kernel/hung_task.c: allow to set checking interval separately from timeout

Currently task hung checking interval is equal to timeout, as the result
hung is detected anywhere between timeout and 2*timeout.  This is fine for
most interactive environments, but this hurts automated testing setups
(syzbot).  In an automated setup we need to strictly order CPU lockup <
RCU stall < workqueue lockup < task hung < silent loss, so that RCU stall
is not detected as task hung and task hung is not detected as silent
machine loss.  The large variance in task hung detection timeout requires
setting silent machine loss timeout to a very large value (e.g.  if task
hung is 3 mins, then silent loss need to be set to ~7 mins).  The
additional 3 minutes significantly reduce testing efficiency because
usually we crash kernel within a minute, and this can add hours to bug
localization process as it needs to do dozens of tests.

Allow setting checking interval separately from timeout.  This allows to
set timeout to, say, 3 minutes, but checking interval to 10 secs.

The interval is controlled via a new hung_task_check_interval_secs sysctl,
similar to the existing hung_task_timeout_secs sysctl.  The default value
of 0 results in the current behavior: checking interval is equal to
timeout.

[[email protected]: update hung_task_timeout_max's comment]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Dmitry Vyukov <[email protected]>
Cc: Paul E. McKenney <[email protected]>
Cc: Tetsuo Handa <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Ingo Molnar <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

Author: dvyukov

Documentation/sysctl/kernel.txt

@@ -38,6 +38,7 @@ show up in /proc/sys/kernel:
 - hung_task_panic
 - hung_task_check_count
 - hung_task_timeout_secs
+- hung_task_check_interval_secs
 - hung_task_warnings
 - hyperv_record_panic_msg
 - kexec_load_disabled
@@ -355,7 +356,7 @@ This file shows up if CONFIG_DETECT_HUNG_TASK is enabled.
 
 hung_task_timeout_secs:
 
-Check interval. When a task in D state did not get scheduled
+When a task in D state did not get scheduled
 for more than this value report a warning.
 This file shows up if CONFIG_DETECT_HUNG_TASK is enabled.
 
@@ -364,6 +365,18 @@ Possible values to set are in range {0..LONG_MAX/HZ}.
 
 ==============================================================
 
+hung_task_check_interval_secs:
+
+Hung task check interval. If hung task checking is enabled
+(see hung_task_timeout_secs), the check is done every
+hung_task_check_interval_secs seconds.
+This file shows up if CONFIG_DETECT_HUNG_TASK is enabled.
+
+0 (default): means use hung_task_timeout_secs as checking interval.
+Possible values to set are in range {0..LONG_MAX/HZ}.
+
+==============================================================
+
 hung_task_warnings:
 
 The maximum number of warnings to report. During a check interval

include/linux/sched.h

@@ -853,6 +853,7 @@ struct task_struct {
 #endif
 #ifdef CONFIG_DETECT_HUNG_TASK
 	unsigned long			last_switch_count;
+	unsigned long			last_switch_time;
 #endif
 	/* Filesystem information: */
 	struct fs_struct		*fs;

include/linux/sched/sysctl.h

@@ -10,6 +10,7 @@ struct ctl_table;
 extern int	     sysctl_hung_task_check_count;
 extern unsigned int  sysctl_hung_task_panic;
 extern unsigned long sysctl_hung_task_timeout_secs;
+extern unsigned long sysctl_hung_task_check_interval_secs;
 extern int sysctl_hung_task_warnings;
 extern int proc_dohung_task_timeout_secs(struct ctl_table *table, int write,
 					 void __user *buffer,

kernel/fork.c

@@ -1302,6 +1302,7 @@ static int copy_mm(unsigned long clone_flags, struct task_struct *tsk)
 	tsk->nvcsw = tsk->nivcsw = 0;
 #ifdef CONFIG_DETECT_HUNG_TASK
 	tsk->last_switch_count = tsk->nvcsw + tsk->nivcsw;
+	tsk->last_switch_time = 0;
 #endif
 
 	tsk->mm = NULL;

kernel/hung_task.c

@@ -40,6 +40,11 @@ int __read_mostly sysctl_hung_task_check_count = PID_MAX_LIMIT;
  */
 unsigned long __read_mostly sysctl_hung_task_timeout_secs = CONFIG_DEFAULT_HUNG_TASK_TIMEOUT;
 
+/*
+ * Zero (default value) means use sysctl_hung_task_timeout_secs:
+ */
+unsigned long __read_mostly sysctl_hung_task_check_interval_secs;
+
 int __read_mostly sysctl_hung_task_warnings = 10;
 
 static int __read_mostly did_panic;
@@ -98,8 +103,11 @@ static void check_hung_task(struct task_struct *t, unsigned long timeout)
 
 	if (switch_count != t->last_switch_count) {
 		t->last_switch_count = switch_count;
+		t->last_switch_time = jiffies;
 		return;
 	}
+	if (time_is_after_jiffies(t->last_switch_time + timeout * HZ))
+		return;
 
 	trace_sched_process_hang(t);
 
@@ -245,8 +253,13 @@ static int watchdog(void *dummy)
 
 	for ( ; ; ) {
 		unsigned long timeout = sysctl_hung_task_timeout_secs;
-		long t = hung_timeout_jiffies(hung_last_checked, timeout);
+		unsigned long interval = sysctl_hung_task_check_interval_secs;
+		long t;
 
+		if (interval == 0)
+			interval = timeout;
+		interval = min_t(unsigned long, interval, timeout);
+		t = hung_timeout_jiffies(hung_last_checked, interval);
 		if (t <= 0) {
 			if (!atomic_xchg(&reset_hung_task, 0))
 				check_hung_uninterruptible_tasks(timeout);

kernel/sysctl.c

@@ -145,7 +145,10 @@ static int minolduid;
 static int ngroups_max = NGROUPS_MAX;
 static const int cap_last_cap = CAP_LAST_CAP;
 
-/*this is needed for proc_doulongvec_minmax of sysctl_hung_task_timeout_secs */
+/*
+ * This is needed for proc_doulongvec_minmax of sysctl_hung_task_timeout_secs
+ * and hung_task_check_interval_secs
+ */
 #ifdef CONFIG_DETECT_HUNG_TASK
 static unsigned long hung_task_timeout_max = (LONG_MAX/HZ);
 #endif
@@ -1090,6 +1093,14 @@ static struct ctl_table kern_table[] = {
 		.proc_handler	= proc_dohung_task_timeout_secs,
 		.extra2		= &hung_task_timeout_max,
 	},
+	{
+		.procname	= "hung_task_check_interval_secs",
+		.data		= &sysctl_hung_task_check_interval_secs,
+		.maxlen		= sizeof(unsigned long),
+		.mode		= 0644,
+		.proc_handler	= proc_dohung_task_timeout_secs,
+		.extra2		= &hung_task_timeout_max,
+	},
 	{
 		.procname	= "hung_task_warnings",
 		.data		= &sysctl_hung_task_warnings,