Emptyness vs. virtualness

[Martin von Loewis]

> But that's what I would have expected.  The rule is that we first lay out
> the non-virtual things (C (not B), A2 (with A)) and then all the virtual
> bases (B).  By the time we get to B, it's A can't be at 0, so we get a new
> one.

In 2.4-III, there is a statement

# Any indirect primary base class E of the current class C, that has
# been chosen as the primary base class of some other base class
# (direct or indirect, virtual or non-virtual) of C, will be allocated
# as part of that other base class, and is not allocated here.

Copying Mark's example again

  struct A {};

  struct B : public A {
    virtual void f () {}
  };

  struct C : public virtual B {};

  struct A2 : public A {};

  struct D : public C, public A2 {
  };

I find that B is the primary base of C, right? Now, I put C at offset
0 of D, as the primary base of D. Since B is an "indirect primary
base" of D, it will be allocated as part of C.

So I would say that C, with B and A, go at offset 0. Then, A2 cannot
go to offset 0.

Regards,
Martin