msi.c 38.7 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6
/*
 * File:	msi.c
 * Purpose:	PCI Message Signaled Interrupt (MSI)
 *
 * Copyright (C) 2003-2004 Intel
 * Copyright (C) Tom Long Nguyen (tom.l.nguyen@intel.com)
7
 * Copyright (C) 2016 Christoph Hellwig.
L
Linus Torvalds 已提交
8 9
 */

10
#include <linux/err.h>
L
Linus Torvalds 已提交
11 12 13
#include <linux/mm.h>
#include <linux/irq.h>
#include <linux/interrupt.h>
14
#include <linux/export.h>
L
Linus Torvalds 已提交
15 16 17
#include <linux/ioport.h>
#include <linux/pci.h>
#include <linux/proc_fs.h>
18
#include <linux/msi.h>
D
Dan Williams 已提交
19
#include <linux/smp.h>
H
Hidetoshi Seto 已提交
20 21
#include <linux/errno.h>
#include <linux/io.h>
22
#include <linux/slab.h>
23
#include <linux/irqdomain.h>
24
#include <linux/of_irq.h>
L
Linus Torvalds 已提交
25 26 27 28

#include "pci.h"

static int pci_msi_enable = 1;
29
int pci_msi_ignore_mask;
L
Linus Torvalds 已提交
30

31 32
#define msix_table_size(flags)	((flags & PCI_MSIX_FLAGS_QSIZE) + 1)

33 34 35 36 37 38 39 40 41
#ifdef CONFIG_PCI_MSI_IRQ_DOMAIN
static struct irq_domain *pci_msi_default_domain;
static DEFINE_MUTEX(pci_msi_domain_lock);

struct irq_domain * __weak arch_get_pci_msi_domain(struct pci_dev *dev)
{
	return pci_msi_default_domain;
}

42 43
static struct irq_domain *pci_msi_get_domain(struct pci_dev *dev)
{
44
	struct irq_domain *domain;
45

46 47 48
	domain = dev_get_msi_domain(&dev->dev);
	if (domain)
		return domain;
49

50
	return arch_get_pci_msi_domain(dev);
51 52
}

53 54 55 56
static int pci_msi_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
{
	struct irq_domain *domain;

57
	domain = pci_msi_get_domain(dev);
58
	if (domain && irq_domain_is_hierarchy(domain))
59 60 61 62 63 64 65 66 67
		return pci_msi_domain_alloc_irqs(domain, dev, nvec, type);

	return arch_setup_msi_irqs(dev, nvec, type);
}

static void pci_msi_teardown_msi_irqs(struct pci_dev *dev)
{
	struct irq_domain *domain;

68
	domain = pci_msi_get_domain(dev);
69
	if (domain && irq_domain_is_hierarchy(domain))
70 71 72 73 74 75 76 77
		pci_msi_domain_free_irqs(domain, dev);
	else
		arch_teardown_msi_irqs(dev);
}
#else
#define pci_msi_setup_msi_irqs		arch_setup_msi_irqs
#define pci_msi_teardown_msi_irqs	arch_teardown_msi_irqs
#endif
78

79 80
/* Arch hooks */

81 82
int __weak arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc)
{
83
	struct msi_controller *chip = dev->bus->msi;
84 85 86 87 88 89 90 91 92 93 94 95
	int err;

	if (!chip || !chip->setup_irq)
		return -EINVAL;

	err = chip->setup_irq(chip, dev, desc);
	if (err < 0)
		return err;

	irq_set_chip_data(desc->irq, chip);

	return 0;
96 97 98
}

void __weak arch_teardown_msi_irq(unsigned int irq)
99
{
100
	struct msi_controller *chip = irq_get_chip_data(irq);
101 102 103 104 105

	if (!chip || !chip->teardown_irq)
		return;

	chip->teardown_irq(chip, irq);
106 107
}

108
int __weak arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
109
{
110
	struct msi_controller *chip = dev->bus->msi;
111 112 113
	struct msi_desc *entry;
	int ret;

114 115
	if (chip && chip->setup_irqs)
		return chip->setup_irqs(chip, dev, nvec, type);
116 117 118 119 120 121 122
	/*
	 * If an architecture wants to support multiple MSI, it needs to
	 * override arch_setup_msi_irqs()
	 */
	if (type == PCI_CAP_ID_MSI && nvec > 1)
		return 1;

123
	for_each_pci_msi_entry(entry, dev) {
124
		ret = arch_setup_msi_irq(dev, entry);
125
		if (ret < 0)
126
			return ret;
127 128
		if (ret > 0)
			return -ENOSPC;
129 130 131 132
	}

	return 0;
}
133

134 135 136 137
/*
 * We have a default implementation available as a separate non-weak
 * function, as it is used by the Xen x86 PCI code
 */
138
void default_teardown_msi_irqs(struct pci_dev *dev)
139
{
140
	int i;
141 142
	struct msi_desc *entry;

143
	for_each_pci_msi_entry(entry, dev)
144 145 146
		if (entry->irq)
			for (i = 0; i < entry->nvec_used; i++)
				arch_teardown_msi_irq(entry->irq + i);
147 148
}

149 150 151 152
void __weak arch_teardown_msi_irqs(struct pci_dev *dev)
{
	return default_teardown_msi_irqs(dev);
}
153

154
static void default_restore_msi_irq(struct pci_dev *dev, int irq)
155 156 157 158 159
{
	struct msi_desc *entry;

	entry = NULL;
	if (dev->msix_enabled) {
160
		for_each_pci_msi_entry(entry, dev) {
161 162 163 164 165 166 167 168
			if (irq == entry->irq)
				break;
		}
	} else if (dev->msi_enabled)  {
		entry = irq_get_msi_desc(irq);
	}

	if (entry)
169
		__pci_write_msi_msg(entry, &entry->msg);
170
}
171

172
void __weak arch_restore_msi_irqs(struct pci_dev *dev)
173
{
174
	return default_restore_msi_irqs(dev);
175
}
176

177 178
static inline __attribute_const__ u32 msi_mask(unsigned x)
{
179 180 181 182
	/* Don't shift by >= width of type */
	if (x >= 5)
		return 0xffffffff;
	return (1 << (1 << x)) - 1;
183 184
}

185 186 187 188 189 190
/*
 * PCI 2.3 does not specify mask bits for each MSI interrupt.  Attempting to
 * mask all MSI interrupts by clearing the MSI enable bit does not work
 * reliably as devices without an INTx disable bit will then generate a
 * level IRQ which will never be cleared.
 */
191
u32 __pci_msi_desc_mask_irq(struct msi_desc *desc, u32 mask, u32 flag)
L
Linus Torvalds 已提交
192
{
193
	u32 mask_bits = desc->masked;
L
Linus Torvalds 已提交
194

195
	if (pci_msi_ignore_mask || !desc->msi_attrib.maskbit)
196
		return 0;
197 198 199

	mask_bits &= ~mask;
	mask_bits |= flag;
200 201
	pci_write_config_dword(msi_desc_to_pci_dev(desc), desc->mask_pos,
			       mask_bits);
202 203 204 205 206 207

	return mask_bits;
}

static void msi_mask_irq(struct msi_desc *desc, u32 mask, u32 flag)
{
208
	desc->masked = __pci_msi_desc_mask_irq(desc, mask, flag);
209 210
}

211 212 213 214 215 216
static void __iomem *pci_msix_desc_addr(struct msi_desc *desc)
{
	return desc->mask_base +
		desc->msi_attrib.entry_nr * PCI_MSIX_ENTRY_SIZE;
}

217 218 219 220 221 222 223
/*
 * This internal function does not flush PCI writes to the device.
 * All users must ensure that they read from the device before either
 * assuming that the device state is up to date, or returning out of this
 * file.  This saves a few milliseconds when initialising devices with lots
 * of MSI-X interrupts.
 */
224
u32 __pci_msix_desc_mask_irq(struct msi_desc *desc, u32 flag)
225 226
{
	u32 mask_bits = desc->masked;
227 228 229 230

	if (pci_msi_ignore_mask)
		return 0;

231 232 233
	mask_bits &= ~PCI_MSIX_ENTRY_CTRL_MASKBIT;
	if (flag)
		mask_bits |= PCI_MSIX_ENTRY_CTRL_MASKBIT;
234
	writel(mask_bits, pci_msix_desc_addr(desc) + PCI_MSIX_ENTRY_VECTOR_CTRL);
235 236 237 238 239 240

	return mask_bits;
}

static void msix_mask_irq(struct msi_desc *desc, u32 flag)
{
241
	desc->masked = __pci_msix_desc_mask_irq(desc, flag);
242
}
243

244
static void msi_set_mask_bit(struct irq_data *data, u32 flag)
245
{
246
	struct msi_desc *desc = irq_data_get_msi_desc(data);
247

248 249 250 251
	if (desc->msi_attrib.is_msix) {
		msix_mask_irq(desc, flag);
		readl(desc->mask_base);		/* Flush write to device */
	} else {
252
		unsigned offset = data->irq - desc->irq;
253
		msi_mask_irq(desc, 1 << offset, flag << offset);
L
Linus Torvalds 已提交
254
	}
255 256
}

257 258 259 260 261
/**
 * pci_msi_mask_irq - Generic irq chip callback to mask PCI/MSI interrupts
 * @data:	pointer to irqdata associated to that interrupt
 */
void pci_msi_mask_irq(struct irq_data *data)
262
{
263
	msi_set_mask_bit(data, 1);
264
}
265
EXPORT_SYMBOL_GPL(pci_msi_mask_irq);
266

267 268 269 270 271
/**
 * pci_msi_unmask_irq - Generic irq chip callback to unmask PCI/MSI interrupts
 * @data:	pointer to irqdata associated to that interrupt
 */
void pci_msi_unmask_irq(struct irq_data *data)
272
{
273
	msi_set_mask_bit(data, 0);
L
Linus Torvalds 已提交
274
}
275
EXPORT_SYMBOL_GPL(pci_msi_unmask_irq);
L
Linus Torvalds 已提交
276

277 278 279 280
void default_restore_msi_irqs(struct pci_dev *dev)
{
	struct msi_desc *entry;

281
	for_each_pci_msi_entry(entry, dev)
282 283 284
		default_restore_msi_irq(dev, entry->irq);
}

285
void __pci_read_msi_msg(struct msi_desc *entry, struct msi_msg *msg)
L
Linus Torvalds 已提交
286
{
287 288 289
	struct pci_dev *dev = msi_desc_to_pci_dev(entry);

	BUG_ON(dev->current_state != PCI_D0);
290 291

	if (entry->msi_attrib.is_msix) {
292
		void __iomem *base = pci_msix_desc_addr(entry);
293 294 295 296 297

		msg->address_lo = readl(base + PCI_MSIX_ENTRY_LOWER_ADDR);
		msg->address_hi = readl(base + PCI_MSIX_ENTRY_UPPER_ADDR);
		msg->data = readl(base + PCI_MSIX_ENTRY_DATA);
	} else {
298
		int pos = dev->msi_cap;
299 300
		u16 data;

301 302
		pci_read_config_dword(dev, pos + PCI_MSI_ADDRESS_LO,
				      &msg->address_lo);
303
		if (entry->msi_attrib.is_64) {
304 305
			pci_read_config_dword(dev, pos + PCI_MSI_ADDRESS_HI,
					      &msg->address_hi);
B
Bjorn Helgaas 已提交
306
			pci_read_config_word(dev, pos + PCI_MSI_DATA_64, &data);
307 308
		} else {
			msg->address_hi = 0;
B
Bjorn Helgaas 已提交
309
			pci_read_config_word(dev, pos + PCI_MSI_DATA_32, &data);
310 311 312 313 314
		}
		msg->data = data;
	}
}

315
void __pci_write_msi_msg(struct msi_desc *entry, struct msi_msg *msg)
Y
Yinghai Lu 已提交
316
{
317 318 319
	struct pci_dev *dev = msi_desc_to_pci_dev(entry);

	if (dev->current_state != PCI_D0) {
320 321
		/* Don't touch the hardware now */
	} else if (entry->msi_attrib.is_msix) {
322
		void __iomem *base = pci_msix_desc_addr(entry);
323

324 325 326
		writel(msg->address_lo, base + PCI_MSIX_ENTRY_LOWER_ADDR);
		writel(msg->address_hi, base + PCI_MSIX_ENTRY_UPPER_ADDR);
		writel(msg->data, base + PCI_MSIX_ENTRY_DATA);
327
	} else {
328
		int pos = dev->msi_cap;
329 330
		u16 msgctl;

331
		pci_read_config_word(dev, pos + PCI_MSI_FLAGS, &msgctl);
332 333
		msgctl &= ~PCI_MSI_FLAGS_QSIZE;
		msgctl |= entry->msi_attrib.multiple << 4;
334
		pci_write_config_word(dev, pos + PCI_MSI_FLAGS, msgctl);
335

336 337
		pci_write_config_dword(dev, pos + PCI_MSI_ADDRESS_LO,
				       msg->address_lo);
338
		if (entry->msi_attrib.is_64) {
339 340
			pci_write_config_dword(dev, pos + PCI_MSI_ADDRESS_HI,
					       msg->address_hi);
B
Bjorn Helgaas 已提交
341 342
			pci_write_config_word(dev, pos + PCI_MSI_DATA_64,
					      msg->data);
343
		} else {
B
Bjorn Helgaas 已提交
344 345
			pci_write_config_word(dev, pos + PCI_MSI_DATA_32,
					      msg->data);
346
		}
L
Linus Torvalds 已提交
347
	}
348
	entry->msg = *msg;
L
Linus Torvalds 已提交
349
}
350

351
void pci_write_msi_msg(unsigned int irq, struct msi_msg *msg)
Y
Yinghai Lu 已提交
352
{
353
	struct msi_desc *entry = irq_get_msi_desc(irq);
Y
Yinghai Lu 已提交
354

355
	__pci_write_msi_msg(entry, msg);
Y
Yinghai Lu 已提交
356
}
357
EXPORT_SYMBOL_GPL(pci_write_msi_msg);
Y
Yinghai Lu 已提交
358

359 360
static void free_msi_irqs(struct pci_dev *dev)
{
361
	struct list_head *msi_list = dev_to_msi_list(&dev->dev);
362
	struct msi_desc *entry, *tmp;
363 364
	struct attribute **msi_attrs;
	struct device_attribute *dev_attr;
365
	int i, count = 0;
366

367
	for_each_pci_msi_entry(entry, dev)
368 369 370
		if (entry->irq)
			for (i = 0; i < entry->nvec_used; i++)
				BUG_ON(irq_has_action(entry->irq + i));
371

372
	pci_msi_teardown_msi_irqs(dev);
373

374
	list_for_each_entry_safe(entry, tmp, msi_list, list) {
375
		if (entry->msi_attrib.is_msix) {
376
			if (list_is_last(&entry->list, msi_list))
377 378
				iounmap(entry->mask_base);
		}
379

380 381 382
		list_del(&entry->list);
		kfree(entry);
	}
383 384 385 386

	if (dev->msi_irq_groups) {
		sysfs_remove_groups(&dev->dev.kobj, dev->msi_irq_groups);
		msi_attrs = dev->msi_irq_groups[0]->attrs;
387
		while (msi_attrs[count]) {
388 389 390 391 392 393 394 395 396 397 398
			dev_attr = container_of(msi_attrs[count],
						struct device_attribute, attr);
			kfree(dev_attr->attr.name);
			kfree(dev_attr);
			++count;
		}
		kfree(msi_attrs);
		kfree(dev->msi_irq_groups[0]);
		kfree(dev->msi_irq_groups);
		dev->msi_irq_groups = NULL;
	}
399
}
S
Satoru Takeuchi 已提交
400

401 402 403 404 405 406
static void pci_intx_for_msi(struct pci_dev *dev, int enable)
{
	if (!(dev->dev_flags & PCI_DEV_FLAGS_MSI_INTX_DISABLE_BUG))
		pci_intx(dev, enable);
}

407
static void __pci_restore_msi_state(struct pci_dev *dev)
408 409
{
	u16 control;
410
	struct msi_desc *entry;
411

412 413 414
	if (!dev->msi_enabled)
		return;

415
	entry = irq_get_msi_desc(dev->irq);
416

417
	pci_intx_for_msi(dev, 0);
418
	pci_msi_set_enable(dev, 0);
419
	arch_restore_msi_irqs(dev);
420

421
	pci_read_config_word(dev, dev->msi_cap + PCI_MSI_FLAGS, &control);
422 423
	msi_mask_irq(entry, msi_mask(entry->msi_attrib.multi_cap),
		     entry->masked);
424
	control &= ~PCI_MSI_FLAGS_QSIZE;
425
	control |= (entry->msi_attrib.multiple << 4) | PCI_MSI_FLAGS_ENABLE;
426
	pci_write_config_word(dev, dev->msi_cap + PCI_MSI_FLAGS, control);
427 428 429
}

static void __pci_restore_msix_state(struct pci_dev *dev)
430 431 432
{
	struct msi_desc *entry;

E
Eric W. Biederman 已提交
433 434
	if (!dev->msix_enabled)
		return;
435
	BUG_ON(list_empty(dev_to_msi_list(&dev->dev)));
E
Eric W. Biederman 已提交
436

437
	/* route the table */
438
	pci_intx_for_msi(dev, 0);
439
	pci_msix_clear_and_set_ctrl(dev, 0,
440
				PCI_MSIX_FLAGS_ENABLE | PCI_MSIX_FLAGS_MASKALL);
441

442
	arch_restore_msi_irqs(dev);
443
	for_each_pci_msi_entry(entry, dev)
444
		msix_mask_irq(entry, entry->masked);
445

446
	pci_msix_clear_and_set_ctrl(dev, PCI_MSIX_FLAGS_MASKALL, 0);
447
}
448 449 450 451 452 453

void pci_restore_msi_state(struct pci_dev *dev)
{
	__pci_restore_msi_state(dev);
	__pci_restore_msix_state(dev);
}
454
EXPORT_SYMBOL_GPL(pci_restore_msi_state);
455

456
static ssize_t msi_mode_show(struct device *dev, struct device_attribute *attr,
457 458
			     char *buf)
{
459 460 461
	struct msi_desc *entry;
	unsigned long irq;
	int retval;
462

463 464 465
	retval = kstrtoul(attr->attr.name, 10, &irq);
	if (retval)
		return retval;
466

467 468 469 470 471
	entry = irq_get_msi_desc(irq);
	if (entry)
		return sprintf(buf, "%s\n",
				entry->msi_attrib.is_msix ? "msix" : "msi");

472
	return -ENODEV;
473 474 475 476
}

static int populate_msi_sysfs(struct pci_dev *pdev)
{
477 478 479 480 481
	struct attribute **msi_attrs;
	struct attribute *msi_attr;
	struct device_attribute *msi_dev_attr;
	struct attribute_group *msi_irq_group;
	const struct attribute_group **msi_irq_groups;
482
	struct msi_desc *entry;
483 484
	int ret = -ENOMEM;
	int num_msi = 0;
485
	int count = 0;
486
	int i;
487

488
	/* Determine how many msi entries we have */
489
	for_each_pci_msi_entry(entry, pdev)
490
		num_msi += entry->nvec_used;
491 492
	if (!num_msi)
		return 0;
493

494 495 496 497
	/* Dynamically create the MSI attributes for the PCI device */
	msi_attrs = kzalloc(sizeof(void *) * (num_msi + 1), GFP_KERNEL);
	if (!msi_attrs)
		return -ENOMEM;
498
	for_each_pci_msi_entry(entry, pdev) {
499 500 501 502 503 504 505 506 507 508 509 510 511 512 513
		for (i = 0; i < entry->nvec_used; i++) {
			msi_dev_attr = kzalloc(sizeof(*msi_dev_attr), GFP_KERNEL);
			if (!msi_dev_attr)
				goto error_attrs;
			msi_attrs[count] = &msi_dev_attr->attr;

			sysfs_attr_init(&msi_dev_attr->attr);
			msi_dev_attr->attr.name = kasprintf(GFP_KERNEL, "%d",
							    entry->irq + i);
			if (!msi_dev_attr->attr.name)
				goto error_attrs;
			msi_dev_attr->attr.mode = S_IRUGO;
			msi_dev_attr->show = msi_mode_show;
			++count;
		}
514 515
	}

516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531
	msi_irq_group = kzalloc(sizeof(*msi_irq_group), GFP_KERNEL);
	if (!msi_irq_group)
		goto error_attrs;
	msi_irq_group->name = "msi_irqs";
	msi_irq_group->attrs = msi_attrs;

	msi_irq_groups = kzalloc(sizeof(void *) * 2, GFP_KERNEL);
	if (!msi_irq_groups)
		goto error_irq_group;
	msi_irq_groups[0] = msi_irq_group;

	ret = sysfs_create_groups(&pdev->dev.kobj, msi_irq_groups);
	if (ret)
		goto error_irq_groups;
	pdev->msi_irq_groups = msi_irq_groups;

532 533
	return 0;

534 535 536 537 538 539 540 541 542 543 544 545 546
error_irq_groups:
	kfree(msi_irq_groups);
error_irq_group:
	kfree(msi_irq_group);
error_attrs:
	count = 0;
	msi_attr = msi_attrs[count];
	while (msi_attr) {
		msi_dev_attr = container_of(msi_attr, struct device_attribute, attr);
		kfree(msi_attr->name);
		kfree(msi_dev_attr);
		++count;
		msi_attr = msi_attrs[count];
547
	}
548
	kfree(msi_attrs);
549 550 551
	return ret;
}

552
static struct msi_desc *msi_setup_entry(struct pci_dev *dev, int nvec)
553 554 555 556 557
{
	u16 control;
	struct msi_desc *entry;

	/* MSI Entry Initialization */
558
	entry = alloc_msi_entry(&dev->dev, nvec, NULL);
559 560 561 562 563 564 565 566 567 568 569
	if (!entry)
		return NULL;

	pci_read_config_word(dev, dev->msi_cap + PCI_MSI_FLAGS, &control);

	entry->msi_attrib.is_msix	= 0;
	entry->msi_attrib.is_64		= !!(control & PCI_MSI_FLAGS_64BIT);
	entry->msi_attrib.entry_nr	= 0;
	entry->msi_attrib.maskbit	= !!(control & PCI_MSI_FLAGS_MASKBIT);
	entry->msi_attrib.default_irq	= dev->irq;	/* Save IOAPIC IRQ */
	entry->msi_attrib.multi_cap	= (control & PCI_MSI_FLAGS_QMASK) >> 1;
570
	entry->msi_attrib.multiple	= ilog2(__roundup_pow_of_two(nvec));
571
	entry->affinity			= dev->irq_affinity;
572 573 574 575 576 577 578 579 580 581 582 583 584

	if (control & PCI_MSI_FLAGS_64BIT)
		entry->mask_pos = dev->msi_cap + PCI_MSI_MASK_64;
	else
		entry->mask_pos = dev->msi_cap + PCI_MSI_MASK_32;

	/* Save the initial mask status */
	if (entry->msi_attrib.maskbit)
		pci_read_config_dword(dev, entry->mask_pos, &entry->masked);

	return entry;
}

585 586 587 588
static int msi_verify_entries(struct pci_dev *dev)
{
	struct msi_desc *entry;

589
	for_each_pci_msi_entry(entry, dev) {
590 591 592 593 594 595 596 597 598
		if (!dev->no_64bit_msi || !entry->msg.address_hi)
			continue;
		dev_err(&dev->dev, "Device has broken 64-bit MSI but arch"
			" tried to assign one above 4G\n");
		return -EIO;
	}
	return 0;
}

L
Linus Torvalds 已提交
599 600 601
/**
 * msi_capability_init - configure device's MSI capability structure
 * @dev: pointer to the pci_dev data structure of MSI device function
602
 * @nvec: number of interrupts to allocate
L
Linus Torvalds 已提交
603
 *
604 605 606 607 608 609 610
 * Setup the MSI capability structure of the device with the requested
 * number of interrupts.  A return value of zero indicates the successful
 * setup of an entry with the new MSI irq.  A negative return value indicates
 * an error, and a positive return value indicates the number of interrupts
 * which could have been allocated.
 */
static int msi_capability_init(struct pci_dev *dev, int nvec)
L
Linus Torvalds 已提交
611 612
{
	struct msi_desc *entry;
613
	int ret;
614
	unsigned mask;
L
Linus Torvalds 已提交
615

616
	pci_msi_set_enable(dev, 0);	/* Disable MSI during set up */
617

618
	entry = msi_setup_entry(dev, nvec);
619 620
	if (!entry)
		return -ENOMEM;
621

622
	/* All MSIs are unmasked by default, Mask them all */
623
	mask = msi_mask(entry->msi_attrib.multi_cap);
624 625
	msi_mask_irq(entry, mask, mask);

626
	list_add_tail(&entry->list, dev_to_msi_list(&dev->dev));
627

L
Linus Torvalds 已提交
628
	/* Configure MSI capability structure */
629
	ret = pci_msi_setup_msi_irqs(dev, nvec, PCI_CAP_ID_MSI);
630
	if (ret) {
631
		msi_mask_irq(entry, mask, ~mask);
632
		free_msi_irqs(dev);
633
		return ret;
634
	}
635

636 637 638 639 640 641 642
	ret = msi_verify_entries(dev);
	if (ret) {
		msi_mask_irq(entry, mask, ~mask);
		free_msi_irqs(dev);
		return ret;
	}

643 644 645 646 647 648 649
	ret = populate_msi_sysfs(dev);
	if (ret) {
		msi_mask_irq(entry, mask, ~mask);
		free_msi_irqs(dev);
		return ret;
	}

L
Linus Torvalds 已提交
650
	/* Set MSI enabled bits	 */
651
	pci_intx_for_msi(dev, 0);
652
	pci_msi_set_enable(dev, 1);
653
	dev->msi_enabled = 1;
L
Linus Torvalds 已提交
654

655
	pcibios_free_irq(dev);
656
	dev->irq = entry->irq;
L
Linus Torvalds 已提交
657 658 659
	return 0;
}

660
static void __iomem *msix_map_region(struct pci_dev *dev, unsigned nr_entries)
661
{
662
	resource_size_t phys_addr;
663
	u32 table_offset;
664
	unsigned long flags;
665 666
	u8 bir;

667 668
	pci_read_config_dword(dev, dev->msix_cap + PCI_MSIX_TABLE,
			      &table_offset);
669
	bir = (u8)(table_offset & PCI_MSIX_TABLE_BIR);
670 671 672 673
	flags = pci_resource_flags(dev, bir);
	if (!flags || (flags & IORESOURCE_UNSET))
		return NULL;

674
	table_offset &= PCI_MSIX_TABLE_OFFSET;
675 676 677 678 679
	phys_addr = pci_resource_start(dev, bir) + table_offset;

	return ioremap_nocache(phys_addr, nr_entries * PCI_MSIX_ENTRY_SIZE);
}

680 681
static int msix_setup_entries(struct pci_dev *dev, void __iomem *base,
			      struct msix_entry *entries, int nvec)
682
{
683
	const struct cpumask *mask = NULL;
684
	struct msi_desc *entry;
685
	int cpu = -1, i;
686 687

	for (i = 0; i < nvec; i++) {
688 689 690 691 692 693 694
		if (dev->irq_affinity) {
			cpu = cpumask_next(cpu, dev->irq_affinity);
			if (cpu >= nr_cpu_ids)
				cpu = cpumask_first(dev->irq_affinity);
			mask = cpumask_of(cpu);
		}

695
		entry = alloc_msi_entry(&dev->dev, 1, NULL);
696 697 698 699 700 701 702 703 704 705 706
		if (!entry) {
			if (!i)
				iounmap(base);
			else
				free_msi_irqs(dev);
			/* No enough memory. Don't try again */
			return -ENOMEM;
		}

		entry->msi_attrib.is_msix	= 1;
		entry->msi_attrib.is_64		= 1;
707 708 709 710
		if (entries)
			entry->msi_attrib.entry_nr = entries[i].entry;
		else
			entry->msi_attrib.entry_nr = i;
711 712
		entry->msi_attrib.default_irq	= dev->irq;
		entry->mask_base		= base;
713
		entry->affinity			= mask;
714

715
		list_add_tail(&entry->list, dev_to_msi_list(&dev->dev));
716 717 718 719 720
	}

	return 0;
}

721
static void msix_program_entries(struct pci_dev *dev,
722
				 struct msix_entry *entries)
723 724 725 726
{
	struct msi_desc *entry;
	int i = 0;

727
	for_each_pci_msi_entry(entry, dev) {
728 729
		if (entries)
			entries[i++].vector = entry->irq;
730 731
		entry->masked = readl(pci_msix_desc_addr(entry) +
				PCI_MSIX_ENTRY_VECTOR_CTRL);
732 733 734 735
		msix_mask_irq(entry, 1);
	}
}

L
Linus Torvalds 已提交
736 737 738
/**
 * msix_capability_init - configure device's MSI-X capability
 * @dev: pointer to the pci_dev data structure of MSI-X device function
R
Randy Dunlap 已提交
739 740
 * @entries: pointer to an array of struct msix_entry entries
 * @nvec: number of @entries
L
Linus Torvalds 已提交
741
 *
742
 * Setup the MSI-X capability structure of device function with a
743 744
 * single MSI-X irq. A return of zero indicates the successful setup of
 * requested MSI-X entries with allocated irqs or non-zero for otherwise.
L
Linus Torvalds 已提交
745 746 747 748
 **/
static int msix_capability_init(struct pci_dev *dev,
				struct msix_entry *entries, int nvec)
{
749
	int ret;
750
	u16 control;
L
Linus Torvalds 已提交
751 752
	void __iomem *base;

753
	/* Ensure MSI-X is disabled while it is set up */
754
	pci_msix_clear_and_set_ctrl(dev, PCI_MSIX_FLAGS_ENABLE, 0);
755

756
	pci_read_config_word(dev, dev->msix_cap + PCI_MSIX_FLAGS, &control);
L
Linus Torvalds 已提交
757
	/* Request & Map MSI-X table region */
758
	base = msix_map_region(dev, msix_table_size(control));
759
	if (!base)
L
Linus Torvalds 已提交
760 761
		return -ENOMEM;

762
	ret = msix_setup_entries(dev, base, entries, nvec);
763 764
	if (ret)
		return ret;
765

766
	ret = pci_msi_setup_msi_irqs(dev, nvec, PCI_CAP_ID_MSIX);
767
	if (ret)
768
		goto out_avail;
769

770 771 772 773 774
	/* Check if all MSI entries honor device restrictions */
	ret = msi_verify_entries(dev);
	if (ret)
		goto out_free;

775 776 777 778 779
	/*
	 * Some devices require MSI-X to be enabled before we can touch the
	 * MSI-X registers.  We need to mask all the vectors to prevent
	 * interrupts coming in before they're fully set up.
	 */
780
	pci_msix_clear_and_set_ctrl(dev, 0,
781
				PCI_MSIX_FLAGS_MASKALL | PCI_MSIX_FLAGS_ENABLE);
782

783
	msix_program_entries(dev, entries);
784

785
	ret = populate_msi_sysfs(dev);
786 787
	if (ret)
		goto out_free;
788

789
	/* Set MSI-X enabled bits and unmask the function */
790
	pci_intx_for_msi(dev, 0);
791
	dev->msix_enabled = 1;
792
	pci_msix_clear_and_set_ctrl(dev, PCI_MSIX_FLAGS_MASKALL, 0);
793

794
	pcibios_free_irq(dev);
L
Linus Torvalds 已提交
795
	return 0;
796

797
out_avail:
798 799 800 801 802
	if (ret < 0) {
		/*
		 * If we had some success, report the number of irqs
		 * we succeeded in setting up.
		 */
803
		struct msi_desc *entry;
804 805
		int avail = 0;

806
		for_each_pci_msi_entry(entry, dev) {
807 808 809 810 811 812 813
			if (entry->irq != 0)
				avail++;
		}
		if (avail != 0)
			ret = avail;
	}

814
out_free:
815 816 817
	free_msi_irqs(dev);

	return ret;
L
Linus Torvalds 已提交
818 819
}

820
/**
821
 * pci_msi_supported - check whether MSI may be enabled on a device
822
 * @dev: pointer to the pci_dev data structure of MSI device function
823
 * @nvec: how many MSIs have been requested ?
824
 *
825
 * Look at global flags, the device itself, and its parent buses
826
 * to determine if MSI/-X are supported for the device. If MSI/-X is
827
 * supported return 1, else return 0.
828
 **/
829
static int pci_msi_supported(struct pci_dev *dev, int nvec)
830 831 832
{
	struct pci_bus *bus;

833
	/* MSI must be globally enabled and supported by the device */
834
	if (!pci_msi_enable)
835
		return 0;
836 837

	if (!dev || dev->no_msi || dev->current_state != PCI_D0)
838
		return 0;
839

840 841 842 843 844 845
	/*
	 * You can't ask to have 0 or less MSIs configured.
	 *  a) it's stupid ..
	 *  b) the list manipulation code assumes nvec >= 1.
	 */
	if (nvec < 1)
846
		return 0;
847

H
Hidetoshi Seto 已提交
848 849 850
	/*
	 * Any bridge which does NOT route MSI transactions from its
	 * secondary bus to its primary bus must set NO_MSI flag on
851 852 853 854
	 * the secondary pci_bus.
	 * We expect only arch-specific PCI host bus controller driver
	 * or quirks for specific PCI bridges to be setting NO_MSI.
	 */
855 856
	for (bus = dev->bus; bus; bus = bus->parent)
		if (bus->bus_flags & PCI_BUS_FLAGS_NO_MSI)
857
			return 0;
858

859
	return 1;
860 861
}

862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886
/**
 * pci_msi_vec_count - Return the number of MSI vectors a device can send
 * @dev: device to report about
 *
 * This function returns the number of MSI vectors a device requested via
 * Multiple Message Capable register. It returns a negative errno if the
 * device is not capable sending MSI interrupts. Otherwise, the call succeeds
 * and returns a power of two, up to a maximum of 2^5 (32), according to the
 * MSI specification.
 **/
int pci_msi_vec_count(struct pci_dev *dev)
{
	int ret;
	u16 msgctl;

	if (!dev->msi_cap)
		return -EINVAL;

	pci_read_config_word(dev, dev->msi_cap + PCI_MSI_FLAGS, &msgctl);
	ret = 1 << ((msgctl & PCI_MSI_FLAGS_QMASK) >> 1);

	return ret;
}
EXPORT_SYMBOL(pci_msi_vec_count);

887
void pci_msi_shutdown(struct pci_dev *dev)
L
Linus Torvalds 已提交
888
{
889 890
	struct msi_desc *desc;
	u32 mask;
L
Linus Torvalds 已提交
891

892
	if (!pci_msi_enable || !dev || !dev->msi_enabled)
E
Eric W. Biederman 已提交
893 894
		return;

895
	BUG_ON(list_empty(dev_to_msi_list(&dev->dev)));
896
	desc = first_pci_msi_entry(dev);
897

898
	pci_msi_set_enable(dev, 0);
899
	pci_intx_for_msi(dev, 1);
900
	dev->msi_enabled = 0;
901

902
	/* Return the device with MSI unmasked as initial states */
903
	mask = msi_mask(desc->msi_attrib.multi_cap);
904
	/* Keep cached state to be restored */
905
	__pci_msi_desc_mask_irq(desc, mask, ~mask);
906 907

	/* Restore dev->irq to its default pin-assertion irq */
908
	dev->irq = desc->msi_attrib.default_irq;
909
	pcibios_alloc_irq(dev);
910
}
911

H
Hidetoshi Seto 已提交
912
void pci_disable_msi(struct pci_dev *dev)
913 914 915 916 917
{
	if (!pci_msi_enable || !dev || !dev->msi_enabled)
		return;

	pci_msi_shutdown(dev);
918
	free_msi_irqs(dev);
L
Linus Torvalds 已提交
919
}
920
EXPORT_SYMBOL(pci_disable_msi);
L
Linus Torvalds 已提交
921

922
/**
923
 * pci_msix_vec_count - return the number of device's MSI-X table entries
924
 * @dev: pointer to the pci_dev data structure of MSI-X device function
925 926 927 928 929 930
 * This function returns the number of device's MSI-X table entries and
 * therefore the number of MSI-X vectors device is capable of sending.
 * It returns a negative errno if the device is not capable of sending MSI-X
 * interrupts.
 **/
int pci_msix_vec_count(struct pci_dev *dev)
931 932 933
{
	u16 control;

934
	if (!dev->msix_cap)
935
		return -EINVAL;
936

937
	pci_read_config_word(dev, dev->msix_cap + PCI_MSIX_FLAGS, &control);
938
	return msix_table_size(control);
939
}
940
EXPORT_SYMBOL(pci_msix_vec_count);
941

L
Linus Torvalds 已提交
942 943 944
/**
 * pci_enable_msix - configure device's MSI-X capability structure
 * @dev: pointer to the pci_dev data structure of MSI-X device function
945
 * @entries: pointer to an array of MSI-X entries (optional)
946
 * @nvec: number of MSI-X irqs requested for allocation by device driver
L
Linus Torvalds 已提交
947 948
 *
 * Setup the MSI-X capability structure of device function with the number
949
 * of requested irqs upon its software driver call to request for
L
Linus Torvalds 已提交
950 951
 * MSI-X mode enabled on its hardware device function. A return of zero
 * indicates the successful configuration of MSI-X capability structure
952
 * with new allocated MSI-X irqs. A return of < 0 indicates a failure.
L
Linus Torvalds 已提交
953
 * Or a return of > 0 indicates that driver request is exceeding the number
954 955
 * of irqs or MSI-X vectors available. Driver should use the returned value to
 * re-send its request.
L
Linus Torvalds 已提交
956
 **/
H
Hidetoshi Seto 已提交
957
int pci_enable_msix(struct pci_dev *dev, struct msix_entry *entries, int nvec)
L
Linus Torvalds 已提交
958
{
959
	int nr_entries;
E
Eric W. Biederman 已提交
960
	int i, j;
L
Linus Torvalds 已提交
961

962 963
	if (!pci_msi_supported(dev, nvec))
		return -EINVAL;
964

965 966 967
	nr_entries = pci_msix_vec_count(dev);
	if (nr_entries < 0)
		return nr_entries;
L
Linus Torvalds 已提交
968
	if (nvec > nr_entries)
969
		return nr_entries;
L
Linus Torvalds 已提交
970

971 972 973 974 975 976 977 978 979
	if (entries) {
		/* Check for any invalid entries */
		for (i = 0; i < nvec; i++) {
			if (entries[i].entry >= nr_entries)
				return -EINVAL;		/* invalid entry */
			for (j = i + 1; j < nvec; j++) {
				if (entries[i].entry == entries[j].entry)
					return -EINVAL;	/* duplicate entry */
			}
L
Linus Torvalds 已提交
980 981
		}
	}
E
Eric W. Biederman 已提交
982
	WARN_ON(!!dev->msix_enabled);
983

984
	/* Check whether driver already requested for MSI irq */
H
Hidetoshi Seto 已提交
985
	if (dev->msi_enabled) {
986
		dev_info(&dev->dev, "can't enable MSI-X (MSI IRQ already assigned)\n");
L
Linus Torvalds 已提交
987 988
		return -EINVAL;
	}
989
	return msix_capability_init(dev, entries, nvec);
L
Linus Torvalds 已提交
990
}
991
EXPORT_SYMBOL(pci_enable_msix);
L
Linus Torvalds 已提交
992

H
Hidetoshi Seto 已提交
993
void pci_msix_shutdown(struct pci_dev *dev)
994
{
995 996
	struct msi_desc *entry;

997
	if (!pci_msi_enable || !dev || !dev->msix_enabled)
E
Eric W. Biederman 已提交
998 999
		return;

1000
	/* Return the device with MSI-X masked as initial states */
1001
	for_each_pci_msi_entry(entry, dev) {
1002
		/* Keep cached states to be restored */
1003
		__pci_msix_desc_mask_irq(entry, 1);
1004 1005
	}

1006
	pci_msix_clear_and_set_ctrl(dev, PCI_MSIX_FLAGS_ENABLE, 0);
1007
	pci_intx_for_msi(dev, 1);
1008
	dev->msix_enabled = 0;
1009
	pcibios_alloc_irq(dev);
1010
}
1011

H
Hidetoshi Seto 已提交
1012
void pci_disable_msix(struct pci_dev *dev)
1013 1014 1015 1016 1017
{
	if (!pci_msi_enable || !dev || !dev->msix_enabled)
		return;

	pci_msix_shutdown(dev);
1018
	free_msi_irqs(dev);
L
Linus Torvalds 已提交
1019
}
1020
EXPORT_SYMBOL(pci_disable_msix);
L
Linus Torvalds 已提交
1021

1022 1023 1024 1025
void pci_no_msi(void)
{
	pci_msi_enable = 0;
}
1026

1027 1028 1029 1030 1031 1032 1033
/**
 * pci_msi_enabled - is MSI enabled?
 *
 * Returns true if MSI has not been disabled by the command-line option
 * pci=nomsi.
 **/
int pci_msi_enabled(void)
1034
{
1035
	return pci_msi_enable;
1036
}
1037
EXPORT_SYMBOL(pci_msi_enabled);
1038

1039 1040
static int __pci_enable_msi_range(struct pci_dev *dev, int minvec, int maxvec,
		unsigned int flags)
1041
{
1042
	int nvec;
1043 1044
	int rc;

1045 1046
	if (!pci_msi_supported(dev, minvec))
		return -EINVAL;
1047 1048 1049 1050 1051 1052 1053 1054 1055 1056

	WARN_ON(!!dev->msi_enabled);

	/* Check whether driver already requested MSI-X irqs */
	if (dev->msix_enabled) {
		dev_info(&dev->dev,
			 "can't enable MSI (MSI-X already enabled)\n");
		return -EINVAL;
	}

1057 1058 1059
	if (maxvec < minvec)
		return -ERANGE;

1060 1061 1062
	nvec = pci_msi_vec_count(dev);
	if (nvec < 0)
		return nvec;
1063
	if (nvec < minvec)
1064
		return -EINVAL;
1065 1066

	if (nvec > maxvec)
1067 1068
		nvec = maxvec;

1069
	for (;;) {
1070
		if (flags & PCI_IRQ_AFFINITY) {
1071 1072 1073 1074 1075
			dev->irq_affinity = irq_create_affinity_mask(&nvec);
			if (nvec < minvec)
				return -ENOSPC;
		}

1076
		rc = msi_capability_init(dev, nvec);
1077 1078 1079 1080 1081 1082 1083
		if (rc == 0)
			return nvec;

		kfree(dev->irq_affinity);
		dev->irq_affinity = NULL;

		if (rc < 0)
1084
			return rc;
1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105
		if (rc < minvec)
			return -ENOSPC;

		nvec = rc;
	}
}

/**
 * pci_enable_msi_range - configure device's MSI capability structure
 * @dev: device to configure
 * @minvec: minimal number of interrupts to configure
 * @maxvec: maximum number of interrupts to configure
 *
 * This function tries to allocate a maximum possible number of interrupts in a
 * range between @minvec and @maxvec. It returns a negative errno if an error
 * occurs. If it succeeds, it returns the actual number of interrupts allocated
 * and updates the @dev's irq member to the lowest new interrupt number;
 * the other interrupt numbers allocated to this device are consecutive.
 **/
int pci_enable_msi_range(struct pci_dev *dev, int minvec, int maxvec)
{
1106
	return __pci_enable_msi_range(dev, minvec, maxvec, 0);
1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120
}
EXPORT_SYMBOL(pci_enable_msi_range);

static int __pci_enable_msix_range(struct pci_dev *dev,
		struct msix_entry *entries, int minvec, int maxvec,
		unsigned int flags)
{
	int nvec = maxvec;
	int rc;

	if (maxvec < minvec)
		return -ERANGE;

	for (;;) {
1121
		if (flags & PCI_IRQ_AFFINITY) {
1122 1123
			dev->irq_affinity = irq_create_affinity_mask(&nvec);
			if (nvec < minvec)
1124 1125 1126
				return -ENOSPC;
		}

1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140
		rc = pci_enable_msix(dev, entries, nvec);
		if (rc == 0)
			return nvec;

		kfree(dev->irq_affinity);
		dev->irq_affinity = NULL;

		if (rc < 0)
			return rc;
		if (rc < minvec)
			return -ENOSPC;

		nvec = rc;
	}
1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158
}

/**
 * pci_enable_msix_range - configure device's MSI-X capability structure
 * @dev: pointer to the pci_dev data structure of MSI-X device function
 * @entries: pointer to an array of MSI-X entries
 * @minvec: minimum number of MSI-X irqs requested
 * @maxvec: maximum number of MSI-X irqs requested
 *
 * Setup the MSI-X capability structure of device function with a maximum
 * possible number of interrupts in the range between @minvec and @maxvec
 * upon its software driver call to request for MSI-X mode enabled on its
 * hardware device function. It returns a negative errno if an error occurs.
 * If it succeeds, it returns the actual number of interrupts allocated and
 * indicates the successful configuration of MSI-X capability structure
 * with new allocated MSI-X interrupts.
 **/
int pci_enable_msix_range(struct pci_dev *dev, struct msix_entry *entries,
1159
		int minvec, int maxvec)
1160
{
1161
	return __pci_enable_msix_range(dev, entries, minvec, maxvec, 0);
1162 1163
}
EXPORT_SYMBOL(pci_enable_msix_range);
1164

1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186
/**
 * pci_alloc_irq_vectors - allocate multiple IRQs for a device
 * @dev:		PCI device to operate on
 * @min_vecs:		minimum number of vectors required (must be >= 1)
 * @max_vecs:		maximum (desired) number of vectors
 * @flags:		flags or quirks for the allocation
 *
 * Allocate up to @max_vecs interrupt vectors for @dev, using MSI-X or MSI
 * vectors if available, and fall back to a single legacy vector
 * if neither is available.  Return the number of vectors allocated,
 * (which might be smaller than @max_vecs) if successful, or a negative
 * error code on error. If less than @min_vecs interrupt vectors are
 * available for @dev the function will fail with -ENOSPC.
 *
 * To get the Linux IRQ number used for a vector that can be passed to
 * request_irq() use the pci_irq_vector() helper.
 */
int pci_alloc_irq_vectors(struct pci_dev *dev, unsigned int min_vecs,
		unsigned int max_vecs, unsigned int flags)
{
	int vecs = -ENOSPC;

1187
	if (flags & PCI_IRQ_MSIX) {
1188 1189
		vecs = __pci_enable_msix_range(dev, NULL, min_vecs, max_vecs,
				flags);
1190 1191 1192 1193
		if (vecs > 0)
			return vecs;
	}

1194
	if (flags & PCI_IRQ_MSI) {
1195
		vecs = __pci_enable_msi_range(dev, min_vecs, max_vecs, flags);
1196 1197 1198 1199 1200
		if (vecs > 0)
			return vecs;
	}

	/* use legacy irq if allowed */
1201 1202
	if ((flags & PCI_IRQ_LEGACY) && min_vecs == 1) {
		pci_intx(dev, 1);
1203
		return 1;
1204 1205
	}

1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256
	return vecs;
}
EXPORT_SYMBOL(pci_alloc_irq_vectors);

/**
 * pci_free_irq_vectors - free previously allocated IRQs for a device
 * @dev:		PCI device to operate on
 *
 * Undoes the allocations and enabling in pci_alloc_irq_vectors().
 */
void pci_free_irq_vectors(struct pci_dev *dev)
{
	pci_disable_msix(dev);
	pci_disable_msi(dev);
}
EXPORT_SYMBOL(pci_free_irq_vectors);

/**
 * pci_irq_vector - return Linux IRQ number of a device vector
 * @dev: PCI device to operate on
 * @nr: device-relative interrupt vector index (0-based).
 */
int pci_irq_vector(struct pci_dev *dev, unsigned int nr)
{
	if (dev->msix_enabled) {
		struct msi_desc *entry;
		int i = 0;

		for_each_pci_msi_entry(entry, dev) {
			if (i == nr)
				return entry->irq;
			i++;
		}
		WARN_ON_ONCE(1);
		return -EINVAL;
	}

	if (dev->msi_enabled) {
		struct msi_desc *entry = first_pci_msi_entry(dev);

		if (WARN_ON_ONCE(nr >= entry->nvec_used))
			return -EINVAL;
	} else {
		if (WARN_ON_ONCE(nr > 0))
			return -EINVAL;
	}

	return dev->irq + nr;
}
EXPORT_SYMBOL(pci_irq_vector);

1257 1258 1259 1260
struct pci_dev *msi_desc_to_pci_dev(struct msi_desc *desc)
{
	return to_pci_dev(desc->dev);
}
1261
EXPORT_SYMBOL(msi_desc_to_pci_dev);
1262

1263 1264 1265 1266 1267 1268 1269 1270
void *msi_desc_to_pci_sysdata(struct msi_desc *desc)
{
	struct pci_dev *dev = msi_desc_to_pci_dev(desc);

	return dev->bus->sysdata;
}
EXPORT_SYMBOL_GPL(msi_desc_to_pci_sysdata);

1271 1272 1273 1274 1275 1276 1277 1278
#ifdef CONFIG_PCI_MSI_IRQ_DOMAIN
/**
 * pci_msi_domain_write_msg - Helper to write MSI message to PCI config space
 * @irq_data:	Pointer to interrupt data of the MSI interrupt
 * @msg:	Pointer to the message
 */
void pci_msi_domain_write_msg(struct irq_data *irq_data, struct msi_msg *msg)
{
1279
	struct msi_desc *desc = irq_data_get_msi_desc(irq_data);
1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385

	/*
	 * For MSI-X desc->irq is always equal to irq_data->irq. For
	 * MSI only the first interrupt of MULTI MSI passes the test.
	 */
	if (desc->irq == irq_data->irq)
		__pci_write_msi_msg(desc, msg);
}

/**
 * pci_msi_domain_calc_hwirq - Generate a unique ID for an MSI source
 * @dev:	Pointer to the PCI device
 * @desc:	Pointer to the msi descriptor
 *
 * The ID number is only used within the irqdomain.
 */
irq_hw_number_t pci_msi_domain_calc_hwirq(struct pci_dev *dev,
					  struct msi_desc *desc)
{
	return (irq_hw_number_t)desc->msi_attrib.entry_nr |
		PCI_DEVID(dev->bus->number, dev->devfn) << 11 |
		(pci_domain_nr(dev->bus) & 0xFFFFFFFF) << 27;
}

static inline bool pci_msi_desc_is_multi_msi(struct msi_desc *desc)
{
	return !desc->msi_attrib.is_msix && desc->nvec_used > 1;
}

/**
 * pci_msi_domain_check_cap - Verify that @domain supports the capabilities for @dev
 * @domain:	The interrupt domain to check
 * @info:	The domain info for verification
 * @dev:	The device to check
 *
 * Returns:
 *  0 if the functionality is supported
 *  1 if Multi MSI is requested, but the domain does not support it
 *  -ENOTSUPP otherwise
 */
int pci_msi_domain_check_cap(struct irq_domain *domain,
			     struct msi_domain_info *info, struct device *dev)
{
	struct msi_desc *desc = first_pci_msi_entry(to_pci_dev(dev));

	/* Special handling to support pci_enable_msi_range() */
	if (pci_msi_desc_is_multi_msi(desc) &&
	    !(info->flags & MSI_FLAG_MULTI_PCI_MSI))
		return 1;
	else if (desc->msi_attrib.is_msix && !(info->flags & MSI_FLAG_PCI_MSIX))
		return -ENOTSUPP;

	return 0;
}

static int pci_msi_domain_handle_error(struct irq_domain *domain,
				       struct msi_desc *desc, int error)
{
	/* Special handling to support pci_enable_msi_range() */
	if (pci_msi_desc_is_multi_msi(desc) && error == -ENOSPC)
		return 1;

	return error;
}

#ifdef GENERIC_MSI_DOMAIN_OPS
static void pci_msi_domain_set_desc(msi_alloc_info_t *arg,
				    struct msi_desc *desc)
{
	arg->desc = desc;
	arg->hwirq = pci_msi_domain_calc_hwirq(msi_desc_to_pci_dev(desc),
					       desc);
}
#else
#define pci_msi_domain_set_desc		NULL
#endif

static struct msi_domain_ops pci_msi_domain_ops_default = {
	.set_desc	= pci_msi_domain_set_desc,
	.msi_check	= pci_msi_domain_check_cap,
	.handle_error	= pci_msi_domain_handle_error,
};

static void pci_msi_domain_update_dom_ops(struct msi_domain_info *info)
{
	struct msi_domain_ops *ops = info->ops;

	if (ops == NULL) {
		info->ops = &pci_msi_domain_ops_default;
	} else {
		if (ops->set_desc == NULL)
			ops->set_desc = pci_msi_domain_set_desc;
		if (ops->msi_check == NULL)
			ops->msi_check = pci_msi_domain_check_cap;
		if (ops->handle_error == NULL)
			ops->handle_error = pci_msi_domain_handle_error;
	}
}

static void pci_msi_domain_update_chip_ops(struct msi_domain_info *info)
{
	struct irq_chip *chip = info->chip;

	BUG_ON(!chip);
	if (!chip->irq_write_msi_msg)
		chip->irq_write_msi_msg = pci_msi_domain_write_msg;
1386 1387 1388 1389
	if (!chip->irq_mask)
		chip->irq_mask = pci_msi_mask_irq;
	if (!chip->irq_unmask)
		chip->irq_unmask = pci_msi_unmask_irq;
1390 1391 1392
}

/**
1393 1394
 * pci_msi_create_irq_domain - Create a MSI interrupt domain
 * @fwnode:	Optional fwnode of the interrupt controller
1395 1396 1397 1398 1399 1400 1401 1402
 * @info:	MSI domain info
 * @parent:	Parent irq domain
 *
 * Updates the domain and chip ops and creates a MSI interrupt domain.
 *
 * Returns:
 * A domain pointer or NULL in case of failure.
 */
1403
struct irq_domain *pci_msi_create_irq_domain(struct fwnode_handle *fwnode,
1404 1405 1406
					     struct msi_domain_info *info,
					     struct irq_domain *parent)
{
1407 1408
	struct irq_domain *domain;

1409 1410 1411 1412 1413
	if (info->flags & MSI_FLAG_USE_DEF_DOM_OPS)
		pci_msi_domain_update_dom_ops(info);
	if (info->flags & MSI_FLAG_USE_DEF_CHIP_OPS)
		pci_msi_domain_update_chip_ops(info);

1414 1415
	info->flags |= MSI_FLAG_ACTIVATE_EARLY;

1416
	domain = msi_create_irq_domain(fwnode, info, parent);
1417 1418 1419 1420 1421
	if (!domain)
		return NULL;

	domain->bus_token = DOMAIN_BUS_PCI_MSI;
	return domain;
1422
}
1423
EXPORT_SYMBOL_GPL(pci_msi_create_irq_domain);
1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449

/**
 * pci_msi_domain_alloc_irqs - Allocate interrupts for @dev in @domain
 * @domain:	The interrupt domain to allocate from
 * @dev:	The device for which to allocate
 * @nvec:	The number of interrupts to allocate
 * @type:	Unused to allow simpler migration from the arch_XXX interfaces
 *
 * Returns:
 * A virtual interrupt number or an error code in case of failure
 */
int pci_msi_domain_alloc_irqs(struct irq_domain *domain, struct pci_dev *dev,
			      int nvec, int type)
{
	return msi_domain_alloc_irqs(domain, &dev->dev, nvec);
}

/**
 * pci_msi_domain_free_irqs - Free interrupts for @dev in @domain
 * @domain:	The interrupt domain
 * @dev:	The device for which to free interrupts
 */
void pci_msi_domain_free_irqs(struct irq_domain *domain, struct pci_dev *dev)
{
	msi_domain_free_irqs(domain, &dev->dev);
}
1450 1451 1452

/**
 * pci_msi_create_default_irq_domain - Create a default MSI interrupt domain
1453
 * @fwnode:	Optional fwnode of the interrupt controller
1454 1455 1456 1457 1458 1459
 * @info:	MSI domain info
 * @parent:	Parent irq domain
 *
 * Returns: A domain pointer or NULL in case of failure. If successful
 * the default PCI/MSI irqdomain pointer is updated.
 */
1460
struct irq_domain *pci_msi_create_default_irq_domain(struct fwnode_handle *fwnode,
1461 1462 1463 1464 1465 1466 1467 1468 1469
		struct msi_domain_info *info, struct irq_domain *parent)
{
	struct irq_domain *domain;

	mutex_lock(&pci_msi_domain_lock);
	if (pci_msi_default_domain) {
		pr_err("PCI: default irq domain for PCI MSI has already been created.\n");
		domain = NULL;
	} else {
1470
		domain = pci_msi_create_irq_domain(fwnode, info, parent);
1471 1472 1473 1474 1475 1476
		pci_msi_default_domain = domain;
	}
	mutex_unlock(&pci_msi_domain_lock);

	return domain;
}
1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507

static int get_msi_id_cb(struct pci_dev *pdev, u16 alias, void *data)
{
	u32 *pa = data;

	*pa = alias;
	return 0;
}
/**
 * pci_msi_domain_get_msi_rid - Get the MSI requester id (RID)
 * @domain:	The interrupt domain
 * @pdev:	The PCI device.
 *
 * The RID for a device is formed from the alias, with a firmware
 * supplied mapping applied
 *
 * Returns: The RID.
 */
u32 pci_msi_domain_get_msi_rid(struct irq_domain *domain, struct pci_dev *pdev)
{
	struct device_node *of_node;
	u32 rid = 0;

	pci_for_each_dma_alias(pdev, get_msi_id_cb, &rid);

	of_node = irq_domain_get_of_node(domain);
	if (of_node)
		rid = of_msi_map_rid(&pdev->dev, of_node, rid);

	return rid;
}
1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524

/**
 * pci_msi_get_device_domain - Get the MSI domain for a given PCI device
 * @pdev:	The PCI device
 *
 * Use the firmware data to find a device-specific MSI domain
 * (i.e. not one that is ste as a default).
 *
 * Returns: The coresponding MSI domain or NULL if none has been found.
 */
struct irq_domain *pci_msi_get_device_domain(struct pci_dev *pdev)
{
	u32 rid = 0;

	pci_for_each_dma_alias(pdev, get_msi_id_cb, &rid);
	return of_msi_map_get_device_domain(&pdev->dev, rid);
}
1525
#endif /* CONFIG_PCI_MSI_IRQ_DOMAIN */