Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[CI] Fix race condition by unsetting merge_ref for Windows tasks #16757

Merged
merged 1 commit into from
Jan 23, 2025

Conversation

aelovikov-intel
Copy link
Contributor

That input parameter was introduced to sycl-windows-build.yml in #16560 but it's unclear why.

Note that linux version has had it for quite some time, but we started unsetting it explicitly in #10720. The reason we had it for Linux in the first place is because GITHUB_SHA pointed to the PR's source branch for the pull_request_target trigger that we used at that time. With switching to pull_request trigger it became unnecessary as GITHUB_SHA points to the merge commit already.

This change should fix current race condition when a post-commit CI job can test change landed after the current commit if there is a delay with runner allocation.

That input parameter was introduced to `sycl-windows-build.yml` in
intel#16560 but it's unclear why.

Note that linux version has had it for quite some time, but we started
unsetting it explicitly in intel#10720. The
reason we had it for Linux in the first place is because `GITHUB_SHA`
pointed to the PR's source branch for the `pull_request_target` trigger
that we used at that time. With switching to `pull_request` trigger it
became unnecessary as `GITHUB_SHA` points to the merge commit already.

This change should fix current race condition when a post-commit CI job
can test change landed after the current commit if there is a delay with
runner allocation.
Copy link
Contributor

@sarnex sarnex left a comment

Choose a reason for hiding this comment

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

the fact we have to explicitly set it to empty is kinda gross but we already do it everywhere so lgtm

@aelovikov-intel
Copy link
Contributor Author

the fact we have to explicitly set it to empty is kinda gross but we already do it everywhere so lgtm

IMO, we should remove or switch the default, but I'd like to hear from @KornevNikita about his use-case first. In the meanwhile, fixing the CI can't wait :)

@aelovikov-intel aelovikov-intel merged commit 4f29c44 into intel:sycl Jan 23, 2025
27 checks passed
@aelovikov-intel aelovikov-intel deleted the unset-merge-ref branch January 23, 2025 23:37
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants