mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Tejun Heo <tj@kernel.org>
To: Yuri Andriaccio <yuri.andriaccio@santannapisa.it>
Cc: luca abeni <luca.abeni@santannapisa.it>,
	Peter Zijlstra <peterz@infradead.org>,
	Yuri Andriaccio <yurand2000@gmail.com>,
	Ingo Molnar <mingo@redhat.com>,
	Juri Lelli <juri.lelli@redhat.com>,
	Vincent Guittot <vincent.guittot@linaro.org>,
	Dietmar Eggemann <dietmar.eggemann@arm.com>,
	Steven Rostedt <rostedt@goodmis.org>,
	Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>,
	Valentin Schneider <vschneid@redhat.com>,
	linux-kernel@vger.kernel.org, hannes@cmpxchg.org,
	mkoutny@suse.com, cgroups@vger.kernel.org
Subject: Re: [RFC PATCH v5 20/29] sched/deadline: Allow deeper hierarchies of RT cgroups
Date: Mon, 18 May 2026 08:47:37 -1000	[thread overview]
Message-ID: <agteySwOHszMVMp8@slm.duckdns.org> (raw)
In-Reply-To: <c51eb4b9-143d-495f-a35b-090fb688cca4@santannapisa.it>

Hello, Yuri.

On Mon, May 18, 2026 at 05:27:17PM +0200, Yuri Andriaccio wrote:
> Interface:
> - File: cpu.rt.max
> - Format: <runtime>|"max" <period>
> - Default value:
>     "max" <parent period> - if the parent schedules on the root runqueue.
>     0 <parent period> - if the parent is instead using HCBS.
> - Meaning (incomplete/dubious):
>     The bandwidth allocated to the specific cgroup and all of its children.
>     Since sum(children bw) <= own bw, a cgroup's servers will be configured
>     with (own bw - sum(children bw)) bandwidth.
>     A cgroup set to "max" whose parents are all set to "max" (root cgroup
> excluded)
>     will run their tasks in the root runqueue.
>     A cgroup set to "max" whose parent has a non-zero reservation will
>     inherit the parent's configuration.
>     The root cgroup's cpu.rt.max file reserves the maximum HCBS bandwidth
> for
>     the whole hierarchy. Root set to "max" disable HCBS (as if set with a
> zero runtime).

I wonder whether it can be generalized more. Would something like the
following work? I'm going to ignore period for the sake of simplicity as it
doesn't seem to affect admission decisions.

- There is no root cgroup.rt.max in line with other control knobs.

- max means running in the nearest ancestor that has budget configuration.
  Obviously, if no one has budget configured, run in root.

- Setting a budget is subject to admission control in both directions - the
  budget source (the nearest budgeted ancestor, or the root pool if none)
  should have enough to give out and the target budget should be big enough
  to contain the actual usages and !max descendants in the subtree. Going
  to max is always fine - the source previously gave the budget out, so it
  has room to take everything back.

It seems like the above would give fairly generic behavior without abrupt
system-wide switches while staying relatively close to the behaviors of
other resource knobs. I could be missing something tho. Would something like
this work?

Thanks.

-- 
tejun

  reply	other threads:[~2026-05-18 18:47 UTC|newest]

Thread overview: 71+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-04-30 21:38 [RFC PATCH v5 00/29] Hierarchical Constant Bandwidth Server Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 01/29] sched/deadline: Fix replenishment logic for non-deferred servers Yuri Andriaccio
2026-05-20  8:34   ` [tip: sched/core] " tip-bot2 for Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 02/29] sched/deadline: Do not access dl_se->rq directly Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 03/29] sched/deadline: Distinguish between dl_rq and my_q Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 04/29] sched/rt: Pass an rt_rq instead of an rq where needed Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 05/29] sched/rt: Move functions from rt.c to sched.h Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 06/29] sched/rt: Disable RT_GROUP_SCHED Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 07/29] sched/rt: Remove unnecessary runqueue pointer in struct rt_rq Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 08/29] sched/rt: Introduce HCBS specific structs in task_group Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 09/29] sched/core: Initialize HCBS specific structures Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 10/29] sched/deadline: Add dl_init_tg Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 11/29] sched/rt: Add {alloc/unregister/free}_rt_sched_group Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 12/29] sched/deadline: Account rt-cgroups bandwidth in deadline tasks schedulability tests Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 13/29] sched/rt: Implement dl-server operations for rt-cgroups Yuri Andriaccio
2026-05-05 13:04   ` Peter Zijlstra
2026-04-30 21:38 ` [RFC PATCH v5 14/29] sched/rt: Update task event callbacks for HCBS scheduling Yuri Andriaccio
2026-05-05 13:16   ` Peter Zijlstra
2026-04-30 21:38 ` [RFC PATCH v5 15/29] sched/rt: Update rt-cgroup schedulability checks Yuri Andriaccio
2026-05-05 14:36   ` Peter Zijlstra
2026-04-30 21:38 ` [RFC PATCH v5 16/29] sched/rt: Allow zeroing the runtime of the root control group Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 17/29] sched/rt: Remove old RT_GROUP_SCHED data structures Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 18/29] sched/core: Cgroup v2 support Yuri Andriaccio
2026-05-05 14:59   ` Peter Zijlstra
2026-05-06 19:58     ` luca abeni
2026-05-07  7:01       ` Peter Zijlstra
2026-05-07 13:30         ` luca abeni
2026-05-07 14:16           ` Peter Zijlstra
2026-04-30 21:38 ` [RFC PATCH v5 19/29] sched/rt: Remove support for cgroups-v1 Yuri Andriaccio
2026-05-05 15:01   ` Peter Zijlstra
2026-05-07 15:35     ` Juri Lelli
2026-05-13 12:15       ` Yuri Andriaccio
2026-05-13 14:37         ` Juri Lelli
2026-04-30 21:38 ` [RFC PATCH v5 20/29] sched/deadline: Allow deeper hierarchies of RT cgroups Yuri Andriaccio
2026-05-05 15:15   ` Peter Zijlstra
2026-05-05 19:56     ` Tejun Heo
2026-05-07 10:53       ` Peter Zijlstra
2026-05-07 15:03         ` Juri Lelli
2026-05-07 15:05           ` Peter Zijlstra
2026-05-07 16:39           ` luca abeni
2026-05-11  9:29             ` Juri Lelli
2026-05-11 17:52               ` Tejun Heo
2026-05-07 16:44         ` luca abeni
2026-05-11  9:40         ` luca abeni
2026-05-11 18:15           ` Tejun Heo
2026-05-11 17:37         ` Tejun Heo
2026-05-07 14:30       ` luca abeni
2026-05-11 18:28         ` Tejun Heo
2026-05-12 17:38           ` Yuri Andriaccio
2026-05-12 18:19             ` Tejun Heo
2026-05-12 18:20               ` Tejun Heo
2026-05-13 12:08                 ` Yuri Andriaccio
2026-05-13 19:10                   ` Tejun Heo
2026-05-14  7:25               ` luca abeni
2026-05-14 22:20                 ` Tejun Heo
2026-05-18 15:27                   ` Yuri Andriaccio
2026-05-18 18:47                     ` Tejun Heo [this message]
2026-05-19 21:02                       ` [BULK] " luca abeni
2026-05-20 18:49                         ` Tejun Heo
2026-04-30 21:38 ` [RFC PATCH v5 21/29] sched/rt: Update default bandwidth for real-time tasks to ONE Yuri Andriaccio
2026-05-20  8:34   ` [tip: sched/core] " tip-bot2 for Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 22/29] sched/rt: Add rt-cgroup migration functions Yuri Andriaccio
2026-05-05 15:20   ` Peter Zijlstra
2026-05-05 15:24   ` Peter Zijlstra
2026-04-30 21:38 ` [RFC PATCH v5 23/29] sched/rt: Hook HCBS " Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 24/29] sched/core: Execute enqueued balance callbacks when changing allowed CPUs Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 25/29] sched/rt: Try pull task on empty server pick Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 26/29] sched/core: Execute enqueued balance callbacks after migrate_disable_switch Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 27/29] Documentation: Update documentation for real-time cgroups Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 28/29] sched/rt: Add debug BUG_ONs for pre-migration code Yuri Andriaccio
2026-04-30 21:38 ` [RFC PATCH v5 29/29] sched/rt: Add debug BUG_ONs in migration code Yuri Andriaccio

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=agteySwOHszMVMp8@slm.duckdns.org \
    --to=tj@kernel.org \
    --cc=bsegall@google.com \
    --cc=cgroups@vger.kernel.org \
    --cc=dietmar.eggemann@arm.com \
    --cc=hannes@cmpxchg.org \
    --cc=juri.lelli@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luca.abeni@santannapisa.it \
    --cc=mgorman@suse.de \
    --cc=mingo@redhat.com \
    --cc=mkoutny@suse.com \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=vincent.guittot@linaro.org \
    --cc=vschneid@redhat.com \
    --cc=yurand2000@gmail.com \
    --cc=yuri.andriaccio@santannapisa.it \
    /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®