[fireairforce]

Summary

Changing a library name or export path changes its entry wrapper even when the module factories remain unchanged. Previously, these builds could share a content-hashed URL and serve a stale wrapper from cache.

Include the library runtime configuration in evaluate-chunk content hashes, including entries with no other chunks. Keep non-entry chunk hashing unchanged and avoid hashing the entry's own output path or generated content.

Add a production regression fixture that checks identical-build stability, filename and body changes for library name/export changes, and the actual browser-global exports.

Refresh the four affected library/server-entry snapshot fixtures. The snapshot changes are limited to entry filenames and their references; shared chunks and map/LICENSE contents are unchanged.

Test Plan

Verified on this branch, based on next at 0603a3f639e0cba5a76d87cffb6c8629e2a83d97:

• cargo fmt --all — passed.
• cargo clippy --offline --locked --all-targets -- -D warnings --no-deps — passed.
• Biome CI and spelling checks — passed.
• git diff --check — passed.
• Reproduced the original snapshot mismatch, then verified all four affected snapshot tests without UPDATE — passed.
• Rebuilt pack-napi --features plugin from this branch and ran npm test --workspace @utoo/pack -- src/__test__/libraryContentHash.test.ts — passed.

The full Rust suite is delegated to CI: local loader tests cannot bind localhost sockets in the sandbox. The original CI failure started with the stale multi_server_entries snapshot; its panic poisoned the shared test lock, causing the subsequent PoisonError failures.

[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 2026-10-28T11:32:16.964709Z	0360829	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_20261028_124248
Generated: 2026-10-28 12:42:48
Trace File: trace_antd.json (0.3GB, 0.84M spans)
Test Project: examples/with-antd

---

Executive Summary

Metric	Value	Assessment
Total Wall Time	6,952.3 ms	Baseline
Total Thread Work (de-duped)	21,278.0 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	844,560	All B/E + X events
Meaningful Spans (>= 10us)	232,778	(27.6% of total)
Tracing Noise (< 10us)	611,782	(72.4% 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	51,214	7,138.7	1,975.1	3,471.1
Parse	8,203	1,432.0	1,080.8	6,243.5
Analyze	152,265	47,726.2	9,961.1	6,149.6
Chunk	5,455	5,991.0	795.9	2,445.6
Codegen	12,304	2,596.6	1,563.7	2,104.9
Emit	33	43.6	21.9	8.0
Other	3,304	7,349.8	4,252.8	6,952.3

---

Workload Distribution by Diagnostic Tier

Category	Spans	Inclusive (ms)	% Work	Self-Time (ms)	% Self
P0: Scheduling & Resolution	204,185	55,507.0	260.9%	12,213.7	57.4%
P1: I/O & Heavy Tasks	2,894	115.7	0.5%	94.0	0.4%
P2: Architecture (Locks/Memory)	0	0.0	0.0%	0.0	0.0%
P3: Asset Pipeline	24,551	10,036.7	47.2%	3,441.9	16.2%
P4: Bridge/Interop	0	0.0	0.0%	0.0	0.0%
Other	1,148	6,618.6	31.1%	3,901.8	18.3%

---

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
4,973.2	27,918.6	103,273	48.2	0.1	21.4	23.4%	module	module (63%)
2,722.2	2,844.2	19	143273.6	348.7	617.9	12.8%	save snapshot	persist (5%)
2,456.2	2,594.1	2,613	940.0	3.1	329.6	11.5%	analyze ecmascript module	module (67%)
1,659.3	16,118.4	36,413	45.6	0.1	9.3	7.8%	process module	process module (83%)
1,292.0	3,472.5	27,893	46.3	0.1	4.7	6.1%	internal resolving	internal resolving (77%)
1,020.5	1,371.7	6,028	169.3	0.6	52.3	4.8%	parse ecmascript	parse ecmascript (65%)
863.8	951.1	10,020	86.2	0.4	6.7	4.1%	precompute code generation	generate merged code (42%)
754.9	862.2	7,311	103.3	0.4	124.9	3.5%	compute async module info	compute async module info (57%)
671.3	3,654.4	22,635	29.7	0.0	4.6	3.2%	resolving	module (56%)
667.1	5,684.1	4,055	164.5	0.2	36.5	3.1%	chunking	chunking (52%)
652.2	2,144.4	878	742.9	1.8	237.2	3.1%	generate merged code	chunking (49%)
455.7	455.7	329	1385.1	1.1	283.7	2.1%	generate source map	code generation (83%)
378.4	806.2	129	2933.4	6.0	215.2	1.8%	emit code	generate merged code (41%)
319.8	586.1	1,718	186.2	0.1	163.0	1.5%	write all entrypoints to disk	write all entrypoints to disk (21%)
244.3	1,189.8	1,955	125.0	0.2	56.0	1.1%	code generation	code generation (83%)
125.0	125.0	8	15619.6	64.0	71.3	0.6%	blocking	save snapshot (50%)
124.9	302.6	1,340	93.2	0.2	11.6	0.6%	compute async chunks	compute async chunks (43%)
61.7	77.4	764	80.8	0.0	29.7	0.3%	compute binding usage info	compute binding usage info (51%)
60.3	60.3	2,173	27.7	0.0	1.8	0.3%	read file	parse ecmascript (91%)
41.9	75.1	1,870	22.4	0.0	14.5	0.2%	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	668.5	3	persist → save snapshot → blocking
2	455.9	6	chunking → generate merged code → emit code → emit code → emit code → read file
3	334.4	9	module → module → process module → ... → process module → analyze ecmascript module → analyze ecmascript module
4	325.1	4	chunking → generate merged code → emit code → generate source map
5	303.9	2	save snapshot → blocking

---

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	36,413	process module (83%)	45.6 us	0.07 ms	1,659.3 ms
internal resolving	27,893	internal resolving (77%)	46.3 us	0.10 ms	1,292.0 ms

---

Duration Distribution

Range	Count	Percentage
<10us	611,782	72.4%
10us-100us	152,326	18.0%
100us-1ms	68,195	8.1%
1ms-10ms	11,993	1.4%
10ms-100ms	221	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