-
-
Notifications
You must be signed in to change notification settings - Fork 454
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
Cysignals not compatible with glibc >= 2.34 (fedora-35, ubuntu-impish) #32576
Comments
comment:3
I was able to build stable and develop branches with the proposed solutions. It seems to be some problems with |
comment:4
I have problems with testing in develop with python3.10. Almost any test yields
in 9.4 with this packages (and also #32828 to be able to build) I get the following errors (I do not know if related with cysignals):
An example:
|
comment:5
Looks like a follow up on #30975 is needed. |
comment:6
I would guess that this goes away with #32828. |
comment:7
Replying to @mkoeppe:
I made the test in two situations. One was in the branch #32828, merged with develop, python3.10, and the corrected cysgnals package. Summarizing, the stable build produced some failed tests (quite a number, actually) but most tests were successful. |
comment:8
I would say that I opened sagemath/cysignals#157 for this. Not sure, how to proceed from here. |
comment:10
@embray, @videlec, please see sagemath/cysignals#157 (comment), |
comment:11
I don't have access to SageMath pypi. |
comment:12
Current "owners" of the project (embray, vdelecroix, jdemeyer) can invite other users to be maintainers or owners using the link https://pypi.org/manage/project/cysignals/collaboration/ |
comment:13
I have just sent out invitations to become owner of some of the PyPI projects that I created (https://pypi.org/project/sage-conf/, https://pypi.org/project/sage-setup/, https://pypi.org/project/sage-docbuild/, https://pypi.org/project/sage-sws2rst/, https://pypi.org/project/sagemath-standard/) to a few people, in order to avoid similar bottlenecks in the future. |
comment:14
Thanks, Vincent, I now have access. |
Commit: |
Branch: public/32576 |
New commits:
|
Author: Jonathan Kliem |
Branch pushed to git repo; I updated commit sha1. New commits:
|
comment:17
Next time I'll hopefully remember to first try the tarball in sage. |
comment:19
it affects sufficiently up to date Gentoo, as well. |
comment:20
Fixes the build on |
Reviewer: Matthias Koeppe |
comment:21
Thank you. |
comment:22
Thanks for your work! |
comment:23
Hi, I am trying to build sage on What do I need to do to tell sage to use this fix ? |
comment:24
|
comment:25
Ah yes indeed it works thanks. I didn't know about this option |
comment:26
You can also just manually download the package by the url in Thanks for verifying. |
comment:27
The cypari testsuite segfaults
|
comment:28
What might be a problem the assembly instructions for sigjmp At the release of 10.3 they were merged into master, but not into 10.x.0. When I started with this release of cysignals, I wasn't aware of this mess. |
comment:29
What's the situation here? Is there a new cysignals release already? |
Changed branch from public/32576 to public/32576-new |
New commits:
|
Branch pushed to git repo; I updated commit sha1. This was a forced push. New commits:
|
comment:32
tesing this on https://github.com/sagemath/sagetrac-mirror/actions/runs/1583913816 |
comment:33
Running tests in Fedora 35 on this branch merged with develop and #33013, most of the failures are gone. The timed out failure is also gone if applied directly.
Problems as in comment:4 do not appear. |
comment:34
By some reason, I built it with python3.9; rebuilt with python3.10 I got similar results as in comment:4 |
comment:35
Tests on GH Actions look OK. |
comment:36
Thanks again. |
Changed branch from public/32576-new to |
As reported on https://groups.google.com/g/sage-devel/c/Wz5Rhjnn4sU/m/rUd1PQbwAwAJ.
Cysignals is not compatible with glibc >= 2.34. Fedora 35 is using it and other systems will follow.
This is the cysignals issue: sagemath/cysignals#150
and this is intended to solve it:
sagemath/cysignals#151.
CC: @embray @videlec @dimpase @vbraun @orlitzky
Component: cython
Author: Jonathan Kliem
Branch/Commit:
8c5551a
Reviewer: Matthias Koeppe
Issue created by migration from https://trac.sagemath.org/ticket/32576
The text was updated successfully, but these errors were encountered: