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

[Sebastian Moeller <[e-mail]>]

Without wanting to take a strong position here (one way or the other) due to this beeing far outside my area of expertise, I want to mention that ATM/AAL5 used trailers*, but each ATM cell carrying such a trailer used a bit in its header to signify that the trailer would reside in the last 8 octets of the 53 byte fixed ATM cell. No idea whether there ever was a "zero-copy" decapsulator in use. I naively assume that with its fixed offset of 41 octets into the ATM cell's payload and a simply way o discover this (via a fixed position bit in the ATM cell header) was considerably more regular and simply than what is discussed for UDP.

Regards
	Sebastian

*) Probably well known here, but I have not seen it come up in the discussion


> On Jul 18, 2021, at 01:06, Joe Touch <[e-mail]> wrote:
> 
> 
> 
>> On Jul 17, 2021, at 2:58 PM, Tom Herbert <[e-mail]> wrote:
>> 
>> If I'm reading the draft correctly, it is possible that UDP options
>> could appear both in the headers and trailers of a packet. If that is
>> correct, then that means an implementation needs to be able to handle
>> both headers and trailers within the *same* packet.
> 
> That is always the case for non-fragmented packets - the only mode that supports transparent use of options that don’t change the payload data with legacy receivers. 
> 
> Joe