3 commits

Author	SHA1	Message	Date
Caio Oliveira	7ac5696157	intel/brw: Remove Gfx8- code from backend passes ... Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27691>	2024-02-28 05:45:38 +00:00
Ian Romanick	8fb37ef985	intel/fs: Add fast path for ballot(true) ... This doesn't help very much now. A later commit adds a NIR optimization pass, tentatively called nir_opt_uniform_subgroup, that converts many kinds of subgroup operations to things involving bitCount(ballot(true)). This commit makes a huge difference in the results of that later commit. No shader-db changes on any Intel platform. Fossil-db results: All Intel platforms had similar results. (Ice Lake shown) Totals: Instrs: 165558033 -> 165557519 (-0.00%) Cycles: 15156188362 -> 15156178922 (-0.00%); split: -0.00%, +0.00% Totals from 299 (0.05% of 656117) affected shaders: Instrs: 88293 -> 87779 (-0.58%) Cycles: 3709498 -> 3700058 (-0.25%); split: -0.28%, +0.03% v2: Rebase on splitting ELK from BRW. Remove devinfo->ver >= 8 check. Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27044>	2024-02-27 08:37:46 -08:00
Caio Oliveira	4fe3498e72	intel/brw: Move small lowering passes into brw_fs_lower.cpp ... Larger lowering passes will go to their own files. Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Acked-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26887>	2024-02-26 20:54:25 +00:00