Group locals by written type in binary writer#8568
Merged
Conversation
We have logic to group locals by type in the binary writer to take advantage of the run-length encoding of locals. But that logic previously grouped locals by their IR types, rather than the types that would actually be written to the binary. These can differ in when the IR uses more precise types than are allowed to be written given the enabled feature set. For example, the IR might use exact types but have to write inexact types because custom descriptors are not enabled. In such cases, it is possible that different groups of locals would be written with the same type, which is suboptimal. Fix the problem by grouping locals by their written types given the enabled features. Fixes #7934.
kripken
approved these changes
Apr 2, 2026
tlively
commented
Apr 2, 2026
tlively
added a commit
that referenced
this pull request
Apr 7, 2026
In #8568 we optimized the grouping of locals in the binary writer to account for how types will be written given the enabled features. However, that change did not properly update the handling of scratch locals accordingly, leading to inconsistencies in the indices assigned to local types in different locations. Fix the problem by reverting the changes from #8568 and handling the mapping from IR types to written types at a lower level; specifically, create a new `TypeIndexMap` type that extends `InsertOrderedMap` but always applies `asWrittenGivenFeatures` to its keys. Use this new map type both for the `numLocalsByType` map and the `scratchLocals` map.
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
We have logic to group locals by type in the binary writer to take advantage of the run-length encoding of locals. But that logic previously grouped locals by their IR types, rather than the types that would actually be written to the binary. These can differ in when the IR uses more precise types than are allowed to be written given the enabled feature set. For example, the IR might use exact types but have to write inexact types because custom descriptors are not enabled. In such cases, it is possible that different groups of locals would be written with the same type, which is suboptimal. Fix the problem by grouping locals by their written types given the enabled features.
Fixes #7934.