perf(moq-mux): SIMD-accelerate MPEG-TS sync byte resync

[kixelated]

Summary

You asked me to look for code that could be SIMD-optimized, but only where it's a simple change. After surveying the byte-scanning hot loops across rs/ and js/, there was exactly one clean win.

The MPEG-TS demuxer (rs/moq-mux/src/container/ts/import.rs) resyncs after a lost sync byte by scanning forward one byte at a time looking for the next 0x47:

if self.scratch[off] != 0x47 {
    off += 1;
    continue;
}

This now jumps straight to the next sync byte with memchr::memchr, which is SIMD-accelerated (SSE2/AVX2/NEON) and was already a dependency of the crate:

if self.scratch[off] != 0x47 {
    off = match memchr::memchr(0x47, &self.scratch[off..]) {
        Some(rel) => off + rel,
        None => self.scratch.len(),
    };
    continue;
}

The end state is identical to the old loop (advance to the next sync byte, or drop the whole buffer when none remains), so resync semantics are unchanged. This matters most when a live capture joins mid-stream or a stream loses sync, where the old path crawled byte-by-byte over the buffer.

What I deliberately left alone

• H.264/H.265 Annex-B start-code scanning (rs/moq-mux/src/codec/annexb.rs) already uses memchr::memmem::find, which is SIMD-accelerated. No change needed.
• There is no TypeScript packet finder to optimize. The MPEG-TS demuxer lives in Rust; the JS side doesn't scan for sync bytes / start codes.
• Other candidates (ADTS frame parsing, ESDS descriptor scan, length-prefix decode) were either too small to matter or not a clean drop-in, so they didn't meet the "simple change" bar.

Test plan

• cargo test -p moq-mux (all 36 container::ts tests pass, including import_resyncs_after_byte_misalignment, stays_desynced_until_next_pusi, tei_continuation_drops_partial_and_resyncs)
• cargo clippy -p moq-mux clean

(Written by Claude)

---

Generated by Claude Code

[coderabbitai]

Warning

Review limit reached

@kixelated, we couldn't start this review because you've reached your PR review rate limit.

More reviews will be available in 3 minutes and 52 seconds. Learn how PR review limits work.

Your organization has run out of usage credits. Purchase more credits in the billing tab to continue.

⌛ How to resolve this issue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

We recommend that you space out your commits to avoid hitting the rate limit.

🚦 How do rate limits work?

CodeRabbit enforces hourly rate limits for each developer per organization.

Our paid plans include higher PR review limits than trial, open-source, and free plans. In all cases, reviews become available again over time. During sustained high-volume PR review activity, CodeRabbit may temporarily slow when the next review becomes available.

Please see our Fair Usage Limits Policy for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: 1e326a37-8833-4cad-97d5-0a81fb849a2f

📥 Commits

Reviewing files that changed from the base of the PR and between 916f7ae and faebdff.

📒 Files selected for processing (1)

• rs/moq-mux/src/container/ts/import.rs

Walkthrough

This PR changes Import::decode's resynchronization: when the byte at the current offset is not 0x47, it now calls memchr to locate the next 0x47 in the remaining scratch buffer. If memchr finds no match, the offset is set to scratch.len() so the packet loop exits.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and concisely summarizes the main change: replacing byte-by-byte MPEG-TS sync resynchronization with SIMD-accelerated memchr. It is specific, directly related to the primary improvement, and accurately reflects the changeset.
Description check	✅ Passed	The description is thorough and directly related to the changeset. It explains the motivation, the old vs. new approach, reasoning for other non-changes, and includes comprehensive test results, all aligned with the actual code modifications.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

✨ Simplify code

• Create PR with simplified code
• Commit simplified code in branch claude/simd-optimization-search-g936s8

---

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 and usage tips.}

[coderabbitai]

🧹 Nitpick comments (1)

rs/moq-mux/src/container/ts/import.rs (1)

145-148: ⚡ Quick win

Use if let for memchr result unwrapping consistency.

This match is only unwrapping Option and setting fallback state; converting to if let aligns with the repo Rust style rule.

Suggested refactor

-				off = match memchr::memchr(0x47, &self.scratch[off..]) {
-					Some(rel) => off + rel,
-					None => self.scratch.len(),
-				};
+				if let Some(rel) = memchr::memchr(0x47, &self.scratch[off..]) {
+					off += rel;
+				} else {
+					off = self.scratch.len();
+				}

As per coding guidelines, "**/*.rs: Prefer if let / let ... else over an unwrapping match when the only job is to unwrap."

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@rs/moq-mux/src/container/ts/import.rs` around lines 145 - 148, Replace the
match that unwraps memchr::memchr with an if let to match repo style: call
memchr::memchr(0x47, &self.scratch[off..]) and use `if let Some(rel) = ... { off
= off + rel; } else { off = self.scratch.len(); }`, preserving the exact
semantics for the `off` update; reference the `off` variable, `self.scratch`,
and the memchr::memchr call in import.rs.

Source: Coding guidelines

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Nitpick comments:
In `@rs/moq-mux/src/container/ts/import.rs`:
- Around line 145-148: Replace the match that unwraps memchr::memchr with an if
let to match repo style: call memchr::memchr(0x47, &self.scratch[off..]) and use
`if let Some(rel) = ... { off = off + rel; } else { off = self.scratch.len();
}`, preserving the exact semantics for the `off` update; reference the `off`
variable, `self.scratch`, and the memchr::memchr call in import.rs.

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: 935f8402-4feb-4659-ba18-da4864e9dc55

📥 Commits

Reviewing files that changed from the base of the PR and between 789cf21 and 5527d4f.

📒 Files selected for processing (1)

• rs/moq-mux/src/container/ts/import.rs