Repository navigation
This release uses Fortran standard parallelism for all compute loops, and OpenMP Target data directives to manually manage GPU-CPU memory.
This release uses Fortran standard parallelism for all compute loops, and OpenMP Target data directives to manually manage GPU-CPU memory.