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 counts>1, 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 with respect to buffering requirements because of retransmissions (for reliability). 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 to do atomic GETs. However, if MPI_GET_ACCUMULATE is not as generic as GET, it should be given a different name (though, if you want to discuss names later, that's fine). -- Pavan -- Pavan Balaji http://www.mcs.anl.gov/~balaji