Created attachment 406215 [details] test case Description of problem: The Fortran interface to the BLACS does not work correctly under OpenMPI Version-Release number of selected component (if applicable): blacs-openmpi-devel-1.1-35.fc12.x86_64 How reproducible: Steps to Reproduce: 1. Compile the attached code with mpif90 -o hello hello.f -lmpiblacsF77init -lmpiblacs 2. run with mpirun -np 1 hello 3. Actual results: [donald:7090] *** An error occurred in MPI_Comm_group [donald:7090] *** on communicator MPI_COMM_WORLD [donald:7090] *** MPI_ERR_COMM: invalid communicator [donald:7090] *** MPI_ERRORS_ARE_FATAL (your MPI job will now abort) -------------------------------------------------------------------------- mpirun has exited due to process rank 0 with PID 7090 on node donald exiting without calling "finalize". This may have caused other processes in the application to be terminated by signals sent by mpirun (as reported here). Expected results: Hello world Additional info: This is almost certainly a build problem due to the setting of TRANSCOMM = -DCSameF77 This setting is correct with MPICH, but wrong with OpenMPI. If I compile myself the BLACS with OpenMPI, and with TRANSCOMM = (i.e. empty) I get the expected "Hello world" string. If I compile my own BLACS with -DCSameF77 I get the same error as with the package.
(In reply to comment #0) > Created an attachment (id=406215) [details] > test case > > Description of problem: > The Fortran interface to the BLACS does not work correctly under OpenMPI > > Version-Release number of selected component (if applicable): > blacs-openmpi-devel-1.1-35.fc12.x86_64 > > How reproducible: > > > Steps to Reproduce: > 1. Compile the attached code with > mpif90 -o hello hello.f -lmpiblacsF77init -lmpiblacs > 2. run with > mpirun -np 1 hello > Forgot to add setp 0: module load openmpi-x86_64 (I don't thinik the bug depends on i386 vs x86_64, but this is what I tested)
Good catch! The actual issue was that TRANSCOMM was being set to -DUseMpich during the openmpi compile, but the fix is the same. Building the fix now.
blacs-1.1-39.fc12 has been submitted as an update for Fedora 12. http://admin.fedoraproject.org/updates/blacs-1.1-39.fc12
blacs-1.1-39.fc13 has been submitted as an update for Fedora 13. http://admin.fedoraproject.org/updates/blacs-1.1-39.fc13
blacs-1.1-39.fc11 has been submitted as an update for Fedora 11. http://admin.fedoraproject.org/updates/blacs-1.1-39.fc11
blacs-1.1-39.fc13 has been pushed to the Fedora 13 testing repository. If problems still persist, please make note of it in this bug report. If you want to test the update, you can install it with su -c 'yum --enablerepo=updates-testing update blacs'. You can provide feedback for this update here: http://admin.fedoraproject.org/updates/blacs-1.1-39.fc13
blacs-1.1-39.fc11 has been pushed to the Fedora 11 testing repository. If problems still persist, please make note of it in this bug report. If you want to test the update, you can install it with su -c 'yum --enablerepo=updates-testing update blacs'. You can provide feedback for this update here: http://admin.fedoraproject.org/updates/blacs-1.1-39.fc11
blacs-1.1-39.fc12 has been pushed to the Fedora 12 testing repository. If problems still persist, please make note of it in this bug report. If you want to test the update, you can install it with su -c 'yum --enablerepo=updates-testing update blacs'. You can provide feedback for this update here: http://admin.fedoraproject.org/updates/blacs-1.1-39.fc12
blacs-1.1-39.fc12 has been pushed to the Fedora 12 stable repository. If problems still persist, please make note of it in this bug report.
blacs-1.1-39.fc11 has been pushed to the Fedora 11 stable repository. If problems still persist, please make note of it in this bug report.
blacs-1.1-39.fc13 has been pushed to the Fedora 13 stable repository. If problems still persist, please make note of it in this bug report.