Hi Joseph,
I believe there was a strong reason for keeping the IID at 64 bits -
because that's what the reset of the IPv6 addressing architecture
tends to assume.
To drive the discussion, can you state why you prefer 'ff:fe'? Simply
to close to, but not exactly with RFC 4944? As I stated before,
trying to avoid global-scoped IEEE addresses is a non-issue since
short addresses have local scope.
--
Jonathan Hui
On Mar 30, 2010, at 10:13 AM, Reddy, Joseph wrote:
Hi Jonathan
I prefer option 1.
I still believe this should be paired with a 112 bit prefix and not
64-bit prefix as you suggest
-Joseph
On Mar 30, 2010, at 05:36, Jonathan Hui wrote:
Typing ...::ff:fe00:xxxx is a bit ugly, granted.
If we were designing 4944 from scratch, I would buy that argument.
However, it was agreed in May 2007, and I'd like to see a better
argument to change it incompatibly.
AFAIK, there are only a handful of independent implementations that
have tested interoperability to this day and even less with the use
of short addresses. W.r.t. 6lowpan-hc there was ambiguity over
whether or not to use the PAN ID in the IID. It seems that we need
to change something to better achieve interoperability. There's
already consensus to never include the PAN ID in the IID, so an
update to Section 6 of RFC 4944 is already necessary.
I believe the vast majority of implementors are subscribed to this
list. So which of the following options do people prefer?
1) IPv6 addrs that have the form aaaa::ff:fe00:xxxx.
2) IPv6 addrs that have the form aaaa::1:xxxx.
In both cases, the prefix and IID is 64 bits. At this point it's
just a matter of how many zeros are involved. It's a minor point, so
I wouldn't dwell on this issue too long.
--
Jonathan Hui
_______________________________________________
6lowpan mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/6lowpan
_______________________________________________
6lowpan mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/6lowpan
