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]
