pci.c 75.0 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>
L
Linus Torvalds 已提交
15 16
#include <linux/module.h>
#include <linux/spinlock.h>
T
Tim Schmielau 已提交
17
#include <linux/string.h>
18
#include <linux/log2.h>
S
Shaohua Li 已提交
19
#include <linux/pci-aspm.h>
20
#include <linux/pm_wakeup.h>
21
#include <linux/interrupt.h>
L
Linus Torvalds 已提交
22
#include <asm/dma.h>	/* isa_dma_bridge_buggy */
23 24
#include <linux/device.h>
#include <asm/setup.h>
25
#include "pci.h"
L
Linus Torvalds 已提交
26

A
Alan Stern 已提交
27 28 29 30 31
const char *pci_power_names[] = {
	"error", "D0", "D1", "D2", "D3hot", "D3cold", "unknown",
};
EXPORT_SYMBOL_GPL(pci_power_names);

32 33 34 35 36 37 38 39 40 41 42
unsigned int pci_pm_d3_delay;

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
43

44 45 46 47
#ifdef CONFIG_PCI_DOMAINS
int pci_domains_supported = 1;
#endif

48 49 50 51 52 53
#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;

54 55 56 57 58 59
#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;

60 61 62 63 64 65
/*
 * 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.
 */
66
u8 pci_dfl_cache_line_size __devinitdata = L1_CACHE_BYTES >> 2;
67 68
u8 pci_cache_line_size;

L
Linus Torvalds 已提交
69 70 71 72 73 74 75
/**
 * 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.
 */
76
unsigned char pci_bus_max_busnr(struct pci_bus* bus)
L
Linus Torvalds 已提交
77 78 79 80
{
	struct list_head *tmp;
	unsigned char max, n;

81
	max = bus->subordinate;
L
Linus Torvalds 已提交
82 83 84 85 86 87 88
	list_for_each(tmp, &bus->children) {
		n = pci_bus_max_busnr(pci_bus_b(tmp));
		if(n > max)
			max = n;
	}
	return max;
}
89
EXPORT_SYMBOL_GPL(pci_bus_max_busnr);
L
Linus Torvalds 已提交
90

A
Andrew Morton 已提交
91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106
#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

107
#if 0
L
Linus Torvalds 已提交
108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128
/**
 * pci_max_busnr - returns maximum PCI bus number
 *
 * Returns the highest PCI bus number present in the system global list of
 * PCI buses.
 */
unsigned char __devinit
pci_max_busnr(void)
{
	struct pci_bus *bus = NULL;
	unsigned char max, n;

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

129 130
#endif  /*  0  */

131 132 133 134
#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)
135 136 137
{
	u8 id;

138
	while ((*ttl)--) {
139 140 141 142 143 144 145 146 147 148 149 150 151 152 153
		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;
}

154 155 156 157 158 159 160 161
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);
}

162 163 164 165 166 167 168
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);

169 170
static int __pci_bus_find_cap_start(struct pci_bus *bus,
				    unsigned int devfn, u8 hdr_type)
L
Linus Torvalds 已提交
171 172 173 174 175 176 177 178 179 180
{
	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:
181
		return PCI_CAPABILITY_LIST;
L
Linus Torvalds 已提交
182
	case PCI_HEADER_TYPE_CARDBUS:
183
		return PCI_CB_CAPABILITY_LIST;
L
Linus Torvalds 已提交
184 185 186
	default:
		return 0;
	}
187 188

	return 0;
L
Linus Torvalds 已提交
189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211
}

/**
 * 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)
{
212 213 214 215 216 217 218
	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
219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235
}

/**
 * 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)
{
236
	int pos;
L
Linus Torvalds 已提交
237 238 239 240
	u8 hdr_type;

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

241 242 243 244 245
	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
246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264
}

/**
 * pci_find_ext_capability - Find an extended capability
 * @dev: PCI device to query
 * @cap: capability code
 *
 * Returns the address of the requested extended capability structure
 * within the device's PCI configuration space or 0 if the device does
 * not support it.  Possible values for @cap:
 *
 *  %PCI_EXT_CAP_ID_ERR		Advanced Error Reporting
 *  %PCI_EXT_CAP_ID_VC		Virtual Channel
 *  %PCI_EXT_CAP_ID_DSN		Device Serial Number
 *  %PCI_EXT_CAP_ID_PWR		Power Budgeting
 */
int pci_find_ext_capability(struct pci_dev *dev, int cap)
{
	u32 header;
265 266
	int ttl;
	int pos = PCI_CFG_SPACE_SIZE;
L
Linus Torvalds 已提交
267

268 269 270 271
	/* 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
272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288
		return 0;

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

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

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

		pos = PCI_EXT_CAP_NEXT(header);
289
		if (pos < PCI_CFG_SPACE_SIZE)
L
Linus Torvalds 已提交
290 291 292 293 294 295 296 297
			break;

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

	return 0;
}
298
EXPORT_SYMBOL_GPL(pci_find_ext_capability);
L
Linus Torvalds 已提交
299

300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342
/**
 * pci_bus_find_ext_capability - find an extended capability
 * @bus:   the PCI bus to query
 * @devfn: PCI device to query
 * @cap:   capability code
 *
 * Like pci_find_ext_capability() but works for pci devices that do not have a
 * pci_dev structure set up yet.
 *
 * Returns the address of the requested capability structure within the
 * device's PCI configuration space or 0 in case the device does not
 * support it.
 */
int pci_bus_find_ext_capability(struct pci_bus *bus, unsigned int devfn,
				int cap)
{
	u32 header;
	int ttl;
	int pos = PCI_CFG_SPACE_SIZE;

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

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

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

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

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

	return 0;
}

343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362
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;

363 364
		pos = __pci_find_next_cap_ttl(dev->bus, dev->devfn,
					      pos + PCI_CAP_LIST_NEXT,
365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411
					      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
412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437
/**
 * 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;
	struct resource *best = NULL;

	for(i = 0; i < PCI_BUS_NUM_RESOURCES; i++) {
		struct resource *r = bus->resource[i];
		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 */
438 439 440 441 442 443
		/* 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
444 445 446 447
	}
	return best;
}

448 449 450 451 452 453 454
/**
 * 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.
 */
455
static void
456 457
pci_restore_bars(struct pci_dev *dev)
{
458
	int i;
459

460
	for (i = 0; i < PCI_BRIDGE_RESOURCES; i++)
461
		pci_update_resource(dev, i);
462 463
}

464 465 466 467
static struct pci_platform_pm_ops *pci_platform_pm;

int pci_set_platform_pm(struct pci_platform_pm_ops *ops)
{
468 469
	if (!ops->is_manageable || !ops->set_state || !ops->choose_state
	    || !ops->sleep_wake || !ops->can_wakeup)
470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490
		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
491

492 493 494 495 496 497 498 499 500 501 502
static inline bool platform_pci_can_wakeup(struct pci_dev *dev)
{
	return pci_platform_pm ? pci_platform_pm->can_wakeup(dev) : false;
}

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

L
Linus Torvalds 已提交
503
/**
504 505 506 507
 * 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
508
 *
509 510 511 512 513 514
 * 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
515
 */
516
static int pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state)
L
Linus Torvalds 已提交
517
{
518
	u16 pmcsr;
519
	bool need_restore = false;
L
Linus Torvalds 已提交
520

521 522 523 524
	/* Check if we're already there */
	if (dev->current_state == state)
		return 0;

525
	if (!dev->pm_cap)
526 527
		return -EIO;

528 529 530
	if (state < PCI_D0 || state > PCI_D3hot)
		return -EINVAL;

L
Linus Torvalds 已提交
531 532 533 534
	/* 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
	 */
535
	if (state != PCI_D0 && dev->current_state <= PCI_D3cold
536
	    && dev->current_state > state) {
537 538
		dev_err(&dev->dev, "invalid power transition "
			"(from state %d to %d)\n", dev->current_state, state);
L
Linus Torvalds 已提交
539
		return -EINVAL;
540
	}
L
Linus Torvalds 已提交
541 542

	/* check if this device supports the desired state */
543 544
	if ((state == PCI_D1 && !dev->d1_support)
	   || (state == PCI_D2 && !dev->d2_support))
545
		return -EIO;
L
Linus Torvalds 已提交
546

547
	pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
548

549
	/* If we're (effectively) in D3, force entire word to 0.
L
Linus Torvalds 已提交
550 551 552
	 * This doesn't affect PME_Status, disables PME_En, and
	 * sets PowerState to 0.
	 */
553
	switch (dev->current_state) {
554 555 556 557 558 559
	case PCI_D0:
	case PCI_D1:
	case PCI_D2:
		pmcsr &= ~PCI_PM_CTRL_STATE_MASK;
		pmcsr |= state;
		break;
560 561
	case PCI_D3hot:
	case PCI_D3cold:
562 563
	case PCI_UNKNOWN: /* Boot-up */
		if ((pmcsr & PCI_PM_CTRL_STATE_MASK) == PCI_D3hot
564
		 && !(pmcsr & PCI_PM_CTRL_NO_SOFT_RESET))
565
			need_restore = true;
566 567
		/* Fall-through: force to D0 */
	default:
568
		pmcsr = 0;
569
		break;
L
Linus Torvalds 已提交
570 571 572
	}

	/* enter specified state */
573
	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, pmcsr);
L
Linus Torvalds 已提交
574 575 576 577

	/* Mandatory power management transition delays */
	/* see PCI PM 1.1 5.6.1 table 18 */
	if (state == PCI_D3hot || dev->current_state == PCI_D3hot)
578
		pci_dev_d3_sleep(dev);
L
Linus Torvalds 已提交
579
	else if (state == PCI_D2 || dev->current_state == PCI_D2)
580
		udelay(PCI_PM_D2_DELAY);
L
Linus Torvalds 已提交
581

582 583 584 585 586
	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);
587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602

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

603
	if (dev->bus->self)
S
Shaohua Li 已提交
604 605
		pcie_aspm_pm_state_change(dev->bus->self);

L
Linus Torvalds 已提交
606 607 608
	return 0;
}

609 610 611 612
/**
 * pci_update_current_state - Read PCI power state of given device from its
 *                            PCI PM registers and cache it
 * @dev: PCI device to handle.
613
 * @state: State to cache in case the device doesn't have the PM capability
614
 */
615
void pci_update_current_state(struct pci_dev *dev, pci_power_t state)
616
{
617
	if (dev->pm_cap) {
618 619
		u16 pmcsr;

620
		pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
621
		dev->current_state = (pmcsr & PCI_PM_CTRL_STATE_MASK);
622 623
	} else {
		dev->current_state = state;
624 625 626
	}
}

627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642
/**
 * pci_platform_power_transition - Use platform to change device power state
 * @dev: PCI device to handle.
 * @state: State to put the device into.
 */
static int pci_platform_power_transition(struct pci_dev *dev, pci_power_t state)
{
	int error;

	if (platform_pci_power_manageable(dev)) {
		error = platform_pci_set_power_state(dev, state);
		if (!error)
			pci_update_current_state(dev, state);
	} else {
		error = -ENODEV;
		/* Fall back to PCI_D0 if native PM is not supported */
643 644
		if (!dev->pm_cap)
			dev->current_state = PCI_D0;
645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674
	}

	return error;
}

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

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

675 676 677 678 679
/**
 * 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.
 *
680
 * Transition a device to a new power state, using the platform firmware and/or
681 682 683 684 685 686 687 688 689 690 691
 * 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)
{
692
	int error;
693 694 695 696 697 698 699 700 701 702 703 704 705 706

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

707 708 709 710
	/* Check if we're already there */
	if (dev->current_state == state)
		return 0;

711 712
	__pci_start_power_transition(dev, state);

713 714 715 716
	/* This device is quirked not to be put into D3, so
	   don't put it in D3 */
	if (state == PCI_D3hot && (dev->dev_flags & PCI_DEV_FLAGS_NO_D3))
		return 0;
717

718
	error = pci_raw_set_power_state(dev, state);
719

720 721
	if (!__pci_complete_power_transition(dev, state))
		error = 0;
722 723 724 725

	return error;
}

L
Linus Torvalds 已提交
726 727 728 729 730 731 732 733 734 735 736 737
/**
 * 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)
{
738
	pci_power_t ret;
739

L
Linus Torvalds 已提交
740 741 742
	if (!pci_find_capability(dev, PCI_CAP_ID_PM))
		return PCI_D0;

743 744 745
	ret = platform_pci_choose_state(dev);
	if (ret != PCI_POWER_ERROR)
		return ret;
746 747 748 749 750

	switch (state.event) {
	case PM_EVENT_ON:
		return PCI_D0;
	case PM_EVENT_FREEZE:
751 752
	case PM_EVENT_PRETHAW:
		/* REVISIT both freeze and pre-thaw "should" use D0 */
753
	case PM_EVENT_SUSPEND:
754
	case PM_EVENT_HIBERNATE:
755
		return PCI_D3hot;
L
Linus Torvalds 已提交
756
	default:
757 758
		dev_info(&dev->dev, "unrecognized suspend event %d\n",
			 state.event);
L
Linus Torvalds 已提交
759 760 761 762 763 764 765
		BUG();
	}
	return PCI_D0;
}

EXPORT_SYMBOL(pci_choose_state);

766 767
#define PCI_EXP_SAVE_REGS	7

768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789
#define pcie_cap_has_devctl(type, flags)	1
#define pcie_cap_has_lnkctl(type, flags)		\
		((flags & PCI_EXP_FLAGS_VERS) > 1 ||	\
		 (type == PCI_EXP_TYPE_ROOT_PORT ||	\
		  type == PCI_EXP_TYPE_ENDPOINT ||	\
		  type == PCI_EXP_TYPE_LEG_END))
#define pcie_cap_has_sltctl(type, flags)		\
		((flags & PCI_EXP_FLAGS_VERS) > 1 ||	\
		 ((type == PCI_EXP_TYPE_ROOT_PORT) ||	\
		  (type == PCI_EXP_TYPE_DOWNSTREAM &&	\
		   (flags & PCI_EXP_FLAGS_SLOT))))
#define pcie_cap_has_rtctl(type, flags)			\
		((flags & PCI_EXP_FLAGS_VERS) > 1 ||	\
		 (type == PCI_EXP_TYPE_ROOT_PORT ||	\
		  type == PCI_EXP_TYPE_RC_EC))
#define pcie_cap_has_devctl2(type, flags)		\
		((flags & PCI_EXP_FLAGS_VERS) > 1)
#define pcie_cap_has_lnkctl2(type, flags)		\
		((flags & PCI_EXP_FLAGS_VERS) > 1)
#define pcie_cap_has_sltctl2(type, flags)		\
		((flags & PCI_EXP_FLAGS_VERS) > 1)

790 791 792 793 794
static int pci_save_pcie_state(struct pci_dev *dev)
{
	int pos, i = 0;
	struct pci_cap_saved_state *save_state;
	u16 *cap;
795
	u16 flags;
796

797 798
	pos = pci_pcie_cap(dev);
	if (!pos)
799 800
		return 0;

801
	save_state = pci_find_saved_cap(dev, PCI_CAP_ID_EXP);
802
	if (!save_state) {
803
		dev_err(&dev->dev, "buffer not found in %s\n", __func__);
804 805 806 807
		return -ENOMEM;
	}
	cap = (u16 *)&save_state->data[0];

808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823
	pci_read_config_word(dev, pos + PCI_EXP_FLAGS, &flags);

	if (pcie_cap_has_devctl(dev->pcie_type, flags))
		pci_read_config_word(dev, pos + PCI_EXP_DEVCTL, &cap[i++]);
	if (pcie_cap_has_lnkctl(dev->pcie_type, flags))
		pci_read_config_word(dev, pos + PCI_EXP_LNKCTL, &cap[i++]);
	if (pcie_cap_has_sltctl(dev->pcie_type, flags))
		pci_read_config_word(dev, pos + PCI_EXP_SLTCTL, &cap[i++]);
	if (pcie_cap_has_rtctl(dev->pcie_type, flags))
		pci_read_config_word(dev, pos + PCI_EXP_RTCTL, &cap[i++]);
	if (pcie_cap_has_devctl2(dev->pcie_type, flags))
		pci_read_config_word(dev, pos + PCI_EXP_DEVCTL2, &cap[i++]);
	if (pcie_cap_has_lnkctl2(dev->pcie_type, flags))
		pci_read_config_word(dev, pos + PCI_EXP_LNKCTL2, &cap[i++]);
	if (pcie_cap_has_sltctl2(dev->pcie_type, flags))
		pci_read_config_word(dev, pos + PCI_EXP_SLTCTL2, &cap[i++]);
824

825 826 827 828 829 830 831 832
	return 0;
}

static void pci_restore_pcie_state(struct pci_dev *dev)
{
	int i = 0, pos;
	struct pci_cap_saved_state *save_state;
	u16 *cap;
833
	u16 flags;
834 835 836 837 838 839 840

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

841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856
	pci_read_config_word(dev, pos + PCI_EXP_FLAGS, &flags);

	if (pcie_cap_has_devctl(dev->pcie_type, flags))
		pci_write_config_word(dev, pos + PCI_EXP_DEVCTL, cap[i++]);
	if (pcie_cap_has_lnkctl(dev->pcie_type, flags))
		pci_write_config_word(dev, pos + PCI_EXP_LNKCTL, cap[i++]);
	if (pcie_cap_has_sltctl(dev->pcie_type, flags))
		pci_write_config_word(dev, pos + PCI_EXP_SLTCTL, cap[i++]);
	if (pcie_cap_has_rtctl(dev->pcie_type, flags))
		pci_write_config_word(dev, pos + PCI_EXP_RTCTL, cap[i++]);
	if (pcie_cap_has_devctl2(dev->pcie_type, flags))
		pci_write_config_word(dev, pos + PCI_EXP_DEVCTL2, cap[i++]);
	if (pcie_cap_has_lnkctl2(dev->pcie_type, flags))
		pci_write_config_word(dev, pos + PCI_EXP_LNKCTL2, cap[i++]);
	if (pcie_cap_has_sltctl2(dev->pcie_type, flags))
		pci_write_config_word(dev, pos + PCI_EXP_SLTCTL2, cap[i++]);
857 858
}

S
Stephen Hemminger 已提交
859 860 861

static int pci_save_pcix_state(struct pci_dev *dev)
{
862
	int pos;
S
Stephen Hemminger 已提交
863 864 865 866 867 868
	struct pci_cap_saved_state *save_state;

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

869
	save_state = pci_find_saved_cap(dev, PCI_CAP_ID_PCIX);
S
Stephen Hemminger 已提交
870
	if (!save_state) {
871
		dev_err(&dev->dev, "buffer not found in %s\n", __func__);
S
Stephen Hemminger 已提交
872 873 874
		return -ENOMEM;
	}

875 876
	pci_read_config_word(dev, pos + PCI_X_CMD, (u16 *)save_state->data);

S
Stephen Hemminger 已提交
877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895
	return 0;
}

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

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

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


L
Linus Torvalds 已提交
896 897 898 899 900 901 902 903 904 905
/**
 * 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++)
906
		pci_read_config_dword(dev, i * 4, &dev->saved_config_space[i]);
907
	dev->state_saved = true;
908 909
	if ((i = pci_save_pcie_state(dev)) != 0)
		return i;
S
Stephen Hemminger 已提交
910 911
	if ((i = pci_save_pcix_state(dev)) != 0)
		return i;
L
Linus Torvalds 已提交
912 913 914 915 916 917 918 919 920 921 922
	return 0;
}

/** 
 * pci_restore_state - Restore the saved state of a PCI device
 * @dev: - PCI device that we're dealing with
 */
int 
pci_restore_state(struct pci_dev *dev)
{
	int i;
A
Al Viro 已提交
923
	u32 val;
L
Linus Torvalds 已提交
924

A
Alek Du 已提交
925 926
	if (!dev->state_saved)
		return 0;
927

928 929 930
	/* PCI Express register must be restored first */
	pci_restore_pcie_state(dev);

931 932 933 934 935
	/*
	 * The Base Address register should be programmed before the command
	 * register(s)
	 */
	for (i = 15; i >= 0; i--) {
936 937
		pci_read_config_dword(dev, i * 4, &val);
		if (val != dev->saved_config_space[i]) {
938 939 940
			dev_printk(KERN_DEBUG, &dev->dev, "restoring config "
				"space at offset %#x (was %#x, writing %#x)\n",
				i, val, (int)dev->saved_config_space[i]);
941 942 943 944
			pci_write_config_dword(dev,i * 4,
				dev->saved_config_space[i]);
		}
	}
S
Stephen Hemminger 已提交
945
	pci_restore_pcix_state(dev);
946
	pci_restore_msi_state(dev);
Y
Yu Zhao 已提交
947
	pci_restore_iov_state(dev);
948

949 950
	dev->state_saved = false;

L
Linus Torvalds 已提交
951 952 953
	return 0;
}

954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969
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;
}

/**
970
 * pci_reenable_device - Resume abandoned device
971 972 973 974 975
 * @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.
 */
976
int pci_reenable_device(struct pci_dev *dev)
977
{
978
	if (pci_is_enabled(dev))
979 980 981 982
		return do_pci_enable_device(dev, (1 << PCI_NUM_RESOURCES) - 1);
	return 0;
}

983 984
static int __pci_enable_device_flags(struct pci_dev *dev,
				     resource_size_t flags)
L
Linus Torvalds 已提交
985 986
{
	int err;
987
	int i, bars = 0;
L
Linus Torvalds 已提交
988

989 990 991
	if (atomic_add_return(1, &dev->enable_cnt) > 1)
		return 0;		/* already enabled */

992 993 994 995
	for (i = 0; i < DEVICE_COUNT_RESOURCE; i++)
		if (dev->resource[i].flags & flags)
			bars |= (1 << i);

996
	err = do_pci_enable_device(dev, bars);
997
	if (err < 0)
998
		atomic_dec(&dev->enable_cnt);
999
	return err;
L
Linus Torvalds 已提交
1000 1001
}

1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027
/**
 * pci_enable_device_io - Initialize a device for use with IO space
 * @dev: PCI device to be initialized
 *
 *  Initialize device before it's used by a driver. Ask low-level code
 *  to enable I/O resources. Wake up the device if it was suspended.
 *  Beware, this function can fail.
 */
int pci_enable_device_io(struct pci_dev *dev)
{
	return __pci_enable_device_flags(dev, IORESOURCE_IO);
}

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

1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040
/**
 * 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)
{
1041
	return __pci_enable_device_flags(dev, IORESOURCE_MEM | IORESOURCE_IO);
1042 1043
}

T
Tejun Heo 已提交
1044 1045 1046 1047 1048 1049 1050
/*
 * 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 {
1051 1052
	unsigned int enabled:1;
	unsigned int pinned:1;
T
Tejun Heo 已提交
1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075
	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);

1076
	if (this->enabled && !this->pinned)
T
Tejun Heo 已提交
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
		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;
1115 1116
	if (dr->enabled)
		return 0;
T
Tejun Heo 已提交
1117 1118 1119 1120

	rc = pci_enable_device(pdev);
	if (!rc) {
		pdev->is_managed = 1;
1121
		dr->enabled = 1;
T
Tejun Heo 已提交
1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138
	}
	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);
1139
	WARN_ON(!dr || !dr->enabled);
T
Tejun Heo 已提交
1140
	if (dr)
1141
		dr->pinned = 1;
T
Tejun Heo 已提交
1142 1143
}

L
Linus Torvalds 已提交
1144 1145 1146 1147 1148 1149 1150 1151 1152 1153
/**
 * pcibios_disable_device - disable arch specific PCI resources for device dev
 * @dev: the PCI device to disable
 *
 * Disables architecture specific PCI resources for the device. This
 * is the default implementation. Architecture implementations can
 * override this.
 */
void __attribute__ ((weak)) pcibios_disable_device (struct pci_dev *dev) {}

1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175
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)
{
1176
	if (pci_is_enabled(dev))
1177 1178 1179
		do_pci_disable_device(dev);
}

L
Linus Torvalds 已提交
1180 1181 1182 1183 1184 1185
/**
 * 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.
1186 1187 1188
 *
 * Note we don't actually disable the device until all callers of
 * pci_device_enable() have called pci_device_disable().
L
Linus Torvalds 已提交
1189 1190 1191 1192
 */
void
pci_disable_device(struct pci_dev *dev)
{
T
Tejun Heo 已提交
1193
	struct pci_devres *dr;
1194

T
Tejun Heo 已提交
1195 1196
	dr = find_pci_dr(dev);
	if (dr)
1197
		dr->enabled = 0;
T
Tejun Heo 已提交
1198

1199 1200 1201
	if (atomic_sub_return(1, &dev->enable_cnt) != 0)
		return;

1202
	do_pci_disable_device(dev);
L
Linus Torvalds 已提交
1203

1204
	dev->is_busmaster = 0;
L
Linus Torvalds 已提交
1205 1206
}

B
Brian King 已提交
1207 1208
/**
 * pcibios_set_pcie_reset_state - set reset state for device dev
1209
 * @dev: the PCIe device reset
B
Brian King 已提交
1210 1211 1212
 * @state: Reset state to enter into
 *
 *
1213
 * Sets the PCIe reset state for the device. This is the default
B
Brian King 已提交
1214 1215 1216 1217 1218 1219 1220 1221 1222 1223
 * implementation. Architecture implementations can override this.
 */
int __attribute__ ((weak)) pcibios_set_pcie_reset_state(struct pci_dev *dev,
							enum pcie_reset_state state)
{
	return -EINVAL;
}

/**
 * pci_set_pcie_reset_state - set reset state for device dev
1224
 * @dev: the PCIe device reset
B
Brian King 已提交
1225 1226 1227 1228 1229 1230 1231 1232 1233 1234
 * @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);
}

1235 1236 1237 1238 1239
/**
 * 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#.
 */
1240
bool pci_pme_capable(struct pci_dev *dev, pci_power_t state)
1241
{
1242
	if (!dev->pm_cap)
1243 1244
		return false;

1245
	return !!(dev->pme_support & (1 << state));
1246 1247 1248 1249 1250 1251 1252 1253 1254 1255
}

/**
 * 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'.
 */
1256
void pci_pme_active(struct pci_dev *dev, bool enable)
1257 1258 1259
{
	u16 pmcsr;

1260
	if (!dev->pm_cap)
1261 1262
		return;

1263
	pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
1264 1265 1266 1267 1268
	/* 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;

1269
	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, pmcsr);
1270

B
Bjorn Helgaas 已提交
1271
	dev_printk(KERN_DEBUG, &dev->dev, "PME# %s\n",
1272 1273 1274
			enable ? "enabled" : "disabled");
}

L
Linus Torvalds 已提交
1275
/**
1276 1277 1278 1279 1280 1281 1282 1283 1284 1285
 * pci_enable_wake - enable PCI device as wakeup event source
 * @dev: PCI device affected
 * @state: PCI state from which device will issue wakeup events
 * @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)
1286
 * always require such platform hooks.
1287
 *
1288 1289 1290 1291 1292
 * 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
1293
 */
1294
int pci_enable_wake(struct pci_dev *dev, pci_power_t state, bool enable)
L
Linus Torvalds 已提交
1295
{
1296
	int ret = 0;
1297

1298
	if (enable && !device_may_wakeup(&dev->dev))
1299
		return -EINVAL;
L
Linus Torvalds 已提交
1300

1301 1302 1303 1304
	/* Don't do the same thing twice in a row for one device. */
	if (!!enable == !!dev->wakeup_prepared)
		return 0;

1305 1306 1307 1308
	/*
	 * 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.
1309
	 */
L
Linus Torvalds 已提交
1310

1311 1312
	if (enable) {
		int error;
L
Linus Torvalds 已提交
1313

1314 1315 1316 1317
		if (pci_pme_capable(dev, state))
			pci_pme_active(dev, true);
		else
			ret = 1;
1318
		error = platform_pci_sleep_wake(dev, true);
1319 1320
		if (ret)
			ret = error;
1321 1322
		if (!ret)
			dev->wakeup_prepared = true;
1323 1324 1325
	} else {
		platform_pci_sleep_wake(dev, false);
		pci_pme_active(dev, false);
1326
		dev->wakeup_prepared = false;
1327
	}
L
Linus Torvalds 已提交
1328

1329
	return ret;
1330
}
L
Linus Torvalds 已提交
1331

1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352
/**
 * 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);
}

1353
/**
J
Jesse Barnes 已提交
1354 1355 1356 1357 1358 1359
 * 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.
1360
 */
1361
pci_power_t pci_target_state(struct pci_dev *dev)
1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382
{
	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;
		}
1383 1384
	} else if (!dev->pm_cap) {
		target_state = PCI_D0;
1385 1386 1387 1388 1389 1390
	} 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#.
		 */
1391 1392 1393 1394
		if (dev->pme_support) {
			while (target_state
			      && !(dev->pme_support & (1 << target_state)))
				target_state--;
1395 1396 1397
		}
	}

1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416
	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;

1417
	pci_enable_wake(dev, target_state, device_may_wakeup(&dev->dev));
1418

1419 1420 1421 1422 1423 1424 1425 1426 1427
	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
1428
 * pci_back_from_sleep - turn PCI device on during system-wide transition into working state
1429 1430 1431 1432 1433 1434 1435 1436 1437 1438
 * @dev: Device to handle.
 *
 * Disable device's sytem wake-up capability and put it into D0.
 */
int pci_back_from_sleep(struct pci_dev *dev)
{
	pci_enable_wake(dev, PCI_D0, false);
	return pci_set_power_state(dev, PCI_D0);
}

1439 1440 1441 1442 1443 1444 1445 1446
/**
 * 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
1447

1448
	dev->wakeup_prepared = false;
1449 1450
	dev->pm_cap = 0;

1451 1452 1453
	/* find PCI PM capability in list */
	pm = pci_find_capability(dev, PCI_CAP_ID_PM);
	if (!pm)
1454
		return;
1455 1456
	/* Check device's ability to generate PME# */
	pci_read_config_word(dev, pm + PCI_PM_PMC, &pmc);
1457

1458 1459 1460
	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);
1461
		return;
1462 1463
	}

1464
	dev->pm_cap = pm;
1465
	dev->d3_delay = PCI_PM_D3_WAIT;
1466 1467 1468 1469

	dev->d1_support = false;
	dev->d2_support = false;
	if (!pci_no_d1d2(dev)) {
B
Bjorn Helgaas 已提交
1470
		if (pmc & PCI_PM_CAP_D1)
1471
			dev->d1_support = true;
B
Bjorn Helgaas 已提交
1472
		if (pmc & PCI_PM_CAP_D2)
1473
			dev->d2_support = true;
B
Bjorn Helgaas 已提交
1474 1475 1476

		if (dev->d1_support || dev->d2_support)
			dev_printk(KERN_DEBUG, &dev->dev, "supports%s%s\n",
1477 1478
				   dev->d1_support ? " D1" : "",
				   dev->d2_support ? " D2" : "");
1479 1480 1481 1482
	}

	pmc &= PCI_PM_CAP_PME_MASK;
	if (pmc) {
B
Bjorn Helgaas 已提交
1483 1484
		dev_printk(KERN_DEBUG, &dev->dev,
			 "PME# supported from%s%s%s%s%s\n",
B
Bjorn Helgaas 已提交
1485 1486 1487 1488 1489
			 (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" : "");
1490
		dev->pme_support = pmc >> PCI_PM_CAP_PME_SHIFT;
1491 1492 1493 1494 1495 1496 1497
		/*
		 * 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);
		device_set_wakeup_enable(&dev->dev, false);
		/* Disable the PME# generation functionality */
1498 1499 1500
		pci_pme_active(dev, false);
	} else {
		dev->pme_support = 0;
1501
	}
L
Linus Torvalds 已提交
1502 1503
}

1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523
/**
 * platform_pci_wakeup_init - init platform wakeup if present
 * @dev: PCI device
 *
 * Some devices don't have PCI PM caps but can still generate wakeup
 * events through platform methods (like ACPI events).  If @dev supports
 * platform wakeup events, set the device flag to indicate as much.  This
 * may be redundant if the device also supports PCI PM caps, but double
 * initialization should be safe in that case.
 */
void platform_pci_wakeup_init(struct pci_dev *dev)
{
	if (!platform_pci_can_wakeup(dev))
		return;

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

1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557
/**
 * pci_add_save_buffer - allocate buffer for saving given capability registers
 * @dev: the PCI device
 * @cap: the capability to allocate the buffer for
 * @size: requested size of the buffer
 */
static int pci_add_cap_save_buffer(
	struct pci_dev *dev, char cap, unsigned int size)
{
	int pos;
	struct pci_cap_saved_state *save_state;

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

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

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

	return 0;
}

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

1558 1559
	error = pci_add_cap_save_buffer(dev, PCI_CAP_ID_EXP,
					PCI_EXP_SAVE_REGS * sizeof(u16));
1560 1561 1562 1563 1564 1565 1566 1567 1568 1569
	if (error)
		dev_err(&dev->dev,
			"unable to preallocate PCI Express save buffer\n");

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

Y
Yu Zhao 已提交
1570 1571 1572 1573 1574 1575 1576 1577 1578
/**
 * pci_enable_ari - enable ARI forwarding if hardware support it
 * @dev: the PCI device
 */
void pci_enable_ari(struct pci_dev *dev)
{
	int pos;
	u32 cap;
	u16 ctrl;
1579
	struct pci_dev *bridge;
Y
Yu Zhao 已提交
1580

1581
	if (!pci_is_pcie(dev) || dev->devfn)
Y
Yu Zhao 已提交
1582 1583
		return;

1584 1585
	pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ARI);
	if (!pos)
Y
Yu Zhao 已提交
1586 1587
		return;

1588
	bridge = dev->bus->self;
1589
	if (!bridge || !pci_is_pcie(bridge))
1590 1591
		return;

1592
	pos = pci_pcie_cap(bridge);
Y
Yu Zhao 已提交
1593 1594 1595
	if (!pos)
		return;

1596
	pci_read_config_dword(bridge, pos + PCI_EXP_DEVCAP2, &cap);
Y
Yu Zhao 已提交
1597 1598 1599
	if (!(cap & PCI_EXP_DEVCAP2_ARI))
		return;

1600
	pci_read_config_word(bridge, pos + PCI_EXP_DEVCTL2, &ctrl);
Y
Yu Zhao 已提交
1601
	ctrl |= PCI_EXP_DEVCTL2_ARI;
1602
	pci_write_config_word(bridge, pos + PCI_EXP_DEVCTL2, ctrl);
Y
Yu Zhao 已提交
1603

1604
	bridge->ari_enabled = 1;
Y
Yu Zhao 已提交
1605 1606
}

C
Chris Wright 已提交
1607 1608 1609 1610 1611 1612 1613 1614 1615 1616
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;
}

1617 1618 1619 1620 1621 1622 1623 1624 1625 1626
/**
 * 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
1627 1628 1629
	if (!pci_acs_enable)
		return;

1630
	if (!pci_is_pcie(dev))
1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654
		return;

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

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

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

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

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

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

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

1655 1656 1657 1658 1659 1660 1661
/**
 * 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
1662 1663 1664
 * 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)
1665 1666 1667
 */
u8 pci_swizzle_interrupt_pin(struct pci_dev *dev, u8 pin)
{
1668 1669 1670 1671 1672 1673 1674 1675
	int slot;

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

	return (((pin - 1) + slot) % 4) + 1;
1676 1677
}

L
Linus Torvalds 已提交
1678 1679 1680 1681 1682
int
pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge)
{
	u8 pin;

1683
	pin = dev->pin;
L
Linus Torvalds 已提交
1684 1685
	if (!pin)
		return -1;
1686

1687
	while (!pci_is_root_bus(dev->bus)) {
1688
		pin = pci_swizzle_interrupt_pin(dev, pin);
L
Linus Torvalds 已提交
1689 1690 1691 1692 1693 1694
		dev = dev->bus->self;
	}
	*bridge = dev;
	return pin;
}

1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706
/**
 * 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;

1707
	while (!pci_is_root_bus(dev->bus)) {
1708 1709 1710 1711 1712 1713 1714
		pin = pci_swizzle_interrupt_pin(dev, pin);
		dev = dev->bus->self;
	}
	*pinp = pin;
	return PCI_SLOT(dev->devfn);
}

L
Linus Torvalds 已提交
1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725
/**
 *	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
1726 1727
	struct pci_devres *dr;

L
Linus Torvalds 已提交
1728 1729 1730 1731 1732 1733 1734 1735
	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
1736 1737 1738 1739

	dr = find_pci_dr(pdev);
	if (dr)
		dr->region_mask &= ~(1 << bar);
L
Linus Torvalds 已提交
1740 1741 1742
}

/**
1743
 *	__pci_request_region - Reserved PCI I/O and memory resource
L
Linus Torvalds 已提交
1744 1745 1746
 *	@pdev: PCI device whose resources are to be reserved
 *	@bar: BAR to be reserved
 *	@res_name: Name to be associated with resource.
1747
 *	@exclusive: whether the region access is exclusive or not
L
Linus Torvalds 已提交
1748 1749 1750 1751 1752 1753
 *
 *	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.
 *
1754 1755 1756 1757
 *	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
1758 1759 1760
 *	Returns 0 on success, or %EBUSY on error.  A warning
 *	message is also printed on failure.
 */
1761 1762
static int __pci_request_region(struct pci_dev *pdev, int bar, const char *res_name,
									int exclusive)
L
Linus Torvalds 已提交
1763
{
T
Tejun Heo 已提交
1764 1765
	struct pci_devres *dr;

L
Linus Torvalds 已提交
1766 1767 1768 1769 1770 1771 1772 1773 1774
	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) {
1775 1776 1777
		if (!__request_mem_region(pci_resource_start(pdev, bar),
					pci_resource_len(pdev, bar), res_name,
					exclusive))
L
Linus Torvalds 已提交
1778 1779
			goto err_out;
	}
T
Tejun Heo 已提交
1780 1781 1782 1783 1784

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

L
Linus Torvalds 已提交
1785 1786 1787
	return 0;

err_out:
1788
	dev_warn(&pdev->dev, "BAR %d: can't reserve %pR\n", bar,
1789
		 &pdev->resource[bar]);
L
Linus Torvalds 已提交
1790 1791 1792
	return -EBUSY;
}

1793
/**
1794
 *	pci_request_region - Reserve PCI I/O and memory resource
1795 1796
 *	@pdev: PCI device whose resources are to be reserved
 *	@bar: BAR to be reserved
1797
 *	@res_name: Name to be associated with resource
1798
 *
1799
 *	Mark the PCI region associated with PCI device @pdev BAR @bar as
1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833
 *	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);
}
1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850
/**
 * 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);
}

1851 1852
int __pci_request_selected_regions(struct pci_dev *pdev, int bars,
				 const char *res_name, int excl)
1853 1854 1855 1856 1857
{
	int i;

	for (i = 0; i < 6; i++)
		if (bars & (1 << i))
1858
			if (__pci_request_region(pdev, i, res_name, excl))
1859 1860 1861 1862 1863 1864 1865 1866 1867 1868
				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
1869

1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889

/**
 * 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
1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900
/**
 *	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)
{
1901
	pci_release_selected_regions(pdev, (1 << 6) - 1);
L
Linus Torvalds 已提交
1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916
}

/**
 *	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.
 */
1917
int pci_request_regions(struct pci_dev *pdev, const char *res_name)
L
Linus Torvalds 已提交
1918
{
1919
	return pci_request_selected_regions(pdev, ((1 << 6) - 1), res_name);
L
Linus Torvalds 已提交
1920 1921
}

1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943
/**
 *	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);
}

1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959
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;
}
1960

L
Linus Torvalds 已提交
1961 1962 1963 1964 1965 1966 1967
/**
 * 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.
 */
1968
void pci_set_master(struct pci_dev *dev)
L
Linus Torvalds 已提交
1969
{
1970
	__pci_set_master(dev, true);
L
Linus Torvalds 已提交
1971 1972 1973
	pcibios_set_master(dev);
}

1974 1975 1976 1977 1978 1979 1980 1981 1982
/**
 * 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
1983
/**
1984 1985
 * 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
1986
 *
1987 1988
 * Helper function for pci_set_mwi.
 * Originally copied from drivers/net/acenic.c.
L
Linus Torvalds 已提交
1989 1990 1991 1992
 * 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
1993
int pci_set_cacheline_size(struct pci_dev *dev)
L
Linus Torvalds 已提交
1994 1995 1996 1997
{
	u8 cacheline_size;

	if (!pci_cache_line_size)
T
Tejun Heo 已提交
1998
		return -EINVAL;
L
Linus Torvalds 已提交
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

	/* 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;

2014 2015
	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
2016 2017 2018

	return -EINVAL;
}
T
Tejun Heo 已提交
2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036
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
2037 2038 2039 2040 2041

/**
 * pci_set_mwi - enables memory-write-invalidate PCI transaction
 * @dev: the PCI device for which MWI is enabled
 *
R
Randy Dunlap 已提交
2042
 * Enables the Memory-Write-Invalidate transaction in %PCI_COMMAND.
L
Linus Torvalds 已提交
2043 2044 2045 2046 2047 2048 2049 2050 2051
 *
 * RETURNS: An appropriate -ERRNO error value on error, or zero for success.
 */
int
pci_set_mwi(struct pci_dev *dev)
{
	int rc;
	u16 cmd;

2052
	rc = pci_set_cacheline_size(dev);
L
Linus Torvalds 已提交
2053 2054 2055 2056 2057
	if (rc)
		return rc;

	pci_read_config_word(dev, PCI_COMMAND, &cmd);
	if (! (cmd & PCI_COMMAND_INVALIDATE)) {
2058
		dev_dbg(&dev->dev, "enabling Mem-Wr-Inval\n");
L
Linus Torvalds 已提交
2059 2060 2061 2062 2063 2064 2065
		cmd |= PCI_COMMAND_INVALIDATE;
		pci_write_config_word(dev, PCI_COMMAND, cmd);
	}
	
	return 0;
}

R
Randy Dunlap 已提交
2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080
/**
 * 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
2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097
/**
 * 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);
	}
}
2098
#endif /* ! PCI_DISABLE_MWI */
L
Linus Torvalds 已提交
2099

B
Brett M Russ 已提交
2100 2101
/**
 * pci_intx - enables/disables PCI INTx for device dev
R
Randy Dunlap 已提交
2102 2103
 * @pdev: the PCI device to operate on
 * @enable: boolean: whether to enable or disable PCI INTx
B
Brett M Russ 已提交
2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120
 *
 * 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
2121 2122
		struct pci_devres *dr;

2123
		pci_write_config_word(pdev, PCI_COMMAND, new);
T
Tejun Heo 已提交
2124 2125 2126 2127 2128 2129

		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
2130 2131 2132
	}
}

2133 2134
/**
 * pci_msi_off - disables any msi or msix capabilities
R
Randy Dunlap 已提交
2135
 * @dev: the PCI device to operate on
2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159
 *
 * 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);
	}
}

L
Linus Torvalds 已提交
2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170
#ifndef HAVE_ARCH_PCI_SET_DMA_MASK
/*
 * These can be overridden by arch-specific implementations
 */
int
pci_set_dma_mask(struct pci_dev *dev, u64 mask)
{
	if (!pci_dma_supported(dev, mask))
		return -EIO;

	dev->dma_mask = mask;
2171
	dev_dbg(&dev->dev, "using %dbit DMA mask\n", fls64(mask));
L
Linus Torvalds 已提交
2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182

	return 0;
}
    
int
pci_set_consistent_dma_mask(struct pci_dev *dev, u64 mask)
{
	if (!pci_dma_supported(dev, mask))
		return -EIO;

	dev->dev.coherent_dma_mask = mask;
2183
	dev_dbg(&dev->dev, "using %dbit consistent DMA mask\n", fls64(mask));
L
Linus Torvalds 已提交
2184 2185 2186 2187

	return 0;
}
#endif
2188

2189 2190 2191 2192 2193 2194 2195 2196
#ifndef HAVE_ARCH_PCI_SET_DMA_MAX_SEGMENT_SIZE
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);
#endif

2197 2198 2199 2200 2201 2202 2203 2204
#ifndef HAVE_ARCH_PCI_SET_DMA_SEGMENT_BOUNDARY
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);
#endif

Y
Yu Zhao 已提交
2205
static int pcie_flr(struct pci_dev *dev, int probe)
2206
{
Y
Yu Zhao 已提交
2207 2208
	int i;
	int pos;
2209
	u32 cap;
2210
	u16 status, control;
2211

2212
	pos = pci_pcie_cap(dev);
Y
Yu Zhao 已提交
2213
	if (!pos)
2214
		return -ENOTTY;
Y
Yu Zhao 已提交
2215 2216

	pci_read_config_dword(dev, pos + PCI_EXP_DEVCAP, &cap);
2217 2218 2219
	if (!(cap & PCI_EXP_DEVCAP_FLR))
		return -ENOTTY;

S
Sheng Yang 已提交
2220 2221 2222
	if (probe)
		return 0;

2223
	/* Wait for Transaction Pending bit clean */
Y
Yu Zhao 已提交
2224 2225 2226
	for (i = 0; i < 4; i++) {
		if (i)
			msleep((1 << (i - 1)) * 100);
S
Sheng Yang 已提交
2227

Y
Yu Zhao 已提交
2228 2229 2230 2231 2232 2233 2234 2235 2236
		pci_read_config_word(dev, pos + PCI_EXP_DEVSTA, &status);
		if (!(status & PCI_EXP_DEVSTA_TRPND))
			goto clear;
	}

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

clear:
2237 2238 2239 2240
	pci_read_config_word(dev, pos + PCI_EXP_DEVCTL, &control);
	control |= PCI_EXP_DEVCTL_BCR_FLR;
	pci_write_config_word(dev, pos + PCI_EXP_DEVCTL, control);

Y
Yu Zhao 已提交
2241
	msleep(100);
2242 2243 2244

	return 0;
}
S
Sheng Yang 已提交
2245

Y
Yu Zhao 已提交
2246
static int pci_af_flr(struct pci_dev *dev, int probe)
2247
{
Y
Yu Zhao 已提交
2248 2249
	int i;
	int pos;
2250
	u8 cap;
Y
Yu Zhao 已提交
2251
	u8 status;
2252

Y
Yu Zhao 已提交
2253 2254
	pos = pci_find_capability(dev, PCI_CAP_ID_AF);
	if (!pos)
2255
		return -ENOTTY;
Y
Yu Zhao 已提交
2256 2257

	pci_read_config_byte(dev, pos + PCI_AF_CAP, &cap);
2258 2259 2260 2261 2262 2263 2264
	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
2265 2266 2267 2268 2269 2270 2271 2272
	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
2273

Y
Yu Zhao 已提交
2274 2275
	dev_err(&dev->dev, "transaction is not cleared; "
			"proceeding with reset anyway\n");
S
Sheng Yang 已提交
2276

Y
Yu Zhao 已提交
2277 2278
clear:
	pci_write_config_byte(dev, pos + PCI_AF_CTRL, PCI_AF_CTRL_FLR);
2279
	msleep(100);
Y
Yu Zhao 已提交
2280

2281 2282 2283
	return 0;
}

2284
static int pci_pm_reset(struct pci_dev *dev, int probe)
S
Sheng Yang 已提交
2285
{
2286 2287 2288 2289
	u16 csr;

	if (!dev->pm_cap)
		return -ENOTTY;
S
Sheng Yang 已提交
2290

2291 2292 2293
	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
2294

2295 2296
	if (probe)
		return 0;
2297

2298 2299 2300 2301 2302 2303
	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);
2304
	pci_dev_d3_sleep(dev);
2305 2306 2307 2308

	csr &= ~PCI_PM_CTRL_STATE_MASK;
	csr |= PCI_D0;
	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, csr);
2309
	pci_dev_d3_sleep(dev);
2310 2311 2312 2313

	return 0;
}

Y
Yu Zhao 已提交
2314 2315 2316 2317 2318
static int pci_parent_bus_reset(struct pci_dev *dev, int probe)
{
	u16 ctrl;
	struct pci_dev *pdev;

2319
	if (pci_is_root_bus(dev->bus) || dev->subordinate || !dev->bus->self)
Y
Yu Zhao 已提交
2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340
		return -ENOTTY;

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

	if (probe)
		return 0;

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

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

	return 0;
}

Y
Yu Zhao 已提交
2341
static int pci_dev_reset(struct pci_dev *dev, int probe)
S
Sheng Yang 已提交
2342
{
Y
Yu Zhao 已提交
2343 2344 2345 2346 2347 2348 2349 2350 2351
	int rc;

	might_sleep();

	if (!probe) {
		pci_block_user_cfg_access(dev);
		/* block PM suspend, driver probe, etc. */
		down(&dev->dev.sem);
	}
S
Sheng Yang 已提交
2352

2353 2354 2355 2356
	rc = pci_dev_specific_reset(dev, probe);
	if (rc != -ENOTTY)
		goto done;

Y
Yu Zhao 已提交
2357 2358 2359
	rc = pcie_flr(dev, probe);
	if (rc != -ENOTTY)
		goto done;
S
Sheng Yang 已提交
2360

Y
Yu Zhao 已提交
2361
	rc = pci_af_flr(dev, probe);
2362 2363 2364 2365
	if (rc != -ENOTTY)
		goto done;

	rc = pci_pm_reset(dev, probe);
Y
Yu Zhao 已提交
2366 2367 2368 2369
	if (rc != -ENOTTY)
		goto done;

	rc = pci_parent_bus_reset(dev, probe);
Y
Yu Zhao 已提交
2370 2371 2372 2373 2374
done:
	if (!probe) {
		up(&dev->dev.sem);
		pci_unblock_user_cfg_access(dev);
	}
2375

Y
Yu Zhao 已提交
2376
	return rc;
S
Sheng Yang 已提交
2377 2378 2379
}

/**
Y
Yu Zhao 已提交
2380 2381
 * __pci_reset_function - reset a PCI device function
 * @dev: PCI device to reset
S
Sheng Yang 已提交
2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392
 *
 * 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
2393
 * Returns 0 if the device function was successfully reset or negative if the
S
Sheng Yang 已提交
2394 2395
 * device doesn't support resetting a single function.
 */
Y
Yu Zhao 已提交
2396
int __pci_reset_function(struct pci_dev *dev)
S
Sheng Yang 已提交
2397
{
Y
Yu Zhao 已提交
2398
	return pci_dev_reset(dev, 0);
S
Sheng Yang 已提交
2399
}
Y
Yu Zhao 已提交
2400
EXPORT_SYMBOL_GPL(__pci_reset_function);
2401

2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417
/**
 * 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);
}

2418
/**
Y
Yu Zhao 已提交
2419 2420
 * pci_reset_function - quiesce and reset a PCI device function
 * @dev: PCI device to reset
2421 2422 2423 2424 2425 2426 2427
 *
 * 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
2428
 * from __pci_reset_function in that it saves and restores device state
2429 2430
 * over the reset.
 *
Y
Yu Zhao 已提交
2431
 * Returns 0 if the device function was successfully reset or negative if the
2432 2433 2434 2435
 * device doesn't support resetting a single function.
 */
int pci_reset_function(struct pci_dev *dev)
{
Y
Yu Zhao 已提交
2436
	int rc;
2437

Y
Yu Zhao 已提交
2438 2439 2440
	rc = pci_dev_reset(dev, 1);
	if (rc)
		return rc;
2441 2442 2443

	pci_save_state(dev);

Y
Yu Zhao 已提交
2444 2445 2446 2447
	/*
	 * both INTx and MSI are disabled after the Interrupt Disable bit
	 * is set and the Bus Master bit is cleared.
	 */
2448 2449
	pci_write_config_word(dev, PCI_COMMAND, PCI_COMMAND_INTX_DISABLE);

Y
Yu Zhao 已提交
2450
	rc = pci_dev_reset(dev, 0);
2451 2452 2453

	pci_restore_state(dev);

Y
Yu Zhao 已提交
2454
	return rc;
2455 2456 2457
}
EXPORT_SYMBOL_GPL(pci_reset_function);

2458 2459 2460 2461 2462 2463 2464 2465 2466
/**
 * 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)
{
2467
	int err, cap;
2468 2469 2470 2471 2472 2473 2474 2475 2476 2477
	u32 stat;

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

	err = pci_read_config_dword(dev, cap + PCI_X_STATUS, &stat);
	if (err)
		return -EINVAL;

2478
	return (stat & PCI_X_STATUS_MAX_READ) >> 12;
2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519
}
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)
{
	int ret, cap;
	u32 cmd;

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

	ret = pci_read_config_dword(dev, cap + PCI_X_CMD, &cmd);
	if (!ret)
		ret = 512 << ((cmd & PCI_X_CMD_MAX_READ) >> 2);

	return ret;
}
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)
{
	int cap, err = -EINVAL;
	u32 stat, cmd, v, o;

2520
	if (mmrbc < 512 || mmrbc > 4096 || !is_power_of_2(mmrbc))
2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566
		goto out;

	v = ffs(mmrbc) - 10;

	cap = pci_find_capability(dev, PCI_CAP_ID_PCIX);
	if (!cap)
		goto out;

	err = pci_read_config_dword(dev, cap + PCI_X_STATUS, &stat);
	if (err)
		goto out;

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

	err = pci_read_config_dword(dev, cap + PCI_X_CMD, &cmd);
	if (err)
		goto out;

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

		cmd &= ~PCI_X_CMD_MAX_READ;
		cmd |= v << 2;
		err = pci_write_config_dword(dev, cap + PCI_X_CMD, cmd);
	}
out:
	return err;
}
EXPORT_SYMBOL(pcix_set_mmrbc);

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

2567
	cap = pci_pcie_cap(dev);
2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581
	if (!cap)
		return -EINVAL;

	ret = pci_read_config_word(dev, cap + PCI_EXP_DEVCTL, &ctl);
	if (!ret)
	ret = 128 << ((ctl & PCI_EXP_DEVCTL_READRQ) >> 12);

	return ret;
}
EXPORT_SYMBOL(pcie_get_readrq);

/**
 * pcie_set_readrq - set PCI Express maximum memory read request
 * @dev: PCI device to query
2582
 * @rq: maximum memory read count in bytes
2583 2584 2585 2586 2587 2588 2589 2590 2591
 *    valid values are 128, 256, 512, 1024, 2048, 4096
 *
 * If possible sets maximum read byte count
 */
int pcie_set_readrq(struct pci_dev *dev, int rq)
{
	int cap, err = -EINVAL;
	u16 ctl, v;

2592
	if (rq < 128 || rq > 4096 || !is_power_of_2(rq))
2593 2594 2595 2596
		goto out;

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

2597
	cap = pci_pcie_cap(dev);
2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615
	if (!cap)
		goto out;

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

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

out:
	return err;
}
EXPORT_SYMBOL(pcie_set_readrq);

2616 2617
/**
 * pci_select_bars - Make BAR mask from the type of resource
2618
 * @dev: the PCI device for which BAR mask is made
2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631
 * @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;
}

2632 2633 2634 2635 2636 2637 2638 2639 2640 2641
/**
 * 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)
{
2642 2643
	int reg;

2644 2645 2646 2647 2648 2649
	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;
2650 2651 2652 2653 2654
	} else if (resno < PCI_BRIDGE_RESOURCES) {
		/* device specific resource */
		reg = pci_iov_resource_bar(dev, resno, type);
		if (reg)
			return reg;
2655 2656
	}

2657
	dev_err(&dev->dev, "BAR %d: invalid resource\n", resno);
2658 2659 2660
	return 0;
}

2661 2662
/**
 * pci_set_vga_state - set VGA decode state on device and parents if requested
R
Randy Dunlap 已提交
2663 2664 2665 2666
 * @dev: the PCI device
 * @decode: true = enable decoding, false = disable decoding
 * @command_bits: PCI_COMMAND_IO and/or PCI_COMMAND_MEMORY
 * @change_bridge: traverse ancestors and change bridges
2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704
 */
int pci_set_vga_state(struct pci_dev *dev, bool decode,
		      unsigned int command_bits, bool change_bridge)
{
	struct pci_bus *bus;
	struct pci_dev *bridge;
	u16 cmd;

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

	pci_read_config_word(dev, PCI_COMMAND, &cmd);
	if (decode == true)
		cmd |= command_bits;
	else
		cmd &= ~command_bits;
	pci_write_config_word(dev, PCI_COMMAND, cmd);

	if (change_bridge == false)
		return 0;

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

2705 2706
#define RESOURCE_ALIGNMENT_PARAM_SIZE COMMAND_LINE_SIZE
static char resource_alignment_param[RESOURCE_ALIGNMENT_PARAM_SIZE] = {0};
2707
static DEFINE_SPINLOCK(resource_alignment_lock);
2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819

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

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

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

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

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

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

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

BUS_ATTR(resource_alignment, 0644, pci_resource_alignment_show,
					pci_resource_alignment_store);

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

late_initcall(pci_resource_alignment_sysfs_init);

2820 2821 2822 2823 2824 2825 2826
static void __devinit pci_no_domains(void)
{
#ifdef CONFIG_PCI_DOMAINS
	pci_domains_supported = 0;
#endif
}

2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839
/**
 * pci_ext_cfg_enabled - can we access extended PCI config space?
 * @dev: The PCI device of the root bridge.
 *
 * Returns 1 if we can access PCI extended config space (offsets
 * greater than 0xff). This is the default implementation. Architecture
 * implementations can override this.
 */
int __attribute__ ((weak)) pci_ext_cfg_avail(struct pci_dev *dev)
{
	return 1;
}

2840 2841 2842 2843 2844
void __weak pci_fixup_cardbus(struct pci_bus *bus)
{
}
EXPORT_SYMBOL(pci_fixup_cardbus);

A
Al Viro 已提交
2845
static int __init pci_setup(char *str)
L
Linus Torvalds 已提交
2846 2847 2848 2849 2850 2851
{
	while (str) {
		char *k = strchr(str, ',');
		if (k)
			*k++ = 0;
		if (*str && (str = pcibios_setup(str)) && *str) {
2852 2853
			if (!strcmp(str, "nomsi")) {
				pci_no_msi();
R
Randy Dunlap 已提交
2854 2855
			} else if (!strcmp(str, "noaer")) {
				pci_no_aer();
2856 2857
			} else if (!strcmp(str, "nodomains")) {
				pci_no_domains();
2858 2859 2860 2861
			} 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);
2862 2863 2864
			} else if (!strncmp(str, "resource_alignment=", 19)) {
				pci_set_resource_alignment_param(str + 19,
							strlen(str + 19));
2865 2866
			} else if (!strncmp(str, "ecrc=", 5)) {
				pcie_ecrc_get_policy(str + 5);
2867 2868 2869 2870
			} 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);
2871 2872 2873 2874
			} else {
				printk(KERN_ERR "PCI: Unknown option `%s'\n",
						str);
			}
L
Linus Torvalds 已提交
2875 2876 2877
		}
		str = k;
	}
2878
	return 0;
L
Linus Torvalds 已提交
2879
}
2880
early_param("pci", pci_setup);
L
Linus Torvalds 已提交
2881

2882
EXPORT_SYMBOL(pci_reenable_device);
2883 2884
EXPORT_SYMBOL(pci_enable_device_io);
EXPORT_SYMBOL(pci_enable_device_mem);
L
Linus Torvalds 已提交
2885
EXPORT_SYMBOL(pci_enable_device);
T
Tejun Heo 已提交
2886 2887
EXPORT_SYMBOL(pcim_enable_device);
EXPORT_SYMBOL(pcim_pin_device);
L
Linus Torvalds 已提交
2888 2889 2890 2891 2892
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);
2893
EXPORT_SYMBOL(pci_request_regions_exclusive);
L
Linus Torvalds 已提交
2894 2895
EXPORT_SYMBOL(pci_release_region);
EXPORT_SYMBOL(pci_request_region);
2896
EXPORT_SYMBOL(pci_request_region_exclusive);
2897 2898
EXPORT_SYMBOL(pci_release_selected_regions);
EXPORT_SYMBOL(pci_request_selected_regions);
2899
EXPORT_SYMBOL(pci_request_selected_regions_exclusive);
L
Linus Torvalds 已提交
2900
EXPORT_SYMBOL(pci_set_master);
2901
EXPORT_SYMBOL(pci_clear_master);
L
Linus Torvalds 已提交
2902
EXPORT_SYMBOL(pci_set_mwi);
R
Randy Dunlap 已提交
2903
EXPORT_SYMBOL(pci_try_set_mwi);
L
Linus Torvalds 已提交
2904
EXPORT_SYMBOL(pci_clear_mwi);
B
Brett M Russ 已提交
2905
EXPORT_SYMBOL_GPL(pci_intx);
L
Linus Torvalds 已提交
2906 2907 2908 2909
EXPORT_SYMBOL(pci_set_dma_mask);
EXPORT_SYMBOL(pci_set_consistent_dma_mask);
EXPORT_SYMBOL(pci_assign_resource);
EXPORT_SYMBOL(pci_find_parent_resource);
2910
EXPORT_SYMBOL(pci_select_bars);
L
Linus Torvalds 已提交
2911 2912 2913 2914

EXPORT_SYMBOL(pci_set_power_state);
EXPORT_SYMBOL(pci_save_state);
EXPORT_SYMBOL(pci_restore_state);
2915
EXPORT_SYMBOL(pci_pme_capable);
2916
EXPORT_SYMBOL(pci_pme_active);
L
Linus Torvalds 已提交
2917
EXPORT_SYMBOL(pci_enable_wake);
2918
EXPORT_SYMBOL(pci_wake_from_d3);
2919
EXPORT_SYMBOL(pci_target_state);
2920 2921
EXPORT_SYMBOL(pci_prepare_to_sleep);
EXPORT_SYMBOL(pci_back_from_sleep);
B
Brian King 已提交
2922
EXPORT_SYMBOL_GPL(pci_set_pcie_reset_state);
L
Linus Torvalds 已提交
2923