Re: 1111 1110 10 equals 0xfe80 to 0xfebf

[Simon Hobson <[e-mail]>]

Alexandre Petrescu <[e-mail]> wrote:

>>> If I have prefix fe80::/10, as described in RFC 4291, the next bit is bit 11. Doing
>>> the same subdivision of the prefix is fe80::/11 and fea0::/11.
>> [Dusan] The hexadecimal definition for LL address is not syntactically correct. The binary 10 bit prefix 1111111010 cannot be presented as hexadecimal FE80::/10. It is rather a range FE80::/10 - FEBF::/10. In this notation, FE80::/10 = FEBF::/10, 
> 
> I agree.  The IPv6 link-local prefix is not one hextet, but a range of hextets: FE80::/10 to FEBF::/10.
> 
> Maybe one can come up with a base in which that range of hextets is represented as just one number.

We don't need any new representation.
As others have already said, this thread has gone on far too long, and is really starting to seem like trolling. It is hard to understand how someone involved in networking does not understand CIDR, and someone involved in an IPv6 working group doesn't understand standard IPv6 CIDR notation. I'm sorry if that sounds like an insult, but these are pretty basic concepts.

There is no problem, there is no ambiguity.

Consider the IPv4 notation where someone says "subnet 123.123.252.0/22". That's unambiguous, meaning a subnet 0111 1011 0111 1011 1111 11 expanded to 32 bits with 00 0000 0000 0000, or in hex, FB FB FC 00/22. The fact that the subnet encompasses addresses from 123.123.252.0 through to 123.123.255.255 doesn't alter that, and we don't insist on saying it's subnet 123.123.(252 to 255).(0 to 255). Everyone knows exactly what is meant - if they understand CIDR, which sadly too many networking people don't (you wouldn't believe the number of people who don't understand anything but a /24 !)

In IPv6 notation, the rules are equally clear. FE80::/10 means exactly one thing - the TEN bits 1111 1110 10 followed by 118 zeros. As a prefix (analogous to a subnet in IPv4) that is clear, unambiguous, and understandable by anyone who understands the basics of IPv6 notation. The fact that the prefix can encompass addresses from FE80:: through to FEBF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF does not change the fact that the PREFIX is still FE80/10. So no, it is not necessary to say FE(80 through BF):<stuff>/10 and it would be very confusing to try and do so.

And all that is completely irrelevant to whether any address is link local or not. For example, an address FE81:1:<something> is still within the prefix FE80::/10. So if you wish to take the view that a link local address is allowed to use all 118 bits then the prefix is still FE80/10 and it is still valid to say that a link local address is in the FE80/10 subnet. So whether 54 bits are all zero or can be non-zero (and thus the hex representation can be anything from FE80:stuff to FEBF:stuff) doesn't change that basic notation.

> What is a nub?

It's a manner of speech in some areas, roughly translated it means "the essential point of the issue".