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

intel/nir: temporarly disable opt_uniform_atomics for RT/CL 

Not had time to investiguate what is going is on but it's definitely a
contributor to failures.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Acked-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16104>
This commit is contained in:
Lionel Landwerlin 2022-04-21 07:32:23 -07:00 committed by Marge Bot
parent bf29ebc0a7
commit f1dd487531
1 changed files with 10 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
src/intel/compiler/brw_nir.c
View File

@ -1236,8 +1236,17 @@ brw_postprocess_nir(nir_shader *nir, const struct brw_compiler *compiler,
/* TODO: Enable nir_opt_uniform_atomics on Gfx7.x too.
* It currently fails Vulkan tests on Haswell for an unknown reason.
*
* TODO: Using this optimization on RT/OpenCL kernels also seems to cause
* issues. Until we can understand those issues, disable it.
*/
if (devinfo->ver >= 8 && OPT(nir_opt_uniform_atomics)) {
bool opt_uniform_atomic_stage_allowed =
devinfo->ver >= 8 &&
nir->info.stage != MESA_SHADER_KERNEL &&
nir->info.stage != MESA_SHADER_RAYGEN &&
!gl_shader_stage_is_callable(nir->info.stage);
if (opt_uniform_atomic_stage_allowed && OPT(nir_opt_uniform_atomics)) {
const nir_lower_subgroups_options subgroups_options = {
.ballot_bit_size = 32,
.ballot_components = 1,