SIMD/Parallelization via OpenMP #36
Description
Referring to https://github.com/tsoding/olive.c/blob/d770d9c995010dbb1dbbce262d01502076c63814/olive.c#L1018C10-L1018C30
@rexim would you accept a PR introducing OpenMP pragmas to automatically offload to gpu (or scale on CPU) based on compiler flags and capabilities? Or is it something you want to explore on your own?
Changes would only be minimally intrusive as OpenMP is designed around the idea of being incrementally added on already existing code.