The attached is a more general test. Please note that these are very sensitive to compilation options with ifort, and probably other compilers as well. In many cases openmpi fails test 1 and perhaps some of the others. Openmpi is not alone; the version of mvapich on my cluster fails a simple "CNTRL/C" at the input leaving zombies!
On Sat, Feb 6, 2010 at 9:24 PM, Laurence Marks <l-ma...@northwestern.edu> wrote: > The following code reproduces the problem for mpif90/ifort > 11.1/openmpi-1.4.1. With an empty test.input (touch test.input) some > not reproducible number of zombies processes are created. > > include "mpif.h" > call MPI_INIT(ierr) > call MPI_COMM_RANK(MPI_COMM_WORLD, irank, ierr) > open (unit=10,file='test.input') > if(irank.lt.3)then > read(10,1,err=20)ii > else > read(10,1)ii > endif > 20 write(6,*)irank,ii > 1 format(i4) > call MPI_FINALIZE(ierr) > end > > N.B., if I deliberately create a format error for the read no zombies remain. > > -- > Laurence Marks > Department of Materials Science and Engineering > MSE Rm 2036 Cook Hall > 2220 N Campus Drive > Northwestern University > Evanston, IL 60208, USA > Tel: (847) 491-3996 Fax: (847) 491-7820 > email: L-marks at northwestern dot edu > Web: www.numis.northwestern.edu > Chair, Commission on Electron Crystallography of IUCR > www.numis.northwestern.edu/ > Electron crystallography is the branch of science that uses electron > scattering and imaging to study the structure of matter. > -- Laurence Marks Department of Materials Science and Engineering MSE Rm 2036 Cook Hall 2220 N Campus Drive Northwestern University Evanston, IL 60208, USA Tel: (847) 491-3996 Fax: (847) 491-7820 email: L-marks at northwestern dot edu Web: www.numis.northwestern.edu Chair, Commission on Electron Crystallography of IUCR www.numis.northwestern.edu/ Electron crystallography is the branch of science that uses electron scattering and imaging to study the structure of matter.
crash_test.f
Description: Binary data
