Re: netcdf-4.0-beta1 release now available on netCDF web site...

NOTE: The netcdf-hdf mailing list is no longer active. The list archives are made available for historical reasons.

Jeff Whitaker wrote:
Roy Mendelssohn wrote:
Hi Jeff:

Are you building it with Fortran enabled. If so, can you tell me what gfortran you are using, and what other settings you have. It fails for me on OSX.


-Roy M.
On Apr 23, 2007, at 11:35 AM, Jeff Whitaker wrote:

Ed Hartnett wrote:
Howdy all!

With the recent release of HDF5 1.8.0-beta1, it is now possible for us
to release netCDF-4.0-beta1, and it is now available from the netCDF
web site:

The netCDF-4.0 beta release requires HDF5 1.8.0-beta1, and
zlib-1.2.3. These can be obtained from the netCDF-4 FTP site:

Build instructions can be found here:

Documentation for this release can be found on the netCDF beta
documentation page:

For more information about this release, see the netCDF-4.0 beta
release notes:

Recent changes in netCDF-4.0 include the use of the new creation
ordering capabilities of HDF5 1.8.0. Since this feature was added to
HDF5 very recently, the HDF5-1.8.0 beta1 release is now required for
netCDF-4.0; previous versions of HDF5 will no longer work.

Build output from a variety of platforms may be found at the netCDF-4
snapshot build page:

The version 4.0 release of netCDF includes many new features, with a
correspondingly higher risk of bugs in the code and the
documentation. Please report any problems to



Roy:  gfortran doesn't work for me either on OS X (error below):

libtool: compile: gfortran -I../libsrc4 -I. -g -O2 -c netcdf4.f90 -o netcdf4.o
   Included at netcdf4.f90:43:

    nf90_fill_ubyte  = Z'FF',                           &
Error: Arithmetic overflow converting INTEGER(8) to INTEGER(1) at (1)
   Included at netcdf4.f90:43:

    nf90_fill_ushort = Z'FFFF',                         &
Error: Arithmetic overflow converting INTEGER(8) to INTEGER(2) at (1)
   Included at netcdf4.f90:43:

    nf90_fill_uint   = Z'FFFFFFFF'
Error: Arithmetic overflow converting INTEGER(8) to INTEGER(4) at (1)
make[2]: *** [netcdf4.lo] Error 1
make[1]: *** [all-recursive] Error 1
make: *** [all] Error 2

which looks like a legitimate error to me.

Setting FC=g95, F77=g95 compiles, but make check fails with a linker error (not finding symbols with two underscores).


Roy: I've got it to work with gfortran - the trick is to use FFLAGS = -fno-range-check.

If you're getting an error in configure that says "error: Fortran could not compile .f90 files", your gfortran is too old. You can update by installing the gcc42 fink package, or grab a binary from


Jeffrey S. Whitaker         Phone  : (303)497-6313
Meteorologist               FAX    : (303)497-6449
NOAA/OAR/PSD  R/PSD1        Email  : Jeffrey.S.Whitaker@xxxxxxxx
325 Broadway                Office : Skaggs Research Cntr 1D-124
Boulder, CO, USA 80303-3328 Web    :

To unsubscribe netcdf-hdf, visit:

  • 2007 messages navigation, sorted by:
    1. Thread
    2. Subject
    3. Author
    4. Date
    5. ↑ Table Of Contents
  • Search the netcdf-hdf archives: