The async and concurrency cheat sheet

unawaited_futures only looks inside async bodies, discarded_futures only inside sync bodies, and both miss arrow functions, tear-offs, and collection literals. Measured case by case on Dart 3.12.2 and Flutter 3.44.8, with the analysis_options.yaml I recommend.

Use await Task.Delay for any timed wait in async code, Thread.Sleep only on a dedicated thread you own, and SpinWait only for sub-microsecond waits inside lock-free code. To wait for a condition, use a signal (ManualResetEventSlim, SemaphoreSlim, TaskCompletionSource) instead of any of the three. Measured on .NET 10.0.10.

Open the log with FileShare.ReadWrite | FileShare.Delete, seek to the end, poll for new bytes, decode them with your own UTF-8 Decoder instead of StreamReader, and handle truncation and rotation explicitly. A tested tail -F implementation for .NET 10 and .NET 11.

Check IsCompleted first and read the result through GetAwaiter().GetResult(), which costs 0 bytes. Only fall back to blocking when the ValueTask is still pending, and do that with a cached event and UnsafeOnCompleted instead of AsTask(), which allocates 72 to 208 bytes per call. Measured on .NET 10 and .NET 11 RC1.

Volatile.Read is an acquire load of one location. Volatile.ReadBarrier, new in .NET 10, is a fence that gives every earlier read acquire semantics. Use Volatile.Read for flags and published references, and ReadBarrier when you need a batch of plain or non-atomic reads to complete before the next memory access, as in a seqlock.

Loading an entity, calling List.Add, and saving rewrites the whole jsonb document and silently loses concurrent appends. Push the append into one UPDATE with the jsonb || operator, either through ExecuteSqlAsync or a mapped function inside ExecuteUpdateAsync, and make it idempotent with a @> guard.

The existence check in your handler is not the guard you think it is. Put a unique index on the inbox table, write the marker in the same SaveChanges as the business change, and let the database pick the winner. Plus the ExecuteUpdate trap that quietly undoes all of it.

Picking 2 or more photos on iOS throws UIKitThreadAccessException in MAUI 10.0.100 and 10.0.101. MAUI reads PHPickerResult.ItemProvider after an await, off the main thread. Pin 10.0.90, upgrade to 10.0.110, or use a PHPicker shim.

Use ZipFile.CreateFromDirectoryAsync, ExtractToDirectoryAsync, ZipArchive.CreateAsync and ZipArchiveEntry.OpenAsync without hidden synchronous I/O. Measured on .NET 11 RC 1 and .NET 10.0.12, including the .NET 10 bug that still blocks Kestrel.

Play Console tells you your ANR rate is over 0.47% and hands you a native stack full of libcoreclr.so frames. Here is how to get from that useless trace to the exact async void event handler that blocked the main thread, using a Looper printer, a SynchronizationContext wrapper that names the state machine, and dotnet-trace over dsrouter.

EF Core 11 still has no lock API. Here is how to take a real row lock with FromSql: WITH (UPDLOCK, ROWLOCK) on SQL Server, FOR UPDATE on PostgreSQL, the subquery trap that silently widens the lock, NOWAIT and SKIP LOCKED, deadlock retries, and what to do when the row does not exist yet.

Eliding async/await in a repository passthrough saves about 6 ns and 72 bytes, and costs you a stack frame, try/catch semantics, and safe disposal. Keep return await unless the method is a pure passthrough on a measured hot path.

The check-then-insert in your handler is not atomic. Catch PostgresException with SqlState 23505, or collapse the whole thing into one INSERT ... ON CONFLICT statement. EnableRetryOnFailure will not help.

A thread shares memory with every other thread in the process. A Dart isolate does not: it owns its heap, runs one event loop, and talks to other isolates only by messages. Here is what that means at the VM level, where isolate groups blur the line, and how it plays out in Flutter, FFI, and on the web.

await Task.WhenAll rethrows only one of the failures. Keep the WhenAll task in a variable and read its Exception.InnerExceptions to see every error instead of one.

CS1998 means an async method has no await, so it runs synchronously. Drop the async modifier and return Task.FromResult, or add the await you forgot.

Add a rowversion concurrency token in EF Core 11: the [Timestamp] and IsRowVersion setup, the SQL EF actually emits, catching DbUpdateConcurrencyException, store-wins vs client-wins vs merge, disconnected APIs with ETags, and the five traps that silently disable the whole thing.

Replace DateTime.UtcNow, Stopwatch, and Task.Delay with System.TimeProvider so tests can control the clock: DI registration, FakeTimeProvider.Advance and SetUtcNow, testing timeouts and PeriodicTimer-based BackgroundServices, plus the Advance-continuation and xUnit v2 gotchas.

A staged playbook for removing sync-over-async from an existing .NET codebase: inventory with analyzers, measure ThreadPool starvation, convert one call chain at a time, and ratchet the count to zero on .NET 11.

await is the right answer almost every time. When you truly must block, GetAwaiter().GetResult() beats .Result and .Wait() because it throws the original exception. A decision matrix for .NET 11 and C# 14.

CS4014 means you called a Task-returning method without awaiting it. Add await, or discard with _ = if fire-and-forget is truly intended, and handle exceptions.

Blocking on an async Task with .Result or .Wait() deadlocks when a SynchronizationContext is present. Here is why it hangs and how to fix it in .NET 11 and C# 14.

Riverpod 3.0 retries a failed provider up to 10 times by default. Pass a retry function that returns null on ProviderScope, ProviderContainer, or an individual provider to turn it off or bound it.

A stream keeps emitting after the user leaves the screen, its onData calls setState on a disposed State, and Flutter throws. Store the subscription, cancel it in dispose before super.dispose, and the callback can never fire on a dead widget. The full pattern for Flutter 3.44.

After an await, the widget may already be disposed, and calling setState throws. Guard the resume with if (!mounted) return; and, better, cancel the work that triggers it. The full pattern for Flutter 3.44.

As the .NET 11 previews progress toward the November release, Runtime Async has graduated: net11.0 projects opt in with a single MSBuild property, and the runtime libraries themselves now ship compiled on it.

A complete guide to IAsyncDisposable in C#: when to use await using, how to write DisposeAsync and DisposeAsyncCore correctly, and the stacking and ConfigureAwait gotchas that leak resources.

Thread a CancellationToken cleanly through every layer of an async call chain in .NET 11: last-parameter convention, default values, linked tokens, ASP.NET Core RequestAborted, and the CA2016 analyzer that catches the ones you drop.

Use CancellationTokenSource.CancelAfter to enforce an async deadline in .NET 11: constructor vs CancelAfter, linked tokens for composing timeouts with caller tokens, exception disambiguation, Task.WaitAsync, TryReset for pooling, and testable timeouts with TimeProvider.

Capture what you need from the context before the await, then guard the resume with if (context.mounted) return. Here is the full pattern, the lint that enforces it, and the edge cases it misses.

ValueTask and ValueTask<T> are structs that let an async method return a result without heap-allocating a Task when it completes synchronously. The win is one fewer allocation on hot paths that usually finish without awaiting. The cost is a strict await-once contract. Here is what the type actually is, how it works, and the narrow set of cases where it earns its keep.

IAsyncEnumerable<T> is the interface for asynchronous streams: a sequence whose elements arrive over time and each one may require an await. Here is what it actually is, how await foreach and yield drive it, and the rule for when to reach for it over Task<List<T>>.

A practical checklist for reverting ValueTask and ValueTask<T> return types to Task and Task<T>, what breaks at the call sites, how to verify each change, and how to know whether the swap was ever worth it.

Your fire-and-forget task captured a request-scoped DbContext that the DI scope already disposed. Resolve a fresh context inside the task with IServiceScopeFactory or IDbContextFactory.

Calling Task.Run from a controller loses work on shutdown, swallows exceptions, and captures disposed scoped services. The safe pattern is a bounded Channel queue drained by a BackgroundService that opens a fresh scope per work item and drains in-flight work on StopAsync.

Use Parallel.ForEach for CPU-bound work over in-memory data, Parallel.ForEachAsync for async I/O over many items with a concurrency cap, and Task.WhenAll for a small fixed fan-out where you want every operation in flight and need the results back.

Four ways to guard a critical section in C#, and a decision matrix for picking one. Use System.Threading.Lock for synchronous mutual exclusion on .NET 9+, SemaphoreSlim when the section spans an await, and Monitor only when you need Wait/Pulse.

Three ways to push work onto the thread pool in C#, and which one to reach for. Use Task.Run for almost everything, ThreadPool.QueueUserWorkItem<TState> for allocation-free fire-and-forget, and Task.Factory.StartNew only for LongRunning or a custom scheduler.

ConfigureAwait(false) is still mandatory in library code that may run under a SynchronizationContext (WinForms, WPF, MAUI). In application code on ASP.NET Core, a console app, or a worker service running on .NET 11, it is a no-op.

Three sequence interfaces, three execution models. Use IQueryable when a database can translate the query, IAsyncEnumerable when the producer is async and you want to stream, IEnumerable for everything else in memory.

async Task is the default and async void is the exception. Use async void only for event handlers, top-level message-loop handlers, and a handful of framework callbacks that demand a void signature. Everywhere else, async Task wins on exceptions, composition, and testability.

EF Core throws when two awaits run in parallel on the same DbContext. Await each call sequentially, or get a new DbContext per concurrent unit of work via IDbContextFactory.

When async/await is not enough: spawn a Dart isolate to run CPU-bound work off the UI thread. Isolate.run, Flutter's compute, long-lived workers with SendPort/ReceivePort, what can cross the boundary, and the JS/web caveat. Tested on Dart 3.11 and Flutter 3.27.1.

System.Threading.Lock arrived in .NET 9 and is the default synchronization primitive on .NET 11 and C# 14. This guide shows how to migrate from lock(object), how EnterScope works, and the gotchas around await, dynamic, and downlevel targets.

System.Threading.Channels is the async-first replacement for BlockingCollection in .NET 11. This guide shows how to migrate, how to choose bounded vs unbounded, and how to handle backpressure, cancellation, and graceful shutdown without deadlocking.

Cooperative cancellation with CancellationToken, CancelAsync, Task.WaitAsync, and linked tokens in .NET 11. Plus the blocking patterns that turn a clean cancel into a deadlock.

Runtime Async in .NET 11 moves async/await handling from compiler-generated state machines into the runtime itself, producing readable stack traces, correct breakpoints, and fewer heap allocations.