-
-
Notifications
You must be signed in to change notification settings - Fork 25.8k
CircleCI fails on pypy3 build #18624
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
Comments
Related with #17994? |
yep, that should probably solve this |
I have had a look to the segfaults. For the sake of completness here is the list of the import giving a segfault with pypy:
|
It's possible to reproduce easily with an env dedicated to pypy from conda-forge as follows:
yields:
However if you replace the pypy artifacts generated by this PR: conda-forge/pypy3.6-feedstock#30 (comment) then it works:
|
I cannot run the full test suite on my laptop: it's too slow and takes too much memory... However, I confirm that with the patched pypy, all the failing imports reported by @cmarmo above pass on in this env. |
I removed the blocker label because I think this will be fixed upstream in the next release of pypy and maybe in a patched release on conda-forge so we do not have to anything here. |
Several pypy3 builds on circleci have been failing recently, example: https://app.circleci.com/pipelines/github/scikit-learn/scikit-learn/8663/workflows/a0349bb9-2c2f-4334-85f2-c48f09a1a30d/jobs/122623
This is kind of a blocker for the 0.24 release since conda-forge now builds for pypy.
Ping @rth
The text was updated successfully, but these errors were encountered: