[cxx-abi-dev] non-public non-static members vs POD layout

[Mark Mitchell]

David Vandevoorde wrote:

>> Also since the ABI predates TC1, the inclusion of pointers to members 
>> on PODs should not apply. I suppose the spec would explicitly say 
>> "PODs as defined prior to TC1"
>>
>> Does anyone object to making that change?
> 
> I think so.  Hereunder is an e-mail to this list (from a few years ago) 
> that seems to indicate we already agreed to use the TC1 definition of PODs.

Well, that's embarassing; clearly my memory is pathetic.

So, yes, we agreed to use TC1 PODs, and thus pointers-to-members are in. 
  In that case, yes, I think we have to say how they are laid out -- but 
I think we already do.

For data members, we say that a pointer to data member is "represented 
as a ptrdiff_t."

And for pointers to function members; we say that they have "the size, 
data size, and alignment of a class containing these two members", after 
saying what members they are.  I think we could probably make that more 
explicit by saying that they layout is as for the following class:

   struct pointer_to_member {
     void (*ptr)();
     ptrdiff_t adj;
   };

That would eliminate ambiguities.  But, I think that's just an editorial 
change.  (Anyone object to that change?  Or think it's a good idea?)

Thoughts?

-- 
Mark Mitchell
CodeSourcery
mark at codesourcery.com
(650) 331-3385 x713