semantic of used_define cpp callback?

[Arnaud Charlet]

> I suppose the idea is to emit each used macro a single time only, the
> code seems to confirm this.

Makes sense indeed (and at least matches what the code does).

> Arnaud> so would the addition of a new callback (e.g.
> Arnaud>   void (*used) (cpp_reader *, unsigned int, cpp_hashnode *);
> Arnaud> ) be acceptable in principle? If so, I'll post a formal patch on
> Arnaud> gcc-patches@
> 
> Yes, this would be fine.

Thanks, I'll prepare such patch then.

Arno