Carsten, sorry, I will write below something I have a hard time
understanding since I participate in 6LoWPAN.
Carsten Bormann a écrit :
On Jun 17, 2009, at 16:55, Alexandru Petrescu wrote:
It is not clear to me why there is a need for 6LoWPAN to manage
fragmentation below the IP layer
RFC 4919, section 4.3 and 5 (item 1).
Indeed:"a fragmentation and reassembly adaptation layer must
be provided at the layer below IP." and then rfc2460:
"On any link that cannot convey a 1280-octet packet in one piece,
link-specific fragmentation and reassembly must be provided at a
layer below IPv6."
I doubt IETF specifies link-specific layers below IPv6. I don't know of
any[*]. Please point to an existing RFC Standards Track defining a
layer below IPv6.
This is why I don't understand the goal of specifying 6LoWPAN
fragmentation at a layer situated below the IPv6 layer.
I am wondering how this work could have been chartered at all... but I
have surely missed earlier discussion and motivation.
Alex
[*] there are IETF IPv6-over-foo documents; they describe mainly
mappings between addresses, constants for encapsulation, MTU values,
etc. Whereas here we talk about _logic_ to be implemented in a
_new_ layer below IP - fragmentation and reassembly is such logic.
Fully specifying such fragm/reassembly layers goes well beyond the
typical IP-over-foo description.
_______________________________________________
6lowpan mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/6lowpan
