AMBER Archive (2009)

Subject: Re: [AMBER] AmberTools-1.3 not compiling on MacOsX

From: Jason Swails (jason.swails_at_gmail.com)
Date: Wed Dec 23 2009 - 05:38:08 CST

I would personally suggest the MacPorts option, since the binaries on
hpc.sourceforge.net are "experimental", and there are known bugs when
trying to use that compiler on some Amber code (AmberTools has
compiled fine, it's bits of Amber I've had problems with). My version
of gfortran from MacPorts (gfortran 4.5) also passes all tests on
AmberTools 1.3 whereas the gfortran4.5 from hpc.sourceforge.net did
not... However, the gfortran 4.5 from MacPorts is also experimental,
so you may want to install gcc44 from MacPorts, and follow Mengjuei's
steps.

The errors you got are as David suggested, your gfortran is creating
only binaries built for the i368 (32-bit) architecture, whereas your
gcc, built for Snow Leopard, is building 64-bit (x86_64) binaries, so
ranlib complains when trying to *compile* these "incompatible" object
files into a library. Such are the pains when Apple attempts to
transition their OS into the 64-bit world, as Snow Leopard is their
first 64-bit kernel (though by default, personal Macs still boot a
64-bit-enabled, 32-bit kernel, which is a source of much of the
trouble).

Good luck!
Jason

On Wed, Dec 23, 2009 at 5:05 AM, Mengjuei Hsieh <mjhsieh_at_gmail.com> wrote:
> On Wed, Dec 23, 2009 at 1:12 AM, I <mjhsieh_at_gmail.com> wrote:
>> What I did was to
>>    1) have a my ~/bin in the PATH
>>    2) ln -s /opt/local/bin/gfortran-mp-4.4 ~/bin/gfortran
>>    3) cd $AMBERHOME/src; ./configure -macAccelerate gnu
>
> Here are some more elaborated steps:
>
> http://amberonmac.blogspot.com/2009/12/step-by-step-procedure-of-macports.html
>
> Best,
> --
> Mengjuei
>
> _______________________________________________
> AMBER mailing list
> AMBER_at_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_at_ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber