From: Dietmar Eggemann <dietmar.eggemann@arm.com>
To: Thara Gopinath <thara.gopinath@linaro.org>,
mingo@redhat.com, peterz@infradead.org, ionela.voinescu@arm.com,
vincent.guittot@linaro.org, rui.zhang@intel.com,
qperret@google.com, daniel.lezcano@linaro.org,
viresh.kumar@linaro.org, rostedt@goodmis.org, will@kernel.org,
catalin.marinas@arm.com, sudeep.holla@arm.com,
juri.lelli@redhat.com, corbet@lwn.net
Cc: linux-kernel@vger.kernel.org, amit.kachhap@gmail.com,
javi.merino@kernel.org, amit.kucheria@verdurent.com
Subject: Re: [Patch v9 0/8] Introduce Thermal Pressure
Date: Mon, 10 Feb 2020 13:07:19 +0100 [thread overview]
Message-ID: <95164e1d-12e4-b155-f0d6-f869ee982aae@arm.com> (raw)
In-Reply-To: <1580250967-4386-1-git-send-email-thara.gopinath@linaro.org>
On 28/01/2020 23:35, Thara Gopinath wrote:
> Thermal governors can respond to an overheat event of a cpu by
> capping the cpu's maximum possible frequency. This in turn
> means that the maximum available compute capacity of the
> cpu is restricted. But today in the kernel, task scheduler is
> not notified of capping of maximum frequency of a cpu.
> In other words, scheduler is unaware of maximum capacity
> restrictions placed on a cpu due to thermal activity.
> This patch series attempts to address this issue.
> The benefits identified are better task placement among available
> cpus in event of overheating which in turn leads to better
> performance numbers.
>
> The reduction in the maximum possible capacity of a cpu due to a
> thermal event can be considered as thermal pressure. Instantaneous
> thermal pressure is hard to record and can sometime be erroneous
> as there can be mismatch between the actual capping of capacity
> and scheduler recording it. Thus solution is to have a weighted
> average per cpu value for thermal pressure over time.
> The weight reflects the amount of time the cpu has spent at a
> capped maximum frequency. Since thermal pressure is recorded as
> an average, it must be decayed periodically. Exisiting algorithm
> in the kernel scheduler pelt framework is re-used to calculate
> the weighted average. This patch series also defines a sysctl
> inerface to allow for a configurable decay period.
>
> Regarding testing, basic build, boot and sanity testing have been
> performed on db845c platform with debian file system.
> Further, dhrystone and hackbench tests have been
> run with the thermal pressure algorithm. During testing, due to
> constraints of step wise governor in dealing with big little systems,
> trip point 0 temperature was made assymetric between cpus in little
> cluster and big cluster; the idea being that
> big core will heat up and cpu cooling device will throttle the
> frequency of the big cores faster, there by limiting the maximum available
> capacity and the scheduler will spread out tasks to little cores as well.
>
> Test Results
>
> Hackbench: 1 group , 30000 loops, 10 runs
> Result SD
> (Secs) (% of mean)
> No Thermal Pressure 14.03 2.69%
> Thermal Pressure PELT Algo. Decay : 32 ms 13.29 0.56%
> Thermal Pressure PELT Algo. Decay : 64 ms 12.57 1.56%
> Thermal Pressure PELT Algo. Decay : 128 ms 12.71 1.04%
> Thermal Pressure PELT Algo. Decay : 256 ms 12.29 1.42%
> Thermal Pressure PELT Algo. Decay : 512 ms 12.42 1.15%
>
> Dhrystone Run Time : 20 threads, 3000 MLOOPS
> Result SD
> (Secs) (% of mean)
> No Thermal Pressure 9.452 4.49%
> Thermal Pressure PELT Algo. Decay : 32 ms 8.793 5.30%
> Thermal Pressure PELT Algo. Decay : 64 ms 8.981 5.29%
> Thermal Pressure PELT Algo. Decay : 128 ms 8.647 6.62%
> Thermal Pressure PELT Algo. Decay : 256 ms 8.774 6.45%
> Thermal Pressure PELT Algo. Decay : 512 ms 8.603 5.41%
What do we do on systems on which one Frequency domain spawns all the
CPUs (e.g. Hikey620)?
perf stat --null --repeat 10 -- perf bench sched messaging -g 10 -l 1000
# Running 'sched/messaging' benchmark:
# 20 sender and receiver processes per group
# 10 groups == 400 processes run
Total time: 4.697 [sec]
# Running 'sched/messaging' benchmark:
[ 8082.882751] hisi_thermal f7030700.tsensor: sensor <2> THERMAL ALARM: 66385 > 65000
# 20 sender and receiver processes per group
# 10 groups == 400 processes run
Total time: 4.910 [sec]
# Running 'sched/messaging' benchmark:
[ 8091.070386] CPU3 cpus=0-7 th_pressure=205
[ 8091.178390] CPU3 cpus=0-7 th_pressure=0
[ 8091.286389] CPU3 cpus=0-7 th_pressure=205
[ 8091.398397] CPU3 cpus=0-7 th_pressure=0
prev parent reply other threads:[~2020-02-10 12:09 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-01-28 22:35 [Patch v9 0/8] Introduce Thermal Pressure Thara Gopinath
2020-01-28 22:36 ` [Patch v9 1/8] sched/pelt: Add support to track thermal pressure Thara Gopinath
2020-02-13 12:29 ` Amit Kucheria
2020-02-13 14:11 ` Thara Gopinath
2020-02-13 14:41 ` Amit Kucheria
2020-01-28 22:36 ` [Patch v9 2/8] sched/topology: Add hook to read per cpu " Thara Gopinath
2020-01-28 22:36 ` [Patch v9 3/8] arm,arm64,drivers:Add infrastructure to store and update instantaneous " Thara Gopinath
2020-02-13 12:25 ` Amit Kucheria
2020-02-13 14:05 ` Thara Gopinath
2020-02-13 14:38 ` Amit Kucheria
2020-02-14 15:01 ` Thara Gopinath
2020-01-28 22:36 ` [Patch v9 4/8] sched/fair: Enable periodic update of average " Thara Gopinath
2020-01-28 22:36 ` [Patch v9 5/8] sched/fair: update cpu_capacity to reflect " Thara Gopinath
2020-02-13 12:47 ` Amit Kucheria
2020-02-13 14:12 ` Thara Gopinath
2020-02-13 13:39 ` Amit Kucheria
2020-02-14 14:52 ` Thara Gopinath
2020-01-28 22:36 ` [Patch v9 6/8] thermal/cpu-cooling: Update thermal pressure in case of a maximum frequency capping Thara Gopinath
2020-01-28 22:36 ` [Patch v9 7/8] sched/fair: Enable tuning of decay period Thara Gopinath
2020-01-28 23:56 ` Randy Dunlap
2020-02-03 12:07 ` Thara Gopinath
2020-02-03 15:55 ` Peter Zijlstra
2020-02-04 8:39 ` Dietmar Eggemann
2020-02-07 22:42 ` Thara Gopinath
2020-02-10 11:59 ` Dietmar Eggemann
2020-02-13 13:54 ` Thara Gopinath
2020-02-14 10:26 ` Dietmar Eggemann
2020-02-18 14:57 ` Thara Gopinath
2020-02-19 9:14 ` Dietmar Eggemann
2020-01-28 22:36 ` [Patch v9 8/8] arm64: Enable averaging of thermal pressure for arm64 based SoCs Thara Gopinath
2020-02-03 8:59 ` Dietmar Eggemann
2020-02-10 12:07 ` Dietmar Eggemann [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=95164e1d-12e4-b155-f0d6-f869ee982aae@arm.com \
--to=dietmar.eggemann@arm.com \
--cc=amit.kachhap@gmail.com \
--cc=amit.kucheria@verdurent.com \
--cc=catalin.marinas@arm.com \
--cc=corbet@lwn.net \
--cc=daniel.lezcano@linaro.org \
--cc=ionela.voinescu@arm.com \
--cc=javi.merino@kernel.org \
--cc=juri.lelli@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=qperret@google.com \
--cc=rostedt@goodmis.org \
--cc=rui.zhang@intel.com \
--cc=sudeep.holla@arm.com \
--cc=thara.gopinath@linaro.org \
--cc=vincent.guittot@linaro.org \
--cc=viresh.kumar@linaro.org \
--cc=will@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).