intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2025-12-25 05:24:02 +08:00
Code Issues Packages Projects Releases Wiki Activity

Dynamic queue size limit in RelaxedOrdering mode

Browse Source 
Related-To: NEO-7458

Signed-off-by: Dunajski, Bartosz <bartosz.dunajski@intel.com>
This commit is contained in:
Dunajski, Bartosz
2023-02-01 17:06:21 +00:00
committed by Compute-Runtime-Automation
parent 756a9a9aaa
commit 6ebdc51fae
13 changed files with 203 additions and 24 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
level_zero/core/source/cmdqueue/cmdqueue.cpp
View File

@@ -114,7 +114,7 @@ NEO::SubmissionStatus CommandQueueImp::submitBatchBuffer(size_t offset, NEO::Res
NEO::BatchBuffer batchBuffer(commandStream.getGraphicsAllocation(), offset, 0, 0, nullptr, false, false,
NEO::QueueThrottle::HIGH, NEO::QueueSliceCount::defaultSliceCount,
commandStream.getUsed(), &commandStream, endingCmdPtr, false, false);
commandStream.getUsed(), &commandStream, endingCmdPtr, csr->getNumClients(), false, false);
commandStream.getGraphicsAllocation()->updateTaskCount(csr->peekTaskCount() + 1, csr->getOsContext().getContextId());
commandStream.getGraphicsAllocation()->updateResidencyTaskCount(csr->peekTaskCount() + 1, csr->getOsContext().getContextId());