hpet.c 29.0 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/slab.h>
8 9
#include <linux/hpet.h>
#include <linux/init.h>
10
#include <linux/cpu.h>
11 12
#include <linux/pm.h>
#include <linux/io.h>
13

14
#include <asm/fixmap.h>
15
#include <asm/i8253.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 30 31
#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

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

32 33 34
/*
 * HPET address is set in acpi/boot.c, when an ACPI entry exists
 */
35
unsigned long				hpet_address;
36
u8					hpet_blockid; /* OS timer block num */
37 38
u8					hpet_msi_disable;

39
#ifdef CONFIG_PCI_MSI
H
Hannes Eder 已提交
40
static unsigned long			hpet_num_timers;
41
#endif
42
static void __iomem			*hpet_virt_address;
43

44
struct hpet_dev {
45 46 47 48 49 50
	struct clock_event_device	evt;
	unsigned int			num;
	int				cpu;
	unsigned int			irq;
	unsigned int			flags;
	char				name[10];
51 52
};

J
Jan Beulich 已提交
53
inline unsigned int hpet_readl(unsigned int a)
54 55 56 57
{
	return readl(hpet_virt_address + a);
}

J
Jan Beulich 已提交
58
static inline void hpet_writel(unsigned int d, unsigned int a)
59 60 61 62
{
	writel(d, hpet_virt_address + a);
}

63 64
#ifdef CONFIG_X86_64
#include <asm/pgtable.h>
65
#endif
66

67 68 69
static inline void hpet_set_mapping(void)
{
	hpet_virt_address = ioremap_nocache(hpet_address, HPET_MMAP_SIZE);
70 71 72
#ifdef CONFIG_X86_64
	__set_fixmap(VSYSCALL_HPET, hpet_address, PAGE_KERNEL_VSYSCALL_NOCACHE);
#endif
73 74 75 76 77 78 79 80
}

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

81 82 83 84
/*
 * HPET command line enable / disable
 */
static int boot_hpet_disable;
T
Thomas Gleixner 已提交
85
int hpet_force_user;
86
static int hpet_verbose;
87

88
static int __init hpet_setup(char *str)
89 90 91 92
{
	if (str) {
		if (!strncmp("disable", str, 7))
			boot_hpet_disable = 1;
T
Thomas Gleixner 已提交
93 94
		if (!strncmp("force", str, 5))
			hpet_force_user = 1;
95 96
		if (!strncmp("verbose", str, 7))
			hpet_verbose = 1;
97 98 99 100 101
	}
	return 1;
}
__setup("hpet=", hpet_setup);

102 103 104 105 106 107 108
static int __init disable_hpet(char *str)
{
	boot_hpet_disable = 1;
	return 1;
}
__setup("nohpet", disable_hpet);

109 110
static inline int is_hpet_capable(void)
{
111
	return !boot_hpet_disable && hpet_address;
112 113 114 115 116 117 118 119 120 121 122 123 124 125
}

/*
 * 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;
}
126
EXPORT_SYMBOL_GPL(is_hpet_enabled);
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
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)

165 166 167 168 169
/*
 * 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
170

V
Venki Pallipadi 已提交
171
static void hpet_reserve_msi_timers(struct hpet_data *hd);
172

J
Jan Beulich 已提交
173
static void hpet_reserve_platform_timers(unsigned int id)
174 175
{
	struct hpet __iomem *hpet = hpet_virt_address;
176 177
	struct hpet_timer __iomem *timer = &hpet->hpet_timers[2];
	unsigned int nrtimers, i;
178 179 180 181
	struct hpet_data hd;

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

182 183 184 185
	memset(&hd, 0, sizeof(hd));
	hd.hd_phys_address	= hpet_address;
	hd.hd_address		= hpet;
	hd.hd_nirqs		= nrtimers;
186 187 188 189 190
	hpet_reserve_timer(&hd, 0);

#ifdef CONFIG_HPET_EMULATE_RTC
	hpet_reserve_timer(&hd, 1);
#endif
191

192 193 194 195 196
	/*
	 * 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.
	 */
197 198 199
	hd.hd_irq[0] = HPET_LEGACY_8254;
	hd.hd_irq[1] = HPET_LEGACY_RTC;

I
Ingo Molnar 已提交
200
	for (i = 2; i < nrtimers; timer++, i++) {
201 202
		hd.hd_irq[i] = (readl(&timer->hpet_config) &
			Tn_INT_ROUTE_CNF_MASK) >> Tn_INT_ROUTE_CNF_SHIFT;
I
Ingo Molnar 已提交
203
	}
204

205
	hpet_reserve_msi_timers(&hd);
206

207
	hpet_alloc(&hd);
208

209 210
}
#else
J
Jan Beulich 已提交
211
static void hpet_reserve_platform_timers(unsigned int id) { }
212 213 214 215 216 217 218
#endif

/*
 * Common hpet info
 */
static unsigned long hpet_period;

219
static void hpet_legacy_set_mode(enum clock_event_mode mode,
220
			  struct clock_event_device *evt);
221
static int hpet_legacy_next_event(unsigned long delta,
222 223 224 225 226 227 228 229
			   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,
230 231
	.set_mode	= hpet_legacy_set_mode,
	.set_next_event = hpet_legacy_next_event,
232 233
	.shift		= 32,
	.irq		= 0,
234
	.rating		= 50,
235 236
};

237
static void hpet_stop_counter(void)
238 239 240 241
{
	unsigned long cfg = hpet_readl(HPET_CFG);
	cfg &= ~HPET_CFG_ENABLE;
	hpet_writel(cfg, HPET_CFG);
242 243 244 245
}

static void hpet_reset_counter(void)
{
246 247
	hpet_writel(0, HPET_COUNTER);
	hpet_writel(0, HPET_COUNTER + 4);
248 249 250 251
}

static void hpet_start_counter(void)
{
J
Jan Beulich 已提交
252
	unsigned int cfg = hpet_readl(HPET_CFG);
253 254 255 256
	cfg |= HPET_CFG_ENABLE;
	hpet_writel(cfg, HPET_CFG);
}

257 258 259
static void hpet_restart_counter(void)
{
	hpet_stop_counter();
260
	hpet_reset_counter();
261 262 263
	hpet_start_counter();
}

264 265
static void hpet_resume_device(void)
{
V
Venki Pallipadi 已提交
266
	force_hpet_resume();
267 268
}

269
static void hpet_resume_counter(struct clocksource *cs)
270 271
{
	hpet_resume_device();
272
	hpet_restart_counter();
273 274
}

275
static void hpet_enable_legacy_int(void)
276
{
J
Jan Beulich 已提交
277
	unsigned int cfg = hpet_readl(HPET_CFG);
278 279 280 281 282 283

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

284 285 286 287 288 289
static void hpet_legacy_clockevent_register(void)
{
	/* Start HPET legacy interrupts */
	hpet_enable_legacy_int();

	/*
290 291 292 293 294 295
	 * The mult factor is defined as (include/linux/clockchips.h)
	 *  mult/2^shift = cyc/ns (in contrast to ns/cyc in clocksource.h)
	 * hpet_period is in units of femtoseconds (per cycle), so
	 *  mult/2^shift = cyc/ns = 10^6/hpet_period
	 *  mult = (10^6 * 2^shift)/hpet_period
	 *  mult = (FSEC_PER_NSEC << hpet_clockevent.shift)/hpet_period
296
	 */
297 298
	hpet_clockevent.mult = div_sc((unsigned long) FSEC_PER_NSEC,
				      hpet_period, hpet_clockevent.shift);
299 300 301
	/* Calculate the min / max delta */
	hpet_clockevent.max_delta_ns = clockevent_delta2ns(0x7FFFFFFF,
							   &hpet_clockevent);
302 303
	/* 5 usec minimum reprogramming delta. */
	hpet_clockevent.min_delta_ns = 5000;
304 305 306 307 308

	/*
	 * Start hpet with the boot cpu mask and make it
	 * global after the IO_APIC has been initialized.
	 */
309
	hpet_clockevent.cpumask = cpumask_of(smp_processor_id());
310 311 312 313 314
	clockevents_register_device(&hpet_clockevent);
	global_clock_event = &hpet_clockevent;
	printk(KERN_DEBUG "hpet clockevent registered\n");
}

315 316
static int hpet_setup_msi_irq(unsigned int irq);

317 318
static void hpet_set_mode(enum clock_event_mode mode,
			  struct clock_event_device *evt, int timer)
319
{
J
Jan Beulich 已提交
320
	unsigned int cfg, cmp, now;
321 322
	uint64_t delta;

323
	switch (mode) {
324
	case CLOCK_EVT_MODE_PERIODIC:
325
		hpet_stop_counter();
326 327
		delta = ((uint64_t)(NSEC_PER_SEC/HZ)) * evt->mult;
		delta >>= evt->shift;
328
		now = hpet_readl(HPET_COUNTER);
J
Jan Beulich 已提交
329
		cmp = now + (unsigned int) delta;
330
		cfg = hpet_readl(HPET_Tn_CFG(timer));
331 332
		/* Make sure we use edge triggered interrupts */
		cfg &= ~HPET_TN_LEVEL;
333 334
		cfg |= HPET_TN_ENABLE | HPET_TN_PERIODIC |
		       HPET_TN_SETVAL | HPET_TN_32BIT;
335
		hpet_writel(cfg, HPET_Tn_CFG(timer));
336 337 338 339 340 341 342 343 344
		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
345
		hpet_writel((unsigned int) delta, HPET_Tn_CMP(timer));
346
		hpet_start_counter();
347
		hpet_print_config();
348 349 350
		break;

	case CLOCK_EVT_MODE_ONESHOT:
351
		cfg = hpet_readl(HPET_Tn_CFG(timer));
352 353
		cfg &= ~HPET_TN_PERIODIC;
		cfg |= HPET_TN_ENABLE | HPET_TN_32BIT;
354
		hpet_writel(cfg, HPET_Tn_CFG(timer));
355 356 357 358
		break;

	case CLOCK_EVT_MODE_UNUSED:
	case CLOCK_EVT_MODE_SHUTDOWN:
359
		cfg = hpet_readl(HPET_Tn_CFG(timer));
360
		cfg &= ~HPET_TN_ENABLE;
361
		hpet_writel(cfg, HPET_Tn_CFG(timer));
362
		break;
T
Thomas Gleixner 已提交
363 364

	case CLOCK_EVT_MODE_RESUME:
365 366 367 368 369 370
		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);
371
			irq_set_affinity(hdev->irq, cpumask_of(hdev->cpu));
372 373
			enable_irq(hdev->irq);
		}
374
		hpet_print_config();
T
Thomas Gleixner 已提交
375
		break;
376 377 378
	}
}

379 380
static int hpet_next_event(unsigned long delta,
			   struct clock_event_device *evt, int timer)
381
{
382
	u32 cnt;
383 384

	cnt = hpet_readl(HPET_COUNTER);
385
	cnt += (u32) delta;
386
	hpet_writel(cnt, HPET_Tn_CMP(timer));
387

388
	/*
389 390 391 392 393 394 395 396
	 * We need to read back the CMP register on certain HPET
	 * implementations (ATI chipsets) which seem to delay the
	 * transfer of the compare register into the internal compare
	 * logic. With small deltas this might actually be too late as
	 * the counter could already be higher than the compare value
	 * at that point and we would wait for the next hpet interrupt
	 * forever. We found out that reading the CMP register back
	 * forces the transfer so we can rely on the comparison with
397 398 399 400 401
	 * the counter register below. If the read back from the
	 * compare register does not match the value we programmed
	 * then we might have a real hardware problem. We can not do
	 * much about it here, but at least alert the user/admin with
	 * a prominent warning.
402
	 *
403 404 405 406
	 * An erratum on some chipsets (ICH9,..), results in
	 * comparator read immediately following a write returning old
	 * value. Workaround for this is to read this value second
	 * time, when first read returns old value.
407
	 *
408 409 410 411 412
	 * In fact the write to the comparator register is delayed up
	 * to two HPET cycles so the workaround we tried to restrict
	 * the readback to those known to be borked ATI chipsets
	 * failed miserably. So we give up on optimizations forever
	 * and penalize all HPET incarnations unconditionally.
413
	 */
414 415
	if (unlikely((u32)hpet_readl(HPET_Tn_CMP(timer)) != cnt)) {
		if (hpet_readl(HPET_Tn_CMP(timer)) != cnt)
416
			printk_once(KERN_WARNING
417
				"hpet: compare register read back failed.\n");
418
	}
419

J
Jan Beulich 已提交
420
	return (s32)(hpet_readl(HPET_COUNTER) - cnt) >= 0 ? -ETIME : 0;
421 422
}

423 424 425 426 427 428 429 430 431 432 433 434
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);
}

435 436 437
/*
 * HPET MSI Support
 */
438
#ifdef CONFIG_PCI_MSI
V
Venki Pallipadi 已提交
439 440 441 442

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

443 444 445
void hpet_msi_unmask(unsigned int irq)
{
	struct hpet_dev *hdev = get_irq_data(irq);
J
Jan Beulich 已提交
446
	unsigned int cfg;
447 448 449 450 451 452 453 454 455

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

void hpet_msi_mask(unsigned int irq)
{
J
Jan Beulich 已提交
456
	unsigned int cfg;
457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481
	struct hpet_dev *hdev = get_irq_data(irq);

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

void hpet_msi_write(unsigned int irq, struct msi_msg *msg)
{
	struct hpet_dev *hdev = get_irq_data(irq);

	hpet_writel(msg->data, HPET_Tn_ROUTE(hdev->num));
	hpet_writel(msg->address_lo, HPET_Tn_ROUTE(hdev->num) + 4);
}

void hpet_msi_read(unsigned int irq, struct msi_msg *msg)
{
	struct hpet_dev *hdev = get_irq_data(irq);

	msg->data = hpet_readl(HPET_Tn_ROUTE(hdev->num));
	msg->address_lo = hpet_readl(HPET_Tn_ROUTE(hdev->num) + 4);
	msg->address_hi = 0;
}

482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497
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)
{
498
	if (arch_setup_hpet_msi(irq, hpet_blockid)) {
499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540
		destroy_irq(irq);
		return -EINVAL;
	}
	return 0;
}

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

	irq = create_irq();
	if (!irq)
		return -EINVAL;

	set_irq_data(irq, dev);

	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,
541 542
			IRQF_TIMER | IRQF_DISABLED | IRQF_NOBALANCING,
			dev->name, dev))
543 544 545
		return -1;

	disable_irq(dev->irq);
546
	irq_set_affinity(dev->irq, cpumask_of(dev->cpu));
547 548
	enable_irq(dev->irq);

Y
Yinghai Lu 已提交
549 550 551
	printk(KERN_DEBUG "hpet: %s irq %d for MSI\n",
			 dev->name, dev->irq);

552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587
	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;
	uint64_t hpet_freq;

	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;
	evt->shift = 32;

	/*
	 * The period is a femto seconds value. We need to calculate the
	 * scaled math multiplication factor for nanosecond to hpet tick
	 * conversion.
	 */
588
	hpet_freq = FSEC_PER_SEC;
589 590 591 592 593 594 595 596
	do_div(hpet_freq, hpet_period);
	evt->mult = div_sc((unsigned long) hpet_freq,
				      NSEC_PER_SEC, evt->shift);
	/* Calculate the max delta */
	evt->max_delta_ns = clockevent_delta2ns(0x7FFFFFFF, evt);
	/* 5 usec minimum reprogramming delta. */
	evt->min_delta_ns = 5000;

597
	evt->cpumask = cpumask_of(hdev->cpu);
598 599 600 601 602 603 604 605 606
	clockevents_register_device(evt);
}

#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
607 608

static void hpet_msi_capability_lookup(unsigned int start_timer)
609 610 611 612 613 614
{
	unsigned int id;
	unsigned int num_timers;
	unsigned int num_timers_used = 0;
	int i;

615 616 617
	if (hpet_msi_disable)
		return;

618 619
	if (boot_cpu_has(X86_FEATURE_ARAT))
		return;
620 621 622 623
	id = hpet_readl(HPET_ID);

	num_timers = ((id & HPET_ID_NUMBER) >> HPET_ID_NUMBER_SHIFT);
	num_timers++; /* Value read out starts from 0 */
624
	hpet_print_config();
625 626 627 628 629 630 631 632 633

	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
634
		unsigned int cfg = hpet_readl(HPET_Tn_CFG(i));
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

		/* 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
660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679
#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

680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728
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:
729
		INIT_DELAYED_WORK_ON_STACK(&work.work, hpet_work);
730 731 732 733
		init_completion(&work.complete);
		/* FIXME: add schedule_work_on() */
		schedule_delayed_work_on(cpu, &work.work, 0);
		wait_for_completion(&work.complete);
734
		destroy_timer_on_stack(&work.work.timer);
735 736 737 738 739 740 741 742 743 744 745 746 747
		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

748 749 750 751
static int hpet_setup_msi_irq(unsigned int irq)
{
	return 0;
}
V
Venki Pallipadi 已提交
752 753 754 755 756 757 758
static void hpet_msi_capability_lookup(unsigned int start_timer)
{
	return;
}

#ifdef CONFIG_HPET
static void hpet_reserve_msi_timers(struct hpet_data *hd)
759 760 761
{
	return;
}
V
Venki Pallipadi 已提交
762
#endif
763 764 765 766 767 768 769 770 771

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

#endif

772 773 774
/*
 * Clock source related code
 */
775
static cycle_t read_hpet(struct clocksource *cs)
776 777 778 779
{
	return (cycle_t)hpet_readl(HPET_COUNTER);
}

780 781 782 783 784 785 786
#ifdef CONFIG_X86_64
static cycle_t __vsyscall_fn vread_hpet(void)
{
	return readl((const void __iomem *)fix_to_virt(VSYSCALL_HPET) + 0xf0);
}
#endif

787 788 789 790 791 792
static struct clocksource clocksource_hpet = {
	.name		= "hpet",
	.rating		= 250,
	.read		= read_hpet,
	.mask		= HPET_MASK,
	.flags		= CLOCK_SOURCE_IS_CONTINUOUS,
793
	.resume		= hpet_resume_counter,
794 795 796
#ifdef CONFIG_X86_64
	.vread		= vread_hpet,
#endif
797 798
};

799
static int hpet_clocksource_register(void)
800
{
801
	u64 start, now;
802
	u64 hpet_freq;
803
	cycle_t t1;
804 805

	/* Start the counter */
806
	hpet_restart_counter();
807

808
	/* Verify whether hpet counter works */
809
	t1 = hpet_readl(HPET_COUNTER);
810 811 812 813 814 815 816 817 818 819 820 821 822
	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);

823
	if (t1 == hpet_readl(HPET_COUNTER)) {
824 825
		printk(KERN_WARNING
		       "HPET counter not counting. HPET disabled\n");
826
		return -ENODEV;
827 828
	}

829 830 831 832 833 834 835
	/*
	 * The definition of mult is (include/linux/clocksource.h)
	 * mult/2^shift = ns/cyc and hpet_period is in units of fsec/cyc
	 * so we first need to convert hpet_period to ns/cyc units:
	 *  mult/2^shift = ns/cyc = hpet_period/10^6
	 *  mult = (hpet_period * 2^shift)/10^6
	 *  mult = (hpet_period << shift)/FSEC_PER_NSEC
836 837
	 */

838 839 840 841 842
	/* Need to convert hpet_period (fsec/cyc) to cyc/sec:
	 *
	 * cyc/sec = FSEC_PER_SEC/hpet_period(fsec/cyc)
	 * cyc/sec = (FSEC_PER_NSEC * NSEC_PER_SEC)/hpet_period
	 */
843
	hpet_freq = FSEC_PER_SEC;
844 845
	do_div(hpet_freq, hpet_period);
	clocksource_register_hz(&clocksource_hpet, (u32)hpet_freq);
846

847 848 849
	return 0;
}

P
Pavel Machek 已提交
850 851
/**
 * hpet_enable - Try to setup the HPET timer. Returns 1 on success.
852 853 854
 */
int __init hpet_enable(void)
{
J
Jan Beulich 已提交
855
	unsigned int id;
856
	int i;
857 858 859 860 861 862 863 864 865 866

	if (!is_hpet_capable())
		return 0;

	hpet_set_mapping();

	/*
	 * Read the period and check for a sane value:
	 */
	hpet_period = hpet_readl(HPET_PERIOD);
867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889

	/*
	 * 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;
		}
	}

890 891 892 893 894 895 896 897
	if (hpet_period < HPET_MIN_PERIOD || hpet_period > HPET_MAX_PERIOD)
		goto out_nohpet;

	/*
	 * Read the HPET ID register to retrieve the IRQ routing
	 * information and the number of channels
	 */
	id = hpet_readl(HPET_ID);
898
	hpet_print_config();
899 900 901 902 903 904 905 906 907 908 909 910 911

#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;

912
	if (id & HPET_ID_LEGSUP) {
913
		hpet_legacy_clockevent_register();
914 915 916
		return 1;
	}
	return 0;
917

918
out_nohpet:
919
	hpet_clear_mapping();
J
Janne Kulmala 已提交
920
	hpet_address = 0;
921 922 923
	return 0;
}

924 925 926 927 928 929 930 931
/*
 * 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)
{
932 933
	int cpu;

934
	if (boot_hpet_disable)
935 936
		return -ENODEV;

937 938 939 940 941 942 943 944
	if (!hpet_address) {
		if (!force_hpet_address)
			return -ENODEV;

		hpet_address = force_hpet_address;
		hpet_enable();
	}

945 946 947
	if (!hpet_virt_address)
		return -ENODEV;

948 949 950 951 952
	if (hpet_readl(HPET_ID) & HPET_ID_LEGSUP)
		hpet_msi_capability_lookup(2);
	else
		hpet_msi_capability_lookup(0);

953
	hpet_reserve_platform_timers(hpet_readl(HPET_ID));
954
	hpet_print_config();
955

956 957 958
	if (hpet_msi_disable)
		return 0;

959 960 961
	if (boot_cpu_has(X86_FEATURE_ARAT))
		return 0;

962 963 964 965 966 967 968
	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);

969 970 971 972
	return 0;
}
fs_initcall(hpet_late_init);

O
OGAWA Hirofumi 已提交
973 974
void hpet_disable(void)
{
975
	if (is_hpet_capable() && hpet_virt_address) {
J
Jan Beulich 已提交
976
		unsigned int cfg = hpet_readl(HPET_CFG);
O
OGAWA Hirofumi 已提交
977 978 979 980 981 982 983 984 985 986

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

987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004
#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>
1005
#include <asm/rtc.h>
1006 1007 1008 1009 1010 1011

#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
1012
static int hpet_prev_update_sec;
1013 1014
static struct rtc_time hpet_alarm_time;
static unsigned long hpet_pie_count;
1015
static u32 hpet_t1_cmp;
J
Jan Beulich 已提交
1016 1017
static u32 hpet_default_delta;
static u32 hpet_pie_delta;
1018 1019
static unsigned long hpet_pie_limit;

1020 1021
static rtc_irq_handler irq_handler;

1022 1023 1024 1025 1026 1027 1028 1029
/*
 * 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;
}

1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059
/*
 * 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);

1060 1061 1062 1063 1064 1065 1066 1067
/*
 * 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
1068 1069
	unsigned int cfg, cnt, delta;
	unsigned long flags;
1070 1071 1072 1073 1074 1075 1076 1077 1078

	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
1079
		hpet_default_delta = clc;
1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101
	}

	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;
}
1102
EXPORT_SYMBOL_GPL(hpet_rtc_timer_init);
1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116

/*
 * 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;
}
1117
EXPORT_SYMBOL_GPL(hpet_mask_rtc_irq_bit);
1118 1119 1120 1121 1122 1123 1124 1125 1126 1127

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
1128 1129 1130
	if ((bit_mask & RTC_UIE) && !(oldbits & RTC_UIE))
		hpet_prev_update_sec = -1;

1131 1132 1133 1134 1135
	if (!oldbits)
		hpet_rtc_timer_init();

	return 1;
}
1136
EXPORT_SYMBOL_GPL(hpet_set_rtc_irq_bit);
1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149

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;
}
1150
EXPORT_SYMBOL_GPL(hpet_set_alarm_time);
1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164

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
1165
		hpet_pie_delta = clc;
1166
		hpet_pie_limit = 0;
1167 1168 1169
	}
	return 1;
}
1170
EXPORT_SYMBOL_GPL(hpet_set_periodic_freq);
1171 1172 1173 1174 1175

int hpet_rtc_dropped_irq(void)
{
	return is_hpet_enabled();
}
1176
EXPORT_SYMBOL_GPL(hpet_rtc_dropped_irq);
1177 1178 1179

static void hpet_rtc_timer_reinit(void)
{
J
Jan Beulich 已提交
1180
	unsigned int cfg, delta;
1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202
	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++;
1203
	} while (!hpet_cnt_ahead(hpet_t1_cmp, hpet_readl(HPET_COUNTER)));
1204 1205 1206 1207 1208

	if (lost_ints) {
		if (hpet_rtc_flags & RTC_PIE)
			hpet_pie_count += lost_ints;
		if (printk_ratelimit())
D
David Brownell 已提交
1209
			printk(KERN_WARNING "hpet1: lost %d rtc interrupts\n",
1210 1211 1212 1213 1214 1215 1216 1217 1218 1219
				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();
1220
	memset(&curr_time, 0, sizeof(struct rtc_time));
1221 1222

	if (hpet_rtc_flags & (RTC_UIE | RTC_AIE))
1223
		get_rtc_time(&curr_time);
1224 1225 1226

	if (hpet_rtc_flags & RTC_UIE &&
	    curr_time.tm_sec != hpet_prev_update_sec) {
D
David Brownell 已提交
1227 1228
		if (hpet_prev_update_sec >= 0)
			rtc_int_flag = RTC_UF;
1229 1230 1231 1232 1233 1234 1235 1236 1237
		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;
	}

1238
	if (hpet_rtc_flags & RTC_AIE &&
1239 1240 1241 1242 1243 1244 1245
	    (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));
1246 1247
		if (irq_handler)
			irq_handler(rtc_int_flag, dev_id);
1248 1249 1250
	}
	return IRQ_HANDLED;
}
1251
EXPORT_SYMBOL_GPL(hpet_rtc_interrupt);
1252
#endif