Re: [PATCH] Fixup for Amigas with memory at 0x200000

[Ilpo Ruotsalainen <lonewolf%iki.fi@localhost>]

On Fri Jan 10 2003 at 18:41:42 +0200, Ilpo Ruotsalainen wrote:
> On Wed Jan 08 2003 at 21:27:20 +0100, Aymeric Vincent wrote:
> > is%netbsd.org@localhost (Ignatios Souvatzis) writes:
> > 
> > > This sounds good. The DraCo is guaranteed to have at least a 68040V CPU
> > > (thats a CMOS 68LC040), which uses the same ITTR + DTTR format and 
> > > registers
> > > as the 68060. Actually, I'm using the same method to switch _off_ 
> > > the MMU at reboot time! Just use my code backwards ;-)
> > 
> > Great. Here is an updated patch which should handle the DraCo.
> > As far as the Draco and 68040 Amigas are concerned, I can just
> > guarantee that "it compiles".
> > 
> > [ the only change from the previous patch is the code in locore.s
> >   which is #ifdef DRACO ]
> > 
> > > Make it so. I'll test on DraCo, and 68040 Amiga. (I don't really want to
> > > pull out the accellerator to test on 68030 Amiga, though).
> > 
> > I tested this patch with -current on my '030 A1200 with both a GENERIC
> > kernel and a stripped down kernel where M68040 and M68060 are
> > undefined.
> 
> Just tried booting my '060 A1200 with patched -current, gray screen
> nothing else... (I know the -current is otherwise good, so it must be
> this patch.)
> 
> Any tips on how to debug the early startup code on amiga? I'm quite new
> to NetBSD (and non-AmigaOS stuff on amiga in general) and I have
> absolutely no ideas how to debug this.
> 
> (Perhaps I could write AGA registers to change the screen color, hmm
> that might work...)

I know this much now: It certainly finishes start_c(). It either doesn't
ever enter start_c_finish() or CUSTOMADDR (or the mappings) are hosed.

-- 
Ilpo Ruotsalainen - <lonewolf%iki.fi@localhost> - http://www.iki.fi/lonewolf/