[Wien] VERSION_05.3: problems with installation of mini

[L. D. Marks]

I don't know if it is the version that you have, or perhaps an
inconsistent version on the web page. The routines sortag & drmngwien
should be in the Makefile.

The routine mdble.c just defines some machine constants. It looks like
your fortran compiler has converted everything to upper case and/or is
being case sensitive, so the C to Fortran communication has got messed up.
A brief google (http://www.hpcf.cam.ac.uk/history/turing.html)
suggests that this may be a general thing with Hitachi compilers. I've
attached a version where I've added upper case versions of the routines
(since C is case sensitive). Copy your previous version somewhere
safe then please let me know if it works. If it does not, you may have to
try some of the things mentioned in the link above or google "Hitachi
Fortran C". There may well be other C routines where something similar
will need to be done.

On Tue, 3 May 2005 Rossitza.Pentcheva at lrz.uni-muenchen.de wrote:

> Dear WIEN-developers,
>
> I am installing the latest WIEN Version on a Hitachi SR8000 Machine and
> have encountered the following problems:
> 1) the subroutine sortag.f is needed both for dstart and lapw2 but OBJS
> (or ORBJS1) in the correspoding Makefiles does not contain sortag.o.
> When icluded in the Makefile, everything is fine.
> 2) the Makefile for mini does not contain drmngwien.o which is called in
> haupt.f.
> OBJS1 =	dfpmin.o  errclr.o  errflg.o  finish.o  func.o    gbass.o \
> 	gtfnam.o  haupt.o   ininos.o  inv.o     latgen.o  lnsrch.o\
> 	maxstp.o  mini.o    mold0.o   mold.o    nn.o      nose.o  \
> 	nose0.o   nwtmin.o  outerr.o  pairdis.o rotate.o  rotdef.o\
> 	wrtscf.o  latmix.o  distxy.o
> #	inter.o   read3.o   integr.o  stern.o   rean3.o \
> #        rean0.o   setffx.o  charge.o  iffpar.o  normal.o choice.o \
> #	 lsqfit.o  mixin.o
> Including it in OBJS1 makes the inclusion of a further set of subroutnes
> necessary (actually a previous Version did contain them). Still, the
> following error message occurs (possibly system specific) during linking:
> f90 -o ./mini atomgrid.o mxpmgrid.o dfpmin.o  errclr.o  errflg.o  finish.o
>  func
> .o    gbass.o  gtfnam.o  haupt.o   ininos.o  inv.o     latgen.o  lnsrch.o
> maxstp
> .o  mini.o    mold0.o   mold.o    nn.o      nose.o   nose0.o   nwtmin.o
> outerr.
> o  pairdis.o rotate.o  rotdef.o wrtscf.o  latmix.o  distxy.o  mdble.o
> d1mach.o d
> itsum.o dr7mdc.o  dv7cpy.o    i7mdcn.o  divset.o drldst.o  dv7dfl.o
> dl7itv.o dr
> mng.o   dv7scp.o stopx.o   dl7ivm.o dv7vmp.o dl7tvm.o drmngwien.o dw7zbf.o
> da7ss
> t.o dl7upd.o  func2.o  dd7dog.o dl7vml.o  dv2axy.o    traptouch.o
> dd7tpr.o dpar
> ck.o  dv2nrm.o     -L../SRC_lib
> ld: 1012 error : Undefined symbol: .MDBLE
> *** Error code 255
>
> Stop.
> I would appreciate any help on that.
> Best regards,
> Rossitza Pentcheva
>
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>

-----------------------------------------------
Laurence Marks
Department of Materials Science and Engineering
MSE Rm 2036 Cook Hall
2220 N Campus Drive
Northwestern University
Evanston, IL 60201, USA
Tel: (847) 491-3996 Fax: (847) 491-7820
email: L - marks @ northwestern . edu
http://www.numis.northwestern.edu
-----------------------------------------------
-------------- next part --------------
/* Return in array double precision system parameters */
#include <float.h>

void mfloat(values)
float values[];
{
	values[0] = FLT_EPSILON;
	values[1] = FLT_MIN;
	values[2] = FLT_MAX;
	values[3] = FLT_MIN_10_EXP;
	values[4] = FLT_MAX_10_EXP;
}
void mdble_(values)
double values[];
{
	values[0] = DBL_EPSILON;
	values[1] = DBL_MIN;
	values[2] = DBL_MAX;
	values[3] = DBL_MIN_10_EXP;
	values[4] = DBL_MAX_10_EXP;
}
void mdble(values)
double values[];
{ mdble_(values);
}
void MDBLE_(values)
double values[];
{
        values[0] = DBL_EPSILON;
        values[1] = DBL_MIN;
        values[2] = DBL_MAX;
        values[3] = DBL_MIN_10_EXP;
        values[4] = DBL_MAX_10_EXP;
}
void MDBLE(values)
double values[];
{ mdble_(values);
}