Harden compiled execution contracts

[SandroMaglione]

Summary

• Give compiled execution plans a runtime-only macrostep and microstep contract instead of claiming to retain generic planner transition diagnostics.
• Name and document the indexed kernel state as process-owned, and funnel its flat-path mutation through an explicit ownership helper.
• Freeze shared empty execution arrays and add direct differential witnesses for targetless, reentering, value-only, simultaneous control-changing, and retained-snapshot behavior.

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 (not applicable; no examples changed)
• Reviewed the automated type-performance report, when the public TypeScript API or inference changed (public API unchanged; local instantiation counts are identical)
• Reviewed the automated runtime-performance report, when runtime behavior changed (five alternating local quick samples are flat or faster in all machine throughput scenarios; hosted full-size comparison remains the merge gate)

[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	0.04s
Machine.defineStates (3 states)	0.10s	0.11s
Machine.make (3 states, 2 events)	0.16s	0.16s
machine.handle (3 states, 2 transitions)	0.25s	0.26s
machine.handle (depth 8)	0.53s	0.53s
machine.handle (depth 12)	0.60s	0.59s
machine.handle (depth 16)	0.65s	0.63s
machine.handle (depth 24)	0.71s	0.73s
machine.handle (wide depth 16)	0.82s	0.79s
machine.handle (parallel/history/choice)	0.67s	0.64s
machine.handle (4 successive calls)	0.59s	0.62s
machine exact input/output/error/services	0.52s	0.53s
execution adapter readiness	0.61s	0.61s

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 7763 64-Core Processor with Node v24.18.0.

Pull request baseline

Scenario	Effect Machine	XState 5	XState 6 alpha
Plan counter transitions	122,484 transitions/s	132,973 transitions/s	13,130 transitions/s
Drain burst with terminal fence	485,250 increments/s	392,344 increments/s	194,727 increments/s
Drain burst with a change observer	456,413 increments/s	388,628 increments/s	197,060 increments/s
Lookup and send to one child	389,511 increments/s	386,435 increments/s	187,293 increments/s
Start and stop a machine	172,980 machines/s	183,824 machines/s	145,921 machines/s
Start and stop a parent with one child	37,147 families/s	68,129 families/s	41,762 families/s
Plan transitions through a compound state	117,913 transitions/s	—	—
Plan transitions through parallel regions	89,556 transitions/s	—	—
Drain burst through a compound state	522,138 events/s	344,459 events/s	192,550 events/s
Drain burst through two parallel regions	477,852 events/s	262,420 events/s	145,009 events/s
Drain a compound-state burst with a change observer	476,620 events/s	—	—

Effect runtime reference points

Scenario	Effect Machine	Effect runtime primitives
Start and stop a raw generic process	18,518 processes/s	—
Start and stop a raw compiled process	68,549 processes/s	—
Start and interrupt a suspended fiber	—	166,085 fibers/s
Start and stop a queue worker	—	115,115 workers/s
Start and stop an actor shell	—	102,270 actors/s
Start and stop two actor shells	—	64,855 families/s
Update an owner-only mutable snapshot	—	316,856,781 updates/s
Update a synchronized snapshot	—	1,512,781 updates/s
Create, resolve, and await a terminal latch	—	1,465,244 latches/s

Memory profile	Effect Machine	XState 5	XState 6 alpha	Effect runtime primitives
Idle machine	1.5 KiB	3.8 KiB	2.2 KiB	—
Raw generic managed process	12.1 KiB	—	—	—
Raw compiled process	2.6 KiB	—	—	—
Two independent idle machines	2.9 KiB	7.3 KiB	3.9 KiB	—
Idle parent with one child	4.8 KiB	5.5 KiB	4.0 KiB	—
Parent with observed child registry	9.0 KiB	—	—	—
Parent with observed invoked child snapshots	5.0 KiB	—	—	—
Suspended Effect fiber	—	—	—	0.6 KiB
Effect queue with waiting fiber	—	—	—	2.8 KiB
Effect mailbox actor shell	—	—	—	3.2 KiB
Two Effect actor shells	—	—	—	6.8 KiB

Effect Machine change from base

Metric	Base	Base variability	PR	PR variability	Difference
Plan counter transitions	130,677 transitions/s	1.2% MAD	122,484 transitions/s	3.7% MAD	-6.3%
Drain burst with terminal fence	519,508 increments/s	3.3% MAD	485,250 increments/s	1.5% MAD	-6.6%
Drain burst with a change observer	478,320 increments/s	2.1% MAD	456,413 increments/s	3.0% MAD	-4.6%
Lookup and send to one child	405,591 increments/s	1.3% MAD	389,511 increments/s	2.7% MAD	-4.0%
Start and stop a machine	168,890 machines/s	0.7% MAD	172,980 machines/s	1.2% MAD	+2.4%
Start and stop a parent with one child	32,941 families/s	0.3% MAD	37,147 families/s	5.2% MAD	+12.8%
Plan transitions through a compound state	117,840 transitions/s	0.8% MAD	117,913 transitions/s	1.7% MAD	+0.1%
Plan transitions through parallel regions	90,905 transitions/s	1.0% MAD	89,556 transitions/s	1.2% MAD	-1.5%
Drain burst through a compound state	523,970 events/s	1.5% MAD	522,138 events/s	1.1% MAD	-0.3%
Drain burst through two parallel regions	474,438 events/s	1.0% MAD	477,852 events/s	0.3% MAD	+0.7%
Drain a compound-state burst with a change observer	475,967 events/s	2.2% MAD	476,620 events/s	1.0% MAD	+0.1%
Idle machine heap per unit	1.5 KiB	0.1% MAD	1.5 KiB	0.1% MAD	-0.0%
Raw generic managed process heap per unit	12.1 KiB	0.1% MAD	12.1 KiB	0.0% MAD	-0.1%
Raw compiled process heap per unit	2.6 KiB	0.0% MAD	2.6 KiB	0.0% MAD	+0.0%
Two independent idle machines heap per unit	2.9 KiB	0.0% MAD	2.9 KiB	0.0% MAD	+0.0%
Idle parent with one child heap per unit	4.8 KiB	0.0% MAD	4.8 KiB	0.0% MAD	-0.0%
Parent with observed child registry heap per unit	9.0 KiB	0.0% MAD	9.0 KiB	0.0% MAD	+0.0%
Parent with observed invoked child snapshots heap per unit	5.0 KiB	0.0% MAD	5.0 KiB	0.0% MAD	-0.0%

Effect runtime reference change from base

Metric	Base	Base variability	PR	PR variability	Difference
Start and stop a raw generic process	18,546 processes/s	2.2% MAD	18,518 processes/s	0.9% MAD	-0.1%
Start and stop a raw compiled process	68,932 processes/s	1.2% MAD	68,549 processes/s	1.1% MAD	-0.6%
Start and interrupt a suspended fiber	164,690 fibers/s	0.2% MAD	166,085 fibers/s	1.3% MAD	+0.8%
Start and stop a queue worker	115,260 workers/s	0.2% MAD	115,115 workers/s	1.1% MAD	-0.1%
Start and stop an actor shell	105,966 actors/s	0.5% MAD	102,270 actors/s	2.2% MAD	-3.5%
Start and stop two actor shells	66,059 families/s	0.1% MAD	64,855 families/s	2.6% MAD	-1.8%
Update an owner-only mutable snapshot	316,856,781 updates/s	0.0% MAD	316,856,781 updates/s	0.0% MAD	0.0%
Update a synchronized snapshot	1,542,234 updates/s	0.9% MAD	1,512,781 updates/s	2.9% MAD	-1.9%
Create, resolve, and await a terminal latch	1,466,162 latches/s	0.2% MAD	1,465,244 latches/s	1.3% MAD	-0.1%

Versions and interpretation

• Effect Machine: 0.3.0
• XState 5: 5.32.5
• XState 6 alpha: 6.0.0-alpha.31
• Effect runtime primitives: 4.0.0-beta.107

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.