Add environment version reporting to CI workflows#538
Merged
Conversation
Reports driver, ROCm, Python, PyTorch, Triton, iris, and system versions after GPU acquisition for documentation and debugging. Co-Authored-By: Claude Opus 4 <noreply@anthropic.com>
github-actions
Bot
added
in-progress
Versions are now rendered as a markdown table on the workflow run summary page via $GITHUB_STEP_SUMMARY, in addition to stdout logging. Co-Authored-By: Claude Opus 4 <noreply@anthropic.com>
Try /sys/module/amdgpu/version and modinfo before amd-smi/rocm-smi for driver version. Remove iris row since it's always the latest commit being tested. Co-Authored-By: Claude Opus 4 <noreply@anthropic.com>
Replaces regex/grep chains with structured JSON from amd-smi version --json and Python introspection. Single Python script inside the container emits JSON, parsed outside for stdout + job summary. Co-Authored-By: Claude Opus 4 <noreply@anthropic.com>
collect_versions.py runs inside the container and emits JSON. report_versions.sh is now a thin wrapper: calls container_exec, captures JSON, formats stdout + job summary on the host side. Co-Authored-By: Claude Opus 4 <noreply@anthropic.com>
Contributor
There was a problem hiding this comment.
Pull request overview
Note
Copilot was unable to run its full agentic suite in this review.
Adds environment/version reporting to GPU CI workflows to improve debuggability and run-to-run comparisons by logging driver/ROCm/Python/PyTorch/Triton/GPU/kernel details.
Changes:
- Added
.github/scripts/collect_versions.pyto collect environment versions inside the container and emit JSON. - Added
.github/scripts/report_versions.shto execute collection in-container and write a GitHub Actions job summary table + stdout report. - Inserted a “Report environment versions” step into four GPU workflows after GPU acquisition.
Reviewed changes
Copilot reviewed 6 out of 6 changed files in this pull request and generated 3 comments.
Show a summary per file
| File | Description |
|---|---|
| .github/workflows/iris-tests.yml | Adds a CI step to report environment versions after GPU acquisition. |
| .github/workflows/iris-performance-regression-test.yml | Adds the environment-report step to benchmark workflow for better traceability. |
| .github/workflows/iris-nightly-triton-test.yml | Adds the environment-report step to nightly Triton workflow logs/summaries. |
| .github/workflows/iris-external-validation-test.yml | Adds the environment-report step to both external validation jobs. |
| .github/scripts/report_versions.sh | New host-side script to run collection in-container and format output + step summary. |
| .github/scripts/collect_versions.py | New in-container collector for driver/ROCm/Python/PyTorch/Triton/GPU/kernel info. |
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.
Summary
report_versions.shscript that logs driver (amd-smi/rocm-smi), ROCm, Python, PyTorch (including HIP version, GPU count, gcnArchName), Triton, iris, and kernel versionsMotivation
CI logs currently don't capture environment versions. When debugging failures or comparing results across runs, we need to know exactly what driver/ROCm/Triton versions were active. This adds that documentation with zero impact on test execution.
Test plan
🤖 Generated with Claude Code