[kixelated]

Plans the follow-ups raised while merging the 2026-10-28 spawn's PRs, decided one at a time with the maintainer.

#7195, which this was stacked on, has landed; main is merged in, so the diff shows only this PR's changes.

Changes

• New: A cancelled miss stops waiting [XS] in m2, a test hook plus a regression test for #7135's cancelled wait.
• Folded into m2/perf/single-delivery-write: write_frames' remaining ~20% per-frame cost from fix(net): preserve transport frame bounds #6922's canonical storage (#7139).
• Folded into m1/drop-set: DropCursor::cut reports a cut that a fetched copy covers (#7053, #7149).

Posted as PR comments instead, since open PRs hold those quests:

• #[kio::test] can live in kio-macros, on #7027.
• The kio-macros workspace-hack entry, on #6985.

Decisions

• A time limit to FETCH a joined group's start anyway (fix(net): finish the browser backend baseline on moq-transport-22 #7135): ✅ skip · [S] m2
• Test for a cancelled fill wait (fix(net): finish the browser backend baseline on moq-transport-22 #7135): ✅ [XS] m2 · [XS] m1 · skip
• write_frames residual (perf(net): resolve the paired network microbenchmark regressions #7139): ✅ fold into single-delivery-write · separate [M] m2/perf · skip
• Covered-cut report (fix(net): an aborted live copy doesn't hold a track's end open #7149): ✅ fold into drop-set · [S] m1 · skip
• Notes for the PRs holding kio-clock and hakari (feat(kio): #[kio::machine] writes straight-line poll bodies #7148): ✅ comment on both · only hakari · neither
• Already declined, so not planned again: gofmt for the test Go clients (build(go): fail just go check on unformatted Go #7158), idling a hidden camera (feat(publish)!: muted and invisible keep capture open #7157), a loom ARM lane (fix(kio): order a log read before the thread's next epoch pin #7134), and ts::Restart beyond the CLI (feat(mux): a TS clock step restarts the import under a new epoch #7144).

quest check passes.

(Written by Claude Opus 5.5)

🤖 Generated with Claude Code

[coderabbitai]

Warning

Review limit reached

You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository.

Next included review available in 5 minutes.

Check out review usage here.

View limit details

Limit details: You’ve used all 4 included reviews currently available.

Learn how review limits work.

Review configuration:

Run configuration

• Source mode: Auto
• Repository: Unavailable
• Branch: Unavailable
• File: No configuration file from this repository applied
• Commit: Unavailable
• Layers: Organization UI
• Review profile: CHILL
• Plan: Advanced
• Run ID: ef3baf7c-2615-48de-89ac-26f9b7d83491

Commits

Reviewing files that changed from the base of the PR and between 3637559 and b392aa8.

Files selected for processing: 4

• quest/m1/drop-set.md
• quest/m2/README.md
• quest/m2/fill-wait-cancel-test.md
• quest/m2/perf/single-delivery-write.md

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-29T15:45:32.031417Z	b392aa8	New commits

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[kixelated]

Automated review by review (OpenAI)

Reviewed commit: 74840c9

No actionable planning bugs found. Accounting for the already-reviewed #7195 stack, the new work is the cancelled-fill-wait test, covered-cut handling folded into drop-set, and batch32 write costs folded into the single-delivery investigation.

Direction: a minimal test-only hook establishes task release without adding a production timeout; covering copies clearing cuts fits the existing shared disposition model; measuring both write paths together preserves canonical-content/reservation guarantees without assuming they share every optimization. These are appropriately scoped additions rather than parallel mechanisms.

Verification: static inspection of the stack delta, relevant plans, cancellation checks and group write paths, with independent reviews. No tests or benchmarks were run, and historical performance figures were not independently reproduced.

[kixelated]

Automated review by review (OpenAI)

Reviewed commit d150204cfcb0a1ee614dc4d6d964169fc842454c.

No actionable planning bugs found in the update. After excluding incoming main changes, the substantive delta replaces the deleted fill-timeout prerequisite with the one-group publisher plan and narrows unknown-range regression coverage to that supported shape. The other reviewed contributions remain equivalent against their respective bases.

Direction: making the one-group narrowing an explicit prerequisite keeps marker scope and tests aligned with the publisher design, rather than implementing obsolete multi-group ordering behavior. The plan still requires both FETCH drafts and the largest-group draft-22 fill to cover a listed group lost before writing.

Verification: static base-aware comparison and inspection of the linked one-group plan. No tests, benchmarks, or interop runs were executed.

[kixelated]

Merge summary:

• Merged main three times. moq-dev/moq#7194 replaced ietf-fill-timeout with ietf-publisher-one-group. The new unknown-range quest now waits on the one-group quest and covers a single listed group.
• moq-dev/moq#7195 then landed as a squash. Its quests take main's copies, and this PR keeps only its own changes: fill-wait-cancel-test, the covered-cut section in drop-set, and the batch32 scope in single-delivery-write.
• Both hand-resolved heads were reviewed clean (the automated review on d150204 and Codex on 6db65a1). The last merge, of moq-dev/moq#7079 and moq-dev/moq#7083, was conflict-free.
• quest check --base origin/main passes.

(Written by Claude Opus 5.5)