Date: Wed, 06 Jul 2005 17:28:28 +0200
From: Brian E Carpenter <brc(_at_)zurich(_dot_)ibm(_dot_)com>
Message-ID: <42CBF89C(_dot_)8040201(_at_)zurich(_dot_)ibm(_dot_)com>
| Well, that is not how I read the text in RFC 2460. It's pretty clear
| to me that there are only 32 option codes and that the other three bits
| don't extend the code space, but rather they modify the meaning of the
| 32 basic options. (e.g. the same option can have a hop-by-hop flavour
| and an e2e flavour).
I'm glad that John Leslie pointed out that IANA isn't interpreting it
that way, which would be truly wild.
What 2460 says is ...
The three high-order bits described above are to be treated as part
of the Option Type, not independent of the Option Type. That is, a
particular option is identified by a full 8-bit Option Type, not just
the low-order 5 bits of an Option Type.
I have no idea how you manage to interpret "identified bt a full 8-bit
Option Type, not just the low-order 5 bits" in the way you say you do,
but if that text isn't clear enough for you, I have no idea what could
possibly be, and perhaps we should just stop talking about anything that
relates to interpretation of the documents, as it seems likely to just
be a waste of time.
On the parenthetcal comment at the end of your message - yes, of course
an option can be both e2e and hbh, in fact, that's the normal case, the
options registered apply to both hbh & dst opt headers (though a particular
option can be defined to have meaning only in one of those situations,
Router Alert as an e2e option would be a fairly meaningless thing...)
What's more relevant here is how you're managing to convolute the
e2e/hbh issue, with the top 3 bits of the option code? They're
orthogonal. None of the top 3 bits have anything at all to do with
which header the option appears in (though the 0x20 bit being set in
an e2e opt header takes a little bit of understanding to appreciate
how it can apply - don't just assume it cannot, it certainly can, but
it isn't going to be common)
In any case, for John Klensin, there's no need to waste time clarifying
this, it is already just about as clear as it is possible to write it.
kre
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf