Łukasz Langa <luk...@langa.pl> added the comment: Well, this is kind of disappointing on my end. I attach the full result of a three-way comparison between `main` at the time of Serhiy's last merge (3f8b23f8ddab75d9b77a3997d54e663187e12cc8) with GH-27986 (20e3149c175a24466c7d1c352f8ff2c11effc489) and GH-28176 (cffa90a8b0057d7e7456571045f2fb7b9ceb426f). The gist is this:
Geometric mean ============== 20e3149c175a24466c7d1c352f8ff2c11effc489-2: 1.01x slower cffa90a8b0057d7e7456571045f2fb7b9ceb426f-2: 1.01x slower At least on my Macbook Pro, all PGO builds, looks like the status quo is on average faster than any of the candidate PRs. ---------- Added file: https://bugs.python.org/file50306/three-way-comparison.log _______________________________________ Python tracker <rep...@bugs.python.org> <https://bugs.python.org/issue45026> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
