⚠️ GitHub issue #39863 has been automatically assigned in GitHub to PR creator.

@ursabot please benchmark lang=C++

Benchmark runs are scheduled for commit 6329711. Watch https://buildkite.com/apache-arrow and https://conbench.ursa.dev for updates. A comment will be posted here when the runs are complete.

Thanks for your patience. Conbench analyzed the 1 benchmarking run that has been run so far on PR commit 6329711.

There weren't enough matching historic benchmark results to make a call on whether there were regressions.

The full Conbench report has more details.

@austin3dickey The wording of the conbench summary at https://github.com/apache/arrow/runs/21102698059 is a bit unexpected, might this be due to the changes in this PR?

@pitrou Thanks for bringing that to my attention - that's unrelated to this PR. I recently upgraded conda on the machine that ran your benchmarks, and it seems to have put a break in Conbench history since the -isystem path changed. I will fix this so new benchmarks can be compared to old benchmarks.

In the meantime, you can check this link for comparisons: https://conbench.ursa.dev/compare/runs/fa53528839744c73945c88a71e31353f...e2a595b8ca9d41d4916b309b6c4ec42b/

(Sorry that page takes so long to load; I will fix that problem soon.)

I fixed the broken history in Conbench and manually reviewed the comparison I linked above. It looks good to me. I don't think this change has much of a performance effect on any of the benchmarks.

@ursabot please benchmark lang=C++

Commit 6329711 already has scheduled benchmark runs.

@Cerdore Could you rebase on main?

⚠️ GitHub issue #39863 has been automatically assigned in GitHub to PR creator.

@Cerdore Could you rebase on main?

Already rebased.

@ursabot please benchmark lang=C++

Benchmark runs are scheduled for commit ddeef70948b2a072db1254ed7846b199ebe747f7. Watch https://buildkite.com/apache-arrow and https://conbench.ursa.dev for updates. A comment will be posted here when the runs are complete.

Thanks for your patience. Conbench analyzed the 1 benchmarking run that has been run so far on PR commit ddeef70948b2a072db1254ed7846b199ebe747f7.

There were 2 benchmark results indicating a performance regression:

• Pull Request Run on ursa-thinkcentre-m75q at 2024-02-04 03:33:35Z
  • BitmapReaderSum (C++) with params=32768, source=cpp-micro, suite=arrow-bit-block-counter-benchmark
  • BitmapReaderSum (C++) with params=4096, source=cpp-micro, suite=arrow-bit-block-counter-benchmark

The full Conbench report has more details.

@github-actions crossbow submit -g cpp

Revision: 946363c

Submitted crossbow builds: ursacomputing/crossbow @ actions-00241391af

Task	Status
test-alpine-linux-cpp
test-build-cpp-fuzz
test-conda-cpp
test-conda-cpp-valgrind
test-cuda-cpp
test-debian-11-cpp-amd64
test-debian-11-cpp-i386
test-fedora-38-cpp
test-ubuntu-20.04-cpp
test-ubuntu-20.04-cpp-bundled
test-ubuntu-20.04-cpp-minimal-with-formats
test-ubuntu-20.04-cpp-thread-sanitizer
test-ubuntu-22.04-cpp
test-ubuntu-22.04-cpp-20
test-ubuntu-22.04-cpp-no-threading

After merging your PR, Conbench analyzed the 6 benchmarking runs that have been run so far on merge-commit b6313a7.

There were no benchmark performance regressions. 🎉

The full Conbench report has more details. It also includes information about 7 possible false positives for unstable benchmarks that are known to sometimes produce them.