[patch] Merge dwarf4 branch

[Cary Coutant]

> Hmm.  So if the function signature changes, the type signature of the
> enclosing class doesn't.  That doesn't affect the layout of the class, but
> it might cause confusing overload resolution in the debugger.
>
> I'm not sure I agree that's preferable to always shallow hashing class
> AT_type.  Certainly it will limit the ODR checking that you can do.

Deep hashing the member functions isn't feasible, since the presence
or absence of a definition for that function would affect the hash.
We'd need instead something in between shallow and deep, where we just
hash the function signature. One way to do that would be to hash the
linkage name instead of the qualified name. Thoughts?

-cary