Re: Initialization order of PCI devices

From: James Sutherland (
Date: Thu May 25 2000 - 14:14:37 EST

On Thu, 25 May 2000, Mark H. Wood wrote:

> On Tue, 23 May 2000, David Woodhouse wrote:
> > said:
> > > Arghh hello world this is becoming a broken record. MAC addresses are
> > > NOT always per card.
> >
> > Not always, true - but if you _have_ to pick an identifying feature and you
> > for some reason don't want to use the order in which they're detected by the
> > kernel, then it's not a bad second option.
> >
> > Yes, someone will probably be able to find a box which both inits its
> > ethernet cards in a nondeterministic order and doesn't have unique MAC
> > addresses, but sorting on MAC would be sufficient for a lot of people, and
> > as it can be done in userspace, why not?
> Please tell us who has been intentionally issuing cards with non-unique
> UAAs.

UAAs? Sun have per-machine MAC addresses for Ethernet, I'm told; Token
Ring NICs have user-definable MAC addresses. This means the cards don't
have unique MAC addresses.


To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
Please read the FAQ at

This archive was generated by hypermail 2b29 : Wed May 31 2000 - 21:00:14 EST