[CPMD-list] pgf77

Axel Kohlmeyer axel.kohlmeyer at theochem.ruhr-uni-bochum.de
Mon Jan 27 10:45:44 CET 2003


On Mon, 27 Jan 2003 10:26:25 +0100  Julia Khalack wrote:

> Dear Juerg,
> 
> I have only pgf77 version 3.1,
> and don't have any pgf90 compiler.
> 
> To compile CPMD  V3.7.0,
> I used PC-PGI option for  Configure.
> 
> Then I tried to compile it using
> PC-PGI-MPI option.
> The compilation was successful
> with the changes I reported before.
> 
> But MPI-executable crashes at the stage
> of reading Vanderbilt potential files.
> 
> The last few lines of the output are:
> -----------------------------------------------------------------
>   ***     DETSP| THE NEW SIZE OF THE PROGRAM IS PGFIO-F-217/unformatted 
> read/unit=22/attempt to read past end of file.
>   File name = ./Pot/008-O-gpbe--bm.uspp    unformatted, sequential 
> access   record = 2
>   In source file ./readvan.f, at line number 174
>     1848 kBYTES ***
> 
> p5_13531:  p4_error: net_recv read:  probable EOF on socket: 1
> ...
> -----------------------------------------------------------------
> 
> I'm not experiensed in unformatted read/write,
> so it looks strange for me that the first record
> read by MPI-executable from the potential file
> (the code line
>        READ(IFN) (IVER(I),I=1,3),(IDMY(I),I=1,3)
> in readvan.f)
> is the following:
> 117440512  50331648  67108864 167772160  33554432  33554432
> 
> while the numbers read by 1-processor executable are:
>   7     3    4     10   2     2
> 
> Could you help me with this?

hi julia!

this is simple. compare the two default configurations carefully.
the PC-PGI-MPI configuration includes the option '-byteswapio'
which enables to read binary/unformatted files from so-called 
'big-endian' machines (e.g. IBM, SGI, Sun and HP workstations).
get rid of that flag, 'make clean' , 'make' and you should have
a working parallel executable.

cheers,
	axel.
	

> 
> Sincerely yours,
> 		Julia
> 
> --------------------------------------------------------------
> Juerg Hutter wrote:
> > Dear Julia
> > 
> > I compiled CPMD V3.7.0 on my PC using the Makefile
> > generated by Configure for PC-PGI.
> > I used the pgf77 and pgf90 compilers (Version 3.2-4).
> > In both cases the program was compiled without any
> > problems.
> > 
> > Do you have a newer version of the PGI compiler?
> > Are your problems also there if you use pgf90?
> > Have you used the compiler options from the Configure
> > script?
> > Is anybody out there who has the same problems?
> > 
> > regards
> > 
> > Juerg
> > 
> 
> -- 
> Julia Khalack
> 
> Division of Physical Chemistry,
> Arrhenius Laboratory,
> Stockholm University,
> S-106 91 Stockholm, Sweden
> 
> Phone: +46-8-16 13 10
> Fax:   +46-8-15 21 87
> 
> E-mail: julia at physc.su.se
> 
> _______________________________________________
> 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.



More information about the CPMD-list mailing list