[OMPI users] OpenMPI + GFortran compile issues on Mac OS X 10.4
Dear list, I'm having a terrible time trying to get OpenMPI running on a G5 with Mac OS X 10.4 and with Fortran support. The F77 example hello_f77.f compiles and runs as expected, but hello_f90.f90, when compiled with mpif90, gives me this mysterious error: Fatal Error: Reading module mpi at line 3562 column 2: Expected integer Here's what I did in detail. I downloaded gfortran 4.3 from http://hpc.sourceforge.net/ and that installed fine. Simple programs compile. I downloaded OpenMPI 1.2.2 from source tarball. Here's what I did for configure: ./configure --with-f77 --with-f90 F77=gfortran FC=gfortran No errors yet at configure. make, sudo make install also yield no errors. cd to examples, mpif77 hello_f77.f, ./a.out works mpif90 hello_f90.f90 gives me the error above: Fatal Error: Reading module mpi at line 3562 column 2: Expected integer mpif90 -showme outputs: gfortran -I/usr/local/include -I/usr/local/lib -Wl,-u,_munmap -Wl,- multiply_defined,suppress -L/usr/local/lib -lmpi_f90 -lmpi_f77 -lmpi - lopen-rte -lopen-pal Any ideas what could be wrong? I think I'm going to be seeing ./ configure script output in my dreams... Thank you, Aaron Thompson Vanderbilt University Dept. of Chemical Engineering
Re: [OMPI users] OpenMPI + GFortran compile issues on Mac OS X 10.4 -- solved
Dr. Simon, My gfortran version was more recent than the official one at the GNU website. I tried it with 4.3, and that gave me that error. However, I just tried it with the stable version 4.2, which is the latest at the GNU website, and it worked! I think what happened is that it compiled with CC pointing to Apple's installed cc version 4.0.1, but with F90 pointing gfortran 4.3 that I installed. This time I forced it to use GCC 4.2 the whole way through and it worked. Thank you for pointing me in the right direction. So for anyone else who may come across this, if you get an error like Fatal Error: Reading module mpi at line 3562 column 2: Expected integer it may mean that you've used a different version of cc than gfortran when configuring OpenMPI. The configure script ought to warn about this! Aaron Thompson Vanderbilt University aaron.p.thomp...@vanderbilt.edu On Jun 3, 2007, at 5:39 PM, Christian Simon wrote: Dear Aaron, Here's what I did in detail. I downloaded gfortran 4.3 from http://hpc.sourceforge.net/ and that installed fine. Simple programs compile. Is your gfortran older or more recent than the official one on http://gcc.gnu.org/fortran/ ? -- Dr. Christian SIMON, Maitre de Conferences Laboratoire LI2C-UMR7612Bat. F74, piece 757 Universite Pierre et Marie CurieTel:+33.1.44.27.32.65 Case 51 Fax:+33.1.44.27.32.28 4 Place Jussieu 75252 Paris Cedex 05 France/Europe ___ users mailing list us...@open-mpi.org http://www.open-mpi.org/mailman/listinfo.cgi/users
[OMPI users] Issues with DL POLY
Hello, Does anyone have experience using DL POLY with OpenMPI? I've gotten it to compile, but when I run a simulation using mpirun with two dual- processor machines, it runs a little *slower* than on one CPU on one machine! Yet the program is running two instances on each node. Any ideas? The test programs included with OpenMPI show that it is running correctly across multiple nodes. Sorry if this is a little off-topic, I wasn't able to find help on the official DL POLY mailing list. Thank you! Aaron Thompson Vanderbilt University aaron.p.thomp...@vanderbilt.edu
Re: [OMPI users] Issues with DL POLY
Very interesting, I certainly hope that my problem is this and not
some kind of error. I'll put the program on some more nodes and run
some tests and see what runs fastest.
My only experience so far with MPI is with LAMMPS, and the simulation
I ran had an almost linear speedup from 1 -> 10 machines (11 -> 1.2
hours), very satisfying!
Aaron Thompson
Vanderbilt University
aaron.p.thomp...@vanderbilt.edu
On Jun 7, 2007, at 8:44 PM, Brock Palen wrote:
We have a few users using DLPOLY with OMPI. Running just fine.
Watch out what kind of simulation you are doing like all MD
software, not all simulations are better in parallel. In some the
comunication overhead is much worse than running on just one cpu. I
see this all the time. You could try just 2 cpus, on one node some
times that is ok (memory access vs network access) But its not
uncommon.
Brock Palen
Center for Advanced Computing
bro...@umich.edu
(734)936-1985
On Jun 7, 2007, at 8:24 PM, Aaron Thompson wrote:
Hello,
Does anyone have experience using DL POLY with OpenMPI? I've gotten
it to compile, but when I run a simulation using mpirun with two
dual-
processor machines, it runs a little *slower* than on one CPU on one
machine! Yet the program is running two instances on each node. Any
ideas? The test programs included with OpenMPI show that it is
running correctly across multiple nodes.
Sorry if this is a little off-topic, I wasn't able to find help on
the official DL POLY mailing list.
Thank you!
Aaron Thompson
Vanderbilt University
aaron.p.thomp...@vanderbilt.edu
___
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users
___
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users
