Re: [AMBER] parallel compiling failure of Amber 11

From: Yubo Fan <yubofan.mail.chem.tamu.edu>
Date: Tue, 22 Jun 2010 11:13:46 -0500

Jason,

I did what you suggested but still get errors. Is there anything wrong
about my settings for Intel compiler? Or the source code downloaded
from www.open-mpi.org.


environment setting in ~/.bash_profile
***********************************************************************************************************************
# .bash_profile

# Get the aliases and functions
if [ -f ~/.bashrc ]; then
         . ~/.bashrc
fi

# User specific environment and startup programs
source /opt/intel/Compiler/11.1/072/bin/intel64/iccvars_intel64.sh
source /opt/intel/Compiler/11.1/072/bin/intel64/ifortvars_intel64.sh
source
/opt/intel/Compiler/11.1/072/mkl/tools/environment/mklvarsem64t.sh

export AMBERHOME="/opt/amber11"

export MKL_HOME="/opt/intel/Compiler/11.1/072/mkl"

unset USERNAME
***********************************************************************************************************************
***********************************************************************************************************************
[root.localhost openmpi-1.4.2]# ./configure
--prefix=/opt/openmpi-1.4.2 F90=ifort, F77=ifort, CC=icc, CXX=icpc
checking for a BSD-compatible install... /usr/bin/install -c
checking whether build environment is sane... yes
checking for a thread-safe mkdir -p... /bin/mkdir -p
checking for gawk... gawk
checking whether make sets $(MAKE)... yes
checking how to create a ustar tar archive... gnutar

============================================================================
== Configuring Open MPI
============================================================================

*** Checking versions
checking Open MPI version... 1.4.2
checking Open MPI release date... May 04, 2010
checking Open MPI Subversion repository version... r23093
checking Open Run-Time Environment version... 1.4.2
checking Open Run-Time Environment release date... May 04, 2010
checking Open Run-Time Environment Subversion repository version...
r23093
checking Open Portable Access Layer version... 1.4.2
checking Open Portable Access Layer release date... May 04, 2010
checking Open Portable Access Layer Subversion repository version...
r23093

*** Initialization, setup
configure: builddir: /opt/openmpi-1.4.2
configure: srcdir: /opt/openmpi-1.4.2
checking build system type... x86_64-unknown-linux-gnu
checking host system type... x86_64-unknown-linux-gnu
installing to directory "/opt/openmpi-1.4.2"

*** Configuration options
checking whether to run code coverage... no
checking whether to compile with branch probabilities... 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 sparse process groups... no
checking if want Fortran 77 bindings... yes
checking if want Fortran 90 bindings... yes
checking desired Fortran 90 bindings "size"... small
checking whether to enable PMPI... yes
checking if want C++ bindings... yes
checking if want MPI::SEEK_SET support... 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 peruse support is required... 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... no
checking if want trace file debugging... no
checking if want full RTE support... yes
checking if want fault tolerance... Disabled fault tolerance
checking if want IPv6 support... yes (if underlying system supports
it)
checking if want orterun "--prefix" behavior to be enabled by
default... no
checking for package/brand string... Open MPI
root.localhost.localdomain Distribution
checking for ident string... 1.4.2
checking whether to add padding to the openib control header... no
checking whether to use an alternative checksum algo for messages...
no

============================================================================
== Compiler and preprocessor tests
============================================================================

*** C compiler and preprocessor
checking for style of include used by make... GNU
checking for gcc... icc,
checking for C compiler default output file name...
configure: error: in `/opt/openmpi-1.4.2':
configure: error: C compiler cannot create executables
See `config.log' for more details.
***********************************************************************************************************************

More helps. Thanks a lot.

Best,
Yubo


On Tue, 22 Jun 2010 11:12:26 -0400
  Jason Swails <jason.swails.gmail.com> wrote:
> Hello,
>
> Your problem is that you're installing openMPI with the gnu
>compilers, but
> you're telling configure to use the intel compilers for some reason.
> Run
> configure like this instead:
>
> ./configure -mpi gnu
>
> Then make parallel. You have to use the compilers used to build the
>MPI
> you're using to compile Amber in parallel. If you want an MPI to
>build with
> the intel compilers, you will have to compile from source, setting
>F90=ifort, F77=ifort, CC=icc, and CXX=icpc.
>
> All the best,
> Jason
>
> On Tue, Jun 22, 2010 at 10:56 AM, Yubo Fan
><yubofan.mail.chem.tamu.edu>wrote:
>
>> Jason,
>>
>> I reinstalled openmpi and it is in /usr/lib/openmpi/1.4-gcc/. I ran
>> ./configure -mpi intel in /opt/amber11/AmberTools/src and the error
>>message
>> came out as:
>> Configuring Parallel NetCDF; (may be time-consuming)
>>
>> Error: parallel NetCDF configure returned 77
>> Parallel NetCDF configure failed! Check the pnetcdf/config.log
>>file.
>> Continuing but parallel NetCDF will be skipped.
>>
>> Configuring NetCDF; (may be time-consuming)
>>
>> NetCDF configure succeeded.
>>
>> Configuring fftw-2.1.5; (may be time-consuming)
>>
>> Error: fftw configure returned 1
>> fftw configure failed! Check the fftw2_config.log file.
>>
>> The two log files are attached.
>>
>> I used "yum install openmpi*" to install openmpi. Do I have install
>>openmpi
>> individually?
>>
>> Thanks,
>> Yubo
>>
>>
>> mpif90 -show
>>>
>> gfortran -I/usr/lib/openmpi/1.4-gcc/include -m32 -pthread
>> -I/usr/lib/openmpi/1.4-gcc/lib -L/usr/lib/openmpi/1.4-gcc/lib
>>-lmpi_f90
>> -lmpi_f77 -lmpi -lopen-rte -lopen-pal -ldl -Wl,--export-dynamic
>>-lnsl -lutil
>> -lm -ldl
>>
>> mpicc -show
>>>
>> gcc -I/usr/lib/openmpi/1.4-gcc/include -m32 -pthread
>> -L/usr/lib/openmpi/1.4-gcc/lib -lmpi -lopen-rte -lopen-pal -ldl
>> -Wl,--export-dynamic -lnsl -lutil -lm -ldl
>>
>> On Mon, 21 Jun 2010 16:29:43 -0400
>> Jason Swails <jason.swails.gmail.com> wrote:
>>
>>> Hello,
>>>
>>> See my comments below.
>>>
>>> On Mon, Jun 21, 2010 at 4:25 PM, Yubo Fan
>>><yubofan.mail.chem.tamu.edu
>>> >wrote:
>>>
>>> Dr. Case,
>>>>
>>>> I got a problem to compile parallel version of Amber on my local
>>>> machine. The commands and error message are shown below:
>>>>
>>>>
>>>>
>>>> =================================================================================
>>>> [root.localhost src]# ./configure -mpi intel
>>>> Your AMBERHOME environment variable should be set to /opt/amber11
>>>>
>>>> Using Intel MKL libraries in
>>>> /opt/intel/Compiler/11.1/072/mkl/lib/em64t
>>>> MKL Version 10 or 11 assumed.
>>>>
>>>> Testing the icc compiler:
>>>> icc -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -o testp testp.c
>>>> OK
>>>>
>>>> Testing the ifort compiler:
>>>> ifort -O0 -o testp testp.f
>>>> OK
>>>>
>>>> Testing mixed C/Fortran compilation:
>>>> icc -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -c -o testp.c.o
>>>> testp.c
>>>> ifort -O0 -c -o testp.f.o testp.f
>>>> icc -o testp testp.c.o testp.f.o
>>>> -L/opt/intel/Compiler/11.1/072/lib/intel64/ -lifport -lifcore -lsvml
>>>> /opt/intel/Compiler/11.1/072/lib/intel64/libimf.so: warning:
>>>>warning:
>>>> feupdateenv is not implemented and will always fail
>>>> OK
>>>>
>>>> Testing pointer size:
>>>> icc -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -o
>>>> test_pointer_size test_pointer_size.c
>>>> Detected 64 bit operating system.
>>>>
>>>> Testing flex:
>>>> OK
>>>>
>>>> Configuring Parallel NetCDF; (may be time-consuming)
>>>>
>>>> Error: parallel NetCDF configure returned 77
>>>> Parallel NetCDF configure failed! Check the pnetcdf/config.log
>>>> file.
>>>>
>>>>
>>> What does pnetcdf/config.log say?
>>>
>>>
>>> Continuing but parallel NetCDF will be skipped.
>>>>
>>>> Configuring NetCDF; (may be time-consuming)
>>>>
>>>> NetCDF configure succeeded.
>>>>
>>>> Configuring fftw-2.1.5; (may be time-consuming)
>>>>
>>>> Error: fftw configure returned 1
>>>> fftw configure failed! Check the fftw2_config.log file.
>>>>
>>>>
>>> What does this file say?
>>>
>>> It may be that you don't have a working MPI, or you're using a
>>>mis-match
>>> of
>>> MPIs. This is especially true for personal systems, since users may
>>>not
>>> have properly built their MPIs. What does "mpif90 -show" and "mpicc
>>> -show"
>>> return on your system?
>>>
>>> Good luck!
>>> Jason
>>>
>>>
>>>
>>>>
>>>> =================================================================================
>>>>
>>>> Any advice?
>>>>
>>>> Regards,
>>>> Yubo
>>>>
>>>> On Mon, 14 Jun 2010 12:34:13 -0400
>>>> case <case.biomaps.rutgers.edu> wrote:
>>>> > On Mon, Jun 14, 2010, Yubo Fan wrote:
>>>> >>
>>>> >> When I ran 'make parallel' at $AMBERHOME/src, the error message
>>>> >> appears as:
>>>> >> Error: config.h is not of type parallel!
>>>> >> Rerun ./configure and specify an MPI implementation.
>>>> >> make: *** [configured_parallel] Error 2
>>>> >>
>>>> >> The making of AmberTools and 'make serial' passed already. I ran
>>>> >> './configure -openmp intel' to create a config.h file at
>>>> >
>>>> > This is wrong, although perhaps an understandable error.
>>>> >
>>>> > 1. AmberTools can be made in parallel in two ways: using openmp
>>>>or
>>>> >MPI.
>>>> >
>>>> > 2. However, Amber itself can only be made in parallel using MPI.
>>>> > So, from
>>>> > the point of view of Amber, using the "-openmp" option is not
>>>>enough
>>>> >to give
>>>> > a "parallel" config.h. You need to run the configure script using
>>>> >the "-mpi"
>>>> > option, not the "openmp" one.
>>>> >
>>>> > ....dac
>>>>
>>>> >
>>>> >
>>>> > _______________________________________________
>>>> > AMBER mailing list
>>>> > AMBER.ambermd.org
>>>> > http://lists.ambermd.org/mailman/listinfo/amber
>>>>
>>>> Yubo Fan
>>>> Chemistry Department
>>>> Texas A & M University
>>>> Tel: (979) 845-5237
>>>> Email: yubofan.mail.chem.tamu.edu
>>>>
>>>>
>>>> _______________________________________________
>>>> 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
>>>
>>
>> Yubo Fan
>> Chemistry Department
>> Texas A & M University
>> Tel: (979) 845-5237
>> Email: yubofan.mail.chem.tamu.edu
>>
>
>
>
> --
> 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

Yubo Fan
Chemistry Department
Texas A & M University
Tel: (979) 845-5237
Email: yubofan.mail.chem.tamu.edu

_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jun 22 2010 - 09:30:03 PDT
Custom Search