linux-stable/kernel/watchdog.c
Michal Hocko 9809b18fcf watchdog: update watchdog_thresh properly
watchdog_tresh controls how often nmi perf event counter checks per-cpu
hrtimer_interrupts counter and blows up if the counter hasn't changed
since the last check.  The counter is updated by per-cpu
watchdog_hrtimer hrtimer which is scheduled with 2/5 watchdog_thresh
period which guarantees that hrtimer is scheduled 2 times per the main
period.  Both hrtimer and perf event are started together when the
watchdog is enabled.

So far so good.  But...

But what happens when watchdog_thresh is updated from sysctl handler?

proc_dowatchdog will set a new sampling period and hrtimer callback
(watchdog_timer_fn) will use the new value in the next round.  The
problem, however, is that nobody tells the perf event that the sampling
period has changed so it is ticking with the period configured when it
has been set up.

This might result in an ear ripping dissonance between perf and hrtimer
parts if the watchdog_thresh is increased.  And even worse it might lead
to KABOOM if the watchdog is configured to panic on such a spurious
lockup.

This patch fixes the issue by updating both nmi perf even counter and
hrtimers if the threshold value has changed.

The nmi one is disabled and then reinitialized from scratch.  This has
an unpleasant side effect that the allocation of the new event might
fail theoretically so the hard lockup detector would be disabled for
such cpus.  On the other hand such a memory allocation failure is very
unlikely because the original event is deallocated right before.

It would be much nicer if we just changed perf event period but there
doesn't seem to be any API to do that right now.  It is also unfortunate
that perf_event_alloc uses GFP_KERNEL allocation unconditionally so we
cannot use on_each_cpu() and do the same thing from the per-cpu context.
The update from the current CPU should be safe because
perf_event_disable removes the event atomically before it clears the
per-cpu watchdog_ev so it cannot change anything under running handler
feet.

The hrtimer is simply restarted (thanks to Don Zickus who has pointed
this out) if it is queued because we cannot rely it will fire&adopt to
the new sampling period before a new nmi event triggers (when the
treshold is decreased).

[akpm@linux-foundation.org: the UP version of __smp_call_function_single ended up in the wrong place]
Signed-off-by: Michal Hocko <mhocko@suse.cz>
Acked-by: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-09-24 17:00:25 -07:00

608 lines
16 KiB
C

/*
* Detect hard and soft lockups on a system
*
* started by Don Zickus, Copyright (C) 2010 Red Hat, Inc.
*
* Note: Most of this code is borrowed heavily from the original softlockup
* detector, so thanks to Ingo for the initial implementation.
* Some chunks also taken from the old x86-specific nmi watchdog code, thanks
* to those contributors as well.
*/
#define pr_fmt(fmt) "NMI watchdog: " fmt
#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/notifier.h>
#include <linux/module.h>
#include <linux/sysctl.h>
#include <linux/smpboot.h>
#include <linux/sched/rt.h>
#include <asm/irq_regs.h>
#include <linux/kvm_para.h>
#include <linux/perf_event.h>
int watchdog_user_enabled = 1;
int __read_mostly watchdog_thresh = 10;
static int __read_mostly watchdog_running;
static u64 __read_mostly sample_period;
static DEFINE_PER_CPU(unsigned long, watchdog_touch_ts);
static DEFINE_PER_CPU(struct task_struct *, softlockup_watchdog);
static DEFINE_PER_CPU(struct hrtimer, watchdog_hrtimer);
static DEFINE_PER_CPU(bool, softlockup_touch_sync);
static DEFINE_PER_CPU(bool, soft_watchdog_warn);
static DEFINE_PER_CPU(unsigned long, hrtimer_interrupts);
static DEFINE_PER_CPU(unsigned long, soft_lockup_hrtimer_cnt);
#ifdef CONFIG_HARDLOCKUP_DETECTOR
static DEFINE_PER_CPU(bool, hard_watchdog_warn);
static DEFINE_PER_CPU(bool, watchdog_nmi_touch);
static DEFINE_PER_CPU(unsigned long, hrtimer_interrupts_saved);
static DEFINE_PER_CPU(struct perf_event *, watchdog_ev);
#endif
/* boot commands */
/*
* Should we panic when a soft-lockup or hard-lockup occurs:
*/
#ifdef CONFIG_HARDLOCKUP_DETECTOR
static int hardlockup_panic =
CONFIG_BOOTPARAM_HARDLOCKUP_PANIC_VALUE;
static int __init hardlockup_panic_setup(char *str)
{
if (!strncmp(str, "panic", 5))
hardlockup_panic = 1;
else if (!strncmp(str, "nopanic", 7))
hardlockup_panic = 0;
else if (!strncmp(str, "0", 1))
watchdog_user_enabled = 0;
return 1;
}
__setup("nmi_watchdog=", hardlockup_panic_setup);
#endif
unsigned int __read_mostly softlockup_panic =
CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC_VALUE;
static int __init softlockup_panic_setup(char *str)
{
softlockup_panic = simple_strtoul(str, NULL, 0);
return 1;
}
__setup("softlockup_panic=", softlockup_panic_setup);
static int __init nowatchdog_setup(char *str)
{
watchdog_user_enabled = 0;
return 1;
}
__setup("nowatchdog", nowatchdog_setup);
/* deprecated */
static int __init nosoftlockup_setup(char *str)
{
watchdog_user_enabled = 0;
return 1;
}
__setup("nosoftlockup", nosoftlockup_setup);
/* */
/*
* Hard-lockup warnings should be triggered after just a few seconds. Soft-
* lockups can have false positives under extreme conditions. So we generally
* want a higher threshold for soft lockups than for hard lockups. So we couple
* the thresholds with a factor: we make the soft threshold twice the amount of
* time the hard threshold is.
*/
static int get_softlockup_thresh(void)
{
return watchdog_thresh * 2;
}
/*
* 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)
{
return local_clock() >> 30LL; /* 2^30 ~= 10^9 */
}
static void set_sample_period(void)
{
/*
* convert watchdog_thresh from seconds to ns
* the divide by 5 is to give hrtimer several chances (two
* or three with the current relation between the soft
* and hard thresholds) to increment before the
* hardlockup detector generates a warning
*/
sample_period = get_softlockup_thresh() * ((u64)NSEC_PER_SEC / 5);
}
/* Commands for resetting the watchdog */
static void __touch_watchdog(void)
{
__this_cpu_write(watchdog_touch_ts, get_timestamp());
}
void touch_softlockup_watchdog(void)
{
__this_cpu_write(watchdog_touch_ts, 0);
}
EXPORT_SYMBOL(touch_softlockup_watchdog);
void touch_all_softlockup_watchdogs(void)
{
int cpu;
/*
* this is done lockless
* do we care if a 0 races with a timestamp?
* all it means is the softlock check starts one cycle later
*/
for_each_online_cpu(cpu)
per_cpu(watchdog_touch_ts, cpu) = 0;
}
#ifdef CONFIG_HARDLOCKUP_DETECTOR
void touch_nmi_watchdog(void)
{
if (watchdog_user_enabled) {
unsigned cpu;
for_each_present_cpu(cpu) {
if (per_cpu(watchdog_nmi_touch, cpu) != true)
per_cpu(watchdog_nmi_touch, cpu) = true;
}
}
touch_softlockup_watchdog();
}
EXPORT_SYMBOL(touch_nmi_watchdog);
#endif
void touch_softlockup_watchdog_sync(void)
{
__raw_get_cpu_var(softlockup_touch_sync) = true;
__raw_get_cpu_var(watchdog_touch_ts) = 0;
}
#ifdef CONFIG_HARDLOCKUP_DETECTOR
/* watchdog detector functions */
static int is_hardlockup(void)
{
unsigned long hrint = __this_cpu_read(hrtimer_interrupts);
if (__this_cpu_read(hrtimer_interrupts_saved) == hrint)
return 1;
__this_cpu_write(hrtimer_interrupts_saved, hrint);
return 0;
}
#endif
static int is_softlockup(unsigned long touch_ts)
{
unsigned long now = get_timestamp();
/* Warn about unreasonable delays: */
if (time_after(now, touch_ts + get_softlockup_thresh()))
return now - touch_ts;
return 0;
}
#ifdef CONFIG_HARDLOCKUP_DETECTOR
static struct perf_event_attr wd_hw_attr = {
.type = PERF_TYPE_HARDWARE,
.config = PERF_COUNT_HW_CPU_CYCLES,
.size = sizeof(struct perf_event_attr),
.pinned = 1,
.disabled = 1,
};
/* Callback function for perf event subsystem */
static void watchdog_overflow_callback(struct perf_event *event,
struct perf_sample_data *data,
struct pt_regs *regs)
{
/* Ensure the watchdog never gets throttled */
event->hw.interrupts = 0;
if (__this_cpu_read(watchdog_nmi_touch) == true) {
__this_cpu_write(watchdog_nmi_touch, false);
return;
}
/* check for a hardlockup
* This is done by making sure our timer interrupt
* is incrementing. The timer interrupt should have
* fired multiple times before we overflow'd. If it hasn't
* then this is a good indication the cpu is stuck
*/
if (is_hardlockup()) {
int this_cpu = smp_processor_id();
/* only print hardlockups once */
if (__this_cpu_read(hard_watchdog_warn) == true)
return;
if (hardlockup_panic)
panic("Watchdog detected hard LOCKUP on cpu %d", this_cpu);
else
WARN(1, "Watchdog detected hard LOCKUP on cpu %d", this_cpu);
__this_cpu_write(hard_watchdog_warn, true);
return;
}
__this_cpu_write(hard_watchdog_warn, false);
return;
}
#endif /* CONFIG_HARDLOCKUP_DETECTOR */
static void watchdog_interrupt_count(void)
{
__this_cpu_inc(hrtimer_interrupts);
}
static int watchdog_nmi_enable(unsigned int cpu);
static void watchdog_nmi_disable(unsigned int cpu);
/* watchdog kicker functions */
static enum hrtimer_restart watchdog_timer_fn(struct hrtimer *hrtimer)
{
unsigned long touch_ts = __this_cpu_read(watchdog_touch_ts);
struct pt_regs *regs = get_irq_regs();
int duration;
/* kick the hardlockup detector */
watchdog_interrupt_count();
/* kick the softlockup detector */
wake_up_process(__this_cpu_read(softlockup_watchdog));
/* .. and repeat */
hrtimer_forward_now(hrtimer, ns_to_ktime(sample_period));
if (touch_ts == 0) {
if (unlikely(__this_cpu_read(softlockup_touch_sync))) {
/*
* If the time stamp was touched atomically
* make sure the scheduler tick is up to date.
*/
__this_cpu_write(softlockup_touch_sync, false);
sched_clock_tick();
}
/* Clear the guest paused flag on watchdog reset */
kvm_check_and_clear_guest_paused();
__touch_watchdog();
return HRTIMER_RESTART;
}
/* check for a softlockup
* This is done by making sure a high priority task is
* being scheduled. The task touches the watchdog to
* indicate it is getting cpu time. If it hasn't then
* this is a good indication some task is hogging the cpu
*/
duration = is_softlockup(touch_ts);
if (unlikely(duration)) {
/*
* If a virtual machine is stopped by the host it can look to
* the watchdog like a soft lockup, check to see if the host
* stopped the vm before we issue the warning
*/
if (kvm_check_and_clear_guest_paused())
return HRTIMER_RESTART;
/* only warn once */
if (__this_cpu_read(soft_watchdog_warn) == true)
return HRTIMER_RESTART;
printk(KERN_EMERG "BUG: soft lockup - CPU#%d stuck for %us! [%s:%d]\n",
smp_processor_id(), duration,
current->comm, task_pid_nr(current));
print_modules();
print_irqtrace_events(current);
if (regs)
show_regs(regs);
else
dump_stack();
if (softlockup_panic)
panic("softlockup: hung tasks");
__this_cpu_write(soft_watchdog_warn, true);
} else
__this_cpu_write(soft_watchdog_warn, false);
return HRTIMER_RESTART;
}
static void watchdog_set_prio(unsigned int policy, unsigned int prio)
{
struct sched_param param = { .sched_priority = prio };
sched_setscheduler(current, policy, &param);
}
static void watchdog_enable(unsigned int cpu)
{
struct hrtimer *hrtimer = &__raw_get_cpu_var(watchdog_hrtimer);
/* kick off the timer for the hardlockup detector */
hrtimer_init(hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
hrtimer->function = watchdog_timer_fn;
/* Enable the perf event */
watchdog_nmi_enable(cpu);
/* done here because hrtimer_start can only pin to smp_processor_id() */
hrtimer_start(hrtimer, ns_to_ktime(sample_period),
HRTIMER_MODE_REL_PINNED);
/* initialize timestamp */
watchdog_set_prio(SCHED_FIFO, MAX_RT_PRIO - 1);
__touch_watchdog();
}
static void watchdog_disable(unsigned int cpu)
{
struct hrtimer *hrtimer = &__raw_get_cpu_var(watchdog_hrtimer);
watchdog_set_prio(SCHED_NORMAL, 0);
hrtimer_cancel(hrtimer);
/* disable the perf event */
watchdog_nmi_disable(cpu);
}
static void watchdog_cleanup(unsigned int cpu, bool online)
{
watchdog_disable(cpu);
}
static int watchdog_should_run(unsigned int cpu)
{
return __this_cpu_read(hrtimer_interrupts) !=
__this_cpu_read(soft_lockup_hrtimer_cnt);
}
/*
* The watchdog thread function - touches the timestamp.
*
* It only runs once every sample_period seconds (4 seconds by
* default) to reset the softlockup timestamp. If this gets delayed
* for more than 2*watchdog_thresh seconds then the debug-printout
* triggers in watchdog_timer_fn().
*/
static void watchdog(unsigned int cpu)
{
__this_cpu_write(soft_lockup_hrtimer_cnt,
__this_cpu_read(hrtimer_interrupts));
__touch_watchdog();
}
#ifdef CONFIG_HARDLOCKUP_DETECTOR
/*
* People like the simple clean cpu node info on boot.
* Reduce the watchdog noise by only printing messages
* that are different from what cpu0 displayed.
*/
static unsigned long cpu0_err;
static int watchdog_nmi_enable(unsigned int cpu)
{
struct perf_event_attr *wd_attr;
struct perf_event *event = per_cpu(watchdog_ev, cpu);
/* is it already setup and enabled? */
if (event && event->state > PERF_EVENT_STATE_OFF)
goto out;
/* it is setup but not enabled */
if (event != NULL)
goto out_enable;
wd_attr = &wd_hw_attr;
wd_attr->sample_period = hw_nmi_get_sample_period(watchdog_thresh);
/* Try to register using hardware perf events */
event = perf_event_create_kernel_counter(wd_attr, cpu, NULL, watchdog_overflow_callback, NULL);
/* save cpu0 error for future comparision */
if (cpu == 0 && IS_ERR(event))
cpu0_err = PTR_ERR(event);
if (!IS_ERR(event)) {
/* only print for cpu0 or different than cpu0 */
if (cpu == 0 || cpu0_err)
pr_info("enabled on all CPUs, permanently consumes one hw-PMU counter.\n");
goto out_save;
}
/* skip displaying the same error again */
if (cpu > 0 && (PTR_ERR(event) == cpu0_err))
return PTR_ERR(event);
/* vary the KERN level based on the returned errno */
if (PTR_ERR(event) == -EOPNOTSUPP)
pr_info("disabled (cpu%i): not supported (no LAPIC?)\n", cpu);
else if (PTR_ERR(event) == -ENOENT)
pr_warning("disabled (cpu%i): hardware events not enabled\n",
cpu);
else
pr_err("disabled (cpu%i): unable to create perf event: %ld\n",
cpu, PTR_ERR(event));
return PTR_ERR(event);
/* success path */
out_save:
per_cpu(watchdog_ev, cpu) = event;
out_enable:
perf_event_enable(per_cpu(watchdog_ev, cpu));
out:
return 0;
}
static void watchdog_nmi_disable(unsigned int cpu)
{
struct perf_event *event = per_cpu(watchdog_ev, cpu);
if (event) {
perf_event_disable(event);
per_cpu(watchdog_ev, cpu) = NULL;
/* should be in cleanup, but blocks oprofile */
perf_event_release_kernel(event);
}
return;
}
#else
static int watchdog_nmi_enable(unsigned int cpu) { return 0; }
static void watchdog_nmi_disable(unsigned int cpu) { return; }
#endif /* CONFIG_HARDLOCKUP_DETECTOR */
static struct smp_hotplug_thread watchdog_threads = {
.store = &softlockup_watchdog,
.thread_should_run = watchdog_should_run,
.thread_fn = watchdog,
.thread_comm = "watchdog/%u",
.setup = watchdog_enable,
.cleanup = watchdog_cleanup,
.park = watchdog_disable,
.unpark = watchdog_enable,
};
static void restart_watchdog_hrtimer(void *info)
{
struct hrtimer *hrtimer = &__raw_get_cpu_var(watchdog_hrtimer);
int ret;
/*
* No need to cancel and restart hrtimer if it is currently executing
* because it will reprogram itself with the new period now.
* We should never see it unqueued here because we are running per-cpu
* with interrupts disabled.
*/
ret = hrtimer_try_to_cancel(hrtimer);
if (ret == 1)
hrtimer_start(hrtimer, ns_to_ktime(sample_period),
HRTIMER_MODE_REL_PINNED);
}
static void update_timers(int cpu)
{
struct call_single_data data = {.func = restart_watchdog_hrtimer};
/*
* Make sure that perf event counter will adopt to a new
* sampling period. Updating the sampling period directly would
* be much nicer but we do not have an API for that now so
* let's use a big hammer.
* Hrtimer will adopt the new period on the next tick but this
* might be late already so we have to restart the timer as well.
*/
watchdog_nmi_disable(cpu);
__smp_call_function_single(cpu, &data, 1);
watchdog_nmi_enable(cpu);
}
static void update_timers_all_cpus(void)
{
int cpu;
get_online_cpus();
preempt_disable();
for_each_online_cpu(cpu)
update_timers(cpu);
preempt_enable();
put_online_cpus();
}
static int watchdog_enable_all_cpus(bool sample_period_changed)
{
int err = 0;
if (!watchdog_running) {
err = smpboot_register_percpu_thread(&watchdog_threads);
if (err)
pr_err("Failed to create watchdog threads, disabled\n");
else
watchdog_running = 1;
} else if (sample_period_changed) {
update_timers_all_cpus();
}
return err;
}
/* prepare/enable/disable routines */
/* sysctl functions */
#ifdef CONFIG_SYSCTL
static void watchdog_disable_all_cpus(void)
{
if (watchdog_running) {
watchdog_running = 0;
smpboot_unregister_percpu_thread(&watchdog_threads);
}
}
/*
* proc handler for /proc/sys/kernel/nmi_watchdog,watchdog_thresh
*/
int proc_dowatchdog(struct ctl_table *table, int write,
void __user *buffer, size_t *lenp, loff_t *ppos)
{
int err, old_thresh, old_enabled;
static DEFINE_MUTEX(watchdog_proc_mutex);
mutex_lock(&watchdog_proc_mutex);
old_thresh = ACCESS_ONCE(watchdog_thresh);
old_enabled = ACCESS_ONCE(watchdog_user_enabled);
err = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
if (err || !write)
goto out;
set_sample_period();
/*
* Watchdog threads shouldn't be enabled if they are
* disabled. The 'watchdog_running' variable check in
* watchdog_*_all_cpus() function takes care of this.
*/
if (watchdog_user_enabled && watchdog_thresh)
err = watchdog_enable_all_cpus(old_thresh != watchdog_thresh);
else
watchdog_disable_all_cpus();
/* Restore old values on failure */
if (err) {
watchdog_thresh = old_thresh;
watchdog_user_enabled = old_enabled;
}
out:
mutex_unlock(&watchdog_proc_mutex);
return err;
}
#endif /* CONFIG_SYSCTL */
void __init lockup_detector_init(void)
{
set_sample_period();
if (watchdog_user_enabled)
watchdog_enable_all_cpus(false);
}