intel/compute-runtime
1
0
Fork 0
mirror of https://github.com/intel/compute-runtime.git synced 2025-12-24 21:18:24 +08:00
Code Issues Packages Projects Releases Wiki Activity

Turn ON DrmGemCloseWorker.

Browse Source 
- For every command buffer that we submit, pass it to gem close worker.
- Gem close worker will do asynchronous cleanup if this resource is meant to
be destroyed.
- if the resource is not meant to be destroyed we will call IOCTL wait for
this batch buffer.
- This will result in bumping up GPU clocks and better performance.

Change-Id: If9f181e411d7748573f31682e875a97c5355abe5
This commit is contained in:
Mrozek, Michal
2018-05-11 09:57:59 +02:00
committed by sys_ocldev
parent 562ededffe
commit cec056f3c4
6 changed files with 30 additions and 46 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
runtime/os_interface/linux/drm_command_stream.h
View File

@@ -48,7 +48,7 @@ class DrmCommandStreamReceiver : public DeviceCommandStreamReceiver<GfxFamily> {
public:
// When drm is null default implementation is used. In this case DrmCommandStreamReceiver is responsible to free drm.
// When drm is passed, DCSR will not free it at destruction
DrmCommandStreamReceiver(const HardwareInfo &hwInfoIn, Drm *drm, gemCloseWorkerMode mode = gemCloseWorkerMode::gemCloseWorkerInactive);
DrmCommandStreamReceiver(const HardwareInfo &hwInfoIn, Drm *drm, gemCloseWorkerMode mode = gemCloseWorkerMode::gemCloseWorkerActive);
FlushStamp flush(BatchBuffer &batchBuffer, EngineType engineType, ResidencyContainer *allocationsForResidency) override;
void makeResident(GraphicsAllocation &gfxAllocation) override;