According to RFC2464, the IPv6 Multicast address will be
mapped to the IEEE802 address as:

>7.  Address Mapping -- Multicast
>
>   An IPv6 packet with a multicast destination address DST, consisting
>   of the sixteen octets DST[1] through DST[16], is transmitted to the
>   Ethernet multicast address whose first two octets are the value 3333
>   hexadecimal and whose last four octets are the last four octets of
>   DST.
>
>                  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>                  |0 0 1 1 0 0 1 1|0 0 1 1 0 0 1 1|
>                  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>                  |   DST[13]     |   DST[14]     |
>                  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>                  |   DST[15]     |   DST[16]     |
>                  +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>

And to respond neighbor solicitation, you have to be able to receive
the corresponding multicast address for your IPv6 address.
It is defined in RFC2373 as:

>   The above multicast addresses identify the group of all IPv6 routers,
>   within scope 1 (node-local), 2 (link-local), or 5 (site-local).
>
>      Solicited-Node Address:  FF02:0:0:0:0:1:FFXX:XXXX

So if your node have MAC address 00:12:34:56:78:9A, then your IPv6 link-local
address will be FE80::212:34FF:FE56:789A, the corresponding
neighbor-solicitation multicast address is FF02::1:FF56:789A,
which will be sent on Ethernet as 33:33:FF:56:78:9A.
Your Ethernet controller have to be able to receive that multicast address
as well as "all-node" multicast 33:33:00:00:00:01(as FF02::1).

Since IPv6 node is supporsed to be able to assign or remove multiple IPv6
address dynamically, so that the Ethernet controller(and the API) should be
able to enable/disable perticuler muticast address.

Unless link-local or stateless-autoconfigured, the IPv6 address is not
directry related to tha MAC address. If a usesr configured additional
IPv6 address on your node, say 3a:5678:9ABC:DEF0, then your Ethernet
controller have to be able to receive multicast address 33:33:FF:BC:DE:F0
as long as 33:33:00:00:00:01 and 33:33:FF:56:78:9A.

Remember, sometimes the neighbor-solicitation multicast address may conflict
(because only last three digits are reflected). For example, when your
node gets stateless-autoconfigured address, say FEC0:1::212:34FF:FE56:789A,
the neighbor-solicitation multicast address is same as the link-local address.
Then when the stateless-autoconfigured address is expired, you SHOUDL NOT
disable to receive 33:33:FF:56:78:9A because it is still used for the
link-local address!




>Hello everyone,
>  One of the things that is stinging me is whether current ethernet
>adapters would have to some how modify themselves for the IPv6 traffic,
>specifically in the case of neighbour discovery on the Solicited node
>Multicast address. Or by default we should configure the Adapter to
>listen and pass the multilink traffic corresponding to the multilink
>address with lower 24 bits the same as its own MAC address. 
>  Please clarify to me, in detail if possible, this (in!)dependence of
>the MAC address to the IPv6 address. (As i understand the routers have
>to listen to additional MAC multilink address corresponding to the IPv6
>multilink addresses for link and site local all-router multicast
>address FF02::1 and FF05::2)
>
>Affan Ahmed.

--------
Yuji Sasaki
Company E-Mail :[EMAIL PROTECTED]

Silex technology america, Inc.
64 East 6400 South, Suite 330
Murray, UT  84107
Phone 801-747-0656
Fax   801-747-0659

---------------------------------------------------------------------
The IPv6 Users Mailing List
Unsubscribe by sending "unsubscribe users" to [EMAIL PROTECTED]

Reply via email to