summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorIngo Molnar <mingo@elte.hu>2006-03-24 03:18:41 -0800
committerLinus Torvalds <torvalds@g5.osdl.org>2006-03-24 07:33:30 -0800
commit6687a97d4041f996f725902d2990e5de6ef5cbe5 (patch)
tree6ab982091cde7179d94cf592f9c669fd22d93a23
parent6a4d11c2abc57ed7ca42041e5f68ae4f7f640a81 (diff)
[PATCH] timer-irq-driven soft-watchdog, cleanups
Make the softlockup detector purely timer-interrupt driven, removing softirq-context (timer) dependencies. This means that if the softlockup watchdog triggers, it has truly observed a longer than 10 seconds scheduling delay of a SCHED_FIFO prio 99 task. (the patch also turns off the softlockup detector during the initial bootup phase and does small style fixes) Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
-rw-r--r--include/linux/sched.h4
-rw-r--r--kernel/softlockup.c54
-rw-r--r--kernel/timer.c2
3 files changed, 33 insertions, 27 deletions
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 2cda439ece4..e0054c1b9a0 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -206,11 +206,11 @@ extern void update_process_times(int user);
extern void scheduler_tick(void);
#ifdef CONFIG_DETECT_SOFTLOCKUP
-extern void softlockup_tick(struct pt_regs *regs);
+extern void softlockup_tick(void);
extern void spawn_softlockup_task(void);
extern void touch_softlockup_watchdog(void);
#else
-static inline void softlockup_tick(struct pt_regs *regs)
+static inline void softlockup_tick(void)
{
}
static inline void spawn_softlockup_task(void)
diff --git a/kernel/softlockup.c b/kernel/softlockup.c
index c67189a25d5..dd9524fa649 100644
--- a/kernel/softlockup.c
+++ b/kernel/softlockup.c
@@ -1,12 +1,11 @@
/*
* Detect Soft Lockups
*
- * started by Ingo Molnar, (C) 2005, Red Hat
+ * started by Ingo Molnar, Copyright (C) 2005, 2006 Red Hat, Inc.
*
* this code detects soft lockups: incidents in where on a CPU
* the kernel does not reschedule for 10 seconds or more.
*/
-
#include <linux/mm.h>
#include <linux/cpu.h>
#include <linux/init.h>
@@ -17,13 +16,14 @@
static DEFINE_SPINLOCK(print_lock);
-static DEFINE_PER_CPU(unsigned long, timestamp) = 0;
-static DEFINE_PER_CPU(unsigned long, print_timestamp) = 0;
+static DEFINE_PER_CPU(unsigned long, touch_timestamp);
+static DEFINE_PER_CPU(unsigned long, print_timestamp);
static DEFINE_PER_CPU(struct task_struct *, watchdog_task);
static int did_panic = 0;
-static int softlock_panic(struct notifier_block *this, unsigned long event,
- void *ptr)
+
+static int
+softlock_panic(struct notifier_block *this, unsigned long event, void *ptr)
{
did_panic = 1;
@@ -36,7 +36,7 @@ static struct notifier_block panic_block = {
void touch_softlockup_watchdog(void)
{
- per_cpu(timestamp, raw_smp_processor_id()) = jiffies;
+ per_cpu(touch_timestamp, raw_smp_processor_id()) = jiffies;
}
EXPORT_SYMBOL(touch_softlockup_watchdog);
@@ -44,25 +44,35 @@ EXPORT_SYMBOL(touch_softlockup_watchdog);
* This callback runs from the timer interrupt, and checks
* whether the watchdog thread has hung or not:
*/
-void softlockup_tick(struct pt_regs *regs)
+void softlockup_tick(void)
{
int this_cpu = smp_processor_id();
- unsigned long timestamp = per_cpu(timestamp, this_cpu);
+ unsigned long touch_timestamp = per_cpu(touch_timestamp, this_cpu);
- if (per_cpu(print_timestamp, this_cpu) == timestamp)
+ /* prevent double reports: */
+ if (per_cpu(print_timestamp, this_cpu) == touch_timestamp ||
+ did_panic ||
+ !per_cpu(watchdog_task, this_cpu))
return;
- /* Do not cause a second panic when there already was one */
- if (did_panic)
+ /* do not print during early bootup: */
+ if (unlikely(system_state != SYSTEM_RUNNING)) {
+ touch_softlockup_watchdog();
return;
+ }
- if (time_after(jiffies, timestamp + 10*HZ)) {
- per_cpu(print_timestamp, this_cpu) = timestamp;
+ /* Wake up the high-prio watchdog task every second: */
+ if (time_after(jiffies, touch_timestamp + HZ))
+ wake_up_process(per_cpu(watchdog_task, this_cpu));
+
+ /* Warn about unreasonable 10+ seconds delays: */
+ if (time_after(jiffies, touch_timestamp + 10*HZ)) {
+ per_cpu(print_timestamp, this_cpu) = touch_timestamp;
spin_lock(&print_lock);
printk(KERN_ERR "BUG: soft lockup detected on CPU#%d!\n",
this_cpu);
- show_regs(regs);
+ dump_stack();
spin_unlock(&print_lock);
}
}
@@ -77,18 +87,16 @@ static int watchdog(void * __bind_cpu)
sched_setscheduler(current, SCHED_FIFO, &param);
current->flags |= PF_NOFREEZE;
- set_current_state(TASK_INTERRUPTIBLE);
-
/*
- * Run briefly once per second - if this gets delayed for
- * more than 10 seconds then the debug-printout triggers
- * in softlockup_tick():
+ * Run briefly once per second to reset the softlockup timestamp.
+ * If this gets delayed for more than 10 seconds then the
+ * debug-printout triggers in softlockup_tick().
*/
while (!kthread_should_stop()) {
- msleep_interruptible(1000);
+ set_current_state(TASK_INTERRUPTIBLE);
touch_softlockup_watchdog();
+ schedule();
}
- __set_current_state(TASK_RUNNING);
return 0;
}
@@ -114,7 +122,6 @@ cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
kthread_bind(p, hotcpu);
break;
case CPU_ONLINE:
-
wake_up_process(per_cpu(watchdog_task, hotcpu));
break;
#ifdef CONFIG_HOTPLUG_CPU
@@ -146,4 +153,3 @@ __init void spawn_softlockup_task(void)
notifier_chain_register(&panic_notifier_list, &panic_block);
}
-
diff --git a/kernel/timer.c b/kernel/timer.c
index 4427e725ccd..17d956cebcb 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -915,6 +915,7 @@ static void run_timer_softirq(struct softirq_action *h)
void run_local_timers(void)
{
raise_softirq(TIMER_SOFTIRQ);
+ softlockup_tick();
}
/*
@@ -945,7 +946,6 @@ void do_timer(struct pt_regs *regs)
/* prevent loading jiffies before storing new jiffies_64 value. */
barrier();
update_times();
- softlockup_tick(regs);
}
#ifdef __ARCH_WANT_SYS_ALARM