Skip to content

[cuda.compute]: add back CI for CTK 12.0 - #10057

Merged
NaderAlAwar merged 13 commits into
NVIDIA:mainfrom
NaderAlAwar:add-python-ctk-12.0
Jul 27, 2026
Merged

[cuda.compute]: add back CI for CTK 12.0#10057
NaderAlAwar merged 13 commits into
NVIDIA:mainfrom
NaderAlAwar:add-python-ctk-12.0

Conversation

@NaderAlAwar

Copy link
Copy Markdown
Contributor

Description

closes #8156

Merge first: #8160

Checklist

  • New or existing tests cover these changes.
  • The documentation is up to date with these changes.

leofang and others added 7 commits March 24, 2026 22:44
Previously, Python test scripts extracted only the major version from
nvcc (e.g. 12) and installed cuda-toolkit==12.*, which floated to the
latest 12.x from PyPI regardless of the container's actual CTK version.
This masked issues like the nvrtc compiler bug in CTK 12.4.

Use PIP_CONSTRAINT to pin cuda-toolkit==X.Y.* (e.g. 12.9.*) matching
the container's nvcc, ensuring CI tests exercise the exact same
cuda-toolkit minor version as the devcontainer.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@copy-pr-bot

copy-pr-bot Bot commented Jul 21, 2026

Copy link
Copy Markdown
Contributor

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@github-project-automation github-project-automation Bot moved this to Todo in CCCL Jul 21, 2026
@cccl-authenticator-app cccl-authenticator-app Bot moved this from Todo to In Progress in CCCL Jul 21, 2026
@NaderAlAwar

Copy link
Copy Markdown
Contributor Author

/ok to test 707ed05

@github-actions

This comment has been minimized.

@NaderAlAwar

Copy link
Copy Markdown
Contributor Author

/ok to test a40ed81

@NaderAlAwar

Copy link
Copy Markdown
Contributor Author

/ok to test eabd8ac

@github-actions

This comment has been minimized.

@github-actions

This comment has been minimized.

@github-actions

This comment has been minimized.

@NaderAlAwar
NaderAlAwar marked this pull request as ready for review July 26, 2026 22:08
@NaderAlAwar
NaderAlAwar requested review from a team as code owners July 26, 2026 22:08
@cccl-authenticator-app cccl-authenticator-app Bot moved this from In Progress to In Review in CCCL Jul 26, 2026
@coderabbitai

coderabbitai Bot commented Jul 26, 2026

Copy link
Copy Markdown
Contributor

Review Change Stack

📝 Walkthrough

Summary by CodeRabbit

  • Bug Fixes

    • Improved compatibility for CUDA Toolkit 12.0–12.2 by skipping cooperative example tests when required CUDA Python support is unavailable.
    • Added clearer test-skip messaging explaining the compatibility limitation.
  • Tests

    • Expanded Python and free-threaded Python CI coverage to include CUDA Toolkit 12.0 across applicable test and wheel-building configurations.

Walkthrough

The PR restores CTK 12.0 coverage across selected Python CI matrix entries and conditionally skips cooperative CUDA examples when nvJitLinkVersion is unavailable.

Changes

CTK 12.0 Python coverage

Layer / File(s) Summary
Expand Python CI CTK coverage
ci/matrix.yaml
Python test, free-threaded, and wheel matrix entries now include CTK 12.0 alongside CTK 12.X, 13.0, and 13.X.
Gate cooperative example tests
python/cuda_cccl/tests/test_examples.py
Dynamic cooperative example tests are skipped when cuda.bindings.nvjitlink.version() is unavailable.

Assessment against linked issues

Objective Addressed Explanation
Restore CTK 12.0 Python CI coverage [#8156]
Add CTK 12.0 testing for cccl.c [#8156] No cccl.c CI matrix entry is changed.

Possibly related PRs

  • NVIDIA/cccl#8160: Updates CTK version selection in the same Python CI matrix areas.

Suggested reviewers: tpn, shwina, wmaxey


Comment @coderabbitai help to get the list of available commands.

@coderabbitai coderabbitai Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Actionable comments posted: 1


ℹ️ Review info
⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 17ce6224-c87c-4313-848c-0caf82ef5b68

📥 Commits

Reviewing files that changed from the base of the PR and between 45b8bb9 and 4e3ede4.

📒 Files selected for processing (2)
  • ci/matrix.yaml
  • python/cuda_cccl/tests/test_examples.py

Comment thread python/cuda_cccl/tests/test_examples.py Outdated
@NaderAlAwar

Copy link
Copy Markdown
Contributor Author

/ok to test 025c04a

@github-actions

Copy link
Copy Markdown
Contributor

🥳 CI Workflow Results

🟩 Finished in 2h 30m: Pass: 100%/532 | Total: 4d 05h | Max: 2h 21m | Hits: 99%/673170

See results here.

@miscco miscco left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

That is great

@NaderAlAwar
NaderAlAwar merged commit 2cb708a into NVIDIA:main Jul 27, 2026
611 of 613 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

Archived in project

Development

Successfully merging this pull request may close these issues.

[cuda.compute]: Add back CI Coverage for CTK 12.0

3 participants