Baymine commented on PR #66164: URL: https://github.com/apache/doris/pull/66164#issuecomment-5102695067
Closing this: a more rigorous benchmark (interleaved runs, pinned core) shows the memchr fast path regresses ~10-14% for sub-6-byte fields (break-even at ~6 bytes, only a clear win at >=20 bytes), which confirms the existing 'for+if beats memchr for short fields' note. Will revisit with a width-gated hybrid if pursued. -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
