From: Christian Loehle <christian.loehle@arm.com>
To: Tom Gebhardt <tomge68@gmail.com>, Qais Yousef <qyousef@layalina.io>
Cc: linux-kernel@vger.kernel.org,
Peter Zijlstra <peterz@infradead.org>,
Ingo Molnar <mingo@redhat.com>,
Vincent Guittot <vincent.guittot@linaro.org>
Subject: Re: [PATCH v2 00/13] sched/fair/schedutil: Better manage system response time
Date: Fri, 15 May 2026 11:01:00 +0100 [thread overview]
Message-ID: <28c59186-4ba0-44f4-b5b8-42f6f73cd3e6@arm.com> (raw)
In-Reply-To: <32dea9123588227a02971341b22e84d3.tomge68@gmail.com>
On 5/15/26 09:24, Tom Gebhardt wrote:
> Hi Qais,
>
> Thanks for the follow-up. Here are the patch isolation results and answers to your questions.
>
> Regarding the governor:
>
> Yes, I'm running `ondemand`, not `schedutil`. My mistake for not mentioning that upfront - I
> assumed the improvement was due to the util_est path being triggered regardless of the governor.
> The improvement is clearly measurable even with `ondemand`, which is surprising given that your
> patches specifically target `schedutil`.
Something is wrong, as Qais mentioned raspberry pi (SMP) with ondemand shouldn't be affected by
util_est changes. In particular with 4 mostly-running workers and 4 CPUs.
Does patch 12 also show similar effects with powersave/performance cpufreq governor?
Qais also split patch 12 out separately and Vincent posted a fix, care to give that a try?
https://lore.kernel.org/lkml/agRyoe1wHyZ-vMk9@vingu-cube/
Thanks for testing these, I'll try to reproduce what you're seeing, too.
> [snip]
next prev parent reply other threads:[~2026-05-15 10:01 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-05-04 1:59 Qais Yousef
2026-05-04 1:59 ` [PATCH v2 01/13] sched: cpufreq: Rename map_util_perf to sugov_apply_dvfs_headroom Qais Yousef
2026-05-04 1:59 ` [PATCH v2 02/13] sched/pelt: Add a new function to approximate the future util_avg value Qais Yousef
2026-05-04 1:59 ` [PATCH v2 03/13] sched/pelt: Add a new function to approximate runtime to reach given util Qais Yousef
2026-06-04 8:50 ` Vincent Guittot
2026-07-14 13:37 ` Qais Yousef
2026-05-04 1:59 ` [PATCH v2 04/13] sched/fair: Remove magic hardcoded margin in fits_capacity() Qais Yousef
2026-06-04 10:08 ` Vincent Guittot
2026-07-14 14:04 ` Qais Yousef
2026-09-23 6:54 ` Zhan Xusheng
2026-09-23 11:34 ` Chen, Yu C
2026-10-05 14:31 ` Zhan Xusheng
2026-11-01 23:48 ` Qais Yousef
2026-05-04 1:59 ` [PATCH v2 05/13] sched: cpufreq: Remove magic 1.25 headroom from sugov_apply_dvfs_headroom() Qais Yousef
2026-05-04 1:59 ` [PATCH v2 06/13] sched/fair: Extend util_est to improve rampup time Qais Yousef
2026-05-04 1:59 ` [PATCH v2 07/13] sched/fair: util_est: Take into account periodic tasks Qais Yousef
2026-05-04 1:59 ` [PATCH v2 RFC 08/13] sched/qos: Add a new sched-qos interface Qais Yousef
2026-05-06 20:38 ` Tim Chen
2026-05-07 9:55 ` Qais Yousef
2026-05-07 14:20 ` Chen, Yu C
2026-05-09 9:39 ` Qais Yousef
2026-05-11 10:57 ` Peter Zijlstra
2026-05-12 7:58 ` Qais Yousef
2026-05-12 8:30 ` Peter Zijlstra
2026-05-12 8:47 ` Qais Yousef
2026-05-19 9:47 ` Peter Zijlstra
2026-05-19 10:56 ` Qais Yousef
2026-06-23 10:28 ` zhidao su
2026-07-14 14:17 ` Qais Yousef
2026-05-04 1:59 ` [PATCH v2 09/13] sched/qos: Add rampup multiplier QoS Qais Yousef
2026-05-11 11:03 ` Peter Zijlstra
2026-05-12 7:59 ` Qais Yousef
2026-05-12 8:37 ` Christian Loehle
2026-05-12 8:53 ` Qais Yousef
2026-06-18 8:30 ` zhidao su (Xiaomi)
2026-07-14 14:08 ` Qais Yousef
2026-05-04 2:00 ` [PATCH v2 10/13] sched/fair: Disable util_est when rampup_multiplier is 0 Qais Yousef
2026-05-04 2:00 ` [PATCH v2 11/13] sched/fair: Don't mess with util_avg post init Qais Yousef
2026-05-04 2:00 ` [PATCH v2 12/13] sched/fair: Call update_util_est() after dequeue_entities() Qais Yousef
2026-05-04 2:00 ` [PATCH v2 RFC 13/13] sched/pelt: Always allow load updates Qais Yousef
2026-05-11 17:58 ` [PATCH v2 00/13] sched/fair/schedutil: Better manage system response time John Stultz
2026-05-12 8:01 ` Qais Yousef
2026-05-13 15:09 ` Tom Gebhardt
2026-05-15 1:42 ` Qais Yousef
2026-05-15 8:24 ` Tom Gebhardt
2026-05-15 10:01 ` Christian Loehle [this message]
2026-05-15 13:57 ` Tom Gebhardt
2026-05-16 13:43 ` Christian Loehle
2026-05-19 7:46 ` Tom Gebhardt
2026-05-25 7:25 ` Tom Gebhardt
2026-05-28 7:38 ` Christian Loehle
2026-05-28 11:08 ` Tom Gebhardt
2026-05-16 3:01 ` Qais Yousef
2026-05-28 12:50 ` Tom Gebhardt
2026-05-29 1:43 ` Qais Yousef
2026-05-29 7:53 ` Tom Gebhardt
2026-05-31 2:35 ` Qais Yousef
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=28c59186-4ba0-44f4-b5b8-42f6f73cd3e6@arm.com \
--to=christian.loehle@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=qyousef@layalina.io \
--cc=tomge68@gmail.com \
--cc=vincent.guittot@linaro.org \
/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®