AMBER Archive (2009)

Subject: [AMBER] Compiling AmberTools 1.2 with intel compilers for Mac Osx in 64 bits

From: Alan (alanwilter_at_gmail.com)
Date: Wed Jul 29 2009 - 11:00:42 CDT

Dear all,
I use Fink in 64 bits mode and intel compilers 11.1 (058) on Mac OSX 10.5.7.

I am studying "amber10/src/configure_at" because I also want to use intel
MKL library.

I see things that I likely need to do like setting:

        processor='x86_64' # because `uname -m` in my mac will return 'i386'

And then, looking at case icc) for $mkl: I set 'export
MKL_HOME=/opt/intel/Compiler/11.1/058/Frameworks/mkl'.

Looking at: tree $MKL_HOME/lib/
/opt/intel/Compiler/11.1/058/Frameworks/mkl/lib/
|-- 32
| |-- libmkl_blas95.a
| |-- libmkl_core.a
| |-- libmkl_core.dylib
| |-- libmkl_intel.a
| |-- libmkl_intel.dylib
| |-- libmkl_intel_thread.a
| |-- libmkl_intel_thread.dylib
| |-- libmkl_lapack.dylib
| |-- libmkl_lapack95.a
| |-- libmkl_p4m.dylib
| |-- libmkl_p4m3.dylib
| |-- libmkl_p4p.dylib
| |-- libmkl_pgi_thread.a
| |-- libmkl_sequential.a
| |-- libmkl_sequential.dylib
| |-- libmkl_solver.a
| |-- libmkl_solver_sequential.a
| |-- libmkl_vml_p4m.dylib
| |-- libmkl_vml_p4m2.dylib
| |-- libmkl_vml_p4m3.dylib
| |-- libmkl_vml_p4p.dylib
| `-- locale
| `-- en_US
| `-- mkl_msg.cat
|-- em64t
| |-- libmkl_blas95_ilp64.a
| |-- libmkl_blas95_lp64.a
| |-- libmkl_core.a
| |-- libmkl_core.dylib
| |-- libmkl_intel_ilp64.a
| |-- libmkl_intel_ilp64.dylib
| |-- libmkl_intel_lp64.a
| |-- libmkl_intel_lp64.dylib
| |-- libmkl_intel_sp2dp.a
| |-- libmkl_intel_sp2dp.dylib
| |-- libmkl_intel_thread.a
| |-- libmkl_intel_thread.dylib
| |-- libmkl_lapack.dylib
| |-- libmkl_lapack95_ilp64.a
| |-- libmkl_lapack95_lp64.a
| |-- libmkl_mc.dylib
| |-- libmkl_mc3.dylib
| |-- libmkl_pgi_thread.a
| |-- libmkl_sequential.a
| |-- libmkl_sequential.dylib
| |-- libmkl_solver_ilp64.a
| |-- libmkl_solver_ilp64_sequential.a
| |-- libmkl_solver_lp64.a
| |-- libmkl_solver_lp64_sequential.a
| |-- libmkl_vml_mc.dylib
| |-- libmkl_vml_mc2.dylib
| |-- libmkl_vml_mc3.dylib
| `-- locale
| `-- en_US
| `-- mkl_msg.cat
`-- universal
    |-- libmkl_core.a
    |-- libmkl_core.dylib
    |-- libmkl_intel_ilp64.a
    |-- libmkl_intel_ilp64.dylib
    |-- libmkl_intel_lp64.a
    |-- libmkl_intel_lp64.dylib
    |-- libmkl_intel_sp2dp.a
    |-- libmkl_intel_sp2dp.dylib
    |-- libmkl_intel_thread.a
    |-- libmkl_intel_thread.dylib
    |-- libmkl_lapack.dylib
    |-- libmkl_mc.dylib
    |-- libmkl_mc3.dylib
    |-- libmkl_p4m.dylib
    |-- libmkl_p4m3.dylib
    |-- libmkl_p4p.dylib
    |-- libmkl_pgi_thread.a
    |-- libmkl_sequential.a
    |-- libmkl_sequential.dylib
    |-- libmkl_solver_ilp64.a
    |-- libmkl_solver_ilp64_sequential.a
    |-- libmkl_solver_lp64.a
    |-- libmkl_solver_lp64_sequential.a
    |-- libmkl_vml_mc.dylib
    |-- libmkl_vml_mc2.dylib
    |-- libmkl_vml_mc3.dylib
    |-- libmkl_vml_p4m.dylib
    |-- libmkl_vml_p4m2.dylib
    |-- libmkl_vml_p4m3.dylib
    `-- libmkl_vml_p4p.dylib

7 directories, 80 files

And I cannot see any file called "libmkl_lapack.a" or "libmkl_em64t.a" and I
don't know, either, there's in an equivalent since, e.g., for lapack I have:

| |-- libmkl_lapack.dylib
| |-- libmkl_lapack95_ilp64.a
| |-- libmkl_lapack95_lp64.a

but not "libmkl_lapack.a".

I was wondering if someone here could give a tip about how to proceed in
changing configure_at (or even Makefile_at) in order to try to make it work
with intel compilers and intel MKL libs.

NOTE:
If I don't touch configure_at and don't set MKL_HOME, simply doing:
./configure_at icc
make -f Makefile_at
cd $AMBERHOME/test; make -f Makefile_at test

It compiled fine, and all test worked except for nab, that gives:
cd $AMBERHOME/test; make -f Makefile_at test
( cd nab; make test )
=====================================================
Running test to make dna duplex:

/bin/sh: line 1: 28400 Segmentation fault ./duplex < duplex.in
make[1]: *** [duplex_test] Error 139

NOTE2:
I didn't find more documentation about './configure_at -h' for options
'-mpi' or '-openmp'. It would enable some programmes in ambertools to use
multi-cores? Which programmes could take advantage of these options?

Many thanks in advance,

Alan 

-- 
Alan Wilter S. da Silva, D.Sc. - CCPN Research Associate
Department of Biochemistry, University of Cambridge.
80 Tennis Court Road, Cambridge CB2 1GA, UK.
>>http://www.bio.cam.ac.uk/~awd28<<
_______________________________________________
AMBER mailing list
AMBER_at_ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber