EPR #408459 - changed the implementation of svmAlloc, so that the first device can create amd::Memory object, and the rest of devices only added gpu memory to it. This is part of changes for mgpu support for svmalloc
code review:
http://ocltc.amd.com/reviews/r/6245/
precheckin testing results:
http://ocltc.amd.com:8111/viewModification.html?modId=43136&personal=true&buildTypeId=&tab=vcsModificationTests
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/cpu/cpudevice.hpp#88 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/device.hpp#233 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#479 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.hpp#133 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/hsa/hsadevice.cpp#87 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/hsa/hsadevice.hpp#44 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/hsa_foundation/hsadevice.cpp#19 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/hsa_foundation/hsadevice.hpp#6 edit
... //depot/stg/opencl/drivers/opencl/runtime/platform/context.cpp#34 edit
[ROCm/clr commit: efbedb25be]
EPR #408185 - Use pinned memory if directaccess is true and remoteAlloc is used.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#478 edit
[ROCm/clr commit: c24b46e708]
EPR #408506 - Extended the reported global memory size(CL_DEVICE_GLOBAL_FREE_MEMORY_AMD) to include a portion of remote memory for APU
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#476 edit
[ROCm/clr commit: dd6074f047]
EPR #407469 - disabled the SVM fine grain buffer support for CZ on mainline
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#472 edit
[ROCm/clr commit: 206bd8113d]
EPR #405824 - On apus, if we run out of local memory to allocate cl_mem objects, ocl runtime will use remote (system) memory. Update maxMemAllocSize_ to include that.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#469 edit
[ROCm/clr commit: 1dfb720281]
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
[ROCm/clr commit: c248f04970]
EPR #406110 - OCL20:Basic subtest fails when running on GPU
- Reduce max prog variable size to 90% of max single allocation
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#465 edit
[ROCm/clr commit: 617422f40f]
EPR #405824 - Back out changelist 1079967. It causes regressions in other apus.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#464 edit
[ROCm/clr commit: 93e05cdac8]
EPR #405824 - On apus, if we run out of local memory to allocate cl_mem objects, ocl runtime will use remote (system) memory. Update maxMemAllocSize_ to include that.
Reviewed by: German
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#463 edit
[ROCm/clr commit: a16bef5482]
EPR #397491 - disable OpenCL 2.0 for mainline when there are multiple devices in the system, because svm test will fail even test on the first device
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#462 edit
[ROCm/clr commit: e7b10515af]
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
[ROCm/clr commit: 2738b30287]
EPR #404714 - [CQE OCL][2.0][DTB]Opencl1.2 WF Conf. Math test failedon Pitcairn and Oland due to CL#1065597
- Add a new MapCacheLock monitor to separate the map cache from the global lock
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#456 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.hpp#131 edit
[ROCm/clr commit: 18b88ee095]
EPR #404714 - [CQE OCL][2.0][DTB]Opencl1.2 WF Conf. Math test failedon Pitcairn and Oland due to CL#1065597
- Add VGPU lock to flush() method, because gsl flush for the same context could be called from multiple threads
- Use new scratchAlloc_ monitor for scratch reallocation
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#455 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.hpp#130 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuvirtual.cpp#331 edit
[ROCm/clr commit: a4bede39eb]
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
[ROCm/clr commit: 8fa184db38]
ECR #304775 - Device enqueuing
- Use atomic fetch for enqueue flags
- Switch to a multithreaded scheduler
- Add a workaround for Linux host_multi_queue failures. Linux has only 2 queues, but the test allocates multiple host queues and the same HW ring can be used
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpublit.cpp#106 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#449 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.hpp#127 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuschedcl.cpp#22 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuvirtual.cpp#325 edit
[ROCm/clr commit: d2b905f18e]
ECR #304775 - OpenCL version should be 2.0 for only CI+
ReviewBoardURL = http://ocltc.amd.com/reviews/r/5311/
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#448 edit
[ROCm/clr commit: db6baeefb6]
EPR #304775 - temporarily disable the SVM fine_grained_buffer support for OpenCL 2.0 on discrete GPUs, because the feature is supposed to release in 14.50. After the 14.40 is branched, we will enable it again on stg.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#445 edit
[ROCm/clr commit: 18d6efdf2c]
foreman