Hi,
The MPI implementation is openmpi 1.4.2_0+gcc43 installed with macports.
And it seems that I have it installed in two places
/usr/include/openmpi/ompi/mpi/cxx/ /opt/local/include/openmpi/openmpi/ompi/mpi/cxx/
The configuration is: # macbook-gcc-config # Whenever this version string changes, the application is configured # and rebuilt from scratch
CPP = cpp-mp-4.3 FPP = cpp-mp-4.3 CC = gcc-mp-4.3 CXX = g++-mp-4.3 F77 = gfortran-mp-4.3 F90 = gfortran-mp-4.3
#CPPFLAGS = -DMPICH_IGNORE_CXX_SEEK FPPFLAGS = -traditional CFLAGS = -g3 -fshow-column -mmacosx-version-min=10.5 -m128bit-long-double -std =gnu99 CXXFLAGS = -g3 -fshow-column -mmacosx-version-min=10.5 -m128bit-long-double F77FLAGS = -g3 -fshow-column -mmacosx-version-min=10.5 -m128bit-long-double -fcr ay-pointer F90FLAGS = -g3 -fshow-column -mmacosx-version-min=10.5 -m128bit-long-double -fcr ay-pointer
LDFLAGS = /System/Library/Frameworks/vecLib.framework/vecLib -L/opt/local/lib -lgfortran
C_LINE_DIRECTIVES = yes F_LINE_DIRECTIVES = yes
REAL16_KIND = 10
DEBUG = no CPP_DEBUG_FLAGS = -DCARPET_DEBUG FPP_DEBUG_FLAGS = -DCARPET_DEBUG C_DEBUG_FLAGS = -fbounds-check -ftrapv -fstack-protector-all CXX_DEBUG_FLAGS = -fbounds-check -ftrapv -fstack-protector-all F77_DEBUG_FLAGS = -fbounds-check -ftrapv -fstack-protector-all F90_DEBUG_FLAGS = -fbounds-check -ftrapv -fstack-protector-all
# Changing ANSI C semantics: OPTIMISE = yes C_OPTIMISE_FLAGS = -O2 CXX_OPTIMISE_FLAGS = -O2 F77_OPTIMISE_FLAGS = -O2 F90_OPTIMISE_FLAGS = -O2
PROFILE = no C_PROFILE_FLAGS = -pg CXX_PROFILE_FLAGS = -pg F77_PROFILE_FLAGS = -pg F90_PROFILE_FLAGS = -pg
BLAS_DIR = /System/Library/Frameworks/vecLib.framework BLAS_LIBS = gfortran
FFTW_DIR = /opt/local FFTW_LIBS = drfftw dfftw m
GSL_DIR = /opt/local
HDF5 = yes HDF5_DIR = /opt/local LIBSZ_DIR = /opt/local
LAPACK = yes LAPACK_DIR = /System/Library/Frameworks/vecLib.framework LAPACK_LIBS =
MPI = OpenMPI MPI_LIBS = mpi #mpi_cxx OPENMPI_DIR = /usr #/opt/local #OPENMPI_LIBS = /usr/lib
PTHREADS = yes
SSL_DIR = /opt/local
X_LIB_DIR = /usr/X11R6/lib
________________________________________ From: Erik Schnetter [schnetter@cct.lsu.edu] Sent: Wednesday, May 26, 2010 7:18 PM To: Babiuc-Hamilton, Maria Cc: Frank Loeffler; users@einsteintoolkit.org Subject: Re: [Users] compilation problem on 10.6.3 macbook pro
Maria
Can you post your current configuration options?
Also, which MPI implementation are you using? Can you tell us the name and version number?
-erik
On May 26, 2010, at 15:02 , Babiuc-Hamilton, Maria wrote:
Hello,
Yes, the "_ompi_" errors were because of that. I recompiled with OpenMPI, and those errors were resolved. However, I still have the second errors: Undefined symbols: "MPI::Comm::Set_errhandler(MPI::Errhandler const&)", referenced from: vtable for MPI::Intercomm in libthorn_CarpetSlab.a(mapping.cc.o) etc... "MPI::Win::Set_errhandler(MPI::Errhandler const&)", referenced from: vtable for MPI::Win in libthorn_CarpetSlab.a(mapping.cc.o) etc... ld: symbol(s) not found collect2: ld returned 1 exit status
Do you have any suggestion about this error?
Maria ________________________________________ From: Frank Loeffler [knarf@cct.lsu.edu] Sent: Wednesday, May 26, 2010 12:42 PM To: Babiuc-Hamilton, Maria Cc: users@einsteintoolkit.org Subject: Re: [Users] compilation problem on 10.6.3 macbook pro
On Wed, May 26, 2010 at 09:34:10AM -0400, Babiuc-Hamilton, Maria wrote:
I am trying to compile a thorn arrangement for the qc0- mclachlan.par parameter file, with the einsteintoolkit.th on a macbook pro 10.6.3 intel core i7. I have attached the configuration file I am using and the output of the compilation errors. The errors are of type: Undefined symbols: "_ompi_mpi_comm_null", referenced from: MPI::Cartcomm::Clone() const in libthorn_CarpetSlab.a (mapping.cc.o) etc.... "MPI::Win::Set_errhandler(MPI::Errhandler const&)", referenced from: vtable for MPI::Win in libthorn_CarpetSlab.a(mapping.cc.o)
ompi_mpi_comm_null is a symbol from openmpi. Your configuration includes MPICH however, but has openmpi commented out. I guess some files are still linked against openmpi. Please delete the whole configuration (under configs/) and start from scratch.
Frank Loeffler
Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users
-- Erik Schnetter schnetter@cct.lsu.edu http://www.cct.lsu.edu/~eschnett/