intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2025-12-30 01:35:20 +08:00
Code Issues Packages Projects Releases Wiki Activity

Align sync buffer address to max atomic type size

Browse Source 
Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
This commit is contained in:
Filip Hazubski
2021-11-19 14:52:40 +00:00
committed by Compute-Runtime-Automation
parent 6ece3534e8
commit cd3930251e
3 changed files with 18 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
shared/source/program/sync_buffer_handler.inl
View File

@@ -9,7 +9,7 @@
template <typename KernelT>
void NEO::SyncBufferHandler::prepareForEnqueue(size_t workGroupsCount, KernelT &kernel) {
auto requiredSize = workGroupsCount;
auto requiredSize = alignUp(workGroupsCount, CommonConstants::maximalSizeOfAtomicType);
std::lock_guard<std::mutex> guard(this->mutex);
bool isCurrentBufferFull = (usedBufferSize + requiredSize > bufferSize);