Re: [AMBER] compilation of amber12 on cray system

From: Jason Swails <jason.swails.gmail.com>
Date: Thu, 11 Jul 2013 15:27:51 -0400

On Thu, Jul 11, 2013 at 3:07 PM, Jorgen Simonsen <jorgen589.gmail.com>wrote:

> Hi all,
>
> I am trying to compile amber12 on a cray-system. For the serial compilation
> everything is fine and the make.test serial runs fine without any issues.
>
> For the mpi-compilation I do have some problems getting it to work fine. I
> generate the config.h file using the following syntax:


> ./configure -noX11 -nobintraj -crayxt5 -mpi pgi
>
> >From the test-run I get the following output:
>
> Obtaining the PGI suite version:
> pgcc -V
> The version is 12.4-0
>
> Testing the cc compiler:
> cc -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -o testp testp.c
> OK
>
> Testing the ftn compiler:
> ftn -O1 -o testp testp.f
> OK
>
> Testing mixed C/Fortran compilation:
> cc -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -c -o testp.c.o
> testp.c
> ftn -O1 -c -o testp.f.o testp.f
> cc -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -o testp testp.c.o
> testp.f.o -pgf90libs
> OK
>
> Testing pointer size:
> cc -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -o test_pointer_size
> test_pointer_size.c
> Detected 64 bit operating system.
>
> Testing flex: OK
>
> Checking for zlib: OK
>
> Checking for libbz2: OK
>
> Configuring fftw-3.3 for mdgx (may be time-consuming)...
>
> fftw-3.3 configure succeeded.
>
> The configuration file, config.h, was successfully created.
>
> The next step is to type 'make install'
>
> Cleaning the src directories. This may take a few moments.
> Configure complete.
>
>
> and the compilation ends with the following error
>
> AT12/amber12/bin/mpinab -c dna3.nab
> nab2c failed!
> make[2]: *** [dna3.o] Error 1
>
> I have seen it on the list but I did not find the solution to it. I have
> tried to compile directly like this
>
> mpinab -c dna3.nab
>
> and no new file is generated as asked for in the thread with the same
> issue. Any help much appreciated
>

My guess is that you will never need an MPI-enabled NAB (OpenMP would
probably be better, anyway, if you really needed a threaded NAB
application).

In this case, you might as well just skip the parallel AmberTools
installation (since MMPBSA.py.MPI has never worked for me on a Cray machine
before).

You can go to $AMBERHOME/src/ and run "make install" there to skip
AmberTools.

HTH,
Jason

-- 
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 Thu Jul 11 2013 - 12:30:04 PDT
Custom Search