--zbGR4y+acU1DwHSi
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Tue, Dec 14, 2004 at 09:02:34PM -0500, Tim Kelly wrote:
> At 5:46 PM -0800 12/14/04, Bill Studenmund wrote:
> >> but the execution instructions were last at 0x600xxx. This looks a lot
> >> like a mismatch between bootxx and ofwboot. bootxx 1.12 would have
> >> loaded ofwboot to 0x600000 but ofwboot 1.10 (?) is linked to 0xe00000,
> >
> >bootxx 1.13 loads ofwboot to 0xe00000.
>=20
> Sure, but since ofwboot was showing up at 0x600000, wouldn't this indicate
> that the bootxx version on the ISO is 1.12? With ofwboot 1.10 linked to
> 0xe00000, it jumps off track as soon it has to start loading any absolute
> addresses.

[snip]

> Although his load-base is 0x600000, bootxx doesn't use load-base and
> instead writes to a hardwired address. For 1.12 this was 0x600000 and for
> 1.13 it is 0xe00000, to match the linking address of ofwboot (background
> for others reading this). It seems to me the only way we could see the
> above data in memory is if bootxx 1.12 loaded ofwboot 1.10.

Sorry, I thought I'd said that.

=46rom looking at the logs, the CD actually has bootxx 1.5 or so.

Luke and I dug into things, and there is a lot of magic in what's going=20
on. For now, it doesn't look easy to pull in a new bootxx.

Take care,

Bill

--zbGR4y+acU1DwHSi
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (NetBSD)

iD8DBQFBv6i4Wz+3JHUci9cRApPaAJ9gyPEYc1SnbO3Iag911o/hoC1kRACfRH11
yb4FvIs6SLz+2CY+mOQPCmQ=
=krpf
-----END PGP SIGNATURE-----

--zbGR4y+acU1DwHSi--