Received-SPF: pass (google.com: domain of linux-kernel+bounces-127412-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Precedence: bulk
MIME-Version: 1.0
References: <20240315015916.21545-1-daweilics@gmail.com> <ZgqOEJ5sCANkkk5N@linux.ibm.com>
In-Reply-To: <ZgqOEJ5sCANkkk5N@linux.ibm.com>
From: Dawei Li <daweilics@gmail.com>
Date: Mon, 1 Apr 2024 22:47:36 -0700
Message-ID: <CAG5MgCq3GT=CVj7Hz8rUMfNG1c9ypVsTSDKNESHV9tY_qWSt2g@mail.gmail.com>
Subject: Re: [PATCH v2] sched/fair: fix initial util_avg calculation
To: Vishal Chourasia <vishalc@linux.ibm.com>
Cc: Ingo Molnar <mingo@redhat.com>, Peter Zijlstra <peterz@infradead.org>, 
	Juri Lelli <juri.lelli@redhat.com>, Vincent Guittot <vincent.guittot@linaro.org>, 
	Dietmar Eggemann <dietmar.eggemann@arm.com>, Steven Rostedt <rostedt@goodmis.org>, 
	Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>, 
	Daniel Bristot de Oliveira <bristot@redhat.com>, Valentin Schneider <vschneid@redhat.com>, linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hi Vishal

Thanks for the comment!
Do you suggest using scale_load_down() in place of se_weight()?
It's a soft bug we should fix one way or another before what the
comment mentions really happens.
I am actually confused that we have both se_weight() and
scale_load_down(), and they do the same thing.

Best regards,
Dawei

On Mon, Apr 1, 2024 at 3:36=E2=80=AFAM Vishal Chourasia <vishalc@linux.ibm.=
com> wrote:
>
> On Thu, Mar 14, 2024 at 06:59:16PM -0700, Dawei Li wrote:
> > Change se->load.weight to se_weight(se) in the calculation for the
> > initial util_avg to avoid unnecessarily inflating the util_avg by 1024
> > times.
> >
> > The reason is that se->load.weight has the unit/scale as the scaled-up
> > load, while cfs_rg->avg.load_avg has the unit/scale as the true task
> > weight (as mapped directly from the task's nice/priority value). With
> > CONFIG_32BIT, the scaled-up load is equal to the true task weight. With
> > CONFIG_64BIT, the scaled-up load is 1024 times the true task weight.
> > Thus, the current code may inflate the util_avg by 1024 times. The
> > follow-up capping will not allow the util_avg value to go wild. But the
> > calculation should have the correct logic.
> >
> > Signed-off-by: Dawei Li <daweilics@gmail.com>
> > ---
> > Changes in v2:
> > - update the commit message
> > ---
> >  kernel/sched/fair.c | 5 +++--
> >  1 file changed, 3 insertions(+), 2 deletions(-)
> >
> > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> > index a19ea290b790..5f98f639bdb9 100644
> > --- a/kernel/sched/fair.c
> > +++ b/kernel/sched/fair.c
> > @@ -1031,7 +1031,8 @@ void init_entity_runnable_average(struct sched_en=
tity *se)
> >   * With new tasks being created, their initial util_avgs are extrapola=
ted
> >   * based on the cfs_rq's current util_avg:
> >   *
> > - *   util_avg =3D cfs_rq->util_avg / (cfs_rq->load_avg + 1) * se.load.=
weight
> > + *   util_avg =3D cfs_rq->avg.util_avg / (cfs_rq->avg.load_avg + 1)
> > + *           * se_weight(se)
> >   *
> >   * However, in many cases, the above util_avg does not give a desired
> >   * value. Moreover, the sum of the util_avgs may be divergent, such
> > @@ -1078,7 +1079,7 @@ void post_init_entity_util_avg(struct task_struct=
 *p)
> >
> >       if (cap > 0) {
> >               if (cfs_rq->avg.util_avg !=3D 0) {
> > -                     sa->util_avg  =3D cfs_rq->avg.util_avg * se->load=
weight;
> > +                     sa->util_avg  =3D cfs_rq->avg.util_avg * se_weigh=
t(se);
> Hi,
>
> The comment above the declaration of se_weight function says we should be
> using full load resolution and get rid of this helper.
>
> Should we be adding new user of the helper?
>
> /*
>  * XXX we want to get rid of these helpers and use the full load resoluti=
on.
>  */
> static inline long se_weight(struct sched_entity *se)
> {
>         return scale_load_down(se->load.weight);
> }
>
>
> >                       sa->util_avg /=3D (cfs_rq->avg.load_avg + 1);
> >
> >                       if (sa->util_avg > cap)
> > --
> > 2.40.1
> >