[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20241202095401.GJ24400@noisy.programming.kicks-ass.net>
Date: Mon, 2 Dec 2024 10:54:01 +0100
From: Peter Zijlstra <peterz@...radead.org>
To: Vincent Guittot <vincent.guittot@...aro.org>
Cc: mingo@...hat.com, juri.lelli@...hat.com, dietmar.eggemann@....com,
rostedt@...dmis.org, bsegall@...gle.com, mgorman@...e.de,
vschneid@...hat.com, linux-kernel@...r.kernel.org,
kprateek.nayak@....com, pauld@...hat.com, efault@....de,
luis.machado@....com
Subject: Re: [PATCH 03/10 v2] sched/fair: Add new cfs_rq.h_nr_runnable
On Fri, Nov 29, 2024 at 05:17:49PM +0100, Vincent Guittot wrote:
> With delayed dequeued feature, a sleeping sched_entity remains queued in
> the rq until its lag has elapsed. As a result, it stays also visible
> in the statistics that are used to balance the system and in particular
> the field cfs.h_nr_queued when the sched_entity is associated to a task.
>
> Create a new h_nr_runnable that tracks all queued and runnable tasks and
> use it when balancing the load on the system.
>
> h_nr_runnable will be used in several places to make decision on load
> balance:
> - PELT runnable_avg
> - deciding if a group is overloaded or has spare capacity
> - numa stats
> - reduced capacity management
> - load balance
> - nohz kick
>
> It should be noticed that the rq->nr_running still counts the delayed
> dequeued tasks as delayed dequeue is a fair feature that is meaningless
> at core level.
Since you're doing a v3, could you please split this into 2 patches, one
adding the accounting, and then a separate patch making use of it?
Powered by blists - more mailing lists