[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <Z3_qA9wZtei_oE-y@vaxr-BM6660-BM6360>
Date: Thu, 9 Jan 2025 23:23:47 +0800
From: I Hsin Cheng <richard120310@...il.com>
To: Peter Zijlstra <peterz@...radead.org>
Cc: mingo@...hat.com, juri.lelli@...hat.com, vincent.guittot@...aro.org,
dietmar.eggemann@....com, rostedt@...dmis.org, bsegall@...gle.com,
mgorman@...e.de, vschneid@...hat.com, linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH] sched/fair: Refactor can_migrate_task() to elimate
looping
On Thu, Jan 09, 2025 at 11:21:22AM +0100, Peter Zijlstra wrote:
> On Thu, Jan 09, 2025 at 06:12:22PM +0800, I Hsin Cheng wrote:
>
> > > > Except you cannot put cpumask on-stack...
> >
> > Oh I'm sorry, may I ask the reason? is it because cpumask tends to be
> > very large?
>
> Yes, when building with NR_CPUS=8192 (as Distos tend to do) it is 1K of
> stack space.
>
Thanks for your explanation! I've made the requested changes using
"cpumask_first_and_and()" so there won't be additional cpumask needed.
Tests are done and I'm sending a v2 rfc patch.
-- Richard.
Powered by blists - more mailing lists