intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2026-01-08 14:02:58 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix: unify path for getting number of grfs per thread

Browse Source 
Related-To: NEO-8043

Signed-off-by: Naklicki, Mateusz <mateusz.naklicki@intel.com>
This commit is contained in:
Naklicki, Mateusz
2023-09-22 16:06:50 +00:00
committed by Compute-Runtime-Automation
parent 9337911742
commit 0461af492d
25 changed files with 91 additions and 41 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
shared/source/command_container/command_encoder_bdw_and_later.inl
View File

@@ -31,7 +31,7 @@ namespace NEO {
template <typename Family>
void EncodeDispatchKernel<Family>::setGrfInfo(INTERFACE_DESCRIPTOR_DATA *pInterfaceDescriptor, uint32_t numGrf,
const size_t &sizeCrossThreadData, const size_t &sizePerThreadData,
const HardwareInfo &hwInfo) {
const RootDeviceEnvironment &rootDeviceEnvironment) {
auto grfSize = sizeof(typename Family::GRF);
DEBUG_BREAK_IF((sizeCrossThreadData % grfSize) != 0);
auto numGrfCrossThreadData = static_cast<uint32_t>(sizeCrossThreadData / grfSize);
@@ -163,7 +163,7 @@ void EncodeDispatchKernel<Family>::encode(CommandContainer &container, EncodeDis
}
EncodeDispatchKernel<Family>::setGrfInfo(&idd, kernelDescriptor.kernelAttributes.numGrfRequired, sizeCrossThreadData,
sizePerThreadData, hwInfo);
sizePerThreadData, rootDeviceEnvironment);
uint32_t sizeThreadData = sizePerThreadDataForWholeGroup + sizeCrossThreadData;
bool isHwLocalIdGeneration = false;