Re: [quicwg/base-drafts] Improve ACK_ECN frame encoding (e.g., use bit-vector) (#1439)

[Kazuho Oku <[e-mail]>]

@ianswett 
> I'm going to reiterate: I don't think duplicate detection is a hard problem. You have an IntervalSet that is used to generate an ack.

That is true, however I do not think that the fact that we can have a bullet-proof duplicate detection means that we should require every QUIC stack to have it, or that the fact justifies the encoding proposed in #1372.

Using the ACK frame (or ACK_ECN frame) to notify which PN had the CE flag set (or other bits of ECN) will simplify the sender implementation, because you can use the signal to determine certain things that the encoding proposed in #1372 cannot (or hard to) tell; e.g., when 1-RTT passes since you've seen the last CE, if there was a consecutive loss of a ECN packet (i.e. blackhole detection).

On the receiver side, I might argue that reapplying the frames without detecting a duplicate could also be a valid approach, considering the fact that we cannot decide to skip decrypting the payload just by checking if the decrypted PN collides with a previously seen PN (see [QUIC-TLS draft section 10.3](https://quicwg.org/base-drafts/draft-ietf-quic-tls.html#rfc.section.10.3)) Using the ACK frame that carries PNs with their ECN flags attached allows this kind of approach.

To summarize, I still think that we should consider one of the following encodings, assuming that the size of the ACK (or ACK_ECN) frame will be acceptable, if not smaller.
* use bit-vector (https://github.com/quicwg/base-drafts/issues/1439#issuecomment-396433945)
* use truncated bit-vector (https://github.com/quicwg/base-drafts/issues/1439#issuecomment-396615367)
* use typed blocks (bottom of https://github.com/quicwg/base-drafts/issues/1439#issuecomment-397133920, https://github.com/quicwg/base-drafts/issues/1439#issuecomment-397157328)


-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/quicwg/base-drafts/issues/1439#issuecomment-397321982