From: Vincent Guittot <vincent.guittot@linaro.org>
To: Qais Yousef <qyousef@layalina.io>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>,
Peter Zijlstra <peterz@infradead.org>,
Kajetan Puchalski <kajetan.puchalski@arm.com>,
Jian-Min Liu <jian-min.liu@mediatek.com>,
Ingo Molnar <mingo@kernel.org>,
Morten Rasmussen <morten.rasmussen@arm.com>,
Vincent Donnefort <vdonnefort@google.com>,
Quentin Perret <qperret@google.com>,
Patrick Bellasi <patrick.bellasi@matbug.net>,
Abhijeet Dharmapurikar <adharmap@quicinc.com>,
Qais Yousef <qais.yousef@arm.com>,
linux-kernel@vger.kernel.org,
Jonathan JMChen <jonathan.jmchen@mediatek.com>
Subject: Re: [RFC PATCH 0/1] sched/pelt: Change PELT halflife at runtime
Date: Wed, 1 Mar 2023 11:39:06 +0100 [thread overview]
Message-ID: <CAKfTPtDVGcvmR5BoJpyoOBE19PcWZP+6NjSD7MnJyBAc7VMnmg@mail.gmail.com> (raw)
In-Reply-To: <20230223153700.55zydy7jyfwidkis@airbuntu>
On Thu, 23 Feb 2023 at 16:37, Qais Yousef <qyousef@layalina.io> wrote:
>
> On 02/09/23 17:16, Vincent Guittot wrote:
>
> > I don't see how util_est_faster can help this 1ms task here ? It's
> > most probably never be preempted during this 1ms. For such an Android
> > Graphics Pipeline short task, hasn't uclamp_min been designed for and
> > a better solution ?
>
> uclamp_min is being used in UI and helping there. But your mileage might vary
> with adoption still.
>
> The major motivation behind this is to help things like gaming as the original
> thread started. It can help UI and other use cases too. Android framework has
> a lot of context on the type of workload that can help it make a decision when
> this helps. And OEMs can have the chance to tune and apply based on the
> characteristics of their device.
>
> > IIUC how util_est_faster works, it removes the waiting time when
> > sharing cpu time with other tasks. So as long as there is no (runnable
> > but not running time), the result is the same as current util_est.
> > util_est_faster makes a difference only when the task alternates
> > between runnable and running slices.
> > Have you considered using runnable_avg metrics in the increase of cpu
> > freq ? This takes into the runnable slice and not only the running
> > time and increase faster than util_avg when tasks compete for the same
> > CPU
>
> Just to understand why we're heading into this direction now.
>
> AFAIU the desired outcome to have faster rampup time (and on HMP faster up
> migration) which both are tied to utilization signal.
>
> Wouldn't make the util response time faster help not just for rampup, but
> rampdown too?
>
> If we improve util response time, couldn't this mean we can remove util_est or
> am I missing something?
not sure because you still have a ramping step whereas util_est
directly gives you the final tager
>
> Currently we have util response which is tweaked by util_est and then that is
> tweaked further by schedutil with that 25% margin when maping util to
> frequency.
the 25% is not related to the ramping time but to the fact that you
always need some margin to cover unexpected events and estimation
error
>
> I think if we can allow improving general util response time by tweaking PELT
> HALFLIFE we can potentially remove util_est and potentially that magic 25%
> margin too.
>
> Why the approach of further tweaking util_est is better?
note that in this case it doesn't really tweak util_est but Dietmar
has taken into account runnable_avg to increase the freq in case of
contention
Also IIUC Dietmar's results, the problem seems more linked to the
selection of a higher freq than increasing the utilization;
runnable_avg tests give similar perf results than shorter half life
and better power consumption.
>
> Recently phoronix reported that schedutil behavior is suboptimal and I wonder
> if the response time is contributing to that
>
> https://www.phoronix.com/review/schedutil-quirky-2023
>
>
> Cheers
>
> --
> Qais Yousef
next prev parent reply other threads:[~2023-03-01 10:39 UTC|newest]
Thread overview: 61+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-08-29 5:54 [RFC PATCH 0/1] sched/pelt: Change PELT halflife at runtime Dietmar Eggemann
2022-08-29 5:54 ` [RFC PATCH 1/1] sched/pelt: Introduce PELT multiplier Dietmar Eggemann
2022-08-29 8:08 ` Peter Zijlstra
2022-08-29 10:02 ` Peter Zijlstra
2022-08-29 10:13 ` Vincent Guittot
2022-08-29 14:23 ` Quentin Perret
2022-08-29 14:34 ` Peter Zijlstra
2022-08-29 15:31 ` Quentin Perret
2022-08-29 15:48 ` Quentin Perret
2022-09-02 7:53 ` Dietmar Eggemann
2022-09-02 8:45 ` Peter Zijlstra
2022-09-06 5:49 ` Vincent Guittot
2022-09-08 6:50 ` Dietmar Eggemann
2022-09-02 7:53 ` Dietmar Eggemann
2022-09-02 8:45 ` Peter Zijlstra
2022-09-20 14:07 ` [RFC PATCH 0/1] sched/pelt: Change PELT halflife at runtime Jian-Min Liu
2022-09-28 17:09 ` Dietmar Eggemann
2022-09-29 9:47 ` Peter Zijlstra
2022-09-29 11:07 ` Dietmar Eggemann
2022-09-29 11:10 ` Kajetan Puchalski
2022-09-29 11:21 ` Peter Zijlstra
2022-09-29 14:41 ` Kajetan Puchalski
2022-10-03 22:57 ` Wei Wang
2022-10-04 9:33 ` Dietmar Eggemann
2022-10-05 16:57 ` Wei Wang
2022-11-07 13:41 ` Peter Zijlstra
2022-11-08 19:48 ` Qais Yousef
2022-11-09 15:49 ` Peter Zijlstra
2022-11-10 13:25 ` Qais Yousef
2023-02-07 10:29 ` Dietmar Eggemann
2023-02-09 16:16 ` Vincent Guittot
2023-02-17 13:54 ` Dietmar Eggemann
2023-02-20 13:54 ` Vincent Guittot
2023-02-21 9:29 ` Vincent Guittot
2023-02-22 20:28 ` Dietmar Eggemann
2023-03-01 10:24 ` Vincent Guittot
2023-02-22 20:13 ` Dietmar Eggemann
2023-03-02 19:36 ` Dietmar Eggemann
2023-02-20 10:13 ` Peter Zijlstra
2023-02-20 13:39 ` Vincent Guittot
2023-02-23 15:37 ` Qais Yousef
2023-03-01 10:39 ` Vincent Guittot [this message]
2023-03-01 17:24 ` Qais Yousef
2023-03-02 8:00 ` Vincent Guittot
2023-03-02 19:39 ` Dietmar Eggemann
2023-03-06 19:11 ` Qais Yousef
2023-03-07 13:22 ` Vincent Guittot
2023-03-11 16:55 ` Qais Yousef
2023-03-23 16:29 ` Dietmar Eggemann
2023-04-03 14:45 ` Qais Yousef
2023-04-06 15:58 ` Dietmar Eggemann
2023-04-11 17:51 ` Qais Yousef
2022-11-09 15:18 ` Lukasz Luba
2022-11-10 11:16 ` Dietmar Eggemann
2022-11-10 13:05 ` Peter Zijlstra
2022-11-10 14:59 ` Dietmar Eggemann
2022-11-10 17:51 ` Peter Zijlstra
2022-11-30 18:14 ` Dietmar Eggemann
2022-12-01 13:37 ` Kajetan Puchalski
2022-11-10 12:45 ` Kajetan Puchalski
2022-11-07 9:41 ` Jian-Min Liu (劉建旻)
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=CAKfTPtDVGcvmR5BoJpyoOBE19PcWZP+6NjSD7MnJyBAc7VMnmg@mail.gmail.com \
--to=vincent.guittot@linaro.org \
--cc=adharmap@quicinc.com \
--cc=dietmar.eggemann@arm.com \
--cc=jian-min.liu@mediatek.com \
--cc=jonathan.jmchen@mediatek.com \
--cc=kajetan.puchalski@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=morten.rasmussen@arm.com \
--cc=patrick.bellasi@matbug.net \
--cc=peterz@infradead.org \
--cc=qais.yousef@arm.com \
--cc=qperret@google.com \
--cc=qyousef@layalina.io \
--cc=vdonnefort@google.com \
/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).