I’m trying to integrate CPU code into GPU code. In particular, I want to do something like system calls embedded in GPU code.
I have already read in a paper that I can only run the Linux system calls on the CPU. Unfortunately, there is no mention of how this is implemented in CUDA code. Can anyone help me?