On Thu, Feb 02, 2017 at 12:49:58PM +0000, Tvrtko Ursulin wrote:
>
> On 02/02/2017 09:08, Chris Wilson wrote:
> >+static bool wait_for_ready(struct igt_wakeup *w)
> >+{
> >+ DEFINE_WAIT(ready);
> >+
> >+ if (atomic_dec_and_test(w->done))
> >+ wake_up_atomic_t(w->done);
> >+
> >+ if (test_bit(STOP, &w->flags))
> >+ goto out;
> >+
> >+ set_bit(IDLE, &w->flags);
>
> I think this needs to be before atomic_dec_and_test(w->done), to
> avoid that same assert racing with the threads. Because immediately
> after the wake_up_atomic above the main loop starts asserting the
> IDLE bit which is not guaranteed to be set yet.
Before wake_up_atomic_t which is the same thing, yup.
-Chris
--
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
