RE: Re: Re: AmigaOS 3.2 kickstart ROM crashing on NetBSD partitions (Was NetBSD/amiga sysinst)

  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.

--8323329-1168949730-1679741988=:2793
Content-Type: TEXT/PLAIN; charset=utf-8
Content-Transfer-Encoding: 8BIT

Hi,

On Sat, 25 Mar 2023, Carlos Milán Figueredo wrote:

> > As I wrote on EAB, this _recoverable_ alert is probably caused
> > by the boot code returning with an error (read: rts while d0 non 0?)
> > to the OS. Just by quickly glancing at it, the boot code makes some
> > assumptions regarding the stack and register layout the boot code
> > is called with, then proceeds to allocate some memory and other
> > things. I did not check if these are "hard truths", or it just
> > happens to work with the previous Kickstart versions.
>
> Although that makes sense, why the bootblock works when running with the
> runbootblock utility from AmigaOS 3.2? The Recoverable Alert will only
> happen when trying to boot NetBSD from the Early Startup Menu. Is there
> something different in the memory layout before actually booting
> AmigaOS?

OK, so my initial hypothesis was wrong, but now I have proof, that on
Kickstart 3.2.x opening console.device fails from the bootblock, in
console.c, function consinit(). That OpenDevice("console.device"...)
call in particular, returns -1:

Video evidence:
https://twitter.com/chainq/status/1639581705332748291

Gist with my patch, to show the evidence:
https://gist.github.com/chainq/0b1a628916a32ee6d93d9749de2d087c

I let the people who actually have interest in these newer Kickstarts
figure it out. :P

I will also post on EAB.

Charlie
--8323329-1168949730-1679741988=:2793--