vtable format

[Jim Dehnert]

dfp at sco.com wrote:
> 
> The only potential issue that I see is that the R_IA_64_IPLT[LM]SB
> relocations have no well-defined addend.  This isn't an issue with
> executable and shared objects since these always occur in plain
> "REL" not "RELA" containers.  But, most relocations for IA64 object
> files use RELA containers.  I suppose you could use a separate
> relocation section just for the vtables instead.  This would make
> it even easier to have ld just pass them on to the dynamic linker.
> 
> Dave Prosser dfp at sco.com (908)790-2358 Rm A332, SCO, Murray Hill, NJ

I think that even this is not a problem.  The current definition
says that the addend from a RELA relocation is added to the referenced
function address, and that seems adequate.

(It does strike me upon rereading our definition that it is never
explicitly stated where the "address of the referenced function" comes
from.  We should say that this relocation uses an S (REL) or S+A (RELA)
relocation calculation, and that the gp value to be used is that for
the DSO where S is defined.  Mats and Sunil?)

Jim

-		Jim Dehnert  x3-4272