super.c 161.8 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>
37
#include <linux/proc_fs.h>
T
Theodore Ts'o 已提交
38
#include <linux/ctype.h>
V
Vignesh Babu 已提交
39
#include <linux/log2.h>
A
Andreas Dilger 已提交
40
#include <linux/crc16.h>
D
Dan Magenheimer 已提交
41
#include <linux/cleancache.h>
42 43
#include <asm/uaccess.h>

44 45 46
#include <linux/kthread.h>
#include <linux/freezer.h>

47
#include "ext4.h"
48
#include "ext4_extents.h"	/* Needed for trace points definition */
49
#include "ext4_jbd2.h"
50 51
#include "xattr.h"
#include "acl.h"
52
#include "mballoc.h"
53

54 55 56
#define CREATE_TRACE_POINTS
#include <trace/events/ext4.h>

57
static struct proc_dir_entry *ext4_proc_root;
T
Theodore Ts'o 已提交
58
static struct kset *ext4_kset;
59 60 61
static struct ext4_lazy_init *ext4_li_info;
static struct mutex ext4_li_mtx;
static struct ext4_features *ext4_feat;
62
static int ext4_mballoc_ready;
63
static struct ratelimit_state ext4_mount_msg_ratelimit;
64

65
static int ext4_load_journal(struct super_block *, struct ext4_super_block *,
66
			     unsigned long journal_devnum);
67
static int ext4_show_options(struct seq_file *seq, struct dentry *root);
68
static int ext4_commit_super(struct super_block *sb, int sync);
69 70 71 72
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);
73
static int ext4_sync_fs(struct super_block *sb, int wait);
74 75
static int ext4_remount(struct super_block *sb, int *flags, char *data);
static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf);
76 77
static int ext4_unfreeze(struct super_block *sb);
static int ext4_freeze(struct super_block *sb);
A
Al Viro 已提交
78 79
static struct dentry *ext4_mount(struct file_system_type *fs_type, int flags,
		       const char *dev_name, void *data);
80 81
static inline int ext2_feature_set_ok(struct super_block *sb);
static inline int ext3_feature_set_ok(struct super_block *sb);
82
static int ext4_feature_set_ok(struct super_block *sb, int readonly);
83 84
static void ext4_destroy_lazyinit_thread(void);
static void ext4_unregister_li_request(struct super_block *sb);
85
static void ext4_clear_request_list(void);
L
Lukas Czerner 已提交
86
static int ext4_reserve_clusters(struct ext4_sb_info *, ext4_fsblk_t);
87

J
Jan Kara 已提交
88
#if !defined(CONFIG_EXT2_FS) && !defined(CONFIG_EXT2_FS_MODULE) && defined(CONFIG_EXT4_USE_FOR_EXT2)
89 90 91 92 93 94 95
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,
};
96
MODULE_ALIAS_FS("ext2");
97
MODULE_ALIAS("ext2");
98 99 100 101 102 103
#define IS_EXT2_SB(sb) ((sb)->s_bdev->bd_holder == &ext2_fs_type)
#else
#define IS_EXT2_SB(sb) (0)
#endif


104 105 106
static struct file_system_type ext3_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "ext3",
A
Al Viro 已提交
107
	.mount		= ext4_mount,
108 109 110
	.kill_sb	= kill_block_super,
	.fs_flags	= FS_REQUIRES_DEV,
};
111
MODULE_ALIAS_FS("ext3");
112
MODULE_ALIAS("ext3");
113
#define IS_EXT3_SB(sb) ((sb)->s_bdev->bd_holder == &ext3_fs_type)
L
Laurent Vivier 已提交
114

115 116 117 118 119 120 121 122 123 124
static int ext4_verify_csum_type(struct super_block *sb,
				 struct ext4_super_block *es)
{
	if (!EXT4_HAS_RO_COMPAT_FEATURE(sb,
					EXT4_FEATURE_RO_COMPAT_METADATA_CSUM))
		return 1;

	return es->s_checksum_type == EXT4_CRC32C_CHKSUM;
}

125 126 127 128 129 130 131 132 133 134 135 136
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);
}

137 138
static int ext4_superblock_csum_verify(struct super_block *sb,
				       struct ext4_super_block *es)
139
{
140
	if (!ext4_has_metadata_csum(sb))
141 142 143 144 145
		return 1;

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

146
void ext4_superblock_csum_set(struct super_block *sb)
147
{
148 149
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;

150
	if (!ext4_has_metadata_csum(sb))
151 152 153 154 155
		return;

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

156 157 158 159
void *ext4_kvmalloc(size_t size, gfp_t flags)
{
	void *ret;

160
	ret = kmalloc(size, flags | __GFP_NOWARN);
161 162 163 164 165 166 167 168 169
	if (!ret)
		ret = __vmalloc(size, flags, PAGE_KERNEL);
	return ret;
}

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

170
	ret = kzalloc(size, flags | __GFP_NOWARN);
171 172 173 174 175
	if (!ret)
		ret = __vmalloc(size, flags | __GFP_ZERO, PAGE_KERNEL);
	return ret;
}

176 177
ext4_fsblk_t ext4_block_bitmap(struct super_block *sb,
			       struct ext4_group_desc *bg)
L
Laurent Vivier 已提交
178
{
179
	return le32_to_cpu(bg->bg_block_bitmap_lo) |
180
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
181
		 (ext4_fsblk_t)le32_to_cpu(bg->bg_block_bitmap_hi) << 32 : 0);
L
Laurent Vivier 已提交
182 183
}

184 185
ext4_fsblk_t ext4_inode_bitmap(struct super_block *sb,
			       struct ext4_group_desc *bg)
L
Laurent Vivier 已提交
186
{
187
	return le32_to_cpu(bg->bg_inode_bitmap_lo) |
188
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
189
		 (ext4_fsblk_t)le32_to_cpu(bg->bg_inode_bitmap_hi) << 32 : 0);
L
Laurent Vivier 已提交
190 191
}

192 193
ext4_fsblk_t ext4_inode_table(struct super_block *sb,
			      struct ext4_group_desc *bg)
L
Laurent Vivier 已提交
194
{
195
	return le32_to_cpu(bg->bg_inode_table_lo) |
196
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
197
		 (ext4_fsblk_t)le32_to_cpu(bg->bg_inode_table_hi) << 32 : 0);
L
Laurent Vivier 已提交
198 199
}

200 201
__u32 ext4_free_group_clusters(struct super_block *sb,
			       struct ext4_group_desc *bg)
202 203 204
{
	return le16_to_cpu(bg->bg_free_blocks_count_lo) |
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
205
		 (__u32)le16_to_cpu(bg->bg_free_blocks_count_hi) << 16 : 0);
206 207 208 209 210 211 212
}

__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 ?
213
		 (__u32)le16_to_cpu(bg->bg_free_inodes_count_hi) << 16 : 0);
214 215 216 217 218 219 220
}

__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 ?
221
		 (__u32)le16_to_cpu(bg->bg_used_dirs_count_hi) << 16 : 0);
222 223 224 225 226 227 228
}

__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 ?
229
		 (__u32)le16_to_cpu(bg->bg_itable_unused_hi) << 16 : 0);
230 231
}

232 233
void ext4_block_bitmap_set(struct super_block *sb,
			   struct ext4_group_desc *bg, ext4_fsblk_t blk)
L
Laurent Vivier 已提交
234
{
235
	bg->bg_block_bitmap_lo = cpu_to_le32((u32)blk);
236 237
	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
238 239
}

240 241
void ext4_inode_bitmap_set(struct super_block *sb,
			   struct ext4_group_desc *bg, ext4_fsblk_t blk)
L
Laurent Vivier 已提交
242
{
243
	bg->bg_inode_bitmap_lo  = cpu_to_le32((u32)blk);
244 245
	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
246 247
}

248 249
void ext4_inode_table_set(struct super_block *sb,
			  struct ext4_group_desc *bg, ext4_fsblk_t blk)
L
Laurent Vivier 已提交
250
{
251
	bg->bg_inode_table_lo = cpu_to_le32((u32)blk);
252 253
	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
254 255
}

256 257
void ext4_free_group_clusters_set(struct super_block *sb,
				  struct ext4_group_desc *bg, __u32 count)
258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287
{
	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);
}

288

289 290 291 292 293 294
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;
295 296
	if (bdev_read_only(sb->s_bdev))
		return;
297 298 299 300 301 302 303 304 305 306 307 308
	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;
	}
309 310 311 312 313 314
	/*
	 * 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);
315
	le32_add_cpu(&es->s_error_count, 1);
316 317 318 319 320 321 322 323 324
}

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);
}

325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340
/*
 * 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
341 342 343 344 345
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);
346
	struct ext4_journal_cb_entry	*jce;
B
Bobi Jam 已提交
347

348
	BUG_ON(txn->t_state == T_FINISHED);
B
Bobi Jam 已提交
349
	spin_lock(&sbi->s_md_lock);
350 351 352
	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
353 354 355 356 357 358 359
		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);
}
360

361 362 363 364
/* 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
365
 * superblock.  That is not possible on ext4, because we may have other
366 367 368 369 370
 * 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.
 *
371
 * We'll just use the jbd2_journal_abort() error code to record an error in
372
 * the journal instead.  On recovery, the journal will complain about
373 374 375
 * that error until we've noted it down and cleared it.
 */

376
static void ext4_handle_error(struct super_block *sb)
377 378 379 380
{
	if (sb->s_flags & MS_RDONLY)
		return;

381
	if (!test_opt(sb, ERRORS_CONT)) {
382
		journal_t *journal = EXT4_SB(sb)->s_journal;
383

384
		EXT4_SB(sb)->s_mount_flags |= EXT4_MF_FS_ABORTED;
385
		if (journal)
386
			jbd2_journal_abort(journal, -EIO);
387
	}
388
	if (test_opt(sb, ERRORS_RO)) {
389
		ext4_msg(sb, KERN_CRIT, "Remounting filesystem read-only");
390 391 392 393 394
		/*
		 * Make sure updated value of ->s_mount_flags will be visible
		 * before ->s_flags update
		 */
		smp_wmb();
395 396 397
		sb->s_flags |= MS_RDONLY;
	}
	if (test_opt(sb, ERRORS_PANIC))
398
		panic("EXT4-fs (device %s): panic forced after error\n",
399 400 401
			sb->s_id);
}

402 403 404 405
#define ext4_error_ratelimit(sb)					\
		___ratelimit(&(EXT4_SB(sb)->s_err_ratelimit_state),	\
			     "EXT4-fs error")

406
void __ext4_error(struct super_block *sb, const char *function,
407
		  unsigned int line, const char *fmt, ...)
408
{
J
Joe Perches 已提交
409
	struct va_format vaf;
410 411
	va_list args;

412 413 414 415 416 417 418 419 420
	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);
	}
421
	save_error_info(sb, function, line);
422
	ext4_handle_error(sb);
423 424
}

425 426 427
void __ext4_error_inode(struct inode *inode, const char *function,
			unsigned int line, ext4_fsblk_t block,
			const char *fmt, ...)
428 429
{
	va_list args;
430
	struct va_format vaf;
431
	struct ext4_super_block *es = EXT4_SB(inode->i_sb)->s_es;
432

433 434
	es->s_last_error_ino = cpu_to_le32(inode->i_ino);
	es->s_last_error_block = cpu_to_le64(block);
435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450
	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);
	}
451
	save_error_info(inode->i_sb, function, line);
452 453 454
	ext4_handle_error(inode->i_sb);
}

455 456 457
void __ext4_error_file(struct file *file, const char *function,
		       unsigned int line, ext4_fsblk_t block,
		       const char *fmt, ...)
458 459
{
	va_list args;
460
	struct va_format vaf;
461
	struct ext4_super_block *es;
A
Al Viro 已提交
462
	struct inode *inode = file_inode(file);
463 464
	char pathname[80], *path;

465 466
	es = EXT4_SB(inode->i_sb)->s_es;
	es->s_last_error_ino = cpu_to_le32(inode->i_ino);
467
	if (ext4_error_ratelimit(inode->i_sb)) {
M
Miklos Szeredi 已提交
468
		path = file_path(file, pathname, sizeof(pathname));
469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487
		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);
	}
488
	save_error_info(inode->i_sb, function, line);
489 490 491
	ext4_handle_error(inode->i_sb);
}

492 493
const char *ext4_decode_error(struct super_block *sb, int errno,
			      char nbuf[16])
494 495 496 497 498 499 500 501 502 503 504
{
	char *errstr = NULL;

	switch (errno) {
	case -EIO:
		errstr = "IO failure";
		break;
	case -ENOMEM:
		errstr = "Out of memory";
		break;
	case -EROFS:
505 506
		if (!sb || (EXT4_SB(sb)->s_journal &&
			    EXT4_SB(sb)->s_journal->j_flags & JBD2_ABORT))
507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525
			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;
}

526
/* __ext4_std_error decodes expected errors from journaling functions
527 528
 * automatically and invokes the appropriate error response.  */

529 530
void __ext4_std_error(struct super_block *sb, const char *function,
		      unsigned int line, int errno)
531 532 533 534 535 536 537 538 539 540 541
{
	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;

542 543 544 545 546
	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);
	}
547

548
	save_error_info(sb, function, line);
549
	ext4_handle_error(sb);
550 551 552
}

/*
553
 * ext4_abort is a much stronger failure handler than ext4_error.  The
554 555 556 557 558 559 560 561
 * 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.
 */

562
void __ext4_abort(struct super_block *sb, const char *function,
563
		unsigned int line, const char *fmt, ...)
564 565 566
{
	va_list args;

567
	save_error_info(sb, function, line);
568
	va_start(args, fmt);
569 570
	printk(KERN_CRIT "EXT4-fs error (device %s): %s:%d: ", sb->s_id,
	       function, line);
571 572 573 574
	vprintk(fmt, args);
	printk("\n");
	va_end(args);

575 576 577
	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;
578 579 580 581 582 583
		/*
		 * Make sure updated value of ->s_mount_flags will be visible
		 * before ->s_flags update
		 */
		smp_wmb();
		sb->s_flags |= MS_RDONLY;
584 585 586 587
		if (EXT4_SB(sb)->s_journal)
			jbd2_journal_abort(EXT4_SB(sb)->s_journal, -EIO);
		save_error_info(sb, function, line);
	}
588
	if (test_opt(sb, ERRORS_PANIC))
589
		panic("EXT4-fs panic from previous error\n");
590 591
}

592 593
void __ext4_msg(struct super_block *sb,
		const char *prefix, const char *fmt, ...)
594
{
J
Joe Perches 已提交
595
	struct va_format vaf;
596 597
	va_list args;

598 599 600
	if (!___ratelimit(&(EXT4_SB(sb)->s_msg_ratelimit_state), "EXT4-fs"))
		return;

601
	va_start(args, fmt);
J
Joe Perches 已提交
602 603 604
	vaf.fmt = fmt;
	vaf.va = &args;
	printk("%sEXT4-fs (%s): %pV\n", prefix, sb->s_id, &vaf);
605 606 607
	va_end(args);
}

608 609 610 611
#define ext4_warning_ratelimit(sb)					\
		___ratelimit(&(EXT4_SB(sb)->s_warning_ratelimit_state),	\
			     "EXT4-fs warning")

612
void __ext4_warning(struct super_block *sb, const char *function,
613
		    unsigned int line, const char *fmt, ...)
614
{
J
Joe Perches 已提交
615
	struct va_format vaf;
616 617
	va_list args;

618
	if (!ext4_warning_ratelimit(sb))
619 620
		return;

621
	va_start(args, fmt);
J
Joe Perches 已提交
622 623 624 625
	vaf.fmt = fmt;
	vaf.va = &args;
	printk(KERN_WARNING "EXT4-fs warning (device %s): %s:%d: %pV\n",
	       sb->s_id, function, line, &vaf);
626 627 628
	va_end(args);
}

629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646
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);
}

647 648 649 650
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, ...)
651 652 653
__releases(bitlock)
__acquires(bitlock)
{
J
Joe Perches 已提交
654
	struct va_format vaf;
655 656 657
	va_list args;
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;

658 659 660
	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
661

662 663 664 665 666 667 668 669 670 671 672 673 674 675
	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);
	}
676 677

	if (test_opt(sb, ERRORS_CONT)) {
678
		ext4_commit_super(sb, 0);
679 680
		return;
	}
681

682 683 684 685 686 687 688 689
	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
690
	 * ext4_grp_locked_error() to distinguish between the
691 692 693 694 695 696 697 698
	 * 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;
}

699
void ext4_update_dynamic_rev(struct super_block *sb)
700
{
701
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;
702

703
	if (le32_to_cpu(es->s_rev_level) > EXT4_GOOD_OLD_REV)
704 705
		return;

706
	ext4_warning(sb,
707 708
		     "updating to rev %d because of new feature flag, "
		     "running e2fsck is recommended",
709
		     EXT4_DYNAMIC_REV);
710

711 712 713
	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);
714 715 716 717 718 719 720 721 722 723 724 725 726
	/* 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
 */
727
static struct block_device *ext4_blkdev_get(dev_t dev, struct super_block *sb)
728 729 730 731
{
	struct block_device *bdev;
	char b[BDEVNAME_SIZE];

732
	bdev = blkdev_get_by_dev(dev, FMODE_READ|FMODE_WRITE|FMODE_EXCL, sb);
733 734 735 736 737
	if (IS_ERR(bdev))
		goto fail;
	return bdev;

fail:
738
	ext4_msg(sb, KERN_ERR, "failed to open journal device %s: %ld",
739 740 741 742 743 744 745
			__bdevname(dev, b), PTR_ERR(bdev));
	return NULL;
}

/*
 * Release the journal device
 */
A
Al Viro 已提交
746
static void ext4_blkdev_put(struct block_device *bdev)
747
{
A
Al Viro 已提交
748
	blkdev_put(bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL);
749 750
}

A
Al Viro 已提交
751
static void ext4_blkdev_remove(struct ext4_sb_info *sbi)
752 753 754 755
{
	struct block_device *bdev;
	bdev = sbi->journal_bdev;
	if (bdev) {
A
Al Viro 已提交
756
		ext4_blkdev_put(bdev);
757 758 759 760 761 762
		sbi->journal_bdev = NULL;
	}
}

static inline struct inode *orphan_list_entry(struct list_head *l)
{
763
	return &list_entry(l, struct ext4_inode_info, i_orphan)->vfs_inode;
764 765
}

766
static void dump_orphan_list(struct super_block *sb, struct ext4_sb_info *sbi)
767 768 769
{
	struct list_head *l;

770 771
	ext4_msg(sb, KERN_ERR, "sb orphan head is %d",
		 le32_to_cpu(sbi->s_es->s_last_orphan));
772 773 774 775 776 777 778 779 780 781 782 783

	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));
	}
}

784
static void ext4_put_super(struct super_block *sb)
785
{
786 787
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_super_block *es = sbi->s_es;
788
	int i, err;
789

790
	ext4_unregister_li_request(sb);
791 792
	dquot_disable(sb, -1, DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED);

793 794
	flush_workqueue(sbi->rsv_conversion_wq);
	destroy_workqueue(sbi->rsv_conversion_wq);
795

796 797 798 799
	if (sbi->s_journal) {
		err = jbd2_journal_destroy(sbi->s_journal);
		sbi->s_journal = NULL;
		if (err < 0)
800
			ext4_abort(sb, "Couldn't clean up the journal");
801
	}
802

803
	ext4_es_unregister_shrinker(sbi);
804
	del_timer_sync(&sbi->s_err_report);
805 806 807 808 809
	ext4_release_system_zone(sb);
	ext4_mb_release(sb);
	ext4_ext_release(sb);
	ext4_xattr_put_super(sb);

810
	if (!(sb->s_flags & MS_RDONLY)) {
811
		EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
812 813
		es->s_state = cpu_to_le16(sbi->s_mount_state);
	}
814
	if (!(sb->s_flags & MS_RDONLY))
815 816
		ext4_commit_super(sb, 1);

817
	if (sbi->s_proc) {
818
		remove_proc_entry("options", sbi->s_proc);
819
		remove_proc_entry(sb->s_id, ext4_proc_root);
820
	}
T
Theodore Ts'o 已提交
821
	kobject_del(&sbi->s_kobj);
822 823 824

	for (i = 0; i < sbi->s_gdb_count; i++)
		brelse(sbi->s_group_desc[i]);
A
Al Viro 已提交
825 826
	kvfree(sbi->s_group_desc);
	kvfree(sbi->s_flex_groups);
827
	percpu_counter_destroy(&sbi->s_freeclusters_counter);
828 829
	percpu_counter_destroy(&sbi->s_freeinodes_counter);
	percpu_counter_destroy(&sbi->s_dirs_counter);
830
	percpu_counter_destroy(&sbi->s_dirtyclusters_counter);
831 832
	brelse(sbi->s_sbh);
#ifdef CONFIG_QUOTA
J
Jan Kara 已提交
833
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
834 835 836 837 838 839 840 841 842 843 844
		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));

845
	sync_blockdev(sb->s_bdev);
846
	invalidate_bdev(sb->s_bdev);
847 848 849 850 851 852 853
	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);
854
		invalidate_bdev(sbi->journal_bdev);
855
		ext4_blkdev_remove(sbi);
856
	}
857 858 859 860
	if (sbi->s_mb_cache) {
		ext4_xattr_destroy_cache(sbi->s_mb_cache);
		sbi->s_mb_cache = NULL;
	}
861 862
	if (sbi->s_mmp_tsk)
		kthread_stop(sbi->s_mmp_tsk);
863
	sb->s_fs_info = NULL;
T
Theodore Ts'o 已提交
864 865 866 867 868 869
	/*
	 * 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);
870 871
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
872
	kfree(sbi->s_blockgroup_lock);
873 874 875
	kfree(sbi);
}

876
static struct kmem_cache *ext4_inode_cachep;
877 878 879 880

/*
 * Called inside transaction, so use GFP_NOFS
 */
881
static struct inode *ext4_alloc_inode(struct super_block *sb)
882
{
883
	struct ext4_inode_info *ei;
884

C
Christoph Lameter 已提交
885
	ei = kmem_cache_alloc(ext4_inode_cachep, GFP_NOFS);
886 887
	if (!ei)
		return NULL;
888

889
	ei->vfs_inode.i_version = 1;
890
	spin_lock_init(&ei->i_raw_lock);
891 892
	INIT_LIST_HEAD(&ei->i_prealloc_list);
	spin_lock_init(&ei->i_prealloc_lock);
Z
Zheng Liu 已提交
893 894
	ext4_es_init_tree(&ei->i_es_tree);
	rwlock_init(&ei->i_es_lock);
895
	INIT_LIST_HEAD(&ei->i_es_list);
896
	ei->i_es_all_nr = 0;
897
	ei->i_es_shk_nr = 0;
898
	ei->i_es_shrink_lblk = 0;
899 900 901
	ei->i_reserved_data_blocks = 0;
	ei->i_reserved_meta_blocks = 0;
	ei->i_allocated_meta_blocks = 0;
902
	ei->i_da_metadata_calc_len = 0;
903
	ei->i_da_metadata_calc_last_lblock = 0;
904
	spin_lock_init(&(ei->i_block_reservation_lock));
905 906
#ifdef CONFIG_QUOTA
	ei->i_reserved_quota = 0;
J
Jan Kara 已提交
907
	memset(&ei->i_dquot, 0, sizeof(ei->i_dquot));
908
#endif
909
	ei->jinode = NULL;
910
	INIT_LIST_HEAD(&ei->i_rsv_conversion_list);
911
	spin_lock_init(&ei->i_completed_io_lock);
912 913
	ei->i_sync_tid = 0;
	ei->i_datasync_tid = 0;
914
	atomic_set(&ei->i_ioend_count, 0);
915
	atomic_set(&ei->i_unwritten, 0);
916
	INIT_WORK(&ei->i_rsv_conversion_work, ext4_end_io_rsv_work);
917
#ifdef CONFIG_EXT4_FS_ENCRYPTION
918
	ei->i_crypt_info = NULL;
919
#endif
920 921 922
	return &ei->vfs_inode;
}

923 924 925 926 927 928 929 930
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
931 932 933 934 935 936
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));
}

937
static void ext4_destroy_inode(struct inode *inode)
938
{
939
	if (!list_empty(&(EXT4_I(inode)->i_orphan))) {
940 941 942
		ext4_msg(inode->i_sb, KERN_ERR,
			 "Inode %lu (%p): orphan list check failed!",
			 inode->i_ino, EXT4_I(inode));
943 944 945 946 947
		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
948
	call_rcu(&inode->i_rcu, ext4_i_callback);
949 950
}

951
static void init_once(void *foo)
952
{
953
	struct ext4_inode_info *ei = (struct ext4_inode_info *) foo;
954

C
Christoph Lameter 已提交
955 956
	INIT_LIST_HEAD(&ei->i_orphan);
	init_rwsem(&ei->xattr_sem);
957
	init_rwsem(&ei->i_data_sem);
C
Christoph Lameter 已提交
958
	inode_init_once(&ei->vfs_inode);
959 960
}

961
static int __init init_inodecache(void)
962
{
963 964
	ext4_inode_cachep = kmem_cache_create("ext4_inode_cache",
					     sizeof(struct ext4_inode_info),
965 966
					     0, (SLAB_RECLAIM_ACCOUNT|
						SLAB_MEM_SPREAD),
967
					     init_once);
968
	if (ext4_inode_cachep == NULL)
969 970 971 972 973 974
		return -ENOMEM;
	return 0;
}

static void destroy_inodecache(void)
{
975 976 977 978 979
	/*
	 * Make sure all delayed rcu free inodes are flushed before we
	 * destroy cache.
	 */
	rcu_barrier();
980
	kmem_cache_destroy(ext4_inode_cachep);
981 982
}

A
Al Viro 已提交
983
void ext4_clear_inode(struct inode *inode)
984
{
A
Al Viro 已提交
985
	invalidate_inode_buffers(inode);
986
	clear_inode(inode);
987
	dquot_drop(inode);
988
	ext4_discard_preallocations(inode);
989
	ext4_es_remove_extent(inode, 0, EXT_MAX_BLOCKS);
990 991 992 993 994 995
	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;
	}
996 997
#ifdef CONFIG_EXT4_FS_ENCRYPTION
	if (EXT4_I(inode)->i_crypt_info)
998
		ext4_free_encryption_info(inode, EXT4_I(inode)->i_crypt_info);
999
#endif
1000 1001
}

C
Christoph Hellwig 已提交
1002
static struct inode *ext4_nfs_get_inode(struct super_block *sb,
1003
					u64 ino, u32 generation)
1004 1005 1006
{
	struct inode *inode;

1007
	if (ino < EXT4_FIRST_INO(sb) && ino != EXT4_ROOT_INO)
1008
		return ERR_PTR(-ESTALE);
1009
	if (ino > le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count))
1010 1011 1012 1013
		return ERR_PTR(-ESTALE);

	/* iget isn't really right if the inode is currently unallocated!!
	 *
1014
	 * ext4_read_inode will return a bad_inode if the inode had been
1015 1016 1017 1018 1019
	 * deleted, so we should be safe.
	 *
	 * Currently we don't know the generation for parent directory, so
	 * a generation of 0 means "accept any"
	 */
1020
	inode = ext4_iget_normal(sb, ino);
1021 1022 1023
	if (IS_ERR(inode))
		return ERR_CAST(inode);
	if (generation && inode->i_generation != generation) {
1024 1025 1026
		iput(inode);
		return ERR_PTR(-ESTALE);
	}
C
Christoph Hellwig 已提交
1027 1028 1029 1030 1031

	return inode;
}

static struct dentry *ext4_fh_to_dentry(struct super_block *sb, struct fid *fid,
1032
					int fh_len, int fh_type)
C
Christoph Hellwig 已提交
1033 1034 1035 1036 1037 1038
{
	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,
1039
					int fh_len, int fh_type)
C
Christoph Hellwig 已提交
1040 1041 1042
{
	return generic_fh_to_parent(sb, fid, fh_len, fh_type,
				    ext4_nfs_get_inode);
1043 1044
}

1045 1046 1047 1048 1049 1050
/*
 * 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.
 */
1051 1052
static int bdev_try_to_free_page(struct super_block *sb, struct page *page,
				 gfp_t wait)
1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064
{
	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,
							wait & ~__GFP_WAIT);
	return try_to_free_buffers(page);
}

1065
#ifdef CONFIG_QUOTA
1066
#define QTYPE2NAME(t) ((t) == USRQUOTA ? "user" : "group")
1067
#define QTYPE2MOPT(on, t) ((t) == USRQUOTA?((on)##USRJQUOTA):((on)##GRPJQUOTA))
1068

1069 1070 1071 1072 1073
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);
1074
static int ext4_quota_on(struct super_block *sb, int type, int format_id,
1075
			 struct path *path);
1076
static int ext4_quota_off(struct super_block *sb, int type);
1077 1078
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,
1079
			       size_t len, loff_t off);
1080
static ssize_t ext4_quota_write(struct super_block *sb, int type,
1081
				const char *data, size_t len, loff_t off);
1082 1083 1084
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);
1085

J
Jan Kara 已提交
1086 1087 1088 1089 1090
static struct dquot **ext4_get_dquots(struct inode *inode)
{
	return EXT4_I(inode)->i_dquot;
}

1091
static const struct dquot_operations ext4_quota_operations = {
1092
	.get_reserved_space = ext4_get_reserved_space,
1093 1094 1095 1096
	.write_dquot	= ext4_write_dquot,
	.acquire_dquot	= ext4_acquire_dquot,
	.release_dquot	= ext4_release_dquot,
	.mark_dirty	= ext4_mark_dquot_dirty,
1097 1098 1099
	.write_info	= ext4_write_info,
	.alloc_dquot	= dquot_alloc,
	.destroy_dquot	= dquot_destroy,
1100 1101
};

1102
static const struct quotactl_ops ext4_qctl_operations = {
1103
	.quota_on	= ext4_quota_on,
1104
	.quota_off	= ext4_quota_off,
1105
	.quota_sync	= dquot_quota_sync,
1106
	.get_state	= dquot_get_state,
1107 1108 1109
	.set_info	= dquot_set_dqinfo,
	.get_dqblk	= dquot_get_dqblk,
	.set_dqblk	= dquot_set_dqblk
1110 1111 1112
};
#endif

1113
static const struct super_operations ext4_sops = {
1114 1115 1116 1117
	.alloc_inode	= ext4_alloc_inode,
	.destroy_inode	= ext4_destroy_inode,
	.write_inode	= ext4_write_inode,
	.dirty_inode	= ext4_dirty_inode,
1118
	.drop_inode	= ext4_drop_inode,
A
Al Viro 已提交
1119
	.evict_inode	= ext4_evict_inode,
1120 1121
	.put_super	= ext4_put_super,
	.sync_fs	= ext4_sync_fs,
1122 1123
	.freeze_fs	= ext4_freeze,
	.unfreeze_fs	= ext4_unfreeze,
1124 1125 1126
	.statfs		= ext4_statfs,
	.remount_fs	= ext4_remount,
	.show_options	= ext4_show_options,
1127
#ifdef CONFIG_QUOTA
1128 1129
	.quota_read	= ext4_quota_read,
	.quota_write	= ext4_quota_write,
J
Jan Kara 已提交
1130
	.get_dquots	= ext4_get_dquots,
1131
#endif
1132
	.bdev_try_to_free_page = bdev_try_to_free_page,
1133 1134
};

1135
static const struct export_operations ext4_export_ops = {
C
Christoph Hellwig 已提交
1136 1137
	.fh_to_dentry = ext4_fh_to_dentry,
	.fh_to_parent = ext4_fh_to_parent,
1138
	.get_parent = ext4_get_parent,
1139 1140 1141 1142 1143
};

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,
1144
	Opt_nouid32, Opt_debug, Opt_removed,
1145
	Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl,
1146
	Opt_auto_da_alloc, Opt_noauto_da_alloc, Opt_noload,
1147 1148
	Opt_commit, Opt_min_batch_time, Opt_max_batch_time, Opt_journal_dev,
	Opt_journal_path, Opt_journal_checksum, Opt_journal_async_commit,
1149
	Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback,
1150
	Opt_data_err_abort, Opt_data_err_ignore, Opt_test_dummy_encryption,
1151
	Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota,
J
Jan Kara 已提交
1152
	Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_jqfmt_vfsv1, Opt_quota,
T
Theodore Ts'o 已提交
1153
	Opt_noquota, Opt_barrier, Opt_nobarrier, Opt_err,
R
Ross Zwisler 已提交
1154
	Opt_usrquota, Opt_grpquota, Opt_i_version, Opt_dax,
1155
	Opt_stripe, Opt_delalloc, Opt_nodelalloc, Opt_mblk_io_submit,
1156
	Opt_lazytime, Opt_nolazytime,
1157
	Opt_nomblk_io_submit, Opt_block_validity, Opt_noblock_validity,
1158
	Opt_inode_readahead_blks, Opt_journal_ioprio,
1159
	Opt_dioread_nolock, Opt_dioread_lock,
1160
	Opt_discard, Opt_nodiscard, Opt_init_itable, Opt_noinit_itable,
1161
	Opt_max_dir_size_kb, Opt_nojournal_checksum,
1162 1163
};

1164
static const match_table_t tokens = {
1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178
	{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"},
1179 1180
	{Opt_removed, "oldalloc"},
	{Opt_removed, "orlov"},
1181 1182 1183 1184
	{Opt_user_xattr, "user_xattr"},
	{Opt_nouser_xattr, "nouser_xattr"},
	{Opt_acl, "acl"},
	{Opt_noacl, "noacl"},
1185
	{Opt_noload, "norecovery"},
1186
	{Opt_noload, "noload"},
1187 1188
	{Opt_removed, "nobh"},
	{Opt_removed, "bh"},
1189
	{Opt_commit, "commit=%u"},
1190 1191
	{Opt_min_batch_time, "min_batch_time=%u"},
	{Opt_max_batch_time, "max_batch_time=%u"},
1192
	{Opt_journal_dev, "journal_dev=%u"},
1193
	{Opt_journal_path, "journal_path=%s"},
1194
	{Opt_journal_checksum, "journal_checksum"},
1195
	{Opt_nojournal_checksum, "nojournal_checksum"},
1196
	{Opt_journal_async_commit, "journal_async_commit"},
1197 1198 1199 1200
	{Opt_abort, "abort"},
	{Opt_data_journal, "data=journal"},
	{Opt_data_ordered, "data=ordered"},
	{Opt_data_writeback, "data=writeback"},
1201 1202
	{Opt_data_err_abort, "data_err=abort"},
	{Opt_data_err_ignore, "data_err=ignore"},
1203 1204 1205 1206 1207 1208
	{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
1209
	{Opt_jqfmt_vfsv1, "jqfmt=vfsv1"},
1210 1211 1212 1213 1214
	{Opt_grpquota, "grpquota"},
	{Opt_noquota, "noquota"},
	{Opt_quota, "quota"},
	{Opt_usrquota, "usrquota"},
	{Opt_barrier, "barrier=%u"},
T
Theodore Ts'o 已提交
1215 1216
	{Opt_barrier, "barrier"},
	{Opt_nobarrier, "nobarrier"},
1217
	{Opt_i_version, "i_version"},
R
Ross Zwisler 已提交
1218
	{Opt_dax, "dax"},
1219
	{Opt_stripe, "stripe=%u"},
1220
	{Opt_delalloc, "delalloc"},
1221 1222
	{Opt_lazytime, "lazytime"},
	{Opt_nolazytime, "nolazytime"},
1223
	{Opt_nodelalloc, "nodelalloc"},
1224 1225
	{Opt_removed, "mblk_io_submit"},
	{Opt_removed, "nomblk_io_submit"},
1226 1227
	{Opt_block_validity, "block_validity"},
	{Opt_noblock_validity, "noblock_validity"},
1228
	{Opt_inode_readahead_blks, "inode_readahead_blks=%u"},
1229
	{Opt_journal_ioprio, "journal_ioprio=%u"},
1230
	{Opt_auto_da_alloc, "auto_da_alloc=%u"},
T
Theodore Ts'o 已提交
1231 1232
	{Opt_auto_da_alloc, "auto_da_alloc"},
	{Opt_noauto_da_alloc, "noauto_da_alloc"},
1233 1234
	{Opt_dioread_nolock, "dioread_nolock"},
	{Opt_dioread_lock, "dioread_lock"},
1235 1236
	{Opt_discard, "discard"},
	{Opt_nodiscard, "nodiscard"},
1237 1238 1239
	{Opt_init_itable, "init_itable=%u"},
	{Opt_init_itable, "init_itable"},
	{Opt_noinit_itable, "noinit_itable"},
1240
	{Opt_max_dir_size_kb, "max_dir_size_kb=%u"},
1241
	{Opt_test_dummy_encryption, "test_dummy_encryption"},
1242 1243 1244 1245 1246
	{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
1247
	{Opt_err, NULL},
1248 1249
};

1250
static ext4_fsblk_t get_sb_block(void **data)
1251
{
1252
	ext4_fsblk_t	sb_block;
1253 1254 1255 1256
	char		*options = (char *) *data;

	if (!options || strncmp(options, "sb=", 3) != 0)
		return 1;	/* Default location */
1257

1258
	options += 3;
1259
	/* TODO: use simple_strtoll with >32bit ext4 */
1260 1261
	sb_block = simple_strtoul(options, &options, 0);
	if (*options && *options != ',') {
1262
		printk(KERN_ERR "EXT4-fs: Invalid sb specification: %s\n",
1263 1264 1265 1266 1267 1268
		       (char *) *data);
		return 1;
	}
	if (*options == ',')
		options++;
	*data = (void *) options;
1269

1270 1271 1272
	return sb_block;
}

1273
#define DEFAULT_JOURNAL_IOPRIO (IOPRIO_PRIO_VALUE(IOPRIO_CLASS_BE, 3))
1274 1275
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";
1276

D
Dmitry Monakhov 已提交
1277 1278 1279 1280 1281
#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;
1282
	int ret = -1;
D
Dmitry Monakhov 已提交
1283 1284 1285 1286 1287 1288

	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");
1289
		return -1;
D
Dmitry Monakhov 已提交
1290
	}
1291 1292 1293 1294 1295
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_QUOTA)) {
		ext4_msg(sb, KERN_ERR, "Cannot set journaled quota options "
			 "when QUOTA feature is enabled");
		return -1;
	}
D
Dmitry Monakhov 已提交
1296 1297 1298 1299
	qname = match_strdup(args);
	if (!qname) {
		ext4_msg(sb, KERN_ERR,
			"Not enough memory for storing quotafile name");
1300
		return -1;
D
Dmitry Monakhov 已提交
1301
	}
1302 1303 1304 1305 1306 1307 1308 1309
	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
1310
	}
1311
	if (strchr(qname, '/')) {
D
Dmitry Monakhov 已提交
1312 1313
		ext4_msg(sb, KERN_ERR,
			"quotafile must be on filesystem root");
1314
		goto errout;
D
Dmitry Monakhov 已提交
1315
	}
1316
	sbi->s_qf_names[qtype] = qname;
1317
	set_opt(sb, QUOTA);
D
Dmitry Monakhov 已提交
1318
	return 1;
1319 1320 1321
errout:
	kfree(qname);
	return ret;
D
Dmitry Monakhov 已提交
1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332
}

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");
1333
		return -1;
D
Dmitry Monakhov 已提交
1334
	}
1335
	kfree(sbi->s_qf_names[qtype]);
D
Dmitry Monakhov 已提交
1336 1337 1338 1339 1340
	sbi->s_qf_names[qtype] = NULL;
	return 1;
}
#endif

1341 1342 1343 1344 1345 1346
#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
1347
#ifdef CONFIG_QUOTA
1348 1349 1350 1351 1352
#define MOPT_Q		0
#define MOPT_QFMT	0x0040
#else
#define MOPT_Q		MOPT_NOSUPPORT
#define MOPT_QFMT	MOPT_NOSUPPORT
1353
#endif
1354
#define MOPT_DATAJ	0x0080
1355 1356 1357
#define MOPT_NO_EXT2	0x0100
#define MOPT_NO_EXT3	0x0200
#define MOPT_EXT4_ONLY	(MOPT_NO_EXT2 | MOPT_NO_EXT3)
1358
#define MOPT_STRING	0x0400
1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370

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},
1371 1372 1373 1374
	{Opt_dioread_nolock, EXT4_MOUNT_DIOREAD_NOLOCK,
	 MOPT_EXT4_ONLY | MOPT_SET},
	{Opt_dioread_lock, EXT4_MOUNT_DIOREAD_NOLOCK,
	 MOPT_EXT4_ONLY | MOPT_CLEAR},
1375 1376
	{Opt_discard, EXT4_MOUNT_DISCARD, MOPT_SET},
	{Opt_nodiscard, EXT4_MOUNT_DISCARD, MOPT_CLEAR},
1377 1378 1379
	{Opt_delalloc, EXT4_MOUNT_DELALLOC,
	 MOPT_EXT4_ONLY | MOPT_SET | MOPT_EXPLICIT},
	{Opt_nodelalloc, EXT4_MOUNT_DELALLOC,
1380
	 MOPT_EXT4_ONLY | MOPT_CLEAR},
1381 1382
	{Opt_nojournal_checksum, EXT4_MOUNT_JOURNAL_CHECKSUM,
	 MOPT_EXT4_ONLY | MOPT_CLEAR},
1383 1384
	{Opt_journal_checksum, EXT4_MOUNT_JOURNAL_CHECKSUM,
	 MOPT_EXT4_ONLY | MOPT_SET},
1385
	{Opt_journal_async_commit, (EXT4_MOUNT_JOURNAL_ASYNC_COMMIT |
1386 1387 1388
				    EXT4_MOUNT_JOURNAL_CHECKSUM),
	 MOPT_EXT4_ONLY | MOPT_SET},
	{Opt_noload, EXT4_MOUNT_NOLOAD, MOPT_NO_EXT2 | MOPT_SET},
1389 1390 1391
	{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},
1392 1393 1394 1395
	{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},
1396 1397 1398 1399 1400 1401 1402 1403 1404 1405
	{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
1406
	{Opt_dax, EXT4_MOUNT_DAX, MOPT_SET},
1407
	{Opt_stripe, 0, MOPT_GTE0},
1408 1409
	{Opt_resuid, 0, MOPT_GTE0},
	{Opt_resgid, 0, MOPT_GTE0},
1410 1411 1412
	{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},
1413 1414 1415 1416
	{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},
1417 1418
	{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
1419
#ifdef CONFIG_EXT4_FS_POSIX_ACL
1420 1421
	{Opt_acl, EXT4_MOUNT_POSIX_ACL, MOPT_SET},
	{Opt_noacl, EXT4_MOUNT_POSIX_ACL, MOPT_CLEAR},
1422
#else
1423 1424
	{Opt_acl, 0, MOPT_NOSUPPORT},
	{Opt_noacl, 0, MOPT_NOSUPPORT},
1425
#endif
1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441
	{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},
1442
	{Opt_max_dir_size_kb, 0, MOPT_GTE0},
1443
	{Opt_test_dummy_encryption, 0, MOPT_GTE0},
1444 1445 1446 1447 1448 1449 1450 1451 1452
	{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;
1453 1454
	kuid_t uid;
	kgid_t gid;
1455 1456
	int arg = 0;

1457 1458 1459 1460 1461 1462 1463 1464 1465 1466
#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
1467
	switch (token) {
1468 1469 1470 1471
	case Opt_noacl:
	case Opt_nouser_xattr:
		ext4_msg(sb, KERN_WARNING, deprecated_msg, opt, "3.5");
		break;
1472 1473 1474
	case Opt_sb:
		return 1;	/* handled by get_sb_block() */
	case Opt_removed:
1475
		ext4_msg(sb, KERN_WARNING, "Ignoring removed %s option", opt);
1476 1477 1478 1479 1480 1481 1482
		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;
1483 1484 1485 1486 1487 1488
	case Opt_lazytime:
		sb->s_flags |= MS_LAZYTIME;
		return 1;
	case Opt_nolazytime:
		sb->s_flags &= ~MS_LAZYTIME;
		return 1;
1489 1490
	}

1491 1492 1493 1494 1495 1496 1497 1498 1499 1500
	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;
	}

1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511
	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;
	}

1512
	if (args->from && !(m->flags & MOPT_STRING) && match_int(args, &arg))
1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536
		return -1;
	if (args->from && (m->flags & MOPT_GTE0) && (arg < 0))
		return -1;
	if (m->flags & MOPT_EXPLICIT)
		set_opt2(sb, EXPLICIT_DELALLOC);
	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) {
1537 1538 1539 1540
		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");
1541
			return -1;
1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556
		}
		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);
1557 1558
			return -1;
		}
1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573
		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;
1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599
	} 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;
		}

1600
		journal_inode = d_inode(path.dentry);
1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611
		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);
1612 1613 1614 1615 1616 1617 1618 1619
	} 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);
1620 1621 1622 1623 1624 1625 1626 1627 1628
	} 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
1629 1630 1631 1632 1633
	} 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) {
1634
				ext4_msg(sb, KERN_ERR,
1635 1636
					 "Cannot change data mode on remount");
				return -1;
1637
			}
1638
		} else {
1639 1640
			clear_opt(sb, DATA_FLAGS);
			sbi->s_mount_opt |= m->mount_opt;
1641
		}
1642 1643 1644 1645 1646 1647 1648 1649
#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;
		}
1650 1651 1652 1653 1654 1655 1656
		if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
					       EXT4_FEATURE_RO_COMPAT_QUOTA)) {
			ext4_msg(sb, KERN_ERR,
				 "Cannot set journaled quota options "
				 "when QUOTA feature is enabled");
			return -1;
		}
1657
		sbi->s_jquota_fmt = m->mount_opt;
R
Ross Zwisler 已提交
1658 1659 1660 1661 1662
#endif
#ifndef CONFIG_FS_DAX
	} else if (token == Opt_dax) {
		ext4_msg(sb, KERN_INFO, "dax option not supported");
		return -1;
1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678
#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;
1679
	}
1680
	return 1;
1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702
}

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.
		 */
1703
		args[0].to = args[0].from = NULL;
1704 1705 1706 1707
		token = match_token(p, tokens, args);
		if (handle_mount_opt(sb, p, token, args, journal_devnum,
				     journal_ioprio, is_remount) < 0)
			return 0;
1708 1709
	}
#ifdef CONFIG_QUOTA
1710 1711 1712 1713 1714 1715
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_QUOTA) &&
	    (test_opt(sb, USRQUOTA) || test_opt(sb, GRPQUOTA))) {
		ext4_msg(sb, KERN_ERR, "Cannot set quota options when QUOTA "
			 "feature is enabled");
		return 0;
	}
1716
	if (sbi->s_qf_names[USRQUOTA] || sbi->s_qf_names[GRPQUOTA]) {
1717
		if (test_opt(sb, USRQUOTA) && sbi->s_qf_names[USRQUOTA])
1718
			clear_opt(sb, USRQUOTA);
1719

1720
		if (test_opt(sb, GRPQUOTA) && sbi->s_qf_names[GRPQUOTA])
1721
			clear_opt(sb, GRPQUOTA);
1722

D
Dmitry Monakhov 已提交
1723
		if (test_opt(sb, GRPQUOTA) || test_opt(sb, USRQUOTA)) {
1724 1725
			ext4_msg(sb, KERN_ERR, "old and new quota "
					"format mixing");
1726 1727 1728 1729
			return 0;
		}

		if (!sbi->s_jquota_fmt) {
1730 1731
			ext4_msg(sb, KERN_ERR, "journaled quota format "
					"not specified");
1732 1733 1734 1735
			return 0;
		}
	}
#endif
J
Jan Kara 已提交
1736 1737 1738 1739 1740 1741 1742 1743 1744 1745
	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;
		}
	}
1746 1747 1748 1749 1750 1751
	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;
	}
1752 1753 1754
	return 1;
}

1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785
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])
		seq_printf(seq, ",usrjquota=%s", sbi->s_qf_names[USRQUOTA]);

	if (sbi->s_qf_names[GRPQUOTA])
		seq_printf(seq, ",grpjquota=%s", sbi->s_qf_names[GRPQUOTA]);
#endif
}

1786 1787
static const char *token2str(int token)
{
1788
	const struct match_token *t;
1789 1790 1791 1792 1793 1794 1795

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

1796 1797 1798 1799 1800
/*
 * Show an option if
 *  - it's set to a non-default value OR
 *  - if the per-sb default is different from the global default
 */
1801 1802
static int _ext4_show_options(struct seq_file *seq, struct super_block *sb,
			      int nodefs)
1803 1804 1805
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_super_block *es = sbi->s_es;
1806
	int def_errors, def_mount_opt = nodefs ? 0 : sbi->s_def_mount_opt;
1807
	const struct mount_opts *m;
1808
	char sep = nodefs ? '\n' : ',';
1809

1810 1811
#define SEQ_OPTS_PUTS(str) seq_printf(seq, "%c" str, sep)
#define SEQ_OPTS_PRINT(str, arg) seq_printf(seq, "%c" str, sep, arg)
1812 1813

	if (sbi->s_sb_block != 1)
1814 1815 1816 1817 1818 1819 1820
		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;
1821
		if (!(m->mount_opt & (sbi->s_mount_opt ^ def_mount_opt)))
1822 1823 1824 1825 1826 1827
			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));
1828
	}
1829

1830
	if (nodefs || !uid_eq(sbi->s_resuid, make_kuid(&init_user_ns, EXT4_DEF_RESUID)) ||
1831
	    le16_to_cpu(es->s_def_resuid) != EXT4_DEF_RESUID)
1832 1833 1834
		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)) ||
1835
	    le16_to_cpu(es->s_def_resgid) != EXT4_DEF_RESGID)
1836 1837
		SEQ_OPTS_PRINT("resgid=%u",
				from_kgid_munged(&init_user_ns, sbi->s_resgid));
1838
	def_errors = nodefs ? -1 : le16_to_cpu(es->s_errors);
1839 1840
	if (test_opt(sb, ERRORS_RO) && def_errors != EXT4_ERRORS_RO)
		SEQ_OPTS_PUTS("errors=remount-ro");
1841
	if (test_opt(sb, ERRORS_CONT) && def_errors != EXT4_ERRORS_CONTINUE)
1842
		SEQ_OPTS_PUTS("errors=continue");
1843
	if (test_opt(sb, ERRORS_PANIC) && def_errors != EXT4_ERRORS_PANIC)
1844
		SEQ_OPTS_PUTS("errors=panic");
1845
	if (nodefs || sbi->s_commit_interval != JBD2_DEFAULT_MAX_COMMIT_AGE*HZ)
1846
		SEQ_OPTS_PRINT("commit=%lu", sbi->s_commit_interval / HZ);
1847
	if (nodefs || sbi->s_min_batch_time != EXT4_DEF_MIN_BATCH_TIME)
1848
		SEQ_OPTS_PRINT("min_batch_time=%u", sbi->s_min_batch_time);
1849
	if (nodefs || sbi->s_max_batch_time != EXT4_DEF_MAX_BATCH_TIME)
1850
		SEQ_OPTS_PRINT("max_batch_time=%u", sbi->s_max_batch_time);
1851
	if (sb->s_flags & MS_I_VERSION)
1852
		SEQ_OPTS_PUTS("i_version");
1853
	if (nodefs || sbi->s_stripe)
1854
		SEQ_OPTS_PRINT("stripe=%lu", sbi->s_stripe);
1855
	if (EXT4_MOUNT_DATA_FLAGS & (sbi->s_mount_opt ^ def_mount_opt)) {
1856 1857 1858 1859 1860 1861 1862
		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");
	}
1863 1864
	if (nodefs ||
	    sbi->s_inode_readahead_blks != EXT4_DEF_INODE_READAHEAD_BLKS)
1865 1866
		SEQ_OPTS_PRINT("inode_readahead_blks=%u",
			       sbi->s_inode_readahead_blks);
1867

1868 1869
	if (nodefs || (test_opt(sb, INIT_INODE_TABLE) &&
		       (sbi->s_li_wait_mult != EXT4_DEF_LI_WAIT_MULT)))
1870
		SEQ_OPTS_PRINT("init_itable=%u", sbi->s_li_wait_mult);
1871 1872
	if (nodefs || sbi->s_max_dir_size_kb)
		SEQ_OPTS_PRINT("max_dir_size_kb=%u", sbi->s_max_dir_size_kb);
1873 1874 1875 1876 1877

	ext4_show_quota_options(seq, sb);
	return 0;
}

1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895
static int ext4_show_options(struct seq_file *seq, struct dentry *root)
{
	return _ext4_show_options(seq, root->d_sb, 0);
}

static int options_seq_show(struct seq_file *seq, void *offset)
{
	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;
}

static int options_open_fs(struct inode *inode, struct file *file)
{
A
Al Viro 已提交
1896
	return single_open(file, options_seq_show, PDE_DATA(inode));
1897 1898 1899 1900 1901 1902 1903 1904 1905 1906
}

static const struct file_operations ext4_seq_options_fops = {
	.owner = THIS_MODULE,
	.open = options_open_fs,
	.read = seq_read,
	.llseek = seq_lseek,
	.release = single_release,
};

1907
static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es,
1908 1909
			    int read_only)
{
1910
	struct ext4_sb_info *sbi = EXT4_SB(sb);
1911 1912
	int res = 0;

1913
	if (le32_to_cpu(es->s_rev_level) > EXT4_MAX_SUPP_REV) {
1914 1915
		ext4_msg(sb, KERN_ERR, "revision level too high, "
			 "forcing read-only mode");
1916 1917 1918
		res = MS_RDONLY;
	}
	if (read_only)
1919
		goto done;
1920
	if (!(sbi->s_mount_state & EXT4_VALID_FS))
1921 1922
		ext4_msg(sb, KERN_WARNING, "warning: mounting unchecked fs, "
			 "running e2fsck is recommended");
1923
	else if (sbi->s_mount_state & EXT4_ERROR_FS)
1924 1925 1926
		ext4_msg(sb, KERN_WARNING,
			 "warning: mounting fs with errors, "
			 "running e2fsck is recommended");
1927
	else if ((__s16) le16_to_cpu(es->s_max_mnt_count) > 0 &&
1928 1929
		 le16_to_cpu(es->s_mnt_count) >=
		 (unsigned short) (__s16) le16_to_cpu(es->s_max_mnt_count))
1930 1931 1932
		ext4_msg(sb, KERN_WARNING,
			 "warning: maximal mount count reached, "
			 "running e2fsck is recommended");
1933 1934 1935
	else if (le32_to_cpu(es->s_checkinterval) &&
		(le32_to_cpu(es->s_lastcheck) +
			le32_to_cpu(es->s_checkinterval) <= get_seconds()))
1936 1937 1938
		ext4_msg(sb, KERN_WARNING,
			 "warning: checktime reached, "
			 "running e2fsck is recommended");
1939
	if (!sbi->s_journal)
1940
		es->s_state &= cpu_to_le16(~EXT4_VALID_FS);
1941
	if (!(__s16) le16_to_cpu(es->s_max_mnt_count))
1942
		es->s_max_mnt_count = cpu_to_le16(EXT4_DFL_MAX_MNT_COUNT);
M
Marcin Slusarz 已提交
1943
	le16_add_cpu(&es->s_mnt_count, 1);
1944
	es->s_mtime = cpu_to_le32(get_seconds());
1945
	ext4_update_dynamic_rev(sb);
1946 1947
	if (sbi->s_journal)
		EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
1948

1949
	ext4_commit_super(sb, 1);
1950
done:
1951
	if (test_opt(sb, DEBUG))
1952
		printk(KERN_INFO "[EXT4 FS bs=%lu, gc=%u, "
1953
				"bpg=%lu, ipg=%lu, mo=%04x, mo2=%04x]\n",
1954 1955
			sb->s_blocksize,
			sbi->s_groups_count,
1956 1957
			EXT4_BLOCKS_PER_GROUP(sb),
			EXT4_INODES_PER_GROUP(sb),
1958
			sbi->s_mount_opt, sbi->s_mount_opt2);
1959

D
Dan Magenheimer 已提交
1960
	cleancache_init_fs(sb);
1961 1962 1963
	return res;
}

1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988
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
1989
		kvfree(sbi->s_flex_groups);
1990 1991 1992 1993 1994 1995
	}
	sbi->s_flex_groups = new_groups;
	sbi->s_flex_groups_allocated = size / sizeof(struct flex_groups);
	return 0;
}

1996 1997 1998 1999 2000
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;
2001
	int i, err;
2002

2003
	sbi->s_log_groups_per_flex = sbi->s_es->s_log_groups_per_flex;
2004
	if (sbi->s_log_groups_per_flex < 1 || sbi->s_log_groups_per_flex > 31) {
2005 2006 2007 2008
		sbi->s_log_groups_per_flex = 0;
		return 1;
	}

2009 2010
	err = ext4_alloc_flex_bg_array(sb, sbi->s_groups_count);
	if (err)
2011
		goto failed;
2012 2013

	for (i = 0; i < sbi->s_groups_count; i++) {
2014
		gdp = ext4_get_group_desc(sb, i, NULL);
2015 2016

		flex_group = ext4_flex_group(sbi, i);
2017 2018
		atomic_add(ext4_free_inodes_count(sb, gdp),
			   &sbi->s_flex_groups[flex_group].free_inodes);
2019 2020
		atomic64_add(ext4_free_group_clusters(sb, gdp),
			     &sbi->s_flex_groups[flex_group].free_clusters);
2021 2022
		atomic_add(ext4_used_dirs_count(sb, gdp),
			   &sbi->s_flex_groups[flex_group].used_dirs);
2023 2024 2025 2026 2027 2028 2029
	}

	return 1;
failed:
	return 0;
}

2030 2031
static __le16 ext4_group_desc_csum(struct ext4_sb_info *sbi, __u32 block_group,
				   struct ext4_group_desc *gdp)
A
Andreas Dilger 已提交
2032
{
2033
	int offset;
A
Andreas Dilger 已提交
2034
	__u16 crc = 0;
2035
	__le32 le_group = cpu_to_le32(block_group);
A
Andreas Dilger 已提交
2036

2037
	if (ext4_has_metadata_csum(sbi->s_sb)) {
2038
		/* Use new metadata_csum algorithm */
2039
		__le16 save_csum;
2040 2041
		__u32 csum32;

2042
		save_csum = gdp->bg_checksum;
2043 2044 2045 2046 2047
		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);
2048
		gdp->bg_checksum = save_csum;
2049 2050 2051

		crc = csum32 & 0xFFFF;
		goto out;
A
Andreas Dilger 已提交
2052 2053
	}

2054
	/* old crc16 code */
2055 2056 2057 2058
	if (!(sbi->s_es->s_feature_ro_compat &
	      cpu_to_le32(EXT4_FEATURE_RO_COMPAT_GDT_CSUM)))
		return 0;

2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073
	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...*/
	if ((sbi->s_es->s_feature_incompat &
	     cpu_to_le32(EXT4_FEATURE_INCOMPAT_64BIT)) &&
	    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
2074 2075 2076
	return cpu_to_le16(crc);
}

2077
int ext4_group_desc_csum_verify(struct super_block *sb, __u32 block_group,
A
Andreas Dilger 已提交
2078 2079
				struct ext4_group_desc *gdp)
{
2080 2081 2082
	if (ext4_has_group_desc_csum(sb) &&
	    (gdp->bg_checksum != ext4_group_desc_csum(EXT4_SB(sb),
						      block_group, gdp)))
A
Andreas Dilger 已提交
2083 2084 2085 2086 2087
		return 0;

	return 1;
}

2088 2089 2090 2091 2092 2093 2094 2095
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;
	gdp->bg_checksum = ext4_group_desc_csum(EXT4_SB(sb), block_group, gdp);
}

2096
/* Called at mount-time, super-block is locked */
2097 2098
static int ext4_check_descriptors(struct super_block *sb,
				  ext4_group_t *first_not_zeroed)
2099
{
2100 2101 2102
	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
2103 2104 2105
	ext4_fsblk_t block_bitmap;
	ext4_fsblk_t inode_bitmap;
	ext4_fsblk_t inode_table;
J
Jose R. Santos 已提交
2106
	int flexbg_flag = 0;
2107
	ext4_group_t i, grp = sbi->s_groups_count;
2108

J
Jose R. Santos 已提交
2109 2110 2111
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG))
		flexbg_flag = 1;

2112
	ext4_debug("Checking group descriptors");
2113

2114 2115 2116
	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
2117
		if (i == sbi->s_groups_count - 1 || flexbg_flag)
L
Laurent Vivier 已提交
2118
			last_block = ext4_blocks_count(sbi->s_es) - 1;
2119 2120
		else
			last_block = first_block +
2121
				(EXT4_BLOCKS_PER_GROUP(sb) - 1);
2122

2123 2124 2125 2126
		if ((grp == sbi->s_groups_count) &&
		   !(gdp->bg_flags & cpu_to_le16(EXT4_BG_INODE_ZEROED)))
			grp = i;

2127
		block_bitmap = ext4_block_bitmap(sb, gdp);
2128
		if (block_bitmap < first_block || block_bitmap > last_block) {
2129
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
2130
			       "Block bitmap for group %u not in group "
2131
			       "(block %llu)!", i, block_bitmap);
2132 2133
			return 0;
		}
2134
		inode_bitmap = ext4_inode_bitmap(sb, gdp);
2135
		if (inode_bitmap < first_block || inode_bitmap > last_block) {
2136
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
2137
			       "Inode bitmap for group %u not in group "
2138
			       "(block %llu)!", i, inode_bitmap);
2139 2140
			return 0;
		}
2141
		inode_table = ext4_inode_table(sb, gdp);
L
Laurent Vivier 已提交
2142
		if (inode_table < first_block ||
2143
		    inode_table + sbi->s_itb_per_group - 1 > last_block) {
2144
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
2145
			       "Inode table for group %u not in group "
2146
			       "(block %llu)!", i, inode_table);
2147 2148
			return 0;
		}
2149
		ext4_lock_group(sb, i);
2150
		if (!ext4_group_desc_csum_verify(sb, i, gdp)) {
2151 2152 2153 2154
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
				 "Checksum for group %u failed (%u!=%u)",
				 i, le16_to_cpu(ext4_group_desc_csum(sbi, i,
				     gdp)), le16_to_cpu(gdp->bg_checksum));
2155
			if (!(sb->s_flags & MS_RDONLY)) {
2156
				ext4_unlock_group(sb, i);
2157
				return 0;
2158
			}
A
Andreas Dilger 已提交
2159
		}
2160
		ext4_unlock_group(sb, i);
J
Jose R. Santos 已提交
2161 2162
		if (!flexbg_flag)
			first_block += EXT4_BLOCKS_PER_GROUP(sb);
2163
	}
2164 2165
	if (NULL != first_not_zeroed)
		*first_not_zeroed = grp;
2166 2167 2168
	return 1;
}

2169
/* ext4_orphan_cleanup() walks a singly-linked list of inodes (starting at
2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181
 * 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
2182
 * ext4_free_inode().  The only reason we would point at a wrong inode is if
2183 2184 2185
 * 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.
 */
2186 2187
static void ext4_orphan_cleanup(struct super_block *sb,
				struct ext4_super_block *es)
2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198
{
	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;
	}

2199
	if (bdev_read_only(sb->s_bdev)) {
2200 2201
		ext4_msg(sb, KERN_ERR, "write access "
			"unavailable, skipping orphan cleanup");
2202 2203 2204
		return;
	}

2205 2206 2207 2208 2209 2210 2211
	/* 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;
	}

2212
	if (EXT4_SB(sb)->s_mount_state & EXT4_ERROR_FS) {
2213 2214
		/* don't clear list on RO mount w/ errors */
		if (es->s_last_orphan && !(s_flags & MS_RDONLY)) {
2215
			ext4_msg(sb, KERN_INFO, "Errors on filesystem, "
2216
				  "clearing orphan list.\n");
2217 2218
			es->s_last_orphan = 0;
		}
2219 2220 2221 2222 2223
		jbd_debug(1, "Skipping orphan recovery on fs with errors.\n");
		return;
	}

	if (s_flags & MS_RDONLY) {
2224
		ext4_msg(sb, KERN_INFO, "orphan cleanup on readonly fs");
2225 2226 2227 2228 2229 2230
		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
2231
	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
2232 2233
		if (EXT4_SB(sb)->s_qf_names[i]) {
			int ret = ext4_quota_on_mount(sb, i);
2234
			if (ret < 0)
2235 2236 2237
				ext4_msg(sb, KERN_ERR,
					"Cannot turn on journaled "
					"quota: error %d", ret);
2238 2239 2240 2241 2242 2243 2244
		}
	}
#endif

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

2245 2246
		inode = ext4_orphan_get(sb, le32_to_cpu(es->s_last_orphan));
		if (IS_ERR(inode)) {
2247 2248 2249 2250
			es->s_last_orphan = 0;
			break;
		}

2251
		list_add(&EXT4_I(inode)->i_orphan, &EXT4_SB(sb)->s_orphan);
2252
		dquot_initialize(inode);
2253
		if (inode->i_nlink) {
2254 2255 2256 2257
			if (test_opt(sb, DEBUG))
				ext4_msg(sb, KERN_DEBUG,
					"%s: truncating inode %lu to %lld bytes",
					__func__, inode->i_ino, inode->i_size);
2258
			jbd_debug(2, "truncating inode %lu to %lld bytes\n",
2259
				  inode->i_ino, inode->i_size);
2260
			mutex_lock(&inode->i_mutex);
2261
			truncate_inode_pages(inode->i_mapping, inode->i_size);
2262
			ext4_truncate(inode);
2263
			mutex_unlock(&inode->i_mutex);
2264 2265
			nr_truncates++;
		} else {
2266 2267 2268 2269
			if (test_opt(sb, DEBUG))
				ext4_msg(sb, KERN_DEBUG,
					"%s: deleting unreferenced inode %lu",
					__func__, inode->i_ino);
2270 2271 2272 2273 2274 2275 2276
			jbd_debug(2, "deleting unreferenced inode %lu\n",
				  inode->i_ino);
			nr_orphans++;
		}
		iput(inode);  /* The delete magic happens here! */
	}

2277
#define PLURAL(x) (x), ((x) == 1) ? "" : "s"
2278 2279

	if (nr_orphans)
2280 2281
		ext4_msg(sb, KERN_INFO, "%d orphan inode%s deleted",
		       PLURAL(nr_orphans));
2282
	if (nr_truncates)
2283 2284
		ext4_msg(sb, KERN_INFO, "%d truncate%s cleaned up",
		       PLURAL(nr_truncates));
2285 2286
#ifdef CONFIG_QUOTA
	/* Turn quotas off */
J
Jan Kara 已提交
2287
	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
2288
		if (sb_dqopt(sb)->files[i])
2289
			dquot_quota_off(sb, i);
2290 2291 2292 2293
	}
#endif
	sb->s_flags = s_flags; /* Restore MS_RDONLY status */
}
2294

2295 2296 2297 2298 2299 2300 2301
/*
 * 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.
 *
2302 2303 2304 2305 2306 2307
 * 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.
 *
2308 2309
 * Note, this does *not* consider any metadata overhead for vfs i_blocks.
 */
2310
static loff_t ext4_max_size(int blkbits, int has_huge_files)
2311 2312 2313 2314 2315
{
	loff_t res;
	loff_t upper_limit = MAX_LFS_FILESIZE;

	/* small i_blocks in vfs inode? */
2316
	if (!has_huge_files || sizeof(blkcnt_t) < sizeof(u64)) {
2317
		/*
2318
		 * CONFIG_LBDAF is not enabled implies the inode
2319 2320 2321 2322 2323 2324 2325 2326 2327 2328
		 * 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;
	}

2329 2330 2331 2332 2333 2334
	/*
	 * 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;
2335 2336 2337 2338 2339 2340 2341 2342
	res <<= blkbits;

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

	return res;
}
2343 2344

/*
2345
 * Maximal bitmap file size.  There is a direct, and {,double-,triple-}indirect
2346 2347
 * 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.
2348
 */
2349
static loff_t ext4_max_bitmap_size(int bits, int has_huge_files)
2350
{
2351
	loff_t res = EXT4_NDIR_BLOCKS;
2352 2353
	int meta_blocks;
	loff_t upper_limit;
2354 2355 2356 2357 2358 2359
	/* 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.
2360 2361
	 */

2362
	if (!has_huge_files || sizeof(blkcnt_t) < sizeof(u64)) {
2363
		/*
2364
		 * !has_huge_files or CONFIG_LBDAF not enabled implies that
2365 2366
		 * the inode i_block field represents total file blocks in
		 * 2^32 512-byte sectors == size of vfs inode i_blocks * 8
2367 2368 2369 2370 2371 2372 2373
		 */
		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
2374 2375 2376 2377 2378 2379
		/*
		 * 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
		 */
2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392
		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;
2393 2394 2395 2396 2397 2398 2399

	res += 1LL << (bits-2);
	res += 1LL << (2*(bits-2));
	res += 1LL << (3*(bits-2));
	res <<= bits;
	if (res > upper_limit)
		res = upper_limit;
2400 2401 2402 2403

	if (res > MAX_LFS_FILESIZE)
		res = MAX_LFS_FILESIZE;

2404 2405 2406
	return res;
}

2407
static ext4_fsblk_t descriptor_loc(struct super_block *sb,
2408
				   ext4_fsblk_t logical_sb_block, int nr)
2409
{
2410
	struct ext4_sb_info *sbi = EXT4_SB(sb);
2411
	ext4_group_t bg, first_meta_bg;
2412 2413 2414 2415
	int has_super = 0;

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

2416
	if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_META_BG) ||
2417
	    nr < first_meta_bg)
2418
		return logical_sb_block + nr + 1;
2419
	bg = sbi->s_desc_per_block * nr;
2420
	if (ext4_bg_has_super(sb, bg))
2421
		has_super = 1;
2422

2423 2424 2425 2426 2427 2428 2429 2430 2431 2432
	/*
	 * 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++;

2433
	return (has_super + ext4_group_first_block_no(sb, bg));
2434 2435
}

2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451
/**
 * 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
2452
	int ret;
2453 2454

	if (sbi->s_stripe && sbi->s_stripe <= sbi->s_blocks_per_group)
D
Dan Ehrenberg 已提交
2455 2456 2457 2458 2459 2460 2461
		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;
2462

D
Dan Ehrenberg 已提交
2463 2464 2465 2466 2467 2468
	/*
	 * 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;
2469

D
Dan Ehrenberg 已提交
2470
	return ret;
2471
}
2472

T
Theodore Ts'o 已提交
2473 2474 2475 2476 2477
/* sysfs supprt */

struct ext4_attr {
	struct attribute attr;
	ssize_t (*show)(struct ext4_attr *, struct ext4_sb_info *, char *);
2478
	ssize_t (*store)(struct ext4_attr *, struct ext4_sb_info *,
T
Theodore Ts'o 已提交
2479
			 const char *, size_t);
2480 2481 2482 2483
	union {
		int offset;
		int deprecated_val;
	} u;
T
Theodore Ts'o 已提交
2484 2485
};

L
Lukas Czerner 已提交
2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496
static int parse_strtoull(const char *buf,
		unsigned long long max, unsigned long long *value)
{
	int ret;

	ret = kstrtoull(skip_spaces(buf), 0, value);
	if (!ret && *value > max)
		ret = -EINVAL;
	return ret;
}

T
Theodore Ts'o 已提交
2497 2498 2499 2500 2501
static ssize_t delayed_allocation_blocks_show(struct ext4_attr *a,
					      struct ext4_sb_info *sbi,
					      char *buf)
{
	return snprintf(buf, PAGE_SIZE, "%llu\n",
2502 2503
		(s64) EXT4_C2B(sbi,
			percpu_counter_sum(&sbi->s_dirtyclusters_counter)));
T
Theodore Ts'o 已提交
2504 2505 2506 2507 2508 2509 2510
}

static ssize_t session_write_kbytes_show(struct ext4_attr *a,
					 struct ext4_sb_info *sbi, char *buf)
{
	struct super_block *sb = sbi->s_buddy_cache->i_sb;

2511 2512
	if (!sb->s_bdev->bd_part)
		return snprintf(buf, PAGE_SIZE, "0\n");
T
Theodore Ts'o 已提交
2513 2514 2515 2516 2517 2518 2519 2520 2521 2522
	return snprintf(buf, PAGE_SIZE, "%lu\n",
			(part_stat_read(sb->s_bdev->bd_part, sectors[1]) -
			 sbi->s_sectors_written_start) >> 1);
}

static ssize_t lifetime_write_kbytes_show(struct ext4_attr *a,
					  struct ext4_sb_info *sbi, char *buf)
{
	struct super_block *sb = sbi->s_buddy_cache->i_sb;

2523 2524
	if (!sb->s_bdev->bd_part)
		return snprintf(buf, PAGE_SIZE, "0\n");
T
Theodore Ts'o 已提交
2525
	return snprintf(buf, PAGE_SIZE, "%llu\n",
2526
			(unsigned long long)(sbi->s_kbytes_written +
T
Theodore Ts'o 已提交
2527
			((part_stat_read(sb->s_bdev->bd_part, sectors[1]) -
2528
			  EXT4_SB(sb)->s_sectors_written_start) >> 1)));
T
Theodore Ts'o 已提交
2529 2530 2531 2532 2533 2534 2535
}

static ssize_t inode_readahead_blks_store(struct ext4_attr *a,
					  struct ext4_sb_info *sbi,
					  const char *buf, size_t count)
{
	unsigned long t;
2536
	int ret;
T
Theodore Ts'o 已提交
2537

2538 2539 2540
	ret = kstrtoul(skip_spaces(buf), 0, &t);
	if (ret)
		return ret;
T
Theodore Ts'o 已提交
2541

2542
	if (t && (!is_power_of_2(t) || t > 0x40000000))
T
Theodore Ts'o 已提交
2543 2544 2545 2546 2547 2548 2549
		return -EINVAL;

	sbi->s_inode_readahead_blks = t;
	return count;
}

static ssize_t sbi_ui_show(struct ext4_attr *a,
2550
			   struct ext4_sb_info *sbi, char *buf)
T
Theodore Ts'o 已提交
2551
{
2552
	unsigned int *ui = (unsigned int *) (((char *) sbi) + a->u.offset);
T
Theodore Ts'o 已提交
2553 2554 2555 2556 2557 2558 2559 2560

	return snprintf(buf, PAGE_SIZE, "%u\n", *ui);
}

static ssize_t sbi_ui_store(struct ext4_attr *a,
			    struct ext4_sb_info *sbi,
			    const char *buf, size_t count)
{
2561
	unsigned int *ui = (unsigned int *) (((char *) sbi) + a->u.offset);
T
Theodore Ts'o 已提交
2562
	unsigned long t;
2563
	int ret;
T
Theodore Ts'o 已提交
2564

2565 2566 2567
	ret = kstrtoul(skip_spaces(buf), 0, &t);
	if (ret)
		return ret;
T
Theodore Ts'o 已提交
2568 2569 2570 2571
	*ui = t;
	return count;
}

2572 2573 2574 2575 2576 2577 2578 2579 2580 2581
static ssize_t es_ui_show(struct ext4_attr *a,
			   struct ext4_sb_info *sbi, char *buf)
{

	unsigned int *ui = (unsigned int *) (((char *) sbi->s_es) +
			   a->u.offset);

	return snprintf(buf, PAGE_SIZE, "%u\n", *ui);
}

L
Lukas Czerner 已提交
2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602
static ssize_t reserved_clusters_show(struct ext4_attr *a,
				  struct ext4_sb_info *sbi, char *buf)
{
	return snprintf(buf, PAGE_SIZE, "%llu\n",
		(unsigned long long) atomic64_read(&sbi->s_resv_clusters));
}

static ssize_t reserved_clusters_store(struct ext4_attr *a,
				   struct ext4_sb_info *sbi,
				   const char *buf, size_t count)
{
	unsigned long long val;
	int ret;

	if (parse_strtoull(buf, -1ULL, &val))
		return -EINVAL;
	ret = ext4_reserve_clusters(sbi, val);

	return ret ? ret : count;
}

2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619
static ssize_t trigger_test_error(struct ext4_attr *a,
				  struct ext4_sb_info *sbi,
				  const char *buf, size_t count)
{
	int len = count;

	if (!capable(CAP_SYS_ADMIN))
		return -EPERM;

	if (len && buf[len-1] == '\n')
		len--;

	if (len)
		ext4_error(sbi->s_sb, "%.*s", len, buf);
	return count;
}

2620 2621 2622 2623 2624 2625
static ssize_t sbi_deprecated_show(struct ext4_attr *a,
				   struct ext4_sb_info *sbi, char *buf)
{
	return snprintf(buf, PAGE_SIZE, "%d\n", a->u.deprecated_val);
}

T
Theodore Ts'o 已提交
2626 2627 2628 2629 2630
#define EXT4_ATTR_OFFSET(_name,_mode,_show,_store,_elname) \
static struct ext4_attr ext4_attr_##_name = {			\
	.attr = {.name = __stringify(_name), .mode = _mode },	\
	.show	= _show,					\
	.store	= _store,					\
2631 2632 2633
	.u = {							\
		.offset = offsetof(struct ext4_sb_info, _elname),\
	},							\
T
Theodore Ts'o 已提交
2634
}
2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645

#define EXT4_ATTR_OFFSET_ES(_name,_mode,_show,_store,_elname)		\
static struct ext4_attr ext4_attr_##_name = {				\
	.attr = {.name = __stringify(_name), .mode = _mode },		\
	.show	= _show,						\
	.store	= _store,						\
	.u = {								\
		.offset = offsetof(struct ext4_super_block, _elname),	\
	},								\
}

T
Theodore Ts'o 已提交
2646 2647 2648
#define EXT4_ATTR(name, mode, show, store) \
static struct ext4_attr ext4_attr_##name = __ATTR(name, mode, show, store)

2649
#define EXT4_INFO_ATTR(name) EXT4_ATTR(name, 0444, NULL, NULL)
T
Theodore Ts'o 已提交
2650 2651
#define EXT4_RO_ATTR(name) EXT4_ATTR(name, 0444, name##_show, NULL)
#define EXT4_RW_ATTR(name) EXT4_ATTR(name, 0644, name##_show, name##_store)
2652 2653 2654

#define EXT4_RO_ATTR_ES_UI(name, elname)	\
	EXT4_ATTR_OFFSET_ES(name, 0444, es_ui_show, NULL, elname)
T
Theodore Ts'o 已提交
2655 2656
#define EXT4_RW_ATTR_SBI_UI(name, elname)	\
	EXT4_ATTR_OFFSET(name, 0644, sbi_ui_show, sbi_ui_store, elname)
2657

T
Theodore Ts'o 已提交
2658
#define ATTR_LIST(name) &ext4_attr_##name.attr
2659 2660 2661 2662 2663 2664 2665 2666
#define EXT4_DEPRECATED_ATTR(_name, _val)	\
static struct ext4_attr ext4_attr_##_name = {			\
	.attr = {.name = __stringify(_name), .mode = 0444 },	\
	.show	= sbi_deprecated_show,				\
	.u = {							\
		.deprecated_val = _val,				\
	},							\
}
T
Theodore Ts'o 已提交
2667 2668 2669 2670

EXT4_RO_ATTR(delayed_allocation_blocks);
EXT4_RO_ATTR(session_write_kbytes);
EXT4_RO_ATTR(lifetime_write_kbytes);
L
Lukas Czerner 已提交
2671
EXT4_RW_ATTR(reserved_clusters);
T
Theodore Ts'o 已提交
2672 2673
EXT4_ATTR_OFFSET(inode_readahead_blks, 0644, sbi_ui_show,
		 inode_readahead_blks_store, s_inode_readahead_blks);
2674
EXT4_RW_ATTR_SBI_UI(inode_goal, s_inode_goal);
T
Theodore Ts'o 已提交
2675 2676 2677 2678 2679 2680
EXT4_RW_ATTR_SBI_UI(mb_stats, s_mb_stats);
EXT4_RW_ATTR_SBI_UI(mb_max_to_scan, s_mb_max_to_scan);
EXT4_RW_ATTR_SBI_UI(mb_min_to_scan, s_mb_min_to_scan);
EXT4_RW_ATTR_SBI_UI(mb_order2_req, s_mb_order2_reqs);
EXT4_RW_ATTR_SBI_UI(mb_stream_req, s_mb_stream_request);
EXT4_RW_ATTR_SBI_UI(mb_group_prealloc, s_mb_group_prealloc);
2681
EXT4_DEPRECATED_ATTR(max_writeback_mb_bump, 128);
2682
EXT4_RW_ATTR_SBI_UI(extent_max_zeroout_kb, s_extent_max_zeroout_kb);
2683
EXT4_ATTR(trigger_fs_error, 0200, NULL, trigger_test_error);
2684 2685 2686 2687 2688 2689
EXT4_RW_ATTR_SBI_UI(err_ratelimit_interval_ms, s_err_ratelimit_state.interval);
EXT4_RW_ATTR_SBI_UI(err_ratelimit_burst, s_err_ratelimit_state.burst);
EXT4_RW_ATTR_SBI_UI(warning_ratelimit_interval_ms, s_warning_ratelimit_state.interval);
EXT4_RW_ATTR_SBI_UI(warning_ratelimit_burst, s_warning_ratelimit_state.burst);
EXT4_RW_ATTR_SBI_UI(msg_ratelimit_interval_ms, s_msg_ratelimit_state.interval);
EXT4_RW_ATTR_SBI_UI(msg_ratelimit_burst, s_msg_ratelimit_state.burst);
2690 2691 2692
EXT4_RO_ATTR_ES_UI(errors_count, s_error_count);
EXT4_RO_ATTR_ES_UI(first_error_time, s_first_error_time);
EXT4_RO_ATTR_ES_UI(last_error_time, s_last_error_time);
T
Theodore Ts'o 已提交
2693 2694 2695 2696 2697

static struct attribute *ext4_attrs[] = {
	ATTR_LIST(delayed_allocation_blocks),
	ATTR_LIST(session_write_kbytes),
	ATTR_LIST(lifetime_write_kbytes),
L
Lukas Czerner 已提交
2698
	ATTR_LIST(reserved_clusters),
T
Theodore Ts'o 已提交
2699
	ATTR_LIST(inode_readahead_blks),
2700
	ATTR_LIST(inode_goal),
T
Theodore Ts'o 已提交
2701 2702 2703 2704 2705 2706
	ATTR_LIST(mb_stats),
	ATTR_LIST(mb_max_to_scan),
	ATTR_LIST(mb_min_to_scan),
	ATTR_LIST(mb_order2_req),
	ATTR_LIST(mb_stream_req),
	ATTR_LIST(mb_group_prealloc),
2707
	ATTR_LIST(max_writeback_mb_bump),
2708
	ATTR_LIST(extent_max_zeroout_kb),
2709
	ATTR_LIST(trigger_fs_error),
2710 2711 2712 2713 2714 2715
	ATTR_LIST(err_ratelimit_interval_ms),
	ATTR_LIST(err_ratelimit_burst),
	ATTR_LIST(warning_ratelimit_interval_ms),
	ATTR_LIST(warning_ratelimit_burst),
	ATTR_LIST(msg_ratelimit_interval_ms),
	ATTR_LIST(msg_ratelimit_burst),
2716 2717 2718
	ATTR_LIST(errors_count),
	ATTR_LIST(first_error_time),
	ATTR_LIST(last_error_time),
T
Theodore Ts'o 已提交
2719 2720 2721
	NULL,
};

2722 2723
/* Features this copy of ext4 supports */
EXT4_INFO_ATTR(lazy_itable_init);
2724
EXT4_INFO_ATTR(batched_discard);
2725
EXT4_INFO_ATTR(meta_bg_resize);
2726
EXT4_INFO_ATTR(encryption);
2727 2728 2729

static struct attribute *ext4_feat_attrs[] = {
	ATTR_LIST(lazy_itable_init),
2730
	ATTR_LIST(batched_discard),
2731
	ATTR_LIST(meta_bg_resize),
2732
	ATTR_LIST(encryption),
2733 2734 2735
	NULL,
};

T
Theodore Ts'o 已提交
2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763
static ssize_t ext4_attr_show(struct kobject *kobj,
			      struct attribute *attr, char *buf)
{
	struct ext4_sb_info *sbi = container_of(kobj, struct ext4_sb_info,
						s_kobj);
	struct ext4_attr *a = container_of(attr, struct ext4_attr, attr);

	return a->show ? a->show(a, sbi, buf) : 0;
}

static ssize_t ext4_attr_store(struct kobject *kobj,
			       struct attribute *attr,
			       const char *buf, size_t len)
{
	struct ext4_sb_info *sbi = container_of(kobj, struct ext4_sb_info,
						s_kobj);
	struct ext4_attr *a = container_of(attr, struct ext4_attr, attr);

	return a->store ? a->store(a, sbi, buf, len) : 0;
}

static void ext4_sb_release(struct kobject *kobj)
{
	struct ext4_sb_info *sbi = container_of(kobj, struct ext4_sb_info,
						s_kobj);
	complete(&sbi->s_kobj_unregister);
}

2764
static const struct sysfs_ops ext4_attr_ops = {
T
Theodore Ts'o 已提交
2765 2766 2767 2768 2769 2770 2771 2772 2773 2774
	.show	= ext4_attr_show,
	.store	= ext4_attr_store,
};

static struct kobj_type ext4_ktype = {
	.default_attrs	= ext4_attrs,
	.sysfs_ops	= &ext4_attr_ops,
	.release	= ext4_sb_release,
};

2775 2776 2777 2778 2779
static void ext4_feat_release(struct kobject *kobj)
{
	complete(&ext4_feat->f_kobj_unregister);
}

2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795
static ssize_t ext4_feat_show(struct kobject *kobj,
			      struct attribute *attr, char *buf)
{
	return snprintf(buf, PAGE_SIZE, "supported\n");
}

/*
 * We can not use ext4_attr_show/store because it relies on the kobject
 * being embedded in the ext4_sb_info structure which is definitely not
 * true in this case.
 */
static const struct sysfs_ops ext4_feat_ops = {
	.show	= ext4_feat_show,
	.store	= NULL,
};

2796 2797
static struct kobj_type ext4_feat_ktype = {
	.default_attrs	= ext4_feat_attrs,
2798
	.sysfs_ops	= &ext4_feat_ops,
2799 2800 2801
	.release	= ext4_feat_release,
};

2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821
/*
 * 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)
{
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, ~EXT4_FEATURE_INCOMPAT_SUPP)) {
		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;

D
Darrick J. Wong 已提交
2822 2823 2824 2825 2826 2827
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_READONLY)) {
		ext4_msg(sb, KERN_INFO, "filesystem is read-only");
		sb->s_flags |= MS_RDONLY;
		return 1;
	}

2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847
	/* Check that feature set is OK for a read-write mount */
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, ~EXT4_FEATURE_RO_COMPAT_SUPP)) {
		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
	 */
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_HUGE_FILE)) {
		if (sizeof(blkcnt_t) < sizeof(u64)) {
			ext4_msg(sb, KERN_ERR, "Filesystem with huge files "
				 "cannot be mounted RDWR without "
				 "CONFIG_LBDAF");
			return 0;
		}
	}
2848 2849 2850 2851 2852 2853 2854
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_BIGALLOC) &&
	    !EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS)) {
		ext4_msg(sb, KERN_ERR,
			 "Can't support bigalloc feature without "
			 "extents feature\n");
		return 0;
	}
2855 2856 2857 2858 2859 2860 2861 2862 2863 2864

#ifndef CONFIG_QUOTA
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_QUOTA) &&
	    !readonly) {
		ext4_msg(sb, KERN_ERR,
			 "Filesystem with quota feature cannot be mounted RDWR "
			 "without CONFIG_QUOTA");
		return 0;
	}
#endif  /* CONFIG_QUOTA */
2865 2866 2867
	return 1;
}

2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881
/*
 * 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)
2882 2883
		/* fsck newer than v1.41.13 is needed to clean this condition. */
		ext4_msg(sb, KERN_NOTICE, "error count since last fsck: %u",
2884 2885
			 le32_to_cpu(es->s_error_count));
	if (es->s_first_error_time) {
2886
		printk(KERN_NOTICE "EXT4-fs (%s): initial error at time %u: %.*s:%d",
2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899
		       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) {
2900
		printk(KERN_NOTICE "EXT4-fs (%s): last error at time %u: %.*s:%d",
2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915
		       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 */
}

2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927
/* 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;

2928
	sb_start_write(sb);
2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939
	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;
	}

2940
	if (group >= ngroups)
2941 2942 2943 2944 2945 2946 2947
		ret = 1;

	if (!ret) {
		timeout = jiffies;
		ret = ext4_init_inode_table(sb, group,
					    elr->lr_timeout ? 0 : 1);
		if (elr->lr_timeout == 0) {
2948 2949
			timeout = (jiffies - timeout) *
				  elr->lr_sbi->s_li_wait_mult;
2950 2951 2952 2953 2954
			elr->lr_timeout = timeout;
		}
		elr->lr_next_sched = jiffies + elr->lr_timeout;
		elr->lr_next_group = group + 1;
	}
2955
	sb_end_write(sb);
2956 2957 2958 2959 2960 2961

	return ret;
}

/*
 * Remove lr_request from the list_request and free the
2962
 * request structure. Should be called with li_list_mtx held
2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979
 */
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)
{
2980 2981 2982
	mutex_lock(&ext4_li_mtx);
	if (!ext4_li_info) {
		mutex_unlock(&ext4_li_mtx);
2983
		return;
2984
	}
2985 2986

	mutex_lock(&ext4_li_info->li_list_mtx);
2987
	ext4_remove_li_request(EXT4_SB(sb)->s_li_request);
2988
	mutex_unlock(&ext4_li_info->li_list_mtx);
2989
	mutex_unlock(&ext4_li_mtx);
2990 2991
}

2992 2993
static struct task_struct *ext4_lazyinit_task;

2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007
/*
 * 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;
3008
	unsigned long next_wakeup, cur;
3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025

	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);

3026 3027 3028 3029 3030 3031
			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;
				}
3032 3033 3034 3035 3036 3037 3038
			}

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

3039
		try_to_freeze();
3040

3041 3042
		cur = jiffies;
		if ((time_after_eq(cur, next_wakeup)) ||
3043
		    (MAX_JIFFY_OFFSET == next_wakeup)) {
3044 3045 3046 3047
			cond_resched();
			continue;
		}

3048 3049
		schedule_timeout_interruptible(next_wakeup - cur);

3050 3051 3052 3053
		if (kthread_should_stop()) {
			ext4_clear_request_list();
			goto exit_thread;
		}
3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095
	}

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)
{
3096 3097 3098 3099
	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);
3100 3101 3102
		ext4_clear_request_list();
		kfree(ext4_li_info);
		ext4_li_info = NULL;
3103
		printk(KERN_CRIT "EXT4-fs: error %d creating inode table "
3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171
				 "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.
	 */
3172 3173
	elr->lr_next_sched = jiffies + (prandom_u32() %
				(EXT4_DEF_LI_MAX_START_DELAY * HZ));
3174 3175 3176
	return elr;
}

3177 3178
int ext4_register_li_request(struct super_block *sb,
			     ext4_group_t first_not_zeroed)
3179 3180
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
3181
	struct ext4_li_request *elr = NULL;
3182
	ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count;
3183
	int ret = 0;
3184

3185
	mutex_lock(&ext4_li_mtx);
3186 3187 3188 3189 3190 3191
	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;
3192
		goto out;
3193
	}
3194 3195 3196

	if (first_not_zeroed == ngroups ||
	    (sb->s_flags & MS_RDONLY) ||
3197
	    !test_opt(sb, INIT_INODE_TABLE))
3198
		goto out;
3199 3200

	elr = ext4_li_request_new(sb, first_not_zeroed);
3201 3202 3203 3204
	if (!elr) {
		ret = -ENOMEM;
		goto out;
	}
3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216

	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;
3217 3218 3219 3220 3221 3222
	/*
	 * 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;
3223 3224 3225 3226 3227 3228 3229

	if (!(ext4_li_info->li_state & EXT4_LAZYINIT_RUNNING)) {
		ret = ext4_run_lazyinit_thread();
		if (ret)
			goto out;
	}
out:
3230 3231
	mutex_unlock(&ext4_li_mtx);
	if (ret)
3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245
		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.
	 */
3246
	if (!ext4_li_info || !ext4_lazyinit_task)
3247 3248
		return;

3249
	kthread_stop(ext4_lazyinit_task);
3250 3251
}

3252 3253 3254 3255 3256 3257
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);

3258
	if (ext4_has_metadata_csum(sb)) {
3259
		/* journal checksum v3 */
3260
		compat = 0;
3261
		incompat = JBD2_FEATURE_INCOMPAT_CSUM_V3;
3262 3263 3264 3265 3266 3267
	} else {
		/* journal checksum v1 */
		compat = JBD2_FEATURE_COMPAT_CHECKSUM;
		incompat = 0;
	}

3268 3269 3270 3271
	jbd2_journal_clear_features(sbi->s_journal,
			JBD2_FEATURE_COMPAT_CHECKSUM, 0,
			JBD2_FEATURE_INCOMPAT_CSUM_V3 |
			JBD2_FEATURE_INCOMPAT_CSUM_V2);
3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283
	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 {
3284 3285
		jbd2_journal_clear_features(sbi->s_journal, 0, 0,
				JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT);
3286 3287 3288 3289 3290
	}

	return ret;
}

3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314
/*
 * 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;

3315 3316 3317 3318
	if (!EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_BIGALLOC))
		return (ext4_bg_has_super(sb, grp) + ext4_bg_num_gdb(sb, grp) +
			sbi->s_itb_per_group + 2);

3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365 3366 3367
	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;
3368
	char *buf = (char *) get_zeroed_page(GFP_NOFS);
3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395

	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();
	}
3396 3397
	/* Add the internal journal blocks as well */
	if (sbi->s_journal && !sbi->journal_bdev)
3398
		overhead += EXT4_NUM_B2C(sbi, sbi->s_journal->j_maxlen);
3399

3400 3401 3402 3403 3404 3405
	sbi->s_overhead = overhead;
	smp_wmb();
	free_page((unsigned long) buf);
	return 0;
}

L
Lukas Czerner 已提交
3406

3407
static ext4_fsblk_t ext4_calculate_resv_clusters(struct super_block *sb)
L
Lukas Czerner 已提交
3408 3409 3410
{
	ext4_fsblk_t resv_clusters;

3411 3412 3413 3414 3415 3416 3417 3418
	/*
	 * 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.
	 */
	if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS))
		return 0;
L
Lukas Czerner 已提交
3419 3420 3421 3422
	/*
	 * 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
3423
	 * unwritten extents in delalloc path. In most cases such
L
Lukas Czerner 已提交
3424 3425 3426
	 * allocation would require 1, or 2 blocks, higher numbers are
	 * very rare.
	 */
3427 3428
	resv_clusters = ext4_blocks_count(EXT4_SB(sb)->s_es) >>
			EXT4_SB(sb)->s_cluster_bits;
L
Lukas Czerner 已提交
3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448

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

	return resv_clusters;
}


static int ext4_reserve_clusters(struct ext4_sb_info *sbi, ext4_fsblk_t count)
{
	ext4_fsblk_t clusters = ext4_blocks_count(sbi->s_es) >>
				sbi->s_cluster_bits;

	if (count >= clusters)
		return -EINVAL;

	atomic64_set(&sbi->s_resv_clusters, count);
	return 0;
}

3449
static int ext4_fill_super(struct super_block *sb, void *data, int silent)
3450
{
3451
	char *orig_data = kstrdup(data, GFP_KERNEL);
3452
	struct buffer_head *bh;
3453 3454 3455 3456
	struct ext4_super_block *es = NULL;
	struct ext4_sb_info *sbi;
	ext4_fsblk_t block;
	ext4_fsblk_t sb_block = get_sb_block(&data);
3457
	ext4_fsblk_t logical_sb_block;
3458 3459 3460 3461
	unsigned long offset = 0;
	unsigned long journal_devnum = 0;
	unsigned long def_mount_opts;
	struct inode *root;
3462
	const char *descr;
3463
	int ret = -ENOMEM;
3464
	int blocksize, clustersize;
3465 3466
	unsigned int db_count;
	unsigned int i;
3467
	int needs_recovery, has_huge_files, has_bigalloc;
L
Laurent Vivier 已提交
3468
	__u64 blocks_count;
3469
	int err = 0;
3470
	unsigned int journal_ioprio = DEFAULT_JOURNAL_IOPRIO;
3471
	ext4_group_t first_not_zeroed;
3472 3473 3474

	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
	if (!sbi)
3475
		goto out_free_orig;
3476 3477 3478 3479 3480

	sbi->s_blockgroup_lock =
		kzalloc(sizeof(struct blockgroup_lock), GFP_KERNEL);
	if (!sbi->s_blockgroup_lock) {
		kfree(sbi);
3481
		goto out_free_orig;
3482
	}
3483
	sb->s_fs_info = sbi;
3484
	sbi->s_sb = sb;
3485
	sbi->s_inode_readahead_blks = EXT4_DEF_INODE_READAHEAD_BLKS;
M
Miklos Szeredi 已提交
3486
	sbi->s_sb_block = sb_block;
3487 3488 3489
	if (sb->s_bdev->bd_part)
		sbi->s_sectors_written_start =
			part_stat_read(sb->s_bdev->bd_part, sectors[1]);
3490

3491
	/* Cleanup superblock name */
3492
	strreplace(sb->s_id, '/', '!');
3493

3494
	/* -EINVAL is default */
3495
	ret = -EINVAL;
3496
	blocksize = sb_min_blocksize(sb, EXT4_MIN_BLOCK_SIZE);
3497
	if (!blocksize) {
3498
		ext4_msg(sb, KERN_ERR, "unable to set blocksize");
3499 3500 3501 3502
		goto out_fail;
	}

	/*
3503
	 * The ext4 superblock will not be buffer aligned for other than 1kB
3504 3505
	 * block sizes.  We need to calculate the offset from buffer start.
	 */
3506
	if (blocksize != EXT4_MIN_BLOCK_SIZE) {
3507 3508
		logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE;
		offset = do_div(logical_sb_block, blocksize);
3509
	} else {
3510
		logical_sb_block = sb_block;
3511 3512
	}

3513
	if (!(bh = sb_bread_unmovable(sb, logical_sb_block))) {
3514
		ext4_msg(sb, KERN_ERR, "unable to read superblock");
3515 3516 3517 3518
		goto out_fail;
	}
	/*
	 * Note: s_es must be initialized as soon as possible because
3519
	 *       some ext4 macro-instructions depend on its value
3520
	 */
3521
	es = (struct ext4_super_block *) (bh->b_data + offset);
3522 3523
	sbi->s_es = es;
	sb->s_magic = le16_to_cpu(es->s_magic);
3524 3525
	if (sb->s_magic != EXT4_SUPER_MAGIC)
		goto cantfind_ext4;
T
Theodore Ts'o 已提交
3526
	sbi->s_kbytes_written = le64_to_cpu(es->s_kbytes_written);
3527

3528 3529 3530 3531
	/* Warn if metadata_csum and gdt_csum are both set. */
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
				       EXT4_FEATURE_RO_COMPAT_METADATA_CSUM) &&
	    EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_GDT_CSUM))
3532
		ext4_warning(sb, "metadata_csum and uninit_bg are "
3533 3534
			     "redundant flags; please run fsck.");

3535 3536 3537 3538 3539 3540 3541 3542
	/* 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;
	}

3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554
	/* Load the checksum driver */
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
				       EXT4_FEATURE_RO_COMPAT_METADATA_CSUM)) {
		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;
		}
	}

3555 3556 3557 3558 3559 3560 3561 3562 3563
	/* 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;
		goto cantfind_ext4;
	}

	/* Precompute checksum seed for all metadata */
3564
	if (ext4_has_metadata_csum(sb))
3565 3566 3567
		sbi->s_csum_seed = ext4_chksum(sbi, ~0, es->s_uuid,
					       sizeof(es->s_uuid));

3568 3569
	/* Set defaults before we parse the mount options */
	def_mount_opts = le32_to_cpu(es->s_default_mount_opts);
3570
	set_opt(sb, INIT_INODE_TABLE);
3571
	if (def_mount_opts & EXT4_DEFM_DEBUG)
3572
		set_opt(sb, DEBUG);
3573
	if (def_mount_opts & EXT4_DEFM_BSDGROUPS)
3574
		set_opt(sb, GRPID);
3575
	if (def_mount_opts & EXT4_DEFM_UID16)
3576
		set_opt(sb, NO_UID32);
3577 3578
	/* xattr user namespace & acls are now defaulted on */
	set_opt(sb, XATTR_USER);
T
Theodore Ts'o 已提交
3579
#ifdef CONFIG_EXT4_FS_POSIX_ACL
3580
	set_opt(sb, POSIX_ACL);
3581
#endif
3582 3583 3584 3585
	/* don't forget to enable journal_csum when metadata_csum is enabled. */
	if (ext4_has_metadata_csum(sb))
		set_opt(sb, JOURNAL_CHECKSUM);

3586
	if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_DATA)
3587
		set_opt(sb, JOURNAL_DATA);
3588
	else if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_ORDERED)
3589
		set_opt(sb, ORDERED_DATA);
3590
	else if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_WBACK)
3591
		set_opt(sb, WRITEBACK_DATA);
3592 3593

	if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_PANIC)
3594
		set_opt(sb, ERRORS_PANIC);
3595
	else if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_CONTINUE)
3596
		set_opt(sb, ERRORS_CONT);
3597
	else
3598
		set_opt(sb, ERRORS_RO);
3599 3600
	/* block_validity enabled by default; disable with noblock_validity */
	set_opt(sb, BLOCK_VALIDITY);
3601
	if (def_mount_opts & EXT4_DEFM_DISCARD)
3602
		set_opt(sb, DISCARD);
3603

3604 3605
	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));
3606 3607 3608
	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;
3609

3610
	if ((def_mount_opts & EXT4_DEFM_NOBARRIER) == 0)
3611
		set_opt(sb, BARRIER);
3612

3613 3614 3615 3616
	/*
	 * enable delayed allocation by default
	 * Use -o nodelalloc to turn it off
	 */
3617
	if (!IS_EXT3_SB(sb) && !IS_EXT2_SB(sb) &&
3618
	    ((def_mount_opts & EXT4_DEFM_NODELALLOC) == 0))
3619
		set_opt(sb, DELALLOC);
3620

3621 3622 3623 3624 3625 3626
	/*
	 * 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;

3627
	if (!parse_options((char *) sbi->s_es->s_mount_opts, sb,
3628
			   &journal_devnum, &journal_ioprio, 0)) {
3629 3630 3631 3632
		ext4_msg(sb, KERN_WARNING,
			 "failed to parse options in superblock: %s",
			 sbi->s_es->s_mount_opts);
	}
3633
	sbi->s_def_mount_opt = sbi->s_mount_opt;
3634
	if (!parse_options((char *) data, sb, &journal_devnum,
3635
			   &journal_ioprio, 0))
3636 3637
		goto failed_mount;

3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648
	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 "
3649
				 "both data=journal and dioread_nolock");
3650 3651
			goto failed_mount;
		}
R
Ross Zwisler 已提交
3652 3653 3654 3655 3656
		if (test_opt(sb, DAX)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "both data=journal and dax");
			goto failed_mount;
		}
3657 3658
		if (test_opt(sb, DELALLOC))
			clear_opt(sb, DELALLOC);
3659 3660
	} else {
		sb->s_iflags |= SB_I_CGROUPWB;
3661 3662
	}

3663
	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
3664
		(test_opt(sb, POSIX_ACL) ? MS_POSIXACL : 0);
3665

3666 3667 3668 3669
	if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV &&
	    (EXT4_HAS_COMPAT_FEATURE(sb, ~0U) ||
	     EXT4_HAS_RO_COMPAT_FEATURE(sb, ~0U) ||
	     EXT4_HAS_INCOMPAT_FEATURE(sb, ~0U)))
3670 3671 3672
		ext4_msg(sb, KERN_WARNING,
		       "feature flags set on rev 0 fs, "
		       "running e2fsck is recommended");
3673

3674 3675 3676 3677 3678 3679 3680 3681 3682 3683
	if (es->s_creator_os == cpu_to_le32(EXT4_OS_HURD)) {
		set_opt2(sb, HURD_COMPAT);
		if (EXT4_HAS_INCOMPAT_FEATURE(sb,
					      EXT4_FEATURE_INCOMPAT_64BIT)) {
			ext4_msg(sb, KERN_ERR,
				 "The Hurd can't support 64-bit file systems");
			goto failed_mount;
		}
	}

3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705
	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;
		}
	}

3706 3707 3708 3709 3710
	/*
	 * 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.
	 */
3711
	if (!ext4_feature_set_ok(sb, (sb->s_flags & MS_RDONLY)))
3712
		goto failed_mount;
3713

J
Jan Kara 已提交
3714
	blocksize = BLOCK_SIZE << le32_to_cpu(es->s_log_block_size);
3715 3716
	if (blocksize < EXT4_MIN_BLOCK_SIZE ||
	    blocksize > EXT4_MAX_BLOCK_SIZE) {
3717 3718
		ext4_msg(sb, KERN_ERR,
		       "Unsupported filesystem blocksize %d", blocksize);
3719 3720 3721
		goto failed_mount;
	}

R
Ross Zwisler 已提交
3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734
	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;
		}
	}

3735 3736 3737 3738 3739 3740 3741
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_ENCRYPT) &&
	    es->s_encryption_level) {
		ext4_msg(sb, KERN_ERR, "Unsupported encryption level %d",
			 es->s_encryption_level);
		goto failed_mount;
	}

3742
	if (sb->s_blocksize != blocksize) {
3743 3744
		/* Validate the filesystem blocksize */
		if (!sb_set_blocksize(sb, blocksize)) {
3745
			ext4_msg(sb, KERN_ERR, "bad block size %d",
3746
					blocksize);
3747 3748 3749
			goto failed_mount;
		}

3750
		brelse(bh);
3751 3752
		logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE;
		offset = do_div(logical_sb_block, blocksize);
3753
		bh = sb_bread_unmovable(sb, logical_sb_block);
3754
		if (!bh) {
3755 3756
			ext4_msg(sb, KERN_ERR,
			       "Can't read superblock on 2nd try");
3757 3758
			goto failed_mount;
		}
3759
		es = (struct ext4_super_block *)(bh->b_data + offset);
3760
		sbi->s_es = es;
3761
		if (es->s_magic != cpu_to_le16(EXT4_SUPER_MAGIC)) {
3762 3763
			ext4_msg(sb, KERN_ERR,
			       "Magic mismatch, very weird!");
3764 3765 3766 3767
			goto failed_mount;
		}
	}

3768 3769
	has_huge_files = EXT4_HAS_RO_COMPAT_FEATURE(sb,
				EXT4_FEATURE_RO_COMPAT_HUGE_FILE);
3770 3771 3772
	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);
3773

3774 3775 3776
	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;
3777 3778 3779
	} else {
		sbi->s_inode_size = le16_to_cpu(es->s_inode_size);
		sbi->s_first_ino = le32_to_cpu(es->s_first_ino);
3780
		if ((sbi->s_inode_size < EXT4_GOOD_OLD_INODE_SIZE) ||
V
Vignesh Babu 已提交
3781
		    (!is_power_of_2(sbi->s_inode_size)) ||
3782
		    (sbi->s_inode_size > blocksize)) {
3783 3784
			ext4_msg(sb, KERN_ERR,
			       "unsupported inode size: %d",
3785
			       sbi->s_inode_size);
3786 3787
			goto failed_mount;
		}
K
Kalpak Shah 已提交
3788 3789
		if (sbi->s_inode_size > EXT4_GOOD_OLD_INODE_SIZE)
			sb->s_time_gran = 1 << (EXT4_EPOCH_BITS - 2);
3790
	}
3791

3792 3793
	sbi->s_desc_size = le16_to_cpu(es->s_desc_size);
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_64BIT)) {
3794
		if (sbi->s_desc_size < EXT4_MIN_DESC_SIZE_64BIT ||
3795
		    sbi->s_desc_size > EXT4_MAX_DESC_SIZE ||
V
vignesh babu 已提交
3796
		    !is_power_of_2(sbi->s_desc_size)) {
3797 3798
			ext4_msg(sb, KERN_ERR,
			       "unsupported descriptor size %lu",
3799 3800 3801 3802 3803
			       sbi->s_desc_size);
			goto failed_mount;
		}
	} else
		sbi->s_desc_size = EXT4_MIN_DESC_SIZE;
3804

3805 3806
	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);
3807
	if (EXT4_INODE_SIZE(sb) == 0 || EXT4_INODES_PER_GROUP(sb) == 0)
3808
		goto cantfind_ext4;
3809

3810
	sbi->s_inodes_per_block = blocksize / EXT4_INODE_SIZE(sb);
3811
	if (sbi->s_inodes_per_block == 0)
3812
		goto cantfind_ext4;
3813 3814
	sbi->s_itb_per_group = sbi->s_inodes_per_group /
					sbi->s_inodes_per_block;
3815
	sbi->s_desc_per_block = blocksize / EXT4_DESC_SIZE(sb);
3816 3817
	sbi->s_sbh = bh;
	sbi->s_mount_state = le16_to_cpu(es->s_state);
3818 3819
	sbi->s_addr_per_block_bits = ilog2(EXT4_ADDR_PER_BLOCK(sb));
	sbi->s_desc_per_block_bits = ilog2(EXT4_DESC_PER_BLOCK(sb));
3820

3821
	for (i = 0; i < 4; i++)
3822 3823
		sbi->s_hash_seed[i] = le32_to_cpu(es->s_hash_seed[i]);
	sbi->s_def_hash_version = es->s_def_hash_version;
3824 3825 3826 3827 3828
	if (EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_DIR_INDEX)) {
		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) {
3829
#ifdef __CHAR_UNSIGNED__
3830 3831 3832 3833
			if (!(sb->s_flags & MS_RDONLY))
				es->s_flags |=
					cpu_to_le32(EXT2_FLAGS_UNSIGNED_HASH);
			sbi->s_hash_unsigned = 3;
3834
#else
3835 3836 3837
			if (!(sb->s_flags & MS_RDONLY))
				es->s_flags |=
					cpu_to_le32(EXT2_FLAGS_SIGNED_HASH);
3838
#endif
3839
		}
3840
	}
3841

3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885
	/* Handle clustersize */
	clustersize = BLOCK_SIZE << le32_to_cpu(es->s_log_cluster_size);
	has_bigalloc = EXT4_HAS_RO_COMPAT_FEATURE(sb,
				EXT4_FEATURE_RO_COMPAT_BIGALLOC);
	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;
3886
	}
3887 3888
	sbi->s_cluster_ratio = clustersize / blocksize;

3889
	if (sbi->s_inodes_per_group > blocksize * 8) {
3890 3891
		ext4_msg(sb, KERN_ERR,
		       "#inodes per group too big: %lu",
3892
		       sbi->s_inodes_per_group);
3893 3894 3895
		goto failed_mount;
	}

3896 3897 3898 3899
	/* Do we have standard group size of clustersize * 8 blocks ? */
	if (sbi->s_blocks_per_group == clustersize << 3)
		set_opt2(sb, STD_GROUP_SIZE);

3900 3901 3902 3903
	/*
	 * Test whether we have more sectors than will fit in sector_t,
	 * and whether the max offset is addressable by the page cache.
	 */
3904
	err = generic_check_addressable(sb->s_blocksize_bits,
3905
					ext4_blocks_count(es));
3906
	if (err) {
3907
		ext4_msg(sb, KERN_ERR, "filesystem"
3908
			 " too large to mount safely on this system");
3909
		if (sizeof(sector_t) < 8)
3910
			ext4_msg(sb, KERN_WARNING, "CONFIG_LBDAF not enabled");
3911 3912 3913
		goto failed_mount;
	}

3914 3915
	if (EXT4_BLOCKS_PER_GROUP(sb) == 0)
		goto cantfind_ext4;
3916

3917 3918 3919
	/* 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) {
3920 3921
		ext4_msg(sb, KERN_WARNING, "bad geometry: block count %llu "
		       "exceeds size of device (%llu blocks)",
3922 3923 3924 3925
		       ext4_blocks_count(es), blocks_count);
		goto failed_mount;
	}

3926 3927 3928 3929 3930
	/*
	 * 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)) {
3931
		ext4_msg(sb, KERN_WARNING, "bad geometry: first data "
3932 3933 3934
			 "block %u is beyond end of filesystem (%llu)",
			 le32_to_cpu(es->s_first_data_block),
			 ext4_blocks_count(es));
3935 3936
		goto failed_mount;
	}
L
Laurent Vivier 已提交
3937 3938 3939 3940
	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));
3941
	if (blocks_count > ((uint64_t)1<<32) - EXT4_DESC_PER_BLOCK(sb)) {
3942
		ext4_msg(sb, KERN_WARNING, "groups count too large: %u "
3943
		       "(block count %llu, first data block %u, "
3944
		       "blocks per group %lu)", sbi->s_groups_count,
3945 3946 3947 3948 3949
		       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
3950
	sbi->s_groups_count = blocks_count;
3951 3952
	sbi->s_blockfile_groups = min_t(ext4_group_t, sbi->s_groups_count,
			(EXT4_MAX_BLOCK_FILE_PHYS / EXT4_BLOCKS_PER_GROUP(sb)));
3953 3954
	db_count = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) /
		   EXT4_DESC_PER_BLOCK(sb);
3955 3956 3957
	sbi->s_group_desc = ext4_kvmalloc(db_count *
					  sizeof(struct buffer_head *),
					  GFP_KERNEL);
3958
	if (sbi->s_group_desc == NULL) {
3959
		ext4_msg(sb, KERN_ERR, "not enough memory");
3960
		ret = -ENOMEM;
3961 3962 3963
		goto failed_mount;
	}

3964 3965
	if (ext4_proc_root)
		sbi->s_proc = proc_mkdir(sb->s_id, ext4_proc_root);
3966

3967 3968 3969 3970
	if (sbi->s_proc)
		proc_create_data("options", S_IRUGO, sbi->s_proc,
				 &ext4_seq_options_fops, sb);

3971
	bgl_lock_init(sbi->s_blockgroup_lock);
3972 3973

	for (i = 0; i < db_count; i++) {
3974
		block = descriptor_loc(sb, logical_sb_block, i);
3975
		sbi->s_group_desc[i] = sb_bread_unmovable(sb, block);
3976
		if (!sbi->s_group_desc[i]) {
3977 3978
			ext4_msg(sb, KERN_ERR,
			       "can't read group descriptor %d", i);
3979 3980 3981 3982
			db_count = i;
			goto failed_mount2;
		}
	}
3983
	if (!ext4_check_descriptors(sb, &first_not_zeroed)) {
3984
		ext4_msg(sb, KERN_ERR, "group descriptors corrupted!");
3985
		goto failed_mount2;
3986
	}
3987

3988
	sbi->s_gdb_count = db_count;
3989 3990 3991
	get_random_bytes(&sbi->s_next_generation, sizeof(u32));
	spin_lock_init(&sbi->s_next_gen_lock);

3992 3993
	setup_timer(&sbi->s_err_report, print_daily_error_info,
		(unsigned long) sb);
3994

3995
	/* Register extent status tree shrinker */
3996
	if (ext4_es_register_shrinker(sbi))
3997 3998
		goto failed_mount3;

3999
	sbi->s_stripe = ext4_get_stripe_size(sbi);
4000
	sbi->s_extent_max_zeroout_kb = 32;
4001

4002 4003 4004
	/*
	 * set up enough so that it can read an inode
	 */
4005
	sb->s_op = &ext4_sops;
4006 4007
	sb->s_export_op = &ext4_export_ops;
	sb->s_xattr = ext4_xattr_handlers;
4008
#ifdef CONFIG_QUOTA
4009
	sb->dq_op = &ext4_quota_operations;
4010
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_QUOTA))
4011
		sb->s_qcop = &dquot_quotactl_sysfile_ops;
4012 4013
	else
		sb->s_qcop = &ext4_qctl_operations;
J
Jan Kara 已提交
4014
	sb->s_quota_types = QTYPE_MASK_USR | QTYPE_MASK_GRP;
4015
#endif
4016 4017
	memcpy(sb->s_uuid, es->s_uuid, sizeof(es->s_uuid));

4018
	INIT_LIST_HEAD(&sbi->s_orphan); /* unlinked but open files */
4019
	mutex_init(&sbi->s_orphan_lock);
4020 4021 4022 4023

	sb->s_root = NULL;

	needs_recovery = (es->s_last_orphan != 0 ||
4024 4025
			  EXT4_HAS_INCOMPAT_FEATURE(sb,
				    EXT4_FEATURE_INCOMPAT_RECOVER));
4026

4027 4028 4029
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_MMP) &&
	    !(sb->s_flags & MS_RDONLY))
		if (ext4_multi_mount_protect(sb, le64_to_cpu(es->s_mmp_block)))
4030
			goto failed_mount3a;
4031

4032 4033 4034 4035 4036
	/*
	 * The first inode we look at is the journal inode.  Don't try
	 * root first: it may be modified in the journal!
	 */
	if (!test_opt(sb, NOLOAD) &&
4037 4038
	    EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL)) {
		if (ext4_load_journal(sb, es, journal_devnum))
4039
			goto failed_mount3a;
4040 4041
	} else if (test_opt(sb, NOLOAD) && !(sb->s_flags & MS_RDONLY) &&
	      EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER)) {
4042 4043
		ext4_msg(sb, KERN_ERR, "required journal recovery "
		       "suppressed and not mounted read-only");
4044
		goto failed_mount_wq;
4045
	} else {
4046
		clear_opt(sb, DATA_FLAGS);
4047 4048 4049
		sbi->s_journal = NULL;
		needs_recovery = 0;
		goto no_journal;
4050 4051
	}

4052
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_64BIT) &&
4053 4054
	    !jbd2_journal_set_features(EXT4_SB(sb)->s_journal, 0, 0,
				       JBD2_FEATURE_INCOMPAT_64BIT)) {
4055
		ext4_msg(sb, KERN_ERR, "Failed to set 64-bit journal feature");
4056
		goto failed_mount_wq;
4057 4058
	}

4059 4060 4061 4062
	if (!set_journal_csum_feature_set(sb)) {
		ext4_msg(sb, KERN_ERR, "Failed to set journal checksum "
			 "feature set");
		goto failed_mount_wq;
4063
	}
4064

4065 4066 4067 4068 4069
	/* 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
4070 4071 4072
		 * capabilities: ORDERED_DATA if the journal can
		 * cope, else JOURNAL_DATA
		 */
4073 4074
		if (jbd2_journal_check_available_features
		    (sbi->s_journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE))
4075
			set_opt(sb, ORDERED_DATA);
4076
		else
4077
			set_opt(sb, JOURNAL_DATA);
4078 4079
		break;

4080 4081
	case EXT4_MOUNT_ORDERED_DATA:
	case EXT4_MOUNT_WRITEBACK_DATA:
4082 4083
		if (!jbd2_journal_check_available_features
		    (sbi->s_journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE)) {
4084 4085
			ext4_msg(sb, KERN_ERR, "Journal does not support "
			       "requested data journaling mode");
4086
			goto failed_mount_wq;
4087 4088 4089 4090
		}
	default:
		break;
	}
4091
	set_task_ioprio(sbi->s_journal->j_task, journal_ioprio);
4092

B
Bobi Jam 已提交
4093 4094
	sbi->s_journal->j_commit_callback = ext4_journal_commit_callback;

4095
no_journal:
4096 4097 4098 4099 4100 4101 4102 4103
	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;
		}
	}

4104 4105 4106 4107 4108 4109 4110 4111 4112
	if ((DUMMY_ENCRYPTION_ENABLED(sbi) ||
	     EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_ENCRYPT)) &&
	    (blocksize != PAGE_CACHE_SIZE)) {
		ext4_msg(sb, KERN_ERR,
			 "Unsupported blocksize for fs encryption");
		goto failed_mount_wq;
	}

	if (DUMMY_ENCRYPTION_ENABLED(sbi) &&
4113 4114 4115 4116 4117 4118
	    !(sb->s_flags & MS_RDONLY) &&
	    !EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_ENCRYPT)) {
		EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_ENCRYPT);
		ext4_commit_super(sb, 1);
	}

4119 4120 4121 4122 4123 4124 4125
	/*
	 * 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 {
4126 4127
		err = ext4_calculate_overhead(sb);
		if (err)
4128 4129 4130
			goto failed_mount_wq;
	}

T
Tejun Heo 已提交
4131 4132 4133 4134
	/*
	 * The maximum number of concurrent works can be high and
	 * concurrency isn't really necessary.  Limit it to 1.
	 */
4135 4136 4137 4138
	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");
4139
		ret = -ENOMEM;
4140 4141 4142
		goto failed_mount4;
	}

4143
	/*
4144
	 * The jbd2_journal_load will have done any necessary log recovery,
4145 4146 4147
	 * so we can safely mount the rest of the filesystem now.
	 */

4148 4149
	root = ext4_iget(sb, EXT4_ROOT_INO);
	if (IS_ERR(root)) {
4150
		ext4_msg(sb, KERN_ERR, "get root inode failed");
4151
		ret = PTR_ERR(root);
4152
		root = NULL;
4153 4154 4155
		goto failed_mount4;
	}
	if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) {
4156
		ext4_msg(sb, KERN_ERR, "corrupt root inode, run e2fsck");
A
Al Viro 已提交
4157
		iput(root);
4158 4159
		goto failed_mount4;
	}
4160
	sb->s_root = d_make_root(root);
4161
	if (!sb->s_root) {
4162
		ext4_msg(sb, KERN_ERR, "get root dentry failed");
4163 4164 4165
		ret = -ENOMEM;
		goto failed_mount4;
	}
4166

4167 4168
	if (ext4_setup_super(sb, es, sb->s_flags & MS_RDONLY))
		sb->s_flags |= MS_RDONLY;
K
Kalpak Shah 已提交
4169 4170 4171 4172 4173 4174 4175 4176 4177 4178 4179 4180 4181 4182 4183 4184 4185 4186 4187 4188 4189 4190

	/* 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;
		if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
				       EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE)) {
			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;
4191 4192
		ext4_msg(sb, KERN_INFO, "required extra inode space not"
			 "available");
K
Kalpak Shah 已提交
4193 4194
	}

4195
	err = ext4_reserve_clusters(sbi, ext4_calculate_resv_clusters(sb));
L
Lukas Czerner 已提交
4196 4197
	if (err) {
		ext4_msg(sb, KERN_ERR, "failed to reserve %llu clusters for "
4198
			 "reserved pool", ext4_calculate_resv_clusters(sb));
4199
		goto failed_mount4a;
L
Lukas Czerner 已提交
4200 4201
	}

4202 4203
	err = ext4_setup_system_zone(sb);
	if (err) {
4204
		ext4_msg(sb, KERN_ERR, "failed to initialize system "
4205
			 "zone (%d)", err);
4206 4207 4208 4209 4210 4211 4212 4213
		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);
4214
		goto failed_mount5;
4215 4216
	}

4217 4218 4219
	block = ext4_count_free_clusters(sb);
	ext4_free_blocks_count_set(sbi->s_es, 
				   EXT4_C2B(sbi, block));
4220 4221
	err = percpu_counter_init(&sbi->s_freeclusters_counter, block,
				  GFP_KERNEL);
4222 4223 4224
	if (!err) {
		unsigned long freei = ext4_count_free_inodes(sb);
		sbi->s_es->s_free_inodes_count = cpu_to_le32(freei);
4225 4226
		err = percpu_counter_init(&sbi->s_freeinodes_counter, freei,
					  GFP_KERNEL);
4227 4228 4229
	}
	if (!err)
		err = percpu_counter_init(&sbi->s_dirs_counter,
4230
					  ext4_count_dirs(sb), GFP_KERNEL);
4231
	if (!err)
4232 4233
		err = percpu_counter_init(&sbi->s_dirtyclusters_counter, 0,
					  GFP_KERNEL);
4234 4235 4236 4237 4238 4239 4240 4241 4242 4243 4244 4245 4246
	if (err) {
		ext4_msg(sb, KERN_ERR, "insufficient memory");
		goto failed_mount6;
	}

	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG))
		if (!ext4_fill_flex_info(sb)) {
			ext4_msg(sb, KERN_ERR,
			       "unable to initialize "
			       "flex_bg meta info!");
			goto failed_mount6;
		}

4247 4248
	err = ext4_register_li_request(sb, first_not_zeroed);
	if (err)
4249
		goto failed_mount6;
4250

T
Theodore Ts'o 已提交
4251 4252 4253 4254
	sbi->s_kobj.kset = ext4_kset;
	init_completion(&sbi->s_kobj_unregister);
	err = kobject_init_and_add(&sbi->s_kobj, &ext4_ktype, NULL,
				   "%s", sb->s_id);
4255 4256
	if (err)
		goto failed_mount7;
T
Theodore Ts'o 已提交
4257

4258 4259 4260 4261 4262 4263 4264 4265 4266 4267
#ifdef CONFIG_QUOTA
	/* Enable quota usage during mount. */
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_QUOTA) &&
	    !(sb->s_flags & MS_RDONLY)) {
		err = ext4_enable_quotas(sb);
		if (err)
			goto failed_mount8;
	}
#endif  /* CONFIG_QUOTA */

4268 4269 4270
	EXT4_SB(sb)->s_mount_state |= EXT4_ORPHAN_FS;
	ext4_orphan_cleanup(sb, es);
	EXT4_SB(sb)->s_mount_state &= ~EXT4_ORPHAN_FS;
4271
	if (needs_recovery) {
4272
		ext4_msg(sb, KERN_INFO, "recovery complete");
4273 4274 4275 4276 4277 4278 4279 4280 4281 4282 4283 4284
		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";

4285 4286 4287 4288 4289 4290 4291 4292
	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");
	}

4293 4294 4295 4296
	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);
4297

4298 4299
	if (es->s_error_count)
		mod_timer(&sbi->s_err_report, jiffies + 300*HZ); /* 5 minutes */
4300

4301 4302 4303 4304 4305
	/* 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);

4306
	kfree(orig_data);
4307 4308
	return 0;

4309
cantfind_ext4:
4310
	if (!silent)
4311
		ext4_msg(sb, KERN_ERR, "VFS: Can't find ext4 filesystem");
4312 4313
	goto failed_mount;

4314 4315 4316 4317
#ifdef CONFIG_QUOTA
failed_mount8:
	kobject_del(&sbi->s_kobj);
#endif
4318 4319 4320
failed_mount7:
	ext4_unregister_li_request(sb);
failed_mount6:
4321
	ext4_mb_release(sb);
4322
	if (sbi->s_flex_groups)
A
Al Viro 已提交
4323
		kvfree(sbi->s_flex_groups);
4324 4325 4326 4327
	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);
4328
failed_mount5:
4329 4330 4331
	ext4_ext_release(sb);
	ext4_release_system_zone(sb);
failed_mount4a:
A
Al Viro 已提交
4332
	dput(sb->s_root);
4333
	sb->s_root = NULL;
A
Al Viro 已提交
4334
failed_mount4:
4335
	ext4_msg(sb, KERN_ERR, "mount failed");
4336 4337
	if (EXT4_SB(sb)->rsv_conversion_wq)
		destroy_workqueue(EXT4_SB(sb)->rsv_conversion_wq);
4338
failed_mount_wq:
4339 4340 4341 4342
	if (sbi->s_journal) {
		jbd2_journal_destroy(sbi->s_journal);
		sbi->s_journal = NULL;
	}
4343
failed_mount3a:
4344
	ext4_es_unregister_shrinker(sbi);
4345
failed_mount3:
4346
	del_timer_sync(&sbi->s_err_report);
4347 4348
	if (sbi->s_mmp_tsk)
		kthread_stop(sbi->s_mmp_tsk);
4349 4350 4351
failed_mount2:
	for (i = 0; i < db_count; i++)
		brelse(sbi->s_group_desc[i]);
A
Al Viro 已提交
4352
	kvfree(sbi->s_group_desc);
4353
failed_mount:
4354 4355
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
4356
	if (sbi->s_proc) {
4357
		remove_proc_entry("options", sbi->s_proc);
4358
		remove_proc_entry(sb->s_id, ext4_proc_root);
4359
	}
4360
#ifdef CONFIG_QUOTA
J
Jan Kara 已提交
4361
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
4362 4363
		kfree(sbi->s_qf_names[i]);
#endif
4364
	ext4_blkdev_remove(sbi);
4365 4366 4367
	brelse(bh);
out_fail:
	sb->s_fs_info = NULL;
4368
	kfree(sbi->s_blockgroup_lock);
4369
	kfree(sbi);
4370
out_free_orig:
4371
	kfree(orig_data);
4372
	return err ? err : ret;
4373 4374 4375 4376 4377 4378 4379
}

/*
 * 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.
 */
4380
static void ext4_init_journal_params(struct super_block *sb, journal_t *journal)
4381
{
4382
	struct ext4_sb_info *sbi = EXT4_SB(sb);
4383

4384 4385 4386
	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;
4387

4388
	write_lock(&journal->j_state_lock);
4389
	if (test_opt(sb, BARRIER))
4390
		journal->j_flags |= JBD2_BARRIER;
4391
	else
4392
		journal->j_flags &= ~JBD2_BARRIER;
4393 4394 4395 4396
	if (test_opt(sb, DATA_ERR_ABORT))
		journal->j_flags |= JBD2_ABORT_ON_SYNCDATA_ERR;
	else
		journal->j_flags &= ~JBD2_ABORT_ON_SYNCDATA_ERR;
4397
	write_unlock(&journal->j_state_lock);
4398 4399
}

4400
static journal_t *ext4_get_journal(struct super_block *sb,
4401 4402 4403 4404 4405
				   unsigned int journal_inum)
{
	struct inode *journal_inode;
	journal_t *journal;

4406 4407
	BUG_ON(!EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL));

4408 4409 4410 4411
	/* 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. */

4412 4413
	journal_inode = ext4_iget(sb, journal_inum);
	if (IS_ERR(journal_inode)) {
4414
		ext4_msg(sb, KERN_ERR, "no journal found");
4415 4416 4417 4418 4419
		return NULL;
	}
	if (!journal_inode->i_nlink) {
		make_bad_inode(journal_inode);
		iput(journal_inode);
4420
		ext4_msg(sb, KERN_ERR, "journal inode is deleted");
4421 4422 4423
		return NULL;
	}

4424
	jbd_debug(2, "Journal inode found at %p: %lld bytes\n",
4425
		  journal_inode, journal_inode->i_size);
4426
	if (!S_ISREG(journal_inode->i_mode)) {
4427
		ext4_msg(sb, KERN_ERR, "invalid journal inode");
4428 4429 4430 4431
		iput(journal_inode);
		return NULL;
	}

4432
	journal = jbd2_journal_init_inode(journal_inode);
4433
	if (!journal) {
4434
		ext4_msg(sb, KERN_ERR, "Could not load journal inode");
4435 4436 4437 4438
		iput(journal_inode);
		return NULL;
	}
	journal->j_private = sb;
4439
	ext4_init_journal_params(sb, journal);
4440 4441 4442
	return journal;
}

4443
static journal_t *ext4_get_dev_journal(struct super_block *sb,
4444 4445
				       dev_t j_dev)
{
4446
	struct buffer_head *bh;
4447
	journal_t *journal;
4448 4449
	ext4_fsblk_t start;
	ext4_fsblk_t len;
4450
	int hblock, blocksize;
4451
	ext4_fsblk_t sb_block;
4452
	unsigned long offset;
4453
	struct ext4_super_block *es;
4454 4455
	struct block_device *bdev;

4456 4457
	BUG_ON(!EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL));

4458
	bdev = ext4_blkdev_get(j_dev, sb);
4459 4460 4461 4462
	if (bdev == NULL)
		return NULL;

	blocksize = sb->s_blocksize;
4463
	hblock = bdev_logical_block_size(bdev);
4464
	if (blocksize < hblock) {
4465 4466
		ext4_msg(sb, KERN_ERR,
			"blocksize too small for journal device");
4467 4468 4469
		goto out_bdev;
	}

4470 4471
	sb_block = EXT4_MIN_BLOCK_SIZE / blocksize;
	offset = EXT4_MIN_BLOCK_SIZE % blocksize;
4472 4473
	set_blocksize(bdev, blocksize);
	if (!(bh = __bread(bdev, sb_block, blocksize))) {
4474 4475
		ext4_msg(sb, KERN_ERR, "couldn't read superblock of "
		       "external journal");
4476 4477 4478
		goto out_bdev;
	}

4479
	es = (struct ext4_super_block *) (bh->b_data + offset);
4480
	if ((le16_to_cpu(es->s_magic) != EXT4_SUPER_MAGIC) ||
4481
	    !(le32_to_cpu(es->s_feature_incompat) &
4482
	      EXT4_FEATURE_INCOMPAT_JOURNAL_DEV)) {
4483 4484
		ext4_msg(sb, KERN_ERR, "external journal has "
					"bad superblock");
4485 4486 4487 4488
		brelse(bh);
		goto out_bdev;
	}

4489 4490 4491 4492 4493 4494 4495 4496 4497
	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;
	}

4498
	if (memcmp(EXT4_SB(sb)->s_es->s_journal_uuid, es->s_uuid, 16)) {
4499
		ext4_msg(sb, KERN_ERR, "journal UUID does not match");
4500 4501 4502 4503
		brelse(bh);
		goto out_bdev;
	}

L
Laurent Vivier 已提交
4504
	len = ext4_blocks_count(es);
4505 4506 4507
	start = sb_block + 1;
	brelse(bh);	/* we're done with the superblock */

4508
	journal = jbd2_journal_init_dev(bdev, sb->s_bdev,
4509 4510
					start, len, blocksize);
	if (!journal) {
4511
		ext4_msg(sb, KERN_ERR, "failed to create device journal");
4512 4513 4514
		goto out_bdev;
	}
	journal->j_private = sb;
4515
	ll_rw_block(READ | REQ_META | REQ_PRIO, 1, &journal->j_sb_buffer);
4516 4517
	wait_on_buffer(journal->j_sb_buffer);
	if (!buffer_uptodate(journal->j_sb_buffer)) {
4518
		ext4_msg(sb, KERN_ERR, "I/O error on journal device");
4519 4520 4521
		goto out_journal;
	}
	if (be32_to_cpu(journal->j_superblock->s_nr_users) != 1) {
4522 4523
		ext4_msg(sb, KERN_ERR, "External journal has more than one "
					"user (unsupported) - %d",
4524 4525 4526
			be32_to_cpu(journal->j_superblock->s_nr_users));
		goto out_journal;
	}
4527 4528
	EXT4_SB(sb)->journal_bdev = bdev;
	ext4_init_journal_params(sb, journal);
4529
	return journal;
4530

4531
out_journal:
4532
	jbd2_journal_destroy(journal);
4533
out_bdev:
4534
	ext4_blkdev_put(bdev);
4535 4536 4537
	return NULL;
}

4538 4539
static int ext4_load_journal(struct super_block *sb,
			     struct ext4_super_block *es,
4540 4541 4542 4543 4544 4545 4546 4547
			     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;

4548 4549
	BUG_ON(!EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL));

4550 4551
	if (journal_devnum &&
	    journal_devnum != le32_to_cpu(es->s_journal_dev)) {
4552 4553
		ext4_msg(sb, KERN_INFO, "external journal device major/minor "
			"numbers have changed");
4554 4555 4556 4557 4558 4559 4560 4561 4562 4563 4564
		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.
	 */
4565
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER)) {
4566
		if (sb->s_flags & MS_RDONLY) {
4567 4568
			ext4_msg(sb, KERN_INFO, "INFO: recovery "
					"required on readonly filesystem");
4569
			if (really_read_only) {
4570 4571
				ext4_msg(sb, KERN_ERR, "write access "
					"unavailable, cannot proceed");
4572 4573
				return -EROFS;
			}
4574 4575
			ext4_msg(sb, KERN_INFO, "write access will "
			       "be enabled during recovery");
4576 4577 4578 4579
		}
	}

	if (journal_inum && journal_dev) {
4580 4581
		ext4_msg(sb, KERN_ERR, "filesystem has both journal "
		       "and inode journals!");
4582 4583 4584 4585
		return -EINVAL;
	}

	if (journal_inum) {
4586
		if (!(journal = ext4_get_journal(sb, journal_inum)))
4587 4588
			return -EINVAL;
	} else {
4589
		if (!(journal = ext4_get_dev_journal(sb, journal_dev)))
4590 4591 4592
			return -EINVAL;
	}

4593
	if (!(journal->j_flags & JBD2_BARRIER))
4594
		ext4_msg(sb, KERN_INFO, "barriers disabled");
4595

4596
	if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER))
4597
		err = jbd2_journal_wipe(journal, !really_read_only);
4598 4599 4600 4601 4602
	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);
4603
		err = jbd2_journal_load(journal);
4604 4605 4606 4607 4608
		if (save)
			memcpy(((char *) es) + EXT4_S_ERR_START,
			       save, EXT4_S_ERR_LEN);
		kfree(save);
	}
4609 4610

	if (err) {
4611
		ext4_msg(sb, KERN_ERR, "error loading journal");
4612
		jbd2_journal_destroy(journal);
4613 4614 4615
		return err;
	}

4616 4617
	EXT4_SB(sb)->s_journal = journal;
	ext4_clear_journal_err(sb, es);
4618

4619
	if (!really_read_only && journal_devnum &&
4620 4621 4622 4623
	    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. */
4624
		ext4_commit_super(sb, 1);
4625 4626 4627 4628 4629
	}

	return 0;
}

4630
static int ext4_commit_super(struct super_block *sb, int sync)
4631
{
4632
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;
4633
	struct buffer_head *sbh = EXT4_SB(sb)->s_sbh;
4634
	int error = 0;
4635

4636
	if (!sbh || block_device_ejected(sb))
4637
		return error;
4638 4639 4640 4641 4642 4643 4644 4645 4646
	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.
		 */
4647 4648
		ext4_msg(sb, KERN_ERR, "previous I/O error to "
		       "superblock detected");
4649 4650 4651
		clear_buffer_write_io_error(sbh);
		set_buffer_uptodate(sbh);
	}
4652 4653 4654 4655 4656 4657 4658 4659 4660 4661 4662 4663
	/*
	 * 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());
4664 4665 4666
	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
4667 4668
			    ((part_stat_read(sb->s_bdev->bd_part, sectors[1]) -
			      EXT4_SB(sb)->s_sectors_written_start) >> 1));
4669 4670 4671
	else
		es->s_kbytes_written =
			cpu_to_le64(EXT4_SB(sb)->s_kbytes_written);
4672 4673
	if (percpu_counter_initialized(&EXT4_SB(sb)->s_freeclusters_counter))
		ext4_free_blocks_count_set(es,
4674 4675
			EXT4_C2B(EXT4_SB(sb), percpu_counter_sum_positive(
				&EXT4_SB(sb)->s_freeclusters_counter)));
4676 4677 4678
	if (percpu_counter_initialized(&EXT4_SB(sb)->s_freeinodes_counter))
		es->s_free_inodes_count =
			cpu_to_le32(percpu_counter_sum_positive(
4679
				&EXT4_SB(sb)->s_freeinodes_counter));
4680
	BUFFER_TRACE(sbh, "marking dirty");
4681
	ext4_superblock_csum_set(sb);
4682
	mark_buffer_dirty(sbh);
4683
	if (sync) {
4684 4685
		error = __sync_dirty_buffer(sbh,
			test_opt(sb, BARRIER) ? WRITE_FUA : WRITE_SYNC);
4686 4687 4688 4689 4690
		if (error)
			return error;

		error = buffer_write_io_error(sbh);
		if (error) {
4691 4692
			ext4_msg(sb, KERN_ERR, "I/O error while writing "
			       "superblock");
4693 4694 4695 4696
			clear_buffer_write_io_error(sbh);
			set_buffer_uptodate(sbh);
		}
	}
4697
	return error;
4698 4699 4700 4701 4702 4703 4704
}

/*
 * 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.
 */
4705 4706
static void ext4_mark_recovery_complete(struct super_block *sb,
					struct ext4_super_block *es)
4707
{
4708
	journal_t *journal = EXT4_SB(sb)->s_journal;
4709

4710 4711 4712 4713
	if (!EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL)) {
		BUG_ON(journal != NULL);
		return;
	}
4714
	jbd2_journal_lock_updates(journal);
4715 4716 4717
	if (jbd2_journal_flush(journal) < 0)
		goto out;

4718
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER) &&
4719
	    sb->s_flags & MS_RDONLY) {
4720
		EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
4721
		ext4_commit_super(sb, 1);
4722
	}
4723 4724

out:
4725
	jbd2_journal_unlock_updates(journal);
4726 4727 4728 4729 4730 4731 4732
}

/*
 * 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.
 */
4733 4734
static void ext4_clear_journal_err(struct super_block *sb,
				   struct ext4_super_block *es)
4735 4736 4737 4738 4739
{
	journal_t *journal;
	int j_errno;
	const char *errstr;

4740 4741
	BUG_ON(!EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL));

4742
	journal = EXT4_SB(sb)->s_journal;
4743 4744 4745

	/*
	 * Now check for any error status which may have been recorded in the
4746
	 * journal by a prior ext4_error() or ext4_abort()
4747 4748
	 */

4749
	j_errno = jbd2_journal_errno(journal);
4750 4751 4752
	if (j_errno) {
		char nbuf[16];

4753
		errstr = ext4_decode_error(sb, j_errno, nbuf);
4754
		ext4_warning(sb, "Filesystem error recorded "
4755
			     "from previous mount: %s", errstr);
4756
		ext4_warning(sb, "Marking fs in need of filesystem check.");
4757

4758 4759
		EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS;
		es->s_state |= cpu_to_le16(EXT4_ERROR_FS);
4760
		ext4_commit_super(sb, 1);
4761

4762
		jbd2_journal_clear_err(journal);
4763
		jbd2_journal_update_sb_errno(journal);
4764 4765 4766 4767 4768 4769 4770
	}
}

/*
 * Force the running and committing transactions to commit,
 * and wait on the commit.
 */
4771
int ext4_force_commit(struct super_block *sb)
4772 4773 4774 4775 4776 4777
{
	journal_t *journal;

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

4778
	journal = EXT4_SB(sb)->s_journal;
4779
	return ext4_journal_force_commit(journal);
4780 4781
}

4782
static int ext4_sync_fs(struct super_block *sb, int wait)
4783
{
4784
	int ret = 0;
4785
	tid_t target;
4786
	bool needs_barrier = false;
4787
	struct ext4_sb_info *sbi = EXT4_SB(sb);
4788

4789
	trace_ext4_sync_fs(sb, wait);
4790
	flush_workqueue(sbi->rsv_conversion_wq);
4791 4792 4793 4794 4795
	/*
	 * Writeback quota in non-journalled quota case - journalled quota has
	 * no dirty dquots
	 */
	dquot_writeback_dquots(sb, -1);
4796 4797 4798 4799 4800
	/*
	 * 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.
	 */
4801 4802 4803 4804 4805 4806 4807 4808 4809 4810 4811 4812
	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))
4813 4814 4815 4816 4817 4818
		needs_barrier = true;
	if (needs_barrier) {
		int err;
		err = blkdev_issue_flush(sb->s_bdev, GFP_KERNEL, NULL);
		if (!ret)
			ret = err;
4819
	}
4820 4821 4822 4823

	return ret;
}

4824 4825 4826
/*
 * 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.
4827 4828
 *
 * Note that only this function cannot bring a filesystem to be in a clean
4829 4830
 * state independently. It relies on upper layer to stop all data & metadata
 * modifications.
4831
 */
4832
static int ext4_freeze(struct super_block *sb)
4833
{
4834 4835
	int error = 0;
	journal_t *journal;
4836

4837 4838
	if (sb->s_flags & MS_RDONLY)
		return 0;
4839

4840
	journal = EXT4_SB(sb)->s_journal;
4841

4842 4843 4844
	if (journal) {
		/* Now we set up the journal barrier. */
		jbd2_journal_lock_updates(journal);
4845

4846 4847 4848 4849 4850 4851 4852
		/*
		 * Don't clear the needs_recovery flag if we failed to
		 * flush the journal.
		 */
		error = jbd2_journal_flush(journal);
		if (error < 0)
			goto out;
4853 4854 4855

		/* Journal blocked and flushed, clear needs_recovery flag. */
		EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
4856
	}
4857 4858

	error = ext4_commit_super(sb, 1);
4859
out:
4860 4861 4862
	if (journal)
		/* we rely on upper layer to stop further updates */
		jbd2_journal_unlock_updates(journal);
4863
	return error;
4864 4865 4866 4867 4868 4869
}

/*
 * 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.
 */
4870
static int ext4_unfreeze(struct super_block *sb)
4871
{
4872 4873 4874
	if (sb->s_flags & MS_RDONLY)
		return 0;

4875 4876 4877 4878 4879
	if (EXT4_SB(sb)->s_journal) {
		/* Reset the needs_recovery flag before the fs is unlocked. */
		EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
	}

4880
	ext4_commit_super(sb, 1);
4881
	return 0;
4882 4883
}

4884 4885 4886 4887 4888
/*
 * Structure to save mount options for ext4_remount's benefit
 */
struct ext4_mount_options {
	unsigned long s_mount_opt;
4889
	unsigned long s_mount_opt2;
4890 4891
	kuid_t s_resuid;
	kgid_t s_resgid;
4892 4893 4894 4895
	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
4896
	char *s_qf_names[EXT4_MAXQUOTAS];
4897 4898 4899
#endif
};

4900
static int ext4_remount(struct super_block *sb, int *flags, char *data)
4901
{
4902
	struct ext4_super_block *es;
4903
	struct ext4_sb_info *sbi = EXT4_SB(sb);
4904
	unsigned long old_sb_flags;
4905
	struct ext4_mount_options old_opts;
4906
	int enable_quota = 0;
4907
	ext4_group_t g;
4908
	unsigned int journal_ioprio = DEFAULT_JOURNAL_IOPRIO;
4909
	int err = 0;
4910
#ifdef CONFIG_QUOTA
4911
	int i, j;
4912
#endif
4913
	char *orig_data = kstrdup(data, GFP_KERNEL);
4914 4915 4916 4917

	/* Store the original options */
	old_sb_flags = sb->s_flags;
	old_opts.s_mount_opt = sbi->s_mount_opt;
4918
	old_opts.s_mount_opt2 = sbi->s_mount_opt2;
4919 4920 4921
	old_opts.s_resuid = sbi->s_resuid;
	old_opts.s_resgid = sbi->s_resgid;
	old_opts.s_commit_interval = sbi->s_commit_interval;
4922 4923
	old_opts.s_min_batch_time = sbi->s_min_batch_time;
	old_opts.s_max_batch_time = sbi->s_max_batch_time;
4924 4925
#ifdef CONFIG_QUOTA
	old_opts.s_jquota_fmt = sbi->s_jquota_fmt;
J
Jan Kara 已提交
4926
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
4927 4928 4929 4930 4931 4932
		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]);
4933
				kfree(orig_data);
4934 4935 4936 4937
				return -ENOMEM;
			}
		} else
			old_opts.s_qf_names[i] = NULL;
4938
#endif
4939 4940
	if (sbi->s_journal && sbi->s_journal->j_task->io_context)
		journal_ioprio = sbi->s_journal->j_task->io_context->ioprio;
4941

4942
	if (!parse_options(data, sb, NULL, &journal_ioprio, 1)) {
4943 4944 4945 4946
		err = -EINVAL;
		goto restore_opts;
	}

4947
	if ((old_opts.s_mount_opt & EXT4_MOUNT_JOURNAL_CHECKSUM) ^
4948 4949
	    test_opt(sb, JOURNAL_CHECKSUM)) {
		ext4_msg(sb, KERN_ERR, "changing journal_checksum "
4950 4951
			 "during remount not supported; ignoring");
		sbi->s_mount_opt ^= EXT4_MOUNT_JOURNAL_CHECKSUM;
4952 4953
	}

4954 4955 4956 4957 4958 4959 4960 4961 4962 4963 4964 4965 4966
	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
4967 4968 4969 4970 4971 4972 4973 4974 4975 4976 4977 4978
		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;
4979 4980
	}

4981
	if (sbi->s_mount_flags & EXT4_MF_FS_ABORTED)
4982
		ext4_abort(sb, "Abort forced by user");
4983 4984

	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
4985
		(test_opt(sb, POSIX_ACL) ? MS_POSIXACL : 0);
4986 4987 4988

	es = sbi->s_es;

4989
	if (sbi->s_journal) {
4990
		ext4_init_journal_params(sb, sbi->s_journal);
4991 4992
		set_task_ioprio(sbi->s_journal->j_task, journal_ioprio);
	}
4993

4994 4995 4996
	if (*flags & MS_LAZYTIME)
		sb->s_flags |= MS_LAZYTIME;

4997
	if ((*flags & MS_RDONLY) != (sb->s_flags & MS_RDONLY)) {
4998
		if (sbi->s_mount_flags & EXT4_MF_FS_ABORTED) {
4999 5000 5001 5002 5003
			err = -EROFS;
			goto restore_opts;
		}

		if (*flags & MS_RDONLY) {
5004 5005 5006
			err = sync_filesystem(sb);
			if (err < 0)
				goto restore_opts;
5007 5008
			err = dquot_suspend(sb, -1);
			if (err < 0)
5009 5010
				goto restore_opts;

5011 5012 5013 5014 5015 5016 5017 5018 5019 5020 5021
			/*
			 * 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.
			 */
5022 5023
			if (!(es->s_state & cpu_to_le16(EXT4_VALID_FS)) &&
			    (sbi->s_mount_state & EXT4_VALID_FS))
5024 5025
				es->s_state = cpu_to_le16(sbi->s_mount_state);

5026
			if (sbi->s_journal)
5027
				ext4_mark_recovery_complete(sb, es);
5028
		} else {
5029
			/* Make sure we can mount this feature set readwrite */
D
Darrick J. Wong 已提交
5030 5031 5032
			if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
					EXT4_FEATURE_RO_COMPAT_READONLY) ||
			    !ext4_feature_set_ok(sb, 0)) {
5033 5034 5035
				err = -EROFS;
				goto restore_opts;
			}
5036 5037
			/*
			 * Make sure the group descriptor checksums
5038
			 * are sane.  If they aren't, refuse to remount r/w.
5039 5040 5041 5042 5043
			 */
			for (g = 0; g < sbi->s_groups_count; g++) {
				struct ext4_group_desc *gdp =
					ext4_get_group_desc(sb, g, NULL);

5044
				if (!ext4_group_desc_csum_verify(sb, g, gdp)) {
5045 5046
					ext4_msg(sb, KERN_ERR,
	       "ext4_remount: Checksum for group %u failed (%u!=%u)",
5047 5048 5049 5050 5051 5052 5053
		g, le16_to_cpu(ext4_group_desc_csum(sbi, g, gdp)),
					       le16_to_cpu(gdp->bg_checksum));
					err = -EINVAL;
					goto restore_opts;
				}
			}

5054 5055 5056 5057 5058 5059
			/*
			 * 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) {
5060
				ext4_msg(sb, KERN_WARNING, "Couldn't "
5061 5062
				       "remount RDWR because of unprocessed "
				       "orphan inode list.  Please "
5063
				       "umount/remount instead");
5064 5065 5066 5067
				err = -EINVAL;
				goto restore_opts;
			}

5068 5069 5070 5071 5072 5073
			/*
			 * 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.)
			 */
5074 5075
			if (sbi->s_journal)
				ext4_clear_journal_err(sb, es);
5076
			sbi->s_mount_state = le16_to_cpu(es->s_state);
5077
			if (!ext4_setup_super(sb, es, 0))
5078
				sb->s_flags &= ~MS_RDONLY;
5079 5080 5081 5082 5083 5084 5085
			if (EXT4_HAS_INCOMPAT_FEATURE(sb,
						     EXT4_FEATURE_INCOMPAT_MMP))
				if (ext4_multi_mount_protect(sb,
						le64_to_cpu(es->s_mmp_block))) {
					err = -EROFS;
					goto restore_opts;
				}
5086
			enable_quota = 1;
5087 5088
		}
	}
5089 5090 5091 5092 5093 5094 5095 5096 5097 5098 5099 5100 5101

	/*
	 * 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);
	}

5102
	ext4_setup_system_zone(sb);
5103
	if (sbi->s_journal == NULL && !(old_sb_flags & MS_RDONLY))
5104
		ext4_commit_super(sb, 1);
5105

5106 5107
#ifdef CONFIG_QUOTA
	/* Release old quota file names */
J
Jan Kara 已提交
5108
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
5109
		kfree(old_opts.s_qf_names[i]);
5110 5111 5112 5113 5114 5115
	if (enable_quota) {
		if (sb_any_quota_suspended(sb))
			dquot_resume(sb, -1);
		else if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
					EXT4_FEATURE_RO_COMPAT_QUOTA)) {
			err = ext4_enable_quotas(sb);
5116
			if (err)
5117 5118 5119
				goto restore_opts;
		}
	}
5120
#endif
5121

5122
	*flags = (*flags & ~MS_LAZYTIME) | (sb->s_flags & MS_LAZYTIME);
5123 5124
	ext4_msg(sb, KERN_INFO, "re-mounted. Opts: %s", orig_data);
	kfree(orig_data);
5125
	return 0;
5126

5127 5128 5129
restore_opts:
	sb->s_flags = old_sb_flags;
	sbi->s_mount_opt = old_opts.s_mount_opt;
5130
	sbi->s_mount_opt2 = old_opts.s_mount_opt2;
5131 5132 5133
	sbi->s_resuid = old_opts.s_resuid;
	sbi->s_resgid = old_opts.s_resgid;
	sbi->s_commit_interval = old_opts.s_commit_interval;
5134 5135
	sbi->s_min_batch_time = old_opts.s_min_batch_time;
	sbi->s_max_batch_time = old_opts.s_max_batch_time;
5136 5137
#ifdef CONFIG_QUOTA
	sbi->s_jquota_fmt = old_opts.s_jquota_fmt;
J
Jan Kara 已提交
5138
	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
5139
		kfree(sbi->s_qf_names[i]);
5140 5141 5142
		sbi->s_qf_names[i] = old_opts.s_qf_names[i];
	}
#endif
5143
	kfree(orig_data);
5144 5145 5146
	return err;
}

5147
static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf)
5148 5149
{
	struct super_block *sb = dentry->d_sb;
5150 5151
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_super_block *es = sbi->s_es;
L
Lukas Czerner 已提交
5152
	ext4_fsblk_t overhead = 0, resv_blocks;
P
Pekka Enberg 已提交
5153
	u64 fsid;
5154
	s64 bfree;
L
Lukas Czerner 已提交
5155
	resv_blocks = EXT4_C2B(sbi, atomic64_read(&sbi->s_resv_clusters));
5156

5157 5158
	if (!test_opt(sb, MINIX_DF))
		overhead = sbi->s_overhead;
5159

5160
	buf->f_type = EXT4_SUPER_MAGIC;
5161
	buf->f_bsize = sb->s_blocksize;
5162
	buf->f_blocks = ext4_blocks_count(es) - EXT4_C2B(sbi, overhead);
5163 5164
	bfree = percpu_counter_sum_positive(&sbi->s_freeclusters_counter) -
		percpu_counter_sum_positive(&sbi->s_dirtyclusters_counter);
5165
	/* prevent underflow in case that few free space is available */
5166
	buf->f_bfree = EXT4_C2B(sbi, max_t(s64, bfree, 0));
L
Lukas Czerner 已提交
5167 5168 5169
	buf->f_bavail = buf->f_bfree -
			(ext4_r_blocks_count(es) + resv_blocks);
	if (buf->f_bfree < (ext4_r_blocks_count(es) + resv_blocks))
5170 5171
		buf->f_bavail = 0;
	buf->f_files = le32_to_cpu(es->s_inodes_count);
P
Peter Zijlstra 已提交
5172
	buf->f_ffree = percpu_counter_sum_positive(&sbi->s_freeinodes_counter);
5173
	buf->f_namelen = EXT4_NAME_LEN;
P
Pekka Enberg 已提交
5174 5175 5176 5177
	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;
5178

5179 5180 5181
	return 0;
}

5182 5183
/* Helper function for writing quotas on sync - we need to start transaction
 * before quota file is locked for write. Otherwise the are possible deadlocks:
5184
 * Process 1                         Process 2
5185
 * ext4_create()                     quota_sync()
5186
 *   jbd2_journal_start()                  write_dquot()
5187
 *   dquot_initialize()                         down(dqio_mutex)
5188
 *     down(dqio_mutex)                    jbd2_journal_start()
5189 5190 5191 5192 5193 5194 5195
 *
 */

#ifdef CONFIG_QUOTA

static inline struct inode *dquot_to_inode(struct dquot *dquot)
{
5196
	return sb_dqopt(dquot->dq_sb)->files[dquot->dq_id.type];
5197 5198
}

5199
static int ext4_write_dquot(struct dquot *dquot)
5200 5201 5202 5203 5204 5205
{
	int ret, err;
	handle_t *handle;
	struct inode *inode;

	inode = dquot_to_inode(dquot);
5206
	handle = ext4_journal_start(inode, EXT4_HT_QUOTA,
5207
				    EXT4_QUOTA_TRANS_BLOCKS(dquot->dq_sb));
5208 5209 5210
	if (IS_ERR(handle))
		return PTR_ERR(handle);
	ret = dquot_commit(dquot);
5211
	err = ext4_journal_stop(handle);
5212 5213 5214 5215 5216
	if (!ret)
		ret = err;
	return ret;
}

5217
static int ext4_acquire_dquot(struct dquot *dquot)
5218 5219 5220 5221
{
	int ret, err;
	handle_t *handle;

5222
	handle = ext4_journal_start(dquot_to_inode(dquot), EXT4_HT_QUOTA,
5223
				    EXT4_QUOTA_INIT_BLOCKS(dquot->dq_sb));
5224 5225 5226
	if (IS_ERR(handle))
		return PTR_ERR(handle);
	ret = dquot_acquire(dquot);
5227
	err = ext4_journal_stop(handle);
5228 5229 5230 5231 5232
	if (!ret)
		ret = err;
	return ret;
}

5233
static int ext4_release_dquot(struct dquot *dquot)
5234 5235 5236 5237
{
	int ret, err;
	handle_t *handle;

5238
	handle = ext4_journal_start(dquot_to_inode(dquot), EXT4_HT_QUOTA,
5239
				    EXT4_QUOTA_DEL_BLOCKS(dquot->dq_sb));
J
Jan Kara 已提交
5240 5241 5242
	if (IS_ERR(handle)) {
		/* Release dquot anyway to avoid endless cycle in dqput() */
		dquot_release(dquot);
5243
		return PTR_ERR(handle);
J
Jan Kara 已提交
5244
	}
5245
	ret = dquot_release(dquot);
5246
	err = ext4_journal_stop(handle);
5247 5248 5249 5250 5251
	if (!ret)
		ret = err;
	return ret;
}

5252
static int ext4_mark_dquot_dirty(struct dquot *dquot)
5253
{
5254 5255 5256
	struct super_block *sb = dquot->dq_sb;
	struct ext4_sb_info *sbi = EXT4_SB(sb);

5257
	/* Are we journaling quotas? */
5258 5259
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_QUOTA) ||
	    sbi->s_qf_names[USRQUOTA] || sbi->s_qf_names[GRPQUOTA]) {
5260
		dquot_mark_dquot_dirty(dquot);
5261
		return ext4_write_dquot(dquot);
5262 5263 5264 5265 5266
	} else {
		return dquot_mark_dquot_dirty(dquot);
	}
}

5267
static int ext4_write_info(struct super_block *sb, int type)
5268 5269 5270 5271 5272
{
	int ret, err;
	handle_t *handle;

	/* Data block + inode block */
5273
	handle = ext4_journal_start(d_inode(sb->s_root), EXT4_HT_QUOTA, 2);
5274 5275 5276
	if (IS_ERR(handle))
		return PTR_ERR(handle);
	ret = dquot_commit_info(sb, type);
5277
	err = ext4_journal_stop(handle);
5278 5279 5280 5281 5282 5283 5284 5285 5286
	if (!ret)
		ret = err;
	return ret;
}

/*
 * Turn on quotas during mount time - we need to find
 * the quota file and such...
 */
5287
static int ext4_quota_on_mount(struct super_block *sb, int type)
5288
{
5289 5290
	return dquot_quota_on_mount(sb, EXT4_SB(sb)->s_qf_names[type],
					EXT4_SB(sb)->s_jquota_fmt, type);
5291 5292 5293 5294 5295
}

/*
 * Standard function to be called on quota_on
 */
5296
static int ext4_quota_on(struct super_block *sb, int type, int format_id,
5297
			 struct path *path)
5298 5299 5300 5301 5302
{
	int err;

	if (!test_opt(sb, QUOTA))
		return -EINVAL;
5303

5304
	/* Quotafile not on the same filesystem? */
5305
	if (path->dentry->d_sb != sb)
5306
		return -EXDEV;
5307 5308
	/* Journaling quota? */
	if (EXT4_SB(sb)->s_qf_names[type]) {
5309
		/* Quotafile not in fs root? */
5310
		if (path->dentry->d_parent != sb->s_root)
5311 5312 5313
			ext4_msg(sb, KERN_WARNING,
				"Quota file not on filesystem root. "
				"Journaled quota will not work");
5314
	}
5315 5316 5317 5318 5319

	/*
	 * When we journal data on quota file, we have to flush journal to see
	 * all updates to the file when we bypass pagecache...
	 */
5320
	if (EXT4_SB(sb)->s_journal &&
5321
	    ext4_should_journal_data(d_inode(path->dentry))) {
5322 5323 5324 5325 5326
		/*
		 * We don't need to lock updates but journal_flush() could
		 * otherwise be livelocked...
		 */
		jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal);
5327
		err = jbd2_journal_flush(EXT4_SB(sb)->s_journal);
5328
		jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal);
5329
		if (err)
5330
			return err;
5331 5332
	}

5333
	return dquot_quota_on(sb, type, format_id, path);
5334 5335
}

5336 5337 5338 5339 5340
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
5341
	unsigned long qf_inums[EXT4_MAXQUOTAS] = {
5342 5343 5344 5345 5346 5347 5348 5349 5350 5351 5352 5353 5354 5355 5356
		le32_to_cpu(EXT4_SB(sb)->s_es->s_usr_quota_inum),
		le32_to_cpu(EXT4_SB(sb)->s_es->s_grp_quota_inum)
	};

	BUG_ON(!EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_QUOTA));

	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
5357 5358
	/* Don't account quota for quota files to avoid recursion */
	qf_inode->i_flags |= S_NOQUOTA;
5359 5360 5361 5362 5363 5364 5365 5366 5367 5368
	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
5369
	unsigned long qf_inums[EXT4_MAXQUOTAS] = {
5370 5371 5372 5373 5374
		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
5375
	for (type = 0; type < EXT4_MAXQUOTAS; type++) {
5376 5377 5378 5379 5380
		if (qf_inums[type]) {
			err = ext4_quota_enable(sb, type, QFMT_VFS_V1,
						DQUOT_USAGE_ENABLED);
			if (err) {
				ext4_warning(sb,
5381 5382 5383
					"Failed to enable quota tracking "
					"(type=%d, err=%d). Please run "
					"e2fsck to fix.", type, err);
5384 5385 5386 5387 5388 5389 5390
				return err;
			}
		}
	}
	return 0;
}

5391 5392
static int ext4_quota_off(struct super_block *sb, int type)
{
5393 5394 5395
	struct inode *inode = sb_dqopt(sb)->files[type];
	handle_t *handle;

5396 5397 5398
	/* Force all delayed allocation blocks to be allocated.
	 * Caller already holds s_umount sem */
	if (test_opt(sb, DELALLOC))
5399 5400
		sync_filesystem(sb);

5401 5402 5403
	if (!inode)
		goto out;

5404 5405
	/* Update modification times of quota files when userspace can
	 * start looking at them */
5406
	handle = ext4_journal_start(inode, EXT4_HT_QUOTA, 1);
5407 5408 5409 5410 5411 5412 5413
	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:
5414 5415 5416
	return dquot_quota_off(sb, type);
}

5417 5418
/* 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
5419
 * itself serializes the operations (and no one else should touch the files)
5420
 * we don't have to be afraid of races */
5421
static ssize_t ext4_quota_read(struct super_block *sb, int type, char *data,
5422 5423 5424
			       size_t len, loff_t off)
{
	struct inode *inode = sb_dqopt(sb)->files[type];
A
Aneesh Kumar K.V 已提交
5425
	ext4_lblk_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
5426 5427 5428 5429 5430 5431 5432 5433 5434 5435 5436 5437 5438 5439
	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;
5440 5441 5442
		bh = ext4_bread(NULL, inode, blk, 0);
		if (IS_ERR(bh))
			return PTR_ERR(bh);
5443 5444 5445 5446 5447 5448 5449 5450 5451 5452 5453 5454 5455 5456 5457
		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) */
5458
static ssize_t ext4_quota_write(struct super_block *sb, int type,
5459 5460 5461
				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
5462
	ext4_lblk_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
5463
	int err, offset = off & (sb->s_blocksize - 1);
5464
	int retries = 0;
5465 5466 5467
	struct buffer_head *bh;
	handle_t *handle = journal_current_handle();

5468
	if (EXT4_SB(sb)->s_journal && !handle) {
5469 5470
		ext4_msg(sb, KERN_WARNING, "Quota write (off=%llu, len=%llu)"
			" cancelled because transaction is not started",
J
Jan Kara 已提交
5471 5472 5473
			(unsigned long long)off, (unsigned long long)len);
		return -EIO;
	}
5474 5475 5476 5477 5478 5479 5480 5481 5482 5483 5484
	/*
	 * 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;
	}

5485 5486 5487 5488 5489 5490
	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));
5491 5492
	if (IS_ERR(bh))
		return PTR_ERR(bh);
5493 5494
	if (!bh)
		goto out;
5495
	BUFFER_TRACE(bh, "get write access");
5496 5497 5498
	err = ext4_journal_get_write_access(handle, bh);
	if (err) {
		brelse(bh);
5499
		return err;
5500
	}
5501 5502 5503 5504
	lock_buffer(bh);
	memcpy(bh->b_data+offset, data, len);
	flush_dcache_page(bh->b_page);
	unlock_buffer(bh);
5505
	err = ext4_handle_dirty_metadata(handle, NULL, bh);
5506
	brelse(bh);
5507
out:
5508 5509
	if (inode->i_size < off + len) {
		i_size_write(inode, off + len);
5510
		EXT4_I(inode)->i_disksize = inode->i_size;
5511
		ext4_mark_inode_dirty(handle, inode);
5512
	}
5513
	return len;
5514 5515 5516 5517
}

#endif

A
Al Viro 已提交
5518 5519
static struct dentry *ext4_mount(struct file_system_type *fs_type, int flags,
		       const char *dev_name, void *data)
5520
{
A
Al Viro 已提交
5521
	return mount_bdev(fs_type, flags, dev_name, data, ext4_fill_super);
5522 5523
}

J
Jan Kara 已提交
5524
#if !defined(CONFIG_EXT2_FS) && !defined(CONFIG_EXT2_FS_MODULE) && defined(CONFIG_EXT4_USE_FOR_EXT2)
5525 5526 5527 5528 5529 5530 5531 5532 5533 5534 5535 5536
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);
}
5537 5538 5539 5540 5541 5542 5543 5544 5545 5546 5547

static inline int ext2_feature_set_ok(struct super_block *sb)
{
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, ~EXT2_FEATURE_INCOMPAT_SUPP))
		return 0;
	if (sb->s_flags & MS_RDONLY)
		return 1;
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, ~EXT2_FEATURE_RO_COMPAT_SUPP))
		return 0;
	return 1;
}
5548 5549 5550
#else
static inline void register_as_ext2(void) { }
static inline void unregister_as_ext2(void) { }
5551
static inline int ext2_feature_set_ok(struct super_block *sb) { return 0; }
5552 5553 5554 5555 5556 5557 5558 5559 5560 5561 5562 5563 5564 5565
#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);
}
5566 5567 5568 5569 5570 5571 5572 5573 5574 5575 5576 5577 5578

static inline int ext3_feature_set_ok(struct super_block *sb)
{
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, ~EXT3_FEATURE_INCOMPAT_SUPP))
		return 0;
	if (!EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL))
		return 0;
	if (sb->s_flags & MS_RDONLY)
		return 1;
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, ~EXT3_FEATURE_RO_COMPAT_SUPP))
		return 0;
	return 1;
}
5579

T
Theodore Ts'o 已提交
5580 5581 5582
static struct file_system_type ext4_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "ext4",
A
Al Viro 已提交
5583
	.mount		= ext4_mount,
T
Theodore Ts'o 已提交
5584 5585 5586
	.kill_sb	= kill_block_super,
	.fs_flags	= FS_REQUIRES_DEV,
};
5587
MODULE_ALIAS_FS("ext4");
T
Theodore Ts'o 已提交
5588

5589
static int __init ext4_init_feat_adverts(void)
5590 5591 5592 5593 5594 5595 5596 5597 5598 5599 5600 5601 5602 5603 5604 5605 5606 5607 5608 5609 5610 5611 5612
{
	struct ext4_features *ef;
	int ret = -ENOMEM;

	ef = kzalloc(sizeof(struct ext4_features), GFP_KERNEL);
	if (!ef)
		goto out;

	ef->f_kobj.kset = ext4_kset;
	init_completion(&ef->f_kobj_unregister);
	ret = kobject_init_and_add(&ef->f_kobj, &ext4_feat_ktype, NULL,
				   "features");
	if (ret) {
		kfree(ef);
		goto out;
	}

	ext4_feat = ef;
	ret = 0;
out:
	return ret;
}

5613 5614 5615 5616 5617 5618 5619
static void ext4_exit_feat_adverts(void)
{
	kobject_put(&ext4_feat->f_kobj);
	wait_for_completion(&ext4_feat->f_kobj_unregister);
	kfree(ext4_feat);
}

5620 5621 5622 5623
/* 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];

5624
static int __init ext4_init_fs(void)
5625
{
5626
	int i, err;
5627

5628
	ratelimit_state_init(&ext4_mount_msg_ratelimit, 30 * HZ, 64);
5629 5630 5631
	ext4_li_info = NULL;
	mutex_init(&ext4_li_mtx);

5632
	/* Build-time check for flags consistency */
5633
	ext4_check_flag_values();
5634 5635 5636 5637 5638 5639

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

5640
	err = ext4_init_es();
5641 5642
	if (err)
		return err;
5643 5644 5645 5646 5647

	err = ext4_init_pageio();
	if (err)
		goto out7;

5648
	err = ext4_init_system_zone();
5649
	if (err)
5650
		goto out6;
T
Theodore Ts'o 已提交
5651
	ext4_kset = kset_create_and_add("ext4", NULL, fs_kobj);
5652 5653
	if (!ext4_kset) {
		err = -ENOMEM;
T
Theodore Ts'o 已提交
5654
		goto out5;
5655
	}
5656
	ext4_proc_root = proc_mkdir("fs/ext4", NULL);
5657 5658

	err = ext4_init_feat_adverts();
T
Theodore Ts'o 已提交
5659 5660
	if (err)
		goto out4;
5661

5662
	err = ext4_init_mballoc();
5663 5664
	if (err)
		goto out2;
5665 5666
	else
		ext4_mballoc_ready = 1;
5667 5668 5669
	err = init_inodecache();
	if (err)
		goto out1;
5670
	register_as_ext3();
5671
	register_as_ext2();
T
Theodore Ts'o 已提交
5672
	err = register_filesystem(&ext4_fs_type);
5673 5674
	if (err)
		goto out;
5675

5676 5677
	return 0;
out:
5678 5679
	unregister_as_ext2();
	unregister_as_ext3();
5680 5681
	destroy_inodecache();
out1:
5682
	ext4_mballoc_ready = 0;
5683
	ext4_exit_mballoc();
5684
out2:
5685
	ext4_exit_feat_adverts();
T
Theodore Ts'o 已提交
5686
out4:
5687 5688
	if (ext4_proc_root)
		remove_proc_entry("fs/ext4", NULL);
5689
	kset_unregister(ext4_kset);
5690
out5:
5691
	ext4_exit_system_zone();
5692
out6:
5693
	ext4_exit_pageio();
5694 5695 5696
out7:
	ext4_exit_es();

5697 5698 5699
	return err;
}

5700
static void __exit ext4_exit_fs(void)
5701
{
5702
	ext4_exit_crypto();
5703
	ext4_destroy_lazyinit_thread();
5704 5705
	unregister_as_ext2();
	unregister_as_ext3();
T
Theodore Ts'o 已提交
5706
	unregister_filesystem(&ext4_fs_type);
5707
	destroy_inodecache();
5708
	ext4_exit_mballoc();
5709
	ext4_exit_feat_adverts();
5710
	remove_proc_entry("fs/ext4", NULL);
T
Theodore Ts'o 已提交
5711
	kset_unregister(ext4_kset);
5712 5713
	ext4_exit_system_zone();
	ext4_exit_pageio();
5714
	ext4_exit_es();
5715 5716 5717
}

MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others");
5718
MODULE_DESCRIPTION("Fourth Extended Filesystem");
5719
MODULE_LICENSE("GPL");
5720 5721
module_init(ext4_init_fs)
module_exit(ext4_exit_fs)