On Fri, Nov 13, 2015 at 04:55:33PM +0100, Neil Roberts wrote: > Hi, > > You don't seem to have included any of the suggestions I made in my > review. Was this deliberate? If not, the main points were: >
It was not intentional, I apologize. It just got lost on top of Chad's feedback which conflicted with yours in these two statements - he asked for something similar and I didn't go back to check exactly what you said, assuming it was the same thing. > • You don't need to call brw_fast_clear_init or use_rectlist in the > function because these are already called before entering it. > > • I don't think it's worth creating a framebuffer. Instead you can just > call _mesa_meta_drawbuffers_from_bitfield(1 << index) in the loop. > Modifying the draw buffers state should be ok because it's saved in > the meta state and it's already done for the Gen8 code path. > > I went ahead and tried the changes in a patch here: > > https://github.com/bpeel/mesa/commit/b2aa8f2d90572392030e5177952bf > > It doesn't cause any Jenkins regressions. Feel free to squash it into > the patch if you want, or of course if you prefer to keep your patch as > it is it's up to you. Thanks a lot, I will squash it in - and sorry again about ignoring your feedback. > > Regards, > - Neil > > Ben Widawsky <benjamin.widaw...@intel.com> writes: [snip] _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
