[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20190729161023.GK31398@hirez.programming.kicks-ass.net>
Date: Mon, 29 Jul 2019 18:10:23 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Dietmar Eggemann <dietmar.eggemann@....com>
Cc: luca abeni <luca.abeni@...tannapisa.it>,
Ingo Molnar <mingo@...nel.org>,
Juri Lelli <juri.lelli@...hat.com>,
Daniel Bristot de Oliveira <bristot@...hat.com>,
Valentin Schneider <Valentin.Schneider@....com>,
Qais Yousef <Qais.Yousef@....com>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/5] sched/deadline: Fix double accounting of rq/running
bw in push_dl_task()
On Mon, Jul 29, 2019 at 09:59:28AM +0100, Dietmar Eggemann wrote:
> On 7/26/19 11:11 AM, luca abeni wrote:
> > Hi Dietmar,
> >
> > On Fri, 26 Jul 2019 09:27:52 +0100
> > Dietmar Eggemann <dietmar.eggemann@....com> wrote:
> >
> >> push_dl_task() always calls deactivate_task() with flags=0 which sets
> >> p->on_rq=TASK_ON_RQ_MIGRATING.
> >
> > Uhm... This is a recent change in the deactivate_task() behaviour,
> > right? Because I tested SCHED_DEADLINE a lot, but I've never seen this
> > issue :)
>
> Looks like it was v5.2 commit 7dd778841164 ("sched/core: Unify p->on_rq
> updates").
Argh, that wasn't intentional (obviously); please post v2 with that as
Fixes and I'll get it in sched/urgent and stable.
Powered by blists - more mailing lists