Hi, how do I make a async execution with 3D arrays?
I need help to execute a 3D matrix array with the function cudaMemcpy3Dasync in cuda Fortran. A single example… PLEASE!
Do you really need to use cudaMemcpy3dAsync(). It is pretty odd. It takes a type(cudaMemcpy3DParms) as an argument. Much easier to just pass a 3D Fortran array via cudaMemcpyAsync()…