intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2026-01-11 00:10:58 +08:00
Code Issues Packages Projects Releases Wiki Activity

Revert "fix: MaxSubSlicesSupported should be set to total non-fused count"

Browse Source 
This reverts commit ee0022d234.

Signed-off-by: Compute-Runtime-Validation <compute-runtime-validation@intel.com>
This commit is contained in:
Compute-Runtime-Validation
2025-08-07 01:39:46 +02:00
committed by Compute-Runtime-Automation
parent 3f1f3697ba
commit b40830c1e6
8 changed files with 62 additions and 62 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
shared/source/gen12lp/gfx_core_helper_gen12lp.cpp
View File

@@ -155,7 +155,7 @@ uint32_t GfxCoreHelperHw<Family>::getComputeUnitsUsedForScratch(const RootDevice
ThreadCount/EUCount=7 is no longer valid, so we have to force 8 in below formula.
This is required to allocate enough scratch space. */
auto hwInfo = rootDeviceEnvironment.getHardwareInfo();
return NEO::GfxCoreHelper::getHighestEnabledDualSubSlice(*hwInfo) * hwInfo->gtSystemInfo.MaxEuPerSubSlice * 8;
return hwInfo->gtSystemInfo.MaxSubSlicesSupported * hwInfo->gtSystemInfo.MaxEuPerSubSlice * 8;
}
template <>