intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2026-01-03 14:55:24 +08:00
Code Issues Packages Projects Releases Wiki Activity

performance: do not free deferred allocations when not needed

Browse Source 
Level Zero doesn't use deferred allocations so no point in paying the
price to check for them in cleanAllocationList.

Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
This commit is contained in:
Michal Mrozek
2023-10-27 16:43:33 +00:00
committed by Compute-Runtime-Automation
parent 39cf653959
commit 80e3632c41
5 changed files with 9 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
shared/source/command_stream/command_stream_receiver_hw_base.inl
View File

@@ -1265,6 +1265,7 @@ TaskCountType CommandStreamReceiverHw<GfxFamily>::flushBcsTask(const BlitPropert
if (blocking) {
const auto waitStatus = waitForTaskCountWithKmdNotifyFallback(newTaskCount, flushStampToWait, false, QueueThrottle::MEDIUM);
internalAllocationStorage->cleanAllocationList(newTaskCount, TEMPORARY_ALLOCATION);
internalAllocationStorage->cleanAllocationList(newTaskCount, DEFERRED_DEALLOCATION);
if (waitStatus == WaitStatus::GpuHang) {
return CompletionStamp::gpuHang;