> The reason being RFC4944 disallows the use of the "zero address" in the
> fake-48.

Does it?

I don't know what

   For either address format, all zero addresses MUST NOT
   be used.

actually means (it doesn't say whether this is about 16-bit, 48-bit or 64-bit 
addresses).
I think this was meant to exclude :: as the IID, but that may be just my 
interpretation.

> With no PAN-ID (ie: zero's for the PAN-ID bits), this disallows
> short address 0. Using some non-zero PAN-ID would remove that problem.

Even if fake-48 == '0'*48  is not allowed:
Is losing one out of 32k 16-bit addresses actually a problem?

> One could use the 0xFFFF PAN-ID, which would not be accidently considered
> valid.

That would be the result of interpreting

   if no PAN ID is known, 16 zero bits may be
   used

as "in non-associated mode, the PAN-ID is known to be 0xFFFF".
We could certainly agree on that, if we want.
That would be a change from current implementation practice, AFAIK.

> Or we could add a note to hc saying that ::ff:fe00:0 is indeed a valid
> address with this scheme. This would be the cleanest solution, I'm not sure
> if it's "allowed"?

If we really need that 32768th address, we could also do that.

Gruesse, Carsten

_______________________________________________
6lowpan mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/6lowpan

Reply via email to