SWDEV-192353 - 12% Performance drop observed while running ROC_OCL_Perf_CompubenchCL_GPU_W64 only on Vega10 with Win7
1. Reduce CmdAllocator sub-allocation size to 4KB.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palvirtual.cpp#137 edit
[ROCm/clr commit: 3669ba964e]
SWDEV-2 - Change OpenCL version number from 2944 to 2945.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2692 edit
[ROCm/clr commit: a8fb74a8b6]
SWDEV-189500 - [HIP] Have to force async=false for host to device case as well
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#61 edit
[ROCm/clr commit: a888b06ede]
[Reason] To be compatible with CUDA [#1133]
Update HIP code, hipify-clang, tests and docs
[TODO] Add support of the corresponding functions on nvcc fallback path
[ROCm/clr commit: f0832fd968]
SWDEV-2 - Change OpenCL version number from 2943 to 2944.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2691 edit
[ROCm/clr commit: ba82794af1]
SWDEV-2 - Change OpenCL version number from 2942 to 2943.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2690 edit
[ROCm/clr commit: f4b8358eb6]
SWDEV-161424 - Fix broken option handling in Comgr path
Introduces another potential bug if any options in Options::llvmOptions contain spaces, but this existed before the switch to Comgr.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/devprogram.cpp#50 edit
[ROCm/clr commit: 93c85b966d]
SWDEV-2 - Change OpenCL version number from 2941 to 2942.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2689 edit
[ROCm/clr commit: 737e74e666]
+ Add option -print-stats-csv to dump statistics to CSV file
+ If -o-dir is specified, CSV file will be dumped there
+ Generate 1 summary file sum_stat.csv in case of multiple sources
[ROCm/clr commit: 60a9143d6d]
SWDEV-2 - Change OpenCL version number from 2940 to 2941.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2688 edit
[ROCm/clr commit: 01b3c9a93c]
SWDEV-2 - Change OpenCL version number from 2939 to 2940.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2687 edit
[ROCm/clr commit: cf619cedf7]
SWDEV-2 - Change OpenCL version number from 2938 to 2939.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2686 edit
[ROCm/clr commit: b1a5dc95c8]
* Remove flags parameter from hipOccupancyMaxPotentialBlockSize
This commit makes the hipOccupancyMaxPotentialBlockSize method consistent with hcc path and the CUDA API.
[ROCm/clr commit: a401997b8e]
module_api_global relies on a HCC only feature which allows host code
to write to device variables. This feature does not exist in CUDA
or hip-clang, which causes the sample not working in CUDA or hip-clang.
This patch fixes the sample by using standard features of CUDA and
hip-clang. The fixed sample works in HCC, CUDA and hip-clang.
[ROCm/clr commit: 3541d18528]
module_api_global relies on a HCC only feature which allows host code
to write to device variables. This feature does not exist in CUDA
or hip-clang, which causes the sample not working in CUDA or hip-clang.
This patch fixes the sample by using standard features of CUDA and
hip-clang. The fixed sample works in HCC, CUDA and hip-clang.
[ROCm/clr commit: 98648828c0]
SWDEV-2 - Change OpenCL version number from 2937 to 2938.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2685 edit
[ROCm/clr commit: 421f8e1b72]
SWDEV-2 - Change OpenCL version number from 2936 to 2937.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2684 edit
[ROCm/clr commit: e4012b8878]
SWDEV-2 - Change OpenCL version number from 2935 to 2936.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2683 edit
[ROCm/clr commit: 4df681f713]
SWDEV-193973 - [OpenCL][NV21] Add support for Navi21
1. Enable OCLOfflineCompilation for gfx1010/1011/1012.
2. Skip gfx1030 as there is no support for this target in the device libraries for now.
ReviewBoardURL = http://ocltc.amd.com/reviews/r/17611/diff/
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/devkernel.cpp#24 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palkernel.cpp#80 edit
... //depot/stg/opencl/drivers/opencl/tests/ocltst/module/runtime/OCLOfflineCompilation.cpp#21 edit
[ROCm/clr commit: 35b8eb2f0e]
SWDEV-193938 - [HIP] RCCL test fails
Set default stream to null-stream
In hipStreamWaitEvent if event_ is null, get the last queued command instead.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_event.cpp#12 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_event.hpp#4 edit
[ROCm/clr commit: ccd7799ada]
SWDEV-145570 - Fix device name mismatch.
Not only gfx906 can have device name with +xnack etc.
Other devices e.g. gfx900 could have that too.
Make the previous fix more generic.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_platform.cpp#33 edit
[ROCm/clr commit: 6768664780]
SWDEV-2 - Change OpenCL version number from 2934 to 2935.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2682 edit
[ROCm/clr commit: c99c79f578]
foreman
Evgeny Mankov
kpyzhov
Yaxun (Sam) Liu
mhbliao
Jatin Chaudhary
Maneesh Gupta
Aryan Salmanpour
darkbuck
searlmc1
Jannik
eshcherb