intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2026-01-11 16:45:25 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix: Unify logic calculating threads per work group part 2

Browse Source 
- use calculateNumThreadsPerThreadGroup instead of getThreadsPerWG to
have same flow and proper values of threads per work groups

Related-To: NEO-8087
Signed-off-by: Cencelewska, Katarzyna <katarzyna.cencelewska@intel.com>
This commit is contained in:
Cencelewska, Katarzyna
2023-07-03 15:40:04 +00:00
committed by Compute-Runtime-Automation
parent c3e2e145c5
commit 2e17c21728
19 changed files with 88 additions and 53 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
shared/source/command_container/command_encoder_bdw_and_later.inl
View File

@@ -183,7 +183,7 @@ void EncodeDispatchKernel<Family>::encode(CommandContainer &container, EncodeDis
auto implicitArgsCrossThreadPtr = ptrOffset(const_cast<uint64_t *>(reinterpret_cast<const uint64_t *>(args.dispatchInterface->getCrossThreadData())), kernelDescriptor.payloadMappings.implicitArgs.implicitArgsBuffer);
*implicitArgsCrossThreadPtr = implicitArgsGpuVA;
ptr = NEO::ImplicitArgsHelper::patchImplicitArgs(ptr, *pImplicitArgs, kernelDescriptor, {});
ptr = NEO::ImplicitArgsHelper::patchImplicitArgs(ptr, *pImplicitArgs, kernelDescriptor, {}, gfxCoreHelper);
}
memcpy_s(ptr, sizeCrossThreadData,