samples: fix the metairq_dispatch sample failed on acrn_ehl_crb

The samples/kernel/metairq_dispatch runs failed on acrn_ehl_crb
due to the potential race condition in accessing the global stat
data. Add a spinlock to prevent this.

Fixes zephyrproject-rtos#40889

Signed-off-by: Enjia Mai <[email protected]>

Author: enjiamai

samples/kernel/metairq_dispatch/src/main.c

@@ -14,6 +14,8 @@ LOG_MODULE_REGISTER(main, LOG_LEVEL_INF);
 /* How many messages can be queued for a single thread */
 #define QUEUE_DEPTH 16
 
+static struct k_spinlock lock;
+
 /* Array of worker threads, and their stacks */
 static struct thread_rec {
 	struct k_thread thread;
@@ -118,6 +120,13 @@ static void record_latencies(struct msg *m, uint32_t latency)
 		return;
 	}
 
+	k_spinlock_key_t key = k_spin_lock(&lock);
+
+	if (stats.num_mirq >= MAX_EVENTS) {
+		k_spin_unlock(&lock, key);
+		return;
+	}
+
 	int t = m->target;
 	int lidx = stats.threads[t].nevt++;
 
@@ -127,6 +136,8 @@ static void record_latencies(struct msg *m, uint32_t latency)
 
 	stats.mirq_latencies[atomic_inc(&stats.num_mirq)] = m->metairq_latency;
 
+	k_spin_unlock(&lock, key);
+
 	/* Once we've logged our final event, print a report.  We use
 	 * a semaphore with an initial count of 1 to ensure that only
 	 * one thread gets to do this.  Also events can be processed