Commits · 6688cc05473b36a0a3d3971e1adf1712919b32eb · jan.koester / Linux · GitLab

Oct 09, 2013
- mm: numa: Do not group on RO pages · 6688cc05
  Peter Zijlstra authored Oct 07, 2013
  6688cc05
- mm: numa: Copy cpupid on page migration · 7851a45c
  Rik van Riel authored Oct 07, 2013
  7851a45c
- sched/numa: Report a NUMA task group ID · e29cf08b
  Mel Gorman authored Oct 07, 2013
  e29cf08b
- sched/numa: Use {cpu, pid} to create task groups for shared faults · 8c8a743c
  Peter Zijlstra authored Oct 07, 2013
  8c8a743c
- mm: numa: Change page last {nid,pid} into {cpu,pid} · 90572890
  Peter Zijlstra authored Oct 07, 2013
  90572890
- sched/numa: Fix placement of workloads spread across multiple nodes · e1dda8a7
  Rik van Riel authored Oct 07, 2013
  e1dda8a7
- sched/numa: Favor placing a task on the preferred node · 2c8a50aa
  Mel Gorman authored Oct 07, 2013
  2c8a50aa
- sched/numa: Use a system-wide search to find swap/migration candidates · fb13c7ee
  Mel Gorman authored Oct 07, 2013
  fb13c7ee
- sched/numa: Introduce migrate_swap() · ac66f547
  Peter Zijlstra authored Oct 07, 2013
  ac66f547
- stop_machine: Introduce stop_two_cpus() · 1be0bd77
  Peter Zijlstra authored Oct 07, 2013
  1be0bd77
- mm: numa: Trap pmd hinting faults only if we would otherwise trap PTE faults · 25cbbef1
  Mel Gorman authored Oct 07, 2013
  25cbbef1
- sched/numa: Do not trap hinting faults for shared libraries · 4591ce4f
  Mel Gorman authored Oct 07, 2013
  4591ce4f
- sched/numa: Increment numa_migrate_seq when task runs in correct location · 06ea5e03
  Rik van Riel authored Oct 07, 2013
  06ea5e03
- sched/numa: Retry migration of tasks to CPU on a preferred node · 6b9a7460
  Mel Gorman authored Oct 07, 2013
  6b9a7460
- sched/numa: Avoid overloading CPUs on a preferred NUMA node · 58d081b5
  Mel Gorman authored Oct 07, 2013
  58d081b5
- mm: numa: Limit NUMA scanning to migrate-on-fault VMAs · fc314724
  Mel Gorman authored Oct 07, 2013
  fc314724
- sched/numa: Do not migrate memory immediately after switching node · 6fe6b2d6
  Rik van Riel authored Oct 07, 2013
  6fe6b2d6
- sched/numa: Set preferred NUMA node based on number of private faults · b795854b
  Mel Gorman authored Oct 07, 2013
  b795854b
- sched/numa: Remove check that skips small VMAs · 073b5bee
  Mel Gorman authored Oct 07, 2013
  073b5bee
- mm: numa: Scan pages with elevated page_mapcount · 1bc115d8
  Mel Gorman authored Oct 07, 2013
  1bc115d8
- sched/numa: Check current->mm before allocating NUMA faults · 9ff1d9ff
  Mel Gorman authored Oct 07, 2013
  9ff1d9ff
- sched/numa: Add infrastructure for split shared/private accounting of NUMA hinting faults · ac8e895b
  Mel Gorman authored Oct 07, 2013
  ac8e895b
- sched/numa: Reschedule task on preferred NUMA node once selected · e6628d5b
  Mel Gorman authored Oct 07, 2013
  e6628d5b
- sched/numa: Resist moving tasks towards nodes with fewer hinting faults · 7a0f3083
  Mel Gorman authored Oct 07, 2013
  7a0f3083
- sched/numa: Favour moving tasks towards the preferred node · 3a7053b3
  Mel Gorman authored Oct 07, 2013
  3a7053b3
- sched/numa: Update NUMA hinting faults once per scan · 745d6147
  Mel Gorman authored Oct 07, 2013
  745d6147
- sched/numa: Select a preferred node with the most numa hinting faults · 688b7585
  Mel Gorman authored Oct 07, 2013
  688b7585
- sched/numa: Track NUMA hinting faults on per-node basis · f809ca9a
  Mel Gorman authored Oct 07, 2013
  f809ca9a
- sched/numa: Slow scan rate if no NUMA hinting faults are being recorded · f307cd1a
  Mel Gorman authored Oct 07, 2013
  f307cd1a
- sched/numa: Set the scan rate proportional to the memory usage of the task being scanned · 598f0ec0
  Mel Gorman authored Oct 07, 2013
  598f0ec0
- sched/numa: Initialise numa_next_scan properly · 7e8d16b6
  Mel Gorman authored Oct 07, 2013
  7e8d16b6
- Revert "mm: sched: numa: Delay PTE scanning until a task is scheduled on a new node" · b726b7df
  Mel Gorman authored Oct 07, 2013
  b726b7df
- sched/numa: Continue PTE scanning even if migrate rate limited · 9e645ab6
  Peter Zijlstra authored Oct 07, 2013
  9e645ab6
- sched/numa: Mitigate chance that same task always updates PTEs · 19a78d11
  Peter Zijlstra authored Oct 07, 2013
  19a78d11
- mm: numa: Do not migrate or account for hinting faults on the zero page · a1a46184
  Mel Gorman authored Oct 07, 2013
  a1a46184
- mm: Only flush TLBs if a transhuge PMD is modified for NUMA pte scanning · f123d74a
  Mel Gorman authored Oct 07, 2013
  f123d74a
- mm: Do not flush TLB during protection change if !pte_present && !migration_entry · e920e14c
  Mel Gorman authored Oct 07, 2013
  e920e14c
- mm: Account for a THP NUMA hinting update as one PTE update · afcae265
  Mel Gorman authored Oct 07, 2013
  afcae265
- mm: Close races between THP migration and PMD numa clearing · a54a407f
  Mel Gorman authored Oct 07, 2013
  a54a407f
- mm: numa: Sanitize task_numa_fault() callsites · 8191acbd
  Mel Gorman authored Oct 07, 2013
  8191acbd