Hi all, GitHub Actions has deprecated the macos-13 image [1], which was the last image to support Intel x86-64 runners. As a result, we won't be able to use our current wheel build setup.
A quick sampling of what others are doing with their support policies: CPython dropped it to Tier 2 [2], Anaconda dropped support [3], conda-forge hasn't decided anything yet [4], PyTorch dropped support already 1.5 years ago [5], Numba is dropping wheels [6]. We will of course continue to accept issues and PR and keep things working for macOS x86-64 either way, just like for many other platforms that either don't have good CI support available or are more niche and for which we don't supply wheels to begin with. However, for wheels we do ship on PyPI, it seems like there should be high enough demand and a good case. It would probably be possible to cross-compile and test x86-64 wheels on arm64 runners with Rosetta. There are a number of downsides to that though, from extra maintenance load, slower CI (the x86-64 builds are already the slowest ones except for PyPy which we'll drop soon as well), and some things potentially not being testable/tested (e.g., the translation to arm code under the hood won't actually exercise SIMD code I'd think). Shipping official wheels that aren't normally testable feels a bit questionable. It's easy enough nowadays to pick older wheels by the user, and `pip install numpy --no-binary` will continue to work just fine. And given that support across the ecosystem will very likely quickly fall away, it feels like effort that could be better spent elsewhere. Hence I'd like to propose to drop macOS x86-64 wheels for the 2.4.0 release series. Cheers, Ralf [1] https://github.blog/changelog/2025-07-11-upcoming-changes-to-macos-hosted-runners-macos-latest-migration-and-xcode-support-policy-updates/#macos-13-is-closing-down [2] https://discuss.python.org/t/dropping-intel-mac-to-tier-2/102100/1 [3] https://www.anaconda.com/blog/intel-mac-package-support-deprecation [4] https://github.com/conda-forge/conda-forge.github.io/issues/2467 [5] https://dev-discuss.pytorch.org/t/pytorch-macos-x86-builds-deprecation-starting-january-2024/1690 [6] https://github.com/numba/numba/issues/10187
_______________________________________________ NumPy-Discussion mailing list -- [email protected] To unsubscribe send an email to [email protected] https://mail.python.org/mailman3//lists/numpy-discussion.python.org Member address: [email protected]
