[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
971111: installing NetCDF 3.3.1 under Linux using fort77(1)
- Subject: 971111: installing NetCDF 3.3.1 under Linux using fort77(1)
- Date: Tue, 11 Nov 97 10:35:34 -0700
Alec,
>Date: Tue, 11 Nov 1997 08:54:01 +0200 (SAT)
>From: Alec Joubert <address@hidden>
>Organization: University of the Witwatersrand/Climatology Research Group
>To: Steve Emmerson <address@hidden>
>Subject: installing NetCDF 3.3.1 using fort77
>Keywords: 199711051121.EAA12731
In the above message, you wrote:
> Hi Steve
>
> I think I've located my most recent problem.
>
> Among the thousands of warning messages which don't appear to be
> troublesome, there are a set like the one below. As I've said, netcdf
> seems to install fine, but errors in compilations of my code seem to refer
> to this warning message below. I've included the output from the
> ./configure command to give you an idea of what netcdf is using.
>
> All of these errors seem to occur when trying to "make all" in the fortran
> directory.
>
> Warning on line 196: inconsistent calling sequences for ncagt,
> arg 4: here integer*2 variable, previously integer*1 variable.
The netCDF Fortran routine NCAGT is part of the netCDF version 2
backward compatibility library and is used to obtain the value of a
netCDF attribute. The value may be one of several types (text, byte,
integer, etc.); consequently, the Fortran compiler doesn't like the fact
that the testing code uses different types to hold the returned value.
This is why the "-w" option to the fort77(1) utility is necessary.
>
> Other netcdf calls that are affected include:
>
> Warning on line 196: inconsistent calling sequences for ncagt,
> Warning on line 494: inconsistent calling sequences for ncapt,
> Warning on line 922: inconsistent calling sequences for ncvgt,
> Warning on line 1039: inconsistent calling sequences for ncvgt1,
> Warning on line 1126: inconsistent calling sequences for ncvpt,
> Warning on line 1217: inconsistent calling sequences for ncvpt1, ~
Same cause; same solution (-w).
--------
Steve Emmerson <http://www.unidata.ucar.edu>