[5.9] SILGen: Provide a reabstracted writeback for inout parameters emitted in a more abstract closure literal context. #67251
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
… in a more abstract closure literal context. When we emit closure literals in the context of a generic parameter, the parameters of the closure may be more abstract than they would normally be for the closure's abstract type. We handled this by reabstracting value parameters to their usual substituted representation in the prolog, but neglected to do so for `inout` parameters too. In this case, we need to take the initial value, reabstract it to provide a local variable with the substituted representation, and then take the final value, reabstract it back, and write back to the original inout parameter on function exit. Fixes rdar://111563642.
|
@swift-ci Please test |
DougGregor
approved these changes
Jul 12, 2023
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Outdated suggestions cannot be applied.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.
Issue: rdar://111563642
Explanation: When a literal closure is passed into a generic context that takes an
(inout T) -> ()function argument, we emit the closure literal with the abstraction pattern expected by the callee, but whenTwas itself a function type, we failed to handle the abstraction difference of theinoutparameter locally, leading to compiler crashes. This patch fixes that by having the closure perform reabstraction on the initial value going in, storing it to a local buffer to represent the inout parameter locally, then reabstracting back to the expected representation and writing back to the original inout parameter.Scope of Issue: Fixes a regression.
Origination: Last year's thunk elimination optimizations.
Risk: Low. The fix is targeted to a relatively rare situation, which currently crashes the compiler, and should have minimal effect on other code.
Reviewed by: @atrick
Cherry picked from: #67239