On Wed, May 04, 2016 at 12:35:09PM +0100, Dave Gordon wrote: > On 29/04/2016 09:49, Chris Wilson wrote: > > On Fri, Apr 29, 2016 at 09:36:37AM +0100, Tvrtko Ursulin wrote: > > On 28/04/16 17:24, Chris Wilson wrote: > > With the introduction of a distinct engine->id vs the hardware id, we need > to fix up the value we use for selecting the target engine when signaling > a semaphore. Note that these values can be merged with engine->guc_id. > > So I broke something more with the decoupling, did not realize. I > suppose it was still worth it. This at least wasn't being used. > > A consolation prize: wean the guc over to a common hw_id :) > -Chris > > > As in, use the GuC's concept of the engine ID for other purposes too? Looks like the HW engine ID predates the GuC and so if they continued to use the same values for the GuC (which for the engines that exist up to and including gen9 they have), the GuC can use the older id and we can have one fewer special GuC value. -Chris -- Chris Wilson, Intel Open Source Technology Centre _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
