[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [MiNT] new FreeMiNT release (was Re: 030-version of gcc)

Well, if we need things recompiled, give me part of the list, and the
tools which should be used (version of GCC, etc), and I can compile, and
rebuild RPMs as needed.  I can't say I would be too much help, at least
not right now, in debugging any problems in the kernel itself.

Might as well put my 68060 to work.  I'm toying with newer versions of
Glib/GTK and GiMP somewhat (trying to take advantage of our XServer!),
also trying to get TRN built, toying with Firebird, and trying to figure
out a good way to assemble a PHP and MODSSL package, but that stuff can
always wait.  Also trying to figure out what I did wrong with Matthias's
060 optimized GCC, but that is a whole different story.

So, let me know how I could assist.

As for Milestones, I would say that January 1 would be a much better date
for any release, as Decemeber 1 is way to close.  Maybe push everything
back a month if that timeline is to be followed.

And, the GCC issue.  I see both arguements, and let me throw in my
thoughts, not that this is anything big.

If you look at the Unix world right now, MOST applications require, as a
minimum, GCC 2.95.2, which we fortunately have right now.  2.95.3 is not
too large, and has been shown to be reliable, so it is good to have.

Now, I do see GREAT benefit in keeping up with the GCC 3.3/3.4 series
compilers.  The biggest concern is that utilizing them will leave people
with lesser resources behind.  While this is an important concern, I also
believe that for the life of FreeMiNT, we need to keep up with the rest of
the GNU world.  Yes, GCC 3.3.x is a beast to compile, and there are issues
that are unresolved.  Hopefully someone would undertake the challenge and
resolve the remaining issues, which should at least keep us in the loop
with GCC 3.3x and most likely GCC 3.4.

An advantage the M68k series of processors have over other CPUs, except
Intel, is the widespread use they enjoy in MANY applications, and because
of that widespread use, work is still being done in optimizations for the
M68k series.  Maybe we will see some noticible changes, but on the
downside, maybe such changes will never happen.