Specialize compiled invoked child startup

[SandroMaglione]

Summary

• start eligible compiled invoked machines directly from their synchronous initial kernel while preserving the existing Effect-based transition, service, scope, observation, and completion semantics
• compile each Machine.invokeMachine process descriptor once while keeping initialization and runtime state isolated per parent instance
• cover input-bearing descriptor reuse and initially-final child completion

Changeset

• Added or updated for a library or package-metadata change
• Not required because this PR does not change src/ or package.json

Validation

• pnpm check
• Relevant example checks, when examples changed (no examples changed)
• Reviewed the automated type-performance report, when the public TypeScript API or inference changed (local budgets pass; no public API change)
• Reviewed the automated runtime-performance report, when runtime behavior changed

The final five-process CI comparison improves parent-plus-child lifecycle from 39,373 to 45,304 families/s (+15.1%) with standalone lifecycle +0.3%. Parent-plus-child heap changes by +0.5% (about 25 bytes); local focused runs measured +13.9% aggregate / +14.7% paired-median lifecycle with all nine pairs positive.

[github-actions]

Type performance

Measured with TypeScript 6.0.3 and skipLibCheck=true.

Scenario	Base	PR	Difference
Effect only	55	55	0 (0.0%)
Import effect-machine	55	55	0 (0.0%)
Machine.defineStates (3 states)	2,965	2,965	0 (0.0%)
Machine.make (3 states, 2 events)	8,715	8,715	0 (0.0%)
machine.handle (3 states, 2 transitions)	23,492	23,492	0 (0.0%)
machine.handle (depth 8)	117,075	117,075	0 (0.0%)
machine.handle (depth 12)	129,713	129,713	0 (0.0%)
machine.handle (depth 16)	145,055	145,055	0 (0.0%)
machine.handle (depth 24)	183,851	183,851	0 (0.0%)
machine.handle (wide depth 16)	214,189	214,189	0 (0.0%)
machine.handle (parallel/history/choice)	132,667	132,667	0 (0.0%)
machine.handle (4 successive calls)	128,674	128,674	0 (0.0%)
machine exact input/output/error/services	110,584	110,584	0 (0.0%)
execution adapter readiness	127,854	127,854	0 (0.0%)

Marginal instantiations are measured against the matching setup without that API call:

Scenario	Base	PR	Difference
Import effect-machine	0	0	0
Machine.defineStates (3 states)	2,910	2,910	0 (0.0%)
Machine.make (3 states, 2 events)	5,742	5,742	0 (0.0%)
machine.handle (3 states, 2 transitions)	14,777	14,777	0 (0.0%)
machine.handle (depth 8)	109,601	109,601	0 (0.0%)
machine.handle (depth 12)	121,039	121,039	0 (0.0%)
machine.handle (depth 16)	135,181	135,181	0 (0.0%)
machine.handle (depth 24)	171,577	171,577	0 (0.0%)
machine.handle (wide depth 16)	202,224	202,224	0 (0.0%)
machine.handle (parallel/history/choice)	114,331	114,331	0 (0.0%)
machine.handle (4 successive calls)	113,703	113,703	0 (0.0%)
machine exact input/output/error/services	100,670	100,670	0 (0.0%)
execution adapter readiness	102,092	102,092	0 (0.0%)

Check times (informational)

Scenario	Base	PR
Effect only	0.03s	0.03s
Import effect-machine	0.03s	0.03s
Machine.defineStates (3 states)	0.12s	0.10s
Machine.make (3 states, 2 events)	0.16s	0.16s
machine.handle (3 states, 2 transitions)	0.24s	0.24s
machine.handle (depth 8)	0.51s	0.52s
machine.handle (depth 12)	0.55s	0.56s
machine.handle (depth 16)	0.61s	0.61s
machine.handle (depth 24)	0.69s	0.70s
machine.handle (wide depth 16)	0.78s	0.78s
machine.handle (parallel/history/choice)	0.60s	0.63s
machine.handle (4 successive calls)	0.58s	0.60s
machine exact input/output/error/services	0.51s	0.50s
execution adapter readiness	0.58s	0.60s

Type instantiations are the comparison metric. Check time varies with runner load and is informational only.

[github-actions]

Runtime performance

Median of 5 independent benchmark processes on AMD EPYC 9V74 80-Core Processor with Node v24.18.0.

Pull request baseline

Scenario	Effect Machine	XState 5	XState 6 alpha
Plan counter transitions	164,199 transitions/s	171,941 transitions/s	142,211 transitions/s
Drain burst with terminal fence	633,097 increments/s	485,631 increments/s	132,816 increments/s
Drain burst with a change observer	603,422 increments/s	481,184 increments/s	132,275 increments/s
Lookup and send to one child	531,737 increments/s	479,745 increments/s	128,853 increments/s
Start and stop a machine	145,985 machines/s	226,912 machines/s	91,861 machines/s
Start and stop a parent with one child	45,304 families/s	90,359 families/s	38,374 families/s
Drain burst through a compound state	629,282 events/s	424,553 events/s	131,785 events/s
Drain burst through two parallel regions	589,876 events/s	325,361 events/s	113,858 events/s

Effect runtime reference points

Scenario	Effect Machine	Effect runtime primitives
Start and stop a raw compiled process	91,017 processes/s	—
Start and interrupt a suspended fiber	—	201,288 fibers/s
Start and stop a queue worker	—	147,275 workers/s
Start and stop an actor shell	—	135,667 actors/s
Start and stop two actor shells	—	85,558 families/s
Update an owner-only mutable snapshot	—	401,123,145 updates/s
Update a synchronized snapshot	—	2,128,991 updates/s
Create, resolve, and await a terminal latch	—	2,104,975 latches/s

Memory profile	Effect Machine	XState 5	XState 6 alpha	Effect runtime primitives
Idle machine	1.6 KiB	3.8 KiB	6.1 KiB	—
Raw generic managed process	12.1 KiB	—	—	—
Raw compiled process	2.6 KiB	—	—	—
Two independent idle machines	3.0 KiB	7.3 KiB	11.8 KiB	—
Idle parent with one child	4.9 KiB	5.5 KiB	8.8 KiB	—
Parent with observed child registry	9.1 KiB	—	—	—
Parent with observed invoked child snapshots	5.0 KiB	—	—	—
Suspended Effect fiber	—	—	—	0.7 KiB
Effect queue with waiting fiber	—	—	—	2.8 KiB
Effect mailbox actor shell	—	—	—	3.2 KiB
Two Effect actor shells	—	—	—	6.9 KiB

Effect Machine change from base

Metric	Base	Base variability	PR	PR variability	Difference
Plan counter transitions	164,415 transitions/s	0.6% MAD	164,199 transitions/s	0.1% MAD	-0.1%
Drain burst with terminal fence	672,676 increments/s	0.6% MAD	633,097 increments/s	3.8% MAD	-5.9%
Drain burst with a change observer	615,003 increments/s	0.3% MAD	603,422 increments/s	1.8% MAD	-1.9%
Lookup and send to one child	543,487 increments/s	0.2% MAD	531,737 increments/s	2.3% MAD	-2.2%
Start and stop a machine	145,560 machines/s	1.3% MAD	145,985 machines/s	0.3% MAD	+0.3%
Start and stop a parent with one child	39,373 families/s	0.5% MAD	45,304 families/s	1.3% MAD	+15.1%
Drain burst through a compound state	639,177 events/s	2.0% MAD	629,282 events/s	3.0% MAD	-1.5%
Drain burst through two parallel regions	584,503 events/s	0.8% MAD	589,876 events/s	1.4% MAD	+0.9%
Idle machine heap per unit	1.6 KiB	0.2% MAD	1.6 KiB	0.0% MAD	-0.1%
Raw generic managed process heap per unit	12.1 KiB	0.0% MAD	12.1 KiB	0.0% MAD	+0.1%
Raw compiled process heap per unit	2.6 KiB	0.0% MAD	2.6 KiB	0.2% MAD	+1.1%
Two independent idle machines heap per unit	3.0 KiB	0.0% MAD	3.0 KiB	0.0% MAD	-0.0%
Idle parent with one child heap per unit	4.9 KiB	0.0% MAD	4.9 KiB	0.0% MAD	+0.5%
Parent with observed child registry heap per unit	9.1 KiB	0.0% MAD	9.1 KiB	0.0% MAD	+0.3%
Parent with observed invoked child snapshots heap per unit	5.0 KiB	0.0% MAD	5.0 KiB	0.0% MAD	+1.4%

Effect runtime reference change from base

Metric	Base	Base variability	PR	PR variability	Difference
Start and stop a raw compiled process	89,550 processes/s	0.4% MAD	91,017 processes/s	0.9% MAD	+1.6%
Start and interrupt a suspended fiber	203,376 fibers/s	0.2% MAD	201,288 fibers/s	0.6% MAD	-1.0%
Start and stop a queue worker	148,368 workers/s	0.7% MAD	147,275 workers/s	0.6% MAD	-0.7%
Start and stop an actor shell	135,117 actors/s	2.1% MAD	135,667 actors/s	1.8% MAD	+0.4%
Start and stop two actor shells	85,273 families/s	0.9% MAD	85,558 families/s	1.1% MAD	+0.3%
Update an owner-only mutable snapshot	400,962,309 updates/s	0.0% MAD	401,123,145 updates/s	0.0% MAD	+0.0%
Update a synchronized snapshot	2,106,571 updates/s	0.3% MAD	2,128,991 updates/s	1.7% MAD	+1.1%
Create, resolve, and await a terminal latch	2,137,602 latches/s	1.3% MAD	2,104,975 latches/s	0.3% MAD	-1.5%

Versions and interpretation

• Effect Machine: 0.3.0
• XState 5: 5.32.5
• XState 6 alpha: 6.0.0-alpha.27
• Effect runtime primitives: 4.0.0-beta.105

Higher throughput is better; lower heap is better. Variability is the median absolute deviation across independent processes, relative to their median. Runtime measurements on shared GitHub-hosted hardware remain informational, so small differences should be confirmed across multiple workflow runs.