Bench: Add clean-heap and forced-GC krausest update probes#235
Merged
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Temporary diagnostic for #223. That PR's per-AST binding-plan cache regresses
update-10th-50by +45% on krausest across two runs, yet the changed code never executes during that op. The leading explanation is suite-ordering GC pressure:update-10th-50runs after create-1k / create-10k / replace-1k / append-1k in one page load, and the cache shifts the heap trajectory those create-heavy ops leave behind.Two probes pin it down. Both are read from the same single page run tachometer already does, so no extra config.
update-isolated-50runs the identical update workload first, on a clean heap with no create-heavy ops before it.update-gc-50runs it in place but forces a full GC right before the measure. Needs--js-flags=--expose-gc, added to the krausest config.The read on #223's bench run: if
update-10th-50regresses whileupdate-isolated-50andupdate-gc-50stay flat, the cost is reclaimable GC pressure from op ordering, not a per-user regression.Reverted once #223's run has read these. Bench harness is pinned from main, so #223 picks these up automatically on its next run.