AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

intel/fs: only avoid SIMD32 if strictly inferior in throughput

Browse Source 
This enabled SIMD32 in blorp shaders and seems to be give a small FPS
bump when using a DG2 GPU as secondary (requires copies to linear
buffers to exchange with main GPU).

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Acked-by: Matt Turner <mattst88@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19341>
This commit is contained in:
Lionel Landwerlin
2022-10-27 13:17:45 +03:00
committed by Marge Bot
parent 480308c6e5
commit 6b494745be
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/intel/compiler/brw_fs.cpp
View File

@@ -7625,7 +7625,7 @@ brw_compile_fs(const struct brw_compiler *compiler,
} else {
const performance &perf = v32->performance_analysis.require();
if (!INTEL_DEBUG(DEBUG_DO32) && throughput >= perf.throughput) {
if (!INTEL_DEBUG(DEBUG_DO32) && throughput > perf.throughput) {
brw_shader_perf_log(compiler, params->log_data,
"SIMD32 shader inefficient\n");
} else {