super.c 150.0 KB
Newer Older
1
/*
2
 *  linux/fs/ext4/super.c
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
 *
 * Copyright (C) 1992, 1993, 1994, 1995
 * Remy Card (card@masi.ibp.fr)
 * Laboratoire MASI - Institut Blaise Pascal
 * Universite Pierre et Marie Curie (Paris VI)
 *
 *  from
 *
 *  linux/fs/minix/inode.c
 *
 *  Copyright (C) 1991, 1992  Linus Torvalds
 *
 *  Big-endian to little-endian byte-swapping/bitmaps by
 *        David S. Miller (davem@caip.rutgers.edu), 1995
 */

#include <linux/module.h>
#include <linux/string.h>
#include <linux/fs.h>
#include <linux/time.h>
23
#include <linux/vmalloc.h>
24 25 26
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/blkdev.h>
27
#include <linux/backing-dev.h>
28 29
#include <linux/parser.h>
#include <linux/buffer_head.h>
30
#include <linux/exportfs.h>
31 32 33 34 35 36
#include <linux/vfs.h>
#include <linux/random.h>
#include <linux/mount.h>
#include <linux/namei.h>
#include <linux/quotaops.h>
#include <linux/seq_file.h>
T
Theodore Ts'o 已提交
37
#include <linux/ctype.h>
V
Vignesh Babu 已提交
38
#include <linux/log2.h>
A
Andreas Dilger 已提交
39
#include <linux/crc16.h>
D
Dan Magenheimer 已提交
40
#include <linux/cleancache.h>
41 42
#include <asm/uaccess.h>

43 44 45
#include <linux/kthread.h>
#include <linux/freezer.h>

46
#include "ext4.h"
47
#include "ext4_extents.h"	/* Needed for trace points definition */
48
#include "ext4_jbd2.h"
49 50
#include "xattr.h"
#include "acl.h"
51
#include "mballoc.h"
52

53 54 55
#define CREATE_TRACE_POINTS
#include <trace/events/ext4.h>

56 57
static struct ext4_lazy_init *ext4_li_info;
static struct mutex ext4_li_mtx;
58
static int ext4_mballoc_ready;
59
static struct ratelimit_state ext4_mount_msg_ratelimit;
60

61
static int ext4_load_journal(struct super_block *, struct ext4_super_block *,
62
			     unsigned long journal_devnum);
63
static int ext4_show_options(struct seq_file *seq, struct dentry *root);
64
static int ext4_commit_super(struct super_block *sb, int sync);
65 66 67 68
static void ext4_mark_recovery_complete(struct super_block *sb,
					struct ext4_super_block *es);
static void ext4_clear_journal_err(struct super_block *sb,
				   struct ext4_super_block *es);
69
static int ext4_sync_fs(struct super_block *sb, int wait);
70 71
static int ext4_remount(struct super_block *sb, int *flags, char *data);
static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf);
72 73
static int ext4_unfreeze(struct super_block *sb);
static int ext4_freeze(struct super_block *sb);
A
Al Viro 已提交
74 75
static struct dentry *ext4_mount(struct file_system_type *fs_type, int flags,
		       const char *dev_name, void *data);
76 77
static inline int ext2_feature_set_ok(struct super_block *sb);
static inline int ext3_feature_set_ok(struct super_block *sb);
78
static int ext4_feature_set_ok(struct super_block *sb, int readonly);
79 80
static void ext4_destroy_lazyinit_thread(void);
static void ext4_unregister_li_request(struct super_block *sb);
81
static void ext4_clear_request_list(void);
82

J
Jan Kara 已提交
83
#if !defined(CONFIG_EXT2_FS) && !defined(CONFIG_EXT2_FS_MODULE) && defined(CONFIG_EXT4_USE_FOR_EXT2)
84 85 86 87 88 89 90
static struct file_system_type ext2_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "ext2",
	.mount		= ext4_mount,
	.kill_sb	= kill_block_super,
	.fs_flags	= FS_REQUIRES_DEV,
};
91
MODULE_ALIAS_FS("ext2");
92
MODULE_ALIAS("ext2");
93 94 95 96 97 98
#define IS_EXT2_SB(sb) ((sb)->s_bdev->bd_holder == &ext2_fs_type)
#else
#define IS_EXT2_SB(sb) (0)
#endif


99 100 101
static struct file_system_type ext3_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "ext3",
A
Al Viro 已提交
102
	.mount		= ext4_mount,
103 104 105
	.kill_sb	= kill_block_super,
	.fs_flags	= FS_REQUIRES_DEV,
};
106
MODULE_ALIAS_FS("ext3");
107
MODULE_ALIAS("ext3");
108
#define IS_EXT3_SB(sb) ((sb)->s_bdev->bd_holder == &ext3_fs_type)
L
Laurent Vivier 已提交
109

110 111 112
static int ext4_verify_csum_type(struct super_block *sb,
				 struct ext4_super_block *es)
{
113
	if (!ext4_has_feature_metadata_csum(sb))
114 115 116 117 118
		return 1;

	return es->s_checksum_type == EXT4_CRC32C_CHKSUM;
}

119 120 121 122 123 124 125 126 127 128 129 130
static __le32 ext4_superblock_csum(struct super_block *sb,
				   struct ext4_super_block *es)
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	int offset = offsetof(struct ext4_super_block, s_checksum);
	__u32 csum;

	csum = ext4_chksum(sbi, ~0, (char *)es, offset);

	return cpu_to_le32(csum);
}

131 132
static int ext4_superblock_csum_verify(struct super_block *sb,
				       struct ext4_super_block *es)
133
{
134
	if (!ext4_has_metadata_csum(sb))
135 136 137 138 139
		return 1;

	return es->s_checksum == ext4_superblock_csum(sb, es);
}

140
void ext4_superblock_csum_set(struct super_block *sb)
141
{
142 143
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;

144
	if (!ext4_has_metadata_csum(sb))
145 146 147 148 149
		return;

	es->s_checksum = ext4_superblock_csum(sb, es);
}

150 151 152 153
void *ext4_kvmalloc(size_t size, gfp_t flags)
{
	void *ret;

154
	ret = kmalloc(size, flags | __GFP_NOWARN);
155 156 157 158 159 160 161 162 163
	if (!ret)
		ret = __vmalloc(size, flags, PAGE_KERNEL);
	return ret;
}

void *ext4_kvzalloc(size_t size, gfp_t flags)
{
	void *ret;

164
	ret = kzalloc(size, flags | __GFP_NOWARN);
165 166 167 168 169
	if (!ret)
		ret = __vmalloc(size, flags | __GFP_ZERO, PAGE_KERNEL);
	return ret;
}

170 171
ext4_fsblk_t ext4_block_bitmap(struct super_block *sb,
			       struct ext4_group_desc *bg)
L
Laurent Vivier 已提交
172
{
173
	return le32_to_cpu(bg->bg_block_bitmap_lo) |
174
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
175
		 (ext4_fsblk_t)le32_to_cpu(bg->bg_block_bitmap_hi) << 32 : 0);
L
Laurent Vivier 已提交
176 177
}

178 179
ext4_fsblk_t ext4_inode_bitmap(struct super_block *sb,
			       struct ext4_group_desc *bg)
L
Laurent Vivier 已提交
180
{
181
	return le32_to_cpu(bg->bg_inode_bitmap_lo) |
182
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
183
		 (ext4_fsblk_t)le32_to_cpu(bg->bg_inode_bitmap_hi) << 32 : 0);
L
Laurent Vivier 已提交
184 185
}

186 187
ext4_fsblk_t ext4_inode_table(struct super_block *sb,
			      struct ext4_group_desc *bg)
L
Laurent Vivier 已提交
188
{
189
	return le32_to_cpu(bg->bg_inode_table_lo) |
190
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
191
		 (ext4_fsblk_t)le32_to_cpu(bg->bg_inode_table_hi) << 32 : 0);
L
Laurent Vivier 已提交
192 193
}

194 195
__u32 ext4_free_group_clusters(struct super_block *sb,
			       struct ext4_group_desc *bg)
196 197 198
{
	return le16_to_cpu(bg->bg_free_blocks_count_lo) |
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
199
		 (__u32)le16_to_cpu(bg->bg_free_blocks_count_hi) << 16 : 0);
200 201 202 203 204 205 206
}

__u32 ext4_free_inodes_count(struct super_block *sb,
			      struct ext4_group_desc *bg)
{
	return le16_to_cpu(bg->bg_free_inodes_count_lo) |
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
207
		 (__u32)le16_to_cpu(bg->bg_free_inodes_count_hi) << 16 : 0);
208 209 210 211 212 213 214
}

__u32 ext4_used_dirs_count(struct super_block *sb,
			      struct ext4_group_desc *bg)
{
	return le16_to_cpu(bg->bg_used_dirs_count_lo) |
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
215
		 (__u32)le16_to_cpu(bg->bg_used_dirs_count_hi) << 16 : 0);
216 217 218 219 220 221 222
}

__u32 ext4_itable_unused_count(struct super_block *sb,
			      struct ext4_group_desc *bg)
{
	return le16_to_cpu(bg->bg_itable_unused_lo) |
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
223
		 (__u32)le16_to_cpu(bg->bg_itable_unused_hi) << 16 : 0);
224 225
}

226 227
void ext4_block_bitmap_set(struct super_block *sb,
			   struct ext4_group_desc *bg, ext4_fsblk_t blk)
L
Laurent Vivier 已提交
228
{
229
	bg->bg_block_bitmap_lo = cpu_to_le32((u32)blk);
230 231
	if (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT)
		bg->bg_block_bitmap_hi = cpu_to_le32(blk >> 32);
L
Laurent Vivier 已提交
232 233
}

234 235
void ext4_inode_bitmap_set(struct super_block *sb,
			   struct ext4_group_desc *bg, ext4_fsblk_t blk)
L
Laurent Vivier 已提交
236
{
237
	bg->bg_inode_bitmap_lo  = cpu_to_le32((u32)blk);
238 239
	if (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT)
		bg->bg_inode_bitmap_hi = cpu_to_le32(blk >> 32);
L
Laurent Vivier 已提交
240 241
}

242 243
void ext4_inode_table_set(struct super_block *sb,
			  struct ext4_group_desc *bg, ext4_fsblk_t blk)
L
Laurent Vivier 已提交
244
{
245
	bg->bg_inode_table_lo = cpu_to_le32((u32)blk);
246 247
	if (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT)
		bg->bg_inode_table_hi = cpu_to_le32(blk >> 32);
L
Laurent Vivier 已提交
248 249
}

250 251
void ext4_free_group_clusters_set(struct super_block *sb,
				  struct ext4_group_desc *bg, __u32 count)
252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281
{
	bg->bg_free_blocks_count_lo = cpu_to_le16((__u16)count);
	if (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT)
		bg->bg_free_blocks_count_hi = cpu_to_le16(count >> 16);
}

void ext4_free_inodes_set(struct super_block *sb,
			  struct ext4_group_desc *bg, __u32 count)
{
	bg->bg_free_inodes_count_lo = cpu_to_le16((__u16)count);
	if (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT)
		bg->bg_free_inodes_count_hi = cpu_to_le16(count >> 16);
}

void ext4_used_dirs_set(struct super_block *sb,
			  struct ext4_group_desc *bg, __u32 count)
{
	bg->bg_used_dirs_count_lo = cpu_to_le16((__u16)count);
	if (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT)
		bg->bg_used_dirs_count_hi = cpu_to_le16(count >> 16);
}

void ext4_itable_unused_set(struct super_block *sb,
			  struct ext4_group_desc *bg, __u32 count)
{
	bg->bg_itable_unused_lo = cpu_to_le16((__u16)count);
	if (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT)
		bg->bg_itable_unused_hi = cpu_to_le16(count >> 16);
}

282

283 284 285 286 287 288
static void __save_error_info(struct super_block *sb, const char *func,
			    unsigned int line)
{
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;

	EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS;
289 290
	if (bdev_read_only(sb->s_bdev))
		return;
291 292 293 294 295 296 297 298 299 300 301 302
	es->s_state |= cpu_to_le16(EXT4_ERROR_FS);
	es->s_last_error_time = cpu_to_le32(get_seconds());
	strncpy(es->s_last_error_func, func, sizeof(es->s_last_error_func));
	es->s_last_error_line = cpu_to_le32(line);
	if (!es->s_first_error_time) {
		es->s_first_error_time = es->s_last_error_time;
		strncpy(es->s_first_error_func, func,
			sizeof(es->s_first_error_func));
		es->s_first_error_line = cpu_to_le32(line);
		es->s_first_error_ino = es->s_last_error_ino;
		es->s_first_error_block = es->s_last_error_block;
	}
303 304 305 306 307 308
	/*
	 * Start the daily error reporting function if it hasn't been
	 * started already
	 */
	if (!es->s_error_count)
		mod_timer(&EXT4_SB(sb)->s_err_report, jiffies + 24*60*60*HZ);
309
	le32_add_cpu(&es->s_error_count, 1);
310 311 312 313 314 315 316 317 318
}

static void save_error_info(struct super_block *sb, const char *func,
			    unsigned int line)
{
	__save_error_info(sb, func, line);
	ext4_commit_super(sb, 1);
}

319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334
/*
 * The del_gendisk() function uninitializes the disk-specific data
 * structures, including the bdi structure, without telling anyone
 * else.  Once this happens, any attempt to call mark_buffer_dirty()
 * (for example, by ext4_commit_super), will cause a kernel OOPS.
 * This is a kludge to prevent these oops until we can put in a proper
 * hook in del_gendisk() to inform the VFS and file system layers.
 */
static int block_device_ejected(struct super_block *sb)
{
	struct inode *bd_inode = sb->s_bdev->bd_inode;
	struct backing_dev_info *bdi = inode_to_bdi(bd_inode);

	return bdi->dev == NULL;
}

B
Bobi Jam 已提交
335 336 337 338 339
static void ext4_journal_commit_callback(journal_t *journal, transaction_t *txn)
{
	struct super_block		*sb = journal->j_private;
	struct ext4_sb_info		*sbi = EXT4_SB(sb);
	int				error = is_journal_aborted(journal);
340
	struct ext4_journal_cb_entry	*jce;
B
Bobi Jam 已提交
341

342
	BUG_ON(txn->t_state == T_FINISHED);
B
Bobi Jam 已提交
343
	spin_lock(&sbi->s_md_lock);
344 345 346
	while (!list_empty(&txn->t_private_list)) {
		jce = list_entry(txn->t_private_list.next,
				 struct ext4_journal_cb_entry, jce_list);
B
Bobi Jam 已提交
347 348 349 350 351 352 353
		list_del_init(&jce->jce_list);
		spin_unlock(&sbi->s_md_lock);
		jce->jce_func(sb, jce, error);
		spin_lock(&sbi->s_md_lock);
	}
	spin_unlock(&sbi->s_md_lock);
}
354

355 356 357 358
/* Deal with the reporting of failure conditions on a filesystem such as
 * inconsistencies detected or read IO failures.
 *
 * On ext2, we can store the error state of the filesystem in the
359
 * superblock.  That is not possible on ext4, because we may have other
360 361 362 363 364
 * write ordering constraints on the superblock which prevent us from
 * writing it out straight away; and given that the journal is about to
 * be aborted, we can't rely on the current, or future, transactions to
 * write out the superblock safely.
 *
365
 * We'll just use the jbd2_journal_abort() error code to record an error in
366
 * the journal instead.  On recovery, the journal will complain about
367 368 369
 * that error until we've noted it down and cleared it.
 */

370
static void ext4_handle_error(struct super_block *sb)
371 372 373 374
{
	if (sb->s_flags & MS_RDONLY)
		return;

375
	if (!test_opt(sb, ERRORS_CONT)) {
376
		journal_t *journal = EXT4_SB(sb)->s_journal;
377

378
		EXT4_SB(sb)->s_mount_flags |= EXT4_MF_FS_ABORTED;
379
		if (journal)
380
			jbd2_journal_abort(journal, -EIO);
381
	}
382
	if (test_opt(sb, ERRORS_RO)) {
383
		ext4_msg(sb, KERN_CRIT, "Remounting filesystem read-only");
384 385 386 387 388
		/*
		 * Make sure updated value of ->s_mount_flags will be visible
		 * before ->s_flags update
		 */
		smp_wmb();
389 390
		sb->s_flags |= MS_RDONLY;
	}
391 392 393 394
	if (test_opt(sb, ERRORS_PANIC)) {
		if (EXT4_SB(sb)->s_journal &&
		  !(EXT4_SB(sb)->s_journal->j_flags & JBD2_REC_ERR))
			return;
395
		panic("EXT4-fs (device %s): panic forced after error\n",
396
			sb->s_id);
397
	}
398 399
}

400 401 402 403
#define ext4_error_ratelimit(sb)					\
		___ratelimit(&(EXT4_SB(sb)->s_err_ratelimit_state),	\
			     "EXT4-fs error")

404
void __ext4_error(struct super_block *sb, const char *function,
405
		  unsigned int line, const char *fmt, ...)
406
{
J
Joe Perches 已提交
407
	struct va_format vaf;
408 409
	va_list args;

410 411 412 413 414 415 416 417 418
	if (ext4_error_ratelimit(sb)) {
		va_start(args, fmt);
		vaf.fmt = fmt;
		vaf.va = &args;
		printk(KERN_CRIT
		       "EXT4-fs error (device %s): %s:%d: comm %s: %pV\n",
		       sb->s_id, function, line, current->comm, &vaf);
		va_end(args);
	}
419
	save_error_info(sb, function, line);
420
	ext4_handle_error(sb);
421 422
}

423 424 425
void __ext4_error_inode(struct inode *inode, const char *function,
			unsigned int line, ext4_fsblk_t block,
			const char *fmt, ...)
426 427
{
	va_list args;
428
	struct va_format vaf;
429
	struct ext4_super_block *es = EXT4_SB(inode->i_sb)->s_es;
430

431 432
	es->s_last_error_ino = cpu_to_le32(inode->i_ino);
	es->s_last_error_block = cpu_to_le64(block);
433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448
	if (ext4_error_ratelimit(inode->i_sb)) {
		va_start(args, fmt);
		vaf.fmt = fmt;
		vaf.va = &args;
		if (block)
			printk(KERN_CRIT "EXT4-fs error (device %s): %s:%d: "
			       "inode #%lu: block %llu: comm %s: %pV\n",
			       inode->i_sb->s_id, function, line, inode->i_ino,
			       block, current->comm, &vaf);
		else
			printk(KERN_CRIT "EXT4-fs error (device %s): %s:%d: "
			       "inode #%lu: comm %s: %pV\n",
			       inode->i_sb->s_id, function, line, inode->i_ino,
			       current->comm, &vaf);
		va_end(args);
	}
449
	save_error_info(inode->i_sb, function, line);
450 451 452
	ext4_handle_error(inode->i_sb);
}

453 454 455
void __ext4_error_file(struct file *file, const char *function,
		       unsigned int line, ext4_fsblk_t block,
		       const char *fmt, ...)
456 457
{
	va_list args;
458
	struct va_format vaf;
459
	struct ext4_super_block *es;
A
Al Viro 已提交
460
	struct inode *inode = file_inode(file);
461 462
	char pathname[80], *path;

463 464
	es = EXT4_SB(inode->i_sb)->s_es;
	es->s_last_error_ino = cpu_to_le32(inode->i_ino);
465
	if (ext4_error_ratelimit(inode->i_sb)) {
M
Miklos Szeredi 已提交
466
		path = file_path(file, pathname, sizeof(pathname));
467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485
		if (IS_ERR(path))
			path = "(unknown)";
		va_start(args, fmt);
		vaf.fmt = fmt;
		vaf.va = &args;
		if (block)
			printk(KERN_CRIT
			       "EXT4-fs error (device %s): %s:%d: inode #%lu: "
			       "block %llu: comm %s: path %s: %pV\n",
			       inode->i_sb->s_id, function, line, inode->i_ino,
			       block, current->comm, path, &vaf);
		else
			printk(KERN_CRIT
			       "EXT4-fs error (device %s): %s:%d: inode #%lu: "
			       "comm %s: path %s: %pV\n",
			       inode->i_sb->s_id, function, line, inode->i_ino,
			       current->comm, path, &vaf);
		va_end(args);
	}
486
	save_error_info(inode->i_sb, function, line);
487 488 489
	ext4_handle_error(inode->i_sb);
}

490 491
const char *ext4_decode_error(struct super_block *sb, int errno,
			      char nbuf[16])
492 493 494 495
{
	char *errstr = NULL;

	switch (errno) {
496 497 498 499 500 501
	case -EFSCORRUPTED:
		errstr = "Corrupt filesystem";
		break;
	case -EFSBADCRC:
		errstr = "Filesystem failed CRC";
		break;
502 503 504 505 506 507 508
	case -EIO:
		errstr = "IO failure";
		break;
	case -ENOMEM:
		errstr = "Out of memory";
		break;
	case -EROFS:
509 510
		if (!sb || (EXT4_SB(sb)->s_journal &&
			    EXT4_SB(sb)->s_journal->j_flags & JBD2_ABORT))
511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529
			errstr = "Journal has aborted";
		else
			errstr = "Readonly filesystem";
		break;
	default:
		/* If the caller passed in an extra buffer for unknown
		 * errors, textualise them now.  Else we just return
		 * NULL. */
		if (nbuf) {
			/* Check for truncated error codes... */
			if (snprintf(nbuf, 16, "error %d", -errno) >= 0)
				errstr = nbuf;
		}
		break;
	}

	return errstr;
}

530
/* __ext4_std_error decodes expected errors from journaling functions
531 532
 * automatically and invokes the appropriate error response.  */

533 534
void __ext4_std_error(struct super_block *sb, const char *function,
		      unsigned int line, int errno)
535 536 537 538 539 540 541 542 543 544 545
{
	char nbuf[16];
	const char *errstr;

	/* Special case: if the error is EROFS, and we're not already
	 * inside a transaction, then there's really no point in logging
	 * an error. */
	if (errno == -EROFS && journal_current_handle() == NULL &&
	    (sb->s_flags & MS_RDONLY))
		return;

546 547 548 549 550
	if (ext4_error_ratelimit(sb)) {
		errstr = ext4_decode_error(sb, errno, nbuf);
		printk(KERN_CRIT "EXT4-fs error (device %s) in %s:%d: %s\n",
		       sb->s_id, function, line, errstr);
	}
551

552
	save_error_info(sb, function, line);
553
	ext4_handle_error(sb);
554 555 556
}

/*
557
 * ext4_abort is a much stronger failure handler than ext4_error.  The
558 559 560 561 562 563 564 565
 * abort function may be used to deal with unrecoverable failures such
 * as journal IO errors or ENOMEM at a critical moment in log management.
 *
 * We unconditionally force the filesystem into an ABORT|READONLY state,
 * unless the error response on the fs has been set to panic in which
 * case we take the easy way out and panic immediately.
 */

566
void __ext4_abort(struct super_block *sb, const char *function,
567
		unsigned int line, const char *fmt, ...)
568 569 570
{
	va_list args;

571
	save_error_info(sb, function, line);
572
	va_start(args, fmt);
573 574
	printk(KERN_CRIT "EXT4-fs error (device %s): %s:%d: ", sb->s_id,
	       function, line);
575 576 577 578
	vprintk(fmt, args);
	printk("\n");
	va_end(args);

579 580 581
	if ((sb->s_flags & MS_RDONLY) == 0) {
		ext4_msg(sb, KERN_CRIT, "Remounting filesystem read-only");
		EXT4_SB(sb)->s_mount_flags |= EXT4_MF_FS_ABORTED;
582 583 584 585 586 587
		/*
		 * Make sure updated value of ->s_mount_flags will be visible
		 * before ->s_flags update
		 */
		smp_wmb();
		sb->s_flags |= MS_RDONLY;
588 589 590 591
		if (EXT4_SB(sb)->s_journal)
			jbd2_journal_abort(EXT4_SB(sb)->s_journal, -EIO);
		save_error_info(sb, function, line);
	}
592 593 594 595
	if (test_opt(sb, ERRORS_PANIC)) {
		if (EXT4_SB(sb)->s_journal &&
		  !(EXT4_SB(sb)->s_journal->j_flags & JBD2_REC_ERR))
			return;
596
		panic("EXT4-fs panic from previous error\n");
597
	}
598 599
}

600 601
void __ext4_msg(struct super_block *sb,
		const char *prefix, const char *fmt, ...)
602
{
J
Joe Perches 已提交
603
	struct va_format vaf;
604 605
	va_list args;

606 607 608
	if (!___ratelimit(&(EXT4_SB(sb)->s_msg_ratelimit_state), "EXT4-fs"))
		return;

609
	va_start(args, fmt);
J
Joe Perches 已提交
610 611 612
	vaf.fmt = fmt;
	vaf.va = &args;
	printk("%sEXT4-fs (%s): %pV\n", prefix, sb->s_id, &vaf);
613 614 615
	va_end(args);
}

616 617 618 619
#define ext4_warning_ratelimit(sb)					\
		___ratelimit(&(EXT4_SB(sb)->s_warning_ratelimit_state),	\
			     "EXT4-fs warning")

620
void __ext4_warning(struct super_block *sb, const char *function,
621
		    unsigned int line, const char *fmt, ...)
622
{
J
Joe Perches 已提交
623
	struct va_format vaf;
624 625
	va_list args;

626
	if (!ext4_warning_ratelimit(sb))
627 628
		return;

629
	va_start(args, fmt);
J
Joe Perches 已提交
630 631 632 633
	vaf.fmt = fmt;
	vaf.va = &args;
	printk(KERN_WARNING "EXT4-fs warning (device %s): %s:%d: %pV\n",
	       sb->s_id, function, line, &vaf);
634 635 636
	va_end(args);
}

637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654
void __ext4_warning_inode(const struct inode *inode, const char *function,
			  unsigned int line, const char *fmt, ...)
{
	struct va_format vaf;
	va_list args;

	if (!ext4_warning_ratelimit(inode->i_sb))
		return;

	va_start(args, fmt);
	vaf.fmt = fmt;
	vaf.va = &args;
	printk(KERN_WARNING "EXT4-fs warning (device %s): %s:%d: "
	       "inode #%lu: comm %s: %pV\n", inode->i_sb->s_id,
	       function, line, inode->i_ino, current->comm, &vaf);
	va_end(args);
}

655 656 657 658
void __ext4_grp_locked_error(const char *function, unsigned int line,
			     struct super_block *sb, ext4_group_t grp,
			     unsigned long ino, ext4_fsblk_t block,
			     const char *fmt, ...)
659 660 661
__releases(bitlock)
__acquires(bitlock)
{
J
Joe Perches 已提交
662
	struct va_format vaf;
663 664 665
	va_list args;
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;

666 667 668
	es->s_last_error_ino = cpu_to_le32(ino);
	es->s_last_error_block = cpu_to_le64(block);
	__save_error_info(sb, function, line);
J
Joe Perches 已提交
669

670 671 672 673 674 675 676 677 678 679 680 681 682 683
	if (ext4_error_ratelimit(sb)) {
		va_start(args, fmt);
		vaf.fmt = fmt;
		vaf.va = &args;
		printk(KERN_CRIT "EXT4-fs error (device %s): %s:%d: group %u, ",
		       sb->s_id, function, line, grp);
		if (ino)
			printk(KERN_CONT "inode %lu: ", ino);
		if (block)
			printk(KERN_CONT "block %llu:",
			       (unsigned long long) block);
		printk(KERN_CONT "%pV\n", &vaf);
		va_end(args);
	}
684 685

	if (test_opt(sb, ERRORS_CONT)) {
686
		ext4_commit_super(sb, 0);
687 688
		return;
	}
689

690 691 692 693 694 695 696 697
	ext4_unlock_group(sb, grp);
	ext4_handle_error(sb);
	/*
	 * We only get here in the ERRORS_RO case; relocking the group
	 * may be dangerous, but nothing bad will happen since the
	 * filesystem will have already been marked read/only and the
	 * journal has been aborted.  We return 1 as a hint to callers
	 * who might what to use the return value from
L
Lucas De Marchi 已提交
698
	 * ext4_grp_locked_error() to distinguish between the
699 700 701 702 703 704 705 706
	 * ERRORS_CONT and ERRORS_RO case, and perhaps return more
	 * aggressively from the ext4 function in question, with a
	 * more appropriate error code.
	 */
	ext4_lock_group(sb, grp);
	return;
}

707
void ext4_update_dynamic_rev(struct super_block *sb)
708
{
709
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;
710

711
	if (le32_to_cpu(es->s_rev_level) > EXT4_GOOD_OLD_REV)
712 713
		return;

714
	ext4_warning(sb,
715 716
		     "updating to rev %d because of new feature flag, "
		     "running e2fsck is recommended",
717
		     EXT4_DYNAMIC_REV);
718

719 720 721
	es->s_first_ino = cpu_to_le32(EXT4_GOOD_OLD_FIRST_INO);
	es->s_inode_size = cpu_to_le16(EXT4_GOOD_OLD_INODE_SIZE);
	es->s_rev_level = cpu_to_le32(EXT4_DYNAMIC_REV);
722 723 724 725 726 727 728 729 730 731 732 733 734
	/* leave es->s_feature_*compat flags alone */
	/* es->s_uuid will be set by e2fsck if empty */

	/*
	 * The rest of the superblock fields should be zero, and if not it
	 * means they are likely already in use, so leave them alone.  We
	 * can leave it up to e2fsck to clean up any inconsistencies there.
	 */
}

/*
 * Open the external journal device
 */
735
static struct block_device *ext4_blkdev_get(dev_t dev, struct super_block *sb)
736 737 738 739
{
	struct block_device *bdev;
	char b[BDEVNAME_SIZE];

740
	bdev = blkdev_get_by_dev(dev, FMODE_READ|FMODE_WRITE|FMODE_EXCL, sb);
741 742 743 744 745
	if (IS_ERR(bdev))
		goto fail;
	return bdev;

fail:
746
	ext4_msg(sb, KERN_ERR, "failed to open journal device %s: %ld",
747 748 749 750 751 752 753
			__bdevname(dev, b), PTR_ERR(bdev));
	return NULL;
}

/*
 * Release the journal device
 */
A
Al Viro 已提交
754
static void ext4_blkdev_put(struct block_device *bdev)
755
{
A
Al Viro 已提交
756
	blkdev_put(bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL);
757 758
}

A
Al Viro 已提交
759
static void ext4_blkdev_remove(struct ext4_sb_info *sbi)
760 761 762 763
{
	struct block_device *bdev;
	bdev = sbi->journal_bdev;
	if (bdev) {
A
Al Viro 已提交
764
		ext4_blkdev_put(bdev);
765 766 767 768 769 770
		sbi->journal_bdev = NULL;
	}
}

static inline struct inode *orphan_list_entry(struct list_head *l)
{
771
	return &list_entry(l, struct ext4_inode_info, i_orphan)->vfs_inode;
772 773
}

774
static void dump_orphan_list(struct super_block *sb, struct ext4_sb_info *sbi)
775 776 777
{
	struct list_head *l;

778 779
	ext4_msg(sb, KERN_ERR, "sb orphan head is %d",
		 le32_to_cpu(sbi->s_es->s_last_orphan));
780 781 782 783 784 785 786 787 788 789 790 791

	printk(KERN_ERR "sb_info orphan list:\n");
	list_for_each(l, &sbi->s_orphan) {
		struct inode *inode = orphan_list_entry(l);
		printk(KERN_ERR "  "
		       "inode %s:%lu at %p: mode %o, nlink %d, next %d\n",
		       inode->i_sb->s_id, inode->i_ino, inode,
		       inode->i_mode, inode->i_nlink,
		       NEXT_ORPHAN(inode));
	}
}

792
static void ext4_put_super(struct super_block *sb)
793
{
794 795
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_super_block *es = sbi->s_es;
796
	int i, err;
797

798
	ext4_unregister_li_request(sb);
799 800
	dquot_disable(sb, -1, DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED);

801 802
	flush_workqueue(sbi->rsv_conversion_wq);
	destroy_workqueue(sbi->rsv_conversion_wq);
803

804 805 806 807
	if (sbi->s_journal) {
		err = jbd2_journal_destroy(sbi->s_journal);
		sbi->s_journal = NULL;
		if (err < 0)
808
			ext4_abort(sb, "Couldn't clean up the journal");
809
	}
810

811
	ext4_unregister_sysfs(sb);
812
	ext4_es_unregister_shrinker(sbi);
813
	del_timer_sync(&sbi->s_err_report);
814 815 816 817 818
	ext4_release_system_zone(sb);
	ext4_mb_release(sb);
	ext4_ext_release(sb);
	ext4_xattr_put_super(sb);

819
	if (!(sb->s_flags & MS_RDONLY)) {
820
		ext4_clear_feature_journal_needs_recovery(sb);
821 822
		es->s_state = cpu_to_le16(sbi->s_mount_state);
	}
823
	if (!(sb->s_flags & MS_RDONLY))
824 825
		ext4_commit_super(sb, 1);

826 827
	for (i = 0; i < sbi->s_gdb_count; i++)
		brelse(sbi->s_group_desc[i]);
A
Al Viro 已提交
828 829
	kvfree(sbi->s_group_desc);
	kvfree(sbi->s_flex_groups);
830
	percpu_counter_destroy(&sbi->s_freeclusters_counter);
831 832
	percpu_counter_destroy(&sbi->s_freeinodes_counter);
	percpu_counter_destroy(&sbi->s_dirs_counter);
833
	percpu_counter_destroy(&sbi->s_dirtyclusters_counter);
834 835
	brelse(sbi->s_sbh);
#ifdef CONFIG_QUOTA
J
Jan Kara 已提交
836
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
837 838 839 840 841 842 843 844 845 846 847
		kfree(sbi->s_qf_names[i]);
#endif

	/* Debugging code just in case the in-memory inode orphan list
	 * isn't empty.  The on-disk one can be non-empty if we've
	 * detected an error and taken the fs readonly, but the
	 * in-memory list had better be clean by this point. */
	if (!list_empty(&sbi->s_orphan))
		dump_orphan_list(sb, sbi);
	J_ASSERT(list_empty(&sbi->s_orphan));

848
	sync_blockdev(sb->s_bdev);
849
	invalidate_bdev(sb->s_bdev);
850 851 852 853 854 855 856
	if (sbi->journal_bdev && sbi->journal_bdev != sb->s_bdev) {
		/*
		 * Invalidate the journal device's buffers.  We don't want them
		 * floating about in memory - the physical journal device may
		 * hotswapped, and it breaks the `ro-after' testing code.
		 */
		sync_blockdev(sbi->journal_bdev);
857
		invalidate_bdev(sbi->journal_bdev);
858
		ext4_blkdev_remove(sbi);
859
	}
860 861 862 863
	if (sbi->s_mb_cache) {
		ext4_xattr_destroy_cache(sbi->s_mb_cache);
		sbi->s_mb_cache = NULL;
	}
864 865
	if (sbi->s_mmp_tsk)
		kthread_stop(sbi->s_mmp_tsk);
866
	sb->s_fs_info = NULL;
T
Theodore Ts'o 已提交
867 868 869 870 871 872
	/*
	 * Now that we are completely done shutting down the
	 * superblock, we need to actually destroy the kobject.
	 */
	kobject_put(&sbi->s_kobj);
	wait_for_completion(&sbi->s_kobj_unregister);
873 874
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
875
	kfree(sbi->s_blockgroup_lock);
876 877 878
	kfree(sbi);
}

879
static struct kmem_cache *ext4_inode_cachep;
880 881 882 883

/*
 * Called inside transaction, so use GFP_NOFS
 */
884
static struct inode *ext4_alloc_inode(struct super_block *sb)
885
{
886
	struct ext4_inode_info *ei;
887

C
Christoph Lameter 已提交
888
	ei = kmem_cache_alloc(ext4_inode_cachep, GFP_NOFS);
889 890
	if (!ei)
		return NULL;
891

892
	ei->vfs_inode.i_version = 1;
893
	spin_lock_init(&ei->i_raw_lock);
894 895
	INIT_LIST_HEAD(&ei->i_prealloc_list);
	spin_lock_init(&ei->i_prealloc_lock);
Z
Zheng Liu 已提交
896 897
	ext4_es_init_tree(&ei->i_es_tree);
	rwlock_init(&ei->i_es_lock);
898
	INIT_LIST_HEAD(&ei->i_es_list);
899
	ei->i_es_all_nr = 0;
900
	ei->i_es_shk_nr = 0;
901
	ei->i_es_shrink_lblk = 0;
902 903 904
	ei->i_reserved_data_blocks = 0;
	ei->i_reserved_meta_blocks = 0;
	ei->i_allocated_meta_blocks = 0;
905
	ei->i_da_metadata_calc_len = 0;
906
	ei->i_da_metadata_calc_last_lblock = 0;
907
	spin_lock_init(&(ei->i_block_reservation_lock));
908 909
#ifdef CONFIG_QUOTA
	ei->i_reserved_quota = 0;
J
Jan Kara 已提交
910
	memset(&ei->i_dquot, 0, sizeof(ei->i_dquot));
911
#endif
912
	ei->jinode = NULL;
913
	INIT_LIST_HEAD(&ei->i_rsv_conversion_list);
914
	spin_lock_init(&ei->i_completed_io_lock);
915 916
	ei->i_sync_tid = 0;
	ei->i_datasync_tid = 0;
917
	atomic_set(&ei->i_ioend_count, 0);
918
	atomic_set(&ei->i_unwritten, 0);
919
	INIT_WORK(&ei->i_rsv_conversion_work, ext4_end_io_rsv_work);
920
#ifdef CONFIG_EXT4_FS_ENCRYPTION
921
	ei->i_crypt_info = NULL;
922
#endif
923 924 925
	return &ei->vfs_inode;
}

926 927 928 929 930 931 932 933
static int ext4_drop_inode(struct inode *inode)
{
	int drop = generic_drop_inode(inode);

	trace_ext4_drop_inode(inode, drop);
	return drop;
}

N
Nick Piggin 已提交
934 935 936 937 938 939
static void ext4_i_callback(struct rcu_head *head)
{
	struct inode *inode = container_of(head, struct inode, i_rcu);
	kmem_cache_free(ext4_inode_cachep, EXT4_I(inode));
}

940
static void ext4_destroy_inode(struct inode *inode)
941
{
942
	if (!list_empty(&(EXT4_I(inode)->i_orphan))) {
943 944 945
		ext4_msg(inode->i_sb, KERN_ERR,
			 "Inode %lu (%p): orphan list check failed!",
			 inode->i_ino, EXT4_I(inode));
946 947 948 949 950
		print_hex_dump(KERN_INFO, "", DUMP_PREFIX_ADDRESS, 16, 4,
				EXT4_I(inode), sizeof(struct ext4_inode_info),
				true);
		dump_stack();
	}
N
Nick Piggin 已提交
951
	call_rcu(&inode->i_rcu, ext4_i_callback);
952 953
}

954
static void init_once(void *foo)
955
{
956
	struct ext4_inode_info *ei = (struct ext4_inode_info *) foo;
957

C
Christoph Lameter 已提交
958 959
	INIT_LIST_HEAD(&ei->i_orphan);
	init_rwsem(&ei->xattr_sem);
960
	init_rwsem(&ei->i_data_sem);
C
Christoph Lameter 已提交
961
	inode_init_once(&ei->vfs_inode);
962 963
}

964
static int __init init_inodecache(void)
965
{
966 967
	ext4_inode_cachep = kmem_cache_create("ext4_inode_cache",
					     sizeof(struct ext4_inode_info),
968 969
					     0, (SLAB_RECLAIM_ACCOUNT|
						SLAB_MEM_SPREAD),
970
					     init_once);
971
	if (ext4_inode_cachep == NULL)
972 973 974 975 976 977
		return -ENOMEM;
	return 0;
}

static void destroy_inodecache(void)
{
978 979 980 981 982
	/*
	 * Make sure all delayed rcu free inodes are flushed before we
	 * destroy cache.
	 */
	rcu_barrier();
983
	kmem_cache_destroy(ext4_inode_cachep);
984 985
}

A
Al Viro 已提交
986
void ext4_clear_inode(struct inode *inode)
987
{
A
Al Viro 已提交
988
	invalidate_inode_buffers(inode);
989
	clear_inode(inode);
990
	dquot_drop(inode);
991
	ext4_discard_preallocations(inode);
992
	ext4_es_remove_extent(inode, 0, EXT_MAX_BLOCKS);
993 994 995 996 997 998
	if (EXT4_I(inode)->jinode) {
		jbd2_journal_release_jbd_inode(EXT4_JOURNAL(inode),
					       EXT4_I(inode)->jinode);
		jbd2_free_inode(EXT4_I(inode)->jinode);
		EXT4_I(inode)->jinode = NULL;
	}
999 1000
#ifdef CONFIG_EXT4_FS_ENCRYPTION
	if (EXT4_I(inode)->i_crypt_info)
1001
		ext4_free_encryption_info(inode, EXT4_I(inode)->i_crypt_info);
1002
#endif
1003 1004
}

C
Christoph Hellwig 已提交
1005
static struct inode *ext4_nfs_get_inode(struct super_block *sb,
1006
					u64 ino, u32 generation)
1007 1008 1009
{
	struct inode *inode;

1010
	if (ino < EXT4_FIRST_INO(sb) && ino != EXT4_ROOT_INO)
1011
		return ERR_PTR(-ESTALE);
1012
	if (ino > le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count))
1013 1014 1015 1016
		return ERR_PTR(-ESTALE);

	/* iget isn't really right if the inode is currently unallocated!!
	 *
1017
	 * ext4_read_inode will return a bad_inode if the inode had been
1018 1019 1020 1021 1022
	 * deleted, so we should be safe.
	 *
	 * Currently we don't know the generation for parent directory, so
	 * a generation of 0 means "accept any"
	 */
1023
	inode = ext4_iget_normal(sb, ino);
1024 1025 1026
	if (IS_ERR(inode))
		return ERR_CAST(inode);
	if (generation && inode->i_generation != generation) {
1027 1028 1029
		iput(inode);
		return ERR_PTR(-ESTALE);
	}
C
Christoph Hellwig 已提交
1030 1031 1032 1033 1034

	return inode;
}

static struct dentry *ext4_fh_to_dentry(struct super_block *sb, struct fid *fid,
1035
					int fh_len, int fh_type)
C
Christoph Hellwig 已提交
1036 1037 1038 1039 1040 1041
{
	return generic_fh_to_dentry(sb, fid, fh_len, fh_type,
				    ext4_nfs_get_inode);
}

static struct dentry *ext4_fh_to_parent(struct super_block *sb, struct fid *fid,
1042
					int fh_len, int fh_type)
C
Christoph Hellwig 已提交
1043 1044 1045
{
	return generic_fh_to_parent(sb, fid, fh_len, fh_type,
				    ext4_nfs_get_inode);
1046 1047
}

1048 1049 1050 1051 1052 1053
/*
 * Try to release metadata pages (indirect blocks, directories) which are
 * mapped via the block device.  Since these pages could have journal heads
 * which would prevent try_to_free_buffers() from freeing them, we must use
 * jbd2 layer's try_to_free_buffers() function to release them.
 */
1054 1055
static int bdev_try_to_free_page(struct super_block *sb, struct page *page,
				 gfp_t wait)
1056 1057 1058 1059 1060 1061 1062 1063
{
	journal_t *journal = EXT4_SB(sb)->s_journal;

	WARN_ON(PageChecked(page));
	if (!page_has_buffers(page))
		return 0;
	if (journal)
		return jbd2_journal_try_to_free_buffers(journal, page,
1064
						wait & ~__GFP_DIRECT_RECLAIM);
1065 1066 1067
	return try_to_free_buffers(page);
}

1068
#ifdef CONFIG_QUOTA
1069
#define QTYPE2NAME(t) ((t) == USRQUOTA ? "user" : "group")
1070
#define QTYPE2MOPT(on, t) ((t) == USRQUOTA?((on)##USRJQUOTA):((on)##GRPJQUOTA))
1071

1072 1073 1074 1075 1076
static int ext4_write_dquot(struct dquot *dquot);
static int ext4_acquire_dquot(struct dquot *dquot);
static int ext4_release_dquot(struct dquot *dquot);
static int ext4_mark_dquot_dirty(struct dquot *dquot);
static int ext4_write_info(struct super_block *sb, int type);
1077
static int ext4_quota_on(struct super_block *sb, int type, int format_id,
1078
			 struct path *path);
1079
static int ext4_quota_off(struct super_block *sb, int type);
1080 1081
static int ext4_quota_on_mount(struct super_block *sb, int type);
static ssize_t ext4_quota_read(struct super_block *sb, int type, char *data,
1082
			       size_t len, loff_t off);
1083
static ssize_t ext4_quota_write(struct super_block *sb, int type,
1084
				const char *data, size_t len, loff_t off);
1085 1086 1087
static int ext4_quota_enable(struct super_block *sb, int type, int format_id,
			     unsigned int flags);
static int ext4_enable_quotas(struct super_block *sb);
1088

J
Jan Kara 已提交
1089 1090 1091 1092 1093
static struct dquot **ext4_get_dquots(struct inode *inode)
{
	return EXT4_I(inode)->i_dquot;
}

1094
static const struct dquot_operations ext4_quota_operations = {
1095
	.get_reserved_space = ext4_get_reserved_space,
1096 1097 1098 1099
	.write_dquot	= ext4_write_dquot,
	.acquire_dquot	= ext4_acquire_dquot,
	.release_dquot	= ext4_release_dquot,
	.mark_dirty	= ext4_mark_dquot_dirty,
1100 1101 1102
	.write_info	= ext4_write_info,
	.alloc_dquot	= dquot_alloc,
	.destroy_dquot	= dquot_destroy,
1103 1104
};

1105
static const struct quotactl_ops ext4_qctl_operations = {
1106
	.quota_on	= ext4_quota_on,
1107
	.quota_off	= ext4_quota_off,
1108
	.quota_sync	= dquot_quota_sync,
1109
	.get_state	= dquot_get_state,
1110 1111 1112
	.set_info	= dquot_set_dqinfo,
	.get_dqblk	= dquot_get_dqblk,
	.set_dqblk	= dquot_set_dqblk
1113 1114 1115
};
#endif

1116
static const struct super_operations ext4_sops = {
1117 1118 1119 1120
	.alloc_inode	= ext4_alloc_inode,
	.destroy_inode	= ext4_destroy_inode,
	.write_inode	= ext4_write_inode,
	.dirty_inode	= ext4_dirty_inode,
1121
	.drop_inode	= ext4_drop_inode,
A
Al Viro 已提交
1122
	.evict_inode	= ext4_evict_inode,
1123 1124
	.put_super	= ext4_put_super,
	.sync_fs	= ext4_sync_fs,
1125 1126
	.freeze_fs	= ext4_freeze,
	.unfreeze_fs	= ext4_unfreeze,
1127 1128 1129
	.statfs		= ext4_statfs,
	.remount_fs	= ext4_remount,
	.show_options	= ext4_show_options,
1130
#ifdef CONFIG_QUOTA
1131 1132
	.quota_read	= ext4_quota_read,
	.quota_write	= ext4_quota_write,
J
Jan Kara 已提交
1133
	.get_dquots	= ext4_get_dquots,
1134
#endif
1135
	.bdev_try_to_free_page = bdev_try_to_free_page,
1136 1137
};

1138
static const struct export_operations ext4_export_ops = {
C
Christoph Hellwig 已提交
1139 1140
	.fh_to_dentry = ext4_fh_to_dentry,
	.fh_to_parent = ext4_fh_to_parent,
1141
	.get_parent = ext4_get_parent,
1142 1143 1144 1145 1146
};

enum {
	Opt_bsd_df, Opt_minix_df, Opt_grpid, Opt_nogrpid,
	Opt_resgid, Opt_resuid, Opt_sb, Opt_err_cont, Opt_err_panic, Opt_err_ro,
1147
	Opt_nouid32, Opt_debug, Opt_removed,
1148
	Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl,
1149
	Opt_auto_da_alloc, Opt_noauto_da_alloc, Opt_noload,
1150 1151
	Opt_commit, Opt_min_batch_time, Opt_max_batch_time, Opt_journal_dev,
	Opt_journal_path, Opt_journal_checksum, Opt_journal_async_commit,
1152
	Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback,
1153
	Opt_data_err_abort, Opt_data_err_ignore, Opt_test_dummy_encryption,
1154
	Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota,
J
Jan Kara 已提交
1155
	Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_jqfmt_vfsv1, Opt_quota,
T
Theodore Ts'o 已提交
1156
	Opt_noquota, Opt_barrier, Opt_nobarrier, Opt_err,
R
Ross Zwisler 已提交
1157
	Opt_usrquota, Opt_grpquota, Opt_i_version, Opt_dax,
1158
	Opt_stripe, Opt_delalloc, Opt_nodelalloc, Opt_mblk_io_submit,
1159
	Opt_lazytime, Opt_nolazytime,
1160
	Opt_nomblk_io_submit, Opt_block_validity, Opt_noblock_validity,
1161
	Opt_inode_readahead_blks, Opt_journal_ioprio,
1162
	Opt_dioread_nolock, Opt_dioread_lock,
1163
	Opt_discard, Opt_nodiscard, Opt_init_itable, Opt_noinit_itable,
1164
	Opt_max_dir_size_kb, Opt_nojournal_checksum,
1165 1166
};

1167
static const match_table_t tokens = {
1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181
	{Opt_bsd_df, "bsddf"},
	{Opt_minix_df, "minixdf"},
	{Opt_grpid, "grpid"},
	{Opt_grpid, "bsdgroups"},
	{Opt_nogrpid, "nogrpid"},
	{Opt_nogrpid, "sysvgroups"},
	{Opt_resgid, "resgid=%u"},
	{Opt_resuid, "resuid=%u"},
	{Opt_sb, "sb=%u"},
	{Opt_err_cont, "errors=continue"},
	{Opt_err_panic, "errors=panic"},
	{Opt_err_ro, "errors=remount-ro"},
	{Opt_nouid32, "nouid32"},
	{Opt_debug, "debug"},
1182 1183
	{Opt_removed, "oldalloc"},
	{Opt_removed, "orlov"},
1184 1185 1186 1187
	{Opt_user_xattr, "user_xattr"},
	{Opt_nouser_xattr, "nouser_xattr"},
	{Opt_acl, "acl"},
	{Opt_noacl, "noacl"},
1188
	{Opt_noload, "norecovery"},
1189
	{Opt_noload, "noload"},
1190 1191
	{Opt_removed, "nobh"},
	{Opt_removed, "bh"},
1192
	{Opt_commit, "commit=%u"},
1193 1194
	{Opt_min_batch_time, "min_batch_time=%u"},
	{Opt_max_batch_time, "max_batch_time=%u"},
1195
	{Opt_journal_dev, "journal_dev=%u"},
1196
	{Opt_journal_path, "journal_path=%s"},
1197
	{Opt_journal_checksum, "journal_checksum"},
1198
	{Opt_nojournal_checksum, "nojournal_checksum"},
1199
	{Opt_journal_async_commit, "journal_async_commit"},
1200 1201 1202 1203
	{Opt_abort, "abort"},
	{Opt_data_journal, "data=journal"},
	{Opt_data_ordered, "data=ordered"},
	{Opt_data_writeback, "data=writeback"},
1204 1205
	{Opt_data_err_abort, "data_err=abort"},
	{Opt_data_err_ignore, "data_err=ignore"},
1206 1207 1208 1209 1210 1211
	{Opt_offusrjquota, "usrjquota="},
	{Opt_usrjquota, "usrjquota=%s"},
	{Opt_offgrpjquota, "grpjquota="},
	{Opt_grpjquota, "grpjquota=%s"},
	{Opt_jqfmt_vfsold, "jqfmt=vfsold"},
	{Opt_jqfmt_vfsv0, "jqfmt=vfsv0"},
J
Jan Kara 已提交
1212
	{Opt_jqfmt_vfsv1, "jqfmt=vfsv1"},
1213 1214 1215 1216 1217
	{Opt_grpquota, "grpquota"},
	{Opt_noquota, "noquota"},
	{Opt_quota, "quota"},
	{Opt_usrquota, "usrquota"},
	{Opt_barrier, "barrier=%u"},
T
Theodore Ts'o 已提交
1218 1219
	{Opt_barrier, "barrier"},
	{Opt_nobarrier, "nobarrier"},
1220
	{Opt_i_version, "i_version"},
R
Ross Zwisler 已提交
1221
	{Opt_dax, "dax"},
1222
	{Opt_stripe, "stripe=%u"},
1223
	{Opt_delalloc, "delalloc"},
1224 1225
	{Opt_lazytime, "lazytime"},
	{Opt_nolazytime, "nolazytime"},
1226
	{Opt_nodelalloc, "nodelalloc"},
1227 1228
	{Opt_removed, "mblk_io_submit"},
	{Opt_removed, "nomblk_io_submit"},
1229 1230
	{Opt_block_validity, "block_validity"},
	{Opt_noblock_validity, "noblock_validity"},
1231
	{Opt_inode_readahead_blks, "inode_readahead_blks=%u"},
1232
	{Opt_journal_ioprio, "journal_ioprio=%u"},
1233
	{Opt_auto_da_alloc, "auto_da_alloc=%u"},
T
Theodore Ts'o 已提交
1234 1235
	{Opt_auto_da_alloc, "auto_da_alloc"},
	{Opt_noauto_da_alloc, "noauto_da_alloc"},
1236 1237
	{Opt_dioread_nolock, "dioread_nolock"},
	{Opt_dioread_lock, "dioread_lock"},
1238 1239
	{Opt_discard, "discard"},
	{Opt_nodiscard, "nodiscard"},
1240 1241 1242
	{Opt_init_itable, "init_itable=%u"},
	{Opt_init_itable, "init_itable"},
	{Opt_noinit_itable, "noinit_itable"},
1243
	{Opt_max_dir_size_kb, "max_dir_size_kb=%u"},
1244
	{Opt_test_dummy_encryption, "test_dummy_encryption"},
1245 1246 1247 1248 1249
	{Opt_removed, "check=none"},	/* mount option from ext2/3 */
	{Opt_removed, "nocheck"},	/* mount option from ext2/3 */
	{Opt_removed, "reservation"},	/* mount option from ext2/3 */
	{Opt_removed, "noreservation"}, /* mount option from ext2/3 */
	{Opt_removed, "journal=%u"},	/* mount option from ext2/3 */
J
Josef Bacik 已提交
1250
	{Opt_err, NULL},
1251 1252
};

1253
static ext4_fsblk_t get_sb_block(void **data)
1254
{
1255
	ext4_fsblk_t	sb_block;
1256 1257 1258 1259
	char		*options = (char *) *data;

	if (!options || strncmp(options, "sb=", 3) != 0)
		return 1;	/* Default location */
1260

1261
	options += 3;
1262
	/* TODO: use simple_strtoll with >32bit ext4 */
1263 1264
	sb_block = simple_strtoul(options, &options, 0);
	if (*options && *options != ',') {
1265
		printk(KERN_ERR "EXT4-fs: Invalid sb specification: %s\n",
1266 1267 1268 1269 1270 1271
		       (char *) *data);
		return 1;
	}
	if (*options == ',')
		options++;
	*data = (void *) options;
1272

1273 1274 1275
	return sb_block;
}

1276
#define DEFAULT_JOURNAL_IOPRIO (IOPRIO_PRIO_VALUE(IOPRIO_CLASS_BE, 3))
1277 1278
static char deprecated_msg[] = "Mount option \"%s\" will be removed by %s\n"
	"Contact linux-ext4@vger.kernel.org if you think we should keep it.\n";
1279

D
Dmitry Monakhov 已提交
1280 1281 1282 1283 1284
#ifdef CONFIG_QUOTA
static int set_qf_name(struct super_block *sb, int qtype, substring_t *args)
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	char *qname;
1285
	int ret = -1;
D
Dmitry Monakhov 已提交
1286 1287 1288 1289 1290 1291

	if (sb_any_quota_loaded(sb) &&
		!sbi->s_qf_names[qtype]) {
		ext4_msg(sb, KERN_ERR,
			"Cannot change journaled "
			"quota options when quota turned on");
1292
		return -1;
D
Dmitry Monakhov 已提交
1293
	}
1294
	if (ext4_has_feature_quota(sb)) {
1295 1296 1297 1298
		ext4_msg(sb, KERN_ERR, "Cannot set journaled quota options "
			 "when QUOTA feature is enabled");
		return -1;
	}
D
Dmitry Monakhov 已提交
1299 1300 1301 1302
	qname = match_strdup(args);
	if (!qname) {
		ext4_msg(sb, KERN_ERR,
			"Not enough memory for storing quotafile name");
1303
		return -1;
D
Dmitry Monakhov 已提交
1304
	}
1305 1306 1307 1308 1309 1310 1311 1312
	if (sbi->s_qf_names[qtype]) {
		if (strcmp(sbi->s_qf_names[qtype], qname) == 0)
			ret = 1;
		else
			ext4_msg(sb, KERN_ERR,
				 "%s quota file already specified",
				 QTYPE2NAME(qtype));
		goto errout;
D
Dmitry Monakhov 已提交
1313
	}
1314
	if (strchr(qname, '/')) {
D
Dmitry Monakhov 已提交
1315 1316
		ext4_msg(sb, KERN_ERR,
			"quotafile must be on filesystem root");
1317
		goto errout;
D
Dmitry Monakhov 已提交
1318
	}
1319
	sbi->s_qf_names[qtype] = qname;
1320
	set_opt(sb, QUOTA);
D
Dmitry Monakhov 已提交
1321
	return 1;
1322 1323 1324
errout:
	kfree(qname);
	return ret;
D
Dmitry Monakhov 已提交
1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335
}

static int clear_qf_name(struct super_block *sb, int qtype)
{

	struct ext4_sb_info *sbi = EXT4_SB(sb);

	if (sb_any_quota_loaded(sb) &&
		sbi->s_qf_names[qtype]) {
		ext4_msg(sb, KERN_ERR, "Cannot change journaled quota options"
			" when quota turned on");
1336
		return -1;
D
Dmitry Monakhov 已提交
1337
	}
1338
	kfree(sbi->s_qf_names[qtype]);
D
Dmitry Monakhov 已提交
1339 1340 1341 1342 1343
	sbi->s_qf_names[qtype] = NULL;
	return 1;
}
#endif

1344 1345 1346 1347 1348 1349
#define MOPT_SET	0x0001
#define MOPT_CLEAR	0x0002
#define MOPT_NOSUPPORT	0x0004
#define MOPT_EXPLICIT	0x0008
#define MOPT_CLEAR_ERR	0x0010
#define MOPT_GTE0	0x0020
1350
#ifdef CONFIG_QUOTA
1351 1352 1353 1354 1355
#define MOPT_Q		0
#define MOPT_QFMT	0x0040
#else
#define MOPT_Q		MOPT_NOSUPPORT
#define MOPT_QFMT	MOPT_NOSUPPORT
1356
#endif
1357
#define MOPT_DATAJ	0x0080
1358 1359 1360
#define MOPT_NO_EXT2	0x0100
#define MOPT_NO_EXT3	0x0200
#define MOPT_EXT4_ONLY	(MOPT_NO_EXT2 | MOPT_NO_EXT3)
1361
#define MOPT_STRING	0x0400
1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373

static const struct mount_opts {
	int	token;
	int	mount_opt;
	int	flags;
} ext4_mount_opts[] = {
	{Opt_minix_df, EXT4_MOUNT_MINIX_DF, MOPT_SET},
	{Opt_bsd_df, EXT4_MOUNT_MINIX_DF, MOPT_CLEAR},
	{Opt_grpid, EXT4_MOUNT_GRPID, MOPT_SET},
	{Opt_nogrpid, EXT4_MOUNT_GRPID, MOPT_CLEAR},
	{Opt_block_validity, EXT4_MOUNT_BLOCK_VALIDITY, MOPT_SET},
	{Opt_noblock_validity, EXT4_MOUNT_BLOCK_VALIDITY, MOPT_CLEAR},
1374 1375 1376 1377
	{Opt_dioread_nolock, EXT4_MOUNT_DIOREAD_NOLOCK,
	 MOPT_EXT4_ONLY | MOPT_SET},
	{Opt_dioread_lock, EXT4_MOUNT_DIOREAD_NOLOCK,
	 MOPT_EXT4_ONLY | MOPT_CLEAR},
1378 1379
	{Opt_discard, EXT4_MOUNT_DISCARD, MOPT_SET},
	{Opt_nodiscard, EXT4_MOUNT_DISCARD, MOPT_CLEAR},
1380 1381 1382
	{Opt_delalloc, EXT4_MOUNT_DELALLOC,
	 MOPT_EXT4_ONLY | MOPT_SET | MOPT_EXPLICIT},
	{Opt_nodelalloc, EXT4_MOUNT_DELALLOC,
1383
	 MOPT_EXT4_ONLY | MOPT_CLEAR},
1384 1385
	{Opt_nojournal_checksum, EXT4_MOUNT_JOURNAL_CHECKSUM,
	 MOPT_EXT4_ONLY | MOPT_CLEAR},
1386
	{Opt_journal_checksum, EXT4_MOUNT_JOURNAL_CHECKSUM,
1387
	 MOPT_EXT4_ONLY | MOPT_SET | MOPT_EXPLICIT},
1388
	{Opt_journal_async_commit, (EXT4_MOUNT_JOURNAL_ASYNC_COMMIT |
1389
				    EXT4_MOUNT_JOURNAL_CHECKSUM),
1390
	 MOPT_EXT4_ONLY | MOPT_SET | MOPT_EXPLICIT},
1391
	{Opt_noload, EXT4_MOUNT_NOLOAD, MOPT_NO_EXT2 | MOPT_SET},
1392 1393 1394
	{Opt_err_panic, EXT4_MOUNT_ERRORS_PANIC, MOPT_SET | MOPT_CLEAR_ERR},
	{Opt_err_ro, EXT4_MOUNT_ERRORS_RO, MOPT_SET | MOPT_CLEAR_ERR},
	{Opt_err_cont, EXT4_MOUNT_ERRORS_CONT, MOPT_SET | MOPT_CLEAR_ERR},
1395 1396 1397 1398
	{Opt_data_err_abort, EXT4_MOUNT_DATA_ERR_ABORT,
	 MOPT_NO_EXT2 | MOPT_SET},
	{Opt_data_err_ignore, EXT4_MOUNT_DATA_ERR_ABORT,
	 MOPT_NO_EXT2 | MOPT_CLEAR},
1399 1400 1401 1402 1403 1404 1405 1406 1407 1408
	{Opt_barrier, EXT4_MOUNT_BARRIER, MOPT_SET},
	{Opt_nobarrier, EXT4_MOUNT_BARRIER, MOPT_CLEAR},
	{Opt_noauto_da_alloc, EXT4_MOUNT_NO_AUTO_DA_ALLOC, MOPT_SET},
	{Opt_auto_da_alloc, EXT4_MOUNT_NO_AUTO_DA_ALLOC, MOPT_CLEAR},
	{Opt_noinit_itable, EXT4_MOUNT_INIT_INODE_TABLE, MOPT_CLEAR},
	{Opt_commit, 0, MOPT_GTE0},
	{Opt_max_batch_time, 0, MOPT_GTE0},
	{Opt_min_batch_time, 0, MOPT_GTE0},
	{Opt_inode_readahead_blks, 0, MOPT_GTE0},
	{Opt_init_itable, 0, MOPT_GTE0},
R
Ross Zwisler 已提交
1409
	{Opt_dax, EXT4_MOUNT_DAX, MOPT_SET},
1410
	{Opt_stripe, 0, MOPT_GTE0},
1411 1412
	{Opt_resuid, 0, MOPT_GTE0},
	{Opt_resgid, 0, MOPT_GTE0},
1413 1414 1415
	{Opt_journal_dev, 0, MOPT_NO_EXT2 | MOPT_GTE0},
	{Opt_journal_path, 0, MOPT_NO_EXT2 | MOPT_STRING},
	{Opt_journal_ioprio, 0, MOPT_NO_EXT2 | MOPT_GTE0},
1416 1417 1418 1419
	{Opt_data_journal, EXT4_MOUNT_JOURNAL_DATA, MOPT_NO_EXT2 | MOPT_DATAJ},
	{Opt_data_ordered, EXT4_MOUNT_ORDERED_DATA, MOPT_NO_EXT2 | MOPT_DATAJ},
	{Opt_data_writeback, EXT4_MOUNT_WRITEBACK_DATA,
	 MOPT_NO_EXT2 | MOPT_DATAJ},
1420 1421
	{Opt_user_xattr, EXT4_MOUNT_XATTR_USER, MOPT_SET},
	{Opt_nouser_xattr, EXT4_MOUNT_XATTR_USER, MOPT_CLEAR},
T
Theodore Ts'o 已提交
1422
#ifdef CONFIG_EXT4_FS_POSIX_ACL
1423 1424
	{Opt_acl, EXT4_MOUNT_POSIX_ACL, MOPT_SET},
	{Opt_noacl, EXT4_MOUNT_POSIX_ACL, MOPT_CLEAR},
1425
#else
1426 1427
	{Opt_acl, 0, MOPT_NOSUPPORT},
	{Opt_noacl, 0, MOPT_NOSUPPORT},
1428
#endif
1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444
	{Opt_nouid32, EXT4_MOUNT_NO_UID32, MOPT_SET},
	{Opt_debug, EXT4_MOUNT_DEBUG, MOPT_SET},
	{Opt_quota, EXT4_MOUNT_QUOTA | EXT4_MOUNT_USRQUOTA, MOPT_SET | MOPT_Q},
	{Opt_usrquota, EXT4_MOUNT_QUOTA | EXT4_MOUNT_USRQUOTA,
							MOPT_SET | MOPT_Q},
	{Opt_grpquota, EXT4_MOUNT_QUOTA | EXT4_MOUNT_GRPQUOTA,
							MOPT_SET | MOPT_Q},
	{Opt_noquota, (EXT4_MOUNT_QUOTA | EXT4_MOUNT_USRQUOTA |
		       EXT4_MOUNT_GRPQUOTA), MOPT_CLEAR | MOPT_Q},
	{Opt_usrjquota, 0, MOPT_Q},
	{Opt_grpjquota, 0, MOPT_Q},
	{Opt_offusrjquota, 0, MOPT_Q},
	{Opt_offgrpjquota, 0, MOPT_Q},
	{Opt_jqfmt_vfsold, QFMT_VFS_OLD, MOPT_QFMT},
	{Opt_jqfmt_vfsv0, QFMT_VFS_V0, MOPT_QFMT},
	{Opt_jqfmt_vfsv1, QFMT_VFS_V1, MOPT_QFMT},
1445
	{Opt_max_dir_size_kb, 0, MOPT_GTE0},
1446
	{Opt_test_dummy_encryption, 0, MOPT_GTE0},
1447 1448 1449 1450 1451 1452 1453 1454 1455
	{Opt_err, 0, 0}
};

static int handle_mount_opt(struct super_block *sb, char *opt, int token,
			    substring_t *args, unsigned long *journal_devnum,
			    unsigned int *journal_ioprio, int is_remount)
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	const struct mount_opts *m;
1456 1457
	kuid_t uid;
	kgid_t gid;
1458 1459
	int arg = 0;

1460 1461 1462 1463 1464 1465 1466 1467 1468 1469
#ifdef CONFIG_QUOTA
	if (token == Opt_usrjquota)
		return set_qf_name(sb, USRQUOTA, &args[0]);
	else if (token == Opt_grpjquota)
		return set_qf_name(sb, GRPQUOTA, &args[0]);
	else if (token == Opt_offusrjquota)
		return clear_qf_name(sb, USRQUOTA);
	else if (token == Opt_offgrpjquota)
		return clear_qf_name(sb, GRPQUOTA);
#endif
1470
	switch (token) {
1471 1472 1473 1474
	case Opt_noacl:
	case Opt_nouser_xattr:
		ext4_msg(sb, KERN_WARNING, deprecated_msg, opt, "3.5");
		break;
1475 1476 1477
	case Opt_sb:
		return 1;	/* handled by get_sb_block() */
	case Opt_removed:
1478
		ext4_msg(sb, KERN_WARNING, "Ignoring removed %s option", opt);
1479 1480 1481 1482 1483 1484 1485
		return 1;
	case Opt_abort:
		sbi->s_mount_flags |= EXT4_MF_FS_ABORTED;
		return 1;
	case Opt_i_version:
		sb->s_flags |= MS_I_VERSION;
		return 1;
1486 1487 1488 1489 1490 1491
	case Opt_lazytime:
		sb->s_flags |= MS_LAZYTIME;
		return 1;
	case Opt_nolazytime:
		sb->s_flags &= ~MS_LAZYTIME;
		return 1;
1492 1493
	}

1494 1495 1496 1497 1498 1499 1500 1501 1502 1503
	for (m = ext4_mount_opts; m->token != Opt_err; m++)
		if (token == m->token)
			break;

	if (m->token == Opt_err) {
		ext4_msg(sb, KERN_ERR, "Unrecognized mount option \"%s\" "
			 "or missing value", opt);
		return -1;
	}

1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514
	if ((m->flags & MOPT_NO_EXT2) && IS_EXT2_SB(sb)) {
		ext4_msg(sb, KERN_ERR,
			 "Mount option \"%s\" incompatible with ext2", opt);
		return -1;
	}
	if ((m->flags & MOPT_NO_EXT3) && IS_EXT3_SB(sb)) {
		ext4_msg(sb, KERN_ERR,
			 "Mount option \"%s\" incompatible with ext3", opt);
		return -1;
	}

1515
	if (args->from && !(m->flags & MOPT_STRING) && match_int(args, &arg))
1516 1517 1518
		return -1;
	if (args->from && (m->flags & MOPT_GTE0) && (arg < 0))
		return -1;
1519 1520 1521
	if (m->flags & MOPT_EXPLICIT) {
		if (m->mount_opt & EXT4_MOUNT_DELALLOC) {
			set_opt2(sb, EXPLICIT_DELALLOC);
1522 1523
		} else if (m->mount_opt & EXT4_MOUNT_JOURNAL_CHECKSUM) {
			set_opt2(sb, EXPLICIT_JOURNAL_CHECKSUM);
1524 1525 1526
		} else
			return -1;
	}
1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545
	if (m->flags & MOPT_CLEAR_ERR)
		clear_opt(sb, ERRORS_MASK);
	if (token == Opt_noquota && sb_any_quota_loaded(sb)) {
		ext4_msg(sb, KERN_ERR, "Cannot change quota "
			 "options when quota turned on");
		return -1;
	}

	if (m->flags & MOPT_NOSUPPORT) {
		ext4_msg(sb, KERN_ERR, "%s option not supported", opt);
	} else if (token == Opt_commit) {
		if (arg == 0)
			arg = JBD2_DEFAULT_MAX_COMMIT_AGE;
		sbi->s_commit_interval = HZ * arg;
	} else if (token == Opt_max_batch_time) {
		sbi->s_max_batch_time = arg;
	} else if (token == Opt_min_batch_time) {
		sbi->s_min_batch_time = arg;
	} else if (token == Opt_inode_readahead_blks) {
1546 1547 1548 1549
		if (arg && (arg > (1 << 30) || !is_power_of_2(arg))) {
			ext4_msg(sb, KERN_ERR,
				 "EXT4-fs: inode_readahead_blks must be "
				 "0 or a power of 2 smaller than 2^31");
1550
			return -1;
1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565
		}
		sbi->s_inode_readahead_blks = arg;
	} else if (token == Opt_init_itable) {
		set_opt(sb, INIT_INODE_TABLE);
		if (!args->from)
			arg = EXT4_DEF_LI_WAIT_MULT;
		sbi->s_li_wait_mult = arg;
	} else if (token == Opt_max_dir_size_kb) {
		sbi->s_max_dir_size_kb = arg;
	} else if (token == Opt_stripe) {
		sbi->s_stripe = arg;
	} else if (token == Opt_resuid) {
		uid = make_kuid(current_user_ns(), arg);
		if (!uid_valid(uid)) {
			ext4_msg(sb, KERN_ERR, "Invalid uid value %d", arg);
1566 1567
			return -1;
		}
1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582
		sbi->s_resuid = uid;
	} else if (token == Opt_resgid) {
		gid = make_kgid(current_user_ns(), arg);
		if (!gid_valid(gid)) {
			ext4_msg(sb, KERN_ERR, "Invalid gid value %d", arg);
			return -1;
		}
		sbi->s_resgid = gid;
	} else if (token == Opt_journal_dev) {
		if (is_remount) {
			ext4_msg(sb, KERN_ERR,
				 "Cannot specify journal on remount");
			return -1;
		}
		*journal_devnum = arg;
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
	} else if (token == Opt_journal_path) {
		char *journal_path;
		struct inode *journal_inode;
		struct path path;
		int error;

		if (is_remount) {
			ext4_msg(sb, KERN_ERR,
				 "Cannot specify journal on remount");
			return -1;
		}
		journal_path = match_strdup(&args[0]);
		if (!journal_path) {
			ext4_msg(sb, KERN_ERR, "error: could not dup "
				"journal device string");
			return -1;
		}

		error = kern_path(journal_path, LOOKUP_FOLLOW, &path);
		if (error) {
			ext4_msg(sb, KERN_ERR, "error: could not find "
				"journal device path: error %d", error);
			kfree(journal_path);
			return -1;
		}

1609
		journal_inode = d_inode(path.dentry);
1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620
		if (!S_ISBLK(journal_inode->i_mode)) {
			ext4_msg(sb, KERN_ERR, "error: journal path %s "
				"is not a block device", journal_path);
			path_put(&path);
			kfree(journal_path);
			return -1;
		}

		*journal_devnum = new_encode_dev(journal_inode->i_rdev);
		path_put(&path);
		kfree(journal_path);
1621 1622 1623 1624 1625 1626 1627 1628
	} else if (token == Opt_journal_ioprio) {
		if (arg > 7) {
			ext4_msg(sb, KERN_ERR, "Invalid journal IO priority"
				 " (must be 0-7)");
			return -1;
		}
		*journal_ioprio =
			IOPRIO_PRIO_VALUE(IOPRIO_CLASS_BE, arg);
1629 1630 1631 1632 1633 1634 1635 1636 1637
	} else if (token == Opt_test_dummy_encryption) {
#ifdef CONFIG_EXT4_FS_ENCRYPTION
		sbi->s_mount_flags |= EXT4_MF_TEST_DUMMY_ENCRYPTION;
		ext4_msg(sb, KERN_WARNING,
			 "Test dummy encryption mode enabled");
#else
		ext4_msg(sb, KERN_WARNING,
			 "Test dummy encryption mount option ignored");
#endif
1638 1639 1640 1641 1642
	} else if (m->flags & MOPT_DATAJ) {
		if (is_remount) {
			if (!sbi->s_journal)
				ext4_msg(sb, KERN_WARNING, "Remounting file system with no journal so ignoring journalled data option");
			else if (test_opt(sb, DATA_FLAGS) != m->mount_opt) {
1643
				ext4_msg(sb, KERN_ERR,
1644 1645
					 "Cannot change data mode on remount");
				return -1;
1646
			}
1647
		} else {
1648 1649
			clear_opt(sb, DATA_FLAGS);
			sbi->s_mount_opt |= m->mount_opt;
1650
		}
1651 1652 1653 1654 1655 1656 1657 1658
#ifdef CONFIG_QUOTA
	} else if (m->flags & MOPT_QFMT) {
		if (sb_any_quota_loaded(sb) &&
		    sbi->s_jquota_fmt != m->mount_opt) {
			ext4_msg(sb, KERN_ERR, "Cannot change journaled "
				 "quota options when quota turned on");
			return -1;
		}
1659
		if (ext4_has_feature_quota(sb)) {
1660 1661 1662 1663 1664
			ext4_msg(sb, KERN_ERR,
				 "Cannot set journaled quota options "
				 "when QUOTA feature is enabled");
			return -1;
		}
1665
		sbi->s_jquota_fmt = m->mount_opt;
R
Ross Zwisler 已提交
1666 1667 1668 1669 1670
#endif
#ifndef CONFIG_FS_DAX
	} else if (token == Opt_dax) {
		ext4_msg(sb, KERN_INFO, "dax option not supported");
		return -1;
1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686
#endif
	} else {
		if (!args->from)
			arg = 1;
		if (m->flags & MOPT_CLEAR)
			arg = !arg;
		else if (unlikely(!(m->flags & MOPT_SET))) {
			ext4_msg(sb, KERN_WARNING,
				 "buggy handling of option %s", opt);
			WARN_ON(1);
			return -1;
		}
		if (arg != 0)
			sbi->s_mount_opt |= m->mount_opt;
		else
			sbi->s_mount_opt &= ~m->mount_opt;
1687
	}
1688
	return 1;
1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710
}

static int parse_options(char *options, struct super_block *sb,
			 unsigned long *journal_devnum,
			 unsigned int *journal_ioprio,
			 int is_remount)
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	char *p;
	substring_t args[MAX_OPT_ARGS];
	int token;

	if (!options)
		return 1;

	while ((p = strsep(&options, ",")) != NULL) {
		if (!*p)
			continue;
		/*
		 * Initialize args struct so we know whether arg was
		 * found; some options take optional arguments.
		 */
1711
		args[0].to = args[0].from = NULL;
1712 1713 1714 1715
		token = match_token(p, tokens, args);
		if (handle_mount_opt(sb, p, token, args, journal_devnum,
				     journal_ioprio, is_remount) < 0)
			return 0;
1716 1717
	}
#ifdef CONFIG_QUOTA
1718
	if (ext4_has_feature_quota(sb) &&
1719 1720 1721 1722 1723
	    (test_opt(sb, USRQUOTA) || test_opt(sb, GRPQUOTA))) {
		ext4_msg(sb, KERN_ERR, "Cannot set quota options when QUOTA "
			 "feature is enabled");
		return 0;
	}
1724
	if (sbi->s_qf_names[USRQUOTA] || sbi->s_qf_names[GRPQUOTA]) {
1725
		if (test_opt(sb, USRQUOTA) && sbi->s_qf_names[USRQUOTA])
1726
			clear_opt(sb, USRQUOTA);
1727

1728
		if (test_opt(sb, GRPQUOTA) && sbi->s_qf_names[GRPQUOTA])
1729
			clear_opt(sb, GRPQUOTA);
1730

D
Dmitry Monakhov 已提交
1731
		if (test_opt(sb, GRPQUOTA) || test_opt(sb, USRQUOTA)) {
1732 1733
			ext4_msg(sb, KERN_ERR, "old and new quota "
					"format mixing");
1734 1735 1736 1737
			return 0;
		}

		if (!sbi->s_jquota_fmt) {
1738 1739
			ext4_msg(sb, KERN_ERR, "journaled quota format "
					"not specified");
1740 1741 1742 1743
			return 0;
		}
	}
#endif
J
Jan Kara 已提交
1744 1745 1746 1747 1748 1749 1750 1751 1752 1753
	if (test_opt(sb, DIOREAD_NOLOCK)) {
		int blocksize =
			BLOCK_SIZE << le32_to_cpu(sbi->s_es->s_log_block_size);

		if (blocksize < PAGE_CACHE_SIZE) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "dioread_nolock if block size != PAGE_SIZE");
			return 0;
		}
	}
1754 1755 1756 1757 1758 1759
	if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA &&
	    test_opt(sb, JOURNAL_ASYNC_COMMIT)) {
		ext4_msg(sb, KERN_ERR, "can't mount with journal_async_commit "
			 "in data=ordered mode");
		return 0;
	}
1760 1761 1762
	return 1;
}

1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786
static inline void ext4_show_quota_options(struct seq_file *seq,
					   struct super_block *sb)
{
#if defined(CONFIG_QUOTA)
	struct ext4_sb_info *sbi = EXT4_SB(sb);

	if (sbi->s_jquota_fmt) {
		char *fmtname = "";

		switch (sbi->s_jquota_fmt) {
		case QFMT_VFS_OLD:
			fmtname = "vfsold";
			break;
		case QFMT_VFS_V0:
			fmtname = "vfsv0";
			break;
		case QFMT_VFS_V1:
			fmtname = "vfsv1";
			break;
		}
		seq_printf(seq, ",jqfmt=%s", fmtname);
	}

	if (sbi->s_qf_names[USRQUOTA])
1787
		seq_show_option(seq, "usrjquota", sbi->s_qf_names[USRQUOTA]);
1788 1789

	if (sbi->s_qf_names[GRPQUOTA])
1790
		seq_show_option(seq, "grpjquota", sbi->s_qf_names[GRPQUOTA]);
1791 1792 1793
#endif
}

1794 1795
static const char *token2str(int token)
{
1796
	const struct match_token *t;
1797 1798 1799 1800 1801 1802 1803

	for (t = tokens; t->token != Opt_err; t++)
		if (t->token == token && !strchr(t->pattern, '='))
			break;
	return t->pattern;
}

1804 1805 1806 1807 1808
/*
 * Show an option if
 *  - it's set to a non-default value OR
 *  - if the per-sb default is different from the global default
 */
1809 1810
static int _ext4_show_options(struct seq_file *seq, struct super_block *sb,
			      int nodefs)
1811 1812 1813
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_super_block *es = sbi->s_es;
1814
	int def_errors, def_mount_opt = nodefs ? 0 : sbi->s_def_mount_opt;
1815
	const struct mount_opts *m;
1816
	char sep = nodefs ? '\n' : ',';
1817

1818 1819
#define SEQ_OPTS_PUTS(str) seq_printf(seq, "%c" str, sep)
#define SEQ_OPTS_PRINT(str, arg) seq_printf(seq, "%c" str, sep, arg)
1820 1821

	if (sbi->s_sb_block != 1)
1822 1823 1824 1825 1826 1827 1828
		SEQ_OPTS_PRINT("sb=%llu", sbi->s_sb_block);

	for (m = ext4_mount_opts; m->token != Opt_err; m++) {
		int want_set = m->flags & MOPT_SET;
		if (((m->flags & (MOPT_SET|MOPT_CLEAR)) == 0) ||
		    (m->flags & MOPT_CLEAR_ERR))
			continue;
1829
		if (!(m->mount_opt & (sbi->s_mount_opt ^ def_mount_opt)))
1830 1831 1832 1833 1834 1835
			continue; /* skip if same as the default */
		if ((want_set &&
		     (sbi->s_mount_opt & m->mount_opt) != m->mount_opt) ||
		    (!want_set && (sbi->s_mount_opt & m->mount_opt)))
			continue; /* select Opt_noFoo vs Opt_Foo */
		SEQ_OPTS_PRINT("%s", token2str(m->token));
1836
	}
1837

1838
	if (nodefs || !uid_eq(sbi->s_resuid, make_kuid(&init_user_ns, EXT4_DEF_RESUID)) ||
1839
	    le16_to_cpu(es->s_def_resuid) != EXT4_DEF_RESUID)
1840 1841 1842
		SEQ_OPTS_PRINT("resuid=%u",
				from_kuid_munged(&init_user_ns, sbi->s_resuid));
	if (nodefs || !gid_eq(sbi->s_resgid, make_kgid(&init_user_ns, EXT4_DEF_RESGID)) ||
1843
	    le16_to_cpu(es->s_def_resgid) != EXT4_DEF_RESGID)
1844 1845
		SEQ_OPTS_PRINT("resgid=%u",
				from_kgid_munged(&init_user_ns, sbi->s_resgid));
1846
	def_errors = nodefs ? -1 : le16_to_cpu(es->s_errors);
1847 1848
	if (test_opt(sb, ERRORS_RO) && def_errors != EXT4_ERRORS_RO)
		SEQ_OPTS_PUTS("errors=remount-ro");
1849
	if (test_opt(sb, ERRORS_CONT) && def_errors != EXT4_ERRORS_CONTINUE)
1850
		SEQ_OPTS_PUTS("errors=continue");
1851
	if (test_opt(sb, ERRORS_PANIC) && def_errors != EXT4_ERRORS_PANIC)
1852
		SEQ_OPTS_PUTS("errors=panic");
1853
	if (nodefs || sbi->s_commit_interval != JBD2_DEFAULT_MAX_COMMIT_AGE*HZ)
1854
		SEQ_OPTS_PRINT("commit=%lu", sbi->s_commit_interval / HZ);
1855
	if (nodefs || sbi->s_min_batch_time != EXT4_DEF_MIN_BATCH_TIME)
1856
		SEQ_OPTS_PRINT("min_batch_time=%u", sbi->s_min_batch_time);
1857
	if (nodefs || sbi->s_max_batch_time != EXT4_DEF_MAX_BATCH_TIME)
1858
		SEQ_OPTS_PRINT("max_batch_time=%u", sbi->s_max_batch_time);
1859
	if (sb->s_flags & MS_I_VERSION)
1860
		SEQ_OPTS_PUTS("i_version");
1861
	if (nodefs || sbi->s_stripe)
1862
		SEQ_OPTS_PRINT("stripe=%lu", sbi->s_stripe);
1863
	if (EXT4_MOUNT_DATA_FLAGS & (sbi->s_mount_opt ^ def_mount_opt)) {
1864 1865 1866 1867 1868 1869 1870
		if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA)
			SEQ_OPTS_PUTS("data=journal");
		else if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA)
			SEQ_OPTS_PUTS("data=ordered");
		else if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA)
			SEQ_OPTS_PUTS("data=writeback");
	}
1871 1872
	if (nodefs ||
	    sbi->s_inode_readahead_blks != EXT4_DEF_INODE_READAHEAD_BLKS)
1873 1874
		SEQ_OPTS_PRINT("inode_readahead_blks=%u",
			       sbi->s_inode_readahead_blks);
1875

1876 1877
	if (nodefs || (test_opt(sb, INIT_INODE_TABLE) &&
		       (sbi->s_li_wait_mult != EXT4_DEF_LI_WAIT_MULT)))
1878
		SEQ_OPTS_PRINT("init_itable=%u", sbi->s_li_wait_mult);
1879 1880
	if (nodefs || sbi->s_max_dir_size_kb)
		SEQ_OPTS_PRINT("max_dir_size_kb=%u", sbi->s_max_dir_size_kb);
1881 1882 1883 1884 1885

	ext4_show_quota_options(seq, sb);
	return 0;
}

1886 1887 1888 1889 1890
static int ext4_show_options(struct seq_file *seq, struct dentry *root)
{
	return _ext4_show_options(seq, root->d_sb, 0);
}

1891
int ext4_seq_options_show(struct seq_file *seq, void *offset)
1892 1893 1894 1895 1896 1897 1898 1899 1900 1901
{
	struct super_block *sb = seq->private;
	int rc;

	seq_puts(seq, (sb->s_flags & MS_RDONLY) ? "ro" : "rw");
	rc = _ext4_show_options(seq, sb, 1);
	seq_puts(seq, "\n");
	return rc;
}

1902
static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es,
1903 1904
			    int read_only)
{
1905
	struct ext4_sb_info *sbi = EXT4_SB(sb);
1906 1907
	int res = 0;

1908
	if (le32_to_cpu(es->s_rev_level) > EXT4_MAX_SUPP_REV) {
1909 1910
		ext4_msg(sb, KERN_ERR, "revision level too high, "
			 "forcing read-only mode");
1911 1912 1913
		res = MS_RDONLY;
	}
	if (read_only)
1914
		goto done;
1915
	if (!(sbi->s_mount_state & EXT4_VALID_FS))
1916 1917
		ext4_msg(sb, KERN_WARNING, "warning: mounting unchecked fs, "
			 "running e2fsck is recommended");
1918
	else if (sbi->s_mount_state & EXT4_ERROR_FS)
1919 1920 1921
		ext4_msg(sb, KERN_WARNING,
			 "warning: mounting fs with errors, "
			 "running e2fsck is recommended");
1922
	else if ((__s16) le16_to_cpu(es->s_max_mnt_count) > 0 &&
1923 1924
		 le16_to_cpu(es->s_mnt_count) >=
		 (unsigned short) (__s16) le16_to_cpu(es->s_max_mnt_count))
1925 1926 1927
		ext4_msg(sb, KERN_WARNING,
			 "warning: maximal mount count reached, "
			 "running e2fsck is recommended");
1928 1929 1930
	else if (le32_to_cpu(es->s_checkinterval) &&
		(le32_to_cpu(es->s_lastcheck) +
			le32_to_cpu(es->s_checkinterval) <= get_seconds()))
1931 1932 1933
		ext4_msg(sb, KERN_WARNING,
			 "warning: checktime reached, "
			 "running e2fsck is recommended");
1934
	if (!sbi->s_journal)
1935
		es->s_state &= cpu_to_le16(~EXT4_VALID_FS);
1936
	if (!(__s16) le16_to_cpu(es->s_max_mnt_count))
1937
		es->s_max_mnt_count = cpu_to_le16(EXT4_DFL_MAX_MNT_COUNT);
M
Marcin Slusarz 已提交
1938
	le16_add_cpu(&es->s_mnt_count, 1);
1939
	es->s_mtime = cpu_to_le32(get_seconds());
1940
	ext4_update_dynamic_rev(sb);
1941
	if (sbi->s_journal)
1942
		ext4_set_feature_journal_needs_recovery(sb);
1943

1944
	ext4_commit_super(sb, 1);
1945
done:
1946
	if (test_opt(sb, DEBUG))
1947
		printk(KERN_INFO "[EXT4 FS bs=%lu, gc=%u, "
1948
				"bpg=%lu, ipg=%lu, mo=%04x, mo2=%04x]\n",
1949 1950
			sb->s_blocksize,
			sbi->s_groups_count,
1951 1952
			EXT4_BLOCKS_PER_GROUP(sb),
			EXT4_INODES_PER_GROUP(sb),
1953
			sbi->s_mount_opt, sbi->s_mount_opt2);
1954

D
Dan Magenheimer 已提交
1955
	cleancache_init_fs(sb);
1956 1957 1958
	return res;
}

1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983
int ext4_alloc_flex_bg_array(struct super_block *sb, ext4_group_t ngroup)
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct flex_groups *new_groups;
	int size;

	if (!sbi->s_log_groups_per_flex)
		return 0;

	size = ext4_flex_group(sbi, ngroup - 1) + 1;
	if (size <= sbi->s_flex_groups_allocated)
		return 0;

	size = roundup_pow_of_two(size * sizeof(struct flex_groups));
	new_groups = ext4_kvzalloc(size, GFP_KERNEL);
	if (!new_groups) {
		ext4_msg(sb, KERN_ERR, "not enough memory for %d flex groups",
			 size / (int) sizeof(struct flex_groups));
		return -ENOMEM;
	}

	if (sbi->s_flex_groups) {
		memcpy(new_groups, sbi->s_flex_groups,
		       (sbi->s_flex_groups_allocated *
			sizeof(struct flex_groups)));
A
Al Viro 已提交
1984
		kvfree(sbi->s_flex_groups);
1985 1986 1987 1988 1989 1990
	}
	sbi->s_flex_groups = new_groups;
	sbi->s_flex_groups_allocated = size / sizeof(struct flex_groups);
	return 0;
}

1991 1992 1993 1994 1995
static int ext4_fill_flex_info(struct super_block *sb)
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_group_desc *gdp = NULL;
	ext4_group_t flex_group;
1996
	int i, err;
1997

1998
	sbi->s_log_groups_per_flex = sbi->s_es->s_log_groups_per_flex;
1999
	if (sbi->s_log_groups_per_flex < 1 || sbi->s_log_groups_per_flex > 31) {
2000 2001 2002 2003
		sbi->s_log_groups_per_flex = 0;
		return 1;
	}

2004 2005
	err = ext4_alloc_flex_bg_array(sb, sbi->s_groups_count);
	if (err)
2006
		goto failed;
2007 2008

	for (i = 0; i < sbi->s_groups_count; i++) {
2009
		gdp = ext4_get_group_desc(sb, i, NULL);
2010 2011

		flex_group = ext4_flex_group(sbi, i);
2012 2013
		atomic_add(ext4_free_inodes_count(sb, gdp),
			   &sbi->s_flex_groups[flex_group].free_inodes);
2014 2015
		atomic64_add(ext4_free_group_clusters(sb, gdp),
			     &sbi->s_flex_groups[flex_group].free_clusters);
2016 2017
		atomic_add(ext4_used_dirs_count(sb, gdp),
			   &sbi->s_flex_groups[flex_group].used_dirs);
2018 2019 2020 2021 2022 2023 2024
	}

	return 1;
failed:
	return 0;
}

2025
static __le16 ext4_group_desc_csum(struct super_block *sb, __u32 block_group,
2026
				   struct ext4_group_desc *gdp)
A
Andreas Dilger 已提交
2027
{
2028
	int offset;
A
Andreas Dilger 已提交
2029
	__u16 crc = 0;
2030
	__le32 le_group = cpu_to_le32(block_group);
2031
	struct ext4_sb_info *sbi = EXT4_SB(sb);
A
Andreas Dilger 已提交
2032

2033
	if (ext4_has_metadata_csum(sbi->s_sb)) {
2034
		/* Use new metadata_csum algorithm */
2035
		__le16 save_csum;
2036 2037
		__u32 csum32;

2038
		save_csum = gdp->bg_checksum;
2039 2040 2041 2042 2043
		gdp->bg_checksum = 0;
		csum32 = ext4_chksum(sbi, sbi->s_csum_seed, (__u8 *)&le_group,
				     sizeof(le_group));
		csum32 = ext4_chksum(sbi, csum32, (__u8 *)gdp,
				     sbi->s_desc_size);
2044
		gdp->bg_checksum = save_csum;
2045 2046 2047

		crc = csum32 & 0xFFFF;
		goto out;
A
Andreas Dilger 已提交
2048 2049
	}

2050
	/* old crc16 code */
2051
	if (!ext4_has_feature_gdt_csum(sb))
2052 2053
		return 0;

2054 2055 2056 2057 2058 2059 2060
	offset = offsetof(struct ext4_group_desc, bg_checksum);

	crc = crc16(~0, sbi->s_es->s_uuid, sizeof(sbi->s_es->s_uuid));
	crc = crc16(crc, (__u8 *)&le_group, sizeof(le_group));
	crc = crc16(crc, (__u8 *)gdp, offset);
	offset += sizeof(gdp->bg_checksum); /* skip checksum */
	/* for checksum of struct ext4_group_desc do the rest...*/
2061
	if (ext4_has_feature_64bit(sb) &&
2062 2063 2064 2065 2066 2067
	    offset < le16_to_cpu(sbi->s_es->s_desc_size))
		crc = crc16(crc, (__u8 *)gdp + offset,
			    le16_to_cpu(sbi->s_es->s_desc_size) -
				offset);

out:
A
Andreas Dilger 已提交
2068 2069 2070
	return cpu_to_le16(crc);
}

2071
int ext4_group_desc_csum_verify(struct super_block *sb, __u32 block_group,
A
Andreas Dilger 已提交
2072 2073
				struct ext4_group_desc *gdp)
{
2074
	if (ext4_has_group_desc_csum(sb) &&
2075
	    (gdp->bg_checksum != ext4_group_desc_csum(sb, block_group, gdp)))
A
Andreas Dilger 已提交
2076 2077 2078 2079 2080
		return 0;

	return 1;
}

2081 2082 2083 2084 2085
void ext4_group_desc_csum_set(struct super_block *sb, __u32 block_group,
			      struct ext4_group_desc *gdp)
{
	if (!ext4_has_group_desc_csum(sb))
		return;
2086
	gdp->bg_checksum = ext4_group_desc_csum(sb, block_group, gdp);
2087 2088
}

2089
/* Called at mount-time, super-block is locked */
2090 2091
static int ext4_check_descriptors(struct super_block *sb,
				  ext4_group_t *first_not_zeroed)
2092
{
2093 2094 2095
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	ext4_fsblk_t first_block = le32_to_cpu(sbi->s_es->s_first_data_block);
	ext4_fsblk_t last_block;
L
Laurent Vivier 已提交
2096 2097 2098
	ext4_fsblk_t block_bitmap;
	ext4_fsblk_t inode_bitmap;
	ext4_fsblk_t inode_table;
J
Jose R. Santos 已提交
2099
	int flexbg_flag = 0;
2100
	ext4_group_t i, grp = sbi->s_groups_count;
2101

2102
	if (ext4_has_feature_flex_bg(sb))
J
Jose R. Santos 已提交
2103 2104
		flexbg_flag = 1;

2105
	ext4_debug("Checking group descriptors");
2106

2107 2108 2109
	for (i = 0; i < sbi->s_groups_count; i++) {
		struct ext4_group_desc *gdp = ext4_get_group_desc(sb, i, NULL);

J
Jose R. Santos 已提交
2110
		if (i == sbi->s_groups_count - 1 || flexbg_flag)
L
Laurent Vivier 已提交
2111
			last_block = ext4_blocks_count(sbi->s_es) - 1;
2112 2113
		else
			last_block = first_block +
2114
				(EXT4_BLOCKS_PER_GROUP(sb) - 1);
2115

2116 2117 2118 2119
		if ((grp == sbi->s_groups_count) &&
		   !(gdp->bg_flags & cpu_to_le16(EXT4_BG_INODE_ZEROED)))
			grp = i;

2120
		block_bitmap = ext4_block_bitmap(sb, gdp);
2121
		if (block_bitmap < first_block || block_bitmap > last_block) {
2122
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
2123
			       "Block bitmap for group %u not in group "
2124
			       "(block %llu)!", i, block_bitmap);
2125 2126
			return 0;
		}
2127
		inode_bitmap = ext4_inode_bitmap(sb, gdp);
2128
		if (inode_bitmap < first_block || inode_bitmap > last_block) {
2129
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
2130
			       "Inode bitmap for group %u not in group "
2131
			       "(block %llu)!", i, inode_bitmap);
2132 2133
			return 0;
		}
2134
		inode_table = ext4_inode_table(sb, gdp);
L
Laurent Vivier 已提交
2135
		if (inode_table < first_block ||
2136
		    inode_table + sbi->s_itb_per_group - 1 > last_block) {
2137
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
2138
			       "Inode table for group %u not in group "
2139
			       "(block %llu)!", i, inode_table);
2140 2141
			return 0;
		}
2142
		ext4_lock_group(sb, i);
2143
		if (!ext4_group_desc_csum_verify(sb, i, gdp)) {
2144 2145
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
				 "Checksum for group %u failed (%u!=%u)",
2146
				 i, le16_to_cpu(ext4_group_desc_csum(sb, i,
2147
				     gdp)), le16_to_cpu(gdp->bg_checksum));
2148
			if (!(sb->s_flags & MS_RDONLY)) {
2149
				ext4_unlock_group(sb, i);
2150
				return 0;
2151
			}
A
Andreas Dilger 已提交
2152
		}
2153
		ext4_unlock_group(sb, i);
J
Jose R. Santos 已提交
2154 2155
		if (!flexbg_flag)
			first_block += EXT4_BLOCKS_PER_GROUP(sb);
2156
	}
2157 2158
	if (NULL != first_not_zeroed)
		*first_not_zeroed = grp;
2159 2160 2161
	return 1;
}

2162
/* ext4_orphan_cleanup() walks a singly-linked list of inodes (starting at
2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174
 * the superblock) which were deleted from all directories, but held open by
 * a process at the time of a crash.  We walk the list and try to delete these
 * inodes at recovery time (only with a read-write filesystem).
 *
 * In order to keep the orphan inode chain consistent during traversal (in
 * case of crash during recovery), we link each inode into the superblock
 * orphan list_head and handle it the same way as an inode deletion during
 * normal operation (which journals the operations for us).
 *
 * We only do an iget() and an iput() on each inode, which is very safe if we
 * accidentally point at an in-use or already deleted inode.  The worst that
 * can happen in this case is that we get a "bit already cleared" message from
2175
 * ext4_free_inode().  The only reason we would point at a wrong inode is if
2176 2177 2178
 * e2fsck was run on this filesystem, and it must have already done the orphan
 * inode cleanup for us, so we can safely abort without any further action.
 */
2179 2180
static void ext4_orphan_cleanup(struct super_block *sb,
				struct ext4_super_block *es)
2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191
{
	unsigned int s_flags = sb->s_flags;
	int nr_orphans = 0, nr_truncates = 0;
#ifdef CONFIG_QUOTA
	int i;
#endif
	if (!es->s_last_orphan) {
		jbd_debug(4, "no orphan inodes to clean up\n");
		return;
	}

2192
	if (bdev_read_only(sb->s_bdev)) {
2193 2194
		ext4_msg(sb, KERN_ERR, "write access "
			"unavailable, skipping orphan cleanup");
2195 2196 2197
		return;
	}

2198 2199 2200 2201 2202 2203 2204
	/* Check if feature set would not allow a r/w mount */
	if (!ext4_feature_set_ok(sb, 0)) {
		ext4_msg(sb, KERN_INFO, "Skipping orphan cleanup due to "
			 "unknown ROCOMPAT features");
		return;
	}

2205
	if (EXT4_SB(sb)->s_mount_state & EXT4_ERROR_FS) {
2206 2207
		/* don't clear list on RO mount w/ errors */
		if (es->s_last_orphan && !(s_flags & MS_RDONLY)) {
2208
			ext4_msg(sb, KERN_INFO, "Errors on filesystem, "
2209
				  "clearing orphan list.\n");
2210 2211
			es->s_last_orphan = 0;
		}
2212 2213 2214 2215 2216
		jbd_debug(1, "Skipping orphan recovery on fs with errors.\n");
		return;
	}

	if (s_flags & MS_RDONLY) {
2217
		ext4_msg(sb, KERN_INFO, "orphan cleanup on readonly fs");
2218 2219 2220 2221 2222 2223
		sb->s_flags &= ~MS_RDONLY;
	}
#ifdef CONFIG_QUOTA
	/* Needed for iput() to work correctly and not trash data */
	sb->s_flags |= MS_ACTIVE;
	/* Turn on quotas so that they are updated correctly */
J
Jan Kara 已提交
2224
	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
2225 2226
		if (EXT4_SB(sb)->s_qf_names[i]) {
			int ret = ext4_quota_on_mount(sb, i);
2227
			if (ret < 0)
2228 2229 2230
				ext4_msg(sb, KERN_ERR,
					"Cannot turn on journaled "
					"quota: error %d", ret);
2231 2232 2233 2234 2235 2236 2237
		}
	}
#endif

	while (es->s_last_orphan) {
		struct inode *inode;

2238 2239
		inode = ext4_orphan_get(sb, le32_to_cpu(es->s_last_orphan));
		if (IS_ERR(inode)) {
2240 2241 2242 2243
			es->s_last_orphan = 0;
			break;
		}

2244
		list_add(&EXT4_I(inode)->i_orphan, &EXT4_SB(sb)->s_orphan);
2245
		dquot_initialize(inode);
2246
		if (inode->i_nlink) {
2247 2248 2249 2250
			if (test_opt(sb, DEBUG))
				ext4_msg(sb, KERN_DEBUG,
					"%s: truncating inode %lu to %lld bytes",
					__func__, inode->i_ino, inode->i_size);
2251
			jbd_debug(2, "truncating inode %lu to %lld bytes\n",
2252
				  inode->i_ino, inode->i_size);
2253
			mutex_lock(&inode->i_mutex);
2254
			truncate_inode_pages(inode->i_mapping, inode->i_size);
2255
			ext4_truncate(inode);
2256
			mutex_unlock(&inode->i_mutex);
2257 2258
			nr_truncates++;
		} else {
2259 2260 2261 2262
			if (test_opt(sb, DEBUG))
				ext4_msg(sb, KERN_DEBUG,
					"%s: deleting unreferenced inode %lu",
					__func__, inode->i_ino);
2263 2264 2265 2266 2267 2268 2269
			jbd_debug(2, "deleting unreferenced inode %lu\n",
				  inode->i_ino);
			nr_orphans++;
		}
		iput(inode);  /* The delete magic happens here! */
	}

2270
#define PLURAL(x) (x), ((x) == 1) ? "" : "s"
2271 2272

	if (nr_orphans)
2273 2274
		ext4_msg(sb, KERN_INFO, "%d orphan inode%s deleted",
		       PLURAL(nr_orphans));
2275
	if (nr_truncates)
2276 2277
		ext4_msg(sb, KERN_INFO, "%d truncate%s cleaned up",
		       PLURAL(nr_truncates));
2278 2279
#ifdef CONFIG_QUOTA
	/* Turn quotas off */
J
Jan Kara 已提交
2280
	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
2281
		if (sb_dqopt(sb)->files[i])
2282
			dquot_quota_off(sb, i);
2283 2284 2285 2286
	}
#endif
	sb->s_flags = s_flags; /* Restore MS_RDONLY status */
}
2287

2288 2289 2290 2291 2292 2293 2294
/*
 * Maximal extent format file size.
 * Resulting logical blkno at s_maxbytes must fit in our on-disk
 * extent format containers, within a sector_t, and within i_blocks
 * in the vfs.  ext4 inode has 48 bits of i_block in fsblock units,
 * so that won't be a limiting factor.
 *
2295 2296 2297 2298 2299 2300
 * However there is other limiting factor. We do store extents in the form
 * of starting block and length, hence the resulting length of the extent
 * covering maximum file size must fit into on-disk format containers as
 * well. Given that length is always by 1 unit bigger than max unit (because
 * we count 0 as well) we have to lower the s_maxbytes by one fs block.
 *
2301 2302
 * Note, this does *not* consider any metadata overhead for vfs i_blocks.
 */
2303
static loff_t ext4_max_size(int blkbits, int has_huge_files)
2304 2305 2306 2307 2308
{
	loff_t res;
	loff_t upper_limit = MAX_LFS_FILESIZE;

	/* small i_blocks in vfs inode? */
2309
	if (!has_huge_files || sizeof(blkcnt_t) < sizeof(u64)) {
2310
		/*
2311
		 * CONFIG_LBDAF is not enabled implies the inode
2312 2313 2314 2315 2316 2317 2318 2319 2320 2321
		 * i_block represent total blocks in 512 bytes
		 * 32 == size of vfs inode i_blocks * 8
		 */
		upper_limit = (1LL << 32) - 1;

		/* total blocks in file system block size */
		upper_limit >>= (blkbits - 9);
		upper_limit <<= blkbits;
	}

2322 2323 2324 2325 2326 2327
	/*
	 * 32-bit extent-start container, ee_block. We lower the maxbytes
	 * by one fs block, so ee_len can cover the extent of maximum file
	 * size
	 */
	res = (1LL << 32) - 1;
2328 2329 2330 2331 2332 2333 2334 2335
	res <<= blkbits;

	/* Sanity check against vm- & vfs- imposed limits */
	if (res > upper_limit)
		res = upper_limit;

	return res;
}
2336 2337

/*
2338
 * Maximal bitmap file size.  There is a direct, and {,double-,triple-}indirect
2339 2340
 * block limit, and also a limit of (2^48 - 1) 512-byte sectors in i_blocks.
 * We need to be 1 filesystem block less than the 2^48 sector limit.
2341
 */
2342
static loff_t ext4_max_bitmap_size(int bits, int has_huge_files)
2343
{
2344
	loff_t res = EXT4_NDIR_BLOCKS;
2345 2346
	int meta_blocks;
	loff_t upper_limit;
2347 2348 2349 2350 2351 2352
	/* This is calculated to be the largest file size for a dense, block
	 * mapped file such that the file's total number of 512-byte sectors,
	 * including data and all indirect blocks, does not exceed (2^48 - 1).
	 *
	 * __u32 i_blocks_lo and _u16 i_blocks_high represent the total
	 * number of 512-byte sectors of the file.
2353 2354
	 */

2355
	if (!has_huge_files || sizeof(blkcnt_t) < sizeof(u64)) {
2356
		/*
2357
		 * !has_huge_files or CONFIG_LBDAF not enabled implies that
2358 2359
		 * the inode i_block field represents total file blocks in
		 * 2^32 512-byte sectors == size of vfs inode i_blocks * 8
2360 2361 2362 2363 2364 2365 2366
		 */
		upper_limit = (1LL << 32) - 1;

		/* total blocks in file system block size */
		upper_limit >>= (bits - 9);

	} else {
A
Aneesh Kumar K.V 已提交
2367 2368 2369 2370 2371 2372
		/*
		 * We use 48 bit ext4_inode i_blocks
		 * With EXT4_HUGE_FILE_FL set the i_blocks
		 * represent total number of blocks in
		 * file system block size
		 */
2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385
		upper_limit = (1LL << 48) - 1;

	}

	/* indirect blocks */
	meta_blocks = 1;
	/* double indirect blocks */
	meta_blocks += 1 + (1LL << (bits-2));
	/* tripple indirect blocks */
	meta_blocks += 1 + (1LL << (bits-2)) + (1LL << (2*(bits-2)));

	upper_limit -= meta_blocks;
	upper_limit <<= bits;
2386 2387 2388 2389 2390 2391 2392

	res += 1LL << (bits-2);
	res += 1LL << (2*(bits-2));
	res += 1LL << (3*(bits-2));
	res <<= bits;
	if (res > upper_limit)
		res = upper_limit;
2393 2394 2395 2396

	if (res > MAX_LFS_FILESIZE)
		res = MAX_LFS_FILESIZE;

2397 2398 2399
	return res;
}

2400
static ext4_fsblk_t descriptor_loc(struct super_block *sb,
2401
				   ext4_fsblk_t logical_sb_block, int nr)
2402
{
2403
	struct ext4_sb_info *sbi = EXT4_SB(sb);
2404
	ext4_group_t bg, first_meta_bg;
2405 2406 2407 2408
	int has_super = 0;

	first_meta_bg = le32_to_cpu(sbi->s_es->s_first_meta_bg);

2409
	if (!ext4_has_feature_meta_bg(sb) || nr < first_meta_bg)
2410
		return logical_sb_block + nr + 1;
2411
	bg = sbi->s_desc_per_block * nr;
2412
	if (ext4_bg_has_super(sb, bg))
2413
		has_super = 1;
2414

2415 2416 2417 2418 2419 2420 2421 2422 2423 2424
	/*
	 * If we have a meta_bg fs with 1k blocks, group 0's GDT is at
	 * block 2, not 1.  If s_first_data_block == 0 (bigalloc is enabled
	 * on modern mke2fs or blksize > 1k on older mke2fs) then we must
	 * compensate.
	 */
	if (sb->s_blocksize == 1024 && nr == 0 &&
	    le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block) == 0)
		has_super++;

2425
	return (has_super + ext4_group_first_block_no(sb, bg));
2426 2427
}

2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443
/**
 * ext4_get_stripe_size: Get the stripe size.
 * @sbi: In memory super block info
 *
 * If we have specified it via mount option, then
 * use the mount option value. If the value specified at mount time is
 * greater than the blocks per group use the super block value.
 * If the super block value is greater than blocks per group return 0.
 * Allocator needs it be less than blocks per group.
 *
 */
static unsigned long ext4_get_stripe_size(struct ext4_sb_info *sbi)
{
	unsigned long stride = le16_to_cpu(sbi->s_es->s_raid_stride);
	unsigned long stripe_width =
			le32_to_cpu(sbi->s_es->s_raid_stripe_width);
D
Dan Ehrenberg 已提交
2444
	int ret;
2445 2446

	if (sbi->s_stripe && sbi->s_stripe <= sbi->s_blocks_per_group)
D
Dan Ehrenberg 已提交
2447 2448 2449 2450 2451 2452 2453
		ret = sbi->s_stripe;
	else if (stripe_width <= sbi->s_blocks_per_group)
		ret = stripe_width;
	else if (stride <= sbi->s_blocks_per_group)
		ret = stride;
	else
		ret = 0;
2454

D
Dan Ehrenberg 已提交
2455 2456 2457 2458 2459 2460
	/*
	 * If the stripe width is 1, this makes no sense and
	 * we set it to 0 to turn off stripe handling code.
	 */
	if (ret <= 1)
		ret = 0;
2461

D
Dan Ehrenberg 已提交
2462
	return ret;
2463
}
2464

2465 2466 2467 2468 2469 2470 2471 2472
/*
 * Check whether this filesystem can be mounted based on
 * the features present and the RDONLY/RDWR mount requested.
 * Returns 1 if this filesystem can be mounted as requested,
 * 0 if it cannot be.
 */
static int ext4_feature_set_ok(struct super_block *sb, int readonly)
{
2473
	if (ext4_has_unknown_ext4_incompat_features(sb)) {
2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484
		ext4_msg(sb, KERN_ERR,
			"Couldn't mount because of "
			"unsupported optional features (%x)",
			(le32_to_cpu(EXT4_SB(sb)->s_es->s_feature_incompat) &
			~EXT4_FEATURE_INCOMPAT_SUPP));
		return 0;
	}

	if (readonly)
		return 1;

2485
	if (ext4_has_feature_readonly(sb)) {
D
Darrick J. Wong 已提交
2486 2487 2488 2489 2490
		ext4_msg(sb, KERN_INFO, "filesystem is read-only");
		sb->s_flags |= MS_RDONLY;
		return 1;
	}

2491
	/* Check that feature set is OK for a read-write mount */
2492
	if (ext4_has_unknown_ext4_ro_compat_features(sb)) {
2493 2494 2495 2496 2497 2498 2499 2500 2501 2502
		ext4_msg(sb, KERN_ERR, "couldn't mount RDWR because of "
			 "unsupported optional features (%x)",
			 (le32_to_cpu(EXT4_SB(sb)->s_es->s_feature_ro_compat) &
				~EXT4_FEATURE_RO_COMPAT_SUPP));
		return 0;
	}
	/*
	 * Large file size enabled file system can only be mounted
	 * read-write on 32-bit systems if kernel is built with CONFIG_LBDAF
	 */
2503
	if (ext4_has_feature_huge_file(sb)) {
2504 2505 2506 2507 2508 2509 2510
		if (sizeof(blkcnt_t) < sizeof(u64)) {
			ext4_msg(sb, KERN_ERR, "Filesystem with huge files "
				 "cannot be mounted RDWR without "
				 "CONFIG_LBDAF");
			return 0;
		}
	}
2511
	if (ext4_has_feature_bigalloc(sb) && !ext4_has_feature_extents(sb)) {
2512 2513 2514 2515 2516
		ext4_msg(sb, KERN_ERR,
			 "Can't support bigalloc feature without "
			 "extents feature\n");
		return 0;
	}
2517 2518

#ifndef CONFIG_QUOTA
2519
	if (ext4_has_feature_quota(sb) && !readonly) {
2520 2521 2522 2523 2524 2525
		ext4_msg(sb, KERN_ERR,
			 "Filesystem with quota feature cannot be mounted RDWR "
			 "without CONFIG_QUOTA");
		return 0;
	}
#endif  /* CONFIG_QUOTA */
2526 2527 2528
	return 1;
}

2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542
/*
 * This function is called once a day if we have errors logged
 * on the file system
 */
static void print_daily_error_info(unsigned long arg)
{
	struct super_block *sb = (struct super_block *) arg;
	struct ext4_sb_info *sbi;
	struct ext4_super_block *es;

	sbi = EXT4_SB(sb);
	es = sbi->s_es;

	if (es->s_error_count)
2543 2544
		/* fsck newer than v1.41.13 is needed to clean this condition. */
		ext4_msg(sb, KERN_NOTICE, "error count since last fsck: %u",
2545 2546
			 le32_to_cpu(es->s_error_count));
	if (es->s_first_error_time) {
2547
		printk(KERN_NOTICE "EXT4-fs (%s): initial error at time %u: %.*s:%d",
2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560
		       sb->s_id, le32_to_cpu(es->s_first_error_time),
		       (int) sizeof(es->s_first_error_func),
		       es->s_first_error_func,
		       le32_to_cpu(es->s_first_error_line));
		if (es->s_first_error_ino)
			printk(": inode %u",
			       le32_to_cpu(es->s_first_error_ino));
		if (es->s_first_error_block)
			printk(": block %llu", (unsigned long long)
			       le64_to_cpu(es->s_first_error_block));
		printk("\n");
	}
	if (es->s_last_error_time) {
2561
		printk(KERN_NOTICE "EXT4-fs (%s): last error at time %u: %.*s:%d",
2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576
		       sb->s_id, le32_to_cpu(es->s_last_error_time),
		       (int) sizeof(es->s_last_error_func),
		       es->s_last_error_func,
		       le32_to_cpu(es->s_last_error_line));
		if (es->s_last_error_ino)
			printk(": inode %u",
			       le32_to_cpu(es->s_last_error_ino));
		if (es->s_last_error_block)
			printk(": block %llu", (unsigned long long)
			       le64_to_cpu(es->s_last_error_block));
		printk("\n");
	}
	mod_timer(&sbi->s_err_report, jiffies + 24*60*60*HZ);  /* Once a day */
}

2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588
/* Find next suitable group and run ext4_init_inode_table */
static int ext4_run_li_request(struct ext4_li_request *elr)
{
	struct ext4_group_desc *gdp = NULL;
	ext4_group_t group, ngroups;
	struct super_block *sb;
	unsigned long timeout = 0;
	int ret = 0;

	sb = elr->lr_super;
	ngroups = EXT4_SB(sb)->s_groups_count;

2589
	sb_start_write(sb);
2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600
	for (group = elr->lr_next_group; group < ngroups; group++) {
		gdp = ext4_get_group_desc(sb, group, NULL);
		if (!gdp) {
			ret = 1;
			break;
		}

		if (!(gdp->bg_flags & cpu_to_le16(EXT4_BG_INODE_ZEROED)))
			break;
	}

2601
	if (group >= ngroups)
2602 2603 2604 2605 2606 2607 2608
		ret = 1;

	if (!ret) {
		timeout = jiffies;
		ret = ext4_init_inode_table(sb, group,
					    elr->lr_timeout ? 0 : 1);
		if (elr->lr_timeout == 0) {
2609 2610
			timeout = (jiffies - timeout) *
				  elr->lr_sbi->s_li_wait_mult;
2611 2612 2613 2614 2615
			elr->lr_timeout = timeout;
		}
		elr->lr_next_sched = jiffies + elr->lr_timeout;
		elr->lr_next_group = group + 1;
	}
2616
	sb_end_write(sb);
2617 2618 2619 2620 2621 2622

	return ret;
}

/*
 * Remove lr_request from the list_request and free the
2623
 * request structure. Should be called with li_list_mtx held
2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640
 */
static void ext4_remove_li_request(struct ext4_li_request *elr)
{
	struct ext4_sb_info *sbi;

	if (!elr)
		return;

	sbi = elr->lr_sbi;

	list_del(&elr->lr_request);
	sbi->s_li_request = NULL;
	kfree(elr);
}

static void ext4_unregister_li_request(struct super_block *sb)
{
2641 2642 2643
	mutex_lock(&ext4_li_mtx);
	if (!ext4_li_info) {
		mutex_unlock(&ext4_li_mtx);
2644
		return;
2645
	}
2646 2647

	mutex_lock(&ext4_li_info->li_list_mtx);
2648
	ext4_remove_li_request(EXT4_SB(sb)->s_li_request);
2649
	mutex_unlock(&ext4_li_info->li_list_mtx);
2650
	mutex_unlock(&ext4_li_mtx);
2651 2652
}

2653 2654
static struct task_struct *ext4_lazyinit_task;

2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668
/*
 * This is the function where ext4lazyinit thread lives. It walks
 * through the request list searching for next scheduled filesystem.
 * When such a fs is found, run the lazy initialization request
 * (ext4_rn_li_request) and keep track of the time spend in this
 * function. Based on that time we compute next schedule time of
 * the request. When walking through the list is complete, compute
 * next waking time and put itself into sleep.
 */
static int ext4_lazyinit_thread(void *arg)
{
	struct ext4_lazy_init *eli = (struct ext4_lazy_init *)arg;
	struct list_head *pos, *n;
	struct ext4_li_request *elr;
2669
	unsigned long next_wakeup, cur;
2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686

	BUG_ON(NULL == eli);

cont_thread:
	while (true) {
		next_wakeup = MAX_JIFFY_OFFSET;

		mutex_lock(&eli->li_list_mtx);
		if (list_empty(&eli->li_request_list)) {
			mutex_unlock(&eli->li_list_mtx);
			goto exit_thread;
		}

		list_for_each_safe(pos, n, &eli->li_request_list) {
			elr = list_entry(pos, struct ext4_li_request,
					 lr_request);

2687 2688 2689 2690 2691 2692
			if (time_after_eq(jiffies, elr->lr_next_sched)) {
				if (ext4_run_li_request(elr) != 0) {
					/* error, remove the lazy_init job */
					ext4_remove_li_request(elr);
					continue;
				}
2693 2694 2695 2696 2697 2698 2699
			}

			if (time_before(elr->lr_next_sched, next_wakeup))
				next_wakeup = elr->lr_next_sched;
		}
		mutex_unlock(&eli->li_list_mtx);

2700
		try_to_freeze();
2701

2702 2703
		cur = jiffies;
		if ((time_after_eq(cur, next_wakeup)) ||
2704
		    (MAX_JIFFY_OFFSET == next_wakeup)) {
2705 2706 2707 2708
			cond_resched();
			continue;
		}

2709 2710
		schedule_timeout_interruptible(next_wakeup - cur);

2711 2712 2713 2714
		if (kthread_should_stop()) {
			ext4_clear_request_list();
			goto exit_thread;
		}
2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756
	}

exit_thread:
	/*
	 * It looks like the request list is empty, but we need
	 * to check it under the li_list_mtx lock, to prevent any
	 * additions into it, and of course we should lock ext4_li_mtx
	 * to atomically free the list and ext4_li_info, because at
	 * this point another ext4 filesystem could be registering
	 * new one.
	 */
	mutex_lock(&ext4_li_mtx);
	mutex_lock(&eli->li_list_mtx);
	if (!list_empty(&eli->li_request_list)) {
		mutex_unlock(&eli->li_list_mtx);
		mutex_unlock(&ext4_li_mtx);
		goto cont_thread;
	}
	mutex_unlock(&eli->li_list_mtx);
	kfree(ext4_li_info);
	ext4_li_info = NULL;
	mutex_unlock(&ext4_li_mtx);

	return 0;
}

static void ext4_clear_request_list(void)
{
	struct list_head *pos, *n;
	struct ext4_li_request *elr;

	mutex_lock(&ext4_li_info->li_list_mtx);
	list_for_each_safe(pos, n, &ext4_li_info->li_request_list) {
		elr = list_entry(pos, struct ext4_li_request,
				 lr_request);
		ext4_remove_li_request(elr);
	}
	mutex_unlock(&ext4_li_info->li_list_mtx);
}

static int ext4_run_lazyinit_thread(void)
{
2757 2758 2759 2760
	ext4_lazyinit_task = kthread_run(ext4_lazyinit_thread,
					 ext4_li_info, "ext4lazyinit");
	if (IS_ERR(ext4_lazyinit_task)) {
		int err = PTR_ERR(ext4_lazyinit_task);
2761 2762 2763
		ext4_clear_request_list();
		kfree(ext4_li_info);
		ext4_li_info = NULL;
2764
		printk(KERN_CRIT "EXT4-fs: error %d creating inode table "
2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832
				 "initialization thread\n",
				 err);
		return err;
	}
	ext4_li_info->li_state |= EXT4_LAZYINIT_RUNNING;
	return 0;
}

/*
 * Check whether it make sense to run itable init. thread or not.
 * If there is at least one uninitialized inode table, return
 * corresponding group number, else the loop goes through all
 * groups and return total number of groups.
 */
static ext4_group_t ext4_has_uninit_itable(struct super_block *sb)
{
	ext4_group_t group, ngroups = EXT4_SB(sb)->s_groups_count;
	struct ext4_group_desc *gdp = NULL;

	for (group = 0; group < ngroups; group++) {
		gdp = ext4_get_group_desc(sb, group, NULL);
		if (!gdp)
			continue;

		if (!(gdp->bg_flags & cpu_to_le16(EXT4_BG_INODE_ZEROED)))
			break;
	}

	return group;
}

static int ext4_li_info_new(void)
{
	struct ext4_lazy_init *eli = NULL;

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

	INIT_LIST_HEAD(&eli->li_request_list);
	mutex_init(&eli->li_list_mtx);

	eli->li_state |= EXT4_LAZYINIT_QUIT;

	ext4_li_info = eli;

	return 0;
}

static struct ext4_li_request *ext4_li_request_new(struct super_block *sb,
					    ext4_group_t start)
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_li_request *elr;

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

	elr->lr_super = sb;
	elr->lr_sbi = sbi;
	elr->lr_next_group = start;

	/*
	 * Randomize first schedule time of the request to
	 * spread the inode table initialization requests
	 * better.
	 */
2833 2834
	elr->lr_next_sched = jiffies + (prandom_u32() %
				(EXT4_DEF_LI_MAX_START_DELAY * HZ));
2835 2836 2837
	return elr;
}

2838 2839
int ext4_register_li_request(struct super_block *sb,
			     ext4_group_t first_not_zeroed)
2840 2841
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
2842
	struct ext4_li_request *elr = NULL;
2843
	ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count;
2844
	int ret = 0;
2845

2846
	mutex_lock(&ext4_li_mtx);
2847 2848 2849 2850 2851 2852
	if (sbi->s_li_request != NULL) {
		/*
		 * Reset timeout so it can be computed again, because
		 * s_li_wait_mult might have changed.
		 */
		sbi->s_li_request->lr_timeout = 0;
2853
		goto out;
2854
	}
2855 2856 2857

	if (first_not_zeroed == ngroups ||
	    (sb->s_flags & MS_RDONLY) ||
2858
	    !test_opt(sb, INIT_INODE_TABLE))
2859
		goto out;
2860 2861

	elr = ext4_li_request_new(sb, first_not_zeroed);
2862 2863 2864 2865
	if (!elr) {
		ret = -ENOMEM;
		goto out;
	}
2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877

	if (NULL == ext4_li_info) {
		ret = ext4_li_info_new();
		if (ret)
			goto out;
	}

	mutex_lock(&ext4_li_info->li_list_mtx);
	list_add(&elr->lr_request, &ext4_li_info->li_request_list);
	mutex_unlock(&ext4_li_info->li_list_mtx);

	sbi->s_li_request = elr;
2878 2879 2880 2881 2882 2883
	/*
	 * set elr to NULL here since it has been inserted to
	 * the request_list and the removal and free of it is
	 * handled by ext4_clear_request_list from now on.
	 */
	elr = NULL;
2884 2885 2886 2887 2888 2889 2890

	if (!(ext4_li_info->li_state & EXT4_LAZYINIT_RUNNING)) {
		ret = ext4_run_lazyinit_thread();
		if (ret)
			goto out;
	}
out:
2891 2892
	mutex_unlock(&ext4_li_mtx);
	if (ret)
2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906
		kfree(elr);
	return ret;
}

/*
 * We do not need to lock anything since this is called on
 * module unload.
 */
static void ext4_destroy_lazyinit_thread(void)
{
	/*
	 * If thread exited earlier
	 * there's nothing to be done.
	 */
2907
	if (!ext4_li_info || !ext4_lazyinit_task)
2908 2909
		return;

2910
	kthread_stop(ext4_lazyinit_task);
2911 2912
}

2913 2914 2915 2916 2917 2918
static int set_journal_csum_feature_set(struct super_block *sb)
{
	int ret = 1;
	int compat, incompat;
	struct ext4_sb_info *sbi = EXT4_SB(sb);

2919
	if (ext4_has_metadata_csum(sb)) {
2920
		/* journal checksum v3 */
2921
		compat = 0;
2922
		incompat = JBD2_FEATURE_INCOMPAT_CSUM_V3;
2923 2924 2925 2926 2927 2928
	} else {
		/* journal checksum v1 */
		compat = JBD2_FEATURE_COMPAT_CHECKSUM;
		incompat = 0;
	}

2929 2930 2931 2932
	jbd2_journal_clear_features(sbi->s_journal,
			JBD2_FEATURE_COMPAT_CHECKSUM, 0,
			JBD2_FEATURE_INCOMPAT_CSUM_V3 |
			JBD2_FEATURE_INCOMPAT_CSUM_V2);
2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944
	if (test_opt(sb, JOURNAL_ASYNC_COMMIT)) {
		ret = jbd2_journal_set_features(sbi->s_journal,
				compat, 0,
				JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT |
				incompat);
	} else if (test_opt(sb, JOURNAL_CHECKSUM)) {
		ret = jbd2_journal_set_features(sbi->s_journal,
				compat, 0,
				incompat);
		jbd2_journal_clear_features(sbi->s_journal, 0, 0,
				JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT);
	} else {
2945 2946
		jbd2_journal_clear_features(sbi->s_journal, 0, 0,
				JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT);
2947 2948 2949 2950 2951
	}

	return ret;
}

2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975
/*
 * Note: calculating the overhead so we can be compatible with
 * historical BSD practice is quite difficult in the face of
 * clusters/bigalloc.  This is because multiple metadata blocks from
 * different block group can end up in the same allocation cluster.
 * Calculating the exact overhead in the face of clustered allocation
 * requires either O(all block bitmaps) in memory or O(number of block
 * groups**2) in time.  We will still calculate the superblock for
 * older file systems --- and if we come across with a bigalloc file
 * system with zero in s_overhead_clusters the estimate will be close to
 * correct especially for very large cluster sizes --- but for newer
 * file systems, it's better to calculate this figure once at mkfs
 * time, and store it in the superblock.  If the superblock value is
 * present (even for non-bigalloc file systems), we will use it.
 */
static int count_overhead(struct super_block *sb, ext4_group_t grp,
			  char *buf)
{
	struct ext4_sb_info	*sbi = EXT4_SB(sb);
	struct ext4_group_desc	*gdp;
	ext4_fsblk_t		first_block, last_block, b;
	ext4_group_t		i, ngroups = ext4_get_groups_count(sb);
	int			s, j, count = 0;

2976
	if (!ext4_has_feature_bigalloc(sb))
2977 2978 2979
		return (ext4_bg_has_super(sb, grp) + ext4_bg_num_gdb(sb, grp) +
			sbi->s_itb_per_group + 2);

2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028
	first_block = le32_to_cpu(sbi->s_es->s_first_data_block) +
		(grp * EXT4_BLOCKS_PER_GROUP(sb));
	last_block = first_block + EXT4_BLOCKS_PER_GROUP(sb) - 1;
	for (i = 0; i < ngroups; i++) {
		gdp = ext4_get_group_desc(sb, i, NULL);
		b = ext4_block_bitmap(sb, gdp);
		if (b >= first_block && b <= last_block) {
			ext4_set_bit(EXT4_B2C(sbi, b - first_block), buf);
			count++;
		}
		b = ext4_inode_bitmap(sb, gdp);
		if (b >= first_block && b <= last_block) {
			ext4_set_bit(EXT4_B2C(sbi, b - first_block), buf);
			count++;
		}
		b = ext4_inode_table(sb, gdp);
		if (b >= first_block && b + sbi->s_itb_per_group <= last_block)
			for (j = 0; j < sbi->s_itb_per_group; j++, b++) {
				int c = EXT4_B2C(sbi, b - first_block);
				ext4_set_bit(c, buf);
				count++;
			}
		if (i != grp)
			continue;
		s = 0;
		if (ext4_bg_has_super(sb, grp)) {
			ext4_set_bit(s++, buf);
			count++;
		}
		for (j = ext4_bg_num_gdb(sb, grp); j > 0; j--) {
			ext4_set_bit(EXT4_B2C(sbi, s++), buf);
			count++;
		}
	}
	if (!count)
		return 0;
	return EXT4_CLUSTERS_PER_GROUP(sb) -
		ext4_count_free(buf, EXT4_CLUSTERS_PER_GROUP(sb) / 8);
}

/*
 * Compute the overhead and stash it in sbi->s_overhead
 */
int ext4_calculate_overhead(struct super_block *sb)
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_super_block *es = sbi->s_es;
	ext4_group_t i, ngroups = ext4_get_groups_count(sb);
	ext4_fsblk_t overhead = 0;
3029
	char *buf = (char *) get_zeroed_page(GFP_NOFS);
3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056

	if (!buf)
		return -ENOMEM;

	/*
	 * Compute the overhead (FS structures).  This is constant
	 * for a given filesystem unless the number of block groups
	 * changes so we cache the previous value until it does.
	 */

	/*
	 * All of the blocks before first_data_block are overhead
	 */
	overhead = EXT4_B2C(sbi, le32_to_cpu(es->s_first_data_block));

	/*
	 * Add the overhead found in each block group
	 */
	for (i = 0; i < ngroups; i++) {
		int blks;

		blks = count_overhead(sb, i, buf);
		overhead += blks;
		if (blks)
			memset(buf, 0, PAGE_SIZE);
		cond_resched();
	}
3057 3058
	/* Add the internal journal blocks as well */
	if (sbi->s_journal && !sbi->journal_bdev)
3059
		overhead += EXT4_NUM_B2C(sbi, sbi->s_journal->j_maxlen);
3060

3061 3062 3063 3064 3065 3066
	sbi->s_overhead = overhead;
	smp_wmb();
	free_page((unsigned long) buf);
	return 0;
}

3067
static void ext4_set_resv_clusters(struct super_block *sb)
L
Lukas Czerner 已提交
3068 3069
{
	ext4_fsblk_t resv_clusters;
3070
	struct ext4_sb_info *sbi = EXT4_SB(sb);
L
Lukas Czerner 已提交
3071

3072 3073 3074 3075 3076 3077
	/*
	 * There's no need to reserve anything when we aren't using extents.
	 * The space estimates are exact, there are no unwritten extents,
	 * hole punching doesn't need new metadata... This is needed especially
	 * to keep ext2/3 backward compatibility.
	 */
3078
	if (!ext4_has_feature_extents(sb))
3079
		return;
L
Lukas Czerner 已提交
3080 3081 3082 3083
	/*
	 * By default we reserve 2% or 4096 clusters, whichever is smaller.
	 * This should cover the situations where we can not afford to run
	 * out of space like for example punch hole, or converting
3084
	 * unwritten extents in delalloc path. In most cases such
L
Lukas Czerner 已提交
3085 3086 3087
	 * allocation would require 1, or 2 blocks, higher numbers are
	 * very rare.
	 */
3088 3089
	resv_clusters = (ext4_blocks_count(sbi->s_es) >>
			 sbi->s_cluster_bits);
L
Lukas Czerner 已提交
3090 3091 3092 3093

	do_div(resv_clusters, 50);
	resv_clusters = min_t(ext4_fsblk_t, resv_clusters, 4096);

3094
	atomic64_set(&sbi->s_resv_clusters, resv_clusters);
L
Lukas Czerner 已提交
3095 3096
}

3097
static int ext4_fill_super(struct super_block *sb, void *data, int silent)
3098
{
3099
	char *orig_data = kstrdup(data, GFP_KERNEL);
3100
	struct buffer_head *bh;
3101 3102 3103 3104
	struct ext4_super_block *es = NULL;
	struct ext4_sb_info *sbi;
	ext4_fsblk_t block;
	ext4_fsblk_t sb_block = get_sb_block(&data);
3105
	ext4_fsblk_t logical_sb_block;
3106 3107 3108 3109
	unsigned long offset = 0;
	unsigned long journal_devnum = 0;
	unsigned long def_mount_opts;
	struct inode *root;
3110
	const char *descr;
3111
	int ret = -ENOMEM;
3112
	int blocksize, clustersize;
3113 3114
	unsigned int db_count;
	unsigned int i;
3115
	int needs_recovery, has_huge_files, has_bigalloc;
L
Laurent Vivier 已提交
3116
	__u64 blocks_count;
3117
	int err = 0;
3118
	unsigned int journal_ioprio = DEFAULT_JOURNAL_IOPRIO;
3119
	ext4_group_t first_not_zeroed;
3120 3121 3122

	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
	if (!sbi)
3123
		goto out_free_orig;
3124 3125 3126 3127 3128

	sbi->s_blockgroup_lock =
		kzalloc(sizeof(struct blockgroup_lock), GFP_KERNEL);
	if (!sbi->s_blockgroup_lock) {
		kfree(sbi);
3129
		goto out_free_orig;
3130
	}
3131
	sb->s_fs_info = sbi;
3132
	sbi->s_sb = sb;
3133
	sbi->s_inode_readahead_blks = EXT4_DEF_INODE_READAHEAD_BLKS;
M
Miklos Szeredi 已提交
3134
	sbi->s_sb_block = sb_block;
3135 3136 3137
	if (sb->s_bdev->bd_part)
		sbi->s_sectors_written_start =
			part_stat_read(sb->s_bdev->bd_part, sectors[1]);
3138

3139
	/* Cleanup superblock name */
3140
	strreplace(sb->s_id, '/', '!');
3141

3142
	/* -EINVAL is default */
3143
	ret = -EINVAL;
3144
	blocksize = sb_min_blocksize(sb, EXT4_MIN_BLOCK_SIZE);
3145
	if (!blocksize) {
3146
		ext4_msg(sb, KERN_ERR, "unable to set blocksize");
3147 3148 3149 3150
		goto out_fail;
	}

	/*
3151
	 * The ext4 superblock will not be buffer aligned for other than 1kB
3152 3153
	 * block sizes.  We need to calculate the offset from buffer start.
	 */
3154
	if (blocksize != EXT4_MIN_BLOCK_SIZE) {
3155 3156
		logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE;
		offset = do_div(logical_sb_block, blocksize);
3157
	} else {
3158
		logical_sb_block = sb_block;
3159 3160
	}

3161
	if (!(bh = sb_bread_unmovable(sb, logical_sb_block))) {
3162
		ext4_msg(sb, KERN_ERR, "unable to read superblock");
3163 3164 3165 3166
		goto out_fail;
	}
	/*
	 * Note: s_es must be initialized as soon as possible because
3167
	 *       some ext4 macro-instructions depend on its value
3168
	 */
3169
	es = (struct ext4_super_block *) (bh->b_data + offset);
3170 3171
	sbi->s_es = es;
	sb->s_magic = le16_to_cpu(es->s_magic);
3172 3173
	if (sb->s_magic != EXT4_SUPER_MAGIC)
		goto cantfind_ext4;
T
Theodore Ts'o 已提交
3174
	sbi->s_kbytes_written = le64_to_cpu(es->s_kbytes_written);
3175

3176
	/* Warn if metadata_csum and gdt_csum are both set. */
3177 3178
	if (ext4_has_feature_metadata_csum(sb) &&
	    ext4_has_feature_gdt_csum(sb))
3179
		ext4_warning(sb, "metadata_csum and uninit_bg are "
3180 3181
			     "redundant flags; please run fsck.");

3182 3183 3184 3185 3186 3187 3188 3189
	/* Check for a known checksum algorithm */
	if (!ext4_verify_csum_type(sb, es)) {
		ext4_msg(sb, KERN_ERR, "VFS: Found ext4 filesystem with "
			 "unknown checksum algorithm.");
		silent = 1;
		goto cantfind_ext4;
	}

3190
	/* Load the checksum driver */
3191
	if (ext4_has_feature_metadata_csum(sb)) {
3192 3193 3194 3195 3196 3197 3198 3199 3200
		sbi->s_chksum_driver = crypto_alloc_shash("crc32c", 0, 0);
		if (IS_ERR(sbi->s_chksum_driver)) {
			ext4_msg(sb, KERN_ERR, "Cannot load crc32c driver.");
			ret = PTR_ERR(sbi->s_chksum_driver);
			sbi->s_chksum_driver = NULL;
			goto failed_mount;
		}
	}

3201 3202 3203 3204 3205
	/* Check superblock checksum */
	if (!ext4_superblock_csum_verify(sb, es)) {
		ext4_msg(sb, KERN_ERR, "VFS: Found ext4 filesystem with "
			 "invalid superblock checksum.  Run e2fsck?");
		silent = 1;
3206
		ret = -EFSBADCRC;
3207 3208 3209 3210
		goto cantfind_ext4;
	}

	/* Precompute checksum seed for all metadata */
3211
	if (ext4_has_feature_csum_seed(sb))
3212 3213
		sbi->s_csum_seed = le32_to_cpu(es->s_checksum_seed);
	else if (ext4_has_metadata_csum(sb))
3214 3215 3216
		sbi->s_csum_seed = ext4_chksum(sbi, ~0, es->s_uuid,
					       sizeof(es->s_uuid));

3217 3218
	/* Set defaults before we parse the mount options */
	def_mount_opts = le32_to_cpu(es->s_default_mount_opts);
3219
	set_opt(sb, INIT_INODE_TABLE);
3220
	if (def_mount_opts & EXT4_DEFM_DEBUG)
3221
		set_opt(sb, DEBUG);
3222
	if (def_mount_opts & EXT4_DEFM_BSDGROUPS)
3223
		set_opt(sb, GRPID);
3224
	if (def_mount_opts & EXT4_DEFM_UID16)
3225
		set_opt(sb, NO_UID32);
3226 3227
	/* xattr user namespace & acls are now defaulted on */
	set_opt(sb, XATTR_USER);
T
Theodore Ts'o 已提交
3228
#ifdef CONFIG_EXT4_FS_POSIX_ACL
3229
	set_opt(sb, POSIX_ACL);
3230
#endif
3231 3232 3233 3234
	/* don't forget to enable journal_csum when metadata_csum is enabled. */
	if (ext4_has_metadata_csum(sb))
		set_opt(sb, JOURNAL_CHECKSUM);

3235
	if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_DATA)
3236
		set_opt(sb, JOURNAL_DATA);
3237
	else if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_ORDERED)
3238
		set_opt(sb, ORDERED_DATA);
3239
	else if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_WBACK)
3240
		set_opt(sb, WRITEBACK_DATA);
3241 3242

	if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_PANIC)
3243
		set_opt(sb, ERRORS_PANIC);
3244
	else if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_CONTINUE)
3245
		set_opt(sb, ERRORS_CONT);
3246
	else
3247
		set_opt(sb, ERRORS_RO);
3248 3249
	/* block_validity enabled by default; disable with noblock_validity */
	set_opt(sb, BLOCK_VALIDITY);
3250
	if (def_mount_opts & EXT4_DEFM_DISCARD)
3251
		set_opt(sb, DISCARD);
3252

3253 3254
	sbi->s_resuid = make_kuid(&init_user_ns, le16_to_cpu(es->s_def_resuid));
	sbi->s_resgid = make_kgid(&init_user_ns, le16_to_cpu(es->s_def_resgid));
3255 3256 3257
	sbi->s_commit_interval = JBD2_DEFAULT_MAX_COMMIT_AGE * HZ;
	sbi->s_min_batch_time = EXT4_DEF_MIN_BATCH_TIME;
	sbi->s_max_batch_time = EXT4_DEF_MAX_BATCH_TIME;
3258

3259
	if ((def_mount_opts & EXT4_DEFM_NOBARRIER) == 0)
3260
		set_opt(sb, BARRIER);
3261

3262 3263 3264 3265
	/*
	 * enable delayed allocation by default
	 * Use -o nodelalloc to turn it off
	 */
3266
	if (!IS_EXT3_SB(sb) && !IS_EXT2_SB(sb) &&
3267
	    ((def_mount_opts & EXT4_DEFM_NODELALLOC) == 0))
3268
		set_opt(sb, DELALLOC);
3269

3270 3271 3272 3273 3274 3275
	/*
	 * set default s_li_wait_mult for lazyinit, for the case there is
	 * no mount option specified.
	 */
	sbi->s_li_wait_mult = EXT4_DEF_LI_WAIT_MULT;

3276
	if (!parse_options((char *) sbi->s_es->s_mount_opts, sb,
3277
			   &journal_devnum, &journal_ioprio, 0)) {
3278 3279 3280 3281
		ext4_msg(sb, KERN_WARNING,
			 "failed to parse options in superblock: %s",
			 sbi->s_es->s_mount_opts);
	}
3282
	sbi->s_def_mount_opt = sbi->s_mount_opt;
3283
	if (!parse_options((char *) data, sb, &journal_devnum,
3284
			   &journal_ioprio, 0))
3285 3286
		goto failed_mount;

3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297
	if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) {
		printk_once(KERN_WARNING "EXT4-fs: Warning: mounting "
			    "with data=journal disables delayed "
			    "allocation and O_DIRECT support!\n");
		if (test_opt2(sb, EXPLICIT_DELALLOC)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "both data=journal and delalloc");
			goto failed_mount;
		}
		if (test_opt(sb, DIOREAD_NOLOCK)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
3298
				 "both data=journal and dioread_nolock");
3299 3300
			goto failed_mount;
		}
R
Ross Zwisler 已提交
3301 3302 3303 3304 3305
		if (test_opt(sb, DAX)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "both data=journal and dax");
			goto failed_mount;
		}
3306 3307
		if (test_opt(sb, DELALLOC))
			clear_opt(sb, DELALLOC);
3308 3309
	} else {
		sb->s_iflags |= SB_I_CGROUPWB;
3310 3311
	}

3312
	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
3313
		(test_opt(sb, POSIX_ACL) ? MS_POSIXACL : 0);
3314

3315
	if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV &&
3316 3317 3318
	    (ext4_has_compat_features(sb) ||
	     ext4_has_ro_compat_features(sb) ||
	     ext4_has_incompat_features(sb)))
3319 3320 3321
		ext4_msg(sb, KERN_WARNING,
		       "feature flags set on rev 0 fs, "
		       "running e2fsck is recommended");
3322

3323 3324
	if (es->s_creator_os == cpu_to_le32(EXT4_OS_HURD)) {
		set_opt2(sb, HURD_COMPAT);
3325
		if (ext4_has_feature_64bit(sb)) {
3326 3327 3328 3329 3330 3331
			ext4_msg(sb, KERN_ERR,
				 "The Hurd can't support 64-bit file systems");
			goto failed_mount;
		}
	}

3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353
	if (IS_EXT2_SB(sb)) {
		if (ext2_feature_set_ok(sb))
			ext4_msg(sb, KERN_INFO, "mounting ext2 file system "
				 "using the ext4 subsystem");
		else {
			ext4_msg(sb, KERN_ERR, "couldn't mount as ext2 due "
				 "to feature incompatibilities");
			goto failed_mount;
		}
	}

	if (IS_EXT3_SB(sb)) {
		if (ext3_feature_set_ok(sb))
			ext4_msg(sb, KERN_INFO, "mounting ext3 file system "
				 "using the ext4 subsystem");
		else {
			ext4_msg(sb, KERN_ERR, "couldn't mount as ext3 due "
				 "to feature incompatibilities");
			goto failed_mount;
		}
	}

3354 3355 3356 3357 3358
	/*
	 * Check feature flags regardless of the revision level, since we
	 * previously didn't change the revision level when setting the flags,
	 * so there is a chance incompat flags are set on a rev 0 filesystem.
	 */
3359
	if (!ext4_feature_set_ok(sb, (sb->s_flags & MS_RDONLY)))
3360
		goto failed_mount;
3361

J
Jan Kara 已提交
3362
	blocksize = BLOCK_SIZE << le32_to_cpu(es->s_log_block_size);
3363 3364
	if (blocksize < EXT4_MIN_BLOCK_SIZE ||
	    blocksize > EXT4_MAX_BLOCK_SIZE) {
3365 3366
		ext4_msg(sb, KERN_ERR,
		       "Unsupported filesystem blocksize %d", blocksize);
3367 3368 3369
		goto failed_mount;
	}

R
Ross Zwisler 已提交
3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382
	if (sbi->s_mount_opt & EXT4_MOUNT_DAX) {
		if (blocksize != PAGE_SIZE) {
			ext4_msg(sb, KERN_ERR,
					"error: unsupported blocksize for dax");
			goto failed_mount;
		}
		if (!sb->s_bdev->bd_disk->fops->direct_access) {
			ext4_msg(sb, KERN_ERR,
					"error: device does not support dax");
			goto failed_mount;
		}
	}

3383
	if (ext4_has_feature_encrypt(sb) && es->s_encryption_level) {
3384 3385 3386 3387 3388
		ext4_msg(sb, KERN_ERR, "Unsupported encryption level %d",
			 es->s_encryption_level);
		goto failed_mount;
	}

3389
	if (sb->s_blocksize != blocksize) {
3390 3391
		/* Validate the filesystem blocksize */
		if (!sb_set_blocksize(sb, blocksize)) {
3392
			ext4_msg(sb, KERN_ERR, "bad block size %d",
3393
					blocksize);
3394 3395 3396
			goto failed_mount;
		}

3397
		brelse(bh);
3398 3399
		logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE;
		offset = do_div(logical_sb_block, blocksize);
3400
		bh = sb_bread_unmovable(sb, logical_sb_block);
3401
		if (!bh) {
3402 3403
			ext4_msg(sb, KERN_ERR,
			       "Can't read superblock on 2nd try");
3404 3405
			goto failed_mount;
		}
3406
		es = (struct ext4_super_block *)(bh->b_data + offset);
3407
		sbi->s_es = es;
3408
		if (es->s_magic != cpu_to_le16(EXT4_SUPER_MAGIC)) {
3409 3410
			ext4_msg(sb, KERN_ERR,
			       "Magic mismatch, very weird!");
3411 3412 3413 3414
			goto failed_mount;
		}
	}

3415
	has_huge_files = ext4_has_feature_huge_file(sb);
3416 3417 3418
	sbi->s_bitmap_maxbytes = ext4_max_bitmap_size(sb->s_blocksize_bits,
						      has_huge_files);
	sb->s_maxbytes = ext4_max_size(sb->s_blocksize_bits, has_huge_files);
3419

3420 3421 3422
	if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV) {
		sbi->s_inode_size = EXT4_GOOD_OLD_INODE_SIZE;
		sbi->s_first_ino = EXT4_GOOD_OLD_FIRST_INO;
3423 3424 3425
	} else {
		sbi->s_inode_size = le16_to_cpu(es->s_inode_size);
		sbi->s_first_ino = le32_to_cpu(es->s_first_ino);
3426
		if ((sbi->s_inode_size < EXT4_GOOD_OLD_INODE_SIZE) ||
V
Vignesh Babu 已提交
3427
		    (!is_power_of_2(sbi->s_inode_size)) ||
3428
		    (sbi->s_inode_size > blocksize)) {
3429 3430
			ext4_msg(sb, KERN_ERR,
			       "unsupported inode size: %d",
3431
			       sbi->s_inode_size);
3432 3433
			goto failed_mount;
		}
K
Kalpak Shah 已提交
3434 3435
		if (sbi->s_inode_size > EXT4_GOOD_OLD_INODE_SIZE)
			sb->s_time_gran = 1 << (EXT4_EPOCH_BITS - 2);
3436
	}
3437

3438
	sbi->s_desc_size = le16_to_cpu(es->s_desc_size);
3439
	if (ext4_has_feature_64bit(sb)) {
3440
		if (sbi->s_desc_size < EXT4_MIN_DESC_SIZE_64BIT ||
3441
		    sbi->s_desc_size > EXT4_MAX_DESC_SIZE ||
V
vignesh babu 已提交
3442
		    !is_power_of_2(sbi->s_desc_size)) {
3443 3444
			ext4_msg(sb, KERN_ERR,
			       "unsupported descriptor size %lu",
3445 3446 3447 3448 3449
			       sbi->s_desc_size);
			goto failed_mount;
		}
	} else
		sbi->s_desc_size = EXT4_MIN_DESC_SIZE;
3450

3451 3452
	sbi->s_blocks_per_group = le32_to_cpu(es->s_blocks_per_group);
	sbi->s_inodes_per_group = le32_to_cpu(es->s_inodes_per_group);
3453
	if (EXT4_INODE_SIZE(sb) == 0 || EXT4_INODES_PER_GROUP(sb) == 0)
3454
		goto cantfind_ext4;
3455

3456
	sbi->s_inodes_per_block = blocksize / EXT4_INODE_SIZE(sb);
3457
	if (sbi->s_inodes_per_block == 0)
3458
		goto cantfind_ext4;
3459 3460
	sbi->s_itb_per_group = sbi->s_inodes_per_group /
					sbi->s_inodes_per_block;
3461
	sbi->s_desc_per_block = blocksize / EXT4_DESC_SIZE(sb);
3462 3463
	sbi->s_sbh = bh;
	sbi->s_mount_state = le16_to_cpu(es->s_state);
3464 3465
	sbi->s_addr_per_block_bits = ilog2(EXT4_ADDR_PER_BLOCK(sb));
	sbi->s_desc_per_block_bits = ilog2(EXT4_DESC_PER_BLOCK(sb));
3466

3467
	for (i = 0; i < 4; i++)
3468 3469
		sbi->s_hash_seed[i] = le32_to_cpu(es->s_hash_seed[i]);
	sbi->s_def_hash_version = es->s_def_hash_version;
3470
	if (ext4_has_feature_dir_index(sb)) {
3471 3472 3473 3474
		i = le32_to_cpu(es->s_flags);
		if (i & EXT2_FLAGS_UNSIGNED_HASH)
			sbi->s_hash_unsigned = 3;
		else if ((i & EXT2_FLAGS_SIGNED_HASH) == 0) {
3475
#ifdef __CHAR_UNSIGNED__
3476 3477 3478 3479
			if (!(sb->s_flags & MS_RDONLY))
				es->s_flags |=
					cpu_to_le32(EXT2_FLAGS_UNSIGNED_HASH);
			sbi->s_hash_unsigned = 3;
3480
#else
3481 3482 3483
			if (!(sb->s_flags & MS_RDONLY))
				es->s_flags |=
					cpu_to_le32(EXT2_FLAGS_SIGNED_HASH);
3484
#endif
3485
		}
3486
	}
3487

3488 3489
	/* Handle clustersize */
	clustersize = BLOCK_SIZE << le32_to_cpu(es->s_log_cluster_size);
3490
	has_bigalloc = ext4_has_feature_bigalloc(sb);
3491 3492 3493 3494 3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 3507 3508 3509 3510 3511 3512 3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528 3529 3530
	if (has_bigalloc) {
		if (clustersize < blocksize) {
			ext4_msg(sb, KERN_ERR,
				 "cluster size (%d) smaller than "
				 "block size (%d)", clustersize, blocksize);
			goto failed_mount;
		}
		sbi->s_cluster_bits = le32_to_cpu(es->s_log_cluster_size) -
			le32_to_cpu(es->s_log_block_size);
		sbi->s_clusters_per_group =
			le32_to_cpu(es->s_clusters_per_group);
		if (sbi->s_clusters_per_group > blocksize * 8) {
			ext4_msg(sb, KERN_ERR,
				 "#clusters per group too big: %lu",
				 sbi->s_clusters_per_group);
			goto failed_mount;
		}
		if (sbi->s_blocks_per_group !=
		    (sbi->s_clusters_per_group * (clustersize / blocksize))) {
			ext4_msg(sb, KERN_ERR, "blocks per group (%lu) and "
				 "clusters per group (%lu) inconsistent",
				 sbi->s_blocks_per_group,
				 sbi->s_clusters_per_group);
			goto failed_mount;
		}
	} else {
		if (clustersize != blocksize) {
			ext4_warning(sb, "fragment/cluster size (%d) != "
				     "block size (%d)", clustersize,
				     blocksize);
			clustersize = blocksize;
		}
		if (sbi->s_blocks_per_group > blocksize * 8) {
			ext4_msg(sb, KERN_ERR,
				 "#blocks per group too big: %lu",
				 sbi->s_blocks_per_group);
			goto failed_mount;
		}
		sbi->s_clusters_per_group = sbi->s_blocks_per_group;
		sbi->s_cluster_bits = 0;
3531
	}
3532 3533
	sbi->s_cluster_ratio = clustersize / blocksize;

3534
	if (sbi->s_inodes_per_group > blocksize * 8) {
3535 3536
		ext4_msg(sb, KERN_ERR,
		       "#inodes per group too big: %lu",
3537
		       sbi->s_inodes_per_group);
3538 3539 3540
		goto failed_mount;
	}

3541 3542 3543 3544
	/* Do we have standard group size of clustersize * 8 blocks ? */
	if (sbi->s_blocks_per_group == clustersize << 3)
		set_opt2(sb, STD_GROUP_SIZE);

3545 3546 3547 3548
	/*
	 * Test whether we have more sectors than will fit in sector_t,
	 * and whether the max offset is addressable by the page cache.
	 */
3549
	err = generic_check_addressable(sb->s_blocksize_bits,
3550
					ext4_blocks_count(es));
3551
	if (err) {
3552
		ext4_msg(sb, KERN_ERR, "filesystem"
3553
			 " too large to mount safely on this system");
3554
		if (sizeof(sector_t) < 8)
3555
			ext4_msg(sb, KERN_WARNING, "CONFIG_LBDAF not enabled");
3556 3557 3558
		goto failed_mount;
	}

3559 3560
	if (EXT4_BLOCKS_PER_GROUP(sb) == 0)
		goto cantfind_ext4;
3561

3562 3563 3564
	/* check blocks count against device size */
	blocks_count = sb->s_bdev->bd_inode->i_size >> sb->s_blocksize_bits;
	if (blocks_count && ext4_blocks_count(es) > blocks_count) {
3565 3566
		ext4_msg(sb, KERN_WARNING, "bad geometry: block count %llu "
		       "exceeds size of device (%llu blocks)",
3567 3568 3569 3570
		       ext4_blocks_count(es), blocks_count);
		goto failed_mount;
	}

3571 3572 3573 3574 3575
	/*
	 * It makes no sense for the first data block to be beyond the end
	 * of the filesystem.
	 */
	if (le32_to_cpu(es->s_first_data_block) >= ext4_blocks_count(es)) {
3576
		ext4_msg(sb, KERN_WARNING, "bad geometry: first data "
3577 3578 3579
			 "block %u is beyond end of filesystem (%llu)",
			 le32_to_cpu(es->s_first_data_block),
			 ext4_blocks_count(es));
3580 3581
		goto failed_mount;
	}
L
Laurent Vivier 已提交
3582 3583 3584 3585
	blocks_count = (ext4_blocks_count(es) -
			le32_to_cpu(es->s_first_data_block) +
			EXT4_BLOCKS_PER_GROUP(sb) - 1);
	do_div(blocks_count, EXT4_BLOCKS_PER_GROUP(sb));
3586
	if (blocks_count > ((uint64_t)1<<32) - EXT4_DESC_PER_BLOCK(sb)) {
3587
		ext4_msg(sb, KERN_WARNING, "groups count too large: %u "
3588
		       "(block count %llu, first data block %u, "
3589
		       "blocks per group %lu)", sbi->s_groups_count,
3590 3591 3592 3593 3594
		       ext4_blocks_count(es),
		       le32_to_cpu(es->s_first_data_block),
		       EXT4_BLOCKS_PER_GROUP(sb));
		goto failed_mount;
	}
L
Laurent Vivier 已提交
3595
	sbi->s_groups_count = blocks_count;
3596 3597
	sbi->s_blockfile_groups = min_t(ext4_group_t, sbi->s_groups_count,
			(EXT4_MAX_BLOCK_FILE_PHYS / EXT4_BLOCKS_PER_GROUP(sb)));
3598 3599
	db_count = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) /
		   EXT4_DESC_PER_BLOCK(sb);
3600 3601 3602
	sbi->s_group_desc = ext4_kvmalloc(db_count *
					  sizeof(struct buffer_head *),
					  GFP_KERNEL);
3603
	if (sbi->s_group_desc == NULL) {
3604
		ext4_msg(sb, KERN_ERR, "not enough memory");
3605
		ret = -ENOMEM;
3606 3607 3608
		goto failed_mount;
	}

3609
	bgl_lock_init(sbi->s_blockgroup_lock);
3610 3611

	for (i = 0; i < db_count; i++) {
3612
		block = descriptor_loc(sb, logical_sb_block, i);
3613
		sbi->s_group_desc[i] = sb_bread_unmovable(sb, block);
3614
		if (!sbi->s_group_desc[i]) {
3615 3616
			ext4_msg(sb, KERN_ERR,
			       "can't read group descriptor %d", i);
3617 3618 3619 3620
			db_count = i;
			goto failed_mount2;
		}
	}
3621
	if (!ext4_check_descriptors(sb, &first_not_zeroed)) {
3622
		ext4_msg(sb, KERN_ERR, "group descriptors corrupted!");
3623
		ret = -EFSCORRUPTED;
3624
		goto failed_mount2;
3625
	}
3626

3627
	sbi->s_gdb_count = db_count;
3628 3629 3630
	get_random_bytes(&sbi->s_next_generation, sizeof(u32));
	spin_lock_init(&sbi->s_next_gen_lock);

3631 3632
	setup_timer(&sbi->s_err_report, print_daily_error_info,
		(unsigned long) sb);
3633

3634
	/* Register extent status tree shrinker */
3635
	if (ext4_es_register_shrinker(sbi))
3636 3637
		goto failed_mount3;

3638
	sbi->s_stripe = ext4_get_stripe_size(sbi);
3639
	sbi->s_extent_max_zeroout_kb = 32;
3640

3641 3642 3643
	/*
	 * set up enough so that it can read an inode
	 */
3644
	sb->s_op = &ext4_sops;
3645 3646
	sb->s_export_op = &ext4_export_ops;
	sb->s_xattr = ext4_xattr_handlers;
3647
#ifdef CONFIG_QUOTA
3648
	sb->dq_op = &ext4_quota_operations;
3649
	if (ext4_has_feature_quota(sb))
3650
		sb->s_qcop = &dquot_quotactl_sysfile_ops;
3651 3652
	else
		sb->s_qcop = &ext4_qctl_operations;
J
Jan Kara 已提交
3653
	sb->s_quota_types = QTYPE_MASK_USR | QTYPE_MASK_GRP;
3654
#endif
3655 3656
	memcpy(sb->s_uuid, es->s_uuid, sizeof(es->s_uuid));

3657
	INIT_LIST_HEAD(&sbi->s_orphan); /* unlinked but open files */
3658
	mutex_init(&sbi->s_orphan_lock);
3659 3660 3661 3662

	sb->s_root = NULL;

	needs_recovery = (es->s_last_orphan != 0 ||
3663
			  ext4_has_feature_journal_needs_recovery(sb));
3664

3665
	if (ext4_has_feature_mmp(sb) && !(sb->s_flags & MS_RDONLY))
3666
		if (ext4_multi_mount_protect(sb, le64_to_cpu(es->s_mmp_block)))
3667
			goto failed_mount3a;
3668

3669 3670 3671 3672
	/*
	 * The first inode we look at is the journal inode.  Don't try
	 * root first: it may be modified in the journal!
	 */
3673
	if (!test_opt(sb, NOLOAD) && ext4_has_feature_journal(sb)) {
3674
		if (ext4_load_journal(sb, es, journal_devnum))
3675
			goto failed_mount3a;
3676
	} else if (test_opt(sb, NOLOAD) && !(sb->s_flags & MS_RDONLY) &&
3677
		   ext4_has_feature_journal_needs_recovery(sb)) {
3678 3679
		ext4_msg(sb, KERN_ERR, "required journal recovery "
		       "suppressed and not mounted read-only");
3680
		goto failed_mount_wq;
3681
	} else {
3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706
		/* Nojournal mode, all journal mount options are illegal */
		if (test_opt2(sb, EXPLICIT_JOURNAL_CHECKSUM)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "journal_checksum, fs mounted w/o journal");
			goto failed_mount_wq;
		}
		if (test_opt(sb, JOURNAL_ASYNC_COMMIT)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "journal_async_commit, fs mounted w/o journal");
			goto failed_mount_wq;
		}
		if (sbi->s_commit_interval != JBD2_DEFAULT_MAX_COMMIT_AGE*HZ) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "commit=%lu, fs mounted w/o journal",
				 sbi->s_commit_interval / HZ);
			goto failed_mount_wq;
		}
		if (EXT4_MOUNT_DATA_FLAGS &
		    (sbi->s_mount_opt ^ sbi->s_def_mount_opt)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "data=, fs mounted w/o journal");
			goto failed_mount_wq;
		}
		sbi->s_def_mount_opt &= EXT4_MOUNT_JOURNAL_CHECKSUM;
		clear_opt(sb, JOURNAL_CHECKSUM);
3707
		clear_opt(sb, DATA_FLAGS);
3708 3709 3710
		sbi->s_journal = NULL;
		needs_recovery = 0;
		goto no_journal;
3711 3712
	}

3713
	if (ext4_has_feature_64bit(sb) &&
3714 3715
	    !jbd2_journal_set_features(EXT4_SB(sb)->s_journal, 0, 0,
				       JBD2_FEATURE_INCOMPAT_64BIT)) {
3716
		ext4_msg(sb, KERN_ERR, "Failed to set 64-bit journal feature");
3717
		goto failed_mount_wq;
3718 3719
	}

3720 3721 3722 3723
	if (!set_journal_csum_feature_set(sb)) {
		ext4_msg(sb, KERN_ERR, "Failed to set journal checksum "
			 "feature set");
		goto failed_mount_wq;
3724
	}
3725

3726 3727 3728 3729 3730
	/* We have now updated the journal if required, so we can
	 * validate the data journaling mode. */
	switch (test_opt(sb, DATA_FLAGS)) {
	case 0:
		/* No mode set, assume a default based on the journal
A
Andrew Morton 已提交
3731 3732 3733
		 * capabilities: ORDERED_DATA if the journal can
		 * cope, else JOURNAL_DATA
		 */
3734 3735
		if (jbd2_journal_check_available_features
		    (sbi->s_journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE))
3736
			set_opt(sb, ORDERED_DATA);
3737
		else
3738
			set_opt(sb, JOURNAL_DATA);
3739 3740
		break;

3741 3742
	case EXT4_MOUNT_ORDERED_DATA:
	case EXT4_MOUNT_WRITEBACK_DATA:
3743 3744
		if (!jbd2_journal_check_available_features
		    (sbi->s_journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE)) {
3745 3746
			ext4_msg(sb, KERN_ERR, "Journal does not support "
			       "requested data journaling mode");
3747
			goto failed_mount_wq;
3748 3749 3750 3751
		}
	default:
		break;
	}
3752
	set_task_ioprio(sbi->s_journal->j_task, journal_ioprio);
3753

B
Bobi Jam 已提交
3754 3755
	sbi->s_journal->j_commit_callback = ext4_journal_commit_callback;

3756
no_journal:
3757 3758 3759 3760 3761 3762 3763 3764
	if (ext4_mballoc_ready) {
		sbi->s_mb_cache = ext4_xattr_create_cache(sb->s_id);
		if (!sbi->s_mb_cache) {
			ext4_msg(sb, KERN_ERR, "Failed to create an mb_cache");
			goto failed_mount_wq;
		}
	}

3765
	if ((DUMMY_ENCRYPTION_ENABLED(sbi) || ext4_has_feature_encrypt(sb)) &&
3766 3767 3768 3769 3770 3771
	    (blocksize != PAGE_CACHE_SIZE)) {
		ext4_msg(sb, KERN_ERR,
			 "Unsupported blocksize for fs encryption");
		goto failed_mount_wq;
	}

3772 3773 3774
	if (DUMMY_ENCRYPTION_ENABLED(sbi) && !(sb->s_flags & MS_RDONLY) &&
	    !ext4_has_feature_encrypt(sb)) {
		ext4_set_feature_encrypt(sb);
3775 3776 3777
		ext4_commit_super(sb, 1);
	}

3778 3779 3780 3781 3782 3783 3784
	/*
	 * Get the # of file system overhead blocks from the
	 * superblock if present.
	 */
	if (es->s_overhead_clusters)
		sbi->s_overhead = le32_to_cpu(es->s_overhead_clusters);
	else {
3785 3786
		err = ext4_calculate_overhead(sb);
		if (err)
3787 3788 3789
			goto failed_mount_wq;
	}

T
Tejun Heo 已提交
3790 3791 3792 3793
	/*
	 * The maximum number of concurrent works can be high and
	 * concurrency isn't really necessary.  Limit it to 1.
	 */
3794 3795 3796 3797
	EXT4_SB(sb)->rsv_conversion_wq =
		alloc_workqueue("ext4-rsv-conversion", WQ_MEM_RECLAIM | WQ_UNBOUND, 1);
	if (!EXT4_SB(sb)->rsv_conversion_wq) {
		printk(KERN_ERR "EXT4-fs: failed to create workqueue\n");
3798
		ret = -ENOMEM;
3799 3800 3801
		goto failed_mount4;
	}

3802
	/*
3803
	 * The jbd2_journal_load will have done any necessary log recovery,
3804 3805 3806
	 * so we can safely mount the rest of the filesystem now.
	 */

3807 3808
	root = ext4_iget(sb, EXT4_ROOT_INO);
	if (IS_ERR(root)) {
3809
		ext4_msg(sb, KERN_ERR, "get root inode failed");
3810
		ret = PTR_ERR(root);
3811
		root = NULL;
3812 3813 3814
		goto failed_mount4;
	}
	if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) {
3815
		ext4_msg(sb, KERN_ERR, "corrupt root inode, run e2fsck");
A
Al Viro 已提交
3816
		iput(root);
3817 3818
		goto failed_mount4;
	}
3819
	sb->s_root = d_make_root(root);
3820
	if (!sb->s_root) {
3821
		ext4_msg(sb, KERN_ERR, "get root dentry failed");
3822 3823 3824
		ret = -ENOMEM;
		goto failed_mount4;
	}
3825

3826 3827
	if (ext4_setup_super(sb, es, sb->s_flags & MS_RDONLY))
		sb->s_flags |= MS_RDONLY;
K
Kalpak Shah 已提交
3828 3829 3830 3831 3832

	/* determine the minimum size of new large inodes, if present */
	if (sbi->s_inode_size > EXT4_GOOD_OLD_INODE_SIZE) {
		sbi->s_want_extra_isize = sizeof(struct ext4_inode) -
						     EXT4_GOOD_OLD_INODE_SIZE;
3833
		if (ext4_has_feature_extra_isize(sb)) {
K
Kalpak Shah 已提交
3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848
			if (sbi->s_want_extra_isize <
			    le16_to_cpu(es->s_want_extra_isize))
				sbi->s_want_extra_isize =
					le16_to_cpu(es->s_want_extra_isize);
			if (sbi->s_want_extra_isize <
			    le16_to_cpu(es->s_min_extra_isize))
				sbi->s_want_extra_isize =
					le16_to_cpu(es->s_min_extra_isize);
		}
	}
	/* Check if enough inode space is available */
	if (EXT4_GOOD_OLD_INODE_SIZE + sbi->s_want_extra_isize >
							sbi->s_inode_size) {
		sbi->s_want_extra_isize = sizeof(struct ext4_inode) -
						       EXT4_GOOD_OLD_INODE_SIZE;
3849 3850
		ext4_msg(sb, KERN_INFO, "required extra inode space not"
			 "available");
K
Kalpak Shah 已提交
3851 3852
	}

3853
	ext4_set_resv_clusters(sb);
L
Lukas Czerner 已提交
3854

3855 3856
	err = ext4_setup_system_zone(sb);
	if (err) {
3857
		ext4_msg(sb, KERN_ERR, "failed to initialize system "
3858
			 "zone (%d)", err);
3859 3860 3861 3862 3863 3864 3865 3866
		goto failed_mount4a;
	}

	ext4_ext_init(sb);
	err = ext4_mb_init(sb);
	if (err) {
		ext4_msg(sb, KERN_ERR, "failed to initialize mballoc (%d)",
			 err);
3867
		goto failed_mount5;
3868 3869
	}

3870 3871 3872
	block = ext4_count_free_clusters(sb);
	ext4_free_blocks_count_set(sbi->s_es, 
				   EXT4_C2B(sbi, block));
3873 3874
	err = percpu_counter_init(&sbi->s_freeclusters_counter, block,
				  GFP_KERNEL);
3875 3876 3877
	if (!err) {
		unsigned long freei = ext4_count_free_inodes(sb);
		sbi->s_es->s_free_inodes_count = cpu_to_le32(freei);
3878 3879
		err = percpu_counter_init(&sbi->s_freeinodes_counter, freei,
					  GFP_KERNEL);
3880 3881 3882
	}
	if (!err)
		err = percpu_counter_init(&sbi->s_dirs_counter,
3883
					  ext4_count_dirs(sb), GFP_KERNEL);
3884
	if (!err)
3885 3886
		err = percpu_counter_init(&sbi->s_dirtyclusters_counter, 0,
					  GFP_KERNEL);
3887 3888 3889 3890 3891
	if (err) {
		ext4_msg(sb, KERN_ERR, "insufficient memory");
		goto failed_mount6;
	}

3892
	if (ext4_has_feature_flex_bg(sb))
3893 3894 3895 3896 3897 3898 3899
		if (!ext4_fill_flex_info(sb)) {
			ext4_msg(sb, KERN_ERR,
			       "unable to initialize "
			       "flex_bg meta info!");
			goto failed_mount6;
		}

3900 3901
	err = ext4_register_li_request(sb, first_not_zeroed);
	if (err)
3902
		goto failed_mount6;
3903

3904
	err = ext4_register_sysfs(sb);
3905 3906
	if (err)
		goto failed_mount7;
T
Theodore Ts'o 已提交
3907

3908 3909
#ifdef CONFIG_QUOTA
	/* Enable quota usage during mount. */
3910
	if (ext4_has_feature_quota(sb) && !(sb->s_flags & MS_RDONLY)) {
3911 3912 3913 3914 3915 3916
		err = ext4_enable_quotas(sb);
		if (err)
			goto failed_mount8;
	}
#endif  /* CONFIG_QUOTA */

3917 3918 3919
	EXT4_SB(sb)->s_mount_state |= EXT4_ORPHAN_FS;
	ext4_orphan_cleanup(sb, es);
	EXT4_SB(sb)->s_mount_state &= ~EXT4_ORPHAN_FS;
3920
	if (needs_recovery) {
3921
		ext4_msg(sb, KERN_INFO, "recovery complete");
3922 3923 3924 3925 3926 3927 3928 3929 3930 3931 3932 3933
		ext4_mark_recovery_complete(sb, es);
	}
	if (EXT4_SB(sb)->s_journal) {
		if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA)
			descr = " journalled data mode";
		else if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA)
			descr = " ordered data mode";
		else
			descr = " writeback data mode";
	} else
		descr = "out journal";

3934 3935 3936 3937 3938 3939 3940 3941
	if (test_opt(sb, DISCARD)) {
		struct request_queue *q = bdev_get_queue(sb->s_bdev);
		if (!blk_queue_discard(q))
			ext4_msg(sb, KERN_WARNING,
				 "mounting with \"discard\" option, but "
				 "the device does not support discard");
	}

3942 3943 3944 3945
	if (___ratelimit(&ext4_mount_msg_ratelimit, "EXT4-fs mount"))
		ext4_msg(sb, KERN_INFO, "mounted filesystem with%s. "
			 "Opts: %s%s%s", descr, sbi->s_es->s_mount_opts,
			 *sbi->s_es->s_mount_opts ? "; " : "", orig_data);
3946

3947 3948
	if (es->s_error_count)
		mod_timer(&sbi->s_err_report, jiffies + 300*HZ); /* 5 minutes */
3949

3950 3951 3952 3953 3954
	/* Enable message ratelimiting. Default is 10 messages per 5 secs. */
	ratelimit_state_init(&sbi->s_err_ratelimit_state, 5 * HZ, 10);
	ratelimit_state_init(&sbi->s_warning_ratelimit_state, 5 * HZ, 10);
	ratelimit_state_init(&sbi->s_msg_ratelimit_state, 5 * HZ, 10);

3955
	kfree(orig_data);
3956 3957
	return 0;

3958
cantfind_ext4:
3959
	if (!silent)
3960
		ext4_msg(sb, KERN_ERR, "VFS: Can't find ext4 filesystem");
3961 3962
	goto failed_mount;

3963 3964
#ifdef CONFIG_QUOTA
failed_mount8:
3965
	ext4_unregister_sysfs(sb);
3966
#endif
3967 3968 3969
failed_mount7:
	ext4_unregister_li_request(sb);
failed_mount6:
3970
	ext4_mb_release(sb);
3971
	if (sbi->s_flex_groups)
A
Al Viro 已提交
3972
		kvfree(sbi->s_flex_groups);
3973 3974 3975 3976
	percpu_counter_destroy(&sbi->s_freeclusters_counter);
	percpu_counter_destroy(&sbi->s_freeinodes_counter);
	percpu_counter_destroy(&sbi->s_dirs_counter);
	percpu_counter_destroy(&sbi->s_dirtyclusters_counter);
3977
failed_mount5:
3978 3979 3980
	ext4_ext_release(sb);
	ext4_release_system_zone(sb);
failed_mount4a:
A
Al Viro 已提交
3981
	dput(sb->s_root);
3982
	sb->s_root = NULL;
A
Al Viro 已提交
3983
failed_mount4:
3984
	ext4_msg(sb, KERN_ERR, "mount failed");
3985 3986
	if (EXT4_SB(sb)->rsv_conversion_wq)
		destroy_workqueue(EXT4_SB(sb)->rsv_conversion_wq);
3987
failed_mount_wq:
3988 3989 3990 3991
	if (sbi->s_journal) {
		jbd2_journal_destroy(sbi->s_journal);
		sbi->s_journal = NULL;
	}
3992
failed_mount3a:
3993
	ext4_es_unregister_shrinker(sbi);
3994
failed_mount3:
3995
	del_timer_sync(&sbi->s_err_report);
3996 3997
	if (sbi->s_mmp_tsk)
		kthread_stop(sbi->s_mmp_tsk);
3998 3999 4000
failed_mount2:
	for (i = 0; i < db_count; i++)
		brelse(sbi->s_group_desc[i]);
A
Al Viro 已提交
4001
	kvfree(sbi->s_group_desc);
4002
failed_mount:
4003 4004
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
4005
#ifdef CONFIG_QUOTA
J
Jan Kara 已提交
4006
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
4007 4008
		kfree(sbi->s_qf_names[i]);
#endif
4009
	ext4_blkdev_remove(sbi);
4010 4011 4012
	brelse(bh);
out_fail:
	sb->s_fs_info = NULL;
4013
	kfree(sbi->s_blockgroup_lock);
4014
	kfree(sbi);
4015
out_free_orig:
4016
	kfree(orig_data);
4017
	return err ? err : ret;
4018 4019 4020 4021 4022 4023 4024
}

/*
 * Setup any per-fs journal parameters now.  We'll do this both on
 * initial mount, once the journal has been initialised but before we've
 * done any recovery; and again on any subsequent remount.
 */
4025
static void ext4_init_journal_params(struct super_block *sb, journal_t *journal)
4026
{
4027
	struct ext4_sb_info *sbi = EXT4_SB(sb);
4028

4029 4030 4031
	journal->j_commit_interval = sbi->s_commit_interval;
	journal->j_min_batch_time = sbi->s_min_batch_time;
	journal->j_max_batch_time = sbi->s_max_batch_time;
4032

4033
	write_lock(&journal->j_state_lock);
4034
	if (test_opt(sb, BARRIER))
4035
		journal->j_flags |= JBD2_BARRIER;
4036
	else
4037
		journal->j_flags &= ~JBD2_BARRIER;
4038 4039 4040 4041
	if (test_opt(sb, DATA_ERR_ABORT))
		journal->j_flags |= JBD2_ABORT_ON_SYNCDATA_ERR;
	else
		journal->j_flags &= ~JBD2_ABORT_ON_SYNCDATA_ERR;
4042
	write_unlock(&journal->j_state_lock);
4043 4044
}

4045
static journal_t *ext4_get_journal(struct super_block *sb,
4046 4047 4048 4049 4050
				   unsigned int journal_inum)
{
	struct inode *journal_inode;
	journal_t *journal;

4051
	BUG_ON(!ext4_has_feature_journal(sb));
4052

4053 4054 4055 4056
	/* First, test for the existence of a valid inode on disk.  Bad
	 * things happen if we iget() an unused inode, as the subsequent
	 * iput() will try to delete it. */

4057 4058
	journal_inode = ext4_iget(sb, journal_inum);
	if (IS_ERR(journal_inode)) {
4059
		ext4_msg(sb, KERN_ERR, "no journal found");
4060 4061 4062 4063 4064
		return NULL;
	}
	if (!journal_inode->i_nlink) {
		make_bad_inode(journal_inode);
		iput(journal_inode);
4065
		ext4_msg(sb, KERN_ERR, "journal inode is deleted");
4066 4067 4068
		return NULL;
	}

4069
	jbd_debug(2, "Journal inode found at %p: %lld bytes\n",
4070
		  journal_inode, journal_inode->i_size);
4071
	if (!S_ISREG(journal_inode->i_mode)) {
4072
		ext4_msg(sb, KERN_ERR, "invalid journal inode");
4073 4074 4075 4076
		iput(journal_inode);
		return NULL;
	}

4077
	journal = jbd2_journal_init_inode(journal_inode);
4078
	if (!journal) {
4079
		ext4_msg(sb, KERN_ERR, "Could not load journal inode");
4080 4081 4082 4083
		iput(journal_inode);
		return NULL;
	}
	journal->j_private = sb;
4084
	ext4_init_journal_params(sb, journal);
4085 4086 4087
	return journal;
}

4088
static journal_t *ext4_get_dev_journal(struct super_block *sb,
4089 4090
				       dev_t j_dev)
{
4091
	struct buffer_head *bh;
4092
	journal_t *journal;
4093 4094
	ext4_fsblk_t start;
	ext4_fsblk_t len;
4095
	int hblock, blocksize;
4096
	ext4_fsblk_t sb_block;
4097
	unsigned long offset;
4098
	struct ext4_super_block *es;
4099 4100
	struct block_device *bdev;

4101
	BUG_ON(!ext4_has_feature_journal(sb));
4102

4103
	bdev = ext4_blkdev_get(j_dev, sb);
4104 4105 4106 4107
	if (bdev == NULL)
		return NULL;

	blocksize = sb->s_blocksize;
4108
	hblock = bdev_logical_block_size(bdev);
4109
	if (blocksize < hblock) {
4110 4111
		ext4_msg(sb, KERN_ERR,
			"blocksize too small for journal device");
4112 4113 4114
		goto out_bdev;
	}

4115 4116
	sb_block = EXT4_MIN_BLOCK_SIZE / blocksize;
	offset = EXT4_MIN_BLOCK_SIZE % blocksize;
4117 4118
	set_blocksize(bdev, blocksize);
	if (!(bh = __bread(bdev, sb_block, blocksize))) {
4119 4120
		ext4_msg(sb, KERN_ERR, "couldn't read superblock of "
		       "external journal");
4121 4122 4123
		goto out_bdev;
	}

4124
	es = (struct ext4_super_block *) (bh->b_data + offset);
4125
	if ((le16_to_cpu(es->s_magic) != EXT4_SUPER_MAGIC) ||
4126
	    !(le32_to_cpu(es->s_feature_incompat) &
4127
	      EXT4_FEATURE_INCOMPAT_JOURNAL_DEV)) {
4128 4129
		ext4_msg(sb, KERN_ERR, "external journal has "
					"bad superblock");
4130 4131 4132 4133
		brelse(bh);
		goto out_bdev;
	}

4134 4135 4136 4137 4138 4139 4140 4141 4142
	if ((le32_to_cpu(es->s_feature_ro_compat) &
	     EXT4_FEATURE_RO_COMPAT_METADATA_CSUM) &&
	    es->s_checksum != ext4_superblock_csum(sb, es)) {
		ext4_msg(sb, KERN_ERR, "external journal has "
				       "corrupt superblock");
		brelse(bh);
		goto out_bdev;
	}

4143
	if (memcmp(EXT4_SB(sb)->s_es->s_journal_uuid, es->s_uuid, 16)) {
4144
		ext4_msg(sb, KERN_ERR, "journal UUID does not match");
4145 4146 4147 4148
		brelse(bh);
		goto out_bdev;
	}

L
Laurent Vivier 已提交
4149
	len = ext4_blocks_count(es);
4150 4151 4152
	start = sb_block + 1;
	brelse(bh);	/* we're done with the superblock */

4153
	journal = jbd2_journal_init_dev(bdev, sb->s_bdev,
4154 4155
					start, len, blocksize);
	if (!journal) {
4156
		ext4_msg(sb, KERN_ERR, "failed to create device journal");
4157 4158 4159
		goto out_bdev;
	}
	journal->j_private = sb;
4160
	ll_rw_block(READ | REQ_META | REQ_PRIO, 1, &journal->j_sb_buffer);
4161 4162
	wait_on_buffer(journal->j_sb_buffer);
	if (!buffer_uptodate(journal->j_sb_buffer)) {
4163
		ext4_msg(sb, KERN_ERR, "I/O error on journal device");
4164 4165 4166
		goto out_journal;
	}
	if (be32_to_cpu(journal->j_superblock->s_nr_users) != 1) {
4167 4168
		ext4_msg(sb, KERN_ERR, "External journal has more than one "
					"user (unsupported) - %d",
4169 4170 4171
			be32_to_cpu(journal->j_superblock->s_nr_users));
		goto out_journal;
	}
4172 4173
	EXT4_SB(sb)->journal_bdev = bdev;
	ext4_init_journal_params(sb, journal);
4174
	return journal;
4175

4176
out_journal:
4177
	jbd2_journal_destroy(journal);
4178
out_bdev:
4179
	ext4_blkdev_put(bdev);
4180 4181 4182
	return NULL;
}

4183 4184
static int ext4_load_journal(struct super_block *sb,
			     struct ext4_super_block *es,
4185 4186 4187 4188 4189 4190 4191 4192
			     unsigned long journal_devnum)
{
	journal_t *journal;
	unsigned int journal_inum = le32_to_cpu(es->s_journal_inum);
	dev_t journal_dev;
	int err = 0;
	int really_read_only;

4193
	BUG_ON(!ext4_has_feature_journal(sb));
4194

4195 4196
	if (journal_devnum &&
	    journal_devnum != le32_to_cpu(es->s_journal_dev)) {
4197 4198
		ext4_msg(sb, KERN_INFO, "external journal device major/minor "
			"numbers have changed");
4199 4200 4201 4202 4203 4204 4205 4206 4207 4208 4209
		journal_dev = new_decode_dev(journal_devnum);
	} else
		journal_dev = new_decode_dev(le32_to_cpu(es->s_journal_dev));

	really_read_only = bdev_read_only(sb->s_bdev);

	/*
	 * Are we loading a blank journal or performing recovery after a
	 * crash?  For recovery, we need to check in advance whether we
	 * can get read-write access to the device.
	 */
4210
	if (ext4_has_feature_journal_needs_recovery(sb)) {
4211
		if (sb->s_flags & MS_RDONLY) {
4212 4213
			ext4_msg(sb, KERN_INFO, "INFO: recovery "
					"required on readonly filesystem");
4214
			if (really_read_only) {
4215 4216
				ext4_msg(sb, KERN_ERR, "write access "
					"unavailable, cannot proceed");
4217 4218
				return -EROFS;
			}
4219 4220
			ext4_msg(sb, KERN_INFO, "write access will "
			       "be enabled during recovery");
4221 4222 4223 4224
		}
	}

	if (journal_inum && journal_dev) {
4225 4226
		ext4_msg(sb, KERN_ERR, "filesystem has both journal "
		       "and inode journals!");
4227 4228 4229 4230
		return -EINVAL;
	}

	if (journal_inum) {
4231
		if (!(journal = ext4_get_journal(sb, journal_inum)))
4232 4233
			return -EINVAL;
	} else {
4234
		if (!(journal = ext4_get_dev_journal(sb, journal_dev)))
4235 4236 4237
			return -EINVAL;
	}

4238
	if (!(journal->j_flags & JBD2_BARRIER))
4239
		ext4_msg(sb, KERN_INFO, "barriers disabled");
4240

4241
	if (!ext4_has_feature_journal_needs_recovery(sb))
4242
		err = jbd2_journal_wipe(journal, !really_read_only);
4243 4244 4245 4246 4247
	if (!err) {
		char *save = kmalloc(EXT4_S_ERR_LEN, GFP_KERNEL);
		if (save)
			memcpy(save, ((char *) es) +
			       EXT4_S_ERR_START, EXT4_S_ERR_LEN);
4248
		err = jbd2_journal_load(journal);
4249 4250 4251 4252 4253
		if (save)
			memcpy(((char *) es) + EXT4_S_ERR_START,
			       save, EXT4_S_ERR_LEN);
		kfree(save);
	}
4254 4255

	if (err) {
4256
		ext4_msg(sb, KERN_ERR, "error loading journal");
4257
		jbd2_journal_destroy(journal);
4258 4259 4260
		return err;
	}

4261 4262
	EXT4_SB(sb)->s_journal = journal;
	ext4_clear_journal_err(sb, es);
4263

4264
	if (!really_read_only && journal_devnum &&
4265 4266 4267 4268
	    journal_devnum != le32_to_cpu(es->s_journal_dev)) {
		es->s_journal_dev = cpu_to_le32(journal_devnum);

		/* Make sure we flush the recovery flag to disk. */
4269
		ext4_commit_super(sb, 1);
4270 4271 4272 4273 4274
	}

	return 0;
}

4275
static int ext4_commit_super(struct super_block *sb, int sync)
4276
{
4277
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;
4278
	struct buffer_head *sbh = EXT4_SB(sb)->s_sbh;
4279
	int error = 0;
4280

4281
	if (!sbh || block_device_ejected(sb))
4282
		return error;
4283 4284 4285 4286 4287 4288 4289 4290 4291
	if (buffer_write_io_error(sbh)) {
		/*
		 * Oh, dear.  A previous attempt to write the
		 * superblock failed.  This could happen because the
		 * USB device was yanked out.  Or it could happen to
		 * be a transient write error and maybe the block will
		 * be remapped.  Nothing we can do but to retry the
		 * write and hope for the best.
		 */
4292 4293
		ext4_msg(sb, KERN_ERR, "previous I/O error to "
		       "superblock detected");
4294 4295 4296
		clear_buffer_write_io_error(sbh);
		set_buffer_uptodate(sbh);
	}
4297 4298 4299 4300 4301 4302 4303 4304 4305 4306 4307 4308
	/*
	 * If the file system is mounted read-only, don't update the
	 * superblock write time.  This avoids updating the superblock
	 * write time when we are mounting the root file system
	 * read/only but we need to replay the journal; at that point,
	 * for people who are east of GMT and who make their clock
	 * tick in localtime for Windows bug-for-bug compatibility,
	 * the clock is set in the future, and this will cause e2fsck
	 * to complain and force a full file system check.
	 */
	if (!(sb->s_flags & MS_RDONLY))
		es->s_wtime = cpu_to_le32(get_seconds());
4309 4310 4311
	if (sb->s_bdev->bd_part)
		es->s_kbytes_written =
			cpu_to_le64(EXT4_SB(sb)->s_kbytes_written +
T
Theodore Ts'o 已提交
4312 4313
			    ((part_stat_read(sb->s_bdev->bd_part, sectors[1]) -
			      EXT4_SB(sb)->s_sectors_written_start) >> 1));
4314 4315 4316
	else
		es->s_kbytes_written =
			cpu_to_le64(EXT4_SB(sb)->s_kbytes_written);
4317 4318
	if (percpu_counter_initialized(&EXT4_SB(sb)->s_freeclusters_counter))
		ext4_free_blocks_count_set(es,
4319 4320
			EXT4_C2B(EXT4_SB(sb), percpu_counter_sum_positive(
				&EXT4_SB(sb)->s_freeclusters_counter)));
4321 4322 4323
	if (percpu_counter_initialized(&EXT4_SB(sb)->s_freeinodes_counter))
		es->s_free_inodes_count =
			cpu_to_le32(percpu_counter_sum_positive(
4324
				&EXT4_SB(sb)->s_freeinodes_counter));
4325
	BUFFER_TRACE(sbh, "marking dirty");
4326
	ext4_superblock_csum_set(sb);
4327
	mark_buffer_dirty(sbh);
4328
	if (sync) {
4329 4330
		error = __sync_dirty_buffer(sbh,
			test_opt(sb, BARRIER) ? WRITE_FUA : WRITE_SYNC);
4331 4332 4333 4334 4335
		if (error)
			return error;

		error = buffer_write_io_error(sbh);
		if (error) {
4336 4337
			ext4_msg(sb, KERN_ERR, "I/O error while writing "
			       "superblock");
4338 4339 4340 4341
			clear_buffer_write_io_error(sbh);
			set_buffer_uptodate(sbh);
		}
	}
4342
	return error;
4343 4344 4345 4346 4347 4348 4349
}

/*
 * Have we just finished recovery?  If so, and if we are mounting (or
 * remounting) the filesystem readonly, then we will end up with a
 * consistent fs on disk.  Record that fact.
 */
4350 4351
static void ext4_mark_recovery_complete(struct super_block *sb,
					struct ext4_super_block *es)
4352
{
4353
	journal_t *journal = EXT4_SB(sb)->s_journal;
4354

4355
	if (!ext4_has_feature_journal(sb)) {
4356 4357 4358
		BUG_ON(journal != NULL);
		return;
	}
4359
	jbd2_journal_lock_updates(journal);
4360 4361 4362
	if (jbd2_journal_flush(journal) < 0)
		goto out;

4363
	if (ext4_has_feature_journal_needs_recovery(sb) &&
4364
	    sb->s_flags & MS_RDONLY) {
4365
		ext4_clear_feature_journal_needs_recovery(sb);
4366
		ext4_commit_super(sb, 1);
4367
	}
4368 4369

out:
4370
	jbd2_journal_unlock_updates(journal);
4371 4372 4373 4374 4375 4376 4377
}

/*
 * If we are mounting (or read-write remounting) a filesystem whose journal
 * has recorded an error from a previous lifetime, move that error to the
 * main filesystem now.
 */
4378 4379
static void ext4_clear_journal_err(struct super_block *sb,
				   struct ext4_super_block *es)
4380 4381 4382 4383 4384
{
	journal_t *journal;
	int j_errno;
	const char *errstr;

4385
	BUG_ON(!ext4_has_feature_journal(sb));
4386

4387
	journal = EXT4_SB(sb)->s_journal;
4388 4389 4390

	/*
	 * Now check for any error status which may have been recorded in the
4391
	 * journal by a prior ext4_error() or ext4_abort()
4392 4393
	 */

4394
	j_errno = jbd2_journal_errno(journal);
4395 4396 4397
	if (j_errno) {
		char nbuf[16];

4398
		errstr = ext4_decode_error(sb, j_errno, nbuf);
4399
		ext4_warning(sb, "Filesystem error recorded "
4400
			     "from previous mount: %s", errstr);
4401
		ext4_warning(sb, "Marking fs in need of filesystem check.");
4402

4403 4404
		EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS;
		es->s_state |= cpu_to_le16(EXT4_ERROR_FS);
4405
		ext4_commit_super(sb, 1);
4406

4407
		jbd2_journal_clear_err(journal);
4408
		jbd2_journal_update_sb_errno(journal);
4409 4410 4411 4412 4413 4414 4415
	}
}

/*
 * Force the running and committing transactions to commit,
 * and wait on the commit.
 */
4416
int ext4_force_commit(struct super_block *sb)
4417 4418 4419 4420 4421 4422
{
	journal_t *journal;

	if (sb->s_flags & MS_RDONLY)
		return 0;

4423
	journal = EXT4_SB(sb)->s_journal;
4424
	return ext4_journal_force_commit(journal);
4425 4426
}

4427
static int ext4_sync_fs(struct super_block *sb, int wait)
4428
{
4429
	int ret = 0;
4430
	tid_t target;
4431
	bool needs_barrier = false;
4432
	struct ext4_sb_info *sbi = EXT4_SB(sb);
4433

4434
	trace_ext4_sync_fs(sb, wait);
4435
	flush_workqueue(sbi->rsv_conversion_wq);
4436 4437 4438 4439 4440
	/*
	 * Writeback quota in non-journalled quota case - journalled quota has
	 * no dirty dquots
	 */
	dquot_writeback_dquots(sb, -1);
4441 4442 4443 4444 4445
	/*
	 * Data writeback is possible w/o journal transaction, so barrier must
	 * being sent at the end of the function. But we can skip it if
	 * transaction_commit will do it for us.
	 */
4446 4447 4448 4449 4450 4451 4452 4453 4454 4455 4456 4457
	if (sbi->s_journal) {
		target = jbd2_get_latest_transaction(sbi->s_journal);
		if (wait && sbi->s_journal->j_flags & JBD2_BARRIER &&
		    !jbd2_trans_will_send_data_barrier(sbi->s_journal, target))
			needs_barrier = true;

		if (jbd2_journal_start_commit(sbi->s_journal, &target)) {
			if (wait)
				ret = jbd2_log_wait_commit(sbi->s_journal,
							   target);
		}
	} else if (wait && test_opt(sb, BARRIER))
4458 4459 4460 4461 4462 4463
		needs_barrier = true;
	if (needs_barrier) {
		int err;
		err = blkdev_issue_flush(sb->s_bdev, GFP_KERNEL, NULL);
		if (!ret)
			ret = err;
4464
	}
4465 4466 4467 4468

	return ret;
}

4469 4470 4471
/*
 * LVM calls this function before a (read-only) snapshot is created.  This
 * gives us a chance to flush the journal completely and mark the fs clean.
4472 4473
 *
 * Note that only this function cannot bring a filesystem to be in a clean
4474 4475
 * state independently. It relies on upper layer to stop all data & metadata
 * modifications.
4476
 */
4477
static int ext4_freeze(struct super_block *sb)
4478
{
4479 4480
	int error = 0;
	journal_t *journal;
4481

4482 4483
	if (sb->s_flags & MS_RDONLY)
		return 0;
4484

4485
	journal = EXT4_SB(sb)->s_journal;
4486

4487 4488 4489
	if (journal) {
		/* Now we set up the journal barrier. */
		jbd2_journal_lock_updates(journal);
4490

4491 4492 4493 4494 4495 4496 4497
		/*
		 * Don't clear the needs_recovery flag if we failed to
		 * flush the journal.
		 */
		error = jbd2_journal_flush(journal);
		if (error < 0)
			goto out;
4498 4499

		/* Journal blocked and flushed, clear needs_recovery flag. */
4500
		ext4_clear_feature_journal_needs_recovery(sb);
4501
	}
4502 4503

	error = ext4_commit_super(sb, 1);
4504
out:
4505 4506 4507
	if (journal)
		/* we rely on upper layer to stop further updates */
		jbd2_journal_unlock_updates(journal);
4508
	return error;
4509 4510 4511 4512 4513 4514
}

/*
 * Called by LVM after the snapshot is done.  We need to reset the RECOVER
 * flag here, even though the filesystem is not technically dirty yet.
 */
4515
static int ext4_unfreeze(struct super_block *sb)
4516
{
4517 4518 4519
	if (sb->s_flags & MS_RDONLY)
		return 0;

4520 4521
	if (EXT4_SB(sb)->s_journal) {
		/* Reset the needs_recovery flag before the fs is unlocked. */
4522
		ext4_set_feature_journal_needs_recovery(sb);
4523 4524
	}

4525
	ext4_commit_super(sb, 1);
4526
	return 0;
4527 4528
}

4529 4530 4531 4532 4533
/*
 * Structure to save mount options for ext4_remount's benefit
 */
struct ext4_mount_options {
	unsigned long s_mount_opt;
4534
	unsigned long s_mount_opt2;
4535 4536
	kuid_t s_resuid;
	kgid_t s_resgid;
4537 4538 4539 4540
	unsigned long s_commit_interval;
	u32 s_min_batch_time, s_max_batch_time;
#ifdef CONFIG_QUOTA
	int s_jquota_fmt;
J
Jan Kara 已提交
4541
	char *s_qf_names[EXT4_MAXQUOTAS];
4542 4543 4544
#endif
};

4545
static int ext4_remount(struct super_block *sb, int *flags, char *data)
4546
{
4547
	struct ext4_super_block *es;
4548
	struct ext4_sb_info *sbi = EXT4_SB(sb);
4549
	unsigned long old_sb_flags;
4550
	struct ext4_mount_options old_opts;
4551
	int enable_quota = 0;
4552
	ext4_group_t g;
4553
	unsigned int journal_ioprio = DEFAULT_JOURNAL_IOPRIO;
4554
	int err = 0;
4555
#ifdef CONFIG_QUOTA
4556
	int i, j;
4557
#endif
4558
	char *orig_data = kstrdup(data, GFP_KERNEL);
4559 4560 4561 4562

	/* Store the original options */
	old_sb_flags = sb->s_flags;
	old_opts.s_mount_opt = sbi->s_mount_opt;
4563
	old_opts.s_mount_opt2 = sbi->s_mount_opt2;
4564 4565 4566
	old_opts.s_resuid = sbi->s_resuid;
	old_opts.s_resgid = sbi->s_resgid;
	old_opts.s_commit_interval = sbi->s_commit_interval;
4567 4568
	old_opts.s_min_batch_time = sbi->s_min_batch_time;
	old_opts.s_max_batch_time = sbi->s_max_batch_time;
4569 4570
#ifdef CONFIG_QUOTA
	old_opts.s_jquota_fmt = sbi->s_jquota_fmt;
J
Jan Kara 已提交
4571
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
4572 4573 4574 4575 4576 4577
		if (sbi->s_qf_names[i]) {
			old_opts.s_qf_names[i] = kstrdup(sbi->s_qf_names[i],
							 GFP_KERNEL);
			if (!old_opts.s_qf_names[i]) {
				for (j = 0; j < i; j++)
					kfree(old_opts.s_qf_names[j]);
4578
				kfree(orig_data);
4579 4580 4581 4582
				return -ENOMEM;
			}
		} else
			old_opts.s_qf_names[i] = NULL;
4583
#endif
4584 4585
	if (sbi->s_journal && sbi->s_journal->j_task->io_context)
		journal_ioprio = sbi->s_journal->j_task->io_context->ioprio;
4586

4587
	if (!parse_options(data, sb, NULL, &journal_ioprio, 1)) {
4588 4589 4590 4591
		err = -EINVAL;
		goto restore_opts;
	}

4592
	if ((old_opts.s_mount_opt & EXT4_MOUNT_JOURNAL_CHECKSUM) ^
4593 4594
	    test_opt(sb, JOURNAL_CHECKSUM)) {
		ext4_msg(sb, KERN_ERR, "changing journal_checksum "
4595 4596
			 "during remount not supported; ignoring");
		sbi->s_mount_opt ^= EXT4_MOUNT_JOURNAL_CHECKSUM;
4597 4598
	}

4599 4600 4601 4602 4603 4604 4605 4606 4607 4608 4609 4610 4611
	if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) {
		if (test_opt2(sb, EXPLICIT_DELALLOC)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "both data=journal and delalloc");
			err = -EINVAL;
			goto restore_opts;
		}
		if (test_opt(sb, DIOREAD_NOLOCK)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "both data=journal and dioread_nolock");
			err = -EINVAL;
			goto restore_opts;
		}
R
Ross Zwisler 已提交
4612 4613 4614 4615 4616 4617 4618 4619 4620 4621 4622 4623
		if (test_opt(sb, DAX)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "both data=journal and dax");
			err = -EINVAL;
			goto restore_opts;
		}
	}

	if ((sbi->s_mount_opt ^ old_opts.s_mount_opt) & EXT4_MOUNT_DAX) {
		ext4_msg(sb, KERN_WARNING, "warning: refusing change of "
			"dax flag with busy inodes while remounting");
		sbi->s_mount_opt ^= EXT4_MOUNT_DAX;
4624 4625
	}

4626
	if (sbi->s_mount_flags & EXT4_MF_FS_ABORTED)
4627
		ext4_abort(sb, "Abort forced by user");
4628 4629

	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
4630
		(test_opt(sb, POSIX_ACL) ? MS_POSIXACL : 0);
4631 4632 4633

	es = sbi->s_es;

4634
	if (sbi->s_journal) {
4635
		ext4_init_journal_params(sb, sbi->s_journal);
4636 4637
		set_task_ioprio(sbi->s_journal->j_task, journal_ioprio);
	}
4638

4639 4640 4641
	if (*flags & MS_LAZYTIME)
		sb->s_flags |= MS_LAZYTIME;

4642
	if ((*flags & MS_RDONLY) != (sb->s_flags & MS_RDONLY)) {
4643
		if (sbi->s_mount_flags & EXT4_MF_FS_ABORTED) {
4644 4645 4646 4647 4648
			err = -EROFS;
			goto restore_opts;
		}

		if (*flags & MS_RDONLY) {
4649 4650 4651
			err = sync_filesystem(sb);
			if (err < 0)
				goto restore_opts;
4652 4653
			err = dquot_suspend(sb, -1);
			if (err < 0)
4654 4655
				goto restore_opts;

4656 4657 4658 4659 4660 4661 4662 4663 4664 4665 4666
			/*
			 * First of all, the unconditional stuff we have to do
			 * to disable replay of the journal when we next remount
			 */
			sb->s_flags |= MS_RDONLY;

			/*
			 * OK, test if we are remounting a valid rw partition
			 * readonly, and if so set the rdonly flag and then
			 * mark the partition as valid again.
			 */
4667 4668
			if (!(es->s_state & cpu_to_le16(EXT4_VALID_FS)) &&
			    (sbi->s_mount_state & EXT4_VALID_FS))
4669 4670
				es->s_state = cpu_to_le16(sbi->s_mount_state);

4671
			if (sbi->s_journal)
4672
				ext4_mark_recovery_complete(sb, es);
4673
		} else {
4674
			/* Make sure we can mount this feature set readwrite */
4675
			if (ext4_has_feature_readonly(sb) ||
D
Darrick J. Wong 已提交
4676
			    !ext4_feature_set_ok(sb, 0)) {
4677 4678 4679
				err = -EROFS;
				goto restore_opts;
			}
4680 4681
			/*
			 * Make sure the group descriptor checksums
4682
			 * are sane.  If they aren't, refuse to remount r/w.
4683 4684 4685 4686 4687
			 */
			for (g = 0; g < sbi->s_groups_count; g++) {
				struct ext4_group_desc *gdp =
					ext4_get_group_desc(sb, g, NULL);

4688
				if (!ext4_group_desc_csum_verify(sb, g, gdp)) {
4689 4690
					ext4_msg(sb, KERN_ERR,
	       "ext4_remount: Checksum for group %u failed (%u!=%u)",
4691
		g, le16_to_cpu(ext4_group_desc_csum(sb, g, gdp)),
4692
					       le16_to_cpu(gdp->bg_checksum));
4693
					err = -EFSBADCRC;
4694 4695 4696 4697
					goto restore_opts;
				}
			}

4698 4699 4700 4701 4702 4703
			/*
			 * If we have an unprocessed orphan list hanging
			 * around from a previously readonly bdev mount,
			 * require a full umount/remount for now.
			 */
			if (es->s_last_orphan) {
4704
				ext4_msg(sb, KERN_WARNING, "Couldn't "
4705 4706
				       "remount RDWR because of unprocessed "
				       "orphan inode list.  Please "
4707
				       "umount/remount instead");
4708 4709 4710 4711
				err = -EINVAL;
				goto restore_opts;
			}

4712 4713 4714 4715 4716 4717
			/*
			 * Mounting a RDONLY partition read-write, so reread
			 * and store the current valid flag.  (It may have
			 * been changed by e2fsck since we originally mounted
			 * the partition.)
			 */
4718 4719
			if (sbi->s_journal)
				ext4_clear_journal_err(sb, es);
4720
			sbi->s_mount_state = le16_to_cpu(es->s_state);
4721
			if (!ext4_setup_super(sb, es, 0))
4722
				sb->s_flags &= ~MS_RDONLY;
4723
			if (ext4_has_feature_mmp(sb))
4724 4725 4726 4727 4728
				if (ext4_multi_mount_protect(sb,
						le64_to_cpu(es->s_mmp_block))) {
					err = -EROFS;
					goto restore_opts;
				}
4729
			enable_quota = 1;
4730 4731
		}
	}
4732 4733 4734 4735 4736 4737 4738 4739 4740 4741 4742 4743 4744

	/*
	 * Reinitialize lazy itable initialization thread based on
	 * current settings
	 */
	if ((sb->s_flags & MS_RDONLY) || !test_opt(sb, INIT_INODE_TABLE))
		ext4_unregister_li_request(sb);
	else {
		ext4_group_t first_not_zeroed;
		first_not_zeroed = ext4_has_uninit_itable(sb);
		ext4_register_li_request(sb, first_not_zeroed);
	}

4745
	ext4_setup_system_zone(sb);
4746
	if (sbi->s_journal == NULL && !(old_sb_flags & MS_RDONLY))
4747
		ext4_commit_super(sb, 1);
4748

4749 4750
#ifdef CONFIG_QUOTA
	/* Release old quota file names */
J
Jan Kara 已提交
4751
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
4752
		kfree(old_opts.s_qf_names[i]);
4753 4754 4755
	if (enable_quota) {
		if (sb_any_quota_suspended(sb))
			dquot_resume(sb, -1);
4756
		else if (ext4_has_feature_quota(sb)) {
4757
			err = ext4_enable_quotas(sb);
4758
			if (err)
4759 4760 4761
				goto restore_opts;
		}
	}
4762
#endif
4763

4764
	*flags = (*flags & ~MS_LAZYTIME) | (sb->s_flags & MS_LAZYTIME);
4765 4766
	ext4_msg(sb, KERN_INFO, "re-mounted. Opts: %s", orig_data);
	kfree(orig_data);
4767
	return 0;
4768

4769 4770 4771
restore_opts:
	sb->s_flags = old_sb_flags;
	sbi->s_mount_opt = old_opts.s_mount_opt;
4772
	sbi->s_mount_opt2 = old_opts.s_mount_opt2;
4773 4774 4775
	sbi->s_resuid = old_opts.s_resuid;
	sbi->s_resgid = old_opts.s_resgid;
	sbi->s_commit_interval = old_opts.s_commit_interval;
4776 4777
	sbi->s_min_batch_time = old_opts.s_min_batch_time;
	sbi->s_max_batch_time = old_opts.s_max_batch_time;
4778 4779
#ifdef CONFIG_QUOTA
	sbi->s_jquota_fmt = old_opts.s_jquota_fmt;
J
Jan Kara 已提交
4780
	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
4781
		kfree(sbi->s_qf_names[i]);
4782 4783 4784
		sbi->s_qf_names[i] = old_opts.s_qf_names[i];
	}
#endif
4785
	kfree(orig_data);
4786 4787 4788
	return err;
}

4789
static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf)
4790 4791
{
	struct super_block *sb = dentry->d_sb;
4792 4793
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_super_block *es = sbi->s_es;
L
Lukas Czerner 已提交
4794
	ext4_fsblk_t overhead = 0, resv_blocks;
P
Pekka Enberg 已提交
4795
	u64 fsid;
4796
	s64 bfree;
L
Lukas Czerner 已提交
4797
	resv_blocks = EXT4_C2B(sbi, atomic64_read(&sbi->s_resv_clusters));
4798

4799 4800
	if (!test_opt(sb, MINIX_DF))
		overhead = sbi->s_overhead;
4801

4802
	buf->f_type = EXT4_SUPER_MAGIC;
4803
	buf->f_bsize = sb->s_blocksize;
4804
	buf->f_blocks = ext4_blocks_count(es) - EXT4_C2B(sbi, overhead);
4805 4806
	bfree = percpu_counter_sum_positive(&sbi->s_freeclusters_counter) -
		percpu_counter_sum_positive(&sbi->s_dirtyclusters_counter);
4807
	/* prevent underflow in case that few free space is available */
4808
	buf->f_bfree = EXT4_C2B(sbi, max_t(s64, bfree, 0));
L
Lukas Czerner 已提交
4809 4810 4811
	buf->f_bavail = buf->f_bfree -
			(ext4_r_blocks_count(es) + resv_blocks);
	if (buf->f_bfree < (ext4_r_blocks_count(es) + resv_blocks))
4812 4813
		buf->f_bavail = 0;
	buf->f_files = le32_to_cpu(es->s_inodes_count);
P
Peter Zijlstra 已提交
4814
	buf->f_ffree = percpu_counter_sum_positive(&sbi->s_freeinodes_counter);
4815
	buf->f_namelen = EXT4_NAME_LEN;
P
Pekka Enberg 已提交
4816 4817 4818 4819
	fsid = le64_to_cpup((void *)es->s_uuid) ^
	       le64_to_cpup((void *)es->s_uuid + sizeof(u64));
	buf->f_fsid.val[0] = fsid & 0xFFFFFFFFUL;
	buf->f_fsid.val[1] = (fsid >> 32) & 0xFFFFFFFFUL;
4820

4821 4822 4823
	return 0;
}

4824 4825
/* Helper function for writing quotas on sync - we need to start transaction
 * before quota file is locked for write. Otherwise the are possible deadlocks:
4826
 * Process 1                         Process 2
4827
 * ext4_create()                     quota_sync()
4828
 *   jbd2_journal_start()                  write_dquot()
4829
 *   dquot_initialize()                         down(dqio_mutex)
4830
 *     down(dqio_mutex)                    jbd2_journal_start()
4831 4832 4833 4834 4835 4836 4837
 *
 */

#ifdef CONFIG_QUOTA

static inline struct inode *dquot_to_inode(struct dquot *dquot)
{
4838
	return sb_dqopt(dquot->dq_sb)->files[dquot->dq_id.type];
4839 4840
}

4841
static int ext4_write_dquot(struct dquot *dquot)
4842 4843 4844 4845 4846 4847
{
	int ret, err;
	handle_t *handle;
	struct inode *inode;

	inode = dquot_to_inode(dquot);
4848
	handle = ext4_journal_start(inode, EXT4_HT_QUOTA,
4849
				    EXT4_QUOTA_TRANS_BLOCKS(dquot->dq_sb));
4850 4851 4852
	if (IS_ERR(handle))
		return PTR_ERR(handle);
	ret = dquot_commit(dquot);
4853
	err = ext4_journal_stop(handle);
4854 4855 4856 4857 4858
	if (!ret)
		ret = err;
	return ret;
}

4859
static int ext4_acquire_dquot(struct dquot *dquot)
4860 4861 4862 4863
{
	int ret, err;
	handle_t *handle;

4864
	handle = ext4_journal_start(dquot_to_inode(dquot), EXT4_HT_QUOTA,
4865
				    EXT4_QUOTA_INIT_BLOCKS(dquot->dq_sb));
4866 4867 4868
	if (IS_ERR(handle))
		return PTR_ERR(handle);
	ret = dquot_acquire(dquot);
4869
	err = ext4_journal_stop(handle);
4870 4871 4872 4873 4874
	if (!ret)
		ret = err;
	return ret;
}

4875
static int ext4_release_dquot(struct dquot *dquot)
4876 4877 4878 4879
{
	int ret, err;
	handle_t *handle;

4880
	handle = ext4_journal_start(dquot_to_inode(dquot), EXT4_HT_QUOTA,
4881
				    EXT4_QUOTA_DEL_BLOCKS(dquot->dq_sb));
J
Jan Kara 已提交
4882 4883 4884
	if (IS_ERR(handle)) {
		/* Release dquot anyway to avoid endless cycle in dqput() */
		dquot_release(dquot);
4885
		return PTR_ERR(handle);
J
Jan Kara 已提交
4886
	}
4887
	ret = dquot_release(dquot);
4888
	err = ext4_journal_stop(handle);
4889 4890 4891 4892 4893
	if (!ret)
		ret = err;
	return ret;
}

4894
static int ext4_mark_dquot_dirty(struct dquot *dquot)
4895
{
4896 4897 4898
	struct super_block *sb = dquot->dq_sb;
	struct ext4_sb_info *sbi = EXT4_SB(sb);

4899
	/* Are we journaling quotas? */
4900
	if (ext4_has_feature_quota(sb) ||
4901
	    sbi->s_qf_names[USRQUOTA] || sbi->s_qf_names[GRPQUOTA]) {
4902
		dquot_mark_dquot_dirty(dquot);
4903
		return ext4_write_dquot(dquot);
4904 4905 4906 4907 4908
	} else {
		return dquot_mark_dquot_dirty(dquot);
	}
}

4909
static int ext4_write_info(struct super_block *sb, int type)
4910 4911 4912 4913 4914
{
	int ret, err;
	handle_t *handle;

	/* Data block + inode block */
4915
	handle = ext4_journal_start(d_inode(sb->s_root), EXT4_HT_QUOTA, 2);
4916 4917 4918
	if (IS_ERR(handle))
		return PTR_ERR(handle);
	ret = dquot_commit_info(sb, type);
4919
	err = ext4_journal_stop(handle);
4920 4921 4922 4923 4924 4925 4926 4927 4928
	if (!ret)
		ret = err;
	return ret;
}

/*
 * Turn on quotas during mount time - we need to find
 * the quota file and such...
 */
4929
static int ext4_quota_on_mount(struct super_block *sb, int type)
4930
{
4931 4932
	return dquot_quota_on_mount(sb, EXT4_SB(sb)->s_qf_names[type],
					EXT4_SB(sb)->s_jquota_fmt, type);
4933 4934 4935 4936 4937
}

/*
 * Standard function to be called on quota_on
 */
4938
static int ext4_quota_on(struct super_block *sb, int type, int format_id,
4939
			 struct path *path)
4940 4941 4942 4943 4944
{
	int err;

	if (!test_opt(sb, QUOTA))
		return -EINVAL;
4945

4946
	/* Quotafile not on the same filesystem? */
4947
	if (path->dentry->d_sb != sb)
4948
		return -EXDEV;
4949 4950
	/* Journaling quota? */
	if (EXT4_SB(sb)->s_qf_names[type]) {
4951
		/* Quotafile not in fs root? */
4952
		if (path->dentry->d_parent != sb->s_root)
4953 4954 4955
			ext4_msg(sb, KERN_WARNING,
				"Quota file not on filesystem root. "
				"Journaled quota will not work");
4956
	}
4957 4958 4959 4960 4961

	/*
	 * When we journal data on quota file, we have to flush journal to see
	 * all updates to the file when we bypass pagecache...
	 */
4962
	if (EXT4_SB(sb)->s_journal &&
4963
	    ext4_should_journal_data(d_inode(path->dentry))) {
4964 4965 4966 4967 4968
		/*
		 * We don't need to lock updates but journal_flush() could
		 * otherwise be livelocked...
		 */
		jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal);
4969
		err = jbd2_journal_flush(EXT4_SB(sb)->s_journal);
4970
		jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal);
4971
		if (err)
4972
			return err;
4973 4974
	}

4975
	return dquot_quota_on(sb, type, format_id, path);
4976 4977
}

4978 4979 4980 4981 4982
static int ext4_quota_enable(struct super_block *sb, int type, int format_id,
			     unsigned int flags)
{
	int err;
	struct inode *qf_inode;
J
Jan Kara 已提交
4983
	unsigned long qf_inums[EXT4_MAXQUOTAS] = {
4984 4985 4986 4987
		le32_to_cpu(EXT4_SB(sb)->s_es->s_usr_quota_inum),
		le32_to_cpu(EXT4_SB(sb)->s_es->s_grp_quota_inum)
	};

4988
	BUG_ON(!ext4_has_feature_quota(sb));
4989 4990 4991 4992 4993 4994 4995 4996 4997 4998

	if (!qf_inums[type])
		return -EPERM;

	qf_inode = ext4_iget(sb, qf_inums[type]);
	if (IS_ERR(qf_inode)) {
		ext4_error(sb, "Bad quota inode # %lu", qf_inums[type]);
		return PTR_ERR(qf_inode);
	}

J
Jan Kara 已提交
4999 5000
	/* Don't account quota for quota files to avoid recursion */
	qf_inode->i_flags |= S_NOQUOTA;
5001 5002 5003 5004 5005 5006 5007 5008 5009 5010
	err = dquot_enable(qf_inode, type, format_id, flags);
	iput(qf_inode);

	return err;
}

/* Enable usage tracking for all quota types. */
static int ext4_enable_quotas(struct super_block *sb)
{
	int type, err = 0;
J
Jan Kara 已提交
5011
	unsigned long qf_inums[EXT4_MAXQUOTAS] = {
5012 5013 5014 5015 5016
		le32_to_cpu(EXT4_SB(sb)->s_es->s_usr_quota_inum),
		le32_to_cpu(EXT4_SB(sb)->s_es->s_grp_quota_inum)
	};

	sb_dqopt(sb)->flags |= DQUOT_QUOTA_SYS_FILE;
J
Jan Kara 已提交
5017
	for (type = 0; type < EXT4_MAXQUOTAS; type++) {
5018 5019 5020 5021 5022
		if (qf_inums[type]) {
			err = ext4_quota_enable(sb, type, QFMT_VFS_V1,
						DQUOT_USAGE_ENABLED);
			if (err) {
				ext4_warning(sb,
5023 5024 5025
					"Failed to enable quota tracking "
					"(type=%d, err=%d). Please run "
					"e2fsck to fix.", type, err);
5026 5027 5028 5029 5030 5031 5032
				return err;
			}
		}
	}
	return 0;
}

5033 5034
static int ext4_quota_off(struct super_block *sb, int type)
{
5035 5036 5037
	struct inode *inode = sb_dqopt(sb)->files[type];
	handle_t *handle;

5038 5039 5040
	/* Force all delayed allocation blocks to be allocated.
	 * Caller already holds s_umount sem */
	if (test_opt(sb, DELALLOC))
5041 5042
		sync_filesystem(sb);

5043 5044 5045
	if (!inode)
		goto out;

5046 5047
	/* Update modification times of quota files when userspace can
	 * start looking at them */
5048
	handle = ext4_journal_start(inode, EXT4_HT_QUOTA, 1);
5049 5050 5051 5052 5053 5054 5055
	if (IS_ERR(handle))
		goto out;
	inode->i_mtime = inode->i_ctime = CURRENT_TIME;
	ext4_mark_inode_dirty(handle, inode);
	ext4_journal_stop(handle);

out:
5056 5057 5058
	return dquot_quota_off(sb, type);
}

5059 5060
/* Read data from quotafile - avoid pagecache and such because we cannot afford
 * acquiring the locks... As quota files are never truncated and quota code
L
Lucas De Marchi 已提交
5061
 * itself serializes the operations (and no one else should touch the files)
5062
 * we don't have to be afraid of races */
5063
static ssize_t ext4_quota_read(struct super_block *sb, int type, char *data,
5064 5065 5066
			       size_t len, loff_t off)
{
	struct inode *inode = sb_dqopt(sb)->files[type];
A
Aneesh Kumar K.V 已提交
5067
	ext4_lblk_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
5068 5069 5070 5071 5072 5073 5074 5075 5076 5077 5078 5079 5080 5081
	int offset = off & (sb->s_blocksize - 1);
	int tocopy;
	size_t toread;
	struct buffer_head *bh;
	loff_t i_size = i_size_read(inode);

	if (off > i_size)
		return 0;
	if (off+len > i_size)
		len = i_size-off;
	toread = len;
	while (toread > 0) {
		tocopy = sb->s_blocksize - offset < toread ?
				sb->s_blocksize - offset : toread;
5082 5083 5084
		bh = ext4_bread(NULL, inode, blk, 0);
		if (IS_ERR(bh))
			return PTR_ERR(bh);
5085 5086 5087 5088 5089 5090 5091 5092 5093 5094 5095 5096 5097 5098 5099
		if (!bh)	/* A hole? */
			memset(data, 0, tocopy);
		else
			memcpy(data, bh->b_data+offset, tocopy);
		brelse(bh);
		offset = 0;
		toread -= tocopy;
		data += tocopy;
		blk++;
	}
	return len;
}

/* Write to quotafile (we know the transaction is already started and has
 * enough credits) */
5100
static ssize_t ext4_quota_write(struct super_block *sb, int type,
5101 5102 5103
				const char *data, size_t len, loff_t off)
{
	struct inode *inode = sb_dqopt(sb)->files[type];
A
Aneesh Kumar K.V 已提交
5104
	ext4_lblk_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
5105
	int err, offset = off & (sb->s_blocksize - 1);
5106
	int retries = 0;
5107 5108 5109
	struct buffer_head *bh;
	handle_t *handle = journal_current_handle();

5110
	if (EXT4_SB(sb)->s_journal && !handle) {
5111 5112
		ext4_msg(sb, KERN_WARNING, "Quota write (off=%llu, len=%llu)"
			" cancelled because transaction is not started",
J
Jan Kara 已提交
5113 5114 5115
			(unsigned long long)off, (unsigned long long)len);
		return -EIO;
	}
5116 5117 5118 5119 5120 5121 5122 5123 5124 5125 5126
	/*
	 * Since we account only one data block in transaction credits,
	 * then it is impossible to cross a block boundary.
	 */
	if (sb->s_blocksize - offset < len) {
		ext4_msg(sb, KERN_WARNING, "Quota write (off=%llu, len=%llu)"
			" cancelled because not block aligned",
			(unsigned long long)off, (unsigned long long)len);
		return -EIO;
	}

5127 5128 5129 5130 5131 5132
	do {
		bh = ext4_bread(handle, inode, blk,
				EXT4_GET_BLOCKS_CREATE |
				EXT4_GET_BLOCKS_METADATA_NOFAIL);
	} while (IS_ERR(bh) && (PTR_ERR(bh) == -ENOSPC) &&
		 ext4_should_retry_alloc(inode->i_sb, &retries));
5133 5134
	if (IS_ERR(bh))
		return PTR_ERR(bh);
5135 5136
	if (!bh)
		goto out;
5137
	BUFFER_TRACE(bh, "get write access");
5138 5139 5140
	err = ext4_journal_get_write_access(handle, bh);
	if (err) {
		brelse(bh);
5141
		return err;
5142
	}
5143 5144 5145 5146
	lock_buffer(bh);
	memcpy(bh->b_data+offset, data, len);
	flush_dcache_page(bh->b_page);
	unlock_buffer(bh);
5147
	err = ext4_handle_dirty_metadata(handle, NULL, bh);
5148
	brelse(bh);
5149
out:
5150 5151
	if (inode->i_size < off + len) {
		i_size_write(inode, off + len);
5152
		EXT4_I(inode)->i_disksize = inode->i_size;
5153
		ext4_mark_inode_dirty(handle, inode);
5154
	}
5155
	return len;
5156 5157 5158 5159
}

#endif

A
Al Viro 已提交
5160 5161
static struct dentry *ext4_mount(struct file_system_type *fs_type, int flags,
		       const char *dev_name, void *data)
5162
{
A
Al Viro 已提交
5163
	return mount_bdev(fs_type, flags, dev_name, data, ext4_fill_super);
5164 5165
}

J
Jan Kara 已提交
5166
#if !defined(CONFIG_EXT2_FS) && !defined(CONFIG_EXT2_FS_MODULE) && defined(CONFIG_EXT4_USE_FOR_EXT2)
5167 5168 5169 5170 5171 5172 5173 5174 5175 5176 5177 5178
static inline void register_as_ext2(void)
{
	int err = register_filesystem(&ext2_fs_type);
	if (err)
		printk(KERN_WARNING
		       "EXT4-fs: Unable to register as ext2 (%d)\n", err);
}

static inline void unregister_as_ext2(void)
{
	unregister_filesystem(&ext2_fs_type);
}
5179 5180 5181

static inline int ext2_feature_set_ok(struct super_block *sb)
{
5182
	if (ext4_has_unknown_ext2_incompat_features(sb))
5183 5184 5185
		return 0;
	if (sb->s_flags & MS_RDONLY)
		return 1;
5186
	if (ext4_has_unknown_ext2_ro_compat_features(sb))
5187 5188 5189
		return 0;
	return 1;
}
5190 5191 5192
#else
static inline void register_as_ext2(void) { }
static inline void unregister_as_ext2(void) { }
5193
static inline int ext2_feature_set_ok(struct super_block *sb) { return 0; }
5194 5195 5196 5197 5198 5199 5200 5201 5202 5203 5204 5205 5206 5207
#endif

static inline void register_as_ext3(void)
{
	int err = register_filesystem(&ext3_fs_type);
	if (err)
		printk(KERN_WARNING
		       "EXT4-fs: Unable to register as ext3 (%d)\n", err);
}

static inline void unregister_as_ext3(void)
{
	unregister_filesystem(&ext3_fs_type);
}
5208 5209 5210

static inline int ext3_feature_set_ok(struct super_block *sb)
{
5211
	if (ext4_has_unknown_ext3_incompat_features(sb))
5212
		return 0;
5213
	if (!ext4_has_feature_journal(sb))
5214 5215 5216
		return 0;
	if (sb->s_flags & MS_RDONLY)
		return 1;
5217
	if (ext4_has_unknown_ext3_ro_compat_features(sb))
5218 5219 5220
		return 0;
	return 1;
}
5221

T
Theodore Ts'o 已提交
5222 5223 5224
static struct file_system_type ext4_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "ext4",
A
Al Viro 已提交
5225
	.mount		= ext4_mount,
T
Theodore Ts'o 已提交
5226 5227 5228
	.kill_sb	= kill_block_super,
	.fs_flags	= FS_REQUIRES_DEV,
};
5229
MODULE_ALIAS_FS("ext4");
T
Theodore Ts'o 已提交
5230

5231 5232 5233 5234
/* Shared across all ext4 file systems */
wait_queue_head_t ext4__ioend_wq[EXT4_WQ_HASH_SZ];
struct mutex ext4__aio_mutex[EXT4_WQ_HASH_SZ];

5235
static int __init ext4_init_fs(void)
5236
{
5237
	int i, err;
5238

5239
	ratelimit_state_init(&ext4_mount_msg_ratelimit, 30 * HZ, 64);
5240 5241 5242
	ext4_li_info = NULL;
	mutex_init(&ext4_li_mtx);

5243
	/* Build-time check for flags consistency */
5244
	ext4_check_flag_values();
5245 5246 5247 5248 5249 5250

	for (i = 0; i < EXT4_WQ_HASH_SZ; i++) {
		mutex_init(&ext4__aio_mutex[i]);
		init_waitqueue_head(&ext4__ioend_wq[i]);
	}

5251
	err = ext4_init_es();
5252 5253
	if (err)
		return err;
5254 5255 5256

	err = ext4_init_pageio();
	if (err)
5257
		goto out5;
5258

5259
	err = ext4_init_system_zone();
5260
	if (err)
5261
		goto out4;
5262

5263
	err = ext4_init_sysfs();
T
Theodore Ts'o 已提交
5264
	if (err)
5265
		goto out3;
5266

5267
	err = ext4_init_mballoc();
5268 5269
	if (err)
		goto out2;
5270 5271
	else
		ext4_mballoc_ready = 1;
5272 5273 5274
	err = init_inodecache();
	if (err)
		goto out1;
5275
	register_as_ext3();
5276
	register_as_ext2();
T
Theodore Ts'o 已提交
5277
	err = register_filesystem(&ext4_fs_type);
5278 5279
	if (err)
		goto out;
5280

5281 5282
	return 0;
out:
5283 5284
	unregister_as_ext2();
	unregister_as_ext3();
5285 5286
	destroy_inodecache();
out1:
5287
	ext4_mballoc_ready = 0;
5288
	ext4_exit_mballoc();
5289
out2:
5290 5291
	ext4_exit_sysfs();
out3:
5292
	ext4_exit_system_zone();
5293
out4:
5294
	ext4_exit_pageio();
5295
out5:
5296 5297
	ext4_exit_es();

5298 5299 5300
	return err;
}

5301
static void __exit ext4_exit_fs(void)
5302
{
5303
	ext4_exit_crypto();
5304
	ext4_destroy_lazyinit_thread();
5305 5306
	unregister_as_ext2();
	unregister_as_ext3();
T
Theodore Ts'o 已提交
5307
	unregister_filesystem(&ext4_fs_type);
5308
	destroy_inodecache();
5309
	ext4_exit_mballoc();
5310
	ext4_exit_sysfs();
5311 5312
	ext4_exit_system_zone();
	ext4_exit_pageio();
5313
	ext4_exit_es();
5314 5315 5316
}

MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others");
5317
MODULE_DESCRIPTION("Fourth Extended Filesystem");
5318
MODULE_LICENSE("GPL");
5319 5320
module_init(ext4_init_fs)
module_exit(ext4_exit_fs)