Re: [OMPI users] unresolvable R_X86_64_64 relocation against symbol `mpi_fortran_*

2011-10-03 Thread Dmitry N. Mikushin
Hi, Here's a reprocase, the same one as mentioned here: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=608901 marcusmae@loveland:~/Programming/mpitest$ cat mpitest.f90 program main include 'mpif.h' integer ierr call mpi_init(ierr) end marcusmae@loveland:~/Programming/mpitest$ mpif90 -g mpitest

Re: [OMPI users] [SOLVED] unresolvable R_X86_64_64 relocation against symbol `mpi_fortran_*

2011-10-03 Thread Dmitry N. Mikushin
Ok, here's the solution: remove --as-needed option out of compiler's internal linker invocation command line. Steps to do this: 1) Dump compiler specs: $ gcc -dumpspecs > specs 2) Open specs file for edit and remove --as-needed from the line *link: %{!r:--build-id} --no-add-needed --as-needed %{!

Re: [OMPI users] Proper way to stop MPI process

2011-10-03 Thread Jeff Squyres
You might want to double check this -- mpirun shouldn't be waiting on you hitting return. Check to make sure you don't just have line-buffered output in python, or somesuch. Or better yet, check from python that the PID has actually disappeared and don't rely on stdout, or something like that.

Re: [OMPI users] Segfault on any MPI communication on head node

2011-10-03 Thread Phillip Vassenkov
I went into the directory that I used to install 1.4.3, did the following: make clean ./configure --enable-debug make -j8 all install and it hangs at this when I try to run my code (I commented out all the host name stuff, so it's just MPI code now) [hostname:16574] [[17705,0],0] ORTE_ERROR_LO

Re: [OMPI users] [SOLVED] unresolvable R_X86_64_64 relocation against symbol `mpi_fortran_*

2011-10-03 Thread Jeff Squyres
Wow -- painful! Glad you figured it out; thanks for posting it back here to make it google-able. On Oct 3, 2011, at 9:21 AM, Dmitry N. Mikushin wrote: > Ok, here's the solution: remove --as-needed option out of compiler's > internal linker invocation command line. Steps to do this: > > 1) Dum

Re: [OMPI users] Segfault on any MPI communication on head node

2011-10-03 Thread Ralph Castain
That means you have mismatched installations around - one configured as debug, and one not. They have to match. Sent from my iPad On Oct 3, 2011, at 2:44 PM, Phillip Vassenkov wrote: > I went into the directory that I used to install 1.4.3, did the following: > make clean > ./configure --enab