cpufreq: Documentation: Minor reformatting (4e660759) · Commits · jan.koester / Linux

Documentation/cpu-freq/governors.txt

+112 −93

Original line number	Original line	Diff line number	Diff line
	@@ -111,68 +111,82 @@ directory.

	The CPUfreq governor "ondemand" sets the CPU depending on the		The CPUfreq governor "ondemand" sets the CPU depending on the
	current usage. To do this the CPU must have the capability to		current usage. To do this the CPU must have the capability to
	switch the frequency very quickly. There are a number of sysfs file		switch the frequency very quickly.
	accessible parameters:
			Sysfs files:
	sampling_rate: measured in uS (10^-6 seconds), this is how often you
	want the kernel to look at the CPU usage and to make decisions on		* sampling_rate:
	what to do about the frequency. Typically this is set to values of
	around '10000' or more. It's default value is (cmp. with users-guide.txt):		Measured in uS (10^-6 seconds), this is how often you want the kernel
	transition_latency * 1000		to look at the CPU usage and to make decisions on what to do about the
	Be aware that transition latency is in ns and sampling_rate is in us, so you		frequency. Typically this is set to values of around '10000' or more.
	get the same sysfs value by default.		It's default value is (cmp. with users-guide.txt): transition_latency
	Sampling rate should always get adjusted considering the transition latency		* 1000. Be aware that transition latency is in ns and sampling_rate
	To set the sampling rate 750 times as high as the transition latency		is in us, so you get the same sysfs value by default. Sampling rate
	in the bash (as said, 1000 is default), do:		should always get adjusted considering the transition latency to set
	echo `$(($(cat cpuinfo_transition_latency) * 750 / 1000)) \		the sampling rate 750 times as high as the transition latency in the
	>ondemand/sampling_rate		bash (as said, 1000 is default), do:

	sampling_rate_min:		$ echo `$(($(cat cpuinfo_transition_latency) * 750 / 1000)) > ondemand/sampling_rate

			* sampling_rate_min:

	The sampling rate is limited by the HW transition latency:		The sampling rate is limited by the HW transition latency:
	transition_latency * 100		transition_latency * 100

	Or by kernel restrictions:		Or by kernel restrictions:
	If CONFIG_NO_HZ_COMMON is set, the limit is 10ms fixed.		- If CONFIG_NO_HZ_COMMON is set, the limit is 10ms fixed.
	If CONFIG_NO_HZ_COMMON is not set or nohz=off boot parameter is used, the		- If CONFIG_NO_HZ_COMMON is not set or nohz=off boot parameter is
	limits depend on the CONFIG_HZ option:		used, the limits depend on the CONFIG_HZ option:
	HZ=1000: min=20000us (20ms)		HZ=1000: min=20000us (20ms)
	HZ=250: min=80000us (80ms)		HZ=250: min=80000us (80ms)
	HZ=100: min=200000us (200ms)		HZ=100: min=200000us (200ms)

	The highest value of kernel and HW latency restrictions is shown and		The highest value of kernel and HW latency restrictions is shown and
	used as the minimum sampling rate.		used as the minimum sampling rate.

	up_threshold: defines what the average CPU usage between the samplings		* up_threshold:
	of 'sampling_rate' needs to be for the kernel to make a decision on
			This defines what the average CPU usage between the samplings of
			'sampling_rate' needs to be for the kernel to make a decision on
	whether it should increase the frequency. For example when it is set		whether it should increase the frequency. For example when it is set
	to its default value of '95' it means that between the checking		to its default value of '95' it means that between the checking
	intervals the CPU needs to be on average more than 95% in use to then		intervals the CPU needs to be on average more than 95% in use to then
	decide that the CPU frequency needs to be increased.		decide that the CPU frequency needs to be increased.

	ignore_nice_load: this parameter takes a value of '0' or '1'. When		* ignore_nice_load:
	set to '0' (its default), all processes are counted towards the
	'cpu utilisation' value. When set to '1', the processes that are		This parameter takes a value of '0' or '1'. When set to '0' (its
	run with a 'nice' value will not count (and thus be ignored) in the		default), all processes are counted towards the 'cpu utilisation'
	overall usage calculation. This is useful if you are running a CPU		value. When set to '1', the processes that are run with a 'nice'
	intensive calculation on your laptop that you do not care how long it		value will not count (and thus be ignored) in the overall usage
	takes to complete as you can 'nice' it and prevent it from taking part		calculation. This is useful if you are running a CPU intensive
	in the deciding process of whether to increase your CPU frequency.		calculation on your laptop that you do not care how long it takes to
			complete as you can 'nice' it and prevent it from taking part in the
	sampling_down_factor: this parameter controls the rate at which the		deciding process of whether to increase your CPU frequency.
	kernel makes a decision on when to decrease the frequency while running
	at top speed. When set to 1 (the default) decisions to reevaluate load		* sampling_down_factor:
	are made at the same interval regardless of current clock speed. But
	when set to greater than 1 (e.g. 100) it acts as a multiplier for the		This parameter controls the rate at which the kernel makes a decision
	scheduling interval for reevaluating load when the CPU is at its top		on when to decrease the frequency while running at top speed. When set
	speed due to high load. This improves performance by reducing the overhead		to 1 (the default) decisions to reevaluate load are made at the same
	of load evaluation and helping the CPU stay at its top speed when truly		interval regardless of current clock speed. But when set to greater
	busy, rather than shifting back and forth in speed. This tunable has no		than 1 (e.g. 100) it acts as a multiplier for the scheduling interval
			for reevaluating load when the CPU is at its top speed due to high
			load. This improves performance by reducing the overhead of load
			evaluation and helping the CPU stay at its top speed when truly busy,
			rather than shifting back and forth in speed. This tunable has no
	effect on behavior at lower speeds/lower CPU loads.		effect on behavior at lower speeds/lower CPU loads.

	powersave_bias: this parameter takes a value between 0 to 1000. It		* powersave_bias:
	defines the percentage (times 10) value of the target frequency that
	will be shaved off of the target. For example, when set to 100 -- 10%,		This parameter takes a value between 0 to 1000. It defines the
	when ondemand governor would have targeted 1000 MHz, it will target		percentage (times 10) value of the target frequency that will be
			shaved off of the target. For example, when set to 100 -- 10%, when
			ondemand governor would have targeted 1000 MHz, it will target
	1000 MHz - (10% of 1000 MHz) = 900 MHz instead. This is set to 0		1000 MHz - (10% of 1000 MHz) = 900 MHz instead. This is set to 0
	(disabled) by default.		(disabled) by default.

	When AMD frequency sensitivity powersave bias driver --		When AMD frequency sensitivity powersave bias driver --
	drivers/cpufreq/amd_freq_sensitivity.c is loaded, this parameter		drivers/cpufreq/amd_freq_sensitivity.c is loaded, this parameter
	defines the workload frequency sensitivity threshold in which a lower		defines the workload frequency sensitivity threshold in which a lower
	@@ -183,8 +197,8 @@ the performance of the workload running on a CPU will change when
	frequency changes. A workload with sensitivity of 0% (memory/IO-bound)		frequency changes. A workload with sensitivity of 0% (memory/IO-bound)
	will not perform any better on higher core frequency, whereas a		will not perform any better on higher core frequency, whereas a
	workload with sensitivity of 100% (CPU-bound) will perform better		workload with sensitivity of 100% (CPU-bound) will perform better
	higher the frequency. When the driver is loaded, this is set to 400		higher the frequency. When the driver is loaded, this is set to 400 by
	by default -- for CPUs running workloads with sensitivity value below		default -- for CPUs running workloads with sensitivity value below
	40%, a lower frequency is chosen. Unloading the driver or writing 0		40%, a lower frequency is chosen. Unloading the driver or writing 0
	will disable this feature.		will disable this feature.

	@@ -200,23 +214,28 @@ CPU. This behaviour more suitable in a battery powered environment.
	The governor is tweaked in the same manner as the "ondemand" governor		The governor is tweaked in the same manner as the "ondemand" governor
	through sysfs with the addition of:		through sysfs with the addition of:

	freq_step: this describes what percentage steps the cpu freq should be		* freq_step:
	increased and decreased smoothly by. By default the cpu frequency will
	increase in 5% chunks of your maximum cpu frequency. You can change this		This describes what percentage steps the cpu freq should be increased
	value to anywhere between 0 and 100 where '0' will effectively lock your		and decreased smoothly by. By default the cpu frequency will increase
	CPU at a speed regardless of its load whilst '100' will, in theory, make		in 5% chunks of your maximum cpu frequency. You can change this value
			to anywhere between 0 and 100 where '0' will effectively lock your CPU
			at a speed regardless of its load whilst '100' will, in theory, make
	it behave identically to the "ondemand" governor.		it behave identically to the "ondemand" governor.

	down_threshold: same as the 'up_threshold' found for the "ondemand"		* down_threshold:
	governor but for the opposite direction. For example when set to its
	default value of '20' it means that if the CPU usage needs to be below		Same as the 'up_threshold' found for the "ondemand" governor but for
	20% between samples to have the frequency decreased.		the opposite direction. For example when set to its default value of
			'20' it means that if the CPU usage needs to be below 20% between
			samples to have the frequency decreased.

			* sampling_down_factor:

	sampling_down_factor: similar functionality as in "ondemand" governor.		Similar functionality as in "ondemand" governor. But in
	But in "conservative", it controls the rate at which the kernel makes		"conservative", it controls the rate at which the kernel makes a
	a decision on when to decrease the frequency while running in any		decision on when to decrease the frequency while running in any speed.
	speed. Load for frequency increase is still evaluated every		Load for frequency increase is still evaluated every sampling rate.
	sampling rate.

	3. The Governor Interface in the CPUfreq Core		3. The Governor Interface in the CPUfreq Core
	=============================================		=============================================