Not enough information. Do you have locks around the send and Irecv?
If you have two threads enter an MPI call concurrently you may get
deadlock. You must assure this never happens.
Dave.
John Korah wrote:
> Hi,
>
> I have a multithreaded application. I use seperate
> threads to do MPI_Send and MPI_Irecv. Before each
> MPI_Send() i use a sleep() function to suspend the
> thread (to simulate network latency). The problem is
> the threads are not moving beyond sleep(). When I set
> sleep(o) or sleep(t) where 0.999<t<0, it works fine...
>
>
> Thanks
> John
>
>
>
>
> __________________________________
> Do you Yahoo!?
> Yahoo! Small Business - Try our new resources site!
> http://smallbusiness.yahoo.com/resources/
> _______________________________________________
> This list is archived at http://www.lam-mpi.org/MailArchives/lam/
>
--
Dr. David Cronk, Ph.D. phone: (865) 974-3735
Research Leader fax: (865) 974-8296
Innovative Computing Lab http://www.cs.utk.edu/~cronk
University of Tennessee, Knoxville
|
