Re: [patch 2.5] PCI: allow alternative methods for probing the BARs

[Benjamin Herrenschmidt]

On Mon, 2003-01-06 at 05:14, Linus Torvalds wrote:

> Alternatively, we could even have a very limited phase #1 that only 
> enumerates _reachable_ devices (ie it doesn't even try to create bus 
> numbers, it only enumerates devices and buses that have already been set 
> up by the firmware, and ignores bridges that aren't set up yet). A pure 
> discovery phase, without any configuration at all.

That would only work for cases where the kernel isn't responsible for
the actual bus numbering/renumbering. For example, since we do not quite
deal with PCI domains yet, I have to force pcibios_assign_all_busses()
on pmac so that my 3 root busses get re-numbered (else I get 3 trees
with conflicting bus numbers all starting at 0). Another case is embedded
which can deal with completely unassigned bus numbers.

Ben.