On Wed, Sep 8, 2010 at 11:43 AM, Jon Smirl <[email protected]> wrote:
> On Wed, Sep 8, 2010 at 10:40 AM, Mariano Alvira <[email protected]> wrote:
>> On Wed, Aug 11, 2010 at 01:35:44AM +0400, Dmitry Eremin-Solenikov wrote:
>>> On 8/11/10, Mariano Alvira <[email protected]> wrote:
>>>
>>> Please refer to the part 'Restricted encapsulated values'.
>>>
>>> IMO you should have generated EUI-64 directly from OUI/extid
>>> EUI-48 -> EUI-64 is really meant for cases where manufacturer
>>> provides 48-bit address (like ethernet card), but we need 64-bit
>>> one (like in IPv6).
>>
>> not that I want to revisit this... but I'm trying to do the right
>> thing and I came across this:
>>
>> in RFC4944, section 6:
>>
>> "The Interface Identifier [RFC4291] for an IEEE 802.15.4 interface may
>> be based on the EUI-64 identifier [EUI64] assigned to the IEEE
>> 802.15.4 device.  In this case, the Interface Identifier is formed
>> from the EUI-64 according to the "IPv6 over Ethernet" specification
>> [RFC2464]."
>
>
> They are referring to the Interface Identifier, not the 48 to 64 bit
> conversion. So starting from the EUI-64 do this....
>
>
>   The Interface Identifier is then formed from the EUI-64 by
>   complementing the "Universal/Local" (U/L) bit, which is the next-to-
>   lowest order bit of the first octet of the EUI-64.  Complementing
>   this bit will generally change a 0 value to a 1, since an interface's
>   built-in address is expected to be from a universally administered
>   address space and hence have a globally unique value.  A universally
>   administered IEEE 802 address or an EUI-64 is signified by a 0 in the
>   U/L bit position, while a globally unique IPv6 Interface Identifier
>   is signified by a 1 in the corresponding position.  For further
>   discussion on this point, see [AARCH].
>
>
> I did that in the link local address code. Flipping the bit turns it
> into a locally administered address - ie link local.


To clarify this, the MAC should have the global form of the EUI64.
When generating the link local IP address I convert to the local form
in software.

>
>
>>
>> And in RFC2464 section 4:
>>
>> "The OUI of the Ethernet address (the first three octets) becomes the
>> company_id of the EUI-64 (the first three octets).  The fourth and
>> fifth octets of the EUI are set to the fixed value FFFE hexadecimal.
>> The last three octets of the Ethernet address become the last three
>> octets of the EUI-64."
>
> This is the description of how to convert a 48b MAC to a 64b one. But
> you already have a 64b one so no conversion is needed. Just don't put
> FFFE in the middle since that is the special key used in converting
> 48/64b.
>
>
>>
>> "For example, the Interface Identifier for an Ethernet interface whose
>> built-in address is, in hexadecimal,
>>
>>                             34-56-78-9A-BC-DE
>>
>> would be
>>
>>                         36-56-78-FF-FE-9A-BC-DE."
>>
>>
>> From my understanding of this, the HW address will be the EUI64 and
>> and the ipv6 address will have the FFFE in the middle.
>>
>> So for Redwire, we could have HW: 00-50-C2-A8-Cx-xx-yy-yy
>>
>> and the "Interface Identifier" would be (I supppose):
>>
>> 00-50-C2-FF-FE-A8-Cx-xx
>>
>> but which group of bits to put as "x-xx" doesn't seem properly defined
>> (at least for IAB owners). It's also annoying that, I have 28 bits in
>> my EUI64, but RFC4944 is saying I can only use 12, even though as you
>> state 802.15.4 uses 64-bit addresses and so does ipv6 so why is
>> RFC4944 making me go from 64-> 48 -> back to 64?
>>
>> Does anyone have any thoughts?
>>
>> -Mar.
>>
>>
>
>
>
> --
> Jon Smirl
> [email protected]
>



-- 
Jon Smirl
[email protected]

------------------------------------------------------------------------------
This SF.net Dev2Dev email is sponsored by:

Show off your parallel programming skills.
Enter the Intel(R) Threading Challenge 2010.
http://p.sf.net/sfu/intel-thread-sfd
_______________________________________________
Linux-zigbee-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-zigbee-devel

Reply via email to