Dandandan commented on pull request #8665: URL: https://github.com/apache/arrow/pull/8665#issuecomment-770384503
@vertexclique I was actually wondering whether AVX-512 is really faster than compiling the kernel with "simd" and the right rustflags , i.e. `RUSTFLAGS='-C target-cpu=native`? I don't have a `AVX-512`-enabled cpu to try this out, but I would have the same intuition as @jhorstmann that packed simd should generate similar `AVX-512` instructions? Would be nice maybe to revisit this once to see whether we need to maintain 2 different implementations, especially when it's going to be in `std` / stable. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
