[Deepin-Kernel-SIG] [Upstream] [linux 6.6-y] sched/fair: Simplify Util_est #572
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here.
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
opsiff
force-pushed
the
linux-6.6.y-2025-01-20-sched
branch
from
df42434 to
035b255
Compare
opsiff
changed the title
…e capacity mainline inclusion from mainline-v6.7-rc1 category: performance Remove the rq::cpu_capacity_orig field and use arch_scale_cpu_capacity() instead. The scheduler uses 3 methods to get access to a CPU's max compute capacity: - arch_scale_cpu_capacity(cpu) which is the default way to get a CPU's capacity. - cpu_capacity_orig field which is periodically updated with arch_scale_cpu_capacity(). - capacity_orig_of(cpu) which encapsulates rq->cpu_capacity_orig. There is no real need to save the value returned by arch_scale_cpu_capacity() in struct rq. arch_scale_cpu_capacity() returns: - either a per_cpu variable. - or a const value for systems which have only one capacity. Remove rq::cpu_capacity_orig and use arch_scale_cpu_capacity() everywhere. No functional changes. Some performance tests on Arm64: - small SMP device (hikey): no noticeable changes - HMP device (RB5): hackbench shows minor improvement (1-2%) - large smp (thx2): hackbench and tbench shows minor improvement (1%) Signed-off-by: Vincent Guittot <[email protected]> Signed-off-by: Ingo Molnar <[email protected]> Reviewed-by: Dietmar Eggemann <[email protected]> Link: https://lore.kernel.org/r/[email protected] (cherry picked from commit 7bc2638)
mainline inclusion
from mainline-v6.8-rc1
category: bugfix
Create a new method to get a unique and fixed max frequency. Currently
cpuinfo.max_freq or the highest (or last) state of performance domain are
used as the max frequency when computing the frequency for a level of
utilization, but:
- cpuinfo_max_freq can change at runtime. boost is one example of
such change.
- cpuinfo.max_freq and last item of the PD can be different leading to
different results between cpufreq and energy model.
We need to save the reference frequency that has been used when computing
the CPUs capacity and use this fixed and coherent value to convert between
frequency and CPU's capacity.
In fact, we already save the frequency that has been used when computing
the capacity of each CPU. We extend the precision to save kHz instead of
MHz currently and we modify the type to be aligned with other variables
used when converting frequency to capacity and the other way.
[ mingo: Minor edits. ]
Signed-off-by: Vincent Guittot <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Tested-by: Lukasz Luba <[email protected]>
Reviewed-by: Lukasz Luba <[email protected]>
Acked-by: Sudeep Holla <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
(cherry picked from commit 9942cb2)
mainline inclusion from mainline-v6.8-rc1 category: bugfix cpuinfo.max_freq can change at runtime because of boost as an example. This implies that the value could be different from the frequency that has been used to compute the capacity of a CPU. The new arch_scale_freq_ref() returns a fixed and coherent frequency that can be used to compute the capacity for a given frequency. [ Also fix a arch_set_freq_scale() newline style wart in <linux/cpufreq.h>. ] Signed-off-by: Vincent Guittot <[email protected]> Signed-off-by: Ingo Molnar <[email protected]> Tested-by: Lukasz Luba <[email protected]> Reviewed-by: Lukasz Luba <[email protected]> Acked-by: Viresh Kumar <[email protected]> Acked-by: Rafael J. Wysocki <[email protected]> Link: https://lore.kernel.org/r/[email protected] (cherry picked from commit 599457b)
mainline inclusion from mainline-v6.8-rc1 category: bugfix cpuinfo.max_freq can change at runtime because of boost as an example. This implies that the value could be different than the one that has been used when computing the capacity of a CPU. The new arch_scale_freq_ref() returns a fixed and coherent reference frequency that can be used when computing a frequency based on utilization. Use this arch_scale_freq_ref() when available and fallback to policy otherwise. Signed-off-by: Vincent Guittot <[email protected]> Signed-off-by: Ingo Molnar <[email protected]> Tested-by: Lukasz Luba <[email protected]> Reviewed-by: Lukasz Luba <[email protected]> Reviewed-by: Dietmar Eggemann <[email protected]> Acked-by: Rafael J. Wysocki <[email protected]> Acked-by: Viresh Kumar <[email protected]> Link: https://lore.kernel.org/r/[email protected] (cherry picked from commit b3edde4)
mainline inclusion from mainline-v6.8-rc1 category: bugfix The last item of a performance domain is not always the performance point that has been used to compute CPU's capacity. This can lead to different target frequency compared with other part of the system like schedutil and would result in wrong energy estimation. A new arch_scale_freq_ref() is available to return a fixed and coherent frequency reference that can be used when computing the CPU's frequency for an level of utilization. Use this function to get this reference frequency. Energy model is never used without defining arch_scale_freq_ref() but can be compiled. Define a default arch_scale_freq_ref() returning 0 in such case. Signed-off-by: Vincent Guittot <[email protected]> Signed-off-by: Ingo Molnar <[email protected]> Tested-by: Lukasz Luba <[email protected]> Reviewed-by: Lukasz Luba <[email protected]> Link: https://lore.kernel.org/r/[email protected] (cherry picked from commit 15cbbd1)
mainline inclusion from mainline-v6.8-rc1 category: bugfix Lukasz Luba reported that a thread's util_est can significantly decrease as a result of sharing the CPU with other threads. The use case can be easily reproduced with a periodic task TA that runs 1ms and sleeps 100us. When the task is alone on the CPU, its max utilization and its util_est is around 888. If another similar task starts to run on the same CPU, TA will have to share the CPU runtime and its maximum utilization will decrease around half the CPU capacity (512) then TA's util_est will follow this new maximum trend which is only the result of sharing the CPU with others tasks. Such situation can be detected with runnable_avg wich is close or equal to util_avg when TA is alone, but increases above util_avg when TA shares the CPU with other threads and wait on the runqueue. [ We prefer an util_est that overestimate rather than under estimate because in 1st case we will not provide enough performance to the task which will remain under-provisioned, whereas in the other case we will create some idle time which will enable to reduce contention and as a result reduces the util_est so the overestimate will be transient whereas the underestimate will remain. ] [ mingo: Refined the changelog, added comments from the LKML discussion. ] Reported-by: Lukasz Luba <[email protected]> Signed-off-by: Vincent Guittot <[email protected]> Signed-off-by: Ingo Molnar <[email protected]> Link: https://lore.kernel.org/lkml/CAKfTPtDd-HhF-YiNTtL9i5k0PfJbF819Yxu4YquzfXgwi7voyw@mail.gmail.com/#t Link: https://lore.kernel.org/r/[email protected] Cc: Hongyan Xia <[email protected]> (cherry picked from commit 50181c0)
mainline inclusion from mainline-v6.8-rc1 category: performance sched_feat(UTIL_EST_FASTUP) has been added to easily disable the feature in order to check for possibly related regressions. After 3 years, it has never been used and no regression has been reported. Let's remove it and make fast increase a permanent behavior. Signed-off-by: Vincent Guittot <[email protected]> Signed-off-by: Ingo Molnar <[email protected]> Tested-by: Lukasz Luba <[email protected]> Reviewed-by: Lukasz Luba <[email protected]> Reviewed-by: Dietmar Eggemann <[email protected]> Reviewed-by: Hongyan Xia <[email protected]> Reviewed-by: Tang Yizhou <[email protected]> Reviewed-by: Yanteng Si <[email protected]> [for the Chinese translation] Reviewed-by: Alex Shi <[email protected]> Link: https://lore.kernel.org/r/[email protected] (cherry picked from commit 7736ae5)
mainline inclusion from mainline-v6.8-rc1 category: performance With UTIL_EST_FASTUP now being permanent, we can take advantage of the fact that the ewma jumps directly to a higher utilization at dequeue to simplify util_est and remove the enqueued field. Signed-off-by: Vincent Guittot <[email protected]> Signed-off-by: Ingo Molnar <[email protected]> Tested-by: Lukasz Luba <[email protected]> Reviewed-by: Lukasz Luba <[email protected]> Reviewed-by: Dietmar Eggemann <[email protected]> Reviewed-by: Hongyan Xia <[email protected]> Reviewed-by: Alex Shi <[email protected]> Link: https://lore.kernel.org/r/[email protected] (cherry picked from commit 11137d3)
mainline inclusion from mainline-v6.8-rc5 category: feature If "capacity-dmips-mhz" is not set, raw_capacity is null and we skip the normalization step which includes setting per_cpu capacity_freq_ref. Always register the notifier but skip the capacity normalization if raw_capacity is null. Fixes: 9942cb2 ("sched/topology: Add a new arch_scale_freq_ref() method") Signed-off-by: Vincent Guittot <[email protected]> Acked-by: Sudeep Holla <[email protected]> Tested-by: Pierre Gondois <[email protected]> Tested-by: Mark Brown <[email protected]> Tested-by: Paul Barker <[email protected]> Reviewed-by: Dietmar Eggemann <[email protected]> Tested-by: Dietmar Eggemann <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Greg Kroah-Hartman <[email protected]> (cherry picked from commit 98323e9)
mainline inclusion from mainline-v6.10-rc2 category: bugfix Commit: 7bc2638 ("sched/topology: Consolidate and clean up access to a CPU's max compute capacity") removed rq->cpu_capacity_orig in favor of using arch_scale_freq_capacity() calls. Export the underlying percpu symbol on x86 so that external trace point helper modules can be made to work again. Signed-off-by: Phil Auld <[email protected]> Signed-off-by: Ingo Molnar <[email protected]> Cc: Peter Zijlstra <[email protected]> Link: https://lore.kernel.org/r/[email protected] (cherry picked from commit d40605a)
opsiff
force-pushed
the
linux-6.6.y-2025-01-20-sched
branch
from
035b255 to
4e4cda7
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
mainline inclusion
Commit:
mainline-v6.7-rc1:
sched/topology: Consolidate and clean up access to a CPU's max compute capacity
mainline-v6.8-rc1:
sched/topology: Add a new arch_scale_freq_ref() method
cpufreq: Use the fixed and coherent frequency for scaling capacity
cpufreq/schedutil: Use a fixed reference frequency
energy_model: Use a fixed reference frequency
sched/pelt: Avoid underestimation of task utilization
sched/fair: Remove SCHED_FEAT(UTIL_EST_FASTUP, true)
sched/fair: Simplify util_est
mainline-v6.8-rc5:
topology: Set capacity_freq_ref in all cases
mainline-v6.10-rc2:
sched/x86: Export 'percpu arch_freq_scale'