What do the block and mutex profiles measure, how do you enable them, and how do they differ?

Both are disabled by default because they add overhead.

• Block profile: where goroutines waited on synchronization primitives — channel send/recv, select, sync.Mutex, sync.Cond, WaitGroup.Wait. Records total delay per stack. Enabled with runtime.SetBlockProfileRate(ns): sample on average one blocking event per ns nanoseconds blocked; 1 = record everything.
• Mutex profile: contention on sync.Mutex/RWMutex, attributed to the stack of the goroutine that held the lock and delayed others (reported at Unlock). Enabled with runtime.SetMutexProfileFraction(n): report ~1/n contention events.

func main() {
    runtime.SetBlockProfileRate(10_000)  // ~ one sample per 10µs blocked
    runtime.SetMutexProfileFraction(100) // 1% of contention events
    startDebugServer()
    // ...
}
// go tool pprof http://127.0.0.1:6060/debug/pprof/mutex
// go tool pprof http://127.0.0.1:6060/debug/pprof/block

Key distinction: block answers "where are my goroutines waiting?" (the victims); mutex answers "which critical section is causing the waiting?" (the culprit). Gotcha: the block profile includes intentional waiting (a worker idling on <-jobs), so filter with -focus. Neither shows time blocked in syscalls/network I/O — for that use the execution trace. In benchmarks: go test -bench=. -blockprofile=b.out -mutexprofile=m.out.