Steven Schäfer

There's two more comments that I couldn't add because they were from earlier commits: --- I don't think the code for type parameters in `coerceInlineClasses` is entirely correct. The `unsafeCoerce`...

I completely missed that. I've changed the code to always compare boxed class types and added a test for boxed against unboxed primitive class comparisons. In effect, this is implementing...

How does this feature interact with special bridge methods for builtins with different binary names on the JVM? If I understand the proposal correctly then users will have to add...

I think there are two ways to deal with renamed declarations from the Kotlin/JVM standard library. In order to allow *any* signatures on overrides we should add an additional override...

@elizarov Maybe I'm reading the specification incorrectly, because from your response it looks like the intention is that declarations with `BinarySignatureNames` can override standard library methods with different (binary) names?...

I'm responsible for the `jvm-abi-gen` changes. While testing this internally we ran into some bugs in `kotlinx-metadata` (which `jvm-abi-gen` bundles). The fixes (https://github.com/JetBrains/kotlin/pull/4614, https://github.com/JetBrains/kotlin/pull/4617) for these issues didn't quite make...

Yes, all of the fixes that are relevant for the compiler made it into the 1.6.10 release and we have been using `jvm-abi-gen` internally with these patches applied for about...

This is one of the things that we are changing for Autosubst 2 (very preliminary report at: https://www.ps.uni-saarland.de/extras/lfmtp17/). In Autosubst 1, I think that we are using Ids_outer somewhere to...