Conversation
Import python-ml-dtypes, pinned to Fedora rawhide (no f43 branch exists yet for this package; only f45/rawhide). One of the TensorFlow Python dependencies identified in the tensorflow-azurelinux build plan. Fedora's %prep pins setuptools >= 80.9, but azl4's python-setuptools is 78.1.1. ml_dtypes does not need setuptools >=80.9-only feature. We loosen the pin via overlay to >= 78.1. This lets us work off fedora dist-git. Upgrading setuptools is a very heavy lift and not recommended. With setuptools version reverted, the build still needs patching to work around failures due to dependencies. At some point when we are able to upgrade setuptools this workaround can be dropped. Verified: smoke-tested python3-ml-dtypes in a mock chroot (float8_e4m3fn conversions, bfloat16 array creation). Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: 83703b8a-3378-491b-88d0-6b29aabc7827
Contributor
Author
|
/azp run |
|
Azure Pipelines: 2 pipeline(s) were filtered out due to trigger conditions. |
Contributor
There was a problem hiding this comment.
Copilot review overview
🟡 Changes recommended
The setuptools overlay lacks the mandatory category and upstream-status metadata.
Review effort: Balanced
Findings: 2
Open (2)
What changed in this PR
Imports python-ml-dtypes as a TensorFlow dependency with Azure Linux compatibility adjustments.
Changes:
- Pins the Fedora Rawhide component and lowers its setuptools requirement.
- Adds Eigen 3.4.0 compatibility and NumPy test patches.
- Adds the lock and rendered RPM sources/spec.
| File | Description |
|---|---|
base/comps/python-ml-dtypes/python-ml-dtypes.comp.toml |
Defines the component and overlays. |
base/comps/python-ml-dtypes/0001-float8-add-eigen-numext-get_integer_by_size-shim.patch |
Adds the Eigen compatibility shim. |
locks/python-ml-dtypes.lock |
Pins upstream resolution. |
specs/p/python-ml-dtypes/python-ml-dtypes.spec |
Provides the rendered RPM spec. |
specs/p/python-ml-dtypes/sources |
Records the source archive checksum. |
specs/p/python-ml-dtypes/04c4dc8b23720d9d92f3cc849ffc387d5798db84.patch |
Backports the NumPy test fix. |
specs/p/python-ml-dtypes/0001-float8-add-eigen-numext-get_integer_by_size-shim.patch |
Provides the rendered compatibility patch. |
💡 Configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
|
|
||
| # Fedora's %prep pins the pyproject.toml build-system requirement to | ||
| # setuptools >= 80.9. AZL4 has 78.1.1 and an upgrade is not feasible. | ||
| # ml_dtypes' is not an actual feature requirement, we can loosen it. |
Comment on lines
+9
to
+14
| [[components.python-ml-dtypes.overlays]] | ||
| description = "Loosen setuptools pin from >=80.9 to >=78.1 to match; ml_dtypes does not use setuptools >=80.9-only feature" | ||
| type = "spec-search-replace" | ||
| section = "%prep" | ||
| regex = '^sed -i .s/"setuptools\.\*",/"setuptools >= 80\.9",/g. pyproject\.toml$' | ||
| replacement = "sed -i 's/\"setuptools.*\",/\"setuptools >= 78.1\",/g' pyproject.toml" |
This branch has not been deployed
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.
Import
python-ml-dtypes, pinned to Fedora rawhide (no f43 branch exists yet for this package; only f45/rawhide). One of the TensorFlow Python dependencies identified in the tensorflow-azurelinux build plan.Fedora's
%preppins setuptools >= 80.9, but azl4'spython-setuptoolsis 78.1.1. ml_dtypes does not need a setuptools >=80.9-only feature. We loosen the pin via overlay to >= 78.1. This lets us work off Fedora dist-git. Upgrading setuptools is a very heavy lift and not recommended.With setuptools version reverted, the build still needs patching to work around failures due to dependencies. At some point when we are able to upgrade setuptools this workaround can be dropped.
Verified: smoke-tested
python3-ml-dtypesin a mock chroot (float8_e4m3fn conversions, bfloat16 array creation).