Re: [OMPI users] Help understanding build option and warning messages, OMPI 1.2.4

On Wed, 27 Feb 2008, David Gunter wrote: We are trying to build OMPI-1.2.4 for a BProc/Ethernet-based cluster. Here are the configure options: ./configure --prefix=${PREFIX} \ --libdir=${LIBDIR} \ --enable-shared \ --with-bproc \ --with-tm=/opt/PBS \ --with-io_romio_flags=--with-file-

[OMPI users] Error compiling open-mpi

Hi i am running Red hat linux in school I am trying to compile open mpi and it gives me this error: make[3]: Entering directory `/home/acct2/babinsk3/research/openmpi-1.2.5/Albert/ ompi/mpi/cxx' /bin/sh ../../../libtool --tag=CXX --mode=link g++ -O3 -DNDEBUG -m64 -finline -functions -pthread -e

Re: [OMPI users] Help understanding build option and warning messages, OMPI 1.2.4

Hello Jenny and David, On Wednesday 27 February 2008 17:42, David Gunter wrote: > We are trying to build OMPI-1.2.4 for a BProc/Ethernet-based cluster. > Here are the configure options: > > ./configure --prefix=${PREFIX} \ >--libdir=${LIBDIR} \ >--enable-shared \ >--with-bproc \ >--

Re: [OMPI users] Cannot build 1.2.5

Scott, I can replicate this on big red. Seems to be a libtool problem. I'll investigate... Thanks, Tim Teige, Scott W wrote: Hi all, Attempting a build of 1.2.5 on a ppc machine, particulars: uname -a Linux s10c2b2 2.6.5-7.286-pseries64-lustre-1.4.10.1 #2 SMP Tue Jun 26 11:36:04 EDT 200

[OMPI users] Cannot build 1.2.5

Hi all, Attempting a build of 1.2.5 on a ppc machine, particulars: uname -a Linux s10c2b2 2.6.5-7.286-pseries64-lustre-1.4.10.1 #2 SMP Tue Jun 26 11:36:04 EDT 2007 ppc64 ppc64 ppc64 GNU/Linux Error message (many times) ../../../opal/.libs/libopen-pal.a(dlopen.o)(.opd+0x0): In function `__a

Re: [OMPI users] OpenMPI 1.2.5 race condition / core dump with MPI_Reduce and MPI_Gather

Brian is completely right. Here is a more detailed description of this problem. Upon receiving a fragment from the BTL (lower layer) we try to match it with an MPI request. If the match fails, then we get a fragment from the free_list (via the blocking call to FREE_LIST_WAIT) and copy the

Re: [OMPI users] Help understanding build option and warning messages, OMPI 1.2.4

This error indicates that the open of a shared library failed. It is generated by the dynamic loader, which is used by Open MPI to load the components. The specific error you get is clear about this: one of the dependencies of the TM RAS is missing. If you log on one of the compute nodes an

[OMPI users] Help understanding build option and warning messages, OMPI 1.2.4

We are trying to build OMPI-1.2.4 for a BProc/Ethernet-based cluster. Here are the configure options: ./configure --prefix=${PREFIX} \ --libdir=${LIBDIR} \ --enable-shared \ --with-bproc \ --with-tm=/opt/PBS \ --with-io_romio_flags=--with-file-system=ufs+nfs \ --disable-pty_support

Re: [OMPI users] OpenMPI 1.2.5 race condition / core dump with MPI_Reduce and MPI_Gather

Bummer; ok. On Feb 27, 2008, at 11:01 AM, Brian W. Barrett wrote: I played with this to fix some things in ORTE at one point, and it's a very dangerous slope -- you're essentially guaranteeing you have a deadlock case. Now instead of running off the stack, you'll deadlock. The issue is th

Re: [OMPI users] OpenMPI 1.2.5 race condition / core dump with MPI_Reduce and MPI_Gather

I played with this to fix some things in ORTE at one point, and it's a very dangerous slope -- you're essentially guaranteeing you have a deadlock case. Now instead of running off the stack, you'll deadlock. The issue is that we call opal_progress to wait for something to happen deep in the bo

Re: [OMPI users] OpenMPI 1.2.5 configure bug for POWERPC64 target

On Feb 23, 2008, at 10:05 AM, Mathias PUETZ wrote: 1. Could you please fix the bug above in the configure script ? Thanks for the detailed ananlysis. Can you confirm that this patch works for you before I commit it: Index: config/ompi_config_asm.m4

Re: [OMPI users] OpenMPI 1.2.5 race condition / core dump with MPI_Reduce and MPI_Gather

Gleb / George -- Is there an easy way for us to put a cap on max recusion down in opal_progress? Just put in a counter in opal_progress() such that if it exceeds some max value, return success without doing anything (if opal_progress_event_flag indicates that nothing *needs* to be done)?