Re: [tsvwg] I-D Action: draft-ietf-tsvwg-udp-options-13.txt

[Joseph Touch <[e-mail]>]

> On Jul 10, 2021, at 11:49 AM, Tom Herbert <[e-mail]> wrote:
> 
> On Sat, Jul 10, 2021 at 9:09 AM Joseph Touch <[e-mail]> wrote:
>> 
>> 
>> UDP options does at least when UDP checksum is
>> non-zero. If processing UDP options walks the list twice,
>> 
>> Only when the SHOULD is not followed; we can certainly add that note.
>> 
>> An implementation could easily walk the TLV and set a flag if options other than OCS or NOP appear. The walk ignores such options until OCS and restarts if OCS works.

And notes when OCS isn’t present but required.

>> So we can note that this “double walk” would incur double work if:
>> - options other than OCS and NOP appear before OCS
> 
> What if OCS is not present?

See above.

> If we ignore options in the first walk to
> account for the possibility that an OCS might be present, doesn't that
> mean that once the walk completes and we haven't found the OCS we need
> to walk again to actually process all the options?

Only if OCS is optional.  If OCS is required, if you haven’t found OCS you would stop.

If OCS is present and early, the walk stops quickly. However, if OCS isn’t required, it’s not clear that if its absence in the first few bytes should interfere with the rest of processing (which would be ignored if OCS shows up later).

>> AND
>> - offloading is used

Again, it boils down to whether OCS is buried deep in the packet or not - that’s why we have a SHOULD. If we have a MUST, we have to decide now what the alignment is and force that overhead on all packets (including alignment).

Again, though, if the WG wants to push that field to the front as fixed - either always the first two bytes or always the first two X-aligned bytes (for X=4 or 8), then it’s easy to add - but we need to pick 4 or 8 now to do that. The issue is that there are no other equivalent fields that are needed, though.

Joe