watchdog.c 24.4 KB
Newer Older
1 2 3 4 5
/*
 * Detect hard and soft lockups on a system
 *
 * started by Don Zickus, Copyright (C) 2010 Red Hat, Inc.
 *
6 7 8
 * 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
9 10 11
 * to those contributors as well.
 */

12 13
#define pr_fmt(fmt) "NMI watchdog: " fmt

14 15 16 17 18 19
#include <linux/mm.h>
#include <linux/cpu.h>
#include <linux/nmi.h>
#include <linux/init.h>
#include <linux/module.h>
#include <linux/sysctl.h>
20
#include <linux/smpboot.h>
21
#include <linux/sched/rt.h>
22
#include <uapi/linux/sched/types.h>
23
#include <linux/tick.h>
24
#include <linux/workqueue.h>
25
#include <linux/sched/clock.h>
26
#include <linux/sched/debug.h>
27 28

#include <asm/irq_regs.h>
29
#include <linux/kvm_para.h>
30
#include <linux/kthread.h>
31

32
/* Watchdog configuration */
33 34
static DEFINE_MUTEX(watchdog_proc_mutex);

35 36 37 38 39
int __read_mostly nmi_watchdog_enabled;

#if defined(CONFIG_HARDLOCKUP_DETECTOR) || defined(CONFIG_HAVE_NMI_WATCHDOG)
unsigned long __read_mostly watchdog_enabled = SOFT_WATCHDOG_ENABLED |
						NMI_WATCHDOG_ENABLED;
40
#else
41
unsigned long __read_mostly watchdog_enabled = SOFT_WATCHDOG_ENABLED;
42
#endif
43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80

#ifdef CONFIG_HARDLOCKUP_DETECTOR
/* boot commands */
/*
 * Should we panic when a soft-lockup or hard-lockup occurs:
 */
unsigned int __read_mostly hardlockup_panic =
			CONFIG_BOOTPARAM_HARDLOCKUP_PANIC_VALUE;
/*
 * We may not want to enable hard lockup detection by default in all cases,
 * for example when running the kernel as a guest on a hypervisor. In these
 * cases this function can be called to disable hard lockup detection. This
 * function should only be executed once by the boot processor before the
 * kernel command line parameters are parsed, because otherwise it is not
 * possible to override this in hardlockup_panic_setup().
 */
void hardlockup_detector_disable(void)
{
	watchdog_enabled &= ~NMI_WATCHDOG_ENABLED;
}

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_enabled &= ~NMI_WATCHDOG_ENABLED;
	else if (!strncmp(str, "1", 1))
		watchdog_enabled |= NMI_WATCHDOG_ENABLED;
	return 1;
}
__setup("nmi_watchdog=", hardlockup_panic_setup);

#endif

#ifdef CONFIG_SOFTLOCKUP_DETECTOR
81
int __read_mostly soft_watchdog_enabled;
82 83
#endif

84
int __read_mostly watchdog_user_enabled;
85
int __read_mostly watchdog_thresh = 10;
86

87 88
#ifdef CONFIG_SMP
int __read_mostly sysctl_softlockup_all_cpu_backtrace;
89
int __read_mostly sysctl_hardlockup_all_cpu_backtrace;
90
#endif
91
struct cpumask watchdog_cpumask __read_mostly;
92 93
unsigned long *watchdog_cpumask_bits = cpumask_bits(&watchdog_cpumask);

94 95 96 97 98
/*
 * The 'watchdog_running' variable is set to 1 when the watchdog threads
 * are registered/started and is set to 0 when the watchdog threads are
 * unregistered/stopped, so it is an indicator whether the threads exist.
 */
99
static int __read_mostly watchdog_running;
100 101 102 103 104 105 106 107 108 109 110 111
/*
 * If a subsystem has a need to deactivate the watchdog temporarily, it
 * can use the suspend/resume interface to achieve this. The content of
 * the 'watchdog_suspended' variable reflects this state. Existing threads
 * are parked/unparked by the lockup_detector_{suspend|resume} functions
 * (see comment blocks pertaining to those functions for further details).
 *
 * 'watchdog_suspended' also prevents threads from being registered/started
 * or unregistered/stopped via parameters in /proc/sys/kernel, so the state
 * of 'watchdog_running' cannot change while the watchdog is deactivated
 * temporarily (see related code in 'proc' handlers).
 */
112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132
int __read_mostly watchdog_suspended;

/*
 * These functions can be overridden if an architecture implements its
 * own hardlockup detector.
 */
int __weak watchdog_nmi_enable(unsigned int cpu)
{
	return 0;
}
void __weak watchdog_nmi_disable(unsigned int cpu)
{
}

#ifdef CONFIG_SOFTLOCKUP_DETECTOR

/* Helper for online, unparked cpus. */
#define for_each_watchdog_cpu(cpu) \
	for_each_cpu_and((cpu), cpu_online_mask, &watchdog_cpumask)

atomic_t watchdog_park_in_progress = ATOMIC_INIT(0);
133

134
static u64 __read_mostly sample_period;
135 136 137 138 139 140

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);
141 142
static DEFINE_PER_CPU(unsigned long, hrtimer_interrupts);
static DEFINE_PER_CPU(unsigned long, soft_lockup_hrtimer_cnt);
143
static DEFINE_PER_CPU(struct task_struct *, softlockup_task_ptr_saved);
144
static DEFINE_PER_CPU(unsigned long, hrtimer_interrupts_saved);
145
static unsigned long soft_lockup_nmi_warn;
146 147 148 149 150 151 152 153 154 155 156 157 158 159

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)
{
160
	watchdog_enabled = 0;
161 162 163 164 165 166
	return 1;
}
__setup("nowatchdog", nowatchdog_setup);

static int __init nosoftlockup_setup(char *str)
{
167
	watchdog_enabled &= ~SOFT_WATCHDOG_ENABLED;
168 169 170
	return 1;
}
__setup("nosoftlockup", nosoftlockup_setup);
171

172 173 174 175 176 177 178 179
#ifdef CONFIG_SMP
static int __init softlockup_all_cpu_backtrace_setup(char *str)
{
	sysctl_softlockup_all_cpu_backtrace =
		!!simple_strtol(str, NULL, 0);
	return 1;
}
__setup("softlockup_all_cpu_backtrace=", softlockup_all_cpu_backtrace_setup);
180
#ifdef CONFIG_HARDLOCKUP_DETECTOR
181 182 183 184 185 186 187
static int __init hardlockup_all_cpu_backtrace_setup(char *str)
{
	sysctl_hardlockup_all_cpu_backtrace =
		!!simple_strtol(str, NULL, 0);
	return 1;
}
__setup("hardlockup_all_cpu_backtrace=", hardlockup_all_cpu_backtrace_setup);
188
#endif
189
#endif
190

191 192 193 194 195 196 197
/*
 * 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.
 */
198
static int get_softlockup_thresh(void)
199 200 201
{
	return watchdog_thresh * 2;
}
202 203 204 205 206 207

/*
 * 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.
 */
208
static unsigned long get_timestamp(void)
209
{
210
	return running_clock() >> 30LL;  /* 2^30 ~= 10^9 */
211 212
}

213
static void set_sample_period(void)
214 215
{
	/*
216
	 * convert watchdog_thresh from seconds to ns
217 218 219 220
	 * 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
221
	 */
222
	sample_period = get_softlockup_thresh() * ((u64)NSEC_PER_SEC / 5);
223 224 225 226 227
}

/* Commands for resetting the watchdog */
static void __touch_watchdog(void)
{
228
	__this_cpu_write(watchdog_touch_ts, get_timestamp());
229 230
}

231 232 233 234 235 236 237 238 239
/**
 * touch_softlockup_watchdog_sched - touch watchdog on scheduler stalls
 *
 * Call when the scheduler may have stalled for legitimate reasons
 * preventing the watchdog task from executing - e.g. the scheduler
 * entering idle state.  This should only be used for scheduler events.
 * Use touch_softlockup_watchdog() for everything else.
 */
void touch_softlockup_watchdog_sched(void)
240
{
241 242 243 244 245
	/*
	 * Preemption can be enabled.  It doesn't matter which CPU's timestamp
	 * gets zeroed here, so use the raw_ operation.
	 */
	raw_cpu_write(watchdog_touch_ts, 0);
246
}
247 248 249 250

void touch_softlockup_watchdog(void)
{
	touch_softlockup_watchdog_sched();
251
	wq_watchdog_touch(raw_smp_processor_id());
252
}
253
EXPORT_SYMBOL(touch_softlockup_watchdog);
254

255
void touch_all_softlockup_watchdogs(void)
256 257 258 259 260 261 262 263
{
	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
	 */
264
	for_each_watchdog_cpu(cpu)
265
		per_cpu(watchdog_touch_ts, cpu) = 0;
266
	wq_watchdog_touch(-1);
267 268 269 270
}

void touch_softlockup_watchdog_sync(void)
{
271 272
	__this_cpu_write(softlockup_touch_sync, true);
	__this_cpu_write(watchdog_touch_ts, 0);
273 274
}

275
static int is_softlockup(unsigned long touch_ts)
276
{
277
	unsigned long now = get_timestamp();
278

279
	if ((watchdog_enabled & SOFT_WATCHDOG_ENABLED) && watchdog_thresh){
280 281 282 283
		/* Warn about unreasonable delays. */
		if (time_after(now, touch_ts + get_softlockup_thresh()))
			return now - touch_ts;
	}
284 285 286
	return 0;
}

287 288
/* watchdog detector functions */
bool is_hardlockup(void)
289
{
290
	unsigned long hrint = __this_cpu_read(hrtimer_interrupts);
291

292 293 294 295 296
	if (__this_cpu_read(hrtimer_interrupts_saved) == hrint)
		return true;

	__this_cpu_write(hrtimer_interrupts_saved, hrint);
	return false;
297
}
298 299

static void watchdog_interrupt_count(void)
300
{
301
	__this_cpu_inc(hrtimer_interrupts);
302
}
303

304 305 306
static int watchdog_enable_all_cpus(void);
static void watchdog_disable_all_cpus(void);

307 308 309
/* watchdog kicker functions */
static enum hrtimer_restart watchdog_timer_fn(struct hrtimer *hrtimer)
{
310
	unsigned long touch_ts = __this_cpu_read(watchdog_touch_ts);
311 312
	struct pt_regs *regs = get_irq_regs();
	int duration;
313
	int softlockup_all_cpu_backtrace = sysctl_softlockup_all_cpu_backtrace;
314

315 316 317
	if (atomic_read(&watchdog_park_in_progress) != 0)
		return HRTIMER_NORESTART;

318 319 320 321
	/* kick the hardlockup detector */
	watchdog_interrupt_count();

	/* kick the softlockup detector */
322
	wake_up_process(__this_cpu_read(softlockup_watchdog));
323 324

	/* .. and repeat */
325
	hrtimer_forward_now(hrtimer, ns_to_ktime(sample_period));
326 327

	if (touch_ts == 0) {
328
		if (unlikely(__this_cpu_read(softlockup_touch_sync))) {
329 330 331 332
			/*
			 * If the time stamp was touched atomically
			 * make sure the scheduler tick is up to date.
			 */
333
			__this_cpu_write(softlockup_touch_sync, false);
334 335
			sched_clock_tick();
		}
336 337 338

		/* Clear the guest paused flag on watchdog reset */
		kvm_check_and_clear_guest_paused();
339 340 341 342 343 344 345 346 347 348
		__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
	 */
349
	duration = is_softlockup(touch_ts);
350
	if (unlikely(duration)) {
351 352 353 354 355 356 357 358
		/*
		 * 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;

359
		/* only warn once */
360 361 362 363 364 365 366 367 368 369 370 371 372 373
		if (__this_cpu_read(soft_watchdog_warn) == true) {
			/*
			 * When multiple processes are causing softlockups the
			 * softlockup detector only warns on the first one
			 * because the code relies on a full quiet cycle to
			 * re-arm.  The second process prevents the quiet cycle
			 * and never gets reported.  Use task pointers to detect
			 * this.
			 */
			if (__this_cpu_read(softlockup_task_ptr_saved) !=
			    current) {
				__this_cpu_write(soft_watchdog_warn, false);
				__touch_watchdog();
			}
374
			return HRTIMER_RESTART;
375
		}
376

377 378 379 380 381 382 383 384 385 386 387
		if (softlockup_all_cpu_backtrace) {
			/* Prevent multiple soft-lockup reports if one cpu is already
			 * engaged in dumping cpu back traces
			 */
			if (test_and_set_bit(0, &soft_lockup_nmi_warn)) {
				/* Someone else will report us. Let's give up */
				__this_cpu_write(soft_watchdog_warn, true);
				return HRTIMER_RESTART;
			}
		}

388
		pr_emerg("BUG: soft lockup - CPU#%d stuck for %us! [%s:%d]\n",
389
			smp_processor_id(), duration,
390
			current->comm, task_pid_nr(current));
391
		__this_cpu_write(softlockup_task_ptr_saved, current);
392 393 394 395 396 397 398
		print_modules();
		print_irqtrace_events(current);
		if (regs)
			show_regs(regs);
		else
			dump_stack();

399 400 401 402 403 404 405 406 407 408 409
		if (softlockup_all_cpu_backtrace) {
			/* Avoid generating two back traces for current
			 * given that one is already made above
			 */
			trigger_allbutself_cpu_backtrace();

			clear_bit(0, &soft_lockup_nmi_warn);
			/* Barrier to sync with other cpus */
			smp_mb__after_atomic();
		}

Josh Hunt's avatar
Josh Hunt committed
410
		add_taint(TAINT_SOFTLOCKUP, LOCKDEP_STILL_OK);
411 412
		if (softlockup_panic)
			panic("softlockup: hung tasks");
413
		__this_cpu_write(soft_watchdog_warn, true);
414
	} else
415
		__this_cpu_write(soft_watchdog_warn, false);
416 417 418 419

	return HRTIMER_RESTART;
}

420 421 422
static void watchdog_set_prio(unsigned int policy, unsigned int prio)
{
	struct sched_param param = { .sched_priority = prio };
423

424 425 426 427
	sched_setscheduler(current, policy, &param);
}

static void watchdog_enable(unsigned int cpu)
428
{
429
	struct hrtimer *hrtimer = raw_cpu_ptr(&watchdog_hrtimer);
430

431 432 433 434
	/* kick off the timer for the hardlockup detector */
	hrtimer_init(hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
	hrtimer->function = watchdog_timer_fn;

435 436
	/* Enable the perf event */
	watchdog_nmi_enable(cpu);
437 438

	/* done here because hrtimer_start can only pin to smp_processor_id() */
439
	hrtimer_start(hrtimer, ns_to_ktime(sample_period),
440 441
		      HRTIMER_MODE_REL_PINNED);

442 443 444 445
	/* initialize timestamp */
	watchdog_set_prio(SCHED_FIFO, MAX_RT_PRIO - 1);
	__touch_watchdog();
}
446

447 448
static void watchdog_disable(unsigned int cpu)
{
449
	struct hrtimer *hrtimer = raw_cpu_ptr(&watchdog_hrtimer);
450

451 452 453 454
	watchdog_set_prio(SCHED_NORMAL, 0);
	hrtimer_cancel(hrtimer);
	/* disable the perf event */
	watchdog_nmi_disable(cpu);
455 456
}

457 458 459 460 461
static void watchdog_cleanup(unsigned int cpu, bool online)
{
	watchdog_disable(cpu);
}

462 463 464 465 466 467 468 469 470
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.
 *
471
 * It only runs once every sample_period seconds (4 seconds by
472 473 474 475 476 477 478 479 480
 * 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();
481 482 483 484 485 486 487 488 489 490 491 492 493 494 495

	/*
	 * watchdog_nmi_enable() clears the NMI_WATCHDOG_ENABLED bit in the
	 * failure path. Check for failures that can occur asynchronously -
	 * for example, when CPUs are on-lined - and shut down the hardware
	 * perf event on each CPU accordingly.
	 *
	 * The only non-obvious place this bit can be cleared is through
	 * watchdog_nmi_enable(), so a pr_info() is placed there.  Placing a
	 * pr_info here would be too noisy as it would result in a message
	 * every few seconds if the hardlockup was disabled but the softlockup
	 * enabled.
	 */
	if (!(watchdog_enabled & NMI_WATCHDOG_ENABLED))
		watchdog_nmi_disable(cpu);
496
}
497

498 499 500 501 502 503 504 505 506 507 508
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,
};

509 510
/*
 * park all watchdog threads that are specified in 'watchdog_cpumask'
511 512 513 514 515 516
 *
 * This function returns an error if kthread_park() of a watchdog thread
 * fails. In this situation, the watchdog threads of some CPUs can already
 * be parked and the watchdog threads of other CPUs can still be runnable.
 * Callers are expected to handle this special condition as appropriate in
 * their context.
517 518 519
 *
 * This function may only be called in a context that is protected against
 * races with CPU hotplug - for example, via get_online_cpus().
520 521 522 523 524
 */
static int watchdog_park_threads(void)
{
	int cpu, ret = 0;

525 526
	atomic_set(&watchdog_park_in_progress, 1);

527 528 529 530 531 532
	for_each_watchdog_cpu(cpu) {
		ret = kthread_park(per_cpu(softlockup_watchdog, cpu));
		if (ret)
			break;
	}

533 534
	atomic_set(&watchdog_park_in_progress, 0);

535 536 537 538 539
	return ret;
}

/*
 * unpark all watchdog threads that are specified in 'watchdog_cpumask'
540 541 542
 *
 * This function may only be called in a context that is protected against
 * races with CPU hotplug - for example, via get_online_cpus().
543 544 545 546 547 548 549 550 551
 */
static void watchdog_unpark_threads(void)
{
	int cpu;

	for_each_watchdog_cpu(cpu)
		kthread_unpark(per_cpu(softlockup_watchdog, cpu));
}

552
static int update_watchdog_all_cpus(void)
553
{
554 555 556 557 558 559
	int ret;

	ret = watchdog_park_threads();
	if (ret)
		return ret;

560
	watchdog_unpark_threads();
561 562

	return 0;
563 564
}

565
static int watchdog_enable_all_cpus(void)
566
{
567
	int err = 0;
568

569
	if (!watchdog_running) {
570 571
		err = smpboot_register_percpu_thread_cpumask(&watchdog_threads,
							     &watchdog_cpumask);
572 573
		if (err)
			pr_err("Failed to create watchdog threads, disabled\n");
574
		else
575
			watchdog_running = 1;
576 577 578 579 580
	} else {
		/*
		 * Enable/disable the lockup detectors or
		 * change the sample period 'on the fly'.
		 */
581 582 583 584 585 586
		err = update_watchdog_all_cpus();

		if (err) {
			watchdog_disable_all_cpus();
			pr_err("Failed to update lockup detectors, disabled\n");
		}
587
	}
588

589 590 591
	if (err)
		watchdog_enabled = 0;

592
	return err;
593 594 595 596
}

static void watchdog_disable_all_cpus(void)
{
597 598
	if (watchdog_running) {
		watchdog_running = 0;
599
		smpboot_unregister_percpu_thread(&watchdog_threads);
600
	}
601 602
}

603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677
#else /* SOFTLOCKUP */
static int watchdog_park_threads(void)
{
	return 0;
}

static void watchdog_unpark_threads(void)
{
}

static int watchdog_enable_all_cpus(void)
{
	return 0;
}

static void watchdog_disable_all_cpus(void)
{
}

static void set_sample_period(void)
{
}
#endif /* SOFTLOCKUP */

/*
 * Suspend the hard and soft lockup detector by parking the watchdog threads.
 */
int lockup_detector_suspend(void)
{
	int ret = 0;

	get_online_cpus();
	mutex_lock(&watchdog_proc_mutex);
	/*
	 * Multiple suspend requests can be active in parallel (counted by
	 * the 'watchdog_suspended' variable). If the watchdog threads are
	 * running, the first caller takes care that they will be parked.
	 * The state of 'watchdog_running' cannot change while a suspend
	 * request is active (see related code in 'proc' handlers).
	 */
	if (watchdog_running && !watchdog_suspended)
		ret = watchdog_park_threads();

	if (ret == 0)
		watchdog_suspended++;
	else {
		watchdog_disable_all_cpus();
		pr_err("Failed to suspend lockup detectors, disabled\n");
		watchdog_enabled = 0;
	}

	mutex_unlock(&watchdog_proc_mutex);

	return ret;
}

/*
 * Resume the hard and soft lockup detector by unparking the watchdog threads.
 */
void lockup_detector_resume(void)
{
	mutex_lock(&watchdog_proc_mutex);

	watchdog_suspended--;
	/*
	 * The watchdog threads are unparked if they were previously running
	 * and if there is no more active suspend request.
	 */
	if (watchdog_running && !watchdog_suspended)
		watchdog_unpark_threads();

	mutex_unlock(&watchdog_proc_mutex);
	put_online_cpus();
}

678 679
#ifdef CONFIG_SYSCTL

680
/*
681 682 683 684 685 686 687 688 689 690 691 692 693 694
 * Update the run state of the lockup detectors.
 */
static int proc_watchdog_update(void)
{
	int err = 0;

	/*
	 * Watchdog threads won't be started if they are already active.
	 * The 'watchdog_running' variable in watchdog_*_all_cpus() takes
	 * care of this. If those threads are already active, the sample
	 * period will be updated and the lockup detectors will be enabled
	 * or disabled 'on the fly'.
	 */
	if (watchdog_enabled && watchdog_thresh)
695
		err = watchdog_enable_all_cpus();
696 697 698 699 700 701 702
	else
		watchdog_disable_all_cpus();

	return err;

}

703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720
/*
 * common function for watchdog, nmi_watchdog and soft_watchdog parameter
 *
 * caller             | table->data points to | 'which' contains the flag(s)
 * -------------------|-----------------------|-----------------------------
 * proc_watchdog      | watchdog_user_enabled | NMI_WATCHDOG_ENABLED or'ed
 *                    |                       | with SOFT_WATCHDOG_ENABLED
 * -------------------|-----------------------|-----------------------------
 * proc_nmi_watchdog  | nmi_watchdog_enabled  | NMI_WATCHDOG_ENABLED
 * -------------------|-----------------------|-----------------------------
 * proc_soft_watchdog | soft_watchdog_enabled | SOFT_WATCHDOG_ENABLED
 */
static int proc_watchdog_common(int which, struct ctl_table *table, int write,
				void __user *buffer, size_t *lenp, loff_t *ppos)
{
	int err, old, new;
	int *watchdog_param = (int *)table->data;

721
	get_online_cpus();
722 723
	mutex_lock(&watchdog_proc_mutex);

724 725 726 727 728 729
	if (watchdog_suspended) {
		/* no parameter changes allowed while watchdog is suspended */
		err = -EAGAIN;
		goto out;
	}

730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762
	/*
	 * If the parameter is being read return the state of the corresponding
	 * bit(s) in 'watchdog_enabled', else update 'watchdog_enabled' and the
	 * run state of the lockup detectors.
	 */
	if (!write) {
		*watchdog_param = (watchdog_enabled & which) != 0;
		err = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
	} else {
		err = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
		if (err)
			goto out;

		/*
		 * There is a race window between fetching the current value
		 * from 'watchdog_enabled' and storing the new value. During
		 * this race window, watchdog_nmi_enable() can sneak in and
		 * clear the NMI_WATCHDOG_ENABLED bit in 'watchdog_enabled'.
		 * The 'cmpxchg' detects this race and the loop retries.
		 */
		do {
			old = watchdog_enabled;
			/*
			 * If the parameter value is not zero set the
			 * corresponding bit(s), else clear it(them).
			 */
			if (*watchdog_param)
				new = old | which;
			else
				new = old & ~which;
		} while (cmpxchg(&watchdog_enabled, old, new) != old);

		/*
763 764 765 766 767
		 * Update the run state of the lockup detectors. There is _no_
		 * need to check the value returned by proc_watchdog_update()
		 * and to restore the previous value of 'watchdog_enabled' as
		 * both lockup detectors are disabled if proc_watchdog_update()
		 * returns an error.
768
		 */
769 770 771
		if (old == new)
			goto out;

772 773 774 775
		err = proc_watchdog_update();
	}
out:
	mutex_unlock(&watchdog_proc_mutex);
776
	put_online_cpus();
777 778 779
	return err;
}

780 781 782 783 784 785 786 787 788 789 790 791
/*
 * /proc/sys/kernel/watchdog
 */
int proc_watchdog(struct ctl_table *table, int write,
		  void __user *buffer, size_t *lenp, loff_t *ppos)
{
	return proc_watchdog_common(NMI_WATCHDOG_ENABLED|SOFT_WATCHDOG_ENABLED,
				    table, write, buffer, lenp, ppos);
}

/*
 * /proc/sys/kernel/nmi_watchdog
792
 */
793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808
int proc_nmi_watchdog(struct ctl_table *table, int write,
		      void __user *buffer, size_t *lenp, loff_t *ppos)
{
	return proc_watchdog_common(NMI_WATCHDOG_ENABLED,
				    table, write, buffer, lenp, ppos);
}

/*
 * /proc/sys/kernel/soft_watchdog
 */
int proc_soft_watchdog(struct ctl_table *table, int write,
			void __user *buffer, size_t *lenp, loff_t *ppos)
{
	return proc_watchdog_common(SOFT_WATCHDOG_ENABLED,
				    table, write, buffer, lenp, ppos);
}
809

810 811 812 813 814
/*
 * /proc/sys/kernel/watchdog_thresh
 */
int proc_watchdog_thresh(struct ctl_table *table, int write,
			 void __user *buffer, size_t *lenp, loff_t *ppos)
815
{
816
	int err, old, new;
817

818
	get_online_cpus();
819
	mutex_lock(&watchdog_proc_mutex);
820

821 822 823 824 825 826
	if (watchdog_suspended) {
		/* no parameter changes allowed while watchdog is suspended */
		err = -EAGAIN;
		goto out;
	}

827
	old = ACCESS_ONCE(watchdog_thresh);
828
	err = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
829

830
	if (err || !write)
831
		goto out;
832

833
	/*
834
	 * Update the sample period. Restore on failure.
835
	 */
836 837 838 839
	new = ACCESS_ONCE(watchdog_thresh);
	if (old == new)
		goto out;

840 841
	set_sample_period();
	err = proc_watchdog_update();
842
	if (err) {
843
		watchdog_thresh = old;
844 845
		set_sample_period();
	}
846 847
out:
	mutex_unlock(&watchdog_proc_mutex);
848
	put_online_cpus();
849
	return err;
850
}
851 852 853 854 855 856 857 858 859 860 861 862

/*
 * The cpumask is the mask of possible cpus that the watchdog can run
 * on, not the mask of cpus it is actually running on.  This allows the
 * user to specify a mask that will include cpus that have not yet
 * been brought online, if desired.
 */
int proc_watchdog_cpumask(struct ctl_table *table, int write,
			  void __user *buffer, size_t *lenp, loff_t *ppos)
{
	int err;

863
	get_online_cpus();
864
	mutex_lock(&watchdog_proc_mutex);
865 866 867 868 869 870 871

	if (watchdog_suspended) {
		/* no parameter changes allowed while watchdog is suspended */
		err = -EAGAIN;
		goto out;
	}

872 873 874 875 876 877 878 879 880 881 882 883
	err = proc_do_large_bitmap(table, write, buffer, lenp, ppos);
	if (!err && write) {
		/* Remove impossible cpus to keep sysctl output cleaner. */
		cpumask_and(&watchdog_cpumask, &watchdog_cpumask,
			    cpu_possible_mask);

		if (watchdog_running) {
			/*
			 * Failure would be due to being unable to allocate
			 * a temporary cpumask, so we are likely not in a
			 * position to do much else to make things better.
			 */
884
#ifdef CONFIG_SOFTLOCKUP_DETECTOR
885 886 887
			if (smpboot_update_cpumask_percpu_thread(
				    &watchdog_threads, &watchdog_cpumask) != 0)
				pr_err("cpumask update failed\n");
888
#endif
889 890
		}
	}
891
out:
892
	mutex_unlock(&watchdog_proc_mutex);
893
	put_online_cpus();
894 895 896
	return err;
}

897 898
#endif /* CONFIG_SYSCTL */

899
void __init lockup_detector_init(void)
900
{
901
	set_sample_period();
902

903 904
#ifdef CONFIG_NO_HZ_FULL
	if (tick_nohz_full_enabled()) {
905 906
		pr_info("Disabling watchdog on nohz_full cores by default\n");
		cpumask_copy(&watchdog_cpumask, housekeeping_mask);
907 908 909 910 911 912
	} else
		cpumask_copy(&watchdog_cpumask, cpu_possible_mask);
#else
	cpumask_copy(&watchdog_cpumask, cpu_possible_mask);
#endif

913
	if (watchdog_enabled)
914
		watchdog_enable_all_cpus();
915
}