scrub.c 91.6 KB
Newer Older
A
Arne Jansen 已提交
1
/*
2
 * Copyright (C) 2011, 2012 STRATO.  All rights reserved.
A
Arne Jansen 已提交
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public
 * License v2 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 021110-1307, USA.
 */

#include <linux/blkdev.h>
20
#include <linux/ratelimit.h>
A
Arne Jansen 已提交
21 22 23 24
#include "ctree.h"
#include "volumes.h"
#include "disk-io.h"
#include "ordered-data.h"
25
#include "transaction.h"
26
#include "backref.h"
27
#include "extent_io.h"
28
#include "dev-replace.h"
29
#include "check-integrity.h"
30
#include "rcu-string.h"
D
David Woodhouse 已提交
31
#include "raid56.h"
A
Arne Jansen 已提交
32 33 34 35 36 37 38 39 40 41 42 43 44 45

/*
 * This is only the first step towards a full-features scrub. It reads all
 * extent and super block and verifies the checksums. In case a bad checksum
 * is found or the extent cannot be read, good data will be written back if
 * any can be found.
 *
 * Future enhancements:
 *  - In case an unrepairable extent is encountered, track which files are
 *    affected and report them
 *  - track and record media errors, throw out bad devices
 *  - add a mode to also read unallocated space
 */

46
struct scrub_block;
47
struct scrub_ctx;
A
Arne Jansen 已提交
48

49 50 51 52 53 54 55 56 57
/*
 * the following three values only influence the performance.
 * The last one configures the number of parallel and outstanding I/O
 * operations. The first two values configure an upper limit for the number
 * of (dynamically allocated) pages that are added to a bio.
 */
#define SCRUB_PAGES_PER_RD_BIO	32	/* 128k per bio */
#define SCRUB_PAGES_PER_WR_BIO	32	/* 128k per bio */
#define SCRUB_BIOS_PER_SCTX	64	/* 8MB per device in flight */
58 59 60 61 62 63

/*
 * the following value times PAGE_SIZE needs to be large enough to match the
 * largest node/leaf/sector size that shall be supported.
 * Values larger than BTRFS_STRIPE_LEN are not supported.
 */
64
#define SCRUB_MAX_PAGES_PER_BLOCK	16	/* 64k per node/leaf/sector */
A
Arne Jansen 已提交
65 66

struct scrub_page {
67 68
	struct scrub_block	*sblock;
	struct page		*page;
69
	struct btrfs_device	*dev;
A
Arne Jansen 已提交
70 71
	u64			flags;  /* extent flags */
	u64			generation;
72 73
	u64			logical;
	u64			physical;
74
	u64			physical_for_dev_replace;
75
	atomic_t		ref_count;
76 77 78 79 80
	struct {
		unsigned int	mirror_num:8;
		unsigned int	have_csum:1;
		unsigned int	io_error:1;
	};
A
Arne Jansen 已提交
81 82 83 84 85
	u8			csum[BTRFS_CSUM_SIZE];
};

struct scrub_bio {
	int			index;
86
	struct scrub_ctx	*sctx;
87
	struct btrfs_device	*dev;
A
Arne Jansen 已提交
88 89 90 91
	struct bio		*bio;
	int			err;
	u64			logical;
	u64			physical;
92 93 94 95 96
#if SCRUB_PAGES_PER_WR_BIO >= SCRUB_PAGES_PER_RD_BIO
	struct scrub_page	*pagev[SCRUB_PAGES_PER_WR_BIO];
#else
	struct scrub_page	*pagev[SCRUB_PAGES_PER_RD_BIO];
#endif
97
	int			page_count;
A
Arne Jansen 已提交
98 99 100 101
	int			next_free;
	struct btrfs_work	work;
};

102
struct scrub_block {
103
	struct scrub_page	*pagev[SCRUB_MAX_PAGES_PER_BLOCK];
104 105 106
	int			page_count;
	atomic_t		outstanding_pages;
	atomic_t		ref_count; /* free mem on transition to zero */
107
	struct scrub_ctx	*sctx;
108 109 110 111
	struct {
		unsigned int	header_error:1;
		unsigned int	checksum_error:1;
		unsigned int	no_io_error_seen:1;
112
		unsigned int	generation_error:1; /* also sets header_error */
113 114 115
	};
};

116 117 118 119 120 121 122 123
struct scrub_wr_ctx {
	struct scrub_bio *wr_curr_bio;
	struct btrfs_device *tgtdev;
	int pages_per_wr_bio; /* <= SCRUB_PAGES_PER_WR_BIO */
	atomic_t flush_all_writes;
	struct mutex wr_lock;
};

124
struct scrub_ctx {
125
	struct scrub_bio	*bios[SCRUB_BIOS_PER_SCTX];
126
	struct btrfs_root	*dev_root;
A
Arne Jansen 已提交
127 128
	int			first_free;
	int			curr;
129 130
	atomic_t		bios_in_flight;
	atomic_t		workers_pending;
A
Arne Jansen 已提交
131 132 133 134 135
	spinlock_t		list_lock;
	wait_queue_head_t	list_wait;
	u16			csum_size;
	struct list_head	csum_list;
	atomic_t		cancel_req;
A
Arne Jansen 已提交
136
	int			readonly;
137
	int			pages_per_rd_bio;
138 139 140
	u32			sectorsize;
	u32			nodesize;
	u32			leafsize;
141 142

	int			is_dev_replace;
143
	struct scrub_wr_ctx	wr_ctx;
144

A
Arne Jansen 已提交
145 146 147 148 149 150 151
	/*
	 * statistics
	 */
	struct btrfs_scrub_progress stat;
	spinlock_t		stat_lock;
};

152
struct scrub_fixup_nodatasum {
153
	struct scrub_ctx	*sctx;
154
	struct btrfs_device	*dev;
155 156 157 158 159 160
	u64			logical;
	struct btrfs_root	*root;
	struct btrfs_work	work;
	int			mirror_num;
};

161 162 163 164 165 166 167
struct scrub_nocow_inode {
	u64			inum;
	u64			offset;
	u64			root;
	struct list_head	list;
};

168 169 170 171 172 173
struct scrub_copy_nocow_ctx {
	struct scrub_ctx	*sctx;
	u64			logical;
	u64			len;
	int			mirror_num;
	u64			physical_for_dev_replace;
174
	struct list_head	inodes;
175 176 177
	struct btrfs_work	work;
};

178 179 180 181 182 183 184 185 186 187 188 189 190
struct scrub_warning {
	struct btrfs_path	*path;
	u64			extent_item_size;
	char			*scratch_buf;
	char			*msg_buf;
	const char		*errstr;
	sector_t		sector;
	u64			logical;
	struct btrfs_device	*dev;
	int			msg_bufsize;
	int			scratch_bufsize;
};

191

192 193 194 195
static void scrub_pending_bio_inc(struct scrub_ctx *sctx);
static void scrub_pending_bio_dec(struct scrub_ctx *sctx);
static void scrub_pending_trans_workers_inc(struct scrub_ctx *sctx);
static void scrub_pending_trans_workers_dec(struct scrub_ctx *sctx);
196
static int scrub_handle_errored_block(struct scrub_block *sblock_to_check);
197
static int scrub_setup_recheck_block(struct scrub_ctx *sctx,
198
				     struct btrfs_fs_info *fs_info,
199
				     struct scrub_block *original_sblock,
200
				     u64 length, u64 logical,
201
				     struct scrub_block *sblocks_for_recheck);
202 203 204 205
static void scrub_recheck_block(struct btrfs_fs_info *fs_info,
				struct scrub_block *sblock, int is_metadata,
				int have_csum, u8 *csum, u64 generation,
				u16 csum_size);
206 207 208 209 210 211 212 213 214 215 216 217
static void scrub_recheck_block_checksum(struct btrfs_fs_info *fs_info,
					 struct scrub_block *sblock,
					 int is_metadata, int have_csum,
					 const u8 *csum, u64 generation,
					 u16 csum_size);
static void scrub_complete_bio_end_io(struct bio *bio, int err);
static int scrub_repair_block_from_good_copy(struct scrub_block *sblock_bad,
					     struct scrub_block *sblock_good,
					     int force_write);
static int scrub_repair_page_from_good_copy(struct scrub_block *sblock_bad,
					    struct scrub_block *sblock_good,
					    int page_num, int force_write);
218 219 220
static void scrub_write_block_to_dev_replace(struct scrub_block *sblock);
static int scrub_write_page_to_dev_replace(struct scrub_block *sblock,
					   int page_num);
221 222 223 224 225
static int scrub_checksum_data(struct scrub_block *sblock);
static int scrub_checksum_tree_block(struct scrub_block *sblock);
static int scrub_checksum_super(struct scrub_block *sblock);
static void scrub_block_get(struct scrub_block *sblock);
static void scrub_block_put(struct scrub_block *sblock);
226 227
static void scrub_page_get(struct scrub_page *spage);
static void scrub_page_put(struct scrub_page *spage);
228 229
static int scrub_add_page_to_rd_bio(struct scrub_ctx *sctx,
				    struct scrub_page *spage);
230
static int scrub_pages(struct scrub_ctx *sctx, u64 logical, u64 len,
231
		       u64 physical, struct btrfs_device *dev, u64 flags,
232 233
		       u64 gen, int mirror_num, u8 *csum, int force,
		       u64 physical_for_dev_replace);
S
Stefan Behrens 已提交
234
static void scrub_bio_end_io(struct bio *bio, int err);
235 236
static void scrub_bio_end_io_worker(struct btrfs_work *work);
static void scrub_block_complete(struct scrub_block *sblock);
237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255
static void scrub_remap_extent(struct btrfs_fs_info *fs_info,
			       u64 extent_logical, u64 extent_len,
			       u64 *extent_physical,
			       struct btrfs_device **extent_dev,
			       int *extent_mirror_num);
static int scrub_setup_wr_ctx(struct scrub_ctx *sctx,
			      struct scrub_wr_ctx *wr_ctx,
			      struct btrfs_fs_info *fs_info,
			      struct btrfs_device *dev,
			      int is_dev_replace);
static void scrub_free_wr_ctx(struct scrub_wr_ctx *wr_ctx);
static int scrub_add_page_to_wr_bio(struct scrub_ctx *sctx,
				    struct scrub_page *spage);
static void scrub_wr_submit(struct scrub_ctx *sctx);
static void scrub_wr_bio_end_io(struct bio *bio, int err);
static void scrub_wr_bio_end_io_worker(struct btrfs_work *work);
static int write_page_nocow(struct scrub_ctx *sctx,
			    u64 physical_for_dev_replace, struct page *page);
static int copy_nocow_pages_for_inode(u64 inum, u64 offset, u64 root,
256
				      struct scrub_copy_nocow_ctx *ctx);
257 258 259
static int copy_nocow_pages(struct scrub_ctx *sctx, u64 logical, u64 len,
			    int mirror_num, u64 physical_for_dev_replace);
static void copy_nocow_pages_worker(struct btrfs_work *work);
S
Stefan Behrens 已提交
260 261


262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314
static void scrub_pending_bio_inc(struct scrub_ctx *sctx)
{
	atomic_inc(&sctx->bios_in_flight);
}

static void scrub_pending_bio_dec(struct scrub_ctx *sctx)
{
	atomic_dec(&sctx->bios_in_flight);
	wake_up(&sctx->list_wait);
}

/*
 * used for workers that require transaction commits (i.e., for the
 * NOCOW case)
 */
static void scrub_pending_trans_workers_inc(struct scrub_ctx *sctx)
{
	struct btrfs_fs_info *fs_info = sctx->dev_root->fs_info;

	/*
	 * increment scrubs_running to prevent cancel requests from
	 * completing as long as a worker is running. we must also
	 * increment scrubs_paused to prevent deadlocking on pause
	 * requests used for transactions commits (as the worker uses a
	 * transaction context). it is safe to regard the worker
	 * as paused for all matters practical. effectively, we only
	 * avoid cancellation requests from completing.
	 */
	mutex_lock(&fs_info->scrub_lock);
	atomic_inc(&fs_info->scrubs_running);
	atomic_inc(&fs_info->scrubs_paused);
	mutex_unlock(&fs_info->scrub_lock);
	atomic_inc(&sctx->workers_pending);
}

/* used for workers that require transaction commits */
static void scrub_pending_trans_workers_dec(struct scrub_ctx *sctx)
{
	struct btrfs_fs_info *fs_info = sctx->dev_root->fs_info;

	/*
	 * see scrub_pending_trans_workers_inc() why we're pretending
	 * to be paused in the scrub counters
	 */
	mutex_lock(&fs_info->scrub_lock);
	atomic_dec(&fs_info->scrubs_running);
	atomic_dec(&fs_info->scrubs_paused);
	mutex_unlock(&fs_info->scrub_lock);
	atomic_dec(&sctx->workers_pending);
	wake_up(&fs_info->scrub_pause_wait);
	wake_up(&sctx->list_wait);
}

315
static void scrub_free_csums(struct scrub_ctx *sctx)
A
Arne Jansen 已提交
316
{
317
	while (!list_empty(&sctx->csum_list)) {
A
Arne Jansen 已提交
318
		struct btrfs_ordered_sum *sum;
319
		sum = list_first_entry(&sctx->csum_list,
A
Arne Jansen 已提交
320 321 322 323 324 325
				       struct btrfs_ordered_sum, list);
		list_del(&sum->list);
		kfree(sum);
	}
}

326
static noinline_for_stack void scrub_free_ctx(struct scrub_ctx *sctx)
A
Arne Jansen 已提交
327 328 329
{
	int i;

330
	if (!sctx)
A
Arne Jansen 已提交
331 332
		return;

333 334
	scrub_free_wr_ctx(&sctx->wr_ctx);

335
	/* this can happen when scrub is cancelled */
336 337
	if (sctx->curr != -1) {
		struct scrub_bio *sbio = sctx->bios[sctx->curr];
338 339

		for (i = 0; i < sbio->page_count; i++) {
340
			WARN_ON(!sbio->pagev[i]->page);
341 342 343 344 345
			scrub_block_put(sbio->pagev[i]->sblock);
		}
		bio_put(sbio->bio);
	}

346
	for (i = 0; i < SCRUB_BIOS_PER_SCTX; ++i) {
347
		struct scrub_bio *sbio = sctx->bios[i];
A
Arne Jansen 已提交
348 349 350 351 352 353

		if (!sbio)
			break;
		kfree(sbio);
	}

354 355
	scrub_free_csums(sctx);
	kfree(sctx);
A
Arne Jansen 已提交
356 357 358
}

static noinline_for_stack
359
struct scrub_ctx *scrub_setup_ctx(struct btrfs_device *dev, int is_dev_replace)
A
Arne Jansen 已提交
360
{
361
	struct scrub_ctx *sctx;
A
Arne Jansen 已提交
362 363
	int		i;
	struct btrfs_fs_info *fs_info = dev->dev_root->fs_info;
364 365
	int pages_per_rd_bio;
	int ret;
A
Arne Jansen 已提交
366

367 368 369 370 371 372 373 374 375 376 377 378
	/*
	 * the setting of pages_per_rd_bio is correct for scrub but might
	 * be wrong for the dev_replace code where we might read from
	 * different devices in the initial huge bios. However, that
	 * code is able to correctly handle the case when adding a page
	 * to a bio fails.
	 */
	if (dev->bdev)
		pages_per_rd_bio = min_t(int, SCRUB_PAGES_PER_RD_BIO,
					 bio_get_nr_vecs(dev->bdev));
	else
		pages_per_rd_bio = SCRUB_PAGES_PER_RD_BIO;
379 380
	sctx = kzalloc(sizeof(*sctx), GFP_NOFS);
	if (!sctx)
A
Arne Jansen 已提交
381
		goto nomem;
382
	sctx->is_dev_replace = is_dev_replace;
383
	sctx->pages_per_rd_bio = pages_per_rd_bio;
384
	sctx->curr = -1;
385
	sctx->dev_root = dev->dev_root;
386
	for (i = 0; i < SCRUB_BIOS_PER_SCTX; ++i) {
A
Arne Jansen 已提交
387 388 389 390 391
		struct scrub_bio *sbio;

		sbio = kzalloc(sizeof(*sbio), GFP_NOFS);
		if (!sbio)
			goto nomem;
392
		sctx->bios[i] = sbio;
A
Arne Jansen 已提交
393 394

		sbio->index = i;
395
		sbio->sctx = sctx;
396 397
		sbio->page_count = 0;
		sbio->work.func = scrub_bio_end_io_worker;
A
Arne Jansen 已提交
398

399
		if (i != SCRUB_BIOS_PER_SCTX - 1)
400
			sctx->bios[i]->next_free = i + 1;
401
		else
402 403 404 405 406 407
			sctx->bios[i]->next_free = -1;
	}
	sctx->first_free = 0;
	sctx->nodesize = dev->dev_root->nodesize;
	sctx->leafsize = dev->dev_root->leafsize;
	sctx->sectorsize = dev->dev_root->sectorsize;
408 409
	atomic_set(&sctx->bios_in_flight, 0);
	atomic_set(&sctx->workers_pending, 0);
410 411 412 413 414 415 416
	atomic_set(&sctx->cancel_req, 0);
	sctx->csum_size = btrfs_super_csum_size(fs_info->super_copy);
	INIT_LIST_HEAD(&sctx->csum_list);

	spin_lock_init(&sctx->list_lock);
	spin_lock_init(&sctx->stat_lock);
	init_waitqueue_head(&sctx->list_wait);
417 418 419 420 421 422 423

	ret = scrub_setup_wr_ctx(sctx, &sctx->wr_ctx, fs_info,
				 fs_info->dev_replace.tgtdev, is_dev_replace);
	if (ret) {
		scrub_free_ctx(sctx);
		return ERR_PTR(ret);
	}
424
	return sctx;
A
Arne Jansen 已提交
425 426

nomem:
427
	scrub_free_ctx(sctx);
A
Arne Jansen 已提交
428 429 430
	return ERR_PTR(-ENOMEM);
}

431 432
static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root,
				     void *warn_ctx)
433 434 435 436 437 438 439
{
	u64 isize;
	u32 nlink;
	int ret;
	int i;
	struct extent_buffer *eb;
	struct btrfs_inode_item *inode_item;
440
	struct scrub_warning *swarn = warn_ctx;
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
	struct btrfs_fs_info *fs_info = swarn->dev->dev_root->fs_info;
	struct inode_fs_paths *ipath = NULL;
	struct btrfs_root *local_root;
	struct btrfs_key root_key;

	root_key.objectid = root;
	root_key.type = BTRFS_ROOT_ITEM_KEY;
	root_key.offset = (u64)-1;
	local_root = btrfs_read_fs_root_no_name(fs_info, &root_key);
	if (IS_ERR(local_root)) {
		ret = PTR_ERR(local_root);
		goto err;
	}

	ret = inode_item_info(inum, 0, local_root, swarn->path);
	if (ret) {
		btrfs_release_path(swarn->path);
		goto err;
	}

	eb = swarn->path->nodes[0];
	inode_item = btrfs_item_ptr(eb, swarn->path->slots[0],
					struct btrfs_inode_item);
	isize = btrfs_inode_size(eb, inode_item);
	nlink = btrfs_inode_nlink(eb, inode_item);
	btrfs_release_path(swarn->path);

	ipath = init_ipath(4096, local_root, swarn->path);
469 470 471 472 473
	if (IS_ERR(ipath)) {
		ret = PTR_ERR(ipath);
		ipath = NULL;
		goto err;
	}
474 475 476 477 478 479 480 481 482 483
	ret = paths_from_inode(inum, ipath);

	if (ret < 0)
		goto err;

	/*
	 * we deliberately ignore the bit ipath might have been too small to
	 * hold all of the paths here
	 */
	for (i = 0; i < ipath->fspath->elem_cnt; ++i)
484
		printk_in_rcu(KERN_WARNING "btrfs: %s at logical %llu on dev "
485 486
			"%s, sector %llu, root %llu, inode %llu, offset %llu, "
			"length %llu, links %u (path: %s)\n", swarn->errstr,
487
			swarn->logical, rcu_str_deref(swarn->dev->name),
488 489
			(unsigned long long)swarn->sector, root, inum, offset,
			min(isize - offset, (u64)PAGE_SIZE), nlink,
490
			(char *)(unsigned long)ipath->fspath->val[i]);
491 492 493 494 495

	free_ipath(ipath);
	return 0;

err:
496
	printk_in_rcu(KERN_WARNING "btrfs: %s at logical %llu on dev "
497 498
		"%s, sector %llu, root %llu, inode %llu, offset %llu: path "
		"resolving failed with ret=%d\n", swarn->errstr,
499
		swarn->logical, rcu_str_deref(swarn->dev->name),
500 501 502 503 504 505
		(unsigned long long)swarn->sector, root, inum, offset, ret);

	free_ipath(ipath);
	return 0;
}

506
static void scrub_print_warning(const char *errstr, struct scrub_block *sblock)
507
{
508 509
	struct btrfs_device *dev;
	struct btrfs_fs_info *fs_info;
510 511 512 513 514
	struct btrfs_path *path;
	struct btrfs_key found_key;
	struct extent_buffer *eb;
	struct btrfs_extent_item *ei;
	struct scrub_warning swarn;
515 516 517
	unsigned long ptr = 0;
	u64 extent_item_pos;
	u64 flags = 0;
518
	u64 ref_root;
519
	u32 item_size;
520 521
	u8 ref_level;
	const int bufsize = 4096;
522
	int ret;
523

524
	WARN_ON(sblock->page_count < 1);
525
	dev = sblock->pagev[0]->dev;
526 527
	fs_info = sblock->sctx->dev_root->fs_info;

528 529 530 531
	path = btrfs_alloc_path();

	swarn.scratch_buf = kmalloc(bufsize, GFP_NOFS);
	swarn.msg_buf = kmalloc(bufsize, GFP_NOFS);
532 533
	swarn.sector = (sblock->pagev[0]->physical) >> 9;
	swarn.logical = sblock->pagev[0]->logical;
534
	swarn.errstr = errstr;
535
	swarn.dev = NULL;
536 537 538 539 540 541
	swarn.msg_bufsize = bufsize;
	swarn.scratch_bufsize = bufsize;

	if (!path || !swarn.scratch_buf || !swarn.msg_buf)
		goto out;

542 543
	ret = extent_from_logical(fs_info, swarn.logical, path, &found_key,
				  &flags);
544 545 546
	if (ret < 0)
		goto out;

J
Jan Schmidt 已提交
547
	extent_item_pos = swarn.logical - found_key.objectid;
548 549 550 551 552 553
	swarn.extent_item_size = found_key.offset;

	eb = path->nodes[0];
	ei = btrfs_item_ptr(eb, path->slots[0], struct btrfs_extent_item);
	item_size = btrfs_item_size_nr(eb, path->slots[0]);

554
	if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) {
555 556 557
		do {
			ret = tree_backref_for_extent(&ptr, eb, ei, item_size,
							&ref_root, &ref_level);
558
			printk_in_rcu(KERN_WARNING
S
Stefan Behrens 已提交
559
				"btrfs: %s at logical %llu on dev %s, "
560
				"sector %llu: metadata %s (level %d) in tree "
561 562
				"%llu\n", errstr, swarn.logical,
				rcu_str_deref(dev->name),
563 564 565 566 567
				(unsigned long long)swarn.sector,
				ref_level ? "node" : "leaf",
				ret < 0 ? -1 : ref_level,
				ret < 0 ? -1 : ref_root);
		} while (ret != 1);
568
		btrfs_release_path(path);
569
	} else {
570
		btrfs_release_path(path);
571
		swarn.path = path;
572
		swarn.dev = dev;
573 574
		iterate_extent_inodes(fs_info, found_key.objectid,
					extent_item_pos, 1,
575 576 577 578 579 580 581 582 583
					scrub_print_warning_inode, &swarn);
	}

out:
	btrfs_free_path(path);
	kfree(swarn.scratch_buf);
	kfree(swarn.msg_buf);
}

584
static int scrub_fixup_readpage(u64 inum, u64 offset, u64 root, void *fixup_ctx)
585
{
586
	struct page *page = NULL;
587
	unsigned long index;
588
	struct scrub_fixup_nodatasum *fixup = fixup_ctx;
589
	int ret;
590
	int corrected = 0;
591
	struct btrfs_key key;
592
	struct inode *inode = NULL;
593
	struct btrfs_fs_info *fs_info;
594 595
	u64 end = offset + PAGE_SIZE - 1;
	struct btrfs_root *local_root;
596
	int srcu_index;
597 598 599 600

	key.objectid = root;
	key.type = BTRFS_ROOT_ITEM_KEY;
	key.offset = (u64)-1;
601 602 603 604 605 606 607

	fs_info = fixup->root->fs_info;
	srcu_index = srcu_read_lock(&fs_info->subvol_srcu);

	local_root = btrfs_read_fs_root_no_name(fs_info, &key);
	if (IS_ERR(local_root)) {
		srcu_read_unlock(&fs_info->subvol_srcu, srcu_index);
608
		return PTR_ERR(local_root);
609
	}
610 611 612 613

	key.type = BTRFS_INODE_ITEM_KEY;
	key.objectid = inum;
	key.offset = 0;
614 615
	inode = btrfs_iget(fs_info->sb, &key, local_root, NULL);
	srcu_read_unlock(&fs_info->subvol_srcu, srcu_index);
616 617 618 619 620 621
	if (IS_ERR(inode))
		return PTR_ERR(inode);

	index = offset >> PAGE_CACHE_SHIFT;

	page = find_or_create_page(inode->i_mapping, index, GFP_NOFS);
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
	if (!page) {
		ret = -ENOMEM;
		goto out;
	}

	if (PageUptodate(page)) {
		if (PageDirty(page)) {
			/*
			 * we need to write the data to the defect sector. the
			 * data that was in that sector is not in memory,
			 * because the page was modified. we must not write the
			 * modified page to that sector.
			 *
			 * TODO: what could be done here: wait for the delalloc
			 *       runner to write out that page (might involve
			 *       COW) and see whether the sector is still
			 *       referenced afterwards.
			 *
			 * For the meantime, we'll treat this error
			 * incorrectable, although there is a chance that a
			 * later scrub will find the bad sector again and that
			 * there's no dirty page in memory, then.
			 */
			ret = -EIO;
			goto out;
		}
648 649
		fs_info = BTRFS_I(inode)->root->fs_info;
		ret = repair_io_failure(fs_info, offset, PAGE_SIZE,
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
					fixup->logical, page,
					fixup->mirror_num);
		unlock_page(page);
		corrected = !ret;
	} else {
		/*
		 * we need to get good data first. the general readpage path
		 * will call repair_io_failure for us, we just have to make
		 * sure we read the bad mirror.
		 */
		ret = set_extent_bits(&BTRFS_I(inode)->io_tree, offset, end,
					EXTENT_DAMAGED, GFP_NOFS);
		if (ret) {
			/* set_extent_bits should give proper error */
			WARN_ON(ret > 0);
			if (ret > 0)
				ret = -EFAULT;
			goto out;
		}

		ret = extent_read_full_page(&BTRFS_I(inode)->io_tree, page,
						btrfs_get_extent,
						fixup->mirror_num);
		wait_on_page_locked(page);

		corrected = !test_range_bit(&BTRFS_I(inode)->io_tree, offset,
						end, EXTENT_DAMAGED, 0, NULL);
		if (!corrected)
			clear_extent_bits(&BTRFS_I(inode)->io_tree, offset, end,
						EXTENT_DAMAGED, GFP_NOFS);
	}

out:
	if (page)
		put_page(page);
	if (inode)
		iput(inode);
687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705

	if (ret < 0)
		return ret;

	if (ret == 0 && corrected) {
		/*
		 * we only need to call readpage for one of the inodes belonging
		 * to this extent. so make iterate_extent_inodes stop
		 */
		return 1;
	}

	return -EIO;
}

static void scrub_fixup_nodatasum(struct btrfs_work *work)
{
	int ret;
	struct scrub_fixup_nodatasum *fixup;
706
	struct scrub_ctx *sctx;
707 708 709 710 711 712
	struct btrfs_trans_handle *trans = NULL;
	struct btrfs_fs_info *fs_info;
	struct btrfs_path *path;
	int uncorrectable = 0;

	fixup = container_of(work, struct scrub_fixup_nodatasum, work);
713
	sctx = fixup->sctx;
714 715 716 717
	fs_info = fixup->root->fs_info;

	path = btrfs_alloc_path();
	if (!path) {
718 719 720
		spin_lock(&sctx->stat_lock);
		++sctx->stat.malloc_errors;
		spin_unlock(&sctx->stat_lock);
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
		uncorrectable = 1;
		goto out;
	}

	trans = btrfs_join_transaction(fixup->root);
	if (IS_ERR(trans)) {
		uncorrectable = 1;
		goto out;
	}

	/*
	 * the idea is to trigger a regular read through the standard path. we
	 * read a page from the (failed) logical address by specifying the
	 * corresponding copynum of the failed sector. thus, that readpage is
	 * expected to fail.
	 * that is the point where on-the-fly error correction will kick in
	 * (once it's finished) and rewrite the failed sector if a good copy
	 * can be found.
	 */
	ret = iterate_inodes_from_logical(fixup->logical, fixup->root->fs_info,
						path, scrub_fixup_readpage,
						fixup);
	if (ret < 0) {
		uncorrectable = 1;
		goto out;
	}
	WARN_ON(ret != 1);

749 750 751
	spin_lock(&sctx->stat_lock);
	++sctx->stat.corrected_errors;
	spin_unlock(&sctx->stat_lock);
752 753 754 755 756

out:
	if (trans && !IS_ERR(trans))
		btrfs_end_transaction(trans, fixup->root);
	if (uncorrectable) {
757 758 759
		spin_lock(&sctx->stat_lock);
		++sctx->stat.uncorrectable_errors;
		spin_unlock(&sctx->stat_lock);
760 761 762
		btrfs_dev_replace_stats_inc(
			&sctx->dev_root->fs_info->dev_replace.
			num_uncorrectable_read_errors);
763
		printk_ratelimited_in_rcu(KERN_ERR
764
			"btrfs: unable to fixup (nodatasum) error at logical %llu on dev %s\n",
765
			fixup->logical, rcu_str_deref(fixup->dev->name));
766 767 768 769 770
	}

	btrfs_free_path(path);
	kfree(fixup);

771
	scrub_pending_trans_workers_dec(sctx);
772 773
}

A
Arne Jansen 已提交
774
/*
775 776 777 778 779 780
 * scrub_handle_errored_block gets called when either verification of the
 * pages failed or the bio failed to read, e.g. with EIO. In the latter
 * case, this function handles all pages in the bio, even though only one
 * may be bad.
 * The goal of this function is to repair the errored block by using the
 * contents of one of the mirrors.
A
Arne Jansen 已提交
781
 */
782
static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
A
Arne Jansen 已提交
783
{
784
	struct scrub_ctx *sctx = sblock_to_check->sctx;
785
	struct btrfs_device *dev;
786 787 788 789 790 791 792 793 794 795 796 797 798 799
	struct btrfs_fs_info *fs_info;
	u64 length;
	u64 logical;
	u64 generation;
	unsigned int failed_mirror_index;
	unsigned int is_metadata;
	unsigned int have_csum;
	u8 *csum;
	struct scrub_block *sblocks_for_recheck; /* holds one for each mirror */
	struct scrub_block *sblock_bad;
	int ret;
	int mirror_index;
	int page_num;
	int success;
800
	static DEFINE_RATELIMIT_STATE(_rs, DEFAULT_RATELIMIT_INTERVAL,
801 802 803
				      DEFAULT_RATELIMIT_BURST);

	BUG_ON(sblock_to_check->page_count < 1);
804
	fs_info = sctx->dev_root->fs_info;
805 806 807 808 809 810 811 812 813 814 815
	if (sblock_to_check->pagev[0]->flags & BTRFS_EXTENT_FLAG_SUPER) {
		/*
		 * if we find an error in a super block, we just report it.
		 * They will get written with the next transaction commit
		 * anyway
		 */
		spin_lock(&sctx->stat_lock);
		++sctx->stat.super_errors;
		spin_unlock(&sctx->stat_lock);
		return 0;
	}
816
	length = sblock_to_check->page_count * PAGE_SIZE;
817 818 819 820 821
	logical = sblock_to_check->pagev[0]->logical;
	generation = sblock_to_check->pagev[0]->generation;
	BUG_ON(sblock_to_check->pagev[0]->mirror_num < 1);
	failed_mirror_index = sblock_to_check->pagev[0]->mirror_num - 1;
	is_metadata = !(sblock_to_check->pagev[0]->flags &
822
			BTRFS_EXTENT_FLAG_DATA);
823 824 825
	have_csum = sblock_to_check->pagev[0]->have_csum;
	csum = sblock_to_check->pagev[0]->csum;
	dev = sblock_to_check->pagev[0]->dev;
826

827 828 829 830 831
	if (sctx->is_dev_replace && !is_metadata && !have_csum) {
		sblocks_for_recheck = NULL;
		goto nodatasum_case;
	}

832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864
	/*
	 * read all mirrors one after the other. This includes to
	 * re-read the extent or metadata block that failed (that was
	 * the cause that this fixup code is called) another time,
	 * page by page this time in order to know which pages
	 * caused I/O errors and which ones are good (for all mirrors).
	 * It is the goal to handle the situation when more than one
	 * mirror contains I/O errors, but the errors do not
	 * overlap, i.e. the data can be repaired by selecting the
	 * pages from those mirrors without I/O error on the
	 * particular pages. One example (with blocks >= 2 * PAGE_SIZE)
	 * would be that mirror #1 has an I/O error on the first page,
	 * the second page is good, and mirror #2 has an I/O error on
	 * the second page, but the first page is good.
	 * Then the first page of the first mirror can be repaired by
	 * taking the first page of the second mirror, and the
	 * second page of the second mirror can be repaired by
	 * copying the contents of the 2nd page of the 1st mirror.
	 * One more note: if the pages of one mirror contain I/O
	 * errors, the checksum cannot be verified. In order to get
	 * the best data for repairing, the first attempt is to find
	 * a mirror without I/O errors and with a validated checksum.
	 * Only if this is not possible, the pages are picked from
	 * mirrors with I/O errors without considering the checksum.
	 * If the latter is the case, at the end, the checksum of the
	 * repaired area is verified in order to correctly maintain
	 * the statistics.
	 */

	sblocks_for_recheck = kzalloc(BTRFS_MAX_MIRRORS *
				     sizeof(*sblocks_for_recheck),
				     GFP_NOFS);
	if (!sblocks_for_recheck) {
865 866 867 868 869
		spin_lock(&sctx->stat_lock);
		sctx->stat.malloc_errors++;
		sctx->stat.read_errors++;
		sctx->stat.uncorrectable_errors++;
		spin_unlock(&sctx->stat_lock);
870
		btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_READ_ERRS);
871
		goto out;
A
Arne Jansen 已提交
872 873
	}

874
	/* setup the context, map the logical blocks and alloc the pages */
875
	ret = scrub_setup_recheck_block(sctx, fs_info, sblock_to_check, length,
876 877
					logical, sblocks_for_recheck);
	if (ret) {
878 879 880 881
		spin_lock(&sctx->stat_lock);
		sctx->stat.read_errors++;
		sctx->stat.uncorrectable_errors++;
		spin_unlock(&sctx->stat_lock);
882
		btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_READ_ERRS);
883 884 885 886
		goto out;
	}
	BUG_ON(failed_mirror_index >= BTRFS_MAX_MIRRORS);
	sblock_bad = sblocks_for_recheck + failed_mirror_index;
887

888
	/* build and submit the bios for the failed mirror, check checksums */
889 890
	scrub_recheck_block(fs_info, sblock_bad, is_metadata, have_csum,
			    csum, generation, sctx->csum_size);
A
Arne Jansen 已提交
891

892 893 894 895 896 897 898 899 900 901
	if (!sblock_bad->header_error && !sblock_bad->checksum_error &&
	    sblock_bad->no_io_error_seen) {
		/*
		 * the error disappeared after reading page by page, or
		 * the area was part of a huge bio and other parts of the
		 * bio caused I/O errors, or the block layer merged several
		 * read requests into one and the error is caused by a
		 * different bio (usually one of the two latter cases is
		 * the cause)
		 */
902 903 904
		spin_lock(&sctx->stat_lock);
		sctx->stat.unverified_errors++;
		spin_unlock(&sctx->stat_lock);
A
Arne Jansen 已提交
905

906 907
		if (sctx->is_dev_replace)
			scrub_write_block_to_dev_replace(sblock_bad);
908
		goto out;
A
Arne Jansen 已提交
909 910
	}

911
	if (!sblock_bad->no_io_error_seen) {
912 913 914
		spin_lock(&sctx->stat_lock);
		sctx->stat.read_errors++;
		spin_unlock(&sctx->stat_lock);
915 916
		if (__ratelimit(&_rs))
			scrub_print_warning("i/o error", sblock_to_check);
917
		btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_READ_ERRS);
918
	} else if (sblock_bad->checksum_error) {
919 920 921
		spin_lock(&sctx->stat_lock);
		sctx->stat.csum_errors++;
		spin_unlock(&sctx->stat_lock);
922 923
		if (__ratelimit(&_rs))
			scrub_print_warning("checksum error", sblock_to_check);
924
		btrfs_dev_stat_inc_and_print(dev,
925
					     BTRFS_DEV_STAT_CORRUPTION_ERRS);
926
	} else if (sblock_bad->header_error) {
927 928 929
		spin_lock(&sctx->stat_lock);
		sctx->stat.verify_errors++;
		spin_unlock(&sctx->stat_lock);
930 931 932
		if (__ratelimit(&_rs))
			scrub_print_warning("checksum/header error",
					    sblock_to_check);
933
		if (sblock_bad->generation_error)
934
			btrfs_dev_stat_inc_and_print(dev,
935 936
				BTRFS_DEV_STAT_GENERATION_ERRS);
		else
937
			btrfs_dev_stat_inc_and_print(dev,
938
				BTRFS_DEV_STAT_CORRUPTION_ERRS);
939
	}
A
Arne Jansen 已提交
940

941
	if (sctx->readonly && !sctx->is_dev_replace)
942
		goto did_not_correct_error;
A
Arne Jansen 已提交
943

944 945
	if (!is_metadata && !have_csum) {
		struct scrub_fixup_nodatasum *fixup_nodatasum;
A
Arne Jansen 已提交
946

947 948 949
nodatasum_case:
		WARN_ON(sctx->is_dev_replace);

950 951 952 953 954 955 956 957 958 959
		/*
		 * !is_metadata and !have_csum, this means that the data
		 * might not be COW'ed, that it might be modified
		 * concurrently. The general strategy to work on the
		 * commit root does not help in the case when COW is not
		 * used.
		 */
		fixup_nodatasum = kzalloc(sizeof(*fixup_nodatasum), GFP_NOFS);
		if (!fixup_nodatasum)
			goto did_not_correct_error;
960
		fixup_nodatasum->sctx = sctx;
961
		fixup_nodatasum->dev = dev;
962 963 964
		fixup_nodatasum->logical = logical;
		fixup_nodatasum->root = fs_info->extent_root;
		fixup_nodatasum->mirror_num = failed_mirror_index + 1;
965
		scrub_pending_trans_workers_inc(sctx);
966 967 968 969
		fixup_nodatasum->work.func = scrub_fixup_nodatasum;
		btrfs_queue_worker(&fs_info->scrub_workers,
				   &fixup_nodatasum->work);
		goto out;
A
Arne Jansen 已提交
970 971
	}

972 973
	/*
	 * now build and submit the bios for the other mirrors, check
974 975
	 * checksums.
	 * First try to pick the mirror which is completely without I/O
976 977 978 979 980 981 982 983 984 985 986 987 988 989 990
	 * errors and also does not have a checksum error.
	 * If one is found, and if a checksum is present, the full block
	 * that is known to contain an error is rewritten. Afterwards
	 * the block is known to be corrected.
	 * If a mirror is found which is completely correct, and no
	 * checksum is present, only those pages are rewritten that had
	 * an I/O error in the block to be repaired, since it cannot be
	 * determined, which copy of the other pages is better (and it
	 * could happen otherwise that a correct page would be
	 * overwritten by a bad one).
	 */
	for (mirror_index = 0;
	     mirror_index < BTRFS_MAX_MIRRORS &&
	     sblocks_for_recheck[mirror_index].page_count > 0;
	     mirror_index++) {
991
		struct scrub_block *sblock_other;
992

993 994 995 996 997
		if (mirror_index == failed_mirror_index)
			continue;
		sblock_other = sblocks_for_recheck + mirror_index;

		/* build and submit the bios, check checksums */
998 999 1000 1001 1002
		scrub_recheck_block(fs_info, sblock_other, is_metadata,
				    have_csum, csum, generation,
				    sctx->csum_size);

		if (!sblock_other->header_error &&
1003 1004
		    !sblock_other->checksum_error &&
		    sblock_other->no_io_error_seen) {
1005 1006 1007 1008 1009 1010 1011 1012 1013
			if (sctx->is_dev_replace) {
				scrub_write_block_to_dev_replace(sblock_other);
			} else {
				int force_write = is_metadata || have_csum;

				ret = scrub_repair_block_from_good_copy(
						sblock_bad, sblock_other,
						force_write);
			}
1014 1015 1016 1017
			if (0 == ret)
				goto corrected_error;
		}
	}
A
Arne Jansen 已提交
1018 1019

	/*
1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 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
	 * for dev_replace, pick good pages and write to the target device.
	 */
	if (sctx->is_dev_replace) {
		success = 1;
		for (page_num = 0; page_num < sblock_bad->page_count;
		     page_num++) {
			int sub_success;

			sub_success = 0;
			for (mirror_index = 0;
			     mirror_index < BTRFS_MAX_MIRRORS &&
			     sblocks_for_recheck[mirror_index].page_count > 0;
			     mirror_index++) {
				struct scrub_block *sblock_other =
					sblocks_for_recheck + mirror_index;
				struct scrub_page *page_other =
					sblock_other->pagev[page_num];

				if (!page_other->io_error) {
					ret = scrub_write_page_to_dev_replace(
							sblock_other, page_num);
					if (ret == 0) {
						/* succeeded for this page */
						sub_success = 1;
						break;
					} else {
						btrfs_dev_replace_stats_inc(
							&sctx->dev_root->
							fs_info->dev_replace.
							num_write_errors);
					}
				}
			}

			if (!sub_success) {
				/*
				 * did not find a mirror to fetch the page
				 * from. scrub_write_page_to_dev_replace()
				 * handles this case (page->io_error), by
				 * filling the block with zeros before
				 * submitting the write request
				 */
				success = 0;
				ret = scrub_write_page_to_dev_replace(
						sblock_bad, page_num);
				if (ret)
					btrfs_dev_replace_stats_inc(
						&sctx->dev_root->fs_info->
						dev_replace.num_write_errors);
			}
		}

		goto out;
	}

	/*
	 * for regular scrub, repair those pages that are errored.
	 * In case of I/O errors in the area that is supposed to be
1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098
	 * repaired, continue by picking good copies of those pages.
	 * Select the good pages from mirrors to rewrite bad pages from
	 * the area to fix. Afterwards verify the checksum of the block
	 * that is supposed to be repaired. This verification step is
	 * only done for the purpose of statistic counting and for the
	 * final scrub report, whether errors remain.
	 * A perfect algorithm could make use of the checksum and try
	 * all possible combinations of pages from the different mirrors
	 * until the checksum verification succeeds. For example, when
	 * the 2nd page of mirror #1 faces I/O errors, and the 2nd page
	 * of mirror #2 is readable but the final checksum test fails,
	 * then the 2nd page of mirror #3 could be tried, whether now
	 * the final checksum succeedes. But this would be a rare
	 * exception and is therefore not implemented. At least it is
	 * avoided that the good copy is overwritten.
	 * A more useful improvement would be to pick the sectors
	 * without I/O error based on sector sizes (512 bytes on legacy
	 * disks) instead of on PAGE_SIZE. Then maybe 512 byte of one
	 * mirror could be repaired by taking 512 byte of a different
	 * mirror, even if other 512 byte sectors in the same PAGE_SIZE
	 * area are unreadable.
A
Arne Jansen 已提交
1099 1100
	 */

1101 1102 1103 1104 1105 1106
	/* can only fix I/O errors from here on */
	if (sblock_bad->no_io_error_seen)
		goto did_not_correct_error;

	success = 1;
	for (page_num = 0; page_num < sblock_bad->page_count; page_num++) {
1107
		struct scrub_page *page_bad = sblock_bad->pagev[page_num];
1108 1109

		if (!page_bad->io_error)
A
Arne Jansen 已提交
1110
			continue;
1111 1112 1113 1114 1115 1116 1117

		for (mirror_index = 0;
		     mirror_index < BTRFS_MAX_MIRRORS &&
		     sblocks_for_recheck[mirror_index].page_count > 0;
		     mirror_index++) {
			struct scrub_block *sblock_other = sblocks_for_recheck +
							   mirror_index;
1118 1119
			struct scrub_page *page_other = sblock_other->pagev[
							page_num];
1120 1121 1122 1123 1124 1125 1126 1127 1128

			if (!page_other->io_error) {
				ret = scrub_repair_page_from_good_copy(
					sblock_bad, sblock_other, page_num, 0);
				if (0 == ret) {
					page_bad->io_error = 0;
					break; /* succeeded for this page */
				}
			}
I
Ilya Dryomov 已提交
1129
		}
A
Arne Jansen 已提交
1130

1131 1132 1133 1134
		if (page_bad->io_error) {
			/* did not find a mirror to copy the page from */
			success = 0;
		}
A
Arne Jansen 已提交
1135 1136
	}

1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147
	if (success) {
		if (is_metadata || have_csum) {
			/*
			 * need to verify the checksum now that all
			 * sectors on disk are repaired (the write
			 * request for data to be repaired is on its way).
			 * Just be lazy and use scrub_recheck_block()
			 * which re-reads the data before the checksum
			 * is verified, but most likely the data comes out
			 * of the page cache.
			 */
1148 1149 1150 1151
			scrub_recheck_block(fs_info, sblock_bad,
					    is_metadata, have_csum, csum,
					    generation, sctx->csum_size);
			if (!sblock_bad->header_error &&
1152 1153 1154 1155 1156 1157 1158
			    !sblock_bad->checksum_error &&
			    sblock_bad->no_io_error_seen)
				goto corrected_error;
			else
				goto did_not_correct_error;
		} else {
corrected_error:
1159 1160 1161
			spin_lock(&sctx->stat_lock);
			sctx->stat.corrected_errors++;
			spin_unlock(&sctx->stat_lock);
1162
			printk_ratelimited_in_rcu(KERN_ERR
1163
				"btrfs: fixed up error at logical %llu on dev %s\n",
1164
				logical, rcu_str_deref(dev->name));
A
Arne Jansen 已提交
1165
		}
1166 1167
	} else {
did_not_correct_error:
1168 1169 1170
		spin_lock(&sctx->stat_lock);
		sctx->stat.uncorrectable_errors++;
		spin_unlock(&sctx->stat_lock);
1171
		printk_ratelimited_in_rcu(KERN_ERR
1172
			"btrfs: unable to fixup (regular) error at logical %llu on dev %s\n",
1173
			logical, rcu_str_deref(dev->name));
I
Ilya Dryomov 已提交
1174
	}
A
Arne Jansen 已提交
1175

1176 1177 1178 1179 1180 1181 1182 1183
out:
	if (sblocks_for_recheck) {
		for (mirror_index = 0; mirror_index < BTRFS_MAX_MIRRORS;
		     mirror_index++) {
			struct scrub_block *sblock = sblocks_for_recheck +
						     mirror_index;
			int page_index;

1184 1185 1186 1187 1188
			for (page_index = 0; page_index < sblock->page_count;
			     page_index++) {
				sblock->pagev[page_index]->sblock = NULL;
				scrub_page_put(sblock->pagev[page_index]);
			}
1189 1190 1191
		}
		kfree(sblocks_for_recheck);
	}
A
Arne Jansen 已提交
1192

1193 1194
	return 0;
}
A
Arne Jansen 已提交
1195

1196
static int scrub_setup_recheck_block(struct scrub_ctx *sctx,
1197
				     struct btrfs_fs_info *fs_info,
1198
				     struct scrub_block *original_sblock,
1199 1200 1201 1202 1203 1204 1205 1206
				     u64 length, u64 logical,
				     struct scrub_block *sblocks_for_recheck)
{
	int page_index;
	int mirror_index;
	int ret;

	/*
1207
	 * note: the two members ref_count and outstanding_pages
1208 1209 1210 1211 1212 1213 1214 1215 1216
	 * are not used (and not set) in the blocks that are used for
	 * the recheck procedure
	 */

	page_index = 0;
	while (length > 0) {
		u64 sublen = min_t(u64, length, PAGE_SIZE);
		u64 mapped_length = sublen;
		struct btrfs_bio *bbio = NULL;
A
Arne Jansen 已提交
1217

1218 1219 1220 1221
		/*
		 * with a length of PAGE_SIZE, each returned stripe
		 * represents one mirror
		 */
1222 1223
		ret = btrfs_map_block(fs_info, REQ_GET_READ_MIRRORS, logical,
				      &mapped_length, &bbio, 0);
1224 1225 1226 1227
		if (ret || !bbio || mapped_length < sublen) {
			kfree(bbio);
			return -EIO;
		}
A
Arne Jansen 已提交
1228

1229
		BUG_ON(page_index >= SCRUB_PAGES_PER_RD_BIO);
1230 1231 1232 1233 1234 1235 1236 1237 1238
		for (mirror_index = 0; mirror_index < (int)bbio->num_stripes;
		     mirror_index++) {
			struct scrub_block *sblock;
			struct scrub_page *page;

			if (mirror_index >= BTRFS_MAX_MIRRORS)
				continue;

			sblock = sblocks_for_recheck + mirror_index;
1239 1240 1241 1242
			sblock->sctx = sctx;
			page = kzalloc(sizeof(*page), GFP_NOFS);
			if (!page) {
leave_nomem:
1243 1244 1245
				spin_lock(&sctx->stat_lock);
				sctx->stat.malloc_errors++;
				spin_unlock(&sctx->stat_lock);
1246
				kfree(bbio);
1247 1248
				return -ENOMEM;
			}
1249 1250 1251 1252
			scrub_page_get(page);
			sblock->pagev[page_index] = page;
			page->logical = logical;
			page->physical = bbio->stripes[mirror_index].physical;
1253 1254 1255 1256
			BUG_ON(page_index >= original_sblock->page_count);
			page->physical_for_dev_replace =
				original_sblock->pagev[page_index]->
				physical_for_dev_replace;
1257 1258 1259
			/* for missing devices, dev->bdev is NULL */
			page->dev = bbio->stripes[mirror_index].dev;
			page->mirror_num = mirror_index + 1;
1260
			sblock->page_count++;
1261 1262 1263
			page->page = alloc_page(GFP_NOFS);
			if (!page->page)
				goto leave_nomem;
1264 1265 1266 1267 1268 1269 1270 1271
		}
		kfree(bbio);
		length -= sublen;
		logical += sublen;
		page_index++;
	}

	return 0;
I
Ilya Dryomov 已提交
1272 1273
}

1274 1275 1276 1277 1278 1279 1280
/*
 * this function will check the on disk data for checksum errors, header
 * errors and read I/O errors. If any I/O errors happen, the exact pages
 * which are errored are marked as being bad. The goal is to enable scrub
 * to take those pages that are not errored from all the mirrors so that
 * the pages that are errored in the just handled mirror can be repaired.
 */
1281 1282 1283 1284
static void scrub_recheck_block(struct btrfs_fs_info *fs_info,
				struct scrub_block *sblock, int is_metadata,
				int have_csum, u8 *csum, u64 generation,
				u16 csum_size)
I
Ilya Dryomov 已提交
1285
{
1286
	int page_num;
I
Ilya Dryomov 已提交
1287

1288 1289 1290
	sblock->no_io_error_seen = 1;
	sblock->header_error = 0;
	sblock->checksum_error = 0;
I
Ilya Dryomov 已提交
1291

1292 1293
	for (page_num = 0; page_num < sblock->page_count; page_num++) {
		struct bio *bio;
1294
		struct scrub_page *page = sblock->pagev[page_num];
1295 1296
		DECLARE_COMPLETION_ONSTACK(complete);

1297
		if (page->dev->bdev == NULL) {
1298 1299 1300 1301 1302
			page->io_error = 1;
			sblock->no_io_error_seen = 0;
			continue;
		}

1303
		WARN_ON(!page->page);
1304
		bio = btrfs_io_bio_alloc(GFP_NOFS, 1);
1305 1306 1307 1308 1309
		if (!bio) {
			page->io_error = 1;
			sblock->no_io_error_seen = 0;
			continue;
		}
1310
		bio->bi_bdev = page->dev->bdev;
1311 1312 1313 1314
		bio->bi_sector = page->physical >> 9;
		bio->bi_end_io = scrub_complete_bio_end_io;
		bio->bi_private = &complete;

1315
		bio_add_page(bio, page->page, PAGE_SIZE, 0);
1316
		btrfsic_submit_bio(READ, bio);
I
Ilya Dryomov 已提交
1317

1318 1319
		/* this will also unplug the queue */
		wait_for_completion(&complete);
I
Ilya Dryomov 已提交
1320

1321 1322 1323 1324 1325
		page->io_error = !test_bit(BIO_UPTODATE, &bio->bi_flags);
		if (!test_bit(BIO_UPTODATE, &bio->bi_flags))
			sblock->no_io_error_seen = 0;
		bio_put(bio);
	}
I
Ilya Dryomov 已提交
1326

1327 1328 1329 1330 1331
	if (sblock->no_io_error_seen)
		scrub_recheck_block_checksum(fs_info, sblock, is_metadata,
					     have_csum, csum, generation,
					     csum_size);

1332
	return;
A
Arne Jansen 已提交
1333 1334
}

1335 1336 1337 1338 1339
static void scrub_recheck_block_checksum(struct btrfs_fs_info *fs_info,
					 struct scrub_block *sblock,
					 int is_metadata, int have_csum,
					 const u8 *csum, u64 generation,
					 u16 csum_size)
A
Arne Jansen 已提交
1340
{
1341 1342 1343 1344 1345
	int page_num;
	u8 calculated_csum[BTRFS_CSUM_SIZE];
	u32 crc = ~(u32)0;
	void *mapped_buffer;

1346
	WARN_ON(!sblock->pagev[0]->page);
1347 1348 1349
	if (is_metadata) {
		struct btrfs_header *h;

1350
		mapped_buffer = kmap_atomic(sblock->pagev[0]->page);
1351 1352
		h = (struct btrfs_header *)mapped_buffer;

1353
		if (sblock->pagev[0]->logical != btrfs_stack_header_bytenr(h) ||
1354 1355
		    memcmp(h->fsid, fs_info->fsid, BTRFS_UUID_SIZE) ||
		    memcmp(h->chunk_tree_uuid, fs_info->chunk_tree_uuid,
1356
			   BTRFS_UUID_SIZE)) {
1357
			sblock->header_error = 1;
1358
		} else if (generation != btrfs_stack_header_generation(h)) {
1359 1360 1361
			sblock->header_error = 1;
			sblock->generation_error = 1;
		}
1362 1363 1364 1365
		csum = h->csum;
	} else {
		if (!have_csum)
			return;
A
Arne Jansen 已提交
1366

1367
		mapped_buffer = kmap_atomic(sblock->pagev[0]->page);
1368
	}
A
Arne Jansen 已提交
1369

1370 1371
	for (page_num = 0;;) {
		if (page_num == 0 && is_metadata)
1372
			crc = btrfs_csum_data(
1373 1374 1375
				((u8 *)mapped_buffer) + BTRFS_CSUM_SIZE,
				crc, PAGE_SIZE - BTRFS_CSUM_SIZE);
		else
1376
			crc = btrfs_csum_data(mapped_buffer, crc, PAGE_SIZE);
1377

1378
		kunmap_atomic(mapped_buffer);
1379 1380 1381
		page_num++;
		if (page_num >= sblock->page_count)
			break;
1382
		WARN_ON(!sblock->pagev[page_num]->page);
1383

1384
		mapped_buffer = kmap_atomic(sblock->pagev[page_num]->page);
1385 1386 1387 1388 1389
	}

	btrfs_csum_final(crc, calculated_csum);
	if (memcmp(calculated_csum, csum, csum_size))
		sblock->checksum_error = 1;
A
Arne Jansen 已提交
1390 1391
}

1392
static void scrub_complete_bio_end_io(struct bio *bio, int err)
A
Arne Jansen 已提交
1393
{
1394 1395
	complete((struct completion *)bio->bi_private);
}
A
Arne Jansen 已提交
1396

1397 1398 1399 1400 1401 1402
static int scrub_repair_block_from_good_copy(struct scrub_block *sblock_bad,
					     struct scrub_block *sblock_good,
					     int force_write)
{
	int page_num;
	int ret = 0;
I
Ilya Dryomov 已提交
1403

1404 1405
	for (page_num = 0; page_num < sblock_bad->page_count; page_num++) {
		int ret_sub;
I
Ilya Dryomov 已提交
1406

1407 1408 1409 1410 1411 1412
		ret_sub = scrub_repair_page_from_good_copy(sblock_bad,
							   sblock_good,
							   page_num,
							   force_write);
		if (ret_sub)
			ret = ret_sub;
A
Arne Jansen 已提交
1413
	}
1414 1415 1416 1417 1418 1419 1420 1421

	return ret;
}

static int scrub_repair_page_from_good_copy(struct scrub_block *sblock_bad,
					    struct scrub_block *sblock_good,
					    int page_num, int force_write)
{
1422 1423
	struct scrub_page *page_bad = sblock_bad->pagev[page_num];
	struct scrub_page *page_good = sblock_good->pagev[page_num];
1424

1425 1426
	BUG_ON(page_bad->page == NULL);
	BUG_ON(page_good->page == NULL);
1427 1428 1429 1430 1431 1432
	if (force_write || sblock_bad->header_error ||
	    sblock_bad->checksum_error || page_bad->io_error) {
		struct bio *bio;
		int ret;
		DECLARE_COMPLETION_ONSTACK(complete);

1433 1434 1435 1436 1437 1438
		if (!page_bad->dev->bdev) {
			printk_ratelimited(KERN_WARNING
				"btrfs: scrub_repair_page_from_good_copy(bdev == NULL) is unexpected!\n");
			return -EIO;
		}

1439
		bio = btrfs_io_bio_alloc(GFP_NOFS, 1);
1440 1441
		if (!bio)
			return -EIO;
1442
		bio->bi_bdev = page_bad->dev->bdev;
1443 1444 1445 1446 1447 1448 1449 1450
		bio->bi_sector = page_bad->physical >> 9;
		bio->bi_end_io = scrub_complete_bio_end_io;
		bio->bi_private = &complete;

		ret = bio_add_page(bio, page_good->page, PAGE_SIZE, 0);
		if (PAGE_SIZE != ret) {
			bio_put(bio);
			return -EIO;
1451
		}
1452 1453 1454 1455
		btrfsic_submit_bio(WRITE, bio);

		/* this will also unplug the queue */
		wait_for_completion(&complete);
1456 1457 1458
		if (!bio_flagged(bio, BIO_UPTODATE)) {
			btrfs_dev_stat_inc_and_print(page_bad->dev,
				BTRFS_DEV_STAT_WRITE_ERRS);
1459 1460 1461
			btrfs_dev_replace_stats_inc(
				&sblock_bad->sctx->dev_root->fs_info->
				dev_replace.num_write_errors);
1462 1463 1464
			bio_put(bio);
			return -EIO;
		}
1465
		bio_put(bio);
A
Arne Jansen 已提交
1466 1467
	}

1468 1469 1470
	return 0;
}

1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529
static void scrub_write_block_to_dev_replace(struct scrub_block *sblock)
{
	int page_num;

	for (page_num = 0; page_num < sblock->page_count; page_num++) {
		int ret;

		ret = scrub_write_page_to_dev_replace(sblock, page_num);
		if (ret)
			btrfs_dev_replace_stats_inc(
				&sblock->sctx->dev_root->fs_info->dev_replace.
				num_write_errors);
	}
}

static int scrub_write_page_to_dev_replace(struct scrub_block *sblock,
					   int page_num)
{
	struct scrub_page *spage = sblock->pagev[page_num];

	BUG_ON(spage->page == NULL);
	if (spage->io_error) {
		void *mapped_buffer = kmap_atomic(spage->page);

		memset(mapped_buffer, 0, PAGE_CACHE_SIZE);
		flush_dcache_page(spage->page);
		kunmap_atomic(mapped_buffer);
	}
	return scrub_add_page_to_wr_bio(sblock->sctx, spage);
}

static int scrub_add_page_to_wr_bio(struct scrub_ctx *sctx,
				    struct scrub_page *spage)
{
	struct scrub_wr_ctx *wr_ctx = &sctx->wr_ctx;
	struct scrub_bio *sbio;
	int ret;

	mutex_lock(&wr_ctx->wr_lock);
again:
	if (!wr_ctx->wr_curr_bio) {
		wr_ctx->wr_curr_bio = kzalloc(sizeof(*wr_ctx->wr_curr_bio),
					      GFP_NOFS);
		if (!wr_ctx->wr_curr_bio) {
			mutex_unlock(&wr_ctx->wr_lock);
			return -ENOMEM;
		}
		wr_ctx->wr_curr_bio->sctx = sctx;
		wr_ctx->wr_curr_bio->page_count = 0;
	}
	sbio = wr_ctx->wr_curr_bio;
	if (sbio->page_count == 0) {
		struct bio *bio;

		sbio->physical = spage->physical_for_dev_replace;
		sbio->logical = spage->logical;
		sbio->dev = wr_ctx->tgtdev;
		bio = sbio->bio;
		if (!bio) {
1530
			bio = btrfs_io_bio_alloc(GFP_NOFS, wr_ctx->pages_per_wr_bio);
1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632
			if (!bio) {
				mutex_unlock(&wr_ctx->wr_lock);
				return -ENOMEM;
			}
			sbio->bio = bio;
		}

		bio->bi_private = sbio;
		bio->bi_end_io = scrub_wr_bio_end_io;
		bio->bi_bdev = sbio->dev->bdev;
		bio->bi_sector = sbio->physical >> 9;
		sbio->err = 0;
	} else if (sbio->physical + sbio->page_count * PAGE_SIZE !=
		   spage->physical_for_dev_replace ||
		   sbio->logical + sbio->page_count * PAGE_SIZE !=
		   spage->logical) {
		scrub_wr_submit(sctx);
		goto again;
	}

	ret = bio_add_page(sbio->bio, spage->page, PAGE_SIZE, 0);
	if (ret != PAGE_SIZE) {
		if (sbio->page_count < 1) {
			bio_put(sbio->bio);
			sbio->bio = NULL;
			mutex_unlock(&wr_ctx->wr_lock);
			return -EIO;
		}
		scrub_wr_submit(sctx);
		goto again;
	}

	sbio->pagev[sbio->page_count] = spage;
	scrub_page_get(spage);
	sbio->page_count++;
	if (sbio->page_count == wr_ctx->pages_per_wr_bio)
		scrub_wr_submit(sctx);
	mutex_unlock(&wr_ctx->wr_lock);

	return 0;
}

static void scrub_wr_submit(struct scrub_ctx *sctx)
{
	struct scrub_wr_ctx *wr_ctx = &sctx->wr_ctx;
	struct scrub_bio *sbio;

	if (!wr_ctx->wr_curr_bio)
		return;

	sbio = wr_ctx->wr_curr_bio;
	wr_ctx->wr_curr_bio = NULL;
	WARN_ON(!sbio->bio->bi_bdev);
	scrub_pending_bio_inc(sctx);
	/* process all writes in a single worker thread. Then the block layer
	 * orders the requests before sending them to the driver which
	 * doubled the write performance on spinning disks when measured
	 * with Linux 3.5 */
	btrfsic_submit_bio(WRITE, sbio->bio);
}

static void scrub_wr_bio_end_io(struct bio *bio, int err)
{
	struct scrub_bio *sbio = bio->bi_private;
	struct btrfs_fs_info *fs_info = sbio->dev->dev_root->fs_info;

	sbio->err = err;
	sbio->bio = bio;

	sbio->work.func = scrub_wr_bio_end_io_worker;
	btrfs_queue_worker(&fs_info->scrub_wr_completion_workers, &sbio->work);
}

static void scrub_wr_bio_end_io_worker(struct btrfs_work *work)
{
	struct scrub_bio *sbio = container_of(work, struct scrub_bio, work);
	struct scrub_ctx *sctx = sbio->sctx;
	int i;

	WARN_ON(sbio->page_count > SCRUB_PAGES_PER_WR_BIO);
	if (sbio->err) {
		struct btrfs_dev_replace *dev_replace =
			&sbio->sctx->dev_root->fs_info->dev_replace;

		for (i = 0; i < sbio->page_count; i++) {
			struct scrub_page *spage = sbio->pagev[i];

			spage->io_error = 1;
			btrfs_dev_replace_stats_inc(&dev_replace->
						    num_write_errors);
		}
	}

	for (i = 0; i < sbio->page_count; i++)
		scrub_page_put(sbio->pagev[i]);

	bio_put(sbio->bio);
	kfree(sbio);
	scrub_pending_bio_dec(sctx);
}

static int scrub_checksum(struct scrub_block *sblock)
1633 1634 1635 1636
{
	u64 flags;
	int ret;

1637 1638
	WARN_ON(sblock->page_count < 1);
	flags = sblock->pagev[0]->flags;
1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649
	ret = 0;
	if (flags & BTRFS_EXTENT_FLAG_DATA)
		ret = scrub_checksum_data(sblock);
	else if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK)
		ret = scrub_checksum_tree_block(sblock);
	else if (flags & BTRFS_EXTENT_FLAG_SUPER)
		(void)scrub_checksum_super(sblock);
	else
		WARN_ON(1);
	if (ret)
		scrub_handle_errored_block(sblock);
1650 1651

	return ret;
A
Arne Jansen 已提交
1652 1653
}

1654
static int scrub_checksum_data(struct scrub_block *sblock)
A
Arne Jansen 已提交
1655
{
1656
	struct scrub_ctx *sctx = sblock->sctx;
A
Arne Jansen 已提交
1657
	u8 csum[BTRFS_CSUM_SIZE];
1658 1659 1660
	u8 *on_disk_csum;
	struct page *page;
	void *buffer;
A
Arne Jansen 已提交
1661 1662
	u32 crc = ~(u32)0;
	int fail = 0;
1663 1664
	u64 len;
	int index;
A
Arne Jansen 已提交
1665

1666
	BUG_ON(sblock->page_count < 1);
1667
	if (!sblock->pagev[0]->have_csum)
A
Arne Jansen 已提交
1668 1669
		return 0;

1670 1671
	on_disk_csum = sblock->pagev[0]->csum;
	page = sblock->pagev[0]->page;
1672
	buffer = kmap_atomic(page);
1673

1674
	len = sctx->sectorsize;
1675 1676 1677 1678
	index = 0;
	for (;;) {
		u64 l = min_t(u64, len, PAGE_SIZE);

1679
		crc = btrfs_csum_data(buffer, crc, l);
1680
		kunmap_atomic(buffer);
1681 1682 1683 1684 1685
		len -= l;
		if (len == 0)
			break;
		index++;
		BUG_ON(index >= sblock->page_count);
1686 1687
		BUG_ON(!sblock->pagev[index]->page);
		page = sblock->pagev[index]->page;
1688
		buffer = kmap_atomic(page);
1689 1690
	}

A
Arne Jansen 已提交
1691
	btrfs_csum_final(crc, csum);
1692
	if (memcmp(csum, on_disk_csum, sctx->csum_size))
A
Arne Jansen 已提交
1693 1694 1695 1696 1697
		fail = 1;

	return fail;
}

1698
static int scrub_checksum_tree_block(struct scrub_block *sblock)
A
Arne Jansen 已提交
1699
{
1700
	struct scrub_ctx *sctx = sblock->sctx;
A
Arne Jansen 已提交
1701
	struct btrfs_header *h;
1702
	struct btrfs_root *root = sctx->dev_root;
A
Arne Jansen 已提交
1703
	struct btrfs_fs_info *fs_info = root->fs_info;
1704 1705 1706 1707 1708 1709
	u8 calculated_csum[BTRFS_CSUM_SIZE];
	u8 on_disk_csum[BTRFS_CSUM_SIZE];
	struct page *page;
	void *mapped_buffer;
	u64 mapped_size;
	void *p;
A
Arne Jansen 已提交
1710 1711 1712
	u32 crc = ~(u32)0;
	int fail = 0;
	int crc_fail = 0;
1713 1714 1715 1716
	u64 len;
	int index;

	BUG_ON(sblock->page_count < 1);
1717
	page = sblock->pagev[0]->page;
1718
	mapped_buffer = kmap_atomic(page);
1719
	h = (struct btrfs_header *)mapped_buffer;
1720
	memcpy(on_disk_csum, h->csum, sctx->csum_size);
A
Arne Jansen 已提交
1721 1722 1723 1724 1725 1726 1727

	/*
	 * we don't use the getter functions here, as we
	 * a) don't have an extent buffer and
	 * b) the page is already kmapped
	 */

1728
	if (sblock->pagev[0]->logical != btrfs_stack_header_bytenr(h))
A
Arne Jansen 已提交
1729 1730
		++fail;

1731
	if (sblock->pagev[0]->generation != btrfs_stack_header_generation(h))
A
Arne Jansen 已提交
1732 1733 1734 1735 1736 1737 1738 1739 1740
		++fail;

	if (memcmp(h->fsid, fs_info->fsid, BTRFS_UUID_SIZE))
		++fail;

	if (memcmp(h->chunk_tree_uuid, fs_info->chunk_tree_uuid,
		   BTRFS_UUID_SIZE))
		++fail;

1741
	WARN_ON(sctx->nodesize != sctx->leafsize);
1742
	len = sctx->nodesize - BTRFS_CSUM_SIZE;
1743 1744 1745 1746 1747 1748
	mapped_size = PAGE_SIZE - BTRFS_CSUM_SIZE;
	p = ((u8 *)mapped_buffer) + BTRFS_CSUM_SIZE;
	index = 0;
	for (;;) {
		u64 l = min_t(u64, len, mapped_size);

1749
		crc = btrfs_csum_data(p, crc, l);
1750
		kunmap_atomic(mapped_buffer);
1751 1752 1753 1754 1755
		len -= l;
		if (len == 0)
			break;
		index++;
		BUG_ON(index >= sblock->page_count);
1756 1757
		BUG_ON(!sblock->pagev[index]->page);
		page = sblock->pagev[index]->page;
1758
		mapped_buffer = kmap_atomic(page);
1759 1760 1761 1762 1763
		mapped_size = PAGE_SIZE;
		p = mapped_buffer;
	}

	btrfs_csum_final(crc, calculated_csum);
1764
	if (memcmp(calculated_csum, on_disk_csum, sctx->csum_size))
A
Arne Jansen 已提交
1765 1766 1767 1768 1769
		++crc_fail;

	return fail || crc_fail;
}

1770
static int scrub_checksum_super(struct scrub_block *sblock)
A
Arne Jansen 已提交
1771 1772
{
	struct btrfs_super_block *s;
1773
	struct scrub_ctx *sctx = sblock->sctx;
1774
	struct btrfs_root *root = sctx->dev_root;
A
Arne Jansen 已提交
1775
	struct btrfs_fs_info *fs_info = root->fs_info;
1776 1777 1778 1779 1780 1781
	u8 calculated_csum[BTRFS_CSUM_SIZE];
	u8 on_disk_csum[BTRFS_CSUM_SIZE];
	struct page *page;
	void *mapped_buffer;
	u64 mapped_size;
	void *p;
A
Arne Jansen 已提交
1782
	u32 crc = ~(u32)0;
1783 1784
	int fail_gen = 0;
	int fail_cor = 0;
1785 1786
	u64 len;
	int index;
A
Arne Jansen 已提交
1787

1788
	BUG_ON(sblock->page_count < 1);
1789
	page = sblock->pagev[0]->page;
1790
	mapped_buffer = kmap_atomic(page);
1791
	s = (struct btrfs_super_block *)mapped_buffer;
1792
	memcpy(on_disk_csum, s->csum, sctx->csum_size);
A
Arne Jansen 已提交
1793

1794
	if (sblock->pagev[0]->logical != btrfs_super_bytenr(s))
1795
		++fail_cor;
A
Arne Jansen 已提交
1796

1797
	if (sblock->pagev[0]->generation != btrfs_super_generation(s))
1798
		++fail_gen;
A
Arne Jansen 已提交
1799 1800

	if (memcmp(s->fsid, fs_info->fsid, BTRFS_UUID_SIZE))
1801
		++fail_cor;
A
Arne Jansen 已提交
1802

1803 1804 1805 1806 1807 1808 1809
	len = BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE;
	mapped_size = PAGE_SIZE - BTRFS_CSUM_SIZE;
	p = ((u8 *)mapped_buffer) + BTRFS_CSUM_SIZE;
	index = 0;
	for (;;) {
		u64 l = min_t(u64, len, mapped_size);

1810
		crc = btrfs_csum_data(p, crc, l);
1811
		kunmap_atomic(mapped_buffer);
1812 1813 1814 1815 1816
		len -= l;
		if (len == 0)
			break;
		index++;
		BUG_ON(index >= sblock->page_count);
1817 1818
		BUG_ON(!sblock->pagev[index]->page);
		page = sblock->pagev[index]->page;
1819
		mapped_buffer = kmap_atomic(page);
1820 1821 1822 1823 1824
		mapped_size = PAGE_SIZE;
		p = mapped_buffer;
	}

	btrfs_csum_final(crc, calculated_csum);
1825
	if (memcmp(calculated_csum, on_disk_csum, sctx->csum_size))
1826
		++fail_cor;
A
Arne Jansen 已提交
1827

1828
	if (fail_cor + fail_gen) {
A
Arne Jansen 已提交
1829 1830 1831 1832 1833
		/*
		 * if we find an error in a super block, we just report it.
		 * They will get written with the next transaction commit
		 * anyway
		 */
1834 1835 1836
		spin_lock(&sctx->stat_lock);
		++sctx->stat.super_errors;
		spin_unlock(&sctx->stat_lock);
1837
		if (fail_cor)
1838
			btrfs_dev_stat_inc_and_print(sblock->pagev[0]->dev,
1839 1840
				BTRFS_DEV_STAT_CORRUPTION_ERRS);
		else
1841
			btrfs_dev_stat_inc_and_print(sblock->pagev[0]->dev,
1842
				BTRFS_DEV_STAT_GENERATION_ERRS);
A
Arne Jansen 已提交
1843 1844
	}

1845
	return fail_cor + fail_gen;
A
Arne Jansen 已提交
1846 1847
}

1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858
static void scrub_block_get(struct scrub_block *sblock)
{
	atomic_inc(&sblock->ref_count);
}

static void scrub_block_put(struct scrub_block *sblock)
{
	if (atomic_dec_and_test(&sblock->ref_count)) {
		int i;

		for (i = 0; i < sblock->page_count; i++)
1859
			scrub_page_put(sblock->pagev[i]);
1860 1861 1862 1863
		kfree(sblock);
	}
}

1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877
static void scrub_page_get(struct scrub_page *spage)
{
	atomic_inc(&spage->ref_count);
}

static void scrub_page_put(struct scrub_page *spage)
{
	if (atomic_dec_and_test(&spage->ref_count)) {
		if (spage->page)
			__free_page(spage->page);
		kfree(spage);
	}
}

1878
static void scrub_submit(struct scrub_ctx *sctx)
A
Arne Jansen 已提交
1879 1880 1881
{
	struct scrub_bio *sbio;

1882
	if (sctx->curr == -1)
S
Stefan Behrens 已提交
1883
		return;
A
Arne Jansen 已提交
1884

1885 1886
	sbio = sctx->bios[sctx->curr];
	sctx->curr = -1;
1887
	scrub_pending_bio_inc(sctx);
A
Arne Jansen 已提交
1888

1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902
	if (!sbio->bio->bi_bdev) {
		/*
		 * this case should not happen. If btrfs_map_block() is
		 * wrong, it could happen for dev-replace operations on
		 * missing devices when no mirrors are available, but in
		 * this case it should already fail the mount.
		 * This case is handled correctly (but _very_ slowly).
		 */
		printk_ratelimited(KERN_WARNING
			"btrfs: scrub_submit(bio bdev == NULL) is unexpected!\n");
		bio_endio(sbio->bio, -EIO);
	} else {
		btrfsic_submit_bio(READ, sbio->bio);
	}
A
Arne Jansen 已提交
1903 1904
}

1905 1906
static int scrub_add_page_to_rd_bio(struct scrub_ctx *sctx,
				    struct scrub_page *spage)
A
Arne Jansen 已提交
1907
{
1908
	struct scrub_block *sblock = spage->sblock;
A
Arne Jansen 已提交
1909
	struct scrub_bio *sbio;
1910
	int ret;
A
Arne Jansen 已提交
1911 1912 1913 1914 1915

again:
	/*
	 * grab a fresh bio or wait for one to become available
	 */
1916 1917 1918 1919 1920 1921 1922 1923
	while (sctx->curr == -1) {
		spin_lock(&sctx->list_lock);
		sctx->curr = sctx->first_free;
		if (sctx->curr != -1) {
			sctx->first_free = sctx->bios[sctx->curr]->next_free;
			sctx->bios[sctx->curr]->next_free = -1;
			sctx->bios[sctx->curr]->page_count = 0;
			spin_unlock(&sctx->list_lock);
A
Arne Jansen 已提交
1924
		} else {
1925 1926
			spin_unlock(&sctx->list_lock);
			wait_event(sctx->list_wait, sctx->first_free != -1);
A
Arne Jansen 已提交
1927 1928
		}
	}
1929
	sbio = sctx->bios[sctx->curr];
1930
	if (sbio->page_count == 0) {
1931 1932
		struct bio *bio;

1933 1934
		sbio->physical = spage->physical;
		sbio->logical = spage->logical;
1935
		sbio->dev = spage->dev;
1936 1937
		bio = sbio->bio;
		if (!bio) {
1938
			bio = btrfs_io_bio_alloc(GFP_NOFS, sctx->pages_per_rd_bio);
1939 1940 1941 1942
			if (!bio)
				return -ENOMEM;
			sbio->bio = bio;
		}
1943 1944 1945

		bio->bi_private = sbio;
		bio->bi_end_io = scrub_bio_end_io;
1946 1947
		bio->bi_bdev = sbio->dev->bdev;
		bio->bi_sector = sbio->physical >> 9;
1948
		sbio->err = 0;
1949 1950 1951
	} else if (sbio->physical + sbio->page_count * PAGE_SIZE !=
		   spage->physical ||
		   sbio->logical + sbio->page_count * PAGE_SIZE !=
1952 1953
		   spage->logical ||
		   sbio->dev != spage->dev) {
1954
		scrub_submit(sctx);
A
Arne Jansen 已提交
1955 1956
		goto again;
	}
1957

1958 1959 1960 1961 1962 1963 1964 1965
	sbio->pagev[sbio->page_count] = spage;
	ret = bio_add_page(sbio->bio, spage->page, PAGE_SIZE, 0);
	if (ret != PAGE_SIZE) {
		if (sbio->page_count < 1) {
			bio_put(sbio->bio);
			sbio->bio = NULL;
			return -EIO;
		}
1966
		scrub_submit(sctx);
1967 1968 1969
		goto again;
	}

1970
	scrub_block_get(sblock); /* one for the page added to the bio */
1971 1972
	atomic_inc(&sblock->outstanding_pages);
	sbio->page_count++;
1973
	if (sbio->page_count == sctx->pages_per_rd_bio)
1974
		scrub_submit(sctx);
1975 1976 1977 1978

	return 0;
}

1979
static int scrub_pages(struct scrub_ctx *sctx, u64 logical, u64 len,
1980
		       u64 physical, struct btrfs_device *dev, u64 flags,
1981 1982
		       u64 gen, int mirror_num, u8 *csum, int force,
		       u64 physical_for_dev_replace)
1983 1984 1985 1986 1987 1988
{
	struct scrub_block *sblock;
	int index;

	sblock = kzalloc(sizeof(*sblock), GFP_NOFS);
	if (!sblock) {
1989 1990 1991
		spin_lock(&sctx->stat_lock);
		sctx->stat.malloc_errors++;
		spin_unlock(&sctx->stat_lock);
1992
		return -ENOMEM;
A
Arne Jansen 已提交
1993
	}
1994

1995 1996
	/* one ref inside this function, plus one for each page added to
	 * a bio later on */
1997
	atomic_set(&sblock->ref_count, 1);
1998
	sblock->sctx = sctx;
1999 2000 2001
	sblock->no_io_error_seen = 1;

	for (index = 0; len > 0; index++) {
2002
		struct scrub_page *spage;
2003 2004
		u64 l = min_t(u64, len, PAGE_SIZE);

2005 2006 2007
		spage = kzalloc(sizeof(*spage), GFP_NOFS);
		if (!spage) {
leave_nomem:
2008 2009 2010
			spin_lock(&sctx->stat_lock);
			sctx->stat.malloc_errors++;
			spin_unlock(&sctx->stat_lock);
2011
			scrub_block_put(sblock);
2012 2013
			return -ENOMEM;
		}
2014 2015 2016
		BUG_ON(index >= SCRUB_MAX_PAGES_PER_BLOCK);
		scrub_page_get(spage);
		sblock->pagev[index] = spage;
2017
		spage->sblock = sblock;
2018
		spage->dev = dev;
2019 2020 2021 2022
		spage->flags = flags;
		spage->generation = gen;
		spage->logical = logical;
		spage->physical = physical;
2023
		spage->physical_for_dev_replace = physical_for_dev_replace;
2024 2025 2026
		spage->mirror_num = mirror_num;
		if (csum) {
			spage->have_csum = 1;
2027
			memcpy(spage->csum, csum, sctx->csum_size);
2028 2029 2030 2031
		} else {
			spage->have_csum = 0;
		}
		sblock->page_count++;
2032 2033 2034
		spage->page = alloc_page(GFP_NOFS);
		if (!spage->page)
			goto leave_nomem;
2035 2036 2037
		len -= l;
		logical += l;
		physical += l;
2038
		physical_for_dev_replace += l;
2039 2040
	}

2041
	WARN_ON(sblock->page_count == 0);
2042
	for (index = 0; index < sblock->page_count; index++) {
2043
		struct scrub_page *spage = sblock->pagev[index];
2044 2045
		int ret;

2046
		ret = scrub_add_page_to_rd_bio(sctx, spage);
2047 2048
		if (ret) {
			scrub_block_put(sblock);
2049
			return ret;
2050
		}
2051
	}
A
Arne Jansen 已提交
2052

2053
	if (force)
2054
		scrub_submit(sctx);
A
Arne Jansen 已提交
2055

2056 2057
	/* last one frees, either here or in bio completion for last page */
	scrub_block_put(sblock);
A
Arne Jansen 已提交
2058 2059 2060
	return 0;
}

2061 2062 2063
static void scrub_bio_end_io(struct bio *bio, int err)
{
	struct scrub_bio *sbio = bio->bi_private;
2064
	struct btrfs_fs_info *fs_info = sbio->dev->dev_root->fs_info;
2065 2066 2067 2068 2069 2070 2071 2072 2073 2074

	sbio->err = err;
	sbio->bio = bio;

	btrfs_queue_worker(&fs_info->scrub_workers, &sbio->work);
}

static void scrub_bio_end_io_worker(struct btrfs_work *work)
{
	struct scrub_bio *sbio = container_of(work, struct scrub_bio, work);
2075
	struct scrub_ctx *sctx = sbio->sctx;
2076 2077
	int i;

2078
	BUG_ON(sbio->page_count > SCRUB_PAGES_PER_RD_BIO);
2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099
	if (sbio->err) {
		for (i = 0; i < sbio->page_count; i++) {
			struct scrub_page *spage = sbio->pagev[i];

			spage->io_error = 1;
			spage->sblock->no_io_error_seen = 0;
		}
	}

	/* now complete the scrub_block items that have all pages completed */
	for (i = 0; i < sbio->page_count; i++) {
		struct scrub_page *spage = sbio->pagev[i];
		struct scrub_block *sblock = spage->sblock;

		if (atomic_dec_and_test(&sblock->outstanding_pages))
			scrub_block_complete(sblock);
		scrub_block_put(sblock);
	}

	bio_put(sbio->bio);
	sbio->bio = NULL;
2100 2101 2102 2103
	spin_lock(&sctx->list_lock);
	sbio->next_free = sctx->first_free;
	sctx->first_free = sbio->index;
	spin_unlock(&sctx->list_lock);
2104 2105 2106 2107 2108 2109 2110 2111

	if (sctx->is_dev_replace &&
	    atomic_read(&sctx->wr_ctx.flush_all_writes)) {
		mutex_lock(&sctx->wr_ctx.wr_lock);
		scrub_wr_submit(sctx);
		mutex_unlock(&sctx->wr_ctx.wr_lock);
	}

2112
	scrub_pending_bio_dec(sctx);
2113 2114 2115 2116
}

static void scrub_block_complete(struct scrub_block *sblock)
{
2117
	if (!sblock->no_io_error_seen) {
2118
		scrub_handle_errored_block(sblock);
2119 2120 2121 2122 2123 2124 2125 2126 2127
	} else {
		/*
		 * if has checksum error, write via repair mechanism in
		 * dev replace case, otherwise write here in dev replace
		 * case.
		 */
		if (!scrub_checksum(sblock) && sblock->sctx->is_dev_replace)
			scrub_write_block_to_dev_replace(sblock);
	}
2128 2129
}

2130
static int scrub_find_csum(struct scrub_ctx *sctx, u64 logical, u64 len,
A
Arne Jansen 已提交
2131 2132 2133
			   u8 *csum)
{
	struct btrfs_ordered_sum *sum = NULL;
2134
	unsigned long index;
A
Arne Jansen 已提交
2135 2136
	unsigned long num_sectors;

2137 2138
	while (!list_empty(&sctx->csum_list)) {
		sum = list_first_entry(&sctx->csum_list,
A
Arne Jansen 已提交
2139 2140 2141 2142 2143 2144
				       struct btrfs_ordered_sum, list);
		if (sum->bytenr > logical)
			return 0;
		if (sum->bytenr + sum->len > logical)
			break;

2145
		++sctx->stat.csum_discards;
A
Arne Jansen 已提交
2146 2147 2148 2149 2150 2151 2152
		list_del(&sum->list);
		kfree(sum);
		sum = NULL;
	}
	if (!sum)
		return 0;

2153
	index = ((u32)(logical - sum->bytenr)) / sctx->sectorsize;
2154
	num_sectors = sum->len / sctx->sectorsize;
2155 2156
	memcpy(csum, sum->sums + index, sctx->csum_size);
	if (index == num_sectors - 1) {
A
Arne Jansen 已提交
2157 2158 2159
		list_del(&sum->list);
		kfree(sum);
	}
2160
	return 1;
A
Arne Jansen 已提交
2161 2162 2163
}

/* scrub extent tries to collect up to 64 kB for each bio */
2164
static int scrub_extent(struct scrub_ctx *sctx, u64 logical, u64 len,
2165
			u64 physical, struct btrfs_device *dev, u64 flags,
2166
			u64 gen, int mirror_num, u64 physical_for_dev_replace)
A
Arne Jansen 已提交
2167 2168 2169
{
	int ret;
	u8 csum[BTRFS_CSUM_SIZE];
2170 2171 2172
	u32 blocksize;

	if (flags & BTRFS_EXTENT_FLAG_DATA) {
2173 2174 2175 2176 2177
		blocksize = sctx->sectorsize;
		spin_lock(&sctx->stat_lock);
		sctx->stat.data_extents_scrubbed++;
		sctx->stat.data_bytes_scrubbed += len;
		spin_unlock(&sctx->stat_lock);
2178
	} else if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) {
2179
		WARN_ON(sctx->nodesize != sctx->leafsize);
2180 2181 2182 2183 2184
		blocksize = sctx->nodesize;
		spin_lock(&sctx->stat_lock);
		sctx->stat.tree_extents_scrubbed++;
		sctx->stat.tree_bytes_scrubbed += len;
		spin_unlock(&sctx->stat_lock);
2185
	} else {
2186
		blocksize = sctx->sectorsize;
2187
		WARN_ON(1);
2188
	}
A
Arne Jansen 已提交
2189 2190

	while (len) {
2191
		u64 l = min_t(u64, len, blocksize);
A
Arne Jansen 已提交
2192 2193 2194 2195
		int have_csum = 0;

		if (flags & BTRFS_EXTENT_FLAG_DATA) {
			/* push csums to sbio */
2196
			have_csum = scrub_find_csum(sctx, logical, l, csum);
A
Arne Jansen 已提交
2197
			if (have_csum == 0)
2198
				++sctx->stat.no_csum;
2199 2200 2201 2202 2203 2204
			if (sctx->is_dev_replace && !have_csum) {
				ret = copy_nocow_pages(sctx, logical, l,
						       mirror_num,
						      physical_for_dev_replace);
				goto behind_scrub_pages;
			}
A
Arne Jansen 已提交
2205
		}
2206
		ret = scrub_pages(sctx, logical, l, physical, dev, flags, gen,
2207 2208 2209
				  mirror_num, have_csum ? csum : NULL, 0,
				  physical_for_dev_replace);
behind_scrub_pages:
A
Arne Jansen 已提交
2210 2211 2212 2213 2214
		if (ret)
			return ret;
		len -= l;
		logical += l;
		physical += l;
2215
		physical_for_dev_replace += l;
A
Arne Jansen 已提交
2216 2217 2218 2219
	}
	return 0;
}

2220
static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
2221 2222
					   struct map_lookup *map,
					   struct btrfs_device *scrub_dev,
2223 2224
					   int num, u64 base, u64 length,
					   int is_dev_replace)
A
Arne Jansen 已提交
2225 2226
{
	struct btrfs_path *path;
2227
	struct btrfs_fs_info *fs_info = sctx->dev_root->fs_info;
A
Arne Jansen 已提交
2228 2229 2230
	struct btrfs_root *root = fs_info->extent_root;
	struct btrfs_root *csum_root = fs_info->csum_root;
	struct btrfs_extent_item *extent;
2231
	struct blk_plug plug;
A
Arne Jansen 已提交
2232 2233 2234 2235 2236 2237 2238 2239
	u64 flags;
	int ret;
	int slot;
	u64 nstripes;
	struct extent_buffer *l;
	struct btrfs_key key;
	u64 physical;
	u64 logical;
L
Liu Bo 已提交
2240
	u64 logic_end;
A
Arne Jansen 已提交
2241
	u64 generation;
2242
	int mirror_num;
A
Arne Jansen 已提交
2243 2244 2245 2246
	struct reada_control *reada1;
	struct reada_control *reada2;
	struct btrfs_key key_start;
	struct btrfs_key key_end;
A
Arne Jansen 已提交
2247 2248
	u64 increment = map->stripe_len;
	u64 offset;
2249 2250 2251 2252 2253
	u64 extent_logical;
	u64 extent_physical;
	u64 extent_len;
	struct btrfs_device *extent_dev;
	int extent_mirror_num;
L
Liu Bo 已提交
2254
	int stop_loop;
A
Arne Jansen 已提交
2255

D
David Woodhouse 已提交
2256 2257 2258 2259 2260 2261 2262
	if (map->type & (BTRFS_BLOCK_GROUP_RAID5 |
			 BTRFS_BLOCK_GROUP_RAID6)) {
		if (num >= nr_data_stripes(map)) {
			return 0;
		}
	}

A
Arne Jansen 已提交
2263 2264 2265 2266 2267 2268
	nstripes = length;
	offset = 0;
	do_div(nstripes, map->stripe_len);
	if (map->type & BTRFS_BLOCK_GROUP_RAID0) {
		offset = map->stripe_len * num;
		increment = map->stripe_len * map->num_stripes;
2269
		mirror_num = 1;
A
Arne Jansen 已提交
2270 2271 2272 2273
	} else if (map->type & BTRFS_BLOCK_GROUP_RAID10) {
		int factor = map->num_stripes / map->sub_stripes;
		offset = map->stripe_len * (num / map->sub_stripes);
		increment = map->stripe_len * factor;
2274
		mirror_num = num % map->sub_stripes + 1;
A
Arne Jansen 已提交
2275 2276
	} else if (map->type & BTRFS_BLOCK_GROUP_RAID1) {
		increment = map->stripe_len;
2277
		mirror_num = num % map->num_stripes + 1;
A
Arne Jansen 已提交
2278 2279
	} else if (map->type & BTRFS_BLOCK_GROUP_DUP) {
		increment = map->stripe_len;
2280
		mirror_num = num % map->num_stripes + 1;
A
Arne Jansen 已提交
2281 2282
	} else {
		increment = map->stripe_len;
2283
		mirror_num = 1;
A
Arne Jansen 已提交
2284 2285 2286 2287 2288 2289
	}

	path = btrfs_alloc_path();
	if (!path)
		return -ENOMEM;

2290 2291 2292 2293 2294
	/*
	 * work on commit root. The related disk blocks are static as
	 * long as COW is applied. This means, it is save to rewrite
	 * them to repair disk errors without any race conditions
	 */
A
Arne Jansen 已提交
2295 2296 2297 2298
	path->search_commit_root = 1;
	path->skip_locking = 1;

	/*
A
Arne Jansen 已提交
2299 2300 2301
	 * trigger the readahead for extent tree csum tree and wait for
	 * completion. During readahead, the scrub is officially paused
	 * to not hold off transaction commits
A
Arne Jansen 已提交
2302 2303 2304
	 */
	logical = base + offset;

2305
	wait_event(sctx->list_wait,
2306
		   atomic_read(&sctx->bios_in_flight) == 0);
A
Arne Jansen 已提交
2307 2308 2309 2310 2311 2312 2313 2314
	atomic_inc(&fs_info->scrubs_paused);
	wake_up(&fs_info->scrub_pause_wait);

	/* FIXME it might be better to start readahead at commit root */
	key_start.objectid = logical;
	key_start.type = BTRFS_EXTENT_ITEM_KEY;
	key_start.offset = (u64)0;
	key_end.objectid = base + offset + nstripes * increment;
2315 2316
	key_end.type = BTRFS_METADATA_ITEM_KEY;
	key_end.offset = (u64)-1;
A
Arne Jansen 已提交
2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337
	reada1 = btrfs_reada_add(root, &key_start, &key_end);

	key_start.objectid = BTRFS_EXTENT_CSUM_OBJECTID;
	key_start.type = BTRFS_EXTENT_CSUM_KEY;
	key_start.offset = logical;
	key_end.objectid = BTRFS_EXTENT_CSUM_OBJECTID;
	key_end.type = BTRFS_EXTENT_CSUM_KEY;
	key_end.offset = base + offset + nstripes * increment;
	reada2 = btrfs_reada_add(csum_root, &key_start, &key_end);

	if (!IS_ERR(reada1))
		btrfs_reada_wait(reada1);
	if (!IS_ERR(reada2))
		btrfs_reada_wait(reada2);

	mutex_lock(&fs_info->scrub_lock);
	while (atomic_read(&fs_info->scrub_pause_req)) {
		mutex_unlock(&fs_info->scrub_lock);
		wait_event(fs_info->scrub_pause_wait,
		   atomic_read(&fs_info->scrub_pause_req) == 0);
		mutex_lock(&fs_info->scrub_lock);
A
Arne Jansen 已提交
2338
	}
A
Arne Jansen 已提交
2339 2340 2341
	atomic_dec(&fs_info->scrubs_paused);
	mutex_unlock(&fs_info->scrub_lock);
	wake_up(&fs_info->scrub_pause_wait);
A
Arne Jansen 已提交
2342 2343 2344 2345 2346

	/*
	 * collect all data csums for the stripe to avoid seeking during
	 * the scrub. This might currently (crc32) end up to be about 1MB
	 */
2347
	blk_start_plug(&plug);
A
Arne Jansen 已提交
2348 2349 2350 2351

	/*
	 * now find all extents for each stripe and scrub them
	 */
A
Arne Jansen 已提交
2352 2353
	logical = base + offset;
	physical = map->stripes[num].physical;
L
Liu Bo 已提交
2354
	logic_end = logical + increment * nstripes;
A
Arne Jansen 已提交
2355
	ret = 0;
L
Liu Bo 已提交
2356
	while (logical < logic_end) {
A
Arne Jansen 已提交
2357 2358 2359 2360
		/*
		 * canceled?
		 */
		if (atomic_read(&fs_info->scrub_cancel_req) ||
2361
		    atomic_read(&sctx->cancel_req)) {
A
Arne Jansen 已提交
2362 2363 2364 2365 2366 2367 2368 2369
			ret = -ECANCELED;
			goto out;
		}
		/*
		 * check to see if we have to pause
		 */
		if (atomic_read(&fs_info->scrub_pause_req)) {
			/* push queued extents */
2370
			atomic_set(&sctx->wr_ctx.flush_all_writes, 1);
2371
			scrub_submit(sctx);
2372 2373 2374
			mutex_lock(&sctx->wr_ctx.wr_lock);
			scrub_wr_submit(sctx);
			mutex_unlock(&sctx->wr_ctx.wr_lock);
2375
			wait_event(sctx->list_wait,
2376
				   atomic_read(&sctx->bios_in_flight) == 0);
2377
			atomic_set(&sctx->wr_ctx.flush_all_writes, 0);
A
Arne Jansen 已提交
2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393
			atomic_inc(&fs_info->scrubs_paused);
			wake_up(&fs_info->scrub_pause_wait);
			mutex_lock(&fs_info->scrub_lock);
			while (atomic_read(&fs_info->scrub_pause_req)) {
				mutex_unlock(&fs_info->scrub_lock);
				wait_event(fs_info->scrub_pause_wait,
				   atomic_read(&fs_info->scrub_pause_req) == 0);
				mutex_lock(&fs_info->scrub_lock);
			}
			atomic_dec(&fs_info->scrubs_paused);
			mutex_unlock(&fs_info->scrub_lock);
			wake_up(&fs_info->scrub_pause_wait);
		}

		key.objectid = logical;
		key.type = BTRFS_EXTENT_ITEM_KEY;
L
Liu Bo 已提交
2394
		key.offset = (u64)-1;
A
Arne Jansen 已提交
2395 2396 2397 2398

		ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
		if (ret < 0)
			goto out;
2399

2400
		if (ret > 0) {
A
Arne Jansen 已提交
2401 2402 2403 2404
			ret = btrfs_previous_item(root, path, 0,
						  BTRFS_EXTENT_ITEM_KEY);
			if (ret < 0)
				goto out;
2405 2406 2407 2408 2409 2410 2411 2412 2413
			if (ret > 0) {
				/* there's no smaller item, so stick with the
				 * larger one */
				btrfs_release_path(path);
				ret = btrfs_search_slot(NULL, root, &key,
							path, 0, 0);
				if (ret < 0)
					goto out;
			}
A
Arne Jansen 已提交
2414 2415
		}

L
Liu Bo 已提交
2416
		stop_loop = 0;
A
Arne Jansen 已提交
2417
		while (1) {
2418 2419
			u64 bytes;

A
Arne Jansen 已提交
2420 2421 2422 2423 2424 2425 2426 2427 2428
			l = path->nodes[0];
			slot = path->slots[0];
			if (slot >= btrfs_header_nritems(l)) {
				ret = btrfs_next_leaf(root, path);
				if (ret == 0)
					continue;
				if (ret < 0)
					goto out;

L
Liu Bo 已提交
2429
				stop_loop = 1;
A
Arne Jansen 已提交
2430 2431 2432 2433
				break;
			}
			btrfs_item_key_to_cpu(l, &key, slot);

2434 2435 2436 2437 2438 2439
			if (key.type == BTRFS_METADATA_ITEM_KEY)
				bytes = root->leafsize;
			else
				bytes = key.offset;

			if (key.objectid + bytes <= logical)
A
Arne Jansen 已提交
2440 2441
				goto next;

L
Liu Bo 已提交
2442 2443 2444
			if (key.type != BTRFS_EXTENT_ITEM_KEY &&
			    key.type != BTRFS_METADATA_ITEM_KEY)
				goto next;
A
Arne Jansen 已提交
2445

L
Liu Bo 已提交
2446 2447 2448 2449 2450 2451
			if (key.objectid >= logical + map->stripe_len) {
				/* out of this device extent */
				if (key.objectid >= logic_end)
					stop_loop = 1;
				break;
			}
A
Arne Jansen 已提交
2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462

			extent = btrfs_item_ptr(l, slot,
						struct btrfs_extent_item);
			flags = btrfs_extent_flags(l, extent);
			generation = btrfs_extent_generation(l, extent);

			if (key.objectid < logical &&
			    (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK)) {
				printk(KERN_ERR
				       "btrfs scrub: tree block %llu spanning "
				       "stripes, ignored. logical=%llu\n",
2463
				       key.objectid, logical);
A
Arne Jansen 已提交
2464 2465 2466
				goto next;
			}

L
Liu Bo 已提交
2467 2468 2469 2470
again:
			extent_logical = key.objectid;
			extent_len = bytes;

A
Arne Jansen 已提交
2471 2472 2473
			/*
			 * trim extent to this stripe
			 */
L
Liu Bo 已提交
2474 2475 2476
			if (extent_logical < logical) {
				extent_len -= logical - extent_logical;
				extent_logical = logical;
A
Arne Jansen 已提交
2477
			}
L
Liu Bo 已提交
2478
			if (extent_logical + extent_len >
A
Arne Jansen 已提交
2479
			    logical + map->stripe_len) {
L
Liu Bo 已提交
2480 2481
				extent_len = logical + map->stripe_len -
					     extent_logical;
A
Arne Jansen 已提交
2482 2483
			}

L
Liu Bo 已提交
2484
			extent_physical = extent_logical - logical + physical;
2485 2486 2487 2488 2489 2490 2491
			extent_dev = scrub_dev;
			extent_mirror_num = mirror_num;
			if (is_dev_replace)
				scrub_remap_extent(fs_info, extent_logical,
						   extent_len, &extent_physical,
						   &extent_dev,
						   &extent_mirror_num);
L
Liu Bo 已提交
2492 2493 2494 2495 2496 2497 2498

			ret = btrfs_lookup_csums_range(csum_root, logical,
						logical + map->stripe_len - 1,
						&sctx->csum_list, 1);
			if (ret)
				goto out;

2499 2500 2501
			ret = scrub_extent(sctx, extent_logical, extent_len,
					   extent_physical, extent_dev, flags,
					   generation, extent_mirror_num,
2502
					   extent_logical - logical + physical);
A
Arne Jansen 已提交
2503 2504 2505
			if (ret)
				goto out;

2506
			scrub_free_csums(sctx);
L
Liu Bo 已提交
2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521
			if (extent_logical + extent_len <
			    key.objectid + bytes) {
				logical += increment;
				physical += map->stripe_len;

				if (logical < key.objectid + bytes) {
					cond_resched();
					goto again;
				}

				if (logical >= logic_end) {
					stop_loop = 1;
					break;
				}
			}
A
Arne Jansen 已提交
2522 2523 2524
next:
			path->slots[0]++;
		}
C
Chris Mason 已提交
2525
		btrfs_release_path(path);
A
Arne Jansen 已提交
2526 2527
		logical += increment;
		physical += map->stripe_len;
2528
		spin_lock(&sctx->stat_lock);
L
Liu Bo 已提交
2529 2530 2531 2532 2533
		if (stop_loop)
			sctx->stat.last_physical = map->stripes[num].physical +
						   length;
		else
			sctx->stat.last_physical = physical;
2534
		spin_unlock(&sctx->stat_lock);
L
Liu Bo 已提交
2535 2536
		if (stop_loop)
			break;
A
Arne Jansen 已提交
2537
	}
2538
out:
A
Arne Jansen 已提交
2539
	/* push queued extents */
2540
	scrub_submit(sctx);
2541 2542 2543
	mutex_lock(&sctx->wr_ctx.wr_lock);
	scrub_wr_submit(sctx);
	mutex_unlock(&sctx->wr_ctx.wr_lock);
A
Arne Jansen 已提交
2544

2545
	blk_finish_plug(&plug);
A
Arne Jansen 已提交
2546 2547 2548 2549
	btrfs_free_path(path);
	return ret < 0 ? ret : 0;
}

2550
static noinline_for_stack int scrub_chunk(struct scrub_ctx *sctx,
2551 2552 2553
					  struct btrfs_device *scrub_dev,
					  u64 chunk_tree, u64 chunk_objectid,
					  u64 chunk_offset, u64 length,
2554
					  u64 dev_offset, int is_dev_replace)
A
Arne Jansen 已提交
2555 2556
{
	struct btrfs_mapping_tree *map_tree =
2557
		&sctx->dev_root->fs_info->mapping_tree;
A
Arne Jansen 已提交
2558 2559 2560
	struct map_lookup *map;
	struct extent_map *em;
	int i;
2561
	int ret = 0;
A
Arne Jansen 已提交
2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577

	read_lock(&map_tree->map_tree.lock);
	em = lookup_extent_mapping(&map_tree->map_tree, chunk_offset, 1);
	read_unlock(&map_tree->map_tree.lock);

	if (!em)
		return -EINVAL;

	map = (struct map_lookup *)em->bdev;
	if (em->start != chunk_offset)
		goto out;

	if (em->len < length)
		goto out;

	for (i = 0; i < map->num_stripes; ++i) {
2578
		if (map->stripes[i].dev->bdev == scrub_dev->bdev &&
2579
		    map->stripes[i].physical == dev_offset) {
2580
			ret = scrub_stripe(sctx, map, scrub_dev, i,
2581 2582
					   chunk_offset, length,
					   is_dev_replace);
A
Arne Jansen 已提交
2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593
			if (ret)
				goto out;
		}
	}
out:
	free_extent_map(em);

	return ret;
}

static noinline_for_stack
2594
int scrub_enumerate_chunks(struct scrub_ctx *sctx,
2595 2596
			   struct btrfs_device *scrub_dev, u64 start, u64 end,
			   int is_dev_replace)
A
Arne Jansen 已提交
2597 2598 2599
{
	struct btrfs_dev_extent *dev_extent = NULL;
	struct btrfs_path *path;
2600
	struct btrfs_root *root = sctx->dev_root;
A
Arne Jansen 已提交
2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611
	struct btrfs_fs_info *fs_info = root->fs_info;
	u64 length;
	u64 chunk_tree;
	u64 chunk_objectid;
	u64 chunk_offset;
	int ret;
	int slot;
	struct extent_buffer *l;
	struct btrfs_key key;
	struct btrfs_key found_key;
	struct btrfs_block_group_cache *cache;
2612
	struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
A
Arne Jansen 已提交
2613 2614 2615 2616 2617 2618 2619 2620 2621

	path = btrfs_alloc_path();
	if (!path)
		return -ENOMEM;

	path->reada = 2;
	path->search_commit_root = 1;
	path->skip_locking = 1;

2622
	key.objectid = scrub_dev->devid;
A
Arne Jansen 已提交
2623 2624 2625 2626 2627 2628
	key.offset = 0ull;
	key.type = BTRFS_DEV_EXTENT_KEY;

	while (1) {
		ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
		if (ret < 0)
2629 2630 2631 2632 2633 2634 2635 2636 2637
			break;
		if (ret > 0) {
			if (path->slots[0] >=
			    btrfs_header_nritems(path->nodes[0])) {
				ret = btrfs_next_leaf(root, path);
				if (ret)
					break;
			}
		}
A
Arne Jansen 已提交
2638 2639 2640 2641 2642 2643

		l = path->nodes[0];
		slot = path->slots[0];

		btrfs_item_key_to_cpu(l, &found_key, slot);

2644
		if (found_key.objectid != scrub_dev->devid)
A
Arne Jansen 已提交
2645 2646
			break;

2647
		if (btrfs_key_type(&found_key) != BTRFS_DEV_EXTENT_KEY)
A
Arne Jansen 已提交
2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660
			break;

		if (found_key.offset >= end)
			break;

		if (found_key.offset < key.offset)
			break;

		dev_extent = btrfs_item_ptr(l, slot, struct btrfs_dev_extent);
		length = btrfs_dev_extent_length(l, dev_extent);

		if (found_key.offset + length <= start) {
			key.offset = found_key.offset + length;
C
Chris Mason 已提交
2661
			btrfs_release_path(path);
A
Arne Jansen 已提交
2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675
			continue;
		}

		chunk_tree = btrfs_dev_extent_chunk_tree(l, dev_extent);
		chunk_objectid = btrfs_dev_extent_chunk_objectid(l, dev_extent);
		chunk_offset = btrfs_dev_extent_chunk_offset(l, dev_extent);

		/*
		 * get a reference on the corresponding block group to prevent
		 * the chunk from going away while we scrub it
		 */
		cache = btrfs_lookup_block_group(fs_info, chunk_offset);
		if (!cache) {
			ret = -ENOENT;
2676
			break;
A
Arne Jansen 已提交
2677
		}
2678 2679 2680
		dev_replace->cursor_right = found_key.offset + length;
		dev_replace->cursor_left = found_key.offset;
		dev_replace->item_needs_writeback = 1;
2681
		ret = scrub_chunk(sctx, scrub_dev, chunk_tree, chunk_objectid,
2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719
				  chunk_offset, length, found_key.offset,
				  is_dev_replace);

		/*
		 * flush, submit all pending read and write bios, afterwards
		 * wait for them.
		 * Note that in the dev replace case, a read request causes
		 * write requests that are submitted in the read completion
		 * worker. Therefore in the current situation, it is required
		 * that all write requests are flushed, so that all read and
		 * write requests are really completed when bios_in_flight
		 * changes to 0.
		 */
		atomic_set(&sctx->wr_ctx.flush_all_writes, 1);
		scrub_submit(sctx);
		mutex_lock(&sctx->wr_ctx.wr_lock);
		scrub_wr_submit(sctx);
		mutex_unlock(&sctx->wr_ctx.wr_lock);

		wait_event(sctx->list_wait,
			   atomic_read(&sctx->bios_in_flight) == 0);
		atomic_set(&sctx->wr_ctx.flush_all_writes, 0);
		atomic_inc(&fs_info->scrubs_paused);
		wake_up(&fs_info->scrub_pause_wait);
		wait_event(sctx->list_wait,
			   atomic_read(&sctx->workers_pending) == 0);

		mutex_lock(&fs_info->scrub_lock);
		while (atomic_read(&fs_info->scrub_pause_req)) {
			mutex_unlock(&fs_info->scrub_lock);
			wait_event(fs_info->scrub_pause_wait,
			   atomic_read(&fs_info->scrub_pause_req) == 0);
			mutex_lock(&fs_info->scrub_lock);
		}
		atomic_dec(&fs_info->scrubs_paused);
		mutex_unlock(&fs_info->scrub_lock);
		wake_up(&fs_info->scrub_pause_wait);

A
Arne Jansen 已提交
2720 2721 2722
		btrfs_put_block_group(cache);
		if (ret)
			break;
2723 2724
		if (is_dev_replace &&
		    atomic64_read(&dev_replace->num_write_errors) > 0) {
2725 2726 2727 2728 2729 2730 2731
			ret = -EIO;
			break;
		}
		if (sctx->stat.malloc_errors > 0) {
			ret = -ENOMEM;
			break;
		}
A
Arne Jansen 已提交
2732

2733 2734 2735
		dev_replace->cursor_left = dev_replace->cursor_right;
		dev_replace->item_needs_writeback = 1;

A
Arne Jansen 已提交
2736
		key.offset = found_key.offset + length;
C
Chris Mason 已提交
2737
		btrfs_release_path(path);
A
Arne Jansen 已提交
2738 2739 2740
	}

	btrfs_free_path(path);
2741 2742 2743 2744 2745 2746

	/*
	 * ret can still be 1 from search_slot or next_leaf,
	 * that's not an error
	 */
	return ret < 0 ? ret : 0;
A
Arne Jansen 已提交
2747 2748
}

2749 2750
static noinline_for_stack int scrub_supers(struct scrub_ctx *sctx,
					   struct btrfs_device *scrub_dev)
A
Arne Jansen 已提交
2751 2752 2753 2754 2755
{
	int	i;
	u64	bytenr;
	u64	gen;
	int	ret;
2756
	struct btrfs_root *root = sctx->dev_root;
A
Arne Jansen 已提交
2757

2758
	if (test_bit(BTRFS_FS_STATE_ERROR, &root->fs_info->fs_state))
2759 2760
		return -EIO;

A
Arne Jansen 已提交
2761 2762 2763 2764
	gen = root->fs_info->last_trans_committed;

	for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) {
		bytenr = btrfs_sb_offset(i);
2765
		if (bytenr + BTRFS_SUPER_INFO_SIZE > scrub_dev->total_bytes)
A
Arne Jansen 已提交
2766 2767
			break;

2768
		ret = scrub_pages(sctx, bytenr, BTRFS_SUPER_INFO_SIZE, bytenr,
2769
				  scrub_dev, BTRFS_EXTENT_FLAG_SUPER, gen, i,
2770
				  NULL, 1, bytenr);
A
Arne Jansen 已提交
2771 2772 2773
		if (ret)
			return ret;
	}
2774
	wait_event(sctx->list_wait, atomic_read(&sctx->bios_in_flight) == 0);
A
Arne Jansen 已提交
2775 2776 2777 2778 2779 2780 2781

	return 0;
}

/*
 * get a reference count on fs_info->scrub_workers. start worker if necessary
 */
2782 2783
static noinline_for_stack int scrub_workers_get(struct btrfs_fs_info *fs_info,
						int is_dev_replace)
A
Arne Jansen 已提交
2784
{
2785
	int ret = 0;
A
Arne Jansen 已提交
2786 2787

	mutex_lock(&fs_info->scrub_lock);
A
Arne Jansen 已提交
2788
	if (fs_info->scrub_workers_refcnt == 0) {
2789 2790 2791 2792 2793 2794 2795
		if (is_dev_replace)
			btrfs_init_workers(&fs_info->scrub_workers, "scrub", 1,
					&fs_info->generic_worker);
		else
			btrfs_init_workers(&fs_info->scrub_workers, "scrub",
					fs_info->thread_pool_size,
					&fs_info->generic_worker);
A
Arne Jansen 已提交
2796
		fs_info->scrub_workers.idle_thresh = 4;
2797 2798 2799
		ret = btrfs_start_workers(&fs_info->scrub_workers);
		if (ret)
			goto out;
2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813
		btrfs_init_workers(&fs_info->scrub_wr_completion_workers,
				   "scrubwrc",
				   fs_info->thread_pool_size,
				   &fs_info->generic_worker);
		fs_info->scrub_wr_completion_workers.idle_thresh = 2;
		ret = btrfs_start_workers(
				&fs_info->scrub_wr_completion_workers);
		if (ret)
			goto out;
		btrfs_init_workers(&fs_info->scrub_nocow_workers, "scrubnc", 1,
				   &fs_info->generic_worker);
		ret = btrfs_start_workers(&fs_info->scrub_nocow_workers);
		if (ret)
			goto out;
A
Arne Jansen 已提交
2814
	}
A
Arne Jansen 已提交
2815
	++fs_info->scrub_workers_refcnt;
2816
out:
A
Arne Jansen 已提交
2817 2818
	mutex_unlock(&fs_info->scrub_lock);

2819
	return ret;
A
Arne Jansen 已提交
2820 2821
}

2822
static noinline_for_stack void scrub_workers_put(struct btrfs_fs_info *fs_info)
A
Arne Jansen 已提交
2823 2824
{
	mutex_lock(&fs_info->scrub_lock);
2825
	if (--fs_info->scrub_workers_refcnt == 0) {
A
Arne Jansen 已提交
2826
		btrfs_stop_workers(&fs_info->scrub_workers);
2827 2828 2829
		btrfs_stop_workers(&fs_info->scrub_wr_completion_workers);
		btrfs_stop_workers(&fs_info->scrub_nocow_workers);
	}
A
Arne Jansen 已提交
2830 2831 2832 2833
	WARN_ON(fs_info->scrub_workers_refcnt < 0);
	mutex_unlock(&fs_info->scrub_lock);
}

2834 2835
int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
		    u64 end, struct btrfs_scrub_progress *progress,
2836
		    int readonly, int is_dev_replace)
A
Arne Jansen 已提交
2837
{
2838
	struct scrub_ctx *sctx;
A
Arne Jansen 已提交
2839 2840 2841
	int ret;
	struct btrfs_device *dev;

2842
	if (btrfs_fs_closing(fs_info))
A
Arne Jansen 已提交
2843 2844 2845 2846 2847
		return -EINVAL;

	/*
	 * check some assumptions
	 */
2848
	if (fs_info->chunk_root->nodesize != fs_info->chunk_root->leafsize) {
2849 2850
		printk(KERN_ERR
		       "btrfs_scrub: size assumption nodesize == leafsize (%d == %d) fails\n",
2851 2852
		       fs_info->chunk_root->nodesize,
		       fs_info->chunk_root->leafsize);
2853 2854 2855
		return -EINVAL;
	}

2856
	if (fs_info->chunk_root->nodesize > BTRFS_STRIPE_LEN) {
2857 2858 2859 2860 2861 2862 2863
		/*
		 * in this case scrub is unable to calculate the checksum
		 * the way scrub is implemented. Do not handle this
		 * situation at all because it won't ever happen.
		 */
		printk(KERN_ERR
		       "btrfs_scrub: size assumption nodesize <= BTRFS_STRIPE_LEN (%d <= %d) fails\n",
2864
		       fs_info->chunk_root->nodesize, BTRFS_STRIPE_LEN);
2865 2866 2867
		return -EINVAL;
	}

2868
	if (fs_info->chunk_root->sectorsize != PAGE_SIZE) {
2869 2870
		/* not supported for data w/o checksums */
		printk(KERN_ERR
2871 2872
		       "btrfs_scrub: size assumption sectorsize != PAGE_SIZE (%d != %lu) fails\n",
		       fs_info->chunk_root->sectorsize, PAGE_SIZE);
A
Arne Jansen 已提交
2873 2874 2875
		return -EINVAL;
	}

2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891
	if (fs_info->chunk_root->nodesize >
	    PAGE_SIZE * SCRUB_MAX_PAGES_PER_BLOCK ||
	    fs_info->chunk_root->sectorsize >
	    PAGE_SIZE * SCRUB_MAX_PAGES_PER_BLOCK) {
		/*
		 * would exhaust the array bounds of pagev member in
		 * struct scrub_block
		 */
		pr_err("btrfs_scrub: size assumption nodesize and sectorsize <= SCRUB_MAX_PAGES_PER_BLOCK (%d <= %d && %d <= %d) fails\n",
		       fs_info->chunk_root->nodesize,
		       SCRUB_MAX_PAGES_PER_BLOCK,
		       fs_info->chunk_root->sectorsize,
		       SCRUB_MAX_PAGES_PER_BLOCK);
		return -EINVAL;
	}

2892
	ret = scrub_workers_get(fs_info, is_dev_replace);
A
Arne Jansen 已提交
2893 2894 2895
	if (ret)
		return ret;

2896 2897
	mutex_lock(&fs_info->fs_devices->device_list_mutex);
	dev = btrfs_find_device(fs_info, devid, NULL, NULL);
2898
	if (!dev || (dev->missing && !is_dev_replace)) {
2899 2900
		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
		scrub_workers_put(fs_info);
A
Arne Jansen 已提交
2901 2902 2903 2904
		return -ENODEV;
	}
	mutex_lock(&fs_info->scrub_lock);

2905
	if (!dev->in_fs_metadata || dev->is_tgtdev_for_dev_replace) {
A
Arne Jansen 已提交
2906
		mutex_unlock(&fs_info->scrub_lock);
2907 2908 2909
		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
		scrub_workers_put(fs_info);
		return -EIO;
A
Arne Jansen 已提交
2910 2911
	}

2912 2913 2914 2915 2916
	btrfs_dev_replace_lock(&fs_info->dev_replace);
	if (dev->scrub_device ||
	    (!is_dev_replace &&
	     btrfs_dev_replace_is_ongoing(&fs_info->dev_replace))) {
		btrfs_dev_replace_unlock(&fs_info->dev_replace);
A
Arne Jansen 已提交
2917
		mutex_unlock(&fs_info->scrub_lock);
2918 2919
		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
		scrub_workers_put(fs_info);
A
Arne Jansen 已提交
2920 2921
		return -EINPROGRESS;
	}
2922
	btrfs_dev_replace_unlock(&fs_info->dev_replace);
2923
	sctx = scrub_setup_ctx(dev, is_dev_replace);
2924
	if (IS_ERR(sctx)) {
A
Arne Jansen 已提交
2925
		mutex_unlock(&fs_info->scrub_lock);
2926 2927
		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
		scrub_workers_put(fs_info);
2928
		return PTR_ERR(sctx);
A
Arne Jansen 已提交
2929
	}
2930 2931
	sctx->readonly = readonly;
	dev->scrub_device = sctx;
A
Arne Jansen 已提交
2932 2933 2934

	atomic_inc(&fs_info->scrubs_running);
	mutex_unlock(&fs_info->scrub_lock);
2935
	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
A
Arne Jansen 已提交
2936

2937 2938 2939 2940 2941
	if (!is_dev_replace) {
		down_read(&fs_info->scrub_super_lock);
		ret = scrub_supers(sctx, dev);
		up_read(&fs_info->scrub_super_lock);
	}
A
Arne Jansen 已提交
2942 2943

	if (!ret)
2944 2945
		ret = scrub_enumerate_chunks(sctx, dev, start, end,
					     is_dev_replace);
A
Arne Jansen 已提交
2946

2947
	wait_event(sctx->list_wait, atomic_read(&sctx->bios_in_flight) == 0);
A
Arne Jansen 已提交
2948 2949 2950
	atomic_dec(&fs_info->scrubs_running);
	wake_up(&fs_info->scrub_pause_wait);

2951
	wait_event(sctx->list_wait, atomic_read(&sctx->workers_pending) == 0);
2952

A
Arne Jansen 已提交
2953
	if (progress)
2954
		memcpy(progress, &sctx->stat, sizeof(*progress));
A
Arne Jansen 已提交
2955 2956 2957 2958 2959

	mutex_lock(&fs_info->scrub_lock);
	dev->scrub_device = NULL;
	mutex_unlock(&fs_info->scrub_lock);

2960
	scrub_free_ctx(sctx);
2961
	scrub_workers_put(fs_info);
A
Arne Jansen 已提交
2962 2963 2964 2965

	return ret;
}

2966
void btrfs_scrub_pause(struct btrfs_root *root)
A
Arne Jansen 已提交
2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982
{
	struct btrfs_fs_info *fs_info = root->fs_info;

	mutex_lock(&fs_info->scrub_lock);
	atomic_inc(&fs_info->scrub_pause_req);
	while (atomic_read(&fs_info->scrubs_paused) !=
	       atomic_read(&fs_info->scrubs_running)) {
		mutex_unlock(&fs_info->scrub_lock);
		wait_event(fs_info->scrub_pause_wait,
			   atomic_read(&fs_info->scrubs_paused) ==
			   atomic_read(&fs_info->scrubs_running));
		mutex_lock(&fs_info->scrub_lock);
	}
	mutex_unlock(&fs_info->scrub_lock);
}

2983
void btrfs_scrub_continue(struct btrfs_root *root)
A
Arne Jansen 已提交
2984 2985 2986 2987 2988 2989 2990
{
	struct btrfs_fs_info *fs_info = root->fs_info;

	atomic_dec(&fs_info->scrub_pause_req);
	wake_up(&fs_info->scrub_pause_wait);
}

2991
void btrfs_scrub_pause_super(struct btrfs_root *root)
A
Arne Jansen 已提交
2992 2993 2994 2995
{
	down_write(&root->fs_info->scrub_super_lock);
}

2996
void btrfs_scrub_continue_super(struct btrfs_root *root)
A
Arne Jansen 已提交
2997 2998 2999 3000
{
	up_write(&root->fs_info->scrub_super_lock);
}

3001
int btrfs_scrub_cancel(struct btrfs_fs_info *fs_info)
A
Arne Jansen 已提交
3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021
{
	mutex_lock(&fs_info->scrub_lock);
	if (!atomic_read(&fs_info->scrubs_running)) {
		mutex_unlock(&fs_info->scrub_lock);
		return -ENOTCONN;
	}

	atomic_inc(&fs_info->scrub_cancel_req);
	while (atomic_read(&fs_info->scrubs_running)) {
		mutex_unlock(&fs_info->scrub_lock);
		wait_event(fs_info->scrub_pause_wait,
			   atomic_read(&fs_info->scrubs_running) == 0);
		mutex_lock(&fs_info->scrub_lock);
	}
	atomic_dec(&fs_info->scrub_cancel_req);
	mutex_unlock(&fs_info->scrub_lock);

	return 0;
}

3022 3023
int btrfs_scrub_cancel_dev(struct btrfs_fs_info *fs_info,
			   struct btrfs_device *dev)
3024
{
3025
	struct scrub_ctx *sctx;
A
Arne Jansen 已提交
3026 3027

	mutex_lock(&fs_info->scrub_lock);
3028 3029
	sctx = dev->scrub_device;
	if (!sctx) {
A
Arne Jansen 已提交
3030 3031 3032
		mutex_unlock(&fs_info->scrub_lock);
		return -ENOTCONN;
	}
3033
	atomic_inc(&sctx->cancel_req);
A
Arne Jansen 已提交
3034 3035 3036 3037 3038 3039 3040 3041 3042 3043
	while (dev->scrub_device) {
		mutex_unlock(&fs_info->scrub_lock);
		wait_event(fs_info->scrub_pause_wait,
			   dev->scrub_device == NULL);
		mutex_lock(&fs_info->scrub_lock);
	}
	mutex_unlock(&fs_info->scrub_lock);

	return 0;
}
S
Stefan Behrens 已提交
3044

A
Arne Jansen 已提交
3045 3046 3047 3048
int btrfs_scrub_progress(struct btrfs_root *root, u64 devid,
			 struct btrfs_scrub_progress *progress)
{
	struct btrfs_device *dev;
3049
	struct scrub_ctx *sctx = NULL;
A
Arne Jansen 已提交
3050 3051

	mutex_lock(&root->fs_info->fs_devices->device_list_mutex);
3052
	dev = btrfs_find_device(root->fs_info, devid, NULL, NULL);
A
Arne Jansen 已提交
3053
	if (dev)
3054 3055 3056
		sctx = dev->scrub_device;
	if (sctx)
		memcpy(progress, &sctx->stat, sizeof(*progress));
A
Arne Jansen 已提交
3057 3058
	mutex_unlock(&root->fs_info->fs_devices->device_list_mutex);

3059
	return dev ? (sctx ? 0 : -ENOTCONN) : -ENODEV;
A
Arne Jansen 已提交
3060
}
3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137

static void scrub_remap_extent(struct btrfs_fs_info *fs_info,
			       u64 extent_logical, u64 extent_len,
			       u64 *extent_physical,
			       struct btrfs_device **extent_dev,
			       int *extent_mirror_num)
{
	u64 mapped_length;
	struct btrfs_bio *bbio = NULL;
	int ret;

	mapped_length = extent_len;
	ret = btrfs_map_block(fs_info, READ, extent_logical,
			      &mapped_length, &bbio, 0);
	if (ret || !bbio || mapped_length < extent_len ||
	    !bbio->stripes[0].dev->bdev) {
		kfree(bbio);
		return;
	}

	*extent_physical = bbio->stripes[0].physical;
	*extent_mirror_num = bbio->mirror_num;
	*extent_dev = bbio->stripes[0].dev;
	kfree(bbio);
}

static int scrub_setup_wr_ctx(struct scrub_ctx *sctx,
			      struct scrub_wr_ctx *wr_ctx,
			      struct btrfs_fs_info *fs_info,
			      struct btrfs_device *dev,
			      int is_dev_replace)
{
	WARN_ON(wr_ctx->wr_curr_bio != NULL);

	mutex_init(&wr_ctx->wr_lock);
	wr_ctx->wr_curr_bio = NULL;
	if (!is_dev_replace)
		return 0;

	WARN_ON(!dev->bdev);
	wr_ctx->pages_per_wr_bio = min_t(int, SCRUB_PAGES_PER_WR_BIO,
					 bio_get_nr_vecs(dev->bdev));
	wr_ctx->tgtdev = dev;
	atomic_set(&wr_ctx->flush_all_writes, 0);
	return 0;
}

static void scrub_free_wr_ctx(struct scrub_wr_ctx *wr_ctx)
{
	mutex_lock(&wr_ctx->wr_lock);
	kfree(wr_ctx->wr_curr_bio);
	wr_ctx->wr_curr_bio = NULL;
	mutex_unlock(&wr_ctx->wr_lock);
}

static int copy_nocow_pages(struct scrub_ctx *sctx, u64 logical, u64 len,
			    int mirror_num, u64 physical_for_dev_replace)
{
	struct scrub_copy_nocow_ctx *nocow_ctx;
	struct btrfs_fs_info *fs_info = sctx->dev_root->fs_info;

	nocow_ctx = kzalloc(sizeof(*nocow_ctx), GFP_NOFS);
	if (!nocow_ctx) {
		spin_lock(&sctx->stat_lock);
		sctx->stat.malloc_errors++;
		spin_unlock(&sctx->stat_lock);
		return -ENOMEM;
	}

	scrub_pending_trans_workers_inc(sctx);

	nocow_ctx->sctx = sctx;
	nocow_ctx->logical = logical;
	nocow_ctx->len = len;
	nocow_ctx->mirror_num = mirror_num;
	nocow_ctx->physical_for_dev_replace = physical_for_dev_replace;
	nocow_ctx->work.func = copy_nocow_pages_worker;
3138
	INIT_LIST_HEAD(&nocow_ctx->inodes);
3139 3140 3141 3142 3143 3144
	btrfs_queue_worker(&fs_info->scrub_nocow_workers,
			   &nocow_ctx->work);

	return 0;
}

3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161
static int record_inode_for_nocow(u64 inum, u64 offset, u64 root, void *ctx)
{
	struct scrub_copy_nocow_ctx *nocow_ctx = ctx;
	struct scrub_nocow_inode *nocow_inode;

	nocow_inode = kzalloc(sizeof(*nocow_inode), GFP_NOFS);
	if (!nocow_inode)
		return -ENOMEM;
	nocow_inode->inum = inum;
	nocow_inode->offset = offset;
	nocow_inode->root = root;
	list_add_tail(&nocow_inode->list, &nocow_ctx->inodes);
	return 0;
}

#define COPY_COMPLETE 1

3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196
static void copy_nocow_pages_worker(struct btrfs_work *work)
{
	struct scrub_copy_nocow_ctx *nocow_ctx =
		container_of(work, struct scrub_copy_nocow_ctx, work);
	struct scrub_ctx *sctx = nocow_ctx->sctx;
	u64 logical = nocow_ctx->logical;
	u64 len = nocow_ctx->len;
	int mirror_num = nocow_ctx->mirror_num;
	u64 physical_for_dev_replace = nocow_ctx->physical_for_dev_replace;
	int ret;
	struct btrfs_trans_handle *trans = NULL;
	struct btrfs_fs_info *fs_info;
	struct btrfs_path *path;
	struct btrfs_root *root;
	int not_written = 0;

	fs_info = sctx->dev_root->fs_info;
	root = fs_info->extent_root;

	path = btrfs_alloc_path();
	if (!path) {
		spin_lock(&sctx->stat_lock);
		sctx->stat.malloc_errors++;
		spin_unlock(&sctx->stat_lock);
		not_written = 1;
		goto out;
	}

	trans = btrfs_join_transaction(root);
	if (IS_ERR(trans)) {
		not_written = 1;
		goto out;
	}

	ret = iterate_inodes_from_logical(logical, fs_info, path,
3197
					  record_inode_for_nocow, nocow_ctx);
3198
	if (ret != 0 && ret != -ENOENT) {
3199 3200 3201
		pr_warn("iterate_inodes_from_logical() failed: log %llu, phys %llu, len %llu, mir %u, ret %d\n",
			logical, physical_for_dev_replace, len, mirror_num,
			ret);
3202 3203 3204 3205
		not_written = 1;
		goto out;
	}

3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223
	btrfs_end_transaction(trans, root);
	trans = NULL;
	while (!list_empty(&nocow_ctx->inodes)) {
		struct scrub_nocow_inode *entry;
		entry = list_first_entry(&nocow_ctx->inodes,
					 struct scrub_nocow_inode,
					 list);
		list_del_init(&entry->list);
		ret = copy_nocow_pages_for_inode(entry->inum, entry->offset,
						 entry->root, nocow_ctx);
		kfree(entry);
		if (ret == COPY_COMPLETE) {
			ret = 0;
			break;
		} else if (ret) {
			break;
		}
	}
3224
out:
3225 3226 3227 3228 3229 3230 3231 3232
	while (!list_empty(&nocow_ctx->inodes)) {
		struct scrub_nocow_inode *entry;
		entry = list_first_entry(&nocow_ctx->inodes,
					 struct scrub_nocow_inode,
					 list);
		list_del_init(&entry->list);
		kfree(entry);
	}
3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244
	if (trans && !IS_ERR(trans))
		btrfs_end_transaction(trans, root);
	if (not_written)
		btrfs_dev_replace_stats_inc(&fs_info->dev_replace.
					    num_uncorrectable_read_errors);

	btrfs_free_path(path);
	kfree(nocow_ctx);

	scrub_pending_trans_workers_dec(sctx);
}

3245 3246
static int copy_nocow_pages_for_inode(u64 inum, u64 offset, u64 root,
				      struct scrub_copy_nocow_ctx *nocow_ctx)
3247
{
3248
	struct btrfs_fs_info *fs_info = nocow_ctx->sctx->dev_root->fs_info;
3249
	struct btrfs_key key;
3250 3251
	struct inode *inode;
	struct page *page;
3252
	struct btrfs_root *local_root;
3253 3254 3255 3256
	struct btrfs_ordered_extent *ordered;
	struct extent_map *em;
	struct extent_state *cached_state = NULL;
	struct extent_io_tree *io_tree;
3257
	u64 physical_for_dev_replace;
3258 3259
	u64 len = nocow_ctx->len;
	u64 lockstart = offset, lockend = offset + len - 1;
3260
	unsigned long index;
3261
	int srcu_index;
3262 3263
	int ret = 0;
	int err = 0;
3264 3265 3266 3267

	key.objectid = root;
	key.type = BTRFS_ROOT_ITEM_KEY;
	key.offset = (u64)-1;
3268 3269 3270

	srcu_index = srcu_read_lock(&fs_info->subvol_srcu);

3271
	local_root = btrfs_read_fs_root_no_name(fs_info, &key);
3272 3273
	if (IS_ERR(local_root)) {
		srcu_read_unlock(&fs_info->subvol_srcu, srcu_index);
3274
		return PTR_ERR(local_root);
3275
	}
3276 3277 3278 3279 3280

	key.type = BTRFS_INODE_ITEM_KEY;
	key.objectid = inum;
	key.offset = 0;
	inode = btrfs_iget(fs_info->sb, &key, local_root, NULL);
3281
	srcu_read_unlock(&fs_info->subvol_srcu, srcu_index);
3282 3283 3284
	if (IS_ERR(inode))
		return PTR_ERR(inode);

3285 3286 3287 3288
	/* Avoid truncate/dio/punch hole.. */
	mutex_lock(&inode->i_mutex);
	inode_dio_wait(inode);

3289
	physical_for_dev_replace = nocow_ctx->physical_for_dev_replace;
3290 3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315
	io_tree = &BTRFS_I(inode)->io_tree;

	lock_extent_bits(io_tree, lockstart, lockend, 0, &cached_state);
	ordered = btrfs_lookup_ordered_range(inode, lockstart, len);
	if (ordered) {
		btrfs_put_ordered_extent(ordered);
		goto out_unlock;
	}

	em = btrfs_get_extent(inode, NULL, 0, lockstart, len, 0);
	if (IS_ERR(em)) {
		ret = PTR_ERR(em);
		goto out_unlock;
	}

	/*
	 * This extent does not actually cover the logical extent anymore,
	 * move on to the next inode.
	 */
	if (em->block_start > nocow_ctx->logical ||
	    em->block_start + em->block_len < nocow_ctx->logical + len) {
		free_extent_map(em);
		goto out_unlock;
	}
	free_extent_map(em);

3316 3317
	while (len >= PAGE_CACHE_SIZE) {
		index = offset >> PAGE_CACHE_SHIFT;
3318
again:
3319 3320 3321 3322
		page = find_or_create_page(inode->i_mapping, index, GFP_NOFS);
		if (!page) {
			pr_err("find_or_create_page() failed\n");
			ret = -ENOMEM;
3323
			goto out;
3324 3325 3326 3327 3328 3329 3330
		}

		if (PageUptodate(page)) {
			if (PageDirty(page))
				goto next_page;
		} else {
			ClearPageError(page);
3331 3332 3333
			err = extent_read_full_page_nolock(io_tree, page,
							   btrfs_get_extent,
							   nocow_ctx->mirror_num);
3334 3335
			if (err) {
				ret = err;
3336 3337
				goto next_page;
			}
3338

3339
			lock_page(page);
3340 3341 3342 3343 3344 3345 3346
			/*
			 * If the page has been remove from the page cache,
			 * the data on it is meaningless, because it may be
			 * old one, the new data may be written into the new
			 * page in the page cache.
			 */
			if (page->mapping != inode->i_mapping) {
3347
				unlock_page(page);
3348 3349 3350
				page_cache_release(page);
				goto again;
			}
3351 3352 3353 3354 3355
			if (!PageUptodate(page)) {
				ret = -EIO;
				goto next_page;
			}
		}
3356 3357 3358 3359
		err = write_page_nocow(nocow_ctx->sctx,
				       physical_for_dev_replace, page);
		if (err)
			ret = err;
3360
next_page:
3361 3362 3363 3364 3365 3366
		unlock_page(page);
		page_cache_release(page);

		if (ret)
			break;

3367 3368 3369 3370
		offset += PAGE_CACHE_SIZE;
		physical_for_dev_replace += PAGE_CACHE_SIZE;
		len -= PAGE_CACHE_SIZE;
	}
3371 3372 3373 3374
	ret = COPY_COMPLETE;
out_unlock:
	unlock_extent_cached(io_tree, lockstart, lockend, &cached_state,
			     GFP_NOFS);
3375
out:
3376
	mutex_unlock(&inode->i_mutex);
3377
	iput(inode);
3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396
	return ret;
}

static int write_page_nocow(struct scrub_ctx *sctx,
			    u64 physical_for_dev_replace, struct page *page)
{
	struct bio *bio;
	struct btrfs_device *dev;
	int ret;
	DECLARE_COMPLETION_ONSTACK(compl);

	dev = sctx->wr_ctx.tgtdev;
	if (!dev)
		return -EIO;
	if (!dev->bdev) {
		printk_ratelimited(KERN_WARNING
			"btrfs: scrub write_page_nocow(bdev == NULL) is unexpected!\n");
		return -EIO;
	}
3397
	bio = btrfs_io_bio_alloc(GFP_NOFS, 1);
3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424
	if (!bio) {
		spin_lock(&sctx->stat_lock);
		sctx->stat.malloc_errors++;
		spin_unlock(&sctx->stat_lock);
		return -ENOMEM;
	}
	bio->bi_private = &compl;
	bio->bi_end_io = scrub_complete_bio_end_io;
	bio->bi_size = 0;
	bio->bi_sector = physical_for_dev_replace >> 9;
	bio->bi_bdev = dev->bdev;
	ret = bio_add_page(bio, page, PAGE_CACHE_SIZE, 0);
	if (ret != PAGE_CACHE_SIZE) {
leave_with_eio:
		bio_put(bio);
		btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_WRITE_ERRS);
		return -EIO;
	}
	btrfsic_submit_bio(WRITE_SYNC, bio);
	wait_for_completion(&compl);

	if (!test_bit(BIO_UPTODATE, &bio->bi_flags))
		goto leave_with_eio;

	bio_put(bio);
	return 0;
}