Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <bb14528b-08c3-c4c0-5bcf-4bec1d75227a@linux.alibaba.com>
In-Reply-To: <bb14528b-08c3-c4c0-5bcf-4bec1d75227a@linux.alibaba.com>
From:   Vincent Guittot <vincent.guittot@linaro.org>
Date:   Fri, 6 Mar 2020 16:06:12 +0100
Message-ID: <CAKfTPtAt43DOT2AnRLyO5tqxDGhaCqFKOc1Ws+WSt2NLtGQ9UQ@mail.gmail.com>
Subject: Re: [PATCH] sched: avoid scale real weight down to zero
To:     =?UTF-8?B?546L6LSH?= <yun.wang@linux.alibaba.com>
Cc:     Ingo Molnar <mingo@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Juri Lelli <juri.lelli@redhat.com>,
        Dietmar Eggemann <dietmar.eggemann@arm.com>,
        Steven Rostedt <rostedt@goodmis.org>,
        Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>,
        "open list:SCHEDULER" <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, 5 Mar 2020 at 03:57, =E7=8E=8B=E8=B4=87 <yun.wang@linux.alibaba.com=
> wrote:
>
> During our testing, we found a case that shares no longer
> working correctly, the cgroup topology is like:
>
>   /sys/fs/cgroup/cpu/A          (shares=3D102400)
>   /sys/fs/cgroup/cpu/A/B        (shares=3D2)
>   /sys/fs/cgroup/cpu/A/B/C      (shares=3D1024)
>
>   /sys/fs/cgroup/cpu/D          (shares=3D1024)
>   /sys/fs/cgroup/cpu/D/E        (shares=3D1024)
>   /sys/fs/cgroup/cpu/D/E/F      (shares=3D1024)
>
> The same benchmark is running in group C & F, no other tasks are
> running, the benchmark is capable to consumed all the CPUs.
>
> We suppose the group C will win more CPU resources since it could
> enjoy all the shares of group A, but it's F who wins much more.
>
> The reason is because we have group B with shares as 2, since
> A->cfs_rq.load.weight =3D=3D B->se.load.weight =3D=3D B->shares/nr_cpus,
> so A->cfs_rq.load.weight become very small.
>
> And in calc_group_shares() we calculate shares as:
>
>   load =3D max(scale_load_down(cfs_rq->load.weight), cfs_rq->avg.load_avg=
);
>   shares =3D (tg_shares * load) / tg_weight;
>
> Since the 'cfs_rq->load.weight' is too small, the load become 0
> after scale down, although 'tg_shares' is 102400, shares of the se
> which stand for group A on root cfs_rq become 2.
>
> While the se of D on root cfs_rq is far more bigger than 2, so it
> wins the battle.
>
> Thus when scale_load_down() scale real weight down to 0, it's no
> longer telling the real story, the caller will have the wrong
> information and the calculation will be buggy.
>
> This patch add check in scale_load_down(), so the real weight will
> be >=3D MIN_SHARES after scale, after applied the group C wins as
> expected.
>
> Cc: Ben Segall <bsegall@google.com>
> Cc: Vincent Guittot <vincent.guittot@linaro.org>
> Suggested-by: Peter Zijlstra <peterz@infradead.org>
> Signed-off-by: Michael Wang <yun.wang@linux.alibaba.com>

Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
> ---
>  kernel/sched/sched.h | 8 +++++++-
>  1 file changed, 7 insertions(+), 1 deletion(-)
>
> diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
> index 2a0caf394dd4..75c283f22256 100644
> --- a/kernel/sched/sched.h
> +++ b/kernel/sched/sched.h
> @@ -118,7 +118,13 @@ extern long calc_load_fold_active(struct rq *this_rq=
, long adjust);
>  #ifdef CONFIG_64BIT
>  # define NICE_0_LOAD_SHIFT     (SCHED_FIXEDPOINT_SHIFT + SCHED_FIXEDPOIN=
T_SHIFT)
>  # define scale_load(w)         ((w) << SCHED_FIXEDPOINT_SHIFT)
> -# define scale_load_down(w)    ((w) >> SCHED_FIXEDPOINT_SHIFT)
> +# define scale_load_down(w) \
> +({ \
> +       unsigned long __w =3D (w); \
> +       if (__w) \
> +               __w =3D max(MIN_SHARES, __w >> SCHED_FIXEDPOINT_SHIFT); \
> +       __w; \
> +})
>  #else
>  # define NICE_0_LOAD_SHIFT     (SCHED_FIXEDPOINT_SHIFT)
>  # define scale_load(w)         (w)
> --
> 2.14.4.44.g2045bb6
>