linux-rt-users.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: John Ogness <john.ogness@linutronix.de>
To: "KHUENY.Gerhard" <Gerhard.KHUENY@bachmann.info>
Cc: "KHUENY.Gerhard" <Gerhard.KHUENY@bachmann.info>,
	"'linux-rt-users@vger.kernel.org'"
	<linux-rt-users@vger.kernel.org>
Subject: Re: Realtime throttling and Real-Time group scheduling
Date: Tue, 21 Oct 2025 16:49:36 +0206	[thread overview]
Message-ID: <87frbce2nb.fsf@jogness.linutronix.de> (raw)
In-Reply-To: <AM9PR05MB77775F0B030DB96B505F00FB9BF2A@AM9PR05MB7777.eurprd05.prod.outlook.com>

On 2025-10-21, KHUENY.Gerhard <Gerhard.KHUENY@bachmann.info> wrote:
>> > I'm running a 6.12.39-rt Linux kernel and have been experimenting with
>> > runtime throttling on my embedded hardware.  However, it seems that
>> > the settings for sched_rt_period_us and sched_rt_runtime_us have no
>> > effect. In my kernel configuration, CONFIG_RT_GROUP_SCHED is not set.
>> >
>> > Is my understanding correct that CONFIG_RT_GROUP_SCHED is not required
>> > to throttle real-time tasks?
>> 
>> Yes. But instead of adjusting "throttling", you adjust the guaranteed
>> bandwidth for !rt tasks. You can read about the fair deadline server in its
>> original commit message [0].
>> 
>> Basically you can adjust these debugfs files:
>> 
>> /sys/kernel/debug/sched/fair_server/cpu*/runtime
>> 
>> to guarantee bandwith for !rt tasks.
>> 
>> And yes, we are aware that proper documentation for this is missing.
>> 
>> John Ogness
>> 
>> [0]
>> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/Do
>> cumentation?&id=a110a81c52a9de73e2e57e826dd3bf0fd4c22226
>
> Thanks for clarification. So this means that if CONFIG_RT_GROUP_SCHED
> is not set, the parameters /proc/sys/kernel/sched_rt_period and
> /proc/sys/kernel/sched_rt_runtime have no effect?

They have no effect on the "RT throttling" feature, which does not exist
for CONFIG_RT_GROUP_SCHED=n. See where it was removed here [1].

However, the parameters are still used for admission control of
SCHED_DEADLINE tasks. And likewise they also affect what values can be
written to /sys/kernel/debug/sched/fair_server/cpu*/runtime to configure
the fair deadline server.

Note that this is all new starting with v6.12.

John Ogness

[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=5f6bd380c7bdbe10f7b4e8ddcceed60ce0714c6d

      reply	other threads:[~2025-10-21 14:43 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-10-20  8:54 Realtime throttling and Real-Time group scheduling KHUENY.Gerhard
2025-10-20  9:36 ` John Ogness
2025-10-21  9:13   ` KHUENY.Gerhard
2025-10-21 14:43     ` John Ogness [this message]

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=87frbce2nb.fsf@jogness.linutronix.de \
    --to=john.ogness@linutronix.de \
    --cc=Gerhard.KHUENY@bachmann.info \
    --cc=linux-rt-users@vger.kernel.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;
as well as URLs for NNTP newsgroup(s).