> At 11:59 AM +0100 on 26/1/2000, Martijn van Buul wrote:
> >
> > I do have one question though:
> >
> > From a hardware-point of view, these machines much more resemble a M68K
> > Mac than a "decent" PowerPC mac. Wouldn't it be a better idea to hack
> > m68kBSD into working, instead of adding NuBUS and non-OF support to
> > the PowerPC port? I assume that this has been asked before, but I couldn't
> > really find it on the archives of this mailinglist on www.netbsd.org.

I don't think so. We want to make the kernel on these machines behave for
programs the same way as the kernel on OF powermacs. The 

> Having done quite a lot of investigation into this, I think it's safe to say
> that both the 68K port and the PPC port are sufficiently different to
> NuBus-like Performa hardware to make the job quite difficult. Certianly,
> the reliance of OF in macppc is a big hurdle and would require some
> major changes. I can't say exactly how difficult the mac68K port
> would be but it wouldn't merge well to the code simply as they aren't
> 68K powered .
> 
> I know there may be objections to this suggestion but perhaps these
> machines are different enough to warrant their own fork?

NetBSD did that recently with the sun3x port. These machines were
basically sparcs but still w/ m68k CPU's (kinda the opposite of what we
have here). For a while, they were their own port. But after a while, it
was merged into the sun3 port.

I'd really recomend we do the same thing here.

Though it's not as bad as it might seem.

First off, we should see how mkLinux boots - what does it get passed in?

> Another problem is the rather severe lack of documentation on
> these machines.

Right. But sharing drivers w/ mac68k will help. :-)

Take care,

Bill