HuaHuaY commented on PR #50281: URL: https://github.com/apache/arrow/pull/50281#issuecomment-4874846654
Currently, `VisitTwoSetBitRuns` materializes the bitmap first and then iterates over it using `VisitSetBitRuns`. The reason I called `BitmapAnd` directly on the full bitmap length is that the bitmap is small, so segmentation seemed unnecessary; furthermore, the performance impact of segmentation is unknown. The performance gap observed in the new benchmarks compared to `BitmapAdd` is due to differences in the extent of automatic compiler inlining. @pitrou Please take a look again. -- 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]
