Re: [RFC PATCH net-next 0/4] tcp: carry a BPF-chosen identifier in TSval low bits

[Jakub Sitnicki]

Hey Stan,

On Thu, Oct 15, 2026 at 02:45 PM -07, Stanislav Fomichev wrote:
> On 10/13, Willem Ferguson wrote:

[...]

>> Questions for reviewers:
>> 
>>  1. Is encoding an identifier in the low bits of TSval acceptable in
>>     principle, given the trade-offs above?
>>  2. If so, what feedback do you have on the implementation?
>
> Don't see any feedback yet, so from the top of my head: looks like you
> control two sides of the communication, so why not a custom (bpf) tcp option?
> Would you need that on every packet and it makes it expensive?
>
> Or any other field you can abuse maybe? IPv6 Flow label?

Nah, we don't control the client unfortunately.

Client here is any device that connects to our edge PoP.

> Implementation-wise, the sock_ops is going away in favor of bpf_tcp_ops,
> so BPF_SOCK_OPS_TCP_TS_ID_CB_FLAG is a no go.

Yeah, we have just recently realized [1].

Thanks for feedback.

[1] https://lore.kernel.org/all/877bjk893u.fsf@xxxxxxxxxxxxxx/