nmi.c 17.8 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
/*
 *  linux/arch/i386/nmi.c
 *
 *  NMI watchdog support on APIC systems
 *
 *  Started by Ingo Molnar <mingo@redhat.com>
 *
 *  Fixes:
 *  Mikael Pettersson	: AMD K7 support for local APIC NMI watchdog.
 *  Mikael Pettersson	: Power Management for local APIC NMI watchdog.
 *  Mikael Pettersson	: Pentium 4 support for local APIC NMI watchdog.
 *  Pavel Machek and
 *  Mikael Pettersson	: PM converted to driver model. Disable/enable API.
 */

#include <linux/config.h>
#include <linux/delay.h>
#include <linux/interrupt.h>
#include <linux/module.h>
#include <linux/nmi.h>
#include <linux/sysdev.h>
#include <linux/sysctl.h>
23
#include <linux/percpu.h>
L
Linus Torvalds 已提交
24 25 26 27 28 29 30 31 32 33 34 35 36

#include <asm/smp.h>
#include <asm/nmi.h>

#include "mach_traps.h"

unsigned int nmi_watchdog = NMI_NONE;
extern int unknown_nmi_panic;
static unsigned int nmi_hz = HZ;
static unsigned int nmi_perfctr_msr;	/* the MSR to reset in NMI handler */
static unsigned int nmi_p4_cccr_val;
extern void show_registers(struct pt_regs *regs);

37 38 39 40 41 42 43 44 45 46 47 48 49 50
/* perfctr_nmi_owner tracks the ownership of the perfctr registers:
 * evtsel_nmi_owner tracks the ownership of the event selection
 * - different performance counters/ event selection may be reserved for
 *   different subsystems this reservation system just tries to coordinate
 *   things a little
 */
static DEFINE_PER_CPU(unsigned long, perfctr_nmi_owner);
static DEFINE_PER_CPU(unsigned long, evntsel_nmi_owner[3]);

/* this number is calculated from Intel's MSR_P4_CRU_ESCR5 register and it's
 * offset from MSR_P4_BSU_ESCR0.  It will be the max for all platforms (for now)
 */
#define NMI_MAX_COUNTER_BITS 66

L
Linus Torvalds 已提交
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 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111
/*
 * lapic_nmi_owner tracks the ownership of the lapic NMI hardware:
 * - it may be reserved by some other driver, or not
 * - when not reserved by some other driver, it may be used for
 *   the NMI watchdog, or not
 *
 * This is maintained separately from nmi_active because the NMI
 * watchdog may also be driven from the I/O APIC timer.
 */
static DEFINE_SPINLOCK(lapic_nmi_owner_lock);
static unsigned int lapic_nmi_owner;
#define LAPIC_NMI_WATCHDOG	(1<<0)
#define LAPIC_NMI_RESERVED	(1<<1)

/* nmi_active:
 * +1: the lapic NMI watchdog is active, but can be disabled
 *  0: the lapic NMI watchdog has not been set up, and cannot
 *     be enabled
 * -1: the lapic NMI watchdog is disabled, but can be enabled
 */
int nmi_active;

#define K7_EVNTSEL_ENABLE	(1 << 22)
#define K7_EVNTSEL_INT		(1 << 20)
#define K7_EVNTSEL_OS		(1 << 17)
#define K7_EVNTSEL_USR		(1 << 16)
#define K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING	0x76
#define K7_NMI_EVENT		K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING

#define P6_EVNTSEL0_ENABLE	(1 << 22)
#define P6_EVNTSEL_INT		(1 << 20)
#define P6_EVNTSEL_OS		(1 << 17)
#define P6_EVNTSEL_USR		(1 << 16)
#define P6_EVENT_CPU_CLOCKS_NOT_HALTED	0x79
#define P6_NMI_EVENT		P6_EVENT_CPU_CLOCKS_NOT_HALTED

#define MSR_P4_MISC_ENABLE	0x1A0
#define MSR_P4_MISC_ENABLE_PERF_AVAIL	(1<<7)
#define MSR_P4_MISC_ENABLE_PEBS_UNAVAIL	(1<<12)
#define MSR_P4_PERFCTR0		0x300
#define MSR_P4_CCCR0		0x360
#define P4_ESCR_EVENT_SELECT(N)	((N)<<25)
#define P4_ESCR_OS		(1<<3)
#define P4_ESCR_USR		(1<<2)
#define P4_CCCR_OVF_PMI0	(1<<26)
#define P4_CCCR_OVF_PMI1	(1<<27)
#define P4_CCCR_THRESHOLD(N)	((N)<<20)
#define P4_CCCR_COMPLEMENT	(1<<19)
#define P4_CCCR_COMPARE		(1<<18)
#define P4_CCCR_REQUIRED	(3<<16)
#define P4_CCCR_ESCR_SELECT(N)	((N)<<13)
#define P4_CCCR_ENABLE		(1<<12)
/* Set up IQ_COUNTER0 to behave like a clock, by having IQ_CCCR0 filter
   CRU_ESCR0 (with any non-null event selector) through a complemented
   max threshold. [IA32-Vol3, Section 14.9.9] */
#define MSR_P4_IQ_COUNTER0	0x30C
#define P4_NMI_CRU_ESCR0	(P4_ESCR_EVENT_SELECT(0x3F)|P4_ESCR_OS|P4_ESCR_USR)
#define P4_NMI_IQ_CCCR0	\
	(P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT|	\
	 P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE)

112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210
/* converts an msr to an appropriate reservation bit */
static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr)
{
	/* returns the bit offset of the performance counter register */
	switch (boot_cpu_data.x86_vendor) {
	case X86_VENDOR_AMD:
		return (msr - MSR_K7_PERFCTR0);
	case X86_VENDOR_INTEL:
		switch (boot_cpu_data.x86) {
		case 6:
			return (msr - MSR_P6_PERFCTR0);
		case 15:
			return (msr - MSR_P4_BPU_PERFCTR0);
		}
	}
	return 0;
}

/* converts an msr to an appropriate reservation bit */
static inline unsigned int nmi_evntsel_msr_to_bit(unsigned int msr)
{
	/* returns the bit offset of the event selection register */
	switch (boot_cpu_data.x86_vendor) {
	case X86_VENDOR_AMD:
		return (msr - MSR_K7_EVNTSEL0);
	case X86_VENDOR_INTEL:
		switch (boot_cpu_data.x86) {
		case 6:
			return (msr - MSR_P6_EVNTSEL0);
		case 15:
			return (msr - MSR_P4_BSU_ESCR0);
		}
	}
	return 0;
}

/* checks for a bit availability (hack for oprofile) */
int avail_to_resrv_perfctr_nmi_bit(unsigned int counter)
{
	BUG_ON(counter > NMI_MAX_COUNTER_BITS);

	return (!test_bit(counter, &__get_cpu_var(perfctr_nmi_owner)));
}

/* checks the an msr for availability */
int avail_to_resrv_perfctr_nmi(unsigned int msr)
{
	unsigned int counter;

	counter = nmi_perfctr_msr_to_bit(msr);
	BUG_ON(counter > NMI_MAX_COUNTER_BITS);

	return (!test_bit(counter, &__get_cpu_var(perfctr_nmi_owner)));
}

int reserve_perfctr_nmi(unsigned int msr)
{
	unsigned int counter;

	counter = nmi_perfctr_msr_to_bit(msr);
	BUG_ON(counter > NMI_MAX_COUNTER_BITS);

	if (!test_and_set_bit(counter, &__get_cpu_var(perfctr_nmi_owner)))
		return 1;
	return 0;
}

void release_perfctr_nmi(unsigned int msr)
{
	unsigned int counter;

	counter = nmi_perfctr_msr_to_bit(msr);
	BUG_ON(counter > NMI_MAX_COUNTER_BITS);

	clear_bit(counter, &__get_cpu_var(perfctr_nmi_owner));
}

int reserve_evntsel_nmi(unsigned int msr)
{
	unsigned int counter;

	counter = nmi_evntsel_msr_to_bit(msr);
	BUG_ON(counter > NMI_MAX_COUNTER_BITS);

	if (!test_and_set_bit(counter, &__get_cpu_var(evntsel_nmi_owner)[0]))
		return 1;
	return 0;
}

void release_evntsel_nmi(unsigned int msr)
{
	unsigned int counter;

	counter = nmi_evntsel_msr_to_bit(msr);
	BUG_ON(counter > NMI_MAX_COUNTER_BITS);

	clear_bit(counter, &__get_cpu_var(evntsel_nmi_owner)[0]);
}

211 212 213 214 215 216 217 218
#ifdef CONFIG_SMP
/* The performance counters used by NMI_LOCAL_APIC don't trigger when
 * the CPU is idle. To make sure the NMI watchdog really ticks on all
 * CPUs during the test make them busy.
 */
static __init void nmi_cpu_busy(void *data)
{
	volatile int *endflag = data;
219
	local_irq_enable_in_hardirq();
220 221 222 223 224 225 226 227 228 229 230
	/* Intentionally don't use cpu_relax here. This is
	   to make sure that the performance counter really ticks,
	   even if there is a simulator or similar that catches the
	   pause instruction. On a real HT machine this is fine because
	   all other CPUs are busy with "useless" delay loops and don't
	   care if they get somewhat less cycles. */
	while (*endflag == 0)
		barrier();
}
#endif

231
static int __init check_nmi_watchdog(void)
L
Linus Torvalds 已提交
232
{
233 234
	volatile int endflag = 0;
	unsigned int *prev_nmi_count;
L
Linus Torvalds 已提交
235 236
	int cpu;

237 238 239
	if (nmi_watchdog == NMI_NONE)
		return 0;

240 241 242 243
	prev_nmi_count = kmalloc(NR_CPUS * sizeof(int), GFP_KERNEL);
	if (!prev_nmi_count)
		return -1;

244
	printk(KERN_INFO "Testing NMI watchdog ... ");
L
Linus Torvalds 已提交
245

246 247 248
	if (nmi_watchdog == NMI_LOCAL_APIC)
		smp_call_function(nmi_cpu_busy, (void *)&endflag, 0, 0);

249
	for_each_possible_cpu(cpu)
L
Linus Torvalds 已提交
250 251 252 253
		prev_nmi_count[cpu] = per_cpu(irq_stat, cpu).__nmi_count;
	local_irq_enable();
	mdelay((10*1000)/nmi_hz); // wait 10 ticks

254
	for_each_possible_cpu(cpu) {
L
Linus Torvalds 已提交
255 256 257 258 259 260 261
#ifdef CONFIG_SMP
		/* Check cpu_callin_map here because that is set
		   after the timer is started. */
		if (!cpu_isset(cpu, cpu_callin_map))
			continue;
#endif
		if (nmi_count(cpu) - prev_nmi_count[cpu] <= 5) {
262 263 264 265 266
			endflag = 1;
			printk("CPU#%d: NMI appears to be stuck (%d->%d)!\n",
				cpu,
				prev_nmi_count[cpu],
				nmi_count(cpu));
L
Linus Torvalds 已提交
267 268
			nmi_active = 0;
			lapic_nmi_owner &= ~LAPIC_NMI_WATCHDOG;
269
			kfree(prev_nmi_count);
L
Linus Torvalds 已提交
270 271 272
			return -1;
		}
	}
273
	endflag = 1;
L
Linus Torvalds 已提交
274 275 276 277 278 279 280
	printk("OK.\n");

	/* now that we know it works we can reduce NMI frequency to
	   something more reasonable; makes a difference in some configs */
	if (nmi_watchdog == NMI_LOCAL_APIC)
		nmi_hz = 1;

281
	kfree(prev_nmi_count);
L
Linus Torvalds 已提交
282 283
	return 0;
}
284 285
/* This needs to happen later in boot so counters are working */
late_initcall(check_nmi_watchdog);
L
Linus Torvalds 已提交
286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339

static int __init setup_nmi_watchdog(char *str)
{
	int nmi;

	get_option(&str, &nmi);

	if (nmi >= NMI_INVALID)
		return 0;
	if (nmi == NMI_NONE)
		nmi_watchdog = nmi;
	/*
	 * If any other x86 CPU has a local APIC, then
	 * please test the NMI stuff there and send me the
	 * missing bits. Right now Intel P6/P4 and AMD K7 only.
	 */
	if ((nmi == NMI_LOCAL_APIC) &&
			(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) &&
			(boot_cpu_data.x86 == 6 || boot_cpu_data.x86 == 15))
		nmi_watchdog = nmi;
	if ((nmi == NMI_LOCAL_APIC) &&
			(boot_cpu_data.x86_vendor == X86_VENDOR_AMD) &&
	  		(boot_cpu_data.x86 == 6 || boot_cpu_data.x86 == 15))
		nmi_watchdog = nmi;
	/*
	 * We can enable the IO-APIC watchdog
	 * unconditionally.
	 */
	if (nmi == NMI_IO_APIC) {
		nmi_active = 1;
		nmi_watchdog = nmi;
	}
	return 1;
}

__setup("nmi_watchdog=", setup_nmi_watchdog);

static void disable_lapic_nmi_watchdog(void)
{
	if (nmi_active <= 0)
		return;
	switch (boot_cpu_data.x86_vendor) {
	case X86_VENDOR_AMD:
		wrmsr(MSR_K7_EVNTSEL0, 0, 0);
		break;
	case X86_VENDOR_INTEL:
		switch (boot_cpu_data.x86) {
		case 6:
			if (boot_cpu_data.x86_model > 0xd)
				break;

			wrmsr(MSR_P6_EVNTSEL0, 0, 0);
			break;
		case 15:
340
			if (boot_cpu_data.x86_model > 0x4)
L
Linus Torvalds 已提交
341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411
				break;

			wrmsr(MSR_P4_IQ_CCCR0, 0, 0);
			wrmsr(MSR_P4_CRU_ESCR0, 0, 0);
			break;
		}
		break;
	}
	nmi_active = -1;
	/* tell do_nmi() and others that we're not active any more */
	nmi_watchdog = 0;
}

static void enable_lapic_nmi_watchdog(void)
{
	if (nmi_active < 0) {
		nmi_watchdog = NMI_LOCAL_APIC;
		setup_apic_nmi_watchdog();
	}
}

int reserve_lapic_nmi(void)
{
	unsigned int old_owner;

	spin_lock(&lapic_nmi_owner_lock);
	old_owner = lapic_nmi_owner;
	lapic_nmi_owner |= LAPIC_NMI_RESERVED;
	spin_unlock(&lapic_nmi_owner_lock);
	if (old_owner & LAPIC_NMI_RESERVED)
		return -EBUSY;
	if (old_owner & LAPIC_NMI_WATCHDOG)
		disable_lapic_nmi_watchdog();
	return 0;
}

void release_lapic_nmi(void)
{
	unsigned int new_owner;

	spin_lock(&lapic_nmi_owner_lock);
	new_owner = lapic_nmi_owner & ~LAPIC_NMI_RESERVED;
	lapic_nmi_owner = new_owner;
	spin_unlock(&lapic_nmi_owner_lock);
	if (new_owner & LAPIC_NMI_WATCHDOG)
		enable_lapic_nmi_watchdog();
}

void disable_timer_nmi_watchdog(void)
{
	if ((nmi_watchdog != NMI_IO_APIC) || (nmi_active <= 0))
		return;

	unset_nmi_callback();
	nmi_active = -1;
	nmi_watchdog = NMI_NONE;
}

void enable_timer_nmi_watchdog(void)
{
	if (nmi_active < 0) {
		nmi_watchdog = NMI_IO_APIC;
		touch_nmi_watchdog();
		nmi_active = 1;
	}
}

#ifdef CONFIG_PM

static int nmi_pm_active; /* nmi_active before suspend */

412
static int lapic_nmi_suspend(struct sys_device *dev, pm_message_t state)
L
Linus Torvalds 已提交
413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459
{
	nmi_pm_active = nmi_active;
	disable_lapic_nmi_watchdog();
	return 0;
}

static int lapic_nmi_resume(struct sys_device *dev)
{
	if (nmi_pm_active > 0)
		enable_lapic_nmi_watchdog();
	return 0;
}


static struct sysdev_class nmi_sysclass = {
	set_kset_name("lapic_nmi"),
	.resume		= lapic_nmi_resume,
	.suspend	= lapic_nmi_suspend,
};

static struct sys_device device_lapic_nmi = {
	.id	= 0,
	.cls	= &nmi_sysclass,
};

static int __init init_lapic_nmi_sysfs(void)
{
	int error;

	if (nmi_active == 0 || nmi_watchdog != NMI_LOCAL_APIC)
		return 0;

	error = sysdev_class_register(&nmi_sysclass);
	if (!error)
		error = sysdev_register(&device_lapic_nmi);
	return error;
}
/* must come after the local APIC's device_initcall() */
late_initcall(init_lapic_nmi_sysfs);

#endif	/* CONFIG_PM */

/*
 * Activate the NMI watchdog via the local APIC.
 * Original code written by Keith Owens.
 */

460
static void write_watchdog_counter(const char *descr)
461 462 463 464 465 466 467 468 469
{
	u64 count = (u64)cpu_khz * 1000;

	do_div(count, nmi_hz);
	if(descr)
		Dprintk("setting %s to -0x%08Lx\n", descr, count);
	wrmsrl(nmi_perfctr_msr, 0 - count);
}

470
static int setup_k7_watchdog(void)
L
Linus Torvalds 已提交
471 472 473 474 475
{
	unsigned int evntsel;

	nmi_perfctr_msr = MSR_K7_PERFCTR0;

476 477 478 479 480 481 482
	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
		goto fail;

	if (!reserve_evntsel_nmi(MSR_K7_EVNTSEL0))
		goto fail1;

	wrmsrl(MSR_K7_PERFCTR0, 0UL);
L
Linus Torvalds 已提交
483 484 485 486 487 488 489

	evntsel = K7_EVNTSEL_INT
		| K7_EVNTSEL_OS
		| K7_EVNTSEL_USR
		| K7_NMI_EVENT;

	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
490
	write_watchdog_counter("K7_PERFCTR0");
L
Linus Torvalds 已提交
491 492 493
	apic_write(APIC_LVTPC, APIC_DM_NMI);
	evntsel |= K7_EVNTSEL_ENABLE;
	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
494 495 496 497 498
	return 1;
fail1:
	release_perfctr_nmi(nmi_perfctr_msr);
fail:
	return 0;
L
Linus Torvalds 已提交
499 500
}

501
static int setup_p6_watchdog(void)
L
Linus Torvalds 已提交
502 503 504 505 506
{
	unsigned int evntsel;

	nmi_perfctr_msr = MSR_P6_PERFCTR0;

507 508 509 510 511
	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
		goto fail;

	if (!reserve_evntsel_nmi(MSR_P6_EVNTSEL0))
		goto fail1;
L
Linus Torvalds 已提交
512 513 514 515 516 517 518

	evntsel = P6_EVNTSEL_INT
		| P6_EVNTSEL_OS
		| P6_EVNTSEL_USR
		| P6_NMI_EVENT;

	wrmsr(MSR_P6_EVNTSEL0, evntsel, 0);
519
	write_watchdog_counter("P6_PERFCTR0");
L
Linus Torvalds 已提交
520 521 522
	apic_write(APIC_LVTPC, APIC_DM_NMI);
	evntsel |= P6_EVNTSEL0_ENABLE;
	wrmsr(MSR_P6_EVNTSEL0, evntsel, 0);
523 524 525 526 527
	return 1;
fail1:
	release_perfctr_nmi(nmi_perfctr_msr);
fail:
	return 0;
L
Linus Torvalds 已提交
528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544
}

static int setup_p4_watchdog(void)
{
	unsigned int misc_enable, dummy;

	rdmsr(MSR_P4_MISC_ENABLE, misc_enable, dummy);
	if (!(misc_enable & MSR_P4_MISC_ENABLE_PERF_AVAIL))
		return 0;

	nmi_perfctr_msr = MSR_P4_IQ_COUNTER0;
	nmi_p4_cccr_val = P4_NMI_IQ_CCCR0;
#ifdef CONFIG_SMP
	if (smp_num_siblings == 2)
		nmi_p4_cccr_val |= P4_CCCR_OVF_PMI1;
#endif

545 546 547 548 549
	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
		goto fail;

	if (!reserve_evntsel_nmi(MSR_P4_CRU_ESCR0))
		goto fail1;
L
Linus Torvalds 已提交
550 551 552

	wrmsr(MSR_P4_CRU_ESCR0, P4_NMI_CRU_ESCR0, 0);
	wrmsr(MSR_P4_IQ_CCCR0, P4_NMI_IQ_CCCR0 & ~P4_CCCR_ENABLE, 0);
553
	write_watchdog_counter("P4_IQ_COUNTER0");
L
Linus Torvalds 已提交
554 555 556
	apic_write(APIC_LVTPC, APIC_DM_NMI);
	wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
	return 1;
557 558 559 560
fail1:
	release_perfctr_nmi(nmi_perfctr_msr);
fail:
	return 0;
L
Linus Torvalds 已提交
561 562 563 564 565 566 567 568
}

void setup_apic_nmi_watchdog (void)
{
	switch (boot_cpu_data.x86_vendor) {
	case X86_VENDOR_AMD:
		if (boot_cpu_data.x86 != 6 && boot_cpu_data.x86 != 15)
			return;
569 570
		if (!setup_k7_watchdog())
			return;
L
Linus Torvalds 已提交
571 572 573 574 575 576 577
		break;
	case X86_VENDOR_INTEL:
		switch (boot_cpu_data.x86) {
		case 6:
			if (boot_cpu_data.x86_model > 0xd)
				return;

578 579
			if(!setup_p6_watchdog())
				return;
L
Linus Torvalds 已提交
580 581
			break;
		case 15:
582
			if (boot_cpu_data.x86_model > 0x4)
L
Linus Torvalds 已提交
583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625
				return;

			if (!setup_p4_watchdog())
				return;
			break;
		default:
			return;
		}
		break;
	default:
		return;
	}
	lapic_nmi_owner = LAPIC_NMI_WATCHDOG;
	nmi_active = 1;
}

/*
 * the best way to detect whether a CPU has a 'hard lockup' problem
 * is to check it's local APIC timer IRQ counts. If they are not
 * changing then that CPU has some problem.
 *
 * as these watchdog NMI IRQs are generated on every CPU, we only
 * have to check the current processor.
 *
 * since NMIs don't listen to _any_ locks, we have to be extremely
 * careful not to rely on unsafe variables. The printk might lock
 * up though, so we have to break up any console locks first ...
 * [when there will be more tty-related locks, break them up
 *  here too!]
 */

static unsigned int
	last_irq_sums [NR_CPUS],
	alert_counter [NR_CPUS];

void touch_nmi_watchdog (void)
{
	int i;

	/*
	 * Just reset the alert counters, (other CPUs might be
	 * spinning on locks we hold):
	 */
626
	for_each_possible_cpu(i)
L
Linus Torvalds 已提交
627
		alert_counter[i] = 0;
I
Ingo Molnar 已提交
628 629 630 631 632

	/*
	 * Tickle the softlockup detector too:
	 */
	touch_softlockup_watchdog();
L
Linus Torvalds 已提交
633
}
634
EXPORT_SYMBOL(touch_nmi_watchdog);
L
Linus Torvalds 已提交
635 636 637 638 639 640 641 642 643 644 645

extern void die_nmi(struct pt_regs *, const char *msg);

void nmi_watchdog_tick (struct pt_regs * regs)
{

	/*
	 * Since current_thread_info()-> is always on the stack, and we
	 * always switch the stack NMI-atomically, it's safe to use
	 * smp_processor_id().
	 */
646 647
	unsigned int sum;
	int cpu = smp_processor_id();
L
Linus Torvalds 已提交
648 649 650 651 652 653 654 655 656 657

	sum = per_cpu(irq_stat, cpu).apic_timer_irqs;

	if (last_irq_sums[cpu] == sum) {
		/*
		 * Ayiee, looks like this CPU is stuck ...
		 * wait a few IRQs (5 seconds) before doing the oops ...
		 */
		alert_counter[cpu]++;
		if (alert_counter[cpu] == 5*nmi_hz)
658 659 660
			/*
			 * die_nmi will return ONLY if NOTIFY_STOP happens..
			 */
661
			die_nmi(regs, "BUG: NMI Watchdog detected LOCKUP");
662
	} else {
L
Linus Torvalds 已提交
663 664 665 666 667 668 669 670 671 672 673 674 675 676 677
		last_irq_sums[cpu] = sum;
		alert_counter[cpu] = 0;
	}
	if (nmi_perfctr_msr) {
		if (nmi_perfctr_msr == MSR_P4_IQ_COUNTER0) {
			/*
			 * P4 quirks:
			 * - An overflown perfctr will assert its interrupt
			 *   until the OVF flag in its CCCR is cleared.
			 * - LVTPC is masked on interrupt and must be
			 *   unmasked by the LVTPC handler.
			 */
			wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
			apic_write(APIC_LVTPC, APIC_DM_NMI);
		}
678
		else if (nmi_perfctr_msr == MSR_P6_PERFCTR0) {
L
Linus Torvalds 已提交
679 680 681 682 683
			/* Only P6 based Pentium M need to re-unmask
			 * the apic vector but it doesn't hurt
			 * other P6 variant */
			apic_write(APIC_LVTPC, APIC_DM_NMI);
		}
684
		write_watchdog_counter(NULL);
L
Linus Torvalds 已提交
685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732
	}
}

#ifdef CONFIG_SYSCTL

static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu)
{
	unsigned char reason = get_nmi_reason();
	char buf[64];

	if (!(reason & 0xc0)) {
		sprintf(buf, "NMI received for unknown reason %02x\n", reason);
		die_nmi(regs, buf);
	}
	return 0;
}

/*
 * proc handler for /proc/sys/kernel/unknown_nmi_panic
 */
int proc_unknown_nmi_panic(ctl_table *table, int write, struct file *file,
			void __user *buffer, size_t *length, loff_t *ppos)
{
	int old_state;

	old_state = unknown_nmi_panic;
	proc_dointvec(table, write, file, buffer, length, ppos);
	if (!!old_state == !!unknown_nmi_panic)
		return 0;

	if (unknown_nmi_panic) {
		if (reserve_lapic_nmi() < 0) {
			unknown_nmi_panic = 0;
			return -EBUSY;
		} else {
			set_nmi_callback(unknown_nmi_panic_callback);
		}
	} else {
		release_lapic_nmi();
		unset_nmi_callback();
	}
	return 0;
}

#endif

EXPORT_SYMBOL(nmi_active);
EXPORT_SYMBOL(nmi_watchdog);
733 734 735 736 737 738
EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi);
EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi_bit);
EXPORT_SYMBOL(reserve_perfctr_nmi);
EXPORT_SYMBOL(release_perfctr_nmi);
EXPORT_SYMBOL(reserve_evntsel_nmi);
EXPORT_SYMBOL(release_evntsel_nmi);
L
Linus Torvalds 已提交
739 740 741 742
EXPORT_SYMBOL(reserve_lapic_nmi);
EXPORT_SYMBOL(release_lapic_nmi);
EXPORT_SYMBOL(disable_timer_nmi_watchdog);
EXPORT_SYMBOL(enable_timer_nmi_watchdog);