From: Marc Gonzalez <marc.w.gonzalez@free.fr>
To: Rui Sousa <rui.sousa@nxp.com>,
John Ogness <john.ogness@linutronix.de>,
Leon Woestenberg <leon@sidebranch.com>,
Daniel Wagner <dwagner@suse.de>
Cc: "linux-rt-users@vger.kernel.org" <linux-rt-users@vger.kernel.org>,
Steven Rostedt <rostedt@goodmis.org>,
Thomas Gleixner <tglx@linutronix.de>,
Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
Daniel Wagner <daniel.wagner@suse.com>,
Clark Williams <williams@redhat.com>,
Pavel Machek <pavel@denx.de>,
Luis Goncalves <lgoncalv@redhat.com>,
John McCalpin <mccalpin@tacc.utexas.edu>
Subject: Re: [EXT] Re: Large(ish) variance induced by SCHED_FIFO
Date: Mon, 8 Sep 2025 19:03:54 +0200 [thread overview]
Message-ID: <7e23f985-e6f2-44b4-a7f3-60d2c8cb76a6@free.fr> (raw)
In-Reply-To: <AS1PR04MB95012A6D68FC8F5D446E3E56E80CA@AS1PR04MB9501.eurprd04.prod.outlook.com>
On 08/09/2025 18:52, Rui Sousa wrote:
> Just replace your last setup script line with:
> echo -1 > /proc/sys/kernel/sched_rt_runtime_us
Launch script consists of:
echo -1 > /proc/sys/kernel/sched_rt_runtime_us
# RUN RT BENCHMARK
echo 950000 > /proc/sys/kernel/sched_rt_runtime_us
https://www.kernel.org/doc/html/latest/scheduler/sched-rt-group.html
"A run time of -1 specifies runtime == period, ie. no limit."
1) Disable RT throttling
2) Run RT benchmark
3) Enable RT throttling
> I believe you are seeing regular tasks being scheduled
> in the middle of the benchmark measurement.
I don't think so.
RT throttling is disabled.
The benchmark runs as SCHED_FIFO 99.
Nothing can interrupt it, except interrupts :)
Am checking Daniel's suggestion.
Regards
next prev parent reply other threads:[~2025-09-08 17:04 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-09-04 23:45 Large(ish) variance induced by SCHED_FIFO Marc Gonzalez
2025-09-05 8:25 ` Daniel Wagner
2025-09-05 13:23 ` Marc Gonzalez
2025-09-05 14:03 ` John Ogness
2025-09-05 15:34 ` Marc Gonzalez
2025-09-05 16:07 ` John Ogness
2025-09-05 17:40 ` Marc Gonzalez
2025-09-08 9:36 ` John Ogness
2025-09-08 15:42 ` Marc Gonzalez
2025-09-08 16:02 ` Daniel Wagner
2025-09-08 18:40 ` Marc Gonzalez
2025-09-09 11:08 ` Unexplained variance in run-time of trivial program Marc Gonzalez
2025-09-09 11:21 ` Daniel Wagner
2025-09-09 12:42 ` Marc Gonzalez
2025-09-09 14:23 ` Steven Rostedt
2025-09-09 12:34 ` John Ogness
2025-09-09 14:08 ` Marc Gonzalez
2025-09-09 18:33 ` Marc Gonzalez
[not found] ` <CAMjhiJjMOQN-nWd+KP4JBBNHf20M+J2fXAuTTvowXctJgvGOcQ@mail.gmail.com>
2025-09-09 20:30 ` Marc Gonzalez
2025-09-10 7:59 ` Daniel Wagner
2025-09-11 21:29 ` Marc Gonzalez
2025-09-11 22:15 ` Marc Gonzalez
2025-09-12 7:44 ` Daniel Wagner
2025-09-13 10:09 ` Marc Gonzalez
2025-09-15 16:30 ` Marc Gonzalez
2025-09-16 7:50 ` Ahmed S. Darwish
2025-09-25 11:38 ` Marc Gonzalez
2025-09-08 16:52 ` [EXT] Re: Large(ish) variance induced by SCHED_FIFO Rui Sousa
2025-09-08 17:03 ` Marc Gonzalez [this message]
[not found] ` <CAMjhiJiVDmnx+pgpvtAy=KarHexjsYs+T9tSkpGvppjqFdtmiw@mail.gmail.com>
2025-09-05 21:05 ` Marc Gonzalez
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=7e23f985-e6f2-44b4-a7f3-60d2c8cb76a6@free.fr \
--to=marc.w.gonzalez@free.fr \
--cc=bigeasy@linutronix.de \
--cc=daniel.wagner@suse.com \
--cc=dwagner@suse.de \
--cc=john.ogness@linutronix.de \
--cc=leon@sidebranch.com \
--cc=lgoncalv@redhat.com \
--cc=linux-rt-users@vger.kernel.org \
--cc=mccalpin@tacc.utexas.edu \
--cc=pavel@denx.de \
--cc=rostedt@goodmis.org \
--cc=rui.sousa@nxp.com \
--cc=tglx@linutronix.de \
--cc=williams@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