iommu.c 64.4 KB
Newer Older
1
// SPDX-License-Identifier: GPL-2.0-only
2 3
/*
 * Copyright (C) 2007-2008 Advanced Micro Devices, Inc.
J
Joerg Roedel 已提交
4
 * Author: Joerg Roedel <jroedel@suse.de>
5 6
 */

7
#define pr_fmt(fmt)    "iommu: " fmt
8

9
#include <linux/device.h>
10
#include <linux/kernel.h>
11 12
#include <linux/bug.h>
#include <linux/types.h>
13 14
#include <linux/init.h>
#include <linux/export.h>
15
#include <linux/slab.h>
16 17
#include <linux/errno.h>
#include <linux/iommu.h>
A
Alex Williamson 已提交
18 19 20
#include <linux/idr.h>
#include <linux/notifier.h>
#include <linux/err.h>
21
#include <linux/pci.h>
22
#include <linux/bitops.h>
R
Robin Murphy 已提交
23
#include <linux/property.h>
24
#include <linux/fsl/mc.h>
25
#include <trace/events/iommu.h>
A
Alex Williamson 已提交
26 27

static struct kset *iommu_group_kset;
H
Heiner Kallweit 已提交
28
static DEFINE_IDA(iommu_group_ida);
29 30

static unsigned int iommu_def_domain_type __read_mostly;
31
static bool iommu_dma_strict __read_mostly = true;
32
static u32 iommu_cmd_line __read_mostly;
A
Alex Williamson 已提交
33 34 35 36 37 38 39 40 41 42 43

struct iommu_group {
	struct kobject kobj;
	struct kobject *devices_kobj;
	struct list_head devices;
	struct mutex mutex;
	struct blocking_notifier_head notifier;
	void *iommu_data;
	void (*iommu_data_release)(void *iommu_data);
	char *name;
	int id;
44
	struct iommu_domain *default_domain;
45
	struct iommu_domain *domain;
A
Alex Williamson 已提交
46 47
};

J
Joerg Roedel 已提交
48
struct group_device {
A
Alex Williamson 已提交
49 50 51 52 53 54 55 56 57 58 59 60
	struct list_head list;
	struct device *dev;
	char *name;
};

struct iommu_group_attribute {
	struct attribute attr;
	ssize_t (*show)(struct iommu_group *group, char *buf);
	ssize_t (*store)(struct iommu_group *group,
			 const char *buf, size_t count);
};

61
static const char * const iommu_group_resv_type_string[] = {
62 63 64 65 66
	[IOMMU_RESV_DIRECT]			= "direct",
	[IOMMU_RESV_DIRECT_RELAXABLE]		= "direct-relaxable",
	[IOMMU_RESV_RESERVED]			= "reserved",
	[IOMMU_RESV_MSI]			= "msi",
	[IOMMU_RESV_SW_MSI]			= "msi",
67 68
};

69 70 71 72 73 74 75
#define IOMMU_CMD_LINE_DMA_API		BIT(0)

static void iommu_set_cmd_line_dma_api(void)
{
	iommu_cmd_line |= IOMMU_CMD_LINE_DMA_API;
}

76
static bool iommu_cmd_line_dma_api(void)
77 78 79 80
{
	return !!(iommu_cmd_line & IOMMU_CMD_LINE_DMA_API);
}

A
Alex Williamson 已提交
81 82 83
#define IOMMU_GROUP_ATTR(_name, _mode, _show, _store)		\
struct iommu_group_attribute iommu_group_attr_##_name =		\
	__ATTR(_name, _mode, _show, _store)
84

A
Alex Williamson 已提交
85 86 87 88
#define to_iommu_group_attr(_attr)	\
	container_of(_attr, struct iommu_group_attribute, attr)
#define to_iommu_group(_kobj)		\
	container_of(_kobj, struct iommu_group, kobj)
89

90 91 92
static LIST_HEAD(iommu_device_list);
static DEFINE_SPINLOCK(iommu_device_lock);

93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114
/*
 * Use a function instead of an array here because the domain-type is a
 * bit-field, so an array would waste memory.
 */
static const char *iommu_domain_type_str(unsigned int t)
{
	switch (t) {
	case IOMMU_DOMAIN_BLOCKED:
		return "Blocked";
	case IOMMU_DOMAIN_IDENTITY:
		return "Passthrough";
	case IOMMU_DOMAIN_UNMANAGED:
		return "Unmanaged";
	case IOMMU_DOMAIN_DMA:
		return "Translated";
	default:
		return "Unknown";
	}
}

static int __init iommu_subsys_init(void)
{
115 116 117 118 119 120 121
	bool cmd_line = iommu_cmd_line_dma_api();

	if (!cmd_line) {
		if (IS_ENABLED(CONFIG_IOMMU_DEFAULT_PASSTHROUGH))
			iommu_set_default_passthrough(false);
		else
			iommu_set_default_translated(false);
122

123 124
		if (iommu_default_passthrough() && mem_encrypt_active()) {
			pr_info("Memory encryption detected - Disabling default IOMMU Passthrough\n");
125 126
			iommu_set_default_translated(false);
		}
127 128 129 130 131
	}

	pr_info("Default domain type: %s %s\n",
		iommu_domain_type_str(iommu_def_domain_type),
		cmd_line ? "(set via kernel command line)" : "");
132 133 134 135 136

	return 0;
}
subsys_initcall(iommu_subsys_init);

137 138 139 140 141 142 143 144 145 146 147 148 149 150 151
int iommu_device_register(struct iommu_device *iommu)
{
	spin_lock(&iommu_device_lock);
	list_add_tail(&iommu->list, &iommu_device_list);
	spin_unlock(&iommu_device_lock);
	return 0;
}

void iommu_device_unregister(struct iommu_device *iommu)
{
	spin_lock(&iommu_device_lock);
	list_del(&iommu->list);
	spin_unlock(&iommu_device_lock);
}

152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173
static struct iommu_param *iommu_get_dev_param(struct device *dev)
{
	struct iommu_param *param = dev->iommu_param;

	if (param)
		return param;

	param = kzalloc(sizeof(*param), GFP_KERNEL);
	if (!param)
		return NULL;

	mutex_init(&param->lock);
	dev->iommu_param = param;
	return param;
}

static void iommu_free_dev_param(struct device *dev)
{
	kfree(dev->iommu_param);
	dev->iommu_param = NULL;
}

174 175 176
int iommu_probe_device(struct device *dev)
{
	const struct iommu_ops *ops = dev->bus->iommu_ops;
177
	int ret;
178 179

	WARN_ON(dev->iommu_group);
180 181
	if (!ops)
		return -EINVAL;
182

183 184
	if (!iommu_get_dev_param(dev))
		return -ENOMEM;
185

186 187 188
	ret = ops->add_device(dev);
	if (ret)
		iommu_free_dev_param(dev);
189 190

	return ret;
191 192 193 194 195 196 197 198
}

void iommu_release_device(struct device *dev)
{
	const struct iommu_ops *ops = dev->bus->iommu_ops;

	if (dev->iommu_group)
		ops->remove_device(dev);
199 200

	iommu_free_dev_param(dev);
201 202
}

203 204
static struct iommu_domain *__iommu_domain_alloc(struct bus_type *bus,
						 unsigned type);
205 206 207 208 209 210
static int __iommu_attach_device(struct iommu_domain *domain,
				 struct device *dev);
static int __iommu_attach_group(struct iommu_domain *domain,
				struct iommu_group *group);
static void __iommu_detach_group(struct iommu_domain *domain,
				 struct iommu_group *group);
211

212 213 214
static int __init iommu_set_def_domain_type(char *str)
{
	bool pt;
215
	int ret;
216

217 218 219
	ret = kstrtobool(str, &pt);
	if (ret)
		return ret;
220

221 222 223 224
	if (pt)
		iommu_set_default_passthrough(true);
	else
		iommu_set_default_translated(true);
225

226 227 228 229
	return 0;
}
early_param("iommu.passthrough", iommu_set_def_domain_type);

230 231 232 233 234 235
static int __init iommu_dma_setup(char *str)
{
	return kstrtobool(str, &iommu_dma_strict);
}
early_param("iommu.strict", iommu_dma_setup);

A
Alex Williamson 已提交
236 237
static ssize_t iommu_group_attr_show(struct kobject *kobj,
				     struct attribute *__attr, char *buf)
238
{
A
Alex Williamson 已提交
239 240 241
	struct iommu_group_attribute *attr = to_iommu_group_attr(__attr);
	struct iommu_group *group = to_iommu_group(kobj);
	ssize_t ret = -EIO;
242

A
Alex Williamson 已提交
243 244 245 246 247 248 249 250 251 252 253 254
	if (attr->show)
		ret = attr->show(group, buf);
	return ret;
}

static ssize_t iommu_group_attr_store(struct kobject *kobj,
				      struct attribute *__attr,
				      const char *buf, size_t count)
{
	struct iommu_group_attribute *attr = to_iommu_group_attr(__attr);
	struct iommu_group *group = to_iommu_group(kobj);
	ssize_t ret = -EIO;
255

A
Alex Williamson 已提交
256 257 258
	if (attr->store)
		ret = attr->store(group, buf, count);
	return ret;
259 260
}

A
Alex Williamson 已提交
261 262 263 264
static const struct sysfs_ops iommu_group_sysfs_ops = {
	.show = iommu_group_attr_show,
	.store = iommu_group_attr_store,
};
265

A
Alex Williamson 已提交
266 267 268 269
static int iommu_group_create_file(struct iommu_group *group,
				   struct iommu_group_attribute *attr)
{
	return sysfs_create_file(&group->kobj, &attr->attr);
270 271
}

A
Alex Williamson 已提交
272 273 274 275 276 277 278 279 280 281 282
static void iommu_group_remove_file(struct iommu_group *group,
				    struct iommu_group_attribute *attr)
{
	sysfs_remove_file(&group->kobj, &attr->attr);
}

static ssize_t iommu_group_show_name(struct iommu_group *group, char *buf)
{
	return sprintf(buf, "%s\n", group->name);
}

E
Eric Auger 已提交
283 284 285 286 287 288
/**
 * iommu_insert_resv_region - Insert a new region in the
 * list of reserved regions.
 * @new: new region to insert
 * @regions: list of regions
 *
289 290
 * Elements are sorted by start address and overlapping segments
 * of the same type are merged.
E
Eric Auger 已提交
291
 */
292 293
int iommu_insert_resv_region(struct iommu_resv_region *new,
			     struct list_head *regions)
E
Eric Auger 已提交
294
{
295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314
	struct iommu_resv_region *iter, *tmp, *nr, *top;
	LIST_HEAD(stack);

	nr = iommu_alloc_resv_region(new->start, new->length,
				     new->prot, new->type);
	if (!nr)
		return -ENOMEM;

	/* First add the new element based on start address sorting */
	list_for_each_entry(iter, regions, list) {
		if (nr->start < iter->start ||
		    (nr->start == iter->start && nr->type <= iter->type))
			break;
	}
	list_add_tail(&nr->list, &iter->list);

	/* Merge overlapping segments of type nr->type in @regions, if any */
	list_for_each_entry_safe(iter, tmp, regions, list) {
		phys_addr_t top_end, iter_end = iter->start + iter->length - 1;

315 316
		/* no merge needed on elements of different types than @new */
		if (iter->type != new->type) {
317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333
			list_move_tail(&iter->list, &stack);
			continue;
		}

		/* look for the last stack element of same type as @iter */
		list_for_each_entry_reverse(top, &stack, list)
			if (top->type == iter->type)
				goto check_overlap;

		list_move_tail(&iter->list, &stack);
		continue;

check_overlap:
		top_end = top->start + top->length - 1;

		if (iter->start > top_end + 1) {
			list_move_tail(&iter->list, &stack);
E
Eric Auger 已提交
334
		} else {
335 336 337
			top->length = max(top_end, iter_end) - top->start + 1;
			list_del(&iter->list);
			kfree(iter);
E
Eric Auger 已提交
338 339
		}
	}
340
	list_splice(&stack, regions);
E
Eric Auger 已提交
341 342 343 344 345 346 347 348
	return 0;
}

static int
iommu_insert_device_resv_regions(struct list_head *dev_resv_regions,
				 struct list_head *group_resv_regions)
{
	struct iommu_resv_region *entry;
349
	int ret = 0;
E
Eric Auger 已提交
350 351 352 353 354 355 356 357 358 359 360 361

	list_for_each_entry(entry, dev_resv_regions, list) {
		ret = iommu_insert_resv_region(entry, group_resv_regions);
		if (ret)
			break;
	}
	return ret;
}

int iommu_get_group_resv_regions(struct iommu_group *group,
				 struct list_head *head)
{
362
	struct group_device *device;
E
Eric Auger 已提交
363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380
	int ret = 0;

	mutex_lock(&group->mutex);
	list_for_each_entry(device, &group->devices, list) {
		struct list_head dev_resv_regions;

		INIT_LIST_HEAD(&dev_resv_regions);
		iommu_get_resv_regions(device->dev, &dev_resv_regions);
		ret = iommu_insert_device_resv_regions(&dev_resv_regions, head);
		iommu_put_resv_regions(device->dev, &dev_resv_regions);
		if (ret)
			break;
	}
	mutex_unlock(&group->mutex);
	return ret;
}
EXPORT_SYMBOL_GPL(iommu_get_group_resv_regions);

381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402
static ssize_t iommu_group_show_resv_regions(struct iommu_group *group,
					     char *buf)
{
	struct iommu_resv_region *region, *next;
	struct list_head group_resv_regions;
	char *str = buf;

	INIT_LIST_HEAD(&group_resv_regions);
	iommu_get_group_resv_regions(group, &group_resv_regions);

	list_for_each_entry_safe(region, next, &group_resv_regions, list) {
		str += sprintf(str, "0x%016llx 0x%016llx %s\n",
			       (long long int)region->start,
			       (long long int)(region->start +
						region->length - 1),
			       iommu_group_resv_type_string[region->type]);
		kfree(region);
	}

	return (str - buf);
}

403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419
static ssize_t iommu_group_show_type(struct iommu_group *group,
				     char *buf)
{
	char *type = "unknown\n";

	if (group->default_domain) {
		switch (group->default_domain->type) {
		case IOMMU_DOMAIN_BLOCKED:
			type = "blocked\n";
			break;
		case IOMMU_DOMAIN_IDENTITY:
			type = "identity\n";
			break;
		case IOMMU_DOMAIN_UNMANAGED:
			type = "unmanaged\n";
			break;
		case IOMMU_DOMAIN_DMA:
420
			type = "DMA\n";
421 422 423 424 425 426 427 428
			break;
		}
	}
	strcpy(buf, type);

	return strlen(type);
}

A
Alex Williamson 已提交
429 430
static IOMMU_GROUP_ATTR(name, S_IRUGO, iommu_group_show_name, NULL);

431 432 433
static IOMMU_GROUP_ATTR(reserved_regions, 0444,
			iommu_group_show_resv_regions, NULL);

434 435
static IOMMU_GROUP_ATTR(type, 0444, iommu_group_show_type, NULL);

A
Alex Williamson 已提交
436 437 438 439
static void iommu_group_release(struct kobject *kobj)
{
	struct iommu_group *group = to_iommu_group(kobj);

440 441
	pr_debug("Releasing group %d\n", group->id);

A
Alex Williamson 已提交
442 443 444
	if (group->iommu_data_release)
		group->iommu_data_release(group->iommu_data);

445
	ida_simple_remove(&iommu_group_ida, group->id);
A
Alex Williamson 已提交
446

447 448 449
	if (group->default_domain)
		iommu_domain_free(group->default_domain);

A
Alex Williamson 已提交
450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470
	kfree(group->name);
	kfree(group);
}

static struct kobj_type iommu_group_ktype = {
	.sysfs_ops = &iommu_group_sysfs_ops,
	.release = iommu_group_release,
};

/**
 * iommu_group_alloc - Allocate a new group
 *
 * This function is called by an iommu driver to allocate a new iommu
 * group.  The iommu group represents the minimum granularity of the iommu.
 * Upon successful return, the caller holds a reference to the supplied
 * group in order to hold the group until devices are added.  Use
 * iommu_group_put() to release this extra reference count, allowing the
 * group to be automatically reclaimed once it has no devices or external
 * references.
 */
struct iommu_group *iommu_group_alloc(void)
471
{
A
Alex Williamson 已提交
472 473 474 475 476 477 478 479 480 481 482 483
	struct iommu_group *group;
	int ret;

	group = kzalloc(sizeof(*group), GFP_KERNEL);
	if (!group)
		return ERR_PTR(-ENOMEM);

	group->kobj.kset = iommu_group_kset;
	mutex_init(&group->mutex);
	INIT_LIST_HEAD(&group->devices);
	BLOCKING_INIT_NOTIFIER_HEAD(&group->notifier);

484 485
	ret = ida_simple_get(&iommu_group_ida, 0, 0, GFP_KERNEL);
	if (ret < 0) {
A
Alex Williamson 已提交
486
		kfree(group);
487
		return ERR_PTR(ret);
A
Alex Williamson 已提交
488
	}
489
	group->id = ret;
490

A
Alex Williamson 已提交
491 492 493
	ret = kobject_init_and_add(&group->kobj, &iommu_group_ktype,
				   NULL, "%d", group->id);
	if (ret) {
494
		ida_simple_remove(&iommu_group_ida, group->id);
A
Alex Williamson 已提交
495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511
		kfree(group);
		return ERR_PTR(ret);
	}

	group->devices_kobj = kobject_create_and_add("devices", &group->kobj);
	if (!group->devices_kobj) {
		kobject_put(&group->kobj); /* triggers .release & free */
		return ERR_PTR(-ENOMEM);
	}

	/*
	 * The devices_kobj holds a reference on the group kobject, so
	 * as long as that exists so will the group.  We can therefore
	 * use the devices_kobj for reference counting.
	 */
	kobject_put(&group->kobj);

512 513 514 515 516
	ret = iommu_group_create_file(group,
				      &iommu_group_attr_reserved_regions);
	if (ret)
		return ERR_PTR(ret);

517 518 519 520
	ret = iommu_group_create_file(group, &iommu_group_attr_type);
	if (ret)
		return ERR_PTR(ret);

521 522
	pr_debug("Allocated group %d\n", group->id);

A
Alex Williamson 已提交
523 524 525 526
	return group;
}
EXPORT_SYMBOL_GPL(iommu_group_alloc);

527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555
struct iommu_group *iommu_group_get_by_id(int id)
{
	struct kobject *group_kobj;
	struct iommu_group *group;
	const char *name;

	if (!iommu_group_kset)
		return NULL;

	name = kasprintf(GFP_KERNEL, "%d", id);
	if (!name)
		return NULL;

	group_kobj = kset_find_obj(iommu_group_kset, name);
	kfree(name);

	if (!group_kobj)
		return NULL;

	group = container_of(group_kobj, struct iommu_group, kobj);
	BUG_ON(group->id != id);

	kobject_get(group->devices_kobj);
	kobject_put(&group->kobj);

	return group;
}
EXPORT_SYMBOL_GPL(iommu_group_get_by_id);

A
Alex Williamson 已提交
556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581
/**
 * iommu_group_get_iommudata - retrieve iommu_data registered for a group
 * @group: the group
 *
 * iommu drivers can store data in the group for use when doing iommu
 * operations.  This function provides a way to retrieve it.  Caller
 * should hold a group reference.
 */
void *iommu_group_get_iommudata(struct iommu_group *group)
{
	return group->iommu_data;
}
EXPORT_SYMBOL_GPL(iommu_group_get_iommudata);

/**
 * iommu_group_set_iommudata - set iommu_data for a group
 * @group: the group
 * @iommu_data: new data
 * @release: release function for iommu_data
 *
 * iommu drivers can store data in the group for use when doing iommu
 * operations.  This function provides a way to set the data after
 * the group has been allocated.  Caller should hold a group reference.
 */
void iommu_group_set_iommudata(struct iommu_group *group, void *iommu_data,
			       void (*release)(void *iommu_data))
582
{
A
Alex Williamson 已提交
583 584 585 586
	group->iommu_data = iommu_data;
	group->iommu_data_release = release;
}
EXPORT_SYMBOL_GPL(iommu_group_set_iommudata);
587

A
Alex Williamson 已提交
588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617
/**
 * iommu_group_set_name - set name for a group
 * @group: the group
 * @name: name
 *
 * Allow iommu driver to set a name for a group.  When set it will
 * appear in a name attribute file under the group in sysfs.
 */
int iommu_group_set_name(struct iommu_group *group, const char *name)
{
	int ret;

	if (group->name) {
		iommu_group_remove_file(group, &iommu_group_attr_name);
		kfree(group->name);
		group->name = NULL;
		if (!name)
			return 0;
	}

	group->name = kstrdup(name, GFP_KERNEL);
	if (!group->name)
		return -ENOMEM;

	ret = iommu_group_create_file(group, &iommu_group_attr_name);
	if (ret) {
		kfree(group->name);
		group->name = NULL;
		return ret;
	}
618 619 620

	return 0;
}
A
Alex Williamson 已提交
621
EXPORT_SYMBOL_GPL(iommu_group_set_name);
622

623 624 625 626
static int iommu_group_create_direct_mappings(struct iommu_group *group,
					      struct device *dev)
{
	struct iommu_domain *domain = group->default_domain;
627
	struct iommu_resv_region *entry;
628 629 630 631 632 633 634
	struct list_head mappings;
	unsigned long pg_size;
	int ret = 0;

	if (!domain || domain->type != IOMMU_DOMAIN_DMA)
		return 0;

635
	BUG_ON(!domain->pgsize_bitmap);
636

637
	pg_size = 1UL << __ffs(domain->pgsize_bitmap);
638 639
	INIT_LIST_HEAD(&mappings);

640
	iommu_get_resv_regions(dev, &mappings);
641 642 643 644 645

	/* We need to consider overlapping regions for different devices */
	list_for_each_entry(entry, &mappings, list) {
		dma_addr_t start, end, addr;

646 647
		if (domain->ops->apply_resv_region)
			domain->ops->apply_resv_region(dev, domain, entry);
648

649 650 651
		start = ALIGN(entry->start, pg_size);
		end   = ALIGN(entry->start + entry->length, pg_size);

652 653
		if (entry->type != IOMMU_RESV_DIRECT &&
		    entry->type != IOMMU_RESV_DIRECT_RELAXABLE)
654 655
			continue;

656 657 658 659 660 661 662 663 664 665 666 667 668 669
		for (addr = start; addr < end; addr += pg_size) {
			phys_addr_t phys_addr;

			phys_addr = iommu_iova_to_phys(domain, addr);
			if (phys_addr)
				continue;

			ret = iommu_map(domain, addr, addr, pg_size, entry->prot);
			if (ret)
				goto out;
		}

	}

670 671
	iommu_flush_tlb_all(domain);

672
out:
673
	iommu_put_resv_regions(dev, &mappings);
674 675 676 677

	return ret;
}

A
Alex Williamson 已提交
678 679 680 681 682 683 684 685 686
/**
 * iommu_group_add_device - add a device to an iommu group
 * @group: the group into which to add the device (reference should be held)
 * @dev: the device
 *
 * This function is called by an iommu driver to add a device into a
 * group.  Adding a device increments the group reference count.
 */
int iommu_group_add_device(struct iommu_group *group, struct device *dev)
687
{
A
Alex Williamson 已提交
688
	int ret, i = 0;
J
Joerg Roedel 已提交
689
	struct group_device *device;
A
Alex Williamson 已提交
690 691 692 693 694 695

	device = kzalloc(sizeof(*device), GFP_KERNEL);
	if (!device)
		return -ENOMEM;

	device->dev = dev;
696

A
Alex Williamson 已提交
697
	ret = sysfs_create_link(&dev->kobj, &group->kobj, "iommu_group");
698 699
	if (ret)
		goto err_free_device;
A
Alex Williamson 已提交
700 701 702 703

	device->name = kasprintf(GFP_KERNEL, "%s", kobject_name(&dev->kobj));
rename:
	if (!device->name) {
704 705
		ret = -ENOMEM;
		goto err_remove_link;
A
Alex Williamson 已提交
706
	}
707

A
Alex Williamson 已提交
708 709 710 711 712 713 714 715
	ret = sysfs_create_link_nowarn(group->devices_kobj,
				       &dev->kobj, device->name);
	if (ret) {
		if (ret == -EEXIST && i >= 0) {
			/*
			 * Account for the slim chance of collision
			 * and append an instance to the name.
			 */
716
			kfree(device->name);
A
Alex Williamson 已提交
717 718 719 720
			device->name = kasprintf(GFP_KERNEL, "%s.%d",
						 kobject_name(&dev->kobj), i++);
			goto rename;
		}
721
		goto err_free_name;
A
Alex Williamson 已提交
722 723 724 725 726 727
	}

	kobject_get(group->devices_kobj);

	dev->iommu_group = group;

728 729
	iommu_group_create_direct_mappings(group, dev);

A
Alex Williamson 已提交
730 731
	mutex_lock(&group->mutex);
	list_add_tail(&device->list, &group->devices);
732
	if (group->domain)
733
		ret = __iommu_attach_device(group->domain, dev);
A
Alex Williamson 已提交
734
	mutex_unlock(&group->mutex);
735 736
	if (ret)
		goto err_put_group;
A
Alex Williamson 已提交
737 738 739 740

	/* Notify any listeners about change to group. */
	blocking_notifier_call_chain(&group->notifier,
				     IOMMU_GROUP_NOTIFY_ADD_DEVICE, dev);
741 742

	trace_add_device_to_group(group->id, dev);
743

744
	dev_info(dev, "Adding to iommu group %d\n", group->id);
745

746
	return 0;
747 748 749 750 751 752 753 754 755 756 757 758 759

err_put_group:
	mutex_lock(&group->mutex);
	list_del(&device->list);
	mutex_unlock(&group->mutex);
	dev->iommu_group = NULL;
	kobject_put(group->devices_kobj);
err_free_name:
	kfree(device->name);
err_remove_link:
	sysfs_remove_link(&dev->kobj, "iommu_group");
err_free_device:
	kfree(device);
760
	dev_err(dev, "Failed to add to iommu group %d: %d\n", group->id, ret);
761
	return ret;
762
}
A
Alex Williamson 已提交
763
EXPORT_SYMBOL_GPL(iommu_group_add_device);
764

A
Alex Williamson 已提交
765 766 767 768 769 770 771 772 773 774
/**
 * iommu_group_remove_device - remove a device from it's current group
 * @dev: device to be removed
 *
 * This function is called by an iommu driver to remove the device from
 * it's current group.  This decrements the iommu group reference count.
 */
void iommu_group_remove_device(struct device *dev)
{
	struct iommu_group *group = dev->iommu_group;
J
Joerg Roedel 已提交
775
	struct group_device *tmp_device, *device = NULL;
A
Alex Williamson 已提交
776

777
	dev_info(dev, "Removing from iommu group %d\n", group->id);
778

A
Alex Williamson 已提交
779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798
	/* Pre-notify listeners that a device is being removed. */
	blocking_notifier_call_chain(&group->notifier,
				     IOMMU_GROUP_NOTIFY_DEL_DEVICE, dev);

	mutex_lock(&group->mutex);
	list_for_each_entry(tmp_device, &group->devices, list) {
		if (tmp_device->dev == dev) {
			device = tmp_device;
			list_del(&device->list);
			break;
		}
	}
	mutex_unlock(&group->mutex);

	if (!device)
		return;

	sysfs_remove_link(group->devices_kobj, device->name);
	sysfs_remove_link(&dev->kobj, "iommu_group");

799 800
	trace_remove_device_from_group(group->id, dev);

A
Alex Williamson 已提交
801 802 803 804 805 806 807
	kfree(device->name);
	kfree(device);
	dev->iommu_group = NULL;
	kobject_put(group->devices_kobj);
}
EXPORT_SYMBOL_GPL(iommu_group_remove_device);

808 809
static int iommu_group_device_count(struct iommu_group *group)
{
J
Joerg Roedel 已提交
810
	struct group_device *entry;
811 812 813 814 815 816 817 818
	int ret = 0;

	list_for_each_entry(entry, &group->devices, list)
		ret++;

	return ret;
}

A
Alex Williamson 已提交
819 820 821 822 823 824 825 826 827 828 829
/**
 * iommu_group_for_each_dev - iterate over each device in the group
 * @group: the group
 * @data: caller opaque data to be passed to callback function
 * @fn: caller supplied callback function
 *
 * This function is called by group users to iterate over group devices.
 * Callers should hold a reference count to the group during callback.
 * The group->mutex is held across callbacks, which will block calls to
 * iommu_group_add/remove_device.
 */
830 831
static int __iommu_group_for_each_dev(struct iommu_group *group, void *data,
				      int (*fn)(struct device *, void *))
A
Alex Williamson 已提交
832
{
J
Joerg Roedel 已提交
833
	struct group_device *device;
A
Alex Williamson 已提交
834 835 836 837 838 839 840
	int ret = 0;

	list_for_each_entry(device, &group->devices, list) {
		ret = fn(device->dev, data);
		if (ret)
			break;
	}
841 842 843 844 845 846 847 848 849 850 851
	return ret;
}


int iommu_group_for_each_dev(struct iommu_group *group, void *data,
			     int (*fn)(struct device *, void *))
{
	int ret;

	mutex_lock(&group->mutex);
	ret = __iommu_group_for_each_dev(group, data, fn);
A
Alex Williamson 已提交
852
	mutex_unlock(&group->mutex);
853

A
Alex Williamson 已提交
854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876
	return ret;
}
EXPORT_SYMBOL_GPL(iommu_group_for_each_dev);

/**
 * iommu_group_get - Return the group for a device and increment reference
 * @dev: get the group that this device belongs to
 *
 * This function is called by iommu drivers and users to get the group
 * for the specified device.  If found, the group is returned and the group
 * reference in incremented, else NULL.
 */
struct iommu_group *iommu_group_get(struct device *dev)
{
	struct iommu_group *group = dev->iommu_group;

	if (group)
		kobject_get(group->devices_kobj);

	return group;
}
EXPORT_SYMBOL_GPL(iommu_group_get);

877 878 879 880 881 882 883 884 885 886 887 888 889
/**
 * iommu_group_ref_get - Increment reference on a group
 * @group: the group to use, must not be NULL
 *
 * This function is called by iommu drivers to take additional references on an
 * existing group.  Returns the given group for convenience.
 */
struct iommu_group *iommu_group_ref_get(struct iommu_group *group)
{
	kobject_get(group->devices_kobj);
	return group;
}

A
Alex Williamson 已提交
890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933
/**
 * iommu_group_put - Decrement group reference
 * @group: the group to use
 *
 * This function is called by iommu drivers and users to release the
 * iommu group.  Once the reference count is zero, the group is released.
 */
void iommu_group_put(struct iommu_group *group)
{
	if (group)
		kobject_put(group->devices_kobj);
}
EXPORT_SYMBOL_GPL(iommu_group_put);

/**
 * iommu_group_register_notifier - Register a notifier for group changes
 * @group: the group to watch
 * @nb: notifier block to signal
 *
 * This function allows iommu group users to track changes in a group.
 * See include/linux/iommu.h for actions sent via this notifier.  Caller
 * should hold a reference to the group throughout notifier registration.
 */
int iommu_group_register_notifier(struct iommu_group *group,
				  struct notifier_block *nb)
{
	return blocking_notifier_chain_register(&group->notifier, nb);
}
EXPORT_SYMBOL_GPL(iommu_group_register_notifier);

/**
 * iommu_group_unregister_notifier - Unregister a notifier
 * @group: the group to watch
 * @nb: notifier block to signal
 *
 * Unregister a previously registered group notifier block.
 */
int iommu_group_unregister_notifier(struct iommu_group *group,
				    struct notifier_block *nb)
{
	return blocking_notifier_chain_unregister(&group->notifier, nb);
}
EXPORT_SYMBOL_GPL(iommu_group_unregister_notifier);

934 935 936 937 938 939 940
/**
 * iommu_register_device_fault_handler() - Register a device fault handler
 * @dev: the device
 * @handler: the fault handler
 * @data: private data passed as argument to the handler
 *
 * When an IOMMU fault event is received, this handler gets called with the
941 942 943 944 945 946 947 948
 * fault event and data as argument. The handler should return 0 on success. If
 * the fault is recoverable (IOMMU_FAULT_PAGE_REQ), the consumer should also
 * complete the fault by calling iommu_page_response() with one of the following
 * response code:
 * - IOMMU_PAGE_RESP_SUCCESS: retry the translation
 * - IOMMU_PAGE_RESP_INVALID: terminate the fault
 * - IOMMU_PAGE_RESP_FAILURE: terminate the fault and stop reporting
 *   page faults if possible.
949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977
 *
 * Return 0 if the fault handler was installed successfully, or an error.
 */
int iommu_register_device_fault_handler(struct device *dev,
					iommu_dev_fault_handler_t handler,
					void *data)
{
	struct iommu_param *param = dev->iommu_param;
	int ret = 0;

	if (!param)
		return -EINVAL;

	mutex_lock(&param->lock);
	/* Only allow one fault handler registered for each device */
	if (param->fault_param) {
		ret = -EBUSY;
		goto done_unlock;
	}

	get_device(dev);
	param->fault_param = kzalloc(sizeof(*param->fault_param), GFP_KERNEL);
	if (!param->fault_param) {
		put_device(dev);
		ret = -ENOMEM;
		goto done_unlock;
	}
	param->fault_param->handler = handler;
	param->fault_param->data = data;
978 979
	mutex_init(&param->fault_param->lock);
	INIT_LIST_HEAD(&param->fault_param->faults);
980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009

done_unlock:
	mutex_unlock(&param->lock);

	return ret;
}
EXPORT_SYMBOL_GPL(iommu_register_device_fault_handler);

/**
 * iommu_unregister_device_fault_handler() - Unregister the device fault handler
 * @dev: the device
 *
 * Remove the device fault handler installed with
 * iommu_register_device_fault_handler().
 *
 * Return 0 on success, or an error.
 */
int iommu_unregister_device_fault_handler(struct device *dev)
{
	struct iommu_param *param = dev->iommu_param;
	int ret = 0;

	if (!param)
		return -EINVAL;

	mutex_lock(&param->lock);

	if (!param->fault_param)
		goto unlock;

1010 1011 1012 1013 1014 1015
	/* we cannot unregister handler if there are pending faults */
	if (!list_empty(&param->fault_param->faults)) {
		ret = -EBUSY;
		goto unlock;
	}

1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031
	kfree(param->fault_param);
	param->fault_param = NULL;
	put_device(dev);
unlock:
	mutex_unlock(&param->lock);

	return ret;
}
EXPORT_SYMBOL_GPL(iommu_unregister_device_fault_handler);

/**
 * iommu_report_device_fault() - Report fault event to device driver
 * @dev: the device
 * @evt: fault event data
 *
 * Called by IOMMU drivers when a fault is detected, typically in a threaded IRQ
1032 1033
 * handler. When this function fails and the fault is recoverable, it is the
 * caller's responsibility to complete the fault.
1034 1035 1036 1037 1038 1039
 *
 * Return 0 on success, or an error.
 */
int iommu_report_device_fault(struct device *dev, struct iommu_fault_event *evt)
{
	struct iommu_param *param = dev->iommu_param;
1040
	struct iommu_fault_event *evt_pending = NULL;
1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053
	struct iommu_fault_param *fparam;
	int ret = 0;

	if (!param || !evt)
		return -EINVAL;

	/* we only report device fault if there is a handler registered */
	mutex_lock(&param->lock);
	fparam = param->fault_param;
	if (!fparam || !fparam->handler) {
		ret = -EINVAL;
		goto done_unlock;
	}
1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067

	if (evt->fault.type == IOMMU_FAULT_PAGE_REQ &&
	    (evt->fault.prm.flags & IOMMU_FAULT_PAGE_REQUEST_LAST_PAGE)) {
		evt_pending = kmemdup(evt, sizeof(struct iommu_fault_event),
				      GFP_KERNEL);
		if (!evt_pending) {
			ret = -ENOMEM;
			goto done_unlock;
		}
		mutex_lock(&fparam->lock);
		list_add_tail(&evt_pending->list, &fparam->faults);
		mutex_unlock(&fparam->lock);
	}

1068
	ret = fparam->handler(&evt->fault, fparam->data);
1069 1070 1071 1072 1073 1074
	if (ret && evt_pending) {
		mutex_lock(&fparam->lock);
		list_del(&evt_pending->list);
		mutex_unlock(&fparam->lock);
		kfree(evt_pending);
	}
1075 1076 1077 1078 1079 1080
done_unlock:
	mutex_unlock(&param->lock);
	return ret;
}
EXPORT_SYMBOL_GPL(iommu_report_device_fault);

1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133
int iommu_page_response(struct device *dev,
			struct iommu_page_response *msg)
{
	bool pasid_valid;
	int ret = -EINVAL;
	struct iommu_fault_event *evt;
	struct iommu_fault_page_request *prm;
	struct iommu_param *param = dev->iommu_param;
	struct iommu_domain *domain = iommu_get_domain_for_dev(dev);

	if (!domain || !domain->ops->page_response)
		return -ENODEV;

	if (!param || !param->fault_param)
		return -EINVAL;

	if (msg->version != IOMMU_PAGE_RESP_VERSION_1 ||
	    msg->flags & ~IOMMU_PAGE_RESP_PASID_VALID)
		return -EINVAL;

	/* Only send response if there is a fault report pending */
	mutex_lock(&param->fault_param->lock);
	if (list_empty(&param->fault_param->faults)) {
		dev_warn_ratelimited(dev, "no pending PRQ, drop response\n");
		goto done_unlock;
	}
	/*
	 * Check if we have a matching page request pending to respond,
	 * otherwise return -EINVAL
	 */
	list_for_each_entry(evt, &param->fault_param->faults, list) {
		prm = &evt->fault.prm;
		pasid_valid = prm->flags & IOMMU_FAULT_PAGE_REQUEST_PASID_VALID;

		if ((pasid_valid && prm->pasid != msg->pasid) ||
		    prm->grpid != msg->grpid)
			continue;

		/* Sanitize the reply */
		msg->flags = pasid_valid ? IOMMU_PAGE_RESP_PASID_VALID : 0;

		ret = domain->ops->page_response(dev, evt, msg);
		list_del(&evt->list);
		kfree(evt);
		break;
	}

done_unlock:
	mutex_unlock(&param->fault_param->lock);
	return ret;
}
EXPORT_SYMBOL_GPL(iommu_page_response);

A
Alex Williamson 已提交
1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144
/**
 * iommu_group_id - Return ID for a group
 * @group: the group to ID
 *
 * Return the unique ID for the group matching the sysfs group number.
 */
int iommu_group_id(struct iommu_group *group)
{
	return group->id;
}
EXPORT_SYMBOL_GPL(iommu_group_id);
1145

1146 1147 1148
static struct iommu_group *get_pci_alias_group(struct pci_dev *pdev,
					       unsigned long *devfns);

1149 1150 1151 1152 1153 1154 1155 1156 1157 1158
/*
 * To consider a PCI device isolated, we require ACS to support Source
 * Validation, Request Redirection, Completer Redirection, and Upstream
 * Forwarding.  This effectively means that devices cannot spoof their
 * requester ID, requests and completions cannot be redirected, and all
 * transactions are forwarded upstream, even as it passes through a
 * bridge where the target device is downstream.
 */
#define REQ_ACS_FLAGS   (PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF)

1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190
/*
 * For multifunction devices which are not isolated from each other, find
 * all the other non-isolated functions and look for existing groups.  For
 * each function, we also need to look for aliases to or from other devices
 * that may already have a group.
 */
static struct iommu_group *get_pci_function_alias_group(struct pci_dev *pdev,
							unsigned long *devfns)
{
	struct pci_dev *tmp = NULL;
	struct iommu_group *group;

	if (!pdev->multifunction || pci_acs_enabled(pdev, REQ_ACS_FLAGS))
		return NULL;

	for_each_pci_dev(tmp) {
		if (tmp == pdev || tmp->bus != pdev->bus ||
		    PCI_SLOT(tmp->devfn) != PCI_SLOT(pdev->devfn) ||
		    pci_acs_enabled(tmp, REQ_ACS_FLAGS))
			continue;

		group = get_pci_alias_group(tmp, devfns);
		if (group) {
			pci_dev_put(tmp);
			return group;
		}
	}

	return NULL;
}

/*
1191 1192
 * Look for aliases to or from the given device for existing groups. DMA
 * aliases are only supported on the same bus, therefore the search
1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216
 * space is quite small (especially since we're really only looking at pcie
 * device, and therefore only expect multiple slots on the root complex or
 * downstream switch ports).  It's conceivable though that a pair of
 * multifunction devices could have aliases between them that would cause a
 * loop.  To prevent this, we use a bitmap to track where we've been.
 */
static struct iommu_group *get_pci_alias_group(struct pci_dev *pdev,
					       unsigned long *devfns)
{
	struct pci_dev *tmp = NULL;
	struct iommu_group *group;

	if (test_and_set_bit(pdev->devfn & 0xff, devfns))
		return NULL;

	group = iommu_group_get(&pdev->dev);
	if (group)
		return group;

	for_each_pci_dev(tmp) {
		if (tmp == pdev || tmp->bus != pdev->bus)
			continue;

		/* We alias them or they alias us */
1217
		if (pci_devs_are_dma_aliases(pdev, tmp)) {
1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234
			group = get_pci_alias_group(tmp, devfns);
			if (group) {
				pci_dev_put(tmp);
				return group;
			}

			group = get_pci_function_alias_group(tmp, devfns);
			if (group) {
				pci_dev_put(tmp);
				return group;
			}
		}
	}

	return NULL;
}

1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253
struct group_for_pci_data {
	struct pci_dev *pdev;
	struct iommu_group *group;
};

/*
 * DMA alias iterator callback, return the last seen device.  Stop and return
 * the IOMMU group if we find one along the way.
 */
static int get_pci_alias_or_group(struct pci_dev *pdev, u16 alias, void *opaque)
{
	struct group_for_pci_data *data = opaque;

	data->pdev = pdev;
	data->group = iommu_group_get(&pdev->dev);

	return data->group != NULL;
}

1254 1255 1256 1257 1258 1259
/*
 * Generic device_group call-back function. It just allocates one
 * iommu-group per device.
 */
struct iommu_group *generic_device_group(struct device *dev)
{
1260
	return iommu_group_alloc();
1261 1262
}

1263 1264 1265 1266
/*
 * Use standard PCI bus topology, isolation features, and DMA alias quirks
 * to find or create an IOMMU group for a device.
 */
1267
struct iommu_group *pci_device_group(struct device *dev)
1268
{
1269
	struct pci_dev *pdev = to_pci_dev(dev);
1270 1271 1272
	struct group_for_pci_data data;
	struct pci_bus *bus;
	struct iommu_group *group = NULL;
1273
	u64 devfns[4] = { 0 };
1274

1275 1276 1277
	if (WARN_ON(!dev_is_pci(dev)))
		return ERR_PTR(-EINVAL);

1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309
	/*
	 * Find the upstream DMA alias for the device.  A device must not
	 * be aliased due to topology in order to have its own IOMMU group.
	 * If we find an alias along the way that already belongs to a
	 * group, use it.
	 */
	if (pci_for_each_dma_alias(pdev, get_pci_alias_or_group, &data))
		return data.group;

	pdev = data.pdev;

	/*
	 * Continue upstream from the point of minimum IOMMU granularity
	 * due to aliases to the point where devices are protected from
	 * peer-to-peer DMA by PCI ACS.  Again, if we find an existing
	 * group, use it.
	 */
	for (bus = pdev->bus; !pci_is_root_bus(bus); bus = bus->parent) {
		if (!bus->self)
			continue;

		if (pci_acs_path_enabled(bus->self, NULL, REQ_ACS_FLAGS))
			break;

		pdev = bus->self;

		group = iommu_group_get(&pdev->dev);
		if (group)
			return group;
	}

	/*
1310 1311
	 * Look for existing groups on device aliases.  If we alias another
	 * device or another device aliases us, use the same group.
1312
	 */
1313 1314 1315
	group = get_pci_alias_group(pdev, (unsigned long *)devfns);
	if (group)
		return group;
1316 1317

	/*
1318 1319 1320
	 * Look for existing groups on non-isolated functions on the same
	 * slot and aliases of those funcions, if any.  No need to clear
	 * the search bitmap, the tested devfns are still valid.
1321
	 */
1322 1323 1324
	group = get_pci_function_alias_group(pdev, (unsigned long *)devfns);
	if (group)
		return group;
1325 1326

	/* No shared group found, allocate new */
1327
	return iommu_group_alloc();
1328 1329
}

1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341
/* Get the IOMMU group for device on fsl-mc bus */
struct iommu_group *fsl_mc_device_group(struct device *dev)
{
	struct device *cont_dev = fsl_mc_cont_dev(dev);
	struct iommu_group *group;

	group = iommu_group_get(cont_dev);
	if (!group)
		group = iommu_group_alloc();
	return group;
}

1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353
/**
 * iommu_group_get_for_dev - Find or create the IOMMU group for a device
 * @dev: target device
 *
 * This function is intended to be called by IOMMU drivers and extended to
 * support common, bus-defined algorithms when determining or creating the
 * IOMMU group for a device.  On success, the caller will hold a reference
 * to the returned IOMMU group, which will already include the provided
 * device.  The reference should be released with iommu_group_put().
 */
struct iommu_group *iommu_group_get_for_dev(struct device *dev)
{
1354
	const struct iommu_ops *ops = dev->bus->iommu_ops;
1355
	struct iommu_group *group;
1356 1357 1358 1359 1360 1361
	int ret;

	group = iommu_group_get(dev);
	if (group)
		return group;

1362 1363
	if (!ops)
		return ERR_PTR(-EINVAL);
1364

1365
	group = ops->device_group(dev);
1366 1367 1368
	if (WARN_ON_ONCE(group == NULL))
		return ERR_PTR(-EINVAL);

1369 1370 1371
	if (IS_ERR(group))
		return group;

1372 1373 1374 1375 1376
	/*
	 * Try to allocate a default domain - needs support from the
	 * IOMMU driver.
	 */
	if (!group->default_domain) {
1377 1378 1379 1380 1381
		struct iommu_domain *dom;

		dom = __iommu_domain_alloc(dev->bus, iommu_def_domain_type);
		if (!dom && iommu_def_domain_type != IOMMU_DOMAIN_DMA) {
			dom = __iommu_domain_alloc(dev->bus, IOMMU_DOMAIN_DMA);
1382 1383 1384 1385 1386
			if (dom) {
				dev_warn(dev,
					 "failed to allocate default IOMMU domain of type %u; falling back to IOMMU_DOMAIN_DMA",
					 iommu_def_domain_type);
			}
1387 1388 1389
		}

		group->default_domain = dom;
1390
		if (!group->domain)
1391
			group->domain = dom;
1392 1393 1394 1395 1396 1397 1398

		if (dom && !iommu_dma_strict) {
			int attr = 1;
			iommu_domain_set_attr(dom,
					      DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE,
					      &attr);
		}
1399 1400
	}

1401 1402 1403 1404 1405 1406 1407 1408 1409
	ret = iommu_group_add_device(group, dev);
	if (ret) {
		iommu_group_put(group);
		return ERR_PTR(ret);
	}

	return group;
}

1410 1411 1412 1413 1414
struct iommu_domain *iommu_group_default_domain(struct iommu_group *group)
{
	return group->default_domain;
}

A
Alex Williamson 已提交
1415
static int add_iommu_group(struct device *dev, void *data)
1416
{
1417
	int ret = iommu_probe_device(dev);
1418 1419 1420 1421 1422 1423 1424 1425 1426 1427

	/*
	 * We ignore -ENODEV errors for now, as they just mean that the
	 * device is not translated by an IOMMU. We still care about
	 * other errors and fail to initialize when they happen.
	 */
	if (ret == -ENODEV)
		ret = 0;

	return ret;
1428 1429
}

1430 1431
static int remove_iommu_group(struct device *dev, void *data)
{
1432
	iommu_release_device(dev);
1433 1434 1435 1436

	return 0;
}

A
Alex Williamson 已提交
1437 1438
static int iommu_bus_notifier(struct notifier_block *nb,
			      unsigned long action, void *data)
1439
{
1440
	unsigned long group_action = 0;
1441
	struct device *dev = data;
A
Alex Williamson 已提交
1442 1443 1444 1445 1446 1447 1448
	struct iommu_group *group;

	/*
	 * ADD/DEL call into iommu driver ops if provided, which may
	 * result in ADD/DEL notifiers to group->notifier
	 */
	if (action == BUS_NOTIFY_ADD_DEVICE) {
1449
		int ret;
1450

1451 1452
		ret = iommu_probe_device(dev);
		return (ret) ? NOTIFY_DONE : NOTIFY_OK;
1453
	} else if (action == BUS_NOTIFY_REMOVED_DEVICE) {
1454 1455
		iommu_release_device(dev);
		return NOTIFY_OK;
A
Alex Williamson 已提交
1456
	}
1457

A
Alex Williamson 已提交
1458 1459 1460 1461 1462 1463 1464
	/*
	 * Remaining BUS_NOTIFYs get filtered and republished to the
	 * group, if anyone is listening
	 */
	group = iommu_group_get(dev);
	if (!group)
		return 0;
1465

A
Alex Williamson 已提交
1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479
	switch (action) {
	case BUS_NOTIFY_BIND_DRIVER:
		group_action = IOMMU_GROUP_NOTIFY_BIND_DRIVER;
		break;
	case BUS_NOTIFY_BOUND_DRIVER:
		group_action = IOMMU_GROUP_NOTIFY_BOUND_DRIVER;
		break;
	case BUS_NOTIFY_UNBIND_DRIVER:
		group_action = IOMMU_GROUP_NOTIFY_UNBIND_DRIVER;
		break;
	case BUS_NOTIFY_UNBOUND_DRIVER:
		group_action = IOMMU_GROUP_NOTIFY_UNBOUND_DRIVER;
		break;
	}
1480

A
Alex Williamson 已提交
1481 1482 1483
	if (group_action)
		blocking_notifier_call_chain(&group->notifier,
					     group_action, dev);
1484

A
Alex Williamson 已提交
1485
	iommu_group_put(group);
1486 1487 1488
	return 0;
}

M
Mark Salter 已提交
1489
static int iommu_bus_init(struct bus_type *bus, const struct iommu_ops *ops)
1490
{
M
Mark Salter 已提交
1491 1492
	int err;
	struct notifier_block *nb;
1493

M
Mark Salter 已提交
1494 1495 1496 1497 1498 1499 1500
	nb = kzalloc(sizeof(struct notifier_block), GFP_KERNEL);
	if (!nb)
		return -ENOMEM;

	nb->notifier_call = iommu_bus_notifier;

	err = bus_register_notifier(bus, nb);
1501 1502
	if (err)
		goto out_free;
1503

L
Lu Baolu 已提交
1504
	err = bus_for_each_dev(bus, NULL, NULL, add_iommu_group);
1505 1506 1507
	if (err)
		goto out_err;

1508 1509

	return 0;
1510 1511 1512

out_err:
	/* Clean up */
L
Lu Baolu 已提交
1513
	bus_for_each_dev(bus, NULL, NULL, remove_iommu_group);
1514 1515 1516 1517 1518 1519
	bus_unregister_notifier(bus, nb);

out_free:
	kfree(nb);

	return err;
1520
}
1521

1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534
/**
 * bus_set_iommu - set iommu-callbacks for the bus
 * @bus: bus.
 * @ops: the callbacks provided by the iommu-driver
 *
 * This function is called by an iommu driver to set the iommu methods
 * used for a particular bus. Drivers for devices on that bus can use
 * the iommu-api after these ops are registered.
 * This special function is needed because IOMMUs are usually devices on
 * the bus itself, so the iommu drivers are not initialized when the bus
 * is set up. With this function the iommu-driver can set the iommu-ops
 * afterwards.
 */
1535
int bus_set_iommu(struct bus_type *bus, const struct iommu_ops *ops)
1536
{
1537 1538
	int err;

1539 1540
	if (bus->iommu_ops != NULL)
		return -EBUSY;
1541

1542 1543 1544
	bus->iommu_ops = ops;

	/* Do IOMMU specific setup for this bus-type */
1545 1546 1547 1548 1549
	err = iommu_bus_init(bus, ops);
	if (err)
		bus->iommu_ops = NULL;

	return err;
1550
}
1551
EXPORT_SYMBOL_GPL(bus_set_iommu);
1552

1553
bool iommu_present(struct bus_type *bus)
1554
{
1555
	return bus->iommu_ops != NULL;
1556
}
1557
EXPORT_SYMBOL_GPL(iommu_present);
1558

1559 1560 1561 1562 1563 1564 1565 1566 1567
bool iommu_capable(struct bus_type *bus, enum iommu_cap cap)
{
	if (!bus->iommu_ops || !bus->iommu_ops->capable)
		return false;

	return bus->iommu_ops->capable(cap);
}
EXPORT_SYMBOL_GPL(iommu_capable);

1568 1569 1570 1571
/**
 * iommu_set_fault_handler() - set a fault handler for an iommu domain
 * @domain: iommu domain
 * @handler: fault handler
1572
 * @token: user data, will be passed back to the fault handler
1573 1574 1575 1576 1577 1578
 *
 * This function should be used by IOMMU users which want to be notified
 * whenever an IOMMU fault happens.
 *
 * The fault handler itself should return 0 on success, and an appropriate
 * error code otherwise.
1579 1580
 */
void iommu_set_fault_handler(struct iommu_domain *domain,
1581 1582
					iommu_fault_handler_t handler,
					void *token)
1583 1584 1585 1586
{
	BUG_ON(!domain);

	domain->handler = handler;
1587
	domain->handler_token = token;
1588
}
1589
EXPORT_SYMBOL_GPL(iommu_set_fault_handler);
1590

1591 1592
static struct iommu_domain *__iommu_domain_alloc(struct bus_type *bus,
						 unsigned type)
1593 1594 1595
{
	struct iommu_domain *domain;

1596
	if (bus == NULL || bus->iommu_ops == NULL)
1597 1598
		return NULL;

1599
	domain = bus->iommu_ops->domain_alloc(type);
1600 1601 1602
	if (!domain)
		return NULL;

1603
	domain->ops  = bus->iommu_ops;
1604
	domain->type = type;
1605 1606
	/* Assume all sizes by default; the driver may override this later */
	domain->pgsize_bitmap  = bus->iommu_ops->pgsize_bitmap;
1607

1608 1609 1610
	return domain;
}

1611 1612 1613
struct iommu_domain *iommu_domain_alloc(struct bus_type *bus)
{
	return __iommu_domain_alloc(bus, IOMMU_DOMAIN_UNMANAGED);
1614 1615 1616 1617 1618
}
EXPORT_SYMBOL_GPL(iommu_domain_alloc);

void iommu_domain_free(struct iommu_domain *domain)
{
1619
	domain->ops->domain_free(domain);
1620 1621 1622
}
EXPORT_SYMBOL_GPL(iommu_domain_free);

1623 1624
static int __iommu_attach_device(struct iommu_domain *domain,
				 struct device *dev)
1625
{
1626
	int ret;
1627 1628 1629 1630
	if ((domain->ops->is_attach_deferred != NULL) &&
	    domain->ops->is_attach_deferred(domain, dev))
		return 0;

1631 1632 1633
	if (unlikely(domain->ops->attach_dev == NULL))
		return -ENODEV;

1634 1635 1636 1637
	ret = domain->ops->attach_dev(domain, dev);
	if (!ret)
		trace_attach_device_to_domain(dev);
	return ret;
1638
}
1639 1640 1641 1642 1643 1644 1645

int iommu_attach_device(struct iommu_domain *domain, struct device *dev)
{
	struct iommu_group *group;
	int ret;

	group = iommu_group_get(dev);
1646 1647 1648
	if (!group)
		return -ENODEV;

1649
	/*
1650
	 * Lock the group to make sure the device-count doesn't
1651 1652 1653 1654 1655 1656 1657
	 * change while we are attaching
	 */
	mutex_lock(&group->mutex);
	ret = -EINVAL;
	if (iommu_group_device_count(group) != 1)
		goto out_unlock;

1658
	ret = __iommu_attach_group(domain, group);
1659 1660 1661 1662 1663 1664 1665

out_unlock:
	mutex_unlock(&group->mutex);
	iommu_group_put(group);

	return ret;
}
1666 1667
EXPORT_SYMBOL_GPL(iommu_attach_device);

Y
Yi L Liu 已提交
1668 1669 1670 1671 1672 1673 1674 1675 1676 1677
int iommu_cache_invalidate(struct iommu_domain *domain, struct device *dev,
			   struct iommu_cache_invalidate_info *inv_info)
{
	if (unlikely(!domain->ops->cache_invalidate))
		return -ENODEV;

	return domain->ops->cache_invalidate(domain, dev, inv_info);
}
EXPORT_SYMBOL_GPL(iommu_cache_invalidate);

1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697
int iommu_sva_bind_gpasid(struct iommu_domain *domain,
			   struct device *dev, struct iommu_gpasid_bind_data *data)
{
	if (unlikely(!domain->ops->sva_bind_gpasid))
		return -ENODEV;

	return domain->ops->sva_bind_gpasid(domain, dev, data);
}
EXPORT_SYMBOL_GPL(iommu_sva_bind_gpasid);

int iommu_sva_unbind_gpasid(struct iommu_domain *domain, struct device *dev,
			     ioasid_t pasid)
{
	if (unlikely(!domain->ops->sva_unbind_gpasid))
		return -ENODEV;

	return domain->ops->sva_unbind_gpasid(dev, pasid);
}
EXPORT_SYMBOL_GPL(iommu_sva_unbind_gpasid);

1698 1699
static void __iommu_detach_device(struct iommu_domain *domain,
				  struct device *dev)
1700
{
1701 1702 1703 1704
	if ((domain->ops->is_attach_deferred != NULL) &&
	    domain->ops->is_attach_deferred(domain, dev))
		return;

1705 1706 1707 1708
	if (unlikely(domain->ops->detach_dev == NULL))
		return;

	domain->ops->detach_dev(domain, dev);
1709
	trace_detach_device_from_domain(dev);
1710
}
1711 1712 1713 1714 1715 1716

void iommu_detach_device(struct iommu_domain *domain, struct device *dev)
{
	struct iommu_group *group;

	group = iommu_group_get(dev);
1717 1718
	if (!group)
		return;
1719 1720 1721 1722 1723 1724 1725

	mutex_lock(&group->mutex);
	if (iommu_group_device_count(group) != 1) {
		WARN_ON(1);
		goto out_unlock;
	}

1726
	__iommu_detach_group(domain, group);
1727 1728 1729 1730 1731

out_unlock:
	mutex_unlock(&group->mutex);
	iommu_group_put(group);
}
1732 1733
EXPORT_SYMBOL_GPL(iommu_detach_device);

1734 1735 1736 1737 1738 1739
struct iommu_domain *iommu_get_domain_for_dev(struct device *dev)
{
	struct iommu_domain *domain;
	struct iommu_group *group;

	group = iommu_group_get(dev);
1740
	if (!group)
1741 1742 1743 1744 1745 1746 1747 1748 1749
		return NULL;

	domain = group->domain;

	iommu_group_put(group);

	return domain;
}
EXPORT_SYMBOL_GPL(iommu_get_domain_for_dev);
1750

A
Alex Williamson 已提交
1751
/*
1752 1753 1754 1755 1756 1757 1758 1759
 * For IOMMU_DOMAIN_DMA implementations which already provide their own
 * guarantees that the group and its default domain are valid and correct.
 */
struct iommu_domain *iommu_get_dma_domain(struct device *dev)
{
	return dev->iommu_group->default_domain;
}

A
Alex Williamson 已提交
1760
/*
R
Rami Rosen 已提交
1761
 * IOMMU groups are really the natural working unit of the IOMMU, but
A
Alex Williamson 已提交
1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773
 * the IOMMU API works on domains and devices.  Bridge that gap by
 * iterating over the devices in a group.  Ideally we'd have a single
 * device which represents the requestor ID of the group, but we also
 * allow IOMMU drivers to create policy defined minimum sets, where
 * the physical hardware may be able to distiguish members, but we
 * wish to group them at a higher level (ex. untrusted multi-function
 * PCI devices).  Thus we attach each device.
 */
static int iommu_group_do_attach_device(struct device *dev, void *data)
{
	struct iommu_domain *domain = data;

1774
	return __iommu_attach_device(domain, dev);
A
Alex Williamson 已提交
1775 1776
}

1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790
static int __iommu_attach_group(struct iommu_domain *domain,
				struct iommu_group *group)
{
	int ret;

	if (group->default_domain && group->domain != group->default_domain)
		return -EBUSY;

	ret = __iommu_group_for_each_dev(group, domain,
					 iommu_group_do_attach_device);
	if (ret == 0)
		group->domain = domain;

	return ret;
A
Alex Williamson 已提交
1791 1792 1793 1794
}

int iommu_attach_group(struct iommu_domain *domain, struct iommu_group *group)
{
1795 1796 1797 1798 1799 1800 1801
	int ret;

	mutex_lock(&group->mutex);
	ret = __iommu_attach_group(domain, group);
	mutex_unlock(&group->mutex);

	return ret;
A
Alex Williamson 已提交
1802 1803 1804 1805 1806 1807 1808
}
EXPORT_SYMBOL_GPL(iommu_attach_group);

static int iommu_group_do_detach_device(struct device *dev, void *data)
{
	struct iommu_domain *domain = data;

1809
	__iommu_detach_device(domain, dev);
A
Alex Williamson 已提交
1810 1811 1812 1813

	return 0;
}

1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837
static void __iommu_detach_group(struct iommu_domain *domain,
				 struct iommu_group *group)
{
	int ret;

	if (!group->default_domain) {
		__iommu_group_for_each_dev(group, domain,
					   iommu_group_do_detach_device);
		group->domain = NULL;
		return;
	}

	if (group->domain == group->default_domain)
		return;

	/* Detach by re-attaching to the default domain */
	ret = __iommu_group_for_each_dev(group, group->default_domain,
					 iommu_group_do_attach_device);
	if (ret != 0)
		WARN_ON(1);
	else
		group->domain = group->default_domain;
}

A
Alex Williamson 已提交
1838 1839
void iommu_detach_group(struct iommu_domain *domain, struct iommu_group *group)
{
1840 1841 1842
	mutex_lock(&group->mutex);
	__iommu_detach_group(domain, group);
	mutex_unlock(&group->mutex);
A
Alex Williamson 已提交
1843 1844 1845
}
EXPORT_SYMBOL_GPL(iommu_detach_group);

1846
phys_addr_t iommu_iova_to_phys(struct iommu_domain *domain, dma_addr_t iova)
1847
{
1848 1849 1850 1851
	if (unlikely(domain->ops->iova_to_phys == NULL))
		return 0;

	return domain->ops->iova_to_phys(domain, iova);
1852 1853
}
EXPORT_SYMBOL_GPL(iommu_iova_to_phys);
S
Sheng Yang 已提交
1854

A
Alex Williamson 已提交
1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874
static size_t iommu_pgsize(struct iommu_domain *domain,
			   unsigned long addr_merge, size_t size)
{
	unsigned int pgsize_idx;
	size_t pgsize;

	/* Max page size that still fits into 'size' */
	pgsize_idx = __fls(size);

	/* need to consider alignment requirements ? */
	if (likely(addr_merge)) {
		/* Max page size allowed by address */
		unsigned int align_pgsize_idx = __ffs(addr_merge);
		pgsize_idx = min(pgsize_idx, align_pgsize_idx);
	}

	/* build a mask of acceptable page sizes */
	pgsize = (1UL << (pgsize_idx + 1)) - 1;

	/* throw away page sizes not supported by the hardware */
1875
	pgsize &= domain->pgsize_bitmap;
A
Alex Williamson 已提交
1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886

	/* make sure we're still sane */
	BUG_ON(!pgsize);

	/* pick the biggest page */
	pgsize_idx = __fls(pgsize);
	pgsize = 1UL << pgsize_idx;

	return pgsize;
}

1887 1888
int __iommu_map(struct iommu_domain *domain, unsigned long iova,
	      phys_addr_t paddr, size_t size, int prot, gfp_t gfp)
1889
{
1890
	const struct iommu_ops *ops = domain->ops;
1891 1892 1893
	unsigned long orig_iova = iova;
	unsigned int min_pagesz;
	size_t orig_size = size;
1894
	phys_addr_t orig_paddr = paddr;
1895
	int ret = 0;
1896

1897
	if (unlikely(ops->map == NULL ||
1898
		     domain->pgsize_bitmap == 0UL))
1899
		return -ENODEV;
1900

1901 1902 1903
	if (unlikely(!(domain->type & __IOMMU_DOMAIN_PAGING)))
		return -EINVAL;

1904
	/* find out the minimum page size supported */
1905
	min_pagesz = 1 << __ffs(domain->pgsize_bitmap);
1906 1907 1908 1909 1910 1911 1912

	/*
	 * both the virtual address and the physical one, as well as
	 * the size of the mapping, must be aligned (at least) to the
	 * size of the smallest page supported by the hardware
	 */
	if (!IS_ALIGNED(iova | paddr | size, min_pagesz)) {
1913
		pr_err("unaligned: iova 0x%lx pa %pa size 0x%zx min_pagesz 0x%x\n",
1914
		       iova, &paddr, size, min_pagesz);
1915 1916 1917
		return -EINVAL;
	}

1918
	pr_debug("map: iova 0x%lx pa %pa size 0x%zx\n", iova, &paddr, size);
1919 1920

	while (size) {
A
Alex Williamson 已提交
1921
		size_t pgsize = iommu_pgsize(domain, iova | paddr, size);
1922

1923
		pr_debug("mapping: iova 0x%lx pa %pa pgsize 0x%zx\n",
1924
			 iova, &paddr, pgsize);
1925
		ret = ops->map(domain, iova, paddr, pgsize, prot, gfp);
1926 1927 1928 1929 1930 1931 1932 1933 1934

		if (ret)
			break;

		iova += pgsize;
		paddr += pgsize;
		size -= pgsize;
	}

1935 1936 1937
	if (ops->iotlb_sync_map)
		ops->iotlb_sync_map(domain);

1938 1939 1940
	/* unroll mapping in case something went wrong */
	if (ret)
		iommu_unmap(domain, orig_iova, orig_size - size);
1941
	else
1942
		trace_map(orig_iova, orig_paddr, orig_size);
1943 1944

	return ret;
1945
}
1946 1947 1948 1949 1950 1951 1952

int iommu_map(struct iommu_domain *domain, unsigned long iova,
	      phys_addr_t paddr, size_t size, int prot)
{
	might_sleep();
	return __iommu_map(domain, iova, paddr, size, prot, GFP_KERNEL);
}
1953 1954
EXPORT_SYMBOL_GPL(iommu_map);

1955 1956 1957 1958 1959 1960 1961
int iommu_map_atomic(struct iommu_domain *domain, unsigned long iova,
	      phys_addr_t paddr, size_t size, int prot)
{
	return __iommu_map(domain, iova, paddr, size, prot, GFP_ATOMIC);
}
EXPORT_SYMBOL_GPL(iommu_map_atomic);

1962 1963
static size_t __iommu_unmap(struct iommu_domain *domain,
			    unsigned long iova, size_t size,
1964
			    struct iommu_iotlb_gather *iotlb_gather)
1965
{
1966
	const struct iommu_ops *ops = domain->ops;
1967
	size_t unmapped_page, unmapped = 0;
1968
	unsigned long orig_iova = iova;
1969
	unsigned int min_pagesz;
1970

1971
	if (unlikely(ops->unmap == NULL ||
1972
		     domain->pgsize_bitmap == 0UL))
1973
		return 0;
1974

1975
	if (unlikely(!(domain->type & __IOMMU_DOMAIN_PAGING)))
1976
		return 0;
1977

1978
	/* find out the minimum page size supported */
1979
	min_pagesz = 1 << __ffs(domain->pgsize_bitmap);
1980 1981 1982 1983 1984 1985 1986

	/*
	 * The virtual address, as well as the size of the mapping, must be
	 * aligned (at least) to the size of the smallest page supported
	 * by the hardware
	 */
	if (!IS_ALIGNED(iova | size, min_pagesz)) {
1987 1988
		pr_err("unaligned: iova 0x%lx size 0x%zx min_pagesz 0x%x\n",
		       iova, size, min_pagesz);
1989
		return 0;
1990 1991
	}

1992
	pr_debug("unmap this: iova 0x%lx size 0x%zx\n", iova, size);
1993 1994 1995 1996 1997 1998

	/*
	 * Keep iterating until we either unmap 'size' bytes (or more)
	 * or we hit an area that isn't mapped.
	 */
	while (unmapped < size) {
A
Alex Williamson 已提交
1999
		size_t pgsize = iommu_pgsize(domain, iova, size - unmapped);
2000

2001
		unmapped_page = ops->unmap(domain, iova, pgsize, iotlb_gather);
2002 2003 2004
		if (!unmapped_page)
			break;

2005 2006
		pr_debug("unmapped: iova 0x%lx size 0x%zx\n",
			 iova, unmapped_page);
2007 2008 2009 2010 2011

		iova += unmapped_page;
		unmapped += unmapped_page;
	}

2012
	trace_unmap(orig_iova, size, unmapped);
2013
	return unmapped;
2014
}
2015 2016 2017 2018

size_t iommu_unmap(struct iommu_domain *domain,
		   unsigned long iova, size_t size)
{
2019 2020 2021 2022 2023 2024 2025 2026
	struct iommu_iotlb_gather iotlb_gather;
	size_t ret;

	iommu_iotlb_gather_init(&iotlb_gather);
	ret = __iommu_unmap(domain, iova, size, &iotlb_gather);
	iommu_tlb_sync(domain, &iotlb_gather);

	return ret;
2027
}
2028
EXPORT_SYMBOL_GPL(iommu_unmap);
2029

2030
size_t iommu_unmap_fast(struct iommu_domain *domain,
2031 2032
			unsigned long iova, size_t size,
			struct iommu_iotlb_gather *iotlb_gather)
2033
{
2034
	return __iommu_unmap(domain, iova, size, iotlb_gather);
2035 2036 2037
}
EXPORT_SYMBOL_GPL(iommu_unmap_fast);

2038 2039 2040
size_t __iommu_map_sg(struct iommu_domain *domain, unsigned long iova,
		    struct scatterlist *sg, unsigned int nents, int prot,
		    gfp_t gfp)
O
Olav Haugan 已提交
2041
{
2042 2043 2044
	size_t len = 0, mapped = 0;
	phys_addr_t start;
	unsigned int i = 0;
2045
	int ret;
O
Olav Haugan 已提交
2046

2047 2048
	while (i <= nents) {
		phys_addr_t s_phys = sg_phys(sg);
2049

2050
		if (len && s_phys != start + len) {
2051 2052 2053
			ret = __iommu_map(domain, iova + mapped, start,
					len, prot, gfp);

2054 2055
			if (ret)
				goto out_err;
2056

2057 2058 2059
			mapped += len;
			len = 0;
		}
2060

2061 2062 2063 2064 2065 2066
		if (len) {
			len += sg->length;
		} else {
			len = sg->length;
			start = s_phys;
		}
2067

2068 2069
		if (++i < nents)
			sg = sg_next(sg);
O
Olav Haugan 已提交
2070 2071 2072
	}

	return mapped;
2073 2074 2075 2076 2077 2078 2079

out_err:
	/* undo mappings already done */
	iommu_unmap(domain, iova, mapped);

	return 0;

O
Olav Haugan 已提交
2080
}
2081 2082 2083 2084 2085 2086 2087

size_t iommu_map_sg(struct iommu_domain *domain, unsigned long iova,
		    struct scatterlist *sg, unsigned int nents, int prot)
{
	might_sleep();
	return __iommu_map_sg(domain, iova, sg, nents, prot, GFP_KERNEL);
}
2088
EXPORT_SYMBOL_GPL(iommu_map_sg);
2089

2090 2091 2092 2093 2094 2095 2096
size_t iommu_map_sg_atomic(struct iommu_domain *domain, unsigned long iova,
		    struct scatterlist *sg, unsigned int nents, int prot)
{
	return __iommu_map_sg(domain, iova, sg, nents, prot, GFP_ATOMIC);
}
EXPORT_SYMBOL_GPL(iommu_map_sg_atomic);

2097
int iommu_domain_window_enable(struct iommu_domain *domain, u32 wnd_nr,
2098
			       phys_addr_t paddr, u64 size, int prot)
2099 2100 2101 2102
{
	if (unlikely(domain->ops->domain_window_enable == NULL))
		return -ENODEV;

2103 2104
	return domain->ops->domain_window_enable(domain, wnd_nr, paddr, size,
						 prot);
2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116
}
EXPORT_SYMBOL_GPL(iommu_domain_window_enable);

void iommu_domain_window_disable(struct iommu_domain *domain, u32 wnd_nr)
{
	if (unlikely(domain->ops->domain_window_disable == NULL))
		return;

	return domain->ops->domain_window_disable(domain, wnd_nr);
}
EXPORT_SYMBOL_GPL(iommu_domain_window_disable);

2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158
/**
 * report_iommu_fault() - report about an IOMMU fault to the IOMMU framework
 * @domain: the iommu domain where the fault has happened
 * @dev: the device where the fault has happened
 * @iova: the faulting address
 * @flags: mmu fault flags (e.g. IOMMU_FAULT_READ/IOMMU_FAULT_WRITE/...)
 *
 * This function should be called by the low-level IOMMU implementations
 * whenever IOMMU faults happen, to allow high-level users, that are
 * interested in such events, to know about them.
 *
 * This event may be useful for several possible use cases:
 * - mere logging of the event
 * - dynamic TLB/PTE loading
 * - if restarting of the faulting device is required
 *
 * Returns 0 on success and an appropriate error code otherwise (if dynamic
 * PTE/TLB loading will one day be supported, implementations will be able
 * to tell whether it succeeded or not according to this return value).
 *
 * Specifically, -ENOSYS is returned if a fault handler isn't installed
 * (though fault handlers can also return -ENOSYS, in case they want to
 * elicit the default behavior of the IOMMU drivers).
 */
int report_iommu_fault(struct iommu_domain *domain, struct device *dev,
		       unsigned long iova, int flags)
{
	int ret = -ENOSYS;

	/*
	 * if upper layers showed interest and installed a fault handler,
	 * invoke it.
	 */
	if (domain->handler)
		ret = domain->handler(domain, dev, iova, flags,
						domain->handler_token);

	trace_io_page_fault(dev, iova, flags);
	return ret;
}
EXPORT_SYMBOL_GPL(report_iommu_fault);

A
Alex Williamson 已提交
2159
static int __init iommu_init(void)
2160
{
A
Alex Williamson 已提交
2161 2162 2163 2164
	iommu_group_kset = kset_create_and_add("iommu_groups",
					       NULL, kernel_kobj);
	BUG_ON(!iommu_group_kset);

2165 2166
	iommu_debugfs_setup();

A
Alex Williamson 已提交
2167
	return 0;
2168
}
2169
core_initcall(iommu_init);
2170 2171 2172 2173

int iommu_domain_get_attr(struct iommu_domain *domain,
			  enum iommu_attr attr, void *data)
{
2174
	struct iommu_domain_geometry *geometry;
2175
	bool *paging;
2176 2177 2178 2179 2180 2181 2182
	int ret = 0;

	switch (attr) {
	case DOMAIN_ATTR_GEOMETRY:
		geometry  = data;
		*geometry = domain->geometry;

2183 2184 2185
		break;
	case DOMAIN_ATTR_PAGING:
		paging  = data;
2186
		*paging = (domain->pgsize_bitmap != 0UL);
2187 2188 2189 2190
		break;
	default:
		if (!domain->ops->domain_get_attr)
			return -EINVAL;
2191

2192 2193 2194 2195
		ret = domain->ops->domain_get_attr(domain, attr, data);
	}

	return ret;
2196 2197 2198 2199 2200 2201
}
EXPORT_SYMBOL_GPL(iommu_domain_get_attr);

int iommu_domain_set_attr(struct iommu_domain *domain,
			  enum iommu_attr attr, void *data)
{
2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212
	int ret = 0;

	switch (attr) {
	default:
		if (domain->ops->domain_set_attr == NULL)
			return -EINVAL;

		ret = domain->ops->domain_set_attr(domain, attr, data);
	}

	return ret;
2213
}
2214
EXPORT_SYMBOL_GPL(iommu_domain_set_attr);
2215

2216
void iommu_get_resv_regions(struct device *dev, struct list_head *list)
2217 2218 2219
{
	const struct iommu_ops *ops = dev->bus->iommu_ops;

2220 2221
	if (ops && ops->get_resv_regions)
		ops->get_resv_regions(dev, list);
2222 2223
}

2224
void iommu_put_resv_regions(struct device *dev, struct list_head *list)
2225 2226 2227
{
	const struct iommu_ops *ops = dev->bus->iommu_ops;

2228 2229
	if (ops && ops->put_resv_regions)
		ops->put_resv_regions(dev, list);
2230
}
2231

E
Eric Auger 已提交
2232
struct iommu_resv_region *iommu_alloc_resv_region(phys_addr_t start,
2233 2234
						  size_t length, int prot,
						  enum iommu_resv_type type)
E
Eric Auger 已提交
2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247
{
	struct iommu_resv_region *region;

	region = kzalloc(sizeof(*region), GFP_KERNEL);
	if (!region)
		return NULL;

	INIT_LIST_HEAD(&region->list);
	region->start = start;
	region->length = length;
	region->prot = prot;
	region->type = type;
	return region;
2248
}
2249

2250 2251
static int
request_default_domain_for_dev(struct device *dev, unsigned long type)
2252
{
2253
	struct iommu_domain *domain;
2254 2255 2256 2257
	struct iommu_group *group;
	int ret;

	/* Device must already be in a group before calling this function */
2258 2259 2260
	group = iommu_group_get(dev);
	if (!group)
		return -EINVAL;
2261 2262 2263 2264

	mutex_lock(&group->mutex);

	ret = 0;
2265
	if (group->default_domain && group->default_domain->type == type)
2266 2267 2268 2269 2270 2271 2272 2273
		goto out;

	/* Don't change mappings of existing devices */
	ret = -EBUSY;
	if (iommu_group_device_count(group) != 1)
		goto out;

	ret = -ENOMEM;
2274 2275
	domain = __iommu_domain_alloc(dev->bus, type);
	if (!domain)
2276 2277 2278
		goto out;

	/* Attach the device to the domain */
2279
	ret = __iommu_attach_group(domain, group);
2280
	if (ret) {
2281
		iommu_domain_free(domain);
2282 2283 2284
		goto out;
	}

2285
	/* Make the domain the default for this group */
2286 2287
	if (group->default_domain)
		iommu_domain_free(group->default_domain);
2288
	group->default_domain = domain;
2289

2290 2291
	iommu_group_create_direct_mappings(group, dev);

2292 2293
	dev_info(dev, "Using iommu %s mapping\n",
		 type == IOMMU_DOMAIN_DMA ? "dma" : "direct");
2294 2295 2296 2297 2298 2299 2300 2301

	ret = 0;
out:
	mutex_unlock(&group->mutex);
	iommu_group_put(group);

	return ret;
}
R
Robin Murphy 已提交
2302

2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314
/* Request that a device is direct mapped by the IOMMU */
int iommu_request_dm_for_dev(struct device *dev)
{
	return request_default_domain_for_dev(dev, IOMMU_DOMAIN_IDENTITY);
}

/* Request that a device can't be direct mapped by the IOMMU */
int iommu_request_dma_domain_for_dev(struct device *dev)
{
	return request_default_domain_for_dev(dev, IOMMU_DOMAIN_DMA);
}

2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336
void iommu_set_default_passthrough(bool cmd_line)
{
	if (cmd_line)
		iommu_set_cmd_line_dma_api();

	iommu_def_domain_type = IOMMU_DOMAIN_IDENTITY;
}

void iommu_set_default_translated(bool cmd_line)
{
	if (cmd_line)
		iommu_set_cmd_line_dma_api();

	iommu_def_domain_type = IOMMU_DOMAIN_DMA;
}

bool iommu_default_passthrough(void)
{
	return iommu_def_domain_type == IOMMU_DOMAIN_IDENTITY;
}
EXPORT_SYMBOL_GPL(iommu_default_passthrough);

2337
const struct iommu_ops *iommu_ops_from_fwnode(struct fwnode_handle *fwnode)
2338 2339
{
	const struct iommu_ops *ops = NULL;
2340
	struct iommu_device *iommu;
2341

2342 2343 2344 2345
	spin_lock(&iommu_device_lock);
	list_for_each_entry(iommu, &iommu_device_list, list)
		if (iommu->fwnode == fwnode) {
			ops = iommu->ops;
2346 2347
			break;
		}
2348
	spin_unlock(&iommu_device_lock);
2349 2350 2351
	return ops;
}

R
Robin Murphy 已提交
2352 2353 2354
int iommu_fwspec_init(struct device *dev, struct fwnode_handle *iommu_fwnode,
		      const struct iommu_ops *ops)
{
2355
	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
R
Robin Murphy 已提交
2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366

	if (fwspec)
		return ops == fwspec->ops ? 0 : -EINVAL;

	fwspec = kzalloc(sizeof(*fwspec), GFP_KERNEL);
	if (!fwspec)
		return -ENOMEM;

	of_node_get(to_of_node(iommu_fwnode));
	fwspec->iommu_fwnode = iommu_fwnode;
	fwspec->ops = ops;
2367
	dev_iommu_fwspec_set(dev, fwspec);
R
Robin Murphy 已提交
2368 2369 2370 2371 2372 2373
	return 0;
}
EXPORT_SYMBOL_GPL(iommu_fwspec_init);

void iommu_fwspec_free(struct device *dev)
{
2374
	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
R
Robin Murphy 已提交
2375 2376 2377 2378

	if (fwspec) {
		fwnode_handle_put(fwspec->iommu_fwnode);
		kfree(fwspec);
2379
		dev_iommu_fwspec_set(dev, NULL);
R
Robin Murphy 已提交
2380 2381 2382 2383 2384 2385
	}
}
EXPORT_SYMBOL_GPL(iommu_fwspec_free);

int iommu_fwspec_add_ids(struct device *dev, u32 *ids, int num_ids)
{
2386
	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
R
Robin Murphy 已提交
2387 2388 2389 2390 2391 2392 2393 2394
	size_t size;
	int i;

	if (!fwspec)
		return -EINVAL;

	size = offsetof(struct iommu_fwspec, ids[fwspec->num_ids + num_ids]);
	if (size > sizeof(*fwspec)) {
2395
		fwspec = krealloc(fwspec, size, GFP_KERNEL);
R
Robin Murphy 已提交
2396 2397
		if (!fwspec)
			return -ENOMEM;
2398

2399
		dev_iommu_fwspec_set(dev, fwspec);
R
Robin Murphy 已提交
2400 2401 2402 2403 2404 2405 2406 2407 2408
	}

	for (i = 0; i < num_ids; i++)
		fwspec->ids[fwspec->num_ids + i] = ids[i];

	fwspec->num_ids += num_ids;
	return 0;
}
EXPORT_SYMBOL_GPL(iommu_fwspec_add_ids);
2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 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

/*
 * Per device IOMMU features.
 */
bool iommu_dev_has_feature(struct device *dev, enum iommu_dev_features feat)
{
	const struct iommu_ops *ops = dev->bus->iommu_ops;

	if (ops && ops->dev_has_feat)
		return ops->dev_has_feat(dev, feat);

	return false;
}
EXPORT_SYMBOL_GPL(iommu_dev_has_feature);

int iommu_dev_enable_feature(struct device *dev, enum iommu_dev_features feat)
{
	const struct iommu_ops *ops = dev->bus->iommu_ops;

	if (ops && ops->dev_enable_feat)
		return ops->dev_enable_feat(dev, feat);

	return -ENODEV;
}
EXPORT_SYMBOL_GPL(iommu_dev_enable_feature);

/*
 * The device drivers should do the necessary cleanups before calling this.
 * For example, before disabling the aux-domain feature, the device driver
 * should detach all aux-domains. Otherwise, this will return -EBUSY.
 */
int iommu_dev_disable_feature(struct device *dev, enum iommu_dev_features feat)
{
	const struct iommu_ops *ops = dev->bus->iommu_ops;

	if (ops && ops->dev_disable_feat)
		return ops->dev_disable_feat(dev, feat);

	return -EBUSY;
}
EXPORT_SYMBOL_GPL(iommu_dev_disable_feature);

bool iommu_dev_feature_enabled(struct device *dev, enum iommu_dev_features feat)
{
	const struct iommu_ops *ops = dev->bus->iommu_ops;

	if (ops && ops->dev_feat_enabled)
		return ops->dev_feat_enabled(dev, feat);

	return false;
}
EXPORT_SYMBOL_GPL(iommu_dev_feature_enabled);

/*
 * Aux-domain specific attach/detach.
 *
 * Only works if iommu_dev_feature_enabled(dev, IOMMU_DEV_FEAT_AUX) returns
 * true. Also, as long as domains are attached to a device through this
 * interface, any tries to call iommu_attach_device() should fail
 * (iommu_detach_device() can't fail, so we fail when trying to re-attach).
 * This should make us safe against a device being attached to a guest as a
 * whole while there are still pasid users on it (aux and sva).
 */
int iommu_aux_attach_device(struct iommu_domain *domain, struct device *dev)
{
	int ret = -ENODEV;

	if (domain->ops->aux_attach_dev)
		ret = domain->ops->aux_attach_dev(domain, dev);

	if (!ret)
		trace_attach_device_to_domain(dev);

	return ret;
}
EXPORT_SYMBOL_GPL(iommu_aux_attach_device);

void iommu_aux_detach_device(struct iommu_domain *domain, struct device *dev)
{
	if (domain->ops->aux_detach_dev) {
		domain->ops->aux_detach_dev(domain, dev);
		trace_detach_device_from_domain(dev);
	}
}
EXPORT_SYMBOL_GPL(iommu_aux_detach_device);

int iommu_aux_get_pasid(struct iommu_domain *domain, struct device *dev)
{
	int ret = -ENODEV;

	if (domain->ops->aux_get_pasid)
		ret = domain->ops->aux_get_pasid(domain, dev);

	return ret;
}
EXPORT_SYMBOL_GPL(iommu_aux_get_pasid);
2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 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 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608

/**
 * iommu_sva_bind_device() - Bind a process address space to a device
 * @dev: the device
 * @mm: the mm to bind, caller must hold a reference to it
 *
 * Create a bond between device and address space, allowing the device to access
 * the mm using the returned PASID. If a bond already exists between @device and
 * @mm, it is returned and an additional reference is taken. Caller must call
 * iommu_sva_unbind_device() to release each reference.
 *
 * iommu_dev_enable_feature(dev, IOMMU_DEV_FEAT_SVA) must be called first, to
 * initialize the required SVA features.
 *
 * On error, returns an ERR_PTR value.
 */
struct iommu_sva *
iommu_sva_bind_device(struct device *dev, struct mm_struct *mm, void *drvdata)
{
	struct iommu_group *group;
	struct iommu_sva *handle = ERR_PTR(-EINVAL);
	const struct iommu_ops *ops = dev->bus->iommu_ops;

	if (!ops || !ops->sva_bind)
		return ERR_PTR(-ENODEV);

	group = iommu_group_get(dev);
	if (!group)
		return ERR_PTR(-ENODEV);

	/* Ensure device count and domain don't change while we're binding */
	mutex_lock(&group->mutex);

	/*
	 * To keep things simple, SVA currently doesn't support IOMMU groups
	 * with more than one device. Existing SVA-capable systems are not
	 * affected by the problems that required IOMMU groups (lack of ACS
	 * isolation, device ID aliasing and other hardware issues).
	 */
	if (iommu_group_device_count(group) != 1)
		goto out_unlock;

	handle = ops->sva_bind(dev, mm, drvdata);

out_unlock:
	mutex_unlock(&group->mutex);
	iommu_group_put(group);

	return handle;
}
EXPORT_SYMBOL_GPL(iommu_sva_bind_device);

/**
 * iommu_sva_unbind_device() - Remove a bond created with iommu_sva_bind_device
 * @handle: the handle returned by iommu_sva_bind_device()
 *
 * Put reference to a bond between device and address space. The device should
 * not be issuing any more transaction for this PASID. All outstanding page
 * requests for this PASID must have been flushed to the IOMMU.
 *
 * Returns 0 on success, or an error value
 */
void iommu_sva_unbind_device(struct iommu_sva *handle)
{
	struct iommu_group *group;
	struct device *dev = handle->dev;
	const struct iommu_ops *ops = dev->bus->iommu_ops;

	if (!ops || !ops->sva_unbind)
		return;

	group = iommu_group_get(dev);
	if (!group)
		return;

	mutex_lock(&group->mutex);
	ops->sva_unbind(handle);
	mutex_unlock(&group->mutex);

	iommu_group_put(group);
}
EXPORT_SYMBOL_GPL(iommu_sva_unbind_device);

int iommu_sva_set_ops(struct iommu_sva *handle,
		      const struct iommu_sva_ops *sva_ops)
{
	if (handle->ops && handle->ops != sva_ops)
		return -EEXIST;

	handle->ops = sva_ops;
	return 0;
}
EXPORT_SYMBOL_GPL(iommu_sva_set_ops);

int iommu_sva_get_pasid(struct iommu_sva *handle)
{
	const struct iommu_ops *ops = handle->dev->bus->iommu_ops;

	if (!ops || !ops->sva_get_pasid)
		return IOMMU_PASID_INVALID;

	return ops->sva_get_pasid(handle);
}
EXPORT_SYMBOL_GPL(iommu_sva_get_pasid);