On Thu, 13 Jul 2006 18:37:31 -0400
Thor Lancelot Simon <tls@rek.tjls.com> wrote:

> Well, to be more precise, the ofppc port is said to run, sort-of,
> modulo severe OpenFirmware bugs in the only firmware that was ever
> made available to BSD developers.
Maybe I should mention that I have a major fsck for port-ofppc in the
pipeline. I removed all the platform code. Things that where done
hardcoded in the platform code at compile time (e.g. bus_space(9)) are
changed to dynamic code that adapt kernel subsystems to the hardware at
runtime based on information provided by the OpenFirmware. Goal is to
make hardware support truly generic instead adding platform code for
every machine. (Modulo machine specific quirks and drivers for machine
specific hardware.) I am now at the point where I have to add support
for real hardware interrupts / interrupt controlers for PCI, ISA, VME,
... devices. My main target at the moment is a RPA / CHRP style IBM
RS/6000 43P-150 and B50. I got distracted by my other hobbys so no
progress on the interrupt code.

If someone wants to have a look at my work:
http://www.unixag-kl.fh-kl.de/~jkunz/NetBSD/ofppc/ofppc-new.tar.gz
--=20


tsch=FC=DF,
       Jochen

Homepage: http://www.unixag-kl.fh-kl.de/~jkunz/