From: Steve Muckle <steve.muckle@linaro.org>
To: Michael Turquette <mturquette@baylibre.com>,
Peter Zijlstra <peterz@infradead.org>
Cc: rjw@rjwysocki.net, linux-kernel@vger.kernel.org,
linux-pm@vger.kernel.org, Juri.Lelli@arm.com,
morten.rasmussen@arm.com, dietmar.eggemann@arm.com,
vincent.guittot@linaro.org,
Michael Turquette <mturquette+renesas@baylibre.com>,
Patrick Bellasi <patrick.bellasi@arm.com>
Subject: Re: [PATCH 4/8] cpufreq/schedutil: sysfs capacity margin tunable
Date: Tue, 15 Mar 2016 20:36:57 -0700 [thread overview]
Message-ID: <56E8D4D9.1060202@linaro.org> (raw)
In-Reply-To: <20160315223701.30639.43127@quark.deferred.io>
On 03/15/2016 03:37 PM, Michael Turquette wrote:
>>>> Yuck sysfs.. I would really rather we did not expose this per default.
>>>> > > > And certainly not in this weird form.
>>> > >
>>> > > I'm happy to change capacity_margin to up_threshold and use a
>>> > > percentage.
>>> > >
>>> > > The sysfs approach has two benefits. First, it is aligned with cpufreq
>>> > > user expectations. Second, there has been rough consensus that this
>>> > > value should be tunable and sysfs gets us there quickly and painlessly.
>>> > > We're already exporting rate_limit_us for schedutil via sysfs. Is there
>>> > > a better way interface you can recommend?
>> >
>> > It really depends on how tunable you want this to be. Do we always want
>> > this to be a tunable, or just now while we're playing about with the
>> > whole thing?
>
> I had considered this myself, and I really think that Steve and Juri
> should chime in as they have spent more time tuning and running the
> numbers.
>
> I'm inclined to think that a debug version would be good enough, as I
> don't imagine this value being changed at run-time by some userspace
> daemon or something.
>
> Then again, maybe this knob will be part of the mythical
> power-vs-performance slider?
Patrick Bellasi's schedtune series [0] (which I think is the referenced
mythical slider) aims to provide a more sophisticated interface for
tuning scheduler-driven frequency selection. In addition to a global
boost value it includes a cgroup controller as well for per-task tuning.
I would definitely expect the margin/boost value to be modified at
runtime, for example if the battery is running low, or the user wants
100% performance for a while, or the userspace framework wants to
temporarily tailor the performance level for a particular set of tasks, etc.
[0] http://article.gmane.org/gmane.linux.kernel/2022959
next prev parent reply other threads:[~2016-03-16 3:36 UTC|newest]
Thread overview: 61+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-03-14 5:22 [PATCH 0/8] schedutil enhancements Michael Turquette
2016-03-14 5:22 ` [PATCH 1/8] sched/cpufreq: remove cpufreq_trigger_update() Michael Turquette
2016-03-15 21:14 ` Peter Zijlstra
[not found] ` <20160315214545.30639.98727@quark.deferred.io>
2016-03-15 21:49 ` Peter Zijlstra
2016-03-16 8:00 ` Peter Zijlstra
2016-03-14 5:22 ` [PATCH 2/8] sched/fair: add margin to utilization update Michael Turquette
2016-03-15 21:16 ` Peter Zijlstra
[not found] ` <20160315212848.30639.38747@quark.deferred.io>
2016-03-15 21:43 ` Peter Zijlstra
2016-03-16 2:52 ` Steve Muckle
2016-03-16 22:12 ` Michael Turquette
2016-03-14 5:22 ` [PATCH 3/8] sched/cpufreq: new cfs capacity margin helpers Michael Turquette
2016-03-15 21:17 ` Peter Zijlstra
2016-03-14 5:22 ` [PATCH 4/8] cpufreq/schedutil: sysfs capacity margin tunable Michael Turquette
2016-03-15 21:20 ` Peter Zijlstra
[not found] ` <20160315214043.30639.75507@quark.deferred.io>
2016-03-15 21:48 ` Peter Zijlstra
[not found] ` <20160315223701.30639.43127@quark.deferred.io>
2016-03-16 3:36 ` Steve Muckle [this message]
2016-03-16 8:05 ` Peter Zijlstra
2016-03-16 10:02 ` Juri Lelli
2016-03-16 17:55 ` Steve Muckle
2016-03-16 22:05 ` Michael Turquette
2016-03-17 9:40 ` Juri Lelli
2016-03-17 13:55 ` Steve Muckle
2016-03-17 15:53 ` Patrick Bellasi
2016-03-17 17:54 ` Juri Lelli
2016-03-17 18:56 ` Michael Turquette
2016-03-17 22:34 ` Rafael J. Wysocki
2016-03-16 12:45 ` Rafael J. Wysocki
2016-03-16 22:03 ` Michael Turquette
2016-03-14 5:22 ` [PATCH 5/8] sched/cpufreq: pass sched class into cpufreq_update_util Michael Turquette
2016-03-15 21:25 ` Peter Zijlstra
[not found] ` <20160315220609.30639.67271@quark.deferred.io>
2016-03-16 3:55 ` Steve Muckle
2016-03-16 7:41 ` Peter Zijlstra
2016-03-16 8:29 ` Vincent Guittot
2016-03-16 8:53 ` Peter Zijlstra
2016-03-16 9:16 ` Vincent Guittot
2016-03-16 12:39 ` Rafael J. Wysocki
2016-03-16 13:10 ` Peter Zijlstra
2016-03-16 13:23 ` Rafael J. Wysocki
2016-03-16 13:43 ` Peter Zijlstra
2016-03-14 5:22 ` [PATCH 6/8] cpufreq/schedutil: sum per-sched class utilization Michael Turquette
2016-03-15 21:29 ` Peter Zijlstra
[not found] ` <20160315220951.30639.12872@quark.deferred.io>
2016-03-16 7:38 ` Peter Zijlstra
2016-03-16 18:20 ` Steve Muckle
2016-03-16 18:36 ` Peter Zijlstra
2016-03-16 19:12 ` Steve Muckle
2016-03-14 5:22 ` [PATCH 7/8] cpufreq: Frequency invariant scheduler load-tracking support Michael Turquette
2016-03-15 19:13 ` Dietmar Eggemann
2016-03-15 20:19 ` Michael Turquette
2016-03-15 21:32 ` Peter Zijlstra
2016-03-16 18:33 ` Dietmar Eggemann
2016-03-15 21:34 ` Peter Zijlstra
2016-03-14 5:22 ` [PATCH 8/8] sched: prefer cpufreq_scale_freq_capacity Michael Turquette
2016-03-15 19:13 ` Dietmar Eggemann
2016-03-15 20:46 ` Michael Turquette
2016-03-16 19:44 ` Dietmar Eggemann
2016-03-16 20:07 ` Peter Zijlstra
2016-03-16 21:32 ` Rafael J. Wysocki
2016-03-15 21:37 ` Peter Zijlstra
[not found] ` <20160315222721.30639.28332@quark.deferred.io>
2016-03-16 7:47 ` Peter Zijlstra
2016-03-16 12:41 ` Peter Zijlstra
2016-03-16 0:08 ` [PATCH 0/8] schedutil enhancements Rafael J. Wysocki
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=56E8D4D9.1060202@linaro.org \
--to=steve.muckle@linaro.org \
--cc=Juri.Lelli@arm.com \
--cc=dietmar.eggemann@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=morten.rasmussen@arm.com \
--cc=mturquette+renesas@baylibre.com \
--cc=mturquette@baylibre.com \
--cc=patrick.bellasi@arm.com \
--cc=peterz@infradead.org \
--cc=rjw@rjwysocki.net \
--cc=vincent.guittot@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).