Jason Ekstrand writes:
> On Fri, Feb 23, 2018 at 9:26 AM, Scott D Phillips <
> scott.d.phill...@intel.com> wrote:
>
> > Jason Ekstrand writes:
> >
> > > There is a potential race between the __sync_fetch_and_add and the
> > > futex_wake where another thread could come in and start waiting.
> > >
On Fri, Feb 23, 2018 at 9:26 AM, Scott D Phillips <
scott.d.phill...@intel.com> wrote:
> Jason Ekstrand writes:
>
> > There is a potential race between the __sync_fetch_and_add and the
> > futex_wake where another thread could come in and start waiting. If we
> > hit this case, the other thread
Jason Ekstrand writes:
> There is a potential race between the __sync_fetch_and_add and the
> futex_wake where another thread could come in and start waiting. If we
> hit this case, the other thread will never get woken back up because the
> futex_wake doesn't get called. We can fix this by cal
There is a potential race between the __sync_fetch_and_add and the
futex_wake where another thread could come in and start waiting. If we
hit this case, the other thread will never get woken back up because the
futex_wake doesn't get called. We can fix this by calling futex_wake
unconditionally.