I just published a new draft with support for out of order packets and it defines no new frame types. https://datatracker.ietf.org/doc/draft-smith-quic-receive-ts/
Thanks, Ian On Thu, Apr 17, 2025 at 4:09 AM Kazuho Oku <[email protected]> wrote: > > > 2025年4月17日(木) 1:00 Ian Swett <[email protected]>: > >> Coming from an author's perspective, I'm happy to remove the extended ACK >> portion and ship a new draft version before adoption. >> >> I wrote a PR that does that and always includes the timestamp section at >> the end of ACK frames, for example: >> https://github.com/wcsmith/draft-quic-receive-ts/pull/8 >> > > I support adoption, and I also think that this PR is in the right > direction. > > Though I kind of wonder if we could agree on how to annotate ACK packets. > At the moment, Multipath QUIC adds a path identifier and uses a different > frame type. This pull request adds a new field for acks, reusing the > original frame type. > > I think people have argued that we do not need to develop an encoding > scheme that allows extensions add new fields to the ACK frame. I fully > agree with that. > > But at the same time, I wonder if extensions could agree on one way of > annotating ACKs rather than trying to define their own ways. > > >> >> Thanks, Ian >> >> On Tue, Apr 15, 2025 at 5:41 AM Mirja Kuehlewind <mirja.kuehlewind= >> [email protected]> wrote: >> >>> Hi chairs, hi all, >>> >>> >>> >>> we had some discussion on list and in the meeting suggesting to rather >>> use a new frame than using an extended ACK. I generally support this work >>> but I would actually rather want to adopt a document that reflects this >>> change. Or what’s the plan here? >>> >>> >>> >>> Mirja >>> >>> >>> >>> >>> >>> >>> >>> *From: *Lucas Pardue <[email protected]> >>> *Date: *Monday, 14. April 2025 at 21:01 >>> *To: *"[email protected]" <[email protected]> >>> *Cc: *QUIC WG Chairs <[email protected]> >>> *Subject: *Re: Adoption call: QUIC Extended Acknowledgement for >>> Reporting Packet Receive Timestamps >>> >>> >>> >>> Reminder: the call for adoption closes at the end of this week >>> >>> >>> >>> On Thu, Apr 3, 2025, at 20:58, Lucas Pardue wrote: >>> >>> Hi folks, >>> >>> >>> >>> At IETF 122 we discussed QUIC Extended Acknowledgement for Reporting >>> Packet Receive Timestamps [1]. The sense of the room was that there was >>> support for the QUIC WG adopting work on ACKs with timestamps. >>> >>> >>> >>> The email launches a formal adoption call >>> for draft-smith-quic-receive-ts-01 as the basis of this work. The call will >>> run until end of day 2025-04-18 anywhere on earth [2] >>> >>> >>> >>> Please post comments in favor or against as replies to this thread. >>> >>> >>> >>> Cheers >>> >>> Lucas & Matt >>> >>> QUIC WG Chairs >>> >>> >>> >>> [1] - https://datatracker.ietf.org/doc/draft-smith-quic-receive-ts/ >>> >>> [2] - https://en.wikipedia.org/wiki/Anywhere_on_Earth >>> >>> >>> >>> > > -- > Kazuho Oku >
