-
-
Notifications
You must be signed in to change notification settings - Fork 31.6k
gh-128588: gh-128550: remove eager tasks optimization that missed and introduced incorrect cancellations #129063
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
gh-128588: gh-128550: remove eager tasks optimization that missed and introduced incorrect cancellations #129063
Conversation
…t missed and introduced incorrect cancellations
Misc/NEWS.d/next/Library/2025-01-20-13-12-39.gh-issue-128550.AJ5TOL.rst
Outdated
Show resolved
Hide resolved
…J5TOL.rst Co-authored-by: Kumar Aditya <kumaraditya@python.org>
Misc/NEWS.d/next/Library/2025-01-20-13-12-39.gh-issue-128550.AJ5TOL.rst
Outdated
Show resolved
Hide resolved
…J5TOL.rst Co-authored-by: Thomas Grainger <tagrain@gmail.com>
Thanks @graingert for the PR, and @kumaraditya303 for merging it 🌮🎉.. I'm working now to backport this PR to: 3.12, 3.13. |
…t missed and introduced incorrect cancellations (pythonGH-129063) (cherry picked from commit ed6934e) Co-authored-by: Thomas Grainger <tagrain@gmail.com> Co-authored-by: Kumar Aditya <kumaraditya@python.org>
Sorry, @graingert and @kumaraditya303, I could not cleanly backport this to
|
GH-129089 is a backport of this pull request to the 3.13 branch. |
…sed and introduced incorrect cancellations (GH-129063) (#129089) gh-128588: gh-128550: remove eager tasks optimization that missed and introduced incorrect cancellations (GH-129063) (cherry picked from commit ed6934e) Co-authored-by: Thomas Grainger <tagrain@gmail.com> Co-authored-by: Kumar Aditya <kumaraditya@python.org>
…t missed and introduced incorrect cancellations (python#129063) Co-authored-by: Kumar Aditya <kumaraditya@python.org>
What's the impact of removing this optimization on the async_tree pyperformance benchmarks? |
I haven't benchmarked yet, I'll do it but I expect an insignificant impact. |
…t missed and introduced incorrect cancellations (python#129063) Co-authored-by: Kumar Aditya <kumaraditya@python.org>
any update on benchmarking results @kumaraditya303 ? benchmarking of the original PR showed up to 4x speedup on some of the async benchmarks. anything changed that you expect removing the optimization would have insignificant impact? |
I had asked Michael to benchmark this last week but I haven't heard back from them, I'll ping them again but meanwhile I compared the rough number by comparing benchmarks with and without this patch and I don't see any large slowdown on the benchmarks: With this PR: |
Hmm, I think I see what's going on here.. Looking more closely at the original PR, it appears the benchmarking used a patched version of pyperformance to add an "eager" flavor to the async tree benchmarks. Since these flavors don't exist in the released version of pyperformance, we don't see them in the benchmark runs... |
… introduced incorrect cancellations