P4 to Git Change 1069927 by skudchad@skudchad_test_win_opencl2 on 2014/08/25 14:51:55

ECR #304775 - Optimization for rectangular copies(Part2). Due to HW restriction of 14bits for src and dst pitch, its advantageous to choose optimal bpp. Higher the bpp the larger the byte pitch. This indirectly helps to reduce the number of packets for buffer copy(line by line vs a single sub_win raw packet)

	ReviewBoardURL = http://ocltc.amd.com/reviews/r/5605/diff/

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpublit.cpp#109 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuresource.cpp#191 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuresource.hpp#76 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLContext.cpp#64 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLContext.h#38 edit

rocclr/runtime/device/gpu/gslbe/src/rt/GSLContext.cpp

@@ -513,7 +513,8 @@ CALGSLContext::copyPartial(GpuEvent&      event,
                            size_t*          destOffset,
                            size_t*          size,
                            CALmemcopyflags  flags,
-                           bool             enableRectCopy)
+                           bool             enableRectCopy,
+                           uint32           bytesPerElement)
 {
     uint64      surfaceSize;
     uint32      mode = GSL_SYNCUPLOAD_IGNORE_ELEMENTSIZE;
@@ -563,7 +564,7 @@ CALGSLContext::copyPartial(GpuEvent&      event,
             }
             m_cs->syncUploadRawRect(srcMem, srcOffset[0], (uint32)srcOffset[1], (uint32)srcOffset[2],
                                     destMem, destOffset[0], (uint32)destOffset[1], (uint32)destOffset[2],
-                                    size[0], (uint32)size[1], (uint32)size[2], mode);
+                                    size[0], (uint32)size[1], (uint32)size[2], mode, bytesPerElement);
         }
         else
         {

rocclr/runtime/device/gpu/gslbe/src/rt/GSLContext.h

@@ -71,7 +71,7 @@ public:
     void             destroyProgramObject(gslProgramObject func);
 
     bool             copyPartial(GpuEvent& event, gslMemObject srcMem, size_t* srcOffset,
-                        gslMemObject destMem, size_t* destOffset, size_t* size, CALmemcopyflags flags, bool enableCopyRect);
+                        gslMemObject destMem, size_t* destOffset, size_t* size, CALmemcopyflags flags, bool enableCopyRect, uint32 bytesPerElement);
 
     void             setSamplerParameter(uint32 sampler, gslTexParameterPname param, CALvoid* vals);