Re: [MiNT] Still don't get it - difference between 68020 and 68020-60

[Alan Hourihane <alanh@fairlite.co.uk>]

On Mon, 2010-06-21 at 09:12 +0200, Miro Kropacek wrote:
> > For me this discussion is about making GCC do what is capable, not the
> > fact whether packages, package maintainers, or developers make use of
> > that support or not. But it gives that flexibility when the time arises.
> But the rpm/bzip/whatever package size will increase drastically (add
> ~8 MB for each new target), it's not only about libgcc but also about
> c++ libs! I really can't agree with 7*2 targets just because "we can".
> How many of you would use these new configurations for your builds?
> It's quite rare case there's m68020-60 version provided for some
> application. I think we should focus on what developers need. If
> Patrice and possibly others think m68030/soft-float is worth of it /
> usable, ok, but please, not another 6*2 targets. As I see it,
> currently we need / use: 68000/mshort, m68030/mlong, m68020-60/mlong
> and m5475/mlong. In case gcc patch could somehow provide this and only
> this set, I celebrate :)

You are talking about packages again. I'm not. Increasing the size of
the package only happens if the package maintainer wants to produce.
He/she can produce a m68000 compiled package and it should work on all
Atari's, and completely ignore the other capabilities of GCC.

And we need -mshort for all targets because the kernel requires -mshort.

Alan.