Cut ~115 seconds from the new 3D adaptivity test gates#389
Open
lmoresi wants to merge 1 commit into
Open
Conversation
… mover run The similarity-class gate ran nine full uniform passes in the pure- Python engine (512x cell growth, 117s) to check a bound the stage-1a oracle already proved at 2.87M cells. The CI gate now checks the sharp part cheaply: the class count must SATURATE between passes five and six (two full isotropic levels) and plateau under the theoretical 36 — a stronger assertion than the ceiling alone, at 9s. The #376 warning-free guard and the redistribute-then-adapt composition test ran an identical mover on an identical mesh; one run now carries both sets of assertions. New 3D test files: ~162s -> ~48s. All 20 pass. Underworld development team with AI support from Claude Code
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
A test-cost trim that was pushed to the #378 branch a few minutes after the squash-merge, so it missed the window — re-applied cleanly on current development (with the #381-#385 callback series in) and verified: the four 3D adaptivity test files pass in 45s instead of ~160s.
Two changes, no loss of coverage:
The similarity-class gate ran nine full uniform refinement passes in the pure-Python engine (512x cell growth, 117 seconds) to check a bound the design-stage oracle already proved at 2.87 million cells. The gate now checks the sharp part cheaply and more strictly: the class count must saturate between passes five and six (two full isotropic levels) and plateau under the theoretical ceiling of 36. Nine seconds.
The Canonical PETSc-sync callback fires on fancy-indexed array copies — serial warning, parallel hang/crash #376 warning-free guard and the redistribute-then-adapt composition test each ran an identical node-redistribution pass on an identical mesh; one mover run now carries both sets of assertions.
Underworld development team with AI support from Claude Code