implemented with Builtin.Word
The definition Swift.Int is becoming heavily platform-dependent, please
avoid using it in SIL and IR tests unless Swift.Int is being tested and
can't be replaced with a fixed-width type (e.g., Int32).
Swift SVN r24720
There's no reason this test shouldn't already be working fine, and indeed it is.
We do have a bit of a problem with the emitted SIL, though: we get two
AST declarations of "the same" function. That's rdar://problem/17632252.
Swift SVN r22979
Our mangling grammar says that entities should always be mangled with a context, so in cases where we weren't (Clang module contexts and top-level code), do so. Additionally, for ObjC protocols, mangle them into the special "So" context so we can catch global ObjC runtime namespace collisions with them. Fixes <rdar://problem/14221244>.
Swift SVN r5851
Map BOOL parameters, return types, and properties to swift.Bool in the Clang importer like we map NSString to swift.String. Generalize SIL's handling of type bridging to accommodate Bool bridging in addition to String bridging. Add convertObjCBoolToBool and v.v. entry points for the compiler to use to insert bridging conversions between the two types. Implements <rdar://problem/14271667>.
Swift SVN r5819
We had an inconsistency where the synthesized Swift getter/setters were bridged but the VarDecl wasn't, causing assertion failures in SILGen.
Swift SVN r5366
If -nsstring-is-string is enabled, lower Strings in cc(c) and cc(objc) function types to NSString, and when calling them, insert calls to StringToNSString/NSStringToString to perform the bridging conversion.
This isn't quite ready for prime-time yet, because we still need to emit the inverse bridging for ObjC method thunks, and I haven't tested the IRGen end of things yet.
Swift SVN r5355
When we intend an ObjC dispatch, indicate so in the SILConstant. This lets us distinguish Swift from ObjC dispatch for Swift classes with ObjC superclasses.
Swift SVN r5207
Port IRGen's calculation of consumed arguments and return value semantics to SILGen, and use it to handle the ownership semantics of calls. Refactor the handling of properties and other clients of emitApply so they can properly hand ownership semantics down to it.
This should let all the moribund cleanup management code in IRGen die. Unfortunately Scope appears to be tied into scoped calculated metadata caching so it's not quite ready to die.
Swift SVN r4834
