since 1.7.x the MPI_THREAD_SERIALIZED is available by default, anf for
MPI_THREAD_MULTIPLE you have to set the --with-verbs flag (1.6.x and older:
--with-openib) at configure time.
Note that activating MPI_THREAD_MULTIPLE silently disable InfiniBand, so your
program may get very bad performance without being verbose.
Is there no bug in MPI_THREAD_MULTIPLE implementation in 1.7.2 and 1.7.3? My
test program just hang now
On 10/23/13 19:47, Jeff Hammond wrote:
> On Wed, Oct 23, 2013 at 12:02 PM, Barrett, Brian W <bwbarre_at_[hidden]> wrote:
>> On 10/22/13 10:23 AM, "Jai Dayal" <dayalsoap_at_[hidden]> wrote:
>> I'm asking because I'm using an open_mpi build ontop of infiniband, and the
>> maximum thread mode is MPI_THREAD_SINGLE.
>> That doesn't seem right; which version of Open MPI are you using?
> The last time I looked at this, OpenMPI only supported
> MPI_THREAD_SINGLE by default and if you ask for any higher thread
> level, you get MPI_THREAD_MULTIPLE, which requires a configure-time
> Maybe something has changed dramatically more recently than I tested.
> Squyres told me some thread-oriented refactoring was going on. All of
> this was over a year ago so it is entirely reasonable for me to be
> wrong about all of this.
Dipl.-Inform. Paul Kapinos - High Performance Computing,
RWTH Aachen University, Center for Computing and Communication
Seffenter Weg 23, D 52074 Aachen (Germany)
Tel: +49 241/80-24915