Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755440AbYGYMku (ORCPT ); Fri, 25 Jul 2008 08:40:50 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752585AbYGYMkm (ORCPT ); Fri, 25 Jul 2008 08:40:42 -0400 Received: from viefep31-int.chello.at ([62.179.121.49]:61230 "EHLO viefep31-int.chello.at" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750927AbYGYMkl (ORCPT ); Fri, 25 Jul 2008 08:40:41 -0400 Subject: Re: [patch, rfc: 2/2] sched, hotplug: ensure a task is on the valid cpu after set_cpus_allowed_ptr() From: Peter Zijlstra To: Dmitry Adamushko Cc: Ingo Molnar , LKML In-Reply-To: <1216937730.5368.16.camel@earth> References: <1216937730.5368.16.camel@earth> Content-Type: text/plain Date: Fri, 25 Jul 2008 14:40:49 +0200 Message-Id: <1216989649.7257.381.camel@twins> Mime-Version: 1.0 X-Mailer: Evolution 2.22.3.1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3001 Lines: 82 On Fri, 2008-07-25 at 00:15 +0200, Dmitry Adamushko wrote: > > From: Dmitry Adamushko > Subject: sched, hotplug: ensure a task is on the valid cpu after > set_cpus_allowed_ptr() > > --- > sched, hotplug: ensure a task is on the valid cpu after set_cpus_allowed_ptr() > > The 'new_mask' may not include task_cpu(p) so we migrate 'p' on another 'cpu'. > In case it can't be placed on this 'cpu' immediately, we submit a request > to the migration thread and wait for its completion. > > Now, by the moment this request gets handled by the migration_thread, > 'cpu' may well be offline/non-active. As a result, 'p' continues > running on its old cpu which is not in the 'new_mask'. > > Fix it: ensure 'p' ends up on a valid cpu. > > Theoreticaly (but unlikely), we may get an endless loop if someone cpu_down()'s > a new cpu we have choosen on each iteration. > > Alternatively, we may introduce a special type of request to migration_thread, > namely "move_to_any_allowed_cpu" (e.g. by specifying dest_cpu == -1). > > Note, any_active_cpu() instead of any_online_cpu() would be better here. Hrmm,.. this is all growing into something of a mess.. defeating the whole purpose of introducing that cpu_active_map stuff. Would the suggested SRCU logic simplify all this? > Signed-off-by: Dmitry Adamushko > > diff --git a/kernel/sched.c b/kernel/sched.c > index b4ccc8b..c3bd78a 100644 > --- a/kernel/sched.c > +++ b/kernel/sched.c > @@ -5774,21 +5774,23 @@ int set_cpus_allowed_ptr(struct task_struct *p, const cpumask_t *new_mask) > } > > /* Can the task run on the task's current CPU? If so, we're done */ > - if (cpu_isset(task_cpu(p), *new_mask)) > - goto out; > + while (!cpu_isset(task_cpu(p), p->cpus_allowed)) { > + int cpu = any_online_cpu(p->cpus_allowed); > > - if (migrate_task(p, any_online_cpu(*new_mask), &req)) { > - /* Need to wait for migration thread (might exit: take ref). */ > - struct task_struct *mt = rq->migration_thread; > + if (migrate_task(p, cpu, &req)) { > + /* Need to wait for migration thread (might exit: take ref). */ > + struct task_struct *mt = rq->migration_thread; > > - get_task_struct(mt); > - task_rq_unlock(rq, &flags); > - wake_up_process(mt); > - put_task_struct(mt); > + get_task_struct(mt); > + task_rq_unlock(rq, &flags); > + wake_up_process(mt); > + put_task_struct(mt); > > - wait_for_completion(&req.done); > - tlb_migrate_finish(p->mm); > - return 0; > + wait_for_completion(&req.done); > + tlb_migrate_finish(p->mm); > + > + rq = task_rq_lock(p, &flags); > + } > } > out: > task_rq_unlock(rq, &flags); > > -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/