mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Gautham R Shenoy <ego@in.ibm.com>
To: Gregory Haskins <ghaskins@novell.com>
Cc: suresh.b.siddha@intel.com, rjw@sisk.pl,
	akpm@linux-foundation.org, dmitry.adamushko@gmail.com,
	mingo@elte.hu, oleg@sign.ru, yi.y.yang@intel.com,
	linux-kernel@vger.kernel.org, tglx@linutronix.de
Subject: Re: [PATCH] keep rd->online and cpu_online_map in sync
Date: Mon, 10 Mar 2008 19:51:17 +0530	[thread overview]
Message-ID: <20080310142117.GG17646@in.ibm.com> (raw)
In-Reply-To: <20080310133755.4689.83217.stgit@novell1.haskins.net>

On Mon, Mar 10, 2008 at 09:39:34AM -0400, Gregory Haskins wrote:
> >>> On Mon, Mar 10, 2008 at  4:14 AM, in message
> <20080310081425.GA11031@in.ibm.com>, Gautham R Shenoy <ego@in.ibm.com> wrote:
> 
> > On Sat, Mar 08, 2008 at 12:10:15AM -0500, Gregory Haskins wrote:
> 
> [ snip ]
> 
> >> @@ -5813,6 +5813,13 @@ migration_call(struct notifier_block *nfb, unsigned 
> > long action, void *hcpu)
> >>  		/* Must be high prio: stop_machine expects to yield to it. */
> >>  		rq = task_rq_lock(p, &flags);
> >>  		__setscheduler(rq, p, SCHED_FIFO, MAX_RT_PRIO-1);
> >> +
> >> +		/* Update our root-domain */
> >> +		if (rq->rd) {
> >> +			BUG_ON(!cpu_isset(cpu, rq->rd->span));
> >> +			cpu_set(cpu, rq->rd->online);
> >> +		}
> > 
> > Hi Greg,
> > 
> > Suppose someone issues a wakeup at some point between CPU_UP_PREPARE
> > and CPU_ONLINE, then isn't there a possibility that the task could
> > be woken up on the cpu which has not yet come online ?
> > Because at this point in find_lowest_cpus()
> > 
> > cpus_and(*lowest_mask, task_rq(task)->rd->online, task->cpus_allowed);
> > 
> > the cpu which has not yet come online is set in both rd->online map
> > and the task->cpus_allowed.
> 
> Hi Gautham,
>    This is a good point.  I think I got it backwards (or rather, I had the ordering more correct before the patch):  I really want to keep the set() on ONLINE as I had it originally.  Its the clear() that is misplaced, as DOWN_PREPARE is too early in the chain.  I probably should have used DYING to defer the clear out to the point right before the cpu_online_map is updated.  Thanks to Dmitry for highlighting the smpboot path which helped me to find the proper notifier symbol. 

I think CPU_DYING is a more apt point to clear the bit.
Vatsa had infact suggested that we do it in cpu_disable(), but since we
have an event, we might as well make use of it.
> 
> > 
> > I wonder if assigning a priority to the update_sched_domains() notifier
> > so that it's called immediately after migration_call() would solve the
> > problem.
> 
> Possibly (and I just saw your patch).  But note that migration_call was previously declared as "should run first" so I am not sure if moving update_sched_domain() above it will not have a ripple effect in some other area.  If that is "ok", then I agree your solution solves the ordering problem, albeit in a bit heavy handed manner.  Otherwise s/CPU_DOWN_PREPARE/CPU_DYING in migrate call will fix the issue as well, I believe.  This will push the update of rd->online to be much more tightly coupled with updates to cpu_online_map.

The reason why migration_call() was declared as "should run first" was
to ensure that all the migration happens before the other subsystems run
their CPU_DEAD events. This is important, else when the subsystems issue
a kthread_stop(subsystem_thread) in their CPU_DEAD case, the
subsystem_thread might still be affined to the offlined cpu, thus
resulting in a deadlock.

That requirement is taken care of in the patch I had sent since
migration_call() will be called immediately after update_sched_domains().

> 
> Ingo/Andrew: Please drop the earlier fix I submitted.  I don't think it is correct on several fronts.  Please try the one below.  As before, I have tested that I can offline/online CPUs, but I dont have s2ram capability here.
> 
> Regards,
> -Greg
> 
> ---------------------------------------------
> keep rd->online and cpu_online_map in sync
> 
> It is possible to allow the root-domain cache of online cpus to
> become out of sync with the global cpu_online_map.  This is because we
> currently trigger removal of cpus too early in the notifier chain.
> Other DOWN_PREPARE handlers may in fact run and reconfigure the
> root-domain topology, thereby stomping on our own offline handling.
> 
> The end result is that rd->online may become out of sync with
> cpu_online_map, which results in potential task misrouting.
> 
> So change the offline handling to be more tightly coupled with the
> global offline process by triggering on CPU_DYING intead of
> CPU_DOWN_PREPARE.
> 
> Signed-off-by: Gregory Haskins <ghaskins@novell.com>
> ---
> 
>  kernel/sched.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/kernel/sched.c b/kernel/sched.c
> index 52b9867..a616fa1 100644
> --- a/kernel/sched.c
> +++ b/kernel/sched.c
> @@ -5881,7 +5881,7 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
>  		spin_unlock_irq(&rq->lock);
>  		break;
> 
> -	case CPU_DOWN_PREPARE:
> +	case CPU_DYING:
>  		/* Update our root-domain */
>  		rq = cpu_rq(cpu);
>  		spin_lock_irqsave(&rq->lock, flags);

-- 
Thanks and Regards
gautham

  reply	other threads:[~2008-03-10 14:21 UTC|newest]

Thread overview: 54+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-03-02 18:42 [BUG 2.6.25-rc3] scheduler/hotplug: some processes are dealocked when cpu is set to offline Yi Yang
2008-03-03 11:54 ` Dmitry Adamushko
2008-03-03 11:56   ` Ingo Molnar
2008-03-03 12:02     ` Dmitry Adamushko
2008-03-03 14:53       ` Yi Yang
2008-03-03 17:37         ` Yi Yang
2008-03-03 15:31 ` Gautham R Shenoy
2008-03-03 14:45   ` Yi Yang
2008-03-04  5:26     ` Gautham R Shenoy
2008-03-04  9:09       ` Gautham R Shenoy
2008-03-03 21:56         ` Yi Yang
2008-03-04 15:01       ` Oleg Nesterov
2008-03-04 14:37         ` Yi Yang
2008-03-06 20:05           ` Yi Yang
2008-03-05 10:05         ` Gautham R Shenoy
2008-03-05 13:53           ` Oleg Nesterov
2008-03-06 11:15             ` Gautham R Shenoy
2008-03-06 12:22               ` Gautham R Shenoy
2008-03-06 13:44         ` Gautham R Shenoy
2008-03-07  2:54           ` Oleg Nesterov
2008-03-07  9:10             ` Gautham R Shenoy
2008-03-07 10:51               ` Gautham R Shenoy
2008-03-06 23:20                 ` Yi Yang
2008-03-07 13:02                 ` Dmitry Adamushko
2008-03-07 13:55                   ` Gautham R Shenoy
2008-03-07 15:50                     ` Gautham R Shenoy
2008-03-07 19:14                       ` [BUG 2.6.25-rc3] scheduler/hotplug: some processes aredealocked " Suresh Siddha
2008-03-07 20:18                   ` [BUG 2.6.25-rc3] scheduler/hotplug: some processes are dealocked " Andrew Morton
2008-03-07 21:36                     ` Rafael J. Wysocki
2008-03-07 23:01                       ` Suresh Siddha
2008-03-07 23:29                         ` Andrew Morton
2008-03-07 23:43                           ` Rafael J. Wysocki
2008-03-08  1:50                             ` Suresh Siddha
2008-03-08  2:09                               ` Andrew Morton
2008-03-08  5:10                               ` [PATCH] adjust root-domain->online span in response to hotplug event Gregory Haskins
2008-03-08  8:41                                 ` Ingo Molnar
2008-03-08 17:50                                   ` [PATCH] adjust root-domain->online span in response to hotplugevent Gregory Haskins
2008-03-09  0:31                                     ` Dmitry Adamushko
2008-03-10 14:12                                       ` Gregory Haskins
2008-03-09  2:35                                 ` [PATCH] adjust root-domain->online span in response to hotplug event Suresh Siddha
2008-03-10 12:41                                   ` Gregory Haskins
2008-03-10  8:14                                 ` Gautham R Shenoy
2008-03-10 13:13                                   ` [PATCH] cpu-hotplug: Register update_sched_domains() notifier with higher prio Gautham R Shenoy
2008-03-10 22:25                                     ` Andrew Morton
2008-03-10 13:39                                   ` [PATCH] keep rd->online and cpu_online_map in sync Gregory Haskins
2008-03-10 14:21                                     ` Gautham R Shenoy [this message]
2008-03-10 18:12                                     ` Suresh Siddha
2008-03-10 22:03                                       ` Rafael J. Wysocki
2008-03-10 22:00                                         ` Gregory Haskins
2008-03-10 22:10                                           ` Suresh Siddha
2008-03-10 21:59                                             ` [PATCH v2] " Gregory Haskins
2008-03-10 23:36                                               ` Andrew Morton
2008-03-11  1:34                                                 ` Suresh Siddha
2008-03-11  4:39                                                   ` Gautham R Shenoy

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20080310142117.GG17646@in.ibm.com \
    --to=ego@in.ibm.com \
    --cc=akpm@linux-foundation.org \
    --cc=dmitry.adamushko@gmail.com \
    --cc=ghaskins@novell.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=oleg@sign.ru \
    --cc=rjw@sisk.pl \
    --cc=suresh.b.siddha@intel.com \
    --cc=tglx@linutronix.de \
    --cc=yi.y.yang@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

Powered by JetHome