From: Quentin Perret <quentin.perret@arm.com>
To: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
"Rafael J. Wysocki" <rjw@rjwysocki.net>,
linux-kernel <linux-kernel@vger.kernel.org>,
"open list:THERMAL" <linux-pm@vger.kernel.org>,
"gregkh@linuxfoundation.org" <gregkh@linuxfoundation.org>,
Ingo Molnar <mingo@redhat.com>,
Dietmar Eggemann <dietmar.eggemann@arm.com>,
Morten Rasmussen <morten.rasmussen@arm.com>,
Chris Redpath <chris.redpath@arm.com>,
Patrick Bellasi <patrick.bellasi@arm.com>,
Valentin Schneider <valentin.schneider@arm.com>,
Thara Gopinath <thara.gopinath@linaro.org>,
viresh kumar <viresh.kumar@linaro.org>,
Todd Kjos <tkjos@google.com>,
Joel Fernandes <joel@joelfernandes.org>,
"Cc: Steve Muckle" <smuckle@google.com>,
adharmap@quicinc.com, "Kannan, Saravana" <skannan@quicinc.com>,
pkondeti@codeaurora.org, Juri Lelli <juri.lelli@redhat.com>,
Eduardo Valentin <edubezval@gmail.com>,
Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>,
currojerez@riseup.net, Javi Merino <javi.merino@kernel.org>
Subject: Re: [PATCH v5 09/14] sched: Add over-utilization/tipping point indicator
Date: Thu, 2 Aug 2018 15:14:24 +0100 [thread overview]
Message-ID: <20180802141424.ju4jxxbk6pxw3kyq@queper01-lin> (raw)
In-Reply-To: <CAKfTPtDcoySXK0fBkDNy4wp1vsRxmiuAGT3CDZBh6Vnwyep2BA@mail.gmail.com>
On Thursday 02 Aug 2018 at 15:48:01 (+0200), Vincent Guittot wrote:
> On Thu, 2 Aug 2018 at 15:19, Quentin Perret <quentin.perret@arm.com> wrote:
> >
> > On Thursday 02 Aug 2018 at 15:08:01 (+0200), Peter Zijlstra wrote:
> > > On Thu, Aug 02, 2018 at 02:03:38PM +0100, Quentin Perret wrote:
> > > > On Thursday 02 Aug 2018 at 14:26:29 (+0200), Peter Zijlstra wrote:
> > > > > On Tue, Jul 24, 2018 at 01:25:16PM +0100, Quentin Perret wrote:
> > > > > > @@ -5100,8 +5118,17 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags)
> > > > > > update_cfs_group(se);
> > > > > > }
> > > > > >
> > > > > > - if (!se)
> > > > > > + if (!se) {
> > > > > > add_nr_running(rq, 1);
> > > > > > + /*
> > > > > > + * The utilization of a new task is 'wrong' so wait for it
> > > > > > + * to build some utilization history before trying to detect
> > > > > > + * the overutilized flag.
> > > > > > + */
> > > > > > + if (flags & ENQUEUE_WAKEUP)
> > > > > > + update_overutilized_status(rq);
> > > > > > +
> > > > > > + }
> > > > > >
> > > > > > hrtick_update(rq);
> > > > > > }
> > > > >
> > > > > That is a somewhat dodgy hack. There is no guarantee what so ever that
> > > > > when the task wakes next its history is any better. The comment doesn't
> > > > > reflect this I feel.
> > > >
> > > > AFAICT the main use-case here is to avoid re-enabling the load balance
> > > > and ruining all the task placement because of a tiny task. I don't
> > > > really see how we can do that differently ...
> > >
> > > Sure I realize that.. but it doesn't completely avoid it. Suppose this
> > > new task instantly blocks and wakes up again. Then its util signal will
> > > be exactly what you didn't want but we'll account it and cause the above
> > > scenario you wanted to avoid.
> >
> > That is true. ... I also realize now that this patch was written long
> > before util_est, and that also has an impact here, especially in the
> > scenario you described where the task blocks. So any wake-up after the
> > first enqueue will risk to overutilize the system, even if the task
> > blocked for ages.
> >
> > Hmm ...
>
> Does a init value set to 0 for util_avg for newly created task can
> help in EAS in this case ?
> Current initial value is computed to prevent packing newly created
> tasks on same CPUs because it hurts performance of some benches. In
> fact it somehow assumes that newly created task will use significant
> part of the remaining capacity of a CPU and want to spread tasks. In
> EAS case, it seems that it prefer to assume that newly created task
> are small and we can pack them and wait a bit to make sure the new
> task will be a big task and will overload the CPU
Good point, setting the util_avg to 0 for new tasks should help
filtering out those tiny tasks too. And that would match with the idea
of letting tasks build their history before looking at their util_avg ...
But there is one difference w.r.t frequency selection. The current code
won't mark the system overutilized, but will let sugov raise the
frequency when a new task is enqueued. So in case of a fork bomb, we
sort of fallback on the existing mainline strategy for both task
placement (because forkees don't go in find_energy_efficient_cpu) and
frequency selection. And I would argue this is the right thing to do
since EAS can't really help in this case.
Thoughts ?
Thanks,
Quentin
next prev parent reply other threads:[~2018-08-02 14:14 UTC|newest]
Thread overview: 72+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-07-24 12:25 [PATCH v5 00/14] Energy Aware Scheduling Quentin Perret
2018-07-24 12:25 ` [PATCH v5 01/14] sched: Relocate arch_scale_cpu_capacity Quentin Perret
2018-07-24 12:25 ` [PATCH v5 02/14] sched/cpufreq: Factor out utilization to frequency mapping Quentin Perret
2018-07-24 12:25 ` [PATCH v5 03/14] PM: Introduce an Energy Model management framework Quentin Perret
2018-08-09 21:52 ` Rafael J. Wysocki
2018-08-10 8:15 ` Quentin Perret
2018-08-10 8:41 ` Rafael J. Wysocki
2018-08-10 9:12 ` Quentin Perret
2018-08-10 11:13 ` Rafael J. Wysocki
2018-08-10 12:30 ` Quentin Perret
2018-08-12 9:49 ` Rafael J. Wysocki
2018-07-24 12:25 ` [PATCH v5 04/14] PM / EM: Expose the Energy Model in sysfs Quentin Perret
2018-07-24 12:25 ` [PATCH v5 05/14] sched/topology: Reference the Energy Model of CPUs when available Quentin Perret
2018-07-24 12:25 ` [PATCH v5 06/14] sched/topology: Lowest energy aware balancing sched_domain level pointer Quentin Perret
2018-07-26 16:00 ` Valentin Schneider
2018-07-26 17:01 ` Quentin Perret
2018-07-24 12:25 ` [PATCH v5 07/14] sched/topology: Introduce sched_energy_present static key Quentin Perret
2018-07-24 12:25 ` [PATCH v5 08/14] sched/fair: Clean-up update_sg_lb_stats parameters Quentin Perret
2018-07-24 12:25 ` [PATCH v5 09/14] sched: Add over-utilization/tipping point indicator Quentin Perret
2018-08-02 12:26 ` Peter Zijlstra
2018-08-02 13:03 ` Quentin Perret
2018-08-02 13:08 ` Peter Zijlstra
2018-08-02 13:18 ` Quentin Perret
2018-08-02 13:48 ` Vincent Guittot
2018-08-02 14:14 ` Quentin Perret [this message]
2018-08-02 15:14 ` Vincent Guittot
2018-08-02 15:30 ` Quentin Perret
2018-08-02 15:55 ` Vincent Guittot
2018-08-02 16:00 ` Quentin Perret
2018-08-02 16:07 ` Vincent Guittot
2018-08-02 16:10 ` Quentin Perret
2018-08-02 16:38 ` Vincent Guittot
2018-08-02 16:59 ` Quentin Perret
2018-08-03 7:48 ` Vincent Guittot
2018-08-03 8:18 ` Quentin Perret
2018-08-03 13:49 ` Vincent Guittot
2018-08-03 14:21 ` Vincent Guittot
2018-08-03 15:55 ` Quentin Perret
2018-08-06 8:40 ` Vincent Guittot
2018-08-06 9:43 ` Quentin Perret
2018-08-06 10:45 ` Vincent Guittot
2018-08-06 11:02 ` Quentin Perret
2018-08-06 10:08 ` Dietmar Eggemann
2018-08-06 10:33 ` Vincent Guittot
2018-08-06 12:29 ` Dietmar Eggemann
2018-08-06 12:37 ` Vincent Guittot
2018-08-06 13:20 ` Dietmar Eggemann
2018-08-09 9:30 ` Vincent Guittot
2018-08-09 9:38 ` Quentin Perret
2018-07-24 12:25 ` [PATCH v5 10/14] sched/cpufreq: Refactor the utilization aggregation method Quentin Perret
2018-07-30 19:35 ` skannan
2018-07-31 7:59 ` Quentin Perret
2018-07-31 19:31 ` skannan
2018-08-01 7:32 ` Rafael J. Wysocki
2018-08-01 8:23 ` Quentin Perret
2018-08-01 8:35 ` Rafael J. Wysocki
2018-08-01 9:23 ` Quentin Perret
2018-08-01 9:40 ` Rafael J. Wysocki
2018-08-02 13:04 ` Peter Zijlstra
2018-08-02 15:39 ` Quentin Perret
2018-08-03 13:04 ` Quentin Perret
2018-08-02 12:33 ` Peter Zijlstra
2018-08-02 12:45 ` Peter Zijlstra
2018-08-02 15:21 ` Quentin Perret
2018-08-02 17:36 ` Peter Zijlstra
2018-08-03 12:42 ` Quentin Perret
2018-07-24 12:25 ` [PATCH v5 11/14] sched/fair: Introduce an energy estimation helper function Quentin Perret
2018-07-24 12:25 ` [PATCH v5 12/14] sched/fair: Select an energy-efficient CPU on task wake-up Quentin Perret
2018-08-02 13:54 ` Peter Zijlstra
2018-08-02 16:21 ` Quentin Perret
2018-07-24 12:25 ` [PATCH v5 13/14] OPTIONAL: arch_topology: Start Energy Aware Scheduling Quentin Perret
2018-07-24 12:25 ` [PATCH v5 14/14] OPTIONAL: cpufreq: dt: Register an Energy Model Quentin Perret
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=20180802141424.ju4jxxbk6pxw3kyq@queper01-lin \
--to=quentin.perret@arm.com \
--cc=adharmap@quicinc.com \
--cc=chris.redpath@arm.com \
--cc=currojerez@riseup.net \
--cc=dietmar.eggemann@arm.com \
--cc=edubezval@gmail.com \
--cc=gregkh@linuxfoundation.org \
--cc=javi.merino@kernel.org \
--cc=joel@joelfernandes.org \
--cc=juri.lelli@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=morten.rasmussen@arm.com \
--cc=patrick.bellasi@arm.com \
--cc=peterz@infradead.org \
--cc=pkondeti@codeaurora.org \
--cc=rjw@rjwysocki.net \
--cc=skannan@quicinc.com \
--cc=smuckle@google.com \
--cc=srinivas.pandruvada@linux.intel.com \
--cc=thara.gopinath@linaro.org \
--cc=tkjos@google.com \
--cc=valentin.schneider@arm.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).