Pointers-to-members

[Christophe de Dinechin]

Mark Mitchell wrote:
> 
> I think I submitted a "proof" of the non-negativity of
> pointers-to-members in all cases.  I believe that you can't cast a
> D::* to a B::* if B is a virtual base.

I believe in the proof. Yet, I'm a bit nervous, because the aCC compiler
currently had large delta between actual pointer to members and NULL pointer to
members. Just paranoia? Maybe. Or maybe there is a case we don't think of...
That's why I proposed 0x800...000 as the NULL value.


Regards
Christophe