intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2026-01-09 06:23:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

performance(ocl): cmd buffer prealloc per cmdqueue

Browse Source 
Add mechanism to preallocate cmd buffer allocations in command stream
receiver reusable allocations list per command queue initialized.

This should limit additional allocations during hot loop.

Needs to be enabled in subsequent commits by setting product helper
method.

Related-To: NEO-8152

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
This commit is contained in:
Dominik Dabek
2023-10-27 15:54:45 +00:00
committed by Compute-Runtime-Automation
parent cfbf6219fe
commit 39cf653959
13 changed files with 120 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
shared/test/common/mocks/mock_product_helper.cpp
View File

@@ -367,6 +367,11 @@ uint32_t ProductHelperHw<IGFX_UNKNOWN>::getMaxNumSamplers() const {
return 0u;
}
template <>
uint32_t ProductHelperHw<IGFX_UNKNOWN>::getCommandBuffersPreallocatedPerCommandQueue() const {
return 0u;
}
template <>
uint32_t L1CachePolicyHelper<IGFX_UNKNOWN>::getL1CachePolicy(bool isDebuggerActive) {
return L1CachePolicyHelper<IGFX_UNKNOWN>::getDefaultL1CachePolicy(isDebuggerActive);