Date: Thu, 31 Mar 2016 11:34:08 +0200
From: Peter Zijlstra <peterz@infradead.org>
To: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Steve Muckle <steve.muckle@linaro.org>,
        Dietmar Eggemann <dietmar.eggemann@arm.com>,
        Ingo Molnar <mingo@redhat.com>,
        linux-kernel <linux-kernel@vger.kernel.org>,
        "linux-pm@vger.kernel.org" <linux-pm@vger.kernel.org>,
        "Rafael J. Wysocki" <rafael@kernel.org>,
        Morten Rasmussen <morten.rasmussen@arm.com>,
        Juri Lelli <Juri.Lelli@arm.com>,
        Patrick Bellasi <patrick.bellasi@arm.com>,
        Michael Turquette <mturquette@baylibre.com>
Subject: Re: [PATCH 1/2] sched/fair: move cpufreq hook to
 update_cfs_rq_load_avg()
Message-ID: <20160331093408.GB12845@twins.programming.kicks-ass.net>
References: <1458606068-7476-1-git-send-email-smuckle@linaro.org>
 <56F91D56.4020007@arm.com>
 <56F95D10.4070400@linaro.org>
 <56F97856.4040804@arm.com>
 <56F98832.3030207@linaro.org>
 <20160330193544.GD407@worktop>
 <CAKfTPtCPfd2-jGDn-9ZLjNKTni4czuvrkH7KFh5L2EtDeTQE5A@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAKfTPtCPfd2-jGDn-9ZLjNKTni4czuvrkH7KFh5L2EtDeTQE5A@mail.gmail.com>
User-Agent: Mutt/1.5.21 (2012-12-30)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1198
Lines: 23

On Thu, Mar 31, 2016 at 11:27:22AM +0200, Vincent Guittot wrote:
> On 30 March 2016 at 21:35, Peter Zijlstra <peterz@infradead.org> wrote:
> > On Mon, Mar 28, 2016 at 12:38:26PM -0700, Steve Muckle wrote:
> >> Without covering all the paths where CFS utilization changes it's
> >> possible to have to wait up to a tick to act on some changes, since the
> >> tick is the only guaranteed regularly-occurring instance of the hook.
> >> That's an unacceptable amount of latency IMO...
> >
> > Note that even with your patches that might still be the case. Remote
> > wakeups might not happen on the destination CPU at all, so it might not
> > be until the next tick (which always happens locally) that we'll
> > 'observe' the utilization change brought with the wakeups.
> >
> > We could force all the remote wakeups to IPI the destination CPU, but
> > that comes at a significant performance cost.
> 
> Isn't a reschedule ipi already sent in this case ?

In what case? Assuming you talk about a remove wakeup, no. Only if that
wakeup results in a preemption, which isn't a given.

And we really don't want to carry the 'has util increased' information
all the way down to where we make that decision.