Re: [PATCH] perf_event: fix loss of notification with multi-eventsampling

[Peter Zijlstra]

On Wed, 2011-11-23 at 11:10 +0100, Stephane Eranian wrote:
> On Wed, Nov 23, 2011 at 11:06 AM, Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
> > On Wed, 2011-11-23 at 10:30 +0100, Stephane Eranian wrote:
> >>
> >> In the case of perf_event, we are not in a producer-consumer model, so
> >> it seems like the behavior we have now is correct. The caller of poll()
> >> gets stuck if the file descriptor is closed.
> >
> > But wouldn't out event->waitq still be referenced by that waiting task?
> > Even though we freed it in our fops->release() callback?
> >
> I suspect you don't end up in fops->release() if you have an ongoing poll
> because you've probably increment the file's refcount along the way.

You're quite right, poll does that internally. Ok, so we don't need to
worry about this bit then.
--
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/