PATCH: third stage of dfp-branch merge

[Ben Elliston]

> > 	* builtins.def: Add new builtins for 32, 64 and 128 bit variants
> > 	of inf, nan, finite, isinf and isnan.
> 
> These functions don't seem to be documented.
> 
> > 	(c_parser_asm_definition): Likewise.
> 
> This entry does not appear to relate to c_parser_asm_definition, but to 
> the comment above the next function.
> 
> > +    pedwarn ("floating constant exceeds range of %qT>", type);
> 
> Stray >.
> 
> > +               pedwarn ("ISO C90 does not support decimal floating point");
> 
> ISO C in any version, not just C90.

OK with these corrections?

Ben