From: Thara Gopinath <thara.gopinath@linaro.org>
To: Lukasz Luba <l.luba@partner.samsung.com>, linux-kernel@vger.kernel.org
Cc: mingo@redhat.com, peterz@infradead.org, rui.zhang@intel.com,
gregkh@linuxfoundation.org, rafael@kernel.org,
amit.kachhap@gmail.com, viresh.kumar@linaro.org,
javi.merino@kernel.org, edubezval@gmail.com,
daniel.lezcano@linaro.org, linux-pm@vger.kernel.org,
quentin.perret@arm.com, ionela.voinescu@arm.com,
vincent.guittot@linaro.org,
Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Subject: Re: [RFC PATCH 0/7] Introduce thermal pressure
Date: Wed, 10 Oct 2018 13:30:57 -0400 [thread overview]
Message-ID: <5BBE3751.7000908@linaro.org> (raw)
In-Reply-To: <20181010153553eucas1p1b8f74f4aa45751ef029805fd118affc1~cSUmU58-F2963929639eucas1p1L@eucas1p1.samsung.com>
Hello Lukasz,
On 10/10/2018 11:35 AM, Lukasz Luba wrote:
> Hi Thara,
>
> I have run it on Exynos5433 mainline.
> When it is enabled with step_wise thermal governor,
> some of my tests are showing ~30-50% regression (i.e. hackbench),
> dhrystone ~10%.
That is interesting. If I understand correctly, dhrystone spawns 16
threads or so and floods the system. In "theory", such a test should not
see any performance improvement and degradation. What is the thermal
activity like in your system? I will try running one of these tests on
hikey960.
>
> Could you tell me which thermal governor was used in your case?
> Please also share the name of that benchmark, i will give it a try.
> Is it single threaded compute-intensive?
Step-wise governor.
I use aobench which is part of phoronix-test-suite.
Regards
Thara
>
> Regards,
> Lukasz
>
> On 10/09/2018 06:24 PM, Thara Gopinath wrote:
>> Thermal governors can respond to an overheat event for 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 linux kernel, in event of maximum
>> frequency capping of a cpu, the maximum available compute
>> capacity of the cpu is not adjusted at all. In other words, scheduler
>> is unware maximum cpu capacity restrictions placed 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 delta between the maximum possible capacity of a cpu and
>> maximum available capacity of a cpu due to 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. To accumulate, average and
>> appropriately decay thermal pressure, this patch series uses pelt
>> signals and reuses the available framework that does a similar
>> bookkeeping of rt/dl task utilization.
>>
>> Regarding testing, basic build, boot and sanity testing have been
>> performed on hikey960 mainline kernel with debian file system.
>> Further aobench (An occlusion renderer for benchmarking realworld
>> floating point performance) showed the following results on hikey960
>> with debain.
>>
>> Result Standard Standard
>> (Time secs) Error Deviation
>> Hikey 960 - no thermal pressure applied 138.67 6.52 11.52%
>> Hikey 960 - thermal pressure applied 122.37 5.78 11.57%
>>
>> Thara Gopinath (7):
>> sched/pelt: Add option to make load and util calculations frequency
>> invariant
>> sched/pelt.c: Add support to track thermal pressure
>> sched: Add infrastructure to store and update instantaneous thermal
>> pressure
>> sched: Initialize per cpu thermal pressure structure
>> sched/fair: Enable CFS periodic tick to update thermal pressure
>> sched/fair: update cpu_capcity to reflect thermal pressure
>> thermal/cpu-cooling: Update thermal pressure in case of a maximum
>> frequency capping
>>
>> drivers/base/arch_topology.c | 1 +
>> drivers/thermal/cpu_cooling.c | 20 ++++++++++++-
>> include/linux/sched.h | 14 +++++++++
>> kernel/sched/Makefile | 2 +-
>> kernel/sched/core.c | 2 ++
>> kernel/sched/fair.c | 4 +++
>> kernel/sched/pelt.c | 40 ++++++++++++++++++--------
>> kernel/sched/pelt.h | 7 +++++
>> kernel/sched/sched.h | 1 +
>> kernel/sched/thermal.c | 66 +++++++++++++++++++++++++++++++++++++++++++
>> kernel/sched/thermal.h | 13 +++++++++
>> 11 files changed, 157 insertions(+), 13 deletions(-)
>> create mode 100644 kernel/sched/thermal.c
>> create mode 100644 kernel/sched/thermal.h
>>
--
Regards
Thara
next prev parent reply other threads:[~2018-10-10 17:30 UTC|newest]
Thread overview: 67+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <CGME20181009162509epcas1p4fdd2e23039caa24586a4a52c6d2e7336@epcas1p4.samsung.com>
2018-10-09 16:24 ` [RFC PATCH 0/7] Introduce thermal pressure Thara Gopinath
2018-10-09 16:24 ` [RFC PATCH 1/7] sched/pelt.c: Add option to make load and util calculations frequency invariant Thara Gopinath
2018-10-09 16:24 ` [RFC PATCH 2/7] sched/pelt.c: Add support to track thermal pressure Thara Gopinath
2018-10-09 16:24 ` [RFC PATCH 3/7] sched: Add infrastructure to store and update instantaneous " Thara Gopinath
2018-10-09 16:24 ` [RFC PATCH 4/7] sched: Initialize per cpu thermal pressure structure Thara Gopinath
2018-10-09 16:25 ` [RFC PATCH 5/7] sched/fair: Enable CFS periodic tick to update thermal pressure Thara Gopinath
2018-12-04 15:43 ` Vincent Guittot
2018-10-09 16:25 ` [RFC PATCH 6/7] sched/fair: update cpu_capcity to reflect " Thara Gopinath
2018-10-10 5:57 ` Javi Merino
2018-10-10 14:22 ` Thara Gopinath
2018-10-09 16:25 ` [RFC PATCH 7/7] thermal/cpu-cooling: Update thermal pressure in case of a maximum frequency capping Thara Gopinath
2018-10-10 5:44 ` [RFC PATCH 0/7] Introduce thermal pressure Javi Merino
2018-10-10 14:15 ` Thara Gopinath
2018-10-10 6:17 ` Ingo Molnar
2018-10-10 8:29 ` Quentin Perret
2018-10-10 8:50 ` Vincent Guittot
2018-10-10 9:55 ` Quentin Perret
2018-10-10 10:14 ` Vincent Guittot
2018-10-10 10:36 ` Quentin Perret
2018-10-10 12:04 ` Vincent Guittot
2018-10-10 12:23 ` Juri Lelli
2018-10-10 12:34 ` Vincent Guittot
2018-10-10 12:50 ` Juri Lelli
2018-10-10 13:08 ` Vincent Guittot
2018-10-10 13:34 ` Juri Lelli
2018-10-10 13:38 ` Vincent Guittot
2018-10-10 17:08 ` Thara Gopinath
2018-10-10 13:11 ` Quentin Perret
2018-10-10 13:05 ` Quentin Perret
2018-10-10 13:27 ` Vincent Guittot
2018-10-10 13:47 ` Quentin Perret
2018-10-10 15:19 ` Vincent Guittot
2018-10-10 16:15 ` Ionela Voinescu
2018-10-10 17:03 ` Thara Gopinath
2018-10-10 15:43 ` Thara Gopinath
2018-10-16 7:33 ` Ingo Molnar
2018-10-16 9:28 ` Lukasz Luba
2018-10-17 16:21 ` Thara Gopinath
2018-10-18 6:48 ` Ingo Molnar
2018-10-18 7:08 ` Rafael J. Wysocki
2018-10-18 7:50 ` Ingo Molnar
2018-10-18 8:14 ` Rafael J. Wysocki
2018-10-18 9:35 ` [PATCH 1/2] sched/cpufreq: Reorganize the cpufreq files Daniel Lezcano
2018-10-18 9:35 ` [PATCH 2/2] sched/cpufreq: Add the SPDX tags Daniel Lezcano
2018-10-18 9:42 ` [PATCH 1/2] sched/cpufreq: Reorganize the cpufreq files Rafael J. Wysocki
2018-10-18 9:54 ` Daniel Lezcano
2018-10-18 10:06 ` Rafael J. Wysocki
2018-10-18 10:13 ` Daniel Lezcano
2018-10-18 9:45 ` Daniel Lezcano
2018-10-19 5:24 ` kbuild test robot
2018-10-19 5:52 ` kbuild test robot
2018-10-18 9:44 ` [PATCH V2 " Daniel Lezcano
2018-10-18 9:44 ` [PATCH V2 2/2] sched/cpufreq: Add the SPDX tags Daniel Lezcano
2018-10-18 16:17 ` [RFC PATCH 0/7] Introduce thermal pressure Thara Gopinath
2018-10-19 8:02 ` Ingo Molnar
2018-10-19 11:29 ` Valentin Schneider
2018-10-10 15:35 ` Lukasz Luba
2018-10-10 16:54 ` Daniel Lezcano
2018-10-11 7:35 ` Lukasz Luba
2018-10-11 8:23 ` Daniel Lezcano
2018-10-12 9:37 ` Lukasz Luba
2018-10-10 17:30 ` Thara Gopinath [this message]
2018-10-11 11:10 ` Lukasz Luba
2018-10-16 17:11 ` Vincent Guittot
2018-10-17 16:24 ` Thara Gopinath
2018-10-18 8:00 ` Lukasz Luba
2018-10-18 8:12 ` Lukasz Luba
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=5BBE3751.7000908@linaro.org \
--to=thara.gopinath@linaro.org \
--cc=amit.kachhap@gmail.com \
--cc=b.zolnierkie@samsung.com \
--cc=daniel.lezcano@linaro.org \
--cc=edubezval@gmail.com \
--cc=gregkh@linuxfoundation.org \
--cc=ionela.voinescu@arm.com \
--cc=javi.merino@kernel.org \
--cc=l.luba@partner.samsung.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=quentin.perret@arm.com \
--cc=rafael@kernel.org \
--cc=rui.zhang@intel.com \
--cc=vincent.guittot@linaro.org \
--cc=viresh.kumar@linaro.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).