[Lldb-commits] [PATCH] D61451: Hide runtime support values such as clang's __vla_expr from frame variable

[Jim Ingham via lldb-commits]

If we did this wrong, then either stopping in a C++ method in a .mm file would suppress "this" (if we mistakenly used the ObjC Runtime's "RuntimeSupportValue") or stopping in an ObjC method in a .mm file would suppress "self and _cmd" (if we used the C++ Runtime's RuntimeSupportValue).  So we need a test that has a .mm file and stops in both a C++ method and an ObjC method and demonstrates that "frame var" shows "this" or "self & _cmd" respectively.

Jim


> On May 2, 2019, at 12:22 PM, Adrian Prantl via Phabricator <reviews at reviews.llvm.org> wrote:
> 
> aprantl added a comment.
> 
> In D61451#1488186 <https://reviews.llvm.org/D61451#1488186>, @jingham wrote:
> 
>> Could you add a comment (probably best in LanguageRuntime.h) saying what a RuntimeSupportValue is?
> 
> 
> Done.
> 
>> Also, so far as I can tell, this patch works because we use the CPP language runtime for C++ methods is an ObjC++ file, and the ObjCLanguageRuntime for ObjC methods in ditto.  That is surely the right thing to do, but it would be good to ensure that there's a test that asserts this behavior.  I'm sure there must be but didn't see on a quick scan.  Can you make sure there's such tests and if not add one?
> 
> I'm not sure what exactly you want me to test there. Can you elaborate?
> 
> 
> Repository:
>  rG LLVM Github Monorepo
> 
> CHANGES SINCE LAST ACTION
>  https://reviews.llvm.org/D61451/new/
> 
> https://reviews.llvm.org/D61451
> 
> 
>