intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2025-12-24 12:23:05 +08:00
Code Issues Packages Projects Releases Wiki Activity

[perf] Set local memory pool for kmd-migrated shared allocation

Browse Source 
Select ccs engine for usm device and shared allocations
(i.e. for transfers from local to local).

Related-To: NEO-7252

Signed-off-by: Milczarek, Slawomir <slawomir.milczarek@intel.com>
This commit is contained in:
Milczarek, Slawomir
2023-03-15 14:19:17 +00:00
committed by Compute-Runtime-Automation
parent a23230ebcf
commit eff7439eca
2 changed files with 4 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
shared/source/os_interface/linux/drm_memory_manager.cpp
View File

@@ -2155,7 +2155,7 @@ GraphicsAllocation *DrmMemoryManager::createSharedUnifiedMemoryAllocation(const
auto gmmHelper = getGmmHelper(allocationData.rootDeviceIndex);
auto canonizedGpuAddress = gmmHelper->canonize(reinterpret_cast<uintptr_t>(cpuPointer));
auto allocation = std::make_unique<DrmAllocation>(allocationData.rootDeviceIndex, allocationData.type, bos, cpuPointer, canonizedGpuAddress, size, MemoryPool::System4KBPages);
auto allocation = std::make_unique<DrmAllocation>(allocationData.rootDeviceIndex, allocationData.type, bos, cpuPointer, canonizedGpuAddress, size, MemoryPool::LocalMemory);
allocation->setMmapPtr(cpuBasePointer);
allocation->setMmapSize(totalSizeToAlloc);
if (!allocation->setCacheRegion(&drm, static_cast<CacheRegion>(allocationData.cacheRegion))) {