[kixelated]

Important

Maintainer prerequisite before merge: enable repository Pages with Settings > Pages > Source: GitHub Actions. Merging does not change repository settings.

Without it, the main-branch Nightly's Benchmark site job still validates the reports and pushes the history to gh-pages, then fails at actions/deploy-pages with "Ensure GitHub Pages has been enabled", so the Nightly run goes red every night. It never skips silently.

Problem

Nightly retained binding results, but most Rust Criterion targets only smoke-tested with --test, and the JS measurements had no durable charts.

Approach

• The Rust bench targets group measures its Criterion targets in quick mode through the existing evidence runner (just gh criterion, bench/run.sh --criterion). The JS group tees each benchmark's CSV. Both write a validated report.json; the binding benchmark already emits one.
• A main-only Benchmark site job validates every bench-* artifact that records an outcome, then stores cost and throughput charts on one serialized gh-pages branch with github-action-benchmark and deploys them explicitly to Pages. Alerts and comments stay off.
• Malformed, empty, nonfinite, or duplicate measurements fail validation. A failed suite contributes no point.
• Fixes stale benchmark callers: the JS catalog bench uses the moq-clock Manual clock, protocol diagnostics stay off CSV stdout, and moq-net's announce_duplicate and reprice_duplicate benches run with a zero update_hold. Both currently fail Nightly on main.

Impact

• Public API: none. Wire: none.
• Nightly measures the selected Rust targets, including the new moq-quic and e2ee benches, and retains Rust and JS evidence for 45 days.
• The "Rust benchmark tests" group (nextest) is not charted; its output is not structured.

Validation

just check and the tooling regressions (sh/gh/nightly.test.py, sh/gh/bench-report.test.py) pass. A measured local just gh nightly js produced 567 valid entries across all 16 suites, including the new e2ee bench (forward stopped early, see Follow-ups). A measured moq-quic wire + receive run through just gh criterion produced 32 Criterion estimates. The moq-net origin bench passes --test again. The pinned upstream action generated both chart families twice in a temporary repository and kept both history points. The full measured Rust nightly suite was not run locally.

Alternatives

Smoke mode supplies no Criterion estimates. Pushing a Pages branch with GITHUB_TOKEN alone does not deploy a site, so the publisher deploys its generated artifact explicitly.

Follow-ups

• After Pages is enabled, verify the first main-branch Nightly deployment at https://moq-dev.github.io/moq/bench/.
• Watch the Rust bench targets job time against its 60-minute limit; quick mode adds measurement to every case.
• Paired regression alerts remain with bench-ci.
• Charting the nextest benchmark tests needs a structured output first.
• Regression on main, outside this PR: js/net/bench/forward.ts hangs at 100% CPU at 2048 held watches since feat(net): announce consumers report when they have synced #5724 (bisected: 866e72783~1 finishes in about 2 s, 866e72783 never finishes). Nightly's JS group will stall on it until fixed, and a failed group publishes no JS point.

(Written by Claude Opus 5.5, building on GPT-6's draft)

🤖 Generated with Claude Code

[kixelated]

Implementation and local checks are complete. This stays a draft for the maintainer's Pages choice: enable Source: GitHub Actions (recommended), then verify the first main-branch Nightly deployment; or keep the site disabled. Repository settings have not changed.

(written by GPT-6)

[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 10 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

• Configuration used: Organization UI
• Review profile: CHILL
• Plan: Advanced
• Run ID: 24eb2fc8-ff88-40fd-8ce7-b103224d3afb

📥 Commits

Reviewing files that changed from the base of the PR and between 8bf21d3 and 1bfcd7e.

📒 Files selected for processing: 15

• .github/justfile
• .github/workflows/nightly.yml
• bench/README.md
• bench/metadata.py
• bench/run.sh
• js/net/bench/frames.ts
• js/net/bench/requests.ts
• js/publish/bench/catalog.ts
• quest/m2/bench-site.md
• rs/moq-net/benches/origin.rs
• sh/gh/bench-publish.sh
• sh/gh/bench-report.py
• sh/gh/bench-report.test.py
• sh/gh/nightly.sh
• sh/gh/nightly.test.py

• 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 2026-10-18T04:26:27.896711Z	1bfcd7e	Draft marked ready

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]

Merging at 1bfcd7e. The maintainer has enabled repository Pages (Source: GitHub Actions), the prerequisite in the description. Codex thumbs-up on this head; CI green. Check the first main deployment at https://moq-dev.github.io/moq/bench/ after the next Nightly.

(Written by Claude Opus 5.5)