[CPMD-list] (no subject)

[Axel Kohlmeyer]

On Mon, 28 Jul 2003 11:57:24 +0100  "Gerardo Gonzalez" wrote:

> dear Axel
> i´m compiling for windows, power station makes is own 
> makefile... the elimination of IMPLICIT NONE was the first 
> thing i´ve tried, because the compiler reported a lot of 
> errors due to the needed of implicit declarations, the 
> pointers works... just de declaration is different from 
> that of most UNIX-based compilers, i was moving all the 
> possible options in the compiler, but nothing is able to 
> solve the above problems...

hmm, i still think your best shot would be configuring cpmd
for one of the pc-linux configurations, preprocessing the
sources to get compileable fortran code (you could use
e.g. the c preprocessor and make from the djgpp package
for that and set the fortran compiler to 'echo') and start 
the port to windows from there.

OTOH, installing linux on your machine would probably be
much less hassle. last time i looked you could still get the 
intel fortran compiler for linux free of charge (unsupported,
noncommercial version), same with the intel math kernel
library for linux and with that setup you would get
very good cpmd performance from your pentium iv.

granted, this approach has much less hack value...


cheers,
	axel.

> 
> i will tried to look into the libraries to tried to fix 
> the problem...
> 
> thanks...
> 
> On Mon, 28 Jul 2003 12:49:48 +0200
>   "Axel Kohlmeyer" 
> <axel.kohlmeyer at theochem.ruhr-uni-bochum.de> wrote:
> >
> >>>> "GG" == Gerardo Gonzalez <gerardo at cv.ua.pt> writes:
> >
> >GG> Dear Dr. Kohlmeyer,
> >
> >hello gerardo,
> >
> >GG> I have comment all the parallel and vector code, 
> >because 
> >GG> the MS Fortran Power Station don´t recognize the 
> >#ifdef
> >
> >you should not do that. the ifdefs are needed to select
> >code depending on architecture, compiler, parallization,
> >and included features. you should run the code through a 
> >c-preprocessor instead to produce the real fortran code.
> >
> >have a look at the makefile. each .F file is run though 
> >$(CPP) with $(CPPFLAGS) to produce the corresponding .f 
> >file.
> >you can try to install the GNU gcc compiler to do that.
> >
> >GG> statement, also the pointer declaration was changed 
> >(a 
> >GG> declaration per pointer word) because the type 
> >declaration 
> >GG> pointer (),() is also unrecognize, and I have 
> >eliminated 
> >
> >the pointer keyword is _essential_ for the dynamic memory 
> >management of cpmd. if your compiler does not support 
> >that,
> >you're out of luck. at least for the time being. sorry.
> >
> >GG> the IMPLICIT NONE statement because
> >GG> that sentence give me an error of implicit type 
> >needed... 
> >GG> is all
> >
> >that also does not sound too good, but it maybe a 
> >consequence
> >of the rather drastic changes you made to the source 
> >code.
> >
> >take care,
> >      axel.
> >
> >GG> cheers Gerardo.
> >
> 
> Gerardo Gonzalez
> Ph.D. Student
> Depto. Eng. Ceramica
> Universidade de Aveiro
> Aveiro 3810-193
> Portugal
> e-mail: gerardo at cv.ua.pt
> _______________________________________________
> CPMD-list mailing list
> CPMD-list at cpmd.org
> http://www.cpmd.org/mailman/listinfo/cpmd-list
> 

--

=======================================================================
Axel Kohlmeyer      e-mail:  axel.kohlmeyer at theochem.ruhr-uni-bochum.de
Lehrstuhl fuer Theoretische Chemie          Phone: ++49 (0)234/32-26673
Ruhr-Universitaet Bochum - NC 03/53         Fax:   ++49 (0)234/32-14045
D-44780 Bochum                   http://www.theochem.ruhr-uni-bochum.de
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.