On 17/03/2017 10:34, Chris Wilson wrote:
On Fri, Mar 17, 2017 at 10:30:10AM +0000, Tvrtko Ursulin wrote:
On 16/03/2017 14:53, Chris Wilson wrote:
Do an early read of the execlists' queue before we take the spinlock and
start checking. This is safe as the first writer to the execlists queue
will cause the tasklet to be run again after a memory barrier.
Which one is the memory barrier, tasklet_hi_schedule? Wouldn't we
need an explicit one after setting engine->execlist_first in
execlists_submit_request now?
Yes, the tasklet scheduling incurs a memory barrier, which we do after
setting execlist_first i.e. we know that after doing so there is always
at least one more invocation of the tasklet.
Okay yes, test_and_set_bit in tasklet_schedule is a memory barrier.
See that Michal beat me to it, but anyway:
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
Regards,
Tvrtko
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
