AMBER Archive (2008)

Subject: RE: AMBER: AMBER 8 problems on RHEL4/Opteron/PGF 7.0-7

From: Ross Walker (ross_at_rosswalker.co.uk)
Date: Wed Feb 13 2008 - 09:20:07 CST


Hi Atro,

Unfortunately AMBER 8 was released before all the complications of x86 vs
x86_64 came about and thus it does not deal well with such architectures.

One option would be to edit $AMBERHOME/src/config.h and add "-tp k8-32" to
the CFLAGS line. Then you can make clean and make again and I think that
will then give you all 32 bit object files such that it links properly
without you having to manually do it.

Good luck,
Ross

/\
\/
|\oss Walker

| Assistant Research Professor |
| San Diego Supercomputer Center |
| Tel: +1 858 822 0854 | EMail:- ross_at_rosswalker.co.uk |
| http://www.rosswalker.co.uk | PGP Key available on request |

Note: Electronic Mail is not secure, has no guarantee of delivery, may not
be read every day, and should not be used for urgent or sensitive issues.

> -----Original Message-----
> From: owner-amber_at_scripps.edu
> [mailto:owner-amber_at_scripps.edu] On Behalf Of Atro Tossavainen
> Sent: Wednesday, February 13, 2008 05:32
> To: amber_at_scripps.edu
> Subject: AMBER: AMBER 8 problems on RHEL4/Opteron/PGF 7.0-7
>
> I'm using
>
> ./configure -opteron pgf90
>
> on csc.fi's Murska cluster. I have applied today's bugfix.all which
> includes everything up to bugfix.64. Everything is fine up until
>
> pgcc -o makeDIST_RST makeDIST_RST.o -lm
> /usr/bin/ld: warning: i386 architecture of input file
> `makeDIST_RST.o' is incompatible with i386:x86-64 output
> makeDIST_RST.o(.text+0xe3d): In function `correct_upperbound':
> : undefined reference to `__mth_i_dlogx'
> makeDIST_RST.o(.text+0xe4a): In function `correct_upperbound':
> : undefined reference to `__mth_i_dexpx'
> make[2]: *** [makeDIST_RST] Error 2
> make[2]: Leaving directory
> `/home/u3/tossavai/amber8/src/nmr_aux/prepare_input'
> make[1]: *** [install] Error 2
> make[1]: Leaving directory `/home/u3/tossavai/amber8/src/nmr_aux'
> make: *** [serial] Error 2
>
> which goes away if the "-tp k8-32" flag is included manually:
>
> ~/amber8/src % cd nmr_aux/prepare_input
> ~/amber8/src/nmr_aux/prepare_input % pgcc -tp k8-32 -o
> makeDIST_RST makeDIST_RST.o -lm
> ~/amber8/src/nmr_aux/prepare_input %
>
> Afterwards, I go back to $AMBERHOME/src, make serial again, and the
> previous step deletes my hand-built binary, so I lose... :-)
>
> Adding $(CFLAGS) to the Makefile in nmr_aux/prepare_input,
> after $(LOADCC)
> and before -o makeWHATEVER, makes this go away. I wonder if it breaks
> stuff on other platforms?
>
> This also happens in src/leap/src/leap with
>
> pgcc -o utilMakeHelp basics.o sysdepend.o stringExtra.o
> varArray.o getline.o avl.o pdb_format.o pdb_read.o
> pdb_sprntf.o pdb_sscanf.o pdb_write.o vector.o zMatrix.o
> sort.o bag.o hash.o dictionary.o database.o nVector.o ring.o
> matrix.o fortran.o displayer.o utilMakeHelp.o -lm
>
> and the same fix makes it go away.
>
> Next, I'm trying to "make test", but sander segfaults... Where do I
> start trying to figure this out?
>
> --
> Atro Tossavainen (Mr.) / The Institute of
> Biotechnology at
> Systems Analyst, Techno-Amish & / the University of
> Helsinki, Finland,
> +358-9-19158939 UNIX Dinosaur / employs me, but my
> opinions are my own.
> < URL : http : / / www . helsinki . fi / %7E atossava / > NO
> FILE ATTACHMENTS
> --------------------------------------------------------------
> ---------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
>

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu