pitrou commented on pull request #7314: URL: https://github.com/apache/arrow/pull/7314#issuecomment-641136788
> Just find a document(https://dl.acm.org/doi/pdf/10.1145/3178433.3178435 PAGE 7, Table 6: Comparison of various SIMD wrappers.) Some SIMD helper(simdpp/xsimd) has performance issue at least on some workload. Interesting comparison, thank you. But that is a single workload and we don't know what the issue was precisely. By the way, this is what the paper says about the approach you're advocating: > The intrinsics or assembly approaches are non-portable, low-level solutions which target specific architectures. They offer maximum control to take advantage of instruction set specificities, and to fine tune register usage. However, it is quite difficult to develop and maintain a low-level code in the long run. ---------------------------------------------------------------- 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: us...@infra.apache.org
