pci.c 102.9 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13
/*
 *	PCI Bus Services, see include/linux/pci.h for further explanation.
 *
 *	Copyright 1993 -- 1997 Drew Eckhardt, Frederic Potter,
 *	David Mosberger-Tang
 *
 *	Copyright 1997 -- 2000 Martin Mares <mj@ucw.cz>
 */

#include <linux/kernel.h>
#include <linux/delay.h>
#include <linux/init.h>
#include <linux/pci.h>
14
#include <linux/pm.h>
15
#include <linux/slab.h>
L
Linus Torvalds 已提交
16 17
#include <linux/module.h>
#include <linux/spinlock.h>
T
Tim Schmielau 已提交
18
#include <linux/string.h>
19
#include <linux/log2.h>
S
Shaohua Li 已提交
20
#include <linux/pci-aspm.h>
21
#include <linux/pm_wakeup.h>
22
#include <linux/interrupt.h>
23
#include <linux/device.h>
24
#include <linux/pm_runtime.h>
25
#include <asm-generic/pci-bridge.h>
26
#include <asm/setup.h>
27
#include "pci.h"
L
Linus Torvalds 已提交
28

A
Alan Stern 已提交
29 30 31 32 33
const char *pci_power_names[] = {
	"error", "D0", "D1", "D2", "D3hot", "D3cold", "unknown",
};
EXPORT_SYMBOL_GPL(pci_power_names);

34 35 36 37 38 39
int isa_dma_bridge_buggy;
EXPORT_SYMBOL(isa_dma_bridge_buggy);

int pci_pci_problems;
EXPORT_SYMBOL(pci_pci_problems);

40 41
unsigned int pci_pm_d3_delay;

42 43 44 45 46 47 48 49 50 51 52 53 54
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 */

55 56 57 58 59 60 61 62 63
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
64

65 66 67 68
#ifdef CONFIG_PCI_DOMAINS
int pci_domains_supported = 1;
#endif

69 70 71 72 73 74
#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;

75 76 77 78 79 80
#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;

81
enum pcie_bus_config_types pcie_bus_config = PCIE_BUS_TUNE_OFF;
82

83 84 85 86 87 88
/*
 * 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
89
u8 pci_dfl_cache_line_size = L1_CACHE_BYTES >> 2;
90 91
u8 pci_cache_line_size;

92 93 94 95 96 97
/*
 * 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;

98 99 100
/* If set, the PCIe ARI capability will not be used. */
static bool pcie_ari_disabled;

L
Linus Torvalds 已提交
101 102 103 104 105 106 107
/**
 * 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.
 */
108
unsigned char pci_bus_max_busnr(struct pci_bus* bus)
L
Linus Torvalds 已提交
109 110 111 112
{
	struct list_head *tmp;
	unsigned char max, n;

113
	max = bus->busn_res.end;
L
Linus Torvalds 已提交
114 115 116 117 118 119 120
	list_for_each(tmp, &bus->children) {
		n = pci_bus_max_busnr(pci_bus_b(tmp));
		if(n > max)
			max = n;
	}
	return max;
}
121
EXPORT_SYMBOL_GPL(pci_bus_max_busnr);
L
Linus Torvalds 已提交
122

A
Andrew Morton 已提交
123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138
#ifdef CONFIG_HAS_IOMEM
void __iomem *pci_ioremap_bar(struct pci_dev *pdev, int bar)
{
	/*
	 * Make sure the BAR is actually a memory resource, not an IO resource
	 */
	if (!(pci_resource_flags(pdev, bar) & IORESOURCE_MEM)) {
		WARN_ON(1);
		return NULL;
	}
	return ioremap_nocache(pci_resource_start(pdev, bar),
				     pci_resource_len(pdev, bar));
}
EXPORT_SYMBOL_GPL(pci_ioremap_bar);
#endif

139 140 141 142
#define PCI_FIND_CAP_TTL	48

static int __pci_find_next_cap_ttl(struct pci_bus *bus, unsigned int devfn,
				   u8 pos, int cap, int *ttl)
143 144 145
{
	u8 id;

146
	while ((*ttl)--) {
147 148 149 150 151 152 153 154 155 156 157 158 159 160 161
		pci_bus_read_config_byte(bus, devfn, pos, &pos);
		if (pos < 0x40)
			break;
		pos &= ~3;
		pci_bus_read_config_byte(bus, devfn, pos + PCI_CAP_LIST_ID,
					 &id);
		if (id == 0xff)
			break;
		if (id == cap)
			return pos;
		pos += PCI_CAP_LIST_NEXT;
	}
	return 0;
}

162 163 164 165 166 167 168 169
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);
}

170 171 172 173 174 175 176
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);

177 178
static int __pci_bus_find_cap_start(struct pci_bus *bus,
				    unsigned int devfn, u8 hdr_type)
L
Linus Torvalds 已提交
179 180 181 182 183 184 185 186 187 188
{
	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:
189
		return PCI_CAPABILITY_LIST;
L
Linus Torvalds 已提交
190
	case PCI_HEADER_TYPE_CARDBUS:
191
		return PCI_CB_CAPABILITY_LIST;
L
Linus Torvalds 已提交
192 193 194
	default:
		return 0;
	}
195 196

	return 0;
L
Linus Torvalds 已提交
197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219
}

/**
 * pci_find_capability - query for devices' capabilities 
 * @dev: PCI device to query
 * @cap: capability code
 *
 * Tell if a device supports a given PCI capability.
 * Returns the address of the requested capability structure within the
 * device's PCI configuration space or 0 in case the device does not
 * support it.  Possible values for @cap:
 *
 *  %PCI_CAP_ID_PM           Power Management 
 *  %PCI_CAP_ID_AGP          Accelerated Graphics Port 
 *  %PCI_CAP_ID_VPD          Vital Product Data 
 *  %PCI_CAP_ID_SLOTID       Slot Identification 
 *  %PCI_CAP_ID_MSI          Message Signalled Interrupts
 *  %PCI_CAP_ID_CHSWP        CompactPCI HotSwap 
 *  %PCI_CAP_ID_PCIX         PCI-X
 *  %PCI_CAP_ID_EXP          PCI Express
 */
int pci_find_capability(struct pci_dev *dev, int cap)
{
220 221 222 223 224 225 226
	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
227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243
}

/**
 * pci_bus_find_capability - query for devices' capabilities 
 * @bus:   the PCI bus to query
 * @devfn: PCI device to query
 * @cap:   capability code
 *
 * Like pci_find_capability() but works for pci devices that do not have a
 * pci_dev structure set up yet. 
 *
 * Returns the address of the requested capability structure within the
 * device's PCI configuration space or 0 in case the device does not
 * support it.
 */
int pci_bus_find_capability(struct pci_bus *bus, unsigned int devfn, int cap)
{
244
	int pos;
L
Linus Torvalds 已提交
245 246 247 248
	u8 hdr_type;

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

249 250 251 252 253
	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
254 255 256
}

/**
257
 * pci_find_next_ext_capability - Find an extended capability
L
Linus Torvalds 已提交
258
 * @dev: PCI device to query
259
 * @start: address at which to start looking (0 to start at beginning of list)
L
Linus Torvalds 已提交
260 261
 * @cap: capability code
 *
262
 * Returns the address of the next matching extended capability structure
L
Linus Torvalds 已提交
263
 * within the device's PCI configuration space or 0 if the device does
264 265
 * 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
266
 */
267
int pci_find_next_ext_capability(struct pci_dev *dev, int start, int cap)
L
Linus Torvalds 已提交
268 269
{
	u32 header;
270 271
	int ttl;
	int pos = PCI_CFG_SPACE_SIZE;
L
Linus Torvalds 已提交
272

273 274 275 276
	/* 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
277 278
		return 0;

279 280 281
	if (start)
		pos = start;

L
Linus Torvalds 已提交
282 283 284 285 286 287 288 289 290 291 292
	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) {
293
		if (PCI_EXT_CAP_ID(header) == cap && pos != start)
L
Linus Torvalds 已提交
294 295 296
			return pos;

		pos = PCI_EXT_CAP_NEXT(header);
297
		if (pos < PCI_CFG_SPACE_SIZE)
L
Linus Torvalds 已提交
298 299 300 301 302 303 304 305
			break;

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

	return 0;
}
306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325
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);
}
326
EXPORT_SYMBOL_GPL(pci_find_ext_capability);
L
Linus Torvalds 已提交
327

328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347
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;

348 349
		pos = __pci_find_next_cap_ttl(dev->bus, dev->devfn,
					      pos + PCI_CAP_LIST_NEXT,
350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396
					      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
397 398 399 400 401 402 403 404 405 406 407 408 409 410
/**
 * pci_find_parent_resource - return resource region of parent bus of given region
 * @dev: PCI device structure contains resources to be searched
 * @res: child resource record for which parent is sought
 *
 *  For given resource region of given device, return the resource
 *  region of parent bus the given region is contained in or where
 *  it should be allocated from.
 */
struct resource *
pci_find_parent_resource(const struct pci_dev *dev, struct resource *res)
{
	const struct pci_bus *bus = dev->bus;
	int i;
411
	struct resource *best = NULL, *r;
L
Linus Torvalds 已提交
412

413
	pci_bus_for_each_resource(bus, r, i) {
L
Linus Torvalds 已提交
414 415 416 417 418 419 420 421
		if (!r)
			continue;
		if (res->start && !(res->start >= r->start && res->end <= r->end))
			continue;	/* Not contained */
		if ((res->flags ^ r->flags) & (IORESOURCE_IO | IORESOURCE_MEM))
			continue;	/* Wrong type */
		if (!((res->flags ^ r->flags) & IORESOURCE_PREFETCH))
			return r;	/* Exact match */
422 423 424 425 426 427
		/* We can't insert a non-prefetch resource inside a prefetchable parent .. */
		if (r->flags & IORESOURCE_PREFETCH)
			continue;
		/* .. but we can put a prefetchable resource inside a non-prefetchable one */
		if (!best)
			best = r;
L
Linus Torvalds 已提交
428 429 430 431
	}
	return best;
}

432 433 434 435 436 437 438
/**
 * pci_restore_bars - restore a devices BAR values (e.g. after wake-up)
 * @dev: PCI device to have its BARs restored
 *
 * Restore the BAR values for a given device, so as to make it
 * accessible by its driver.
 */
439
static void
440 441
pci_restore_bars(struct pci_dev *dev)
{
442
	int i;
443

444
	for (i = 0; i < PCI_BRIDGE_RESOURCES; i++)
445
		pci_update_resource(dev, i);
446 447
}

448 449 450 451
static struct pci_platform_pm_ops *pci_platform_pm;

int pci_set_platform_pm(struct pci_platform_pm_ops *ops)
{
452
	if (!ops->is_manageable || !ops->set_state || !ops->choose_state
453
	    || !ops->sleep_wake)
454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474
		return -EINVAL;
	pci_platform_pm = ops;
	return 0;
}

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

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

static inline pci_power_t platform_pci_choose_state(struct pci_dev *dev)
{
	return pci_platform_pm ?
			pci_platform_pm->choose_state(dev) : PCI_POWER_ERROR;
}
R
Randy Dunlap 已提交
475

476 477 478 479 480 481
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;
}

482 483 484 485 486 487
static inline int platform_pci_run_wake(struct pci_dev *dev, bool enable)
{
	return pci_platform_pm ?
			pci_platform_pm->run_wake(dev, enable) : -ENODEV;
}

L
Linus Torvalds 已提交
488
/**
489 490 491 492
 * 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
493
 *
494 495 496 497 498 499
 * 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
500
 */
501
static int pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state)
L
Linus Torvalds 已提交
502
{
503
	u16 pmcsr;
504
	bool need_restore = false;
L
Linus Torvalds 已提交
505

506 507 508 509
	/* Check if we're already there */
	if (dev->current_state == state)
		return 0;

510
	if (!dev->pm_cap)
511 512
		return -EIO;

513 514 515
	if (state < PCI_D0 || state > PCI_D3hot)
		return -EINVAL;

L
Linus Torvalds 已提交
516 517 518 519
	/* Validate current state:
	 * Can enter D0 from any state, but if we can only go deeper 
	 * to sleep if we're already in a low power state
	 */
520
	if (state != PCI_D0 && dev->current_state <= PCI_D3cold
521
	    && dev->current_state > state) {
522 523
		dev_err(&dev->dev, "invalid power transition "
			"(from state %d to %d)\n", dev->current_state, state);
L
Linus Torvalds 已提交
524
		return -EINVAL;
525
	}
L
Linus Torvalds 已提交
526 527

	/* check if this device supports the desired state */
528 529
	if ((state == PCI_D1 && !dev->d1_support)
	   || (state == PCI_D2 && !dev->d2_support))
530
		return -EIO;
L
Linus Torvalds 已提交
531

532
	pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
533

534
	/* If we're (effectively) in D3, force entire word to 0.
L
Linus Torvalds 已提交
535 536 537
	 * This doesn't affect PME_Status, disables PME_En, and
	 * sets PowerState to 0.
	 */
538
	switch (dev->current_state) {
539 540 541 542 543 544
	case PCI_D0:
	case PCI_D1:
	case PCI_D2:
		pmcsr &= ~PCI_PM_CTRL_STATE_MASK;
		pmcsr |= state;
		break;
545 546
	case PCI_D3hot:
	case PCI_D3cold:
547 548
	case PCI_UNKNOWN: /* Boot-up */
		if ((pmcsr & PCI_PM_CTRL_STATE_MASK) == PCI_D3hot
549
		 && !(pmcsr & PCI_PM_CTRL_NO_SOFT_RESET))
550
			need_restore = true;
551 552
		/* Fall-through: force to D0 */
	default:
553
		pmcsr = 0;
554
		break;
L
Linus Torvalds 已提交
555 556 557
	}

	/* enter specified state */
558
	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, pmcsr);
L
Linus Torvalds 已提交
559 560 561 562

	/* Mandatory power management transition delays */
	/* see PCI PM 1.1 5.6.1 table 18 */
	if (state == PCI_D3hot || dev->current_state == PCI_D3hot)
563
		pci_dev_d3_sleep(dev);
L
Linus Torvalds 已提交
564
	else if (state == PCI_D2 || dev->current_state == PCI_D2)
565
		udelay(PCI_PM_D2_DELAY);
L
Linus Torvalds 已提交
566

567 568 569 570 571
	pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
	dev->current_state = (pmcsr & PCI_PM_CTRL_STATE_MASK);
	if (dev->current_state != state && printk_ratelimit())
		dev_info(&dev->dev, "Refused to change power state, "
			"currently in D%d\n", dev->current_state);
572

573 574
	/*
	 * According to section 5.4.1 of the "PCI BUS POWER MANAGEMENT
575 576 577 578 579 580 581 582 583 584 585 586 587 588
	 * 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);

589
	if (dev->bus->self)
S
Shaohua Li 已提交
590 591
		pcie_aspm_pm_state_change(dev->bus->self);

L
Linus Torvalds 已提交
592 593 594
	return 0;
}

595 596 597 598
/**
 * pci_update_current_state - Read PCI power state of given device from its
 *                            PCI PM registers and cache it
 * @dev: PCI device to handle.
599
 * @state: State to cache in case the device doesn't have the PM capability
600
 */
601
void pci_update_current_state(struct pci_dev *dev, pci_power_t state)
602
{
603
	if (dev->pm_cap) {
604 605
		u16 pmcsr;

606 607 608 609 610 611 612 613 614 615
		/*
		 * 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;
		}
616
		pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
617
		dev->current_state = (pmcsr & PCI_PM_CTRL_STATE_MASK);
618 619
	} else {
		dev->current_state = state;
620 621 622
	}
}

623 624 625 626 627 628 629 630 631 632 633 634 635
/**
 * 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);
}

636 637 638 639 640 641 642 643 644 645 646 647 648
/**
 * 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);
649 650 651
		/* Fall back to PCI_D0 if native PM is not supported */
		if (!dev->pm_cap)
			dev->current_state = PCI_D0;
652 653 654
	} else {
		error = -ENODEV;
		/* Fall back to PCI_D0 if native PM is not supported */
655 656
		if (!dev->pm_cap)
			dev->current_state = PCI_D0;
657 658 659 660 661 662 663 664 665 666 667 668
	}

	return error;
}

/**
 * __pci_start_power_transition - Start power transition of a PCI device
 * @dev: PCI device to handle.
 * @state: State to put the device into.
 */
static void __pci_start_power_transition(struct pci_dev *dev, pci_power_t state)
{
669
	if (state == PCI_D0) {
670
		pci_platform_power_transition(dev, PCI_D0);
671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712
		/*
		 * 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);
713 714 715 716 717 718 719 720 721 722 723
}

/**
 * __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)
{
724 725
	int ret;

726
	if (state <= PCI_D0)
727 728 729 730 731 732
		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;
733 734 735
}
EXPORT_SYMBOL_GPL(__pci_complete_power_transition);

736 737 738 739 740
/**
 * 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.
 *
741
 * Transition a device to a new power state, using the platform firmware and/or
742 743 744 745 746 747 748 749 750 751 752
 * 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)
{
753
	int error;
754 755

	/* bound the state we're entering */
756 757
	if (state > PCI_D3cold)
		state = PCI_D3cold;
758 759 760 761 762 763 764 765 766 767
	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;

768 769 770 771
	/* Check if we're already there */
	if (dev->current_state == state)
		return 0;

772 773
	__pci_start_power_transition(dev, state);

774 775
	/* This device is quirked not to be put into D3, so
	   don't put it in D3 */
776
	if (state >= PCI_D3hot && (dev->dev_flags & PCI_DEV_FLAGS_NO_D3))
777
		return 0;
778

779 780 781 782 783 784
	/*
	 * 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);
785

786 787
	if (!__pci_complete_power_transition(dev, state))
		error = 0;
788 789 790 791 792 793
	/*
	 * When aspm_policy is "powersave" this call ensures
	 * that ASPM is configured.
	 */
	if (!error && dev->bus->self)
		pcie_aspm_powersave_config_link(dev->bus->self);
794 795 796 797

	return error;
}

L
Linus Torvalds 已提交
798 799 800 801 802 803 804 805 806 807 808 809
/**
 * 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)
{
810
	pci_power_t ret;
811

L
Linus Torvalds 已提交
812 813 814
	if (!pci_find_capability(dev, PCI_CAP_ID_PM))
		return PCI_D0;

815 816 817
	ret = platform_pci_choose_state(dev);
	if (ret != PCI_POWER_ERROR)
		return ret;
818 819 820 821 822

	switch (state.event) {
	case PM_EVENT_ON:
		return PCI_D0;
	case PM_EVENT_FREEZE:
823 824
	case PM_EVENT_PRETHAW:
		/* REVISIT both freeze and pre-thaw "should" use D0 */
825
	case PM_EVENT_SUSPEND:
826
	case PM_EVENT_HIBERNATE:
827
		return PCI_D3hot;
L
Linus Torvalds 已提交
828
	default:
829 830
		dev_info(&dev->dev, "unrecognized suspend event %d\n",
			 state.event);
L
Linus Torvalds 已提交
831 832 833 834 835 836 837
		BUG();
	}
	return PCI_D0;
}

EXPORT_SYMBOL(pci_choose_state);

838 839
#define PCI_EXP_SAVE_REGS	7

840

841 842 843 844 845
static struct pci_cap_saved_state *pci_find_saved_cap(
	struct pci_dev *pci_dev, char cap)
{
	struct pci_cap_saved_state *tmp;

846
	hlist_for_each_entry(tmp, &pci_dev->saved_cap_space, next) {
847 848 849 850 851 852
		if (tmp->cap.cap_nr == cap)
			return tmp;
	}
	return NULL;
}

853 854
static int pci_save_pcie_state(struct pci_dev *dev)
{
855
	int i = 0;
856 857 858
	struct pci_cap_saved_state *save_state;
	u16 *cap;

859
	if (!pci_is_pcie(dev))
860 861
		return 0;

862
	save_state = pci_find_saved_cap(dev, PCI_CAP_ID_EXP);
863
	if (!save_state) {
864
		dev_err(&dev->dev, "buffer not found in %s\n", __func__);
865 866
		return -ENOMEM;
	}
867

868 869 870 871 872 873 874 875
	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++]);
876

877 878 879 880 881
	return 0;
}

static void pci_restore_pcie_state(struct pci_dev *dev)
{
882
	int i = 0;
883 884 885 886
	struct pci_cap_saved_state *save_state;
	u16 *cap;

	save_state = pci_find_saved_cap(dev, PCI_CAP_ID_EXP);
887
	if (!save_state)
888 889
		return;

890 891 892 893 894 895 896 897
	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++]);
898 899
}

S
Stephen Hemminger 已提交
900 901 902

static int pci_save_pcix_state(struct pci_dev *dev)
{
903
	int pos;
S
Stephen Hemminger 已提交
904 905 906 907 908 909
	struct pci_cap_saved_state *save_state;

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

910
	save_state = pci_find_saved_cap(dev, PCI_CAP_ID_PCIX);
S
Stephen Hemminger 已提交
911
	if (!save_state) {
912
		dev_err(&dev->dev, "buffer not found in %s\n", __func__);
S
Stephen Hemminger 已提交
913 914 915
		return -ENOMEM;
	}

916 917
	pci_read_config_word(dev, pos + PCI_X_CMD,
			     (u16 *)save_state->cap.data);
918

S
Stephen Hemminger 已提交
919 920 921 922 923 924 925 926 927 928 929 930 931
	return 0;
}

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

	save_state = pci_find_saved_cap(dev, PCI_CAP_ID_PCIX);
	pos = pci_find_capability(dev, PCI_CAP_ID_PCIX);
	if (!save_state || pos <= 0)
		return;
932
	cap = (u16 *)&save_state->cap.data[0];
S
Stephen Hemminger 已提交
933 934 935 936 937

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


L
Linus Torvalds 已提交
938 939 940 941 942 943 944 945 946 947
/**
 * pci_save_state - save the PCI configuration space of a device before suspending
 * @dev: - PCI device that we're dealing with
 */
int
pci_save_state(struct pci_dev *dev)
{
	int i;
	/* XXX: 100% dword access ok here? */
	for (i = 0; i < 16; i++)
948
		pci_read_config_dword(dev, i * 4, &dev->saved_config_space[i]);
949
	dev->state_saved = true;
950 951
	if ((i = pci_save_pcie_state(dev)) != 0)
		return i;
S
Stephen Hemminger 已提交
952 953
	if ((i = pci_save_pcix_state(dev)) != 0)
		return i;
L
Linus Torvalds 已提交
954 955 956
	return 0;
}

957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980
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 (;;) {
		dev_dbg(&pdev->dev, "restoring config space at offset "
			"%#x (was %#x, writing %#x)\n", offset, val, saved_val);
		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);
	}
}

981 982
static void pci_restore_config_space_range(struct pci_dev *pdev,
					   int start, int end, int retry)
983 984 985 986 987 988 989 990 991
{
	int index;

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

992 993 994 995 996 997 998 999 1000 1001 1002 1003
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);
	}
}

L
Linus Torvalds 已提交
1004 1005 1006 1007
/** 
 * pci_restore_state - Restore the saved state of a PCI device
 * @dev: - PCI device that we're dealing with
 */
1008
void pci_restore_state(struct pci_dev *dev)
L
Linus Torvalds 已提交
1009
{
A
Alek Du 已提交
1010
	if (!dev->state_saved)
1011
		return;
1012

1013 1014
	/* PCI Express register must be restored first */
	pci_restore_pcie_state(dev);
1015
	pci_restore_ats_state(dev);
1016

1017
	pci_restore_config_space(dev);
1018

S
Stephen Hemminger 已提交
1019
	pci_restore_pcix_state(dev);
1020
	pci_restore_msi_state(dev);
Y
Yu Zhao 已提交
1021
	pci_restore_iov_state(dev);
1022

1023
	dev->state_saved = false;
L
Linus Torvalds 已提交
1024 1025
}

1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049
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
 *
 * Rerturn NULL if no state or error.
 */
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);

1050
	hlist_for_each_entry(tmp, &dev->saved_cap_space, next)
1051 1052 1053 1054 1055 1056 1057 1058 1059 1060
		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;
1061
	hlist_for_each_entry(tmp, &dev->saved_cap_space, next) {
1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122
		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()
 */
int pci_load_saved_state(struct pci_dev *dev, struct pci_saved_state *state)
{
	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;

		tmp = pci_find_saved_cap(dev, cap->cap_nr);
		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;
}
EXPORT_SYMBOL_GPL(pci_load_saved_state);

/**
 * 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);

1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138
static int do_pci_enable_device(struct pci_dev *dev, int bars)
{
	int err;

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

	return 0;
}

/**
1139
 * pci_reenable_device - Resume abandoned device
1140 1141 1142 1143 1144
 * @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.
 */
1145
int pci_reenable_device(struct pci_dev *dev)
1146
{
1147
	if (pci_is_enabled(dev))
1148 1149 1150 1151
		return do_pci_enable_device(dev, (1 << PCI_NUM_RESOURCES) - 1);
	return 0;
}

1152
static int pci_enable_device_flags(struct pci_dev *dev, unsigned long flags)
L
Linus Torvalds 已提交
1153 1154
{
	int err;
1155
	int i, bars = 0;
L
Linus Torvalds 已提交
1156

1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168
	/*
	 * 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);
	}

1169
	if (atomic_inc_return(&dev->enable_cnt) > 1)
1170 1171
		return 0;		/* already enabled */

1172 1173 1174 1175 1176
	/* 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++)
1177 1178 1179
		if (dev->resource[i].flags & flags)
			bars |= (1 << i);

1180
	err = do_pci_enable_device(dev, bars);
1181
	if (err < 0)
1182
		atomic_dec(&dev->enable_cnt);
1183
	return err;
L
Linus Torvalds 已提交
1184 1185
}

1186 1187 1188 1189 1190 1191 1192 1193 1194 1195
/**
 * 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)
{
1196
	return pci_enable_device_flags(dev, IORESOURCE_IO);
1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208
}

/**
 * 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)
{
1209
	return pci_enable_device_flags(dev, IORESOURCE_MEM);
1210 1211
}

1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224
/**
 * 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)
{
1225
	return pci_enable_device_flags(dev, IORESOURCE_MEM | IORESOURCE_IO);
1226 1227
}

T
Tejun Heo 已提交
1228 1229 1230 1231 1232 1233 1234
/*
 * 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 {
1235 1236
	unsigned int enabled:1;
	unsigned int pinned:1;
T
Tejun Heo 已提交
1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259
	unsigned int orig_intx:1;
	unsigned int restore_intx:1;
	u32 region_mask;
};

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

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

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

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

1260
	if (this->enabled && !this->pinned)
T
Tejun Heo 已提交
1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298
		pci_disable_device(dev);
}

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

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

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

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

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

	dr = get_pci_dr(pdev);
	if (unlikely(!dr))
		return -ENOMEM;
1299 1300
	if (dr->enabled)
		return 0;
T
Tejun Heo 已提交
1301 1302 1303 1304

	rc = pci_enable_device(pdev);
	if (!rc) {
		pdev->is_managed = 1;
1305
		dr->enabled = 1;
T
Tejun Heo 已提交
1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322
	}
	return rc;
}

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

	dr = find_pci_dr(pdev);
1323
	WARN_ON(!dr || !dr->enabled);
T
Tejun Heo 已提交
1324
	if (dr)
1325
		dr->pinned = 1;
T
Tejun Heo 已提交
1326 1327
}

M
Matthew Garrett 已提交
1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340
/*
 * 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.
 */
int __weak pcibios_add_device (struct pci_dev *dev)
{
	return 0;
}

L
Linus Torvalds 已提交
1341 1342 1343 1344 1345 1346 1347 1348
/**
 * 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
Bjorn Helgaas 已提交
1349
void __weak pcibios_disable_device (struct pci_dev *dev) {}
L
Linus Torvalds 已提交
1350

1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372
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)
{
1373
	if (pci_is_enabled(dev))
1374 1375 1376
		do_pci_disable_device(dev);
}

L
Linus Torvalds 已提交
1377 1378 1379 1380 1381 1382
/**
 * 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.
1383 1384
 *
 * Note we don't actually disable the device until all callers of
1385
 * pci_enable_device() have called pci_disable_device().
L
Linus Torvalds 已提交
1386 1387 1388 1389
 */
void
pci_disable_device(struct pci_dev *dev)
{
T
Tejun Heo 已提交
1390
	struct pci_devres *dr;
1391

T
Tejun Heo 已提交
1392 1393
	dr = find_pci_dr(dev);
	if (dr)
1394
		dr->enabled = 0;
T
Tejun Heo 已提交
1395

1396 1397 1398
	dev_WARN_ONCE(&dev->dev, atomic_read(&dev->enable_cnt) <= 0,
		      "disabling already-disabled device");

1399
	if (atomic_dec_return(&dev->enable_cnt) != 0)
1400 1401
		return;

1402
	do_pci_disable_device(dev);
L
Linus Torvalds 已提交
1403

1404
	dev->is_busmaster = 0;
L
Linus Torvalds 已提交
1405 1406
}

B
Brian King 已提交
1407 1408
/**
 * pcibios_set_pcie_reset_state - set reset state for device dev
1409
 * @dev: the PCIe device reset
B
Brian King 已提交
1410 1411 1412
 * @state: Reset state to enter into
 *
 *
1413
 * Sets the PCIe reset state for the device. This is the default
B
Brian King 已提交
1414 1415
 * implementation. Architecture implementations can override this.
 */
B
Bjorn Helgaas 已提交
1416 1417
int __weak pcibios_set_pcie_reset_state(struct pci_dev *dev,
					enum pcie_reset_state state)
B
Brian King 已提交
1418 1419 1420 1421 1422 1423
{
	return -EINVAL;
}

/**
 * pci_set_pcie_reset_state - set reset state for device dev
1424
 * @dev: the PCIe device reset
B
Brian King 已提交
1425 1426 1427 1428 1429 1430 1431 1432 1433 1434
 * @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);
}

1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469
/**
 * 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;
}

1470 1471 1472
/**
 * pci_pme_wakeup - Wake up a PCI device if its PME Status bit is set.
 * @dev: Device to handle.
1473
 * @pme_poll_reset: Whether or not to reset the device's pme_poll flag.
1474 1475 1476 1477
 *
 * Check if @dev has generated PME and queue a resume request for it in that
 * case.
 */
1478
static int pci_pme_wakeup(struct pci_dev *dev, void *pme_poll_reset)
1479
{
1480 1481 1482
	if (pme_poll_reset && dev->pme_poll)
		dev->pme_poll = false;

1483 1484
	if (pci_check_pme_status(dev)) {
		pci_wakeup_event(dev);
1485
		pm_request_resume(&dev->dev);
1486
	}
1487 1488 1489 1490 1491 1492 1493 1494 1495 1496
	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)
1497
		pci_walk_bus(bus, pci_pme_wakeup, (void *)true);
1498 1499
}

1500 1501
/**
 * pci_wakeup - Wake up a PCI device
1502
 * @pci_dev: Device to handle.
1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521
 * @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.
 */
void pci_wakeup_bus(struct pci_bus *bus)
{
	if (bus)
		pci_walk_bus(bus, pci_wakeup, NULL);
}

1522 1523 1524 1525 1526
/**
 * 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#.
 */
1527
bool pci_pme_capable(struct pci_dev *dev, pci_power_t state)
1528
{
1529
	if (!dev->pm_cap)
1530 1531
		return false;

1532
	return !!(dev->pme_support & (1 << state));
1533 1534
}

1535 1536
static void pci_pme_list_scan(struct work_struct *work)
{
1537
	struct pci_pme_device *pme_dev, *n;
1538 1539 1540

	mutex_lock(&pci_pme_list_mutex);
	if (!list_empty(&pci_pme_list)) {
1541 1542
		list_for_each_entry_safe(pme_dev, n, &pci_pme_list, list) {
			if (pme_dev->dev->pme_poll) {
1543 1544 1545 1546 1547 1548 1549 1550 1551 1552
				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;
1553 1554 1555 1556 1557 1558 1559 1560 1561
				pci_pme_wakeup(pme_dev->dev, NULL);
			} else {
				list_del(&pme_dev->list);
				kfree(pme_dev);
			}
		}
		if (!list_empty(&pci_pme_list))
			schedule_delayed_work(&pci_pme_work,
					      msecs_to_jiffies(PME_TIMEOUT));
1562 1563 1564 1565
	}
	mutex_unlock(&pci_pme_list_mutex);
}

1566 1567 1568 1569 1570 1571 1572 1573
/**
 * 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'.
 */
1574
void pci_pme_active(struct pci_dev *dev, bool enable)
1575 1576 1577
{
	u16 pmcsr;

1578
	if (!dev->pm_cap)
1579 1580
		return;

1581
	pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
1582 1583 1584 1585 1586
	/* 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;

1587
	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, pmcsr);
1588

1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607
	/*
	 * 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.
	 */
1608

1609
	if (dev->pme_poll) {
1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636
		struct pci_pme_device *pme_dev;
		if (enable) {
			pme_dev = kmalloc(sizeof(struct pci_pme_device),
					  GFP_KERNEL);
			if (!pme_dev)
				goto out;
			pme_dev->dev = dev;
			mutex_lock(&pci_pme_list_mutex);
			list_add(&pme_dev->list, &pci_pme_list);
			if (list_is_singular(&pci_pme_list))
				schedule_delayed_work(&pci_pme_work,
						      msecs_to_jiffies(PME_TIMEOUT));
			mutex_unlock(&pci_pme_list_mutex);
		} else {
			mutex_lock(&pci_pme_list_mutex);
			list_for_each_entry(pme_dev, &pci_pme_list, list) {
				if (pme_dev->dev == dev) {
					list_del(&pme_dev->list);
					kfree(pme_dev);
					break;
				}
			}
			mutex_unlock(&pci_pme_list_mutex);
		}
	}

out:
1637
	dev_dbg(&dev->dev, "PME# %s\n", enable ? "enabled" : "disabled");
1638 1639
}

L
Linus Torvalds 已提交
1640
/**
1641
 * __pci_enable_wake - enable PCI device as wakeup event source
1642 1643
 * @dev: PCI device affected
 * @state: PCI state from which device will issue wakeup events
1644
 * @runtime: True if the events are to be generated at run time
1645 1646 1647 1648 1649 1650 1651
 * @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)
1652
 * always require such platform hooks.
1653
 *
1654 1655 1656 1657 1658
 * 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
1659
 */
1660 1661
int __pci_enable_wake(struct pci_dev *dev, pci_power_t state,
		      bool runtime, bool enable)
L
Linus Torvalds 已提交
1662
{
1663
	int ret = 0;
1664

1665
	if (enable && !runtime && !device_may_wakeup(&dev->dev))
1666
		return -EINVAL;
L
Linus Torvalds 已提交
1667

1668 1669 1670 1671
	/* Don't do the same thing twice in a row for one device. */
	if (!!enable == !!dev->wakeup_prepared)
		return 0;

1672 1673 1674 1675
	/*
	 * 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.
1676
	 */
L
Linus Torvalds 已提交
1677

1678 1679
	if (enable) {
		int error;
L
Linus Torvalds 已提交
1680

1681 1682 1683 1684
		if (pci_pme_capable(dev, state))
			pci_pme_active(dev, true);
		else
			ret = 1;
1685 1686
		error = runtime ? platform_pci_run_wake(dev, true) :
					platform_pci_sleep_wake(dev, true);
1687 1688
		if (ret)
			ret = error;
1689 1690
		if (!ret)
			dev->wakeup_prepared = true;
1691
	} else {
1692 1693 1694 1695
		if (runtime)
			platform_pci_run_wake(dev, false);
		else
			platform_pci_sleep_wake(dev, false);
1696
		pci_pme_active(dev, false);
1697
		dev->wakeup_prepared = false;
1698
	}
L
Linus Torvalds 已提交
1699

1700
	return ret;
1701
}
1702
EXPORT_SYMBOL(__pci_enable_wake);
L
Linus Torvalds 已提交
1703

1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724
/**
 * 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);
}

1725
/**
J
Jesse Barnes 已提交
1726 1727 1728 1729 1730 1731
 * 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.
1732
 */
1733
pci_power_t pci_target_state(struct pci_dev *dev)
1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754
{
	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;
		}
1755 1756
	} else if (!dev->pm_cap) {
		target_state = PCI_D0;
1757 1758 1759 1760 1761 1762
	} 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#.
		 */
1763 1764 1765 1766
		if (dev->pme_support) {
			while (target_state
			      && !(dev->pme_support & (1 << target_state)))
				target_state--;
1767 1768 1769
		}
	}

1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788
	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;

1789 1790 1791 1792
	/* D3cold during system suspend/hibernate is not supported */
	if (target_state > PCI_D3hot)
		target_state = PCI_D3hot;

1793
	pci_enable_wake(dev, target_state, device_may_wakeup(&dev->dev));
1794

1795 1796 1797 1798 1799 1800 1801 1802 1803
	error = pci_set_power_state(dev, target_state);

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

	return error;
}

/**
R
Randy Dunlap 已提交
1804
 * pci_back_from_sleep - turn PCI device on during system-wide transition into working state
1805 1806
 * @dev: Device to handle.
 *
T
Thomas Weber 已提交
1807
 * Disable device's system wake-up capability and put it into D0.
1808 1809 1810 1811 1812 1813 1814
 */
int pci_back_from_sleep(struct pci_dev *dev)
{
	pci_enable_wake(dev, PCI_D0, false);
	return pci_set_power_state(dev, PCI_D0);
}

1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829
/**
 * 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;

1830 1831
	dev->runtime_d3cold = target_state == PCI_D3cold;

1832 1833 1834 1835
	__pci_enable_wake(dev, target_state, true, pci_dev_run_wake(dev));

	error = pci_set_power_state(dev, target_state);

1836
	if (error) {
1837
		__pci_enable_wake(dev, target_state, true, false);
1838 1839
		dev->runtime_d3cold = false;
	}
1840 1841 1842 1843

	return error;
}

1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878
/**
 * pci_dev_run_wake - Check if device can generate run-time wake-up events.
 * @dev: Device to check.
 *
 * Return true if the device itself is cabable of generating wake-up events
 * (through the platform or using the native PCIe PME) or if the device supports
 * PME and one of its upstream bridges can generate wake-up events.
 */
bool pci_dev_run_wake(struct pci_dev *dev)
{
	struct pci_bus *bus = dev->bus;

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

	if (!dev->pme_support)
		return false;

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

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

		bus = bus->parent;
	}

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

	return false;
}
EXPORT_SYMBOL_GPL(pci_dev_run_wake);

1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910
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);
}

1911 1912 1913 1914 1915 1916 1917 1918
/**
 * 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
1919

1920
	pm_runtime_forbid(&dev->dev);
1921 1922
	pm_runtime_set_active(&dev->dev);
	pm_runtime_enable(&dev->dev);
1923
	device_enable_async_suspend(&dev->dev);
1924
	dev->wakeup_prepared = false;
1925

1926 1927
	dev->pm_cap = 0;

1928 1929 1930
	/* find PCI PM capability in list */
	pm = pci_find_capability(dev, PCI_CAP_ID_PM);
	if (!pm)
1931
		return;
1932 1933
	/* Check device's ability to generate PME# */
	pci_read_config_word(dev, pm + PCI_PM_PMC, &pmc);
1934

1935 1936 1937
	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);
1938
		return;
1939 1940
	}

1941
	dev->pm_cap = pm;
1942
	dev->d3_delay = PCI_PM_D3_WAIT;
1943
	dev->d3cold_delay = PCI_PM_D3COLD_WAIT;
1944
	dev->d3cold_allowed = true;
1945 1946 1947 1948

	dev->d1_support = false;
	dev->d2_support = false;
	if (!pci_no_d1d2(dev)) {
B
Bjorn Helgaas 已提交
1949
		if (pmc & PCI_PM_CAP_D1)
1950
			dev->d1_support = true;
B
Bjorn Helgaas 已提交
1951
		if (pmc & PCI_PM_CAP_D2)
1952
			dev->d2_support = true;
B
Bjorn Helgaas 已提交
1953 1954 1955

		if (dev->d1_support || dev->d2_support)
			dev_printk(KERN_DEBUG, &dev->dev, "supports%s%s\n",
1956 1957
				   dev->d1_support ? " D1" : "",
				   dev->d2_support ? " D2" : "");
1958 1959 1960 1961
	}

	pmc &= PCI_PM_CAP_PME_MASK;
	if (pmc) {
B
Bjorn Helgaas 已提交
1962 1963
		dev_printk(KERN_DEBUG, &dev->dev,
			 "PME# supported from%s%s%s%s%s\n",
B
Bjorn Helgaas 已提交
1964 1965 1966 1967 1968
			 (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" : "");
1969
		dev->pme_support = pmc >> PCI_PM_CAP_PME_SHIFT;
1970
		dev->pme_poll = true;
1971 1972 1973 1974 1975 1976
		/*
		 * 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 */
1977 1978 1979
		pci_pme_active(dev, false);
	} else {
		dev->pme_support = 0;
1980
	}
L
Linus Torvalds 已提交
1981 1982
}

1983 1984 1985 1986 1987 1988
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);
}

1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
/**
 * pci_add_save_buffer - allocate buffer for saving given capability registers
 * @dev: the PCI device
 * @cap: the capability to allocate the buffer for
 * @size: requested size of the buffer
 */
static int pci_add_cap_save_buffer(
	struct pci_dev *dev, char cap, unsigned int size)
{
	int pos;
	struct pci_cap_saved_state *save_state;

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

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

2009 2010
	save_state->cap.cap_nr = cap;
	save_state->cap.size = size;
2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023
	pci_add_saved_cap(dev, save_state);

	return 0;
}

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

2024 2025
	error = pci_add_cap_save_buffer(dev, PCI_CAP_ID_EXP,
					PCI_EXP_SAVE_REGS * sizeof(u16));
2026 2027 2028 2029 2030 2031 2032 2033 2034 2035
	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");
}

2036 2037 2038
void pci_free_cap_save_buffers(struct pci_dev *dev)
{
	struct pci_cap_saved_state *tmp;
2039
	struct hlist_node *n;
2040

2041
	hlist_for_each_entry_safe(tmp, n, &dev->saved_cap_space, next)
2042 2043 2044
		kfree(tmp);
}

Y
Yu Zhao 已提交
2045
/**
2046
 * pci_configure_ari - enable or disable ARI forwarding
Y
Yu Zhao 已提交
2047
 * @dev: the PCI device
2048 2049 2050
 *
 * 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
2051
 */
2052
void pci_configure_ari(struct pci_dev *dev)
Y
Yu Zhao 已提交
2053 2054
{
	u32 cap;
2055
	struct pci_dev *bridge;
Y
Yu Zhao 已提交
2056

2057
	if (pcie_ari_disabled || !pci_is_pcie(dev) || dev->devfn)
Y
Yu Zhao 已提交
2058 2059
		return;

2060
	bridge = dev->bus->self;
2061
	if (!bridge)
2062 2063
		return;

2064
	pcie_capability_read_dword(bridge, PCI_EXP_DEVCAP2, &cap);
Y
Yu Zhao 已提交
2065 2066 2067
	if (!(cap & PCI_EXP_DEVCAP2_ARI))
		return;

2068 2069 2070 2071 2072 2073 2074 2075 2076
	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
2077 2078
}

2079
/**
2080
 * pci_enable_ido - enable ID-based Ordering on a device
2081 2082 2083 2084 2085 2086 2087 2088 2089
 * @dev: the PCI device
 * @type: which types of IDO to enable
 *
 * Enable ID-based ordering on @dev.  @type can contain the bits
 * %PCI_EXP_IDO_REQUEST and/or %PCI_EXP_IDO_COMPLETION to indicate
 * which types of transactions are allowed to be re-ordered.
 */
void pci_enable_ido(struct pci_dev *dev, unsigned long type)
{
2090
	u16 ctrl = 0;
2091 2092 2093 2094 2095

	if (type & PCI_EXP_IDO_REQUEST)
		ctrl |= PCI_EXP_IDO_REQ_EN;
	if (type & PCI_EXP_IDO_COMPLETION)
		ctrl |= PCI_EXP_IDO_CMP_EN;
2096 2097
	if (ctrl)
		pcie_capability_set_word(dev, PCI_EXP_DEVCTL2, ctrl);
2098 2099 2100 2101 2102 2103 2104 2105 2106 2107
}
EXPORT_SYMBOL(pci_enable_ido);

/**
 * pci_disable_ido - disable ID-based ordering on a device
 * @dev: the PCI device
 * @type: which types of IDO to disable
 */
void pci_disable_ido(struct pci_dev *dev, unsigned long type)
{
2108
	u16 ctrl = 0;
2109 2110

	if (type & PCI_EXP_IDO_REQUEST)
2111
		ctrl |= PCI_EXP_IDO_REQ_EN;
2112
	if (type & PCI_EXP_IDO_COMPLETION)
2113 2114 2115
		ctrl |= PCI_EXP_IDO_CMP_EN;
	if (ctrl)
		pcie_capability_clear_word(dev, PCI_EXP_DEVCTL2, ctrl);
2116 2117 2118
}
EXPORT_SYMBOL(pci_disable_ido);

2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143
/**
 * pci_enable_obff - enable optimized buffer flush/fill
 * @dev: PCI device
 * @type: type of signaling to use
 *
 * Try to enable @type OBFF signaling on @dev.  It will try using WAKE#
 * signaling if possible, falling back to message signaling only if
 * WAKE# isn't supported.  @type should indicate whether the PCIe link
 * be brought out of L0s or L1 to send the message.  It should be either
 * %PCI_EXP_OBFF_SIGNAL_ALWAYS or %PCI_OBFF_SIGNAL_L0.
 *
 * If your device can benefit from receiving all messages, even at the
 * power cost of bringing the link back up from a low power state, use
 * %PCI_EXP_OBFF_SIGNAL_ALWAYS.  Otherwise, use %PCI_OBFF_SIGNAL_L0 (the
 * preferred type).
 *
 * RETURNS:
 * Zero on success, appropriate error number on failure.
 */
int pci_enable_obff(struct pci_dev *dev, enum pci_obff_signal_type type)
{
	u32 cap;
	u16 ctrl;
	int ret;

2144
	pcie_capability_read_dword(dev, PCI_EXP_DEVCAP2, &cap);
2145 2146 2147 2148
	if (!(cap & PCI_EXP_OBFF_MASK))
		return -ENOTSUPP; /* no OBFF support at all */

	/* Make sure the topology supports OBFF as well */
2149
	if (dev->bus->self) {
2150 2151 2152 2153 2154
		ret = pci_enable_obff(dev->bus->self, type);
		if (ret)
			return ret;
	}

2155
	pcie_capability_read_word(dev, PCI_EXP_DEVCTL2, &ctrl);
2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172
	if (cap & PCI_EXP_OBFF_WAKE)
		ctrl |= PCI_EXP_OBFF_WAKE_EN;
	else {
		switch (type) {
		case PCI_EXP_OBFF_SIGNAL_L0:
			if (!(ctrl & PCI_EXP_OBFF_WAKE_EN))
				ctrl |= PCI_EXP_OBFF_MSGA_EN;
			break;
		case PCI_EXP_OBFF_SIGNAL_ALWAYS:
			ctrl &= ~PCI_EXP_OBFF_WAKE_EN;
			ctrl |= PCI_EXP_OBFF_MSGB_EN;
			break;
		default:
			WARN(1, "bad OBFF signal type\n");
			return -ENOTSUPP;
		}
	}
2173
	pcie_capability_write_word(dev, PCI_EXP_DEVCTL2, ctrl);
2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186

	return 0;
}
EXPORT_SYMBOL(pci_enable_obff);

/**
 * pci_disable_obff - disable optimized buffer flush/fill
 * @dev: PCI device
 *
 * Disable OBFF on @dev.
 */
void pci_disable_obff(struct pci_dev *dev)
{
2187
	pcie_capability_clear_word(dev, PCI_EXP_DEVCTL2, PCI_EXP_OBFF_WAKE_EN);
2188 2189 2190
}
EXPORT_SYMBOL(pci_disable_obff);

2191 2192 2193 2194 2195 2196 2197
/**
 * pci_ltr_supported - check whether a device supports LTR
 * @dev: PCI device
 *
 * RETURNS:
 * True if @dev supports latency tolerance reporting, false otherwise.
 */
2198
static bool pci_ltr_supported(struct pci_dev *dev)
2199 2200 2201
{
	u32 cap;

2202
	pcie_capability_read_dword(dev, PCI_EXP_DEVCAP2, &cap);
2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224

	return cap & PCI_EXP_DEVCAP2_LTR;
}

/**
 * pci_enable_ltr - enable latency tolerance reporting
 * @dev: PCI device
 *
 * Enable LTR on @dev if possible, which means enabling it first on
 * upstream ports.
 *
 * RETURNS:
 * Zero on success, errno on failure.
 */
int pci_enable_ltr(struct pci_dev *dev)
{
	int ret;

	/* Only primary function can enable/disable LTR */
	if (PCI_FUNC(dev->devfn) != 0)
		return -EINVAL;

2225 2226 2227
	if (!pci_ltr_supported(dev))
		return -ENOTSUPP;

2228
	/* Enable upstream ports first */
2229
	if (dev->bus->self) {
2230 2231 2232 2233 2234
		ret = pci_enable_ltr(dev->bus->self);
		if (ret)
			return ret;
	}

2235
	return pcie_capability_set_word(dev, PCI_EXP_DEVCTL2, PCI_EXP_LTR_EN);
2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248
}
EXPORT_SYMBOL(pci_enable_ltr);

/**
 * pci_disable_ltr - disable latency tolerance reporting
 * @dev: PCI device
 */
void pci_disable_ltr(struct pci_dev *dev)
{
	/* Only primary function can enable/disable LTR */
	if (PCI_FUNC(dev->devfn) != 0)
		return;

2249 2250 2251 2252
	if (!pci_ltr_supported(dev))
		return;

	pcie_capability_clear_word(dev, PCI_EXP_DEVCTL2, PCI_EXP_LTR_EN);
2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311
}
EXPORT_SYMBOL(pci_disable_ltr);

static int __pci_ltr_scale(int *val)
{
	int scale = 0;

	while (*val > 1023) {
		*val = (*val + 31) / 32;
		scale++;
	}
	return scale;
}

/**
 * pci_set_ltr - set LTR latency values
 * @dev: PCI device
 * @snoop_lat_ns: snoop latency in nanoseconds
 * @nosnoop_lat_ns: nosnoop latency in nanoseconds
 *
 * Figure out the scale and set the LTR values accordingly.
 */
int pci_set_ltr(struct pci_dev *dev, int snoop_lat_ns, int nosnoop_lat_ns)
{
	int pos, ret, snoop_scale, nosnoop_scale;
	u16 val;

	if (!pci_ltr_supported(dev))
		return -ENOTSUPP;

	snoop_scale = __pci_ltr_scale(&snoop_lat_ns);
	nosnoop_scale = __pci_ltr_scale(&nosnoop_lat_ns);

	if (snoop_lat_ns > PCI_LTR_VALUE_MASK ||
	    nosnoop_lat_ns > PCI_LTR_VALUE_MASK)
		return -EINVAL;

	if ((snoop_scale > (PCI_LTR_SCALE_MASK >> PCI_LTR_SCALE_SHIFT)) ||
	    (nosnoop_scale > (PCI_LTR_SCALE_MASK >> PCI_LTR_SCALE_SHIFT)))
		return -EINVAL;

	pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_LTR);
	if (!pos)
		return -ENOTSUPP;

	val = (snoop_scale << PCI_LTR_SCALE_SHIFT) | snoop_lat_ns;
	ret = pci_write_config_word(dev, pos + PCI_LTR_MAX_SNOOP_LAT, val);
	if (ret != 4)
		return -EIO;

	val = (nosnoop_scale << PCI_LTR_SCALE_SHIFT) | nosnoop_lat_ns;
	ret = pci_write_config_word(dev, pos + PCI_LTR_MAX_NOSNOOP_LAT, val);
	if (ret != 4)
		return -EIO;

	return 0;
}
EXPORT_SYMBOL(pci_set_ltr);

C
Chris Wright 已提交
2312 2313 2314 2315 2316 2317 2318 2319 2320 2321
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;
}

2322 2323 2324 2325 2326 2327 2328 2329 2330 2331
/**
 * pci_enable_acs - enable ACS if hardware support it
 * @dev: the PCI device
 */
void pci_enable_acs(struct pci_dev *dev)
{
	int pos;
	u16 cap;
	u16 ctrl;

C
Chris Wright 已提交
2332 2333 2334
	if (!pci_acs_enable)
		return;

2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356
	pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ACS);
	if (!pos)
		return;

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

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

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

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

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

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

2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381
/**
 * 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.
 */
bool pci_acs_enabled(struct pci_dev *pdev, u16 acs_flags)
{
	int pos, ret;
	u16 ctrl;

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

	if (!pci_is_pcie(pdev))
		return false;

	/* Filter out flags not applicable to multifunction */
	if (pdev->multifunction)
		acs_flags &= (PCI_ACS_RR | PCI_ACS_CR |
			      PCI_ACS_EC | PCI_ACS_DT);

2382 2383
	if (pci_pcie_type(pdev) == PCI_EXP_TYPE_DOWNSTREAM ||
	    pci_pcie_type(pdev) == PCI_EXP_TYPE_ROOT_PORT ||
2384 2385 2386 2387 2388 2389 2390 2391 2392 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 2421 2422 2423 2424 2425
	    pdev->multifunction) {
		pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_ACS);
		if (!pos)
			return false;

		pci_read_config_word(pdev, pos + PCI_ACS_CTRL, &ctrl);
		if ((ctrl & acs_flags) != acs_flags)
			return false;
	}

	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;
}

2426 2427 2428 2429 2430 2431 2432
/**
 * pci_swizzle_interrupt_pin - swizzle INTx for device behind bridge
 * @dev: the PCI device
 * @pin: the INTx pin (1=INTA, 2=INTB, 3=INTD, 4=INTD)
 *
 * Perform INTx swizzling for a device behind one level of bridge.  This is
 * required by section 9.1 of the PCI-to-PCI bridge specification for devices
2433 2434 2435
 * 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)
2436
 */
2437
u8 pci_swizzle_interrupt_pin(const struct pci_dev *dev, u8 pin)
2438
{
2439 2440 2441 2442 2443 2444 2445 2446
	int slot;

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

	return (((pin - 1) + slot) % 4) + 1;
2447 2448
}

L
Linus Torvalds 已提交
2449 2450 2451 2452 2453
int
pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge)
{
	u8 pin;

2454
	pin = dev->pin;
L
Linus Torvalds 已提交
2455 2456
	if (!pin)
		return -1;
2457

2458
	while (!pci_is_root_bus(dev->bus)) {
2459
		pin = pci_swizzle_interrupt_pin(dev, pin);
L
Linus Torvalds 已提交
2460 2461 2462 2463 2464 2465
		dev = dev->bus->self;
	}
	*bridge = dev;
	return pin;
}

2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477
/**
 * 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;

2478
	while (!pci_is_root_bus(dev->bus)) {
2479 2480 2481 2482 2483 2484 2485
		pin = pci_swizzle_interrupt_pin(dev, pin);
		dev = dev->bus->self;
	}
	*pinp = pin;
	return PCI_SLOT(dev->devfn);
}

L
Linus Torvalds 已提交
2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496
/**
 *	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
2497 2498
	struct pci_devres *dr;

L
Linus Torvalds 已提交
2499 2500 2501 2502 2503 2504 2505 2506
	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
2507 2508 2509 2510

	dr = find_pci_dr(pdev);
	if (dr)
		dr->region_mask &= ~(1 << bar);
L
Linus Torvalds 已提交
2511 2512 2513
}

/**
2514
 *	__pci_request_region - Reserved PCI I/O and memory resource
L
Linus Torvalds 已提交
2515 2516 2517
 *	@pdev: PCI device whose resources are to be reserved
 *	@bar: BAR to be reserved
 *	@res_name: Name to be associated with resource.
2518
 *	@exclusive: whether the region access is exclusive or not
L
Linus Torvalds 已提交
2519 2520 2521 2522 2523 2524
 *
 *	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.
 *
2525 2526 2527 2528
 *	If @exclusive is set, then the region is marked so that userspace
 *	is explicitly not allowed to map the resource via /dev/mem or
 * 	sysfs MMIO access.
 *
L
Linus Torvalds 已提交
2529 2530 2531
 *	Returns 0 on success, or %EBUSY on error.  A warning
 *	message is also printed on failure.
 */
2532 2533
static int __pci_request_region(struct pci_dev *pdev, int bar, const char *res_name,
									int exclusive)
L
Linus Torvalds 已提交
2534
{
T
Tejun Heo 已提交
2535 2536
	struct pci_devres *dr;

L
Linus Torvalds 已提交
2537 2538 2539 2540 2541 2542 2543 2544 2545
	if (pci_resource_len(pdev, bar) == 0)
		return 0;
		
	if (pci_resource_flags(pdev, bar) & IORESOURCE_IO) {
		if (!request_region(pci_resource_start(pdev, bar),
			    pci_resource_len(pdev, bar), res_name))
			goto err_out;
	}
	else if (pci_resource_flags(pdev, bar) & IORESOURCE_MEM) {
2546 2547 2548
		if (!__request_mem_region(pci_resource_start(pdev, bar),
					pci_resource_len(pdev, bar), res_name,
					exclusive))
L
Linus Torvalds 已提交
2549 2550
			goto err_out;
	}
T
Tejun Heo 已提交
2551 2552 2553 2554 2555

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

L
Linus Torvalds 已提交
2556 2557 2558
	return 0;

err_out:
2559
	dev_warn(&pdev->dev, "BAR %d: can't reserve %pR\n", bar,
2560
		 &pdev->resource[bar]);
L
Linus Torvalds 已提交
2561 2562 2563
	return -EBUSY;
}

2564
/**
2565
 *	pci_request_region - Reserve PCI I/O and memory resource
2566 2567
 *	@pdev: PCI device whose resources are to be reserved
 *	@bar: BAR to be reserved
2568
 *	@res_name: Name to be associated with resource
2569
 *
2570
 *	Mark the PCI region associated with PCI device @pdev BAR @bar as
2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604
 *	being reserved by owner @res_name.  Do not access any
 *	address inside the PCI regions unless this call returns
 *	successfully.
 *
 *	Returns 0 on success, or %EBUSY on error.  A warning
 *	message is also printed on failure.
 */
int pci_request_region(struct pci_dev *pdev, int bar, const char *res_name)
{
	return __pci_request_region(pdev, bar, res_name, 0);
}

/**
 *	pci_request_region_exclusive - Reserved PCI I/O and memory resource
 *	@pdev: PCI device whose resources are to be reserved
 *	@bar: BAR to be reserved
 *	@res_name: Name to be associated with resource.
 *
 *	Mark the PCI region associated with PCI device @pdev BR @bar as
 *	being reserved by owner @res_name.  Do not access any
 *	address inside the PCI regions unless this call returns
 *	successfully.
 *
 *	Returns 0 on success, or %EBUSY on error.  A warning
 *	message is also printed on failure.
 *
 *	The key difference that _exclusive makes it that userspace is
 *	explicitly not allowed to map the resource via /dev/mem or
 * 	sysfs.
 */
int pci_request_region_exclusive(struct pci_dev *pdev, int bar, const char *res_name)
{
	return __pci_request_region(pdev, bar, res_name, IORESOURCE_EXCLUSIVE);
}
2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621
/**
 * 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);
}

2622 2623
int __pci_request_selected_regions(struct pci_dev *pdev, int bars,
				 const char *res_name, int excl)
2624 2625 2626 2627 2628
{
	int i;

	for (i = 0; i < 6; i++)
		if (bars & (1 << i))
2629
			if (__pci_request_region(pdev, i, res_name, excl))
2630 2631 2632 2633 2634 2635 2636 2637 2638 2639
				goto err_out;
	return 0;

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

	return -EBUSY;
}
L
Linus Torvalds 已提交
2640

2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660

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

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

L
Linus Torvalds 已提交
2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671
/**
 *	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)
{
2672
	pci_release_selected_regions(pdev, (1 << 6) - 1);
L
Linus Torvalds 已提交
2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687
}

/**
 *	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.
 */
2688
int pci_request_regions(struct pci_dev *pdev, const char *res_name)
L
Linus Torvalds 已提交
2689
{
2690
	return pci_request_selected_regions(pdev, ((1 << 6) - 1), res_name);
L
Linus Torvalds 已提交
2691 2692
}

2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714
/**
 *	pci_request_regions_exclusive - Reserved PCI I/O and memory resources
 *	@pdev: PCI device whose resources are to be reserved
 *	@res_name: Name to be associated with resource.
 *
 *	Mark all PCI regions associated with PCI device @pdev as
 *	being reserved by owner @res_name.  Do not access any
 *	address inside the PCI regions unless this call returns
 *	successfully.
 *
 *	pci_request_regions_exclusive() will mark the region so that
 * 	/dev/mem and the sysfs MMIO access will not be allowed.
 *
 *	Returns 0 on success, or %EBUSY on error.  A warning
 *	message is also printed on failure.
 */
int pci_request_regions_exclusive(struct pci_dev *pdev, const char *res_name)
{
	return pci_request_selected_regions_exclusive(pdev,
					((1 << 6) - 1), res_name);
}

2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730
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;
}
2731

2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743
/**
 * 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;
}

2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755
/**
 * 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;

2756 2757 2758 2759
	/* The latency timer doesn't apply to PCIe (either Type 0 or Type 1) */
	if (pci_is_pcie(dev))
		return;

2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770
	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;
	dev_printk(KERN_DEBUG, &dev->dev, "setting latency timer to %d\n", lat);
	pci_write_config_byte(dev, PCI_LATENCY_TIMER, lat);
}

L
Linus Torvalds 已提交
2771 2772 2773 2774 2775 2776 2777
/**
 * 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.
 */
2778
void pci_set_master(struct pci_dev *dev)
L
Linus Torvalds 已提交
2779
{
2780
	__pci_set_master(dev, true);
L
Linus Torvalds 已提交
2781 2782 2783
	pcibios_set_master(dev);
}

2784 2785 2786 2787 2788 2789 2790 2791 2792
/**
 * pci_clear_master - disables bus-mastering for device dev
 * @dev: the PCI device to disable
 */
void pci_clear_master(struct pci_dev *dev)
{
	__pci_set_master(dev, false);
}

L
Linus Torvalds 已提交
2793
/**
2794 2795
 * 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
2796
 *
2797 2798
 * Helper function for pci_set_mwi.
 * Originally copied from drivers/net/acenic.c.
L
Linus Torvalds 已提交
2799 2800 2801 2802
 * 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
2803
int pci_set_cacheline_size(struct pci_dev *dev)
L
Linus Torvalds 已提交
2804 2805 2806 2807
{
	u8 cacheline_size;

	if (!pci_cache_line_size)
T
Tejun Heo 已提交
2808
		return -EINVAL;
L
Linus Torvalds 已提交
2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823

	/* 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;

2824 2825
	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
2826 2827 2828

	return -EINVAL;
}
T
Tejun Heo 已提交
2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846
EXPORT_SYMBOL_GPL(pci_set_cacheline_size);

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

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

void pci_clear_mwi(struct pci_dev *dev)
{
}

#else
L
Linus Torvalds 已提交
2847 2848 2849 2850 2851

/**
 * pci_set_mwi - enables memory-write-invalidate PCI transaction
 * @dev: the PCI device for which MWI is enabled
 *
R
Randy Dunlap 已提交
2852
 * Enables the Memory-Write-Invalidate transaction in %PCI_COMMAND.
L
Linus Torvalds 已提交
2853 2854 2855 2856 2857 2858 2859 2860 2861
 *
 * RETURNS: An appropriate -ERRNO error value on error, or zero for success.
 */
int
pci_set_mwi(struct pci_dev *dev)
{
	int rc;
	u16 cmd;

2862
	rc = pci_set_cacheline_size(dev);
L
Linus Torvalds 已提交
2863 2864 2865 2866 2867
	if (rc)
		return rc;

	pci_read_config_word(dev, PCI_COMMAND, &cmd);
	if (! (cmd & PCI_COMMAND_INVALIDATE)) {
2868
		dev_dbg(&dev->dev, "enabling Mem-Wr-Inval\n");
L
Linus Torvalds 已提交
2869 2870 2871 2872 2873 2874 2875
		cmd |= PCI_COMMAND_INVALIDATE;
		pci_write_config_word(dev, PCI_COMMAND, cmd);
	}
	
	return 0;
}

R
Randy Dunlap 已提交
2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890
/**
 * pci_try_set_mwi - enables memory-write-invalidate PCI transaction
 * @dev: the PCI device for which MWI is enabled
 *
 * Enables the Memory-Write-Invalidate transaction in %PCI_COMMAND.
 * Callers are not required to check the return value.
 *
 * RETURNS: An appropriate -ERRNO error value on error, or zero for success.
 */
int pci_try_set_mwi(struct pci_dev *dev)
{
	int rc = pci_set_mwi(dev);
	return rc;
}

L
Linus Torvalds 已提交
2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907
/**
 * pci_clear_mwi - disables Memory-Write-Invalidate for device dev
 * @dev: the PCI device to disable
 *
 * Disables PCI Memory-Write-Invalidate transaction on the device
 */
void
pci_clear_mwi(struct pci_dev *dev)
{
	u16 cmd;

	pci_read_config_word(dev, PCI_COMMAND, &cmd);
	if (cmd & PCI_COMMAND_INVALIDATE) {
		cmd &= ~PCI_COMMAND_INVALIDATE;
		pci_write_config_word(dev, PCI_COMMAND, cmd);
	}
}
2908
#endif /* ! PCI_DISABLE_MWI */
L
Linus Torvalds 已提交
2909

B
Brett M Russ 已提交
2910 2911
/**
 * pci_intx - enables/disables PCI INTx for device dev
R
Randy Dunlap 已提交
2912 2913
 * @pdev: the PCI device to operate on
 * @enable: boolean: whether to enable or disable PCI INTx
B
Brett M Russ 已提交
2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930
 *
 * Enables/disables PCI INTx for device dev
 */
void
pci_intx(struct pci_dev *pdev, int enable)
{
	u16 pci_command, new;

	pci_read_config_word(pdev, PCI_COMMAND, &pci_command);

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

	if (new != pci_command) {
T
Tejun Heo 已提交
2931 2932
		struct pci_devres *dr;

2933
		pci_write_config_word(pdev, PCI_COMMAND, new);
T
Tejun Heo 已提交
2934 2935 2936 2937 2938 2939

		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
2940 2941 2942
	}
}

2943 2944
/**
 * pci_intx_mask_supported - probe for INTx masking support
2945
 * @dev: the PCI device to operate on
2946 2947 2948 2949 2950 2951 2952 2953 2954
 *
 * 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;

2955 2956 2957
	if (dev->broken_intx_masking)
		return false;

2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029
	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) {
		dev_err(&dev->dev, "Command register changed from "
			"0x%x to 0x%x: driver or hardware bug?\n", orig, new);
	} 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
3030
 * @dev: the PCI device to operate on
3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043
 *
 * 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);

/**
 * pci_check_and_mask_intx - unmask INTx of no interrupt is pending
3044
 * @dev: the PCI device to operate on
3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055
 *
 * 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);

3056 3057
/**
 * pci_msi_off - disables any msi or msix capabilities
R
Randy Dunlap 已提交
3058
 * @dev: the PCI device to operate on
3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081
 *
 * If you want to use msi see pci_enable_msi and friends.
 * This is a lower level primitive that allows us to disable
 * msi operation at the device level.
 */
void pci_msi_off(struct pci_dev *dev)
{
	int pos;
	u16 control;

	pos = pci_find_capability(dev, PCI_CAP_ID_MSI);
	if (pos) {
		pci_read_config_word(dev, pos + PCI_MSI_FLAGS, &control);
		control &= ~PCI_MSI_FLAGS_ENABLE;
		pci_write_config_word(dev, pos + PCI_MSI_FLAGS, control);
	}
	pos = pci_find_capability(dev, PCI_CAP_ID_MSIX);
	if (pos) {
		pci_read_config_word(dev, pos + PCI_MSIX_FLAGS, &control);
		control &= ~PCI_MSIX_FLAGS_ENABLE;
		pci_write_config_word(dev, pos + PCI_MSIX_FLAGS, control);
	}
}
3082
EXPORT_SYMBOL_GPL(pci_msi_off);
3083

3084 3085 3086 3087 3088 3089
int pci_set_dma_max_seg_size(struct pci_dev *dev, unsigned int size)
{
	return dma_set_max_seg_size(&dev->dev, size);
}
EXPORT_SYMBOL(pci_set_dma_max_seg_size);

3090 3091 3092 3093 3094 3095
int pci_set_dma_seg_boundary(struct pci_dev *dev, unsigned long mask)
{
	return dma_set_seg_boundary(&dev->dev, mask);
}
EXPORT_SYMBOL(pci_set_dma_seg_boundary);

Y
Yu Zhao 已提交
3096
static int pcie_flr(struct pci_dev *dev, int probe)
3097
{
Y
Yu Zhao 已提交
3098
	int i;
3099
	u32 cap;
3100
	u16 status;
Y
Yu Zhao 已提交
3101

3102
	pcie_capability_read_dword(dev, PCI_EXP_DEVCAP, &cap);
3103 3104 3105
	if (!(cap & PCI_EXP_DEVCAP_FLR))
		return -ENOTTY;

S
Sheng Yang 已提交
3106 3107 3108
	if (probe)
		return 0;

3109
	/* Wait for Transaction Pending bit clean */
Y
Yu Zhao 已提交
3110 3111 3112
	for (i = 0; i < 4; i++) {
		if (i)
			msleep((1 << (i - 1)) * 100);
S
Sheng Yang 已提交
3113

3114
		pcie_capability_read_word(dev, PCI_EXP_DEVSTA, &status);
Y
Yu Zhao 已提交
3115 3116 3117 3118 3119 3120 3121 3122
		if (!(status & PCI_EXP_DEVSTA_TRPND))
			goto clear;
	}

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

clear:
3123
	pcie_capability_set_word(dev, PCI_EXP_DEVCTL, PCI_EXP_DEVCTL_BCR_FLR);
3124

Y
Yu Zhao 已提交
3125
	msleep(100);
3126 3127 3128

	return 0;
}
S
Sheng Yang 已提交
3129

Y
Yu Zhao 已提交
3130
static int pci_af_flr(struct pci_dev *dev, int probe)
3131
{
Y
Yu Zhao 已提交
3132 3133
	int i;
	int pos;
3134
	u8 cap;
Y
Yu Zhao 已提交
3135
	u8 status;
3136

Y
Yu Zhao 已提交
3137 3138
	pos = pci_find_capability(dev, PCI_CAP_ID_AF);
	if (!pos)
3139
		return -ENOTTY;
Y
Yu Zhao 已提交
3140 3141

	pci_read_config_byte(dev, pos + PCI_AF_CAP, &cap);
3142 3143 3144 3145 3146 3147 3148
	if (!(cap & PCI_AF_CAP_TP) || !(cap & PCI_AF_CAP_FLR))
		return -ENOTTY;

	if (probe)
		return 0;

	/* Wait for Transaction Pending bit clean */
Y
Yu Zhao 已提交
3149 3150 3151 3152 3153 3154 3155 3156
	for (i = 0; i < 4; i++) {
		if (i)
			msleep((1 << (i - 1)) * 100);

		pci_read_config_byte(dev, pos + PCI_AF_STATUS, &status);
		if (!(status & PCI_AF_STATUS_TP))
			goto clear;
	}
S
Sheng Yang 已提交
3157

Y
Yu Zhao 已提交
3158 3159
	dev_err(&dev->dev, "transaction is not cleared; "
			"proceeding with reset anyway\n");
S
Sheng Yang 已提交
3160

Y
Yu Zhao 已提交
3161 3162
clear:
	pci_write_config_byte(dev, pos + PCI_AF_CTRL, PCI_AF_CTRL_FLR);
3163
	msleep(100);
Y
Yu Zhao 已提交
3164

3165 3166 3167
	return 0;
}

3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182
/**
 * pci_pm_reset - Put device into PCI_D3 and back into PCI_D0.
 * @dev: Device to reset.
 * @probe: If set, only check if the device can be reset this way.
 *
 * If @dev supports native PCI PM and its PCI_PM_CTRL_NO_SOFT_RESET flag is
 * unset, it will be reinitialized internally when going from PCI_D3hot to
 * PCI_D0.  If that's the case and the device is not in a low-power state
 * already, force it into PCI_D3hot and back to PCI_D0, causing it to be reset.
 *
 * NOTE: This causes the caller to sleep for twice the device power transition
 * cooldown period, which for the D0->D3hot and D3hot->D0 transitions is 10 ms
 * by devault (i.e. unless the @dev's d3_delay field has a different value).
 * Moreover, only devices in D0 can be reset by this function.
 */
3183
static int pci_pm_reset(struct pci_dev *dev, int probe)
S
Sheng Yang 已提交
3184
{
3185 3186 3187 3188
	u16 csr;

	if (!dev->pm_cap)
		return -ENOTTY;
S
Sheng Yang 已提交
3189

3190 3191 3192
	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
3193

3194 3195
	if (probe)
		return 0;
3196

3197 3198 3199 3200 3201 3202
	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);
3203
	pci_dev_d3_sleep(dev);
3204 3205 3206 3207

	csr &= ~PCI_PM_CTRL_STATE_MASK;
	csr |= PCI_D0;
	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, csr);
3208
	pci_dev_d3_sleep(dev);
3209 3210 3211 3212

	return 0;
}

Y
Yu Zhao 已提交
3213 3214 3215 3216 3217
static int pci_parent_bus_reset(struct pci_dev *dev, int probe)
{
	u16 ctrl;
	struct pci_dev *pdev;

3218
	if (pci_is_root_bus(dev->bus) || dev->subordinate || !dev->bus->self)
Y
Yu Zhao 已提交
3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239
		return -ENOTTY;

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

	if (probe)
		return 0;

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

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

	return 0;
}

3240
static int __pci_dev_reset(struct pci_dev *dev, int probe)
S
Sheng Yang 已提交
3241
{
Y
Yu Zhao 已提交
3242 3243 3244 3245
	int rc;

	might_sleep();

3246 3247 3248 3249
	rc = pci_dev_specific_reset(dev, probe);
	if (rc != -ENOTTY)
		goto done;

Y
Yu Zhao 已提交
3250 3251 3252
	rc = pcie_flr(dev, probe);
	if (rc != -ENOTTY)
		goto done;
S
Sheng Yang 已提交
3253

Y
Yu Zhao 已提交
3254
	rc = pci_af_flr(dev, probe);
3255 3256 3257 3258
	if (rc != -ENOTTY)
		goto done;

	rc = pci_pm_reset(dev, probe);
Y
Yu Zhao 已提交
3259 3260 3261 3262
	if (rc != -ENOTTY)
		goto done;

	rc = pci_parent_bus_reset(dev, probe);
Y
Yu Zhao 已提交
3263
done:
3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278
	return rc;
}

static int pci_dev_reset(struct pci_dev *dev, int probe)
{
	int rc;

	if (!probe) {
		pci_cfg_access_lock(dev);
		/* block PM suspend, driver probe, etc. */
		device_lock(&dev->dev);
	}

	rc = __pci_dev_reset(dev, probe);

Y
Yu Zhao 已提交
3279
	if (!probe) {
3280
		device_unlock(&dev->dev);
3281
		pci_cfg_access_unlock(dev);
Y
Yu Zhao 已提交
3282 3283
	}
	return rc;
S
Sheng Yang 已提交
3284 3285
}
/**
Y
Yu Zhao 已提交
3286 3287
 * __pci_reset_function - reset a PCI device function
 * @dev: PCI device to reset
S
Sheng Yang 已提交
3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298
 *
 * 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
3299
 * Returns 0 if the device function was successfully reset or negative if the
S
Sheng Yang 已提交
3300 3301
 * device doesn't support resetting a single function.
 */
Y
Yu Zhao 已提交
3302
int __pci_reset_function(struct pci_dev *dev)
S
Sheng Yang 已提交
3303
{
Y
Yu Zhao 已提交
3304
	return pci_dev_reset(dev, 0);
S
Sheng Yang 已提交
3305
}
Y
Yu Zhao 已提交
3306
EXPORT_SYMBOL_GPL(__pci_reset_function);
3307

3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328
/**
 * __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)
{
3329
	return __pci_dev_reset(dev, 0);
3330 3331 3332
}
EXPORT_SYMBOL_GPL(__pci_reset_function_locked);

3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348
/**
 * 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);
}

3349
/**
Y
Yu Zhao 已提交
3350 3351
 * pci_reset_function - quiesce and reset a PCI device function
 * @dev: PCI device to reset
3352 3353 3354 3355 3356 3357 3358
 *
 * 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
3359
 * from __pci_reset_function in that it saves and restores device state
3360 3361
 * over the reset.
 *
Y
Yu Zhao 已提交
3362
 * Returns 0 if the device function was successfully reset or negative if the
3363 3364 3365 3366
 * device doesn't support resetting a single function.
 */
int pci_reset_function(struct pci_dev *dev)
{
Y
Yu Zhao 已提交
3367
	int rc;
3368

Y
Yu Zhao 已提交
3369 3370 3371
	rc = pci_dev_reset(dev, 1);
	if (rc)
		return rc;
3372 3373 3374

	pci_save_state(dev);

Y
Yu Zhao 已提交
3375 3376 3377 3378
	/*
	 * both INTx and MSI are disabled after the Interrupt Disable bit
	 * is set and the Bus Master bit is cleared.
	 */
3379 3380
	pci_write_config_word(dev, PCI_COMMAND, PCI_COMMAND_INTX_DISABLE);

Y
Yu Zhao 已提交
3381
	rc = pci_dev_reset(dev, 0);
3382 3383 3384

	pci_restore_state(dev);

Y
Yu Zhao 已提交
3385
	return rc;
3386 3387 3388
}
EXPORT_SYMBOL_GPL(pci_reset_function);

3389 3390 3391 3392 3393 3394 3395 3396 3397
/**
 * 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)
{
3398
	int cap;
3399 3400 3401 3402 3403 3404
	u32 stat;

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

3405
	if (pci_read_config_dword(dev, cap + PCI_X_STATUS, &stat))
3406 3407
		return -EINVAL;

3408
	return 512 << ((stat & PCI_X_STATUS_MAX_READ) >> 21);
3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420
}
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)
{
3421
	int cap;
3422
	u16 cmd;
3423 3424 3425 3426 3427

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

3428 3429
	if (pci_read_config_word(dev, cap + PCI_X_CMD, &cmd))
		return -EINVAL;
3430

3431
	return 512 << ((cmd & PCI_X_CMD_MAX_READ) >> 2);
3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445
}
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)
{
3446
	int cap;
3447 3448
	u32 stat, v, o;
	u16 cmd;
3449

3450
	if (mmrbc < 512 || mmrbc > 4096 || !is_power_of_2(mmrbc))
3451
		return -EINVAL;
3452 3453 3454 3455 3456

	v = ffs(mmrbc) - 10;

	cap = pci_find_capability(dev, PCI_CAP_ID_PCIX);
	if (!cap)
3457
		return -EINVAL;
3458

3459 3460
	if (pci_read_config_dword(dev, cap + PCI_X_STATUS, &stat))
		return -EINVAL;
3461 3462 3463 3464

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

3465 3466
	if (pci_read_config_word(dev, cap + PCI_X_CMD, &cmd))
		return -EINVAL;
3467 3468 3469

	o = (cmd & PCI_X_CMD_MAX_READ) >> 2;
	if (o != v) {
3470
		if (v > o && (dev->bus->bus_flags & PCI_BUS_FLAGS_NO_MMRBC))
3471 3472 3473 3474
			return -EIO;

		cmd &= ~PCI_X_CMD_MAX_READ;
		cmd |= v << 2;
3475 3476
		if (pci_write_config_word(dev, cap + PCI_X_CMD, cmd))
			return -EIO;
3477
	}
3478
	return 0;
3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492
}
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;

3493
	pcie_capability_read_word(dev, PCI_EXP_DEVCTL, &ctl);
3494

3495
	return 128 << ((ctl & PCI_EXP_DEVCTL_READRQ) >> 12);
3496 3497 3498 3499 3500 3501
}
EXPORT_SYMBOL(pcie_get_readrq);

/**
 * pcie_set_readrq - set PCI Express maximum memory read request
 * @dev: PCI device to query
3502
 * @rq: maximum memory read count in bytes
3503 3504
 *    valid values are 128, 256, 512, 1024, 2048, 4096
 *
3505
 * If possible sets maximum memory read request in bytes
3506 3507 3508
 */
int pcie_set_readrq(struct pci_dev *dev, int rq)
{
3509
	u16 v;
3510

3511
	if (rq < 128 || rq > 4096 || !is_power_of_2(rq))
3512
		return -EINVAL;
3513

3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528 3529
	/*
	 * 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 < 0)
			return mps;
		if (mps < rq)
			rq = mps;
	}

	v = (ffs(rq) - 8) << 12;
3530

3531 3532
	return pcie_capability_clear_and_set_word(dev, PCI_EXP_DEVCTL,
						  PCI_EXP_DEVCTL_READRQ, v);
3533 3534 3535
}
EXPORT_SYMBOL(pcie_set_readrq);

3536 3537 3538 3539 3540 3541 3542 3543 3544 3545 3546
/**
 * pcie_get_mps - get PCI Express maximum payload size
 * @dev: PCI device to query
 *
 * Returns maximum payload size in bytes
 *    or appropriate error value.
 */
int pcie_get_mps(struct pci_dev *dev)
{
	u16 ctl;

3547
	pcie_capability_read_word(dev, PCI_EXP_DEVCTL, &ctl);
3548

3549
	return 128 << ((ctl & PCI_EXP_DEVCTL_PAYLOAD) >> 5);
3550 3551 3552 3553 3554
}

/**
 * pcie_set_mps - set PCI Express maximum payload size
 * @dev: PCI device to query
3555
 * @mps: maximum payload size in bytes
3556 3557 3558 3559 3560 3561
 *    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)
{
3562
	u16 v;
3563 3564

	if (mps < 128 || mps > 4096 || !is_power_of_2(mps))
3565
		return -EINVAL;
3566 3567 3568

	v = ffs(mps) - 8;
	if (v > dev->pcie_mpss) 
3569
		return -EINVAL;
3570 3571
	v <<= 5;

3572 3573
	return pcie_capability_clear_and_set_word(dev, PCI_EXP_DEVCTL,
						  PCI_EXP_DEVCTL_PAYLOAD, v);
3574 3575
}

3576 3577
/**
 * pci_select_bars - Make BAR mask from the type of resource
3578
 * @dev: the PCI device for which BAR mask is made
3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591
 * @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;
}

3592 3593 3594 3595 3596 3597 3598 3599 3600 3601
/**
 * 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)
{
3602 3603
	int reg;

3604 3605 3606 3607 3608 3609
	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;
3610 3611 3612 3613 3614
	} else if (resno < PCI_BRIDGE_RESOURCES) {
		/* device specific resource */
		reg = pci_iov_resource_bar(dev, resno, type);
		if (reg)
			return reg;
3615 3616
	}

3617
	dev_err(&dev->dev, "BAR %d: invalid resource\n", resno);
3618 3619 3620
	return 0;
}

3621 3622 3623 3624 3625 3626 3627 3628 3629
/* 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,
3630
		      unsigned int command_bits, u32 flags)
3631 3632 3633
{
	if (arch_set_vga_state)
		return arch_set_vga_state(dev, decode, command_bits,
3634
						flags);
3635 3636 3637
	return 0;
}

3638 3639
/**
 * pci_set_vga_state - set VGA decode state on device and parents if requested
R
Randy Dunlap 已提交
3640 3641 3642
 * @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
3643
 * @flags: traverse ancestors and change bridges
3644
 * CHANGE_BRIDGE_ONLY / CHANGE_BRIDGE
3645 3646
 */
int pci_set_vga_state(struct pci_dev *dev, bool decode,
3647
		      unsigned int command_bits, u32 flags)
3648 3649 3650 3651
{
	struct pci_bus *bus;
	struct pci_dev *bridge;
	u16 cmd;
3652
	int rc;
3653

3654
	WARN_ON((flags & PCI_VGA_STATE_CHANGE_DECODES) & (command_bits & ~(PCI_COMMAND_IO|PCI_COMMAND_MEMORY)));
3655

3656
	/* ARCH specific VGA enables */
3657
	rc = pci_set_vga_state_arch(dev, decode, command_bits, flags);
3658 3659 3660
	if (rc)
		return rc;

3661 3662 3663 3664 3665 3666 3667 3668
	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);
	}
3669

3670
	if (!(flags & PCI_VGA_STATE_CHANGE_BRIDGE))
3671 3672 3673 3674 3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690
		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;
}

3691 3692
#define RESOURCE_ALIGNMENT_PARAM_SIZE COMMAND_LINE_SIZE
static char resource_alignment_param[RESOURCE_ALIGNMENT_PARAM_SIZE] = {0};
3693
static DEFINE_SPINLOCK(resource_alignment_lock);
3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 3720 3721 3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734 3735 3736 3737 3738 3739 3740 3741 3742 3743 3744 3745 3746 3747 3748 3749 3750 3751

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

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

3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765
/*
 * 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
3766 3767 3768
	/* check if specified PCI is target device to reassign */
	align = pci_specified_resource_alignment(dev);
	if (!align)
3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814
		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);
		}
		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;
			r->end = resource_size(r) - 1;
			r->start = 0;
		}
		pci_disable_bridge_window(dev);
	}
}

3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856
ssize_t pci_set_resource_alignment_param(const char *buf, size_t count)
{
	if (count > RESOURCE_ALIGNMENT_PARAM_SIZE - 1)
		count = RESOURCE_ALIGNMENT_PARAM_SIZE - 1;
	spin_lock(&resource_alignment_lock);
	strncpy(resource_alignment_param, buf, count);
	resource_alignment_param[count] = '\0';
	spin_unlock(&resource_alignment_lock);
	return count;
}

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

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

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

BUS_ATTR(resource_alignment, 0644, pci_resource_alignment_show,
					pci_resource_alignment_store);

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

late_initcall(pci_resource_alignment_sysfs_init);

B
Bill Pemberton 已提交
3857
static void pci_no_domains(void)
3858 3859 3860 3861 3862 3863
{
#ifdef CONFIG_PCI_DOMAINS
	pci_domains_supported = 0;
#endif
}

3864
/**
3865
 * pci_ext_cfg_avail - can we access extended PCI config space?
3866 3867 3868 3869 3870
 *
 * Returns 1 if we can access PCI extended config space (offsets
 * greater than 0xff). This is the default implementation. Architecture
 * implementations can override this.
 */
3871
int __weak pci_ext_cfg_avail(void)
3872 3873 3874 3875
{
	return 1;
}

3876 3877 3878 3879 3880
void __weak pci_fixup_cardbus(struct pci_bus *bus)
{
}
EXPORT_SYMBOL(pci_fixup_cardbus);

A
Al Viro 已提交
3881
static int __init pci_setup(char *str)
L
Linus Torvalds 已提交
3882 3883 3884 3885 3886 3887
{
	while (str) {
		char *k = strchr(str, ',');
		if (k)
			*k++ = 0;
		if (*str && (str = pcibios_setup(str)) && *str) {
3888 3889
			if (!strcmp(str, "nomsi")) {
				pci_no_msi();
R
Randy Dunlap 已提交
3890 3891
			} else if (!strcmp(str, "noaer")) {
				pci_no_aer();
3892 3893
			} else if (!strncmp(str, "realloc=", 8)) {
				pci_realloc_get_opt(str + 8);
3894
			} else if (!strncmp(str, "realloc", 7)) {
3895
				pci_realloc_get_opt("on");
3896 3897
			} else if (!strcmp(str, "nodomains")) {
				pci_no_domains();
3898 3899
			} else if (!strncmp(str, "noari", 5)) {
				pcie_ari_disabled = true;
3900 3901 3902 3903
			} 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);
3904 3905 3906
			} else if (!strncmp(str, "resource_alignment=", 19)) {
				pci_set_resource_alignment_param(str + 19,
							strlen(str + 19));
3907 3908
			} else if (!strncmp(str, "ecrc=", 5)) {
				pcie_ecrc_get_policy(str + 5);
3909 3910 3911 3912
			} 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);
3913 3914
			} else if (!strncmp(str, "pcie_bus_tune_off", 17)) {
				pcie_bus_config = PCIE_BUS_TUNE_OFF;
3915 3916 3917 3918
			} 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;
3919 3920
			} else if (!strncmp(str, "pcie_bus_peer2peer", 18)) {
				pcie_bus_config = PCIE_BUS_PEER2PEER;
3921 3922
			} else if (!strncmp(str, "pcie_scan_all", 13)) {
				pci_add_flags(PCI_SCAN_ALL_PCIE_DEVS);
3923 3924 3925 3926
			} else {
				printk(KERN_ERR "PCI: Unknown option `%s'\n",
						str);
			}
L
Linus Torvalds 已提交
3927 3928 3929
		}
		str = k;
	}
3930
	return 0;
L
Linus Torvalds 已提交
3931
}
3932
early_param("pci", pci_setup);
L
Linus Torvalds 已提交
3933

3934
EXPORT_SYMBOL(pci_reenable_device);
3935 3936
EXPORT_SYMBOL(pci_enable_device_io);
EXPORT_SYMBOL(pci_enable_device_mem);
L
Linus Torvalds 已提交
3937
EXPORT_SYMBOL(pci_enable_device);
T
Tejun Heo 已提交
3938 3939
EXPORT_SYMBOL(pcim_enable_device);
EXPORT_SYMBOL(pcim_pin_device);
L
Linus Torvalds 已提交
3940 3941 3942 3943 3944
EXPORT_SYMBOL(pci_disable_device);
EXPORT_SYMBOL(pci_find_capability);
EXPORT_SYMBOL(pci_bus_find_capability);
EXPORT_SYMBOL(pci_release_regions);
EXPORT_SYMBOL(pci_request_regions);
3945
EXPORT_SYMBOL(pci_request_regions_exclusive);
L
Linus Torvalds 已提交
3946 3947
EXPORT_SYMBOL(pci_release_region);
EXPORT_SYMBOL(pci_request_region);
3948
EXPORT_SYMBOL(pci_request_region_exclusive);
3949 3950
EXPORT_SYMBOL(pci_release_selected_regions);
EXPORT_SYMBOL(pci_request_selected_regions);
3951
EXPORT_SYMBOL(pci_request_selected_regions_exclusive);
L
Linus Torvalds 已提交
3952
EXPORT_SYMBOL(pci_set_master);
3953
EXPORT_SYMBOL(pci_clear_master);
L
Linus Torvalds 已提交
3954
EXPORT_SYMBOL(pci_set_mwi);
R
Randy Dunlap 已提交
3955
EXPORT_SYMBOL(pci_try_set_mwi);
L
Linus Torvalds 已提交
3956
EXPORT_SYMBOL(pci_clear_mwi);
B
Brett M Russ 已提交
3957
EXPORT_SYMBOL_GPL(pci_intx);
L
Linus Torvalds 已提交
3958 3959
EXPORT_SYMBOL(pci_assign_resource);
EXPORT_SYMBOL(pci_find_parent_resource);
3960
EXPORT_SYMBOL(pci_select_bars);
L
Linus Torvalds 已提交
3961 3962 3963 3964

EXPORT_SYMBOL(pci_set_power_state);
EXPORT_SYMBOL(pci_save_state);
EXPORT_SYMBOL(pci_restore_state);
3965
EXPORT_SYMBOL(pci_pme_capable);
3966
EXPORT_SYMBOL(pci_pme_active);
3967
EXPORT_SYMBOL(pci_wake_from_d3);
3968
EXPORT_SYMBOL(pci_target_state);
3969 3970
EXPORT_SYMBOL(pci_prepare_to_sleep);
EXPORT_SYMBOL(pci_back_from_sleep);
B
Brian King 已提交
3971
EXPORT_SYMBOL_GPL(pci_set_pcie_reset_state);