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

intel/blorp: Program DispatchWalkOrder and ThreadGroupBatchSize with optimized values for regular computer walkers

Browse Source 
It was only added to indirect compute walkers while HSD don't say
anything about this optimization be specific to indirect compute
walkers.

Reviewed-by: Sagar Ghuge <sagar.ghuge@intel.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36058>
This commit is contained in:
José Roberto de Souza
2025-07-10 11:53:05 -07:00
committed by Marge Bot
parent 06997a42a3
commit aea519cbc2
1 changed files with 5 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/intel/blorp/blorp_genX_exec_brw.h
+5
View File
@@ -1785,6 +1785,11 @@ blorp_exec_compute(struct blorp_batch *batch, const struct blorp_params *params)
.TileLayout = cs_prog_data->walk_order == INTEL_WALK_ORDER_YXZ ?
TileY32bpe : Linear,
#endif
#if GFX_VER >= 30
/* HSD 14016252163 */
.DispatchWalkOrder = cs_prog_data->uses_sampler ? MortonWalk : LinearWalk,
.ThreadGroupBatchSize = cs_prog_data->uses_sampler ? TG_BATCH_4 : TG_BATCH_1,
#endif
.InterfaceDescriptor = (struct GENX(INTERFACE_DESCRIPTOR_DATA)) {
.KernelStartPointer = params->cs_prog_kernel,