linux-kernel - Re: [PATCH v2] sched/eevdf: Fix wakeup-preempt by checking cfs_rq->nr

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: <fea9b64f-4ede-475d-8788-73bce88b2e3a@126.com>
Date: Fri, 27 Sep 2024 09:54:28 +0800
From: Honglei Wang <jameshongleiwang@....com>
To: Chen Yu <yu.c.chen@...el.com>, Ingo Molnar <mingo@...hat.com>,
 Peter Zijlstra <peterz@...radead.org>, Juri Lelli <juri.lelli@...hat.com>,
 Vincent Guittot <vincent.guittot@...aro.org>
Cc: Dietmar Eggemann <dietmar.eggemann@....com>,
 Valentin Schneider <vschneid@...hat.com>,
 Chunxin Zang <zangchunxin@...iang.com>, linux-kernel@...r.kernel.org,
 Chen Yu <yu.chen.surf@...il.com>, kernel test robot <oliver.sang@...el.com>,
 K Prateek Nayak <kprateek.nayak@....com>
Subject: Re: [PATCH v2] sched/eevdf: Fix wakeup-preempt by checking
 cfs_rq->nr_running



On 2024/9/25 16:54, Chen Yu wrote:
> Commit 85e511df3cec ("sched/eevdf: Allow shorter slices to wakeup-preempt")
> introduced a mechanism that a wakee with shorter slice could preempt
> the current running task. It also lower the bar for the current task
> to be preempted, by checking the rq->nr_running instead of cfs_rq->nr_running
> when the current task has ran out of time slice. But there is a scenario
> that is problematic. Say, if there is 1 cfs task and 1 rt task, before
> 85e511df3cec, update_deadline() will not trigger a reschedule, and after
> 85e511df3cec, since rq->nr_running is 2 and resched is true, a resched_curr()
> would happen.
> 
> Some workloads (like the hackbench reported by lkp) do not like
> over-scheduling. We can see that the preemption rate has been
> increased by 2.2%:
> 
> 1.654e+08            +2.2%   1.69e+08        hackbench.time.involuntary_context_switches
> 
> Restore its previous check criterion.
> 
> Fixes: 85e511df3cec ("sched/eevdf: Allow shorter slices to wakeup-preempt")
> Reported-by: kernel test robot <oliver.sang@...el.com>
> Closes: https://lore.kernel.org/oe-lkp/202409231416.9403c2e9-oliver.sang@intel.com
> Suggested-by: K Prateek Nayak <kprateek.nayak@....com>
> Tested-by: K Prateek Nayak <kprateek.nayak@....com>
> Signed-off-by: Chen Yu <yu.c.chen@...el.com>
> ---
> v1->v2:
>      Check cfs_rq->nr_running instead of rq->nr_running(K Prateek Nayak)
> ---
>   kernel/sched/fair.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 225b31aaee55..53a351b18740 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -1247,7 +1247,7 @@ static void update_curr(struct cfs_rq *cfs_rq)
>   
>   	account_cfs_rq_runtime(cfs_rq, delta_exec);
>   
> -	if (rq->nr_running == 1)
> +	if (cfs_rq->nr_running == 1)
>   		return;
>  
Hi Yu,

I'm wondering if commit 85e511df3cec wants to give more chances to do 
resched just in case there are 'short slices' tasks ready in the other 
cfs hierarchy. Does something like rq->cfs->nr_running == 1 make more 
sense? But maybe it helps less than 'cfs_rq->nr_running == 1' in this 
hackbench case.


Thanks,
Honglei

>   	if (resched || did_preempt_short(cfs_rq, curr)) {