[fireairforce]

Summary

Explicit externals for URI requests such as virtual:request/h5 were skipped by the module-name filter. With an alias, the target was bundled alongside the direct external import, splitting mutable state between two instances. Without an alias, the original URI was loaded instead of the configured external target.

Match complete, literal request names before alias resolution so these imports use the configured external instance. Add executable Node regressions for alias plus externals, externals only, unscoped package subpaths, and literal glob characters.

Closes #3427.

CI pins Ubuntu 24.04 and the tested ecosystem harness revision from utooland/utoo-ecosystem-ci#10. This restores the supported environment after ubuntu-latest moved to Ubuntu 26.04, which broke the wasm-bindgen installer and EVJS Playwright browser installation before the candidate tests ran.

Test Plan

• cargo test --locked -p pack-tests --test snapshot externals
• cargo fmt --check
• cargo clippy --all-targets -- -D warnings --no-deps
• npx tombi format --check, npx biome ci, and typos
• Actionlint for both modified workflows; ecosystem harness tests (45 passed), lint, and workflow policy checks

[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 2026-10-27T09:19:07.382430Z	850ef7f	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_20261027_100047
Generated: 2026-10-27 10:00:47
Trace File: trace_antd.json (0.3GB, 0.84M spans)
Test Project: examples/with-antd

---

Executive Summary

Metric	Value	Assessment
Total Wall Time	5,849.7 ms	Baseline
Total Thread Work (de-duped)	17,993.6 ms	Non-overlapping busy time
Effective Parallelism	3.1x	thread_work / wall_time
Working Threads	10	Threads with actual spans
Thread Utilization	30.8%	⚠️ Suboptimal
Total Spans	844,069	All B/E + X events
Meaningful Spans (>= 10us)	183,084	(21.7% of total)
Tracing Noise (< 10us)	660,985	(78.3% 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	37,765	5,515.7	1,451.7	2,727.4
Parse	8,148	1,174.5	871.4	4,963.5
Analyze	122,190	35,918.4	7,674.2	4,888.2
Chunk	4,773	4,615.0	763.4	1,964.9
Codegen	8,054	1,771.5	1,142.9	1,703.5
Emit	29	36.3	18.2	7.8
Other	2,125	7,416.3	4,544.3	5,849.7

---

Workload Distribution by Diagnostic Tier

Category	Spans	Inclusive (ms)	% Work	Self-Time (ms)	% Self
P0: Scheduling & Resolution	160,218	41,909.1	232.9%	9,329.8	51.9%
P1: I/O & Heavy Tasks	2,602	104.7	0.6%	86.6	0.5%
P2: Architecture (Locks/Memory)	0	0.0	0.0%	0.0	0.0%
P3: Asset Pipeline	19,436	7,565.3	42.0%	2,769.6	15.4%
P4: Bridge/Interop	0	0.0	0.0%	0.0	0.0%
Other	828	6,868.5	38.2%	4,280.1	23.8%

---

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
3,684.2	21,246.5	82,524	44.6	0.1	14.5	20.5%	module	module (60%)
3,252.0	3,400.7	18	180665.1	373.2	701.7	18.1%	save snapshot	persist (6%)
2,034.0	2,169.4	2,280	892.1	3.0	209.3	11.3%	analyze ecmascript module	module (76%)
1,306.7	11,661.9	29,718	44.0	0.1	11.1	7.3%	process module	process module (81%)
986.4	2,725.2	20,937	47.1	0.1	6.0	5.5%	internal resolving	internal resolving (71%)
814.4	1,117.5	5,975	136.3	0.5	58.7	4.5%	parse ecmascript	parse ecmascript (66%)
675.3	4,395.4	3,873	174.4	0.1	151.8	3.8%	chunking	chunking (46%)
618.9	679.9	5,985	103.4	0.4	7.0	3.4%	precompute code generation	generate merged code (54%)
565.1	671.9	5,630	100.4	0.3	99.5	3.1%	compute async module info	compute async module info (50%)
539.2	1,765.0	598	901.7	2.1	175.7	3.0%	generate merged code	chunking (76%)
453.9	2,779.1	16,428	27.6	0.0	5.1	2.5%	resolving	module (62%)
374.4	374.4	329	1138.0	0.7	225.1	2.1%	generate source map	code generation (83%)
314.8	678.3	129	2440.4	4.0	188.1	1.7%	emit code	generate merged code (41%)
247.1	449.9	974	253.7	0.1	137.7	1.4%	write all entrypoints to disk	write all entrypoints to disk (10%)
149.6	717.2	1,740	86.0	0.2	39.3	0.8%	code generation	code generation (83%)
149.5	149.5	8	18685.0	67.8	82.1	0.8%	blocking	save snapshot (75%)
85.7	216.6	868	98.7	0.2	9.2	0.5%	compute async chunks	compute async chunks (55%)
57.0	57.0	2,170	26.3	0.0	7.1	0.3%	read file	parse ecmascript (91%)
49.0	61.3	634	77.2	0.0	26.2	0.3%	compute binding usage info	compute binding usage info (56%)
24.3	48.0	1,387	17.5	0.0	9.3	0.1%	collect mergeable modules	collect mergeable modules (100%)

---

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	783.8	3	persist → save snapshot → blocking
2	366.7	6	chunking → generate merged code → emit code → emit code → emit code → read file
3	260.4	4	chunking → generate merged code → emit code → generate source map
4	245.7	2	save snapshot → blocking
5	209.4	3	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	29,718	process module (81%)	44.0 us	0.05 ms	1,306.7 ms
internal resolving	20,937	internal resolving (71%)	47.1 us	0.07 ms	986.4 ms

---

Duration Distribution

Range	Count	Percentage
<10us	660,985	78.3%
10us-100us	115,443	13.7%
100us-1ms	58,379	6.9%
1ms-10ms	9,033	1.1%
10ms-100ms	191	0.0%
>100ms	38	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

[PeachScript]

+1