mboehme marked an inline comment as done.
mboehme added inline comments.
================
Comment at: clang/lib/Sema/SemaDecl.cpp:4633-4637
Sema::ParsedFreeStandingDeclSpec(Scope *S, AccessSpecifier AS, DeclSpec &DS,
RecordDecl *&AnonRecord) {
- return ParsedFreeStandingDeclSpec(S, AS, DS, MultiTemplateParamsArg(), false,
+ return ParsedFreeStandingDeclSpec(S, AS, DS, ParsedAttributesView::none(),
+ MultiTemplateParamsArg(), false,
AnonRecord);
----------------
aaron.ballman wrote:
> It's surprising that we don't need to update this interface as well; is that
> inconsistency desired?
This is OK because every caller of this overload happens to do
`ProhibitAttributes(DeclAttrs)`.
But now that you bring this up, I realize that this is risky. What if we add
another caller in the future that doesn’t do `ProhibitAttribubtes(DeclAttrs)`?
And even in the current situation, I think it makes sense to be explicit that
we’re passing an empty list of attributes.
So I’ve added the `DeclAttrs` parameter to this function too, and now pass
`ParsedAttributesView::none()` at the callsites.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D128499/new/
https://reviews.llvm.org/D128499
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
