iommu.c 46.5 KB
Newer Older
1 2
/*
 * Copyright (C) 2007-2008 Advanced Micro Devices, Inc.
J
Joerg Roedel 已提交
3
 * Author: Joerg Roedel <jroedel@suse.de>
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
 *
 * This program is free software; you can redistribute it and/or modify it
 * under the terms of the GNU General Public License version 2 as published
 * by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
 */

19
#define pr_fmt(fmt)    "iommu: " fmt
20

21
#include <linux/device.h>
22
#include <linux/kernel.h>
23 24
#include <linux/bug.h>
#include <linux/types.h>
25 26
#include <linux/module.h>
#include <linux/slab.h>
27 28
#include <linux/errno.h>
#include <linux/iommu.h>
A
Alex Williamson 已提交
29 30 31
#include <linux/idr.h>
#include <linux/notifier.h>
#include <linux/err.h>
32
#include <linux/pci.h>
33
#include <linux/bitops.h>
R
Robin Murphy 已提交
34
#include <linux/property.h>
35
#include <trace/events/iommu.h>
A
Alex Williamson 已提交
36 37

static struct kset *iommu_group_kset;
H
Heiner Kallweit 已提交
38
static DEFINE_IDA(iommu_group_ida);
A
Alex Williamson 已提交
39

40 41 42 43
struct iommu_callback_data {
	const struct iommu_ops *ops;
};

A
Alex Williamson 已提交
44 45 46 47 48 49 50 51 52 53
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;
54
	struct iommu_domain *default_domain;
55
	struct iommu_domain *domain;
A
Alex Williamson 已提交
56 57
};

J
Joerg Roedel 已提交
58
struct group_device {
A
Alex Williamson 已提交
59 60 61 62 63 64 65 66 67 68 69 70
	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);
};

71 72 73 74 75 76
static const char * const iommu_group_resv_type_string[] = {
	[IOMMU_RESV_DIRECT]	= "direct",
	[IOMMU_RESV_RESERVED]	= "reserved",
	[IOMMU_RESV_MSI]	= "msi",
};

A
Alex Williamson 已提交
77 78 79
#define IOMMU_GROUP_ATTR(_name, _mode, _show, _store)		\
struct iommu_group_attribute iommu_group_attr_##_name =		\
	__ATTR(_name, _mode, _show, _store)
80

A
Alex Williamson 已提交
81 82 83 84
#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)
85

86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104
static LIST_HEAD(iommu_device_list);
static DEFINE_SPINLOCK(iommu_device_lock);

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);
}

105 106
static struct iommu_domain *__iommu_domain_alloc(struct bus_type *bus,
						 unsigned type);
107 108 109 110 111 112
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);
113

A
Alex Williamson 已提交
114 115
static ssize_t iommu_group_attr_show(struct kobject *kobj,
				     struct attribute *__attr, char *buf)
116
{
A
Alex Williamson 已提交
117 118 119
	struct iommu_group_attribute *attr = to_iommu_group_attr(__attr);
	struct iommu_group *group = to_iommu_group(kobj);
	ssize_t ret = -EIO;
120

A
Alex Williamson 已提交
121 122 123 124 125 126 127 128 129 130 131 132
	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;
133

A
Alex Williamson 已提交
134 135 136
	if (attr->store)
		ret = attr->store(group, buf, count);
	return ret;
137 138
}

A
Alex Williamson 已提交
139 140 141 142
static const struct sysfs_ops iommu_group_sysfs_ops = {
	.show = iommu_group_attr_show,
	.store = iommu_group_attr_store,
};
143

A
Alex Williamson 已提交
144 145 146 147
static int iommu_group_create_file(struct iommu_group *group,
				   struct iommu_group_attribute *attr)
{
	return sysfs_create_file(&group->kobj, &attr->attr);
148 149
}

A
Alex Williamson 已提交
150 151 152 153 154 155 156 157 158 159 160
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
161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226
/**
 * iommu_insert_resv_region - Insert a new region in the
 * list of reserved regions.
 * @new: new region to insert
 * @regions: list of regions
 *
 * The new element is sorted by address with respect to the other
 * regions of the same type. In case it overlaps with another
 * region of the same type, regions are merged. In case it
 * overlaps with another region of different type, regions are
 * not merged.
 */
static int iommu_insert_resv_region(struct iommu_resv_region *new,
				    struct list_head *regions)
{
	struct iommu_resv_region *region;
	phys_addr_t start = new->start;
	phys_addr_t end = new->start + new->length - 1;
	struct list_head *pos = regions->next;

	while (pos != regions) {
		struct iommu_resv_region *entry =
			list_entry(pos, struct iommu_resv_region, list);
		phys_addr_t a = entry->start;
		phys_addr_t b = entry->start + entry->length - 1;
		int type = entry->type;

		if (end < a) {
			goto insert;
		} else if (start > b) {
			pos = pos->next;
		} else if ((start >= a) && (end <= b)) {
			if (new->type == type)
				goto done;
			else
				pos = pos->next;
		} else {
			if (new->type == type) {
				phys_addr_t new_start = min(a, start);
				phys_addr_t new_end = max(b, end);

				list_del(&entry->list);
				entry->start = new_start;
				entry->length = new_end - new_start + 1;
				iommu_insert_resv_region(entry, regions);
			} else {
				pos = pos->next;
			}
		}
	}
insert:
	region = iommu_alloc_resv_region(new->start, new->length,
					 new->prot, new->type);
	if (!region)
		return -ENOMEM;

	list_add_tail(&region->list, pos);
done:
	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;
227
	int ret = 0;
E
Eric Auger 已提交
228 229 230 231 232 233 234 235 236 237 238 239

	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)
{
240
	struct group_device *device;
E
Eric Auger 已提交
241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258
	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);

259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280
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);
}

A
Alex Williamson 已提交
281 282
static IOMMU_GROUP_ATTR(name, S_IRUGO, iommu_group_show_name, NULL);

283 284 285
static IOMMU_GROUP_ATTR(reserved_regions, 0444,
			iommu_group_show_resv_regions, NULL);

A
Alex Williamson 已提交
286 287 288 289
static void iommu_group_release(struct kobject *kobj)
{
	struct iommu_group *group = to_iommu_group(kobj);

290 291
	pr_debug("Releasing group %d\n", group->id);

A
Alex Williamson 已提交
292 293 294
	if (group->iommu_data_release)
		group->iommu_data_release(group->iommu_data);

295
	ida_simple_remove(&iommu_group_ida, group->id);
A
Alex Williamson 已提交
296

297 298 299
	if (group->default_domain)
		iommu_domain_free(group->default_domain);

A
Alex Williamson 已提交
300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321
	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
 * @name: Optional name to associate with group, visible in sysfs
 *
 * 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)
322
{
A
Alex Williamson 已提交
323 324 325 326 327 328 329 330 331 332 333 334
	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);

335 336
	ret = ida_simple_get(&iommu_group_ida, 0, 0, GFP_KERNEL);
	if (ret < 0) {
A
Alex Williamson 已提交
337
		kfree(group);
338
		return ERR_PTR(ret);
A
Alex Williamson 已提交
339
	}
340
	group->id = ret;
341

A
Alex Williamson 已提交
342 343 344
	ret = kobject_init_and_add(&group->kobj, &iommu_group_ktype,
				   NULL, "%d", group->id);
	if (ret) {
345
		ida_simple_remove(&iommu_group_ida, group->id);
A
Alex Williamson 已提交
346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362
		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);

363 364 365 366 367
	ret = iommu_group_create_file(group,
				      &iommu_group_attr_reserved_regions);
	if (ret)
		return ERR_PTR(ret);

368 369
	pr_debug("Allocated group %d\n", group->id);

A
Alex Williamson 已提交
370 371 372 373
	return group;
}
EXPORT_SYMBOL_GPL(iommu_group_alloc);

374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402
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
403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428
/**
 * 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))
429
{
A
Alex Williamson 已提交
430 431 432 433
	group->iommu_data = iommu_data;
	group->iommu_data_release = release;
}
EXPORT_SYMBOL_GPL(iommu_group_set_iommudata);
434

A
Alex Williamson 已提交
435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464
/**
 * 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;
	}
465 466 467

	return 0;
}
A
Alex Williamson 已提交
468
EXPORT_SYMBOL_GPL(iommu_group_set_name);
469

470 471 472 473
static int iommu_group_create_direct_mappings(struct iommu_group *group,
					      struct device *dev)
{
	struct iommu_domain *domain = group->default_domain;
474
	struct iommu_resv_region *entry;
475 476 477 478 479 480 481
	struct list_head mappings;
	unsigned long pg_size;
	int ret = 0;

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

482
	BUG_ON(!domain->pgsize_bitmap);
483

484
	pg_size = 1UL << __ffs(domain->pgsize_bitmap);
485 486
	INIT_LIST_HEAD(&mappings);

487
	iommu_get_resv_regions(dev, &mappings);
488 489 490 491 492

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

493 494
		if (domain->ops->apply_resv_region)
			domain->ops->apply_resv_region(dev, domain, entry);
495

496 497 498
		start = ALIGN(entry->start, pg_size);
		end   = ALIGN(entry->start + entry->length, pg_size);

499 500 501
		if (entry->type != IOMMU_RESV_DIRECT)
			continue;

502 503 504 505 506 507 508 509 510 511 512 513 514 515 516
		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;
		}

	}

out:
517
	iommu_put_resv_regions(dev, &mappings);
518 519 520 521

	return ret;
}

A
Alex Williamson 已提交
522 523 524 525 526 527 528 529 530
/**
 * 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)
531
{
A
Alex Williamson 已提交
532
	int ret, i = 0;
J
Joerg Roedel 已提交
533
	struct group_device *device;
A
Alex Williamson 已提交
534 535 536 537 538 539

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

	device->dev = dev;
540

A
Alex Williamson 已提交
541
	ret = sysfs_create_link(&dev->kobj, &group->kobj, "iommu_group");
542 543
	if (ret)
		goto err_free_device;
A
Alex Williamson 已提交
544 545 546 547

	device->name = kasprintf(GFP_KERNEL, "%s", kobject_name(&dev->kobj));
rename:
	if (!device->name) {
548 549
		ret = -ENOMEM;
		goto err_remove_link;
A
Alex Williamson 已提交
550
	}
551

A
Alex Williamson 已提交
552 553 554 555 556 557 558 559
	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.
			 */
560
			kfree(device->name);
A
Alex Williamson 已提交
561 562 563 564
			device->name = kasprintf(GFP_KERNEL, "%s.%d",
						 kobject_name(&dev->kobj), i++);
			goto rename;
		}
565
		goto err_free_name;
A
Alex Williamson 已提交
566 567 568 569 570 571
	}

	kobject_get(group->devices_kobj);

	dev->iommu_group = group;

572 573
	iommu_group_create_direct_mappings(group, dev);

A
Alex Williamson 已提交
574 575
	mutex_lock(&group->mutex);
	list_add_tail(&device->list, &group->devices);
576
	if (group->domain)
577
		ret = __iommu_attach_device(group->domain, dev);
A
Alex Williamson 已提交
578
	mutex_unlock(&group->mutex);
579 580
	if (ret)
		goto err_put_group;
A
Alex Williamson 已提交
581 582 583 584

	/* Notify any listeners about change to group. */
	blocking_notifier_call_chain(&group->notifier,
				     IOMMU_GROUP_NOTIFY_ADD_DEVICE, dev);
585 586

	trace_add_device_to_group(group->id, dev);
587 588 589

	pr_info("Adding device %s to group %d\n", dev_name(dev), group->id);

590
	return 0;
591 592 593 594 595 596 597 598 599 600 601 602 603 604 605

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);
	pr_err("Failed to add device %s to group %d: %d\n", dev_name(dev), group->id, ret);
	return ret;
606
}
A
Alex Williamson 已提交
607
EXPORT_SYMBOL_GPL(iommu_group_add_device);
608

A
Alex Williamson 已提交
609 610 611 612 613 614 615 616 617 618
/**
 * 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
619
	struct group_device *tmp_device, *device = NULL;
A
Alex Williamson 已提交
620

621 622
	pr_info("Removing device %s from group %d\n", dev_name(dev), group->id);

A
Alex Williamson 已提交
623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642
	/* 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");

643 644
	trace_remove_device_from_group(group->id, dev);

A
Alex Williamson 已提交
645 646 647 648 649 650 651
	kfree(device->name);
	kfree(device);
	dev->iommu_group = NULL;
	kobject_put(group->devices_kobj);
}
EXPORT_SYMBOL_GPL(iommu_group_remove_device);

652 653
static int iommu_group_device_count(struct iommu_group *group)
{
J
Joerg Roedel 已提交
654
	struct group_device *entry;
655 656 657 658 659 660 661 662
	int ret = 0;

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

	return ret;
}

A
Alex Williamson 已提交
663 664 665 666 667 668 669 670 671 672 673
/**
 * 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.
 */
674 675
static int __iommu_group_for_each_dev(struct iommu_group *group, void *data,
				      int (*fn)(struct device *, void *))
A
Alex Williamson 已提交
676
{
J
Joerg Roedel 已提交
677
	struct group_device *device;
A
Alex Williamson 已提交
678 679 680 681 682 683 684
	int ret = 0;

	list_for_each_entry(device, &group->devices, list) {
		ret = fn(device->dev, data);
		if (ret)
			break;
	}
685 686 687 688 689 690 691 692 693 694 695
	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
696
	mutex_unlock(&group->mutex);
697

A
Alex Williamson 已提交
698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720
	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);

721 722 723 724 725 726 727 728 729 730 731 732 733
/**
 * 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
734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788
/**
 * 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);

/**
 * 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);
789

790 791 792
static struct iommu_group *get_pci_alias_group(struct pci_dev *pdev,
					       unsigned long *devfns);

793 794 795 796 797 798 799 800 801 802
/*
 * 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)

803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834
/*
 * 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;
}

/*
835 836
 * Look for aliases to or from the given device for existing groups. DMA
 * aliases are only supported on the same bus, therefore the search
837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860
 * 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 */
861
		if (pci_devs_are_dma_aliases(pdev, tmp)) {
862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878
			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;
}

879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897
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;
}

898 899 900 901 902 903 904 905 906 907 908 909 910 911 912
/*
 * Generic device_group call-back function. It just allocates one
 * iommu-group per device.
 */
struct iommu_group *generic_device_group(struct device *dev)
{
	struct iommu_group *group;

	group = iommu_group_alloc();
	if (IS_ERR(group))
		return NULL;

	return group;
}

913 914 915 916
/*
 * Use standard PCI bus topology, isolation features, and DMA alias quirks
 * to find or create an IOMMU group for a device.
 */
917
struct iommu_group *pci_device_group(struct device *dev)
918
{
919
	struct pci_dev *pdev = to_pci_dev(dev);
920 921 922
	struct group_for_pci_data data;
	struct pci_bus *bus;
	struct iommu_group *group = NULL;
923
	u64 devfns[4] = { 0 };
924

925 926 927
	if (WARN_ON(!dev_is_pci(dev)))
		return ERR_PTR(-EINVAL);

928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959
	/*
	 * 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;
	}

	/*
960 961
	 * Look for existing groups on device aliases.  If we alias another
	 * device or another device aliases us, use the same group.
962
	 */
963 964 965
	group = get_pci_alias_group(pdev, (unsigned long *)devfns);
	if (group)
		return group;
966 967

	/*
968 969 970
	 * 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.
971
	 */
972 973 974
	group = get_pci_function_alias_group(pdev, (unsigned long *)devfns);
	if (group)
		return group;
975 976

	/* No shared group found, allocate new */
977
	group = iommu_group_alloc();
978 979 980
	if (IS_ERR(group))
		return NULL;

981
	return group;
982 983 984 985 986 987 988 989 990 991 992 993 994 995
}

/**
 * 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)
{
996
	const struct iommu_ops *ops = dev->bus->iommu_ops;
997
	struct iommu_group *group;
998 999 1000 1001 1002 1003
	int ret;

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

1004
	group = ERR_PTR(-EINVAL);
1005

1006 1007
	if (ops && ops->device_group)
		group = ops->device_group(dev);
1008 1009 1010 1011

	if (IS_ERR(group))
		return group;

1012 1013 1014 1015 1016 1017 1018
	/*
	 * Try to allocate a default domain - needs support from the
	 * IOMMU driver.
	 */
	if (!group->default_domain) {
		group->default_domain = __iommu_domain_alloc(dev->bus,
							     IOMMU_DOMAIN_DMA);
1019 1020
		if (!group->domain)
			group->domain = group->default_domain;
1021 1022
	}

1023 1024 1025 1026 1027 1028 1029 1030 1031
	ret = iommu_group_add_device(group, dev);
	if (ret) {
		iommu_group_put(group);
		return ERR_PTR(ret);
	}

	return group;
}

1032 1033 1034 1035 1036
struct iommu_domain *iommu_group_default_domain(struct iommu_group *group)
{
	return group->default_domain;
}

A
Alex Williamson 已提交
1037
static int add_iommu_group(struct device *dev, void *data)
1038
{
1039 1040
	struct iommu_callback_data *cb = data;
	const struct iommu_ops *ops = cb->ops;
1041
	int ret;
A
Alex Williamson 已提交
1042 1043

	if (!ops->add_device)
1044
		return 0;
1045

A
Alex Williamson 已提交
1046 1047
	WARN_ON(dev->iommu_group);

1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058
	ret = ops->add_device(dev);

	/*
	 * 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;
1059 1060
}

1061 1062 1063 1064 1065 1066 1067
static int remove_iommu_group(struct device *dev, void *data)
{
	struct iommu_callback_data *cb = data;
	const struct iommu_ops *ops = cb->ops;

	if (ops->remove_device && dev->iommu_group)
		ops->remove_device(dev);
1068 1069 1070 1071

	return 0;
}

A
Alex Williamson 已提交
1072 1073
static int iommu_bus_notifier(struct notifier_block *nb,
			      unsigned long action, void *data)
1074 1075
{
	struct device *dev = data;
1076
	const struct iommu_ops *ops = dev->bus->iommu_ops;
A
Alex Williamson 已提交
1077 1078 1079 1080 1081 1082 1083 1084
	struct iommu_group *group;
	unsigned long group_action = 0;

	/*
	 * 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) {
1085 1086 1087 1088 1089 1090
		if (ops->add_device) {
			int ret;

			ret = ops->add_device(dev);
			return (ret) ? NOTIFY_DONE : NOTIFY_OK;
		}
1091
	} else if (action == BUS_NOTIFY_REMOVED_DEVICE) {
A
Alex Williamson 已提交
1092 1093 1094 1095 1096
		if (ops->remove_device && dev->iommu_group) {
			ops->remove_device(dev);
			return 0;
		}
	}
1097

A
Alex Williamson 已提交
1098 1099 1100 1101 1102 1103 1104
	/*
	 * Remaining BUS_NOTIFYs get filtered and republished to the
	 * group, if anyone is listening
	 */
	group = iommu_group_get(dev);
	if (!group)
		return 0;
1105

A
Alex Williamson 已提交
1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119
	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;
	}
1120

A
Alex Williamson 已提交
1121 1122 1123
	if (group_action)
		blocking_notifier_call_chain(&group->notifier,
					     group_action, dev);
1124

A
Alex Williamson 已提交
1125
	iommu_group_put(group);
1126 1127 1128
	return 0;
}

M
Mark Salter 已提交
1129
static int iommu_bus_init(struct bus_type *bus, const struct iommu_ops *ops)
1130
{
M
Mark Salter 已提交
1131 1132
	int err;
	struct notifier_block *nb;
1133 1134 1135 1136
	struct iommu_callback_data cb = {
		.ops = ops,
	};

M
Mark Salter 已提交
1137 1138 1139 1140 1141 1142 1143
	nb = kzalloc(sizeof(struct notifier_block), GFP_KERNEL);
	if (!nb)
		return -ENOMEM;

	nb->notifier_call = iommu_bus_notifier;

	err = bus_register_notifier(bus, nb);
1144 1145
	if (err)
		goto out_free;
1146 1147

	err = bus_for_each_dev(bus, NULL, &cb, add_iommu_group);
1148 1149 1150
	if (err)
		goto out_err;

1151 1152

	return 0;
1153 1154 1155 1156 1157 1158 1159 1160 1161 1162

out_err:
	/* Clean up */
	bus_for_each_dev(bus, NULL, &cb, remove_iommu_group);
	bus_unregister_notifier(bus, nb);

out_free:
	kfree(nb);

	return err;
1163
}
1164

1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177
/**
 * 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.
 */
1178
int bus_set_iommu(struct bus_type *bus, const struct iommu_ops *ops)
1179
{
1180 1181
	int err;

1182 1183
	if (bus->iommu_ops != NULL)
		return -EBUSY;
1184

1185 1186 1187
	bus->iommu_ops = ops;

	/* Do IOMMU specific setup for this bus-type */
1188 1189 1190 1191 1192
	err = iommu_bus_init(bus, ops);
	if (err)
		bus->iommu_ops = NULL;

	return err;
1193
}
1194
EXPORT_SYMBOL_GPL(bus_set_iommu);
1195

1196
bool iommu_present(struct bus_type *bus)
1197
{
1198
	return bus->iommu_ops != NULL;
1199
}
1200
EXPORT_SYMBOL_GPL(iommu_present);
1201

1202 1203 1204 1205 1206 1207 1208 1209 1210
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);

1211 1212 1213 1214
/**
 * iommu_set_fault_handler() - set a fault handler for an iommu domain
 * @domain: iommu domain
 * @handler: fault handler
1215
 * @token: user data, will be passed back to the fault handler
1216 1217 1218 1219 1220 1221
 *
 * 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.
1222 1223
 */
void iommu_set_fault_handler(struct iommu_domain *domain,
1224 1225
					iommu_fault_handler_t handler,
					void *token)
1226 1227 1228 1229
{
	BUG_ON(!domain);

	domain->handler = handler;
1230
	domain->handler_token = token;
1231
}
1232
EXPORT_SYMBOL_GPL(iommu_set_fault_handler);
1233

1234 1235
static struct iommu_domain *__iommu_domain_alloc(struct bus_type *bus,
						 unsigned type)
1236 1237 1238
{
	struct iommu_domain *domain;

1239
	if (bus == NULL || bus->iommu_ops == NULL)
1240 1241
		return NULL;

1242
	domain = bus->iommu_ops->domain_alloc(type);
1243 1244 1245
	if (!domain)
		return NULL;

1246
	domain->ops  = bus->iommu_ops;
1247
	domain->type = type;
1248 1249
	/* Assume all sizes by default; the driver may override this later */
	domain->pgsize_bitmap  = bus->iommu_ops->pgsize_bitmap;
1250

1251 1252 1253
	return domain;
}

1254 1255 1256
struct iommu_domain *iommu_domain_alloc(struct bus_type *bus)
{
	return __iommu_domain_alloc(bus, IOMMU_DOMAIN_UNMANAGED);
1257 1258 1259 1260 1261
}
EXPORT_SYMBOL_GPL(iommu_domain_alloc);

void iommu_domain_free(struct iommu_domain *domain)
{
1262
	domain->ops->domain_free(domain);
1263 1264 1265
}
EXPORT_SYMBOL_GPL(iommu_domain_free);

1266 1267
static int __iommu_attach_device(struct iommu_domain *domain,
				 struct device *dev)
1268
{
1269
	int ret;
1270 1271 1272
	if (unlikely(domain->ops->attach_dev == NULL))
		return -ENODEV;

1273 1274 1275 1276
	ret = domain->ops->attach_dev(domain, dev);
	if (!ret)
		trace_attach_device_to_domain(dev);
	return ret;
1277
}
1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297

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

	group = iommu_group_get(dev);
	/* FIXME: Remove this when groups a mandatory for iommu drivers */
	if (group == NULL)
		return __iommu_attach_device(domain, dev);

	/*
	 * We have a group - lock it to make sure the device-count doesn't
	 * change while we are attaching
	 */
	mutex_lock(&group->mutex);
	ret = -EINVAL;
	if (iommu_group_device_count(group) != 1)
		goto out_unlock;

1298
	ret = __iommu_attach_group(domain, group);
1299 1300 1301 1302 1303 1304 1305

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

	return ret;
}
1306 1307
EXPORT_SYMBOL_GPL(iommu_attach_device);

1308 1309
static void __iommu_detach_device(struct iommu_domain *domain,
				  struct device *dev)
1310
{
1311 1312 1313 1314
	if (unlikely(domain->ops->detach_dev == NULL))
		return;

	domain->ops->detach_dev(domain, dev);
1315
	trace_detach_device_from_domain(dev);
1316
}
1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332

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

	group = iommu_group_get(dev);
	/* FIXME: Remove this when groups a mandatory for iommu drivers */
	if (group == NULL)
		return __iommu_detach_device(domain, dev);

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

1333
	__iommu_detach_group(domain, group);
1334 1335 1336 1337 1338

out_unlock:
	mutex_unlock(&group->mutex);
	iommu_group_put(group);
}
1339 1340
EXPORT_SYMBOL_GPL(iommu_detach_device);

1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357
struct iommu_domain *iommu_get_domain_for_dev(struct device *dev)
{
	struct iommu_domain *domain;
	struct iommu_group *group;

	group = iommu_group_get(dev);
	/* FIXME: Remove this when groups a mandatory for iommu drivers */
	if (group == NULL)
		return NULL;

	domain = group->domain;

	iommu_group_put(group);

	return domain;
}
EXPORT_SYMBOL_GPL(iommu_get_domain_for_dev);
1358

A
Alex Williamson 已提交
1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372
/*
 * IOMMU groups are really the natrual working unit of the IOMMU, but
 * 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;

1373
	return __iommu_attach_device(domain, dev);
A
Alex Williamson 已提交
1374 1375
}

1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389
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
1390 1391 1392 1393
}

int iommu_attach_group(struct iommu_domain *domain, struct iommu_group *group)
{
1394 1395 1396 1397 1398 1399 1400
	int ret;

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

	return ret;
A
Alex Williamson 已提交
1401 1402 1403 1404 1405 1406 1407
}
EXPORT_SYMBOL_GPL(iommu_attach_group);

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

1408
	__iommu_detach_device(domain, dev);
A
Alex Williamson 已提交
1409 1410 1411 1412

	return 0;
}

1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436
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
1437 1438
void iommu_detach_group(struct iommu_domain *domain, struct iommu_group *group)
{
1439 1440 1441
	mutex_lock(&group->mutex);
	__iommu_detach_group(domain, group);
	mutex_unlock(&group->mutex);
A
Alex Williamson 已提交
1442 1443 1444
}
EXPORT_SYMBOL_GPL(iommu_detach_group);

1445
phys_addr_t iommu_iova_to_phys(struct iommu_domain *domain, dma_addr_t iova)
1446
{
1447 1448 1449 1450
	if (unlikely(domain->ops->iova_to_phys == NULL))
		return 0;

	return domain->ops->iova_to_phys(domain, iova);
1451 1452
}
EXPORT_SYMBOL_GPL(iommu_iova_to_phys);
S
Sheng Yang 已提交
1453

A
Alex Williamson 已提交
1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473
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 */
1474
	pgsize &= domain->pgsize_bitmap;
A
Alex Williamson 已提交
1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485

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

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

	return pgsize;
}

1486
int iommu_map(struct iommu_domain *domain, unsigned long iova,
1487
	      phys_addr_t paddr, size_t size, int prot)
1488
{
1489 1490 1491
	unsigned long orig_iova = iova;
	unsigned int min_pagesz;
	size_t orig_size = size;
1492
	phys_addr_t orig_paddr = paddr;
1493
	int ret = 0;
1494

1495
	if (unlikely(domain->ops->map == NULL ||
1496
		     domain->pgsize_bitmap == 0UL))
1497
		return -ENODEV;
1498

1499 1500 1501
	if (unlikely(!(domain->type & __IOMMU_DOMAIN_PAGING)))
		return -EINVAL;

1502
	/* find out the minimum page size supported */
1503
	min_pagesz = 1 << __ffs(domain->pgsize_bitmap);
1504 1505 1506 1507 1508 1509 1510

	/*
	 * 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)) {
1511
		pr_err("unaligned: iova 0x%lx pa %pa size 0x%zx min_pagesz 0x%x\n",
1512
		       iova, &paddr, size, min_pagesz);
1513 1514 1515
		return -EINVAL;
	}

1516
	pr_debug("map: iova 0x%lx pa %pa size 0x%zx\n", iova, &paddr, size);
1517 1518

	while (size) {
A
Alex Williamson 已提交
1519
		size_t pgsize = iommu_pgsize(domain, iova | paddr, size);
1520

1521
		pr_debug("mapping: iova 0x%lx pa %pa pgsize 0x%zx\n",
1522
			 iova, &paddr, pgsize);
1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535

		ret = domain->ops->map(domain, iova, paddr, pgsize, prot);
		if (ret)
			break;

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

	/* unroll mapping in case something went wrong */
	if (ret)
		iommu_unmap(domain, orig_iova, orig_size - size);
1536
	else
1537
		trace_map(orig_iova, orig_paddr, orig_size);
1538 1539

	return ret;
1540 1541 1542
}
EXPORT_SYMBOL_GPL(iommu_map);

1543
size_t iommu_unmap(struct iommu_domain *domain, unsigned long iova, size_t size)
1544
{
1545 1546
	size_t unmapped_page, unmapped = 0;
	unsigned int min_pagesz;
1547
	unsigned long orig_iova = iova;
1548

1549
	if (unlikely(domain->ops->unmap == NULL ||
1550
		     domain->pgsize_bitmap == 0UL))
1551 1552
		return -ENODEV;

1553 1554 1555
	if (unlikely(!(domain->type & __IOMMU_DOMAIN_PAGING)))
		return -EINVAL;

1556
	/* find out the minimum page size supported */
1557
	min_pagesz = 1 << __ffs(domain->pgsize_bitmap);
1558 1559 1560 1561 1562 1563 1564

	/*
	 * 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)) {
1565 1566
		pr_err("unaligned: iova 0x%lx size 0x%zx min_pagesz 0x%x\n",
		       iova, size, min_pagesz);
1567 1568 1569
		return -EINVAL;
	}

1570
	pr_debug("unmap this: iova 0x%lx size 0x%zx\n", iova, size);
1571 1572 1573 1574 1575 1576

	/*
	 * 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
1577
		size_t pgsize = iommu_pgsize(domain, iova, size - unmapped);
1578

A
Alex Williamson 已提交
1579
		unmapped_page = domain->ops->unmap(domain, iova, pgsize);
1580 1581 1582
		if (!unmapped_page)
			break;

1583 1584
		pr_debug("unmapped: iova 0x%lx size 0x%zx\n",
			 iova, unmapped_page);
1585 1586 1587 1588 1589

		iova += unmapped_page;
		unmapped += unmapped_page;
	}

1590
	trace_unmap(orig_iova, size, unmapped);
1591
	return unmapped;
1592 1593
}
EXPORT_SYMBOL_GPL(iommu_unmap);
1594

O
Olav Haugan 已提交
1595 1596 1597
size_t default_iommu_map_sg(struct iommu_domain *domain, unsigned long iova,
			 struct scatterlist *sg, unsigned int nents, int prot)
{
1598
	struct scatterlist *s;
O
Olav Haugan 已提交
1599
	size_t mapped = 0;
1600
	unsigned int i, min_pagesz;
1601
	int ret;
O
Olav Haugan 已提交
1602

1603
	if (unlikely(domain->pgsize_bitmap == 0UL))
1604
		return 0;
O
Olav Haugan 已提交
1605

1606
	min_pagesz = 1 << __ffs(domain->pgsize_bitmap);
1607 1608

	for_each_sg(sg, s, nents, i) {
D
Dan Williams 已提交
1609
		phys_addr_t phys = page_to_phys(sg_page(s)) + s->offset;
1610 1611 1612 1613 1614 1615 1616 1617

		/*
		 * We are mapping on IOMMU page boundaries, so offset within
		 * the page must be 0. However, the IOMMU may support pages
		 * smaller than PAGE_SIZE, so s->offset may still represent
		 * an offset of that boundary within the CPU page.
		 */
		if (!IS_ALIGNED(s->offset, min_pagesz))
1618 1619 1620 1621 1622 1623 1624
			goto out_err;

		ret = iommu_map(domain, iova + mapped, phys, s->length, prot);
		if (ret)
			goto out_err;

		mapped += s->length;
O
Olav Haugan 已提交
1625 1626 1627
	}

	return mapped;
1628 1629 1630 1631 1632 1633 1634

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

	return 0;

O
Olav Haugan 已提交
1635 1636
}
EXPORT_SYMBOL_GPL(default_iommu_map_sg);
1637 1638

int iommu_domain_window_enable(struct iommu_domain *domain, u32 wnd_nr,
1639
			       phys_addr_t paddr, u64 size, int prot)
1640 1641 1642 1643
{
	if (unlikely(domain->ops->domain_window_enable == NULL))
		return -ENODEV;

1644 1645
	return domain->ops->domain_window_enable(domain, wnd_nr, paddr, size,
						 prot);
1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657
}
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);

A
Alex Williamson 已提交
1658
static int __init iommu_init(void)
1659
{
A
Alex Williamson 已提交
1660 1661 1662 1663 1664
	iommu_group_kset = kset_create_and_add("iommu_groups",
					       NULL, kernel_kobj);
	BUG_ON(!iommu_group_kset);

	return 0;
1665
}
1666
core_initcall(iommu_init);
1667 1668 1669 1670

int iommu_domain_get_attr(struct iommu_domain *domain,
			  enum iommu_attr attr, void *data)
{
1671
	struct iommu_domain_geometry *geometry;
1672
	bool *paging;
1673
	int ret = 0;
1674
	u32 *count;
1675 1676 1677 1678 1679 1680

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

1681 1682 1683
		break;
	case DOMAIN_ATTR_PAGING:
		paging  = data;
1684
		*paging = (domain->pgsize_bitmap != 0UL);
1685 1686 1687 1688 1689 1690 1691 1692 1693
		break;
	case DOMAIN_ATTR_WINDOWS:
		count = data;

		if (domain->ops->domain_get_windows != NULL)
			*count = domain->ops->domain_get_windows(domain);
		else
			ret = -ENODEV;

1694 1695 1696 1697
		break;
	default:
		if (!domain->ops->domain_get_attr)
			return -EINVAL;
1698

1699 1700 1701 1702
		ret = domain->ops->domain_get_attr(domain, attr, data);
	}

	return ret;
1703 1704 1705 1706 1707 1708
}
EXPORT_SYMBOL_GPL(iommu_domain_get_attr);

int iommu_domain_set_attr(struct iommu_domain *domain,
			  enum iommu_attr attr, void *data)
{
1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719
	int ret = 0;
	u32 *count;

	switch (attr) {
	case DOMAIN_ATTR_WINDOWS:
		count = data;

		if (domain->ops->domain_set_windows != NULL)
			ret = domain->ops->domain_set_windows(domain, *count);
		else
			ret = -ENODEV;
1720

1721 1722 1723 1724 1725 1726 1727 1728 1729
		break;
	default:
		if (domain->ops->domain_set_attr == NULL)
			return -EINVAL;

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

	return ret;
1730
}
1731
EXPORT_SYMBOL_GPL(iommu_domain_set_attr);
1732

1733
void iommu_get_resv_regions(struct device *dev, struct list_head *list)
1734 1735 1736
{
	const struct iommu_ops *ops = dev->bus->iommu_ops;

1737 1738
	if (ops && ops->get_resv_regions)
		ops->get_resv_regions(dev, list);
1739 1740
}

1741
void iommu_put_resv_regions(struct device *dev, struct list_head *list)
1742 1743 1744
{
	const struct iommu_ops *ops = dev->bus->iommu_ops;

1745 1746
	if (ops && ops->put_resv_regions)
		ops->put_resv_regions(dev, list);
1747
}
1748

E
Eric Auger 已提交
1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764
struct iommu_resv_region *iommu_alloc_resv_region(phys_addr_t start,
						  size_t length,
						  int prot, int type)
{
	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;
1765
}
1766 1767 1768 1769 1770 1771 1772 1773 1774 1775

/* Request that a device is direct mapped by the IOMMU */
int iommu_request_dm_for_dev(struct device *dev)
{
	struct iommu_domain *dm_domain;
	struct iommu_group *group;
	int ret;

	/* Device must already be in a group before calling this function */
	group = iommu_group_get_for_dev(dev);
1776 1777
	if (IS_ERR(group))
		return PTR_ERR(group);
1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818

	mutex_lock(&group->mutex);

	/* Check if the default domain is already direct mapped */
	ret = 0;
	if (group->default_domain &&
	    group->default_domain->type == IOMMU_DOMAIN_IDENTITY)
		goto out;

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

	/* Allocate a direct mapped domain */
	ret = -ENOMEM;
	dm_domain = __iommu_domain_alloc(dev->bus, IOMMU_DOMAIN_IDENTITY);
	if (!dm_domain)
		goto out;

	/* Attach the device to the domain */
	ret = __iommu_attach_group(dm_domain, group);
	if (ret) {
		iommu_domain_free(dm_domain);
		goto out;
	}

	/* Make the direct mapped domain the default for this group */
	if (group->default_domain)
		iommu_domain_free(group->default_domain);
	group->default_domain = dm_domain;

	pr_info("Using direct mapping for device %s\n", dev_name(dev));

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

	return ret;
}
R
Robin Murphy 已提交
1819

1820
const struct iommu_ops *iommu_ops_from_fwnode(struct fwnode_handle *fwnode)
1821 1822
{
	const struct iommu_ops *ops = NULL;
1823
	struct iommu_device *iommu;
1824

1825 1826 1827 1828
	spin_lock(&iommu_device_lock);
	list_for_each_entry(iommu, &iommu_device_list, list)
		if (iommu->fwnode == fwnode) {
			ops = iommu->ops;
1829 1830
			break;
		}
1831
	spin_unlock(&iommu_device_lock);
1832 1833 1834
	return ops;
}

R
Robin Murphy 已提交
1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880
int iommu_fwspec_init(struct device *dev, struct fwnode_handle *iommu_fwnode,
		      const struct iommu_ops *ops)
{
	struct iommu_fwspec *fwspec = dev->iommu_fwspec;

	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;
	dev->iommu_fwspec = fwspec;
	return 0;
}
EXPORT_SYMBOL_GPL(iommu_fwspec_init);

void iommu_fwspec_free(struct device *dev)
{
	struct iommu_fwspec *fwspec = dev->iommu_fwspec;

	if (fwspec) {
		fwnode_handle_put(fwspec->iommu_fwnode);
		kfree(fwspec);
		dev->iommu_fwspec = NULL;
	}
}
EXPORT_SYMBOL_GPL(iommu_fwspec_free);

int iommu_fwspec_add_ids(struct device *dev, u32 *ids, int num_ids)
{
	struct iommu_fwspec *fwspec = dev->iommu_fwspec;
	size_t size;
	int i;

	if (!fwspec)
		return -EINVAL;

	size = offsetof(struct iommu_fwspec, ids[fwspec->num_ids + num_ids]);
	if (size > sizeof(*fwspec)) {
		fwspec = krealloc(dev->iommu_fwspec, size, GFP_KERNEL);
		if (!fwspec)
			return -ENOMEM;
1881 1882

		dev->iommu_fwspec = fwspec;
R
Robin Murphy 已提交
1883 1884 1885 1886 1887 1888 1889 1890 1891
	}

	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);