intel/fs: Add support for subgroup quad operations
authorJason Ekstrand <jason.ekstrand@intel.com>
Fri, 1 Sep 2017 22:18:02 +0000 (15:18 -0700)
committerJason Ekstrand <jason.ekstrand@intel.com>
Wed, 7 Mar 2018 20:13:47 +0000 (12:13 -0800)
commit8b4a5e641bc3cb9cf0cfe7d0487926127fc25de7
tree677f320e6c04883a55a4f2592549428154972423
parent2292b20b2969c9e3e0494ccc55c6216f330762ae
intel/fs: Add support for subgroup quad operations

NIR has code to lower these away for us but we can do significantly
better in many cases with register regioning and SIMD4x2.

Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
src/intel/compiler/brw_eu_defines.h
src/intel/compiler/brw_fs.cpp
src/intel/compiler/brw_fs_generator.cpp
src/intel/compiler/brw_fs_nir.cpp
src/intel/compiler/brw_shader.cpp