[Bug target/64368] [5 Regression] Several libstdc++ test failures on darwin after r218964.

[howarth at bromo dot med.uc.edu]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64368

--- Comment #5 from howarth at bromo dot med.uc.edu ---
(In reply to Francois-Xavier Coudert from comment #3)
> Appeared between 218883 (works) and 218985 (fails) [I'm glad my nightly
> tester is useful!]
> 
> Culprits are:
> 
> r218984 | redi | 2014-12-20 01:19:40 +0100 (Sat, 20 Dec 2014) | 4 lines
> 
> Fix duplicate symbol errors in Darwin bootstrap.
> 
>         * src/c++98/locale-inst.cc (__add_grouping, __pad, __int_to_char):
>         Only compile for old ABI.
> 
> r218964 | redi | 2014-12-19 19:16:39 +0100 (Fri, 19 Dec 2014) | 183 lines
> 
> New std::string implementation.

The r218964 commit can't be at fault as I have posted testing at r218877...

https://gcc.gnu.org/ml/gcc-testresults/2014-12/msg02450.html

and the libstdc++ results are pristine. The commit at r218984 seems bizarre as
it has no associated PR for the supposed bug.