P4 to Git Change 1982729 by gandryey@gera-win10 on 2019/08/13 17:40:55

SWDEV-79445 - OCL generic changes and code clean-up
	- Use max number of waves per SIMD in the scratch calculation to allow async kernel execution with the scratch buffer

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldevice.cpp#155 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldevice.hpp#43 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palvirtual.cpp#147 edit

rocclr/runtime/device/pal/palvirtual.cpp

@@ -2444,7 +2444,7 @@ bool VirtualGPU::submitKernelInternal(const amd::NDRangeContainer& sizes, const
       // Use maximum available slots for all dispatches to allow async on the same queue
       // HW value loaded into SGPR is an offset value calculated as
       // wave_slot * COMPUTE_TMPRING_SIZE.WAVESIZE
-      dispatchParam.workitemPrivateSegmentSize = scratch->privateMemSize_;
+      dispatchParam.workitemPrivateSegmentSize = std::max(hsaKernel.spillSegSize(), scratch->privateMemSize_);
     }
     dispatchParam.pCpuAqlCode = hsaKernel.cpuAqlCode();
     dispatchParam.hsaQueueVa = hsaQueueMem_->vmAddress();