Fix determinism bug in type name updating#8359
Merged
Conversation
There was previously a determinism bug where the result of updating type names could depend on the iteration order of the oldToNewTypes map. The bug occurred when the new types were a shuffling of the old types. The update loop updated the module's type names in-place for the new types, and those in-place updates could affect later results if the updated new types were later visited as old types. Fix the bug by collecting all changes to apply before applying them to the module's types and indices. This does not affect any existing tests, but it will unbreak CI for #8217.
kripken
reviewed
Feb 23, 2026
| // Do not overwrite the entry for the old type if it has already appeared | ||
| // as a new type. | ||
| if (newTypeNames.insert({old, names}).second) { | ||
| seenTypeNames.insert(names.name); |
Member
There was a problem hiding this comment.
Isn't it ok to unconditionally do this insert? I'm not sure what the comment is saying is dangerous to overwrite, as this is just a set of names - we can't overwrite a value for a key, given there are only keys?
Member
Author
There was a problem hiding this comment.
The comment is referring to the use of newTypeNames.insert({old, names}) rather than newTypeNames[old] = names.
Member
Author
There was a problem hiding this comment.
Although looking at this again, I'm not sure that's even necessary. We're iterating over the old types, so they can only ever appear once as keys. I think this insert probably always succeeds.
Member
Author
There was a problem hiding this comment.
Oh no, nevermind, because we already do newTypeNames[new_] = ... above.
kripken
approved these changes
Feb 23, 2026
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.
There was previously a determinism bug where the result of updating type
names could depend on the iteration order of the oldToNewTypes map. The
bug occurred when the new types were a shuffling of the old types. The
update loop updated the module's type names in-place for the new types,
and those in-place updates could affect later results if the updated new
types were later visited as old types.
Fix the bug by collecting all changes to apply before applying them to the module's types and indices. This does not affect any existing tests,
but it will unbreak CI for #8217.