From: Hongyan Xia <hongyan.xia2@arm.com>
To: Xuewen Yan <xuewen.yan@unisoc.com>,
dietmar.eggemann@arm.com, vincent.guittot@linaro.org,
mingo@redhat.com, peterz@infradead.org, juri.lelli@redhat.com
Cc: rostedt@goodmis.org, bsegall@google.com, mgorman@suse.de,
vschneid@redhat.co, linux-kernel@vger.kernel.org,
qyousef@layalina.io, ke.wang@unisoc.com, di.shen@unisoc.com,
xuewen.yan94@gmail.com
Subject: Re: [RFC PATCH] sched/util_est: Do not sub the delayed-task's util-est
Date: Fri, 14 Mar 2025 13:14:33 +0000 [thread overview]
Message-ID: <7dbd7dd3-dbf7-42e0-97c4-c5eb50250947@arm.com> (raw)
In-Reply-To: <20250314090909.8404-1-xuewen.yan@unisoc.com>
On 14/03/2025 09:09, Xuewen Yan wrote:
> In cpu_util_without, When the task is in rq, we should
> sub the task's util_est, however, the delayed_task->on_rq
> is true, however, the delayed_task's util had been sub
> when sleep, so there is no need to sub the delayed task's
> util-est. So add the checking of delayed-task.
>
> On the other hand, as said in [1], the logic of util_est's
> enqueue/dequeue could be simplified.
> So simplify it by aligning with the conditions of uclamp.
>
> [1]https://lore.kernel.org/all/CAB8ipk8pEvOtCm-d0o1rsekwxPWUHk9iBGtt9TLTWW-iWTQKiA@mail.gmail.com/
>
> Signed-off-by: Xuewen Yan <xuewen.yan@unisoc.com>
> ---
> kernel/sched/fair.c | 7 ++++---
> 1 file changed, 4 insertions(+), 3 deletions(-)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index c798d2795243..bebf40a0fa4e 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))
> 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)
> util_est_dequeue(&rq->cfs, p);
>
> util_est_update(&rq->cfs, p, flags & DEQUEUE_SLEEP);
> @@ -8037,7 +8037,8 @@ cpu_util(int cpu, struct task_struct *p, int dst_cpu, int boost)
> */
> if (dst_cpu == cpu)
> util_est += _task_util_est(p);
> - else if (p && unlikely(task_on_rq_queued(p) || current == p))
> + else if (p && unlikely(current == p ||
> + (task_on_rq_queued(p) && !p->se.sched_delayed)))
> lsub_positive(&util_est, _task_util_est(p));
>
> util = max(util, util_est);
Have you tested the above changes to make sure util_est enqueue dequeue
are balanced? util_est was broken for quite a while when merging delayed
dequeue because now enqueue_ and dequeue_task() do not always appear in
pairs. Since then, I always have a local patch like this (may be a bit
out of date now) to make sure util_est is balanced
https://lore.kernel.org/all/752ae417c02b9277ca3ec18893747c54dd5f277f.1724245193.git.hongyan.xia2@arm.com/
next prev parent reply other threads:[~2025-03-14 13:14 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-14 9:09 [RFC PATCH] sched/util_est: Do not sub the delayed-task's util-est Xuewen Yan
2025-03-14 13:14 ` Hongyan Xia [this message]
2025-03-21 6:33 ` Xuewen Yan
2025-03-19 13:32 ` Dietmar Eggemann
2025-03-21 6:44 ` Xuewen Yan
2025-03-24 12:10 ` Hongyan Xia
2025-03-24 13:22 ` Dietmar Eggemann
2025-04-14 13:39 ` Hongyan Xia
2025-04-16 10:55 ` Dietmar Eggemann
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=7dbd7dd3-dbf7-42e0-97c4-c5eb50250947@arm.com \
--to=hongyan.xia2@arm.com \
--cc=bsegall@google.com \
--cc=di.shen@unisoc.com \
--cc=dietmar.eggemann@arm.com \
--cc=juri.lelli@redhat.com \
--cc=ke.wang@unisoc.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mgorman@suse.de \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=qyousef@layalina.io \
--cc=rostedt@goodmis.org \
--cc=vincent.guittot@linaro.org \
--cc=vschneid@redhat.co \
--cc=xuewen.yan94@gmail.com \
--cc=xuewen.yan@unisoc.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox