I was talking about functionality of MPI_Iprobe only.
Thanks :)
On Sun, Dec 12, 2010 at 1:18 AM, Eugene Loh <eugene....@oracle.com> wrote:
> Alaukik Aggarwal wrote:
>
>> Thanks for your reply. I used this to solve the problem.
>>
>> But I think there should be an in-built construct for this.
>>
>
> What would such a construct look like? If you need information from the
> remote processes, they need to send messages (in the two-sided model). If
> you want to time out after a while, you can have MPI_Iprobe() checks for
> in-coming messages and then give up after some period of time. I just don't
> know what you'd be looking for.
>
> If you have concrete ideas you would really want, you should address them to
> the MPI Forum, which is in charge of defining MPI calls.
>
>> On Sat, Dec 11, 2010 at 10:28 AM, Eugene Loh <eugene....@oracle.com>
>> wrote:
>>
>>>
>>> Alaukik Aggarwal wrote:
>>>
>>>>
>>>> I am using Open MPI 1.4.3.
>>>>
>>>> I have to perform a receive operation from processes that are sending
>>>> data. It might happen that some of the processes don't send data
>>>> (might have completed in-fact).
>>>>
>>>> So, how do I perform check on which processes to receive data from and
>>>> which processes to skip?
>>>>
>>>> [code]
>>>> if(id != master)
>>>> MPI::COMM_WORLD.Send(&dist, NUM_VERTEX, MPI_LONG, master, 1234);
>>>> if(id == master)
>>>> {
>>>> for(int eachId = 1; eachId<procs ; eachId++)
>>>> MPI::COMM_WORLD.Recv(&dist1, NUM_VERTEX, MPI_LONG, eachId, 1234);
>>>> }
>>>> [/code]
>>>>
>
> _______________________________________________
> users mailing list
> us...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/users
>
