When I pushed "Send" on this email I thought immediately: " ... hmm, Ralph or Jeff will say this is wrong ..."
Wow! Support to singletons! I haven't read this word since long forgotten readings in Set Theory. So, if you run a single process, you can do away with mpiexec, and pretend that the code were serial, right? Amazing. You guys think of all edge cases! Happy Holidays! Gus Correa [standing corrected] On Dec 21, 2011, at 4:44 PM, Ralph Castain wrote: > Not really - we support singletons, so that should work. The key is to have > LD_LIBRARY_PATH set correctly in the environment. > > On Dec 21, 2011, at 1:08 PM, Gustavo Correa wrote: > >> You probably need also to launch the program with mpiexec (mpiexec -np 4 >> ./hello_c), >> not just ./hello_c as your email suggests you did. >> >> On Dec 21, 2011, at 2:12 PM, Ralph Castain wrote: >> >>> Did you remember to set your LD_LIBRARY_PATH to include /opt/openmpi, per >>> your configure line? >>> >>> >>> On Dec 21, 2011, at 11:56 AM, amosl...@gmail.com wrote: >>> >>>> Dear OMPI Users, >>>> I have just read the messages from Martin Rushton and Jeff Squyres >>>> and have been having the same problem trying to get openmp-1.4.4 to work. >>>> My specs are below: >>>> Xeon(R) CPU 5335 2.00 GHz >>>> Linux SUSE 11.4 (x86_64) >>>> Linux 2.6.371-1.2 desktop x86_64 >>>> I go through the compilation process with the commands: >>>> ./configure --prefix=/opt/openmpi CC=icc CXX=icpc >>>> F77=ifort F90=ifort "FCFLAGS=-O3 -i8" "FFLAGS=-O3 -i8" 2>&1 | tee >>>> config.out >>>> make -j 4 all 2>&1 | tee make.out >>>> make install 2>&1 | tee install.out. >>>> The entire process seems to go properly but when I try to use an example >>>> it doesn't work properly. >>>> mpicc hello_c.c -o hello_c >>>> compiles properly. However, >>>> "./hello_c" gives an error message that it cannot >>>> find the file libmpi_so.0. There are at least 3 copies of the file >>>> present as found by the search command but none of these are found. I >>>> have checked the permissions and they seem to be OK so I am at the same >>>> point as Martin Rushton. I hope that somebody comes up with an anser soon. >>>> >>>> Amos Leffler >>>> _______________________________________________ >>>> 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 >> >> >> _______________________________________________ >> 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
