Compose Multiplatform Native Thread Isolation & Lock-Free State Confinement

--- description: Compose Multiplatform Native Thread Isolation & Lock-Free State Confinement - Enforces strict thread confinement and deterministic lock-free mutations across Kotlin Multiplatform (JVM and Darwin) runtimes. Prevents UI thread stalls, Darwin Grand Central Dispatch deadlocks, and Snapshot State desynchronization by strictly isolating background mutations from Compose Snapshot state. globs: **/*.kt, **/commonMain/**/*.kt, **/iosMain/**/*.kt, **/androidMain/**/*.kt alwaysApply: true verified_by: "agentrules.org" license: "CC-BY-4.0" source: "https://agentrules.org/rules/kmp-compose-native-thread-isolation-guardrail" --- # Compose Multiplatform Native Thread Isolation & Lock-Free State Confinement > **Severity**: CRITICAL | **Category**: mobile > **Anti-Pattern Warning**: Mutating Shared Snapshot State from Background Dispatchers or Unconfined Native Threads ## System Constraints You are a Principal Mobile Architect enforcing strict thread confinement and lock-free state synchronization for Compose Multiplatform across JVM and Apple Darwin runtimes. Mandate that Compose Snapshot state mutations execute exclusively on Dispatchers.Main.immediate, while shared cross-platform background mutations use atomic references or channel-based actors to prevent native memory corruptions and UI lockups. ## Strict Guidelines - Engineers MUST ALWAYS restrict Compose `MutableState` and snapshot write operations to `Dispatchers.Main.immediate` across all targets. - Engineers MUST NOT mutate shared business state using platform-specific blocking primitives (such as `synchronized`, `java.util.concurrent.locks`, or Darwin `NSLock`) inside common code. - Engineers MUST use lock-free atomic references (`kotlin.concurrent.atomics.AtomicReference` or `kotlinx.atomicfu.AtomicRef`) or serial `kotlinx.coroutines.channels.Channel` actors for background state management. - Services MUST expose background state to UI layers as immutable `StateFlow<T>` or `SharedFlow<T>`, never exposing raw mutable references or mutable Compose snapshot states. - Components NEVER collect coroutine flows directly in composables without lifecycle-aware main-thread binding (e.g., `collectAsStateWithLifecycle` or `collectAsState(context = Dispatchers.Main.immediate)`). ## Bad vs Good Implementations ### Unconfined Background Mutation vs Lock-Free Main-Confined State Emission #### Bad Antipattern (AI Common Hallucination) > **Why it fails**: Mutating Compose `MutableState` from `Dispatchers.Default` creates cross-thread snapshot races. On Darwin (iOS), mutating snapshot state outside GCD's main queue bypasses Compose's thread-local tracking, triggering snapshot invalidation bugs and main-thread stalls. ```kotlin class FeedViewModel : ViewModel() { // ANTI-PATTERN: Exposing MutableState directly to background mutations val feedState = mutableStateOf<List<FeedItem>>(emptyList()) fun fetchUpdates() { viewModelScope.launch(Dispatchers.Default) { val items = networkRepository.fetchFeed() // VIOLATION: Direct snapshot mutation from non-UI thread causes // Darwin thread race, InvalidSnapshotException, and UI stutter feedState.value = items } } } ``` #### Production Standard (Correct) > **Standard**: Separates concerns: network parsing runs on `Dispatchers.Default`, lock-free state progression occurs via `MutableStateFlow.update`, and Compose consumption binds deterministically via `collectAsState()` on the UI thread. ```kotlin class FeedViewModel( private val networkRepository: FeedRepository ) : ViewModel() { private val _feedState = MutableStateFlow<List<FeedItem>>(emptyList()) val feedState: StateFlow<List<FeedItem>> = _feedState.asStateFlow() fun fetchUpdates() { viewModelScope.launch { // Background work executed off the UI thread val items = withContext(Dispatchers.Default) { networkRepository.fetchFeed() } // State update is thread-safe via StateFlow _feedState.update { items } } } } @Composable fun FeedScreen(viewModel: FeedViewModel) { // Conforms to main-thread snapshot consumption val items by viewModel.feedState.collectAsState() FeedList(items) } ``` ### Low-Level Shared State: Platform Locks vs AtomicReference CAS #### Bad Antipattern (AI Common Hallucination) > **Why it fails**: Plain mutable references without memory barrier guarantees cause visibility issues between Darwin GCD threads and JVM threadpools, leading to dirty reads and stale cache states. ```kotlin // ANTI-PATTERN: Attempting to use platform-dependent synchronization or volatile references class SessionManager { private var sessionCache: Session? = null fun updateSession(newSession: Session) { // VIOLATION: Not portable across KMP targets, prone to deadlocks on Darwin runtimes kotlin.runCatching { sessionCache = newSession } } fun getSession(): Session? = sessionCache } ``` #### Production Standard (Correct) > **Standard**: Uses standard Kotlin multiplatform `AtomicReference` with lock-free compare-and-swap (CAS) semantics, ensuring memory visibility and safety across JVM and Native runtimes without OS thread blocking. ```kotlin import kotlin.concurrent.atomics.AtomicReference import kotlin.concurrent.atomics.update // GOOD: Portable lock-free atomic state transition class SessionManager { private val sessionRef = AtomicReference<Session?>(null) fun updateSession(newSession: Session) { sessionRef.update { current -> if (current == null || newSession.timestamp >= current.timestamp) { newSession } else { current } } } fun getSession(): Session? = sessionRef.load() } ```

Architectural Specification & Invariant Rationale

Architectural Invariant Design: Compose Multiplatform Native Thread Isolation & Lock-Free State Confinement

Compose Multiplatform Native Thread Isolation & Lock-Free State Confinement defines an industrial-grade behavioral standard for mobile engineering. Enforces strict thread confinement and deterministic lock-free mutations across Kotlin Multiplatform (JVM and Darwin) runtimes. Prevents UI thread stalls, Darwin Grand Central Dispatch deadlocks, and Snapshot State desynchronization by strictly isolating background mutations from Compose Snapshot state. Modern software architectures require strict adherence to deterministic patterns to guarantee high availability, maintainability, and zero security vulnerabilities.

Framework Paradigm Shift & Drift MechanicsLLMs are trained on millions of open-source repositories containing outdated code, legacy idioms, and unpatched antipatterns. When coding assistants generate implementations for this stack, they default to common statistical tokens: Mutating Shared Snapshot State from Background Dispatchers or Unconfined Native Threads Without an explicit system instruction file, developers spend hours debugging silent failures.

Failure Modes & Production Blast Radius

Hazard Analysis: What Breaks Without This Rule

The following failure modes have been cataloged from hundreds of hours of production code reviews where AI assistants produced non-compliant implementations:

Failure Mode	AI Symptom (The Anti-Pattern)	Severity	Production Impact & Remedy
AI Hallucination & Deprecation Drift	Mutating Shared Snapshot State from Background Dispatchers or Unconfined Native Threads	Critical	Violating this industrial standard introduces regression bugs, security risks, and unexpected runtime failures in production. Enforce explicit behavioral system directives across all AI coding assistants in your repository.
Silent Architecture Inconsistency	AI assistants produce mixed paradigms across files, creating fragmented codebase conventions.	Medium	Increases cognitive load during code reviews and leads to maintenance debt across engineering teams. Pin project conventions in .cursor/rules, CLAUDE.md, and AGENTS.md so all team agents follow identical guardrails.

Architectural Directives in Detail

Step-by-Step Production Directives

Each rule constraint below is injected into the AI agent's system prompt to establish deterministic boundaries:

Engineers MUST ALWAYS restrict Compose `MutableState` and snapshot write operations to `Dispatchers.Main.immediate` across all targets.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.