On Mar 10, 2006, at 5:45 PM, Peeyush Jain wrote:
> I setup lam mpi -7.1.1 with option:
>
> $./configure --prefix=/home/peeyush/lammpi --with-shared --with-
> rsh="ssh -x"
> $make
> $make install
>
> it works properly fine when I use this on computing on one single
> file and on only localhost also. But when I try to run it on more
> than one .c files with a header file then it gives the following
> error:
>
> $mpirun -np 1 ./a
> ----------------------------------------------------------------------
> -------
> One of the processes started by mpirun has exited with a nonzero exit
> code. This typically indicates that the process finished in error.
> If your process did not finish in error, be sure to include a "return
> 0" or "exit(0)" in your C code before exiting the application.
>
> PID 4813 failed on node n0 (172.26.117.167) due to signal 11.
> ----------------------------------------------------------------------
> -------
I'd recommend looking at the LAM/MPI FAQ on debugging MPI
applications. There's quite a bit of useful information on there.
http://www.lam-mpi.org/faq/category6.php3
The problem you are running into is with your application, not LAM/
MPI. The error message above indicates that your application did
something that caused the process to segmentation fault.
If you followed the instructions for running your application under a
debugger, you will likely find that the segmentation fault occurs
inside a LAM/MPI function. This is because you are using an
uninitialized MPI Communicator comm for almost all the functions that
take a comm variable. You need to initialize the variable comm,
either assigning it the value of one of the predefined communicators
or creating a new communicator with MPI_COMM_DUP or the like.
Hope this helps,
Brian
--
Brian Barrett
LAM/MPI developer and all around nice guy
Have a LAM/MPI day: http://www.lam-mpi.org/
|
