linux-kernel - Re: [PATCH v2] sched/uclamp: Align uclamp and util

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAKfTPtA06D1EGg2VdVHt-P0PDE-1XZVJOp3myZPSNknnFT3nNg@mail.gmail.com>
Date: Tue, 15 Apr 2025 19:04:57 +0200
From: Vincent Guittot <vincent.guittot@...aro.org>
To: Xuewen Yan <xuewen.yan@...soc.com>
Cc: dietmar.eggemann@....com, mingo@...hat.com, peterz@...radead.org, 
	juri.lelli@...hat.com, rostedt@...dmis.org, bsegall@...gle.com, 
	mgorman@...e.de, vschneid@...hat.com, hongyan.xia2@....com, 
	qyousef@...alina.io, ke.wang@...soc.com, di.shen@...soc.com, 
	linux-kernel@...r.kernel.org, xuewen.yan94@...il.com
Subject: Re: [PATCH v2] sched/uclamp: Align uclamp and util_est and call
 before freq update

On Tue, 25 Mar 2025 at 02:48, Xuewen Yan <xuewen.yan@...soc.com> wrote:
>
> When task's uclamp is set, we hope that the CPU frequency
> can increase as quickly as possible when the task is enqueued.
> Because the cpu frequency updating happens during the enqueue_task(),

Strictly speaking, it doesn't happen during enqueue_task but when :
- attach/detach tasks when migrating
- update_load_avg decayed
- io_wait

This often happens during an enqueue but not always ...

> so the rq's uclamp needs to be updated before the task is enqueued,

this doesn't ensure that new rq's uclamp will be taken into account

> just like util_est.

just like util_est

> So, aline the uclamp and util_est and call before freq update.

nit s/aline/align/ ?

>
> For sched-delayed tasks, the rq uclamp/util_est should only be updated
> when they are enqueued upon being awakened.
> So simply the logic of util_est's enqueue/dequeue check.
>
> Signed-off-by: Xuewen Yan <xuewen.yan@...soc.com>
> ---
> v2:
> - simply the util-est's en/dequeue check;
> ---
> Previous discussion:
> https://lore.kernel.org/all/CAB8ipk8pEvOtCm-d0o1rsekwxPWUHk9iBGtt9TLTWW-iWTQKiA@mail.gmail.com/
> https://lore.kernel.org/all/84441660bef0a5e67fd09dc3787178d0276dad31.1740664400.git.hongyan.xia2@arm.com/T/#u
> https://lore.kernel.org/all/CAB8ipk9LpbiUDnbcV6+59+Sa=Ai7tFzO===mpLD3obNdV4=J-A@mail.gmail.com/T/#u
> https://lore.kernel.org/all/aa8baf67-a8ec-4ad8-a6a8-afdcd7036771@arm.com/
> ---
>  kernel/sched/core.c | 17 ++++++++++-------
>  kernel/sched/fair.c |  4 ++--
>  2 files changed, 12 insertions(+), 9 deletions(-)
>
> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> index 042351c7afce..72fbe2031e54 100644
> --- a/kernel/sched/core.c
> +++ b/kernel/sched/core.c
> @@ -1747,7 +1747,7 @@ static inline void uclamp_rq_dec_id(struct rq *rq, struct task_struct *p,
>         }
>  }
>
> -static inline void uclamp_rq_inc(struct rq *rq, struct task_struct *p)
> +static inline void uclamp_rq_inc(struct rq *rq, struct task_struct *p, int flags)
>  {
>         enum uclamp_id clamp_id;
>
> @@ -1763,7 +1763,8 @@ static inline void uclamp_rq_inc(struct rq *rq, struct task_struct *p)
>         if (unlikely(!p->sched_class->uclamp_enabled))
>                 return;
>
> -       if (p->se.sched_delayed)
> +       /* Only inc the delayed task which being woken up. */
> +       if (p->se.sched_delayed && !(flags & ENQUEUE_DELAYED))
>                 return;
>
>         for_each_clamp_id(clamp_id)
> @@ -2031,7 +2032,7 @@ static void __init init_uclamp(void)
>  }
>
>  #else /* !CONFIG_UCLAMP_TASK */
> -static inline void uclamp_rq_inc(struct rq *rq, struct task_struct *p) { }
> +static inline void uclamp_rq_inc(struct rq *rq, struct task_struct *p, int flags) { }
>  static inline void uclamp_rq_dec(struct rq *rq, struct task_struct *p) { }
>  static inline void uclamp_fork(struct task_struct *p) { }
>  static inline void uclamp_post_fork(struct task_struct *p) { }
> @@ -2067,12 +2068,14 @@ void enqueue_task(struct rq *rq, struct task_struct *p, int flags)
>         if (!(flags & ENQUEUE_NOCLOCK))
>                 update_rq_clock(rq);
>
> -       p->sched_class->enqueue_task(rq, p, flags);
>         /*
> -        * Must be after ->enqueue_task() because ENQUEUE_DELAYED can clear
> -        * ->sched_delayed.
> +        * Can be before ->enqueue_task() because uclamp considers the
> +        * ENQUEUE_DELAYED task before its ->sched_delayed gets cleared
> +        * in ->enqueue_task().
>          */
> -       uclamp_rq_inc(rq, p);
> +       uclamp_rq_inc(rq, p, flags);
> +
> +       p->sched_class->enqueue_task(rq, p, flags);
>
>         psi_enqueue(p, flags);
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index c798d2795243..c92fee07fb7b 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -6930,7 +6930,7 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags)
>          * Let's add the task's estimated utilization to the cfs_rq's
>          * estimated utilization, before we update schedutil.
>          */
> -       if (!(p->se.sched_delayed && (task_on_rq_migrating(p) || (flags & ENQUEUE_RESTORE))))
> +       if (!p->se.sched_delayed || (flags & ENQUEUE_DELAYED))

commit message doesn't explain why you change util_est condition

>                 util_est_enqueue(&rq->cfs, p);
>
>         if (flags & ENQUEUE_DELAYED) {
> @@ -7168,7 +7168,7 @@ static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags)
>   */
>  static bool dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags)
>  {
> -       if (!(p->se.sched_delayed && (task_on_rq_migrating(p) || (flags & DEQUEUE_SAVE))))
> +       if (!p->se.sched_delayed)

same here, you should explain in commit message why it's okay to do so

>                 util_est_dequeue(&rq->cfs, p);
>
>         util_est_update(&rq->cfs, p, flags & DEQUEUE_SLEEP);
> --
> 2.25.1
>