[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <ZxKhxaM8Wv8EEXMd@gpd3>
Date: Fri, 18 Oct 2024 19:58:29 +0200
From: Andrea Righi <andrea.righi@...ux.dev>
To: Tejun Heo <tj@...nel.org>
Cc: David Vernet <void@...ifault.com>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] sched_ext: improve WAKE_SYNC behavior for default idle
CPU selection
On Fri, Oct 18, 2024 at 07:38:10AM -1000, Tejun Heo wrote:
> Hello,
>
> On Fri, Oct 18, 2024 at 11:40:33AM +0200, Andrea Righi wrote:
> ...
> > + * If WAKE_SYNC, try to migrate the wakee to the waker's CPU.
> > + */
> > + if (wake_flags & SCX_WAKE_SYNC) {
> > + cpu = smp_processor_id();
> > +
> > + /*
> > + * If the waker's CPU is cache affine and prev_cpu is idle,
> > + * then avoid a migration.
> > + */
> > + if (cpus_share_cache(cpu, prev_cpu) &&
> > + test_and_clear_cpu_idle(prev_cpu))
> > goto cpu_found;
>
> Shouldn't this do set cpu to prev_cpu before jumping to cpu_found?
Ah! You're absolutely right!
Let me send a v2 (and test it).
Thanks,
-Andrea
>
> Thanks.
>
> --
> tejun
Powered by blists - more mailing lists