This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Re: does gfortran backport any g95 code?
- From: Steve Kargl <sgk at troutmask dot apl dot washington dot edu>
- To: Jack Howarth <howarth at bromo dot msbb dot uc dot edu>
- Cc: fortran at sources dot redhat dot com
- Date: Mon, 25 Jul 2005 18:52:27 -0700
- Subject: Re: does gfortran backport any g95 code?
- References: <20050726002444.D71A51DC06D@bromo.msbb.uc.edu>
On Mon, Jul 25, 2005 at 08:24:44PM -0400, Jack Howarth wrote:
> Do the gfortran developers ever look at the current g95 source
> tree for opportunities to backport code (specifically related to
> intrinsics)? I ask because I have just finished building xplor-nih
> (a complex mix of c, c++ and fortran) on MacOS X 10.4 using the
> Apple Xcode 2.1 gnu gcc/g++ compilers and the current g95 compiler.
> The resulting xplor-nih build passes all of its testsuite. However
> I can't build the same package with gfortran 4.0.1 on Fedora Core 4
> due to the absence of the LOC intrinsic. It seems a shame that, in the
> area of intrinsics, gfortran can't leverage the current work from
> g95.
g95 and gfortran have diverged to the point were porting
code from g95 is a waste of time if it is outside of the
scanner/parser/matcher code. That is, the internal handling
of intrinsics in gfortran is much different than what is
done in g95. The lack of comments in g95 and Andy's old
practice of abfuscating his code is also a hinderance (
which I'm sure pleases Andy).
--
Steve