construction vtables broken

[Mark Mitchell]

>>>>> "Jason" == Jason Merrill <jason at redhat.com> writes:

    Jason> Really?  It seems very clear to me.  Your testcase is

Yes, it's a nice way of thinking about it.

But, I think the other nice thing about Nathan's observation is that
it simplifies the whole criterion -- we had a special case for virtual
functions overridden along virtual paths, and now there's an easier
way to describe the whole process, without even mentioning the notion
of a virtual path.  

The criteria is now just "has a virtual base, or is a (morally)
virtual base" which is a lot easier to understand!

--
Mark Mitchell                   mark at codesourcery.com
CodeSourcery, LLC               http://www.codesourcery.com