Re: [tcpm] 793bis and RFC3168

[t petch <[e-mail]>]

On 07/10/2021 08:55, Gorry Fairhurst wrote:
> On 06/10/2021 13:00, tom petch wrote:
>> Does 793bis warrant an
>> Updates 3168?
>>
>> It does change the format of the IANA registry as created by RFC3168
>> and so would seem an update.  I realise that the IANA registry will
>> point to 793bis; I am unclear whether the revised registry will have
>> any mention of RFC3168 not that it changes my view on Updates.
>>
>> Tom Petch
>> _______________________________________________
>> tcpm mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/tcpm
>
> Thanks for asking, I looked at this.
>
> RFC 3168 gave initial contents of the registry and stated:
>
>     The Transmission Control Protocol (TCP) included a 6-bit Reserved
>     field defined inRFC 793
> <https://datatracker.ietf.org/doc/html/rfc793>, reserved for future use,
> in bytes 13 and 14
>     of the TCP header, as illustrated below.  The other six Control bits
>     are defined separately byRFC 793
> <https://datatracker.ietf.org/doc/html/rfc793>.
>
> This could have been handled more elegantly, but  I think the RFC3168
> text remains correct, in that the base TCP spec (RFC793, or it's bis)
> define the 6 other bits.  To me, there is no change to the meaning of
> RFC3168 intended by 793bis, and I think updating a reference would not
> need to be flagged as an update.

Yeees, I was looking at the format where 'Bit' is updated to 'Bit 
Offset' and 'Assignment Notes' is added.  An update?  Perhaps not.

Tom Petch

> Best wishes,
>
> Gorry Fairhurst
>
> (as co-chair tsvwg).
>