zanmato1984 commented on issue #43687: URL: https://github.com/apache/arrow/issues/43687#issuecomment-2290919509
> We can't promise that the compiler won't generate AVX512 instructions if we allow it to. So, no, it looks like a bug. > > Thanks for the clarification. If this is a bug, then my other question quoted below stands no more. > I appreciate the explanation. I now have a further question: so the minmax implementations for both AVX2 and AVX512 are the same (both being AVX2-level optimized). Dose this mean that there are two identical kernels with identical SIMD level in the scalar agg function minmax? (Suppose both compiler switches ARROW_HAVE_RUNTIME_AVX2/512 are on - which should normally be the case.) -- 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. To unsubscribe, e-mail: github-unsubscr...@arrow.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
