EPR #397491 - Replace "switch" with "if" so that new ASIC id doesn't need to be added.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.cpp#103 edit
EPR #403341 - [Regression][OCL] P2P SDI to GPU (RGBA) and P2P SDI to GPU (RGB) show corruption on sdi output
- Enable GL Acquire/Release calls for more resource types. GL can enable color compression, but compute doesn't support it. Thus decompression can be required.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuresource.cpp#194 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.h#41 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDeviceGL.cpp#18 edit
EPR #406328 - changed the device open algorithm so that the we only open the first OpenCL device. This is the OPENCL runtime changes, but this will be removed once we implemented multiple device support for SVM.
the code review and precheckin test:
http://ocltc.amd.com/reviews/r/5942/http://ocltc.amd.com:8111/viewModification.html?modId=40902&personal=true&buildTypeId=&tab=vcsModificationBuilds&show_all_builds=true
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#466 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpusettings.cpp#287 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpusettings.hpp#87 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/include/cal/cal.h#29 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.cpp#100 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.h#40 edit
EPR #398128 - Windows 2015, WDDM2.0, New Residency Model
- Modify MarkUsedInCmdBuf in IOL to make sure that MakeResident is called for OpenCL (Part 2)
ReviewBoardURL = http://ocltc.amd.com/reviews/r/5684/diff/
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.cpp#99 edit
ECR #304775 - Add batching to the device enqueue for possible asynchronous execution
- Increase the max device queue size to 512KB. That will allow to pass conformance tests that enqueue more jobs than the queue size.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#459 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpusched.hpp#13 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuschedcl.cpp#28 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuvirtual.cpp#333 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLContext.cpp#65 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLContext.h#39 edit
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
ECR #304775 - Device enqueuing
- Provide scratch buffer offset for generic address space
- Use single scratch buffer for all available queues. Each queue will have a unique subbuffer in the global buffer
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#454 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.hpp#129 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpusched.hpp#11 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuschedcl.cpp#24 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuvirtual.cpp#329 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuvirtual.hpp#120 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLContext.cpp#63 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLContext.h#37 edit
ECR #304775 - Add new flag GPU_IFH_MODE to allow IFH mode testing without using the inifile.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.cpp#97 edit
... //depot/stg/opencl/drivers/opencl/runtime/utils/flags.hpp#211 edit
EPR #403896 - [OpenGL] - Use GFXIP instead of asic family name
This is the first step to using a GFXIP version instead of the asic family names wherever possible in our driver. It is only being used for a few checks right now, and there will be more changes after this to use it in more places.
- Added GFXIP so that we can reference IP version instead of checking asic family names. This is being done so it's more clear about which version of gfx ip a given asic is, since marketing often uses a different family name than engineering (ex. Hawaii is a VI part from marketing standpoint). This also reflects the HW team's move towards using the gfxip version instead of a family name. Another benefit is it will clean up our asic family checks quite a bit, and they should be a bit more forward looking as well.
- Added some asserts for when gfxip 9 comes along we know the places we need to validate. It hasn't been determined yet if gfxip 9 will use the same hwl, so if it doesn't, these asserts will be removed then.
- Removed all references of Kauai, Tiran, and Maui from our driver since these asics were all cancelled at some point previously.
- Changed Distributed Tessellation enablement criteria to be gfxip 8+ && numSE > 1 instead of checking asic names.
ReviewBoardURL = http://ocltc.amd.com/reviews/r/5450/
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.cpp#96 edit
ECR #304775 - Device enqueuing
- Add L2 cache flush after the scheduler execution. Although CP has to work with L2 cache, it seems some functionality relies on direct memory access and without explicit L2 flush CP can pick old values in the template.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuvirtual.cpp#324 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLContext.cpp#61 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLContext.h#35 edit
foreman