ROCm/gpufort

Reenable CUDA Fortran

Closed this issue · 2 comments

  • GPUFORT has seen dramatic changes in the past to enable OpenACC
  • We need to reassess the CUDA Fortran porting capabilities of the code
  • We further need to adopt the new HIPFORT Fortran interfaces

Some CUF loop kernel -> OMP loop kernel conversion works again but this still needs a few days

CUDA Fortran translation to HIP works again. Two examples have been added to demonstrate the feature.