feat(pack): support externalType: promise

[fireairforce]

Summary

ref: utooland/next.js#142

For business support:

Test Plan

add snapshot test.

[gemini-code-assist]

Code Review

This pull request introduces support for "promise" type externals across the core Rust logic, configuration schemas, and TypeScript compatibility layers. Key changes include adding the "promise" variant to external type enums, updating resolution logic to handle the "promise" prefix, and implementing a new snapshot test. Feedback suggests using the matches! macro in tests for better conciseness and refactoring duplicated logic in the Webpack compatibility module to improve maintainability.

[github-actions]

📊 Performance Benchmark Report (with-antd)

Utoopack Performance Report

Report ID: utoopack_performance_report_20260428_081754
Generated: 2026-04-28 08:17:54
Trace File: trace_antd.json (0.5GB, 1.57M spans)
Test Project: examples/with-antd

---

Executive Summary

Metric	Value	Assessment
Total Wall Time	11,392.5 ms	Baseline
Total Thread Work (de-duped)	21,462.3 ms	Non-overlapping busy time
Effective Parallelism	1.9x	thread_work / wall_time
Working Threads	5	Threads with actual spans
Thread Utilization	37.7%	⚠️ Suboptimal
Total Spans	1,567,470	All B/E + X events
Meaningful Spans (>= 10us)	361,094	(23.0% of total)
Tracing Noise (< 10us)	1,206,376	(77.0% 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	79,671	2,178.1	1,754.2	8,317.7
Parse	9,861	1,823.3	1,535.4	11,266.4
Analyze	213,986	12,872.8	8,686.8	11,022.9
Chunk	19,812	1,690.2	1,559.2	8,337.6
Codegen	28,596	3,804.3	2,819.2	8,485.8
Emit	39	42.6	21.2	7,571.4
Other	9,129	1,194.5	1,040.4	11,392.5

---

Workload Distribution by Diagnostic Tier

Category	Spans	Inclusive (ms)	% Work	Self-Time (ms)	% Self
P0: Scheduling & Resolution	300,177	15,829.2	73.8%	11,082.7	51.6%
P1: I/O & Heavy Tasks	3,045	165.3	0.8%	144.0	0.7%
P2: Architecture (Locks/Memory)	0	0.0	0.0%	0.0	0.0%
P3: Asset Pipeline	56,764	7,258.0	33.8%	5,854.0	27.3%
P4: Bridge/Interop	0	0.0	0.0%	0.0	0.0%
Other	1,108	353.2	1.6%	335.7	1.6%

---

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
4,413.5	5,234.9	114,015	38.7	0.1	11.1	20.6%	module	write all entrypoints to disk (1%)
2,474.8	3,494.8	35,595	69.5	0.1	178.5	11.5%	analyze ecmascript module	process module (80%)
1,618.9	2,604.0	16,820	96.2	0.4	44.6	7.5%	code generation	chunking (1%)
1,134.3	1,249.4	6,959	163.0	0.6	36.3	5.3%	parse ecmascript	analyze ecmascript module (32%)
1,036.0	3,354.7	55,751	18.6	0.0	6.2	4.8%	process module	module (9%)
1,022.9	1,100.3	44,550	23.0	0.0	8.0	4.8%	internal resolving	resolving (29%)
991.5	1,100.8	11,679	84.9	0.2	21.5	4.6%	chunking	write all entrypoints to disk (0%)
841.5	841.5	9,694	86.8	0.3	7.4	3.9%	precompute code generation	code generation (58%)
724.0	1,070.4	34,461	21.0	0.0	6.0	3.4%	resolving	module (11%)
669.6	801.9	7,453	89.8	0.0	108.5	3.1%	write all entrypoints to disk	None (0%)
667.1	667.1	6,783	98.4	0.4	57.4	3.1%	compute async module info	chunking (0%)
547.1	547.4	8,044	68.0	0.0	32.3	2.5%	compute async chunks	write all entrypoints to disk (0%)
358.8	358.8	2,082	172.3	0.4	16.8	1.7%	generate source map	code generation (96%)
296.1	306.3	1,010	293.2	1.4	14.0	1.4%	webpack loader	parse css (11%)
285.7	458.4	556	513.8	2.4	61.5	1.3%	parse css	module (8%)
115.4	115.4	2,341	49.3	0.2	3.1	0.5%	read file	parse ecmascript (91%)
55.6	55.6	841	66.2	0.0	11.2	0.3%	compute binding usage info	write all entrypoints to disk (0%)
35.1	39.4	568	61.7	0.1	6.2	0.2%	async reference	write all entrypoints to disk (1%)
24.3	24.3	979	24.8	0.0	6.3	0.1%	collect mergeable modules	compute merged modules (0%)
20.8	20.8	36	578.1	3.1	6.6	0.1%	write file	apply effects (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	178.6	2	process module → analyze ecmascript module
2	97.4	2	process module → analyze ecmascript module
3	64.2	2	module → parse css
4	61.4	2	code generation → generate source map
5	41.0	2	code generation → generate source map

---

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
analyze ecmascript module	35,595	process module (80%)	69.5 us	0.14 ms	2,474.8 ms

---

Duration Distribution

Range	Count	Percentage
<10us	1,206,376	77.0%
10us-100us	338,803	21.6%
100us-1ms	18,478	1.2%
1ms-10ms	3,712	0.2%
10ms-100ms	97	0.0%
>100ms	4	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