26_numerics/fabs_inline.cc

[Gabriel Dos Reis]

Andreas Jaeger <aj@suse.de> writes:

| Mark Mitchell <mark@codesourcery.com> writes:
| 
| > This appears to be a brand-new test which fails due to the changes to
| > the handling of builtins.  The compiler issues a (seemingly correct)
| > error on the using directive that puts multiple declarations of `fabs'
| > with the same type in the same scope.  I think the
| >  
| >   #define fabs __glibcpp_fabs
| > 
| > stuff is necessary in std_cmath.h too, with the changes that we just
| > made to the builtins.
| > 
| > This is easy to do, but I don't have time before I turn into a
| > pumpkin.  Would one of you take care of it?  Or, if you like, just
| > XFAIL the test on the branch?  We're just banging our heads on the
| > same issue over and over here with the global/std namespace clashes.
| 
| If it fails, it's a regression from 2.95,

Sigh.  You're right.  Will just face the problem.

-- Gaby
CodeSourcery, LLC                       http://www.codesourcery.com