Re: [c++] Keep tm, div_t, ldiv_t, lconv mangling on Solaris (PR libstdc++-v3/1773)

[Rainer Orth <ro at CeBiTec dot Uni-Bielefeld dot DE>]

Marc,

>> 	* config/sol2-cxx.c
>
> You are missing a closing parenthesis in there, on the line with
> DECL_NAMESPACE_STD_P. I hadn't dared modify DECL_CONTEXT(decl) because I

drats, I had fixed that in my tree, but forgot to refresh the patch.

> was afraid of side effects.
>
> For instance, in:
> namespace std { struct tm {}; }
> //void f(std::tm){}
> void g(tm){}
>
> uncommenting the middle line turns the error message from:
> e.cc:3:10: error: variable or field 'g' declared void
> e.cc:3:8: error: 'tm' was not declared in this scope
> e.cc:3:8: note: suggested alternative:
> e.cc:1:24: note:   'std::tm'
>
> to:
> e.cc:3:10: error: variable or field 'g' declared void
> e.cc:3:8: error: 'tm' was not declared in this scope
> e.cc:3:8: note: suggested alternative:
> e.cc:1:24: note:   'tm'
>
> It also breaks argument-dependent lookup:
> namespace std { struct tm {}; void h(tm){} }
> void g(std::tm x){ h(x); }
>
> On the other hand, there probably isn't much ADL going on with these 4
> types...
>
> It might be less invasive to have decl_mangling_context return
> global_namespace without actually modifying the expr?

Maybe.  Let's wait for Jason's comments.

> (as already mentioned, I don't know much about gcc internals, I am just
> commenting from my experience writing the previous draft patch)

Me neither, especially when it gets to the non-port-specific code.
Without your draft, I wouldn't probably have arrived at anything
remotely sensible.

Thanks.
        Rainer

-- 
-----------------------------------------------------------------------------
Rainer Orth, Center for Biotechnology, Bielefeld University