SWDEV-159036 - HBCC is off by default in OCL on WX9100
- Add HBCC size to the global memory. It seems KMD fixed all regressions.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/flags.hpp#292 edit
[ROCm/clr commit: abefff3eb2]
SWDEV-2 - Change OpenCL version number from 2710 to 2711.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2458 edit
[ROCm/clr commit: 60bea15438]
SWDEV-145570 - [HIP] - Fix some issues in hip runtime
- Set stream for event
- Free mem needs to be reported in bytes but runtime backends reports in Kb
ReviewBoardURL = http://ocltc.amd.com/reviews/r/15586/diff/
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#40 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_module.cpp#15 edit
[ROCm/clr commit: b4ccaeffe1]
SWDEV-145570 - [HIP] Increase the default size for SVM to 16GB for HIP
ReviewBoardURL = http://ocltc.amd.com/reviews/r/15578/diff/
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldevice.cpp#100 edit
[ROCm/clr commit: aa99feb153]
SWDEV-2 - Change OpenCL version number from 2709 to 2710.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2457 edit
[ROCm/clr commit: f889d2ebba]
SWDEV-2 - Change OpenCL version number from 2708 to 2709.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2456 edit
[ROCm/clr commit: 6d82075654]
SWDEV-145570 - [HIP] Refactored some g_* stuff
Refactored g_functions into a platform state.
Added a _vars for registered variables.
Added an execution stack similar to Hcc-clang.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_platform.cpp#14 edit
[ROCm/clr commit: 2ff260c727]
SWDEV-2 - Change OpenCL version number from 2707 to 2708.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2455 edit
[ROCm/clr commit: 7d877e1589]
SWDEV-145570 - [HIP] - Release eStop if it not a nullptr.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_module.cpp#13 edit
[ROCm/clr commit: e638c1377f]
SWDEV-155434 - Add SQTT instrumentation tokens for OpenCL dispatches for RGP support
- Switch to the workgroup size report for the dispatch info.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palvirtual.cpp#118 edit
[ROCm/clr commit: 19968ab78a]
SWDEV-2 - Change OpenCL version number from 2706 to 2707.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2454 edit
[ROCm/clr commit: 37677e98a5]
SWDEV-145570 - [HIP] refactor hipMemcpy* functions to correctly handle copies using prepinned memory
The current implementation of hipMemcpy functions picks the copy type based on a flag that the user passes. However, one can use the hipMemcpyHostToDevice/hipMemcpyDeviceToHost flag in a combination with prepinned memory. By using the WriteMemoryCommand/ReadMemoyCommand in this case, we will pin the same host memory twice. This is fine on PAL/Linux, since pinning the same VA range is a noop, but this will start failing once we switch to using device memory with HIP/VDI/HSA.
The solution is to ignore the hipMemcpyKind flag and let the runtime decide what kind of copy is best to do. Except for the case when hipMemcpyHostToHost is passed, since both host pointers may be prepinned.
ReviewBoardURL = http://ocltc.amd.com/reviews/r/15482/diff/
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#37 edit
[ROCm/clr commit: b02040e934]
SWDEV-2 - Change OpenCL version number from 2705 to 2706.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2453 edit
[ROCm/clr commit: ce73c861d6]
Implement the vector value types in terms of SCALAR_TYPE##SIZE_vector_value_type for float, int, uint and sizes 2, 3, 4, 8, and 16.
[ROCm/clr commit: ff1e49d191]
When shareWithAll memory (e.g., host memory) is allocated, set appId
in hc::AmPointerInfo to -1 to indicate that this memory is not mapped
to any device. Peer checking in ihipStream_t::canSeeMemory is not
necessary if memory is shared with all devices. Thus, it is skipped.
Note that earlier host memory is always mapped to device 0 and HIP
always performs peer checking for all kinds of hipMemcpy. Since the
peer checking process requires context locking, hipMemcpy from/to host
memory always grabs device 0's context lock. Therefore, if there is
another thread holding the context lock of device 0 (e.g.,
hipDeviceSynchronize on device 0), hipMemcpy will have to wait for the
lock until it can actually perform memcpy. This can significantly
deteriorate execution performance.
Signed-off-by: Sarunya Pumma <sarunya.pumma@amd.com>
[ROCm/clr commit: 84aadb9274]
Yaxun Sam Liu
foreman
sunway513
Wen-Heng (Jack) Chung
Maneesh Gupta
Rahul Garg
Aaron Enye Shi
Evgeny Mankov
Evgeny Mankov
lthakur
Sarunya Pumma