Re: Version numbers

[Fergus Henderson <fjh at cs dot mu dot oz dot au>]

On 17-Feb-2001, Toon Moene <toon@moene.indiv.nluug.nl> wrote:
> Fergus Henderson wrote:
> 
> > On 16-Feb-2001, Toon Moene <toon@moene.indiv.nluug.nl> wrote:
> 
> > > "Joseph S. Myers" wrote:
> 
> > > > * CVS date.
> > > > * GCC version (3.0, etc.).
> > > > * g77 version (0.5.26, etc.).
> > > > * CHILL version (1.5.2, etc.).
> > > > * gcov version (1.5, etc.).
> > > >
> > > > and any others.
> > >
> > > This sounds like a *very* good idea to me !
> > 
> > I'm not so sure that this is a good idea.  Doing it this way would
> > mean that adding a new front-end can't be done by just providing new
> > files in a new directory.  Instead, patches to the language-independent
> > stuff would be required.  Furthermore, such patches are very likely to
> > have conflicts.  That could be a real pain.
> 
> Ah - you're clearly from a different church than I am.  The first
> two-and-a-half years of its public existence, g77 carried a perpetually
> growing "GNU Back End" patch just to get in some basic functionality we
> needed or to be able to adapt optimisation heuristics to the Fortran
> Reality.

For the Mercury front-end, so far I've needed only one small patch,
<http://gcc.gnu.org/ml/gcc-patches/2001-01/msg00577.html>,
to allow the front-end to define main(). 
(Incidentally, that patch is still unreviewed...)

I also have a change planned to allow the front-end to tell the
back-end when it should do tail recursion/sibling call optimization.
But apart from that, I don't anticipate any Mercury-specific patches
to the gcc back-end code.  And I hope that those two changes can be
incorporated in the gcc cvs repository, so that we won't need any
Mercury-specific patch kit at all.

-- 
Fergus Henderson <fjh@cs.mu.oz.au>  |  "I have always known that the pursuit
                                    |  of excellence is a lethal habit"
WWW: <http://www.cs.mu.oz.au/~fjh>  |     -- the last words of T. S. Garp.