On Thu, Nov 03, 2016 at 08:33:58AM +0000, Tvrtko Ursulin wrote: > On 02/11/2016 17:34, Chris Wilson wrote: > >It looks ok, but I just don't see the point. wq->flags is private to the > >wq->func callback. > > A very superficial skim shows that wake_up_common at least looks at > the flags. So I thought, as long as wake_up_something gets called > form somewhere on these ones, it would be safer not to potentially > silently collide with some core flag. > > Or are you saying no one ever touches them outside i915_sw_fence.c ? That would break the encapsulation of the waitqueue being inside the fence. I was not intending for people to call wake_up_all(fence.wait), kfence_wake_up_all() [bad name, I was trying to borrow the concept from wake_up_all() but it is not the same, it is for the completion of a deferred signal notify] completes the fence in addition to signaling its waiters. -Chris -- Chris Wilson, Intel Open Source Technology Centre _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
