Open MPI logo

Open MPI User's Mailing List Archives

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

From: Jeff Squyres \(jsquyres\) (jsquyres_at_[hidden])
Date: 2006-05-24 18:55:28


There is a known issue with OpenIB on PPC machines at the moment -- see:
 
    https://svn.open-mpi.org/trac/ompi/ticket/23
<https://svn.open-mpi.org/trac/ompi/ticket/23>
 
A temporary workaround is to either use the SRQ or disable eager RDMA.
See the bug for the details of both of these options.
 

________________________________

        From: users-bounces_at_[hidden]
[mailto:users-bounces_at_[hidden]] On Behalf Of Paul
        Sent: Wednesday, May 24, 2006 6:53 PM
        To: users_at_[hidden]
        Subject: [OMPI users] pallas assistance ?
        
        
        So I have 64bit ppc versions of openmpi, openib and the pallas
files (IMB_MP1 being the important one). ldd checks out okay and shows
nothing missing. However when I try to execute the pallas run it dies
like so:
        
        
        
        [root_at_thing ompi]# /opt/ompi/bin/mpirun -np 2 -machinefile
machine.list ./IMB-MPI1
        Signal:11 info.si_errno:0(Success) si_code:1(SEGV_MAPERR)
        Failing at addr:0x3000100a819d
        [0] func:/opt/ompi/lib/libopal.so.0 [0x80001c6e18]
        [1] func:[0x1ffffffdfd0]
        [2] func:/opt/ompi/lib/libmpi.so.0 [0x800006516c]
        [3] func:/opt/ompi/lib/libmpi.so.0 [0x80000652b4]
        [4] func:/opt/ompi/lib/openmpi/mca_btl_openib.so [0x800056f2f0]
        [5] func:/opt/ompi/lib/libmpi.so.0 [0x80000d0540]
        [6] func:/opt/ompi/lib/openmpi/mca_bml_r2.so [0x80005548a8]
        [7] func:/opt/ompi/lib/libmpi.so.0 [0x80000cfc8c]
        [8] func:/opt/ompi/lib/openmpi/mca_pml_ob1.so [0x8000533d9c]
        [9] func:/opt/ompi/lib/libmpi.so.0 [0x80000d9988]
        [10] func:/opt/ompi/lib/libmpi.so.0 [0x8000087a80]
        [11] func:/opt/ompi/lib/libmpi.so.0 [0x80000b09ac]
        [12] func:./IMB-MPI1 [0x10003328]
        [13] func:/lib64/tls/libc.so.6 [0x8064e9415c]
        [14] func:/lib64/tls/libc.so.6 [0x8064e942e4]
        *** End of error message ***
        
        are there any special things that need to be done with pallas,
open-ib, open-mpi ? Pallas compiled fine and linked okay with the needed
libraries. Currently machine.list is just localhost twice.