[molpro-user] 2006.1 compilation, main_cvb.f, iunset

Nick Wilson WilsonNT at Cardiff.ac.uk
Wed Jun 7 13:05:58 BST 2006


Dear Mikael,

It looks as if this happens if you don't select the default during 
configure for:
  Enter max number of states per symmetry
This piece of code is still hard coded to the default of 20.

I'm building a patch now and will deposit it when it has been fully tested

Best wishes,
Nick Wilson

Mikael Johansson wrote:
> 
> Hello!
> 
> I'm compiling 2006.1 (patchlevel 1) on various systems, and some 
> complain about line 295 in src/casvb/main_cvb.f :
> 
>       data iunset/0,1,1,1,1, 1,1,1,1,1, 1,1,1,1,1, 1,1,1,1,1,
>      >            0,1,1,1,1, 1,1,1,1,1, 1,1,1,1,1, 1,1,1,1,1/
> 
> For example f90 on Solaris 9 says:
> main_cvb.f: preprocessing ... compiling ... failed:
> f90 -dalign -xtypemap=real:64,double:64,integer:64 -c -I. -M../global 
> -fast -ftrap=%none -xarch=v9b -fsimple=2 main_cvb.f
> 
>       data iunset/0,1,1,1,1, 1,1,1,1,1, 1,1,1,1,1, 1,1,1,1,1,
>            ^
> "main_cvb.f", Line = 295, Column = 12: ERROR: The number of entities to 
> be initialized exceeds the number of values.
> 
> A similar message appeared with PGI pgf90 6.0-4 on Linux/Opteron. 
> Pathscale 2.4, on the other hand, didn't complain. It also compiled 
> without problems on IBM AIX and xlf.
> 
> Any hints on a workaround/fix?
> 
> Have a nice day,
>     Mikael J.
> 
> -- 
> Mikael Johansson, application specialist, science support, chemistry
> CSC - IT Center for Science, P.O. Box 405, FI-02101 Espoo, Finland
> Tel.: +358-9-457 2934, www.csc.fi, e-mail: mikael.johansson at csc.fi
> 



More information about the Molpro-user mailing list