Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20220810223313.386614-1-libo.chen@oracle.com> <YvonlUOgMbla6dSh@worktop.programming.kicks-ass.net>
 <aac4a43d-4999-1da5-a617-b512ae0b3982@oracle.com>
In-Reply-To: <aac4a43d-4999-1da5-a617-b512ae0b3982@oracle.com>
From:   Vincent Guittot <vincent.guittot@linaro.org>
Date:   Wed, 17 Aug 2022 15:20:33 +0200
Message-ID: <CAKfTPtA3UTrsc3qZf_GB-QA7O7DJvoFqEk-VuWeFVRp8G_iUkg@mail.gmail.com>
Subject: Re: [PATCH 1/1] sched/fair: Fix inaccurate tally of ttwu_move_affine
To:     Libo Chen <libo.chen@oracle.com>
Cc:     Peter Zijlstra <peterz@infradead.org>, mingo@redhat.com,
        juri.lelli@redhat.com, dietmar.eggemann@arm.com, mgorman@suse.de,
        linux-kernel@vger.kernel.org,
        Daniel Jordan <daniel.m.jordan@oracle.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Mon, 15 Aug 2022 at 21:19, Libo Chen <libo.chen@oracle.com> wrote:
>
>
>
> On 8/15/22 04:01, Peter Zijlstra wrote:
> > On Wed, Aug 10, 2022 at 03:33:13PM -0700, Libo Chen wrote:
> >> There are scenarios where non-affine wakeups are incorrectly counted as
> >> affine wakeups by schedstats.
> >>
> >> When wake_affine_idle() returns prev_cpu which doesn't equal to
> >> nr_cpumask_bits, it will slip through the check: target == nr_cpumask_bits
> >> in wake_affine() and be counted as if target == this_cpu in schedstats.
> >>
> >> Replace target == nr_cpumask_bits with target != this_cpu to make sure
> >> affine wakeups are accurately tallied.
> >>
> >> Fixes: 806486c377e33 (sched/fair: Do not migrate if the prev_cpu is idle)
> >> Suggested-by: Daniel Jordan <daniel.m.jordan@oracle.com>
> >> Signed-off-by: Libo Chen <libo.chen@oracle.com>
> >> ---
> >>   kernel/sched/fair.c | 2 +-
> >>   1 file changed, 1 insertion(+), 1 deletion(-)
> >>
> >> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> >> index da388657d5ac..b179da4f8105 100644
> >> --- a/kernel/sched/fair.c
> >> +++ b/kernel/sched/fair.c
> >> @@ -6114,7 +6114,7 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p,
> >>              target = wake_affine_weight(sd, p, this_cpu, prev_cpu, sync);
> >>
> >>      schedstat_inc(p->stats.nr_wakeups_affine_attempts);
> >> -    if (target == nr_cpumask_bits)
> >> +    if (target != this_cpu)
> >>              return prev_cpu;
> >>
> >>      schedstat_inc(sd->ttwu_move_affine);
> > This not only changes the accounting but also the placement, no?
> No, it should only change the accounting. wake_affine() still returns
> prev_cpu if target equals to prev_cpu or nr_cpumask_bits, the same
> behavior as before.

Looks good to me

Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>

>
>
> Libo