[fireairforce]

Keep the SIGTERM handler active until dev server shutdown settles, preventing repeated SIGTERM from interrupting cache persistence. Preserve forced exit on a second Ctrl+C and add subprocess regression tests.

Validated with 62 targeted tests and a native cache-persistence smoke test.

[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 2026-11-12T10:37:22.007170Z	6beb056	PR opened

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_20261112_105043
Generated: 2026-11-12 10:50:43
Trace File: trace_antd.json (0.3GB, 0.88M spans)
Test Project: examples/with-antd

---

Executive Summary

Metric	Value	Assessment
Total Wall Time	4,263.1 ms	Baseline
Total Thread Work (de-duped)	13,134.1 ms	Non-overlapping busy time
Effective Parallelism	3.1x	thread_work / wall_time
Working Threads	9	Threads with actual spans
Thread Utilization	34.2%	⚠️ Suboptimal
Total Spans	879,490	All B/E + X events
Meaningful Spans (>= 10us)	162,988	(18.5% of total)
Tracing Noise (< 10us)	716,502	(81.5% 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	31,790	3,629.6	884.0	1,937.9
Parse	8,136	898.3	666.2	3,475.6
Analyze	107,872	26,621.0	4,829.9	3,431.8
Chunk	4,535	3,178.0	450.7	1,400.8
Codegen	8,661	1,313.3	844.0	1,398.9
Emit	28	31.8	15.9	7.8
Other	1,966	7,654.1	4,088.8	4,263.1

---

Workload Distribution by Diagnostic Tier

Category	Spans	Inclusive (ms)	% Work	Self-Time (ms)	% Self
P0: Scheduling & Resolution	140,240	31,977.8	243.5%	6,043.2	46.0%
P1: I/O & Heavy Tasks	2,688	84.1	0.6%	68.2	0.5%
P2: Architecture (Locks/Memory)	0	0.0	0.0%	0.0	0.0%
P3: Asset Pipeline	19,169	5,350.5	40.7%	1,921.7	14.6%
P4: Bridge/Interop	0	0.0	0.0%	0.0	0.0%
Other	891	5,913.7	45.0%	3,746.3	28.5%

---

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
2,862.0	16,922.9	80,177	35.7	0.1	15.6	21.8%	module	module (60%)
2,831.9	2,969.6	19	149047.1	407.4	620.5	21.6%	save snapshot	persist (5%)
1,218.5	1,273.6	2,072	588.1	1.7	172.0	9.3%	analyze ecmascript module	module (83%)
731.7	8,399.2	25,597	28.6	0.0	13.3	5.6%	process module	process module (79%)
621.7	1,808.3	19,374	32.1	0.1	14.9	4.7%	internal resolving	internal resolving (71%)
618.6	850.8	5,963	103.7	0.4	38.2	4.7%	parse ecmascript	parse ecmascript (66%)
474.0	1,453.0	662	716.0	1.5	179.3	3.6%	generate merged code	chunking (92%)
435.6	489.6	6,679	65.2	0.2	11.4	3.3%	precompute code generation	generate merged code (77%)
356.5	2,968.4	3,784	94.2	0.1	25.5	2.7%	chunking	chunking (23%)
332.2	1,678.1	792	419.4	0.3	69.7	2.5%	write all entrypoints to disk	write all entrypoints to disk (13%)
292.5	292.5	328	891.7	0.6	191.7	2.2%	generate source map	code generation (83%)
280.6	572.3	130	2158.7	4.3	170.3	2.1%	emit code	emit code (48%)
257.5	1,816.6	11,929	21.6	0.0	4.1	2.0%	resolving	module (51%)
138.6	138.6	10	13859.8	72.2	124.6	1.1%	blocking	save snapshot (50%)
115.9	531.2	1,654	70.1	0.1	35.8	0.9%	code generation	code generation (83%)
91.9	206.8	724	126.9	0.3	17.5	0.7%	compute async chunks	compute async chunks (60%)
47.5	47.5	2,170	21.9	0.0	1.1	0.4%	read file	parse ecmascript (91%)
15.8	15.8	12	1316.6	4.7	5.1	0.1%	write file	apply effects (100%)
12.9	13.1	7	1848.9	8.7	12.2	0.1%	create project	None (0%)
10.3	62.2	283	36.5	0.1	1.2	0.1%	async reference	async reference (45%)

---

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	745.1	3	persist → save snapshot → blocking
2	351.7	6	chunking → generate merged code → emit code → emit code → emit code → read file
3	225.3	4	chunking → generate merged code → emit code → generate source map
4	179.5	2	save snapshot → blocking
5	172.3	10	module → module → process module → ... → analyze ecmascript module → analyze ecmascript 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	25,597	process module (79%)	28.6 us	0.04 ms	731.7 ms
internal resolving	19,374	internal resolving (71%)	32.1 us	0.05 ms	621.7 ms
precompute code generation	6,679	generate merged code (77%)	65.2 us	0.23 ms	435.6 ms

---

Duration Distribution

Range	Count	Percentage
<10us	716,502	81.5%
10us-100us	103,850	11.8%
100us-1ms	53,025	6.0%
1ms-10ms	5,929	0.7%
10ms-100ms	150	0.0%
>100ms	34	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