emitting vtables (was rtti data layout notes)

[Nathan Sidwell]

Jim Dehnert wrote:
> 
> OK, I've added a comment about this to the ABI layout document --
> please see if it looks OK, everyone.  
Looks good to me.

> But beyond that, do we assume
> that all references to any function members of these classes are in the
> library, and therefore inherently consistent, or do we need to define
> the function member interface, too?
No, we should not constrain the implementation in this way. The
entry points are abi::__dynamic_cast and <catch matcher> (which
might be implementation defined anyway, when the unwind stuff has settled).

nathan

-- 
Dr Nathan Sidwell :: Computer Science Department :: Bristol University
Never hand someone a gun unless you are sure where they will point it
nathan at acm.org  http://www.cs.bris.ac.uk/~nathan/  nathan at cs.bris.ac.uk