Resolution of B-2 not reflected in B-6 (aka covariant returns)

[Martin von Loewis]

> Note that this rule does not provide what I originally assumed was
> meant by "multiple entries for covariant returns," as it makes use of
> the multiple entries already present in secondary vtables when it
> can.

That would have been my proposal as well.

> Am I missing anything?  Is this clear?  Other comments?

I like it. Regarding Jason's comment, the critical phrase is

# and the derived class overrides a virtual function in the primary
# base's primary vtable with a different return type

I think this is also clear, saying that you need another entry if the
return type differs, even if no adjustments will be made. I think this
is fine, given that it only requires a slot in the vtable - the
function could be the same one as in the base vtable part.

Martin