Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Compatibility with scipy 1.12 #37123

Merged
merged 1 commit into from
Feb 2, 2024
Merged

Compatibility with scipy 1.12 #37123

merged 1 commit into from
Feb 2, 2024

Conversation

antonio-rojas
Copy link
Contributor

Drop usage of some numpy function aliases which are no longer provided in 1.12

@tornaria
Copy link
Contributor

LGTM, will test later.

BTW, there are some changes in openblas 0.3.26 that cause some regressions on scipy (either scipy 1.12 or 1.11.4 is broken). The fix in scipy/scipy#19909 didn't make it to scipy 1.12.

@github-actions GitHub Actions
Copy link

Documentation preview for this PR (built with commit 78fee0a; changes) is ready! 🎉

@tornaria tornaria mentioned this pull request Jan 21, 2024
Closed
@antonio-rojas
Copy link
Contributor Author

BTW, there are some changes in openblas 0.3.26 that cause some regressions on scipy (either scipy 1.12 or 1.11.4 is broken). The fix in scipy/scipy#19909 didn't make it to scipy 1.12.

We've been shipping 0.3.26 for a while, seen no issues on either scipy of sage test suites.

@kiwifb
Copy link
Member

kiwifb commented Jan 21, 2024

Definitely needed, and it appears those are the only instances that need changing. At least those functions.

vbraun pushed a commit to vbraun/sage that referenced this pull request Jan 22, 2024
sagemathgh-37123: Compatibility with scipy 1.12
fb8288a 
Drop usage of some numpy function aliases which are no longer provided
in 1.12

URL: sagemath#37123
Reported by: Antonio Rojas
Reviewer(s): François Bissey
tornaria added a commit to tornaria/void-packages that referenced this pull request Jan 22, 2024
@tornaria
sagemath: patch for scipy 1.12.
cd2f090 
See: sagemath/sage#37123

Also apply a few minor fixes.
@tornaria tornaria mentioned this pull request Jan 22, 2024
Closed
@tornaria
Copy link
Contributor

Everything looks fine: void-linux/void-packages#48317

There's an unrelated failure on i686 there, a timeout in:

...
sage: R.<x> = PolynomialRing(GF(65537), implementation="FLINT") ## line 701 ##
sage: f = R.random_element(9973) * R.random_element(10007) ## line 702 ##
sage: alarm(0.5); f.factor() ## line 703 ##

Have you seen this? I think these things occasionally happen (around the alarm()) calls, not specific to i686. My guess is the alarm signal gets lost. The test looks stupid but I think it's there to make sure f.factor() can be interrupted (say, by hitting ctrl-C, if that signal gets lots one can hit ctrl-C again ofc). Not sure what to do about this (my first reflex is to add a second alarm so if the first one is lost the second one gets through, but alarm(2) doesn't allow for more than one alarm.

tornaria added a commit to tornaria/void-packages that referenced this pull request Jan 22, 2024
@tornaria
sagemath: patch for scipy 1.12.
9636e16 
See: sagemath/sage#37123

Also apply a few minor fixes.
@antonio-rojas
Copy link
Contributor Author

Have you seen this?

I haven't, but we only build for x86_64

@kiwifb
Copy link
Member

kiwifb commented Jan 22, 2024

Have you seen this?

I haven't, but we only build for x86_64

Same here, I am only building for x86_64 right now.

@tornaria
Copy link
Contributor

Have you seen this?

I haven't, but we only build for x86_64

Same here, I am only building for x86_64 right now.

I don't think this has anything to do with 32 bit. It seems more like the alarm signal is lost because of some race condition (?), and it triggers in CI since gh runners are sometimes very heavily loaded.

ahesford pushed a commit to void-linux/void-packages that referenced this pull request Jan 23, 2024
@tornaria @ahesford
sagemath: patch for scipy 1.12.
26bdd8b 
See: sagemath/sage#37123

Also apply a few minor fixes.

Closes: #48317 [via git-merge-pr]
vbraun pushed a commit to vbraun/sage that referenced this pull request Jan 24, 2024
sagemathgh-37123: Compatibility with scipy 1.12
54eec46 
    
Drop usage of some numpy function aliases which are no longer provided
in 1.12
    
URL: sagemath#37123
Reported by: Antonio Rojas
Reviewer(s): François Bissey, Gonzalo Tornaría
vbraun pushed a commit to vbraun/sage that referenced this pull request Jan 27, 2024
sagemathgh-37123: Compatibility with scipy 1.12
05b0d79 
    
Drop usage of some numpy function aliases which are no longer provided
in 1.12
    
URL: sagemath#37123
Reported by: Antonio Rojas
Reviewer(s): François Bissey, Gonzalo Tornaría
vbraun pushed a commit to vbraun/sage that referenced this pull request Jan 29, 2024
sagemathgh-37123: Compatibility with scipy 1.12
4700ebe 
    
Drop usage of some numpy function aliases which are no longer provided
in 1.12
    
URL: sagemath#37123
Reported by: Antonio Rojas
Reviewer(s): François Bissey, Gonzalo Tornaría
vbraun pushed a commit to vbraun/sage that referenced this pull request Jan 30, 2024
sagemathgh-37123: Compatibility with scipy 1.12
d8a5e02 
    
Drop usage of some numpy function aliases which are no longer provided
in 1.12
    
URL: sagemath#37123
Reported by: Antonio Rojas
Reviewer(s): François Bissey, Gonzalo Tornaría
@vbraun vbraun merged commit 31eee6e into sagemath:develop Feb 2, 2024
14 of 20 checks passed
@antonio-rojas antonio-rojas deleted the scipy-1.12 branch February 2, 2024 21:43
@tornaria tornaria mentioned this pull request Feb 9, 2024
Merged
3 tasks
vbraun pushed a commit to vbraun/sage that referenced this pull request Feb 11, 2024
sagemathgh-37270: Allow python 3.12 and scipy 1.12
75afadb 
    
This is a minimal PR to allow building with python 3.12, given the
circumstances surrounding sagemath#36181.

I've tested this works with system python 3.12 by:
```
./bootstrap
./configure --disable-doc --disable-editable --enable-system-site-
packages
env 'MAKE=make -j36' make
./sage -tp 36 --all
./sage -tp 36 --all --long
```

It only gave the expected failures in
`src/sage/matroids/database_collections.py` from sagemath#37140.

Since scipy 1.12 works ok after sagemath#37123 and it's just a one-liner to
change the required version, I included it here.

No attempt is made at upgrading anything in sage-the-distro.

### 📝 Checklist

- [x] The title is concise, informative, and self-explanatory.
- [x] The description explains in detail what this PR is about.
- [x] I have linked a relevant issue or discussion.

EDIT: rebased on top of sagemath#36983 to address reviewer suggestion.
    
URL: sagemath#37270
Reported by: Gonzalo Tornaría
Reviewer(s): Aliaksei Urbanski, Gonzalo Tornaría, Tobias Diez
@mkoeppe mkoeppe added this to the sage-10.3 milestone Mar 7, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@tornaria tornaria tornaria approved these changes

@kiwifb kiwifb kiwifb approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
sage-10.3
Development

Successfully merging this pull request may close these issues.
5 participants
@antonio-rojas @tornaria @kiwifb @vbraun @mkoeppe