[MPICH] Problem with MPICH2 and Opt++
Rajeev Thakur
thakur at mcs.anl.gov
Mon Apr 2 11:07:30 CDT 2007
If the MPICH2 examples and tests run, it means MPICH2 has been installed
correctly. Try running the cpi example from the examples directory (mpiexec
-n 4 cpi). Then run "make testing" in the top-level MPICH2 directory, which
will run the entire test suite in test/mpi (could take an hour).
Rajeev
> -----Original Message-----
> From: owner-mpich-discuss at mcs.anl.gov
> [mailto:owner-mpich-discuss at mcs.anl.gov] On Behalf Of Brett Gordon
> Sent: Monday, April 02, 2007 10:25 AM
> To: mpich-discuss at mcs.anl.gov
> Subject: [MPICH] Problem with MPICH2 and Opt++
>
> Hello,
>
> I am trying to get OPT++, a numerical library, working with MPICH2 but
> I'm running into some problems. I can't figure out if the problem is
> with OPT++, perhaps in terms of their parallel implementations, or
> with the way I have MPICH2 configured.
>
> When running the 'make checks' on OPT++, I get several fails, one of
> which looks like this:
>
> /bin/sh: line 1: 10141 Segmentation fault ${dir}$tst
> FAIL: tstfdnewtpds
> Signal:11 info.si_errno:0(Success) si_code:1(SEGV_MAPERR)
> Failing at addr:0x44000098
> [0] func:/usr/local/lib/libopal.so.0 [0x2aaaab440bb8]
> [1] func:/lib64/tls/libpthread.so.0 [0x2aaaab9f62c0]
> [2] func:/usr/local/lib/libmpi.so.0(PMPI_Comm_rank+0x4a)
> [0x2aaaab18cd1a]
> [3] func:./tstnewtpds(main+0x4f) [0x41ac4f]
> [4] func:/lib64/tls/libc.so.6(__libc_start_main+0xda) [0x2aaaabb1b5aa]
> [5] func:./tstnewtpds(__gxx_personality_v0+0x7a) [0x41aaea]
>
>
> I know that libopal is not the source of the error. I have recompiled
> the program and inserted a number of print statements, which confirm
> that the error occurs at [2]. The code around the MPI_Comm_rank
> statement is simple enough:
>
> int main (int argc, char* argv[]){
> int me = -1;
> MPI_Init(&argc, &argv);
> MPI_Comm_rank(MPI_COMM_WORLD, &me);
>
> .....
> }
>
> This looks standard way, so I'm not sure where to proceed. I have no
> experience with MPI-related programming. Does anyone have any
> suggestions? Does this look like a problem with my configuration of
> MPI or is it within OPT++? If more information is needed, please let
> me know.
>
> Running on openSUSE 10.0
> Linux myhost 2.6.13-15.15-smp #1 SMP Mon Feb 26 14:11:33 UTC 2007
> x86_64 x86_64 x86_64 GNU/Linux
>
> Thanks,
> Brett
>
>
More information about the mpich-discuss
mailing list