pci.c 129.7 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12
/*
 *	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>
13 14
#include <linux/of.h>
#include <linux/of_pci.h>
L
Linus Torvalds 已提交
15
#include <linux/pci.h>
16
#include <linux/pm.h>
17
#include <linux/slab.h>
L
Linus Torvalds 已提交
18 19
#include <linux/module.h>
#include <linux/spinlock.h>
T
Tim Schmielau 已提交
20
#include <linux/string.h>
21
#include <linux/log2.h>
S
Shaohua Li 已提交
22
#include <linux/pci-aspm.h>
23
#include <linux/pm_wakeup.h>
24
#include <linux/interrupt.h>
25
#include <linux/device.h>
26
#include <linux/pm_runtime.h>
27
#include <linux/pci_hotplug.h>
28
#include <asm/setup.h>
29
#include <linux/aer.h>
30
#include "pci.h"
L
Linus Torvalds 已提交
31

A
Alan Stern 已提交
32 33 34 35 36
const char *pci_power_names[] = {
	"error", "D0", "D1", "D2", "D3hot", "D3cold", "unknown",
};
EXPORT_SYMBOL_GPL(pci_power_names);

37 38 39 40 41 42
int isa_dma_bridge_buggy;
EXPORT_SYMBOL(isa_dma_bridge_buggy);

int pci_pci_problems;
EXPORT_SYMBOL(pci_pci_problems);

43 44
unsigned int pci_pm_d3_delay;

45 46 47 48 49 50 51 52 53 54 55 56 57
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 */

58 59 60 61 62 63 64 65 66
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
67

68 69 70 71
#ifdef CONFIG_PCI_DOMAINS
int pci_domains_supported = 1;
#endif

72 73 74 75 76 77
#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;

78 79 80 81 82 83
#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;

84
enum pcie_bus_config_types pcie_bus_config = PCIE_BUS_DEFAULT;
85

86 87 88 89 90 91
/*
 * 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.
 */
B
Bill Pemberton 已提交
92
u8 pci_dfl_cache_line_size = L1_CACHE_BYTES >> 2;
93 94
u8 pci_cache_line_size;

95 96 97 98 99 100
/*
 * If we set up a device for bus mastering, we need to check the latency
 * timer as certain BIOSes forget to set it properly.
 */
unsigned int pcibios_max_latency = 255;

101 102 103
/* If set, the PCIe ARI capability will not be used. */
static bool pcie_ari_disabled;

L
Linus Torvalds 已提交
104 105 106 107 108 109 110
/**
 * 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.
 */
111
unsigned char pci_bus_max_busnr(struct pci_bus *bus)
L
Linus Torvalds 已提交
112
{
113
	struct pci_bus *tmp;
L
Linus Torvalds 已提交
114 115
	unsigned char max, n;

116
	max = bus->busn_res.end;
117 118
	list_for_each_entry(tmp, &bus->children, node) {
		n = pci_bus_max_busnr(tmp);
R
Ryan Desfosses 已提交
119
		if (n > max)
L
Linus Torvalds 已提交
120 121 122 123
			max = n;
	}
	return max;
}
124
EXPORT_SYMBOL_GPL(pci_bus_max_busnr);
L
Linus Torvalds 已提交
125

A
Andrew Morton 已提交
126 127 128
#ifdef CONFIG_HAS_IOMEM
void __iomem *pci_ioremap_bar(struct pci_dev *pdev, int bar)
{
129 130
	struct resource *res = &pdev->resource[bar];

A
Andrew Morton 已提交
131 132 133
	/*
	 * Make sure the BAR is actually a memory resource, not an IO resource
	 */
134
	if (res->flags & IORESOURCE_UNSET || !(res->flags & IORESOURCE_MEM)) {
135
		dev_warn(&pdev->dev, "can't ioremap BAR %d: %pR\n", bar, res);
A
Andrew Morton 已提交
136 137
		return NULL;
	}
138
	return ioremap_nocache(res->start, resource_size(res));
A
Andrew Morton 已提交
139 140
}
EXPORT_SYMBOL_GPL(pci_ioremap_bar);
141 142 143 144 145 146 147 148 149 150 151 152 153 154

void __iomem *pci_ioremap_wc_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_wc(pci_resource_start(pdev, bar),
			  pci_resource_len(pdev, bar));
}
EXPORT_SYMBOL_GPL(pci_ioremap_wc_bar);
A
Andrew Morton 已提交
155 156
#endif

157 158 159

static int __pci_find_next_cap_ttl(struct pci_bus *bus, unsigned int devfn,
				   u8 pos, int cap, int *ttl)
160 161
{
	u8 id;
162 163 164
	u16 ent;

	pci_bus_read_config_byte(bus, devfn, pos, &pos);
165

166
	while ((*ttl)--) {
167 168 169
		if (pos < 0x40)
			break;
		pos &= ~3;
170 171 172
		pci_bus_read_config_word(bus, devfn, pos, &ent);

		id = ent & 0xff;
173 174 175 176
		if (id == 0xff)
			break;
		if (id == cap)
			return pos;
177
		pos = (ent >> 8);
178 179 180 181
	}
	return 0;
}

182 183 184 185 186 187 188 189
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);
}

190 191 192 193 194 195 196
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);

197 198
static int __pci_bus_find_cap_start(struct pci_bus *bus,
				    unsigned int devfn, u8 hdr_type)
L
Linus Torvalds 已提交
199 200 201 202 203 204 205 206 207 208
{
	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:
209
		return PCI_CAPABILITY_LIST;
L
Linus Torvalds 已提交
210
	case PCI_HEADER_TYPE_CARDBUS:
211
		return PCI_CB_CAPABILITY_LIST;
L
Linus Torvalds 已提交
212
	}
213 214

	return 0;
L
Linus Torvalds 已提交
215 216 217
}

/**
218
 * pci_find_capability - query for devices' capabilities
L
Linus Torvalds 已提交
219 220 221 222 223 224 225 226
 * @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:
 *
227 228 229 230
 *  %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
L
Linus Torvalds 已提交
231
 *  %PCI_CAP_ID_MSI          Message Signalled Interrupts
232
 *  %PCI_CAP_ID_CHSWP        CompactPCI HotSwap
L
Linus Torvalds 已提交
233 234 235 236 237
 *  %PCI_CAP_ID_PCIX         PCI-X
 *  %PCI_CAP_ID_EXP          PCI Express
 */
int pci_find_capability(struct pci_dev *dev, int cap)
{
238 239 240 241 242 243 244
	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
245
}
246
EXPORT_SYMBOL(pci_find_capability);
L
Linus Torvalds 已提交
247 248

/**
249
 * pci_bus_find_capability - query for devices' capabilities
L
Linus Torvalds 已提交
250 251 252 253 254
 * @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
255
 * pci_dev structure set up yet.
L
Linus Torvalds 已提交
256 257 258 259 260 261 262
 *
 * 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)
{
263
	int pos;
L
Linus Torvalds 已提交
264 265 266 267
	u8 hdr_type;

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

268 269 270 271 272
	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
273
}
274
EXPORT_SYMBOL(pci_bus_find_capability);
L
Linus Torvalds 已提交
275 276

/**
277
 * pci_find_next_ext_capability - Find an extended capability
L
Linus Torvalds 已提交
278
 * @dev: PCI device to query
279
 * @start: address at which to start looking (0 to start at beginning of list)
L
Linus Torvalds 已提交
280 281
 * @cap: capability code
 *
282
 * Returns the address of the next matching extended capability structure
L
Linus Torvalds 已提交
283
 * within the device's PCI configuration space or 0 if the device does
284 285
 * not support it.  Some capabilities can occur several times, e.g., the
 * vendor-specific capability, and this provides a way to find them all.
L
Linus Torvalds 已提交
286
 */
287
int pci_find_next_ext_capability(struct pci_dev *dev, int start, int cap)
L
Linus Torvalds 已提交
288 289
{
	u32 header;
290 291
	int ttl;
	int pos = PCI_CFG_SPACE_SIZE;
L
Linus Torvalds 已提交
292

293 294 295 296
	/* 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
297 298
		return 0;

299 300 301
	if (start)
		pos = start;

L
Linus Torvalds 已提交
302 303 304 305 306 307 308 309 310 311 312
	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) {
313
		if (PCI_EXT_CAP_ID(header) == cap && pos != start)
L
Linus Torvalds 已提交
314 315 316
			return pos;

		pos = PCI_EXT_CAP_NEXT(header);
317
		if (pos < PCI_CFG_SPACE_SIZE)
L
Linus Torvalds 已提交
318 319 320 321 322 323 324 325
			break;

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

	return 0;
}
326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345
EXPORT_SYMBOL_GPL(pci_find_next_ext_capability);

/**
 * 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)
{
	return pci_find_next_ext_capability(dev, 0, cap);
}
346
EXPORT_SYMBOL_GPL(pci_find_ext_capability);
L
Linus Torvalds 已提交
347

348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367
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;

368 369
		pos = __pci_find_next_cap_ttl(dev->bus, dev->devfn,
					      pos + PCI_CAP_LIST_NEXT,
370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416
					      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
417 418 419 420 421 422
/**
 * 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
423
 *  region of parent bus the given region is contained in.
L
Linus Torvalds 已提交
424
 */
R
Ryan Desfosses 已提交
425 426
struct resource *pci_find_parent_resource(const struct pci_dev *dev,
					  struct resource *res)
L
Linus Torvalds 已提交
427 428
{
	const struct pci_bus *bus = dev->bus;
429
	struct resource *r;
L
Linus Torvalds 已提交
430 431
	int i;

432
	pci_bus_for_each_resource(bus, r, i) {
L
Linus Torvalds 已提交
433 434
		if (!r)
			continue;
435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454
		if (res->start && resource_contains(r, res)) {

			/*
			 * If the window is prefetchable but the BAR is
			 * not, the allocator made a mistake.
			 */
			if (r->flags & IORESOURCE_PREFETCH &&
			    !(res->flags & IORESOURCE_PREFETCH))
				return NULL;

			/*
			 * If we're below a transparent bridge, there may
			 * be both a positively-decoded aperture and a
			 * subtractively-decoded region that contain the BAR.
			 * We want the positively-decoded one, so this depends
			 * on pci_bus_for_each_resource() giving us those
			 * first.
			 */
			return r;
		}
L
Linus Torvalds 已提交
455
	}
456
	return NULL;
L
Linus Torvalds 已提交
457
}
458
EXPORT_SYMBOL(pci_find_parent_resource);
L
Linus Torvalds 已提交
459

460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483
/**
 * pci_find_pcie_root_port - return PCIe Root Port
 * @dev: PCI device to query
 *
 * Traverse up the parent chain and return the PCIe Root Port PCI Device
 * for a given PCI Device.
 */
struct pci_dev *pci_find_pcie_root_port(struct pci_dev *dev)
{
	struct pci_dev *bridge, *highest_pcie_bridge = NULL;

	bridge = pci_upstream_bridge(dev);
	while (bridge && pci_is_pcie(bridge)) {
		highest_pcie_bridge = bridge;
		bridge = pci_upstream_bridge(bridge);
	}

	if (pci_pcie_type(highest_pcie_bridge) != PCI_EXP_TYPE_ROOT_PORT)
		return NULL;

	return highest_pcie_bridge;
}
EXPORT_SYMBOL(pci_find_pcie_root_port);

484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509
/**
 * pci_wait_for_pending - wait for @mask bit(s) to clear in status word @pos
 * @dev: the PCI device to operate on
 * @pos: config space offset of status word
 * @mask: mask of bit(s) to care about in status word
 *
 * Return 1 when mask bit(s) in status word clear, 0 otherwise.
 */
int pci_wait_for_pending(struct pci_dev *dev, int pos, u16 mask)
{
	int i;

	/* Wait for Transaction Pending bit clean */
	for (i = 0; i < 4; i++) {
		u16 status;
		if (i)
			msleep((1 << (i - 1)) * 100);

		pci_read_config_word(dev, pos, &status);
		if (!(status & mask))
			return 1;
	}

	return 0;
}

510
/**
W
Wei Yang 已提交
511
 * pci_restore_bars - restore a device's BAR values (e.g. after wake-up)
512 513 514 515 516
 * @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.
 */
R
Ryan Desfosses 已提交
517
static void pci_restore_bars(struct pci_dev *dev)
518
{
519
	int i;
520

W
Wei Yang 已提交
521 522 523 524
	/* Per SR-IOV spec 3.4.1.11, VF BARs are RO zero */
	if (dev->is_virtfn)
		return;

525
	for (i = 0; i < PCI_BRIDGE_RESOURCES; i++)
526
		pci_update_resource(dev, i);
527 528
}

529
static const struct pci_platform_pm_ops *pci_platform_pm;
530

531
int pci_set_platform_pm(const struct pci_platform_pm_ops *ops)
532
{
533
	if (!ops->is_manageable || !ops->set_state || !ops->choose_state
534
	    || !ops->sleep_wake)
535 536 537 538 539 540 541 542 543 544 545
		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,
R
Ryan Desfosses 已提交
546
					       pci_power_t t)
547 548 549 550 551 552 553 554 555
{
	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
556

557 558 559 560 561 562
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;
}

563 564 565 566 567 568
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;
}

569 570 571 572 573
static inline bool platform_pci_need_resume(struct pci_dev *dev)
{
	return pci_platform_pm ? pci_platform_pm->need_resume(dev) : false;
}

L
Linus Torvalds 已提交
574
/**
575 576 577 578
 * 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
579
 *
580 581 582 583 584 585
 * 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
586
 */
587
static int pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state)
L
Linus Torvalds 已提交
588
{
589
	u16 pmcsr;
590
	bool need_restore = false;
L
Linus Torvalds 已提交
591

592 593 594 595
	/* Check if we're already there */
	if (dev->current_state == state)
		return 0;

596
	if (!dev->pm_cap)
597 598
		return -EIO;

599 600 601
	if (state < PCI_D0 || state > PCI_D3hot)
		return -EINVAL;

L
Linus Torvalds 已提交
602
	/* Validate current state:
603
	 * Can enter D0 from any state, but if we can only go deeper
L
Linus Torvalds 已提交
604 605
	 * to sleep if we're already in a low power state
	 */
606
	if (state != PCI_D0 && dev->current_state <= PCI_D3cold
607
	    && dev->current_state > state) {
608 609
		dev_err(&dev->dev, "invalid power transition (from state %d to %d)\n",
			dev->current_state, state);
L
Linus Torvalds 已提交
610
		return -EINVAL;
611
	}
L
Linus Torvalds 已提交
612 613

	/* check if this device supports the desired state */
614 615
	if ((state == PCI_D1 && !dev->d1_support)
	   || (state == PCI_D2 && !dev->d2_support))
616
		return -EIO;
L
Linus Torvalds 已提交
617

618
	pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
619

620
	/* If we're (effectively) in D3, force entire word to 0.
L
Linus Torvalds 已提交
621 622 623
	 * This doesn't affect PME_Status, disables PME_En, and
	 * sets PowerState to 0.
	 */
624
	switch (dev->current_state) {
625 626 627 628 629 630
	case PCI_D0:
	case PCI_D1:
	case PCI_D2:
		pmcsr &= ~PCI_PM_CTRL_STATE_MASK;
		pmcsr |= state;
		break;
631 632
	case PCI_D3hot:
	case PCI_D3cold:
633 634
	case PCI_UNKNOWN: /* Boot-up */
		if ((pmcsr & PCI_PM_CTRL_STATE_MASK) == PCI_D3hot
635
		 && !(pmcsr & PCI_PM_CTRL_NO_SOFT_RESET))
636
			need_restore = true;
637 638
		/* Fall-through: force to D0 */
	default:
639
		pmcsr = 0;
640
		break;
L
Linus Torvalds 已提交
641 642 643
	}

	/* enter specified state */
644
	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, pmcsr);
L
Linus Torvalds 已提交
645 646 647 648

	/* Mandatory power management transition delays */
	/* see PCI PM 1.1 5.6.1 table 18 */
	if (state == PCI_D3hot || dev->current_state == PCI_D3hot)
649
		pci_dev_d3_sleep(dev);
L
Linus Torvalds 已提交
650
	else if (state == PCI_D2 || dev->current_state == PCI_D2)
651
		udelay(PCI_PM_D2_DELAY);
L
Linus Torvalds 已提交
652

653 654 655
	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())
656 657
		dev_info(&dev->dev, "Refused to change power state, currently in D%d\n",
			 dev->current_state);
658

659 660
	/*
	 * According to section 5.4.1 of the "PCI BUS POWER MANAGEMENT
661 662 663 664 665 666 667 668 669 670 671 672 673 674
	 * 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);

675
	if (dev->bus->self)
S
Shaohua Li 已提交
676 677
		pcie_aspm_pm_state_change(dev->bus->self);

L
Linus Torvalds 已提交
678 679 680
	return 0;
}

681 682 683 684
/**
 * pci_update_current_state - Read PCI power state of given device from its
 *                            PCI PM registers and cache it
 * @dev: PCI device to handle.
685
 * @state: State to cache in case the device doesn't have the PM capability
686
 */
687
void pci_update_current_state(struct pci_dev *dev, pci_power_t state)
688
{
689
	if (dev->pm_cap) {
690 691
		u16 pmcsr;

692 693 694 695 696 697 698 699 700 701
		/*
		 * Configuration space is not accessible for device in
		 * D3cold, so just keep or set D3cold for safety
		 */
		if (dev->current_state == PCI_D3cold)
			return;
		if (state == PCI_D3cold) {
			dev->current_state = PCI_D3cold;
			return;
		}
702
		pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
703
		dev->current_state = (pmcsr & PCI_PM_CTRL_STATE_MASK);
704 705
	} else {
		dev->current_state = state;
706 707 708
	}
}

709 710 711 712 713 714 715 716 717 718 719 720 721
/**
 * pci_power_up - Put the given device into D0 forcibly
 * @dev: PCI device to power up
 */
void pci_power_up(struct pci_dev *dev)
{
	if (platform_pci_power_manageable(dev))
		platform_pci_set_power_state(dev, PCI_D0);

	pci_raw_set_power_state(dev, PCI_D0);
	pci_update_current_state(dev, PCI_D0);
}

722 723 724 725 726 727 728 729 730 731 732 733 734
/**
 * 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);
735
	} else
736
		error = -ENODEV;
737 738 739

	if (error && !dev->pm_cap) /* Fall back to PCI_D0 */
		dev->current_state = PCI_D0;
740 741 742 743

	return error;
}

744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765
/**
 * pci_wakeup - Wake up a PCI device
 * @pci_dev: Device to handle.
 * @ign: ignored parameter
 */
static int pci_wakeup(struct pci_dev *pci_dev, void *ign)
{
	pci_wakeup_event(pci_dev);
	pm_request_resume(&pci_dev->dev);
	return 0;
}

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

766 767 768 769 770 771 772
/**
 * __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)
{
773
	if (state == PCI_D0) {
774
		pci_platform_power_transition(dev, PCI_D0);
775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816
		/*
		 * Mandatory power management transition delays, see
		 * PCI Express Base Specification Revision 2.0 Section
		 * 6.6.1: Conventional Reset.  Do not delay for
		 * devices powered on/off by corresponding bridge,
		 * because have already delayed for the bridge.
		 */
		if (dev->runtime_d3cold) {
			msleep(dev->d3cold_delay);
			/*
			 * When powering on a bridge from D3cold, the
			 * whole hierarchy may be powered on into
			 * D0uninitialized state, resume them to give
			 * them a chance to suspend again
			 */
			pci_wakeup_bus(dev->subordinate);
		}
	}
}

/**
 * __pci_dev_set_current_state - Set current state of a PCI device
 * @dev: Device to handle
 * @data: pointer to state to be set
 */
static int __pci_dev_set_current_state(struct pci_dev *dev, void *data)
{
	pci_power_t state = *(pci_power_t *)data;

	dev->current_state = state;
	return 0;
}

/**
 * __pci_bus_set_current_state - Walk given bus and set current state of devices
 * @bus: Top bus of the subtree to walk.
 * @state: state to be set
 */
static void __pci_bus_set_current_state(struct pci_bus *bus, pci_power_t state)
{
	if (bus)
		pci_walk_bus(bus, __pci_dev_set_current_state, &state);
817 818 819 820 821 822 823 824 825 826 827
}

/**
 * __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)
{
828 829
	int ret;

830
	if (state <= PCI_D0)
831 832 833 834 835 836
		return -EINVAL;
	ret = pci_platform_power_transition(dev, state);
	/* Power off the bridge may power off the whole hierarchy */
	if (!ret && state == PCI_D3cold)
		__pci_bus_set_current_state(dev->subordinate, PCI_D3cold);
	return ret;
837 838 839
}
EXPORT_SYMBOL_GPL(__pci_complete_power_transition);

840 841 842 843 844
/**
 * 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.
 *
845
 * Transition a device to a new power state, using the platform firmware and/or
846 847 848 849 850 851 852 853 854 855 856
 * 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)
{
857
	int error;
858 859

	/* bound the state we're entering */
860 861
	if (state > PCI_D3cold)
		state = PCI_D3cold;
862 863 864 865 866 867 868 869 870 871
	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;

872 873 874 875
	/* Check if we're already there */
	if (dev->current_state == state)
		return 0;

876 877
	__pci_start_power_transition(dev, state);

878 879
	/* This device is quirked not to be put into D3, so
	   don't put it in D3 */
880
	if (state >= PCI_D3hot && (dev->dev_flags & PCI_DEV_FLAGS_NO_D3))
881
		return 0;
882

883 884 885 886 887 888
	/*
	 * To put device in D3cold, we put device into D3hot in native
	 * way, then put device into D3cold with platform ops
	 */
	error = pci_raw_set_power_state(dev, state > PCI_D3hot ?
					PCI_D3hot : state);
889

890 891
	if (!__pci_complete_power_transition(dev, state))
		error = 0;
892 893 894

	return error;
}
895
EXPORT_SYMBOL(pci_set_power_state);
896

L
Linus Torvalds 已提交
897 898 899 900 901 902 903 904 905 906 907 908
/**
 * 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)
{
909
	pci_power_t ret;
910

911
	if (!dev->pm_cap)
L
Linus Torvalds 已提交
912 913
		return PCI_D0;

914 915 916
	ret = platform_pci_choose_state(dev);
	if (ret != PCI_POWER_ERROR)
		return ret;
917 918 919 920 921

	switch (state.event) {
	case PM_EVENT_ON:
		return PCI_D0;
	case PM_EVENT_FREEZE:
922 923
	case PM_EVENT_PRETHAW:
		/* REVISIT both freeze and pre-thaw "should" use D0 */
924
	case PM_EVENT_SUSPEND:
925
	case PM_EVENT_HIBERNATE:
926
		return PCI_D3hot;
L
Linus Torvalds 已提交
927
	default:
928 929
		dev_info(&dev->dev, "unrecognized suspend event %d\n",
			 state.event);
L
Linus Torvalds 已提交
930 931 932 933 934 935
		BUG();
	}
	return PCI_D0;
}
EXPORT_SYMBOL(pci_choose_state);

936 937
#define PCI_EXP_SAVE_REGS	7

938 939
static struct pci_cap_saved_state *_pci_find_saved_cap(struct pci_dev *pci_dev,
						       u16 cap, bool extended)
940 941 942
{
	struct pci_cap_saved_state *tmp;

943
	hlist_for_each_entry(tmp, &pci_dev->saved_cap_space, next) {
944
		if (tmp->cap.cap_extended == extended && tmp->cap.cap_nr == cap)
945 946 947 948 949
			return tmp;
	}
	return NULL;
}

950 951 952 953 954 955 956 957 958 959
struct pci_cap_saved_state *pci_find_saved_cap(struct pci_dev *dev, char cap)
{
	return _pci_find_saved_cap(dev, cap, false);
}

struct pci_cap_saved_state *pci_find_saved_ext_cap(struct pci_dev *dev, u16 cap)
{
	return _pci_find_saved_cap(dev, cap, true);
}

960 961
static int pci_save_pcie_state(struct pci_dev *dev)
{
962
	int i = 0;
963 964 965
	struct pci_cap_saved_state *save_state;
	u16 *cap;

966
	if (!pci_is_pcie(dev))
967 968
		return 0;

969
	save_state = pci_find_saved_cap(dev, PCI_CAP_ID_EXP);
970
	if (!save_state) {
971
		dev_err(&dev->dev, "buffer not found in %s\n", __func__);
972 973
		return -ENOMEM;
	}
974

975 976 977 978 979 980 981 982
	cap = (u16 *)&save_state->cap.data[0];
	pcie_capability_read_word(dev, PCI_EXP_DEVCTL, &cap[i++]);
	pcie_capability_read_word(dev, PCI_EXP_LNKCTL, &cap[i++]);
	pcie_capability_read_word(dev, PCI_EXP_SLTCTL, &cap[i++]);
	pcie_capability_read_word(dev, PCI_EXP_RTCTL,  &cap[i++]);
	pcie_capability_read_word(dev, PCI_EXP_DEVCTL2, &cap[i++]);
	pcie_capability_read_word(dev, PCI_EXP_LNKCTL2, &cap[i++]);
	pcie_capability_read_word(dev, PCI_EXP_SLTCTL2, &cap[i++]);
983

984 985 986 987 988
	return 0;
}

static void pci_restore_pcie_state(struct pci_dev *dev)
{
989
	int i = 0;
990 991 992 993
	struct pci_cap_saved_state *save_state;
	u16 *cap;

	save_state = pci_find_saved_cap(dev, PCI_CAP_ID_EXP);
994
	if (!save_state)
995 996
		return;

997 998 999 1000 1001 1002 1003 1004
	cap = (u16 *)&save_state->cap.data[0];
	pcie_capability_write_word(dev, PCI_EXP_DEVCTL, cap[i++]);
	pcie_capability_write_word(dev, PCI_EXP_LNKCTL, cap[i++]);
	pcie_capability_write_word(dev, PCI_EXP_SLTCTL, cap[i++]);
	pcie_capability_write_word(dev, PCI_EXP_RTCTL, cap[i++]);
	pcie_capability_write_word(dev, PCI_EXP_DEVCTL2, cap[i++]);
	pcie_capability_write_word(dev, PCI_EXP_LNKCTL2, cap[i++]);
	pcie_capability_write_word(dev, PCI_EXP_SLTCTL2, cap[i++]);
1005 1006
}

S
Stephen Hemminger 已提交
1007 1008 1009

static int pci_save_pcix_state(struct pci_dev *dev)
{
1010
	int pos;
S
Stephen Hemminger 已提交
1011 1012 1013
	struct pci_cap_saved_state *save_state;

	pos = pci_find_capability(dev, PCI_CAP_ID_PCIX);
1014
	if (!pos)
S
Stephen Hemminger 已提交
1015 1016
		return 0;

1017
	save_state = pci_find_saved_cap(dev, PCI_CAP_ID_PCIX);
S
Stephen Hemminger 已提交
1018
	if (!save_state) {
1019
		dev_err(&dev->dev, "buffer not found in %s\n", __func__);
S
Stephen Hemminger 已提交
1020 1021 1022
		return -ENOMEM;
	}

1023 1024
	pci_read_config_word(dev, pos + PCI_X_CMD,
			     (u16 *)save_state->cap.data);
1025

S
Stephen Hemminger 已提交
1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036
	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);
1037
	if (!save_state || !pos)
S
Stephen Hemminger 已提交
1038
		return;
1039
	cap = (u16 *)&save_state->cap.data[0];
S
Stephen Hemminger 已提交
1040 1041 1042 1043 1044

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


L
Linus Torvalds 已提交
1045 1046 1047 1048
/**
 * pci_save_state - save the PCI configuration space of a device before suspending
 * @dev: - PCI device that we're dealing with
 */
R
Ryan Desfosses 已提交
1049
int pci_save_state(struct pci_dev *dev)
L
Linus Torvalds 已提交
1050 1051 1052 1053
{
	int i;
	/* XXX: 100% dword access ok here? */
	for (i = 0; i < 16; i++)
1054
		pci_read_config_dword(dev, i * 4, &dev->saved_config_space[i]);
1055
	dev->state_saved = true;
1056 1057 1058

	i = pci_save_pcie_state(dev);
	if (i != 0)
1059
		return i;
1060 1061 1062

	i = pci_save_pcix_state(dev);
	if (i != 0)
S
Stephen Hemminger 已提交
1063
		return i;
1064

1065
	return pci_save_vc_state(dev);
L
Linus Torvalds 已提交
1066
}
1067
EXPORT_SYMBOL(pci_save_state);
L
Linus Torvalds 已提交
1068

1069 1070 1071 1072 1073 1074 1075 1076 1077 1078
static void pci_restore_config_dword(struct pci_dev *pdev, int offset,
				     u32 saved_val, int retry)
{
	u32 val;

	pci_read_config_dword(pdev, offset, &val);
	if (val == saved_val)
		return;

	for (;;) {
1079 1080
		dev_dbg(&pdev->dev, "restoring config space at offset %#x (was %#x, writing %#x)\n",
			offset, val, saved_val);
1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092
		pci_write_config_dword(pdev, offset, saved_val);
		if (retry-- <= 0)
			return;

		pci_read_config_dword(pdev, offset, &val);
		if (val == saved_val)
			return;

		mdelay(1);
	}
}

1093 1094
static void pci_restore_config_space_range(struct pci_dev *pdev,
					   int start, int end, int retry)
1095 1096 1097 1098 1099 1100 1101 1102 1103
{
	int index;

	for (index = end; index >= start; index--)
		pci_restore_config_dword(pdev, 4 * index,
					 pdev->saved_config_space[index],
					 retry);
}

1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115
static void pci_restore_config_space(struct pci_dev *pdev)
{
	if (pdev->hdr_type == PCI_HEADER_TYPE_NORMAL) {
		pci_restore_config_space_range(pdev, 10, 15, 0);
		/* Restore BARs before the command register. */
		pci_restore_config_space_range(pdev, 4, 9, 10);
		pci_restore_config_space_range(pdev, 0, 3, 0);
	} else {
		pci_restore_config_space_range(pdev, 0, 15, 0);
	}
}

1116
/**
L
Linus Torvalds 已提交
1117 1118 1119
 * pci_restore_state - Restore the saved state of a PCI device
 * @dev: - PCI device that we're dealing with
 */
1120
void pci_restore_state(struct pci_dev *dev)
L
Linus Torvalds 已提交
1121
{
A
Alek Du 已提交
1122
	if (!dev->state_saved)
1123
		return;
1124

1125 1126
	/* PCI Express register must be restored first */
	pci_restore_pcie_state(dev);
1127
	pci_restore_ats_state(dev);
1128
	pci_restore_vc_state(dev);
1129

1130 1131
	pci_cleanup_aer_error_status_regs(dev);

1132
	pci_restore_config_space(dev);
1133

S
Stephen Hemminger 已提交
1134
	pci_restore_pcix_state(dev);
1135
	pci_restore_msi_state(dev);
1136 1137 1138

	/* Restore ACS and IOV configuration state */
	pci_enable_acs(dev);
Y
Yu Zhao 已提交
1139
	pci_restore_iov_state(dev);
1140

1141
	dev->state_saved = false;
L
Linus Torvalds 已提交
1142
}
1143
EXPORT_SYMBOL(pci_restore_state);
L
Linus Torvalds 已提交
1144

1145 1146 1147 1148 1149 1150 1151 1152 1153 1154
struct pci_saved_state {
	u32 config_space[16];
	struct pci_cap_saved_data cap[0];
};

/**
 * pci_store_saved_state - Allocate and return an opaque struct containing
 *			   the device saved state.
 * @dev: PCI device that we're dealing with
 *
1155
 * Return NULL if no state or error.
1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168
 */
struct pci_saved_state *pci_store_saved_state(struct pci_dev *dev)
{
	struct pci_saved_state *state;
	struct pci_cap_saved_state *tmp;
	struct pci_cap_saved_data *cap;
	size_t size;

	if (!dev->state_saved)
		return NULL;

	size = sizeof(*state) + sizeof(struct pci_cap_saved_data);

1169
	hlist_for_each_entry(tmp, &dev->saved_cap_space, next)
1170 1171 1172 1173 1174 1175 1176 1177 1178 1179
		size += sizeof(struct pci_cap_saved_data) + tmp->cap.size;

	state = kzalloc(size, GFP_KERNEL);
	if (!state)
		return NULL;

	memcpy(state->config_space, dev->saved_config_space,
	       sizeof(state->config_space));

	cap = state->cap;
1180
	hlist_for_each_entry(tmp, &dev->saved_cap_space, next) {
1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195
		size_t len = sizeof(struct pci_cap_saved_data) + tmp->cap.size;
		memcpy(cap, &tmp->cap, len);
		cap = (struct pci_cap_saved_data *)((u8 *)cap + len);
	}
	/* Empty cap_save terminates list */

	return state;
}
EXPORT_SYMBOL_GPL(pci_store_saved_state);

/**
 * pci_load_saved_state - Reload the provided save state into struct pci_dev.
 * @dev: PCI device that we're dealing with
 * @state: Saved state returned from pci_store_saved_state()
 */
1196 1197
int pci_load_saved_state(struct pci_dev *dev,
			 struct pci_saved_state *state)
1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212
{
	struct pci_cap_saved_data *cap;

	dev->state_saved = false;

	if (!state)
		return 0;

	memcpy(dev->saved_config_space, state->config_space,
	       sizeof(state->config_space));

	cap = state->cap;
	while (cap->size) {
		struct pci_cap_saved_state *tmp;

1213
		tmp = _pci_find_saved_cap(dev, cap->cap_nr, cap->cap_extended);
1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224
		if (!tmp || tmp->cap.size != cap->size)
			return -EINVAL;

		memcpy(tmp->cap.data, cap->data, tmp->cap.size);
		cap = (struct pci_cap_saved_data *)((u8 *)cap +
		       sizeof(struct pci_cap_saved_data) + cap->size);
	}

	dev->state_saved = true;
	return 0;
}
1225
EXPORT_SYMBOL_GPL(pci_load_saved_state);
1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242

/**
 * pci_load_and_free_saved_state - Reload the save state pointed to by state,
 *				   and free the memory allocated for it.
 * @dev: PCI device that we're dealing with
 * @state: Pointer to saved state returned from pci_store_saved_state()
 */
int pci_load_and_free_saved_state(struct pci_dev *dev,
				  struct pci_saved_state **state)
{
	int ret = pci_load_saved_state(dev, *state);
	kfree(*state);
	*state = NULL;
	return ret;
}
EXPORT_SYMBOL_GPL(pci_load_and_free_saved_state);

1243 1244 1245 1246 1247
int __weak pcibios_enable_device(struct pci_dev *dev, int bars)
{
	return pci_enable_resources(dev, bars);
}

1248 1249 1250
static int do_pci_enable_device(struct pci_dev *dev, int bars)
{
	int err;
1251
	struct pci_dev *bridge;
1252 1253
	u16 cmd;
	u8 pin;
1254 1255 1256 1257

	err = pci_set_power_state(dev, PCI_D0);
	if (err < 0 && err != -EIO)
		return err;
1258 1259 1260 1261 1262

	bridge = pci_upstream_bridge(dev);
	if (bridge)
		pcie_aspm_powersave_config_link(bridge);

1263 1264 1265 1266 1267
	err = pcibios_enable_device(dev, bars);
	if (err < 0)
		return err;
	pci_fixup_device(pci_fixup_enable, dev);

1268 1269 1270
	if (dev->msi_enabled || dev->msix_enabled)
		return 0;

1271 1272 1273 1274 1275 1276 1277 1278
	pci_read_config_byte(dev, PCI_INTERRUPT_PIN, &pin);
	if (pin) {
		pci_read_config_word(dev, PCI_COMMAND, &cmd);
		if (cmd & PCI_COMMAND_INTX_DISABLE)
			pci_write_config_word(dev, PCI_COMMAND,
					      cmd & ~PCI_COMMAND_INTX_DISABLE);
	}

1279 1280 1281 1282
	return 0;
}

/**
1283
 * pci_reenable_device - Resume abandoned device
1284 1285 1286 1287 1288
 * @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.
 */
1289
int pci_reenable_device(struct pci_dev *dev)
1290
{
1291
	if (pci_is_enabled(dev))
1292 1293 1294
		return do_pci_enable_device(dev, (1 << PCI_NUM_RESOURCES) - 1);
	return 0;
}
1295
EXPORT_SYMBOL(pci_reenable_device);
1296

1297 1298
static void pci_enable_bridge(struct pci_dev *dev)
{
1299
	struct pci_dev *bridge;
1300 1301
	int retval;

1302 1303 1304
	bridge = pci_upstream_bridge(dev);
	if (bridge)
		pci_enable_bridge(bridge);
1305

1306
	if (pci_is_enabled(dev)) {
1307
		if (!dev->is_busmaster)
1308
			pci_set_master(dev);
1309
		return;
1310 1311
	}

1312 1313 1314 1315 1316 1317 1318
	retval = pci_enable_device(dev);
	if (retval)
		dev_err(&dev->dev, "Error enabling bridge (%d), continuing\n",
			retval);
	pci_set_master(dev);
}

1319
static int pci_enable_device_flags(struct pci_dev *dev, unsigned long flags)
L
Linus Torvalds 已提交
1320
{
1321
	struct pci_dev *bridge;
L
Linus Torvalds 已提交
1322
	int err;
1323
	int i, bars = 0;
L
Linus Torvalds 已提交
1324

1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336
	/*
	 * 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);
	}

1337
	if (atomic_inc_return(&dev->enable_cnt) > 1)
1338 1339
		return 0;		/* already enabled */

1340 1341 1342
	bridge = pci_upstream_bridge(dev);
	if (bridge)
		pci_enable_bridge(bridge);
1343

1344 1345 1346 1347 1348
	/* only skip sriov related */
	for (i = 0; i <= PCI_ROM_RESOURCE; i++)
		if (dev->resource[i].flags & flags)
			bars |= (1 << i);
	for (i = PCI_BRIDGE_RESOURCES; i < DEVICE_COUNT_RESOURCE; i++)
1349 1350 1351
		if (dev->resource[i].flags & flags)
			bars |= (1 << i);

1352
	err = do_pci_enable_device(dev, bars);
1353
	if (err < 0)
1354
		atomic_dec(&dev->enable_cnt);
1355
	return err;
L
Linus Torvalds 已提交
1356 1357
}

1358 1359 1360 1361 1362 1363 1364 1365 1366 1367
/**
 * 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)
{
1368
	return pci_enable_device_flags(dev, IORESOURCE_IO);
1369
}
1370
EXPORT_SYMBOL(pci_enable_device_io);
1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381

/**
 * 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)
{
1382
	return pci_enable_device_flags(dev, IORESOURCE_MEM);
1383
}
1384
EXPORT_SYMBOL(pci_enable_device_mem);
1385

1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398
/**
 * 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)
{
1399
	return pci_enable_device_flags(dev, IORESOURCE_MEM | IORESOURCE_IO);
1400
}
1401
EXPORT_SYMBOL(pci_enable_device);
1402

T
Tejun Heo 已提交
1403 1404 1405 1406 1407 1408 1409
/*
 * 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 {
1410 1411
	unsigned int enabled:1;
	unsigned int pinned:1;
T
Tejun Heo 已提交
1412 1413 1414 1415 1416 1417 1418
	unsigned int orig_intx:1;
	unsigned int restore_intx:1;
	u32 region_mask;
};

static void pcim_release(struct device *gendev, void *res)
{
1419
	struct pci_dev *dev = to_pci_dev(gendev);
T
Tejun Heo 已提交
1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434
	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);

1435
	if (this->enabled && !this->pinned)
T
Tejun Heo 已提交
1436 1437 1438
		pci_disable_device(dev);
}

1439
static struct pci_devres *get_pci_dr(struct pci_dev *pdev)
T
Tejun Heo 已提交
1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452
{
	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);
}

1453
static struct pci_devres *find_pci_dr(struct pci_dev *pdev)
T
Tejun Heo 已提交
1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473
{
	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;
1474 1475
	if (dr->enabled)
		return 0;
T
Tejun Heo 已提交
1476 1477 1478 1479

	rc = pci_enable_device(pdev);
	if (!rc) {
		pdev->is_managed = 1;
1480
		dr->enabled = 1;
T
Tejun Heo 已提交
1481 1482 1483
	}
	return rc;
}
1484
EXPORT_SYMBOL(pcim_enable_device);
T
Tejun Heo 已提交
1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498

/**
 * 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);
1499
	WARN_ON(!dr || !dr->enabled);
T
Tejun Heo 已提交
1500
	if (dr)
1501
		dr->pinned = 1;
T
Tejun Heo 已提交
1502
}
1503
EXPORT_SYMBOL(pcim_pin_device);
T
Tejun Heo 已提交
1504

M
Matthew Garrett 已提交
1505 1506 1507 1508 1509 1510 1511 1512
/*
 * pcibios_add_device - provide arch specific hooks when adding device dev
 * @dev: the PCI device being added
 *
 * Permits the platform to provide architecture specific functionality when
 * devices are added. This is the default implementation. Architecture
 * implementations can override this.
 */
R
Ryan Desfosses 已提交
1513
int __weak pcibios_add_device(struct pci_dev *dev)
M
Matthew Garrett 已提交
1514 1515 1516 1517
{
	return 0;
}

1518 1519 1520 1521 1522 1523 1524 1525 1526 1527
/**
 * pcibios_release_device - provide arch specific hooks when releasing device dev
 * @dev: the PCI device being released
 *
 * Permits the platform to provide architecture specific functionality when
 * devices are released. This is the default implementation. Architecture
 * implementations can override this.
 */
void __weak pcibios_release_device(struct pci_dev *dev) {}

L
Linus Torvalds 已提交
1528 1529 1530 1531 1532 1533 1534 1535
/**
 * 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.
 */
B
Bogicevic Sasa 已提交
1536
void __weak pcibios_disable_device(struct pci_dev *dev) {}
L
Linus Torvalds 已提交
1537

1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548
/**
 * pcibios_penalize_isa_irq - penalize an ISA IRQ
 * @irq: ISA IRQ to penalize
 * @active: IRQ active or not
 *
 * Permits the platform to provide architecture-specific functionality when
 * penalizing ISA IRQs. This is the default implementation. Architecture
 * implementations can override this.
 */
void __weak pcibios_penalize_isa_irq(int irq, int active) {}

1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570
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)
{
1571
	if (pci_is_enabled(dev))
1572 1573 1574
		do_pci_disable_device(dev);
}

L
Linus Torvalds 已提交
1575 1576 1577 1578 1579 1580
/**
 * 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.
1581 1582
 *
 * Note we don't actually disable the device until all callers of
1583
 * pci_enable_device() have called pci_disable_device().
L
Linus Torvalds 已提交
1584
 */
R
Ryan Desfosses 已提交
1585
void pci_disable_device(struct pci_dev *dev)
L
Linus Torvalds 已提交
1586
{
T
Tejun Heo 已提交
1587
	struct pci_devres *dr;
1588

T
Tejun Heo 已提交
1589 1590
	dr = find_pci_dr(dev);
	if (dr)
1591
		dr->enabled = 0;
T
Tejun Heo 已提交
1592

1593 1594 1595
	dev_WARN_ONCE(&dev->dev, atomic_read(&dev->enable_cnt) <= 0,
		      "disabling already-disabled device");

1596
	if (atomic_dec_return(&dev->enable_cnt) != 0)
1597 1598
		return;

1599
	do_pci_disable_device(dev);
L
Linus Torvalds 已提交
1600

1601
	dev->is_busmaster = 0;
L
Linus Torvalds 已提交
1602
}
1603
EXPORT_SYMBOL(pci_disable_device);
L
Linus Torvalds 已提交
1604

B
Brian King 已提交
1605 1606
/**
 * pcibios_set_pcie_reset_state - set reset state for device dev
1607
 * @dev: the PCIe device reset
B
Brian King 已提交
1608 1609 1610
 * @state: Reset state to enter into
 *
 *
1611
 * Sets the PCIe reset state for the device. This is the default
B
Brian King 已提交
1612 1613
 * implementation. Architecture implementations can override this.
 */
B
Bjorn Helgaas 已提交
1614 1615
int __weak pcibios_set_pcie_reset_state(struct pci_dev *dev,
					enum pcie_reset_state state)
B
Brian King 已提交
1616 1617 1618 1619 1620 1621
{
	return -EINVAL;
}

/**
 * pci_set_pcie_reset_state - set reset state for device dev
1622
 * @dev: the PCIe device reset
B
Brian King 已提交
1623 1624 1625 1626 1627 1628 1629 1630 1631
 * @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);
}
1632
EXPORT_SYMBOL_GPL(pci_set_pcie_reset_state);
B
Brian King 已提交
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 1668
/**
 * 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;
}

1669 1670 1671
/**
 * pci_pme_wakeup - Wake up a PCI device if its PME Status bit is set.
 * @dev: Device to handle.
1672
 * @pme_poll_reset: Whether or not to reset the device's pme_poll flag.
1673 1674 1675 1676
 *
 * Check if @dev has generated PME and queue a resume request for it in that
 * case.
 */
1677
static int pci_pme_wakeup(struct pci_dev *dev, void *pme_poll_reset)
1678
{
1679 1680 1681
	if (pme_poll_reset && dev->pme_poll)
		dev->pme_poll = false;

1682 1683
	if (pci_check_pme_status(dev)) {
		pci_wakeup_event(dev);
1684
		pm_request_resume(&dev->dev);
1685
	}
1686 1687 1688 1689 1690 1691 1692 1693 1694 1695
	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)
1696
		pci_walk_bus(bus, pci_pme_wakeup, (void *)true);
1697 1698
}

1699

1700 1701 1702 1703 1704
/**
 * 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#.
 */
1705
bool pci_pme_capable(struct pci_dev *dev, pci_power_t state)
1706
{
1707
	if (!dev->pm_cap)
1708 1709
		return false;

1710
	return !!(dev->pme_support & (1 << state));
1711
}
1712
EXPORT_SYMBOL(pci_pme_capable);
1713

1714 1715
static void pci_pme_list_scan(struct work_struct *work)
{
1716
	struct pci_pme_device *pme_dev, *n;
1717 1718

	mutex_lock(&pci_pme_list_mutex);
1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734
	list_for_each_entry_safe(pme_dev, n, &pci_pme_list, list) {
		if (pme_dev->dev->pme_poll) {
			struct pci_dev *bridge;

			bridge = pme_dev->dev->bus->self;
			/*
			 * If bridge is in low power state, the
			 * configuration space of subordinate devices
			 * may be not accessible
			 */
			if (bridge && bridge->current_state != PCI_D0)
				continue;
			pci_pme_wakeup(pme_dev->dev, NULL);
		} else {
			list_del(&pme_dev->list);
			kfree(pme_dev);
1735
		}
1736
	}
1737 1738 1739
	if (!list_empty(&pci_pme_list))
		schedule_delayed_work(&pci_pme_work,
				      msecs_to_jiffies(PME_TIMEOUT));
1740 1741 1742
	mutex_unlock(&pci_pme_list_mutex);
}

1743
static void __pci_pme_active(struct pci_dev *dev, bool enable)
1744 1745 1746
{
	u16 pmcsr;

1747
	if (!dev->pme_support)
1748 1749
		return;

1750
	pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
1751 1752 1753 1754 1755
	/* 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;

1756
	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, pmcsr);
1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769
}

/**
 * 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'.
 */
void pci_pme_active(struct pci_dev *dev, bool enable)
{
	__pci_pme_active(dev, enable);
1770

1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789
	/*
	 * 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.
	 *
	 * Although PCIe uses in-band PME message instead of PME# line
	 * to report PME, PME does not work for some PCIe devices in
	 * reality.  For example, there are devices that set their PME
	 * status bits, but don't really bother to send a PME message;
	 * there are PCI Express Root Ports that don't bother to
	 * trigger interrupts when they receive PME messages from the
	 * devices below.  So PME poll is used for PCIe devices too.
	 */
1790

1791
	if (dev->pme_poll) {
1792 1793 1794 1795
		struct pci_pme_device *pme_dev;
		if (enable) {
			pme_dev = kmalloc(sizeof(struct pci_pme_device),
					  GFP_KERNEL);
1796 1797 1798 1799
			if (!pme_dev) {
				dev_warn(&dev->dev, "can't enable PME#\n");
				return;
			}
1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819
			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);
		}
	}

1820
	dev_dbg(&dev->dev, "PME# %s\n", enable ? "enabled" : "disabled");
1821
}
1822
EXPORT_SYMBOL(pci_pme_active);
1823

L
Linus Torvalds 已提交
1824
/**
1825
 * __pci_enable_wake - enable PCI device as wakeup event source
1826 1827
 * @dev: PCI device affected
 * @state: PCI state from which device will issue wakeup events
1828
 * @runtime: True if the events are to be generated at run time
1829 1830 1831 1832 1833 1834 1835
 * @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)
1836
 * always require such platform hooks.
1837
 *
1838 1839 1840 1841 1842
 * 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
1843
 */
1844 1845
int __pci_enable_wake(struct pci_dev *dev, pci_power_t state,
		      bool runtime, bool enable)
L
Linus Torvalds 已提交
1846
{
1847
	int ret = 0;
1848

1849
	if (enable && !runtime && !device_may_wakeup(&dev->dev))
1850
		return -EINVAL;
L
Linus Torvalds 已提交
1851

1852 1853 1854 1855
	/* Don't do the same thing twice in a row for one device. */
	if (!!enable == !!dev->wakeup_prepared)
		return 0;

1856 1857 1858 1859
	/*
	 * 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.
1860
	 */
L
Linus Torvalds 已提交
1861

1862 1863
	if (enable) {
		int error;
L
Linus Torvalds 已提交
1864

1865 1866 1867 1868
		if (pci_pme_capable(dev, state))
			pci_pme_active(dev, true);
		else
			ret = 1;
1869 1870
		error = runtime ? platform_pci_run_wake(dev, true) :
					platform_pci_sleep_wake(dev, true);
1871 1872
		if (ret)
			ret = error;
1873 1874
		if (!ret)
			dev->wakeup_prepared = true;
1875
	} else {
1876 1877 1878 1879
		if (runtime)
			platform_pci_run_wake(dev, false);
		else
			platform_pci_sleep_wake(dev, false);
1880
		pci_pme_active(dev, false);
1881
		dev->wakeup_prepared = false;
1882
	}
L
Linus Torvalds 已提交
1883

1884
	return ret;
1885
}
1886
EXPORT_SYMBOL(__pci_enable_wake);
L
Linus Torvalds 已提交
1887

1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907
/**
 * 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);
}
1908
EXPORT_SYMBOL(pci_wake_from_d3);
1909

1910
/**
J
Jesse Barnes 已提交
1911 1912 1913 1914 1915 1916
 * 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.
1917
 */
1918
static pci_power_t pci_target_state(struct pci_dev *dev)
1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939
{
	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;
		}
1940 1941
	} else if (!dev->pm_cap) {
		target_state = PCI_D0;
1942 1943 1944 1945 1946 1947
	} 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#.
		 */
1948 1949 1950 1951
		if (dev->pme_support) {
			while (target_state
			      && !(dev->pme_support & (1 << target_state)))
				target_state--;
1952 1953 1954
		}
	}

1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973
	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;

1974
	pci_enable_wake(dev, target_state, device_may_wakeup(&dev->dev));
1975

1976 1977 1978 1979 1980 1981 1982
	error = pci_set_power_state(dev, target_state);

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

	return error;
}
1983
EXPORT_SYMBOL(pci_prepare_to_sleep);
1984 1985

/**
R
Randy Dunlap 已提交
1986
 * pci_back_from_sleep - turn PCI device on during system-wide transition into working state
1987 1988
 * @dev: Device to handle.
 *
T
Thomas Weber 已提交
1989
 * Disable device's system wake-up capability and put it into D0.
1990 1991 1992 1993 1994 1995
 */
int pci_back_from_sleep(struct pci_dev *dev)
{
	pci_enable_wake(dev, PCI_D0, false);
	return pci_set_power_state(dev, PCI_D0);
}
1996
EXPORT_SYMBOL(pci_back_from_sleep);
1997

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
/**
 * 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;

2013 2014
	dev->runtime_d3cold = target_state == PCI_D3cold;

2015 2016 2017 2018
	__pci_enable_wake(dev, target_state, true, pci_dev_run_wake(dev));

	error = pci_set_power_state(dev, target_state);

2019
	if (error) {
2020
		__pci_enable_wake(dev, target_state, true, false);
2021 2022
		dev->runtime_d3cold = false;
	}
2023 2024 2025 2026

	return error;
}

2027 2028 2029 2030
/**
 * pci_dev_run_wake - Check if device can generate run-time wake-up events.
 * @dev: Device to check.
 *
2031
 * Return true if the device itself is capable of generating wake-up events
2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061
 * (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);

2062 2063 2064 2065 2066 2067 2068 2069
/**
 * pci_dev_keep_suspended - Check if the device can stay in the suspended state.
 * @pci_dev: Device to check.
 *
 * Return 'true' if the device is runtime-suspended, it doesn't have to be
 * reconfigured due to wakeup settings difference between system and runtime
 * suspend and the current power state of it is suitable for the upcoming
 * (system) transition.
2070 2071 2072
 *
 * If the device is not configured for system wakeup, disable PME for it before
 * returning 'true' to prevent it from waking up the system unnecessarily.
2073 2074 2075 2076 2077 2078
 */
bool pci_dev_keep_suspended(struct pci_dev *pci_dev)
{
	struct device *dev = &pci_dev->dev;

	if (!pm_runtime_suspended(dev)
2079
	    || pci_target_state(pci_dev) != pci_dev->current_state
2080 2081 2082
	    || platform_pci_need_resume(pci_dev))
		return false;

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 2117 2118 2119 2120 2121 2122 2123
	/*
	 * At this point the device is good to go unless it's been configured
	 * to generate PME at the runtime suspend time, but it is not supposed
	 * to wake up the system.  In that case, simply disable PME for it
	 * (it will have to be re-enabled on exit from system resume).
	 *
	 * If the device's power state is D3cold and the platform check above
	 * hasn't triggered, the device's configuration is suitable and we don't
	 * need to manipulate it at all.
	 */
	spin_lock_irq(&dev->power.lock);

	if (pm_runtime_suspended(dev) && pci_dev->current_state < PCI_D3cold &&
	    !device_may_wakeup(dev))
		__pci_pme_active(pci_dev, false);

	spin_unlock_irq(&dev->power.lock);
	return true;
}

/**
 * pci_dev_complete_resume - Finalize resume from system sleep for a device.
 * @pci_dev: Device to handle.
 *
 * If the device is runtime suspended and wakeup-capable, enable PME for it as
 * it might have been disabled during the prepare phase of system suspend if
 * the device was not configured for system wakeup.
 */
void pci_dev_complete_resume(struct pci_dev *pci_dev)
{
	struct device *dev = &pci_dev->dev;

	if (!pci_dev_run_wake(pci_dev))
		return;

	spin_lock_irq(&dev->power.lock);

	if (pm_runtime_suspended(dev) && pci_dev->current_state < PCI_D3cold)
		__pci_pme_active(pci_dev, true);

	spin_unlock_irq(&dev->power.lock);
2124 2125
}

2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157
void pci_config_pm_runtime_get(struct pci_dev *pdev)
{
	struct device *dev = &pdev->dev;
	struct device *parent = dev->parent;

	if (parent)
		pm_runtime_get_sync(parent);
	pm_runtime_get_noresume(dev);
	/*
	 * pdev->current_state is set to PCI_D3cold during suspending,
	 * so wait until suspending completes
	 */
	pm_runtime_barrier(dev);
	/*
	 * Only need to resume devices in D3cold, because config
	 * registers are still accessible for devices suspended but
	 * not in D3cold.
	 */
	if (pdev->current_state == PCI_D3cold)
		pm_runtime_resume(dev);
}

void pci_config_pm_runtime_put(struct pci_dev *pdev)
{
	struct device *dev = &pdev->dev;
	struct device *parent = dev->parent;

	pm_runtime_put(dev);
	if (parent)
		pm_runtime_put_sync(parent);
}

2158 2159 2160 2161 2162 2163 2164 2165
/**
 * 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
2166

2167
	pm_runtime_forbid(&dev->dev);
2168 2169
	pm_runtime_set_active(&dev->dev);
	pm_runtime_enable(&dev->dev);
2170
	device_enable_async_suspend(&dev->dev);
2171
	dev->wakeup_prepared = false;
2172

2173
	dev->pm_cap = 0;
2174
	dev->pme_support = 0;
2175

2176 2177 2178
	/* find PCI PM capability in list */
	pm = pci_find_capability(dev, PCI_CAP_ID_PM);
	if (!pm)
2179
		return;
2180 2181
	/* Check device's ability to generate PME# */
	pci_read_config_word(dev, pm + PCI_PM_PMC, &pmc);
2182

2183 2184 2185
	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);
2186
		return;
2187 2188
	}

2189
	dev->pm_cap = pm;
2190
	dev->d3_delay = PCI_PM_D3_WAIT;
2191
	dev->d3cold_delay = PCI_PM_D3COLD_WAIT;
2192
	dev->d3cold_allowed = true;
2193 2194 2195 2196

	dev->d1_support = false;
	dev->d2_support = false;
	if (!pci_no_d1d2(dev)) {
B
Bjorn Helgaas 已提交
2197
		if (pmc & PCI_PM_CAP_D1)
2198
			dev->d1_support = true;
B
Bjorn Helgaas 已提交
2199
		if (pmc & PCI_PM_CAP_D2)
2200
			dev->d2_support = true;
B
Bjorn Helgaas 已提交
2201 2202 2203

		if (dev->d1_support || dev->d2_support)
			dev_printk(KERN_DEBUG, &dev->dev, "supports%s%s\n",
2204 2205
				   dev->d1_support ? " D1" : "",
				   dev->d2_support ? " D2" : "");
2206 2207 2208 2209
	}

	pmc &= PCI_PM_CAP_PME_MASK;
	if (pmc) {
B
Bjorn Helgaas 已提交
2210 2211
		dev_printk(KERN_DEBUG, &dev->dev,
			 "PME# supported from%s%s%s%s%s\n",
B
Bjorn Helgaas 已提交
2212 2213 2214 2215 2216
			 (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" : "");
2217
		dev->pme_support = pmc >> PCI_PM_CAP_PME_SHIFT;
2218
		dev->pme_poll = true;
2219 2220 2221 2222 2223 2224
		/*
		 * 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 */
2225
		pci_pme_active(dev, false);
2226
	}
L
Linus Torvalds 已提交
2227 2228
}

2229 2230
static unsigned long pci_ea_flags(struct pci_dev *dev, u8 prop)
{
2231
	unsigned long flags = IORESOURCE_PCI_FIXED | IORESOURCE_PCI_EA_BEI;
2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256

	switch (prop) {
	case PCI_EA_P_MEM:
	case PCI_EA_P_VF_MEM:
		flags |= IORESOURCE_MEM;
		break;
	case PCI_EA_P_MEM_PREFETCH:
	case PCI_EA_P_VF_MEM_PREFETCH:
		flags |= IORESOURCE_MEM | IORESOURCE_PREFETCH;
		break;
	case PCI_EA_P_IO:
		flags |= IORESOURCE_IO;
		break;
	default:
		return 0;
	}

	return flags;
}

static struct resource *pci_ea_get_resource(struct pci_dev *dev, u8 bei,
					    u8 prop)
{
	if (bei <= PCI_EA_BEI_BAR5 && prop <= PCI_EA_P_IO)
		return &dev->resource[bei];
2257 2258 2259 2260 2261 2262
#ifdef CONFIG_PCI_IOV
	else if (bei >= PCI_EA_BEI_VF_BAR0 && bei <= PCI_EA_BEI_VF_BAR5 &&
		 (prop == PCI_EA_P_VF_MEM || prop == PCI_EA_P_VF_MEM_PREFETCH))
		return &dev->resource[PCI_IOV_RESOURCES +
				      bei - PCI_EA_BEI_VF_BAR0];
#endif
2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275
	else if (bei == PCI_EA_BEI_ROM)
		return &dev->resource[PCI_ROM_RESOURCE];
	else
		return NULL;
}

/* Read an Enhanced Allocation (EA) entry */
static int pci_ea_read(struct pci_dev *dev, int offset)
{
	struct resource *res;
	int ent_size, ent_offset = offset;
	resource_size_t start, end;
	unsigned long flags;
2276
	u32 dw0, bei, base, max_offset;
2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288
	u8 prop;
	bool support_64 = (sizeof(resource_size_t) >= 8);

	pci_read_config_dword(dev, ent_offset, &dw0);
	ent_offset += 4;

	/* Entry size field indicates DWORDs after 1st */
	ent_size = ((dw0 & PCI_EA_ES) + 1) << 2;

	if (!(dw0 & PCI_EA_ENABLE)) /* Entry not enabled */
		goto out;

2289 2290 2291
	bei = (dw0 & PCI_EA_BEI) >> 4;
	prop = (dw0 & PCI_EA_PP) >> 8;

2292 2293 2294 2295 2296
	/*
	 * If the Property is in the reserved range, try the Secondary
	 * Property instead.
	 */
	if (prop > PCI_EA_P_BRIDGE_IO && prop < PCI_EA_P_MEM_RESERVED)
2297
		prop = (dw0 & PCI_EA_SP) >> 16;
2298 2299 2300
	if (prop > PCI_EA_P_BRIDGE_IO)
		goto out;

2301
	res = pci_ea_get_resource(dev, bei, prop);
2302
	if (!res) {
2303
		dev_err(&dev->dev, "Unsupported EA entry BEI: %u\n", bei);
2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373
		goto out;
	}

	flags = pci_ea_flags(dev, prop);
	if (!flags) {
		dev_err(&dev->dev, "Unsupported EA properties: %#x\n", prop);
		goto out;
	}

	/* Read Base */
	pci_read_config_dword(dev, ent_offset, &base);
	start = (base & PCI_EA_FIELD_MASK);
	ent_offset += 4;

	/* Read MaxOffset */
	pci_read_config_dword(dev, ent_offset, &max_offset);
	ent_offset += 4;

	/* Read Base MSBs (if 64-bit entry) */
	if (base & PCI_EA_IS_64) {
		u32 base_upper;

		pci_read_config_dword(dev, ent_offset, &base_upper);
		ent_offset += 4;

		flags |= IORESOURCE_MEM_64;

		/* entry starts above 32-bit boundary, can't use */
		if (!support_64 && base_upper)
			goto out;

		if (support_64)
			start |= ((u64)base_upper << 32);
	}

	end = start + (max_offset | 0x03);

	/* Read MaxOffset MSBs (if 64-bit entry) */
	if (max_offset & PCI_EA_IS_64) {
		u32 max_offset_upper;

		pci_read_config_dword(dev, ent_offset, &max_offset_upper);
		ent_offset += 4;

		flags |= IORESOURCE_MEM_64;

		/* entry too big, can't use */
		if (!support_64 && max_offset_upper)
			goto out;

		if (support_64)
			end += ((u64)max_offset_upper << 32);
	}

	if (end < start) {
		dev_err(&dev->dev, "EA Entry crosses address boundary\n");
		goto out;
	}

	if (ent_size != ent_offset - offset) {
		dev_err(&dev->dev,
			"EA Entry Size (%d) does not match length read (%d)\n",
			ent_size, ent_offset - offset);
		goto out;
	}

	res->name = pci_name(dev);
	res->start = start;
	res->end = end;
	res->flags = flags;
2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387

	if (bei <= PCI_EA_BEI_BAR5)
		dev_printk(KERN_DEBUG, &dev->dev, "BAR %d: %pR (from Enhanced Allocation, properties %#02x)\n",
			   bei, res, prop);
	else if (bei == PCI_EA_BEI_ROM)
		dev_printk(KERN_DEBUG, &dev->dev, "ROM: %pR (from Enhanced Allocation, properties %#02x)\n",
			   res, prop);
	else if (bei >= PCI_EA_BEI_VF_BAR0 && bei <= PCI_EA_BEI_VF_BAR5)
		dev_printk(KERN_DEBUG, &dev->dev, "VF BAR %d: %pR (from Enhanced Allocation, properties %#02x)\n",
			   bei - PCI_EA_BEI_VF_BAR0, res, prop);
	else
		dev_printk(KERN_DEBUG, &dev->dev, "BEI %d res: %pR (from Enhanced Allocation, properties %#02x)\n",
			   bei, res, prop);

2388 2389 2390 2391
out:
	return offset + ent_size;
}

C
Colin Ian King 已提交
2392
/* Enhanced Allocation Initialization */
2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420
void pci_ea_init(struct pci_dev *dev)
{
	int ea;
	u8 num_ent;
	int offset;
	int i;

	/* find PCI EA capability in list */
	ea = pci_find_capability(dev, PCI_CAP_ID_EA);
	if (!ea)
		return;

	/* determine the number of entries */
	pci_bus_read_config_byte(dev->bus, dev->devfn, ea + PCI_EA_NUM_ENT,
					&num_ent);
	num_ent &= PCI_EA_NUM_ENT_MASK;

	offset = ea + PCI_EA_FIRST_ENT;

	/* Skip DWORD 2 for type 1 functions */
	if (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE)
		offset += 4;

	/* parse each EA entry */
	for (i = 0; i < num_ent; ++i)
		offset = pci_ea_read(dev, offset);
}

2421 2422 2423 2424 2425 2426
static void pci_add_saved_cap(struct pci_dev *pci_dev,
	struct pci_cap_saved_state *new_cap)
{
	hlist_add_head(&new_cap->next, &pci_dev->saved_cap_space);
}

2427
/**
2428 2429
 * _pci_add_cap_save_buffer - allocate buffer for saving given
 *                            capability registers
2430 2431
 * @dev: the PCI device
 * @cap: the capability to allocate the buffer for
2432
 * @extended: Standard or Extended capability ID
2433 2434
 * @size: requested size of the buffer
 */
2435 2436
static int _pci_add_cap_save_buffer(struct pci_dev *dev, u16 cap,
				    bool extended, unsigned int size)
2437 2438 2439 2440
{
	int pos;
	struct pci_cap_saved_state *save_state;

2441 2442 2443 2444 2445
	if (extended)
		pos = pci_find_ext_capability(dev, cap);
	else
		pos = pci_find_capability(dev, cap);

2446
	if (!pos)
2447 2448 2449 2450 2451 2452
		return 0;

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

2453
	save_state->cap.cap_nr = cap;
2454
	save_state->cap.cap_extended = extended;
2455
	save_state->cap.size = size;
2456 2457 2458 2459 2460
	pci_add_saved_cap(dev, save_state);

	return 0;
}

2461 2462 2463 2464 2465 2466 2467 2468 2469 2470
int pci_add_cap_save_buffer(struct pci_dev *dev, char cap, unsigned int size)
{
	return _pci_add_cap_save_buffer(dev, cap, false, size);
}

int pci_add_ext_cap_save_buffer(struct pci_dev *dev, u16 cap, unsigned int size)
{
	return _pci_add_cap_save_buffer(dev, cap, true, size);
}

2471 2472 2473 2474 2475 2476 2477 2478
/**
 * 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;

2479 2480
	error = pci_add_cap_save_buffer(dev, PCI_CAP_ID_EXP,
					PCI_EXP_SAVE_REGS * sizeof(u16));
2481 2482 2483 2484 2485 2486 2487 2488
	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");
2489 2490

	pci_allocate_vc_save_buffers(dev);
2491 2492
}

2493 2494 2495
void pci_free_cap_save_buffers(struct pci_dev *dev)
{
	struct pci_cap_saved_state *tmp;
2496
	struct hlist_node *n;
2497

2498
	hlist_for_each_entry_safe(tmp, n, &dev->saved_cap_space, next)
2499 2500 2501
		kfree(tmp);
}

Y
Yu Zhao 已提交
2502
/**
2503
 * pci_configure_ari - enable or disable ARI forwarding
Y
Yu Zhao 已提交
2504
 * @dev: the PCI device
2505 2506 2507
 *
 * If @dev and its upstream bridge both support ARI, enable ARI in the
 * bridge.  Otherwise, disable ARI in the bridge.
Y
Yu Zhao 已提交
2508
 */
2509
void pci_configure_ari(struct pci_dev *dev)
Y
Yu Zhao 已提交
2510 2511
{
	u32 cap;
2512
	struct pci_dev *bridge;
Y
Yu Zhao 已提交
2513

2514
	if (pcie_ari_disabled || !pci_is_pcie(dev) || dev->devfn)
Y
Yu Zhao 已提交
2515 2516
		return;

2517
	bridge = dev->bus->self;
2518
	if (!bridge)
2519 2520
		return;

2521
	pcie_capability_read_dword(bridge, PCI_EXP_DEVCAP2, &cap);
Y
Yu Zhao 已提交
2522 2523 2524
	if (!(cap & PCI_EXP_DEVCAP2_ARI))
		return;

2525 2526 2527 2528 2529 2530 2531 2532 2533
	if (pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ARI)) {
		pcie_capability_set_word(bridge, PCI_EXP_DEVCTL2,
					 PCI_EXP_DEVCTL2_ARI);
		bridge->ari_enabled = 1;
	} else {
		pcie_capability_clear_word(bridge, PCI_EXP_DEVCTL2,
					   PCI_EXP_DEVCTL2_ARI);
		bridge->ari_enabled = 0;
	}
Y
Yu Zhao 已提交
2534 2535
}

C
Chris Wright 已提交
2536 2537 2538 2539 2540 2541 2542 2543 2544 2545
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;
}

2546
/**
2547
 * pci_std_enable_acs - enable ACS on devices using standard ACS capabilites
2548 2549
 * @dev: the PCI device
 */
2550
static void pci_std_enable_acs(struct pci_dev *dev)
2551 2552 2553 2554 2555 2556 2557
{
	int pos;
	u16 cap;
	u16 ctrl;

	pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ACS);
	if (!pos)
2558
		return;
2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575

	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);
2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586
}

/**
 * pci_enable_acs - enable ACS if hardware support it
 * @dev: the PCI device
 */
void pci_enable_acs(struct pci_dev *dev)
{
	if (!pci_acs_enable)
		return;

2587
	if (!pci_dev_specific_enable_acs(dev))
2588 2589
		return;

2590
	pci_std_enable_acs(dev);
2591 2592
}

2593 2594 2595
static bool pci_acs_flags_enabled(struct pci_dev *pdev, u16 acs_flags)
{
	int pos;
2596
	u16 cap, ctrl;
2597 2598 2599 2600 2601

	pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_ACS);
	if (!pos)
		return false;

2602 2603 2604 2605 2606 2607 2608 2609
	/*
	 * Except for egress control, capabilities are either required
	 * or only required if controllable.  Features missing from the
	 * capability field can therefore be assumed as hard-wired enabled.
	 */
	pci_read_config_word(pdev, pos + PCI_ACS_CAP, &cap);
	acs_flags &= (cap | PCI_ACS_EC);

2610 2611 2612 2613
	pci_read_config_word(pdev, pos + PCI_ACS_CTRL, &ctrl);
	return (ctrl & acs_flags) == acs_flags;
}

2614 2615 2616 2617 2618 2619 2620
/**
 * pci_acs_enabled - test ACS against required flags for a given device
 * @pdev: device to test
 * @acs_flags: required PCI ACS flags
 *
 * Return true if the device supports the provided flags.  Automatically
 * filters out flags that are not implemented on multifunction devices.
2621 2622 2623 2624 2625 2626 2627 2628
 *
 * Note that this interface checks the effective ACS capabilities of the
 * device rather than the actual capabilities.  For instance, most single
 * function endpoints are not required to support ACS because they have no
 * opportunity for peer-to-peer access.  We therefore return 'true'
 * regardless of whether the device exposes an ACS capability.  This makes
 * it much easier for callers of this function to ignore the actual type
 * or topology of the device when testing ACS support.
2629 2630 2631
 */
bool pci_acs_enabled(struct pci_dev *pdev, u16 acs_flags)
{
2632
	int ret;
2633 2634 2635 2636 2637

	ret = pci_dev_specific_acs_enabled(pdev, acs_flags);
	if (ret >= 0)
		return ret > 0;

2638 2639 2640 2641 2642
	/*
	 * Conventional PCI and PCI-X devices never support ACS, either
	 * effectively or actually.  The shared bus topology implies that
	 * any device on the bus can receive or snoop DMA.
	 */
2643 2644 2645
	if (!pci_is_pcie(pdev))
		return false;

2646 2647 2648
	switch (pci_pcie_type(pdev)) {
	/*
	 * PCI/X-to-PCIe bridges are not specifically mentioned by the spec,
2649
	 * but since their primary interface is PCI/X, we conservatively
2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672
	 * handle them as we would a non-PCIe device.
	 */
	case PCI_EXP_TYPE_PCIE_BRIDGE:
	/*
	 * PCIe 3.0, 6.12.1 excludes ACS on these devices.  "ACS is never
	 * applicable... must never implement an ACS Extended Capability...".
	 * This seems arbitrary, but we take a conservative interpretation
	 * of this statement.
	 */
	case PCI_EXP_TYPE_PCI_BRIDGE:
	case PCI_EXP_TYPE_RC_EC:
		return false;
	/*
	 * PCIe 3.0, 6.12.1.1 specifies that downstream and root ports should
	 * implement ACS in order to indicate their peer-to-peer capabilities,
	 * regardless of whether they are single- or multi-function devices.
	 */
	case PCI_EXP_TYPE_DOWNSTREAM:
	case PCI_EXP_TYPE_ROOT_PORT:
		return pci_acs_flags_enabled(pdev, acs_flags);
	/*
	 * PCIe 3.0, 6.12.1.2 specifies ACS capabilities that should be
	 * implemented by the remaining PCIe types to indicate peer-to-peer
2673
	 * capabilities, but only when they are part of a multifunction
2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684
	 * device.  The footnote for section 6.12 indicates the specific
	 * PCIe types included here.
	 */
	case PCI_EXP_TYPE_ENDPOINT:
	case PCI_EXP_TYPE_UPSTREAM:
	case PCI_EXP_TYPE_LEG_END:
	case PCI_EXP_TYPE_RC_END:
		if (!pdev->multifunction)
			break;

		return pci_acs_flags_enabled(pdev, acs_flags);
2685 2686
	}

2687
	/*
2688
	 * PCIe 3.0, 6.12.1.3 specifies no ACS capabilities are applicable
2689 2690
	 * to single function devices with the exception of downstream ports.
	 */
2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722
	return true;
}

/**
 * pci_acs_path_enable - test ACS flags from start to end in a hierarchy
 * @start: starting downstream device
 * @end: ending upstream device or NULL to search to the root bus
 * @acs_flags: required flags
 *
 * Walk up a device tree from start to end testing PCI ACS support.  If
 * any step along the way does not support the required flags, return false.
 */
bool pci_acs_path_enabled(struct pci_dev *start,
			  struct pci_dev *end, u16 acs_flags)
{
	struct pci_dev *pdev, *parent = start;

	do {
		pdev = parent;

		if (!pci_acs_enabled(pdev, acs_flags))
			return false;

		if (pci_is_root_bus(pdev->bus))
			return (end == NULL);

		parent = pdev->bus->self;
	} while (pdev != end);

	return true;
}

2723 2724 2725
/**
 * pci_swizzle_interrupt_pin - swizzle INTx for device behind bridge
 * @dev: the PCI device
2726
 * @pin: the INTx pin (1=INTA, 2=INTB, 3=INTC, 4=INTD)
2727 2728 2729
 *
 * 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
2730 2731 2732
 * 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)
2733
 */
2734
u8 pci_swizzle_interrupt_pin(const struct pci_dev *dev, u8 pin)
2735
{
2736 2737 2738 2739 2740 2741 2742 2743
	int slot;

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

	return (((pin - 1) + slot) % 4) + 1;
2744 2745
}

R
Ryan Desfosses 已提交
2746
int pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge)
L
Linus Torvalds 已提交
2747 2748 2749
{
	u8 pin;

2750
	pin = dev->pin;
L
Linus Torvalds 已提交
2751 2752
	if (!pin)
		return -1;
2753

2754
	while (!pci_is_root_bus(dev->bus)) {
2755
		pin = pci_swizzle_interrupt_pin(dev, pin);
L
Linus Torvalds 已提交
2756 2757 2758 2759 2760 2761
		dev = dev->bus->self;
	}
	*bridge = dev;
	return pin;
}

2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773
/**
 * 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;

2774
	while (!pci_is_root_bus(dev->bus)) {
2775 2776 2777 2778 2779 2780
		pin = pci_swizzle_interrupt_pin(dev, pin);
		dev = dev->bus->self;
	}
	*pinp = pin;
	return PCI_SLOT(dev->devfn);
}
2781
EXPORT_SYMBOL_GPL(pci_common_swizzle);
2782

L
Linus Torvalds 已提交
2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793
/**
 *	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
2794 2795
	struct pci_devres *dr;

L
Linus Torvalds 已提交
2796 2797 2798 2799 2800 2801 2802 2803
	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
2804 2805 2806 2807

	dr = find_pci_dr(pdev);
	if (dr)
		dr->region_mask &= ~(1 << bar);
L
Linus Torvalds 已提交
2808
}
2809
EXPORT_SYMBOL(pci_release_region);
L
Linus Torvalds 已提交
2810 2811

/**
2812
 *	__pci_request_region - Reserved PCI I/O and memory resource
L
Linus Torvalds 已提交
2813 2814 2815
 *	@pdev: PCI device whose resources are to be reserved
 *	@bar: BAR to be reserved
 *	@res_name: Name to be associated with resource.
2816
 *	@exclusive: whether the region access is exclusive or not
L
Linus Torvalds 已提交
2817 2818 2819 2820 2821 2822
 *
 *	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.
 *
2823 2824
 *	If @exclusive is set, then the region is marked so that userspace
 *	is explicitly not allowed to map the resource via /dev/mem or
2825
 *	sysfs MMIO access.
2826
 *
L
Linus Torvalds 已提交
2827 2828 2829
 *	Returns 0 on success, or %EBUSY on error.  A warning
 *	message is also printed on failure.
 */
R
Ryan Desfosses 已提交
2830 2831
static int __pci_request_region(struct pci_dev *pdev, int bar,
				const char *res_name, int exclusive)
L
Linus Torvalds 已提交
2832
{
T
Tejun Heo 已提交
2833 2834
	struct pci_devres *dr;

L
Linus Torvalds 已提交
2835 2836
	if (pci_resource_len(pdev, bar) == 0)
		return 0;
2837

L
Linus Torvalds 已提交
2838 2839 2840 2841
	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;
R
Ryan Desfosses 已提交
2842
	} else if (pci_resource_flags(pdev, bar) & IORESOURCE_MEM) {
2843 2844 2845
		if (!__request_mem_region(pci_resource_start(pdev, bar),
					pci_resource_len(pdev, bar), res_name,
					exclusive))
L
Linus Torvalds 已提交
2846 2847
			goto err_out;
	}
T
Tejun Heo 已提交
2848 2849 2850 2851 2852

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

L
Linus Torvalds 已提交
2853 2854 2855
	return 0;

err_out:
2856
	dev_warn(&pdev->dev, "BAR %d: can't reserve %pR\n", bar,
2857
		 &pdev->resource[bar]);
L
Linus Torvalds 已提交
2858 2859 2860
	return -EBUSY;
}

2861
/**
2862
 *	pci_request_region - Reserve PCI I/O and memory resource
2863 2864
 *	@pdev: PCI device whose resources are to be reserved
 *	@bar: BAR to be reserved
2865
 *	@res_name: Name to be associated with resource
2866
 *
2867
 *	Mark the PCI region associated with PCI device @pdev BAR @bar as
2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878
 *	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);
}
2879
EXPORT_SYMBOL(pci_request_region);
2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896

/**
 *	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
2897
 *	sysfs.
2898
 */
R
Ryan Desfosses 已提交
2899 2900
int pci_request_region_exclusive(struct pci_dev *pdev, int bar,
				 const char *res_name)
2901 2902 2903
{
	return __pci_request_region(pdev, bar, res_name, IORESOURCE_EXCLUSIVE);
}
2904 2905
EXPORT_SYMBOL(pci_request_region_exclusive);

2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921
/**
 * 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);
}
2922
EXPORT_SYMBOL(pci_release_selected_regions);
2923

2924
static int __pci_request_selected_regions(struct pci_dev *pdev, int bars,
R
Ryan Desfosses 已提交
2925
					  const char *res_name, int excl)
2926 2927 2928 2929 2930
{
	int i;

	for (i = 0; i < 6; i++)
		if (bars & (1 << i))
2931
			if (__pci_request_region(pdev, i, res_name, excl))
2932 2933 2934 2935
				goto err_out;
	return 0;

err_out:
R
Ryan Desfosses 已提交
2936
	while (--i >= 0)
2937 2938 2939 2940 2941
		if (bars & (1 << i))
			pci_release_region(pdev, i);

	return -EBUSY;
}
L
Linus Torvalds 已提交
2942

2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954

/**
 * 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);
}
2955
EXPORT_SYMBOL(pci_request_selected_regions);
2956

R
Ryan Desfosses 已提交
2957 2958
int pci_request_selected_regions_exclusive(struct pci_dev *pdev, int bars,
					   const char *res_name)
2959 2960 2961 2962
{
	return __pci_request_selected_regions(pdev, bars, res_name,
			IORESOURCE_EXCLUSIVE);
}
2963
EXPORT_SYMBOL(pci_request_selected_regions_exclusive);
2964

L
Linus Torvalds 已提交
2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975
/**
 *	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)
{
2976
	pci_release_selected_regions(pdev, (1 << 6) - 1);
L
Linus Torvalds 已提交
2977
}
2978
EXPORT_SYMBOL(pci_release_regions);
L
Linus Torvalds 已提交
2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992

/**
 *	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.
 */
2993
int pci_request_regions(struct pci_dev *pdev, const char *res_name)
L
Linus Torvalds 已提交
2994
{
2995
	return pci_request_selected_regions(pdev, ((1 << 6) - 1), res_name);
L
Linus Torvalds 已提交
2996
}
2997
EXPORT_SYMBOL(pci_request_regions);
L
Linus Torvalds 已提交
2998

2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009
/**
 *	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
3010
 *	/dev/mem and the sysfs MMIO access will not be allowed.
3011 3012 3013 3014 3015 3016 3017 3018 3019
 *
 *	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);
}
3020
EXPORT_SYMBOL(pci_request_regions_exclusive);
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 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136
#ifdef PCI_IOBASE
struct io_range {
	struct list_head list;
	phys_addr_t start;
	resource_size_t size;
};

static LIST_HEAD(io_range_list);
static DEFINE_SPINLOCK(io_range_lock);
#endif

/*
 * Record the PCI IO range (expressed as CPU physical address + size).
 * Return a negative value if an error has occured, zero otherwise
 */
int __weak pci_register_io_range(phys_addr_t addr, resource_size_t size)
{
	int err = 0;

#ifdef PCI_IOBASE
	struct io_range *range;
	resource_size_t allocated_size = 0;

	/* check if the range hasn't been previously recorded */
	spin_lock(&io_range_lock);
	list_for_each_entry(range, &io_range_list, list) {
		if (addr >= range->start && addr + size <= range->start + size) {
			/* range already registered, bail out */
			goto end_register;
		}
		allocated_size += range->size;
	}

	/* range not registed yet, check for available space */
	if (allocated_size + size - 1 > IO_SPACE_LIMIT) {
		/* if it's too big check if 64K space can be reserved */
		if (allocated_size + SZ_64K - 1 > IO_SPACE_LIMIT) {
			err = -E2BIG;
			goto end_register;
		}

		size = SZ_64K;
		pr_warn("Requested IO range too big, new size set to 64K\n");
	}

	/* add the range to the list */
	range = kzalloc(sizeof(*range), GFP_ATOMIC);
	if (!range) {
		err = -ENOMEM;
		goto end_register;
	}

	range->start = addr;
	range->size = size;

	list_add_tail(&range->list, &io_range_list);

end_register:
	spin_unlock(&io_range_lock);
#endif

	return err;
}

phys_addr_t pci_pio_to_address(unsigned long pio)
{
	phys_addr_t address = (phys_addr_t)OF_BAD_ADDR;

#ifdef PCI_IOBASE
	struct io_range *range;
	resource_size_t allocated_size = 0;

	if (pio > IO_SPACE_LIMIT)
		return address;

	spin_lock(&io_range_lock);
	list_for_each_entry(range, &io_range_list, list) {
		if (pio >= allocated_size && pio < allocated_size + range->size) {
			address = range->start + pio - allocated_size;
			break;
		}
		allocated_size += range->size;
	}
	spin_unlock(&io_range_lock);
#endif

	return address;
}

unsigned long __weak pci_address_to_pio(phys_addr_t address)
{
#ifdef PCI_IOBASE
	struct io_range *res;
	resource_size_t offset = 0;
	unsigned long addr = -1;

	spin_lock(&io_range_lock);
	list_for_each_entry(res, &io_range_list, list) {
		if (address >= res->start && address < res->start + res->size) {
			addr = address - res->start + offset;
			break;
		}
		offset += res->size;
	}
	spin_unlock(&io_range_lock);

	return addr;
#else
	if (address > IO_SPACE_LIMIT)
		return (unsigned long)-1;

	return (unsigned long) address;
#endif
}

3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167
/**
 *	pci_remap_iospace - Remap the memory mapped I/O space
 *	@res: Resource describing the I/O space
 *	@phys_addr: physical address of range to be mapped
 *
 *	Remap the memory mapped I/O space described by the @res
 *	and the CPU physical address @phys_addr into virtual address space.
 *	Only architectures that have memory mapped IO functions defined
 *	(and the PCI_IOBASE value defined) should call this function.
 */
int __weak pci_remap_iospace(const struct resource *res, phys_addr_t phys_addr)
{
#if defined(PCI_IOBASE) && defined(CONFIG_MMU)
	unsigned long vaddr = (unsigned long)PCI_IOBASE + res->start;

	if (!(res->flags & IORESOURCE_IO))
		return -EINVAL;

	if (res->end > IO_SPACE_LIMIT)
		return -EINVAL;

	return ioremap_page_range(vaddr, vaddr + resource_size(res), phys_addr,
				  pgprot_device(PAGE_KERNEL));
#else
	/* this architecture does not have memory mapped I/O space,
	   so this function should never be called */
	WARN_ONCE(1, "This architecture does not support memory mapped I/O\n");
	return -ENODEV;
#endif
}

3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183
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;
}
3184

3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196
/**
 * pcibios_setup - process "pci=" kernel boot arguments
 * @str: string used to pass in "pci=" kernel boot arguments
 *
 * Process kernel boot arguments.  This is the default implementation.
 * Architecture specific implementations can override this as necessary.
 */
char * __weak __init pcibios_setup(char *str)
{
	return str;
}

3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208
/**
 * pcibios_set_master - enable PCI bus-mastering for device dev
 * @dev: the PCI device to enable
 *
 * Enables PCI bus-mastering for the device.  This is the default
 * implementation.  Architecture specific implementations can override
 * this if necessary.
 */
void __weak pcibios_set_master(struct pci_dev *dev)
{
	u8 lat;

3209 3210 3211 3212
	/* The latency timer doesn't apply to PCIe (either Type 0 or Type 1) */
	if (pci_is_pcie(dev))
		return;

3213 3214 3215 3216 3217 3218 3219
	pci_read_config_byte(dev, PCI_LATENCY_TIMER, &lat);
	if (lat < 16)
		lat = (64 <= pcibios_max_latency) ? 64 : pcibios_max_latency;
	else if (lat > pcibios_max_latency)
		lat = pcibios_max_latency;
	else
		return;
3220

3221 3222 3223
	pci_write_config_byte(dev, PCI_LATENCY_TIMER, lat);
}

L
Linus Torvalds 已提交
3224 3225 3226 3227 3228 3229 3230
/**
 * 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.
 */
3231
void pci_set_master(struct pci_dev *dev)
L
Linus Torvalds 已提交
3232
{
3233
	__pci_set_master(dev, true);
L
Linus Torvalds 已提交
3234 3235
	pcibios_set_master(dev);
}
3236
EXPORT_SYMBOL(pci_set_master);
L
Linus Torvalds 已提交
3237

3238 3239 3240 3241 3242 3243 3244 3245
/**
 * 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);
}
3246
EXPORT_SYMBOL(pci_clear_master);
3247

L
Linus Torvalds 已提交
3248
/**
3249 3250
 * 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
3251
 *
3252 3253
 * Helper function for pci_set_mwi.
 * Originally copied from drivers/net/acenic.c.
L
Linus Torvalds 已提交
3254 3255 3256 3257
 * 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
3258
int pci_set_cacheline_size(struct pci_dev *dev)
L
Linus Torvalds 已提交
3259 3260 3261 3262
{
	u8 cacheline_size;

	if (!pci_cache_line_size)
T
Tejun Heo 已提交
3263
		return -EINVAL;
L
Linus Torvalds 已提交
3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278

	/* 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;

3279 3280
	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
3281 3282 3283

	return -EINVAL;
}
T
Tejun Heo 已提交
3284 3285
EXPORT_SYMBOL_GPL(pci_set_cacheline_size);

L
Linus Torvalds 已提交
3286 3287 3288 3289
/**
 * pci_set_mwi - enables memory-write-invalidate PCI transaction
 * @dev: the PCI device for which MWI is enabled
 *
R
Randy Dunlap 已提交
3290
 * Enables the Memory-Write-Invalidate transaction in %PCI_COMMAND.
L
Linus Torvalds 已提交
3291 3292 3293
 *
 * RETURNS: An appropriate -ERRNO error value on error, or zero for success.
 */
R
Ryan Desfosses 已提交
3294
int pci_set_mwi(struct pci_dev *dev)
L
Linus Torvalds 已提交
3295
{
3296 3297 3298
#ifdef PCI_DISABLE_MWI
	return 0;
#else
L
Linus Torvalds 已提交
3299 3300 3301
	int rc;
	u16 cmd;

3302
	rc = pci_set_cacheline_size(dev);
L
Linus Torvalds 已提交
3303 3304 3305 3306
	if (rc)
		return rc;

	pci_read_config_word(dev, PCI_COMMAND, &cmd);
R
Ryan Desfosses 已提交
3307
	if (!(cmd & PCI_COMMAND_INVALIDATE)) {
3308
		dev_dbg(&dev->dev, "enabling Mem-Wr-Inval\n");
L
Linus Torvalds 已提交
3309 3310 3311 3312
		cmd |= PCI_COMMAND_INVALIDATE;
		pci_write_config_word(dev, PCI_COMMAND, cmd);
	}
	return 0;
3313
#endif
L
Linus Torvalds 已提交
3314
}
3315
EXPORT_SYMBOL(pci_set_mwi);
L
Linus Torvalds 已提交
3316

R
Randy Dunlap 已提交
3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327
/**
 * 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)
{
3328 3329 3330 3331 3332
#ifdef PCI_DISABLE_MWI
	return 0;
#else
	return pci_set_mwi(dev);
#endif
R
Randy Dunlap 已提交
3333
}
3334
EXPORT_SYMBOL(pci_try_set_mwi);
R
Randy Dunlap 已提交
3335

L
Linus Torvalds 已提交
3336 3337 3338 3339 3340 3341
/**
 * 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
 */
R
Ryan Desfosses 已提交
3342
void pci_clear_mwi(struct pci_dev *dev)
L
Linus Torvalds 已提交
3343
{
3344
#ifndef PCI_DISABLE_MWI
L
Linus Torvalds 已提交
3345 3346 3347 3348 3349 3350 3351
	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);
	}
3352
#endif
L
Linus Torvalds 已提交
3353
}
3354
EXPORT_SYMBOL(pci_clear_mwi);
L
Linus Torvalds 已提交
3355

B
Brett M Russ 已提交
3356 3357
/**
 * pci_intx - enables/disables PCI INTx for device dev
R
Randy Dunlap 已提交
3358 3359
 * @pdev: the PCI device to operate on
 * @enable: boolean: whether to enable or disable PCI INTx
B
Brett M Russ 已提交
3360 3361 3362
 *
 * Enables/disables PCI INTx for device dev
 */
R
Ryan Desfosses 已提交
3363
void pci_intx(struct pci_dev *pdev, int enable)
B
Brett M Russ 已提交
3364 3365 3366 3367 3368
{
	u16 pci_command, new;

	pci_read_config_word(pdev, PCI_COMMAND, &pci_command);

R
Ryan Desfosses 已提交
3369
	if (enable)
B
Brett M Russ 已提交
3370
		new = pci_command & ~PCI_COMMAND_INTX_DISABLE;
R
Ryan Desfosses 已提交
3371
	else
B
Brett M Russ 已提交
3372 3373 3374
		new = pci_command | PCI_COMMAND_INTX_DISABLE;

	if (new != pci_command) {
T
Tejun Heo 已提交
3375 3376
		struct pci_devres *dr;

3377
		pci_write_config_word(pdev, PCI_COMMAND, new);
T
Tejun Heo 已提交
3378 3379 3380 3381 3382 3383

		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
3384 3385
	}
}
3386
EXPORT_SYMBOL_GPL(pci_intx);
B
Brett M Russ 已提交
3387

3388 3389
/**
 * pci_intx_mask_supported - probe for INTx masking support
3390
 * @dev: the PCI device to operate on
3391 3392 3393 3394 3395 3396 3397 3398 3399
 *
 * Check if the device dev support INTx masking via the config space
 * command word.
 */
bool pci_intx_mask_supported(struct pci_dev *dev)
{
	bool mask_supported = false;
	u16 orig, new;

3400 3401 3402
	if (dev->broken_intx_masking)
		return false;

3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415
	pci_cfg_access_lock(dev);

	pci_read_config_word(dev, PCI_COMMAND, &orig);
	pci_write_config_word(dev, PCI_COMMAND,
			      orig ^ PCI_COMMAND_INTX_DISABLE);
	pci_read_config_word(dev, PCI_COMMAND, &new);

	/*
	 * There's no way to protect against hardware bugs or detect them
	 * reliably, but as long as we know what the value should be, let's
	 * go ahead and check it.
	 */
	if ((new ^ orig) & ~PCI_COMMAND_INTX_DISABLE) {
3416 3417
		dev_err(&dev->dev, "Command register changed from 0x%x to 0x%x: driver or hardware bug?\n",
			orig, new);
3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474
	} else if ((new ^ orig) & PCI_COMMAND_INTX_DISABLE) {
		mask_supported = true;
		pci_write_config_word(dev, PCI_COMMAND, orig);
	}

	pci_cfg_access_unlock(dev);
	return mask_supported;
}
EXPORT_SYMBOL_GPL(pci_intx_mask_supported);

static bool pci_check_and_set_intx_mask(struct pci_dev *dev, bool mask)
{
	struct pci_bus *bus = dev->bus;
	bool mask_updated = true;
	u32 cmd_status_dword;
	u16 origcmd, newcmd;
	unsigned long flags;
	bool irq_pending;

	/*
	 * We do a single dword read to retrieve both command and status.
	 * Document assumptions that make this possible.
	 */
	BUILD_BUG_ON(PCI_COMMAND % 4);
	BUILD_BUG_ON(PCI_COMMAND + 2 != PCI_STATUS);

	raw_spin_lock_irqsave(&pci_lock, flags);

	bus->ops->read(bus, dev->devfn, PCI_COMMAND, 4, &cmd_status_dword);

	irq_pending = (cmd_status_dword >> 16) & PCI_STATUS_INTERRUPT;

	/*
	 * Check interrupt status register to see whether our device
	 * triggered the interrupt (when masking) or the next IRQ is
	 * already pending (when unmasking).
	 */
	if (mask != irq_pending) {
		mask_updated = false;
		goto done;
	}

	origcmd = cmd_status_dword;
	newcmd = origcmd & ~PCI_COMMAND_INTX_DISABLE;
	if (mask)
		newcmd |= PCI_COMMAND_INTX_DISABLE;
	if (newcmd != origcmd)
		bus->ops->write(bus, dev->devfn, PCI_COMMAND, 2, newcmd);

done:
	raw_spin_unlock_irqrestore(&pci_lock, flags);

	return mask_updated;
}

/**
 * pci_check_and_mask_intx - mask INTx on pending interrupt
3475
 * @dev: the PCI device to operate on
3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487
 *
 * Check if the device dev has its INTx line asserted, mask it and
 * return true in that case. False is returned if not interrupt was
 * pending.
 */
bool pci_check_and_mask_intx(struct pci_dev *dev)
{
	return pci_check_and_set_intx_mask(dev, true);
}
EXPORT_SYMBOL_GPL(pci_check_and_mask_intx);

/**
3488
 * pci_check_and_unmask_intx - unmask INTx if no interrupt is pending
3489
 * @dev: the PCI device to operate on
3490 3491 3492 3493 3494 3495 3496 3497 3498 3499 3500
 *
 * Check if the device dev has its INTx line asserted, unmask it if not
 * and return true. False is returned and the mask remains active if
 * there was still an interrupt pending.
 */
bool pci_check_and_unmask_intx(struct pci_dev *dev)
{
	return pci_check_and_set_intx_mask(dev, false);
}
EXPORT_SYMBOL_GPL(pci_check_and_unmask_intx);

3501 3502 3503 3504 3505 3506 3507
/**
 * pci_wait_for_pending_transaction - waits for pending transaction
 * @dev: the PCI device to operate on
 *
 * Return 0 if transaction is pending 1 otherwise.
 */
int pci_wait_for_pending_transaction(struct pci_dev *dev)
3508
{
3509 3510
	if (!pci_is_pcie(dev))
		return 1;
Y
Yu Zhao 已提交
3511

3512 3513
	return pci_wait_for_pending(dev, pci_pcie_cap(dev) + PCI_EXP_DEVSTA,
				    PCI_EXP_DEVSTA_TRPND);
3514 3515 3516
}
EXPORT_SYMBOL(pci_wait_for_pending_transaction);

3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528 3529 3530 3531 3532 3533 3534 3535 3536 3537 3538 3539
/*
 * We should only need to wait 100ms after FLR, but some devices take longer.
 * Wait for up to 1000ms for config space to return something other than -1.
 * Intel IGD requires this when an LCD panel is attached.  We read the 2nd
 * dword because VFs don't implement the 1st dword.
 */
static void pci_flr_wait(struct pci_dev *dev)
{
	int i = 0;
	u32 id;

	do {
		msleep(100);
		pci_read_config_dword(dev, PCI_COMMAND, &id);
	} while (i++ < 10 && id == ~0);

	if (id == ~0)
		dev_warn(&dev->dev, "Failed to return from FLR\n");
	else if (i > 1)
		dev_info(&dev->dev, "Required additional %dms to return from FLR\n",
			 (i - 1) * 100);
}

3540 3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551
static int pcie_flr(struct pci_dev *dev, int probe)
{
	u32 cap;

	pcie_capability_read_dword(dev, PCI_EXP_DEVCAP, &cap);
	if (!(cap & PCI_EXP_DEVCAP_FLR))
		return -ENOTTY;

	if (probe)
		return 0;

	if (!pci_wait_for_pending_transaction(dev))
3552
		dev_err(&dev->dev, "timed out waiting for pending transaction; performing function level reset anyway\n");
Y
Yu Zhao 已提交
3553

3554
	pcie_capability_set_word(dev, PCI_EXP_DEVCTL, PCI_EXP_DEVCTL_BCR_FLR);
3555
	pci_flr_wait(dev);
3556 3557
	return 0;
}
S
Sheng Yang 已提交
3558

Y
Yu Zhao 已提交
3559
static int pci_af_flr(struct pci_dev *dev, int probe)
3560
{
Y
Yu Zhao 已提交
3561
	int pos;
3562 3563
	u8 cap;

Y
Yu Zhao 已提交
3564 3565
	pos = pci_find_capability(dev, PCI_CAP_ID_AF);
	if (!pos)
3566
		return -ENOTTY;
Y
Yu Zhao 已提交
3567 3568

	pci_read_config_byte(dev, pos + PCI_AF_CAP, &cap);
3569 3570 3571 3572 3573 3574
	if (!(cap & PCI_AF_CAP_TP) || !(cap & PCI_AF_CAP_FLR))
		return -ENOTTY;

	if (probe)
		return 0;

3575 3576 3577 3578 3579
	/*
	 * Wait for Transaction Pending bit to clear.  A word-aligned test
	 * is used, so we use the conrol offset rather than status and shift
	 * the test bit to match.
	 */
3580
	if (!pci_wait_for_pending(dev, pos + PCI_AF_CTRL,
3581
				 PCI_AF_STATUS_TP << 8))
3582
		dev_err(&dev->dev, "timed out waiting for pending transaction; performing AF function level reset anyway\n");
S
Sheng Yang 已提交
3583

Y
Yu Zhao 已提交
3584
	pci_write_config_byte(dev, pos + PCI_AF_CTRL, PCI_AF_CTRL_FLR);
3585
	pci_flr_wait(dev);
3586 3587 3588
	return 0;
}

3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600
/**
 * 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
3601
 * by default (i.e. unless the @dev's d3_delay field has a different value).
3602 3603
 * Moreover, only devices in D0 can be reset by this function.
 */
3604
static int pci_pm_reset(struct pci_dev *dev, int probe)
S
Sheng Yang 已提交
3605
{
3606 3607
	u16 csr;

3608
	if (!dev->pm_cap || dev->dev_flags & PCI_DEV_FLAGS_NO_PM_RESET)
3609
		return -ENOTTY;
S
Sheng Yang 已提交
3610

3611 3612 3613
	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
3614

3615 3616
	if (probe)
		return 0;
3617

3618 3619 3620 3621 3622 3623
	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);
3624
	pci_dev_d3_sleep(dev);
3625 3626 3627 3628

	csr &= ~PCI_PM_CTRL_STATE_MASK;
	csr |= PCI_D0;
	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, csr);
3629
	pci_dev_d3_sleep(dev);
3630 3631 3632 3633

	return 0;
}

3634
void pci_reset_secondary_bus(struct pci_dev *dev)
Y
Yu Zhao 已提交
3635 3636
{
	u16 ctrl;
3637 3638 3639 3640

	pci_read_config_word(dev, PCI_BRIDGE_CONTROL, &ctrl);
	ctrl |= PCI_BRIDGE_CTL_BUS_RESET;
	pci_write_config_word(dev, PCI_BRIDGE_CONTROL, ctrl);
3641 3642
	/*
	 * PCI spec v3.0 7.6.4.2 requires minimum Trst of 1ms.  Double
3643
	 * this to 2ms to ensure that we meet the minimum requirement.
3644 3645
	 */
	msleep(2);
3646 3647 3648

	ctrl &= ~PCI_BRIDGE_CTL_BUS_RESET;
	pci_write_config_word(dev, PCI_BRIDGE_CONTROL, ctrl);
3649 3650 3651 3652 3653 3654 3655 3656 3657

	/*
	 * Trhfa for conventional PCI is 2^25 clock cycles.
	 * Assuming a minimum 33MHz clock this results in a 1s
	 * delay before we can consider subordinate devices to
	 * be re-initialized.  PCIe has some ways to shorten this,
	 * but we don't make use of them yet.
	 */
	ssleep(1);
3658
}
3659

3660 3661 3662 3663 3664
void __weak pcibios_reset_secondary_bus(struct pci_dev *dev)
{
	pci_reset_secondary_bus(dev);
}

3665 3666 3667 3668 3669 3670 3671 3672 3673 3674 3675
/**
 * pci_reset_bridge_secondary_bus - Reset the secondary bus on a PCI bridge.
 * @dev: Bridge device
 *
 * Use the bridge control register to assert reset on the secondary bus.
 * Devices on the secondary bus are left in power-on state.
 */
void pci_reset_bridge_secondary_bus(struct pci_dev *dev)
{
	pcibios_reset_secondary_bus(dev);
}
3676 3677 3678 3679
EXPORT_SYMBOL_GPL(pci_reset_bridge_secondary_bus);

static int pci_parent_bus_reset(struct pci_dev *dev, int probe)
{
Y
Yu Zhao 已提交
3680 3681
	struct pci_dev *pdev;

3682 3683
	if (pci_is_root_bus(dev->bus) || dev->subordinate ||
	    !dev->bus->self || dev->dev_flags & PCI_DEV_FLAGS_NO_BUS_RESET)
Y
Yu Zhao 已提交
3684 3685 3686 3687 3688 3689 3690 3691 3692
		return -ENOTTY;

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

	if (probe)
		return 0;

3693
	pci_reset_bridge_secondary_bus(dev->bus->self);
Y
Yu Zhao 已提交
3694 3695 3696 3697

	return 0;
}

3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711 3712 3713 3714 3715 3716
static int pci_reset_hotplug_slot(struct hotplug_slot *hotplug, int probe)
{
	int rc = -ENOTTY;

	if (!hotplug || !try_module_get(hotplug->ops->owner))
		return rc;

	if (hotplug->ops->reset_slot)
		rc = hotplug->ops->reset_slot(hotplug, probe);

	module_put(hotplug->ops->owner);

	return rc;
}

static int pci_dev_reset_slot_function(struct pci_dev *dev, int probe)
{
	struct pci_dev *pdev;

3717 3718
	if (dev->subordinate || !dev->slot ||
	    dev->dev_flags & PCI_DEV_FLAGS_NO_BUS_RESET)
3719 3720 3721 3722 3723 3724 3725 3726 3727
		return -ENOTTY;

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

	return pci_reset_hotplug_slot(dev->slot->hotplug, probe);
}

3728
static int __pci_dev_reset(struct pci_dev *dev, int probe)
S
Sheng Yang 已提交
3729
{
Y
Yu Zhao 已提交
3730 3731 3732 3733
	int rc;

	might_sleep();

3734 3735 3736 3737
	rc = pci_dev_specific_reset(dev, probe);
	if (rc != -ENOTTY)
		goto done;

Y
Yu Zhao 已提交
3738 3739 3740
	rc = pcie_flr(dev, probe);
	if (rc != -ENOTTY)
		goto done;
S
Sheng Yang 已提交
3741

Y
Yu Zhao 已提交
3742
	rc = pci_af_flr(dev, probe);
3743 3744 3745 3746
	if (rc != -ENOTTY)
		goto done;

	rc = pci_pm_reset(dev, probe);
Y
Yu Zhao 已提交
3747 3748 3749
	if (rc != -ENOTTY)
		goto done;

3750 3751 3752 3753
	rc = pci_dev_reset_slot_function(dev, probe);
	if (rc != -ENOTTY)
		goto done;

Y
Yu Zhao 已提交
3754
	rc = pci_parent_bus_reset(dev, probe);
Y
Yu Zhao 已提交
3755
done:
3756 3757 3758
	return rc;
}

3759 3760 3761 3762 3763 3764 3765
static void pci_dev_lock(struct pci_dev *dev)
{
	pci_cfg_access_lock(dev);
	/* block PM suspend, driver probe, etc. */
	device_lock(&dev->dev);
}

3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777
/* Return 1 on successful lock, 0 on contention */
static int pci_dev_trylock(struct pci_dev *dev)
{
	if (pci_cfg_access_trylock(dev)) {
		if (device_trylock(&dev->dev))
			return 1;
		pci_cfg_access_unlock(dev);
	}

	return 0;
}

3778 3779 3780 3781 3782 3783
static void pci_dev_unlock(struct pci_dev *dev)
{
	device_unlock(&dev->dev);
	pci_cfg_access_unlock(dev);
}

3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800
/**
 * pci_reset_notify - notify device driver of reset
 * @dev: device to be notified of reset
 * @prepare: 'true' if device is about to be reset; 'false' if reset attempt
 *           completed
 *
 * Must be called prior to device access being disabled and after device
 * access is restored.
 */
static void pci_reset_notify(struct pci_dev *dev, bool prepare)
{
	const struct pci_error_handlers *err_handler =
			dev->driver ? dev->driver->err_handler : NULL;
	if (err_handler && err_handler->reset_notify)
		err_handler->reset_notify(dev, prepare);
}

3801 3802
static void pci_dev_save_and_disable(struct pci_dev *dev)
{
3803 3804
	pci_reset_notify(dev, true);

3805 3806 3807 3808 3809 3810 3811
	/*
	 * Wake-up device prior to save.  PM registers default to D0 after
	 * reset and a simple register restore doesn't reliably return
	 * to a non-D0 state anyway.
	 */
	pci_set_power_state(dev, PCI_D0);

3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825
	pci_save_state(dev);
	/*
	 * Disable the device by clearing the Command register, except for
	 * INTx-disable which is set.  This not only disables MMIO and I/O port
	 * BARs, but also prevents the device from being Bus Master, preventing
	 * DMA from the device including MSI/MSI-X interrupts.  For PCI 2.3
	 * compliant devices, INTx-disable prevents legacy interrupts.
	 */
	pci_write_config_word(dev, PCI_COMMAND, PCI_COMMAND_INTX_DISABLE);
}

static void pci_dev_restore(struct pci_dev *dev)
{
	pci_restore_state(dev);
3826
	pci_reset_notify(dev, false);
3827 3828
}

3829 3830 3831 3832
static int pci_dev_reset(struct pci_dev *dev, int probe)
{
	int rc;

3833 3834
	if (!probe)
		pci_dev_lock(dev);
3835 3836 3837

	rc = __pci_dev_reset(dev, probe);

3838 3839 3840
	if (!probe)
		pci_dev_unlock(dev);

Y
Yu Zhao 已提交
3841
	return rc;
S
Sheng Yang 已提交
3842
}
3843

S
Sheng Yang 已提交
3844
/**
Y
Yu Zhao 已提交
3845 3846
 * __pci_reset_function - reset a PCI device function
 * @dev: PCI device to reset
S
Sheng Yang 已提交
3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857
 *
 * 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
3858
 * Returns 0 if the device function was successfully reset or negative if the
S
Sheng Yang 已提交
3859 3860
 * device doesn't support resetting a single function.
 */
Y
Yu Zhao 已提交
3861
int __pci_reset_function(struct pci_dev *dev)
S
Sheng Yang 已提交
3862
{
Y
Yu Zhao 已提交
3863
	return pci_dev_reset(dev, 0);
S
Sheng Yang 已提交
3864
}
Y
Yu Zhao 已提交
3865
EXPORT_SYMBOL_GPL(__pci_reset_function);
3866

3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887
/**
 * __pci_reset_function_locked - reset a PCI device function while holding
 * the @dev mutex lock.
 * @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.
 *
 * The device function is presumed to be unused and the caller is holding
 * the device mutex lock 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.
 *
 * Returns 0 if the device function was successfully reset or negative if the
 * device doesn't support resetting a single function.
 */
int __pci_reset_function_locked(struct pci_dev *dev)
{
3888
	return __pci_dev_reset(dev, 0);
3889 3890 3891
}
EXPORT_SYMBOL_GPL(__pci_reset_function_locked);

3892 3893 3894 3895 3896 3897 3898 3899 3900 3901 3902 3903 3904 3905 3906 3907
/**
 * 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);
}

3908
/**
Y
Yu Zhao 已提交
3909 3910
 * pci_reset_function - quiesce and reset a PCI device function
 * @dev: PCI device to reset
3911 3912 3913 3914 3915 3916 3917
 *
 * 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
3918
 * from __pci_reset_function in that it saves and restores device state
3919 3920
 * over the reset.
 *
Y
Yu Zhao 已提交
3921
 * Returns 0 if the device function was successfully reset or negative if the
3922 3923 3924 3925
 * device doesn't support resetting a single function.
 */
int pci_reset_function(struct pci_dev *dev)
{
Y
Yu Zhao 已提交
3926
	int rc;
3927

Y
Yu Zhao 已提交
3928 3929 3930
	rc = pci_dev_reset(dev, 1);
	if (rc)
		return rc;
3931

3932
	pci_dev_save_and_disable(dev);
3933

Y
Yu Zhao 已提交
3934
	rc = pci_dev_reset(dev, 0);
3935

3936
	pci_dev_restore(dev);
3937

Y
Yu Zhao 已提交
3938
	return rc;
3939 3940 3941
}
EXPORT_SYMBOL_GPL(pci_reset_function);

3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969
/**
 * pci_try_reset_function - quiesce and reset a PCI device function
 * @dev: PCI device to reset
 *
 * Same as above, except return -EAGAIN if unable to lock device.
 */
int pci_try_reset_function(struct pci_dev *dev)
{
	int rc;

	rc = pci_dev_reset(dev, 1);
	if (rc)
		return rc;

	pci_dev_save_and_disable(dev);

	if (pci_dev_trylock(dev)) {
		rc = __pci_dev_reset(dev, 0);
		pci_dev_unlock(dev);
	} else
		rc = -EAGAIN;

	pci_dev_restore(dev);

	return rc;
}
EXPORT_SYMBOL_GPL(pci_try_reset_function);

3970 3971 3972 3973 3974 3975 3976 3977 3978 3979 3980 3981 3982 3983
/* Do any devices on or below this bus prevent a bus reset? */
static bool pci_bus_resetable(struct pci_bus *bus)
{
	struct pci_dev *dev;

	list_for_each_entry(dev, &bus->devices, bus_list) {
		if (dev->dev_flags & PCI_DEV_FLAGS_NO_BUS_RESET ||
		    (dev->subordinate && !pci_bus_resetable(dev->subordinate)))
			return false;
	}

	return true;
}

3984 3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004 4005 4006 4007
/* Lock devices from the top of the tree down */
static void pci_bus_lock(struct pci_bus *bus)
{
	struct pci_dev *dev;

	list_for_each_entry(dev, &bus->devices, bus_list) {
		pci_dev_lock(dev);
		if (dev->subordinate)
			pci_bus_lock(dev->subordinate);
	}
}

/* Unlock devices from the bottom of the tree up */
static void pci_bus_unlock(struct pci_bus *bus)
{
	struct pci_dev *dev;

	list_for_each_entry(dev, &bus->devices, bus_list) {
		if (dev->subordinate)
			pci_bus_unlock(dev->subordinate);
		pci_dev_unlock(dev);
	}
}

4008 4009 4010 4011 4012 4013 4014 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028 4029 4030 4031 4032 4033
/* Return 1 on successful lock, 0 on contention */
static int pci_bus_trylock(struct pci_bus *bus)
{
	struct pci_dev *dev;

	list_for_each_entry(dev, &bus->devices, bus_list) {
		if (!pci_dev_trylock(dev))
			goto unlock;
		if (dev->subordinate) {
			if (!pci_bus_trylock(dev->subordinate)) {
				pci_dev_unlock(dev);
				goto unlock;
			}
		}
	}
	return 1;

unlock:
	list_for_each_entry_continue_reverse(dev, &bus->devices, bus_list) {
		if (dev->subordinate)
			pci_bus_unlock(dev->subordinate);
		pci_dev_unlock(dev);
	}
	return 0;
}

4034 4035 4036 4037 4038 4039 4040 4041 4042 4043 4044 4045 4046 4047 4048 4049
/* Do any devices on or below this slot prevent a bus reset? */
static bool pci_slot_resetable(struct pci_slot *slot)
{
	struct pci_dev *dev;

	list_for_each_entry(dev, &slot->bus->devices, bus_list) {
		if (!dev->slot || dev->slot != slot)
			continue;
		if (dev->dev_flags & PCI_DEV_FLAGS_NO_BUS_RESET ||
		    (dev->subordinate && !pci_bus_resetable(dev->subordinate)))
			return false;
	}

	return true;
}

4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067 4068 4069 4070 4071 4072 4073 4074 4075 4076 4077
/* Lock devices from the top of the tree down */
static void pci_slot_lock(struct pci_slot *slot)
{
	struct pci_dev *dev;

	list_for_each_entry(dev, &slot->bus->devices, bus_list) {
		if (!dev->slot || dev->slot != slot)
			continue;
		pci_dev_lock(dev);
		if (dev->subordinate)
			pci_bus_lock(dev->subordinate);
	}
}

/* Unlock devices from the bottom of the tree up */
static void pci_slot_unlock(struct pci_slot *slot)
{
	struct pci_dev *dev;

	list_for_each_entry(dev, &slot->bus->devices, bus_list) {
		if (!dev->slot || dev->slot != slot)
			continue;
		if (dev->subordinate)
			pci_bus_unlock(dev->subordinate);
		pci_dev_unlock(dev);
	}
}

4078 4079 4080 4081 4082 4083 4084 4085 4086 4087 4088 4089 4090 4091 4092 4093 4094 4095 4096 4097 4098 4099 4100 4101 4102 4103 4104 4105 4106 4107 4108
/* Return 1 on successful lock, 0 on contention */
static int pci_slot_trylock(struct pci_slot *slot)
{
	struct pci_dev *dev;

	list_for_each_entry(dev, &slot->bus->devices, bus_list) {
		if (!dev->slot || dev->slot != slot)
			continue;
		if (!pci_dev_trylock(dev))
			goto unlock;
		if (dev->subordinate) {
			if (!pci_bus_trylock(dev->subordinate)) {
				pci_dev_unlock(dev);
				goto unlock;
			}
		}
	}
	return 1;

unlock:
	list_for_each_entry_continue_reverse(dev,
					     &slot->bus->devices, bus_list) {
		if (!dev->slot || dev->slot != slot)
			continue;
		if (dev->subordinate)
			pci_bus_unlock(dev->subordinate);
		pci_dev_unlock(dev);
	}
	return 0;
}

4109 4110 4111 4112 4113 4114 4115 4116 4117 4118 4119 4120 4121 4122 4123 4124 4125 4126 4127 4128 4129 4130 4131 4132 4133 4134 4135 4136 4137 4138 4139 4140 4141 4142 4143 4144 4145 4146 4147 4148 4149 4150 4151 4152 4153 4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170
/* Save and disable devices from the top of the tree down */
static void pci_bus_save_and_disable(struct pci_bus *bus)
{
	struct pci_dev *dev;

	list_for_each_entry(dev, &bus->devices, bus_list) {
		pci_dev_save_and_disable(dev);
		if (dev->subordinate)
			pci_bus_save_and_disable(dev->subordinate);
	}
}

/*
 * Restore devices from top of the tree down - parent bridges need to be
 * restored before we can get to subordinate devices.
 */
static void pci_bus_restore(struct pci_bus *bus)
{
	struct pci_dev *dev;

	list_for_each_entry(dev, &bus->devices, bus_list) {
		pci_dev_restore(dev);
		if (dev->subordinate)
			pci_bus_restore(dev->subordinate);
	}
}

/* Save and disable devices from the top of the tree down */
static void pci_slot_save_and_disable(struct pci_slot *slot)
{
	struct pci_dev *dev;

	list_for_each_entry(dev, &slot->bus->devices, bus_list) {
		if (!dev->slot || dev->slot != slot)
			continue;
		pci_dev_save_and_disable(dev);
		if (dev->subordinate)
			pci_bus_save_and_disable(dev->subordinate);
	}
}

/*
 * Restore devices from top of the tree down - parent bridges need to be
 * restored before we can get to subordinate devices.
 */
static void pci_slot_restore(struct pci_slot *slot)
{
	struct pci_dev *dev;

	list_for_each_entry(dev, &slot->bus->devices, bus_list) {
		if (!dev->slot || dev->slot != slot)
			continue;
		pci_dev_restore(dev);
		if (dev->subordinate)
			pci_bus_restore(dev->subordinate);
	}
}

static int pci_slot_reset(struct pci_slot *slot, int probe)
{
	int rc;

4171
	if (!slot || !pci_slot_resetable(slot))
4172 4173 4174 4175 4176 4177 4178 4179 4180 4181 4182 4183 4184 4185 4186
		return -ENOTTY;

	if (!probe)
		pci_slot_lock(slot);

	might_sleep();

	rc = pci_reset_hotplug_slot(slot->hotplug, probe);

	if (!probe)
		pci_slot_unlock(slot);

	return rc;
}

4187 4188 4189 4190 4191 4192 4193 4194 4195 4196 4197 4198
/**
 * pci_probe_reset_slot - probe whether a PCI slot can be reset
 * @slot: PCI slot to probe
 *
 * Return 0 if slot can be reset, negative if a slot reset is not supported.
 */
int pci_probe_reset_slot(struct pci_slot *slot)
{
	return pci_slot_reset(slot, 1);
}
EXPORT_SYMBOL_GPL(pci_probe_reset_slot);

4199 4200 4201 4202 4203 4204 4205 4206 4207 4208 4209 4210 4211 4212 4213 4214 4215 4216 4217 4218 4219 4220 4221 4222 4223 4224 4225 4226 4227 4228 4229 4230 4231
/**
 * pci_reset_slot - reset a PCI slot
 * @slot: PCI slot to reset
 *
 * A PCI bus may host multiple slots, each slot may support a reset mechanism
 * independent of other slots.  For instance, some slots may support slot power
 * control.  In the case of a 1:1 bus to slot architecture, this function may
 * wrap the bus reset to avoid spurious slot related events such as hotplug.
 * Generally a slot reset should be attempted before a bus reset.  All of the
 * function of the slot and any subordinate buses behind the slot are reset
 * through this function.  PCI config space of all devices in the slot and
 * behind the slot is saved before and restored after reset.
 *
 * Return 0 on success, non-zero on error.
 */
int pci_reset_slot(struct pci_slot *slot)
{
	int rc;

	rc = pci_slot_reset(slot, 1);
	if (rc)
		return rc;

	pci_slot_save_and_disable(slot);

	rc = pci_slot_reset(slot, 0);

	pci_slot_restore(slot);

	return rc;
}
EXPORT_SYMBOL_GPL(pci_reset_slot);

4232 4233 4234 4235 4236 4237 4238 4239 4240 4241 4242 4243 4244 4245 4246 4247 4248 4249 4250 4251 4252 4253 4254 4255 4256 4257 4258 4259 4260
/**
 * pci_try_reset_slot - Try to reset a PCI slot
 * @slot: PCI slot to reset
 *
 * Same as above except return -EAGAIN if the slot cannot be locked
 */
int pci_try_reset_slot(struct pci_slot *slot)
{
	int rc;

	rc = pci_slot_reset(slot, 1);
	if (rc)
		return rc;

	pci_slot_save_and_disable(slot);

	if (pci_slot_trylock(slot)) {
		might_sleep();
		rc = pci_reset_hotplug_slot(slot->hotplug, 0);
		pci_slot_unlock(slot);
	} else
		rc = -EAGAIN;

	pci_slot_restore(slot);

	return rc;
}
EXPORT_SYMBOL_GPL(pci_try_reset_slot);

4261 4262
static int pci_bus_reset(struct pci_bus *bus, int probe)
{
4263
	if (!bus->self || !pci_bus_resetable(bus))
4264 4265 4266 4267 4268 4269 4270 4271 4272 4273 4274 4275 4276 4277 4278 4279
		return -ENOTTY;

	if (probe)
		return 0;

	pci_bus_lock(bus);

	might_sleep();

	pci_reset_bridge_secondary_bus(bus->self);

	pci_bus_unlock(bus);

	return 0;
}

4280 4281 4282 4283 4284 4285 4286 4287 4288 4289 4290 4291
/**
 * pci_probe_reset_bus - probe whether a PCI bus can be reset
 * @bus: PCI bus to probe
 *
 * Return 0 if bus can be reset, negative if a bus reset is not supported.
 */
int pci_probe_reset_bus(struct pci_bus *bus)
{
	return pci_bus_reset(bus, 1);
}
EXPORT_SYMBOL_GPL(pci_probe_reset_bus);

4292 4293 4294 4295 4296 4297 4298 4299 4300 4301 4302 4303 4304 4305 4306 4307 4308 4309 4310 4311 4312 4313 4314 4315 4316 4317 4318
/**
 * pci_reset_bus - reset a PCI bus
 * @bus: top level PCI bus to reset
 *
 * Do a bus reset on the given bus and any subordinate buses, saving
 * and restoring state of all devices.
 *
 * Return 0 on success, non-zero on error.
 */
int pci_reset_bus(struct pci_bus *bus)
{
	int rc;

	rc = pci_bus_reset(bus, 1);
	if (rc)
		return rc;

	pci_bus_save_and_disable(bus);

	rc = pci_bus_reset(bus, 0);

	pci_bus_restore(bus);

	return rc;
}
EXPORT_SYMBOL_GPL(pci_reset_bus);

4319 4320 4321 4322 4323 4324 4325 4326 4327 4328 4329 4330 4331 4332 4333 4334 4335 4336 4337 4338 4339 4340 4341 4342 4343 4344 4345 4346 4347
/**
 * pci_try_reset_bus - Try to reset a PCI bus
 * @bus: top level PCI bus to reset
 *
 * Same as above except return -EAGAIN if the bus cannot be locked
 */
int pci_try_reset_bus(struct pci_bus *bus)
{
	int rc;

	rc = pci_bus_reset(bus, 1);
	if (rc)
		return rc;

	pci_bus_save_and_disable(bus);

	if (pci_bus_trylock(bus)) {
		might_sleep();
		pci_reset_bridge_secondary_bus(bus->self);
		pci_bus_unlock(bus);
	} else
		rc = -EAGAIN;

	pci_bus_restore(bus);

	return rc;
}
EXPORT_SYMBOL_GPL(pci_try_reset_bus);

4348 4349 4350 4351 4352 4353 4354 4355 4356
/**
 * 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)
{
4357
	int cap;
4358 4359 4360 4361 4362 4363
	u32 stat;

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

4364
	if (pci_read_config_dword(dev, cap + PCI_X_STATUS, &stat))
4365 4366
		return -EINVAL;

4367
	return 512 << ((stat & PCI_X_STATUS_MAX_READ) >> 21);
4368 4369 4370 4371 4372 4373 4374 4375 4376 4377 4378 4379
}
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)
{
4380
	int cap;
4381
	u16 cmd;
4382 4383 4384 4385 4386

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

4387 4388
	if (pci_read_config_word(dev, cap + PCI_X_CMD, &cmd))
		return -EINVAL;
4389

4390
	return 512 << ((cmd & PCI_X_CMD_MAX_READ) >> 2);
4391 4392 4393 4394 4395 4396 4397 4398 4399 4400 4401 4402 4403 4404
}
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)
{
4405
	int cap;
4406 4407
	u32 stat, v, o;
	u16 cmd;
4408

4409
	if (mmrbc < 512 || mmrbc > 4096 || !is_power_of_2(mmrbc))
4410
		return -EINVAL;
4411 4412 4413 4414 4415

	v = ffs(mmrbc) - 10;

	cap = pci_find_capability(dev, PCI_CAP_ID_PCIX);
	if (!cap)
4416
		return -EINVAL;
4417

4418 4419
	if (pci_read_config_dword(dev, cap + PCI_X_STATUS, &stat))
		return -EINVAL;
4420 4421 4422 4423

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

4424 4425
	if (pci_read_config_word(dev, cap + PCI_X_CMD, &cmd))
		return -EINVAL;
4426 4427 4428

	o = (cmd & PCI_X_CMD_MAX_READ) >> 2;
	if (o != v) {
4429
		if (v > o && (dev->bus->bus_flags & PCI_BUS_FLAGS_NO_MMRBC))
4430 4431 4432 4433
			return -EIO;

		cmd &= ~PCI_X_CMD_MAX_READ;
		cmd |= v << 2;
4434 4435
		if (pci_write_config_word(dev, cap + PCI_X_CMD, cmd))
			return -EIO;
4436
	}
4437
	return 0;
4438 4439 4440 4441 4442 4443 4444 4445 4446 4447 4448 4449 4450 4451
}
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)
{
	u16 ctl;

4452
	pcie_capability_read_word(dev, PCI_EXP_DEVCTL, &ctl);
4453

4454
	return 128 << ((ctl & PCI_EXP_DEVCTL_READRQ) >> 12);
4455 4456 4457 4458 4459 4460
}
EXPORT_SYMBOL(pcie_get_readrq);

/**
 * pcie_set_readrq - set PCI Express maximum memory read request
 * @dev: PCI device to query
4461
 * @rq: maximum memory read count in bytes
4462 4463
 *    valid values are 128, 256, 512, 1024, 2048, 4096
 *
4464
 * If possible sets maximum memory read request in bytes
4465 4466 4467
 */
int pcie_set_readrq(struct pci_dev *dev, int rq)
{
4468
	u16 v;
4469

4470
	if (rq < 128 || rq > 4096 || !is_power_of_2(rq))
4471
		return -EINVAL;
4472

4473 4474 4475 4476 4477 4478 4479 4480 4481 4482 4483 4484 4485 4486
	/*
	 * If using the "performance" PCIe config, we clamp the
	 * read rq size to the max packet size to prevent the
	 * host bridge generating requests larger than we can
	 * cope with
	 */
	if (pcie_bus_config == PCIE_BUS_PERFORMANCE) {
		int mps = pcie_get_mps(dev);

		if (mps < rq)
			rq = mps;
	}

	v = (ffs(rq) - 8) << 12;
4487

4488 4489
	return pcie_capability_clear_and_set_word(dev, PCI_EXP_DEVCTL,
						  PCI_EXP_DEVCTL_READRQ, v);
4490 4491 4492
}
EXPORT_SYMBOL(pcie_set_readrq);

4493 4494 4495 4496 4497 4498 4499 4500 4501 4502
/**
 * pcie_get_mps - get PCI Express maximum payload size
 * @dev: PCI device to query
 *
 * Returns maximum payload size in bytes
 */
int pcie_get_mps(struct pci_dev *dev)
{
	u16 ctl;

4503
	pcie_capability_read_word(dev, PCI_EXP_DEVCTL, &ctl);
4504

4505
	return 128 << ((ctl & PCI_EXP_DEVCTL_PAYLOAD) >> 5);
4506
}
4507
EXPORT_SYMBOL(pcie_get_mps);
4508 4509 4510 4511

/**
 * pcie_set_mps - set PCI Express maximum payload size
 * @dev: PCI device to query
4512
 * @mps: maximum payload size in bytes
4513 4514 4515 4516 4517 4518
 *    valid values are 128, 256, 512, 1024, 2048, 4096
 *
 * If possible sets maximum payload size
 */
int pcie_set_mps(struct pci_dev *dev, int mps)
{
4519
	u16 v;
4520 4521

	if (mps < 128 || mps > 4096 || !is_power_of_2(mps))
4522
		return -EINVAL;
4523 4524

	v = ffs(mps) - 8;
4525
	if (v > dev->pcie_mpss)
4526
		return -EINVAL;
4527 4528
	v <<= 5;

4529 4530
	return pcie_capability_clear_and_set_word(dev, PCI_EXP_DEVCTL,
						  PCI_EXP_DEVCTL_PAYLOAD, v);
4531
}
4532
EXPORT_SYMBOL(pcie_set_mps);
4533

4534 4535 4536 4537 4538 4539 4540 4541 4542 4543 4544 4545 4546 4547 4548 4549 4550 4551 4552 4553 4554 4555 4556 4557 4558 4559 4560 4561 4562 4563 4564 4565 4566 4567 4568 4569 4570 4571 4572 4573 4574 4575 4576
/**
 * pcie_get_minimum_link - determine minimum link settings of a PCI device
 * @dev: PCI device to query
 * @speed: storage for minimum speed
 * @width: storage for minimum width
 *
 * This function will walk up the PCI device chain and determine the minimum
 * link width and speed of the device.
 */
int pcie_get_minimum_link(struct pci_dev *dev, enum pci_bus_speed *speed,
			  enum pcie_link_width *width)
{
	int ret;

	*speed = PCI_SPEED_UNKNOWN;
	*width = PCIE_LNK_WIDTH_UNKNOWN;

	while (dev) {
		u16 lnksta;
		enum pci_bus_speed next_speed;
		enum pcie_link_width next_width;

		ret = pcie_capability_read_word(dev, PCI_EXP_LNKSTA, &lnksta);
		if (ret)
			return ret;

		next_speed = pcie_link_speed[lnksta & PCI_EXP_LNKSTA_CLS];
		next_width = (lnksta & PCI_EXP_LNKSTA_NLW) >>
			PCI_EXP_LNKSTA_NLW_SHIFT;

		if (next_speed < *speed)
			*speed = next_speed;

		if (next_width < *width)
			*width = next_width;

		dev = dev->bus->self;
	}

	return 0;
}
EXPORT_SYMBOL(pcie_get_minimum_link);

4577 4578
/**
 * pci_select_bars - Make BAR mask from the type of resource
4579
 * @dev: the PCI device for which BAR mask is made
4580 4581 4582 4583 4584 4585 4586 4587 4588 4589 4590 4591
 * @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;
}
4592
EXPORT_SYMBOL(pci_select_bars);
4593

4594 4595 4596 4597 4598 4599 4600 4601 4602 4603
/**
 * 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)
{
4604 4605
	int reg;

4606 4607 4608 4609 4610 4611
	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;
4612 4613
	} else if (resno < PCI_BRIDGE_RESOURCES) {
		/* device specific resource */
4614 4615
		*type = pci_bar_unknown;
		reg = pci_iov_resource_bar(dev, resno);
4616 4617
		if (reg)
			return reg;
4618 4619
	}

4620
	dev_err(&dev->dev, "BAR %d: invalid resource\n", resno);
4621 4622 4623
	return 0;
}

4624 4625 4626 4627 4628 4629 4630 4631 4632
/* 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,
R
Ryan Desfosses 已提交
4633
				  unsigned int command_bits, u32 flags)
4634 4635 4636
{
	if (arch_set_vga_state)
		return arch_set_vga_state(dev, decode, command_bits,
4637
						flags);
4638 4639 4640
	return 0;
}

4641 4642
/**
 * pci_set_vga_state - set VGA decode state on device and parents if requested
R
Randy Dunlap 已提交
4643 4644 4645
 * @dev: the PCI device
 * @decode: true = enable decoding, false = disable decoding
 * @command_bits: PCI_COMMAND_IO and/or PCI_COMMAND_MEMORY
R
Randy Dunlap 已提交
4646
 * @flags: traverse ancestors and change bridges
4647
 * CHANGE_BRIDGE_ONLY / CHANGE_BRIDGE
4648 4649
 */
int pci_set_vga_state(struct pci_dev *dev, bool decode,
4650
		      unsigned int command_bits, u32 flags)
4651 4652 4653 4654
{
	struct pci_bus *bus;
	struct pci_dev *bridge;
	u16 cmd;
4655
	int rc;
4656

4657
	WARN_ON((flags & PCI_VGA_STATE_CHANGE_DECODES) && (command_bits & ~(PCI_COMMAND_IO|PCI_COMMAND_MEMORY)));
4658

4659
	/* ARCH specific VGA enables */
4660
	rc = pci_set_vga_state_arch(dev, decode, command_bits, flags);
4661 4662 4663
	if (rc)
		return rc;

4664 4665 4666 4667 4668 4669 4670 4671
	if (flags & PCI_VGA_STATE_CHANGE_DECODES) {
		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);
	}
4672

4673
	if (!(flags & PCI_VGA_STATE_CHANGE_BRIDGE))
4674 4675 4676 4677 4678 4679 4680 4681 4682 4683 4684 4685 4686 4687 4688 4689 4690 4691 4692 4693
		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;
}

4694 4695 4696 4697 4698 4699 4700 4701 4702 4703
/**
 * pci_add_dma_alias - Add a DMA devfn alias for a device
 * @dev: the PCI device for which alias is added
 * @devfn: alias slot and function
 *
 * This helper encodes 8-bit devfn as bit number in dma_alias_mask.
 * It should be called early, preferably as PCI fixup header quirk.
 */
void pci_add_dma_alias(struct pci_dev *dev, u8 devfn)
{
4704 4705 4706 4707 4708 4709 4710 4711 4712
	if (!dev->dma_alias_mask)
		dev->dma_alias_mask = kcalloc(BITS_TO_LONGS(U8_MAX),
					      sizeof(long), GFP_KERNEL);
	if (!dev->dma_alias_mask) {
		dev_warn(&dev->dev, "Unable to allocate DMA alias mask\n");
		return;
	}

	set_bit(devfn, dev->dma_alias_mask);
4713 4714
	dev_info(&dev->dev, "Enabling fixed DMA alias to %02x.%d\n",
		 PCI_SLOT(devfn), PCI_FUNC(devfn));
4715 4716
}

4717 4718 4719 4720 4721 4722 4723 4724
bool pci_devs_are_dma_aliases(struct pci_dev *dev1, struct pci_dev *dev2)
{
	return (dev1->dma_alias_mask &&
		test_bit(dev2->devfn, dev1->dma_alias_mask)) ||
	       (dev2->dma_alias_mask &&
		test_bit(dev1->devfn, dev2->dma_alias_mask));
}

4725 4726 4727 4728 4729 4730 4731 4732
bool pci_device_is_present(struct pci_dev *pdev)
{
	u32 v;

	return pci_bus_read_dev_vendor_id(pdev->bus, pdev->devfn, &v, 0);
}
EXPORT_SYMBOL_GPL(pci_device_is_present);

4733 4734 4735 4736 4737 4738 4739 4740 4741 4742 4743
void pci_ignore_hotplug(struct pci_dev *dev)
{
	struct pci_dev *bridge = dev->bus->self;

	dev->ignore_hotplug = 1;
	/* Propagate the "ignore hotplug" setting to the parent bridge. */
	if (bridge)
		bridge->ignore_hotplug = 1;
}
EXPORT_SYMBOL_GPL(pci_ignore_hotplug);

4744 4745
#define RESOURCE_ALIGNMENT_PARAM_SIZE COMMAND_LINE_SIZE
static char resource_alignment_param[RESOURCE_ALIGNMENT_PARAM_SIZE] = {0};
4746
static DEFINE_SPINLOCK(resource_alignment_lock);
4747 4748 4749 4750 4751 4752 4753 4754

/**
 * 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.
 */
4755
static resource_size_t pci_specified_resource_alignment(struct pci_dev *dev)
4756 4757 4758 4759 4760 4761 4762 4763 4764 4765 4766 4767 4768 4769 4770 4771 4772 4773 4774 4775 4776 4777 4778 4779 4780 4781 4782 4783 4784 4785 4786
{
	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)) {
R
Ryan Desfosses 已提交
4787
			if (align_order == -1)
4788
				align = PAGE_SIZE;
R
Ryan Desfosses 已提交
4789
			else
4790 4791 4792 4793 4794 4795 4796 4797 4798 4799 4800 4801 4802 4803
				align = 1 << align_order;
			/* Found */
			break;
		}
		if (*p != ';' && *p != ',') {
			/* End of param or invalid format */
			break;
		}
		p++;
	}
	spin_unlock(&resource_alignment_lock);
	return align;
}

4804 4805 4806 4807 4808 4809 4810 4811 4812 4813 4814 4815 4816 4817
/*
 * This function disables memory decoding and releases memory resources
 * of the device specified by kernel's boot parameter 'pci=resource_alignment='.
 * It also rounds up size to specified alignment.
 * Later on, the kernel will assign page-aligned memory resource back
 * to the device.
 */
void pci_reassigndev_resource_alignment(struct pci_dev *dev)
{
	int i;
	struct resource *r;
	resource_size_t align, size;
	u16 command;

Y
Yinghai Lu 已提交
4818 4819 4820
	/* check if specified PCI is target device to reassign */
	align = pci_specified_resource_alignment(dev);
	if (!align)
4821 4822 4823 4824 4825 4826 4827 4828 4829 4830 4831 4832 4833 4834 4835 4836 4837 4838 4839 4840 4841 4842 4843 4844 4845 4846
		return;

	if (dev->hdr_type == PCI_HEADER_TYPE_NORMAL &&
	    (dev->class >> 8) == PCI_CLASS_BRIDGE_HOST) {
		dev_warn(&dev->dev,
			"Can't reassign resources to host bridge.\n");
		return;
	}

	dev_info(&dev->dev,
		"Disabling memory decoding and releasing memory resources.\n");
	pci_read_config_word(dev, PCI_COMMAND, &command);
	command &= ~PCI_COMMAND_MEMORY;
	pci_write_config_word(dev, PCI_COMMAND, command);

	for (i = 0; i < PCI_BRIDGE_RESOURCES; i++) {
		r = &dev->resource[i];
		if (!(r->flags & IORESOURCE_MEM))
			continue;
		size = resource_size(r);
		if (size < align) {
			size = align;
			dev_info(&dev->dev,
				"Rounding up size of resource #%d to %#llx.\n",
				i, (unsigned long long)size);
		}
4847
		r->flags |= IORESOURCE_UNSET;
4848 4849 4850 4851 4852 4853 4854 4855 4856 4857 4858 4859 4860
		r->end = size - 1;
		r->start = 0;
	}
	/* Need to disable bridge's resource window,
	 * to enable the kernel to reassign new resource
	 * window later on.
	 */
	if (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE &&
	    (dev->class >> 8) == PCI_CLASS_BRIDGE_PCI) {
		for (i = PCI_BRIDGE_RESOURCES; i < PCI_NUM_RESOURCES; i++) {
			r = &dev->resource[i];
			if (!(r->flags & IORESOURCE_MEM))
				continue;
4861
			r->flags |= IORESOURCE_UNSET;
4862 4863 4864 4865 4866 4867 4868
			r->end = resource_size(r) - 1;
			r->start = 0;
		}
		pci_disable_bridge_window(dev);
	}
}

4869
static ssize_t pci_set_resource_alignment_param(const char *buf, size_t count)
4870 4871 4872 4873 4874 4875 4876 4877 4878 4879
{
	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;
}

4880
static ssize_t pci_get_resource_alignment_param(char *buf, size_t size)
4881 4882 4883 4884 4885 4886 4887 4888 4889 4890 4891 4892 4893 4894 4895 4896 4897 4898 4899 4900 4901 4902 4903 4904 4905 4906 4907 4908 4909
{
	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);

B
Bill Pemberton 已提交
4910
static void pci_no_domains(void)
4911 4912 4913 4914 4915 4916
{
#ifdef CONFIG_PCI_DOMAINS
	pci_domains_supported = 0;
#endif
}

4917 4918 4919 4920 4921 4922 4923
#ifdef CONFIG_PCI_DOMAINS
static atomic_t __domain_nr = ATOMIC_INIT(-1);

int pci_get_new_domain_nr(void)
{
	return atomic_inc_return(&__domain_nr);
}
4924 4925 4926 4927 4928

#ifdef CONFIG_PCI_DOMAINS_GENERIC
void pci_bus_assign_domain_nr(struct pci_bus *bus, struct device *parent)
{
	static int use_dt_domains = -1;
4929
	int domain = -1;
4930

4931 4932
	if (parent)
		domain = of_get_pci_domain_nr(parent->of_node);
4933 4934 4935 4936 4937 4938 4939 4940 4941 4942 4943 4944 4945 4946 4947 4948 4949 4950 4951 4952 4953 4954 4955 4956 4957 4958 4959 4960 4961 4962 4963 4964 4965 4966 4967 4968 4969 4970 4971 4972
	/*
	 * Check DT domain and use_dt_domains values.
	 *
	 * If DT domain property is valid (domain >= 0) and
	 * use_dt_domains != 0, the DT assignment is valid since this means
	 * we have not previously allocated a domain number by using
	 * pci_get_new_domain_nr(); we should also update use_dt_domains to
	 * 1, to indicate that we have just assigned a domain number from
	 * DT.
	 *
	 * If DT domain property value is not valid (ie domain < 0), and we
	 * have not previously assigned a domain number from DT
	 * (use_dt_domains != 1) we should assign a domain number by
	 * using the:
	 *
	 * pci_get_new_domain_nr()
	 *
	 * API and update the use_dt_domains value to keep track of method we
	 * are using to assign domain numbers (use_dt_domains = 0).
	 *
	 * All other combinations imply we have a platform that is trying
	 * to mix domain numbers obtained from DT and pci_get_new_domain_nr(),
	 * which is a recipe for domain mishandling and it is prevented by
	 * invalidating the domain value (domain = -1) and printing a
	 * corresponding error.
	 */
	if (domain >= 0 && use_dt_domains) {
		use_dt_domains = 1;
	} else if (domain < 0 && use_dt_domains != 1) {
		use_dt_domains = 0;
		domain = pci_get_new_domain_nr();
	} else {
		dev_err(parent, "Node %s has inconsistent \"linux,pci-domain\" property in DT\n",
			parent->of_node->full_name);
		domain = -1;
	}

	bus->domain_nr = domain;
}
#endif
4973 4974
#endif

4975
/**
4976
 * pci_ext_cfg_avail - can we access extended PCI config space?
4977 4978 4979 4980 4981
 *
 * Returns 1 if we can access PCI extended config space (offsets
 * greater than 0xff). This is the default implementation. Architecture
 * implementations can override this.
 */
4982
int __weak pci_ext_cfg_avail(void)
4983 4984 4985 4986
{
	return 1;
}

4987 4988 4989 4990 4991
void __weak pci_fixup_cardbus(struct pci_bus *bus)
{
}
EXPORT_SYMBOL(pci_fixup_cardbus);

A
Al Viro 已提交
4992
static int __init pci_setup(char *str)
L
Linus Torvalds 已提交
4993 4994 4995 4996 4997 4998
{
	while (str) {
		char *k = strchr(str, ',');
		if (k)
			*k++ = 0;
		if (*str && (str = pcibios_setup(str)) && *str) {
4999 5000
			if (!strcmp(str, "nomsi")) {
				pci_no_msi();
R
Randy Dunlap 已提交
5001 5002
			} else if (!strcmp(str, "noaer")) {
				pci_no_aer();
5003 5004
			} else if (!strncmp(str, "realloc=", 8)) {
				pci_realloc_get_opt(str + 8);
5005
			} else if (!strncmp(str, "realloc", 7)) {
5006
				pci_realloc_get_opt("on");
5007 5008
			} else if (!strcmp(str, "nodomains")) {
				pci_no_domains();
5009 5010
			} else if (!strncmp(str, "noari", 5)) {
				pcie_ari_disabled = true;
5011 5012 5013 5014
			} 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);
5015 5016 5017
			} else if (!strncmp(str, "resource_alignment=", 19)) {
				pci_set_resource_alignment_param(str + 19,
							strlen(str + 19));
5018 5019
			} else if (!strncmp(str, "ecrc=", 5)) {
				pcie_ecrc_get_policy(str + 5);
5020 5021 5022 5023
			} 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);
5024 5025
			} else if (!strncmp(str, "pcie_bus_tune_off", 17)) {
				pcie_bus_config = PCIE_BUS_TUNE_OFF;
5026 5027 5028 5029
			} else if (!strncmp(str, "pcie_bus_safe", 13)) {
				pcie_bus_config = PCIE_BUS_SAFE;
			} else if (!strncmp(str, "pcie_bus_perf", 13)) {
				pcie_bus_config = PCIE_BUS_PERFORMANCE;
5030 5031
			} else if (!strncmp(str, "pcie_bus_peer2peer", 18)) {
				pcie_bus_config = PCIE_BUS_PEER2PEER;
5032 5033
			} else if (!strncmp(str, "pcie_scan_all", 13)) {
				pci_add_flags(PCI_SCAN_ALL_PCIE_DEVS);
5034 5035 5036 5037
			} else {
				printk(KERN_ERR "PCI: Unknown option `%s'\n",
						str);
			}
L
Linus Torvalds 已提交
5038 5039 5040
		}
		str = k;
	}
5041
	return 0;
L
Linus Torvalds 已提交
5042
}
5043
early_param("pci", pci_setup);