dm.c 23.7 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
/*
 * Copyright (C) 2001, 2002 Sistina Software (UK) Limited.
 * Copyright (C) 2004 Red Hat, Inc. All rights reserved.
 *
 * This file is released under the GPL.
 */

#include "dm.h"
#include "dm-bio-list.h"

#include <linux/init.h>
#include <linux/module.h>
#include <linux/moduleparam.h>
#include <linux/blkpg.h>
#include <linux/bio.h>
#include <linux/buffer_head.h>
#include <linux/mempool.h>
#include <linux/slab.h>
#include <linux/idr.h>

static const char *_name = DM_NAME;

static unsigned int major = 0;
static unsigned int _major = 0;

/*
 * One of these is allocated per bio.
 */
struct dm_io {
	struct mapped_device *md;
	int error;
	struct bio *bio;
	atomic_t io_count;
};

/*
 * One of these is allocated per target within a bio.  Hopefully
 * this will be simplified out one day.
 */
struct target_io {
	struct dm_io *io;
	struct dm_target *ti;
	union map_info info;
};

union map_info *dm_get_mapinfo(struct bio *bio)
{
        if (bio && bio->bi_private)
                return &((struct target_io *)bio->bi_private)->info;
        return NULL;
}

/*
 * Bits for the md->flags field.
 */
#define DMF_BLOCK_IO 0
#define DMF_SUSPENDED 1
58
#define DMF_FROZEN 2
Linus Torvalds's avatar
Linus Torvalds committed
59 60

struct mapped_device {
61 62
	struct rw_semaphore io_lock;
	struct semaphore suspend_lock;
Linus Torvalds's avatar
Linus Torvalds committed
63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100
	rwlock_t map_lock;
	atomic_t holders;

	unsigned long flags;

	request_queue_t *queue;
	struct gendisk *disk;

	void *interface_ptr;

	/*
	 * A list of ios that arrived while we were suspended.
	 */
	atomic_t pending;
	wait_queue_head_t wait;
 	struct bio_list deferred;

	/*
	 * The current mapping.
	 */
	struct dm_table *map;

	/*
	 * io objects are allocated from here.
	 */
	mempool_t *io_pool;
	mempool_t *tio_pool;

	/*
	 * Event handling.
	 */
	atomic_t event_nr;
	wait_queue_head_t eventq;

	/*
	 * freeze/thaw support require holding onto a super block
	 */
	struct super_block *frozen_sb;
101
	struct block_device *suspended_bdev;
Linus Torvalds's avatar
Linus Torvalds committed
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251
};

#define MIN_IOS 256
static kmem_cache_t *_io_cache;
static kmem_cache_t *_tio_cache;

static struct bio_set *dm_set;

static int __init local_init(void)
{
	int r;

	dm_set = bioset_create(16, 16, 4);
	if (!dm_set)
		return -ENOMEM;

	/* allocate a slab for the dm_ios */
	_io_cache = kmem_cache_create("dm_io",
				      sizeof(struct dm_io), 0, 0, NULL, NULL);
	if (!_io_cache)
		return -ENOMEM;

	/* allocate a slab for the target ios */
	_tio_cache = kmem_cache_create("dm_tio", sizeof(struct target_io),
				       0, 0, NULL, NULL);
	if (!_tio_cache) {
		kmem_cache_destroy(_io_cache);
		return -ENOMEM;
	}

	_major = major;
	r = register_blkdev(_major, _name);
	if (r < 0) {
		kmem_cache_destroy(_tio_cache);
		kmem_cache_destroy(_io_cache);
		return r;
	}

	if (!_major)
		_major = r;

	return 0;
}

static void local_exit(void)
{
	kmem_cache_destroy(_tio_cache);
	kmem_cache_destroy(_io_cache);

	bioset_free(dm_set);

	if (unregister_blkdev(_major, _name) < 0)
		DMERR("devfs_unregister_blkdev failed");

	_major = 0;

	DMINFO("cleaned up");
}

int (*_inits[])(void) __initdata = {
	local_init,
	dm_target_init,
	dm_linear_init,
	dm_stripe_init,
	dm_interface_init,
};

void (*_exits[])(void) = {
	local_exit,
	dm_target_exit,
	dm_linear_exit,
	dm_stripe_exit,
	dm_interface_exit,
};

static int __init dm_init(void)
{
	const int count = ARRAY_SIZE(_inits);

	int r, i;

	for (i = 0; i < count; i++) {
		r = _inits[i]();
		if (r)
			goto bad;
	}

	return 0;

      bad:
	while (i--)
		_exits[i]();

	return r;
}

static void __exit dm_exit(void)
{
	int i = ARRAY_SIZE(_exits);

	while (i--)
		_exits[i]();
}

/*
 * Block device functions
 */
static int dm_blk_open(struct inode *inode, struct file *file)
{
	struct mapped_device *md;

	md = inode->i_bdev->bd_disk->private_data;
	dm_get(md);
	return 0;
}

static int dm_blk_close(struct inode *inode, struct file *file)
{
	struct mapped_device *md;

	md = inode->i_bdev->bd_disk->private_data;
	dm_put(md);
	return 0;
}

static inline struct dm_io *alloc_io(struct mapped_device *md)
{
	return mempool_alloc(md->io_pool, GFP_NOIO);
}

static inline void free_io(struct mapped_device *md, struct dm_io *io)
{
	mempool_free(io, md->io_pool);
}

static inline struct target_io *alloc_tio(struct mapped_device *md)
{
	return mempool_alloc(md->tio_pool, GFP_NOIO);
}

static inline void free_tio(struct mapped_device *md, struct target_io *tio)
{
	mempool_free(tio, md->tio_pool);
}

/*
 * Add the bio to the list of deferred io.
 */
static int queue_io(struct mapped_device *md, struct bio *bio)
{
252
	down_write(&md->io_lock);
Linus Torvalds's avatar
Linus Torvalds committed
253 254

	if (!test_bit(DMF_BLOCK_IO, &md->flags)) {
255
		up_write(&md->io_lock);
Linus Torvalds's avatar
Linus Torvalds committed
256 257 258 259 260
		return 1;
	}

	bio_list_add(&md->deferred, bio);

261
	up_write(&md->io_lock);
Linus Torvalds's avatar
Linus Torvalds committed
262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295
	return 0;		/* deferred successfully */
}

/*
 * Everyone (including functions in this file), should use this
 * function to access the md->map field, and make sure they call
 * dm_table_put() when finished.
 */
struct dm_table *dm_get_table(struct mapped_device *md)
{
	struct dm_table *t;

	read_lock(&md->map_lock);
	t = md->map;
	if (t)
		dm_table_get(t);
	read_unlock(&md->map_lock);

	return t;
}

/*-----------------------------------------------------------------
 * CRUD START:
 *   A more elegant soln is in the works that uses the queue
 *   merge fn, unfortunately there are a couple of changes to
 *   the block layer that I want to make for this.  So in the
 *   interests of getting something for people to use I give
 *   you this clearly demarcated crap.
 *---------------------------------------------------------------*/

/*
 * Decrements the number of outstanding ios that a bio has been
 * cloned into, completing the original io if necc.
 */
296
static void dec_pending(struct dm_io *io, int error)
Linus Torvalds's avatar
Linus Torvalds committed
297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387
{
	if (error)
		io->error = error;

	if (atomic_dec_and_test(&io->io_count)) {
		if (atomic_dec_and_test(&io->md->pending))
			/* nudge anyone waiting on suspend queue */
			wake_up(&io->md->wait);

		bio_endio(io->bio, io->bio->bi_size, io->error);
		free_io(io->md, io);
	}
}

static int clone_endio(struct bio *bio, unsigned int done, int error)
{
	int r = 0;
	struct target_io *tio = bio->bi_private;
	struct dm_io *io = tio->io;
	dm_endio_fn endio = tio->ti->type->end_io;

	if (bio->bi_size)
		return 1;

	if (!bio_flagged(bio, BIO_UPTODATE) && !error)
		error = -EIO;

	if (endio) {
		r = endio(tio->ti, bio, error, &tio->info);
		if (r < 0)
			error = r;

		else if (r > 0)
			/* the target wants another shot at the io */
			return 1;
	}

	free_tio(io->md, tio);
	dec_pending(io, error);
	bio_put(bio);
	return r;
}

static sector_t max_io_len(struct mapped_device *md,
			   sector_t sector, struct dm_target *ti)
{
	sector_t offset = sector - ti->begin;
	sector_t len = ti->len - offset;

	/*
	 * Does the target need to split even further ?
	 */
	if (ti->split_io) {
		sector_t boundary;
		boundary = ((offset + ti->split_io) & ~(ti->split_io - 1))
			   - offset;
		if (len > boundary)
			len = boundary;
	}

	return len;
}

static void __map_bio(struct dm_target *ti, struct bio *clone,
		      struct target_io *tio)
{
	int r;

	/*
	 * Sanity checks.
	 */
	BUG_ON(!clone->bi_size);

	clone->bi_end_io = clone_endio;
	clone->bi_private = tio;

	/*
	 * Map the clone.  If r == 0 we don't need to do
	 * anything, the target has assumed ownership of
	 * this io.
	 */
	atomic_inc(&tio->io->io_count);
	r = ti->type->map(ti, clone, &tio->info);
	if (r > 0)
		/* the bio has been remapped so dispatch it */
		generic_make_request(clone);

	else if (r < 0) {
		/* error the io and bail out */
		struct dm_io *io = tio->io;
		free_tio(tio->io->md, tio);
388
		dec_pending(io, r);
Linus Torvalds's avatar
Linus Torvalds committed
389 390 391 392 393 394 395 396 397 398 399 400 401 402
		bio_put(clone);
	}
}

struct clone_info {
	struct mapped_device *md;
	struct dm_table *map;
	struct bio *bio;
	struct dm_io *io;
	sector_t sector;
	sector_t sector_count;
	unsigned short idx;
};

Peter Osterlund's avatar
Peter Osterlund committed
403 404 405 406 407
static void dm_bio_destructor(struct bio *bio)
{
	bio_free(bio, dm_set);
}

Linus Torvalds's avatar
Linus Torvalds committed
408 409 410 411 412 413 414 415 416 417 418
/*
 * Creates a little bio that is just does part of a bvec.
 */
static struct bio *split_bvec(struct bio *bio, sector_t sector,
			      unsigned short idx, unsigned int offset,
			      unsigned int len)
{
	struct bio *clone;
	struct bio_vec *bv = bio->bi_io_vec + idx;

	clone = bio_alloc_bioset(GFP_NOIO, 1, dm_set);
Peter Osterlund's avatar
Peter Osterlund committed
419
	clone->bi_destructor = dm_bio_destructor;
Linus Torvalds's avatar
Linus Torvalds committed
420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577
	*clone->bi_io_vec = *bv;

	clone->bi_sector = sector;
	clone->bi_bdev = bio->bi_bdev;
	clone->bi_rw = bio->bi_rw;
	clone->bi_vcnt = 1;
	clone->bi_size = to_bytes(len);
	clone->bi_io_vec->bv_offset = offset;
	clone->bi_io_vec->bv_len = clone->bi_size;

	return clone;
}

/*
 * Creates a bio that consists of range of complete bvecs.
 */
static struct bio *clone_bio(struct bio *bio, sector_t sector,
			     unsigned short idx, unsigned short bv_count,
			     unsigned int len)
{
	struct bio *clone;

	clone = bio_clone(bio, GFP_NOIO);
	clone->bi_sector = sector;
	clone->bi_idx = idx;
	clone->bi_vcnt = idx + bv_count;
	clone->bi_size = to_bytes(len);
	clone->bi_flags &= ~(1 << BIO_SEG_VALID);

	return clone;
}

static void __clone_and_map(struct clone_info *ci)
{
	struct bio *clone, *bio = ci->bio;
	struct dm_target *ti = dm_table_find_target(ci->map, ci->sector);
	sector_t len = 0, max = max_io_len(ci->md, ci->sector, ti);
	struct target_io *tio;

	/*
	 * Allocate a target io object.
	 */
	tio = alloc_tio(ci->md);
	tio->io = ci->io;
	tio->ti = ti;
	memset(&tio->info, 0, sizeof(tio->info));

	if (ci->sector_count <= max) {
		/*
		 * Optimise for the simple case where we can do all of
		 * the remaining io with a single clone.
		 */
		clone = clone_bio(bio, ci->sector, ci->idx,
				  bio->bi_vcnt - ci->idx, ci->sector_count);
		__map_bio(ti, clone, tio);
		ci->sector_count = 0;

	} else if (to_sector(bio->bi_io_vec[ci->idx].bv_len) <= max) {
		/*
		 * There are some bvecs that don't span targets.
		 * Do as many of these as possible.
		 */
		int i;
		sector_t remaining = max;
		sector_t bv_len;

		for (i = ci->idx; remaining && (i < bio->bi_vcnt); i++) {
			bv_len = to_sector(bio->bi_io_vec[i].bv_len);

			if (bv_len > remaining)
				break;

			remaining -= bv_len;
			len += bv_len;
		}

		clone = clone_bio(bio, ci->sector, ci->idx, i - ci->idx, len);
		__map_bio(ti, clone, tio);

		ci->sector += len;
		ci->sector_count -= len;
		ci->idx = i;

	} else {
		/*
		 * Create two copy bios to deal with io that has
		 * been split across a target.
		 */
		struct bio_vec *bv = bio->bi_io_vec + ci->idx;

		clone = split_bvec(bio, ci->sector, ci->idx,
				   bv->bv_offset, max);
		__map_bio(ti, clone, tio);

		ci->sector += max;
		ci->sector_count -= max;
		ti = dm_table_find_target(ci->map, ci->sector);

		len = to_sector(bv->bv_len) - max;
		clone = split_bvec(bio, ci->sector, ci->idx,
				   bv->bv_offset + to_bytes(max), len);
		tio = alloc_tio(ci->md);
		tio->io = ci->io;
		tio->ti = ti;
		memset(&tio->info, 0, sizeof(tio->info));
		__map_bio(ti, clone, tio);

		ci->sector += len;
		ci->sector_count -= len;
		ci->idx++;
	}
}

/*
 * Split the bio into several clones.
 */
static void __split_bio(struct mapped_device *md, struct bio *bio)
{
	struct clone_info ci;

	ci.map = dm_get_table(md);
	if (!ci.map) {
		bio_io_error(bio, bio->bi_size);
		return;
	}

	ci.md = md;
	ci.bio = bio;
	ci.io = alloc_io(md);
	ci.io->error = 0;
	atomic_set(&ci.io->io_count, 1);
	ci.io->bio = bio;
	ci.io->md = md;
	ci.sector = bio->bi_sector;
	ci.sector_count = bio_sectors(bio);
	ci.idx = bio->bi_idx;

	atomic_inc(&md->pending);
	while (ci.sector_count)
		__clone_and_map(&ci);

	/* drop the extra reference count */
	dec_pending(ci.io, 0);
	dm_table_put(ci.map);
}
/*-----------------------------------------------------------------
 * CRUD END
 *---------------------------------------------------------------*/

/*
 * The request function that just remaps the bio built up by
 * dm_merge_bvec.
 */
static int dm_request(request_queue_t *q, struct bio *bio)
{
	int r;
	struct mapped_device *md = q->queuedata;

578
	down_read(&md->io_lock);
Linus Torvalds's avatar
Linus Torvalds committed
579 580 581 582 583 584

	/*
	 * If we're suspended we have to queue
	 * this io for later.
	 */
	while (test_bit(DMF_BLOCK_IO, &md->flags)) {
585
		up_read(&md->io_lock);
Linus Torvalds's avatar
Linus Torvalds committed
586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603

		if (bio_rw(bio) == READA) {
			bio_io_error(bio, bio->bi_size);
			return 0;
		}

		r = queue_io(md, bio);
		if (r < 0) {
			bio_io_error(bio, bio->bi_size);
			return 0;

		} else if (r == 0)
			return 0;	/* deferred successfully */

		/*
		 * We're in a while loop, because someone could suspend
		 * before we get to the following read lock.
		 */
604
		down_read(&md->io_lock);
Linus Torvalds's avatar
Linus Torvalds committed
605 606 607
	}

	__split_bio(md, bio);
608
	up_read(&md->io_lock);
Linus Torvalds's avatar
Linus Torvalds committed
609 610 611 612 613 614 615 616 617 618 619
	return 0;
}

static int dm_flush_all(request_queue_t *q, struct gendisk *disk,
			sector_t *error_sector)
{
	struct mapped_device *md = q->queuedata;
	struct dm_table *map = dm_get_table(md);
	int ret = -ENXIO;

	if (map) {
620
		ret = dm_table_flush_all(map);
Linus Torvalds's avatar
Linus Torvalds committed
621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756
		dm_table_put(map);
	}

	return ret;
}

static void dm_unplug_all(request_queue_t *q)
{
	struct mapped_device *md = q->queuedata;
	struct dm_table *map = dm_get_table(md);

	if (map) {
		dm_table_unplug_all(map);
		dm_table_put(map);
	}
}

static int dm_any_congested(void *congested_data, int bdi_bits)
{
	int r;
	struct mapped_device *md = (struct mapped_device *) congested_data;
	struct dm_table *map = dm_get_table(md);

	if (!map || test_bit(DMF_BLOCK_IO, &md->flags))
		r = bdi_bits;
	else
		r = dm_table_any_congested(map, bdi_bits);

	dm_table_put(map);
	return r;
}

/*-----------------------------------------------------------------
 * An IDR is used to keep track of allocated minor numbers.
 *---------------------------------------------------------------*/
static DECLARE_MUTEX(_minor_lock);
static DEFINE_IDR(_minor_idr);

static void free_minor(unsigned int minor)
{
	down(&_minor_lock);
	idr_remove(&_minor_idr, minor);
	up(&_minor_lock);
}

/*
 * See if the device with a specific minor # is free.
 */
static int specific_minor(struct mapped_device *md, unsigned int minor)
{
	int r, m;

	if (minor >= (1 << MINORBITS))
		return -EINVAL;

	down(&_minor_lock);

	if (idr_find(&_minor_idr, minor)) {
		r = -EBUSY;
		goto out;
	}

	r = idr_pre_get(&_minor_idr, GFP_KERNEL);
	if (!r) {
		r = -ENOMEM;
		goto out;
	}

	r = idr_get_new_above(&_minor_idr, md, minor, &m);
	if (r) {
		goto out;
	}

	if (m != minor) {
		idr_remove(&_minor_idr, m);
		r = -EBUSY;
		goto out;
	}

out:
	up(&_minor_lock);
	return r;
}

static int next_free_minor(struct mapped_device *md, unsigned int *minor)
{
	int r;
	unsigned int m;

	down(&_minor_lock);

	r = idr_pre_get(&_minor_idr, GFP_KERNEL);
	if (!r) {
		r = -ENOMEM;
		goto out;
	}

	r = idr_get_new(&_minor_idr, md, &m);
	if (r) {
		goto out;
	}

	if (m >= (1 << MINORBITS)) {
		idr_remove(&_minor_idr, m);
		r = -ENOSPC;
		goto out;
	}

	*minor = m;

out:
	up(&_minor_lock);
	return r;
}

static struct block_device_operations dm_blk_dops;

/*
 * Allocate and initialise a blank device with a given minor.
 */
static struct mapped_device *alloc_dev(unsigned int minor, int persistent)
{
	int r;
	struct mapped_device *md = kmalloc(sizeof(*md), GFP_KERNEL);

	if (!md) {
		DMWARN("unable to allocate device, out of memory.");
		return NULL;
	}

	/* get a minor number for the dev */
	r = persistent ? specific_minor(md, minor) : next_free_minor(md, &minor);
	if (r < 0)
		goto bad1;

	memset(md, 0, sizeof(*md));
757 758
	init_rwsem(&md->io_lock);
	init_MUTEX(&md->suspend_lock);
Linus Torvalds's avatar
Linus Torvalds committed
759 760 761 762 763 764 765 766 767 768 769 770
	rwlock_init(&md->map_lock);
	atomic_set(&md->holders, 1);
	atomic_set(&md->event_nr, 0);

	md->queue = blk_alloc_queue(GFP_KERNEL);
	if (!md->queue)
		goto bad1;

	md->queue->queuedata = md;
	md->queue->backing_dev_info.congested_fn = dm_any_congested;
	md->queue->backing_dev_info.congested_data = md;
	blk_queue_make_request(md->queue, dm_request);
771
	blk_queue_bounce_limit(md->queue, BLK_BOUNCE_ANY);
Linus Torvalds's avatar
Linus Torvalds committed
772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836
	md->queue->unplug_fn = dm_unplug_all;
	md->queue->issue_flush_fn = dm_flush_all;

	md->io_pool = mempool_create(MIN_IOS, mempool_alloc_slab,
				     mempool_free_slab, _io_cache);
 	if (!md->io_pool)
 		goto bad2;

	md->tio_pool = mempool_create(MIN_IOS, mempool_alloc_slab,
				      mempool_free_slab, _tio_cache);
	if (!md->tio_pool)
		goto bad3;

	md->disk = alloc_disk(1);
	if (!md->disk)
		goto bad4;

	md->disk->major = _major;
	md->disk->first_minor = minor;
	md->disk->fops = &dm_blk_dops;
	md->disk->queue = md->queue;
	md->disk->private_data = md;
	sprintf(md->disk->disk_name, "dm-%d", minor);
	add_disk(md->disk);

	atomic_set(&md->pending, 0);
	init_waitqueue_head(&md->wait);
	init_waitqueue_head(&md->eventq);

	return md;

 bad4:
	mempool_destroy(md->tio_pool);
 bad3:
	mempool_destroy(md->io_pool);
 bad2:
	blk_put_queue(md->queue);
	free_minor(minor);
 bad1:
	kfree(md);
	return NULL;
}

static void free_dev(struct mapped_device *md)
{
	free_minor(md->disk->first_minor);
	mempool_destroy(md->tio_pool);
	mempool_destroy(md->io_pool);
	del_gendisk(md->disk);
	put_disk(md->disk);
	blk_put_queue(md->queue);
	kfree(md);
}

/*
 * Bind a table to the device.
 */
static void event_callback(void *context)
{
	struct mapped_device *md = (struct mapped_device *) context;

	atomic_inc(&md->event_nr);
	wake_up(&md->eventq);
}

837
static void __set_size(struct mapped_device *md, sector_t size)
Linus Torvalds's avatar
Linus Torvalds committed
838
{
839
	set_capacity(md->disk, size);
Linus Torvalds's avatar
Linus Torvalds committed
840

841
	mutex_lock(&md->suspended_bdev->bd_inode->i_mutex);
842
	i_size_write(md->suspended_bdev->bd_inode, (loff_t)size << SECTOR_SHIFT);
843
	mutex_unlock(&md->suspended_bdev->bd_inode->i_mutex);
Linus Torvalds's avatar
Linus Torvalds committed
844 845 846 847 848 849 850 851
}

static int __bind(struct mapped_device *md, struct dm_table *t)
{
	request_queue_t *q = md->queue;
	sector_t size;

	size = dm_table_get_size(t);
852
	__set_size(md, size);
Linus Torvalds's avatar
Linus Torvalds committed
853 854 855
	if (size == 0)
		return 0;

856 857 858
	dm_table_get(t);
	dm_table_event_callback(t, event_callback, md);

Linus Torvalds's avatar
Linus Torvalds committed
859 860
	write_lock(&md->map_lock);
	md->map = t;
861
	dm_table_set_restrictions(t, q);
Linus Torvalds's avatar
Linus Torvalds committed
862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906
	write_unlock(&md->map_lock);

	return 0;
}

static void __unbind(struct mapped_device *md)
{
	struct dm_table *map = md->map;

	if (!map)
		return;

	dm_table_event_callback(map, NULL, NULL);
	write_lock(&md->map_lock);
	md->map = NULL;
	write_unlock(&md->map_lock);
	dm_table_put(map);
}

/*
 * Constructor for a new device.
 */
static int create_aux(unsigned int minor, int persistent,
		      struct mapped_device **result)
{
	struct mapped_device *md;

	md = alloc_dev(minor, persistent);
	if (!md)
		return -ENXIO;

	*result = md;
	return 0;
}

int dm_create(struct mapped_device **result)
{
	return create_aux(0, 0, result);
}

int dm_create_with_minor(unsigned int minor, struct mapped_device **result)
{
	return create_aux(minor, 1, result);
}

907
static struct mapped_device *dm_find_md(dev_t dev)
Linus Torvalds's avatar
Linus Torvalds committed
908 909 910 911 912 913 914 915 916 917
{
	struct mapped_device *md;
	unsigned minor = MINOR(dev);

	if (MAJOR(dev) != _major || minor >= (1 << MINORBITS))
		return NULL;

	down(&_minor_lock);

	md = idr_find(&_minor_idr, minor);
918 919
	if (!md || (dm_disk(md)->first_minor != minor))
		md = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
920 921 922

	up(&_minor_lock);

923 924 925
	return md;
}

926 927 928 929 930 931 932 933 934 935
struct mapped_device *dm_get_md(dev_t dev)
{
	struct mapped_device *md = dm_find_md(dev);

	if (md)
		dm_get(md);

	return md;
}

936 937 938 939 940 941 942 943
void *dm_get_mdptr(dev_t dev)
{
	struct mapped_device *md;
	void *mdptr = NULL;

	md = dm_find_md(dev);
	if (md)
		mdptr = md->interface_ptr;
Linus Torvalds's avatar
Linus Torvalds committed
944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961
	return mdptr;
}

void dm_set_mdptr(struct mapped_device *md, void *ptr)
{
	md->interface_ptr = ptr;
}

void dm_get(struct mapped_device *md)
{
	atomic_inc(&md->holders);
}

void dm_put(struct mapped_device *md)
{
	struct dm_table *map = dm_get_table(md);

	if (atomic_dec_and_test(&md->holders)) {
962
		if (!dm_suspended(md)) {
Linus Torvalds's avatar
Linus Torvalds committed
963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992
			dm_table_presuspend_targets(map);
			dm_table_postsuspend_targets(map);
		}
		__unbind(md);
		free_dev(md);
	}

	dm_table_put(map);
}

/*
 * Process the deferred bios
 */
static void __flush_deferred_io(struct mapped_device *md, struct bio *c)
{
	struct bio *n;

	while (c) {
		n = c->bi_next;
		c->bi_next = NULL;
		__split_bio(md, c);
		c = n;
	}
}

/*
 * Swap in a new table (destroying old one).
 */
int dm_swap_table(struct mapped_device *md, struct dm_table *table)
{
993
	int r = -EINVAL;
Linus Torvalds's avatar
Linus Torvalds committed
994

995
	down(&md->suspend_lock);
Linus Torvalds's avatar
Linus Torvalds committed
996 997

	/* device must be suspended */
998
	if (!dm_suspended(md))
999
		goto out;
Linus Torvalds's avatar
Linus Torvalds committed
1000 1001 1002 1003

	__unbind(md);
	r = __bind(md, table);

1004
out:
1005
	up(&md->suspend_lock);
1006
	return r;
Linus Torvalds's avatar
Linus Torvalds committed
1007 1008 1009 1010 1011 1012
}

/*
 * Functions to lock and unlock any filesystem running on the
 * device.
 */
1013
static int lock_fs(struct mapped_device *md)
Linus Torvalds's avatar
Linus Torvalds committed
1014
{
1015
	int r;
Linus Torvalds's avatar
Linus Torvalds committed
1016 1017

	WARN_ON(md->frozen_sb);
1018

1019
	md->frozen_sb = freeze_bdev(md->suspended_bdev);
1020
	if (IS_ERR(md->frozen_sb)) {
1021
		r = PTR_ERR(md->frozen_sb);
1022 1023
		md->frozen_sb = NULL;
		return r;
1024 1025
	}

1026 1027
	set_bit(DMF_FROZEN, &md->flags);

Linus Torvalds's avatar
Linus Torvalds committed
1028
	/* don't bdput right now, we don't want the bdev
1029
	 * to go away while it is locked.
Linus Torvalds's avatar
Linus Torvalds committed
1030 1031 1032 1033
	 */
	return 0;
}

1034
static void unlock_fs(struct mapped_device *md)
Linus Torvalds's avatar
Linus Torvalds committed
1035
{
1036 1037 1038
	if (!test_bit(DMF_FROZEN, &md->flags))
		return;

1039
	thaw_bdev(md->suspended_bdev, md->frozen_sb);
Linus Torvalds's avatar
Linus Torvalds committed
1040
	md->frozen_sb = NULL;
1041
	clear_bit(DMF_FROZEN, &md->flags);
Linus Torvalds's avatar
Linus Torvalds committed
1042 1043 1044 1045 1046 1047 1048 1049 1050
}

/*
 * We need to be able to change a mapping table under a mounted
 * filesystem.  For example we might want to move some data in
 * the background.  Before the table can be swapped with
 * dm_bind_table, dm_suspend must be called to flush any in
 * flight bios and ensure that any further io gets deferred.
 */
1051
int dm_suspend(struct mapped_device *md, int do_lockfs)
Linus Torvalds's avatar
Linus Torvalds committed
1052
{
1053
	struct dm_table *map = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
1054
	DECLARE_WAITQUEUE(wait, current);
1055
	int r = -EINVAL;
Linus Torvalds's avatar
Linus Torvalds committed
1056

1057 1058 1059 1060
	down(&md->suspend_lock);

	if (dm_suspended(md))
		goto out;
Linus Torvalds's avatar
Linus Torvalds committed
1061 1062 1063

	map = dm_get_table(md);

1064 1065 1066
	/* This does not get reverted if there's an error later. */
	dm_table_presuspend_targets(map);

1067 1068 1069 1070 1071 1072 1073
	md->suspended_bdev = bdget_disk(md->disk, 0);
	if (!md->suspended_bdev) {
		DMWARN("bdget failed in dm_suspend");
		r = -ENOMEM;
		goto out;
	}

1074
	/* Flush I/O to the device. */
1075 1076 1077 1078 1079
	if (do_lockfs) {
		r = lock_fs(md);
		if (r)
			goto out;
	}
Linus Torvalds's avatar
Linus Torvalds committed
1080 1081

	/*
1082
	 * First we set the BLOCK_IO flag so no more ios will be mapped.
Linus Torvalds's avatar
Linus Torvalds committed
1083
	 */
1084 1085
	down_write(&md->io_lock);
	set_bit(DMF_BLOCK_IO, &md->flags);
Linus Torvalds's avatar
Linus Torvalds committed
1086 1087

	add_wait_queue(&md->wait, &wait);
1088
	up_write(&md->io_lock);
Linus Torvalds's avatar
Linus Torvalds committed
1089 1090

	/* unplug */
1091
	if (map)
Linus Torvalds's avatar
Linus Torvalds committed
1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107
		dm_table_unplug_all(map);

	/*
	 * Then we wait for the already mapped ios to
	 * complete.
	 */
	while (1) {
		set_current_state(TASK_INTERRUPTIBLE);

		if (!atomic_read(&md->pending) || signal_pending(current))
			break;

		io_schedule();
	}
	set_current_state(TASK_RUNNING);

1108
	down_write(&md->io_lock);
Linus Torvalds's avatar
Linus Torvalds committed
1109 1110 1111
	remove_wait_queue(&md->wait, &wait);

	/* were we interrupted ? */
1112
	r = -EINTR;
1113 1114 1115 1116 1117 1118 1119
	if (atomic_read(&md->pending)) {
		up_write(&md->io_lock);
		unlock_fs(md);
		clear_bit(DMF_BLOCK_IO, &md->flags);
		goto out;
	}
	up_write(&md->io_lock);
Linus Torvalds's avatar
Linus Torvalds committed
1120

1121
	dm_table_postsuspend_targets(map);
Linus Torvalds's avatar
Linus Torvalds committed
1122

1123
	set_bit(DMF_SUSPENDED, &md->flags);
1124

1125
	r = 0;
1126

1127
out:
1128 1129 1130 1131 1132
	if (r && md->suspended_bdev) {
		bdput(md->suspended_bdev);
		md->suspended_bdev = NULL;
	}

1133 1134
	dm_table_put(map);
	up(&md->suspend_lock);
1135
	return r;
Linus Torvalds's avatar
Linus Torvalds committed
1136 1137 1138 1139
}

int dm_resume(struct mapped_device *md)
{
1140
	int r = -EINVAL;
Linus Torvalds's avatar
Linus Torvalds committed
1141
	struct bio *def;
1142
	struct dm_table *map = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
1143

1144 1145
	down(&md->suspend_lock);
	if (!dm_suspended(md))
1146 1147 1148
		goto out;

	map = dm_get_table(md);
1149
	if (!map || !dm_table_get_size(map))
1150
		goto out;
Linus Torvalds's avatar
Linus Torvalds committed
1151 1152

	dm_table_resume_targets(map);
1153 1154

	down_write(&md->io_lock);
Linus Torvalds's avatar
Linus Torvalds committed
1155 1156 1157 1158
	clear_bit(DMF_BLOCK_IO, &md->flags);

	def = bio_list_get(&md->deferred);
	__flush_deferred_io(md, def);
1159 1160 1161 1162
	up_write(&md->io_lock);

	unlock_fs(md);

1163 1164 1165
	bdput(md->suspended_bdev);
	md->suspended_bdev = NULL;

1166 1167
	clear_bit(DMF_SUSPENDED, &md->flags);

Linus Torvalds's avatar
Linus Torvalds committed
1168 1169
	dm_table_unplug_all(map);

1170
	r = 0;
1171

1172 1173
out:
	dm_table_put(map);
1174 1175
	up(&md->suspend_lock);

1176
	return r;
Linus Torvalds's avatar
Linus Torvalds committed
1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225
}

/*-----------------------------------------------------------------
 * Event notification.
 *---------------------------------------------------------------*/
uint32_t dm_get_event_nr(struct mapped_device *md)
{
	return atomic_read(&md->event_nr);
}

int dm_wait_event(struct mapped_device *md, int event_nr)
{
	return wait_event_interruptible(md->eventq,
			(event_nr != atomic_read(&md->event_nr)));
}

/*
 * The gendisk is only valid as long as you have a reference
 * count on 'md'.
 */
struct gendisk *dm_disk(struct mapped_device *md)
{
	return md->disk;
}

int dm_suspended(struct mapped_device *md)
{
	return test_bit(DMF_SUSPENDED, &md->flags);
}

static struct block_device_operations dm_blk_dops = {
	.open = dm_blk_open,
	.release = dm_blk_close,
	.owner = THIS_MODULE
};

EXPORT_SYMBOL(dm_get_mapinfo);

/*
 * module hooks
 */
module_init(dm_init);
module_exit(dm_exit);

module_param(major, uint, 0);
MODULE_PARM_DESC(major, "The major number of the device mapper");
MODULE_DESCRIPTION(DM_NAME " driver");
MODULE_AUTHOR("Joe Thornber <dm-devel@redhat.com>");
MODULE_LICENSE("GPL");