LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Quentin Perret <firstname.lastname@example.org>
To: Viresh Kumar <email@example.com>
Cc: Dietmar Eggemann <firstname.lastname@example.org>,
Peter Zijlstra <email@example.com>,
Ingo Molnar <firstname.lastname@example.org>,
email@example.com, Pavan Kondeti <firstname.lastname@example.org>,
"Rafael J . Wysocki" <email@example.com>,
Juri Lelli <firstname.lastname@example.org>,
Joel Fernandes <email@example.com>,
Patrick Bellasi <firstname.lastname@example.org>
Subject: Re: [PATCH] Revert "cpufreq: schedutil: Don't restrict kthread to related_cpus unnecessarily"
Date: Tue, 8 May 2018 11:02:27 +0100 [thread overview]
Message-ID: <20180508100227.GB3752@e108498-lin.cambridge.arm.com> (raw)
On Tuesday 08 May 2018 at 15:15:26 (+0530), Viresh Kumar wrote:
> On 08-05-18, 11:09, Dietmar Eggemann wrote:
> > This would make sure that the kthreads are bound to the correct set of cpus
> > for platforms with those cpufreq drivers (cpufreq-dt (h960), scmi-cpufreq,
> > scpi-cpufreq) but it will also change the logic (e.g.
> > sugov_should_update_freq() -> cpufreq_can_do_remote_dvfs()).
> Yeah, I misunderstood your patch a bit. So you are not disabling
> remote updates but only limiting the CPUs where the kthread runs.
> That still looks to be a big little specific problem to me right now
> and I am not sure why should we specially handle these kthreads ?
> Isn't the same true for any other threads/tasks in the kernel which
> may end up running on big CPUs ? And this problem still occurs with
> the EAS patches applied ? As I thought we may end up keeping such
> small tasks on little cores then.
The sugov kthreads are DL tasks so they're not impacted by EAS. But even
if you take EAS out of the picture, those kthreads are assigned to a
"random" CPU at boot time and stay there forever (because that's how DL
works). Is this what we want ?
> > I'm still struggling to understand when a driver/platform should set
> > dvfs_possible_from_any_cpu to true and what the actual benefit would be.
> Ideally it should be set by default for all ARM platforms at least
> which have more than one cpufreq policy,
Looking at the commit you mention below it seems that you did the
testing on the old hikey which has only one cpufreq policy. Did you try
on other platforms as well ?
> as there is no hardware
> limitation for changing frequency from other CPUs. If you look at the
> commit logs of patches which added remote updates, you will see
> interesting cases where this can be very useful.
> commit 674e75411fc2 ("sched: cpufreq: Allow remote cpufreq callbacks")
next prev parent reply other threads:[~2018-05-08 10:02 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-05-08 7:33 Dietmar Eggemann
2018-05-08 8:22 ` Viresh Kumar
2018-05-08 9:09 ` Dietmar Eggemann
2018-05-08 9:42 ` Quentin Perret
2018-05-13 5:19 ` Joel Fernandes
2018-05-17 19:10 ` Saravana Kannan
2018-05-17 19:13 ` Joel Fernandes
2018-05-08 9:45 ` Viresh Kumar
2018-05-08 10:02 ` Quentin Perret [this message]
2018-05-08 10:34 ` Viresh Kumar
2018-05-08 11:00 ` Quentin Perret
2018-05-08 11:14 ` Viresh Kumar
2018-05-08 11:24 ` Quentin Perret
2018-05-08 12:20 ` Juri Lelli
2018-05-08 20:36 ` Rafael J. Wysocki
2018-05-09 4:55 ` Viresh Kumar
2018-05-08 10:36 ` Dietmar Eggemann
2018-05-08 10:53 ` Viresh Kumar
2018-05-08 12:17 ` Juri Lelli
2018-05-09 4:55 ` Viresh Kumar
2018-05-17 10:32 ` Rafael J. Wysocki
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:
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--subject='Re: [PATCH] Revert "cpufreq: schedutil: Don'\''t restrict kthread to related_cpus unnecessarily"' \
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
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).