[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