pci.c 82.7 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13
/*
 *	PCI Bus Services, see include/linux/pci.h for further explanation.
 *
 *	Copyright 1993 -- 1997 Drew Eckhardt, Frederic Potter,
 *	David Mosberger-Tang
 *
 *	Copyright 1997 -- 2000 Martin Mares <mj@ucw.cz>
 */

#include <linux/kernel.h>
#include <linux/delay.h>
#include <linux/init.h>
#include <linux/pci.h>
14
#include <linux/pm.h>
15
#include <linux/slab.h>
L
Linus Torvalds 已提交
16 17
#include <linux/module.h>
#include <linux/spinlock.h>
T
Tim Schmielau 已提交
18
#include <linux/string.h>
19
#include <linux/log2.h>
S
Shaohua Li 已提交
20
#include <linux/pci-aspm.h>
21
#include <linux/pm_wakeup.h>
22
#include <linux/interrupt.h>
23
#include <linux/device.h>
24
#include <linux/pm_runtime.h>
25
#include <asm/setup.h>
26
#include "pci.h"
L
Linus Torvalds 已提交
27

A
Alan Stern 已提交
28 29 30 31 32
const char *pci_power_names[] = {
	"error", "D0", "D1", "D2", "D3hot", "D3cold", "unknown",
};
EXPORT_SYMBOL_GPL(pci_power_names);

33 34 35 36 37 38
int isa_dma_bridge_buggy;
EXPORT_SYMBOL(isa_dma_bridge_buggy);

int pci_pci_problems;
EXPORT_SYMBOL(pci_pci_problems);

39 40
unsigned int pci_pm_d3_delay;

41 42 43 44 45 46 47 48 49 50 51 52 53
static void pci_pme_list_scan(struct work_struct *work);

static LIST_HEAD(pci_pme_list);
static DEFINE_MUTEX(pci_pme_list_mutex);
static DECLARE_DELAYED_WORK(pci_pme_work, pci_pme_list_scan);

struct pci_pme_device {
	struct list_head list;
	struct pci_dev *dev;
};

#define PME_TIMEOUT 1000 /* How long between PME checks */

54 55 56 57 58 59 60 61 62
static void pci_dev_d3_sleep(struct pci_dev *dev)
{
	unsigned int delay = dev->d3_delay;

	if (delay < pci_pm_d3_delay)
		delay = pci_pm_d3_delay;

	msleep(delay);
}
L
Linus Torvalds 已提交
63

64 65 66 67
#ifdef CONFIG_PCI_DOMAINS
int pci_domains_supported = 1;
#endif

68 69 70 71 72 73
#define DEFAULT_CARDBUS_IO_SIZE		(256)
#define DEFAULT_CARDBUS_MEM_SIZE	(64*1024*1024)
/* pci=cbmemsize=nnM,cbiosize=nn can override this */
unsigned long pci_cardbus_io_size = DEFAULT_CARDBUS_IO_SIZE;
unsigned long pci_cardbus_mem_size = DEFAULT_CARDBUS_MEM_SIZE;

74 75 76 77 78 79
#define DEFAULT_HOTPLUG_IO_SIZE		(256)
#define DEFAULT_HOTPLUG_MEM_SIZE	(2*1024*1024)
/* pci=hpmemsize=nnM,hpiosize=nn can override this */
unsigned long pci_hotplug_io_size  = DEFAULT_HOTPLUG_IO_SIZE;
unsigned long pci_hotplug_mem_size = DEFAULT_HOTPLUG_MEM_SIZE;

80 81 82 83 84 85
/*
 * The default CLS is used if arch didn't set CLS explicitly and not
 * all pci devices agree on the same value.  Arch can override either
 * the dfl or actual value as it sees fit.  Don't forget this is
 * measured in 32-bit words, not bytes.
 */
86
u8 pci_dfl_cache_line_size __devinitdata = L1_CACHE_BYTES >> 2;
87 88
u8 pci_cache_line_size;

L
Linus Torvalds 已提交
89 90 91 92 93 94 95
/**
 * pci_bus_max_busnr - returns maximum PCI bus number of given bus' children
 * @bus: pointer to PCI bus structure to search
 *
 * Given a PCI bus, returns the highest PCI bus number present in the set
 * including the given PCI bus and its list of child PCI buses.
 */
96
unsigned char pci_bus_max_busnr(struct pci_bus* bus)
L
Linus Torvalds 已提交
97 98 99 100
{
	struct list_head *tmp;
	unsigned char max, n;

101
	max = bus->subordinate;
L
Linus Torvalds 已提交
102 103 104 105 106 107 108
	list_for_each(tmp, &bus->children) {
		n = pci_bus_max_busnr(pci_bus_b(tmp));
		if(n > max)
			max = n;
	}
	return max;
}
109
EXPORT_SYMBOL_GPL(pci_bus_max_busnr);
L
Linus Torvalds 已提交
110

A
Andrew Morton 已提交
111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126
#ifdef CONFIG_HAS_IOMEM
void __iomem *pci_ioremap_bar(struct pci_dev *pdev, int bar)
{
	/*
	 * Make sure the BAR is actually a memory resource, not an IO resource
	 */
	if (!(pci_resource_flags(pdev, bar) & IORESOURCE_MEM)) {
		WARN_ON(1);
		return NULL;
	}
	return ioremap_nocache(pci_resource_start(pdev, bar),
				     pci_resource_len(pdev, bar));
}
EXPORT_SYMBOL_GPL(pci_ioremap_bar);
#endif

127
#if 0
L
Linus Torvalds 已提交
128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148
/**
 * pci_max_busnr - returns maximum PCI bus number
 *
 * Returns the highest PCI bus number present in the system global list of
 * PCI buses.
 */
unsigned char __devinit
pci_max_busnr(void)
{
	struct pci_bus *bus = NULL;
	unsigned char max, n;

	max = 0;
	while ((bus = pci_find_next_bus(bus)) != NULL) {
		n = pci_bus_max_busnr(bus);
		if(n > max)
			max = n;
	}
	return max;
}

149 150
#endif  /*  0  */

151 152 153 154
#define PCI_FIND_CAP_TTL	48

static int __pci_find_next_cap_ttl(struct pci_bus *bus, unsigned int devfn,
				   u8 pos, int cap, int *ttl)
155 156 157
{
	u8 id;

158
	while ((*ttl)--) {
159 160 161 162 163 164 165 166 167 168 169 170 171 172 173
		pci_bus_read_config_byte(bus, devfn, pos, &pos);
		if (pos < 0x40)
			break;
		pos &= ~3;
		pci_bus_read_config_byte(bus, devfn, pos + PCI_CAP_LIST_ID,
					 &id);
		if (id == 0xff)
			break;
		if (id == cap)
			return pos;
		pos += PCI_CAP_LIST_NEXT;
	}
	return 0;
}

174 175 176 177 178 179 180 181
static int __pci_find_next_cap(struct pci_bus *bus, unsigned int devfn,
			       u8 pos, int cap)
{
	int ttl = PCI_FIND_CAP_TTL;

	return __pci_find_next_cap_ttl(bus, devfn, pos, cap, &ttl);
}

182 183 184 185 186 187 188
int pci_find_next_capability(struct pci_dev *dev, u8 pos, int cap)
{
	return __pci_find_next_cap(dev->bus, dev->devfn,
				   pos + PCI_CAP_LIST_NEXT, cap);
}
EXPORT_SYMBOL_GPL(pci_find_next_capability);

189 190
static int __pci_bus_find_cap_start(struct pci_bus *bus,
				    unsigned int devfn, u8 hdr_type)
L
Linus Torvalds 已提交
191 192 193 194 195 196 197 198 199 200
{
	u16 status;

	pci_bus_read_config_word(bus, devfn, PCI_STATUS, &status);
	if (!(status & PCI_STATUS_CAP_LIST))
		return 0;

	switch (hdr_type) {
	case PCI_HEADER_TYPE_NORMAL:
	case PCI_HEADER_TYPE_BRIDGE:
201
		return PCI_CAPABILITY_LIST;
L
Linus Torvalds 已提交
202
	case PCI_HEADER_TYPE_CARDBUS:
203
		return PCI_CB_CAPABILITY_LIST;
L
Linus Torvalds 已提交
204 205 206
	default:
		return 0;
	}
207 208

	return 0;
L
Linus Torvalds 已提交
209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231
}

/**
 * pci_find_capability - query for devices' capabilities 
 * @dev: PCI device to query
 * @cap: capability code
 *
 * Tell if a device supports a given PCI capability.
 * Returns the address of the requested capability structure within the
 * device's PCI configuration space or 0 in case the device does not
 * support it.  Possible values for @cap:
 *
 *  %PCI_CAP_ID_PM           Power Management 
 *  %PCI_CAP_ID_AGP          Accelerated Graphics Port 
 *  %PCI_CAP_ID_VPD          Vital Product Data 
 *  %PCI_CAP_ID_SLOTID       Slot Identification 
 *  %PCI_CAP_ID_MSI          Message Signalled Interrupts
 *  %PCI_CAP_ID_CHSWP        CompactPCI HotSwap 
 *  %PCI_CAP_ID_PCIX         PCI-X
 *  %PCI_CAP_ID_EXP          PCI Express
 */
int pci_find_capability(struct pci_dev *dev, int cap)
{
232 233 234 235 236 237 238
	int pos;

	pos = __pci_bus_find_cap_start(dev->bus, dev->devfn, dev->hdr_type);
	if (pos)
		pos = __pci_find_next_cap(dev->bus, dev->devfn, pos, cap);

	return pos;
L
Linus Torvalds 已提交
239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255
}

/**
 * pci_bus_find_capability - query for devices' capabilities 
 * @bus:   the PCI bus to query
 * @devfn: PCI device to query
 * @cap:   capability code
 *
 * Like pci_find_capability() but works for pci devices that do not have a
 * pci_dev structure set up yet. 
 *
 * Returns the address of the requested capability structure within the
 * device's PCI configuration space or 0 in case the device does not
 * support it.
 */
int pci_bus_find_capability(struct pci_bus *bus, unsigned int devfn, int cap)
{
256
	int pos;
L
Linus Torvalds 已提交
257 258 259 260
	u8 hdr_type;

	pci_bus_read_config_byte(bus, devfn, PCI_HEADER_TYPE, &hdr_type);

261 262 263 264 265
	pos = __pci_bus_find_cap_start(bus, devfn, hdr_type & 0x7f);
	if (pos)
		pos = __pci_find_next_cap(bus, devfn, pos, cap);

	return pos;
L
Linus Torvalds 已提交
266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284
}

/**
 * pci_find_ext_capability - Find an extended capability
 * @dev: PCI device to query
 * @cap: capability code
 *
 * Returns the address of the requested extended capability structure
 * within the device's PCI configuration space or 0 if the device does
 * not support it.  Possible values for @cap:
 *
 *  %PCI_EXT_CAP_ID_ERR		Advanced Error Reporting
 *  %PCI_EXT_CAP_ID_VC		Virtual Channel
 *  %PCI_EXT_CAP_ID_DSN		Device Serial Number
 *  %PCI_EXT_CAP_ID_PWR		Power Budgeting
 */
int pci_find_ext_capability(struct pci_dev *dev, int cap)
{
	u32 header;
285 286
	int ttl;
	int pos = PCI_CFG_SPACE_SIZE;
L
Linus Torvalds 已提交
287

288 289 290 291
	/* minimum 8 bytes per capability */
	ttl = (PCI_CFG_SPACE_EXP_SIZE - PCI_CFG_SPACE_SIZE) / 8;

	if (dev->cfg_size <= PCI_CFG_SPACE_SIZE)
L
Linus Torvalds 已提交
292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308
		return 0;

	if (pci_read_config_dword(dev, pos, &header) != PCIBIOS_SUCCESSFUL)
		return 0;

	/*
	 * If we have no capabilities, this is indicated by cap ID,
	 * cap version and next pointer all being 0.
	 */
	if (header == 0)
		return 0;

	while (ttl-- > 0) {
		if (PCI_EXT_CAP_ID(header) == cap)
			return pos;

		pos = PCI_EXT_CAP_NEXT(header);
309
		if (pos < PCI_CFG_SPACE_SIZE)
L
Linus Torvalds 已提交
310 311 312 313 314 315 316 317
			break;

		if (pci_read_config_dword(dev, pos, &header) != PCIBIOS_SUCCESSFUL)
			break;
	}

	return 0;
}
318
EXPORT_SYMBOL_GPL(pci_find_ext_capability);
L
Linus Torvalds 已提交
319

320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362
/**
 * pci_bus_find_ext_capability - find an extended capability
 * @bus:   the PCI bus to query
 * @devfn: PCI device to query
 * @cap:   capability code
 *
 * Like pci_find_ext_capability() but works for pci devices that do not have a
 * pci_dev structure set up yet.
 *
 * Returns the address of the requested capability structure within the
 * device's PCI configuration space or 0 in case the device does not
 * support it.
 */
int pci_bus_find_ext_capability(struct pci_bus *bus, unsigned int devfn,
				int cap)
{
	u32 header;
	int ttl;
	int pos = PCI_CFG_SPACE_SIZE;

	/* minimum 8 bytes per capability */
	ttl = (PCI_CFG_SPACE_EXP_SIZE - PCI_CFG_SPACE_SIZE) / 8;

	if (!pci_bus_read_config_dword(bus, devfn, pos, &header))
		return 0;
	if (header == 0xffffffff || header == 0)
		return 0;

	while (ttl-- > 0) {
		if (PCI_EXT_CAP_ID(header) == cap)
			return pos;

		pos = PCI_EXT_CAP_NEXT(header);
		if (pos < PCI_CFG_SPACE_SIZE)
			break;

		if (!pci_bus_read_config_dword(bus, devfn, pos, &header))
			break;
	}

	return 0;
}

363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382
static int __pci_find_next_ht_cap(struct pci_dev *dev, int pos, int ht_cap)
{
	int rc, ttl = PCI_FIND_CAP_TTL;
	u8 cap, mask;

	if (ht_cap == HT_CAPTYPE_SLAVE || ht_cap == HT_CAPTYPE_HOST)
		mask = HT_3BIT_CAP_MASK;
	else
		mask = HT_5BIT_CAP_MASK;

	pos = __pci_find_next_cap_ttl(dev->bus, dev->devfn, pos,
				      PCI_CAP_ID_HT, &ttl);
	while (pos) {
		rc = pci_read_config_byte(dev, pos + 3, &cap);
		if (rc != PCIBIOS_SUCCESSFUL)
			return 0;

		if ((cap & mask) == ht_cap)
			return pos;

383 384
		pos = __pci_find_next_cap_ttl(dev->bus, dev->devfn,
					      pos + PCI_CAP_LIST_NEXT,
385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431
					      PCI_CAP_ID_HT, &ttl);
	}

	return 0;
}
/**
 * pci_find_next_ht_capability - query a device's Hypertransport capabilities
 * @dev: PCI device to query
 * @pos: Position from which to continue searching
 * @ht_cap: Hypertransport capability code
 *
 * To be used in conjunction with pci_find_ht_capability() to search for
 * all capabilities matching @ht_cap. @pos should always be a value returned
 * from pci_find_ht_capability().
 *
 * NB. To be 100% safe against broken PCI devices, the caller should take
 * steps to avoid an infinite loop.
 */
int pci_find_next_ht_capability(struct pci_dev *dev, int pos, int ht_cap)
{
	return __pci_find_next_ht_cap(dev, pos + PCI_CAP_LIST_NEXT, ht_cap);
}
EXPORT_SYMBOL_GPL(pci_find_next_ht_capability);

/**
 * pci_find_ht_capability - query a device's Hypertransport capabilities
 * @dev: PCI device to query
 * @ht_cap: Hypertransport capability code
 *
 * Tell if a device supports a given Hypertransport capability.
 * Returns an address within the device's PCI configuration space
 * or 0 in case the device does not support the request capability.
 * The address points to the PCI capability, of type PCI_CAP_ID_HT,
 * which has a Hypertransport capability matching @ht_cap.
 */
int pci_find_ht_capability(struct pci_dev *dev, int ht_cap)
{
	int pos;

	pos = __pci_bus_find_cap_start(dev->bus, dev->devfn, dev->hdr_type);
	if (pos)
		pos = __pci_find_next_ht_cap(dev, pos, ht_cap);

	return pos;
}
EXPORT_SYMBOL_GPL(pci_find_ht_capability);

L
Linus Torvalds 已提交
432 433 434 435 436 437 438 439 440 441 442 443 444 445
/**
 * pci_find_parent_resource - return resource region of parent bus of given region
 * @dev: PCI device structure contains resources to be searched
 * @res: child resource record for which parent is sought
 *
 *  For given resource region of given device, return the resource
 *  region of parent bus the given region is contained in or where
 *  it should be allocated from.
 */
struct resource *
pci_find_parent_resource(const struct pci_dev *dev, struct resource *res)
{
	const struct pci_bus *bus = dev->bus;
	int i;
446
	struct resource *best = NULL, *r;
L
Linus Torvalds 已提交
447

448
	pci_bus_for_each_resource(bus, r, i) {
L
Linus Torvalds 已提交
449 450 451 452 453 454 455 456
		if (!r)
			continue;
		if (res->start && !(res->start >= r->start && res->end <= r->end))
			continue;	/* Not contained */
		if ((res->flags ^ r->flags) & (IORESOURCE_IO | IORESOURCE_MEM))
			continue;	/* Wrong type */
		if (!((res->flags ^ r->flags) & IORESOURCE_PREFETCH))
			return r;	/* Exact match */
457 458 459 460 461 462
		/* We can't insert a non-prefetch resource inside a prefetchable parent .. */
		if (r->flags & IORESOURCE_PREFETCH)
			continue;
		/* .. but we can put a prefetchable resource inside a non-prefetchable one */
		if (!best)
			best = r;
L
Linus Torvalds 已提交
463 464 465 466
	}
	return best;
}

467 468 469 470 471 472 473
/**
 * pci_restore_bars - restore a devices BAR values (e.g. after wake-up)
 * @dev: PCI device to have its BARs restored
 *
 * Restore the BAR values for a given device, so as to make it
 * accessible by its driver.
 */
474
static void
475 476
pci_restore_bars(struct pci_dev *dev)
{
477
	int i;
478

479
	for (i = 0; i < PCI_BRIDGE_RESOURCES; i++)
480
		pci_update_resource(dev, i);
481 482
}

483 484 485 486
static struct pci_platform_pm_ops *pci_platform_pm;

int pci_set_platform_pm(struct pci_platform_pm_ops *ops)
{
487 488
	if (!ops->is_manageable || !ops->set_state || !ops->choose_state
	    || !ops->sleep_wake || !ops->can_wakeup)
489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509
		return -EINVAL;
	pci_platform_pm = ops;
	return 0;
}

static inline bool platform_pci_power_manageable(struct pci_dev *dev)
{
	return pci_platform_pm ? pci_platform_pm->is_manageable(dev) : false;
}

static inline int platform_pci_set_power_state(struct pci_dev *dev,
                                                pci_power_t t)
{
	return pci_platform_pm ? pci_platform_pm->set_state(dev, t) : -ENOSYS;
}

static inline pci_power_t platform_pci_choose_state(struct pci_dev *dev)
{
	return pci_platform_pm ?
			pci_platform_pm->choose_state(dev) : PCI_POWER_ERROR;
}
R
Randy Dunlap 已提交
510

511 512 513 514 515 516 517 518 519 520 521
static inline bool platform_pci_can_wakeup(struct pci_dev *dev)
{
	return pci_platform_pm ? pci_platform_pm->can_wakeup(dev) : false;
}

static inline int platform_pci_sleep_wake(struct pci_dev *dev, bool enable)
{
	return pci_platform_pm ?
			pci_platform_pm->sleep_wake(dev, enable) : -ENODEV;
}

522 523 524 525 526 527
static inline int platform_pci_run_wake(struct pci_dev *dev, bool enable)
{
	return pci_platform_pm ?
			pci_platform_pm->run_wake(dev, enable) : -ENODEV;
}

L
Linus Torvalds 已提交
528
/**
529 530 531 532
 * pci_raw_set_power_state - Use PCI PM registers to set the power state of
 *                           given PCI device
 * @dev: PCI device to handle.
 * @state: PCI power state (D0, D1, D2, D3hot) to put the device into.
L
Linus Torvalds 已提交
533
 *
534 535 536 537 538 539
 * RETURN VALUE:
 * -EINVAL if the requested state is invalid.
 * -EIO if device does not support PCI PM or its PM capabilities register has a
 * wrong version, or device doesn't support the requested state.
 * 0 if device already is in the requested state.
 * 0 if device's power state has been successfully changed.
L
Linus Torvalds 已提交
540
 */
541
static int pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state)
L
Linus Torvalds 已提交
542
{
543
	u16 pmcsr;
544
	bool need_restore = false;
L
Linus Torvalds 已提交
545

546 547 548 549
	/* Check if we're already there */
	if (dev->current_state == state)
		return 0;

550
	if (!dev->pm_cap)
551 552
		return -EIO;

553 554 555
	if (state < PCI_D0 || state > PCI_D3hot)
		return -EINVAL;

L
Linus Torvalds 已提交
556 557 558 559
	/* Validate current state:
	 * Can enter D0 from any state, but if we can only go deeper 
	 * to sleep if we're already in a low power state
	 */
560
	if (state != PCI_D0 && dev->current_state <= PCI_D3cold
561
	    && dev->current_state > state) {
562 563
		dev_err(&dev->dev, "invalid power transition "
			"(from state %d to %d)\n", dev->current_state, state);
L
Linus Torvalds 已提交
564
		return -EINVAL;
565
	}
L
Linus Torvalds 已提交
566 567

	/* check if this device supports the desired state */
568 569
	if ((state == PCI_D1 && !dev->d1_support)
	   || (state == PCI_D2 && !dev->d2_support))
570
		return -EIO;
L
Linus Torvalds 已提交
571

572
	pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
573

574
	/* If we're (effectively) in D3, force entire word to 0.
L
Linus Torvalds 已提交
575 576 577
	 * This doesn't affect PME_Status, disables PME_En, and
	 * sets PowerState to 0.
	 */
578
	switch (dev->current_state) {
579 580 581 582 583 584
	case PCI_D0:
	case PCI_D1:
	case PCI_D2:
		pmcsr &= ~PCI_PM_CTRL_STATE_MASK;
		pmcsr |= state;
		break;
585 586
	case PCI_D3hot:
	case PCI_D3cold:
587 588
	case PCI_UNKNOWN: /* Boot-up */
		if ((pmcsr & PCI_PM_CTRL_STATE_MASK) == PCI_D3hot
589
		 && !(pmcsr & PCI_PM_CTRL_NO_SOFT_RESET))
590
			need_restore = true;
591 592
		/* Fall-through: force to D0 */
	default:
593
		pmcsr = 0;
594
		break;
L
Linus Torvalds 已提交
595 596 597
	}

	/* enter specified state */
598
	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, pmcsr);
L
Linus Torvalds 已提交
599 600 601 602

	/* Mandatory power management transition delays */
	/* see PCI PM 1.1 5.6.1 table 18 */
	if (state == PCI_D3hot || dev->current_state == PCI_D3hot)
603
		pci_dev_d3_sleep(dev);
L
Linus Torvalds 已提交
604
	else if (state == PCI_D2 || dev->current_state == PCI_D2)
605
		udelay(PCI_PM_D2_DELAY);
L
Linus Torvalds 已提交
606

607 608 609 610 611
	pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
	dev->current_state = (pmcsr & PCI_PM_CTRL_STATE_MASK);
	if (dev->current_state != state && printk_ratelimit())
		dev_info(&dev->dev, "Refused to change power state, "
			"currently in D%d\n", dev->current_state);
612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627

	/* According to section 5.4.1 of the "PCI BUS POWER MANAGEMENT
	 * INTERFACE SPECIFICATION, REV. 1.2", a device transitioning
	 * from D3hot to D0 _may_ perform an internal reset, thereby
	 * going to "D0 Uninitialized" rather than "D0 Initialized".
	 * For example, at least some versions of the 3c905B and the
	 * 3c556B exhibit this behaviour.
	 *
	 * At least some laptop BIOSen (e.g. the Thinkpad T21) leave
	 * devices in a D3hot state at boot.  Consequently, we need to
	 * restore at least the BARs so that the device will be
	 * accessible to its driver.
	 */
	if (need_restore)
		pci_restore_bars(dev);

628
	if (dev->bus->self)
S
Shaohua Li 已提交
629 630
		pcie_aspm_pm_state_change(dev->bus->self);

L
Linus Torvalds 已提交
631 632 633
	return 0;
}

634 635 636 637
/**
 * pci_update_current_state - Read PCI power state of given device from its
 *                            PCI PM registers and cache it
 * @dev: PCI device to handle.
638
 * @state: State to cache in case the device doesn't have the PM capability
639
 */
640
void pci_update_current_state(struct pci_dev *dev, pci_power_t state)
641
{
642
	if (dev->pm_cap) {
643 644
		u16 pmcsr;

645
		pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
646
		dev->current_state = (pmcsr & PCI_PM_CTRL_STATE_MASK);
647 648
	} else {
		dev->current_state = state;
649 650 651
	}
}

652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667
/**
 * pci_platform_power_transition - Use platform to change device power state
 * @dev: PCI device to handle.
 * @state: State to put the device into.
 */
static int pci_platform_power_transition(struct pci_dev *dev, pci_power_t state)
{
	int error;

	if (platform_pci_power_manageable(dev)) {
		error = platform_pci_set_power_state(dev, state);
		if (!error)
			pci_update_current_state(dev, state);
	} else {
		error = -ENODEV;
		/* Fall back to PCI_D0 if native PM is not supported */
668 669
		if (!dev->pm_cap)
			dev->current_state = PCI_D0;
670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694
	}

	return error;
}

/**
 * __pci_start_power_transition - Start power transition of a PCI device
 * @dev: PCI device to handle.
 * @state: State to put the device into.
 */
static void __pci_start_power_transition(struct pci_dev *dev, pci_power_t state)
{
	if (state == PCI_D0)
		pci_platform_power_transition(dev, PCI_D0);
}

/**
 * __pci_complete_power_transition - Complete power transition of a PCI device
 * @dev: PCI device to handle.
 * @state: State to put the device into.
 *
 * This function should not be called directly by device drivers.
 */
int __pci_complete_power_transition(struct pci_dev *dev, pci_power_t state)
{
695
	return state >= PCI_D0 ?
696 697 698 699
			pci_platform_power_transition(dev, state) : -EINVAL;
}
EXPORT_SYMBOL_GPL(__pci_complete_power_transition);

700 701 702 703 704
/**
 * pci_set_power_state - Set the power state of a PCI device
 * @dev: PCI device to handle.
 * @state: PCI power state (D0, D1, D2, D3hot) to put the device into.
 *
705
 * Transition a device to a new power state, using the platform firmware and/or
706 707 708 709 710 711 712 713 714 715 716
 * the device's PCI PM registers.
 *
 * RETURN VALUE:
 * -EINVAL if the requested state is invalid.
 * -EIO if device does not support PCI PM or its PM capabilities register has a
 * wrong version, or device doesn't support the requested state.
 * 0 if device already is in the requested state.
 * 0 if device's power state has been successfully changed.
 */
int pci_set_power_state(struct pci_dev *dev, pci_power_t state)
{
717
	int error;
718 719 720 721 722 723 724 725 726 727 728 729 730 731

	/* bound the state we're entering */
	if (state > PCI_D3hot)
		state = PCI_D3hot;
	else if (state < PCI_D0)
		state = PCI_D0;
	else if ((state == PCI_D1 || state == PCI_D2) && pci_no_d1d2(dev))
		/*
		 * If the device or the parent bridge do not support PCI PM,
		 * ignore the request if we're doing anything other than putting
		 * it into D0 (which would only happen on boot).
		 */
		return 0;

732 733
	__pci_start_power_transition(dev, state);

734 735 736 737
	/* This device is quirked not to be put into D3, so
	   don't put it in D3 */
	if (state == PCI_D3hot && (dev->dev_flags & PCI_DEV_FLAGS_NO_D3))
		return 0;
738

739
	error = pci_raw_set_power_state(dev, state);
740

741 742
	if (!__pci_complete_power_transition(dev, state))
		error = 0;
743 744 745 746 747 748
	/*
	 * When aspm_policy is "powersave" this call ensures
	 * that ASPM is configured.
	 */
	if (!error && dev->bus->self)
		pcie_aspm_powersave_config_link(dev->bus->self);
749 750 751 752

	return error;
}

L
Linus Torvalds 已提交
753 754 755 756 757 758 759 760 761 762 763 764
/**
 * pci_choose_state - Choose the power state of a PCI device
 * @dev: PCI device to be suspended
 * @state: target sleep state for the whole system. This is the value
 *	that is passed to suspend() function.
 *
 * Returns PCI power state suitable for given device and given system
 * message.
 */

pci_power_t pci_choose_state(struct pci_dev *dev, pm_message_t state)
{
765
	pci_power_t ret;
766

L
Linus Torvalds 已提交
767 768 769
	if (!pci_find_capability(dev, PCI_CAP_ID_PM))
		return PCI_D0;

770 771 772
	ret = platform_pci_choose_state(dev);
	if (ret != PCI_POWER_ERROR)
		return ret;
773 774 775 776 777

	switch (state.event) {
	case PM_EVENT_ON:
		return PCI_D0;
	case PM_EVENT_FREEZE:
778 779
	case PM_EVENT_PRETHAW:
		/* REVISIT both freeze and pre-thaw "should" use D0 */
780
	case PM_EVENT_SUSPEND:
781
	case PM_EVENT_HIBERNATE:
782
		return PCI_D3hot;
L
Linus Torvalds 已提交
783
	default:
784 785
		dev_info(&dev->dev, "unrecognized suspend event %d\n",
			 state.event);
L
Linus Torvalds 已提交
786 787 788 789 790 791 792
		BUG();
	}
	return PCI_D0;
}

EXPORT_SYMBOL(pci_choose_state);

793 794
#define PCI_EXP_SAVE_REGS	7

795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816
#define pcie_cap_has_devctl(type, flags)	1
#define pcie_cap_has_lnkctl(type, flags)		\
		((flags & PCI_EXP_FLAGS_VERS) > 1 ||	\
		 (type == PCI_EXP_TYPE_ROOT_PORT ||	\
		  type == PCI_EXP_TYPE_ENDPOINT ||	\
		  type == PCI_EXP_TYPE_LEG_END))
#define pcie_cap_has_sltctl(type, flags)		\
		((flags & PCI_EXP_FLAGS_VERS) > 1 ||	\
		 ((type == PCI_EXP_TYPE_ROOT_PORT) ||	\
		  (type == PCI_EXP_TYPE_DOWNSTREAM &&	\
		   (flags & PCI_EXP_FLAGS_SLOT))))
#define pcie_cap_has_rtctl(type, flags)			\
		((flags & PCI_EXP_FLAGS_VERS) > 1 ||	\
		 (type == PCI_EXP_TYPE_ROOT_PORT ||	\
		  type == PCI_EXP_TYPE_RC_EC))
#define pcie_cap_has_devctl2(type, flags)		\
		((flags & PCI_EXP_FLAGS_VERS) > 1)
#define pcie_cap_has_lnkctl2(type, flags)		\
		((flags & PCI_EXP_FLAGS_VERS) > 1)
#define pcie_cap_has_sltctl2(type, flags)		\
		((flags & PCI_EXP_FLAGS_VERS) > 1)

817 818 819 820 821
static int pci_save_pcie_state(struct pci_dev *dev)
{
	int pos, i = 0;
	struct pci_cap_saved_state *save_state;
	u16 *cap;
822
	u16 flags;
823

824 825
	pos = pci_pcie_cap(dev);
	if (!pos)
826 827
		return 0;

828
	save_state = pci_find_saved_cap(dev, PCI_CAP_ID_EXP);
829
	if (!save_state) {
830
		dev_err(&dev->dev, "buffer not found in %s\n", __func__);
831 832 833 834
		return -ENOMEM;
	}
	cap = (u16 *)&save_state->data[0];

835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850
	pci_read_config_word(dev, pos + PCI_EXP_FLAGS, &flags);

	if (pcie_cap_has_devctl(dev->pcie_type, flags))
		pci_read_config_word(dev, pos + PCI_EXP_DEVCTL, &cap[i++]);
	if (pcie_cap_has_lnkctl(dev->pcie_type, flags))
		pci_read_config_word(dev, pos + PCI_EXP_LNKCTL, &cap[i++]);
	if (pcie_cap_has_sltctl(dev->pcie_type, flags))
		pci_read_config_word(dev, pos + PCI_EXP_SLTCTL, &cap[i++]);
	if (pcie_cap_has_rtctl(dev->pcie_type, flags))
		pci_read_config_word(dev, pos + PCI_EXP_RTCTL, &cap[i++]);
	if (pcie_cap_has_devctl2(dev->pcie_type, flags))
		pci_read_config_word(dev, pos + PCI_EXP_DEVCTL2, &cap[i++]);
	if (pcie_cap_has_lnkctl2(dev->pcie_type, flags))
		pci_read_config_word(dev, pos + PCI_EXP_LNKCTL2, &cap[i++]);
	if (pcie_cap_has_sltctl2(dev->pcie_type, flags))
		pci_read_config_word(dev, pos + PCI_EXP_SLTCTL2, &cap[i++]);
851

852 853 854 855 856 857 858 859
	return 0;
}

static void pci_restore_pcie_state(struct pci_dev *dev)
{
	int i = 0, pos;
	struct pci_cap_saved_state *save_state;
	u16 *cap;
860
	u16 flags;
861 862 863 864 865 866 867

	save_state = pci_find_saved_cap(dev, PCI_CAP_ID_EXP);
	pos = pci_find_capability(dev, PCI_CAP_ID_EXP);
	if (!save_state || pos <= 0)
		return;
	cap = (u16 *)&save_state->data[0];

868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883
	pci_read_config_word(dev, pos + PCI_EXP_FLAGS, &flags);

	if (pcie_cap_has_devctl(dev->pcie_type, flags))
		pci_write_config_word(dev, pos + PCI_EXP_DEVCTL, cap[i++]);
	if (pcie_cap_has_lnkctl(dev->pcie_type, flags))
		pci_write_config_word(dev, pos + PCI_EXP_LNKCTL, cap[i++]);
	if (pcie_cap_has_sltctl(dev->pcie_type, flags))
		pci_write_config_word(dev, pos + PCI_EXP_SLTCTL, cap[i++]);
	if (pcie_cap_has_rtctl(dev->pcie_type, flags))
		pci_write_config_word(dev, pos + PCI_EXP_RTCTL, cap[i++]);
	if (pcie_cap_has_devctl2(dev->pcie_type, flags))
		pci_write_config_word(dev, pos + PCI_EXP_DEVCTL2, cap[i++]);
	if (pcie_cap_has_lnkctl2(dev->pcie_type, flags))
		pci_write_config_word(dev, pos + PCI_EXP_LNKCTL2, cap[i++]);
	if (pcie_cap_has_sltctl2(dev->pcie_type, flags))
		pci_write_config_word(dev, pos + PCI_EXP_SLTCTL2, cap[i++]);
884 885
}

S
Stephen Hemminger 已提交
886 887 888

static int pci_save_pcix_state(struct pci_dev *dev)
{
889
	int pos;
S
Stephen Hemminger 已提交
890 891 892 893 894 895
	struct pci_cap_saved_state *save_state;

	pos = pci_find_capability(dev, PCI_CAP_ID_PCIX);
	if (pos <= 0)
		return 0;

896
	save_state = pci_find_saved_cap(dev, PCI_CAP_ID_PCIX);
S
Stephen Hemminger 已提交
897
	if (!save_state) {
898
		dev_err(&dev->dev, "buffer not found in %s\n", __func__);
S
Stephen Hemminger 已提交
899 900 901
		return -ENOMEM;
	}

902 903
	pci_read_config_word(dev, pos + PCI_X_CMD, (u16 *)save_state->data);

S
Stephen Hemminger 已提交
904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922
	return 0;
}

static void pci_restore_pcix_state(struct pci_dev *dev)
{
	int i = 0, pos;
	struct pci_cap_saved_state *save_state;
	u16 *cap;

	save_state = pci_find_saved_cap(dev, PCI_CAP_ID_PCIX);
	pos = pci_find_capability(dev, PCI_CAP_ID_PCIX);
	if (!save_state || pos <= 0)
		return;
	cap = (u16 *)&save_state->data[0];

	pci_write_config_word(dev, pos + PCI_X_CMD, cap[i++]);
}


L
Linus Torvalds 已提交
923 924 925 926 927 928 929 930 931 932
/**
 * pci_save_state - save the PCI configuration space of a device before suspending
 * @dev: - PCI device that we're dealing with
 */
int
pci_save_state(struct pci_dev *dev)
{
	int i;
	/* XXX: 100% dword access ok here? */
	for (i = 0; i < 16; i++)
933
		pci_read_config_dword(dev, i * 4, &dev->saved_config_space[i]);
934
	dev->state_saved = true;
935 936
	if ((i = pci_save_pcie_state(dev)) != 0)
		return i;
S
Stephen Hemminger 已提交
937 938
	if ((i = pci_save_pcix_state(dev)) != 0)
		return i;
L
Linus Torvalds 已提交
939 940 941 942 943 944 945
	return 0;
}

/** 
 * pci_restore_state - Restore the saved state of a PCI device
 * @dev: - PCI device that we're dealing with
 */
946
void pci_restore_state(struct pci_dev *dev)
L
Linus Torvalds 已提交
947 948
{
	int i;
A
Al Viro 已提交
949
	u32 val;
L
Linus Torvalds 已提交
950

A
Alek Du 已提交
951
	if (!dev->state_saved)
952
		return;
953

954 955 956
	/* PCI Express register must be restored first */
	pci_restore_pcie_state(dev);

957 958 959 960 961
	/*
	 * The Base Address register should be programmed before the command
	 * register(s)
	 */
	for (i = 15; i >= 0; i--) {
962 963
		pci_read_config_dword(dev, i * 4, &val);
		if (val != dev->saved_config_space[i]) {
964 965 966
			dev_printk(KERN_DEBUG, &dev->dev, "restoring config "
				"space at offset %#x (was %#x, writing %#x)\n",
				i, val, (int)dev->saved_config_space[i]);
967 968 969 970
			pci_write_config_dword(dev,i * 4,
				dev->saved_config_space[i]);
		}
	}
S
Stephen Hemminger 已提交
971
	pci_restore_pcix_state(dev);
972
	pci_restore_msi_state(dev);
Y
Yu Zhao 已提交
973
	pci_restore_iov_state(dev);
974

975
	dev->state_saved = false;
L
Linus Torvalds 已提交
976 977
}

978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993
static int do_pci_enable_device(struct pci_dev *dev, int bars)
{
	int err;

	err = pci_set_power_state(dev, PCI_D0);
	if (err < 0 && err != -EIO)
		return err;
	err = pcibios_enable_device(dev, bars);
	if (err < 0)
		return err;
	pci_fixup_device(pci_fixup_enable, dev);

	return 0;
}

/**
994
 * pci_reenable_device - Resume abandoned device
995 996 997 998 999
 * @dev: PCI device to be resumed
 *
 *  Note this function is a backend of pci_default_resume and is not supposed
 *  to be called by normal code, write proper resume handler and use it instead.
 */
1000
int pci_reenable_device(struct pci_dev *dev)
1001
{
1002
	if (pci_is_enabled(dev))
1003 1004 1005 1006
		return do_pci_enable_device(dev, (1 << PCI_NUM_RESOURCES) - 1);
	return 0;
}

1007 1008
static int __pci_enable_device_flags(struct pci_dev *dev,
				     resource_size_t flags)
L
Linus Torvalds 已提交
1009 1010
{
	int err;
1011
	int i, bars = 0;
L
Linus Torvalds 已提交
1012

1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024
	/*
	 * Power state could be unknown at this point, either due to a fresh
	 * boot or a device removal call.  So get the current power state
	 * so that things like MSI message writing will behave as expected
	 * (e.g. if the device really is in D0 at enable time).
	 */
	if (dev->pm_cap) {
		u16 pmcsr;
		pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
		dev->current_state = (pmcsr & PCI_PM_CTRL_STATE_MASK);
	}

1025 1026 1027
	if (atomic_add_return(1, &dev->enable_cnt) > 1)
		return 0;		/* already enabled */

1028 1029 1030 1031
	for (i = 0; i < DEVICE_COUNT_RESOURCE; i++)
		if (dev->resource[i].flags & flags)
			bars |= (1 << i);

1032
	err = do_pci_enable_device(dev, bars);
1033
	if (err < 0)
1034
		atomic_dec(&dev->enable_cnt);
1035
	return err;
L
Linus Torvalds 已提交
1036 1037
}

1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063
/**
 * pci_enable_device_io - Initialize a device for use with IO space
 * @dev: PCI device to be initialized
 *
 *  Initialize device before it's used by a driver. Ask low-level code
 *  to enable I/O resources. Wake up the device if it was suspended.
 *  Beware, this function can fail.
 */
int pci_enable_device_io(struct pci_dev *dev)
{
	return __pci_enable_device_flags(dev, IORESOURCE_IO);
}

/**
 * pci_enable_device_mem - Initialize a device for use with Memory space
 * @dev: PCI device to be initialized
 *
 *  Initialize device before it's used by a driver. Ask low-level code
 *  to enable Memory resources. Wake up the device if it was suspended.
 *  Beware, this function can fail.
 */
int pci_enable_device_mem(struct pci_dev *dev)
{
	return __pci_enable_device_flags(dev, IORESOURCE_MEM);
}

1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076
/**
 * pci_enable_device - Initialize device before it's used by a driver.
 * @dev: PCI device to be initialized
 *
 *  Initialize device before it's used by a driver. Ask low-level code
 *  to enable I/O and memory. Wake up the device if it was suspended.
 *  Beware, this function can fail.
 *
 *  Note we don't actually enable the device many times if we call
 *  this function repeatedly (we just increment the count).
 */
int pci_enable_device(struct pci_dev *dev)
{
1077
	return __pci_enable_device_flags(dev, IORESOURCE_MEM | IORESOURCE_IO);
1078 1079
}

T
Tejun Heo 已提交
1080 1081 1082 1083 1084 1085 1086
/*
 * Managed PCI resources.  This manages device on/off, intx/msi/msix
 * on/off and BAR regions.  pci_dev itself records msi/msix status, so
 * there's no need to track it separately.  pci_devres is initialized
 * when a device is enabled using managed PCI device enable interface.
 */
struct pci_devres {
1087 1088
	unsigned int enabled:1;
	unsigned int pinned:1;
T
Tejun Heo 已提交
1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111
	unsigned int orig_intx:1;
	unsigned int restore_intx:1;
	u32 region_mask;
};

static void pcim_release(struct device *gendev, void *res)
{
	struct pci_dev *dev = container_of(gendev, struct pci_dev, dev);
	struct pci_devres *this = res;
	int i;

	if (dev->msi_enabled)
		pci_disable_msi(dev);
	if (dev->msix_enabled)
		pci_disable_msix(dev);

	for (i = 0; i < DEVICE_COUNT_RESOURCE; i++)
		if (this->region_mask & (1 << i))
			pci_release_region(dev, i);

	if (this->restore_intx)
		pci_intx(dev, this->orig_intx);

1112
	if (this->enabled && !this->pinned)
T
Tejun Heo 已提交
1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150
		pci_disable_device(dev);
}

static struct pci_devres * get_pci_dr(struct pci_dev *pdev)
{
	struct pci_devres *dr, *new_dr;

	dr = devres_find(&pdev->dev, pcim_release, NULL, NULL);
	if (dr)
		return dr;

	new_dr = devres_alloc(pcim_release, sizeof(*new_dr), GFP_KERNEL);
	if (!new_dr)
		return NULL;
	return devres_get(&pdev->dev, new_dr, NULL, NULL);
}

static struct pci_devres * find_pci_dr(struct pci_dev *pdev)
{
	if (pci_is_managed(pdev))
		return devres_find(&pdev->dev, pcim_release, NULL, NULL);
	return NULL;
}

/**
 * pcim_enable_device - Managed pci_enable_device()
 * @pdev: PCI device to be initialized
 *
 * Managed pci_enable_device().
 */
int pcim_enable_device(struct pci_dev *pdev)
{
	struct pci_devres *dr;
	int rc;

	dr = get_pci_dr(pdev);
	if (unlikely(!dr))
		return -ENOMEM;
1151 1152
	if (dr->enabled)
		return 0;
T
Tejun Heo 已提交
1153 1154 1155 1156

	rc = pci_enable_device(pdev);
	if (!rc) {
		pdev->is_managed = 1;
1157
		dr->enabled = 1;
T
Tejun Heo 已提交
1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174
	}
	return rc;
}

/**
 * pcim_pin_device - Pin managed PCI device
 * @pdev: PCI device to pin
 *
 * Pin managed PCI device @pdev.  Pinned device won't be disabled on
 * driver detach.  @pdev must have been enabled with
 * pcim_enable_device().
 */
void pcim_pin_device(struct pci_dev *pdev)
{
	struct pci_devres *dr;

	dr = find_pci_dr(pdev);
1175
	WARN_ON(!dr || !dr->enabled);
T
Tejun Heo 已提交
1176
	if (dr)
1177
		dr->pinned = 1;
T
Tejun Heo 已提交
1178 1179
}

L
Linus Torvalds 已提交
1180 1181 1182 1183 1184 1185 1186 1187 1188 1189
/**
 * pcibios_disable_device - disable arch specific PCI resources for device dev
 * @dev: the PCI device to disable
 *
 * Disables architecture specific PCI resources for the device. This
 * is the default implementation. Architecture implementations can
 * override this.
 */
void __attribute__ ((weak)) pcibios_disable_device (struct pci_dev *dev) {}

1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211
static void do_pci_disable_device(struct pci_dev *dev)
{
	u16 pci_command;

	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
	if (pci_command & PCI_COMMAND_MASTER) {
		pci_command &= ~PCI_COMMAND_MASTER;
		pci_write_config_word(dev, PCI_COMMAND, pci_command);
	}

	pcibios_disable_device(dev);
}

/**
 * pci_disable_enabled_device - Disable device without updating enable_cnt
 * @dev: PCI device to disable
 *
 * NOTE: This function is a backend of PCI power management routines and is
 * not supposed to be called drivers.
 */
void pci_disable_enabled_device(struct pci_dev *dev)
{
1212
	if (pci_is_enabled(dev))
1213 1214 1215
		do_pci_disable_device(dev);
}

L
Linus Torvalds 已提交
1216 1217 1218 1219 1220 1221
/**
 * pci_disable_device - Disable PCI device after use
 * @dev: PCI device to be disabled
 *
 * Signal to the system that the PCI device is not in use by the system
 * anymore.  This only involves disabling PCI bus-mastering, if active.
1222 1223
 *
 * Note we don't actually disable the device until all callers of
1224
 * pci_enable_device() have called pci_disable_device().
L
Linus Torvalds 已提交
1225 1226 1227 1228
 */
void
pci_disable_device(struct pci_dev *dev)
{
T
Tejun Heo 已提交
1229
	struct pci_devres *dr;
1230

T
Tejun Heo 已提交
1231 1232
	dr = find_pci_dr(dev);
	if (dr)
1233
		dr->enabled = 0;
T
Tejun Heo 已提交
1234

1235 1236 1237
	if (atomic_sub_return(1, &dev->enable_cnt) != 0)
		return;

1238
	do_pci_disable_device(dev);
L
Linus Torvalds 已提交
1239

1240
	dev->is_busmaster = 0;
L
Linus Torvalds 已提交
1241 1242
}

B
Brian King 已提交
1243 1244
/**
 * pcibios_set_pcie_reset_state - set reset state for device dev
1245
 * @dev: the PCIe device reset
B
Brian King 已提交
1246 1247 1248
 * @state: Reset state to enter into
 *
 *
1249
 * Sets the PCIe reset state for the device. This is the default
B
Brian King 已提交
1250 1251 1252 1253 1254 1255 1256 1257 1258 1259
 * implementation. Architecture implementations can override this.
 */
int __attribute__ ((weak)) pcibios_set_pcie_reset_state(struct pci_dev *dev,
							enum pcie_reset_state state)
{
	return -EINVAL;
}

/**
 * pci_set_pcie_reset_state - set reset state for device dev
1260
 * @dev: the PCIe device reset
B
Brian King 已提交
1261 1262 1263 1264 1265 1266 1267 1268 1269 1270
 * @state: Reset state to enter into
 *
 *
 * Sets the PCI reset state for the device.
 */
int pci_set_pcie_reset_state(struct pci_dev *dev, enum pcie_reset_state state)
{
	return pcibios_set_pcie_reset_state(dev, state);
}

1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305
/**
 * pci_check_pme_status - Check if given device has generated PME.
 * @dev: Device to check.
 *
 * Check the PME status of the device and if set, clear it and clear PME enable
 * (if set).  Return 'true' if PME status and PME enable were both set or
 * 'false' otherwise.
 */
bool pci_check_pme_status(struct pci_dev *dev)
{
	int pmcsr_pos;
	u16 pmcsr;
	bool ret = false;

	if (!dev->pm_cap)
		return false;

	pmcsr_pos = dev->pm_cap + PCI_PM_CTRL;
	pci_read_config_word(dev, pmcsr_pos, &pmcsr);
	if (!(pmcsr & PCI_PM_CTRL_PME_STATUS))
		return false;

	/* Clear PME status. */
	pmcsr |= PCI_PM_CTRL_PME_STATUS;
	if (pmcsr & PCI_PM_CTRL_PME_ENABLE) {
		/* Disable PME to avoid interrupt flood. */
		pmcsr &= ~PCI_PM_CTRL_PME_ENABLE;
		ret = true;
	}

	pci_write_config_word(dev, pmcsr_pos, pmcsr);

	return ret;
}

1306 1307 1308 1309 1310 1311 1312 1313 1314 1315
/**
 * pci_pme_wakeup - Wake up a PCI device if its PME Status bit is set.
 * @dev: Device to handle.
 * @ign: Ignored.
 *
 * Check if @dev has generated PME and queue a resume request for it in that
 * case.
 */
static int pci_pme_wakeup(struct pci_dev *dev, void *ign)
{
1316 1317
	if (pci_check_pme_status(dev)) {
		pci_wakeup_event(dev);
1318
		pm_request_resume(&dev->dev);
1319
	}
1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332
	return 0;
}

/**
 * pci_pme_wakeup_bus - Walk given bus and wake up devices on it, if necessary.
 * @bus: Top bus of the subtree to walk.
 */
void pci_pme_wakeup_bus(struct pci_bus *bus)
{
	if (bus)
		pci_walk_bus(bus, pci_pme_wakeup, NULL);
}

1333 1334 1335 1336 1337
/**
 * pci_pme_capable - check the capability of PCI device to generate PME#
 * @dev: PCI device to handle.
 * @state: PCI state from which device will issue PME#.
 */
1338
bool pci_pme_capable(struct pci_dev *dev, pci_power_t state)
1339
{
1340
	if (!dev->pm_cap)
1341 1342
		return false;

1343
	return !!(dev->pme_support & (1 << state));
1344 1345
}

1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371
static void pci_pme_list_scan(struct work_struct *work)
{
	struct pci_pme_device *pme_dev;

	mutex_lock(&pci_pme_list_mutex);
	if (!list_empty(&pci_pme_list)) {
		list_for_each_entry(pme_dev, &pci_pme_list, list)
			pci_pme_wakeup(pme_dev->dev, NULL);
		schedule_delayed_work(&pci_pme_work, msecs_to_jiffies(PME_TIMEOUT));
	}
	mutex_unlock(&pci_pme_list_mutex);
}

/**
 * pci_external_pme - is a device an external PCI PME source?
 * @dev: PCI device to check
 *
 */

static bool pci_external_pme(struct pci_dev *dev)
{
	if (pci_is_pcie(dev) || dev->bus->number == 0)
		return false;
	return true;
}

1372 1373 1374 1375 1376 1377 1378 1379
/**
 * pci_pme_active - enable or disable PCI device's PME# function
 * @dev: PCI device to handle.
 * @enable: 'true' to enable PME# generation; 'false' to disable it.
 *
 * The caller must verify that the device is capable of generating PME# before
 * calling this function with @enable equal to 'true'.
 */
1380
void pci_pme_active(struct pci_dev *dev, bool enable)
1381 1382 1383
{
	u16 pmcsr;

1384
	if (!dev->pm_cap)
1385 1386
		return;

1387
	pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
1388 1389 1390 1391 1392
	/* Clear PME_Status by writing 1 to it and enable PME# */
	pmcsr |= PCI_PM_CTRL_PME_STATUS | PCI_PM_CTRL_PME_ENABLE;
	if (!enable)
		pmcsr &= ~PCI_PM_CTRL_PME_ENABLE;

1393
	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, pmcsr);
1394

1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432
	/* PCI (as opposed to PCIe) PME requires that the device have
	   its PME# line hooked up correctly. Not all hardware vendors
	   do this, so the PME never gets delivered and the device
	   remains asleep. The easiest way around this is to
	   periodically walk the list of suspended devices and check
	   whether any have their PME flag set. The assumption is that
	   we'll wake up often enough anyway that this won't be a huge
	   hit, and the power savings from the devices will still be a
	   win. */

	if (pci_external_pme(dev)) {
		struct pci_pme_device *pme_dev;
		if (enable) {
			pme_dev = kmalloc(sizeof(struct pci_pme_device),
					  GFP_KERNEL);
			if (!pme_dev)
				goto out;
			pme_dev->dev = dev;
			mutex_lock(&pci_pme_list_mutex);
			list_add(&pme_dev->list, &pci_pme_list);
			if (list_is_singular(&pci_pme_list))
				schedule_delayed_work(&pci_pme_work,
						      msecs_to_jiffies(PME_TIMEOUT));
			mutex_unlock(&pci_pme_list_mutex);
		} else {
			mutex_lock(&pci_pme_list_mutex);
			list_for_each_entry(pme_dev, &pci_pme_list, list) {
				if (pme_dev->dev == dev) {
					list_del(&pme_dev->list);
					kfree(pme_dev);
					break;
				}
			}
			mutex_unlock(&pci_pme_list_mutex);
		}
	}

out:
B
Bjorn Helgaas 已提交
1433
	dev_printk(KERN_DEBUG, &dev->dev, "PME# %s\n",
1434 1435 1436
			enable ? "enabled" : "disabled");
}

L
Linus Torvalds 已提交
1437
/**
1438
 * __pci_enable_wake - enable PCI device as wakeup event source
1439 1440
 * @dev: PCI device affected
 * @state: PCI state from which device will issue wakeup events
1441
 * @runtime: True if the events are to be generated at run time
1442 1443 1444 1445 1446 1447 1448
 * @enable: True to enable event generation; false to disable
 *
 * This enables the device as a wakeup event source, or disables it.
 * When such events involves platform-specific hooks, those hooks are
 * called automatically by this routine.
 *
 * Devices with legacy power management (no standard PCI PM capabilities)
1449
 * always require such platform hooks.
1450
 *
1451 1452 1453 1454 1455
 * RETURN VALUE:
 * 0 is returned on success
 * -EINVAL is returned if device is not supposed to wake up the system
 * Error code depending on the platform is returned if both the platform and
 * the native mechanism fail to enable the generation of wake-up events
L
Linus Torvalds 已提交
1456
 */
1457 1458
int __pci_enable_wake(struct pci_dev *dev, pci_power_t state,
		      bool runtime, bool enable)
L
Linus Torvalds 已提交
1459
{
1460
	int ret = 0;
1461

1462
	if (enable && !runtime && !device_may_wakeup(&dev->dev))
1463
		return -EINVAL;
L
Linus Torvalds 已提交
1464

1465 1466 1467 1468
	/* Don't do the same thing twice in a row for one device. */
	if (!!enable == !!dev->wakeup_prepared)
		return 0;

1469 1470 1471 1472
	/*
	 * According to "PCI System Architecture" 4th ed. by Tom Shanley & Don
	 * Anderson we should be doing PME# wake enable followed by ACPI wake
	 * enable.  To disable wake-up we call the platform first, for symmetry.
1473
	 */
L
Linus Torvalds 已提交
1474

1475 1476
	if (enable) {
		int error;
L
Linus Torvalds 已提交
1477

1478 1479 1480 1481
		if (pci_pme_capable(dev, state))
			pci_pme_active(dev, true);
		else
			ret = 1;
1482 1483
		error = runtime ? platform_pci_run_wake(dev, true) :
					platform_pci_sleep_wake(dev, true);
1484 1485
		if (ret)
			ret = error;
1486 1487
		if (!ret)
			dev->wakeup_prepared = true;
1488
	} else {
1489 1490 1491 1492
		if (runtime)
			platform_pci_run_wake(dev, false);
		else
			platform_pci_sleep_wake(dev, false);
1493
		pci_pme_active(dev, false);
1494
		dev->wakeup_prepared = false;
1495
	}
L
Linus Torvalds 已提交
1496

1497
	return ret;
1498
}
1499
EXPORT_SYMBOL(__pci_enable_wake);
L
Linus Torvalds 已提交
1500

1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521
/**
 * pci_wake_from_d3 - enable/disable device to wake up from D3_hot or D3_cold
 * @dev: PCI device to prepare
 * @enable: True to enable wake-up event generation; false to disable
 *
 * Many drivers want the device to wake up the system from D3_hot or D3_cold
 * and this function allows them to set that up cleanly - pci_enable_wake()
 * should not be called twice in a row to enable wake-up due to PCI PM vs ACPI
 * ordering constraints.
 *
 * This function only returns error code if the device is not capable of
 * generating PME# from both D3_hot and D3_cold, and the platform is unable to
 * enable wake-up power for it.
 */
int pci_wake_from_d3(struct pci_dev *dev, bool enable)
{
	return pci_pme_capable(dev, PCI_D3cold) ?
			pci_enable_wake(dev, PCI_D3cold, enable) :
			pci_enable_wake(dev, PCI_D3hot, enable);
}

1522
/**
J
Jesse Barnes 已提交
1523 1524 1525 1526 1527 1528
 * pci_target_state - find an appropriate low power state for a given PCI dev
 * @dev: PCI device
 *
 * Use underlying platform code to find a supported low power state for @dev.
 * If the platform can't manage @dev, return the deepest state from which it
 * can generate wake events, based on any available PME info.
1529
 */
1530
pci_power_t pci_target_state(struct pci_dev *dev)
1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551
{
	pci_power_t target_state = PCI_D3hot;

	if (platform_pci_power_manageable(dev)) {
		/*
		 * Call the platform to choose the target state of the device
		 * and enable wake-up from this state if supported.
		 */
		pci_power_t state = platform_pci_choose_state(dev);

		switch (state) {
		case PCI_POWER_ERROR:
		case PCI_UNKNOWN:
			break;
		case PCI_D1:
		case PCI_D2:
			if (pci_no_d1d2(dev))
				break;
		default:
			target_state = state;
		}
1552 1553
	} else if (!dev->pm_cap) {
		target_state = PCI_D0;
1554 1555 1556 1557 1558 1559
	} else if (device_may_wakeup(&dev->dev)) {
		/*
		 * Find the deepest state from which the device can generate
		 * wake-up events, make it the target state and enable device
		 * to generate PME#.
		 */
1560 1561 1562 1563
		if (dev->pme_support) {
			while (target_state
			      && !(dev->pme_support & (1 << target_state)))
				target_state--;
1564 1565 1566
		}
	}

1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585
	return target_state;
}

/**
 * pci_prepare_to_sleep - prepare PCI device for system-wide transition into a sleep state
 * @dev: Device to handle.
 *
 * Choose the power state appropriate for the device depending on whether
 * it can wake up the system and/or is power manageable by the platform
 * (PCI_D3hot is the default) and put the device into that state.
 */
int pci_prepare_to_sleep(struct pci_dev *dev)
{
	pci_power_t target_state = pci_target_state(dev);
	int error;

	if (target_state == PCI_POWER_ERROR)
		return -EIO;

1586
	pci_enable_wake(dev, target_state, device_may_wakeup(&dev->dev));
1587

1588 1589 1590 1591 1592 1593 1594 1595 1596
	error = pci_set_power_state(dev, target_state);

	if (error)
		pci_enable_wake(dev, target_state, false);

	return error;
}

/**
R
Randy Dunlap 已提交
1597
 * pci_back_from_sleep - turn PCI device on during system-wide transition into working state
1598 1599
 * @dev: Device to handle.
 *
T
Thomas Weber 已提交
1600
 * Disable device's system wake-up capability and put it into D0.
1601 1602 1603 1604 1605 1606 1607
 */
int pci_back_from_sleep(struct pci_dev *dev)
{
	pci_enable_wake(dev, PCI_D0, false);
	return pci_set_power_state(dev, PCI_D0);
}

1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632
/**
 * pci_finish_runtime_suspend - Carry out PCI-specific part of runtime suspend.
 * @dev: PCI device being suspended.
 *
 * Prepare @dev to generate wake-up events at run time and put it into a low
 * power state.
 */
int pci_finish_runtime_suspend(struct pci_dev *dev)
{
	pci_power_t target_state = pci_target_state(dev);
	int error;

	if (target_state == PCI_POWER_ERROR)
		return -EIO;

	__pci_enable_wake(dev, target_state, true, pci_dev_run_wake(dev));

	error = pci_set_power_state(dev, target_state);

	if (error)
		__pci_enable_wake(dev, target_state, true, false);

	return error;
}

1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667
/**
 * pci_dev_run_wake - Check if device can generate run-time wake-up events.
 * @dev: Device to check.
 *
 * Return true if the device itself is cabable of generating wake-up events
 * (through the platform or using the native PCIe PME) or if the device supports
 * PME and one of its upstream bridges can generate wake-up events.
 */
bool pci_dev_run_wake(struct pci_dev *dev)
{
	struct pci_bus *bus = dev->bus;

	if (device_run_wake(&dev->dev))
		return true;

	if (!dev->pme_support)
		return false;

	while (bus->parent) {
		struct pci_dev *bridge = bus->self;

		if (device_run_wake(&bridge->dev))
			return true;

		bus = bus->parent;
	}

	/* We have reached the root bus. */
	if (bus->bridge)
		return device_run_wake(bus->bridge);

	return false;
}
EXPORT_SYMBOL_GPL(pci_dev_run_wake);

1668 1669 1670 1671 1672 1673 1674 1675
/**
 * pci_pm_init - Initialize PM functions of given PCI device
 * @dev: PCI device to handle.
 */
void pci_pm_init(struct pci_dev *dev)
{
	int pm;
	u16 pmc;
L
Linus Torvalds 已提交
1676

1677
	pm_runtime_forbid(&dev->dev);
1678
	device_enable_async_suspend(&dev->dev);
1679
	dev->wakeup_prepared = false;
1680

1681 1682
	dev->pm_cap = 0;

1683 1684 1685
	/* find PCI PM capability in list */
	pm = pci_find_capability(dev, PCI_CAP_ID_PM);
	if (!pm)
1686
		return;
1687 1688
	/* Check device's ability to generate PME# */
	pci_read_config_word(dev, pm + PCI_PM_PMC, &pmc);
1689

1690 1691 1692
	if ((pmc & PCI_PM_CAP_VER_MASK) > 3) {
		dev_err(&dev->dev, "unsupported PM cap regs version (%u)\n",
			pmc & PCI_PM_CAP_VER_MASK);
1693
		return;
1694 1695
	}

1696
	dev->pm_cap = pm;
1697
	dev->d3_delay = PCI_PM_D3_WAIT;
1698 1699 1700 1701

	dev->d1_support = false;
	dev->d2_support = false;
	if (!pci_no_d1d2(dev)) {
B
Bjorn Helgaas 已提交
1702
		if (pmc & PCI_PM_CAP_D1)
1703
			dev->d1_support = true;
B
Bjorn Helgaas 已提交
1704
		if (pmc & PCI_PM_CAP_D2)
1705
			dev->d2_support = true;
B
Bjorn Helgaas 已提交
1706 1707 1708

		if (dev->d1_support || dev->d2_support)
			dev_printk(KERN_DEBUG, &dev->dev, "supports%s%s\n",
1709 1710
				   dev->d1_support ? " D1" : "",
				   dev->d2_support ? " D2" : "");
1711 1712 1713 1714
	}

	pmc &= PCI_PM_CAP_PME_MASK;
	if (pmc) {
B
Bjorn Helgaas 已提交
1715 1716
		dev_printk(KERN_DEBUG, &dev->dev,
			 "PME# supported from%s%s%s%s%s\n",
B
Bjorn Helgaas 已提交
1717 1718 1719 1720 1721
			 (pmc & PCI_PM_CAP_PME_D0) ? " D0" : "",
			 (pmc & PCI_PM_CAP_PME_D1) ? " D1" : "",
			 (pmc & PCI_PM_CAP_PME_D2) ? " D2" : "",
			 (pmc & PCI_PM_CAP_PME_D3) ? " D3hot" : "",
			 (pmc & PCI_PM_CAP_PME_D3cold) ? " D3cold" : "");
1722
		dev->pme_support = pmc >> PCI_PM_CAP_PME_SHIFT;
1723 1724 1725 1726 1727 1728
		/*
		 * Make device's PM flags reflect the wake-up capability, but
		 * let the user space enable it to wake up the system as needed.
		 */
		device_set_wakeup_capable(&dev->dev, true);
		/* Disable the PME# generation functionality */
1729 1730 1731
		pci_pme_active(dev, false);
	} else {
		dev->pme_support = 0;
1732
	}
L
Linus Torvalds 已提交
1733 1734
}

1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753
/**
 * platform_pci_wakeup_init - init platform wakeup if present
 * @dev: PCI device
 *
 * Some devices don't have PCI PM caps but can still generate wakeup
 * events through platform methods (like ACPI events).  If @dev supports
 * platform wakeup events, set the device flag to indicate as much.  This
 * may be redundant if the device also supports PCI PM caps, but double
 * initialization should be safe in that case.
 */
void platform_pci_wakeup_init(struct pci_dev *dev)
{
	if (!platform_pci_can_wakeup(dev))
		return;

	device_set_wakeup_capable(&dev->dev, true);
	platform_pci_sleep_wake(dev, false);
}

1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787
/**
 * pci_add_save_buffer - allocate buffer for saving given capability registers
 * @dev: the PCI device
 * @cap: the capability to allocate the buffer for
 * @size: requested size of the buffer
 */
static int pci_add_cap_save_buffer(
	struct pci_dev *dev, char cap, unsigned int size)
{
	int pos;
	struct pci_cap_saved_state *save_state;

	pos = pci_find_capability(dev, cap);
	if (pos <= 0)
		return 0;

	save_state = kzalloc(sizeof(*save_state) + size, GFP_KERNEL);
	if (!save_state)
		return -ENOMEM;

	save_state->cap_nr = cap;
	pci_add_saved_cap(dev, save_state);

	return 0;
}

/**
 * pci_allocate_cap_save_buffers - allocate buffers for saving capabilities
 * @dev: the PCI device
 */
void pci_allocate_cap_save_buffers(struct pci_dev *dev)
{
	int error;

1788 1789
	error = pci_add_cap_save_buffer(dev, PCI_CAP_ID_EXP,
					PCI_EXP_SAVE_REGS * sizeof(u16));
1790 1791 1792 1793 1794 1795 1796 1797 1798 1799
	if (error)
		dev_err(&dev->dev,
			"unable to preallocate PCI Express save buffer\n");

	error = pci_add_cap_save_buffer(dev, PCI_CAP_ID_PCIX, sizeof(u16));
	if (error)
		dev_err(&dev->dev,
			"unable to preallocate PCI-X save buffer\n");
}

Y
Yu Zhao 已提交
1800 1801 1802 1803 1804 1805 1806 1807 1808
/**
 * pci_enable_ari - enable ARI forwarding if hardware support it
 * @dev: the PCI device
 */
void pci_enable_ari(struct pci_dev *dev)
{
	int pos;
	u32 cap;
	u16 ctrl;
1809
	struct pci_dev *bridge;
Y
Yu Zhao 已提交
1810

1811
	if (!pci_is_pcie(dev) || dev->devfn)
Y
Yu Zhao 已提交
1812 1813
		return;

1814 1815
	pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ARI);
	if (!pos)
Y
Yu Zhao 已提交
1816 1817
		return;

1818
	bridge = dev->bus->self;
1819
	if (!bridge || !pci_is_pcie(bridge))
1820 1821
		return;

1822
	pos = pci_pcie_cap(bridge);
Y
Yu Zhao 已提交
1823 1824 1825
	if (!pos)
		return;

1826
	pci_read_config_dword(bridge, pos + PCI_EXP_DEVCAP2, &cap);
Y
Yu Zhao 已提交
1827 1828 1829
	if (!(cap & PCI_EXP_DEVCAP2_ARI))
		return;

1830
	pci_read_config_word(bridge, pos + PCI_EXP_DEVCTL2, &ctrl);
Y
Yu Zhao 已提交
1831
	ctrl |= PCI_EXP_DEVCTL2_ARI;
1832
	pci_write_config_word(bridge, pos + PCI_EXP_DEVCTL2, ctrl);
Y
Yu Zhao 已提交
1833

1834
	bridge->ari_enabled = 1;
Y
Yu Zhao 已提交
1835 1836
}

1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889
/**
 * pci_enable_ido - enable ID-based ordering on a device
 * @dev: the PCI device
 * @type: which types of IDO to enable
 *
 * Enable ID-based ordering on @dev.  @type can contain the bits
 * %PCI_EXP_IDO_REQUEST and/or %PCI_EXP_IDO_COMPLETION to indicate
 * which types of transactions are allowed to be re-ordered.
 */
void pci_enable_ido(struct pci_dev *dev, unsigned long type)
{
	int pos;
	u16 ctrl;

	pos = pci_pcie_cap(dev);
	if (!pos)
		return;

	pci_read_config_word(dev, pos + PCI_EXP_DEVCTL2, &ctrl);
	if (type & PCI_EXP_IDO_REQUEST)
		ctrl |= PCI_EXP_IDO_REQ_EN;
	if (type & PCI_EXP_IDO_COMPLETION)
		ctrl |= PCI_EXP_IDO_CMP_EN;
	pci_write_config_word(dev, pos + PCI_EXP_DEVCTL2, ctrl);
}
EXPORT_SYMBOL(pci_enable_ido);

/**
 * pci_disable_ido - disable ID-based ordering on a device
 * @dev: the PCI device
 * @type: which types of IDO to disable
 */
void pci_disable_ido(struct pci_dev *dev, unsigned long type)
{
	int pos;
	u16 ctrl;

	if (!pci_is_pcie(dev))
		return;

	pos = pci_pcie_cap(dev);
	if (!pos)
		return;

	pci_read_config_word(dev, pos + PCI_EXP_DEVCTL2, &ctrl);
	if (type & PCI_EXP_IDO_REQUEST)
		ctrl &= ~PCI_EXP_IDO_REQ_EN;
	if (type & PCI_EXP_IDO_COMPLETION)
		ctrl &= ~PCI_EXP_IDO_CMP_EN;
	pci_write_config_word(dev, pos + PCI_EXP_DEVCTL2, ctrl);
}
EXPORT_SYMBOL(pci_disable_ido);

C
Chris Wright 已提交
1890 1891 1892 1893 1894 1895 1896 1897 1898 1899
static int pci_acs_enable;

/**
 * pci_request_acs - ask for ACS to be enabled if supported
 */
void pci_request_acs(void)
{
	pci_acs_enable = 1;
}

1900 1901 1902 1903 1904 1905 1906 1907 1908 1909
/**
 * pci_enable_acs - enable ACS if hardware support it
 * @dev: the PCI device
 */
void pci_enable_acs(struct pci_dev *dev)
{
	int pos;
	u16 cap;
	u16 ctrl;

C
Chris Wright 已提交
1910 1911 1912
	if (!pci_acs_enable)
		return;

1913
	if (!pci_is_pcie(dev))
1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937
		return;

	pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ACS);
	if (!pos)
		return;

	pci_read_config_word(dev, pos + PCI_ACS_CAP, &cap);
	pci_read_config_word(dev, pos + PCI_ACS_CTRL, &ctrl);

	/* Source Validation */
	ctrl |= (cap & PCI_ACS_SV);

	/* P2P Request Redirect */
	ctrl |= (cap & PCI_ACS_RR);

	/* P2P Completion Redirect */
	ctrl |= (cap & PCI_ACS_CR);

	/* Upstream Forwarding */
	ctrl |= (cap & PCI_ACS_UF);

	pci_write_config_word(dev, pos + PCI_ACS_CTRL, ctrl);
}

1938 1939 1940 1941 1942 1943 1944
/**
 * pci_swizzle_interrupt_pin - swizzle INTx for device behind bridge
 * @dev: the PCI device
 * @pin: the INTx pin (1=INTA, 2=INTB, 3=INTD, 4=INTD)
 *
 * Perform INTx swizzling for a device behind one level of bridge.  This is
 * required by section 9.1 of the PCI-to-PCI bridge specification for devices
1945 1946 1947
 * behind bridges on add-in cards.  For devices with ARI enabled, the slot
 * number is always 0 (see the Implementation Note in section 2.2.8.1 of
 * the PCI Express Base Specification, Revision 2.1)
1948 1949 1950
 */
u8 pci_swizzle_interrupt_pin(struct pci_dev *dev, u8 pin)
{
1951 1952 1953 1954 1955 1956 1957 1958
	int slot;

	if (pci_ari_enabled(dev->bus))
		slot = 0;
	else
		slot = PCI_SLOT(dev->devfn);

	return (((pin - 1) + slot) % 4) + 1;
1959 1960
}

L
Linus Torvalds 已提交
1961 1962 1963 1964 1965
int
pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge)
{
	u8 pin;

1966
	pin = dev->pin;
L
Linus Torvalds 已提交
1967 1968
	if (!pin)
		return -1;
1969

1970
	while (!pci_is_root_bus(dev->bus)) {
1971
		pin = pci_swizzle_interrupt_pin(dev, pin);
L
Linus Torvalds 已提交
1972 1973 1974 1975 1976 1977
		dev = dev->bus->self;
	}
	*bridge = dev;
	return pin;
}

1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989
/**
 * pci_common_swizzle - swizzle INTx all the way to root bridge
 * @dev: the PCI device
 * @pinp: pointer to the INTx pin value (1=INTA, 2=INTB, 3=INTD, 4=INTD)
 *
 * Perform INTx swizzling for a device.  This traverses through all PCI-to-PCI
 * bridges all the way up to a PCI root bus.
 */
u8 pci_common_swizzle(struct pci_dev *dev, u8 *pinp)
{
	u8 pin = *pinp;

1990
	while (!pci_is_root_bus(dev->bus)) {
1991 1992 1993 1994 1995 1996 1997
		pin = pci_swizzle_interrupt_pin(dev, pin);
		dev = dev->bus->self;
	}
	*pinp = pin;
	return PCI_SLOT(dev->devfn);
}

L
Linus Torvalds 已提交
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
/**
 *	pci_release_region - Release a PCI bar
 *	@pdev: PCI device whose resources were previously reserved by pci_request_region
 *	@bar: BAR to release
 *
 *	Releases the PCI I/O and memory resources previously reserved by a
 *	successful call to pci_request_region.  Call this function only
 *	after all use of the PCI regions has ceased.
 */
void pci_release_region(struct pci_dev *pdev, int bar)
{
T
Tejun Heo 已提交
2009 2010
	struct pci_devres *dr;

L
Linus Torvalds 已提交
2011 2012 2013 2014 2015 2016 2017 2018
	if (pci_resource_len(pdev, bar) == 0)
		return;
	if (pci_resource_flags(pdev, bar) & IORESOURCE_IO)
		release_region(pci_resource_start(pdev, bar),
				pci_resource_len(pdev, bar));
	else if (pci_resource_flags(pdev, bar) & IORESOURCE_MEM)
		release_mem_region(pci_resource_start(pdev, bar),
				pci_resource_len(pdev, bar));
T
Tejun Heo 已提交
2019 2020 2021 2022

	dr = find_pci_dr(pdev);
	if (dr)
		dr->region_mask &= ~(1 << bar);
L
Linus Torvalds 已提交
2023 2024 2025
}

/**
2026
 *	__pci_request_region - Reserved PCI I/O and memory resource
L
Linus Torvalds 已提交
2027 2028 2029
 *	@pdev: PCI device whose resources are to be reserved
 *	@bar: BAR to be reserved
 *	@res_name: Name to be associated with resource.
2030
 *	@exclusive: whether the region access is exclusive or not
L
Linus Torvalds 已提交
2031 2032 2033 2034 2035 2036
 *
 *	Mark the PCI region associated with PCI device @pdev BR @bar as
 *	being reserved by owner @res_name.  Do not access any
 *	address inside the PCI regions unless this call returns
 *	successfully.
 *
2037 2038 2039 2040
 *	If @exclusive is set, then the region is marked so that userspace
 *	is explicitly not allowed to map the resource via /dev/mem or
 * 	sysfs MMIO access.
 *
L
Linus Torvalds 已提交
2041 2042 2043
 *	Returns 0 on success, or %EBUSY on error.  A warning
 *	message is also printed on failure.
 */
2044 2045
static int __pci_request_region(struct pci_dev *pdev, int bar, const char *res_name,
									int exclusive)
L
Linus Torvalds 已提交
2046
{
T
Tejun Heo 已提交
2047 2048
	struct pci_devres *dr;

L
Linus Torvalds 已提交
2049 2050 2051 2052 2053 2054 2055 2056 2057
	if (pci_resource_len(pdev, bar) == 0)
		return 0;
		
	if (pci_resource_flags(pdev, bar) & IORESOURCE_IO) {
		if (!request_region(pci_resource_start(pdev, bar),
			    pci_resource_len(pdev, bar), res_name))
			goto err_out;
	}
	else if (pci_resource_flags(pdev, bar) & IORESOURCE_MEM) {
2058 2059 2060
		if (!__request_mem_region(pci_resource_start(pdev, bar),
					pci_resource_len(pdev, bar), res_name,
					exclusive))
L
Linus Torvalds 已提交
2061 2062
			goto err_out;
	}
T
Tejun Heo 已提交
2063 2064 2065 2066 2067

	dr = find_pci_dr(pdev);
	if (dr)
		dr->region_mask |= 1 << bar;

L
Linus Torvalds 已提交
2068 2069 2070
	return 0;

err_out:
2071
	dev_warn(&pdev->dev, "BAR %d: can't reserve %pR\n", bar,
2072
		 &pdev->resource[bar]);
L
Linus Torvalds 已提交
2073 2074 2075
	return -EBUSY;
}

2076
/**
2077
 *	pci_request_region - Reserve PCI I/O and memory resource
2078 2079
 *	@pdev: PCI device whose resources are to be reserved
 *	@bar: BAR to be reserved
2080
 *	@res_name: Name to be associated with resource
2081
 *
2082
 *	Mark the PCI region associated with PCI device @pdev BAR @bar as
2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116
 *	being reserved by owner @res_name.  Do not access any
 *	address inside the PCI regions unless this call returns
 *	successfully.
 *
 *	Returns 0 on success, or %EBUSY on error.  A warning
 *	message is also printed on failure.
 */
int pci_request_region(struct pci_dev *pdev, int bar, const char *res_name)
{
	return __pci_request_region(pdev, bar, res_name, 0);
}

/**
 *	pci_request_region_exclusive - Reserved PCI I/O and memory resource
 *	@pdev: PCI device whose resources are to be reserved
 *	@bar: BAR to be reserved
 *	@res_name: Name to be associated with resource.
 *
 *	Mark the PCI region associated with PCI device @pdev BR @bar as
 *	being reserved by owner @res_name.  Do not access any
 *	address inside the PCI regions unless this call returns
 *	successfully.
 *
 *	Returns 0 on success, or %EBUSY on error.  A warning
 *	message is also printed on failure.
 *
 *	The key difference that _exclusive makes it that userspace is
 *	explicitly not allowed to map the resource via /dev/mem or
 * 	sysfs.
 */
int pci_request_region_exclusive(struct pci_dev *pdev, int bar, const char *res_name)
{
	return __pci_request_region(pdev, bar, res_name, IORESOURCE_EXCLUSIVE);
}
2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133
/**
 * pci_release_selected_regions - Release selected PCI I/O and memory resources
 * @pdev: PCI device whose resources were previously reserved
 * @bars: Bitmask of BARs to be released
 *
 * Release selected PCI I/O and memory resources previously reserved.
 * Call this function only after all use of the PCI regions has ceased.
 */
void pci_release_selected_regions(struct pci_dev *pdev, int bars)
{
	int i;

	for (i = 0; i < 6; i++)
		if (bars & (1 << i))
			pci_release_region(pdev, i);
}

2134 2135
int __pci_request_selected_regions(struct pci_dev *pdev, int bars,
				 const char *res_name, int excl)
2136 2137 2138 2139 2140
{
	int i;

	for (i = 0; i < 6; i++)
		if (bars & (1 << i))
2141
			if (__pci_request_region(pdev, i, res_name, excl))
2142 2143 2144 2145 2146 2147 2148 2149 2150 2151
				goto err_out;
	return 0;

err_out:
	while(--i >= 0)
		if (bars & (1 << i))
			pci_release_region(pdev, i);

	return -EBUSY;
}
L
Linus Torvalds 已提交
2152

2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172

/**
 * pci_request_selected_regions - Reserve selected PCI I/O and memory resources
 * @pdev: PCI device whose resources are to be reserved
 * @bars: Bitmask of BARs to be requested
 * @res_name: Name to be associated with resource
 */
int pci_request_selected_regions(struct pci_dev *pdev, int bars,
				 const char *res_name)
{
	return __pci_request_selected_regions(pdev, bars, res_name, 0);
}

int pci_request_selected_regions_exclusive(struct pci_dev *pdev,
				 int bars, const char *res_name)
{
	return __pci_request_selected_regions(pdev, bars, res_name,
			IORESOURCE_EXCLUSIVE);
}

L
Linus Torvalds 已提交
2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183
/**
 *	pci_release_regions - Release reserved PCI I/O and memory resources
 *	@pdev: PCI device whose resources were previously reserved by pci_request_regions
 *
 *	Releases all PCI I/O and memory resources previously reserved by a
 *	successful call to pci_request_regions.  Call this function only
 *	after all use of the PCI regions has ceased.
 */

void pci_release_regions(struct pci_dev *pdev)
{
2184
	pci_release_selected_regions(pdev, (1 << 6) - 1);
L
Linus Torvalds 已提交
2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199
}

/**
 *	pci_request_regions - Reserved PCI I/O and memory resources
 *	@pdev: PCI device whose resources are to be reserved
 *	@res_name: Name to be associated with resource.
 *
 *	Mark all PCI regions associated with PCI device @pdev as
 *	being reserved by owner @res_name.  Do not access any
 *	address inside the PCI regions unless this call returns
 *	successfully.
 *
 *	Returns 0 on success, or %EBUSY on error.  A warning
 *	message is also printed on failure.
 */
2200
int pci_request_regions(struct pci_dev *pdev, const char *res_name)
L
Linus Torvalds 已提交
2201
{
2202
	return pci_request_selected_regions(pdev, ((1 << 6) - 1), res_name);
L
Linus Torvalds 已提交
2203 2204
}

2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226
/**
 *	pci_request_regions_exclusive - Reserved PCI I/O and memory resources
 *	@pdev: PCI device whose resources are to be reserved
 *	@res_name: Name to be associated with resource.
 *
 *	Mark all PCI regions associated with PCI device @pdev as
 *	being reserved by owner @res_name.  Do not access any
 *	address inside the PCI regions unless this call returns
 *	successfully.
 *
 *	pci_request_regions_exclusive() will mark the region so that
 * 	/dev/mem and the sysfs MMIO access will not be allowed.
 *
 *	Returns 0 on success, or %EBUSY on error.  A warning
 *	message is also printed on failure.
 */
int pci_request_regions_exclusive(struct pci_dev *pdev, const char *res_name)
{
	return pci_request_selected_regions_exclusive(pdev,
					((1 << 6) - 1), res_name);
}

2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242
static void __pci_set_master(struct pci_dev *dev, bool enable)
{
	u16 old_cmd, cmd;

	pci_read_config_word(dev, PCI_COMMAND, &old_cmd);
	if (enable)
		cmd = old_cmd | PCI_COMMAND_MASTER;
	else
		cmd = old_cmd & ~PCI_COMMAND_MASTER;
	if (cmd != old_cmd) {
		dev_dbg(&dev->dev, "%s bus mastering\n",
			enable ? "enabling" : "disabling");
		pci_write_config_word(dev, PCI_COMMAND, cmd);
	}
	dev->is_busmaster = enable;
}
2243

L
Linus Torvalds 已提交
2244 2245 2246 2247 2248 2249 2250
/**
 * pci_set_master - enables bus-mastering for device dev
 * @dev: the PCI device to enable
 *
 * Enables bus-mastering on the device and calls pcibios_set_master()
 * to do the needed arch specific settings.
 */
2251
void pci_set_master(struct pci_dev *dev)
L
Linus Torvalds 已提交
2252
{
2253
	__pci_set_master(dev, true);
L
Linus Torvalds 已提交
2254 2255 2256
	pcibios_set_master(dev);
}

2257 2258 2259 2260 2261 2262 2263 2264 2265
/**
 * pci_clear_master - disables bus-mastering for device dev
 * @dev: the PCI device to disable
 */
void pci_clear_master(struct pci_dev *dev)
{
	__pci_set_master(dev, false);
}

L
Linus Torvalds 已提交
2266
/**
2267 2268
 * pci_set_cacheline_size - ensure the CACHE_LINE_SIZE register is programmed
 * @dev: the PCI device for which MWI is to be enabled
L
Linus Torvalds 已提交
2269
 *
2270 2271
 * Helper function for pci_set_mwi.
 * Originally copied from drivers/net/acenic.c.
L
Linus Torvalds 已提交
2272 2273 2274 2275
 * Copyright 1998-2001 by Jes Sorensen, <jes@trained-monkey.org>.
 *
 * RETURNS: An appropriate -ERRNO error value on error, or zero for success.
 */
T
Tejun Heo 已提交
2276
int pci_set_cacheline_size(struct pci_dev *dev)
L
Linus Torvalds 已提交
2277 2278 2279 2280
{
	u8 cacheline_size;

	if (!pci_cache_line_size)
T
Tejun Heo 已提交
2281
		return -EINVAL;
L
Linus Torvalds 已提交
2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296

	/* Validate current setting: the PCI_CACHE_LINE_SIZE must be
	   equal to or multiple of the right value. */
	pci_read_config_byte(dev, PCI_CACHE_LINE_SIZE, &cacheline_size);
	if (cacheline_size >= pci_cache_line_size &&
	    (cacheline_size % pci_cache_line_size) == 0)
		return 0;

	/* Write the correct value. */
	pci_write_config_byte(dev, PCI_CACHE_LINE_SIZE, pci_cache_line_size);
	/* Read it back. */
	pci_read_config_byte(dev, PCI_CACHE_LINE_SIZE, &cacheline_size);
	if (cacheline_size == pci_cache_line_size)
		return 0;

2297 2298
	dev_printk(KERN_DEBUG, &dev->dev, "cache line size of %d is not "
		   "supported\n", pci_cache_line_size << 2);
L
Linus Torvalds 已提交
2299 2300 2301

	return -EINVAL;
}
T
Tejun Heo 已提交
2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319
EXPORT_SYMBOL_GPL(pci_set_cacheline_size);

#ifdef PCI_DISABLE_MWI
int pci_set_mwi(struct pci_dev *dev)
{
	return 0;
}

int pci_try_set_mwi(struct pci_dev *dev)
{
	return 0;
}

void pci_clear_mwi(struct pci_dev *dev)
{
}

#else
L
Linus Torvalds 已提交
2320 2321 2322 2323 2324

/**
 * pci_set_mwi - enables memory-write-invalidate PCI transaction
 * @dev: the PCI device for which MWI is enabled
 *
R
Randy Dunlap 已提交
2325
 * Enables the Memory-Write-Invalidate transaction in %PCI_COMMAND.
L
Linus Torvalds 已提交
2326 2327 2328 2329 2330 2331 2332 2333 2334
 *
 * RETURNS: An appropriate -ERRNO error value on error, or zero for success.
 */
int
pci_set_mwi(struct pci_dev *dev)
{
	int rc;
	u16 cmd;

2335
	rc = pci_set_cacheline_size(dev);
L
Linus Torvalds 已提交
2336 2337 2338 2339 2340
	if (rc)
		return rc;

	pci_read_config_word(dev, PCI_COMMAND, &cmd);
	if (! (cmd & PCI_COMMAND_INVALIDATE)) {
2341
		dev_dbg(&dev->dev, "enabling Mem-Wr-Inval\n");
L
Linus Torvalds 已提交
2342 2343 2344 2345 2346 2347 2348
		cmd |= PCI_COMMAND_INVALIDATE;
		pci_write_config_word(dev, PCI_COMMAND, cmd);
	}
	
	return 0;
}

R
Randy Dunlap 已提交
2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363
/**
 * pci_try_set_mwi - enables memory-write-invalidate PCI transaction
 * @dev: the PCI device for which MWI is enabled
 *
 * Enables the Memory-Write-Invalidate transaction in %PCI_COMMAND.
 * Callers are not required to check the return value.
 *
 * RETURNS: An appropriate -ERRNO error value on error, or zero for success.
 */
int pci_try_set_mwi(struct pci_dev *dev)
{
	int rc = pci_set_mwi(dev);
	return rc;
}

L
Linus Torvalds 已提交
2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380
/**
 * pci_clear_mwi - disables Memory-Write-Invalidate for device dev
 * @dev: the PCI device to disable
 *
 * Disables PCI Memory-Write-Invalidate transaction on the device
 */
void
pci_clear_mwi(struct pci_dev *dev)
{
	u16 cmd;

	pci_read_config_word(dev, PCI_COMMAND, &cmd);
	if (cmd & PCI_COMMAND_INVALIDATE) {
		cmd &= ~PCI_COMMAND_INVALIDATE;
		pci_write_config_word(dev, PCI_COMMAND, cmd);
	}
}
2381
#endif /* ! PCI_DISABLE_MWI */
L
Linus Torvalds 已提交
2382

B
Brett M Russ 已提交
2383 2384
/**
 * pci_intx - enables/disables PCI INTx for device dev
R
Randy Dunlap 已提交
2385 2386
 * @pdev: the PCI device to operate on
 * @enable: boolean: whether to enable or disable PCI INTx
B
Brett M Russ 已提交
2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403
 *
 * Enables/disables PCI INTx for device dev
 */
void
pci_intx(struct pci_dev *pdev, int enable)
{
	u16 pci_command, new;

	pci_read_config_word(pdev, PCI_COMMAND, &pci_command);

	if (enable) {
		new = pci_command & ~PCI_COMMAND_INTX_DISABLE;
	} else {
		new = pci_command | PCI_COMMAND_INTX_DISABLE;
	}

	if (new != pci_command) {
T
Tejun Heo 已提交
2404 2405
		struct pci_devres *dr;

2406
		pci_write_config_word(pdev, PCI_COMMAND, new);
T
Tejun Heo 已提交
2407 2408 2409 2410 2411 2412

		dr = find_pci_dr(pdev);
		if (dr && !dr->restore_intx) {
			dr->restore_intx = 1;
			dr->orig_intx = !enable;
		}
B
Brett M Russ 已提交
2413 2414 2415
	}
}

2416 2417
/**
 * pci_msi_off - disables any msi or msix capabilities
R
Randy Dunlap 已提交
2418
 * @dev: the PCI device to operate on
2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441
 *
 * If you want to use msi see pci_enable_msi and friends.
 * This is a lower level primitive that allows us to disable
 * msi operation at the device level.
 */
void pci_msi_off(struct pci_dev *dev)
{
	int pos;
	u16 control;

	pos = pci_find_capability(dev, PCI_CAP_ID_MSI);
	if (pos) {
		pci_read_config_word(dev, pos + PCI_MSI_FLAGS, &control);
		control &= ~PCI_MSI_FLAGS_ENABLE;
		pci_write_config_word(dev, pos + PCI_MSI_FLAGS, control);
	}
	pos = pci_find_capability(dev, PCI_CAP_ID_MSIX);
	if (pos) {
		pci_read_config_word(dev, pos + PCI_MSIX_FLAGS, &control);
		control &= ~PCI_MSIX_FLAGS_ENABLE;
		pci_write_config_word(dev, pos + PCI_MSIX_FLAGS, control);
	}
}
2442
EXPORT_SYMBOL_GPL(pci_msi_off);
2443

2444 2445 2446 2447 2448 2449
int pci_set_dma_max_seg_size(struct pci_dev *dev, unsigned int size)
{
	return dma_set_max_seg_size(&dev->dev, size);
}
EXPORT_SYMBOL(pci_set_dma_max_seg_size);

2450 2451 2452 2453 2454 2455
int pci_set_dma_seg_boundary(struct pci_dev *dev, unsigned long mask)
{
	return dma_set_seg_boundary(&dev->dev, mask);
}
EXPORT_SYMBOL(pci_set_dma_seg_boundary);

Y
Yu Zhao 已提交
2456
static int pcie_flr(struct pci_dev *dev, int probe)
2457
{
Y
Yu Zhao 已提交
2458 2459
	int i;
	int pos;
2460
	u32 cap;
2461
	u16 status, control;
2462

2463
	pos = pci_pcie_cap(dev);
Y
Yu Zhao 已提交
2464
	if (!pos)
2465
		return -ENOTTY;
Y
Yu Zhao 已提交
2466 2467

	pci_read_config_dword(dev, pos + PCI_EXP_DEVCAP, &cap);
2468 2469 2470
	if (!(cap & PCI_EXP_DEVCAP_FLR))
		return -ENOTTY;

S
Sheng Yang 已提交
2471 2472 2473
	if (probe)
		return 0;

2474
	/* Wait for Transaction Pending bit clean */
Y
Yu Zhao 已提交
2475 2476 2477
	for (i = 0; i < 4; i++) {
		if (i)
			msleep((1 << (i - 1)) * 100);
S
Sheng Yang 已提交
2478

Y
Yu Zhao 已提交
2479 2480 2481 2482 2483 2484 2485 2486 2487
		pci_read_config_word(dev, pos + PCI_EXP_DEVSTA, &status);
		if (!(status & PCI_EXP_DEVSTA_TRPND))
			goto clear;
	}

	dev_err(&dev->dev, "transaction is not cleared; "
			"proceeding with reset anyway\n");

clear:
2488 2489 2490 2491
	pci_read_config_word(dev, pos + PCI_EXP_DEVCTL, &control);
	control |= PCI_EXP_DEVCTL_BCR_FLR;
	pci_write_config_word(dev, pos + PCI_EXP_DEVCTL, control);

Y
Yu Zhao 已提交
2492
	msleep(100);
2493 2494 2495

	return 0;
}
S
Sheng Yang 已提交
2496

Y
Yu Zhao 已提交
2497
static int pci_af_flr(struct pci_dev *dev, int probe)
2498
{
Y
Yu Zhao 已提交
2499 2500
	int i;
	int pos;
2501
	u8 cap;
Y
Yu Zhao 已提交
2502
	u8 status;
2503

Y
Yu Zhao 已提交
2504 2505
	pos = pci_find_capability(dev, PCI_CAP_ID_AF);
	if (!pos)
2506
		return -ENOTTY;
Y
Yu Zhao 已提交
2507 2508

	pci_read_config_byte(dev, pos + PCI_AF_CAP, &cap);
2509 2510 2511 2512 2513 2514 2515
	if (!(cap & PCI_AF_CAP_TP) || !(cap & PCI_AF_CAP_FLR))
		return -ENOTTY;

	if (probe)
		return 0;

	/* Wait for Transaction Pending bit clean */
Y
Yu Zhao 已提交
2516 2517 2518 2519 2520 2521 2522 2523
	for (i = 0; i < 4; i++) {
		if (i)
			msleep((1 << (i - 1)) * 100);

		pci_read_config_byte(dev, pos + PCI_AF_STATUS, &status);
		if (!(status & PCI_AF_STATUS_TP))
			goto clear;
	}
S
Sheng Yang 已提交
2524

Y
Yu Zhao 已提交
2525 2526
	dev_err(&dev->dev, "transaction is not cleared; "
			"proceeding with reset anyway\n");
S
Sheng Yang 已提交
2527

Y
Yu Zhao 已提交
2528 2529
clear:
	pci_write_config_byte(dev, pos + PCI_AF_CTRL, PCI_AF_CTRL_FLR);
2530
	msleep(100);
Y
Yu Zhao 已提交
2531

2532 2533 2534
	return 0;
}

2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549
/**
 * pci_pm_reset - Put device into PCI_D3 and back into PCI_D0.
 * @dev: Device to reset.
 * @probe: If set, only check if the device can be reset this way.
 *
 * If @dev supports native PCI PM and its PCI_PM_CTRL_NO_SOFT_RESET flag is
 * unset, it will be reinitialized internally when going from PCI_D3hot to
 * PCI_D0.  If that's the case and the device is not in a low-power state
 * already, force it into PCI_D3hot and back to PCI_D0, causing it to be reset.
 *
 * NOTE: This causes the caller to sleep for twice the device power transition
 * cooldown period, which for the D0->D3hot and D3hot->D0 transitions is 10 ms
 * by devault (i.e. unless the @dev's d3_delay field has a different value).
 * Moreover, only devices in D0 can be reset by this function.
 */
2550
static int pci_pm_reset(struct pci_dev *dev, int probe)
S
Sheng Yang 已提交
2551
{
2552 2553 2554 2555
	u16 csr;

	if (!dev->pm_cap)
		return -ENOTTY;
S
Sheng Yang 已提交
2556

2557 2558 2559
	pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &csr);
	if (csr & PCI_PM_CTRL_NO_SOFT_RESET)
		return -ENOTTY;
S
Sheng Yang 已提交
2560

2561 2562
	if (probe)
		return 0;
2563

2564 2565 2566 2567 2568 2569
	if (dev->current_state != PCI_D0)
		return -EINVAL;

	csr &= ~PCI_PM_CTRL_STATE_MASK;
	csr |= PCI_D3hot;
	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, csr);
2570
	pci_dev_d3_sleep(dev);
2571 2572 2573 2574

	csr &= ~PCI_PM_CTRL_STATE_MASK;
	csr |= PCI_D0;
	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, csr);
2575
	pci_dev_d3_sleep(dev);
2576 2577 2578 2579

	return 0;
}

Y
Yu Zhao 已提交
2580 2581 2582 2583 2584
static int pci_parent_bus_reset(struct pci_dev *dev, int probe)
{
	u16 ctrl;
	struct pci_dev *pdev;

2585
	if (pci_is_root_bus(dev->bus) || dev->subordinate || !dev->bus->self)
Y
Yu Zhao 已提交
2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606
		return -ENOTTY;

	list_for_each_entry(pdev, &dev->bus->devices, bus_list)
		if (pdev != dev)
			return -ENOTTY;

	if (probe)
		return 0;

	pci_read_config_word(dev->bus->self, PCI_BRIDGE_CONTROL, &ctrl);
	ctrl |= PCI_BRIDGE_CTL_BUS_RESET;
	pci_write_config_word(dev->bus->self, PCI_BRIDGE_CONTROL, ctrl);
	msleep(100);

	ctrl &= ~PCI_BRIDGE_CTL_BUS_RESET;
	pci_write_config_word(dev->bus->self, PCI_BRIDGE_CONTROL, ctrl);
	msleep(100);

	return 0;
}

Y
Yu Zhao 已提交
2607
static int pci_dev_reset(struct pci_dev *dev, int probe)
S
Sheng Yang 已提交
2608
{
Y
Yu Zhao 已提交
2609 2610 2611 2612 2613 2614 2615
	int rc;

	might_sleep();

	if (!probe) {
		pci_block_user_cfg_access(dev);
		/* block PM suspend, driver probe, etc. */
2616
		device_lock(&dev->dev);
Y
Yu Zhao 已提交
2617
	}
S
Sheng Yang 已提交
2618

2619 2620 2621 2622
	rc = pci_dev_specific_reset(dev, probe);
	if (rc != -ENOTTY)
		goto done;

Y
Yu Zhao 已提交
2623 2624 2625
	rc = pcie_flr(dev, probe);
	if (rc != -ENOTTY)
		goto done;
S
Sheng Yang 已提交
2626

Y
Yu Zhao 已提交
2627
	rc = pci_af_flr(dev, probe);
2628 2629 2630 2631
	if (rc != -ENOTTY)
		goto done;

	rc = pci_pm_reset(dev, probe);
Y
Yu Zhao 已提交
2632 2633 2634 2635
	if (rc != -ENOTTY)
		goto done;

	rc = pci_parent_bus_reset(dev, probe);
Y
Yu Zhao 已提交
2636 2637
done:
	if (!probe) {
2638
		device_unlock(&dev->dev);
Y
Yu Zhao 已提交
2639 2640
		pci_unblock_user_cfg_access(dev);
	}
2641

Y
Yu Zhao 已提交
2642
	return rc;
S
Sheng Yang 已提交
2643 2644 2645
}

/**
Y
Yu Zhao 已提交
2646 2647
 * __pci_reset_function - reset a PCI device function
 * @dev: PCI device to reset
S
Sheng Yang 已提交
2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658
 *
 * Some devices allow an individual function to be reset without affecting
 * other functions in the same device.  The PCI device must be responsive
 * to PCI config space in order to use this function.
 *
 * The device function is presumed to be unused when this function is called.
 * Resetting the device will make the contents of PCI configuration space
 * random, so any caller of this must be prepared to reinitialise the
 * device including MSI, bus mastering, BARs, decoding IO and memory spaces,
 * etc.
 *
Y
Yu Zhao 已提交
2659
 * Returns 0 if the device function was successfully reset or negative if the
S
Sheng Yang 已提交
2660 2661
 * device doesn't support resetting a single function.
 */
Y
Yu Zhao 已提交
2662
int __pci_reset_function(struct pci_dev *dev)
S
Sheng Yang 已提交
2663
{
Y
Yu Zhao 已提交
2664
	return pci_dev_reset(dev, 0);
S
Sheng Yang 已提交
2665
}
Y
Yu Zhao 已提交
2666
EXPORT_SYMBOL_GPL(__pci_reset_function);
2667

2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683
/**
 * pci_probe_reset_function - check whether the device can be safely reset
 * @dev: PCI device to reset
 *
 * Some devices allow an individual function to be reset without affecting
 * other functions in the same device.  The PCI device must be responsive
 * to PCI config space in order to use this function.
 *
 * Returns 0 if the device function can be reset or negative if the
 * device doesn't support resetting a single function.
 */
int pci_probe_reset_function(struct pci_dev *dev)
{
	return pci_dev_reset(dev, 1);
}

2684
/**
Y
Yu Zhao 已提交
2685 2686
 * pci_reset_function - quiesce and reset a PCI device function
 * @dev: PCI device to reset
2687 2688 2689 2690 2691 2692 2693
 *
 * Some devices allow an individual function to be reset without affecting
 * other functions in the same device.  The PCI device must be responsive
 * to PCI config space in order to use this function.
 *
 * This function does not just reset the PCI portion of a device, but
 * clears all the state associated with the device.  This function differs
Y
Yu Zhao 已提交
2694
 * from __pci_reset_function in that it saves and restores device state
2695 2696
 * over the reset.
 *
Y
Yu Zhao 已提交
2697
 * Returns 0 if the device function was successfully reset or negative if the
2698 2699 2700 2701
 * device doesn't support resetting a single function.
 */
int pci_reset_function(struct pci_dev *dev)
{
Y
Yu Zhao 已提交
2702
	int rc;
2703

Y
Yu Zhao 已提交
2704 2705 2706
	rc = pci_dev_reset(dev, 1);
	if (rc)
		return rc;
2707 2708 2709

	pci_save_state(dev);

Y
Yu Zhao 已提交
2710 2711 2712 2713
	/*
	 * both INTx and MSI are disabled after the Interrupt Disable bit
	 * is set and the Bus Master bit is cleared.
	 */
2714 2715
	pci_write_config_word(dev, PCI_COMMAND, PCI_COMMAND_INTX_DISABLE);

Y
Yu Zhao 已提交
2716
	rc = pci_dev_reset(dev, 0);
2717 2718 2719

	pci_restore_state(dev);

Y
Yu Zhao 已提交
2720
	return rc;
2721 2722 2723
}
EXPORT_SYMBOL_GPL(pci_reset_function);

2724 2725 2726 2727 2728 2729 2730 2731 2732
/**
 * pcix_get_max_mmrbc - get PCI-X maximum designed memory read byte count
 * @dev: PCI device to query
 *
 * Returns mmrbc: maximum designed memory read count in bytes
 *    or appropriate error value.
 */
int pcix_get_max_mmrbc(struct pci_dev *dev)
{
2733
	int cap;
2734 2735 2736 2737 2738 2739
	u32 stat;

	cap = pci_find_capability(dev, PCI_CAP_ID_PCIX);
	if (!cap)
		return -EINVAL;

2740
	if (pci_read_config_dword(dev, cap + PCI_X_STATUS, &stat))
2741 2742
		return -EINVAL;

2743
	return 512 << ((stat & PCI_X_STATUS_MAX_READ) >> 21);
2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755
}
EXPORT_SYMBOL(pcix_get_max_mmrbc);

/**
 * pcix_get_mmrbc - get PCI-X maximum memory read byte count
 * @dev: PCI device to query
 *
 * Returns mmrbc: maximum memory read count in bytes
 *    or appropriate error value.
 */
int pcix_get_mmrbc(struct pci_dev *dev)
{
2756
	int cap;
2757
	u16 cmd;
2758 2759 2760 2761 2762

	cap = pci_find_capability(dev, PCI_CAP_ID_PCIX);
	if (!cap)
		return -EINVAL;

2763 2764
	if (pci_read_config_word(dev, cap + PCI_X_CMD, &cmd))
		return -EINVAL;
2765

2766
	return 512 << ((cmd & PCI_X_CMD_MAX_READ) >> 2);
2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780
}
EXPORT_SYMBOL(pcix_get_mmrbc);

/**
 * pcix_set_mmrbc - set PCI-X maximum memory read byte count
 * @dev: PCI device to query
 * @mmrbc: maximum memory read count in bytes
 *    valid values are 512, 1024, 2048, 4096
 *
 * If possible sets maximum memory read byte count, some bridges have erratas
 * that prevent this.
 */
int pcix_set_mmrbc(struct pci_dev *dev, int mmrbc)
{
2781
	int cap;
2782 2783
	u32 stat, v, o;
	u16 cmd;
2784

2785
	if (mmrbc < 512 || mmrbc > 4096 || !is_power_of_2(mmrbc))
2786
		return -EINVAL;
2787 2788 2789 2790 2791

	v = ffs(mmrbc) - 10;

	cap = pci_find_capability(dev, PCI_CAP_ID_PCIX);
	if (!cap)
2792
		return -EINVAL;
2793

2794 2795
	if (pci_read_config_dword(dev, cap + PCI_X_STATUS, &stat))
		return -EINVAL;
2796 2797 2798 2799

	if (v > (stat & PCI_X_STATUS_MAX_READ) >> 21)
		return -E2BIG;

2800 2801
	if (pci_read_config_word(dev, cap + PCI_X_CMD, &cmd))
		return -EINVAL;
2802 2803 2804 2805 2806 2807 2808 2809 2810

	o = (cmd & PCI_X_CMD_MAX_READ) >> 2;
	if (o != v) {
		if (v > o && dev->bus &&
		   (dev->bus->bus_flags & PCI_BUS_FLAGS_NO_MMRBC))
			return -EIO;

		cmd &= ~PCI_X_CMD_MAX_READ;
		cmd |= v << 2;
2811 2812
		if (pci_write_config_word(dev, cap + PCI_X_CMD, cmd))
			return -EIO;
2813
	}
2814
	return 0;
2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829
}
EXPORT_SYMBOL(pcix_set_mmrbc);

/**
 * pcie_get_readrq - get PCI Express read request size
 * @dev: PCI device to query
 *
 * Returns maximum memory read request in bytes
 *    or appropriate error value.
 */
int pcie_get_readrq(struct pci_dev *dev)
{
	int ret, cap;
	u16 ctl;

2830
	cap = pci_pcie_cap(dev);
2831 2832 2833 2834 2835
	if (!cap)
		return -EINVAL;

	ret = pci_read_config_word(dev, cap + PCI_EXP_DEVCTL, &ctl);
	if (!ret)
2836
		ret = 128 << ((ctl & PCI_EXP_DEVCTL_READRQ) >> 12);
2837 2838 2839 2840 2841 2842 2843 2844

	return ret;
}
EXPORT_SYMBOL(pcie_get_readrq);

/**
 * pcie_set_readrq - set PCI Express maximum memory read request
 * @dev: PCI device to query
2845
 * @rq: maximum memory read count in bytes
2846 2847 2848 2849 2850 2851 2852 2853 2854
 *    valid values are 128, 256, 512, 1024, 2048, 4096
 *
 * If possible sets maximum read byte count
 */
int pcie_set_readrq(struct pci_dev *dev, int rq)
{
	int cap, err = -EINVAL;
	u16 ctl, v;

2855
	if (rq < 128 || rq > 4096 || !is_power_of_2(rq))
2856 2857 2858 2859
		goto out;

	v = (ffs(rq) - 8) << 12;

2860
	cap = pci_pcie_cap(dev);
2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878
	if (!cap)
		goto out;

	err = pci_read_config_word(dev, cap + PCI_EXP_DEVCTL, &ctl);
	if (err)
		goto out;

	if ((ctl & PCI_EXP_DEVCTL_READRQ) != v) {
		ctl &= ~PCI_EXP_DEVCTL_READRQ;
		ctl |= v;
		err = pci_write_config_dword(dev, cap + PCI_EXP_DEVCTL, ctl);
	}

out:
	return err;
}
EXPORT_SYMBOL(pcie_set_readrq);

2879 2880
/**
 * pci_select_bars - Make BAR mask from the type of resource
2881
 * @dev: the PCI device for which BAR mask is made
2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894
 * @flags: resource type mask to be selected
 *
 * This helper routine makes bar mask from the type of resource.
 */
int pci_select_bars(struct pci_dev *dev, unsigned long flags)
{
	int i, bars = 0;
	for (i = 0; i < PCI_NUM_RESOURCES; i++)
		if (pci_resource_flags(dev, i) & flags)
			bars |= (1 << i);
	return bars;
}

2895 2896 2897 2898 2899 2900 2901 2902 2903 2904
/**
 * pci_resource_bar - get position of the BAR associated with a resource
 * @dev: the PCI device
 * @resno: the resource number
 * @type: the BAR type to be filled in
 *
 * Returns BAR position in config space, or 0 if the BAR is invalid.
 */
int pci_resource_bar(struct pci_dev *dev, int resno, enum pci_bar_type *type)
{
2905 2906
	int reg;

2907 2908 2909 2910 2911 2912
	if (resno < PCI_ROM_RESOURCE) {
		*type = pci_bar_unknown;
		return PCI_BASE_ADDRESS_0 + 4 * resno;
	} else if (resno == PCI_ROM_RESOURCE) {
		*type = pci_bar_mem32;
		return dev->rom_base_reg;
2913 2914 2915 2916 2917
	} else if (resno < PCI_BRIDGE_RESOURCES) {
		/* device specific resource */
		reg = pci_iov_resource_bar(dev, resno, type);
		if (reg)
			return reg;
2918 2919
	}

2920
	dev_err(&dev->dev, "BAR %d: invalid resource\n", resno);
2921 2922 2923
	return 0;
}

2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940
/* Some architectures require additional programming to enable VGA */
static arch_set_vga_state_t arch_set_vga_state;

void __init pci_register_set_vga_state(arch_set_vga_state_t func)
{
	arch_set_vga_state = func;	/* NULL disables */
}

static int pci_set_vga_state_arch(struct pci_dev *dev, bool decode,
		      unsigned int command_bits, bool change_bridge)
{
	if (arch_set_vga_state)
		return arch_set_vga_state(dev, decode, command_bits,
						change_bridge);
	return 0;
}

2941 2942
/**
 * pci_set_vga_state - set VGA decode state on device and parents if requested
R
Randy Dunlap 已提交
2943 2944 2945 2946
 * @dev: the PCI device
 * @decode: true = enable decoding, false = disable decoding
 * @command_bits: PCI_COMMAND_IO and/or PCI_COMMAND_MEMORY
 * @change_bridge: traverse ancestors and change bridges
2947 2948 2949 2950 2951 2952 2953
 */
int pci_set_vga_state(struct pci_dev *dev, bool decode,
		      unsigned int command_bits, bool change_bridge)
{
	struct pci_bus *bus;
	struct pci_dev *bridge;
	u16 cmd;
2954
	int rc;
2955 2956 2957

	WARN_ON(command_bits & ~(PCI_COMMAND_IO|PCI_COMMAND_MEMORY));

2958 2959 2960 2961 2962
	/* ARCH specific VGA enables */
	rc = pci_set_vga_state_arch(dev, decode, command_bits, change_bridge);
	if (rc)
		return rc;

2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990
	pci_read_config_word(dev, PCI_COMMAND, &cmd);
	if (decode == true)
		cmd |= command_bits;
	else
		cmd &= ~command_bits;
	pci_write_config_word(dev, PCI_COMMAND, cmd);

	if (change_bridge == false)
		return 0;

	bus = dev->bus;
	while (bus) {
		bridge = bus->self;
		if (bridge) {
			pci_read_config_word(bridge, PCI_BRIDGE_CONTROL,
					     &cmd);
			if (decode == true)
				cmd |= PCI_BRIDGE_CTL_VGA;
			else
				cmd &= ~PCI_BRIDGE_CTL_VGA;
			pci_write_config_word(bridge, PCI_BRIDGE_CONTROL,
					      cmd);
		}
		bus = bus->parent;
	}
	return 0;
}

2991 2992
#define RESOURCE_ALIGNMENT_PARAM_SIZE COMMAND_LINE_SIZE
static char resource_alignment_param[RESOURCE_ALIGNMENT_PARAM_SIZE] = {0};
2993
static DEFINE_SPINLOCK(resource_alignment_lock);
2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105

/**
 * pci_specified_resource_alignment - get resource alignment specified by user.
 * @dev: the PCI device to get
 *
 * RETURNS: Resource alignment if it is specified.
 *          Zero if it is not specified.
 */
resource_size_t pci_specified_resource_alignment(struct pci_dev *dev)
{
	int seg, bus, slot, func, align_order, count;
	resource_size_t align = 0;
	char *p;

	spin_lock(&resource_alignment_lock);
	p = resource_alignment_param;
	while (*p) {
		count = 0;
		if (sscanf(p, "%d%n", &align_order, &count) == 1 &&
							p[count] == '@') {
			p += count + 1;
		} else {
			align_order = -1;
		}
		if (sscanf(p, "%x:%x:%x.%x%n",
			&seg, &bus, &slot, &func, &count) != 4) {
			seg = 0;
			if (sscanf(p, "%x:%x.%x%n",
					&bus, &slot, &func, &count) != 3) {
				/* Invalid format */
				printk(KERN_ERR "PCI: Can't parse resource_alignment parameter: %s\n",
					p);
				break;
			}
		}
		p += count;
		if (seg == pci_domain_nr(dev->bus) &&
			bus == dev->bus->number &&
			slot == PCI_SLOT(dev->devfn) &&
			func == PCI_FUNC(dev->devfn)) {
			if (align_order == -1) {
				align = PAGE_SIZE;
			} else {
				align = 1 << align_order;
			}
			/* Found */
			break;
		}
		if (*p != ';' && *p != ',') {
			/* End of param or invalid format */
			break;
		}
		p++;
	}
	spin_unlock(&resource_alignment_lock);
	return align;
}

/**
 * pci_is_reassigndev - check if specified PCI is target device to reassign
 * @dev: the PCI device to check
 *
 * RETURNS: non-zero for PCI device is a target device to reassign,
 *          or zero is not.
 */
int pci_is_reassigndev(struct pci_dev *dev)
{
	return (pci_specified_resource_alignment(dev) != 0);
}

ssize_t pci_set_resource_alignment_param(const char *buf, size_t count)
{
	if (count > RESOURCE_ALIGNMENT_PARAM_SIZE - 1)
		count = RESOURCE_ALIGNMENT_PARAM_SIZE - 1;
	spin_lock(&resource_alignment_lock);
	strncpy(resource_alignment_param, buf, count);
	resource_alignment_param[count] = '\0';
	spin_unlock(&resource_alignment_lock);
	return count;
}

ssize_t pci_get_resource_alignment_param(char *buf, size_t size)
{
	size_t count;
	spin_lock(&resource_alignment_lock);
	count = snprintf(buf, size, "%s", resource_alignment_param);
	spin_unlock(&resource_alignment_lock);
	return count;
}

static ssize_t pci_resource_alignment_show(struct bus_type *bus, char *buf)
{
	return pci_get_resource_alignment_param(buf, PAGE_SIZE);
}

static ssize_t pci_resource_alignment_store(struct bus_type *bus,
					const char *buf, size_t count)
{
	return pci_set_resource_alignment_param(buf, count);
}

BUS_ATTR(resource_alignment, 0644, pci_resource_alignment_show,
					pci_resource_alignment_store);

static int __init pci_resource_alignment_sysfs_init(void)
{
	return bus_create_file(&pci_bus_type,
					&bus_attr_resource_alignment);
}

late_initcall(pci_resource_alignment_sysfs_init);

3106 3107 3108 3109 3110 3111 3112
static void __devinit pci_no_domains(void)
{
#ifdef CONFIG_PCI_DOMAINS
	pci_domains_supported = 0;
#endif
}

3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125
/**
 * pci_ext_cfg_enabled - can we access extended PCI config space?
 * @dev: The PCI device of the root bridge.
 *
 * Returns 1 if we can access PCI extended config space (offsets
 * greater than 0xff). This is the default implementation. Architecture
 * implementations can override this.
 */
int __attribute__ ((weak)) pci_ext_cfg_avail(struct pci_dev *dev)
{
	return 1;
}

3126 3127 3128 3129 3130
void __weak pci_fixup_cardbus(struct pci_bus *bus)
{
}
EXPORT_SYMBOL(pci_fixup_cardbus);

A
Al Viro 已提交
3131
static int __init pci_setup(char *str)
L
Linus Torvalds 已提交
3132 3133 3134 3135 3136 3137
{
	while (str) {
		char *k = strchr(str, ',');
		if (k)
			*k++ = 0;
		if (*str && (str = pcibios_setup(str)) && *str) {
3138 3139
			if (!strcmp(str, "nomsi")) {
				pci_no_msi();
R
Randy Dunlap 已提交
3140 3141
			} else if (!strcmp(str, "noaer")) {
				pci_no_aer();
3142 3143
			} else if (!strcmp(str, "nodomains")) {
				pci_no_domains();
3144 3145 3146 3147
			} else if (!strncmp(str, "cbiosize=", 9)) {
				pci_cardbus_io_size = memparse(str + 9, &str);
			} else if (!strncmp(str, "cbmemsize=", 10)) {
				pci_cardbus_mem_size = memparse(str + 10, &str);
3148 3149 3150
			} else if (!strncmp(str, "resource_alignment=", 19)) {
				pci_set_resource_alignment_param(str + 19,
							strlen(str + 19));
3151 3152
			} else if (!strncmp(str, "ecrc=", 5)) {
				pcie_ecrc_get_policy(str + 5);
3153 3154 3155 3156
			} else if (!strncmp(str, "hpiosize=", 9)) {
				pci_hotplug_io_size = memparse(str + 9, &str);
			} else if (!strncmp(str, "hpmemsize=", 10)) {
				pci_hotplug_mem_size = memparse(str + 10, &str);
3157 3158 3159 3160
			} else {
				printk(KERN_ERR "PCI: Unknown option `%s'\n",
						str);
			}
L
Linus Torvalds 已提交
3161 3162 3163
		}
		str = k;
	}
3164
	return 0;
L
Linus Torvalds 已提交
3165
}
3166
early_param("pci", pci_setup);
L
Linus Torvalds 已提交
3167

3168
EXPORT_SYMBOL(pci_reenable_device);
3169 3170
EXPORT_SYMBOL(pci_enable_device_io);
EXPORT_SYMBOL(pci_enable_device_mem);
L
Linus Torvalds 已提交
3171
EXPORT_SYMBOL(pci_enable_device);
T
Tejun Heo 已提交
3172 3173
EXPORT_SYMBOL(pcim_enable_device);
EXPORT_SYMBOL(pcim_pin_device);
L
Linus Torvalds 已提交
3174 3175 3176 3177 3178
EXPORT_SYMBOL(pci_disable_device);
EXPORT_SYMBOL(pci_find_capability);
EXPORT_SYMBOL(pci_bus_find_capability);
EXPORT_SYMBOL(pci_release_regions);
EXPORT_SYMBOL(pci_request_regions);
3179
EXPORT_SYMBOL(pci_request_regions_exclusive);
L
Linus Torvalds 已提交
3180 3181
EXPORT_SYMBOL(pci_release_region);
EXPORT_SYMBOL(pci_request_region);
3182
EXPORT_SYMBOL(pci_request_region_exclusive);
3183 3184
EXPORT_SYMBOL(pci_release_selected_regions);
EXPORT_SYMBOL(pci_request_selected_regions);
3185
EXPORT_SYMBOL(pci_request_selected_regions_exclusive);
L
Linus Torvalds 已提交
3186
EXPORT_SYMBOL(pci_set_master);
3187
EXPORT_SYMBOL(pci_clear_master);
L
Linus Torvalds 已提交
3188
EXPORT_SYMBOL(pci_set_mwi);
R
Randy Dunlap 已提交
3189
EXPORT_SYMBOL(pci_try_set_mwi);
L
Linus Torvalds 已提交
3190
EXPORT_SYMBOL(pci_clear_mwi);
B
Brett M Russ 已提交
3191
EXPORT_SYMBOL_GPL(pci_intx);
L
Linus Torvalds 已提交
3192 3193
EXPORT_SYMBOL(pci_assign_resource);
EXPORT_SYMBOL(pci_find_parent_resource);
3194
EXPORT_SYMBOL(pci_select_bars);
L
Linus Torvalds 已提交
3195 3196 3197 3198

EXPORT_SYMBOL(pci_set_power_state);
EXPORT_SYMBOL(pci_save_state);
EXPORT_SYMBOL(pci_restore_state);
3199
EXPORT_SYMBOL(pci_pme_capable);
3200
EXPORT_SYMBOL(pci_pme_active);
3201
EXPORT_SYMBOL(pci_wake_from_d3);
3202
EXPORT_SYMBOL(pci_target_state);
3203 3204
EXPORT_SYMBOL(pci_prepare_to_sleep);
EXPORT_SYMBOL(pci_back_from_sleep);
B
Brian King 已提交
3205
EXPORT_SYMBOL_GPL(pci_set_pcie_reset_state);