'-Ofast' and '-march=native' cause 2x-speedup in machines with SSE (but no AVX) instructions. Should help other platforms, too.
parent
603f97ba11
commit
49b2f9bd66
Loading…
Reference in new issue