[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20170705091715.lv57k25kntqh4nic@hirez.programming.kicks-ass.net>
Date: Wed, 5 Jul 2017 11:17:15 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Thomas Gleixner <tglx@...utronix.de>
Cc: Vikram Mulukutla <markivx@...eaurora.org>,
Rusty Russell <rusty@...tcorp.com.au>,
Tejun Heo <tj@...nel.org>,
Andrew Morton <akpm@...ux-foundation.org>,
LKML <linux-kernel@...r.kernel.org>,
Sebastian Sewior <bigeasy@...utronix.de>
Subject: Re: [PATCH] smp/hotplug: Move unparking of percpu threads to the
control CPU
On Wed, Jul 05, 2017 at 11:07:34AM +0200, Thomas Gleixner wrote:
> On Wed, 5 Jul 2017, Peter Zijlstra wrote:
> > OK, so if I get this right we do something like:
> >
> >
> > BP AP
> >
> > bringup_cpu();
> > __cpu_up() ------------> /* stuff */
> > bringup_wait_for_ap()
> > wait_for_completion();
> > cpuhp_online_idle();
> > <------------ complete(&st->done);
> > unpark()
> > while(1)
> > do_idle();
>
> actually I added after unpark():
>
> kick_ap()
> wait_for_completion()
>
> So the AP will execute the online callbacks in its own hotplug thread.
Indeed, but I stopped after the unpark() bits bcause well, that's what
you changed here :-)
Powered by blists - more mailing lists