Open MPI logo

Open MPI User's Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Open MPI User's mailing list

Subject: [OMPI users] problem with -npernode
From: David Turner (dpturner_at_[hidden])
Date: 2010-06-17 17:50:07


Hi,

Recently, Christopher Maestas reported a problem with -npernode in
Open MPI 1.4.2 ("running a ompi 1.4.2 job with -np versus -npernode").
I have also encountered this problem, with a simple "hello, world"
program:

% mpirun -np 16 ./a.out
  myrank, icount = 0 16
  myrank, icount = 2 16
  myrank, icount = 5 16
  myrank, icount = 7 16
  myrank, icount = 1 16
  myrank, icount = 4 16
  myrank, icount = 6 16
  myrank, icount = 3 16
  myrank, icount = 8 16
  myrank, icount = 9 16
  myrank, icount = 10 16
  myrank, icount = 12 16
  myrank, icount = 13 16
  myrank, icount = 15 16
  myrank, icount = 11 16
  myrank, icount = 14 16
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
 

% mpirun -np 16 -npernode 8 ./a.out
[c1146:15313] *** Process received signal ***
[c1146:15313] Signal: Segmentation fault (11)
[c1146:15313] Signal code: Address not mapped (1)
[c1146:15313] Failing at address: 0x50
[c1146:15313] *** End of error message ***
Segmentation fault
[c1138:26571] [[62315,0],1] routed:binomial: Connection to lifeline
[[62315,0],0] lost
 
          % module swap openmpi openmpi/1.4.1
 
          % mpirun -np 16 -npernode 8 ./a.out
  myrank, icount = 8 16
  myrank, icount = 13 16
  myrank, icount = 10 16
  myrank, icount = 11 16
  myrank, icount = 15 16
  myrank, icount = 14 16
  myrank, icount = 12 16
  myrank, icount = 5 16
  myrank, icount = 2 16
  myrank, icount = 3 16
  myrank, icount = 1 16
  myrank, icount = 0 16
  myrank, icount = 9 16
  myrank, icount = 6 16
  myrank, icount = 7 16
  myrank, icount = 4 16
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP
FORTRAN STOP

Compilers are PGI/10.5, OS is Scientific Linux 5.4, resource manager is
torque 2.4.5. Please let me know if you need more information. Thanks!

-- 
Best regards,
David Turner
User Services Group        email: dpturner_at_[hidden]
NERSC Division             phone: (510) 486-4027
Lawrence Berkeley Lab        fax: (510) 486-4316