SWDEV-102733 - [OCL-LC-ROCm] Cmake build Write CMakeLists.txt to enable building with and without the DK environment
- Set the default triple to amdgcn-amd-amdhsa-opencl
Affected files ...
... //depot/stg/opencl/drivers/opencl/CMakeLists.txt#9 edit
[ROCm/clr commit: 3623f0d69e]
SWDEV-102733 - [OCL-LC-ROCm] Cmake build Write CMakeLists.txt to enable building with and without the DK environment
- Set the default triple to amdgcn-amd-amdhsa-opencl
Affected files ...
... //depot/stg/opencl/drivers/opencl/CMakeLists.txt#9 edit
SWDEV-2 - Change OpenCL version number from 2426 to 2427.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2174 edit
[ROCm/clr commit: 5e85545c43]
SWDEV-2 - Change OpenCL version number from 2426 to 2427.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2174 edit
Provide per-allocation control over coherent/non-coherent mem.
These overrid the default HIP_COHERENT_HOST_ALLOC setting.
[ROCm/clr commit: dda70ae514]
SWDEV-2 - Change OpenCL version number from 2425 to 2426.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2173 edit
[ROCm/clr commit: 2e71ba3a73]
SWDEV-2 - Change OpenCL version number from 2425 to 2426.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2173 edit
https://github.com/GPUOpen-ProfessionalCompute-Tools/HIP/issues/80
Example from CUDA 8.0.44 sample (CUDASamples\0_Simple\matrixMulDrv\matrixMulDrv.cpp):
CUjit_option *jitOptions = new CUjit_option[jitNumOptions];
where CUjit_option is enum, should be:
hipJitOption *jitOptions = new hipJitOption[jitNumOptions];
[ROCm/clr commit: 21d74f09b9]
https://github.com/GPUOpen-ProfessionalCompute-Tools/HIP/issues/80
Example from CUDA 8.0.44 sample (CUDASamples\0_Simple\matrixMulDrv\matrixMulDrv.cpp):
CUjit_option *jitOptions = new CUjit_option[jitNumOptions];
where CUjit_option is enum, should be:
hipJitOption *jitOptions = new hipJitOption[jitNumOptions];
https://github.com/GPUOpen-ProfessionalCompute-Tools/HIP/issues/80
Example from CUDA 8.0.44 sample (CUDASamples\0_Simple\matrixMulDrv\matrixMulDrv.cpp):
CUjit_option *jitOptions = new CUjit_option[jitNumOptions];
where CUjit_option is enum, should be:
hipJitOption *jitOptions = new hipJitOption[jitNumOptions];
SWDEV-2 - Change OpenCL version number from 2424 to 2425.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2172 edit
[ROCm/clr commit: 421c10fd87]
SWDEV-2 - Change OpenCL version number from 2424 to 2425.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2172 edit
https://github.com/GPUOpen-ProfessionalCompute-Tools/HIP/issues/81
1. Do not use JIT options on HCC path, call hipModuleLoadData instead.
2. NVCC path is unchanged, to call cuModuleLoadDataEx with all options.
3. Get rid of manual hipification, based on #ifdef #else for NVCC/HIP.
4. Update documentation accordingly.
[ROCm/clr commit: ae9f14ef9c]
https://github.com/GPUOpen-ProfessionalCompute-Tools/HIP/issues/81
1. Do not use JIT options on HCC path, call hipModuleLoadData instead.
2. NVCC path is unchanged, to call cuModuleLoadDataEx with all options.
3. Get rid of manual hipification, based on #ifdef #else for NVCC/HIP.
4. Update documentation accordingly.
https://github.com/GPUOpen-ProfessionalCompute-Tools/HIP/issues/81
1. Do not use JIT options on HCC path, call hipModuleLoadData instead.
2. NVCC path is unchanged, to call cuModuleLoadDataEx with all options.
3. Get rid of manual hipification, based on #ifdef #else for NVCC/HIP.
4. Update documentation accordingly.
SWDEV-2 - Change OpenCL version number from 2423 to 2424.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2171 edit
[ROCm/clr commit: 738528f6f3]
SWDEV-2 - Change OpenCL version number from 2423 to 2424.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2171 edit