mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: Chris Mason <clm@meta.com>
Cc: linux-kernel@vger.kernel.org, Ingo Molnar <mingo@kernel.org>,
	dietmar.eggemann@arm.com, vschneid@redhat.com
Subject: Re: scheduler performance regression since v6.11
Date: Tue, 13 May 2025 09:15:51 +0200	[thread overview]
Message-ID: <20250513071551.GC25763@noisy.programming.kicks-ass.net> (raw)
In-Reply-To: <d3c8527f-ffaf-4463-a305-17ca21a06ce8@meta.com>

On Mon, May 12, 2025 at 06:35:24PM -0400, Chris Mason wrote:

> If I take 54a58a787791 or 54a58a787791 and turn off the
> DELAY_DEQUEUE/ZERO features at run time, I don't get the performance
> back.  But, if I patch them such that DELAY_DEQUEUE/ZERO default off
> 
> diff --git a/kernel/sched/features.h b/kernel/sched/features.h
> index 7fdeb5576188c..94409e9831e97 100644
> --- a/kernel/sched/features.h
> +++ b/kernel/sched/features.h
> @@ -37,8 +37,8 @@ SCHED_FEAT(CACHE_HOT_BUDDY, true)
>   *
>   * DELAY_ZERO clips the lag on dequeue (or wakeup) to 0.
>   */
> -SCHED_FEAT(DELAY_DEQUEUE, true)
> -SCHED_FEAT(DELAY_ZERO, true)
> +SCHED_FEAT(DELAY_DEQUEUE, false)
> +SCHED_FEAT(DELAY_ZERO, false)
> 
> It runs at 2M QPS again.  If I enable DELAY_QUEUE/ZERO, I go back to 1.95M

Well, that's odd... there should not be residual effects like that. I'll
see if I can spot anything.

  reply	other threads:[~2025-05-13  7:15 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-05-07 23:13 Chris Mason
2025-05-09 19:49 ` Peter Zijlstra
2025-05-12 18:08   ` Peter Zijlstra
2025-05-12 19:39     ` Chris Mason
2025-05-12 22:35       ` Chris Mason
2025-05-13  7:15         ` Peter Zijlstra [this message]
2025-05-16 10:18         ` Peter Zijlstra
2025-05-20 14:38           ` Dietmar Eggemann
2025-05-20 14:53             ` Chris Mason
2025-05-21 13:59               ` Dietmar Eggemann
2025-05-21 14:32                 ` Chris Mason
2025-05-20 19:38             ` Peter Zijlstra
2025-05-21 14:02               ` Dietmar Eggemann
2025-05-21 15:02                 ` Peter Zijlstra
2025-05-21 19:00                   ` Peter Zijlstra
2025-05-21 14:54               ` Peter Zijlstra
2025-05-22  8:48                 ` Peter Zijlstra
2025-05-22 15:00                   ` Johannes Weiner
2025-05-23 15:40                     ` Peter Zijlstra
2025-05-23 12:27                   ` Dietmar Eggemann
2025-07-10 12:46                   ` [tip: sched/core] sched/psi: Optimize psi_group_change() cpu_clock() usage tip-bot2 for Peter Zijlstra

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=20250513071551.GC25763@noisy.programming.kicks-ass.net \
    --to=peterz@infradead.org \
    --cc=clm@meta.com \
    --cc=dietmar.eggemann@arm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=vschneid@redhat.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

all inboxes | Powered by JetHome®