Remove some usages of InOutType #17043
Conversation
|
@CodaFi I still have to update the AST printer, there's some stuff in code completion that's suspect as well. |
|
@CodaFi Also I'd like to get rid of |
slavapestov
force-pushed
the
small-progress-on-killing-inout-type
branch
from
54e3dcf to
44971fa
Compare
|
@swift-ci Please smoke test |
slavapestov
changed the title
|
@swift-ci Please smoke test |
NFC for now, but by not looking at the FunctionType of the AST Decl, we no longer pass an InOutType to type lowering here, which won't be supported soon.
… InOutType Most callers did not want the InOutType here, and checked the ParamDecl's flags instead.
There is still some unfortunate hackery around InOutType in SILGenProlog.cpp, but I don't want to clean it up yet.
slavapestov
force-pushed
the
small-progress-on-killing-inout-type
branch
from
44971fa to
1955778
Compare
| // FIXME: Should this check if the lowered SILType is address only | ||
| // instead? Otherwise optionals of archetypes etc will still have | ||
| // 'Unwrap' set to false. | ||
| bool Unwrap = VarInfo->Constant || SILTy.is<ArchetypeType>(); |
There was a problem hiding this comment.
@dcci @adrian-prantl It looks like this wasn't doing anything, so I removed it.
There was a problem hiding this comment.
Looks fine to me if it sticks together.
|
@swift-ci Please smoke test |
|
@swift-ci Please smoke test macOS |
|
@swift-ci test |
|
Build failed |
|
@swift-ci Please smoke test macOS |
|
Build failed |
|
@swift-ci Please test source compatibility |
1 similar comment
|
@swift-ci Please test source compatibility |
Most callers did not want the InOutType here, and checked
the ParamDecl's flags instead.