mqy commented on pull request #9211: URL: https://github.com/apache/arrow/pull/9211#issuecomment-761745923
I saw some performance regression: - `equal_nulls_512` regressed about +22% - `equal_string_nulls_512` regressed about +17%. It looks like this is caused by `lhs_start + lhs.offset()` being executed for every bit and the `rhs` case too. For both clarity and performance reasons, perhaps it make sense to define variables like `let lhs_bits_start = lhs_start + lhs.offset();` above `(0..len).all(...` and let `get_bit()` call with `lhs_bits_start+i`. ---------------------------------------------------------------- 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: [email protected]
