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

Re: 20040630: netcdf-3.5.1 on Solaris x86

Matt,

>Date: Wed, 30 Jun 2004 15:36:02 -0400 (EDT)
>From: <address@hidden>
>Organization: MESO, Inc.
>To: <address@hidden>
>Subject: Re: 20040630: netcdf-3.5.1 on Solaris x86
>Keywords: 200406181713.i5IHDnWb026226

The above message contained the following:

> Ok, they had gcc on there and we were able to get "make" to run to
> completion.  I attached make.log, but there were a couple of warnings that
> did not get piped into that file:
> 
> alonsom@acmes3:/data3/cm-testing/matt/wrf/netcdf-3.5.1/src>make > make.log
> Warning: ncgentab.c is out-of-date with respect to ncgen.y
> Warning: It should be recreated via yacc on an OSF/1 system
> Not making `cxx/all' because no C++ compiler

The above warnings can be safely ignored.  You should strive to redirect
both standard output and standard error streams to the same file,
however.

> One thing I found out was that I needed to use '-tp p6' as an FFLAG
> because of the manner in which everything is setup on their system.  I'm
> wondering if their is a similar CFLAG that is needed, maybe you have heard
> of one that does the same thing.

Sorry.  I've never heard of the "-tp p6" option.

> When I tried "make test" it crashed (see
> make_test.log).

The output from "make test":

    Making `test' in directory 
/data3/cm-testing/matt/wrf/netcdf-3.5.1/src/libsrc

    /usr/local/bin/gcc -c -O -I.  -DNDEBUG -DpgiFortran t_nc.c
    ar cru libnetcdf.a attr.o  dim.o  error.o  libvers.o  nc.o  ncio.o  ncx.o  
putget.o  string.o  v1hpg.o  v2i.o  var.o
    ranlib libnetcdf.a
    /usr/local/bin/gcc -o t_nc -O  t_nc.o -L. -lnetcdf
    ./t_nc
    dimrename: IXX
    nc_close ret = 0

    reopen id = 3 for filename test.nc
    NC done
    t_nc.c:491: failed assertion `nc_inq_att(id, NC_GLOBAL, adesc->mnem, 
&(adesc->type), &(adesc->len))== NC_NOERR'
    GATTR *** Signal 6
    make: Fatal error: Command failed for target `test'
    Current working directory /data3/cm-testing/matt/wrf/netcdf-3.5.1/src/libsrc
    *** Error code 1
    make: Fatal error: Command failed for target `subdir_target'
    Current working directory /data3/cm-testing/matt/wrf/netcdf-3.5.1/src
    *** Error code 1
    make: Fatal error: Command failed for target `libsrc/test'

makes no sense because the netCDF package has been built and tested
in this way on hundreds (if not thousands) of Linux systems using the
gcc(1) compiler.

Please send me the output from the following commands:

    uname -a
    /usr/local/bin/gcc -v

Also, please send me the file "configure.log" IF it contains both the
standard output and standard error streams.  If it doesn't, then please
recreate it first by re-executing the configure script and redirecting
both streams into the file.

Regards,
Steve Emmerson