Re: [PATCH 3/3] [RFC] ipc/sem.c: replace busy loop with completion

[Manfred Spraul]

On 09/26/2011 10:00 AM, Peter Zijlstra wrote:
> On Sat, 2011-09-24 at 19:37 +0200, Manfred Spraul wrote:
>
> 	sma = sem_lock(ns, semid);
>
> > @@ -1456,8 +1414,17 @@ sleep_again:
> >          /*
> >           * Wait until it's guaranteed that no wakeup_sem_queue_do() is ongoing.
> >           */
> > -       error = get_queue_result(&queue);
> > -
> > +       error = queue.status;
> > +       if (error != -EINTR) {
> > +               /* If there is a return code, then we can leave immediately. */
> > +               if (!IS_ERR(sma)) {
> > +                       sem_unlock(sma);
> > +               }
> > +               /* Except that we must wait for the hands-off */
> > +               wait_for_completion(&queue.done);
> > +               goto out_free;
> 		you just lost a sem_unlock() in the IS_ERR() case.
>
> > +       }
If IS_ERR(), then sem_lock() failed - thus there must be no unlock.

--
    Manfred
--
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/