[Mpi-forum] ABI

Jed Brown jedbrown at mcs.anl.gov
Thu Dec 5 10:43:20 CST 2013

"Supalov, Alexander" <alexander.supalov at intel.com> writes:

> I hope so. Just click on the "Attachments" tab and look into the DOC files there.

That just defines some terms and scope, but Analysis, Solution, and
Issues are TBD.  Then there are various emails, but they are mainly
high-level motivation.  The pointer vs. integer sizes thing came up in
one email, where Jeff Squyres said

> For an ABI, you need to standardize both the size and the value.  It
> may not be difficult to agree on the sizes: sizeof(void*)/64 bits
> because you can treat it as an int or a pointer.  But the value is  
> where the religious debates come in.  For example: MPI_COMM_WORLD --  
> it is "0" or is it a pointer resolved at compile/link time?

which I think is a red herring since the value can be a link-time
constant and an implementation is free to "optimize" internally.

Is there more technical content somewhere?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 835 bytes
Desc: not available
URL: <http://lists.mpi-forum.org/pipermail/mpi-forum/attachments/20131205/362d5055/attachment-0001.pgp>

More information about the mpi-forum mailing list