Excludes are not replaced, for instance, CHECK_CUDA_ERROR and CUDA_SAFE_CALL.
Add check for excludes in MacroExpands and CallExpr routines.
[ROCm/clr commit: 5d04a17a44]
To run clang-hipify with hipblas support please specify corresponding hipblas include directory, for example:
./hipify-clang --print-stats matrixMult.cpp.cuda -- -I/srv/git/HIP/HIP-Examples/hipblas/include
Additionally:
+ typedef translation support is added (cudaEvent_t and cudaStream_t are converted now).
+ anonymous typedef enum translation support.
+ function name macro expansion support.
+ clang options propagation is restored.
P.S. In order to avoid the following error message:
“Could not auto-detect compilation database for file "… .cu"
No compilation database found in /srv/git/HIP/build/clang-hipify or any parent directory
json-compilation-database: Error while opening JSON database: No such file or directory
Running without flags.”
please append “--“ to the end of clang-hipify’s command line.
[ROCm/clr commit: aa75bff6ee]
On HIP path property obtaining done through hsa_iterate_agents and counting the devices of HSA_DEVICE_TYPE_GPU type.
P.S.
On multi-boards systems it might be problems with detection what board a GPU plugged into (not tested).
[ROCm/clr commit: 7bb0f17656]
Device property MaxSharedMemoryPerMultiprocessor set equal to totalGlobalMem (HIP path).
Reason: MaxSharedMemoryPerMultiprocessor should be as the same as group memory size. Group memory will not be paged out, so, the physical memory size = total shared memory size = group region size. NVCC path remains untouched: CUDA's device property MaxSharedMemoryPerMultiprocessor is reported.
hipify is updated as well.
[ROCm/clr commit: 460b501cbb]
Reason: maxThreadsPerMultiProcessor should be as the same as group memory size. Group memory will not be paged out, so, the physical memory size = total shared memory size = group region size.
NVCC path remains untouched: CUDA's device property maxThreadsPerMultiProcessor is reported.
[ROCm/clr commit: 1025341300]
For HCC path concurrentKernels is set to true since all ROCR hardware supports this feature.
For NVCC path concurrentKernels is obtained from CUDA's device property cudaDeviceProp::concurrentKernels.
[ROCm/clr commit: 4d4ca3ef3f]
Evgeny Mankov