Skip to content

ci: disable unused coverage instrumentation - #68

Merged
bernalde merged 1 commit into
mainfrom
codex/issue-67-disable-unused-coverage
Jul 21, 2026
Merged

ci: disable unused coverage instrumentation#68
bernalde merged 1 commit into
mainfrom
codex/issue-67-disable-unused-coverage

Conversation

@bernalde

@bernalde bernalde commented Jul 21, 2026

Copy link
Copy Markdown
Member

Summary

  • disable unused coverage instrumentation across the full DWave Julia/OS matrix
  • retain every backend, parity, topology, and QUBODrivers test on every existing matrix leg
  • remove stale commented coverage/upload configuration

Validation

  • actionlint
  • git diff --check
  • full Julia 1.10/current × Ubuntu/Windows CI matrix (post-push)

Measurement

Issue #67 recorded 34.6 aggregate runner-minutes. Current-Julia test logs showed 92 dependencies precompiled under coverage in 166 seconds on Ubuntu and 217 seconds on Windows.

  • attempt 1: 32.5 aggregate runner-minutes; longest job 11m07s
  • warm rerun: 29.6 aggregate runner-minutes; longest job 10m06s
  • sampled CI logs show coverage=false; both complete four-job matrices passed with every backend and interface test retained

The warm rerun used 2.9 fewer aggregate runner-minutes than attempt 1 and 5.0 fewer than the audited baseline. These timings still include material runner/cache variance, so the reliable change is removal of unused coverage work from all four jobs.

Branch Hygiene

Closes #67

@bernalde
bernalde marked this pull request as ready for review July 21, 2026 08:54
@bernalde
bernalde merged commit 7996029 into main Jul 21, 2026
8 checks passed
@bernalde
bernalde deleted the codex/issue-67-disable-unused-coverage branch July 21, 2026 15:51
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.

Reduce DWave CI cost without weakening platform coverage

1 participant