AMBER Archive (2005)

Subject: Re: AMBER: tleap load pdb file

From: David A. Case (case_at_scripps.edu)
Date: Tue Feb 08 2005 - 13:50:08 CST

On Mon, Feb 07, 2005, xiaowen wrote:
>
> machine is IBM-AIX, Operation system is UNIX
> xlf for fortran, xlc for C, xlC for C++
>
>
> Today, I followed the instruction to recompile amber8. But cannot finish "make"

Did you previously compile Amber on this machine? You should follow what you
did earlier, if you can remember it.

> (1) I used the same fold name "amber8" and change the old one to amber;
> (2) then copy config.h to $AMBERHOME/src, change:
> AMBER_SRC=/nsfchem/xwfang/amber8/src (the path is the one generated
> by ./configure command)
> (3)make clean >& try01.log &
> (4)make >& try.log &
>
> the try01.log and try.log were attached with this email!

Here is the error from try.log:

make: 1254-005 Ignored error code 2 from last command.
        cd addles; make install
        xlf90 -qfixed -c -g -g -c lesmain.f
"lesmain.f", line 105.0: 1515-005 (S) Continuation line is not permitted at
+beginning of program or after INCLUDE, EJECT, @PROCESS statements or comment
+directive. Line is ignored.

Something is very wrong with your configuration, since the Makefile is not
using the rule that should be specified to compile fortran programs (i.e.,
it is supposed to run the source through cpp to get a file called _lesmain.f,
and then to compile that.

Your config.h file should end with the following lines:

# default rules for Fortran and C compilation:

.f.o: $<
    $(FPP) $< > $(FPP_PREFIX)$<
    $(FC) -c $(FFLAGS) -o $@ $(FPP_PREFIX)$<

.f90.o: $<
    $(FPP) $< > $(FPP_PREFIX)$<
    $(FC) -c $(FFLAGS) -o $@ $(FPP_PREFIX)$<

.c.o:
    $(CC) -c $(CFLAGS) $(CPPFLAGS) -o $@ $<

If these are missing, something is wrong with config.h. If they are present
but being ignored, try changing the following line in config.h, from

.SUFFIXES: .f90

to

.SUFFIXES: .f .c .o .f90

It may be that you will need to use GNU make, rather than IBM's make, but
this seems like a long shot.

Also, about the "nuc.pdb" file: you need to insert a TER card between the two
chains of the DNA in the pdb file. But I doubt that this will fix the problem
you had reported.

...dac

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