jonkeane commented on PR #12928: URL: https://github.com/apache/arrow/pull/12928#issuecomment-1104251092
There definitely are regressions in throughput on those C++ benchmarks. Interestingly, it looks like there was a pretty big regression (larger than here, even!) in some of these already around https://github.com/apache/arrow/pull/12634 which is odd since that's a java + documentation only ticket. Though it does look like our compiler version on the benchmarking machine went from 9.4.0 to 10.3.0 coincidentally at that commit, which might be the real source of that regression. -- 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]
