Hattonuri commented on PR #39124: URL: https://github.com/apache/arrow/pull/39124#issuecomment-1846946694
> 🤔Though benchmark using SystemAllocator is a bit weird, should we use something like jemalloc? > > Also ZSTD_decompressSequences_bmi2 shouldn't introduce any differences if the page we read doesn't changed, except we read more pages? i compile all my project with tcmalloc > Are you just comparing percentages? This is silly, isn't it? I think that it's not in that case because page faults increased and zstd increased - so we can still see that that it is changed I will try to binary search breaking commit. But I think that it needs another issue because we see that the problem is not with that PR -- 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]
