Repository navigation
ci: allow eight hosted template legs concurrently - #434
Merged
Merged
Conversation
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The template nonvisual matrix currently admits four hosted legs at a time, so a full 13-kit run needs four waves. Allow eight existing
ubuntu-24.04legs concurrently, reducing that to two waves while retaining every selected kit/scenario,fail-fast: false, routing and timeouts.In queue run 37204698238, 13 successful legs used 70m13 runner time and spanned 23m18. Replaying those fixed durations in observed start order models 23m11 at cap 4 versus 18m09 at cap 8 (~5m02 matrix saving). This assumes available hosted slots and unchanged duration; it is not a measured speedup or removal of the Android critical path. The organization reports Free; GitHub documents a default 20 concurrent standard hosted jobs, shared with other jobs/workflows. Eight is a matrix ceiling, not reserved capacity.
Validation: 240 CPU workflow/planner/efficiency tests passed on the final head rebased onto landed #430 (
b31808f). Independent review found no functional defect or dependency on the unpublished full-queue/Integration changes. Own JS/DTS package build used at most 2 workers; normal bounded pre-push drift hooks run without bypass. No local native/GPU work or manual workflow dispatch.