Re: more on the isa IO panic

[Jochen Kunz <jkunz%unixag-kl.fh-kl.de@localhost>]

On Wed, 27 Jun 2007 11:35:39 -0700 (MST)
Tim Rightnour <root%garbled.net@localhost> wrote:

> Ok.. let me try to explain the way the ofwoea_map_space function
> works, in the context of isa.  It's possible I made some incorrect
> assumptions here.. but I'm hoping it's correct..
I don't have time to investigate your work in detail at the moment.
Maybe you can look at my patches to the old ofppc code, file
ofppc/pci/ofpci.c. There are some comments about that problem in front
of the function init_spaces(). This function also does the guesswork to
sort out which PCI space maps to which ISA space.

About the console IO stuff: You can delay attachment of the console
driver until attachment of com(4). port-sparc does it that way and
port-hp700 does similar things. It just uses the OFW console IO
functions until it finds a "better" driver. This is also nice as you
don't need a working bus_space(9) very early in bootup just for console
IO. You can wait with the initialization of bus_space(9) until the PCI
host bridge gets attached during autoconf(9).

Here is my old stuff:
http://www.unixag-kl.fh-kl.de/~jkunz/NetBSD/ofppc/ofppc.tar.gz
-- 


tschüß,
       Jochen

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