Repository navigation
BUG: Pin numpy <2.5 for manylinux2014 python test venv - #2659
Merged
blowekamp merged 1 commit intoJul 29, 2026
Merged
Conversation
blowekamp
marked this pull request as ready for review
July 28, 2026 20:18
Numpy 2.5.x requires GCC >= 10.3, but manylinux2014's devtoolset-10 is 10.2.1. Cap numpy at <2.5 for building cp314t from source, and add VERBATIM so CMake escapes '<' in the pip install command.
blowekamp
force-pushed
the
fix/manylinux-numpy-cp314t-gcc-pin
branch
from
July 29, 2026 10:40
29fb840 to
0863690
Compare
Contributor
There was a problem hiding this comment.
Pull request overview
This PR updates the legacy Python packaging/test virtualenv bootstrap to avoid NumPy 2.5+ source builds that fail on manylinux2014 (CentOS 7 / GCC 10.2) for Python ABIs that lack manylinux2014-tagged NumPy wheels (e.g., free-threaded cp314t).
Changes:
- Pin NumPy to
<2.5during CI test-venv creation to avoid GCC >= 10.3 build requirements on manylinux2014. - Add explanatory comments for the pin rationale.
- Add
VERBATIMto theadd_custom_commandinvocation for safer argument escaping.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
zivy
approved these changes
Jul 29, 2026
dave3d
approved these changes
Jul 29, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Problem
The
Package.ymlpackage-dockerjob (manylinux2014, both x86_64 and aarch64) fails when building thecp314-cp314t(free-threaded Python 3.14) leg:manylinux2014 (CentOS 7) is fixed at devtoolset-10 (GCC 10.2.1), which is the newest toolchain that image ships. NumPy publishes wheels for
cp314tonly taggedmanylinux_2_27/manylinux_2_28, so pip falls back to building numpy 2.5.x from source on manylinux2014 aarch64/x86_64, which now hard-requires GCC >= 10.3 and aborts the wholemakestep (this happens while creating the local python virtualenv used for CI testing, perWrapping/Python/LegacyPackaging.cmake, before the wheel is even packaged).Fix
NumPy 2.4.x already supports Python 3NumPy 2.4.x already supports Python 3NumPy 2.4.x already supports Python 3NumPy 2.4.x already supports Python 3NumPyleNumPy 2.4.x already supports Python 3NumPy 2.4.x already supposfully NumPy 2.4.x already jNumPy 2.4.x already supports Python 3NumPy 2.4.x already supports Python 3NumPy 2.4.x already supports Python 3NumPy 2.4.x already supports Python 3NumPylePythNumPyrsNumPy 2.4.x already supports Python 3NumPy 2.4.x already supports Pytg/Python/LegacyPackaging.cmake`) — it does not change the distributed wheel's build or runtime requirements.
Follow-up
Long-term, manylinux2014's toolchain will keep falling further behind newer numpy/Python releases. Worth tracking whether to move newer Python ABI legs (e.g. free-threaded builds) to a manylinux_2_28-based image.