ci: expose cpu_parallel_mode and run planner tests with -n 0#8378
Conversation
Add a `cpu_parallel_mode` input (default `auto`) to shared-test.yml, build-test-distribute-flavor.yml, and the matrix wrapper, wired into the pytest action's `parallel_mode`. Planner callers in pr.yaml and post-merge-ci.yml override it to `'0'` to address OOM / flakiness observed under xdist `auto`. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (5)
WalkthroughGitHub Actions workflows are updated to add a configurable Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~8 minutes 🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
planner-build already produces a multi-arch (amd64 + arm64) image, so extend planner-test's platform matrix to cover both. arm64 dispatches to prod-tester-arm-v1 via shared-test.yml. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
Summary
cpu_parallel_modeinput (default'auto') toshared-test.yml,build-test-distribute-flavor.yml, and the matrix wrapper; forward it to the pytest action's existingparallel_mode.'0'for planner callers inpr.yaml(planner-test) andpost-merge-ci.yml(planner-pipeline) to address OOM / flakiness observed with xdistauto.autodefault — no behavior change outside planner.Test plan
planner-teststep logs📊 Parallelization: disabled (sequential execution) for GPU runsand passes.📊 Parallelization: auto.-n 0after merge.🤖 Generated with Claude Code
Summary by CodeRabbit
auto,none, or custom worker counts) across build and test pipelines instead of using hardcoded defaults.