alamb commented on PR #9022: URL: https://github.com/apache/arrow-rs/pull/9022#issuecomment-3682240368
Ok, given that totally specializing the slicing didn't help, I am now 100% convinced that the difference in performance reported for _sliced24 is due to alignment or some other crazy thing. What I plan to do now is: 1. Try and update the benchmarks to 1) allocate input arrays before any other allocations happen and 2) include more / multiple different alignments (will allocate a few different ways) -- 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]
