Re: [tsvwg] UDP options and header-data split (zero copy)

[Tom Herbert <[e-mail]>]

On Fri, Jul 30, 2021 at 8:35 PM Joseph Touch <[e-mail]> wrote:
>
> Hi, Tom,
>
> > On Jul 20, 2021, at 7:49 AM, Tom Herbert <[e-mail]> wrote:
> >
> > On Mon, Jul 19, 2021 at 5:58 PM Joe Touch <[e-mail]> wrote:
> >>
> >>
> >>
> >>> On Jul 19, 2021, at 3:11 PM, Tom Herbert <[e-mail]> wrote:
> >>>
> >>> Please check the list archives, I have been pointing out the problems and deployment impediments with protocol trailers and for a while.
> >>
> >> You have. I though you were limiting your comments to fragments at this time.
> >>
> >>> Neither, AFAIK, has consensus been established on this.
> >>
> >> It has never been consensus to limit use to only fragment mode. In fact, that mode has been argued as optional most recently.
> >>
> > Limiting use to fragment mode is not what I am suggesting. There are
> > applications that need legacy mode and those that don't (DNS on the
> > Internet needs it, UDP tunnels in a datacenter don't). There are
> > implementations that support protocol trailers and those that don't
> > (like the routers I mentioned). Protocol trailers are technically
> > required in legacy mode, but there is no technical rationale why
> > non-legacy mode needs trailers. If trailers are required in all use
> > cases, then devices that don't support them can't use UDP options even
> > if they're only interested in non-legacy mode uses. For instance, UDP
> > options is a means to bring extensibility to otherwise non-extensible
> > UDP encapsulation protocols and augmenting VXLAN with a fragmentation
> > capability is compelling. But if protocol trailers are a forced
> > requirement, then I believe that would preclude us from ever using UDP
> > options with UDP encapsulation-- that would be unfortunate.
>
> I don’t follow this line of logic.
>
> You claim that you’re not limiting use to frag-only, but then talk about “if trailers are required”.
>
> There’s no IF. They are.
>

Joe,

That is an opinion, not an established fact. If you want to convince
us that trailers are a requirement in all use cases, specifically
non-legacy mode, then please provide at least one example of options
that could not be correctly conveyed in protocol headers. Given that
we have many examples of successful protocols that have protocol
options in headers, IP options, TCP options, and IPv6 which provides
the example of how per fragment options and options for the
reassembled packet, I am dubious the need for trailers can be proven.

> If an implementation can’t deal with trailers, it can’t support UDP options - the core set is “all or none”, not “pick what you want”.

If an implementation can't deal with trailers, then that means that
users of those implementations can't use UDP options. If users can use
the protocol we're defining then what is the point of bringing it to
IETF?

Tom

>
> Joe