ci(windows): make PR checks opt-in and main jobs advisory - #3268
Merged
Conversation
Signed-off-by: Piotr Mlocek <pmlocek@nvidia.com>
pimlock
requested review from
a team,
derekwaynecarr,
mrunalp and
sjenning
as code owners
September 11, 2026 01:13
drew
approved these changes
Sep 11, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Windows CI currently runs expensive lint/tests on every PR mirror and merge group. Make PR validation opt-in with
test:windows, remove merge-group runs, and keep main/manual validation advisory so Windows failures do not fail those workflows.The sample x64 PR job took 24m 53s: 9m 49s linting, 9m 14s compiling/running tests, and 4m 52s cleaning/saving the cache. Test execution itself took only 70 seconds. The target-cache save failed with an explicit GitHub rate-limit warning after cleanup and compression. sccache recorded 213 hits, 792 misses, and 792 write errors, so caching is only partly effective. Rate limiting may also explain the sccache write failures, but the statistics alone do not establish that cause.
These timings and cache failures motivate reducing automatic Windows runs while cache reliability and warm-run performance are investigated. This PR changes when validation runs and how failures affect workflow status; it does not fix the cache failures.
Related Issue
No issue required: localized CI configuration maintenance requested by a maintainer; no product or runtime behavior changes.
Changes
test:windowsthrough the existing PR gate for PR lint/tests. Created the repository label; applying it requires re-running the current mirror workflow or pushing another mirrored commit.continue-on-error: trueon main/manual cache-seed and binary-build jobs. Opt-in PR failures remain visible normally.Testing
mise run pre-commitpassesactionlintpasses for.github/workflows/windows-msvc.ymlgit diff --checkpassesChecklist