AMBER Archive (2008)

Subject: Re: AMBER: Problems installing Amber Tools for Amber 10

From: Sam Danziger (sam.danziger_at_gmail.com)
Date: Fri May 16 2008 - 13:29:32 CDT

I apologize for the numerous posts, but I'm attempting to characterize
the problem as well as possible.

I was able to compile with gcc 4.3.0 by making the following changes
to the configure_at file:

$diff configure_at.bk configure_at.gfortran
215c215
< fc=g77 

---
>     fc="gfortran -fno-automatic"
439,440c439,440
<     fc=g77
<     echo "Mopac and resp will be compiled with g77"
---
>     fc=gfortran -fno-automatic
>     echo "Mopac and resp will be compiled with gfortran"
442c442
<     fflags="-O0"
---
>     fflags="-O1"

When I ran make -f Makefile_at test

all of the tests in the ncsu directory failed with an error that
looked like this:

>  ** NCSU-Error ** : expected list value for key 'i', got '<EMPTY>' instead
---------------------------------------
possible FAILURE:  check pmd.txt.dif
/extra/p530/software/amber10/test/ncsu/pmd

Other than that the antechamber tests failed with errors like:

/extra/p530/software/amber10/test/antechamber/fluorescein
8c8
<       2 C2          1.9170    3.7710    0.1310 ca        1 FLR      0.591300
>       2 C2          1.9170    3.7710    0.1310 ca        1 FLR      0.589100

-Sam

On Fri, May 16, 2008 at 10:41 AM, Sam Danziger <sam.danziger_at_gmail.com> wrote:
> 4.3.0 update:
>
> I checked the lib directory and found that the copy of
> libpdb.a
> was two days old, although all of the other .a's were recently regenerated.
>
> I deleted libpdb.a and then reran make -f Makefile_at.
>
> It looks like Amber Tools compiled and installed correctly after that,
> although this is without any of the fortran parts as 4.3.0 does not
> include g77.
>
> -Sam
>
> On Fri, May 16, 2008 at 10:06 AM, Sam Danziger <sam.danziger_at_gmail.com> wrote:
>> David,
>>
>> Thanks for the response, I was starting to fear that my problem had
>> dissapeared into the great beyond.
>>
>> Here is the gcc 4.3.0 error message.  Note: I had to change to
>> gfortran and not compile mopac to get this far.
>>
>> Thank you,
>> -Sam
>>
>> gcc -o rdparm main.o rdparm.o dispatch.o help.o utility.o second.o io.o trajecto
>> ry.o evec.o torsion.o mask.o rms.o display.o interface.o energy.o experimental.o
>>  ptraj.o actions.o analyze.o thermo.o pubfft.o cluster.o clusterLib.o /extra/p53
>> 0/software/amber10/lib/libpdb.a /extra/p530/software/amber10/lib/libsym.a /extra
>> /p530/software/amber10/lib/arpack.a /extra/p530/software/amber10/lib/lapack.a /e
>> xtra/p530/software/amber10/lib/blas.a /extra/p530/software/amber10/lib/f2c.a
>> ../netcdf/lib/libnetcdf.a -lm
>> /extra/p530/software/amber10/lib/libpdb.a(pdb_read.o): In function `pdb_read_str
>> ing.':
>> pdb_read.c:(.text+0x1d32): undefined reference to `_intel_fast_memset'
>> /extra/p530/software/amber10/lib/libpdb.a(pdb_sscanf.o): In function `pdb_sscanf
>> ':
>> pdb_sscanf.c:(.text+0x2a0): undefined reference to `_intel_fast_memset'
>> /extra/p530/software/amber10/lib/libpdb.a(pdb_sprntf.o): In function `pdb_sprint
>> f':
>> pdb_sprntf.c:(.text+0x2b9): undefined reference to `_intel_fast_memset'
>> pdb_sprntf.c:(.text+0x4de): undefined reference to `_intel_fast_memset'
>> pdb_sprntf.c:(.text+0x7ea): undefined reference to `_intel_fast_memset'
>> /extra/p530/software/amber10/lib/libpdb.a(pdb_sprntf.o): In function `outexp':
>> pdb_sprntf.c:(.text+0xbab): undefined reference to `__intel_f2int'
>> pdb_sprntf.c:(.text+0xbd7): undefined reference to `__intel_f2int'
>> pdb_sprntf.c:(.text+0xd4a): undefined reference to `_intel_fast_memset'
>> /extra/p530/software/amber10/lib/libpdb.a(pdb_sprntf.o): In function `outfloat':
>>
>> pdb_sprntf.c:(.text+0xdd7): undefined reference to `__intel_f2int'
>> pdb_sprntf.c:(.text+0xe09): undefined reference to `_intel_fast_memset'
>> pdb_sprntf.c:(.text+0xf91): undefined reference to `_intel_fast_memcpy'
>> pdb_sprntf.c:(.text+0x100d): undefined reference to `_intel_fast_memset'
>> pdb_sprntf.c:(.text+0x1033): undefined reference to `_intel_fast_memset'
>> /extra/p530/software/amber10/lib/libpdb.a(pdb_sprntf.o): In function `outstr':
>> pdb_sprntf.c:(.text+0x1151): undefined reference to `_intel_fast_memcpy'
>> pdb_sprntf.c:(.text+0x119b): undefined reference to `_intel_fast_memset'
>> /extra/p530/software/amber10/lib/libpdb.a(pdb_sprntf.o): In function `outunsigne
>> d':
>> pdb_sprntf.c:(.text+0x12a7): undefined reference to `_intel_fast_memset'
>> /extra/p530/software/amber10/lib/libpdb.a(pdb_sprntf.o): In function `outint':
>> pdb_sprntf.c:(.text+0x13f7): undefined reference to `_intel_fast_memset'
>> collect2: ld returned 1 exit status
>> make[1]: *** [rdparm] Error 1
>> make[1]: Leaving directory `/extra/p530/software/amber10/src/ptraj'
>> make: *** [install] Error 2
>>
>>
>> On Thu, May 15, 2008 at 9:47 PM, David A. Case <case_at_scripps.edu> wrote:
>>> On Mon, May 12, 2008, Sam Danziger wrote:
>>>
>>>> I've tried to install Amber Tools 1.0, but get an error message when I
>>>> try to compile.
>>>>
>>>> I'm using Gentoo linux (Linux version 2.6.22-gentoo-r18), and have
>>>> tried gcc 3.4.1, 3.4.4 and 4.3.0 (4.3.0 gives me a different error
>>>> than below).
>>>>
>>>
>>> Thanks for the report; aren't C++ error messages awe-inspiring sometimes?
>>>
>>> I've entered this into bugzilla.  Probably no developer has used Gentoo,
>>> although other Linux and Mac and cygwin versions of gcc seem OK here.
>>>
>>> It might be helpful to post the 4.3.0 error message as well...you never know
>>> when a clue will pop up.
>>>
>>> ...regards...dac
>>>
>>> -----------------------------------------------------------------------
>>> The AMBER Mail Reflector
>>> To post, send mail to amber_at_scripps.edu
>>> To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
>>>      to majordomo_at_scripps.edu
>>>
>>
>>
>>
>> --
>> Samuel A. Danziger
>> Dept. of Biomedical Engineering
>> PhD Candidate
>> University of California
>> Irvine, CA 92697
>> sdanzige_at_uci.edu
>> (908)246-4502
>> http://samdanziger.com/
>>
>> Associated Graduate Students
>> VP Financial Affairs
>> vpfinance_at_ags.uci.edu
>>
>> Truth Cannot Contradict Truth
>>
>
>
>
> --
> Samuel A. Danziger
> Dept. of Biomedical Engineering
> PhD Candidate
> University of California
> Irvine, CA 92697
> sdanzige_at_uci.edu
> (908)246-4502
> http://samdanziger.com/
>
> Associated Graduate Students
> VP Financial Affairs
> vpfinance_at_ags.uci.edu
>
> Truth Cannot Contradict Truth
>

-- 
Samuel A. Danziger
Dept. of Biomedical Engineering
PhD Candidate
University of California
Irvine, CA 92697
sdanzige_at_uci.edu
(908)246-4502
http://samdanziger.com/

Associated Graduate Students
VP Financial Affairs
vpfinance_at_ags.uci.edu

Truth Cannot Contradict Truth
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
      to majordomo_at_scripps.edu