On Tue, 31 Aug 1999, Dan Winship wrote:

> > Well the problem is more that we assume that OF has configured the card
> > (set up the PCI mappings) when infact it hasn't.
> 
> OF configures devices at startup time, not when you open them. (If you
> browse the device tree in OF before booting, you'll see that all of
> the addresses and interrupts have already been assigned.)

Ahhh. I thought the problem was they weren't being assigned.

> (Oh, and bm is a built-in device attached to obio, not a PCI card.)
> 
> > To teach NetBSD how to do this in general is a big job. :-(
> 
> I'm not convinced. Linux and Darwin don't have this problem (I think?
> Can anyone confirm?) I suspect we're doing something wrong somewhere
> (failing to set some grackle-specific status bit or something) which
> is causing these problems.

If that's the problem, you're right it shouldn't be that hard to fix. :-)

Take care,

Bill