Merge: tracing/osnoise: Do not cancel timer if no kthread and add cpumask to identify user threads

MR: https://gitlab.com/redhat/centos-stream/src/kernel/centos-stream-9/-/merge_requests/5167

JIRA: https://issues.redhat.com/browse/RHEL-39968

rtla crashes the timerlat tracer if killed with SIGTERM and subsequently re-run:
    while true; do rtla timerlat top -u -q & PID=$!; sleep 5; kill -INT $PID; sleep 0.001; kill -TERM $PID; wait $PID; done
Fix two separate causes of the issue:
- Check for an active kthread before doing hrtimer_cancel() in both timerlat_fd_release() and tlat_var_reset(). Add necessary locking to prevent race between the kthread read and the timer cancel.
- Identify CPUs on which a user workload is running with a cpumask and prevent calling of kthread_stop() on the thread such CPUs.

Signed-off-by: Tomas Glozar <[email protected]>

Approved-by: Luis Claudio R. Goncalves <[email protected]>
Approved-by: Andrew Halaney <[email protected]>
Approved-by: CKI KWF Bot <[email protected]>

Merged-by: Rado Vrbovsky <[email protected]>

Author: emmil

kernel/trace/trace_osnoise.c

@@ -228,6 +228,11 @@ static inline struct osnoise_variables *this_cpu_osn_var(void)
 	return this_cpu_ptr(&per_cpu_osnoise_var);
 }
 
+/*
+ * Protect the interface.
+ */
+static struct mutex interface_lock;
+
 #ifdef CONFIG_TIMERLAT_TRACER
 /*
  * Runtime information for the timer mode.
@@ -259,14 +264,20 @@ static inline void tlat_var_reset(void)
 {
 	struct timerlat_variables *tlat_var;
 	int cpu;
+
+	/* Synchronize with the timerlat interfaces */
+	mutex_lock(&interface_lock);
 	/*
 	 * So far, all the values are initialized as 0, so
 	 * zeroing the structure is perfect.
 	 */
 	for_each_cpu(cpu, cpu_online_mask) {
 		tlat_var = per_cpu_ptr(&per_cpu_timerlat_var, cpu);
+		if (tlat_var->kthread)
+			hrtimer_cancel(&tlat_var->timer);
 		memset(tlat_var, 0, sizeof(*tlat_var));
 	}
+	mutex_unlock(&interface_lock);
 }
 #else /* CONFIG_TIMERLAT_TRACER */
 #define tlat_var_reset()	do {} while (0)
@@ -331,11 +342,6 @@ struct timerlat_sample {
 };
 #endif
 
-/*
- * Protect the interface.
- */
-static struct mutex interface_lock;
-
 /*
  * Tracer data.
  */
@@ -1612,6 +1618,7 @@ static int run_osnoise(void)
 
 static struct cpumask osnoise_cpumask;
 static struct cpumask save_cpumask;
+static struct cpumask kthread_cpumask;
 
 /*
  * osnoise_sleep - sleep until the next period
@@ -1675,6 +1682,7 @@ static inline int osnoise_migration_pending(void)
 	 */
 	mutex_lock(&interface_lock);
 	this_cpu_osn_var()->kthread = NULL;
+	cpumask_clear_cpu(smp_processor_id(), &kthread_cpumask);
 	mutex_unlock(&interface_lock);
 
 	return 1;
@@ -1945,11 +1953,16 @@ static void stop_kthread(unsigned int cpu)
 {
 	struct task_struct *kthread;
 
+	mutex_lock(&interface_lock);
 	kthread = per_cpu(per_cpu_osnoise_var, cpu).kthread;
 	if (kthread) {
-		if (test_bit(OSN_WORKLOAD, &osnoise_options)) {
+		per_cpu(per_cpu_osnoise_var, cpu).kthread = NULL;
+		mutex_unlock(&interface_lock);
+
+		if (cpumask_test_and_clear_cpu(cpu, &kthread_cpumask) &&
+		    !WARN_ON(!test_bit(OSN_WORKLOAD, &osnoise_options))) {
 			kthread_stop(kthread);
-		} else {
+		} else if (!WARN_ON(test_bit(OSN_WORKLOAD, &osnoise_options))) {
 			/*
 			 * This is a user thread waiting on the timerlat_fd. We need
 			 * to close all users, and the best way to guarantee this is
@@ -1958,16 +1971,15 @@ static void stop_kthread(unsigned int cpu)
 			kill_pid(kthread->thread_pid, SIGKILL, 1);
 			put_task_struct(kthread);
 		}
-		per_cpu(per_cpu_osnoise_var, cpu).kthread = NULL;
 	} else {
+		mutex_unlock(&interface_lock);
 		/* if no workload, just return */
 		if (!test_bit(OSN_WORKLOAD, &osnoise_options)) {
 			/*
 			 * This is set in the osnoise tracer case.
 			 */
 			per_cpu(per_cpu_osnoise_var, cpu).sampling = false;
 			barrier();
-			return;
 		}
 	}
 }
@@ -1982,12 +1994,8 @@ static void stop_per_cpu_kthreads(void)
 {
 	int cpu;
 
-	cpus_read_lock();
-
-	for_each_online_cpu(cpu)
+	for_each_possible_cpu(cpu)
 		stop_kthread(cpu);
-
-	cpus_read_unlock();
 }
 
 /*
@@ -2021,6 +2029,7 @@ static int start_kthread(unsigned int cpu)
 	}
 
 	per_cpu(per_cpu_osnoise_var, cpu).kthread = kthread;
+	cpumask_set_cpu(cpu, &kthread_cpumask);
 
 	return 0;
 }
@@ -2048,8 +2057,16 @@ static int start_per_cpu_kthreads(void)
 	 */
 	cpumask_and(current_mask, cpu_online_mask, &osnoise_cpumask);
 
-	for_each_possible_cpu(cpu)
+	for_each_possible_cpu(cpu) {
+		if (cpumask_test_and_clear_cpu(cpu, &kthread_cpumask)) {
+			struct task_struct *kthread;
+
+			kthread = per_cpu(per_cpu_osnoise_var, cpu).kthread;
+			if (!WARN_ON(!kthread))
+				kthread_stop(kthread);
+		}
 		per_cpu(per_cpu_osnoise_var, cpu).kthread = NULL;
+	}
 
 	for_each_cpu(cpu, current_mask) {
 		retval = start_kthread(cpu);
@@ -2579,7 +2596,8 @@ static int timerlat_fd_release(struct inode *inode, struct file *file)
 	osn_var = per_cpu_ptr(&per_cpu_osnoise_var, cpu);
 	tlat_var = per_cpu_ptr(&per_cpu_timerlat_var, cpu);
 
-	hrtimer_cancel(&tlat_var->timer);
+	if (tlat_var->kthread)
+		hrtimer_cancel(&tlat_var->timer);
 	memset(tlat_var, 0, sizeof(*tlat_var));
 
 	osn_var->sampling = 0;