On Thu, Jul 29, 1999, Dan Winship <danw@MIT.EDU> wrote:

>This problem only happens if Open Firmware touches the IDE controller.
>So if you netboot, you'll be fine, but if you boot off the IDE disks,
>you'll lose. So it seems like an ofw bug, but Darwin and MacOS somehow
>don't lose the same way...

I beleive that whatever OF do, it must be possible to fix it later by
software.

I don't have one of those beasts here so I can't test, but it would be
interesting if you could get a doc of the chipset, and dump various
regisers (both config space and i/o) with and without netboot and then
compare.

Once that's done, specific fixup code may be added to the driver.

-- 
           E-Mail: <mailto:bh40@calva.net>
BenH.      Web   : <http://calvaweb.calvacom.fr/bh40/>