hemi
hemi copied to clipboard
(feat req) OpenMP acceleration in parallel_for
It would be great if, when CUDA is disabled, hemi::paralell_for ran on all available CPUs. I tried adding a simple #pragma omp for myself, but for some reason OpenMP refused to cooperate...
I agree. I would like to look at this in the future if somebody doesn't get to it before me.
does this meet the requirement?
https://github.com/louietsai/hemi/commit/216f912d87f72f13ca8d7463ce21b275a634a452