ozgrakkurt commented on PR #4441: URL: https://github.com/apache/arrow-rs/pull/4441#issuecomment-1609483886
I didn’t check assembly code myself. My implementation is adapted from the c impl of impala. There is a big difference is benchmarks so I don’t think it optimizes to same assembly especially for avx2 On 27 Jun 2023 Tue at 19:07 Jörn Horstmann ***@***.***> wrote: > I would actually expect the existing code to auto-vectorize and generate > nearly the same assembly as with using intrinsics. > > Maybe check needs a tiny refactoring and not do that early exit though: > https://rust.godbolt.org/z/GG6YqTena > > Are there more optimizations in the crate that I'm missing? > > — > Reply to this email directly, view it on GitHub > <https://github.com/apache/arrow-rs/pull/4441#issuecomment-1609469322>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AV37FAYLPGLXN35YS7P6ERTXNLLIBANCNFSM6AAAAAAZPEUWEA> > . > You are receiving this because you authored the thread.Message ID: > ***@***.***> > -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
