Hi Maurice:

Thanks for bringing CentOS into the discussion. That is a clone of Red
Hat Enterprise so I did some further searching to see what was going
on for g77 for the vanilla Red Hat Enterprise case. According to 
http://docs.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/6/html/Developer_Guide/compilers.html#gcc.compatibility.language
g77 _for enterprise 4.x_ is still available as a deprecated package
even for Enterprise 6 (and presumably Enterprise 5).  However,

"These packages are intended to be used as a temporary aid in
transitioning sources to newer compilers with changed behavior, or as
a convenient way to otherwise isolate differences in the system
environment from the compile environment."

"Please be advised that Red Hat may remove these packages in future Red
Hat Enterprise Linux releases."

I assume this same legacy enterprise 4.x package for g77 is also
available for Red Hat Enterprise 5.x and its CentOS clone.

Scientific Linux (another clone of RedHat Enterprise) handled the
g77-gfortran transition a different way.

For that case according to
http://www-zeuthen.desy.de/technisches_seminar/texte/sl5z.pdf, g77 was
dropped in favor of gfortran as of Scientific Linux 5.x.

On 2011-11-06 17:22-0700 Maurice LeBrun wrote:

> On Sunday, November 6, 2011 at 12:51:12 (-0800) Alan W. Irwin writes:
> > Does anyone here know the g77 package availability for enterprise
> > RedHat and SuSe?
>
> On my CentOS 5.x box, g77 is indeed part of the standard installation, with
> /usr/bin/f77 symlinked to /usr/bin/g77.

Which Fortran compiler (g77 or gfortran) do you use to build PLplot on
CentOS 5.x?  (Both the cmake output and output from "make VERBOSE=1"
should tell you that.) I believe if you have both the gfortran and g77
compilers available on your system, that our build system will choose
gfortran by default without you having to fiddle with the FC
environment variable.  But it would be good to get confirmation of
that (and also confirmation that gfortran produces good PLplot Fortran
results) for CentOS 5.x before we make any decision about dropping
support for the g77 compiler.

The most convenient run-time test you could do of our fortran bindings
and examples is to configure cmake with -DBUILD_TEST=ON and run the
test_diff_psc target, e.g., "make VERBOSE=1 test_diff_psc".

Note if we do decide to drop support for the g77 compiler, providing a
soft landing only requires a tiny modification to our build system. 
Right now we test if the Fortran compiler provides support for Fortran
95, and we set ENABLE_f95 to OFF for the case where there is no such
support.  But as far as I know g77 is the only Fortran compiler left
that provides Fortran 77 support but no Fortran 95 support so all we
have to do is set ENABLE_f77 to OFF whenever Fortran 95 is not
supported to make sure the g77 compiler is never used.

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________

Linux-powered Science
__________________________

------------------------------------------------------------------------------
RSA(R) Conference 2012
Save $700 by Nov 18
Register now
http://p.sf.net/sfu/rsa-sfdev2dev1
_______________________________________________
Plplot-devel mailing list
Plplot-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/plplot-devel

Reply via email to