Skip to content

Conversation

bgilbert
Copy link
Collaborator

They're only available in mesonbuild/wrapdb, not in forks. As a result the emulated ppc64le job is potentially still useful, but it would likely bitrot since it wouldn't run in the main repo. Drop that job.

Self-hosted runners don't count toward the concurrent job limit so this actually improves our parallelism.

In forks, the new jobs would sit indefinitely waiting for a runner to pick them up, blocking the workflow from completing. We want to skip those jobs, but the job's if: directive can't read the matrix context. Instead, preprocess the matrix from the prelude job.

@bgilbert
Copy link
Collaborator Author

Enablement is tracked in IBM/actionspz#10.

@bgilbert
Copy link
Collaborator Author

Prep commits broken out into #2422.

@bgilbert bgilbert closed this Sep 29, 2025
@bgilbert bgilbert deleted the runners branch September 29, 2025 02:34
@bgilbert bgilbert restored the runners branch September 29, 2025 02:38
@bgilbert bgilbert reopened this Sep 29, 2025
They're only available in mesonbuild/wrapdb, not in forks.  As a result
the emulated ppc64le job is potentially still useful, but it would likely
bitrot since it wouldn't run in the main repo.  Drop that job.

Self-hosted runners don't count toward the concurrent job limit so this
actually improves our parallelism.
In forks, the ppc64le and s390x jobs sit indefinitely waiting for a
runner to pick them up, blocking the workflow from completing.  We want
to skip those jobs, but the job's if: directive can't read the matrix
context.  Instead, preprocess the matrix from the prelude job.
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.

1 participant