Yes, but you have to remove all references to intel-specific compiler flags
(and don't forget to do a make clean before trying to build again), and make
sure F90 = mpif90 and LD = mpif90.
"How do I know which flags are intel-specific," you say? Whichever flags
your compiler complains about ;). For starters, I would remove the flags
-limf -lsvml -Wl,-rpath=$(IFORT_RPATH)
-ip -wX
since I'm pretty sure those are intel-specific.
HTH,
Jason
On Mon, Aug 8, 2011 at 9:27 AM, hari krishna <haricoolguy111.gmail.com>wrote:
> currently i dont have intel MPI library can i use gfortran as a compiler
>
> On Mon, Aug 8, 2011 at 6:45 PM, Daniel Roe <daniel.r.roe.gmail.com> wrote:
>
> > Hi,
> > Your MPI_HOME is pointing to a gnu compilation, but you are attempting to
> > compile with intel compilers. For ./configure ifort you should use an MPI
> > installation compiled with intel compilers as well.
> >
> > -Dan
> >
> > On Aug 8, 2011, at 8:42 AM, hari krishna <haricoolguy111.gmail.com>
> wrote:
> >
> > > I am using AMBER10 below is the config.h file
> > > platform: X86_64
> > > compiler: ifort
> > > mpi: mpich2
> > > not using netcdf as default
> > >
> > > *./configure linux64_opteron ifort mpich2 pubfft nobintraj* (this is
> the
> > one
> > > i tried now)
> > >
> > > MATH_DEFINES =
> > > MATH_LIBS =
> > > IFORT_RPATH =
> > >
> >
> /home/p_kumar/intel/composerxe-2011.5.220/compiler/lib/intel64:/home/p_kumar/intel/composerxe-2011.5.220/compiler/lib/intel64:/home/p_kumar/intel/composerxe-2011.5.220/mkl/lib/intel64:/home/p_kumar/intel/composerxe-2011.5.220/compiler/lib/ia32:/home/p_kumar/intel/composerxe-2011.5.220/compiler/lib/ia32:/home/p_kumar/intel/composerxe-2011.5.220/mkl/lib/ia32:/home/p_kumar/intel/composerxe-2011.5.220/compiler/lib/intel64:/home/p_kumar/intel/composerxe-2011.5.220/compiler/lib/intel64:/home/p_kumar/intel/composerxe-2011.5.220/mkl/lib/intel64:/home/p_kumar/intel/composerxe-2011.5.220/compiler/lib/ia32:/home/p_kumar/intel/composerxe-2011.5.220/compiler/lib/ia32:/home/p_kumar/intel/composerxe-2011.5.220/mkl/lib/ia32:/home/p_kumar/intel/composerxe-2011.5.220/mpirt/lib/ia32:/home/p_kumar/intel/composerxe-2011.5.220/mpirt/lib/intel64:/home/p_kumar/intel/composerxe-2011.5.220/mpirt/lib/ia32:/home/p_kumar/intel/composerxe-2011.5.220/mpirt/lib/intel64
> > > FFT_DEFINES = -DPUBFFT
> > > FFT_INCLUDE =
> > > FFT_LIBS =
> > > NETCDF_HOME =
> > > NETCDF_DEFINES =
> > > NETCDF_MOD =
> > > NETCDF_LIBS =
> > > MPI_HOME = /opt/mpich2/gnu
> > > MPI_DEFINES = -DMPI -DSLOW_NONBLOCKING_MPI
> > > MPI_INCLUDE = -I$(MPI_HOME)/include
> > > MPI_LIBDIR = $(MPI_HOME)/lib
> > > MPI_LIBS =
> > > DIRFRC_DEFINES = -DDIRFRC_EFS -DDIRFRC_NOVEC
> > > CPP = /lib/cpp
> > > CPPFLAGS = -traditional -P
> > > F90_DEFINES = -DFFTLOADBAL_2PROC
> > >
> > > F90 = mpif90
> > > MODULE_SUFFIX = mod
> > > F90FLAGS = -c -auto
> > > F90_OPT_DBG = -g -traceback
> > > F90_OPT_LO = -O0
> > > F90_OPT_MED = -O2
> > > F90_OPT_HI = -xW -ip -O3
> > > CC = gcc
> > > CFLAGS =
> > >
> > > LOAD = mpif90
> > > LOADFLAGS =
> > > LOADLIBS = -limf -lsvml -Wl,-rpath=$(IFORT_RPATH)
> > >
> > > Error Message
> > >
> > >
> >
> /home/p_kumar/intel/composerxe-2011.5.220/compiler/lib/intel64/libsvml.so:
> > > undefined reference to `__intel_cpu_indicator_init'
> > >
> >
> /home/p_kumar/intel/composerxe-2011.5.220/compiler/lib/intel64/libsvml.so:
> > > undefined reference to `__intel_cpu_indicator'
> > > collect2: ld returned 1 exit status
> > > make[1]: *** [pmemd] Error 1
> > > make[1]: Leaving directory `/home/p_kumar/amber10/src/pmemd/src'
> > > make: *** [all] Error 2
> > >
> > >
> > > On Mon, Aug 8, 2011 at 5:45 PM, David A Case <case.biomaps.rutgers.edu
> > >wrote:
> > >
> > >> On Mon, Aug 08, 2011, hari krishna wrote:
> > >>>
> > >>> *./configure linux64_opteron gfortran mpi pubfft bintraj *
> > >>
> > >> This looks like a very old version of Amber; what version is it?
> > >>
> > >>> ---
> > >>> *make *
> > >>>
> > >>> *end module gbl_constants_mod
> > >>> gbl_constants.f90
> > >>> make[1]: gbl_constants.f90: Command not found
> > >>
> > >> We need more details of the error message. Also, does sander compile
> > >> correctly? It's probably also good to post the config.h file that is
> > >> created
> > >> by configure.
> > >>
> > >> ...dac
> > >>
> > >>
> > >> _______________________________________________
> > >> 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
> >
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Candidate
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Aug 08 2011 - 07:00:02 PDT
