[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <10626645.IplgtjqW4B@vostro.rjw.lan>
Date: Tue, 13 Sep 2016 00:25:54 +0200
From: "Rafael J. Wysocki" <rjw@...ysocki.net>
To: Peter Zijlstra <peterz@...radead.org>,
Steve Muckle <steve.muckle@...aro.org>
Cc: Ingo Molnar <mingo@...hat.com>,
"Rafael J . Wysocki" <rafael@...nel.org>,
linux-kernel@...r.kernel.org, linux-pm@...r.kernel.org,
Vincent Guittot <vincent.guittot@...aro.org>,
Morten Rasmussen <morten.rasmussen@....com>,
Dietmar Eggemann <dietmar.eggemann@....com>,
Juri Lelli <Juri.Lelli@....com>,
Patrick Bellasi <patrick.bellasi@....com>,
Steve Muckle <smuckle@...aro.org>
Subject: Re: [PATCH 1/2] sched: cpufreq: ignore SMT when determining max cpu capacity
On Wednesday, August 31, 2016 04:34:37 PM Peter Zijlstra wrote:
> On Wed, Aug 31, 2016 at 03:27:30AM +0200, Rafael J. Wysocki wrote:
> > On Friday, August 26, 2016 11:40:47 AM Steve Muckle wrote:
> > > PELT does not consider SMT when scaling its utilization values via
> > > arch_scale_cpu_capacity(). The value in rq->cpu_capacity_orig does
> > > take SMT into consideration though and therefore may be smaller than
> > > the utilization reported by PELT.
> > >
> > > On an Intel i7-3630QM for example rq->cpu_capacity_orig is 589 but
> > > util_avg scales up to 1024. This means that a 50% utilized CPU will show
> > > up in schedutil as ~86% busy.
> > >
> > > Fix this by using the same CPU scaling value in schedutil as that which
> > > is used by PELT.
> > >
> > > Signed-off-by: Steve Muckle <smuckle@...aro.org>
> >
> > This is fine by me.
> >
> > Peter, any objections?
>
> No, looks good. Thanks
OK, thanks!
Applied.
Powered by blists - more mailing lists