Build and test with CUDA 13.3.0

[bdice]

Contributes to rapidsai/build-planning#286

• uses CUDA 13.3.0 to build and test
• updates to CUDA 13.3.0 devcontainers

Notes for Reviewers

This switches GitHub Actions workflows to the cuda-13.3.0 branch from here: rapidsai/shared-workflows#574

A future round of PRs will revert that back to main, once all of RAPIDS is migrated.

[copy-pr-bot]

Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually.

Contributors can view more details about this message here.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: dc8e8e82-edb2-40ce-8c5a-e627326aa3a2

📥 Commits

Reviewing files that changed from the base of the PR and between 0c41093 and aefc2a5.

📒 Files selected for processing (1)

• dependencies.yaml

🚧 Files skipped from review as they are similar to previous changes (1)

• dependencies.yaml

---

📝 Walkthrough

Summary by CodeRabbit

• Chores

  • Updated development environment and container configurations from CUDA 13.2 to 13.3.
  • Pinned CI/CD reusable workflow references and updated CI matrices to CUDA 13.3.
  • Adjusted build/tooling and environment naming to align with CUDA 13.3.

• Documentation

  • Updated setup and contribution docs to reference CUDA 13.3 environment files and commands.

Walkthrough

This PR updates devcontainer configs, CI workflow references, conda environment specs, dependency matrices, and docs to replace CUDA 13.2 with CUDA 13.3 across the repository.

Changes

CUDA 13.3 infrastructure update

Layer / File(s)	Summary
Devcontainer CUDA 13.3 configuration .devcontainer/cuda13.3-conda/devcontainer.json, .devcontainer/cuda13.3-pip/devcontainer.json	Both conda and pip devcontainer variants update base image/build args, cache references, container names, conda/venv environment paths, and lifecycle commands to target CUDA 13.3.
CI/CD workflow references pinned to CUDA 13.3.0 .github/workflows/build.yaml, .github/workflows/pr.yaml, .github/workflows/pr_issue_status_automation.yml, .github/workflows/test.yaml, .github/workflows/trigger-breaking-change-alert.yaml	All shared-workflow uses: refs replaced with @cuda-13.3.0 across build, test, PR validation, and automation workflows; devcontainer CUDA matrices updated to include 13.3.
Conda environment specifications for CUDA 13.3 conda/environments/all_cuda-133_arch-*.yaml, conda/environments/clang_tidy_cuda-133_arch-x86_64.yaml, conda/environments/cpp_all_cuda-133_arch-x86_64.yaml	Conda environment YAML files update cuda-version=13.3 pins and are renamed to reflect the new CUDA variant across architectures and profiles.
Dependency matrix and documentation updates dependencies.yaml, BUILD.md, CONTRIBUTING.md	dependencies.yaml adds cuda: "13.3" entries for conda and requirements outputs; BUILD.md and CONTRIBUTING.md reference the CUDA 13.3 environment filenames.

🎯 2 (Simple) | ⏱️ ~8 minutes

Possibly related PRs

• rapidsai/cuml#8081: Implements the same CUDA 13.3 version bump pattern across devcontainer configurations, workflow pinning, and conda environment specifications.
• rapidsai/cuml#7907: Addresses related dependencies.yaml matrix logic for CUDA version handling and nvidia-nvjitlink toolkit selection across CUDA minor versions.

Suggested labels

ci, CUDA/C++

Suggested reviewers

• jameslamb
• msarahan
• dantegd

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The PR title accurately and concisely describes the main change: upgrading build and test infrastructure to use CUDA 13.3.0, which aligns with the comprehensive changeset updating devcontainers, workflows, and dependency configurations.
Description check	✅ Passed	The PR description is directly related to the changeset, referencing the issue it contributes to, explaining the CUDA version upgrade, and clarifying the workflow reference changes to the cuda-13.3.0 branch.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@CONTRIBUTING.md`:
- Line 147: Update the CONTRIBUTING.md entry that runs "conda env create --yes
-n cuml-clang-tidy -f conda/environments/clang_tidy_cuda-133_arch-$(uname
-m).yaml" to be explicit about x86_64-only support: replace or annotate that
command to target the x86_64 YAML (e.g., use arch=x86_64 instead of $(uname -m))
and add the suggested follow-up note telling users the conda clang-tidy
environment is provided for x86_64 only and to use the Docker-based clang-tidy
workflow on other architectures.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 0f8f56be-9c2c-4f99-b2e1-5dc9892275fa

📥 Commits

Reviewing files that changed from the base of the PR and between 7e086f7 and 0c41093.

📒 Files selected for processing (14)

• .devcontainer/cuda13.3-conda/devcontainer.json
• .devcontainer/cuda13.3-pip/devcontainer.json
• .github/workflows/build.yaml
• .github/workflows/pr.yaml
• .github/workflows/pr_issue_status_automation.yml
• .github/workflows/test.yaml
• .github/workflows/trigger-breaking-change-alert.yaml
• BUILD.md
• CONTRIBUTING.md
• conda/environments/all_cuda-133_arch-aarch64.yaml
• conda/environments/all_cuda-133_arch-x86_64.yaml
• conda/environments/clang_tidy_cuda-133_arch-x86_64.yaml
• conda/environments/cpp_all_cuda-133_arch-x86_64.yaml
• dependencies.yaml

[coderabbitai]

⚠️ Potential issue | 🟡 Minor | ⚡ Quick win

Make the clang-tidy env command explicit about x86_64 support.

This command uses $(uname -m), but the CUDA 13.3 clang-tidy env in this PR is x86_64-only. On aarch64 hosts, this resolves to a non-existent file and fails setup.

Suggested doc fix

-        conda env create --yes -n cuml-clang-tidy -f conda/environments/clang_tidy_cuda-133_arch-$(uname -m).yaml
+        conda env create --yes -n cuml-clang-tidy -f conda/environments/clang_tidy_cuda-133_arch-x86_64.yaml

# Optional follow-up note near this command:
# "The conda clang-tidy environment is currently provided for x86_64. On other
# architectures, use the Docker-based clang-tidy workflow above."

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@CONTRIBUTING.md` at line 147, Update the CONTRIBUTING.md entry that runs
"conda env create --yes -n cuml-clang-tidy -f
conda/environments/clang_tidy_cuda-133_arch-$(uname -m).yaml" to be explicit
about x86_64-only support: replace or annotate that command to target the x86_64
YAML (e.g., use arch=x86_64 instead of $(uname -m)) and add the suggested
follow-up note telling users the conda clang-tidy environment is provided for
x86_64 only and to use the Docker-based clang-tidy workflow on other
architectures.

[bdice]

/merge

[jameslamb]

clang-tidy is failing like this:

/__w/cuml/cuml/cpp/tests/sg/genetic/node_test.cpp: /__w/cuml/cuml/cpp/.clang-tidy:5:1: error: unknown key 'AnalyzeTemporaryDtors'
AnalyzeTemporaryDtors: false
^~~~~~~~~~~~~~~~~~~~~
Error parsing /__w/cuml/cuml/cpp/.clang-tidy: Invalid argument
/__w/cuml/cuml/cpp/.clang-tidy:5:1: error: unknown key 'AnalyzeTemporaryDtors'
AnalyzeTemporaryDtors: false
^~~~~~~~~~~~~~~~~~~~~
Error parsing /__w/cuml/cuml/cpp/.clang-tidy: Invalid argument
/__w/cuml/cuml/cpp/.clang-tidy:5:1: error: unknown key 'AnalyzeTemporaryDtors'
AnalyzeTemporaryDtors: false
^~~~~~~~~~~~~~~~~~~~~
Error parsing /__w/cuml/cuml/cpp/.clang-tidy: Invalid argument

(build link)

This looks relevant: NVIDIA/raft#2945

And looks like Dante fixed it for this repo in #8251

I'll merge in main and re-run CI.