[Mpi3-rma] Proposal 1 discussion points for next telecon
wgropp at illinois.edu
Tue Oct 26 14:41:19 CDT 2010
Names are *always* discussed *only* after the semantics are settled.
On Oct 25, 2010, at 11:35 PM, Pavan Balaji wrote:
> On 10/25/2010 10:58 PM, Torsten Hoefler wrote:
>> Hi Pavan,
>>> This proposal doesn't have an atomic GET operation.
>>> MPI_ACCUMULATE with MPI_REPLACE is an atomic PUT.
>>> MPI_GET_ACCUMULATE with MPI_NO_OP does not work as an atomic GET
>>> as it
>>> does not take more than 1 count, or non-predefined datatypes.
>> Correct, we had count>1 and ddts in an earlier version and changed it
>> after heated discussions about buffering. I forgot to add it to the
>> discussion items. I have no huge problems to allow ddts and
>> however, I believe Brian and Keith were against it.
>> We should see if an advice to users/implementers that MPI_NO_OP is a
>> special case that doesn't require buffering and that all other
>> operations might be really really slow with large data.
>> Please consider this item 7 on the discussion list!
> Yes, I remember the discussion and it is a valid argument -- it was
> respect to buffering requirements because of retransmissions (for
> Note that I'm not proposing that MPI_GET_ACCUMULATE be made more
> generic. I'm just saying that we need a (possible separate) function
> do atomic GETs. However, if MPI_GET_ACCUMULATE is not as generic as
> it should be given a different name (though, if you want to discuss
> names later, that's fine).
> -- Pavan
> Pavan Balaji
> mpi3-rma mailing list
> mpi3-rma at lists.mpi-forum.org
Deputy Director for Research
Institute for Advanced Computing Applications and Technologies
Paul and Cynthia Saylor Professor of Computer Science
University of Illinois Urbana-Champaign
More information about the mpiwg-rma