Received-SPF: pass (google.com: domain of linux-kernel+bounces-60371-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
Precedence: bulk
MIME-Version: 1.0
References: <20240207034704.935774-1-alexs@kernel.org> <20240207034704.935774-4-alexs@kernel.org>
 <20240210011223.GA28896@ranerica-svr.sc.intel.com>
In-Reply-To: <20240210011223.GA28896@ranerica-svr.sc.intel.com>
From: Alex Shi <seakeel@gmail.com>
Date: Sat, 10 Feb 2024 19:08:45 +0800
Message-ID: <CAJy-Am=9R=AwV8Dpu=Pb8ohEKPZGDx2aJR9eH9hfWpTpXdGx9A@mail.gmail.com>
Subject: Re: [PATCH v4 4/4] sched/fair: Check the SD_ASYM_PACKING flag in sched_use_asym_prio()
To: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Cc: alexs@kernel.org, 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>, 
	Daniel Bristot de Oliveira <bristot@redhat.com>, Valentin Schneider <vschneid@redhat.com>, linux-kernel@vger.kernel.org, 
	yangyicong@hisilicon.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Ricardo Neri <ricardo.neri-calderon@linux.intel.com> =E4=BA=8E2024=E5=B9=B4=
2=E6=9C=8810=E6=97=A5=E5=91=A8=E5=85=AD 09:11=E5=86=99=E9=81=93=EF=BC=9A
>
> On Wed, Feb 07, 2024 at 11:47:04AM +0800, alexs@kernel.org wrote:
> > From: Alex Shi <alexs@kernel.org>
> >
> > sched_use_asym_prio() checks whether CPU priorities should be used. It
> > makes sense to check for the SD_ASYM_PACKING() inside the function.
> > Since both sched_asym() and sched_group_asym() use sched_use_asym_prio(=
),
> > remove the now superfluous checks for the flag in various places.
>
> Reviewed-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
>
> Tested on Alder Lake and Meteor Lake, which use asym_packing.
>
> Tested-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>

It's the best gift for my lunar new year! :)
Next version with your Tested and Reviewed is coming.

Thanks a lot!
Alex
>
> >
> > Signed-off-by: Alex Shi <alexs@kernel.org>
> > To: linux-kernel@vger.kernel.org
> > To: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
> > To: Ben Segall <bsegall@google.com>
> > To: Steven Rostedt <rostedt@goodmis.org>
> > To: Dietmar Eggemann <dietmar.eggemann@arm.com>
> > To: Valentin Schneider <vschneid@redhat.com>
> > To: Daniel Bristot de Oliveira <bristot@redhat.com>
> > To: Vincent Guittot <vincent.guittot@linaro.org>
> > To: Juri Lelli <juri.lelli@redhat.com>
> > To: Peter Zijlstra <peterz@infradead.org>
> > To: Ingo Molnar <mingo@redhat.com>
> > ---
> >  kernel/sched/fair.c | 16 +++++++---------
> >  1 file changed, 7 insertions(+), 9 deletions(-)
> >
> > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> > index 942b6358f683..10ae28e1c088 100644
> > --- a/kernel/sched/fair.c
> > +++ b/kernel/sched/fair.c
> > @@ -9740,6 +9740,9 @@ group_type group_classify(unsigned int imbalance_=
pct,
> >   */
> >  static bool sched_use_asym_prio(struct sched_domain *sd, int cpu)
> >  {
> > +     if (!(sd->flags & SD_ASYM_PACKING))
> > +             return false;
> > +
> >       if (!sched_smt_active())
> >               return true;
> >
> > @@ -9941,11 +9944,9 @@ static inline void update_sg_lb_stats(struct lb_=
env *env,
> >       sgs->group_weight =3D group->group_weight;
> >
> >       /* Check if dst CPU is idle and preferred to this group */
> > -     if (!local_group && env->sd->flags & SD_ASYM_PACKING &&
> > -         env->idle !=3D CPU_NOT_IDLE && sgs->sum_h_nr_running &&
> > -         sched_group_asym(env, sgs, group)) {
> > +     if (!local_group && env->idle !=3D CPU_NOT_IDLE && sgs->sum_h_nr_=
running &&
> > +         sched_group_asym(env, sgs, group))
> >               sgs->group_asym_packing =3D 1;
> > -     }
> >
> >       /* Check for loaded SMT group to be balanced to dst CPU */
> >       if (!local_group && smt_balance(env, sgs, group))
> > @@ -11041,9 +11042,7 @@ static struct rq *find_busiest_queue(struct lb_=
env *env,
> >                * If balancing between cores, let lower priority CPUs he=
lp
> >                * SMT cores with more than one busy sibling.
> >                */
> > -             if ((env->sd->flags & SD_ASYM_PACKING) &&
> > -                 sched_asym(env->sd, i, env->dst_cpu) &&
> > -                 nr_running =3D=3D 1)
> > +             if (sched_asym(env->sd, i, env->dst_cpu) && nr_running =
=3D=3D 1)
> >                       continue;
> >
> >               switch (env->migration_type) {
> > @@ -11139,8 +11138,7 @@ asym_active_balance(struct lb_env *env)
> >        * the lower priority @env::dst_cpu help it. Do not follow
> >        * CPU priority.
> >        */
> > -     return env->idle !=3D CPU_NOT_IDLE && (env->sd->flags & SD_ASYM_P=
ACKING) &&
> > -            sched_use_asym_prio(env->sd, env->dst_cpu) &&
> > +     return env->idle !=3D CPU_NOT_IDLE && sched_use_asym_prio(env->sd=
, env->dst_cpu) &&
> >              (sched_asym_prefer(env->dst_cpu, env->src_cpu) ||
> >               !sched_use_asym_prio(env->sd, env->src_cpu));
> >  }
> > --
> > 2.43.0
> >