super.c 160.6 KB
Newer Older
1
/*
2
 *  linux/fs/ext4/super.c
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
 *
 * Copyright (C) 1992, 1993, 1994, 1995
 * Remy Card (card@masi.ibp.fr)
 * Laboratoire MASI - Institut Blaise Pascal
 * Universite Pierre et Marie Curie (Paris VI)
 *
 *  from
 *
 *  linux/fs/minix/inode.c
 *
 *  Copyright (C) 1991, 1992  Linus Torvalds
 *
 *  Big-endian to little-endian byte-swapping/bitmaps by
 *        David S. Miller (davem@caip.rutgers.edu), 1995
 */

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

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

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

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

56 57
static struct ext4_lazy_init *ext4_li_info;
static struct mutex ext4_li_mtx;
58
static 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;

441 442 443 444 445 446 447 448 449
	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);
	}
450
	save_error_info(sb, function, line);
451
	ext4_handle_error(sb);
452 453
}

454 455 456
void __ext4_error_inode(struct inode *inode, const char *function,
			unsigned int line, ext4_fsblk_t block,
			const char *fmt, ...)
457 458
{
	va_list args;
459
	struct va_format vaf;
460
	struct ext4_super_block *es = EXT4_SB(inode->i_sb)->s_es;
461

462 463
	es->s_last_error_ino = cpu_to_le32(inode->i_ino);
	es->s_last_error_block = cpu_to_le64(block);
464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479
	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);
	}
480
	save_error_info(inode->i_sb, function, line);
481 482 483
	ext4_handle_error(inode->i_sb);
}

484 485 486
void __ext4_error_file(struct file *file, const char *function,
		       unsigned int line, ext4_fsblk_t block,
		       const char *fmt, ...)
487 488
{
	va_list args;
489
	struct va_format vaf;
490
	struct ext4_super_block *es;
A
Al Viro 已提交
491
	struct inode *inode = file_inode(file);
492 493
	char pathname[80], *path;

494 495
	es = EXT4_SB(inode->i_sb)->s_es;
	es->s_last_error_ino = cpu_to_le32(inode->i_ino);
496
	if (ext4_error_ratelimit(inode->i_sb)) {
M
Miklos Szeredi 已提交
497
		path = file_path(file, pathname, sizeof(pathname));
498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516
		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);
	}
517
	save_error_info(inode->i_sb, function, line);
518 519 520
	ext4_handle_error(inode->i_sb);
}

521 522
const char *ext4_decode_error(struct super_block *sb, int errno,
			      char nbuf[16])
523 524 525 526
{
	char *errstr = NULL;

	switch (errno) {
527 528 529 530 531 532
	case -EFSCORRUPTED:
		errstr = "Corrupt filesystem";
		break;
	case -EFSBADCRC:
		errstr = "Filesystem failed CRC";
		break;
533 534 535 536 537 538 539
	case -EIO:
		errstr = "IO failure";
		break;
	case -ENOMEM:
		errstr = "Out of memory";
		break;
	case -EROFS:
540 541
		if (!sb || (EXT4_SB(sb)->s_journal &&
			    EXT4_SB(sb)->s_journal->j_flags & JBD2_ABORT))
542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560
			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;
}

561
/* __ext4_std_error decodes expected errors from journaling functions
562 563
 * automatically and invokes the appropriate error response.  */

564 565
void __ext4_std_error(struct super_block *sb, const char *function,
		      unsigned int line, int errno)
566 567 568 569 570 571 572 573 574 575 576
{
	char nbuf[16];
	const char *errstr;

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

577 578 579 580 581
	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);
	}
582

583
	save_error_info(sb, function, line);
584
	ext4_handle_error(sb);
585 586 587
}

/*
588
 * ext4_abort is a much stronger failure handler than ext4_error.  The
589 590 591 592 593 594 595 596
 * 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.
 */

597
void __ext4_abort(struct super_block *sb, const char *function,
598
		unsigned int line, const char *fmt, ...)
599
{
600
	struct va_format vaf;
601 602
	va_list args;

603
	save_error_info(sb, function, line);
604
	va_start(args, fmt);
605 606 607 608
	vaf.fmt = fmt;
	vaf.va = &args;
	printk(KERN_CRIT "EXT4-fs error (device %s): %s:%d: %pV\n",
	       sb->s_id, function, line, &vaf);
609 610
	va_end(args);

611 612 613
	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;
614 615 616 617 618 619
		/*
		 * Make sure updated value of ->s_mount_flags will be visible
		 * before ->s_flags update
		 */
		smp_wmb();
		sb->s_flags |= MS_RDONLY;
620 621 622 623
		if (EXT4_SB(sb)->s_journal)
			jbd2_journal_abort(EXT4_SB(sb)->s_journal, -EIO);
		save_error_info(sb, function, line);
	}
624 625 626 627
	if (test_opt(sb, ERRORS_PANIC)) {
		if (EXT4_SB(sb)->s_journal &&
		  !(EXT4_SB(sb)->s_journal->j_flags & JBD2_REC_ERR))
			return;
628
		panic("EXT4-fs panic from previous error\n");
629
	}
630 631
}

632 633
void __ext4_msg(struct super_block *sb,
		const char *prefix, const char *fmt, ...)
634
{
J
Joe Perches 已提交
635
	struct va_format vaf;
636 637
	va_list args;

638 639 640
	if (!___ratelimit(&(EXT4_SB(sb)->s_msg_ratelimit_state), "EXT4-fs"))
		return;

641
	va_start(args, fmt);
J
Joe Perches 已提交
642 643 644
	vaf.fmt = fmt;
	vaf.va = &args;
	printk("%sEXT4-fs (%s): %pV\n", prefix, sb->s_id, &vaf);
645 646 647
	va_end(args);
}

648 649 650 651
#define ext4_warning_ratelimit(sb)					\
		___ratelimit(&(EXT4_SB(sb)->s_warning_ratelimit_state),	\
			     "EXT4-fs warning")

652
void __ext4_warning(struct super_block *sb, const char *function,
653
		    unsigned int line, const char *fmt, ...)
654
{
J
Joe Perches 已提交
655
	struct va_format vaf;
656 657
	va_list args;

658
	if (!ext4_warning_ratelimit(sb))
659 660
		return;

661
	va_start(args, fmt);
J
Joe Perches 已提交
662 663 664 665
	vaf.fmt = fmt;
	vaf.va = &args;
	printk(KERN_WARNING "EXT4-fs warning (device %s): %s:%d: %pV\n",
	       sb->s_id, function, line, &vaf);
666 667 668
	va_end(args);
}

669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686
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);
}

687 688 689 690
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, ...)
691 692 693
__releases(bitlock)
__acquires(bitlock)
{
J
Joe Perches 已提交
694
	struct va_format vaf;
695 696 697
	va_list args;
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;

698 699 700
	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
701

702 703 704 705 706 707 708 709 710 711 712 713 714 715
	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);
	}
716 717

	if (test_opt(sb, ERRORS_CONT)) {
718
		ext4_commit_super(sb, 0);
719 720
		return;
	}
721

722 723 724 725 726 727 728 729
	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
730
	 * ext4_grp_locked_error() to distinguish between the
731 732 733 734 735 736 737 738
	 * 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;
}

739
void ext4_update_dynamic_rev(struct super_block *sb)
740
{
741
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;
742

743
	if (le32_to_cpu(es->s_rev_level) > EXT4_GOOD_OLD_REV)
744 745
		return;

746
	ext4_warning(sb,
747 748
		     "updating to rev %d because of new feature flag, "
		     "running e2fsck is recommended",
749
		     EXT4_DYNAMIC_REV);
750

751 752 753
	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);
754 755 756 757 758 759 760 761 762 763 764 765 766
	/* 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
 */
767
static struct block_device *ext4_blkdev_get(dev_t dev, struct super_block *sb)
768 769 770 771
{
	struct block_device *bdev;
	char b[BDEVNAME_SIZE];

772
	bdev = blkdev_get_by_dev(dev, FMODE_READ|FMODE_WRITE|FMODE_EXCL, sb);
773 774 775 776 777
	if (IS_ERR(bdev))
		goto fail;
	return bdev;

fail:
778
	ext4_msg(sb, KERN_ERR, "failed to open journal device %s: %ld",
779 780 781 782 783 784 785
			__bdevname(dev, b), PTR_ERR(bdev));
	return NULL;
}

/*
 * Release the journal device
 */
A
Al Viro 已提交
786
static void ext4_blkdev_put(struct block_device *bdev)
787
{
A
Al Viro 已提交
788
	blkdev_put(bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL);
789 790
}

A
Al Viro 已提交
791
static void ext4_blkdev_remove(struct ext4_sb_info *sbi)
792 793 794 795
{
	struct block_device *bdev;
	bdev = sbi->journal_bdev;
	if (bdev) {
A
Al Viro 已提交
796
		ext4_blkdev_put(bdev);
797 798 799 800 801 802
		sbi->journal_bdev = NULL;
	}
}

static inline struct inode *orphan_list_entry(struct list_head *l)
{
803
	return &list_entry(l, struct ext4_inode_info, i_orphan)->vfs_inode;
804 805
}

806
static void dump_orphan_list(struct super_block *sb, struct ext4_sb_info *sbi)
807 808 809
{
	struct list_head *l;

810 811
	ext4_msg(sb, KERN_ERR, "sb orphan head is %d",
		 le32_to_cpu(sbi->s_es->s_last_orphan));
812 813 814 815 816 817 818 819 820 821 822 823

	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));
	}
}

824
static void ext4_put_super(struct super_block *sb)
825
{
826 827
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_super_block *es = sbi->s_es;
828
	int i, err;
829

830
	ext4_unregister_li_request(sb);
831 832
	dquot_disable(sb, -1, DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED);

833 834
	flush_workqueue(sbi->rsv_conversion_wq);
	destroy_workqueue(sbi->rsv_conversion_wq);
835

836 837 838 839
	if (sbi->s_journal) {
		err = jbd2_journal_destroy(sbi->s_journal);
		sbi->s_journal = NULL;
		if (err < 0)
840
			ext4_abort(sb, "Couldn't clean up the journal");
841
	}
842

843
	ext4_unregister_sysfs(sb);
844
	ext4_es_unregister_shrinker(sbi);
845
	del_timer_sync(&sbi->s_err_report);
846 847 848 849
	ext4_release_system_zone(sb);
	ext4_mb_release(sb);
	ext4_ext_release(sb);

850
	if (!(sb->s_flags & MS_RDONLY)) {
851
		ext4_clear_feature_journal_needs_recovery(sb);
852 853
		es->s_state = cpu_to_le16(sbi->s_mount_state);
	}
854
	if (!(sb->s_flags & MS_RDONLY))
855 856
		ext4_commit_super(sb, 1);

857 858
	for (i = 0; i < sbi->s_gdb_count; i++)
		brelse(sbi->s_group_desc[i]);
A
Al Viro 已提交
859 860
	kvfree(sbi->s_group_desc);
	kvfree(sbi->s_flex_groups);
861
	percpu_counter_destroy(&sbi->s_freeclusters_counter);
862 863
	percpu_counter_destroy(&sbi->s_freeinodes_counter);
	percpu_counter_destroy(&sbi->s_dirs_counter);
864
	percpu_counter_destroy(&sbi->s_dirtyclusters_counter);
865
	percpu_free_rwsem(&sbi->s_journal_flag_rwsem);
866 867
	brelse(sbi->s_sbh);
#ifdef CONFIG_QUOTA
J
Jan Kara 已提交
868
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
869 870 871 872 873 874 875 876 877 878 879
		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));

880
	sync_blockdev(sb->s_bdev);
881
	invalidate_bdev(sb->s_bdev);
882 883 884 885 886 887 888
	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);
889
		invalidate_bdev(sbi->journal_bdev);
890
		ext4_blkdev_remove(sbi);
891
	}
892 893 894 895
	if (sbi->s_mb_cache) {
		ext4_xattr_destroy_cache(sbi->s_mb_cache);
		sbi->s_mb_cache = NULL;
	}
896 897
	if (sbi->s_mmp_tsk)
		kthread_stop(sbi->s_mmp_tsk);
898
	sb->s_fs_info = NULL;
T
Theodore Ts'o 已提交
899 900 901 902 903 904
	/*
	 * 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);
905 906
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
907
	kfree(sbi->s_blockgroup_lock);
908 909 910
	kfree(sbi);
}

911
static struct kmem_cache *ext4_inode_cachep;
912 913 914 915

/*
 * Called inside transaction, so use GFP_NOFS
 */
916
static struct inode *ext4_alloc_inode(struct super_block *sb)
917
{
918
	struct ext4_inode_info *ei;
919

C
Christoph Lameter 已提交
920
	ei = kmem_cache_alloc(ext4_inode_cachep, GFP_NOFS);
921 922
	if (!ei)
		return NULL;
923

924
	ei->vfs_inode.i_version = 1;
925
	spin_lock_init(&ei->i_raw_lock);
926 927
	INIT_LIST_HEAD(&ei->i_prealloc_list);
	spin_lock_init(&ei->i_prealloc_lock);
Z
Zheng Liu 已提交
928 929
	ext4_es_init_tree(&ei->i_es_tree);
	rwlock_init(&ei->i_es_lock);
930
	INIT_LIST_HEAD(&ei->i_es_list);
931
	ei->i_es_all_nr = 0;
932
	ei->i_es_shk_nr = 0;
933
	ei->i_es_shrink_lblk = 0;
934 935 936
	ei->i_reserved_data_blocks = 0;
	ei->i_reserved_meta_blocks = 0;
	ei->i_allocated_meta_blocks = 0;
937
	ei->i_da_metadata_calc_len = 0;
938
	ei->i_da_metadata_calc_last_lblock = 0;
939
	spin_lock_init(&(ei->i_block_reservation_lock));
940 941
#ifdef CONFIG_QUOTA
	ei->i_reserved_quota = 0;
J
Jan Kara 已提交
942
	memset(&ei->i_dquot, 0, sizeof(ei->i_dquot));
943
#endif
944
	ei->jinode = NULL;
945
	INIT_LIST_HEAD(&ei->i_rsv_conversion_list);
946
	spin_lock_init(&ei->i_completed_io_lock);
947 948
	ei->i_sync_tid = 0;
	ei->i_datasync_tid = 0;
949
	atomic_set(&ei->i_unwritten, 0);
950
	INIT_WORK(&ei->i_rsv_conversion_work, ext4_end_io_rsv_work);
951 952 953
	return &ei->vfs_inode;
}

954 955 956 957 958 959 960 961
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
962 963 964 965 966 967
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));
}

968
static void ext4_destroy_inode(struct inode *inode)
969
{
970
	if (!list_empty(&(EXT4_I(inode)->i_orphan))) {
971 972 973
		ext4_msg(inode->i_sb, KERN_ERR,
			 "Inode %lu (%p): orphan list check failed!",
			 inode->i_ino, EXT4_I(inode));
974 975 976 977 978
		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
979
	call_rcu(&inode->i_rcu, ext4_i_callback);
980 981
}

982
static void init_once(void *foo)
983
{
984
	struct ext4_inode_info *ei = (struct ext4_inode_info *) foo;
985

C
Christoph Lameter 已提交
986 987
	INIT_LIST_HEAD(&ei->i_orphan);
	init_rwsem(&ei->xattr_sem);
988
	init_rwsem(&ei->i_data_sem);
989
	init_rwsem(&ei->i_mmap_sem);
C
Christoph Lameter 已提交
990
	inode_init_once(&ei->vfs_inode);
991 992
}

993
static int __init init_inodecache(void)
994
{
995 996
	ext4_inode_cachep = kmem_cache_create("ext4_inode_cache",
					     sizeof(struct ext4_inode_info),
997
					     0, (SLAB_RECLAIM_ACCOUNT|
998
						SLAB_MEM_SPREAD|SLAB_ACCOUNT),
999
					     init_once);
1000
	if (ext4_inode_cachep == NULL)
1001 1002 1003 1004 1005 1006
		return -ENOMEM;
	return 0;
}

static void destroy_inodecache(void)
{
1007 1008 1009 1010 1011
	/*
	 * Make sure all delayed rcu free inodes are flushed before we
	 * destroy cache.
	 */
	rcu_barrier();
1012
	kmem_cache_destroy(ext4_inode_cachep);
1013 1014
}

A
Al Viro 已提交
1015
void ext4_clear_inode(struct inode *inode)
1016
{
A
Al Viro 已提交
1017
	invalidate_inode_buffers(inode);
1018
	clear_inode(inode);
1019
	dquot_drop(inode);
1020
	ext4_discard_preallocations(inode);
1021
	ext4_es_remove_extent(inode, 0, EXT_MAX_BLOCKS);
1022 1023 1024 1025 1026 1027
	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;
	}
1028
#ifdef CONFIG_EXT4_FS_ENCRYPTION
1029
	fscrypt_put_encryption_info(inode, NULL);
1030
#endif
1031 1032
}

C
Christoph Hellwig 已提交
1033
static struct inode *ext4_nfs_get_inode(struct super_block *sb,
1034
					u64 ino, u32 generation)
1035 1036 1037
{
	struct inode *inode;

1038
	if (ino < EXT4_FIRST_INO(sb) && ino != EXT4_ROOT_INO)
1039
		return ERR_PTR(-ESTALE);
1040
	if (ino > le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count))
1041 1042 1043 1044
		return ERR_PTR(-ESTALE);

	/* iget isn't really right if the inode is currently unallocated!!
	 *
1045
	 * ext4_read_inode will return a bad_inode if the inode had been
1046 1047 1048 1049 1050
	 * deleted, so we should be safe.
	 *
	 * Currently we don't know the generation for parent directory, so
	 * a generation of 0 means "accept any"
	 */
1051
	inode = ext4_iget_normal(sb, ino);
1052 1053 1054
	if (IS_ERR(inode))
		return ERR_CAST(inode);
	if (generation && inode->i_generation != generation) {
1055 1056 1057
		iput(inode);
		return ERR_PTR(-ESTALE);
	}
C
Christoph Hellwig 已提交
1058 1059 1060 1061 1062

	return inode;
}

static struct dentry *ext4_fh_to_dentry(struct super_block *sb, struct fid *fid,
1063
					int fh_len, int fh_type)
C
Christoph Hellwig 已提交
1064 1065 1066 1067 1068 1069
{
	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,
1070
					int fh_len, int fh_type)
C
Christoph Hellwig 已提交
1071 1072 1073
{
	return generic_fh_to_parent(sb, fid, fh_len, fh_type,
				    ext4_nfs_get_inode);
1074 1075
}

1076 1077 1078 1079 1080 1081
/*
 * 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.
 */
1082 1083
static int bdev_try_to_free_page(struct super_block *sb, struct page *page,
				 gfp_t wait)
1084 1085 1086 1087 1088 1089 1090 1091
{
	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,
1092
						wait & ~__GFP_DIRECT_RECLAIM);
1093 1094 1095
	return try_to_free_buffers(page);
}

1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116
#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_key_prefix(struct inode *inode, u8 **key)
{
	*key = EXT4_SB(inode->i_sb)->key_prefix;
	return EXT4_SB(inode->i_sb)->key_prefix_size;
}

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)
{
1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132
	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.
	 */

	if (handle) {
		res = ext4_xattr_set_handle(handle, inode,
					    EXT4_XATTR_INDEX_ENCRYPTION,
					    EXT4_XATTR_NAME_ENCRYPTION_CONTEXT,
					    ctx, len, 0);
1133 1134 1135 1136
		if (!res) {
			ext4_set_inode_flag(inode, EXT4_INODE_ENCRYPT);
			ext4_clear_inode_state(inode,
					EXT4_STATE_MAY_INLINE_DATA);
1137 1138 1139 1140
			/*
			 * Update inode->i_flags - e.g. S_DAX may get disabled
			 */
			ext4_set_inode_flags(inode);
1141 1142 1143 1144
		}
		return res;
	}

1145
retry:
1146 1147 1148 1149 1150
	handle = ext4_journal_start(inode, EXT4_HT_MISC,
			ext4_jbd2_credits_xattr(inode));
	if (IS_ERR(handle))
		return PTR_ERR(handle);

1151 1152 1153
	res = ext4_xattr_set_handle(handle, inode, EXT4_XATTR_INDEX_ENCRYPTION,
				    EXT4_XATTR_NAME_ENCRYPTION_CONTEXT,
				    ctx, len, 0);
1154 1155
	if (!res) {
		ext4_set_inode_flag(inode, EXT4_INODE_ENCRYPT);
1156 1157
		/* Update inode->i_flags - e.g. S_DAX may get disabled */
		ext4_set_inode_flags(inode);
1158 1159 1160 1161 1162
		res = ext4_mark_inode_dirty(handle, inode);
		if (res)
			EXT4_ERROR_INODE(inode, "Failed to mark inode dirty");
	}
	res2 = ext4_journal_stop(handle);
1163 1164 1165

	if (res == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
		goto retry;
1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197
	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;
}

static struct fscrypt_operations ext4_cryptops = {
	.get_context		= ext4_get_context,
	.key_prefix		= ext4_key_prefix,
	.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
static struct fscrypt_operations ext4_cryptops = {
	.is_encrypted		= ext4_encrypted_inode,
};
#endif

1198
#ifdef CONFIG_QUOTA
L
Li Xi 已提交
1199 1200
static char *quotatypes[] = INITQFNAMES;
#define QTYPE2NAME(t) (quotatypes[t])
1201

1202 1203 1204 1205 1206
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);
1207
static int ext4_quota_on(struct super_block *sb, int type, int format_id,
1208
			 struct path *path);
1209
static int ext4_quota_off(struct super_block *sb, int type);
1210 1211
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,
1212
			       size_t len, loff_t off);
1213
static ssize_t ext4_quota_write(struct super_block *sb, int type,
1214
				const char *data, size_t len, loff_t off);
1215 1216 1217
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);
1218
static int ext4_get_next_id(struct super_block *sb, struct kqid *qid);
1219

J
Jan Kara 已提交
1220 1221 1222 1223 1224
static struct dquot **ext4_get_dquots(struct inode *inode)
{
	return EXT4_I(inode)->i_dquot;
}

1225
static const struct dquot_operations ext4_quota_operations = {
1226
	.get_reserved_space = ext4_get_reserved_space,
1227 1228 1229 1230
	.write_dquot	= ext4_write_dquot,
	.acquire_dquot	= ext4_acquire_dquot,
	.release_dquot	= ext4_release_dquot,
	.mark_dirty	= ext4_mark_dquot_dirty,
1231 1232 1233
	.write_info	= ext4_write_info,
	.alloc_dquot	= dquot_alloc,
	.destroy_dquot	= dquot_destroy,
L
Li Xi 已提交
1234
	.get_projid	= ext4_get_projid,
1235
	.get_next_id	= ext4_get_next_id,
1236 1237
};

1238
static const struct quotactl_ops ext4_qctl_operations = {
1239
	.quota_on	= ext4_quota_on,
1240
	.quota_off	= ext4_quota_off,
1241
	.quota_sync	= dquot_quota_sync,
1242
	.get_state	= dquot_get_state,
1243 1244
	.set_info	= dquot_set_dqinfo,
	.get_dqblk	= dquot_get_dqblk,
1245 1246
	.set_dqblk	= dquot_set_dqblk,
	.get_nextdqblk	= dquot_get_next_dqblk,
1247 1248 1249
};
#endif

1250
static const struct super_operations ext4_sops = {
1251 1252 1253 1254
	.alloc_inode	= ext4_alloc_inode,
	.destroy_inode	= ext4_destroy_inode,
	.write_inode	= ext4_write_inode,
	.dirty_inode	= ext4_dirty_inode,
1255
	.drop_inode	= ext4_drop_inode,
A
Al Viro 已提交
1256
	.evict_inode	= ext4_evict_inode,
1257 1258
	.put_super	= ext4_put_super,
	.sync_fs	= ext4_sync_fs,
1259 1260
	.freeze_fs	= ext4_freeze,
	.unfreeze_fs	= ext4_unfreeze,
1261 1262 1263
	.statfs		= ext4_statfs,
	.remount_fs	= ext4_remount,
	.show_options	= ext4_show_options,
1264
#ifdef CONFIG_QUOTA
1265 1266
	.quota_read	= ext4_quota_read,
	.quota_write	= ext4_quota_write,
J
Jan Kara 已提交
1267
	.get_dquots	= ext4_get_dquots,
1268
#endif
1269
	.bdev_try_to_free_page = bdev_try_to_free_page,
1270 1271
};

1272
static const struct export_operations ext4_export_ops = {
C
Christoph Hellwig 已提交
1273 1274
	.fh_to_dentry = ext4_fh_to_dentry,
	.fh_to_parent = ext4_fh_to_parent,
1275
	.get_parent = ext4_get_parent,
1276 1277 1278 1279 1280
};

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,
1281
	Opt_nouid32, Opt_debug, Opt_removed,
1282
	Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl,
1283
	Opt_auto_da_alloc, Opt_noauto_da_alloc, Opt_noload,
1284 1285
	Opt_commit, Opt_min_batch_time, Opt_max_batch_time, Opt_journal_dev,
	Opt_journal_path, Opt_journal_checksum, Opt_journal_async_commit,
1286
	Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback,
1287
	Opt_data_err_abort, Opt_data_err_ignore, Opt_test_dummy_encryption,
1288
	Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota,
J
Jan Kara 已提交
1289
	Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_jqfmt_vfsv1, Opt_quota,
T
Theodore Ts'o 已提交
1290
	Opt_noquota, Opt_barrier, Opt_nobarrier, Opt_err,
1291
	Opt_usrquota, Opt_grpquota, Opt_prjquota, Opt_i_version, Opt_dax,
1292
	Opt_stripe, Opt_delalloc, Opt_nodelalloc, Opt_mblk_io_submit,
1293
	Opt_lazytime, Opt_nolazytime,
1294
	Opt_nomblk_io_submit, Opt_block_validity, Opt_noblock_validity,
1295
	Opt_inode_readahead_blks, Opt_journal_ioprio,
1296
	Opt_dioread_nolock, Opt_dioread_lock,
1297
	Opt_discard, Opt_nodiscard, Opt_init_itable, Opt_noinit_itable,
1298
	Opt_max_dir_size_kb, Opt_nojournal_checksum,
1299 1300
};

1301
static const match_table_t tokens = {
1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315
	{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"},
1316 1317
	{Opt_removed, "oldalloc"},
	{Opt_removed, "orlov"},
1318 1319 1320 1321
	{Opt_user_xattr, "user_xattr"},
	{Opt_nouser_xattr, "nouser_xattr"},
	{Opt_acl, "acl"},
	{Opt_noacl, "noacl"},
1322
	{Opt_noload, "norecovery"},
1323
	{Opt_noload, "noload"},
1324 1325
	{Opt_removed, "nobh"},
	{Opt_removed, "bh"},
1326
	{Opt_commit, "commit=%u"},
1327 1328
	{Opt_min_batch_time, "min_batch_time=%u"},
	{Opt_max_batch_time, "max_batch_time=%u"},
1329
	{Opt_journal_dev, "journal_dev=%u"},
1330
	{Opt_journal_path, "journal_path=%s"},
1331
	{Opt_journal_checksum, "journal_checksum"},
1332
	{Opt_nojournal_checksum, "nojournal_checksum"},
1333
	{Opt_journal_async_commit, "journal_async_commit"},
1334 1335 1336 1337
	{Opt_abort, "abort"},
	{Opt_data_journal, "data=journal"},
	{Opt_data_ordered, "data=ordered"},
	{Opt_data_writeback, "data=writeback"},
1338 1339
	{Opt_data_err_abort, "data_err=abort"},
	{Opt_data_err_ignore, "data_err=ignore"},
1340 1341 1342 1343 1344 1345
	{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
1346
	{Opt_jqfmt_vfsv1, "jqfmt=vfsv1"},
1347 1348 1349 1350
	{Opt_grpquota, "grpquota"},
	{Opt_noquota, "noquota"},
	{Opt_quota, "quota"},
	{Opt_usrquota, "usrquota"},
1351
	{Opt_prjquota, "prjquota"},
1352
	{Opt_barrier, "barrier=%u"},
T
Theodore Ts'o 已提交
1353 1354
	{Opt_barrier, "barrier"},
	{Opt_nobarrier, "nobarrier"},
1355
	{Opt_i_version, "i_version"},
R
Ross Zwisler 已提交
1356
	{Opt_dax, "dax"},
1357
	{Opt_stripe, "stripe=%u"},
1358
	{Opt_delalloc, "delalloc"},
1359 1360
	{Opt_lazytime, "lazytime"},
	{Opt_nolazytime, "nolazytime"},
1361
	{Opt_nodelalloc, "nodelalloc"},
1362 1363
	{Opt_removed, "mblk_io_submit"},
	{Opt_removed, "nomblk_io_submit"},
1364 1365
	{Opt_block_validity, "block_validity"},
	{Opt_noblock_validity, "noblock_validity"},
1366
	{Opt_inode_readahead_blks, "inode_readahead_blks=%u"},
1367
	{Opt_journal_ioprio, "journal_ioprio=%u"},
1368
	{Opt_auto_da_alloc, "auto_da_alloc=%u"},
T
Theodore Ts'o 已提交
1369 1370
	{Opt_auto_da_alloc, "auto_da_alloc"},
	{Opt_noauto_da_alloc, "noauto_da_alloc"},
1371 1372
	{Opt_dioread_nolock, "dioread_nolock"},
	{Opt_dioread_lock, "dioread_lock"},
1373 1374
	{Opt_discard, "discard"},
	{Opt_nodiscard, "nodiscard"},
1375 1376 1377
	{Opt_init_itable, "init_itable=%u"},
	{Opt_init_itable, "init_itable"},
	{Opt_noinit_itable, "noinit_itable"},
1378
	{Opt_max_dir_size_kb, "max_dir_size_kb=%u"},
1379
	{Opt_test_dummy_encryption, "test_dummy_encryption"},
1380 1381 1382 1383 1384
	{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
1385
	{Opt_err, NULL},
1386 1387
};

1388
static ext4_fsblk_t get_sb_block(void **data)
1389
{
1390
	ext4_fsblk_t	sb_block;
1391 1392 1393 1394
	char		*options = (char *) *data;

	if (!options || strncmp(options, "sb=", 3) != 0)
		return 1;	/* Default location */
1395

1396
	options += 3;
1397
	/* TODO: use simple_strtoll with >32bit ext4 */
1398 1399
	sb_block = simple_strtoul(options, &options, 0);
	if (*options && *options != ',') {
1400
		printk(KERN_ERR "EXT4-fs: Invalid sb specification: %s\n",
1401 1402 1403 1404 1405 1406
		       (char *) *data);
		return 1;
	}
	if (*options == ',')
		options++;
	*data = (void *) options;
1407

1408 1409 1410
	return sb_block;
}

1411
#define DEFAULT_JOURNAL_IOPRIO (IOPRIO_PRIO_VALUE(IOPRIO_CLASS_BE, 3))
1412 1413
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";
1414

D
Dmitry Monakhov 已提交
1415 1416 1417 1418 1419
#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;
1420
	int ret = -1;
D
Dmitry Monakhov 已提交
1421 1422 1423 1424 1425 1426

	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");
1427
		return -1;
D
Dmitry Monakhov 已提交
1428
	}
1429
	if (ext4_has_feature_quota(sb)) {
1430 1431 1432
		ext4_msg(sb, KERN_INFO, "Journaled quota options "
			 "ignored when QUOTA feature is enabled");
		return 1;
1433
	}
D
Dmitry Monakhov 已提交
1434 1435 1436 1437
	qname = match_strdup(args);
	if (!qname) {
		ext4_msg(sb, KERN_ERR,
			"Not enough memory for storing quotafile name");
1438
		return -1;
D
Dmitry Monakhov 已提交
1439
	}
1440 1441 1442 1443 1444 1445 1446 1447
	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
1448
	}
1449
	if (strchr(qname, '/')) {
D
Dmitry Monakhov 已提交
1450 1451
		ext4_msg(sb, KERN_ERR,
			"quotafile must be on filesystem root");
1452
		goto errout;
D
Dmitry Monakhov 已提交
1453
	}
1454
	sbi->s_qf_names[qtype] = qname;
1455
	set_opt(sb, QUOTA);
D
Dmitry Monakhov 已提交
1456
	return 1;
1457 1458 1459
errout:
	kfree(qname);
	return ret;
D
Dmitry Monakhov 已提交
1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470
}

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");
1471
		return -1;
D
Dmitry Monakhov 已提交
1472
	}
1473
	kfree(sbi->s_qf_names[qtype]);
D
Dmitry Monakhov 已提交
1474 1475 1476 1477 1478
	sbi->s_qf_names[qtype] = NULL;
	return 1;
}
#endif

1479 1480 1481 1482 1483 1484
#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
1485
#ifdef CONFIG_QUOTA
1486 1487 1488 1489 1490
#define MOPT_Q		0
#define MOPT_QFMT	0x0040
#else
#define MOPT_Q		MOPT_NOSUPPORT
#define MOPT_QFMT	MOPT_NOSUPPORT
1491
#endif
1492
#define MOPT_DATAJ	0x0080
1493 1494 1495
#define MOPT_NO_EXT2	0x0100
#define MOPT_NO_EXT3	0x0200
#define MOPT_EXT4_ONLY	(MOPT_NO_EXT2 | MOPT_NO_EXT3)
1496
#define MOPT_STRING	0x0400
1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508

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},
1509 1510 1511 1512
	{Opt_dioread_nolock, EXT4_MOUNT_DIOREAD_NOLOCK,
	 MOPT_EXT4_ONLY | MOPT_SET},
	{Opt_dioread_lock, EXT4_MOUNT_DIOREAD_NOLOCK,
	 MOPT_EXT4_ONLY | MOPT_CLEAR},
1513 1514
	{Opt_discard, EXT4_MOUNT_DISCARD, MOPT_SET},
	{Opt_nodiscard, EXT4_MOUNT_DISCARD, MOPT_CLEAR},
1515 1516 1517
	{Opt_delalloc, EXT4_MOUNT_DELALLOC,
	 MOPT_EXT4_ONLY | MOPT_SET | MOPT_EXPLICIT},
	{Opt_nodelalloc, EXT4_MOUNT_DELALLOC,
1518
	 MOPT_EXT4_ONLY | MOPT_CLEAR},
1519 1520
	{Opt_nojournal_checksum, EXT4_MOUNT_JOURNAL_CHECKSUM,
	 MOPT_EXT4_ONLY | MOPT_CLEAR},
1521
	{Opt_journal_checksum, EXT4_MOUNT_JOURNAL_CHECKSUM,
1522
	 MOPT_EXT4_ONLY | MOPT_SET | MOPT_EXPLICIT},
1523
	{Opt_journal_async_commit, (EXT4_MOUNT_JOURNAL_ASYNC_COMMIT |
1524
				    EXT4_MOUNT_JOURNAL_CHECKSUM),
1525
	 MOPT_EXT4_ONLY | MOPT_SET | MOPT_EXPLICIT},
1526
	{Opt_noload, EXT4_MOUNT_NOLOAD, MOPT_NO_EXT2 | MOPT_SET},
1527 1528 1529
	{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},
1530
	{Opt_data_err_abort, EXT4_MOUNT_DATA_ERR_ABORT,
1531
	 MOPT_NO_EXT2},
1532
	{Opt_data_err_ignore, EXT4_MOUNT_DATA_ERR_ABORT,
1533
	 MOPT_NO_EXT2},
1534 1535 1536 1537 1538 1539 1540 1541 1542 1543
	{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
1544
	{Opt_dax, EXT4_MOUNT_DAX, MOPT_SET},
1545
	{Opt_stripe, 0, MOPT_GTE0},
1546 1547
	{Opt_resuid, 0, MOPT_GTE0},
	{Opt_resgid, 0, MOPT_GTE0},
1548 1549 1550
	{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},
1551 1552 1553 1554
	{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},
1555 1556
	{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
1557
#ifdef CONFIG_EXT4_FS_POSIX_ACL
1558 1559
	{Opt_acl, EXT4_MOUNT_POSIX_ACL, MOPT_SET},
	{Opt_noacl, EXT4_MOUNT_POSIX_ACL, MOPT_CLEAR},
1560
#else
1561 1562
	{Opt_acl, 0, MOPT_NOSUPPORT},
	{Opt_noacl, 0, MOPT_NOSUPPORT},
1563
#endif
1564 1565 1566 1567 1568 1569 1570
	{Opt_nouid32, EXT4_MOUNT_NO_UID32, MOPT_SET},
	{Opt_debug, EXT4_MOUNT_DEBUG, MOPT_SET},
	{Opt_quota, EXT4_MOUNT_QUOTA | EXT4_MOUNT_USRQUOTA, MOPT_SET | MOPT_Q},
	{Opt_usrquota, EXT4_MOUNT_QUOTA | EXT4_MOUNT_USRQUOTA,
							MOPT_SET | MOPT_Q},
	{Opt_grpquota, EXT4_MOUNT_QUOTA | EXT4_MOUNT_GRPQUOTA,
							MOPT_SET | MOPT_Q},
1571 1572
	{Opt_prjquota, EXT4_MOUNT_QUOTA | EXT4_MOUNT_PRJQUOTA,
							MOPT_SET | MOPT_Q},
1573
	{Opt_noquota, (EXT4_MOUNT_QUOTA | EXT4_MOUNT_USRQUOTA |
1574 1575
		       EXT4_MOUNT_GRPQUOTA | EXT4_MOUNT_PRJQUOTA),
							MOPT_CLEAR | MOPT_Q},
1576 1577 1578 1579 1580 1581 1582
	{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},
1583
	{Opt_max_dir_size_kb, 0, MOPT_GTE0},
1584
	{Opt_test_dummy_encryption, 0, MOPT_GTE0},
1585 1586 1587 1588 1589 1590 1591 1592 1593
	{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;
1594 1595
	kuid_t uid;
	kgid_t gid;
1596 1597
	int arg = 0;

1598 1599 1600 1601 1602 1603 1604 1605 1606 1607
#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
1608
	switch (token) {
1609 1610 1611 1612
	case Opt_noacl:
	case Opt_nouser_xattr:
		ext4_msg(sb, KERN_WARNING, deprecated_msg, opt, "3.5");
		break;
1613 1614 1615
	case Opt_sb:
		return 1;	/* handled by get_sb_block() */
	case Opt_removed:
1616
		ext4_msg(sb, KERN_WARNING, "Ignoring removed %s option", opt);
1617 1618 1619 1620 1621 1622 1623
		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;
1624 1625 1626 1627 1628 1629
	case Opt_lazytime:
		sb->s_flags |= MS_LAZYTIME;
		return 1;
	case Opt_nolazytime:
		sb->s_flags &= ~MS_LAZYTIME;
		return 1;
1630 1631
	}

1632 1633 1634 1635 1636 1637 1638 1639 1640 1641
	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;
	}

1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652
	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;
	}

1653
	if (args->from && !(m->flags & MOPT_STRING) && match_int(args, &arg))
1654 1655 1656
		return -1;
	if (args->from && (m->flags & MOPT_GTE0) && (arg < 0))
		return -1;
1657 1658 1659
	if (m->flags & MOPT_EXPLICIT) {
		if (m->mount_opt & EXT4_MOUNT_DELALLOC) {
			set_opt2(sb, EXPLICIT_DELALLOC);
1660 1661
		} else if (m->mount_opt & EXT4_MOUNT_JOURNAL_CHECKSUM) {
			set_opt2(sb, EXPLICIT_JOURNAL_CHECKSUM);
1662 1663 1664
		} else
			return -1;
	}
1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683
	if (m->flags & MOPT_CLEAR_ERR)
		clear_opt(sb, ERRORS_MASK);
	if (token == Opt_noquota && sb_any_quota_loaded(sb)) {
		ext4_msg(sb, KERN_ERR, "Cannot change quota "
			 "options when quota turned on");
		return -1;
	}

	if (m->flags & MOPT_NOSUPPORT) {
		ext4_msg(sb, KERN_ERR, "%s option not supported", opt);
	} else if (token == Opt_commit) {
		if (arg == 0)
			arg = JBD2_DEFAULT_MAX_COMMIT_AGE;
		sbi->s_commit_interval = HZ * arg;
	} else if (token == Opt_max_batch_time) {
		sbi->s_max_batch_time = arg;
	} else if (token == Opt_min_batch_time) {
		sbi->s_min_batch_time = arg;
	} else if (token == Opt_inode_readahead_blks) {
1684 1685 1686 1687
		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");
1688
			return -1;
1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703
		}
		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);
1704 1705
			return -1;
		}
1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720
		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;
1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746
	} 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;
		}

1747
		journal_inode = d_inode(path.dentry);
1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758
		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);
1759 1760 1761 1762 1763 1764 1765 1766
	} 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);
1767 1768 1769 1770 1771 1772 1773 1774 1775
	} 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
1776 1777 1778 1779 1780
	} 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) {
1781
				ext4_msg(sb, KERN_ERR,
1782 1783
					 "Cannot change data mode on remount");
				return -1;
1784
			}
1785
		} else {
1786 1787
			clear_opt(sb, DATA_FLAGS);
			sbi->s_mount_opt |= m->mount_opt;
1788
		}
1789 1790 1791 1792 1793 1794 1795 1796
#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;
		}
1797
		if (ext4_has_feature_quota(sb)) {
1798 1799
			ext4_msg(sb, KERN_INFO,
				 "Quota format mount options ignored "
1800
				 "when QUOTA feature is enabled");
1801
			return 1;
1802
		}
1803
		sbi->s_jquota_fmt = m->mount_opt;
R
Ross Zwisler 已提交
1804 1805
#endif
	} else if (token == Opt_dax) {
1806 1807 1808 1809 1810
#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
1811 1812
		ext4_msg(sb, KERN_INFO, "dax option not supported");
		return -1;
1813
#endif
1814 1815 1816 1817
	} 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;
1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832
	} 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;
1833
	}
1834
	return 1;
1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856
}

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.
		 */
1857
		args[0].to = args[0].from = NULL;
1858 1859 1860 1861
		token = match_token(p, tokens, args);
		if (handle_mount_opt(sb, p, token, args, journal_devnum,
				     journal_ioprio, is_remount) < 0)
			return 0;
1862 1863
	}
#ifdef CONFIG_QUOTA
1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874
	/*
	 * 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]) {
1875
		if (test_opt(sb, USRQUOTA) && sbi->s_qf_names[USRQUOTA])
1876
			clear_opt(sb, USRQUOTA);
1877

1878
		if (test_opt(sb, GRPQUOTA) && sbi->s_qf_names[GRPQUOTA])
1879
			clear_opt(sb, GRPQUOTA);
1880

D
Dmitry Monakhov 已提交
1881
		if (test_opt(sb, GRPQUOTA) || test_opt(sb, USRQUOTA)) {
1882 1883
			ext4_msg(sb, KERN_ERR, "old and new quota "
					"format mixing");
1884 1885 1886 1887
			return 0;
		}

		if (!sbi->s_jquota_fmt) {
1888 1889
			ext4_msg(sb, KERN_ERR, "journaled quota format "
					"not specified");
1890 1891 1892 1893
			return 0;
		}
	}
#endif
J
Jan Kara 已提交
1894 1895 1896 1897
	if (test_opt(sb, DIOREAD_NOLOCK)) {
		int blocksize =
			BLOCK_SIZE << le32_to_cpu(sbi->s_es->s_log_block_size);

1898
		if (blocksize < PAGE_SIZE) {
J
Jan Kara 已提交
1899 1900 1901 1902 1903
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "dioread_nolock if block size != PAGE_SIZE");
			return 0;
		}
	}
1904 1905 1906 1907 1908 1909
	if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA &&
	    test_opt(sb, JOURNAL_ASYNC_COMMIT)) {
		ext4_msg(sb, KERN_ERR, "can't mount with journal_async_commit "
			 "in data=ordered mode");
		return 0;
	}
1910 1911 1912
	return 1;
}

1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936
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])
1937
		seq_show_option(seq, "usrjquota", sbi->s_qf_names[USRQUOTA]);
1938 1939

	if (sbi->s_qf_names[GRPQUOTA])
1940
		seq_show_option(seq, "grpjquota", sbi->s_qf_names[GRPQUOTA]);
1941 1942 1943
#endif
}

1944 1945
static const char *token2str(int token)
{
1946
	const struct match_token *t;
1947 1948 1949 1950 1951 1952 1953

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

1954 1955 1956 1957 1958
/*
 * Show an option if
 *  - it's set to a non-default value OR
 *  - if the per-sb default is different from the global default
 */
1959 1960
static int _ext4_show_options(struct seq_file *seq, struct super_block *sb,
			      int nodefs)
1961 1962 1963
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_super_block *es = sbi->s_es;
1964
	int def_errors, def_mount_opt = nodefs ? 0 : sbi->s_def_mount_opt;
1965
	const struct mount_opts *m;
1966
	char sep = nodefs ? '\n' : ',';
1967

1968 1969
#define SEQ_OPTS_PUTS(str) seq_printf(seq, "%c" str, sep)
#define SEQ_OPTS_PRINT(str, arg) seq_printf(seq, "%c" str, sep, arg)
1970 1971

	if (sbi->s_sb_block != 1)
1972 1973 1974 1975 1976 1977 1978
		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;
1979
		if (!(m->mount_opt & (sbi->s_mount_opt ^ def_mount_opt)))
1980 1981 1982 1983 1984 1985
			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));
1986
	}
1987

1988
	if (nodefs || !uid_eq(sbi->s_resuid, make_kuid(&init_user_ns, EXT4_DEF_RESUID)) ||
1989
	    le16_to_cpu(es->s_def_resuid) != EXT4_DEF_RESUID)
1990 1991 1992
		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)) ||
1993
	    le16_to_cpu(es->s_def_resgid) != EXT4_DEF_RESGID)
1994 1995
		SEQ_OPTS_PRINT("resgid=%u",
				from_kgid_munged(&init_user_ns, sbi->s_resgid));
1996
	def_errors = nodefs ? -1 : le16_to_cpu(es->s_errors);
1997 1998
	if (test_opt(sb, ERRORS_RO) && def_errors != EXT4_ERRORS_RO)
		SEQ_OPTS_PUTS("errors=remount-ro");
1999
	if (test_opt(sb, ERRORS_CONT) && def_errors != EXT4_ERRORS_CONTINUE)
2000
		SEQ_OPTS_PUTS("errors=continue");
2001
	if (test_opt(sb, ERRORS_PANIC) && def_errors != EXT4_ERRORS_PANIC)
2002
		SEQ_OPTS_PUTS("errors=panic");
2003
	if (nodefs || sbi->s_commit_interval != JBD2_DEFAULT_MAX_COMMIT_AGE*HZ)
2004
		SEQ_OPTS_PRINT("commit=%lu", sbi->s_commit_interval / HZ);
2005
	if (nodefs || sbi->s_min_batch_time != EXT4_DEF_MIN_BATCH_TIME)
2006
		SEQ_OPTS_PRINT("min_batch_time=%u", sbi->s_min_batch_time);
2007
	if (nodefs || sbi->s_max_batch_time != EXT4_DEF_MAX_BATCH_TIME)
2008
		SEQ_OPTS_PRINT("max_batch_time=%u", sbi->s_max_batch_time);
2009
	if (sb->s_flags & MS_I_VERSION)
2010
		SEQ_OPTS_PUTS("i_version");
2011
	if (nodefs || sbi->s_stripe)
2012
		SEQ_OPTS_PRINT("stripe=%lu", sbi->s_stripe);
2013
	if (EXT4_MOUNT_DATA_FLAGS & (sbi->s_mount_opt ^ def_mount_opt)) {
2014 2015 2016 2017 2018 2019 2020
		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");
	}
2021 2022
	if (nodefs ||
	    sbi->s_inode_readahead_blks != EXT4_DEF_INODE_READAHEAD_BLKS)
2023 2024
		SEQ_OPTS_PRINT("inode_readahead_blks=%u",
			       sbi->s_inode_readahead_blks);
2025

2026 2027
	if (nodefs || (test_opt(sb, INIT_INODE_TABLE) &&
		       (sbi->s_li_wait_mult != EXT4_DEF_LI_WAIT_MULT)))
2028
		SEQ_OPTS_PRINT("init_itable=%u", sbi->s_li_wait_mult);
2029 2030
	if (nodefs || sbi->s_max_dir_size_kb)
		SEQ_OPTS_PRINT("max_dir_size_kb=%u", sbi->s_max_dir_size_kb);
2031 2032
	if (test_opt(sb, DATA_ERR_ABORT))
		SEQ_OPTS_PUTS("data_err=abort");
2033 2034 2035 2036 2037

	ext4_show_quota_options(seq, sb);
	return 0;
}

2038 2039 2040 2041 2042
static int ext4_show_options(struct seq_file *seq, struct dentry *root)
{
	return _ext4_show_options(seq, root->d_sb, 0);
}

2043
int ext4_seq_options_show(struct seq_file *seq, void *offset)
2044 2045 2046 2047 2048 2049 2050 2051 2052 2053
{
	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;
}

2054
static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es,
2055 2056
			    int read_only)
{
2057
	struct ext4_sb_info *sbi = EXT4_SB(sb);
2058 2059
	int res = 0;

2060
	if (le32_to_cpu(es->s_rev_level) > EXT4_MAX_SUPP_REV) {
2061 2062
		ext4_msg(sb, KERN_ERR, "revision level too high, "
			 "forcing read-only mode");
2063 2064 2065
		res = MS_RDONLY;
	}
	if (read_only)
2066
		goto done;
2067
	if (!(sbi->s_mount_state & EXT4_VALID_FS))
2068 2069
		ext4_msg(sb, KERN_WARNING, "warning: mounting unchecked fs, "
			 "running e2fsck is recommended");
2070
	else if (sbi->s_mount_state & EXT4_ERROR_FS)
2071 2072 2073
		ext4_msg(sb, KERN_WARNING,
			 "warning: mounting fs with errors, "
			 "running e2fsck is recommended");
2074
	else if ((__s16) le16_to_cpu(es->s_max_mnt_count) > 0 &&
2075 2076
		 le16_to_cpu(es->s_mnt_count) >=
		 (unsigned short) (__s16) le16_to_cpu(es->s_max_mnt_count))
2077 2078 2079
		ext4_msg(sb, KERN_WARNING,
			 "warning: maximal mount count reached, "
			 "running e2fsck is recommended");
2080 2081 2082
	else if (le32_to_cpu(es->s_checkinterval) &&
		(le32_to_cpu(es->s_lastcheck) +
			le32_to_cpu(es->s_checkinterval) <= get_seconds()))
2083 2084 2085
		ext4_msg(sb, KERN_WARNING,
			 "warning: checktime reached, "
			 "running e2fsck is recommended");
2086
	if (!sbi->s_journal)
2087
		es->s_state &= cpu_to_le16(~EXT4_VALID_FS);
2088
	if (!(__s16) le16_to_cpu(es->s_max_mnt_count))
2089
		es->s_max_mnt_count = cpu_to_le16(EXT4_DFL_MAX_MNT_COUNT);
M
Marcin Slusarz 已提交
2090
	le16_add_cpu(&es->s_mnt_count, 1);
2091
	es->s_mtime = cpu_to_le32(get_seconds());
2092
	ext4_update_dynamic_rev(sb);
2093
	if (sbi->s_journal)
2094
		ext4_set_feature_journal_needs_recovery(sb);
2095

2096
	ext4_commit_super(sb, 1);
2097
done:
2098
	if (test_opt(sb, DEBUG))
2099
		printk(KERN_INFO "[EXT4 FS bs=%lu, gc=%u, "
2100
				"bpg=%lu, ipg=%lu, mo=%04x, mo2=%04x]\n",
2101 2102
			sb->s_blocksize,
			sbi->s_groups_count,
2103 2104
			EXT4_BLOCKS_PER_GROUP(sb),
			EXT4_INODES_PER_GROUP(sb),
2105
			sbi->s_mount_opt, sbi->s_mount_opt2);
2106

D
Dan Magenheimer 已提交
2107
	cleancache_init_fs(sb);
2108 2109 2110
	return res;
}

2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135
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
2136
		kvfree(sbi->s_flex_groups);
2137 2138 2139 2140 2141 2142
	}
	sbi->s_flex_groups = new_groups;
	sbi->s_flex_groups_allocated = size / sizeof(struct flex_groups);
	return 0;
}

2143 2144 2145 2146 2147
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;
2148
	int i, err;
2149

2150
	sbi->s_log_groups_per_flex = sbi->s_es->s_log_groups_per_flex;
2151
	if (sbi->s_log_groups_per_flex < 1 || sbi->s_log_groups_per_flex > 31) {
2152 2153 2154 2155
		sbi->s_log_groups_per_flex = 0;
		return 1;
	}

2156 2157
	err = ext4_alloc_flex_bg_array(sb, sbi->s_groups_count);
	if (err)
2158
		goto failed;
2159 2160

	for (i = 0; i < sbi->s_groups_count; i++) {
2161
		gdp = ext4_get_group_desc(sb, i, NULL);
2162 2163

		flex_group = ext4_flex_group(sbi, i);
2164 2165
		atomic_add(ext4_free_inodes_count(sb, gdp),
			   &sbi->s_flex_groups[flex_group].free_inodes);
2166 2167
		atomic64_add(ext4_free_group_clusters(sb, gdp),
			     &sbi->s_flex_groups[flex_group].free_clusters);
2168 2169
		atomic_add(ext4_used_dirs_count(sb, gdp),
			   &sbi->s_flex_groups[flex_group].used_dirs);
2170 2171 2172 2173 2174 2175 2176
	}

	return 1;
failed:
	return 0;
}

2177
static __le16 ext4_group_desc_csum(struct super_block *sb, __u32 block_group,
2178
				   struct ext4_group_desc *gdp)
A
Andreas Dilger 已提交
2179
{
2180
	int offset = offsetof(struct ext4_group_desc, bg_checksum);
A
Andreas Dilger 已提交
2181
	__u16 crc = 0;
2182
	__le32 le_group = cpu_to_le32(block_group);
2183
	struct ext4_sb_info *sbi = EXT4_SB(sb);
A
Andreas Dilger 已提交
2184

2185
	if (ext4_has_metadata_csum(sbi->s_sb)) {
2186 2187
		/* Use new metadata_csum algorithm */
		__u32 csum32;
2188
		__u16 dummy_csum = 0;
2189 2190 2191

		csum32 = ext4_chksum(sbi, sbi->s_csum_seed, (__u8 *)&le_group,
				     sizeof(le_group));
2192 2193 2194 2195 2196 2197 2198
		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);
2199 2200 2201

		crc = csum32 & 0xFFFF;
		goto out;
A
Andreas Dilger 已提交
2202 2203
	}

2204
	/* old crc16 code */
2205
	if (!ext4_has_feature_gdt_csum(sb))
2206 2207
		return 0;

2208 2209 2210 2211 2212
	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...*/
2213
	if (ext4_has_feature_64bit(sb) &&
2214 2215 2216 2217 2218 2219
	    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
2220 2221 2222
	return cpu_to_le16(crc);
}

2223
int ext4_group_desc_csum_verify(struct super_block *sb, __u32 block_group,
A
Andreas Dilger 已提交
2224 2225
				struct ext4_group_desc *gdp)
{
2226
	if (ext4_has_group_desc_csum(sb) &&
2227
	    (gdp->bg_checksum != ext4_group_desc_csum(sb, block_group, gdp)))
A
Andreas Dilger 已提交
2228 2229 2230 2231 2232
		return 0;

	return 1;
}

2233 2234 2235 2236 2237
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;
2238
	gdp->bg_checksum = ext4_group_desc_csum(sb, block_group, gdp);
2239 2240
}

2241
/* Called at mount-time, super-block is locked */
2242
static int ext4_check_descriptors(struct super_block *sb,
2243
				  ext4_fsblk_t sb_block,
2244
				  ext4_group_t *first_not_zeroed)
2245
{
2246 2247 2248
	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
2249 2250 2251
	ext4_fsblk_t block_bitmap;
	ext4_fsblk_t inode_bitmap;
	ext4_fsblk_t inode_table;
J
Jose R. Santos 已提交
2252
	int flexbg_flag = 0;
2253
	ext4_group_t i, grp = sbi->s_groups_count;
2254

2255
	if (ext4_has_feature_flex_bg(sb))
J
Jose R. Santos 已提交
2256 2257
		flexbg_flag = 1;

2258
	ext4_debug("Checking group descriptors");
2259

2260 2261 2262
	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
2263
		if (i == sbi->s_groups_count - 1 || flexbg_flag)
L
Laurent Vivier 已提交
2264
			last_block = ext4_blocks_count(sbi->s_es) - 1;
2265 2266
		else
			last_block = first_block +
2267
				(EXT4_BLOCKS_PER_GROUP(sb) - 1);
2268

2269 2270 2271 2272
		if ((grp == sbi->s_groups_count) &&
		   !(gdp->bg_flags & cpu_to_le16(EXT4_BG_INODE_ZEROED)))
			grp = i;

2273
		block_bitmap = ext4_block_bitmap(sb, gdp);
2274 2275 2276 2277 2278
		if (block_bitmap == sb_block) {
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
				 "Block bitmap for group %u overlaps "
				 "superblock", i);
		}
2279
		if (block_bitmap < first_block || block_bitmap > last_block) {
2280
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
2281
			       "Block bitmap for group %u not in group "
2282
			       "(block %llu)!", i, block_bitmap);
2283 2284
			return 0;
		}
2285
		inode_bitmap = ext4_inode_bitmap(sb, gdp);
2286 2287 2288 2289 2290
		if (inode_bitmap == sb_block) {
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
				 "Inode bitmap for group %u overlaps "
				 "superblock", i);
		}
2291
		if (inode_bitmap < first_block || inode_bitmap > last_block) {
2292
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
2293
			       "Inode bitmap for group %u not in group "
2294
			       "(block %llu)!", i, inode_bitmap);
2295 2296
			return 0;
		}
2297
		inode_table = ext4_inode_table(sb, gdp);
2298 2299 2300 2301 2302
		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
2303
		if (inode_table < first_block ||
2304
		    inode_table + sbi->s_itb_per_group - 1 > last_block) {
2305
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
2306
			       "Inode table for group %u not in group "
2307
			       "(block %llu)!", i, inode_table);
2308 2309
			return 0;
		}
2310
		ext4_lock_group(sb, i);
2311
		if (!ext4_group_desc_csum_verify(sb, i, gdp)) {
2312 2313
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
				 "Checksum for group %u failed (%u!=%u)",
2314
				 i, le16_to_cpu(ext4_group_desc_csum(sb, i,
2315
				     gdp)), le16_to_cpu(gdp->bg_checksum));
2316
			if (!(sb->s_flags & MS_RDONLY)) {
2317
				ext4_unlock_group(sb, i);
2318
				return 0;
2319
			}
A
Andreas Dilger 已提交
2320
		}
2321
		ext4_unlock_group(sb, i);
J
Jose R. Santos 已提交
2322 2323
		if (!flexbg_flag)
			first_block += EXT4_BLOCKS_PER_GROUP(sb);
2324
	}
2325 2326
	if (NULL != first_not_zeroed)
		*first_not_zeroed = grp;
2327 2328 2329
	return 1;
}

2330
/* ext4_orphan_cleanup() walks a singly-linked list of inodes (starting at
2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342
 * 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
2343
 * ext4_free_inode().  The only reason we would point at a wrong inode is if
2344 2345 2346
 * 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.
 */
2347 2348
static void ext4_orphan_cleanup(struct super_block *sb,
				struct ext4_super_block *es)
2349 2350
{
	unsigned int s_flags = sb->s_flags;
2351
	int ret, nr_orphans = 0, nr_truncates = 0;
2352 2353 2354 2355 2356 2357 2358 2359
#ifdef CONFIG_QUOTA
	int i;
#endif
	if (!es->s_last_orphan) {
		jbd_debug(4, "no orphan inodes to clean up\n");
		return;
	}

2360
	if (bdev_read_only(sb->s_bdev)) {
2361 2362
		ext4_msg(sb, KERN_ERR, "write access "
			"unavailable, skipping orphan cleanup");
2363 2364 2365
		return;
	}

2366 2367 2368 2369 2370 2371 2372
	/* 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;
	}

2373
	if (EXT4_SB(sb)->s_mount_state & EXT4_ERROR_FS) {
2374 2375
		/* don't clear list on RO mount w/ errors */
		if (es->s_last_orphan && !(s_flags & MS_RDONLY)) {
2376
			ext4_msg(sb, KERN_INFO, "Errors on filesystem, "
2377
				  "clearing orphan list.\n");
2378 2379
			es->s_last_orphan = 0;
		}
2380 2381 2382 2383 2384
		jbd_debug(1, "Skipping orphan recovery on fs with errors.\n");
		return;
	}

	if (s_flags & MS_RDONLY) {
2385
		ext4_msg(sb, KERN_INFO, "orphan cleanup on readonly fs");
2386 2387 2388 2389 2390 2391
		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
2392
	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
2393 2394
		if (EXT4_SB(sb)->s_qf_names[i]) {
			int ret = ext4_quota_on_mount(sb, i);
2395
			if (ret < 0)
2396 2397 2398
				ext4_msg(sb, KERN_ERR,
					"Cannot turn on journaled "
					"quota: error %d", ret);
2399 2400 2401 2402 2403 2404 2405
		}
	}
#endif

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

2406 2407 2408 2409 2410 2411 2412 2413 2414 2415
		/*
		 * 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;
		}

2416 2417
		inode = ext4_orphan_get(sb, le32_to_cpu(es->s_last_orphan));
		if (IS_ERR(inode)) {
2418 2419 2420 2421
			es->s_last_orphan = 0;
			break;
		}

2422
		list_add(&EXT4_I(inode)->i_orphan, &EXT4_SB(sb)->s_orphan);
2423
		dquot_initialize(inode);
2424
		if (inode->i_nlink) {
2425 2426 2427 2428
			if (test_opt(sb, DEBUG))
				ext4_msg(sb, KERN_DEBUG,
					"%s: truncating inode %lu to %lld bytes",
					__func__, inode->i_ino, inode->i_size);
2429
			jbd_debug(2, "truncating inode %lu to %lld bytes\n",
2430
				  inode->i_ino, inode->i_size);
A
Al Viro 已提交
2431
			inode_lock(inode);
2432
			truncate_inode_pages(inode->i_mapping, inode->i_size);
2433 2434 2435
			ret = ext4_truncate(inode);
			if (ret)
				ext4_std_error(inode->i_sb, ret);
A
Al Viro 已提交
2436
			inode_unlock(inode);
2437 2438
			nr_truncates++;
		} else {
2439 2440 2441 2442
			if (test_opt(sb, DEBUG))
				ext4_msg(sb, KERN_DEBUG,
					"%s: deleting unreferenced inode %lu",
					__func__, inode->i_ino);
2443 2444 2445 2446 2447 2448 2449
			jbd_debug(2, "deleting unreferenced inode %lu\n",
				  inode->i_ino);
			nr_orphans++;
		}
		iput(inode);  /* The delete magic happens here! */
	}

2450
#define PLURAL(x) (x), ((x) == 1) ? "" : "s"
2451 2452

	if (nr_orphans)
2453 2454
		ext4_msg(sb, KERN_INFO, "%d orphan inode%s deleted",
		       PLURAL(nr_orphans));
2455
	if (nr_truncates)
2456 2457
		ext4_msg(sb, KERN_INFO, "%d truncate%s cleaned up",
		       PLURAL(nr_truncates));
2458 2459
#ifdef CONFIG_QUOTA
	/* Turn quotas off */
J
Jan Kara 已提交
2460
	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
2461
		if (sb_dqopt(sb)->files[i])
2462
			dquot_quota_off(sb, i);
2463 2464 2465 2466
	}
#endif
	sb->s_flags = s_flags; /* Restore MS_RDONLY status */
}
2467

2468 2469 2470 2471 2472 2473 2474
/*
 * 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.
 *
2475 2476 2477 2478 2479 2480
 * 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.
 *
2481 2482
 * Note, this does *not* consider any metadata overhead for vfs i_blocks.
 */
2483
static loff_t ext4_max_size(int blkbits, int has_huge_files)
2484 2485 2486 2487 2488
{
	loff_t res;
	loff_t upper_limit = MAX_LFS_FILESIZE;

	/* small i_blocks in vfs inode? */
2489
	if (!has_huge_files || sizeof(blkcnt_t) < sizeof(u64)) {
2490
		/*
2491
		 * CONFIG_LBDAF is not enabled implies the inode
2492 2493 2494 2495 2496 2497 2498 2499 2500 2501
		 * 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;
	}

2502 2503 2504 2505 2506 2507
	/*
	 * 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;
2508 2509 2510 2511 2512 2513 2514 2515
	res <<= blkbits;

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

	return res;
}
2516 2517

/*
2518
 * Maximal bitmap file size.  There is a direct, and {,double-,triple-}indirect
2519 2520
 * 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.
2521
 */
2522
static loff_t ext4_max_bitmap_size(int bits, int has_huge_files)
2523
{
2524
	loff_t res = EXT4_NDIR_BLOCKS;
2525 2526
	int meta_blocks;
	loff_t upper_limit;
2527 2528 2529 2530 2531 2532
	/* 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.
2533 2534
	 */

2535
	if (!has_huge_files || sizeof(blkcnt_t) < sizeof(u64)) {
2536
		/*
2537
		 * !has_huge_files or CONFIG_LBDAF not enabled implies that
2538 2539
		 * the inode i_block field represents total file blocks in
		 * 2^32 512-byte sectors == size of vfs inode i_blocks * 8
2540 2541 2542 2543 2544 2545 2546
		 */
		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
2547 2548 2549 2550 2551 2552
		/*
		 * 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
		 */
2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565
		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;
2566 2567 2568 2569 2570 2571 2572

	res += 1LL << (bits-2);
	res += 1LL << (2*(bits-2));
	res += 1LL << (3*(bits-2));
	res <<= bits;
	if (res > upper_limit)
		res = upper_limit;
2573 2574 2575 2576

	if (res > MAX_LFS_FILESIZE)
		res = MAX_LFS_FILESIZE;

2577 2578 2579
	return res;
}

2580
static ext4_fsblk_t descriptor_loc(struct super_block *sb,
2581
				   ext4_fsblk_t logical_sb_block, int nr)
2582
{
2583
	struct ext4_sb_info *sbi = EXT4_SB(sb);
2584
	ext4_group_t bg, first_meta_bg;
2585 2586 2587 2588
	int has_super = 0;

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

2589
	if (!ext4_has_feature_meta_bg(sb) || nr < first_meta_bg)
2590
		return logical_sb_block + nr + 1;
2591
	bg = sbi->s_desc_per_block * nr;
2592
	if (ext4_bg_has_super(sb, bg))
2593
		has_super = 1;
2594

2595 2596 2597 2598 2599 2600 2601 2602 2603 2604
	/*
	 * 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++;

2605
	return (has_super + ext4_group_first_block_no(sb, bg));
2606 2607
}

2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623
/**
 * 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
2624
	int ret;
2625 2626

	if (sbi->s_stripe && sbi->s_stripe <= sbi->s_blocks_per_group)
D
Dan Ehrenberg 已提交
2627 2628 2629 2630 2631 2632 2633
		ret = sbi->s_stripe;
	else if (stripe_width <= sbi->s_blocks_per_group)
		ret = stripe_width;
	else if (stride <= sbi->s_blocks_per_group)
		ret = stride;
	else
		ret = 0;
2634

D
Dan Ehrenberg 已提交
2635 2636 2637 2638 2639 2640
	/*
	 * 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;
2641

D
Dan Ehrenberg 已提交
2642
	return ret;
2643
}
2644

2645 2646 2647 2648 2649 2650 2651 2652
/*
 * 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)
{
2653
	if (ext4_has_unknown_ext4_incompat_features(sb)) {
2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664
		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;

2665
	if (ext4_has_feature_readonly(sb)) {
D
Darrick J. Wong 已提交
2666 2667 2668 2669 2670
		ext4_msg(sb, KERN_INFO, "filesystem is read-only");
		sb->s_flags |= MS_RDONLY;
		return 1;
	}

2671
	/* Check that feature set is OK for a read-write mount */
2672
	if (ext4_has_unknown_ext4_ro_compat_features(sb)) {
2673 2674 2675 2676 2677 2678 2679 2680 2681 2682
		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
	 */
2683
	if (ext4_has_feature_huge_file(sb)) {
2684 2685 2686 2687 2688 2689 2690
		if (sizeof(blkcnt_t) < sizeof(u64)) {
			ext4_msg(sb, KERN_ERR, "Filesystem with huge files "
				 "cannot be mounted RDWR without "
				 "CONFIG_LBDAF");
			return 0;
		}
	}
2691
	if (ext4_has_feature_bigalloc(sb) && !ext4_has_feature_extents(sb)) {
2692 2693 2694 2695 2696
		ext4_msg(sb, KERN_ERR,
			 "Can't support bigalloc feature without "
			 "extents feature\n");
		return 0;
	}
2697 2698

#ifndef CONFIG_QUOTA
2699
	if (ext4_has_feature_quota(sb) && !readonly) {
2700 2701 2702 2703 2704
		ext4_msg(sb, KERN_ERR,
			 "Filesystem with quota feature cannot be mounted RDWR "
			 "without CONFIG_QUOTA");
		return 0;
	}
L
Li Xi 已提交
2705 2706 2707 2708 2709 2710
	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;
	}
2711
#endif  /* CONFIG_QUOTA */
2712 2713 2714
	return 1;
}

2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728
/*
 * 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)
2729 2730
		/* fsck newer than v1.41.13 is needed to clean this condition. */
		ext4_msg(sb, KERN_NOTICE, "error count since last fsck: %u",
2731 2732
			 le32_to_cpu(es->s_error_count));
	if (es->s_first_error_time) {
2733
		printk(KERN_NOTICE "EXT4-fs (%s): initial error at time %u: %.*s:%d",
2734 2735 2736 2737 2738
		       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)
2739
			printk(KERN_CONT ": inode %u",
2740 2741
			       le32_to_cpu(es->s_first_error_ino));
		if (es->s_first_error_block)
2742
			printk(KERN_CONT ": block %llu", (unsigned long long)
2743
			       le64_to_cpu(es->s_first_error_block));
2744
		printk(KERN_CONT "\n");
2745 2746
	}
	if (es->s_last_error_time) {
2747
		printk(KERN_NOTICE "EXT4-fs (%s): last error at time %u: %.*s:%d",
2748 2749 2750 2751 2752
		       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)
2753
			printk(KERN_CONT ": inode %u",
2754 2755
			       le32_to_cpu(es->s_last_error_ino));
		if (es->s_last_error_block)
2756
			printk(KERN_CONT ": block %llu", (unsigned long long)
2757
			       le64_to_cpu(es->s_last_error_block));
2758
		printk(KERN_CONT "\n");
2759 2760 2761 2762
	}
	mod_timer(&sbi->s_err_report, jiffies + 24*60*60*HZ);  /* Once a day */
}

2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785
/* 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;
	}

2786
	if (group >= ngroups)
2787 2788 2789 2790 2791 2792 2793
		ret = 1;

	if (!ret) {
		timeout = jiffies;
		ret = ext4_init_inode_table(sb, group,
					    elr->lr_timeout ? 0 : 1);
		if (elr->lr_timeout == 0) {
2794 2795
			timeout = (jiffies - timeout) *
				  elr->lr_sbi->s_li_wait_mult;
2796 2797 2798 2799 2800 2801 2802 2803 2804 2805
			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
2806
 * request structure. Should be called with li_list_mtx held
2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823
 */
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)
{
2824 2825 2826
	mutex_lock(&ext4_li_mtx);
	if (!ext4_li_info) {
		mutex_unlock(&ext4_li_mtx);
2827
		return;
2828
	}
2829 2830

	mutex_lock(&ext4_li_info->li_list_mtx);
2831
	ext4_remove_li_request(EXT4_SB(sb)->s_li_request);
2832
	mutex_unlock(&ext4_li_info->li_list_mtx);
2833
	mutex_unlock(&ext4_li_mtx);
2834 2835
}

2836 2837
static struct task_struct *ext4_lazyinit_task;

2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851
/*
 * 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;
2852
	unsigned long next_wakeup, cur;
2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865

	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) {
2866 2867
			int err = 0;
			int progress = 0;
2868 2869 2870
			elr = list_entry(pos, struct ext4_li_request,
					 lr_request);

2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888
			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;
2889
				}
2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900
				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));
2901 2902 2903 2904 2905 2906
			}
			if (time_before(elr->lr_next_sched, next_wakeup))
				next_wakeup = elr->lr_next_sched;
		}
		mutex_unlock(&eli->li_list_mtx);

2907
		try_to_freeze();
2908

2909 2910
		cur = jiffies;
		if ((time_after_eq(cur, next_wakeup)) ||
2911
		    (MAX_JIFFY_OFFSET == next_wakeup)) {
2912 2913 2914 2915
			cond_resched();
			continue;
		}

2916 2917
		schedule_timeout_interruptible(next_wakeup - cur);

2918 2919 2920 2921
		if (kthread_should_stop()) {
			ext4_clear_request_list();
			goto exit_thread;
		}
2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 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
	}

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)
{
2964 2965 2966 2967
	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);
2968 2969 2970
		ext4_clear_request_list();
		kfree(ext4_li_info);
		ext4_li_info = NULL;
2971
		printk(KERN_CRIT "EXT4-fs: error %d creating inode table "
2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039
				 "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.
	 */
3040 3041
	elr->lr_next_sched = jiffies + (prandom_u32() %
				(EXT4_DEF_LI_MAX_START_DELAY * HZ));
3042 3043 3044
	return elr;
}

3045 3046
int ext4_register_li_request(struct super_block *sb,
			     ext4_group_t first_not_zeroed)
3047 3048
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
3049
	struct ext4_li_request *elr = NULL;
3050
	ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count;
3051
	int ret = 0;
3052

3053
	mutex_lock(&ext4_li_mtx);
3054 3055 3056 3057 3058 3059
	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;
3060
		goto out;
3061
	}
3062 3063 3064

	if (first_not_zeroed == ngroups ||
	    (sb->s_flags & MS_RDONLY) ||
3065
	    !test_opt(sb, INIT_INODE_TABLE))
3066
		goto out;
3067 3068

	elr = ext4_li_request_new(sb, first_not_zeroed);
3069 3070 3071 3072
	if (!elr) {
		ret = -ENOMEM;
		goto out;
	}
3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084

	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;
3085 3086 3087 3088 3089 3090
	/*
	 * 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;
3091 3092 3093 3094 3095 3096 3097

	if (!(ext4_li_info->li_state & EXT4_LAZYINIT_RUNNING)) {
		ret = ext4_run_lazyinit_thread();
		if (ret)
			goto out;
	}
out:
3098 3099
	mutex_unlock(&ext4_li_mtx);
	if (ret)
3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113
		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.
	 */
3114
	if (!ext4_li_info || !ext4_lazyinit_task)
3115 3116
		return;

3117
	kthread_stop(ext4_lazyinit_task);
3118 3119
}

3120 3121 3122 3123 3124 3125
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);

3126
	if (ext4_has_metadata_csum(sb)) {
3127
		/* journal checksum v3 */
3128
		compat = 0;
3129
		incompat = JBD2_FEATURE_INCOMPAT_CSUM_V3;
3130 3131 3132 3133 3134 3135
	} else {
		/* journal checksum v1 */
		compat = JBD2_FEATURE_COMPAT_CHECKSUM;
		incompat = 0;
	}

3136 3137 3138 3139
	jbd2_journal_clear_features(sbi->s_journal,
			JBD2_FEATURE_COMPAT_CHECKSUM, 0,
			JBD2_FEATURE_INCOMPAT_CSUM_V3 |
			JBD2_FEATURE_INCOMPAT_CSUM_V2);
3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151
	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 {
3152 3153
		jbd2_journal_clear_features(sbi->s_journal, 0, 0,
				JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT);
3154 3155 3156 3157 3158
	}

	return ret;
}

3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182
/*
 * 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;

3183
	if (!ext4_has_feature_bigalloc(sb))
3184 3185 3186
		return (ext4_bg_has_super(sb, grp) + ext4_bg_num_gdb(sb, grp) +
			sbi->s_itb_per_group + 2);

3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215
	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++;
		}
3216 3217 3218 3219 3220
		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;
3221
		}
3222 3223 3224
		count += j;
		for (; j > 0; j--)
			ext4_set_bit(EXT4_B2C(sbi, s++), buf);
3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238
	}
	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;
3239 3240
	struct inode *j_inode;
	unsigned int j_blocks, j_inum = le32_to_cpu(es->s_journal_inum);
3241 3242
	ext4_group_t i, ngroups = ext4_get_groups_count(sb);
	ext4_fsblk_t overhead = 0;
3243
	char *buf = (char *) get_zeroed_page(GFP_NOFS);
3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270

	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();
	}
3271 3272 3273 3274 3275

	/*
	 * Add the internal journal blocks whether the journal has been
	 * loaded or not
	 */
3276
	if (sbi->s_journal && !sbi->journal_bdev)
3277
		overhead += EXT4_NUM_B2C(sbi, sbi->s_journal->j_maxlen);
3278 3279 3280 3281 3282 3283 3284 3285 3286 3287
	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");
		}
	}
3288 3289 3290 3291 3292 3293
	sbi->s_overhead = overhead;
	smp_wmb();
	free_page((unsigned long) buf);
	return 0;
}

3294
static void ext4_set_resv_clusters(struct super_block *sb)
L
Lukas Czerner 已提交
3295 3296
{
	ext4_fsblk_t resv_clusters;
3297
	struct ext4_sb_info *sbi = EXT4_SB(sb);
L
Lukas Czerner 已提交
3298

3299 3300 3301 3302 3303 3304
	/*
	 * 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.
	 */
3305
	if (!ext4_has_feature_extents(sb))
3306
		return;
L
Lukas Czerner 已提交
3307 3308 3309 3310
	/*
	 * 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
3311
	 * unwritten extents in delalloc path. In most cases such
L
Lukas Czerner 已提交
3312 3313 3314
	 * allocation would require 1, or 2 blocks, higher numbers are
	 * very rare.
	 */
3315 3316
	resv_clusters = (ext4_blocks_count(sbi->s_es) >>
			 sbi->s_cluster_bits);
L
Lukas Czerner 已提交
3317 3318 3319 3320

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

3321
	atomic64_set(&sbi->s_resv_clusters, resv_clusters);
L
Lukas Czerner 已提交
3322 3323
}

3324
static int ext4_fill_super(struct super_block *sb, void *data, int silent)
3325
{
3326
	char *orig_data = kstrdup(data, GFP_KERNEL);
3327
	struct buffer_head *bh;
3328
	struct ext4_super_block *es = NULL;
3329
	struct ext4_sb_info *sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
3330 3331
	ext4_fsblk_t block;
	ext4_fsblk_t sb_block = get_sb_block(&data);
3332
	ext4_fsblk_t logical_sb_block;
3333 3334 3335 3336
	unsigned long offset = 0;
	unsigned long journal_devnum = 0;
	unsigned long def_mount_opts;
	struct inode *root;
3337
	const char *descr;
3338
	int ret = -ENOMEM;
3339
	int blocksize, clustersize;
3340 3341
	unsigned int db_count;
	unsigned int i;
3342
	int needs_recovery, has_huge_files, has_bigalloc;
L
Laurent Vivier 已提交
3343
	__u64 blocks_count;
3344
	int err = 0;
3345
	unsigned int journal_ioprio = DEFAULT_JOURNAL_IOPRIO;
3346
	ext4_group_t first_not_zeroed;
3347

3348 3349
	if ((data && !orig_data) || !sbi)
		goto out_free_base;
3350 3351 3352

	sbi->s_blockgroup_lock =
		kzalloc(sizeof(struct blockgroup_lock), GFP_KERNEL);
3353 3354 3355
	if (!sbi->s_blockgroup_lock)
		goto out_free_base;

3356
	sb->s_fs_info = sbi;
3357
	sbi->s_sb = sb;
3358
	sbi->s_inode_readahead_blks = EXT4_DEF_INODE_READAHEAD_BLKS;
M
Miklos Szeredi 已提交
3359
	sbi->s_sb_block = sb_block;
3360 3361 3362
	if (sb->s_bdev->bd_part)
		sbi->s_sectors_written_start =
			part_stat_read(sb->s_bdev->bd_part, sectors[1]);
3363

3364
	/* Cleanup superblock name */
3365
	strreplace(sb->s_id, '/', '!');
3366

3367
	/* -EINVAL is default */
3368
	ret = -EINVAL;
3369
	blocksize = sb_min_blocksize(sb, EXT4_MIN_BLOCK_SIZE);
3370
	if (!blocksize) {
3371
		ext4_msg(sb, KERN_ERR, "unable to set blocksize");
3372 3373 3374 3375
		goto out_fail;
	}

	/*
3376
	 * The ext4 superblock will not be buffer aligned for other than 1kB
3377 3378
	 * block sizes.  We need to calculate the offset from buffer start.
	 */
3379
	if (blocksize != EXT4_MIN_BLOCK_SIZE) {
3380 3381
		logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE;
		offset = do_div(logical_sb_block, blocksize);
3382
	} else {
3383
		logical_sb_block = sb_block;
3384 3385
	}

3386
	if (!(bh = sb_bread_unmovable(sb, logical_sb_block))) {
3387
		ext4_msg(sb, KERN_ERR, "unable to read superblock");
3388 3389 3390 3391
		goto out_fail;
	}
	/*
	 * Note: s_es must be initialized as soon as possible because
3392
	 *       some ext4 macro-instructions depend on its value
3393
	 */
3394
	es = (struct ext4_super_block *) (bh->b_data + offset);
3395 3396
	sbi->s_es = es;
	sb->s_magic = le16_to_cpu(es->s_magic);
3397 3398
	if (sb->s_magic != EXT4_SUPER_MAGIC)
		goto cantfind_ext4;
T
Theodore Ts'o 已提交
3399
	sbi->s_kbytes_written = le64_to_cpu(es->s_kbytes_written);
3400

3401
	/* Warn if metadata_csum and gdt_csum are both set. */
3402 3403
	if (ext4_has_feature_metadata_csum(sb) &&
	    ext4_has_feature_gdt_csum(sb))
3404
		ext4_warning(sb, "metadata_csum and uninit_bg are "
3405 3406
			     "redundant flags; please run fsck.");

3407 3408 3409 3410 3411 3412 3413 3414
	/* 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;
	}

3415
	/* Load the checksum driver */
3416
	if (ext4_has_feature_metadata_csum(sb)) {
3417 3418 3419 3420 3421 3422 3423 3424 3425
		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;
		}
	}

3426 3427 3428 3429 3430
	/* 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;
3431
		ret = -EFSBADCRC;
3432 3433 3434 3435
		goto cantfind_ext4;
	}

	/* Precompute checksum seed for all metadata */
3436
	if (ext4_has_feature_csum_seed(sb))
3437 3438
		sbi->s_csum_seed = le32_to_cpu(es->s_checksum_seed);
	else if (ext4_has_metadata_csum(sb))
3439 3440 3441
		sbi->s_csum_seed = ext4_chksum(sbi, ~0, es->s_uuid,
					       sizeof(es->s_uuid));

3442 3443
	/* Set defaults before we parse the mount options */
	def_mount_opts = le32_to_cpu(es->s_default_mount_opts);
3444
	set_opt(sb, INIT_INODE_TABLE);
3445
	if (def_mount_opts & EXT4_DEFM_DEBUG)
3446
		set_opt(sb, DEBUG);
3447
	if (def_mount_opts & EXT4_DEFM_BSDGROUPS)
3448
		set_opt(sb, GRPID);
3449
	if (def_mount_opts & EXT4_DEFM_UID16)
3450
		set_opt(sb, NO_UID32);
3451 3452
	/* xattr user namespace & acls are now defaulted on */
	set_opt(sb, XATTR_USER);
T
Theodore Ts'o 已提交
3453
#ifdef CONFIG_EXT4_FS_POSIX_ACL
3454
	set_opt(sb, POSIX_ACL);
3455
#endif
3456 3457 3458 3459
	/* don't forget to enable journal_csum when metadata_csum is enabled. */
	if (ext4_has_metadata_csum(sb))
		set_opt(sb, JOURNAL_CHECKSUM);

3460
	if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_DATA)
3461
		set_opt(sb, JOURNAL_DATA);
3462
	else if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_ORDERED)
3463
		set_opt(sb, ORDERED_DATA);
3464
	else if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_WBACK)
3465
		set_opt(sb, WRITEBACK_DATA);
3466 3467

	if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_PANIC)
3468
		set_opt(sb, ERRORS_PANIC);
3469
	else if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_CONTINUE)
3470
		set_opt(sb, ERRORS_CONT);
3471
	else
3472
		set_opt(sb, ERRORS_RO);
3473 3474
	/* block_validity enabled by default; disable with noblock_validity */
	set_opt(sb, BLOCK_VALIDITY);
3475
	if (def_mount_opts & EXT4_DEFM_DISCARD)
3476
		set_opt(sb, DISCARD);
3477

3478 3479
	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));
3480 3481 3482
	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;
3483

3484
	if ((def_mount_opts & EXT4_DEFM_NOBARRIER) == 0)
3485
		set_opt(sb, BARRIER);
3486

3487 3488 3489 3490
	/*
	 * enable delayed allocation by default
	 * Use -o nodelalloc to turn it off
	 */
3491
	if (!IS_EXT3_SB(sb) && !IS_EXT2_SB(sb) &&
3492
	    ((def_mount_opts & EXT4_DEFM_NODELALLOC) == 0))
3493
		set_opt(sb, DELALLOC);
3494

3495 3496 3497 3498 3499 3500
	/*
	 * 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;

3501 3502 3503 3504 3505 3506 3507 3508 3509 3510 3511 3512 3513
	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);
3514
	}
3515
	sbi->s_def_mount_opt = sbi->s_mount_opt;
3516
	if (!parse_options((char *) data, sb, &journal_devnum,
3517
			   &journal_ioprio, 0))
3518 3519
		goto failed_mount;

3520 3521 3522 3523 3524 3525 3526 3527 3528 3529 3530
	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 "
3531
				 "both data=journal and dioread_nolock");
3532 3533
			goto failed_mount;
		}
R
Ross Zwisler 已提交
3534 3535 3536 3537 3538
		if (test_opt(sb, DAX)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "both data=journal and dax");
			goto failed_mount;
		}
3539 3540
		if (test_opt(sb, DELALLOC))
			clear_opt(sb, DELALLOC);
3541 3542
	} else {
		sb->s_iflags |= SB_I_CGROUPWB;
3543 3544
	}

3545
	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
3546
		(test_opt(sb, POSIX_ACL) ? MS_POSIXACL : 0);
3547

3548
	if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV &&
3549 3550 3551
	    (ext4_has_compat_features(sb) ||
	     ext4_has_ro_compat_features(sb) ||
	     ext4_has_incompat_features(sb)))
3552 3553 3554
		ext4_msg(sb, KERN_WARNING,
		       "feature flags set on rev 0 fs, "
		       "running e2fsck is recommended");
3555

3556 3557
	if (es->s_creator_os == cpu_to_le32(EXT4_OS_HURD)) {
		set_opt2(sb, HURD_COMPAT);
3558
		if (ext4_has_feature_64bit(sb)) {
3559 3560 3561 3562 3563 3564
			ext4_msg(sb, KERN_ERR,
				 "The Hurd can't support 64-bit file systems");
			goto failed_mount;
		}
	}

3565 3566 3567 3568 3569 3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586
	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;
		}
	}

3587 3588 3589 3590 3591
	/*
	 * 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.
	 */
3592
	if (!ext4_feature_set_ok(sb, (sb->s_flags & MS_RDONLY)))
3593
		goto failed_mount;
3594

J
Jan Kara 已提交
3595
	blocksize = BLOCK_SIZE << le32_to_cpu(es->s_log_block_size);
3596 3597
	if (blocksize < EXT4_MIN_BLOCK_SIZE ||
	    blocksize > EXT4_MAX_BLOCK_SIZE) {
3598
		ext4_msg(sb, KERN_ERR,
3599 3600 3601 3602 3603 3604 3605 3606 3607
		       "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));
3608 3609 3610
		goto failed_mount;
	}

3611 3612 3613 3614 3615 3616 3617
	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
3618
	if (sbi->s_mount_opt & EXT4_MOUNT_DAX) {
3619 3620
		err = bdev_dax_supported(sb, blocksize);
		if (err)
R
Ross Zwisler 已提交
3621 3622 3623
			goto failed_mount;
	}

3624
	if (ext4_has_feature_encrypt(sb) && es->s_encryption_level) {
3625 3626 3627 3628 3629
		ext4_msg(sb, KERN_ERR, "Unsupported encryption level %d",
			 es->s_encryption_level);
		goto failed_mount;
	}

3630
	if (sb->s_blocksize != blocksize) {
3631 3632
		/* Validate the filesystem blocksize */
		if (!sb_set_blocksize(sb, blocksize)) {
3633
			ext4_msg(sb, KERN_ERR, "bad block size %d",
3634
					blocksize);
3635 3636 3637
			goto failed_mount;
		}

3638
		brelse(bh);
3639 3640
		logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE;
		offset = do_div(logical_sb_block, blocksize);
3641
		bh = sb_bread_unmovable(sb, logical_sb_block);
3642
		if (!bh) {
3643 3644
			ext4_msg(sb, KERN_ERR,
			       "Can't read superblock on 2nd try");
3645 3646
			goto failed_mount;
		}
3647
		es = (struct ext4_super_block *)(bh->b_data + offset);
3648
		sbi->s_es = es;
3649
		if (es->s_magic != cpu_to_le16(EXT4_SUPER_MAGIC)) {
3650 3651
			ext4_msg(sb, KERN_ERR,
			       "Magic mismatch, very weird!");
3652 3653 3654 3655
			goto failed_mount;
		}
	}

3656
	has_huge_files = ext4_has_feature_huge_file(sb);
3657 3658 3659
	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);
3660

3661 3662 3663
	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;
3664 3665 3666
	} else {
		sbi->s_inode_size = le16_to_cpu(es->s_inode_size);
		sbi->s_first_ino = le32_to_cpu(es->s_first_ino);
3667
		if ((sbi->s_inode_size < EXT4_GOOD_OLD_INODE_SIZE) ||
V
Vignesh Babu 已提交
3668
		    (!is_power_of_2(sbi->s_inode_size)) ||
3669
		    (sbi->s_inode_size > blocksize)) {
3670 3671
			ext4_msg(sb, KERN_ERR,
			       "unsupported inode size: %d",
3672
			       sbi->s_inode_size);
3673 3674
			goto failed_mount;
		}
K
Kalpak Shah 已提交
3675 3676
		if (sbi->s_inode_size > EXT4_GOOD_OLD_INODE_SIZE)
			sb->s_time_gran = 1 << (EXT4_EPOCH_BITS - 2);
3677
	}
3678

3679
	sbi->s_desc_size = le16_to_cpu(es->s_desc_size);
3680
	if (ext4_has_feature_64bit(sb)) {
3681
		if (sbi->s_desc_size < EXT4_MIN_DESC_SIZE_64BIT ||
3682
		    sbi->s_desc_size > EXT4_MAX_DESC_SIZE ||
V
vignesh babu 已提交
3683
		    !is_power_of_2(sbi->s_desc_size)) {
3684 3685
			ext4_msg(sb, KERN_ERR,
			       "unsupported descriptor size %lu",
3686 3687 3688 3689 3690
			       sbi->s_desc_size);
			goto failed_mount;
		}
	} else
		sbi->s_desc_size = EXT4_MIN_DESC_SIZE;
3691

3692 3693
	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);
3694

3695
	sbi->s_inodes_per_block = blocksize / EXT4_INODE_SIZE(sb);
3696
	if (sbi->s_inodes_per_block == 0)
3697
		goto cantfind_ext4;
3698 3699 3700 3701 3702 3703
	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;
	}
3704 3705
	sbi->s_itb_per_group = sbi->s_inodes_per_group /
					sbi->s_inodes_per_block;
3706
	sbi->s_desc_per_block = blocksize / EXT4_DESC_SIZE(sb);
3707 3708
	sbi->s_sbh = bh;
	sbi->s_mount_state = le16_to_cpu(es->s_state);
3709 3710
	sbi->s_addr_per_block_bits = ilog2(EXT4_ADDR_PER_BLOCK(sb));
	sbi->s_desc_per_block_bits = ilog2(EXT4_DESC_PER_BLOCK(sb));
3711

3712
	for (i = 0; i < 4; i++)
3713 3714
		sbi->s_hash_seed[i] = le32_to_cpu(es->s_hash_seed[i]);
	sbi->s_def_hash_version = es->s_def_hash_version;
3715
	if (ext4_has_feature_dir_index(sb)) {
3716 3717 3718 3719
		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) {
3720
#ifdef __CHAR_UNSIGNED__
3721 3722 3723 3724
			if (!(sb->s_flags & MS_RDONLY))
				es->s_flags |=
					cpu_to_le32(EXT2_FLAGS_UNSIGNED_HASH);
			sbi->s_hash_unsigned = 3;
3725
#else
3726 3727 3728
			if (!(sb->s_flags & MS_RDONLY))
				es->s_flags |=
					cpu_to_le32(EXT2_FLAGS_SIGNED_HASH);
3729
#endif
3730
		}
3731
	}
3732

3733 3734
	/* Handle clustersize */
	clustersize = BLOCK_SIZE << le32_to_cpu(es->s_log_cluster_size);
3735
	has_bigalloc = ext4_has_feature_bigalloc(sb);
3736 3737 3738 3739 3740 3741 3742
	if (has_bigalloc) {
		if (clustersize < blocksize) {
			ext4_msg(sb, KERN_ERR,
				 "cluster size (%d) smaller than "
				 "block size (%d)", clustersize, blocksize);
			goto failed_mount;
		}
3743 3744 3745 3746 3747 3748 3749
		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;
		}
3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782
		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;
3783
	}
3784 3785
	sbi->s_cluster_ratio = clustersize / blocksize;

3786 3787 3788 3789
	/* Do we have standard group size of clustersize * 8 blocks ? */
	if (sbi->s_blocks_per_group == clustersize << 3)
		set_opt2(sb, STD_GROUP_SIZE);

3790 3791 3792 3793
	/*
	 * Test whether we have more sectors than will fit in sector_t,
	 * and whether the max offset is addressable by the page cache.
	 */
3794
	err = generic_check_addressable(sb->s_blocksize_bits,
3795
					ext4_blocks_count(es));
3796
	if (err) {
3797
		ext4_msg(sb, KERN_ERR, "filesystem"
3798
			 " too large to mount safely on this system");
3799
		if (sizeof(sector_t) < 8)
3800
			ext4_msg(sb, KERN_WARNING, "CONFIG_LBDAF not enabled");
3801 3802 3803
		goto failed_mount;
	}

3804 3805
	if (EXT4_BLOCKS_PER_GROUP(sb) == 0)
		goto cantfind_ext4;
3806

3807 3808 3809
	/* 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) {
3810 3811
		ext4_msg(sb, KERN_WARNING, "bad geometry: block count %llu "
		       "exceeds size of device (%llu blocks)",
3812 3813 3814 3815
		       ext4_blocks_count(es), blocks_count);
		goto failed_mount;
	}

3816 3817 3818 3819 3820
	/*
	 * 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)) {
3821
		ext4_msg(sb, KERN_WARNING, "bad geometry: first data "
3822 3823 3824
			 "block %u is beyond end of filesystem (%llu)",
			 le32_to_cpu(es->s_first_data_block),
			 ext4_blocks_count(es));
3825 3826
		goto failed_mount;
	}
L
Laurent Vivier 已提交
3827 3828 3829 3830
	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));
3831
	if (blocks_count > ((uint64_t)1<<32) - EXT4_DESC_PER_BLOCK(sb)) {
3832
		ext4_msg(sb, KERN_WARNING, "groups count too large: %u "
3833
		       "(block count %llu, first data block %u, "
3834
		       "blocks per group %lu)", sbi->s_groups_count,
3835 3836 3837 3838 3839
		       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
3840
	sbi->s_groups_count = blocks_count;
3841 3842
	sbi->s_blockfile_groups = min_t(ext4_group_t, sbi->s_groups_count,
			(EXT4_MAX_BLOCK_FILE_PHYS / EXT4_BLOCKS_PER_GROUP(sb)));
3843 3844
	db_count = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) /
		   EXT4_DESC_PER_BLOCK(sb);
3845 3846 3847
	sbi->s_group_desc = ext4_kvmalloc(db_count *
					  sizeof(struct buffer_head *),
					  GFP_KERNEL);
3848
	if (sbi->s_group_desc == NULL) {
3849
		ext4_msg(sb, KERN_ERR, "not enough memory");
3850
		ret = -ENOMEM;
3851 3852 3853
		goto failed_mount;
	}

3854
	bgl_lock_init(sbi->s_blockgroup_lock);
3855 3856

	for (i = 0; i < db_count; i++) {
3857
		block = descriptor_loc(sb, logical_sb_block, i);
3858
		sbi->s_group_desc[i] = sb_bread_unmovable(sb, block);
3859
		if (!sbi->s_group_desc[i]) {
3860 3861
			ext4_msg(sb, KERN_ERR,
			       "can't read group descriptor %d", i);
3862 3863 3864 3865
			db_count = i;
			goto failed_mount2;
		}
	}
3866
	if (!ext4_check_descriptors(sb, logical_sb_block, &first_not_zeroed)) {
3867
		ext4_msg(sb, KERN_ERR, "group descriptors corrupted!");
3868
		ret = -EFSCORRUPTED;
3869
		goto failed_mount2;
3870
	}
3871

3872
	sbi->s_gdb_count = db_count;
3873 3874 3875
	get_random_bytes(&sbi->s_next_generation, sizeof(u32));
	spin_lock_init(&sbi->s_next_gen_lock);

3876 3877
	setup_timer(&sbi->s_err_report, print_daily_error_info,
		(unsigned long) sb);
3878

3879
	/* Register extent status tree shrinker */
3880
	if (ext4_es_register_shrinker(sbi))
3881 3882
		goto failed_mount3;

3883
	sbi->s_stripe = ext4_get_stripe_size(sbi);
3884
	sbi->s_extent_max_zeroout_kb = 32;
3885

3886 3887 3888
	/*
	 * set up enough so that it can read an inode
	 */
3889
	sb->s_op = &ext4_sops;
3890 3891
	sb->s_export_op = &ext4_export_ops;
	sb->s_xattr = ext4_xattr_handlers;
3892
	sb->s_cop = &ext4_cryptops;
3893
#ifdef CONFIG_QUOTA
3894
	sb->dq_op = &ext4_quota_operations;
3895
	if (ext4_has_feature_quota(sb))
3896
		sb->s_qcop = &dquot_quotactl_sysfile_ops;
3897 3898
	else
		sb->s_qcop = &ext4_qctl_operations;
L
Li Xi 已提交
3899
	sb->s_quota_types = QTYPE_MASK_USR | QTYPE_MASK_GRP | QTYPE_MASK_PRJ;
3900
#endif
3901 3902
	memcpy(sb->s_uuid, es->s_uuid, sizeof(es->s_uuid));

3903
	INIT_LIST_HEAD(&sbi->s_orphan); /* unlinked but open files */
3904
	mutex_init(&sbi->s_orphan_lock);
3905 3906 3907 3908

	sb->s_root = NULL;

	needs_recovery = (es->s_last_orphan != 0 ||
3909
			  ext4_has_feature_journal_needs_recovery(sb));
3910

3911
	if (ext4_has_feature_mmp(sb) && !(sb->s_flags & MS_RDONLY))
3912
		if (ext4_multi_mount_protect(sb, le64_to_cpu(es->s_mmp_block)))
3913
			goto failed_mount3a;
3914

3915 3916 3917 3918
	/*
	 * The first inode we look at is the journal inode.  Don't try
	 * root first: it may be modified in the journal!
	 */
3919
	if (!test_opt(sb, NOLOAD) && ext4_has_feature_journal(sb)) {
3920
		if (ext4_load_journal(sb, es, journal_devnum))
3921
			goto failed_mount3a;
3922
	} else if (test_opt(sb, NOLOAD) && !(sb->s_flags & MS_RDONLY) &&
3923
		   ext4_has_feature_journal_needs_recovery(sb)) {
3924 3925
		ext4_msg(sb, KERN_ERR, "required journal recovery "
		       "suppressed and not mounted read-only");
3926
		goto failed_mount_wq;
3927
	} else {
3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952
		/* 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);
3953
		clear_opt(sb, DATA_FLAGS);
3954 3955 3956
		sbi->s_journal = NULL;
		needs_recovery = 0;
		goto no_journal;
3957 3958
	}

3959
	if (ext4_has_feature_64bit(sb) &&
3960 3961
	    !jbd2_journal_set_features(EXT4_SB(sb)->s_journal, 0, 0,
				       JBD2_FEATURE_INCOMPAT_64BIT)) {
3962
		ext4_msg(sb, KERN_ERR, "Failed to set 64-bit journal feature");
3963
		goto failed_mount_wq;
3964 3965
	}

3966 3967 3968 3969
	if (!set_journal_csum_feature_set(sb)) {
		ext4_msg(sb, KERN_ERR, "Failed to set journal checksum "
			 "feature set");
		goto failed_mount_wq;
3970
	}
3971

3972 3973 3974 3975 3976
	/* 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
3977 3978 3979
		 * capabilities: ORDERED_DATA if the journal can
		 * cope, else JOURNAL_DATA
		 */
3980 3981
		if (jbd2_journal_check_available_features
		    (sbi->s_journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE))
3982
			set_opt(sb, ORDERED_DATA);
3983
		else
3984
			set_opt(sb, JOURNAL_DATA);
3985 3986
		break;

3987 3988
	case EXT4_MOUNT_ORDERED_DATA:
	case EXT4_MOUNT_WRITEBACK_DATA:
3989 3990
		if (!jbd2_journal_check_available_features
		    (sbi->s_journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE)) {
3991 3992
			ext4_msg(sb, KERN_ERR, "Journal does not support "
			       "requested data journaling mode");
3993
			goto failed_mount_wq;
3994 3995 3996 3997
		}
	default:
		break;
	}
3998
	set_task_ioprio(sbi->s_journal->j_task, journal_ioprio);
3999

B
Bobi Jam 已提交
4000 4001
	sbi->s_journal->j_commit_callback = ext4_journal_commit_callback;

4002
no_journal:
4003 4004 4005 4006
	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;
4007 4008
	}

4009
	if ((DUMMY_ENCRYPTION_ENABLED(sbi) || ext4_has_feature_encrypt(sb)) &&
4010
	    (blocksize != PAGE_SIZE)) {
4011 4012 4013 4014 4015
		ext4_msg(sb, KERN_ERR,
			 "Unsupported blocksize for fs encryption");
		goto failed_mount_wq;
	}

4016 4017 4018
	if (DUMMY_ENCRYPTION_ENABLED(sbi) && !(sb->s_flags & MS_RDONLY) &&
	    !ext4_has_feature_encrypt(sb)) {
		ext4_set_feature_encrypt(sb);
4019 4020 4021
		ext4_commit_super(sb, 1);
	}

4022 4023 4024 4025 4026 4027 4028
	/*
	 * 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 {
4029 4030
		err = ext4_calculate_overhead(sb);
		if (err)
4031 4032 4033
			goto failed_mount_wq;
	}

T
Tejun Heo 已提交
4034 4035 4036 4037
	/*
	 * The maximum number of concurrent works can be high and
	 * concurrency isn't really necessary.  Limit it to 1.
	 */
4038 4039 4040 4041
	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");
4042
		ret = -ENOMEM;
4043 4044 4045
		goto failed_mount4;
	}

4046
	/*
4047
	 * The jbd2_journal_load will have done any necessary log recovery,
4048 4049 4050
	 * so we can safely mount the rest of the filesystem now.
	 */

4051 4052
	root = ext4_iget(sb, EXT4_ROOT_INO);
	if (IS_ERR(root)) {
4053
		ext4_msg(sb, KERN_ERR, "get root inode failed");
4054
		ret = PTR_ERR(root);
4055
		root = NULL;
4056 4057 4058
		goto failed_mount4;
	}
	if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) {
4059
		ext4_msg(sb, KERN_ERR, "corrupt root inode, run e2fsck");
A
Al Viro 已提交
4060
		iput(root);
4061 4062
		goto failed_mount4;
	}
4063
	sb->s_root = d_make_root(root);
4064
	if (!sb->s_root) {
4065
		ext4_msg(sb, KERN_ERR, "get root dentry failed");
4066 4067 4068
		ret = -ENOMEM;
		goto failed_mount4;
	}
4069

4070 4071
	if (ext4_setup_super(sb, es, sb->s_flags & MS_RDONLY))
		sb->s_flags |= MS_RDONLY;
K
Kalpak Shah 已提交
4072 4073 4074 4075 4076

	/* determine the minimum size of new large inodes, if present */
	if (sbi->s_inode_size > EXT4_GOOD_OLD_INODE_SIZE) {
		sbi->s_want_extra_isize = sizeof(struct ext4_inode) -
						     EXT4_GOOD_OLD_INODE_SIZE;
4077
		if (ext4_has_feature_extra_isize(sb)) {
K
Kalpak Shah 已提交
4078 4079 4080 4081 4082 4083 4084 4085 4086 4087 4088 4089 4090 4091 4092
			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;
4093 4094
		ext4_msg(sb, KERN_INFO, "required extra inode space not"
			 "available");
K
Kalpak Shah 已提交
4095 4096
	}

4097
	ext4_set_resv_clusters(sb);
L
Lukas Czerner 已提交
4098

4099 4100
	err = ext4_setup_system_zone(sb);
	if (err) {
4101
		ext4_msg(sb, KERN_ERR, "failed to initialize system "
4102
			 "zone (%d)", err);
4103 4104 4105 4106 4107 4108 4109 4110
		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);
4111
		goto failed_mount5;
4112 4113
	}

4114 4115 4116
	block = ext4_count_free_clusters(sb);
	ext4_free_blocks_count_set(sbi->s_es, 
				   EXT4_C2B(sbi, block));
4117 4118
	err = percpu_counter_init(&sbi->s_freeclusters_counter, block,
				  GFP_KERNEL);
4119 4120 4121
	if (!err) {
		unsigned long freei = ext4_count_free_inodes(sb);
		sbi->s_es->s_free_inodes_count = cpu_to_le32(freei);
4122 4123
		err = percpu_counter_init(&sbi->s_freeinodes_counter, freei,
					  GFP_KERNEL);
4124 4125 4126
	}
	if (!err)
		err = percpu_counter_init(&sbi->s_dirs_counter,
4127
					  ext4_count_dirs(sb), GFP_KERNEL);
4128
	if (!err)
4129 4130
		err = percpu_counter_init(&sbi->s_dirtyclusters_counter, 0,
					  GFP_KERNEL);
4131 4132 4133
	if (!err)
		err = percpu_init_rwsem(&sbi->s_journal_flag_rwsem);

4134 4135 4136 4137 4138
	if (err) {
		ext4_msg(sb, KERN_ERR, "insufficient memory");
		goto failed_mount6;
	}

4139
	if (ext4_has_feature_flex_bg(sb))
4140 4141 4142 4143 4144 4145 4146
		if (!ext4_fill_flex_info(sb)) {
			ext4_msg(sb, KERN_ERR,
			       "unable to initialize "
			       "flex_bg meta info!");
			goto failed_mount6;
		}

4147 4148
	err = ext4_register_li_request(sb, first_not_zeroed);
	if (err)
4149
		goto failed_mount6;
4150

4151
	err = ext4_register_sysfs(sb);
4152 4153
	if (err)
		goto failed_mount7;
T
Theodore Ts'o 已提交
4154

4155 4156
#ifdef CONFIG_QUOTA
	/* Enable quota usage during mount. */
4157
	if (ext4_has_feature_quota(sb) && !(sb->s_flags & MS_RDONLY)) {
4158 4159 4160 4161 4162 4163
		err = ext4_enable_quotas(sb);
		if (err)
			goto failed_mount8;
	}
#endif  /* CONFIG_QUOTA */

4164 4165 4166
	EXT4_SB(sb)->s_mount_state |= EXT4_ORPHAN_FS;
	ext4_orphan_cleanup(sb, es);
	EXT4_SB(sb)->s_mount_state &= ~EXT4_ORPHAN_FS;
4167
	if (needs_recovery) {
4168
		ext4_msg(sb, KERN_INFO, "recovery complete");
4169 4170 4171 4172 4173 4174 4175 4176 4177 4178 4179 4180
		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";

4181 4182 4183 4184 4185 4186 4187 4188
	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");
	}

4189 4190
	if (___ratelimit(&ext4_mount_msg_ratelimit, "EXT4-fs mount"))
		ext4_msg(sb, KERN_INFO, "mounted filesystem with%s. "
4191 4192 4193
			 "Opts: %.*s%s%s", descr,
			 (int) sizeof(sbi->s_es->s_mount_opts),
			 sbi->s_es->s_mount_opts,
4194
			 *sbi->s_es->s_mount_opts ? "; " : "", orig_data);
4195

4196 4197
	if (es->s_error_count)
		mod_timer(&sbi->s_err_report, jiffies + 300*HZ); /* 5 minutes */
4198

4199 4200 4201 4202 4203
	/* 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);

4204
	kfree(orig_data);
4205 4206 4207 4208 4209
#ifdef CONFIG_EXT4_FS_ENCRYPTION
	memcpy(sbi->key_prefix, EXT4_KEY_DESC_PREFIX,
				EXT4_KEY_DESC_PREFIX_SIZE);
	sbi->key_prefix_size = EXT4_KEY_DESC_PREFIX_SIZE;
#endif
4210 4211
	return 0;

4212
cantfind_ext4:
4213
	if (!silent)
4214
		ext4_msg(sb, KERN_ERR, "VFS: Can't find ext4 filesystem");
4215 4216
	goto failed_mount;

4217 4218
#ifdef CONFIG_QUOTA
failed_mount8:
4219
	ext4_unregister_sysfs(sb);
4220
#endif
4221 4222 4223
failed_mount7:
	ext4_unregister_li_request(sb);
failed_mount6:
4224
	ext4_mb_release(sb);
4225
	if (sbi->s_flex_groups)
A
Al Viro 已提交
4226
		kvfree(sbi->s_flex_groups);
4227 4228 4229 4230
	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);
4231
failed_mount5:
4232 4233 4234
	ext4_ext_release(sb);
	ext4_release_system_zone(sb);
failed_mount4a:
A
Al Viro 已提交
4235
	dput(sb->s_root);
4236
	sb->s_root = NULL;
A
Al Viro 已提交
4237
failed_mount4:
4238
	ext4_msg(sb, KERN_ERR, "mount failed");
4239 4240
	if (EXT4_SB(sb)->rsv_conversion_wq)
		destroy_workqueue(EXT4_SB(sb)->rsv_conversion_wq);
4241
failed_mount_wq:
J
Jan Kara 已提交
4242 4243 4244 4245
	if (sbi->s_mb_cache) {
		ext4_xattr_destroy_cache(sbi->s_mb_cache);
		sbi->s_mb_cache = NULL;
	}
4246 4247 4248 4249
	if (sbi->s_journal) {
		jbd2_journal_destroy(sbi->s_journal);
		sbi->s_journal = NULL;
	}
4250
failed_mount3a:
4251
	ext4_es_unregister_shrinker(sbi);
4252
failed_mount3:
4253
	del_timer_sync(&sbi->s_err_report);
4254 4255
	if (sbi->s_mmp_tsk)
		kthread_stop(sbi->s_mmp_tsk);
4256 4257 4258
failed_mount2:
	for (i = 0; i < db_count; i++)
		brelse(sbi->s_group_desc[i]);
A
Al Viro 已提交
4259
	kvfree(sbi->s_group_desc);
4260
failed_mount:
4261 4262
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
4263
#ifdef CONFIG_QUOTA
J
Jan Kara 已提交
4264
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
4265 4266
		kfree(sbi->s_qf_names[i]);
#endif
4267
	ext4_blkdev_remove(sbi);
4268 4269 4270
	brelse(bh);
out_fail:
	sb->s_fs_info = NULL;
4271
	kfree(sbi->s_blockgroup_lock);
4272
out_free_base:
4273
	kfree(sbi);
4274
	kfree(orig_data);
4275
	return err ? err : ret;
4276 4277 4278 4279 4280 4281 4282
}

/*
 * 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.
 */
4283
static void ext4_init_journal_params(struct super_block *sb, journal_t *journal)
4284
{
4285
	struct ext4_sb_info *sbi = EXT4_SB(sb);
4286

4287 4288 4289
	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;
4290

4291
	write_lock(&journal->j_state_lock);
4292
	if (test_opt(sb, BARRIER))
4293
		journal->j_flags |= JBD2_BARRIER;
4294
	else
4295
		journal->j_flags &= ~JBD2_BARRIER;
4296 4297 4298 4299
	if (test_opt(sb, DATA_ERR_ABORT))
		journal->j_flags |= JBD2_ABORT_ON_SYNCDATA_ERR;
	else
		journal->j_flags &= ~JBD2_ABORT_ON_SYNCDATA_ERR;
4300
	write_unlock(&journal->j_state_lock);
4301 4302
}

4303 4304
static struct inode *ext4_get_journal_inode(struct super_block *sb,
					     unsigned int journal_inum)
4305 4306 4307
{
	struct inode *journal_inode;

4308 4309 4310 4311 4312
	/*
	 * 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.
	 */
4313 4314
	journal_inode = ext4_iget(sb, journal_inum);
	if (IS_ERR(journal_inode)) {
4315
		ext4_msg(sb, KERN_ERR, "no journal found");
4316 4317 4318 4319 4320
		return NULL;
	}
	if (!journal_inode->i_nlink) {
		make_bad_inode(journal_inode);
		iput(journal_inode);
4321
		ext4_msg(sb, KERN_ERR, "journal inode is deleted");
4322 4323 4324
		return NULL;
	}

4325
	jbd_debug(2, "Journal inode found at %p: %lld bytes\n",
4326
		  journal_inode, journal_inode->i_size);
4327
	if (!S_ISREG(journal_inode->i_mode)) {
4328
		ext4_msg(sb, KERN_ERR, "invalid journal inode");
4329 4330 4331
		iput(journal_inode);
		return NULL;
	}
4332 4333 4334 4335 4336 4337 4338 4339 4340 4341 4342 4343 4344 4345
	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;
4346

4347
	journal = jbd2_journal_init_inode(journal_inode);
4348
	if (!journal) {
4349
		ext4_msg(sb, KERN_ERR, "Could not load journal inode");
4350 4351 4352 4353
		iput(journal_inode);
		return NULL;
	}
	journal->j_private = sb;
4354
	ext4_init_journal_params(sb, journal);
4355 4356 4357
	return journal;
}

4358
static journal_t *ext4_get_dev_journal(struct super_block *sb,
4359 4360
				       dev_t j_dev)
{
4361
	struct buffer_head *bh;
4362
	journal_t *journal;
4363 4364
	ext4_fsblk_t start;
	ext4_fsblk_t len;
4365
	int hblock, blocksize;
4366
	ext4_fsblk_t sb_block;
4367
	unsigned long offset;
4368
	struct ext4_super_block *es;
4369 4370
	struct block_device *bdev;

4371
	BUG_ON(!ext4_has_feature_journal(sb));
4372

4373
	bdev = ext4_blkdev_get(j_dev, sb);
4374 4375 4376 4377
	if (bdev == NULL)
		return NULL;

	blocksize = sb->s_blocksize;
4378
	hblock = bdev_logical_block_size(bdev);
4379
	if (blocksize < hblock) {
4380 4381
		ext4_msg(sb, KERN_ERR,
			"blocksize too small for journal device");
4382 4383 4384
		goto out_bdev;
	}

4385 4386
	sb_block = EXT4_MIN_BLOCK_SIZE / blocksize;
	offset = EXT4_MIN_BLOCK_SIZE % blocksize;
4387 4388
	set_blocksize(bdev, blocksize);
	if (!(bh = __bread(bdev, sb_block, blocksize))) {
4389 4390
		ext4_msg(sb, KERN_ERR, "couldn't read superblock of "
		       "external journal");
4391 4392 4393
		goto out_bdev;
	}

4394
	es = (struct ext4_super_block *) (bh->b_data + offset);
4395
	if ((le16_to_cpu(es->s_magic) != EXT4_SUPER_MAGIC) ||
4396
	    !(le32_to_cpu(es->s_feature_incompat) &
4397
	      EXT4_FEATURE_INCOMPAT_JOURNAL_DEV)) {
4398 4399
		ext4_msg(sb, KERN_ERR, "external journal has "
					"bad superblock");
4400 4401 4402 4403
		brelse(bh);
		goto out_bdev;
	}

4404 4405 4406 4407 4408 4409 4410 4411 4412
	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;
	}

4413
	if (memcmp(EXT4_SB(sb)->s_es->s_journal_uuid, es->s_uuid, 16)) {
4414
		ext4_msg(sb, KERN_ERR, "journal UUID does not match");
4415 4416 4417 4418
		brelse(bh);
		goto out_bdev;
	}

L
Laurent Vivier 已提交
4419
	len = ext4_blocks_count(es);
4420 4421 4422
	start = sb_block + 1;
	brelse(bh);	/* we're done with the superblock */

4423
	journal = jbd2_journal_init_dev(bdev, sb->s_bdev,
4424 4425
					start, len, blocksize);
	if (!journal) {
4426
		ext4_msg(sb, KERN_ERR, "failed to create device journal");
4427 4428 4429
		goto out_bdev;
	}
	journal->j_private = sb;
4430
	ll_rw_block(REQ_OP_READ, REQ_META | REQ_PRIO, 1, &journal->j_sb_buffer);
4431 4432
	wait_on_buffer(journal->j_sb_buffer);
	if (!buffer_uptodate(journal->j_sb_buffer)) {
4433
		ext4_msg(sb, KERN_ERR, "I/O error on journal device");
4434 4435 4436
		goto out_journal;
	}
	if (be32_to_cpu(journal->j_superblock->s_nr_users) != 1) {
4437 4438
		ext4_msg(sb, KERN_ERR, "External journal has more than one "
					"user (unsupported) - %d",
4439 4440 4441
			be32_to_cpu(journal->j_superblock->s_nr_users));
		goto out_journal;
	}
4442 4443
	EXT4_SB(sb)->journal_bdev = bdev;
	ext4_init_journal_params(sb, journal);
4444
	return journal;
4445

4446
out_journal:
4447
	jbd2_journal_destroy(journal);
4448
out_bdev:
4449
	ext4_blkdev_put(bdev);
4450 4451 4452
	return NULL;
}

4453 4454
static int ext4_load_journal(struct super_block *sb,
			     struct ext4_super_block *es,
4455 4456 4457 4458 4459 4460 4461 4462
			     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;

4463
	BUG_ON(!ext4_has_feature_journal(sb));
4464

4465 4466
	if (journal_devnum &&
	    journal_devnum != le32_to_cpu(es->s_journal_dev)) {
4467 4468
		ext4_msg(sb, KERN_INFO, "external journal device major/minor "
			"numbers have changed");
4469 4470 4471 4472 4473 4474 4475 4476 4477 4478 4479
		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.
	 */
4480
	if (ext4_has_feature_journal_needs_recovery(sb)) {
4481
		if (sb->s_flags & MS_RDONLY) {
4482 4483
			ext4_msg(sb, KERN_INFO, "INFO: recovery "
					"required on readonly filesystem");
4484
			if (really_read_only) {
4485 4486
				ext4_msg(sb, KERN_ERR, "write access "
					"unavailable, cannot proceed");
4487 4488
				return -EROFS;
			}
4489 4490
			ext4_msg(sb, KERN_INFO, "write access will "
			       "be enabled during recovery");
4491 4492 4493 4494
		}
	}

	if (journal_inum && journal_dev) {
4495 4496
		ext4_msg(sb, KERN_ERR, "filesystem has both journal "
		       "and inode journals!");
4497 4498 4499 4500
		return -EINVAL;
	}

	if (journal_inum) {
4501
		if (!(journal = ext4_get_journal(sb, journal_inum)))
4502 4503
			return -EINVAL;
	} else {
4504
		if (!(journal = ext4_get_dev_journal(sb, journal_dev)))
4505 4506 4507
			return -EINVAL;
	}

4508
	if (!(journal->j_flags & JBD2_BARRIER))
4509
		ext4_msg(sb, KERN_INFO, "barriers disabled");
4510

4511
	if (!ext4_has_feature_journal_needs_recovery(sb))
4512
		err = jbd2_journal_wipe(journal, !really_read_only);
4513 4514 4515 4516 4517
	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);
4518
		err = jbd2_journal_load(journal);
4519 4520 4521 4522 4523
		if (save)
			memcpy(((char *) es) + EXT4_S_ERR_START,
			       save, EXT4_S_ERR_LEN);
		kfree(save);
	}
4524 4525

	if (err) {
4526
		ext4_msg(sb, KERN_ERR, "error loading journal");
4527
		jbd2_journal_destroy(journal);
4528 4529 4530
		return err;
	}

4531 4532
	EXT4_SB(sb)->s_journal = journal;
	ext4_clear_journal_err(sb, es);
4533

4534
	if (!really_read_only && journal_devnum &&
4535 4536 4537 4538
	    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. */
4539
		ext4_commit_super(sb, 1);
4540 4541 4542 4543 4544
	}

	return 0;
}

4545
static int ext4_commit_super(struct super_block *sb, int sync)
4546
{
4547
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;
4548
	struct buffer_head *sbh = EXT4_SB(sb)->s_sbh;
4549
	int error = 0;
4550

4551
	if (!sbh || block_device_ejected(sb))
4552
		return error;
4553 4554 4555 4556 4557 4558 4559 4560 4561 4562 4563 4564
	/*
	 * 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());
4565 4566 4567
	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
4568 4569
			    ((part_stat_read(sb->s_bdev->bd_part, sectors[1]) -
			      EXT4_SB(sb)->s_sectors_written_start) >> 1));
4570 4571 4572
	else
		es->s_kbytes_written =
			cpu_to_le64(EXT4_SB(sb)->s_kbytes_written);
4573 4574
	if (percpu_counter_initialized(&EXT4_SB(sb)->s_freeclusters_counter))
		ext4_free_blocks_count_set(es,
4575 4576
			EXT4_C2B(EXT4_SB(sb), percpu_counter_sum_positive(
				&EXT4_SB(sb)->s_freeclusters_counter)));
4577 4578 4579
	if (percpu_counter_initialized(&EXT4_SB(sb)->s_freeinodes_counter))
		es->s_free_inodes_count =
			cpu_to_le32(percpu_counter_sum_positive(
4580
				&EXT4_SB(sb)->s_freeinodes_counter));
4581
	BUFFER_TRACE(sbh, "marking dirty");
4582
	ext4_superblock_csum_set(sb);
4583 4584
	if (sync)
		lock_buffer(sbh);
4585 4586 4587 4588 4589 4590 4591 4592 4593 4594 4595 4596 4597 4598
	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);
	}
4599
	mark_buffer_dirty(sbh);
4600
	if (sync) {
4601
		unlock_buffer(sbh);
4602 4603
		error = __sync_dirty_buffer(sbh,
			test_opt(sb, BARRIER) ? WRITE_FUA : WRITE_SYNC);
4604 4605 4606 4607 4608
		if (error)
			return error;

		error = buffer_write_io_error(sbh);
		if (error) {
4609 4610
			ext4_msg(sb, KERN_ERR, "I/O error while writing "
			       "superblock");
4611 4612 4613 4614
			clear_buffer_write_io_error(sbh);
			set_buffer_uptodate(sbh);
		}
	}
4615
	return error;
4616 4617 4618 4619 4620 4621 4622
}

/*
 * 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.
 */
4623 4624
static void ext4_mark_recovery_complete(struct super_block *sb,
					struct ext4_super_block *es)
4625
{
4626
	journal_t *journal = EXT4_SB(sb)->s_journal;
4627

4628
	if (!ext4_has_feature_journal(sb)) {
4629 4630 4631
		BUG_ON(journal != NULL);
		return;
	}
4632
	jbd2_journal_lock_updates(journal);
4633 4634 4635
	if (jbd2_journal_flush(journal) < 0)
		goto out;

4636
	if (ext4_has_feature_journal_needs_recovery(sb) &&
4637
	    sb->s_flags & MS_RDONLY) {
4638
		ext4_clear_feature_journal_needs_recovery(sb);
4639
		ext4_commit_super(sb, 1);
4640
	}
4641 4642

out:
4643
	jbd2_journal_unlock_updates(journal);
4644 4645 4646 4647 4648 4649 4650
}

/*
 * 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.
 */
4651 4652
static void ext4_clear_journal_err(struct super_block *sb,
				   struct ext4_super_block *es)
4653 4654 4655 4656 4657
{
	journal_t *journal;
	int j_errno;
	const char *errstr;

4658
	BUG_ON(!ext4_has_feature_journal(sb));
4659

4660
	journal = EXT4_SB(sb)->s_journal;
4661 4662 4663

	/*
	 * Now check for any error status which may have been recorded in the
4664
	 * journal by a prior ext4_error() or ext4_abort()
4665 4666
	 */

4667
	j_errno = jbd2_journal_errno(journal);
4668 4669 4670
	if (j_errno) {
		char nbuf[16];

4671
		errstr = ext4_decode_error(sb, j_errno, nbuf);
4672
		ext4_warning(sb, "Filesystem error recorded "
4673
			     "from previous mount: %s", errstr);
4674
		ext4_warning(sb, "Marking fs in need of filesystem check.");
4675

4676 4677
		EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS;
		es->s_state |= cpu_to_le16(EXT4_ERROR_FS);
4678
		ext4_commit_super(sb, 1);
4679

4680
		jbd2_journal_clear_err(journal);
4681
		jbd2_journal_update_sb_errno(journal);
4682 4683 4684 4685 4686 4687 4688
	}
}

/*
 * Force the running and committing transactions to commit,
 * and wait on the commit.
 */
4689
int ext4_force_commit(struct super_block *sb)
4690 4691 4692 4693 4694 4695
{
	journal_t *journal;

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

4696
	journal = EXT4_SB(sb)->s_journal;
4697
	return ext4_journal_force_commit(journal);
4698 4699
}

4700
static int ext4_sync_fs(struct super_block *sb, int wait)
4701
{
4702
	int ret = 0;
4703
	tid_t target;
4704
	bool needs_barrier = false;
4705
	struct ext4_sb_info *sbi = EXT4_SB(sb);
4706

4707
	trace_ext4_sync_fs(sb, wait);
4708
	flush_workqueue(sbi->rsv_conversion_wq);
4709 4710 4711 4712 4713
	/*
	 * Writeback quota in non-journalled quota case - journalled quota has
	 * no dirty dquots
	 */
	dquot_writeback_dquots(sb, -1);
4714 4715 4716 4717 4718
	/*
	 * 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.
	 */
4719 4720 4721 4722 4723 4724 4725 4726 4727 4728 4729 4730
	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))
4731 4732 4733 4734 4735 4736
		needs_barrier = true;
	if (needs_barrier) {
		int err;
		err = blkdev_issue_flush(sb->s_bdev, GFP_KERNEL, NULL);
		if (!ret)
			ret = err;
4737
	}
4738 4739 4740 4741

	return ret;
}

4742 4743 4744
/*
 * 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.
4745 4746
 *
 * Note that only this function cannot bring a filesystem to be in a clean
4747 4748
 * state independently. It relies on upper layer to stop all data & metadata
 * modifications.
4749
 */
4750
static int ext4_freeze(struct super_block *sb)
4751
{
4752 4753
	int error = 0;
	journal_t *journal;
4754

4755 4756
	if (sb->s_flags & MS_RDONLY)
		return 0;
4757

4758
	journal = EXT4_SB(sb)->s_journal;
4759

4760 4761 4762
	if (journal) {
		/* Now we set up the journal barrier. */
		jbd2_journal_lock_updates(journal);
4763

4764 4765 4766 4767 4768 4769 4770
		/*
		 * Don't clear the needs_recovery flag if we failed to
		 * flush the journal.
		 */
		error = jbd2_journal_flush(journal);
		if (error < 0)
			goto out;
4771 4772

		/* Journal blocked and flushed, clear needs_recovery flag. */
4773
		ext4_clear_feature_journal_needs_recovery(sb);
4774
	}
4775 4776

	error = ext4_commit_super(sb, 1);
4777
out:
4778 4779 4780
	if (journal)
		/* we rely on upper layer to stop further updates */
		jbd2_journal_unlock_updates(journal);
4781
	return error;
4782 4783 4784 4785 4786 4787
}

/*
 * 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.
 */
4788
static int ext4_unfreeze(struct super_block *sb)
4789
{
4790 4791 4792
	if (sb->s_flags & MS_RDONLY)
		return 0;

4793 4794
	if (EXT4_SB(sb)->s_journal) {
		/* Reset the needs_recovery flag before the fs is unlocked. */
4795
		ext4_set_feature_journal_needs_recovery(sb);
4796 4797
	}

4798
	ext4_commit_super(sb, 1);
4799
	return 0;
4800 4801
}

4802 4803 4804 4805 4806
/*
 * Structure to save mount options for ext4_remount's benefit
 */
struct ext4_mount_options {
	unsigned long s_mount_opt;
4807
	unsigned long s_mount_opt2;
4808 4809
	kuid_t s_resuid;
	kgid_t s_resgid;
4810 4811 4812 4813
	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
4814
	char *s_qf_names[EXT4_MAXQUOTAS];
4815 4816 4817
#endif
};

4818
static int ext4_remount(struct super_block *sb, int *flags, char *data)
4819
{
4820
	struct ext4_super_block *es;
4821
	struct ext4_sb_info *sbi = EXT4_SB(sb);
4822
	unsigned long old_sb_flags;
4823
	struct ext4_mount_options old_opts;
4824
	int enable_quota = 0;
4825
	ext4_group_t g;
4826
	unsigned int journal_ioprio = DEFAULT_JOURNAL_IOPRIO;
4827
	int err = 0;
4828
#ifdef CONFIG_QUOTA
4829
	int i, j;
4830
#endif
4831
	char *orig_data = kstrdup(data, GFP_KERNEL);
4832 4833 4834 4835

	/* Store the original options */
	old_sb_flags = sb->s_flags;
	old_opts.s_mount_opt = sbi->s_mount_opt;
4836
	old_opts.s_mount_opt2 = sbi->s_mount_opt2;
4837 4838 4839
	old_opts.s_resuid = sbi->s_resuid;
	old_opts.s_resgid = sbi->s_resgid;
	old_opts.s_commit_interval = sbi->s_commit_interval;
4840 4841
	old_opts.s_min_batch_time = sbi->s_min_batch_time;
	old_opts.s_max_batch_time = sbi->s_max_batch_time;
4842 4843
#ifdef CONFIG_QUOTA
	old_opts.s_jquota_fmt = sbi->s_jquota_fmt;
J
Jan Kara 已提交
4844
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
4845 4846 4847 4848 4849 4850
		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]);
4851
				kfree(orig_data);
4852 4853 4854 4855
				return -ENOMEM;
			}
		} else
			old_opts.s_qf_names[i] = NULL;
4856
#endif
4857 4858
	if (sbi->s_journal && sbi->s_journal->j_task->io_context)
		journal_ioprio = sbi->s_journal->j_task->io_context->ioprio;
4859

4860
	if (!parse_options(data, sb, NULL, &journal_ioprio, 1)) {
4861 4862 4863 4864
		err = -EINVAL;
		goto restore_opts;
	}

4865
	if ((old_opts.s_mount_opt & EXT4_MOUNT_JOURNAL_CHECKSUM) ^
4866 4867
	    test_opt(sb, JOURNAL_CHECKSUM)) {
		ext4_msg(sb, KERN_ERR, "changing journal_checksum "
4868 4869
			 "during remount not supported; ignoring");
		sbi->s_mount_opt ^= EXT4_MOUNT_JOURNAL_CHECKSUM;
4870 4871
	}

4872 4873 4874 4875 4876 4877 4878 4879 4880 4881 4882 4883 4884
	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
4885 4886 4887 4888 4889 4890 4891 4892 4893 4894 4895 4896
		if (test_opt(sb, DAX)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "both data=journal and dax");
			err = -EINVAL;
			goto restore_opts;
		}
	}

	if ((sbi->s_mount_opt ^ old_opts.s_mount_opt) & EXT4_MOUNT_DAX) {
		ext4_msg(sb, KERN_WARNING, "warning: refusing change of "
			"dax flag with busy inodes while remounting");
		sbi->s_mount_opt ^= EXT4_MOUNT_DAX;
4897 4898
	}

4899
	if (sbi->s_mount_flags & EXT4_MF_FS_ABORTED)
4900
		ext4_abort(sb, "Abort forced by user");
4901 4902

	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
4903
		(test_opt(sb, POSIX_ACL) ? MS_POSIXACL : 0);
4904 4905 4906

	es = sbi->s_es;

4907
	if (sbi->s_journal) {
4908
		ext4_init_journal_params(sb, sbi->s_journal);
4909 4910
		set_task_ioprio(sbi->s_journal->j_task, journal_ioprio);
	}
4911

4912 4913 4914
	if (*flags & MS_LAZYTIME)
		sb->s_flags |= MS_LAZYTIME;

4915
	if ((*flags & MS_RDONLY) != (sb->s_flags & MS_RDONLY)) {
4916
		if (sbi->s_mount_flags & EXT4_MF_FS_ABORTED) {
4917 4918 4919 4920 4921
			err = -EROFS;
			goto restore_opts;
		}

		if (*flags & MS_RDONLY) {
4922 4923 4924
			err = sync_filesystem(sb);
			if (err < 0)
				goto restore_opts;
4925 4926
			err = dquot_suspend(sb, -1);
			if (err < 0)
4927 4928
				goto restore_opts;

4929 4930 4931 4932 4933 4934 4935 4936 4937 4938 4939
			/*
			 * 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.
			 */
4940 4941
			if (!(es->s_state & cpu_to_le16(EXT4_VALID_FS)) &&
			    (sbi->s_mount_state & EXT4_VALID_FS))
4942 4943
				es->s_state = cpu_to_le16(sbi->s_mount_state);

4944
			if (sbi->s_journal)
4945
				ext4_mark_recovery_complete(sb, es);
4946
		} else {
4947
			/* Make sure we can mount this feature set readwrite */
4948
			if (ext4_has_feature_readonly(sb) ||
D
Darrick J. Wong 已提交
4949
			    !ext4_feature_set_ok(sb, 0)) {
4950 4951 4952
				err = -EROFS;
				goto restore_opts;
			}
4953 4954
			/*
			 * Make sure the group descriptor checksums
4955
			 * are sane.  If they aren't, refuse to remount r/w.
4956 4957 4958 4959 4960
			 */
			for (g = 0; g < sbi->s_groups_count; g++) {
				struct ext4_group_desc *gdp =
					ext4_get_group_desc(sb, g, NULL);

4961
				if (!ext4_group_desc_csum_verify(sb, g, gdp)) {
4962 4963
					ext4_msg(sb, KERN_ERR,
	       "ext4_remount: Checksum for group %u failed (%u!=%u)",
4964
		g, le16_to_cpu(ext4_group_desc_csum(sb, g, gdp)),
4965
					       le16_to_cpu(gdp->bg_checksum));
4966
					err = -EFSBADCRC;
4967 4968 4969 4970
					goto restore_opts;
				}
			}

4971 4972 4973 4974 4975 4976
			/*
			 * 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) {
4977
				ext4_msg(sb, KERN_WARNING, "Couldn't "
4978 4979
				       "remount RDWR because of unprocessed "
				       "orphan inode list.  Please "
4980
				       "umount/remount instead");
4981 4982 4983 4984
				err = -EINVAL;
				goto restore_opts;
			}

4985 4986 4987 4988 4989 4990
			/*
			 * 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.)
			 */
4991 4992
			if (sbi->s_journal)
				ext4_clear_journal_err(sb, es);
4993
			sbi->s_mount_state = le16_to_cpu(es->s_state);
4994
			if (!ext4_setup_super(sb, es, 0))
4995
				sb->s_flags &= ~MS_RDONLY;
4996
			if (ext4_has_feature_mmp(sb))
4997 4998 4999 5000 5001
				if (ext4_multi_mount_protect(sb,
						le64_to_cpu(es->s_mmp_block))) {
					err = -EROFS;
					goto restore_opts;
				}
5002
			enable_quota = 1;
5003 5004
		}
	}
5005 5006 5007 5008 5009 5010 5011 5012 5013 5014 5015 5016 5017

	/*
	 * 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);
	}

5018
	ext4_setup_system_zone(sb);
5019
	if (sbi->s_journal == NULL && !(old_sb_flags & MS_RDONLY))
5020
		ext4_commit_super(sb, 1);
5021

5022 5023
#ifdef CONFIG_QUOTA
	/* Release old quota file names */
J
Jan Kara 已提交
5024
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
5025
		kfree(old_opts.s_qf_names[i]);
5026 5027 5028
	if (enable_quota) {
		if (sb_any_quota_suspended(sb))
			dquot_resume(sb, -1);
5029
		else if (ext4_has_feature_quota(sb)) {
5030
			err = ext4_enable_quotas(sb);
5031
			if (err)
5032 5033 5034
				goto restore_opts;
		}
	}
5035
#endif
5036

5037
	*flags = (*flags & ~MS_LAZYTIME) | (sb->s_flags & MS_LAZYTIME);
5038 5039
	ext4_msg(sb, KERN_INFO, "re-mounted. Opts: %s", orig_data);
	kfree(orig_data);
5040
	return 0;
5041

5042 5043 5044
restore_opts:
	sb->s_flags = old_sb_flags;
	sbi->s_mount_opt = old_opts.s_mount_opt;
5045
	sbi->s_mount_opt2 = old_opts.s_mount_opt2;
5046 5047 5048
	sbi->s_resuid = old_opts.s_resuid;
	sbi->s_resgid = old_opts.s_resgid;
	sbi->s_commit_interval = old_opts.s_commit_interval;
5049 5050
	sbi->s_min_batch_time = old_opts.s_min_batch_time;
	sbi->s_max_batch_time = old_opts.s_max_batch_time;
5051 5052
#ifdef CONFIG_QUOTA
	sbi->s_jquota_fmt = old_opts.s_jquota_fmt;
J
Jan Kara 已提交
5053
	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
5054
		kfree(sbi->s_qf_names[i]);
5055 5056 5057
		sbi->s_qf_names[i] = old_opts.s_qf_names[i];
	}
#endif
5058
	kfree(orig_data);
5059 5060 5061
	return err;
}

L
Li Xi 已提交
5062 5063 5064 5065 5066 5067 5068 5069 5070 5071 5072 5073 5074 5075 5076 5077 5078 5079 5080 5081 5082 5083 5084 5085 5086 5087 5088 5089 5090 5091 5092 5093 5094 5095 5096 5097 5098 5099 5100 5101 5102 5103
#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

5104
static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf)
5105 5106
{
	struct super_block *sb = dentry->d_sb;
5107 5108
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_super_block *es = sbi->s_es;
L
Lukas Czerner 已提交
5109
	ext4_fsblk_t overhead = 0, resv_blocks;
P
Pekka Enberg 已提交
5110
	u64 fsid;
5111
	s64 bfree;
L
Lukas Czerner 已提交
5112
	resv_blocks = EXT4_C2B(sbi, atomic64_read(&sbi->s_resv_clusters));
5113

5114 5115
	if (!test_opt(sb, MINIX_DF))
		overhead = sbi->s_overhead;
5116

5117
	buf->f_type = EXT4_SUPER_MAGIC;
5118
	buf->f_bsize = sb->s_blocksize;
5119
	buf->f_blocks = ext4_blocks_count(es) - EXT4_C2B(sbi, overhead);
5120 5121
	bfree = percpu_counter_sum_positive(&sbi->s_freeclusters_counter) -
		percpu_counter_sum_positive(&sbi->s_dirtyclusters_counter);
5122
	/* prevent underflow in case that few free space is available */
5123
	buf->f_bfree = EXT4_C2B(sbi, max_t(s64, bfree, 0));
L
Lukas Czerner 已提交
5124 5125 5126
	buf->f_bavail = buf->f_bfree -
			(ext4_r_blocks_count(es) + resv_blocks);
	if (buf->f_bfree < (ext4_r_blocks_count(es) + resv_blocks))
5127 5128
		buf->f_bavail = 0;
	buf->f_files = le32_to_cpu(es->s_inodes_count);
P
Peter Zijlstra 已提交
5129
	buf->f_ffree = percpu_counter_sum_positive(&sbi->s_freeinodes_counter);
5130
	buf->f_namelen = EXT4_NAME_LEN;
P
Pekka Enberg 已提交
5131 5132 5133 5134
	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;
5135

L
Li Xi 已提交
5136 5137 5138 5139 5140
#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
5141 5142 5143
	return 0;
}

5144 5145
/* Helper function for writing quotas on sync - we need to start transaction
 * before quota file is locked for write. Otherwise the are possible deadlocks:
5146
 * Process 1                         Process 2
5147
 * ext4_create()                     quota_sync()
5148
 *   jbd2_journal_start()                  write_dquot()
5149
 *   dquot_initialize()                         down(dqio_mutex)
5150
 *     down(dqio_mutex)                    jbd2_journal_start()
5151 5152 5153 5154 5155 5156 5157
 *
 */

#ifdef CONFIG_QUOTA

static inline struct inode *dquot_to_inode(struct dquot *dquot)
{
5158
	return sb_dqopt(dquot->dq_sb)->files[dquot->dq_id.type];
5159 5160
}

5161
static int ext4_write_dquot(struct dquot *dquot)
5162 5163 5164 5165 5166 5167
{
	int ret, err;
	handle_t *handle;
	struct inode *inode;

	inode = dquot_to_inode(dquot);
5168
	handle = ext4_journal_start(inode, EXT4_HT_QUOTA,
5169
				    EXT4_QUOTA_TRANS_BLOCKS(dquot->dq_sb));
5170 5171 5172
	if (IS_ERR(handle))
		return PTR_ERR(handle);
	ret = dquot_commit(dquot);
5173
	err = ext4_journal_stop(handle);
5174 5175 5176 5177 5178
	if (!ret)
		ret = err;
	return ret;
}

5179
static int ext4_acquire_dquot(struct dquot *dquot)
5180 5181 5182 5183
{
	int ret, err;
	handle_t *handle;

5184
	handle = ext4_journal_start(dquot_to_inode(dquot), EXT4_HT_QUOTA,
5185
				    EXT4_QUOTA_INIT_BLOCKS(dquot->dq_sb));
5186 5187 5188
	if (IS_ERR(handle))
		return PTR_ERR(handle);
	ret = dquot_acquire(dquot);
5189
	err = ext4_journal_stop(handle);
5190 5191 5192 5193 5194
	if (!ret)
		ret = err;
	return ret;
}

5195
static int ext4_release_dquot(struct dquot *dquot)
5196 5197 5198 5199
{
	int ret, err;
	handle_t *handle;

5200
	handle = ext4_journal_start(dquot_to_inode(dquot), EXT4_HT_QUOTA,
5201
				    EXT4_QUOTA_DEL_BLOCKS(dquot->dq_sb));
J
Jan Kara 已提交
5202 5203 5204
	if (IS_ERR(handle)) {
		/* Release dquot anyway to avoid endless cycle in dqput() */
		dquot_release(dquot);
5205
		return PTR_ERR(handle);
J
Jan Kara 已提交
5206
	}
5207
	ret = dquot_release(dquot);
5208
	err = ext4_journal_stop(handle);
5209 5210 5211 5212 5213
	if (!ret)
		ret = err;
	return ret;
}

5214
static int ext4_mark_dquot_dirty(struct dquot *dquot)
5215
{
5216 5217 5218
	struct super_block *sb = dquot->dq_sb;
	struct ext4_sb_info *sbi = EXT4_SB(sb);

5219
	/* Are we journaling quotas? */
5220
	if (ext4_has_feature_quota(sb) ||
5221
	    sbi->s_qf_names[USRQUOTA] || sbi->s_qf_names[GRPQUOTA]) {
5222
		dquot_mark_dquot_dirty(dquot);
5223
		return ext4_write_dquot(dquot);
5224 5225 5226 5227 5228
	} else {
		return dquot_mark_dquot_dirty(dquot);
	}
}

5229
static int ext4_write_info(struct super_block *sb, int type)
5230 5231 5232 5233 5234
{
	int ret, err;
	handle_t *handle;

	/* Data block + inode block */
5235
	handle = ext4_journal_start(d_inode(sb->s_root), EXT4_HT_QUOTA, 2);
5236 5237 5238
	if (IS_ERR(handle))
		return PTR_ERR(handle);
	ret = dquot_commit_info(sb, type);
5239
	err = ext4_journal_stop(handle);
5240 5241 5242 5243 5244 5245 5246 5247 5248
	if (!ret)
		ret = err;
	return ret;
}

/*
 * Turn on quotas during mount time - we need to find
 * the quota file and such...
 */
5249
static int ext4_quota_on_mount(struct super_block *sb, int type)
5250
{
5251 5252
	return dquot_quota_on_mount(sb, EXT4_SB(sb)->s_qf_names[type],
					EXT4_SB(sb)->s_jquota_fmt, type);
5253 5254
}

5255 5256 5257 5258 5259 5260 5261 5262 5263 5264 5265 5266 5267 5268
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);
}

5269 5270 5271
/*
 * Standard function to be called on quota_on
 */
5272
static int ext4_quota_on(struct super_block *sb, int type, int format_id,
5273
			 struct path *path)
5274 5275 5276 5277 5278
{
	int err;

	if (!test_opt(sb, QUOTA))
		return -EINVAL;
5279

5280
	/* Quotafile not on the same filesystem? */
5281
	if (path->dentry->d_sb != sb)
5282
		return -EXDEV;
5283 5284
	/* Journaling quota? */
	if (EXT4_SB(sb)->s_qf_names[type]) {
5285
		/* Quotafile not in fs root? */
5286
		if (path->dentry->d_parent != sb->s_root)
5287 5288 5289
			ext4_msg(sb, KERN_WARNING,
				"Quota file not on filesystem root. "
				"Journaled quota will not work");
5290
	}
5291 5292 5293 5294 5295

	/*
	 * When we journal data on quota file, we have to flush journal to see
	 * all updates to the file when we bypass pagecache...
	 */
5296
	if (EXT4_SB(sb)->s_journal &&
5297
	    ext4_should_journal_data(d_inode(path->dentry))) {
5298 5299 5300 5301 5302
		/*
		 * We don't need to lock updates but journal_flush() could
		 * otherwise be livelocked...
		 */
		jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal);
5303
		err = jbd2_journal_flush(EXT4_SB(sb)->s_journal);
5304
		jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal);
5305
		if (err)
5306
			return err;
5307
	}
5308 5309 5310 5311 5312 5313
	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;
5314 5315
}

5316 5317 5318 5319 5320
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
5321
	unsigned long qf_inums[EXT4_MAXQUOTAS] = {
5322
		le32_to_cpu(EXT4_SB(sb)->s_es->s_usr_quota_inum),
L
Li Xi 已提交
5323 5324
		le32_to_cpu(EXT4_SB(sb)->s_es->s_grp_quota_inum),
		le32_to_cpu(EXT4_SB(sb)->s_es->s_prj_quota_inum)
5325 5326
	};

5327
	BUG_ON(!ext4_has_feature_quota(sb));
5328 5329 5330 5331 5332 5333 5334 5335 5336 5337

	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
5338 5339
	/* Don't account quota for quota files to avoid recursion */
	qf_inode->i_flags |= S_NOQUOTA;
5340
	lockdep_set_quota_inode(qf_inode, I_DATA_SEM_QUOTA);
5341 5342
	err = dquot_enable(qf_inode, type, format_id, flags);
	iput(qf_inode);
5343 5344
	if (err)
		lockdep_set_quota_inode(qf_inode, I_DATA_SEM_NORMAL);
5345 5346 5347 5348 5349 5350 5351 5352

	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
5353
	unsigned long qf_inums[EXT4_MAXQUOTAS] = {
5354
		le32_to_cpu(EXT4_SB(sb)->s_es->s_usr_quota_inum),
L
Li Xi 已提交
5355 5356
		le32_to_cpu(EXT4_SB(sb)->s_es->s_grp_quota_inum),
		le32_to_cpu(EXT4_SB(sb)->s_es->s_prj_quota_inum)
5357
	};
5358 5359 5360 5361 5362
	bool quota_mopt[EXT4_MAXQUOTAS] = {
		test_opt(sb, USRQUOTA),
		test_opt(sb, GRPQUOTA),
		test_opt(sb, PRJQUOTA),
	};
5363 5364

	sb_dqopt(sb)->flags |= DQUOT_QUOTA_SYS_FILE;
J
Jan Kara 已提交
5365
	for (type = 0; type < EXT4_MAXQUOTAS; type++) {
5366 5367
		if (qf_inums[type]) {
			err = ext4_quota_enable(sb, type, QFMT_VFS_V1,
5368 5369
				DQUOT_USAGE_ENABLED |
				(quota_mopt[type] ? DQUOT_LIMITS_ENABLED : 0));
5370 5371
			if (err) {
				ext4_warning(sb,
5372 5373 5374
					"Failed to enable quota tracking "
					"(type=%d, err=%d). Please run "
					"e2fsck to fix.", type, err);
5375 5376 5377 5378 5379 5380 5381
				return err;
			}
		}
	}
	return 0;
}

5382 5383
static int ext4_quota_off(struct super_block *sb, int type)
{
5384 5385 5386
	struct inode *inode = sb_dqopt(sb)->files[type];
	handle_t *handle;

5387 5388 5389
	/* Force all delayed allocation blocks to be allocated.
	 * Caller already holds s_umount sem */
	if (test_opt(sb, DELALLOC))
5390 5391
		sync_filesystem(sb);

5392 5393 5394
	if (!inode)
		goto out;

5395 5396
	/* Update modification times of quota files when userspace can
	 * start looking at them */
5397
	handle = ext4_journal_start(inode, EXT4_HT_QUOTA, 1);
5398 5399
	if (IS_ERR(handle))
		goto out;
5400
	inode->i_mtime = inode->i_ctime = current_time(inode);
5401 5402 5403 5404
	ext4_mark_inode_dirty(handle, inode);
	ext4_journal_stop(handle);

out:
5405 5406 5407
	return dquot_quota_off(sb, type);
}

5408 5409
/* 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
5410
 * itself serializes the operations (and no one else should touch the files)
5411
 * we don't have to be afraid of races */
5412
static ssize_t ext4_quota_read(struct super_block *sb, int type, char *data,
5413 5414 5415
			       size_t len, loff_t off)
{
	struct inode *inode = sb_dqopt(sb)->files[type];
A
Aneesh Kumar K.V 已提交
5416
	ext4_lblk_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
5417 5418 5419 5420 5421 5422 5423 5424 5425 5426 5427 5428 5429 5430
	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;
5431 5432 5433
		bh = ext4_bread(NULL, inode, blk, 0);
		if (IS_ERR(bh))
			return PTR_ERR(bh);
5434 5435 5436 5437 5438 5439 5440 5441 5442 5443 5444 5445 5446 5447 5448
		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) */
5449
static ssize_t ext4_quota_write(struct super_block *sb, int type,
5450 5451 5452
				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
5453
	ext4_lblk_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
5454
	int err, offset = off & (sb->s_blocksize - 1);
5455
	int retries = 0;
5456 5457 5458
	struct buffer_head *bh;
	handle_t *handle = journal_current_handle();

5459
	if (EXT4_SB(sb)->s_journal && !handle) {
5460 5461
		ext4_msg(sb, KERN_WARNING, "Quota write (off=%llu, len=%llu)"
			" cancelled because transaction is not started",
J
Jan Kara 已提交
5462 5463 5464
			(unsigned long long)off, (unsigned long long)len);
		return -EIO;
	}
5465 5466 5467 5468 5469 5470 5471 5472 5473 5474 5475
	/*
	 * 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;
	}

5476 5477 5478 5479 5480 5481
	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));
5482 5483
	if (IS_ERR(bh))
		return PTR_ERR(bh);
5484 5485
	if (!bh)
		goto out;
5486
	BUFFER_TRACE(bh, "get write access");
5487 5488 5489
	err = ext4_journal_get_write_access(handle, bh);
	if (err) {
		brelse(bh);
5490
		return err;
5491
	}
5492 5493 5494 5495
	lock_buffer(bh);
	memcpy(bh->b_data+offset, data, len);
	flush_dcache_page(bh->b_page);
	unlock_buffer(bh);
5496
	err = ext4_handle_dirty_metadata(handle, NULL, bh);
5497
	brelse(bh);
5498
out:
5499 5500
	if (inode->i_size < off + len) {
		i_size_write(inode, off + len);
5501
		EXT4_I(inode)->i_disksize = inode->i_size;
5502
		ext4_mark_inode_dirty(handle, inode);
5503
	}
5504
	return len;
5505 5506
}

5507 5508 5509 5510 5511 5512 5513 5514 5515 5516 5517
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);
}
5518 5519
#endif

A
Al Viro 已提交
5520 5521
static struct dentry *ext4_mount(struct file_system_type *fs_type, int flags,
		       const char *dev_name, void *data)
5522
{
A
Al Viro 已提交
5523
	return mount_bdev(fs_type, flags, dev_name, data, ext4_fill_super);
5524 5525
}

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

static inline int ext2_feature_set_ok(struct super_block *sb)
{
5542
	if (ext4_has_unknown_ext2_incompat_features(sb))
5543 5544 5545
		return 0;
	if (sb->s_flags & MS_RDONLY)
		return 1;
5546
	if (ext4_has_unknown_ext2_ro_compat_features(sb))
5547 5548 5549
		return 0;
	return 1;
}
5550 5551 5552
#else
static inline void register_as_ext2(void) { }
static inline void unregister_as_ext2(void) { }
5553
static inline int ext2_feature_set_ok(struct super_block *sb) { return 0; }
5554 5555 5556 5557 5558 5559 5560 5561 5562 5563 5564 5565 5566 5567
#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);
}
5568 5569 5570

static inline int ext3_feature_set_ok(struct super_block *sb)
{
5571
	if (ext4_has_unknown_ext3_incompat_features(sb))
5572
		return 0;
5573
	if (!ext4_has_feature_journal(sb))
5574 5575 5576
		return 0;
	if (sb->s_flags & MS_RDONLY)
		return 1;
5577
	if (ext4_has_unknown_ext3_ro_compat_features(sb))
5578 5579 5580
		return 0;
	return 1;
}
5581

T
Theodore Ts'o 已提交
5582 5583 5584
static struct file_system_type ext4_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "ext4",
A
Al Viro 已提交
5585
	.mount		= ext4_mount,
T
Theodore Ts'o 已提交
5586 5587 5588
	.kill_sb	= kill_block_super,
	.fs_flags	= FS_REQUIRES_DEV,
};
5589
MODULE_ALIAS_FS("ext4");
T
Theodore Ts'o 已提交
5590

5591 5592 5593
/* Shared across all ext4 file systems */
wait_queue_head_t ext4__ioend_wq[EXT4_WQ_HASH_SZ];

5594
static int __init ext4_init_fs(void)
5595
{
5596
	int i, err;
5597

5598
	ratelimit_state_init(&ext4_mount_msg_ratelimit, 30 * HZ, 64);
5599 5600 5601
	ext4_li_info = NULL;
	mutex_init(&ext4_li_mtx);

5602
	/* Build-time check for flags consistency */
5603
	ext4_check_flag_values();
5604

5605
	for (i = 0; i < EXT4_WQ_HASH_SZ; i++)
5606 5607
		init_waitqueue_head(&ext4__ioend_wq[i]);

5608
	err = ext4_init_es();
5609 5610
	if (err)
		return err;
5611 5612 5613

	err = ext4_init_pageio();
	if (err)
5614
		goto out5;
5615

5616
	err = ext4_init_system_zone();
5617
	if (err)
5618
		goto out4;
5619

5620
	err = ext4_init_sysfs();
T
Theodore Ts'o 已提交
5621
	if (err)
5622
		goto out3;
5623

5624
	err = ext4_init_mballoc();
5625 5626
	if (err)
		goto out2;
5627 5628 5629
	err = init_inodecache();
	if (err)
		goto out1;
5630
	register_as_ext3();
5631
	register_as_ext2();
T
Theodore Ts'o 已提交
5632
	err = register_filesystem(&ext4_fs_type);
5633 5634
	if (err)
		goto out;
5635

5636 5637
	return 0;
out:
5638 5639
	unregister_as_ext2();
	unregister_as_ext3();
5640 5641
	destroy_inodecache();
out1:
5642
	ext4_exit_mballoc();
5643
out2:
5644 5645
	ext4_exit_sysfs();
out3:
5646
	ext4_exit_system_zone();
5647
out4:
5648
	ext4_exit_pageio();
5649
out5:
5650 5651
	ext4_exit_es();

5652 5653 5654
	return err;
}

5655
static void __exit ext4_exit_fs(void)
5656
{
5657
	ext4_destroy_lazyinit_thread();
5658 5659
	unregister_as_ext2();
	unregister_as_ext3();
T
Theodore Ts'o 已提交
5660
	unregister_filesystem(&ext4_fs_type);
5661
	destroy_inodecache();
5662
	ext4_exit_mballoc();
5663
	ext4_exit_sysfs();
5664 5665
	ext4_exit_system_zone();
	ext4_exit_pageio();
5666
	ext4_exit_es();
5667 5668 5669
}

MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others");
5670
MODULE_DESCRIPTION("Fourth Extended Filesystem");
5671
MODULE_LICENSE("GPL");
5672 5673
module_init(ext4_init_fs)
module_exit(ext4_exit_fs)