Jim Bound wrote:
>
> Alex,
>
> > Jim,
> > [...]
> > I am afraid, there is a misunderstanding. Nobody advocated
> > transport+port to promote routing or as a strategy.
> > [...]
> > As I do not expect everyone to understand or care about bit shifting, I
> > should mention just for the sake of the record, that it was shown, t
> > hat the details can be crafted to eliminate the risk [...]
>
> I saw that mail. That is not the only bug. The other bug will be
> assuming the extensions and when to update the length. Adding this to the
> IPv6 processing would need to be reviewed very carefully at this point.
> Because it can be bit shifted does not mean it will work.
>
You must have skipped one or more of my earlier messages.
> > [...]
> >
> > No question, IPv6 extensibility which is the ability to define and stack
> > options in the 'hop by hop' headers, or 'destination' headers, is a
> > plus.
>
> Also completely new prototcols.
>
What does "completely new protocols" have to do with IPv6
extensibility??? What do you mean?
> [...]
> I agree and that was for the address too. But if one follows the
> extension chain all works correctly no matter what size or how many
> extensions exist.
>
> > Minimizing the price paid is even more necessary, when we examine the
> > gain on extensibility, by counting the IPv6 options which were
> > standardized. and compare, them to the IPv4 standard options.
>
> IPv4 options don't work? I am missing the point?
>
You missed some of the earlier messages - mine, Thomas Eklund, etc...
Sequential header processing is OK in software, or in host, but prevents
parallel header fields and headers processing in hardware
implementations, in routers, where fast IPv6 header processing is
critical, and the price paid is high.
The IPv4 options worked, but as it turned out, few were useful, so few
are mandatory.
This should be an additional indication, that options, in reality, are
not that valuable.
Based on the number of current standard IPv6 options -- how many? a
handful? -- we should try
really hard to minimize the cost of extensibility 'cause I see more pain
than gain.
Alex
S/MIME Cryptographic Signature
