[OMPI users] openmpi-1.0.2a12 on solaris opteron
Hello, I am attempting to run openmpi-1.0.2a12 on solaris opteron. I compile in 64 bit mode, with Studio11 compilers The configure and the make run fine. I compile my application with : mpicc -o myapp myapp.c 1/ I have the following hieroglyphs and error message when launching mpirun : icare> mpirun -np 4 -host icare ./myapp T0<ú²þÿÿ+½T0<ú²þÿÿ4½T0<ú²þÿÿ=½T0<ú²þÿÿF½ld.so.1: myapp: fatal: relocation error: file /users/roch/lib/openmpi-1.0.2a12/lib/libmca_common_sm.so.0: symbol nanosleep: referenced symbol not found ld.so.1: all2all: fatal: relocation error: file /users/roch/lib/openmpi-1.0.2a12/lib/libmca_common_sm.so.0: symbol nanosleep: referenced symbol not found ld.so.1: all2all: fatal: relocation error: file /users/roch/lib/openmpi-1.0.2a12/lib/libmca_common_sm.so.0: symbol nanosleep: referenced symbol not found ld.so.1: all2all: fatal: relocation error: file /users/roch/lib/openmpi-1.0.2a12/lib/libmca_common_sm.so.0: symbol nanosleep: referenced symbol not found The problem is fixed by compiling with -lrt option : mpicc -lrt -o myapp myapp.c 2/ I attempted to force interface exclusion without success : n14 > mpirun --mca btl_tcp_if_exclude bge1 -np 2 -host p14,p39 myapp 9ÿýD $ ½T Ö¡¸ý $ ½Process 0 is alive on n14 Process 1 is alive on n39 [n39:26824] *** An error occurred in MPI_Barrier [n39:26824] *** on communicator MPI_COMM_WORLD [n39:26824] *** MPI_ERR_INTERN: internal error [n39:26824] *** MPI_ERRORS_ARE_FATAL (goodbye) Best Regards, Francoise --
[OMPI users] XMPI ?
XMPI is a GUI debugger that works with LAM/MPI. Is there anything similar that works with OpenMPI? Michael
Re: [OMPI users] XMPI ?
On Mar 29, 2006, at 8:38 AM, Michael Kluskens wrote: XMPI is a GUI debugger that works with LAM/MPI. Is there anything similar that works with OpenMPI? Not quite in the same format as XMPI or as free as XMPI. There are a number of projects/products that use the profiling layer to get much of the same information, generally in a file that can be read after the fact. MPE is one example. I believe that some of the commercial debuggers also include support for looking at message transfer information. Unfortunately, we don't support the unexpected message queue debugging for TotalView at this point, although it is on our long term to-do list. Brian -- Brian Barrett Open MPI developer http://www.open-mpi.org/
[OMPI users] OMPI 1.0.1, CentOS 4.2 and gcc4
dIt seems like this should be a simple problem. I am trying to get OpenMPI to compile on a CentOS 4.2 (like Redhat EL 4.2) box. It has installed gcc 3.4, and gcc 4.0. I want to compile OMPI with gcc4, but I am getting this error. What am I doing wrong? [root@Cent01 openmpi-1.0.1]# CC=gcc4 CPP=gcc4 CXX=g++4 CXXCPP=g++4 F77=gfortran FC=gfortran ./configure checking for a BSD-compatible install... /usr/bin/install -c checking whether build environment is sane... yes checking for gawk... gawk checking whether make sets $(MAKE)... yes == Configuring Open MPI *** Checking versions checking Open MPI version... 1.0.1 checking Open MPI Subversion repository version... r8453 checking Open Run-Time Environment (ORTE) version... 1.0.1 checking ORTE Subversion repository version... r8453 checking Open Portable Access Layer (OPAL) version... 1.0.1 checking OPAL Subversion repository version... r8453 *** Initialization, setup configure: builddir: /root/Desktop/openmpi-1.0.1 configure: srcdir: /root/Desktop/openmpi-1.0.1 checking build system type... i686-pc-linux-gnu checking host system type... i686-pc-linux-gnu checking for prefix by checking for ompi_clean... no installing to directory "/usr/local" *** Configuration options checking Whether to run code coverage... no checking whether to debug memory usage... no checking whether to profile memory usage... no checking if want developer-level compiler pickyness... no checking if want developer-level debugging code... no checking if want Fortran 77 bindings... yes checking if want Fortran 90 bindings... yes checking whether to enable PMPI... yes checking if want C++ bindings... yes checking if want to enable weak symbol support... yes checking if want run-time MPI parameter checking... runtime checking if want to install OMPI header files... no checking if want pretty-print stacktrace... yes checking if want deprecated executable names... no checking if want MPI-2 one-sided empty shell functions... no checking max supported array dimension in F90 MPI bindings... 4 checking if pty support should be enabled... yes checking if user wants dlopen support... yes checking if heterogeneous support should be enabled... yes checking if want trace file debugging... no == Compiler and preprocessor tests *** C compiler and preprocessor checking for style of include used by make... GNU checking for gcc... gcc4 checking for C compiler default output file name... a.out checking whether the C compiler works... yes checking whether we are cross compiling... no checking for suffix of executables... checking for suffix of object files... o checking whether we are using the GNU C compiler... yes checking whether gcc4 accepts -g... yes checking for gcc4 option to accept ANSI C... none needed checking dependency style of gcc4... gcc3 checking whether gcc4 and cc understand -c and -o together... yes checking if compiler impersonates gcc... no checking if gcc4 supports -finline-functions... yes checking if gcc4 supports -fno-strict-aliasing... yes configure: WARNING: -fno-strict-aliasing has been added to CFLAGS checking for C optimization flags... -O3 -DNDEBUG -fno-strict-aliasing checking how to run the C preprocessor... gcc4 configure: error: C preprocessor "gcc4" fails sanity check See `config.log' for more details. [root@Cent01 openmpi-1.0.1]#
Re: [OMPI users] OMPI 1.0.1, CentOS 4.2 and gcc4
On Mar 29, 2006, at 11:35 AM, Adams Samuel D Contr AFRL/HEDR wrote: dIt seems like this should be a simple problem. I am trying to get OpenMPI to compile on a CentOS 4.2 (like Redhat EL 4.2) box. It has installed gcc 3.4, and gcc 4.0. I want to compile OMPI with gcc4, but I am getting this error. What am I doing wrong? checking how to run the C preprocessor... gcc4 configure: error: C preprocessor "gcc4" fails sanity check See `config.log' for more details. [root@Cent01 openmpi-1.0.1]# Could you include the config.log file -- it will have the information we need to help you. Thanks, Brian -- Brian Barrett Open MPI developer http://www.open-mpi.org/
[OMPI users] sed :36: unescaped newline in pattern substitution
I tried building openmpi on two different powerbooks, a titanium G4 (800mhz) and an aluminum G4 (1.6ghz) both running Mac OS X 10.4.5 and Xcode 2.2. The sed error ONLY occurs if I try and build the Fortran 90 bindings ( yes I do need them). I am using Absoft 8.0 with service pack 6. In the mailing list I saw that someone else solved this problem by rebooting the machine, so I tried it on both machines. No luck. I set F77="/Applications/Absoft/bin/f77 -f -N15" and FC="/Applications/Absoft/bin/f95 -YEXT_NAMES=LCS -YEXT_SFX=_" While I don't see how this could be the problem, if I don't include the flags with the name of the compiler, the FFLAGS get passed to both the f77 and the f95 compiler and the FCFLAGS are ignored. . config.log.gz Description: GNU Zip compressed data outfile.txt.gz Description: GNU Zip compressed data
Re: [OMPI users] openmpi-1.0.2a12 on solaris opteron
Doh! This is a problem that was fixed on the trunk long ago and it somehow never made it over to the v1.0 branch. Brian ported it over earlier today; it'll be in the 1.0.2a13 that should be on the web site in a few hours. Thanks for bringing it to our attention. > -Original Message- > From: users-boun...@open-mpi.org > [mailto:users-boun...@open-mpi.org] On Behalf Of Francoise Roch > Sent: Wednesday, March 29, 2006 7:08 AM > To: us...@open-mpi.org > Subject: [OMPI users] openmpi-1.0.2a12 on solaris opteron > > > Hello, > > I am attempting to run openmpi-1.0.2a12 on solaris opteron. > I compile in 64 bit mode, with Studio11 compilers > > The configure and the make run fine. > > I compile my application with : > mpicc -o myapp myapp.c > > 1/ I have the following hieroglyphs and error message when launching > mpirun : > > icare> mpirun -np 4 -host icare ./myapp > T0<ú²þÿÿ+½T0<ú²þÿÿ4½T0<ú²þÿÿ=½T0<ú²þÿÿF½ld.so.1: myapp: fatal: > relocation error: file > /users/roch/lib/openmpi-1.0.2a12/lib/libmca_common_sm.so.0: symbol > nanosleep: referenced symbol not found > ld.so.1: all2all: fatal: relocation error: file > /users/roch/lib/openmpi-1.0.2a12/lib/libmca_common_sm.so.0: symbol > nanosleep: referenced symbol not found > ld.so.1: all2all: fatal: relocation error: file > /users/roch/lib/openmpi-1.0.2a12/lib/libmca_common_sm.so.0: symbol > nanosleep: referenced symbol not found > ld.so.1: all2all: fatal: relocation error: file > /users/roch/lib/openmpi-1.0.2a12/lib/libmca_common_sm.so.0: symbol > nanosleep: referenced symbol not found > > The problem is fixed by compiling with -lrt option : > mpicc -lrt -o myapp myapp.c > > 2/ I attempted to force interface exclusion without success : > > n14 > mpirun --mca btl_tcp_if_exclude bge1 -np 2 -host p14,p39 myapp > 9ÿýD $ ½T Ö¡¸ý $ ½Process 0 is alive on n14 > Process 1 is alive on n39 > [n39:26824] *** An error occurred in MPI_Barrier > [n39:26824] *** on communicator MPI_COMM_WORLD > [n39:26824] *** MPI_ERR_INTERN: internal error > [n39:26824] *** MPI_ERRORS_ARE_FATAL (goodbye) > > > Best Regards, > Francoise > > -- > > ___ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users >
