build under HUPX v8.0

• To: netcdfgroup@xxxxxxxxxxxxxxxx (netcdfgroup)
• Subject: build under HUPX v8.0
• From: Rich Schramm <scri@xxxxxxxxx>
• Date: Fri, 7 Aug 1992 12:47:49 PDT

Please add me to your netcdf mailing list...

Also....

I have just ftp-ed the netcdf distribution over
to my hp9000s800 and am having dificulties getting
it to build under HUPX v8.0.  Its dying in when
testing under ncgen.  In xdb I can determine that
its giving a bus errors when trying to do an assignment
in module getfill.c (cdf_getfill) using values of FILL_DOUBLE
like:

     case NC_DOUBLE:
       gval->doublev = FILL_DOUBLE;
       return;

I believe that FILL_DOUBLE evaluates to xdr_d_infinity
which is type long cast to a double???   Anyway...
even adding an fprintf(stderr,"%lf\n",FILL_DOUBLE);
for debugging dies - saying FILL_DOUBLE is not a number.
The other assignments above such as FILL_FLOAT seem to work
OK.

(See also netcdf/src/netcdf.h for define FILL_DOUBLE and
 netcdf/src/array.c for xdr_d_infinity)

Does anyone have any ideas on what to try next??????
Any help would be greatly appreciated - Thanks

Rich Schramm
Monterey Bay Aquarium Research Institute
(408)647-3712