hpet.c 27.3 KB
Newer Older
1
#include <linux/clocksource.h>
2
#include <linux/clockchips.h>
3 4
#include <linux/interrupt.h>
#include <linux/sysdev.h>
5
#include <linux/delay.h>
6
#include <linux/errno.h>
7
#include <linux/i8253.h>
8
#include <linux/slab.h>
9 10
#include <linux/hpet.h>
#include <linux/init.h>
11
#include <linux/cpu.h>
12 13
#include <linux/pm.h>
#include <linux/io.h>
14

15
#include <asm/fixmap.h>
16
#include <asm/hpet.h>
17

18
#define HPET_MASK			CLOCKSOURCE_MASK(32)
19

P
Pavel Machek 已提交
20 21
/* FSEC = 10^-15
   NSEC = 10^-9 */
22
#define FSEC_PER_NSEC			1000000L
23

24 25 26 27 28 29
#define HPET_DEV_USED_BIT		2
#define HPET_DEV_USED			(1 << HPET_DEV_USED_BIT)
#define HPET_DEV_VALID			0x8
#define HPET_DEV_FSB_CAP		0x1000
#define HPET_DEV_PERI_CAP		0x2000

30 31 32
#define HPET_MIN_CYCLES			128
#define HPET_MIN_PROG_DELTA		(HPET_MIN_CYCLES + (HPET_MIN_CYCLES >> 1))

33 34
#define EVT_TO_HPET_DEV(evt) container_of(evt, struct hpet_dev, evt)

35 36 37
/*
 * HPET address is set in acpi/boot.c, when an ACPI entry exists
 */
38
unsigned long				hpet_address;
39
u8					hpet_blockid; /* OS timer block num */
40 41
u8					hpet_msi_disable;

42
#ifdef CONFIG_PCI_MSI
H
Hannes Eder 已提交
43
static unsigned long			hpet_num_timers;
44
#endif
45
static void __iomem			*hpet_virt_address;
46

47
struct hpet_dev {
48 49 50 51 52 53
	struct clock_event_device	evt;
	unsigned int			num;
	int				cpu;
	unsigned int			irq;
	unsigned int			flags;
	char				name[10];
54 55
};

J
Jan Beulich 已提交
56
inline unsigned int hpet_readl(unsigned int a)
57 58 59 60
{
	return readl(hpet_virt_address + a);
}

J
Jan Beulich 已提交
61
static inline void hpet_writel(unsigned int d, unsigned int a)
62 63 64 65
{
	writel(d, hpet_virt_address + a);
}

66 67
#ifdef CONFIG_X86_64
#include <asm/pgtable.h>
68
#endif
69

70 71 72
static inline void hpet_set_mapping(void)
{
	hpet_virt_address = ioremap_nocache(hpet_address, HPET_MMAP_SIZE);
73 74 75
#ifdef CONFIG_X86_64
	__set_fixmap(VSYSCALL_HPET, hpet_address, PAGE_KERNEL_VSYSCALL_NOCACHE);
#endif
76 77 78 79 80 81 82 83
}

static inline void hpet_clear_mapping(void)
{
	iounmap(hpet_virt_address);
	hpet_virt_address = NULL;
}

84 85 86 87
/*
 * HPET command line enable / disable
 */
static int boot_hpet_disable;
T
Thomas Gleixner 已提交
88
int hpet_force_user;
89
static int hpet_verbose;
90

91
static int __init hpet_setup(char *str)
92 93 94 95
{
	if (str) {
		if (!strncmp("disable", str, 7))
			boot_hpet_disable = 1;
T
Thomas Gleixner 已提交
96 97
		if (!strncmp("force", str, 5))
			hpet_force_user = 1;
98 99
		if (!strncmp("verbose", str, 7))
			hpet_verbose = 1;
100 101 102 103 104
	}
	return 1;
}
__setup("hpet=", hpet_setup);

105 106 107 108 109 110 111
static int __init disable_hpet(char *str)
{
	boot_hpet_disable = 1;
	return 1;
}
__setup("nohpet", disable_hpet);

112 113
static inline int is_hpet_capable(void)
{
114
	return !boot_hpet_disable && hpet_address;
115 116 117 118 119 120 121 122 123 124 125 126 127 128
}

/*
 * HPET timer interrupt enable / disable
 */
static int hpet_legacy_int_enabled;

/**
 * is_hpet_enabled - check whether the hpet timer interrupt is enabled
 */
int is_hpet_enabled(void)
{
	return is_hpet_capable() && hpet_legacy_int_enabled;
}
129
EXPORT_SYMBOL_GPL(is_hpet_enabled);
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
static void _hpet_print_config(const char *function, int line)
{
	u32 i, timers, l, h;
	printk(KERN_INFO "hpet: %s(%d):\n", function, line);
	l = hpet_readl(HPET_ID);
	h = hpet_readl(HPET_PERIOD);
	timers = ((l & HPET_ID_NUMBER) >> HPET_ID_NUMBER_SHIFT) + 1;
	printk(KERN_INFO "hpet: ID: 0x%x, PERIOD: 0x%x\n", l, h);
	l = hpet_readl(HPET_CFG);
	h = hpet_readl(HPET_STATUS);
	printk(KERN_INFO "hpet: CFG: 0x%x, STATUS: 0x%x\n", l, h);
	l = hpet_readl(HPET_COUNTER);
	h = hpet_readl(HPET_COUNTER+4);
	printk(KERN_INFO "hpet: COUNTER_l: 0x%x, COUNTER_h: 0x%x\n", l, h);

	for (i = 0; i < timers; i++) {
		l = hpet_readl(HPET_Tn_CFG(i));
		h = hpet_readl(HPET_Tn_CFG(i)+4);
		printk(KERN_INFO "hpet: T%d: CFG_l: 0x%x, CFG_h: 0x%x\n",
		       i, l, h);
		l = hpet_readl(HPET_Tn_CMP(i));
		h = hpet_readl(HPET_Tn_CMP(i)+4);
		printk(KERN_INFO "hpet: T%d: CMP_l: 0x%x, CMP_h: 0x%x\n",
		       i, l, h);
		l = hpet_readl(HPET_Tn_ROUTE(i));
		h = hpet_readl(HPET_Tn_ROUTE(i)+4);
		printk(KERN_INFO "hpet: T%d ROUTE_l: 0x%x, ROUTE_h: 0x%x\n",
		       i, l, h);
	}
}

#define hpet_print_config()					\
do {								\
	if (hpet_verbose)					\
		_hpet_print_config(__FUNCTION__, __LINE__);	\
} while (0)

168 169 170 171 172
/*
 * When the hpet driver (/dev/hpet) is enabled, we need to reserve
 * timer 0 and timer 1 in case of RTC emulation.
 */
#ifdef CONFIG_HPET
173

V
Venki Pallipadi 已提交
174
static void hpet_reserve_msi_timers(struct hpet_data *hd);
175

J
Jan Beulich 已提交
176
static void hpet_reserve_platform_timers(unsigned int id)
177 178
{
	struct hpet __iomem *hpet = hpet_virt_address;
179 180
	struct hpet_timer __iomem *timer = &hpet->hpet_timers[2];
	unsigned int nrtimers, i;
181 182 183 184
	struct hpet_data hd;

	nrtimers = ((id & HPET_ID_NUMBER) >> HPET_ID_NUMBER_SHIFT) + 1;

185 186 187 188
	memset(&hd, 0, sizeof(hd));
	hd.hd_phys_address	= hpet_address;
	hd.hd_address		= hpet;
	hd.hd_nirqs		= nrtimers;
189 190 191 192 193
	hpet_reserve_timer(&hd, 0);

#ifdef CONFIG_HPET_EMULATE_RTC
	hpet_reserve_timer(&hd, 1);
#endif
194

195 196 197 198 199
	/*
	 * NOTE that hd_irq[] reflects IOAPIC input pins (LEGACY_8254
	 * is wrong for i8259!) not the output IRQ.  Many BIOS writers
	 * don't bother configuring *any* comparator interrupts.
	 */
200 201 202
	hd.hd_irq[0] = HPET_LEGACY_8254;
	hd.hd_irq[1] = HPET_LEGACY_RTC;

I
Ingo Molnar 已提交
203
	for (i = 2; i < nrtimers; timer++, i++) {
204 205
		hd.hd_irq[i] = (readl(&timer->hpet_config) &
			Tn_INT_ROUTE_CNF_MASK) >> Tn_INT_ROUTE_CNF_SHIFT;
I
Ingo Molnar 已提交
206
	}
207

208
	hpet_reserve_msi_timers(&hd);
209

210
	hpet_alloc(&hd);
211

212 213
}
#else
J
Jan Beulich 已提交
214
static void hpet_reserve_platform_timers(unsigned int id) { }
215 216 217 218 219
#endif

/*
 * Common hpet info
 */
220
static unsigned long hpet_freq;
221

222
static void hpet_legacy_set_mode(enum clock_event_mode mode,
223
			  struct clock_event_device *evt);
224
static int hpet_legacy_next_event(unsigned long delta,
225 226 227 228 229 230 231 232
			   struct clock_event_device *evt);

/*
 * The hpet clock event device
 */
static struct clock_event_device hpet_clockevent = {
	.name		= "hpet",
	.features	= CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
233 234
	.set_mode	= hpet_legacy_set_mode,
	.set_next_event = hpet_legacy_next_event,
235
	.irq		= 0,
236
	.rating		= 50,
237 238
};

239
static void hpet_stop_counter(void)
240 241 242 243
{
	unsigned long cfg = hpet_readl(HPET_CFG);
	cfg &= ~HPET_CFG_ENABLE;
	hpet_writel(cfg, HPET_CFG);
244 245 246 247
}

static void hpet_reset_counter(void)
{
248 249
	hpet_writel(0, HPET_COUNTER);
	hpet_writel(0, HPET_COUNTER + 4);
250 251 252 253
}

static void hpet_start_counter(void)
{
J
Jan Beulich 已提交
254
	unsigned int cfg = hpet_readl(HPET_CFG);
255 256 257 258
	cfg |= HPET_CFG_ENABLE;
	hpet_writel(cfg, HPET_CFG);
}

259 260 261
static void hpet_restart_counter(void)
{
	hpet_stop_counter();
262
	hpet_reset_counter();
263 264 265
	hpet_start_counter();
}

266 267
static void hpet_resume_device(void)
{
V
Venki Pallipadi 已提交
268
	force_hpet_resume();
269 270
}

271
static void hpet_resume_counter(struct clocksource *cs)
272 273
{
	hpet_resume_device();
274
	hpet_restart_counter();
275 276
}

277
static void hpet_enable_legacy_int(void)
278
{
J
Jan Beulich 已提交
279
	unsigned int cfg = hpet_readl(HPET_CFG);
280 281 282 283 284 285

	cfg |= HPET_CFG_LEGACY;
	hpet_writel(cfg, HPET_CFG);
	hpet_legacy_int_enabled = 1;
}

286 287 288 289 290 291 292 293 294
static void hpet_legacy_clockevent_register(void)
{
	/* Start HPET legacy interrupts */
	hpet_enable_legacy_int();

	/*
	 * Start hpet with the boot cpu mask and make it
	 * global after the IO_APIC has been initialized.
	 */
295
	hpet_clockevent.cpumask = cpumask_of(smp_processor_id());
296 297
	clockevents_config_and_register(&hpet_clockevent, hpet_freq,
					HPET_MIN_PROG_DELTA, 0x7FFFFFFF);
298 299 300 301
	global_clock_event = &hpet_clockevent;
	printk(KERN_DEBUG "hpet clockevent registered\n");
}

302 303
static int hpet_setup_msi_irq(unsigned int irq);

304 305
static void hpet_set_mode(enum clock_event_mode mode,
			  struct clock_event_device *evt, int timer)
306
{
J
Jan Beulich 已提交
307
	unsigned int cfg, cmp, now;
308 309
	uint64_t delta;

310
	switch (mode) {
311
	case CLOCK_EVT_MODE_PERIODIC:
312
		hpet_stop_counter();
313 314
		delta = ((uint64_t)(NSEC_PER_SEC/HZ)) * evt->mult;
		delta >>= evt->shift;
315
		now = hpet_readl(HPET_COUNTER);
J
Jan Beulich 已提交
316
		cmp = now + (unsigned int) delta;
317
		cfg = hpet_readl(HPET_Tn_CFG(timer));
318 319
		/* Make sure we use edge triggered interrupts */
		cfg &= ~HPET_TN_LEVEL;
320 321
		cfg |= HPET_TN_ENABLE | HPET_TN_PERIODIC |
		       HPET_TN_SETVAL | HPET_TN_32BIT;
322
		hpet_writel(cfg, HPET_Tn_CFG(timer));
323 324 325 326 327 328 329 330 331
		hpet_writel(cmp, HPET_Tn_CMP(timer));
		udelay(1);
		/*
		 * HPET on AMD 81xx needs a second write (with HPET_TN_SETVAL
		 * cleared) to T0_CMP to set the period. The HPET_TN_SETVAL
		 * bit is automatically cleared after the first write.
		 * (See AMD-8111 HyperTransport I/O Hub Data Sheet,
		 * Publication # 24674)
		 */
J
Jan Beulich 已提交
332
		hpet_writel((unsigned int) delta, HPET_Tn_CMP(timer));
333
		hpet_start_counter();
334
		hpet_print_config();
335 336 337
		break;

	case CLOCK_EVT_MODE_ONESHOT:
338
		cfg = hpet_readl(HPET_Tn_CFG(timer));
339 340
		cfg &= ~HPET_TN_PERIODIC;
		cfg |= HPET_TN_ENABLE | HPET_TN_32BIT;
341
		hpet_writel(cfg, HPET_Tn_CFG(timer));
342 343 344 345
		break;

	case CLOCK_EVT_MODE_UNUSED:
	case CLOCK_EVT_MODE_SHUTDOWN:
346
		cfg = hpet_readl(HPET_Tn_CFG(timer));
347
		cfg &= ~HPET_TN_ENABLE;
348
		hpet_writel(cfg, HPET_Tn_CFG(timer));
349
		break;
T
Thomas Gleixner 已提交
350 351

	case CLOCK_EVT_MODE_RESUME:
352 353 354 355 356 357
		if (timer == 0) {
			hpet_enable_legacy_int();
		} else {
			struct hpet_dev *hdev = EVT_TO_HPET_DEV(evt);
			hpet_setup_msi_irq(hdev->irq);
			disable_irq(hdev->irq);
358
			irq_set_affinity(hdev->irq, cpumask_of(hdev->cpu));
359 360
			enable_irq(hdev->irq);
		}
361
		hpet_print_config();
T
Thomas Gleixner 已提交
362
		break;
363 364 365
	}
}

366 367
static int hpet_next_event(unsigned long delta,
			   struct clock_event_device *evt, int timer)
368
{
369
	u32 cnt;
370
	s32 res;
371 372

	cnt = hpet_readl(HPET_COUNTER);
373
	cnt += (u32) delta;
374
	hpet_writel(cnt, HPET_Tn_CMP(timer));
375

376
	/*
377 378 379 380 381 382
	 * HPETs are a complete disaster. The compare register is
	 * based on a equal comparison and neither provides a less
	 * than or equal functionality (which would require to take
	 * the wraparound into account) nor a simple count down event
	 * mode. Further the write to the comparator register is
	 * delayed internally up to two HPET clock cycles in certain
383 384 385 386 387 388
	 * chipsets (ATI, ICH9,10). Some newer AMD chipsets have even
	 * longer delays. We worked around that by reading back the
	 * compare register, but that required another workaround for
	 * ICH9,10 chips where the first readout after write can
	 * return the old stale value. We already had a minimum
	 * programming delta of 5us enforced, but a NMI or SMI hitting
389 390 391 392
	 * between the counter readout and the comparator write can
	 * move us behind that point easily. Now instead of reading
	 * the compare register back several times, we make the ETIME
	 * decision based on the following: Return ETIME if the
393
	 * counter value after the write is less than HPET_MIN_CYCLES
394
	 * away from the event or if the counter is already ahead of
395 396
	 * the event. The minimum programming delta for the generic
	 * clockevents code is set to 1.5 * HPET_MIN_CYCLES.
397
	 */
398
	res = (s32)(cnt - hpet_readl(HPET_COUNTER));
399

400
	return res < HPET_MIN_CYCLES ? -ETIME : 0;
401 402
}

403 404 405 406 407 408 409 410 411 412 413 414
static void hpet_legacy_set_mode(enum clock_event_mode mode,
			struct clock_event_device *evt)
{
	hpet_set_mode(mode, evt, 0);
}

static int hpet_legacy_next_event(unsigned long delta,
			struct clock_event_device *evt)
{
	return hpet_next_event(delta, evt, 0);
}

415 416 417
/*
 * HPET MSI Support
 */
418
#ifdef CONFIG_PCI_MSI
V
Venki Pallipadi 已提交
419 420 421 422

static DEFINE_PER_CPU(struct hpet_dev *, cpu_hpet_dev);
static struct hpet_dev	*hpet_devs;

423
void hpet_msi_unmask(struct irq_data *data)
424
{
425
	struct hpet_dev *hdev = data->handler_data;
J
Jan Beulich 已提交
426
	unsigned int cfg;
427 428 429 430 431 432 433

	/* unmask it */
	cfg = hpet_readl(HPET_Tn_CFG(hdev->num));
	cfg |= HPET_TN_FSB;
	hpet_writel(cfg, HPET_Tn_CFG(hdev->num));
}

434
void hpet_msi_mask(struct irq_data *data)
435
{
436
	struct hpet_dev *hdev = data->handler_data;
J
Jan Beulich 已提交
437
	unsigned int cfg;
438 439 440 441 442 443 444

	/* mask it */
	cfg = hpet_readl(HPET_Tn_CFG(hdev->num));
	cfg &= ~HPET_TN_FSB;
	hpet_writel(cfg, HPET_Tn_CFG(hdev->num));
}

445
void hpet_msi_write(struct hpet_dev *hdev, struct msi_msg *msg)
446 447 448 449 450
{
	hpet_writel(msg->data, HPET_Tn_ROUTE(hdev->num));
	hpet_writel(msg->address_lo, HPET_Tn_ROUTE(hdev->num) + 4);
}

451
void hpet_msi_read(struct hpet_dev *hdev, struct msi_msg *msg)
452 453 454 455 456 457
{
	msg->data = hpet_readl(HPET_Tn_ROUTE(hdev->num));
	msg->address_lo = hpet_readl(HPET_Tn_ROUTE(hdev->num) + 4);
	msg->address_hi = 0;
}

458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473
static void hpet_msi_set_mode(enum clock_event_mode mode,
				struct clock_event_device *evt)
{
	struct hpet_dev *hdev = EVT_TO_HPET_DEV(evt);
	hpet_set_mode(mode, evt, hdev->num);
}

static int hpet_msi_next_event(unsigned long delta,
				struct clock_event_device *evt)
{
	struct hpet_dev *hdev = EVT_TO_HPET_DEV(evt);
	return hpet_next_event(delta, evt, hdev->num);
}

static int hpet_setup_msi_irq(unsigned int irq)
{
474
	if (arch_setup_hpet_msi(irq, hpet_blockid)) {
475 476 477 478 479 480 481 482 483 484
		destroy_irq(irq);
		return -EINVAL;
	}
	return 0;
}

static int hpet_assign_irq(struct hpet_dev *dev)
{
	unsigned int irq;

485
	irq = create_irq_nr(0, -1);
486 487 488
	if (!irq)
		return -EINVAL;

489
	irq_set_handler_data(irq, dev);
490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516

	if (hpet_setup_msi_irq(irq))
		return -EINVAL;

	dev->irq = irq;
	return 0;
}

static irqreturn_t hpet_interrupt_handler(int irq, void *data)
{
	struct hpet_dev *dev = (struct hpet_dev *)data;
	struct clock_event_device *hevt = &dev->evt;

	if (!hevt->event_handler) {
		printk(KERN_INFO "Spurious HPET timer interrupt on HPET timer %d\n",
				dev->num);
		return IRQ_HANDLED;
	}

	hevt->event_handler(hevt);
	return IRQ_HANDLED;
}

static int hpet_setup_irq(struct hpet_dev *dev)
{

	if (request_irq(dev->irq, hpet_interrupt_handler,
517 518
			IRQF_TIMER | IRQF_DISABLED | IRQF_NOBALANCING,
			dev->name, dev))
519 520 521
		return -1;

	disable_irq(dev->irq);
522
	irq_set_affinity(dev->irq, cpumask_of(dev->cpu));
523 524
	enable_irq(dev->irq);

Y
Yinghai Lu 已提交
525 526 527
	printk(KERN_DEBUG "hpet: %s irq %d for MSI\n",
			 dev->name, dev->irq);

528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555
	return 0;
}

/* This should be called in specific @cpu */
static void init_one_hpet_msi_clockevent(struct hpet_dev *hdev, int cpu)
{
	struct clock_event_device *evt = &hdev->evt;

	WARN_ON(cpu != smp_processor_id());
	if (!(hdev->flags & HPET_DEV_VALID))
		return;

	if (hpet_setup_msi_irq(hdev->irq))
		return;

	hdev->cpu = cpu;
	per_cpu(cpu_hpet_dev, cpu) = hdev;
	evt->name = hdev->name;
	hpet_setup_irq(hdev);
	evt->irq = hdev->irq;

	evt->rating = 110;
	evt->features = CLOCK_EVT_FEAT_ONESHOT;
	if (hdev->flags & HPET_DEV_PERI_CAP)
		evt->features |= CLOCK_EVT_FEAT_PERIODIC;

	evt->set_mode = hpet_msi_set_mode;
	evt->set_next_event = hpet_msi_next_event;
556
	evt->cpumask = cpumask_of(hdev->cpu);
557 558 559

	clockevents_config_and_register(evt, hpet_freq, HPET_MIN_PROG_DELTA,
					0x7FFFFFFF);
560 561 562 563 564 565 566 567
}

#ifdef CONFIG_HPET
/* Reserve at least one timer for userspace (/dev/hpet) */
#define RESERVE_TIMERS 1
#else
#define RESERVE_TIMERS 0
#endif
V
Venki Pallipadi 已提交
568 569

static void hpet_msi_capability_lookup(unsigned int start_timer)
570 571 572 573 574 575
{
	unsigned int id;
	unsigned int num_timers;
	unsigned int num_timers_used = 0;
	int i;

576 577 578
	if (hpet_msi_disable)
		return;

579 580
	if (boot_cpu_has(X86_FEATURE_ARAT))
		return;
581 582 583 584
	id = hpet_readl(HPET_ID);

	num_timers = ((id & HPET_ID_NUMBER) >> HPET_ID_NUMBER_SHIFT);
	num_timers++; /* Value read out starts from 0 */
585
	hpet_print_config();
586 587 588 589 590 591 592 593 594

	hpet_devs = kzalloc(sizeof(struct hpet_dev) * num_timers, GFP_KERNEL);
	if (!hpet_devs)
		return;

	hpet_num_timers = num_timers;

	for (i = start_timer; i < num_timers - RESERVE_TIMERS; i++) {
		struct hpet_dev *hdev = &hpet_devs[num_timers_used];
J
Jan Beulich 已提交
595
		unsigned int cfg = hpet_readl(HPET_Tn_CFG(i));
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

		/* Only consider HPET timer with MSI support */
		if (!(cfg & HPET_TN_FSB_CAP))
			continue;

		hdev->flags = 0;
		if (cfg & HPET_TN_PERIODIC_CAP)
			hdev->flags |= HPET_DEV_PERI_CAP;
		hdev->num = i;

		sprintf(hdev->name, "hpet%d", i);
		if (hpet_assign_irq(hdev))
			continue;

		hdev->flags |= HPET_DEV_FSB_CAP;
		hdev->flags |= HPET_DEV_VALID;
		num_timers_used++;
		if (num_timers_used == num_possible_cpus())
			break;
	}

	printk(KERN_INFO "HPET: %d timers in total, %d timers will be used for per-cpu timer\n",
		num_timers, num_timers_used);
}

V
Venki Pallipadi 已提交
621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640
#ifdef CONFIG_HPET
static void hpet_reserve_msi_timers(struct hpet_data *hd)
{
	int i;

	if (!hpet_devs)
		return;

	for (i = 0; i < hpet_num_timers; i++) {
		struct hpet_dev *hdev = &hpet_devs[i];

		if (!(hdev->flags & HPET_DEV_VALID))
			continue;

		hd->hd_irq[hdev->num] = hdev->irq;
		hpet_reserve_timer(hd, hdev->num);
	}
}
#endif

641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689
static struct hpet_dev *hpet_get_unused_timer(void)
{
	int i;

	if (!hpet_devs)
		return NULL;

	for (i = 0; i < hpet_num_timers; i++) {
		struct hpet_dev *hdev = &hpet_devs[i];

		if (!(hdev->flags & HPET_DEV_VALID))
			continue;
		if (test_and_set_bit(HPET_DEV_USED_BIT,
			(unsigned long *)&hdev->flags))
			continue;
		return hdev;
	}
	return NULL;
}

struct hpet_work_struct {
	struct delayed_work work;
	struct completion complete;
};

static void hpet_work(struct work_struct *w)
{
	struct hpet_dev *hdev;
	int cpu = smp_processor_id();
	struct hpet_work_struct *hpet_work;

	hpet_work = container_of(w, struct hpet_work_struct, work.work);

	hdev = hpet_get_unused_timer();
	if (hdev)
		init_one_hpet_msi_clockevent(hdev, cpu);

	complete(&hpet_work->complete);
}

static int hpet_cpuhp_notify(struct notifier_block *n,
		unsigned long action, void *hcpu)
{
	unsigned long cpu = (unsigned long)hcpu;
	struct hpet_work_struct work;
	struct hpet_dev *hdev = per_cpu(cpu_hpet_dev, cpu);

	switch (action & 0xf) {
	case CPU_ONLINE:
A
Andrew Morton 已提交
690
		INIT_DELAYED_WORK_ONSTACK(&work.work, hpet_work);
691 692 693 694
		init_completion(&work.complete);
		/* FIXME: add schedule_work_on() */
		schedule_delayed_work_on(cpu, &work.work, 0);
		wait_for_completion(&work.complete);
695
		destroy_timer_on_stack(&work.work.timer);
696 697 698 699 700 701 702 703 704 705 706 707 708
		break;
	case CPU_DEAD:
		if (hdev) {
			free_irq(hdev->irq, hdev);
			hdev->flags &= ~HPET_DEV_USED;
			per_cpu(cpu_hpet_dev, cpu) = NULL;
		}
		break;
	}
	return NOTIFY_OK;
}
#else

709 710 711 712
static int hpet_setup_msi_irq(unsigned int irq)
{
	return 0;
}
V
Venki Pallipadi 已提交
713 714 715 716 717 718 719
static void hpet_msi_capability_lookup(unsigned int start_timer)
{
	return;
}

#ifdef CONFIG_HPET
static void hpet_reserve_msi_timers(struct hpet_data *hd)
720 721 722
{
	return;
}
V
Venki Pallipadi 已提交
723
#endif
724 725 726 727 728 729 730 731 732

static int hpet_cpuhp_notify(struct notifier_block *n,
		unsigned long action, void *hcpu)
{
	return NOTIFY_OK;
}

#endif

733 734 735
/*
 * Clock source related code
 */
736
static cycle_t read_hpet(struct clocksource *cs)
737 738 739 740
{
	return (cycle_t)hpet_readl(HPET_COUNTER);
}

741 742 743 744 745 746 747
#ifdef CONFIG_X86_64
static cycle_t __vsyscall_fn vread_hpet(void)
{
	return readl((const void __iomem *)fix_to_virt(VSYSCALL_HPET) + 0xf0);
}
#endif

748 749 750 751 752 753
static struct clocksource clocksource_hpet = {
	.name		= "hpet",
	.rating		= 250,
	.read		= read_hpet,
	.mask		= HPET_MASK,
	.flags		= CLOCK_SOURCE_IS_CONTINUOUS,
754
	.resume		= hpet_resume_counter,
755 756 757
#ifdef CONFIG_X86_64
	.vread		= vread_hpet,
#endif
758 759
};

760
static int hpet_clocksource_register(void)
761
{
762
	u64 start, now;
763
	cycle_t t1;
764 765

	/* Start the counter */
766
	hpet_restart_counter();
767

768
	/* Verify whether hpet counter works */
769
	t1 = hpet_readl(HPET_COUNTER);
770 771 772 773 774 775 776 777 778 779 780 781 782
	rdtscll(start);

	/*
	 * We don't know the TSC frequency yet, but waiting for
	 * 200000 TSC cycles is safe:
	 * 4 GHz == 50us
	 * 1 GHz == 200us
	 */
	do {
		rep_nop();
		rdtscll(now);
	} while ((now - start) < 200000UL);

783
	if (t1 == hpet_readl(HPET_COUNTER)) {
784 785
		printk(KERN_WARNING
		       "HPET counter not counting. HPET disabled\n");
786
		return -ENODEV;
787 788
	}

789
	clocksource_register_hz(&clocksource_hpet, (u32)hpet_freq);
790 791 792
	return 0;
}

P
Pavel Machek 已提交
793 794
/**
 * hpet_enable - Try to setup the HPET timer. Returns 1 on success.
795 796 797
 */
int __init hpet_enable(void)
{
798
	unsigned long hpet_period;
J
Jan Beulich 已提交
799
	unsigned int id;
800
	u64 freq;
801
	int i;
802 803 804 805 806 807 808 809 810 811

	if (!is_hpet_capable())
		return 0;

	hpet_set_mapping();

	/*
	 * Read the period and check for a sane value:
	 */
	hpet_period = hpet_readl(HPET_PERIOD);
812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834

	/*
	 * AMD SB700 based systems with spread spectrum enabled use a
	 * SMM based HPET emulation to provide proper frequency
	 * setting. The SMM code is initialized with the first HPET
	 * register access and takes some time to complete. During
	 * this time the config register reads 0xffffffff. We check
	 * for max. 1000 loops whether the config register reads a non
	 * 0xffffffff value to make sure that HPET is up and running
	 * before we go further. A counting loop is safe, as the HPET
	 * access takes thousands of CPU cycles. On non SB700 based
	 * machines this check is only done once and has no side
	 * effects.
	 */
	for (i = 0; hpet_readl(HPET_CFG) == 0xFFFFFFFF; i++) {
		if (i == 1000) {
			printk(KERN_WARNING
			       "HPET config register value = 0xFFFFFFFF. "
			       "Disabling HPET\n");
			goto out_nohpet;
		}
	}

835 836 837
	if (hpet_period < HPET_MIN_PERIOD || hpet_period > HPET_MAX_PERIOD)
		goto out_nohpet;

838 839 840 841 842 843 844 845
	/*
	 * The period is a femto seconds value. Convert it to a
	 * frequency.
	 */
	freq = FSEC_PER_SEC;
	do_div(freq, hpet_period);
	hpet_freq = freq;

846 847 848 849 850
	/*
	 * Read the HPET ID register to retrieve the IRQ routing
	 * information and the number of channels
	 */
	id = hpet_readl(HPET_ID);
851
	hpet_print_config();
852 853 854 855 856 857 858 859 860 861 862 863 864

#ifdef CONFIG_HPET_EMULATE_RTC
	/*
	 * The legacy routing mode needs at least two channels, tick timer
	 * and the rtc emulation channel.
	 */
	if (!(id & HPET_ID_NUMBER))
		goto out_nohpet;
#endif

	if (hpet_clocksource_register())
		goto out_nohpet;

865
	if (id & HPET_ID_LEGSUP) {
866
		hpet_legacy_clockevent_register();
867 868 869
		return 1;
	}
	return 0;
870

871
out_nohpet:
872
	hpet_clear_mapping();
J
Janne Kulmala 已提交
873
	hpet_address = 0;
874 875 876
	return 0;
}

877 878 879 880 881 882 883 884
/*
 * Needs to be late, as the reserve_timer code calls kalloc !
 *
 * Not a problem on i386 as hpet_enable is called from late_time_init,
 * but on x86_64 it is necessary !
 */
static __init int hpet_late_init(void)
{
885 886
	int cpu;

887
	if (boot_hpet_disable)
888 889
		return -ENODEV;

890 891 892 893 894 895 896 897
	if (!hpet_address) {
		if (!force_hpet_address)
			return -ENODEV;

		hpet_address = force_hpet_address;
		hpet_enable();
	}

898 899 900
	if (!hpet_virt_address)
		return -ENODEV;

901 902 903 904 905
	if (hpet_readl(HPET_ID) & HPET_ID_LEGSUP)
		hpet_msi_capability_lookup(2);
	else
		hpet_msi_capability_lookup(0);

906
	hpet_reserve_platform_timers(hpet_readl(HPET_ID));
907
	hpet_print_config();
908

909 910 911
	if (hpet_msi_disable)
		return 0;

912 913 914
	if (boot_cpu_has(X86_FEATURE_ARAT))
		return 0;

915 916 917 918 919 920 921
	for_each_online_cpu(cpu) {
		hpet_cpuhp_notify(NULL, CPU_ONLINE, (void *)(long)cpu);
	}

	/* This notifier should be called after workqueue is ready */
	hotcpu_notifier(hpet_cpuhp_notify, -20);

922 923 924 925
	return 0;
}
fs_initcall(hpet_late_init);

O
OGAWA Hirofumi 已提交
926 927
void hpet_disable(void)
{
928
	if (is_hpet_capable() && hpet_virt_address) {
J
Jan Beulich 已提交
929
		unsigned int cfg = hpet_readl(HPET_CFG);
O
OGAWA Hirofumi 已提交
930 931 932 933 934 935 936 937 938 939

		if (hpet_legacy_int_enabled) {
			cfg &= ~HPET_CFG_LEGACY;
			hpet_legacy_int_enabled = 0;
		}
		cfg &= ~HPET_CFG_ENABLE;
		hpet_writel(cfg, HPET_CFG);
	}
}

940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957
#ifdef CONFIG_HPET_EMULATE_RTC

/* HPET in LegacyReplacement Mode eats up RTC interrupt line. When, HPET
 * is enabled, we support RTC interrupt functionality in software.
 * RTC has 3 kinds of interrupts:
 * 1) Update Interrupt - generate an interrupt, every sec, when RTC clock
 *    is updated
 * 2) Alarm Interrupt - generate an interrupt at a specific time of day
 * 3) Periodic Interrupt - generate periodic interrupt, with frequencies
 *    2Hz-8192Hz (2Hz-64Hz for non-root user) (all freqs in powers of 2)
 * (1) and (2) above are implemented using polling at a frequency of
 * 64 Hz. The exact frequency is a tradeoff between accuracy and interrupt
 * overhead. (DEFAULT_RTC_INT_FREQ)
 * For (3), we use interrupts at 64Hz or user specified periodic
 * frequency, whichever is higher.
 */
#include <linux/mc146818rtc.h>
#include <linux/rtc.h>
958
#include <asm/rtc.h>
959 960 961 962 963 964

#define DEFAULT_RTC_INT_FREQ	64
#define DEFAULT_RTC_SHIFT	6
#define RTC_NUM_INTS		1

static unsigned long hpet_rtc_flags;
D
David Brownell 已提交
965
static int hpet_prev_update_sec;
966 967
static struct rtc_time hpet_alarm_time;
static unsigned long hpet_pie_count;
968
static u32 hpet_t1_cmp;
J
Jan Beulich 已提交
969 970
static u32 hpet_default_delta;
static u32 hpet_pie_delta;
971 972
static unsigned long hpet_pie_limit;

973 974
static rtc_irq_handler irq_handler;

975 976 977 978 979 980 981 982
/*
 * Check that the hpet counter c1 is ahead of the c2
 */
static inline int hpet_cnt_ahead(u32 c1, u32 c2)
{
	return (s32)(c2 - c1) < 0;
}

983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012
/*
 * Registers a IRQ handler.
 */
int hpet_register_irq_handler(rtc_irq_handler handler)
{
	if (!is_hpet_enabled())
		return -ENODEV;
	if (irq_handler)
		return -EBUSY;

	irq_handler = handler;

	return 0;
}
EXPORT_SYMBOL_GPL(hpet_register_irq_handler);

/*
 * Deregisters the IRQ handler registered with hpet_register_irq_handler()
 * and does cleanup.
 */
void hpet_unregister_irq_handler(rtc_irq_handler handler)
{
	if (!is_hpet_enabled())
		return;

	irq_handler = NULL;
	hpet_rtc_flags = 0;
}
EXPORT_SYMBOL_GPL(hpet_unregister_irq_handler);

1013 1014 1015 1016 1017 1018 1019 1020
/*
 * Timer 1 for RTC emulation. We use one shot mode, as periodic mode
 * is not supported by all HPET implementations for timer 1.
 *
 * hpet_rtc_timer_init() is called when the rtc is initialized.
 */
int hpet_rtc_timer_init(void)
{
J
Jan Beulich 已提交
1021 1022
	unsigned int cfg, cnt, delta;
	unsigned long flags;
1023 1024 1025 1026 1027 1028 1029 1030 1031

	if (!is_hpet_enabled())
		return 0;

	if (!hpet_default_delta) {
		uint64_t clc;

		clc = (uint64_t) hpet_clockevent.mult * NSEC_PER_SEC;
		clc >>= hpet_clockevent.shift + DEFAULT_RTC_SHIFT;
J
Jan Beulich 已提交
1032
		hpet_default_delta = clc;
1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054
	}

	if (!(hpet_rtc_flags & RTC_PIE) || hpet_pie_limit)
		delta = hpet_default_delta;
	else
		delta = hpet_pie_delta;

	local_irq_save(flags);

	cnt = delta + hpet_readl(HPET_COUNTER);
	hpet_writel(cnt, HPET_T1_CMP);
	hpet_t1_cmp = cnt;

	cfg = hpet_readl(HPET_T1_CFG);
	cfg &= ~HPET_TN_PERIODIC;
	cfg |= HPET_TN_ENABLE | HPET_TN_32BIT;
	hpet_writel(cfg, HPET_T1_CFG);

	local_irq_restore(flags);

	return 1;
}
1055
EXPORT_SYMBOL_GPL(hpet_rtc_timer_init);
1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069

/*
 * The functions below are called from rtc driver.
 * Return 0 if HPET is not being used.
 * Otherwise do the necessary changes and return 1.
 */
int hpet_mask_rtc_irq_bit(unsigned long bit_mask)
{
	if (!is_hpet_enabled())
		return 0;

	hpet_rtc_flags &= ~bit_mask;
	return 1;
}
1070
EXPORT_SYMBOL_GPL(hpet_mask_rtc_irq_bit);
1071 1072 1073 1074 1075 1076 1077 1078 1079 1080

int hpet_set_rtc_irq_bit(unsigned long bit_mask)
{
	unsigned long oldbits = hpet_rtc_flags;

	if (!is_hpet_enabled())
		return 0;

	hpet_rtc_flags |= bit_mask;

D
David Brownell 已提交
1081 1082 1083
	if ((bit_mask & RTC_UIE) && !(oldbits & RTC_UIE))
		hpet_prev_update_sec = -1;

1084 1085 1086 1087 1088
	if (!oldbits)
		hpet_rtc_timer_init();

	return 1;
}
1089
EXPORT_SYMBOL_GPL(hpet_set_rtc_irq_bit);
1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102

int hpet_set_alarm_time(unsigned char hrs, unsigned char min,
			unsigned char sec)
{
	if (!is_hpet_enabled())
		return 0;

	hpet_alarm_time.tm_hour = hrs;
	hpet_alarm_time.tm_min = min;
	hpet_alarm_time.tm_sec = sec;

	return 1;
}
1103
EXPORT_SYMBOL_GPL(hpet_set_alarm_time);
1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117

int hpet_set_periodic_freq(unsigned long freq)
{
	uint64_t clc;

	if (!is_hpet_enabled())
		return 0;

	if (freq <= DEFAULT_RTC_INT_FREQ)
		hpet_pie_limit = DEFAULT_RTC_INT_FREQ / freq;
	else {
		clc = (uint64_t) hpet_clockevent.mult * NSEC_PER_SEC;
		do_div(clc, freq);
		clc >>= hpet_clockevent.shift;
J
Jan Beulich 已提交
1118
		hpet_pie_delta = clc;
1119
		hpet_pie_limit = 0;
1120 1121 1122
	}
	return 1;
}
1123
EXPORT_SYMBOL_GPL(hpet_set_periodic_freq);
1124 1125 1126 1127 1128

int hpet_rtc_dropped_irq(void)
{
	return is_hpet_enabled();
}
1129
EXPORT_SYMBOL_GPL(hpet_rtc_dropped_irq);
1130 1131 1132

static void hpet_rtc_timer_reinit(void)
{
J
Jan Beulich 已提交
1133
	unsigned int cfg, delta;
1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155
	int lost_ints = -1;

	if (unlikely(!hpet_rtc_flags)) {
		cfg = hpet_readl(HPET_T1_CFG);
		cfg &= ~HPET_TN_ENABLE;
		hpet_writel(cfg, HPET_T1_CFG);
		return;
	}

	if (!(hpet_rtc_flags & RTC_PIE) || hpet_pie_limit)
		delta = hpet_default_delta;
	else
		delta = hpet_pie_delta;

	/*
	 * Increment the comparator value until we are ahead of the
	 * current count.
	 */
	do {
		hpet_t1_cmp += delta;
		hpet_writel(hpet_t1_cmp, HPET_T1_CMP);
		lost_ints++;
1156
	} while (!hpet_cnt_ahead(hpet_t1_cmp, hpet_readl(HPET_COUNTER)));
1157 1158 1159 1160 1161

	if (lost_ints) {
		if (hpet_rtc_flags & RTC_PIE)
			hpet_pie_count += lost_ints;
		if (printk_ratelimit())
D
David Brownell 已提交
1162
			printk(KERN_WARNING "hpet1: lost %d rtc interrupts\n",
1163 1164 1165 1166 1167 1168 1169 1170 1171 1172
				lost_ints);
	}
}

irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id)
{
	struct rtc_time curr_time;
	unsigned long rtc_int_flag = 0;

	hpet_rtc_timer_reinit();
1173
	memset(&curr_time, 0, sizeof(struct rtc_time));
1174 1175

	if (hpet_rtc_flags & (RTC_UIE | RTC_AIE))
1176
		get_rtc_time(&curr_time);
1177 1178 1179

	if (hpet_rtc_flags & RTC_UIE &&
	    curr_time.tm_sec != hpet_prev_update_sec) {
D
David Brownell 已提交
1180 1181
		if (hpet_prev_update_sec >= 0)
			rtc_int_flag = RTC_UF;
1182 1183 1184 1185 1186 1187 1188 1189 1190
		hpet_prev_update_sec = curr_time.tm_sec;
	}

	if (hpet_rtc_flags & RTC_PIE &&
	    ++hpet_pie_count >= hpet_pie_limit) {
		rtc_int_flag |= RTC_PF;
		hpet_pie_count = 0;
	}

1191
	if (hpet_rtc_flags & RTC_AIE &&
1192 1193 1194 1195 1196 1197 1198
	    (curr_time.tm_sec == hpet_alarm_time.tm_sec) &&
	    (curr_time.tm_min == hpet_alarm_time.tm_min) &&
	    (curr_time.tm_hour == hpet_alarm_time.tm_hour))
			rtc_int_flag |= RTC_AF;

	if (rtc_int_flag) {
		rtc_int_flag |= (RTC_IRQF | (RTC_NUM_INTS << 8));
1199 1200
		if (irq_handler)
			irq_handler(rtc_int_flag, dev_id);
1201 1202 1203
	}
	return IRQ_HANDLED;
}
1204
EXPORT_SYMBOL_GPL(hpet_rtc_interrupt);
1205
#endif