noq-proto: L4S support: configurable ECT(1) codepoint, echo validated against what was sent, CE bytes to the controller

Open
#877 2 comments 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
4/5
Estimated time
3-5 days
Newbie friendliness
25/100
Issue type
Feature
Clarity
Mostly clear
Activity status
Active
Tech stack
rust, wasm
Domain
networking

Research direction

The ECN marking lives in poll_transmit in noq-proto, and validation is in detect_ecn, which currently rejects any ECT(1) increase in the peer's ACK. Read the config code for TransportConfig and the congestion Controller trait to see where ecn_codepoint and on_ecn would sit. Done means ECT(1) can be configured and validated against the codepoint actually sent, with the ecn_validation_* tests and a CE-bytes test passing.

Written by the indexing model from the issue text.

Description

API congestion-control

noq-proto marks every packet ECT(0) (poll_transmit), and detect_ecn treats any ECT(1) increase in the peer's ACK as "ECN corruption". So an application can't use L4S (RFC 9330/9331): it can't send ECT(1), and if it did, validation would turn ECN off at the first ACK. The congestion controller also never learns how many of the acked bytes were CE-marked, which an L4S (scalable) controller needs.

Proposal:

  • TransportConfig::ecn_codepoint(EcnCodepoint), default Ect0, so nothing changes for existing users. It's stored per path, and detect_ecn validates the echo against the codepoint actually sent (RFC 9000 §13.4.2.1; with Ect0 this is exactly today's rule).
  • Controller::on_ecn(acked_bytes, ce_bytes), a default no-op called for every validated ECN block.

We have both working in a vendored copy, with tests (ecn_validation_*, a CE-bytes test). Would you take a PR, or two PRs if you'd rather split them? Happy to adjust the API shape.

Dominant language
Rust
Stars
454
Forks
98
Avg merge
2d 8h
Merged PRs (30d)
19

Getting set up

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from n0-computer/noq

All issues in n0-computer/noq

Similar issues

More Rust issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.