super.c 161.9 KB
Newer Older
1
/*
2
 *  linux/fs/ext4/super.c
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
 *
 * Copyright (C) 1992, 1993, 1994, 1995
 * Remy Card (card@masi.ibp.fr)
 * Laboratoire MASI - Institut Blaise Pascal
 * Universite Pierre et Marie Curie (Paris VI)
 *
 *  from
 *
 *  linux/fs/minix/inode.c
 *
 *  Copyright (C) 1991, 1992  Linus Torvalds
 *
 *  Big-endian to little-endian byte-swapping/bitmaps by
 *        David S. Miller (davem@caip.rutgers.edu), 1995
 */

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

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

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

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

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

60
static int ext4_load_journal(struct super_block *, struct ext4_super_block *,
61
			     unsigned long journal_devnum);
62
static int ext4_show_options(struct seq_file *seq, struct dentry *root);
63
static int ext4_commit_super(struct super_block *sb, int sync);
64 65 66 67
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);
68
static int ext4_sync_fs(struct super_block *sb, int wait);
69 70
static int ext4_remount(struct super_block *sb, int *flags, char *data);
static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf);
71 72
static int ext4_unfreeze(struct super_block *sb);
static int ext4_freeze(struct super_block *sb);
A
Al Viro 已提交
73 74
static struct dentry *ext4_mount(struct file_system_type *fs_type, int flags,
		       const char *dev_name, void *data);
75 76
static inline int ext2_feature_set_ok(struct super_block *sb);
static inline int ext3_feature_set_ok(struct super_block *sb);
77
static int ext4_feature_set_ok(struct super_block *sb, int readonly);
78 79
static void ext4_destroy_lazyinit_thread(void);
static void ext4_unregister_li_request(struct super_block *sb);
80
static void ext4_clear_request_list(void);
81 82
static struct inode *ext4_get_journal_inode(struct super_block *sb,
					    unsigned int journal_inum);
83

J
Jan Kara 已提交
84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113
/*
 * Lock ordering
 *
 * Note the difference between i_mmap_sem (EXT4_I(inode)->i_mmap_sem) and
 * i_mmap_rwsem (inode->i_mmap_rwsem)!
 *
 * page fault path:
 * mmap_sem -> sb_start_pagefault -> i_mmap_sem (r) -> transaction start ->
 *   page lock -> i_data_sem (rw)
 *
 * buffered write path:
 * sb_start_write -> i_mutex -> mmap_sem
 * sb_start_write -> i_mutex -> transaction start -> page lock ->
 *   i_data_sem (rw)
 *
 * truncate:
 * sb_start_write -> i_mutex -> EXT4_STATE_DIOREAD_LOCK (w) -> i_mmap_sem (w) ->
 *   i_mmap_rwsem (w) -> page lock
 * sb_start_write -> i_mutex -> EXT4_STATE_DIOREAD_LOCK (w) -> i_mmap_sem (w) ->
 *   transaction start -> i_data_sem (rw)
 *
 * direct IO:
 * sb_start_write -> i_mutex -> EXT4_STATE_DIOREAD_LOCK (r) -> mmap_sem
 * sb_start_write -> i_mutex -> EXT4_STATE_DIOREAD_LOCK (r) ->
 *   transaction start -> i_data_sem (rw)
 *
 * writepages:
 * transaction start -> page lock(s) -> i_data_sem (rw)
 */

J
Jan Kara 已提交
114
#if !defined(CONFIG_EXT2_FS) && !defined(CONFIG_EXT2_FS_MODULE) && defined(CONFIG_EXT4_USE_FOR_EXT2)
115 116 117 118 119 120 121
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,
};
122
MODULE_ALIAS_FS("ext2");
123
MODULE_ALIAS("ext2");
124 125 126 127 128 129
#define IS_EXT2_SB(sb) ((sb)->s_bdev->bd_holder == &ext2_fs_type)
#else
#define IS_EXT2_SB(sb) (0)
#endif


130 131 132
static struct file_system_type ext3_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "ext3",
A
Al Viro 已提交
133
	.mount		= ext4_mount,
134 135 136
	.kill_sb	= kill_block_super,
	.fs_flags	= FS_REQUIRES_DEV,
};
137
MODULE_ALIAS_FS("ext3");
138
MODULE_ALIAS("ext3");
139
#define IS_EXT3_SB(sb) ((sb)->s_bdev->bd_holder == &ext3_fs_type)
L
Laurent Vivier 已提交
140

141 142 143
static int ext4_verify_csum_type(struct super_block *sb,
				 struct ext4_super_block *es)
{
144
	if (!ext4_has_feature_metadata_csum(sb))
145 146 147 148 149
		return 1;

	return es->s_checksum_type == EXT4_CRC32C_CHKSUM;
}

150 151 152 153 154 155 156 157 158 159 160 161
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);
}

162 163
static int ext4_superblock_csum_verify(struct super_block *sb,
				       struct ext4_super_block *es)
164
{
165
	if (!ext4_has_metadata_csum(sb))
166 167 168 169 170
		return 1;

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

171
void ext4_superblock_csum_set(struct super_block *sb)
172
{
173 174
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;

175
	if (!ext4_has_metadata_csum(sb))
176 177 178 179 180
		return;

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

181 182 183 184
void *ext4_kvmalloc(size_t size, gfp_t flags)
{
	void *ret;

185
	ret = kmalloc(size, flags | __GFP_NOWARN);
186 187 188 189 190 191 192 193 194
	if (!ret)
		ret = __vmalloc(size, flags, PAGE_KERNEL);
	return ret;
}

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

195
	ret = kzalloc(size, flags | __GFP_NOWARN);
196 197 198 199 200
	if (!ret)
		ret = __vmalloc(size, flags | __GFP_ZERO, PAGE_KERNEL);
	return ret;
}

201 202
ext4_fsblk_t ext4_block_bitmap(struct super_block *sb,
			       struct ext4_group_desc *bg)
L
Laurent Vivier 已提交
203
{
204
	return le32_to_cpu(bg->bg_block_bitmap_lo) |
205
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
206
		 (ext4_fsblk_t)le32_to_cpu(bg->bg_block_bitmap_hi) << 32 : 0);
L
Laurent Vivier 已提交
207 208
}

209 210
ext4_fsblk_t ext4_inode_bitmap(struct super_block *sb,
			       struct ext4_group_desc *bg)
L
Laurent Vivier 已提交
211
{
212
	return le32_to_cpu(bg->bg_inode_bitmap_lo) |
213
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
214
		 (ext4_fsblk_t)le32_to_cpu(bg->bg_inode_bitmap_hi) << 32 : 0);
L
Laurent Vivier 已提交
215 216
}

217 218
ext4_fsblk_t ext4_inode_table(struct super_block *sb,
			      struct ext4_group_desc *bg)
L
Laurent Vivier 已提交
219
{
220
	return le32_to_cpu(bg->bg_inode_table_lo) |
221
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
222
		 (ext4_fsblk_t)le32_to_cpu(bg->bg_inode_table_hi) << 32 : 0);
L
Laurent Vivier 已提交
223 224
}

225 226
__u32 ext4_free_group_clusters(struct super_block *sb,
			       struct ext4_group_desc *bg)
227 228 229
{
	return le16_to_cpu(bg->bg_free_blocks_count_lo) |
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
230
		 (__u32)le16_to_cpu(bg->bg_free_blocks_count_hi) << 16 : 0);
231 232 233 234 235 236 237
}

__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 ?
238
		 (__u32)le16_to_cpu(bg->bg_free_inodes_count_hi) << 16 : 0);
239 240 241 242 243 244 245
}

__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 ?
246
		 (__u32)le16_to_cpu(bg->bg_used_dirs_count_hi) << 16 : 0);
247 248 249 250 251 252 253
}

__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 ?
254
		 (__u32)le16_to_cpu(bg->bg_itable_unused_hi) << 16 : 0);
255 256
}

257 258
void ext4_block_bitmap_set(struct super_block *sb,
			   struct ext4_group_desc *bg, ext4_fsblk_t blk)
L
Laurent Vivier 已提交
259
{
260
	bg->bg_block_bitmap_lo = cpu_to_le32((u32)blk);
261 262
	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
263 264
}

265 266
void ext4_inode_bitmap_set(struct super_block *sb,
			   struct ext4_group_desc *bg, ext4_fsblk_t blk)
L
Laurent Vivier 已提交
267
{
268
	bg->bg_inode_bitmap_lo  = cpu_to_le32((u32)blk);
269 270
	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
271 272
}

273 274
void ext4_inode_table_set(struct super_block *sb,
			  struct ext4_group_desc *bg, ext4_fsblk_t blk)
L
Laurent Vivier 已提交
275
{
276
	bg->bg_inode_table_lo = cpu_to_le32((u32)blk);
277 278
	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
279 280
}

281 282
void ext4_free_group_clusters_set(struct super_block *sb,
				  struct ext4_group_desc *bg, __u32 count)
283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312
{
	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);
}

313

314 315 316 317 318 319
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;
320 321
	if (bdev_read_only(sb->s_bdev))
		return;
322 323 324 325 326 327 328 329 330 331 332 333
	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;
	}
334 335 336 337 338 339
	/*
	 * 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);
340
	le32_add_cpu(&es->s_error_count, 1);
341 342 343 344 345 346 347 348 349
}

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);
}

350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365
/*
 * 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
366 367 368 369 370
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);
371
	struct ext4_journal_cb_entry	*jce;
B
Bobi Jam 已提交
372

373
	BUG_ON(txn->t_state == T_FINISHED);
B
Bobi Jam 已提交
374
	spin_lock(&sbi->s_md_lock);
375 376 377
	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
378 379 380 381 382 383 384
		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);
}
385

386 387 388 389
/* 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
390
 * superblock.  That is not possible on ext4, because we may have other
391 392 393 394 395
 * 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.
 *
396
 * We'll just use the jbd2_journal_abort() error code to record an error in
397
 * the journal instead.  On recovery, the journal will complain about
398 399 400
 * that error until we've noted it down and cleared it.
 */

401
static void ext4_handle_error(struct super_block *sb)
402 403 404 405
{
	if (sb->s_flags & MS_RDONLY)
		return;

406
	if (!test_opt(sb, ERRORS_CONT)) {
407
		journal_t *journal = EXT4_SB(sb)->s_journal;
408

409
		EXT4_SB(sb)->s_mount_flags |= EXT4_MF_FS_ABORTED;
410
		if (journal)
411
			jbd2_journal_abort(journal, -EIO);
412
	}
413
	if (test_opt(sb, ERRORS_RO)) {
414
		ext4_msg(sb, KERN_CRIT, "Remounting filesystem read-only");
415 416 417 418 419
		/*
		 * Make sure updated value of ->s_mount_flags will be visible
		 * before ->s_flags update
		 */
		smp_wmb();
420 421
		sb->s_flags |= MS_RDONLY;
	}
422 423 424 425
	if (test_opt(sb, ERRORS_PANIC)) {
		if (EXT4_SB(sb)->s_journal &&
		  !(EXT4_SB(sb)->s_journal->j_flags & JBD2_REC_ERR))
			return;
426
		panic("EXT4-fs (device %s): panic forced after error\n",
427
			sb->s_id);
428
	}
429 430
}

431 432 433 434
#define ext4_error_ratelimit(sb)					\
		___ratelimit(&(EXT4_SB(sb)->s_err_ratelimit_state),	\
			     "EXT4-fs error")

435
void __ext4_error(struct super_block *sb, const char *function,
436
		  unsigned int line, const char *fmt, ...)
437
{
J
Joe Perches 已提交
438
	struct va_format vaf;
439 440
	va_list args;

441 442 443
	if (unlikely(ext4_forced_shutdown(EXT4_SB(sb))))
		return;

444 445 446 447 448 449 450 451 452
	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);
	}
453
	save_error_info(sb, function, line);
454
	ext4_handle_error(sb);
455 456
}

457 458 459
void __ext4_error_inode(struct inode *inode, const char *function,
			unsigned int line, ext4_fsblk_t block,
			const char *fmt, ...)
460 461
{
	va_list args;
462
	struct va_format vaf;
463
	struct ext4_super_block *es = EXT4_SB(inode->i_sb)->s_es;
464

465 466 467
	if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb))))
		return;

468 469
	es->s_last_error_ino = cpu_to_le32(inode->i_ino);
	es->s_last_error_block = cpu_to_le64(block);
470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485
	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);
	}
486
	save_error_info(inode->i_sb, function, line);
487 488 489
	ext4_handle_error(inode->i_sb);
}

490 491 492
void __ext4_error_file(struct file *file, const char *function,
		       unsigned int line, ext4_fsblk_t block,
		       const char *fmt, ...)
493 494
{
	va_list args;
495
	struct va_format vaf;
496
	struct ext4_super_block *es;
A
Al Viro 已提交
497
	struct inode *inode = file_inode(file);
498 499
	char pathname[80], *path;

500 501 502
	if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb))))
		return;

503 504
	es = EXT4_SB(inode->i_sb)->s_es;
	es->s_last_error_ino = cpu_to_le32(inode->i_ino);
505
	if (ext4_error_ratelimit(inode->i_sb)) {
M
Miklos Szeredi 已提交
506
		path = file_path(file, pathname, sizeof(pathname));
507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525
		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);
	}
526
	save_error_info(inode->i_sb, function, line);
527 528 529
	ext4_handle_error(inode->i_sb);
}

530 531
const char *ext4_decode_error(struct super_block *sb, int errno,
			      char nbuf[16])
532 533 534 535
{
	char *errstr = NULL;

	switch (errno) {
536 537 538 539 540 541
	case -EFSCORRUPTED:
		errstr = "Corrupt filesystem";
		break;
	case -EFSBADCRC:
		errstr = "Filesystem failed CRC";
		break;
542 543 544 545 546 547 548
	case -EIO:
		errstr = "IO failure";
		break;
	case -ENOMEM:
		errstr = "Out of memory";
		break;
	case -EROFS:
549 550
		if (!sb || (EXT4_SB(sb)->s_journal &&
			    EXT4_SB(sb)->s_journal->j_flags & JBD2_ABORT))
551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569
			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;
}

570
/* __ext4_std_error decodes expected errors from journaling functions
571 572
 * automatically and invokes the appropriate error response.  */

573 574
void __ext4_std_error(struct super_block *sb, const char *function,
		      unsigned int line, int errno)
575 576 577 578
{
	char nbuf[16];
	const char *errstr;

579 580 581
	if (unlikely(ext4_forced_shutdown(EXT4_SB(sb))))
		return;

582 583 584 585 586 587 588
	/* 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;

589 590 591 592 593
	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);
	}
594

595
	save_error_info(sb, function, line);
596
	ext4_handle_error(sb);
597 598 599
}

/*
600
 * ext4_abort is a much stronger failure handler than ext4_error.  The
601 602 603 604 605 606 607 608
 * 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.
 */

609
void __ext4_abort(struct super_block *sb, const char *function,
610
		unsigned int line, const char *fmt, ...)
611
{
612
	struct va_format vaf;
613 614
	va_list args;

615 616 617
	if (unlikely(ext4_forced_shutdown(EXT4_SB(sb))))
		return;

618
	save_error_info(sb, function, line);
619
	va_start(args, fmt);
620 621 622 623
	vaf.fmt = fmt;
	vaf.va = &args;
	printk(KERN_CRIT "EXT4-fs error (device %s): %s:%d: %pV\n",
	       sb->s_id, function, line, &vaf);
624 625
	va_end(args);

626 627 628
	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;
629 630 631 632 633 634
		/*
		 * Make sure updated value of ->s_mount_flags will be visible
		 * before ->s_flags update
		 */
		smp_wmb();
		sb->s_flags |= MS_RDONLY;
635 636 637 638
		if (EXT4_SB(sb)->s_journal)
			jbd2_journal_abort(EXT4_SB(sb)->s_journal, -EIO);
		save_error_info(sb, function, line);
	}
639 640 641 642
	if (test_opt(sb, ERRORS_PANIC)) {
		if (EXT4_SB(sb)->s_journal &&
		  !(EXT4_SB(sb)->s_journal->j_flags & JBD2_REC_ERR))
			return;
643
		panic("EXT4-fs panic from previous error\n");
644
	}
645 646
}

647 648
void __ext4_msg(struct super_block *sb,
		const char *prefix, const char *fmt, ...)
649
{
J
Joe Perches 已提交
650
	struct va_format vaf;
651 652
	va_list args;

653 654 655
	if (!___ratelimit(&(EXT4_SB(sb)->s_msg_ratelimit_state), "EXT4-fs"))
		return;

656
	va_start(args, fmt);
J
Joe Perches 已提交
657 658 659
	vaf.fmt = fmt;
	vaf.va = &args;
	printk("%sEXT4-fs (%s): %pV\n", prefix, sb->s_id, &vaf);
660 661 662
	va_end(args);
}

663 664 665 666
#define ext4_warning_ratelimit(sb)					\
		___ratelimit(&(EXT4_SB(sb)->s_warning_ratelimit_state),	\
			     "EXT4-fs warning")

667
void __ext4_warning(struct super_block *sb, const char *function,
668
		    unsigned int line, const char *fmt, ...)
669
{
J
Joe Perches 已提交
670
	struct va_format vaf;
671 672
	va_list args;

673
	if (!ext4_warning_ratelimit(sb))
674 675
		return;

676
	va_start(args, fmt);
J
Joe Perches 已提交
677 678 679 680
	vaf.fmt = fmt;
	vaf.va = &args;
	printk(KERN_WARNING "EXT4-fs warning (device %s): %s:%d: %pV\n",
	       sb->s_id, function, line, &vaf);
681 682 683
	va_end(args);
}

684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701
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);
}

702 703 704 705
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, ...)
706 707 708
__releases(bitlock)
__acquires(bitlock)
{
J
Joe Perches 已提交
709
	struct va_format vaf;
710 711 712
	va_list args;
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;

713 714 715
	if (unlikely(ext4_forced_shutdown(EXT4_SB(sb))))
		return;

716 717 718
	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
719

720 721 722 723 724 725 726 727 728 729 730 731 732 733
	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);
	}
734 735

	if (test_opt(sb, ERRORS_CONT)) {
736
		ext4_commit_super(sb, 0);
737 738
		return;
	}
739

740 741 742 743 744 745 746 747
	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
748
	 * ext4_grp_locked_error() to distinguish between the
749 750 751 752 753 754 755 756
	 * 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;
}

757
void ext4_update_dynamic_rev(struct super_block *sb)
758
{
759
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;
760

761
	if (le32_to_cpu(es->s_rev_level) > EXT4_GOOD_OLD_REV)
762 763
		return;

764
	ext4_warning(sb,
765 766
		     "updating to rev %d because of new feature flag, "
		     "running e2fsck is recommended",
767
		     EXT4_DYNAMIC_REV);
768

769 770 771
	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);
772 773 774 775 776 777 778 779 780 781 782 783 784
	/* 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
 */
785
static struct block_device *ext4_blkdev_get(dev_t dev, struct super_block *sb)
786 787 788 789
{
	struct block_device *bdev;
	char b[BDEVNAME_SIZE];

790
	bdev = blkdev_get_by_dev(dev, FMODE_READ|FMODE_WRITE|FMODE_EXCL, sb);
791 792 793 794 795
	if (IS_ERR(bdev))
		goto fail;
	return bdev;

fail:
796
	ext4_msg(sb, KERN_ERR, "failed to open journal device %s: %ld",
797 798 799 800 801 802 803
			__bdevname(dev, b), PTR_ERR(bdev));
	return NULL;
}

/*
 * Release the journal device
 */
A
Al Viro 已提交
804
static void ext4_blkdev_put(struct block_device *bdev)
805
{
A
Al Viro 已提交
806
	blkdev_put(bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL);
807 808
}

A
Al Viro 已提交
809
static void ext4_blkdev_remove(struct ext4_sb_info *sbi)
810 811 812 813
{
	struct block_device *bdev;
	bdev = sbi->journal_bdev;
	if (bdev) {
A
Al Viro 已提交
814
		ext4_blkdev_put(bdev);
815 816 817 818 819 820
		sbi->journal_bdev = NULL;
	}
}

static inline struct inode *orphan_list_entry(struct list_head *l)
{
821
	return &list_entry(l, struct ext4_inode_info, i_orphan)->vfs_inode;
822 823
}

824
static void dump_orphan_list(struct super_block *sb, struct ext4_sb_info *sbi)
825 826 827
{
	struct list_head *l;

828 829
	ext4_msg(sb, KERN_ERR, "sb orphan head is %d",
		 le32_to_cpu(sbi->s_es->s_last_orphan));
830 831 832 833 834 835 836 837 838 839 840 841

	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));
	}
}

842
static void ext4_put_super(struct super_block *sb)
843
{
844 845
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_super_block *es = sbi->s_es;
846
	int aborted = 0;
847
	int i, err;
848

849
	ext4_unregister_li_request(sb);
850 851
	dquot_disable(sb, -1, DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED);

852 853
	flush_workqueue(sbi->rsv_conversion_wq);
	destroy_workqueue(sbi->rsv_conversion_wq);
854

855
	if (sbi->s_journal) {
856
		aborted = is_journal_aborted(sbi->s_journal);
857 858
		err = jbd2_journal_destroy(sbi->s_journal);
		sbi->s_journal = NULL;
859
		if ((err < 0) && !aborted)
860
			ext4_abort(sb, "Couldn't clean up the journal");
861
	}
862

863
	ext4_unregister_sysfs(sb);
864
	ext4_es_unregister_shrinker(sbi);
865
	del_timer_sync(&sbi->s_err_report);
866 867 868 869
	ext4_release_system_zone(sb);
	ext4_mb_release(sb);
	ext4_ext_release(sb);

870
	if (!(sb->s_flags & MS_RDONLY) && !aborted) {
871
		ext4_clear_feature_journal_needs_recovery(sb);
872 873
		es->s_state = cpu_to_le16(sbi->s_mount_state);
	}
874
	if (!(sb->s_flags & MS_RDONLY))
875 876
		ext4_commit_super(sb, 1);

877 878
	for (i = 0; i < sbi->s_gdb_count; i++)
		brelse(sbi->s_group_desc[i]);
A
Al Viro 已提交
879 880
	kvfree(sbi->s_group_desc);
	kvfree(sbi->s_flex_groups);
881
	percpu_counter_destroy(&sbi->s_freeclusters_counter);
882 883
	percpu_counter_destroy(&sbi->s_freeinodes_counter);
	percpu_counter_destroy(&sbi->s_dirs_counter);
884
	percpu_counter_destroy(&sbi->s_dirtyclusters_counter);
885
	percpu_free_rwsem(&sbi->s_journal_flag_rwsem);
886
#ifdef CONFIG_QUOTA
J
Jan Kara 已提交
887
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
888 889 890 891 892 893 894 895 896 897 898
		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));

899
	sync_blockdev(sb->s_bdev);
900
	invalidate_bdev(sb->s_bdev);
901 902 903 904 905 906 907
	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);
908
		invalidate_bdev(sbi->journal_bdev);
909
		ext4_blkdev_remove(sbi);
910
	}
911 912 913 914
	if (sbi->s_mb_cache) {
		ext4_xattr_destroy_cache(sbi->s_mb_cache);
		sbi->s_mb_cache = NULL;
	}
915 916
	if (sbi->s_mmp_tsk)
		kthread_stop(sbi->s_mmp_tsk);
917
	brelse(sbi->s_sbh);
918
	sb->s_fs_info = NULL;
T
Theodore Ts'o 已提交
919 920 921 922 923 924
	/*
	 * 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);
925 926
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
927
	kfree(sbi->s_blockgroup_lock);
928 929 930
	kfree(sbi);
}

931
static struct kmem_cache *ext4_inode_cachep;
932 933 934 935

/*
 * Called inside transaction, so use GFP_NOFS
 */
936
static struct inode *ext4_alloc_inode(struct super_block *sb)
937
{
938
	struct ext4_inode_info *ei;
939

C
Christoph Lameter 已提交
940
	ei = kmem_cache_alloc(ext4_inode_cachep, GFP_NOFS);
941 942
	if (!ei)
		return NULL;
943

944
	ei->vfs_inode.i_version = 1;
945
	spin_lock_init(&ei->i_raw_lock);
946 947
	INIT_LIST_HEAD(&ei->i_prealloc_list);
	spin_lock_init(&ei->i_prealloc_lock);
Z
Zheng Liu 已提交
948 949
	ext4_es_init_tree(&ei->i_es_tree);
	rwlock_init(&ei->i_es_lock);
950
	INIT_LIST_HEAD(&ei->i_es_list);
951
	ei->i_es_all_nr = 0;
952
	ei->i_es_shk_nr = 0;
953
	ei->i_es_shrink_lblk = 0;
954 955 956
	ei->i_reserved_data_blocks = 0;
	ei->i_reserved_meta_blocks = 0;
	ei->i_allocated_meta_blocks = 0;
957
	ei->i_da_metadata_calc_len = 0;
958
	ei->i_da_metadata_calc_last_lblock = 0;
959
	spin_lock_init(&(ei->i_block_reservation_lock));
960 961
#ifdef CONFIG_QUOTA
	ei->i_reserved_quota = 0;
J
Jan Kara 已提交
962
	memset(&ei->i_dquot, 0, sizeof(ei->i_dquot));
963
#endif
964
	ei->jinode = NULL;
965
	INIT_LIST_HEAD(&ei->i_rsv_conversion_list);
966
	spin_lock_init(&ei->i_completed_io_lock);
967 968
	ei->i_sync_tid = 0;
	ei->i_datasync_tid = 0;
969
	atomic_set(&ei->i_unwritten, 0);
970
	INIT_WORK(&ei->i_rsv_conversion_work, ext4_end_io_rsv_work);
971 972 973
	return &ei->vfs_inode;
}

974 975 976 977 978 979 980 981
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
982 983 984 985 986 987
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));
}

988
static void ext4_destroy_inode(struct inode *inode)
989
{
990
	if (!list_empty(&(EXT4_I(inode)->i_orphan))) {
991 992 993
		ext4_msg(inode->i_sb, KERN_ERR,
			 "Inode %lu (%p): orphan list check failed!",
			 inode->i_ino, EXT4_I(inode));
994 995 996 997 998
		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
999
	call_rcu(&inode->i_rcu, ext4_i_callback);
1000 1001
}

1002
static void init_once(void *foo)
1003
{
1004
	struct ext4_inode_info *ei = (struct ext4_inode_info *) foo;
1005

C
Christoph Lameter 已提交
1006 1007
	INIT_LIST_HEAD(&ei->i_orphan);
	init_rwsem(&ei->xattr_sem);
1008
	init_rwsem(&ei->i_data_sem);
1009
	init_rwsem(&ei->i_mmap_sem);
C
Christoph Lameter 已提交
1010
	inode_init_once(&ei->vfs_inode);
1011 1012
}

1013
static int __init init_inodecache(void)
1014
{
1015 1016
	ext4_inode_cachep = kmem_cache_create("ext4_inode_cache",
					     sizeof(struct ext4_inode_info),
1017
					     0, (SLAB_RECLAIM_ACCOUNT|
1018
						SLAB_MEM_SPREAD|SLAB_ACCOUNT),
1019
					     init_once);
1020
	if (ext4_inode_cachep == NULL)
1021 1022 1023 1024 1025 1026
		return -ENOMEM;
	return 0;
}

static void destroy_inodecache(void)
{
1027 1028 1029 1030 1031
	/*
	 * Make sure all delayed rcu free inodes are flushed before we
	 * destroy cache.
	 */
	rcu_barrier();
1032
	kmem_cache_destroy(ext4_inode_cachep);
1033 1034
}

A
Al Viro 已提交
1035
void ext4_clear_inode(struct inode *inode)
1036
{
A
Al Viro 已提交
1037
	invalidate_inode_buffers(inode);
1038
	clear_inode(inode);
1039
	dquot_drop(inode);
1040
	ext4_discard_preallocations(inode);
1041
	ext4_es_remove_extent(inode, 0, EXT_MAX_BLOCKS);
1042 1043 1044 1045 1046 1047
	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;
	}
1048
#ifdef CONFIG_EXT4_FS_ENCRYPTION
1049
	fscrypt_put_encryption_info(inode, NULL);
1050
#endif
1051 1052
}

C
Christoph Hellwig 已提交
1053
static struct inode *ext4_nfs_get_inode(struct super_block *sb,
1054
					u64 ino, u32 generation)
1055 1056 1057
{
	struct inode *inode;

1058
	if (ino < EXT4_FIRST_INO(sb) && ino != EXT4_ROOT_INO)
1059
		return ERR_PTR(-ESTALE);
1060
	if (ino > le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count))
1061 1062 1063 1064
		return ERR_PTR(-ESTALE);

	/* iget isn't really right if the inode is currently unallocated!!
	 *
1065
	 * ext4_read_inode will return a bad_inode if the inode had been
1066 1067 1068 1069 1070
	 * deleted, so we should be safe.
	 *
	 * Currently we don't know the generation for parent directory, so
	 * a generation of 0 means "accept any"
	 */
1071
	inode = ext4_iget_normal(sb, ino);
1072 1073 1074
	if (IS_ERR(inode))
		return ERR_CAST(inode);
	if (generation && inode->i_generation != generation) {
1075 1076 1077
		iput(inode);
		return ERR_PTR(-ESTALE);
	}
C
Christoph Hellwig 已提交
1078 1079 1080 1081 1082

	return inode;
}

static struct dentry *ext4_fh_to_dentry(struct super_block *sb, struct fid *fid,
1083
					int fh_len, int fh_type)
C
Christoph Hellwig 已提交
1084 1085 1086 1087 1088 1089
{
	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,
1090
					int fh_len, int fh_type)
C
Christoph Hellwig 已提交
1091 1092 1093
{
	return generic_fh_to_parent(sb, fid, fh_len, fh_type,
				    ext4_nfs_get_inode);
1094 1095
}

1096 1097 1098 1099 1100 1101
/*
 * 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.
 */
1102 1103
static int bdev_try_to_free_page(struct super_block *sb, struct page *page,
				 gfp_t wait)
1104 1105 1106 1107 1108 1109 1110 1111
{
	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,
1112
						wait & ~__GFP_DIRECT_RECLAIM);
1113 1114 1115
	return try_to_free_buffers(page);
}

1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130
#ifdef CONFIG_EXT4_FS_ENCRYPTION
static int ext4_get_context(struct inode *inode, void *ctx, size_t len)
{
	return ext4_xattr_get(inode, EXT4_XATTR_INDEX_ENCRYPTION,
				 EXT4_XATTR_NAME_ENCRYPTION_CONTEXT, ctx, len);
}

static int ext4_prepare_context(struct inode *inode)
{
	return ext4_convert_inline_data(inode);
}

static int ext4_set_context(struct inode *inode, const void *ctx, size_t len,
							void *fs_data)
{
1131 1132 1133 1134 1135 1136 1137 1138 1139 1140
	handle_t *handle = fs_data;
	int res, res2, retries = 0;

	/*
	 * If a journal handle was specified, then the encryption context is
	 * being set on a new inode via inheritance and is part of a larger
	 * transaction to create the inode.  Otherwise the encryption context is
	 * being set on an existing inode in its own transaction.  Only in the
	 * latter case should the "retry on ENOSPC" logic be used.
	 */
1141

1142 1143 1144 1145 1146
	if (handle) {
		res = ext4_xattr_set_handle(handle, inode,
					    EXT4_XATTR_INDEX_ENCRYPTION,
					    EXT4_XATTR_NAME_ENCRYPTION_CONTEXT,
					    ctx, len, 0);
1147 1148 1149 1150
		if (!res) {
			ext4_set_inode_flag(inode, EXT4_INODE_ENCRYPT);
			ext4_clear_inode_state(inode,
					EXT4_STATE_MAY_INLINE_DATA);
1151 1152 1153 1154
			/*
			 * Update inode->i_flags - e.g. S_DAX may get disabled
			 */
			ext4_set_inode_flags(inode);
1155 1156 1157 1158
		}
		return res;
	}

1159
retry:
1160 1161 1162 1163 1164
	handle = ext4_journal_start(inode, EXT4_HT_MISC,
			ext4_jbd2_credits_xattr(inode));
	if (IS_ERR(handle))
		return PTR_ERR(handle);

1165 1166 1167
	res = ext4_xattr_set_handle(handle, inode, EXT4_XATTR_INDEX_ENCRYPTION,
				    EXT4_XATTR_NAME_ENCRYPTION_CONTEXT,
				    ctx, len, 0);
1168 1169
	if (!res) {
		ext4_set_inode_flag(inode, EXT4_INODE_ENCRYPT);
1170 1171
		/* Update inode->i_flags - e.g. S_DAX may get disabled */
		ext4_set_inode_flags(inode);
1172 1173 1174 1175 1176
		res = ext4_mark_inode_dirty(handle, inode);
		if (res)
			EXT4_ERROR_INODE(inode, "Failed to mark inode dirty");
	}
	res2 = ext4_journal_stop(handle);
1177 1178 1179

	if (res == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
		goto retry;
1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195
	if (!res)
		res = res2;
	return res;
}

static int ext4_dummy_context(struct inode *inode)
{
	return DUMMY_ENCRYPTION_ENABLED(EXT4_SB(inode->i_sb));
}

static unsigned ext4_max_namelen(struct inode *inode)
{
	return S_ISLNK(inode->i_mode) ? inode->i_sb->s_blocksize :
		EXT4_NAME_LEN;
}

1196
static const struct fscrypt_operations ext4_cryptops = {
1197
	.key_prefix		= "ext4:",
1198 1199 1200 1201 1202 1203 1204 1205 1206
	.get_context		= ext4_get_context,
	.prepare_context	= ext4_prepare_context,
	.set_context		= ext4_set_context,
	.dummy_context		= ext4_dummy_context,
	.is_encrypted		= ext4_encrypted_inode,
	.empty_dir		= ext4_empty_dir,
	.max_namelen		= ext4_max_namelen,
};
#else
1207
static const struct fscrypt_operations ext4_cryptops = {
1208 1209 1210 1211
	.is_encrypted		= ext4_encrypted_inode,
};
#endif

1212
#ifdef CONFIG_QUOTA
L
Li Xi 已提交
1213 1214
static char *quotatypes[] = INITQFNAMES;
#define QTYPE2NAME(t) (quotatypes[t])
1215

1216 1217 1218 1219 1220
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);
1221
static int ext4_quota_on(struct super_block *sb, int type, int format_id,
A
Al Viro 已提交
1222
			 const struct path *path);
1223
static int ext4_quota_off(struct super_block *sb, int type);
1224 1225
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,
1226
			       size_t len, loff_t off);
1227
static ssize_t ext4_quota_write(struct super_block *sb, int type,
1228
				const char *data, size_t len, loff_t off);
1229 1230 1231
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);
1232
static int ext4_get_next_id(struct super_block *sb, struct kqid *qid);
1233

J
Jan Kara 已提交
1234 1235 1236 1237 1238
static struct dquot **ext4_get_dquots(struct inode *inode)
{
	return EXT4_I(inode)->i_dquot;
}

1239
static const struct dquot_operations ext4_quota_operations = {
1240
	.get_reserved_space = ext4_get_reserved_space,
1241 1242 1243 1244
	.write_dquot	= ext4_write_dquot,
	.acquire_dquot	= ext4_acquire_dquot,
	.release_dquot	= ext4_release_dquot,
	.mark_dirty	= ext4_mark_dquot_dirty,
1245 1246 1247
	.write_info	= ext4_write_info,
	.alloc_dquot	= dquot_alloc,
	.destroy_dquot	= dquot_destroy,
L
Li Xi 已提交
1248
	.get_projid	= ext4_get_projid,
1249
	.get_next_id	= ext4_get_next_id,
1250 1251
};

1252
static const struct quotactl_ops ext4_qctl_operations = {
1253
	.quota_on	= ext4_quota_on,
1254
	.quota_off	= ext4_quota_off,
1255
	.quota_sync	= dquot_quota_sync,
1256
	.get_state	= dquot_get_state,
1257 1258
	.set_info	= dquot_set_dqinfo,
	.get_dqblk	= dquot_get_dqblk,
1259 1260
	.set_dqblk	= dquot_set_dqblk,
	.get_nextdqblk	= dquot_get_next_dqblk,
1261 1262 1263
};
#endif

1264
static const struct super_operations ext4_sops = {
1265 1266 1267 1268
	.alloc_inode	= ext4_alloc_inode,
	.destroy_inode	= ext4_destroy_inode,
	.write_inode	= ext4_write_inode,
	.dirty_inode	= ext4_dirty_inode,
1269
	.drop_inode	= ext4_drop_inode,
A
Al Viro 已提交
1270
	.evict_inode	= ext4_evict_inode,
1271 1272
	.put_super	= ext4_put_super,
	.sync_fs	= ext4_sync_fs,
1273 1274
	.freeze_fs	= ext4_freeze,
	.unfreeze_fs	= ext4_unfreeze,
1275 1276 1277
	.statfs		= ext4_statfs,
	.remount_fs	= ext4_remount,
	.show_options	= ext4_show_options,
1278
#ifdef CONFIG_QUOTA
1279 1280
	.quota_read	= ext4_quota_read,
	.quota_write	= ext4_quota_write,
J
Jan Kara 已提交
1281
	.get_dquots	= ext4_get_dquots,
1282
#endif
1283
	.bdev_try_to_free_page = bdev_try_to_free_page,
1284 1285
};

1286
static const struct export_operations ext4_export_ops = {
C
Christoph Hellwig 已提交
1287 1288
	.fh_to_dentry = ext4_fh_to_dentry,
	.fh_to_parent = ext4_fh_to_parent,
1289
	.get_parent = ext4_get_parent,
1290 1291 1292 1293 1294
};

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,
1295
	Opt_nouid32, Opt_debug, Opt_removed,
1296
	Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl,
1297
	Opt_auto_da_alloc, Opt_noauto_da_alloc, Opt_noload,
1298 1299
	Opt_commit, Opt_min_batch_time, Opt_max_batch_time, Opt_journal_dev,
	Opt_journal_path, Opt_journal_checksum, Opt_journal_async_commit,
1300
	Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback,
1301
	Opt_data_err_abort, Opt_data_err_ignore, Opt_test_dummy_encryption,
1302
	Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota,
J
Jan Kara 已提交
1303
	Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_jqfmt_vfsv1, Opt_quota,
T
Theodore Ts'o 已提交
1304
	Opt_noquota, Opt_barrier, Opt_nobarrier, Opt_err,
1305
	Opt_usrquota, Opt_grpquota, Opt_prjquota, Opt_i_version, Opt_dax,
1306
	Opt_stripe, Opt_delalloc, Opt_nodelalloc, Opt_mblk_io_submit,
1307
	Opt_lazytime, Opt_nolazytime, Opt_debug_want_extra_isize,
1308
	Opt_nomblk_io_submit, Opt_block_validity, Opt_noblock_validity,
1309
	Opt_inode_readahead_blks, Opt_journal_ioprio,
1310
	Opt_dioread_nolock, Opt_dioread_lock,
1311
	Opt_discard, Opt_nodiscard, Opt_init_itable, Opt_noinit_itable,
1312
	Opt_max_dir_size_kb, Opt_nojournal_checksum,
1313 1314
};

1315
static const match_table_t tokens = {
1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329
	{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"},
1330 1331
	{Opt_removed, "oldalloc"},
	{Opt_removed, "orlov"},
1332 1333 1334 1335
	{Opt_user_xattr, "user_xattr"},
	{Opt_nouser_xattr, "nouser_xattr"},
	{Opt_acl, "acl"},
	{Opt_noacl, "noacl"},
1336
	{Opt_noload, "norecovery"},
1337
	{Opt_noload, "noload"},
1338 1339
	{Opt_removed, "nobh"},
	{Opt_removed, "bh"},
1340
	{Opt_commit, "commit=%u"},
1341 1342
	{Opt_min_batch_time, "min_batch_time=%u"},
	{Opt_max_batch_time, "max_batch_time=%u"},
1343
	{Opt_journal_dev, "journal_dev=%u"},
1344
	{Opt_journal_path, "journal_path=%s"},
1345
	{Opt_journal_checksum, "journal_checksum"},
1346
	{Opt_nojournal_checksum, "nojournal_checksum"},
1347
	{Opt_journal_async_commit, "journal_async_commit"},
1348 1349 1350 1351
	{Opt_abort, "abort"},
	{Opt_data_journal, "data=journal"},
	{Opt_data_ordered, "data=ordered"},
	{Opt_data_writeback, "data=writeback"},
1352 1353
	{Opt_data_err_abort, "data_err=abort"},
	{Opt_data_err_ignore, "data_err=ignore"},
1354 1355 1356 1357 1358 1359
	{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
1360
	{Opt_jqfmt_vfsv1, "jqfmt=vfsv1"},
1361 1362 1363 1364
	{Opt_grpquota, "grpquota"},
	{Opt_noquota, "noquota"},
	{Opt_quota, "quota"},
	{Opt_usrquota, "usrquota"},
1365
	{Opt_prjquota, "prjquota"},
1366
	{Opt_barrier, "barrier=%u"},
T
Theodore Ts'o 已提交
1367 1368
	{Opt_barrier, "barrier"},
	{Opt_nobarrier, "nobarrier"},
1369
	{Opt_i_version, "i_version"},
R
Ross Zwisler 已提交
1370
	{Opt_dax, "dax"},
1371
	{Opt_stripe, "stripe=%u"},
1372
	{Opt_delalloc, "delalloc"},
1373 1374
	{Opt_lazytime, "lazytime"},
	{Opt_nolazytime, "nolazytime"},
1375
	{Opt_debug_want_extra_isize, "debug_want_extra_isize=%u"},
1376
	{Opt_nodelalloc, "nodelalloc"},
1377 1378
	{Opt_removed, "mblk_io_submit"},
	{Opt_removed, "nomblk_io_submit"},
1379 1380
	{Opt_block_validity, "block_validity"},
	{Opt_noblock_validity, "noblock_validity"},
1381
	{Opt_inode_readahead_blks, "inode_readahead_blks=%u"},
1382
	{Opt_journal_ioprio, "journal_ioprio=%u"},
1383
	{Opt_auto_da_alloc, "auto_da_alloc=%u"},
T
Theodore Ts'o 已提交
1384 1385
	{Opt_auto_da_alloc, "auto_da_alloc"},
	{Opt_noauto_da_alloc, "noauto_da_alloc"},
1386 1387
	{Opt_dioread_nolock, "dioread_nolock"},
	{Opt_dioread_lock, "dioread_lock"},
1388 1389
	{Opt_discard, "discard"},
	{Opt_nodiscard, "nodiscard"},
1390 1391 1392
	{Opt_init_itable, "init_itable=%u"},
	{Opt_init_itable, "init_itable"},
	{Opt_noinit_itable, "noinit_itable"},
1393
	{Opt_max_dir_size_kb, "max_dir_size_kb=%u"},
1394
	{Opt_test_dummy_encryption, "test_dummy_encryption"},
1395 1396 1397 1398 1399
	{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
1400
	{Opt_err, NULL},
1401 1402
};

1403
static ext4_fsblk_t get_sb_block(void **data)
1404
{
1405
	ext4_fsblk_t	sb_block;
1406 1407 1408 1409
	char		*options = (char *) *data;

	if (!options || strncmp(options, "sb=", 3) != 0)
		return 1;	/* Default location */
1410

1411
	options += 3;
1412
	/* TODO: use simple_strtoll with >32bit ext4 */
1413 1414
	sb_block = simple_strtoul(options, &options, 0);
	if (*options && *options != ',') {
1415
		printk(KERN_ERR "EXT4-fs: Invalid sb specification: %s\n",
1416 1417 1418 1419 1420 1421
		       (char *) *data);
		return 1;
	}
	if (*options == ',')
		options++;
	*data = (void *) options;
1422

1423 1424 1425
	return sb_block;
}

1426
#define DEFAULT_JOURNAL_IOPRIO (IOPRIO_PRIO_VALUE(IOPRIO_CLASS_BE, 3))
1427 1428
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";
1429

D
Dmitry Monakhov 已提交
1430 1431 1432 1433 1434
#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;
1435
	int ret = -1;
D
Dmitry Monakhov 已提交
1436 1437 1438 1439 1440 1441

	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");
1442
		return -1;
D
Dmitry Monakhov 已提交
1443
	}
1444
	if (ext4_has_feature_quota(sb)) {
1445 1446 1447
		ext4_msg(sb, KERN_INFO, "Journaled quota options "
			 "ignored when QUOTA feature is enabled");
		return 1;
1448
	}
D
Dmitry Monakhov 已提交
1449 1450 1451 1452
	qname = match_strdup(args);
	if (!qname) {
		ext4_msg(sb, KERN_ERR,
			"Not enough memory for storing quotafile name");
1453
		return -1;
D
Dmitry Monakhov 已提交
1454
	}
1455 1456 1457 1458 1459 1460 1461 1462
	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
1463
	}
1464
	if (strchr(qname, '/')) {
D
Dmitry Monakhov 已提交
1465 1466
		ext4_msg(sb, KERN_ERR,
			"quotafile must be on filesystem root");
1467
		goto errout;
D
Dmitry Monakhov 已提交
1468
	}
1469
	sbi->s_qf_names[qtype] = qname;
1470
	set_opt(sb, QUOTA);
D
Dmitry Monakhov 已提交
1471
	return 1;
1472 1473 1474
errout:
	kfree(qname);
	return ret;
D
Dmitry Monakhov 已提交
1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485
}

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");
1486
		return -1;
D
Dmitry Monakhov 已提交
1487
	}
1488
	kfree(sbi->s_qf_names[qtype]);
D
Dmitry Monakhov 已提交
1489 1490 1491 1492 1493
	sbi->s_qf_names[qtype] = NULL;
	return 1;
}
#endif

1494 1495 1496 1497 1498 1499
#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
1500
#ifdef CONFIG_QUOTA
1501 1502 1503 1504 1505
#define MOPT_Q		0
#define MOPT_QFMT	0x0040
#else
#define MOPT_Q		MOPT_NOSUPPORT
#define MOPT_QFMT	MOPT_NOSUPPORT
1506
#endif
1507
#define MOPT_DATAJ	0x0080
1508 1509 1510
#define MOPT_NO_EXT2	0x0100
#define MOPT_NO_EXT3	0x0200
#define MOPT_EXT4_ONLY	(MOPT_NO_EXT2 | MOPT_NO_EXT3)
1511
#define MOPT_STRING	0x0400
1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523

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},
1524 1525 1526 1527
	{Opt_dioread_nolock, EXT4_MOUNT_DIOREAD_NOLOCK,
	 MOPT_EXT4_ONLY | MOPT_SET},
	{Opt_dioread_lock, EXT4_MOUNT_DIOREAD_NOLOCK,
	 MOPT_EXT4_ONLY | MOPT_CLEAR},
1528 1529
	{Opt_discard, EXT4_MOUNT_DISCARD, MOPT_SET},
	{Opt_nodiscard, EXT4_MOUNT_DISCARD, MOPT_CLEAR},
1530 1531 1532
	{Opt_delalloc, EXT4_MOUNT_DELALLOC,
	 MOPT_EXT4_ONLY | MOPT_SET | MOPT_EXPLICIT},
	{Opt_nodelalloc, EXT4_MOUNT_DELALLOC,
1533
	 MOPT_EXT4_ONLY | MOPT_CLEAR},
1534 1535
	{Opt_nojournal_checksum, EXT4_MOUNT_JOURNAL_CHECKSUM,
	 MOPT_EXT4_ONLY | MOPT_CLEAR},
1536
	{Opt_journal_checksum, EXT4_MOUNT_JOURNAL_CHECKSUM,
1537
	 MOPT_EXT4_ONLY | MOPT_SET | MOPT_EXPLICIT},
1538
	{Opt_journal_async_commit, (EXT4_MOUNT_JOURNAL_ASYNC_COMMIT |
1539
				    EXT4_MOUNT_JOURNAL_CHECKSUM),
1540
	 MOPT_EXT4_ONLY | MOPT_SET | MOPT_EXPLICIT},
1541
	{Opt_noload, EXT4_MOUNT_NOLOAD, MOPT_NO_EXT2 | MOPT_SET},
1542 1543 1544
	{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},
1545
	{Opt_data_err_abort, EXT4_MOUNT_DATA_ERR_ABORT,
1546
	 MOPT_NO_EXT2},
1547
	{Opt_data_err_ignore, EXT4_MOUNT_DATA_ERR_ABORT,
1548
	 MOPT_NO_EXT2},
1549 1550 1551 1552 1553 1554 1555 1556 1557 1558
	{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
1559
	{Opt_dax, EXT4_MOUNT_DAX, MOPT_SET},
1560
	{Opt_stripe, 0, MOPT_GTE0},
1561 1562
	{Opt_resuid, 0, MOPT_GTE0},
	{Opt_resgid, 0, MOPT_GTE0},
1563 1564 1565
	{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},
1566 1567 1568 1569
	{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},
1570 1571
	{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
1572
#ifdef CONFIG_EXT4_FS_POSIX_ACL
1573 1574
	{Opt_acl, EXT4_MOUNT_POSIX_ACL, MOPT_SET},
	{Opt_noacl, EXT4_MOUNT_POSIX_ACL, MOPT_CLEAR},
1575
#else
1576 1577
	{Opt_acl, 0, MOPT_NOSUPPORT},
	{Opt_noacl, 0, MOPT_NOSUPPORT},
1578
#endif
1579 1580
	{Opt_nouid32, EXT4_MOUNT_NO_UID32, MOPT_SET},
	{Opt_debug, EXT4_MOUNT_DEBUG, MOPT_SET},
1581
	{Opt_debug_want_extra_isize, 0, MOPT_GTE0},
1582 1583 1584 1585 1586
	{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},
1587 1588
	{Opt_prjquota, EXT4_MOUNT_QUOTA | EXT4_MOUNT_PRJQUOTA,
							MOPT_SET | MOPT_Q},
1589
	{Opt_noquota, (EXT4_MOUNT_QUOTA | EXT4_MOUNT_USRQUOTA |
1590 1591
		       EXT4_MOUNT_GRPQUOTA | EXT4_MOUNT_PRJQUOTA),
							MOPT_CLEAR | MOPT_Q},
1592 1593 1594 1595 1596 1597 1598
	{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},
1599
	{Opt_max_dir_size_kb, 0, MOPT_GTE0},
1600
	{Opt_test_dummy_encryption, 0, MOPT_GTE0},
1601 1602 1603 1604 1605 1606 1607 1608 1609
	{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;
1610 1611
	kuid_t uid;
	kgid_t gid;
1612 1613
	int arg = 0;

1614 1615 1616 1617 1618 1619 1620 1621 1622 1623
#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
1624
	switch (token) {
1625 1626 1627 1628
	case Opt_noacl:
	case Opt_nouser_xattr:
		ext4_msg(sb, KERN_WARNING, deprecated_msg, opt, "3.5");
		break;
1629 1630 1631
	case Opt_sb:
		return 1;	/* handled by get_sb_block() */
	case Opt_removed:
1632
		ext4_msg(sb, KERN_WARNING, "Ignoring removed %s option", opt);
1633 1634 1635 1636 1637 1638 1639
		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;
1640 1641 1642 1643 1644 1645
	case Opt_lazytime:
		sb->s_flags |= MS_LAZYTIME;
		return 1;
	case Opt_nolazytime:
		sb->s_flags &= ~MS_LAZYTIME;
		return 1;
1646 1647
	}

1648 1649 1650 1651 1652 1653 1654 1655 1656 1657
	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;
	}

1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668
	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;
	}

1669
	if (args->from && !(m->flags & MOPT_STRING) && match_int(args, &arg))
1670 1671 1672
		return -1;
	if (args->from && (m->flags & MOPT_GTE0) && (arg < 0))
		return -1;
1673 1674 1675
	if (m->flags & MOPT_EXPLICIT) {
		if (m->mount_opt & EXT4_MOUNT_DELALLOC) {
			set_opt2(sb, EXPLICIT_DELALLOC);
1676 1677
		} else if (m->mount_opt & EXT4_MOUNT_JOURNAL_CHECKSUM) {
			set_opt2(sb, EXPLICIT_JOURNAL_CHECKSUM);
1678 1679 1680
		} else
			return -1;
	}
1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694
	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;
1695 1696
	} else if (token == Opt_debug_want_extra_isize) {
		sbi->s_want_extra_isize = arg;
1697 1698 1699 1700 1701
	} 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) {
1702 1703 1704 1705
		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");
1706
			return -1;
1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721
		}
		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);
1722 1723
			return -1;
		}
1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738
		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;
1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764
	} 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;
		}

1765
		journal_inode = d_inode(path.dentry);
1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776
		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);
1777 1778 1779 1780 1781 1782 1783 1784
	} 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);
1785 1786 1787 1788 1789 1790 1791 1792 1793
	} 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
1794 1795 1796 1797 1798
	} 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) {
1799
				ext4_msg(sb, KERN_ERR,
1800 1801
					 "Cannot change data mode on remount");
				return -1;
1802
			}
1803
		} else {
1804 1805
			clear_opt(sb, DATA_FLAGS);
			sbi->s_mount_opt |= m->mount_opt;
1806
		}
1807 1808 1809 1810 1811 1812 1813 1814
#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;
		}
1815
		if (ext4_has_feature_quota(sb)) {
1816 1817
			ext4_msg(sb, KERN_INFO,
				 "Quota format mount options ignored "
1818
				 "when QUOTA feature is enabled");
1819
			return 1;
1820
		}
1821
		sbi->s_jquota_fmt = m->mount_opt;
R
Ross Zwisler 已提交
1822 1823
#endif
	} else if (token == Opt_dax) {
1824 1825 1826 1827 1828
#ifdef CONFIG_FS_DAX
		ext4_msg(sb, KERN_WARNING,
		"DAX enabled. Warning: EXPERIMENTAL, use at your own risk");
			sbi->s_mount_opt |= m->mount_opt;
#else
R
Ross Zwisler 已提交
1829 1830
		ext4_msg(sb, KERN_INFO, "dax option not supported");
		return -1;
1831
#endif
1832 1833 1834 1835
	} else if (token == Opt_data_err_abort) {
		sbi->s_mount_opt |= m->mount_opt;
	} else if (token == Opt_data_err_ignore) {
		sbi->s_mount_opt &= ~m->mount_opt;
1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850
	} 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;
1851
	}
1852
	return 1;
1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874
}

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.
		 */
1875
		args[0].to = args[0].from = NULL;
1876 1877 1878 1879
		token = match_token(p, tokens, args);
		if (handle_mount_opt(sb, p, token, args, journal_devnum,
				     journal_ioprio, is_remount) < 0)
			return 0;
1880 1881
	}
#ifdef CONFIG_QUOTA
1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892
	/*
	 * We do the test below only for project quotas. 'usrquota' and
	 * 'grpquota' mount options are allowed even without quota feature
	 * to support legacy quotas in quota files.
	 */
	if (test_opt(sb, PRJQUOTA) && !ext4_has_feature_project(sb)) {
		ext4_msg(sb, KERN_ERR, "Project quota feature not enabled. "
			 "Cannot enable project quota enforcement.");
		return 0;
	}
	if (sbi->s_qf_names[USRQUOTA] || sbi->s_qf_names[GRPQUOTA]) {
1893
		if (test_opt(sb, USRQUOTA) && sbi->s_qf_names[USRQUOTA])
1894
			clear_opt(sb, USRQUOTA);
1895

1896
		if (test_opt(sb, GRPQUOTA) && sbi->s_qf_names[GRPQUOTA])
1897
			clear_opt(sb, GRPQUOTA);
1898

D
Dmitry Monakhov 已提交
1899
		if (test_opt(sb, GRPQUOTA) || test_opt(sb, USRQUOTA)) {
1900 1901
			ext4_msg(sb, KERN_ERR, "old and new quota "
					"format mixing");
1902 1903 1904 1905
			return 0;
		}

		if (!sbi->s_jquota_fmt) {
1906 1907
			ext4_msg(sb, KERN_ERR, "journaled quota format "
					"not specified");
1908 1909 1910 1911
			return 0;
		}
	}
#endif
J
Jan Kara 已提交
1912 1913 1914 1915
	if (test_opt(sb, DIOREAD_NOLOCK)) {
		int blocksize =
			BLOCK_SIZE << le32_to_cpu(sbi->s_es->s_log_block_size);

1916
		if (blocksize < PAGE_SIZE) {
J
Jan Kara 已提交
1917 1918 1919 1920 1921
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "dioread_nolock if block size != PAGE_SIZE");
			return 0;
		}
	}
1922 1923 1924
	return 1;
}

1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948
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])
1949
		seq_show_option(seq, "usrjquota", sbi->s_qf_names[USRQUOTA]);
1950 1951

	if (sbi->s_qf_names[GRPQUOTA])
1952
		seq_show_option(seq, "grpjquota", sbi->s_qf_names[GRPQUOTA]);
1953 1954 1955
#endif
}

1956 1957
static const char *token2str(int token)
{
1958
	const struct match_token *t;
1959 1960 1961 1962 1963 1964 1965

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

1966 1967 1968 1969 1970
/*
 * Show an option if
 *  - it's set to a non-default value OR
 *  - if the per-sb default is different from the global default
 */
1971 1972
static int _ext4_show_options(struct seq_file *seq, struct super_block *sb,
			      int nodefs)
1973 1974 1975
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_super_block *es = sbi->s_es;
1976
	int def_errors, def_mount_opt = nodefs ? 0 : sbi->s_def_mount_opt;
1977
	const struct mount_opts *m;
1978
	char sep = nodefs ? '\n' : ',';
1979

1980 1981
#define SEQ_OPTS_PUTS(str) seq_printf(seq, "%c" str, sep)
#define SEQ_OPTS_PRINT(str, arg) seq_printf(seq, "%c" str, sep, arg)
1982 1983

	if (sbi->s_sb_block != 1)
1984 1985 1986 1987 1988 1989 1990
		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;
1991
		if (!(m->mount_opt & (sbi->s_mount_opt ^ def_mount_opt)))
1992 1993 1994 1995 1996 1997
			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));
1998
	}
1999

2000
	if (nodefs || !uid_eq(sbi->s_resuid, make_kuid(&init_user_ns, EXT4_DEF_RESUID)) ||
2001
	    le16_to_cpu(es->s_def_resuid) != EXT4_DEF_RESUID)
2002 2003 2004
		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)) ||
2005
	    le16_to_cpu(es->s_def_resgid) != EXT4_DEF_RESGID)
2006 2007
		SEQ_OPTS_PRINT("resgid=%u",
				from_kgid_munged(&init_user_ns, sbi->s_resgid));
2008
	def_errors = nodefs ? -1 : le16_to_cpu(es->s_errors);
2009 2010
	if (test_opt(sb, ERRORS_RO) && def_errors != EXT4_ERRORS_RO)
		SEQ_OPTS_PUTS("errors=remount-ro");
2011
	if (test_opt(sb, ERRORS_CONT) && def_errors != EXT4_ERRORS_CONTINUE)
2012
		SEQ_OPTS_PUTS("errors=continue");
2013
	if (test_opt(sb, ERRORS_PANIC) && def_errors != EXT4_ERRORS_PANIC)
2014
		SEQ_OPTS_PUTS("errors=panic");
2015
	if (nodefs || sbi->s_commit_interval != JBD2_DEFAULT_MAX_COMMIT_AGE*HZ)
2016
		SEQ_OPTS_PRINT("commit=%lu", sbi->s_commit_interval / HZ);
2017
	if (nodefs || sbi->s_min_batch_time != EXT4_DEF_MIN_BATCH_TIME)
2018
		SEQ_OPTS_PRINT("min_batch_time=%u", sbi->s_min_batch_time);
2019
	if (nodefs || sbi->s_max_batch_time != EXT4_DEF_MAX_BATCH_TIME)
2020
		SEQ_OPTS_PRINT("max_batch_time=%u", sbi->s_max_batch_time);
2021
	if (sb->s_flags & MS_I_VERSION)
2022
		SEQ_OPTS_PUTS("i_version");
2023
	if (nodefs || sbi->s_stripe)
2024
		SEQ_OPTS_PRINT("stripe=%lu", sbi->s_stripe);
2025
	if (EXT4_MOUNT_DATA_FLAGS & (sbi->s_mount_opt ^ def_mount_opt)) {
2026 2027 2028 2029 2030 2031 2032
		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");
	}
2033 2034
	if (nodefs ||
	    sbi->s_inode_readahead_blks != EXT4_DEF_INODE_READAHEAD_BLKS)
2035 2036
		SEQ_OPTS_PRINT("inode_readahead_blks=%u",
			       sbi->s_inode_readahead_blks);
2037

2038 2039
	if (nodefs || (test_opt(sb, INIT_INODE_TABLE) &&
		       (sbi->s_li_wait_mult != EXT4_DEF_LI_WAIT_MULT)))
2040
		SEQ_OPTS_PRINT("init_itable=%u", sbi->s_li_wait_mult);
2041 2042
	if (nodefs || sbi->s_max_dir_size_kb)
		SEQ_OPTS_PRINT("max_dir_size_kb=%u", sbi->s_max_dir_size_kb);
2043 2044
	if (test_opt(sb, DATA_ERR_ABORT))
		SEQ_OPTS_PUTS("data_err=abort");
2045 2046 2047 2048 2049

	ext4_show_quota_options(seq, sb);
	return 0;
}

2050 2051 2052 2053 2054
static int ext4_show_options(struct seq_file *seq, struct dentry *root)
{
	return _ext4_show_options(seq, root->d_sb, 0);
}

2055
int ext4_seq_options_show(struct seq_file *seq, void *offset)
2056 2057 2058 2059 2060 2061 2062 2063 2064 2065
{
	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;
}

2066
static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es,
2067 2068
			    int read_only)
{
2069
	struct ext4_sb_info *sbi = EXT4_SB(sb);
2070 2071
	int res = 0;

2072
	if (le32_to_cpu(es->s_rev_level) > EXT4_MAX_SUPP_REV) {
2073 2074
		ext4_msg(sb, KERN_ERR, "revision level too high, "
			 "forcing read-only mode");
2075 2076 2077
		res = MS_RDONLY;
	}
	if (read_only)
2078
		goto done;
2079
	if (!(sbi->s_mount_state & EXT4_VALID_FS))
2080 2081
		ext4_msg(sb, KERN_WARNING, "warning: mounting unchecked fs, "
			 "running e2fsck is recommended");
2082
	else if (sbi->s_mount_state & EXT4_ERROR_FS)
2083 2084 2085
		ext4_msg(sb, KERN_WARNING,
			 "warning: mounting fs with errors, "
			 "running e2fsck is recommended");
2086
	else if ((__s16) le16_to_cpu(es->s_max_mnt_count) > 0 &&
2087 2088
		 le16_to_cpu(es->s_mnt_count) >=
		 (unsigned short) (__s16) le16_to_cpu(es->s_max_mnt_count))
2089 2090 2091
		ext4_msg(sb, KERN_WARNING,
			 "warning: maximal mount count reached, "
			 "running e2fsck is recommended");
2092 2093 2094
	else if (le32_to_cpu(es->s_checkinterval) &&
		(le32_to_cpu(es->s_lastcheck) +
			le32_to_cpu(es->s_checkinterval) <= get_seconds()))
2095 2096 2097
		ext4_msg(sb, KERN_WARNING,
			 "warning: checktime reached, "
			 "running e2fsck is recommended");
2098
	if (!sbi->s_journal)
2099
		es->s_state &= cpu_to_le16(~EXT4_VALID_FS);
2100
	if (!(__s16) le16_to_cpu(es->s_max_mnt_count))
2101
		es->s_max_mnt_count = cpu_to_le16(EXT4_DFL_MAX_MNT_COUNT);
M
Marcin Slusarz 已提交
2102
	le16_add_cpu(&es->s_mnt_count, 1);
2103
	es->s_mtime = cpu_to_le32(get_seconds());
2104
	ext4_update_dynamic_rev(sb);
2105
	if (sbi->s_journal)
2106
		ext4_set_feature_journal_needs_recovery(sb);
2107

2108
	ext4_commit_super(sb, 1);
2109
done:
2110
	if (test_opt(sb, DEBUG))
2111
		printk(KERN_INFO "[EXT4 FS bs=%lu, gc=%u, "
2112
				"bpg=%lu, ipg=%lu, mo=%04x, mo2=%04x]\n",
2113 2114
			sb->s_blocksize,
			sbi->s_groups_count,
2115 2116
			EXT4_BLOCKS_PER_GROUP(sb),
			EXT4_INODES_PER_GROUP(sb),
2117
			sbi->s_mount_opt, sbi->s_mount_opt2);
2118

D
Dan Magenheimer 已提交
2119
	cleancache_init_fs(sb);
2120 2121 2122
	return res;
}

2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147
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
2148
		kvfree(sbi->s_flex_groups);
2149 2150 2151 2152 2153 2154
	}
	sbi->s_flex_groups = new_groups;
	sbi->s_flex_groups_allocated = size / sizeof(struct flex_groups);
	return 0;
}

2155 2156 2157 2158 2159
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;
2160
	int i, err;
2161

2162
	sbi->s_log_groups_per_flex = sbi->s_es->s_log_groups_per_flex;
2163
	if (sbi->s_log_groups_per_flex < 1 || sbi->s_log_groups_per_flex > 31) {
2164 2165 2166 2167
		sbi->s_log_groups_per_flex = 0;
		return 1;
	}

2168 2169
	err = ext4_alloc_flex_bg_array(sb, sbi->s_groups_count);
	if (err)
2170
		goto failed;
2171 2172

	for (i = 0; i < sbi->s_groups_count; i++) {
2173
		gdp = ext4_get_group_desc(sb, i, NULL);
2174 2175

		flex_group = ext4_flex_group(sbi, i);
2176 2177
		atomic_add(ext4_free_inodes_count(sb, gdp),
			   &sbi->s_flex_groups[flex_group].free_inodes);
2178 2179
		atomic64_add(ext4_free_group_clusters(sb, gdp),
			     &sbi->s_flex_groups[flex_group].free_clusters);
2180 2181
		atomic_add(ext4_used_dirs_count(sb, gdp),
			   &sbi->s_flex_groups[flex_group].used_dirs);
2182 2183 2184 2185 2186 2187 2188
	}

	return 1;
failed:
	return 0;
}

2189
static __le16 ext4_group_desc_csum(struct super_block *sb, __u32 block_group,
2190
				   struct ext4_group_desc *gdp)
A
Andreas Dilger 已提交
2191
{
2192
	int offset = offsetof(struct ext4_group_desc, bg_checksum);
A
Andreas Dilger 已提交
2193
	__u16 crc = 0;
2194
	__le32 le_group = cpu_to_le32(block_group);
2195
	struct ext4_sb_info *sbi = EXT4_SB(sb);
A
Andreas Dilger 已提交
2196

2197
	if (ext4_has_metadata_csum(sbi->s_sb)) {
2198 2199
		/* Use new metadata_csum algorithm */
		__u32 csum32;
2200
		__u16 dummy_csum = 0;
2201 2202 2203

		csum32 = ext4_chksum(sbi, sbi->s_csum_seed, (__u8 *)&le_group,
				     sizeof(le_group));
2204 2205 2206 2207 2208 2209 2210
		csum32 = ext4_chksum(sbi, csum32, (__u8 *)gdp, offset);
		csum32 = ext4_chksum(sbi, csum32, (__u8 *)&dummy_csum,
				     sizeof(dummy_csum));
		offset += sizeof(dummy_csum);
		if (offset < sbi->s_desc_size)
			csum32 = ext4_chksum(sbi, csum32, (__u8 *)gdp + offset,
					     sbi->s_desc_size - offset);
2211 2212 2213

		crc = csum32 & 0xFFFF;
		goto out;
A
Andreas Dilger 已提交
2214 2215
	}

2216
	/* old crc16 code */
2217
	if (!ext4_has_feature_gdt_csum(sb))
2218 2219
		return 0;

2220 2221 2222 2223 2224
	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...*/
2225
	if (ext4_has_feature_64bit(sb) &&
2226 2227 2228 2229 2230 2231
	    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
2232 2233 2234
	return cpu_to_le16(crc);
}

2235
int ext4_group_desc_csum_verify(struct super_block *sb, __u32 block_group,
A
Andreas Dilger 已提交
2236 2237
				struct ext4_group_desc *gdp)
{
2238
	if (ext4_has_group_desc_csum(sb) &&
2239
	    (gdp->bg_checksum != ext4_group_desc_csum(sb, block_group, gdp)))
A
Andreas Dilger 已提交
2240 2241 2242 2243 2244
		return 0;

	return 1;
}

2245 2246 2247 2248 2249
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;
2250
	gdp->bg_checksum = ext4_group_desc_csum(sb, block_group, gdp);
2251 2252
}

2253
/* Called at mount-time, super-block is locked */
2254
static int ext4_check_descriptors(struct super_block *sb,
2255
				  ext4_fsblk_t sb_block,
2256
				  ext4_group_t *first_not_zeroed)
2257
{
2258 2259 2260
	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
2261 2262 2263
	ext4_fsblk_t block_bitmap;
	ext4_fsblk_t inode_bitmap;
	ext4_fsblk_t inode_table;
J
Jose R. Santos 已提交
2264
	int flexbg_flag = 0;
2265
	ext4_group_t i, grp = sbi->s_groups_count;
2266

2267
	if (ext4_has_feature_flex_bg(sb))
J
Jose R. Santos 已提交
2268 2269
		flexbg_flag = 1;

2270
	ext4_debug("Checking group descriptors");
2271

2272 2273 2274
	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
2275
		if (i == sbi->s_groups_count - 1 || flexbg_flag)
L
Laurent Vivier 已提交
2276
			last_block = ext4_blocks_count(sbi->s_es) - 1;
2277 2278
		else
			last_block = first_block +
2279
				(EXT4_BLOCKS_PER_GROUP(sb) - 1);
2280

2281 2282 2283 2284
		if ((grp == sbi->s_groups_count) &&
		   !(gdp->bg_flags & cpu_to_le16(EXT4_BG_INODE_ZEROED)))
			grp = i;

2285
		block_bitmap = ext4_block_bitmap(sb, gdp);
2286 2287 2288 2289 2290
		if (block_bitmap == sb_block) {
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
				 "Block bitmap for group %u overlaps "
				 "superblock", i);
		}
2291
		if (block_bitmap < first_block || block_bitmap > last_block) {
2292
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
2293
			       "Block bitmap for group %u not in group "
2294
			       "(block %llu)!", i, block_bitmap);
2295 2296
			return 0;
		}
2297
		inode_bitmap = ext4_inode_bitmap(sb, gdp);
2298 2299 2300 2301 2302
		if (inode_bitmap == sb_block) {
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
				 "Inode bitmap for group %u overlaps "
				 "superblock", i);
		}
2303
		if (inode_bitmap < first_block || inode_bitmap > last_block) {
2304
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
2305
			       "Inode bitmap for group %u not in group "
2306
			       "(block %llu)!", i, inode_bitmap);
2307 2308
			return 0;
		}
2309
		inode_table = ext4_inode_table(sb, gdp);
2310 2311 2312 2313 2314
		if (inode_table == sb_block) {
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
				 "Inode table for group %u overlaps "
				 "superblock", i);
		}
L
Laurent Vivier 已提交
2315
		if (inode_table < first_block ||
2316
		    inode_table + sbi->s_itb_per_group - 1 > last_block) {
2317
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
2318
			       "Inode table for group %u not in group "
2319
			       "(block %llu)!", i, inode_table);
2320 2321
			return 0;
		}
2322
		ext4_lock_group(sb, i);
2323
		if (!ext4_group_desc_csum_verify(sb, i, gdp)) {
2324 2325
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
				 "Checksum for group %u failed (%u!=%u)",
2326
				 i, le16_to_cpu(ext4_group_desc_csum(sb, i,
2327
				     gdp)), le16_to_cpu(gdp->bg_checksum));
2328
			if (!(sb->s_flags & MS_RDONLY)) {
2329
				ext4_unlock_group(sb, i);
2330
				return 0;
2331
			}
A
Andreas Dilger 已提交
2332
		}
2333
		ext4_unlock_group(sb, i);
J
Jose R. Santos 已提交
2334 2335
		if (!flexbg_flag)
			first_block += EXT4_BLOCKS_PER_GROUP(sb);
2336
	}
2337 2338
	if (NULL != first_not_zeroed)
		*first_not_zeroed = grp;
2339 2340 2341
	return 1;
}

2342
/* ext4_orphan_cleanup() walks a singly-linked list of inodes (starting at
2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354
 * 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
2355
 * ext4_free_inode().  The only reason we would point at a wrong inode is if
2356 2357 2358
 * 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.
 */
2359 2360
static void ext4_orphan_cleanup(struct super_block *sb,
				struct ext4_super_block *es)
2361 2362
{
	unsigned int s_flags = sb->s_flags;
2363
	int ret, nr_orphans = 0, nr_truncates = 0;
2364 2365 2366 2367 2368 2369 2370 2371
#ifdef CONFIG_QUOTA
	int i;
#endif
	if (!es->s_last_orphan) {
		jbd_debug(4, "no orphan inodes to clean up\n");
		return;
	}

2372
	if (bdev_read_only(sb->s_bdev)) {
2373 2374
		ext4_msg(sb, KERN_ERR, "write access "
			"unavailable, skipping orphan cleanup");
2375 2376 2377
		return;
	}

2378 2379 2380 2381 2382 2383 2384
	/* 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;
	}

2385
	if (EXT4_SB(sb)->s_mount_state & EXT4_ERROR_FS) {
2386 2387
		/* don't clear list on RO mount w/ errors */
		if (es->s_last_orphan && !(s_flags & MS_RDONLY)) {
2388
			ext4_msg(sb, KERN_INFO, "Errors on filesystem, "
2389
				  "clearing orphan list.\n");
2390 2391
			es->s_last_orphan = 0;
		}
2392 2393 2394 2395 2396
		jbd_debug(1, "Skipping orphan recovery on fs with errors.\n");
		return;
	}

	if (s_flags & MS_RDONLY) {
2397
		ext4_msg(sb, KERN_INFO, "orphan cleanup on readonly fs");
2398 2399 2400 2401 2402 2403
		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
2404
	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
2405 2406
		if (EXT4_SB(sb)->s_qf_names[i]) {
			int ret = ext4_quota_on_mount(sb, i);
2407
			if (ret < 0)
2408 2409 2410
				ext4_msg(sb, KERN_ERR,
					"Cannot turn on journaled "
					"quota: error %d", ret);
2411 2412 2413 2414 2415 2416 2417
		}
	}
#endif

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

2418 2419 2420 2421 2422 2423 2424 2425 2426 2427
		/*
		 * We may have encountered an error during cleanup; if
		 * so, skip the rest.
		 */
		if (EXT4_SB(sb)->s_mount_state & EXT4_ERROR_FS) {
			jbd_debug(1, "Skipping orphan recovery on fs with errors.\n");
			es->s_last_orphan = 0;
			break;
		}

2428 2429
		inode = ext4_orphan_get(sb, le32_to_cpu(es->s_last_orphan));
		if (IS_ERR(inode)) {
2430 2431 2432 2433
			es->s_last_orphan = 0;
			break;
		}

2434
		list_add(&EXT4_I(inode)->i_orphan, &EXT4_SB(sb)->s_orphan);
2435
		dquot_initialize(inode);
2436
		if (inode->i_nlink) {
2437 2438 2439 2440
			if (test_opt(sb, DEBUG))
				ext4_msg(sb, KERN_DEBUG,
					"%s: truncating inode %lu to %lld bytes",
					__func__, inode->i_ino, inode->i_size);
2441
			jbd_debug(2, "truncating inode %lu to %lld bytes\n",
2442
				  inode->i_ino, inode->i_size);
A
Al Viro 已提交
2443
			inode_lock(inode);
2444
			truncate_inode_pages(inode->i_mapping, inode->i_size);
2445 2446 2447
			ret = ext4_truncate(inode);
			if (ret)
				ext4_std_error(inode->i_sb, ret);
A
Al Viro 已提交
2448
			inode_unlock(inode);
2449 2450
			nr_truncates++;
		} else {
2451 2452 2453 2454
			if (test_opt(sb, DEBUG))
				ext4_msg(sb, KERN_DEBUG,
					"%s: deleting unreferenced inode %lu",
					__func__, inode->i_ino);
2455 2456 2457 2458 2459 2460 2461
			jbd_debug(2, "deleting unreferenced inode %lu\n",
				  inode->i_ino);
			nr_orphans++;
		}
		iput(inode);  /* The delete magic happens here! */
	}

2462
#define PLURAL(x) (x), ((x) == 1) ? "" : "s"
2463 2464

	if (nr_orphans)
2465 2466
		ext4_msg(sb, KERN_INFO, "%d orphan inode%s deleted",
		       PLURAL(nr_orphans));
2467
	if (nr_truncates)
2468 2469
		ext4_msg(sb, KERN_INFO, "%d truncate%s cleaned up",
		       PLURAL(nr_truncates));
2470 2471
#ifdef CONFIG_QUOTA
	/* Turn quotas off */
J
Jan Kara 已提交
2472
	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
2473
		if (sb_dqopt(sb)->files[i])
2474
			dquot_quota_off(sb, i);
2475 2476 2477 2478
	}
#endif
	sb->s_flags = s_flags; /* Restore MS_RDONLY status */
}
2479

2480 2481 2482 2483 2484 2485 2486
/*
 * 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.
 *
2487 2488 2489 2490 2491 2492
 * 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.
 *
2493 2494
 * Note, this does *not* consider any metadata overhead for vfs i_blocks.
 */
2495
static loff_t ext4_max_size(int blkbits, int has_huge_files)
2496 2497 2498 2499 2500
{
	loff_t res;
	loff_t upper_limit = MAX_LFS_FILESIZE;

	/* small i_blocks in vfs inode? */
2501
	if (!has_huge_files || sizeof(blkcnt_t) < sizeof(u64)) {
2502
		/*
2503
		 * CONFIG_LBDAF is not enabled implies the inode
2504 2505 2506 2507 2508 2509 2510 2511 2512 2513
		 * 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;
	}

2514 2515 2516 2517 2518 2519
	/*
	 * 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;
2520 2521 2522 2523 2524 2525 2526 2527
	res <<= blkbits;

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

	return res;
}
2528 2529

/*
2530
 * Maximal bitmap file size.  There is a direct, and {,double-,triple-}indirect
2531 2532
 * 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.
2533
 */
2534
static loff_t ext4_max_bitmap_size(int bits, int has_huge_files)
2535
{
2536
	loff_t res = EXT4_NDIR_BLOCKS;
2537 2538
	int meta_blocks;
	loff_t upper_limit;
2539 2540 2541 2542 2543 2544
	/* 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.
2545 2546
	 */

2547
	if (!has_huge_files || sizeof(blkcnt_t) < sizeof(u64)) {
2548
		/*
2549
		 * !has_huge_files or CONFIG_LBDAF not enabled implies that
2550 2551
		 * the inode i_block field represents total file blocks in
		 * 2^32 512-byte sectors == size of vfs inode i_blocks * 8
2552 2553 2554 2555 2556 2557 2558
		 */
		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
2559 2560 2561 2562 2563 2564
		/*
		 * 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
		 */
2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577
		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;
2578 2579 2580 2581 2582 2583 2584

	res += 1LL << (bits-2);
	res += 1LL << (2*(bits-2));
	res += 1LL << (3*(bits-2));
	res <<= bits;
	if (res > upper_limit)
		res = upper_limit;
2585 2586 2587 2588

	if (res > MAX_LFS_FILESIZE)
		res = MAX_LFS_FILESIZE;

2589 2590 2591
	return res;
}

2592
static ext4_fsblk_t descriptor_loc(struct super_block *sb,
2593
				   ext4_fsblk_t logical_sb_block, int nr)
2594
{
2595
	struct ext4_sb_info *sbi = EXT4_SB(sb);
2596
	ext4_group_t bg, first_meta_bg;
2597 2598 2599 2600
	int has_super = 0;

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

2601
	if (!ext4_has_feature_meta_bg(sb) || nr < first_meta_bg)
2602
		return logical_sb_block + nr + 1;
2603
	bg = sbi->s_desc_per_block * nr;
2604
	if (ext4_bg_has_super(sb, bg))
2605
		has_super = 1;
2606

2607 2608 2609 2610 2611 2612 2613 2614 2615 2616
	/*
	 * 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++;

2617
	return (has_super + ext4_group_first_block_no(sb, bg));
2618 2619
}

2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635
/**
 * 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
2636
	int ret;
2637 2638

	if (sbi->s_stripe && sbi->s_stripe <= sbi->s_blocks_per_group)
D
Dan Ehrenberg 已提交
2639
		ret = sbi->s_stripe;
2640
	else if (stripe_width && stripe_width <= sbi->s_blocks_per_group)
D
Dan Ehrenberg 已提交
2641
		ret = stripe_width;
2642
	else if (stride && stride <= sbi->s_blocks_per_group)
D
Dan Ehrenberg 已提交
2643 2644 2645
		ret = stride;
	else
		ret = 0;
2646

D
Dan Ehrenberg 已提交
2647 2648 2649 2650 2651 2652
	/*
	 * 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;
2653

D
Dan Ehrenberg 已提交
2654
	return ret;
2655
}
2656

2657 2658 2659 2660 2661 2662 2663 2664
/*
 * 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)
{
2665
	if (ext4_has_unknown_ext4_incompat_features(sb)) {
2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676
		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;

2677
	if (ext4_has_feature_readonly(sb)) {
D
Darrick J. Wong 已提交
2678 2679 2680 2681 2682
		ext4_msg(sb, KERN_INFO, "filesystem is read-only");
		sb->s_flags |= MS_RDONLY;
		return 1;
	}

2683
	/* Check that feature set is OK for a read-write mount */
2684
	if (ext4_has_unknown_ext4_ro_compat_features(sb)) {
2685 2686 2687 2688 2689 2690 2691 2692 2693 2694
		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
	 */
2695
	if (ext4_has_feature_huge_file(sb)) {
2696 2697 2698 2699 2700 2701 2702
		if (sizeof(blkcnt_t) < sizeof(u64)) {
			ext4_msg(sb, KERN_ERR, "Filesystem with huge files "
				 "cannot be mounted RDWR without "
				 "CONFIG_LBDAF");
			return 0;
		}
	}
2703
	if (ext4_has_feature_bigalloc(sb) && !ext4_has_feature_extents(sb)) {
2704 2705 2706 2707 2708
		ext4_msg(sb, KERN_ERR,
			 "Can't support bigalloc feature without "
			 "extents feature\n");
		return 0;
	}
2709 2710

#ifndef CONFIG_QUOTA
2711
	if (ext4_has_feature_quota(sb) && !readonly) {
2712 2713 2714 2715 2716
		ext4_msg(sb, KERN_ERR,
			 "Filesystem with quota feature cannot be mounted RDWR "
			 "without CONFIG_QUOTA");
		return 0;
	}
L
Li Xi 已提交
2717 2718 2719 2720 2721 2722
	if (ext4_has_feature_project(sb) && !readonly) {
		ext4_msg(sb, KERN_ERR,
			 "Filesystem with project quota feature cannot be mounted RDWR "
			 "without CONFIG_QUOTA");
		return 0;
	}
2723
#endif  /* CONFIG_QUOTA */
2724 2725 2726
	return 1;
}

2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740
/*
 * 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)
2741 2742
		/* fsck newer than v1.41.13 is needed to clean this condition. */
		ext4_msg(sb, KERN_NOTICE, "error count since last fsck: %u",
2743 2744
			 le32_to_cpu(es->s_error_count));
	if (es->s_first_error_time) {
2745
		printk(KERN_NOTICE "EXT4-fs (%s): initial error at time %u: %.*s:%d",
2746 2747 2748 2749 2750
		       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)
2751
			printk(KERN_CONT ": inode %u",
2752 2753
			       le32_to_cpu(es->s_first_error_ino));
		if (es->s_first_error_block)
2754
			printk(KERN_CONT ": block %llu", (unsigned long long)
2755
			       le64_to_cpu(es->s_first_error_block));
2756
		printk(KERN_CONT "\n");
2757 2758
	}
	if (es->s_last_error_time) {
2759
		printk(KERN_NOTICE "EXT4-fs (%s): last error at time %u: %.*s:%d",
2760 2761 2762 2763 2764
		       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)
2765
			printk(KERN_CONT ": inode %u",
2766 2767
			       le32_to_cpu(es->s_last_error_ino));
		if (es->s_last_error_block)
2768
			printk(KERN_CONT ": block %llu", (unsigned long long)
2769
			       le64_to_cpu(es->s_last_error_block));
2770
		printk(KERN_CONT "\n");
2771 2772 2773 2774
	}
	mod_timer(&sbi->s_err_report, jiffies + 24*60*60*HZ);  /* Once a day */
}

2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797
/* 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;

	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;
	}

2798
	if (group >= ngroups)
2799 2800 2801 2802 2803 2804 2805
		ret = 1;

	if (!ret) {
		timeout = jiffies;
		ret = ext4_init_inode_table(sb, group,
					    elr->lr_timeout ? 0 : 1);
		if (elr->lr_timeout == 0) {
2806 2807
			timeout = (jiffies - timeout) *
				  elr->lr_sbi->s_li_wait_mult;
2808 2809 2810 2811 2812 2813 2814 2815 2816 2817
			elr->lr_timeout = timeout;
		}
		elr->lr_next_sched = jiffies + elr->lr_timeout;
		elr->lr_next_group = group + 1;
	}
	return ret;
}

/*
 * Remove lr_request from the list_request and free the
2818
 * request structure. Should be called with li_list_mtx held
2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835
 */
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)
{
2836 2837 2838
	mutex_lock(&ext4_li_mtx);
	if (!ext4_li_info) {
		mutex_unlock(&ext4_li_mtx);
2839
		return;
2840
	}
2841 2842

	mutex_lock(&ext4_li_info->li_list_mtx);
2843
	ext4_remove_li_request(EXT4_SB(sb)->s_li_request);
2844
	mutex_unlock(&ext4_li_info->li_list_mtx);
2845
	mutex_unlock(&ext4_li_mtx);
2846 2847
}

2848 2849
static struct task_struct *ext4_lazyinit_task;

2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863
/*
 * 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;
2864
	unsigned long next_wakeup, cur;
2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877

	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) {
2878 2879
			int err = 0;
			int progress = 0;
2880 2881 2882
			elr = list_entry(pos, struct ext4_li_request,
					 lr_request);

2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900
			if (time_before(jiffies, elr->lr_next_sched)) {
				if (time_before(elr->lr_next_sched, next_wakeup))
					next_wakeup = elr->lr_next_sched;
				continue;
			}
			if (down_read_trylock(&elr->lr_super->s_umount)) {
				if (sb_start_write_trylock(elr->lr_super)) {
					progress = 1;
					/*
					 * We hold sb->s_umount, sb can not
					 * be removed from the list, it is
					 * now safe to drop li_list_mtx
					 */
					mutex_unlock(&eli->li_list_mtx);
					err = ext4_run_li_request(elr);
					sb_end_write(elr->lr_super);
					mutex_lock(&eli->li_list_mtx);
					n = pos->next;
2901
				}
2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912
				up_read((&elr->lr_super->s_umount));
			}
			/* error, remove the lazy_init job */
			if (err) {
				ext4_remove_li_request(elr);
				continue;
			}
			if (!progress) {
				elr->lr_next_sched = jiffies +
					(prandom_u32()
					 % (EXT4_DEF_LI_MAX_START_DELAY * HZ));
2913 2914 2915 2916 2917 2918
			}
			if (time_before(elr->lr_next_sched, next_wakeup))
				next_wakeup = elr->lr_next_sched;
		}
		mutex_unlock(&eli->li_list_mtx);

2919
		try_to_freeze();
2920

2921 2922
		cur = jiffies;
		if ((time_after_eq(cur, next_wakeup)) ||
2923
		    (MAX_JIFFY_OFFSET == next_wakeup)) {
2924 2925 2926 2927
			cond_resched();
			continue;
		}

2928 2929
		schedule_timeout_interruptible(next_wakeup - cur);

2930 2931 2932 2933
		if (kthread_should_stop()) {
			ext4_clear_request_list();
			goto exit_thread;
		}
2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975
	}

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)
{
2976 2977 2978 2979
	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);
2980 2981 2982
		ext4_clear_request_list();
		kfree(ext4_li_info);
		ext4_li_info = NULL;
2983
		printk(KERN_CRIT "EXT4-fs: error %d creating inode table "
2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051
				 "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.
	 */
3052 3053
	elr->lr_next_sched = jiffies + (prandom_u32() %
				(EXT4_DEF_LI_MAX_START_DELAY * HZ));
3054 3055 3056
	return elr;
}

3057 3058
int ext4_register_li_request(struct super_block *sb,
			     ext4_group_t first_not_zeroed)
3059 3060
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
3061
	struct ext4_li_request *elr = NULL;
3062
	ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count;
3063
	int ret = 0;
3064

3065
	mutex_lock(&ext4_li_mtx);
3066 3067 3068 3069 3070 3071
	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;
3072
		goto out;
3073
	}
3074 3075 3076

	if (first_not_zeroed == ngroups ||
	    (sb->s_flags & MS_RDONLY) ||
3077
	    !test_opt(sb, INIT_INODE_TABLE))
3078
		goto out;
3079 3080

	elr = ext4_li_request_new(sb, first_not_zeroed);
3081 3082 3083 3084
	if (!elr) {
		ret = -ENOMEM;
		goto out;
	}
3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096

	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;
3097 3098 3099 3100 3101 3102
	/*
	 * 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;
3103 3104 3105 3106 3107 3108 3109

	if (!(ext4_li_info->li_state & EXT4_LAZYINIT_RUNNING)) {
		ret = ext4_run_lazyinit_thread();
		if (ret)
			goto out;
	}
out:
3110 3111
	mutex_unlock(&ext4_li_mtx);
	if (ret)
3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125
		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.
	 */
3126
	if (!ext4_li_info || !ext4_lazyinit_task)
3127 3128
		return;

3129
	kthread_stop(ext4_lazyinit_task);
3130 3131
}

3132 3133 3134 3135 3136 3137
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);

3138
	if (ext4_has_metadata_csum(sb)) {
3139
		/* journal checksum v3 */
3140
		compat = 0;
3141
		incompat = JBD2_FEATURE_INCOMPAT_CSUM_V3;
3142 3143 3144 3145 3146 3147
	} else {
		/* journal checksum v1 */
		compat = JBD2_FEATURE_COMPAT_CHECKSUM;
		incompat = 0;
	}

3148 3149 3150 3151
	jbd2_journal_clear_features(sbi->s_journal,
			JBD2_FEATURE_COMPAT_CHECKSUM, 0,
			JBD2_FEATURE_INCOMPAT_CSUM_V3 |
			JBD2_FEATURE_INCOMPAT_CSUM_V2);
3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163
	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 {
3164 3165
		jbd2_journal_clear_features(sbi->s_journal, 0, 0,
				JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT);
3166 3167 3168 3169 3170
	}

	return ret;
}

3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194
/*
 * 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;

3195
	if (!ext4_has_feature_bigalloc(sb))
3196 3197 3198
		return (ext4_bg_has_super(sb, grp) + ext4_bg_num_gdb(sb, grp) +
			sbi->s_itb_per_group + 2);

3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227
	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++;
		}
3228 3229 3230 3231 3232
		j = ext4_bg_num_gdb(sb, grp);
		if (s + j > EXT4_BLOCKS_PER_GROUP(sb)) {
			ext4_error(sb, "Invalid number of block group "
				   "descriptor blocks: %d", j);
			j = EXT4_BLOCKS_PER_GROUP(sb) - s;
3233
		}
3234 3235 3236
		count += j;
		for (; j > 0; j--)
			ext4_set_bit(EXT4_B2C(sbi, s++), buf);
3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248 3249 3250
	}
	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;
3251 3252
	struct inode *j_inode;
	unsigned int j_blocks, j_inum = le32_to_cpu(es->s_journal_inum);
3253 3254
	ext4_group_t i, ngroups = ext4_get_groups_count(sb);
	ext4_fsblk_t overhead = 0;
3255
	char *buf = (char *) get_zeroed_page(GFP_NOFS);
3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282

	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();
	}
3283 3284 3285 3286 3287

	/*
	 * Add the internal journal blocks whether the journal has been
	 * loaded or not
	 */
3288
	if (sbi->s_journal && !sbi->journal_bdev)
3289
		overhead += EXT4_NUM_B2C(sbi, sbi->s_journal->j_maxlen);
3290 3291 3292 3293 3294 3295 3296 3297 3298 3299
	else if (ext4_has_feature_journal(sb) && !sbi->s_journal) {
		j_inode = ext4_get_journal_inode(sb, j_inum);
		if (j_inode) {
			j_blocks = j_inode->i_size >> sb->s_blocksize_bits;
			overhead += EXT4_NUM_B2C(sbi, j_blocks);
			iput(j_inode);
		} else {
			ext4_msg(sb, KERN_ERR, "can't get journal size");
		}
	}
3300 3301 3302 3303 3304 3305
	sbi->s_overhead = overhead;
	smp_wmb();
	free_page((unsigned long) buf);
	return 0;
}

3306
static void ext4_set_resv_clusters(struct super_block *sb)
L
Lukas Czerner 已提交
3307 3308
{
	ext4_fsblk_t resv_clusters;
3309
	struct ext4_sb_info *sbi = EXT4_SB(sb);
L
Lukas Czerner 已提交
3310

3311 3312 3313 3314 3315 3316
	/*
	 * 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.
	 */
3317
	if (!ext4_has_feature_extents(sb))
3318
		return;
L
Lukas Czerner 已提交
3319 3320 3321 3322
	/*
	 * 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
3323
	 * unwritten extents in delalloc path. In most cases such
L
Lukas Czerner 已提交
3324 3325 3326
	 * allocation would require 1, or 2 blocks, higher numbers are
	 * very rare.
	 */
3327 3328
	resv_clusters = (ext4_blocks_count(sbi->s_es) >>
			 sbi->s_cluster_bits);
L
Lukas Czerner 已提交
3329 3330 3331 3332

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

3333
	atomic64_set(&sbi->s_resv_clusters, resv_clusters);
L
Lukas Czerner 已提交
3334 3335
}

3336
static int ext4_fill_super(struct super_block *sb, void *data, int silent)
3337
{
3338
	char *orig_data = kstrdup(data, GFP_KERNEL);
3339
	struct buffer_head *bh;
3340
	struct ext4_super_block *es = NULL;
3341
	struct ext4_sb_info *sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
3342 3343
	ext4_fsblk_t block;
	ext4_fsblk_t sb_block = get_sb_block(&data);
3344
	ext4_fsblk_t logical_sb_block;
3345 3346 3347 3348
	unsigned long offset = 0;
	unsigned long journal_devnum = 0;
	unsigned long def_mount_opts;
	struct inode *root;
3349
	const char *descr;
3350
	int ret = -ENOMEM;
3351
	int blocksize, clustersize;
3352 3353
	unsigned int db_count;
	unsigned int i;
3354
	int needs_recovery, has_huge_files, has_bigalloc;
L
Laurent Vivier 已提交
3355
	__u64 blocks_count;
3356
	int err = 0;
3357
	unsigned int journal_ioprio = DEFAULT_JOURNAL_IOPRIO;
3358
	ext4_group_t first_not_zeroed;
3359

3360 3361
	if ((data && !orig_data) || !sbi)
		goto out_free_base;
3362 3363 3364

	sbi->s_blockgroup_lock =
		kzalloc(sizeof(struct blockgroup_lock), GFP_KERNEL);
3365 3366 3367
	if (!sbi->s_blockgroup_lock)
		goto out_free_base;

3368
	sb->s_fs_info = sbi;
3369
	sbi->s_sb = sb;
3370
	sbi->s_inode_readahead_blks = EXT4_DEF_INODE_READAHEAD_BLKS;
M
Miklos Szeredi 已提交
3371
	sbi->s_sb_block = sb_block;
3372 3373 3374
	if (sb->s_bdev->bd_part)
		sbi->s_sectors_written_start =
			part_stat_read(sb->s_bdev->bd_part, sectors[1]);
3375

3376
	/* Cleanup superblock name */
3377
	strreplace(sb->s_id, '/', '!');
3378

3379
	/* -EINVAL is default */
3380
	ret = -EINVAL;
3381
	blocksize = sb_min_blocksize(sb, EXT4_MIN_BLOCK_SIZE);
3382
	if (!blocksize) {
3383
		ext4_msg(sb, KERN_ERR, "unable to set blocksize");
3384 3385 3386 3387
		goto out_fail;
	}

	/*
3388
	 * The ext4 superblock will not be buffer aligned for other than 1kB
3389 3390
	 * block sizes.  We need to calculate the offset from buffer start.
	 */
3391
	if (blocksize != EXT4_MIN_BLOCK_SIZE) {
3392 3393
		logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE;
		offset = do_div(logical_sb_block, blocksize);
3394
	} else {
3395
		logical_sb_block = sb_block;
3396 3397
	}

3398
	if (!(bh = sb_bread_unmovable(sb, logical_sb_block))) {
3399
		ext4_msg(sb, KERN_ERR, "unable to read superblock");
3400 3401 3402 3403
		goto out_fail;
	}
	/*
	 * Note: s_es must be initialized as soon as possible because
3404
	 *       some ext4 macro-instructions depend on its value
3405
	 */
3406
	es = (struct ext4_super_block *) (bh->b_data + offset);
3407 3408
	sbi->s_es = es;
	sb->s_magic = le16_to_cpu(es->s_magic);
3409 3410
	if (sb->s_magic != EXT4_SUPER_MAGIC)
		goto cantfind_ext4;
T
Theodore Ts'o 已提交
3411
	sbi->s_kbytes_written = le64_to_cpu(es->s_kbytes_written);
3412

3413
	/* Warn if metadata_csum and gdt_csum are both set. */
3414 3415
	if (ext4_has_feature_metadata_csum(sb) &&
	    ext4_has_feature_gdt_csum(sb))
3416
		ext4_warning(sb, "metadata_csum and uninit_bg are "
3417 3418
			     "redundant flags; please run fsck.");

3419 3420 3421 3422 3423 3424 3425 3426
	/* 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;
	}

3427
	/* Load the checksum driver */
3428
	if (ext4_has_feature_metadata_csum(sb)) {
3429 3430 3431 3432 3433 3434 3435 3436 3437
		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;
		}
	}

3438 3439 3440 3441 3442
	/* 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;
3443
		ret = -EFSBADCRC;
3444 3445 3446 3447
		goto cantfind_ext4;
	}

	/* Precompute checksum seed for all metadata */
3448
	if (ext4_has_feature_csum_seed(sb))
3449 3450
		sbi->s_csum_seed = le32_to_cpu(es->s_checksum_seed);
	else if (ext4_has_metadata_csum(sb))
3451 3452 3453
		sbi->s_csum_seed = ext4_chksum(sbi, ~0, es->s_uuid,
					       sizeof(es->s_uuid));

3454 3455
	/* Set defaults before we parse the mount options */
	def_mount_opts = le32_to_cpu(es->s_default_mount_opts);
3456
	set_opt(sb, INIT_INODE_TABLE);
3457
	if (def_mount_opts & EXT4_DEFM_DEBUG)
3458
		set_opt(sb, DEBUG);
3459
	if (def_mount_opts & EXT4_DEFM_BSDGROUPS)
3460
		set_opt(sb, GRPID);
3461
	if (def_mount_opts & EXT4_DEFM_UID16)
3462
		set_opt(sb, NO_UID32);
3463 3464
	/* xattr user namespace & acls are now defaulted on */
	set_opt(sb, XATTR_USER);
T
Theodore Ts'o 已提交
3465
#ifdef CONFIG_EXT4_FS_POSIX_ACL
3466
	set_opt(sb, POSIX_ACL);
3467
#endif
3468 3469 3470 3471
	/* don't forget to enable journal_csum when metadata_csum is enabled. */
	if (ext4_has_metadata_csum(sb))
		set_opt(sb, JOURNAL_CHECKSUM);

3472
	if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_DATA)
3473
		set_opt(sb, JOURNAL_DATA);
3474
	else if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_ORDERED)
3475
		set_opt(sb, ORDERED_DATA);
3476
	else if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_WBACK)
3477
		set_opt(sb, WRITEBACK_DATA);
3478 3479

	if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_PANIC)
3480
		set_opt(sb, ERRORS_PANIC);
3481
	else if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_CONTINUE)
3482
		set_opt(sb, ERRORS_CONT);
3483
	else
3484
		set_opt(sb, ERRORS_RO);
3485 3486
	/* block_validity enabled by default; disable with noblock_validity */
	set_opt(sb, BLOCK_VALIDITY);
3487
	if (def_mount_opts & EXT4_DEFM_DISCARD)
3488
		set_opt(sb, DISCARD);
3489

3490 3491
	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));
3492 3493 3494
	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;
3495

3496
	if ((def_mount_opts & EXT4_DEFM_NOBARRIER) == 0)
3497
		set_opt(sb, BARRIER);
3498

3499 3500 3501 3502
	/*
	 * enable delayed allocation by default
	 * Use -o nodelalloc to turn it off
	 */
3503
	if (!IS_EXT3_SB(sb) && !IS_EXT2_SB(sb) &&
3504
	    ((def_mount_opts & EXT4_DEFM_NODELALLOC) == 0))
3505
		set_opt(sb, DELALLOC);
3506

3507 3508 3509 3510 3511 3512
	/*
	 * 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;

3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525
	if (sbi->s_es->s_mount_opts[0]) {
		char *s_mount_opts = kstrndup(sbi->s_es->s_mount_opts,
					      sizeof(sbi->s_es->s_mount_opts),
					      GFP_KERNEL);
		if (!s_mount_opts)
			goto failed_mount;
		if (!parse_options(s_mount_opts, sb, &journal_devnum,
				   &journal_ioprio, 0)) {
			ext4_msg(sb, KERN_WARNING,
				 "failed to parse options in superblock: %s",
				 s_mount_opts);
		}
		kfree(s_mount_opts);
3526
	}
3527
	sbi->s_def_mount_opt = sbi->s_mount_opt;
3528
	if (!parse_options((char *) data, sb, &journal_devnum,
3529
			   &journal_ioprio, 0))
3530 3531
		goto failed_mount;

3532 3533 3534 3535 3536 3537 3538 3539 3540 3541 3542
	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 "
3543
				 "both data=journal and dioread_nolock");
3544 3545
			goto failed_mount;
		}
R
Ross Zwisler 已提交
3546 3547 3548 3549 3550
		if (test_opt(sb, DAX)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "both data=journal and dax");
			goto failed_mount;
		}
3551 3552 3553 3554 3555
		if (ext4_has_feature_encrypt(sb)) {
			ext4_msg(sb, KERN_WARNING,
				 "encrypted files will use data=ordered "
				 "instead of data journaling mode");
		}
3556 3557
		if (test_opt(sb, DELALLOC))
			clear_opt(sb, DELALLOC);
3558 3559
	} else {
		sb->s_iflags |= SB_I_CGROUPWB;
3560 3561
	}

3562
	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
3563
		(test_opt(sb, POSIX_ACL) ? MS_POSIXACL : 0);
3564

3565
	if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV &&
3566 3567 3568
	    (ext4_has_compat_features(sb) ||
	     ext4_has_ro_compat_features(sb) ||
	     ext4_has_incompat_features(sb)))
3569 3570 3571
		ext4_msg(sb, KERN_WARNING,
		       "feature flags set on rev 0 fs, "
		       "running e2fsck is recommended");
3572

3573 3574
	if (es->s_creator_os == cpu_to_le32(EXT4_OS_HURD)) {
		set_opt2(sb, HURD_COMPAT);
3575
		if (ext4_has_feature_64bit(sb)) {
3576 3577 3578 3579 3580 3581
			ext4_msg(sb, KERN_ERR,
				 "The Hurd can't support 64-bit file systems");
			goto failed_mount;
		}
	}

3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603
	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;
		}
	}

3604 3605 3606 3607 3608
	/*
	 * 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.
	 */
3609
	if (!ext4_feature_set_ok(sb, (sb->s_flags & MS_RDONLY)))
3610
		goto failed_mount;
3611

J
Jan Kara 已提交
3612
	blocksize = BLOCK_SIZE << le32_to_cpu(es->s_log_block_size);
3613 3614
	if (blocksize < EXT4_MIN_BLOCK_SIZE ||
	    blocksize > EXT4_MAX_BLOCK_SIZE) {
3615
		ext4_msg(sb, KERN_ERR,
3616 3617 3618 3619 3620 3621 3622 3623 3624
		       "Unsupported filesystem blocksize %d (%d log_block_size)",
			 blocksize, le32_to_cpu(es->s_log_block_size));
		goto failed_mount;
	}
	if (le32_to_cpu(es->s_log_block_size) >
	    (EXT4_MAX_BLOCK_LOG_SIZE - EXT4_MIN_BLOCK_LOG_SIZE)) {
		ext4_msg(sb, KERN_ERR,
			 "Invalid log block size: %u",
			 le32_to_cpu(es->s_log_block_size));
3625 3626 3627
		goto failed_mount;
	}

3628 3629 3630 3631 3632 3633 3634
	if (le16_to_cpu(sbi->s_es->s_reserved_gdt_blocks) > (blocksize / 4)) {
		ext4_msg(sb, KERN_ERR,
			 "Number of reserved GDT blocks insanely large: %d",
			 le16_to_cpu(sbi->s_es->s_reserved_gdt_blocks));
		goto failed_mount;
	}

R
Ross Zwisler 已提交
3635
	if (sbi->s_mount_opt & EXT4_MOUNT_DAX) {
3636 3637
		err = bdev_dax_supported(sb, blocksize);
		if (err)
R
Ross Zwisler 已提交
3638 3639 3640
			goto failed_mount;
	}

3641
	if (ext4_has_feature_encrypt(sb) && es->s_encryption_level) {
3642 3643 3644 3645 3646
		ext4_msg(sb, KERN_ERR, "Unsupported encryption level %d",
			 es->s_encryption_level);
		goto failed_mount;
	}

3647
	if (sb->s_blocksize != blocksize) {
3648 3649
		/* Validate the filesystem blocksize */
		if (!sb_set_blocksize(sb, blocksize)) {
3650
			ext4_msg(sb, KERN_ERR, "bad block size %d",
3651
					blocksize);
3652 3653 3654
			goto failed_mount;
		}

3655
		brelse(bh);
3656 3657
		logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE;
		offset = do_div(logical_sb_block, blocksize);
3658
		bh = sb_bread_unmovable(sb, logical_sb_block);
3659
		if (!bh) {
3660 3661
			ext4_msg(sb, KERN_ERR,
			       "Can't read superblock on 2nd try");
3662 3663
			goto failed_mount;
		}
3664
		es = (struct ext4_super_block *)(bh->b_data + offset);
3665
		sbi->s_es = es;
3666
		if (es->s_magic != cpu_to_le16(EXT4_SUPER_MAGIC)) {
3667 3668
			ext4_msg(sb, KERN_ERR,
			       "Magic mismatch, very weird!");
3669 3670 3671 3672
			goto failed_mount;
		}
	}

3673
	has_huge_files = ext4_has_feature_huge_file(sb);
3674 3675 3676
	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);
3677

3678 3679 3680
	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;
3681 3682 3683
	} else {
		sbi->s_inode_size = le16_to_cpu(es->s_inode_size);
		sbi->s_first_ino = le32_to_cpu(es->s_first_ino);
3684
		if ((sbi->s_inode_size < EXT4_GOOD_OLD_INODE_SIZE) ||
V
Vignesh Babu 已提交
3685
		    (!is_power_of_2(sbi->s_inode_size)) ||
3686
		    (sbi->s_inode_size > blocksize)) {
3687 3688
			ext4_msg(sb, KERN_ERR,
			       "unsupported inode size: %d",
3689
			       sbi->s_inode_size);
3690 3691
			goto failed_mount;
		}
K
Kalpak Shah 已提交
3692 3693
		if (sbi->s_inode_size > EXT4_GOOD_OLD_INODE_SIZE)
			sb->s_time_gran = 1 << (EXT4_EPOCH_BITS - 2);
3694
	}
3695

3696
	sbi->s_desc_size = le16_to_cpu(es->s_desc_size);
3697
	if (ext4_has_feature_64bit(sb)) {
3698
		if (sbi->s_desc_size < EXT4_MIN_DESC_SIZE_64BIT ||
3699
		    sbi->s_desc_size > EXT4_MAX_DESC_SIZE ||
V
vignesh babu 已提交
3700
		    !is_power_of_2(sbi->s_desc_size)) {
3701 3702
			ext4_msg(sb, KERN_ERR,
			       "unsupported descriptor size %lu",
3703 3704 3705 3706 3707
			       sbi->s_desc_size);
			goto failed_mount;
		}
	} else
		sbi->s_desc_size = EXT4_MIN_DESC_SIZE;
3708

3709 3710
	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);
3711

3712
	sbi->s_inodes_per_block = blocksize / EXT4_INODE_SIZE(sb);
3713
	if (sbi->s_inodes_per_block == 0)
3714
		goto cantfind_ext4;
3715 3716 3717 3718 3719 3720
	if (sbi->s_inodes_per_group < sbi->s_inodes_per_block ||
	    sbi->s_inodes_per_group > blocksize * 8) {
		ext4_msg(sb, KERN_ERR, "invalid inodes per group: %lu\n",
			 sbi->s_blocks_per_group);
		goto failed_mount;
	}
3721 3722
	sbi->s_itb_per_group = sbi->s_inodes_per_group /
					sbi->s_inodes_per_block;
3723
	sbi->s_desc_per_block = blocksize / EXT4_DESC_SIZE(sb);
3724 3725
	sbi->s_sbh = bh;
	sbi->s_mount_state = le16_to_cpu(es->s_state);
3726 3727
	sbi->s_addr_per_block_bits = ilog2(EXT4_ADDR_PER_BLOCK(sb));
	sbi->s_desc_per_block_bits = ilog2(EXT4_DESC_PER_BLOCK(sb));
3728

3729
	for (i = 0; i < 4; i++)
3730 3731
		sbi->s_hash_seed[i] = le32_to_cpu(es->s_hash_seed[i]);
	sbi->s_def_hash_version = es->s_def_hash_version;
3732
	if (ext4_has_feature_dir_index(sb)) {
3733 3734 3735 3736
		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) {
3737
#ifdef __CHAR_UNSIGNED__
3738 3739 3740 3741
			if (!(sb->s_flags & MS_RDONLY))
				es->s_flags |=
					cpu_to_le32(EXT2_FLAGS_UNSIGNED_HASH);
			sbi->s_hash_unsigned = 3;
3742
#else
3743 3744 3745
			if (!(sb->s_flags & MS_RDONLY))
				es->s_flags |=
					cpu_to_le32(EXT2_FLAGS_SIGNED_HASH);
3746
#endif
3747
		}
3748
	}
3749

3750 3751
	/* Handle clustersize */
	clustersize = BLOCK_SIZE << le32_to_cpu(es->s_log_cluster_size);
3752
	has_bigalloc = ext4_has_feature_bigalloc(sb);
3753 3754 3755 3756 3757 3758 3759
	if (has_bigalloc) {
		if (clustersize < blocksize) {
			ext4_msg(sb, KERN_ERR,
				 "cluster size (%d) smaller than "
				 "block size (%d)", clustersize, blocksize);
			goto failed_mount;
		}
3760 3761 3762 3763 3764 3765 3766
		if (le32_to_cpu(es->s_log_cluster_size) >
		    (EXT4_MAX_CLUSTER_LOG_SIZE - EXT4_MIN_BLOCK_LOG_SIZE)) {
			ext4_msg(sb, KERN_ERR,
				 "Invalid log cluster size: %u",
				 le32_to_cpu(es->s_log_cluster_size));
			goto failed_mount;
		}
3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799
		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;
3800
	}
3801 3802
	sbi->s_cluster_ratio = clustersize / blocksize;

3803 3804 3805 3806
	/* Do we have standard group size of clustersize * 8 blocks ? */
	if (sbi->s_blocks_per_group == clustersize << 3)
		set_opt2(sb, STD_GROUP_SIZE);

3807 3808 3809 3810
	/*
	 * Test whether we have more sectors than will fit in sector_t,
	 * and whether the max offset is addressable by the page cache.
	 */
3811
	err = generic_check_addressable(sb->s_blocksize_bits,
3812
					ext4_blocks_count(es));
3813
	if (err) {
3814
		ext4_msg(sb, KERN_ERR, "filesystem"
3815
			 " too large to mount safely on this system");
3816
		if (sizeof(sector_t) < 8)
3817
			ext4_msg(sb, KERN_WARNING, "CONFIG_LBDAF not enabled");
3818 3819 3820
		goto failed_mount;
	}

3821 3822
	if (EXT4_BLOCKS_PER_GROUP(sb) == 0)
		goto cantfind_ext4;
3823

3824 3825 3826
	/* 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) {
3827 3828
		ext4_msg(sb, KERN_WARNING, "bad geometry: block count %llu "
		       "exceeds size of device (%llu blocks)",
3829 3830 3831 3832
		       ext4_blocks_count(es), blocks_count);
		goto failed_mount;
	}

3833 3834 3835 3836 3837
	/*
	 * 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)) {
3838
		ext4_msg(sb, KERN_WARNING, "bad geometry: first data "
3839 3840 3841
			 "block %u is beyond end of filesystem (%llu)",
			 le32_to_cpu(es->s_first_data_block),
			 ext4_blocks_count(es));
3842 3843
		goto failed_mount;
	}
L
Laurent Vivier 已提交
3844 3845 3846 3847
	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));
3848
	if (blocks_count > ((uint64_t)1<<32) - EXT4_DESC_PER_BLOCK(sb)) {
3849
		ext4_msg(sb, KERN_WARNING, "groups count too large: %u "
3850
		       "(block count %llu, first data block %u, "
3851
		       "blocks per group %lu)", sbi->s_groups_count,
3852 3853 3854 3855 3856
		       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
3857
	sbi->s_groups_count = blocks_count;
3858 3859
	sbi->s_blockfile_groups = min_t(ext4_group_t, sbi->s_groups_count,
			(EXT4_MAX_BLOCK_FILE_PHYS / EXT4_BLOCKS_PER_GROUP(sb)));
3860 3861
	db_count = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) /
		   EXT4_DESC_PER_BLOCK(sb);
3862
	if (ext4_has_feature_meta_bg(sb)) {
3863
		if (le32_to_cpu(es->s_first_meta_bg) > db_count) {
3864 3865 3866 3867 3868 3869 3870
			ext4_msg(sb, KERN_WARNING,
				 "first meta block group too large: %u "
				 "(group descriptor block count %u)",
				 le32_to_cpu(es->s_first_meta_bg), db_count);
			goto failed_mount;
		}
	}
3871 3872 3873
	sbi->s_group_desc = ext4_kvmalloc(db_count *
					  sizeof(struct buffer_head *),
					  GFP_KERNEL);
3874
	if (sbi->s_group_desc == NULL) {
3875
		ext4_msg(sb, KERN_ERR, "not enough memory");
3876
		ret = -ENOMEM;
3877 3878 3879
		goto failed_mount;
	}

3880
	bgl_lock_init(sbi->s_blockgroup_lock);
3881 3882

	for (i = 0; i < db_count; i++) {
3883
		block = descriptor_loc(sb, logical_sb_block, i);
3884
		sbi->s_group_desc[i] = sb_bread_unmovable(sb, block);
3885
		if (!sbi->s_group_desc[i]) {
3886 3887
			ext4_msg(sb, KERN_ERR,
			       "can't read group descriptor %d", i);
3888 3889 3890 3891
			db_count = i;
			goto failed_mount2;
		}
	}
3892
	if (!ext4_check_descriptors(sb, logical_sb_block, &first_not_zeroed)) {
3893
		ext4_msg(sb, KERN_ERR, "group descriptors corrupted!");
3894
		ret = -EFSCORRUPTED;
3895
		goto failed_mount2;
3896
	}
3897

3898
	sbi->s_gdb_count = db_count;
3899 3900 3901
	get_random_bytes(&sbi->s_next_generation, sizeof(u32));
	spin_lock_init(&sbi->s_next_gen_lock);

3902 3903
	setup_timer(&sbi->s_err_report, print_daily_error_info,
		(unsigned long) sb);
3904

3905
	/* Register extent status tree shrinker */
3906
	if (ext4_es_register_shrinker(sbi))
3907 3908
		goto failed_mount3;

3909
	sbi->s_stripe = ext4_get_stripe_size(sbi);
3910
	sbi->s_extent_max_zeroout_kb = 32;
3911

3912 3913 3914
	/*
	 * set up enough so that it can read an inode
	 */
3915
	sb->s_op = &ext4_sops;
3916 3917
	sb->s_export_op = &ext4_export_ops;
	sb->s_xattr = ext4_xattr_handlers;
3918
	sb->s_cop = &ext4_cryptops;
3919
#ifdef CONFIG_QUOTA
3920
	sb->dq_op = &ext4_quota_operations;
3921
	if (ext4_has_feature_quota(sb))
3922
		sb->s_qcop = &dquot_quotactl_sysfile_ops;
3923 3924
	else
		sb->s_qcop = &ext4_qctl_operations;
L
Li Xi 已提交
3925
	sb->s_quota_types = QTYPE_MASK_USR | QTYPE_MASK_GRP | QTYPE_MASK_PRJ;
3926
#endif
3927 3928
	memcpy(sb->s_uuid, es->s_uuid, sizeof(es->s_uuid));

3929
	INIT_LIST_HEAD(&sbi->s_orphan); /* unlinked but open files */
3930
	mutex_init(&sbi->s_orphan_lock);
3931 3932 3933 3934

	sb->s_root = NULL;

	needs_recovery = (es->s_last_orphan != 0 ||
3935
			  ext4_has_feature_journal_needs_recovery(sb));
3936

3937
	if (ext4_has_feature_mmp(sb) && !(sb->s_flags & MS_RDONLY))
3938
		if (ext4_multi_mount_protect(sb, le64_to_cpu(es->s_mmp_block)))
3939
			goto failed_mount3a;
3940

3941 3942 3943 3944
	/*
	 * The first inode we look at is the journal inode.  Don't try
	 * root first: it may be modified in the journal!
	 */
3945
	if (!test_opt(sb, NOLOAD) && ext4_has_feature_journal(sb)) {
3946 3947
		err = ext4_load_journal(sb, es, journal_devnum);
		if (err)
3948
			goto failed_mount3a;
3949
	} else if (test_opt(sb, NOLOAD) && !(sb->s_flags & MS_RDONLY) &&
3950
		   ext4_has_feature_journal_needs_recovery(sb)) {
3951 3952
		ext4_msg(sb, KERN_ERR, "required journal recovery "
		       "suppressed and not mounted read-only");
3953
		goto failed_mount_wq;
3954
	} else {
3955 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976 3977 3978 3979
		/* Nojournal mode, all journal mount options are illegal */
		if (test_opt2(sb, EXPLICIT_JOURNAL_CHECKSUM)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "journal_checksum, fs mounted w/o journal");
			goto failed_mount_wq;
		}
		if (test_opt(sb, JOURNAL_ASYNC_COMMIT)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "journal_async_commit, fs mounted w/o journal");
			goto failed_mount_wq;
		}
		if (sbi->s_commit_interval != JBD2_DEFAULT_MAX_COMMIT_AGE*HZ) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "commit=%lu, fs mounted w/o journal",
				 sbi->s_commit_interval / HZ);
			goto failed_mount_wq;
		}
		if (EXT4_MOUNT_DATA_FLAGS &
		    (sbi->s_mount_opt ^ sbi->s_def_mount_opt)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "data=, fs mounted w/o journal");
			goto failed_mount_wq;
		}
		sbi->s_def_mount_opt &= EXT4_MOUNT_JOURNAL_CHECKSUM;
		clear_opt(sb, JOURNAL_CHECKSUM);
3980
		clear_opt(sb, DATA_FLAGS);
3981 3982 3983
		sbi->s_journal = NULL;
		needs_recovery = 0;
		goto no_journal;
3984 3985
	}

3986
	if (ext4_has_feature_64bit(sb) &&
3987 3988
	    !jbd2_journal_set_features(EXT4_SB(sb)->s_journal, 0, 0,
				       JBD2_FEATURE_INCOMPAT_64BIT)) {
3989
		ext4_msg(sb, KERN_ERR, "Failed to set 64-bit journal feature");
3990
		goto failed_mount_wq;
3991 3992
	}

3993 3994 3995 3996
	if (!set_journal_csum_feature_set(sb)) {
		ext4_msg(sb, KERN_ERR, "Failed to set journal checksum "
			 "feature set");
		goto failed_mount_wq;
3997
	}
3998

3999 4000 4001 4002 4003
	/* 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
4004 4005 4006
		 * capabilities: ORDERED_DATA if the journal can
		 * cope, else JOURNAL_DATA
		 */
4007 4008
		if (jbd2_journal_check_available_features
		    (sbi->s_journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE))
4009
			set_opt(sb, ORDERED_DATA);
4010
		else
4011
			set_opt(sb, JOURNAL_DATA);
4012 4013
		break;

4014 4015
	case EXT4_MOUNT_ORDERED_DATA:
	case EXT4_MOUNT_WRITEBACK_DATA:
4016 4017
		if (!jbd2_journal_check_available_features
		    (sbi->s_journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE)) {
4018 4019
			ext4_msg(sb, KERN_ERR, "Journal does not support "
			       "requested data journaling mode");
4020
			goto failed_mount_wq;
4021 4022 4023 4024
		}
	default:
		break;
	}
4025 4026 4027 4028 4029 4030 4031 4032

	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");
		goto failed_mount_wq;
	}

4033
	set_task_ioprio(sbi->s_journal->j_task, journal_ioprio);
4034

B
Bobi Jam 已提交
4035 4036
	sbi->s_journal->j_commit_callback = ext4_journal_commit_callback;

4037
no_journal:
4038 4039 4040 4041
	sbi->s_mb_cache = ext4_xattr_create_cache();
	if (!sbi->s_mb_cache) {
		ext4_msg(sb, KERN_ERR, "Failed to create an mb_cache");
		goto failed_mount_wq;
4042 4043
	}

4044
	if ((DUMMY_ENCRYPTION_ENABLED(sbi) || ext4_has_feature_encrypt(sb)) &&
4045
	    (blocksize != PAGE_SIZE)) {
4046 4047 4048 4049 4050
		ext4_msg(sb, KERN_ERR,
			 "Unsupported blocksize for fs encryption");
		goto failed_mount_wq;
	}

4051 4052 4053
	if (DUMMY_ENCRYPTION_ENABLED(sbi) && !(sb->s_flags & MS_RDONLY) &&
	    !ext4_has_feature_encrypt(sb)) {
		ext4_set_feature_encrypt(sb);
4054 4055 4056
		ext4_commit_super(sb, 1);
	}

4057 4058 4059 4060 4061 4062 4063
	/*
	 * 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 {
4064 4065
		err = ext4_calculate_overhead(sb);
		if (err)
4066 4067 4068
			goto failed_mount_wq;
	}

T
Tejun Heo 已提交
4069 4070 4071 4072
	/*
	 * The maximum number of concurrent works can be high and
	 * concurrency isn't really necessary.  Limit it to 1.
	 */
4073 4074 4075 4076
	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");
4077
		ret = -ENOMEM;
4078 4079 4080
		goto failed_mount4;
	}

4081
	/*
4082
	 * The jbd2_journal_load will have done any necessary log recovery,
4083 4084 4085
	 * so we can safely mount the rest of the filesystem now.
	 */

4086 4087
	root = ext4_iget(sb, EXT4_ROOT_INO);
	if (IS_ERR(root)) {
4088
		ext4_msg(sb, KERN_ERR, "get root inode failed");
4089
		ret = PTR_ERR(root);
4090
		root = NULL;
4091 4092 4093
		goto failed_mount4;
	}
	if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) {
4094
		ext4_msg(sb, KERN_ERR, "corrupt root inode, run e2fsck");
A
Al Viro 已提交
4095
		iput(root);
4096 4097
		goto failed_mount4;
	}
4098
	sb->s_root = d_make_root(root);
4099
	if (!sb->s_root) {
4100
		ext4_msg(sb, KERN_ERR, "get root dentry failed");
4101 4102 4103
		ret = -ENOMEM;
		goto failed_mount4;
	}
4104

4105 4106
	if (ext4_setup_super(sb, es, sb->s_flags & MS_RDONLY))
		sb->s_flags |= MS_RDONLY;
K
Kalpak Shah 已提交
4107 4108

	/* determine the minimum size of new large inodes, if present */
4109 4110
	if (sbi->s_inode_size > EXT4_GOOD_OLD_INODE_SIZE &&
	    sbi->s_want_extra_isize == 0) {
K
Kalpak Shah 已提交
4111 4112
		sbi->s_want_extra_isize = sizeof(struct ext4_inode) -
						     EXT4_GOOD_OLD_INODE_SIZE;
4113
		if (ext4_has_feature_extra_isize(sb)) {
K
Kalpak Shah 已提交
4114 4115 4116 4117 4118 4119 4120 4121 4122 4123 4124 4125 4126 4127 4128
			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;
4129 4130
		ext4_msg(sb, KERN_INFO, "required extra inode space not"
			 "available");
K
Kalpak Shah 已提交
4131 4132
	}

4133
	ext4_set_resv_clusters(sb);
L
Lukas Czerner 已提交
4134

4135 4136
	err = ext4_setup_system_zone(sb);
	if (err) {
4137
		ext4_msg(sb, KERN_ERR, "failed to initialize system "
4138
			 "zone (%d)", err);
4139 4140 4141 4142 4143 4144 4145 4146
		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);
4147
		goto failed_mount5;
4148 4149
	}

4150 4151 4152
	block = ext4_count_free_clusters(sb);
	ext4_free_blocks_count_set(sbi->s_es, 
				   EXT4_C2B(sbi, block));
4153 4154
	err = percpu_counter_init(&sbi->s_freeclusters_counter, block,
				  GFP_KERNEL);
4155 4156 4157
	if (!err) {
		unsigned long freei = ext4_count_free_inodes(sb);
		sbi->s_es->s_free_inodes_count = cpu_to_le32(freei);
4158 4159
		err = percpu_counter_init(&sbi->s_freeinodes_counter, freei,
					  GFP_KERNEL);
4160 4161 4162
	}
	if (!err)
		err = percpu_counter_init(&sbi->s_dirs_counter,
4163
					  ext4_count_dirs(sb), GFP_KERNEL);
4164
	if (!err)
4165 4166
		err = percpu_counter_init(&sbi->s_dirtyclusters_counter, 0,
					  GFP_KERNEL);
4167 4168 4169
	if (!err)
		err = percpu_init_rwsem(&sbi->s_journal_flag_rwsem);

4170 4171 4172 4173 4174
	if (err) {
		ext4_msg(sb, KERN_ERR, "insufficient memory");
		goto failed_mount6;
	}

4175
	if (ext4_has_feature_flex_bg(sb))
4176 4177 4178 4179 4180 4181 4182
		if (!ext4_fill_flex_info(sb)) {
			ext4_msg(sb, KERN_ERR,
			       "unable to initialize "
			       "flex_bg meta info!");
			goto failed_mount6;
		}

4183 4184
	err = ext4_register_li_request(sb, first_not_zeroed);
	if (err)
4185
		goto failed_mount6;
4186

4187
	err = ext4_register_sysfs(sb);
4188 4189
	if (err)
		goto failed_mount7;
T
Theodore Ts'o 已提交
4190

4191 4192
#ifdef CONFIG_QUOTA
	/* Enable quota usage during mount. */
4193
	if (ext4_has_feature_quota(sb) && !(sb->s_flags & MS_RDONLY)) {
4194 4195 4196 4197 4198 4199
		err = ext4_enable_quotas(sb);
		if (err)
			goto failed_mount8;
	}
#endif  /* CONFIG_QUOTA */

4200 4201 4202
	EXT4_SB(sb)->s_mount_state |= EXT4_ORPHAN_FS;
	ext4_orphan_cleanup(sb, es);
	EXT4_SB(sb)->s_mount_state &= ~EXT4_ORPHAN_FS;
4203
	if (needs_recovery) {
4204
		ext4_msg(sb, KERN_INFO, "recovery complete");
4205 4206 4207 4208 4209 4210 4211 4212 4213 4214 4215 4216
		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";

4217 4218 4219 4220 4221 4222 4223 4224
	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");
	}

4225 4226
	if (___ratelimit(&ext4_mount_msg_ratelimit, "EXT4-fs mount"))
		ext4_msg(sb, KERN_INFO, "mounted filesystem with%s. "
4227 4228 4229
			 "Opts: %.*s%s%s", descr,
			 (int) sizeof(sbi->s_es->s_mount_opts),
			 sbi->s_es->s_mount_opts,
4230
			 *sbi->s_es->s_mount_opts ? "; " : "", orig_data);
4231

4232 4233
	if (es->s_error_count)
		mod_timer(&sbi->s_err_report, jiffies + 300*HZ); /* 5 minutes */
4234

4235 4236 4237 4238 4239
	/* 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);

4240
	kfree(orig_data);
4241 4242
	return 0;

4243
cantfind_ext4:
4244
	if (!silent)
4245
		ext4_msg(sb, KERN_ERR, "VFS: Can't find ext4 filesystem");
4246 4247
	goto failed_mount;

4248 4249
#ifdef CONFIG_QUOTA
failed_mount8:
4250
	ext4_unregister_sysfs(sb);
4251
#endif
4252 4253 4254
failed_mount7:
	ext4_unregister_li_request(sb);
failed_mount6:
4255
	ext4_mb_release(sb);
4256
	if (sbi->s_flex_groups)
A
Al Viro 已提交
4257
		kvfree(sbi->s_flex_groups);
4258 4259 4260 4261
	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);
4262
failed_mount5:
4263 4264 4265
	ext4_ext_release(sb);
	ext4_release_system_zone(sb);
failed_mount4a:
A
Al Viro 已提交
4266
	dput(sb->s_root);
4267
	sb->s_root = NULL;
A
Al Viro 已提交
4268
failed_mount4:
4269
	ext4_msg(sb, KERN_ERR, "mount failed");
4270 4271
	if (EXT4_SB(sb)->rsv_conversion_wq)
		destroy_workqueue(EXT4_SB(sb)->rsv_conversion_wq);
4272
failed_mount_wq:
J
Jan Kara 已提交
4273 4274 4275 4276
	if (sbi->s_mb_cache) {
		ext4_xattr_destroy_cache(sbi->s_mb_cache);
		sbi->s_mb_cache = NULL;
	}
4277 4278 4279 4280
	if (sbi->s_journal) {
		jbd2_journal_destroy(sbi->s_journal);
		sbi->s_journal = NULL;
	}
4281
failed_mount3a:
4282
	ext4_es_unregister_shrinker(sbi);
4283
failed_mount3:
4284
	del_timer_sync(&sbi->s_err_report);
4285 4286
	if (sbi->s_mmp_tsk)
		kthread_stop(sbi->s_mmp_tsk);
4287 4288 4289
failed_mount2:
	for (i = 0; i < db_count; i++)
		brelse(sbi->s_group_desc[i]);
A
Al Viro 已提交
4290
	kvfree(sbi->s_group_desc);
4291
failed_mount:
4292 4293
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
4294
#ifdef CONFIG_QUOTA
J
Jan Kara 已提交
4295
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
4296 4297
		kfree(sbi->s_qf_names[i]);
#endif
4298
	ext4_blkdev_remove(sbi);
4299 4300 4301
	brelse(bh);
out_fail:
	sb->s_fs_info = NULL;
4302
	kfree(sbi->s_blockgroup_lock);
4303
out_free_base:
4304
	kfree(sbi);
4305
	kfree(orig_data);
4306
	return err ? err : ret;
4307 4308 4309 4310 4311 4312 4313
}

/*
 * 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.
 */
4314
static void ext4_init_journal_params(struct super_block *sb, journal_t *journal)
4315
{
4316
	struct ext4_sb_info *sbi = EXT4_SB(sb);
4317

4318 4319 4320
	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;
4321

4322
	write_lock(&journal->j_state_lock);
4323
	if (test_opt(sb, BARRIER))
4324
		journal->j_flags |= JBD2_BARRIER;
4325
	else
4326
		journal->j_flags &= ~JBD2_BARRIER;
4327 4328 4329 4330
	if (test_opt(sb, DATA_ERR_ABORT))
		journal->j_flags |= JBD2_ABORT_ON_SYNCDATA_ERR;
	else
		journal->j_flags &= ~JBD2_ABORT_ON_SYNCDATA_ERR;
4331
	write_unlock(&journal->j_state_lock);
4332 4333
}

4334 4335
static struct inode *ext4_get_journal_inode(struct super_block *sb,
					     unsigned int journal_inum)
4336 4337 4338
{
	struct inode *journal_inode;

4339 4340 4341 4342 4343
	/*
	 * 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.
	 */
4344 4345
	journal_inode = ext4_iget(sb, journal_inum);
	if (IS_ERR(journal_inode)) {
4346
		ext4_msg(sb, KERN_ERR, "no journal found");
4347 4348 4349 4350 4351
		return NULL;
	}
	if (!journal_inode->i_nlink) {
		make_bad_inode(journal_inode);
		iput(journal_inode);
4352
		ext4_msg(sb, KERN_ERR, "journal inode is deleted");
4353 4354 4355
		return NULL;
	}

4356
	jbd_debug(2, "Journal inode found at %p: %lld bytes\n",
4357
		  journal_inode, journal_inode->i_size);
4358
	if (!S_ISREG(journal_inode->i_mode)) {
4359
		ext4_msg(sb, KERN_ERR, "invalid journal inode");
4360 4361 4362
		iput(journal_inode);
		return NULL;
	}
4363 4364 4365 4366 4367 4368 4369 4370 4371 4372 4373 4374 4375 4376
	return journal_inode;
}

static journal_t *ext4_get_journal(struct super_block *sb,
				   unsigned int journal_inum)
{
	struct inode *journal_inode;
	journal_t *journal;

	BUG_ON(!ext4_has_feature_journal(sb));

	journal_inode = ext4_get_journal_inode(sb, journal_inum);
	if (!journal_inode)
		return NULL;
4377

4378
	journal = jbd2_journal_init_inode(journal_inode);
4379
	if (!journal) {
4380
		ext4_msg(sb, KERN_ERR, "Could not load journal inode");
4381 4382 4383 4384
		iput(journal_inode);
		return NULL;
	}
	journal->j_private = sb;
4385
	ext4_init_journal_params(sb, journal);
4386 4387 4388
	return journal;
}

4389
static journal_t *ext4_get_dev_journal(struct super_block *sb,
4390 4391
				       dev_t j_dev)
{
4392
	struct buffer_head *bh;
4393
	journal_t *journal;
4394 4395
	ext4_fsblk_t start;
	ext4_fsblk_t len;
4396
	int hblock, blocksize;
4397
	ext4_fsblk_t sb_block;
4398
	unsigned long offset;
4399
	struct ext4_super_block *es;
4400 4401
	struct block_device *bdev;

4402
	BUG_ON(!ext4_has_feature_journal(sb));
4403

4404
	bdev = ext4_blkdev_get(j_dev, sb);
4405 4406 4407 4408
	if (bdev == NULL)
		return NULL;

	blocksize = sb->s_blocksize;
4409
	hblock = bdev_logical_block_size(bdev);
4410
	if (blocksize < hblock) {
4411 4412
		ext4_msg(sb, KERN_ERR,
			"blocksize too small for journal device");
4413 4414 4415
		goto out_bdev;
	}

4416 4417
	sb_block = EXT4_MIN_BLOCK_SIZE / blocksize;
	offset = EXT4_MIN_BLOCK_SIZE % blocksize;
4418 4419
	set_blocksize(bdev, blocksize);
	if (!(bh = __bread(bdev, sb_block, blocksize))) {
4420 4421
		ext4_msg(sb, KERN_ERR, "couldn't read superblock of "
		       "external journal");
4422 4423 4424
		goto out_bdev;
	}

4425
	es = (struct ext4_super_block *) (bh->b_data + offset);
4426
	if ((le16_to_cpu(es->s_magic) != EXT4_SUPER_MAGIC) ||
4427
	    !(le32_to_cpu(es->s_feature_incompat) &
4428
	      EXT4_FEATURE_INCOMPAT_JOURNAL_DEV)) {
4429 4430
		ext4_msg(sb, KERN_ERR, "external journal has "
					"bad superblock");
4431 4432 4433 4434
		brelse(bh);
		goto out_bdev;
	}

4435 4436 4437 4438 4439 4440 4441 4442 4443
	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;
	}

4444
	if (memcmp(EXT4_SB(sb)->s_es->s_journal_uuid, es->s_uuid, 16)) {
4445
		ext4_msg(sb, KERN_ERR, "journal UUID does not match");
4446 4447 4448 4449
		brelse(bh);
		goto out_bdev;
	}

L
Laurent Vivier 已提交
4450
	len = ext4_blocks_count(es);
4451 4452 4453
	start = sb_block + 1;
	brelse(bh);	/* we're done with the superblock */

4454
	journal = jbd2_journal_init_dev(bdev, sb->s_bdev,
4455 4456
					start, len, blocksize);
	if (!journal) {
4457
		ext4_msg(sb, KERN_ERR, "failed to create device journal");
4458 4459 4460
		goto out_bdev;
	}
	journal->j_private = sb;
4461
	ll_rw_block(REQ_OP_READ, REQ_META | REQ_PRIO, 1, &journal->j_sb_buffer);
4462 4463
	wait_on_buffer(journal->j_sb_buffer);
	if (!buffer_uptodate(journal->j_sb_buffer)) {
4464
		ext4_msg(sb, KERN_ERR, "I/O error on journal device");
4465 4466 4467
		goto out_journal;
	}
	if (be32_to_cpu(journal->j_superblock->s_nr_users) != 1) {
4468 4469
		ext4_msg(sb, KERN_ERR, "External journal has more than one "
					"user (unsupported) - %d",
4470 4471 4472
			be32_to_cpu(journal->j_superblock->s_nr_users));
		goto out_journal;
	}
4473 4474
	EXT4_SB(sb)->journal_bdev = bdev;
	ext4_init_journal_params(sb, journal);
4475
	return journal;
4476

4477
out_journal:
4478
	jbd2_journal_destroy(journal);
4479
out_bdev:
4480
	ext4_blkdev_put(bdev);
4481 4482 4483
	return NULL;
}

4484 4485
static int ext4_load_journal(struct super_block *sb,
			     struct ext4_super_block *es,
4486 4487 4488 4489 4490 4491 4492 4493
			     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;

4494
	BUG_ON(!ext4_has_feature_journal(sb));
4495

4496 4497
	if (journal_devnum &&
	    journal_devnum != le32_to_cpu(es->s_journal_dev)) {
4498 4499
		ext4_msg(sb, KERN_INFO, "external journal device major/minor "
			"numbers have changed");
4500 4501 4502 4503 4504 4505 4506 4507 4508 4509 4510
		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.
	 */
4511
	if (ext4_has_feature_journal_needs_recovery(sb)) {
4512
		if (sb->s_flags & MS_RDONLY) {
4513 4514
			ext4_msg(sb, KERN_INFO, "INFO: recovery "
					"required on readonly filesystem");
4515
			if (really_read_only) {
4516 4517
				ext4_msg(sb, KERN_ERR, "write access "
					"unavailable, cannot proceed");
4518 4519
				return -EROFS;
			}
4520 4521
			ext4_msg(sb, KERN_INFO, "write access will "
			       "be enabled during recovery");
4522 4523 4524 4525
		}
	}

	if (journal_inum && journal_dev) {
4526 4527
		ext4_msg(sb, KERN_ERR, "filesystem has both journal "
		       "and inode journals!");
4528 4529 4530 4531
		return -EINVAL;
	}

	if (journal_inum) {
4532
		if (!(journal = ext4_get_journal(sb, journal_inum)))
4533 4534
			return -EINVAL;
	} else {
4535
		if (!(journal = ext4_get_dev_journal(sb, journal_dev)))
4536 4537 4538
			return -EINVAL;
	}

4539
	if (!(journal->j_flags & JBD2_BARRIER))
4540
		ext4_msg(sb, KERN_INFO, "barriers disabled");
4541

4542
	if (!ext4_has_feature_journal_needs_recovery(sb))
4543
		err = jbd2_journal_wipe(journal, !really_read_only);
4544 4545 4546 4547 4548
	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);
4549
		err = jbd2_journal_load(journal);
4550 4551 4552 4553 4554
		if (save)
			memcpy(((char *) es) + EXT4_S_ERR_START,
			       save, EXT4_S_ERR_LEN);
		kfree(save);
	}
4555 4556

	if (err) {
4557
		ext4_msg(sb, KERN_ERR, "error loading journal");
4558
		jbd2_journal_destroy(journal);
4559 4560 4561
		return err;
	}

4562 4563
	EXT4_SB(sb)->s_journal = journal;
	ext4_clear_journal_err(sb, es);
4564

4565
	if (!really_read_only && journal_devnum &&
4566 4567 4568 4569
	    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. */
4570
		ext4_commit_super(sb, 1);
4571 4572 4573 4574 4575
	}

	return 0;
}

4576
static int ext4_commit_super(struct super_block *sb, int sync)
4577
{
4578
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;
4579
	struct buffer_head *sbh = EXT4_SB(sb)->s_sbh;
4580
	int error = 0;
4581

4582
	if (!sbh || block_device_ejected(sb))
4583
		return error;
4584 4585 4586 4587 4588 4589 4590 4591 4592 4593 4594 4595
	/*
	 * 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());
4596 4597 4598
	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
4599 4600
			    ((part_stat_read(sb->s_bdev->bd_part, sectors[1]) -
			      EXT4_SB(sb)->s_sectors_written_start) >> 1));
4601 4602 4603
	else
		es->s_kbytes_written =
			cpu_to_le64(EXT4_SB(sb)->s_kbytes_written);
4604 4605
	if (percpu_counter_initialized(&EXT4_SB(sb)->s_freeclusters_counter))
		ext4_free_blocks_count_set(es,
4606 4607
			EXT4_C2B(EXT4_SB(sb), percpu_counter_sum_positive(
				&EXT4_SB(sb)->s_freeclusters_counter)));
4608 4609 4610
	if (percpu_counter_initialized(&EXT4_SB(sb)->s_freeinodes_counter))
		es->s_free_inodes_count =
			cpu_to_le32(percpu_counter_sum_positive(
4611
				&EXT4_SB(sb)->s_freeinodes_counter));
4612
	BUFFER_TRACE(sbh, "marking dirty");
4613
	ext4_superblock_csum_set(sb);
4614 4615
	if (sync)
		lock_buffer(sbh);
4616 4617 4618 4619 4620 4621 4622 4623 4624 4625 4626 4627 4628 4629
	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.
		 */
		ext4_msg(sb, KERN_ERR, "previous I/O error to "
		       "superblock detected");
		clear_buffer_write_io_error(sbh);
		set_buffer_uptodate(sbh);
	}
4630
	mark_buffer_dirty(sbh);
4631
	if (sync) {
4632
		unlock_buffer(sbh);
4633
		error = __sync_dirty_buffer(sbh,
4634
			test_opt(sb, BARRIER) ? REQ_FUA : REQ_SYNC);
4635 4636 4637 4638 4639
		if (error)
			return error;

		error = buffer_write_io_error(sbh);
		if (error) {
4640 4641
			ext4_msg(sb, KERN_ERR, "I/O error while writing "
			       "superblock");
4642 4643 4644 4645
			clear_buffer_write_io_error(sbh);
			set_buffer_uptodate(sbh);
		}
	}
4646
	return error;
4647 4648 4649 4650 4651 4652 4653
}

/*
 * 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.
 */
4654 4655
static void ext4_mark_recovery_complete(struct super_block *sb,
					struct ext4_super_block *es)
4656
{
4657
	journal_t *journal = EXT4_SB(sb)->s_journal;
4658

4659
	if (!ext4_has_feature_journal(sb)) {
4660 4661 4662
		BUG_ON(journal != NULL);
		return;
	}
4663
	jbd2_journal_lock_updates(journal);
4664 4665 4666
	if (jbd2_journal_flush(journal) < 0)
		goto out;

4667
	if (ext4_has_feature_journal_needs_recovery(sb) &&
4668
	    sb->s_flags & MS_RDONLY) {
4669
		ext4_clear_feature_journal_needs_recovery(sb);
4670
		ext4_commit_super(sb, 1);
4671
	}
4672 4673

out:
4674
	jbd2_journal_unlock_updates(journal);
4675 4676 4677 4678 4679 4680 4681
}

/*
 * 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.
 */
4682 4683
static void ext4_clear_journal_err(struct super_block *sb,
				   struct ext4_super_block *es)
4684 4685 4686 4687 4688
{
	journal_t *journal;
	int j_errno;
	const char *errstr;

4689
	BUG_ON(!ext4_has_feature_journal(sb));
4690

4691
	journal = EXT4_SB(sb)->s_journal;
4692 4693 4694

	/*
	 * Now check for any error status which may have been recorded in the
4695
	 * journal by a prior ext4_error() or ext4_abort()
4696 4697
	 */

4698
	j_errno = jbd2_journal_errno(journal);
4699 4700 4701
	if (j_errno) {
		char nbuf[16];

4702
		errstr = ext4_decode_error(sb, j_errno, nbuf);
4703
		ext4_warning(sb, "Filesystem error recorded "
4704
			     "from previous mount: %s", errstr);
4705
		ext4_warning(sb, "Marking fs in need of filesystem check.");
4706

4707 4708
		EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS;
		es->s_state |= cpu_to_le16(EXT4_ERROR_FS);
4709
		ext4_commit_super(sb, 1);
4710

4711
		jbd2_journal_clear_err(journal);
4712
		jbd2_journal_update_sb_errno(journal);
4713 4714 4715 4716 4717 4718 4719
	}
}

/*
 * Force the running and committing transactions to commit,
 * and wait on the commit.
 */
4720
int ext4_force_commit(struct super_block *sb)
4721 4722 4723 4724 4725 4726
{
	journal_t *journal;

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

4727
	journal = EXT4_SB(sb)->s_journal;
4728
	return ext4_journal_force_commit(journal);
4729 4730
}

4731
static int ext4_sync_fs(struct super_block *sb, int wait)
4732
{
4733
	int ret = 0;
4734
	tid_t target;
4735
	bool needs_barrier = false;
4736
	struct ext4_sb_info *sbi = EXT4_SB(sb);
4737

4738 4739 4740
	if (unlikely(ext4_forced_shutdown(EXT4_SB(sb))))
		return 0;

4741
	trace_ext4_sync_fs(sb, wait);
4742
	flush_workqueue(sbi->rsv_conversion_wq);
4743 4744 4745 4746 4747
	/*
	 * Writeback quota in non-journalled quota case - journalled quota has
	 * no dirty dquots
	 */
	dquot_writeback_dquots(sb, -1);
4748 4749 4750 4751 4752
	/*
	 * 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.
	 */
4753 4754 4755 4756 4757 4758 4759 4760 4761 4762 4763 4764
	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))
4765 4766 4767 4768 4769 4770
		needs_barrier = true;
	if (needs_barrier) {
		int err;
		err = blkdev_issue_flush(sb->s_bdev, GFP_KERNEL, NULL);
		if (!ret)
			ret = err;
4771
	}
4772 4773 4774 4775

	return ret;
}

4776 4777 4778
/*
 * 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.
4779 4780
 *
 * Note that only this function cannot bring a filesystem to be in a clean
4781 4782
 * state independently. It relies on upper layer to stop all data & metadata
 * modifications.
4783
 */
4784
static int ext4_freeze(struct super_block *sb)
4785
{
4786 4787
	int error = 0;
	journal_t *journal;
4788

4789 4790
	if (sb->s_flags & MS_RDONLY)
		return 0;
4791

4792
	journal = EXT4_SB(sb)->s_journal;
4793

4794 4795 4796
	if (journal) {
		/* Now we set up the journal barrier. */
		jbd2_journal_lock_updates(journal);
4797

4798 4799 4800 4801 4802 4803 4804
		/*
		 * Don't clear the needs_recovery flag if we failed to
		 * flush the journal.
		 */
		error = jbd2_journal_flush(journal);
		if (error < 0)
			goto out;
4805 4806

		/* Journal blocked and flushed, clear needs_recovery flag. */
4807
		ext4_clear_feature_journal_needs_recovery(sb);
4808
	}
4809 4810

	error = ext4_commit_super(sb, 1);
4811
out:
4812 4813 4814
	if (journal)
		/* we rely on upper layer to stop further updates */
		jbd2_journal_unlock_updates(journal);
4815
	return error;
4816 4817 4818 4819 4820 4821
}

/*
 * 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.
 */
4822
static int ext4_unfreeze(struct super_block *sb)
4823
{
4824
	if ((sb->s_flags & MS_RDONLY) || ext4_forced_shutdown(EXT4_SB(sb)))
4825 4826
		return 0;

4827 4828
	if (EXT4_SB(sb)->s_journal) {
		/* Reset the needs_recovery flag before the fs is unlocked. */
4829
		ext4_set_feature_journal_needs_recovery(sb);
4830 4831
	}

4832
	ext4_commit_super(sb, 1);
4833
	return 0;
4834 4835
}

4836 4837 4838 4839 4840
/*
 * Structure to save mount options for ext4_remount's benefit
 */
struct ext4_mount_options {
	unsigned long s_mount_opt;
4841
	unsigned long s_mount_opt2;
4842 4843
	kuid_t s_resuid;
	kgid_t s_resgid;
4844 4845 4846 4847
	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
4848
	char *s_qf_names[EXT4_MAXQUOTAS];
4849 4850 4851
#endif
};

4852
static int ext4_remount(struct super_block *sb, int *flags, char *data)
4853
{
4854
	struct ext4_super_block *es;
4855
	struct ext4_sb_info *sbi = EXT4_SB(sb);
4856
	unsigned long old_sb_flags;
4857
	struct ext4_mount_options old_opts;
4858
	int enable_quota = 0;
4859
	ext4_group_t g;
4860
	unsigned int journal_ioprio = DEFAULT_JOURNAL_IOPRIO;
4861
	int err = 0;
4862
#ifdef CONFIG_QUOTA
4863
	int i, j;
4864
#endif
4865
	char *orig_data = kstrdup(data, GFP_KERNEL);
4866 4867 4868 4869

	/* Store the original options */
	old_sb_flags = sb->s_flags;
	old_opts.s_mount_opt = sbi->s_mount_opt;
4870
	old_opts.s_mount_opt2 = sbi->s_mount_opt2;
4871 4872 4873
	old_opts.s_resuid = sbi->s_resuid;
	old_opts.s_resgid = sbi->s_resgid;
	old_opts.s_commit_interval = sbi->s_commit_interval;
4874 4875
	old_opts.s_min_batch_time = sbi->s_min_batch_time;
	old_opts.s_max_batch_time = sbi->s_max_batch_time;
4876 4877
#ifdef CONFIG_QUOTA
	old_opts.s_jquota_fmt = sbi->s_jquota_fmt;
J
Jan Kara 已提交
4878
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
4879 4880 4881 4882 4883 4884
		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]);
4885
				kfree(orig_data);
4886 4887 4888 4889
				return -ENOMEM;
			}
		} else
			old_opts.s_qf_names[i] = NULL;
4890
#endif
4891 4892
	if (sbi->s_journal && sbi->s_journal->j_task->io_context)
		journal_ioprio = sbi->s_journal->j_task->io_context->ioprio;
4893

4894
	if (!parse_options(data, sb, NULL, &journal_ioprio, 1)) {
4895 4896 4897 4898
		err = -EINVAL;
		goto restore_opts;
	}

4899
	if ((old_opts.s_mount_opt & EXT4_MOUNT_JOURNAL_CHECKSUM) ^
4900 4901
	    test_opt(sb, JOURNAL_CHECKSUM)) {
		ext4_msg(sb, KERN_ERR, "changing journal_checksum "
4902 4903
			 "during remount not supported; ignoring");
		sbi->s_mount_opt ^= EXT4_MOUNT_JOURNAL_CHECKSUM;
4904 4905
	}

4906 4907 4908 4909 4910 4911 4912 4913 4914 4915 4916 4917 4918
	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
4919 4920 4921 4922 4923 4924
		if (test_opt(sb, DAX)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "both data=journal and dax");
			err = -EINVAL;
			goto restore_opts;
		}
4925 4926 4927 4928 4929 4930 4931
	} else if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA) {
		if (test_opt(sb, JOURNAL_ASYNC_COMMIT)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				"journal_async_commit in data=ordered mode");
			err = -EINVAL;
			goto restore_opts;
		}
R
Ross Zwisler 已提交
4932 4933 4934 4935 4936 4937
	}

	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;
4938 4939
	}

4940
	if (sbi->s_mount_flags & EXT4_MF_FS_ABORTED)
4941
		ext4_abort(sb, "Abort forced by user");
4942 4943

	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
4944
		(test_opt(sb, POSIX_ACL) ? MS_POSIXACL : 0);
4945 4946 4947

	es = sbi->s_es;

4948
	if (sbi->s_journal) {
4949
		ext4_init_journal_params(sb, sbi->s_journal);
4950 4951
		set_task_ioprio(sbi->s_journal->j_task, journal_ioprio);
	}
4952

4953 4954 4955
	if (*flags & MS_LAZYTIME)
		sb->s_flags |= MS_LAZYTIME;

4956
	if ((*flags & MS_RDONLY) != (sb->s_flags & MS_RDONLY)) {
4957
		if (sbi->s_mount_flags & EXT4_MF_FS_ABORTED) {
4958 4959 4960 4961 4962
			err = -EROFS;
			goto restore_opts;
		}

		if (*flags & MS_RDONLY) {
4963 4964 4965
			err = sync_filesystem(sb);
			if (err < 0)
				goto restore_opts;
4966 4967
			err = dquot_suspend(sb, -1);
			if (err < 0)
4968 4969
				goto restore_opts;

4970 4971 4972 4973 4974 4975 4976 4977 4978 4979 4980
			/*
			 * 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.
			 */
4981 4982
			if (!(es->s_state & cpu_to_le16(EXT4_VALID_FS)) &&
			    (sbi->s_mount_state & EXT4_VALID_FS))
4983 4984
				es->s_state = cpu_to_le16(sbi->s_mount_state);

4985
			if (sbi->s_journal)
4986
				ext4_mark_recovery_complete(sb, es);
4987
		} else {
4988
			/* Make sure we can mount this feature set readwrite */
4989
			if (ext4_has_feature_readonly(sb) ||
D
Darrick J. Wong 已提交
4990
			    !ext4_feature_set_ok(sb, 0)) {
4991 4992 4993
				err = -EROFS;
				goto restore_opts;
			}
4994 4995
			/*
			 * Make sure the group descriptor checksums
4996
			 * are sane.  If they aren't, refuse to remount r/w.
4997 4998 4999 5000 5001
			 */
			for (g = 0; g < sbi->s_groups_count; g++) {
				struct ext4_group_desc *gdp =
					ext4_get_group_desc(sb, g, NULL);

5002
				if (!ext4_group_desc_csum_verify(sb, g, gdp)) {
5003 5004
					ext4_msg(sb, KERN_ERR,
	       "ext4_remount: Checksum for group %u failed (%u!=%u)",
5005
		g, le16_to_cpu(ext4_group_desc_csum(sb, g, gdp)),
5006
					       le16_to_cpu(gdp->bg_checksum));
5007
					err = -EFSBADCRC;
5008 5009 5010 5011
					goto restore_opts;
				}
			}

5012 5013 5014 5015 5016 5017
			/*
			 * 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) {
5018
				ext4_msg(sb, KERN_WARNING, "Couldn't "
5019 5020
				       "remount RDWR because of unprocessed "
				       "orphan inode list.  Please "
5021
				       "umount/remount instead");
5022 5023 5024 5025
				err = -EINVAL;
				goto restore_opts;
			}

5026 5027 5028 5029 5030 5031
			/*
			 * 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.)
			 */
5032 5033
			if (sbi->s_journal)
				ext4_clear_journal_err(sb, es);
5034
			sbi->s_mount_state = le16_to_cpu(es->s_state);
5035
			if (!ext4_setup_super(sb, es, 0))
5036
				sb->s_flags &= ~MS_RDONLY;
5037
			if (ext4_has_feature_mmp(sb))
5038 5039 5040 5041 5042
				if (ext4_multi_mount_protect(sb,
						le64_to_cpu(es->s_mmp_block))) {
					err = -EROFS;
					goto restore_opts;
				}
5043
			enable_quota = 1;
5044 5045
		}
	}
5046 5047 5048 5049 5050 5051 5052 5053 5054 5055 5056 5057 5058

	/*
	 * 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);
	}

5059
	ext4_setup_system_zone(sb);
5060
	if (sbi->s_journal == NULL && !(old_sb_flags & MS_RDONLY))
5061
		ext4_commit_super(sb, 1);
5062

5063 5064
#ifdef CONFIG_QUOTA
	/* Release old quota file names */
J
Jan Kara 已提交
5065
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
5066
		kfree(old_opts.s_qf_names[i]);
5067 5068 5069
	if (enable_quota) {
		if (sb_any_quota_suspended(sb))
			dquot_resume(sb, -1);
5070
		else if (ext4_has_feature_quota(sb)) {
5071
			err = ext4_enable_quotas(sb);
5072
			if (err)
5073 5074 5075
				goto restore_opts;
		}
	}
5076
#endif
5077

5078
	*flags = (*flags & ~MS_LAZYTIME) | (sb->s_flags & MS_LAZYTIME);
5079 5080
	ext4_msg(sb, KERN_INFO, "re-mounted. Opts: %s", orig_data);
	kfree(orig_data);
5081
	return 0;
5082

5083 5084 5085
restore_opts:
	sb->s_flags = old_sb_flags;
	sbi->s_mount_opt = old_opts.s_mount_opt;
5086
	sbi->s_mount_opt2 = old_opts.s_mount_opt2;
5087 5088 5089
	sbi->s_resuid = old_opts.s_resuid;
	sbi->s_resgid = old_opts.s_resgid;
	sbi->s_commit_interval = old_opts.s_commit_interval;
5090 5091
	sbi->s_min_batch_time = old_opts.s_min_batch_time;
	sbi->s_max_batch_time = old_opts.s_max_batch_time;
5092 5093
#ifdef CONFIG_QUOTA
	sbi->s_jquota_fmt = old_opts.s_jquota_fmt;
J
Jan Kara 已提交
5094
	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
5095
		kfree(sbi->s_qf_names[i]);
5096 5097 5098
		sbi->s_qf_names[i] = old_opts.s_qf_names[i];
	}
#endif
5099
	kfree(orig_data);
5100 5101 5102
	return err;
}

L
Li Xi 已提交
5103 5104 5105 5106 5107 5108 5109 5110 5111 5112 5113 5114 5115 5116 5117 5118 5119 5120 5121 5122 5123 5124 5125 5126 5127 5128 5129 5130 5131 5132 5133 5134 5135 5136 5137 5138 5139 5140 5141 5142 5143 5144
#ifdef CONFIG_QUOTA
static int ext4_statfs_project(struct super_block *sb,
			       kprojid_t projid, struct kstatfs *buf)
{
	struct kqid qid;
	struct dquot *dquot;
	u64 limit;
	u64 curblock;

	qid = make_kqid_projid(projid);
	dquot = dqget(sb, qid);
	if (IS_ERR(dquot))
		return PTR_ERR(dquot);
	spin_lock(&dq_data_lock);

	limit = (dquot->dq_dqb.dqb_bsoftlimit ?
		 dquot->dq_dqb.dqb_bsoftlimit :
		 dquot->dq_dqb.dqb_bhardlimit) >> sb->s_blocksize_bits;
	if (limit && buf->f_blocks > limit) {
		curblock = dquot->dq_dqb.dqb_curspace >> sb->s_blocksize_bits;
		buf->f_blocks = limit;
		buf->f_bfree = buf->f_bavail =
			(buf->f_blocks > curblock) ?
			 (buf->f_blocks - curblock) : 0;
	}

	limit = dquot->dq_dqb.dqb_isoftlimit ?
		dquot->dq_dqb.dqb_isoftlimit :
		dquot->dq_dqb.dqb_ihardlimit;
	if (limit && buf->f_files > limit) {
		buf->f_files = limit;
		buf->f_ffree =
			(buf->f_files > dquot->dq_dqb.dqb_curinodes) ?
			 (buf->f_files - dquot->dq_dqb.dqb_curinodes) : 0;
	}

	spin_unlock(&dq_data_lock);
	dqput(dquot);
	return 0;
}
#endif

5145
static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf)
5146 5147
{
	struct super_block *sb = dentry->d_sb;
5148 5149
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_super_block *es = sbi->s_es;
L
Lukas Czerner 已提交
5150
	ext4_fsblk_t overhead = 0, resv_blocks;
P
Pekka Enberg 已提交
5151
	u64 fsid;
5152
	s64 bfree;
L
Lukas Czerner 已提交
5153
	resv_blocks = EXT4_C2B(sbi, atomic64_read(&sbi->s_resv_clusters));
5154

5155 5156
	if (!test_opt(sb, MINIX_DF))
		overhead = sbi->s_overhead;
5157

5158
	buf->f_type = EXT4_SUPER_MAGIC;
5159
	buf->f_bsize = sb->s_blocksize;
5160
	buf->f_blocks = ext4_blocks_count(es) - EXT4_C2B(sbi, overhead);
5161 5162
	bfree = percpu_counter_sum_positive(&sbi->s_freeclusters_counter) -
		percpu_counter_sum_positive(&sbi->s_dirtyclusters_counter);
5163
	/* prevent underflow in case that few free space is available */
5164
	buf->f_bfree = EXT4_C2B(sbi, max_t(s64, bfree, 0));
L
Lukas Czerner 已提交
5165 5166 5167
	buf->f_bavail = buf->f_bfree -
			(ext4_r_blocks_count(es) + resv_blocks);
	if (buf->f_bfree < (ext4_r_blocks_count(es) + resv_blocks))
5168 5169
		buf->f_bavail = 0;
	buf->f_files = le32_to_cpu(es->s_inodes_count);
P
Peter Zijlstra 已提交
5170
	buf->f_ffree = percpu_counter_sum_positive(&sbi->s_freeinodes_counter);
5171
	buf->f_namelen = EXT4_NAME_LEN;
P
Pekka Enberg 已提交
5172 5173 5174 5175
	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;
5176

L
Li Xi 已提交
5177 5178 5179 5180 5181
#ifdef CONFIG_QUOTA
	if (ext4_test_inode_flag(dentry->d_inode, EXT4_INODE_PROJINHERIT) &&
	    sb_has_quota_limits_enabled(sb, PRJQUOTA))
		ext4_statfs_project(sb, EXT4_I(dentry->d_inode)->i_projid, buf);
#endif
5182 5183 5184
	return 0;
}

5185 5186
/* Helper function for writing quotas on sync - we need to start transaction
 * before quota file is locked for write. Otherwise the are possible deadlocks:
5187
 * Process 1                         Process 2
5188
 * ext4_create()                     quota_sync()
5189
 *   jbd2_journal_start()                  write_dquot()
5190
 *   dquot_initialize()                         down(dqio_mutex)
5191
 *     down(dqio_mutex)                    jbd2_journal_start()
5192 5193 5194 5195 5196 5197 5198
 *
 */

#ifdef CONFIG_QUOTA

static inline struct inode *dquot_to_inode(struct dquot *dquot)
{
5199
	return sb_dqopt(dquot->dq_sb)->files[dquot->dq_id.type];
5200 5201
}

5202
static int ext4_write_dquot(struct dquot *dquot)
5203 5204 5205 5206 5207 5208
{
	int ret, err;
	handle_t *handle;
	struct inode *inode;

	inode = dquot_to_inode(dquot);
5209
	handle = ext4_journal_start(inode, EXT4_HT_QUOTA,
5210
				    EXT4_QUOTA_TRANS_BLOCKS(dquot->dq_sb));
5211 5212 5213
	if (IS_ERR(handle))
		return PTR_ERR(handle);
	ret = dquot_commit(dquot);
5214
	err = ext4_journal_stop(handle);
5215 5216 5217 5218 5219
	if (!ret)
		ret = err;
	return ret;
}

5220
static int ext4_acquire_dquot(struct dquot *dquot)
5221 5222 5223 5224
{
	int ret, err;
	handle_t *handle;

5225
	handle = ext4_journal_start(dquot_to_inode(dquot), EXT4_HT_QUOTA,
5226
				    EXT4_QUOTA_INIT_BLOCKS(dquot->dq_sb));
5227 5228 5229
	if (IS_ERR(handle))
		return PTR_ERR(handle);
	ret = dquot_acquire(dquot);
5230
	err = ext4_journal_stop(handle);
5231 5232 5233 5234 5235
	if (!ret)
		ret = err;
	return ret;
}

5236
static int ext4_release_dquot(struct dquot *dquot)
5237 5238 5239 5240
{
	int ret, err;
	handle_t *handle;

5241
	handle = ext4_journal_start(dquot_to_inode(dquot), EXT4_HT_QUOTA,
5242
				    EXT4_QUOTA_DEL_BLOCKS(dquot->dq_sb));
J
Jan Kara 已提交
5243 5244 5245
	if (IS_ERR(handle)) {
		/* Release dquot anyway to avoid endless cycle in dqput() */
		dquot_release(dquot);
5246
		return PTR_ERR(handle);
J
Jan Kara 已提交
5247
	}
5248
	ret = dquot_release(dquot);
5249
	err = ext4_journal_stop(handle);
5250 5251 5252 5253 5254
	if (!ret)
		ret = err;
	return ret;
}

5255
static int ext4_mark_dquot_dirty(struct dquot *dquot)
5256
{
5257 5258 5259
	struct super_block *sb = dquot->dq_sb;
	struct ext4_sb_info *sbi = EXT4_SB(sb);

5260
	/* Are we journaling quotas? */
5261
	if (ext4_has_feature_quota(sb) ||
5262
	    sbi->s_qf_names[USRQUOTA] || sbi->s_qf_names[GRPQUOTA]) {
5263
		dquot_mark_dquot_dirty(dquot);
5264
		return ext4_write_dquot(dquot);
5265 5266 5267 5268 5269
	} else {
		return dquot_mark_dquot_dirty(dquot);
	}
}

5270
static int ext4_write_info(struct super_block *sb, int type)
5271 5272 5273 5274 5275
{
	int ret, err;
	handle_t *handle;

	/* Data block + inode block */
5276
	handle = ext4_journal_start(d_inode(sb->s_root), EXT4_HT_QUOTA, 2);
5277 5278 5279
	if (IS_ERR(handle))
		return PTR_ERR(handle);
	ret = dquot_commit_info(sb, type);
5280
	err = ext4_journal_stop(handle);
5281 5282 5283 5284 5285 5286 5287 5288 5289
	if (!ret)
		ret = err;
	return ret;
}

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

5296 5297 5298 5299 5300 5301 5302 5303 5304 5305 5306 5307 5308 5309
static void lockdep_set_quota_inode(struct inode *inode, int subclass)
{
	struct ext4_inode_info *ei = EXT4_I(inode);

	/* The first argument of lockdep_set_subclass has to be
	 * *exactly* the same as the argument to init_rwsem() --- in
	 * this case, in init_once() --- or lockdep gets unhappy
	 * because the name of the lock is set using the
	 * stringification of the argument to init_rwsem().
	 */
	(void) ei;	/* shut up clang warning if !CONFIG_LOCKDEP */
	lockdep_set_subclass(&ei->i_data_sem, subclass);
}

5310 5311 5312
/*
 * Standard function to be called on quota_on
 */
5313
static int ext4_quota_on(struct super_block *sb, int type, int format_id,
A
Al Viro 已提交
5314
			 const struct path *path)
5315 5316 5317 5318 5319
{
	int err;

	if (!test_opt(sb, QUOTA))
		return -EINVAL;
5320

5321
	/* Quotafile not on the same filesystem? */
5322
	if (path->dentry->d_sb != sb)
5323
		return -EXDEV;
5324 5325
	/* Journaling quota? */
	if (EXT4_SB(sb)->s_qf_names[type]) {
5326
		/* Quotafile not in fs root? */
5327
		if (path->dentry->d_parent != sb->s_root)
5328 5329 5330
			ext4_msg(sb, KERN_WARNING,
				"Quota file not on filesystem root. "
				"Journaled quota will not work");
5331
	}
5332 5333 5334 5335 5336

	/*
	 * When we journal data on quota file, we have to flush journal to see
	 * all updates to the file when we bypass pagecache...
	 */
5337
	if (EXT4_SB(sb)->s_journal &&
5338
	    ext4_should_journal_data(d_inode(path->dentry))) {
5339 5340 5341 5342 5343
		/*
		 * We don't need to lock updates but journal_flush() could
		 * otherwise be livelocked...
		 */
		jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal);
5344
		err = jbd2_journal_flush(EXT4_SB(sb)->s_journal);
5345
		jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal);
5346
		if (err)
5347
			return err;
5348
	}
5349 5350 5351 5352 5353 5354
	lockdep_set_quota_inode(path->dentry->d_inode, I_DATA_SEM_QUOTA);
	err = dquot_quota_on(sb, type, format_id, path);
	if (err)
		lockdep_set_quota_inode(path->dentry->d_inode,
					     I_DATA_SEM_NORMAL);
	return err;
5355 5356
}

5357 5358 5359 5360 5361
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
5362
	unsigned long qf_inums[EXT4_MAXQUOTAS] = {
5363
		le32_to_cpu(EXT4_SB(sb)->s_es->s_usr_quota_inum),
L
Li Xi 已提交
5364 5365
		le32_to_cpu(EXT4_SB(sb)->s_es->s_grp_quota_inum),
		le32_to_cpu(EXT4_SB(sb)->s_es->s_prj_quota_inum)
5366 5367
	};

5368
	BUG_ON(!ext4_has_feature_quota(sb));
5369 5370 5371 5372 5373 5374 5375 5376 5377 5378

	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
5379 5380
	/* Don't account quota for quota files to avoid recursion */
	qf_inode->i_flags |= S_NOQUOTA;
5381
	lockdep_set_quota_inode(qf_inode, I_DATA_SEM_QUOTA);
5382 5383
	err = dquot_enable(qf_inode, type, format_id, flags);
	iput(qf_inode);
5384 5385
	if (err)
		lockdep_set_quota_inode(qf_inode, I_DATA_SEM_NORMAL);
5386 5387 5388 5389 5390 5391 5392 5393

	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
5394
	unsigned long qf_inums[EXT4_MAXQUOTAS] = {
5395
		le32_to_cpu(EXT4_SB(sb)->s_es->s_usr_quota_inum),
L
Li Xi 已提交
5396 5397
		le32_to_cpu(EXT4_SB(sb)->s_es->s_grp_quota_inum),
		le32_to_cpu(EXT4_SB(sb)->s_es->s_prj_quota_inum)
5398
	};
5399 5400 5401 5402 5403
	bool quota_mopt[EXT4_MAXQUOTAS] = {
		test_opt(sb, USRQUOTA),
		test_opt(sb, GRPQUOTA),
		test_opt(sb, PRJQUOTA),
	};
5404 5405

	sb_dqopt(sb)->flags |= DQUOT_QUOTA_SYS_FILE;
J
Jan Kara 已提交
5406
	for (type = 0; type < EXT4_MAXQUOTAS; type++) {
5407 5408
		if (qf_inums[type]) {
			err = ext4_quota_enable(sb, type, QFMT_VFS_V1,
5409 5410
				DQUOT_USAGE_ENABLED |
				(quota_mopt[type] ? DQUOT_LIMITS_ENABLED : 0));
5411 5412
			if (err) {
				ext4_warning(sb,
5413 5414 5415
					"Failed to enable quota tracking "
					"(type=%d, err=%d). Please run "
					"e2fsck to fix.", type, err);
5416 5417 5418 5419 5420 5421 5422
				return err;
			}
		}
	}
	return 0;
}

5423 5424
static int ext4_quota_off(struct super_block *sb, int type)
{
5425 5426 5427
	struct inode *inode = sb_dqopt(sb)->files[type];
	handle_t *handle;

5428 5429 5430
	/* Force all delayed allocation blocks to be allocated.
	 * Caller already holds s_umount sem */
	if (test_opt(sb, DELALLOC))
5431 5432
		sync_filesystem(sb);

5433 5434 5435
	if (!inode)
		goto out;

5436 5437
	/* Update modification times of quota files when userspace can
	 * start looking at them */
5438
	handle = ext4_journal_start(inode, EXT4_HT_QUOTA, 1);
5439 5440
	if (IS_ERR(handle))
		goto out;
5441
	inode->i_mtime = inode->i_ctime = current_time(inode);
5442 5443 5444 5445
	ext4_mark_inode_dirty(handle, inode);
	ext4_journal_stop(handle);

out:
5446 5447 5448
	return dquot_quota_off(sb, type);
}

5449 5450
/* 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
5451
 * itself serializes the operations (and no one else should touch the files)
5452
 * we don't have to be afraid of races */
5453
static ssize_t ext4_quota_read(struct super_block *sb, int type, char *data,
5454 5455 5456
			       size_t len, loff_t off)
{
	struct inode *inode = sb_dqopt(sb)->files[type];
A
Aneesh Kumar K.V 已提交
5457
	ext4_lblk_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
5458 5459 5460 5461 5462 5463 5464 5465 5466 5467 5468 5469 5470 5471
	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;
5472 5473 5474
		bh = ext4_bread(NULL, inode, blk, 0);
		if (IS_ERR(bh))
			return PTR_ERR(bh);
5475 5476 5477 5478 5479 5480 5481 5482 5483 5484 5485 5486 5487 5488 5489
		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) */
5490
static ssize_t ext4_quota_write(struct super_block *sb, int type,
5491 5492 5493
				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
5494
	ext4_lblk_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
5495
	int err, offset = off & (sb->s_blocksize - 1);
5496
	int retries = 0;
5497 5498 5499
	struct buffer_head *bh;
	handle_t *handle = journal_current_handle();

5500
	if (EXT4_SB(sb)->s_journal && !handle) {
5501 5502
		ext4_msg(sb, KERN_WARNING, "Quota write (off=%llu, len=%llu)"
			" cancelled because transaction is not started",
J
Jan Kara 已提交
5503 5504 5505
			(unsigned long long)off, (unsigned long long)len);
		return -EIO;
	}
5506 5507 5508 5509 5510 5511 5512 5513 5514 5515 5516
	/*
	 * 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;
	}

5517 5518 5519 5520 5521 5522
	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));
5523 5524
	if (IS_ERR(bh))
		return PTR_ERR(bh);
5525 5526
	if (!bh)
		goto out;
5527
	BUFFER_TRACE(bh, "get write access");
5528 5529 5530
	err = ext4_journal_get_write_access(handle, bh);
	if (err) {
		brelse(bh);
5531
		return err;
5532
	}
5533 5534 5535 5536
	lock_buffer(bh);
	memcpy(bh->b_data+offset, data, len);
	flush_dcache_page(bh->b_page);
	unlock_buffer(bh);
5537
	err = ext4_handle_dirty_metadata(handle, NULL, bh);
5538
	brelse(bh);
5539
out:
5540 5541
	if (inode->i_size < off + len) {
		i_size_write(inode, off + len);
5542
		EXT4_I(inode)->i_disksize = inode->i_size;
5543
		ext4_mark_inode_dirty(handle, inode);
5544
	}
5545
	return len;
5546 5547
}

5548 5549 5550 5551 5552 5553 5554 5555 5556 5557 5558
static int ext4_get_next_id(struct super_block *sb, struct kqid *qid)
{
	const struct quota_format_ops	*ops;

	if (!sb_has_quota_loaded(sb, qid->type))
		return -ESRCH;
	ops = sb_dqopt(sb)->ops[qid->type];
	if (!ops || !ops->get_next_id)
		return -ENOSYS;
	return dquot_get_next_id(sb, qid);
}
5559 5560
#endif

A
Al Viro 已提交
5561 5562
static struct dentry *ext4_mount(struct file_system_type *fs_type, int flags,
		       const char *dev_name, void *data)
5563
{
A
Al Viro 已提交
5564
	return mount_bdev(fs_type, flags, dev_name, data, ext4_fill_super);
5565 5566
}

J
Jan Kara 已提交
5567
#if !defined(CONFIG_EXT2_FS) && !defined(CONFIG_EXT2_FS_MODULE) && defined(CONFIG_EXT4_USE_FOR_EXT2)
5568 5569 5570 5571 5572 5573 5574 5575 5576 5577 5578 5579
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);
}
5580 5581 5582

static inline int ext2_feature_set_ok(struct super_block *sb)
{
5583
	if (ext4_has_unknown_ext2_incompat_features(sb))
5584 5585 5586
		return 0;
	if (sb->s_flags & MS_RDONLY)
		return 1;
5587
	if (ext4_has_unknown_ext2_ro_compat_features(sb))
5588 5589 5590
		return 0;
	return 1;
}
5591 5592 5593
#else
static inline void register_as_ext2(void) { }
static inline void unregister_as_ext2(void) { }
5594
static inline int ext2_feature_set_ok(struct super_block *sb) { return 0; }
5595 5596 5597 5598 5599 5600 5601 5602 5603 5604 5605 5606 5607 5608
#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);
}
5609 5610 5611

static inline int ext3_feature_set_ok(struct super_block *sb)
{
5612
	if (ext4_has_unknown_ext3_incompat_features(sb))
5613
		return 0;
5614
	if (!ext4_has_feature_journal(sb))
5615 5616 5617
		return 0;
	if (sb->s_flags & MS_RDONLY)
		return 1;
5618
	if (ext4_has_unknown_ext3_ro_compat_features(sb))
5619 5620 5621
		return 0;
	return 1;
}
5622

T
Theodore Ts'o 已提交
5623 5624 5625
static struct file_system_type ext4_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "ext4",
A
Al Viro 已提交
5626
	.mount		= ext4_mount,
T
Theodore Ts'o 已提交
5627 5628 5629
	.kill_sb	= kill_block_super,
	.fs_flags	= FS_REQUIRES_DEV,
};
5630
MODULE_ALIAS_FS("ext4");
T
Theodore Ts'o 已提交
5631

5632 5633 5634
/* Shared across all ext4 file systems */
wait_queue_head_t ext4__ioend_wq[EXT4_WQ_HASH_SZ];

5635
static int __init ext4_init_fs(void)
5636
{
5637
	int i, err;
5638

5639
	ratelimit_state_init(&ext4_mount_msg_ratelimit, 30 * HZ, 64);
5640 5641 5642
	ext4_li_info = NULL;
	mutex_init(&ext4_li_mtx);

5643
	/* Build-time check for flags consistency */
5644
	ext4_check_flag_values();
5645

5646
	for (i = 0; i < EXT4_WQ_HASH_SZ; i++)
5647 5648
		init_waitqueue_head(&ext4__ioend_wq[i]);

5649
	err = ext4_init_es();
5650 5651
	if (err)
		return err;
5652 5653 5654

	err = ext4_init_pageio();
	if (err)
5655
		goto out5;
5656

5657
	err = ext4_init_system_zone();
5658
	if (err)
5659
		goto out4;
5660

5661
	err = ext4_init_sysfs();
T
Theodore Ts'o 已提交
5662
	if (err)
5663
		goto out3;
5664

5665
	err = ext4_init_mballoc();
5666 5667
	if (err)
		goto out2;
5668 5669 5670
	err = init_inodecache();
	if (err)
		goto out1;
5671
	register_as_ext3();
5672
	register_as_ext2();
T
Theodore Ts'o 已提交
5673
	err = register_filesystem(&ext4_fs_type);
5674 5675
	if (err)
		goto out;
5676

5677 5678
	return 0;
out:
5679 5680
	unregister_as_ext2();
	unregister_as_ext3();
5681 5682
	destroy_inodecache();
out1:
5683
	ext4_exit_mballoc();
5684
out2:
5685 5686
	ext4_exit_sysfs();
out3:
5687
	ext4_exit_system_zone();
5688
out4:
5689
	ext4_exit_pageio();
5690
out5:
5691 5692
	ext4_exit_es();

5693 5694 5695
	return err;
}

5696
static void __exit ext4_exit_fs(void)
5697
{
5698
	ext4_destroy_lazyinit_thread();
5699 5700
	unregister_as_ext2();
	unregister_as_ext3();
T
Theodore Ts'o 已提交
5701
	unregister_filesystem(&ext4_fs_type);
5702
	destroy_inodecache();
5703
	ext4_exit_mballoc();
5704
	ext4_exit_sysfs();
5705 5706
	ext4_exit_system_zone();
	ext4_exit_pageio();
5707
	ext4_exit_es();
5708 5709 5710
}

MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others");
5711
MODULE_DESCRIPTION("Fourth Extended Filesystem");
5712
MODULE_LICENSE("GPL");
5713 5714
module_init(ext4_init_fs)
module_exit(ext4_exit_fs)