Re: GNAT and GCC 3.0

[Stephen Leake <stephen.a.leake.1@gsfc.nasa.gov>]

Florian Weimer <fw@deneb.enyo.de> writes:

> dewar@gnat.com (Robert Dewar) writes:
> 
> > Although it can sometimes be the other way round if we fall a few
> > days behind in keeping the trees synchronized.
> 
> Do you really think "a few days behind" is an accurate description of
> the current situation?

Since Robert is in a position to know, this seems a silly question.

I'll go out on a limb and guess that Florian is complaining that the
current public binary release of GNAT (3.13p) is 18 months behind the
GCC CVS tree. That is true. But I believe Robert was talking about
keeping the GCC CVS tree in sync with the internal GNAT development
tree (currently heading for a 3.15 release). I'm actually surprised
that ACT is doing that, but it should certainly satisfy anyone who
wants access to the "cutting edge" GNAT.

I hope ACT continues to make binary releases, at least for Windows and
Solaris, even if they are "way behind". The various Linux maintainers
could start making binary releases for their systems. Even if I want
the "latest" GNAT from the CVS tree, it helps tremendously to have a
known good binary to install first, since that is required to compile
the CVS release.

-- 
-- Stephe