intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2026-01-07 12:42:54 +08:00
Code Issues Packages Projects Releases Wiki Activity

feature: adding flag to block dispatch implicit scaling commands

Browse Source 
- this feature is part of making compute walker command view
- compute walker is programed for implicit scaling but not dispatched
- together with new flag, comes the refactor to reduce number of arguments

Related-To: NEO-11972

Signed-off-by: Zbigniew Zdanowicz <zbigniew.zdanowicz@intel.com>
This commit is contained in:
Zbigniew Zdanowicz
2024-07-29 13:37:30 +00:00
committed by Compute-Runtime-Automation
parent 2b71ffa7ce
commit b33fe6ccf1
16 changed files with 371 additions and 221 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
shared/source/xe_hpg_core/implicit_scaling_xe_hpg_core.cpp
View File

@@ -19,10 +19,8 @@ template <>
bool ImplicitScalingDispatch<Family>::pipeControlStallRequired = true;
template struct ImplicitScalingDispatch<Family>;
template void ImplicitScalingDispatch<Family>::dispatchCommands<DefaultWalkerType>(LinearStream &commandStream, DefaultWalkerType &walkerCmd, void **outWalkerPtr,
const DeviceBitfield &devices, RequiredPartitionDim requiredPartitionDim, uint32_t &partitionCount,
bool useSecondaryBatchBuffer, bool apiSelfCleanup, bool dcFlush, bool forceExecutionOnSingleTile,
uint64_t workPartitionAllocationGpuVa, const HardwareInfo &hwInfo);
template void ImplicitScalingDispatch<Family>::dispatchCommands<DefaultWalkerType>(LinearStream &commandStream, DefaultWalkerType &walkerCmd, const DeviceBitfield &devices,
ImplicitScalingDispatchCommandArgs &dispatchCommandArgs);
template size_t ImplicitScalingDispatch<Family>::getSize<DefaultWalkerType>(bool apiSelfCleanup, bool preferStaticPartitioning, const DeviceBitfield &devices, const Vec3<size_t> &groupStart, const Vec3<size_t> &groupCount);
} // namespace NEO