[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Thu, 3 Mar 2016 18:45:11 +0100 (CET)
From: Thomas Gleixner <tglx@...utronix.de>
To: Peter Zijlstra <peterz@...radead.org>
cc: Ingo Molnar <mingo@...nel.org>, linux-kernel@...r.kernel.org,
jschoenh@...zon.de, oleg@...hat.com
Subject: Re: [PATCH] sched: Allow per-cpu kernel threads to run on online &&
!active
On Thu, 3 Mar 2016, Thomas Gleixner wrote:
> On Thu, 3 Mar 2016, Peter Zijlstra wrote:
> > > +++ b/kernel/sched/core.c
> > > @@ -5726,23 +5726,6 @@ static int sched_cpu_active(struct notif
> > > case CPU_STARTING:
> > > set_cpu_rq_start_time();
> >
> > If we stick that in notify_starting() we can kill all of this.
>
> Yep. I'm going to kill that one by one :)
After I'm done fixing the wreckage which is caused by moving set_active()
alone. The numa and cpuset muck rely on it ....
But now that we can delay random workload on the upcoming/outgoing cpu we can
move that stuff alltogether.
Thanks,
tglx
Powered by blists - more mailing lists