Re: [OMPI users] Python Binding problem with openmpi-1.3.2

2009-07-13 Thread Neil Ludban
> Date: Mon, 13 Jul 2009 14:27:13 -0400
> From: Rainer Keller 
> Subject: Re: [OMPI users] Python Binding problem with openmpi-1.3.2
> To: us...@open-mpi.org
> Cc: "Evans, Thomas M." 
> Message-ID: <200907131427.13656.kel...@ornl.gov>
> Content-Type: Text/Plain; charset=iso-8859-1
> 
> Hi Tom,
> as discussed, the following threads may help You?
> 
> http://www.open-mpi.org/community/lists/users/2009/02/8158.php
> http://www.open-mpi.org/faq/?category=running#loading-libmpi-dynamically

We ran into this a few months ago with the bcMPI interpreter bindings
and I forgot to post our code for possible inclusion in the FAQ.
(Having never used dlopen before, it took some trial and error to get
it to work right.)

The C code below was put in a separate extension *not* linked against
MPI and called before any MPI extension functions.  Note we didn't
bother with a matching dlclose since the program is expected to exit,
this hasn't been a problem so far.

#include 

   void *handle;
   handle = dlopen("libmpi.so", RTLD_LAZY|RTLD_GLOBAL);
   if (handle == NULL) {
  fprintf(stderr, "%s\n", dlerror());
   }


Re: [OMPI users] Receiving MPI messages of unknown size

2009-06-04 Thread Neil Ludban
> Date: Thu, 4 Jun 2009 11:14:16 +1000
> From: Lars Andersson 
> Subject: [OMPI users] Receiving MPI messages of unknown size
> To: us...@open-mpi.org
> 
> When using blocking message passing, I have simply solved the problem
> by first sending a small, fixed size header containing the size of
> rest of the data, sent in the following mpi message. When using
> non-blocking message passing, this doesn't seem to be such a good
> idea, since we cant post the main data transfer until we have received
> the message header... It seems to take away most of the advantages on
> non-blocking io in the first place.

If enough messages are small enough, a medium sized message with
fixed header and variable data eliminates most of the second message
overhead.