Re: TR29124 C++ Special Maths - Make <math.h> pull functions into global namespace.

[Jonathan Wakely <jwakely at redhat dot com>]

On 21/01/16 19:07 -0500, Ed Smith-Rowland wrote:
> On 01/21/2016 07:29 AM, Jonathan Wakely wrote:
> > On 20/01/16 20:30 -0500, Ed Smith-Rowland wrote:
> > > Now that libstdc++ installs a proper math.h we can piggyback on 
> > > that to put in the last bit of TR29124.
> > >
> > > This patch adds the math special functions to 
> > > c_compatibility/math.h in the global namespace.
> > > I remove the XFAILs from the compile_2.cc tests.
> > >
> > > This converts 21 XFAILs into 21 PASSes.
> > >
> > > Tested on x86_64-linux.
> > >
> > > I understand if this is too late.
> > > I'll put it up on trunk and backport after stage 1 reopens.
> > >
> > > Meanwhile I'll commit this to the tr29124 branch.
> >
> > I'm inclined to say the change is OK for trunk now. We've added math.h
> > and we've added the special functions, it would be good if the two new
> > things work together.
> >
> > However ...
> >
> > > Index: include/c_compatibility/math.h
> > > ===================================================================
> > > --- include/c_compatibility/math.h    (revision 232610)
> > > +++ include/c_compatibility/math.h    (working copy)
> > > @@ -75,70 +75,71 @@
> > > #endif
> > >
> > > #if __STDCPP_WANT_MATH_SPEC_FUNCS__ == 1
> > > -using std::assoc_laguerref
> >
> > This doesn't look like the right patch, because these lines aren't in
> > the version on trunk.
> I must have given you a negative patch relative to the tr29124 
> branch...  Sigh.
> I'll make a new one.
> Those lines have to be *added*.
> Sorry.

No problem. It looks like it was a partial patch, adding semi-colons
to the using declarations, so you just need to diff against the
previous revision instead.