Repository navigation
ci: run Python CodeQL on standard Ubuntu ARM64 - #3834
Merged
Daniel Meppiel (danielmeppiel) merged 1 commit intoNov 18, 2026
Merged
Conversation
In a same-source, four-core hosted A/B, Python scan runtime fell from 7m02s to 2m56s with identical rules and findings. Keep the matrix, analysis category and Actions runner unchanged. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Daniel Meppiel (danielmeppiel)
requested a review
from Sergio Sisternes (sergio-sisternes-epam)
as a code owner
November 18, 2026 07:51
Contributor
There was a problem hiding this comment.
🟢 Approval recommended
No active review issues are recorded for the CodeQL runner split and accompanying documentation update.
No open issues 🎉
What changed in this PR
This follow-up to #3830 targets CodeQL scan time as part of the CI speedup work tracked in #3581, without changing analysis settings or release gates.
Changes:
- Run Python CodeQL on
ubuntu-24.04-arm, while keeping GitHub Actions analysis onubuntu-latest. - Document the runner split in the contributor development guide.
| File | Description |
|---|---|
docs/src/content/docs/contributing/development-guide.md |
Documents ARM64 for Python analysis and x86_64 for Actions analysis. |
.github/workflows/codeql.yml |
Selects the runner based on the existing language matrix. |
Review effort: Balanced
Daniel Meppiel (danielmeppiel)
deleted the
danielmeppiel-codeql-python-arm
branch
November 18, 2026 08:04
9 of 16 tasks
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
Move only Python CodeQL to the existing
ubuntu-24.04-armrunner. The Actions-language job stays onubuntu-latest. No matrix axis, check name, triggers, queries, reporting category, upload, permissions, cache settings or release gates change. The contributor guide documents the runner split.Measured: on the same-source, same-query comparison, Python took 7:06 on x86 vs 3:02 on ARM, including the initial 4s / 6s of queue time. Actual runner occupation was 7:02 vs 2:56: approximately 4.1 fewer runner-minutes per scan, not a billing estimate. Both had four CPU threads and about 14.5 GB available to CodeQL. The uploaded SARIF contained identical rule IDs, findings, file locations and partial fingerprints: 43 rules, 26 findings; 2,426 of 2,426 Python files scanned on both. No scan status or evidence was removed.
Why this job: CodeQL finished last on five of the first eight successful build-bearing pushes after #3830. On the latest comparison base, all checks took 8:01; the build took 7:15. A faster Python scan would move that push's critical path to the build, a projected 46s end-to-end saving, not four minutes. Even a free Windows binary build would have saved zero seconds on the median push, so no additional shards or jobs are proposed.
Note
Both isolated benchmark jobs used the pinned production CodeQL 2.27.4 and
overlay-baseto include main-style extraction and queries. A branch run doesn't publish the default branch's overlay cache or database bundles; the 7:15 figure is a projection, not a second production measurement. Production keeps its automatic overlay/cache and database upload behavior.Issue and approved scope
Issue: Refs #3581; follow-up to #3830, not a substitute for #3582. This does not complete #3581.
Human scope-approval comment: Daniel requested measured, uncomplicated follow-on CI speedups and ordinary review and landing through Lane L in the coordinating session. No separate public approval comment for this runner change is cited; the older release-only approval is not treated as one.
Type of change
Testing
Tested locally
All existing tests pass
Added tests for new functionality (if applicable)
actionlint -shellcheck= .github/workflows/codeql.yml— passed.uv run --frozen --extra dev pytest -q tests/unit/test_security_workflow_contracts.py— 4 passed. Covers unchanged trigger, language and upload contracts. No product behavior or new test selection.git diff --check— passed; local CI lint mirror passed.Hosted x86 and ARM runs both completed on the first attempt and uploaded accepted SARIF. No added capacity or settings change.
Spec conformance (OpenAPM v0.1)
If this PR changes behaviour that an OpenAPM v0.1
req-XXXcovers,confirm the three-step ritual in the
development guide:
docs/src/content/docs/specs/openapm-v0.1.mdupdated(new/changed
<a id="req-XXX"></a>anchor + prose + Appendix Crow).
docs/src/content/docs/specs/manifests/openapm-v0.1.requirements.ymlupdated.
@pytest.mark.req("req-XXX")test undertests/spec_conformance/added or extended.CONFORMANCE.{md,json}regenerated viauv run --extra dev python -m tests.spec_conformance.gen_statementand committed.
Co-authored-by: Copilot 223556219+Copilot@users.noreply.github.com