softlockup: decouple hung tasks check from softlockup detection

Decoupling allows:

* hung tasks check to happen at very low priority

* hung tasks check and softlockup to be enabled/disabled independently
  at compile and/or run-time

* individual panic settings to be enabled disabled independently
  at compile and/or run-time

* softlockup threshold to be reduced without increasing hung tasks
  poll frequency (hung task check is expensive relative to softlock watchdog)

* hung task check to be zero over-head when disabled at run-time

Signed-off-by: Mandeep Singh Baines <msb@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

6 files changed, 261 insertions, 105 deletions

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 54cbabf3b871..f2f94d532302 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -297,9 +297,6 @@ extern int proc_dosoftlockup_thresh(struct ctl_table *table, int write,
                                     struct file *filp, void __user *buffer,
                                     size_t *lenp, loff_t *ppos);
 extern unsigned int  softlockup_panic;
-extern unsigned long sysctl_hung_task_check_count;
-extern unsigned long sysctl_hung_task_timeout_secs;
-extern unsigned long sysctl_hung_task_warnings;
 extern int softlockup_thresh;
 #else
 static inline void softlockup_tick(void)
@@ -316,6 +313,15 @@ static inline void touch_all_softlockup_watchdogs(void)
 }
 #endif
 
+#ifdef CONFIG_DETECT_HUNG_TASK
+extern unsigned int  sysctl_hung_task_panic;
+extern unsigned long sysctl_hung_task_check_count;
+extern unsigned long sysctl_hung_task_timeout_secs;
+extern unsigned long sysctl_hung_task_warnings;
+extern int proc_dohung_task_timeout_secs(struct ctl_table *table, int write,
+                                         struct file *filp, void __user *buffer,
+                                         size_t *lenp, loff_t *ppos);
+#endif
 
 /* Attach to any functions which should be ignored in wchan output. */
 #define __sched         __attribute__((__section__(".sched.text")))
@@ -1236,7 +1242,7 @@ struct task_struct {
 /* ipc stuff */
         struct sysv_sem sysvsem;
 #endif
-#ifdef CONFIG_DETECT_SOFTLOCKUP
+#ifdef CONFIG_DETECT_HUNG_TASK
 /* hung task detection */
         unsigned long last_switch_timestamp;
         unsigned long last_switch_count;
diff --git a/kernel/Makefile b/kernel/Makefile
index 2aebc4cd7878..979745f1b4bc 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -74,6 +74,7 @@ obj-$(CONFIG_AUDIT_TREE) += audit_tree.o
 obj-$(CONFIG_KPROBES) += kprobes.o
 obj-$(CONFIG_KGDB) += kgdb.o
 obj-$(CONFIG_DETECT_SOFTLOCKUP) += softlockup.o
+obj-$(CONFIG_DETECT_HUNG_TASK) += hung_task.o
 obj-$(CONFIG_GENERIC_HARDIRQS) += irq/
 obj-$(CONFIG_SECCOMP) += seccomp.o
 obj-$(CONFIG_RCU_TORTURE_TEST) += rcutorture.o
diff --git a/kernel/hung_task.c b/kernel/hung_task.c
new file mode 100644
index 000000000000..ba5a77cad3bb
--- /dev/null
+++ b/kernel/hung_task.c
@@ -0,0 +1,198 @@
+/*
+ * Detect Hung Task
+ *
+ * kernel/hung_task.c - kernel thread for detecting tasks stuck in D state
+ *
+ */
+
+#include <linux/mm.h>
+#include <linux/cpu.h>
+#include <linux/nmi.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/freezer.h>
+#include <linux/kthread.h>
+#include <linux/lockdep.h>
+#include <linux/module.h>
+#include <linux/sysctl.h>
+
+/*
+ * Have a reasonable limit on the number of tasks checked:
+ */
+unsigned long __read_mostly sysctl_hung_task_check_count = 1024;
+
+/*
+ * Zero means infinite timeout - no checking done:
+ */
+unsigned long __read_mostly sysctl_hung_task_timeout_secs = 120;
+static unsigned long __read_mostly hung_task_poll_jiffies;
+
+unsigned long __read_mostly sysctl_hung_task_warnings = 10;
+
+static int __read_mostly did_panic;
+
+static struct task_struct *watchdog_task;
+
+/*
+ * Should we panic (and reboot, if panic_timeout= is set) when a
+ * hung task is detected:
+ */
+unsigned int __read_mostly sysctl_hung_task_panic =
+                                CONFIG_BOOTPARAM_HUNG_TASK_PANIC_VALUE;
+
+static int __init hung_task_panic_setup(char *str)
+{
+        sysctl_hung_task_panic = simple_strtoul(str, NULL, 0);
+
+        return 1;
+}
+__setup("hung_task_panic=", hung_task_panic_setup);
+
+static int
+hung_task_panic(struct notifier_block *this, unsigned long event, void *ptr)
+{
+        did_panic = 1;
+
+        return NOTIFY_DONE;
+}
+
+static struct notifier_block panic_block = {
+        .notifier_call = hung_task_panic,
+};
+
+/*
+ * Returns seconds, approximately.  We don't need nanosecond
+ * resolution, and we don't need to waste time with a big divide when
+ * 2^30ns == 1.074s.
+ */
+static unsigned long get_timestamp(void)
+{
+        int this_cpu = raw_smp_processor_id();
+
+        return cpu_clock(this_cpu) >> 30LL;  /* 2^30 ~= 10^9 */
+}
+
+static void check_hung_task(struct task_struct *t, unsigned long now)
+{
+        unsigned long switch_count = t->nvcsw + t->nivcsw;
+
+        if (t->flags & PF_FROZEN)
+                return;
+
+        if (switch_count != t->last_switch_count || !t->last_switch_timestamp) {
+                t->last_switch_count = switch_count;
+                t->last_switch_timestamp = now;
+                return;
+        }
+        if ((long)(now - t->last_switch_timestamp) <
+                                        sysctl_hung_task_timeout_secs)
+                return;
+        if (!sysctl_hung_task_warnings)
+                return;
+        sysctl_hung_task_warnings--;
+
+        /*
+         * Ok, the task did not get scheduled for more than 2 minutes,
+         * complain:
+         */
+        printk(KERN_ERR "INFO: task %s:%d blocked for more than "
+                        "%ld seconds.\n", t->comm, t->pid,
+                        sysctl_hung_task_timeout_secs);
+        printk(KERN_ERR "\"echo 0 > /proc/sys/kernel/hung_task_timeout_secs\""
+                        " disables this message.\n");
+        sched_show_task(t);
+        __debug_show_held_locks(t);
+
+        t->last_switch_timestamp = now;
+        touch_nmi_watchdog();
+
+        if (sysctl_hung_task_panic)
+                panic("hung_task: blocked tasks");
+}
+
+/*
+ * Check whether a TASK_UNINTERRUPTIBLE does not get woken up for
+ * a really long time (120 seconds). If that happens, print out
+ * a warning.
+ */
+static void check_hung_uninterruptible_tasks(void)
+{
+        int max_count = sysctl_hung_task_check_count;
+        unsigned long now = get_timestamp();
+        struct task_struct *g, *t;
+
+        /*
+         * If the system crashed already then all bets are off,
+         * do not report extra hung tasks:
+         */
+        if (test_taint(TAINT_DIE) || did_panic)
+                return;
+
+        read_lock(&tasklist_lock);
+        do_each_thread(g, t) {
+                if (!--max_count)
+                        goto unlock;
+                /* use "==" to skip the TASK_KILLABLE tasks waiting on NFS */
+                if (t->state == TASK_UNINTERRUPTIBLE)
+                        check_hung_task(t, now);
+        } while_each_thread(g, t);
+ unlock:
+        read_unlock(&tasklist_lock);
+}
+
+static void update_poll_jiffies(void)
+{
+        /* timeout of 0 will disable the watchdog */
+        if (sysctl_hung_task_timeout_secs == 0)
+                hung_task_poll_jiffies = MAX_SCHEDULE_TIMEOUT;
+        else
+                hung_task_poll_jiffies = sysctl_hung_task_timeout_secs * HZ / 2;
+}
+
+/*
+ * Process updating of timeout sysctl
+ */
+int proc_dohung_task_timeout_secs(struct ctl_table *table, int write,
+                                  struct file *filp, void __user *buffer,
+                                  size_t *lenp, loff_t *ppos)
+{
+        int ret;
+
+        ret = proc_doulongvec_minmax(table, write, filp, buffer, lenp, ppos);
+
+        if (ret || !write)
+                goto out;
+
+        update_poll_jiffies();
+
+        wake_up_process(watchdog_task);
+
+ out:
+        return ret;
+}
+
+/*
+ * kthread which checks for tasks stuck in D state
+ */
+static int watchdog(void *dummy)
+{
+        set_user_nice(current, 0);
+        update_poll_jiffies();
+
+        for ( ; ; ) {
+                while (schedule_timeout_interruptible(hung_task_poll_jiffies));
+                check_hung_uninterruptible_tasks();
+        }
+
+        return 0;
+}
+
+static int __init hung_task_init(void)
+{
+        atomic_notifier_chain_register(&panic_notifier_list, &panic_block);
+        watchdog_task = kthread_run(watchdog, NULL, "khungtaskd");
+
+        return 0;
+}
+
+module_init(hung_task_init);
diff --git a/kernel/softlockup.c b/kernel/softlockup.c
index 85d5a2455103..88796c330838 100644
--- a/kernel/softlockup.c
+++ b/kernel/softlockup.c
@@ -166,97 +166,11 @@ void softlockup_tick(void)
 }
 
 /*
- * Have a reasonable limit on the number of tasks checked:
- */
-unsigned long __read_mostly sysctl_hung_task_check_count = 1024;
-
-/*
- * Zero means infinite timeout - no checking done:
- */
-unsigned long __read_mostly sysctl_hung_task_timeout_secs = 480;
-
-unsigned long __read_mostly sysctl_hung_task_warnings = 10;
-
-/*
- * Only do the hung-tasks check on one CPU:
- */
-static int check_cpu __read_mostly = -1;
-
-static void check_hung_task(struct task_struct *t, unsigned long now)
-{
-        unsigned long switch_count = t->nvcsw + t->nivcsw;
-
-        if (t->flags & PF_FROZEN)
-                return;
-
-        if (switch_count != t->last_switch_count || !t->last_switch_timestamp) {
-                t->last_switch_count = switch_count;
-                t->last_switch_timestamp = now;
-                return;
-        }
-        if ((long)(now - t->last_switch_timestamp) <
-                                        sysctl_hung_task_timeout_secs)
-                return;
-        if (!sysctl_hung_task_warnings)
-                return;
-        sysctl_hung_task_warnings--;
-
-        /*
-         * Ok, the task did not get scheduled for more than 2 minutes,
-         * complain:
-         */
-        printk(KERN_ERR "INFO: task %s:%d blocked for more than "
-                        "%ld seconds.\n", t->comm, t->pid,
-                        sysctl_hung_task_timeout_secs);
-        printk(KERN_ERR "\"echo 0 > /proc/sys/kernel/hung_task_timeout_secs\""
-                        " disables this message.\n");
-        sched_show_task(t);
-        __debug_show_held_locks(t);
-
-        t->last_switch_timestamp = now;
-        touch_nmi_watchdog();
-
-        if (softlockup_panic)
-                panic("softlockup: blocked tasks");
-}
-
-/*
- * Check whether a TASK_UNINTERRUPTIBLE does not get woken up for
- * a really long time (120 seconds). If that happens, print out
- * a warning.
- */
-static void check_hung_uninterruptible_tasks(int this_cpu)
-{
-        int max_count = sysctl_hung_task_check_count;
-        unsigned long now = get_timestamp(this_cpu);
-        struct task_struct *g, *t;
-
-        /*
-         * If the system crashed already then all bets are off,
-         * do not report extra hung tasks:
-         */
-        if (test_taint(TAINT_DIE) || did_panic)
-                return;
-
-        read_lock(&tasklist_lock);
-        do_each_thread(g, t) {
-                if (!--max_count)
-                        goto unlock;
-                /* use "==" to skip the TASK_KILLABLE tasks waiting on NFS */
-                if (t->state == TASK_UNINTERRUPTIBLE)
-                        check_hung_task(t, now);
-        } while_each_thread(g, t);
- unlock:
-        read_unlock(&tasklist_lock);
-}
-
-/*
  * The watchdog thread - runs every second and touches the timestamp.
  */
 static int watchdog(void *__bind_cpu)
 {
         struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 };
-        int this_cpu = (long)__bind_cpu;
 
         sched_setscheduler(current, SCHED_FIFO, &param);
 
@@ -276,11 +190,6 @@ static int watchdog(void *__bind_cpu)
                 if (kthread_should_stop())
                         break;
 
-                if (this_cpu == check_cpu) {
-                        if (sysctl_hung_task_timeout_secs)
-                                check_hung_uninterruptible_tasks(this_cpu);
-                }
-
                 set_current_state(TASK_INTERRUPTIBLE);
         }
         __set_current_state(TASK_RUNNING);
@@ -312,18 +221,9 @@ cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
                 break;
         case CPU_ONLINE:
         case CPU_ONLINE_FROZEN:
-                check_cpu = cpumask_any(cpu_online_mask);
                 wake_up_process(per_cpu(watchdog_task, hotcpu));
                 break;
 #ifdef CONFIG_HOTPLUG_CPU
-        case CPU_DOWN_PREPARE:
-        case CPU_DOWN_PREPARE_FROZEN:
-                if (hotcpu == check_cpu) {
-                        /* Pick any other online cpu. */
-                        check_cpu = cpumask_any_but(cpu_online_mask, hotcpu);
-                }
-                break;
-
         case CPU_UP_CANCELED:
         case CPU_UP_CANCELED_FROZEN:
                 if (!per_cpu(watchdog_task, hotcpu))
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 596dc31a7116..2481ed30d2b5 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -805,6 +805,19 @@ static struct ctl_table kern_table[] = {
                 .extra1         = &neg_one,
                 .extra2         = &sixty,
         },
+#endif
+#ifdef CONFIG_DETECT_HUNG_TASK
+        {
+                .ctl_name       = CTL_UNNUMBERED,
+                .procname       = "hung_task_panic",
+                .data           = &sysctl_hung_task_panic,
+                .maxlen         = sizeof(int),
+                .mode           = 0644,
+                .proc_handler   = &proc_dointvec_minmax,
+                .strategy       = &sysctl_intvec,
+                .extra1         = &zero,
+                .extra2         = &one,
+        },
         {
                 .ctl_name       = CTL_UNNUMBERED,
                 .procname       = "hung_task_check_count",
@@ -820,7 +833,7 @@ static struct ctl_table kern_table[] = {
                 .data           = &sysctl_hung_task_timeout_secs,
                 .maxlen         = sizeof(unsigned long),
                 .mode           = 0644,
-                .proc_handler   = &proc_doulongvec_minmax,
+                .proc_handler   = &proc_dohung_task_timeout_secs,
                 .strategy       = &sysctl_intvec,
         },
         {
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 4c9ae6085c75..883ecea22f37 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -186,6 +186,44 @@ config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE
         default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC
         default 1 if BOOTPARAM_SOFTLOCKUP_PANIC
 
+config DETECT_HUNG_TASK
+        bool "Detect Hung Tasks"
+        depends on DEBUG_KERNEL
+        default y
+        help
+          Say Y here to enable the kernel to detect "hung tasks",
+          which are bugs that cause the task to be stuck in
+          uninterruptible "D" state indefinitiley.
+
+          When a hung task is detected, the kernel will print the
+          current stack trace (which you should report), but the
+          task will stay in uninterruptible state. If lockdep is
+          enabled then all held locks will also be reported. This
+          feature has negligible overhead.
+
+config BOOTPARAM_HUNG_TASK_PANIC
+        bool "Panic (Reboot) On Hung Tasks"
+        depends on DETECT_HUNG_TASK
+        help
+          Say Y here to enable the kernel to panic on "hung tasks",
+          which are bugs that cause the kernel to leave a task stuck
+          in uninterruptible "D" state.
+
+          The panic can be used in combination with panic_timeout,
+          to cause the system to reboot automatically after a
+          hung task has been detected. This feature is useful for
+          high-availability systems that have uptime guarantees and
+          where a hung tasks must be resolved ASAP.
+
+          Say N if unsure.
+
+config BOOTPARAM_HUNG_TASK_PANIC_VALUE
+        int
+        depends on DETECT_HUNG_TASK
+        range 0 1
+        default 0 if !BOOTPARAM_HUNG_TASK_PANIC
+        default 1 if BOOTPARAM_HUNG_TASK_PANIC
+
 config SCHED_DEBUG
         bool "Collect scheduler debugging info"
         depends on DEBUG_KERNEL && PROC_FS