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

Re: 19990602: netcdf "make test" failure under Red Hat Linux 2.2.5

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: Re: 19990602: netcdf "make test" failure under Red Hat Linux 2.2.5
  • Date: Wed, 02 Jun 1999 13:30:01 -0600
Graham,

>Date: Wed, 02 Jun 1999 18:49:13 +0000
>From: Graham Campbell <address@hidden>
>Organization: Brookhaven National Laboratory
>To: Steve Emmerson <address@hidden>
>Subject: Re: 19990601: netcdf "make test" failure under Red Hat Linux 2.2.5
>Keywords: 199905282013.OAA17107

In the above message, you wrote:

> This is a multi-part message in MIME format.
> --------------C4FFF63C40E7C8908E77CD02
> Content-Type: text/plain; charset=us-ascii
> Content-Transfer-Encoding: 7bit
> 
> Here is output from running t_nc under gdb:
> 
> (gdb) run
> Starting program: /home/gc/netcdf-linux/netcdf-3.4/src/libsrc/./t_nc
> 
> Program received signal SIGSEGV, Segmentation fault.
> 0x804f067 in nc_def_dim (ncid=6, name=0x2 <Address 0x2 out of bounds>,
> size=134714472,
>     dimidp=0x8079478) at dim.c:307
> 307     {
> (gdb) where
> #0  0x804f067 in nc_def_dim (ncid=6, name=0x2 <Address 0x2 out of
> bounds>, size=134714472,
>     dimidp=0x8079478) at dim.c:307
> #1  0x8048ced in createtestdims (cdfid=6, num_dims=2, sizes=0x8079468,
> dim_names=0x8079478)
>     at t_nc.c:116
> #2  0x80495b8 in main (ac=1, av=0xbffffa04) at t_nc.c:384
> #3  0x40031cb3 in __libc_start_main (main=0x8049394 <main>, argc=1,
> argv=0xbffffa04,
>     init=0x80487fc <_init>, fini=0x80741ec <_fini>, rtld_fini=0x4000a350
> <_dl_fini>,
>     stack_end=0xbffff9fc) at ../sysdeps/generic/libc-start.c:78
> (gdb)

Bizarre.  I don't see how the "name" and "size" arguments to
nc_def_dim() could be incorrect -- especialy on your system where the
size of a "size_t" is the same size as that of an "int" or "long" (4
bytes).  It only makes sense if the compiler misinterprets the meaning
of declarations like "char *foo[]".  As a consequence, I'm beginning to
think the impossible and suspect your C compiler.

What is the version of your C compiler, e.g.

    /usr/bin/cc --version

Do you have the "egcs" compiler available?  If so, what's its version,
e.g.

    /usr/bin/egcs --version

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