[fireairforce]

Adds opt-in module.references to replace exact browser modules before parsing, loaders, or dependency traversal. Factories receive JSON data; Node builds and unconfigured applications retain their original behavior. Noncanonical resource registrations fail explicitly.

Covers aliases, symlink imports, barrel re-exports, queries, implementation isolation, and legacy browser output with live bindings and cycles. Includes the submodule fix in utooland/next.js#211. Framework integration: EVJS #162.

Validation: 17 focused native/capability/worker tests on Node 20, Rust check/clippy/fmt, and Biome passed.

[chatgpt-codex-connector]

Codex Review Summary: ✅ Completed

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

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-11-10T07:45:59.174752Z	863dd40	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.

[github-actions]

📊 Performance Benchmark Report (with-antd)

Utoopack Performance Report

Report ID: utoopack_performance_report_20261110_102908
Generated: 2026-11-10 10:29:08
Trace File: trace_antd.json (0.3GB, 0.88M spans)
Test Project: examples/with-antd

---

Executive Summary

Metric	Value	Assessment
Total Wall Time	5,495.7 ms	Baseline
Total Thread Work (de-duped)	16,836.4 ms	Non-overlapping busy time
Effective Parallelism	3.1x	thread_work / wall_time
Working Threads	10	Threads with actual spans
Thread Utilization	30.6%	⚠️ Suboptimal
Total Spans	879,322	All B/E + X events
Meaningful Spans (>= 10us)	193,328	(22.0% of total)
Tracing Noise (< 10us)	685,994	(78.0% of total)

---

Build Phase Timeline

Shows when each build phase is active and how much CPU it consumes.
Self-Time is the time spent exclusively in that phase (excluding children).

Phase	Spans	Inclusive (ms)	Self-Time (ms)	Wall Range (ms)
Resolve	38,635	4,940.0	1,270.7	2,617.2
Parse	8,161	1,104.7	802.8	4,524.9
Analyze	128,232	36,862.9	6,787.2	4,471.8
Chunk	5,010	4,005.8	547.1	1,749.9
Codegen	11,149	1,735.0	1,111.3	1,742.5
Emit	28	35.0	17.5	7.6
Other	2,113	9,464.9	4,752.8	5,495.7

---

Workload Distribution by Diagnostic Tier

Category	Spans	Inclusive (ms)	% Work	Self-Time (ms)	% Self
P0: Scheduling & Resolution	167,434	44,163.2	262.3%	8,425.1	50.0%
P1: I/O & Heavy Tasks	2,809	100.4	0.6%	82.9	0.5%
P2: Architecture (Locks/Memory)	0	0.0	0.0%	0.0	0.0%
P3: Asset Pipeline	22,155	6,802.4	40.4%	2,418.0	14.4%
P4: Bridge/Interop	0	0.0	0.0%	0.0	0.0%
Other	930	7,082.4	42.1%	4,363.4	25.9%

---

Top 20 Tasks by Self-Time

Self-time is the exclusive duration: time spent in the task itself, not in sub-tasks.
This is the most accurate indicator of where CPU cycles are actually spent.

Self (ms)	Inclusive (ms)	Count	Avg Self (us)	P95 Self (ms)	Max Self (ms)	% Work	Task Name	Top Caller
4,025.1	23,573.5	97,322	41.4	0.1	21.4	23.9%	module	module (64%)
3,192.7	3,382.9	17	187806.2	496.5	755.8	19.0%	save snapshot	persist (6%)
1,699.2	1,777.1	2,150	790.3	2.4	185.1	10.1%	analyze ecmascript module	module (81%)
1,037.4	11,476.1	28,727	36.1	0.1	5.4	6.2%	process module	process module (81%)
837.4	2,417.9	21,392	39.1	0.1	7.0	5.0%	internal resolving	internal resolving (72%)
746.3	1,048.2	5,986	124.7	0.5	32.9	4.4%	parse ecmascript	parse ecmascript (66%)
608.1	679.1	9,077	67.0	0.2	6.8	3.6%	precompute code generation	generate merged code (67%)
605.1	1,836.8	699	865.7	2.0	200.0	3.6%	generate merged code	chunking (90%)
461.0	3,824.3	3,988	115.6	0.1	32.6	2.7%	chunking	chunking (24%)
424.4	2,513.2	16,637	25.5	0.0	6.0	2.5%	resolving	module (62%)
374.2	2,293.0	808	463.1	0.3	73.2	2.2%	write all entrypoints to disk	write all entrypoints to disk (13%)
355.3	355.3	329	1080.0	0.8	226.2	2.1%	generate source map	code generation (83%)
345.7	694.3	130	2659.2	3.6	208.2	2.1%	emit code	emit code (43%)
202.2	202.2	11	18378.2	93.3	179.5	1.2%	blocking	save snapshot (45%)
147.9	700.6	1,743	84.8	0.2	42.2	0.9%	code generation	code generation (83%)
82.8	177.8	969	85.5	0.1	8.1	0.5%	compute async chunks	compute async chunks (49%)
56.4	56.4	2,172	26.0	0.0	2.0	0.3%	read file	parse ecmascript (91%)
17.4	17.4	12	1450.7	5.1	5.5	0.1%	write file	apply effects (100%)
15.2	89.6	375	40.6	0.1	2.8	0.1%	async reference	async reference (42%)
13.3	13.3	10	1332.1	5.7	8.1	0.1%	compute binding usage info	write all entrypoints to disk (20%)

---

Critical Path Analysis

The longest sequential dependency chains that determine wall-clock time.
Focus on reducing the depth of these chains to improve parallelism.

Rank	Self-Time (ms)	Depth	Path
1	935.4	3	persist → save snapshot → blocking
2	411.0	6	chunking → generate merged code → emit code → emit code → emit code → read file
3	263.2	4	chunking → generate merged code → emit code → generate source map
4	262.4	2	save snapshot → blocking
5	185.3	8	module → module → process module → process module → process module → process module → process module → analyze ecmascript module

---

Batching Candidates

High-volume tasks dominated by a single parent. If the parent can batch them,
it drastically reduces scheduler overhead.

Task Name	Count	Top Caller (Attribution)	Avg Self	P95 Self	Total Self
process module	28,727	process module (81%)	36.1 us	0.05 ms	1,037.4 ms
internal resolving	21,392	internal resolving (72%)	39.1 us	0.07 ms	837.4 ms

---

Duration Distribution

Range	Count	Percentage
<10us	685,994	78.0%
10us-100us	121,665	13.8%
100us-1ms	63,068	7.2%
1ms-10ms	8,375	1.0%
10ms-100ms	177	0.0%
>100ms	43	0.0%

---

Action Items

1. [P0] Focus on tasks with the highest Self-Time — these are where CPU cycles are actually spent.
2. [P0] Use Batching Candidates to identify callers that should use try_join or reduce #[turbo_tasks::function] granularity.
3. [P1] Check Build Phase Timeline for phases with disproportionate wall range vs. self-time (= serialization).
4. [P1] Inspect P95 Self (ms) for heavy monolith tasks. Focus on long-tail outliers, not averages.
5. [P1] Review Critical Paths — reducing the longest chain depth directly improves wall-clock time.
6. [P2] If Thread Utilization < 60%, investigate scheduling gaps (lock contention or deep dependency chains).

Report generated by Utoopack Performance Analysis Agent