On Aug 2, 2016 12:36 PM, "Matt Turner" <<a href="mailto:mattst88@gmail.com">mattst88@gmail.com</a>> wrote: > > On Mon, Aug 1, 2016 at 8:59 PM, Kenneth Graunke <<a href="mailto:kenneth@whitecape.org">kenneth@whitecape.org</a>> wrote: > > Previously, for every input, we moved the dispatch mask to the flag > > register, then emitted two predicated PLN instructions, one with > > centroid barycentric coordinates (for normal pixels), and one with > > pixel barycentric coordinates (for unlit helper pixels). > > > > Instead, we can simply emit a set of predicated MOVs at the top of > > the program which copy the pixel barycentric coordinates over the > > centroid ones for unlit helper pixel channels. Then, we can just > > use normal PLNs. I like this approach *much* better than what we did before. Reviewed-by: Jason Ekstrand <<a href="mailto:jason@jlekstrand.net">jason@jlekstrand.net</a>> > > On Sandybridge: > > > > total instructions in shared programs: 7538470 -> 7534500 (-0.05%) > > instructions in affected programs: 101268 -> 97298 (-3.92%) > > helped: 705 > > HURT: 9 (all of which are SIMD16 programs) > > Reviewed-by: Matt Turner <<a href="mailto:mattst88@gmail.com">mattst88@gmail.com</a>> > > Maybe wait a few days before pushing so we can confirm that my patch > indeed doesn't cause any regressions on Gen7. They're pretty much orthogonal so I'm not too worried about that but sure. --Jason