dm-table.c 37.6 KB
Newer Older
L
Linus Torvalds 已提交
1 2
/*
 * Copyright (C) 2001 Sistina Software (UK) Limited.
3
 * Copyright (C) 2004-2008 Red Hat, Inc. All rights reserved.
L
Linus Torvalds 已提交
4 5 6 7 8 9 10 11 12 13 14
 *
 * This file is released under the GPL.
 */

#include "dm.h"

#include <linux/module.h>
#include <linux/vmalloc.h>
#include <linux/blkdev.h>
#include <linux/namei.h>
#include <linux/ctype.h>
15
#include <linux/string.h>
L
Linus Torvalds 已提交
16 17
#include <linux/slab.h>
#include <linux/interrupt.h>
A
Arjan van de Ven 已提交
18
#include <linux/mutex.h>
19
#include <linux/delay.h>
A
Arun Sharma 已提交
20
#include <linux/atomic.h>
L
Linus Torvalds 已提交
21

22 23
#define DM_MSG_PREFIX "table"

L
Linus Torvalds 已提交
24 25 26 27 28
#define MAX_DEPTH 16
#define NODE_SIZE L1_CACHE_BYTES
#define KEYS_PER_NODE (NODE_SIZE / sizeof(sector_t))
#define CHILDREN_PER_NODE (KEYS_PER_NODE + 1)

29 30 31 32 33 34 35 36 37 38 39 40 41
/*
 * The table has always exactly one reference from either mapped_device->map
 * or hash_cell->new_map. This reference is not counted in table->holders.
 * A pair of dm_create_table/dm_destroy_table functions is used for table
 * creation/destruction.
 *
 * Temporary references from the other code increase table->holders. A pair
 * of dm_table_get/dm_table_put functions is used to manipulate it.
 *
 * When the table is about to be destroyed, we wait for table->holders to
 * drop to zero.
 */

L
Linus Torvalds 已提交
42
struct dm_table {
M
Mike Anderson 已提交
43
	struct mapped_device *md;
L
Linus Torvalds 已提交
44
	atomic_t holders;
K
Kiyoshi Ueda 已提交
45
	unsigned type;
L
Linus Torvalds 已提交
46 47 48 49 50 51 52 53 54 55 56

	/* btree table */
	unsigned int depth;
	unsigned int counts[MAX_DEPTH];	/* in nodes */
	sector_t *index[MAX_DEPTH];

	unsigned int num_targets;
	unsigned int num_allocated;
	sector_t *highs;
	struct dm_target *targets;

57
	struct target_type *immutable_target_type;
58
	unsigned integrity_supported:1;
59
	unsigned singleton:1;
M
Mike Snitzer 已提交
60

L
Linus Torvalds 已提交
61 62 63 64 65
	/*
	 * Indicates the rw permissions for the new logical
	 * device.  This should be a combination of FMODE_READ
	 * and FMODE_WRITE.
	 */
66
	fmode_t mode;
L
Linus Torvalds 已提交
67 68 69 70 71 72 73

	/* a list of devices used by this table */
	struct list_head devices;

	/* events get handed up using this callback */
	void (*event_fn)(void *);
	void *event_context;
K
Kiyoshi Ueda 已提交
74 75

	struct dm_md_mempools *mempools;
76 77

	struct list_head target_callbacks;
L
Linus Torvalds 已提交
78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157
};

/*
 * Similar to ceiling(log_size(n))
 */
static unsigned int int_log(unsigned int n, unsigned int base)
{
	int result = 0;

	while (n > 1) {
		n = dm_div_up(n, base);
		result++;
	}

	return result;
}

/*
 * Calculate the index of the child node of the n'th node k'th key.
 */
static inline unsigned int get_child(unsigned int n, unsigned int k)
{
	return (n * CHILDREN_PER_NODE) + k;
}

/*
 * Return the n'th node of level l from table t.
 */
static inline sector_t *get_node(struct dm_table *t,
				 unsigned int l, unsigned int n)
{
	return t->index[l] + (n * KEYS_PER_NODE);
}

/*
 * Return the highest key that you could lookup from the n'th
 * node on level l of the btree.
 */
static sector_t high(struct dm_table *t, unsigned int l, unsigned int n)
{
	for (; l < t->depth - 1; l++)
		n = get_child(n, CHILDREN_PER_NODE - 1);

	if (n >= t->counts[l])
		return (sector_t) - 1;

	return get_node(t, l, n)[KEYS_PER_NODE - 1];
}

/*
 * Fills in a level of the btree based on the highs of the level
 * below it.
 */
static int setup_btree_index(unsigned int l, struct dm_table *t)
{
	unsigned int n, k;
	sector_t *node;

	for (n = 0U; n < t->counts[l]; n++) {
		node = get_node(t, l, n);

		for (k = 0U; k < KEYS_PER_NODE; k++)
			node[k] = high(t, l + 1, get_child(n, k));
	}

	return 0;
}

void *dm_vcalloc(unsigned long nmemb, unsigned long elem_size)
{
	unsigned long size;
	void *addr;

	/*
	 * Check that we're not going to overflow.
	 */
	if (nmemb > (ULONG_MAX / elem_size))
		return NULL;

	size = nmemb * elem_size;
J
Joe Perches 已提交
158
	addr = vzalloc(size);
L
Linus Torvalds 已提交
159 160 161

	return addr;
}
162
EXPORT_SYMBOL(dm_vcalloc);
L
Linus Torvalds 已提交
163 164 165 166 167 168 169 170 171 172 173 174 175

/*
 * highs, and targets are managed as dynamic arrays during a
 * table load.
 */
static int alloc_targets(struct dm_table *t, unsigned int num)
{
	sector_t *n_highs;
	struct dm_target *n_targets;
	int n = t->num_targets;

	/*
	 * Allocate both the target array and offset array at once.
176 177
	 * Append an empty entry to catch sectors beyond the end of
	 * the device.
L
Linus Torvalds 已提交
178
	 */
179
	n_highs = (sector_t *) dm_vcalloc(num + 1, sizeof(struct dm_target) +
L
Linus Torvalds 已提交
180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
					  sizeof(sector_t));
	if (!n_highs)
		return -ENOMEM;

	n_targets = (struct dm_target *) (n_highs + num);

	if (n) {
		memcpy(n_highs, t->highs, sizeof(*n_highs) * n);
		memcpy(n_targets, t->targets, sizeof(*n_targets) * n);
	}

	memset(n_highs + n, -1, sizeof(*n_highs) * (num - n));
	vfree(t->highs);

	t->num_allocated = num;
	t->highs = n_highs;
	t->targets = n_targets;

	return 0;
}

201
int dm_table_create(struct dm_table **result, fmode_t mode,
M
Mike Anderson 已提交
202
		    unsigned num_targets, struct mapped_device *md)
L
Linus Torvalds 已提交
203
{
D
Dmitry Monakhov 已提交
204
	struct dm_table *t = kzalloc(sizeof(*t), GFP_KERNEL);
L
Linus Torvalds 已提交
205 206 207 208 209

	if (!t)
		return -ENOMEM;

	INIT_LIST_HEAD(&t->devices);
210
	INIT_LIST_HEAD(&t->target_callbacks);
211
	atomic_set(&t->holders, 0);
L
Linus Torvalds 已提交
212 213 214 215 216 217 218 219 220 221 222 223 224

	if (!num_targets)
		num_targets = KEYS_PER_NODE;

	num_targets = dm_round_up(num_targets, KEYS_PER_NODE);

	if (alloc_targets(t, num_targets)) {
		kfree(t);
		t = NULL;
		return -ENOMEM;
	}

	t->mode = mode;
M
Mike Anderson 已提交
225
	t->md = md;
L
Linus Torvalds 已提交
226 227 228 229 230 231 232 233
	*result = t;
	return 0;
}

static void free_devices(struct list_head *devices)
{
	struct list_head *tmp, *next;

P
Paul Jimenez 已提交
234
	list_for_each_safe(tmp, next, devices) {
M
Mikulas Patocka 已提交
235 236
		struct dm_dev_internal *dd =
		    list_entry(tmp, struct dm_dev_internal, list);
237 238
		DMWARN("dm_table_destroy: dm_put_device call missing for %s",
		       dd->dm_dev.name);
L
Linus Torvalds 已提交
239 240 241 242
		kfree(dd);
	}
}

243
void dm_table_destroy(struct dm_table *t)
L
Linus Torvalds 已提交
244 245 246
{
	unsigned int i;

247 248 249
	if (!t)
		return;

250 251 252 253
	while (atomic_read(&t->holders))
		msleep(1);
	smp_mb();

254
	/* free the indexes */
L
Linus Torvalds 已提交
255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270
	if (t->depth >= 2)
		vfree(t->index[t->depth - 2]);

	/* free the targets */
	for (i = 0; i < t->num_targets; i++) {
		struct dm_target *tgt = t->targets + i;

		if (tgt->type->dtr)
			tgt->type->dtr(tgt);

		dm_put_target_type(tgt->type);
	}

	vfree(t->highs);

	/* free the device list */
271
	free_devices(&t->devices);
L
Linus Torvalds 已提交
272

K
Kiyoshi Ueda 已提交
273 274
	dm_free_md_mempools(t->mempools);

L
Linus Torvalds 已提交
275 276 277 278 279 280 281
	kfree(t);
}

void dm_table_get(struct dm_table *t)
{
	atomic_inc(&t->holders);
}
282
EXPORT_SYMBOL(dm_table_get);
L
Linus Torvalds 已提交
283 284 285 286 287 288

void dm_table_put(struct dm_table *t)
{
	if (!t)
		return;

289 290
	smp_mb__before_atomic_dec();
	atomic_dec(&t->holders);
L
Linus Torvalds 已提交
291
}
292
EXPORT_SYMBOL(dm_table_put);
L
Linus Torvalds 已提交
293 294 295 296 297 298 299 300 301 302 303 304 305 306 307

/*
 * Checks to see if we need to extend highs or targets.
 */
static inline int check_space(struct dm_table *t)
{
	if (t->num_targets >= t->num_allocated)
		return alloc_targets(t, t->num_allocated * 2);

	return 0;
}

/*
 * See if we've already got a device in the list.
 */
M
Mikulas Patocka 已提交
308
static struct dm_dev_internal *find_device(struct list_head *l, dev_t dev)
L
Linus Torvalds 已提交
309
{
M
Mikulas Patocka 已提交
310
	struct dm_dev_internal *dd;
L
Linus Torvalds 已提交
311 312

	list_for_each_entry (dd, l, list)
M
Mikulas Patocka 已提交
313
		if (dd->dm_dev.bdev->bd_dev == dev)
L
Linus Torvalds 已提交
314 315 316 317 318 319 320 321
			return dd;

	return NULL;
}

/*
 * Open a device so we can use it as a map destination.
 */
M
Mikulas Patocka 已提交
322 323
static int open_dev(struct dm_dev_internal *d, dev_t dev,
		    struct mapped_device *md)
L
Linus Torvalds 已提交
324 325 326 327 328 329
{
	static char *_claim_ptr = "I belong to device-mapper";
	struct block_device *bdev;

	int r;

M
Mikulas Patocka 已提交
330
	BUG_ON(d->dm_dev.bdev);
L
Linus Torvalds 已提交
331

332
	bdev = blkdev_get_by_dev(dev, d->dm_dev.mode | FMODE_EXCL, _claim_ptr);
L
Linus Torvalds 已提交
333 334
	if (IS_ERR(bdev))
		return PTR_ERR(bdev);
335 336 337

	r = bd_link_disk_holder(bdev, dm_disk(md));
	if (r) {
338
		blkdev_put(bdev, d->dm_dev.mode | FMODE_EXCL);
339 340 341 342 343
		return r;
	}

	d->dm_dev.bdev = bdev;
	return 0;
L
Linus Torvalds 已提交
344 345 346 347 348
}

/*
 * Close a device that we've been using.
 */
M
Mikulas Patocka 已提交
349
static void close_dev(struct dm_dev_internal *d, struct mapped_device *md)
L
Linus Torvalds 已提交
350
{
M
Mikulas Patocka 已提交
351
	if (!d->dm_dev.bdev)
L
Linus Torvalds 已提交
352 353
		return;

354
	bd_unlink_disk_holder(d->dm_dev.bdev, dm_disk(md));
355
	blkdev_put(d->dm_dev.bdev, d->dm_dev.mode | FMODE_EXCL);
M
Mikulas Patocka 已提交
356
	d->dm_dev.bdev = NULL;
L
Linus Torvalds 已提交
357 358 359
}

/*
360
 * If possible, this checks an area of a destination device is invalid.
L
Linus Torvalds 已提交
361
 */
362 363
static int device_area_is_invalid(struct dm_target *ti, struct dm_dev *dev,
				  sector_t start, sector_t len, void *data)
L
Linus Torvalds 已提交
364
{
365
	struct request_queue *q;
366 367 368 369
	struct queue_limits *limits = data;
	struct block_device *bdev = dev->bdev;
	sector_t dev_size =
		i_size_read(bdev->bd_inode) >> SECTOR_SHIFT;
370
	unsigned short logical_block_size_sectors =
371
		limits->logical_block_size >> SECTOR_SHIFT;
372
	char b[BDEVNAME_SIZE];
M
Mike Anderson 已提交
373

374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389
	/*
	 * Some devices exist without request functions,
	 * such as loop devices not yet bound to backing files.
	 * Forbid the use of such devices.
	 */
	q = bdev_get_queue(bdev);
	if (!q || !q->make_request_fn) {
		DMWARN("%s: %s is not yet initialised: "
		       "start=%llu, len=%llu, dev_size=%llu",
		       dm_device_name(ti->table->md), bdevname(bdev, b),
		       (unsigned long long)start,
		       (unsigned long long)len,
		       (unsigned long long)dev_size);
		return 1;
	}

M
Mike Anderson 已提交
390
	if (!dev_size)
391
		return 0;
M
Mike Anderson 已提交
392

393
	if ((start >= dev_size) || (start + len > dev_size)) {
394 395 396 397 398 399
		DMWARN("%s: %s too small for target: "
		       "start=%llu, len=%llu, dev_size=%llu",
		       dm_device_name(ti->table->md), bdevname(bdev, b),
		       (unsigned long long)start,
		       (unsigned long long)len,
		       (unsigned long long)dev_size);
400
		return 1;
401 402 403
	}

	if (logical_block_size_sectors <= 1)
404
		return 0;
405 406 407

	if (start & (logical_block_size_sectors - 1)) {
		DMWARN("%s: start=%llu not aligned to h/w "
408
		       "logical block size %u of %s",
409 410
		       dm_device_name(ti->table->md),
		       (unsigned long long)start,
411
		       limits->logical_block_size, bdevname(bdev, b));
412
		return 1;
413 414
	}

415
	if (len & (logical_block_size_sectors - 1)) {
416
		DMWARN("%s: len=%llu not aligned to h/w "
417
		       "logical block size %u of %s",
418
		       dm_device_name(ti->table->md),
419
		       (unsigned long long)len,
420
		       limits->logical_block_size, bdevname(bdev, b));
421
		return 1;
422 423
	}

424
	return 0;
L
Linus Torvalds 已提交
425 426 427
}

/*
428
 * This upgrades the mode on an already open dm_dev, being
L
Linus Torvalds 已提交
429
 * careful to leave things as they were if we fail to reopen the
430 431
 * device and not to touch the existing bdev field in case
 * it is accessed concurrently inside dm_table_any_congested().
L
Linus Torvalds 已提交
432
 */
433
static int upgrade_mode(struct dm_dev_internal *dd, fmode_t new_mode,
M
Mikulas Patocka 已提交
434
			struct mapped_device *md)
L
Linus Torvalds 已提交
435 436
{
	int r;
437
	struct dm_dev_internal dd_new, dd_old;
L
Linus Torvalds 已提交
438

439 440 441 442 443 444 445 446
	dd_new = dd_old = *dd;

	dd_new.dm_dev.mode |= new_mode;
	dd_new.dm_dev.bdev = NULL;

	r = open_dev(&dd_new, dd->dm_dev.bdev->bd_dev, md);
	if (r)
		return r;
L
Linus Torvalds 已提交
447

M
Mikulas Patocka 已提交
448
	dd->dm_dev.mode |= new_mode;
449
	close_dev(&dd_old, md);
L
Linus Torvalds 已提交
450

451
	return 0;
L
Linus Torvalds 已提交
452 453 454 455 456 457
}

/*
 * Add a device to the list, or just increment the usage count if
 * it's already present.
 */
458 459
int dm_get_device(struct dm_target *ti, const char *path, fmode_t mode,
		  struct dm_dev **result)
L
Linus Torvalds 已提交
460 461
{
	int r;
A
Andrew Morton 已提交
462
	dev_t uninitialized_var(dev);
M
Mikulas Patocka 已提交
463
	struct dm_dev_internal *dd;
L
Linus Torvalds 已提交
464
	unsigned int major, minor;
465
	struct dm_table *t = ti->table;
466
	char dummy;
L
Linus Torvalds 已提交
467

468
	BUG_ON(!t);
L
Linus Torvalds 已提交
469

470
	if (sscanf(path, "%u:%u%c", &major, &minor, &dummy) == 2) {
L
Linus Torvalds 已提交
471 472 473 474 475 476
		/* Extract the major/minor numbers */
		dev = MKDEV(major, minor);
		if (MAJOR(dev) != major || MINOR(dev) != minor)
			return -EOVERFLOW;
	} else {
		/* convert the path to a device */
477 478 479 480 481 482
		struct block_device *bdev = lookup_bdev(path);

		if (IS_ERR(bdev))
			return PTR_ERR(bdev);
		dev = bdev->bd_dev;
		bdput(bdev);
L
Linus Torvalds 已提交
483 484 485 486 487 488 489 490
	}

	dd = find_device(&t->devices, dev);
	if (!dd) {
		dd = kmalloc(sizeof(*dd), GFP_KERNEL);
		if (!dd)
			return -ENOMEM;

M
Mikulas Patocka 已提交
491 492
		dd->dm_dev.mode = mode;
		dd->dm_dev.bdev = NULL;
L
Linus Torvalds 已提交
493

494
		if ((r = open_dev(dd, dev, t->md))) {
L
Linus Torvalds 已提交
495 496 497 498
			kfree(dd);
			return r;
		}

M
Mikulas Patocka 已提交
499
		format_dev_t(dd->dm_dev.name, dev);
L
Linus Torvalds 已提交
500 501 502 503

		atomic_set(&dd->count, 0);
		list_add(&dd->list, &t->devices);

M
Mikulas Patocka 已提交
504
	} else if (dd->dm_dev.mode != (mode | dd->dm_dev.mode)) {
505
		r = upgrade_mode(dd, mode, t->md);
L
Linus Torvalds 已提交
506 507 508 509 510
		if (r)
			return r;
	}
	atomic_inc(&dd->count);

M
Mikulas Patocka 已提交
511
	*result = &dd->dm_dev;
L
Linus Torvalds 已提交
512 513
	return 0;
}
514
EXPORT_SYMBOL(dm_get_device);
L
Linus Torvalds 已提交
515

516
int dm_set_device_limits(struct dm_target *ti, struct dm_dev *dev,
517
			 sector_t start, sector_t len, void *data)
L
Linus Torvalds 已提交
518
{
519 520
	struct queue_limits *limits = data;
	struct block_device *bdev = dev->bdev;
521
	struct request_queue *q = bdev_get_queue(bdev);
A
Alasdair G Kergon 已提交
522 523 524 525 526
	char b[BDEVNAME_SIZE];

	if (unlikely(!q)) {
		DMWARN("%s: Cannot set limits for nonexistent device %s",
		       dm_device_name(ti->table->md), bdevname(bdev, b));
527
		return 0;
A
Alasdair G Kergon 已提交
528
	}
529

530 531
	if (bdev_stack_limits(limits, bdev, start) < 0)
		DMWARN("%s: adding target device %s caused an alignment inconsistency: "
532 533 534 535 536 537
		       "physical_block_size=%u, logical_block_size=%u, "
		       "alignment_offset=%u, start=%llu",
		       dm_device_name(ti->table->md), bdevname(bdev, b),
		       q->limits.physical_block_size,
		       q->limits.logical_block_size,
		       q->limits.alignment_offset,
538
		       (unsigned long long) start << SECTOR_SHIFT);
539

540 541 542 543
	/*
	 * Check if merge fn is supported.
	 * If not we'll force DM to use PAGE_SIZE or
	 * smaller I/O, just to be safe.
544
	 */
545
	if (dm_queue_merge_is_compulsory(q) && !ti->type->merge)
546 547
		blk_limits_max_hw_sectors(limits,
					  (unsigned int) (PAGE_SIZE >> 9));
548
	return 0;
549 550
}
EXPORT_SYMBOL_GPL(dm_set_device_limits);
551

L
Linus Torvalds 已提交
552
/*
553
 * Decrement a device's use count and remove it if necessary.
L
Linus Torvalds 已提交
554
 */
M
Mikulas Patocka 已提交
555
void dm_put_device(struct dm_target *ti, struct dm_dev *d)
L
Linus Torvalds 已提交
556
{
M
Mikulas Patocka 已提交
557 558 559
	struct dm_dev_internal *dd = container_of(d, struct dm_dev_internal,
						  dm_dev);

L
Linus Torvalds 已提交
560
	if (atomic_dec_and_test(&dd->count)) {
561
		close_dev(dd, ti->table->md);
L
Linus Torvalds 已提交
562 563 564 565
		list_del(&dd->list);
		kfree(dd);
	}
}
566
EXPORT_SYMBOL(dm_put_device);
L
Linus Torvalds 已提交
567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609

/*
 * Checks to see if the target joins onto the end of the table.
 */
static int adjoin(struct dm_table *table, struct dm_target *ti)
{
	struct dm_target *prev;

	if (!table->num_targets)
		return !ti->begin;

	prev = &table->targets[table->num_targets - 1];
	return (ti->begin == (prev->begin + prev->len));
}

/*
 * Used to dynamically allocate the arg array.
 */
static char **realloc_argv(unsigned *array_size, char **old_argv)
{
	char **argv;
	unsigned new_size;

	new_size = *array_size ? *array_size * 2 : 64;
	argv = kmalloc(new_size * sizeof(*argv), GFP_KERNEL);
	if (argv) {
		memcpy(argv, old_argv, *array_size * sizeof(*argv));
		*array_size = new_size;
	}

	kfree(old_argv);
	return argv;
}

/*
 * Destructively splits up the argument list to pass to ctr.
 */
int dm_split_args(int *argc, char ***argvp, char *input)
{
	char *start, *end = input, *out, **argv = NULL;
	unsigned array_size = 0;

	*argc = 0;
610 611 612 613 614 615

	if (!input) {
		*argvp = NULL;
		return 0;
	}

L
Linus Torvalds 已提交
616 617 618 619 620 621
	argv = realloc_argv(&array_size, argv);
	if (!argv)
		return -ENOMEM;

	while (1) {
		/* Skip whitespace */
622
		start = skip_spaces(end);
L
Linus Torvalds 已提交
623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663

		if (!*start)
			break;	/* success, we hit the end */

		/* 'out' is used to remove any back-quotes */
		end = out = start;
		while (*end) {
			/* Everything apart from '\0' can be quoted */
			if (*end == '\\' && *(end + 1)) {
				*out++ = *(end + 1);
				end += 2;
				continue;
			}

			if (isspace(*end))
				break;	/* end of token */

			*out++ = *end++;
		}

		/* have we already filled the array ? */
		if ((*argc + 1) > array_size) {
			argv = realloc_argv(&array_size, argv);
			if (!argv)
				return -ENOMEM;
		}

		/* we know this is whitespace */
		if (*end)
			end++;

		/* terminate the string and put it in the array */
		*out = '\0';
		argv[*argc] = start;
		(*argc)++;
	}

	*argvp = argv;
	return 0;
}

664 665 666 667 668 669 670
/*
 * Impose necessary and sufficient conditions on a devices's table such
 * that any incoming bio which respects its logical_block_size can be
 * processed successfully.  If it falls across the boundary between
 * two or more targets, the size of each piece it gets split into must
 * be compatible with the logical_block_size of the target processing it.
 */
671 672
static int validate_hardware_logical_block_alignment(struct dm_table *table,
						 struct queue_limits *limits)
673 674 675 676 677 678
{
	/*
	 * This function uses arithmetic modulo the logical_block_size
	 * (in units of 512-byte sectors).
	 */
	unsigned short device_logical_block_size_sects =
679
		limits->logical_block_size >> SECTOR_SHIFT;
680 681 682 683 684 685 686 687 688 689 690 691 692

	/*
	 * Offset of the start of the next table entry, mod logical_block_size.
	 */
	unsigned short next_target_start = 0;

	/*
	 * Given an aligned bio that extends beyond the end of a
	 * target, how many sectors must the next target handle?
	 */
	unsigned short remaining = 0;

	struct dm_target *uninitialized_var(ti);
693
	struct queue_limits ti_limits;
694 695 696 697 698 699 700 701
	unsigned i = 0;

	/*
	 * Check each entry in the table in turn.
	 */
	while (i < dm_table_get_num_targets(table)) {
		ti = dm_table_get_target(table, i++);

702
		blk_set_stacking_limits(&ti_limits);
703 704 705 706 707 708

		/* combine all target devices' limits */
		if (ti->type->iterate_devices)
			ti->type->iterate_devices(ti, dm_set_device_limits,
						  &ti_limits);

709 710 711 712 713
		/*
		 * If the remaining sectors fall entirely within this
		 * table entry are they compatible with its logical_block_size?
		 */
		if (remaining < ti->len &&
714
		    remaining & ((ti_limits.logical_block_size >>
715 716 717 718 719 720 721 722 723 724 725 726
				  SECTOR_SHIFT) - 1))
			break;	/* Error */

		next_target_start =
		    (unsigned short) ((next_target_start + ti->len) &
				      (device_logical_block_size_sects - 1));
		remaining = next_target_start ?
		    device_logical_block_size_sects - next_target_start : 0;
	}

	if (remaining) {
		DMWARN("%s: table line %u (start sect %llu len %llu) "
727
		       "not aligned to h/w logical block size %u",
728 729 730
		       dm_device_name(table->md), i,
		       (unsigned long long) ti->begin,
		       (unsigned long long) ti->len,
731
		       limits->logical_block_size);
732 733 734 735 736 737
		return -EINVAL;
	}

	return 0;
}

L
Linus Torvalds 已提交
738 739 740 741 742 743 744
int dm_table_add_target(struct dm_table *t, const char *type,
			sector_t start, sector_t len, char *params)
{
	int r = -EINVAL, argc;
	char **argv;
	struct dm_target *tgt;

745 746 747 748 749 750
	if (t->singleton) {
		DMERR("%s: target type %s must appear alone in table",
		      dm_device_name(t->md), t->targets->type->name);
		return -EINVAL;
	}

L
Linus Torvalds 已提交
751 752 753 754 755 756 757
	if ((r = check_space(t)))
		return r;

	tgt = t->targets + t->num_targets;
	memset(tgt, 0, sizeof(*tgt));

	if (!len) {
758
		DMERR("%s: zero-length target", dm_device_name(t->md));
L
Linus Torvalds 已提交
759 760 761 762 763
		return -EINVAL;
	}

	tgt->type = dm_get_target_type(type);
	if (!tgt->type) {
764 765
		DMERR("%s: %s: unknown target type", dm_device_name(t->md),
		      type);
L
Linus Torvalds 已提交
766 767 768
		return -EINVAL;
	}

769 770 771 772 773 774 775 776 777
	if (dm_target_needs_singleton(tgt->type)) {
		if (t->num_targets) {
			DMERR("%s: target type %s must appear alone in table",
			      dm_device_name(t->md), type);
			return -EINVAL;
		}
		t->singleton = 1;
	}

778 779 780 781 782 783
	if (dm_target_always_writeable(tgt->type) && !(t->mode & FMODE_WRITE)) {
		DMERR("%s: target type %s may not be included in read-only tables",
		      dm_device_name(t->md), type);
		return -EINVAL;
	}

784 785 786 787 788 789 790 791 792 793 794 795 796 797 798
	if (t->immutable_target_type) {
		if (t->immutable_target_type != tgt->type) {
			DMERR("%s: immutable target type %s cannot be mixed with other target types",
			      dm_device_name(t->md), t->immutable_target_type->name);
			return -EINVAL;
		}
	} else if (dm_target_is_immutable(tgt->type)) {
		if (t->num_targets) {
			DMERR("%s: immutable target type %s cannot be mixed with other target types",
			      dm_device_name(t->md), tgt->type->name);
			return -EINVAL;
		}
		t->immutable_target_type = tgt->type;
	}

L
Linus Torvalds 已提交
799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825
	tgt->table = t;
	tgt->begin = start;
	tgt->len = len;
	tgt->error = "Unknown error";

	/*
	 * Does this target adjoin the previous one ?
	 */
	if (!adjoin(t, tgt)) {
		tgt->error = "Gap in table";
		r = -EINVAL;
		goto bad;
	}

	r = dm_split_args(&argc, &argv, params);
	if (r) {
		tgt->error = "couldn't split parameters (insufficient memory)";
		goto bad;
	}

	r = tgt->type->ctr(tgt, argc, argv);
	kfree(argv);
	if (r)
		goto bad;

	t->highs[t->num_targets++] = tgt->begin + tgt->len - 1;

M
Mike Snitzer 已提交
826 827 828
	if (!tgt->num_discard_requests && tgt->discards_supported)
		DMWARN("%s: %s: ignoring discards_supported because num_discard_requests is zero.",
		       dm_device_name(t->md), type);
M
Mike Snitzer 已提交
829

L
Linus Torvalds 已提交
830 831 832
	return 0;

 bad:
833
	DMERR("%s: %s: %s", dm_device_name(t->md), type, tgt->error);
L
Linus Torvalds 已提交
834 835 836 837
	dm_put_target_type(tgt->type);
	return r;
}

838 839 840 841 842 843 844
/*
 * Target argument parsing helpers.
 */
static int validate_next_arg(struct dm_arg *arg, struct dm_arg_set *arg_set,
			     unsigned *value, char **error, unsigned grouped)
{
	const char *arg_str = dm_shift_arg(arg_set);
845
	char dummy;
846 847

	if (!arg_str ||
848
	    (sscanf(arg_str, "%u%c", value, &dummy) != 1) ||
849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895
	    (*value < arg->min) ||
	    (*value > arg->max) ||
	    (grouped && arg_set->argc < *value)) {
		*error = arg->error;
		return -EINVAL;
	}

	return 0;
}

int dm_read_arg(struct dm_arg *arg, struct dm_arg_set *arg_set,
		unsigned *value, char **error)
{
	return validate_next_arg(arg, arg_set, value, error, 0);
}
EXPORT_SYMBOL(dm_read_arg);

int dm_read_arg_group(struct dm_arg *arg, struct dm_arg_set *arg_set,
		      unsigned *value, char **error)
{
	return validate_next_arg(arg, arg_set, value, error, 1);
}
EXPORT_SYMBOL(dm_read_arg_group);

const char *dm_shift_arg(struct dm_arg_set *as)
{
	char *r;

	if (as->argc) {
		as->argc--;
		r = *as->argv;
		as->argv++;
		return r;
	}

	return NULL;
}
EXPORT_SYMBOL(dm_shift_arg);

void dm_consume_args(struct dm_arg_set *as, unsigned num_args)
{
	BUG_ON(as->argc < num_args);
	as->argc -= num_args;
	as->argv += num_args;
}
EXPORT_SYMBOL(dm_consume_args);

896
static int dm_table_set_type(struct dm_table *t)
K
Kiyoshi Ueda 已提交
897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956
{
	unsigned i;
	unsigned bio_based = 0, request_based = 0;
	struct dm_target *tgt;
	struct dm_dev_internal *dd;
	struct list_head *devices;

	for (i = 0; i < t->num_targets; i++) {
		tgt = t->targets + i;
		if (dm_target_request_based(tgt))
			request_based = 1;
		else
			bio_based = 1;

		if (bio_based && request_based) {
			DMWARN("Inconsistent table: different target types"
			       " can't be mixed up");
			return -EINVAL;
		}
	}

	if (bio_based) {
		/* We must use this table as bio-based */
		t->type = DM_TYPE_BIO_BASED;
		return 0;
	}

	BUG_ON(!request_based); /* No targets in this table */

	/* Non-request-stackable devices can't be used for request-based dm */
	devices = dm_table_get_devices(t);
	list_for_each_entry(dd, devices, list) {
		if (!blk_queue_stackable(bdev_get_queue(dd->dm_dev.bdev))) {
			DMWARN("table load rejected: including"
			       " non-request-stackable devices");
			return -EINVAL;
		}
	}

	/*
	 * Request-based dm supports only tables that have a single target now.
	 * To support multiple targets, request splitting support is needed,
	 * and that needs lots of changes in the block-layer.
	 * (e.g. request completion process for partial completion.)
	 */
	if (t->num_targets > 1) {
		DMWARN("Request-based dm doesn't support multiple targets yet");
		return -EINVAL;
	}

	t->type = DM_TYPE_REQUEST_BASED;

	return 0;
}

unsigned dm_table_get_type(struct dm_table *t)
{
	return t->type;
}

957 958 959 960 961
struct target_type *dm_table_get_immutable_target_type(struct dm_table *t)
{
	return t->immutable_target_type;
}

K
Kiyoshi Ueda 已提交
962 963 964 965 966 967 968 969 970 971 972 973 974 975
bool dm_table_request_based(struct dm_table *t)
{
	return dm_table_get_type(t) == DM_TYPE_REQUEST_BASED;
}

int dm_table_alloc_md_mempools(struct dm_table *t)
{
	unsigned type = dm_table_get_type(t);

	if (unlikely(type == DM_TYPE_NONE)) {
		DMWARN("no table type is set, can't allocate mempools");
		return -EINVAL;
	}

976
	t->mempools = dm_alloc_md_mempools(type, t->integrity_supported);
K
Kiyoshi Ueda 已提交
977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993
	if (!t->mempools)
		return -ENOMEM;

	return 0;
}

void dm_table_free_md_mempools(struct dm_table *t)
{
	dm_free_md_mempools(t->mempools);
	t->mempools = NULL;
}

struct dm_md_mempools *dm_table_get_md_mempools(struct dm_table *t)
{
	return t->mempools;
}

L
Linus Torvalds 已提交
994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010
static int setup_indexes(struct dm_table *t)
{
	int i;
	unsigned int total = 0;
	sector_t *indexes;

	/* allocate the space for *all* the indexes */
	for (i = t->depth - 2; i >= 0; i--) {
		t->counts[i] = dm_div_up(t->counts[i + 1], CHILDREN_PER_NODE);
		total += t->counts[i];
	}

	indexes = (sector_t *) dm_vcalloc(total, (unsigned long) NODE_SIZE);
	if (!indexes)
		return -ENOMEM;

	/* set up internal nodes, bottom-up */
J
Jun'ichi Nomura 已提交
1011
	for (i = t->depth - 2; i >= 0; i--) {
L
Linus Torvalds 已提交
1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022
		t->index[i] = indexes;
		indexes += (KEYS_PER_NODE * t->counts[i]);
		setup_btree_index(i, t);
	}

	return 0;
}

/*
 * Builds the btree to index the map.
 */
1023
static int dm_table_build_index(struct dm_table *t)
L
Linus Torvalds 已提交
1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041
{
	int r = 0;
	unsigned int leaf_nodes;

	/* how many indexes will the btree have ? */
	leaf_nodes = dm_div_up(t->num_targets, KEYS_PER_NODE);
	t->depth = 1 + int_log(leaf_nodes, CHILDREN_PER_NODE);

	/* leaf layer has already been set up */
	t->counts[t->depth - 1] = leaf_nodes;
	t->index[t->depth - 1] = t->highs;

	if (t->depth >= 2)
		r = setup_indexes(t);

	return r;
}

1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078
/*
 * Get a disk whose integrity profile reflects the table's profile.
 * If %match_all is true, all devices' profiles must match.
 * If %match_all is false, all devices must at least have an
 * allocated integrity profile; but uninitialized is ok.
 * Returns NULL if integrity support was inconsistent or unavailable.
 */
static struct gendisk * dm_table_get_integrity_disk(struct dm_table *t,
						    bool match_all)
{
	struct list_head *devices = dm_table_get_devices(t);
	struct dm_dev_internal *dd = NULL;
	struct gendisk *prev_disk = NULL, *template_disk = NULL;

	list_for_each_entry(dd, devices, list) {
		template_disk = dd->dm_dev.bdev->bd_disk;
		if (!blk_get_integrity(template_disk))
			goto no_integrity;
		if (!match_all && !blk_integrity_is_initialized(template_disk))
			continue; /* skip uninitialized profiles */
		else if (prev_disk &&
			 blk_integrity_compare(prev_disk, template_disk) < 0)
			goto no_integrity;
		prev_disk = template_disk;
	}

	return template_disk;

no_integrity:
	if (prev_disk)
		DMWARN("%s: integrity not set: %s and %s profile mismatch",
		       dm_device_name(t->md),
		       prev_disk->disk_name,
		       template_disk->disk_name);
	return NULL;
}

1079 1080
/*
 * Register the mapped device for blk_integrity support if
1081 1082 1083 1084 1085 1086 1087
 * the underlying devices have an integrity profile.  But all devices
 * may not have matching profiles (checking all devices isn't reliable
 * during table load because this table may use other DM device(s) which
 * must be resumed before they will have an initialized integity profile).
 * Stacked DM devices force a 2 stage integrity profile validation:
 * 1 - during load, validate all initialized integrity profiles match
 * 2 - during resume, validate all integrity profiles match
1088 1089 1090
 */
static int dm_table_prealloc_integrity(struct dm_table *t, struct mapped_device *md)
{
1091
	struct gendisk *template_disk = NULL;
1092

1093 1094 1095
	template_disk = dm_table_get_integrity_disk(t, false);
	if (!template_disk)
		return 0;
1096

1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116
	if (!blk_integrity_is_initialized(dm_disk(md))) {
		t->integrity_supported = 1;
		return blk_integrity_register(dm_disk(md), NULL);
	}

	/*
	 * If DM device already has an initalized integrity
	 * profile the new profile should not conflict.
	 */
	if (blk_integrity_is_initialized(template_disk) &&
	    blk_integrity_compare(dm_disk(md), template_disk) < 0) {
		DMWARN("%s: conflict with existing integrity profile: "
		       "%s profile mismatch",
		       dm_device_name(t->md),
		       template_disk->disk_name);
		return 1;
	}

	/* Preserve existing initialized integrity profile */
	t->integrity_supported = 1;
1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152
	return 0;
}

/*
 * Prepares the table for use by building the indices,
 * setting the type, and allocating mempools.
 */
int dm_table_complete(struct dm_table *t)
{
	int r;

	r = dm_table_set_type(t);
	if (r) {
		DMERR("unable to set table type");
		return r;
	}

	r = dm_table_build_index(t);
	if (r) {
		DMERR("unable to build btrees");
		return r;
	}

	r = dm_table_prealloc_integrity(t, t->md);
	if (r) {
		DMERR("could not register integrity profile.");
		return r;
	}

	r = dm_table_alloc_md_mempools(t);
	if (r)
		DMERR("unable to allocate mempools");

	return r;
}

A
Arjan van de Ven 已提交
1153
static DEFINE_MUTEX(_event_lock);
L
Linus Torvalds 已提交
1154 1155 1156
void dm_table_event_callback(struct dm_table *t,
			     void (*fn)(void *), void *context)
{
A
Arjan van de Ven 已提交
1157
	mutex_lock(&_event_lock);
L
Linus Torvalds 已提交
1158 1159
	t->event_fn = fn;
	t->event_context = context;
A
Arjan van de Ven 已提交
1160
	mutex_unlock(&_event_lock);
L
Linus Torvalds 已提交
1161 1162 1163 1164 1165 1166 1167 1168 1169 1170
}

void dm_table_event(struct dm_table *t)
{
	/*
	 * You can no longer call dm_table_event() from interrupt
	 * context, use a bottom half instead.
	 */
	BUG_ON(in_interrupt());

A
Arjan van de Ven 已提交
1171
	mutex_lock(&_event_lock);
L
Linus Torvalds 已提交
1172 1173
	if (t->event_fn)
		t->event_fn(t->event_context);
A
Arjan van de Ven 已提交
1174
	mutex_unlock(&_event_lock);
L
Linus Torvalds 已提交
1175
}
1176
EXPORT_SYMBOL(dm_table_event);
L
Linus Torvalds 已提交
1177 1178 1179 1180 1181

sector_t dm_table_get_size(struct dm_table *t)
{
	return t->num_targets ? (t->highs[t->num_targets - 1] + 1) : 0;
}
1182
EXPORT_SYMBOL(dm_table_get_size);
L
Linus Torvalds 已提交
1183 1184 1185

struct dm_target *dm_table_get_target(struct dm_table *t, unsigned int index)
{
1186
	if (index >= t->num_targets)
L
Linus Torvalds 已提交
1187 1188 1189 1190 1191 1192 1193
		return NULL;

	return t->targets + index;
}

/*
 * Search the btree for the correct target.
1194 1195 1196
 *
 * Caller should check returned pointer with dm_target_is_valid()
 * to trap I/O beyond end of device.
L
Linus Torvalds 已提交
1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214
 */
struct dm_target *dm_table_find_target(struct dm_table *t, sector_t sector)
{
	unsigned int l, n = 0, k = 0;
	sector_t *node;

	for (l = 0; l < t->depth; l++) {
		n = get_child(n, k);
		node = get_node(t, l, n);

		for (k = 0; k < KEYS_PER_NODE; k++)
			if (node[k] >= sector)
				break;
	}

	return &t->targets[(KEYS_PER_NODE * n) + k];
}

1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249
static int count_device(struct dm_target *ti, struct dm_dev *dev,
			sector_t start, sector_t len, void *data)
{
	unsigned *num_devices = data;

	(*num_devices)++;

	return 0;
}

/*
 * Check whether a table has no data devices attached using each
 * target's iterate_devices method.
 * Returns false if the result is unknown because a target doesn't
 * support iterate_devices.
 */
bool dm_table_has_no_data_devices(struct dm_table *table)
{
	struct dm_target *uninitialized_var(ti);
	unsigned i = 0, num_devices = 0;

	while (i < dm_table_get_num_targets(table)) {
		ti = dm_table_get_target(table, i++);

		if (!ti->type->iterate_devices)
			return false;

		ti->type->iterate_devices(ti, count_device, &num_devices);
		if (num_devices)
			return false;
	}

	return true;
}

1250 1251 1252 1253 1254 1255 1256 1257 1258 1259
/*
 * Establish the new table's queue_limits and validate them.
 */
int dm_calculate_queue_limits(struct dm_table *table,
			      struct queue_limits *limits)
{
	struct dm_target *uninitialized_var(ti);
	struct queue_limits ti_limits;
	unsigned i = 0;

1260
	blk_set_stacking_limits(limits);
1261 1262

	while (i < dm_table_get_num_targets(table)) {
1263
		blk_set_stacking_limits(&ti_limits);
1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275

		ti = dm_table_get_target(table, i++);

		if (!ti->type->iterate_devices)
			goto combine_limits;

		/*
		 * Combine queue limits of all the devices this target uses.
		 */
		ti->type->iterate_devices(ti, dm_set_device_limits,
					  &ti_limits);

1276 1277 1278 1279
		/* Set I/O hints portion of queue limits */
		if (ti->type->io_hints)
			ti->type->io_hints(ti, &ti_limits);

1280 1281 1282 1283
		/*
		 * Check each device area is consistent with the target's
		 * overall queue limits.
		 */
1284 1285
		if (ti->type->iterate_devices(ti, device_area_is_invalid,
					      &ti_limits))
1286 1287 1288 1289 1290 1291 1292 1293
			return -EINVAL;

combine_limits:
		/*
		 * Merge this target's queue limits into the overall limits
		 * for the table.
		 */
		if (blk_stack_limits(limits, &ti_limits, 0) < 0)
1294
			DMWARN("%s: adding target device "
1295
			       "(start sect %llu len %llu) "
1296
			       "caused an alignment inconsistency",
1297 1298 1299 1300 1301 1302 1303 1304
			       dm_device_name(table->md),
			       (unsigned long long) ti->begin,
			       (unsigned long long) ti->len);
	}

	return validate_hardware_logical_block_alignment(table, limits);
}

M
Martin K. Petersen 已提交
1305 1306
/*
 * Set the integrity profile for this device if all devices used have
1307 1308 1309 1310
 * matching profiles.  We're quite deep in the resume path but still
 * don't know if all devices (particularly DM devices this device
 * may be stacked on) have matching profiles.  Even if the profiles
 * don't match we have no way to fail (to resume) at this point.
M
Martin K. Petersen 已提交
1311 1312 1313
 */
static void dm_table_set_integrity(struct dm_table *t)
{
1314
	struct gendisk *template_disk = NULL;
M
Martin K. Petersen 已提交
1315 1316 1317 1318

	if (!blk_get_integrity(dm_disk(t->md)))
		return;

1319
	template_disk = dm_table_get_integrity_disk(t, true);
1320 1321 1322 1323
	if (template_disk)
		blk_integrity_register(dm_disk(t->md),
				       blk_get_integrity(template_disk));
	else if (blk_integrity_is_initialized(dm_disk(t->md)))
1324 1325
		DMWARN("%s: device no longer has a valid integrity profile",
		       dm_device_name(t->md));
1326 1327 1328
	else
		DMWARN("%s: unable to establish an integrity profile",
		       dm_device_name(t->md));
M
Martin K. Petersen 已提交
1329 1330
}

1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356
static int device_flush_capable(struct dm_target *ti, struct dm_dev *dev,
				sector_t start, sector_t len, void *data)
{
	unsigned flush = (*(unsigned *)data);
	struct request_queue *q = bdev_get_queue(dev->bdev);

	return q && (q->flush_flags & flush);
}

static bool dm_table_supports_flush(struct dm_table *t, unsigned flush)
{
	struct dm_target *ti;
	unsigned i = 0;

	/*
	 * Require at least one underlying device to support flushes.
	 * t->devices includes internal dm devices such as mirror logs
	 * so we need to use iterate_devices here, which targets
	 * supporting flushes must provide.
	 */
	while (i < dm_table_get_num_targets(t)) {
		ti = dm_table_get_target(t, i++);

		if (!ti->num_flush_requests)
			continue;

1357 1358 1359
		if (ti->flush_supported)
			return 1;

1360 1361 1362 1363 1364 1365 1366 1367
		if (ti->type->iterate_devices &&
		    ti->type->iterate_devices(ti, device_flush_capable, &flush))
			return 1;
	}

	return 0;
}

1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383
static bool dm_table_discard_zeroes_data(struct dm_table *t)
{
	struct dm_target *ti;
	unsigned i = 0;

	/* Ensure that all targets supports discard_zeroes_data. */
	while (i < dm_table_get_num_targets(t)) {
		ti = dm_table_get_target(t, i++);

		if (ti->discard_zeroes_data_unsupported)
			return 0;
	}

	return 1;
}

1384 1385 1386 1387 1388 1389 1390 1391
static int device_is_nonrot(struct dm_target *ti, struct dm_dev *dev,
			    sector_t start, sector_t len, void *data)
{
	struct request_queue *q = bdev_get_queue(dev->bdev);

	return q && blk_queue_nonrot(q);
}

1392 1393 1394 1395 1396 1397 1398 1399 1400 1401
static int device_is_not_random(struct dm_target *ti, struct dm_dev *dev,
			     sector_t start, sector_t len, void *data)
{
	struct request_queue *q = bdev_get_queue(dev->bdev);

	return q && !blk_queue_add_random(q);
}

static bool dm_table_all_devices_attribute(struct dm_table *t,
					   iterate_devices_callout_fn func)
1402 1403 1404 1405 1406 1407 1408 1409
{
	struct dm_target *ti;
	unsigned i = 0;

	while (i < dm_table_get_num_targets(t)) {
		ti = dm_table_get_target(t, i++);

		if (!ti->type->iterate_devices ||
1410
		    !ti->type->iterate_devices(ti, func, NULL))
1411 1412 1413 1414 1415 1416
			return 0;
	}

	return 1;
}

1417 1418
void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
			       struct queue_limits *limits)
L
Linus Torvalds 已提交
1419
{
1420 1421
	unsigned flush = 0;

L
Linus Torvalds 已提交
1422
	/*
1423
	 * Copy table's limits to the DM device's request_queue
L
Linus Torvalds 已提交
1424
	 */
1425
	q->limits = *limits;
1426

M
Mike Snitzer 已提交
1427 1428 1429 1430 1431
	if (!dm_table_supports_discards(t))
		queue_flag_clear_unlocked(QUEUE_FLAG_DISCARD, q);
	else
		queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, q);

1432 1433 1434 1435 1436 1437 1438
	if (dm_table_supports_flush(t, REQ_FLUSH)) {
		flush |= REQ_FLUSH;
		if (dm_table_supports_flush(t, REQ_FUA))
			flush |= REQ_FUA;
	}
	blk_queue_flush(q, flush);

1439 1440 1441
	if (!dm_table_discard_zeroes_data(t))
		q->limits.discard_zeroes_data = 0;

1442 1443
	/* Ensure that all underlying devices are non-rotational. */
	if (dm_table_all_devices_attribute(t, device_is_nonrot))
1444 1445 1446 1447
		queue_flag_set_unlocked(QUEUE_FLAG_NONROT, q);
	else
		queue_flag_clear_unlocked(QUEUE_FLAG_NONROT, q);

M
Martin K. Petersen 已提交
1448
	dm_table_set_integrity(t);
K
Kiyoshi Ueda 已提交
1449

1450 1451 1452 1453 1454 1455 1456 1457 1458
	/*
	 * Determine whether or not this queue's I/O timings contribute
	 * to the entropy pool, Only request-based targets use this.
	 * Clear QUEUE_FLAG_ADD_RANDOM if any underlying device does not
	 * have it set.
	 */
	if (blk_queue_add_random(q) && dm_table_all_devices_attribute(t, device_is_not_random))
		queue_flag_clear_unlocked(QUEUE_FLAG_ADD_RANDOM, q);

K
Kiyoshi Ueda 已提交
1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470
	/*
	 * QUEUE_FLAG_STACKABLE must be set after all queue settings are
	 * visible to other CPUs because, once the flag is set, incoming bios
	 * are processed by request-based dm, which refers to the queue
	 * settings.
	 * Until the flag set, bios are passed to bio-based dm and queued to
	 * md->deferred where queue settings are not needed yet.
	 * Those bios are passed to request-based dm at the resume time.
	 */
	smp_mb();
	if (dm_table_request_based(t))
		queue_flag_set_unlocked(QUEUE_FLAG_STACKABLE, q);
L
Linus Torvalds 已提交
1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482
}

unsigned int dm_table_get_num_targets(struct dm_table *t)
{
	return t->num_targets;
}

struct list_head *dm_table_get_devices(struct dm_table *t)
{
	return &t->devices;
}

1483
fmode_t dm_table_get_mode(struct dm_table *t)
L
Linus Torvalds 已提交
1484 1485 1486
{
	return t->mode;
}
1487
EXPORT_SYMBOL(dm_table_get_mode);
L
Linus Torvalds 已提交
1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506

static void suspend_targets(struct dm_table *t, unsigned postsuspend)
{
	int i = t->num_targets;
	struct dm_target *ti = t->targets;

	while (i--) {
		if (postsuspend) {
			if (ti->type->postsuspend)
				ti->type->postsuspend(ti);
		} else if (ti->type->presuspend)
			ti->type->presuspend(ti);

		ti++;
	}
}

void dm_table_presuspend_targets(struct dm_table *t)
{
1507 1508 1509
	if (!t)
		return;

1510
	suspend_targets(t, 0);
L
Linus Torvalds 已提交
1511 1512 1513 1514
}

void dm_table_postsuspend_targets(struct dm_table *t)
{
1515 1516 1517
	if (!t)
		return;

1518
	suspend_targets(t, 1);
L
Linus Torvalds 已提交
1519 1520
}

1521
int dm_table_resume_targets(struct dm_table *t)
L
Linus Torvalds 已提交
1522
{
1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534
	int i, r = 0;

	for (i = 0; i < t->num_targets; i++) {
		struct dm_target *ti = t->targets + i;

		if (!ti->type->preresume)
			continue;

		r = ti->type->preresume(ti);
		if (r)
			return r;
	}
L
Linus Torvalds 已提交
1535 1536 1537 1538 1539 1540 1541

	for (i = 0; i < t->num_targets; i++) {
		struct dm_target *ti = t->targets + i;

		if (ti->type->resume)
			ti->type->resume(ti);
	}
1542 1543

	return 0;
L
Linus Torvalds 已提交
1544 1545
}

1546 1547 1548 1549 1550 1551
void dm_table_add_target_callbacks(struct dm_table *t, struct dm_target_callbacks *cb)
{
	list_add(&cb->list, &t->target_callbacks);
}
EXPORT_SYMBOL_GPL(dm_table_add_target_callbacks);

L
Linus Torvalds 已提交
1552 1553
int dm_table_any_congested(struct dm_table *t, int bdi_bits)
{
M
Mikulas Patocka 已提交
1554
	struct dm_dev_internal *dd;
P
Paul Jimenez 已提交
1555
	struct list_head *devices = dm_table_get_devices(t);
1556
	struct dm_target_callbacks *cb;
L
Linus Torvalds 已提交
1557 1558
	int r = 0;

P
Paul Jimenez 已提交
1559
	list_for_each_entry(dd, devices, list) {
M
Mikulas Patocka 已提交
1560
		struct request_queue *q = bdev_get_queue(dd->dm_dev.bdev);
A
Alasdair G Kergon 已提交
1561 1562 1563 1564 1565 1566 1567 1568
		char b[BDEVNAME_SIZE];

		if (likely(q))
			r |= bdi_congested(&q->backing_dev_info, bdi_bits);
		else
			DMWARN_LIMIT("%s: any_congested: nonexistent device %s",
				     dm_device_name(t->md),
				     bdevname(dd->dm_dev.bdev, b));
L
Linus Torvalds 已提交
1569 1570
	}

1571 1572 1573 1574
	list_for_each_entry(cb, &t->target_callbacks, list)
		if (cb->congested_fn)
			r |= cb->congested_fn(cb, bdi_bits);

L
Linus Torvalds 已提交
1575 1576 1577
	return r;
}

1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591
int dm_table_any_busy_target(struct dm_table *t)
{
	unsigned i;
	struct dm_target *ti;

	for (i = 0; i < t->num_targets; i++) {
		ti = t->targets + i;
		if (ti->type->busy && ti->type->busy(ti))
			return 1;
	}

	return 0;
}

M
Mike Anderson 已提交
1592 1593 1594 1595
struct mapped_device *dm_table_get_md(struct dm_table *t)
{
	return t->md;
}
1596
EXPORT_SYMBOL(dm_table_get_md);
M
Mike Anderson 已提交
1597

M
Mike Snitzer 已提交
1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611
static int device_discard_capable(struct dm_target *ti, struct dm_dev *dev,
				  sector_t start, sector_t len, void *data)
{
	struct request_queue *q = bdev_get_queue(dev->bdev);

	return q && blk_queue_discard(q);
}

bool dm_table_supports_discards(struct dm_table *t)
{
	struct dm_target *ti;
	unsigned i = 0;

	/*
1612 1613
	 * Unless any target used by the table set discards_supported,
	 * require at least one underlying device to support discards.
M
Mike Snitzer 已提交
1614 1615
	 * t->devices includes internal dm devices such as mirror logs
	 * so we need to use iterate_devices here, which targets
M
Mike Snitzer 已提交
1616
	 * supporting discard selectively must provide.
M
Mike Snitzer 已提交
1617 1618 1619 1620
	 */
	while (i < dm_table_get_num_targets(t)) {
		ti = dm_table_get_target(t, i++);

M
Mike Snitzer 已提交
1621 1622 1623
		if (!ti->num_discard_requests)
			continue;

1624 1625 1626
		if (ti->discards_supported)
			return 1;

M
Mike Snitzer 已提交
1627 1628 1629 1630 1631 1632 1633
		if (ti->type->iterate_devices &&
		    ti->type->iterate_devices(ti, device_discard_capable, NULL))
			return 1;
	}

	return 0;
}