Avoid unnecessarily creating TypeReferences on import #970
Conversation
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
|
@jbevain I don't think the linux CI failure is related to my changes. |
This is related to actions/runner-images#11101 I fixed it in #972 |
|
Now you can just sync your branch |
When importing a TypeSpecification, an element type or generic argument could be a TypeDefinition within the current module. When this happens, a new TypeReference does not need to be created. This can lead to a TypeReference being added to the typeref table for a type that is already in the assembly. This seems to hold together, although I don't think it's ideal. And really my goal is to avoid failing this logic in the ILLink test framework https://github.com/dotnet/runtime/blob/cec44d6dff9de95421f199f65bbc80b8296da1c0/src/tools/illink/test/Mono.Linker.Tests/TestCasesRunner/ResultChecker.cs#L56 This fix superceeds jbevain#138. I've left that fix in since that API is exposed publically and others could be depending on it. While I was adjusting this logic, I thought I would apply the same logic to a TypeReference where the scope is the current modules assembly. This case is a bit contrived as it shouldn't happen, but if it did, it would result in the same circular reference problem as jbevain#138
e1f3f5a to
a3c2663
Compare
|
@jbevain What do you think about these changes? |
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Outdated suggestions cannot be applied.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.
When importing a TypeSpecification, an element type or generic argument could be a TypeDefinition within the current module. When this happens, a new TypeReference does not need to be created. This can lead to a TypeReference being added to the typeref table for a type that is already in the assembly. This seems to hold together, although I don't think it's ideal. And really my goal is to avoid failing this logic in the ILLink test framework https://github.com/dotnet/runtime/blob/cec44d6dff9de95421f199f65bbc80b8296da1c0/src/tools/illink/test/Mono.Linker.Tests/TestCasesRunner/ResultChecker.cs#L56
This fix superceeds #138. I've left that fix in since that API is exposed publically and others could be depending on it.
While I was adjusting this logic, I thought I would apply the same logic to a TypeReference where the scope is the current modules assembly. This case is a bit contrived as it shouldn't happen, but if it did, it would result in the same circular reference problem as #138