How is the Ethernet MAC address for usb0 defined?

[Mikko J Rauhala]

On to, 2007-08-09 at 09:07 -0400, Dr. H. Nikolaus Schaller wrote:
> > Bluetooth MAC into a different link-local MAC usable on the USB
> > interface would be as simple as to flip the second least significant
> > bit of the most significant byte into a 1¹.
> 
> That sounds pretty simple - and IMHO the probability of conflict is  
> as low as the number of Bluetooth devices out there...

Actually much less than that. First, there is absolutely nil chance of
conflict with any device that translates any universal MAC (or EUI-48)
into a link-local MAC via the same means; thus at least no conflicts
between Neos, ever (unless one would override this MAC policy, of
course).

Conflict with any given (pseudo)randomly assigned link-local MAC has a
probability of 1/2^46, the same as if it were (pseudo)randomly assigned
on the Neo as well. (48 bits minus the uni/multicast bit and the
OUI/link-local bit which are fixed for our purposes.)

-- 
Mikko J Rauhala <mjrauhal at cc.helsinki.fi>
University of Helsinki