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: <44fa1cee-08db-e4ab-e5ab-08d6fbd421d7@linux.alibaba.com> <20200303195245.GF2596@hirez.programming.kicks-ass.net>
In-Reply-To: <20200303195245.GF2596@hirez.programming.kicks-ass.net>
From:   Vincent Guittot <vincent.guittot@linaro.org>
Date:   Wed, 4 Mar 2020 09:45:44 +0100
Message-ID: <CAKfTPtCPTcfFs2OjaY9O2BtSZ_M6Gr-rGgFO2b8-wfvQWAZ1Zg@mail.gmail.com>
Subject: Re: [RFC PATCH] sched: fix the nonsense shares when load of cfs_rq is
 too, small
To:     Peter Zijlstra <peterz@infradead.org>
Cc:     =?UTF-8?B?546L6LSH?= <yun.wang@linux.alibaba.com>,
        Ingo Molnar <mingo@redhat.com>,
        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 Tue, 3 Mar 2020 at 20:52, Peter Zijlstra <peterz@infradead.org> wrote:
>
> On Tue, Mar 03, 2020 at 10:17:03PM +0800, =E7=8E=8B=E8=B4=87 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, which make
> > the group A 'cfs_rq->load.weight' 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_a=
vg);
> >   shares =3D (tg_shares * load) / tg_weight;
> >
> > Since the 'cfs_rq->load.weight' is too small, the load become 0
> > in here, although 'tg_shares' is 102400, shares of the se which
> > stand for group A on root cfs_rq become 2.
>
> Argh, because A->cfs_rq.load.weight is B->se.load.weight which is
> B->shares/nr_cpus.
>
> > While the se of D on root cfs_rq is far more bigger than 2, so it
> > wins the battle.
> >
> > This patch add a check on the zero load and make it as MIN_SHARES
> > to fix the nonsense shares, after applied the group C wins as
> > expected.
> >
> > Signed-off-by: Michael Wang <yun.wang@linux.alibaba.com>
> > ---
> >  kernel/sched/fair.c | 2 ++
> >  1 file changed, 2 insertions(+)
> >
> > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> > index 84594f8aeaf8..53d705f75fa4 100644
> > --- a/kernel/sched/fair.c
> > +++ b/kernel/sched/fair.c
> > @@ -3182,6 +3182,8 @@ static long calc_group_shares(struct cfs_rq *cfs_=
rq)
> >       tg_shares =3D READ_ONCE(tg->shares);
> >
> >       load =3D max(scale_load_down(cfs_rq->load.weight), cfs_rq->avg.lo=
ad_avg);
> > +     if (!load && cfs_rq->load.weight)
> > +             load =3D MIN_SHARES;
> >
> >       tg_weight =3D atomic_long_read(&tg->load_avg);
>
> Yeah, I suppose that'll do. Hurmph, wants a comment though.
>
> But that has me looking at other users of scale_load_down(), and doesn't
> at least update_tg_cfs_load() suffer the same problem?

yes and other places like the load_avg that will stay to 0 or the fact
that weight !=3D 0 is used to assume that se on enqueued and to not
remove the cfs from the leaf_cfs_rq_list even if load_avg is null