Re: [PATCH]CPU hotplug breaks wake_up_new_task

[Nick Piggin]

Shaohua Li wrote:
> On Tue, 2005-05-31 at 18:50 +1000, Nick Piggin wrote:
>
> > Shaohua Li wrote:
> >
> >
> > > I must be over considering. Ok, how does this updated one look?
> >
> > Looks like you've found a race, alright. Nice work!
> >
> > I think it would be preferable to do the check in kernel/fork.c,
> > after the tasklist lock is taken (and you'll need to rediff the
> > patch for the -mm tree).
>
> It seems there is still a race between copy_process and wake_up_new_task
> to me (cpu offline after copy_process). Am I missing anything?

Offlining the CPU takes the tasklist lock to migrate off tasks.
So either the CPU will be offline first, in which case the a
check in kernel/fork.c will pick that up; or the task will be
added to the tasklist first, in which case CPU hotplug should
correctly migrate it away.

I think?

--
SUSE Labs, Novell Inc.

Send instant messages to your online friends http://au.messenger.yahoo.com 
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/