[mpich-discuss] Non-blocking collectives

[rreddypsc at gmail.com]

This is a very good example! Thanks for the clarification!

Thanks,

Raghu

--On Saturday, June 29, 2013 3:08 PM -0500 Rajeev Thakur 
<thakur at mcs.anl.gov> wrote:

> Yes.
>
> On Jun 29, 2013, at 3:05 PM, Jed Brown wrote:
>
>> Rajeev Thakur <thakur at mcs.anl.gov> writes:
>>
>>> It doesn't have to be a matching Wait. The implementation is required
>>> to make progress on other pending communication while it is blocked in
>>> a Wait.
>>
>> Or blocked in any of a number of other places, if non-blocking
>> collective requests have the same completion semantics as point-to-point
>> requests.  E.g., this shouldn't deadlock:
>>
>>  MPI_Iallreduce(...,COMM_WORLD,&req);
>>  if (!rank) MPI_Wait(&req);
>>  MPI_Allgather(...,COMM_WORLD);
>>  if (rank) MPI_Wait(&req);
>
> _______________________________________________
> discuss mailing list     discuss at mpich.org
> To manage subscription options or unsubscribe:
> https://lists.mpich.org/mailman/listinfo/discuss