linux-kernel - Re: [PATCH 1/2] sched/fair: Fix cfs_rq_clock_pelt() for throttled cfs

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <xm26tub4vcg9.fsf@bsegall-linux.svl.corp.google.com>
Date:   Thu, 07 Apr 2022 13:57:26 -0700
From:   Benjamin Segall <bsegall@...gle.com>
To:     Chengming Zhou <zhouchengming@...edance.com>
Cc:     mingo@...hat.com, peterz@...radead.org, juri.lelli@...hat.com,
        vincent.guittot@...aro.org, dietmar.eggemann@....com,
        rostedt@...dmis.org, mgorman@...e.de, bristot@...hat.com,
        linux-kernel@...r.kernel.org, duanxiongchun@...edance.com,
        songmuchun@...edance.com
Subject: Re: [PATCH 1/2] sched/fair: Fix cfs_rq_clock_pelt() for throttled
 cfs_rq

Chengming Zhou <zhouchengming@...edance.com> writes:

> Since commit 23127296889f ("sched/fair: Update scale invariance of PELT")
> change to use rq_clock_pelt() instead of rq_clock_task(), we should also
> use rq_clock_pelt() for throttled_clock_task_time and throttled_clock_task
> accounting.

I think this patch is indeed what we want, despite the confusing
interactions between pelt slowdown+skipping and throttle_clock_task
trying to freeze time.

I think it would be slightly better to rename
throttled_clock_task(_time) to be clock_pelt rather than clock_task, but
that's minor.

Reviewed-by: Ben Segall <bsegall@...gle.com>

>
> Fixes: 23127296889f ("sched/fair: Update scale invariance of PELT")
> Signed-off-by: Chengming Zhou <zhouchengming@...edance.com>
> ---
>  kernel/sched/fair.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index d4bd299d67ab..e6fa5d1141b4 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -4846,7 +4846,7 @@ static int tg_unthrottle_up(struct task_group *tg, void *data)
>  
>  	cfs_rq->throttle_count--;
>  	if (!cfs_rq->throttle_count) {
> -		cfs_rq->throttled_clock_task_time += rq_clock_task(rq) -
> +		cfs_rq->throttled_clock_task_time += rq_clock_pelt(rq) -
>  					     cfs_rq->throttled_clock_task;
>  
>  		/* Add cfs_rq with load or one or more already running entities to the list */
> @@ -4864,7 +4864,7 @@ static int tg_throttle_down(struct task_group *tg, void *data)
>  
>  	/* group is entering throttled state, stop time */
>  	if (!cfs_rq->throttle_count) {
> -		cfs_rq->throttled_clock_task = rq_clock_task(rq);
> +		cfs_rq->throttled_clock_task = rq_clock_pelt(rq);
>  		list_del_leaf_cfs_rq(cfs_rq);
>  	}
>  	cfs_rq->throttle_count++;
> @@ -5308,7 +5308,7 @@ static void sync_throttle(struct task_group *tg, int cpu)
>  	pcfs_rq = tg->parent->cfs_rq[cpu];
>  
>  	cfs_rq->throttle_count = pcfs_rq->throttle_count;
> -	cfs_rq->throttled_clock_task = rq_clock_task(cpu_rq(cpu));
> +	cfs_rq->throttled_clock_task = rq_clock_pelt(cpu_rq(cpu));
>  }
>  
>  /* conditionally throttle active cfs_rq's from put_prev_entity() */