ctrl, true); dp_aux_init(dp->aux); dp->core_initialized = true; @@ -489,11 +493,6 @@ static int dp_display_usbpd_configure_cb(struct device *dev) return dp_display_process_hpd_high(dp); } -static int dp_display_usbpd_disconnect_cb(struct device *dev) -{ - return 0; -} - static int dp_display_notify_disconnect(struct device *dev) { struct dp_display_private *dp = dev_get_dp_display_private(dev); @@ -578,13 +577,9 @@ static int dp_display_usbpd_attention_cb(struct device *dev) static int dp_hpd_plug_handle(struct dp_display_private *dp, u32 data) { - struct dp_usbpd *hpd = dp->usbpd; u32 state; int ret; - if (!hpd) - return 0; - mutex_lock(&dp->event_mutex); state = dp->hpd_state; @@ -615,12 +610,6 @@ static int dp_hpd_plug_handle(struct dp_display_private *dp, u32 data) dp->hpd_state = ST_MAINLINK_READY; } - /* enable HDP irq_hpd/replug interrupt */ - if (dp->dp_display.internal_hpd) - dp_catalog_hpd_config_intr(dp->catalog, - DP_DP_IRQ_HPD_INT_MASK | DP_DP_HPD_REPLUG_INT_MASK, - true); - drm_dbg_dp(dp->drm_dev, "After, type=%d hpd_state=%d\n", dp->dp_display.connector_type, state); mutex_unlock(&dp->event_mutex); @@ -645,12 +634,8 @@ static void dp_display_handle_plugged_change(struct msm_dp *dp_display, static int dp_hpd_unplug_handle(struct dp_display_private *dp, u32 data) { - struct dp_usbpd *hpd = dp->usbpd; u32 state; - if (!hpd) - return 0; - mutex_lock(&dp->event_mutex); state = dp->hpd_state; @@ -658,12 +643,6 @@ static int dp_hpd_unplug_handle(struct dp_display_private *dp, u32 data) drm_dbg_dp(dp->drm_dev, "Before, type=%d hpd_state=%d\n", dp->dp_display.connector_type, state); - /* disable irq_hpd/replug interrupts */ - if (dp->dp_display.internal_hpd) - dp_catalog_hpd_config_intr(dp->catalog, - DP_DP_IRQ_HPD_INT_MASK | DP_DP_HPD_REPLUG_INT_MASK, - false); - /* unplugged, no more irq_hpd handle */ dp_del_event(dp, EV_IRQ_HPD_INT); @@ -687,10 +666,6 @@ static int dp_hpd_unplug_handle(struct dp_display_private *dp, u32 data) return 0; } - /* disable HPD plug interrupts */ - if (dp->dp_display.internal_hpd) - dp_catalog_hpd_config_intr(dp->catalog, DP_DP_HPD_PLUG_INT_MASK, false); - /* * We don't need separate work for disconnect as * connect/attention interrupts are disabled @@ -706,10 +681,6 @@ static int dp_hpd_unplug_handle(struct dp_display_private *dp, u32 data) /* signal the disconnect event early to ensure proper teardown */ dp_display_handle_plugged_change(&dp->dp_display, false); - /* enable HDP plug interrupt to prepare for next plugin */ - if (dp->dp_display.internal_hpd) - dp_catalog_hpd_config_intr(dp->catalog, DP_DP_HPD_PLUG_INT_MASK, true); - drm_dbg_dp(dp->drm_dev, "After, type=%d hpd_state=%d\n", dp->dp_display.connector_type, state); @@ -763,24 +734,10 @@ static int dp_init_sub_modules(struct dp_display_private *dp) { int rc = 0; struct device *dev = &dp->pdev->dev; - struct dp_usbpd_cb *cb = &dp->usbpd_cb; struct dp_panel_in panel_in = { .dev = dev, }; - /* Callback APIs used for cable status change event */ - cb->configure = dp_display_usbpd_configure_cb; - cb->disconnect = dp_display_usbpd_disconnect_cb; - cb->attention = dp_display_usbpd_attention_cb; - - dp->usbpd = dp_hpd_get(dev, cb); - if (IS_ERR(dp->usbpd)) { - rc = PTR_ERR(dp->usbpd); - DRM_ERROR("failed to initialize hpd, rc = %d\n", rc); - dp->usbpd = NULL; - goto error; - } - dp->parser = dp_parser_get(dp->pdev); if (IS_ERR(dp->parser)) { rc = PTR_ERR(dp->parser); @@ -1082,26 +1039,6 @@ void msm_dp_snapshot(struct msm_disp_state *disp_state, struct msm_dp *dp) mutex_unlock(&dp_display->event_mutex); } -static void dp_display_config_hpd(struct dp_display_private *dp) -{ - - dp_display_host_init(dp); - dp_catalog_ctrl_hpd_config(dp->catalog); - - /* Enable plug and unplug interrupts only if requested */ - if (dp->dp_display.internal_hpd) - dp_catalog_hpd_config_intr(dp->catalog, - DP_DP_HPD_PLUG_INT_MASK | - DP_DP_HPD_UNPLUG_INT_MASK, - true); - - /* Enable interrupt first time - * we are leaving dp clocks on during disconnect - * and never disable interrupt - */ - enable_irq(dp->irq); -} - void dp_display_set_psr(struct msm_dp *dp_display, bool enter) { struct dp_display_private *dp; @@ -1176,7 +1113,7 @@ static int hpd_event_thread(void *data) switch (todo->event_id) { case EV_HPD_INIT_SETUP: - dp_display_config_hpd(dp_priv); + dp_display_host_init(dp_priv); break; case EV_HPD_PLUG_INT: dp_hpd_plug_handle(dp_priv, todo->data); @@ -1282,7 +1219,6 @@ int dp_display_request_irq(struct msm_dp *dp_display) dp->irq, rc); return rc; } - disable_irq(dp->irq); return 0; } @@ -1364,9 +1300,9 @@ static int dp_display_remove(struct platform_device *pdev) { struct dp_display_private *dp = dev_get_dp_display_private(&pdev->dev); + component_del(&pdev->dev, &dp_display_comp_ops); dp_display_deinit_sub_modules(dp); - component_del(&pdev->dev, &dp_display_comp_ops); platform_set_drvdata(pdev, NULL); return 0; @@ -1394,13 +1330,8 @@ static int dp_pm_resume(struct device *dev) /* turn on dp ctrl/phy */ dp_display_host_init(dp); - dp_catalog_ctrl_hpd_config(dp->catalog); - - if (dp->dp_display.internal_hpd) - dp_catalog_hpd_config_intr(dp->catalog, - DP_DP_HPD_PLUG_INT_MASK | - DP_DP_HPD_UNPLUG_INT_MASK, - true); + if (dp_display->is_edp) + dp_catalog_ctrl_hpd_enable(dp->catalog); if (dp_catalog_link_is_connected(dp->catalog)) { /* @@ -1540,7 +1471,7 @@ void msm_dp_debugfs_init(struct msm_dp *dp_display, struct drm_minor *minor) dp = container_of(dp_display, struct dp_display_private, dp_display); dev = &dp->pdev->dev; - dp->debug = dp_debug_get(dev, dp->panel, dp->usbpd, + dp->debug = dp_debug_get(dev, dp->panel, dp->link, dp->dp_display.connector, minor); if (IS_ERR(dp->debug)) { @@ -1550,11 +1481,6 @@ void msm_dp_debugfs_init(struct msm_dp *dp_display, struct drm_minor *minor) } } -static void of_dp_aux_depopulate_bus_void(void *data) -{ - of_dp_aux_depopulate_bus(data); -} - static int dp_display_get_next_bridge(struct msm_dp *dp) { int rc; @@ -1568,9 +1494,8 @@ static int dp_display_get_next_bridge(struct msm_dp *dp) if (aux_bus && dp->is_edp) { dp_display_host_init(dp_priv); - dp_catalog_ctrl_hpd_config(dp_priv->catalog); + dp_catalog_ctrl_hpd_enable(dp_priv->catalog); dp_display_host_phy_init(dp_priv); - enable_irq(dp_priv->irq); /* * The code below assumes that the panel will finish probing @@ -1583,12 +1508,6 @@ static int dp_display_get_next_bridge(struct msm_dp *dp) of_node_put(aux_bus); if (rc) goto error; - - rc = devm_add_action_or_reset(dp->drm_dev->dev, - of_dp_aux_depopulate_bus_void, - dp_priv->aux); - if (rc) - goto error; } else if (dp->is_edp) { DRM_ERROR("eDP aux_bus not found\n"); return -ENODEV; @@ -1612,7 +1531,7 @@ static int dp_display_get_next_bridge(struct msm_dp *dp) error: if (dp->is_edp) { - disable_irq(dp_priv->irq); + of_dp_aux_depopulate_bus(dp_priv->aux); dp_display_host_phy_exit(dp_priv); dp_display_host_deinit(dp_priv); } @@ -1801,16 +1720,31 @@ void dp_bridge_hpd_enable(struct drm_bridge *bridge) { struct msm_dp_bridge *dp_bridge = to_dp_bridge(bridge); struct msm_dp *dp_display = dp_bridge->dp_display; + struct dp_display_private *dp = container_of(dp_display, struct dp_display_private, dp_display); + + mutex_lock(&dp->event_mutex); + dp_catalog_ctrl_hpd_enable(dp->catalog); + + /* enable HDP interrupts */ + dp_catalog_hpd_config_intr(dp->catalog, DP_DP_HPD_INT_MASK, true); dp_display->internal_hpd = true; + mutex_unlock(&dp->event_mutex); } void dp_bridge_hpd_disable(struct drm_bridge *bridge) { struct msm_dp_bridge *dp_bridge = to_dp_bridge(bridge); struct msm_dp *dp_display = dp_bridge->dp_display; + struct dp_display_private *dp = container_of(dp_display, struct dp_display_private, dp_display); + + mutex_lock(&dp->event_mutex); + /* disable HDP interrupts */ + dp_catalog_hpd_config_intr(dp->catalog, DP_DP_HPD_INT_MASK, false); + dp_catalog_ctrl_hpd_disable(dp->catalog); dp_display->internal_hpd = false; + mutex_unlock(&dp->event_mutex); } void dp_bridge_hpd_notify(struct drm_bridge *bridge, diff --git a/drivers/gpu/drm/msm/dp/dp_hpd.c b/drivers/gpu/drm/msm/dp/dp_hpd.c deleted file mode 100644 index db98a1d431eb6a586b5e5c5ed25583410fa36e04..0000000000000000000000000000000000000000 --- a/drivers/gpu/drm/msm/dp/dp_hpd.c +++ /dev/null @@ -1,67 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved. - */ - -#define pr_fmt(fmt) "[drm-dp] %s: " fmt, __func__ - -#include -#include - -#include "dp_hpd.h" - -/* DP specific VDM commands */ -#define DP_USBPD_VDM_STATUS 0x10 -#define DP_USBPD_VDM_CONFIGURE 0x11 - -/* USBPD-TypeC specific Macros */ -#define VDM_VERSION 0x0 -#define USB_C_DP_SID 0xFF01 - -struct dp_hpd_private { - struct device *dev; - struct dp_usbpd_cb *dp_cb; - struct dp_usbpd dp_usbpd; -}; - -int dp_hpd_connect(struct dp_usbpd *dp_usbpd, bool hpd) -{ - int rc = 0; - struct dp_hpd_private *hpd_priv; - - hpd_priv = container_of(dp_usbpd, struct dp_hpd_private, - dp_usbpd); - - if (!hpd_priv->dp_cb || !hpd_priv->dp_cb->configure - || !hpd_priv->dp_cb->disconnect) { - pr_err("hpd dp_cb not initialized\n"); - return -EINVAL; - } - if (hpd) - hpd_priv->dp_cb->configure(hpd_priv->dev); - else - hpd_priv->dp_cb->disconnect(hpd_priv->dev); - - return rc; -} - -struct dp_usbpd *dp_hpd_get(struct device *dev, struct dp_usbpd_cb *cb) -{ - struct dp_hpd_private *dp_hpd; - - if (!cb) { - pr_err("invalid cb data\n"); - return ERR_PTR(-EINVAL); - } - - dp_hpd = devm_kzalloc(dev, sizeof(*dp_hpd), GFP_KERNEL); - if (!dp_hpd) - return ERR_PTR(-ENOMEM); - - dp_hpd->dev = dev; - dp_hpd->dp_cb = cb; - - dp_hpd->dp_usbpd.connect = dp_hpd_connect; - - return &dp_hpd->dp_usbpd; -} diff --git a/drivers/gpu/drm/msm/dp/dp_hpd.h b/drivers/gpu/drm/msm/dp/dp_hpd.h deleted file mode 100644 index 8feec5aa50271830d9faf09100e9a269a9192dc5..0000000000000000000000000000000000000000 --- a/drivers/gpu/drm/msm/dp/dp_hpd.h +++ /dev/null @@ -1,78 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved. - */ - -#ifndef _DP_HPD_H_ -#define _DP_HPD_H_ - -//#include - -#include -#include - -enum plug_orientation { - ORIENTATION_NONE, - ORIENTATION_CC1, - ORIENTATION_CC2, -}; - -/** - * struct dp_usbpd - DisplayPort status - * - * @orientation: plug orientation configuration - * @low_pow_st: low power state - * @adaptor_dp_en: adaptor functionality enabled - * @multi_func: multi-function preferred - * @usb_config_req: request to switch to usb - * @exit_dp_mode: request exit from displayport mode - * @hpd_irq: Change in the status since last message - * @alt_mode_cfg_done: bool to specify alt mode status - * @debug_en: bool to specify debug mode - * @connect: simulate disconnect or connect for debug mode - */ -struct dp_usbpd { - enum plug_orientation orientation; - bool low_pow_st; - bool adaptor_dp_en; - bool multi_func; - bool usb_config_req; - bool exit_dp_mode; - bool hpd_irq; - bool alt_mode_cfg_done; - bool debug_en; - - int (*connect)(struct dp_usbpd *dp_usbpd, bool hpd); -}; - -/** - * struct dp_usbpd_cb - callback functions provided by the client - * - * @configure: called by usbpd module when PD communication has - * been completed and the usb peripheral has been configured on - * dp mode. - * @disconnect: notify the cable disconnect issued by usb. - * @attention: notify any attention message issued by usb. - */ -struct dp_usbpd_cb { - int (*configure)(struct device *dev); - int (*disconnect)(struct device *dev); - int (*attention)(struct device *dev); -}; - -/** - * dp_hpd_get() - setup hpd module - * - * @dev: device instance of the caller - * @cb: struct containing callback function pointers. - * - * This function allows the client to initialize the usbpd - * module. The module will communicate with HPD module. - */ -struct dp_usbpd *dp_hpd_get(struct device *dev, struct dp_usbpd_cb *cb); - -int dp_hpd_register(struct dp_usbpd *dp_usbpd); -void dp_hpd_unregister(struct dp_usbpd *dp_usbpd); -int dp_hpd_connect(struct dp_usbpd *dp_usbpd, bool hpd); - -#endif /* _DP_HPD_H_ */ diff --git a/drivers/gpu/drm/msm/dp/dp_panel.h b/drivers/gpu/drm/msm/dp/dp_panel.h index 45208b45eb5357b8a40c7c99ae126c1e07e71031..ed1030e17e1b89bd2e6be18f3d1a832efb986cfa 100644 --- a/drivers/gpu/drm/msm/dp/dp_panel.h +++ b/drivers/gpu/drm/msm/dp/dp_panel.h @@ -10,7 +10,6 @@ #include "dp_aux.h" #include "dp_link.h" -#include "dp_hpd.h" struct edid; diff --git a/drivers/gpu/drm/msm/dp/dp_power.c b/drivers/gpu/drm/msm/dp/dp_power.c index c0aaabb033894c4b031fb1769543505d01097d4f..5cb84ca40e9ed6a97bbeb774015fc7ad0a72e043 100644 --- a/drivers/gpu/drm/msm/dp/dp_power.c +++ b/drivers/gpu/drm/msm/dp/dp_power.c @@ -14,7 +14,6 @@ struct dp_power_private { struct dp_parser *parser; - struct platform_device *pdev; struct device *dev; struct drm_device *drm_dev; struct clk *link_clk_src; @@ -28,32 +27,23 @@ static int dp_power_clk_init(struct dp_power_private *power) { int rc = 0; struct dss_module_power *core, *ctrl, *stream; - struct device *dev = &power->pdev->dev; + struct device *dev = power->dev; core = &power->parser->mp[DP_CORE_PM]; ctrl = &power->parser->mp[DP_CTRL_PM]; stream = &power->parser->mp[DP_STREAM_PM]; rc = devm_clk_bulk_get(dev, core->num_clk, core->clocks); - if (rc) { - DRM_ERROR("failed to get %s clk. err=%d\n", - dp_parser_pm_name(DP_CORE_PM), rc); + if (rc) return rc; - } rc = devm_clk_bulk_get(dev, ctrl->num_clk, ctrl->clocks); - if (rc) { - DRM_ERROR("failed to get %s clk. err=%d\n", - dp_parser_pm_name(DP_CTRL_PM), rc); + if (rc) return -ENODEV; - } rc = devm_clk_bulk_get(dev, stream->num_clk, stream->clocks); - if (rc) { - DRM_ERROR("failed to get %s clk. err=%d\n", - dp_parser_pm_name(DP_CTRL_PM), rc); + if (rc) return -ENODEV; - } return 0; } @@ -121,11 +111,9 @@ int dp_power_clk_enable(struct dp_power *dp_power, mp = &power->parser->mp[DP_CORE_PM]; rc = clk_bulk_prepare_enable(mp->num_clk, mp->clocks); - if (rc) { - DRM_ERROR("fail to enable clks: %s. err=%d\n", - dp_parser_pm_name(DP_CORE_PM), rc); + if (rc) return rc; - } + dp_power->core_clks_on = true; } } @@ -133,10 +121,8 @@ int dp_power_clk_enable(struct dp_power *dp_power, mp = &power->parser->mp[pm_type]; if (enable) { rc = clk_bulk_prepare_enable(mp->num_clk, mp->clocks); - if (rc) { - DRM_ERROR("failed to enable clks, err: %d\n", rc); + if (rc) return rc; - } } else { clk_bulk_disable_unprepare(mp->num_clk, mp->clocks); } @@ -162,63 +148,37 @@ int dp_power_clk_enable(struct dp_power *dp_power, int dp_power_client_init(struct dp_power *dp_power) { - int rc = 0; struct dp_power_private *power; - if (!dp_power) { - DRM_ERROR("invalid power data\n"); - return -EINVAL; - } - power = container_of(dp_power, struct dp_power_private, dp_power); - pm_runtime_enable(&power->pdev->dev); - - rc = dp_power_clk_init(power); - if (rc) - DRM_ERROR("failed to init clocks %d\n", rc); + pm_runtime_enable(power->dev); - return rc; + return dp_power_clk_init(power); } void dp_power_client_deinit(struct dp_power *dp_power) { struct dp_power_private *power; - if (!dp_power) { - DRM_ERROR("invalid power data\n"); - return; - } - power = container_of(dp_power, struct dp_power_private, dp_power); - pm_runtime_disable(&power->pdev->dev); + pm_runtime_disable(power->dev); } -int dp_power_init(struct dp_power *dp_power, bool flip) +int dp_power_init(struct dp_power *dp_power) { int rc = 0; struct dp_power_private *power = NULL; - if (!dp_power) { - DRM_ERROR("invalid power data\n"); - return -EINVAL; - } - power = container_of(dp_power, struct dp_power_private, dp_power); - pm_runtime_get_sync(&power->pdev->dev); + pm_runtime_get_sync(power->dev); rc = dp_power_clk_enable(dp_power, DP_CORE_PM, true); - if (rc) { - DRM_ERROR("failed to enable DP core clocks, %d\n", rc); - goto exit; - } - - return 0; + if (rc) + pm_runtime_put_sync(power->dev); -exit: - pm_runtime_put_sync(&power->pdev->dev); return rc; } @@ -229,7 +189,7 @@ int dp_power_deinit(struct dp_power *dp_power) power = container_of(dp_power, struct dp_power_private, dp_power); dp_power_clk_enable(dp_power, DP_CORE_PM, false); - pm_runtime_put_sync(&power->pdev->dev); + pm_runtime_put_sync(power->dev); return 0; } @@ -238,17 +198,11 @@ struct dp_power *dp_power_get(struct device *dev, struct dp_parser *parser) struct dp_power_private *power; struct dp_power *dp_power; - if (!parser) { - DRM_ERROR("invalid input\n"); - return ERR_PTR(-EINVAL); - } - - power = devm_kzalloc(&parser->pdev->dev, sizeof(*power), GFP_KERNEL); + power = devm_kzalloc(dev, sizeof(*power), GFP_KERNEL); if (!power) return ERR_PTR(-ENOMEM); power->parser = parser; - power->pdev = parser->pdev; power->dev = dev; dp_power = &power->dp_power; diff --git a/drivers/gpu/drm/msm/dp/dp_power.h b/drivers/gpu/drm/msm/dp/dp_power.h index e3f959ffae12153d7fdd1b9602c864f731a4547e..a3dec200785e5f47cb344b00ee47ac390693983f 100644 --- a/drivers/gpu/drm/msm/dp/dp_power.h +++ b/drivers/gpu/drm/msm/dp/dp_power.h @@ -26,13 +26,12 @@ struct dp_power { * dp_power_init() - enable power supplies for display controller * * @power: instance of power module - * @flip: bool for flipping gpio direction * return: 0 if success or error if failure. * * This API will turn on the regulators and configures gpio's * aux/hpd. */ -int dp_power_init(struct dp_power *power, bool flip); +int dp_power_init(struct dp_power *power); /** * dp_power_deinit() - turn off regulators and gpios. diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.c b/drivers/gpu/drm/msm/dsi/dsi_cfg.c index 29ccd755cc2e2606775ea06f70c81c0672aea5ee..8a5fb6df72103229d75ddf3cbddae50dedec09b2 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_cfg.c +++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.c @@ -245,6 +245,8 @@ static const struct msm_dsi_cfg_handler dsi_cfg_handlers[] = { &apq8064_dsi_cfg, &msm_dsi_v2_host_ops}, {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_0, &msm8974_apq8084_dsi_cfg, &msm_dsi_6g_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_0_2, + &msm8974_apq8084_dsi_cfg, &msm_dsi_6g_host_ops}, {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_1, &msm8974_apq8084_dsi_cfg, &msm_dsi_6g_host_ops}, {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_1_1, diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.h b/drivers/gpu/drm/msm/dsi/dsi_cfg.h index 91bdaf50bb1a4a26c64b781a77902464428220f1..43f0dd74edb656628ecceb18001f71e4c9ec3ee3 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_cfg.h +++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.h @@ -11,6 +11,7 @@ #define MSM_DSI_VER_MAJOR_V2 0x02 #define MSM_DSI_VER_MAJOR_6G 0x03 #define MSM_DSI_6G_VER_MINOR_V1_0 0x10000000 +#define MSM_DSI_6G_VER_MINOR_V1_0_2 0x10000002 #define MSM_DSI_6G_VER_MINOR_V1_1 0x10010000 #define MSM_DSI_6G_VER_MINOR_V1_1_1 0x10010001 #define MSM_DSI_6G_VER_MINOR_V1_2 0x10020000 diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c index 961689a255c476cba49d3deb9e8151467de07894..3f6dfb4f9d5a6dabf33a78a5b3c43aa1ed3bbe45 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_host.c +++ b/drivers/gpu/drm/msm/dsi/dsi_host.c @@ -28,6 +28,7 @@ #include "dsi.xml.h" #include "sfpb.xml.h" #include "dsi_cfg.h" +#include "msm_dsc_helper.h" #include "msm_kms.h" #include "msm_gem.h" #include "phy/dsi_phy.h" @@ -117,8 +118,6 @@ struct msm_dsi_host { struct clk *byte_clk; struct clk *esc_clk; struct clk *pixel_clk; - struct clk *byte_clk_src; - struct clk *pixel_clk_src; struct clk *byte_intf_clk; unsigned long byte_clk_rate; @@ -128,8 +127,6 @@ struct msm_dsi_host { /* DSI v2 specific clocks */ struct clk *src_clk; - struct clk *esc_clk_src; - struct clk *dsi_clk_src; unsigned long src_clk_rate; @@ -266,21 +263,6 @@ int dsi_clk_init_v2(struct msm_dsi_host *msm_host) return ret; } - msm_host->esc_clk_src = clk_get_parent(msm_host->esc_clk); - if (!msm_host->esc_clk_src) { - ret = -ENODEV; - pr_err("%s: can't get esc clock parent. ret=%d\n", - __func__, ret); - return ret; - } - - msm_host->dsi_clk_src = clk_get_parent(msm_host->src_clk); - if (!msm_host->dsi_clk_src) { - ret = -ENODEV; - pr_err("%s: can't get src clock parent. ret=%d\n", - __func__, ret); - } - return ret; } @@ -345,20 +327,6 @@ static int dsi_clk_init(struct msm_dsi_host *msm_host) goto exit; } - msm_host->byte_clk_src = clk_get_parent(msm_host->byte_clk); - if (IS_ERR(msm_host->byte_clk_src)) { - ret = PTR_ERR(msm_host->byte_clk_src); - pr_err("%s: can't find byte_clk clock. ret=%d\n", __func__, ret); - goto exit; - } - - msm_host->pixel_clk_src = clk_get_parent(msm_host->pixel_clk); - if (IS_ERR(msm_host->pixel_clk_src)) { - ret = PTR_ERR(msm_host->pixel_clk_src); - pr_err("%s: can't find pixel_clk clock. ret=%d\n", __func__, ret); - goto exit; - } - if (cfg_hnd->ops->clk_init_ver) ret = cfg_hnd->ops->clk_init_ver(msm_host); exit: @@ -560,12 +528,27 @@ void dsi_link_clk_disable_v2(struct msm_dsi_host *msm_host) clk_disable_unprepare(msm_host->byte_clk); } -static unsigned long dsi_get_pclk_rate(const struct drm_display_mode *mode, bool is_bonded_dsi) +static unsigned long dsi_adjust_pclk_for_compression(const struct drm_display_mode *mode, + const struct drm_dsc_config *dsc) +{ + int new_hdisplay = DIV_ROUND_UP(mode->hdisplay * drm_dsc_get_bpp_int(dsc), + dsc->bits_per_component * 3); + + int new_htotal = mode->htotal - mode->hdisplay + new_hdisplay; + + return new_htotal * mode->vtotal * drm_mode_vrefresh(mode); +} + +static unsigned long dsi_get_pclk_rate(const struct drm_display_mode *mode, + const struct drm_dsc_config *dsc, bool is_bonded_dsi) { unsigned long pclk_rate; pclk_rate = mode->clock * 1000; + if (dsc) + pclk_rate = dsi_adjust_pclk_for_compression(mode, dsc); + /* * For bonded DSI mode, the current DRM mode has the complete width of the * panel. Since, the complete panel is driven by two DSI controllers, @@ -584,8 +567,8 @@ unsigned long dsi_byte_clk_get_rate(struct mipi_dsi_host *host, bool is_bonded_d struct msm_dsi_host *msm_host = to_msm_dsi_host(host); u8 lanes = msm_host->lanes; u32 bpp = dsi_get_bpp(msm_host->format); - unsigned long pclk_rate = dsi_get_pclk_rate(mode, is_bonded_dsi); - u64 pclk_bpp = (u64)pclk_rate * bpp; + unsigned long pclk_rate = dsi_get_pclk_rate(mode, msm_host->dsc, is_bonded_dsi); + unsigned long pclk_bpp; if (lanes == 0) { pr_err("%s: forcing mdss_dsi lanes to 1\n", __func__); @@ -594,16 +577,16 @@ unsigned long dsi_byte_clk_get_rate(struct mipi_dsi_host *host, bool is_bonded_d /* CPHY "byte_clk" is in units of 16 bits */ if (msm_host->cphy_mode) - do_div(pclk_bpp, (16 * lanes)); + pclk_bpp = mult_frac(pclk_rate, bpp, 16 * lanes); else - do_div(pclk_bpp, (8 * lanes)); + pclk_bpp = mult_frac(pclk_rate, bpp, 8 * lanes); return pclk_bpp; } static void dsi_calc_pclk(struct msm_dsi_host *msm_host, bool is_bonded_dsi) { - msm_host->pixel_clk_rate = dsi_get_pclk_rate(msm_host->mode, is_bonded_dsi); + msm_host->pixel_clk_rate = dsi_get_pclk_rate(msm_host->mode, msm_host->dsc, is_bonded_dsi); msm_host->byte_clk_rate = dsi_byte_clk_get_rate(&msm_host->base, is_bonded_dsi, msm_host->mode); @@ -627,15 +610,12 @@ int dsi_calc_clk_rate_6g(struct msm_dsi_host *msm_host, bool is_bonded_dsi) int dsi_calc_clk_rate_v2(struct msm_dsi_host *msm_host, bool is_bonded_dsi) { u32 bpp = dsi_get_bpp(msm_host->format); - u64 pclk_bpp; unsigned int esc_mhz, esc_div; unsigned long byte_mhz; dsi_calc_pclk(msm_host, is_bonded_dsi); - pclk_bpp = (u64)dsi_get_pclk_rate(msm_host->mode, is_bonded_dsi) * bpp; - do_div(pclk_bpp, 8); - msm_host->src_clk_rate = pclk_bpp; + msm_host->src_clk_rate = mult_frac(msm_host->pixel_clk_rate, bpp, 8); /* * esc clock is byte clock followed by a 4 bit divider, @@ -725,7 +705,12 @@ static inline enum dsi_cmd_dst_format dsi_get_cmd_fmt( } } -static void dsi_ctrl_config(struct msm_dsi_host *msm_host, bool enable, +static void dsi_ctrl_disable(struct msm_dsi_host *msm_host) +{ + dsi_write(msm_host, REG_DSI_CTRL, 0); +} + +static void dsi_ctrl_enable(struct msm_dsi_host *msm_host, struct msm_dsi_phy_shared_timings *phy_shared_timings, struct msm_dsi_phy *phy) { u32 flags = msm_host->mode_flags; @@ -733,11 +718,6 @@ static void dsi_ctrl_config(struct msm_dsi_host *msm_host, bool enable, const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd; u32 data = 0, lane_ctrl = 0; - if (!enable) { - dsi_write(msm_host, REG_DSI_CTRL, 0); - return; - } - if (flags & MIPI_DSI_MODE_VIDEO) { if (flags & MIPI_DSI_MODE_VIDEO_HSE) data |= DSI_VID_CFG0_PULSE_MODE_HSA_HE; @@ -822,7 +802,7 @@ static void dsi_ctrl_config(struct msm_dsi_host *msm_host, bool enable, if (!(flags & MIPI_DSI_CLOCK_NON_CONTINUOUS)) { lane_ctrl = dsi_read(msm_host, REG_DSI_LANE_CTRL); - if (msm_dsi_phy_set_continuous_clock(phy, enable)) + if (msm_dsi_phy_set_continuous_clock(phy, true)) lane_ctrl &= ~DSI_LANE_CTRL_HS_REQ_SEL_PHY; dsi_write(msm_host, REG_DSI_LANE_CTRL, @@ -848,20 +828,19 @@ static void dsi_update_dsc_timing(struct msm_dsi_host *msm_host, bool is_cmd_mod /* first calculate dsc parameters and then program * compress mode registers */ - slice_per_intf = DIV_ROUND_UP(hdisplay, dsc->slice_width); - - /* - * If slice_count is greater than slice_per_intf - * then default to 1. This can happen during partial - * update. - */ - if (dsc->slice_count > slice_per_intf) - dsc->slice_count = 1; + slice_per_intf = msm_dsc_get_slices_per_intf(dsc, hdisplay); total_bytes_per_intf = dsc->slice_chunk_size * slice_per_intf; eol_byte_num = total_bytes_per_intf % 3; - pkt_per_line = slice_per_intf / dsc->slice_count; + + /* + * Typically, pkt_per_line = slice_per_intf * slice_per_pkt. + * + * Since the current driver only supports slice_per_pkt = 1, + * pkt_per_line will be equal to slice per intf for now. + */ + pkt_per_line = slice_per_intf; if (is_cmd_mode) /* packet data type */ reg = DSI_COMMAND_COMPRESSION_MODE_CTRL_STREAM0_DATATYPE(MIPI_DSI_DCS_LONG_WRITE); @@ -951,7 +930,7 @@ static void dsi_timing_setup(struct msm_dsi_host *msm_host, bool is_bonded_dsi) * pulse width same */ h_total -= hdisplay; - hdisplay /= 3; + hdisplay = DIV_ROUND_UP(msm_dsc_get_bytes_per_line(msm_host->dsc), 3); h_total += hdisplay; ha_end = ha_start + hdisplay; } @@ -985,7 +964,14 @@ static void dsi_timing_setup(struct msm_dsi_host *msm_host, bool is_bonded_dsi) if (!msm_host->dsc) wc = hdisplay * dsi_get_bpp(msm_host->format) / 8 + 1; else - wc = msm_host->dsc->slice_chunk_size * msm_host->dsc->slice_count + 1; + /* + * When DSC is enabled, WC = slice_chunk_size * slice_per_pkt + 1. + * Currently, the driver only supports default value of slice_per_pkt = 1 + * + * TODO: Expand mipi_dsi_device struct to hold slice_per_pkt info + * and adjust DSC math to account for slice_per_pkt. + */ + wc = msm_host->dsc->slice_chunk_size + 1; dsi_write(msm_host, REG_DSI_CMD_MDP_STREAM0_CTRL, DSI_CMD_MDP_STREAM0_CTRL_WORD_COUNT(wc) | @@ -1731,28 +1717,9 @@ static int dsi_host_parse_lane_data(struct msm_dsi_host *msm_host, return -EINVAL; } -static u32 dsi_dsc_rc_buf_thresh[DSC_NUM_BUF_RANGES - 1] = { - 0x0e, 0x1c, 0x2a, 0x38, 0x46, 0x54, 0x62, - 0x69, 0x70, 0x77, 0x79, 0x7b, 0x7d, 0x7e -}; - -/* only 8bpc, 8bpp added */ -static char min_qp[DSC_NUM_BUF_RANGES] = { - 0, 0, 1, 1, 3, 3, 3, 3, 3, 3, 5, 5, 5, 7, 13 -}; - -static char max_qp[DSC_NUM_BUF_RANGES] = { - 4, 4, 5, 6, 7, 7, 7, 8, 9, 10, 11, 12, 13, 13, 15 -}; - -static char bpg_offset[DSC_NUM_BUF_RANGES] = { - 2, 0, 0, -2, -4, -6, -8, -8, -8, -10, -10, -12, -12, -12, -12 -}; - static int dsi_populate_dsc_params(struct msm_dsi_host *msm_host, struct drm_dsc_config *dsc) { - int i; - u16 bpp = dsc->bits_per_pixel >> 4; + int ret; if (dsc->bits_per_pixel & 0xf) { DRM_DEV_ERROR(&msm_host->pdev->dev, "DSI does not support fractional bits_per_pixel\n"); @@ -1764,49 +1731,23 @@ static int dsi_populate_dsc_params(struct msm_dsi_host *msm_host, struct drm_dsc return -EOPNOTSUPP; } - dsc->rc_model_size = 8192; - dsc->first_line_bpg_offset = 12; - dsc->rc_edge_factor = 6; - dsc->rc_tgt_offset_high = 3; - dsc->rc_tgt_offset_low = 3; dsc->simple_422 = 0; dsc->convert_rgb = 1; dsc->vbr_enable = 0; - /* handle only bpp = bpc = 8 */ - for (i = 0; i < DSC_NUM_BUF_RANGES - 1 ; i++) - dsc->rc_buf_thresh[i] = dsi_dsc_rc_buf_thresh[i]; + drm_dsc_set_const_params(dsc); + drm_dsc_set_rc_buf_thresh(dsc); - for (i = 0; i < DSC_NUM_BUF_RANGES; i++) { - dsc->rc_range_params[i].range_min_qp = min_qp[i]; - dsc->rc_range_params[i].range_max_qp = max_qp[i]; - /* - * Range BPG Offset contains two's-complement signed values that fill - * 8 bits, yet the registers and DCS PPS field are only 6 bits wide. - */ - dsc->rc_range_params[i].range_bpg_offset = bpg_offset[i] & DSC_RANGE_BPG_OFFSET_MASK; + /* handle only bpp = bpc = 8, pre-SCR panels */ + ret = drm_dsc_setup_rc_params(dsc, DRM_DSC_1_1_PRE_SCR); + if (ret) { + DRM_DEV_ERROR(&msm_host->pdev->dev, "could not find DSC RC parameters\n"); + return ret; } - dsc->initial_offset = 6144; /* Not bpp 12 */ - if (bpp != 8) - dsc->initial_offset = 2048; /* bpp = 12 */ - - if (dsc->bits_per_component <= 10) - dsc->mux_word_size = DSC_MUX_WORD_SIZE_8_10_BPC; - else - dsc->mux_word_size = DSC_MUX_WORD_SIZE_12_BPC; - - dsc->initial_xmit_delay = 512; - dsc->initial_scale_value = 32; - dsc->first_line_bpg_offset = 12; + dsc->initial_scale_value = drm_dsc_initial_scale_value(dsc); dsc->line_buf_depth = dsc->bits_per_component + 1; - /* bpc 8 */ - dsc->flatness_min_qp = 3; - dsc->flatness_max_qp = 12; - dsc->rc_quant_incr_limit0 = 11; - dsc->rc_quant_incr_limit1 = 11; - return drm_dsc_compute_rc_parameters(dsc); } @@ -2417,7 +2358,7 @@ int msm_dsi_host_power_on(struct mipi_dsi_host *host, dsi_timing_setup(msm_host, is_bonded_dsi); dsi_sw_reset(msm_host); - dsi_ctrl_config(msm_host, true, phy_shared_timings, phy); + dsi_ctrl_enable(msm_host, phy_shared_timings, phy); if (msm_host->disp_en_gpio) gpiod_set_value(msm_host->disp_en_gpio, 1); @@ -2449,7 +2390,7 @@ int msm_dsi_host_power_off(struct mipi_dsi_host *host) goto unlock_ret; } - dsi_ctrl_config(msm_host, false, NULL, NULL); + dsi_ctrl_disable(msm_host); if (msm_host->disp_en_gpio) gpiod_set_value(msm_host->disp_en_gpio, 0); diff --git a/drivers/gpu/drm/msm/dsi/dsi_manager.c b/drivers/gpu/drm/msm/dsi/dsi_manager.c index 1bbac72dad35ba079606f16752dfac93e59e9bf5..28b8012a21f2b3ba9d202797dcbeb93b73961f7a 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_manager.c +++ b/drivers/gpu/drm/msm/dsi/dsi_manager.c @@ -34,32 +34,6 @@ static struct msm_dsi_manager msm_dsim_glb; #define IS_SYNC_NEEDED() (msm_dsim_glb.is_sync_needed) #define IS_MASTER_DSI_LINK(id) (msm_dsim_glb.master_dsi_link_id == id) -#ifdef CONFIG_OF -static bool dsi_mgr_power_on_early(struct drm_bridge *bridge) -{ - struct drm_bridge *next_bridge = drm_bridge_get_next_bridge(bridge); - - /* - * If the next bridge in the chain is the Parade ps8640 bridge chip - * then don't power on early since it seems to violate the expectations - * of the firmware that the bridge chip is running. - * - * NOTE: this is expected to be a temporary special case. It's expected - * that we'll eventually have a framework that allows the next level - * bridge to indicate whether it needs us to power on before it or - * after it. When that framework is in place then we'll use it and - * remove this special case. - */ - return !(next_bridge && next_bridge->of_node && - of_device_is_compatible(next_bridge->of_node, "parade,ps8640")); -} -#else -static inline bool dsi_mgr_power_on_early(struct drm_bridge *bridge) -{ - return true; -} -#endif - static inline struct msm_dsi *dsi_mgr_get_dsi(int id) { return msm_dsim_glb.dsi[id]; @@ -254,7 +228,7 @@ static void msm_dsi_manager_set_split_display(u8 id) } } -static void dsi_mgr_bridge_power_on(struct drm_bridge *bridge) +static int dsi_mgr_bridge_power_on(struct drm_bridge *bridge) { int id = dsi_mgr_bridge_get_id(bridge); struct msm_dsi *msm_dsi = dsi_mgr_get_dsi(id); @@ -265,12 +239,6 @@ static void dsi_mgr_bridge_power_on(struct drm_bridge *bridge) int ret; DBG("id=%d", id); - if (!msm_dsi_device_connected(msm_dsi)) - return; - - /* Do nothing with the host if it is slave-DSI in case of bonded DSI */ - if (is_bonded_dsi && !IS_MASTER_DSI_LINK(id)) - return; ret = dsi_mgr_phy_enable(id, phy_shared_timings); if (ret) @@ -300,14 +268,31 @@ static void dsi_mgr_bridge_power_on(struct drm_bridge *bridge) if (is_bonded_dsi && msm_dsi1) msm_dsi_host_enable_irq(msm_dsi1->host); - return; + return 0; host1_on_fail: msm_dsi_host_power_off(host); host_on_fail: dsi_mgr_phy_disable(id); phy_en_fail: - return; + return ret; +} + +static void dsi_mgr_bridge_power_off(struct drm_bridge *bridge) +{ + int id = dsi_mgr_bridge_get_id(bridge); + struct msm_dsi *msm_dsi = dsi_mgr_get_dsi(id); + struct msm_dsi *msm_dsi1 = dsi_mgr_get_dsi(DSI_1); + struct mipi_dsi_host *host = msm_dsi->host; + bool is_bonded_dsi = IS_BONDED_DSI(); + + msm_dsi_host_disable_irq(host); + if (is_bonded_dsi && msm_dsi1) { + msm_dsi_host_disable_irq(msm_dsi1->host); + msm_dsi_host_power_off(msm_dsi1->host); + } + msm_dsi_host_power_off(host); + dsi_mgr_phy_disable(id); } static void dsi_mgr_bridge_pre_enable(struct drm_bridge *bridge) @@ -327,8 +312,11 @@ static void dsi_mgr_bridge_pre_enable(struct drm_bridge *bridge) if (is_bonded_dsi && !IS_MASTER_DSI_LINK(id)) return; - if (!dsi_mgr_power_on_early(bridge)) - dsi_mgr_bridge_power_on(bridge); + ret = dsi_mgr_bridge_power_on(bridge); + if (ret) { + dev_err(&msm_dsi->pdev->dev, "Power on failed: %d\n", ret); + return; + } ret = msm_dsi_host_enable(host); if (ret) { @@ -349,8 +337,7 @@ static void dsi_mgr_bridge_pre_enable(struct drm_bridge *bridge) host1_en_fail: msm_dsi_host_disable(host); host_en_fail: - - return; + dsi_mgr_bridge_power_off(bridge); } void msm_dsi_manager_tpg_enable(void) @@ -438,9 +425,6 @@ static void dsi_mgr_bridge_mode_set(struct drm_bridge *bridge, msm_dsi_host_set_display_mode(host, adjusted_mode); if (is_bonded_dsi && other_dsi) msm_dsi_host_set_display_mode(other_dsi->host, adjusted_mode); - - if (dsi_mgr_power_on_early(bridge)) - dsi_mgr_bridge_power_on(bridge); } static enum drm_mode_status dsi_mgr_bridge_mode_valid(struct drm_bridge *bridge, diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c index bb09cbe8ff86ba1f8fc6e71021eb3a016f9173c1..9d5795c58a98436f48e35c60a01c462f0e5273d0 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c @@ -541,6 +541,8 @@ static const struct of_device_id dsi_phy_dt_match[] = { .data = &dsi_phy_28nm_hpm_famb_cfgs }, { .compatible = "qcom,dsi-phy-28nm-lp", .data = &dsi_phy_28nm_lp_cfgs }, + { .compatible = "qcom,dsi-phy-28nm-8226", + .data = &dsi_phy_28nm_8226_cfgs }, #endif #ifdef CONFIG_DRM_MSM_DSI_20NM_PHY { .compatible = "qcom,dsi-phy-20nm", diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h index 7137a17ae5238c2d6c281b031927b324dbb0cb67..8b640d174785161bfdeeabbdafc7d500046f1a51 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h @@ -46,8 +46,9 @@ struct msm_dsi_phy_cfg { extern const struct msm_dsi_phy_cfg dsi_phy_28nm_hpm_cfgs; extern const struct msm_dsi_phy_cfg dsi_phy_28nm_hpm_famb_cfgs; extern const struct msm_dsi_phy_cfg dsi_phy_28nm_lp_cfgs; -extern const struct msm_dsi_phy_cfg dsi_phy_20nm_cfgs; +extern const struct msm_dsi_phy_cfg dsi_phy_28nm_8226_cfgs; extern const struct msm_dsi_phy_cfg dsi_phy_28nm_8960_cfgs; +extern const struct msm_dsi_phy_cfg dsi_phy_20nm_cfgs; extern const struct msm_dsi_phy_cfg dsi_phy_14nm_cfgs; extern const struct msm_dsi_phy_cfg dsi_phy_14nm_660_cfgs; extern const struct msm_dsi_phy_cfg dsi_phy_14nm_2290_cfgs; diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c index 9f488adea7f543511de686e1725039fd489f8d97..3ce45b023e637a9cb5ae5e907231a463a0ee17d0 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c @@ -539,6 +539,9 @@ static int dsi_pll_14nm_vco_prepare(struct clk_hw *hw) if (unlikely(pll_14nm->phy->pll_on)) return 0; + if (dsi_pll_14nm_vco_recalc_rate(hw, VCO_REF_CLK_RATE) == 0) + dsi_pll_14nm_vco_set_rate(hw, pll_14nm->phy->cfg->min_pll_rate, VCO_REF_CLK_RATE); + dsi_phy_write(base + REG_DSI_14nm_PHY_PLL_VREF_CFG1, 0x10); dsi_phy_write(cmn_base + REG_DSI_14nm_PHY_CMN_PLL_CNTRL, 1); diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c index 4c1bf55c5f382f552c7b733bb7b5e49e4e0f5f48..ceec7bb87bf138c0193eb4974d151b26c7ab1458 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c @@ -37,6 +37,7 @@ /* v2.0.0 28nm LP implementation */ #define DSI_PHY_28NM_QUIRK_PHY_LP BIT(0) +#define DSI_PHY_28NM_QUIRK_PHY_8226 BIT(1) #define LPFR_LUT_SIZE 10 struct lpfr_cfg { @@ -377,6 +378,74 @@ static int dsi_pll_28nm_vco_prepare_hpm(struct clk_hw *hw) return ret; } +static int dsi_pll_28nm_vco_prepare_8226(struct clk_hw *hw) +{ + struct dsi_pll_28nm *pll_28nm = to_pll_28nm(hw); + struct device *dev = &pll_28nm->phy->pdev->dev; + void __iomem *base = pll_28nm->phy->pll_base; + u32 max_reads = 5, timeout_us = 100; + bool locked; + u32 val; + int i; + + DBG("id=%d", pll_28nm->phy->id); + + pll_28nm_software_reset(pll_28nm); + + /* + * PLL power up sequence. + * Add necessary delays recommended by hardware. + */ + dsi_phy_write(base + REG_DSI_28nm_PHY_PLL_CAL_CFG1, 0x34); + + val = DSI_28nm_PHY_PLL_GLB_CFG_PLL_PWRDN_B; + dsi_phy_write_udelay(base + REG_DSI_28nm_PHY_PLL_GLB_CFG, val, 200); + + val |= DSI_28nm_PHY_PLL_GLB_CFG_PLL_PWRGEN_PWRDN_B; + dsi_phy_write_udelay(base + REG_DSI_28nm_PHY_PLL_GLB_CFG, val, 200); + + val |= DSI_28nm_PHY_PLL_GLB_CFG_PLL_LDO_PWRDN_B; + val |= DSI_28nm_PHY_PLL_GLB_CFG_PLL_ENABLE; + dsi_phy_write_udelay(base + REG_DSI_28nm_PHY_PLL_GLB_CFG, val, 600); + + for (i = 0; i < 7; i++) { + /* DSI Uniphy lock detect setting */ + dsi_phy_write(base + REG_DSI_28nm_PHY_PLL_LKDET_CFG2, 0x0d); + dsi_phy_write_udelay(base + REG_DSI_28nm_PHY_PLL_LKDET_CFG2, + 0x0c, 100); + dsi_phy_write(base + REG_DSI_28nm_PHY_PLL_LKDET_CFG2, 0x0d); + + /* poll for PLL ready status */ + locked = pll_28nm_poll_for_ready(pll_28nm, + max_reads, timeout_us); + if (locked) + break; + + pll_28nm_software_reset(pll_28nm); + + /* + * PLL power up sequence. + * Add necessary delays recommended by hardware. + */ + dsi_phy_write_udelay(base + REG_DSI_28nm_PHY_PLL_PWRGEN_CFG, 0x00, 50); + + val = DSI_28nm_PHY_PLL_GLB_CFG_PLL_PWRDN_B; + val |= DSI_28nm_PHY_PLL_GLB_CFG_PLL_PWRGEN_PWRDN_B; + dsi_phy_write_udelay(base + REG_DSI_28nm_PHY_PLL_GLB_CFG, val, 100); + + val |= DSI_28nm_PHY_PLL_GLB_CFG_PLL_LDO_PWRDN_B; + val |= DSI_28nm_PHY_PLL_GLB_CFG_PLL_ENABLE; + dsi_phy_write_udelay(base + REG_DSI_28nm_PHY_PLL_GLB_CFG, val, 600); + } + + if (unlikely(!locked)) + DRM_DEV_ERROR(dev, "DSI PLL lock failed\n"); + else + DBG("DSI PLL Lock success"); + + return locked ? 0 : -EINVAL; +} + static int dsi_pll_28nm_vco_prepare_lp(struct clk_hw *hw) { struct dsi_pll_28nm *pll_28nm = to_pll_28nm(hw); @@ -471,6 +540,15 @@ static const struct clk_ops clk_ops_dsi_pll_28nm_vco_lp = { .is_enabled = dsi_pll_28nm_clk_is_enabled, }; +static const struct clk_ops clk_ops_dsi_pll_28nm_vco_8226 = { + .round_rate = dsi_pll_28nm_clk_round_rate, + .set_rate = dsi_pll_28nm_clk_set_rate, + .recalc_rate = dsi_pll_28nm_clk_recalc_rate, + .prepare = dsi_pll_28nm_vco_prepare_8226, + .unprepare = dsi_pll_28nm_vco_unprepare, + .is_enabled = dsi_pll_28nm_clk_is_enabled, +}; + /* * PLL Callbacks */ @@ -536,6 +614,8 @@ static int pll_28nm_register(struct dsi_pll_28nm *pll_28nm, struct clk_hw **prov if (pll_28nm->phy->cfg->quirks & DSI_PHY_28NM_QUIRK_PHY_LP) vco_init.ops = &clk_ops_dsi_pll_28nm_vco_lp; + else if (pll_28nm->phy->cfg->quirks & DSI_PHY_28NM_QUIRK_PHY_8226) + vco_init.ops = &clk_ops_dsi_pll_28nm_vco_8226; else vco_init.ops = &clk_ops_dsi_pll_28nm_vco_hpm; @@ -820,3 +900,20 @@ const struct msm_dsi_phy_cfg dsi_phy_28nm_lp_cfgs = { .quirks = DSI_PHY_28NM_QUIRK_PHY_LP, }; +const struct msm_dsi_phy_cfg dsi_phy_28nm_8226_cfgs = { + .has_phy_regulator = true, + .regulator_data = dsi_phy_28nm_regulators, + .num_regulators = ARRAY_SIZE(dsi_phy_28nm_regulators), + .ops = { + .enable = dsi_28nm_phy_enable, + .disable = dsi_28nm_phy_disable, + .pll_init = dsi_pll_28nm_init, + .save_pll_state = dsi_28nm_pll_save_state, + .restore_pll_state = dsi_28nm_pll_restore_state, + }, + .min_pll_rate = VCO_MIN_RATE, + .max_pll_rate = VCO_MAX_RATE, + .io_start = { 0xfd922b00 }, + .num_dsi_phy = 1, + .quirks = DSI_PHY_28NM_QUIRK_PHY_8226, +}; diff --git a/drivers/gpu/drm/msm/msm_atomic.c b/drivers/gpu/drm/msm/msm_atomic.c index d77fa9793c54dc7adb001f9e81dd12ecad891f2a..9c45d641b5212c11078ab38c13a519663d85e10a 100644 --- a/drivers/gpu/drm/msm/msm_atomic.c +++ b/drivers/gpu/drm/msm/msm_atomic.c @@ -155,6 +155,8 @@ static bool can_do_async(struct drm_atomic_state *state, for_each_new_crtc_in_state(state, crtc, crtc_state, i) { if (drm_atomic_crtc_needs_modeset(crtc_state)) return false; + if (!crtc_state->active) + return false; if (++num_crtcs > 1) return false; *async_crtc = crtc; diff --git a/drivers/gpu/drm/msm/msm_debugfs.c b/drivers/gpu/drm/msm/msm_debugfs.c index 9c0e633a3a6114f755502a519b95a1f97c0c76bc..a0a936f80ae350c1633c5345ece5215f7960e579 100644 --- a/drivers/gpu/drm/msm/msm_debugfs.c +++ b/drivers/gpu/drm/msm/msm_debugfs.c @@ -211,7 +211,7 @@ DEFINE_DEBUGFS_ATTRIBUTE(shrink_fops, static int msm_gem_show(struct seq_file *m, void *arg) { - struct drm_info_node *node = (struct drm_info_node *) m->private; + struct drm_info_node *node = m->private; struct drm_device *dev = node->minor->dev; struct msm_drm_private *priv = dev->dev_private; int ret; @@ -229,7 +229,7 @@ static int msm_gem_show(struct seq_file *m, void *arg) static int msm_mm_show(struct seq_file *m, void *arg) { - struct drm_info_node *node = (struct drm_info_node *) m->private; + struct drm_info_node *node = m->private; struct drm_device *dev = node->minor->dev; struct drm_printer p = drm_seq_file_printer(m); @@ -240,7 +240,7 @@ static int msm_mm_show(struct seq_file *m, void *arg) static int msm_fb_show(struct seq_file *m, void *arg) { - struct drm_info_node *node = (struct drm_info_node *) m->private; + struct drm_info_node *node = m->private; struct drm_device *dev = node->minor->dev; struct drm_framebuffer *fb, *fbdev_fb = NULL; diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 0023a6d56804dda6f3b9c279b3fe7f9b692eec42..47efa3c4492c4b1a6bb186cd2caad09d8c35be29 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -449,6 +449,8 @@ static int msm_drm_init(struct device *dev, const struct drm_driver *drv) if (ret) goto err_cleanup_mode_config; + dma_set_max_seg_size(dev, UINT_MAX); + /* Bind all our sub-components: */ ret = component_bind_all(dev, ddev); if (ret) @@ -459,8 +461,6 @@ static int msm_drm_init(struct device *dev, const struct drm_driver *drv) if (ret) goto err_msm_uninit; - dma_set_max_seg_size(dev, UINT_MAX); - msm_gem_shrinker_init(ddev); if (priv->kms_init) { diff --git a/drivers/gpu/drm/msm/msm_dsc_helper.h b/drivers/gpu/drm/msm/msm_dsc_helper.h new file mode 100644 index 0000000000000000000000000000000000000000..b9049fe1e2790703a6f42dd7e6cd3fa5eea29389 --- /dev/null +++ b/drivers/gpu/drm/msm/msm_dsc_helper.h @@ -0,0 +1,38 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved + * + * Helper methods for MSM-specific DSC calculations that are common between timing engine, + * DSI, and DP. + */ + +#ifndef MSM_DSC_HELPER_H_ +#define MSM_DSC_HELPER_H_ + +#include +#include + +/** + * msm_dsc_get_slices_per_intf() - calculate number of slices per interface + * @dsc: Pointer to drm dsc config struct + * @intf_width: interface width in pixels + * Returns: Integer representing the number of slices for the given interface + */ +static inline u32 msm_dsc_get_slices_per_intf(const struct drm_dsc_config *dsc, u32 intf_width) +{ + return DIV_ROUND_UP(intf_width, dsc->slice_width); +} + +/** + * msm_dsc_get_bytes_per_line() - calculate bytes per line + * @dsc: Pointer to drm dsc config struct + * Returns: Integer value representing bytes per line. DSI and DP need + * to perform further calculations to turn this into pclk_per_intf, + * such as dividing by different values depending on if widebus is enabled. + */ +static inline u32 msm_dsc_get_bytes_per_line(const struct drm_dsc_config *dsc) +{ + return dsc->slice_count * dsc->slice_chunk_size; +} + +#endif /* MSM_DSC_HELPER_H_ */ diff --git a/drivers/gpu/drm/msm/msm_fbdev.c b/drivers/gpu/drm/msm/msm_fbdev.c index fa9c1cbffae3c67d623490db3f939c5150ced949..b933a85420f6d54237e6fb9e18b6cb62d4b45f54 100644 --- a/drivers/gpu/drm/msm/msm_fbdev.c +++ b/drivers/gpu/drm/msm/msm_fbdev.c @@ -139,8 +139,28 @@ static int msm_fbdev_create(struct drm_fb_helper *helper, return ret; } +static int msm_fbdev_fb_dirty(struct drm_fb_helper *helper, + struct drm_clip_rect *clip) +{ + struct drm_device *dev = helper->dev; + int ret; + + /* Call damage handlers only if necessary */ + if (!(clip->x1 < clip->x2 && clip->y1 < clip->y2)) + return 0; + + if (helper->fb->funcs->dirty) { + ret = helper->fb->funcs->dirty(helper->fb, NULL, 0, 0, clip, 1); + if (drm_WARN_ONCE(dev, ret, "Dirty helper failed: ret=%d\n", ret)) + return ret; + } + + return 0; +} + static const struct drm_fb_helper_funcs msm_fb_helper_funcs = { .fb_probe = msm_fbdev_create, + .fb_dirty = msm_fbdev_fb_dirty, }; /* diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c index 649688b5a8409bc20615a87d0686a53f4995ef78..635744bc4765fc2c29e959ecc433840d48e8bb17 100644 --- a/drivers/gpu/drm/msm/msm_gem.c +++ b/drivers/gpu/drm/msm/msm_gem.c @@ -219,7 +219,8 @@ static void put_pages(struct drm_gem_object *obj) } } -static struct page **msm_gem_pin_pages_locked(struct drm_gem_object *obj) +static struct page **msm_gem_pin_pages_locked(struct drm_gem_object *obj, + unsigned madv) { struct msm_drm_private *priv = obj->dev->dev_private; struct msm_gem_object *msm_obj = to_msm_bo(obj); @@ -227,7 +228,9 @@ static struct page **msm_gem_pin_pages_locked(struct drm_gem_object *obj) msm_gem_assert_locked(obj); - if (GEM_WARN_ON(msm_obj->madv != MSM_MADV_WILLNEED)) { + if (GEM_WARN_ON(msm_obj->madv > madv)) { + DRM_DEV_ERROR(obj->dev->dev, "Invalid madv state: %u vs %u\n", + msm_obj->madv, madv); return ERR_PTR(-EBUSY); } @@ -248,7 +251,7 @@ struct page **msm_gem_pin_pages(struct drm_gem_object *obj) struct page **p; msm_gem_lock(obj); - p = msm_gem_pin_pages_locked(obj); + p = msm_gem_pin_pages_locked(obj, MSM_MADV_WILLNEED); msm_gem_unlock(obj); return p; @@ -473,10 +476,7 @@ int msm_gem_pin_vma_locked(struct drm_gem_object *obj, struct msm_gem_vma *vma) msm_gem_assert_locked(obj); - if (GEM_WARN_ON(msm_obj->madv != MSM_MADV_WILLNEED)) - return -EBUSY; - - pages = msm_gem_pin_pages_locked(obj); + pages = msm_gem_pin_pages_locked(obj, MSM_MADV_WILLNEED); if (IS_ERR(pages)) return PTR_ERR(pages); @@ -699,13 +699,7 @@ static void *get_vaddr(struct drm_gem_object *obj, unsigned madv) if (obj->import_attach) return ERR_PTR(-ENODEV); - if (GEM_WARN_ON(msm_obj->madv > madv)) { - DRM_DEV_ERROR(obj->dev->dev, "Invalid madv state: %u vs %u\n", - msm_obj->madv, madv); - return ERR_PTR(-EBUSY); - } - - pages = msm_gem_pin_pages_locked(obj); + pages = msm_gem_pin_pages_locked(obj, madv); if (IS_ERR(pages)) return ERR_CAST(pages); diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c index aff18c2f600ab2050d52cd05ba924d3b4ae68a20..3f1aa4de3b87302aaa1f48f999b37ef8d336454a 100644 --- a/drivers/gpu/drm/msm/msm_gem_submit.c +++ b/drivers/gpu/drm/msm/msm_gem_submit.c @@ -558,8 +558,7 @@ static struct drm_syncobj **msm_parse_deps(struct msm_gem_submit *submit, struct drm_file *file, uint64_t in_syncobjs_addr, uint32_t nr_in_syncobjs, - size_t syncobj_stride, - struct msm_ringbuffer *ring) + size_t syncobj_stride) { struct drm_syncobj **syncobjs = NULL; struct drm_msm_gem_submit_syncobj syncobj_desc = {0}; @@ -722,7 +721,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, struct msm_drm_private *priv = dev->dev_private; struct drm_msm_gem_submit *args = data; struct msm_file_private *ctx = file->driver_priv; - struct msm_gem_submit *submit; + struct msm_gem_submit *submit = NULL; struct msm_gpu *gpu = priv->gpu; struct msm_gpu_submitqueue *queue; struct msm_ringbuffer *ring; @@ -769,13 +768,15 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, out_fence_fd = get_unused_fd_flags(O_CLOEXEC); if (out_fence_fd < 0) { ret = out_fence_fd; - return ret; + goto out_post_unlock; } } submit = submit_create(dev, gpu, queue, args->nr_bos, args->nr_cmds); - if (IS_ERR(submit)) - return PTR_ERR(submit); + if (IS_ERR(submit)) { + ret = PTR_ERR(submit); + goto out_post_unlock; + } trace_msm_gpu_submit(pid_nr(submit->pid), ring->id, submit->ident, args->nr_bos, args->nr_cmds); @@ -806,7 +807,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, syncobjs_to_reset = msm_parse_deps(submit, file, args->in_syncobjs, args->nr_in_syncobjs, - args->syncobj_stride, ring); + args->syncobj_stride); if (IS_ERR(syncobjs_to_reset)) { ret = PTR_ERR(syncobjs_to_reset); goto out_unlock; @@ -947,6 +948,8 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, /* The scheduler owns a ref now: */ msm_gem_submit_get(submit); + msm_rd_dump_submit(priv->rd, submit, NULL); + drm_sched_entity_push_job(&submit->base); args->fence = submit->fence_id; @@ -962,11 +965,20 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, if (has_ww_ticket) ww_acquire_fini(&submit->ticket); out_unlock: - if (ret && (out_fence_fd >= 0)) - put_unused_fd(out_fence_fd); mutex_unlock(&queue->lock); out_post_unlock: - msm_gem_submit_put(submit); + if (ret && (out_fence_fd >= 0)) + put_unused_fd(out_fence_fd); + + if (!IS_ERR_OR_NULL(submit)) { + msm_gem_submit_put(submit); + } else { + /* + * If the submit hasn't yet taken ownership of the queue + * then we need to drop the reference ourself: + */ + msm_submitqueue_put(queue); + } if (!IS_ERR_OR_NULL(post_deps)) { for (i = 0; i < args->nr_out_syncobjs; ++i) { kfree(post_deps[i].chain); diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c index c403912d13ab795b85c82c85bf2c31d31b8860f4..52db90e34eadc37c8fc109c121910351348e367d 100644 --- a/drivers/gpu/drm/msm/msm_gpu.c +++ b/drivers/gpu/drm/msm/msm_gpu.c @@ -746,8 +746,6 @@ void msm_gpu_retire(struct msm_gpu *gpu) /* add bo's to gpu's ring, and kick gpu: */ void msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) { - struct drm_device *dev = gpu->dev; - struct msm_drm_private *priv = dev->dev_private; struct msm_ringbuffer *ring = submit->ring; unsigned long flags; @@ -759,8 +757,6 @@ void msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) submit->seqno = submit->hw_fence->seqno; - msm_rd_dump_submit(priv->rd, submit, NULL); - update_sw_cntrs(gpu); /* diff --git a/drivers/gpu/drm/msm/msm_iommu.c b/drivers/gpu/drm/msm/msm_iommu.c index 418e1e06cddefd83b7cfd5aa57f2197de2f4f489..5cc8d358cc9759307a444cd62bce83c62b3dcdb7 100644 --- a/drivers/gpu/drm/msm/msm_iommu.c +++ b/drivers/gpu/drm/msm/msm_iommu.c @@ -234,7 +234,12 @@ struct msm_mmu *msm_iommu_pagetable_create(struct msm_mmu *parent) /* Get the pagetable configuration from the domain */ if (adreno_smmu->cookie) ttbr1_cfg = adreno_smmu->get_ttbr1_cfg(adreno_smmu->cookie); - if (!ttbr1_cfg) + + /* + * If you hit this WARN_ONCE() you are probably missing an entry in + * qcom_smmu_impl_of_match[] in arm-smmu-qcom.c + */ + if (WARN_ONCE(!ttbr1_cfg, "No per-process page tables")) return ERR_PTR(-ENODEV); pagetable = kzalloc(sizeof(*pagetable), GFP_KERNEL); @@ -410,7 +415,7 @@ struct msm_mmu *msm_iommu_gpu_new(struct device *dev, struct msm_gpu *gpu, unsig struct msm_mmu *mmu; mmu = msm_iommu_new(dev, quirks); - if (IS_ERR(mmu)) + if (IS_ERR_OR_NULL(mmu)) return mmu; iommu = to_msm_iommu(mmu); diff --git a/drivers/gpu/drm/msm/msm_mdss.c b/drivers/gpu/drm/msm/msm_mdss.c index e8c93731aaa18f769d54a835dfac6612d1683d27..05648c910c68efe2785c0251606eb17368d9d755 100644 --- a/drivers/gpu/drm/msm/msm_mdss.c +++ b/drivers/gpu/drm/msm/msm_mdss.c @@ -538,6 +538,14 @@ static const struct msm_mdss_data sdm845_data = { .highest_bank_bit = 2, }; +static const struct msm_mdss_data sm6350_data = { + .ubwc_version = UBWC_2_0, + .ubwc_dec_version = UBWC_2_0, + .ubwc_swizzle = 6, + .ubwc_static = 0x1e, + .highest_bank_bit = 1, +}; + static const struct msm_mdss_data sm8150_data = { .ubwc_version = UBWC_3_0, .ubwc_dec_version = UBWC_3_0, @@ -571,6 +579,8 @@ static const struct of_device_id mdss_dt_match[] = { { .compatible = "qcom,sc8180x-mdss", .data = &sc8180x_data }, { .compatible = "qcom,sc8280xp-mdss", .data = &sc8280xp_data }, { .compatible = "qcom,sm6115-mdss", .data = &sm6115_data }, + { .compatible = "qcom,sm6350-mdss", .data = &sm6350_data }, + { .compatible = "qcom,sm6375-mdss", .data = &sm6350_data }, { .compatible = "qcom,sm8150-mdss", .data = &sm8150_data }, { .compatible = "qcom,sm8250-mdss", .data = &sm8250_data }, { .compatible = "qcom,sm8350-mdss", .data = &sm8250_data }, diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c index db2f847c8535fc2c0768a62483db36da6a5e66e1..8d5687d5ed786ee2f3d7ba2bd5fbcf00fdd95838 100644 --- a/drivers/gpu/drm/msm/msm_rd.c +++ b/drivers/gpu/drm/msm/msm_rd.c @@ -83,15 +83,10 @@ struct msm_rd_state { bool open; - /* current submit to read out: */ - struct msm_gem_submit *submit; - /* fifo access is synchronized on the producer side by - * gpu->lock held by submit code (otherwise we could - * end up w/ cmds logged in different order than they - * were executed). And read_lock synchronizes the reads + * write_lock. And read_lock synchronizes the reads */ - struct mutex read_lock; + struct mutex read_lock, write_lock; wait_queue_head_t fifo_event; struct circ_buf fifo; @@ -243,6 +238,7 @@ static void rd_cleanup(struct msm_rd_state *rd) return; mutex_destroy(&rd->read_lock); + mutex_destroy(&rd->write_lock); kfree(rd); } @@ -258,6 +254,7 @@ static struct msm_rd_state *rd_init(struct drm_minor *minor, const char *name) rd->fifo.buf = rd->buf; mutex_init(&rd->read_lock); + mutex_init(&rd->write_lock); init_waitqueue_head(&rd->fifo_event); @@ -338,19 +335,15 @@ static void snapshot_buf(struct msm_rd_state *rd, if (!(submit->bos[idx].flags & MSM_SUBMIT_BO_READ)) return; - msm_gem_lock(&obj->base); buf = msm_gem_get_vaddr_active(&obj->base); if (IS_ERR(buf)) - goto out_unlock; + return; buf += offset; rd_write_section(rd, RD_BUFFER_CONTENTS, buf, size); msm_gem_put_vaddr_locked(&obj->base); - -out_unlock: - msm_gem_unlock(&obj->base); } /* called under gpu->lock */ @@ -364,10 +357,7 @@ void msm_rd_dump_submit(struct msm_rd_state *rd, struct msm_gem_submit *submit, if (!rd->open) return; - /* writing into fifo is serialized by caller, and - * rd->read_lock is used to serialize the reads - */ - WARN_ON(!mutex_is_locked(&submit->gpu->lock)); + mutex_lock(&rd->write_lock); if (fmt) { va_list args; @@ -424,5 +414,7 @@ void msm_rd_dump_submit(struct msm_rd_state *rd, struct msm_gem_submit *submit, break; } } + + mutex_unlock(&rd->write_lock); } #endif diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c b/drivers/gpu/drm/nouveau/dispnv50/disp.c index 9b6824f6b9e4b0ecbb40dcaf5e65476beffd5117..42e1665ba11a3ff5b1dd88b7fdf355ce80c1be34 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/disp.c +++ b/drivers/gpu/drm/nouveau/dispnv50/disp.c @@ -1359,22 +1359,26 @@ nv50_mstm_service(struct nouveau_drm *drm, u8 esi[8] = {}; while (handled) { + u8 ack[8] = {}; + rc = drm_dp_dpcd_read(aux, DP_SINK_COUNT_ESI, esi, 8); if (rc != 8) { ret = false; break; } - drm_dp_mst_hpd_irq(&mstm->mgr, esi, &handled); + drm_dp_mst_hpd_irq_handle_event(&mstm->mgr, esi, ack, &handled); if (!handled) break; - rc = drm_dp_dpcd_write(aux, DP_SINK_COUNT_ESI + 1, &esi[1], - 3); - if (rc != 3) { + rc = drm_dp_dpcd_writeb(aux, DP_SINK_COUNT_ESI + 1, ack[1]); + + if (rc != 1) { ret = false; break; } + + drm_dp_mst_hpd_irq_send_new_request(&mstm->mgr); } if (!ret) diff --git a/drivers/gpu/drm/nouveau/nouveau_acpi.c b/drivers/gpu/drm/nouveau/nouveau_acpi.c index 8cf096f841a908f9403a75b640526c24939020c0..a2ae8c21e4dcea81c18d87913c5961e46f36234d 100644 --- a/drivers/gpu/drm/nouveau/nouveau_acpi.c +++ b/drivers/gpu/drm/nouveau/nouveau_acpi.c @@ -220,6 +220,9 @@ static void nouveau_dsm_pci_probe(struct pci_dev *pdev, acpi_handle *dhandle_out int optimus_funcs; struct pci_dev *parent_pdev; + if (pdev->vendor != PCI_VENDOR_ID_NVIDIA) + return; + *has_pr3 = false; parent_pdev = pci_upstream_bridge(pdev); if (parent_pdev) { diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c index 086b66b60d918a488dcda36db23d816e281ebebf..f75c6f09dd2af5fca8431f1144eb3439511f81c0 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.c +++ b/drivers/gpu/drm/nouveau/nouveau_connector.c @@ -730,7 +730,8 @@ nouveau_connector_detect_lvds(struct drm_connector *connector, bool force) #endif nouveau_connector_set_edid(nv_connector, edid); - nouveau_connector_set_encoder(connector, nv_encoder); + if (nv_encoder) + nouveau_connector_set_encoder(connector, nv_encoder); return status; } @@ -966,7 +967,7 @@ nouveau_connector_get_modes(struct drm_connector *connector) /* Determine display colour depth for everything except LVDS now, * DP requires this before mode_valid() is called. */ - if (connector->connector_type != DRM_MODE_CONNECTOR_LVDS) + if (connector->connector_type != DRM_MODE_CONNECTOR_LVDS && nv_connector->native_mode) nouveau_connector_detect_depth(connector); /* Find the native mode if this is a digital panel, if we didn't @@ -987,7 +988,7 @@ nouveau_connector_get_modes(struct drm_connector *connector) * "native" mode as some VBIOS tables require us to use the * pixel clock as part of the lookup... */ - if (connector->connector_type == DRM_MODE_CONNECTOR_LVDS) + if (connector->connector_type == DRM_MODE_CONNECTOR_LVDS && nv_connector->native_mode) nouveau_connector_detect_depth(connector); if (nv_encoder->dcb->type == DCB_OUTPUT_TV) diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c index 4199b8294e6e88984fb264687f8432a7ad5e3075..51f1918b44d36d62448b82aa01961879a885f192 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drm.c +++ b/drivers/gpu/drm/nouveau/nouveau_drm.c @@ -137,10 +137,16 @@ nouveau_name(struct drm_device *dev) static inline bool nouveau_cli_work_ready(struct dma_fence *fence) { - if (!dma_fence_is_signaled(fence)) - return false; - dma_fence_put(fence); - return true; + bool ret = true; + + spin_lock_irq(fence->lock); + if (!dma_fence_is_signaled_locked(fence)) + ret = false; + spin_unlock_irq(fence->lock); + + if (ret == true) + dma_fence_put(fence); + return ret; } static void diff --git a/drivers/gpu/drm/pl111/pl111_display.c b/drivers/gpu/drm/pl111/pl111_display.c index 6afdf260a4e226492637fa1f27f5a6b33de95918..b9fe926a49e8bb42116ac382010829e333edbf8f 100644 --- a/drivers/gpu/drm/pl111/pl111_display.c +++ b/drivers/gpu/drm/pl111/pl111_display.c @@ -53,7 +53,7 @@ pl111_mode_valid(struct drm_simple_display_pipe *pipe, { struct drm_device *drm = pipe->crtc.dev; struct pl111_drm_dev_private *priv = drm->dev_private; - u32 cpp = priv->variant->fb_bpp / 8; + u32 cpp = DIV_ROUND_UP(priv->variant->fb_depth, 8); u64 bw; /* diff --git a/drivers/gpu/drm/pl111/pl111_drm.h b/drivers/gpu/drm/pl111/pl111_drm.h index 2a46b5bd857649eb811e5d27b122ca3515397d2e..d1fe756444ee52056da07b6189d39d6fa441cb8d 100644 --- a/drivers/gpu/drm/pl111/pl111_drm.h +++ b/drivers/gpu/drm/pl111/pl111_drm.h @@ -114,7 +114,7 @@ struct drm_minor; * extensions to the control register * @formats: array of supported pixel formats on this variant * @nformats: the length of the array of supported pixel formats - * @fb_bpp: desired bits per pixel on the default framebuffer + * @fb_depth: desired depth per pixel on the default framebuffer */ struct pl111_variant_data { const char *name; @@ -126,7 +126,7 @@ struct pl111_variant_data { bool st_bitmux_control; const u32 *formats; unsigned int nformats; - unsigned int fb_bpp; + unsigned int fb_depth; }; struct pl111_drm_dev_private { diff --git a/drivers/gpu/drm/pl111/pl111_drv.c b/drivers/gpu/drm/pl111/pl111_drv.c index 512fc4cfbbf061b8de4a07279bef74fc1ae9702e..c4b8357ea999616054fdd36b25648a26d0831892 100644 --- a/drivers/gpu/drm/pl111/pl111_drv.c +++ b/drivers/gpu/drm/pl111/pl111_drv.c @@ -307,7 +307,7 @@ static int pl111_amba_probe(struct amba_device *amba_dev, if (ret < 0) goto dev_put; - drm_fbdev_dma_setup(drm, priv->variant->fb_bpp); + drm_fbdev_dma_setup(drm, priv->variant->fb_depth); return 0; @@ -350,7 +350,7 @@ static const struct pl111_variant_data pl110_variant = { .is_pl110 = true, .formats = pl110_pixel_formats, .nformats = ARRAY_SIZE(pl110_pixel_formats), - .fb_bpp = 16, + .fb_depth = 16, }; /* RealView, Versatile Express etc use this modern variant */ @@ -375,7 +375,7 @@ static const struct pl111_variant_data pl111_variant = { .name = "PL111", .formats = pl111_pixel_formats, .nformats = ARRAY_SIZE(pl111_pixel_formats), - .fb_bpp = 32, + .fb_depth = 32, }; static const u32 pl110_nomadik_pixel_formats[] = { @@ -404,7 +404,7 @@ static const struct pl111_variant_data pl110_nomadik_variant = { .is_lcdc = true, .st_bitmux_control = true, .broken_vblank = true, - .fb_bpp = 16, + .fb_depth = 16, }; static const struct amba_id pl111_id_table[] = { diff --git a/drivers/gpu/drm/pl111/pl111_versatile.c b/drivers/gpu/drm/pl111/pl111_versatile.c index 1b436b75fd396f2bf71ab067a873873ccd484fd5..00c3ebd32359b975c553e1b275c7e3b64d727879 100644 --- a/drivers/gpu/drm/pl111/pl111_versatile.c +++ b/drivers/gpu/drm/pl111/pl111_versatile.c @@ -316,7 +316,7 @@ static const struct pl111_variant_data pl110_integrator = { .broken_vblank = true, .formats = pl110_integrator_pixel_formats, .nformats = ARRAY_SIZE(pl110_integrator_pixel_formats), - .fb_bpp = 16, + .fb_depth = 16, }; /* @@ -330,7 +330,7 @@ static const struct pl111_variant_data pl110_impd1 = { .broken_vblank = true, .formats = pl110_integrator_pixel_formats, .nformats = ARRAY_SIZE(pl110_integrator_pixel_formats), - .fb_bpp = 16, + .fb_depth = 15, }; /* @@ -343,7 +343,7 @@ static const struct pl111_variant_data pl110_versatile = { .external_bgr = true, .formats = pl110_versatile_pixel_formats, .nformats = ARRAY_SIZE(pl110_versatile_pixel_formats), - .fb_bpp = 16, + .fb_depth = 16, }; /* @@ -355,7 +355,7 @@ static const struct pl111_variant_data pl111_realview = { .name = "PL111 RealView", .formats = pl111_realview_pixel_formats, .nformats = ARRAY_SIZE(pl111_realview_pixel_formats), - .fb_bpp = 16, + .fb_depth = 16, }; /* @@ -367,7 +367,7 @@ static const struct pl111_variant_data pl111_vexpress = { .name = "PL111 Versatile Express", .formats = pl111_realview_pixel_formats, .nformats = ARRAY_SIZE(pl111_realview_pixel_formats), - .fb_bpp = 16, + .fb_depth = 16, .broken_clockdivider = true, }; diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c index 8ef25ab305ae7a5d5cf19b4ca71dc0e49d179ddf..b8f4dac68d85043f76bdb440d5af47058950dc28 100644 --- a/drivers/gpu/drm/radeon/ci_dpm.c +++ b/drivers/gpu/drm/radeon/ci_dpm.c @@ -5517,6 +5517,7 @@ static int ci_parse_power_table(struct radeon_device *rdev) u8 frev, crev; u8 *power_state_offset; struct ci_ps *ps; + int ret; if (!atom_parse_data_header(mode_info->atom_context, index, NULL, &frev, &crev, &data_offset)) @@ -5546,11 +5547,15 @@ static int ci_parse_power_table(struct radeon_device *rdev) non_clock_array_index = power_state->v2.nonClockInfoIndex; non_clock_info = (struct _ATOM_PPLIB_NONCLOCK_INFO *) &non_clock_info_array->nonClockInfo[non_clock_array_index]; - if (!rdev->pm.power_state[i].clock_info) - return -EINVAL; + if (!rdev->pm.power_state[i].clock_info) { + ret = -EINVAL; + goto err_free_ps; + } ps = kzalloc(sizeof(struct ci_ps), GFP_KERNEL); - if (ps == NULL) - return -ENOMEM; + if (ps == NULL) { + ret = -ENOMEM; + goto err_free_ps; + } rdev->pm.dpm.ps[i].ps_priv = ps; ci_parse_pplib_non_clock_info(rdev, &rdev->pm.dpm.ps[i], non_clock_info, @@ -5590,6 +5595,12 @@ static int ci_parse_power_table(struct radeon_device *rdev) } return 0; + +err_free_ps: + for (i = 0; i < rdev->pm.dpm.num_ps; i++) + kfree(rdev->pm.dpm.ps[i].ps_priv); + kfree(rdev->pm.dpm.ps); + return ret; } static int ci_get_vbios_boot_values(struct radeon_device *rdev, @@ -5678,25 +5689,26 @@ int ci_dpm_init(struct radeon_device *rdev) ret = ci_get_vbios_boot_values(rdev, &pi->vbios_boot_state); if (ret) { - ci_dpm_fini(rdev); + kfree(rdev->pm.dpm.priv); return ret; } ret = r600_get_platform_caps(rdev); if (ret) { - ci_dpm_fini(rdev); + kfree(rdev->pm.dpm.priv); return ret; } ret = r600_parse_extended_power_table(rdev); if (ret) { - ci_dpm_fini(rdev); + kfree(rdev->pm.dpm.priv); return ret; } ret = ci_parse_power_table(rdev); if (ret) { - ci_dpm_fini(rdev); + kfree(rdev->pm.dpm.priv); + r600_free_extended_power_table(rdev); return ret; } diff --git a/drivers/gpu/drm/radeon/cypress_dpm.c b/drivers/gpu/drm/radeon/cypress_dpm.c index fdddbbaecbb7401aa9e84cf81996b0da639b7a87..72a0768df00f7701b8aa6862a2d9e0b5bf61a460 100644 --- a/drivers/gpu/drm/radeon/cypress_dpm.c +++ b/drivers/gpu/drm/radeon/cypress_dpm.c @@ -557,8 +557,12 @@ static int cypress_populate_mclk_value(struct radeon_device *rdev, ASIC_INTERNAL_MEMORY_SS, vco_freq)) { u32 reference_clock = rdev->clock.mpll.reference_freq; u32 decoded_ref = rv740_get_decoded_reference_divider(dividers.ref_div); - u32 clk_s = reference_clock * 5 / (decoded_ref * ss.rate); - u32 clk_v = ss.percentage * + u32 clk_s, clk_v; + + if (!decoded_ref) + return -EINVAL; + clk_s = reference_clock * 5 / (decoded_ref * ss.rate); + clk_v = ss.percentage * (0x4000 * dividers.whole_fb_div + 0x800 * dividers.frac_fb_div) / (clk_s * 625); mpll_ss1 &= ~CLKV_MASK; diff --git a/drivers/gpu/drm/radeon/ni_dpm.c b/drivers/gpu/drm/radeon/ni_dpm.c index 672d2239293e013104e9596e3fc7085331a46263..3e1c1a392fb7b3b2e7c039d476e89adb793dc0de 100644 --- a/drivers/gpu/drm/radeon/ni_dpm.c +++ b/drivers/gpu/drm/radeon/ni_dpm.c @@ -2241,8 +2241,12 @@ static int ni_populate_mclk_value(struct radeon_device *rdev, ASIC_INTERNAL_MEMORY_SS, vco_freq)) { u32 reference_clock = rdev->clock.mpll.reference_freq; u32 decoded_ref = rv740_get_decoded_reference_divider(dividers.ref_div); - u32 clk_s = reference_clock * 5 / (decoded_ref * ss.rate); - u32 clk_v = ss.percentage * + u32 clk_s, clk_v; + + if (!decoded_ref) + return -EINVAL; + clk_s = reference_clock * 5 / (decoded_ref * ss.rate); + clk_v = ss.percentage * (0x4000 * dividers.whole_fb_div + 0x800 * dividers.frac_fb_div) / (clk_s * 625); mpll_ss1 &= ~CLKV_MASK; diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c index d4f09ecc3d2214f1158ac90a07df834a005dc342..affa9e0309b274ede232694c5df4ec9f7f705594 100644 --- a/drivers/gpu/drm/radeon/r100.c +++ b/drivers/gpu/drm/radeon/r100.c @@ -2929,7 +2929,7 @@ static void r100_set_safe_registers(struct radeon_device *rdev) #if defined(CONFIG_DEBUG_FS) static int r100_debugfs_rbbm_info_show(struct seq_file *m, void *unused) { - struct radeon_device *rdev = (struct radeon_device *)m->private; + struct radeon_device *rdev = m->private; uint32_t reg, value; unsigned i; @@ -2948,7 +2948,7 @@ static int r100_debugfs_rbbm_info_show(struct seq_file *m, void *unused) static int r100_debugfs_cp_ring_info_show(struct seq_file *m, void *unused) { - struct radeon_device *rdev = (struct radeon_device *)m->private; + struct radeon_device *rdev = m->private; struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]; uint32_t rdp, wdp; unsigned count, i, j; @@ -2974,7 +2974,7 @@ static int r100_debugfs_cp_ring_info_show(struct seq_file *m, void *unused) static int r100_debugfs_cp_csq_fifo_show(struct seq_file *m, void *unused) { - struct radeon_device *rdev = (struct radeon_device *)m->private; + struct radeon_device *rdev = m->private; uint32_t csq_stat, csq2_stat, tmp; unsigned r_rptr, r_wptr, ib1_rptr, ib1_wptr, ib2_rptr, ib2_wptr; unsigned i; @@ -3022,7 +3022,7 @@ static int r100_debugfs_cp_csq_fifo_show(struct seq_file *m, void *unused) static int r100_debugfs_mc_info_show(struct seq_file *m, void *unused) { - struct radeon_device *rdev = (struct radeon_device *)m->private; + struct radeon_device *rdev = m->private; uint32_t tmp; tmp = RREG32(RADEON_CONFIG_MEMSIZE); diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c index 7b0cfeaddcec35e1dd455317ce3ae0ebad26fea4..9c1a92fa2af6dd162ac6508e210403f409232b7b 100644 --- a/drivers/gpu/drm/radeon/r300.c +++ b/drivers/gpu/drm/radeon/r300.c @@ -589,7 +589,7 @@ int rv370_get_pcie_lanes(struct radeon_device *rdev) #if defined(CONFIG_DEBUG_FS) static int rv370_debugfs_pcie_gart_info_show(struct seq_file *m, void *unused) { - struct radeon_device *rdev = (struct radeon_device *)m->private; + struct radeon_device *rdev = m->private; uint32_t tmp; tmp = RREG32_PCIE(RADEON_PCIE_TX_GART_CNTL); diff --git a/drivers/gpu/drm/radeon/r420.c b/drivers/gpu/drm/radeon/r420.c index 7e6320e8c6a0c2ec63fc05f42c9d959691b75e21..eae8a6389f5ea2be24b8688b3c2f32bb05754cfa 100644 --- a/drivers/gpu/drm/radeon/r420.c +++ b/drivers/gpu/drm/radeon/r420.c @@ -474,7 +474,7 @@ int r420_init(struct radeon_device *rdev) #if defined(CONFIG_DEBUG_FS) static int r420_debugfs_pipes_info_show(struct seq_file *m, void *unused) { - struct radeon_device *rdev = (struct radeon_device *)m->private; + struct radeon_device *rdev = m->private; uint32_t tmp; tmp = RREG32(R400_GB_PIPE_SELECT); diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index dd78fc4994024815e0758ad73aef807693a42496..382795a8b3c064ba8602184d715571c5330e0e24 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -4345,7 +4345,7 @@ int r600_irq_process(struct radeon_device *rdev) static int r600_debugfs_mc_info_show(struct seq_file *m, void *unused) { - struct radeon_device *rdev = (struct radeon_device *)m->private; + struct radeon_device *rdev = m->private; DREG32_SYS(m, rdev, R_000E50_SRBM_STATUS); DREG32_SYS(m, rdev, VM_L2_STATUS); diff --git a/drivers/gpu/drm/radeon/radeon_acpi.h b/drivers/gpu/drm/radeon/radeon_acpi.h index 35202a453e660320404b84a30607dac6ca817407..974fbb4ce2c29a4933345ca633ef2e5b484ba30e 100644 --- a/drivers/gpu/drm/radeon/radeon_acpi.h +++ b/drivers/gpu/drm/radeon/radeon_acpi.h @@ -453,4 +453,13 @@ struct acpi_bus_event; * BYTE - number of active lanes */ +#if defined(CONFIG_VGA_SWITCHEROO) +void radeon_register_atpx_handler(void); +void radeon_unregister_atpx_handler(void); +bool radeon_has_atpx_dgpu_power_cntl(void); +bool radeon_is_atpx_hybrid(void); +bool radeon_has_atpx(void); +bool radeon_atpx_dgpu_req_power_for_displays(void); +#endif + #endif diff --git a/drivers/gpu/drm/radeon/radeon_atpx_handler.c b/drivers/gpu/drm/radeon/radeon_atpx_handler.c index 6f93f54bf65197d3e349613982e52122aa25c4fb..d0b450a0650601ec9f92512365b97548caa1f21c 100644 --- a/drivers/gpu/drm/radeon/radeon_atpx_handler.c +++ b/drivers/gpu/drm/radeon/radeon_atpx_handler.c @@ -147,7 +147,7 @@ static void radeon_atpx_parse_functions(struct radeon_atpx_functions *f, u32 mas } /** - * radeon_atpx_validate_functions - validate ATPX functions + * radeon_atpx_validate() - validate ATPX functions * * @atpx: radeon atpx struct * diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c index 46a27ebf4588a2241d4ab7f8f00be4b8ebff9057..a6700d7278bf3ee6967c8a43e67922ae436a5f49 100644 --- a/drivers/gpu/drm/radeon/radeon_cs.c +++ b/drivers/gpu/drm/radeon/radeon_cs.c @@ -270,7 +270,8 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data) { struct drm_radeon_cs *cs = data; uint64_t *chunk_array_ptr; - unsigned size, i; + u64 size; + unsigned i; u32 ring = RADEON_CS_RING_GFX; s32 priority = 0; diff --git a/drivers/gpu/drm/radeon/radeon_fbdev.c b/drivers/gpu/drm/radeon/radeon_fbdev.c index 28212c2d6c986c4137ff436f754bf81075df26a5..ab9c1abbac979e50934f8a28c8341877aaf19e28 100644 --- a/drivers/gpu/drm/radeon/radeon_fbdev.c +++ b/drivers/gpu/drm/radeon/radeon_fbdev.c @@ -304,6 +304,7 @@ static void radeon_fbdev_client_unregister(struct drm_client_dev *client) if (fb_helper->info) { vga_switcheroo_client_fb_set(rdev->pdev, NULL); + drm_helper_force_disable_all(dev); drm_fb_helper_unregister_info(fb_helper); } else { drm_client_release(&fb_helper->client); diff --git a/drivers/gpu/drm/radeon/radeon_fence.c b/drivers/gpu/drm/radeon/radeon_fence.c index 73e3117420bf0eae166a9ac0237d82986936494e..2749dde5838f1ce4956723e7afeeb856e2bfdcad 100644 --- a/drivers/gpu/drm/radeon/radeon_fence.c +++ b/drivers/gpu/drm/radeon/radeon_fence.c @@ -955,7 +955,7 @@ void radeon_fence_driver_force_completion(struct radeon_device *rdev, int ring) #if defined(CONFIG_DEBUG_FS) static int radeon_debugfs_fence_info_show(struct seq_file *m, void *data) { - struct radeon_device *rdev = (struct radeon_device *)m->private; + struct radeon_device *rdev = m->private; int i, j; for (i = 0; i < RADEON_NUM_RINGS; ++i) { diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c index bdc5af23f005accc418ea77a9fe2d59e9ae409db..d0119c5f7eb3e6eb95ed820df362b0103a298294 100644 --- a/drivers/gpu/drm/radeon/radeon_gem.c +++ b/drivers/gpu/drm/radeon/radeon_gem.c @@ -459,7 +459,6 @@ int radeon_gem_set_domain_ioctl(struct drm_device *dev, void *data, struct radeon_device *rdev = dev->dev_private; struct drm_radeon_gem_set_domain *args = data; struct drm_gem_object *gobj; - struct radeon_bo *robj; int r; /* for now if someone requests domain CPU - @@ -472,13 +471,12 @@ int radeon_gem_set_domain_ioctl(struct drm_device *dev, void *data, up_read(&rdev->exclusive_lock); return -ENOENT; } - robj = gem_to_radeon_bo(gobj); r = radeon_gem_set_domain(gobj, args->read_domains, args->write_domain); drm_gem_object_put(gobj); up_read(&rdev->exclusive_lock); - r = radeon_gem_handle_lockup(robj->rdev, r); + r = radeon_gem_handle_lockup(rdev, r); return r; } @@ -879,7 +877,7 @@ int radeon_mode_dumb_create(struct drm_file *file_priv, #if defined(CONFIG_DEBUG_FS) static int radeon_debugfs_gem_info_show(struct seq_file *m, void *unused) { - struct radeon_device *rdev = (struct radeon_device *)m->private; + struct radeon_device *rdev = m->private; struct radeon_bo *rbo; unsigned i = 0; diff --git a/drivers/gpu/drm/radeon/radeon_ib.c b/drivers/gpu/drm/radeon/radeon_ib.c index 6a45a72488f97f62008ad1f27167a4b9ec5cb5b4..fb9ecf5dbe2b7a99802c704dac71e48146b46432 100644 --- a/drivers/gpu/drm/radeon/radeon_ib.c +++ b/drivers/gpu/drm/radeon/radeon_ib.c @@ -292,7 +292,7 @@ int radeon_ib_ring_tests(struct radeon_device *rdev) static int radeon_debugfs_sa_info_show(struct seq_file *m, void *unused) { - struct radeon_device *rdev = (struct radeon_device *)m->private; + struct radeon_device *rdev = m->private; radeon_sa_bo_dump_debug_info(&rdev->ring_tmp_bo, m); diff --git a/drivers/gpu/drm/radeon/radeon_irq_kms.c b/drivers/gpu/drm/radeon/radeon_irq_kms.c index 3377fbc71f65478b630085fccdbc786306f0157e..c4dda908666cfc4f9a312517698c7c0c7870af1f 100644 --- a/drivers/gpu/drm/radeon/radeon_irq_kms.c +++ b/drivers/gpu/drm/radeon/radeon_irq_kms.c @@ -99,6 +99,16 @@ static void radeon_hotplug_work_func(struct work_struct *work) static void radeon_dp_work_func(struct work_struct *work) { + struct radeon_device *rdev = container_of(work, struct radeon_device, + dp_work); + struct drm_device *dev = rdev->ddev; + struct drm_mode_config *mode_config = &dev->mode_config; + struct drm_connector *connector; + + mutex_lock(&mode_config->mutex); + list_for_each_entry(connector, &mode_config->connector_list, head) + radeon_connector_hotplug(connector); + mutex_unlock(&mode_config->mutex); } /** diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c index cbc554928bccf6c53e63256d74f290c709b52478..b73fd9ab02522a7fa5b2310bffa8f6c80893289f 100644 --- a/drivers/gpu/drm/radeon/radeon_pm.c +++ b/drivers/gpu/drm/radeon/radeon_pm.c @@ -1916,7 +1916,7 @@ static void radeon_dynpm_idle_work_handler(struct work_struct *work) static int radeon_debugfs_pm_info_show(struct seq_file *m, void *unused) { - struct radeon_device *rdev = (struct radeon_device *)m->private; + struct radeon_device *rdev = m->private; struct drm_device *ddev = rdev->ddev; if ((rdev->flags & RADEON_IS_PX) && diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c index 7e207276df374f84439e8acbf3ffae8bcc1a962c..e6534fa9f1fb5443b0bc3d13a773f263d4f16ec5 100644 --- a/drivers/gpu/drm/radeon/radeon_ring.c +++ b/drivers/gpu/drm/radeon/radeon_ring.c @@ -464,7 +464,7 @@ void radeon_ring_fini(struct radeon_device *rdev, struct radeon_ring *ring) static int radeon_debugfs_ring_info_show(struct seq_file *m, void *unused) { - struct radeon_ring *ring = (struct radeon_ring *) m->private; + struct radeon_ring *ring = m->private; struct radeon_device *rdev = ring->rdev; uint32_t rptr, wptr, rptr_next; diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c index 2220cdf6a3f68014ae2c6fbba3564f14101ee052..10794be3023903abf89426355f92b165c1659ed3 100644 --- a/drivers/gpu/drm/radeon/radeon_ttm.c +++ b/drivers/gpu/drm/radeon/radeon_ttm.c @@ -36,7 +36,6 @@ #include #include #include -#include #include #include @@ -780,7 +779,7 @@ void radeon_ttm_set_active_vram_size(struct radeon_device *rdev, u64 size) static int radeon_ttm_page_pool_show(struct seq_file *m, void *data) { - struct radeon_device *rdev = (struct radeon_device *)m->private; + struct radeon_device *rdev = m->private; return ttm_pool_debugfs(&rdev->mman.bdev.pool, m); } diff --git a/drivers/gpu/drm/radeon/rs400.c b/drivers/gpu/drm/radeon/rs400.c index 6383f7a34bd8cf9be272a73e5d4a8aabd394571e..922a29e58880270f1edc9325f9e805bd9076a862 100644 --- a/drivers/gpu/drm/radeon/rs400.c +++ b/drivers/gpu/drm/radeon/rs400.c @@ -307,7 +307,7 @@ void rs400_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v) #if defined(CONFIG_DEBUG_FS) static int rs400_debugfs_gart_info_show(struct seq_file *m, void *unused) { - struct radeon_device *rdev = (struct radeon_device *)m->private; + struct radeon_device *rdev = m->private; uint32_t tmp; tmp = RREG32(RADEON_HOST_PATH_CNTL); diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c index 63fb06e8e2d7ca2fbdf24de2e7e77ec69d2d5bfb..76260fdfbaa725c35c9ec5abc67b37bd029a9f91 100644 --- a/drivers/gpu/drm/radeon/rv515.c +++ b/drivers/gpu/drm/radeon/rv515.c @@ -221,7 +221,7 @@ void rv515_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v) #if defined(CONFIG_DEBUG_FS) static int rv515_debugfs_pipes_info_show(struct seq_file *m, void *unused) { - struct radeon_device *rdev = (struct radeon_device *)m->private; + struct radeon_device *rdev = m->private; uint32_t tmp; tmp = RREG32(GB_PIPE_SELECT); @@ -237,7 +237,7 @@ static int rv515_debugfs_pipes_info_show(struct seq_file *m, void *unused) static int rv515_debugfs_ga_info_show(struct seq_file *m, void *unused) { - struct radeon_device *rdev = (struct radeon_device *)m->private; + struct radeon_device *rdev = m->private; uint32_t tmp; tmp = RREG32(0x2140); diff --git a/drivers/gpu/drm/radeon/rv740_dpm.c b/drivers/gpu/drm/radeon/rv740_dpm.c index d57a3e1df8d63fe1778e48a66b6bdaeacbc476fc..4464fd21a30296f76b1a21e2f66632af17ed904b 100644 --- a/drivers/gpu/drm/radeon/rv740_dpm.c +++ b/drivers/gpu/drm/radeon/rv740_dpm.c @@ -249,8 +249,12 @@ int rv740_populate_mclk_value(struct radeon_device *rdev, ASIC_INTERNAL_MEMORY_SS, vco_freq)) { u32 reference_clock = rdev->clock.mpll.reference_freq; u32 decoded_ref = rv740_get_decoded_reference_divider(dividers.ref_div); - u32 clk_s = reference_clock * 5 / (decoded_ref * ss.rate); - u32 clk_v = 0x40000 * ss.percentage * + u32 clk_s, clk_v; + + if (!decoded_ref) + return -EINVAL; + clk_s = reference_clock * 5 / (decoded_ref * ss.rate); + clk_v = 0x40000 * ss.percentage * (dividers.whole_fb_div + (dividers.frac_fb_div / 8)) / (clk_s * 10000); mpll_ss1 &= ~CLKV_MASK; diff --git a/drivers/gpu/drm/renesas/Kconfig b/drivers/gpu/drm/renesas/Kconfig new file mode 100644 index 0000000000000000000000000000000000000000..3777dad17f81dee89a7ecb6f7bcb2f0a1b2462bb --- /dev/null +++ b/drivers/gpu/drm/renesas/Kconfig @@ -0,0 +1,4 @@ +# SPDX-License-Identifier: GPL-2.0-only + +source "drivers/gpu/drm/renesas/rcar-du/Kconfig" +source "drivers/gpu/drm/renesas/shmobile/Kconfig" diff --git a/drivers/gpu/drm/renesas/Makefile b/drivers/gpu/drm/renesas/Makefile new file mode 100644 index 0000000000000000000000000000000000000000..ec0e89e7a592682cd8b599d9d29c874953e93ad3 --- /dev/null +++ b/drivers/gpu/drm/renesas/Makefile @@ -0,0 +1,4 @@ +# SPDX-License-Identifier: GPL-2.0 + +obj-y += rcar-du/ +obj-$(CONFIG_DRM_SHMOBILE) += shmobile/ diff --git a/drivers/gpu/drm/rcar-du/Kconfig b/drivers/gpu/drm/renesas/rcar-du/Kconfig similarity index 100% rename from drivers/gpu/drm/rcar-du/Kconfig rename to drivers/gpu/drm/renesas/rcar-du/Kconfig diff --git a/drivers/gpu/drm/rcar-du/Makefile b/drivers/gpu/drm/renesas/rcar-du/Makefile similarity index 100% rename from drivers/gpu/drm/rcar-du/Makefile rename to drivers/gpu/drm/renesas/rcar-du/Makefile diff --git a/drivers/gpu/drm/rcar-du/rcar_cmm.c b/drivers/gpu/drm/renesas/rcar-du/rcar_cmm.c similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_cmm.c rename to drivers/gpu/drm/renesas/rcar-du/rcar_cmm.c diff --git a/drivers/gpu/drm/rcar-du/rcar_cmm.h b/drivers/gpu/drm/renesas/rcar-du/rcar_cmm.h similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_cmm.h rename to drivers/gpu/drm/renesas/rcar-du/rcar_cmm.h diff --git a/drivers/gpu/drm/rcar-du/rcar_du_crtc.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_crtc.c similarity index 96% rename from drivers/gpu/drm/rcar-du/rcar_du_crtc.c rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_crtc.c index d6d29be6b4f482411b5c5c437eab69abb50371df..7e175dbfd89246fbcdea96c8f3a204d7d816512b 100644 --- a/drivers/gpu/drm/rcar-du/rcar_du_crtc.c +++ b/drivers/gpu/drm/renesas/rcar-du/rcar_du_crtc.c @@ -223,20 +223,6 @@ static void rcar_du_crtc_set_display_timing(struct rcar_du_crtc *rcrtc) * DU channels that have a display PLL can't use the internal * system clock, and have no internal clock divider. */ - - /* - * The H3 ES1.x exhibits dot clock duty cycle stability issues. - * We can work around them by configuring the DPLL to twice the - * desired frequency, coupled with a /2 post-divider. Restrict - * the workaround to H3 ES1.x as ES2.0 and all other SoCs have - * no post-divider when a display PLL is present (as shown by - * the workaround breaking HDMI output on M3-W during testing). - */ - if (rcdu->info->quirks & RCAR_DU_QUIRK_H3_ES1_PCLK_STABILITY) { - target *= 2; - div = 1; - } - extclk = clk_get_rate(rcrtc->extclock); rcar_du_dpll_divider(rcrtc, &dpll, extclk, target); @@ -245,30 +231,13 @@ static void rcar_du_crtc_set_display_timing(struct rcar_du_crtc *rcrtc) | DPLLCR_N(dpll.n) | DPLLCR_M(dpll.m) | DPLLCR_STBY; - if (rcrtc->index == 1) { + if (rcrtc->index == 1) dpllcr |= DPLLCR_PLCS1 | DPLLCR_INCS_DOTCLKIN1; - } else { - dpllcr |= DPLLCR_PLCS0_PLL + else + dpllcr |= DPLLCR_PLCS0 | DPLLCR_INCS_DOTCLKIN0; - /* - * On ES2.x we have a single mux controlled via bit 21, - * which selects between DCLKIN source (bit 21 = 0) and - * a PLL source (bit 21 = 1), where the PLL is always - * PLL1. - * - * On ES1.x we have an additional mux, controlled - * via bit 20, for choosing between PLL0 (bit 20 = 0) - * and PLL1 (bit 20 = 1). We always want to use PLL1, - * so on ES1.x, in addition to setting bit 21, we need - * to set the bit 20. - */ - - if (rcdu->info->quirks & RCAR_DU_QUIRK_H3_ES1_PLL) - dpllcr |= DPLLCR_PLCS0_H3ES1X_PLL1; - } - rcar_du_group_write(rcrtc->group, DPLLCR, dpllcr); escr = ESCR_DCLKSEL_DCLKIN | div; diff --git a/drivers/gpu/drm/rcar-du/rcar_du_crtc.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_crtc.h similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_du_crtc.h rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_crtc.h diff --git a/drivers/gpu/drm/rcar-du/rcar_du_drv.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_drv.c similarity index 93% rename from drivers/gpu/drm/rcar-du/rcar_du_drv.c rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_drv.c index 602d78eba3076930e2f62535cf7acf29636c6fcf..4280ff5fa91f276a49ffff530f6fe419e0070fad 100644 --- a/drivers/gpu/drm/rcar-du/rcar_du_drv.c +++ b/drivers/gpu/drm/renesas/rcar-du/rcar_du_drv.c @@ -16,7 +16,6 @@ #include #include #include -#include #include #include @@ -387,43 +386,6 @@ static const struct rcar_du_device_info rcar_du_r8a7795_info = { .dpll_mask = BIT(2) | BIT(1), }; -static const struct rcar_du_device_info rcar_du_r8a7795_es1_info = { - .gen = 3, - .features = RCAR_DU_FEATURE_CRTC_IRQ - | RCAR_DU_FEATURE_CRTC_CLOCK - | RCAR_DU_FEATURE_VSP1_SOURCE - | RCAR_DU_FEATURE_INTERLACED - | RCAR_DU_FEATURE_TVM_SYNC, - .quirks = RCAR_DU_QUIRK_H3_ES1_PCLK_STABILITY - | RCAR_DU_QUIRK_H3_ES1_PLL, - .channels_mask = BIT(3) | BIT(2) | BIT(1) | BIT(0), - .routes = { - /* - * R8A7795 has one RGB output, two HDMI outputs and one - * LVDS output. - */ - [RCAR_DU_OUTPUT_DPAD0] = { - .possible_crtcs = BIT(3), - .port = 0, - }, - [RCAR_DU_OUTPUT_HDMI0] = { - .possible_crtcs = BIT(1), - .port = 1, - }, - [RCAR_DU_OUTPUT_HDMI1] = { - .possible_crtcs = BIT(2), - .port = 2, - }, - [RCAR_DU_OUTPUT_LVDS0] = { - .possible_crtcs = BIT(0), - .port = 3, - }, - }, - .num_lvds = 1, - .num_rpf = 5, - .dpll_mask = BIT(2) | BIT(1), -}; - static const struct rcar_du_device_info rcar_du_r8a7796_info = { .gen = 3, .features = RCAR_DU_FEATURE_CRTC_IRQ @@ -614,11 +576,6 @@ static const struct of_device_id rcar_du_of_table[] = { MODULE_DEVICE_TABLE(of, rcar_du_of_table); -static const struct soc_device_attribute rcar_du_soc_table[] = { - { .soc_id = "r8a7795", .revision = "ES1.*", .data = &rcar_du_r8a7795_es1_info }, - { /* sentinel */ } -}; - const char *rcar_du_output_name(enum rcar_du_output output) { static const char * const names[] = { @@ -704,7 +661,6 @@ static void rcar_du_shutdown(struct platform_device *pdev) static int rcar_du_probe(struct platform_device *pdev) { - const struct soc_device_attribute *soc_attr; struct rcar_du_device *rcdu; unsigned int mask; int ret; @@ -722,10 +678,6 @@ static int rcar_du_probe(struct platform_device *pdev) rcdu->info = of_device_get_match_data(rcdu->dev); - soc_attr = soc_device_match(rcar_du_soc_table); - if (soc_attr) - rcdu->info = soc_attr->data; - platform_set_drvdata(pdev, rcdu); /* I/O resources */ diff --git a/drivers/gpu/drm/rcar-du/rcar_du_drv.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_drv.h similarity index 96% rename from drivers/gpu/drm/rcar-du/rcar_du_drv.h rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_drv.h index acc3673fefe181ad9d905815747190a6acb4da4e..5cfa2bb7ad93dc2453a361934bb48406b1605a50 100644 --- a/drivers/gpu/drm/rcar-du/rcar_du_drv.h +++ b/drivers/gpu/drm/renesas/rcar-du/rcar_du_drv.h @@ -34,8 +34,6 @@ struct rcar_du_device; #define RCAR_DU_FEATURE_NO_BLENDING BIT(5) /* PnMR.SPIM does not have ALP nor EOR bits */ #define RCAR_DU_QUIRK_ALIGN_128B BIT(0) /* Align pitches to 128 bytes */ -#define RCAR_DU_QUIRK_H3_ES1_PCLK_STABILITY BIT(1) /* H3 ES1 has pclk stability issue */ -#define RCAR_DU_QUIRK_H3_ES1_PLL BIT(2) /* H3 ES1 PLL setup differs from non-ES1 */ enum rcar_du_output { RCAR_DU_OUTPUT_DPAD0, diff --git a/drivers/gpu/drm/rcar-du/rcar_du_encoder.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_encoder.c similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_du_encoder.c rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_encoder.c diff --git a/drivers/gpu/drm/rcar-du/rcar_du_encoder.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_encoder.h similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_du_encoder.h rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_encoder.h diff --git a/drivers/gpu/drm/rcar-du/rcar_du_group.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_group.c similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_du_group.c rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_group.c diff --git a/drivers/gpu/drm/rcar-du/rcar_du_group.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_group.h similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_du_group.h rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_group.h diff --git a/drivers/gpu/drm/rcar-du/rcar_du_kms.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_kms.c similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_du_kms.c rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_kms.c diff --git a/drivers/gpu/drm/rcar-du/rcar_du_kms.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_kms.h similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_du_kms.h rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_kms.h diff --git a/drivers/gpu/drm/rcar-du/rcar_du_plane.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_plane.c similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_du_plane.c rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_plane.c diff --git a/drivers/gpu/drm/rcar-du/rcar_du_plane.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_plane.h similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_du_plane.h rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_plane.h diff --git a/drivers/gpu/drm/rcar-du/rcar_du_regs.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_regs.h similarity index 99% rename from drivers/gpu/drm/rcar-du/rcar_du_regs.h rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_regs.h index 6c750fab6ebb736ea1a342a414cc1b11c6cfba1a..391de6661d8bc4114ef47cc7883edc41ca63da38 100644 --- a/drivers/gpu/drm/rcar-du/rcar_du_regs.h +++ b/drivers/gpu/drm/renesas/rcar-du/rcar_du_regs.h @@ -283,8 +283,7 @@ #define DPLLCR 0x20044 #define DPLLCR_CODE (0x95 << 24) #define DPLLCR_PLCS1 (1 << 23) -#define DPLLCR_PLCS0_PLL (1 << 21) -#define DPLLCR_PLCS0_H3ES1X_PLL1 (1 << 20) +#define DPLLCR_PLCS0 (1 << 21) #define DPLLCR_CLKE (1 << 18) #define DPLLCR_FDPLL(n) ((n) << 12) #define DPLLCR_N(n) ((n) << 5) diff --git a/drivers/gpu/drm/rcar-du/rcar_du_vsp.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_vsp.c similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_du_vsp.c rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_vsp.c diff --git a/drivers/gpu/drm/rcar-du/rcar_du_vsp.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_vsp.h similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_du_vsp.h rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_vsp.h diff --git a/drivers/gpu/drm/rcar-du/rcar_du_writeback.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_writeback.c similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_du_writeback.c rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_writeback.c diff --git a/drivers/gpu/drm/rcar-du/rcar_du_writeback.h b/drivers/gpu/drm/renesas/rcar-du/rcar_du_writeback.h similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_du_writeback.h rename to drivers/gpu/drm/renesas/rcar-du/rcar_du_writeback.h diff --git a/drivers/gpu/drm/rcar-du/rcar_dw_hdmi.c b/drivers/gpu/drm/renesas/rcar-du/rcar_dw_hdmi.c similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_dw_hdmi.c rename to drivers/gpu/drm/renesas/rcar-du/rcar_dw_hdmi.c diff --git a/drivers/gpu/drm/rcar-du/rcar_lvds.c b/drivers/gpu/drm/renesas/rcar-du/rcar_lvds.c similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_lvds.c rename to drivers/gpu/drm/renesas/rcar-du/rcar_lvds.c diff --git a/drivers/gpu/drm/rcar-du/rcar_lvds.h b/drivers/gpu/drm/renesas/rcar-du/rcar_lvds.h similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_lvds.h rename to drivers/gpu/drm/renesas/rcar-du/rcar_lvds.h diff --git a/drivers/gpu/drm/rcar-du/rcar_lvds_regs.h b/drivers/gpu/drm/renesas/rcar-du/rcar_lvds_regs.h similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_lvds_regs.h rename to drivers/gpu/drm/renesas/rcar-du/rcar_lvds_regs.h diff --git a/drivers/gpu/drm/rcar-du/rcar_mipi_dsi.c b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_mipi_dsi.c rename to drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c diff --git a/drivers/gpu/drm/rcar-du/rcar_mipi_dsi.h b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.h similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_mipi_dsi.h rename to drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.h diff --git a/drivers/gpu/drm/rcar-du/rcar_mipi_dsi_regs.h b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi_regs.h similarity index 100% rename from drivers/gpu/drm/rcar-du/rcar_mipi_dsi_regs.h rename to drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi_regs.h diff --git a/drivers/gpu/drm/rcar-du/rzg2l_mipi_dsi.c b/drivers/gpu/drm/renesas/rcar-du/rzg2l_mipi_dsi.c similarity index 100% rename from drivers/gpu/drm/rcar-du/rzg2l_mipi_dsi.c rename to drivers/gpu/drm/renesas/rcar-du/rzg2l_mipi_dsi.c diff --git a/drivers/gpu/drm/rcar-du/rzg2l_mipi_dsi_regs.h b/drivers/gpu/drm/renesas/rcar-du/rzg2l_mipi_dsi_regs.h similarity index 100% rename from drivers/gpu/drm/rcar-du/rzg2l_mipi_dsi_regs.h rename to drivers/gpu/drm/renesas/rcar-du/rzg2l_mipi_dsi_regs.h diff --git a/drivers/gpu/drm/shmobile/Kconfig b/drivers/gpu/drm/renesas/shmobile/Kconfig similarity index 100% rename from drivers/gpu/drm/shmobile/Kconfig rename to drivers/gpu/drm/renesas/shmobile/Kconfig diff --git a/drivers/gpu/drm/shmobile/Makefile b/drivers/gpu/drm/renesas/shmobile/Makefile similarity index 100% rename from drivers/gpu/drm/shmobile/Makefile rename to drivers/gpu/drm/renesas/shmobile/Makefile diff --git a/drivers/gpu/drm/shmobile/shmob_drm_backlight.c b/drivers/gpu/drm/renesas/shmobile/shmob_drm_backlight.c similarity index 100% rename from drivers/gpu/drm/shmobile/shmob_drm_backlight.c rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_backlight.c diff --git a/drivers/gpu/drm/shmobile/shmob_drm_backlight.h b/drivers/gpu/drm/renesas/shmobile/shmob_drm_backlight.h similarity index 100% rename from drivers/gpu/drm/shmobile/shmob_drm_backlight.h rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_backlight.h diff --git a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c b/drivers/gpu/drm/renesas/shmobile/shmob_drm_crtc.c similarity index 100% rename from drivers/gpu/drm/shmobile/shmob_drm_crtc.c rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_crtc.c diff --git a/drivers/gpu/drm/shmobile/shmob_drm_crtc.h b/drivers/gpu/drm/renesas/shmobile/shmob_drm_crtc.h similarity index 100% rename from drivers/gpu/drm/shmobile/shmob_drm_crtc.h rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_crtc.h diff --git a/drivers/gpu/drm/shmobile/shmob_drm_drv.c b/drivers/gpu/drm/renesas/shmobile/shmob_drm_drv.c similarity index 100% rename from drivers/gpu/drm/shmobile/shmob_drm_drv.c rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_drv.c diff --git a/drivers/gpu/drm/shmobile/shmob_drm_drv.h b/drivers/gpu/drm/renesas/shmobile/shmob_drm_drv.h similarity index 100% rename from drivers/gpu/drm/shmobile/shmob_drm_drv.h rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_drv.h diff --git a/drivers/gpu/drm/shmobile/shmob_drm_kms.c b/drivers/gpu/drm/renesas/shmobile/shmob_drm_kms.c similarity index 100% rename from drivers/gpu/drm/shmobile/shmob_drm_kms.c rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_kms.c diff --git a/drivers/gpu/drm/shmobile/shmob_drm_kms.h b/drivers/gpu/drm/renesas/shmobile/shmob_drm_kms.h similarity index 100% rename from drivers/gpu/drm/shmobile/shmob_drm_kms.h rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_kms.h diff --git a/drivers/gpu/drm/shmobile/shmob_drm_plane.c b/drivers/gpu/drm/renesas/shmobile/shmob_drm_plane.c similarity index 100% rename from drivers/gpu/drm/shmobile/shmob_drm_plane.c rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_plane.c diff --git a/drivers/gpu/drm/shmobile/shmob_drm_plane.h b/drivers/gpu/drm/renesas/shmobile/shmob_drm_plane.h similarity index 100% rename from drivers/gpu/drm/shmobile/shmob_drm_plane.h rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_plane.h diff --git a/drivers/gpu/drm/shmobile/shmob_drm_regs.h b/drivers/gpu/drm/renesas/shmobile/shmob_drm_regs.h similarity index 100% rename from drivers/gpu/drm/shmobile/shmob_drm_regs.h rename to drivers/gpu/drm/renesas/shmobile/shmob_drm_regs.h diff --git a/drivers/gpu/drm/scheduler/sched_entity.c b/drivers/gpu/drm/scheduler/sched_entity.c index 68e807ae136ad819a1cf80dbe2d4f6f19e199f85..b2bbc8a68b30562d14b53cf8fc2e5bcf3bb3c75d 100644 --- a/drivers/gpu/drm/scheduler/sched_entity.c +++ b/drivers/gpu/drm/scheduler/sched_entity.c @@ -384,7 +384,7 @@ static bool drm_sched_entity_add_dependency_cb(struct drm_sched_entity *entity) } s_fence = to_drm_sched_fence(fence); - if (s_fence && s_fence->sched == sched && + if (!fence->error && s_fence && s_fence->sched == sched && !test_bit(DRM_SCHED_FENCE_DONT_PIPELINE, &fence->flags)) { /* diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c index 394010a608217df276f53c5d2950e6d80f417e9b..7b2bfc10c1a54c497020451841a5ffafef512399 100644 --- a/drivers/gpu/drm/scheduler/sched_main.c +++ b/drivers/gpu/drm/scheduler/sched_main.c @@ -1142,9 +1142,6 @@ void drm_sched_fini(struct drm_gpu_scheduler *sched) for (i = DRM_SCHED_PRIORITY_COUNT - 1; i >= DRM_SCHED_PRIORITY_MIN; i--) { struct drm_sched_rq *rq = &sched->sched_rq[i]; - if (!rq) - continue; - spin_lock(&rq->lock); list_for_each_entry(s_entity, &rq->entities, list) /* diff --git a/drivers/gpu/drm/ttm/ttm_device.c b/drivers/gpu/drm/ttm/ttm_device.c index df4cf5468e7fd648b23212422ddbb283fd5b4801..7726a72befc5446bd73d9da51b7ccd8175c0ab5d 100644 --- a/drivers/gpu/drm/ttm/ttm_device.c +++ b/drivers/gpu/drm/ttm/ttm_device.c @@ -213,7 +213,7 @@ int ttm_device_init(struct ttm_device *bdev, const struct ttm_device_funcs *func bdev->funcs = funcs; ttm_sys_man_init(bdev); - ttm_pool_init(&bdev->pool, dev, use_dma_alloc, use_dma32); + ttm_pool_init(&bdev->pool, dev, NUMA_NO_NODE, use_dma_alloc, use_dma32); bdev->vma_manager = vma_manager; spin_lock_init(&bdev->lru_lock); diff --git a/drivers/gpu/drm/ttm/ttm_pool.c b/drivers/gpu/drm/ttm/ttm_pool.c index 4db3982057be8d18fc20a423c91d4ab708bd7514..cddb9151d20f4465bfc269e62c3072ad51a990c0 100644 --- a/drivers/gpu/drm/ttm/ttm_pool.c +++ b/drivers/gpu/drm/ttm/ttm_pool.c @@ -93,7 +93,7 @@ static struct page *ttm_pool_alloc_page(struct ttm_pool *pool, gfp_t gfp_flags, __GFP_KSWAPD_RECLAIM; if (!pool->use_dma_alloc) { - p = alloc_pages(gfp_flags, order); + p = alloc_pages_node(pool->nid, gfp_flags, order); if (p) p->private = order; return p; @@ -287,7 +287,7 @@ static struct ttm_pool_type *ttm_pool_select_type(struct ttm_pool *pool, enum ttm_caching caching, unsigned int order) { - if (pool->use_dma_alloc) + if (pool->use_dma_alloc || pool->nid != NUMA_NO_NODE) return &pool->caching[caching].orders[order]; #ifdef CONFIG_X86 @@ -545,29 +545,32 @@ EXPORT_SYMBOL(ttm_pool_free); * * @pool: the pool to initialize * @dev: device for DMA allocations and mappings + * @nid: NUMA node to use for allocations * @use_dma_alloc: true if coherent DMA alloc should be used * @use_dma32: true if GFP_DMA32 should be used * * Initialize the pool and its pool types. */ void ttm_pool_init(struct ttm_pool *pool, struct device *dev, - bool use_dma_alloc, bool use_dma32) + int nid, bool use_dma_alloc, bool use_dma32) { unsigned int i, j; WARN_ON(!dev && use_dma_alloc); pool->dev = dev; + pool->nid = nid; pool->use_dma_alloc = use_dma_alloc; pool->use_dma32 = use_dma32; - if (use_dma_alloc) { + if (use_dma_alloc || nid != NUMA_NO_NODE) { for (i = 0; i < TTM_NUM_CACHING_TYPES; ++i) for (j = 0; j <= MAX_ORDER; ++j) ttm_pool_type_init(&pool->caching[i].orders[j], pool, i, j); } } +EXPORT_SYMBOL(ttm_pool_init); /** * ttm_pool_fini - Cleanup a pool @@ -581,7 +584,7 @@ void ttm_pool_fini(struct ttm_pool *pool) { unsigned int i, j; - if (pool->use_dma_alloc) { + if (pool->use_dma_alloc || pool->nid != NUMA_NO_NODE) { for (i = 0; i < TTM_NUM_CACHING_TYPES; ++i) for (j = 0; j <= MAX_ORDER; ++j) ttm_pool_type_fini(&pool->caching[i].orders[j]); @@ -592,6 +595,7 @@ void ttm_pool_fini(struct ttm_pool *pool) */ synchronize_shrinkers(); } +EXPORT_SYMBOL(ttm_pool_fini); /* As long as pages are available make sure to release at least one */ static unsigned long ttm_pool_shrinker_scan(struct shrinker *shrink, diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c index 1ce4b36ab33be191f1d238e5799e2bda5c5969d4..e0a77671edd6c155329119e498e271a520a4eaab 100644 --- a/drivers/gpu/drm/ttm/ttm_tt.c +++ b/drivers/gpu/drm/ttm/ttm_tt.c @@ -449,3 +449,9 @@ ttm_kmap_iter_tt_init(struct ttm_kmap_iter_tt *iter_tt, return &iter_tt->base; } EXPORT_SYMBOL(ttm_kmap_iter_tt_init); + +unsigned long ttm_tt_pages_limit(void) +{ + return ttm_pages_limit; +} +EXPORT_SYMBOL(ttm_tt_pages_limit); diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c index 1e55576089376e623effa0e04a7d89c630d62d13..56a950e6d7775db14ab5e8278b77750f12854d5c 100644 --- a/drivers/gpu/drm/vc4/vc4_hdmi.c +++ b/drivers/gpu/drm/vc4/vc4_hdmi.c @@ -742,7 +742,7 @@ static int vc4_hdmi_connector_init(struct drm_device *dev, if (ret) return ret; - ret = drm_mode_create_hdmi_colorspace_property(connector); + ret = drm_mode_create_hdmi_colorspace_property(connector, 0); if (ret) return ret; diff --git a/drivers/hid/hid-google-hammer.c b/drivers/hid/hid-google-hammer.c index 7ae5f27df54dd6731b1dcef653a6153323cdbfc2..c6bdb9c4ef3e004cf27fba53b6f5e0eccbe21b55 100644 --- a/drivers/hid/hid-google-hammer.c +++ b/drivers/hid/hid-google-hammer.c @@ -586,6 +586,8 @@ static const struct hid_device_id hammer_devices[] = { USB_VENDOR_ID_GOOGLE, USB_DEVICE_ID_GOOGLE_EEL) }, { HID_DEVICE(BUS_USB, HID_GROUP_GENERIC, USB_VENDOR_ID_GOOGLE, USB_DEVICE_ID_GOOGLE_HAMMER) }, + { HID_DEVICE(BUS_USB, HID_GROUP_GENERIC, + USB_VENDOR_ID_GOOGLE, USB_DEVICE_ID_GOOGLE_JEWEL) }, { HID_DEVICE(BUS_USB, HID_GROUP_GENERIC, USB_VENDOR_ID_GOOGLE, USB_DEVICE_ID_GOOGLE_MAGNEMITE) }, { HID_DEVICE(BUS_USB, HID_GROUP_GENERIC, diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h index d79e946acdcbed9fe7afe132ada5366b0a784dbb..5d29abac2300e40c75b0b6eedb803969cac1a84a 100644 --- a/drivers/hid/hid-ids.h +++ b/drivers/hid/hid-ids.h @@ -529,6 +529,7 @@ #define USB_DEVICE_ID_GOOGLE_MOONBALL 0x5044 #define USB_DEVICE_ID_GOOGLE_DON 0x5050 #define USB_DEVICE_ID_GOOGLE_EEL 0x5057 +#define USB_DEVICE_ID_GOOGLE_JEWEL 0x5061 #define USB_VENDOR_ID_GOTOP 0x08f2 #define USB_DEVICE_ID_SUPER_Q2 0x007f diff --git a/drivers/hid/hid-logitech-hidpp.c b/drivers/hid/hid-logitech-hidpp.c index 0fcfd85fea0feff39a650348d31380daa0f26706..5e1a412fd28fa125ba84ee35ce570cc4827d755f 100644 --- a/drivers/hid/hid-logitech-hidpp.c +++ b/drivers/hid/hid-logitech-hidpp.c @@ -286,7 +286,7 @@ static int hidpp_send_message_sync(struct hidpp_device *hidpp, struct hidpp_report *message, struct hidpp_report *response) { - int ret; + int ret = -1; int max_retries = 3; mutex_lock(&hidpp->send_mutex); @@ -300,13 +300,13 @@ static int hidpp_send_message_sync(struct hidpp_device *hidpp, */ *response = *message; - for (; max_retries != 0; max_retries--) { + for (; max_retries != 0 && ret; max_retries--) { ret = __hidpp_send_report(hidpp->hid_dev, message); if (ret) { dbg_hid("__hidpp_send_report returned err: %d\n", ret); memset(response, 0, sizeof(struct hidpp_report)); - goto exit; + break; } if (!wait_event_timeout(hidpp->wait, hidpp->answer_available, @@ -314,13 +314,14 @@ static int hidpp_send_message_sync(struct hidpp_device *hidpp, dbg_hid("%s:timeout waiting for response\n", __func__); memset(response, 0, sizeof(struct hidpp_report)); ret = -ETIMEDOUT; + break; } if (response->report_id == REPORT_ID_HIDPP_SHORT && response->rap.sub_id == HIDPP_ERROR) { ret = response->rap.params[1]; dbg_hid("%s:got hidpp error %02X\n", __func__, ret); - goto exit; + break; } if ((response->report_id == REPORT_ID_HIDPP_LONG || @@ -329,13 +330,12 @@ static int hidpp_send_message_sync(struct hidpp_device *hidpp, ret = response->fap.params[1]; if (ret != HIDPP20_ERROR_BUSY) { dbg_hid("%s:got hidpp 2.0 error %02X\n", __func__, ret); - goto exit; + break; } dbg_hid("%s:got busy hidpp 2.0 error %02X, retrying\n", __func__, ret); } } -exit: mutex_unlock(&hidpp->send_mutex); return ret; diff --git a/drivers/hid/wacom_sys.c b/drivers/hid/wacom_sys.c index 8214896adadaddcb17ebccfc644d2c0f8ad496f3..76e5353aca0c7440253d407ceacd41513e18158a 100644 --- a/drivers/hid/wacom_sys.c +++ b/drivers/hid/wacom_sys.c @@ -2224,7 +2224,9 @@ static void wacom_update_name(struct wacom *wacom, const char *suffix) } else if (strstr(product_name, "Wacom") || strstr(product_name, "wacom") || strstr(product_name, "WACOM")) { - strscpy(name, product_name, sizeof(name)); + if (strscpy(name, product_name, sizeof(name)) < 0) { + hid_warn(wacom->hdev, "String overflow while assembling device name"); + } } else { snprintf(name, sizeof(name), "Wacom %s", product_name); } @@ -2242,7 +2244,9 @@ static void wacom_update_name(struct wacom *wacom, const char *suffix) if (name[strlen(name)-1] == ' ') name[strlen(name)-1] = '\0'; } else { - strscpy(name, features->name, sizeof(name)); + if (strscpy(name, features->name, sizeof(name)) < 0) { + hid_warn(wacom->hdev, "String overflow while assembling device name"); + } } snprintf(wacom_wac->name, sizeof(wacom_wac->name), "%s%s", @@ -2410,8 +2414,13 @@ static int wacom_parse_and_register(struct wacom *wacom, bool wireless) goto fail_quirks; } - if (features->device_type & WACOM_DEVICETYPE_WL_MONITOR) + if (features->device_type & WACOM_DEVICETYPE_WL_MONITOR) { error = hid_hw_open(hdev); + if (error) { + hid_err(hdev, "hw open failed\n"); + goto fail_quirks; + } + } wacom_set_shared_values(wacom_wac); devres_close_group(&hdev->dev, wacom); @@ -2500,8 +2509,10 @@ static void wacom_wireless_work(struct work_struct *work) goto fail; } - strscpy(wacom_wac->name, wacom_wac1->name, - sizeof(wacom_wac->name)); + if (strscpy(wacom_wac->name, wacom_wac1->name, + sizeof(wacom_wac->name)) < 0) { + hid_warn(wacom->hdev, "String overflow while assembling device name"); + } } return; diff --git a/drivers/hid/wacom_wac.c b/drivers/hid/wacom_wac.c index dc0f7d9a992c10977a5bc3bcaf100cb5c548129b..2ccf838371343da1628bec555896caba7843d2eb 100644 --- a/drivers/hid/wacom_wac.c +++ b/drivers/hid/wacom_wac.c @@ -831,7 +831,7 @@ static int wacom_intuos_inout(struct wacom_wac *wacom) /* Enter report */ if ((data[1] & 0xfc) == 0xc0) { /* serial number of the tool */ - wacom->serial[idx] = ((data[3] & 0x0f) << 28) + + wacom->serial[idx] = ((__u64)(data[3] & 0x0f) << 28) + (data[4] << 20) + (data[5] << 12) + (data[6] << 4) + (data[7] >> 4); diff --git a/drivers/hwtracing/coresight/coresight-etm-perf.c b/drivers/hwtracing/coresight/coresight-etm-perf.c index 711f451b69469404cc4a0835c266e9854860e3be..89e8ed214ea4967620c20ad1c1a1e39f9d068400 100644 --- a/drivers/hwtracing/coresight/coresight-etm-perf.c +++ b/drivers/hwtracing/coresight/coresight-etm-perf.c @@ -402,6 +402,7 @@ static void *etm_setup_aux(struct perf_event *event, void **pages, trace_id = coresight_trace_id_get_cpu_id(cpu); if (!IS_VALID_CS_TRACE_ID(trace_id)) { cpumask_clear_cpu(cpu, mask); + coresight_release_path(path); continue; } diff --git a/drivers/hwtracing/coresight/coresight-tmc-etr.c b/drivers/hwtracing/coresight/coresight-tmc-etr.c index 918d461fcf4a68637cea80d6079ebb2e0f8eccb3..eaa296ced167893770a3695a1e9b8fb24098d861 100644 --- a/drivers/hwtracing/coresight/coresight-tmc-etr.c +++ b/drivers/hwtracing/coresight/coresight-tmc-etr.c @@ -942,7 +942,7 @@ tmc_etr_buf_insert_barrier_packet(struct etr_buf *etr_buf, u64 offset) len = tmc_etr_buf_get_data(etr_buf, offset, CORESIGHT_BARRIER_PKT_SIZE, &bufp); - if (WARN_ON(len < CORESIGHT_BARRIER_PKT_SIZE)) + if (WARN_ON(len < 0 || len < CORESIGHT_BARRIER_PKT_SIZE)) return -EINVAL; coresight_insert_barrier_packet(bufp); return offset + CORESIGHT_BARRIER_PKT_SIZE; diff --git a/drivers/i2c/busses/i2c-designware-core.h b/drivers/i2c/busses/i2c-designware-core.h index c5d87aae39c666db553afe3f57c18b27288f462d..bf23bfb51aead1e45c00cb2e44e28d609ecb40d1 100644 --- a/drivers/i2c/busses/i2c-designware-core.h +++ b/drivers/i2c/busses/i2c-designware-core.h @@ -40,6 +40,7 @@ #define DW_IC_CON_BUS_CLEAR_CTRL BIT(11) #define DW_IC_DATA_CMD_DAT GENMASK(7, 0) +#define DW_IC_DATA_CMD_FIRST_DATA_BYTE BIT(11) /* * Registers offset diff --git a/drivers/i2c/busses/i2c-designware-slave.c b/drivers/i2c/busses/i2c-designware-slave.c index cec25054bb244fadd946d36dc234d499aa831260..2e079cf20bb5b40c09004bdc02a8124352fab958 100644 --- a/drivers/i2c/busses/i2c-designware-slave.c +++ b/drivers/i2c/busses/i2c-designware-slave.c @@ -176,6 +176,10 @@ static irqreturn_t i2c_dw_isr_slave(int this_irq, void *dev_id) do { regmap_read(dev->map, DW_IC_DATA_CMD, &tmp); + if (tmp & DW_IC_DATA_CMD_FIRST_DATA_BYTE) + i2c_slave_event(dev->slave, + I2C_SLAVE_WRITE_REQUESTED, + &val); val = tmp; i2c_slave_event(dev->slave, I2C_SLAVE_WRITE_RECEIVED, &val); diff --git a/drivers/i2c/busses/i2c-img-scb.c b/drivers/i2c/busses/i2c-img-scb.c index 8e987945ed4503d641bd44bcdf2fe36e6e078c66..39c479f96eb58564b75e91f4650cf31e1345433f 100644 --- a/drivers/i2c/busses/i2c-img-scb.c +++ b/drivers/i2c/busses/i2c-img-scb.c @@ -257,7 +257,7 @@ #define IMG_I2C_TIMEOUT (msecs_to_jiffies(1000)) /* - * Worst incs are 1 (innacurate) and 16*256 (irregular). + * Worst incs are 1 (inaccurate) and 16*256 (irregular). * So a sensible inc is the logarithmic mean: 64 (2^6), which is * in the middle of the valid range (0-127). */ diff --git a/drivers/i2c/busses/i2c-mchp-pci1xxxx.c b/drivers/i2c/busses/i2c-mchp-pci1xxxx.c index b21ffd6df9276b34ec14267d6db6fe9cd93c9c66..5ef136c3ecb1262144f36be9e54336740d9e5528 100644 --- a/drivers/i2c/busses/i2c-mchp-pci1xxxx.c +++ b/drivers/i2c/busses/i2c-mchp-pci1xxxx.c @@ -1118,8 +1118,10 @@ static int pci1xxxx_i2c_resume(struct device *dev) static DEFINE_SIMPLE_DEV_PM_OPS(pci1xxxx_i2c_pm_ops, pci1xxxx_i2c_suspend, pci1xxxx_i2c_resume); -static void pci1xxxx_i2c_shutdown(struct pci1xxxx_i2c *i2c) +static void pci1xxxx_i2c_shutdown(void *data) { + struct pci1xxxx_i2c *i2c = data; + pci1xxxx_i2c_config_padctrl(i2c, false); pci1xxxx_i2c_configure_core_reg(i2c, false); } @@ -1156,7 +1158,7 @@ static int pci1xxxx_i2c_probe_pci(struct pci_dev *pdev, init_completion(&i2c->i2c_xfer_done); pci1xxxx_i2c_init(i2c); - ret = devm_add_action(dev, (void (*)(void *))pci1xxxx_i2c_shutdown, i2c); + ret = devm_add_action(dev, pci1xxxx_i2c_shutdown, i2c); if (ret) return ret; diff --git a/drivers/i2c/busses/i2c-mv64xxx.c b/drivers/i2c/busses/i2c-mv64xxx.c index 047dfef7a657750ef7ef75646b026d4ea3c168b9..878c076ebdc6b2ab5e30f8441094db198da28fcd 100644 --- a/drivers/i2c/busses/i2c-mv64xxx.c +++ b/drivers/i2c/busses/i2c-mv64xxx.c @@ -520,6 +520,17 @@ mv64xxx_i2c_intr(int irq, void *dev_id) while (readl(drv_data->reg_base + drv_data->reg_offsets.control) & MV64XXX_I2C_REG_CONTROL_IFLG) { + /* + * It seems that sometime the controller updates the status + * register only after it asserts IFLG in control register. + * This may result in weird bugs when in atomic mode. A delay + * of 100 ns before reading the status register solves this + * issue. This bug does not seem to appear when using + * interrupts. + */ + if (drv_data->atomic) + ndelay(100); + status = readl(drv_data->reg_base + drv_data->reg_offsets.status); mv64xxx_i2c_fsm(drv_data, status); mv64xxx_i2c_do_action(drv_data); diff --git a/drivers/i2c/busses/i2c-sprd.c b/drivers/i2c/busses/i2c-sprd.c index 4fe15cd78907e065a496724a0b84c7c644c21eb4..ffc54fbf814dde0a8a9ada780d811ea6138edbe6 100644 --- a/drivers/i2c/busses/i2c-sprd.c +++ b/drivers/i2c/busses/i2c-sprd.c @@ -576,12 +576,14 @@ static int sprd_i2c_remove(struct platform_device *pdev) struct sprd_i2c *i2c_dev = platform_get_drvdata(pdev); int ret; - ret = pm_runtime_resume_and_get(i2c_dev->dev); + ret = pm_runtime_get_sync(i2c_dev->dev); if (ret < 0) - return ret; + dev_err(&pdev->dev, "Failed to resume device (%pe)\n", ERR_PTR(ret)); i2c_del_adapter(&i2c_dev->adap); - clk_disable_unprepare(i2c_dev->clk); + + if (ret >= 0) + clk_disable_unprepare(i2c_dev->clk); pm_runtime_put_noidle(i2c_dev->dev); pm_runtime_disable(i2c_dev->dev); diff --git a/drivers/iio/accel/kionix-kx022a.c b/drivers/iio/accel/kionix-kx022a.c index f98393d746668f0ebbf3ba7d5d909adf25b1d680..b8636fa8eaeb68d740300d75e06365f9e887dd25 100644 --- a/drivers/iio/accel/kionix-kx022a.c +++ b/drivers/iio/accel/kionix-kx022a.c @@ -1048,7 +1048,7 @@ int kx022a_probe_internal(struct device *dev) data->ien_reg = KX022A_REG_INC4; } else { irq = fwnode_irq_get_byname(fwnode, "INT2"); - if (irq <= 0) + if (irq < 0) return dev_err_probe(dev, irq, "No suitable IRQ\n"); data->inc_reg = KX022A_REG_INC5; diff --git a/drivers/iio/accel/st_accel_core.c b/drivers/iio/accel/st_accel_core.c index 5f7d81b44b1d28ce2eab297ec16c459bb9feed33..282e539157f87cb223c2ad985d964b0cd5a7a0b3 100644 --- a/drivers/iio/accel/st_accel_core.c +++ b/drivers/iio/accel/st_accel_core.c @@ -1291,12 +1291,12 @@ static int apply_acpi_orientation(struct iio_dev *indio_dev) adev = ACPI_COMPANION(indio_dev->dev.parent); if (!adev) - return 0; + return -ENXIO; /* Read _ONT data, which should be a package of 6 integers. */ status = acpi_evaluate_object(adev->handle, "_ONT", NULL, &buffer); if (status == AE_NOT_FOUND) { - return 0; + return -ENXIO; } else if (ACPI_FAILURE(status)) { dev_warn(&indio_dev->dev, "failed to execute _ONT: %d\n", status); diff --git a/drivers/iio/adc/ad4130.c b/drivers/iio/adc/ad4130.c index 38394341fd6e7401a94662fd5524eb512ea402b4..5a5dd5e87ffc4481c9776d509d32c319ba83135a 100644 --- a/drivers/iio/adc/ad4130.c +++ b/drivers/iio/adc/ad4130.c @@ -1817,6 +1817,11 @@ static const struct clk_ops ad4130_int_clk_ops = { .unprepare = ad4130_int_clk_unprepare, }; +static void ad4130_clk_del_provider(void *of_node) +{ + of_clk_del_provider(of_node); +} + static int ad4130_setup_int_clk(struct ad4130_state *st) { struct device *dev = &st->spi->dev; @@ -1824,6 +1829,7 @@ static int ad4130_setup_int_clk(struct ad4130_state *st) struct clk_init_data init; const char *clk_name; struct clk *clk; + int ret; if (st->int_pin_sel == AD4130_INT_PIN_CLK || st->mclk_sel != AD4130_MCLK_76_8KHZ) @@ -1843,7 +1849,11 @@ static int ad4130_setup_int_clk(struct ad4130_state *st) if (IS_ERR(clk)) return PTR_ERR(clk); - return of_clk_add_provider(of_node, of_clk_src_simple_get, clk); + ret = of_clk_add_provider(of_node, of_clk_src_simple_get, clk); + if (ret) + return ret; + + return devm_add_action_or_reset(dev, ad4130_clk_del_provider, of_node); } static int ad4130_setup(struct iio_dev *indio_dev) diff --git a/drivers/iio/adc/ad7192.c b/drivers/iio/adc/ad7192.c index 55a6ab5910160725415b004ab523f453c97b5337..99bb604b78c8cf904271b1832bde8a91a1c13bc3 100644 --- a/drivers/iio/adc/ad7192.c +++ b/drivers/iio/adc/ad7192.c @@ -897,10 +897,6 @@ static const struct iio_info ad7195_info = { __AD719x_CHANNEL(_si, _channel1, -1, _address, NULL, IIO_VOLTAGE, \ BIT(IIO_CHAN_INFO_SCALE), ad7192_calibsys_ext_info) -#define AD719x_SHORTED_CHANNEL(_si, _channel1, _address) \ - __AD719x_CHANNEL(_si, _channel1, -1, _address, "shorted", IIO_VOLTAGE, \ - BIT(IIO_CHAN_INFO_SCALE), ad7192_calibsys_ext_info) - #define AD719x_TEMP_CHANNEL(_si, _address) \ __AD719x_CHANNEL(_si, 0, -1, _address, NULL, IIO_TEMP, 0, NULL) @@ -908,7 +904,7 @@ static const struct iio_chan_spec ad7192_channels[] = { AD719x_DIFF_CHANNEL(0, 1, 2, AD7192_CH_AIN1P_AIN2M), AD719x_DIFF_CHANNEL(1, 3, 4, AD7192_CH_AIN3P_AIN4M), AD719x_TEMP_CHANNEL(2, AD7192_CH_TEMP), - AD719x_SHORTED_CHANNEL(3, 2, AD7192_CH_AIN2P_AIN2M), + AD719x_DIFF_CHANNEL(3, 2, 2, AD7192_CH_AIN2P_AIN2M), AD719x_CHANNEL(4, 1, AD7192_CH_AIN1), AD719x_CHANNEL(5, 2, AD7192_CH_AIN2), AD719x_CHANNEL(6, 3, AD7192_CH_AIN3), @@ -922,7 +918,7 @@ static const struct iio_chan_spec ad7193_channels[] = { AD719x_DIFF_CHANNEL(2, 5, 6, AD7193_CH_AIN5P_AIN6M), AD719x_DIFF_CHANNEL(3, 7, 8, AD7193_CH_AIN7P_AIN8M), AD719x_TEMP_CHANNEL(4, AD7193_CH_TEMP), - AD719x_SHORTED_CHANNEL(5, 2, AD7193_CH_AIN2P_AIN2M), + AD719x_DIFF_CHANNEL(5, 2, 2, AD7193_CH_AIN2P_AIN2M), AD719x_CHANNEL(6, 1, AD7193_CH_AIN1), AD719x_CHANNEL(7, 2, AD7193_CH_AIN2), AD719x_CHANNEL(8, 3, AD7193_CH_AIN3), diff --git a/drivers/iio/adc/ad_sigma_delta.c b/drivers/iio/adc/ad_sigma_delta.c index d8570f620785af8d8b6dfb1a11ee3b63ef579526..7e21928707437baf9a2522f265d72194745205e6 100644 --- a/drivers/iio/adc/ad_sigma_delta.c +++ b/drivers/iio/adc/ad_sigma_delta.c @@ -584,6 +584,10 @@ static int devm_ad_sd_probe_trigger(struct device *dev, struct iio_dev *indio_de init_completion(&sigma_delta->completion); sigma_delta->irq_dis = true; + + /* the IRQ core clears IRQ_DISABLE_UNLAZY flag when freeing an IRQ */ + irq_set_status_flags(sigma_delta->spi->irq, IRQ_DISABLE_UNLAZY); + ret = devm_request_irq(dev, sigma_delta->spi->irq, ad_sd_data_rdy_trig_poll, sigma_delta->info->irq_flags | IRQF_NO_AUTOEN, diff --git a/drivers/iio/adc/imx93_adc.c b/drivers/iio/adc/imx93_adc.c index a775d2e4056718c0f4b83239617e71d26a14cb1d..dce9ec91e4a778db91f866b6adac6d606496b786 100644 --- a/drivers/iio/adc/imx93_adc.c +++ b/drivers/iio/adc/imx93_adc.c @@ -236,8 +236,7 @@ static int imx93_adc_read_raw(struct iio_dev *indio_dev, { struct imx93_adc *adc = iio_priv(indio_dev); struct device *dev = adc->dev; - long ret; - u32 vref_uv; + int ret; switch (mask) { case IIO_CHAN_INFO_RAW: @@ -253,10 +252,10 @@ static int imx93_adc_read_raw(struct iio_dev *indio_dev, return IIO_VAL_INT; case IIO_CHAN_INFO_SCALE: - ret = vref_uv = regulator_get_voltage(adc->vref); + ret = regulator_get_voltage(adc->vref); if (ret < 0) return ret; - *val = vref_uv / 1000; + *val = ret / 1000; *val2 = 12; return IIO_VAL_FRACTIONAL_LOG2; diff --git a/drivers/iio/adc/mt6370-adc.c b/drivers/iio/adc/mt6370-adc.c index bc62e5a9d50d11d42e6791065f2a3266d4dc2b06..0bc112135bca1e5922ae19a7bd591f555e863240 100644 --- a/drivers/iio/adc/mt6370-adc.c +++ b/drivers/iio/adc/mt6370-adc.c @@ -19,6 +19,7 @@ #include +#define MT6370_REG_DEV_INFO 0x100 #define MT6370_REG_CHG_CTRL3 0x113 #define MT6370_REG_CHG_CTRL7 0x117 #define MT6370_REG_CHG_ADC 0x121 @@ -27,6 +28,7 @@ #define MT6370_ADC_START_MASK BIT(0) #define MT6370_ADC_IN_SEL_MASK GENMASK(7, 4) #define MT6370_AICR_ICHG_MASK GENMASK(7, 2) +#define MT6370_VENID_MASK GENMASK(7, 4) #define MT6370_AICR_100_mA 0x0 #define MT6370_AICR_150_mA 0x1 @@ -47,6 +49,10 @@ #define ADC_CONV_TIME_MS 35 #define ADC_CONV_POLLING_TIME_US 1000 +#define MT6370_VID_RT5081 0x8 +#define MT6370_VID_RT5081A 0xA +#define MT6370_VID_MT6370 0xE + struct mt6370_adc_data { struct device *dev; struct regmap *regmap; @@ -55,6 +61,7 @@ struct mt6370_adc_data { * from being read at the same time. */ struct mutex adc_lock; + unsigned int vid; }; static int mt6370_adc_read_channel(struct mt6370_adc_data *priv, int chan, @@ -98,6 +105,30 @@ static int mt6370_adc_read_channel(struct mt6370_adc_data *priv, int chan, return ret; } +static int mt6370_adc_get_ibus_scale(struct mt6370_adc_data *priv) +{ + switch (priv->vid) { + case MT6370_VID_RT5081: + case MT6370_VID_RT5081A: + case MT6370_VID_MT6370: + return 3350; + default: + return 3875; + } +} + +static int mt6370_adc_get_ibat_scale(struct mt6370_adc_data *priv) +{ + switch (priv->vid) { + case MT6370_VID_RT5081: + case MT6370_VID_RT5081A: + case MT6370_VID_MT6370: + return 2680; + default: + return 3870; + } +} + static int mt6370_adc_read_scale(struct mt6370_adc_data *priv, int chan, int *val1, int *val2) { @@ -123,7 +154,7 @@ static int mt6370_adc_read_scale(struct mt6370_adc_data *priv, case MT6370_AICR_250_mA: case MT6370_AICR_300_mA: case MT6370_AICR_350_mA: - *val1 = 3350; + *val1 = mt6370_adc_get_ibus_scale(priv); break; default: *val1 = 5000; @@ -150,7 +181,7 @@ static int mt6370_adc_read_scale(struct mt6370_adc_data *priv, case MT6370_ICHG_600_mA: case MT6370_ICHG_700_mA: case MT6370_ICHG_800_mA: - *val1 = 2680; + *val1 = mt6370_adc_get_ibat_scale(priv); break; default: *val1 = 5000; @@ -251,6 +282,20 @@ static const struct iio_chan_spec mt6370_adc_channels[] = { MT6370_ADC_CHAN(TEMP_JC, IIO_TEMP, 12, BIT(IIO_CHAN_INFO_OFFSET)), }; +static int mt6370_get_vendor_info(struct mt6370_adc_data *priv) +{ + unsigned int dev_info; + int ret; + + ret = regmap_read(priv->regmap, MT6370_REG_DEV_INFO, &dev_info); + if (ret) + return ret; + + priv->vid = FIELD_GET(MT6370_VENID_MASK, dev_info); + + return 0; +} + static int mt6370_adc_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; @@ -272,6 +317,10 @@ static int mt6370_adc_probe(struct platform_device *pdev) priv->regmap = regmap; mutex_init(&priv->adc_lock); + ret = mt6370_get_vendor_info(priv); + if (ret) + return dev_err_probe(dev, ret, "Failed to get vid\n"); + ret = regmap_write(priv->regmap, MT6370_REG_CHG_ADC, 0); if (ret) return dev_err_probe(dev, ret, "Failed to reset ADC\n"); diff --git a/drivers/iio/adc/mxs-lradc-adc.c b/drivers/iio/adc/mxs-lradc-adc.c index bca79a93cbe43186fb8de002ab2f8a63bba54499..a50f39143d3ea3d7f23cf353081049b7069ca64b 100644 --- a/drivers/iio/adc/mxs-lradc-adc.c +++ b/drivers/iio/adc/mxs-lradc-adc.c @@ -757,13 +757,13 @@ static int mxs_lradc_adc_probe(struct platform_device *pdev) ret = mxs_lradc_adc_trigger_init(iio); if (ret) - goto err_trig; + return ret; ret = iio_triggered_buffer_setup(iio, &iio_pollfunc_store_time, &mxs_lradc_adc_trigger_handler, &mxs_lradc_adc_buffer_ops); if (ret) - return ret; + goto err_trig; adc->vref_mv = mxs_lradc_adc_vref_mv[lradc->soc]; @@ -801,9 +801,9 @@ static int mxs_lradc_adc_probe(struct platform_device *pdev) err_dev: mxs_lradc_adc_hw_stop(adc); - mxs_lradc_adc_trigger_remove(iio); -err_trig: iio_triggered_buffer_cleanup(iio); +err_trig: + mxs_lradc_adc_trigger_remove(iio); return ret; } @@ -814,8 +814,8 @@ static int mxs_lradc_adc_remove(struct platform_device *pdev) iio_device_unregister(iio); mxs_lradc_adc_hw_stop(adc); - mxs_lradc_adc_trigger_remove(iio); iio_triggered_buffer_cleanup(iio); + mxs_lradc_adc_trigger_remove(iio); return 0; } diff --git a/drivers/iio/adc/palmas_gpadc.c b/drivers/iio/adc/palmas_gpadc.c index c1c439215aeb5fc5a20313572e2b7e6b2ec41716..7dfc9c927a23f121e6ea08ddb0b06a15601d4aa3 100644 --- a/drivers/iio/adc/palmas_gpadc.c +++ b/drivers/iio/adc/palmas_gpadc.c @@ -547,7 +547,7 @@ static int palmas_gpadc_read_raw(struct iio_dev *indio_dev, int adc_chan = chan->channel; int ret = 0; - if (adc_chan > PALMAS_ADC_CH_MAX) + if (adc_chan >= PALMAS_ADC_CH_MAX) return -EINVAL; mutex_lock(&adc->lock); @@ -595,7 +595,7 @@ static int palmas_gpadc_read_event_config(struct iio_dev *indio_dev, int adc_chan = chan->channel; int ret = 0; - if (adc_chan > PALMAS_ADC_CH_MAX || type != IIO_EV_TYPE_THRESH) + if (adc_chan >= PALMAS_ADC_CH_MAX || type != IIO_EV_TYPE_THRESH) return -EINVAL; mutex_lock(&adc->lock); @@ -684,7 +684,7 @@ static int palmas_gpadc_write_event_config(struct iio_dev *indio_dev, int adc_chan = chan->channel; int ret; - if (adc_chan > PALMAS_ADC_CH_MAX || type != IIO_EV_TYPE_THRESH) + if (adc_chan >= PALMAS_ADC_CH_MAX || type != IIO_EV_TYPE_THRESH) return -EINVAL; mutex_lock(&adc->lock); @@ -710,7 +710,7 @@ static int palmas_gpadc_read_event_value(struct iio_dev *indio_dev, int adc_chan = chan->channel; int ret; - if (adc_chan > PALMAS_ADC_CH_MAX || type != IIO_EV_TYPE_THRESH) + if (adc_chan >= PALMAS_ADC_CH_MAX || type != IIO_EV_TYPE_THRESH) return -EINVAL; mutex_lock(&adc->lock); @@ -744,7 +744,7 @@ static int palmas_gpadc_write_event_value(struct iio_dev *indio_dev, int old; int ret; - if (adc_chan > PALMAS_ADC_CH_MAX || type != IIO_EV_TYPE_THRESH) + if (adc_chan >= PALMAS_ADC_CH_MAX || type != IIO_EV_TYPE_THRESH) return -EINVAL; mutex_lock(&adc->lock); diff --git a/drivers/iio/adc/stm32-adc.c b/drivers/iio/adc/stm32-adc.c index 1aadb2ad2cabdf12130030d4bf8b341b0fb93c42..bd7e2408bf281da5a90e40472cf7a2da3703f2bd 100644 --- a/drivers/iio/adc/stm32-adc.c +++ b/drivers/iio/adc/stm32-adc.c @@ -2006,16 +2006,15 @@ static int stm32_adc_get_legacy_chan_count(struct iio_dev *indio_dev, struct stm * to get the *real* number of channels. */ ret = device_property_count_u32(dev, "st,adc-diff-channels"); - if (ret < 0) - return ret; - - ret /= (int)(sizeof(struct stm32_adc_diff_channel) / sizeof(u32)); - if (ret > adc_info->max_channels) { - dev_err(&indio_dev->dev, "Bad st,adc-diff-channels?\n"); - return -EINVAL; - } else if (ret > 0) { - adc->num_diff = ret; - num_channels += ret; + if (ret > 0) { + ret /= (int)(sizeof(struct stm32_adc_diff_channel) / sizeof(u32)); + if (ret > adc_info->max_channels) { + dev_err(&indio_dev->dev, "Bad st,adc-diff-channels?\n"); + return -EINVAL; + } else if (ret > 0) { + adc->num_diff = ret; + num_channels += ret; + } } /* Optional sample time is provided either for each, or all channels */ @@ -2037,6 +2036,7 @@ static int stm32_adc_legacy_chan_init(struct iio_dev *indio_dev, struct stm32_adc_diff_channel diff[STM32_ADC_CH_MAX]; struct device *dev = &indio_dev->dev; u32 num_diff = adc->num_diff; + int num_se = nchans - num_diff; int size = num_diff * sizeof(*diff) / sizeof(u32); int scan_index = 0, ret, i, c; u32 smp = 0, smps[STM32_ADC_CH_MAX], chans[STM32_ADC_CH_MAX]; @@ -2063,29 +2063,32 @@ static int stm32_adc_legacy_chan_init(struct iio_dev *indio_dev, scan_index++; } } - - ret = device_property_read_u32_array(dev, "st,adc-channels", chans, - nchans); - if (ret) - return ret; - - for (c = 0; c < nchans; c++) { - if (chans[c] >= adc_info->max_channels) { - dev_err(&indio_dev->dev, "Invalid channel %d\n", - chans[c]); - return -EINVAL; + if (num_se > 0) { + ret = device_property_read_u32_array(dev, "st,adc-channels", chans, num_se); + if (ret) { + dev_err(&indio_dev->dev, "Failed to get st,adc-channels %d\n", ret); + return ret; } - /* Channel can't be configured both as single-ended & diff */ - for (i = 0; i < num_diff; i++) { - if (chans[c] == diff[i].vinp) { - dev_err(&indio_dev->dev, "channel %d misconfigured\n", chans[c]); + for (c = 0; c < num_se; c++) { + if (chans[c] >= adc_info->max_channels) { + dev_err(&indio_dev->dev, "Invalid channel %d\n", + chans[c]); return -EINVAL; } + + /* Channel can't be configured both as single-ended & diff */ + for (i = 0; i < num_diff; i++) { + if (chans[c] == diff[i].vinp) { + dev_err(&indio_dev->dev, "channel %d misconfigured\n", + chans[c]); + return -EINVAL; + } + } + stm32_adc_chan_init_one(indio_dev, &channels[scan_index], + chans[c], 0, scan_index, false); + scan_index++; } - stm32_adc_chan_init_one(indio_dev, &channels[scan_index], - chans[c], 0, scan_index, false); - scan_index++; } if (adc->nsmps > 0) { @@ -2306,7 +2309,7 @@ static int stm32_adc_chan_fw_init(struct iio_dev *indio_dev, bool timestamping) if (legacy) ret = stm32_adc_legacy_chan_init(indio_dev, adc, channels, - num_channels); + timestamping ? num_channels - 1 : num_channels); else ret = stm32_adc_generic_chan_init(indio_dev, adc, channels); if (ret < 0) diff --git a/drivers/iio/addac/ad74413r.c b/drivers/iio/addac/ad74413r.c index 07e9f6ae16a83dc612dafbb212a63e888c934fda..e3366cf5eb31943c006e059eda6fc26558073431 100644 --- a/drivers/iio/addac/ad74413r.c +++ b/drivers/iio/addac/ad74413r.c @@ -1007,7 +1007,7 @@ static int ad74413r_read_raw(struct iio_dev *indio_dev, ret = ad74413r_get_single_adc_result(indio_dev, chan->channel, val); - if (ret) + if (ret < 0) return ret; ad74413r_adc_to_resistance_result(*val, val); diff --git a/drivers/iio/dac/Makefile b/drivers/iio/dac/Makefile index 6c74fea21736a426414689aabdb19422e5258356..addd97a788389cfb1305d41e872111534898c047 100644 --- a/drivers/iio/dac/Makefile +++ b/drivers/iio/dac/Makefile @@ -17,7 +17,7 @@ obj-$(CONFIG_AD5592R_BASE) += ad5592r-base.o obj-$(CONFIG_AD5592R) += ad5592r.o obj-$(CONFIG_AD5593R) += ad5593r.o obj-$(CONFIG_AD5755) += ad5755.o -obj-$(CONFIG_AD5755) += ad5758.o +obj-$(CONFIG_AD5758) += ad5758.o obj-$(CONFIG_AD5761) += ad5761.o obj-$(CONFIG_AD5764) += ad5764.o obj-$(CONFIG_AD5766) += ad5766.o diff --git a/drivers/iio/dac/mcp4725.c b/drivers/iio/dac/mcp4725.c index 46bf758760f85bbd9500da020c3740fef1529e11..3f5661a3718fe87577d30c0ac039fdbfda4c9292 100644 --- a/drivers/iio/dac/mcp4725.c +++ b/drivers/iio/dac/mcp4725.c @@ -47,12 +47,18 @@ static int mcp4725_suspend(struct device *dev) struct mcp4725_data *data = iio_priv(i2c_get_clientdata( to_i2c_client(dev))); u8 outbuf[2]; + int ret; outbuf[0] = (data->powerdown_mode + 1) << 4; outbuf[1] = 0; data->powerdown = true; - return i2c_master_send(data->client, outbuf, 2); + ret = i2c_master_send(data->client, outbuf, 2); + if (ret < 0) + return ret; + else if (ret != 2) + return -EIO; + return 0; } static int mcp4725_resume(struct device *dev) @@ -60,13 +66,19 @@ static int mcp4725_resume(struct device *dev) struct mcp4725_data *data = iio_priv(i2c_get_clientdata( to_i2c_client(dev))); u8 outbuf[2]; + int ret; /* restore previous DAC value */ outbuf[0] = (data->dac_value >> 8) & 0xf; outbuf[1] = data->dac_value & 0xff; data->powerdown = false; - return i2c_master_send(data->client, outbuf, 2); + ret = i2c_master_send(data->client, outbuf, 2); + if (ret < 0) + return ret; + else if (ret != 2) + return -EIO; + return 0; } static DEFINE_SIMPLE_DEV_PM_OPS(mcp4725_pm_ops, mcp4725_suspend, mcp4725_resume); diff --git a/drivers/iio/imu/inv_icm42600/inv_icm42600_buffer.c b/drivers/iio/imu/inv_icm42600/inv_icm42600_buffer.c index 99576b2c171f4d3d47f18de4704511ec69d1842e..32d7f836423036851b33493ded7e12007eadb47d 100644 --- a/drivers/iio/imu/inv_icm42600/inv_icm42600_buffer.c +++ b/drivers/iio/imu/inv_icm42600/inv_icm42600_buffer.c @@ -275,9 +275,14 @@ static int inv_icm42600_buffer_preenable(struct iio_dev *indio_dev) { struct inv_icm42600_state *st = iio_device_get_drvdata(indio_dev); struct device *dev = regmap_get_device(st->map); + struct inv_icm42600_timestamp *ts = iio_priv(indio_dev); pm_runtime_get_sync(dev); + mutex_lock(&st->lock); + inv_icm42600_timestamp_reset(ts); + mutex_unlock(&st->lock); + return 0; } @@ -375,7 +380,6 @@ static int inv_icm42600_buffer_postdisable(struct iio_dev *indio_dev) struct device *dev = regmap_get_device(st->map); unsigned int sensor; unsigned int *watermark; - struct inv_icm42600_timestamp *ts; struct inv_icm42600_sensor_conf conf = INV_ICM42600_SENSOR_CONF_INIT; unsigned int sleep_temp = 0; unsigned int sleep_sensor = 0; @@ -385,11 +389,9 @@ static int inv_icm42600_buffer_postdisable(struct iio_dev *indio_dev) if (indio_dev == st->indio_gyro) { sensor = INV_ICM42600_SENSOR_GYRO; watermark = &st->fifo.watermark.gyro; - ts = iio_priv(st->indio_gyro); } else if (indio_dev == st->indio_accel) { sensor = INV_ICM42600_SENSOR_ACCEL; watermark = &st->fifo.watermark.accel; - ts = iio_priv(st->indio_accel); } else { return -EINVAL; } @@ -417,8 +419,6 @@ static int inv_icm42600_buffer_postdisable(struct iio_dev *indio_dev) if (!st->fifo.on) ret = inv_icm42600_set_temp_conf(st, false, &sleep_temp); - inv_icm42600_timestamp_reset(ts); - out_unlock: mutex_unlock(&st->lock); diff --git a/drivers/iio/industrialio-gts-helper.c b/drivers/iio/industrialio-gts-helper.c index 8bb68975b2598a48682af26d798b0af9340de311..7653261d2dc2bf4cb916f9ca30b239261597a018 100644 --- a/drivers/iio/industrialio-gts-helper.c +++ b/drivers/iio/industrialio-gts-helper.c @@ -337,6 +337,17 @@ static int iio_gts_build_avail_scale_table(struct iio_gts *gts) return ret; } +static void iio_gts_us_to_int_micro(int *time_us, int *int_micro_times, + int num_times) +{ + int i; + + for (i = 0; i < num_times; i++) { + int_micro_times[i * 2] = time_us[i] / 1000000; + int_micro_times[i * 2 + 1] = time_us[i] % 1000000; + } +} + /** * iio_gts_build_avail_time_table - build table of available integration times * @gts: Gain time scale descriptor @@ -351,7 +362,7 @@ static int iio_gts_build_avail_scale_table(struct iio_gts *gts) */ static int iio_gts_build_avail_time_table(struct iio_gts *gts) { - int *times, i, j, idx = 0; + int *times, i, j, idx = 0, *int_micro_times; if (!gts->num_itime) return 0; @@ -378,13 +389,24 @@ static int iio_gts_build_avail_time_table(struct iio_gts *gts) } } } - gts->avail_time_tables = times; - /* - * This is just to survive a unlikely corner-case where times in the - * given time table were not unique. Else we could just trust the - * gts->num_itime. - */ - gts->num_avail_time_tables = idx; + + /* create a list of times formatted as list of IIO_VAL_INT_PLUS_MICRO */ + int_micro_times = kcalloc(idx, sizeof(int) * 2, GFP_KERNEL); + if (int_micro_times) { + /* + * This is just to survive a unlikely corner-case where times in + * the given time table were not unique. Else we could just + * trust the gts->num_itime. + */ + gts->num_avail_time_tables = idx; + iio_gts_us_to_int_micro(times, int_micro_times, idx); + } + + gts->avail_time_tables = int_micro_times; + kfree(times); + + if (!int_micro_times) + return -ENOMEM; return 0; } @@ -683,8 +705,8 @@ int iio_gts_avail_times(struct iio_gts *gts, const int **vals, int *type, return -EINVAL; *vals = gts->avail_time_tables; - *type = IIO_VAL_INT; - *length = gts->num_avail_time_tables; + *type = IIO_VAL_INT_PLUS_MICRO; + *length = gts->num_avail_time_tables * 2; return IIO_AVAIL_LIST; } diff --git a/drivers/iio/light/rohm-bu27034.c b/drivers/iio/light/rohm-bu27034.c index e486dcf35eba7d60fc84e52f75fc566e52193a72..f85194fda6b093d5de2c487fb268e89ea3b2b14d 100644 --- a/drivers/iio/light/rohm-bu27034.c +++ b/drivers/iio/light/rohm-bu27034.c @@ -231,6 +231,9 @@ struct bu27034_result { static const struct regmap_range bu27034_volatile_ranges[] = { { + .range_min = BU27034_REG_SYSTEM_CONTROL, + .range_max = BU27034_REG_SYSTEM_CONTROL, + }, { .range_min = BU27034_REG_MODE_CONTROL4, .range_max = BU27034_REG_MODE_CONTROL4, }, { @@ -1167,11 +1170,12 @@ static int bu27034_read_raw(struct iio_dev *idev, switch (mask) { case IIO_CHAN_INFO_INT_TIME: - *val = bu27034_get_int_time(data); - if (*val < 0) - return *val; + *val = 0; + *val2 = bu27034_get_int_time(data); + if (*val2 < 0) + return *val2; - return IIO_VAL_INT; + return IIO_VAL_INT_PLUS_MICRO; case IIO_CHAN_INFO_SCALE: return bu27034_get_scale(data, chan->channel, val, val2); @@ -1229,7 +1233,10 @@ static int bu27034_write_raw(struct iio_dev *idev, ret = bu27034_set_scale(data, chan->channel, val, val2); break; case IIO_CHAN_INFO_INT_TIME: - ret = bu27034_try_set_int_time(data, val); + if (!val) + ret = bu27034_try_set_int_time(data, val2); + else + ret = -EINVAL; break; default: ret = -EINVAL; @@ -1268,12 +1275,19 @@ static int bu27034_chip_init(struct bu27034_data *data) int ret, sel; /* Reset */ - ret = regmap_update_bits(data->regmap, BU27034_REG_SYSTEM_CONTROL, + ret = regmap_write_bits(data->regmap, BU27034_REG_SYSTEM_CONTROL, BU27034_MASK_SW_RESET, BU27034_MASK_SW_RESET); if (ret) return dev_err_probe(data->dev, ret, "Sensor reset failed\n"); msleep(1); + + ret = regmap_reinit_cache(data->regmap, &bu27034_regmap); + if (ret) { + dev_err(data->dev, "Failed to reinit reg cache\n"); + return ret; + } + /* * Read integration time here to ensure it is in regmap cache. We do * this to speed-up the int-time acquisition in the start of the buffer diff --git a/drivers/iio/light/vcnl4035.c b/drivers/iio/light/vcnl4035.c index 14e29330e972eb91f094a284913fb57e79e50572..94f5d611e98c3c80225737dd837a3d25d834aa9d 100644 --- a/drivers/iio/light/vcnl4035.c +++ b/drivers/iio/light/vcnl4035.c @@ -8,6 +8,7 @@ * TODO: Proximity */ #include +#include #include #include #include @@ -42,6 +43,7 @@ #define VCNL4035_ALS_PERS_MASK GENMASK(3, 2) #define VCNL4035_INT_ALS_IF_H_MASK BIT(12) #define VCNL4035_INT_ALS_IF_L_MASK BIT(13) +#define VCNL4035_DEV_ID_MASK GENMASK(7, 0) /* Default values */ #define VCNL4035_MODE_ALS_ENABLE BIT(0) @@ -413,6 +415,7 @@ static int vcnl4035_init(struct vcnl4035_data *data) return ret; } + id = FIELD_GET(VCNL4035_DEV_ID_MASK, id); if (id != VCNL4035_DEV_ID_VAL) { dev_err(&data->client->dev, "Wrong id, got %x, expected %x\n", id, VCNL4035_DEV_ID_VAL); diff --git a/drivers/iio/magnetometer/tmag5273.c b/drivers/iio/magnetometer/tmag5273.c index 28bb7efe8df8c07bba0b4810c9beae3e60c5ef8c..e155a75b3cd29d3563b59b8a9fd98889f0ac9bf9 100644 --- a/drivers/iio/magnetometer/tmag5273.c +++ b/drivers/iio/magnetometer/tmag5273.c @@ -296,12 +296,13 @@ static int tmag5273_read_raw(struct iio_dev *indio_dev, return ret; ret = tmag5273_get_measure(data, &t, &x, &y, &z, &angle, &magnitude); - if (ret) - return ret; pm_runtime_mark_last_busy(data->dev); pm_runtime_put_autosuspend(data->dev); + if (ret) + return ret; + switch (chan->address) { case TEMPERATURE: *val = t; diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index 93a1c48d0c3281ff617f6d1daed911266cf3a0bb..6b3f4384e46acbfcbc0f6cee7ed8e31d048c924f 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -3295,7 +3295,7 @@ static int cma_resolve_iboe_route(struct rdma_id_private *id_priv) route->path_rec->traffic_class = tos; route->path_rec->mtu = iboe_get_mtu(ndev->mtu); route->path_rec->rate_selector = IB_SA_EQ; - route->path_rec->rate = iboe_get_rate(ndev); + route->path_rec->rate = IB_RATE_PORT_CURRENT; dev_put(ndev); route->path_rec->packet_life_time_selector = IB_SA_EQ; /* In case ACK timeout is set, use this value to calculate @@ -4964,7 +4964,7 @@ static int cma_iboe_join_multicast(struct rdma_id_private *id_priv, if (!ndev) return -ENODEV; - ib.rec.rate = iboe_get_rate(ndev); + ib.rec.rate = IB_RATE_PORT_CURRENT; ib.rec.hop_limit = 1; ib.rec.mtu = iboe_get_mtu(ndev->mtu); diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 4796f6a8828caa47ba0f1b980cdee3b1ec1fb97f..e836c9c477f6755e97a469b5f0cf588d8c1aa21c 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -1850,8 +1850,13 @@ static int modify_qp(struct uverbs_attr_bundle *attrs, attr->path_mtu = cmd->base.path_mtu; if (cmd->base.attr_mask & IB_QP_PATH_MIG_STATE) attr->path_mig_state = cmd->base.path_mig_state; - if (cmd->base.attr_mask & IB_QP_QKEY) + if (cmd->base.attr_mask & IB_QP_QKEY) { + if (cmd->base.qkey & IB_QP_SET_QKEY && !capable(CAP_NET_RAW)) { + ret = -EPERM; + goto release_qp; + } attr->qkey = cmd->base.qkey; + } if (cmd->base.attr_mask & IB_QP_RQ_PSN) attr->rq_psn = cmd->base.rq_psn; if (cmd->base.attr_mask & IB_QP_SQ_PSN) diff --git a/drivers/infiniband/core/uverbs_main.c b/drivers/infiniband/core/uverbs_main.c index fbace69672ca921f5d13ba9ab64462dda8d95e21..7c9c79c139411cc9efca2b12ad4ed56935f14491 100644 --- a/drivers/infiniband/core/uverbs_main.c +++ b/drivers/infiniband/core/uverbs_main.c @@ -222,8 +222,12 @@ static ssize_t ib_uverbs_event_read(struct ib_uverbs_event_queue *ev_queue, spin_lock_irq(&ev_queue->lock); while (list_empty(&ev_queue->event_list)) { - spin_unlock_irq(&ev_queue->lock); + if (ev_queue->is_closed) { + spin_unlock_irq(&ev_queue->lock); + return -EIO; + } + spin_unlock_irq(&ev_queue->lock); if (filp->f_flags & O_NONBLOCK) return -EAGAIN; @@ -233,12 +237,6 @@ static ssize_t ib_uverbs_event_read(struct ib_uverbs_event_queue *ev_queue, return -ERESTARTSYS; spin_lock_irq(&ev_queue->lock); - - /* If device was disassociated and no event exists set an error */ - if (list_empty(&ev_queue->event_list) && ev_queue->is_closed) { - spin_unlock_irq(&ev_queue->lock); - return -EIO; - } } event = list_entry(ev_queue->event_list.next, struct ib_uverbs_event, list); diff --git a/drivers/infiniband/hw/bnxt_re/bnxt_re.h b/drivers/infiniband/hw/bnxt_re/bnxt_re.h index 5a2baf49ecaa47c022ad848a620b12ed50a2e580..2c95e6f3d47acc6c43aceb627907f66579c142c8 100644 --- a/drivers/infiniband/hw/bnxt_re/bnxt_re.h +++ b/drivers/infiniband/hw/bnxt_re/bnxt_re.h @@ -135,8 +135,6 @@ struct bnxt_re_dev { struct delayed_work worker; u8 cur_prio_map; - u16 active_speed; - u8 active_width; /* FP Notification Queue (CQ & SRQ) */ struct tasklet_struct nq_task; diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.c b/drivers/infiniband/hw/bnxt_re/ib_verbs.c index e86afecfbe46e67374f38e799f0d01acd332b8b4..952811c40c54be64d52532f455399a1a1417d96c 100644 --- a/drivers/infiniband/hw/bnxt_re/ib_verbs.c +++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.c @@ -199,6 +199,7 @@ int bnxt_re_query_port(struct ib_device *ibdev, u32 port_num, { struct bnxt_re_dev *rdev = to_bnxt_re_dev(ibdev, ibdev); struct bnxt_qplib_dev_attr *dev_attr = &rdev->dev_attr; + int rc; memset(port_attr, 0, sizeof(*port_attr)); @@ -228,10 +229,10 @@ int bnxt_re_query_port(struct ib_device *ibdev, u32 port_num, port_attr->sm_sl = 0; port_attr->subnet_timeout = 0; port_attr->init_type_reply = 0; - port_attr->active_speed = rdev->active_speed; - port_attr->active_width = rdev->active_width; + rc = ib_get_eth_speed(&rdev->ibdev, port_num, &port_attr->active_speed, + &port_attr->active_width); - return 0; + return rc; } int bnxt_re_get_port_immutable(struct ib_device *ibdev, u32 port_num, @@ -3341,9 +3342,7 @@ static int bnxt_re_process_raw_qp_pkt_rx(struct bnxt_re_qp *gsi_qp, udwr.remote_qkey = gsi_sqp->qplib_qp.qkey; /* post data received in the send queue */ - rc = bnxt_re_post_send_shadow_qp(rdev, gsi_sqp, swr); - - return 0; + return bnxt_re_post_send_shadow_qp(rdev, gsi_sqp, swr); } static void bnxt_re_process_res_rawqp1_wc(struct ib_wc *wc, diff --git a/drivers/infiniband/hw/bnxt_re/main.c b/drivers/infiniband/hw/bnxt_re/main.c index b9e2f89337e85e9a26b1d7ce63ceef906fd4fd67..3073398a218341e67f5742f00247461276bc4bdf 100644 --- a/drivers/infiniband/hw/bnxt_re/main.c +++ b/drivers/infiniband/hw/bnxt_re/main.c @@ -1077,8 +1077,6 @@ static int bnxt_re_ib_init(struct bnxt_re_dev *rdev) return rc; } dev_info(rdev_to_dev(rdev), "Device registered with IB successfully"); - ib_get_eth_speed(&rdev->ibdev, 1, &rdev->active_speed, - &rdev->active_width); set_bit(BNXT_RE_FLAG_ISSUE_ROCE_STATS, &rdev->flags); event = netif_running(rdev->netdev) && netif_carrier_ok(rdev->netdev) ? @@ -1336,6 +1334,10 @@ static void bnxt_re_setup_cc(struct bnxt_re_dev *rdev, bool enable) { struct bnxt_qplib_cc_param cc_param = {}; + /* Do not enable congestion control on VFs */ + if (rdev->is_virtfn) + return; + /* Currently enabling only for GenP5 adapters */ if (!bnxt_qplib_is_chip_gen_p5(rdev->chip_ctx)) return; diff --git a/drivers/infiniband/hw/bnxt_re/qplib_fp.c b/drivers/infiniband/hw/bnxt_re/qplib_fp.c index f139d4cd17128a9b5a79bb02f400092b7b4d986a..8974f6235cfaa83a9e658faf08e2085a23738f4e 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_fp.c +++ b/drivers/infiniband/hw/bnxt_re/qplib_fp.c @@ -2056,6 +2056,12 @@ int bnxt_qplib_create_cq(struct bnxt_qplib_res *res, struct bnxt_qplib_cq *cq) u32 pg_sz_lvl; int rc; + if (!cq->dpi) { + dev_err(&rcfw->pdev->dev, + "FP: CREATE_CQ failed due to NULL DPI\n"); + return -EINVAL; + } + hwq_attr.res = res; hwq_attr.depth = cq->max_wqe; hwq_attr.stride = sizeof(struct cq_base); @@ -2069,11 +2075,6 @@ int bnxt_qplib_create_cq(struct bnxt_qplib_res *res, struct bnxt_qplib_cq *cq) CMDQ_BASE_OPCODE_CREATE_CQ, sizeof(req)); - if (!cq->dpi) { - dev_err(&rcfw->pdev->dev, - "FP: CREATE_CQ failed due to NULL DPI\n"); - return -EINVAL; - } req.dpi = cpu_to_le32(cq->dpi->dpi); req.cq_handle = cpu_to_le64(cq->cq_handle); req.cq_size = cpu_to_le32(cq->hwq.max_elements); diff --git a/drivers/infiniband/hw/bnxt_re/qplib_res.c b/drivers/infiniband/hw/bnxt_re/qplib_res.c index 126d4f26f75ad722f9f71857b0711e3e31b0f451..81b0c5e879f9e857a77875dd104530eb4edaee91 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_res.c +++ b/drivers/infiniband/hw/bnxt_re/qplib_res.c @@ -215,17 +215,9 @@ int bnxt_qplib_alloc_init_hwq(struct bnxt_qplib_hwq *hwq, return -EINVAL; hwq_attr->sginfo->npages = npages; } else { - unsigned long sginfo_num_pages = ib_umem_num_dma_blocks( - hwq_attr->sginfo->umem, hwq_attr->sginfo->pgsize); - + npages = ib_umem_num_dma_blocks(hwq_attr->sginfo->umem, + hwq_attr->sginfo->pgsize); hwq->is_user = true; - npages = sginfo_num_pages; - npages = (npages * PAGE_SIZE) / - BIT_ULL(hwq_attr->sginfo->pgshft); - if ((sginfo_num_pages * PAGE_SIZE) % - BIT_ULL(hwq_attr->sginfo->pgshft)) - if (!npages) - npages++; } if (npages == MAX_PBL_LVL_0_PGS && !hwq_attr->sginfo->nopte) { diff --git a/drivers/infiniband/hw/bnxt_re/qplib_sp.c b/drivers/infiniband/hw/bnxt_re/qplib_sp.c index 1714a1e231132fcf3b6eee90ca03277475d88d7a..b967a17a44bebcc282547488eb9209781dda3450 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_sp.c +++ b/drivers/infiniband/hw/bnxt_re/qplib_sp.c @@ -617,16 +617,15 @@ int bnxt_qplib_reg_mr(struct bnxt_qplib_res *res, struct bnxt_qplib_mrw *mr, /* Free the hwq if it already exist, must be a rereg */ if (mr->hwq.max_elements) bnxt_qplib_free_hwq(res, &mr->hwq); - /* Use system PAGE_SIZE */ hwq_attr.res = res; hwq_attr.depth = pages; - hwq_attr.stride = buf_pg_size; + hwq_attr.stride = sizeof(dma_addr_t); hwq_attr.type = HWQ_TYPE_MR; hwq_attr.sginfo = &sginfo; hwq_attr.sginfo->umem = umem; hwq_attr.sginfo->npages = pages; - hwq_attr.sginfo->pgsize = PAGE_SIZE; - hwq_attr.sginfo->pgshft = PAGE_SHIFT; + hwq_attr.sginfo->pgsize = buf_pg_size; + hwq_attr.sginfo->pgshft = ilog2(buf_pg_size); rc = bnxt_qplib_alloc_init_hwq(&mr->hwq, &hwq_attr); if (rc) { dev_err(&res->pdev->dev, diff --git a/drivers/infiniband/hw/efa/efa_verbs.c b/drivers/infiniband/hw/efa/efa_verbs.c index 8eca6c14d0cfca3db186810789058aaeecbcae21..2a195c4b0f17de077a87b6ff7a8fed4e837f03ad 100644 --- a/drivers/infiniband/hw/efa/efa_verbs.c +++ b/drivers/infiniband/hw/efa/efa_verbs.c @@ -1403,7 +1403,7 @@ static int pbl_continuous_initialize(struct efa_dev *dev, */ static int pbl_indirect_initialize(struct efa_dev *dev, struct pbl_context *pbl) { - u32 size_in_pages = DIV_ROUND_UP(pbl->pbl_buf_size_in_bytes, PAGE_SIZE); + u32 size_in_pages = DIV_ROUND_UP(pbl->pbl_buf_size_in_bytes, EFA_CHUNK_PAYLOAD_SIZE); struct scatterlist *sgl; int sg_dma_cnt, err; diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index 84f1167de1d9e7a9c2affa62adeb9034245fac5f..d4c6b9bc0a4ea222e7d4bff78ab3010d452a049c 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -4583,11 +4583,9 @@ static int modify_qp_init_to_rtr(struct ib_qp *ibqp, mtu = ib_mtu_enum_to_int(ib_mtu); if (WARN_ON(mtu <= 0)) return -EINVAL; -#define MAX_LP_MSG_LEN 16384 - /* MTU * (2 ^ LP_PKTN_INI) shouldn't be bigger than 16KB */ - lp_pktn_ini = ilog2(MAX_LP_MSG_LEN / mtu); - if (WARN_ON(lp_pktn_ini >= 0xF)) - return -EINVAL; +#define MIN_LP_MSG_LEN 1024 + /* mtu * (2 ^ lp_pktn_ini) should be in the range of 1024 to mtu */ + lp_pktn_ini = ilog2(max(mtu, MIN_LP_MSG_LEN) / mtu); if (attr_mask & IB_QP_PATH_MTU) { hr_reg_write(context, QPC_MTU, ib_mtu); @@ -5012,7 +5010,6 @@ static int hns_roce_v2_set_abs_fields(struct ib_qp *ibqp, static bool check_qp_timeout_cfg_range(struct hns_roce_dev *hr_dev, u8 *timeout) { #define QP_ACK_TIMEOUT_MAX_HIP08 20 -#define QP_ACK_TIMEOUT_OFFSET 10 #define QP_ACK_TIMEOUT_MAX 31 if (hr_dev->pci_dev->revision == PCI_REVISION_ID_HIP08) { @@ -5021,7 +5018,7 @@ static bool check_qp_timeout_cfg_range(struct hns_roce_dev *hr_dev, u8 *timeout) "local ACK timeout shall be 0 to 20.\n"); return false; } - *timeout += QP_ACK_TIMEOUT_OFFSET; + *timeout += HNS_ROCE_V2_QP_ACK_TIMEOUT_OFS_HIP08; } else if (hr_dev->pci_dev->revision > PCI_REVISION_ID_HIP08) { if (*timeout > QP_ACK_TIMEOUT_MAX) { ibdev_warn(&hr_dev->ib_dev, @@ -5307,6 +5304,18 @@ static int hns_roce_v2_query_qpc(struct hns_roce_dev *hr_dev, u32 qpn, return ret; } +static u8 get_qp_timeout_attr(struct hns_roce_dev *hr_dev, + struct hns_roce_v2_qp_context *context) +{ + u8 timeout; + + timeout = (u8)hr_reg_read(context, QPC_AT); + if (hr_dev->pci_dev->revision == PCI_REVISION_ID_HIP08) + timeout -= HNS_ROCE_V2_QP_ACK_TIMEOUT_OFS_HIP08; + + return timeout; +} + static int hns_roce_v2_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, int qp_attr_mask, struct ib_qp_init_attr *qp_init_attr) @@ -5384,7 +5393,7 @@ static int hns_roce_v2_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *qp_attr, qp_attr->max_dest_rd_atomic = 1 << hr_reg_read(&context, QPC_RR_MAX); qp_attr->min_rnr_timer = (u8)hr_reg_read(&context, QPC_MIN_RNR_TIME); - qp_attr->timeout = (u8)hr_reg_read(&context, QPC_AT); + qp_attr->timeout = get_qp_timeout_attr(hr_dev, &context); qp_attr->retry_cnt = hr_reg_read(&context, QPC_RETRY_NUM_INIT); qp_attr->rnr_retry = hr_reg_read(&context, QPC_RNR_NUM_INIT); diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.h b/drivers/infiniband/hw/hns/hns_roce_hw_v2.h index 1b44d2434ab415a756cb36fe6d3df6ee5d8bf637..7033eae2407c541c88a07d4a8543678c1fb630f7 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.h +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.h @@ -44,6 +44,8 @@ #define HNS_ROCE_V2_MAX_XRCD_NUM 0x1000000 #define HNS_ROCE_V2_RSV_XRCD_NUM 0 +#define HNS_ROCE_V2_QP_ACK_TIMEOUT_OFS_HIP08 10 + #define HNS_ROCE_V3_SCCC_SZ 64 #define HNS_ROCE_V3_GMV_ENTRY_SZ 32 diff --git a/drivers/infiniband/hw/hns/hns_roce_mr.c b/drivers/infiniband/hw/hns/hns_roce_mr.c index 37a5cf62f88b48b12293946e311bdd5d067edbaf..14376490ac226abce81b3f4f27f58726f5994a29 100644 --- a/drivers/infiniband/hw/hns/hns_roce_mr.c +++ b/drivers/infiniband/hw/hns/hns_roce_mr.c @@ -33,6 +33,7 @@ #include #include +#include #include "hns_roce_device.h" #include "hns_roce_cmd.h" #include "hns_roce_hem.h" @@ -909,6 +910,44 @@ static int mtr_init_buf_cfg(struct hns_roce_dev *hr_dev, return page_cnt; } +static u64 cal_pages_per_l1ba(unsigned int ba_per_bt, unsigned int hopnum) +{ + return int_pow(ba_per_bt, hopnum - 1); +} + +static unsigned int cal_best_bt_pg_sz(struct hns_roce_dev *hr_dev, + struct hns_roce_mtr *mtr, + unsigned int pg_shift) +{ + unsigned long cap = hr_dev->caps.page_size_cap; + struct hns_roce_buf_region *re; + unsigned int pgs_per_l1ba; + unsigned int ba_per_bt; + unsigned int ba_num; + int i; + + for_each_set_bit_from(pg_shift, &cap, sizeof(cap) * BITS_PER_BYTE) { + if (!(BIT(pg_shift) & cap)) + continue; + + ba_per_bt = BIT(pg_shift) / BA_BYTE_LEN; + ba_num = 0; + for (i = 0; i < mtr->hem_cfg.region_count; i++) { + re = &mtr->hem_cfg.region[i]; + if (re->hopnum == 0) + continue; + + pgs_per_l1ba = cal_pages_per_l1ba(ba_per_bt, re->hopnum); + ba_num += DIV_ROUND_UP(re->count, pgs_per_l1ba); + } + + if (ba_num <= ba_per_bt) + return pg_shift; + } + + return 0; +} + static int mtr_alloc_mtt(struct hns_roce_dev *hr_dev, struct hns_roce_mtr *mtr, unsigned int ba_page_shift) { @@ -917,6 +956,10 @@ static int mtr_alloc_mtt(struct hns_roce_dev *hr_dev, struct hns_roce_mtr *mtr, hns_roce_hem_list_init(&mtr->hem_list); if (!cfg->is_direct) { + ba_page_shift = cal_best_bt_pg_sz(hr_dev, mtr, ba_page_shift); + if (!ba_page_shift) + return -ERANGE; + ret = hns_roce_hem_list_request(hr_dev, &mtr->hem_list, cfg->region, cfg->region_count, ba_page_shift); diff --git a/drivers/infiniband/hw/irdma/verbs.c b/drivers/infiniband/hw/irdma/verbs.c index ab5cdf78278521123c92f8ff4482144797b765f6..eaa12c124598285beaa891761d8a2ec6aaa17ada 100644 --- a/drivers/infiniband/hw/irdma/verbs.c +++ b/drivers/infiniband/hw/irdma/verbs.c @@ -522,11 +522,6 @@ static int irdma_destroy_qp(struct ib_qp *ibqp, struct ib_udata *udata) if (!iwqp->user_mode) cancel_delayed_work_sync(&iwqp->dwork_flush); - irdma_qp_rem_ref(&iwqp->ibqp); - wait_for_completion(&iwqp->free_qp); - irdma_free_lsmm_rsrc(iwqp); - irdma_cqp_qp_destroy_cmd(&iwdev->rf->sc_dev, &iwqp->sc_qp); - if (!iwqp->user_mode) { if (iwqp->iwscq) { irdma_clean_cqes(iwqp, iwqp->iwscq); @@ -534,6 +529,12 @@ static int irdma_destroy_qp(struct ib_qp *ibqp, struct ib_udata *udata) irdma_clean_cqes(iwqp, iwqp->iwrcq); } } + + irdma_qp_rem_ref(&iwqp->ibqp); + wait_for_completion(&iwqp->free_qp); + irdma_free_lsmm_rsrc(iwqp); + irdma_cqp_qp_destroy_cmd(&iwdev->rf->sc_dev, &iwqp->sc_qp); + irdma_remove_push_mmap_entries(iwqp); irdma_free_qp_rsrc(iwqp); @@ -3291,6 +3292,7 @@ static int irdma_post_send(struct ib_qp *ibqp, break; case IB_WR_LOCAL_INV: info.op_type = IRDMA_OP_TYPE_INV_STAG; + info.local_fence = info.read_fence; info.op.inv_local_stag.target_stag = ib_wr->ex.invalidate_rkey; err = irdma_uk_stag_local_invalidate(ukqp, &info, true); break; diff --git a/drivers/infiniband/hw/mlx5/counters.c b/drivers/infiniband/hw/mlx5/counters.c index 1c06920505d25fb8f975bcac64f25eadbb610aa2..93257fa5aae8eb6f59b5cf58cda945246933c681 100644 --- a/drivers/infiniband/hw/mlx5/counters.c +++ b/drivers/infiniband/hw/mlx5/counters.c @@ -209,7 +209,8 @@ static const struct mlx5_ib_counters *get_counters(struct mlx5_ib_dev *dev, !vport_qcounters_supported(dev)) || !port_num) return &dev->port[0].cnts; - return &dev->port[port_num - 1].cnts; + return is_mdev_switchdev_mode(dev->mdev) ? + &dev->port[1].cnts : &dev->port[port_num - 1].cnts; } /** @@ -262,7 +263,7 @@ static struct rdma_hw_stats * mlx5_ib_alloc_hw_port_stats(struct ib_device *ibdev, u32 port_num) { struct mlx5_ib_dev *dev = to_mdev(ibdev); - const struct mlx5_ib_counters *cnts = &dev->port[port_num - 1].cnts; + const struct mlx5_ib_counters *cnts = get_counters(dev, port_num); return do_alloc_stats(cnts); } @@ -329,6 +330,7 @@ static int mlx5_ib_query_q_counters_vport(struct mlx5_ib_dev *dev, { u32 out[MLX5_ST_SZ_DW(query_q_counter_out)] = {}; u32 in[MLX5_ST_SZ_DW(query_q_counter_in)] = {}; + struct mlx5_core_dev *mdev; __be32 val; int ret, i; @@ -336,12 +338,16 @@ static int mlx5_ib_query_q_counters_vport(struct mlx5_ib_dev *dev, dev->port[port_num].rep->vport == MLX5_VPORT_UPLINK) return 0; + mdev = mlx5_eswitch_get_core_dev(dev->port[port_num].rep->esw); + if (!mdev) + return -EOPNOTSUPP; + MLX5_SET(query_q_counter_in, in, opcode, MLX5_CMD_OP_QUERY_Q_COUNTER); MLX5_SET(query_q_counter_in, in, other_vport, 1); MLX5_SET(query_q_counter_in, in, vport_number, dev->port[port_num].rep->vport); MLX5_SET(query_q_counter_in, in, aggregate, 1); - ret = mlx5_cmd_exec_inout(dev->mdev, query_q_counter, in, out); + ret = mlx5_cmd_exec_inout(mdev, query_q_counter, in, out); if (ret) return ret; @@ -575,43 +581,53 @@ static void mlx5_ib_fill_counters(struct mlx5_ib_dev *dev, bool is_vport = is_mdev_switchdev_mode(dev->mdev) && port_num != MLX5_VPORT_PF; const struct mlx5_ib_counter *names; - int j = 0, i; + int j = 0, i, size; names = is_vport ? vport_basic_q_cnts : basic_q_cnts; - for (i = 0; i < ARRAY_SIZE(basic_q_cnts); i++, j++) { + size = is_vport ? ARRAY_SIZE(vport_basic_q_cnts) : + ARRAY_SIZE(basic_q_cnts); + for (i = 0; i < size; i++, j++) { descs[j].name = names[i].name; - offsets[j] = basic_q_cnts[i].offset; + offsets[j] = names[i].offset; } names = is_vport ? vport_out_of_seq_q_cnts : out_of_seq_q_cnts; + size = is_vport ? ARRAY_SIZE(vport_out_of_seq_q_cnts) : + ARRAY_SIZE(out_of_seq_q_cnts); if (MLX5_CAP_GEN(dev->mdev, out_of_seq_cnt)) { - for (i = 0; i < ARRAY_SIZE(out_of_seq_q_cnts); i++, j++) { + for (i = 0; i < size; i++, j++) { descs[j].name = names[i].name; - offsets[j] = out_of_seq_q_cnts[i].offset; + offsets[j] = names[i].offset; } } names = is_vport ? vport_retrans_q_cnts : retrans_q_cnts; + size = is_vport ? ARRAY_SIZE(vport_retrans_q_cnts) : + ARRAY_SIZE(retrans_q_cnts); if (MLX5_CAP_GEN(dev->mdev, retransmission_q_counters)) { - for (i = 0; i < ARRAY_SIZE(retrans_q_cnts); i++, j++) { + for (i = 0; i < size; i++, j++) { descs[j].name = names[i].name; - offsets[j] = retrans_q_cnts[i].offset; + offsets[j] = names[i].offset; } } names = is_vport ? vport_extended_err_cnts : extended_err_cnts; + size = is_vport ? ARRAY_SIZE(vport_extended_err_cnts) : + ARRAY_SIZE(extended_err_cnts); if (MLX5_CAP_GEN(dev->mdev, enhanced_error_q_counters)) { - for (i = 0; i < ARRAY_SIZE(extended_err_cnts); i++, j++) { + for (i = 0; i < size; i++, j++) { descs[j].name = names[i].name; - offsets[j] = extended_err_cnts[i].offset; + offsets[j] = names[i].offset; } } names = is_vport ? vport_roce_accl_cnts : roce_accl_cnts; + size = is_vport ? ARRAY_SIZE(vport_roce_accl_cnts) : + ARRAY_SIZE(roce_accl_cnts); if (MLX5_CAP_GEN(dev->mdev, roce_accl)) { - for (i = 0; i < ARRAY_SIZE(roce_accl_cnts); i++, j++) { + for (i = 0; i < size; i++, j++) { descs[j].name = names[i].name; - offsets[j] = roce_accl_cnts[i].offset; + offsets[j] = names[i].offset; } } @@ -661,25 +677,37 @@ static void mlx5_ib_fill_counters(struct mlx5_ib_dev *dev, static int __mlx5_ib_alloc_counters(struct mlx5_ib_dev *dev, struct mlx5_ib_counters *cnts, u32 port_num) { - u32 num_counters, num_op_counters = 0; + bool is_vport = is_mdev_switchdev_mode(dev->mdev) && + port_num != MLX5_VPORT_PF; + u32 num_counters, num_op_counters = 0, size; - num_counters = ARRAY_SIZE(basic_q_cnts); + size = is_vport ? ARRAY_SIZE(vport_basic_q_cnts) : + ARRAY_SIZE(basic_q_cnts); + num_counters = size; + size = is_vport ? ARRAY_SIZE(vport_out_of_seq_q_cnts) : + ARRAY_SIZE(out_of_seq_q_cnts); if (MLX5_CAP_GEN(dev->mdev, out_of_seq_cnt)) - num_counters += ARRAY_SIZE(out_of_seq_q_cnts); + num_counters += size; + size = is_vport ? ARRAY_SIZE(vport_retrans_q_cnts) : + ARRAY_SIZE(retrans_q_cnts); if (MLX5_CAP_GEN(dev->mdev, retransmission_q_counters)) - num_counters += ARRAY_SIZE(retrans_q_cnts); + num_counters += size; + size = is_vport ? ARRAY_SIZE(vport_extended_err_cnts) : + ARRAY_SIZE(extended_err_cnts); if (MLX5_CAP_GEN(dev->mdev, enhanced_error_q_counters)) - num_counters += ARRAY_SIZE(extended_err_cnts); + num_counters += size; + size = is_vport ? ARRAY_SIZE(vport_roce_accl_cnts) : + ARRAY_SIZE(roce_accl_cnts); if (MLX5_CAP_GEN(dev->mdev, roce_accl)) - num_counters += ARRAY_SIZE(roce_accl_cnts); + num_counters += size; cnts->num_q_counters = num_counters; - if (is_mdev_switchdev_mode(dev->mdev) && port_num != MLX5_VPORT_PF) + if (is_vport) goto skip_non_qcounters; if (MLX5_CAP_GEN(dev->mdev, cc_query_allowed)) { @@ -725,11 +753,11 @@ static int __mlx5_ib_alloc_counters(struct mlx5_ib_dev *dev, static void mlx5_ib_dealloc_counters(struct mlx5_ib_dev *dev) { u32 in[MLX5_ST_SZ_DW(dealloc_q_counter_in)] = {}; - int num_cnt_ports; + int num_cnt_ports = dev->num_ports; int i, j; - num_cnt_ports = (!is_mdev_switchdev_mode(dev->mdev) || - vport_qcounters_supported(dev)) ? dev->num_ports : 1; + if (is_mdev_switchdev_mode(dev->mdev)) + num_cnt_ports = min(2, num_cnt_ports); MLX5_SET(dealloc_q_counter_in, in, opcode, MLX5_CMD_OP_DEALLOC_Q_COUNTER); @@ -761,15 +789,22 @@ static int mlx5_ib_alloc_counters(struct mlx5_ib_dev *dev) { u32 out[MLX5_ST_SZ_DW(alloc_q_counter_out)] = {}; u32 in[MLX5_ST_SZ_DW(alloc_q_counter_in)] = {}; - int num_cnt_ports; + int num_cnt_ports = dev->num_ports; int err = 0; int i; bool is_shared; MLX5_SET(alloc_q_counter_in, in, opcode, MLX5_CMD_OP_ALLOC_Q_COUNTER); is_shared = MLX5_CAP_GEN(dev->mdev, log_max_uctx) != 0; - num_cnt_ports = (!is_mdev_switchdev_mode(dev->mdev) || - vport_qcounters_supported(dev)) ? dev->num_ports : 1; + + /* + * In switchdev we need to allocate two ports, one that is used for + * the device Q_counters and it is essentially the real Q_counters of + * this device, while the other is used as a helper for PF to be able to + * query all other vports. + */ + if (is_mdev_switchdev_mode(dev->mdev)) + num_cnt_ports = min(2, num_cnt_ports); for (i = 0; i < num_cnt_ports; i++) { err = __mlx5_ib_alloc_counters(dev, &dev->port[i].cnts, i); diff --git a/drivers/infiniband/hw/mlx5/fs.c b/drivers/infiniband/hw/mlx5/fs.c index 3008632a6c2064a81c5c8d5ded5e2f98385319de..1e419e080b5352bbcdf3c7faa4277cca97f6e3ce 100644 --- a/drivers/infiniband/hw/mlx5/fs.c +++ b/drivers/infiniband/hw/mlx5/fs.c @@ -695,8 +695,6 @@ static struct mlx5_ib_flow_prio *_get_prio(struct mlx5_ib_dev *dev, struct mlx5_flow_table_attr ft_attr = {}; struct mlx5_flow_table *ft; - if (mlx5_ib_shared_ft_allowed(&dev->ib_dev)) - ft_attr.uid = MLX5_SHARED_RESOURCE_UID; ft_attr.prio = priority; ft_attr.max_fte = num_entries; ft_attr.flags = flags; @@ -2025,6 +2023,237 @@ static int flow_matcher_cleanup(struct ib_uobject *uobject, return 0; } +static int steering_anchor_create_ft(struct mlx5_ib_dev *dev, + struct mlx5_ib_flow_prio *ft_prio, + enum mlx5_flow_namespace_type ns_type) +{ + struct mlx5_flow_table_attr ft_attr = {}; + struct mlx5_flow_namespace *ns; + struct mlx5_flow_table *ft; + + if (ft_prio->anchor.ft) + return 0; + + ns = mlx5_get_flow_namespace(dev->mdev, ns_type); + if (!ns) + return -EOPNOTSUPP; + + ft_attr.flags = MLX5_FLOW_TABLE_UNMANAGED; + ft_attr.uid = MLX5_SHARED_RESOURCE_UID; + ft_attr.prio = 0; + ft_attr.max_fte = 2; + ft_attr.level = 1; + + ft = mlx5_create_flow_table(ns, &ft_attr); + if (IS_ERR(ft)) + return PTR_ERR(ft); + + ft_prio->anchor.ft = ft; + + return 0; +} + +static void steering_anchor_destroy_ft(struct mlx5_ib_flow_prio *ft_prio) +{ + if (ft_prio->anchor.ft) { + mlx5_destroy_flow_table(ft_prio->anchor.ft); + ft_prio->anchor.ft = NULL; + } +} + +static int +steering_anchor_create_fg_drop(struct mlx5_ib_flow_prio *ft_prio) +{ + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct mlx5_flow_group *fg; + void *flow_group_in; + int err = 0; + + if (ft_prio->anchor.fg_drop) + return 0; + + flow_group_in = kvzalloc(inlen, GFP_KERNEL); + if (!flow_group_in) + return -ENOMEM; + + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 1); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 1); + + fg = mlx5_create_flow_group(ft_prio->anchor.ft, flow_group_in); + if (IS_ERR(fg)) { + err = PTR_ERR(fg); + goto out; + } + + ft_prio->anchor.fg_drop = fg; + +out: + kvfree(flow_group_in); + + return err; +} + +static void +steering_anchor_destroy_fg_drop(struct mlx5_ib_flow_prio *ft_prio) +{ + if (ft_prio->anchor.fg_drop) { + mlx5_destroy_flow_group(ft_prio->anchor.fg_drop); + ft_prio->anchor.fg_drop = NULL; + } +} + +static int +steering_anchor_create_fg_goto_table(struct mlx5_ib_flow_prio *ft_prio) +{ + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct mlx5_flow_group *fg; + void *flow_group_in; + int err = 0; + + if (ft_prio->anchor.fg_goto_table) + return 0; + + flow_group_in = kvzalloc(inlen, GFP_KERNEL); + if (!flow_group_in) + return -ENOMEM; + + fg = mlx5_create_flow_group(ft_prio->anchor.ft, flow_group_in); + if (IS_ERR(fg)) { + err = PTR_ERR(fg); + goto out; + } + ft_prio->anchor.fg_goto_table = fg; + +out: + kvfree(flow_group_in); + + return err; +} + +static void +steering_anchor_destroy_fg_goto_table(struct mlx5_ib_flow_prio *ft_prio) +{ + if (ft_prio->anchor.fg_goto_table) { + mlx5_destroy_flow_group(ft_prio->anchor.fg_goto_table); + ft_prio->anchor.fg_goto_table = NULL; + } +} + +static int +steering_anchor_create_rule_drop(struct mlx5_ib_flow_prio *ft_prio) +{ + struct mlx5_flow_act flow_act = {}; + struct mlx5_flow_handle *handle; + + if (ft_prio->anchor.rule_drop) + return 0; + + flow_act.fg = ft_prio->anchor.fg_drop; + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP; + + handle = mlx5_add_flow_rules(ft_prio->anchor.ft, NULL, &flow_act, + NULL, 0); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + ft_prio->anchor.rule_drop = handle; + + return 0; +} + +static void steering_anchor_destroy_rule_drop(struct mlx5_ib_flow_prio *ft_prio) +{ + if (ft_prio->anchor.rule_drop) { + mlx5_del_flow_rules(ft_prio->anchor.rule_drop); + ft_prio->anchor.rule_drop = NULL; + } +} + +static int +steering_anchor_create_rule_goto_table(struct mlx5_ib_flow_prio *ft_prio) +{ + struct mlx5_flow_destination dest = {}; + struct mlx5_flow_act flow_act = {}; + struct mlx5_flow_handle *handle; + + if (ft_prio->anchor.rule_goto_table) + return 0; + + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + flow_act.flags |= FLOW_ACT_IGNORE_FLOW_LEVEL; + flow_act.fg = ft_prio->anchor.fg_goto_table; + + dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; + dest.ft = ft_prio->flow_table; + + handle = mlx5_add_flow_rules(ft_prio->anchor.ft, NULL, &flow_act, + &dest, 1); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + ft_prio->anchor.rule_goto_table = handle; + + return 0; +} + +static void +steering_anchor_destroy_rule_goto_table(struct mlx5_ib_flow_prio *ft_prio) +{ + if (ft_prio->anchor.rule_goto_table) { + mlx5_del_flow_rules(ft_prio->anchor.rule_goto_table); + ft_prio->anchor.rule_goto_table = NULL; + } +} + +static int steering_anchor_create_res(struct mlx5_ib_dev *dev, + struct mlx5_ib_flow_prio *ft_prio, + enum mlx5_flow_namespace_type ns_type) +{ + int err; + + err = steering_anchor_create_ft(dev, ft_prio, ns_type); + if (err) + return err; + + err = steering_anchor_create_fg_drop(ft_prio); + if (err) + goto destroy_ft; + + err = steering_anchor_create_fg_goto_table(ft_prio); + if (err) + goto destroy_fg_drop; + + err = steering_anchor_create_rule_drop(ft_prio); + if (err) + goto destroy_fg_goto_table; + + err = steering_anchor_create_rule_goto_table(ft_prio); + if (err) + goto destroy_rule_drop; + + return 0; + +destroy_rule_drop: + steering_anchor_destroy_rule_drop(ft_prio); +destroy_fg_goto_table: + steering_anchor_destroy_fg_goto_table(ft_prio); +destroy_fg_drop: + steering_anchor_destroy_fg_drop(ft_prio); +destroy_ft: + steering_anchor_destroy_ft(ft_prio); + + return err; +} + +static void mlx5_steering_anchor_destroy_res(struct mlx5_ib_flow_prio *ft_prio) +{ + steering_anchor_destroy_rule_goto_table(ft_prio); + steering_anchor_destroy_rule_drop(ft_prio); + steering_anchor_destroy_fg_goto_table(ft_prio); + steering_anchor_destroy_fg_drop(ft_prio); + steering_anchor_destroy_ft(ft_prio); +} + static int steering_anchor_cleanup(struct ib_uobject *uobject, enum rdma_remove_reason why, struct uverbs_attr_bundle *attrs) @@ -2035,6 +2264,9 @@ static int steering_anchor_cleanup(struct ib_uobject *uobject, return -EBUSY; mutex_lock(&obj->dev->flow_db->lock); + if (!--obj->ft_prio->anchor.rule_goto_table_ref) + steering_anchor_destroy_rule_goto_table(obj->ft_prio); + put_flow_table(obj->dev, obj->ft_prio, true); mutex_unlock(&obj->dev->flow_db->lock); @@ -2042,6 +2274,24 @@ static int steering_anchor_cleanup(struct ib_uobject *uobject, return 0; } +static void fs_cleanup_anchor(struct mlx5_ib_flow_prio *prio, + int count) +{ + while (count--) + mlx5_steering_anchor_destroy_res(&prio[count]); +} + +void mlx5_ib_fs_cleanup_anchor(struct mlx5_ib_dev *dev) +{ + fs_cleanup_anchor(dev->flow_db->prios, MLX5_IB_NUM_FLOW_FT); + fs_cleanup_anchor(dev->flow_db->egress_prios, MLX5_IB_NUM_FLOW_FT); + fs_cleanup_anchor(dev->flow_db->sniffer, MLX5_IB_NUM_SNIFFER_FTS); + fs_cleanup_anchor(dev->flow_db->egress, MLX5_IB_NUM_EGRESS_FTS); + fs_cleanup_anchor(dev->flow_db->fdb, MLX5_IB_NUM_FDB_FTS); + fs_cleanup_anchor(dev->flow_db->rdma_rx, MLX5_IB_NUM_FLOW_FT); + fs_cleanup_anchor(dev->flow_db->rdma_tx, MLX5_IB_NUM_FLOW_FT); +} + static int mlx5_ib_matcher_ns(struct uverbs_attr_bundle *attrs, struct mlx5_ib_flow_matcher *obj) { @@ -2182,21 +2432,31 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_STEERING_ANCHOR_CREATE)( return -ENOMEM; mutex_lock(&dev->flow_db->lock); + ft_prio = _get_flow_table(dev, priority, ns_type, 0); if (IS_ERR(ft_prio)) { - mutex_unlock(&dev->flow_db->lock); err = PTR_ERR(ft_prio); goto free_obj; } ft_prio->refcount++; - ft_id = mlx5_flow_table_id(ft_prio->flow_table); - mutex_unlock(&dev->flow_db->lock); + + if (!ft_prio->anchor.rule_goto_table_ref) { + err = steering_anchor_create_res(dev, ft_prio, ns_type); + if (err) + goto put_flow_table; + } + + ft_prio->anchor.rule_goto_table_ref++; + + ft_id = mlx5_flow_table_id(ft_prio->anchor.ft); err = uverbs_copy_to(attrs, MLX5_IB_ATTR_STEERING_ANCHOR_FT_ID, &ft_id, sizeof(ft_id)); if (err) - goto put_flow_table; + goto destroy_res; + + mutex_unlock(&dev->flow_db->lock); uobj->object = obj; obj->dev = dev; @@ -2205,8 +2465,10 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_STEERING_ANCHOR_CREATE)( return 0; +destroy_res: + --ft_prio->anchor.rule_goto_table_ref; + mlx5_steering_anchor_destroy_res(ft_prio); put_flow_table: - mutex_lock(&dev->flow_db->lock); put_flow_table(dev, ft_prio, true); mutex_unlock(&dev->flow_db->lock); free_obj: diff --git a/drivers/infiniband/hw/mlx5/fs.h b/drivers/infiniband/hw/mlx5/fs.h index ad320adaf32172e50a67401ac33eec9441b3be5a..b9734904f5f01ca89427c79c4851e0fc378318dd 100644 --- a/drivers/infiniband/hw/mlx5/fs.h +++ b/drivers/infiniband/hw/mlx5/fs.h @@ -10,6 +10,7 @@ #if IS_ENABLED(CONFIG_INFINIBAND_USER_ACCESS) int mlx5_ib_fs_init(struct mlx5_ib_dev *dev); +void mlx5_ib_fs_cleanup_anchor(struct mlx5_ib_dev *dev); #else static inline int mlx5_ib_fs_init(struct mlx5_ib_dev *dev) { @@ -21,9 +22,24 @@ static inline int mlx5_ib_fs_init(struct mlx5_ib_dev *dev) mutex_init(&dev->flow_db->lock); return 0; } + +inline void mlx5_ib_fs_cleanup_anchor(struct mlx5_ib_dev *dev) {} #endif + static inline void mlx5_ib_fs_cleanup(struct mlx5_ib_dev *dev) { + /* When a steering anchor is created, a special flow table is also + * created for the user to reference. Since the user can reference it, + * the kernel cannot trust that when the user destroys the steering + * anchor, they no longer reference the flow table. + * + * To address this issue, when a user destroys a steering anchor, only + * the flow steering rule in the table is destroyed, but the table + * itself is kept to deal with the above scenario. The remaining + * resources are only removed when the RDMA device is destroyed, which + * is a safe assumption that all references are gone. + */ + mlx5_ib_fs_cleanup_anchor(dev); kfree(dev->flow_db); } #endif /* _MLX5_IB_FS_H */ diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 5d45de223c43a9afed62a70265b9e32d75db4e10..f0b394ed7452a652461c42c99ef47f0e2d540a37 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -4275,6 +4275,9 @@ const struct mlx5_ib_profile raw_eth_profile = { STAGE_CREATE(MLX5_IB_STAGE_POST_IB_REG_UMR, mlx5_ib_stage_post_ib_reg_umr_init, NULL), + STAGE_CREATE(MLX5_IB_STAGE_DELAY_DROP, + mlx5_ib_stage_delay_drop_init, + mlx5_ib_stage_delay_drop_cleanup), STAGE_CREATE(MLX5_IB_STAGE_RESTRACK, mlx5_ib_restrack_init, NULL), diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h index efa4dc6e7dee1516fb15e5873ddb8bca928da6dd..2dfa6f49a6f4816c7c0bad61d5928a6e38187e5f 100644 --- a/drivers/infiniband/hw/mlx5/mlx5_ib.h +++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h @@ -237,8 +237,19 @@ enum { #define MLX5_IB_NUM_SNIFFER_FTS 2 #define MLX5_IB_NUM_EGRESS_FTS 1 #define MLX5_IB_NUM_FDB_FTS MLX5_BY_PASS_NUM_REGULAR_PRIOS + +struct mlx5_ib_anchor { + struct mlx5_flow_table *ft; + struct mlx5_flow_group *fg_goto_table; + struct mlx5_flow_group *fg_drop; + struct mlx5_flow_handle *rule_goto_table; + struct mlx5_flow_handle *rule_drop; + unsigned int rule_goto_table_ref; +}; + struct mlx5_ib_flow_prio { struct mlx5_flow_table *flow_table; + struct mlx5_ib_anchor anchor; unsigned int refcount; }; @@ -1587,6 +1598,9 @@ static inline bool mlx5_ib_lag_should_assign_affinity(struct mlx5_ib_dev *dev) MLX5_CAP_PORT_SELECTION(dev->mdev, port_select_flow_table_bypass)) return 0; + if (mlx5_lag_is_lacp_owner(dev->mdev) && !dev->lag_active) + return 0; + return dev->lag_active || (MLX5_CAP_GEN(dev->mdev, num_lag_ports) > 1 && MLX5_CAP_GEN(dev->mdev, lag_tx_port_affinity)); diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index 70ca8ffa9256b1d11a48bbc5d50a6506a5813671..78b96bfb4e6ac9df9bb24f0f5a29a76ed4520b0a 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -1237,6 +1237,9 @@ static int create_raw_packet_qp_tis(struct mlx5_ib_dev *dev, MLX5_SET(create_tis_in, in, uid, to_mpd(pd)->uid); MLX5_SET(tisc, tisc, transport_domain, tdn); + if (!mlx5_ib_lag_should_assign_affinity(dev) && + mlx5_lag_is_lacp_owner(dev->mdev)) + MLX5_SET(tisc, tisc, strict_lag_tx_port_affinity, 1); if (qp->flags & IB_QP_CREATE_SOURCE_QPN) MLX5_SET(tisc, tisc, underlay_qpn, qp->underlay_qpn); diff --git a/drivers/infiniband/sw/rxe/rxe_comp.c b/drivers/infiniband/sw/rxe/rxe_comp.c index db18ace74d2b53a8e0bbc68bc3d7cded8463d263..f46c5a5fd0aea4d4c0530d054949a8867013053a 100644 --- a/drivers/infiniband/sw/rxe/rxe_comp.c +++ b/drivers/infiniband/sw/rxe/rxe_comp.c @@ -115,15 +115,16 @@ static enum ib_wc_opcode wr_to_wc_opcode(enum ib_wr_opcode opcode) void retransmit_timer(struct timer_list *t) { struct rxe_qp *qp = from_timer(qp, t, retrans_timer); + unsigned long flags; rxe_dbg_qp(qp, "retransmit timer fired\n"); - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); if (qp->valid) { qp->comp.timeout = 1; rxe_sched_task(&qp->comp.task); } - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); } void rxe_comp_queue_pkt(struct rxe_qp *qp, struct sk_buff *skb) @@ -481,11 +482,13 @@ static void do_complete(struct rxe_qp *qp, struct rxe_send_wqe *wqe) static void comp_check_sq_drain_done(struct rxe_qp *qp) { - spin_lock_bh(&qp->state_lock); + unsigned long flags; + + spin_lock_irqsave(&qp->state_lock, flags); if (unlikely(qp_state(qp) == IB_QPS_SQD)) { if (qp->attr.sq_draining && qp->comp.psn == qp->req.psn) { qp->attr.sq_draining = 0; - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); if (qp->ibqp.event_handler) { struct ib_event ev; @@ -499,7 +502,7 @@ static void comp_check_sq_drain_done(struct rxe_qp *qp) return; } } - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); } static inline enum comp_state complete_ack(struct rxe_qp *qp, @@ -625,13 +628,15 @@ static void free_pkt(struct rxe_pkt_info *pkt) */ static void reset_retry_timer(struct rxe_qp *qp) { + unsigned long flags; + if (qp_type(qp) == IB_QPT_RC && qp->qp_timeout_jiffies) { - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); if (qp_state(qp) >= IB_QPS_RTS && psn_compare(qp->req.psn, qp->comp.psn) > 0) mod_timer(&qp->retrans_timer, jiffies + qp->qp_timeout_jiffies); - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); } } @@ -643,18 +648,19 @@ int rxe_completer(struct rxe_qp *qp) struct rxe_pkt_info *pkt = NULL; enum comp_state state; int ret; + unsigned long flags; - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); if (!qp->valid || qp_state(qp) == IB_QPS_ERR || qp_state(qp) == IB_QPS_RESET) { bool notify = qp->valid && (qp_state(qp) == IB_QPS_ERR); drain_resp_pkts(qp); flush_send_queue(qp, notify); - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); goto exit; } - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); if (qp->comp.timeout) { qp->comp.timeout_retry = 1; diff --git a/drivers/infiniband/sw/rxe/rxe_cq.c b/drivers/infiniband/sw/rxe/rxe_cq.c index 20ff0c0c46052efdcc6fc896be6d0667cda81da2..6ca2a05b6a2ab262594b7a3fbbc799f86fba4eff 100644 --- a/drivers/infiniband/sw/rxe/rxe_cq.c +++ b/drivers/infiniband/sw/rxe/rxe_cq.c @@ -113,8 +113,6 @@ int rxe_cq_post(struct rxe_cq *cq, struct rxe_cqe *cqe, int solicited) queue_advance_producer(cq->queue, QUEUE_TYPE_TO_CLIENT); - spin_unlock_irqrestore(&cq->cq_lock, flags); - if ((cq->notify == IB_CQ_NEXT_COMP) || (cq->notify == IB_CQ_SOLICITED && solicited)) { cq->notify = 0; @@ -122,6 +120,8 @@ int rxe_cq_post(struct rxe_cq *cq, struct rxe_cqe *cqe, int solicited) cq->ibcq.comp_handler(&cq->ibcq, cq->ibcq.cq_context); } + spin_unlock_irqrestore(&cq->cq_lock, flags); + return 0; } diff --git a/drivers/infiniband/sw/rxe/rxe_net.c b/drivers/infiniband/sw/rxe/rxe_net.c index 2bc7361152ea70f95c78f743098c4b6cc547fa16..cd59666158b18aa4a035f0e898e70d591274eed9 100644 --- a/drivers/infiniband/sw/rxe/rxe_net.c +++ b/drivers/infiniband/sw/rxe/rxe_net.c @@ -159,6 +159,9 @@ static int rxe_udp_encap_recv(struct sock *sk, struct sk_buff *skb) pkt->mask = RXE_GRH_MASK; pkt->paylen = be16_to_cpu(udph->len) - sizeof(*udph); + /* remove udp header */ + skb_pull(skb, sizeof(struct udphdr)); + rxe_rcv(skb); return 0; @@ -401,6 +404,9 @@ static int rxe_loopback(struct sk_buff *skb, struct rxe_pkt_info *pkt) return -EIO; } + /* remove udp header */ + skb_pull(skb, sizeof(struct udphdr)); + rxe_rcv(skb); return 0; @@ -412,15 +418,16 @@ int rxe_xmit_packet(struct rxe_qp *qp, struct rxe_pkt_info *pkt, int err; int is_request = pkt->mask & RXE_REQ_MASK; struct rxe_dev *rxe = to_rdev(qp->ibqp.device); + unsigned long flags; - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); if ((is_request && (qp_state(qp) < IB_QPS_RTS)) || (!is_request && (qp_state(qp) < IB_QPS_RTR))) { - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); rxe_dbg_qp(qp, "Packet dropped. QP is not in ready state\n"); goto drop; } - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); rxe_icrc_generate(skb, pkt); diff --git a/drivers/infiniband/sw/rxe/rxe_qp.c b/drivers/infiniband/sw/rxe/rxe_qp.c index c5451a4488ca35b4cd7cb21930f7fab1ca482784..a0f206431cf8eb82d21573888290f2826367b803 100644 --- a/drivers/infiniband/sw/rxe/rxe_qp.c +++ b/drivers/infiniband/sw/rxe/rxe_qp.c @@ -176,6 +176,9 @@ static void rxe_qp_init_misc(struct rxe_dev *rxe, struct rxe_qp *qp, spin_lock_init(&qp->rq.producer_lock); spin_lock_init(&qp->rq.consumer_lock); + skb_queue_head_init(&qp->req_pkts); + skb_queue_head_init(&qp->resp_pkts); + atomic_set(&qp->ssn, 0); atomic_set(&qp->skb_out, 0); } @@ -234,8 +237,6 @@ static int rxe_qp_init_req(struct rxe_dev *rxe, struct rxe_qp *qp, qp->req.opcode = -1; qp->comp.opcode = -1; - skb_queue_head_init(&qp->req_pkts); - rxe_init_task(&qp->req.task, qp, rxe_requester); rxe_init_task(&qp->comp.task, qp, rxe_completer); @@ -279,8 +280,6 @@ static int rxe_qp_init_resp(struct rxe_dev *rxe, struct rxe_qp *qp, } } - skb_queue_head_init(&qp->resp_pkts); - rxe_init_task(&qp->resp.task, qp, rxe_responder); qp->resp.opcode = OPCODE_NONE; @@ -300,6 +299,7 @@ int rxe_qp_from_init(struct rxe_dev *rxe, struct rxe_qp *qp, struct rxe_pd *pd, struct rxe_cq *rcq = to_rcq(init->recv_cq); struct rxe_cq *scq = to_rcq(init->send_cq); struct rxe_srq *srq = init->srq ? to_rsrq(init->srq) : NULL; + unsigned long flags; rxe_get(pd); rxe_get(rcq); @@ -325,10 +325,10 @@ int rxe_qp_from_init(struct rxe_dev *rxe, struct rxe_qp *qp, struct rxe_pd *pd, if (err) goto err2; - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); qp->attr.qp_state = IB_QPS_RESET; qp->valid = 1; - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); return 0; @@ -492,24 +492,28 @@ static void rxe_qp_reset(struct rxe_qp *qp) /* move the qp to the error state */ void rxe_qp_error(struct rxe_qp *qp) { - spin_lock_bh(&qp->state_lock); + unsigned long flags; + + spin_lock_irqsave(&qp->state_lock, flags); qp->attr.qp_state = IB_QPS_ERR; /* drain work and packet queues */ rxe_sched_task(&qp->resp.task); rxe_sched_task(&qp->comp.task); rxe_sched_task(&qp->req.task); - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); } static void rxe_qp_sqd(struct rxe_qp *qp, struct ib_qp_attr *attr, int mask) { - spin_lock_bh(&qp->state_lock); + unsigned long flags; + + spin_lock_irqsave(&qp->state_lock, flags); qp->attr.sq_draining = 1; rxe_sched_task(&qp->comp.task); rxe_sched_task(&qp->req.task); - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); } /* caller should hold qp->state_lock */ @@ -555,14 +559,16 @@ int rxe_qp_from_attr(struct rxe_qp *qp, struct ib_qp_attr *attr, int mask, qp->attr.cur_qp_state = attr->qp_state; if (mask & IB_QP_STATE) { - spin_lock_bh(&qp->state_lock); + unsigned long flags; + + spin_lock_irqsave(&qp->state_lock, flags); err = __qp_chk_state(qp, attr, mask); if (!err) { qp->attr.qp_state = attr->qp_state; rxe_dbg_qp(qp, "state -> %s\n", qps2str[attr->qp_state]); } - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); if (err) return err; @@ -688,6 +694,8 @@ int rxe_qp_from_attr(struct rxe_qp *qp, struct ib_qp_attr *attr, int mask, /* called by the query qp verb */ int rxe_qp_to_attr(struct rxe_qp *qp, struct ib_qp_attr *attr, int mask) { + unsigned long flags; + *attr = qp->attr; attr->rq_psn = qp->resp.psn; @@ -708,12 +716,13 @@ int rxe_qp_to_attr(struct rxe_qp *qp, struct ib_qp_attr *attr, int mask) /* Applications that get this state typically spin on it. * Yield the processor */ - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); if (qp->attr.sq_draining) { - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); cond_resched(); + } else { + spin_unlock_irqrestore(&qp->state_lock, flags); } - spin_unlock_bh(&qp->state_lock); return 0; } @@ -736,10 +745,11 @@ int rxe_qp_chk_destroy(struct rxe_qp *qp) static void rxe_qp_do_cleanup(struct work_struct *work) { struct rxe_qp *qp = container_of(work, typeof(*qp), cleanup_work.work); + unsigned long flags; - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); qp->valid = 0; - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); qp->qp_timeout_jiffies = 0; if (qp_type(qp) == IB_QPT_RC) { diff --git a/drivers/infiniband/sw/rxe/rxe_recv.c b/drivers/infiniband/sw/rxe/rxe_recv.c index 2f953cc74256d90c9015bd413520be8369a6ff2b..5861e42440490dd1014efad9c0a52c2d9ebb5bc4 100644 --- a/drivers/infiniband/sw/rxe/rxe_recv.c +++ b/drivers/infiniband/sw/rxe/rxe_recv.c @@ -14,6 +14,7 @@ static int check_type_state(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, struct rxe_qp *qp) { unsigned int pkt_type; + unsigned long flags; if (unlikely(!qp->valid)) return -EINVAL; @@ -38,19 +39,19 @@ static int check_type_state(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, return -EINVAL; } - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); if (pkt->mask & RXE_REQ_MASK) { if (unlikely(qp_state(qp) < IB_QPS_RTR)) { - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); return -EINVAL; } } else { if (unlikely(qp_state(qp) < IB_QPS_RTS)) { - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); return -EINVAL; } } - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); return 0; } diff --git a/drivers/infiniband/sw/rxe/rxe_req.c b/drivers/infiniband/sw/rxe/rxe_req.c index 65134a9aefe7bfe5f9ea71f57d237af5f61170ad..5fe7cbae30313c0ce5f2ab61e00e33e6778b19ad 100644 --- a/drivers/infiniband/sw/rxe/rxe_req.c +++ b/drivers/infiniband/sw/rxe/rxe_req.c @@ -99,17 +99,18 @@ static void req_retry(struct rxe_qp *qp) void rnr_nak_timer(struct timer_list *t) { struct rxe_qp *qp = from_timer(qp, t, rnr_nak_timer); + unsigned long flags; rxe_dbg_qp(qp, "nak timer fired\n"); - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); if (qp->valid) { /* request a send queue retry */ qp->req.need_retry = 1; qp->req.wait_for_rnr_timer = 0; rxe_sched_task(&qp->req.task); } - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); } static void req_check_sq_drain_done(struct rxe_qp *qp) @@ -118,8 +119,9 @@ static void req_check_sq_drain_done(struct rxe_qp *qp) unsigned int index; unsigned int cons; struct rxe_send_wqe *wqe; + unsigned long flags; - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); if (qp_state(qp) == IB_QPS_SQD) { q = qp->sq.queue; index = qp->req.wqe_index; @@ -140,7 +142,7 @@ static void req_check_sq_drain_done(struct rxe_qp *qp) break; qp->attr.sq_draining = 0; - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); if (qp->ibqp.event_handler) { struct ib_event ev; @@ -154,7 +156,7 @@ static void req_check_sq_drain_done(struct rxe_qp *qp) return; } while (0); } - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); } static struct rxe_send_wqe *__req_next_wqe(struct rxe_qp *qp) @@ -173,6 +175,7 @@ static struct rxe_send_wqe *__req_next_wqe(struct rxe_qp *qp) static struct rxe_send_wqe *req_next_wqe(struct rxe_qp *qp) { struct rxe_send_wqe *wqe; + unsigned long flags; req_check_sq_drain_done(qp); @@ -180,13 +183,13 @@ static struct rxe_send_wqe *req_next_wqe(struct rxe_qp *qp) if (wqe == NULL) return NULL; - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); if (unlikely((qp_state(qp) == IB_QPS_SQD) && (wqe->state != wqe_state_processing))) { - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); return NULL; } - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); wqe->mask = wr_opcode_mask(wqe->wr.opcode, qp); return wqe; @@ -676,16 +679,17 @@ int rxe_requester(struct rxe_qp *qp) struct rxe_queue *q = qp->sq.queue; struct rxe_ah *ah; struct rxe_av *av; + unsigned long flags; - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); if (unlikely(!qp->valid)) { - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); goto exit; } if (unlikely(qp_state(qp) == IB_QPS_ERR)) { wqe = __req_next_wqe(qp); - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); if (wqe) goto err; else @@ -700,10 +704,10 @@ int rxe_requester(struct rxe_qp *qp) qp->req.wait_psn = 0; qp->req.need_retry = 0; qp->req.wait_for_rnr_timer = 0; - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); goto exit; } - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); /* we come here if the retransmit timer has fired * or if the rnr timer has fired. If the retransmit diff --git a/drivers/infiniband/sw/rxe/rxe_resp.c b/drivers/infiniband/sw/rxe/rxe_resp.c index 68f6cd188d8ed2f943d89d15438f2189ed4c16fe..ee68306555b990ca169175bae2c31003737d62aa 100644 --- a/drivers/infiniband/sw/rxe/rxe_resp.c +++ b/drivers/infiniband/sw/rxe/rxe_resp.c @@ -489,8 +489,9 @@ static enum resp_states check_rkey(struct rxe_qp *qp, if (mw->access & IB_ZERO_BASED) qp->resp.offset = mw->addr; - rxe_put(mw); rxe_get(mr); + rxe_put(mw); + mw = NULL; } else { mr = lookup_mr(qp->pd, access, rkey, RXE_LOOKUP_REMOTE); if (!mr) { @@ -1047,6 +1048,7 @@ static enum resp_states do_complete(struct rxe_qp *qp, struct ib_uverbs_wc *uwc = &cqe.uibwc; struct rxe_recv_wqe *wqe = qp->resp.wqe; struct rxe_dev *rxe = to_rdev(qp->ibqp.device); + unsigned long flags; if (!wqe) goto finish; @@ -1137,12 +1139,12 @@ static enum resp_states do_complete(struct rxe_qp *qp, return RESPST_ERR_CQ_OVERFLOW; finish: - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); if (unlikely(qp_state(qp) == IB_QPS_ERR)) { - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); return RESPST_CHK_RESOURCE; } - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); if (unlikely(!pkt)) return RESPST_DONE; @@ -1468,18 +1470,19 @@ int rxe_responder(struct rxe_qp *qp) enum resp_states state; struct rxe_pkt_info *pkt = NULL; int ret; + unsigned long flags; - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); if (!qp->valid || qp_state(qp) == IB_QPS_ERR || qp_state(qp) == IB_QPS_RESET) { bool notify = qp->valid && (qp_state(qp) == IB_QPS_ERR); drain_req_pkts(qp); flush_recv_queue(qp, notify); - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); goto exit; } - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); qp->resp.aeth_syndrome = AETH_ACK_UNLIMITED; diff --git a/drivers/infiniband/sw/rxe/rxe_verbs.c b/drivers/infiniband/sw/rxe/rxe_verbs.c index dea605b7f68335f78c94d8464594071aef2dd351..4d8f6b8051ff7db5509b2e9961af54c1479373cb 100644 --- a/drivers/infiniband/sw/rxe/rxe_verbs.c +++ b/drivers/infiniband/sw/rxe/rxe_verbs.c @@ -904,10 +904,10 @@ static int rxe_post_send_kernel(struct rxe_qp *qp, if (!err) rxe_sched_task(&qp->req.task); - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); if (qp_state(qp) == IB_QPS_ERR) rxe_sched_task(&qp->comp.task); - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); return err; } @@ -917,22 +917,23 @@ static int rxe_post_send(struct ib_qp *ibqp, const struct ib_send_wr *wr, { struct rxe_qp *qp = to_rqp(ibqp); int err; + unsigned long flags; - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); /* caller has already called destroy_qp */ if (WARN_ON_ONCE(!qp->valid)) { - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); rxe_err_qp(qp, "qp has been destroyed"); return -EINVAL; } if (unlikely(qp_state(qp) < IB_QPS_RTS)) { - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); *bad_wr = wr; rxe_err_qp(qp, "qp not ready to send"); return -EINVAL; } - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); if (qp->is_user) { /* Utilize process context to do protocol processing */ @@ -1008,22 +1009,22 @@ static int rxe_post_recv(struct ib_qp *ibqp, const struct ib_recv_wr *wr, struct rxe_rq *rq = &qp->rq; unsigned long flags; - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); /* caller has already called destroy_qp */ if (WARN_ON_ONCE(!qp->valid)) { - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); rxe_err_qp(qp, "qp has been destroyed"); return -EINVAL; } /* see C10-97.2.1 */ if (unlikely((qp_state(qp) < IB_QPS_INIT))) { - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); *bad_wr = wr; rxe_dbg_qp(qp, "qp not ready to post recv"); return -EINVAL; } - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); if (unlikely(qp->srq)) { *bad_wr = wr; @@ -1044,10 +1045,10 @@ static int rxe_post_recv(struct ib_qp *ibqp, const struct ib_recv_wr *wr, spin_unlock_irqrestore(&rq->producer_lock, flags); - spin_lock_bh(&qp->state_lock); + spin_lock_irqsave(&qp->state_lock, flags); if (qp_state(qp) == IB_QPS_ERR) rxe_sched_task(&qp->resp.task); - spin_unlock_bh(&qp->state_lock); + spin_unlock_irqrestore(&qp->state_lock, flags); return err; } diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index f290cd49698ea27b1f143587fbc666c89c28dcc8..92e1e7587af8b4131f37e7c6887747f7649b86c1 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -657,9 +657,13 @@ static int isert_connect_error(struct rdma_cm_id *cma_id) { struct isert_conn *isert_conn = cma_id->qp->qp_context; + struct isert_np *isert_np = cma_id->context; ib_drain_qp(isert_conn->qp); + + mutex_lock(&isert_np->mutex); list_del_init(&isert_conn->node); + mutex_unlock(&isert_np->mutex); isert_conn->cm_id = NULL; isert_put_conn(isert_conn); @@ -2431,6 +2435,7 @@ isert_free_np(struct iscsi_np *np) { struct isert_np *isert_np = np->np_context; struct isert_conn *isert_conn, *n; + LIST_HEAD(drop_conn_list); if (isert_np->cm_id) rdma_destroy_id(isert_np->cm_id); @@ -2450,7 +2455,7 @@ isert_free_np(struct iscsi_np *np) node) { isert_info("cleaning isert_conn %p state (%d)\n", isert_conn, isert_conn->state); - isert_connect_release(isert_conn); + list_move_tail(&isert_conn->node, &drop_conn_list); } } @@ -2461,11 +2466,16 @@ isert_free_np(struct iscsi_np *np) node) { isert_info("cleaning isert_conn %p state (%d)\n", isert_conn, isert_conn->state); - isert_connect_release(isert_conn); + list_move_tail(&isert_conn->node, &drop_conn_list); } } mutex_unlock(&isert_np->mutex); + list_for_each_entry_safe(isert_conn, n, &drop_conn_list, node) { + list_del_init(&isert_conn->node); + isert_connect_release(isert_conn); + } + np->np_context = NULL; kfree(isert_np); } @@ -2560,8 +2570,6 @@ static void isert_wait_conn(struct iscsit_conn *conn) isert_put_unsol_pending_cmds(conn); isert_wait4cmds(conn); isert_wait4logout(isert_conn); - - queue_work(isert_release_wq, &isert_conn->release_work); } static void isert_free_conn(struct iscsit_conn *conn) diff --git a/drivers/infiniband/ulp/rtrs/rtrs-clt.c b/drivers/infiniband/ulp/rtrs/rtrs-clt.c index edb2e3a25880b47bd044f08c5039a8a330fec729..cfb50bfe53c3c9a1eeecd5398d2883ecdeb5a2cf 100644 --- a/drivers/infiniband/ulp/rtrs/rtrs-clt.c +++ b/drivers/infiniband/ulp/rtrs/rtrs-clt.c @@ -2040,6 +2040,7 @@ static int rtrs_clt_rdma_cm_handler(struct rdma_cm_id *cm_id, return 0; } +/* The caller should do the cleanup in case of error */ static int create_cm(struct rtrs_clt_con *con) { struct rtrs_path *s = con->c.path; @@ -2062,14 +2063,14 @@ static int create_cm(struct rtrs_clt_con *con) err = rdma_set_reuseaddr(cm_id, 1); if (err != 0) { rtrs_err(s, "Set address reuse failed, err: %d\n", err); - goto destroy_cm; + return err; } err = rdma_resolve_addr(cm_id, (struct sockaddr *)&clt_path->s.src_addr, (struct sockaddr *)&clt_path->s.dst_addr, RTRS_CONNECT_TIMEOUT_MS); if (err) { rtrs_err(s, "Failed to resolve address, err: %d\n", err); - goto destroy_cm; + return err; } /* * Combine connection status and session events. This is needed @@ -2084,29 +2085,15 @@ static int create_cm(struct rtrs_clt_con *con) if (err == 0) err = -ETIMEDOUT; /* Timedout or interrupted */ - goto errr; - } - if (con->cm_err < 0) { - err = con->cm_err; - goto errr; + return err; } - if (READ_ONCE(clt_path->state) != RTRS_CLT_CONNECTING) { + if (con->cm_err < 0) + return con->cm_err; + if (READ_ONCE(clt_path->state) != RTRS_CLT_CONNECTING) /* Device removal */ - err = -ECONNABORTED; - goto errr; - } + return -ECONNABORTED; return 0; - -errr: - stop_cm(con); - mutex_lock(&con->con_mutex); - destroy_con_cq_qp(con); - mutex_unlock(&con->con_mutex); -destroy_cm: - destroy_cm(con); - - return err; } static void rtrs_clt_path_up(struct rtrs_clt_path *clt_path) @@ -2334,7 +2321,7 @@ static void rtrs_clt_close_work(struct work_struct *work) static int init_conns(struct rtrs_clt_path *clt_path) { unsigned int cid; - int err; + int err, i; /* * On every new session connections increase reconnect counter @@ -2350,10 +2337,8 @@ static int init_conns(struct rtrs_clt_path *clt_path) goto destroy; err = create_cm(to_clt_con(clt_path->s.con[cid])); - if (err) { - destroy_con(to_clt_con(clt_path->s.con[cid])); + if (err) goto destroy; - } } err = alloc_path_reqs(clt_path); if (err) @@ -2364,15 +2349,21 @@ static int init_conns(struct rtrs_clt_path *clt_path) return 0; destroy: - while (cid--) { - struct rtrs_clt_con *con = to_clt_con(clt_path->s.con[cid]); + /* Make sure we do the cleanup in the order they are created */ + for (i = 0; i <= cid; i++) { + struct rtrs_clt_con *con; - stop_cm(con); + if (!clt_path->s.con[i]) + break; - mutex_lock(&con->con_mutex); - destroy_con_cq_qp(con); - mutex_unlock(&con->con_mutex); - destroy_cm(con); + con = to_clt_con(clt_path->s.con[i]); + if (con->c.cm_id) { + stop_cm(con); + mutex_lock(&con->con_mutex); + destroy_con_cq_qp(con); + mutex_unlock(&con->con_mutex); + destroy_cm(con); + } destroy_con(con); } /* diff --git a/drivers/infiniband/ulp/rtrs/rtrs.c b/drivers/infiniband/ulp/rtrs/rtrs.c index 4bf9d868cc522bd8b7631f0f28aba9199385800c..3696f367ff5151333234c707ccad326385fdd59c 100644 --- a/drivers/infiniband/ulp/rtrs/rtrs.c +++ b/drivers/infiniband/ulp/rtrs/rtrs.c @@ -37,8 +37,10 @@ struct rtrs_iu *rtrs_iu_alloc(u32 iu_num, size_t size, gfp_t gfp_mask, goto err; iu->dma_addr = ib_dma_map_single(dma_dev, iu->buf, size, dir); - if (ib_dma_mapping_error(dma_dev, iu->dma_addr)) + if (ib_dma_mapping_error(dma_dev, iu->dma_addr)) { + kfree(iu->buf); goto err; + } iu->cqe.done = done; iu->size = size; diff --git a/drivers/input/input.c b/drivers/input/input.c index 37e876d45eb9c23775465da445860bd20aa961aa..641eb86f276e6da05e86ea1ddd9ae10069cefe7a 100644 --- a/drivers/input/input.c +++ b/drivers/input/input.c @@ -703,7 +703,7 @@ void input_close_device(struct input_handle *handle) __input_release_device(handle); - if (!dev->inhibited && !--dev->users) { + if (!--dev->users && !dev->inhibited) { if (dev->poller) input_dev_poller_stop(dev->poller); if (dev->close) diff --git a/drivers/input/joystick/xpad.c b/drivers/input/joystick/xpad.c index 28be88e0e96abed47f0c0453e741b934e53ba15b..f33622fe946f657380155fd1db111ab9d23e39f3 100644 --- a/drivers/input/joystick/xpad.c +++ b/drivers/input/joystick/xpad.c @@ -281,7 +281,6 @@ static const struct xpad_device { { 0x1430, 0xf801, "RedOctane Controller", 0, XTYPE_XBOX360 }, { 0x146b, 0x0601, "BigBen Interactive XBOX 360 Controller", 0, XTYPE_XBOX360 }, { 0x146b, 0x0604, "Bigben Interactive DAIJA Arcade Stick", MAP_TRIGGERS_TO_BUTTONS, XTYPE_XBOX360 }, - { 0x1532, 0x0037, "Razer Sabertooth", 0, XTYPE_XBOX360 }, { 0x1532, 0x0a00, "Razer Atrox Arcade Stick", MAP_TRIGGERS_TO_BUTTONS, XTYPE_XBOXONE }, { 0x1532, 0x0a03, "Razer Wildcat", 0, XTYPE_XBOXONE }, { 0x15e4, 0x3f00, "Power A Mini Pro Elite", 0, XTYPE_XBOX360 }, diff --git a/drivers/input/misc/soc_button_array.c b/drivers/input/misc/soc_button_array.c index 09489380afda7a78d1f331e1c092f6d90d7b0f66..e79f5497948b8c047039c29c504317d009b5ef4c 100644 --- a/drivers/input/misc/soc_button_array.c +++ b/drivers/input/misc/soc_button_array.c @@ -108,6 +108,27 @@ static const struct dmi_system_id dmi_use_low_level_irq[] = { {} /* Terminating entry */ }; +/* + * Some devices have a wrong entry which points to a GPIO which is + * required in another driver, so this driver must not claim it. + */ +static const struct dmi_system_id dmi_invalid_acpi_index[] = { + { + /* + * Lenovo Yoga Book X90F / X90L, the PNP0C40 home button entry + * points to a GPIO which is not a home button and which is + * required by the lenovo-yogabook driver. + */ + .matches = { + DMI_EXACT_MATCH(DMI_SYS_VENDOR, "Intel Corporation"), + DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "CHERRYVIEW D1 PLATFORM"), + DMI_EXACT_MATCH(DMI_PRODUCT_VERSION, "YETI-11"), + }, + .driver_data = (void *)1l, + }, + {} /* Terminating entry */ +}; + /* * Get the Nth GPIO number from the ACPI object. */ @@ -137,6 +158,8 @@ soc_button_device_create(struct platform_device *pdev, struct platform_device *pd; struct gpio_keys_button *gpio_keys; struct gpio_keys_platform_data *gpio_keys_pdata; + const struct dmi_system_id *dmi_id; + int invalid_acpi_index = -1; int error, gpio, irq; int n_buttons = 0; @@ -154,10 +177,17 @@ soc_button_device_create(struct platform_device *pdev, gpio_keys = (void *)(gpio_keys_pdata + 1); n_buttons = 0; + dmi_id = dmi_first_match(dmi_invalid_acpi_index); + if (dmi_id) + invalid_acpi_index = (long)dmi_id->driver_data; + for (info = button_info; info->name; info++) { if (info->autorepeat != autorepeat) continue; + if (info->acpi_index == invalid_acpi_index) + continue; + error = soc_button_lookup_gpio(&pdev->dev, info->acpi_index, &gpio, &irq); if (error || irq < 0) { /* diff --git a/drivers/input/mouse/elantech.c b/drivers/input/mouse/elantech.c index ece97f8c6a3e399365b3f7c6a7e5d3f923c06ca7..2118b2075f437c2bfda8d3ec68b97ff831c3d74a 100644 --- a/drivers/input/mouse/elantech.c +++ b/drivers/input/mouse/elantech.c @@ -674,10 +674,11 @@ static void process_packet_head_v4(struct psmouse *psmouse) struct input_dev *dev = psmouse->dev; struct elantech_data *etd = psmouse->private; unsigned char *packet = psmouse->packet; - int id = ((packet[3] & 0xe0) >> 5) - 1; + int id; int pres, traces; - if (id < 0) + id = ((packet[3] & 0xe0) >> 5) - 1; + if (id < 0 || id >= ETP_MAX_FINGERS) return; etd->mt[id].x = ((packet[1] & 0x0f) << 8) | packet[2]; @@ -707,7 +708,7 @@ static void process_packet_motion_v4(struct psmouse *psmouse) int id, sid; id = ((packet[0] & 0xe0) >> 5) - 1; - if (id < 0) + if (id < 0 || id >= ETP_MAX_FINGERS) return; sid = ((packet[3] & 0xe0) >> 5) - 1; @@ -728,7 +729,7 @@ static void process_packet_motion_v4(struct psmouse *psmouse) input_report_abs(dev, ABS_MT_POSITION_X, etd->mt[id].x); input_report_abs(dev, ABS_MT_POSITION_Y, etd->mt[id].y); - if (sid >= 0) { + if (sid >= 0 && sid < ETP_MAX_FINGERS) { etd->mt[sid].x += delta_x2 * weight; etd->mt[sid].y -= delta_y2 * weight; input_mt_slot(dev, sid); diff --git a/drivers/input/touchscreen/cyttsp5.c b/drivers/input/touchscreen/cyttsp5.c index 30102cb80fac823ae47a27cc64a540a78211e656..3c9d07218f48dfc52bb59fe0be008d8e4915f586 100644 --- a/drivers/input/touchscreen/cyttsp5.c +++ b/drivers/input/touchscreen/cyttsp5.c @@ -560,7 +560,7 @@ static int cyttsp5_hid_output_get_sysinfo(struct cyttsp5 *ts) static int cyttsp5_hid_output_bl_launch_app(struct cyttsp5 *ts) { int rc; - u8 cmd[HID_OUTPUT_BL_LAUNCH_APP]; + u8 cmd[HID_OUTPUT_BL_LAUNCH_APP_SIZE]; u16 crc; put_unaligned_le16(HID_OUTPUT_BL_LAUNCH_APP_SIZE, cmd); diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig index db98c3f86e8c8b87f4c9c7e45b1f1b775f72c2f0..4d800601e8ecd6002c462383e4fc6d2cc186cb03 100644 --- a/drivers/iommu/Kconfig +++ b/drivers/iommu/Kconfig @@ -282,6 +282,7 @@ config EXYNOS_IOMMU_DEBUG config IPMMU_VMSA bool "Renesas VMSA-compatible IPMMU" depends on ARCH_RENESAS || COMPILE_TEST + depends on ARM || ARM64 || COMPILE_TEST depends on !GENERIC_ATOMIC64 # for IOMMU_IO_PGTABLE_LPAE select IOMMU_API select IOMMU_IO_PGTABLE_LPAE @@ -417,22 +418,6 @@ config S390_IOMMU help Support for the IOMMU API for s390 PCI devices. -config S390_CCW_IOMMU - bool "S390 CCW IOMMU Support" - depends on S390 && CCW || COMPILE_TEST - select IOMMU_API - help - Enables bits of IOMMU API required by VFIO. The iommu_ops - is not implemented as it is not necessary for VFIO. - -config S390_AP_IOMMU - bool "S390 AP IOMMU Support" - depends on S390 && ZCRYPT || COMPILE_TEST - select IOMMU_API - help - Enables bits of IOMMU API required by VFIO. The iommu_ops - is not implemented as it is not necessary for VFIO. - config MTK_IOMMU tristate "MediaTek IOMMU Support" depends on ARCH_MEDIATEK || COMPILE_TEST diff --git a/drivers/iommu/amd/amd_iommu.h b/drivers/iommu/amd/amd_iommu.h index e98f20a9bdd82e6be58e4d9938094f18979482b6..9beeceb9d825dfeae9e2aa9229e031513680ba09 100644 --- a/drivers/iommu/amd/amd_iommu.h +++ b/drivers/iommu/amd/amd_iommu.h @@ -15,9 +15,7 @@ extern irqreturn_t amd_iommu_int_thread(int irq, void *data); extern irqreturn_t amd_iommu_int_handler(int irq, void *data); extern void amd_iommu_apply_erratum_63(struct amd_iommu *iommu, u16 devid); extern void amd_iommu_restart_event_logging(struct amd_iommu *iommu); -extern int amd_iommu_init_devices(void); -extern void amd_iommu_uninit_devices(void); -extern void amd_iommu_init_notifier(void); +extern void amd_iommu_restart_ga_log(struct amd_iommu *iommu); extern void amd_iommu_set_rlookup_table(struct amd_iommu *iommu, u16 devid); #ifdef CONFIG_AMD_IOMMU_DEBUGFS diff --git a/drivers/iommu/amd/init.c b/drivers/iommu/amd/init.c index 329a406cc37de8f84e88acde441eea20ce65599b..c2d80a4e5fb066c69cfe0f8cab1856894bfbb2e3 100644 --- a/drivers/iommu/amd/init.c +++ b/drivers/iommu/amd/init.c @@ -758,6 +758,30 @@ void amd_iommu_restart_event_logging(struct amd_iommu *iommu) iommu_feature_enable(iommu, CONTROL_EVT_LOG_EN); } +/* + * This function restarts event logging in case the IOMMU experienced + * an GA log overflow. + */ +void amd_iommu_restart_ga_log(struct amd_iommu *iommu) +{ + u32 status; + + status = readl(iommu->mmio_base + MMIO_STATUS_OFFSET); + if (status & MMIO_STATUS_GALOG_RUN_MASK) + return; + + pr_info_ratelimited("IOMMU GA Log restarting\n"); + + iommu_feature_disable(iommu, CONTROL_GALOG_EN); + iommu_feature_disable(iommu, CONTROL_GAINT_EN); + + writel(MMIO_STATUS_GALOG_OVERFLOW_MASK, + iommu->mmio_base + MMIO_STATUS_OFFSET); + + iommu_feature_enable(iommu, CONTROL_GAINT_EN); + iommu_feature_enable(iommu, CONTROL_GALOG_EN); +} + /* * This function resets the command buffer if the IOMMU stopped fetching * commands from it. diff --git a/drivers/iommu/amd/iommu.c b/drivers/iommu/amd/iommu.c index 4a314647d1f792e3ef74538baedb654b8e8a2f42..dc1ec684977542ea5aa1af73a055ff962cdbe3d0 100644 --- a/drivers/iommu/amd/iommu.c +++ b/drivers/iommu/amd/iommu.c @@ -845,6 +845,7 @@ amd_iommu_set_pci_msi_domain(struct device *dev, struct amd_iommu *iommu) { } (MMIO_STATUS_EVT_OVERFLOW_INT_MASK | \ MMIO_STATUS_EVT_INT_MASK | \ MMIO_STATUS_PPR_INT_MASK | \ + MMIO_STATUS_GALOG_OVERFLOW_MASK | \ MMIO_STATUS_GALOG_INT_MASK) irqreturn_t amd_iommu_int_thread(int irq, void *data) @@ -868,10 +869,16 @@ irqreturn_t amd_iommu_int_thread(int irq, void *data) } #ifdef CONFIG_IRQ_REMAP - if (status & MMIO_STATUS_GALOG_INT_MASK) { + if (status & (MMIO_STATUS_GALOG_INT_MASK | + MMIO_STATUS_GALOG_OVERFLOW_MASK)) { pr_devel("Processing IOMMU GA Log\n"); iommu_poll_ga_log(iommu); } + + if (status & MMIO_STATUS_GALOG_OVERFLOW_MASK) { + pr_info_ratelimited("IOMMU GA Log overflow\n"); + amd_iommu_restart_ga_log(iommu); + } #endif if (status & MMIO_STATUS_EVT_OVERFLOW_INT_MASK) { @@ -2067,7 +2074,7 @@ static struct protection_domain *protection_domain_alloc(unsigned int type) { struct io_pgtable_ops *pgtbl_ops; struct protection_domain *domain; - int pgtable = amd_iommu_pgtable; + int pgtable; int mode = DEFAULT_PGTABLE_LEVEL; int ret; @@ -2084,6 +2091,10 @@ static struct protection_domain *protection_domain_alloc(unsigned int type) mode = PAGE_MODE_NONE; } else if (type == IOMMU_DOMAIN_UNMANAGED) { pgtable = AMD_IOMMU_V1; + } else if (type == IOMMU_DOMAIN_DMA || type == IOMMU_DOMAIN_DMA_FQ) { + pgtable = amd_iommu_pgtable; + } else { + return NULL; } switch (pgtable) { @@ -2118,6 +2129,15 @@ static struct protection_domain *protection_domain_alloc(unsigned int type) return NULL; } +static inline u64 dma_max_address(void) +{ + if (amd_iommu_pgtable == AMD_IOMMU_V1) + return ~0ULL; + + /* V2 with 4/5 level page table */ + return ((1ULL << PM_LEVEL_SHIFT(amd_iommu_gpt_level)) - 1); +} + static struct iommu_domain *amd_iommu_domain_alloc(unsigned type) { struct protection_domain *domain; @@ -2134,7 +2154,7 @@ static struct iommu_domain *amd_iommu_domain_alloc(unsigned type) return NULL; domain->domain.geometry.aperture_start = 0; - domain->domain.geometry.aperture_end = ~0ULL; + domain->domain.geometry.aperture_end = dma_max_address(); domain->domain.geometry.force_aperture = true; return &domain->domain; @@ -2387,7 +2407,7 @@ static void amd_iommu_iotlb_sync(struct iommu_domain *domain, unsigned long flags; spin_lock_irqsave(&dom->lock, flags); - domain_flush_pages(dom, gather->start, gather->end - gather->start, 1); + domain_flush_pages(dom, gather->start, gather->end - gather->start + 1, 1); amd_iommu_domain_flush_complete(dom); spin_unlock_irqrestore(&dom->lock, flags); } @@ -3493,8 +3513,7 @@ int amd_iommu_activate_guest_mode(void *data) struct irte_ga *entry = (struct irte_ga *) ir_data->entry; u64 valid; - if (!AMD_IOMMU_GUEST_IR_VAPIC(amd_iommu_guest_ir) || - !entry || entry->lo.fields_vapic.guest_mode) + if (!AMD_IOMMU_GUEST_IR_VAPIC(amd_iommu_guest_ir) || !entry) return 0; valid = entry->lo.fields_vapic.valid; diff --git a/drivers/iommu/arm/arm-smmu/arm-smmu-qcom.c b/drivers/iommu/arm/arm-smmu/arm-smmu-qcom.c index ae09c627bc8448b962f95f581ef213ff06b0ab58..c71afda79d6442f57026b945fd7ff8fe08cd35a4 100644 --- a/drivers/iommu/arm/arm-smmu/arm-smmu-qcom.c +++ b/drivers/iommu/arm/arm-smmu/arm-smmu-qcom.c @@ -517,6 +517,7 @@ static const struct of_device_id __maybe_unused qcom_smmu_impl_of_match[] = { { .compatible = "qcom,qcm2290-smmu-500", .data = &qcom_smmu_500_impl0_data }, { .compatible = "qcom,qdu1000-smmu-500", .data = &qcom_smmu_500_impl0_data }, { .compatible = "qcom,sc7180-smmu-500", .data = &qcom_smmu_500_impl0_data }, + { .compatible = "qcom,sc7180-smmu-v2", .data = &qcom_smmu_v2_data }, { .compatible = "qcom,sc7280-smmu-500", .data = &qcom_smmu_500_impl0_data }, { .compatible = "qcom,sc8180x-smmu-500", .data = &qcom_smmu_500_impl0_data }, { .compatible = "qcom,sc8280xp-smmu-500", .data = &qcom_smmu_500_impl0_data }, @@ -561,5 +562,14 @@ struct arm_smmu_device *qcom_smmu_impl_init(struct arm_smmu_device *smmu) if (match) return qcom_smmu_create(smmu, match->data); + /* + * If you hit this WARN_ON() you are missing an entry in the + * qcom_smmu_impl_of_match[] table, and GPU per-process page- + * tables will be broken. + */ + WARN(of_device_is_compatible(np, "qcom,adreno-smmu"), + "Missing qcom_smmu_impl_of_match entry for: %s", + dev_name(smmu->dev)); + return smmu; } diff --git a/drivers/iommu/mtk_iommu.c b/drivers/iommu/mtk_iommu.c index aecc7d154f28eeb1cfb493b9e61686f1fba2f4bf..e93906d6e112e826facaea508e118e36b78fb666 100644 --- a/drivers/iommu/mtk_iommu.c +++ b/drivers/iommu/mtk_iommu.c @@ -781,7 +781,8 @@ static void mtk_iommu_flush_iotlb_all(struct iommu_domain *domain) { struct mtk_iommu_domain *dom = to_mtk_domain(domain); - mtk_iommu_tlb_flush_all(dom->bank->parent_data); + if (dom->bank) + mtk_iommu_tlb_flush_all(dom->bank->parent_data); } static void mtk_iommu_iotlb_sync(struct iommu_domain *domain, diff --git a/drivers/iommu/rockchip-iommu.c b/drivers/iommu/rockchip-iommu.c index ea5a3088bb7e8a566ab9285d644009eb2bcd1d6a..4054030c32379533eca2e96003016a8a6a51481a 100644 --- a/drivers/iommu/rockchip-iommu.c +++ b/drivers/iommu/rockchip-iommu.c @@ -1335,20 +1335,22 @@ static int rk_iommu_probe(struct platform_device *pdev) for (i = 0; i < iommu->num_irq; i++) { int irq = platform_get_irq(pdev, i); - if (irq < 0) - return irq; + if (irq < 0) { + err = irq; + goto err_pm_disable; + } err = devm_request_irq(iommu->dev, irq, rk_iommu_irq, IRQF_SHARED, dev_name(dev), iommu); - if (err) { - pm_runtime_disable(dev); - goto err_remove_sysfs; - } + if (err) + goto err_pm_disable; } dma_set_mask_and_coherent(dev, rk_ops->dma_bit_mask); return 0; +err_pm_disable: + pm_runtime_disable(dev); err_remove_sysfs: iommu_device_sysfs_remove(&iommu->iommu); err_put_group: diff --git a/drivers/irqchip/irq-gic-common.c b/drivers/irqchip/irq-gic-common.c index a610821c8ff2af1e8485046480b047c6d3538a27..afd6a1841715a152afd572dac15e4366bfd96d1f 100644 --- a/drivers/irqchip/irq-gic-common.c +++ b/drivers/irqchip/irq-gic-common.c @@ -16,7 +16,13 @@ void gic_enable_of_quirks(const struct device_node *np, const struct gic_quirk *quirks, void *data) { for (; quirks->desc; quirks++) { - if (!of_device_is_compatible(np, quirks->compatible)) + if (!quirks->compatible && !quirks->property) + continue; + if (quirks->compatible && + !of_device_is_compatible(np, quirks->compatible)) + continue; + if (quirks->property && + !of_property_read_bool(np, quirks->property)) continue; if (quirks->init(data)) pr_info("GIC: enabling workaround for %s\n", @@ -28,7 +34,7 @@ void gic_enable_quirks(u32 iidr, const struct gic_quirk *quirks, void *data) { for (; quirks->desc; quirks++) { - if (quirks->compatible) + if (quirks->compatible || quirks->property) continue; if (quirks->iidr != (quirks->mask & iidr)) continue; diff --git a/drivers/irqchip/irq-gic-common.h b/drivers/irqchip/irq-gic-common.h index 27e3d4ed4f32861a4c46eebcc4634eb835a8e45b..3db4592cda1c01ad8a27e73f62b9d0890019667d 100644 --- a/drivers/irqchip/irq-gic-common.h +++ b/drivers/irqchip/irq-gic-common.h @@ -13,6 +13,7 @@ struct gic_quirk { const char *desc; const char *compatible; + const char *property; bool (*init)(void *data); u32 iidr; u32 mask; diff --git a/drivers/irqchip/irq-gic-v3.c b/drivers/irqchip/irq-gic-v3.c index 6fcee221f201720929f1481aa64d3367d6395737..a605aa79435a4dca64dd1eb012f0f1fe7f394ff3 100644 --- a/drivers/irqchip/irq-gic-v3.c +++ b/drivers/irqchip/irq-gic-v3.c @@ -39,6 +39,7 @@ #define FLAGS_WORKAROUND_GICR_WAKER_MSM8996 (1ULL << 0) #define FLAGS_WORKAROUND_CAVIUM_ERRATUM_38539 (1ULL << 1) +#define FLAGS_WORKAROUND_MTK_GICR_SAVE (1ULL << 2) #define GIC_IRQ_TYPE_PARTITION (GIC_IRQ_TYPE_LPI + 1) @@ -1720,6 +1721,15 @@ static bool gic_enable_quirk_msm8996(void *data) return true; } +static bool gic_enable_quirk_mtk_gicr(void *data) +{ + struct gic_chip_data *d = data; + + d->flags |= FLAGS_WORKAROUND_MTK_GICR_SAVE; + + return true; +} + static bool gic_enable_quirk_cavium_38539(void *data) { struct gic_chip_data *d = data; @@ -1792,6 +1802,11 @@ static const struct gic_quirk gic_quirks[] = { .compatible = "qcom,msm8996-gic-v3", .init = gic_enable_quirk_msm8996, }, + { + .desc = "GICv3: Mediatek Chromebook GICR save problem", + .property = "mediatek,broken-save-restore-fw", + .init = gic_enable_quirk_mtk_gicr, + }, { .desc = "GICv3: HIP06 erratum 161010803", .iidr = 0x0204043b, @@ -1834,6 +1849,11 @@ static void gic_enable_nmi_support(void) if (!gic_prio_masking_enabled()) return; + if (gic_data.flags & FLAGS_WORKAROUND_MTK_GICR_SAVE) { + pr_warn("Skipping NMI enable due to firmware issues\n"); + return; + } + ppi_nmi_refs = kcalloc(gic_data.ppi_nr, sizeof(*ppi_nmi_refs), GFP_KERNEL); if (!ppi_nmi_refs) return; diff --git a/drivers/irqchip/irq-mbigen.c b/drivers/irqchip/irq-mbigen.c index eada5e0e3eb956c939be7d6eef06e1a0a4aedd39..5101a3fb11df5bef53122db9db3c194669d754e7 100644 --- a/drivers/irqchip/irq-mbigen.c +++ b/drivers/irqchip/irq-mbigen.c @@ -240,26 +240,27 @@ static int mbigen_of_create_domain(struct platform_device *pdev, struct irq_domain *domain; struct device_node *np; u32 num_pins; + int ret = 0; + + parent = bus_get_dev_root(&platform_bus_type); + if (!parent) + return -ENODEV; for_each_child_of_node(pdev->dev.of_node, np) { if (!of_property_read_bool(np, "interrupt-controller")) continue; - parent = bus_get_dev_root(&platform_bus_type); - if (parent) { - child = of_platform_device_create(np, NULL, parent); - put_device(parent); - if (!child) { - of_node_put(np); - return -ENOMEM; - } + child = of_platform_device_create(np, NULL, parent); + if (!child) { + ret = -ENOMEM; + break; } if (of_property_read_u32(child->dev.of_node, "num-pins", &num_pins) < 0) { dev_err(&pdev->dev, "No num-pins property\n"); - of_node_put(np); - return -EINVAL; + ret = -EINVAL; + break; } domain = platform_msi_create_device_domain(&child->dev, num_pins, @@ -267,12 +268,16 @@ static int mbigen_of_create_domain(struct platform_device *pdev, &mbigen_domain_ops, mgn_chip); if (!domain) { - of_node_put(np); - return -ENOMEM; + ret = -ENOMEM; + break; } } - return 0; + put_device(parent); + if (ret) + of_node_put(np); + + return ret; } #ifdef CONFIG_ACPI diff --git a/drivers/irqchip/irq-meson-gpio.c b/drivers/irqchip/irq-meson-gpio.c index 2aaa9aad3e87acf666e7cf1fbe97feb9a175d797..7da18ef95211984bd2998af59e3bac07b5bedf98 100644 --- a/drivers/irqchip/irq-meson-gpio.c +++ b/drivers/irqchip/irq-meson-gpio.c @@ -150,7 +150,7 @@ static const struct meson_gpio_irq_params s4_params = { INIT_MESON_S4_COMMON_DATA(82) }; -static const struct of_device_id meson_irq_gpio_matches[] = { +static const struct of_device_id meson_irq_gpio_matches[] __maybe_unused = { { .compatible = "amlogic,meson8-gpio-intc", .data = &meson8_params }, { .compatible = "amlogic,meson8b-gpio-intc", .data = &meson8b_params }, { .compatible = "amlogic,meson-gxbb-gpio-intc", .data = &gxbb_params }, diff --git a/drivers/irqchip/irq-mips-gic.c b/drivers/irqchip/irq-mips-gic.c index 046c355e120b6485abf5266490670af9ce21be21..6d5ecc10a87032dfccb0835964999bfb07d912bd 100644 --- a/drivers/irqchip/irq-mips-gic.c +++ b/drivers/irqchip/irq-mips-gic.c @@ -50,7 +50,7 @@ void __iomem *mips_gic_base; static DEFINE_PER_CPU_READ_MOSTLY(unsigned long[GIC_MAX_LONGS], pcpu_masks); -static DEFINE_SPINLOCK(gic_lock); +static DEFINE_RAW_SPINLOCK(gic_lock); static struct irq_domain *gic_irq_domain; static int gic_shared_intrs; static unsigned int gic_cpu_pin; @@ -210,7 +210,7 @@ static int gic_set_type(struct irq_data *d, unsigned int type) irq = GIC_HWIRQ_TO_SHARED(d->hwirq); - spin_lock_irqsave(&gic_lock, flags); + raw_spin_lock_irqsave(&gic_lock, flags); switch (type & IRQ_TYPE_SENSE_MASK) { case IRQ_TYPE_EDGE_FALLING: pol = GIC_POL_FALLING_EDGE; @@ -250,7 +250,7 @@ static int gic_set_type(struct irq_data *d, unsigned int type) else irq_set_chip_handler_name_locked(d, &gic_level_irq_controller, handle_level_irq, NULL); - spin_unlock_irqrestore(&gic_lock, flags); + raw_spin_unlock_irqrestore(&gic_lock, flags); return 0; } @@ -268,7 +268,7 @@ static int gic_set_affinity(struct irq_data *d, const struct cpumask *cpumask, return -EINVAL; /* Assumption : cpumask refers to a single CPU */ - spin_lock_irqsave(&gic_lock, flags); + raw_spin_lock_irqsave(&gic_lock, flags); /* Re-route this IRQ */ write_gic_map_vp(irq, BIT(mips_cm_vp_id(cpu))); @@ -279,7 +279,7 @@ static int gic_set_affinity(struct irq_data *d, const struct cpumask *cpumask, set_bit(irq, per_cpu_ptr(pcpu_masks, cpu)); irq_data_update_effective_affinity(d, cpumask_of(cpu)); - spin_unlock_irqrestore(&gic_lock, flags); + raw_spin_unlock_irqrestore(&gic_lock, flags); return IRQ_SET_MASK_OK; } @@ -357,12 +357,12 @@ static void gic_mask_local_irq_all_vpes(struct irq_data *d) cd = irq_data_get_irq_chip_data(d); cd->mask = false; - spin_lock_irqsave(&gic_lock, flags); + raw_spin_lock_irqsave(&gic_lock, flags); for_each_online_cpu(cpu) { write_gic_vl_other(mips_cm_vp_id(cpu)); write_gic_vo_rmask(BIT(intr)); } - spin_unlock_irqrestore(&gic_lock, flags); + raw_spin_unlock_irqrestore(&gic_lock, flags); } static void gic_unmask_local_irq_all_vpes(struct irq_data *d) @@ -375,12 +375,12 @@ static void gic_unmask_local_irq_all_vpes(struct irq_data *d) cd = irq_data_get_irq_chip_data(d); cd->mask = true; - spin_lock_irqsave(&gic_lock, flags); + raw_spin_lock_irqsave(&gic_lock, flags); for_each_online_cpu(cpu) { write_gic_vl_other(mips_cm_vp_id(cpu)); write_gic_vo_smask(BIT(intr)); } - spin_unlock_irqrestore(&gic_lock, flags); + raw_spin_unlock_irqrestore(&gic_lock, flags); } static void gic_all_vpes_irq_cpu_online(void) @@ -393,19 +393,21 @@ static void gic_all_vpes_irq_cpu_online(void) unsigned long flags; int i; - spin_lock_irqsave(&gic_lock, flags); + raw_spin_lock_irqsave(&gic_lock, flags); for (i = 0; i < ARRAY_SIZE(local_intrs); i++) { unsigned int intr = local_intrs[i]; struct gic_all_vpes_chip_data *cd; + if (!gic_local_irq_is_routable(intr)) + continue; cd = &gic_all_vpes_chip_data[intr]; write_gic_vl_map(mips_gic_vx_map_reg(intr), cd->map); if (cd->mask) write_gic_vl_smask(BIT(intr)); } - spin_unlock_irqrestore(&gic_lock, flags); + raw_spin_unlock_irqrestore(&gic_lock, flags); } static struct irq_chip gic_all_vpes_local_irq_controller = { @@ -435,11 +437,11 @@ static int gic_shared_irq_domain_map(struct irq_domain *d, unsigned int virq, data = irq_get_irq_data(virq); - spin_lock_irqsave(&gic_lock, flags); + raw_spin_lock_irqsave(&gic_lock, flags); write_gic_map_pin(intr, GIC_MAP_PIN_MAP_TO_PIN | gic_cpu_pin); write_gic_map_vp(intr, BIT(mips_cm_vp_id(cpu))); irq_data_update_effective_affinity(data, cpumask_of(cpu)); - spin_unlock_irqrestore(&gic_lock, flags); + raw_spin_unlock_irqrestore(&gic_lock, flags); return 0; } @@ -531,12 +533,12 @@ static int gic_irq_domain_map(struct irq_domain *d, unsigned int virq, if (!gic_local_irq_is_routable(intr)) return -EPERM; - spin_lock_irqsave(&gic_lock, flags); + raw_spin_lock_irqsave(&gic_lock, flags); for_each_online_cpu(cpu) { write_gic_vl_other(mips_cm_vp_id(cpu)); write_gic_vo_map(mips_gic_vx_map_reg(intr), map); } - spin_unlock_irqrestore(&gic_lock, flags); + raw_spin_unlock_irqrestore(&gic_lock, flags); return 0; } diff --git a/drivers/leds/rgb/leds-qcom-lpg.c b/drivers/leds/rgb/leds-qcom-lpg.c index 55a037234df1b07c44fab18846bfa17e705c1af1..1c849814a4917fe5805a3f1aa8bf87831366a759 100644 --- a/drivers/leds/rgb/leds-qcom-lpg.c +++ b/drivers/leds/rgb/leds-qcom-lpg.c @@ -312,14 +312,14 @@ static int lpg_calc_freq(struct lpg_channel *chan, uint64_t period) max_res = LPG_RESOLUTION_9BIT; } - min_period = (u64)NSEC_PER_SEC * - div64_u64((1 << pwm_resolution_arr[0]), clk_rate_arr[clk_len - 1]); + min_period = div64_u64((u64)NSEC_PER_SEC * (1 << pwm_resolution_arr[0]), + clk_rate_arr[clk_len - 1]); if (period <= min_period) return -EINVAL; /* Limit period to largest possible value, to avoid overflows */ - max_period = (u64)NSEC_PER_SEC * max_res * LPG_MAX_PREDIV * - div64_u64((1 << LPG_MAX_M), 1024); + max_period = div64_u64((u64)NSEC_PER_SEC * max_res * LPG_MAX_PREDIV * (1 << LPG_MAX_M), + 1024); if (period > max_period) period = max_period; diff --git a/drivers/mailbox/mailbox-test.c b/drivers/mailbox/mailbox-test.c index c4a705c30331421e08e31606a12285d56a6d8682..fc6a12a51b403863d7dcb36a63a54b5a76c2f135 100644 --- a/drivers/mailbox/mailbox-test.c +++ b/drivers/mailbox/mailbox-test.c @@ -98,6 +98,7 @@ static ssize_t mbox_test_message_write(struct file *filp, size_t count, loff_t *ppos) { struct mbox_test_device *tdev = filp->private_data; + char *message; void *data; int ret; @@ -113,12 +114,13 @@ static ssize_t mbox_test_message_write(struct file *filp, return -EINVAL; } - mutex_lock(&tdev->mutex); - - tdev->message = kzalloc(MBOX_MAX_MSG_LEN, GFP_KERNEL); - if (!tdev->message) + message = kzalloc(MBOX_MAX_MSG_LEN, GFP_KERNEL); + if (!message) return -ENOMEM; + mutex_lock(&tdev->mutex); + + tdev->message = message; ret = copy_from_user(tdev->message, userbuf, count); if (ret) { ret = -EFAULT; diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index cc77cf3d410921432eb0c62cdede7d55b9aa674a..7d5c9c582ed2d639f244ad998836c1e49fbbaad5 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -1168,13 +1168,10 @@ static int do_resume(struct dm_ioctl *param) /* Do we need to load a new map ? */ if (new_map) { sector_t old_size, new_size; - int srcu_idx; /* Suspend if it isn't already suspended */ - old_map = dm_get_live_table(md, &srcu_idx); - if ((param->flags & DM_SKIP_LOCKFS_FLAG) || !old_map) + if (param->flags & DM_SKIP_LOCKFS_FLAG) suspend_flags &= ~DM_SUSPEND_LOCKFS_FLAG; - dm_put_live_table(md, srcu_idx); if (param->flags & DM_NOFLUSH_FLAG) suspend_flags |= DM_SUSPEND_NOFLUSH_FLAG; if (!dm_suspended_md(md)) diff --git a/drivers/md/dm-thin-metadata.c b/drivers/md/dm-thin-metadata.c index 9f5cb52c57632cac44c9f182cef25dac606bcc97..b9461faa9f0df1dccef30684c37dff23577dc61f 100644 --- a/drivers/md/dm-thin-metadata.c +++ b/drivers/md/dm-thin-metadata.c @@ -1756,13 +1756,15 @@ int dm_thin_remove_range(struct dm_thin_device *td, int dm_pool_block_is_shared(struct dm_pool_metadata *pmd, dm_block_t b, bool *result) { - int r; + int r = -EINVAL; uint32_t ref_count; down_read(&pmd->root_lock); - r = dm_sm_get_count(pmd->data_sm, b, &ref_count); - if (!r) - *result = (ref_count > 1); + if (!pmd->fail_io) { + r = dm_sm_get_count(pmd->data_sm, b, &ref_count); + if (!r) + *result = (ref_count > 1); + } up_read(&pmd->root_lock); return r; @@ -1770,10 +1772,11 @@ int dm_pool_block_is_shared(struct dm_pool_metadata *pmd, dm_block_t b, bool *re int dm_pool_inc_data_range(struct dm_pool_metadata *pmd, dm_block_t b, dm_block_t e) { - int r = 0; + int r = -EINVAL; pmd_write_lock(pmd); - r = dm_sm_inc_blocks(pmd->data_sm, b, e); + if (!pmd->fail_io) + r = dm_sm_inc_blocks(pmd->data_sm, b, e); pmd_write_unlock(pmd); return r; @@ -1781,10 +1784,11 @@ int dm_pool_inc_data_range(struct dm_pool_metadata *pmd, dm_block_t b, dm_block_ int dm_pool_dec_data_range(struct dm_pool_metadata *pmd, dm_block_t b, dm_block_t e) { - int r = 0; + int r = -EINVAL; pmd_write_lock(pmd); - r = dm_sm_dec_blocks(pmd->data_sm, b, e); + if (!pmd->fail_io) + r = dm_sm_dec_blocks(pmd->data_sm, b, e); pmd_write_unlock(pmd); return r; diff --git a/drivers/md/dm-thin.c b/drivers/md/dm-thin.c index 2b13c949bd726bf13cc82c08e9d10676d32e4f47..39410bf186cf6ecbf4056814b271f4b3ce80330b 100644 --- a/drivers/md/dm-thin.c +++ b/drivers/md/dm-thin.c @@ -401,8 +401,7 @@ static int issue_discard(struct discard_op *op, dm_block_t data_b, dm_block_t da sector_t s = block_to_sectors(tc->pool, data_b); sector_t len = block_to_sectors(tc->pool, data_e - data_b); - return __blkdev_issue_discard(tc->pool_dev->bdev, s, len, GFP_NOWAIT, - &op->bio); + return __blkdev_issue_discard(tc->pool_dev->bdev, s, len, GFP_NOIO, &op->bio); } static void end_discard(struct discard_op *op, int r) diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 3b694ba3a106e68d4c0d5e64cd9136cf7abce237..fffb0cbe2ac8a14fe9bc0c9dc442ce6e4a0a02aa 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -1172,7 +1172,8 @@ static inline sector_t max_io_len_target_boundary(struct dm_target *ti, } static sector_t __max_io_len(struct dm_target *ti, sector_t sector, - unsigned int max_granularity) + unsigned int max_granularity, + unsigned int max_sectors) { sector_t target_offset = dm_target_offset(ti, sector); sector_t len = max_io_len_target_boundary(ti, target_offset); @@ -1186,13 +1187,13 @@ static sector_t __max_io_len(struct dm_target *ti, sector_t sector, if (!max_granularity) return len; return min_t(sector_t, len, - min(queue_max_sectors(ti->table->md->queue), + min(max_sectors ? : queue_max_sectors(ti->table->md->queue), blk_chunk_sectors_left(target_offset, max_granularity))); } static inline sector_t max_io_len(struct dm_target *ti, sector_t sector) { - return __max_io_len(ti, sector, ti->max_io_len); + return __max_io_len(ti, sector, ti->max_io_len, 0); } int dm_set_target_max_io_len(struct dm_target *ti, sector_t len) @@ -1581,12 +1582,13 @@ static void __send_empty_flush(struct clone_info *ci) static void __send_changing_extent_only(struct clone_info *ci, struct dm_target *ti, unsigned int num_bios, - unsigned int max_granularity) + unsigned int max_granularity, + unsigned int max_sectors) { unsigned int len, bios; len = min_t(sector_t, ci->sector_count, - __max_io_len(ti, ci->sector, max_granularity)); + __max_io_len(ti, ci->sector, max_granularity, max_sectors)); atomic_add(num_bios, &ci->io->io_count); bios = __send_duplicate_bios(ci, ti, num_bios, &len); @@ -1623,23 +1625,27 @@ static blk_status_t __process_abnormal_io(struct clone_info *ci, { unsigned int num_bios = 0; unsigned int max_granularity = 0; + unsigned int max_sectors = 0; struct queue_limits *limits = dm_get_queue_limits(ti->table->md); switch (bio_op(ci->bio)) { case REQ_OP_DISCARD: num_bios = ti->num_discard_bios; + max_sectors = limits->max_discard_sectors; if (ti->max_discard_granularity) - max_granularity = limits->max_discard_sectors; + max_granularity = max_sectors; break; case REQ_OP_SECURE_ERASE: num_bios = ti->num_secure_erase_bios; + max_sectors = limits->max_secure_erase_sectors; if (ti->max_secure_erase_granularity) - max_granularity = limits->max_secure_erase_sectors; + max_granularity = max_sectors; break; case REQ_OP_WRITE_ZEROES: num_bios = ti->num_write_zeroes_bios; + max_sectors = limits->max_write_zeroes_sectors; if (ti->max_write_zeroes_granularity) - max_granularity = limits->max_write_zeroes_sectors; + max_granularity = max_sectors; break; default: break; @@ -1654,7 +1660,8 @@ static blk_status_t __process_abnormal_io(struct clone_info *ci, if (unlikely(!num_bios)) return BLK_STS_NOTSUPP; - __send_changing_extent_only(ci, ti, num_bios, max_granularity); + __send_changing_extent_only(ci, ti, num_bios, + max_granularity, max_sectors); return BLK_STS_OK; } @@ -2808,6 +2815,10 @@ int dm_suspend(struct mapped_device *md, unsigned int suspend_flags) } map = rcu_dereference_protected(md->map, lockdep_is_held(&md->suspend_lock)); + if (!map) { + /* avoid deadlock with fs/namespace.c:do_mount() */ + suspend_flags &= ~DM_SUSPEND_LOCKFS_FLAG; + } r = __dm_suspend(md, map, suspend_flags, TASK_INTERRUPTIBLE, DMF_SUSPENDED); if (r) diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 4739ed891e756eb3a8af3e7c48c6be0950b82db7..9ea285fbc4a654703f363f7d450c903d5e8c6d77 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -5516,7 +5516,7 @@ static int raid5_read_one_chunk(struct mddev *mddev, struct bio *raid_bio) sector = raid5_compute_sector(conf, raid_bio->bi_iter.bi_sector, 0, &dd_idx, NULL); - end_sector = bio_end_sector(raid_bio); + end_sector = sector + bio_sectors(raid_bio); rcu_read_lock(); if (r5c_big_stripe_cached(conf, sector)) diff --git a/drivers/media/cec/core/cec-adap.c b/drivers/media/cec/core/cec-adap.c index 769ea6b2e1d0b6776f16e78a0ccb71d4c03e8668..241b1621b197c4f64677a6a16108f72b7322db11 100644 --- a/drivers/media/cec/core/cec-adap.c +++ b/drivers/media/cec/core/cec-adap.c @@ -1091,7 +1091,8 @@ void cec_received_msg_ts(struct cec_adapter *adap, mutex_lock(&adap->lock); dprintk(2, "%s: %*ph\n", __func__, msg->len, msg->msg); - adap->last_initiator = 0xff; + if (!adap->transmit_in_progress) + adap->last_initiator = 0xff; /* Check if this message was for us (directed or broadcast). */ if (!cec_msg_is_broadcast(msg)) { @@ -1585,7 +1586,7 @@ static void cec_claim_log_addrs(struct cec_adapter *adap, bool block) * * This function is called with adap->lock held. */ -static int cec_adap_enable(struct cec_adapter *adap) +int cec_adap_enable(struct cec_adapter *adap) { bool enable; int ret = 0; @@ -1595,6 +1596,9 @@ static int cec_adap_enable(struct cec_adapter *adap) if (adap->needs_hpd) enable = enable && adap->phys_addr != CEC_PHYS_ADDR_INVALID; + if (adap->devnode.unregistered) + enable = false; + if (enable == adap->is_enabled) return 0; diff --git a/drivers/media/cec/core/cec-core.c b/drivers/media/cec/core/cec-core.c index af358e901b5f35c6cf05f7a7bc5fd954ae3b5854..7e153c5cad04f50d58824da520cc019409bd4a87 100644 --- a/drivers/media/cec/core/cec-core.c +++ b/drivers/media/cec/core/cec-core.c @@ -191,6 +191,8 @@ static void cec_devnode_unregister(struct cec_adapter *adap) mutex_lock(&adap->lock); __cec_s_phys_addr(adap, CEC_PHYS_ADDR_INVALID, false); __cec_s_log_addrs(adap, NULL, false); + // Disable the adapter (since adap->devnode.unregistered is true) + cec_adap_enable(adap); mutex_unlock(&adap->lock); cdev_device_del(&devnode->cdev, &devnode->dev); diff --git a/drivers/media/cec/core/cec-priv.h b/drivers/media/cec/core/cec-priv.h index b78df931aa74b04a2699bbb11f0c140cccf7de46..ed1f8c67626bf9cee44f4596307b9c95aa62e4e5 100644 --- a/drivers/media/cec/core/cec-priv.h +++ b/drivers/media/cec/core/cec-priv.h @@ -47,6 +47,7 @@ int cec_monitor_pin_cnt_inc(struct cec_adapter *adap); void cec_monitor_pin_cnt_dec(struct cec_adapter *adap); int cec_adap_status(struct seq_file *file, void *priv); int cec_thread_func(void *_adap); +int cec_adap_enable(struct cec_adapter *adap); void __cec_s_phys_addr(struct cec_adapter *adap, u16 phys_addr, bool block); int __cec_s_log_addrs(struct cec_adapter *adap, struct cec_log_addrs *log_addrs, bool block); diff --git a/drivers/media/dvb-core/dvb_ca_en50221.c b/drivers/media/dvb-core/dvb_ca_en50221.c index c2d2792227f865ff934f0effa3eda98ecef7802d..baf64540dc00a127c5f781034b1fec192e7e4ca2 100644 --- a/drivers/media/dvb-core/dvb_ca_en50221.c +++ b/drivers/media/dvb-core/dvb_ca_en50221.c @@ -151,6 +151,12 @@ struct dvb_ca_private { /* mutex serializing ioctls */ struct mutex ioctl_mutex; + + /* A mutex used when a device is disconnected */ + struct mutex remove_mutex; + + /* Whether the device is disconnected */ + int exit; }; static void dvb_ca_private_free(struct dvb_ca_private *ca) @@ -187,7 +193,7 @@ static void dvb_ca_en50221_thread_wakeup(struct dvb_ca_private *ca); static int dvb_ca_en50221_read_data(struct dvb_ca_private *ca, int slot, u8 *ebuf, int ecount); static int dvb_ca_en50221_write_data(struct dvb_ca_private *ca, int slot, - u8 *ebuf, int ecount); + u8 *ebuf, int ecount, int size_write_flag); /** * findstr - Safely find needle in haystack. @@ -370,7 +376,7 @@ static int dvb_ca_en50221_link_init(struct dvb_ca_private *ca, int slot) ret = dvb_ca_en50221_wait_if_status(ca, slot, STATUSREG_FR, HZ / 10); if (ret) return ret; - ret = dvb_ca_en50221_write_data(ca, slot, buf, 2); + ret = dvb_ca_en50221_write_data(ca, slot, buf, 2, CMDREG_SW); if (ret != 2) return -EIO; ret = ca->pub->write_cam_control(ca->pub, slot, CTRLIF_COMMAND, IRQEN); @@ -778,11 +784,13 @@ static int dvb_ca_en50221_read_data(struct dvb_ca_private *ca, int slot, * @buf: The data in this buffer is treated as a complete link-level packet to * be written. * @bytes_write: Size of ebuf. + * @size_write_flag: A flag on Command Register which says whether the link size + * information will be writen or not. * * return: Number of bytes written, or < 0 on error. */ static int dvb_ca_en50221_write_data(struct dvb_ca_private *ca, int slot, - u8 *buf, int bytes_write) + u8 *buf, int bytes_write, int size_write_flag) { struct dvb_ca_slot *sl = &ca->slot_info[slot]; int status; @@ -817,7 +825,7 @@ static int dvb_ca_en50221_write_data(struct dvb_ca_private *ca, int slot, /* OK, set HC bit */ status = ca->pub->write_cam_control(ca->pub, slot, CTRLIF_COMMAND, - IRQEN | CMDREG_HC); + IRQEN | CMDREG_HC | size_write_flag); if (status) goto exit; @@ -1508,7 +1516,7 @@ static ssize_t dvb_ca_en50221_io_write(struct file *file, mutex_lock(&sl->slot_lock); status = dvb_ca_en50221_write_data(ca, slot, fragbuf, - fraglen + 2); + fraglen + 2, 0); mutex_unlock(&sl->slot_lock); if (status == (fraglen + 2)) { written = 1; @@ -1709,12 +1717,22 @@ static int dvb_ca_en50221_io_open(struct inode *inode, struct file *file) dprintk("%s\n", __func__); - if (!try_module_get(ca->pub->owner)) + mutex_lock(&ca->remove_mutex); + + if (ca->exit) { + mutex_unlock(&ca->remove_mutex); + return -ENODEV; + } + + if (!try_module_get(ca->pub->owner)) { + mutex_unlock(&ca->remove_mutex); return -EIO; + } err = dvb_generic_open(inode, file); if (err < 0) { module_put(ca->pub->owner); + mutex_unlock(&ca->remove_mutex); return err; } @@ -1739,6 +1757,7 @@ static int dvb_ca_en50221_io_open(struct inode *inode, struct file *file) dvb_ca_private_get(ca); + mutex_unlock(&ca->remove_mutex); return 0; } @@ -1758,6 +1777,8 @@ static int dvb_ca_en50221_io_release(struct inode *inode, struct file *file) dprintk("%s\n", __func__); + mutex_lock(&ca->remove_mutex); + /* mark the CA device as closed */ ca->open = 0; dvb_ca_en50221_thread_update_delay(ca); @@ -1768,6 +1789,13 @@ static int dvb_ca_en50221_io_release(struct inode *inode, struct file *file) dvb_ca_private_put(ca); + if (dvbdev->users == 1 && ca->exit == 1) { + mutex_unlock(&ca->remove_mutex); + wake_up(&dvbdev->wait_queue); + } else { + mutex_unlock(&ca->remove_mutex); + } + return err; } @@ -1891,6 +1919,7 @@ int dvb_ca_en50221_init(struct dvb_adapter *dvb_adapter, } mutex_init(&ca->ioctl_mutex); + mutex_init(&ca->remove_mutex); if (signal_pending(current)) { ret = -EINTR; @@ -1933,6 +1962,14 @@ void dvb_ca_en50221_release(struct dvb_ca_en50221 *pubca) dprintk("%s\n", __func__); + mutex_lock(&ca->remove_mutex); + ca->exit = 1; + mutex_unlock(&ca->remove_mutex); + + if (ca->dvbdev->users < 1) + wait_event(ca->dvbdev->wait_queue, + ca->dvbdev->users == 1); + /* shutdown the thread if there was one */ kthread_stop(ca->thread); diff --git a/drivers/media/dvb-core/dvb_demux.c b/drivers/media/dvb-core/dvb_demux.c index 398c86279b5b09555590c5c3ac4d3a8798c9c103..7c4d86bfdd6c9d259ab99b57cd3ce355f3b341ad 100644 --- a/drivers/media/dvb-core/dvb_demux.c +++ b/drivers/media/dvb-core/dvb_demux.c @@ -115,12 +115,12 @@ static inline int dvb_dmx_swfilter_payload(struct dvb_demux_feed *feed, cc = buf[3] & 0x0f; ccok = ((feed->cc + 1) & 0x0f) == cc; - feed->cc = cc; if (!ccok) { set_buf_flags(feed, DMX_BUFFER_FLAG_DISCONTINUITY_DETECTED); dprintk_sect_loss("missed packet: %d instead of %d!\n", cc, (feed->cc + 1) & 0x0f); } + feed->cc = cc; if (buf[1] & 0x40) // PUSI ? feed->peslen = 0xfffa; @@ -300,7 +300,6 @@ static int dvb_dmx_swfilter_section_packet(struct dvb_demux_feed *feed, cc = buf[3] & 0x0f; ccok = ((feed->cc + 1) & 0x0f) == cc; - feed->cc = cc; if (buf[3] & 0x20) { /* adaption field present, check for discontinuity_indicator */ @@ -336,6 +335,7 @@ static int dvb_dmx_swfilter_section_packet(struct dvb_demux_feed *feed, feed->pusi_seen = false; dvb_dmx_swfilter_section_new(feed); } + feed->cc = cc; if (buf[1] & 0x40) { /* PUSI=1 (is set), section boundary is here */ diff --git a/drivers/media/dvb-core/dvb_frontend.c b/drivers/media/dvb-core/dvb_frontend.c index cc0a789f09ae5360c43aaaa8d989704790ff6d17..9293b058ab9974885da8938095597dabd96e603c 100644 --- a/drivers/media/dvb-core/dvb_frontend.c +++ b/drivers/media/dvb-core/dvb_frontend.c @@ -293,14 +293,22 @@ static int dvb_frontend_get_event(struct dvb_frontend *fe, } if (events->eventw == events->eventr) { - int ret; + struct wait_queue_entry wait; + int ret = 0; if (flags & O_NONBLOCK) return -EWOULDBLOCK; - ret = wait_event_interruptible(events->wait_queue, - dvb_frontend_test_event(fepriv, events)); - + init_waitqueue_entry(&wait, current); + add_wait_queue(&events->wait_queue, &wait); + while (!dvb_frontend_test_event(fepriv, events)) { + wait_woken(&wait, TASK_INTERRUPTIBLE, 0); + if (signal_pending(current)) { + ret = -ERESTARTSYS; + break; + } + } + remove_wait_queue(&events->wait_queue, &wait); if (ret < 0) return ret; } diff --git a/drivers/media/dvb-core/dvb_net.c b/drivers/media/dvb-core/dvb_net.c index 8a2febf33ce28fde43676e66f7381e59d032f4d1..8bb8dd34c223e91119059965851de57cac9af8e5 100644 --- a/drivers/media/dvb-core/dvb_net.c +++ b/drivers/media/dvb-core/dvb_net.c @@ -1564,15 +1564,43 @@ static long dvb_net_ioctl(struct file *file, return dvb_usercopy(file, cmd, arg, dvb_net_do_ioctl); } +static int locked_dvb_net_open(struct inode *inode, struct file *file) +{ + struct dvb_device *dvbdev = file->private_data; + struct dvb_net *dvbnet = dvbdev->priv; + int ret; + + if (mutex_lock_interruptible(&dvbnet->remove_mutex)) + return -ERESTARTSYS; + + if (dvbnet->exit) { + mutex_unlock(&dvbnet->remove_mutex); + return -ENODEV; + } + + ret = dvb_generic_open(inode, file); + + mutex_unlock(&dvbnet->remove_mutex); + + return ret; +} + static int dvb_net_close(struct inode *inode, struct file *file) { struct dvb_device *dvbdev = file->private_data; struct dvb_net *dvbnet = dvbdev->priv; + mutex_lock(&dvbnet->remove_mutex); + dvb_generic_release(inode, file); - if(dvbdev->users == 1 && dvbnet->exit == 1) + if (dvbdev->users == 1 && dvbnet->exit == 1) { + mutex_unlock(&dvbnet->remove_mutex); wake_up(&dvbdev->wait_queue); + } else { + mutex_unlock(&dvbnet->remove_mutex); + } + return 0; } @@ -1580,7 +1608,7 @@ static int dvb_net_close(struct inode *inode, struct file *file) static const struct file_operations dvb_net_fops = { .owner = THIS_MODULE, .unlocked_ioctl = dvb_net_ioctl, - .open = dvb_generic_open, + .open = locked_dvb_net_open, .release = dvb_net_close, .llseek = noop_llseek, }; @@ -1599,10 +1627,13 @@ void dvb_net_release (struct dvb_net *dvbnet) { int i; + mutex_lock(&dvbnet->remove_mutex); dvbnet->exit = 1; + mutex_unlock(&dvbnet->remove_mutex); + if (dvbnet->dvbdev->users < 1) wait_event(dvbnet->dvbdev->wait_queue, - dvbnet->dvbdev->users==1); + dvbnet->dvbdev->users == 1); dvb_unregister_device(dvbnet->dvbdev); @@ -1621,6 +1652,7 @@ int dvb_net_init (struct dvb_adapter *adap, struct dvb_net *dvbnet, int i; mutex_init(&dvbnet->ioctl_mutex); + mutex_init(&dvbnet->remove_mutex); dvbnet->demux = dmx; for (i=0; i static DEFINE_MUTEX(dvbdev_mutex); +static LIST_HEAD(dvbdevfops_list); static int dvbdev_debug; module_param(dvbdev_debug, int, 0644); @@ -453,14 +454,15 @@ int dvb_register_device(struct dvb_adapter *adap, struct dvb_device **pdvbdev, enum dvb_device_type type, int demux_sink_pads) { struct dvb_device *dvbdev; - struct file_operations *dvbdevfops; + struct file_operations *dvbdevfops = NULL; + struct dvbdevfops_node *node = NULL, *new_node = NULL; struct device *clsdev; int minor; int id, ret; mutex_lock(&dvbdev_register_lock); - if ((id = dvbdev_get_free_id (adap, type)) < 0){ + if ((id = dvbdev_get_free_id (adap, type)) < 0) { mutex_unlock(&dvbdev_register_lock); *pdvbdev = NULL; pr_err("%s: couldn't find free device id\n", __func__); @@ -468,18 +470,45 @@ int dvb_register_device(struct dvb_adapter *adap, struct dvb_device **pdvbdev, } *pdvbdev = dvbdev = kzalloc(sizeof(*dvbdev), GFP_KERNEL); - if (!dvbdev){ mutex_unlock(&dvbdev_register_lock); return -ENOMEM; } - dvbdevfops = kmemdup(template->fops, sizeof(*dvbdevfops), GFP_KERNEL); + /* + * When a device of the same type is probe()d more than once, + * the first allocated fops are used. This prevents memory leaks + * that can occur when the same device is probe()d repeatedly. + */ + list_for_each_entry(node, &dvbdevfops_list, list_head) { + if (node->fops->owner == adap->module && + node->type == type && + node->template == template) { + dvbdevfops = node->fops; + break; + } + } - if (!dvbdevfops){ - kfree (dvbdev); - mutex_unlock(&dvbdev_register_lock); - return -ENOMEM; + if (dvbdevfops == NULL) { + dvbdevfops = kmemdup(template->fops, sizeof(*dvbdevfops), GFP_KERNEL); + if (!dvbdevfops) { + kfree(dvbdev); + mutex_unlock(&dvbdev_register_lock); + return -ENOMEM; + } + + new_node = kzalloc(sizeof(struct dvbdevfops_node), GFP_KERNEL); + if (!new_node) { + kfree(dvbdevfops); + kfree(dvbdev); + mutex_unlock(&dvbdev_register_lock); + return -ENOMEM; + } + + new_node->fops = dvbdevfops; + new_node->type = type; + new_node->template = template; + list_add_tail (&new_node->list_head, &dvbdevfops_list); } memcpy(dvbdev, template, sizeof(struct dvb_device)); @@ -490,20 +519,20 @@ int dvb_register_device(struct dvb_adapter *adap, struct dvb_device **pdvbdev, dvbdev->priv = priv; dvbdev->fops = dvbdevfops; init_waitqueue_head (&dvbdev->wait_queue); - dvbdevfops->owner = adap->module; - list_add_tail (&dvbdev->list_head, &adap->device_list); - down_write(&minor_rwsem); #ifdef CONFIG_DVB_DYNAMIC_MINORS for (minor = 0; minor < MAX_DVB_MINORS; minor++) if (dvb_minors[minor] == NULL) break; - if (minor == MAX_DVB_MINORS) { + if (new_node) { + list_del (&new_node->list_head); + kfree(dvbdevfops); + kfree(new_node); + } list_del (&dvbdev->list_head); - kfree(dvbdevfops); kfree(dvbdev); up_write(&minor_rwsem); mutex_unlock(&dvbdev_register_lock); @@ -512,41 +541,47 @@ int dvb_register_device(struct dvb_adapter *adap, struct dvb_device **pdvbdev, #else minor = nums2minor(adap->num, type, id); #endif - dvbdev->minor = minor; dvb_minors[minor] = dvb_device_get(dvbdev); up_write(&minor_rwsem); - ret = dvb_register_media_device(dvbdev, type, minor, demux_sink_pads); if (ret) { pr_err("%s: dvb_register_media_device failed to create the mediagraph\n", __func__); - + if (new_node) { + list_del (&new_node->list_head); + kfree(dvbdevfops); + kfree(new_node); + } dvb_media_device_free(dvbdev); list_del (&dvbdev->list_head); - kfree(dvbdevfops); kfree(dvbdev); mutex_unlock(&dvbdev_register_lock); return ret; } - mutex_unlock(&dvbdev_register_lock); - clsdev = device_create(dvb_class, adap->device, MKDEV(DVB_MAJOR, minor), dvbdev, "dvb%d.%s%d", adap->num, dnames[type], id); if (IS_ERR(clsdev)) { pr_err("%s: failed to create device dvb%d.%s%d (%ld)\n", __func__, adap->num, dnames[type], id, PTR_ERR(clsdev)); + if (new_node) { + list_del (&new_node->list_head); + kfree(dvbdevfops); + kfree(new_node); + } dvb_media_device_free(dvbdev); list_del (&dvbdev->list_head); - kfree(dvbdevfops); kfree(dvbdev); + mutex_unlock(&dvbdev_register_lock); return PTR_ERR(clsdev); } + dprintk("DVB: register adapter%d/%s%d @ minor: %i (0x%02x)\n", adap->num, dnames[type], id, minor, minor); + mutex_unlock(&dvbdev_register_lock); return 0; } EXPORT_SYMBOL(dvb_register_device); @@ -575,7 +610,6 @@ static void dvb_free_device(struct kref *ref) { struct dvb_device *dvbdev = container_of(ref, struct dvb_device, ref); - kfree (dvbdev->fops); kfree (dvbdev); } @@ -1081,9 +1115,17 @@ static int __init init_dvbdev(void) static void __exit exit_dvbdev(void) { + struct dvbdevfops_node *node, *next; + class_destroy(dvb_class); cdev_del(&dvb_device_cdev); unregister_chrdev_region(MKDEV(DVB_MAJOR, 0), MAX_DVB_MINORS); + + list_for_each_entry_safe(node, next, &dvbdevfops_list, list_head) { + list_del (&node->list_head); + kfree(node->fops); + kfree(node); + } } subsys_initcall(init_dvbdev); diff --git a/drivers/media/dvb-frontends/mn88443x.c b/drivers/media/dvb-frontends/mn88443x.c index 1f1753f2ab1a3ed5f1b52a87d9848e132cb94645..0782f8377eb2f333731fe93de70f0804d95ffcc5 100644 --- a/drivers/media/dvb-frontends/mn88443x.c +++ b/drivers/media/dvb-frontends/mn88443x.c @@ -798,7 +798,7 @@ MODULE_DEVICE_TABLE(i2c, mn88443x_i2c_id); static struct i2c_driver mn88443x_driver = { .driver = { .name = "mn88443x", - .of_match_table = of_match_ptr(mn88443x_of_match), + .of_match_table = mn88443x_of_match, }, .probe_new = mn88443x_probe, .remove = mn88443x_remove, diff --git a/drivers/media/pci/netup_unidvb/netup_unidvb_core.c b/drivers/media/pci/netup_unidvb/netup_unidvb_core.c index 8287851b5ffdc8d591bc2f1354eaa136168e057d..d85bfbb77a2509eaf5da9b4716d86eb3d05e3bc1 100644 --- a/drivers/media/pci/netup_unidvb/netup_unidvb_core.c +++ b/drivers/media/pci/netup_unidvb/netup_unidvb_core.c @@ -697,7 +697,7 @@ static void netup_unidvb_dma_fini(struct netup_unidvb_dev *ndev, int num) netup_unidvb_dma_enable(dma, 0); msleep(50); cancel_work_sync(&dma->work); - del_timer(&dma->timeout); + del_timer_sync(&dma->timeout); } static int netup_unidvb_dma_setup(struct netup_unidvb_dev *ndev) @@ -887,12 +887,7 @@ static int netup_unidvb_initdev(struct pci_dev *pci_dev, ndev->lmmio0, (u32)pci_resource_len(pci_dev, 0), ndev->lmmio1, (u32)pci_resource_len(pci_dev, 1), pci_dev->irq); - if (request_irq(pci_dev->irq, netup_unidvb_isr, IRQF_SHARED, - "netup_unidvb", pci_dev) < 0) { - dev_err(&pci_dev->dev, - "%s(): can't get IRQ %d\n", __func__, pci_dev->irq); - goto irq_request_err; - } + ndev->dma_size = 2 * 188 * NETUP_DMA_BLOCKS_COUNT * NETUP_DMA_PACKETS_COUNT; ndev->dma_virt = dma_alloc_coherent(&pci_dev->dev, @@ -933,6 +928,14 @@ static int netup_unidvb_initdev(struct pci_dev *pci_dev, dev_err(&pci_dev->dev, "netup_unidvb: DMA setup failed\n"); goto dma_setup_err; } + + if (request_irq(pci_dev->irq, netup_unidvb_isr, IRQF_SHARED, + "netup_unidvb", pci_dev) < 0) { + dev_err(&pci_dev->dev, + "%s(): can't get IRQ %d\n", __func__, pci_dev->irq); + goto dma_setup_err; + } + dev_info(&pci_dev->dev, "netup_unidvb: device has been initialized\n"); return 0; @@ -951,8 +954,6 @@ static int netup_unidvb_initdev(struct pci_dev *pci_dev, dma_free_coherent(&pci_dev->dev, ndev->dma_size, ndev->dma_virt, ndev->dma_phys); dma_alloc_err: - free_irq(pci_dev->irq, pci_dev); -irq_request_err: iounmap(ndev->lmmio1); pci_bar1_error: iounmap(ndev->lmmio0); diff --git a/drivers/media/platform/mediatek/vcodec/mtk_vcodec_dec_stateful.c b/drivers/media/platform/mediatek/vcodec/mtk_vcodec_dec_stateful.c index 29991551cf6140671e486490ff648871b0860250..0fbd030026c729620314e7f15355bb9c24f2797a 100644 --- a/drivers/media/platform/mediatek/vcodec/mtk_vcodec_dec_stateful.c +++ b/drivers/media/platform/mediatek/vcodec/mtk_vcodec_dec_stateful.c @@ -584,6 +584,9 @@ static void mtk_init_vdec_params(struct mtk_vcodec_ctx *ctx) if (!(ctx->dev->dec_capability & VCODEC_CAPABILITY_4K_DISABLED)) { for (i = 0; i < num_supported_formats; i++) { + if (mtk_video_formats[i].type != MTK_FMT_DEC) + continue; + mtk_video_formats[i].frmsize.max_width = VCODEC_DEC_4K_CODED_WIDTH; mtk_video_formats[i].frmsize.max_height = diff --git a/drivers/media/platform/qcom/camss/camss-video.c b/drivers/media/platform/qcom/camss/camss-video.c index 898f32177b12a5cc2b6aa6a65b9c7f1b8719bd45..8640db306026804270e82cb4642f9592f51b5c0b 100644 --- a/drivers/media/platform/qcom/camss/camss-video.c +++ b/drivers/media/platform/qcom/camss/camss-video.c @@ -353,7 +353,6 @@ static int video_get_subdev_format(struct camss_video *video, if (subdev == NULL) return -EPIPE; - memset(&fmt, 0, sizeof(fmt)); fmt.pad = pad; ret = v4l2_subdev_call(subdev, pad, get_fmt, NULL, &fmt); diff --git a/drivers/media/platform/verisilicon/hantro_v4l2.c b/drivers/media/platform/verisilicon/hantro_v4l2.c index 835518534e3b94dc6f352cf3a8fe0330db4d79c5..61cfaaf4e927b35a81e55ccae6e2d4b8b2f13dd5 100644 --- a/drivers/media/platform/verisilicon/hantro_v4l2.c +++ b/drivers/media/platform/verisilicon/hantro_v4l2.c @@ -397,10 +397,12 @@ hantro_reset_raw_fmt(struct hantro_ctx *ctx, int bit_depth) if (!raw_vpu_fmt) return -EINVAL; - if (ctx->is_encoder) + if (ctx->is_encoder) { encoded_fmt = &ctx->dst_fmt; - else + ctx->vpu_src_fmt = raw_vpu_fmt; + } else { encoded_fmt = &ctx->src_fmt; + } hantro_reset_fmt(&raw_fmt, raw_vpu_fmt); raw_fmt.width = encoded_fmt->width; diff --git a/drivers/media/usb/dvb-usb-v2/ce6230.c b/drivers/media/usb/dvb-usb-v2/ce6230.c index 44540de1a206686fcc366a2a8701bf15fe5b221b..d3b5cb4a24daf9c47a4903a3618aee977abeafd3 100644 --- a/drivers/media/usb/dvb-usb-v2/ce6230.c +++ b/drivers/media/usb/dvb-usb-v2/ce6230.c @@ -101,6 +101,10 @@ static int ce6230_i2c_master_xfer(struct i2c_adapter *adap, if (num > i + 1 && (msg[i+1].flags & I2C_M_RD)) { if (msg[i].addr == ce6230_zl10353_config.demod_address) { + if (msg[i].len < 1) { + i = -EOPNOTSUPP; + break; + } req.cmd = DEMOD_READ; req.value = msg[i].addr >> 1; req.index = msg[i].buf[0]; @@ -117,6 +121,10 @@ static int ce6230_i2c_master_xfer(struct i2c_adapter *adap, } else { if (msg[i].addr == ce6230_zl10353_config.demod_address) { + if (msg[i].len < 1) { + i = -EOPNOTSUPP; + break; + } req.cmd = DEMOD_WRITE; req.value = msg[i].addr >> 1; req.index = msg[i].buf[0]; diff --git a/drivers/media/usb/dvb-usb-v2/ec168.c b/drivers/media/usb/dvb-usb-v2/ec168.c index 7ed0ab9e429b1457abc5e91b9f4795e96cdda631..0e4773fc025c9e74f8c86a451e9df60b513ccc8f 100644 --- a/drivers/media/usb/dvb-usb-v2/ec168.c +++ b/drivers/media/usb/dvb-usb-v2/ec168.c @@ -115,6 +115,10 @@ static int ec168_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msg[], while (i < num) { if (num > i + 1 && (msg[i+1].flags & I2C_M_RD)) { if (msg[i].addr == ec168_ec100_config.demod_address) { + if (msg[i].len < 1) { + i = -EOPNOTSUPP; + break; + } req.cmd = READ_DEMOD; req.value = 0; req.index = 0xff00 + msg[i].buf[0]; /* reg */ @@ -131,6 +135,10 @@ static int ec168_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msg[], } } else { if (msg[i].addr == ec168_ec100_config.demod_address) { + if (msg[i].len < 1) { + i = -EOPNOTSUPP; + break; + } req.cmd = WRITE_DEMOD; req.value = msg[i].buf[1]; /* val */ req.index = 0xff00 + msg[i].buf[0]; /* reg */ @@ -139,6 +147,10 @@ static int ec168_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msg[], ret = ec168_ctrl_msg(d, &req); i += 1; } else { + if (msg[i].len < 1) { + i = -EOPNOTSUPP; + break; + } req.cmd = WRITE_I2C; req.value = msg[i].buf[0]; /* val */ req.index = 0x0100 + msg[i].addr; /* I2C addr */ diff --git a/drivers/media/usb/dvb-usb-v2/rtl28xxu.c b/drivers/media/usb/dvb-usb-v2/rtl28xxu.c index 795a012d402002de74559c0aab4e0d80dc826e3e..f7884bb56fccff29205c95f20b00fd098cec3ee3 100644 --- a/drivers/media/usb/dvb-usb-v2/rtl28xxu.c +++ b/drivers/media/usb/dvb-usb-v2/rtl28xxu.c @@ -176,6 +176,10 @@ static int rtl28xxu_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msg[], ret = -EOPNOTSUPP; goto err_mutex_unlock; } else if (msg[0].addr == 0x10) { + if (msg[0].len < 1 || msg[1].len < 1) { + ret = -EOPNOTSUPP; + goto err_mutex_unlock; + } /* method 1 - integrated demod */ if (msg[0].buf[0] == 0x00) { /* return demod page from driver cache */ @@ -189,6 +193,10 @@ static int rtl28xxu_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msg[], ret = rtl28xxu_ctrl_msg(d, &req); } } else if (msg[0].len < 2) { + if (msg[0].len < 1) { + ret = -EOPNOTSUPP; + goto err_mutex_unlock; + } /* method 2 - old I2C */ req.value = (msg[0].buf[0] << 8) | (msg[0].addr << 1); req.index = CMD_I2C_RD; @@ -217,8 +225,16 @@ static int rtl28xxu_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msg[], ret = -EOPNOTSUPP; goto err_mutex_unlock; } else if (msg[0].addr == 0x10) { + if (msg[0].len < 1) { + ret = -EOPNOTSUPP; + goto err_mutex_unlock; + } /* method 1 - integrated demod */ if (msg[0].buf[0] == 0x00) { + if (msg[0].len < 2) { + ret = -EOPNOTSUPP; + goto err_mutex_unlock; + } /* save demod page for later demod access */ dev->page = msg[0].buf[1]; ret = 0; @@ -231,6 +247,10 @@ static int rtl28xxu_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msg[], ret = rtl28xxu_ctrl_msg(d, &req); } } else if ((msg[0].len < 23) && (!dev->new_i2c_write)) { + if (msg[0].len < 1) { + ret = -EOPNOTSUPP; + goto err_mutex_unlock; + } /* method 2 - old I2C */ req.value = (msg[0].buf[0] << 8) | (msg[0].addr << 1); req.index = CMD_I2C_WR; diff --git a/drivers/media/usb/dvb-usb/az6027.c b/drivers/media/usb/dvb-usb/az6027.c index 7d78ee09be5e1dc40d813777d5e26e2abeb8aba3..a31c6f82f4e90fbcd08bc600ada4948180a9970f 100644 --- a/drivers/media/usb/dvb-usb/az6027.c +++ b/drivers/media/usb/dvb-usb/az6027.c @@ -988,6 +988,10 @@ static int az6027_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msg[], int n /* write/read request */ if (i + 1 < num && (msg[i + 1].flags & I2C_M_RD)) { req = 0xB9; + if (msg[i].len < 1) { + i = -EOPNOTSUPP; + break; + } index = (((msg[i].buf[0] << 8) & 0xff00) | (msg[i].buf[1] & 0x00ff)); value = msg[i].addr + (msg[i].len << 8); length = msg[i + 1].len + 6; @@ -1001,6 +1005,10 @@ static int az6027_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msg[], int n /* demod 16bit addr */ req = 0xBD; + if (msg[i].len < 1) { + i = -EOPNOTSUPP; + break; + } index = (((msg[i].buf[0] << 8) & 0xff00) | (msg[i].buf[1] & 0x00ff)); value = msg[i].addr + (2 << 8); length = msg[i].len - 2; @@ -1026,6 +1034,10 @@ static int az6027_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msg[], int n } else { req = 0xBD; + if (msg[i].len < 1) { + i = -EOPNOTSUPP; + break; + } index = msg[i].buf[0] & 0x00FF; value = msg[i].addr + (1 << 8); length = msg[i].len - 1; diff --git a/drivers/media/usb/dvb-usb/digitv.c b/drivers/media/usb/dvb-usb/digitv.c index 2756815a780bca7e2c2fc07f549d0c3ca99651a4..32134be1691489fa3bed90cc97f68f8eea8179a0 100644 --- a/drivers/media/usb/dvb-usb/digitv.c +++ b/drivers/media/usb/dvb-usb/digitv.c @@ -63,6 +63,10 @@ static int digitv_i2c_xfer(struct i2c_adapter *adap,struct i2c_msg msg[],int num warn("more than 2 i2c messages at a time is not handled yet. TODO."); for (i = 0; i < num; i++) { + if (msg[i].len < 1) { + i = -EOPNOTSUPP; + break; + } /* write/read request */ if (i+1 < num && (msg[i+1].flags & I2C_M_RD)) { if (digitv_ctrl_msg(d, USB_READ_COFDM, msg[i].buf[0], NULL, 0, diff --git a/drivers/media/usb/dvb-usb/dw2102.c b/drivers/media/usb/dvb-usb/dw2102.c index 0ca764282c767eb1b12aa7cb515453068fb2d4a7..8747960e614611a71f2d5ec78ecac77f633207f9 100644 --- a/drivers/media/usb/dvb-usb/dw2102.c +++ b/drivers/media/usb/dvb-usb/dw2102.c @@ -946,7 +946,7 @@ static int su3000_read_mac_address(struct dvb_usb_device *d, u8 mac[6]) for (i = 0; i < 6; i++) { obuf[1] = 0xf0 + i; if (i2c_transfer(&d->i2c_adap, msg, 2) != 2) - break; + return -1; else mac[i] = ibuf[0]; } diff --git a/drivers/media/usb/pvrusb2/Kconfig b/drivers/media/usb/pvrusb2/Kconfig index 9501b10b31aa57d1683f7ace1a0c05ebc7c604e8..0df10270dbdfcf47a1e428d095182e42cfc686ee 100644 --- a/drivers/media/usb/pvrusb2/Kconfig +++ b/drivers/media/usb/pvrusb2/Kconfig @@ -37,6 +37,7 @@ config VIDEO_PVRUSB2_DVB bool "pvrusb2 ATSC/DVB support" default y depends on VIDEO_PVRUSB2 && DVB_CORE + depends on VIDEO_PVRUSB2=m || DVB_CORE=y select DVB_LGDT330X if MEDIA_SUBDRV_AUTOSELECT select DVB_S5H1409 if MEDIA_SUBDRV_AUTOSELECT select DVB_S5H1411 if MEDIA_SUBDRV_AUTOSELECT diff --git a/drivers/media/usb/ttusb-dec/ttusb_dec.c b/drivers/media/usb/ttusb-dec/ttusb_dec.c index 38822cedd93a97c23486de57fa8931f9723f92fa..c4474d4c44e28adc84a26265ac45dd619bef3df2 100644 --- a/drivers/media/usb/ttusb-dec/ttusb_dec.c +++ b/drivers/media/usb/ttusb-dec/ttusb_dec.c @@ -1544,8 +1544,7 @@ static void ttusb_dec_exit_dvb(struct ttusb_dec *dec) dvb_dmx_release(&dec->demux); if (dec->fe) { dvb_unregister_frontend(dec->fe); - if (dec->fe->ops.release) - dec->fe->ops.release(dec->fe); + dvb_frontend_detach(dec->fe); } dvb_unregister_adapter(&dec->adapter); } diff --git a/drivers/media/usb/uvc/uvc_driver.c b/drivers/media/usb/uvc/uvc_driver.c index 7aefa76a42b316f30cb6643d833a92b94547e9bc..d631ce4f9f7bb298d2b4c309d0bf242cdbd61a93 100644 --- a/drivers/media/usb/uvc/uvc_driver.c +++ b/drivers/media/usb/uvc/uvc_driver.c @@ -251,14 +251,17 @@ static int uvc_parse_format(struct uvc_device *dev, /* Find the format descriptor from its GUID. */ fmtdesc = uvc_format_by_guid(&buffer[5]); - if (fmtdesc != NULL) { - format->fcc = fmtdesc->fcc; - } else { + if (!fmtdesc) { + /* + * Unknown video formats are not fatal errors, the + * caller will skip this descriptor. + */ dev_info(&streaming->intf->dev, "Unknown video format %pUl\n", &buffer[5]); - format->fcc = 0; + return 0; } + format->fcc = fmtdesc->fcc; format->bpp = buffer[21]; /* @@ -675,7 +678,7 @@ static int uvc_parse_streaming(struct uvc_device *dev, interval = (u32 *)&frame[nframes]; streaming->format = format; - streaming->nformats = nformats; + streaming->nformats = 0; /* Parse the format descriptors. */ while (buflen > 2 && buffer[1] == USB_DT_CS_INTERFACE) { @@ -689,7 +692,10 @@ static int uvc_parse_streaming(struct uvc_device *dev, &interval, buffer, buflen); if (ret < 0) goto error; + if (!ret) + break; + streaming->nformats++; frame += format->nframes; format++; diff --git a/drivers/media/v4l2-core/v4l2-mc.c b/drivers/media/v4l2-core/v4l2-mc.c index bf0c18100664f56366a188ee22a1f354184d3b39..22fe08fce0a9abe8a3fb359a014b2471acab2ac3 100644 --- a/drivers/media/v4l2-core/v4l2-mc.c +++ b/drivers/media/v4l2-core/v4l2-mc.c @@ -314,8 +314,7 @@ int v4l2_create_fwnode_links_to_pad(struct v4l2_subdev *src_sd, { struct fwnode_handle *endpoint; - if (!(sink->flags & MEDIA_PAD_FL_SINK) || - !is_media_entity_v4l2_subdev(sink->entity)) + if (!(sink->flags & MEDIA_PAD_FL_SINK)) return -EINVAL; fwnode_graph_for_each_endpoint(dev_fwnode(src_sd->dev), endpoint) { diff --git a/drivers/misc/eeprom/Kconfig b/drivers/misc/eeprom/Kconfig index f0a7531f354c1215df6c25b55590cb5e925a5290..2d240bfa819f8bd8bdd6f854609a91adb737f5b3 100644 --- a/drivers/misc/eeprom/Kconfig +++ b/drivers/misc/eeprom/Kconfig @@ -6,6 +6,7 @@ config EEPROM_AT24 depends on I2C && SYSFS select NVMEM select NVMEM_SYSFS + select REGMAP select REGMAP_I2C help Enable this driver to get read/write support to most I2C EEPROMs diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c index f48466960f1b9c0f9b78ff4c6d0038c7719e124f..30d4d0476248ffdfeec0f5c2c2c7a71840e26d20 100644 --- a/drivers/misc/fastrpc.c +++ b/drivers/misc/fastrpc.c @@ -316,12 +316,14 @@ static void fastrpc_free_map(struct kref *ref) if (map->table) { if (map->attr & FASTRPC_ATTR_SECUREMAP) { struct qcom_scm_vmperm perm; + int vmid = map->fl->cctx->vmperms[0].vmid; + u64 src_perms = BIT(QCOM_SCM_VMID_HLOS) | BIT(vmid); int err = 0; perm.vmid = QCOM_SCM_VMID_HLOS; perm.perm = QCOM_SCM_PERM_RWX; err = qcom_scm_assign_mem(map->phys, map->size, - &map->fl->cctx->perms, &perm, 1); + &src_perms, &perm, 1); if (err) { dev_err(map->fl->sctx->dev, "Failed to assign memory phys 0x%llx size 0x%llx err %d", map->phys, map->size, err); @@ -787,8 +789,12 @@ static int fastrpc_map_create(struct fastrpc_user *fl, int fd, goto map_err; } - map->phys = sg_dma_address(map->table->sgl); - map->phys += ((u64)fl->sctx->sid << 32); + if (attr & FASTRPC_ATTR_SECUREMAP) { + map->phys = sg_phys(map->table->sgl); + } else { + map->phys = sg_dma_address(map->table->sgl); + map->phys += ((u64)fl->sctx->sid << 32); + } map->size = len; map->va = sg_virt(map->table->sgl); map->len = len; @@ -798,9 +804,15 @@ static int fastrpc_map_create(struct fastrpc_user *fl, int fd, * If subsystem VMIDs are defined in DTSI, then do * hyp_assign from HLOS to those VM(s) */ + u64 src_perms = BIT(QCOM_SCM_VMID_HLOS); + struct qcom_scm_vmperm dst_perms[2] = {0}; + + dst_perms[0].vmid = QCOM_SCM_VMID_HLOS; + dst_perms[0].perm = QCOM_SCM_PERM_RW; + dst_perms[1].vmid = fl->cctx->vmperms[0].vmid; + dst_perms[1].perm = QCOM_SCM_PERM_RWX; map->attr = attr; - err = qcom_scm_assign_mem(map->phys, (u64)map->size, &fl->cctx->perms, - fl->cctx->vmperms, fl->cctx->vmcount); + err = qcom_scm_assign_mem(map->phys, (u64)map->size, &src_perms, dst_perms, 2); if (err) { dev_err(sess->dev, "Failed to assign memory with phys 0x%llx size 0x%llx err %d", map->phys, map->size, err); @@ -1892,7 +1904,7 @@ static int fastrpc_req_mmap(struct fastrpc_user *fl, char __user *argp) req.vaddrout = rsp_msg.vaddr; /* Add memory to static PD pool, protection thru hypervisor */ - if (req.flags != ADSP_MMAP_REMOTE_HEAP_ADDR && fl->cctx->vmcount) { + if (req.flags == ADSP_MMAP_REMOTE_HEAP_ADDR && fl->cctx->vmcount) { struct qcom_scm_vmperm perm; perm.vmid = QCOM_SCM_VMID_HLOS; @@ -2337,8 +2349,10 @@ static void fastrpc_notify_users(struct fastrpc_user *user) struct fastrpc_invoke_ctx *ctx; spin_lock(&user->lock); - list_for_each_entry(ctx, &user->pending, node) + list_for_each_entry(ctx, &user->pending, node) { + ctx->retval = -EPIPE; complete(&ctx->work); + } spin_unlock(&user->lock); } @@ -2349,7 +2363,9 @@ static void fastrpc_rpmsg_remove(struct rpmsg_device *rpdev) struct fastrpc_user *user; unsigned long flags; + /* No invocations past this point */ spin_lock_irqsave(&cctx->lock, flags); + cctx->rpdev = NULL; list_for_each_entry(user, &cctx->users, user) fastrpc_notify_users(user); spin_unlock_irqrestore(&cctx->lock, flags); @@ -2368,7 +2384,6 @@ static void fastrpc_rpmsg_remove(struct rpmsg_device *rpdev) of_platform_depopulate(&rpdev->dev); - cctx->rpdev = NULL; fastrpc_channel_ctx_put(cctx); } diff --git a/drivers/misc/mei/hdcp/mei_hdcp.c b/drivers/misc/mei/hdcp/mei_hdcp.c index 85cbfc3413ee2f040fcb3206d3c4aebbb7c805c2..51359cc5ece9a109d1d3dc96c32baf49e95500c4 100644 --- a/drivers/misc/mei/hdcp/mei_hdcp.c +++ b/drivers/misc/mei/hdcp/mei_hdcp.c @@ -735,13 +735,13 @@ static const struct i915_hdcp_ops mei_hdcp_ops = { static int mei_component_master_bind(struct device *dev) { struct mei_cl_device *cldev = to_mei_cl_device(dev); - struct i915_hdcp_master *comp_master = mei_cldev_get_drvdata(cldev); + struct i915_hdcp_arbiter *comp_arbiter = mei_cldev_get_drvdata(cldev); int ret; dev_dbg(dev, "%s\n", __func__); - comp_master->ops = &mei_hdcp_ops; - comp_master->hdcp_dev = dev; - ret = component_bind_all(dev, comp_master); + comp_arbiter->ops = &mei_hdcp_ops; + comp_arbiter->hdcp_dev = dev; + ret = component_bind_all(dev, comp_arbiter); if (ret < 0) return ret; @@ -751,10 +751,10 @@ static int mei_component_master_bind(struct device *dev) static void mei_component_master_unbind(struct device *dev) { struct mei_cl_device *cldev = to_mei_cl_device(dev); - struct i915_hdcp_master *comp_master = mei_cldev_get_drvdata(cldev); + struct i915_hdcp_arbiter *comp_arbiter = mei_cldev_get_drvdata(cldev); dev_dbg(dev, "%s\n", __func__); - component_unbind_all(dev, comp_master); + component_unbind_all(dev, comp_arbiter); } static const struct component_master_ops mei_component_master_ops = { @@ -799,7 +799,7 @@ static int mei_hdcp_component_match(struct device *dev, int subcomponent, static int mei_hdcp_probe(struct mei_cl_device *cldev, const struct mei_cl_device_id *id) { - struct i915_hdcp_master *comp_master; + struct i915_hdcp_arbiter *comp_arbiter; struct component_match *master_match; int ret; @@ -809,8 +809,8 @@ static int mei_hdcp_probe(struct mei_cl_device *cldev, goto enable_err_exit; } - comp_master = kzalloc(sizeof(*comp_master), GFP_KERNEL); - if (!comp_master) { + comp_arbiter = kzalloc(sizeof(*comp_arbiter), GFP_KERNEL); + if (!comp_arbiter) { ret = -ENOMEM; goto err_exit; } @@ -823,7 +823,7 @@ static int mei_hdcp_probe(struct mei_cl_device *cldev, goto err_exit; } - mei_cldev_set_drvdata(cldev, comp_master); + mei_cldev_set_drvdata(cldev, comp_arbiter); ret = component_master_add_with_match(&cldev->dev, &mei_component_master_ops, master_match); @@ -836,7 +836,7 @@ static int mei_hdcp_probe(struct mei_cl_device *cldev, err_exit: mei_cldev_set_drvdata(cldev, NULL); - kfree(comp_master); + kfree(comp_arbiter); mei_cldev_disable(cldev); enable_err_exit: return ret; @@ -844,11 +844,11 @@ static int mei_hdcp_probe(struct mei_cl_device *cldev, static void mei_hdcp_remove(struct mei_cl_device *cldev) { - struct i915_hdcp_master *comp_master = mei_cldev_get_drvdata(cldev); + struct i915_hdcp_arbiter *comp_arbiter = mei_cldev_get_drvdata(cldev); int ret; component_master_del(&cldev->dev, &mei_component_master_ops); - kfree(comp_master); + kfree(comp_arbiter); mei_cldev_set_drvdata(cldev, NULL); ret = mei_cldev_disable(cldev); diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c index 00c33edb9fb94ed4c6cd7c4f0830ea5d71d901e1..d920c417838930214b7d9c77fc4208d66c177e6e 100644 --- a/drivers/mmc/core/block.c +++ b/drivers/mmc/core/block.c @@ -264,6 +264,7 @@ static ssize_t power_ro_lock_store(struct device *dev, goto out_put; } req_to_mmc_queue_req(req)->drv_op = MMC_DRV_OP_BOOT_WP; + req_to_mmc_queue_req(req)->drv_op_result = -EIO; blk_execute_rq(req, false); ret = req_to_mmc_queue_req(req)->drv_op_result; blk_mq_free_request(req); @@ -651,6 +652,7 @@ static int mmc_blk_ioctl_cmd(struct mmc_blk_data *md, idatas[0] = idata; req_to_mmc_queue_req(req)->drv_op = rpmb ? MMC_DRV_OP_IOCTL_RPMB : MMC_DRV_OP_IOCTL; + req_to_mmc_queue_req(req)->drv_op_result = -EIO; req_to_mmc_queue_req(req)->drv_op_data = idatas; req_to_mmc_queue_req(req)->ioc_count = 1; blk_execute_rq(req, false); @@ -722,6 +724,7 @@ static int mmc_blk_ioctl_multi_cmd(struct mmc_blk_data *md, } req_to_mmc_queue_req(req)->drv_op = rpmb ? MMC_DRV_OP_IOCTL_RPMB : MMC_DRV_OP_IOCTL; + req_to_mmc_queue_req(req)->drv_op_result = -EIO; req_to_mmc_queue_req(req)->drv_op_data = idata; req_to_mmc_queue_req(req)->ioc_count = n; blk_execute_rq(req, false); @@ -2806,6 +2809,7 @@ static int mmc_dbg_card_status_get(void *data, u64 *val) if (IS_ERR(req)) return PTR_ERR(req); req_to_mmc_queue_req(req)->drv_op = MMC_DRV_OP_GET_CARD_STATUS; + req_to_mmc_queue_req(req)->drv_op_result = -EIO; blk_execute_rq(req, false); ret = req_to_mmc_queue_req(req)->drv_op_result; if (ret >= 0) { @@ -2844,6 +2848,7 @@ static int mmc_ext_csd_open(struct inode *inode, struct file *filp) goto out_free; } req_to_mmc_queue_req(req)->drv_op = MMC_DRV_OP_GET_EXT_CSD; + req_to_mmc_queue_req(req)->drv_op_result = -EIO; req_to_mmc_queue_req(req)->drv_op_data = &ext_csd; blk_execute_rq(req, false); err = req_to_mmc_queue_req(req)->drv_op_result; diff --git a/drivers/mmc/core/pwrseq_sd8787.c b/drivers/mmc/core/pwrseq_sd8787.c index 2e120ad83020f8f51d4bb529b1a470f9de65bf78..0c5f5e371e1f8311f3454e1fc0c462cde2841787 100644 --- a/drivers/mmc/core/pwrseq_sd8787.c +++ b/drivers/mmc/core/pwrseq_sd8787.c @@ -28,7 +28,6 @@ struct mmc_pwrseq_sd8787 { struct mmc_pwrseq pwrseq; struct gpio_desc *reset_gpio; struct gpio_desc *pwrdn_gpio; - u32 reset_pwrdwn_delay_ms; }; #define to_pwrseq_sd8787(p) container_of(p, struct mmc_pwrseq_sd8787, pwrseq) @@ -39,7 +38,7 @@ static void mmc_pwrseq_sd8787_pre_power_on(struct mmc_host *host) gpiod_set_value_cansleep(pwrseq->reset_gpio, 1); - msleep(pwrseq->reset_pwrdwn_delay_ms); + msleep(300); gpiod_set_value_cansleep(pwrseq->pwrdn_gpio, 1); } @@ -51,17 +50,37 @@ static void mmc_pwrseq_sd8787_power_off(struct mmc_host *host) gpiod_set_value_cansleep(pwrseq->reset_gpio, 0); } +static void mmc_pwrseq_wilc1000_pre_power_on(struct mmc_host *host) +{ + struct mmc_pwrseq_sd8787 *pwrseq = to_pwrseq_sd8787(host->pwrseq); + + /* The pwrdn_gpio is really CHIP_EN, reset_gpio is RESETN */ + gpiod_set_value_cansleep(pwrseq->pwrdn_gpio, 1); + msleep(5); + gpiod_set_value_cansleep(pwrseq->reset_gpio, 1); +} + +static void mmc_pwrseq_wilc1000_power_off(struct mmc_host *host) +{ + struct mmc_pwrseq_sd8787 *pwrseq = to_pwrseq_sd8787(host->pwrseq); + + gpiod_set_value_cansleep(pwrseq->reset_gpio, 0); + gpiod_set_value_cansleep(pwrseq->pwrdn_gpio, 0); +} + static const struct mmc_pwrseq_ops mmc_pwrseq_sd8787_ops = { .pre_power_on = mmc_pwrseq_sd8787_pre_power_on, .power_off = mmc_pwrseq_sd8787_power_off, }; -static const u32 sd8787_delay_ms = 300; -static const u32 wilc1000_delay_ms = 5; +static const struct mmc_pwrseq_ops mmc_pwrseq_wilc1000_ops = { + .pre_power_on = mmc_pwrseq_wilc1000_pre_power_on, + .power_off = mmc_pwrseq_wilc1000_power_off, +}; static const struct of_device_id mmc_pwrseq_sd8787_of_match[] = { - { .compatible = "mmc-pwrseq-sd8787", .data = &sd8787_delay_ms }, - { .compatible = "mmc-pwrseq-wilc1000", .data = &wilc1000_delay_ms }, + { .compatible = "mmc-pwrseq-sd8787", .data = &mmc_pwrseq_sd8787_ops }, + { .compatible = "mmc-pwrseq-wilc1000", .data = &mmc_pwrseq_wilc1000_ops }, {/* sentinel */}, }; MODULE_DEVICE_TABLE(of, mmc_pwrseq_sd8787_of_match); @@ -77,7 +96,6 @@ static int mmc_pwrseq_sd8787_probe(struct platform_device *pdev) return -ENOMEM; match = of_match_node(mmc_pwrseq_sd8787_of_match, pdev->dev.of_node); - pwrseq->reset_pwrdwn_delay_ms = *(u32 *)match->data; pwrseq->pwrdn_gpio = devm_gpiod_get(dev, "powerdown", GPIOD_OUT_LOW); if (IS_ERR(pwrseq->pwrdn_gpio)) @@ -88,7 +106,7 @@ static int mmc_pwrseq_sd8787_probe(struct platform_device *pdev) return PTR_ERR(pwrseq->reset_gpio); pwrseq->pwrseq.dev = dev; - pwrseq->pwrseq.ops = &mmc_pwrseq_sd8787_ops; + pwrseq->pwrseq.ops = match->data; pwrseq->pwrseq.owner = THIS_MODULE; platform_set_drvdata(pdev, pwrseq); diff --git a/drivers/mmc/host/sdhci-cadence.c b/drivers/mmc/host/sdhci-cadence.c index b24aa27da50c40dd861212d8b7e7cfc848e962cd..d2f62505468932b069e3411f2a4b7418ffece517 100644 --- a/drivers/mmc/host/sdhci-cadence.c +++ b/drivers/mmc/host/sdhci-cadence.c @@ -540,9 +540,11 @@ static int sdhci_cdns_probe(struct platform_device *pdev) if (host->mmc->caps & MMC_CAP_HW_RESET) { priv->rst_hw = devm_reset_control_get_optional_exclusive(dev, NULL); - if (IS_ERR(priv->rst_hw)) - return dev_err_probe(mmc_dev(host->mmc), PTR_ERR(priv->rst_hw), - "reset controller error\n"); + if (IS_ERR(priv->rst_hw)) { + ret = dev_err_probe(mmc_dev(host->mmc), PTR_ERR(priv->rst_hw), + "reset controller error\n"); + goto free; + } if (priv->rst_hw) host->mmc_host_ops.card_hw_reset = sdhci_cdns_mmc_hw_reset; } diff --git a/drivers/mmc/host/sdhci-esdhc-imx.c b/drivers/mmc/host/sdhci-esdhc-imx.c index d7c0c0b9e26c5e10563337cfbfca23d739956dd6..eebf94604a7fdd45dc578bb1fa7d4217ad996927 100644 --- a/drivers/mmc/host/sdhci-esdhc-imx.c +++ b/drivers/mmc/host/sdhci-esdhc-imx.c @@ -1634,6 +1634,10 @@ sdhci_esdhc_imx_probe_dt(struct platform_device *pdev, if (ret) return ret; + /* HS400/HS400ES require 8 bit bus */ + if (!(host->mmc->caps & MMC_CAP_8_BIT_DATA)) + host->mmc->caps2 &= ~(MMC_CAP2_HS400 | MMC_CAP2_HS400_ES); + if (mmc_gpio_get_cd(host->mmc) >= 0) host->quirks &= ~SDHCI_QUIRK_BROKEN_CARD_DETECTION; @@ -1724,10 +1728,6 @@ static int sdhci_esdhc_imx_probe(struct platform_device *pdev) host->mmc_host_ops.init_card = usdhc_init_card; } - err = sdhci_esdhc_imx_probe_dt(pdev, host, imx_data); - if (err) - goto disable_ahb_clk; - if (imx_data->socdata->flags & ESDHC_FLAG_MAN_TUNING) sdhci_esdhc_ops.platform_execute_tuning = esdhc_executing_tuning; @@ -1735,15 +1735,13 @@ static int sdhci_esdhc_imx_probe(struct platform_device *pdev) if (imx_data->socdata->flags & ESDHC_FLAG_ERR004536) host->quirks |= SDHCI_QUIRK_BROKEN_ADMA; - if (host->mmc->caps & MMC_CAP_8_BIT_DATA && - imx_data->socdata->flags & ESDHC_FLAG_HS400) + if (imx_data->socdata->flags & ESDHC_FLAG_HS400) host->mmc->caps2 |= MMC_CAP2_HS400; if (imx_data->socdata->flags & ESDHC_FLAG_BROKEN_AUTO_CMD23) host->quirks2 |= SDHCI_QUIRK2_ACMD23_BROKEN; - if (host->mmc->caps & MMC_CAP_8_BIT_DATA && - imx_data->socdata->flags & ESDHC_FLAG_HS400_ES) { + if (imx_data->socdata->flags & ESDHC_FLAG_HS400_ES) { host->mmc->caps2 |= MMC_CAP2_HS400_ES; host->mmc_host_ops.hs400_enhanced_strobe = esdhc_hs400_enhanced_strobe; @@ -1765,6 +1763,10 @@ static int sdhci_esdhc_imx_probe(struct platform_device *pdev) goto disable_ahb_clk; } + err = sdhci_esdhc_imx_probe_dt(pdev, host, imx_data); + if (err) + goto disable_ahb_clk; + sdhci_esdhc_imx_hwinit(host); err = sdhci_add_host(host); diff --git a/drivers/mmc/host/vub300.c b/drivers/mmc/host/vub300.c index e4c4bfac3763bae1ec808dcdb3d06bb9fb049769..9ec593d52f0fa930daee67dbf2bf6979695e7e5e 100644 --- a/drivers/mmc/host/vub300.c +++ b/drivers/mmc/host/vub300.c @@ -1713,6 +1713,9 @@ static void construct_request_response(struct vub300_mmc_host *vub300, int bytes = 3 & less_cmd; int words = less_cmd >> 2; u8 *r = vub300->resp.response.command_response; + + if (!resp_len) + return; if (bytes == 3) { cmd->resp[words] = (r[1 + (words << 2)] << 24) | (r[2 + (words << 2)] << 16) diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c index 01f1c6792df9c790622d2b039ee0fab992e6d0da..8dc4f5c493fcbac3abe17818cd5f7b2adcc70dd9 100644 --- a/drivers/mtd/mtdchar.c +++ b/drivers/mtd/mtdchar.c @@ -590,8 +590,8 @@ static void adjust_oob_length(struct mtd_info *mtd, uint64_t start, (end_page - start_page + 1) * oob_per_page); } -static int mtdchar_write_ioctl(struct mtd_info *mtd, - struct mtd_write_req __user *argp) +static noinline_for_stack int +mtdchar_write_ioctl(struct mtd_info *mtd, struct mtd_write_req __user *argp) { struct mtd_info *master = mtd_get_master(mtd); struct mtd_write_req req; @@ -688,8 +688,8 @@ static int mtdchar_write_ioctl(struct mtd_info *mtd, return ret; } -static int mtdchar_read_ioctl(struct mtd_info *mtd, - struct mtd_read_req __user *argp) +static noinline_for_stack int +mtdchar_read_ioctl(struct mtd_info *mtd, struct mtd_read_req __user *argp) { struct mtd_info *master = mtd_get_master(mtd); struct mtd_read_req req; diff --git a/drivers/mtd/nand/raw/ingenic/ingenic_ecc.h b/drivers/mtd/nand/raw/ingenic/ingenic_ecc.h index 2cda439b5e11bff0fc091a4ba31109b27cfdbb98..017868f59f2221611256b3f34edc09a196ea2581 100644 --- a/drivers/mtd/nand/raw/ingenic/ingenic_ecc.h +++ b/drivers/mtd/nand/raw/ingenic/ingenic_ecc.h @@ -36,25 +36,25 @@ int ingenic_ecc_correct(struct ingenic_ecc *ecc, void ingenic_ecc_release(struct ingenic_ecc *ecc); struct ingenic_ecc *of_ingenic_ecc_get(struct device_node *np); #else /* CONFIG_MTD_NAND_INGENIC_ECC */ -int ingenic_ecc_calculate(struct ingenic_ecc *ecc, +static inline int ingenic_ecc_calculate(struct ingenic_ecc *ecc, struct ingenic_ecc_params *params, const u8 *buf, u8 *ecc_code) { return -ENODEV; } -int ingenic_ecc_correct(struct ingenic_ecc *ecc, +static inline int ingenic_ecc_correct(struct ingenic_ecc *ecc, struct ingenic_ecc_params *params, u8 *buf, u8 *ecc_code) { return -ENODEV; } -void ingenic_ecc_release(struct ingenic_ecc *ecc) +static inline void ingenic_ecc_release(struct ingenic_ecc *ecc) { } -struct ingenic_ecc *of_ingenic_ecc_get(struct device_node *np) +static inline struct ingenic_ecc *of_ingenic_ecc_get(struct device_node *np) { return ERR_PTR(-ENODEV); } diff --git a/drivers/mtd/nand/raw/marvell_nand.c b/drivers/mtd/nand/raw/marvell_nand.c index afb424579f0b89777037726f54060fd21c2e0e73..30c15e4e1cc0d2899bf2fa943c1f7b95cc1d0e79 100644 --- a/drivers/mtd/nand/raw/marvell_nand.c +++ b/drivers/mtd/nand/raw/marvell_nand.c @@ -2457,6 +2457,12 @@ static int marvell_nfc_setup_interface(struct nand_chip *chip, int chipnr, NDTR1_WAIT_MODE; } + /* + * Reset nfc->selected_chip so the next command will cause the timing + * registers to be updated in marvell_nfc_select_target(). + */ + nfc->selected_chip = NULL; + return 0; } @@ -2894,10 +2900,6 @@ static int marvell_nfc_init(struct marvell_nfc *nfc) regmap_update_bits(sysctrl_base, GENCONF_CLK_GATING_CTRL, GENCONF_CLK_GATING_CTRL_ND_GATE, GENCONF_CLK_GATING_CTRL_ND_GATE); - - regmap_update_bits(sysctrl_base, GENCONF_ND_CLK_CTRL, - GENCONF_ND_CLK_CTRL_EN, - GENCONF_ND_CLK_CTRL_EN); } /* Configure the DMA if appropriate */ diff --git a/drivers/mtd/spi-nor/core.c b/drivers/mtd/spi-nor/core.c index 0bb0ad14a2fc01c97d8e20144f8844186aea17df..5f29fac8669a301e8010566eee02edfdcdb66620 100644 --- a/drivers/mtd/spi-nor/core.c +++ b/drivers/mtd/spi-nor/core.c @@ -2018,6 +2018,7 @@ static const struct spi_nor_manufacturer *manufacturers[] = { static const struct flash_info spi_nor_generic_flash = { .name = "spi-nor-generic", + .n_banks = 1, /* * JESD216 rev A doesn't specify the page size, therefore we need a * sane default. @@ -2921,7 +2922,8 @@ static void spi_nor_late_init_params(struct spi_nor *nor) if (nor->flags & SNOR_F_HAS_LOCK && !nor->params->locking_ops) spi_nor_init_default_locking_ops(nor); - nor->params->bank_size = div64_u64(nor->params->size, nor->info->n_banks); + if (nor->info->n_banks > 1) + params->bank_size = div64_u64(params->size, nor->info->n_banks); } /** @@ -2987,6 +2989,7 @@ static void spi_nor_init_default_params(struct spi_nor *nor) /* Set SPI NOR sizes. */ params->writesize = 1; params->size = (u64)info->sector_size * info->n_sectors; + params->bank_size = params->size; params->page_size = info->page_size; if (!(info->flags & SPI_NOR_NO_FR)) { diff --git a/drivers/mtd/spi-nor/spansion.c b/drivers/mtd/spi-nor/spansion.c index 15f9a80c10b9bd58096d95214b921dee5c72f48f..36876aa849ede7cda8d9900589fccbcb879c6d93 100644 --- a/drivers/mtd/spi-nor/spansion.c +++ b/drivers/mtd/spi-nor/spansion.c @@ -361,7 +361,7 @@ static int cypress_nor_determine_addr_mode_by_sr1(struct spi_nor *nor, */ static int cypress_nor_set_addr_mode_nbytes(struct spi_nor *nor) { - struct spi_mem_op op; + struct spi_mem_op op = {}; u8 addr_mode; int ret; @@ -492,7 +492,7 @@ s25fs256t_post_bfpt_fixup(struct spi_nor *nor, const struct sfdp_parameter_header *bfpt_header, const struct sfdp_bfpt *bfpt) { - struct spi_mem_op op; + struct spi_mem_op op = {}; int ret; ret = cypress_nor_set_addr_mode_nbytes(nor); diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 3fed888629f7b5d08fede3187fb5639d74185c13..edbaa1444f8ecd9bf344a50f6f599d7eaaf4ff3e 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -3947,7 +3947,11 @@ static int bond_slave_netdev_event(unsigned long event, unblock_netpoll_tx(); break; case NETDEV_FEAT_CHANGE: - bond_compute_features(bond); + if (!bond->notifier_ctx) { + bond->notifier_ctx = true; + bond_compute_features(bond); + bond->notifier_ctx = false; + } break; case NETDEV_RESEND_IGMP: /* Propagate to master device */ @@ -6342,6 +6346,8 @@ static int bond_init(struct net_device *bond_dev) if (!bond->wq) return -ENOMEM; + bond->notifier_ctx = false; + spin_lock_init(&bond->stats_lock); netdev_lockdep_set_classes(bond_dev); diff --git a/drivers/net/can/Kconfig b/drivers/net/can/Kconfig index 3ceccafd701b2a319535791edd73d69513df45bc..b190007c01bec5f4e3b6bf11a60026560e09f333 100644 --- a/drivers/net/can/Kconfig +++ b/drivers/net/can/Kconfig @@ -95,7 +95,7 @@ config CAN_AT91 config CAN_BXCAN tristate "STM32 Basic Extended CAN (bxCAN) devices" - depends on OF || ARCH_STM32 || COMPILE_TEST + depends on ARCH_STM32 || COMPILE_TEST depends on HAS_IOMEM select CAN_RX_OFFLOAD help diff --git a/drivers/net/can/bxcan.c b/drivers/net/can/bxcan.c index e26ccd41e3cba5ae4a13256e3b4dd3bca4ba3b0b..027a8a162fe49fb3457dd943b8d84cd55e4f319e 100644 --- a/drivers/net/can/bxcan.c +++ b/drivers/net/can/bxcan.c @@ -118,7 +118,7 @@ #define BXCAN_FiR1_REG(b) (0x40 + (b) * 8) #define BXCAN_FiR2_REG(b) (0x44 + (b) * 8) -#define BXCAN_FILTER_ID(primary) (primary ? 0 : 14) +#define BXCAN_FILTER_ID(cfg) ((cfg) == BXCAN_CFG_DUAL_SECONDARY ? 14 : 0) /* Filter primary register (FMR) bits */ #define BXCAN_FMR_CANSB_MASK GENMASK(13, 8) @@ -135,6 +135,12 @@ enum bxcan_lec_code { BXCAN_LEC_UNUSED }; +enum bxcan_cfg { + BXCAN_CFG_SINGLE = 0, + BXCAN_CFG_DUAL_PRIMARY, + BXCAN_CFG_DUAL_SECONDARY +}; + /* Structure of the message buffer */ struct bxcan_mb { u32 id; /* can identifier */ @@ -167,7 +173,7 @@ struct bxcan_priv { struct regmap *gcan; int tx_irq; int sce_irq; - bool primary; + enum bxcan_cfg cfg; struct clk *clk; spinlock_t rmw_lock; /* lock for read-modify-write operations */ unsigned int tx_head; @@ -202,17 +208,17 @@ static inline void bxcan_rmw(struct bxcan_priv *priv, void __iomem *addr, spin_unlock_irqrestore(&priv->rmw_lock, flags); } -static void bxcan_disable_filters(struct bxcan_priv *priv, bool primary) +static void bxcan_disable_filters(struct bxcan_priv *priv, enum bxcan_cfg cfg) { - unsigned int fid = BXCAN_FILTER_ID(primary); + unsigned int fid = BXCAN_FILTER_ID(cfg); u32 fmask = BIT(fid); regmap_update_bits(priv->gcan, BXCAN_FA1R_REG, fmask, 0); } -static void bxcan_enable_filters(struct bxcan_priv *priv, bool primary) +static void bxcan_enable_filters(struct bxcan_priv *priv, enum bxcan_cfg cfg) { - unsigned int fid = BXCAN_FILTER_ID(primary); + unsigned int fid = BXCAN_FILTER_ID(cfg); u32 fmask = BIT(fid); /* Filter settings: @@ -680,7 +686,7 @@ static int bxcan_chip_start(struct net_device *ndev) BXCAN_BTR_BRP_MASK | BXCAN_BTR_TS1_MASK | BXCAN_BTR_TS2_MASK | BXCAN_BTR_SJW_MASK, set); - bxcan_enable_filters(priv, priv->primary); + bxcan_enable_filters(priv, priv->cfg); /* Clear all internal status */ priv->tx_head = 0; @@ -806,7 +812,7 @@ static void bxcan_chip_stop(struct net_device *ndev) BXCAN_IER_EPVIE | BXCAN_IER_EWGIE | BXCAN_IER_FOVIE1 | BXCAN_IER_FFIE1 | BXCAN_IER_FMPIE1 | BXCAN_IER_FOVIE0 | BXCAN_IER_FFIE0 | BXCAN_IER_FMPIE0 | BXCAN_IER_TMEIE, 0); - bxcan_disable_filters(priv, priv->primary); + bxcan_disable_filters(priv, priv->cfg); bxcan_enter_sleep_mode(priv); priv->can.state = CAN_STATE_STOPPED; } @@ -931,7 +937,7 @@ static int bxcan_probe(struct platform_device *pdev) struct clk *clk = NULL; void __iomem *regs; struct regmap *gcan; - bool primary; + enum bxcan_cfg cfg; int err, rx_irq, tx_irq, sce_irq; regs = devm_platform_ioremap_resource(pdev, 0); @@ -946,7 +952,13 @@ static int bxcan_probe(struct platform_device *pdev) return PTR_ERR(gcan); } - primary = of_property_read_bool(np, "st,can-primary"); + if (of_property_read_bool(np, "st,can-primary")) + cfg = BXCAN_CFG_DUAL_PRIMARY; + else if (of_property_read_bool(np, "st,can-secondary")) + cfg = BXCAN_CFG_DUAL_SECONDARY; + else + cfg = BXCAN_CFG_SINGLE; + clk = devm_clk_get(dev, NULL); if (IS_ERR(clk)) { dev_err(dev, "failed to get clock\n"); @@ -992,7 +1004,7 @@ static int bxcan_probe(struct platform_device *pdev) priv->clk = clk; priv->tx_irq = tx_irq; priv->sce_irq = sce_irq; - priv->primary = primary; + priv->cfg = cfg; priv->can.clock.freq = clk_get_rate(clk); spin_lock_init(&priv->rmw_lock); priv->tx_head = 0; diff --git a/drivers/net/can/dev/skb.c b/drivers/net/can/dev/skb.c index 241ec636e91fd09f5006b694a3f32fec46662af6..f6d05b3ef59abf85a0bd7f40f9333983b44e4a90 100644 --- a/drivers/net/can/dev/skb.c +++ b/drivers/net/can/dev/skb.c @@ -54,7 +54,8 @@ int can_put_echo_skb(struct sk_buff *skb, struct net_device *dev, /* check flag whether this packet has to be looped back */ if (!(dev->flags & IFF_ECHO) || (skb->protocol != htons(ETH_P_CAN) && - skb->protocol != htons(ETH_P_CANFD))) { + skb->protocol != htons(ETH_P_CANFD) && + skb->protocol != htons(ETH_P_CANXL))) { kfree_skb(skb); return 0; } diff --git a/drivers/net/can/kvaser_pciefd.c b/drivers/net/can/kvaser_pciefd.c index 53e8a914c88b57884b76a81c9a8750535550ca92..be189edb256ceb13590b170e3294ef72264a5092 100644 --- a/drivers/net/can/kvaser_pciefd.c +++ b/drivers/net/can/kvaser_pciefd.c @@ -71,10 +71,12 @@ MODULE_DESCRIPTION("CAN driver for Kvaser CAN/PCIe devices"); #define KVASER_PCIEFD_SYSID_BUILD_REG (KVASER_PCIEFD_SYSID_BASE + 0x14) /* Shared receive buffer registers */ #define KVASER_PCIEFD_SRB_BASE 0x1f200 +#define KVASER_PCIEFD_SRB_FIFO_LAST_REG (KVASER_PCIEFD_SRB_BASE + 0x1f4) #define KVASER_PCIEFD_SRB_CMD_REG (KVASER_PCIEFD_SRB_BASE + 0x200) #define KVASER_PCIEFD_SRB_IEN_REG (KVASER_PCIEFD_SRB_BASE + 0x204) #define KVASER_PCIEFD_SRB_IRQ_REG (KVASER_PCIEFD_SRB_BASE + 0x20c) #define KVASER_PCIEFD_SRB_STAT_REG (KVASER_PCIEFD_SRB_BASE + 0x210) +#define KVASER_PCIEFD_SRB_RX_NR_PACKETS_REG (KVASER_PCIEFD_SRB_BASE + 0x214) #define KVASER_PCIEFD_SRB_CTRL_REG (KVASER_PCIEFD_SRB_BASE + 0x218) /* EPCS flash controller registers */ #define KVASER_PCIEFD_SPI_BASE 0x1fc00 @@ -111,6 +113,9 @@ MODULE_DESCRIPTION("CAN driver for Kvaser CAN/PCIe devices"); /* DMA support */ #define KVASER_PCIEFD_SRB_STAT_DMA BIT(24) +/* SRB current packet level */ +#define KVASER_PCIEFD_SRB_RX_NR_PACKETS_MASK 0xff + /* DMA Enable */ #define KVASER_PCIEFD_SRB_CTRL_DMA_ENABLE BIT(0) @@ -526,7 +531,7 @@ static int kvaser_pciefd_set_tx_irq(struct kvaser_pciefd_can *can) KVASER_PCIEFD_KCAN_IRQ_TOF | KVASER_PCIEFD_KCAN_IRQ_ABD | KVASER_PCIEFD_KCAN_IRQ_TAE | KVASER_PCIEFD_KCAN_IRQ_TAL | KVASER_PCIEFD_KCAN_IRQ_FDIC | KVASER_PCIEFD_KCAN_IRQ_BPP | - KVASER_PCIEFD_KCAN_IRQ_TAR | KVASER_PCIEFD_KCAN_IRQ_TFD; + KVASER_PCIEFD_KCAN_IRQ_TAR; iowrite32(msk, can->reg_base + KVASER_PCIEFD_KCAN_IEN_REG); @@ -554,6 +559,8 @@ static void kvaser_pciefd_setup_controller(struct kvaser_pciefd_can *can) if (can->can.ctrlmode & CAN_CTRLMODE_LISTENONLY) mode |= KVASER_PCIEFD_KCAN_MODE_LOM; + else + mode &= ~KVASER_PCIEFD_KCAN_MODE_LOM; mode |= KVASER_PCIEFD_KCAN_MODE_EEN; mode |= KVASER_PCIEFD_KCAN_MODE_EPEN; @@ -572,7 +579,7 @@ static void kvaser_pciefd_start_controller_flush(struct kvaser_pciefd_can *can) spin_lock_irqsave(&can->lock, irq); iowrite32(-1, can->reg_base + KVASER_PCIEFD_KCAN_IRQ_REG); - iowrite32(KVASER_PCIEFD_KCAN_IRQ_ABD | KVASER_PCIEFD_KCAN_IRQ_TFD, + iowrite32(KVASER_PCIEFD_KCAN_IRQ_ABD, can->reg_base + KVASER_PCIEFD_KCAN_IEN_REG); status = ioread32(can->reg_base + KVASER_PCIEFD_KCAN_STAT_REG); @@ -615,7 +622,7 @@ static int kvaser_pciefd_bus_on(struct kvaser_pciefd_can *can) iowrite32(0, can->reg_base + KVASER_PCIEFD_KCAN_IEN_REG); iowrite32(-1, can->reg_base + KVASER_PCIEFD_KCAN_IRQ_REG); - iowrite32(KVASER_PCIEFD_KCAN_IRQ_ABD | KVASER_PCIEFD_KCAN_IRQ_TFD, + iowrite32(KVASER_PCIEFD_KCAN_IRQ_ABD, can->reg_base + KVASER_PCIEFD_KCAN_IEN_REG); mode = ioread32(can->reg_base + KVASER_PCIEFD_KCAN_MODE_REG); @@ -719,6 +726,7 @@ static int kvaser_pciefd_stop(struct net_device *netdev) iowrite32(0, can->reg_base + KVASER_PCIEFD_KCAN_IEN_REG); del_timer(&can->bec_poll_timer); } + can->can.state = CAN_STATE_STOPPED; close_candev(netdev); return ret; @@ -1007,8 +1015,7 @@ static int kvaser_pciefd_setup_can_ctrls(struct kvaser_pciefd *pcie) SET_NETDEV_DEV(netdev, &pcie->pci->dev); iowrite32(-1, can->reg_base + KVASER_PCIEFD_KCAN_IRQ_REG); - iowrite32(KVASER_PCIEFD_KCAN_IRQ_ABD | - KVASER_PCIEFD_KCAN_IRQ_TFD, + iowrite32(KVASER_PCIEFD_KCAN_IRQ_ABD, can->reg_base + KVASER_PCIEFD_KCAN_IEN_REG); pcie->can[i] = can; @@ -1058,6 +1065,7 @@ static int kvaser_pciefd_setup_dma(struct kvaser_pciefd *pcie) { int i; u32 srb_status; + u32 srb_packet_count; dma_addr_t dma_addr[KVASER_PCIEFD_DMA_COUNT]; /* Disable the DMA */ @@ -1085,6 +1093,15 @@ static int kvaser_pciefd_setup_dma(struct kvaser_pciefd *pcie) KVASER_PCIEFD_SRB_CMD_RDB1, pcie->reg_base + KVASER_PCIEFD_SRB_CMD_REG); + /* Empty Rx FIFO */ + srb_packet_count = ioread32(pcie->reg_base + KVASER_PCIEFD_SRB_RX_NR_PACKETS_REG) & + KVASER_PCIEFD_SRB_RX_NR_PACKETS_MASK; + while (srb_packet_count) { + /* Drop current packet in FIFO */ + ioread32(pcie->reg_base + KVASER_PCIEFD_SRB_FIFO_LAST_REG); + srb_packet_count--; + } + srb_status = ioread32(pcie->reg_base + KVASER_PCIEFD_SRB_STAT_REG); if (!(srb_status & KVASER_PCIEFD_SRB_STAT_DI)) { dev_err(&pcie->pci->dev, "DMA not idle before enabling\n"); @@ -1425,9 +1442,6 @@ static int kvaser_pciefd_handle_status_packet(struct kvaser_pciefd *pcie, cmd = KVASER_PCIEFD_KCAN_CMD_AT; cmd |= ++can->cmd_seq << KVASER_PCIEFD_KCAN_CMD_SEQ_SHIFT; iowrite32(cmd, can->reg_base + KVASER_PCIEFD_KCAN_CMD_REG); - - iowrite32(KVASER_PCIEFD_KCAN_IRQ_TFD, - can->reg_base + KVASER_PCIEFD_KCAN_IEN_REG); } else if (p->header[0] & KVASER_PCIEFD_SPACK_IDET && p->header[0] & KVASER_PCIEFD_SPACK_IRM && cmdseq == (p->header[1] & KVASER_PCIEFD_PACKET_SEQ_MSK) && @@ -1714,15 +1728,6 @@ static int kvaser_pciefd_transmit_irq(struct kvaser_pciefd_can *can) if (irq & KVASER_PCIEFD_KCAN_IRQ_TOF) netdev_err(can->can.dev, "Tx FIFO overflow\n"); - if (irq & KVASER_PCIEFD_KCAN_IRQ_TFD) { - u8 count = ioread32(can->reg_base + - KVASER_PCIEFD_KCAN_TX_NPACKETS_REG) & 0xff; - - if (count == 0) - iowrite32(KVASER_PCIEFD_KCAN_CTRL_EFLUSH, - can->reg_base + KVASER_PCIEFD_KCAN_CTRL_REG); - } - if (irq & KVASER_PCIEFD_KCAN_IRQ_BPP) netdev_err(can->can.dev, "Fail to change bittiming, when not in reset mode\n"); @@ -1824,6 +1829,11 @@ static int kvaser_pciefd_probe(struct pci_dev *pdev, if (err) goto err_teardown_can_ctrls; + err = request_irq(pcie->pci->irq, kvaser_pciefd_irq_handler, + IRQF_SHARED, KVASER_PCIEFD_DRV_NAME, pcie); + if (err) + goto err_teardown_can_ctrls; + iowrite32(KVASER_PCIEFD_SRB_IRQ_DPD0 | KVASER_PCIEFD_SRB_IRQ_DPD1, pcie->reg_base + KVASER_PCIEFD_SRB_IRQ_REG); @@ -1844,11 +1854,6 @@ static int kvaser_pciefd_probe(struct pci_dev *pdev, iowrite32(KVASER_PCIEFD_SRB_CMD_RDB1, pcie->reg_base + KVASER_PCIEFD_SRB_CMD_REG); - err = request_irq(pcie->pci->irq, kvaser_pciefd_irq_handler, - IRQF_SHARED, KVASER_PCIEFD_DRV_NAME, pcie); - if (err) - goto err_teardown_can_ctrls; - err = kvaser_pciefd_reg_candev(pcie); if (err) goto err_free_irq; @@ -1856,6 +1861,8 @@ static int kvaser_pciefd_probe(struct pci_dev *pdev, return 0; err_free_irq: + /* Disable PCI interrupts */ + iowrite32(0, pcie->reg_base + KVASER_PCIEFD_IEN_REG); free_irq(pcie->pci->irq, pcie); err_teardown_can_ctrls: diff --git a/drivers/net/dsa/lan9303-core.c b/drivers/net/dsa/lan9303-core.c index cbe831875347170dee071996a6d9e10a5992d3dd..c0215a8770f49d71c9847853fe38492a8a01d80e 100644 --- a/drivers/net/dsa/lan9303-core.c +++ b/drivers/net/dsa/lan9303-core.c @@ -1188,8 +1188,6 @@ static int lan9303_port_fdb_add(struct dsa_switch *ds, int port, struct lan9303 *chip = ds->priv; dev_dbg(chip->dev, "%s(%d, %pM, %d)\n", __func__, port, addr, vid); - if (vid) - return -EOPNOTSUPP; return lan9303_alr_add_port(chip, addr, port, false); } @@ -1201,8 +1199,6 @@ static int lan9303_port_fdb_del(struct dsa_switch *ds, int port, struct lan9303 *chip = ds->priv; dev_dbg(chip->dev, "%s(%d, %pM, %d)\n", __func__, port, addr, vid); - if (vid) - return -EOPNOTSUPP; lan9303_alr_del_port(chip, addr, port); return 0; diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index 64a2f2f83735d04fdaa343cc80e24de8f1f3b97e..08a46ffd53af99524dc751dd05f2dbece4a6d71f 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -7170,7 +7170,7 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) goto out; } if (chip->reset) - usleep_range(1000, 2000); + usleep_range(10000, 20000); /* Detect if the device is configured in single chip addressing mode, * otherwise continue with address specific smi init/detection. diff --git a/drivers/net/dsa/mv88e6xxx/port.h b/drivers/net/dsa/mv88e6xxx/port.h index aec9d4fd20e3631cbb9cc6eb3951b8e2d74ff644..d19b6303b91f0aafc6924d6098a93f93fc07213c 100644 --- a/drivers/net/dsa/mv88e6xxx/port.h +++ b/drivers/net/dsa/mv88e6xxx/port.h @@ -276,7 +276,7 @@ /* Offset 0x10: Extended Port Control Command */ #define MV88E6393X_PORT_EPC_CMD 0x10 #define MV88E6393X_PORT_EPC_CMD_BUSY 0x8000 -#define MV88E6393X_PORT_EPC_CMD_WRITE 0x0300 +#define MV88E6393X_PORT_EPC_CMD_WRITE 0x3000 #define MV88E6393X_PORT_EPC_INDEX_PORT_ETYPE 0x02 /* Offset 0x11: Extended Port Control Data */ diff --git a/drivers/net/dsa/ocelot/felix_vsc9959.c b/drivers/net/dsa/ocelot/felix_vsc9959.c index cfb3faeaa5bfa9ee282fe28e3084bf8dbc5d1e5c..d172a3e9736c426516e687e6f99ae1302c2c73d6 100644 --- a/drivers/net/dsa/ocelot/felix_vsc9959.c +++ b/drivers/net/dsa/ocelot/felix_vsc9959.c @@ -1263,7 +1263,7 @@ static void vsc9959_tas_guard_bands_update(struct ocelot *ocelot, int port) /* Consider the standard Ethernet overhead of 8 octets preamble+SFD, * 4 octets FCS, 12 octets IFG. */ - needed_bit_time_ps = (maxlen + 24) * picos_per_byte; + needed_bit_time_ps = (u64)(maxlen + 24) * picos_per_byte; dev_dbg(ocelot->dev, "port %d: max frame size %d needs %llu ps at speed %d\n", diff --git a/drivers/net/dsa/qca/Kconfig b/drivers/net/dsa/qca/Kconfig index 4347b42c50fd24762c64b5f41e262570a53ba60c..de9da469908bcff10cf658c3b9f46c4770d015ad 100644 --- a/drivers/net/dsa/qca/Kconfig +++ b/drivers/net/dsa/qca/Kconfig @@ -20,6 +20,7 @@ config NET_DSA_QCA8K_LEDS_SUPPORT bool "Qualcomm Atheros QCA8K Ethernet switch family LEDs support" depends on NET_DSA_QCA8K depends on LEDS_CLASS=y || LEDS_CLASS=NET_DSA_QCA8K + depends on LEDS_TRIGGERS help This enabled support for LEDs present on the Qualcomm Atheros QCA8K Ethernet switch chips. diff --git a/drivers/net/dsa/rzn1_a5psw.c b/drivers/net/dsa/rzn1_a5psw.c index 919027cf20124350e31fcc9ca13f0a19071fee11..c37d2e5372302bd62a9f3c14cd0c768033946ec8 100644 --- a/drivers/net/dsa/rzn1_a5psw.c +++ b/drivers/net/dsa/rzn1_a5psw.c @@ -120,6 +120,22 @@ static void a5psw_port_mgmtfwd_set(struct a5psw *a5psw, int port, bool enable) a5psw_port_pattern_set(a5psw, port, A5PSW_PATTERN_MGMTFWD, enable); } +static void a5psw_port_tx_enable(struct a5psw *a5psw, int port, bool enable) +{ + u32 mask = A5PSW_PORT_ENA_TX(port); + u32 reg = enable ? mask : 0; + + /* Even though the port TX is disabled through TXENA bit in the + * PORT_ENA register, it can still send BPDUs. This depends on the tag + * configuration added when sending packets from the CPU port to the + * switch port. Indeed, when using forced forwarding without filtering, + * even disabled ports will be able to send packets that are tagged. + * This allows to implement STP support when ports are in a state where + * forwarding traffic should be stopped but BPDUs should still be sent. + */ + a5psw_reg_rmw(a5psw, A5PSW_PORT_ENA, mask, reg); +} + static void a5psw_port_enable_set(struct a5psw *a5psw, int port, bool enable) { u32 port_ena = 0; @@ -292,6 +308,22 @@ static int a5psw_set_ageing_time(struct dsa_switch *ds, unsigned int msecs) return 0; } +static void a5psw_port_learning_set(struct a5psw *a5psw, int port, bool learn) +{ + u32 mask = A5PSW_INPUT_LEARN_DIS(port); + u32 reg = !learn ? mask : 0; + + a5psw_reg_rmw(a5psw, A5PSW_INPUT_LEARN, mask, reg); +} + +static void a5psw_port_rx_block_set(struct a5psw *a5psw, int port, bool block) +{ + u32 mask = A5PSW_INPUT_LEARN_BLOCK(port); + u32 reg = block ? mask : 0; + + a5psw_reg_rmw(a5psw, A5PSW_INPUT_LEARN, mask, reg); +} + static void a5psw_flooding_set_resolution(struct a5psw *a5psw, int port, bool set) { @@ -308,6 +340,14 @@ static void a5psw_flooding_set_resolution(struct a5psw *a5psw, int port, a5psw_reg_writel(a5psw, offsets[i], a5psw->bridged_ports); } +static void a5psw_port_set_standalone(struct a5psw *a5psw, int port, + bool standalone) +{ + a5psw_port_learning_set(a5psw, port, !standalone); + a5psw_flooding_set_resolution(a5psw, port, !standalone); + a5psw_port_mgmtfwd_set(a5psw, port, standalone); +} + static int a5psw_port_bridge_join(struct dsa_switch *ds, int port, struct dsa_bridge bridge, bool *tx_fwd_offload, @@ -323,8 +363,7 @@ static int a5psw_port_bridge_join(struct dsa_switch *ds, int port, } a5psw->br_dev = bridge.dev; - a5psw_flooding_set_resolution(a5psw, port, true); - a5psw_port_mgmtfwd_set(a5psw, port, false); + a5psw_port_set_standalone(a5psw, port, false); return 0; } @@ -334,8 +373,7 @@ static void a5psw_port_bridge_leave(struct dsa_switch *ds, int port, { struct a5psw *a5psw = ds->priv; - a5psw_flooding_set_resolution(a5psw, port, false); - a5psw_port_mgmtfwd_set(a5psw, port, true); + a5psw_port_set_standalone(a5psw, port, true); /* No more ports bridged */ if (a5psw->bridged_ports == BIT(A5PSW_CPU_PORT)) @@ -344,28 +382,35 @@ static void a5psw_port_bridge_leave(struct dsa_switch *ds, int port, static void a5psw_port_stp_state_set(struct dsa_switch *ds, int port, u8 state) { - u32 mask = A5PSW_INPUT_LEARN_DIS(port) | A5PSW_INPUT_LEARN_BLOCK(port); + bool learning_enabled, rx_enabled, tx_enabled; struct a5psw *a5psw = ds->priv; - u32 reg = 0; switch (state) { case BR_STATE_DISABLED: case BR_STATE_BLOCKING: - reg |= A5PSW_INPUT_LEARN_DIS(port); - reg |= A5PSW_INPUT_LEARN_BLOCK(port); - break; case BR_STATE_LISTENING: - reg |= A5PSW_INPUT_LEARN_DIS(port); + rx_enabled = false; + tx_enabled = false; + learning_enabled = false; break; case BR_STATE_LEARNING: - reg |= A5PSW_INPUT_LEARN_BLOCK(port); + rx_enabled = false; + tx_enabled = false; + learning_enabled = true; break; case BR_STATE_FORWARDING: - default: + rx_enabled = true; + tx_enabled = true; + learning_enabled = true; break; + default: + dev_err(ds->dev, "invalid STP state: %d\n", state); + return; } - a5psw_reg_rmw(a5psw, A5PSW_INPUT_LEARN, mask, reg); + a5psw_port_learning_set(a5psw, port, learning_enabled); + a5psw_port_rx_block_set(a5psw, port, !rx_enabled); + a5psw_port_tx_enable(a5psw, port, tx_enabled); } static void a5psw_port_fast_age(struct dsa_switch *ds, int port) @@ -673,7 +718,7 @@ static int a5psw_setup(struct dsa_switch *ds) } /* Configure management port */ - reg = A5PSW_CPU_PORT | A5PSW_MGMT_CFG_DISCARD; + reg = A5PSW_CPU_PORT | A5PSW_MGMT_CFG_ENABLE; a5psw_reg_writel(a5psw, A5PSW_MGMT_CFG, reg); /* Set pattern 0 to forward all frame to mgmt port */ @@ -722,13 +767,15 @@ static int a5psw_setup(struct dsa_switch *ds) if (dsa_port_is_unused(dp)) continue; - /* Enable egress flooding for CPU port */ - if (dsa_port_is_cpu(dp)) + /* Enable egress flooding and learning for CPU port */ + if (dsa_port_is_cpu(dp)) { a5psw_flooding_set_resolution(a5psw, port, true); + a5psw_port_learning_set(a5psw, port, true); + } - /* Enable management forward only for user ports */ + /* Enable standalone mode for user ports */ if (dsa_port_is_user(dp)) - a5psw_port_mgmtfwd_set(a5psw, port, true); + a5psw_port_set_standalone(a5psw, port, true); } return 0; diff --git a/drivers/net/dsa/rzn1_a5psw.h b/drivers/net/dsa/rzn1_a5psw.h index c67abd49c013d356a5d00c293070c4cdde8e1ce9..b869192eef3f71bb224b8308f3f6301e1efa5218 100644 --- a/drivers/net/dsa/rzn1_a5psw.h +++ b/drivers/net/dsa/rzn1_a5psw.h @@ -19,6 +19,7 @@ #define A5PSW_PORT_OFFSET(port) (0x400 * (port)) #define A5PSW_PORT_ENA 0x8 +#define A5PSW_PORT_ENA_TX(port) BIT(port) #define A5PSW_PORT_ENA_RX_SHIFT 16 #define A5PSW_PORT_ENA_TX_RX(port) (BIT((port) + A5PSW_PORT_ENA_RX_SHIFT) | \ BIT(port)) @@ -36,7 +37,7 @@ #define A5PSW_INPUT_LEARN_BLOCK(p) BIT(p) #define A5PSW_MGMT_CFG 0x20 -#define A5PSW_MGMT_CFG_DISCARD BIT(7) +#define A5PSW_MGMT_CFG_ENABLE BIT(6) #define A5PSW_MODE_CFG 0x24 #define A5PSW_MODE_STATS_RESET BIT(31) diff --git a/drivers/net/ethernet/3com/3c515.c b/drivers/net/ethernet/3com/3c515.c index d2f4358cc5503e939d1fd23fbb37933bb494aa36..ba3e7aa1a28fb577185be24a6592d37f298e9eef 100644 --- a/drivers/net/ethernet/3com/3c515.c +++ b/drivers/net/ethernet/3com/3c515.c @@ -66,8 +66,10 @@ static int max_interrupt_work = 20; #include #include #include - #include + +#include + #include #include diff --git a/drivers/net/ethernet/3com/3c589_cs.c b/drivers/net/ethernet/3com/3c589_cs.c index 82f94b1635bf816150e5f3c4f7d3e71417d81263..5267e9dcd87ef90927e7806206dba7d108e96ddc 100644 --- a/drivers/net/ethernet/3com/3c589_cs.c +++ b/drivers/net/ethernet/3com/3c589_cs.c @@ -195,6 +195,7 @@ static int tc589_probe(struct pcmcia_device *link) { struct el3_private *lp; struct net_device *dev; + int ret; dev_dbg(&link->dev, "3c589_attach()\n"); @@ -218,7 +219,15 @@ static int tc589_probe(struct pcmcia_device *link) dev->ethtool_ops = &netdev_ethtool_ops; - return tc589_config(link); + ret = tc589_config(link); + if (ret) + goto err_free_netdev; + + return 0; + +err_free_netdev: + free_netdev(dev); + return ret; } static void tc589_detach(struct pcmcia_device *link) diff --git a/drivers/net/ethernet/8390/ne.c b/drivers/net/ethernet/8390/ne.c index 0a9118b8be0c64b481f8e43988f5a8679c461e5c..bc9c81dc00fd20c37311da008e9d6bc04dc8f864 100644 --- a/drivers/net/ethernet/8390/ne.c +++ b/drivers/net/ethernet/8390/ne.c @@ -52,6 +52,7 @@ static const char version2[] = #include #include #include +#include #include diff --git a/drivers/net/ethernet/8390/smc-ultra.c b/drivers/net/ethernet/8390/smc-ultra.c index 6e62c37c9400565616fd35a061aece8eb4beca8f..7465650c80780594203516d5a4915a9122d04448 100644 --- a/drivers/net/ethernet/8390/smc-ultra.c +++ b/drivers/net/ethernet/8390/smc-ultra.c @@ -66,6 +66,7 @@ static const char version[] = #include #include #include +#include #include #include diff --git a/drivers/net/ethernet/8390/wd.c b/drivers/net/ethernet/8390/wd.c index 5b00c452bede64a71519d34dfab79a0d50e05f64..119021d41451e88726af07852b587b4d33673efd 100644 --- a/drivers/net/ethernet/8390/wd.c +++ b/drivers/net/ethernet/8390/wd.c @@ -37,6 +37,7 @@ static const char version[] = #include #include #include +#include #include diff --git a/drivers/net/ethernet/amd/lance.c b/drivers/net/ethernet/amd/lance.c index 8971665a4b2ac3adc53de5fd292dfe479deec304..6cf38180cc01934f6252d2dcc4a3b3b9c8fecfe1 100644 --- a/drivers/net/ethernet/amd/lance.c +++ b/drivers/net/ethernet/amd/lance.c @@ -59,6 +59,7 @@ static const char version[] = "lance.c:v1.16 2006/11/09 dplatt@3do.com, becker@c #include #include #include +#include #include #include diff --git a/drivers/net/ethernet/amd/pds_core/dev.c b/drivers/net/ethernet/amd/pds_core/dev.c index f7c597ea5daf3dd0582b5fc98737a8b20f0a0f63..debe5216fe29e2e8ada260ffd3fd4887f5b161d6 100644 --- a/drivers/net/ethernet/amd/pds_core/dev.c +++ b/drivers/net/ethernet/amd/pds_core/dev.c @@ -68,9 +68,15 @@ bool pdsc_is_fw_running(struct pdsc *pdsc) bool pdsc_is_fw_good(struct pdsc *pdsc) { - u8 gen = pdsc->fw_status & PDS_CORE_FW_STS_F_GENERATION; + bool fw_running = pdsc_is_fw_running(pdsc); + u8 gen; - return pdsc_is_fw_running(pdsc) && gen == pdsc->fw_generation; + /* Make sure to update the cached fw_status by calling + * pdsc_is_fw_running() before getting the generation + */ + gen = pdsc->fw_status & PDS_CORE_FW_STS_F_GENERATION; + + return fw_running && gen == pdsc->fw_generation; } static u8 pdsc_devcmd_status(struct pdsc *pdsc) diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c index 33a9574e9e0434fbdfcfd022b9731329c6fbad8f..32d2c6fac65266baee9bc36477f21418b50f67e0 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c @@ -1329,7 +1329,7 @@ static enum xgbe_mode xgbe_phy_status_aneg(struct xgbe_prv_data *pdata) return pdata->phy_if.phy_impl.an_outcome(pdata); } -static void xgbe_phy_status_result(struct xgbe_prv_data *pdata) +static bool xgbe_phy_status_result(struct xgbe_prv_data *pdata) { struct ethtool_link_ksettings *lks = &pdata->phy.lks; enum xgbe_mode mode; @@ -1367,8 +1367,13 @@ static void xgbe_phy_status_result(struct xgbe_prv_data *pdata) pdata->phy.duplex = DUPLEX_FULL; - if (xgbe_set_mode(pdata, mode) && pdata->an_again) + if (!xgbe_set_mode(pdata, mode)) + return false; + + if (pdata->an_again) xgbe_phy_reconfig_aneg(pdata); + + return true; } static void xgbe_phy_status(struct xgbe_prv_data *pdata) @@ -1398,7 +1403,8 @@ static void xgbe_phy_status(struct xgbe_prv_data *pdata) return; } - xgbe_phy_status_result(pdata); + if (xgbe_phy_status_result(pdata)) + return; if (test_bit(XGBE_LINK_INIT, &pdata->dev_state)) clear_bit(XGBE_LINK_INIT, &pdata->dev_state); diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c index 38d0cdaf22a5b3bd9d5ae396d02f54609e1254d7..bf1611cce974a95d6e53df6cf52e0a47ecf25311 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.c +++ b/drivers/net/ethernet/broadcom/bcmsysport.c @@ -2531,9 +2531,9 @@ static int bcm_sysport_probe(struct platform_device *pdev) priv->irq0 = platform_get_irq(pdev, 0); if (!priv->is_lite) { priv->irq1 = platform_get_irq(pdev, 1); - priv->wol_irq = platform_get_irq(pdev, 2); + priv->wol_irq = platform_get_irq_optional(pdev, 2); } else { - priv->wol_irq = platform_get_irq(pdev, 1); + priv->wol_irq = platform_get_irq_optional(pdev, 1); } if (priv->irq0 <= 0 || (priv->irq1 <= 0 && !priv->is_lite)) { ret = -EINVAL; diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 637d162bbcfaf662407c819468e7e2379918b9f2..1e7a6f1d422337f855fc5d0e578f9bfe137017f5 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -14294,11 +14294,16 @@ static void bnx2x_io_resume(struct pci_dev *pdev) bp->fw_seq = SHMEM_RD(bp, func_mb[BP_FW_MB_IDX(bp)].drv_mb_header) & DRV_MSG_SEQ_NUMBER_MASK; - if (netif_running(dev)) - bnx2x_nic_load(bp, LOAD_NORMAL); + if (netif_running(dev)) { + if (bnx2x_nic_load(bp, LOAD_NORMAL)) { + netdev_err(bp->dev, "Error during driver initialization, try unloading/reloading the driver\n"); + goto done; + } + } netif_device_attach(dev); +done: rtnl_unlock(); } diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index dcd9367f05afd476afa1294a2fb18902e2d08ffd..b499bc9c4e06701c3d5bc02e0898d5c79ee1b267 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -692,7 +692,7 @@ static void bnxt_tx_int(struct bnxt *bp, struct bnxt_napi *bnapi, int nr_pkts) __netif_txq_completed_wake(txq, nr_pkts, tx_bytes, bnxt_tx_avail(bp, txr), bp->tx_wake_thresh, - READ_ONCE(txr->dev_state) != BNXT_DEV_STATE_CLOSING); + READ_ONCE(txr->dev_state) == BNXT_DEV_STATE_CLOSING); } static struct page *__bnxt_alloc_rx_page(struct bnxt *bp, dma_addr_t *mapping, @@ -2365,6 +2365,9 @@ static int bnxt_async_event_process(struct bnxt *bp, struct bnxt_ptp_cfg *ptp = bp->ptp_cfg; u64 ns; + if (!ptp) + goto async_event_process_exit; + spin_lock_bh(&ptp->ptp_lock); bnxt_ptp_update_current_time(bp); ns = (((u64)BNXT_EVENT_PHC_RTC_UPDATE(data1) << @@ -4763,6 +4766,9 @@ int bnxt_hwrm_func_drv_rgtr(struct bnxt *bp, unsigned long *bmap, int bmap_size, if (event_id == ASYNC_EVENT_CMPL_EVENT_ID_ERROR_RECOVERY && !(bp->fw_cap & BNXT_FW_CAP_ERROR_RECOVERY)) continue; + if (event_id == ASYNC_EVENT_CMPL_EVENT_ID_PHC_UPDATE && + !bp->ptp_cfg) + continue; __set_bit(bnxt_async_events_arr[i], async_events_bmap); } if (bmap && bmap_size) { @@ -5350,6 +5356,7 @@ static void bnxt_hwrm_update_rss_hash_cfg(struct bnxt *bp) if (hwrm_req_init(bp, req, HWRM_VNIC_RSS_QCFG)) return; + req->vnic_id = cpu_to_le16(vnic->fw_vnic_id); /* all contexts configured to same hash_type, zero always exists */ req->rss_ctx_idx = cpu_to_le16(vnic->fw_rss_cos_lb_ctx[0]); resp = hwrm_req_hold(bp, req); @@ -8812,6 +8819,9 @@ static int bnxt_init_chip(struct bnxt *bp, bool irq_re_init) goto err_out; } + if (BNXT_VF(bp)) + bnxt_hwrm_func_qcfg(bp); + rc = bnxt_setup_vnic(bp, 0); if (rc) goto err_out; @@ -11598,6 +11608,7 @@ static void bnxt_tx_timeout(struct net_device *dev, unsigned int txqueue) static void bnxt_fw_health_check(struct bnxt *bp) { struct bnxt_fw_health *fw_health = bp->fw_health; + struct pci_dev *pdev = bp->pdev; u32 val; if (!fw_health->enabled || test_bit(BNXT_STATE_IN_FW_RESET, &bp->state)) @@ -11611,7 +11622,7 @@ static void bnxt_fw_health_check(struct bnxt *bp) } val = bnxt_fw_health_readl(bp, BNXT_FW_HEARTBEAT_REG); - if (val == fw_health->last_fw_heartbeat) { + if (val == fw_health->last_fw_heartbeat && pci_device_is_present(pdev)) { fw_health->arrests++; goto fw_reset; } @@ -11619,7 +11630,7 @@ static void bnxt_fw_health_check(struct bnxt *bp) fw_health->last_fw_heartbeat = val; val = bnxt_fw_health_readl(bp, BNXT_FW_RESET_CNT_REG); - if (val != fw_health->last_fw_reset_cnt) { + if (val != fw_health->last_fw_reset_cnt && pci_device_is_present(pdev)) { fw_health->discoveries++; goto fw_reset; } @@ -13025,26 +13036,37 @@ static void bnxt_cfg_ntp_filters(struct bnxt *bp) #endif /* CONFIG_RFS_ACCEL */ -static int bnxt_udp_tunnel_sync(struct net_device *netdev, unsigned int table) +static int bnxt_udp_tunnel_set_port(struct net_device *netdev, unsigned int table, + unsigned int entry, struct udp_tunnel_info *ti) { struct bnxt *bp = netdev_priv(netdev); - struct udp_tunnel_info ti; unsigned int cmd; - udp_tunnel_nic_get_port(netdev, table, 0, &ti); - if (ti.type == UDP_TUNNEL_TYPE_VXLAN) + if (ti->type == UDP_TUNNEL_TYPE_VXLAN) cmd = TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_VXLAN; else cmd = TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE; - if (ti.port) - return bnxt_hwrm_tunnel_dst_port_alloc(bp, ti.port, cmd); + return bnxt_hwrm_tunnel_dst_port_alloc(bp, ti->port, cmd); +} + +static int bnxt_udp_tunnel_unset_port(struct net_device *netdev, unsigned int table, + unsigned int entry, struct udp_tunnel_info *ti) +{ + struct bnxt *bp = netdev_priv(netdev); + unsigned int cmd; + + if (ti->type == UDP_TUNNEL_TYPE_VXLAN) + cmd = TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_VXLAN; + else + cmd = TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE; return bnxt_hwrm_tunnel_dst_port_free(bp, cmd); } static const struct udp_tunnel_nic_info bnxt_udp_tunnels = { - .sync_table = bnxt_udp_tunnel_sync, + .set_port = bnxt_udp_tunnel_set_port, + .unset_port = bnxt_udp_tunnel_unset_port, .flags = UDP_TUNNEL_NIC_INFO_MAY_SLEEP | UDP_TUNNEL_NIC_INFO_OPEN_ONLY, .tables = { diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index 2dd8ee4a6f75b76b0ea794925a979400bac85930..8fd5071d8b09951402d217946300a1fed40912d7 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -3831,7 +3831,7 @@ static int bnxt_reset(struct net_device *dev, u32 *flags) } } - if (req & BNXT_FW_RESET_AP) { + if (!BNXT_CHIP_P4_PLUS(bp) && (req & BNXT_FW_RESET_AP)) { /* This feature is not supported in older firmware versions */ if (bp->hwrm_spec_code >= 0x10803) { if (!bnxt_firmware_reset_ap(dev)) { diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c index e46689128e323fd1d7748c2d509a38b308a27f3d..f3886710e77873a83e8b8aad3eb2f3e2ba7d465e 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c @@ -952,6 +952,7 @@ int bnxt_ptp_init(struct bnxt *bp, bool phc_cfg) bnxt_ptp_timecounter_init(bp, true); bnxt_ptp_adjfine_rtc(bp, 0); } + bnxt_hwrm_func_drv_rgtr(bp, NULL, 0, true); ptp->ptp_info = bnxt_ptp_caps; if ((bp->fw_cap & BNXT_FW_CAP_PTP_PPS)) { diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index f28ffc31df2200660954cbebea08a5e06232b947..2b5761ad2f92f1f0be680e56981a6ca8266c7d9b 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -1272,7 +1272,8 @@ static void bcmgenet_get_ethtool_stats(struct net_device *dev, } } -static void bcmgenet_eee_enable_set(struct net_device *dev, bool enable) +void bcmgenet_eee_enable_set(struct net_device *dev, bool enable, + bool tx_lpi_enabled) { struct bcmgenet_priv *priv = netdev_priv(dev); u32 off = priv->hw_params->tbuf_offset + TBUF_ENERGY_CTRL; @@ -1292,7 +1293,7 @@ static void bcmgenet_eee_enable_set(struct net_device *dev, bool enable) /* Enable EEE and switch to a 27Mhz clock automatically */ reg = bcmgenet_readl(priv->base + off); - if (enable) + if (tx_lpi_enabled) reg |= TBUF_EEE_EN | TBUF_PM_EN; else reg &= ~(TBUF_EEE_EN | TBUF_PM_EN); @@ -1313,6 +1314,7 @@ static void bcmgenet_eee_enable_set(struct net_device *dev, bool enable) priv->eee.eee_enabled = enable; priv->eee.eee_active = enable; + priv->eee.tx_lpi_enabled = tx_lpi_enabled; } static int bcmgenet_get_eee(struct net_device *dev, struct ethtool_eee *e) @@ -1328,6 +1330,7 @@ static int bcmgenet_get_eee(struct net_device *dev, struct ethtool_eee *e) e->eee_enabled = p->eee_enabled; e->eee_active = p->eee_active; + e->tx_lpi_enabled = p->tx_lpi_enabled; e->tx_lpi_timer = bcmgenet_umac_readl(priv, UMAC_EEE_LPI_TIMER); return phy_ethtool_get_eee(dev->phydev, e); @@ -1337,7 +1340,6 @@ static int bcmgenet_set_eee(struct net_device *dev, struct ethtool_eee *e) { struct bcmgenet_priv *priv = netdev_priv(dev); struct ethtool_eee *p = &priv->eee; - int ret = 0; if (GENET_IS_V1(priv)) return -EOPNOTSUPP; @@ -1348,16 +1350,11 @@ static int bcmgenet_set_eee(struct net_device *dev, struct ethtool_eee *e) p->eee_enabled = e->eee_enabled; if (!p->eee_enabled) { - bcmgenet_eee_enable_set(dev, false); + bcmgenet_eee_enable_set(dev, false, false); } else { - ret = phy_init_eee(dev->phydev, false); - if (ret) { - netif_err(priv, hw, dev, "EEE initialization failed\n"); - return ret; - } - + p->eee_active = phy_init_eee(dev->phydev, false) >= 0; bcmgenet_umac_writel(priv, e->tx_lpi_timer, UMAC_EEE_LPI_TIMER); - bcmgenet_eee_enable_set(dev, true); + bcmgenet_eee_enable_set(dev, p->eee_active, e->tx_lpi_enabled); } return phy_ethtool_set_eee(dev->phydev, e); @@ -3450,7 +3447,7 @@ static int bcmgenet_open(struct net_device *dev) return ret; } -static void bcmgenet_netif_stop(struct net_device *dev) +static void bcmgenet_netif_stop(struct net_device *dev, bool stop_phy) { struct bcmgenet_priv *priv = netdev_priv(dev); @@ -3465,6 +3462,8 @@ static void bcmgenet_netif_stop(struct net_device *dev) /* Disable MAC transmit. TX DMA disabled must be done before this */ umac_enable_set(priv, CMD_TX_EN, false); + if (stop_phy) + phy_stop(dev->phydev); bcmgenet_disable_rx_napi(priv); bcmgenet_intr_disable(priv); @@ -3485,7 +3484,7 @@ static int bcmgenet_close(struct net_device *dev) netif_dbg(priv, ifdown, dev, "bcmgenet_close\n"); - bcmgenet_netif_stop(dev); + bcmgenet_netif_stop(dev, false); /* Really kill the PHY state machine and disconnect from it */ phy_disconnect(dev->phydev); @@ -4277,9 +4276,6 @@ static int bcmgenet_resume(struct device *d) if (!device_may_wakeup(d)) phy_resume(dev->phydev); - if (priv->eee.eee_enabled) - bcmgenet_eee_enable_set(dev, true); - bcmgenet_netif_start(dev); netif_device_attach(dev); @@ -4303,7 +4299,7 @@ static int bcmgenet_suspend(struct device *d) netif_device_detach(dev); - bcmgenet_netif_stop(dev); + bcmgenet_netif_stop(dev, true); if (!device_may_wakeup(d)) phy_suspend(dev->phydev); diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.h b/drivers/net/ethernet/broadcom/genet/bcmgenet.h index 946f6e283c4e6b24d4588ae6f42687ed58d84f41..1985c0ec4da2ab05af224113b2b57c6cf33b7b74 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.h +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.h @@ -703,4 +703,7 @@ int bcmgenet_wol_power_down_cfg(struct bcmgenet_priv *priv, void bcmgenet_wol_power_up_cfg(struct bcmgenet_priv *priv, enum bcmgenet_power_mode mode); +void bcmgenet_eee_enable_set(struct net_device *dev, bool enable, + bool tx_lpi_enabled); + #endif /* __BCMGENET_H__ */ diff --git a/drivers/net/ethernet/broadcom/genet/bcmmii.c b/drivers/net/ethernet/broadcom/genet/bcmmii.c index be042905ada2a41a35250aafde6b8c984d865c3d..c15ed0acdb777deef01816683e4024dcc27e5277 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmmii.c +++ b/drivers/net/ethernet/broadcom/genet/bcmmii.c @@ -87,6 +87,11 @@ static void bcmgenet_mac_config(struct net_device *dev) reg |= CMD_TX_EN | CMD_RX_EN; } bcmgenet_umac_writel(priv, reg, UMAC_CMD); + + priv->eee.eee_active = phy_init_eee(phydev, 0) >= 0; + bcmgenet_eee_enable_set(dev, + priv->eee.eee_enabled && priv->eee.eee_active, + priv->eee.tx_lpi_enabled); } /* setup netdev link state when PHY link status change and diff --git a/drivers/net/ethernet/cirrus/cs89x0.c b/drivers/net/ethernet/cirrus/cs89x0.c index 06a0c00af99c7562d334c4e391686233267b4671..276c32c3926a7d9ae12991cacf2ae1ee87e0f0d5 100644 --- a/drivers/net/ethernet/cirrus/cs89x0.c +++ b/drivers/net/ethernet/cirrus/cs89x0.c @@ -72,6 +72,8 @@ #include #include +#include + #include #include #if ALLOW_DMA diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c index 3c4fa26f0f9b1c55a46fceae945234e5dfe068bc..9e1b2536e9a916f5541fbe3f9fbfd51109767e2c 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc.c +++ b/drivers/net/ethernet/freescale/enetc/enetc.c @@ -1229,7 +1229,13 @@ static int enetc_clean_rx_ring(struct enetc_bdr *rx_ring, if (!skb) break; - rx_byte_cnt += skb->len; + /* When set, the outer VLAN header is extracted and reported + * in the receive buffer descriptor. So rx_byte_cnt should + * add the length of the extracted VLAN header. + */ + if (bd_status & ENETC_RXBD_FLAG_VLAN) + rx_byte_cnt += VLAN_HLEN; + rx_byte_cnt += skb->len + ETH_HLEN; rx_frm_cnt++; napi_gro_receive(napi, skb); @@ -1565,6 +1571,14 @@ static int enetc_clean_rx_ring_xdp(struct enetc_bdr *rx_ring, enetc_build_xdp_buff(rx_ring, bd_status, &rxbd, &i, &cleaned_cnt, &xdp_buff); + /* When set, the outer VLAN header is extracted and reported + * in the receive buffer descriptor. So rx_byte_cnt should + * add the length of the extracted VLAN header. + */ + if (bd_status & ENETC_RXBD_FLAG_VLAN) + rx_byte_cnt += VLAN_HLEN; + rx_byte_cnt += xdp_get_buff_len(&xdp_buff); + xdp_act = bpf_prog_run_xdp(prog, &xdp_buff); switch (xdp_act) { diff --git a/drivers/net/ethernet/freescale/enetc/enetc_qos.c b/drivers/net/ethernet/freescale/enetc/enetc_qos.c index 83c27bbbc6edffa294d9f20618583622aeb031e2..126007ab70f619f3497e0d1ad76e2317fb78a71c 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_qos.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_qos.c @@ -181,8 +181,8 @@ int enetc_setup_tc_cbs(struct net_device *ndev, void *type_data) int bw_sum = 0; u8 bw; - prio_top = netdev_get_prio_tc_map(ndev, tc_nums - 1); - prio_next = netdev_get_prio_tc_map(ndev, tc_nums - 2); + prio_top = tc_nums - 1; + prio_next = tc_nums - 2; /* Support highest prio and second prio tc in cbs mode */ if (tc != prio_top && tc != prio_next) diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 42ec6ca3bf035a87e11a7dfc92b19483e843cae1..38e5b5abe067cf6ce8a2bf8144c9dd7245ae2e8c 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -3798,7 +3798,6 @@ static int fec_enet_txq_xmit_frame(struct fec_enet_private *fep, entries_free = fec_enet_get_free_txdesc_num(txq); if (entries_free < MAX_SKB_FRAGS + 1) { netdev_err(fep->netdev, "NOT enough BD for SG!\n"); - xdp_return_frame(frame); return NETDEV_TX_BUSY; } @@ -3835,6 +3834,11 @@ static int fec_enet_txq_xmit_frame(struct fec_enet_private *fep, index = fec_enet_get_bd_index(last_bdp, &txq->bd); txq->tx_skbuff[index] = NULL; + /* Make sure the updates to rest of the descriptor are performed before + * transferring ownership. + */ + dma_wmb(); + /* Send it on its way. Tell FEC it's ready, interrupt when done, * it's the last BD of the frame, and to put the CRC on the end. */ @@ -3844,8 +3848,14 @@ static int fec_enet_txq_xmit_frame(struct fec_enet_private *fep, /* If this was the last BD in the ring, start at the beginning again. */ bdp = fec_enet_get_nextdesc(last_bdp, &txq->bd); + /* Make sure the update to bdp are performed before txq->bd.cur. */ + dma_wmb(); + txq->bd.cur = bdp; + /* Trigger transmission start */ + writel(0, txq->bd.reg_desc_active); + return 0; } @@ -3874,12 +3884,6 @@ static int fec_enet_xdp_xmit(struct net_device *dev, sent_frames++; } - /* Make sure the update to bdp and tx_skbuff are performed. */ - wmb(); - - /* Trigger transmission start */ - writel(0, txq->bd.reg_desc_active); - __netif_tx_unlock(nq); return sent_frames; @@ -4478,9 +4482,11 @@ fec_drv_remove(struct platform_device *pdev) struct device_node *np = pdev->dev.of_node; int ret; - ret = pm_runtime_resume_and_get(&pdev->dev); + ret = pm_runtime_get_sync(&pdev->dev); if (ret < 0) - return ret; + dev_err(&pdev->dev, + "Failed to resume device in remove callback (%pe)\n", + ERR_PTR(ret)); cancel_work_sync(&fep->tx_timeout_work); fec_ptp_stop(pdev); @@ -4493,8 +4499,13 @@ fec_drv_remove(struct platform_device *pdev) of_phy_deregister_fixed_link(np); of_node_put(fep->phy_node); - clk_disable_unprepare(fep->clk_ahb); - clk_disable_unprepare(fep->clk_ipg); + /* After pm_runtime_get_sync() failed, the clks are still off, so skip + * disabling them again. + */ + if (ret >= 0) { + clk_disable_unprepare(fep->clk_ahb); + clk_disable_unprepare(fep->clk_ipg); + } pm_runtime_put_noidle(&pdev->dev); pm_runtime_disable(&pdev->dev); diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.c b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.c index cbbab5b2b402b70d4b1a2f680fa8b773c1cb7744..b85c412683ddc227ad06bc71b5357f80e9d50c93 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.c @@ -331,9 +331,25 @@ static int hclge_comm_cmd_csq_done(struct hclge_comm_hw *hw) return head == hw->cmq.csq.next_to_use; } -static void hclge_comm_wait_for_resp(struct hclge_comm_hw *hw, +static u32 hclge_get_cmdq_tx_timeout(u16 opcode, u32 tx_timeout) +{ + static const struct hclge_cmdq_tx_timeout_map cmdq_tx_timeout_map[] = { + {HCLGE_OPC_CFG_RST_TRIGGER, HCLGE_COMM_CMDQ_TX_TIMEOUT_500MS}, + }; + u32 i; + + for (i = 0; i < ARRAY_SIZE(cmdq_tx_timeout_map); i++) + if (cmdq_tx_timeout_map[i].opcode == opcode) + return cmdq_tx_timeout_map[i].tx_timeout; + + return tx_timeout; +} + +static void hclge_comm_wait_for_resp(struct hclge_comm_hw *hw, u16 opcode, bool *is_completed) { + u32 cmdq_tx_timeout = hclge_get_cmdq_tx_timeout(opcode, + hw->cmq.tx_timeout); u32 timeout = 0; do { @@ -343,7 +359,7 @@ static void hclge_comm_wait_for_resp(struct hclge_comm_hw *hw, } udelay(1); timeout++; - } while (timeout < hw->cmq.tx_timeout); + } while (timeout < cmdq_tx_timeout); } static int hclge_comm_cmd_convert_err_code(u16 desc_ret) @@ -407,7 +423,8 @@ static int hclge_comm_cmd_check_result(struct hclge_comm_hw *hw, * if multi descriptors to be sent, use the first one to check */ if (HCLGE_COMM_SEND_SYNC(le16_to_cpu(desc->flag))) - hclge_comm_wait_for_resp(hw, &is_completed); + hclge_comm_wait_for_resp(hw, le16_to_cpu(desc->opcode), + &is_completed); if (!is_completed) ret = -EBADE; @@ -529,7 +546,7 @@ int hclge_comm_cmd_queue_init(struct pci_dev *pdev, struct hclge_comm_hw *hw) cmdq->crq.desc_num = HCLGE_COMM_NIC_CMQ_DESC_NUM; /* Setup Tx write back timeout */ - cmdq->tx_timeout = HCLGE_COMM_CMDQ_TX_TIMEOUT; + cmdq->tx_timeout = HCLGE_COMM_CMDQ_TX_TIMEOUT_DEFAULT; /* Setup queue rings */ ret = hclge_comm_alloc_cmd_queue(hw, HCLGE_COMM_TYPE_CSQ); diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.h b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.h index de72ecbfd5ad8549653dc730e986a1b3a0f37af9..18f1b4bf362da9b83f4fe9fd8d4885842ff4d7dc 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.h @@ -54,7 +54,8 @@ #define HCLGE_COMM_NIC_SW_RST_RDY BIT(HCLGE_COMM_NIC_SW_RST_RDY_B) #define HCLGE_COMM_NIC_CMQ_DESC_NUM_S 3 #define HCLGE_COMM_NIC_CMQ_DESC_NUM 1024 -#define HCLGE_COMM_CMDQ_TX_TIMEOUT 30000 +#define HCLGE_COMM_CMDQ_TX_TIMEOUT_DEFAULT 30000 +#define HCLGE_COMM_CMDQ_TX_TIMEOUT_500MS 500000 enum hclge_opcode_type { /* Generic commands */ @@ -360,6 +361,11 @@ struct hclge_comm_caps_bit_map { u16 local_bit; }; +struct hclge_cmdq_tx_timeout_map { + u32 opcode; + u32 tx_timeout; +}; + struct hclge_comm_firmware_compat_cmd { __le32 compat; u8 rsv[20]; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c index 4c3e90a1c4d07368ad4757926aeced0ea8832808..d385ffc21876601d94022104c545fa95be60ddcb 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c @@ -130,7 +130,7 @@ static struct hns3_dbg_cmd_info hns3_dbg_cmd[] = { .name = "tx_bd_queue", .cmd = HNAE3_DBG_CMD_TX_BD, .dentry = HNS3_DBG_DENTRY_TX_BD, - .buf_len = HNS3_DBG_READ_LEN_4MB, + .buf_len = HNS3_DBG_READ_LEN_5MB, .init = hns3_dbg_bd_file_init, }, { diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.h b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.h index 97578eabb7d8b7a2defc1ac1422bb77f29b1f6bf..4a5ef8a90a1046883416f00648e5f47ff2a42721 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.h @@ -10,6 +10,7 @@ #define HNS3_DBG_READ_LEN_128KB 0x20000 #define HNS3_DBG_READ_LEN_1MB 0x100000 #define HNS3_DBG_READ_LEN_4MB 0x400000 +#define HNS3_DBG_READ_LEN_5MB 0x500000 #define HNS3_DBG_WRITE_LEN 1024 #define HNS3_DBG_DATA_STR_LEN 32 diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 4fb5406c1951dce7095010a4e7c7528d8e59091b..2689b108f7df7b4e94efdeec45135b8b297a15d7 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -8053,12 +8053,15 @@ static void hclge_ae_stop(struct hnae3_handle *handle) /* If it is not PF reset or FLR, the firmware will disable the MAC, * so it only need to stop phy here. */ - if (test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state) && - hdev->reset_type != HNAE3_FUNC_RESET && - hdev->reset_type != HNAE3_FLR_RESET) { - hclge_mac_stop_phy(hdev); - hclge_update_link_status(hdev); - return; + if (test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state)) { + hclge_pfc_pause_en_cfg(hdev, HCLGE_PFC_TX_RX_DISABLE, + HCLGE_PFC_DISABLE); + if (hdev->reset_type != HNAE3_FUNC_RESET && + hdev->reset_type != HNAE3_FLR_RESET) { + hclge_mac_stop_phy(hdev); + hclge_update_link_status(hdev); + return; + } } hclge_reset_tqp(handle); diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c index 4a33f65190e2b431ea201c42ce81b86ed5c99d54..922c0da3660c7b9ff466828c1896676da7206ca2 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c @@ -171,8 +171,8 @@ int hclge_mac_pause_en_cfg(struct hclge_dev *hdev, bool tx, bool rx) return hclge_cmd_send(&hdev->hw, &desc, 1); } -static int hclge_pfc_pause_en_cfg(struct hclge_dev *hdev, u8 tx_rx_bitmap, - u8 pfc_bitmap) +int hclge_pfc_pause_en_cfg(struct hclge_dev *hdev, u8 tx_rx_bitmap, + u8 pfc_bitmap) { struct hclge_desc desc; struct hclge_pfc_en_cmd *pfc = (struct hclge_pfc_en_cmd *)desc.data; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.h index 68f28a98e380bc529b1808e39d74a679decbbc43..dd6f1fd486cf24d326c808e4d902732ca9329cf7 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.h @@ -164,6 +164,9 @@ struct hclge_bp_to_qs_map_cmd { u32 rsvd1; }; +#define HCLGE_PFC_DISABLE 0 +#define HCLGE_PFC_TX_RX_DISABLE 0 + struct hclge_pfc_en_cmd { u8 tx_rx_en_bitmap; u8 pri_en_bitmap; @@ -235,6 +238,8 @@ void hclge_tm_schd_info_update(struct hclge_dev *hdev, u8 num_tc); void hclge_tm_pfc_info_update(struct hclge_dev *hdev); int hclge_tm_dwrr_cfg(struct hclge_dev *hdev); int hclge_tm_init_hw(struct hclge_dev *hdev, bool init); +int hclge_pfc_pause_en_cfg(struct hclge_dev *hdev, u8 tx_rx_bitmap, + u8 pfc_bitmap); int hclge_mac_pause_en_cfg(struct hclge_dev *hdev, bool tx, bool rx); int hclge_pause_addr_cfg(struct hclge_dev *hdev, const u8 *mac_addr); void hclge_pfc_rx_stats_get(struct hclge_dev *hdev, u64 *stats); diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c index f240462503419035c7b2e733434b66969372adee..dd08989a4c7c1d122955c8b2839f07b32ff2283d 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c @@ -1436,7 +1436,10 @@ static int hclgevf_reset_wait(struct hclgevf_dev *hdev) * might happen in case reset assertion was made by PF. Yes, this also * means we might end up waiting bit more even for VF reset. */ - msleep(5000); + if (hdev->reset_type == HNAE3_VF_FULL_RESET) + msleep(5000); + else + msleep(500); return 0; } diff --git a/drivers/net/ethernet/intel/iavf/iavf.h b/drivers/net/ethernet/intel/iavf/iavf.h index 9abaff1f2affbc456545a2ccdeb2e7b2da6cf062..39d0fe76a38ff80e7ce895a48473d0f676dc6a35 100644 --- a/drivers/net/ethernet/intel/iavf/iavf.h +++ b/drivers/net/ethernet/intel/iavf/iavf.h @@ -525,7 +525,7 @@ void iavf_set_ethtool_ops(struct net_device *netdev); void iavf_update_stats(struct iavf_adapter *adapter); void iavf_reset_interrupt_capability(struct iavf_adapter *adapter); int iavf_init_interrupt_scheme(struct iavf_adapter *adapter); -void iavf_irq_enable_queues(struct iavf_adapter *adapter, u32 mask); +void iavf_irq_enable_queues(struct iavf_adapter *adapter); void iavf_free_all_tx_resources(struct iavf_adapter *adapter); void iavf_free_all_rx_resources(struct iavf_adapter *adapter); diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c index 2de4baff4c20501baab55b7ffddb84b6f8e20aee..4a66873882d12a70b6d79eedd40684d3f5a77b11 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_main.c +++ b/drivers/net/ethernet/intel/iavf/iavf_main.c @@ -359,21 +359,18 @@ static void iavf_irq_disable(struct iavf_adapter *adapter) } /** - * iavf_irq_enable_queues - Enable interrupt for specified queues + * iavf_irq_enable_queues - Enable interrupt for all queues * @adapter: board private structure - * @mask: bitmap of queues to enable **/ -void iavf_irq_enable_queues(struct iavf_adapter *adapter, u32 mask) +void iavf_irq_enable_queues(struct iavf_adapter *adapter) { struct iavf_hw *hw = &adapter->hw; int i; for (i = 1; i < adapter->num_msix_vectors; i++) { - if (mask & BIT(i - 1)) { - wr32(hw, IAVF_VFINT_DYN_CTLN1(i - 1), - IAVF_VFINT_DYN_CTLN1_INTENA_MASK | - IAVF_VFINT_DYN_CTLN1_ITR_INDX_MASK); - } + wr32(hw, IAVF_VFINT_DYN_CTLN1(i - 1), + IAVF_VFINT_DYN_CTLN1_INTENA_MASK | + IAVF_VFINT_DYN_CTLN1_ITR_INDX_MASK); } } @@ -387,7 +384,7 @@ void iavf_irq_enable(struct iavf_adapter *adapter, bool flush) struct iavf_hw *hw = &adapter->hw; iavf_misc_irq_enable(adapter); - iavf_irq_enable_queues(adapter, ~0); + iavf_irq_enable_queues(adapter); if (flush) iavf_flush(hw); diff --git a/drivers/net/ethernet/intel/iavf/iavf_register.h b/drivers/net/ethernet/intel/iavf/iavf_register.h index bf793332fc9d562cc09772f665a470865ce97552..a19e88898a0bb0a9d1755abaa4ff6439b9c4a373 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_register.h +++ b/drivers/net/ethernet/intel/iavf/iavf_register.h @@ -40,7 +40,7 @@ #define IAVF_VFINT_DYN_CTL01_INTENA_MASK IAVF_MASK(0x1, IAVF_VFINT_DYN_CTL01_INTENA_SHIFT) #define IAVF_VFINT_DYN_CTL01_ITR_INDX_SHIFT 3 #define IAVF_VFINT_DYN_CTL01_ITR_INDX_MASK IAVF_MASK(0x3, IAVF_VFINT_DYN_CTL01_ITR_INDX_SHIFT) -#define IAVF_VFINT_DYN_CTLN1(_INTVF) (0x00003800 + ((_INTVF) * 4)) /* _i=0...15 */ /* Reset: VFR */ +#define IAVF_VFINT_DYN_CTLN1(_INTVF) (0x00003800 + ((_INTVF) * 4)) /* _i=0...63 */ /* Reset: VFR */ #define IAVF_VFINT_DYN_CTLN1_INTENA_SHIFT 0 #define IAVF_VFINT_DYN_CTLN1_INTENA_MASK IAVF_MASK(0x1, IAVF_VFINT_DYN_CTLN1_INTENA_SHIFT) #define IAVF_VFINT_DYN_CTLN1_SWINT_TRIG_SHIFT 2 diff --git a/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c b/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c index 9afbbdac35903f1ae018c00fe28213ece3227d3f..7c0578b5457b96bd9d4b4f53a19f64cb4ca8560b 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c +++ b/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c @@ -2238,11 +2238,6 @@ void iavf_virtchnl_completion(struct iavf_adapter *adapter, iavf_process_config(adapter); adapter->flags |= IAVF_FLAG_SETUP_NETDEV_FEATURES; - /* Request VLAN offload settings */ - if (VLAN_V2_ALLOWED(adapter)) - iavf_set_vlan_offload_features(adapter, 0, - netdev->features); - iavf_set_queue_vlan_tag_loc(adapter); was_mac_changed = !ether_addr_equal(netdev->dev_addr, diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c index 0157f6e98d3e24e3e77224e45a92d8698e3fc84a..eb2dc09837769ce33a7d5d1efdfae4a23fcf7a0c 100644 --- a/drivers/net/ethernet/intel/ice/ice_common.c +++ b/drivers/net/ethernet/intel/ice/ice_common.c @@ -5160,7 +5160,7 @@ ice_aq_read_i2c(struct ice_hw *hw, struct ice_aqc_link_topo_addr topo_addr, */ int ice_aq_write_i2c(struct ice_hw *hw, struct ice_aqc_link_topo_addr topo_addr, - u16 bus_addr, __le16 addr, u8 params, u8 *data, + u16 bus_addr, __le16 addr, u8 params, const u8 *data, struct ice_sq_cd *cd) { struct ice_aq_desc desc = { 0 }; diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h index 8ba5f935a092bf1525137c0467ae63b4c8ac16bd..81961a7d65985cae2068ed366512040ebc501303 100644 --- a/drivers/net/ethernet/intel/ice/ice_common.h +++ b/drivers/net/ethernet/intel/ice/ice_common.h @@ -229,7 +229,7 @@ ice_aq_read_i2c(struct ice_hw *hw, struct ice_aqc_link_topo_addr topo_addr, struct ice_sq_cd *cd); int ice_aq_write_i2c(struct ice_hw *hw, struct ice_aqc_link_topo_addr topo_addr, - u16 bus_addr, __le16 addr, u8 params, u8 *data, + u16 bus_addr, __le16 addr, u8 params, const u8 *data, struct ice_sq_cd *cd); bool ice_fw_supports_report_dflt_cfg(struct ice_hw *hw); #endif /* _ICE_COMMON_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_dcb_lib.c b/drivers/net/ethernet/intel/ice/ice_dcb_lib.c index c6d4926f0fcf50df80bec065e8b150ffe45b0831..850db8e0e6b0021ab0f70718634c4dfb496f0928 100644 --- a/drivers/net/ethernet/intel/ice/ice_dcb_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_dcb_lib.c @@ -932,10 +932,9 @@ ice_tx_prepare_vlan_flags_dcb(struct ice_tx_ring *tx_ring, if ((first->tx_flags & ICE_TX_FLAGS_HW_VLAN || first->tx_flags & ICE_TX_FLAGS_HW_OUTER_SINGLE_VLAN) || skb->priority != TC_PRIO_CONTROL) { - first->tx_flags &= ~ICE_TX_FLAGS_VLAN_PR_M; + first->vid &= ~VLAN_PRIO_MASK; /* Mask the lower 3 bits to set the 802.1p priority */ - first->tx_flags |= (skb->priority & 0x7) << - ICE_TX_FLAGS_VLAN_PR_S; + first->vid |= (skb->priority << VLAN_PRIO_SHIFT) & VLAN_PRIO_MASK; /* if this is not already set it means a VLAN 0 + priority needs * to be offloaded */ diff --git a/drivers/net/ethernet/intel/ice/ice_gnss.c b/drivers/net/ethernet/intel/ice/ice_gnss.c index 2ea8a2b11bcda7d905d3f06c12c4eab78c698043..75c9de675f2028611c2d91d1b2ed055401dbeadd 100644 --- a/drivers/net/ethernet/intel/ice/ice_gnss.c +++ b/drivers/net/ethernet/intel/ice/ice_gnss.c @@ -16,8 +16,8 @@ * * number of bytes written - success * * negative - error code */ -static unsigned int -ice_gnss_do_write(struct ice_pf *pf, unsigned char *buf, unsigned int size) +static int +ice_gnss_do_write(struct ice_pf *pf, const unsigned char *buf, unsigned int size) { struct ice_aqc_link_topo_addr link_topo; struct ice_hw *hw = &pf->hw; @@ -72,39 +72,7 @@ ice_gnss_do_write(struct ice_pf *pf, unsigned char *buf, unsigned int size) dev_err(ice_pf_to_dev(pf), "GNSS failed to write, offset=%u, size=%u, err=%d\n", offset, size, err); - return offset; -} - -/** - * ice_gnss_write_pending - Write all pending data to internal GNSS - * @work: GNSS write work structure - */ -static void ice_gnss_write_pending(struct kthread_work *work) -{ - struct gnss_serial *gnss = container_of(work, struct gnss_serial, - write_work); - struct ice_pf *pf = gnss->back; - - if (!pf) - return; - - if (!test_bit(ICE_FLAG_GNSS, pf->flags)) - return; - - if (!list_empty(&gnss->queue)) { - struct gnss_write_buf *write_buf = NULL; - unsigned int bytes; - - write_buf = list_first_entry(&gnss->queue, - struct gnss_write_buf, queue); - - bytes = ice_gnss_do_write(pf, write_buf->buf, write_buf->size); - dev_dbg(ice_pf_to_dev(pf), "%u bytes written to GNSS\n", bytes); - - list_del(&write_buf->queue); - kfree(write_buf->buf); - kfree(write_buf); - } + return err; } /** @@ -128,12 +96,7 @@ static void ice_gnss_read(struct kthread_work *work) int err = 0; pf = gnss->back; - if (!pf) { - err = -EFAULT; - goto exit; - } - - if (!test_bit(ICE_FLAG_GNSS, pf->flags)) + if (!pf || !test_bit(ICE_FLAG_GNSS, pf->flags)) return; hw = &pf->hw; @@ -191,7 +154,6 @@ static void ice_gnss_read(struct kthread_work *work) free_page((unsigned long)buf); requeue: kthread_queue_delayed_work(gnss->kworker, &gnss->read_work, delay); -exit: if (err) dev_dbg(ice_pf_to_dev(pf), "GNSS failed to read err=%d\n", err); } @@ -220,8 +182,6 @@ static struct gnss_serial *ice_gnss_struct_init(struct ice_pf *pf) pf->gnss_serial = gnss; kthread_init_delayed_work(&gnss->read_work, ice_gnss_read); - INIT_LIST_HEAD(&gnss->queue); - kthread_init_work(&gnss->write_work, ice_gnss_write_pending); kworker = kthread_create_worker(0, "ice-gnss-%s", dev_name(dev)); if (IS_ERR(kworker)) { kfree(gnss); @@ -281,7 +241,6 @@ static void ice_gnss_close(struct gnss_device *gdev) if (!gnss) return; - kthread_cancel_work_sync(&gnss->write_work); kthread_cancel_delayed_work_sync(&gnss->read_work); } @@ -300,10 +259,7 @@ ice_gnss_write(struct gnss_device *gdev, const unsigned char *buf, size_t count) { struct ice_pf *pf = gnss_get_drvdata(gdev); - struct gnss_write_buf *write_buf; struct gnss_serial *gnss; - unsigned char *cmd_buf; - int err = count; /* We cannot write a single byte using our I2C implementation. */ if (count <= 1 || count > ICE_GNSS_TTY_WRITE_BUF) @@ -319,24 +275,7 @@ ice_gnss_write(struct gnss_device *gdev, const unsigned char *buf, if (!gnss) return -ENODEV; - cmd_buf = kcalloc(count, sizeof(*buf), GFP_KERNEL); - if (!cmd_buf) - return -ENOMEM; - - memcpy(cmd_buf, buf, count); - write_buf = kzalloc(sizeof(*write_buf), GFP_KERNEL); - if (!write_buf) { - kfree(cmd_buf); - return -ENOMEM; - } - - write_buf->buf = cmd_buf; - write_buf->size = count; - INIT_LIST_HEAD(&write_buf->queue); - list_add_tail(&write_buf->queue, &gnss->queue); - kthread_queue_work(gnss->kworker, &gnss->write_work); - - return err; + return ice_gnss_do_write(pf, buf, count); } static const struct gnss_operations ice_gnss_ops = { @@ -432,7 +371,6 @@ void ice_gnss_exit(struct ice_pf *pf) if (pf->gnss_serial) { struct gnss_serial *gnss = pf->gnss_serial; - kthread_cancel_work_sync(&gnss->write_work); kthread_cancel_delayed_work_sync(&gnss->read_work); kthread_destroy_worker(gnss->kworker); gnss->kworker = NULL; diff --git a/drivers/net/ethernet/intel/ice/ice_gnss.h b/drivers/net/ethernet/intel/ice/ice_gnss.h index b8bb8b63d081723b13719e75fb974df2e865c143..75e567ad70594528c9cb3cb2aa966b9c7353ad7c 100644 --- a/drivers/net/ethernet/intel/ice/ice_gnss.h +++ b/drivers/net/ethernet/intel/ice/ice_gnss.h @@ -22,26 +22,16 @@ */ #define ICE_GNSS_UBX_WRITE_BYTES (ICE_MAX_I2C_WRITE_BYTES + 1) -struct gnss_write_buf { - struct list_head queue; - unsigned int size; - unsigned char *buf; -}; - /** * struct gnss_serial - data used to initialize GNSS TTY port * @back: back pointer to PF * @kworker: kwork thread for handling periodic work * @read_work: read_work function for handling GNSS reads - * @write_work: write_work function for handling GNSS writes - * @queue: write buffers queue */ struct gnss_serial { struct ice_pf *back; struct kthread_worker *kworker; struct kthread_delayed_work read_work; - struct kthread_work write_work; - struct list_head queue; }; #if IS_ENABLED(CONFIG_GNSS) diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c index 450317dfcca7360cb1acdfca470bd4295d11e3db..11ae0e41f518a13b1c77c56a15404338495b746a 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_lib.c @@ -2745,6 +2745,8 @@ ice_vsi_cfg_def(struct ice_vsi *vsi, struct ice_vsi_cfg_params *params) goto unroll_vector_base; ice_vsi_map_rings_to_vectors(vsi); + vsi->stat_offsets_loaded = false; + if (ice_is_xdp_ena_vsi(vsi)) { ret = ice_vsi_determine_xdp_res(vsi); if (ret) @@ -2793,6 +2795,9 @@ ice_vsi_cfg_def(struct ice_vsi *vsi, struct ice_vsi_cfg_params *params) ret = ice_vsi_alloc_ring_stats(vsi); if (ret) goto unroll_vector_base; + + vsi->stat_offsets_loaded = false; + /* Do not exit if configuring RSS had an issue, at least * receive traffic on first queue. Hence no need to capture * return value diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index a1f7c8edc22f341eb68bb30a507981c7c4baaa2d..42c318ceff61837495c33fa4e19ecc3db0e9bf63 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -4802,9 +4802,13 @@ static int ice_init_dev(struct ice_pf *pf) static void ice_deinit_dev(struct ice_pf *pf) { ice_free_irq_msix_misc(pf); - ice_clear_interrupt_scheme(pf); ice_deinit_pf(pf); ice_deinit_hw(&pf->hw); + + /* Service task is already stopped, so call reset directly. */ + ice_reset(&pf->hw, ICE_RESET_PFR); + pci_wait_for_pending_transaction(pf->pdev); + ice_clear_interrupt_scheme(pf); } static void ice_init_features(struct ice_pf *pf) @@ -5094,10 +5098,6 @@ int ice_load(struct ice_pf *pf) struct ice_vsi *vsi; int err; - err = ice_reset(&pf->hw, ICE_RESET_PFR); - if (err) - return err; - err = ice_init_dev(pf); if (err) return err; @@ -5354,12 +5354,6 @@ static void ice_remove(struct pci_dev *pdev) ice_setup_mc_magic_wake(pf); ice_set_wake(pf); - /* Issue a PFR as part of the prescribed driver unload flow. Do not - * do it via ice_schedule_reset() since there is no need to rebuild - * and the service task is already stopped. - */ - ice_reset(&pf->hw, ICE_RESET_PFR); - pci_wait_for_pending_transaction(pdev); pci_disable_device(pdev); } @@ -7056,6 +7050,10 @@ int ice_down(struct ice_vsi *vsi) ice_for_each_txq(vsi, i) ice_clean_tx_ring(vsi->tx_rings[i]); + if (ice_is_xdp_ena_vsi(vsi)) + ice_for_each_xdp_txq(vsi, i) + ice_clean_tx_ring(vsi->xdp_rings[i]); + ice_for_each_rxq(vsi, i) ice_clean_rx_ring(vsi->rx_rings[i]); diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.c b/drivers/net/ethernet/intel/ice/ice_sriov.c index f1dca59bd8449ea91706451c3226fe4ff912cc40..588ad8696756defb6154022fe4fc9b7c75eed366 100644 --- a/drivers/net/ethernet/intel/ice/ice_sriov.c +++ b/drivers/net/ethernet/intel/ice/ice_sriov.c @@ -1171,7 +1171,7 @@ int ice_set_vf_spoofchk(struct net_device *netdev, int vf_id, bool ena) if (!vf) return -EINVAL; - ret = ice_check_vf_ready_for_cfg(vf); + ret = ice_check_vf_ready_for_reset(vf); if (ret) goto out_put_vf; @@ -1286,7 +1286,7 @@ int ice_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac) goto out_put_vf; } - ret = ice_check_vf_ready_for_cfg(vf); + ret = ice_check_vf_ready_for_reset(vf); if (ret) goto out_put_vf; @@ -1340,7 +1340,7 @@ int ice_set_vf_trust(struct net_device *netdev, int vf_id, bool trusted) return -EOPNOTSUPP; } - ret = ice_check_vf_ready_for_cfg(vf); + ret = ice_check_vf_ready_for_reset(vf); if (ret) goto out_put_vf; @@ -1653,7 +1653,7 @@ ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos, if (!vf) return -EINVAL; - ret = ice_check_vf_ready_for_cfg(vf); + ret = ice_check_vf_ready_for_reset(vf); if (ret) goto out_put_vf; diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index 4fcf2d07eb853b10be4b5b483b33e0f95db9446a..52d0a126eb6161852bb96457b9025a35e5641578 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -1152,11 +1152,11 @@ int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) unsigned int total_rx_bytes = 0, total_rx_pkts = 0; unsigned int offset = rx_ring->rx_offset; struct xdp_buff *xdp = &rx_ring->xdp; + u32 cached_ntc = rx_ring->first_desc; struct ice_tx_ring *xdp_ring = NULL; struct bpf_prog *xdp_prog = NULL; u32 ntc = rx_ring->next_to_clean; u32 cnt = rx_ring->count; - u32 cached_ntc = ntc; u32 xdp_xmit = 0; u32 cached_ntu; bool failure; @@ -1664,8 +1664,7 @@ ice_tx_map(struct ice_tx_ring *tx_ring, struct ice_tx_buf *first, if (first->tx_flags & ICE_TX_FLAGS_HW_VLAN) { td_cmd |= (u64)ICE_TX_DESC_CMD_IL2TAG1; - td_tag = (first->tx_flags & ICE_TX_FLAGS_VLAN_M) >> - ICE_TX_FLAGS_VLAN_S; + td_tag = first->vid; } dma = dma_map_single(tx_ring->dev, skb->data, size, DMA_TO_DEVICE); @@ -1998,7 +1997,7 @@ ice_tx_prepare_vlan_flags(struct ice_tx_ring *tx_ring, struct ice_tx_buf *first) * VLAN offloads exclusively so we only care about the VLAN ID here */ if (skb_vlan_tag_present(skb)) { - first->tx_flags |= skb_vlan_tag_get(skb) << ICE_TX_FLAGS_VLAN_S; + first->vid = skb_vlan_tag_get(skb); if (tx_ring->flags & ICE_TX_FLAGS_RING_VLAN_L2TAG2) first->tx_flags |= ICE_TX_FLAGS_HW_OUTER_SINGLE_VLAN; else @@ -2388,8 +2387,7 @@ ice_xmit_frame_ring(struct sk_buff *skb, struct ice_tx_ring *tx_ring) offload.cd_qw1 |= (u64)(ICE_TX_DESC_DTYPE_CTX | (ICE_TX_CTX_DESC_IL2TAG2 << ICE_TXD_CTX_QW1_CMD_S)); - offload.cd_l2tag2 = (first->tx_flags & ICE_TX_FLAGS_VLAN_M) >> - ICE_TX_FLAGS_VLAN_S; + offload.cd_l2tag2 = first->vid; } /* set up TSO offload */ diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h index fff0efe28373affc555994c335f3c412d8844cdb..166413fc33f48f71a459009819a75d000601b03f 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.h +++ b/drivers/net/ethernet/intel/ice/ice_txrx.h @@ -127,10 +127,6 @@ static inline int ice_skb_pad(void) #define ICE_TX_FLAGS_IPV6 BIT(6) #define ICE_TX_FLAGS_TUNNEL BIT(7) #define ICE_TX_FLAGS_HW_OUTER_SINGLE_VLAN BIT(8) -#define ICE_TX_FLAGS_VLAN_M 0xffff0000 -#define ICE_TX_FLAGS_VLAN_PR_M 0xe0000000 -#define ICE_TX_FLAGS_VLAN_PR_S 29 -#define ICE_TX_FLAGS_VLAN_S 16 #define ICE_XDP_PASS 0 #define ICE_XDP_CONSUMED BIT(0) @@ -182,8 +178,9 @@ struct ice_tx_buf { unsigned int gso_segs; unsigned int nr_frags; /* used for mbuf XDP */ }; - u32 type:16; /* &ice_tx_buf_type */ - u32 tx_flags:16; + u32 tx_flags:12; + u32 type:4; /* &ice_tx_buf_type */ + u32 vid:16; DEFINE_DMA_UNMAP_LEN(len); DEFINE_DMA_UNMAP_ADDR(dma); }; diff --git a/drivers/net/ethernet/intel/ice/ice_vf_lib.c b/drivers/net/ethernet/intel/ice/ice_vf_lib.c index 89fd6982df0934b53b27a538856bdc7b3c30c75b..bf74a2f3a4f8c3d8ebf5a1492782ec06c79b3ddd 100644 --- a/drivers/net/ethernet/intel/ice/ice_vf_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_vf_lib.c @@ -185,6 +185,25 @@ int ice_check_vf_ready_for_cfg(struct ice_vf *vf) return 0; } +/** + * ice_check_vf_ready_for_reset - check if VF is ready to be reset + * @vf: VF to check if it's ready to be reset + * + * The purpose of this function is to ensure that the VF is not in reset, + * disabled, and is both initialized and active, thus enabling us to safely + * initialize another reset. + */ +int ice_check_vf_ready_for_reset(struct ice_vf *vf) +{ + int ret; + + ret = ice_check_vf_ready_for_cfg(vf); + if (!ret && !test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) + ret = -EAGAIN; + + return ret; +} + /** * ice_trigger_vf_reset - Reset a VF on HW * @vf: pointer to the VF structure diff --git a/drivers/net/ethernet/intel/ice/ice_vf_lib.h b/drivers/net/ethernet/intel/ice/ice_vf_lib.h index e3cda6fb71ab1e61221557abb0692ad9c1b19d14..a38ef00a367942b52f45c8b9d87abe4d95f52fef 100644 --- a/drivers/net/ethernet/intel/ice/ice_vf_lib.h +++ b/drivers/net/ethernet/intel/ice/ice_vf_lib.h @@ -215,6 +215,7 @@ u16 ice_get_num_vfs(struct ice_pf *pf); struct ice_vsi *ice_get_vf_vsi(struct ice_vf *vf); bool ice_is_vf_disabled(struct ice_vf *vf); int ice_check_vf_ready_for_cfg(struct ice_vf *vf); +int ice_check_vf_ready_for_reset(struct ice_vf *vf); void ice_set_vf_state_dis(struct ice_vf *vf); bool ice_is_any_vf_in_unicast_promisc(struct ice_pf *pf); void diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl.c b/drivers/net/ethernet/intel/ice/ice_virtchnl.c index 97243c616d5d6d939196fcca9b4ea04724dbbf0a..f4a524f80b110f1dc3d60d706eedb31a20740000 100644 --- a/drivers/net/ethernet/intel/ice/ice_virtchnl.c +++ b/drivers/net/ethernet/intel/ice/ice_virtchnl.c @@ -3955,6 +3955,7 @@ void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event, ice_vc_notify_vf_link_state(vf); break; case VIRTCHNL_OP_RESET_VF: + clear_bit(ICE_VF_STATE_ACTIVE, vf->vf_states); ops->reset_vf(vf); break; case VIRTCHNL_OP_ADD_ETH_ADDR: diff --git a/drivers/net/ethernet/intel/igb/e1000_mac.c b/drivers/net/ethernet/intel/igb/e1000_mac.c index 205d577bdbbaad57af46694858f6d4f28617e3a7..caf91c6f52b4d03cabef2cc7fe6a4d545a0cc899 100644 --- a/drivers/net/ethernet/intel/igb/e1000_mac.c +++ b/drivers/net/ethernet/intel/igb/e1000_mac.c @@ -426,7 +426,7 @@ void igb_mta_set(struct e1000_hw *hw, u32 hash_value) static u32 igb_hash_mc_addr(struct e1000_hw *hw, u8 *mc_addr) { u32 hash_value, hash_mask; - u8 bit_shift = 0; + u8 bit_shift = 1; /* Register count multiplied by bits per register */ hash_mask = (hw->mac.mta_reg_count * 32) - 1; @@ -434,7 +434,7 @@ static u32 igb_hash_mc_addr(struct e1000_hw *hw, u8 *mc_addr) /* For a mc_filter_type of 0, bit_shift is the number of left-shifts * where 0xFF would still fall within the hash mask. */ - while (hash_mask >> bit_shift != 0xFF) + while (hash_mask >> bit_shift != 0xFF && bit_shift < 4) bit_shift++; /* The portion of the address that is used for the hash table diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c b/drivers/net/ethernet/intel/igb/igb_ethtool.c index 7d60da1b7bf4182ffebdcbd3ff2c4dbe7df47d6b..319ed601eaa1e9827e7fab48c6d13b01e2bedff1 100644 --- a/drivers/net/ethernet/intel/igb/igb_ethtool.c +++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c @@ -822,6 +822,8 @@ static int igb_set_eeprom(struct net_device *netdev, */ ret_val = hw->nvm.ops.read(hw, last_word, 1, &eeprom_buff[last_word - first_word]); + if (ret_val) + goto out; } /* Device's eeprom is always little-endian, word addressable */ @@ -841,6 +843,7 @@ static int igb_set_eeprom(struct net_device *netdev, hw->nvm.ops.update(hw); igb_set_fw_version(adapter); +out: kfree(eeprom_buff); return ret_val; } diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c index 58872a4c254015f527093076a9124b615d63243e..bb3db387d49cf6de899fec0acbcb1882a5f2452e 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -6947,6 +6947,7 @@ static void igb_extts(struct igb_adapter *adapter, int tsintr_tt) struct e1000_hw *hw = &adapter->hw; struct ptp_clock_event event; struct timespec64 ts; + unsigned long flags; if (pin < 0 || pin >= IGB_N_SDP) return; @@ -6954,9 +6955,12 @@ static void igb_extts(struct igb_adapter *adapter, int tsintr_tt) if (hw->mac.type == e1000_82580 || hw->mac.type == e1000_i354 || hw->mac.type == e1000_i350) { - s64 ns = rd32(auxstmpl); + u64 ns = rd32(auxstmpl); - ns += ((s64)(rd32(auxstmph) & 0xFF)) << 32; + ns += ((u64)(rd32(auxstmph) & 0xFF)) << 32; + spin_lock_irqsave(&adapter->tmreg_lock, flags); + ns = timecounter_cyc2time(&adapter->tc, ns); + spin_unlock_irqrestore(&adapter->tmreg_lock, flags); ts = ns_to_timespec64(ns); } else { ts.tv_nsec = rd32(auxstmpl); diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c index 1c46768820825ff20bb401b6c4a519fabbcdd429..fa764190f27050a28dc0778e426eec18467c9389 100644 --- a/drivers/net/ethernet/intel/igc/igc_main.c +++ b/drivers/net/ethernet/intel/igc/igc_main.c @@ -254,6 +254,13 @@ static void igc_clean_tx_ring(struct igc_ring *tx_ring) /* reset BQL for queue */ netdev_tx_reset_queue(txring_txq(tx_ring)); + /* Zero out the buffer ring */ + memset(tx_ring->tx_buffer_info, 0, + sizeof(*tx_ring->tx_buffer_info) * tx_ring->count); + + /* Zero out the descriptor ring */ + memset(tx_ring->desc, 0, tx_ring->size); + /* reset next_to_use and next_to_clean */ tx_ring->next_to_use = 0; tx_ring->next_to_clean = 0; @@ -267,7 +274,7 @@ static void igc_clean_tx_ring(struct igc_ring *tx_ring) */ void igc_free_tx_resources(struct igc_ring *tx_ring) { - igc_clean_tx_ring(tx_ring); + igc_disable_tx_ring(tx_ring); vfree(tx_ring->tx_buffer_info); tx_ring->tx_buffer_info = NULL; @@ -6723,6 +6730,9 @@ static void igc_remove(struct pci_dev *pdev) igc_ptp_stop(adapter); + pci_disable_ptm(pdev); + pci_clear_master(pdev); + set_bit(__IGC_DOWN, &adapter->state); del_timer_sync(&adapter->watchdog_timer); diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 5d83c887a3fceb9f131abfc7c41798a982772fd2..1726297f2e0df09e92e8e5b327162b8a87571040 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -1256,7 +1256,7 @@ static bool ixgbe_clean_tx_irq(struct ixgbe_q_vector *q_vector, if (!__netif_txq_completed_wake(txq, total_packets, total_bytes, ixgbe_desc_unused(tx_ring), TX_WAKE_THRESHOLD, - netif_carrier_ok(tx_ring->netdev) && + !netif_carrier_ok(tx_ring->netdev) || test_bit(__IXGBE_DOWN, &adapter->state))) ++tx_ring->tx_stats.restart_queue; diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c index e1853da280f961b3f14099fa2ff51bc4b2fb7312..43eb6e871351101f5f917ec90105f1202b21f50e 100644 --- a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c +++ b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c @@ -981,6 +981,9 @@ int octep_device_setup(struct octep_device *oct) oct->mmio[i].hw_addr = ioremap(pci_resource_start(oct->pdev, i * 2), pci_resource_len(oct->pdev, i * 2)); + if (!oct->mmio[i].hw_addr) + goto unmap_prev; + oct->mmio[i].mapped = 1; } @@ -1015,7 +1018,9 @@ int octep_device_setup(struct octep_device *oct) return 0; unsupported_dev: - for (i = 0; i < OCTEP_MMIO_REGIONS; i++) + i = OCTEP_MMIO_REGIONS; +unmap_prev: + while (i--) iounmap(oct->mmio[i].hw_addr); kfree(oct->conf); diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c index 4ad707e758b9f700cd675f0f093b59f5b088779d..f01d057ad025ae7f2aca0d06c16238bc2d005227 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c @@ -1878,7 +1878,8 @@ static int nix_check_txschq_alloc_req(struct rvu *rvu, int lvl, u16 pcifunc, free_cnt = rvu_rsrc_free_count(&txsch->schq); } - if (free_cnt < req_schq || req_schq > MAX_TXSCHQ_PER_FUNC) + if (free_cnt < req_schq || req->schq[lvl] > MAX_TXSCHQ_PER_FUNC || + req->schq_contig[lvl] > MAX_TXSCHQ_PER_FUNC) return NIX_AF_ERR_TLX_ALLOC_FAIL; /* If contiguous queues are needed, check for availability */ @@ -4080,10 +4081,6 @@ int rvu_mbox_handler_nix_set_rx_cfg(struct rvu *rvu, struct nix_rx_cfg *req, static u64 rvu_get_lbk_link_credits(struct rvu *rvu, u16 lbk_max_frs) { - /* CN10k supports 72KB FIFO size and max packet size of 64k */ - if (rvu->hw->lbk_bufsize == 0x12000) - return (rvu->hw->lbk_bufsize - lbk_max_frs) / 16; - return 1600; /* 16 * max LBK datarate = 16 * 100Gbps */ } diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_hash.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_hash.c index 51209119f0f2faf431ab46df5ee71e706472c9e0..9f11c1e4073737f3f8d187da7f29497d925e2b8b 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_hash.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_hash.c @@ -1164,10 +1164,8 @@ static u16 __rvu_npc_exact_cmd_rules_cnt_update(struct rvu *rvu, int drop_mcam_i { struct npc_exact_table *table; u16 *cnt, old_cnt; - bool promisc; table = rvu->hw->table; - promisc = table->promisc_mode[drop_mcam_idx]; cnt = &table->cnt_cmd_rules[drop_mcam_idx]; old_cnt = *cnt; @@ -1179,16 +1177,13 @@ static u16 __rvu_npc_exact_cmd_rules_cnt_update(struct rvu *rvu, int drop_mcam_i *enable_or_disable_cam = false; - if (promisc) - goto done; - - /* If all rules are deleted and not already in promisc mode; disable cam */ + /* If all rules are deleted, disable cam */ if (!*cnt && val < 0) { *enable_or_disable_cam = true; goto done; } - /* If rule got added and not already in promisc mode; enable cam */ + /* If rule got added, enable cam */ if (!old_cnt && val > 0) { *enable_or_disable_cam = true; goto done; @@ -1443,7 +1438,6 @@ int rvu_npc_exact_promisc_disable(struct rvu *rvu, u16 pcifunc) u32 drop_mcam_idx; bool *promisc; bool rc; - u32 cnt; table = rvu->hw->table; @@ -1466,17 +1460,8 @@ int rvu_npc_exact_promisc_disable(struct rvu *rvu, u16 pcifunc) return LMAC_AF_ERR_INVALID_PARAM; } *promisc = false; - cnt = __rvu_npc_exact_cmd_rules_cnt_update(rvu, drop_mcam_idx, 0, NULL); mutex_unlock(&table->lock); - /* If no dmac filter entries configured, disable drop rule */ - if (!cnt) - rvu_npc_enable_mcam_by_entry_index(rvu, drop_mcam_idx, NIX_INTF_RX, false); - else - rvu_npc_enable_mcam_by_entry_index(rvu, drop_mcam_idx, NIX_INTF_RX, !*promisc); - - dev_dbg(rvu->dev, "%s: disabled promisc mode (cgx=%d lmac=%d, cnt=%d)\n", - __func__, cgx_id, lmac_id, cnt); return 0; } @@ -1494,7 +1479,6 @@ int rvu_npc_exact_promisc_enable(struct rvu *rvu, u16 pcifunc) u32 drop_mcam_idx; bool *promisc; bool rc; - u32 cnt; table = rvu->hw->table; @@ -1517,17 +1501,8 @@ int rvu_npc_exact_promisc_enable(struct rvu *rvu, u16 pcifunc) return LMAC_AF_ERR_INVALID_PARAM; } *promisc = true; - cnt = __rvu_npc_exact_cmd_rules_cnt_update(rvu, drop_mcam_idx, 0, NULL); mutex_unlock(&table->lock); - /* If no dmac filter entries configured, disable drop rule */ - if (!cnt) - rvu_npc_enable_mcam_by_entry_index(rvu, drop_mcam_idx, NIX_INTF_RX, false); - else - rvu_npc_enable_mcam_by_entry_index(rvu, drop_mcam_idx, NIX_INTF_RX, !*promisc); - - dev_dbg(rvu->dev, "%s: Enabled promisc mode (cgx=%d lmac=%d cnt=%d)\n", - __func__, cgx_id, lmac_id, cnt); return 0; } diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c index 7045fedfd73a04ba63cdb2d657754722f5291de8..7af223b0a37f51e129c5d28d9d500262a4c0b620 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c @@ -652,9 +652,7 @@ static void otx2_sqe_add_ext(struct otx2_nic *pfvf, struct otx2_snd_queue *sq, htons(ext->lso_sb - skb_network_offset(skb)); } else if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6) { ext->lso_format = pfvf->hw.lso_tsov6_idx; - - ipv6_hdr(skb)->payload_len = - htons(ext->lso_sb - skb_network_offset(skb)); + ipv6_hdr(skb)->payload_len = htons(tcp_hdrlen(skb)); } else if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) { __be16 l3_proto = vlan_get_protocol(skb); struct udphdr *udph = udp_hdr(skb); diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index a75fd072082c6e415043398a8293323c246ac4a1..834c644b67db51b69724d4b8671f3bb65821e7b0 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -3269,18 +3269,14 @@ static int mtk_open(struct net_device *dev) eth->dsa_meta[i] = md_dst; } } else { - /* Hardware special tag parsing needs to be disabled if at least - * one MAC does not use DSA. + /* Hardware DSA untagging and VLAN RX offloading need to be + * disabled if at least one MAC does not use DSA. */ u32 val = mtk_r32(eth, MTK_CDMP_IG_CTRL); val &= ~MTK_CDMP_STAG_EN; mtk_w32(eth, val, MTK_CDMP_IG_CTRL); - val = mtk_r32(eth, MTK_CDMQ_IG_CTRL); - val &= ~MTK_CDMQ_STAG_EN; - mtk_w32(eth, val, MTK_CDMQ_IG_CTRL); - mtk_w32(eth, 0, MTK_CDMP_EG_CTRL); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index d53de39539a8fd90d455fbe83f87dba0225dde14..d532883b42d7c30bf2f1230097a1cd55d46e4cf4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -1920,9 +1920,10 @@ static void mlx5_cmd_err_trace(struct mlx5_core_dev *dev, u16 opcode, u16 op_mod static void cmd_status_log(struct mlx5_core_dev *dev, u16 opcode, u8 status, u32 syndrome, int err) { + const char *namep = mlx5_command_str(opcode); struct mlx5_cmd_stats *stats; - if (!err) + if (!err || !(strcmp(namep, "unknown command opcode"))) return; stats = &dev->cmd.stats[opcode]; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c index f40497823e65fada81f0d289e0b4c68842ffc940..7c0f2adbea000d2453add53348a48c49ccc75110 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c @@ -490,7 +490,7 @@ static void poll_trace(struct mlx5_fw_tracer *tracer, (u64)timestamp_low; break; default: - if (tracer_event->event_id >= tracer->str_db.first_string_trace || + if (tracer_event->event_id >= tracer->str_db.first_string_trace && tracer_event->event_id <= tracer->str_db.first_string_trace + tracer->str_db.num_string_trace) { tracer_event->type = TRACER_EVENT_TYPE_STRING; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index b8987a404d7526d2f6fe3cd48c0a5f94e22d9aee..8e999f238194812bb44e5b57e1f3b22cf08c67dd 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -327,6 +327,7 @@ struct mlx5e_params { unsigned int sw_mtu; int hard_mtu; bool ptp_rx; + __be32 terminate_lkey_be; }; static inline u8 mlx5e_get_dcb_num_tc(struct mlx5e_params *params) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c b/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c index 7ac1ad9c46de0352a0a323f815f8a0b28f2281d2..7e8e96cc5cd083e220e36334dd499220f2ed542a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c @@ -51,7 +51,7 @@ int mlx5e_port_query_buffer(struct mlx5e_priv *priv, if (err) goto out; - for (i = 0; i < MLX5E_MAX_BUFFER; i++) { + for (i = 0; i < MLX5E_MAX_NETWORK_BUFFER; i++) { buffer = MLX5_ADDR_OF(pbmc_reg, out, buffer[i]); port_buffer->buffer[i].lossy = MLX5_GET(bufferx_reg, buffer, lossy); @@ -73,14 +73,24 @@ int mlx5e_port_query_buffer(struct mlx5e_priv *priv, port_buffer->buffer[i].lossy); } - port_buffer->headroom_size = total_used; + port_buffer->internal_buffers_size = 0; + for (i = MLX5E_MAX_NETWORK_BUFFER; i < MLX5E_TOTAL_BUFFERS; i++) { + buffer = MLX5_ADDR_OF(pbmc_reg, out, buffer[i]); + port_buffer->internal_buffers_size += + MLX5_GET(bufferx_reg, buffer, size) * port_buff_cell_sz; + } + port_buffer->port_buffer_size = MLX5_GET(pbmc_reg, out, port_buffer_size) * port_buff_cell_sz; - port_buffer->spare_buffer_size = - port_buffer->port_buffer_size - total_used; - - mlx5e_dbg(HW, priv, "total buffer size=%d, spare buffer size=%d\n", - port_buffer->port_buffer_size, + port_buffer->headroom_size = total_used; + port_buffer->spare_buffer_size = port_buffer->port_buffer_size - + port_buffer->internal_buffers_size - + port_buffer->headroom_size; + + mlx5e_dbg(HW, priv, + "total buffer size=%u, headroom buffer size=%u, internal buffers size=%u, spare buffer size=%u\n", + port_buffer->port_buffer_size, port_buffer->headroom_size, + port_buffer->internal_buffers_size, port_buffer->spare_buffer_size); out: kfree(out); @@ -206,11 +216,11 @@ static int port_update_pool_cfg(struct mlx5_core_dev *mdev, if (!MLX5_CAP_GEN(mdev, sbcam_reg)) return 0; - for (i = 0; i < MLX5E_MAX_BUFFER; i++) + for (i = 0; i < MLX5E_MAX_NETWORK_BUFFER; i++) lossless_buff_count += ((port_buffer->buffer[i].size) && (!(port_buffer->buffer[i].lossy))); - for (i = 0; i < MLX5E_MAX_BUFFER; i++) { + for (i = 0; i < MLX5E_MAX_NETWORK_BUFFER; i++) { p = select_sbcm_params(&port_buffer->buffer[i], lossless_buff_count); err = mlx5e_port_set_sbcm(mdev, 0, i, MLX5_INGRESS_DIR, @@ -293,7 +303,7 @@ static int port_set_buffer(struct mlx5e_priv *priv, if (err) goto out; - for (i = 0; i < MLX5E_MAX_BUFFER; i++) { + for (i = 0; i < MLX5E_MAX_NETWORK_BUFFER; i++) { void *buffer = MLX5_ADDR_OF(pbmc_reg, in, buffer[i]); u64 size = port_buffer->buffer[i].size; u64 xoff = port_buffer->buffer[i].xoff; @@ -351,7 +361,7 @@ static int update_xoff_threshold(struct mlx5e_port_buffer *port_buffer, { int i; - for (i = 0; i < MLX5E_MAX_BUFFER; i++) { + for (i = 0; i < MLX5E_MAX_NETWORK_BUFFER; i++) { if (port_buffer->buffer[i].lossy) { port_buffer->buffer[i].xoff = 0; port_buffer->buffer[i].xon = 0; @@ -408,7 +418,7 @@ static int update_buffer_lossy(struct mlx5_core_dev *mdev, int err; int i; - for (i = 0; i < MLX5E_MAX_BUFFER; i++) { + for (i = 0; i < MLX5E_MAX_NETWORK_BUFFER; i++) { prio_count = 0; lossy_count = 0; @@ -432,11 +442,11 @@ static int update_buffer_lossy(struct mlx5_core_dev *mdev, } if (changed) { - err = port_update_pool_cfg(mdev, port_buffer); + err = update_xoff_threshold(port_buffer, xoff, max_mtu, port_buff_cell_sz); if (err) return err; - err = update_xoff_threshold(port_buffer, xoff, max_mtu, port_buff_cell_sz); + err = port_update_pool_cfg(mdev, port_buffer); if (err) return err; @@ -515,7 +525,7 @@ int mlx5e_port_manual_buffer_config(struct mlx5e_priv *priv, if (change & MLX5E_PORT_BUFFER_PRIO2BUFFER) { update_prio2buffer = true; - for (i = 0; i < MLX5E_MAX_BUFFER; i++) + for (i = 0; i < MLX5E_MAX_NETWORK_BUFFER; i++) mlx5e_dbg(HW, priv, "%s: requested to map prio[%d] to buffer %d\n", __func__, i, prio2buffer[i]); @@ -530,7 +540,7 @@ int mlx5e_port_manual_buffer_config(struct mlx5e_priv *priv, } if (change & MLX5E_PORT_BUFFER_SIZE) { - for (i = 0; i < MLX5E_MAX_BUFFER; i++) { + for (i = 0; i < MLX5E_MAX_NETWORK_BUFFER; i++) { mlx5e_dbg(HW, priv, "%s: buffer[%d]=%d\n", __func__, i, buffer_size[i]); if (!port_buffer.buffer[i].lossy && !buffer_size[i]) { mlx5e_dbg(HW, priv, "%s: lossless buffer[%d] size cannot be zero\n", @@ -544,7 +554,9 @@ int mlx5e_port_manual_buffer_config(struct mlx5e_priv *priv, mlx5e_dbg(HW, priv, "%s: total buffer requested=%d\n", __func__, total_used); - if (total_used > port_buffer.port_buffer_size) + if (total_used > port_buffer.headroom_size && + (total_used - port_buffer.headroom_size) > + port_buffer.spare_buffer_size) return -EINVAL; update_buffer = true; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.h b/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.h index a6ef118de758f88591181d521d83cd2493a08bda..f4a19ffbb641c07b82291091e4f67f8524378e82 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.h @@ -35,7 +35,8 @@ #include "en.h" #include "port.h" -#define MLX5E_MAX_BUFFER 8 +#define MLX5E_MAX_NETWORK_BUFFER 8 +#define MLX5E_TOTAL_BUFFERS 10 #define MLX5E_DEFAULT_CABLE_LEN 7 /* 7 meters */ #define MLX5_BUFFER_SUPPORTED(mdev) (MLX5_CAP_GEN(mdev, pcam_reg) && \ @@ -60,8 +61,9 @@ struct mlx5e_bufferx_reg { struct mlx5e_port_buffer { u32 port_buffer_size; u32 spare_buffer_size; - u32 headroom_size; - struct mlx5e_bufferx_reg buffer[MLX5E_MAX_BUFFER]; + u32 headroom_size; /* Buffers 0-7 */ + u32 internal_buffers_size; /* Buffers 8-9 */ + struct mlx5e_bufferx_reg buffer[MLX5E_MAX_NETWORK_BUFFER]; }; int mlx5e_port_manual_buffer_config(struct mlx5e_priv *priv, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c index eb5abd0e55d9060ae62ea7542eeb2b111128c4ab..3cbebfba582bdbce79741c4c7d798dd13846af64 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c @@ -175,6 +175,8 @@ static bool mlx5e_ptp_poll_ts_cq(struct mlx5e_cq *cq, int budget) /* ensure cq space is freed before enabling more cqes */ wmb(); + mlx5e_txqsq_wake(&ptpsq->txqsq); + return work_done == budget; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c index fc923a99b6a483d5eaddba1f01c76d66549b74a3..0380a04c3691cbc72d13c873886a4ffd3ec96d01 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c @@ -84,7 +84,7 @@ mlx5e_tc_act_init_parse_state(struct mlx5e_tc_act_parse_state *parse_state, int mlx5e_tc_act_post_parse(struct mlx5e_tc_act_parse_state *parse_state, - struct flow_action *flow_action, + struct flow_action *flow_action, int from, int to, struct mlx5_flow_attr *attr, enum mlx5_flow_namespace_type ns_type) { @@ -96,6 +96,11 @@ mlx5e_tc_act_post_parse(struct mlx5e_tc_act_parse_state *parse_state, priv = parse_state->flow->priv; flow_action_for_each(i, act, flow_action) { + if (i < from) + continue; + else if (i > to) + break; + tc_act = mlx5e_tc_act_get(act->id, ns_type); if (!tc_act || !tc_act->post_parse) continue; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h index 0e6e1872ac62e291aef03dcafd4d413fcb9d6bb7..d6c12d0ea55ba5458bda9d0ec1199ab2b4891c2d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h @@ -112,7 +112,7 @@ mlx5e_tc_act_init_parse_state(struct mlx5e_tc_act_parse_state *parse_state, int mlx5e_tc_act_post_parse(struct mlx5e_tc_act_parse_state *parse_state, - struct flow_action *flow_action, + struct flow_action *flow_action, int from, int to, struct mlx5_flow_attr *attr, enum mlx5_flow_namespace_type ns_type); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c index 20c2d2ecaf9356da8d253e8e33a9b45add25bf73..f0c3464f037f4a0d0bc37047e7ace1b50ba8ff6c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c @@ -492,6 +492,19 @@ void mlx5e_encap_put(struct mlx5e_priv *priv, struct mlx5e_encap_entry *e) mlx5e_encap_dealloc(priv, e); } +static void mlx5e_encap_put_locked(struct mlx5e_priv *priv, struct mlx5e_encap_entry *e) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + + lockdep_assert_held(&esw->offloads.encap_tbl_lock); + + if (!refcount_dec_and_test(&e->refcnt)) + return; + list_del(&e->route_list); + hash_del_rcu(&e->encap_hlist); + mlx5e_encap_dealloc(priv, e); +} + static void mlx5e_decap_put(struct mlx5e_priv *priv, struct mlx5e_decap_entry *d) { struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; @@ -816,6 +829,8 @@ int mlx5e_attach_encap(struct mlx5e_priv *priv, uintptr_t hash_key; int err = 0; + lockdep_assert_held(&esw->offloads.encap_tbl_lock); + parse_attr = attr->parse_attr; tun_info = parse_attr->tun_info[out_index]; mpls_info = &parse_attr->mpls_info[out_index]; @@ -829,7 +844,6 @@ int mlx5e_attach_encap(struct mlx5e_priv *priv, hash_key = hash_encap_info(&key); - mutex_lock(&esw->offloads.encap_tbl_lock); e = mlx5e_encap_get(priv, &key, hash_key); /* must verify if encap is valid or not */ @@ -840,15 +854,6 @@ int mlx5e_attach_encap(struct mlx5e_priv *priv, goto out_err; } - mutex_unlock(&esw->offloads.encap_tbl_lock); - wait_for_completion(&e->res_ready); - - /* Protect against concurrent neigh update. */ - mutex_lock(&esw->offloads.encap_tbl_lock); - if (e->compl_result < 0) { - err = -EREMOTEIO; - goto out_err; - } goto attach_flow; } @@ -877,15 +882,12 @@ int mlx5e_attach_encap(struct mlx5e_priv *priv, INIT_LIST_HEAD(&e->flows); hash_add_rcu(esw->offloads.encap_tbl, &e->encap_hlist, hash_key); tbl_time_before = mlx5e_route_tbl_get_last_update(priv); - mutex_unlock(&esw->offloads.encap_tbl_lock); if (family == AF_INET) err = mlx5e_tc_tun_create_header_ipv4(priv, mirred_dev, e); else if (family == AF_INET6) err = mlx5e_tc_tun_create_header_ipv6(priv, mirred_dev, e); - /* Protect against concurrent neigh update. */ - mutex_lock(&esw->offloads.encap_tbl_lock); complete_all(&e->res_ready); if (err) { e->compl_result = err; @@ -920,18 +922,15 @@ int mlx5e_attach_encap(struct mlx5e_priv *priv, } else { flow_flag_set(flow, SLOW); } - mutex_unlock(&esw->offloads.encap_tbl_lock); return err; out_err: - mutex_unlock(&esw->offloads.encap_tbl_lock); if (e) - mlx5e_encap_put(priv, e); + mlx5e_encap_put_locked(priv, e); return err; out_err_init: - mutex_unlock(&esw->offloads.encap_tbl_lock); kfree(tun_info); kfree(e); return err; @@ -1016,6 +1015,93 @@ int mlx5e_attach_decap(struct mlx5e_priv *priv, return err; } +int mlx5e_tc_tun_encap_dests_set(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, + struct mlx5_flow_attr *attr, + struct netlink_ext_ack *extack, + bool *vf_tun) +{ + struct mlx5e_tc_flow_parse_attr *parse_attr; + struct mlx5_esw_flow_attr *esw_attr; + struct net_device *encap_dev = NULL; + struct mlx5e_rep_priv *rpriv; + struct mlx5e_priv *out_priv; + struct mlx5_eswitch *esw; + int out_index; + int err = 0; + + if (!mlx5e_is_eswitch_flow(flow)) + return 0; + + parse_attr = attr->parse_attr; + esw_attr = attr->esw_attr; + *vf_tun = false; + + esw = priv->mdev->priv.eswitch; + mutex_lock(&esw->offloads.encap_tbl_lock); + for (out_index = 0; out_index < MLX5_MAX_FLOW_FWD_VPORTS; out_index++) { + struct net_device *out_dev; + int mirred_ifindex; + + if (!(esw_attr->dests[out_index].flags & MLX5_ESW_DEST_ENCAP)) + continue; + + mirred_ifindex = parse_attr->mirred_ifindex[out_index]; + out_dev = dev_get_by_index(dev_net(priv->netdev), mirred_ifindex); + if (!out_dev) { + NL_SET_ERR_MSG_MOD(extack, "Requested mirred device not found"); + err = -ENODEV; + goto out; + } + err = mlx5e_attach_encap(priv, flow, attr, out_dev, out_index, + extack, &encap_dev); + dev_put(out_dev); + if (err) + goto out; + + if (esw_attr->dests[out_index].flags & + MLX5_ESW_DEST_CHAIN_WITH_SRC_PORT_CHANGE && + !esw_attr->dest_int_port) + *vf_tun = true; + + out_priv = netdev_priv(encap_dev); + rpriv = out_priv->ppriv; + esw_attr->dests[out_index].rep = rpriv->rep; + esw_attr->dests[out_index].mdev = out_priv->mdev; + } + + if (*vf_tun && esw_attr->out_count > 1) { + NL_SET_ERR_MSG_MOD(extack, "VF tunnel encap with mirroring is not supported"); + err = -EOPNOTSUPP; + goto out; + } + +out: + mutex_unlock(&esw->offloads.encap_tbl_lock); + return err; +} + +void mlx5e_tc_tun_encap_dests_unset(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, + struct mlx5_flow_attr *attr) +{ + struct mlx5_esw_flow_attr *esw_attr; + int out_index; + + if (!mlx5e_is_eswitch_flow(flow)) + return; + + esw_attr = attr->esw_attr; + + for (out_index = 0; out_index < MLX5_MAX_FLOW_FWD_VPORTS; out_index++) { + if (!(esw_attr->dests[out_index].flags & MLX5_ESW_DEST_ENCAP)) + continue; + + mlx5e_detach_encap(flow->priv, flow, attr, out_index); + kfree(attr->parse_attr->tun_info[out_index]); + } +} + static int cmp_route_info(struct mlx5e_route_key *a, struct mlx5e_route_key *b) { @@ -1369,11 +1455,13 @@ static void mlx5e_invalidate_encap(struct mlx5e_priv *priv, struct mlx5e_tc_flow *flow; list_for_each_entry(flow, encap_flows, tmp_list) { - struct mlx5_flow_attr *attr = flow->attr; struct mlx5_esw_flow_attr *esw_attr; + struct mlx5_flow_attr *attr; if (!mlx5e_is_offloaded_flow(flow)) continue; + + attr = mlx5e_tc_get_encap_attr(flow); esw_attr = attr->esw_attr; if (flow_flag_test(flow, SLOW)) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.h index 8ad273dde40ee5b41ad1cdc4963ce8ad9d10fccb..5d7d67687cbcd29204ae121e4af99b1292335b02 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.h @@ -30,6 +30,15 @@ int mlx5e_attach_decap_route(struct mlx5e_priv *priv, void mlx5e_detach_decap_route(struct mlx5e_priv *priv, struct mlx5e_tc_flow *flow); +int mlx5e_tc_tun_encap_dests_set(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, + struct mlx5_flow_attr *attr, + struct netlink_ext_ack *extack, + bool *vf_tun); +void mlx5e_tc_tun_encap_dests_unset(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, + struct mlx5_flow_attr *attr); + struct ip_tunnel_info *mlx5e_dup_tun_info(const struct ip_tunnel_info *tun_info); int mlx5e_tc_set_attr_rx_tun(struct mlx5e_tc_flow *flow, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h index 47381e949f1f5fc7006fddeb2b5e19e1dee7a354..879d698b6119303836915d5dc19f6f910d506890 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h @@ -193,6 +193,8 @@ static inline u16 mlx5e_txqsq_get_next_pi(struct mlx5e_txqsq *sq, u16 size) return pi; } +void mlx5e_txqsq_wake(struct mlx5e_txqsq *sq); + static inline u16 mlx5e_shampo_get_cqe_header_index(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe) { return be16_to_cpu(cqe->shampo.header_entry_index) & (rq->mpwqe.shampo->hd_per_wq - 1); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c index 1f90594499c6098d06f37f1beed887d200b1a1a4..41c396e764579492b21b2f1cedb3f54a471894c4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c @@ -150,10 +150,8 @@ int mlx5e_refresh_tirs(struct mlx5e_priv *priv, bool enable_uc_lb, inlen = MLX5_ST_SZ_BYTES(modify_tir_in); in = kvzalloc(inlen, GFP_KERNEL); - if (!in) { - err = -ENOMEM; - goto out; - } + if (!in) + return -ENOMEM; if (enable_uc_lb) lb_flags = MLX5_TIRC_SELF_LB_BLOCK_BLOCK_UNICAST; @@ -171,14 +169,13 @@ int mlx5e_refresh_tirs(struct mlx5e_priv *priv, bool enable_uc_lb, tirn = tir->tirn; err = mlx5_core_modify_tir(mdev, tirn, in); if (err) - goto out; + break; } + mutex_unlock(&mdev->mlx5e_res.hw_objs.td.list_lock); -out: kvfree(in); if (err) netdev_err(priv->netdev, "refresh tir(0x%x) failed, %d\n", tirn, err); - mutex_unlock(&mdev->mlx5e_res.hw_objs.td.list_lock); return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c index 89de92d0648363a899793f1aa5ea4fb324a9ddc8..ebee52a8361aa7f7102683c4d62dbc4571ed1a71 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c @@ -926,9 +926,10 @@ static int mlx5e_dcbnl_getbuffer(struct net_device *dev, if (err) return err; - for (i = 0; i < MLX5E_MAX_BUFFER; i++) + for (i = 0; i < MLX5E_MAX_NETWORK_BUFFER; i++) dcb_buffer->buffer_size[i] = port_buffer.buffer[i].size; - dcb_buffer->total_size = port_buffer.port_buffer_size; + dcb_buffer->total_size = port_buffer.port_buffer_size - + port_buffer.internal_buffers_size; return 0; } @@ -970,7 +971,7 @@ static int mlx5e_dcbnl_setbuffer(struct net_device *dev, if (err) return err; - for (i = 0; i < MLX5E_MAX_BUFFER; i++) { + for (i = 0; i < MLX5E_MAX_NETWORK_BUFFER; i++) { if (port_buffer.buffer[i].size != dcb_buffer->buffer_size[i]) { changed |= MLX5E_PORT_BUFFER_SIZE; buffer_size = dcb_buffer->buffer_size; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 2944691f06added0b8420c962f514a7bee413e49..a7c526ee5024748097a39abf2f718c0e9ea28f9d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -727,26 +727,6 @@ static void mlx5e_rq_free_shampo(struct mlx5e_rq *rq) mlx5e_rq_shampo_hd_free(rq); } -static __be32 mlx5e_get_terminate_scatter_list_mkey(struct mlx5_core_dev *dev) -{ - u32 out[MLX5_ST_SZ_DW(query_special_contexts_out)] = {}; - u32 in[MLX5_ST_SZ_DW(query_special_contexts_in)] = {}; - int res; - - if (!MLX5_CAP_GEN(dev, terminate_scatter_list_mkey)) - return MLX5_TERMINATE_SCATTER_LIST_LKEY; - - MLX5_SET(query_special_contexts_in, in, opcode, - MLX5_CMD_OP_QUERY_SPECIAL_CONTEXTS); - res = mlx5_cmd_exec_inout(dev, query_special_contexts, in, out); - if (res) - return MLX5_TERMINATE_SCATTER_LIST_LKEY; - - res = MLX5_GET(query_special_contexts_out, out, - terminate_scatter_list_mkey); - return cpu_to_be32(res); -} - static int mlx5e_alloc_rq(struct mlx5e_params *params, struct mlx5e_xsk_param *xsk, struct mlx5e_rq_param *rqp, @@ -908,7 +888,7 @@ static int mlx5e_alloc_rq(struct mlx5e_params *params, /* check if num_frags is not a pow of two */ if (rq->wqe.info.num_frags < (1 << rq->wqe.info.log_num_frags)) { wqe->data[f].byte_count = 0; - wqe->data[f].lkey = mlx5e_get_terminate_scatter_list_mkey(mdev); + wqe->data[f].lkey = params->terminate_lkey_be; wqe->data[f].addr = 0; } } @@ -5007,6 +4987,8 @@ void mlx5e_build_nic_params(struct mlx5e_priv *priv, struct mlx5e_xsk *xsk, u16 /* RQ */ mlx5e_build_rq_params(mdev, params); + params->terminate_lkey_be = mlx5_core_get_terminate_scatter_list_mkey(mdev); + params->packet_merge.timeout = mlx5e_choose_lro_timeout(mdev, MLX5E_DEFAULT_LRO_TIMEOUT); /* CQ moderation params */ @@ -5279,12 +5261,16 @@ static int mlx5e_nic_init(struct mlx5_core_dev *mdev, mlx5e_timestamp_init(priv); + priv->dfs_root = debugfs_create_dir("nic", + mlx5_debugfs_get_dev_root(mdev)); + fs = mlx5e_fs_init(priv->profile, mdev, !test_bit(MLX5E_STATE_DESTROYING, &priv->state), priv->dfs_root); if (!fs) { err = -ENOMEM; mlx5_core_err(mdev, "FS initialization failed, %d\n", err); + debugfs_remove_recursive(priv->dfs_root); return err; } priv->fs = fs; @@ -5305,6 +5291,7 @@ static void mlx5e_nic_cleanup(struct mlx5e_priv *priv) mlx5e_health_destroy_reporters(priv); mlx5e_ktls_cleanup(priv); mlx5e_fs_cleanup(priv->fs); + debugfs_remove_recursive(priv->dfs_root); priv->fs = NULL; } @@ -5851,8 +5838,8 @@ void mlx5e_detach_netdev(struct mlx5e_priv *priv) } static int -mlx5e_netdev_attach_profile(struct net_device *netdev, struct mlx5_core_dev *mdev, - const struct mlx5e_profile *new_profile, void *new_ppriv) +mlx5e_netdev_init_profile(struct net_device *netdev, struct mlx5_core_dev *mdev, + const struct mlx5e_profile *new_profile, void *new_ppriv) { struct mlx5e_priv *priv = netdev_priv(netdev); int err; @@ -5868,6 +5855,25 @@ mlx5e_netdev_attach_profile(struct net_device *netdev, struct mlx5_core_dev *mde err = new_profile->init(priv->mdev, priv->netdev); if (err) goto priv_cleanup; + + return 0; + +priv_cleanup: + mlx5e_priv_cleanup(priv); + return err; +} + +static int +mlx5e_netdev_attach_profile(struct net_device *netdev, struct mlx5_core_dev *mdev, + const struct mlx5e_profile *new_profile, void *new_ppriv) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + int err; + + err = mlx5e_netdev_init_profile(netdev, mdev, new_profile, new_ppriv); + if (err) + return err; + err = mlx5e_attach_netdev(priv); if (err) goto profile_cleanup; @@ -5875,7 +5881,6 @@ mlx5e_netdev_attach_profile(struct net_device *netdev, struct mlx5_core_dev *mde profile_cleanup: new_profile->cleanup(priv); -priv_cleanup: mlx5e_priv_cleanup(priv); return err; } @@ -5894,6 +5899,12 @@ int mlx5e_netdev_change_profile(struct mlx5e_priv *priv, priv->profile->cleanup(priv); mlx5e_priv_cleanup(priv); + if (mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR) { + mlx5e_netdev_init_profile(netdev, mdev, new_profile, new_ppriv); + set_bit(MLX5E_STATE_DESTROYING, &priv->state); + return -EIO; + } + err = mlx5e_netdev_attach_profile(netdev, mdev, new_profile, new_ppriv); if (err) { /* roll back to original profile */ netdev_warn(netdev, "%s: new profile init failed, %d\n", __func__, err); @@ -5955,8 +5966,11 @@ static int mlx5e_suspend(struct auxiliary_device *adev, pm_message_t state) struct net_device *netdev = priv->netdev; struct mlx5_core_dev *mdev = priv->mdev; - if (!netif_device_present(netdev)) + if (!netif_device_present(netdev)) { + if (test_bit(MLX5E_STATE_DESTROYING, &priv->state)) + mlx5e_destroy_mdev_resources(mdev); return -ENODEV; + } mlx5e_detach_netdev(priv); mlx5e_destroy_mdev_resources(mdev); @@ -6002,9 +6016,6 @@ static int mlx5e_probe(struct auxiliary_device *adev, priv->profile = profile; priv->ppriv = NULL; - priv->dfs_root = debugfs_create_dir("nic", - mlx5_debugfs_get_dev_root(priv->mdev)); - err = profile->init(mdev, netdev); if (err) { mlx5_core_err(mdev, "mlx5e_nic_profile init failed, %d\n", err); @@ -6033,7 +6044,6 @@ static int mlx5e_probe(struct auxiliary_device *adev, err_profile_cleanup: profile->cleanup(priv); err_destroy_netdev: - debugfs_remove_recursive(priv->dfs_root); mlx5e_destroy_netdev(priv); err_devlink_port_unregister: mlx5e_devlink_port_unregister(mlx5e_dev); @@ -6053,7 +6063,6 @@ static void mlx5e_remove(struct auxiliary_device *adev) unregister_netdev(priv->netdev); mlx5e_suspend(adev, state); priv->profile->cleanup(priv); - debugfs_remove_recursive(priv->dfs_root); mlx5e_destroy_netdev(priv); mlx5e_devlink_port_unregister(mlx5e_dev); mlx5e_destroy_devlink(mlx5e_dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c index 1fc386eccaf8f9f1345d1a2c640b1039781ca0e1..3e7041bd5705e8761258d3513c5bc358a602cf68 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c @@ -30,6 +30,7 @@ * SOFTWARE. */ +#include #include #include #include @@ -812,11 +813,15 @@ static int mlx5e_init_ul_rep(struct mlx5_core_dev *mdev, { struct mlx5e_priv *priv = netdev_priv(netdev); + priv->dfs_root = debugfs_create_dir("nic", + mlx5_debugfs_get_dev_root(mdev)); + priv->fs = mlx5e_fs_init(priv->profile, mdev, !test_bit(MLX5E_STATE_DESTROYING, &priv->state), priv->dfs_root); if (!priv->fs) { netdev_err(priv->netdev, "FS allocation failed\n"); + debugfs_remove_recursive(priv->dfs_root); return -ENOMEM; } @@ -829,6 +834,7 @@ static int mlx5e_init_ul_rep(struct mlx5_core_dev *mdev, static void mlx5e_cleanup_rep(struct mlx5e_priv *priv) { mlx5e_fs_cleanup(priv->fs); + debugfs_remove_recursive(priv->dfs_root); priv->fs = NULL; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 728b82ce4031a2b6ce3e0434064980f39985ff2e..8a5a8703f0a31b0c4b2feadfb39a5cbcdb1d9196 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -1665,11 +1665,9 @@ bool mlx5e_tc_is_vf_tunnel(struct net_device *out_dev, struct net_device *route_ int mlx5e_tc_query_route_vport(struct net_device *out_dev, struct net_device *route_dev, u16 *vport) { struct mlx5e_priv *out_priv, *route_priv; - struct mlx5_devcom *devcom = NULL; struct mlx5_core_dev *route_mdev; struct mlx5_eswitch *esw; u16 vhca_id; - int err; out_priv = netdev_priv(out_dev); esw = out_priv->mdev->priv.eswitch; @@ -1678,6 +1676,9 @@ int mlx5e_tc_query_route_vport(struct net_device *out_dev, struct net_device *ro vhca_id = MLX5_CAP_GEN(route_mdev, vhca_id); if (mlx5_lag_is_active(out_priv->mdev)) { + struct mlx5_devcom *devcom; + int err; + /* In lag case we may get devices from different eswitch instances. * If we failed to get vport num, it means, mostly, that we on the wrong * eswitch. @@ -1686,101 +1687,16 @@ int mlx5e_tc_query_route_vport(struct net_device *out_dev, struct net_device *ro if (err != -ENOENT) return err; + rcu_read_lock(); devcom = out_priv->mdev->priv.devcom; - esw = mlx5_devcom_get_peer_data(devcom, MLX5_DEVCOM_ESW_OFFLOADS); - if (!esw) - return -ENODEV; - } - - err = mlx5_eswitch_vhca_id_to_vport(esw, vhca_id, vport); - if (devcom) - mlx5_devcom_release_peer_data(devcom, MLX5_DEVCOM_ESW_OFFLOADS); - return err; -} - -static int -set_encap_dests(struct mlx5e_priv *priv, - struct mlx5e_tc_flow *flow, - struct mlx5_flow_attr *attr, - struct netlink_ext_ack *extack, - bool *vf_tun) -{ - struct mlx5e_tc_flow_parse_attr *parse_attr; - struct mlx5_esw_flow_attr *esw_attr; - struct net_device *encap_dev = NULL; - struct mlx5e_rep_priv *rpriv; - struct mlx5e_priv *out_priv; - int out_index; - int err = 0; - - if (!mlx5e_is_eswitch_flow(flow)) - return 0; - - parse_attr = attr->parse_attr; - esw_attr = attr->esw_attr; - *vf_tun = false; - - for (out_index = 0; out_index < MLX5_MAX_FLOW_FWD_VPORTS; out_index++) { - struct net_device *out_dev; - int mirred_ifindex; - - if (!(esw_attr->dests[out_index].flags & MLX5_ESW_DEST_ENCAP)) - continue; - - mirred_ifindex = parse_attr->mirred_ifindex[out_index]; - out_dev = dev_get_by_index(dev_net(priv->netdev), mirred_ifindex); - if (!out_dev) { - NL_SET_ERR_MSG_MOD(extack, "Requested mirred device not found"); - err = -ENODEV; - goto out; - } - err = mlx5e_attach_encap(priv, flow, attr, out_dev, out_index, - extack, &encap_dev); - dev_put(out_dev); - if (err) - goto out; + esw = mlx5_devcom_get_peer_data_rcu(devcom, MLX5_DEVCOM_ESW_OFFLOADS); + err = esw ? mlx5_eswitch_vhca_id_to_vport(esw, vhca_id, vport) : -ENODEV; + rcu_read_unlock(); - if (esw_attr->dests[out_index].flags & - MLX5_ESW_DEST_CHAIN_WITH_SRC_PORT_CHANGE && - !esw_attr->dest_int_port) - *vf_tun = true; - - out_priv = netdev_priv(encap_dev); - rpriv = out_priv->ppriv; - esw_attr->dests[out_index].rep = rpriv->rep; - esw_attr->dests[out_index].mdev = out_priv->mdev; - } - - if (*vf_tun && esw_attr->out_count > 1) { - NL_SET_ERR_MSG_MOD(extack, "VF tunnel encap with mirroring is not supported"); - err = -EOPNOTSUPP; - goto out; + return err; } -out: - return err; -} - -static void -clean_encap_dests(struct mlx5e_priv *priv, - struct mlx5e_tc_flow *flow, - struct mlx5_flow_attr *attr) -{ - struct mlx5_esw_flow_attr *esw_attr; - int out_index; - - if (!mlx5e_is_eswitch_flow(flow)) - return; - - esw_attr = attr->esw_attr; - - for (out_index = 0; out_index < MLX5_MAX_FLOW_FWD_VPORTS; out_index++) { - if (!(esw_attr->dests[out_index].flags & MLX5_ESW_DEST_ENCAP)) - continue; - - mlx5e_detach_encap(priv, flow, attr, out_index); - kfree(attr->parse_attr->tun_info[out_index]); - } + return mlx5_eswitch_vhca_id_to_vport(esw, vhca_id, vport); } static int @@ -1819,7 +1735,7 @@ post_process_attr(struct mlx5e_tc_flow *flow, if (err) goto err_out; - err = set_encap_dests(flow->priv, flow, attr, extack, &vf_tun); + err = mlx5e_tc_tun_encap_dests_set(flow->priv, flow, attr, extack, &vf_tun); if (err) goto err_out; @@ -3943,8 +3859,8 @@ parse_tc_actions(struct mlx5e_tc_act_parse_state *parse_state, struct mlx5_flow_attr *prev_attr; struct flow_action_entry *act; struct mlx5e_tc_act *tc_act; + int err, i, i_split = 0; bool is_missable; - int err, i; ns_type = mlx5e_get_flow_namespace(flow); list_add(&attr->list, &flow->attrs); @@ -3985,7 +3901,8 @@ parse_tc_actions(struct mlx5e_tc_act_parse_state *parse_state, i < flow_action->num_entries - 1)) { is_missable = tc_act->is_missable ? tc_act->is_missable(act) : false; - err = mlx5e_tc_act_post_parse(parse_state, flow_action, attr, ns_type); + err = mlx5e_tc_act_post_parse(parse_state, flow_action, i_split, i, attr, + ns_type); if (err) goto out_free_post_acts; @@ -3995,6 +3912,7 @@ parse_tc_actions(struct mlx5e_tc_act_parse_state *parse_state, goto out_free_post_acts; } + i_split = i + 1; list_add(&attr->list, &flow->attrs); } @@ -4009,7 +3927,7 @@ parse_tc_actions(struct mlx5e_tc_act_parse_state *parse_state, } } - err = mlx5e_tc_act_post_parse(parse_state, flow_action, attr, ns_type); + err = mlx5e_tc_act_post_parse(parse_state, flow_action, i_split, i, attr, ns_type); if (err) goto out_free_post_acts; @@ -4323,7 +4241,7 @@ mlx5_free_flow_attr_actions(struct mlx5e_tc_flow *flow, struct mlx5_flow_attr *a if (attr->post_act_handle) mlx5e_tc_post_act_del(get_post_action(flow->priv), attr->post_act_handle); - clean_encap_dests(flow->priv, flow, attr); + mlx5e_tc_tun_encap_dests_unset(flow->priv, flow, attr); if (attr->action & MLX5_FLOW_CONTEXT_ACTION_COUNT) mlx5_fc_destroy(counter_dev, attr->counter); @@ -5301,6 +5219,8 @@ int mlx5e_tc_esw_init(struct mlx5_rep_uplink_priv *uplink_priv) goto err_action_counter; } + mlx5_esw_offloads_devcom_init(esw); + return 0; err_action_counter: @@ -5329,7 +5249,7 @@ void mlx5e_tc_esw_cleanup(struct mlx5_rep_uplink_priv *uplink_priv) priv = netdev_priv(rpriv->netdev); esw = priv->mdev->priv.eswitch; - mlx5e_tc_clean_fdb_peer_flows(esw); + mlx5_esw_offloads_devcom_cleanup(esw); mlx5e_tc_tun_cleanup(uplink_priv->encap); @@ -5643,22 +5563,43 @@ bool mlx5e_tc_update_skb_nic(struct mlx5_cqe64 *cqe, struct sk_buff *skb) 0, NULL); } +static struct mapping_ctx * +mlx5e_get_priv_obj_mapping(struct mlx5e_priv *priv) +{ + struct mlx5e_tc_table *tc; + struct mlx5_eswitch *esw; + struct mapping_ctx *ctx; + + if (is_mdev_switchdev_mode(priv->mdev)) { + esw = priv->mdev->priv.eswitch; + ctx = esw->offloads.reg_c0_obj_pool; + } else { + tc = mlx5e_fs_get_tc(priv->fs); + ctx = tc->mapping; + } + + return ctx; +} + int mlx5e_tc_action_miss_mapping_get(struct mlx5e_priv *priv, struct mlx5_flow_attr *attr, u64 act_miss_cookie, u32 *act_miss_mapping) { - struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; struct mlx5_mapped_obj mapped_obj = {}; + struct mlx5_eswitch *esw; struct mapping_ctx *ctx; int err; - ctx = esw->offloads.reg_c0_obj_pool; - + ctx = mlx5e_get_priv_obj_mapping(priv); mapped_obj.type = MLX5_MAPPED_OBJ_ACT_MISS; mapped_obj.act_miss_cookie = act_miss_cookie; err = mapping_add(ctx, &mapped_obj, act_miss_mapping); if (err) return err; + if (!is_mdev_switchdev_mode(priv->mdev)) + return 0; + + esw = priv->mdev->priv.eswitch; attr->act_id_restore_rule = esw_add_restore_rule(esw, *act_miss_mapping); if (IS_ERR(attr->act_id_restore_rule)) goto err_rule; @@ -5673,10 +5614,9 @@ int mlx5e_tc_action_miss_mapping_get(struct mlx5e_priv *priv, struct mlx5_flow_a void mlx5e_tc_action_miss_mapping_put(struct mlx5e_priv *priv, struct mlx5_flow_attr *attr, u32 act_miss_mapping) { - struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; - struct mapping_ctx *ctx; + struct mapping_ctx *ctx = mlx5e_get_priv_obj_mapping(priv); - ctx = esw->offloads.reg_c0_obj_pool; - mlx5_del_flow_rules(attr->act_id_restore_rule); + if (is_mdev_switchdev_mode(priv->mdev)) + mlx5_del_flow_rules(attr->act_id_restore_rule); mapping_remove(ctx, act_miss_mapping); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index df5e780e8e6a6537b23c8fc8b0fb18234480e108..c7eb6b238c2ba15510533a904920afd9790dca5e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -762,6 +762,17 @@ static void mlx5e_tx_wi_consume_fifo_skbs(struct mlx5e_txqsq *sq, struct mlx5e_t } } +void mlx5e_txqsq_wake(struct mlx5e_txqsq *sq) +{ + if (netif_tx_queue_stopped(sq->txq) && + mlx5e_wqc_has_room_for(&sq->wq, sq->cc, sq->pc, sq->stop_room) && + mlx5e_ptpsq_fifo_has_room(sq) && + !test_bit(MLX5E_SQ_STATE_RECOVERING, &sq->state)) { + netif_tx_wake_queue(sq->txq); + sq->stats->wake++; + } +} + bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) { struct mlx5e_sq_stats *stats; @@ -861,13 +872,7 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) netdev_tx_completed_queue(sq->txq, npkts, nbytes); - if (netif_tx_queue_stopped(sq->txq) && - mlx5e_wqc_has_room_for(&sq->wq, sq->cc, sq->pc, sq->stop_room) && - mlx5e_ptpsq_fifo_has_room(sq) && - !test_bit(MLX5E_SQ_STATE_RECOVERING, &sq->state)) { - netif_tx_wake_queue(sq->txq); - stats->wake++; - } + mlx5e_txqsq_wake(sq); return (i == MLX5E_TX_CQ_POLL_BUDGET); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c index a50bfda18e96d95796ecb10a80c839d17455d569..fbb2d963fb7e1bc3afce053980f360f2467d1bf1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c @@ -161,20 +161,22 @@ int mlx5e_napi_poll(struct napi_struct *napi, int budget) } } + /* budget=0 means we may be in IRQ context, do as little as possible */ + if (unlikely(!budget)) + goto out; + busy |= mlx5e_poll_xdpsq_cq(&c->xdpsq.cq); if (c->xdp) busy |= mlx5e_poll_xdpsq_cq(&c->rq_xdpsq.cq); - if (likely(budget)) { /* budget=0 means: don't poll rx rings */ - if (xsk_open) - work_done = mlx5e_poll_rx_cq(&xskrq->cq, budget); + if (xsk_open) + work_done = mlx5e_poll_rx_cq(&xskrq->cq, budget); - if (likely(budget - work_done)) - work_done += mlx5e_poll_rx_cq(&rq->cq, budget - work_done); + if (likely(budget - work_done)) + work_done += mlx5e_poll_rx_cq(&rq->cq, budget - work_done); - busy |= work_done == budget; - } + busy |= work_done == budget; mlx5e_poll_ico_cq(&c->icosq.cq); if (mlx5e_poll_ico_cq(&c->async_icosq.cq)) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c index 1c35d721a31d7d250705aa7630ea666cffedc639..3db4866d7880fc4c41b05971e8fc4224e8f391ac 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c @@ -824,7 +824,7 @@ static int comp_irqs_request_pci(struct mlx5_core_dev *dev) ncomp_eqs = table->num_comp_eqs; cpus = kcalloc(ncomp_eqs, sizeof(*cpus), GFP_KERNEL); if (!cpus) - ret = -ENOMEM; + return -ENOMEM; i = 0; rcu_read_lock(); @@ -1104,7 +1104,7 @@ void mlx5_core_eq_free_irqs(struct mlx5_core_dev *dev) struct mlx5_eq_table *table = dev->priv.eq_table; mutex_lock(&table->lock); /* sync with create/destroy_async_eq */ - mlx5_irq_table_destroy(dev); + mlx5_irq_table_free_irqs(dev); mutex_unlock(&table->lock); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 1a042c9817131e7860a046c77ed7cc18b0b6e7e6..add6cfa432a5b7dd2c5cd4830a576d3ec9a00e7e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -342,6 +342,7 @@ struct mlx5_eswitch { u32 large_group_num; } params; struct blocking_notifier_head n_head; + bool paired[MLX5_MAX_PORTS]; }; void esw_offloads_disable(struct mlx5_eswitch *esw); @@ -369,6 +370,8 @@ int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs); void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw, bool clear_vf); void mlx5_eswitch_disable_locked(struct mlx5_eswitch *esw); void mlx5_eswitch_disable(struct mlx5_eswitch *esw); +void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw); +void mlx5_esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw); int mlx5_eswitch_set_vport_mac(struct mlx5_eswitch *esw, u16 vport, const u8 *mac); int mlx5_eswitch_set_vport_state(struct mlx5_eswitch *esw, @@ -767,6 +770,8 @@ static inline void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw) {} static inline int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs) { return 0; } static inline void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw, bool clear_vf) {} static inline void mlx5_eswitch_disable(struct mlx5_eswitch *esw) {} +static inline void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw) {} +static inline void mlx5_esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw) {} static inline bool mlx5_eswitch_is_funcs_handler(struct mlx5_core_dev *dev) { return false; } static inline int mlx5_eswitch_set_vport_state(struct mlx5_eswitch *esw, u16 vport, int link_state) { return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 69215ffb9999532d839f5c21f7881f716ee35c96..8d19c20d3447e67cb63b4e7b4ead294ba3147b66 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -2742,6 +2742,9 @@ static int mlx5_esw_offloads_devcom_event(int event, mlx5_eswitch_vport_match_metadata_enabled(peer_esw)) break; + if (esw->paired[mlx5_get_dev_index(peer_esw->dev)]) + break; + err = mlx5_esw_offloads_set_ns_peer(esw, peer_esw, true); if (err) goto err_out; @@ -2753,14 +2756,18 @@ static int mlx5_esw_offloads_devcom_event(int event, if (err) goto err_pair; + esw->paired[mlx5_get_dev_index(peer_esw->dev)] = true; + peer_esw->paired[mlx5_get_dev_index(esw->dev)] = true; mlx5_devcom_set_paired(devcom, MLX5_DEVCOM_ESW_OFFLOADS, true); break; case ESW_OFFLOADS_DEVCOM_UNPAIR: - if (!mlx5_devcom_is_paired(devcom, MLX5_DEVCOM_ESW_OFFLOADS)) + if (!esw->paired[mlx5_get_dev_index(peer_esw->dev)]) break; mlx5_devcom_set_paired(devcom, MLX5_DEVCOM_ESW_OFFLOADS, false); + esw->paired[mlx5_get_dev_index(peer_esw->dev)] = false; + peer_esw->paired[mlx5_get_dev_index(esw->dev)] = false; mlx5_esw_offloads_unpair(peer_esw); mlx5_esw_offloads_unpair(esw); mlx5_esw_offloads_set_ns_peer(esw, peer_esw, false); @@ -2779,7 +2786,7 @@ static int mlx5_esw_offloads_devcom_event(int event, return err; } -static void esw_offloads_devcom_init(struct mlx5_eswitch *esw) +void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw) { struct mlx5_devcom *devcom = esw->dev->priv.devcom; @@ -2802,7 +2809,7 @@ static void esw_offloads_devcom_init(struct mlx5_eswitch *esw) ESW_OFFLOADS_DEVCOM_PAIR, esw); } -static void esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw) +void mlx5_esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw) { struct mlx5_devcom *devcom = esw->dev->priv.devcom; @@ -3250,8 +3257,6 @@ int esw_offloads_enable(struct mlx5_eswitch *esw) if (err) goto err_vports; - esw_offloads_devcom_init(esw); - return 0; err_vports: @@ -3292,7 +3297,6 @@ static int esw_offloads_stop(struct mlx5_eswitch *esw, void esw_offloads_disable(struct mlx5_eswitch *esw) { - esw_offloads_devcom_cleanup(esw); mlx5_eswitch_disable_pf_vf_vports(esw); esw_offloads_unload_rep(esw, MLX5_VPORT_UPLINK); esw_set_passing_vport_metadata(esw, false); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c index adefde3ea941010f7133184a1dd40c405c06841f..b7d779d08d8371082415d439b547e54cf6f6b8e5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c @@ -3,6 +3,7 @@ #include #include "lib/devcom.h" +#include "mlx5_core.h" static LIST_HEAD(devcom_list); @@ -13,7 +14,7 @@ static LIST_HEAD(devcom_list); struct mlx5_devcom_component { struct { - void *data; + void __rcu *data; } device[MLX5_DEVCOM_PORTS_SUPPORTED]; mlx5_devcom_event_handler_t handler; @@ -77,6 +78,7 @@ struct mlx5_devcom *mlx5_devcom_register_device(struct mlx5_core_dev *dev) if (MLX5_CAP_GEN(dev, num_lag_ports) != MLX5_DEVCOM_PORTS_SUPPORTED) return NULL; + mlx5_dev_list_lock(); sguid0 = mlx5_query_nic_system_image_guid(dev); list_for_each_entry(iter, &devcom_list, list) { struct mlx5_core_dev *tmp_dev = NULL; @@ -102,8 +104,10 @@ struct mlx5_devcom *mlx5_devcom_register_device(struct mlx5_core_dev *dev) if (!priv) { priv = mlx5_devcom_list_alloc(); - if (!priv) - return ERR_PTR(-ENOMEM); + if (!priv) { + devcom = ERR_PTR(-ENOMEM); + goto out; + } idx = 0; new_priv = true; @@ -112,13 +116,16 @@ struct mlx5_devcom *mlx5_devcom_register_device(struct mlx5_core_dev *dev) priv->devs[idx] = dev; devcom = mlx5_devcom_alloc(priv, idx); if (!devcom) { - kfree(priv); - return ERR_PTR(-ENOMEM); + if (new_priv) + kfree(priv); + devcom = ERR_PTR(-ENOMEM); + goto out; } if (new_priv) list_add(&priv->list, &devcom_list); - +out: + mlx5_dev_list_unlock(); return devcom; } @@ -131,6 +138,7 @@ void mlx5_devcom_unregister_device(struct mlx5_devcom *devcom) if (IS_ERR_OR_NULL(devcom)) return; + mlx5_dev_list_lock(); priv = devcom->priv; priv->devs[devcom->idx] = NULL; @@ -141,10 +149,12 @@ void mlx5_devcom_unregister_device(struct mlx5_devcom *devcom) break; if (i != MLX5_DEVCOM_PORTS_SUPPORTED) - return; + goto out; list_del(&priv->list); kfree(priv); +out: + mlx5_dev_list_unlock(); } void mlx5_devcom_register_component(struct mlx5_devcom *devcom, @@ -162,7 +172,7 @@ void mlx5_devcom_register_component(struct mlx5_devcom *devcom, comp = &devcom->priv->components[id]; down_write(&comp->sem); comp->handler = handler; - comp->device[devcom->idx].data = data; + rcu_assign_pointer(comp->device[devcom->idx].data, data); up_write(&comp->sem); } @@ -176,8 +186,9 @@ void mlx5_devcom_unregister_component(struct mlx5_devcom *devcom, comp = &devcom->priv->components[id]; down_write(&comp->sem); - comp->device[devcom->idx].data = NULL; + RCU_INIT_POINTER(comp->device[devcom->idx].data, NULL); up_write(&comp->sem); + synchronize_rcu(); } int mlx5_devcom_send_event(struct mlx5_devcom *devcom, @@ -193,12 +204,15 @@ int mlx5_devcom_send_event(struct mlx5_devcom *devcom, comp = &devcom->priv->components[id]; down_write(&comp->sem); - for (i = 0; i < MLX5_DEVCOM_PORTS_SUPPORTED; i++) - if (i != devcom->idx && comp->device[i].data) { - err = comp->handler(event, comp->device[i].data, - event_data); + for (i = 0; i < MLX5_DEVCOM_PORTS_SUPPORTED; i++) { + void *data = rcu_dereference_protected(comp->device[i].data, + lockdep_is_held(&comp->sem)); + + if (i != devcom->idx && data) { + err = comp->handler(event, data, event_data); break; } + } up_write(&comp->sem); return err; @@ -213,7 +227,7 @@ void mlx5_devcom_set_paired(struct mlx5_devcom *devcom, comp = &devcom->priv->components[id]; WARN_ON(!rwsem_is_locked(&comp->sem)); - comp->paired = paired; + WRITE_ONCE(comp->paired, paired); } bool mlx5_devcom_is_paired(struct mlx5_devcom *devcom, @@ -222,7 +236,7 @@ bool mlx5_devcom_is_paired(struct mlx5_devcom *devcom, if (IS_ERR_OR_NULL(devcom)) return false; - return devcom->priv->components[id].paired; + return READ_ONCE(devcom->priv->components[id].paired); } void *mlx5_devcom_get_peer_data(struct mlx5_devcom *devcom, @@ -236,7 +250,7 @@ void *mlx5_devcom_get_peer_data(struct mlx5_devcom *devcom, comp = &devcom->priv->components[id]; down_read(&comp->sem); - if (!comp->paired) { + if (!READ_ONCE(comp->paired)) { up_read(&comp->sem); return NULL; } @@ -245,7 +259,29 @@ void *mlx5_devcom_get_peer_data(struct mlx5_devcom *devcom, if (i != devcom->idx) break; - return comp->device[i].data; + return rcu_dereference_protected(comp->device[i].data, lockdep_is_held(&comp->sem)); +} + +void *mlx5_devcom_get_peer_data_rcu(struct mlx5_devcom *devcom, enum mlx5_devcom_components id) +{ + struct mlx5_devcom_component *comp; + int i; + + if (IS_ERR_OR_NULL(devcom)) + return NULL; + + for (i = 0; i < MLX5_DEVCOM_PORTS_SUPPORTED; i++) + if (i != devcom->idx) + break; + + comp = &devcom->priv->components[id]; + /* This can change concurrently, however 'data' pointer will remain + * valid for the duration of RCU read section. + */ + if (!READ_ONCE(comp->paired)) + return NULL; + + return rcu_dereference(comp->device[i].data); } void mlx5_devcom_release_peer_data(struct mlx5_devcom *devcom, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h index 94313c18bb6470d593e8e15e161188a279b58401..9a496f4722dadd690388a98c5cba59f0cfa9371b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h @@ -41,6 +41,7 @@ bool mlx5_devcom_is_paired(struct mlx5_devcom *devcom, void *mlx5_devcom_get_peer_data(struct mlx5_devcom *devcom, enum mlx5_devcom_components id); +void *mlx5_devcom_get_peer_data_rcu(struct mlx5_devcom *devcom, enum mlx5_devcom_components id); void mlx5_devcom_release_peer_data(struct mlx5_devcom *devcom, enum mlx5_devcom_components id); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 995eb2d5ace0b2681718920d7ebdb09c94625156..d6ee016deae176aa2ac54401304a5e8e1dda4c8f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -923,7 +923,6 @@ static int mlx5_pci_init(struct mlx5_core_dev *dev, struct pci_dev *pdev, } mlx5_pci_vsc_init(dev); - dev->caps.embedded_cpu = mlx5_read_embedded_cpu(dev); return 0; err_clr_master: @@ -1049,7 +1048,7 @@ static int mlx5_init_once(struct mlx5_core_dev *dev) dev->dm = mlx5_dm_create(dev); if (IS_ERR(dev->dm)) - mlx5_core_warn(dev, "Failed to init device memory%d\n", err); + mlx5_core_warn(dev, "Failed to init device memory %ld\n", PTR_ERR(dev->dm)); dev->tracer = mlx5_fw_tracer_create(dev); dev->hv_vhca = mlx5_hv_vhca_create(dev); @@ -1155,6 +1154,7 @@ static int mlx5_function_setup(struct mlx5_core_dev *dev, bool boot, u64 timeout goto err_cmd_cleanup; } + dev->caps.embedded_cpu = mlx5_read_embedded_cpu(dev); mlx5_cmd_set_state(dev, MLX5_CMDIF_STATE_UP); mlx5_start_health_poll(dev); @@ -1802,15 +1802,16 @@ static void remove_one(struct pci_dev *pdev) struct devlink *devlink = priv_to_devlink(dev); set_bit(MLX5_BREAK_FW_WAIT, &dev->intf_state); - /* mlx5_drain_fw_reset() is using devlink APIs. Hence, we must drain - * fw_reset before unregistering the devlink. + /* mlx5_drain_fw_reset() and mlx5_drain_health_wq() are using + * devlink notify APIs. + * Hence, we must drain them before unregistering the devlink. */ mlx5_drain_fw_reset(dev); + mlx5_drain_health_wq(dev); devlink_unregister(devlink); mlx5_sriov_disable(pdev); mlx5_thermal_uninit(dev); mlx5_crdump_disable(dev); - mlx5_drain_health_wq(dev); mlx5_uninit_one(dev); mlx5_pci_close(dev); mlx5_mdev_uninit(dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h index 1d879374acaa4e2ffceabc6acf5bd4c5896ac31c..229520405d4ae9d74abc7e844e2bea82184c6bf0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h @@ -276,18 +276,6 @@ static inline bool mlx5_sriov_is_enabled(struct mlx5_core_dev *dev) return pci_num_vf(dev->pdev) ? true : false; } -static inline int mlx5_lag_is_lacp_owner(struct mlx5_core_dev *dev) -{ - /* LACP owner conditions: - * 1) Function is physical. - * 2) LAG is supported by FW. - * 3) LAG is managed by driver (currently the only option). - */ - return MLX5_CAP_GEN(dev, vport_group_manager) && - (MLX5_CAP_GEN(dev, num_lag_ports) > 1) && - MLX5_CAP_GEN(dev, lag_master); -} - int mlx5_rescan_drivers_locked(struct mlx5_core_dev *dev); static inline int mlx5_rescan_drivers(struct mlx5_core_dev *dev) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h index efd0c299c5c73f54a3e8356aff4ea30a57459dec..aa403a5ea34ed22d30f987c84260423c444b37ec 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h @@ -15,6 +15,7 @@ int mlx5_irq_table_init(struct mlx5_core_dev *dev); void mlx5_irq_table_cleanup(struct mlx5_core_dev *dev); int mlx5_irq_table_create(struct mlx5_core_dev *dev); void mlx5_irq_table_destroy(struct mlx5_core_dev *dev); +void mlx5_irq_table_free_irqs(struct mlx5_core_dev *dev); int mlx5_irq_table_get_num_comp(struct mlx5_irq_table *table); int mlx5_irq_table_get_sfs_vec(struct mlx5_irq_table *table); struct mlx5_irq_table *mlx5_irq_table_get(struct mlx5_core_dev *dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mr.c b/drivers/net/ethernet/mellanox/mlx5/core/mr.c index 9d735c343a3b8c0df0a1a90ceacab0e9439d86b7..678f0be8137520bc91e3d6fcc322529e20b1062a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mr.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/mr.c @@ -32,6 +32,7 @@ #include #include +#include #include "mlx5_core.h" int mlx5_core_create_mkey(struct mlx5_core_dev *dev, u32 *mkey, u32 *in, @@ -122,3 +123,23 @@ int mlx5_core_destroy_psv(struct mlx5_core_dev *dev, int psv_num) return mlx5_cmd_exec_in(dev, destroy_psv, in); } EXPORT_SYMBOL(mlx5_core_destroy_psv); + +__be32 mlx5_core_get_terminate_scatter_list_mkey(struct mlx5_core_dev *dev) +{ + u32 out[MLX5_ST_SZ_DW(query_special_contexts_out)] = {}; + u32 in[MLX5_ST_SZ_DW(query_special_contexts_in)] = {}; + u32 mkey; + + if (!MLX5_CAP_GEN(dev, terminate_scatter_list_mkey)) + return MLX5_TERMINATE_SCATTER_LIST_LKEY; + + MLX5_SET(query_special_contexts_in, in, opcode, + MLX5_CMD_OP_QUERY_SPECIAL_CONTEXTS); + if (mlx5_cmd_exec_inout(dev, query_special_contexts, in, out)) + return MLX5_TERMINATE_SCATTER_LIST_LKEY; + + mkey = MLX5_GET(query_special_contexts_out, out, + terminate_scatter_list_mkey); + return cpu_to_be32(mkey); +} +EXPORT_SYMBOL(mlx5_core_get_terminate_scatter_list_mkey); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c index 2245d3b2f393bcd60affd0963d3340016502570a..843da89a903501dd8964bf5ba7de7967113fd5aa 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c @@ -32,6 +32,7 @@ struct mlx5_irq { struct mlx5_irq_pool *pool; int refcount; struct msi_map map; + u32 pool_index; }; struct mlx5_irq_table { @@ -132,7 +133,7 @@ static void irq_release(struct mlx5_irq *irq) struct cpu_rmap *rmap; #endif - xa_erase(&pool->irqs, irq->map.index); + xa_erase(&pool->irqs, irq->pool_index); /* free_irq requires that affinity_hint and rmap will be cleared before * calling it. To satisfy this requirement, we call * irq_cpu_rmap_remove() to remove the notifier @@ -140,7 +141,7 @@ static void irq_release(struct mlx5_irq *irq) irq_update_affinity_hint(irq->map.virq, NULL); #ifdef CONFIG_RFS_ACCEL rmap = mlx5_eq_table_get_rmap(pool->dev); - if (rmap && irq->map.index) + if (rmap) irq_cpu_rmap_remove(rmap, irq->map.virq); #endif @@ -231,12 +232,13 @@ struct mlx5_irq *mlx5_irq_alloc(struct mlx5_irq_pool *pool, int i, if (!irq) return ERR_PTR(-ENOMEM); if (!i || !pci_msix_can_alloc_dyn(dev->pdev)) { - /* The vector at index 0 was already allocated. - * Just get the irq number. If dynamic irq is not supported - * vectors have also been allocated. + /* The vector at index 0 is always statically allocated. If + * dynamic irq is not supported all vectors are statically + * allocated. In both cases just get the irq number and set + * the index. */ irq->map.virq = pci_irq_vector(dev->pdev, i); - irq->map.index = 0; + irq->map.index = i; } else { irq->map = pci_msix_alloc_irq_at(dev->pdev, MSI_ANY_INDEX, af_desc); if (!irq->map.virq) { @@ -276,11 +278,11 @@ struct mlx5_irq *mlx5_irq_alloc(struct mlx5_irq_pool *pool, int i, } irq->pool = pool; irq->refcount = 1; - irq->map.index = i; - err = xa_err(xa_store(&pool->irqs, irq->map.index, irq, GFP_KERNEL)); + irq->pool_index = i; + err = xa_err(xa_store(&pool->irqs, irq->pool_index, irq, GFP_KERNEL)); if (err) { mlx5_core_err(dev, "Failed to alloc xa entry for irq(%u). err = %d\n", - irq->map.index, err); + irq->pool_index, err); goto err_xa; } return irq; @@ -567,13 +569,13 @@ int mlx5_irqs_request_vectors(struct mlx5_core_dev *dev, u16 *cpus, int nirqs, struct mlx5_irq *irq; int i; - af_desc.is_managed = 1; + af_desc.is_managed = false; for (i = 0; i < nirqs; i++) { + cpumask_clear(&af_desc.mask); cpumask_set_cpu(cpus[i], &af_desc.mask); irq = mlx5_irq_request(dev, i + 1, &af_desc, rmap); if (IS_ERR(irq)) break; - cpumask_clear(&af_desc.mask); irqs[i] = irq; } @@ -691,6 +693,24 @@ static void irq_pools_destroy(struct mlx5_irq_table *table) irq_pool_free(table->pcif_pool); } +static void mlx5_irq_pool_free_irqs(struct mlx5_irq_pool *pool) +{ + struct mlx5_irq *irq; + unsigned long index; + + xa_for_each(&pool->irqs, index, irq) + free_irq(irq->map.virq, &irq->nh); +} + +static void mlx5_irq_pools_free_irqs(struct mlx5_irq_table *table) +{ + if (table->sf_ctrl_pool) { + mlx5_irq_pool_free_irqs(table->sf_comp_pool); + mlx5_irq_pool_free_irqs(table->sf_ctrl_pool); + } + mlx5_irq_pool_free_irqs(table->pcif_pool); +} + /* irq_table API */ int mlx5_irq_table_init(struct mlx5_core_dev *dev) @@ -774,6 +794,17 @@ void mlx5_irq_table_destroy(struct mlx5_core_dev *dev) pci_free_irq_vectors(dev->pdev); } +void mlx5_irq_table_free_irqs(struct mlx5_core_dev *dev) +{ + struct mlx5_irq_table *table = dev->priv.irq_table; + + if (mlx5_core_is_sf(dev)) + return; + + mlx5_irq_pools_free_irqs(table); + pci_free_irq_vectors(dev->pdev); +} + int mlx5_irq_table_get_sfs_vec(struct mlx5_irq_table *table) { if (table->sf_comp_pool) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c index e2f26d0bc61541f98d637d0c45f9c710c4325230..0692363cf80e4d8f2d31e4133faab797a03abf10 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/driver.c @@ -63,6 +63,7 @@ static void mlx5_sf_dev_remove(struct auxiliary_device *adev) struct mlx5_sf_dev *sf_dev = container_of(adev, struct mlx5_sf_dev, adev); struct devlink *devlink = priv_to_devlink(sf_dev->mdev); + mlx5_drain_health_wq(sf_dev->mdev); devlink_unregister(devlink); mlx5_uninit_one(sf_dev->mdev); iounmap(sf_dev->mdev->iseg); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c index 3835ba3f4ddac6a3b82aab154121d15b8ec4fb6a..1aa525e509f109277b03d53c11f51e18886fde95 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c @@ -117,6 +117,8 @@ int mlx5dr_cmd_query_device(struct mlx5_core_dev *mdev, caps->gvmi = MLX5_CAP_GEN(mdev, vhca_id); caps->flex_protocols = MLX5_CAP_GEN(mdev, flex_parser_protocols); caps->sw_format_ver = MLX5_CAP_GEN(mdev, steering_format_version); + caps->roce_caps.fl_rc_qp_when_roce_disabled = + MLX5_CAP_GEN(mdev, fl_rc_qp_when_roce_disabled); if (MLX5_CAP_GEN(mdev, roce)) { err = dr_cmd_query_nic_vport_roce_en(mdev, 0, &roce_en); @@ -124,7 +126,7 @@ int mlx5dr_cmd_query_device(struct mlx5_core_dev *mdev, return err; caps->roce_caps.roce_en = roce_en; - caps->roce_caps.fl_rc_qp_when_roce_disabled = + caps->roce_caps.fl_rc_qp_when_roce_disabled |= MLX5_CAP_ROCE(mdev, fl_rc_qp_when_roce_disabled); caps->roce_caps.fl_rc_qp_when_roce_enabled = MLX5_CAP_ROCE(mdev, fl_rc_qp_when_roce_enabled); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ptrn.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ptrn.c index 13e06a6a6b22af75704204eae5243f0288b71049..d6947fe13d560d141e8e2efa9c7be49a485abcc0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ptrn.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ptrn.c @@ -213,6 +213,8 @@ struct mlx5dr_ptrn_mgr *mlx5dr_ptrn_mgr_create(struct mlx5dr_domain *dmn) } INIT_LIST_HEAD(&mgr->ptrn_list); + mutex_init(&mgr->modify_hdr_mutex); + return mgr; free_mgr: @@ -237,5 +239,6 @@ void mlx5dr_ptrn_mgr_destroy(struct mlx5dr_ptrn_mgr *mgr) } mlx5dr_icm_pool_destroy(mgr->ptrn_icm_pool); + mutex_destroy(&mgr->modify_hdr_mutex); kfree(mgr); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c index 9413aaf51251b798cc5fc70c861599041efbd687..e94fbb015efad7e1988e738980d2e1378e4d4d50 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c @@ -15,7 +15,8 @@ static u32 dr_ste_crc32_calc(const void *input_data, size_t length) { u32 crc = crc32(0, input_data, length); - return (__force u32)htonl(crc); + return (__force u32)((crc >> 24) & 0xff) | ((crc << 8) & 0xff0000) | + ((crc >> 8) & 0xff00) | ((crc << 24) & 0xff000000); } bool mlx5dr_ste_supp_ttl_cs_recalc(struct mlx5dr_cmd_caps *caps) diff --git a/drivers/net/ethernet/mellanox/mlxbf_gige/mlxbf_gige_rx.c b/drivers/net/ethernet/mellanox/mlxbf_gige/mlxbf_gige_rx.c index afa3b92a6905f691dbd63649ccf4190f20e9bd48..0d5a41a2ae010914b50fac74b5c719c6aa3fa2cd 100644 --- a/drivers/net/ethernet/mellanox/mlxbf_gige/mlxbf_gige_rx.c +++ b/drivers/net/ethernet/mellanox/mlxbf_gige/mlxbf_gige_rx.c @@ -245,12 +245,6 @@ static bool mlxbf_gige_rx_packet(struct mlxbf_gige *priv, int *rx_pkts) skb = priv->rx_skb[rx_pi_rem]; - skb_put(skb, datalen); - - skb->ip_summed = CHECKSUM_NONE; /* device did not checksum packet */ - - skb->protocol = eth_type_trans(skb, netdev); - /* Alloc another RX SKB for this same index */ rx_skb = mlxbf_gige_alloc_skb(priv, MLXBF_GIGE_DEFAULT_BUF_SZ, &rx_buf_dma, DMA_FROM_DEVICE); @@ -259,6 +253,13 @@ static bool mlxbf_gige_rx_packet(struct mlxbf_gige *priv, int *rx_pkts) priv->rx_skb[rx_pi_rem] = rx_skb; dma_unmap_single(priv->dev, *rx_wqe_addr, MLXBF_GIGE_DEFAULT_BUF_SZ, DMA_FROM_DEVICE); + + skb_put(skb, datalen); + + skb->ip_summed = CHECKSUM_NONE; /* device did not checksum packet */ + + skb->protocol = eth_type_trans(skb, netdev); + *rx_wqe_addr = rx_buf_dma; } else if (rx_cqe & MLXBF_GIGE_RX_CQE_PKT_STATUS_MAC_ERR) { priv->stats.rx_mac_errors++; diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_main.c b/drivers/net/ethernet/microchip/lan966x/lan966x_main.c index 2b6e046e1d10bbc7fcec9f334e344eec90bb7d11..ee2698698d71a43bb71256a085f91daf603ac095 100644 --- a/drivers/net/ethernet/microchip/lan966x/lan966x_main.c +++ b/drivers/net/ethernet/microchip/lan966x/lan966x_main.c @@ -1039,6 +1039,16 @@ static int lan966x_reset_switch(struct lan966x *lan966x) reset_control_reset(switch_reset); + /* Don't reinitialize the switch core, if it is already initialized. In + * case it is initialized twice, some pointers inside the queue system + * in HW will get corrupted and then after a while the queue system gets + * full and no traffic is passing through the switch. The issue is seen + * when loading and unloading the driver and sending traffic through the + * switch. + */ + if (lan_rd(lan966x, SYS_RESET_CFG) & SYS_RESET_CFG_CORE_ENA) + return 0; + lan_wr(SYS_RESET_CFG_CORE_ENA_SET(0), lan966x, SYS_RESET_CFG); lan_wr(SYS_RAM_INIT_RAM_INIT_SET(1), lan966x, SYS_RAM_INIT); ret = readx_poll_timeout(lan966x_ram_init, lan966x, diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c index 06d6292e09b332d9261806784738651e29d4dd8a..d907727c7b7a5324b66e319683d4d611638cd09c 100644 --- a/drivers/net/ethernet/microsoft/mana/mana_en.c +++ b/drivers/net/ethernet/microsoft/mana/mana_en.c @@ -1279,8 +1279,6 @@ static void mana_poll_tx_cq(struct mana_cq *cq) if (comp_read < 1) return; - apc->eth_stats.tx_cqes = comp_read; - for (i = 0; i < comp_read; i++) { struct mana_tx_comp_oob *cqe_oob; @@ -1363,8 +1361,6 @@ static void mana_poll_tx_cq(struct mana_cq *cq) WARN_ON_ONCE(1); cq->work_done = pkt_transmitted; - - apc->eth_stats.tx_cqes -= pkt_transmitted; } static void mana_post_pkt_rxq(struct mana_rxq *rxq) @@ -1626,15 +1622,11 @@ static void mana_poll_rx_cq(struct mana_cq *cq) { struct gdma_comp *comp = cq->gdma_comp_buf; struct mana_rxq *rxq = cq->rxq; - struct mana_port_context *apc; int comp_read, i; - apc = netdev_priv(rxq->ndev); - comp_read = mana_gd_poll_cq(cq->gdma_cq, comp, CQE_POLLING_BUFFER); WARN_ON_ONCE(comp_read > CQE_POLLING_BUFFER); - apc->eth_stats.rx_cqes = comp_read; rxq->xdp_flush = false; for (i = 0; i < comp_read; i++) { @@ -1646,8 +1638,6 @@ static void mana_poll_rx_cq(struct mana_cq *cq) return; mana_process_rx_cqe(rxq, cq, &comp[i]); - - apc->eth_stats.rx_cqes--; } if (rxq->xdp_flush) diff --git a/drivers/net/ethernet/microsoft/mana/mana_ethtool.c b/drivers/net/ethernet/microsoft/mana/mana_ethtool.c index a64c81410dc14cfcd44d263f0e79cc3337805ca4..0dc78679f62096ebf79da9d9af12b0519acb1e95 100644 --- a/drivers/net/ethernet/microsoft/mana/mana_ethtool.c +++ b/drivers/net/ethernet/microsoft/mana/mana_ethtool.c @@ -13,11 +13,9 @@ static const struct { } mana_eth_stats[] = { {"stop_queue", offsetof(struct mana_ethtool_stats, stop_queue)}, {"wake_queue", offsetof(struct mana_ethtool_stats, wake_queue)}, - {"tx_cqes", offsetof(struct mana_ethtool_stats, tx_cqes)}, {"tx_cq_err", offsetof(struct mana_ethtool_stats, tx_cqe_err)}, {"tx_cqe_unknown_type", offsetof(struct mana_ethtool_stats, tx_cqe_unknown_type)}, - {"rx_cqes", offsetof(struct mana_ethtool_stats, rx_cqes)}, {"rx_coalesced_err", offsetof(struct mana_ethtool_stats, rx_coalesced_err)}, {"rx_cqe_unknown_type", offsetof(struct mana_ethtool_stats, diff --git a/drivers/net/ethernet/netronome/nfp/nic/main.h b/drivers/net/ethernet/netronome/nfp/nic/main.h index 094374df42b8c41513f968e44455f3e286fdcd84..38b8b10b03cd3b5c94a52925751e3171a5601a7e 100644 --- a/drivers/net/ethernet/netronome/nfp/nic/main.h +++ b/drivers/net/ethernet/netronome/nfp/nic/main.h @@ -8,7 +8,7 @@ #ifdef CONFIG_DCB /* DCB feature definitions */ -#define NFP_NET_MAX_DSCP 4 +#define NFP_NET_MAX_DSCP 64 #define NFP_NET_MAX_TC IEEE_8021QAZ_MAX_TCS #define NFP_NET_MAX_PRIO 8 #define NFP_DCB_CFG_STRIDE 256 diff --git a/drivers/net/ethernet/nvidia/forcedeth.c b/drivers/net/ethernet/nvidia/forcedeth.c index 0605d1ee490dd0599dc6d69b31a52955d71ef9c8..7a549b834e970a5ac40330513e3220466d78acef 100644 --- a/drivers/net/ethernet/nvidia/forcedeth.c +++ b/drivers/net/ethernet/nvidia/forcedeth.c @@ -6138,6 +6138,7 @@ static int nv_probe(struct pci_dev *pci_dev, const struct pci_device_id *id) return 0; out_error: + nv_mgmt_release_sema(dev); if (phystate_orig) writel(phystate|NVREG_ADAPTCTL_RUNNING, base + NvRegAdapterControl); out_freering: diff --git a/drivers/net/ethernet/qlogic/qed/qed_l2.c b/drivers/net/ethernet/qlogic/qed/qed_l2.c index 2edd6bf64a3cc29ea1934eb10c7bc1b23edb767a..7776d3bdd459a54fe3f93c0166f824c6878e50e6 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_l2.c +++ b/drivers/net/ethernet/qlogic/qed/qed_l2.c @@ -1903,7 +1903,7 @@ void qed_get_vport_stats(struct qed_dev *cdev, struct qed_eth_stats *stats) { u32 i; - if (!cdev) { + if (!cdev || cdev->recov_in_prog) { memset(stats, 0, sizeof(*stats)); return; } diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h index f9931ecb7baa252c9e7b05c57461e29da9f60313..4d83ceebdc498dfd700531f822721edea96ed6a5 100644 --- a/drivers/net/ethernet/qlogic/qede/qede.h +++ b/drivers/net/ethernet/qlogic/qede/qede.h @@ -269,6 +269,10 @@ struct qede_dev { #define QEDE_ERR_WARN 3 struct qede_dump_info dump_info; + struct delayed_work periodic_task; + unsigned long stats_coal_ticks; + u32 stats_coal_usecs; + spinlock_t stats_lock; /* lock for vport stats access */ }; enum QEDE_STATE { diff --git a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c index 374a86b875a3e5db52182258c642dd97a4fc4d3c..95820cf1cd6cc7b3ba95463adb3a1a9b41f33b1e 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c +++ b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c @@ -429,6 +429,8 @@ static void qede_get_ethtool_stats(struct net_device *dev, } } + spin_lock(&edev->stats_lock); + for (i = 0; i < QEDE_NUM_STATS; i++) { if (qede_is_irrelevant_stat(edev, i)) continue; @@ -438,6 +440,8 @@ static void qede_get_ethtool_stats(struct net_device *dev, buf++; } + spin_unlock(&edev->stats_lock); + __qede_unlock(edev); } @@ -829,6 +833,7 @@ static int qede_get_coalesce(struct net_device *dev, coal->rx_coalesce_usecs = rx_coal; coal->tx_coalesce_usecs = tx_coal; + coal->stats_block_coalesce_usecs = edev->stats_coal_usecs; return rc; } @@ -842,6 +847,19 @@ int qede_set_coalesce(struct net_device *dev, struct ethtool_coalesce *coal, int i, rc = 0; u16 rxc, txc; + if (edev->stats_coal_usecs != coal->stats_block_coalesce_usecs) { + edev->stats_coal_usecs = coal->stats_block_coalesce_usecs; + if (edev->stats_coal_usecs) { + edev->stats_coal_ticks = usecs_to_jiffies(edev->stats_coal_usecs); + schedule_delayed_work(&edev->periodic_task, 0); + + DP_INFO(edev, "Configured stats coal ticks=%lu jiffies\n", + edev->stats_coal_ticks); + } else { + cancel_delayed_work_sync(&edev->periodic_task); + } + } + if (!netif_running(dev)) { DP_INFO(edev, "Interface is down\n"); return -EINVAL; @@ -2252,7 +2270,8 @@ static int qede_get_per_coalesce(struct net_device *dev, } static const struct ethtool_ops qede_ethtool_ops = { - .supported_coalesce_params = ETHTOOL_COALESCE_USECS, + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_STATS_BLOCK_USECS, .get_link_ksettings = qede_get_link_ksettings, .set_link_ksettings = qede_set_link_ksettings, .get_drvinfo = qede_get_drvinfo, @@ -2303,7 +2322,8 @@ static const struct ethtool_ops qede_ethtool_ops = { }; static const struct ethtool_ops qede_vf_ethtool_ops = { - .supported_coalesce_params = ETHTOOL_COALESCE_USECS, + .supported_coalesce_params = ETHTOOL_COALESCE_USECS | + ETHTOOL_COALESCE_STATS_BLOCK_USECS, .get_link_ksettings = qede_get_link_ksettings, .get_drvinfo = qede_get_drvinfo, .get_msglevel = qede_get_msglevel, diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index 4c6c685820e3f0dc66d6375f0a5242b59cbbd3d2..4b004a7281903fe72e351ee2a5a22bc1c52a731e 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -307,6 +307,8 @@ void qede_fill_by_demand_stats(struct qede_dev *edev) edev->ops->get_vport_stats(edev->cdev, &stats); + spin_lock(&edev->stats_lock); + p_common->no_buff_discards = stats.common.no_buff_discards; p_common->packet_too_big_discard = stats.common.packet_too_big_discard; p_common->ttl0_discard = stats.common.ttl0_discard; @@ -404,6 +406,8 @@ void qede_fill_by_demand_stats(struct qede_dev *edev) p_ah->tx_1519_to_max_byte_packets = stats.ah.tx_1519_to_max_byte_packets; } + + spin_unlock(&edev->stats_lock); } static void qede_get_stats64(struct net_device *dev, @@ -412,9 +416,10 @@ static void qede_get_stats64(struct net_device *dev, struct qede_dev *edev = netdev_priv(dev); struct qede_stats_common *p_common; - qede_fill_by_demand_stats(edev); p_common = &edev->stats.common; + spin_lock(&edev->stats_lock); + stats->rx_packets = p_common->rx_ucast_pkts + p_common->rx_mcast_pkts + p_common->rx_bcast_pkts; stats->tx_packets = p_common->tx_ucast_pkts + p_common->tx_mcast_pkts + @@ -434,6 +439,8 @@ static void qede_get_stats64(struct net_device *dev, stats->collisions = edev->stats.bb.tx_total_collisions; stats->rx_crc_errors = p_common->rx_crc_errors; stats->rx_frame_errors = p_common->rx_align_errors; + + spin_unlock(&edev->stats_lock); } #ifdef CONFIG_QED_SRIOV @@ -1063,6 +1070,23 @@ static void qede_unlock(struct qede_dev *edev) rtnl_unlock(); } +static void qede_periodic_task(struct work_struct *work) +{ + struct qede_dev *edev = container_of(work, struct qede_dev, + periodic_task.work); + + qede_fill_by_demand_stats(edev); + schedule_delayed_work(&edev->periodic_task, edev->stats_coal_ticks); +} + +static void qede_init_periodic_task(struct qede_dev *edev) +{ + INIT_DELAYED_WORK(&edev->periodic_task, qede_periodic_task); + spin_lock_init(&edev->stats_lock); + edev->stats_coal_usecs = USEC_PER_SEC; + edev->stats_coal_ticks = usecs_to_jiffies(USEC_PER_SEC); +} + static void qede_sp_task(struct work_struct *work) { struct qede_dev *edev = container_of(work, struct qede_dev, @@ -1082,6 +1106,7 @@ static void qede_sp_task(struct work_struct *work) */ if (test_and_clear_bit(QEDE_SP_RECOVERY, &edev->sp_flags)) { + cancel_delayed_work_sync(&edev->periodic_task); #ifdef CONFIG_QED_SRIOV /* SRIOV must be disabled outside the lock to avoid a deadlock. * The recovery of the active VFs is currently not supported. @@ -1272,6 +1297,7 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, */ INIT_DELAYED_WORK(&edev->sp_task, qede_sp_task); mutex_init(&edev->qede_lock); + qede_init_periodic_task(edev); rc = register_netdev(edev->ndev); if (rc) { @@ -1296,6 +1322,11 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, edev->rx_copybreak = QEDE_RX_HDR_SIZE; qede_log_probe(edev); + + /* retain user config (for example - after recovery) */ + if (edev->stats_coal_usecs) + schedule_delayed_work(&edev->periodic_task, 0); + return 0; err4: @@ -1364,6 +1395,7 @@ static void __qede_remove(struct pci_dev *pdev, enum qede_remove_mode mode) unregister_netdev(ndev); cancel_delayed_work_sync(&edev->sp_task); + cancel_delayed_work_sync(&edev->periodic_task); edev->ops->common->set_power_state(cdev, PCI_D0); diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index a7e376e7e689c016e1d7cad675b27b0821002883..4b19803a7dd011c3fa13f2c65b6f886e8a291b7d 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -616,10 +616,10 @@ struct rtl8169_private { struct work_struct work; } wk; - spinlock_t config25_lock; - spinlock_t mac_ocp_lock; + raw_spinlock_t config25_lock; + raw_spinlock_t mac_ocp_lock; - spinlock_t cfg9346_usage_lock; + raw_spinlock_t cfg9346_usage_lock; int cfg9346_usage_count; unsigned supports_gmii:1; @@ -671,20 +671,20 @@ static void rtl_lock_config_regs(struct rtl8169_private *tp) { unsigned long flags; - spin_lock_irqsave(&tp->cfg9346_usage_lock, flags); + raw_spin_lock_irqsave(&tp->cfg9346_usage_lock, flags); if (!--tp->cfg9346_usage_count) RTL_W8(tp, Cfg9346, Cfg9346_Lock); - spin_unlock_irqrestore(&tp->cfg9346_usage_lock, flags); + raw_spin_unlock_irqrestore(&tp->cfg9346_usage_lock, flags); } static void rtl_unlock_config_regs(struct rtl8169_private *tp) { unsigned long flags; - spin_lock_irqsave(&tp->cfg9346_usage_lock, flags); + raw_spin_lock_irqsave(&tp->cfg9346_usage_lock, flags); if (!tp->cfg9346_usage_count++) RTL_W8(tp, Cfg9346, Cfg9346_Unlock); - spin_unlock_irqrestore(&tp->cfg9346_usage_lock, flags); + raw_spin_unlock_irqrestore(&tp->cfg9346_usage_lock, flags); } static void rtl_pci_commit(struct rtl8169_private *tp) @@ -698,10 +698,10 @@ static void rtl_mod_config2(struct rtl8169_private *tp, u8 clear, u8 set) unsigned long flags; u8 val; - spin_lock_irqsave(&tp->config25_lock, flags); + raw_spin_lock_irqsave(&tp->config25_lock, flags); val = RTL_R8(tp, Config2); RTL_W8(tp, Config2, (val & ~clear) | set); - spin_unlock_irqrestore(&tp->config25_lock, flags); + raw_spin_unlock_irqrestore(&tp->config25_lock, flags); } static void rtl_mod_config5(struct rtl8169_private *tp, u8 clear, u8 set) @@ -709,10 +709,10 @@ static void rtl_mod_config5(struct rtl8169_private *tp, u8 clear, u8 set) unsigned long flags; u8 val; - spin_lock_irqsave(&tp->config25_lock, flags); + raw_spin_lock_irqsave(&tp->config25_lock, flags); val = RTL_R8(tp, Config5); RTL_W8(tp, Config5, (val & ~clear) | set); - spin_unlock_irqrestore(&tp->config25_lock, flags); + raw_spin_unlock_irqrestore(&tp->config25_lock, flags); } static bool rtl_is_8125(struct rtl8169_private *tp) @@ -899,9 +899,9 @@ static void r8168_mac_ocp_write(struct rtl8169_private *tp, u32 reg, u32 data) { unsigned long flags; - spin_lock_irqsave(&tp->mac_ocp_lock, flags); + raw_spin_lock_irqsave(&tp->mac_ocp_lock, flags); __r8168_mac_ocp_write(tp, reg, data); - spin_unlock_irqrestore(&tp->mac_ocp_lock, flags); + raw_spin_unlock_irqrestore(&tp->mac_ocp_lock, flags); } static u16 __r8168_mac_ocp_read(struct rtl8169_private *tp, u32 reg) @@ -919,9 +919,9 @@ static u16 r8168_mac_ocp_read(struct rtl8169_private *tp, u32 reg) unsigned long flags; u16 val; - spin_lock_irqsave(&tp->mac_ocp_lock, flags); + raw_spin_lock_irqsave(&tp->mac_ocp_lock, flags); val = __r8168_mac_ocp_read(tp, reg); - spin_unlock_irqrestore(&tp->mac_ocp_lock, flags); + raw_spin_unlock_irqrestore(&tp->mac_ocp_lock, flags); return val; } @@ -932,10 +932,10 @@ static void r8168_mac_ocp_modify(struct rtl8169_private *tp, u32 reg, u16 mask, unsigned long flags; u16 data; - spin_lock_irqsave(&tp->mac_ocp_lock, flags); + raw_spin_lock_irqsave(&tp->mac_ocp_lock, flags); data = __r8168_mac_ocp_read(tp, reg); __r8168_mac_ocp_write(tp, reg, (data & ~mask) | set); - spin_unlock_irqrestore(&tp->mac_ocp_lock, flags); + raw_spin_unlock_irqrestore(&tp->mac_ocp_lock, flags); } /* Work around a hw issue with RTL8168g PHY, the quirk disables @@ -1420,14 +1420,14 @@ static void __rtl8169_set_wol(struct rtl8169_private *tp, u32 wolopts) r8168_mac_ocp_modify(tp, 0xc0b6, BIT(0), 0); } - spin_lock_irqsave(&tp->config25_lock, flags); + raw_spin_lock_irqsave(&tp->config25_lock, flags); for (i = 0; i < tmp; i++) { options = RTL_R8(tp, cfg[i].reg) & ~cfg[i].mask; if (wolopts & cfg[i].opt) options |= cfg[i].mask; RTL_W8(tp, cfg[i].reg, options); } - spin_unlock_irqrestore(&tp->config25_lock, flags); + raw_spin_unlock_irqrestore(&tp->config25_lock, flags); switch (tp->mac_version) { case RTL_GIGA_MAC_VER_02 ... RTL_GIGA_MAC_VER_06: @@ -5179,9 +5179,9 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) tp->eee_adv = -1; tp->ocp_base = OCP_STD_PHY_BASE; - spin_lock_init(&tp->cfg9346_usage_lock); - spin_lock_init(&tp->config25_lock); - spin_lock_init(&tp->mac_ocp_lock); + raw_spin_lock_init(&tp->cfg9346_usage_lock); + raw_spin_lock_init(&tp->config25_lock); + raw_spin_lock_init(&tp->mac_ocp_lock); dev->tstats = devm_netdev_alloc_pcpu_stats(&pdev->dev, struct pcpu_sw_netstats); diff --git a/drivers/net/ethernet/renesas/rswitch.c b/drivers/net/ethernet/renesas/rswitch.c index 29afaddb598db4b32acebdfa34667365ab503c85..fa6d6202b129d083a025cdec6cf1ff9f30aaddba 100644 --- a/drivers/net/ethernet/renesas/rswitch.c +++ b/drivers/net/ethernet/renesas/rswitch.c @@ -347,17 +347,6 @@ static int rswitch_gwca_queue_alloc(struct net_device *ndev, return -ENOMEM; } -static int rswitch_gwca_ts_queue_alloc(struct rswitch_private *priv) -{ - struct rswitch_gwca_queue *gq = &priv->gwca.ts_queue; - - gq->ring_size = TS_RING_SIZE; - gq->ts_ring = dma_alloc_coherent(&priv->pdev->dev, - sizeof(struct rswitch_ts_desc) * - (gq->ring_size + 1), &gq->ring_dma, GFP_KERNEL); - return !gq->ts_ring ? -ENOMEM : 0; -} - static void rswitch_desc_set_dptr(struct rswitch_desc *desc, dma_addr_t addr) { desc->dptrl = cpu_to_le32(lower_32_bits(addr)); @@ -533,6 +522,28 @@ static void rswitch_gwca_linkfix_free(struct rswitch_private *priv) gwca->linkfix_table = NULL; } +static int rswitch_gwca_ts_queue_alloc(struct rswitch_private *priv) +{ + struct rswitch_gwca_queue *gq = &priv->gwca.ts_queue; + struct rswitch_ts_desc *desc; + + gq->ring_size = TS_RING_SIZE; + gq->ts_ring = dma_alloc_coherent(&priv->pdev->dev, + sizeof(struct rswitch_ts_desc) * + (gq->ring_size + 1), &gq->ring_dma, GFP_KERNEL); + + if (!gq->ts_ring) + return -ENOMEM; + + rswitch_gwca_ts_queue_fill(priv, 0, TS_RING_SIZE); + desc = &gq->ts_ring[gq->ring_size]; + desc->desc.die_dt = DT_LINKFIX; + rswitch_desc_set_dptr(&desc->desc, gq->ring_dma); + INIT_LIST_HEAD(&priv->gwca.ts_info_list); + + return 0; +} + static struct rswitch_gwca_queue *rswitch_gwca_get(struct rswitch_private *priv) { struct rswitch_gwca_queue *gq; @@ -1485,7 +1496,7 @@ static netdev_tx_t rswitch_start_xmit(struct sk_buff *skb, struct net_device *nd if (rswitch_get_num_cur_queues(gq) >= gq->ring_size - 1) { netif_stop_subqueue(ndev, 0); - return ret; + return NETDEV_TX_BUSY; } if (skb_put_padto(skb, ETH_ZLEN)) @@ -1780,9 +1791,6 @@ static int rswitch_init(struct rswitch_private *priv) if (err < 0) goto err_ts_queue_alloc; - rswitch_gwca_ts_queue_fill(priv, 0, TS_RING_SIZE); - INIT_LIST_HEAD(&priv->gwca.ts_info_list); - for (i = 0; i < RSWITCH_NUM_PORTS; i++) { err = rswitch_device_alloc(priv, i); if (err < 0) { diff --git a/drivers/net/ethernet/sfc/ef100_netdev.c b/drivers/net/ethernet/sfc/ef100_netdev.c index d916877b5a9ad92414c36f317ea8aae41706cecd..be395cd8770bc71db641b710b90c3d00dc2f43fe 100644 --- a/drivers/net/ethernet/sfc/ef100_netdev.c +++ b/drivers/net/ethernet/sfc/ef100_netdev.c @@ -378,7 +378,9 @@ int ef100_probe_netdev(struct efx_probe_data *probe_data) efx->net_dev = net_dev; SET_NETDEV_DEV(net_dev, &efx->pci_dev->dev); - net_dev->features |= efx->type->offload_features; + /* enable all supported features except rx-fcs and rx-all */ + net_dev->features |= efx->type->offload_features & + ~(NETIF_F_RXFCS | NETIF_F_RXALL); net_dev->hw_features |= efx->type->offload_features; net_dev->hw_enc_features |= efx->type->offload_features; net_dev->vlan_features |= NETIF_F_HW_CSUM | NETIF_F_SG | diff --git a/drivers/net/ethernet/sfc/efx_channels.c b/drivers/net/ethernet/sfc/efx_channels.c index fcea3ea809d77b315959fcc241b30e283f603ad6..41b33a75333cd40224282a958a76e241a407b566 100644 --- a/drivers/net/ethernet/sfc/efx_channels.c +++ b/drivers/net/ethernet/sfc/efx_channels.c @@ -301,6 +301,7 @@ int efx_probe_interrupts(struct efx_nic *efx) efx->tx_channel_offset = 0; efx->n_xdp_channels = 0; efx->xdp_channel_offset = efx->n_channels; + efx->xdp_txq_queues_mode = EFX_XDP_TX_QUEUES_BORROWED; rc = pci_enable_msi(efx->pci_dev); if (rc == 0) { efx_get_channel(efx, 0)->irq = efx->pci_dev->irq; @@ -322,6 +323,7 @@ int efx_probe_interrupts(struct efx_nic *efx) efx->tx_channel_offset = efx_separate_tx_channels ? 1 : 0; efx->n_xdp_channels = 0; efx->xdp_channel_offset = efx->n_channels; + efx->xdp_txq_queues_mode = EFX_XDP_TX_QUEUES_BORROWED; efx->legacy_irq = efx->pci_dev->irq; } diff --git a/drivers/net/ethernet/sfc/efx_devlink.c b/drivers/net/ethernet/sfc/efx_devlink.c index 381b805659d39a11d5ceef2cf9d67cb2fba3bc8d..ef9971cbb695d2e82fb58268522295ce853ebffe 100644 --- a/drivers/net/ethernet/sfc/efx_devlink.c +++ b/drivers/net/ethernet/sfc/efx_devlink.c @@ -171,9 +171,14 @@ static int efx_devlink_info_nvram_partition(struct efx_nic *efx, rc = efx_mcdi_nvram_metadata(efx, partition_type, NULL, version, NULL, 0); + + /* If the partition does not exist, that is not an error. */ + if (rc == -ENOENT) + return 0; + if (rc) { - netif_err(efx, drv, efx->net_dev, "mcdi nvram %s: failed\n", - version_name); + netif_err(efx, drv, efx->net_dev, "mcdi nvram %s: failed (rc=%d)\n", + version_name, rc); return rc; } @@ -187,36 +192,33 @@ static int efx_devlink_info_nvram_partition(struct efx_nic *efx, static int efx_devlink_info_stored_versions(struct efx_nic *efx, struct devlink_info_req *req) { - int rc; - - rc = efx_devlink_info_nvram_partition(efx, req, - NVRAM_PARTITION_TYPE_BUNDLE, - DEVLINK_INFO_VERSION_GENERIC_FW_BUNDLE_ID); - if (rc) - return rc; - - rc = efx_devlink_info_nvram_partition(efx, req, - NVRAM_PARTITION_TYPE_MC_FIRMWARE, - DEVLINK_INFO_VERSION_GENERIC_FW_MGMT); - if (rc) - return rc; - - rc = efx_devlink_info_nvram_partition(efx, req, - NVRAM_PARTITION_TYPE_SUC_FIRMWARE, - EFX_DEVLINK_INFO_VERSION_FW_MGMT_SUC); - if (rc) - return rc; - - rc = efx_devlink_info_nvram_partition(efx, req, - NVRAM_PARTITION_TYPE_EXPANSION_ROM, - EFX_DEVLINK_INFO_VERSION_FW_EXPROM); - if (rc) - return rc; + int err; - rc = efx_devlink_info_nvram_partition(efx, req, - NVRAM_PARTITION_TYPE_EXPANSION_UEFI, - EFX_DEVLINK_INFO_VERSION_FW_UEFI); - return rc; + /* We do not care here about the specific error but just if an error + * happened. The specific error will be reported inside the call + * through system messages, and if any error happened in any call + * below, we report it through extack. + */ + err = efx_devlink_info_nvram_partition(efx, req, + NVRAM_PARTITION_TYPE_BUNDLE, + DEVLINK_INFO_VERSION_GENERIC_FW_BUNDLE_ID); + + err |= efx_devlink_info_nvram_partition(efx, req, + NVRAM_PARTITION_TYPE_MC_FIRMWARE, + DEVLINK_INFO_VERSION_GENERIC_FW_MGMT); + + err |= efx_devlink_info_nvram_partition(efx, req, + NVRAM_PARTITION_TYPE_SUC_FIRMWARE, + EFX_DEVLINK_INFO_VERSION_FW_MGMT_SUC); + + err |= efx_devlink_info_nvram_partition(efx, req, + NVRAM_PARTITION_TYPE_EXPANSION_ROM, + EFX_DEVLINK_INFO_VERSION_FW_EXPROM); + + err |= efx_devlink_info_nvram_partition(efx, req, + NVRAM_PARTITION_TYPE_EXPANSION_UEFI, + EFX_DEVLINK_INFO_VERSION_FW_UEFI); + return err; } #define EFX_VER_FLAG(_f) \ @@ -587,27 +589,20 @@ static int efx_devlink_info_get(struct devlink *devlink, { struct efx_devlink *devlink_private = devlink_priv(devlink); struct efx_nic *efx = devlink_private->efx; - int rc; + int err; - /* Several different MCDI commands are used. We report first error - * through extack returning at that point. Specific error - * information via system messages. + /* Several different MCDI commands are used. We report if errors + * happened through extack. Specific error information via system + * messages inside the calls. */ - rc = efx_devlink_info_board_cfg(efx, req); - if (rc) { - NL_SET_ERR_MSG_MOD(extack, "Getting board info failed"); - return rc; - } - rc = efx_devlink_info_stored_versions(efx, req); - if (rc) { - NL_SET_ERR_MSG_MOD(extack, "Getting stored versions failed"); - return rc; - } - rc = efx_devlink_info_running_versions(efx, req); - if (rc) { - NL_SET_ERR_MSG_MOD(extack, "Getting running versions failed"); - return rc; - } + err = efx_devlink_info_board_cfg(efx, req); + + err |= efx_devlink_info_stored_versions(efx, req); + + err |= efx_devlink_info_running_versions(efx, req); + + if (err) + NL_SET_ERR_MSG_MOD(extack, "Errors when getting device info. Check system messages"); return 0; } diff --git a/drivers/net/ethernet/sfc/siena/efx_channels.c b/drivers/net/ethernet/sfc/siena/efx_channels.c index 06ed74994e366201ec468e2fd05ff14f474b8e7f..1776f7f8a7a90ea4043086008847ab2a52c532d8 100644 --- a/drivers/net/ethernet/sfc/siena/efx_channels.c +++ b/drivers/net/ethernet/sfc/siena/efx_channels.c @@ -302,6 +302,7 @@ int efx_siena_probe_interrupts(struct efx_nic *efx) efx->tx_channel_offset = 0; efx->n_xdp_channels = 0; efx->xdp_channel_offset = efx->n_channels; + efx->xdp_txq_queues_mode = EFX_XDP_TX_QUEUES_BORROWED; rc = pci_enable_msi(efx->pci_dev); if (rc == 0) { efx_get_channel(efx, 0)->irq = efx->pci_dev->irq; @@ -323,6 +324,7 @@ int efx_siena_probe_interrupts(struct efx_nic *efx) efx->tx_channel_offset = efx_siena_separate_tx_channels ? 1 : 0; efx->n_xdp_channels = 0; efx->xdp_channel_offset = efx->n_channels; + efx->xdp_txq_queues_mode = EFX_XDP_TX_QUEUES_BORROWED; efx->legacy_irq = efx->pci_dev->irq; } diff --git a/drivers/net/ethernet/sfc/tc.c b/drivers/net/ethernet/sfc/tc.c index 0327639a628a91adccd4229bca56c30fb75b5c78..c004443c1d58c3eebe042489fb18b1430688cd27 100644 --- a/drivers/net/ethernet/sfc/tc.c +++ b/drivers/net/ethernet/sfc/tc.c @@ -624,13 +624,12 @@ static int efx_tc_flower_replace_foreign(struct efx_nic *efx, if (!found) { /* We don't care. */ netif_dbg(efx, drv, efx->net_dev, "Ignoring foreign filter that doesn't egdev us\n"); - rc = -EOPNOTSUPP; - goto release; + return -EOPNOTSUPP; } rc = efx_mae_match_check_caps(efx, &match.mask, NULL); if (rc) - goto release; + return rc; if (efx_tc_match_is_encap(&match.mask)) { enum efx_encap_type type; @@ -639,8 +638,7 @@ static int efx_tc_flower_replace_foreign(struct efx_nic *efx, if (type == EFX_ENCAP_TYPE_NONE) { NL_SET_ERR_MSG_MOD(extack, "Egress encap match on unsupported tunnel device"); - rc = -EOPNOTSUPP; - goto release; + return -EOPNOTSUPP; } rc = efx_mae_check_encap_type_supported(efx, type); @@ -648,25 +646,24 @@ static int efx_tc_flower_replace_foreign(struct efx_nic *efx, NL_SET_ERR_MSG_FMT_MOD(extack, "Firmware reports no support for %s encap match", efx_tc_encap_type_name(type)); - goto release; + return rc; } rc = efx_tc_flower_record_encap_match(efx, &match, type, extack); if (rc) - goto release; + return rc; } else { /* This is not a tunnel decap rule, ignore it */ netif_dbg(efx, drv, efx->net_dev, "Ignoring foreign filter without encap match\n"); - rc = -EOPNOTSUPP; - goto release; + return -EOPNOTSUPP; } rule = kzalloc(sizeof(*rule), GFP_USER); if (!rule) { rc = -ENOMEM; - goto release; + goto out_free; } INIT_LIST_HEAD(&rule->acts.list); rule->cookie = tc->cookie; @@ -678,7 +675,7 @@ static int efx_tc_flower_replace_foreign(struct efx_nic *efx, "Ignoring already-offloaded rule (cookie %lx)\n", tc->cookie); rc = -EEXIST; - goto release; + goto out_free; } act = kzalloc(sizeof(*act), GFP_USER); @@ -843,6 +840,7 @@ static int efx_tc_flower_replace_foreign(struct efx_nic *efx, efx_tc_match_action_ht_params); efx_tc_free_action_set_list(efx, &rule->acts, false); } +out_free: kfree(rule); if (match.encap) efx_tc_flower_release_encap_match(efx, match.encap); @@ -899,8 +897,7 @@ static int efx_tc_flower_replace(struct efx_nic *efx, return rc; if (efx_tc_match_is_encap(&match.mask)) { NL_SET_ERR_MSG_MOD(extack, "Ingress enc_key matches not supported"); - rc = -EOPNOTSUPP; - goto release; + return -EOPNOTSUPP; } if (tc->common.chain_index) { @@ -924,9 +921,9 @@ static int efx_tc_flower_replace(struct efx_nic *efx, if (old) { netif_dbg(efx, drv, efx->net_dev, "Already offloaded rule (cookie %lx)\n", tc->cookie); - rc = -EEXIST; NL_SET_ERR_MSG_MOD(extack, "Rule already offloaded"); - goto release; + kfree(rule); + return -EEXIST; } /* Parse actions */ diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c index 16a8c361283b4f90a04472669671a80db29dfcf7..f07905f00f9888e64cc177e58edb16c220a8d1cc 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c @@ -644,7 +644,8 @@ static int qcom_ethqos_probe(struct platform_device *pdev) plat_dat->fix_mac_speed = ethqos_fix_mac_speed; plat_dat->dump_debug_regs = rgmii_dump; plat_dat->has_gmac4 = 1; - plat_dat->dwmac4_addrs = &data->dwmac4_addrs; + if (ethqos->has_emac3) + plat_dat->dwmac4_addrs = &data->dwmac4_addrs; plat_dat->pmt = 1; plat_dat->tso_en = of_property_read_bool(np, "snps,tso"); if (of_device_is_compatible(np, "qcom,qcs404-ethqos")) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 0fca81507a779acafc11faa52aafeb41326b1f8e..87510951f4e80b4f6e3aa52424439886216fe195 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -3873,7 +3873,6 @@ static int __stmmac_open(struct net_device *dev, stmmac_hw_teardown(dev); init_error: - free_dma_desc_resources(priv, &priv->dma_conf); phylink_disconnect_phy(priv->phylink); init_phy_error: pm_runtime_put(priv->device); @@ -3891,6 +3890,9 @@ static int stmmac_open(struct net_device *dev) return PTR_ERR(dma_conf); ret = __stmmac_open(dev, dma_conf); + if (ret) + free_dma_desc_resources(priv, dma_conf); + kfree(dma_conf); return ret; } @@ -5633,12 +5635,15 @@ static int stmmac_change_mtu(struct net_device *dev, int new_mtu) stmmac_release(dev); ret = __stmmac_open(dev, dma_conf); - kfree(dma_conf); if (ret) { + free_dma_desc_resources(priv, dma_conf); + kfree(dma_conf); netdev_err(priv->dev, "failed reopening the interface after MTU change\n"); return ret; } + kfree(dma_conf); + stmmac_set_rx_mode(dev); } @@ -7233,8 +7238,7 @@ int stmmac_dvr_probe(struct device *device, ndev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_RXCSUM; ndev->xdp_features = NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT | - NETDEV_XDP_ACT_XSK_ZEROCOPY | - NETDEV_XDP_ACT_NDO_XMIT; + NETDEV_XDP_ACT_XSK_ZEROCOPY; ret = stmmac_tc_init(priv, priv); if (!ret) { diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_xdp.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_xdp.c index 9d4d8c3dad0a3f769650d571ace87a4e77d292a7..aa6f16d3df649e8742584e0aacf83204acc1030c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_xdp.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_xdp.c @@ -117,6 +117,9 @@ int stmmac_xdp_set_prog(struct stmmac_priv *priv, struct bpf_prog *prog, return -EOPNOTSUPP; } + if (!prog) + xdp_features_clear_redirect_target(dev); + need_update = !!priv->xdp_prog != !!prog; if (if_running && need_update) stmmac_xdp_release(dev); @@ -131,5 +134,8 @@ int stmmac_xdp_set_prog(struct stmmac_priv *priv, struct bpf_prog *prog, if (if_running && need_update) stmmac_xdp_open(dev); + if (prog) + xdp_features_set_redirect_target(dev, false); + return 0; } diff --git a/drivers/net/ethernet/sun/cassini.c b/drivers/net/ethernet/sun/cassini.c index 4ef05bad4613c9309a81b00ba896b365a698eec7..d61dfa250feb722b67352104b272dbeebd056f0d 100644 --- a/drivers/net/ethernet/sun/cassini.c +++ b/drivers/net/ethernet/sun/cassini.c @@ -5077,6 +5077,8 @@ static int cas_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) cas_shutdown(cp); mutex_unlock(&cp->pm_mutex); + vfree(cp->fw_data); + pci_iounmap(pdev, cp->regs); diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c index 11cbcd9e2c72549f799e2058ebb925476fbdf63f..bebcfd5e6b579ef115588ae106eba0f87b15637a 100644 --- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c +++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c @@ -2068,7 +2068,7 @@ static int am65_cpsw_nuss_init_slave_ports(struct am65_cpsw_common *common) /* Initialize the Serdes PHY for the port */ ret = am65_cpsw_init_serdes_phy(dev, port_np, port); if (ret) - return ret; + goto of_node_put; port->slave.mac_only = of_property_read_bool(port_np, "ti,mac-only"); diff --git a/drivers/net/ipa/ipa_endpoint.c b/drivers/net/ipa/ipa_endpoint.c index 2ee80ed140b72b1cf1702b19bf9bad69c36c7d47..afa1d56d9095c13fe65a09f76fbc99679d9d8624 100644 --- a/drivers/net/ipa/ipa_endpoint.c +++ b/drivers/net/ipa/ipa_endpoint.c @@ -119,7 +119,7 @@ enum ipa_status_field_id { }; /* Size in bytes of an IPA packet status structure */ -#define IPA_STATUS_SIZE sizeof(__le32[4]) +#define IPA_STATUS_SIZE sizeof(__le32[8]) /* IPA status structure decoder; looks up field values for a structure */ static u32 ipa_status_extract(struct ipa *ipa, const void *data, diff --git a/drivers/net/ipvlan/ipvlan_l3s.c b/drivers/net/ipvlan/ipvlan_l3s.c index 71712ea25403dd71322f36297bb54f782d789f2c..d5b05e8032199a7bd10380547df7d5e82e667f69 100644 --- a/drivers/net/ipvlan/ipvlan_l3s.c +++ b/drivers/net/ipvlan/ipvlan_l3s.c @@ -102,6 +102,10 @@ static unsigned int ipvlan_nf_input(void *priv, struct sk_buff *skb, skb->dev = addr->master->dev; skb->skb_iif = skb->dev->ifindex; +#if IS_ENABLED(CONFIG_IPV6) + if (addr->atype == IPVL_IPV6) + IP6CB(skb)->iif = skb->dev->ifindex; +#endif len = skb->len + ETH_HLEN; ipvlan_count_rx(addr->master, len, true, false); out: diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index 3427993f94f748ccb60af1989f0cb37da2cdad7b..984dfa5d6c11cb3ae6900597edeb1612f27decc3 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -3997,17 +3997,15 @@ static int macsec_add_dev(struct net_device *dev, sci_t sci, u8 icv_len) return -ENOMEM; secy->tx_sc.stats = netdev_alloc_pcpu_stats(struct pcpu_tx_sc_stats); - if (!secy->tx_sc.stats) { - free_percpu(macsec->stats); + if (!secy->tx_sc.stats) return -ENOMEM; - } secy->tx_sc.md_dst = metadata_dst_alloc(0, METADATA_MACSEC, GFP_KERNEL); - if (!secy->tx_sc.md_dst) { - free_percpu(secy->tx_sc.stats); - free_percpu(macsec->stats); + if (!secy->tx_sc.md_dst) + /* macsec and secy percpu stats will be freed when unregistering + * net_device in macsec_free_netdev() + */ return -ENOMEM; - } if (sci == MACSEC_UNDEF_SCI) sci = dev_to_sci(dev, MACSEC_PORT_ES); diff --git a/drivers/net/mdio/mdio-i2c.c b/drivers/net/mdio/mdio-i2c.c index 1e0c206d0f2e68175e671b7d6ec65a869af6fc4f..da2001ea1f9937dd5c4e633dd3d781ef7d5f1dbb 100644 --- a/drivers/net/mdio/mdio-i2c.c +++ b/drivers/net/mdio/mdio-i2c.c @@ -291,7 +291,8 @@ static int i2c_rollball_mii_cmd(struct mii_bus *bus, int bus_addr, u8 cmd, return i2c_transfer_rollball(i2c, msgs, ARRAY_SIZE(msgs)); } -static int i2c_mii_read_rollball(struct mii_bus *bus, int phy_id, int reg) +static int i2c_mii_read_rollball(struct mii_bus *bus, int phy_id, int devad, + int reg) { u8 buf[4], res[6]; int bus_addr, ret; @@ -302,7 +303,7 @@ static int i2c_mii_read_rollball(struct mii_bus *bus, int phy_id, int reg) return 0xffff; buf[0] = ROLLBALL_DATA_ADDR; - buf[1] = (reg >> 16) & 0x1f; + buf[1] = devad; buf[2] = (reg >> 8) & 0xff; buf[3] = reg & 0xff; @@ -322,8 +323,8 @@ static int i2c_mii_read_rollball(struct mii_bus *bus, int phy_id, int reg) return val; } -static int i2c_mii_write_rollball(struct mii_bus *bus, int phy_id, int reg, - u16 val) +static int i2c_mii_write_rollball(struct mii_bus *bus, int phy_id, int devad, + int reg, u16 val) { int bus_addr, ret; u8 buf[6]; @@ -333,7 +334,7 @@ static int i2c_mii_write_rollball(struct mii_bus *bus, int phy_id, int reg, return 0; buf[0] = ROLLBALL_DATA_ADDR; - buf[1] = (reg >> 16) & 0x1f; + buf[1] = devad; buf[2] = (reg >> 8) & 0xff; buf[3] = reg & 0xff; buf[4] = val >> 8; @@ -405,8 +406,8 @@ struct mii_bus *mdio_i2c_alloc(struct device *parent, struct i2c_adapter *i2c, return ERR_PTR(ret); } - mii->read = i2c_mii_read_rollball; - mii->write = i2c_mii_write_rollball; + mii->read_c45 = i2c_mii_read_rollball; + mii->write_c45 = i2c_mii_write_rollball; break; default: mii->read = i2c_mii_read_default_c22; diff --git a/drivers/net/pcs/pcs-xpcs.c b/drivers/net/pcs/pcs-xpcs.c index f19d48c94fe0e9ccd66c3a7c451268af7d713248..72f25e778840f9330ca862e611732aabd11c340c 100644 --- a/drivers/net/pcs/pcs-xpcs.c +++ b/drivers/net/pcs/pcs-xpcs.c @@ -873,7 +873,7 @@ int xpcs_do_config(struct dw_xpcs *xpcs, phy_interface_t interface, switch (compat->an_mode) { case DW_AN_C73: - if (phylink_autoneg_inband(mode)) { + if (test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT, advertising)) { ret = xpcs_config_aneg_c73(xpcs, compat); if (ret) return ret; diff --git a/drivers/net/phy/dp83867.c b/drivers/net/phy/dp83867.c index d75f526a20a49b6a65889c41adeb1395dcea002c..76f5a2402fb060dcc6c85aa4f6f6727f9c105317 100644 --- a/drivers/net/phy/dp83867.c +++ b/drivers/net/phy/dp83867.c @@ -44,6 +44,7 @@ #define DP83867_STRAP_STS1 0x006E #define DP83867_STRAP_STS2 0x006f #define DP83867_RGMIIDCTL 0x0086 +#define DP83867_DSP_FFE_CFG 0x012c #define DP83867_RXFCFG 0x0134 #define DP83867_RXFPMD1 0x0136 #define DP83867_RXFPMD2 0x0137 @@ -941,8 +942,27 @@ static int dp83867_phy_reset(struct phy_device *phydev) usleep_range(10, 20); - return phy_modify(phydev, MII_DP83867_PHYCTRL, + err = phy_modify(phydev, MII_DP83867_PHYCTRL, DP83867_PHYCR_FORCE_LINK_GOOD, 0); + if (err < 0) + return err; + + /* Configure the DSP Feedforward Equalizer Configuration register to + * improve short cable (< 1 meter) performance. This will not affect + * long cable performance. + */ + err = phy_write_mmd(phydev, DP83867_DEVADDR, DP83867_DSP_FFE_CFG, + 0x0e81); + if (err < 0) + return err; + + err = phy_write(phydev, DP83867_CTRL, DP83867_SW_RESTART); + if (err < 0) + return err; + + usleep_range(10, 20); + + return 0; } static void dp83867_link_change_notify(struct phy_device *phydev) diff --git a/drivers/net/phy/mscc/mscc.h b/drivers/net/phy/mscc/mscc.h index a50235fdf7d99d2f5db21dc8e002a0830223ae08..defe5cc6d4fcef4754edfb0aa5a1d6e562e0b282 100644 --- a/drivers/net/phy/mscc/mscc.h +++ b/drivers/net/phy/mscc/mscc.h @@ -179,6 +179,7 @@ enum rgmii_clock_delay { #define VSC8502_RGMII_CNTL 20 #define VSC8502_RGMII_RX_DELAY_MASK 0x0070 #define VSC8502_RGMII_TX_DELAY_MASK 0x0007 +#define VSC8502_RGMII_RX_CLK_DISABLE 0x0800 #define MSCC_PHY_WOL_LOWER_MAC_ADDR 21 #define MSCC_PHY_WOL_MID_MAC_ADDR 22 @@ -276,6 +277,7 @@ enum rgmii_clock_delay { /* Microsemi PHY ID's * Code assumes lowest nibble is 0 */ +#define PHY_ID_VSC8501 0x00070530 #define PHY_ID_VSC8502 0x00070630 #define PHY_ID_VSC8504 0x000704c0 #define PHY_ID_VSC8514 0x00070670 diff --git a/drivers/net/phy/mscc/mscc_main.c b/drivers/net/phy/mscc/mscc_main.c index 62bf99e45af16d86ff80428af812e34183888715..28df8a2e4230f3ffd6f559e8ac9521cedc7d0100 100644 --- a/drivers/net/phy/mscc/mscc_main.c +++ b/drivers/net/phy/mscc/mscc_main.c @@ -519,16 +519,27 @@ static int vsc85xx_mac_if_set(struct phy_device *phydev, * * 2.0 ns (which causes the data to be sampled at exactly half way between * clock transitions at 1000 Mbps) if delays should be enabled */ -static int vsc85xx_rgmii_set_skews(struct phy_device *phydev, u32 rgmii_cntl, - u16 rgmii_rx_delay_mask, - u16 rgmii_tx_delay_mask) +static int vsc85xx_update_rgmii_cntl(struct phy_device *phydev, u32 rgmii_cntl, + u16 rgmii_rx_delay_mask, + u16 rgmii_tx_delay_mask) { u16 rgmii_rx_delay_pos = ffs(rgmii_rx_delay_mask) - 1; u16 rgmii_tx_delay_pos = ffs(rgmii_tx_delay_mask) - 1; u16 reg_val = 0; - int rc; + u16 mask = 0; + int rc = 0; - mutex_lock(&phydev->lock); + /* For traffic to pass, the VSC8502 family needs the RX_CLK disable bit + * to be unset for all PHY modes, so do that as part of the paged + * register modification. + * For some family members (like VSC8530/31/40/41) this bit is reserved + * and read-only, and the RX clock is enabled by default. + */ + if (rgmii_cntl == VSC8502_RGMII_CNTL) + mask |= VSC8502_RGMII_RX_CLK_DISABLE; + + if (phy_interface_is_rgmii(phydev)) + mask |= rgmii_rx_delay_mask | rgmii_tx_delay_mask; if (phydev->interface == PHY_INTERFACE_MODE_RGMII_RXID || phydev->interface == PHY_INTERFACE_MODE_RGMII_ID) @@ -537,31 +548,20 @@ static int vsc85xx_rgmii_set_skews(struct phy_device *phydev, u32 rgmii_cntl, phydev->interface == PHY_INTERFACE_MODE_RGMII_ID) reg_val |= RGMII_CLK_DELAY_2_0_NS << rgmii_tx_delay_pos; - rc = phy_modify_paged(phydev, MSCC_PHY_PAGE_EXTENDED_2, - rgmii_cntl, - rgmii_rx_delay_mask | rgmii_tx_delay_mask, - reg_val); - - mutex_unlock(&phydev->lock); + if (mask) + rc = phy_modify_paged(phydev, MSCC_PHY_PAGE_EXTENDED_2, + rgmii_cntl, mask, reg_val); return rc; } static int vsc85xx_default_config(struct phy_device *phydev) { - int rc; - phydev->mdix_ctrl = ETH_TP_MDI_AUTO; - if (phy_interface_mode_is_rgmii(phydev->interface)) { - rc = vsc85xx_rgmii_set_skews(phydev, VSC8502_RGMII_CNTL, - VSC8502_RGMII_RX_DELAY_MASK, - VSC8502_RGMII_TX_DELAY_MASK); - if (rc) - return rc; - } - - return 0; + return vsc85xx_update_rgmii_cntl(phydev, VSC8502_RGMII_CNTL, + VSC8502_RGMII_RX_DELAY_MASK, + VSC8502_RGMII_TX_DELAY_MASK); } static int vsc85xx_get_tunable(struct phy_device *phydev, @@ -1758,13 +1758,11 @@ static int vsc8584_config_init(struct phy_device *phydev) if (ret) return ret; - if (phy_interface_is_rgmii(phydev)) { - ret = vsc85xx_rgmii_set_skews(phydev, VSC8572_RGMII_CNTL, - VSC8572_RGMII_RX_DELAY_MASK, - VSC8572_RGMII_TX_DELAY_MASK); - if (ret) - return ret; - } + ret = vsc85xx_update_rgmii_cntl(phydev, VSC8572_RGMII_CNTL, + VSC8572_RGMII_RX_DELAY_MASK, + VSC8572_RGMII_TX_DELAY_MASK); + if (ret) + return ret; ret = genphy_soft_reset(phydev); if (ret) @@ -2316,6 +2314,30 @@ static int vsc85xx_probe(struct phy_device *phydev) /* Microsemi VSC85xx PHYs */ static struct phy_driver vsc85xx_driver[] = { +{ + .phy_id = PHY_ID_VSC8501, + .name = "Microsemi GE VSC8501 SyncE", + .phy_id_mask = 0xfffffff0, + /* PHY_BASIC_FEATURES */ + .soft_reset = &genphy_soft_reset, + .config_init = &vsc85xx_config_init, + .config_aneg = &vsc85xx_config_aneg, + .read_status = &vsc85xx_read_status, + .handle_interrupt = vsc85xx_handle_interrupt, + .config_intr = &vsc85xx_config_intr, + .suspend = &genphy_suspend, + .resume = &genphy_resume, + .probe = &vsc85xx_probe, + .set_wol = &vsc85xx_wol_set, + .get_wol = &vsc85xx_wol_get, + .get_tunable = &vsc85xx_get_tunable, + .set_tunable = &vsc85xx_set_tunable, + .read_page = &vsc85xx_phy_read_page, + .write_page = &vsc85xx_phy_write_page, + .get_sset_count = &vsc85xx_get_sset_count, + .get_strings = &vsc85xx_get_strings, + .get_stats = &vsc85xx_get_stats, +}, { .phy_id = PHY_ID_VSC8502, .name = "Microsemi GE VSC8502 SyncE", @@ -2656,6 +2678,8 @@ static struct phy_driver vsc85xx_driver[] = { module_phy_driver(vsc85xx_driver); static struct mdio_device_id __maybe_unused vsc85xx_tbl[] = { + { PHY_ID_VSC8501, 0xfffffff0, }, + { PHY_ID_VSC8502, 0xfffffff0, }, { PHY_ID_VSC8504, 0xfffffff0, }, { PHY_ID_VSC8514, 0xfffffff0, }, { PHY_ID_VSC8530, 0xfffffff0, }, diff --git a/drivers/net/phy/mxl-gpy.c b/drivers/net/phy/mxl-gpy.c index 6301a9abfb95596647e81e9071ca7828e54d6bcf..ea1073adc5a16a6b531e9f1f2c51869dad8ead80 100644 --- a/drivers/net/phy/mxl-gpy.c +++ b/drivers/net/phy/mxl-gpy.c @@ -274,13 +274,6 @@ static int gpy_config_init(struct phy_device *phydev) return ret < 0 ? ret : 0; } -static bool gpy_has_broken_mdint(struct phy_device *phydev) -{ - /* At least these PHYs are known to have broken interrupt handling */ - return phydev->drv->phy_id == PHY_ID_GPY215B || - phydev->drv->phy_id == PHY_ID_GPY215C; -} - static int gpy_probe(struct phy_device *phydev) { struct device *dev = &phydev->mdio.dev; @@ -300,8 +293,7 @@ static int gpy_probe(struct phy_device *phydev) phydev->priv = priv; mutex_init(&priv->mbox_lock); - if (gpy_has_broken_mdint(phydev) && - !device_property_present(dev, "maxlinear,use-broken-interrupts")) + if (!device_property_present(dev, "maxlinear,use-broken-interrupts")) phydev->dev_flags |= PHY_F_NO_IRQ; fw_version = phy_read(phydev, PHY_FWV); @@ -659,11 +651,9 @@ static irqreturn_t gpy_handle_interrupt(struct phy_device *phydev) * frame. Therefore, polling is the best we can do and won't do any more * harm. * It was observed that this bug happens on link state and link speed - * changes on a GPY215B and GYP215C independent of the firmware version - * (which doesn't mean that this list is exhaustive). + * changes independent of the firmware version. */ - if (gpy_has_broken_mdint(phydev) && - (reg & (PHY_IMASK_LSTC | PHY_IMASK_LSPC))) { + if (reg & (PHY_IMASK_LSTC | PHY_IMASK_LSPC)) { reg = gpy_mbox_read(phydev, REG_GPIO0_OUT); if (reg < 0) { phy_error(phydev); diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c index a4111f1be375738033042f3c86fcba8197c528bb..5efdeb59f4b2972131b79453692591da481b8baa 100644 --- a/drivers/net/phy/phylink.c +++ b/drivers/net/phy/phylink.c @@ -188,6 +188,7 @@ static int phylink_interface_max_speed(phy_interface_t interface) case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII: case PHY_INTERFACE_MODE_QSGMII: + case PHY_INTERFACE_MODE_QUSGMII: case PHY_INTERFACE_MODE_SGMII: case PHY_INTERFACE_MODE_GMII: return SPEED_1000; @@ -204,7 +205,6 @@ static int phylink_interface_max_speed(phy_interface_t interface) case PHY_INTERFACE_MODE_10GBASER: case PHY_INTERFACE_MODE_10GKR: case PHY_INTERFACE_MODE_USXGMII: - case PHY_INTERFACE_MODE_QUSGMII: return SPEED_10000; case PHY_INTERFACE_MODE_25GBASER: @@ -2226,6 +2226,12 @@ int phylink_ethtool_ksettings_set(struct phylink *pl, ASSERT_RTNL(); if (pl->phydev) { + struct ethtool_link_ksettings phy_kset = *kset; + + linkmode_and(phy_kset.link_modes.advertising, + phy_kset.link_modes.advertising, + pl->supported); + /* We can rely on phylib for this update; we also do not need * to update the pl->link_config settings: * - the configuration returned via ksettings_get() will come @@ -2244,11 +2250,10 @@ int phylink_ethtool_ksettings_set(struct phylink *pl, * the presence of a PHY, this should not be changed as that * should be determined from the media side advertisement. */ - return phy_ethtool_ksettings_set(pl->phydev, kset); + return phy_ethtool_ksettings_set(pl->phydev, &phy_kset); } config = pl->link_config; - /* Mask out unsupported advertisements */ linkmode_and(config.advertising, kset->link_modes.advertising, pl->supported); @@ -3293,6 +3298,41 @@ void phylink_decode_usxgmii_word(struct phylink_link_state *state, } EXPORT_SYMBOL_GPL(phylink_decode_usxgmii_word); +/** + * phylink_decode_usgmii_word() - decode the USGMII word from a MAC PCS + * @state: a pointer to a struct phylink_link_state. + * @lpa: a 16 bit value which stores the USGMII auto-negotiation word + * + * Helper for MAC PCS supporting the USGMII protocol and the auto-negotiation + * code word. Decode the USGMII code word and populate the corresponding fields + * (speed, duplex) into the phylink_link_state structure. The structure for this + * word is the same as the USXGMII word, except it only supports speeds up to + * 1Gbps. + */ +static void phylink_decode_usgmii_word(struct phylink_link_state *state, + uint16_t lpa) +{ + switch (lpa & MDIO_USXGMII_SPD_MASK) { + case MDIO_USXGMII_10: + state->speed = SPEED_10; + break; + case MDIO_USXGMII_100: + state->speed = SPEED_100; + break; + case MDIO_USXGMII_1000: + state->speed = SPEED_1000; + break; + default: + state->link = false; + return; + } + + if (lpa & MDIO_USXGMII_FULL_DUPLEX) + state->duplex = DUPLEX_FULL; + else + state->duplex = DUPLEX_HALF; +} + /** * phylink_mii_c22_pcs_decode_state() - Decode MAC PCS state from MII registers * @state: a pointer to a &struct phylink_link_state. @@ -3330,9 +3370,11 @@ void phylink_mii_c22_pcs_decode_state(struct phylink_link_state *state, case PHY_INTERFACE_MODE_SGMII: case PHY_INTERFACE_MODE_QSGMII: - case PHY_INTERFACE_MODE_QUSGMII: phylink_decode_sgmii_word(state, lpa); break; + case PHY_INTERFACE_MODE_QUSGMII: + phylink_decode_usgmii_word(state, lpa); + break; default: state->link = false; diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index d10606f257c43345e1d293dd462cdf77fc37497b..555b0b1e9a7893ac828fa2e9488d1d3d7df3c064 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -1629,6 +1629,7 @@ static int team_init(struct net_device *dev) team->dev = dev; team_set_no_mode(team); + team->notifier_ctx = false; team->pcpu_stats = netdev_alloc_pcpu_stats(struct team_pcpu_stats); if (!team->pcpu_stats) @@ -3022,7 +3023,11 @@ static int team_device_event(struct notifier_block *unused, team_del_slave(port->team->dev, dev); break; case NETDEV_FEAT_CHANGE: - team_compute_features(port->team); + if (!port->team->notifier_ctx) { + port->team->notifier_ctx = true; + team_compute_features(port->team); + port->team->notifier_ctx = false; + } break; case NETDEV_PRECHANGEMTU: /* Forbid to change mtu of underlaying device */ diff --git a/drivers/net/tun.c b/drivers/net/tun.c index d4d0a41a905a7e7fb332af8279136041b9229f23..d75456adc62ac82a4f9da10c50c48266c5a9a0c0 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -1977,6 +1977,14 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile, int queue_len; spin_lock_bh(&queue->lock); + + if (unlikely(tfile->detached)) { + spin_unlock_bh(&queue->lock); + rcu_read_unlock(); + err = -EBUSY; + goto free_skb; + } + __skb_queue_tail(queue, skb); queue_len = skb_queue_len(queue); spin_unlock(&queue->lock); @@ -2512,6 +2520,13 @@ static int tun_xdp_one(struct tun_struct *tun, if (tfile->napi_enabled) { queue = &tfile->sk.sk_write_queue; spin_lock(&queue->lock); + + if (unlikely(tfile->detached)) { + spin_unlock(&queue->lock); + kfree_skb(skb); + return -EBUSY; + } + __skb_queue_tail(queue, skb); spin_unlock(&queue->lock); ret = 1; diff --git a/drivers/net/usb/cdc_ncm.c b/drivers/net/usb/cdc_ncm.c index 6ce8f4f0c70e898c5ebb7198c46ac6a38f5b588f..db05622f1f703ec6fc147935c296640c55499a4b 100644 --- a/drivers/net/usb/cdc_ncm.c +++ b/drivers/net/usb/cdc_ncm.c @@ -181,9 +181,12 @@ static u32 cdc_ncm_check_tx_max(struct usbnet *dev, u32 new_tx) else min = ctx->max_datagram_size + ctx->max_ndp_size + sizeof(struct usb_cdc_ncm_nth32); - max = min_t(u32, CDC_NCM_NTB_MAX_SIZE_TX, le32_to_cpu(ctx->ncm_parm.dwNtbOutMaxSize)); - if (max == 0) + if (le32_to_cpu(ctx->ncm_parm.dwNtbOutMaxSize) == 0) max = CDC_NCM_NTB_MAX_SIZE_TX; /* dwNtbOutMaxSize not set */ + else + max = clamp_t(u32, le32_to_cpu(ctx->ncm_parm.dwNtbOutMaxSize), + USB_CDC_NCM_NTB_MIN_OUT_SIZE, + CDC_NCM_NTB_MAX_SIZE_TX); /* some devices set dwNtbOutMaxSize too low for the above default */ min = min(min, max); @@ -1244,6 +1247,9 @@ cdc_ncm_fill_tx_frame(struct usbnet *dev, struct sk_buff *skb, __le32 sign) * further. */ if (skb_out == NULL) { + /* If even the smallest allocation fails, abort. */ + if (ctx->tx_curr_size == USB_CDC_NCM_NTB_MIN_OUT_SIZE) + goto alloc_failed; ctx->tx_low_mem_max_cnt = min(ctx->tx_low_mem_max_cnt + 1, (unsigned)CDC_NCM_LOW_MEM_MAX_CNT); ctx->tx_low_mem_val = ctx->tx_low_mem_max_cnt; @@ -1262,13 +1268,8 @@ cdc_ncm_fill_tx_frame(struct usbnet *dev, struct sk_buff *skb, __le32 sign) skb_out = alloc_skb(ctx->tx_curr_size, GFP_ATOMIC); /* No allocation possible so we will abort */ - if (skb_out == NULL) { - if (skb != NULL) { - dev_kfree_skb_any(skb); - dev->net->stats.tx_dropped++; - } - goto exit_no_skb; - } + if (!skb_out) + goto alloc_failed; ctx->tx_low_mem_val--; } if (ctx->is_ndp16) { @@ -1461,6 +1462,11 @@ cdc_ncm_fill_tx_frame(struct usbnet *dev, struct sk_buff *skb, __le32 sign) return skb_out; +alloc_failed: + if (skb) { + dev_kfree_skb_any(skb); + dev->net->stats.tx_dropped++; + } exit_no_skb: /* Start timer, if there is a remaining non-empty skb */ if (ctx->tx_curr_skb != NULL && n > 0) diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c index 571e37e67f9ce1706de16d3f4a4f9c883a6e8992..2e7c7b0cdc549e6b9e8799ea38083dca982c6db9 100644 --- a/drivers/net/usb/qmi_wwan.c +++ b/drivers/net/usb/qmi_wwan.c @@ -1220,7 +1220,9 @@ static const struct usb_device_id products[] = { {QMI_FIXED_INTF(0x05c6, 0x9080, 8)}, {QMI_FIXED_INTF(0x05c6, 0x9083, 3)}, {QMI_FIXED_INTF(0x05c6, 0x9084, 4)}, + {QMI_QUIRK_SET_DTR(0x05c6, 0x9091, 2)}, /* Compal RXM-G1 */ {QMI_FIXED_INTF(0x05c6, 0x90b2, 3)}, /* ublox R410M */ + {QMI_QUIRK_SET_DTR(0x05c6, 0x90db, 2)}, /* Compal RXM-G1 */ {QMI_FIXED_INTF(0x05c6, 0x920d, 0)}, {QMI_FIXED_INTF(0x05c6, 0x920d, 5)}, {QMI_QUIRK_SET_DTR(0x05c6, 0x9625, 4)}, /* YUGA CLM920-NC5 */ @@ -1325,7 +1327,7 @@ static const struct usb_device_id products[] = { {QMI_FIXED_INTF(0x2001, 0x7e3d, 4)}, /* D-Link DWM-222 A2 */ {QMI_FIXED_INTF(0x2020, 0x2031, 4)}, /* Olicard 600 */ {QMI_FIXED_INTF(0x2020, 0x2033, 4)}, /* BroadMobi BM806U */ - {QMI_FIXED_INTF(0x2020, 0x2060, 4)}, /* BroadMobi BM818 */ + {QMI_QUIRK_SET_DTR(0x2020, 0x2060, 4)}, /* BroadMobi BM818 */ {QMI_FIXED_INTF(0x0f3d, 0x68a2, 8)}, /* Sierra Wireless MC7700 */ {QMI_FIXED_INTF(0x114f, 0x68a2, 8)}, /* Sierra Wireless MC7750 */ {QMI_FIXED_INTF(0x1199, 0x68a2, 8)}, /* Sierra Wireless MC7710 in QMI mode */ diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index a12ae26db0e223e57b691b0dc0728b97734d0659..486b5849033dc96cf2ff3ded546c7736f840f61d 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -205,6 +205,8 @@ struct control_buf { __virtio16 vid; __virtio64 offloads; struct virtio_net_ctrl_rss rss; + struct virtio_net_ctrl_coal_tx coal_tx; + struct virtio_net_ctrl_coal_rx coal_rx; }; struct virtnet_info { @@ -1868,6 +1870,38 @@ static int virtnet_poll(struct napi_struct *napi, int budget) return received; } +static void virtnet_disable_queue_pair(struct virtnet_info *vi, int qp_index) +{ + virtnet_napi_tx_disable(&vi->sq[qp_index].napi); + napi_disable(&vi->rq[qp_index].napi); + xdp_rxq_info_unreg(&vi->rq[qp_index].xdp_rxq); +} + +static int virtnet_enable_queue_pair(struct virtnet_info *vi, int qp_index) +{ + struct net_device *dev = vi->dev; + int err; + + err = xdp_rxq_info_reg(&vi->rq[qp_index].xdp_rxq, dev, qp_index, + vi->rq[qp_index].napi.napi_id); + if (err < 0) + return err; + + err = xdp_rxq_info_reg_mem_model(&vi->rq[qp_index].xdp_rxq, + MEM_TYPE_PAGE_SHARED, NULL); + if (err < 0) + goto err_xdp_reg_mem_model; + + virtnet_napi_enable(vi->rq[qp_index].vq, &vi->rq[qp_index].napi); + virtnet_napi_tx_enable(vi, vi->sq[qp_index].vq, &vi->sq[qp_index].napi); + + return 0; + +err_xdp_reg_mem_model: + xdp_rxq_info_unreg(&vi->rq[qp_index].xdp_rxq); + return err; +} + static int virtnet_open(struct net_device *dev) { struct virtnet_info *vi = netdev_priv(dev); @@ -1881,22 +1915,20 @@ static int virtnet_open(struct net_device *dev) if (!try_fill_recv(vi, &vi->rq[i], GFP_KERNEL)) schedule_delayed_work(&vi->refill, 0); - err = xdp_rxq_info_reg(&vi->rq[i].xdp_rxq, dev, i, vi->rq[i].napi.napi_id); + err = virtnet_enable_queue_pair(vi, i); if (err < 0) - return err; - - err = xdp_rxq_info_reg_mem_model(&vi->rq[i].xdp_rxq, - MEM_TYPE_PAGE_SHARED, NULL); - if (err < 0) { - xdp_rxq_info_unreg(&vi->rq[i].xdp_rxq); - return err; - } - - virtnet_napi_enable(vi->rq[i].vq, &vi->rq[i].napi); - virtnet_napi_tx_enable(vi, vi->sq[i].vq, &vi->sq[i].napi); + goto err_enable_qp; } return 0; + +err_enable_qp: + disable_delayed_refill(vi); + cancel_delayed_work_sync(&vi->refill); + + for (i--; i >= 0; i--) + virtnet_disable_queue_pair(vi, i); + return err; } static int virtnet_poll_tx(struct napi_struct *napi, int budget) @@ -2305,11 +2337,8 @@ static int virtnet_close(struct net_device *dev) /* Make sure refill_work doesn't re-enable napi! */ cancel_delayed_work_sync(&vi->refill); - for (i = 0; i < vi->max_queue_pairs; i++) { - virtnet_napi_tx_disable(&vi->sq[i].napi); - napi_disable(&vi->rq[i].napi); - xdp_rxq_info_unreg(&vi->rq[i].xdp_rxq); - } + for (i = 0; i < vi->max_queue_pairs; i++) + virtnet_disable_queue_pair(vi, i); return 0; } @@ -2907,12 +2936,10 @@ static int virtnet_send_notf_coal_cmds(struct virtnet_info *vi, struct ethtool_coalesce *ec) { struct scatterlist sgs_tx, sgs_rx; - struct virtio_net_ctrl_coal_tx coal_tx; - struct virtio_net_ctrl_coal_rx coal_rx; - coal_tx.tx_usecs = cpu_to_le32(ec->tx_coalesce_usecs); - coal_tx.tx_max_packets = cpu_to_le32(ec->tx_max_coalesced_frames); - sg_init_one(&sgs_tx, &coal_tx, sizeof(coal_tx)); + vi->ctrl->coal_tx.tx_usecs = cpu_to_le32(ec->tx_coalesce_usecs); + vi->ctrl->coal_tx.tx_max_packets = cpu_to_le32(ec->tx_max_coalesced_frames); + sg_init_one(&sgs_tx, &vi->ctrl->coal_tx, sizeof(vi->ctrl->coal_tx)); if (!virtnet_send_command(vi, VIRTIO_NET_CTRL_NOTF_COAL, VIRTIO_NET_CTRL_NOTF_COAL_TX_SET, @@ -2923,9 +2950,9 @@ static int virtnet_send_notf_coal_cmds(struct virtnet_info *vi, vi->tx_usecs = ec->tx_coalesce_usecs; vi->tx_max_packets = ec->tx_max_coalesced_frames; - coal_rx.rx_usecs = cpu_to_le32(ec->rx_coalesce_usecs); - coal_rx.rx_max_packets = cpu_to_le32(ec->rx_max_coalesced_frames); - sg_init_one(&sgs_rx, &coal_rx, sizeof(coal_rx)); + vi->ctrl->coal_rx.rx_usecs = cpu_to_le32(ec->rx_coalesce_usecs); + vi->ctrl->coal_rx.rx_max_packets = cpu_to_le32(ec->rx_max_coalesced_frames); + sg_init_one(&sgs_rx, &vi->ctrl->coal_rx, sizeof(vi->ctrl->coal_rx)); if (!virtnet_send_command(vi, VIRTIO_NET_CTRL_NOTF_COAL, VIRTIO_NET_CTRL_NOTF_COAL_RX_SET, diff --git a/drivers/net/wan/lapbether.c b/drivers/net/wan/lapbether.c index d62a904d2e422d7e48cf67cba829df0568e99b01..56326f38fe8a30f45e88cdce7efd43e18041e52a 100644 --- a/drivers/net/wan/lapbether.c +++ b/drivers/net/wan/lapbether.c @@ -384,6 +384,9 @@ static int lapbeth_new_device(struct net_device *dev) ASSERT_RTNL(); + if (dev->type != ARPHRD_ETHER) + return -EINVAL; + ndev = alloc_netdev(sizeof(*lapbeth), "lapb%d", NET_NAME_UNKNOWN, lapbeth_setup); if (!ndev) diff --git a/drivers/net/wireless/broadcom/b43/b43.h b/drivers/net/wireless/broadcom/b43/b43.h index 9fc7c088a539e3728631ef41bf49004e60fa17f5..67b4bac048e585bd336196b8004fa6555d6abd0c 100644 --- a/drivers/net/wireless/broadcom/b43/b43.h +++ b/drivers/net/wireless/broadcom/b43/b43.h @@ -651,7 +651,7 @@ struct b43_iv { union { __be16 d16; __be32 d32; - } data __packed; + } __packed data; } __packed; diff --git a/drivers/net/wireless/broadcom/b43legacy/b43legacy.h b/drivers/net/wireless/broadcom/b43legacy/b43legacy.h index 6b0cec467938fc555f23f0d464086afec69181d9..f49365d14619f39d05535e0b82b4e7f2228fbfa5 100644 --- a/drivers/net/wireless/broadcom/b43legacy/b43legacy.h +++ b/drivers/net/wireless/broadcom/b43legacy/b43legacy.h @@ -379,7 +379,7 @@ struct b43legacy_iv { union { __be16 d16; __be32 d32; - } data __packed; + } __packed data; } __packed; #define B43legacy_PHYMODE(phytype) (1 << (phytype)) diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c index ff710b0b5071a19b60e8f5d3a7fe1e22cd7b75a9..00679a990e3dac9bc9b71c0941a737ec0338ccc4 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c @@ -1039,6 +1039,11 @@ static int brcmf_ops_sdio_probe(struct sdio_func *func, struct brcmf_sdio_dev *sdiodev; struct brcmf_bus *bus_if; + if (!id) { + dev_err(&func->dev, "Error no sdio_device_id passed for %x:%x\n", func->vendor, func->device); + return -ENODEV; + } + brcmf_dbg(SDIO, "Enter\n"); brcmf_dbg(SDIO, "Class=%x\n", func->class); brcmf_dbg(SDIO, "sdio vendor ID: 0x%04x\n", func->vendor); diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c index 59f3e9c5e13907bec4902210ddc7024829b2024d..80220685f5e4514782a1179d3195cbd34cd890b3 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c @@ -2394,6 +2394,9 @@ static void brcmf_pcie_debugfs_create(struct device *dev) } #endif +/* Forward declaration for pci_match_id() call */ +static const struct pci_device_id brcmf_pcie_devid_table[]; + static int brcmf_pcie_probe(struct pci_dev *pdev, const struct pci_device_id *id) { @@ -2404,6 +2407,14 @@ brcmf_pcie_probe(struct pci_dev *pdev, const struct pci_device_id *id) struct brcmf_core *core; struct brcmf_bus *bus; + if (!id) { + id = pci_match_id(brcmf_pcie_devid_table, pdev); + if (!id) { + pci_err(pdev, "Error could not find pci_device_id for %x:%x\n", pdev->vendor, pdev->device); + return -ENODEV; + } + } + brcmf_dbg(PCIE, "Enter %x:%x\n", pdev->vendor, pdev->device); ret = -ENOMEM; diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c index 246843aeb6964af25f5fc9b6a64d0b9e169dc4de..2178675ae1a44d0e04afdd4d26ad6aa0625da8cf 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c @@ -1331,6 +1331,9 @@ brcmf_usb_disconnect_cb(struct brcmf_usbdev_info *devinfo) brcmf_usb_detach(devinfo); } +/* Forward declaration for usb_match_id() call */ +static const struct usb_device_id brcmf_usb_devid_table[]; + static int brcmf_usb_probe(struct usb_interface *intf, const struct usb_device_id *id) { @@ -1342,6 +1345,14 @@ brcmf_usb_probe(struct usb_interface *intf, const struct usb_device_id *id) u32 num_of_eps; u8 endpoint_num, ep; + if (!id) { + id = usb_match_id(intf, brcmf_usb_devid_table); + if (!id) { + dev_err(&intf->dev, "Error could not find matching usb_device_id\n"); + return -ENODEV; + } + } + brcmf_dbg(USB, "Enter 0x%04x:0x%04x\n", id->idVendor, id->idProduct); devinfo = kzalloc(sizeof(*devinfo), GFP_ATOMIC); diff --git a/drivers/net/wireless/intel/iwlwifi/fw/acpi.c b/drivers/net/wireless/intel/iwlwifi/fw/acpi.c index 5f4a51310add082ecd5c931883ff4926fae0ed5d..cb9181f05501130b3cd6fb9cf348996397f6082c 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/acpi.c +++ b/drivers/net/wireless/intel/iwlwifi/fw/acpi.c @@ -38,7 +38,7 @@ static const struct dmi_system_id dmi_ppag_approved_list[] = { }, { .ident = "ASUS", .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "ASUSTek COMPUTER INC."), + DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), }, }, {} diff --git a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c index d9faaae01abd22b149566dfbd6eda80672c59ea6..55219974b92bf73c6ab70aa3e53ca4408257753a 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c +++ b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c @@ -1664,14 +1664,10 @@ static __le32 iwl_get_mon_reg(struct iwl_fw_runtime *fwrt, u32 alloc_id, } static void * -iwl_dump_ini_mon_fill_header(struct iwl_fw_runtime *fwrt, - struct iwl_dump_ini_region_data *reg_data, +iwl_dump_ini_mon_fill_header(struct iwl_fw_runtime *fwrt, u32 alloc_id, struct iwl_fw_ini_monitor_dump *data, const struct iwl_fw_mon_regs *addrs) { - struct iwl_fw_ini_region_tlv *reg = (void *)reg_data->reg_tlv->data; - u32 alloc_id = le32_to_cpu(reg->dram_alloc_id); - if (!iwl_trans_grab_nic_access(fwrt->trans)) { IWL_ERR(fwrt, "Failed to get monitor header\n"); return NULL; @@ -1702,8 +1698,10 @@ iwl_dump_ini_mon_dram_fill_header(struct iwl_fw_runtime *fwrt, void *data, u32 data_len) { struct iwl_fw_ini_monitor_dump *mon_dump = (void *)data; + struct iwl_fw_ini_region_tlv *reg = (void *)reg_data->reg_tlv->data; + u32 alloc_id = le32_to_cpu(reg->dram_alloc_id); - return iwl_dump_ini_mon_fill_header(fwrt, reg_data, mon_dump, + return iwl_dump_ini_mon_fill_header(fwrt, alloc_id, mon_dump, &fwrt->trans->cfg->mon_dram_regs); } @@ -1713,8 +1711,10 @@ iwl_dump_ini_mon_smem_fill_header(struct iwl_fw_runtime *fwrt, void *data, u32 data_len) { struct iwl_fw_ini_monitor_dump *mon_dump = (void *)data; + struct iwl_fw_ini_region_tlv *reg = (void *)reg_data->reg_tlv->data; + u32 alloc_id = le32_to_cpu(reg->internal_buffer.alloc_id); - return iwl_dump_ini_mon_fill_header(fwrt, reg_data, mon_dump, + return iwl_dump_ini_mon_fill_header(fwrt, alloc_id, mon_dump, &fwrt->trans->cfg->mon_smem_regs); } @@ -1725,7 +1725,10 @@ iwl_dump_ini_mon_dbgi_fill_header(struct iwl_fw_runtime *fwrt, { struct iwl_fw_ini_monitor_dump *mon_dump = (void *)data; - return iwl_dump_ini_mon_fill_header(fwrt, reg_data, mon_dump, + return iwl_dump_ini_mon_fill_header(fwrt, + /* no offset calculation later */ + IWL_FW_INI_ALLOCATION_ID_DBGC1, + mon_dump, &fwrt->trans->cfg->mon_dbgi_regs); } diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c index 37aa4676dc94852a1ae6add76d52d0ab5f8c8912..6d1007f24b4aa91ba33e4e60d1019eb309a35984 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c @@ -2732,17 +2732,13 @@ static bool iwl_mvm_wait_d3_notif(struct iwl_notif_wait_data *notif_wait, if (wowlan_info_ver < 2) { struct iwl_wowlan_info_notif_v1 *notif_v1 = (void *)pkt->data; - notif = kmemdup(notif_v1, - offsetofend(struct iwl_wowlan_info_notif, - received_beacons), - GFP_ATOMIC); - + notif = kmemdup(notif_v1, sizeof(*notif), GFP_ATOMIC); if (!notif) return false; notif->tid_tear_down = notif_v1->tid_tear_down; notif->station_id = notif_v1->station_id; - + memset_after(notif, 0, station_id); } else { notif = (void *)pkt->data; } diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c index 3963a0d4ed0427ba004e0bb5f846df5ab4eb8332..652a603c4500efed3be6162d4f26a520eeda3f93 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c @@ -526,6 +526,11 @@ iwl_mvm_ftm_put_target(struct iwl_mvm *mvm, struct ieee80211_vif *vif, rcu_read_lock(); sta = rcu_dereference(mvm->fw_id_to_mac_id[mvmvif->deflink.ap_sta_id]); + if (WARN_ON_ONCE(IS_ERR_OR_NULL(sta))) { + rcu_read_unlock(); + return PTR_ERR_OR_ZERO(sta); + } + if (sta->mfp && (peer->ftm.trigger_based || peer->ftm.non_trigger_based)) FTM_PUT_FLAG(PMF); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c index b35c96cf7ad24d5f6f2a3da03568a7db3db4fe6c..205c09bc986342ff8d98866310a4056dd0366602 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c @@ -1091,7 +1091,7 @@ static const struct dmi_system_id dmi_tas_approved_list[] = { }, { .ident = "LENOVO", .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "Lenovo"), + DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), }, }, { .ident = "DELL", @@ -1727,8 +1727,7 @@ int iwl_mvm_up(struct iwl_mvm *mvm) iwl_mvm_tas_init(mvm); iwl_mvm_leds_sync(mvm); - if (fw_has_capa(&mvm->fw->ucode_capa, - IWL_UCODE_TLV_CAPA_RFIM_SUPPORT)) { + if (iwl_rfi_supported(mvm)) { if (iwl_mvm_eval_dsm_rfi(mvm) == DSM_VALUE_RFI_ENABLE) iwl_rfi_send_config_cmd(mvm, NULL); } diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/link.c b/drivers/net/wireless/intel/iwlwifi/mvm/link.c index eb828de40a3c6d9491dca4839fea28ce66911b4e..3814915cb1a67df435b3b84a973af9fa9e29505f 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/link.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/link.c @@ -123,11 +123,13 @@ int iwl_mvm_link_changed(struct iwl_mvm *mvm, struct ieee80211_vif *vif, if (mvmvif->link[i]->phy_ctxt) count++; - /* FIXME: IWL_MVM_FW_MAX_ACTIVE_LINKS_NUM should be - * defined per HW - */ - if (count >= IWL_MVM_FW_MAX_ACTIVE_LINKS_NUM) - return -EINVAL; + if (vif->type == NL80211_IFTYPE_AP) { + if (count > mvm->fw->ucode_capa.num_beacons) + return -EOPNOTSUPP; + /* this should be per HW or such */ + } else if (count >= IWL_MVM_FW_MAX_ACTIVE_LINKS_NUM) { + return -EOPNOTSUPP; + } } /* Catch early if driver tries to activate or deactivate a link diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c index 0f01b62357c6fded37ad07cfcacbd7eec8cf8118..17f788a5ff6ba686e0d9743c3e8677dfecf9dc6d 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause /* - * Copyright (C) 2012-2014, 2018-2022 Intel Corporation + * Copyright (C) 2012-2014, 2018-2023 Intel Corporation * Copyright (C) 2013-2015 Intel Mobile Communications GmbH * Copyright (C) 2016-2017 Intel Deutschland GmbH */ @@ -3607,7 +3607,8 @@ static bool iwl_mvm_vif_conf_from_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif, struct ieee80211_sta *sta) { - unsigned int i; + struct ieee80211_link_sta *link_sta; + unsigned int link_id; /* Beacon interval check - firmware will crash if the beacon * interval is less than 16. We can't avoid connecting at all, @@ -3616,14 +3617,11 @@ static bool iwl_mvm_vif_conf_from_sta(struct iwl_mvm *mvm, * wpa_s will blocklist the AP... */ - for_each_set_bit(i, (unsigned long *)&sta->valid_links, - IEEE80211_MLD_MAX_NUM_LINKS) { - struct ieee80211_link_sta *link_sta = - link_sta_dereference_protected(sta, i); + for_each_sta_active_link(vif, sta, link_sta, link_id) { struct ieee80211_bss_conf *link_conf = - link_conf_dereference_protected(vif, i); + link_conf_dereference_protected(vif, link_id); - if (!link_conf || !link_sta) + if (!link_conf) continue; if (link_conf->beacon_int < IWL_MVM_MIN_BEACON_INTERVAL_TU) { @@ -3645,24 +3643,23 @@ static void iwl_mvm_vif_set_he_support(struct ieee80211_hw *hw, bool is_sta) { struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif); - unsigned int i; + struct ieee80211_link_sta *link_sta; + unsigned int link_id; - for_each_set_bit(i, (unsigned long *)&sta->valid_links, - IEEE80211_MLD_MAX_NUM_LINKS) { - struct ieee80211_link_sta *link_sta = - link_sta_dereference_protected(sta, i); + for_each_sta_active_link(vif, sta, link_sta, link_id) { struct ieee80211_bss_conf *link_conf = - link_conf_dereference_protected(vif, i); + link_conf_dereference_protected(vif, link_id); - if (!link_conf || !link_sta || !mvmvif->link[i]) + if (!link_conf || !mvmvif->link[link_id]) continue; link_conf->he_support = link_sta->he_cap.has_he; if (is_sta) { - mvmvif->link[i]->he_ru_2mhz_block = false; + mvmvif->link[link_id]->he_ru_2mhz_block = false; if (link_sta->he_cap.has_he) - iwl_mvm_check_he_obss_narrow_bw_ru(hw, vif, i, + iwl_mvm_check_he_obss_narrow_bw_ru(hw, vif, + link_id, link_conf); } } @@ -3675,6 +3672,7 @@ iwl_mvm_sta_state_notexist_to_none(struct iwl_mvm *mvm, struct iwl_mvm_sta_state_ops *callbacks) { struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif); + struct ieee80211_link_sta *link_sta; unsigned int i; int ret; @@ -3699,15 +3697,9 @@ iwl_mvm_sta_state_notexist_to_none(struct iwl_mvm *mvm, NL80211_TDLS_SETUP); } - for (i = 0; i < ARRAY_SIZE(sta->link); i++) { - struct ieee80211_link_sta *link_sta; - - link_sta = link_sta_dereference_protected(sta, i); - if (!link_sta) - continue; - + for_each_sta_active_link(vif, sta, link_sta, i) link_sta->agg.max_rc_amsdu_len = 1; - } + ieee80211_sta_recalc_aggregates(sta); if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls) @@ -3725,7 +3717,8 @@ iwl_mvm_sta_state_auth_to_assoc(struct ieee80211_hw *hw, { struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif); struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta); - unsigned int i; + struct ieee80211_link_sta *link_sta; + unsigned int link_id; lockdep_assert_held(&mvm->mutex); @@ -3751,14 +3744,13 @@ iwl_mvm_sta_state_auth_to_assoc(struct ieee80211_hw *hw, if (!mvm->mld_api_is_used) goto out; - for_each_set_bit(i, (unsigned long *)&sta->valid_links, - IEEE80211_MLD_MAX_NUM_LINKS) { + for_each_sta_active_link(vif, sta, link_sta, link_id) { struct ieee80211_bss_conf *link_conf = - link_conf_dereference_protected(vif, i); + link_conf_dereference_protected(vif, link_id); if (WARN_ON(!link_conf)) return -EINVAL; - if (!mvmvif->link[i]) + if (!mvmvif->link[link_id]) continue; iwl_mvm_link_changed(mvm, vif, link_conf, @@ -3889,6 +3881,9 @@ int iwl_mvm_mac_sta_state_common(struct ieee80211_hw *hw, * from the AP now. */ iwl_mvm_reset_cca_40mhz_workaround(mvm, vif); + + /* Also free dup data just in case any assertions below fail */ + kfree(mvm_sta->dup_data); } mutex_lock(&mvm->mutex); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c index fbc2d5ed100690a583f750e0db0a91603aa7881c..7fb66c5709596af2ffa7c1168e1bcf96879643a4 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c @@ -906,11 +906,12 @@ iwl_mvm_mld_change_vif_links(struct ieee80211_hw *hw, n_active++; } - if (vif->type == NL80211_IFTYPE_AP && - n_active > mvm->fw->ucode_capa.num_beacons) - return -EOPNOTSUPP; - else if (n_active > 1) + if (vif->type == NL80211_IFTYPE_AP) { + if (n_active > mvm->fw->ucode_capa.num_beacons) + return -EOPNOTSUPP; + } else if (n_active > 1) { return -EOPNOTSUPP; + } } for (i = 0; i < IEEE80211_MLD_MAX_NUM_LINKS; i++) { diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c index 0bfdf446275508a1c82b8a3335048af2a1591c3a..85a4ce8449ade824dc067961ac2938b4e4cfc661 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c @@ -667,15 +667,15 @@ int iwl_mvm_mld_add_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif, ret = iwl_mvm_mld_alloc_sta_links(mvm, vif, sta); if (ret) return ret; - } - spin_lock_init(&mvm_sta->lock); + spin_lock_init(&mvm_sta->lock); - if (test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status)) - ret = iwl_mvm_alloc_sta_after_restart(mvm, vif, sta); - else ret = iwl_mvm_sta_init(mvm, vif, sta, IWL_MVM_INVALID_STA, STATION_TYPE_PEER); + } else { + ret = iwl_mvm_alloc_sta_after_restart(mvm, vif, sta); + } + if (ret) goto err; @@ -728,7 +728,7 @@ int iwl_mvm_mld_update_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif, struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta); struct ieee80211_link_sta *link_sta; unsigned int link_id; - int ret = 0; + int ret = -EINVAL; lockdep_assert_held(&mvm->mutex); @@ -791,8 +791,6 @@ int iwl_mvm_mld_rm_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif, lockdep_assert_held(&mvm->mutex); - kfree(mvm_sta->dup_data); - /* flush its queues here since we are freeing mvm_sta */ for_each_sta_active_link(vif, sta, link_sta, link_id) { struct iwl_mvm_link_sta *mvm_link_sta = diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h index 6e7470d3a826d35ded09d840985893907205d671..9e5008e0e47f5c1824854c654a101cda4600caa3 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h @@ -2347,6 +2347,7 @@ int iwl_mvm_mld_update_sta_keys(struct iwl_mvm *mvm, u32 old_sta_mask, u32 new_sta_mask); +bool iwl_rfi_supported(struct iwl_mvm *mvm); int iwl_rfi_send_config_cmd(struct iwl_mvm *mvm, struct iwl_rfi_lut_entry *rfi_table); struct iwl_rfi_freq_table_resp_cmd *iwl_rfi_get_freq_table(struct iwl_mvm *mvm); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c index 6d18a1fd649b90e52006f7869663541e7320360a..fdf60afb0f3f26ec9e96f9132aacf45d61564ae4 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c @@ -445,6 +445,11 @@ iwl_mvm_update_mcc(struct iwl_mvm *mvm, const char *alpha2, struct iwl_mcc_update_resp *mcc_resp = (void *)pkt->data; n_channels = __le32_to_cpu(mcc_resp->n_channels); + if (iwl_rx_packet_payload_len(pkt) != + struct_size(mcc_resp, channels, n_channels)) { + resp_cp = ERR_PTR(-EINVAL); + goto exit; + } resp_len = sizeof(struct iwl_mcc_update_resp) + n_channels * sizeof(__le32); resp_cp = kmemdup(mcc_resp, resp_len, GFP_KERNEL); @@ -456,6 +461,11 @@ iwl_mvm_update_mcc(struct iwl_mvm *mvm, const char *alpha2, struct iwl_mcc_update_resp_v3 *mcc_resp_v3 = (void *)pkt->data; n_channels = __le32_to_cpu(mcc_resp_v3->n_channels); + if (iwl_rx_packet_payload_len(pkt) != + struct_size(mcc_resp_v3, channels, n_channels)) { + resp_cp = ERR_PTR(-EINVAL); + goto exit; + } resp_len = sizeof(struct iwl_mcc_update_resp) + n_channels * sizeof(__le32); resp_cp = kzalloc(resp_len, GFP_KERNEL); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rfi.c b/drivers/net/wireless/intel/iwlwifi/mvm/rfi.c index bb77bc9aa8218af83b21fa8b353d118b12c3685b..2ecd32bed752ff55734d5e9f78f36efc213cb107 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rfi.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rfi.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause /* - * Copyright (C) 2020 - 2021 Intel Corporation + * Copyright (C) 2020 - 2022 Intel Corporation */ #include "mvm.h" @@ -70,6 +70,16 @@ static const struct iwl_rfi_lut_entry iwl_rfi_table[IWL_RFI_LUT_SIZE] = { PHY_BAND_6, PHY_BAND_6,}}, }; +bool iwl_rfi_supported(struct iwl_mvm *mvm) +{ + /* The feature depends on a platform bugfix, so for now + * it's always disabled. + * When the platform support detection is implemented we should + * check FW TLV and platform support instead. + */ + return false; +} + int iwl_rfi_send_config_cmd(struct iwl_mvm *mvm, struct iwl_rfi_lut_entry *rfi_table) { int ret; @@ -81,7 +91,7 @@ int iwl_rfi_send_config_cmd(struct iwl_mvm *mvm, struct iwl_rfi_lut_entry *rfi_t .len[0] = sizeof(cmd), }; - if (!fw_has_capa(&mvm->fw->ucode_capa, IWL_UCODE_TLV_CAPA_RFIM_SUPPORT)) + if (!iwl_rfi_supported(mvm)) return -EOPNOTSUPP; lockdep_assert_held(&mvm->mutex); @@ -113,7 +123,7 @@ struct iwl_rfi_freq_table_resp_cmd *iwl_rfi_get_freq_table(struct iwl_mvm *mvm) .flags = CMD_WANT_SKB, }; - if (!fw_has_capa(&mvm->fw->ucode_capa, IWL_UCODE_TLV_CAPA_RFIM_SUPPORT)) + if (!iwl_rfi_supported(mvm)) return ERR_PTR(-EOPNOTSUPP); mutex_lock(&mvm->mutex); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c index a4c1e3bf4ff1de8db144095f157984ea11943999..9a20468345e49db727cfe087da652f73d71b7026 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c @@ -2691,6 +2691,8 @@ static void rs_drv_get_rate(void *mvm_r, struct ieee80211_sta *sta, return; lq_sta = mvm_sta; + + spin_lock_bh(&lq_sta->pers.lock); iwl_mvm_hwrate_to_tx_rate_v1(lq_sta->last_rate_n_flags, info->band, &info->control.rates[0]); info->control.rates[0].count = 1; @@ -2705,6 +2707,7 @@ static void rs_drv_get_rate(void *mvm_r, struct ieee80211_sta *sta, iwl_mvm_hwrate_to_tx_rate_v1(last_ucode_rate, info->band, &txrc->reported_rate); } + spin_unlock_bh(&lq_sta->pers.lock); } static void *rs_drv_alloc_sta(void *mvm_rate, struct ieee80211_sta *sta, @@ -3261,11 +3264,11 @@ void iwl_mvm_rs_tx_status(struct iwl_mvm *mvm, struct ieee80211_sta *sta, /* If it's locked we are in middle of init flow * just wait for next tx status to update the lq_sta data */ - if (!spin_trylock(&mvmsta->deflink.lq_sta.rs_drv.pers.lock)) + if (!spin_trylock_bh(&mvmsta->deflink.lq_sta.rs_drv.pers.lock)) return; __iwl_mvm_rs_tx_status(mvm, sta, tid, info, ndp); - spin_unlock(&mvmsta->deflink.lq_sta.rs_drv.pers.lock); + spin_unlock_bh(&mvmsta->deflink.lq_sta.rs_drv.pers.lock); } #ifdef CONFIG_MAC80211_DEBUGFS @@ -4114,9 +4117,9 @@ void iwl_mvm_rs_rate_init(struct iwl_mvm *mvm, } else { struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta); - spin_lock(&mvmsta->deflink.lq_sta.rs_drv.pers.lock); + spin_lock_bh(&mvmsta->deflink.lq_sta.rs_drv.pers.lock); rs_drv_rate_init(mvm, sta, band); - spin_unlock(&mvmsta->deflink.lq_sta.rs_drv.pers.lock); + spin_unlock_bh(&mvmsta->deflink.lq_sta.rs_drv.pers.lock); } } diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c index e1d02c260e69d504a35f6ddf3cf8d04b23751a9a..6226e4e54a51d2775a19174f7683d2c1db5ec729 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c @@ -691,6 +691,11 @@ void iwl_mvm_reorder_timer_expired(struct timer_list *t) rcu_read_lock(); sta = rcu_dereference(buf->mvm->fw_id_to_mac_id[sta_id]); + if (WARN_ON_ONCE(IS_ERR_OR_NULL(sta))) { + rcu_read_unlock(); + goto out; + } + mvmsta = iwl_mvm_sta_from_mac80211(sta); /* SN is set to the last expired frame + 1 */ @@ -712,6 +717,8 @@ void iwl_mvm_reorder_timer_expired(struct timer_list *t) entries[index].e.reorder_time + 1 + RX_REORDER_BUF_TIMEOUT_MQ); } + +out: spin_unlock(&buf->lock); } @@ -2512,7 +2519,7 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi, RCU_INIT_POINTER(mvm->csa_tx_blocked_vif, NULL); /* Unblock BCAST / MCAST station */ iwl_mvm_modify_all_sta_disable_tx(mvm, mvmvif, false); - cancel_delayed_work_sync(&mvm->cs_tx_unblock_dwork); + cancel_delayed_work(&mvm->cs_tx_unblock_dwork); } } diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c index 5469d634e28998a12b2c9bd2fefde11dbddbdd0a..05a54a69c1357c59f121803204c8d7d448d4c0ef 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c @@ -281,7 +281,7 @@ static void iwl_mvm_rx_agg_session_expired(struct timer_list *t) * A-MDPU and hence the timer continues to run. Then, the * timer expires and sta is NULL. */ - if (!sta) + if (IS_ERR_OR_NULL(sta)) goto unlock; mvm_sta = iwl_mvm_sta_from_mac80211(sta); @@ -2089,9 +2089,6 @@ int iwl_mvm_rm_sta(struct iwl_mvm *mvm, lockdep_assert_held(&mvm->mutex); - if (iwl_mvm_has_new_rx_api(mvm)) - kfree(mvm_sta->dup_data); - ret = iwl_mvm_drain_sta(mvm, mvm_sta, true); if (ret) return ret; @@ -3785,6 +3782,9 @@ static inline u8 *iwl_mvm_get_mac_addr(struct iwl_mvm *mvm, u8 sta_id = mvmvif->deflink.ap_sta_id; sta = rcu_dereference_protected(mvm->fw_id_to_mac_id[sta_id], lockdep_is_held(&mvm->mutex)); + if (WARN_ON_ONCE(IS_ERR_OR_NULL(sta))) + return NULL; + return sta->addr; } @@ -3822,6 +3822,11 @@ static int __iwl_mvm_set_sta_key(struct iwl_mvm *mvm, if (keyconf->cipher == WLAN_CIPHER_SUITE_TKIP) { addr = iwl_mvm_get_mac_addr(mvm, vif, sta); + if (!addr) { + IWL_ERR(mvm, "Failed to find mac address\n"); + return -EINVAL; + } + /* get phase 1 key from mac80211 */ ieee80211_get_key_rx_seq(keyconf, 0, &seq); ieee80211_get_tkip_rx_p1k(keyconf, addr, seq.tkip.iv32, p1k); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c index 10d7178f10718ebe350494c40235368eacfeeb6c..00719e1304386454f350677322764e46483fab57 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c @@ -1875,7 +1875,7 @@ static void iwl_mvm_rx_tx_cmd_agg(struct iwl_mvm *mvm, mvmsta = iwl_mvm_sta_from_staid_rcu(mvm, sta_id); sta = rcu_dereference(mvm->fw_id_to_mac_id[sta_id]); - if (WARN_ON_ONCE(!sta || !sta->wme)) { + if (WARN_ON_ONCE(IS_ERR_OR_NULL(sta) || !sta->wme)) { rcu_read_unlock(); return; } diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mac.c b/drivers/net/wireless/mediatek/mt76/mt7615/mac.c index da1d17b73a2553c39556d64a60d1b014de8f233e..64002484ccadbc27ece0021b8fcaff664433a626 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7615/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7615/mac.c @@ -914,7 +914,10 @@ void mt7615_mac_sta_poll(struct mt7615_dev *dev) msta = list_first_entry(&sta_poll_list, struct mt7615_sta, poll_list); + + spin_lock_bh(&dev->sta_poll_lock); list_del_init(&msta->poll_list); + spin_unlock_bh(&dev->sta_poll_lock); addr = mt7615_mac_wtbl_addr(dev, msta->wcid.idx) + 19 * 4; diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac2_mac.h b/drivers/net/wireless/mediatek/mt76/mt76_connac2_mac.h index a5ec0f6313850a54c7cb5170b7d9c051b8d554ca..fabf637bdf7f99f5d6b6d59d1c1dd6d7b7a1ec90 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76_connac2_mac.h +++ b/drivers/net/wireless/mediatek/mt76/mt76_connac2_mac.h @@ -173,7 +173,7 @@ enum { #define MT_TXS5_MPDU_TX_CNT GENMASK(31, 23) #define MT_TXS6_MPDU_FAIL_CNT GENMASK(31, 23) - +#define MT_TXS7_MPDU_RETRY_BYTE GENMASK(22, 0) #define MT_TXS7_MPDU_RETRY_CNT GENMASK(31, 23) /* RXD DW0 */ diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c index ee0fbfcd07d6469d94ce9a3f62c6868ef1fb753b..d39a3cc5e381f8b01e9ca2c8d0559215969ebeba 100644 --- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c @@ -608,7 +608,8 @@ bool mt76_connac2_mac_fill_txs(struct mt76_dev *dev, struct mt76_wcid *wcid, /* PPDU based reporting */ if (FIELD_GET(MT_TXS0_TXS_FORMAT, txs) > 1) { stats->tx_bytes += - le32_get_bits(txs_data[5], MT_TXS5_MPDU_TX_BYTE); + le32_get_bits(txs_data[5], MT_TXS5_MPDU_TX_BYTE) - + le32_get_bits(txs_data[7], MT_TXS7_MPDU_RETRY_BYTE); stats->tx_packets += le32_get_bits(txs_data[5], MT_TXS5_MPDU_TX_CNT); stats->tx_failed += diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c index 130eb7b4fd914c02f9b938b49e945365c36a1122..9b0f6053e0fa6280a58c48e029a91d8a0061d4e3 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mac.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mac.c @@ -1004,10 +1004,10 @@ void mt7996_mac_write_txwi(struct mt7996_dev *dev, __le32 *txwi, { struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); struct ieee80211_vif *vif = info->control.vif; - struct mt7996_vif *mvif = (struct mt7996_vif *)vif->drv_priv; u8 band_idx = (info->hw_queue & MT_TX_HW_QUEUE_PHY) >> 2; u8 p_fmt, q_idx, omac_idx = 0, wmm_idx = 0; bool is_8023 = info->flags & IEEE80211_TX_CTL_HW_80211_ENCAP; + struct mt7996_vif *mvif; u16 tx_count = 15; u32 val; bool beacon = !!(changed & (BSS_CHANGED_BEACON | @@ -1015,7 +1015,8 @@ void mt7996_mac_write_txwi(struct mt7996_dev *dev, __le32 *txwi, bool inband_disc = !!(changed & (BSS_CHANGED_UNSOL_BCAST_PROBE_RESP | BSS_CHANGED_FILS_DISCOVERY)); - if (vif) { + mvif = vif ? (struct mt7996_vif *)vif->drv_priv : NULL; + if (mvif) { omac_idx = mvif->mt76.omac_idx; wmm_idx = mvif->mt76.wmm_idx; band_idx = mvif->mt76.band_idx; @@ -1081,14 +1082,18 @@ void mt7996_mac_write_txwi(struct mt7996_dev *dev, __le32 *txwi, struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data; bool mcast = ieee80211_is_data(hdr->frame_control) && is_multicast_ether_addr(hdr->addr1); - u8 idx = mvif->basic_rates_idx; + u8 idx = MT7996_BASIC_RATES_TBL; - if (mcast && mvif->mcast_rates_idx) - idx = mvif->mcast_rates_idx; - else if (beacon && mvif->beacon_rates_idx) - idx = mvif->beacon_rates_idx; + if (mvif) { + if (mcast && mvif->mcast_rates_idx) + idx = mvif->mcast_rates_idx; + else if (beacon && mvif->beacon_rates_idx) + idx = mvif->beacon_rates_idx; + else + idx = mvif->basic_rates_idx; + } - txwi[6] |= FIELD_PREP(MT_TXD6_TX_RATE, idx); + txwi[6] |= cpu_to_le32(FIELD_PREP(MT_TXD6_TX_RATE, idx)); txwi[3] |= cpu_to_le32(MT_TXD3_BA_DISABLE); } } diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h index 8eafbf1cee718dc31045bdfb71efe94dfb0a6a55..808c1c895113dacf4190674fa6aae9e7ffe121e9 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h @@ -1803,6 +1803,7 @@ struct rtl8xxxu_priv { u32 rege9c; u32 regeb4; u32 regebc; + u32 regrcr; int next_mbox; int nr_out_eps; diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c index fd8c8c6d53d60b771da8ff0332c669167b203be4..831639d73657be9c338ec21d6cf4108331c40ac3 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c @@ -4171,6 +4171,7 @@ static int rtl8xxxu_init_device(struct ieee80211_hw *hw) RCR_ACCEPT_MGMT_FRAME | RCR_HTC_LOC_CTRL | RCR_APPEND_PHYSTAT | RCR_APPEND_ICV | RCR_APPEND_MIC; rtl8xxxu_write32(priv, REG_RCR, val32); + priv->regrcr = val32; if (fops->init_reg_rxfltmap) { /* Accept all data frames */ @@ -6501,7 +6502,7 @@ static void rtl8xxxu_configure_filter(struct ieee80211_hw *hw, unsigned int *total_flags, u64 multicast) { struct rtl8xxxu_priv *priv = hw->priv; - u32 rcr = rtl8xxxu_read32(priv, REG_RCR); + u32 rcr = priv->regrcr; dev_dbg(&priv->udev->dev, "%s: changed_flags %08x, total_flags %08x\n", __func__, changed_flags, *total_flags); @@ -6547,6 +6548,7 @@ static void rtl8xxxu_configure_filter(struct ieee80211_hw *hw, */ rtl8xxxu_write32(priv, REG_RCR, rcr); + priv->regrcr = rcr; *total_flags &= (FIF_ALLMULTI | FIF_FCSFAIL | FIF_BCN_PRBRESP_PROMISC | FIF_CONTROL | FIF_OTHER_BSS | FIF_PSPOLL | diff --git a/drivers/net/wireless/realtek/rtw88/mac80211.c b/drivers/net/wireless/realtek/rtw88/mac80211.c index 7aa6edad0d012738887d45febf34a54c8065d6e5..144618bb94c8654ad731dc872286dc068d0df801 100644 --- a/drivers/net/wireless/realtek/rtw88/mac80211.c +++ b/drivers/net/wireless/realtek/rtw88/mac80211.c @@ -88,15 +88,6 @@ static int rtw_ops_config(struct ieee80211_hw *hw, u32 changed) } } - if (changed & IEEE80211_CONF_CHANGE_PS) { - if (hw->conf.flags & IEEE80211_CONF_PS) { - rtwdev->ps_enabled = true; - } else { - rtwdev->ps_enabled = false; - rtw_leave_lps(rtwdev); - } - } - if (changed & IEEE80211_CONF_CHANGE_CHANNEL) rtw_set_channel(rtwdev); @@ -213,6 +204,7 @@ static int rtw_ops_add_interface(struct ieee80211_hw *hw, config |= PORT_SET_BCN_CTRL; rtw_vif_port_config(rtwdev, rtwvif, config); rtw_core_port_switch(rtwdev, vif); + rtw_recalc_lps(rtwdev, vif); mutex_unlock(&rtwdev->mutex); @@ -244,6 +236,7 @@ static void rtw_ops_remove_interface(struct ieee80211_hw *hw, config |= PORT_SET_BCN_CTRL; rtw_vif_port_config(rtwdev, rtwvif, config); clear_bit(rtwvif->port, rtwdev->hw_port); + rtw_recalc_lps(rtwdev, NULL); mutex_unlock(&rtwdev->mutex); } @@ -438,6 +431,9 @@ static void rtw_ops_bss_info_changed(struct ieee80211_hw *hw, if (changed & BSS_CHANGED_ERP_SLOT) rtw_conf_tx(rtwdev, rtwvif); + if (changed & BSS_CHANGED_PS) + rtw_recalc_lps(rtwdev, NULL); + rtw_vif_port_config(rtwdev, rtwvif, config); mutex_unlock(&rtwdev->mutex); @@ -918,7 +914,7 @@ static void rtw_ops_sta_rc_update(struct ieee80211_hw *hw, struct rtw_sta_info *si = (struct rtw_sta_info *)sta->drv_priv; if (changed & IEEE80211_RC_BW_CHANGED) - rtw_update_sta_info(rtwdev, si, true); + ieee80211_queue_work(rtwdev->hw, &si->rc_work); } const struct ieee80211_ops rtw_ops = { diff --git a/drivers/net/wireless/realtek/rtw88/main.c b/drivers/net/wireless/realtek/rtw88/main.c index 5bf6b45815578feb766453db223e7a71c93119e7..9447a3aae3b5e16d4beb86c2129e283847c5fc70 100644 --- a/drivers/net/wireless/realtek/rtw88/main.c +++ b/drivers/net/wireless/realtek/rtw88/main.c @@ -271,8 +271,8 @@ static void rtw_watch_dog_work(struct work_struct *work) * more than two stations associated to the AP, then we can not enter * lps, because fw does not handle the overlapped beacon interval * - * mac80211 should iterate vifs and determine if driver can enter - * ps by passing IEEE80211_CONF_PS to us, all we need to do is to + * rtw_recalc_lps() iterate vifs and determine if driver can enter + * ps by vif->type and vif->cfg.ps, all we need to do here is to * get that vif and check if device is having traffic more than the * threshold. */ @@ -319,6 +319,17 @@ static u8 rtw_acquire_macid(struct rtw_dev *rtwdev) return mac_id; } +static void rtw_sta_rc_work(struct work_struct *work) +{ + struct rtw_sta_info *si = container_of(work, struct rtw_sta_info, + rc_work); + struct rtw_dev *rtwdev = si->rtwdev; + + mutex_lock(&rtwdev->mutex); + rtw_update_sta_info(rtwdev, si, true); + mutex_unlock(&rtwdev->mutex); +} + int rtw_sta_add(struct rtw_dev *rtwdev, struct ieee80211_sta *sta, struct ieee80211_vif *vif) { @@ -329,12 +340,14 @@ int rtw_sta_add(struct rtw_dev *rtwdev, struct ieee80211_sta *sta, if (si->mac_id >= RTW_MAX_MAC_ID_NUM) return -ENOSPC; + si->rtwdev = rtwdev; si->sta = sta; si->vif = vif; si->init_ra_lv = 1; ewma_rssi_init(&si->avg_rssi); for (i = 0; i < ARRAY_SIZE(sta->txq); i++) rtw_txq_init(rtwdev, sta->txq[i]); + INIT_WORK(&si->rc_work, rtw_sta_rc_work); rtw_update_sta_info(rtwdev, si, true); rtw_fw_media_status_report(rtwdev, si->mac_id, true); @@ -353,6 +366,8 @@ void rtw_sta_remove(struct rtw_dev *rtwdev, struct ieee80211_sta *sta, struct rtw_sta_info *si = (struct rtw_sta_info *)sta->drv_priv; int i; + cancel_work_sync(&si->rc_work); + rtw_release_macid(rtwdev, si->mac_id); if (fw_exist) rtw_fw_media_status_report(rtwdev, si->mac_id, false); diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h index a563285e90ede577f287d7329f1246cdb3569a14..9e841f6991a9a4d0d984fe5892282bd8d473ade4 100644 --- a/drivers/net/wireless/realtek/rtw88/main.h +++ b/drivers/net/wireless/realtek/rtw88/main.h @@ -743,6 +743,7 @@ struct rtw_txq { DECLARE_EWMA(rssi, 10, 16); struct rtw_sta_info { + struct rtw_dev *rtwdev; struct ieee80211_sta *sta; struct ieee80211_vif *vif; @@ -767,6 +768,8 @@ struct rtw_sta_info { bool use_cfg_mask; struct cfg80211_bitrate_mask *mask; + + struct work_struct rc_work; }; enum rtw_bfee_role { diff --git a/drivers/net/wireless/realtek/rtw88/ps.c b/drivers/net/wireless/realtek/rtw88/ps.c index 996365575f44fe7d16e0ce10a7d77586e8ca4367..53933fb38a33084dcbdb81f68ba8f524e73feb02 100644 --- a/drivers/net/wireless/realtek/rtw88/ps.c +++ b/drivers/net/wireless/realtek/rtw88/ps.c @@ -299,3 +299,46 @@ void rtw_leave_lps_deep(struct rtw_dev *rtwdev) __rtw_leave_lps_deep(rtwdev); } + +struct rtw_vif_recalc_lps_iter_data { + struct rtw_dev *rtwdev; + struct ieee80211_vif *found_vif; + int count; +}; + +static void __rtw_vif_recalc_lps(struct rtw_vif_recalc_lps_iter_data *data, + struct ieee80211_vif *vif) +{ + if (data->count < 0) + return; + + if (vif->type != NL80211_IFTYPE_STATION) { + data->count = -1; + return; + } + + data->count++; + data->found_vif = vif; +} + +static void rtw_vif_recalc_lps_iter(void *data, u8 *mac, + struct ieee80211_vif *vif) +{ + __rtw_vif_recalc_lps(data, vif); +} + +void rtw_recalc_lps(struct rtw_dev *rtwdev, struct ieee80211_vif *new_vif) +{ + struct rtw_vif_recalc_lps_iter_data data = { .rtwdev = rtwdev }; + + if (new_vif) + __rtw_vif_recalc_lps(&data, new_vif); + rtw_iterate_vifs(rtwdev, rtw_vif_recalc_lps_iter, &data); + + if (data.count == 1 && data.found_vif->cfg.ps) { + rtwdev->ps_enabled = true; + } else { + rtwdev->ps_enabled = false; + rtw_leave_lps(rtwdev); + } +} diff --git a/drivers/net/wireless/realtek/rtw88/ps.h b/drivers/net/wireless/realtek/rtw88/ps.h index c194386f6db53a378489255e4dfe188e2733d983..5ae83d2526cfd2730daaeed07f1d0ca56527a068 100644 --- a/drivers/net/wireless/realtek/rtw88/ps.h +++ b/drivers/net/wireless/realtek/rtw88/ps.h @@ -23,4 +23,6 @@ void rtw_enter_lps(struct rtw_dev *rtwdev, u8 port_id); void rtw_leave_lps(struct rtw_dev *rtwdev); void rtw_leave_lps_deep(struct rtw_dev *rtwdev); enum rtw_lps_deep_mode rtw_get_lps_deep_mode(struct rtw_dev *rtwdev); +void rtw_recalc_lps(struct rtw_dev *rtwdev, struct ieee80211_vif *new_vif); + #endif diff --git a/drivers/net/wireless/realtek/rtw88/sdio.c b/drivers/net/wireless/realtek/rtw88/sdio.c index af0459a79899f65bd675d7078d00b44c8f617b4b..06fce7c3addaa395bc5c622621bf8bc7c0537dae 100644 --- a/drivers/net/wireless/realtek/rtw88/sdio.c +++ b/drivers/net/wireless/realtek/rtw88/sdio.c @@ -87,11 +87,6 @@ static void rtw_sdio_writew(struct rtw_dev *rtwdev, u16 val, u32 addr, u8 buf[2]; int i; - if (rtw_sdio_use_memcpy_io(rtwdev, addr, 2)) { - sdio_writew(rtwsdio->sdio_func, val, addr, err_ret); - return; - } - *(__le16 *)buf = cpu_to_le16(val); for (i = 0; i < 2; i++) { @@ -125,9 +120,6 @@ static u16 rtw_sdio_readw(struct rtw_dev *rtwdev, u32 addr, int *err_ret) u8 buf[2]; int i; - if (rtw_sdio_use_memcpy_io(rtwdev, addr, 2)) - return sdio_readw(rtwsdio->sdio_func, addr, err_ret); - for (i = 0; i < 2; i++) { buf[i] = sdio_readb(rtwsdio->sdio_func, addr + i, err_ret); if (*err_ret) diff --git a/drivers/net/wireless/realtek/rtw88/usb.h b/drivers/net/wireless/realtek/rtw88/usb.h index 30647f0dd61c66fb6433c7976752b916d8be3e53..ad1d7955c6a51e545b92c5e5bbd316f9120bf438 100644 --- a/drivers/net/wireless/realtek/rtw88/usb.h +++ b/drivers/net/wireless/realtek/rtw88/usb.h @@ -78,7 +78,7 @@ struct rtw_usb { u8 pipe_interrupt; u8 pipe_in; u8 out_ep[RTW_USB_EP_MAX]; - u8 qsel_to_ep[TX_DESC_QSEL_MAX]; + int qsel_to_ep[TX_DESC_QSEL_MAX]; u8 usb_txagg_num; struct workqueue_struct *txwq, *rxwq; diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c index 7fc0a26a4d7317cde259f2521607d5f6378d70e0..bad864d56bd5cc7a16dc4c3eca81121db00efa89 100644 --- a/drivers/net/wireless/realtek/rtw89/core.c +++ b/drivers/net/wireless/realtek/rtw89/core.c @@ -2531,9 +2531,6 @@ static void rtw89_vif_enter_lps(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwv rtwvif->tdls_peer) return; - if (rtwdev->total_sta_assoc > 1) - return; - if (rtwvif->offchan) return; diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c index b8019cfc11b20dd5889b6106ba905d5999f4aab3..512de491a064b1b741ad34802b423e7a2b59a968 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.c +++ b/drivers/net/wireless/realtek/rtw89/mac.c @@ -1425,6 +1425,8 @@ const struct rtw89_mac_size_set rtw89_mac_size = { .wde_size4 = {RTW89_WDE_PG_64, 0, 4096,}, /* PCIE 64 */ .wde_size6 = {RTW89_WDE_PG_64, 512, 0,}, + /* 8852B PCIE SCC */ + .wde_size7 = {RTW89_WDE_PG_64, 510, 2,}, /* DLFW */ .wde_size9 = {RTW89_WDE_PG_64, 0, 1024,}, /* 8852C DLFW */ @@ -1449,6 +1451,8 @@ const struct rtw89_mac_size_set rtw89_mac_size = { .wde_qt4 = {0, 0, 0, 0,}, /* PCIE 64 */ .wde_qt6 = {448, 48, 0, 16,}, + /* 8852B PCIE SCC */ + .wde_qt7 = {446, 48, 0, 16,}, /* 8852C DLFW */ .wde_qt17 = {0, 0, 0, 0,}, /* 8852C PCIE SCC */ diff --git a/drivers/net/wireless/realtek/rtw89/mac.h b/drivers/net/wireless/realtek/rtw89/mac.h index a8d9847ef0b49e1df2f7ffa19abc9729cf4db7b9..6ba633ccdd0377ff571a5c0bbf2017c8bb11807e 100644 --- a/drivers/net/wireless/realtek/rtw89/mac.h +++ b/drivers/net/wireless/realtek/rtw89/mac.h @@ -792,6 +792,7 @@ struct rtw89_mac_size_set { const struct rtw89_dle_size wde_size0; const struct rtw89_dle_size wde_size4; const struct rtw89_dle_size wde_size6; + const struct rtw89_dle_size wde_size7; const struct rtw89_dle_size wde_size9; const struct rtw89_dle_size wde_size18; const struct rtw89_dle_size wde_size19; @@ -804,6 +805,7 @@ struct rtw89_mac_size_set { const struct rtw89_wde_quota wde_qt0; const struct rtw89_wde_quota wde_qt4; const struct rtw89_wde_quota wde_qt6; + const struct rtw89_wde_quota wde_qt7; const struct rtw89_wde_quota wde_qt17; const struct rtw89_wde_quota wde_qt18; const struct rtw89_ple_quota ple_qt4; diff --git a/drivers/net/wireless/realtek/rtw89/mac80211.c b/drivers/net/wireless/realtek/rtw89/mac80211.c index ee4588b61b8f66bd6157e5ada41a34e96f47d9c0..c42e310690351a37fcd7053a30362918155b6cfe 100644 --- a/drivers/net/wireless/realtek/rtw89/mac80211.c +++ b/drivers/net/wireless/realtek/rtw89/mac80211.c @@ -89,15 +89,6 @@ static int rtw89_ops_config(struct ieee80211_hw *hw, u32 changed) !(hw->conf.flags & IEEE80211_CONF_IDLE)) rtw89_leave_ips(rtwdev); - if (changed & IEEE80211_CONF_CHANGE_PS) { - if (hw->conf.flags & IEEE80211_CONF_PS) { - rtwdev->lps_enabled = true; - } else { - rtw89_leave_lps(rtwdev); - rtwdev->lps_enabled = false; - } - } - if (changed & IEEE80211_CONF_CHANGE_CHANNEL) { rtw89_config_entity_chandef(rtwdev, RTW89_SUB_ENTITY_0, &hw->conf.chandef); @@ -168,6 +159,8 @@ static int rtw89_ops_add_interface(struct ieee80211_hw *hw, rtw89_core_txq_init(rtwdev, vif->txq); rtw89_btc_ntfy_role_info(rtwdev, rtwvif, NULL, BTC_ROLE_START); + + rtw89_recalc_lps(rtwdev); out: mutex_unlock(&rtwdev->mutex); @@ -192,6 +185,7 @@ static void rtw89_ops_remove_interface(struct ieee80211_hw *hw, rtw89_mac_remove_vif(rtwdev, rtwvif); rtw89_core_release_bit_map(rtwdev->hw_port, rtwvif->port); list_del_init(&rtwvif->list); + rtw89_recalc_lps(rtwdev); rtw89_enter_ips_by_hwflags(rtwdev); mutex_unlock(&rtwdev->mutex); @@ -451,6 +445,9 @@ static void rtw89_ops_bss_info_changed(struct ieee80211_hw *hw, if (changed & BSS_CHANGED_CQM) rtw89_fw_h2c_set_bcn_fltr_cfg(rtwdev, vif, true); + if (changed & BSS_CHANGED_PS) + rtw89_recalc_lps(rtwdev); + mutex_unlock(&rtwdev->mutex); } diff --git a/drivers/net/wireless/realtek/rtw89/ps.c b/drivers/net/wireless/realtek/rtw89/ps.c index fa94335f699aa47eba2f76a686f36b7cce8507ec..84201ef19c176c9426ac2cbeb3bbbc7a760d83a5 100644 --- a/drivers/net/wireless/realtek/rtw89/ps.c +++ b/drivers/net/wireless/realtek/rtw89/ps.c @@ -252,3 +252,29 @@ void rtw89_process_p2p_ps(struct rtw89_dev *rtwdev, struct ieee80211_vif *vif) rtw89_p2p_disable_all_noa(rtwdev, vif); rtw89_p2p_update_noa(rtwdev, vif); } + +void rtw89_recalc_lps(struct rtw89_dev *rtwdev) +{ + struct ieee80211_vif *vif, *found_vif = NULL; + struct rtw89_vif *rtwvif; + int count = 0; + + rtw89_for_each_rtwvif(rtwdev, rtwvif) { + vif = rtwvif_to_vif(rtwvif); + + if (vif->type != NL80211_IFTYPE_STATION) { + count = 0; + break; + } + + count++; + found_vif = vif; + } + + if (count == 1 && found_vif->cfg.ps) { + rtwdev->lps_enabled = true; + } else { + rtw89_leave_lps(rtwdev); + rtwdev->lps_enabled = false; + } +} diff --git a/drivers/net/wireless/realtek/rtw89/ps.h b/drivers/net/wireless/realtek/rtw89/ps.h index 73c008db042643e953ff500f72f8afa2647dbeda..4c18f49204b285aaaf2ce3a7cf6ea57b0622163e 100644 --- a/drivers/net/wireless/realtek/rtw89/ps.h +++ b/drivers/net/wireless/realtek/rtw89/ps.h @@ -15,6 +15,7 @@ void rtw89_enter_ips(struct rtw89_dev *rtwdev); void rtw89_leave_ips(struct rtw89_dev *rtwdev); void rtw89_set_coex_ctrl_lps(struct rtw89_dev *rtwdev, bool btc_ctrl); void rtw89_process_p2p_ps(struct rtw89_dev *rtwdev, struct ieee80211_vif *vif); +void rtw89_recalc_lps(struct rtw89_dev *rtwdev); static inline void rtw89_leave_ips_by_hwflags(struct rtw89_dev *rtwdev) { diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c index eaa2ea0586bc621852d2b318d107a4eb95099ee4..6da1b603a9a95c096fc531b727ea080c84ebeaa2 100644 --- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c +++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c @@ -18,25 +18,25 @@ RTW8852B_FW_BASENAME "-" __stringify(RTW8852B_FW_FORMAT_MAX) ".bin" static const struct rtw89_hfc_ch_cfg rtw8852b_hfc_chcfg_pcie[] = { - {5, 343, grp_0}, /* ACH 0 */ - {5, 343, grp_0}, /* ACH 1 */ - {5, 343, grp_0}, /* ACH 2 */ - {5, 343, grp_0}, /* ACH 3 */ + {5, 341, grp_0}, /* ACH 0 */ + {5, 341, grp_0}, /* ACH 1 */ + {4, 342, grp_0}, /* ACH 2 */ + {4, 342, grp_0}, /* ACH 3 */ {0, 0, grp_0}, /* ACH 4 */ {0, 0, grp_0}, /* ACH 5 */ {0, 0, grp_0}, /* ACH 6 */ {0, 0, grp_0}, /* ACH 7 */ - {4, 344, grp_0}, /* B0MGQ */ - {4, 344, grp_0}, /* B0HIQ */ + {4, 342, grp_0}, /* B0MGQ */ + {4, 342, grp_0}, /* B0HIQ */ {0, 0, grp_0}, /* B1MGQ */ {0, 0, grp_0}, /* B1HIQ */ {40, 0, 0} /* FWCMDQ */ }; static const struct rtw89_hfc_pub_cfg rtw8852b_hfc_pubcfg_pcie = { - 448, /* Group 0 */ + 446, /* Group 0 */ 0, /* Group 1 */ - 448, /* Public Max */ + 446, /* Public Max */ 0 /* WP threshold */ }; @@ -49,13 +49,13 @@ static const struct rtw89_hfc_param_ini rtw8852b_hfc_param_ini_pcie[] = { }; static const struct rtw89_dle_mem rtw8852b_dle_mem_pcie[] = { - [RTW89_QTA_SCC] = {RTW89_QTA_SCC, &rtw89_mac_size.wde_size6, - &rtw89_mac_size.ple_size6, &rtw89_mac_size.wde_qt6, - &rtw89_mac_size.wde_qt6, &rtw89_mac_size.ple_qt18, + [RTW89_QTA_SCC] = {RTW89_QTA_SCC, &rtw89_mac_size.wde_size7, + &rtw89_mac_size.ple_size6, &rtw89_mac_size.wde_qt7, + &rtw89_mac_size.wde_qt7, &rtw89_mac_size.ple_qt18, &rtw89_mac_size.ple_qt58}, - [RTW89_QTA_WOW] = {RTW89_QTA_WOW, &rtw89_mac_size.wde_size6, - &rtw89_mac_size.ple_size6, &rtw89_mac_size.wde_qt6, - &rtw89_mac_size.wde_qt6, &rtw89_mac_size.ple_qt18, + [RTW89_QTA_WOW] = {RTW89_QTA_WOW, &rtw89_mac_size.wde_size7, + &rtw89_mac_size.ple_size6, &rtw89_mac_size.wde_qt7, + &rtw89_mac_size.wde_qt7, &rtw89_mac_size.ple_qt18, &rtw89_mac_size.ple_qt_52b_wow}, [RTW89_QTA_DLFW] = {RTW89_QTA_DLFW, &rtw89_mac_size.wde_size9, &rtw89_mac_size.ple_size8, &rtw89_mac_size.wde_qt4, diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.c b/drivers/net/wireless/virtual/mac80211_hwsim.c index 9a8faaf4c6b642e9960ea593072f476f12ec3985..89c7a1420381d369930cf7e6907d08740638290f 100644 --- a/drivers/net/wireless/virtual/mac80211_hwsim.c +++ b/drivers/net/wireless/virtual/mac80211_hwsim.c @@ -5964,10 +5964,11 @@ static int hwsim_new_radio_nl(struct sk_buff *msg, struct genl_info *info) ret = -ENOMEM; goto out_free; } + param.pmsr_capa = pmsr_capa; + ret = parse_pmsr_capa(info->attrs[HWSIM_ATTR_PMSR_SUPPORT], pmsr_capa, info); if (ret) goto out_free; - param.pmsr_capa = pmsr_capa; } ret = mac80211_hwsim_new_radio(info, ¶m); diff --git a/drivers/net/wwan/iosm/iosm_ipc_imem.c b/drivers/net/wwan/iosm/iosm_ipc_imem.c index c066b0040a3fe5fb6668bd35d45694495dc1647a..829515a601b379d3acb57927b67076a35ebf16c9 100644 --- a/drivers/net/wwan/iosm/iosm_ipc_imem.c +++ b/drivers/net/wwan/iosm/iosm_ipc_imem.c @@ -565,24 +565,32 @@ static void ipc_imem_run_state_worker(struct work_struct *instance) struct ipc_mux_config mux_cfg; struct iosm_imem *ipc_imem; u8 ctrl_chl_idx = 0; + int ret; ipc_imem = container_of(instance, struct iosm_imem, run_state_worker); if (ipc_imem->phase != IPC_P_RUN) { dev_err(ipc_imem->dev, "Modem link down. Exit run state worker."); - return; + goto err_out; } if (test_and_clear_bit(IOSM_DEVLINK_INIT, &ipc_imem->flag)) ipc_devlink_deinit(ipc_imem->ipc_devlink); - if (!ipc_imem_setup_cp_mux_cap_init(ipc_imem, &mux_cfg)) - ipc_imem->mux = ipc_mux_init(&mux_cfg, ipc_imem); + ret = ipc_imem_setup_cp_mux_cap_init(ipc_imem, &mux_cfg); + if (ret < 0) + goto err_out; + + ipc_imem->mux = ipc_mux_init(&mux_cfg, ipc_imem); + if (!ipc_imem->mux) + goto err_out; + + ret = ipc_imem_wwan_channel_init(ipc_imem, mux_cfg.protocol); + if (ret < 0) + goto err_ipc_mux_deinit; - ipc_imem_wwan_channel_init(ipc_imem, mux_cfg.protocol); - if (ipc_imem->mux) - ipc_imem->mux->wwan = ipc_imem->wwan; + ipc_imem->mux->wwan = ipc_imem->wwan; while (ctrl_chl_idx < IPC_MEM_MAX_CHANNELS) { if (!ipc_chnl_cfg_get(&chnl_cfg_port, ctrl_chl_idx)) { @@ -622,6 +630,13 @@ static void ipc_imem_run_state_worker(struct work_struct *instance) /* Complete all memory stores after setting bit */ smp_mb__after_atomic(); + + return; + +err_ipc_mux_deinit: + ipc_mux_deinit(ipc_imem->mux); +err_out: + ipc_uevent_send(ipc_imem->dev, UEVENT_CD_READY_LINK_DOWN); } static void ipc_imem_handle_irq(struct iosm_imem *ipc_imem, int irq) diff --git a/drivers/net/wwan/iosm/iosm_ipc_imem_ops.c b/drivers/net/wwan/iosm/iosm_ipc_imem_ops.c index 66b90cc4c3460f4d046037c6e51181322720f56e..109cf89304888bc15e045536daca29979bc8a8f2 100644 --- a/drivers/net/wwan/iosm/iosm_ipc_imem_ops.c +++ b/drivers/net/wwan/iosm/iosm_ipc_imem_ops.c @@ -77,8 +77,8 @@ int ipc_imem_sys_wwan_transmit(struct iosm_imem *ipc_imem, } /* Initialize wwan channel */ -void ipc_imem_wwan_channel_init(struct iosm_imem *ipc_imem, - enum ipc_mux_protocol mux_type) +int ipc_imem_wwan_channel_init(struct iosm_imem *ipc_imem, + enum ipc_mux_protocol mux_type) { struct ipc_chnl_cfg chnl_cfg = { 0 }; @@ -87,7 +87,7 @@ void ipc_imem_wwan_channel_init(struct iosm_imem *ipc_imem, /* If modem version is invalid (0xffffffff), do not initialize WWAN. */ if (ipc_imem->cp_version == -1) { dev_err(ipc_imem->dev, "invalid CP version"); - return; + return -EIO; } ipc_chnl_cfg_get(&chnl_cfg, ipc_imem->nr_of_channels); @@ -104,9 +104,13 @@ void ipc_imem_wwan_channel_init(struct iosm_imem *ipc_imem, /* WWAN registration. */ ipc_imem->wwan = ipc_wwan_init(ipc_imem, ipc_imem->dev); - if (!ipc_imem->wwan) + if (!ipc_imem->wwan) { dev_err(ipc_imem->dev, "failed to register the ipc_wwan interfaces"); + return -ENOMEM; + } + + return 0; } /* Map SKB to DMA for transfer */ diff --git a/drivers/net/wwan/iosm/iosm_ipc_imem_ops.h b/drivers/net/wwan/iosm/iosm_ipc_imem_ops.h index f8afb217d9e2fb8acad63e6b6de8b4f726542db9..026c5bd0f9992f7afeb03667149e4fbbd8dc83cc 100644 --- a/drivers/net/wwan/iosm/iosm_ipc_imem_ops.h +++ b/drivers/net/wwan/iosm/iosm_ipc_imem_ops.h @@ -91,9 +91,11 @@ int ipc_imem_sys_wwan_transmit(struct iosm_imem *ipc_imem, int if_id, * MUX. * @ipc_imem: Pointer to iosm_imem struct. * @mux_type: Type of mux protocol. + * + * Return: 0 on success and failure value on error */ -void ipc_imem_wwan_channel_init(struct iosm_imem *ipc_imem, - enum ipc_mux_protocol mux_type); +int ipc_imem_wwan_channel_init(struct iosm_imem *ipc_imem, + enum ipc_mux_protocol mux_type); /** * ipc_imem_sys_devlink_open - Open a Flash/CD Channel link to CP diff --git a/drivers/net/wwan/t7xx/t7xx_pci.c b/drivers/net/wwan/t7xx/t7xx_pci.c index 226fc1703e90f2d56728cfbf73b64c9f5bf49849..91256e005b846f17f869d62ceb6ac645d698c5fc 100644 --- a/drivers/net/wwan/t7xx/t7xx_pci.c +++ b/drivers/net/wwan/t7xx/t7xx_pci.c @@ -45,6 +45,7 @@ #define T7XX_PCI_IREG_BASE 0 #define T7XX_PCI_EREG_BASE 2 +#define T7XX_INIT_TIMEOUT 20 #define PM_SLEEP_DIS_TIMEOUT_MS 20 #define PM_ACK_TIMEOUT_MS 1500 #define PM_AUTOSUSPEND_MS 20000 @@ -96,6 +97,7 @@ static int t7xx_pci_pm_init(struct t7xx_pci_dev *t7xx_dev) spin_lock_init(&t7xx_dev->md_pm_lock); init_completion(&t7xx_dev->sleep_lock_acquire); init_completion(&t7xx_dev->pm_sr_ack); + init_completion(&t7xx_dev->init_done); atomic_set(&t7xx_dev->md_pm_state, MTK_PM_INIT); device_init_wakeup(&pdev->dev, true); @@ -124,6 +126,7 @@ void t7xx_pci_pm_init_late(struct t7xx_pci_dev *t7xx_dev) pm_runtime_mark_last_busy(&t7xx_dev->pdev->dev); pm_runtime_allow(&t7xx_dev->pdev->dev); pm_runtime_put_noidle(&t7xx_dev->pdev->dev); + complete_all(&t7xx_dev->init_done); } static int t7xx_pci_pm_reinit(struct t7xx_pci_dev *t7xx_dev) @@ -529,6 +532,20 @@ static void t7xx_pci_shutdown(struct pci_dev *pdev) __t7xx_pci_pm_suspend(pdev); } +static int t7xx_pci_pm_prepare(struct device *dev) +{ + struct pci_dev *pdev = to_pci_dev(dev); + struct t7xx_pci_dev *t7xx_dev; + + t7xx_dev = pci_get_drvdata(pdev); + if (!wait_for_completion_timeout(&t7xx_dev->init_done, T7XX_INIT_TIMEOUT * HZ)) { + dev_warn(dev, "Not ready for system sleep.\n"); + return -ETIMEDOUT; + } + + return 0; +} + static int t7xx_pci_pm_suspend(struct device *dev) { return __t7xx_pci_pm_suspend(to_pci_dev(dev)); @@ -555,6 +572,7 @@ static int t7xx_pci_pm_runtime_resume(struct device *dev) } static const struct dev_pm_ops t7xx_pci_pm_ops = { + .prepare = t7xx_pci_pm_prepare, .suspend = t7xx_pci_pm_suspend, .resume = t7xx_pci_pm_resume, .resume_noirq = t7xx_pci_pm_resume_noirq, diff --git a/drivers/net/wwan/t7xx/t7xx_pci.h b/drivers/net/wwan/t7xx/t7xx_pci.h index 112efa534eaceffb47a8b049d5126eae79c971e3..f08f1ab7446917b0b84f053fe31a0f9902d76f90 100644 --- a/drivers/net/wwan/t7xx/t7xx_pci.h +++ b/drivers/net/wwan/t7xx/t7xx_pci.h @@ -69,6 +69,7 @@ struct t7xx_pci_dev { struct t7xx_modem *md; struct t7xx_ccmni_ctrl *ccmni_ctlb; bool rgu_pci_irq_en; + struct completion init_done; /* Low Power Items */ struct list_head md_pm_entities; diff --git a/drivers/nfc/nfcsim.c b/drivers/nfc/nfcsim.c index 44eeb17ae48d911fbb75b92d0d5d8987d5e5bea6..a55381f80cd6e4d98eaece05340269ab6a7f45ed 100644 --- a/drivers/nfc/nfcsim.c +++ b/drivers/nfc/nfcsim.c @@ -336,10 +336,6 @@ static struct dentry *nfcsim_debugfs_root; static void nfcsim_debugfs_init(void) { nfcsim_debugfs_root = debugfs_create_dir("nfcsim", NULL); - - if (!nfcsim_debugfs_root) - pr_err("Could not create debugfs entry\n"); - } static void nfcsim_debugfs_remove(void) diff --git a/drivers/nvme/host/constants.c b/drivers/nvme/host/constants.c index bc523ca0225486776eba989fab6ed758b4c569d0..5e4f8848dce08e912d5c4d85b84ae70cf55c14e5 100644 --- a/drivers/nvme/host/constants.c +++ b/drivers/nvme/host/constants.c @@ -21,7 +21,7 @@ static const char * const nvme_ops[] = { [nvme_cmd_resv_release] = "Reservation Release", [nvme_cmd_zone_mgmt_send] = "Zone Management Send", [nvme_cmd_zone_mgmt_recv] = "Zone Management Receive", - [nvme_cmd_zone_append] = "Zone Management Append", + [nvme_cmd_zone_append] = "Zone Append", }; static const char * const nvme_admin_ops[] = { diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index ccb6eb1282f82d8b801b38172261be1052119a6a..3ec38e2b91732691ae53ac50eb7fc8a45add5465 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -397,7 +397,16 @@ void nvme_complete_rq(struct request *req) trace_nvme_complete_rq(req); nvme_cleanup_cmd(req); - if (ctrl->kas) + /* + * Completions of long-running commands should not be able to + * defer sending of periodic keep alives, since the controller + * may have completed processing such commands a long time ago + * (arbitrarily close to command submission time). + * req->deadline - req->timeout is the command submission time + * in jiffies. + */ + if (ctrl->kas && + req->deadline - req->timeout >= ctrl->ka_last_check_time) ctrl->comp_seen = true; switch (nvme_decide_disposition(req)) { @@ -1115,7 +1124,7 @@ u32 nvme_passthru_start(struct nvme_ctrl *ctrl, struct nvme_ns *ns, u8 opcode) } EXPORT_SYMBOL_NS_GPL(nvme_passthru_start, NVME_TARGET_PASSTHRU); -void nvme_passthru_end(struct nvme_ctrl *ctrl, u32 effects, +void nvme_passthru_end(struct nvme_ctrl *ctrl, struct nvme_ns *ns, u32 effects, struct nvme_command *cmd, int status) { if (effects & NVME_CMD_EFFECTS_CSE_MASK) { @@ -1132,6 +1141,8 @@ void nvme_passthru_end(struct nvme_ctrl *ctrl, u32 effects, nvme_queue_scan(ctrl); flush_work(&ctrl->scan_work); } + if (ns) + return; switch (cmd->common.opcode) { case nvme_admin_set_features: @@ -1161,9 +1172,25 @@ EXPORT_SYMBOL_NS_GPL(nvme_passthru_end, NVME_TARGET_PASSTHRU); * The host should send Keep Alive commands at half of the Keep Alive Timeout * accounting for transport roundtrip times [..]. */ +static unsigned long nvme_keep_alive_work_period(struct nvme_ctrl *ctrl) +{ + unsigned long delay = ctrl->kato * HZ / 2; + + /* + * When using Traffic Based Keep Alive, we need to run + * nvme_keep_alive_work at twice the normal frequency, as one + * command completion can postpone sending a keep alive command + * by up to twice the delay between runs. + */ + if (ctrl->ctratt & NVME_CTRL_ATTR_TBKAS) + delay /= 2; + return delay; +} + static void nvme_queue_keep_alive_work(struct nvme_ctrl *ctrl) { - queue_delayed_work(nvme_wq, &ctrl->ka_work, ctrl->kato * HZ / 2); + queue_delayed_work(nvme_wq, &ctrl->ka_work, + nvme_keep_alive_work_period(ctrl)); } static enum rq_end_io_ret nvme_keep_alive_end_io(struct request *rq, @@ -1172,6 +1199,20 @@ static enum rq_end_io_ret nvme_keep_alive_end_io(struct request *rq, struct nvme_ctrl *ctrl = rq->end_io_data; unsigned long flags; bool startka = false; + unsigned long rtt = jiffies - (rq->deadline - rq->timeout); + unsigned long delay = nvme_keep_alive_work_period(ctrl); + + /* + * Subtract off the keepalive RTT so nvme_keep_alive_work runs + * at the desired frequency. + */ + if (rtt <= delay) { + delay -= rtt; + } else { + dev_warn(ctrl->device, "long keepalive RTT (%u ms)\n", + jiffies_to_msecs(rtt)); + delay = 0; + } blk_mq_free_request(rq); @@ -1182,6 +1223,7 @@ static enum rq_end_io_ret nvme_keep_alive_end_io(struct request *rq, return RQ_END_IO_NONE; } + ctrl->ka_last_check_time = jiffies; ctrl->comp_seen = false; spin_lock_irqsave(&ctrl->lock, flags); if (ctrl->state == NVME_CTRL_LIVE || @@ -1189,7 +1231,7 @@ static enum rq_end_io_ret nvme_keep_alive_end_io(struct request *rq, startka = true; spin_unlock_irqrestore(&ctrl->lock, flags); if (startka) - nvme_queue_keep_alive_work(ctrl); + queue_delayed_work(nvme_wq, &ctrl->ka_work, delay); return RQ_END_IO_NONE; } @@ -1200,6 +1242,8 @@ static void nvme_keep_alive_work(struct work_struct *work) bool comp_seen = ctrl->comp_seen; struct request *rq; + ctrl->ka_last_check_time = jiffies; + if ((ctrl->ctratt & NVME_CTRL_ATTR_TBKAS) && comp_seen) { dev_dbg(ctrl->device, "reschedule traffic based keep-alive timer\n"); @@ -3585,6 +3629,9 @@ static ssize_t nvme_sysfs_delete(struct device *dev, { struct nvme_ctrl *ctrl = dev_get_drvdata(dev); + if (!test_bit(NVME_CTRL_STARTED_ONCE, &ctrl->flags)) + return -EBUSY; + if (device_remove_file_self(dev, attr)) nvme_delete_ctrl_sync(ctrl); return count; @@ -5045,7 +5092,7 @@ void nvme_start_ctrl(struct nvme_ctrl *ctrl) * that were missed. We identify persistent discovery controllers by * checking that they started once before, hence are reconnecting back. */ - if (test_and_set_bit(NVME_CTRL_STARTED_ONCE, &ctrl->flags) && + if (test_bit(NVME_CTRL_STARTED_ONCE, &ctrl->flags) && nvme_discovery_ctrl(ctrl)) nvme_change_uevent(ctrl, "NVME_EVENT=rediscover"); @@ -5056,6 +5103,7 @@ void nvme_start_ctrl(struct nvme_ctrl *ctrl) } nvme_change_uevent(ctrl, "NVME_EVENT=connected"); + set_bit(NVME_CTRL_STARTED_ONCE, &ctrl->flags); } EXPORT_SYMBOL_GPL(nvme_start_ctrl); diff --git a/drivers/nvme/host/hwmon.c b/drivers/nvme/host/hwmon.c index 9e6e56c20ec993bfca3ee2eb18d2e371d7225230..316f3e4ca7cc60da613e5f246f9e5a861d60e14f 100644 --- a/drivers/nvme/host/hwmon.c +++ b/drivers/nvme/host/hwmon.c @@ -163,7 +163,9 @@ static umode_t nvme_hwmon_is_visible(const void *_data, case hwmon_temp_max: case hwmon_temp_min: if ((!channel && data->ctrl->wctemp) || - (channel && data->log->temp_sensor[channel - 1])) { + (channel && data->log->temp_sensor[channel - 1] && + !(data->ctrl->quirks & + NVME_QUIRK_NO_SECONDARY_TEMP_THRESH))) { if (data->ctrl->quirks & NVME_QUIRK_NO_TEMP_THRESH_CHANGE) return 0444; diff --git a/drivers/nvme/host/ioctl.c b/drivers/nvme/host/ioctl.c index 81c5c9e38477470c1134506b78bdd09ceaa36be5..f15e7330b75ac5c310bf05f88cd1275469461097 100644 --- a/drivers/nvme/host/ioctl.c +++ b/drivers/nvme/host/ioctl.c @@ -254,7 +254,7 @@ static int nvme_submit_user_cmd(struct request_queue *q, blk_mq_free_request(req); if (effects) - nvme_passthru_end(ctrl, effects, cmd, ret); + nvme_passthru_end(ctrl, ns, effects, cmd, ret); return ret; } diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c index 9171452e2f6d4e2eed95d0c6ef9ec0cbf05bc7ef..2bc159a318ff0af19a06aa668aeb94bf8167ff1d 100644 --- a/drivers/nvme/host/multipath.c +++ b/drivers/nvme/host/multipath.c @@ -884,7 +884,6 @@ void nvme_mpath_remove_disk(struct nvme_ns_head *head) { if (!head->disk) return; - blk_mark_disk_dead(head->disk); /* make sure all pending bios are cleaned up */ kblockd_schedule_work(&head->requeue_work); flush_work(&head->requeue_work); diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h index bf46f122e9e1e476c7127186c356c3857c37c30a..8657811f8b8875e21a53732252d7f0738c85d678 100644 --- a/drivers/nvme/host/nvme.h +++ b/drivers/nvme/host/nvme.h @@ -149,6 +149,11 @@ enum nvme_quirks { * Reports garbage in the namespace identifiers (eui64, nguid, uuid). */ NVME_QUIRK_BOGUS_NID = (1 << 18), + + /* + * No temperature thresholds for channels other than 0 (Composite). + */ + NVME_QUIRK_NO_SECONDARY_TEMP_THRESH = (1 << 19), }; /* @@ -323,6 +328,7 @@ struct nvme_ctrl { struct delayed_work ka_work; struct delayed_work failfast_work; struct nvme_command ka_cmd; + unsigned long ka_last_check_time; struct work_struct fw_act_work; unsigned long events; @@ -1072,7 +1078,7 @@ u32 nvme_command_effects(struct nvme_ctrl *ctrl, struct nvme_ns *ns, u8 opcode); u32 nvme_passthru_start(struct nvme_ctrl *ctrl, struct nvme_ns *ns, u8 opcode); int nvme_execute_rq(struct request *rq, bool at_head); -void nvme_passthru_end(struct nvme_ctrl *ctrl, u32 effects, +void nvme_passthru_end(struct nvme_ctrl *ctrl, struct nvme_ns *ns, u32 effects, struct nvme_command *cmd, int status); struct nvme_ctrl *nvme_ctrl_from_file(struct file *file); struct nvme_ns *nvme_find_get_ns(struct nvme_ctrl *ctrl, unsigned nsid); diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index 7f25c0fe3a0bf5baf92dcd5f81613ecac75bb593..492f319ebdf37ad838dacbb5237f1a804b243840 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -2956,7 +2956,7 @@ static struct nvme_dev *nvme_pci_alloc_dev(struct pci_dev *pdev, * over a single page. */ dev->ctrl.max_hw_sectors = min_t(u32, - NVME_MAX_KB_SZ << 1, dma_max_mapping_size(&pdev->dev) >> 9); + NVME_MAX_KB_SZ << 1, dma_opt_mapping_size(&pdev->dev) >> 9); dev->ctrl.max_segments = NVME_MAX_SEGS; /* @@ -3402,6 +3402,8 @@ static const struct pci_device_id nvme_id_table[] = { .driver_data = NVME_QUIRK_NO_DEEPEST_PS, }, { PCI_DEVICE(0x2646, 0x2263), /* KINGSTON A2000 NVMe SSD */ .driver_data = NVME_QUIRK_NO_DEEPEST_PS, }, + { PCI_DEVICE(0x2646, 0x5013), /* Kingston KC3000, Kingston FURY Renegade */ + .driver_data = NVME_QUIRK_NO_SECONDARY_TEMP_THRESH, }, { PCI_DEVICE(0x2646, 0x5018), /* KINGSTON OM8SFP4xxxxP OS21012 NVMe SSD */ .driver_data = NVME_QUIRK_DISABLE_WRITE_ZEROES, }, { PCI_DEVICE(0x2646, 0x5016), /* KINGSTON OM3PGP4xxxxP OS21011 NVMe SSD */ @@ -3422,6 +3424,8 @@ static const struct pci_device_id nvme_id_table[] = { .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x1e4B, 0x1202), /* MAXIO MAP1202 */ .driver_data = NVME_QUIRK_BOGUS_NID, }, + { PCI_DEVICE(0x1e4B, 0x1602), /* MAXIO MAP1602 */ + .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x1cc1, 0x5350), /* ADATA XPG GAMMIX S50 */ .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x1dbe, 0x5236), /* ADATA XPG GAMMIX S70 */ @@ -3441,6 +3445,10 @@ static const struct pci_device_id nvme_id_table[] = { NVME_QUIRK_IGNORE_DEV_SUBNQN, }, { PCI_DEVICE(0x10ec, 0x5763), /* TEAMGROUP T-FORCE CARDEA ZERO Z330 SSD */ .driver_data = NVME_QUIRK_BOGUS_NID, }, + { PCI_DEVICE(0x1e4b, 0x1602), /* HS-SSD-FUTURE 2048G */ + .driver_data = NVME_QUIRK_BOGUS_NID, }, + { PCI_DEVICE(0x10ec, 0x5765), /* TEAMGROUP MP33 2TB SSD */ + .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(PCI_VENDOR_ID_AMAZON, 0x0061), .driver_data = NVME_QUIRK_DMA_ADDRESS_BITS_48, }, { PCI_DEVICE(PCI_VENDOR_ID_AMAZON, 0x0065), diff --git a/drivers/nvme/target/passthru.c b/drivers/nvme/target/passthru.c index 511c980d538df3af57fa9f1d998607393c417008..71a9c1cc57f59c265478b8c816c8e4f5a9b65f61 100644 --- a/drivers/nvme/target/passthru.c +++ b/drivers/nvme/target/passthru.c @@ -243,7 +243,7 @@ static void nvmet_passthru_execute_cmd_work(struct work_struct *w) blk_mq_free_request(rq); if (effects) - nvme_passthru_end(ctrl, effects, req->cmd, status); + nvme_passthru_end(ctrl, ns, effects, req->cmd, status); } static enum rq_end_io_ret nvmet_passthru_req_done(struct request *rq, diff --git a/drivers/of/overlay.c b/drivers/of/overlay.c index 2e01960f1aeb3e3f7695849bbff6cc4ce67ad4f9..7feb643f13707d34b4012e365c6011f18cb37373 100644 --- a/drivers/of/overlay.c +++ b/drivers/of/overlay.c @@ -811,6 +811,7 @@ static int init_overlay_changeset(struct overlay_changeset *ovcs) if (!fragment->target) { pr_err("symbols in overlay, but not in live tree\n"); ret = -EINVAL; + of_node_put(node); goto err_out; } diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index f4e2a88729fd144bfa125a7378153b4cd26fd34c..c525867760bf819c0af9ccf601844b2c37d0ca49 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -6003,8 +6003,9 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x56c1, aspm_l1_acceptable_latency #ifdef CONFIG_PCIE_DPC /* - * Intel Tiger Lake and Alder Lake BIOS has a bug that clears the DPC - * RP PIO Log Size of the integrated Thunderbolt PCIe Root Ports. + * Intel Ice Lake, Tiger Lake and Alder Lake BIOS has a bug that clears + * the DPC RP PIO Log Size of the integrated Thunderbolt PCIe Root + * Ports. */ static void dpc_log_size(struct pci_dev *dev) { @@ -6027,6 +6028,10 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x461f, dpc_log_size); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x462f, dpc_log_size); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x463f, dpc_log_size); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x466e, dpc_log_size); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x8a1d, dpc_log_size); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x8a1f, dpc_log_size); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x8a21, dpc_log_size); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x8a23, dpc_log_size); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x9a23, dpc_log_size); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x9a25, dpc_log_size); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x9a27, dpc_log_size); diff --git a/drivers/phy/amlogic/phy-meson-g12a-mipi-dphy-analog.c b/drivers/phy/amlogic/phy-meson-g12a-mipi-dphy-analog.c index c14089fa7db495b869556e785a9ff1651419d0a4..cabdddbbabfd7dc251a7027b6e5745dea398d273 100644 --- a/drivers/phy/amlogic/phy-meson-g12a-mipi-dphy-analog.c +++ b/drivers/phy/amlogic/phy-meson-g12a-mipi-dphy-analog.c @@ -70,7 +70,7 @@ static int phy_g12a_mipi_dphy_analog_power_on(struct phy *phy) HHI_MIPI_CNTL1_BANDGAP); regmap_write(priv->regmap, HHI_MIPI_CNTL2, - FIELD_PREP(HHI_MIPI_CNTL2_DIF_TX_CTL0, 0x459) | + FIELD_PREP(HHI_MIPI_CNTL2_DIF_TX_CTL0, 0x45a) | FIELD_PREP(HHI_MIPI_CNTL2_DIF_TX_CTL1, 0x2680)); reg = DSI_LANE_CLK; diff --git a/drivers/phy/mediatek/phy-mtk-hdmi-mt8195.c b/drivers/phy/mediatek/phy-mtk-hdmi-mt8195.c index caa953780beebb85a07fe2e582ad70fbfec43f55..8aa7251de4a96e48c4a0148950e0961d3ba6d9a0 100644 --- a/drivers/phy/mediatek/phy-mtk-hdmi-mt8195.c +++ b/drivers/phy/mediatek/phy-mtk-hdmi-mt8195.c @@ -237,11 +237,11 @@ static int mtk_hdmi_pll_calc(struct mtk_hdmi_phy *hdmi_phy, struct clk_hw *hw, */ if (tmds_clk < 54 * MEGA) txposdiv = 8; - else if (tmds_clk >= 54 * MEGA && tmds_clk < 148.35 * MEGA) + else if (tmds_clk >= 54 * MEGA && (tmds_clk * 100) < 14835 * MEGA) txposdiv = 4; - else if (tmds_clk >= 148.35 * MEGA && tmds_clk < 296.7 * MEGA) + else if ((tmds_clk * 100) >= 14835 * MEGA && (tmds_clk * 10) < 2967 * MEGA) txposdiv = 2; - else if (tmds_clk >= 296.7 * MEGA && tmds_clk <= 594 * MEGA) + else if ((tmds_clk * 10) >= 2967 * MEGA && tmds_clk <= 594 * MEGA) txposdiv = 1; else return -EINVAL; @@ -324,12 +324,12 @@ static int mtk_hdmi_pll_drv_setting(struct clk_hw *hw) clk_channel_bias = 0x34; /* 20mA */ impedance_en = 0xf; impedance = 0x36; /* 100ohm */ - } else if (pixel_clk >= 74.175 * MEGA && pixel_clk <= 300 * MEGA) { + } else if (((u64)pixel_clk * 1000) >= 74175 * MEGA && pixel_clk <= 300 * MEGA) { data_channel_bias = 0x34; /* 20mA */ clk_channel_bias = 0x2c; /* 16mA */ impedance_en = 0xf; impedance = 0x36; /* 100ohm */ - } else if (pixel_clk >= 27 * MEGA && pixel_clk < 74.175 * MEGA) { + } else if (pixel_clk >= 27 * MEGA && ((u64)pixel_clk * 1000) < 74175 * MEGA) { data_channel_bias = 0x14; /* 10mA */ clk_channel_bias = 0x14; /* 10mA */ impedance_en = 0x0; diff --git a/drivers/phy/qualcomm/phy-qcom-qmp-combo.c b/drivers/phy/qualcomm/phy-qcom-qmp-combo.c index 6850e04c329b8e679814890d799e7f83e78176cc..87b17e5877ab8c6d9ed1ab082cda5ba974ab617b 100644 --- a/drivers/phy/qualcomm/phy-qcom-qmp-combo.c +++ b/drivers/phy/qualcomm/phy-qcom-qmp-combo.c @@ -2472,7 +2472,7 @@ static int qmp_combo_com_init(struct qmp_combo *qmp) ret = regulator_bulk_enable(cfg->num_vregs, qmp->vregs); if (ret) { dev_err(qmp->dev, "failed to enable regulators, err=%d\n", ret); - goto err_unlock; + goto err_decrement_count; } ret = reset_control_bulk_assert(cfg->num_resets, qmp->resets); @@ -2522,7 +2522,8 @@ static int qmp_combo_com_init(struct qmp_combo *qmp) reset_control_bulk_assert(cfg->num_resets, qmp->resets); err_disable_regulators: regulator_bulk_disable(cfg->num_vregs, qmp->vregs); -err_unlock: +err_decrement_count: + qmp->init_count--; mutex_unlock(&qmp->phy_mutex); return ret; diff --git a/drivers/phy/qualcomm/phy-qcom-qmp-pcie-msm8996.c b/drivers/phy/qualcomm/phy-qcom-qmp-pcie-msm8996.c index 09824be088c96c0b6354ff0488068b17e573a34b..0c603bc06e0998a95a0853072a10fffba87958ff 100644 --- a/drivers/phy/qualcomm/phy-qcom-qmp-pcie-msm8996.c +++ b/drivers/phy/qualcomm/phy-qcom-qmp-pcie-msm8996.c @@ -379,7 +379,7 @@ static int qmp_pcie_msm8996_com_init(struct qmp_phy *qphy) ret = regulator_bulk_enable(cfg->num_vregs, qmp->vregs); if (ret) { dev_err(qmp->dev, "failed to enable regulators, err=%d\n", ret); - goto err_unlock; + goto err_decrement_count; } ret = reset_control_bulk_assert(cfg->num_resets, qmp->resets); @@ -409,7 +409,8 @@ static int qmp_pcie_msm8996_com_init(struct qmp_phy *qphy) reset_control_bulk_assert(cfg->num_resets, qmp->resets); err_disable_regulators: regulator_bulk_disable(cfg->num_vregs, qmp->vregs); -err_unlock: +err_decrement_count: + qmp->init_count--; mutex_unlock(&qmp->phy_mutex); return ret; diff --git a/drivers/phy/qualcomm/phy-qcom-snps-femto-v2.c b/drivers/phy/qualcomm/phy-qcom-snps-femto-v2.c index a590635962140f817e0ee3287d576bbd043ffa3c..6c237f3cc66db8235dd3ae100ce11a487a8bf06c 100644 --- a/drivers/phy/qualcomm/phy-qcom-snps-femto-v2.c +++ b/drivers/phy/qualcomm/phy-qcom-snps-femto-v2.c @@ -115,11 +115,11 @@ struct phy_override_seq { * * @cfg_ahb_clk: AHB2PHY interface clock * @ref_clk: phy reference clock - * @iface_clk: phy interface clock * @phy_reset: phy reset control * @vregs: regulator supplies bulk data * @phy_initialized: if PHY has been initialized correctly * @mode: contains the current mode the PHY is in + * @update_seq_cfg: tuning parameters for phy init */ struct qcom_snps_hsphy { struct phy *phy; diff --git a/drivers/pinctrl/meson/pinctrl-meson-axg.c b/drivers/pinctrl/meson/pinctrl-meson-axg.c index 7bfecdfba17797620070ca9a8eeee2326837cd09..d249a035c2b9b37bea0f0e48fa5dbc75366cf8be 100644 --- a/drivers/pinctrl/meson/pinctrl-meson-axg.c +++ b/drivers/pinctrl/meson/pinctrl-meson-axg.c @@ -400,6 +400,7 @@ static struct meson_pmx_group meson_axg_periphs_groups[] = { GPIO_GROUP(GPIOA_15), GPIO_GROUP(GPIOA_16), GPIO_GROUP(GPIOA_17), + GPIO_GROUP(GPIOA_18), GPIO_GROUP(GPIOA_19), GPIO_GROUP(GPIOA_20), diff --git a/drivers/platform/mellanox/mlxbf-pmc.c b/drivers/platform/mellanox/mlxbf-pmc.c index c2c9b0d3244cbf2bc0cfad486169a0741d0d931e..be967d797c28e04c6ce726b434e0edb322737f88 100644 --- a/drivers/platform/mellanox/mlxbf-pmc.c +++ b/drivers/platform/mellanox/mlxbf-pmc.c @@ -1348,9 +1348,8 @@ static int mlxbf_pmc_map_counters(struct device *dev) for (i = 0; i < pmc->total_blocks; ++i) { if (strstr(pmc->block_name[i], "tile")) { - ret = sscanf(pmc->block_name[i], "tile%d", &tile_num); - if (ret < 0) - return ret; + if (sscanf(pmc->block_name[i], "tile%d", &tile_num) != 1) + return -EINVAL; if (tile_num >= pmc->tile_count) continue; diff --git a/drivers/platform/surface/aggregator/controller.c b/drivers/platform/surface/aggregator/controller.c index 535581c0471c509cc6836aedc66b32a345b0561d..7fc602e01487d42542338f4c767cac28b34fcd31 100644 --- a/drivers/platform/surface/aggregator/controller.c +++ b/drivers/platform/surface/aggregator/controller.c @@ -825,7 +825,7 @@ static int ssam_cplt_init(struct ssam_cplt *cplt, struct device *dev) cplt->dev = dev; - cplt->wq = create_workqueue(SSAM_CPLT_WQ_NAME); + cplt->wq = alloc_workqueue(SSAM_CPLT_WQ_NAME, WQ_UNBOUND | WQ_MEM_RECLAIM, 0); if (!cplt->wq) return -ENOMEM; diff --git a/drivers/platform/surface/surface_aggregator_tabletsw.c b/drivers/platform/surface/surface_aggregator_tabletsw.c index 8f52b62d1c19567b041a9558d92337fae2631474..c0a1a5869246edde0e5e65f1e1b75a019fe1ced0 100644 --- a/drivers/platform/surface/surface_aggregator_tabletsw.c +++ b/drivers/platform/surface/surface_aggregator_tabletsw.c @@ -210,6 +210,7 @@ enum ssam_kip_cover_state { SSAM_KIP_COVER_STATE_LAPTOP = 0x03, SSAM_KIP_COVER_STATE_FOLDED_CANVAS = 0x04, SSAM_KIP_COVER_STATE_FOLDED_BACK = 0x05, + SSAM_KIP_COVER_STATE_BOOK = 0x06, }; static const char *ssam_kip_cover_state_name(struct ssam_tablet_sw *sw, @@ -231,6 +232,9 @@ static const char *ssam_kip_cover_state_name(struct ssam_tablet_sw *sw, case SSAM_KIP_COVER_STATE_FOLDED_BACK: return "folded-back"; + case SSAM_KIP_COVER_STATE_BOOK: + return "book"; + default: dev_warn(&sw->sdev->dev, "unknown KIP cover state: %u\n", state->state); return ""; @@ -244,6 +248,7 @@ static bool ssam_kip_cover_state_is_tablet_mode(struct ssam_tablet_sw *sw, case SSAM_KIP_COVER_STATE_DISCONNECTED: case SSAM_KIP_COVER_STATE_FOLDED_CANVAS: case SSAM_KIP_COVER_STATE_FOLDED_BACK: + case SSAM_KIP_COVER_STATE_BOOK: return true; case SSAM_KIP_COVER_STATE_CLOSED: @@ -335,6 +340,7 @@ enum ssam_pos_state_cover { SSAM_POS_COVER_LAPTOP = 0x03, SSAM_POS_COVER_FOLDED_CANVAS = 0x04, SSAM_POS_COVER_FOLDED_BACK = 0x05, + SSAM_POS_COVER_BOOK = 0x06, }; enum ssam_pos_state_sls { @@ -367,6 +373,9 @@ static const char *ssam_pos_state_name_cover(struct ssam_tablet_sw *sw, u32 stat case SSAM_POS_COVER_FOLDED_BACK: return "folded-back"; + case SSAM_POS_COVER_BOOK: + return "book"; + default: dev_warn(&sw->sdev->dev, "unknown device posture for type-cover: %u\n", state); return ""; @@ -416,6 +425,7 @@ static bool ssam_pos_state_is_tablet_mode_cover(struct ssam_tablet_sw *sw, u32 s case SSAM_POS_COVER_DISCONNECTED: case SSAM_POS_COVER_FOLDED_CANVAS: case SSAM_POS_COVER_FOLDED_BACK: + case SSAM_POS_COVER_BOOK: return true; case SSAM_POS_COVER_CLOSED: diff --git a/drivers/platform/x86/amd/pmf/core.c b/drivers/platform/x86/amd/pmf/core.c index d5bb775dadcf20328294da3cea153075c6456138..ee5f124f78b6d712a58e944fe114f817e6c76ae3 100644 --- a/drivers/platform/x86/amd/pmf/core.c +++ b/drivers/platform/x86/amd/pmf/core.c @@ -245,24 +245,29 @@ static const struct pci_device_id pmf_pci_ids[] = { { } }; -int amd_pmf_init_metrics_table(struct amd_pmf_dev *dev) +static void amd_pmf_set_dram_addr(struct amd_pmf_dev *dev) { u64 phys_addr; u32 hi, low; - INIT_DELAYED_WORK(&dev->work_buffer, amd_pmf_get_metrics); + phys_addr = virt_to_phys(dev->buf); + hi = phys_addr >> 32; + low = phys_addr & GENMASK(31, 0); + + amd_pmf_send_cmd(dev, SET_DRAM_ADDR_HIGH, 0, hi, NULL); + amd_pmf_send_cmd(dev, SET_DRAM_ADDR_LOW, 0, low, NULL); +} +int amd_pmf_init_metrics_table(struct amd_pmf_dev *dev) +{ /* Get Metrics Table Address */ dev->buf = kzalloc(sizeof(dev->m_table), GFP_KERNEL); if (!dev->buf) return -ENOMEM; - phys_addr = virt_to_phys(dev->buf); - hi = phys_addr >> 32; - low = phys_addr & GENMASK(31, 0); + INIT_DELAYED_WORK(&dev->work_buffer, amd_pmf_get_metrics); - amd_pmf_send_cmd(dev, SET_DRAM_ADDR_HIGH, 0, hi, NULL); - amd_pmf_send_cmd(dev, SET_DRAM_ADDR_LOW, 0, low, NULL); + amd_pmf_set_dram_addr(dev); /* * Start collecting the metrics data after a small delay @@ -273,6 +278,18 @@ int amd_pmf_init_metrics_table(struct amd_pmf_dev *dev) return 0; } +static int amd_pmf_resume_handler(struct device *dev) +{ + struct amd_pmf_dev *pdev = dev_get_drvdata(dev); + + if (pdev->buf) + amd_pmf_set_dram_addr(pdev); + + return 0; +} + +static DEFINE_SIMPLE_DEV_PM_OPS(amd_pmf_pm, NULL, amd_pmf_resume_handler); + static void amd_pmf_init_features(struct amd_pmf_dev *dev) { int ret; @@ -413,6 +430,7 @@ static struct platform_driver amd_pmf_driver = { .name = "amd-pmf", .acpi_match_table = amd_pmf_acpi_ids, .dev_groups = amd_pmf_driver_groups, + .pm = pm_sleep_ptr(&amd_pmf_pm), }, .probe = amd_pmf_probe, .remove_new = amd_pmf_remove, diff --git a/drivers/platform/x86/asus-nb-wmi.c b/drivers/platform/x86/asus-nb-wmi.c index e2c9a68d12df9cc00ce3deda7964fb12bfd23a0f..fdf7da06af3067495fe7132e725a52542010ff8d 100644 --- a/drivers/platform/x86/asus-nb-wmi.c +++ b/drivers/platform/x86/asus-nb-wmi.c @@ -555,6 +555,7 @@ static const struct key_entry asus_nb_wmi_keymap[] = { { KE_KEY, 0x71, { KEY_F13 } }, /* General-purpose button */ { KE_IGNORE, 0x79, }, /* Charger type dectection notification */ { KE_KEY, 0x7a, { KEY_ALS_TOGGLE } }, /* Ambient Light Sensor Toggle */ + { KE_IGNORE, 0x7B, }, /* Charger connect/disconnect notification */ { KE_KEY, 0x7c, { KEY_MICMUTE } }, { KE_KEY, 0x7D, { KEY_BLUETOOTH } }, /* Bluetooth Enable */ { KE_KEY, 0x7E, { KEY_BLUETOOTH } }, /* Bluetooth Disable */ @@ -584,6 +585,7 @@ static const struct key_entry asus_nb_wmi_keymap[] = { { KE_KEY, 0xAE, { KEY_FN_F5 } }, /* Fn+F5 fan mode on 2020+ */ { KE_KEY, 0xB3, { KEY_PROG4 } }, /* AURA */ { KE_KEY, 0xB5, { KEY_CALC } }, + { KE_IGNORE, 0xC0, }, /* External display connect/disconnect notification */ { KE_KEY, 0xC4, { KEY_KBDILLUMUP } }, { KE_KEY, 0xC5, { KEY_KBDILLUMDOWN } }, { KE_IGNORE, 0xC6, }, /* Ambient Light Sensor notification */ diff --git a/drivers/platform/x86/intel/ifs/load.c b/drivers/platform/x86/intel/ifs/load.c index 61dffb4c8a1dcd07c42623b78b90d7eda8dba884..e6ae8265f3a37eab17d9698a78363fbadf562265 100644 --- a/drivers/platform/x86/intel/ifs/load.c +++ b/drivers/platform/x86/intel/ifs/load.c @@ -208,7 +208,7 @@ static int scan_chunks_sanity_check(struct device *dev) continue; reinit_completion(&ifs_done); local_work.dev = dev; - INIT_WORK(&local_work.w, copy_hashes_authenticate_chunks); + INIT_WORK_ONSTACK(&local_work.w, copy_hashes_authenticate_chunks); schedule_work_on(cpu, &local_work.w); wait_for_completion(&ifs_done); if (ifsd->loading_error) { diff --git a/drivers/platform/x86/intel/int3472/clk_and_regulator.c b/drivers/platform/x86/intel/int3472/clk_and_regulator.c index 1086c3d8349454f98ade335daf046bafc7d5116f..399f0623ca1b5ca817db2a30e5e4b425ef37cbed 100644 --- a/drivers/platform/x86/intel/int3472/clk_and_regulator.c +++ b/drivers/platform/x86/intel/int3472/clk_and_regulator.c @@ -101,9 +101,11 @@ int skl_int3472_register_clock(struct int3472_discrete_device *int3472, int3472->clock.ena_gpio = acpi_get_and_request_gpiod(path, agpio->pin_table[0], "int3472,clk-enable"); - if (IS_ERR(int3472->clock.ena_gpio)) - return dev_err_probe(int3472->dev, PTR_ERR(int3472->clock.ena_gpio), - "getting clk-enable GPIO\n"); + if (IS_ERR(int3472->clock.ena_gpio)) { + ret = PTR_ERR(int3472->clock.ena_gpio); + int3472->clock.ena_gpio = NULL; + return dev_err_probe(int3472->dev, ret, "getting clk-enable GPIO\n"); + } if (polarity == GPIO_ACTIVE_LOW) gpiod_toggle_active_low(int3472->clock.ena_gpio); @@ -199,8 +201,9 @@ int skl_int3472_register_regulator(struct int3472_discrete_device *int3472, int3472->regulator.gpio = acpi_get_and_request_gpiod(path, agpio->pin_table[0], "int3472,regulator"); if (IS_ERR(int3472->regulator.gpio)) { - dev_err(int3472->dev, "Failed to get regulator GPIO line\n"); - return PTR_ERR(int3472->regulator.gpio); + ret = PTR_ERR(int3472->regulator.gpio); + int3472->regulator.gpio = NULL; + return dev_err_probe(int3472->dev, ret, "getting regulator GPIO\n"); } /* Ensure the pin is in output mode and non-active state */ diff --git a/drivers/platform/x86/intel/speed_select_if/isst_if_common.c b/drivers/platform/x86/intel/speed_select_if/isst_if_common.c index e0572a29212e850c28313bcca2822ca0097ba3c1..02fe360a59c7cf427c7293f5a0f3d89c78fe92ae 100644 --- a/drivers/platform/x86/intel/speed_select_if/isst_if_common.c +++ b/drivers/platform/x86/intel/speed_select_if/isst_if_common.c @@ -304,14 +304,13 @@ struct isst_if_pkg_info { static struct isst_if_cpu_info *isst_cpu_info; static struct isst_if_pkg_info *isst_pkg_info; -#define ISST_MAX_PCI_DOMAINS 8 - static struct pci_dev *_isst_if_get_pci_dev(int cpu, int bus_no, int dev, int fn) { struct pci_dev *matched_pci_dev = NULL; struct pci_dev *pci_dev = NULL; + struct pci_dev *_pci_dev = NULL; int no_matches = 0, pkg_id; - int i, bus_number; + int bus_number; if (bus_no < 0 || bus_no >= ISST_MAX_BUS_NUMBER || cpu < 0 || cpu >= nr_cpu_ids || cpu >= num_possible_cpus()) @@ -323,12 +322,11 @@ static struct pci_dev *_isst_if_get_pci_dev(int cpu, int bus_no, int dev, int fn if (bus_number < 0) return NULL; - for (i = 0; i < ISST_MAX_PCI_DOMAINS; ++i) { - struct pci_dev *_pci_dev; + for_each_pci_dev(_pci_dev) { int node; - _pci_dev = pci_get_domain_bus_and_slot(i, bus_number, PCI_DEVFN(dev, fn)); - if (!_pci_dev) + if (_pci_dev->bus->number != bus_number || + _pci_dev->devfn != PCI_DEVFN(dev, fn)) continue; ++no_matches; diff --git a/drivers/power/supply/ab8500_btemp.c b/drivers/power/supply/ab8500_btemp.c index 307ee6f71042e953f26a8f9c552aef63e10a6a13..6f83e99d2eb72e34b5666a76ecff1cf7ffd367ab 100644 --- a/drivers/power/supply/ab8500_btemp.c +++ b/drivers/power/supply/ab8500_btemp.c @@ -624,10 +624,8 @@ static int ab8500_btemp_get_ext_psy_data(struct device *dev, void *data) */ static void ab8500_btemp_external_power_changed(struct power_supply *psy) { - struct ab8500_btemp *di = power_supply_get_drvdata(psy); - - class_for_each_device(power_supply_class, NULL, - di->btemp_psy, ab8500_btemp_get_ext_psy_data); + class_for_each_device(power_supply_class, NULL, psy, + ab8500_btemp_get_ext_psy_data); } /* ab8500 btemp driver interrupts and their respective isr */ diff --git a/drivers/power/supply/ab8500_fg.c b/drivers/power/supply/ab8500_fg.c index 41a7bff9ac3765639a463b24aa94da8ef105dc51..53560fbb6dcd34bb08db8543ce9eb7e338afdb40 100644 --- a/drivers/power/supply/ab8500_fg.c +++ b/drivers/power/supply/ab8500_fg.c @@ -2407,10 +2407,8 @@ static int ab8500_fg_init_hw_registers(struct ab8500_fg *di) */ static void ab8500_fg_external_power_changed(struct power_supply *psy) { - struct ab8500_fg *di = power_supply_get_drvdata(psy); - - class_for_each_device(power_supply_class, NULL, - di->fg_psy, ab8500_fg_get_ext_psy_data); + class_for_each_device(power_supply_class, NULL, psy, + ab8500_fg_get_ext_psy_data); } /** diff --git a/drivers/power/supply/axp288_fuel_gauge.c b/drivers/power/supply/axp288_fuel_gauge.c index 05f41317846296b067f41d050abc66d26cf09ff0..3be6f3b10ea4286e218f876798a0c7f0b99e4aa0 100644 --- a/drivers/power/supply/axp288_fuel_gauge.c +++ b/drivers/power/supply/axp288_fuel_gauge.c @@ -507,7 +507,7 @@ static void fuel_gauge_external_power_changed(struct power_supply *psy) mutex_lock(&info->lock); info->valid = 0; /* Force updating of the cached registers */ mutex_unlock(&info->lock); - power_supply_changed(info->bat); + power_supply_changed(psy); } static struct power_supply_desc fuel_gauge_desc = { diff --git a/drivers/power/supply/bq24190_charger.c b/drivers/power/supply/bq24190_charger.c index de67b985f0a913a355ea3c27e3897d7373b78db2..dc33f00fcc0682a3c69ead86186df4219a1cd879 100644 --- a/drivers/power/supply/bq24190_charger.c +++ b/drivers/power/supply/bq24190_charger.c @@ -1262,6 +1262,7 @@ static void bq24190_input_current_limit_work(struct work_struct *work) bq24190_charger_set_property(bdi->charger, POWER_SUPPLY_PROP_INPUT_CURRENT_LIMIT, &val); + power_supply_changed(bdi->charger); } /* Sync the input-current-limit with our parent supply (if we have one) */ diff --git a/drivers/power/supply/bq25890_charger.c b/drivers/power/supply/bq25890_charger.c index 22cde35eb144c612edc491307c06f5dddad9e3ab..f8636cf86505a7749eb7835eb422eda819137cf6 100644 --- a/drivers/power/supply/bq25890_charger.c +++ b/drivers/power/supply/bq25890_charger.c @@ -750,7 +750,7 @@ static void bq25890_charger_external_power_changed(struct power_supply *psy) if (bq->chip_version != BQ25892) return; - ret = power_supply_get_property_from_supplier(bq->charger, + ret = power_supply_get_property_from_supplier(psy, POWER_SUPPLY_PROP_USB_TYPE, &val); if (ret) @@ -775,6 +775,7 @@ static void bq25890_charger_external_power_changed(struct power_supply *psy) } bq25890_field_write(bq, F_IINLIM, input_current_limit); + power_supply_changed(psy); } static int bq25890_get_chip_state(struct bq25890_device *bq, @@ -1106,6 +1107,8 @@ static void bq25890_pump_express_work(struct work_struct *data) dev_info(bq->dev, "Hi-voltage charging requested, input voltage is %d mV\n", voltage); + power_supply_changed(bq->charger); + return; error_print: bq25890_field_write(bq, F_PUMPX_EN, 0); diff --git a/drivers/power/supply/bq27xxx_battery.c b/drivers/power/supply/bq27xxx_battery.c index 5ff6f44fd47b22e709813fbfdb3eedd7a4320aa5..4296600e8912a3988c45286f58420ffabb547345 100644 --- a/drivers/power/supply/bq27xxx_battery.c +++ b/drivers/power/supply/bq27xxx_battery.c @@ -1083,10 +1083,8 @@ static int poll_interval_param_set(const char *val, const struct kernel_param *k return ret; mutex_lock(&bq27xxx_list_lock); - list_for_each_entry(di, &bq27xxx_battery_devices, list) { - cancel_delayed_work_sync(&di->work); - schedule_delayed_work(&di->work, 0); - } + list_for_each_entry(di, &bq27xxx_battery_devices, list) + mod_delayed_work(system_wq, &di->work, 0); mutex_unlock(&bq27xxx_list_lock); return ret; @@ -1761,60 +1759,6 @@ static int bq27xxx_battery_read_health(struct bq27xxx_device_info *di) return POWER_SUPPLY_HEALTH_GOOD; } -void bq27xxx_battery_update(struct bq27xxx_device_info *di) -{ - struct bq27xxx_reg_cache cache = {0, }; - bool has_singe_flag = di->opts & BQ27XXX_O_ZERO; - - cache.flags = bq27xxx_read(di, BQ27XXX_REG_FLAGS, has_singe_flag); - if ((cache.flags & 0xff) == 0xff) - cache.flags = -1; /* read error */ - if (cache.flags >= 0) { - cache.temperature = bq27xxx_battery_read_temperature(di); - if (di->regs[BQ27XXX_REG_TTE] != INVALID_REG_ADDR) - cache.time_to_empty = bq27xxx_battery_read_time(di, BQ27XXX_REG_TTE); - if (di->regs[BQ27XXX_REG_TTECP] != INVALID_REG_ADDR) - cache.time_to_empty_avg = bq27xxx_battery_read_time(di, BQ27XXX_REG_TTECP); - if (di->regs[BQ27XXX_REG_TTF] != INVALID_REG_ADDR) - cache.time_to_full = bq27xxx_battery_read_time(di, BQ27XXX_REG_TTF); - - cache.charge_full = bq27xxx_battery_read_fcc(di); - cache.capacity = bq27xxx_battery_read_soc(di); - if (di->regs[BQ27XXX_REG_AE] != INVALID_REG_ADDR) - cache.energy = bq27xxx_battery_read_energy(di); - di->cache.flags = cache.flags; - cache.health = bq27xxx_battery_read_health(di); - if (di->regs[BQ27XXX_REG_CYCT] != INVALID_REG_ADDR) - cache.cycle_count = bq27xxx_battery_read_cyct(di); - - /* We only have to read charge design full once */ - if (di->charge_design_full <= 0) - di->charge_design_full = bq27xxx_battery_read_dcap(di); - } - - if ((di->cache.capacity != cache.capacity) || - (di->cache.flags != cache.flags)) - power_supply_changed(di->bat); - - if (memcmp(&di->cache, &cache, sizeof(cache)) != 0) - di->cache = cache; - - di->last_update = jiffies; -} -EXPORT_SYMBOL_GPL(bq27xxx_battery_update); - -static void bq27xxx_battery_poll(struct work_struct *work) -{ - struct bq27xxx_device_info *di = - container_of(work, struct bq27xxx_device_info, - work.work); - - bq27xxx_battery_update(di); - - if (poll_interval > 0) - schedule_delayed_work(&di->work, poll_interval * HZ); -} - static bool bq27xxx_battery_is_full(struct bq27xxx_device_info *di, int flags) { if (di->opts & BQ27XXX_O_ZERO) @@ -1833,7 +1777,8 @@ static bool bq27xxx_battery_is_full(struct bq27xxx_device_info *di, int flags) static int bq27xxx_battery_current_and_status( struct bq27xxx_device_info *di, union power_supply_propval *val_curr, - union power_supply_propval *val_status) + union power_supply_propval *val_status, + struct bq27xxx_reg_cache *cache) { bool single_flags = (di->opts & BQ27XXX_O_ZERO); int curr; @@ -1845,10 +1790,14 @@ static int bq27xxx_battery_current_and_status( return curr; } - flags = bq27xxx_read(di, BQ27XXX_REG_FLAGS, single_flags); - if (flags < 0) { - dev_err(di->dev, "error reading flags\n"); - return flags; + if (cache) { + flags = cache->flags; + } else { + flags = bq27xxx_read(di, BQ27XXX_REG_FLAGS, single_flags); + if (flags < 0) { + dev_err(di->dev, "error reading flags\n"); + return flags; + } } if (di->opts & BQ27XXX_O_ZERO) { @@ -1883,6 +1832,78 @@ static int bq27xxx_battery_current_and_status( return 0; } +static void bq27xxx_battery_update_unlocked(struct bq27xxx_device_info *di) +{ + union power_supply_propval status = di->last_status; + struct bq27xxx_reg_cache cache = {0, }; + bool has_singe_flag = di->opts & BQ27XXX_O_ZERO; + + cache.flags = bq27xxx_read(di, BQ27XXX_REG_FLAGS, has_singe_flag); + if ((cache.flags & 0xff) == 0xff) + cache.flags = -1; /* read error */ + if (cache.flags >= 0) { + cache.temperature = bq27xxx_battery_read_temperature(di); + if (di->regs[BQ27XXX_REG_TTE] != INVALID_REG_ADDR) + cache.time_to_empty = bq27xxx_battery_read_time(di, BQ27XXX_REG_TTE); + if (di->regs[BQ27XXX_REG_TTECP] != INVALID_REG_ADDR) + cache.time_to_empty_avg = bq27xxx_battery_read_time(di, BQ27XXX_REG_TTECP); + if (di->regs[BQ27XXX_REG_TTF] != INVALID_REG_ADDR) + cache.time_to_full = bq27xxx_battery_read_time(di, BQ27XXX_REG_TTF); + + cache.charge_full = bq27xxx_battery_read_fcc(di); + cache.capacity = bq27xxx_battery_read_soc(di); + if (di->regs[BQ27XXX_REG_AE] != INVALID_REG_ADDR) + cache.energy = bq27xxx_battery_read_energy(di); + di->cache.flags = cache.flags; + cache.health = bq27xxx_battery_read_health(di); + if (di->regs[BQ27XXX_REG_CYCT] != INVALID_REG_ADDR) + cache.cycle_count = bq27xxx_battery_read_cyct(di); + + /* + * On gauges with signed current reporting the current must be + * checked to detect charging <-> discharging status changes. + */ + if (!(di->opts & BQ27XXX_O_ZERO)) + bq27xxx_battery_current_and_status(di, NULL, &status, &cache); + + /* We only have to read charge design full once */ + if (di->charge_design_full <= 0) + di->charge_design_full = bq27xxx_battery_read_dcap(di); + } + + if ((di->cache.capacity != cache.capacity) || + (di->cache.flags != cache.flags) || + (di->last_status.intval != status.intval)) { + di->last_status.intval = status.intval; + power_supply_changed(di->bat); + } + + if (memcmp(&di->cache, &cache, sizeof(cache)) != 0) + di->cache = cache; + + di->last_update = jiffies; + + if (!di->removed && poll_interval > 0) + mod_delayed_work(system_wq, &di->work, poll_interval * HZ); +} + +void bq27xxx_battery_update(struct bq27xxx_device_info *di) +{ + mutex_lock(&di->lock); + bq27xxx_battery_update_unlocked(di); + mutex_unlock(&di->lock); +} +EXPORT_SYMBOL_GPL(bq27xxx_battery_update); + +static void bq27xxx_battery_poll(struct work_struct *work) +{ + struct bq27xxx_device_info *di = + container_of(work, struct bq27xxx_device_info, + work.work); + + bq27xxx_battery_update(di); +} + /* * Get the average power in µW * Return < 0 if something fails. @@ -1985,10 +2006,8 @@ static int bq27xxx_battery_get_property(struct power_supply *psy, struct bq27xxx_device_info *di = power_supply_get_drvdata(psy); mutex_lock(&di->lock); - if (time_is_before_jiffies(di->last_update + 5 * HZ)) { - cancel_delayed_work_sync(&di->work); - bq27xxx_battery_poll(&di->work.work); - } + if (time_is_before_jiffies(di->last_update + 5 * HZ)) + bq27xxx_battery_update_unlocked(di); mutex_unlock(&di->lock); if (psp != POWER_SUPPLY_PROP_PRESENT && di->cache.flags < 0) @@ -1996,7 +2015,7 @@ static int bq27xxx_battery_get_property(struct power_supply *psy, switch (psp) { case POWER_SUPPLY_PROP_STATUS: - ret = bq27xxx_battery_current_and_status(di, NULL, val); + ret = bq27xxx_battery_current_and_status(di, NULL, val, NULL); break; case POWER_SUPPLY_PROP_VOLTAGE_NOW: ret = bq27xxx_battery_voltage(di, val); @@ -2005,7 +2024,7 @@ static int bq27xxx_battery_get_property(struct power_supply *psy, val->intval = di->cache.flags < 0 ? 0 : 1; break; case POWER_SUPPLY_PROP_CURRENT_NOW: - ret = bq27xxx_battery_current_and_status(di, val, NULL); + ret = bq27xxx_battery_current_and_status(di, val, NULL, NULL); break; case POWER_SUPPLY_PROP_CAPACITY: ret = bq27xxx_simple_value(di->cache.capacity, val); @@ -2078,8 +2097,8 @@ static void bq27xxx_external_power_changed(struct power_supply *psy) { struct bq27xxx_device_info *di = power_supply_get_drvdata(psy); - cancel_delayed_work_sync(&di->work); - schedule_delayed_work(&di->work, 0); + /* After charger plug in/out wait 0.5s for things to stabilize */ + mod_delayed_work(system_wq, &di->work, HZ / 2); } int bq27xxx_battery_setup(struct bq27xxx_device_info *di) @@ -2127,22 +2146,18 @@ EXPORT_SYMBOL_GPL(bq27xxx_battery_setup); void bq27xxx_battery_teardown(struct bq27xxx_device_info *di) { - /* - * power_supply_unregister call bq27xxx_battery_get_property which - * call bq27xxx_battery_poll. - * Make sure that bq27xxx_battery_poll will not call - * schedule_delayed_work again after unregister (which cause OOPS). - */ - poll_interval = 0; - - cancel_delayed_work_sync(&di->work); - - power_supply_unregister(di->bat); - mutex_lock(&bq27xxx_list_lock); list_del(&di->list); mutex_unlock(&bq27xxx_list_lock); + /* Set removed to avoid bq27xxx_battery_update() re-queuing the work */ + mutex_lock(&di->lock); + di->removed = true; + mutex_unlock(&di->lock); + + cancel_delayed_work_sync(&di->work); + + power_supply_unregister(di->bat); mutex_destroy(&di->lock); } EXPORT_SYMBOL_GPL(bq27xxx_battery_teardown); diff --git a/drivers/power/supply/bq27xxx_battery_i2c.c b/drivers/power/supply/bq27xxx_battery_i2c.c index f8768997333bc59412d040cee7427efad086a7f2..6d3c748763390675f759388cc1ceef3d73afee02 100644 --- a/drivers/power/supply/bq27xxx_battery_i2c.c +++ b/drivers/power/supply/bq27xxx_battery_i2c.c @@ -179,7 +179,7 @@ static int bq27xxx_battery_i2c_probe(struct i2c_client *client) i2c_set_clientdata(client, di); if (client->irq) { - ret = devm_request_threaded_irq(&client->dev, client->irq, + ret = request_threaded_irq(client->irq, NULL, bq27xxx_battery_irq_handler_thread, IRQF_ONESHOT, di->name, di); @@ -209,6 +209,7 @@ static void bq27xxx_battery_i2c_remove(struct i2c_client *client) { struct bq27xxx_device_info *di = i2c_get_clientdata(client); + free_irq(client->irq, di); bq27xxx_battery_teardown(di); mutex_lock(&battery_mutex); diff --git a/drivers/power/supply/mt6360_charger.c b/drivers/power/supply/mt6360_charger.c index 92e48e3a485364dd154e8489398488494215874c..1305cba61edd4b957313d7d86fb05158975b3086 100644 --- a/drivers/power/supply/mt6360_charger.c +++ b/drivers/power/supply/mt6360_charger.c @@ -796,7 +796,9 @@ static int mt6360_charger_probe(struct platform_device *pdev) mci->vinovp = 6500000; mutex_init(&mci->chgdet_lock); platform_set_drvdata(pdev, mci); - devm_work_autocancel(&pdev->dev, &mci->chrdet_work, mt6360_chrdet_work); + ret = devm_work_autocancel(&pdev->dev, &mci->chrdet_work, mt6360_chrdet_work); + if (ret) + return dev_err_probe(&pdev->dev, ret, "Failed to set delayed work\n"); ret = device_property_read_u32(&pdev->dev, "richtek,vinovp-microvolt", &mci->vinovp); if (ret) diff --git a/drivers/power/supply/power_supply_core.c b/drivers/power/supply/power_supply_core.c index ab986dbace16a309e8598d802095efab4ae4d3a9..3791aec69ddc675404ccc5974710a01ca49be9ad 100644 --- a/drivers/power/supply/power_supply_core.c +++ b/drivers/power/supply/power_supply_core.c @@ -348,6 +348,10 @@ static int __power_supply_is_system_supplied(struct device *dev, void *data) struct power_supply *psy = dev_get_drvdata(dev); unsigned int *count = data; + if (!psy->desc->get_property(psy, POWER_SUPPLY_PROP_SCOPE, &ret)) + if (ret.intval == POWER_SUPPLY_SCOPE_DEVICE) + return 0; + (*count)++; if (psy->desc->type != POWER_SUPPLY_TYPE_BATTERY) if (!psy->desc->get_property(psy, POWER_SUPPLY_PROP_ONLINE, @@ -366,8 +370,8 @@ int power_supply_is_system_supplied(void) __power_supply_is_system_supplied); /* - * If no power class device was found at all, most probably we are - * running on a desktop system, so assume we are on mains power. + * If no system scope power class device was found at all, most probably we + * are running on a desktop system, so assume we are on mains power. */ if (count == 0) return 1; @@ -573,7 +577,7 @@ int power_supply_get_battery_info(struct power_supply *psy, struct power_supply_battery_info *info; struct device_node *battery_np = NULL; struct fwnode_reference_args args; - struct fwnode_handle *fwnode; + struct fwnode_handle *fwnode = NULL; const char *value; int err, len, index; const __be32 *list; @@ -585,7 +589,7 @@ int power_supply_get_battery_info(struct power_supply *psy, return -ENODEV; fwnode = fwnode_handle_get(of_fwnode_handle(battery_np)); - } else { + } else if (psy->dev.parent) { err = fwnode_property_get_reference_args( dev_fwnode(psy->dev.parent), "monitored-battery", NULL, 0, 0, &args); @@ -595,6 +599,9 @@ int power_supply_get_battery_info(struct power_supply *psy, fwnode = args.fwnode; } + if (!fwnode) + return -ENOENT; + err = fwnode_property_read_string(fwnode, "compatible", &value); if (err) goto out_put_node; diff --git a/drivers/power/supply/power_supply_leds.c b/drivers/power/supply/power_supply_leds.c index 702bf83f6e6d256e7012c9657b750ee3f51d2ca4..0674483279d77c2b7bd1e8e687839332074d1ca8 100644 --- a/drivers/power/supply/power_supply_leds.c +++ b/drivers/power/supply/power_supply_leds.c @@ -35,8 +35,9 @@ static void power_supply_update_bat_leds(struct power_supply *psy) led_trigger_event(psy->charging_full_trig, LED_FULL); led_trigger_event(psy->charging_trig, LED_OFF); led_trigger_event(psy->full_trig, LED_FULL); - led_trigger_event(psy->charging_blink_full_solid_trig, - LED_FULL); + /* Going from blink to LED on requires a LED_OFF event to stop blink */ + led_trigger_event(psy->charging_blink_full_solid_trig, LED_OFF); + led_trigger_event(psy->charging_blink_full_solid_trig, LED_FULL); break; case POWER_SUPPLY_STATUS_CHARGING: led_trigger_event(psy->charging_full_trig, LED_FULL); diff --git a/drivers/power/supply/power_supply_sysfs.c b/drivers/power/supply/power_supply_sysfs.c index ba3b125cd66e432631d6170b048bc089870aedd7..06e5b6b0e255ce8bf159c2f5f0e012916fe12551 100644 --- a/drivers/power/supply/power_supply_sysfs.c +++ b/drivers/power/supply/power_supply_sysfs.c @@ -286,7 +286,8 @@ static ssize_t power_supply_show_property(struct device *dev, if (ret < 0) { if (ret == -ENODATA) - dev_dbg(dev, "driver has no data for `%s' property\n", + dev_dbg_ratelimited(dev, + "driver has no data for `%s' property\n", attr->attr.name); else if (ret != -ENODEV && ret != -EAGAIN) dev_err_ratelimited(dev, diff --git a/drivers/power/supply/rt9467-charger.c b/drivers/power/supply/rt9467-charger.c index 73f744a3155d4ce55011d26310173b6eb69803cb..ea33693b69779ac0928d112ec8ec5d45dc5a90e3 100644 --- a/drivers/power/supply/rt9467-charger.c +++ b/drivers/power/supply/rt9467-charger.c @@ -1023,7 +1023,7 @@ static int rt9467_request_interrupt(struct rt9467_chg_data *data) for (i = 0; i < num_chg_irqs; i++) { virq = regmap_irq_get_virq(data->irq_chip_data, chg_irqs[i].hwirq); if (virq <= 0) - return dev_err_probe(dev, virq, "Failed to get (%s) irq\n", + return dev_err_probe(dev, -EINVAL, "Failed to get (%s) irq\n", chg_irqs[i].name); ret = devm_request_threaded_irq(dev, virq, NULL, chg_irqs[i].handler, diff --git a/drivers/power/supply/sbs-charger.c b/drivers/power/supply/sbs-charger.c index 75ebcbf0a7883750b009ef21a6bd10692f138b1a..a14e89ac0369cae0a0990775eb41fec84d537c4b 100644 --- a/drivers/power/supply/sbs-charger.c +++ b/drivers/power/supply/sbs-charger.c @@ -24,7 +24,7 @@ #define SBS_CHARGER_REG_STATUS 0x13 #define SBS_CHARGER_REG_ALARM_WARNING 0x16 -#define SBS_CHARGER_STATUS_CHARGE_INHIBITED BIT(1) +#define SBS_CHARGER_STATUS_CHARGE_INHIBITED BIT(0) #define SBS_CHARGER_STATUS_RES_COLD BIT(9) #define SBS_CHARGER_STATUS_RES_HOT BIT(10) #define SBS_CHARGER_STATUS_BATTERY_PRESENT BIT(14) diff --git a/drivers/power/supply/sc27xx_fuel_gauge.c b/drivers/power/supply/sc27xx_fuel_gauge.c index 632977f84b954e6355e87a1edcac7b24020f16c1..bd23c4d9fed43482e972ccc086311e7bfcec2d54 100644 --- a/drivers/power/supply/sc27xx_fuel_gauge.c +++ b/drivers/power/supply/sc27xx_fuel_gauge.c @@ -733,13 +733,6 @@ static int sc27xx_fgu_set_property(struct power_supply *psy, return ret; } -static void sc27xx_fgu_external_power_changed(struct power_supply *psy) -{ - struct sc27xx_fgu_data *data = power_supply_get_drvdata(psy); - - power_supply_changed(data->battery); -} - static int sc27xx_fgu_property_is_writeable(struct power_supply *psy, enum power_supply_property psp) { @@ -774,7 +767,7 @@ static const struct power_supply_desc sc27xx_fgu_desc = { .num_properties = ARRAY_SIZE(sc27xx_fgu_props), .get_property = sc27xx_fgu_get_property, .set_property = sc27xx_fgu_set_property, - .external_power_changed = sc27xx_fgu_external_power_changed, + .external_power_changed = power_supply_changed, .property_is_writeable = sc27xx_fgu_property_is_writeable, .no_thermal = true, }; diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c index dc741ac156c32c084c7f469e44feccf297863ece..698ab7f5004bf6b74e2e57ac224f6138139f4666 100644 --- a/drivers/regulator/core.c +++ b/drivers/regulator/core.c @@ -5256,7 +5256,7 @@ static void rdev_init_debugfs(struct regulator_dev *rdev) } rdev->debugfs = debugfs_create_dir(rname, debugfs_root); - if (!rdev->debugfs) { + if (IS_ERR(rdev->debugfs)) { rdev_warn(rdev, "Failed to create debugfs directory\n"); return; } @@ -6178,7 +6178,7 @@ static int __init regulator_init(void) ret = class_register(®ulator_class); debugfs_root = debugfs_create_dir("regulator", NULL); - if (!debugfs_root) + if (IS_ERR(debugfs_root)) pr_warn("regulator: Failed to create debugfs directory\n"); #ifdef CONFIG_DEBUG_FS diff --git a/drivers/regulator/mt6359-regulator.c b/drivers/regulator/mt6359-regulator.c index 1849566784ab5e60b75cba257b965baa3bc388e7..3eb86ec21d08800bac3838f443f6615e75cb1562 100644 --- a/drivers/regulator/mt6359-regulator.c +++ b/drivers/regulator/mt6359-regulator.c @@ -951,9 +951,12 @@ static int mt6359_regulator_probe(struct platform_device *pdev) struct regulator_config config = {}; struct regulator_dev *rdev; struct mt6359_regulator_info *mt6359_info; - int i, hw_ver; + int i, hw_ver, ret; + + ret = regmap_read(mt6397->regmap, MT6359P_HWCID, &hw_ver); + if (ret) + return ret; - regmap_read(mt6397->regmap, MT6359P_HWCID, &hw_ver); if (hw_ver >= MT6359P_CHIP_VER) mt6359_info = mt6359p_regulators; else diff --git a/drivers/regulator/pca9450-regulator.c b/drivers/regulator/pca9450-regulator.c index 87a746dcb5169759f5413dfd8fe150933e267c3e..e75dd92f86ca728435247b1523a1a3283e0e0efc 100644 --- a/drivers/regulator/pca9450-regulator.c +++ b/drivers/regulator/pca9450-regulator.c @@ -264,7 +264,7 @@ static const struct pca9450_regulator_desc pca9450a_regulators[] = { .vsel_reg = PCA9450_REG_BUCK2OUT_DVS0, .vsel_mask = BUCK2OUT_DVS0_MASK, .enable_reg = PCA9450_REG_BUCK2CTRL, - .enable_mask = BUCK1_ENMODE_MASK, + .enable_mask = BUCK2_ENMODE_MASK, .ramp_reg = PCA9450_REG_BUCK2CTRL, .ramp_mask = BUCK2_RAMP_MASK, .ramp_delay_table = pca9450_dvs_buck_ramp_table, @@ -502,7 +502,7 @@ static const struct pca9450_regulator_desc pca9450bc_regulators[] = { .vsel_reg = PCA9450_REG_BUCK2OUT_DVS0, .vsel_mask = BUCK2OUT_DVS0_MASK, .enable_reg = PCA9450_REG_BUCK2CTRL, - .enable_mask = BUCK1_ENMODE_MASK, + .enable_mask = BUCK2_ENMODE_MASK, .ramp_reg = PCA9450_REG_BUCK2CTRL, .ramp_mask = BUCK2_RAMP_MASK, .ramp_delay_table = pca9450_dvs_buck_ramp_table, diff --git a/drivers/regulator/qcom-rpmh-regulator.c b/drivers/regulator/qcom-rpmh-regulator.c index b0a58c62b1e2c2f35dee8720fc5f0f85a5b5a5f1..f3b280af0773786999b36877f00f1eb52cf88d21 100644 --- a/drivers/regulator/qcom-rpmh-regulator.c +++ b/drivers/regulator/qcom-rpmh-regulator.c @@ -1057,21 +1057,21 @@ static const struct rpmh_vreg_init_data pm8450_vreg_data[] = { }; static const struct rpmh_vreg_init_data pm8550_vreg_data[] = { - RPMH_VREG("ldo1", "ldo%s1", &pmic5_pldo, "vdd-l1-l4-l10"), + RPMH_VREG("ldo1", "ldo%s1", &pmic5_nldo515, "vdd-l1-l4-l10"), RPMH_VREG("ldo2", "ldo%s2", &pmic5_pldo, "vdd-l2-l13-l14"), - RPMH_VREG("ldo3", "ldo%s3", &pmic5_nldo, "vdd-l3"), - RPMH_VREG("ldo4", "ldo%s4", &pmic5_nldo, "vdd-l1-l4-l10"), + RPMH_VREG("ldo3", "ldo%s3", &pmic5_nldo515, "vdd-l3"), + RPMH_VREG("ldo4", "ldo%s4", &pmic5_nldo515, "vdd-l1-l4-l10"), RPMH_VREG("ldo5", "ldo%s5", &pmic5_pldo, "vdd-l5-l16"), - RPMH_VREG("ldo6", "ldo%s6", &pmic5_pldo_lv, "vdd-l6-l7"), - RPMH_VREG("ldo7", "ldo%s7", &pmic5_pldo_lv, "vdd-l6-l7"), - RPMH_VREG("ldo8", "ldo%s8", &pmic5_pldo_lv, "vdd-l8-l9"), + RPMH_VREG("ldo6", "ldo%s6", &pmic5_pldo, "vdd-l6-l7"), + RPMH_VREG("ldo7", "ldo%s7", &pmic5_pldo, "vdd-l6-l7"), + RPMH_VREG("ldo8", "ldo%s8", &pmic5_pldo, "vdd-l8-l9"), RPMH_VREG("ldo9", "ldo%s9", &pmic5_pldo, "vdd-l8-l9"), - RPMH_VREG("ldo10", "ldo%s10", &pmic5_nldo, "vdd-l1-l4-l10"), - RPMH_VREG("ldo11", "ldo%s11", &pmic5_nldo, "vdd-l11"), + RPMH_VREG("ldo10", "ldo%s10", &pmic5_nldo515, "vdd-l1-l4-l10"), + RPMH_VREG("ldo11", "ldo%s11", &pmic5_nldo515, "vdd-l11"), RPMH_VREG("ldo12", "ldo%s12", &pmic5_pldo, "vdd-l12"), RPMH_VREG("ldo13", "ldo%s13", &pmic5_pldo, "vdd-l2-l13-l14"), RPMH_VREG("ldo14", "ldo%s14", &pmic5_pldo, "vdd-l2-l13-l14"), - RPMH_VREG("ldo15", "ldo%s15", &pmic5_pldo, "vdd-l15"), + RPMH_VREG("ldo15", "ldo%s15", &pmic5_nldo515, "vdd-l15"), RPMH_VREG("ldo16", "ldo%s16", &pmic5_pldo, "vdd-l5-l16"), RPMH_VREG("ldo17", "ldo%s17", &pmic5_pldo, "vdd-l17"), RPMH_VREG("bob1", "bob%s1", &pmic5_bob, "vdd-bob1"), @@ -1086,9 +1086,9 @@ static const struct rpmh_vreg_init_data pm8550vs_vreg_data[] = { RPMH_VREG("smps4", "smp%s4", &pmic5_ftsmps525_lv, "vdd-s4"), RPMH_VREG("smps5", "smp%s5", &pmic5_ftsmps525_lv, "vdd-s5"), RPMH_VREG("smps6", "smp%s6", &pmic5_ftsmps525_mv, "vdd-s6"), - RPMH_VREG("ldo1", "ldo%s1", &pmic5_nldo, "vdd-l1"), - RPMH_VREG("ldo2", "ldo%s2", &pmic5_nldo, "vdd-l2"), - RPMH_VREG("ldo3", "ldo%s3", &pmic5_nldo, "vdd-l3"), + RPMH_VREG("ldo1", "ldo%s1", &pmic5_nldo515, "vdd-l1"), + RPMH_VREG("ldo2", "ldo%s2", &pmic5_nldo515, "vdd-l2"), + RPMH_VREG("ldo3", "ldo%s3", &pmic5_nldo515, "vdd-l3"), {} }; @@ -1101,9 +1101,9 @@ static const struct rpmh_vreg_init_data pm8550ve_vreg_data[] = { RPMH_VREG("smps6", "smp%s6", &pmic5_ftsmps525_lv, "vdd-s6"), RPMH_VREG("smps7", "smp%s7", &pmic5_ftsmps525_lv, "vdd-s7"), RPMH_VREG("smps8", "smp%s8", &pmic5_ftsmps525_lv, "vdd-s8"), - RPMH_VREG("ldo1", "ldo%s1", &pmic5_nldo, "vdd-l1"), - RPMH_VREG("ldo2", "ldo%s2", &pmic5_nldo, "vdd-l2"), - RPMH_VREG("ldo3", "ldo%s3", &pmic5_nldo, "vdd-l3"), + RPMH_VREG("ldo1", "ldo%s1", &pmic5_nldo515, "vdd-l1"), + RPMH_VREG("ldo2", "ldo%s2", &pmic5_nldo515, "vdd-l2"), + RPMH_VREG("ldo3", "ldo%s3", &pmic5_nldo515, "vdd-l3"), {} }; diff --git a/drivers/s390/block/dasd_eckd.c b/drivers/s390/block/dasd_eckd.c index ade1369fe5ed3056706b3f8361af2efe8dc6e327..113c509bf6d052bf70b3642adcf4bb6ad41ffeea 100644 --- a/drivers/s390/block/dasd_eckd.c +++ b/drivers/s390/block/dasd_eckd.c @@ -127,6 +127,8 @@ static int prepare_itcw(struct itcw *, unsigned int, unsigned int, int, struct dasd_device *, struct dasd_device *, unsigned int, int, unsigned int, unsigned int, unsigned int, unsigned int); +static int dasd_eckd_query_pprc_status(struct dasd_device *, + struct dasd_pprc_data_sc4 *); /* initial attempt at a probe function. this can be simplified once * the other detection code is gone */ @@ -3733,6 +3735,26 @@ static int count_exts(unsigned int from, unsigned int to, int trks_per_ext) return count; } +static int dasd_in_copy_relation(struct dasd_device *device) +{ + struct dasd_pprc_data_sc4 *temp; + int rc; + + if (!dasd_eckd_pprc_enabled(device)) + return 0; + + temp = kzalloc(sizeof(*temp), GFP_KERNEL); + if (!temp) + return -ENOMEM; + + rc = dasd_eckd_query_pprc_status(device, temp); + if (!rc) + rc = temp->dev_info[0].state; + + kfree(temp); + return rc; +} + /* * Release allocated space for a given range or an entire volume. */ @@ -3749,6 +3771,7 @@ dasd_eckd_dso_ras(struct dasd_device *device, struct dasd_block *block, int cur_to_trk, cur_from_trk; struct dasd_ccw_req *cqr; u32 beg_cyl, end_cyl; + int copy_relation; struct ccw1 *ccw; int trks_per_ext; size_t ras_size; @@ -3760,6 +3783,10 @@ dasd_eckd_dso_ras(struct dasd_device *device, struct dasd_block *block, if (dasd_eckd_ras_sanity_checks(device, first_trk, last_trk)) return ERR_PTR(-EINVAL); + copy_relation = dasd_in_copy_relation(device); + if (copy_relation < 0) + return ERR_PTR(copy_relation); + rq = req ? blk_mq_rq_to_pdu(req) : NULL; features = &private->features; @@ -3788,9 +3815,11 @@ dasd_eckd_dso_ras(struct dasd_device *device, struct dasd_block *block, /* * This bit guarantees initialisation of tracks within an extent that is * not fully specified, but is only supported with a certain feature - * subset. + * subset and for devices not in a copy relation. */ - ras_data->op_flags.guarantee_init = !!(features->feature[56] & 0x01); + if (features->feature[56] & 0x01 && !copy_relation) + ras_data->op_flags.guarantee_init = 1; + ras_data->lss = private->conf.ned->ID; ras_data->dev_addr = private->conf.ned->unit_addr; ras_data->nr_exts = nr_exts; diff --git a/drivers/s390/block/dasd_ioctl.c b/drivers/s390/block/dasd_ioctl.c index 9327dcdd6e5e43271276226ebdda360130389ee8..8fca725b3daeceda4f80be4200d65464349aae9f 100644 --- a/drivers/s390/block/dasd_ioctl.c +++ b/drivers/s390/block/dasd_ioctl.c @@ -552,10 +552,10 @@ static int __dasd_ioctl_information(struct dasd_block *block, memcpy(dasd_info->type, base->discipline->name, 4); - spin_lock_irqsave(&block->queue_lock, flags); + spin_lock_irqsave(get_ccwdev_lock(base->cdev), flags); list_for_each(l, &base->ccw_queue) dasd_info->chanq_len++; - spin_unlock_irqrestore(&block->queue_lock, flags); + spin_unlock_irqrestore(get_ccwdev_lock(base->cdev), flags); return 0; } diff --git a/drivers/s390/cio/device.c b/drivers/s390/cio/device.c index 8eb089b99cde92bad4670656b281b4cfcc8425b6..c0d620ffea618e87da0ca248cc2effbc3800cc62 100644 --- a/drivers/s390/cio/device.c +++ b/drivers/s390/cio/device.c @@ -1111,6 +1111,8 @@ static void io_subchannel_verify(struct subchannel *sch) cdev = sch_get_cdev(sch); if (cdev) dev_fsm_event(cdev, DEV_EVENT_VERIFY); + else + css_schedule_eval(sch->schid); } static void io_subchannel_terminate_path(struct subchannel *sch, u8 mask) @@ -1374,6 +1376,7 @@ void ccw_device_set_notoper(struct ccw_device *cdev) enum io_sch_action { IO_SCH_UNREG, IO_SCH_ORPH_UNREG, + IO_SCH_UNREG_CDEV, IO_SCH_ATTACH, IO_SCH_UNREG_ATTACH, IO_SCH_ORPH_ATTACH, @@ -1406,7 +1409,7 @@ static enum io_sch_action sch_get_action(struct subchannel *sch) } if ((sch->schib.pmcw.pam & sch->opm) == 0) { if (ccw_device_notify(cdev, CIO_NO_PATH) != NOTIFY_OK) - return IO_SCH_UNREG; + return IO_SCH_UNREG_CDEV; return IO_SCH_DISC; } if (device_is_disconnected(cdev)) @@ -1468,6 +1471,7 @@ static int io_subchannel_sch_event(struct subchannel *sch, int process) case IO_SCH_ORPH_ATTACH: ccw_device_set_disconnected(cdev); break; + case IO_SCH_UNREG_CDEV: case IO_SCH_UNREG_ATTACH: case IO_SCH_UNREG: if (!cdev) @@ -1501,6 +1505,7 @@ static int io_subchannel_sch_event(struct subchannel *sch, int process) if (rc) goto out; break; + case IO_SCH_UNREG_CDEV: case IO_SCH_UNREG_ATTACH: spin_lock_irqsave(sch->lock, flags); sch_set_cdev(sch, NULL); diff --git a/drivers/s390/cio/qdio.h b/drivers/s390/cio/qdio.h index 5ea6249d81803ad2358592f2e4c40cb167692828..641f0dbb65a90b15577b27a941a8b65718b09159 100644 --- a/drivers/s390/cio/qdio.h +++ b/drivers/s390/cio/qdio.h @@ -95,7 +95,7 @@ static inline int do_sqbs(u64 token, unsigned char state, int queue, " lgr 1,%[token]\n" " .insn rsy,0xeb000000008a,%[qs],%[ccq],0(%[state])" : [ccq] "+&d" (_ccq), [qs] "+&d" (_queuestart) - : [state] "d" ((unsigned long)state), [token] "d" (token) + : [state] "a" ((unsigned long)state), [token] "d" (token) : "memory", "cc", "1"); *count = _ccq & 0xff; *start = _queuestart & 0xff; diff --git a/drivers/s390/crypto/pkey_api.c b/drivers/s390/crypto/pkey_api.c index 5a05d1cdfec20adf5bc43a6fad6fa2f219fc9805..a8def50c149bd433b1324cbd6d3798ea4130bc7e 100644 --- a/drivers/s390/crypto/pkey_api.c +++ b/drivers/s390/crypto/pkey_api.c @@ -1293,6 +1293,7 @@ static long pkey_unlocked_ioctl(struct file *filp, unsigned int cmd, return PTR_ERR(kkey); rc = pkey_keyblob2pkey(kkey, ktp.keylen, &ktp.protkey); DEBUG_DBG("%s pkey_keyblob2pkey()=%d\n", __func__, rc); + memzero_explicit(kkey, ktp.keylen); kfree(kkey); if (rc) break; @@ -1426,6 +1427,7 @@ static long pkey_unlocked_ioctl(struct file *filp, unsigned int cmd, kkey, ktp.keylen, &ktp.protkey); DEBUG_DBG("%s pkey_keyblob2pkey2()=%d\n", __func__, rc); kfree(apqns); + memzero_explicit(kkey, ktp.keylen); kfree(kkey); if (rc) break; @@ -1552,6 +1554,7 @@ static long pkey_unlocked_ioctl(struct file *filp, unsigned int cmd, protkey, &protkeylen); DEBUG_DBG("%s pkey_keyblob2pkey3()=%d\n", __func__, rc); kfree(apqns); + memzero_explicit(kkey, ktp.keylen); kfree(kkey); if (rc) { kfree(protkey); diff --git a/drivers/s390/net/ism_drv.c b/drivers/s390/net/ism_drv.c index 8acb9eba691b5ec9f804f8cf647f9752ee54a345..c2096e4bba319960de45c4768943d0aa1634f87a 100644 --- a/drivers/s390/net/ism_drv.c +++ b/drivers/s390/net/ism_drv.c @@ -771,14 +771,6 @@ static int __init ism_init(void) static void __exit ism_exit(void) { - struct ism_dev *ism; - - mutex_lock(&ism_dev_list.mutex); - list_for_each_entry(ism, &ism_dev_list.list, list) { - ism_dev_exit(ism); - } - mutex_unlock(&ism_dev_list.mutex); - pci_unregister_driver(&ism_driver); debug_unregister(ism_debug_info); } diff --git a/drivers/scsi/aacraid/aacraid.h b/drivers/scsi/aacraid/aacraid.h index 5e115e8b2ba462bae6d821e195a6f9462e10c70f..7c6efde75da664b3819161d5aa96466b0d26aaa1 100644 --- a/drivers/scsi/aacraid/aacraid.h +++ b/drivers/scsi/aacraid/aacraid.h @@ -1678,6 +1678,7 @@ struct aac_dev u32 handle_pci_error; bool init_reset; u8 soft_reset_support; + u8 use_map_queue; }; #define aac_adapter_interrupt(dev) \ diff --git a/drivers/scsi/aacraid/commsup.c b/drivers/scsi/aacraid/commsup.c index deb32c9f4b3e6fca8041c3aa093671d04a58b53d..3f062e4013ab6c3b9cb54e46f608bee974b5e9a1 100644 --- a/drivers/scsi/aacraid/commsup.c +++ b/drivers/scsi/aacraid/commsup.c @@ -223,8 +223,12 @@ int aac_fib_setup(struct aac_dev * dev) struct fib *aac_fib_alloc_tag(struct aac_dev *dev, struct scsi_cmnd *scmd) { struct fib *fibptr; + u32 blk_tag; + int i; - fibptr = &dev->fibs[scsi_cmd_to_rq(scmd)->tag]; + blk_tag = blk_mq_unique_tag(scsi_cmd_to_rq(scmd)); + i = blk_mq_unique_tag_to_tag(blk_tag); + fibptr = &dev->fibs[i]; /* * Null out fields that depend on being zero at the start of * each I/O diff --git a/drivers/scsi/aacraid/linit.c b/drivers/scsi/aacraid/linit.c index 68f4dbcfff49250a632a7c1e2e10f74e303a0e19..c4a36c0be527cd86b04713dbfcab4c1206e30618 100644 --- a/drivers/scsi/aacraid/linit.c +++ b/drivers/scsi/aacraid/linit.c @@ -19,6 +19,7 @@ #include #include +#include #include #include #include @@ -504,6 +505,15 @@ static int aac_slave_configure(struct scsi_device *sdev) return 0; } +static void aac_map_queues(struct Scsi_Host *shost) +{ + struct aac_dev *aac = (struct aac_dev *)shost->hostdata; + + blk_mq_pci_map_queues(&shost->tag_set.map[HCTX_TYPE_DEFAULT], + aac->pdev, 0); + aac->use_map_queue = true; +} + /** * aac_change_queue_depth - alter queue depths * @sdev: SCSI device we are considering @@ -1488,6 +1498,7 @@ static const struct scsi_host_template aac_driver_template = { .bios_param = aac_biosparm, .shost_groups = aac_host_groups, .slave_configure = aac_slave_configure, + .map_queues = aac_map_queues, .change_queue_depth = aac_change_queue_depth, .sdev_groups = aac_dev_groups, .eh_abort_handler = aac_eh_abort, @@ -1775,6 +1786,8 @@ static int aac_probe_one(struct pci_dev *pdev, const struct pci_device_id *id) shost->max_lun = AAC_MAX_LUN; pci_set_drvdata(pdev, shost); + shost->nr_hw_queues = aac->max_msix; + shost->host_tagset = 1; error = scsi_add_host(shost, &pdev->dev); if (error) @@ -1906,6 +1919,7 @@ static void aac_remove_one(struct pci_dev *pdev) struct aac_dev *aac = (struct aac_dev *)shost->hostdata; aac_cancel_rescan_worker(aac); + aac->use_map_queue = false; scsi_remove_host(shost); __aac_shutdown(aac); diff --git a/drivers/scsi/aacraid/src.c b/drivers/scsi/aacraid/src.c index 11ef58204e96f179227c166433eb4932ed7dbafa..61949f3741886ba1b439292b473bf27d50d1b227 100644 --- a/drivers/scsi/aacraid/src.c +++ b/drivers/scsi/aacraid/src.c @@ -493,6 +493,10 @@ static int aac_src_deliver_message(struct fib *fib) #endif u16 vector_no; + struct scsi_cmnd *scmd; + u32 blk_tag; + struct Scsi_Host *shost = dev->scsi_host_ptr; + struct blk_mq_queue_map *qmap; atomic_inc(&q->numpending); @@ -505,8 +509,25 @@ static int aac_src_deliver_message(struct fib *fib) if ((dev->comm_interface == AAC_COMM_MESSAGE_TYPE3) && dev->sa_firmware) vector_no = aac_get_vector(dev); - else - vector_no = fib->vector_no; + else { + if (!fib->vector_no || !fib->callback_data) { + if (shost && dev->use_map_queue) { + qmap = &shost->tag_set.map[HCTX_TYPE_DEFAULT]; + vector_no = qmap->mq_map[raw_smp_processor_id()]; + } + /* + * We hardcode the vector_no for + * reserved commands as a valid shost is + * absent during the init + */ + else + vector_no = 0; + } else { + scmd = (struct scsi_cmnd *)fib->callback_data; + blk_tag = blk_mq_unique_tag(scsi_cmd_to_rq(scmd)); + vector_no = blk_mq_unique_tag_to_hwq(blk_tag); + } + } if (native_hba) { if (fib->flags & FIB_CONTEXT_FLAG_NATIVE_HBA_TMF) { diff --git a/drivers/scsi/lpfc/lpfc_bsg.c b/drivers/scsi/lpfc/lpfc_bsg.c index 9a322a3a215020609513faf15f0b822558beaba3..595dca92e8db5ad76dffddac8181fd561408372b 100644 --- a/drivers/scsi/lpfc/lpfc_bsg.c +++ b/drivers/scsi/lpfc/lpfc_bsg.c @@ -889,7 +889,7 @@ lpfc_bsg_ct_unsol_event(struct lpfc_hba *phba, struct lpfc_sli_ring *pring, struct lpfc_iocbq *piocbq) { uint32_t evt_req_id = 0; - uint32_t cmd; + u16 cmd; struct lpfc_dmabuf *dmabuf = NULL; struct lpfc_bsg_event *evt; struct event_data *evt_dat = NULL; @@ -915,7 +915,7 @@ lpfc_bsg_ct_unsol_event(struct lpfc_hba *phba, struct lpfc_sli_ring *pring, ct_req = (struct lpfc_sli_ct_request *)bdeBuf1->virt; evt_req_id = ct_req->FsType; - cmd = ct_req->CommandResponse.bits.CmdRsp; + cmd = be16_to_cpu(ct_req->CommandResponse.bits.CmdRsp); spin_lock_irqsave(&phba->ct_ev_lock, flags); list_for_each_entry(evt, &phba->ct_ev_waiters, node) { @@ -3186,8 +3186,8 @@ lpfc_bsg_diag_loopback_run(struct bsg_job *job) ctreq->RevisionId.bits.InId = 0; ctreq->FsType = SLI_CT_ELX_LOOPBACK; ctreq->FsSubType = 0; - ctreq->CommandResponse.bits.CmdRsp = ELX_LOOPBACK_DATA; - ctreq->CommandResponse.bits.Size = size; + ctreq->CommandResponse.bits.CmdRsp = cpu_to_be16(ELX_LOOPBACK_DATA); + ctreq->CommandResponse.bits.Size = cpu_to_be16(size); segment_offset = ELX_LOOPBACK_HEADER_SZ; } else segment_offset = 0; diff --git a/drivers/scsi/qla2xxx/qla_def.h b/drivers/scsi/qla2xxx/qla_def.h index df5e5b7fdcfe73accb080e412f43372c44974cbe..84aa3571be6d4de1f8e7cf7e2e0a011a6c4b6fd1 100644 --- a/drivers/scsi/qla2xxx/qla_def.h +++ b/drivers/scsi/qla2xxx/qla_def.h @@ -3796,6 +3796,7 @@ struct qla_qpair { uint64_t retry_term_jiff; struct qla_tgt_counters tgt_counters; uint16_t cpuid; + bool cpu_mapped; struct qla_fw_resources fwres ____cacheline_aligned; struct qla_buf_pool buf_pool; u32 cmd_cnt; diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c index ec0423ec66817c1a54ea00d508aeaccc02def258..1a955c3ff3d6ce2732e3fd03e65203a5162dbc9a 100644 --- a/drivers/scsi/qla2xxx/qla_init.c +++ b/drivers/scsi/qla2xxx/qla_init.c @@ -9426,6 +9426,9 @@ struct qla_qpair *qla2xxx_create_qpair(struct scsi_qla_host *vha, int qos, qpair->rsp->req = qpair->req; qpair->rsp->qpair = qpair; + if (!qpair->cpu_mapped) + qla_cpu_update(qpair, raw_smp_processor_id()); + if (IS_T10_PI_CAPABLE(ha) && ql2xenabledif) { if (ha->fw_attributes & BIT_4) qpair->difdix_supported = 1; diff --git a/drivers/scsi/qla2xxx/qla_inline.h b/drivers/scsi/qla2xxx/qla_inline.h index cce6e425c1214a9aabaf9fe2677ba1ef0419cdb9..7b42558a8839ad51e0b92ede49c89656a16b03f7 100644 --- a/drivers/scsi/qla2xxx/qla_inline.h +++ b/drivers/scsi/qla2xxx/qla_inline.h @@ -539,11 +539,14 @@ qla_mapq_init_qp_cpu_map(struct qla_hw_data *ha, if (!ha->qp_cpu_map) return; mask = pci_irq_get_affinity(ha->pdev, msix->vector_base0); + if (!mask) + return; qpair->cpuid = cpumask_first(mask); for_each_cpu(cpu, mask) { ha->qp_cpu_map[cpu] = qpair; } msix->cpuid = qpair->cpuid; + qpair->cpu_mapped = true; } static inline void diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c index 71feda2cdb63046a713c5018ae64948abed1084f..245e3a5d81fd33b3040df3000712d5d961ec9d6e 100644 --- a/drivers/scsi/qla2xxx/qla_isr.c +++ b/drivers/scsi/qla2xxx/qla_isr.c @@ -3770,6 +3770,9 @@ void qla24xx_process_response_queue(struct scsi_qla_host *vha, if (rsp->qpair->cpuid != smp_processor_id() || !rsp->qpair->rcv_intr) { rsp->qpair->rcv_intr = 1; + + if (!rsp->qpair->cpu_mapped) + qla_cpu_update(rsp->qpair, raw_smp_processor_id()); } #define __update_rsp_in(_is_shadow_hba, _rsp, _rsp_in) \ diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index b7c569a42aa477551597ff0629fc6c04888277fb..0226c9279cef60b0daf3cebfba77c467200975c9 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -1463,6 +1463,8 @@ static int scsi_dispatch_cmd(struct scsi_cmnd *cmd) struct Scsi_Host *host = cmd->device->host; int rtn = 0; + atomic_inc(&cmd->device->iorequest_cnt); + /* check if the device is still usable */ if (unlikely(cmd->device->sdev_state == SDEV_DEL)) { /* in SDEV_DEL we error all commands. DID_NO_CONNECT @@ -1483,6 +1485,7 @@ static int scsi_dispatch_cmd(struct scsi_cmnd *cmd) */ SCSI_LOG_MLQUEUE(3, scmd_printk(KERN_INFO, cmd, "queuecommand : device blocked\n")); + atomic_dec(&cmd->device->iorequest_cnt); return SCSI_MLQUEUE_DEVICE_BUSY; } @@ -1515,6 +1518,7 @@ static int scsi_dispatch_cmd(struct scsi_cmnd *cmd) trace_scsi_dispatch_cmd_start(cmd); rtn = host->hostt->queuecommand(host, cmd); if (rtn) { + atomic_dec(&cmd->device->iorequest_cnt); trace_scsi_dispatch_cmd_error(cmd, rtn); if (rtn != SCSI_MLQUEUE_DEVICE_BUSY && rtn != SCSI_MLQUEUE_TARGET_BUSY) @@ -1761,7 +1765,6 @@ static blk_status_t scsi_queue_rq(struct blk_mq_hw_ctx *hctx, goto out_dec_host_busy; } - atomic_inc(&cmd->device->iorequest_cnt); return BLK_STS_OK; out_dec_host_busy: diff --git a/drivers/scsi/stex.c b/drivers/scsi/stex.c index 5b230e149c3dfe346588121e03ed81efd64cde95..8ffb75be99bc8fab964fc6df5663f9392c844258 100644 --- a/drivers/scsi/stex.c +++ b/drivers/scsi/stex.c @@ -109,7 +109,9 @@ enum { TASK_ATTRIBUTE_HEADOFQUEUE = 0x1, TASK_ATTRIBUTE_ORDERED = 0x2, TASK_ATTRIBUTE_ACA = 0x4, +}; +enum { SS_STS_NORMAL = 0x80000000, SS_STS_DONE = 0x40000000, SS_STS_HANDSHAKE = 0x20000000, @@ -121,7 +123,9 @@ enum { SS_I2H_REQUEST_RESET = 0x2000, SS_MU_OPERATIONAL = 0x80000000, +}; +enum { STEX_CDB_LENGTH = 16, STATUS_VAR_LEN = 128, diff --git a/drivers/scsi/storvsc_drv.c b/drivers/scsi/storvsc_drv.c index d9ce379c4d2e836476a04a724622d80a0a85a36a..659196a2f63ad33d3f16e9cb0e26449e78ef46a9 100644 --- a/drivers/scsi/storvsc_drv.c +++ b/drivers/scsi/storvsc_drv.c @@ -1567,6 +1567,8 @@ static int storvsc_device_configure(struct scsi_device *sdevice) { blk_queue_rq_timeout(sdevice->request_queue, (storvsc_timeout * HZ)); + /* storvsc devices don't support MAINTENANCE_IN SCSI cmd */ + sdevice->no_report_opcodes = 1; sdevice->no_write_same = 1; /* @@ -1780,7 +1782,7 @@ static int storvsc_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *scmnd) length = scsi_bufflen(scmnd); payload = (struct vmbus_packet_mpb_array *)&cmd_request->mpb; - payload_sz = sizeof(cmd_request->mpb); + payload_sz = 0; if (scsi_sg_count(scmnd)) { unsigned long offset_in_hvpg = offset_in_hvpage(sgl->offset); @@ -1789,10 +1791,10 @@ static int storvsc_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *scmnd) unsigned long hvpfn, hvpfns_to_add; int j, i = 0, sg_count; - if (hvpg_count > MAX_PAGE_BUFFER_COUNT) { + payload_sz = (hvpg_count * sizeof(u64) + + sizeof(struct vmbus_packet_mpb_array)); - payload_sz = (hvpg_count * sizeof(u64) + - sizeof(struct vmbus_packet_mpb_array)); + if (hvpg_count > MAX_PAGE_BUFFER_COUNT) { payload = kzalloc(payload_sz, GFP_ATOMIC); if (!payload) return SCSI_MLQUEUE_DEVICE_BUSY; diff --git a/drivers/soc/fsl/qe/Kconfig b/drivers/soc/fsl/qe/Kconfig index 7268c2fbcbc1d80b7b72b45ae4668584016f92c8..e0d096607fefbaf4f9ee93921046ec675a9af05f 100644 --- a/drivers/soc/fsl/qe/Kconfig +++ b/drivers/soc/fsl/qe/Kconfig @@ -36,7 +36,7 @@ config UCC config CPM_TSA tristate "CPM TSA support" depends on OF && HAS_IOMEM - depends on CPM1 || COMPILE_TEST + depends on CPM1 || (CPM && COMPILE_TEST) help Freescale CPM Time Slot Assigner (TSA) controller. @@ -47,7 +47,7 @@ config CPM_TSA config CPM_QMC tristate "CPM QMC support" depends on OF && HAS_IOMEM - depends on CPM1 || (FSL_SOC && COMPILE_TEST) + depends on CPM1 || (FSL_SOC && CPM && COMPILE_TEST) depends on CPM_TSA help Freescale CPM QUICC Multichannel Controller diff --git a/drivers/soc/qcom/Makefile b/drivers/soc/qcom/Makefile index 0f43a88b4894e4675b7e456a08d906890c538829..89b775512befa1cc2d12462a09dd567cf5d6a726 100644 --- a/drivers/soc/qcom/Makefile +++ b/drivers/soc/qcom/Makefile @@ -32,4 +32,5 @@ obj-$(CONFIG_QCOM_RPMHPD) += rpmhpd.o obj-$(CONFIG_QCOM_RPMPD) += rpmpd.o obj-$(CONFIG_QCOM_KRYO_L2_ACCESSORS) += kryo-l2-accessors.o obj-$(CONFIG_QCOM_ICC_BWMON) += icc-bwmon.o -obj-$(CONFIG_QCOM_INLINE_CRYPTO_ENGINE) += ice.o +qcom_ice-objs += ice.o +obj-$(CONFIG_QCOM_INLINE_CRYPTO_ENGINE) += qcom_ice.o diff --git a/drivers/soc/qcom/icc-bwmon.c b/drivers/soc/qcom/icc-bwmon.c index fd58c5b698978765d3213b8689d7db4dab39d415..f65bfeca7ed6b9802e7a2eb038f6fe72710b98af 100644 --- a/drivers/soc/qcom/icc-bwmon.c +++ b/drivers/soc/qcom/icc-bwmon.c @@ -773,12 +773,12 @@ static int bwmon_probe(struct platform_device *pdev) bwmon->max_bw_kbps = UINT_MAX; opp = dev_pm_opp_find_bw_floor(dev, &bwmon->max_bw_kbps, 0); if (IS_ERR(opp)) - return dev_err_probe(dev, ret, "failed to find max peak bandwidth\n"); + return dev_err_probe(dev, PTR_ERR(opp), "failed to find max peak bandwidth\n"); bwmon->min_bw_kbps = 0; opp = dev_pm_opp_find_bw_ceil(dev, &bwmon->min_bw_kbps, 0); if (IS_ERR(opp)) - return dev_err_probe(dev, ret, "failed to find min peak bandwidth\n"); + return dev_err_probe(dev, PTR_ERR(opp), "failed to find min peak bandwidth\n"); bwmon->dev = dev; diff --git a/drivers/soc/qcom/ramp_controller.c b/drivers/soc/qcom/ramp_controller.c index dc74d2a19de2bd9ef89e758764f34523bab000fb..5e3ba0be090354fb05d84c450879c8ddb9a2625a 100644 --- a/drivers/soc/qcom/ramp_controller.c +++ b/drivers/soc/qcom/ramp_controller.c @@ -296,7 +296,7 @@ static int qcom_ramp_controller_probe(struct platform_device *pdev) return -ENOMEM; qrc->desc = device_get_match_data(&pdev->dev); - if (!qrc) + if (!qrc->desc) return -EINVAL; qrc->regmap = devm_regmap_init_mmio(&pdev->dev, base, &qrc_regmap_config); diff --git a/drivers/soc/qcom/rmtfs_mem.c b/drivers/soc/qcom/rmtfs_mem.c index ce48a9f3b4c87f284eed596f5d0c578a3dea74f2..f83811f51175352777ccc95886df4d39d4ffd8cb 100644 --- a/drivers/soc/qcom/rmtfs_mem.c +++ b/drivers/soc/qcom/rmtfs_mem.c @@ -233,6 +233,7 @@ static int qcom_rmtfs_mem_probe(struct platform_device *pdev) num_vmids = 0; } else if (num_vmids < 0) { dev_err(&pdev->dev, "failed to count qcom,vmid elements: %d\n", num_vmids); + ret = num_vmids; goto remove_cdev; } else if (num_vmids > NUM_MAX_VMIDS) { dev_warn(&pdev->dev, diff --git a/drivers/soc/qcom/rpmh-rsc.c b/drivers/soc/qcom/rpmh-rsc.c index f93544f6d7961a01e18858f9b8cfee42f3da5645..0dd4363ebac8f77ffc36b5798697a87475977d41 100644 --- a/drivers/soc/qcom/rpmh-rsc.c +++ b/drivers/soc/qcom/rpmh-rsc.c @@ -1073,7 +1073,7 @@ static int rpmh_rsc_probe(struct platform_device *pdev) drv->ver.minor = rsc_id & (MINOR_VER_MASK << MINOR_VER_SHIFT); drv->ver.minor >>= MINOR_VER_SHIFT; - if (drv->ver.major == 3 && drv->ver.minor >= 0) + if (drv->ver.major == 3) drv->regs = rpmh_rsc_reg_offset_ver_3_0; else drv->regs = rpmh_rsc_reg_offset_ver_2_7; diff --git a/drivers/soc/qcom/rpmhpd.c b/drivers/soc/qcom/rpmhpd.c index f20e2a49a6695757385020e25a6f96bd53f773ad..63c35a32065ba4463324a67bcefa9588f1179c28 100644 --- a/drivers/soc/qcom/rpmhpd.c +++ b/drivers/soc/qcom/rpmhpd.c @@ -342,6 +342,21 @@ static const struct rpmhpd_desc sm8150_desc = { .num_pds = ARRAY_SIZE(sm8150_rpmhpds), }; +static struct rpmhpd *sa8155p_rpmhpds[] = { + [SA8155P_CX] = &cx_w_mx_parent, + [SA8155P_CX_AO] = &cx_ao_w_mx_parent, + [SA8155P_EBI] = &ebi, + [SA8155P_GFX] = &gfx, + [SA8155P_MSS] = &mss, + [SA8155P_MX] = &mx, + [SA8155P_MX_AO] = &mx_ao, +}; + +static const struct rpmhpd_desc sa8155p_desc = { + .rpmhpds = sa8155p_rpmhpds, + .num_pds = ARRAY_SIZE(sa8155p_rpmhpds), +}; + /* SM8250 RPMH powerdomains */ static struct rpmhpd *sm8250_rpmhpds[] = { [SM8250_CX] = &cx_w_mx_parent, @@ -519,6 +534,7 @@ static const struct rpmhpd_desc sc8280xp_desc = { static const struct of_device_id rpmhpd_match_table[] = { { .compatible = "qcom,qdu1000-rpmhpd", .data = &qdu1000_desc }, + { .compatible = "qcom,sa8155p-rpmhpd", .data = &sa8155p_desc }, { .compatible = "qcom,sa8540p-rpmhpd", .data = &sa8540p_desc }, { .compatible = "qcom,sa8775p-rpmhpd", .data = &sa8775p_desc }, { .compatible = "qcom,sc7180-rpmhpd", .data = &sc7180_desc }, diff --git a/drivers/soundwire/dmi-quirks.c b/drivers/soundwire/dmi-quirks.c index 58ea013fa918a105edf1d002bc44e6fce815f1e6..2a1096dab63d3c06c1d4dc9ab44a9a409273b048 100644 --- a/drivers/soundwire/dmi-quirks.c +++ b/drivers/soundwire/dmi-quirks.c @@ -99,6 +99,13 @@ static const struct dmi_system_id adr_remap_quirk_table[] = { }, .driver_data = (void *)intel_tgl_bios, }, + { + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "HP"), + DMI_MATCH(DMI_BOARD_NAME, "8709"), + }, + .driver_data = (void *)intel_tgl_bios, + }, { /* quirk used for NUC15 'Bishop County' LAPBC510 and LAPBC710 skews */ .matches = { diff --git a/drivers/soundwire/qcom.c b/drivers/soundwire/qcom.c index c296e0bf897b756e0f00428c8b558d40e50ae5ef..280455f047a3608ae4689d8126482c25903f8ec5 100644 --- a/drivers/soundwire/qcom.c +++ b/drivers/soundwire/qcom.c @@ -1099,8 +1099,10 @@ static int qcom_swrm_startup(struct snd_pcm_substream *substream, } sruntime = sdw_alloc_stream(dai->name); - if (!sruntime) - return -ENOMEM; + if (!sruntime) { + ret = -ENOMEM; + goto err_alloc; + } ctrl->sruntime[dai->id] = sruntime; @@ -1110,12 +1112,19 @@ static int qcom_swrm_startup(struct snd_pcm_substream *substream, if (ret < 0 && ret != -ENOTSUPP) { dev_err(dai->dev, "Failed to set sdw stream on %s\n", codec_dai->name); - sdw_release_stream(sruntime); - return ret; + goto err_set_stream; } } return 0; + +err_set_stream: + sdw_release_stream(sruntime); +err_alloc: + pm_runtime_mark_last_busy(ctrl->dev); + pm_runtime_put_autosuspend(ctrl->dev); + + return ret; } static void qcom_swrm_shutdown(struct snd_pcm_substream *substream, diff --git a/drivers/soundwire/stream.c b/drivers/soundwire/stream.c index c2191c07442b02f0a1e4f2725eb35b2222bdb0b7..379228f2218696ec42801fcbb8a080dc6b87f850 100644 --- a/drivers/soundwire/stream.c +++ b/drivers/soundwire/stream.c @@ -2021,8 +2021,10 @@ int sdw_stream_add_slave(struct sdw_slave *slave, skip_alloc_master_rt: s_rt = sdw_slave_rt_find(slave, stream); - if (s_rt) + if (s_rt) { + alloc_slave_rt = false; goto skip_alloc_slave_rt; + } s_rt = sdw_slave_rt_alloc(slave, m_rt); if (!s_rt) { diff --git a/drivers/spi/spi-cadence-quadspi.c b/drivers/spi/spi-cadence-quadspi.c index 6ddb2dfc0f005fedf1fbd71b62812c8beb29d62e..32449bef4415ad889931a442dba44df016b20d85 100644 --- a/drivers/spi/spi-cadence-quadspi.c +++ b/drivers/spi/spi-cadence-quadspi.c @@ -1756,8 +1756,11 @@ static int cqspi_probe(struct platform_device *pdev) cqspi->slow_sram = true; if (of_device_is_compatible(pdev->dev.of_node, - "xlnx,versal-ospi-1.0")) - dma_set_mask(&pdev->dev, DMA_BIT_MASK(64)); + "xlnx,versal-ospi-1.0")) { + ret = dma_set_mask(&pdev->dev, DMA_BIT_MASK(64)); + if (ret) + goto probe_reset_failed; + } } ret = devm_request_irq(dev, irq, cqspi_irq_handler, 0, diff --git a/drivers/spi/spi-cadence.c b/drivers/spi/spi-cadence.c index ac85d5562212705a9fd8c0bb5f13c73f024fbf67..26e6633693196ef20ae95b93114627a9d97a2e7b 100644 --- a/drivers/spi/spi-cadence.c +++ b/drivers/spi/spi-cadence.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include @@ -301,49 +302,43 @@ static int cdns_spi_setup_transfer(struct spi_device *spi, } /** - * cdns_spi_fill_tx_fifo - Fills the TX FIFO with as many bytes as possible + * cdns_spi_process_fifo - Fills the TX FIFO, and drain the RX FIFO * @xspi: Pointer to the cdns_spi structure + * @ntx: Number of bytes to pack into the TX FIFO + * @nrx: Number of bytes to drain from the RX FIFO */ -static void cdns_spi_fill_tx_fifo(struct cdns_spi *xspi) +static void cdns_spi_process_fifo(struct cdns_spi *xspi, int ntx, int nrx) { - unsigned long trans_cnt = 0; + ntx = clamp(ntx, 0, xspi->tx_bytes); + nrx = clamp(nrx, 0, xspi->rx_bytes); - while ((trans_cnt < xspi->tx_fifo_depth) && - (xspi->tx_bytes > 0)) { + xspi->tx_bytes -= ntx; + xspi->rx_bytes -= nrx; + while (ntx || nrx) { /* When xspi in busy condition, bytes may send failed, * then spi control did't work thoroughly, add one byte delay */ - if (cdns_spi_read(xspi, CDNS_SPI_ISR) & - CDNS_SPI_IXR_TXFULL) + if (cdns_spi_read(xspi, CDNS_SPI_ISR) & CDNS_SPI_IXR_TXFULL) udelay(10); - if (xspi->txbuf) - cdns_spi_write(xspi, CDNS_SPI_TXD, *xspi->txbuf++); - else - cdns_spi_write(xspi, CDNS_SPI_TXD, 0); + if (ntx) { + if (xspi->txbuf) + cdns_spi_write(xspi, CDNS_SPI_TXD, *xspi->txbuf++); + else + cdns_spi_write(xspi, CDNS_SPI_TXD, 0); - xspi->tx_bytes--; - trans_cnt++; - } -} + ntx--; + } -/** - * cdns_spi_read_rx_fifo - Reads the RX FIFO with as many bytes as possible - * @xspi: Pointer to the cdns_spi structure - * @count: Read byte count - */ -static void cdns_spi_read_rx_fifo(struct cdns_spi *xspi, unsigned long count) -{ - u8 data; - - /* Read out the data from the RX FIFO */ - while (count > 0) { - data = cdns_spi_read(xspi, CDNS_SPI_RXD); - if (xspi->rxbuf) - *xspi->rxbuf++ = data; - xspi->rx_bytes--; - count--; + if (nrx) { + u8 data = cdns_spi_read(xspi, CDNS_SPI_RXD); + + if (xspi->rxbuf) + *xspi->rxbuf++ = data; + + nrx--; + } } } @@ -381,33 +376,22 @@ static irqreturn_t cdns_spi_irq(int irq, void *dev_id) spi_finalize_current_transfer(ctlr); status = IRQ_HANDLED; } else if (intr_status & CDNS_SPI_IXR_TXOW) { - int trans_cnt = cdns_spi_read(xspi, CDNS_SPI_THLD); + int threshold = cdns_spi_read(xspi, CDNS_SPI_THLD); + int trans_cnt = xspi->rx_bytes - xspi->tx_bytes; + + if (threshold > 1) + trans_cnt -= threshold; + /* Set threshold to one if number of pending are * less than half fifo */ if (xspi->tx_bytes < xspi->tx_fifo_depth >> 1) cdns_spi_write(xspi, CDNS_SPI_THLD, 1); - while (trans_cnt) { - cdns_spi_read_rx_fifo(xspi, 1); - - if (xspi->tx_bytes) { - if (xspi->txbuf) - cdns_spi_write(xspi, CDNS_SPI_TXD, - *xspi->txbuf++); - else - cdns_spi_write(xspi, CDNS_SPI_TXD, 0); - xspi->tx_bytes--; - } - trans_cnt--; - } - if (!xspi->tx_bytes) { - /* Fixed delay due to controller limitation with - * RX_NEMPTY incorrect status - * Xilinx AR:65885 contains more details - */ - udelay(10); - cdns_spi_read_rx_fifo(xspi, xspi->rx_bytes); + if (xspi->tx_bytes) { + cdns_spi_process_fifo(xspi, trans_cnt, trans_cnt); + } else { + cdns_spi_process_fifo(xspi, 0, trans_cnt); cdns_spi_write(xspi, CDNS_SPI_IDR, CDNS_SPI_IXR_DEFAULT); spi_finalize_current_transfer(ctlr); @@ -450,16 +434,17 @@ static int cdns_transfer_one(struct spi_controller *ctlr, xspi->tx_bytes = transfer->len; xspi->rx_bytes = transfer->len; - if (!spi_controller_is_slave(ctlr)) + if (!spi_controller_is_slave(ctlr)) { cdns_spi_setup_transfer(spi, transfer); + } else { + /* Set TX empty threshold to half of FIFO depth + * only if TX bytes are more than half FIFO depth. + */ + if (xspi->tx_bytes > xspi->tx_fifo_depth) + cdns_spi_write(xspi, CDNS_SPI_THLD, xspi->tx_fifo_depth >> 1); + } - /* Set TX empty threshold to half of FIFO depth - * only if TX bytes are more than half FIFO depth. - */ - if (xspi->tx_bytes > (xspi->tx_fifo_depth >> 1)) - cdns_spi_write(xspi, CDNS_SPI_THLD, xspi->tx_fifo_depth >> 1); - - cdns_spi_fill_tx_fifo(xspi); + cdns_spi_process_fifo(xspi, xspi->tx_fifo_depth, 0); spi_transfer_delay_exec(transfer); cdns_spi_write(xspi, CDNS_SPI_IER, CDNS_SPI_IXR_DEFAULT); diff --git a/drivers/spi/spi-dw-mmio.c b/drivers/spi/spi-dw-mmio.c index 5e6faa98aa8523d0229454afebcdb6beb50af109..15f5e9cb54ad45717c61e21ab8005b79d53307b0 100644 --- a/drivers/spi/spi-dw-mmio.c +++ b/drivers/spi/spi-dw-mmio.c @@ -264,17 +264,17 @@ static void dw_spi_elba_set_cs(struct spi_device *spi, bool enable) struct regmap *syscon = dwsmmio->priv; u8 cs; - cs = spi->chip_select; + cs = spi_get_chipselect(spi, 0); if (cs < 2) - dw_spi_elba_override_cs(syscon, spi->chip_select, enable); + dw_spi_elba_override_cs(syscon, spi_get_chipselect(spi, 0), enable); /* * The DW SPI controller needs a native CS bit selected to start * the serial engine. */ - spi->chip_select = 0; + spi_set_chipselect(spi, 0, 0); dw_spi_set_cs(spi, enable); - spi->chip_select = cs; + spi_set_chipselect(spi, 0, cs); } static int dw_spi_elba_init(struct platform_device *pdev, diff --git a/drivers/spi/spi-fsl-dspi.c b/drivers/spi/spi-fsl-dspi.c index 4339485d202c05c7fac4f9897ea7d36872851bb8..674cfe05f4118abf53995ce8d7283ea675d99587 100644 --- a/drivers/spi/spi-fsl-dspi.c +++ b/drivers/spi/spi-fsl-dspi.c @@ -1002,7 +1002,9 @@ static int dspi_transfer_one_message(struct spi_controller *ctlr, static int dspi_setup(struct spi_device *spi) { struct fsl_dspi *dspi = spi_controller_get_devdata(spi->controller); + u32 period_ns = DIV_ROUND_UP(NSEC_PER_SEC, spi->max_speed_hz); unsigned char br = 0, pbr = 0, pcssck = 0, cssck = 0; + u32 quarter_period_ns = DIV_ROUND_UP(period_ns, 4); u32 cs_sck_delay = 0, sck_cs_delay = 0; struct fsl_dspi_platform_data *pdata; unsigned char pasc = 0, asc = 0; @@ -1031,6 +1033,19 @@ static int dspi_setup(struct spi_device *spi) sck_cs_delay = pdata->sck_cs_delay; } + /* Since tCSC and tASC apply to continuous transfers too, avoid SCK + * glitches of half a cycle by never allowing tCSC + tASC to go below + * half a SCK period. + */ + if (cs_sck_delay < quarter_period_ns) + cs_sck_delay = quarter_period_ns; + if (sck_cs_delay < quarter_period_ns) + sck_cs_delay = quarter_period_ns; + + dev_dbg(&spi->dev, + "DSPI controller timing params: CS-to-SCK delay %u ns, SCK-to-CS delay %u ns\n", + cs_sck_delay, sck_cs_delay); + clkrate = clk_get_rate(dspi->clk); hz_to_spi_baud(&pbr, &br, spi->max_speed_hz, clkrate); diff --git a/drivers/spi/spi-fsl-lpspi.c b/drivers/spi/spi-fsl-lpspi.c index f2341ab99556649fa308931afc7ede4e8bc7cd52..4b70038ceb6b28e16d1cd2936547074834764f13 100644 --- a/drivers/spi/spi-fsl-lpspi.c +++ b/drivers/spi/spi-fsl-lpspi.c @@ -910,9 +910,14 @@ static int fsl_lpspi_probe(struct platform_device *pdev) ret = fsl_lpspi_dma_init(&pdev->dev, fsl_lpspi, controller); if (ret == -EPROBE_DEFER) goto out_pm_get; - if (ret < 0) dev_err(&pdev->dev, "dma setup error %d, use pio\n", ret); + else + /* + * disable LPSPI module IRQ when enable DMA mode successfully, + * to prevent the unexpected LPSPI module IRQ events. + */ + disable_irq(irq); ret = devm_spi_register_controller(&pdev->dev, controller); if (ret < 0) { diff --git a/drivers/spi/spi-geni-qcom.c b/drivers/spi/spi-geni-qcom.c index ba7be505ec4ef5594ea9e9d4ee6e24c247f10140..a98b781b103ab31b93aed20fc3bf6b167a3c55cb 100644 --- a/drivers/spi/spi-geni-qcom.c +++ b/drivers/spi/spi-geni-qcom.c @@ -294,6 +294,8 @@ static void spi_geni_set_cs(struct spi_device *slv, bool set_flag) mas->cs_flag = set_flag; /* set xfer_mode to FIFO to complete cs_done in isr */ mas->cur_xfer_mode = GENI_SE_FIFO; + geni_se_select_mode(se, mas->cur_xfer_mode); + reinit_completion(&mas->cs_done); if (set_flag) geni_se_setup_m_cmd(se, SPI_CS_ASSERT, 0); diff --git a/drivers/spi/spi-mt65xx.c b/drivers/spi/spi-mt65xx.c index 21c321f437667495c5a3ed94b3d73bf483672ffa..d7432e2219d850685d3ad83a5f8a60b3f7266dde 100644 --- a/drivers/spi/spi-mt65xx.c +++ b/drivers/spi/spi-mt65xx.c @@ -1275,6 +1275,9 @@ static int mtk_spi_remove(struct platform_device *pdev) struct mtk_spi *mdata = spi_master_get_devdata(master); int ret; + if (mdata->use_spimem && !completion_done(&mdata->spimem_done)) + complete(&mdata->spimem_done); + ret = pm_runtime_resume_and_get(&pdev->dev); if (ret < 0) return ret; diff --git a/drivers/spi/spi-qup.c b/drivers/spi/spi-qup.c index 944ef6b42bce6f2c22d4627f0c88716e7c2ac894..00e5e88e72c49d9c143b1ec942c38ebf2f482cae 100644 --- a/drivers/spi/spi-qup.c +++ b/drivers/spi/spi-qup.c @@ -1028,23 +1028,8 @@ static int spi_qup_probe(struct platform_device *pdev) return -ENXIO; } - ret = clk_prepare_enable(cclk); - if (ret) { - dev_err(dev, "cannot enable core clock\n"); - return ret; - } - - ret = clk_prepare_enable(iclk); - if (ret) { - clk_disable_unprepare(cclk); - dev_err(dev, "cannot enable iface clock\n"); - return ret; - } - master = spi_alloc_master(dev, sizeof(struct spi_qup)); if (!master) { - clk_disable_unprepare(cclk); - clk_disable_unprepare(iclk); dev_err(dev, "cannot allocate master\n"); return -ENOMEM; } @@ -1092,6 +1077,19 @@ static int spi_qup_probe(struct platform_device *pdev) spin_lock_init(&controller->lock); init_completion(&controller->done); + ret = clk_prepare_enable(cclk); + if (ret) { + dev_err(dev, "cannot enable core clock\n"); + goto error_dma; + } + + ret = clk_prepare_enable(iclk); + if (ret) { + clk_disable_unprepare(cclk); + dev_err(dev, "cannot enable iface clock\n"); + goto error_dma; + } + iomode = readl_relaxed(base + QUP_IO_M_MODES); size = QUP_IO_M_OUTPUT_BLOCK_SIZE(iomode); @@ -1121,7 +1119,7 @@ static int spi_qup_probe(struct platform_device *pdev) ret = spi_qup_set_state(controller, QUP_STATE_RESET); if (ret) { dev_err(dev, "cannot set RESET state\n"); - goto error_dma; + goto error_clk; } writel_relaxed(0, base + QUP_OPERATIONAL); @@ -1145,7 +1143,7 @@ static int spi_qup_probe(struct platform_device *pdev) ret = devm_request_irq(dev, irq, spi_qup_qup_irq, IRQF_TRIGGER_HIGH, pdev->name, controller); if (ret) - goto error_dma; + goto error_clk; pm_runtime_set_autosuspend_delay(dev, MSEC_PER_SEC); pm_runtime_use_autosuspend(dev); @@ -1160,11 +1158,12 @@ static int spi_qup_probe(struct platform_device *pdev) disable_pm: pm_runtime_disable(&pdev->dev); +error_clk: + clk_disable_unprepare(cclk); + clk_disable_unprepare(iclk); error_dma: spi_qup_release_dma(master); error: - clk_disable_unprepare(cclk); - clk_disable_unprepare(iclk); spi_master_put(master); return ret; } diff --git a/drivers/staging/media/atomisp/i2c/atomisp-ov2680.c b/drivers/staging/media/atomisp/i2c/atomisp-ov2680.c index 63de214916f59ad7ceb27a2ac7761ae4aa6862a6..c079368019e87f5d7d5695ad9c2cff47e922042a 100644 --- a/drivers/staging/media/atomisp/i2c/atomisp-ov2680.c +++ b/drivers/staging/media/atomisp/i2c/atomisp-ov2680.c @@ -373,7 +373,7 @@ static int ov2680_get_fmt(struct v4l2_subdev *sd, static int ov2680_detect(struct i2c_client *client) { struct i2c_adapter *adapter = client->adapter; - u32 high, low; + u32 high = 0, low = 0; int ret; u16 id; u8 revision; @@ -383,7 +383,7 @@ static int ov2680_detect(struct i2c_client *client) ret = ov_read_reg8(client, OV2680_SC_CMMN_CHIP_ID_H, &high); if (ret) { - dev_err(&client->dev, "sensor_id_high = 0x%x\n", high); + dev_err(&client->dev, "sensor_id_high read failed (%d)\n", ret); return -ENODEV; } ret = ov_read_reg8(client, OV2680_SC_CMMN_CHIP_ID_L, &low); diff --git a/drivers/staging/media/imx/imx8mq-mipi-csi2.c b/drivers/staging/media/imx/imx8mq-mipi-csi2.c index 32700cb8bc4d551055b8d6f36618acbe07088042..ca2efcc21efe3063df15e811d2cb9a317bd3b8c1 100644 --- a/drivers/staging/media/imx/imx8mq-mipi-csi2.c +++ b/drivers/staging/media/imx/imx8mq-mipi-csi2.c @@ -354,7 +354,7 @@ static int imx8mq_mipi_csi_start_stream(struct csi_state *state, struct v4l2_subdev_state *sd_state) { int ret; - u32 hs_settle; + u32 hs_settle = 0; ret = imx8mq_mipi_csi_sw_reset(state); if (ret) diff --git a/drivers/staging/octeon/TODO b/drivers/staging/octeon/TODO index 67a0a1f6b922edcdedb2d25f0912c273d80d6097..044e48e3d65f662daf4bfa53adb3254c26ad2de2 100644 --- a/drivers/staging/octeon/TODO +++ b/drivers/staging/octeon/TODO @@ -6,4 +6,3 @@ TODO: - make driver self-contained instead of being split between staging and arch/mips/cavium-octeon. -Contact: Aaro Koskinen diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c index 834cce50f9b00ae5588abaf1c2c657203176ee39..b516c2893420bc142412061aa5db056d6732b685 100644 --- a/drivers/target/iscsi/iscsi_target.c +++ b/drivers/target/iscsi/iscsi_target.c @@ -364,8 +364,6 @@ struct iscsi_np *iscsit_add_np( init_completion(&np->np_restart_comp); INIT_LIST_HEAD(&np->np_list); - timer_setup(&np->np_login_timer, iscsi_handle_login_thread_timeout, 0); - ret = iscsi_target_setup_login_socket(np, sockaddr); if (ret != 0) { kfree(np); diff --git a/drivers/target/iscsi/iscsi_target_login.c b/drivers/target/iscsi/iscsi_target_login.c index 274bdd7845ca936a038dc5e602dc34e3750cbcbd..90b870f234f03cd9033418387482a34bf9d4ede1 100644 --- a/drivers/target/iscsi/iscsi_target_login.c +++ b/drivers/target/iscsi/iscsi_target_login.c @@ -811,59 +811,6 @@ void iscsi_post_login_handler( iscsit_dec_conn_usage_count(conn); } -void iscsi_handle_login_thread_timeout(struct timer_list *t) -{ - struct iscsi_np *np = from_timer(np, t, np_login_timer); - - spin_lock_bh(&np->np_thread_lock); - pr_err("iSCSI Login timeout on Network Portal %pISpc\n", - &np->np_sockaddr); - - if (np->np_login_timer_flags & ISCSI_TF_STOP) { - spin_unlock_bh(&np->np_thread_lock); - return; - } - - if (np->np_thread) - send_sig(SIGINT, np->np_thread, 1); - - np->np_login_timer_flags &= ~ISCSI_TF_RUNNING; - spin_unlock_bh(&np->np_thread_lock); -} - -static void iscsi_start_login_thread_timer(struct iscsi_np *np) -{ - /* - * This used the TA_LOGIN_TIMEOUT constant because at this - * point we do not have access to ISCSI_TPG_ATTRIB(tpg)->login_timeout - */ - spin_lock_bh(&np->np_thread_lock); - np->np_login_timer_flags &= ~ISCSI_TF_STOP; - np->np_login_timer_flags |= ISCSI_TF_RUNNING; - mod_timer(&np->np_login_timer, jiffies + TA_LOGIN_TIMEOUT * HZ); - - pr_debug("Added timeout timer to iSCSI login request for" - " %u seconds.\n", TA_LOGIN_TIMEOUT); - spin_unlock_bh(&np->np_thread_lock); -} - -static void iscsi_stop_login_thread_timer(struct iscsi_np *np) -{ - spin_lock_bh(&np->np_thread_lock); - if (!(np->np_login_timer_flags & ISCSI_TF_RUNNING)) { - spin_unlock_bh(&np->np_thread_lock); - return; - } - np->np_login_timer_flags |= ISCSI_TF_STOP; - spin_unlock_bh(&np->np_thread_lock); - - del_timer_sync(&np->np_login_timer); - - spin_lock_bh(&np->np_thread_lock); - np->np_login_timer_flags &= ~ISCSI_TF_RUNNING; - spin_unlock_bh(&np->np_thread_lock); -} - int iscsit_setup_np( struct iscsi_np *np, struct sockaddr_storage *sockaddr) @@ -1123,10 +1070,13 @@ static struct iscsit_conn *iscsit_alloc_conn(struct iscsi_np *np) spin_lock_init(&conn->nopin_timer_lock); spin_lock_init(&conn->response_queue_lock); spin_lock_init(&conn->state_lock); + spin_lock_init(&conn->login_worker_lock); + spin_lock_init(&conn->login_timer_lock); timer_setup(&conn->nopin_response_timer, iscsit_handle_nopin_response_timeout, 0); timer_setup(&conn->nopin_timer, iscsit_handle_nopin_timeout, 0); + timer_setup(&conn->login_timer, iscsit_login_timeout, 0); if (iscsit_conn_set_transport(conn, np->np_transport) < 0) goto free_conn; @@ -1304,7 +1254,7 @@ static int __iscsi_target_login_thread(struct iscsi_np *np) goto new_sess_out; } - iscsi_start_login_thread_timer(np); + iscsit_start_login_timer(conn, current); pr_debug("Moving to TARG_CONN_STATE_XPT_UP.\n"); conn->conn_state = TARG_CONN_STATE_XPT_UP; @@ -1417,8 +1367,6 @@ static int __iscsi_target_login_thread(struct iscsi_np *np) if (ret < 0) goto new_sess_out; - iscsi_stop_login_thread_timer(np); - if (ret == 1) { tpg_np = conn->tpg_np; @@ -1434,7 +1382,7 @@ static int __iscsi_target_login_thread(struct iscsi_np *np) new_sess_out: new_sess = true; old_sess_out: - iscsi_stop_login_thread_timer(np); + iscsit_stop_login_timer(conn); tpg_np = conn->tpg_np; iscsi_target_login_sess_out(conn, zero_tsih, new_sess); new_sess = false; @@ -1448,7 +1396,6 @@ static int __iscsi_target_login_thread(struct iscsi_np *np) return 1; exit: - iscsi_stop_login_thread_timer(np); spin_lock_bh(&np->np_thread_lock); np->np_thread_state = ISCSI_NP_THREAD_EXIT; spin_unlock_bh(&np->np_thread_lock); diff --git a/drivers/target/iscsi/iscsi_target_nego.c b/drivers/target/iscsi/iscsi_target_nego.c index 24040c118e49d26145eceaaa3b6ce34331366a8f..fa3fb5f4e6bc44db94eb42740dec672c099dd889 100644 --- a/drivers/target/iscsi/iscsi_target_nego.c +++ b/drivers/target/iscsi/iscsi_target_nego.c @@ -535,25 +535,6 @@ static void iscsi_target_login_drop(struct iscsit_conn *conn, struct iscsi_login iscsi_target_login_sess_out(conn, zero_tsih, true); } -struct conn_timeout { - struct timer_list timer; - struct iscsit_conn *conn; -}; - -static void iscsi_target_login_timeout(struct timer_list *t) -{ - struct conn_timeout *timeout = from_timer(timeout, t, timer); - struct iscsit_conn *conn = timeout->conn; - - pr_debug("Entering iscsi_target_login_timeout >>>>>>>>>>>>>>>>>>>\n"); - - if (conn->login_kworker) { - pr_debug("Sending SIGINT to conn->login_kworker %s/%d\n", - conn->login_kworker->comm, conn->login_kworker->pid); - send_sig(SIGINT, conn->login_kworker, 1); - } -} - static void iscsi_target_do_login_rx(struct work_struct *work) { struct iscsit_conn *conn = container_of(work, @@ -562,12 +543,15 @@ static void iscsi_target_do_login_rx(struct work_struct *work) struct iscsi_np *np = login->np; struct iscsi_portal_group *tpg = conn->tpg; struct iscsi_tpg_np *tpg_np = conn->tpg_np; - struct conn_timeout timeout; int rc, zero_tsih = login->zero_tsih; bool state; pr_debug("entering iscsi_target_do_login_rx, conn: %p, %s:%d\n", conn, current->comm, current->pid); + + spin_lock(&conn->login_worker_lock); + set_bit(LOGIN_FLAGS_WORKER_RUNNING, &conn->login_flags); + spin_unlock(&conn->login_worker_lock); /* * If iscsi_target_do_login_rx() has been invoked by ->sk_data_ready() * before initial PDU processing in iscsi_target_start_negotiation() @@ -597,19 +581,16 @@ static void iscsi_target_do_login_rx(struct work_struct *work) goto err; } - conn->login_kworker = current; allow_signal(SIGINT); - - timeout.conn = conn; - timer_setup_on_stack(&timeout.timer, iscsi_target_login_timeout, 0); - mod_timer(&timeout.timer, jiffies + TA_LOGIN_TIMEOUT * HZ); - pr_debug("Starting login timer for %s/%d\n", current->comm, current->pid); + rc = iscsit_set_login_timer_kworker(conn, current); + if (rc < 0) { + /* The login timer has already expired */ + pr_debug("iscsi_target_do_login_rx, login failed\n"); + goto err; + } rc = conn->conn_transport->iscsit_get_login_rx(conn, login); - del_timer_sync(&timeout.timer); - destroy_timer_on_stack(&timeout.timer); flush_signals(current); - conn->login_kworker = NULL; if (rc < 0) goto err; @@ -646,7 +627,17 @@ static void iscsi_target_do_login_rx(struct work_struct *work) if (iscsi_target_sk_check_and_clear(conn, LOGIN_FLAGS_WRITE_ACTIVE)) goto err; + + /* + * Set the login timer thread pointer to NULL to prevent the + * login process from getting stuck if the initiator + * stops sending data. + */ + rc = iscsit_set_login_timer_kworker(conn, NULL); + if (rc < 0) + goto err; } else if (rc == 1) { + iscsit_stop_login_timer(conn); cancel_delayed_work(&conn->login_work); iscsi_target_nego_release(conn); iscsi_post_login_handler(np, conn, zero_tsih); @@ -656,6 +647,7 @@ static void iscsi_target_do_login_rx(struct work_struct *work) err: iscsi_target_restore_sock_callbacks(conn); + iscsit_stop_login_timer(conn); cancel_delayed_work(&conn->login_work); iscsi_target_login_drop(conn, login); iscsit_deaccess_np(np, tpg, tpg_np); @@ -1130,6 +1122,7 @@ int iscsi_target_locate_portal( iscsi_target_set_sock_callbacks(conn); login->np = np; + conn->tpg = NULL; login_req = (struct iscsi_login_req *) login->req; payload_length = ntoh24(login_req->dlength); @@ -1197,7 +1190,6 @@ int iscsi_target_locate_portal( */ sessiontype = strncmp(s_buf, DISCOVERY, 9); if (!sessiontype) { - conn->tpg = iscsit_global->discovery_tpg; if (!login->leading_connection) goto get_target; @@ -1214,9 +1206,11 @@ int iscsi_target_locate_portal( * Serialize access across the discovery struct iscsi_portal_group to * process login attempt. */ + conn->tpg = iscsit_global->discovery_tpg; if (iscsit_access_np(np, conn->tpg) < 0) { iscsit_tx_login_rsp(conn, ISCSI_STATUS_CLS_TARGET_ERR, ISCSI_LOGIN_STATUS_SVC_UNAVAILABLE); + conn->tpg = NULL; ret = -1; goto out; } @@ -1368,14 +1362,30 @@ int iscsi_target_start_negotiation( * and perform connection cleanup now. */ ret = iscsi_target_do_login(conn, login); - if (!ret && iscsi_target_sk_check_and_clear(conn, LOGIN_FLAGS_INITIAL_PDU)) - ret = -1; + if (!ret) { + spin_lock(&conn->login_worker_lock); + + if (iscsi_target_sk_check_and_clear(conn, LOGIN_FLAGS_INITIAL_PDU)) + ret = -1; + else if (!test_bit(LOGIN_FLAGS_WORKER_RUNNING, &conn->login_flags)) { + if (iscsit_set_login_timer_kworker(conn, NULL) < 0) { + /* + * The timeout has expired already. + * Schedule login_work to perform the cleanup. + */ + schedule_delayed_work(&conn->login_work, 0); + } + } + + spin_unlock(&conn->login_worker_lock); + } if (ret < 0) { iscsi_target_restore_sock_callbacks(conn); iscsi_remove_failed_auth_entry(conn); } if (ret != 0) { + iscsit_stop_login_timer(conn); cancel_delayed_work_sync(&conn->login_work); iscsi_target_nego_release(conn); } diff --git a/drivers/target/iscsi/iscsi_target_util.c b/drivers/target/iscsi/iscsi_target_util.c index 26dc8ed3045b6ad9aee85d543bd2ed144e77d2ed..b14835fcb03344780b764bbeeee21bbf53838b86 100644 --- a/drivers/target/iscsi/iscsi_target_util.c +++ b/drivers/target/iscsi/iscsi_target_util.c @@ -1040,6 +1040,57 @@ void iscsit_stop_nopin_timer(struct iscsit_conn *conn) spin_unlock_bh(&conn->nopin_timer_lock); } +void iscsit_login_timeout(struct timer_list *t) +{ + struct iscsit_conn *conn = from_timer(conn, t, login_timer); + struct iscsi_login *login = conn->login; + + pr_debug("Entering iscsi_target_login_timeout >>>>>>>>>>>>>>>>>>>\n"); + + spin_lock_bh(&conn->login_timer_lock); + login->login_failed = 1; + + if (conn->login_kworker) { + pr_debug("Sending SIGINT to conn->login_kworker %s/%d\n", + conn->login_kworker->comm, conn->login_kworker->pid); + send_sig(SIGINT, conn->login_kworker, 1); + } else { + schedule_delayed_work(&conn->login_work, 0); + } + spin_unlock_bh(&conn->login_timer_lock); +} + +void iscsit_start_login_timer(struct iscsit_conn *conn, struct task_struct *kthr) +{ + pr_debug("Login timer started\n"); + + conn->login_kworker = kthr; + mod_timer(&conn->login_timer, jiffies + TA_LOGIN_TIMEOUT * HZ); +} + +int iscsit_set_login_timer_kworker(struct iscsit_conn *conn, struct task_struct *kthr) +{ + struct iscsi_login *login = conn->login; + int ret = 0; + + spin_lock_bh(&conn->login_timer_lock); + if (login->login_failed) { + /* The timer has already expired */ + ret = -1; + } else { + conn->login_kworker = kthr; + } + spin_unlock_bh(&conn->login_timer_lock); + + return ret; +} + +void iscsit_stop_login_timer(struct iscsit_conn *conn) +{ + pr_debug("Login timer stopped\n"); + timer_delete_sync(&conn->login_timer); +} + int iscsit_send_tx_data( struct iscsit_cmd *cmd, struct iscsit_conn *conn, diff --git a/drivers/target/iscsi/iscsi_target_util.h b/drivers/target/iscsi/iscsi_target_util.h index 33ea799a08508f16c59ca931e8182d9afeec9dac..24b8e577575a6ba567b0ca33cc21d3e8e10dd678 100644 --- a/drivers/target/iscsi/iscsi_target_util.h +++ b/drivers/target/iscsi/iscsi_target_util.h @@ -56,6 +56,10 @@ extern void iscsit_handle_nopin_timeout(struct timer_list *t); extern void __iscsit_start_nopin_timer(struct iscsit_conn *); extern void iscsit_start_nopin_timer(struct iscsit_conn *); extern void iscsit_stop_nopin_timer(struct iscsit_conn *); +extern void iscsit_login_timeout(struct timer_list *t); +extern void iscsit_start_login_timer(struct iscsit_conn *, struct task_struct *kthr); +extern void iscsit_stop_login_timer(struct iscsit_conn *); +extern int iscsit_set_login_timer_kworker(struct iscsit_conn *, struct task_struct *kthr); extern int iscsit_send_tx_data(struct iscsit_cmd *, struct iscsit_conn *, int); extern int iscsit_fe_sendpage_sg(struct iscsit_cmd *, struct iscsit_conn *); extern int iscsit_tx_login_rsp(struct iscsit_conn *, u8, u8); diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c index 86adff2a86eddb5baa9e436f58074be4d2478c30..687adc9e086ca942ba85dd591ab3b7bf76cc9192 100644 --- a/drivers/target/target_core_transport.c +++ b/drivers/target/target_core_transport.c @@ -504,6 +504,8 @@ target_setup_session(struct se_portal_group *tpg, free_sess: transport_free_session(sess); + return ERR_PTR(rc); + free_cnt: target_free_cmd_counter(cmd_cnt); return ERR_PTR(rc); diff --git a/drivers/tee/amdtee/amdtee_if.h b/drivers/tee/amdtee/amdtee_if.h index ff48c3e4737503934acd9912f719ac656a9ac11b..e2014e21530acb86d50eb5995e2124f52acb5e7c 100644 --- a/drivers/tee/amdtee/amdtee_if.h +++ b/drivers/tee/amdtee/amdtee_if.h @@ -118,16 +118,18 @@ struct tee_cmd_unmap_shared_mem { /** * struct tee_cmd_load_ta - load Trusted Application (TA) binary into TEE - * @low_addr: [in] bits [31:0] of the physical address of the TA binary - * @hi_addr: [in] bits [63:32] of the physical address of the TA binary - * @size: [in] size of TA binary in bytes - * @ta_handle: [out] return handle of the loaded TA + * @low_addr: [in] bits [31:0] of the physical address of the TA binary + * @hi_addr: [in] bits [63:32] of the physical address of the TA binary + * @size: [in] size of TA binary in bytes + * @ta_handle: [out] return handle of the loaded TA + * @return_origin: [out] origin of return code after TEE processing */ struct tee_cmd_load_ta { u32 low_addr; u32 hi_addr; u32 size; u32 ta_handle; + u32 return_origin; }; /** diff --git a/drivers/tee/amdtee/call.c b/drivers/tee/amdtee/call.c index e8cd9aaa346750fd7e47a2a861813ecb512b51b5..e9b63dcb3194cf4cef47b16a698a12f944200ef1 100644 --- a/drivers/tee/amdtee/call.c +++ b/drivers/tee/amdtee/call.c @@ -423,19 +423,23 @@ int handle_load_ta(void *data, u32 size, struct tee_ioctl_open_session_arg *arg) if (ret) { arg->ret_origin = TEEC_ORIGIN_COMMS; arg->ret = TEEC_ERROR_COMMUNICATION; - } else if (arg->ret == TEEC_SUCCESS) { - ret = get_ta_refcount(load_cmd.ta_handle); - if (!ret) { - arg->ret_origin = TEEC_ORIGIN_COMMS; - arg->ret = TEEC_ERROR_OUT_OF_MEMORY; - - /* Unload the TA on error */ - unload_cmd.ta_handle = load_cmd.ta_handle; - psp_tee_process_cmd(TEE_CMD_ID_UNLOAD_TA, - (void *)&unload_cmd, - sizeof(unload_cmd), &ret); - } else { - set_session_id(load_cmd.ta_handle, 0, &arg->session); + } else { + arg->ret_origin = load_cmd.return_origin; + + if (arg->ret == TEEC_SUCCESS) { + ret = get_ta_refcount(load_cmd.ta_handle); + if (!ret) { + arg->ret_origin = TEEC_ORIGIN_COMMS; + arg->ret = TEEC_ERROR_OUT_OF_MEMORY; + + /* Unload the TA on error */ + unload_cmd.ta_handle = load_cmd.ta_handle; + psp_tee_process_cmd(TEE_CMD_ID_UNLOAD_TA, + (void *)&unload_cmd, + sizeof(unload_cmd), &ret); + } else { + set_session_id(load_cmd.ta_handle, 0, &arg->session); + } } } mutex_unlock(&ta_refcount_mutex); diff --git a/drivers/tee/optee/smc_abi.c b/drivers/tee/optee/smc_abi.c index 49702cb08f4fb3a56f28f3bd9f643268b68cba1f..3861ae06d902f072451bd46d99206ab8bb8e53f5 100644 --- a/drivers/tee/optee/smc_abi.c +++ b/drivers/tee/optee/smc_abi.c @@ -1004,8 +1004,10 @@ static u32 get_async_notif_value(optee_invoke_fn *invoke_fn, bool *value_valid, invoke_fn(OPTEE_SMC_GET_ASYNC_NOTIF_VALUE, 0, 0, 0, 0, 0, 0, 0, &res); - if (res.a0) + if (res.a0) { + *value_valid = false; return 0; + } *value_valid = (res.a2 & OPTEE_SMC_ASYNC_NOTIF_VALUE_VALID); *value_pending = (res.a2 & OPTEE_SMC_ASYNC_NOTIF_VALUE_PENDING); return res.a1; diff --git a/drivers/thermal/intel/int340x_thermal/int3400_thermal.c b/drivers/thermal/intel/int340x_thermal/int3400_thermal.c index 810231b59dcdcc7b1e2943a603b099ac40a3e33b..5e1164226ada21b93d2ef2eb954d17a5d214b4bc 100644 --- a/drivers/thermal/intel/int340x_thermal/int3400_thermal.c +++ b/drivers/thermal/intel/int340x_thermal/int3400_thermal.c @@ -131,7 +131,7 @@ static ssize_t available_uuids_show(struct device *dev, for (i = 0; i < INT3400_THERMAL_MAXIMUM_UUID; i++) { if (priv->uuid_bitmap & (1 << i)) - length += sysfs_emit_at(buf, length, int3400_thermal_uuids[i]); + length += sysfs_emit_at(buf, length, "%s\n", int3400_thermal_uuids[i]); } return length; @@ -149,7 +149,7 @@ static ssize_t current_uuid_show(struct device *dev, for (i = 0; i <= INT3400_THERMAL_CRITICAL; i++) { if (priv->os_uuid_mask & BIT(i)) - length += sysfs_emit_at(buf, length, int3400_thermal_uuids[i]); + length += sysfs_emit_at(buf, length, "%s\n", int3400_thermal_uuids[i]); } if (length) diff --git a/drivers/thunderbolt/dma_test.c b/drivers/thunderbolt/dma_test.c index 3bedecb236e0d15f67e9c09586eddb008276c87a..14bb6dec6c4b023c38ff671da20965c18a82d2b3 100644 --- a/drivers/thunderbolt/dma_test.c +++ b/drivers/thunderbolt/dma_test.c @@ -192,9 +192,9 @@ static int dma_test_start_rings(struct dma_test *dt) } ret = tb_xdomain_enable_paths(dt->xd, dt->tx_hopid, - dt->tx_ring ? dt->tx_ring->hop : 0, + dt->tx_ring ? dt->tx_ring->hop : -1, dt->rx_hopid, - dt->rx_ring ? dt->rx_ring->hop : 0); + dt->rx_ring ? dt->rx_ring->hop : -1); if (ret) { dma_test_free_rings(dt); return ret; @@ -218,9 +218,9 @@ static void dma_test_stop_rings(struct dma_test *dt) tb_ring_stop(dt->tx_ring); ret = tb_xdomain_disable_paths(dt->xd, dt->tx_hopid, - dt->tx_ring ? dt->tx_ring->hop : 0, + dt->tx_ring ? dt->tx_ring->hop : -1, dt->rx_hopid, - dt->rx_ring ? dt->rx_ring->hop : 0); + dt->rx_ring ? dt->rx_ring->hop : -1); if (ret) dev_warn(&dt->svc->dev, "failed to disable DMA paths\n"); diff --git a/drivers/thunderbolt/nhi.c b/drivers/thunderbolt/nhi.c index d76e923fbc6a01fcc8ce93bb998d4231511ffa39..e58beac442958c922a55bc876d9c39c8912cd4c7 100644 --- a/drivers/thunderbolt/nhi.c +++ b/drivers/thunderbolt/nhi.c @@ -54,6 +54,26 @@ static int ring_interrupt_index(const struct tb_ring *ring) return bit; } +static void nhi_mask_interrupt(struct tb_nhi *nhi, int mask, int ring) +{ + if (nhi->quirks & QUIRK_AUTO_CLEAR_INT) { + u32 val; + + val = ioread32(nhi->iobase + REG_RING_INTERRUPT_BASE + ring); + iowrite32(val & ~mask, nhi->iobase + REG_RING_INTERRUPT_BASE + ring); + } else { + iowrite32(mask, nhi->iobase + REG_RING_INTERRUPT_MASK_CLEAR_BASE + ring); + } +} + +static void nhi_clear_interrupt(struct tb_nhi *nhi, int ring) +{ + if (nhi->quirks & QUIRK_AUTO_CLEAR_INT) + ioread32(nhi->iobase + REG_RING_NOTIFY_BASE + ring); + else + iowrite32(~0, nhi->iobase + REG_RING_INT_CLEAR + ring); +} + /* * ring_interrupt_active() - activate/deactivate interrupts for a single ring * @@ -61,8 +81,8 @@ static int ring_interrupt_index(const struct tb_ring *ring) */ static void ring_interrupt_active(struct tb_ring *ring, bool active) { - int reg = REG_RING_INTERRUPT_BASE + - ring_interrupt_index(ring) / 32 * 4; + int index = ring_interrupt_index(ring) / 32 * 4; + int reg = REG_RING_INTERRUPT_BASE + index; int interrupt_bit = ring_interrupt_index(ring) & 31; int mask = 1 << interrupt_bit; u32 old, new; @@ -123,7 +143,11 @@ static void ring_interrupt_active(struct tb_ring *ring, bool active) "interrupt for %s %d is already %s\n", RING_TYPE(ring), ring->hop, active ? "enabled" : "disabled"); - iowrite32(new, ring->nhi->iobase + reg); + + if (active) + iowrite32(new, ring->nhi->iobase + reg); + else + nhi_mask_interrupt(ring->nhi, mask, index); } /* @@ -136,11 +160,11 @@ static void nhi_disable_interrupts(struct tb_nhi *nhi) int i = 0; /* disable interrupts */ for (i = 0; i < RING_INTERRUPT_REG_COUNT(nhi); i++) - iowrite32(0, nhi->iobase + REG_RING_INTERRUPT_BASE + 4 * i); + nhi_mask_interrupt(nhi, ~0, 4 * i); /* clear interrupt status bits */ for (i = 0; i < RING_NOTIFY_REG_COUNT(nhi); i++) - ioread32(nhi->iobase + REG_RING_NOTIFY_BASE + 4 * i); + nhi_clear_interrupt(nhi, 4 * i); } /* ring helper methods */ diff --git a/drivers/thunderbolt/nhi_regs.h b/drivers/thunderbolt/nhi_regs.h index faef165a919ccda93a00fea018151210f68aec1b..6ba2958154770f9d2f2e11251b754c818baf360e 100644 --- a/drivers/thunderbolt/nhi_regs.h +++ b/drivers/thunderbolt/nhi_regs.h @@ -93,6 +93,8 @@ struct ring_desc { #define REG_RING_INTERRUPT_BASE 0x38200 #define RING_INTERRUPT_REG_COUNT(nhi) ((31 + 2 * nhi->hop_count) / 32) +#define REG_RING_INTERRUPT_MASK_CLEAR_BASE 0x38208 + #define REG_INT_THROTTLING_RATE 0x38c00 /* Interrupt Vector Allocation */ diff --git a/drivers/thunderbolt/tb.c b/drivers/thunderbolt/tb.c index 7bfbc9ca9ba4fe3bf59d4b635ef74137e2b76739..c1af712ca728827ac3b11ea9d101989fcbab4d95 100644 --- a/drivers/thunderbolt/tb.c +++ b/drivers/thunderbolt/tb.c @@ -737,6 +737,7 @@ static void tb_scan_port(struct tb_port *port) { struct tb_cm *tcm = tb_priv(port->sw->tb); struct tb_port *upstream_port; + bool discovery = false; struct tb_switch *sw; int ret; @@ -804,8 +805,10 @@ static void tb_scan_port(struct tb_port *port) * tunnels and know which switches were authorized already by * the boot firmware. */ - if (!tcm->hotplug_active) + if (!tcm->hotplug_active) { dev_set_uevent_suppress(&sw->dev, true); + discovery = true; + } /* * At the moment Thunderbolt 2 and beyond (devices with LC) we @@ -835,10 +838,14 @@ static void tb_scan_port(struct tb_port *port) * CL0s and CL1 are enabled and supported together. * Silently ignore CLx enabling in case CLx is not supported. */ - ret = tb_switch_enable_clx(sw, TB_CL1); - if (ret && ret != -EOPNOTSUPP) - tb_sw_warn(sw, "failed to enable %s on upstream port\n", - tb_switch_clx_name(TB_CL1)); + if (discovery) { + tb_sw_dbg(sw, "discovery, not touching CL states\n"); + } else { + ret = tb_switch_enable_clx(sw, TB_CL1); + if (ret && ret != -EOPNOTSUPP) + tb_sw_warn(sw, "failed to enable %s on upstream port\n", + tb_switch_clx_name(TB_CL1)); + } if (tb_switch_is_clx_enabled(sw, TB_CL1)) /* diff --git a/drivers/thunderbolt/tunnel.c b/drivers/thunderbolt/tunnel.c index 9099ae73e78f39417e3d67b1706722f610450f6e..4f222673d651d945dff7ea01565d1500b84fa4ed 100644 --- a/drivers/thunderbolt/tunnel.c +++ b/drivers/thunderbolt/tunnel.c @@ -526,7 +526,7 @@ static int tb_dp_xchg_caps(struct tb_tunnel *tunnel) * Perform connection manager handshake between IN and OUT ports * before capabilities exchange can take place. */ - ret = tb_dp_cm_handshake(in, out, 1500); + ret = tb_dp_cm_handshake(in, out, 3000); if (ret) return ret; diff --git a/drivers/tty/serial/8250/8250_bcm7271.c b/drivers/tty/serial/8250/8250_bcm7271.c index f801b1f5b46c0cd98df6ed19e9c76d7342765585..af0e1c07018795d406bab65eea73181f79920c57 100644 --- a/drivers/tty/serial/8250/8250_bcm7271.c +++ b/drivers/tty/serial/8250/8250_bcm7271.c @@ -1012,7 +1012,7 @@ static int brcmuart_probe(struct platform_device *pdev) of_property_read_u32(np, "clock-frequency", &clk_rate); /* See if a Baud clock has been specified */ - baud_mux_clk = of_clk_get_by_name(np, "sw_baud"); + baud_mux_clk = devm_clk_get(dev, "sw_baud"); if (IS_ERR(baud_mux_clk)) { if (PTR_ERR(baud_mux_clk) == -EPROBE_DEFER) { ret = -EPROBE_DEFER; @@ -1032,7 +1032,7 @@ static int brcmuart_probe(struct platform_device *pdev) if (clk_rate == 0) { dev_err(dev, "clock-frequency or clk not defined\n"); ret = -EINVAL; - goto release_dma; + goto err_clk_disable; } dev_dbg(dev, "DMA is %senabled\n", priv->dma_enabled ? "" : "not "); @@ -1119,6 +1119,8 @@ static int brcmuart_probe(struct platform_device *pdev) serial8250_unregister_port(priv->line); err: brcmuart_free_bufs(dev, priv); +err_clk_disable: + clk_disable_unprepare(baud_mux_clk); release_dma: if (priv->dma_enabled) brcmuart_arbitration(priv, 0); @@ -1133,6 +1135,7 @@ static int brcmuart_remove(struct platform_device *pdev) hrtimer_cancel(&priv->hrt); serial8250_unregister_port(priv->line); brcmuart_free_bufs(&pdev->dev, priv); + clk_disable_unprepare(priv->baud_mux_clk); if (priv->dma_enabled) brcmuart_arbitration(priv, 0); return 0; diff --git a/drivers/tty/serial/8250/8250_exar.c b/drivers/tty/serial/8250/8250_exar.c index 64770c62bbec5436d332821ffaf57074276b2422..b406cba10b0eb4b6bb97aacb18d938000efb1903 100644 --- a/drivers/tty/serial/8250/8250_exar.c +++ b/drivers/tty/serial/8250/8250_exar.c @@ -40,9 +40,13 @@ #define PCI_DEVICE_ID_COMMTECH_4224PCIE 0x0020 #define PCI_DEVICE_ID_COMMTECH_4228PCIE 0x0021 #define PCI_DEVICE_ID_COMMTECH_4222PCIE 0x0022 + #define PCI_DEVICE_ID_EXAR_XR17V4358 0x4358 #define PCI_DEVICE_ID_EXAR_XR17V8358 0x8358 +#define PCI_SUBDEVICE_ID_USR_2980 0x0128 +#define PCI_SUBDEVICE_ID_USR_2981 0x0129 + #define PCI_DEVICE_ID_SEALEVEL_710xC 0x1001 #define PCI_DEVICE_ID_SEALEVEL_720xC 0x1002 #define PCI_DEVICE_ID_SEALEVEL_740xC 0x1004 @@ -829,6 +833,15 @@ static const struct exar8250_board pbn_exar_XR17V8358 = { (kernel_ulong_t)&bd \ } +#define USR_DEVICE(devid, sdevid, bd) { \ + PCI_DEVICE_SUB( \ + PCI_VENDOR_ID_USR, \ + PCI_DEVICE_ID_EXAR_##devid, \ + PCI_VENDOR_ID_EXAR, \ + PCI_SUBDEVICE_ID_USR_##sdevid), 0, 0, \ + (kernel_ulong_t)&bd \ + } + static const struct pci_device_id exar_pci_tbl[] = { EXAR_DEVICE(ACCESSIO, COM_2S, pbn_exar_XR17C15x), EXAR_DEVICE(ACCESSIO, COM_4S, pbn_exar_XR17C15x), @@ -853,6 +866,10 @@ static const struct pci_device_id exar_pci_tbl[] = { IBM_DEVICE(XR17C152, SATURN_SERIAL_ONE_PORT, pbn_exar_ibm_saturn), + /* USRobotics USR298x-OEM PCI Modems */ + USR_DEVICE(XR17C152, 2980, pbn_exar_XR17C15x), + USR_DEVICE(XR17C152, 2981, pbn_exar_XR17C15x), + /* Exar Corp. XR17C15[248] Dual/Quad/Octal UART */ EXAR_DEVICE(EXAR, XR17C152, pbn_exar_XR17C15x), EXAR_DEVICE(EXAR, XR17C154, pbn_exar_XR17C15x), diff --git a/drivers/tty/serial/8250/8250_pci.c b/drivers/tty/serial/8250/8250_pci.c index c55be6fda0cab1fb9cedc6014fb8b90e5094a492..e80c4f6551a1c26c701729d2eb6d22623016c3d4 100644 --- a/drivers/tty/serial/8250/8250_pci.c +++ b/drivers/tty/serial/8250/8250_pci.c @@ -1920,6 +1920,8 @@ pci_moxa_setup(struct serial_private *priv, #define PCI_SUBDEVICE_ID_SIIG_DUAL_30 0x2530 #define PCI_VENDOR_ID_ADVANTECH 0x13fe #define PCI_DEVICE_ID_INTEL_CE4100_UART 0x2e66 +#define PCI_DEVICE_ID_ADVANTECH_PCI1600 0x1600 +#define PCI_DEVICE_ID_ADVANTECH_PCI1600_1611 0x1611 #define PCI_DEVICE_ID_ADVANTECH_PCI3620 0x3620 #define PCI_DEVICE_ID_ADVANTECH_PCI3618 0x3618 #define PCI_DEVICE_ID_ADVANTECH_PCIf618 0xf618 @@ -4085,6 +4087,9 @@ static SIMPLE_DEV_PM_OPS(pciserial_pm_ops, pciserial_suspend_one, pciserial_resume_one); static const struct pci_device_id serial_pci_tbl[] = { + { PCI_VENDOR_ID_ADVANTECH, PCI_DEVICE_ID_ADVANTECH_PCI1600, + PCI_DEVICE_ID_ADVANTECH_PCI1600_1611, PCI_ANY_ID, 0, 0, + pbn_b0_4_921600 }, /* Advantech use PCI_DEVICE_ID_ADVANTECH_PCI3620 (0x3620) as 'PCI_SUBVENDOR_ID' */ { PCI_VENDOR_ID_ADVANTECH, PCI_DEVICE_ID_ADVANTECH_PCI3620, PCI_DEVICE_ID_ADVANTECH_PCI3620, 0x0001, 0, 0, diff --git a/drivers/tty/serial/8250/8250_port.c b/drivers/tty/serial/8250/8250_port.c index fe8d79c4ae95e5c0cb944c6906f9b286832f46e6..c153ba3a018a255bd4304099d951a017f3482502 100644 --- a/drivers/tty/serial/8250/8250_port.c +++ b/drivers/tty/serial/8250/8250_port.c @@ -669,6 +669,7 @@ EXPORT_SYMBOL_GPL(serial8250_em485_supported); /** * serial8250_em485_config() - generic ->rs485_config() callback * @port: uart port + * @termios: termios structure * @rs485: rs485 settings * * Generic callback usable by 8250 uart drivers to activate rs485 settings diff --git a/drivers/tty/serial/8250/8250_tegra.c b/drivers/tty/serial/8250/8250_tegra.c index 2509e7f74ccf2bcd7a366c90dee0724a9b38ef06..89956bbf34d9e4d4ad6b921db7a13b039629741a 100644 --- a/drivers/tty/serial/8250/8250_tegra.c +++ b/drivers/tty/serial/8250/8250_tegra.c @@ -113,13 +113,15 @@ static int tegra_uart_probe(struct platform_device *pdev) ret = serial8250_register_8250_port(&port8250); if (ret < 0) - goto err_clkdisable; + goto err_ctrl_assert; platform_set_drvdata(pdev, uart); uart->line = ret; return 0; +err_ctrl_assert: + reset_control_assert(uart->rst); err_clkdisable: clk_disable_unprepare(uart->clk); diff --git a/drivers/tty/serial/Kconfig b/drivers/tty/serial/Kconfig index 398e5aac2e77a326686310d4b3c2355502765ce2..3e3fb377d90d4c9f15ad55055025d4d6d8883c25 100644 --- a/drivers/tty/serial/Kconfig +++ b/drivers/tty/serial/Kconfig @@ -762,7 +762,7 @@ config SERIAL_PMACZILOG_CONSOLE config SERIAL_CPM tristate "CPM SCC/SMC serial port support" - depends on CPM2 || CPM1 || (PPC32 && COMPILE_TEST) + depends on CPM2 || CPM1 select SERIAL_CORE help This driver supports the SCC and SMC serial ports on Motorola diff --git a/drivers/tty/serial/arc_uart.c b/drivers/tty/serial/arc_uart.c index 59e25f2b6632215cbb06a60a4c17e57402abc53e..4b2512eef577b431748cf539609cb4e72678633c 100644 --- a/drivers/tty/serial/arc_uart.c +++ b/drivers/tty/serial/arc_uart.c @@ -606,10 +606,11 @@ static int arc_serial_probe(struct platform_device *pdev) } uart->baud = val; - port->membase = of_iomap(np, 0); - if (!port->membase) + port->membase = devm_platform_ioremap_resource(pdev, 0); + if (IS_ERR(port->membase)) { /* No point of dev_err since UART itself is hosed here */ - return -ENXIO; + return PTR_ERR(port->membase); + } port->irq = irq_of_parse_and_map(np, 0); diff --git a/drivers/tty/serial/cpm_uart/cpm_uart.h b/drivers/tty/serial/cpm_uart/cpm_uart.h index 0577618e78c041af601b2f238f86b67de4ab2f1a..46c03ed71c31bb346e67e48cd998500f903b53fa 100644 --- a/drivers/tty/serial/cpm_uart/cpm_uart.h +++ b/drivers/tty/serial/cpm_uart/cpm_uart.h @@ -19,8 +19,6 @@ struct gpio_desc; #include "cpm_uart_cpm2.h" #elif defined(CONFIG_CPM1) #include "cpm_uart_cpm1.h" -#elif defined(CONFIG_COMPILE_TEST) -#include "cpm_uart_cpm2.h" #endif #define SERIAL_CPM_MAJOR 204 diff --git a/drivers/tty/serial/fsl_lpuart.c b/drivers/tty/serial/fsl_lpuart.c index c91916e13648302545d8e60727d8615fa94ca436..7fd30fcc10c6238c2f3961ddc215a39a8cd61082 100644 --- a/drivers/tty/serial/fsl_lpuart.c +++ b/drivers/tty/serial/fsl_lpuart.c @@ -310,7 +310,7 @@ static const struct lpuart_soc_data ls1021a_data = { static const struct lpuart_soc_data ls1028a_data = { .devtype = LS1028A_LPUART, .iotype = UPIO_MEM32, - .rx_watermark = 1, + .rx_watermark = 0, }; static struct lpuart_soc_data imx7ulp_data = { @@ -1495,34 +1495,36 @@ static void lpuart_break_ctl(struct uart_port *port, int break_state) static void lpuart32_break_ctl(struct uart_port *port, int break_state) { - unsigned long temp, modem; - struct tty_struct *tty; - unsigned int cflag = 0; - - tty = tty_port_tty_get(&port->state->port); - if (tty) { - cflag = tty->termios.c_cflag; - tty_kref_put(tty); - } + unsigned long temp; - temp = lpuart32_read(port, UARTCTRL) & ~UARTCTRL_SBK; - modem = lpuart32_read(port, UARTMODIR); + temp = lpuart32_read(port, UARTCTRL); + /* + * LPUART IP now has two known bugs, one is CTS has higher priority than the + * break signal, which causes the break signal sending through UARTCTRL_SBK + * may impacted by the CTS input if the HW flow control is enabled. It + * exists on all platforms we support in this driver. + * Another bug is i.MX8QM LPUART may have an additional break character + * being sent after SBK was cleared. + * To avoid above two bugs, we use Transmit Data Inversion function to send + * the break signal instead of UARTCTRL_SBK. + */ if (break_state != 0) { - temp |= UARTCTRL_SBK; /* - * LPUART CTS has higher priority than SBK, need to disable CTS before - * asserting SBK to avoid any interference if flow control is enabled. + * Disable the transmitter to prevent any data from being sent out + * during break, then invert the TX line to send break. */ - if (cflag & CRTSCTS && modem & UARTMODIR_TXCTSE) - lpuart32_write(port, modem & ~UARTMODIR_TXCTSE, UARTMODIR); + temp &= ~UARTCTRL_TE; + lpuart32_write(port, temp, UARTCTRL); + temp |= UARTCTRL_TXINV; + lpuart32_write(port, temp, UARTCTRL); } else { - /* Re-enable the CTS when break off. */ - if (cflag & CRTSCTS && !(modem & UARTMODIR_TXCTSE)) - lpuart32_write(port, modem | UARTMODIR_TXCTSE, UARTMODIR); + /* Disable the TXINV to turn off break and re-enable transmitter. */ + temp &= ~UARTCTRL_TXINV; + lpuart32_write(port, temp, UARTCTRL); + temp |= UARTCTRL_TE; + lpuart32_write(port, temp, UARTCTRL); } - - lpuart32_write(port, temp, UARTCTRL); } static void lpuart_setup_watermark(struct lpuart_port *sport) diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c index a58e9277dfad8752b10af8ca5bd8ad7383984531..f1387f1024dbb336b2304966d990b563ba42f0db 100644 --- a/drivers/tty/serial/lantiq.c +++ b/drivers/tty/serial/lantiq.c @@ -250,6 +250,7 @@ lqasc_err_int(int irq, void *_port) struct ltq_uart_port *ltq_port = to_ltq_uart_port(port); spin_lock_irqsave(<q_port->lock, flags); + __raw_writel(ASC_IRNCR_EIR, port->membase + LTQ_ASC_IRNCR); /* clear any pending interrupts */ asc_update_bits(0, ASCWHBSTATE_CLRPE | ASCWHBSTATE_CLRFE | ASCWHBSTATE_CLRROE, port->membase + LTQ_ASC_WHBSTATE); diff --git a/drivers/tty/serial/qcom_geni_serial.c b/drivers/tty/serial/qcom_geni_serial.c index 08dc3e2a729c76af3cb1ccb2658831c83873b45a..8582479f0211a082cfe1209dcee1aacc3cdc1459 100644 --- a/drivers/tty/serial/qcom_geni_serial.c +++ b/drivers/tty/serial/qcom_geni_serial.c @@ -1664,19 +1664,18 @@ static int qcom_geni_serial_probe(struct platform_device *pdev) uport->private_data = &port->private_data; platform_set_drvdata(pdev, port); - ret = uart_add_one_port(drv, uport); - if (ret) - return ret; - irq_set_status_flags(uport->irq, IRQ_NOAUTOEN); ret = devm_request_irq(uport->dev, uport->irq, qcom_geni_serial_isr, IRQF_TRIGGER_HIGH, port->name, uport); if (ret) { dev_err(uport->dev, "Failed to get IRQ ret %d\n", ret); - uart_remove_one_port(drv, uport); return ret; } + ret = uart_add_one_port(drv, uport); + if (ret) + return ret; + /* * Set pm_runtime status as ACTIVE so that wakeup_irq gets * enabled/disabled from dev_pm_arm_wake_irq during system diff --git a/drivers/tty/vt/vc_screen.c b/drivers/tty/vt/vc_screen.c index 498ba9c0ee9374f971854009ce39f64dfef345b9..829c4be66f3b036e1941a7e6fab286f90d9e8725 100644 --- a/drivers/tty/vt/vc_screen.c +++ b/drivers/tty/vt/vc_screen.c @@ -656,10 +656,17 @@ vcs_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos) } } - /* The vcs_size might have changed while we slept to grab - * the user buffer, so recheck. + /* The vc might have been freed or vcs_size might have changed + * while we slept to grab the user buffer, so recheck. * Return data written up to now on failure. */ + vc = vcs_vc(inode, &viewed); + if (!vc) { + if (written) + break; + ret = -ENXIO; + goto unlock_out; + } size = vcs_size(vc, attr, false); if (size < 0) { if (written) diff --git a/drivers/ufs/core/ufs-mcq.c b/drivers/ufs/core/ufs-mcq.c index 202ff71e1b582132f474c91379a63b158074a928..51b3c6ae781df4a4ab93f5ac5a61f71e622c27e2 100644 --- a/drivers/ufs/core/ufs-mcq.c +++ b/drivers/ufs/core/ufs-mcq.c @@ -150,7 +150,8 @@ static int ufshcd_mcq_config_nr_queues(struct ufs_hba *hba) u32 hba_maxq, rem, tot_queues; struct Scsi_Host *host = hba->host; - hba_maxq = FIELD_GET(MAX_QUEUE_SUP, hba->mcq_capabilities); + /* maxq is 0 based value */ + hba_maxq = FIELD_GET(MAX_QUEUE_SUP, hba->mcq_capabilities) + 1; tot_queues = UFS_MCQ_NUM_DEV_CMD_QUEUES + read_queues + poll_queues + rw_queues; @@ -265,7 +266,7 @@ static int ufshcd_mcq_get_tag(struct ufs_hba *hba, addr = (le64_to_cpu(cqe->command_desc_base_addr) & CQE_UCD_BA) - hba->ucdl_dma_addr; - return div_u64(addr, sizeof(struct utp_transfer_cmd_desc)); + return div_u64(addr, ufshcd_get_ucd_size(hba)); } static void ufshcd_mcq_process_cqe(struct ufs_hba *hba, diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c index 45fd374fe56c13e6b834ddee59ceb092c0c94701..e7e79f515e14182c627069ea231fd8c2a0f0baaf 100644 --- a/drivers/ufs/core/ufshcd.c +++ b/drivers/ufs/core/ufshcd.c @@ -2849,10 +2849,10 @@ static void ufshcd_map_queues(struct Scsi_Host *shost) static void ufshcd_init_lrb(struct ufs_hba *hba, struct ufshcd_lrb *lrb, int i) { struct utp_transfer_cmd_desc *cmd_descp = (void *)hba->ucdl_base_addr + - i * sizeof_utp_transfer_cmd_desc(hba); + i * ufshcd_get_ucd_size(hba); struct utp_transfer_req_desc *utrdlp = hba->utrdl_base_addr; dma_addr_t cmd_desc_element_addr = hba->ucdl_dma_addr + - i * sizeof_utp_transfer_cmd_desc(hba); + i * ufshcd_get_ucd_size(hba); u16 response_offset = offsetof(struct utp_transfer_cmd_desc, response_upiu); u16 prdt_offset = offsetof(struct utp_transfer_cmd_desc, prd_table); @@ -3761,7 +3761,7 @@ static int ufshcd_memory_alloc(struct ufs_hba *hba) size_t utmrdl_size, utrdl_size, ucdl_size; /* Allocate memory for UTP command descriptors */ - ucdl_size = sizeof_utp_transfer_cmd_desc(hba) * hba->nutrs; + ucdl_size = ufshcd_get_ucd_size(hba) * hba->nutrs; hba->ucdl_base_addr = dmam_alloc_coherent(hba->dev, ucdl_size, &hba->ucdl_dma_addr, @@ -3861,7 +3861,7 @@ static void ufshcd_host_memory_configure(struct ufs_hba *hba) prdt_offset = offsetof(struct utp_transfer_cmd_desc, prd_table); - cmd_desc_size = sizeof_utp_transfer_cmd_desc(hba); + cmd_desc_size = ufshcd_get_ucd_size(hba); cmd_desc_dma_addr = hba->ucdl_dma_addr; for (i = 0; i < hba->nutrs; i++) { @@ -8452,7 +8452,7 @@ static void ufshcd_release_sdb_queue(struct ufs_hba *hba, int nutrs) { size_t ucdl_size, utrdl_size; - ucdl_size = sizeof(struct utp_transfer_cmd_desc) * nutrs; + ucdl_size = ufshcd_get_ucd_size(hba) * nutrs; dmam_free_coherent(hba->dev, ucdl_size, hba->ucdl_base_addr, hba->ucdl_dma_addr); diff --git a/drivers/usb/cdns3/cdns3-gadget.c b/drivers/usb/cdns3/cdns3-gadget.c index ccfaebca6faa7512ac92ad0beeea239f467bbb52..1dcadef933e3a6fb3e163ec5061a8f391960baa5 100644 --- a/drivers/usb/cdns3/cdns3-gadget.c +++ b/drivers/usb/cdns3/cdns3-gadget.c @@ -2097,6 +2097,19 @@ int cdns3_ep_config(struct cdns3_endpoint *priv_ep, bool enable) else priv_ep->trb_burst_size = 16; + /* + * In versions preceding DEV_VER_V2, for example, iMX8QM, there exit the bugs + * in the DMA. These bugs occur when the trb_burst_size exceeds 16 and the + * address is not aligned to 128 Bytes (which is a product of the 64-bit AXI + * and AXI maximum burst length of 16 or 0xF+1, dma_axi_ctrl0[3:0]). This + * results in data corruption when it crosses the 4K border. The corruption + * specifically occurs from the position (4K - (address & 0x7F)) to 4K. + * + * So force trb_burst_size to 16 at such platform. + */ + if (priv_dev->dev_ver < DEV_VER_V2) + priv_ep->trb_burst_size = 16; + mult = min_t(u8, mult, EP_CFG_MULT_MAX); buffering = min_t(u8, buffering, EP_CFG_BUFFERING_MAX); maxburst = min_t(u8, maxburst, EP_CFG_MAXBURST_MAX); diff --git a/drivers/usb/class/usbtmc.c b/drivers/usb/class/usbtmc.c index 4bb6d304eb4b21fe3049f44ca000826af2c40ff0..311007b1d9046566b8a155c21ef8188cca13a819 100644 --- a/drivers/usb/class/usbtmc.c +++ b/drivers/usb/class/usbtmc.c @@ -1928,6 +1928,8 @@ static int usbtmc_ioctl_request(struct usbtmc_device_data *data, if (request.req.wLength > USBTMC_BUFSIZE) return -EMSGSIZE; + if (request.req.wLength == 0) /* Length-0 requests are never IN */ + request.req.bRequestType &= ~USB_DIR_IN; is_in = request.req.bRequestType & USB_DIR_IN; diff --git a/drivers/usb/core/buffer.c b/drivers/usb/core/buffer.c index fbb087b728dc987fbcda7d56fbba97b0d6a948a5..268ccbec88f951ab44ad5f410dc19a3a22e3c7a7 100644 --- a/drivers/usb/core/buffer.c +++ b/drivers/usb/core/buffer.c @@ -172,3 +172,44 @@ void hcd_buffer_free( } dma_free_coherent(hcd->self.sysdev, size, addr, dma); } + +void *hcd_buffer_alloc_pages(struct usb_hcd *hcd, + size_t size, gfp_t mem_flags, dma_addr_t *dma) +{ + if (size == 0) + return NULL; + + if (hcd->localmem_pool) + return gen_pool_dma_alloc_align(hcd->localmem_pool, + size, dma, PAGE_SIZE); + + /* some USB hosts just use PIO */ + if (!hcd_uses_dma(hcd)) { + *dma = DMA_MAPPING_ERROR; + return (void *)__get_free_pages(mem_flags, + get_order(size)); + } + + return dma_alloc_coherent(hcd->self.sysdev, + size, dma, mem_flags); +} + +void hcd_buffer_free_pages(struct usb_hcd *hcd, + size_t size, void *addr, dma_addr_t dma) +{ + if (!addr) + return; + + if (hcd->localmem_pool) { + gen_pool_free(hcd->localmem_pool, + (unsigned long)addr, size); + return; + } + + if (!hcd_uses_dma(hcd)) { + free_pages((unsigned long)addr, get_order(size)); + return; + } + + dma_free_coherent(hcd->self.sysdev, size, addr, dma); +} diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c index e501a03d6c700b930b0a152bcb9c87dfe6f5ae7e..fcf68818e999286296e51626e72b25461ab3ea7f 100644 --- a/drivers/usb/core/devio.c +++ b/drivers/usb/core/devio.c @@ -186,6 +186,7 @@ static int connected(struct usb_dev_state *ps) static void dec_usb_memory_use_count(struct usb_memory *usbm, int *count) { struct usb_dev_state *ps = usbm->ps; + struct usb_hcd *hcd = bus_to_hcd(ps->dev->bus); unsigned long flags; spin_lock_irqsave(&ps->lock, flags); @@ -194,8 +195,8 @@ static void dec_usb_memory_use_count(struct usb_memory *usbm, int *count) list_del(&usbm->memlist); spin_unlock_irqrestore(&ps->lock, flags); - usb_free_coherent(ps->dev, usbm->size, usbm->mem, - usbm->dma_handle); + hcd_buffer_free_pages(hcd, usbm->size, + usbm->mem, usbm->dma_handle); usbfs_decrease_memory_usage( usbm->size + sizeof(struct usb_memory)); kfree(usbm); @@ -234,7 +235,7 @@ static int usbdev_mmap(struct file *file, struct vm_area_struct *vma) size_t size = vma->vm_end - vma->vm_start; void *mem; unsigned long flags; - dma_addr_t dma_handle; + dma_addr_t dma_handle = DMA_MAPPING_ERROR; int ret; ret = usbfs_increase_memory_usage(size + sizeof(struct usb_memory)); @@ -247,8 +248,8 @@ static int usbdev_mmap(struct file *file, struct vm_area_struct *vma) goto error_decrease_mem; } - mem = usb_alloc_coherent(ps->dev, size, GFP_USER | __GFP_NOWARN, - &dma_handle); + mem = hcd_buffer_alloc_pages(hcd, + size, GFP_USER | __GFP_NOWARN, &dma_handle); if (!mem) { ret = -ENOMEM; goto error_free_usbm; @@ -264,7 +265,14 @@ static int usbdev_mmap(struct file *file, struct vm_area_struct *vma) usbm->vma_use_count = 1; INIT_LIST_HEAD(&usbm->memlist); - if (hcd->localmem_pool || !hcd_uses_dma(hcd)) { + /* + * In DMA-unavailable cases, hcd_buffer_alloc_pages allocates + * normal pages and assigns DMA_MAPPING_ERROR to dma_handle. Check + * whether we are in such cases, and then use remap_pfn_range (or + * dma_mmap_coherent) to map normal (or DMA) pages into the user + * space, respectively. + */ + if (dma_handle == DMA_MAPPING_ERROR) { if (remap_pfn_range(vma, vma->vm_start, virt_to_phys(usbm->mem) >> PAGE_SHIFT, size, vma->vm_page_prot) < 0) { diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c index 0beaab932e7db2122e05bf52ef84c0c4b538d434..d68958e151a78d70e7f823899ec9152ec6bf568b 100644 --- a/drivers/usb/dwc3/core.c +++ b/drivers/usb/dwc3/core.c @@ -1137,7 +1137,7 @@ static int dwc3_core_init(struct dwc3 *dwc) dwc3_set_incr_burst_type(dwc); - dwc3_phy_power_on(dwc); + ret = dwc3_phy_power_on(dwc); if (ret) goto err_exit_phy; @@ -1929,6 +1929,11 @@ static int dwc3_remove(struct platform_device *pdev) pm_runtime_disable(&pdev->dev); pm_runtime_dont_use_autosuspend(&pdev->dev); pm_runtime_put_noidle(&pdev->dev); + /* + * HACK: Clear the driver data, which is currently accessed by parent + * glue drivers, before allowing the parent to suspend. + */ + platform_set_drvdata(pdev, NULL); pm_runtime_set_suspended(&pdev->dev); dwc3_free_event_buffers(dwc); diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h index d56457c02996b669c60a096533a506c324ad4a98..1f043c31a0969ffc427f62b56974162c0cf0651b 100644 --- a/drivers/usb/dwc3/core.h +++ b/drivers/usb/dwc3/core.h @@ -1116,6 +1116,7 @@ struct dwc3_scratchpad_array { * @dis_metastability_quirk: set to disable metastability quirk. * @dis_split_quirk: set to disable split boundary. * @wakeup_configured: set if the device is configured for remote wakeup. + * @suspended: set to track suspend event due to U3/L2. * @imod_interval: set the interrupt moderation interval in 250ns * increments or 0 to disable. * @max_cfg_eps: current max number of IN eps used across all USB configs. @@ -1332,6 +1333,7 @@ struct dwc3 { unsigned dis_split_quirk:1; unsigned async_callbacks:1; unsigned wakeup_configured:1; + unsigned suspended:1; u16 imod_interval; diff --git a/drivers/usb/dwc3/debugfs.c b/drivers/usb/dwc3/debugfs.c index e4a2560b9dc0c670d9a5c775a73d17c340b5a992..ebf03468fac4a9f89cefa093c6ba346f0b1035dd 100644 --- a/drivers/usb/dwc3/debugfs.c +++ b/drivers/usb/dwc3/debugfs.c @@ -332,6 +332,11 @@ static int dwc3_lsp_show(struct seq_file *s, void *unused) unsigned int current_mode; unsigned long flags; u32 reg; + int ret; + + ret = pm_runtime_resume_and_get(dwc->dev); + if (ret < 0) + return ret; spin_lock_irqsave(&dwc->lock, flags); reg = dwc3_readl(dwc->regs, DWC3_GSTS); @@ -350,6 +355,8 @@ static int dwc3_lsp_show(struct seq_file *s, void *unused) } spin_unlock_irqrestore(&dwc->lock, flags); + pm_runtime_put_sync(dwc->dev); + return 0; } @@ -395,6 +402,11 @@ static int dwc3_mode_show(struct seq_file *s, void *unused) struct dwc3 *dwc = s->private; unsigned long flags; u32 reg; + int ret; + + ret = pm_runtime_resume_and_get(dwc->dev); + if (ret < 0) + return ret; spin_lock_irqsave(&dwc->lock, flags); reg = dwc3_readl(dwc->regs, DWC3_GCTL); @@ -414,6 +426,8 @@ static int dwc3_mode_show(struct seq_file *s, void *unused) seq_printf(s, "UNKNOWN %08x\n", DWC3_GCTL_PRTCAP(reg)); } + pm_runtime_put_sync(dwc->dev); + return 0; } @@ -463,6 +477,11 @@ static int dwc3_testmode_show(struct seq_file *s, void *unused) struct dwc3 *dwc = s->private; unsigned long flags; u32 reg; + int ret; + + ret = pm_runtime_resume_and_get(dwc->dev); + if (ret < 0) + return ret; spin_lock_irqsave(&dwc->lock, flags); reg = dwc3_readl(dwc->regs, DWC3_DCTL); @@ -493,6 +512,8 @@ static int dwc3_testmode_show(struct seq_file *s, void *unused) seq_printf(s, "UNKNOWN %d\n", reg); } + pm_runtime_put_sync(dwc->dev); + return 0; } @@ -509,6 +530,7 @@ static ssize_t dwc3_testmode_write(struct file *file, unsigned long flags; u32 testmode = 0; char buf[32]; + int ret; if (copy_from_user(&buf, ubuf, min_t(size_t, sizeof(buf) - 1, count))) return -EFAULT; @@ -526,10 +548,16 @@ static ssize_t dwc3_testmode_write(struct file *file, else testmode = 0; + ret = pm_runtime_resume_and_get(dwc->dev); + if (ret < 0) + return ret; + spin_lock_irqsave(&dwc->lock, flags); dwc3_gadget_set_test_mode(dwc, testmode); spin_unlock_irqrestore(&dwc->lock, flags); + pm_runtime_put_sync(dwc->dev); + return count; } @@ -548,12 +576,18 @@ static int dwc3_link_state_show(struct seq_file *s, void *unused) enum dwc3_link_state state; u32 reg; u8 speed; + int ret; + + ret = pm_runtime_resume_and_get(dwc->dev); + if (ret < 0) + return ret; spin_lock_irqsave(&dwc->lock, flags); reg = dwc3_readl(dwc->regs, DWC3_GSTS); if (DWC3_GSTS_CURMOD(reg) != DWC3_GSTS_CURMOD_DEVICE) { seq_puts(s, "Not available\n"); spin_unlock_irqrestore(&dwc->lock, flags); + pm_runtime_put_sync(dwc->dev); return 0; } @@ -566,6 +600,8 @@ static int dwc3_link_state_show(struct seq_file *s, void *unused) dwc3_gadget_hs_link_string(state)); spin_unlock_irqrestore(&dwc->lock, flags); + pm_runtime_put_sync(dwc->dev); + return 0; } @@ -584,6 +620,7 @@ static ssize_t dwc3_link_state_write(struct file *file, char buf[32]; u32 reg; u8 speed; + int ret; if (copy_from_user(&buf, ubuf, min_t(size_t, sizeof(buf) - 1, count))) return -EFAULT; @@ -603,10 +640,15 @@ static ssize_t dwc3_link_state_write(struct file *file, else return -EINVAL; + ret = pm_runtime_resume_and_get(dwc->dev); + if (ret < 0) + return ret; + spin_lock_irqsave(&dwc->lock, flags); reg = dwc3_readl(dwc->regs, DWC3_GSTS); if (DWC3_GSTS_CURMOD(reg) != DWC3_GSTS_CURMOD_DEVICE) { spin_unlock_irqrestore(&dwc->lock, flags); + pm_runtime_put_sync(dwc->dev); return -EINVAL; } @@ -616,12 +658,15 @@ static ssize_t dwc3_link_state_write(struct file *file, if (speed < DWC3_DSTS_SUPERSPEED && state != DWC3_LINK_STATE_RECOV) { spin_unlock_irqrestore(&dwc->lock, flags); + pm_runtime_put_sync(dwc->dev); return -EINVAL; } dwc3_gadget_set_link_state(dwc, state); spin_unlock_irqrestore(&dwc->lock, flags); + pm_runtime_put_sync(dwc->dev); + return count; } @@ -645,6 +690,11 @@ static int dwc3_tx_fifo_size_show(struct seq_file *s, void *unused) unsigned long flags; u32 mdwidth; u32 val; + int ret; + + ret = pm_runtime_resume_and_get(dwc->dev); + if (ret < 0) + return ret; spin_lock_irqsave(&dwc->lock, flags); val = dwc3_core_fifo_space(dep, DWC3_TXFIFO); @@ -657,6 +707,8 @@ static int dwc3_tx_fifo_size_show(struct seq_file *s, void *unused) seq_printf(s, "%u\n", val); spin_unlock_irqrestore(&dwc->lock, flags); + pm_runtime_put_sync(dwc->dev); + return 0; } @@ -667,6 +719,11 @@ static int dwc3_rx_fifo_size_show(struct seq_file *s, void *unused) unsigned long flags; u32 mdwidth; u32 val; + int ret; + + ret = pm_runtime_resume_and_get(dwc->dev); + if (ret < 0) + return ret; spin_lock_irqsave(&dwc->lock, flags); val = dwc3_core_fifo_space(dep, DWC3_RXFIFO); @@ -679,6 +736,8 @@ static int dwc3_rx_fifo_size_show(struct seq_file *s, void *unused) seq_printf(s, "%u\n", val); spin_unlock_irqrestore(&dwc->lock, flags); + pm_runtime_put_sync(dwc->dev); + return 0; } @@ -688,12 +747,19 @@ static int dwc3_tx_request_queue_show(struct seq_file *s, void *unused) struct dwc3 *dwc = dep->dwc; unsigned long flags; u32 val; + int ret; + + ret = pm_runtime_resume_and_get(dwc->dev); + if (ret < 0) + return ret; spin_lock_irqsave(&dwc->lock, flags); val = dwc3_core_fifo_space(dep, DWC3_TXREQQ); seq_printf(s, "%u\n", val); spin_unlock_irqrestore(&dwc->lock, flags); + pm_runtime_put_sync(dwc->dev); + return 0; } @@ -703,12 +769,19 @@ static int dwc3_rx_request_queue_show(struct seq_file *s, void *unused) struct dwc3 *dwc = dep->dwc; unsigned long flags; u32 val; + int ret; + + ret = pm_runtime_resume_and_get(dwc->dev); + if (ret < 0) + return ret; spin_lock_irqsave(&dwc->lock, flags); val = dwc3_core_fifo_space(dep, DWC3_RXREQQ); seq_printf(s, "%u\n", val); spin_unlock_irqrestore(&dwc->lock, flags); + pm_runtime_put_sync(dwc->dev); + return 0; } @@ -718,12 +791,19 @@ static int dwc3_rx_info_queue_show(struct seq_file *s, void *unused) struct dwc3 *dwc = dep->dwc; unsigned long flags; u32 val; + int ret; + + ret = pm_runtime_resume_and_get(dwc->dev); + if (ret < 0) + return ret; spin_lock_irqsave(&dwc->lock, flags); val = dwc3_core_fifo_space(dep, DWC3_RXINFOQ); seq_printf(s, "%u\n", val); spin_unlock_irqrestore(&dwc->lock, flags); + pm_runtime_put_sync(dwc->dev); + return 0; } @@ -733,12 +813,19 @@ static int dwc3_descriptor_fetch_queue_show(struct seq_file *s, void *unused) struct dwc3 *dwc = dep->dwc; unsigned long flags; u32 val; + int ret; + + ret = pm_runtime_resume_and_get(dwc->dev); + if (ret < 0) + return ret; spin_lock_irqsave(&dwc->lock, flags); val = dwc3_core_fifo_space(dep, DWC3_DESCFETCHQ); seq_printf(s, "%u\n", val); spin_unlock_irqrestore(&dwc->lock, flags); + pm_runtime_put_sync(dwc->dev); + return 0; } @@ -748,12 +835,19 @@ static int dwc3_event_queue_show(struct seq_file *s, void *unused) struct dwc3 *dwc = dep->dwc; unsigned long flags; u32 val; + int ret; + + ret = pm_runtime_resume_and_get(dwc->dev); + if (ret < 0) + return ret; spin_lock_irqsave(&dwc->lock, flags); val = dwc3_core_fifo_space(dep, DWC3_EVENTQ); seq_printf(s, "%u\n", val); spin_unlock_irqrestore(&dwc->lock, flags); + pm_runtime_put_sync(dwc->dev); + return 0; } @@ -798,6 +892,11 @@ static int dwc3_trb_ring_show(struct seq_file *s, void *unused) struct dwc3 *dwc = dep->dwc; unsigned long flags; int i; + int ret; + + ret = pm_runtime_resume_and_get(dwc->dev); + if (ret < 0) + return ret; spin_lock_irqsave(&dwc->lock, flags); if (dep->number <= 1) { @@ -827,6 +926,8 @@ static int dwc3_trb_ring_show(struct seq_file *s, void *unused) out: spin_unlock_irqrestore(&dwc->lock, flags); + pm_runtime_put_sync(dwc->dev); + return 0; } @@ -839,6 +940,11 @@ static int dwc3_ep_info_register_show(struct seq_file *s, void *unused) u32 lower_32_bits; u32 upper_32_bits; u32 reg; + int ret; + + ret = pm_runtime_resume_and_get(dwc->dev); + if (ret < 0) + return ret; spin_lock_irqsave(&dwc->lock, flags); reg = DWC3_GDBGLSPMUX_EPSELECT(dep->number); @@ -851,6 +957,8 @@ static int dwc3_ep_info_register_show(struct seq_file *s, void *unused) seq_printf(s, "0x%016llx\n", ep_info); spin_unlock_irqrestore(&dwc->lock, flags); + pm_runtime_put_sync(dwc->dev); + return 0; } @@ -910,6 +1018,7 @@ void dwc3_debugfs_init(struct dwc3 *dwc) dwc->regset->regs = dwc3_regs; dwc->regset->nregs = ARRAY_SIZE(dwc3_regs); dwc->regset->base = dwc->regs - DWC3_GLOBALS_REGS_START; + dwc->regset->dev = dwc->dev; root = debugfs_create_dir(dev_name(dwc->dev), usb_debug_root); dwc->debug_root = root; diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c index 959fc925ca7c52316b4df8f23157fd4cedc02257..79b22abf97276fee67f594837d028262e7ff5729 100644 --- a/drivers/usb/dwc3/dwc3-qcom.c +++ b/drivers/usb/dwc3/dwc3-qcom.c @@ -308,7 +308,16 @@ static void dwc3_qcom_interconnect_exit(struct dwc3_qcom *qcom) /* Only usable in contexts where the role can not change. */ static bool dwc3_qcom_is_host(struct dwc3_qcom *qcom) { - struct dwc3 *dwc = platform_get_drvdata(qcom->dwc3); + struct dwc3 *dwc; + + /* + * FIXME: Fix this layering violation. + */ + dwc = platform_get_drvdata(qcom->dwc3); + + /* Core driver may not have probed yet. */ + if (!dwc) + return false; return dwc->xhci; } diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c index c0ca4d12f95d96b189128456d1b7c33455c02723..b78599dd705c28a9bf38943a649d710ec5226709 100644 --- a/drivers/usb/dwc3/gadget.c +++ b/drivers/usb/dwc3/gadget.c @@ -198,6 +198,7 @@ static void dwc3_gadget_del_and_unmap_request(struct dwc3_ep *dep, list_del(&req->list); req->remaining = 0; req->needs_extra_trb = false; + req->num_trbs = 0; if (req->request.status == -EINPROGRESS) req->request.status = status; @@ -2440,6 +2441,7 @@ static int dwc3_gadget_func_wakeup(struct usb_gadget *g, int intf_id) return -EINVAL; } dwc3_resume_gadget(dwc); + dwc->suspended = false; dwc->link_state = DWC3_LINK_STATE_U0; } @@ -2699,6 +2701,21 @@ static int dwc3_gadget_soft_disconnect(struct dwc3 *dwc) return ret; } +static int dwc3_gadget_soft_connect(struct dwc3 *dwc) +{ + /* + * In the Synopsys DWC_usb31 1.90a programming guide section + * 4.1.9, it specifies that for a reconnect after a + * device-initiated disconnect requires a core soft reset + * (DCTL.CSftRst) before enabling the run/stop bit. + */ + dwc3_core_soft_reset(dwc); + + dwc3_event_buffers_setup(dwc); + __dwc3_gadget_start(dwc); + return dwc3_gadget_run_stop(dwc, true); +} + static int dwc3_gadget_pullup(struct usb_gadget *g, int is_on) { struct dwc3 *dwc = gadget_to_dwc(g); @@ -2737,21 +2754,10 @@ static int dwc3_gadget_pullup(struct usb_gadget *g, int is_on) synchronize_irq(dwc->irq_gadget); - if (!is_on) { + if (!is_on) ret = dwc3_gadget_soft_disconnect(dwc); - } else { - /* - * In the Synopsys DWC_usb31 1.90a programming guide section - * 4.1.9, it specifies that for a reconnect after a - * device-initiated disconnect requires a core soft reset - * (DCTL.CSftRst) before enabling the run/stop bit. - */ - dwc3_core_soft_reset(dwc); - - dwc3_event_buffers_setup(dwc); - __dwc3_gadget_start(dwc); - ret = dwc3_gadget_run_stop(dwc, true); - } + else + ret = dwc3_gadget_soft_connect(dwc); pm_runtime_put(dwc->dev); @@ -3938,6 +3944,8 @@ static void dwc3_gadget_disconnect_interrupt(struct dwc3 *dwc) { int reg; + dwc->suspended = false; + dwc3_gadget_set_link_state(dwc, DWC3_LINK_STATE_RX_DET); reg = dwc3_readl(dwc->regs, DWC3_DCTL); @@ -3962,6 +3970,8 @@ static void dwc3_gadget_reset_interrupt(struct dwc3 *dwc) { u32 reg; + dwc->suspended = false; + /* * Ideally, dwc3_reset_gadget() would trigger the function * drivers to stop any active transfers through ep disable. @@ -4180,6 +4190,8 @@ static void dwc3_gadget_conndone_interrupt(struct dwc3 *dwc) static void dwc3_gadget_wakeup_interrupt(struct dwc3 *dwc, unsigned int evtinfo) { + dwc->suspended = false; + /* * TODO take core out of low power mode when that's * implemented. @@ -4277,6 +4289,7 @@ static void dwc3_gadget_linksts_change_interrupt(struct dwc3 *dwc, if (dwc->gadget->wakeup_armed) { dwc3_gadget_enable_linksts_evts(dwc, false); dwc3_resume_gadget(dwc); + dwc->suspended = false; } break; case DWC3_LINK_STATE_U1: @@ -4303,8 +4316,10 @@ static void dwc3_gadget_suspend_interrupt(struct dwc3 *dwc, { enum dwc3_link_state next = evtinfo & DWC3_LINK_STATE_MASK; - if (dwc->link_state != next && next == DWC3_LINK_STATE_U3) + if (!dwc->suspended && next == DWC3_LINK_STATE_U3) { + dwc->suspended = true; dwc3_suspend_gadget(dwc); + } dwc->link_state = next; } @@ -4655,42 +4670,39 @@ void dwc3_gadget_exit(struct dwc3 *dwc) int dwc3_gadget_suspend(struct dwc3 *dwc) { unsigned long flags; + int ret; if (!dwc->gadget_driver) return 0; - dwc3_gadget_run_stop(dwc, false); + ret = dwc3_gadget_soft_disconnect(dwc); + if (ret) + goto err; spin_lock_irqsave(&dwc->lock, flags); dwc3_disconnect_gadget(dwc); - __dwc3_gadget_stop(dwc); spin_unlock_irqrestore(&dwc->lock, flags); return 0; + +err: + /* + * Attempt to reset the controller's state. Likely no + * communication can be established until the host + * performs a port reset. + */ + if (dwc->softconnect) + dwc3_gadget_soft_connect(dwc); + + return ret; } int dwc3_gadget_resume(struct dwc3 *dwc) { - int ret; - if (!dwc->gadget_driver || !dwc->softconnect) return 0; - ret = __dwc3_gadget_start(dwc); - if (ret < 0) - goto err0; - - ret = dwc3_gadget_run_stop(dwc, true); - if (ret < 0) - goto err1; - - return 0; - -err1: - __dwc3_gadget_stop(dwc); - -err0: - return ret; + return dwc3_gadget_soft_connect(dwc); } void dwc3_gadget_process_pending_events(struct dwc3 *dwc) diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c index a13c946e0663c2c41736bc04f0dcfae44895a93b..f41a385a5c421adbdeee11f1f010fb3a885a0b30 100644 --- a/drivers/usb/gadget/function/f_fs.c +++ b/drivers/usb/gadget/function/f_fs.c @@ -3535,6 +3535,7 @@ static void ffs_func_unbind(struct usb_configuration *c, /* Drain any pending AIO completions */ drain_workqueue(ffs->io_completion_wq); + ffs_event_add(ffs, FUNCTIONFS_UNBIND); if (!--opts->refcnt) functionfs_unbind(ffs); @@ -3559,7 +3560,6 @@ static void ffs_func_unbind(struct usb_configuration *c, func->function.ssp_descriptors = NULL; func->interfaces_nums = NULL; - ffs_event_add(ffs, FUNCTIONFS_UNBIND); } static struct usb_function *ffs_alloc(struct usb_function_instance *fi) diff --git a/drivers/usb/gadget/function/u_ether.c b/drivers/usb/gadget/function/u_ether.c index 6956ad8ba8dd86e5c1a3fd6e10bc25adc8a734f2..a366abb456236c7033b6fbd1c7cfb9b4dcc1a7cf 100644 --- a/drivers/usb/gadget/function/u_ether.c +++ b/drivers/usb/gadget/function/u_ether.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include "u_ether.h" @@ -965,6 +966,8 @@ int gether_get_host_addr_cdc(struct net_device *net, char *host_addr, int len) dev = netdev_priv(net); snprintf(host_addr, len, "%pm", dev->host_mac); + string_upper(host_addr, host_addr); + return strlen(host_addr); } EXPORT_SYMBOL_GPL(gether_get_host_addr_cdc); diff --git a/drivers/usb/gadget/udc/amd5536udc_pci.c b/drivers/usb/gadget/udc/amd5536udc_pci.c index c80f9bd51b7501eef578fde52dae2d40dc712ca6..a36913ae31f9ef56f5814e61bc01d6da1769f8a4 100644 --- a/drivers/usb/gadget/udc/amd5536udc_pci.c +++ b/drivers/usb/gadget/udc/amd5536udc_pci.c @@ -170,6 +170,9 @@ static int udc_pci_probe( retval = -ENODEV; goto err_probe; } + + udc = dev; + return 0; err_probe: diff --git a/drivers/usb/gadget/udc/core.c b/drivers/usb/gadget/udc/core.c index 4641153e9706b161889ec3836d5807ec71552309..83fd1de14784f29ca78be8ea9fb9eb4a7cc6ab06 100644 --- a/drivers/usb/gadget/udc/core.c +++ b/drivers/usb/gadget/udc/core.c @@ -37,10 +37,14 @@ static const struct bus_type gadget_bus_type; * @vbus: for udcs who care about vbus status, this value is real vbus status; * for udcs who do not care about vbus status, this value is always true * @started: the UDC's started state. True if the UDC had started. - * @connect_lock: protects udc->vbus, udc->started, gadget->connect, gadget->deactivate related - * functions. usb_gadget_connect_locked, usb_gadget_disconnect_locked, - * usb_udc_connect_control_locked, usb_gadget_udc_start_locked, usb_gadget_udc_stop_locked are - * called with this lock held. + * @allow_connect: Indicates whether UDC is allowed to be pulled up. + * Set/cleared by gadget_(un)bind_driver() after gadget driver is bound or + * unbound. + * @connect_lock: protects udc->started, gadget->connect, + * gadget->allow_connect and gadget->deactivate. The routines + * usb_gadget_connect_locked(), usb_gadget_disconnect_locked(), + * usb_udc_connect_control_locked(), usb_gadget_udc_start_locked() and + * usb_gadget_udc_stop_locked() are called with this lock held. * * This represents the internal data structure which is used by the UDC-class * to hold information about udc driver and gadget together. @@ -52,6 +56,8 @@ struct usb_udc { struct list_head list; bool vbus; bool started; + bool allow_connect; + struct work_struct vbus_work; struct mutex connect_lock; }; @@ -692,7 +698,6 @@ int usb_gadget_vbus_disconnect(struct usb_gadget *gadget) } EXPORT_SYMBOL_GPL(usb_gadget_vbus_disconnect); -/* Internal version of usb_gadget_connect needs to be called with connect_lock held. */ static int usb_gadget_connect_locked(struct usb_gadget *gadget) __must_hold(&gadget->udc->connect_lock) { @@ -703,15 +708,12 @@ static int usb_gadget_connect_locked(struct usb_gadget *gadget) goto out; } - if (gadget->connected) - goto out; - - if (gadget->deactivated || !gadget->udc->started) { + if (gadget->deactivated || !gadget->udc->allow_connect || !gadget->udc->started) { /* - * If gadget is deactivated we only save new state. - * Gadget will be connected automatically after activation. - * - * udc first needs to be started before gadget can be pulled up. + * If the gadget isn't usable (because it is deactivated, + * unbound, or not yet started), we only save the new state. + * The gadget will be connected automatically when it is + * activated/bound/started. */ gadget->connected = true; goto out; @@ -749,7 +751,6 @@ int usb_gadget_connect(struct usb_gadget *gadget) } EXPORT_SYMBOL_GPL(usb_gadget_connect); -/* Internal version of usb_gadget_disconnect needs to be called with connect_lock held. */ static int usb_gadget_disconnect_locked(struct usb_gadget *gadget) __must_hold(&gadget->udc->connect_lock) { @@ -767,8 +768,6 @@ static int usb_gadget_disconnect_locked(struct usb_gadget *gadget) /* * If gadget is deactivated we only save new state. * Gadget will stay disconnected after activation. - * - * udc should have been started before gadget being pulled down. */ gadget->connected = false; goto out; @@ -829,10 +828,10 @@ int usb_gadget_deactivate(struct usb_gadget *gadget) { int ret = 0; + mutex_lock(&gadget->udc->connect_lock); if (gadget->deactivated) - goto out; + goto unlock; - mutex_lock(&gadget->udc->connect_lock); if (gadget->connected) { ret = usb_gadget_disconnect_locked(gadget); if (ret) @@ -848,7 +847,6 @@ int usb_gadget_deactivate(struct usb_gadget *gadget) unlock: mutex_unlock(&gadget->udc->connect_lock); -out: trace_usb_gadget_deactivate(gadget, ret); return ret; @@ -868,10 +866,10 @@ int usb_gadget_activate(struct usb_gadget *gadget) { int ret = 0; + mutex_lock(&gadget->udc->connect_lock); if (!gadget->deactivated) - goto out; + goto unlock; - mutex_lock(&gadget->udc->connect_lock); gadget->deactivated = false; /* @@ -882,7 +880,8 @@ int usb_gadget_activate(struct usb_gadget *gadget) ret = usb_gadget_connect_locked(gadget); mutex_unlock(&gadget->udc->connect_lock); -out: +unlock: + mutex_unlock(&gadget->udc->connect_lock); trace_usb_gadget_activate(gadget, ret); return ret; @@ -1124,12 +1123,21 @@ EXPORT_SYMBOL_GPL(usb_gadget_set_state); /* Acquire connect_lock before calling this function. */ static void usb_udc_connect_control_locked(struct usb_udc *udc) __must_hold(&udc->connect_lock) { - if (udc->vbus && udc->started) + if (udc->vbus) usb_gadget_connect_locked(udc->gadget); else usb_gadget_disconnect_locked(udc->gadget); } +static void vbus_event_work(struct work_struct *work) +{ + struct usb_udc *udc = container_of(work, struct usb_udc, vbus_work); + + mutex_lock(&udc->connect_lock); + usb_udc_connect_control_locked(udc); + mutex_unlock(&udc->connect_lock); +} + /** * usb_udc_vbus_handler - updates the udc core vbus status, and try to * connect or disconnect gadget @@ -1138,17 +1146,23 @@ static void usb_udc_connect_control_locked(struct usb_udc *udc) __must_hold(&udc * * The udc driver calls it when it wants to connect or disconnect gadget * according to vbus status. + * + * This function can be invoked from interrupt context by irq handlers of + * the gadget drivers, however, usb_udc_connect_control() has to run in + * non-atomic context due to the following: + * a. Some of the gadget driver implementations expect the ->pullup + * callback to be invoked in non-atomic context. + * b. usb_gadget_disconnect() acquires udc_lock which is a mutex. + * Hence offload invocation of usb_udc_connect_control() to workqueue. */ void usb_udc_vbus_handler(struct usb_gadget *gadget, bool status) { struct usb_udc *udc = gadget->udc; - mutex_lock(&udc->connect_lock); if (udc) { udc->vbus = status; - usb_udc_connect_control_locked(udc); + schedule_work(&udc->vbus_work); } - mutex_unlock(&udc->connect_lock); } EXPORT_SYMBOL_GPL(usb_udc_vbus_handler); @@ -1381,6 +1395,7 @@ int usb_add_gadget(struct usb_gadget *gadget) mutex_lock(&udc_lock); list_add_tail(&udc->list, &udc_list); mutex_unlock(&udc_lock); + INIT_WORK(&udc->vbus_work, vbus_event_work); ret = device_add(&udc->dev); if (ret) @@ -1512,6 +1527,7 @@ void usb_del_gadget(struct usb_gadget *gadget) flush_work(&gadget->work); device_del(&gadget->dev); ida_free(&gadget_id_numbers, gadget->id_number); + cancel_work_sync(&udc->vbus_work); device_unregister(&udc->dev); } EXPORT_SYMBOL_GPL(usb_del_gadget); @@ -1583,6 +1599,7 @@ static int gadget_bind_driver(struct device *dev) goto err_start; } usb_gadget_enable_async_callbacks(udc); + udc->allow_connect = true; usb_udc_connect_control_locked(udc); mutex_unlock(&udc->connect_lock); @@ -1615,6 +1632,8 @@ static void gadget_unbind_driver(struct device *dev) kobject_uevent(&udc->dev.kobj, KOBJ_CHANGE); + udc->allow_connect = false; + cancel_work_sync(&udc->vbus_work); mutex_lock(&udc->connect_lock); usb_gadget_disconnect_locked(gadget); usb_gadget_disable_async_callbacks(udc); diff --git a/drivers/usb/gadget/udc/renesas_usb3.c b/drivers/usb/gadget/udc/renesas_usb3.c index aac8bc185afa906eb0fa9891a59675478ec9b2e9..eb008e873361fcaaf5f7c5e8f3571e041bfe8ac0 100644 --- a/drivers/usb/gadget/udc/renesas_usb3.c +++ b/drivers/usb/gadget/udc/renesas_usb3.c @@ -2877,9 +2877,9 @@ static int renesas_usb3_probe(struct platform_device *pdev) struct rzv2m_usb3drd *ddata = dev_get_drvdata(pdev->dev.parent); usb3->drd_reg = ddata->reg; - ret = devm_request_irq(ddata->dev, ddata->drd_irq, + ret = devm_request_irq(&pdev->dev, ddata->drd_irq, renesas_usb3_otg_irq, 0, - dev_name(ddata->dev), usb3); + dev_name(&pdev->dev), usb3); if (ret < 0) return ret; } diff --git a/drivers/usb/host/uhci-pci.c b/drivers/usb/host/uhci-pci.c index 3592f757fe05ddfbce622b7bf91b93cc76b30ce2..7bd2fddde770ae4a895dc2988bc1fda0e8d56a55 100644 --- a/drivers/usb/host/uhci-pci.c +++ b/drivers/usb/host/uhci-pci.c @@ -119,11 +119,13 @@ static int uhci_pci_init(struct usb_hcd *hcd) uhci->rh_numports = uhci_count_ports(hcd); - /* Intel controllers report the OverCurrent bit active on. - * VIA controllers report it active off, so we'll adjust the - * bit value. (It's not standardized in the UHCI spec.) + /* + * Intel controllers report the OverCurrent bit active on. VIA + * and ZHAOXIN controllers report it active off, so we'll adjust + * the bit value. (It's not standardized in the UHCI spec.) */ - if (to_pci_dev(uhci_dev(uhci))->vendor == PCI_VENDOR_ID_VIA) + if (to_pci_dev(uhci_dev(uhci))->vendor == PCI_VENDOR_ID_VIA || + to_pci_dev(uhci_dev(uhci))->vendor == PCI_VENDOR_ID_ZHAOXIN) uhci->oc_low = 1; /* HP's server management chip requires a longer port reset delay. */ diff --git a/drivers/usb/host/xhci-pci.c b/drivers/usb/host/xhci-pci.c index ddb79f23fb3b7ce0a54b93f388e46db6508aecff..79b3691f373f34e8f4d325b1e9ae57949c23a54d 100644 --- a/drivers/usb/host/xhci-pci.c +++ b/drivers/usb/host/xhci-pci.c @@ -13,6 +13,7 @@ #include #include #include +#include #include "xhci.h" #include "xhci-trace.h" @@ -387,7 +388,7 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci) if (pdev->vendor == PCI_VENDOR_ID_AMD && pdev->device == PCI_DEVICE_ID_AMD_RENOIR_XHCI) - xhci->quirks |= XHCI_BROKEN_D3COLD; + xhci->quirks |= XHCI_BROKEN_D3COLD_S2I; if (pdev->vendor == PCI_VENDOR_ID_INTEL) { xhci->quirks |= XHCI_LPM_SUPPORT; @@ -801,9 +802,16 @@ static int xhci_pci_suspend(struct usb_hcd *hcd, bool do_wakeup) * Systems with the TI redriver that loses port status change events * need to have the registers polled during D3, so avoid D3cold. */ - if (xhci->quirks & (XHCI_COMP_MODE_QUIRK | XHCI_BROKEN_D3COLD)) + if (xhci->quirks & XHCI_COMP_MODE_QUIRK) pci_d3cold_disable(pdev); +#ifdef CONFIG_SUSPEND + /* d3cold is broken, but only when s2idle is used */ + if (pm_suspend_target_state == PM_SUSPEND_TO_IDLE && + xhci->quirks & (XHCI_BROKEN_D3COLD_S2I)) + pci_d3cold_disable(pdev); +#endif + if (xhci->quirks & XHCI_PME_STUCK_QUIRK) xhci_pme_quirk(hcd); diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index 1ad12d5a48572b15e3d877df43dd80cf480ccc00..2bc82b3a2f984c33e5d32b23336dce57acdf5a24 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -276,6 +276,26 @@ static void inc_enq(struct xhci_hcd *xhci, struct xhci_ring *ring, trace_xhci_inc_enq(ring); } +static int xhci_num_trbs_to(struct xhci_segment *start_seg, union xhci_trb *start, + struct xhci_segment *end_seg, union xhci_trb *end, + unsigned int num_segs) +{ + union xhci_trb *last_on_seg; + int num = 0; + int i = 0; + + do { + if (start_seg == end_seg && end >= start) + return num + (end - start); + last_on_seg = &start_seg->trbs[TRBS_PER_SEGMENT - 1]; + num += last_on_seg - start; + start_seg = start_seg->next; + start = start_seg->trbs; + } while (i++ <= num_segs); + + return -EINVAL; +} + /* * Check to see if there's room to enqueue num_trbs on the ring and make sure * enqueue pointer will not advance into dequeue segment. See rules above. @@ -2140,6 +2160,7 @@ static int finish_td(struct xhci_hcd *xhci, struct xhci_virt_ep *ep, u32 trb_comp_code) { struct xhci_ep_ctx *ep_ctx; + int trbs_freed; ep_ctx = xhci_get_ep_ctx(xhci, ep->vdev->out_ctx, ep->ep_index); @@ -2209,9 +2230,15 @@ static int finish_td(struct xhci_hcd *xhci, struct xhci_virt_ep *ep, } /* Update ring dequeue pointer */ + trbs_freed = xhci_num_trbs_to(ep_ring->deq_seg, ep_ring->dequeue, + td->last_trb_seg, td->last_trb, + ep_ring->num_segs); + if (trbs_freed < 0) + xhci_dbg(xhci, "Failed to count freed trbs at TD finish\n"); + else + ep_ring->num_trbs_free += trbs_freed; ep_ring->dequeue = td->last_trb; ep_ring->deq_seg = td->last_trb_seg; - ep_ring->num_trbs_free += td->num_trbs - 1; inc_deq(xhci, ep_ring); return xhci_td_cleanup(xhci, td, ep_ring, td->status); diff --git a/drivers/usb/host/xhci.h b/drivers/usb/host/xhci.h index 08d721921b7bb22606d82cea5b4736208beb0a9c..6b690ec91ff3ae1044905657dd1f2f0f0f749d80 100644 --- a/drivers/usb/host/xhci.h +++ b/drivers/usb/host/xhci.h @@ -1901,7 +1901,7 @@ struct xhci_hcd { #define XHCI_DISABLE_SPARSE BIT_ULL(38) #define XHCI_SG_TRB_CACHE_SIZE_QUIRK BIT_ULL(39) #define XHCI_NO_SOFT_RETRY BIT_ULL(40) -#define XHCI_BROKEN_D3COLD BIT_ULL(41) +#define XHCI_BROKEN_D3COLD_S2I BIT_ULL(41) #define XHCI_EP_CTX_BROKEN_DCS BIT_ULL(42) #define XHCI_SUSPEND_RESUME_CLKS BIT_ULL(43) #define XHCI_RESET_TO_DEFAULT BIT_ULL(44) diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c index 644a55447fd7f2fe88c957d8003f2e482914edda..fd42e3a0bd1879a2cb9de7fc11d42666b2166beb 100644 --- a/drivers/usb/serial/option.c +++ b/drivers/usb/serial/option.c @@ -248,6 +248,8 @@ static void option_instat_callback(struct urb *urb); #define QUECTEL_VENDOR_ID 0x2c7c /* These Quectel products use Quectel's vendor ID */ #define QUECTEL_PRODUCT_EC21 0x0121 +#define QUECTEL_PRODUCT_EM061K_LTA 0x0123 +#define QUECTEL_PRODUCT_EM061K_LMS 0x0124 #define QUECTEL_PRODUCT_EC25 0x0125 #define QUECTEL_PRODUCT_EG91 0x0191 #define QUECTEL_PRODUCT_EG95 0x0195 @@ -266,6 +268,8 @@ static void option_instat_callback(struct urb *urb); #define QUECTEL_PRODUCT_RM520N 0x0801 #define QUECTEL_PRODUCT_EC200U 0x0901 #define QUECTEL_PRODUCT_EC200S_CN 0x6002 +#define QUECTEL_PRODUCT_EM061K_LWW 0x6008 +#define QUECTEL_PRODUCT_EM061K_LCN 0x6009 #define QUECTEL_PRODUCT_EC200T 0x6026 #define QUECTEL_PRODUCT_RM500K 0x7001 @@ -1189,6 +1193,18 @@ static const struct usb_device_id option_ids[] = { { USB_DEVICE_AND_INTERFACE_INFO(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EM060K, 0xff, 0x00, 0x40) }, { USB_DEVICE_AND_INTERFACE_INFO(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EM060K, 0xff, 0xff, 0x30) }, { USB_DEVICE_AND_INTERFACE_INFO(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EM060K, 0xff, 0xff, 0x40) }, + { USB_DEVICE_AND_INTERFACE_INFO(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EM061K_LCN, 0xff, 0xff, 0x30) }, + { USB_DEVICE_AND_INTERFACE_INFO(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EM061K_LCN, 0xff, 0x00, 0x40) }, + { USB_DEVICE_AND_INTERFACE_INFO(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EM061K_LCN, 0xff, 0xff, 0x40) }, + { USB_DEVICE_AND_INTERFACE_INFO(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EM061K_LMS, 0xff, 0xff, 0x30) }, + { USB_DEVICE_AND_INTERFACE_INFO(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EM061K_LMS, 0xff, 0x00, 0x40) }, + { USB_DEVICE_AND_INTERFACE_INFO(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EM061K_LMS, 0xff, 0xff, 0x40) }, + { USB_DEVICE_AND_INTERFACE_INFO(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EM061K_LTA, 0xff, 0xff, 0x30) }, + { USB_DEVICE_AND_INTERFACE_INFO(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EM061K_LTA, 0xff, 0x00, 0x40) }, + { USB_DEVICE_AND_INTERFACE_INFO(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EM061K_LTA, 0xff, 0xff, 0x40) }, + { USB_DEVICE_AND_INTERFACE_INFO(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EM061K_LWW, 0xff, 0xff, 0x30) }, + { USB_DEVICE_AND_INTERFACE_INFO(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EM061K_LWW, 0xff, 0x00, 0x40) }, + { USB_DEVICE_AND_INTERFACE_INFO(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EM061K_LWW, 0xff, 0xff, 0x40) }, { USB_DEVICE_AND_INTERFACE_INFO(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EM12, 0xff, 0xff, 0xff), .driver_info = RSVD(1) | RSVD(2) | RSVD(3) | RSVD(4) | NUMEP2 }, { USB_DEVICE_AND_INTERFACE_INFO(QUECTEL_VENDOR_ID, QUECTEL_PRODUCT_EM12, 0xff, 0, 0) }, diff --git a/drivers/usb/storage/scsiglue.c b/drivers/usb/storage/scsiglue.c index 8931df5a85fd9e49de0e8b8f399771405168ed06..c54e9805da536a0ec139ad789017b79131c88561 100644 --- a/drivers/usb/storage/scsiglue.c +++ b/drivers/usb/storage/scsiglue.c @@ -406,22 +406,25 @@ static DEF_SCSI_QCMD(queuecommand) ***********************************************************************/ /* Command timeout and abort */ -static int command_abort(struct scsi_cmnd *srb) +static int command_abort_matching(struct us_data *us, struct scsi_cmnd *srb_match) { - struct us_data *us = host_to_us(srb->device->host); - - usb_stor_dbg(us, "%s called\n", __func__); - /* * us->srb together with the TIMED_OUT, RESETTING, and ABORTING * bits are protected by the host lock. */ scsi_lock(us_to_host(us)); - /* Is this command still active? */ - if (us->srb != srb) { + /* is there any active pending command to abort ? */ + if (!us->srb) { scsi_unlock(us_to_host(us)); usb_stor_dbg(us, "-- nothing to abort\n"); + return SUCCESS; + } + + /* Does the command match the passed srb if any ? */ + if (srb_match && us->srb != srb_match) { + scsi_unlock(us_to_host(us)); + usb_stor_dbg(us, "-- pending command mismatch\n"); return FAILED; } @@ -444,6 +447,14 @@ static int command_abort(struct scsi_cmnd *srb) return SUCCESS; } +static int command_abort(struct scsi_cmnd *srb) +{ + struct us_data *us = host_to_us(srb->device->host); + + usb_stor_dbg(us, "%s called\n", __func__); + return command_abort_matching(us, srb); +} + /* * This invokes the transport reset mechanism to reset the state of the * device @@ -455,6 +466,9 @@ static int device_reset(struct scsi_cmnd *srb) usb_stor_dbg(us, "%s called\n", __func__); + /* abort any pending command before reset */ + command_abort_matching(us, NULL); + /* lock the device pointers and do the reset */ mutex_lock(&(us->dev_mutex)); result = us->transport_reset(us); diff --git a/drivers/usb/typec/altmodes/displayport.c b/drivers/usb/typec/altmodes/displayport.c index 8f3e884222adef512a424e16623ad3cad8565143..66de880b28d01a60d592a490d820aa78f87a9aa0 100644 --- a/drivers/usb/typec/altmodes/displayport.c +++ b/drivers/usb/typec/altmodes/displayport.c @@ -516,6 +516,10 @@ static ssize_t pin_assignment_show(struct device *dev, mutex_unlock(&dp->lock); + /* get_current_pin_assignments can return 0 when no matching pin assignments are found */ + if (len == 0) + len++; + buf[len - 1] = '\n'; return len; } diff --git a/drivers/usb/typec/pd.c b/drivers/usb/typec/pd.c index 0bcde1ff4d39b9a212f061448035e53819924fcb..8cc66e4467c48f3f6175f612092c661b484a44e0 100644 --- a/drivers/usb/typec/pd.c +++ b/drivers/usb/typec/pd.c @@ -95,7 +95,7 @@ peak_current_show(struct device *dev, struct device_attribute *attr, char *buf) static ssize_t fast_role_swap_current_show(struct device *dev, struct device_attribute *attr, char *buf) { - return sysfs_emit(buf, "%u\n", to_pdo(dev)->pdo >> PDO_FIXED_FRS_CURR_SHIFT) & 3; + return sysfs_emit(buf, "%u\n", (to_pdo(dev)->pdo >> PDO_FIXED_FRS_CURR_SHIFT) & 3); } static DEVICE_ATTR_RO(fast_role_swap_current); diff --git a/drivers/usb/typec/tipd/core.c b/drivers/usb/typec/tipd/core.c index 8b075ca82ef6de80b4b6f31fa316b675141705c3..603dbd44deba9aaab55ad671526a6bf14befd339 100644 --- a/drivers/usb/typec/tipd/core.c +++ b/drivers/usb/typec/tipd/core.c @@ -886,6 +886,9 @@ static void tps6598x_remove(struct i2c_client *client) { struct tps6598x *tps = i2c_get_clientdata(client); + if (!client->irq) + cancel_delayed_work_sync(&tps->wq_poll); + tps6598x_disconnect(tps, 0); typec_unregister_port(tps->port); usb_role_switch_put(tps->role_sw); @@ -917,7 +920,7 @@ static int __maybe_unused tps6598x_resume(struct device *dev) enable_irq(client->irq); } - if (client->irq) + if (!client->irq) queue_delayed_work(system_power_efficient_wq, &tps->wq_poll, msecs_to_jiffies(POLL_INTERVAL)); diff --git a/drivers/usb/typec/ucsi/ucsi.c b/drivers/usb/typec/ucsi/ucsi.c index 2b472ec01dc42e84c7c355c6eae4073428b9a2ac..b664ecbb798be6012d988e15d7622c6071f20fe8 100644 --- a/drivers/usb/typec/ucsi/ucsi.c +++ b/drivers/usb/typec/ucsi/ucsi.c @@ -132,10 +132,8 @@ static int ucsi_exec_command(struct ucsi *ucsi, u64 cmd) if (ret) return ret; - if (cci & UCSI_CCI_BUSY) { - ucsi->ops->async_write(ucsi, UCSI_CANCEL, NULL, 0); - return -EBUSY; - } + if (cmd != UCSI_CANCEL && cci & UCSI_CCI_BUSY) + return ucsi_exec_command(ucsi, UCSI_CANCEL); if (!(cci & UCSI_CCI_COMMAND_COMPLETE)) return -EIO; @@ -149,6 +147,11 @@ static int ucsi_exec_command(struct ucsi *ucsi, u64 cmd) return ucsi_read_error(ucsi); } + if (cmd == UCSI_CANCEL && cci & UCSI_CCI_CANCEL_COMPLETE) { + ret = ucsi_acknowledge_command(ucsi); + return ret ? ret : -EBUSY; + } + return UCSI_CCI_LENGTH(cci); } diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c index e29e32b306adc683e9e9e7414244d709f71f2b12..279ac6a558d29a7124204c9229fe13f7ee4eafd3 100644 --- a/drivers/vdpa/mlx5/net/mlx5_vnet.c +++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c @@ -3349,10 +3349,10 @@ static void mlx5_vdpa_dev_del(struct vdpa_mgmt_dev *v_mdev, struct vdpa_device * mlx5_vdpa_remove_debugfs(ndev->debugfs); ndev->debugfs = NULL; unregister_link_notifier(ndev); + _vdpa_unregister_device(dev); wq = mvdev->wq; mvdev->wq = NULL; destroy_workqueue(wq); - _vdpa_unregister_device(dev); mgtdev->ndev = NULL; } diff --git a/drivers/vdpa/vdpa_user/vduse_dev.c b/drivers/vdpa/vdpa_user/vduse_dev.c index de97e38c3b82fea80204d4be6ccfbd1b08cf4063..5f5c21674fdce5050fa1bc0c672b3b3606871da9 100644 --- a/drivers/vdpa/vdpa_user/vduse_dev.c +++ b/drivers/vdpa/vdpa_user/vduse_dev.c @@ -1685,6 +1685,9 @@ static bool vduse_validate_config(struct vduse_dev_config *config) if (config->vq_num > 0xffff) return false; + if (!config->name[0]) + return false; + if (!device_is_allowed(config->device_id)) return false; diff --git a/drivers/vfio/vfio_iommu_type1.c b/drivers/vfio/vfio_iommu_type1.c index 3d4dd9420c30701453be9323a0e3e46bfbf58755..0d2f805468e190c37b5e79527f864d13a42d311e 100644 --- a/drivers/vfio/vfio_iommu_type1.c +++ b/drivers/vfio/vfio_iommu_type1.c @@ -860,6 +860,11 @@ static int vfio_iommu_type1_pin_pages(void *iommu_data, if (ret) goto pin_unwind; + if (!pfn_valid(phys_pfn)) { + ret = -EINVAL; + goto pin_unwind; + } + ret = vfio_add_to_pfn_list(dma, iova, phys_pfn); if (ret) { if (put_pfn(phys_pfn, dma->prot) && do_accounting) diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c index 07181cd8d52e638126cf6583eeba3bbfe17ffdb0..ae2273196b0c902c472b3f5cf45249cb6cae37bf 100644 --- a/drivers/vhost/net.c +++ b/drivers/vhost/net.c @@ -935,13 +935,18 @@ static void handle_tx_zerocopy(struct vhost_net *net, struct socket *sock) err = sock->ops->sendmsg(sock, &msg, len); if (unlikely(err < 0)) { + bool retry = err == -EAGAIN || err == -ENOMEM || err == -ENOBUFS; + if (zcopy_used) { if (vq->heads[ubuf->desc].len == VHOST_DMA_IN_PROGRESS) vhost_net_ubuf_put(ubufs); - nvq->upend_idx = ((unsigned)nvq->upend_idx - 1) - % UIO_MAXIOV; + if (retry) + nvq->upend_idx = ((unsigned)nvq->upend_idx - 1) + % UIO_MAXIOV; + else + vq->heads[ubuf->desc].len = VHOST_DMA_DONE_LEN; } - if (err == -EAGAIN || err == -ENOMEM || err == -ENOBUFS) { + if (retry) { vhost_discard_vq_desc(vq, 1); vhost_net_enable_vq(net, vq); break; diff --git a/drivers/vhost/vdpa.c b/drivers/vhost/vdpa.c index 8c1aefc865f0854ca7c07873a2c757c1688d30f0..bf77924d5b602389fd79b9faadd8409cb1422943 100644 --- a/drivers/vhost/vdpa.c +++ b/drivers/vhost/vdpa.c @@ -407,7 +407,10 @@ static long vhost_vdpa_set_features(struct vhost_vdpa *v, u64 __user *featurep) { struct vdpa_device *vdpa = v->vdpa; const struct vdpa_config_ops *ops = vdpa->config; + struct vhost_dev *d = &v->vdev; + u64 actual_features; u64 features; + int i; /* * It's not allowed to change the features after they have @@ -422,6 +425,16 @@ static long vhost_vdpa_set_features(struct vhost_vdpa *v, u64 __user *featurep) if (vdpa_set_features(vdpa, features)) return -EINVAL; + /* let the vqs know what has been configured */ + actual_features = ops->get_driver_features(vdpa); + for (i = 0; i < d->nvqs; ++i) { + struct vhost_virtqueue *vq = d->vqs[i]; + + mutex_lock(&vq->mutex); + vq->acked_features = actual_features; + mutex_unlock(&vq->mutex); + } + return 0; } @@ -594,7 +607,14 @@ static long vhost_vdpa_vring_ioctl(struct vhost_vdpa *v, unsigned int cmd, if (r) return r; - vq->last_avail_idx = vq_state.split.avail_index; + if (vhost_has_feature(vq, VIRTIO_F_RING_PACKED)) { + vq->last_avail_idx = vq_state.packed.last_avail_idx | + (vq_state.packed.last_avail_counter << 15); + vq->last_used_idx = vq_state.packed.last_used_idx | + (vq_state.packed.last_used_counter << 15); + } else { + vq->last_avail_idx = vq_state.split.avail_index; + } break; } @@ -612,9 +632,15 @@ static long vhost_vdpa_vring_ioctl(struct vhost_vdpa *v, unsigned int cmd, break; case VHOST_SET_VRING_BASE: - vq_state.split.avail_index = vq->last_avail_idx; - if (ops->set_vq_state(vdpa, idx, &vq_state)) - r = -EINVAL; + if (vhost_has_feature(vq, VIRTIO_F_RING_PACKED)) { + vq_state.packed.last_avail_idx = vq->last_avail_idx & 0x7fff; + vq_state.packed.last_avail_counter = !!(vq->last_avail_idx & 0x8000); + vq_state.packed.last_used_idx = vq->last_used_idx & 0x7fff; + vq_state.packed.last_used_counter = !!(vq->last_used_idx & 0x8000); + } else { + vq_state.split.avail_index = vq->last_avail_idx; + } + r = ops->set_vq_state(vdpa, idx, &vq_state); break; case VHOST_SET_VRING_CALL: diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index a92af08e786450640b2bb6fa848810dec402fcbb..60c9ebd629dd159d0ceb8a0c14f96091be193753 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -235,7 +235,7 @@ void vhost_dev_flush(struct vhost_dev *dev) { struct vhost_flush_struct flush; - if (dev->worker) { + if (dev->worker.vtsk) { init_completion(&flush.wait_event); vhost_work_init(&flush.work, vhost_flush_work); @@ -247,7 +247,7 @@ EXPORT_SYMBOL_GPL(vhost_dev_flush); void vhost_work_queue(struct vhost_dev *dev, struct vhost_work *work) { - if (!dev->worker) + if (!dev->worker.vtsk) return; if (!test_and_set_bit(VHOST_WORK_QUEUED, &work->flags)) { @@ -255,8 +255,8 @@ void vhost_work_queue(struct vhost_dev *dev, struct vhost_work *work) * sure it was not in the list. * test_and_set_bit() implies a memory barrier. */ - llist_add(&work->node, &dev->worker->work_list); - wake_up_process(dev->worker->vtsk->task); + llist_add(&work->node, &dev->worker.work_list); + vhost_task_wake(dev->worker.vtsk); } } EXPORT_SYMBOL_GPL(vhost_work_queue); @@ -264,7 +264,7 @@ EXPORT_SYMBOL_GPL(vhost_work_queue); /* A lockless hint for busy polling code to exit the loop */ bool vhost_has_work(struct vhost_dev *dev) { - return dev->worker && !llist_empty(&dev->worker->work_list); + return !llist_empty(&dev->worker.work_list); } EXPORT_SYMBOL_GPL(vhost_has_work); @@ -333,31 +333,21 @@ static void vhost_vq_reset(struct vhost_dev *dev, __vhost_vq_meta_reset(vq); } -static int vhost_worker(void *data) +static bool vhost_worker(void *data) { struct vhost_worker *worker = data; struct vhost_work *work, *work_next; struct llist_node *node; - for (;;) { - /* mb paired w/ kthread_stop */ - set_current_state(TASK_INTERRUPTIBLE); - - if (vhost_task_should_stop(worker->vtsk)) { - __set_current_state(TASK_RUNNING); - break; - } - - node = llist_del_all(&worker->work_list); - if (!node) - schedule(); + node = llist_del_all(&worker->work_list); + if (node) { + __set_current_state(TASK_RUNNING); node = llist_reverse_order(node); /* make sure flag is seen after deletion */ smp_wmb(); llist_for_each_entry_safe(work, work_next, node, node) { clear_bit(VHOST_WORK_QUEUED, &work->flags); - __set_current_state(TASK_RUNNING); kcov_remote_start_common(worker->kcov_handle); work->fn(work); kcov_remote_stop(); @@ -365,7 +355,7 @@ static int vhost_worker(void *data) } } - return 0; + return !!node; } static void vhost_vq_free_iovecs(struct vhost_virtqueue *vq) @@ -468,7 +458,8 @@ void vhost_dev_init(struct vhost_dev *dev, dev->umem = NULL; dev->iotlb = NULL; dev->mm = NULL; - dev->worker = NULL; + memset(&dev->worker, 0, sizeof(dev->worker)); + init_llist_head(&dev->worker.work_list); dev->iov_limit = iov_limit; dev->weight = weight; dev->byte_weight = byte_weight; @@ -542,47 +533,30 @@ static void vhost_detach_mm(struct vhost_dev *dev) static void vhost_worker_free(struct vhost_dev *dev) { - struct vhost_worker *worker = dev->worker; - - if (!worker) + if (!dev->worker.vtsk) return; - dev->worker = NULL; - WARN_ON(!llist_empty(&worker->work_list)); - vhost_task_stop(worker->vtsk); - kfree(worker); + WARN_ON(!llist_empty(&dev->worker.work_list)); + vhost_task_stop(dev->worker.vtsk); + dev->worker.kcov_handle = 0; + dev->worker.vtsk = NULL; } static int vhost_worker_create(struct vhost_dev *dev) { - struct vhost_worker *worker; struct vhost_task *vtsk; char name[TASK_COMM_LEN]; - int ret; - - worker = kzalloc(sizeof(*worker), GFP_KERNEL_ACCOUNT); - if (!worker) - return -ENOMEM; - dev->worker = worker; - worker->kcov_handle = kcov_common_handle(); - init_llist_head(&worker->work_list); snprintf(name, sizeof(name), "vhost-%d", current->pid); - vtsk = vhost_task_create(vhost_worker, worker, name); - if (!vtsk) { - ret = -ENOMEM; - goto free_worker; - } + vtsk = vhost_task_create(vhost_worker, &dev->worker, name); + if (!vtsk) + return -ENOMEM; - worker->vtsk = vtsk; + dev->worker.kcov_handle = kcov_common_handle(); + dev->worker.vtsk = vtsk; vhost_task_start(vtsk); return 0; - -free_worker: - kfree(worker); - dev->worker = NULL; - return ret; } /* Caller should have device mutex */ @@ -1626,17 +1600,25 @@ long vhost_vring_ioctl(struct vhost_dev *d, unsigned int ioctl, void __user *arg r = -EFAULT; break; } - if (s.num > 0xffff) { - r = -EINVAL; - break; + if (vhost_has_feature(vq, VIRTIO_F_RING_PACKED)) { + vq->last_avail_idx = s.num & 0xffff; + vq->last_used_idx = (s.num >> 16) & 0xffff; + } else { + if (s.num > 0xffff) { + r = -EINVAL; + break; + } + vq->last_avail_idx = s.num; } - vq->last_avail_idx = s.num; /* Forget the cached index value. */ vq->avail_idx = vq->last_avail_idx; break; case VHOST_GET_VRING_BASE: s.index = idx; - s.num = vq->last_avail_idx; + if (vhost_has_feature(vq, VIRTIO_F_RING_PACKED)) + s.num = (u32)vq->last_avail_idx | ((u32)vq->last_used_idx << 16); + else + s.num = vq->last_avail_idx; if (copy_to_user(argp, &s, sizeof s)) r = -EFAULT; break; @@ -2575,12 +2557,11 @@ EXPORT_SYMBOL_GPL(vhost_disable_notify); /* Create a new message. */ struct vhost_msg_node *vhost_new_msg(struct vhost_virtqueue *vq, int type) { - struct vhost_msg_node *node = kmalloc(sizeof *node, GFP_KERNEL); + /* Make sure all padding within the structure is initialized. */ + struct vhost_msg_node *node = kzalloc(sizeof(*node), GFP_KERNEL); if (!node) return NULL; - /* Make sure all padding within the structure is initialized. */ - memset(&node->msg, 0, sizeof node->msg); node->vq = vq; node->msg.type = type; return node; diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h index 0308638cdeeebef46ca0ad98994eb8c6edd803f4..fc900be504b38e48d7c74f96e78d825d4a2a909e 100644 --- a/drivers/vhost/vhost.h +++ b/drivers/vhost/vhost.h @@ -92,13 +92,17 @@ struct vhost_virtqueue { /* The routine to call when the Guest pings us, or timeout. */ vhost_work_fn_t handle_kick; - /* Last available index we saw. */ + /* Last available index we saw. + * Values are limited to 0x7fff, and the high bit is used as + * a wrap counter when using VIRTIO_F_RING_PACKED. */ u16 last_avail_idx; /* Caches available index value from user. */ u16 avail_idx; - /* Last index we used. */ + /* Last index we used. + * Values are limited to 0x7fff, and the high bit is used as + * a wrap counter when using VIRTIO_F_RING_PACKED. */ u16 last_used_idx; /* Used flags */ @@ -154,7 +158,7 @@ struct vhost_dev { struct vhost_virtqueue **vqs; int nvqs; struct eventfd_ctx *log_ctx; - struct vhost_worker *worker; + struct vhost_worker worker; struct vhost_iotlb *umem; struct vhost_iotlb *iotlb; spinlock_t iotlb_lock; diff --git a/drivers/video/fbdev/Kconfig b/drivers/video/fbdev/Kconfig index 6e4b6ad9d8e62e7e94edd7cee5fccc692106326f..6df9bd09454a26b45559bf49168c166cb016d180 100644 --- a/drivers/video/fbdev/Kconfig +++ b/drivers/video/fbdev/Kconfig @@ -124,7 +124,7 @@ config FB_PROVIDE_GET_FB_UNMAPPED_AREA depends on FB help Allow generic frame-buffer to provide get_fb_unmapped_area - function. + function to provide shareable character device support on nommu. menuconfig FB_FOREIGN_ENDIAN bool "Framebuffer foreign endianness support" diff --git a/drivers/video/fbdev/arcfb.c b/drivers/video/fbdev/arcfb.c index 3d926c3ada7300f7b3977969fd79b03783ad797d..9aaea3be82811a4ff81f18df78c94262d48cda04 100644 --- a/drivers/video/fbdev/arcfb.c +++ b/drivers/video/fbdev/arcfb.c @@ -594,7 +594,7 @@ static int arcfb_probe(struct platform_device *dev) return retval; } -static int arcfb_remove(struct platform_device *dev) +static void arcfb_remove(struct platform_device *dev) { struct fb_info *info = platform_get_drvdata(dev); @@ -605,12 +605,11 @@ static int arcfb_remove(struct platform_device *dev) vfree(info->screen_buffer); framebuffer_release(info); } - return 0; } static struct platform_driver arcfb_driver = { .probe = arcfb_probe, - .remove = arcfb_remove, + .remove_new = arcfb_remove, .driver = { .name = "arcfb", }, diff --git a/drivers/video/fbdev/aty/atyfb_base.c b/drivers/video/fbdev/aty/atyfb_base.c index b02e4e645035c44716740805bc3ffdb666d12c0d..cba2b113b28b0533418e017e8658138c97d21fd7 100644 --- a/drivers/video/fbdev/aty/atyfb_base.c +++ b/drivers/video/fbdev/aty/atyfb_base.c @@ -3498,11 +3498,6 @@ static int atyfb_setup_generic(struct pci_dev *pdev, struct fb_info *info, if (ret) goto atyfb_setup_generic_fail; #endif - if (!(aty_ld_le32(CRTC_GEN_CNTL, par) & CRTC_EXT_DISP_EN)) - par->clk_wr_offset = (inb(R_GENMO) & 0x0CU) >> 2; - else - par->clk_wr_offset = aty_ld_8(CLOCK_CNTL, par) & 0x03U; - /* according to ATI, we should use clock 3 for acelerated mode */ par->clk_wr_offset = 3; diff --git a/drivers/video/fbdev/au1100fb.c b/drivers/video/fbdev/au1100fb.c index 519313b8bb004fa245def79b0e74525a4e2edd42..648d6cac86e8fb1386c459d05852f6f603c00fd7 100644 --- a/drivers/video/fbdev/au1100fb.c +++ b/drivers/video/fbdev/au1100fb.c @@ -520,13 +520,10 @@ static int au1100fb_drv_probe(struct platform_device *dev) return -ENODEV; } -int au1100fb_drv_remove(struct platform_device *dev) +void au1100fb_drv_remove(struct platform_device *dev) { struct au1100fb_device *fbdev = NULL; - if (!dev) - return -ENODEV; - fbdev = platform_get_drvdata(dev); #if !defined(CONFIG_FRAMEBUFFER_CONSOLE) && defined(CONFIG_LOGO) @@ -543,8 +540,6 @@ int au1100fb_drv_remove(struct platform_device *dev) clk_disable_unprepare(fbdev->lcdclk); clk_put(fbdev->lcdclk); } - - return 0; } #ifdef CONFIG_PM @@ -593,9 +588,9 @@ static struct platform_driver au1100fb_driver = { .name = "au1100-lcd", }, .probe = au1100fb_drv_probe, - .remove = au1100fb_drv_remove, + .remove_new = au1100fb_drv_remove, .suspend = au1100fb_drv_suspend, - .resume = au1100fb_drv_resume, + .resume = au1100fb_drv_resume, }; module_platform_driver(au1100fb_driver); diff --git a/drivers/video/fbdev/au1200fb.c b/drivers/video/fbdev/au1200fb.c index 847f82ed52ff35e2795b65ffec50936826297d77..5c232eb137247ba6d454a8024b1dc62fd5689266 100644 --- a/drivers/video/fbdev/au1200fb.c +++ b/drivers/video/fbdev/au1200fb.c @@ -1765,7 +1765,7 @@ static int au1200fb_drv_probe(struct platform_device *dev) return ret; } -static int au1200fb_drv_remove(struct platform_device *dev) +static void au1200fb_drv_remove(struct platform_device *dev) { struct au1200fb_platdata *pd = platform_get_drvdata(dev); struct fb_info *fbi; @@ -1788,8 +1788,6 @@ static int au1200fb_drv_remove(struct platform_device *dev) } free_irq(platform_get_irq(dev, 0), (void *)dev); - - return 0; } #ifdef CONFIG_PM @@ -1840,7 +1838,7 @@ static struct platform_driver au1200fb_driver = { .pm = AU1200FB_PMOPS, }, .probe = au1200fb_drv_probe, - .remove = au1200fb_drv_remove, + .remove_new = au1200fb_drv_remove, }; module_platform_driver(au1200fb_driver); diff --git a/drivers/video/fbdev/broadsheetfb.c b/drivers/video/fbdev/broadsheetfb.c index e9c5d5c04062bd6b41ee6be1111d9f623da3e60b..5f18af88e7408deda392308bf90202cdd8d04171 100644 --- a/drivers/video/fbdev/broadsheetfb.c +++ b/drivers/video/fbdev/broadsheetfb.c @@ -1193,7 +1193,7 @@ static int broadsheetfb_probe(struct platform_device *dev) } -static int broadsheetfb_remove(struct platform_device *dev) +static void broadsheetfb_remove(struct platform_device *dev) { struct fb_info *info = platform_get_drvdata(dev); @@ -1209,12 +1209,11 @@ static int broadsheetfb_remove(struct platform_device *dev) module_put(par->board->owner); framebuffer_release(info); } - return 0; } static struct platform_driver broadsheetfb_driver = { .probe = broadsheetfb_probe, - .remove = broadsheetfb_remove, + .remove_new = broadsheetfb_remove, .driver = { .name = "broadsheetfb", }, diff --git a/drivers/video/fbdev/bw2.c b/drivers/video/fbdev/bw2.c index 9cbadcd18b256c4ff0e525f7d09867d377d1028f..025d663dc6fdcae85bc10e8e1813cd5305ca97e5 100644 --- a/drivers/video/fbdev/bw2.c +++ b/drivers/video/fbdev/bw2.c @@ -352,7 +352,7 @@ static int bw2_probe(struct platform_device *op) return err; } -static int bw2_remove(struct platform_device *op) +static void bw2_remove(struct platform_device *op) { struct fb_info *info = dev_get_drvdata(&op->dev); struct bw2_par *par = info->par; @@ -363,8 +363,6 @@ static int bw2_remove(struct platform_device *op) of_iounmap(&op->resource[0], info->screen_base, info->fix.smem_len); framebuffer_release(info); - - return 0; } static const struct of_device_id bw2_match[] = { @@ -381,7 +379,7 @@ static struct platform_driver bw2_driver = { .of_match_table = bw2_match, }, .probe = bw2_probe, - .remove = bw2_remove, + .remove_new = bw2_remove, }; static int __init bw2_init(void) diff --git a/drivers/video/fbdev/core/bitblit.c b/drivers/video/fbdev/core/bitblit.c index f98e8f298bc19a0ae63a8d493ae2e316d8e2e38a..8587c9da067003f74f50d3234a759bf3f218bf12 100644 --- a/drivers/video/fbdev/core/bitblit.c +++ b/drivers/video/fbdev/core/bitblit.c @@ -247,6 +247,9 @@ static void bit_cursor(struct vc_data *vc, struct fb_info *info, int mode, cursor.set = 0; + if (!vc->vc_font.data) + return; + c = scr_readw((u16 *) vc->vc_pos); attribute = get_attribute(info, c); src = vc->vc_font.data + ((c & charmask) * (w * vc->vc_font.height)); diff --git a/drivers/video/fbdev/core/fbmem.c b/drivers/video/fbdev/core/fbmem.c index d5f897b5ba54f0361e04dac8fb72b783fb6a710d..329d16e49a9004c18cc76c279ce2530a1657b104 100644 --- a/drivers/video/fbdev/core/fbmem.c +++ b/drivers/video/fbdev/core/fbmem.c @@ -1363,7 +1363,7 @@ __releases(&info->lock) } #if defined(CONFIG_FB_PROVIDE_GET_FB_UNMAPPED_AREA) && !defined(CONFIG_MMU) -unsigned long get_fb_unmapped_area(struct file *filp, +static unsigned long get_fb_unmapped_area(struct file *filp, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { diff --git a/drivers/video/fbdev/i810/i810_dvt.c b/drivers/video/fbdev/i810/i810_dvt.c index b4b3670667ab8ce42dce3d58cc2d65be02d72eaa..2082b5c92e8f3b9f72deff934b6fc4b8a958e8fc 100644 --- a/drivers/video/fbdev/i810/i810_dvt.c +++ b/drivers/video/fbdev/i810/i810_dvt.c @@ -14,6 +14,7 @@ #include "i810_regs.h" #include "i810.h" +#include "i810_main.h" struct mode_registers std_modes[] = { /* 640x480 @ 60Hz */ @@ -276,7 +277,7 @@ void i810fb_fill_var_timings(struct fb_var_screeninfo *var) var->upper_margin = total - (yres + var->lower_margin + var->vsync_len); } -u32 i810_get_watermark(struct fb_var_screeninfo *var, +u32 i810_get_watermark(const struct fb_var_screeninfo *var, struct i810fb_par *par) { struct mode_registers *params = &par->regs; diff --git a/drivers/video/fbdev/imsttfb.c b/drivers/video/fbdev/imsttfb.c index 975dd682fae4bb59187b0589a4487eee1acb8cd9..ee7d01ad14068687b968a42f1558ed730664e47b 100644 --- a/drivers/video/fbdev/imsttfb.c +++ b/drivers/video/fbdev/imsttfb.c @@ -1452,9 +1452,13 @@ static int init_imstt(struct fb_info *info) FBINFO_HWACCEL_FILLRECT | FBINFO_HWACCEL_YPAN; - fb_alloc_cmap(&info->cmap, 0, 0); + if (fb_alloc_cmap(&info->cmap, 0, 0)) { + framebuffer_release(info); + return -ENODEV; + } if (register_framebuffer(info) < 0) { + fb_dealloc_cmap(&info->cmap); framebuffer_release(info); return -ENODEV; } @@ -1531,8 +1535,10 @@ static int imsttfb_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto error; info->pseudo_palette = par->palette; ret = init_imstt(info); - if (!ret) - pci_set_drvdata(pdev, info); + if (ret) + goto error; + + pci_set_drvdata(pdev, info); return ret; error: diff --git a/drivers/video/fbdev/matrox/matroxfb_maven.c b/drivers/video/fbdev/matrox/matroxfb_maven.c index 727a10a59811b38d701f12159ce5201ee8d18bea..b15a8ad92ba704e234e7e16874897dffb2427cf7 100644 --- a/drivers/video/fbdev/matrox/matroxfb_maven.c +++ b/drivers/video/fbdev/matrox/matroxfb_maven.c @@ -1291,7 +1291,7 @@ static struct i2c_driver maven_driver={ .driver = { .name = "maven", }, - .probe_new = maven_probe, + .probe = maven_probe, .remove = maven_remove, .id_table = maven_id, }; diff --git a/drivers/video/fbdev/omap2/omapfb/displays/panel-tpo-td043mtea1.c b/drivers/video/fbdev/omap2/omapfb/displays/panel-tpo-td043mtea1.c index 1eaa35c278359836ebd4860af036b42fce2d010b..477789cff8e08d5e620a0754feb1f1dba6008370 100644 --- a/drivers/video/fbdev/omap2/omapfb/displays/panel-tpo-td043mtea1.c +++ b/drivers/video/fbdev/omap2/omapfb/displays/panel-tpo-td043mtea1.c @@ -491,7 +491,8 @@ static int tpo_td043_probe(struct spi_device *spi) ddata->vcc_reg = devm_regulator_get(&spi->dev, "vcc"); if (IS_ERR(ddata->vcc_reg)) { - r = dev_err_probe(&spi->dev, r, "failed to get LCD VCC regulator\n"); + r = dev_err_probe(&spi->dev, PTR_ERR(ddata->vcc_reg), + "failed to get LCD VCC regulator\n"); goto err_regulator; } diff --git a/drivers/video/fbdev/ssd1307fb.c b/drivers/video/fbdev/ssd1307fb.c index a8f2975de76bebb104f36fecc01e7030b86e7ed5..11c3737982790cda958221c50ddfa18c7074453a 100644 --- a/drivers/video/fbdev/ssd1307fb.c +++ b/drivers/video/fbdev/ssd1307fb.c @@ -847,7 +847,7 @@ static const struct i2c_device_id ssd1307fb_i2c_id[] = { MODULE_DEVICE_TABLE(i2c, ssd1307fb_i2c_id); static struct i2c_driver ssd1307fb_driver = { - .probe_new = ssd1307fb_probe, + .probe = ssd1307fb_probe, .remove = ssd1307fb_remove, .id_table = ssd1307fb_i2c_id, .driver = { diff --git a/drivers/video/fbdev/udlfb.c b/drivers/video/fbdev/udlfb.c index 09cf9381075aa459309067fd21d0dded39969318..a4a21b4ac28c6fd8affa7e6ba4d5dea4bde7a7e2 100644 --- a/drivers/video/fbdev/udlfb.c +++ b/drivers/video/fbdev/udlfb.c @@ -27,6 +27,8 @@ #include