SIMD support
bkmgit opened this issue · 3 comments
bkmgit commented
Would it be of interest to add SIMD support? For example Numpy uses https://github.com/mathlibrary/usimd
bkmgit commented
Thanks. This adds SSE instructions. However, AVX, neon and possibly SVE instructions may also be helpful:
- https://github.com/gcc-mirror/gcc/blob/master/gcc/config/aarch64
- https://github.com/gcc-mirror/gcc/tree/master/gcc/config/i386
Further conditional compilation options could be used
barracuda156 commented
It would be nice to have AltiVec as well, but presently the core doesn’t even compile: #215