8000 BLD: use sonoma image on Cirrus for wheel build by mattip · Pull Request #29061 · numpy/numpy · GitHub
[go: up one dir, main page]
Skip to content

BLD: use sonoma image on Cirrus for wheel build #29061

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
Merged
merged 1 commit into from
May 27, 2025
Merged

BLD: use sonoma image on Cirrus for wheel build #29061

merged 1 commit into from
May 27, 2025

Conversation

mattip
Copy link
Member
@mattip mattip commented May 27, 2025

As noted in #29039, cirrus no longer supports monteray. The only image they support is sonoma. This is for the wheel build with OpenBLAS, targeting macos_11_arm64. Maybe we should move that build to github, which provides free runners?

When bumping the OpenBLAS version in #29039, I see many failures related to OpenBLAS. So this is an attempt to change only one parameter to see if the errors are due to the image/xcode version, or due to an OpenBLAS regression.

@github-actions github-actions bot added the 36 - Build Build related PR label May 27, 2025
@mattip mattip mentioned this pull request May 27, 2025
Merged
@andyfaff
Copy link
Member

The job that's causing the issues is attempting to build a wheel compatible back to the earliest arm64 versions, built against OpenBLAS because Accelerate wasn't usable for our purposes until v14.

I'm trying to run the test at home, see https://github.com/numpy/numpy/wiki/Debugging-CI-guidelines#running-cirrus-ci-configurations-locally, but it takes forever to download an image and times out.

@mattip
Copy link
Member Author
mattip commented May 27, 2025

The build here succeeded, Three wheels were built. The logs are truncated, the last build has some warnings during testing. I think it is fine to merge this as is, no?

=============================== warnings summary ===============================
linalg/tests/test_linalg.py::TestDet::test_types[complex64]
linalg/tests/test_linalg.py::TestDet::test_types[complex128]
  /private/var/folders/7m/kxwv39y54d9g6lmkzwbqwk8r0000gn/T/cibw-run-7qmfdqox/cp312-macosx_arm64/venv-test-arm64/lib/python3.12/site-packages/numpy/linalg/_linalg.py:2430: RuntimeWarning: divide by zero encountered in det
    r = _umath_linalg.det(a, signature=signature)
linalg/tests/test_linalg.py::TestDet::test_types[complex64]
linalg/tests/test_linalg.py::TestDet::test_types[complex128]
  /private/var/folders/7m/kxwv39y54d9g6lmkzwbqwk8r0000gn/T/cibw-run-7qmfdqox/cp312-macosx_arm64/venv-test-arm64/lib/python3.12/site-packages/numpy/linalg/_linalg.py:2430: RuntimeWarning: invalid value encountered in det
    r = _umath_linalg.det(a, signature=signature)
linalg/tests/test_linalg.py::TestDet::test_types[complex64]
linalg/tests/test_linalg.py::TestDet::test_types[complex128]
  /private/var/folders/7m/kxwv39y54d9g6lmkzwbqwk8r0000gn/T/cibw-run-7qmfdqox/cp312-macosx_arm64/venv-test-arm64/lib/python3.12/site-packages/numpy/linalg/_linalg.py:2373: RuntimeWarning: divide by zero encountered in slogdet
    sign, logdet = _umath_linalg.slogdet(a, signature=signature)
linalg/tests/test_linalg.py::TestDet::test_types[complex64]
linalg/tests/test_linalg.py::TestDet::test_types[complex128]
  /private/var/folders/7m/kxwv39y54d9g6lmkzwbqwk8r0000gn/T/cibw-run-7qmfdqox/cp312-macosx_arm64/venv-test-arm64/lib/python3.12/site-packages/numpy/linalg/_linalg.py:2373: RuntimeWarning: invalid value encountered in slogdet
    sign, logdet = _umath_linalg.slogdet(a, signature=signature)
linalg/tests/test_linalg.py::test_xerbla_override
  /private/var/folders/7m/kxwv39y54d9g6lmkzwbqwk8r0000gn/T/cibw-run-7qmfdqox/cp312-macosx_arm64/venv-test-arm64/lib/python3.12/site-packages/numpy/linalg/tests/test_linalg.py:1984: DeprecationWarning: This process (pid=23099) is multi-threaded, use of fork() may lead to deadlocks in the child.
    pid = os.fork()
-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
============================= slowest 10 durations =============================
15.84s call     linalg/tests/test_regression.py::TestRegression::test_openblas_threading
7.48s call     random/tests/test_extending.py::test_cython
5.85s call     _core/tests/test_mem_overlap.py::test_may_share_memory_harder_fuzz
5.63s call     tests/test_warnings.py::test_warning_calls
5.47s call     _core/tests/test_mem_overlap.py::test_may_share_memory_easy_fuzz
5.10s call     _core/tests/test_multiarray.py::TestBool::test_count_nonzero_all
4.94s call     linalg/tests/test_linalg.py::TestCond::test_generalized_sq_cases
3.78s setup    _core/tests/test_cython.py::test_is_timedelta64_object
3.40s setup    f2py/tests/test_character.py::TestStringAssumedLength::test_gh24008
3.29s setup    f2py/tests/test_character.py::TestMiscCharacter::test_gh18684
47539 passed, 532 skipped, 45 xfailed, 5 xpassed, 9 warnings in 176.92s (0:02:56)

@andyfaff
Copy link
Member
8000 andyfaff commented May 27, 2025

The failing run in the other test was using Sequoia, this runs with Sonoma. Not sure why there's a fail with Sequoia, but not Sonoma. Anyway, I don't think we have a choice with upgrading. Let's merge this.

@andyfaff andyfaff merged commit 005ce76 into numpy:main May 27, 2025
7 checks passed
@charris charris added the 09 - Backport-Candidate PRs tagged should be backported label May 27, 2025
@charris charris mentioned this pull request May 27, 2025
Merged
@charris charris removed the 09 - Backport-Candidate PRs tagged should be backported label May 29, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
36 - Build Build related PR
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@mattip @andyfaff @charris
0