Hi Jorgen,
The config.h file looks okay. With mpif90 you do not need the -lmpi. Change
gcc to mpicc. You can also take out the -tpp7. Check that all the paths are
correct. That mpif90 -show points to the compiler you expect. Look at what
'which mpif90' gives you.
Also make sure you do a full make clean before compiling again.
All the best
Ross
> -----Original Message-----
> From: Jorgen Simonsen [mailto:jorgen589.gmail.com]
> Sent: Tuesday, August 10, 2010 9:37 AM
> To: AMBER Mailing List
> Subject: Re: [AMBER] error when compiling pmemd
>
> here is the config.h file
>
> MATH_DEFINES =
> MATH_LIBS =
> IFORT_RPATH =
> /opt/intel/Compiler/11.0/083/idb/lib/intel64:/opt/intel/Compiler/11.0/0
> 83/lib/intel64:/opt/openmpi/1.3.3-
> 1.el5.fys.ifort.11.0/lib64:/usr/local/lib
> MATH_DEFINES = -DMKL
> MATH_LIBS = /opt/intel/mkl/10.1.3.027/lib/em64t/libmkl_em64t.a-
> L/opt/intel/mkl/
> 10.1.3.027/lib/em64t/ -lguide -lpthread
> FFT_DEFINES = -DPUBFFT
> FFT_INCLUDE =
> FFT_LIBS =
> NETCDF_HOME =
> NETCDF_DEFINES =
> NETCDF_MOD =
> NETCDF_LIBS =
> MPI_HOME = /opt/openmpi/1.3.3-1.el5.fys.ifort.11.0
> MPI_DEFINES = -DMPI
> MPI_INCLUDE = -I$(MPI_HOME)/include
> MPI_LIBDIR = $(MPI_HOME)/lib64
> MPI_LIBS = -L$(MPI_LIBDIR)
> DIRFRC_DEFINES = -DDIRFRC_EFS -DDIRFRC_COMTRANS -DDIRFRC_NOVEC
> CPP = /lib/cpp
> CPPFLAGS = -traditional -P
> F90_DEFINES = -DFFTLOADBAL_2PROC
> F90 = mpif90
> MODULE_SUFFIX = mod
> F90FLAGS = -c -auto
> F90_OPT_DBG = -g -traceback
> F90_OPT_LO = -tpp7 -O0
> F90_OPT_MED = -tpp7 -O2
> F90_OPT_HI = -tpp7 -xP -ip -O3
> F90_OPT_DFLT = $(F90_OPT_HI)
>
> CC = gcc
> CFLAGS =
>
> LOAD = ifort
> LOADFLAGS =
> LOADLIBS = -limf -lsvml -Wl,-rpath=$(IFORT_RPATH)
>
> I have tried with and without -lmpi in the mpilib parameter but gave
> same
> error.
>
>
>
>
>
> On Tue, Aug 10, 2010 at 3:05 PM, Jason Swails
> <jason.swails.gmail.com>wrote:
>
> > Can you post your config.h file here?
> >
> > On Tue, Aug 10, 2010 at 2:55 AM, Jorgen Simonsen <jorgen589.gmail.com
> > >wrote:
> >
> > > Hi
> > >
> > > I remove this from the -lpiif from the MPI_LIBS in the config.h
> file but
> > > get an error on undefined mpi links
> > >
> > > mdin_ctrl_dat.o: In function
> `mdin_ctrl_dat_mod_mp_bcast_mdin_ctrl_dat_':
> > > mdin_ctrl_dat.f90:(.text+0x21): undefined reference to `mpi_bcast_'
> > > mdin_ctrl_dat.f90:(.text+0x45): undefined reference to `mpi_bcast_'
> > > mdin_ewald_dat.o: In function
> > > `mdin_ewald_dat_mod_mp_bcast_mdin_ewald_dat_':
> > > mdin_ewald_dat.f90:(.text+0x281): undefined reference to
> `mpi_bcast_'
> > > mdin_ewald_dat.f90:(.text+0x2a5): undefined reference to
> `mpi_bcast_'
> > > prmtop_dat.o: In function
> `prmtop_dat_mod_mp_bcast_amber_prmtop_dat_':
> > > prmtop_dat.f90:(.text+0xa64): undefined reference to `mpi_bcast_'
> > > prmtop_dat.o:prmtop_dat.f90:(.text+0xaad): more undefined
> references to
> > > `mpi_bcast_' follow
> > > dynamics_dat.o: In function
> `dynamics_dat_mod_mp_destroy_communicators_':
> > > dynamics_dat.f90:(.text+0x1b0): undefined reference to
> `mpi_comm_free_'
> > > dynamics_dat.f90:(.text+0x1fe): undefined reference to
> `mpi_group_free_'
> > > dynamics_dat.o: In function
> `dynamics_dat_mod_mp_create_communicators_':
> > > dynamics_dat.f90:(.text+0x3c7): undefined reference to
> `mpi_group_incl_'
> > > dynamics_dat.f90:(.text+0x3f1): undefined reference to
> `mpi_comm_create_'
> > > parallel.o: In function `parallel_mod_mp_pvt_mpi_gathervec_':
> > > parallel.f90:(.text+0x5ce4): undefined reference to `mpi_gatherv_'
> > > parallel.o: In function `parallel_mod_mp_mpi_gathervec_':
> > > parallel.f90:(.text+0x5f5d): undefined reference to `mpi_gatherv_'
> > > parallel.o: In function `parallel_mod_mp_pvt_mpi_allgathervec_':
> > > etc etc
> > >
> > > On Mon, Aug 9, 2010 at 9:57 PM, Jason Swails
> <jason.swails.gmail.com>
> > > wrote:
> > >
> > > > On Mon, Aug 9, 2010 at 12:20 PM, Jorgen Simonsen
> <jorgen589.gmail.com
> > > > >wrote:
> > > >
> > > > > Hi
> > > > >
> > > > > I fix it it was something with the path to mpi libraries but I
> get
> > > > another
> > > > > error later on in the compilation
> > > > >
> > > > > ld: cannot find -lmpiif
> > > > >
> > > >
> > > > If you are using mpif90 as the compiler and linker, remove all
> mpi
> > > > libraries
> > > > from MPI_LIBRARY (or something like that) in the config.h file,
> since
> > > > mpif90
> > > > implicitly takes care of those.
> > > >
> > > > Good luck!
> > > > Jason
> > > >
> > > >
> > > > >
> > > > > during this I get the warning
> > > > >
> > > > > ifort: command line remark #10148: option '-tp' not supported
> > > > >
> > > > >
> > > > > can I ignore this one?
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > On Mon, Aug 9, 2010 at 3:06 PM, Ross Walker
> <ross.rosswalker.co.uk>
> > > > wrote:
> > > > >
> > > > > > Hi Jorgen,
> > > > > >
> > > > > > > the error message is
> > > > > > >
> > > > > > > module gbl_constants_mod
> > > > > > >
> > > > > > > implicit none
> > > > > > >
> > > > > > > ! Global constants:
> > > > > > ...
> > > > > > > end module gbl_constants_mod
> > > > > > > make[1]: *** [gbl_constants.o] Error 1
> > > > > > > make[1]: Leaving directory
> `/home/Programs/amber10/src/pmemd/src'
> > > > > > > make: *** [install] Error 2
> > > > > >
> > > > > > This is an extremely weird error message since it is not
> actually
> > an
> > > > > error
> > > > > > message it is simply echoing the file to the screen. I
> suspect
> > there
> > > > are
> > > > > > issues with your precompiler. Can you please post your
> config.h
> > file
> > > > > along
> > > > > > with details of your system, gcc, cpp versions etc. Also
> please
> > > confirm
> > > > > > that
> > > > > > you can compile the rest of AMBER fine using your current
> login and
> > > > > > environment setup - try it again just to be sure.
> > > > > >
> > > > > > Also did you apply all of the bugfixes?
> > > > > >
> > > > > > All the best
> > > > > > Ross
> > > > > >
> > > > > > /\
> > > > > > \/
> > > > > > |\oss Walker
> > > > > >
> > > > > > ---------------------------------------------------------
> > > > > > | Assistant Research Professor |
> > > > > > | San Diego Supercomputer Center |
> > > > > > | Adjunct Assistant Professor |
> > > > > > | Dept. of Chemistry and Biochemistry |
> > > > > > | University of California San Diego |
> > > > > > | http://www.rosswalker.co.uk | http://www.wmd-lab.org/ |
> > > > > > | Tel: +1 858 822 0854 | EMail:- ross.rosswalker.co.uk |
> > > > > > ---------------------------------------------------------
> > > > > >
> > > > > > Note: Electronic Mail is not secure, has no guarantee of
> delivery,
> > > may
> > > > > not
> > > > > > be read every day, and should not be used for urgent or
> sensitive
> > > > issues.
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > _______________________________________________
> > > > > > AMBER mailing list
> > > > > > AMBER.ambermd.org
> > > > > > http://lists.ambermd.org/mailman/listinfo/amber
> > > > > >
> > > > > _______________________________________________
> > > > > AMBER mailing list
> > > > > AMBER.ambermd.org
> > > > > http://lists.ambermd.org/mailman/listinfo/amber
> > > > >
> > > >
> > > >
> > > >
> > > > --
> > > > Jason M. Swails
> > > > Quantum Theory Project,
> > > > University of Florida
> > > > Ph.D. Graduate Student
> > > > 352-392-4032
> > > > _______________________________________________
> > > > AMBER mailing list
> > > > AMBER.ambermd.org
> > > > http://lists.ambermd.org/mailman/listinfo/amber
> > > >
> > > _______________________________________________
> > > AMBER mailing list
> > > AMBER.ambermd.org
> > > http://lists.ambermd.org/mailman/listinfo/amber
> > >
> >
> >
> >
> > --
> > Jason M. Swails
> > Quantum Theory Project,
> > University of Florida
> > Ph.D. Graduate Student
> > 352-392-4032
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> >
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Aug 10 2010 - 11:00:06 PDT
