[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

20020515: netcdf on UNICOS: Fortran-90 8-byte integers



Verena,

>Date: Wed, 15 May 2002 16:04:08 +0200
>From: Verena Cals <address@hidden>
>Organization: Stratosphäre / Forschungszentrum Jülich GmbH
>To: Steve Emmerson <address@hidden>
>Subject: Re: 20020514: netcdf on UNICOS: non-support of primitive F90 types
>Keywords: 200204250907.g3P97sa28336

The above message contained the following:

> I have verified the assertions, but it does not apply anything. The
> first, second and fourth assertion is correct. 
> 
> 3. But in netcdf-3.5.0/src/f90/netcdf_overloads.f90 there are only the
> following subroutines for handling 1-D vectors: 'nf90_put_var_1D_text'
> and 'nf90_put_var_1D_EightByteReal'. There is no
> 'nf90_put_var_1D_EightByteInt'.
> 
> 5. 'nf90_put_var_1D_EightByteInt' is NOT defined in
> netcdf-3.5.0/src/f90/netcdf_expanded.f90
> 
> 6.Because 'nf90_put_var_1D_EightByteInt' does not exist, it is not able
> to handle 1-D integer vectors of type-kind 'EightByteInt'
> 
> I have also verified the assertions above in netcdf-3.5.1-beta3 (without
> ..._OneByteInt, ..._TwoByteInt, ..._FourByteInt and
> ..._FourByteReal in src/f90/netcdf_overloads.f90). There everything is
> correct, but executing configure seems to be not complete.
> 
> Should I integrate the missing assertions (3., 5. and 6.) in
> netcdf-3.5.0?

I was working with a more advanced version of the netCDF package --
apparently, one that already supports 8-byte integers in the Fortran-90
interface. We're not sure if this capability will make it into the
next release, however, because we're not sure that ALL Fortran-90
implementations support 8-byte integers.

Try making the modifications to the files netcdf_overloads.f90
and netcdf_expanded.f90 and see if that works on your system.  If it
does, then that is a good argument for putting those changes in the next
release.

Regards,
Steve Emmerson   <http://www.unidata.ucar.edu>