fix(pack): circular dependency reexport

[fireairforce]

Summary

fix circurlar reexport cause runtime error

ref pr: utooland/next.js#144

closes: #2582

Test Plan

add test case here

[gemini-code-assist]

Code Review

This pull request updates and introduces several test snapshots for Turbopack, focusing on circular dependencies and tree-shaking optimizations. It adds a new test case for circular dependencies involving barrel re-exports and a regression test for a bug where imports from side-effect-free barrels were incorrectly dropped during optimization. Additionally, existing snapshots are updated to reflect fixes in module concatenation and unused import removal, and the next.js subproject reference is updated. I have no feedback to provide as there are no review comments.

[github-actions]

📊 Performance Benchmark Report (with-antd)

Utoopack Performance Report

Report ID: utoopack_performance_report_20260506_102130
Generated: 2026-05-06 10:21:30
Trace File: trace_antd.json (0.5GB, 1.57M spans)
Test Project: examples/with-antd

---

Executive Summary

Metric	Value	Assessment
Total Wall Time	12,338.2 ms	Baseline
Total Thread Work (de-duped)	37,861.7 ms	Non-overlapping busy time
Effective Parallelism	3.1x	thread_work / wall_time
Working Threads	5	Threads with actual spans
Thread Utilization	61.4%	🆗 Average
Total Spans	1,567,749	All B/E + X events
Meaningful Spans (>= 10us)	409,515	(26.1% of total)
Tracing Noise (< 10us)	1,158,234	(73.9% 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	94,702	6,310.7	5,131.8	8,523.9
Parse	11,111	2,773.9	2,448.2	12,215.6
Analyze	242,755	26,893.0	19,695.2	11,794.6
Chunk	20,513	2,200.0	2,041.3	9,588.9
Codegen	31,146	4,245.3	3,142.4	9,666.4
Emit	38	41.5	20.6	9,693.2
Other	9,250	1,700.6	1,282.2	12,233.8

---

Workload Distribution by Diagnostic Tier

Category	Spans	Inclusive (ms)	% Work	Self-Time (ms)	% Self
P0: Scheduling & Resolution	343,995	34,351.7	90.7%	25,579.3	67.6%
P1: I/O & Heavy Tasks	3,074	180.2	0.5%	159.3	0.4%
P2: Architecture (Locks/Memory)	1	0.0	0.0%	0.0	0.0%
P3: Asset Pipeline	61,263	9,165.3	24.2%	7,577.9	20.0%
P4: Bridge/Interop	0	0.0	0.0%	0.0	0.0%
Other	1,182	467.8	1.2%	445.1	1.2%

---

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
12,345.1	16,409.0	157,399	78.4	0.1	43.6	32.6%	module	write all entrypoints to disk (1%)
3,564.6	3,667.6	28,948	123.1	0.2	137.3	9.4%	analyze ecmascript module	module (66%)
2,858.5	5,858.4	46,161	61.9	0.1	25.6	7.5%	process module	module (16%)
2,691.5	2,853.0	51,515	52.2	0.0	21.6	7.1%	internal resolving	resolving (30%)
2,429.5	3,446.9	42,498	57.2	0.0	20.1	6.4%	resolving	module (20%)
2,067.0	2,194.5	8,006	258.2	0.8	77.6	5.5%	parse ecmascript	process module (28%)
1,831.7	2,934.6	18,498	99.0	0.4	66.7	4.8%	code generation	chunking (2%)
1,370.5	1,501.6	13,169	104.1	0.3	98.0	3.6%	chunking	write all entrypoints to disk (0%)
979.4	979.4	10,565	92.7	0.3	6.4	2.6%	precompute code generation	code generation (55%)
816.7	1,205.8	7,416	110.1	0.0	188.7	2.2%	write all entrypoints to disk	None (0%)
805.5	805.5	7,587	106.2	0.4	97.7	2.1%	compute async module info	chunking (0%)
650.3	650.7	7,259	89.6	0.1	60.2	1.7%	compute async chunks	webpack loader (0%)
409.8	422.9	1,058	387.3	1.9	23.3	1.1%	webpack loader	parse css (14%)
331.3	331.3	2,083	159.1	0.4	12.8	0.9%	generate source map	code generation (96%)
253.3	451.4	758	334.1	1.2	23.4	0.7%	parse css	module (6%)
127.8	127.8	2,341	54.6	0.1	13.5	0.3%	read file	parse ecmascript (91%)
74.0	74.0	840	88.1	0.0	20.8	0.2%	compute binding usage info	write all entrypoints to disk (1%)
30.7	30.7	1,793	17.1	0.0	7.0	0.1%	collect mergeable modules	compute merged modules (0%)
20.5	47.7	85	241.1	0.4	15.1	0.1%	make production chunks	chunking (6%)
20.4	27.0	651	31.4	0.1	2.7	0.1%	async reference	write all entrypoints to disk (1%)

---

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	137.4	2	process module → analyze ecmascript module
2	79.5	2	code generation → generate source map
3	77.7	2	process module → parse ecmascript
4	71.4	2	module → analyze ecmascript module
5	62.1	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
No obvious batching candidates found	-	-	-	-	-

---

Duration Distribution

Range	Count	Percentage
<10us	1,158,234	73.9%
10us-100us	382,262	24.4%
100us-1ms	20,294	1.3%
1ms-10ms	6,485	0.4%
10ms-100ms	470	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