Except FunctionID (or DomainID in CUDA) support, because cudaDeviceProp::pciDomainID is not reported by CUDA. [ROCm/clr commit: 658e9f0484]
658e9f0484
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]
4d4ca3ef3f
[ROCm/clr commit: afe45964ae]
afe45964ae
[ROCm/clr commit: 304171c1a2]
304171c1a2