Browse Source

hung_task: add method to reset detector

In certain occasions it is possible for a hung task detector
positive to be false: continuation from a paused VM, for example.

Add a method to reset detection, similar as is done
with other kernel watchdogs.

Acked-by: Don Zickus <dzickus@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Marcelo Tosatti 11 years ago
parent
commit
8b414521bc
3 changed files with 20 additions and 0 deletions
  1. 1 0
      arch/x86/kernel/pvclock.c
  2. 8 0
      include/linux/sched.h
  3. 11 0
      kernel/hung_task.c

+ 1 - 0
arch/x86/kernel/pvclock.c

@@ -48,6 +48,7 @@ void pvclock_touch_watchdogs(void)
 	touch_softlockup_watchdog_sync();
 	clocksource_touch_watchdog();
 	rcu_cpu_stall_reset();
+	reset_hung_task_detector();
 }
 
 static atomic64_t last_value = ATOMIC64_INIT(0);

+ 8 - 0
include/linux/sched.h

@@ -285,6 +285,14 @@ static inline void lockup_detector_init(void)
 }
 #endif
 
+#ifdef CONFIG_DETECT_HUNG_TASK
+void reset_hung_task_detector(void);
+#else
+static inline void reset_hung_task_detector(void)
+{
+}
+#endif
+
 /* Attach to any functions which should be ignored in wchan output. */
 #define __sched		__attribute__((__section__(".sched.text")))
 

+ 11 - 0
kernel/hung_task.c

@@ -203,6 +203,14 @@ int proc_dohung_task_timeout_secs(struct ctl_table *table, int write,
 	return ret;
 }
 
+static atomic_t reset_hung_task = ATOMIC_INIT(0);
+
+void reset_hung_task_detector(void)
+{
+	atomic_set(&reset_hung_task, 1);
+}
+EXPORT_SYMBOL_GPL(reset_hung_task_detector);
+
 /*
  * kthread which checks for tasks stuck in D state
  */
@@ -216,6 +224,9 @@ static int watchdog(void *dummy)
 		while (schedule_timeout_interruptible(timeout_jiffies(timeout)))
 			timeout = sysctl_hung_task_timeout_secs;
 
+		if (atomic_xchg(&reset_hung_task, 0))
+			continue;
+
 		check_hung_uninterruptible_tasks(timeout);
 	}