[mvapich-discuss] one-sided passive communications

[sreeram potluri]

Jeff, thats true with MVAPICH2 as well. Thanks.

Sreeram Potluri

On Wed, Dec 5, 2012 at 1:57 PM, Jeff Hammond <jhammond at alcf.anl.gov> wrote:

> >> If helper threads are not used, what is needed to get progress in the
> >> lock/unlock calls? Is it necessary to enter in any specific MPI call in
> the
> >> target process?
> >
> > Any of the MPI Test/Probe/Wait calls on the target will ensure progress.
>
> To add to this, I ~think~ that any other MPI calls that internally
> call the progress engine will also drive progress, including blocking
> send/recv and collectives.  I could be wrong though; my understanding
> here is based upon MPICH, not MVAPICH2.
>
> Jeff
>
> --
> Jeff Hammond
> Argonne Leadership Computing Facility
> University of Chicago Computation Institute
> jhammond at alcf.anl.gov / (630) 252-5381
> http://www.linkedin.com/in/jeffhammond
> https://wiki.alcf.anl.gov/parts/index.php/User:Jhammond
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.cse.ohio-state.edu/pipermail/mvapich-discuss/attachments/20121205/1c144d8e/attachment.html