super.c 161.0 KB
Newer Older
1
/*
2
 *  linux/fs/ext4/super.c
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
 *
 * Copyright (C) 1992, 1993, 1994, 1995
 * Remy Card (card@masi.ibp.fr)
 * Laboratoire MASI - Institut Blaise Pascal
 * Universite Pierre et Marie Curie (Paris VI)
 *
 *  from
 *
 *  linux/fs/minix/inode.c
 *
 *  Copyright (C) 1991, 1992  Linus Torvalds
 *
 *  Big-endian to little-endian byte-swapping/bitmaps by
 *        David S. Miller (davem@caip.rutgers.edu), 1995
 */

#include <linux/module.h>
#include <linux/string.h>
#include <linux/fs.h>
#include <linux/time.h>
23
#include <linux/vmalloc.h>
24 25 26 27 28
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/blkdev.h>
#include <linux/parser.h>
#include <linux/buffer_head.h>
29
#include <linux/exportfs.h>
30 31 32 33 34 35
#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>
36
#include <linux/proc_fs.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
static struct proc_dir_entry *ext4_proc_root;
T
Theodore Ts'o 已提交
57
static struct kset *ext4_kset;
58 59 60
static struct ext4_lazy_init *ext4_li_info;
static struct mutex ext4_li_mtx;
static struct ext4_features *ext4_feat;
61
static int ext4_mballoc_ready;
62

63
static int ext4_load_journal(struct super_block *, struct ext4_super_block *,
64
			     unsigned long journal_devnum);
65
static int ext4_show_options(struct seq_file *seq, struct dentry *root);
66
static int ext4_commit_super(struct super_block *sb, int sync);
67 68 69 70
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);
71
static int ext4_sync_fs(struct super_block *sb, int wait);
72 73
static int ext4_remount(struct super_block *sb, int *flags, char *data);
static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf);
74 75
static int ext4_unfreeze(struct super_block *sb);
static int ext4_freeze(struct super_block *sb);
A
Al Viro 已提交
76 77
static struct dentry *ext4_mount(struct file_system_type *fs_type, int flags,
		       const char *dev_name, void *data);
78 79
static inline int ext2_feature_set_ok(struct super_block *sb);
static inline int ext3_feature_set_ok(struct super_block *sb);
80
static int ext4_feature_set_ok(struct super_block *sb, int readonly);
81 82
static void ext4_destroy_lazyinit_thread(void);
static void ext4_unregister_li_request(struct super_block *sb);
83
static void ext4_clear_request_list(void);
L
Lukas Czerner 已提交
84
static int ext4_reserve_clusters(struct ext4_sb_info *, ext4_fsblk_t);
85

86 87 88 89 90 91 92 93
#if !defined(CONFIG_EXT2_FS) && !defined(CONFIG_EXT2_FS_MODULE) && defined(CONFIG_EXT4_USE_FOR_EXT23)
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,
};
94
MODULE_ALIAS_FS("ext2");
95
MODULE_ALIAS("ext2");
96 97 98 99 100 101
#define IS_EXT2_SB(sb) ((sb)->s_bdev->bd_holder == &ext2_fs_type)
#else
#define IS_EXT2_SB(sb) (0)
#endif


102 103 104 105
#if !defined(CONFIG_EXT3_FS) && !defined(CONFIG_EXT3_FS_MODULE) && defined(CONFIG_EXT4_USE_FOR_EXT23)
static struct file_system_type ext3_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "ext3",
A
Al Viro 已提交
106
	.mount		= ext4_mount,
107 108 109
	.kill_sb	= kill_block_super,
	.fs_flags	= FS_REQUIRES_DEV,
};
110
MODULE_ALIAS_FS("ext3");
111
MODULE_ALIAS("ext3");
112 113 114 115
#define IS_EXT3_SB(sb) ((sb)->s_bdev->bd_holder == &ext3_fs_type)
#else
#define IS_EXT3_SB(sb) (0)
#endif
L
Laurent Vivier 已提交
116

117 118 119 120 121 122 123 124 125 126
static int ext4_verify_csum_type(struct super_block *sb,
				 struct ext4_super_block *es)
{
	if (!EXT4_HAS_RO_COMPAT_FEATURE(sb,
					EXT4_FEATURE_RO_COMPAT_METADATA_CSUM))
		return 1;

	return es->s_checksum_type == EXT4_CRC32C_CHKSUM;
}

127 128 129 130 131 132 133 134 135 136 137 138
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);
}

139 140
static int ext4_superblock_csum_verify(struct super_block *sb,
				       struct ext4_super_block *es)
141
{
142
	if (!ext4_has_metadata_csum(sb))
143 144 145 146 147
		return 1;

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

148
void ext4_superblock_csum_set(struct super_block *sb)
149
{
150 151
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;

152
	if (!ext4_has_metadata_csum(sb))
153 154 155 156 157
		return;

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

158 159 160 161
void *ext4_kvmalloc(size_t size, gfp_t flags)
{
	void *ret;

162
	ret = kmalloc(size, flags | __GFP_NOWARN);
163 164 165 166 167 168 169 170 171
	if (!ret)
		ret = __vmalloc(size, flags, PAGE_KERNEL);
	return ret;
}

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

172
	ret = kzalloc(size, flags | __GFP_NOWARN);
173 174 175 176 177
	if (!ret)
		ret = __vmalloc(size, flags | __GFP_ZERO, PAGE_KERNEL);
	return ret;
}

178 179
ext4_fsblk_t ext4_block_bitmap(struct super_block *sb,
			       struct ext4_group_desc *bg)
L
Laurent Vivier 已提交
180
{
181
	return le32_to_cpu(bg->bg_block_bitmap_lo) |
182
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
183
		 (ext4_fsblk_t)le32_to_cpu(bg->bg_block_bitmap_hi) << 32 : 0);
L
Laurent Vivier 已提交
184 185
}

186 187
ext4_fsblk_t ext4_inode_bitmap(struct super_block *sb,
			       struct ext4_group_desc *bg)
L
Laurent Vivier 已提交
188
{
189
	return le32_to_cpu(bg->bg_inode_bitmap_lo) |
190
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
191
		 (ext4_fsblk_t)le32_to_cpu(bg->bg_inode_bitmap_hi) << 32 : 0);
L
Laurent Vivier 已提交
192 193
}

194 195
ext4_fsblk_t ext4_inode_table(struct super_block *sb,
			      struct ext4_group_desc *bg)
L
Laurent Vivier 已提交
196
{
197
	return le32_to_cpu(bg->bg_inode_table_lo) |
198
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
199
		 (ext4_fsblk_t)le32_to_cpu(bg->bg_inode_table_hi) << 32 : 0);
L
Laurent Vivier 已提交
200 201
}

202 203
__u32 ext4_free_group_clusters(struct super_block *sb,
			       struct ext4_group_desc *bg)
204 205 206
{
	return le16_to_cpu(bg->bg_free_blocks_count_lo) |
		(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
207
		 (__u32)le16_to_cpu(bg->bg_free_blocks_count_hi) << 16 : 0);
208 209 210 211 212 213 214
}

__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 ?
215
		 (__u32)le16_to_cpu(bg->bg_free_inodes_count_hi) << 16 : 0);
216 217 218 219 220 221 222
}

__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 ?
223
		 (__u32)le16_to_cpu(bg->bg_used_dirs_count_hi) << 16 : 0);
224 225 226 227 228 229 230
}

__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 ?
231
		 (__u32)le16_to_cpu(bg->bg_itable_unused_hi) << 16 : 0);
232 233
}

234 235
void ext4_block_bitmap_set(struct super_block *sb,
			   struct ext4_group_desc *bg, ext4_fsblk_t blk)
L
Laurent Vivier 已提交
236
{
237
	bg->bg_block_bitmap_lo = cpu_to_le32((u32)blk);
238 239
	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
240 241
}

242 243
void ext4_inode_bitmap_set(struct super_block *sb,
			   struct ext4_group_desc *bg, ext4_fsblk_t blk)
L
Laurent Vivier 已提交
244
{
245
	bg->bg_inode_bitmap_lo  = cpu_to_le32((u32)blk);
246 247
	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
248 249
}

250 251
void ext4_inode_table_set(struct super_block *sb,
			  struct ext4_group_desc *bg, ext4_fsblk_t blk)
L
Laurent Vivier 已提交
252
{
253
	bg->bg_inode_table_lo = cpu_to_le32((u32)blk);
254 255
	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
256 257
}

258 259
void ext4_free_group_clusters_set(struct super_block *sb,
				  struct ext4_group_desc *bg, __u32 count)
260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289
{
	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);
}

290

291 292 293 294 295 296
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;
297 298
	if (bdev_read_only(sb->s_bdev))
		return;
299 300 301 302 303 304 305 306 307 308 309 310
	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;
	}
311 312 313 314 315 316
	/*
	 * 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);
317
	le32_add_cpu(&es->s_error_count, 1);
318 319 320 321 322 323 324 325 326
}

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);
}

B
Bobi Jam 已提交
327 328 329 330 331
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);
332
	struct ext4_journal_cb_entry	*jce;
B
Bobi Jam 已提交
333

334
	BUG_ON(txn->t_state == T_FINISHED);
B
Bobi Jam 已提交
335
	spin_lock(&sbi->s_md_lock);
336 337 338
	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
339 340 341 342 343 344 345
		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);
}
346

347 348 349 350
/* 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
351
 * superblock.  That is not possible on ext4, because we may have other
352 353 354 355 356
 * 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.
 *
357
 * We'll just use the jbd2_journal_abort() error code to record an error in
358
 * the journal instead.  On recovery, the journal will complain about
359 360 361
 * that error until we've noted it down and cleared it.
 */

362
static void ext4_handle_error(struct super_block *sb)
363 364 365 366
{
	if (sb->s_flags & MS_RDONLY)
		return;

367
	if (!test_opt(sb, ERRORS_CONT)) {
368
		journal_t *journal = EXT4_SB(sb)->s_journal;
369

370
		EXT4_SB(sb)->s_mount_flags |= EXT4_MF_FS_ABORTED;
371
		if (journal)
372
			jbd2_journal_abort(journal, -EIO);
373
	}
374
	if (test_opt(sb, ERRORS_RO)) {
375
		ext4_msg(sb, KERN_CRIT, "Remounting filesystem read-only");
376 377 378 379 380
		/*
		 * Make sure updated value of ->s_mount_flags will be visible
		 * before ->s_flags update
		 */
		smp_wmb();
381 382 383
		sb->s_flags |= MS_RDONLY;
	}
	if (test_opt(sb, ERRORS_PANIC))
384
		panic("EXT4-fs (device %s): panic forced after error\n",
385 386 387
			sb->s_id);
}

388 389 390 391
#define ext4_error_ratelimit(sb)					\
		___ratelimit(&(EXT4_SB(sb)->s_err_ratelimit_state),	\
			     "EXT4-fs error")

392
void __ext4_error(struct super_block *sb, const char *function,
393
		  unsigned int line, const char *fmt, ...)
394
{
J
Joe Perches 已提交
395
	struct va_format vaf;
396 397
	va_list args;

398 399 400 401 402 403 404 405 406
	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);
	}
407
	save_error_info(sb, function, line);
408
	ext4_handle_error(sb);
409 410
}

411 412 413
void __ext4_error_inode(struct inode *inode, const char *function,
			unsigned int line, ext4_fsblk_t block,
			const char *fmt, ...)
414 415
{
	va_list args;
416
	struct va_format vaf;
417
	struct ext4_super_block *es = EXT4_SB(inode->i_sb)->s_es;
418

419 420
	es->s_last_error_ino = cpu_to_le32(inode->i_ino);
	es->s_last_error_block = cpu_to_le64(block);
421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436
	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);
	}
437
	save_error_info(inode->i_sb, function, line);
438 439 440
	ext4_handle_error(inode->i_sb);
}

441 442 443
void __ext4_error_file(struct file *file, const char *function,
		       unsigned int line, ext4_fsblk_t block,
		       const char *fmt, ...)
444 445
{
	va_list args;
446
	struct va_format vaf;
447
	struct ext4_super_block *es;
A
Al Viro 已提交
448
	struct inode *inode = file_inode(file);
449 450
	char pathname[80], *path;

451 452
	es = EXT4_SB(inode->i_sb)->s_es;
	es->s_last_error_ino = cpu_to_le32(inode->i_ino);
453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473
	if (ext4_error_ratelimit(inode->i_sb)) {
		path = d_path(&(file->f_path), pathname, sizeof(pathname));
		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);
	}
474
	save_error_info(inode->i_sb, function, line);
475 476 477
	ext4_handle_error(inode->i_sb);
}

478 479
const char *ext4_decode_error(struct super_block *sb, int errno,
			      char nbuf[16])
480 481 482 483 484 485 486 487 488 489 490
{
	char *errstr = NULL;

	switch (errno) {
	case -EIO:
		errstr = "IO failure";
		break;
	case -ENOMEM:
		errstr = "Out of memory";
		break;
	case -EROFS:
491 492
		if (!sb || (EXT4_SB(sb)->s_journal &&
			    EXT4_SB(sb)->s_journal->j_flags & JBD2_ABORT))
493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511
			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;
}

512
/* __ext4_std_error decodes expected errors from journaling functions
513 514
 * automatically and invokes the appropriate error response.  */

515 516
void __ext4_std_error(struct super_block *sb, const char *function,
		      unsigned int line, int errno)
517 518 519 520 521 522 523 524 525 526 527
{
	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;

528 529 530 531 532
	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);
	}
533

534
	save_error_info(sb, function, line);
535
	ext4_handle_error(sb);
536 537 538
}

/*
539
 * ext4_abort is a much stronger failure handler than ext4_error.  The
540 541 542 543 544 545 546 547
 * 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.
 */

548
void __ext4_abort(struct super_block *sb, const char *function,
549
		unsigned int line, const char *fmt, ...)
550 551 552
{
	va_list args;

553
	save_error_info(sb, function, line);
554
	va_start(args, fmt);
555 556
	printk(KERN_CRIT "EXT4-fs error (device %s): %s:%d: ", sb->s_id,
	       function, line);
557 558 559 560
	vprintk(fmt, args);
	printk("\n");
	va_end(args);

561 562 563
	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;
564 565 566 567 568 569
		/*
		 * Make sure updated value of ->s_mount_flags will be visible
		 * before ->s_flags update
		 */
		smp_wmb();
		sb->s_flags |= MS_RDONLY;
570 571 572 573
		if (EXT4_SB(sb)->s_journal)
			jbd2_journal_abort(EXT4_SB(sb)->s_journal, -EIO);
		save_error_info(sb, function, line);
	}
574
	if (test_opt(sb, ERRORS_PANIC))
575
		panic("EXT4-fs panic from previous error\n");
576 577
}

578 579
void __ext4_msg(struct super_block *sb,
		const char *prefix, const char *fmt, ...)
580
{
J
Joe Perches 已提交
581
	struct va_format vaf;
582 583
	va_list args;

584 585 586
	if (!___ratelimit(&(EXT4_SB(sb)->s_msg_ratelimit_state), "EXT4-fs"))
		return;

587
	va_start(args, fmt);
J
Joe Perches 已提交
588 589 590
	vaf.fmt = fmt;
	vaf.va = &args;
	printk("%sEXT4-fs (%s): %pV\n", prefix, sb->s_id, &vaf);
591 592 593
	va_end(args);
}

594 595 596 597
#define ext4_warning_ratelimit(sb)					\
		___ratelimit(&(EXT4_SB(sb)->s_warning_ratelimit_state),	\
			     "EXT4-fs warning")

598
void __ext4_warning(struct super_block *sb, const char *function,
599
		    unsigned int line, const char *fmt, ...)
600
{
J
Joe Perches 已提交
601
	struct va_format vaf;
602 603
	va_list args;

604
	if (!ext4_warning_ratelimit(sb))
605 606
		return;

607
	va_start(args, fmt);
J
Joe Perches 已提交
608 609 610 611
	vaf.fmt = fmt;
	vaf.va = &args;
	printk(KERN_WARNING "EXT4-fs warning (device %s): %s:%d: %pV\n",
	       sb->s_id, function, line, &vaf);
612 613 614
	va_end(args);
}

615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632
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);
}

633 634 635 636
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, ...)
637 638 639
__releases(bitlock)
__acquires(bitlock)
{
J
Joe Perches 已提交
640
	struct va_format vaf;
641 642 643
	va_list args;
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;

644 645 646
	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
647

648 649 650 651 652 653 654 655 656 657 658 659 660 661
	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);
	}
662 663

	if (test_opt(sb, ERRORS_CONT)) {
664
		ext4_commit_super(sb, 0);
665 666
		return;
	}
667

668 669 670 671 672 673 674 675
	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
676
	 * ext4_grp_locked_error() to distinguish between the
677 678 679 680 681 682 683 684
	 * 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;
}

685
void ext4_update_dynamic_rev(struct super_block *sb)
686
{
687
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;
688

689
	if (le32_to_cpu(es->s_rev_level) > EXT4_GOOD_OLD_REV)
690 691
		return;

692
	ext4_warning(sb,
693 694
		     "updating to rev %d because of new feature flag, "
		     "running e2fsck is recommended",
695
		     EXT4_DYNAMIC_REV);
696

697 698 699
	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);
700 701 702 703 704 705 706 707 708 709 710 711 712
	/* 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
 */
713
static struct block_device *ext4_blkdev_get(dev_t dev, struct super_block *sb)
714 715 716 717
{
	struct block_device *bdev;
	char b[BDEVNAME_SIZE];

718
	bdev = blkdev_get_by_dev(dev, FMODE_READ|FMODE_WRITE|FMODE_EXCL, sb);
719 720 721 722 723
	if (IS_ERR(bdev))
		goto fail;
	return bdev;

fail:
724
	ext4_msg(sb, KERN_ERR, "failed to open journal device %s: %ld",
725 726 727 728 729 730 731
			__bdevname(dev, b), PTR_ERR(bdev));
	return NULL;
}

/*
 * Release the journal device
 */
A
Al Viro 已提交
732
static void ext4_blkdev_put(struct block_device *bdev)
733
{
A
Al Viro 已提交
734
	blkdev_put(bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL);
735 736
}

A
Al Viro 已提交
737
static void ext4_blkdev_remove(struct ext4_sb_info *sbi)
738 739 740 741
{
	struct block_device *bdev;
	bdev = sbi->journal_bdev;
	if (bdev) {
A
Al Viro 已提交
742
		ext4_blkdev_put(bdev);
743 744 745 746 747 748
		sbi->journal_bdev = NULL;
	}
}

static inline struct inode *orphan_list_entry(struct list_head *l)
{
749
	return &list_entry(l, struct ext4_inode_info, i_orphan)->vfs_inode;
750 751
}

752
static void dump_orphan_list(struct super_block *sb, struct ext4_sb_info *sbi)
753 754 755
{
	struct list_head *l;

756 757
	ext4_msg(sb, KERN_ERR, "sb orphan head is %d",
		 le32_to_cpu(sbi->s_es->s_last_orphan));
758 759 760 761 762 763 764 765 766 767 768 769

	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));
	}
}

770
static void ext4_put_super(struct super_block *sb)
771
{
772 773
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_super_block *es = sbi->s_es;
774
	int i, err;
775

776
	ext4_unregister_li_request(sb);
777 778
	dquot_disable(sb, -1, DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED);

779 780
	flush_workqueue(sbi->rsv_conversion_wq);
	destroy_workqueue(sbi->rsv_conversion_wq);
781

782 783 784 785
	if (sbi->s_journal) {
		err = jbd2_journal_destroy(sbi->s_journal);
		sbi->s_journal = NULL;
		if (err < 0)
786
			ext4_abort(sb, "Couldn't clean up the journal");
787
	}
788

789
	ext4_es_unregister_shrinker(sbi);
790
	del_timer_sync(&sbi->s_err_report);
791 792 793 794 795
	ext4_release_system_zone(sb);
	ext4_mb_release(sb);
	ext4_ext_release(sb);
	ext4_xattr_put_super(sb);

796
	if (!(sb->s_flags & MS_RDONLY)) {
797
		EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
798 799
		es->s_state = cpu_to_le16(sbi->s_mount_state);
	}
800
	if (!(sb->s_flags & MS_RDONLY))
801 802
		ext4_commit_super(sb, 1);

803
	if (sbi->s_proc) {
804
		remove_proc_entry("options", sbi->s_proc);
805
		remove_proc_entry(sb->s_id, ext4_proc_root);
806
	}
T
Theodore Ts'o 已提交
807
	kobject_del(&sbi->s_kobj);
808 809 810

	for (i = 0; i < sbi->s_gdb_count; i++)
		brelse(sbi->s_group_desc[i]);
A
Al Viro 已提交
811 812
	kvfree(sbi->s_group_desc);
	kvfree(sbi->s_flex_groups);
813
	percpu_counter_destroy(&sbi->s_freeclusters_counter);
814 815
	percpu_counter_destroy(&sbi->s_freeinodes_counter);
	percpu_counter_destroy(&sbi->s_dirs_counter);
816
	percpu_counter_destroy(&sbi->s_dirtyclusters_counter);
817 818
	brelse(sbi->s_sbh);
#ifdef CONFIG_QUOTA
J
Jan Kara 已提交
819
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
820 821 822 823 824 825 826 827 828 829 830
		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));

831
	invalidate_bdev(sb->s_bdev);
832 833 834 835 836 837 838
	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);
839
		invalidate_bdev(sbi->journal_bdev);
840
		ext4_blkdev_remove(sbi);
841
	}
842 843 844 845
	if (sbi->s_mb_cache) {
		ext4_xattr_destroy_cache(sbi->s_mb_cache);
		sbi->s_mb_cache = NULL;
	}
846 847
	if (sbi->s_mmp_tsk)
		kthread_stop(sbi->s_mmp_tsk);
848
	sb->s_fs_info = NULL;
T
Theodore Ts'o 已提交
849 850 851 852 853 854
	/*
	 * 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);
855 856
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
857
	kfree(sbi->s_blockgroup_lock);
858 859 860
	kfree(sbi);
}

861
static struct kmem_cache *ext4_inode_cachep;
862 863 864 865

/*
 * Called inside transaction, so use GFP_NOFS
 */
866
static struct inode *ext4_alloc_inode(struct super_block *sb)
867
{
868
	struct ext4_inode_info *ei;
869

C
Christoph Lameter 已提交
870
	ei = kmem_cache_alloc(ext4_inode_cachep, GFP_NOFS);
871 872
	if (!ei)
		return NULL;
873

874
	ei->vfs_inode.i_version = 1;
875
	spin_lock_init(&ei->i_raw_lock);
876 877
	INIT_LIST_HEAD(&ei->i_prealloc_list);
	spin_lock_init(&ei->i_prealloc_lock);
Z
Zheng Liu 已提交
878 879
	ext4_es_init_tree(&ei->i_es_tree);
	rwlock_init(&ei->i_es_lock);
880
	INIT_LIST_HEAD(&ei->i_es_list);
881
	ei->i_es_all_nr = 0;
882
	ei->i_es_shk_nr = 0;
883
	ei->i_es_shrink_lblk = 0;
884 885 886
	ei->i_reserved_data_blocks = 0;
	ei->i_reserved_meta_blocks = 0;
	ei->i_allocated_meta_blocks = 0;
887
	ei->i_da_metadata_calc_len = 0;
888
	ei->i_da_metadata_calc_last_lblock = 0;
889
	spin_lock_init(&(ei->i_block_reservation_lock));
890 891
#ifdef CONFIG_QUOTA
	ei->i_reserved_quota = 0;
J
Jan Kara 已提交
892
	memset(&ei->i_dquot, 0, sizeof(ei->i_dquot));
893
#endif
894
	ei->jinode = NULL;
895
	INIT_LIST_HEAD(&ei->i_rsv_conversion_list);
896
	spin_lock_init(&ei->i_completed_io_lock);
897 898
	ei->i_sync_tid = 0;
	ei->i_datasync_tid = 0;
899
	atomic_set(&ei->i_ioend_count, 0);
900
	atomic_set(&ei->i_unwritten, 0);
901
	INIT_WORK(&ei->i_rsv_conversion_work, ext4_end_io_rsv_work);
902
#ifdef CONFIG_EXT4_FS_ENCRYPTION
903
	ei->i_crypt_info = NULL;
904
#endif
905 906 907
	return &ei->vfs_inode;
}

908 909 910 911 912 913 914 915
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
916 917 918 919 920 921
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));
}

922
static void ext4_destroy_inode(struct inode *inode)
923
{
924
	if (!list_empty(&(EXT4_I(inode)->i_orphan))) {
925 926 927
		ext4_msg(inode->i_sb, KERN_ERR,
			 "Inode %lu (%p): orphan list check failed!",
			 inode->i_ino, EXT4_I(inode));
928 929 930 931 932
		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
933
	call_rcu(&inode->i_rcu, ext4_i_callback);
934 935
}

936
static void init_once(void *foo)
937
{
938
	struct ext4_inode_info *ei = (struct ext4_inode_info *) foo;
939

C
Christoph Lameter 已提交
940 941
	INIT_LIST_HEAD(&ei->i_orphan);
	init_rwsem(&ei->xattr_sem);
942
	init_rwsem(&ei->i_data_sem);
C
Christoph Lameter 已提交
943
	inode_init_once(&ei->vfs_inode);
944 945
}

946
static int __init init_inodecache(void)
947
{
948 949
	ext4_inode_cachep = kmem_cache_create("ext4_inode_cache",
					     sizeof(struct ext4_inode_info),
950 951
					     0, (SLAB_RECLAIM_ACCOUNT|
						SLAB_MEM_SPREAD),
952
					     init_once);
953
	if (ext4_inode_cachep == NULL)
954 955 956 957 958 959
		return -ENOMEM;
	return 0;
}

static void destroy_inodecache(void)
{
960 961 962 963 964
	/*
	 * Make sure all delayed rcu free inodes are flushed before we
	 * destroy cache.
	 */
	rcu_barrier();
965
	kmem_cache_destroy(ext4_inode_cachep);
966 967
}

A
Al Viro 已提交
968
void ext4_clear_inode(struct inode *inode)
969
{
A
Al Viro 已提交
970
	invalidate_inode_buffers(inode);
971
	clear_inode(inode);
972
	dquot_drop(inode);
973
	ext4_discard_preallocations(inode);
974
	ext4_es_remove_extent(inode, 0, EXT_MAX_BLOCKS);
975 976 977 978 979 980
	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;
	}
981 982
#ifdef CONFIG_EXT4_FS_ENCRYPTION
	if (EXT4_I(inode)->i_crypt_info)
983
		ext4_free_encryption_info(inode, EXT4_I(inode)->i_crypt_info);
984
#endif
985 986
}

C
Christoph Hellwig 已提交
987
static struct inode *ext4_nfs_get_inode(struct super_block *sb,
988
					u64 ino, u32 generation)
989 990 991
{
	struct inode *inode;

992
	if (ino < EXT4_FIRST_INO(sb) && ino != EXT4_ROOT_INO)
993
		return ERR_PTR(-ESTALE);
994
	if (ino > le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count))
995 996 997 998
		return ERR_PTR(-ESTALE);

	/* iget isn't really right if the inode is currently unallocated!!
	 *
999
	 * ext4_read_inode will return a bad_inode if the inode had been
1000 1001 1002 1003 1004
	 * deleted, so we should be safe.
	 *
	 * Currently we don't know the generation for parent directory, so
	 * a generation of 0 means "accept any"
	 */
1005
	inode = ext4_iget_normal(sb, ino);
1006 1007 1008
	if (IS_ERR(inode))
		return ERR_CAST(inode);
	if (generation && inode->i_generation != generation) {
1009 1010 1011
		iput(inode);
		return ERR_PTR(-ESTALE);
	}
C
Christoph Hellwig 已提交
1012 1013 1014 1015 1016

	return inode;
}

static struct dentry *ext4_fh_to_dentry(struct super_block *sb, struct fid *fid,
1017
					int fh_len, int fh_type)
C
Christoph Hellwig 已提交
1018 1019 1020 1021 1022 1023
{
	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,
1024
					int fh_len, int fh_type)
C
Christoph Hellwig 已提交
1025 1026 1027
{
	return generic_fh_to_parent(sb, fid, fh_len, fh_type,
				    ext4_nfs_get_inode);
1028 1029
}

1030 1031 1032 1033 1034 1035
/*
 * 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.
 */
1036 1037
static int bdev_try_to_free_page(struct super_block *sb, struct page *page,
				 gfp_t wait)
1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049
{
	journal_t *journal = EXT4_SB(sb)->s_journal;

	WARN_ON(PageChecked(page));
	if (!page_has_buffers(page))
		return 0;
	if (journal)
		return jbd2_journal_try_to_free_buffers(journal, page,
							wait & ~__GFP_WAIT);
	return try_to_free_buffers(page);
}

1050
#ifdef CONFIG_QUOTA
1051
#define QTYPE2NAME(t) ((t) == USRQUOTA ? "user" : "group")
1052
#define QTYPE2MOPT(on, t) ((t) == USRQUOTA?((on)##USRJQUOTA):((on)##GRPJQUOTA))
1053

1054 1055 1056 1057 1058
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);
1059
static int ext4_quota_on(struct super_block *sb, int type, int format_id,
1060
			 struct path *path);
1061
static int ext4_quota_off(struct super_block *sb, int type);
1062 1063
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,
1064
			       size_t len, loff_t off);
1065
static ssize_t ext4_quota_write(struct super_block *sb, int type,
1066
				const char *data, size_t len, loff_t off);
1067 1068 1069
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);
1070

J
Jan Kara 已提交
1071 1072 1073 1074 1075
static struct dquot **ext4_get_dquots(struct inode *inode)
{
	return EXT4_I(inode)->i_dquot;
}

1076
static const struct dquot_operations ext4_quota_operations = {
1077
	.get_reserved_space = ext4_get_reserved_space,
1078 1079 1080 1081
	.write_dquot	= ext4_write_dquot,
	.acquire_dquot	= ext4_acquire_dquot,
	.release_dquot	= ext4_release_dquot,
	.mark_dirty	= ext4_mark_dquot_dirty,
1082 1083 1084
	.write_info	= ext4_write_info,
	.alloc_dquot	= dquot_alloc,
	.destroy_dquot	= dquot_destroy,
1085 1086
};

1087
static const struct quotactl_ops ext4_qctl_operations = {
1088
	.quota_on	= ext4_quota_on,
1089
	.quota_off	= ext4_quota_off,
1090
	.quota_sync	= dquot_quota_sync,
1091
	.get_state	= dquot_get_state,
1092 1093 1094
	.set_info	= dquot_set_dqinfo,
	.get_dqblk	= dquot_get_dqblk,
	.set_dqblk	= dquot_set_dqblk
1095 1096 1097
};
#endif

1098
static const struct super_operations ext4_sops = {
1099 1100 1101 1102
	.alloc_inode	= ext4_alloc_inode,
	.destroy_inode	= ext4_destroy_inode,
	.write_inode	= ext4_write_inode,
	.dirty_inode	= ext4_dirty_inode,
1103
	.drop_inode	= ext4_drop_inode,
A
Al Viro 已提交
1104
	.evict_inode	= ext4_evict_inode,
1105 1106
	.put_super	= ext4_put_super,
	.sync_fs	= ext4_sync_fs,
1107 1108
	.freeze_fs	= ext4_freeze,
	.unfreeze_fs	= ext4_unfreeze,
1109 1110 1111
	.statfs		= ext4_statfs,
	.remount_fs	= ext4_remount,
	.show_options	= ext4_show_options,
1112
#ifdef CONFIG_QUOTA
1113 1114
	.quota_read	= ext4_quota_read,
	.quota_write	= ext4_quota_write,
J
Jan Kara 已提交
1115
	.get_dquots	= ext4_get_dquots,
1116
#endif
1117
	.bdev_try_to_free_page = bdev_try_to_free_page,
1118 1119
};

1120
static const struct export_operations ext4_export_ops = {
C
Christoph Hellwig 已提交
1121 1122
	.fh_to_dentry = ext4_fh_to_dentry,
	.fh_to_parent = ext4_fh_to_parent,
1123
	.get_parent = ext4_get_parent,
1124 1125 1126 1127 1128
};

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,
1129
	Opt_nouid32, Opt_debug, Opt_removed,
1130
	Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl,
1131
	Opt_auto_da_alloc, Opt_noauto_da_alloc, Opt_noload,
1132 1133
	Opt_commit, Opt_min_batch_time, Opt_max_batch_time, Opt_journal_dev,
	Opt_journal_path, Opt_journal_checksum, Opt_journal_async_commit,
1134
	Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback,
1135
	Opt_data_err_abort, Opt_data_err_ignore, Opt_test_dummy_encryption,
1136
	Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota,
J
Jan Kara 已提交
1137
	Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_jqfmt_vfsv1, Opt_quota,
T
Theodore Ts'o 已提交
1138
	Opt_noquota, Opt_barrier, Opt_nobarrier, Opt_err,
R
Ross Zwisler 已提交
1139
	Opt_usrquota, Opt_grpquota, Opt_i_version, Opt_dax,
1140
	Opt_stripe, Opt_delalloc, Opt_nodelalloc, Opt_mblk_io_submit,
1141
	Opt_lazytime, Opt_nolazytime,
1142
	Opt_nomblk_io_submit, Opt_block_validity, Opt_noblock_validity,
1143
	Opt_inode_readahead_blks, Opt_journal_ioprio,
1144
	Opt_dioread_nolock, Opt_dioread_lock,
1145
	Opt_discard, Opt_nodiscard, Opt_init_itable, Opt_noinit_itable,
1146
	Opt_max_dir_size_kb, Opt_nojournal_checksum,
1147 1148
};

1149
static const match_table_t tokens = {
1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163
	{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"},
1164 1165
	{Opt_removed, "oldalloc"},
	{Opt_removed, "orlov"},
1166 1167 1168 1169
	{Opt_user_xattr, "user_xattr"},
	{Opt_nouser_xattr, "nouser_xattr"},
	{Opt_acl, "acl"},
	{Opt_noacl, "noacl"},
1170
	{Opt_noload, "norecovery"},
1171
	{Opt_noload, "noload"},
1172 1173
	{Opt_removed, "nobh"},
	{Opt_removed, "bh"},
1174
	{Opt_commit, "commit=%u"},
1175 1176
	{Opt_min_batch_time, "min_batch_time=%u"},
	{Opt_max_batch_time, "max_batch_time=%u"},
1177
	{Opt_journal_dev, "journal_dev=%u"},
1178
	{Opt_journal_path, "journal_path=%s"},
1179
	{Opt_journal_checksum, "journal_checksum"},
1180
	{Opt_nojournal_checksum, "nojournal_checksum"},
1181
	{Opt_journal_async_commit, "journal_async_commit"},
1182 1183 1184 1185
	{Opt_abort, "abort"},
	{Opt_data_journal, "data=journal"},
	{Opt_data_ordered, "data=ordered"},
	{Opt_data_writeback, "data=writeback"},
1186 1187
	{Opt_data_err_abort, "data_err=abort"},
	{Opt_data_err_ignore, "data_err=ignore"},
1188 1189 1190 1191 1192 1193
	{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
1194
	{Opt_jqfmt_vfsv1, "jqfmt=vfsv1"},
1195 1196 1197 1198 1199
	{Opt_grpquota, "grpquota"},
	{Opt_noquota, "noquota"},
	{Opt_quota, "quota"},
	{Opt_usrquota, "usrquota"},
	{Opt_barrier, "barrier=%u"},
T
Theodore Ts'o 已提交
1200 1201
	{Opt_barrier, "barrier"},
	{Opt_nobarrier, "nobarrier"},
1202
	{Opt_i_version, "i_version"},
R
Ross Zwisler 已提交
1203
	{Opt_dax, "dax"},
1204
	{Opt_stripe, "stripe=%u"},
1205
	{Opt_delalloc, "delalloc"},
1206 1207
	{Opt_lazytime, "lazytime"},
	{Opt_nolazytime, "nolazytime"},
1208
	{Opt_nodelalloc, "nodelalloc"},
1209 1210
	{Opt_removed, "mblk_io_submit"},
	{Opt_removed, "nomblk_io_submit"},
1211 1212
	{Opt_block_validity, "block_validity"},
	{Opt_noblock_validity, "noblock_validity"},
1213
	{Opt_inode_readahead_blks, "inode_readahead_blks=%u"},
1214
	{Opt_journal_ioprio, "journal_ioprio=%u"},
1215
	{Opt_auto_da_alloc, "auto_da_alloc=%u"},
T
Theodore Ts'o 已提交
1216 1217
	{Opt_auto_da_alloc, "auto_da_alloc"},
	{Opt_noauto_da_alloc, "noauto_da_alloc"},
1218 1219
	{Opt_dioread_nolock, "dioread_nolock"},
	{Opt_dioread_lock, "dioread_lock"},
1220 1221
	{Opt_discard, "discard"},
	{Opt_nodiscard, "nodiscard"},
1222 1223 1224
	{Opt_init_itable, "init_itable=%u"},
	{Opt_init_itable, "init_itable"},
	{Opt_noinit_itable, "noinit_itable"},
1225
	{Opt_max_dir_size_kb, "max_dir_size_kb=%u"},
1226
	{Opt_test_dummy_encryption, "test_dummy_encryption"},
1227 1228 1229 1230 1231
	{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
1232
	{Opt_err, NULL},
1233 1234
};

1235
static ext4_fsblk_t get_sb_block(void **data)
1236
{
1237
	ext4_fsblk_t	sb_block;
1238 1239 1240 1241
	char		*options = (char *) *data;

	if (!options || strncmp(options, "sb=", 3) != 0)
		return 1;	/* Default location */
1242

1243
	options += 3;
1244
	/* TODO: use simple_strtoll with >32bit ext4 */
1245 1246
	sb_block = simple_strtoul(options, &options, 0);
	if (*options && *options != ',') {
1247
		printk(KERN_ERR "EXT4-fs: Invalid sb specification: %s\n",
1248 1249 1250 1251 1252 1253
		       (char *) *data);
		return 1;
	}
	if (*options == ',')
		options++;
	*data = (void *) options;
1254

1255 1256 1257
	return sb_block;
}

1258
#define DEFAULT_JOURNAL_IOPRIO (IOPRIO_PRIO_VALUE(IOPRIO_CLASS_BE, 3))
1259 1260
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";
1261

D
Dmitry Monakhov 已提交
1262 1263 1264 1265 1266
#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;
1267
	int ret = -1;
D
Dmitry Monakhov 已提交
1268 1269 1270 1271 1272 1273

	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");
1274
		return -1;
D
Dmitry Monakhov 已提交
1275
	}
1276 1277 1278 1279 1280
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_QUOTA)) {
		ext4_msg(sb, KERN_ERR, "Cannot set journaled quota options "
			 "when QUOTA feature is enabled");
		return -1;
	}
D
Dmitry Monakhov 已提交
1281 1282 1283 1284
	qname = match_strdup(args);
	if (!qname) {
		ext4_msg(sb, KERN_ERR,
			"Not enough memory for storing quotafile name");
1285
		return -1;
D
Dmitry Monakhov 已提交
1286
	}
1287 1288 1289 1290 1291 1292 1293 1294
	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
1295
	}
1296
	if (strchr(qname, '/')) {
D
Dmitry Monakhov 已提交
1297 1298
		ext4_msg(sb, KERN_ERR,
			"quotafile must be on filesystem root");
1299
		goto errout;
D
Dmitry Monakhov 已提交
1300
	}
1301
	sbi->s_qf_names[qtype] = qname;
1302
	set_opt(sb, QUOTA);
D
Dmitry Monakhov 已提交
1303
	return 1;
1304 1305 1306
errout:
	kfree(qname);
	return ret;
D
Dmitry Monakhov 已提交
1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317
}

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");
1318
		return -1;
D
Dmitry Monakhov 已提交
1319
	}
1320
	kfree(sbi->s_qf_names[qtype]);
D
Dmitry Monakhov 已提交
1321 1322 1323 1324 1325
	sbi->s_qf_names[qtype] = NULL;
	return 1;
}
#endif

1326 1327 1328 1329 1330 1331
#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
1332
#ifdef CONFIG_QUOTA
1333 1334 1335 1336 1337
#define MOPT_Q		0
#define MOPT_QFMT	0x0040
#else
#define MOPT_Q		MOPT_NOSUPPORT
#define MOPT_QFMT	MOPT_NOSUPPORT
1338
#endif
1339
#define MOPT_DATAJ	0x0080
1340 1341 1342
#define MOPT_NO_EXT2	0x0100
#define MOPT_NO_EXT3	0x0200
#define MOPT_EXT4_ONLY	(MOPT_NO_EXT2 | MOPT_NO_EXT3)
1343
#define MOPT_STRING	0x0400
1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355

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},
1356 1357 1358 1359
	{Opt_dioread_nolock, EXT4_MOUNT_DIOREAD_NOLOCK,
	 MOPT_EXT4_ONLY | MOPT_SET},
	{Opt_dioread_lock, EXT4_MOUNT_DIOREAD_NOLOCK,
	 MOPT_EXT4_ONLY | MOPT_CLEAR},
1360 1361
	{Opt_discard, EXT4_MOUNT_DISCARD, MOPT_SET},
	{Opt_nodiscard, EXT4_MOUNT_DISCARD, MOPT_CLEAR},
1362 1363 1364
	{Opt_delalloc, EXT4_MOUNT_DELALLOC,
	 MOPT_EXT4_ONLY | MOPT_SET | MOPT_EXPLICIT},
	{Opt_nodelalloc, EXT4_MOUNT_DELALLOC,
1365
	 MOPT_EXT4_ONLY | MOPT_CLEAR},
1366 1367
	{Opt_nojournal_checksum, EXT4_MOUNT_JOURNAL_CHECKSUM,
	 MOPT_EXT4_ONLY | MOPT_CLEAR},
1368 1369
	{Opt_journal_checksum, EXT4_MOUNT_JOURNAL_CHECKSUM,
	 MOPT_EXT4_ONLY | MOPT_SET},
1370
	{Opt_journal_async_commit, (EXT4_MOUNT_JOURNAL_ASYNC_COMMIT |
1371 1372 1373
				    EXT4_MOUNT_JOURNAL_CHECKSUM),
	 MOPT_EXT4_ONLY | MOPT_SET},
	{Opt_noload, EXT4_MOUNT_NOLOAD, MOPT_NO_EXT2 | MOPT_SET},
1374 1375 1376
	{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},
1377 1378 1379 1380
	{Opt_data_err_abort, EXT4_MOUNT_DATA_ERR_ABORT,
	 MOPT_NO_EXT2 | MOPT_SET},
	{Opt_data_err_ignore, EXT4_MOUNT_DATA_ERR_ABORT,
	 MOPT_NO_EXT2 | MOPT_CLEAR},
1381 1382 1383 1384 1385 1386 1387 1388 1389 1390
	{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
1391
	{Opt_dax, EXT4_MOUNT_DAX, MOPT_SET},
1392
	{Opt_stripe, 0, MOPT_GTE0},
1393 1394 1395
	{Opt_resuid, 0, MOPT_GTE0},
	{Opt_resgid, 0, MOPT_GTE0},
	{Opt_journal_dev, 0, MOPT_GTE0},
1396
	{Opt_journal_path, 0, MOPT_STRING},
1397
	{Opt_journal_ioprio, 0, MOPT_GTE0},
1398 1399 1400 1401
	{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},
1402 1403
	{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
1404
#ifdef CONFIG_EXT4_FS_POSIX_ACL
1405 1406
	{Opt_acl, EXT4_MOUNT_POSIX_ACL, MOPT_SET},
	{Opt_noacl, EXT4_MOUNT_POSIX_ACL, MOPT_CLEAR},
1407
#else
1408 1409
	{Opt_acl, 0, MOPT_NOSUPPORT},
	{Opt_noacl, 0, MOPT_NOSUPPORT},
1410
#endif
1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426
	{Opt_nouid32, EXT4_MOUNT_NO_UID32, MOPT_SET},
	{Opt_debug, EXT4_MOUNT_DEBUG, MOPT_SET},
	{Opt_quota, EXT4_MOUNT_QUOTA | EXT4_MOUNT_USRQUOTA, MOPT_SET | MOPT_Q},
	{Opt_usrquota, EXT4_MOUNT_QUOTA | EXT4_MOUNT_USRQUOTA,
							MOPT_SET | MOPT_Q},
	{Opt_grpquota, EXT4_MOUNT_QUOTA | EXT4_MOUNT_GRPQUOTA,
							MOPT_SET | MOPT_Q},
	{Opt_noquota, (EXT4_MOUNT_QUOTA | EXT4_MOUNT_USRQUOTA |
		       EXT4_MOUNT_GRPQUOTA), MOPT_CLEAR | MOPT_Q},
	{Opt_usrjquota, 0, MOPT_Q},
	{Opt_grpjquota, 0, MOPT_Q},
	{Opt_offusrjquota, 0, MOPT_Q},
	{Opt_offgrpjquota, 0, MOPT_Q},
	{Opt_jqfmt_vfsold, QFMT_VFS_OLD, MOPT_QFMT},
	{Opt_jqfmt_vfsv0, QFMT_VFS_V0, MOPT_QFMT},
	{Opt_jqfmt_vfsv1, QFMT_VFS_V1, MOPT_QFMT},
1427
	{Opt_max_dir_size_kb, 0, MOPT_GTE0},
1428
	{Opt_test_dummy_encryption, 0, MOPT_GTE0},
1429 1430 1431 1432 1433 1434 1435 1436 1437
	{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;
1438 1439
	kuid_t uid;
	kgid_t gid;
1440 1441
	int arg = 0;

1442 1443 1444 1445 1446 1447 1448 1449 1450 1451
#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
1452
	switch (token) {
1453 1454 1455 1456
	case Opt_noacl:
	case Opt_nouser_xattr:
		ext4_msg(sb, KERN_WARNING, deprecated_msg, opt, "3.5");
		break;
1457 1458 1459
	case Opt_sb:
		return 1;	/* handled by get_sb_block() */
	case Opt_removed:
1460
		ext4_msg(sb, KERN_WARNING, "Ignoring removed %s option", opt);
1461 1462 1463 1464 1465 1466 1467
		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;
1468 1469 1470 1471 1472 1473
	case Opt_lazytime:
		sb->s_flags |= MS_LAZYTIME;
		return 1;
	case Opt_nolazytime:
		sb->s_flags &= ~MS_LAZYTIME;
		return 1;
1474 1475
	}

1476 1477 1478 1479 1480 1481 1482 1483 1484 1485
	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;
	}

1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496
	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;
	}

1497
	if (args->from && !(m->flags & MOPT_STRING) && match_int(args, &arg))
1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521
		return -1;
	if (args->from && (m->flags & MOPT_GTE0) && (arg < 0))
		return -1;
	if (m->flags & MOPT_EXPLICIT)
		set_opt2(sb, EXPLICIT_DELALLOC);
	if (m->flags & MOPT_CLEAR_ERR)
		clear_opt(sb, ERRORS_MASK);
	if (token == Opt_noquota && sb_any_quota_loaded(sb)) {
		ext4_msg(sb, KERN_ERR, "Cannot change quota "
			 "options when quota turned on");
		return -1;
	}

	if (m->flags & MOPT_NOSUPPORT) {
		ext4_msg(sb, KERN_ERR, "%s option not supported", opt);
	} else if (token == Opt_commit) {
		if (arg == 0)
			arg = JBD2_DEFAULT_MAX_COMMIT_AGE;
		sbi->s_commit_interval = HZ * arg;
	} else if (token == Opt_max_batch_time) {
		sbi->s_max_batch_time = arg;
	} else if (token == Opt_min_batch_time) {
		sbi->s_min_batch_time = arg;
	} else if (token == Opt_inode_readahead_blks) {
1522 1523 1524 1525
		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");
1526
			return -1;
1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541
		}
		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);
1542 1543
			return -1;
		}
1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558
		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;
1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584
	} 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;
		}

1585
		journal_inode = d_inode(path.dentry);
1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596
		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);
1597 1598 1599 1600 1601 1602 1603 1604
	} 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);
1605 1606 1607 1608 1609 1610 1611 1612 1613
	} 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
1614 1615 1616 1617 1618
	} 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) {
1619
				ext4_msg(sb, KERN_ERR,
1620 1621
					 "Cannot change data mode on remount");
				return -1;
1622
			}
1623
		} else {
1624 1625
			clear_opt(sb, DATA_FLAGS);
			sbi->s_mount_opt |= m->mount_opt;
1626
		}
1627 1628 1629 1630 1631 1632 1633 1634
#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;
		}
1635 1636 1637 1638 1639 1640 1641
		if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
					       EXT4_FEATURE_RO_COMPAT_QUOTA)) {
			ext4_msg(sb, KERN_ERR,
				 "Cannot set journaled quota options "
				 "when QUOTA feature is enabled");
			return -1;
		}
1642
		sbi->s_jquota_fmt = m->mount_opt;
R
Ross Zwisler 已提交
1643 1644 1645 1646 1647
#endif
#ifndef CONFIG_FS_DAX
	} else if (token == Opt_dax) {
		ext4_msg(sb, KERN_INFO, "dax option not supported");
		return -1;
1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663
#endif
	} else {
		if (!args->from)
			arg = 1;
		if (m->flags & MOPT_CLEAR)
			arg = !arg;
		else if (unlikely(!(m->flags & MOPT_SET))) {
			ext4_msg(sb, KERN_WARNING,
				 "buggy handling of option %s", opt);
			WARN_ON(1);
			return -1;
		}
		if (arg != 0)
			sbi->s_mount_opt |= m->mount_opt;
		else
			sbi->s_mount_opt &= ~m->mount_opt;
1664
	}
1665
	return 1;
1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687
}

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.
		 */
1688
		args[0].to = args[0].from = NULL;
1689 1690 1691 1692
		token = match_token(p, tokens, args);
		if (handle_mount_opt(sb, p, token, args, journal_devnum,
				     journal_ioprio, is_remount) < 0)
			return 0;
1693 1694
	}
#ifdef CONFIG_QUOTA
1695 1696 1697 1698 1699 1700
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_QUOTA) &&
	    (test_opt(sb, USRQUOTA) || test_opt(sb, GRPQUOTA))) {
		ext4_msg(sb, KERN_ERR, "Cannot set quota options when QUOTA "
			 "feature is enabled");
		return 0;
	}
1701
	if (sbi->s_qf_names[USRQUOTA] || sbi->s_qf_names[GRPQUOTA]) {
1702
		if (test_opt(sb, USRQUOTA) && sbi->s_qf_names[USRQUOTA])
1703
			clear_opt(sb, USRQUOTA);
1704

1705
		if (test_opt(sb, GRPQUOTA) && sbi->s_qf_names[GRPQUOTA])
1706
			clear_opt(sb, GRPQUOTA);
1707

D
Dmitry Monakhov 已提交
1708
		if (test_opt(sb, GRPQUOTA) || test_opt(sb, USRQUOTA)) {
1709 1710
			ext4_msg(sb, KERN_ERR, "old and new quota "
					"format mixing");
1711 1712 1713 1714
			return 0;
		}

		if (!sbi->s_jquota_fmt) {
1715 1716
			ext4_msg(sb, KERN_ERR, "journaled quota format "
					"not specified");
1717 1718 1719 1720
			return 0;
		}
	}
#endif
J
Jan Kara 已提交
1721 1722 1723 1724 1725 1726 1727 1728 1729 1730
	if (test_opt(sb, DIOREAD_NOLOCK)) {
		int blocksize =
			BLOCK_SIZE << le32_to_cpu(sbi->s_es->s_log_block_size);

		if (blocksize < PAGE_CACHE_SIZE) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "dioread_nolock if block size != PAGE_SIZE");
			return 0;
		}
	}
1731 1732 1733 1734 1735 1736
	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;
	}
1737 1738 1739
	return 1;
}

1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770
static inline void ext4_show_quota_options(struct seq_file *seq,
					   struct super_block *sb)
{
#if defined(CONFIG_QUOTA)
	struct ext4_sb_info *sbi = EXT4_SB(sb);

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

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

	if (sbi->s_qf_names[USRQUOTA])
		seq_printf(seq, ",usrjquota=%s", sbi->s_qf_names[USRQUOTA]);

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

1771 1772
static const char *token2str(int token)
{
1773
	const struct match_token *t;
1774 1775 1776 1777 1778 1779 1780

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

1781 1782 1783 1784 1785
/*
 * Show an option if
 *  - it's set to a non-default value OR
 *  - if the per-sb default is different from the global default
 */
1786 1787
static int _ext4_show_options(struct seq_file *seq, struct super_block *sb,
			      int nodefs)
1788 1789 1790
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_super_block *es = sbi->s_es;
1791
	int def_errors, def_mount_opt = nodefs ? 0 : sbi->s_def_mount_opt;
1792
	const struct mount_opts *m;
1793
	char sep = nodefs ? '\n' : ',';
1794

1795 1796
#define SEQ_OPTS_PUTS(str) seq_printf(seq, "%c" str, sep)
#define SEQ_OPTS_PRINT(str, arg) seq_printf(seq, "%c" str, sep, arg)
1797 1798

	if (sbi->s_sb_block != 1)
1799 1800 1801 1802 1803 1804 1805
		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;
1806
		if (!(m->mount_opt & (sbi->s_mount_opt ^ def_mount_opt)))
1807 1808 1809 1810 1811 1812
			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));
1813
	}
1814

1815
	if (nodefs || !uid_eq(sbi->s_resuid, make_kuid(&init_user_ns, EXT4_DEF_RESUID)) ||
1816
	    le16_to_cpu(es->s_def_resuid) != EXT4_DEF_RESUID)
1817 1818 1819
		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)) ||
1820
	    le16_to_cpu(es->s_def_resgid) != EXT4_DEF_RESGID)
1821 1822
		SEQ_OPTS_PRINT("resgid=%u",
				from_kgid_munged(&init_user_ns, sbi->s_resgid));
1823
	def_errors = nodefs ? -1 : le16_to_cpu(es->s_errors);
1824 1825
	if (test_opt(sb, ERRORS_RO) && def_errors != EXT4_ERRORS_RO)
		SEQ_OPTS_PUTS("errors=remount-ro");
1826
	if (test_opt(sb, ERRORS_CONT) && def_errors != EXT4_ERRORS_CONTINUE)
1827
		SEQ_OPTS_PUTS("errors=continue");
1828
	if (test_opt(sb, ERRORS_PANIC) && def_errors != EXT4_ERRORS_PANIC)
1829
		SEQ_OPTS_PUTS("errors=panic");
1830
	if (nodefs || sbi->s_commit_interval != JBD2_DEFAULT_MAX_COMMIT_AGE*HZ)
1831
		SEQ_OPTS_PRINT("commit=%lu", sbi->s_commit_interval / HZ);
1832
	if (nodefs || sbi->s_min_batch_time != EXT4_DEF_MIN_BATCH_TIME)
1833
		SEQ_OPTS_PRINT("min_batch_time=%u", sbi->s_min_batch_time);
1834
	if (nodefs || sbi->s_max_batch_time != EXT4_DEF_MAX_BATCH_TIME)
1835
		SEQ_OPTS_PRINT("max_batch_time=%u", sbi->s_max_batch_time);
1836
	if (sb->s_flags & MS_I_VERSION)
1837
		SEQ_OPTS_PUTS("i_version");
1838
	if (nodefs || sbi->s_stripe)
1839
		SEQ_OPTS_PRINT("stripe=%lu", sbi->s_stripe);
1840
	if (EXT4_MOUNT_DATA_FLAGS & (sbi->s_mount_opt ^ def_mount_opt)) {
1841 1842 1843 1844 1845 1846 1847
		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");
	}
1848 1849
	if (nodefs ||
	    sbi->s_inode_readahead_blks != EXT4_DEF_INODE_READAHEAD_BLKS)
1850 1851
		SEQ_OPTS_PRINT("inode_readahead_blks=%u",
			       sbi->s_inode_readahead_blks);
1852

1853 1854
	if (nodefs || (test_opt(sb, INIT_INODE_TABLE) &&
		       (sbi->s_li_wait_mult != EXT4_DEF_LI_WAIT_MULT)))
1855
		SEQ_OPTS_PRINT("init_itable=%u", sbi->s_li_wait_mult);
1856 1857
	if (nodefs || sbi->s_max_dir_size_kb)
		SEQ_OPTS_PRINT("max_dir_size_kb=%u", sbi->s_max_dir_size_kb);
1858 1859 1860 1861 1862

	ext4_show_quota_options(seq, sb);
	return 0;
}

1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880
static int ext4_show_options(struct seq_file *seq, struct dentry *root)
{
	return _ext4_show_options(seq, root->d_sb, 0);
}

static int options_seq_show(struct seq_file *seq, void *offset)
{
	struct super_block *sb = seq->private;
	int rc;

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

static int options_open_fs(struct inode *inode, struct file *file)
{
A
Al Viro 已提交
1881
	return single_open(file, options_seq_show, PDE_DATA(inode));
1882 1883 1884 1885 1886 1887 1888 1889 1890 1891
}

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

1892
static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es,
1893 1894
			    int read_only)
{
1895
	struct ext4_sb_info *sbi = EXT4_SB(sb);
1896 1897
	int res = 0;

1898
	if (le32_to_cpu(es->s_rev_level) > EXT4_MAX_SUPP_REV) {
1899 1900
		ext4_msg(sb, KERN_ERR, "revision level too high, "
			 "forcing read-only mode");
1901 1902 1903
		res = MS_RDONLY;
	}
	if (read_only)
1904
		goto done;
1905
	if (!(sbi->s_mount_state & EXT4_VALID_FS))
1906 1907
		ext4_msg(sb, KERN_WARNING, "warning: mounting unchecked fs, "
			 "running e2fsck is recommended");
1908
	else if (sbi->s_mount_state & EXT4_ERROR_FS)
1909 1910 1911
		ext4_msg(sb, KERN_WARNING,
			 "warning: mounting fs with errors, "
			 "running e2fsck is recommended");
1912
	else if ((__s16) le16_to_cpu(es->s_max_mnt_count) > 0 &&
1913 1914
		 le16_to_cpu(es->s_mnt_count) >=
		 (unsigned short) (__s16) le16_to_cpu(es->s_max_mnt_count))
1915 1916 1917
		ext4_msg(sb, KERN_WARNING,
			 "warning: maximal mount count reached, "
			 "running e2fsck is recommended");
1918 1919 1920
	else if (le32_to_cpu(es->s_checkinterval) &&
		(le32_to_cpu(es->s_lastcheck) +
			le32_to_cpu(es->s_checkinterval) <= get_seconds()))
1921 1922 1923
		ext4_msg(sb, KERN_WARNING,
			 "warning: checktime reached, "
			 "running e2fsck is recommended");
1924
	if (!sbi->s_journal)
1925
		es->s_state &= cpu_to_le16(~EXT4_VALID_FS);
1926
	if (!(__s16) le16_to_cpu(es->s_max_mnt_count))
1927
		es->s_max_mnt_count = cpu_to_le16(EXT4_DFL_MAX_MNT_COUNT);
M
Marcin Slusarz 已提交
1928
	le16_add_cpu(&es->s_mnt_count, 1);
1929
	es->s_mtime = cpu_to_le32(get_seconds());
1930
	ext4_update_dynamic_rev(sb);
1931 1932
	if (sbi->s_journal)
		EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
1933

1934
	ext4_commit_super(sb, 1);
1935
done:
1936
	if (test_opt(sb, DEBUG))
1937
		printk(KERN_INFO "[EXT4 FS bs=%lu, gc=%u, "
1938
				"bpg=%lu, ipg=%lu, mo=%04x, mo2=%04x]\n",
1939 1940
			sb->s_blocksize,
			sbi->s_groups_count,
1941 1942
			EXT4_BLOCKS_PER_GROUP(sb),
			EXT4_INODES_PER_GROUP(sb),
1943
			sbi->s_mount_opt, sbi->s_mount_opt2);
1944

D
Dan Magenheimer 已提交
1945
	cleancache_init_fs(sb);
1946 1947 1948
	return res;
}

1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973
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
1974
		kvfree(sbi->s_flex_groups);
1975 1976 1977 1978 1979 1980
	}
	sbi->s_flex_groups = new_groups;
	sbi->s_flex_groups_allocated = size / sizeof(struct flex_groups);
	return 0;
}

1981 1982 1983 1984 1985
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;
1986
	int i, err;
1987

1988
	sbi->s_log_groups_per_flex = sbi->s_es->s_log_groups_per_flex;
1989
	if (sbi->s_log_groups_per_flex < 1 || sbi->s_log_groups_per_flex > 31) {
1990 1991 1992 1993
		sbi->s_log_groups_per_flex = 0;
		return 1;
	}

1994 1995
	err = ext4_alloc_flex_bg_array(sb, sbi->s_groups_count);
	if (err)
1996
		goto failed;
1997 1998

	for (i = 0; i < sbi->s_groups_count; i++) {
1999
		gdp = ext4_get_group_desc(sb, i, NULL);
2000 2001

		flex_group = ext4_flex_group(sbi, i);
2002 2003
		atomic_add(ext4_free_inodes_count(sb, gdp),
			   &sbi->s_flex_groups[flex_group].free_inodes);
2004 2005
		atomic64_add(ext4_free_group_clusters(sb, gdp),
			     &sbi->s_flex_groups[flex_group].free_clusters);
2006 2007
		atomic_add(ext4_used_dirs_count(sb, gdp),
			   &sbi->s_flex_groups[flex_group].used_dirs);
2008 2009 2010 2011 2012 2013 2014
	}

	return 1;
failed:
	return 0;
}

2015 2016
static __le16 ext4_group_desc_csum(struct ext4_sb_info *sbi, __u32 block_group,
				   struct ext4_group_desc *gdp)
A
Andreas Dilger 已提交
2017
{
2018
	int offset;
A
Andreas Dilger 已提交
2019
	__u16 crc = 0;
2020
	__le32 le_group = cpu_to_le32(block_group);
A
Andreas Dilger 已提交
2021

2022
	if (ext4_has_metadata_csum(sbi->s_sb)) {
2023
		/* Use new metadata_csum algorithm */
2024
		__le16 save_csum;
2025 2026
		__u32 csum32;

2027
		save_csum = gdp->bg_checksum;
2028 2029 2030 2031 2032
		gdp->bg_checksum = 0;
		csum32 = ext4_chksum(sbi, sbi->s_csum_seed, (__u8 *)&le_group,
				     sizeof(le_group));
		csum32 = ext4_chksum(sbi, csum32, (__u8 *)gdp,
				     sbi->s_desc_size);
2033
		gdp->bg_checksum = save_csum;
2034 2035 2036

		crc = csum32 & 0xFFFF;
		goto out;
A
Andreas Dilger 已提交
2037 2038
	}

2039
	/* old crc16 code */
2040 2041 2042 2043
	if (!(sbi->s_es->s_feature_ro_compat &
	      cpu_to_le32(EXT4_FEATURE_RO_COMPAT_GDT_CSUM)))
		return 0;

2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058
	offset = offsetof(struct ext4_group_desc, bg_checksum);

	crc = crc16(~0, sbi->s_es->s_uuid, sizeof(sbi->s_es->s_uuid));
	crc = crc16(crc, (__u8 *)&le_group, sizeof(le_group));
	crc = crc16(crc, (__u8 *)gdp, offset);
	offset += sizeof(gdp->bg_checksum); /* skip checksum */
	/* for checksum of struct ext4_group_desc do the rest...*/
	if ((sbi->s_es->s_feature_incompat &
	     cpu_to_le32(EXT4_FEATURE_INCOMPAT_64BIT)) &&
	    offset < le16_to_cpu(sbi->s_es->s_desc_size))
		crc = crc16(crc, (__u8 *)gdp + offset,
			    le16_to_cpu(sbi->s_es->s_desc_size) -
				offset);

out:
A
Andreas Dilger 已提交
2059 2060 2061
	return cpu_to_le16(crc);
}

2062
int ext4_group_desc_csum_verify(struct super_block *sb, __u32 block_group,
A
Andreas Dilger 已提交
2063 2064
				struct ext4_group_desc *gdp)
{
2065 2066 2067
	if (ext4_has_group_desc_csum(sb) &&
	    (gdp->bg_checksum != ext4_group_desc_csum(EXT4_SB(sb),
						      block_group, gdp)))
A
Andreas Dilger 已提交
2068 2069 2070 2071 2072
		return 0;

	return 1;
}

2073 2074 2075 2076 2077 2078 2079 2080
void ext4_group_desc_csum_set(struct super_block *sb, __u32 block_group,
			      struct ext4_group_desc *gdp)
{
	if (!ext4_has_group_desc_csum(sb))
		return;
	gdp->bg_checksum = ext4_group_desc_csum(EXT4_SB(sb), block_group, gdp);
}

2081
/* Called at mount-time, super-block is locked */
2082 2083
static int ext4_check_descriptors(struct super_block *sb,
				  ext4_group_t *first_not_zeroed)
2084
{
2085 2086 2087
	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
2088 2089 2090
	ext4_fsblk_t block_bitmap;
	ext4_fsblk_t inode_bitmap;
	ext4_fsblk_t inode_table;
J
Jose R. Santos 已提交
2091
	int flexbg_flag = 0;
2092
	ext4_group_t i, grp = sbi->s_groups_count;
2093

J
Jose R. Santos 已提交
2094 2095 2096
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG))
		flexbg_flag = 1;

2097
	ext4_debug("Checking group descriptors");
2098

2099 2100 2101
	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
2102
		if (i == sbi->s_groups_count - 1 || flexbg_flag)
L
Laurent Vivier 已提交
2103
			last_block = ext4_blocks_count(sbi->s_es) - 1;
2104 2105
		else
			last_block = first_block +
2106
				(EXT4_BLOCKS_PER_GROUP(sb) - 1);
2107

2108 2109 2110 2111
		if ((grp == sbi->s_groups_count) &&
		   !(gdp->bg_flags & cpu_to_le16(EXT4_BG_INODE_ZEROED)))
			grp = i;

2112
		block_bitmap = ext4_block_bitmap(sb, gdp);
2113
		if (block_bitmap < first_block || block_bitmap > last_block) {
2114
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
2115
			       "Block bitmap for group %u not in group "
2116
			       "(block %llu)!", i, block_bitmap);
2117 2118
			return 0;
		}
2119
		inode_bitmap = ext4_inode_bitmap(sb, gdp);
2120
		if (inode_bitmap < first_block || inode_bitmap > last_block) {
2121
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
2122
			       "Inode bitmap for group %u not in group "
2123
			       "(block %llu)!", i, inode_bitmap);
2124 2125
			return 0;
		}
2126
		inode_table = ext4_inode_table(sb, gdp);
L
Laurent Vivier 已提交
2127
		if (inode_table < first_block ||
2128
		    inode_table + sbi->s_itb_per_group - 1 > last_block) {
2129
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
2130
			       "Inode table for group %u not in group "
2131
			       "(block %llu)!", i, inode_table);
2132 2133
			return 0;
		}
2134
		ext4_lock_group(sb, i);
2135
		if (!ext4_group_desc_csum_verify(sb, i, gdp)) {
2136 2137 2138 2139
			ext4_msg(sb, KERN_ERR, "ext4_check_descriptors: "
				 "Checksum for group %u failed (%u!=%u)",
				 i, le16_to_cpu(ext4_group_desc_csum(sbi, i,
				     gdp)), le16_to_cpu(gdp->bg_checksum));
2140
			if (!(sb->s_flags & MS_RDONLY)) {
2141
				ext4_unlock_group(sb, i);
2142
				return 0;
2143
			}
A
Andreas Dilger 已提交
2144
		}
2145
		ext4_unlock_group(sb, i);
J
Jose R. Santos 已提交
2146 2147
		if (!flexbg_flag)
			first_block += EXT4_BLOCKS_PER_GROUP(sb);
2148
	}
2149 2150
	if (NULL != first_not_zeroed)
		*first_not_zeroed = grp;
2151 2152 2153
	return 1;
}

2154
/* ext4_orphan_cleanup() walks a singly-linked list of inodes (starting at
2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166
 * 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
2167
 * ext4_free_inode().  The only reason we would point at a wrong inode is if
2168 2169 2170
 * 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.
 */
2171 2172
static void ext4_orphan_cleanup(struct super_block *sb,
				struct ext4_super_block *es)
2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183
{
	unsigned int s_flags = sb->s_flags;
	int nr_orphans = 0, nr_truncates = 0;
#ifdef CONFIG_QUOTA
	int i;
#endif
	if (!es->s_last_orphan) {
		jbd_debug(4, "no orphan inodes to clean up\n");
		return;
	}

2184
	if (bdev_read_only(sb->s_bdev)) {
2185 2186
		ext4_msg(sb, KERN_ERR, "write access "
			"unavailable, skipping orphan cleanup");
2187 2188 2189
		return;
	}

2190 2191 2192 2193 2194 2195 2196
	/* 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;
	}

2197
	if (EXT4_SB(sb)->s_mount_state & EXT4_ERROR_FS) {
2198 2199
		/* don't clear list on RO mount w/ errors */
		if (es->s_last_orphan && !(s_flags & MS_RDONLY)) {
2200
			ext4_msg(sb, KERN_INFO, "Errors on filesystem, "
2201
				  "clearing orphan list.\n");
2202 2203
			es->s_last_orphan = 0;
		}
2204 2205 2206 2207 2208
		jbd_debug(1, "Skipping orphan recovery on fs with errors.\n");
		return;
	}

	if (s_flags & MS_RDONLY) {
2209
		ext4_msg(sb, KERN_INFO, "orphan cleanup on readonly fs");
2210 2211 2212 2213 2214 2215
		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
2216
	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
2217 2218
		if (EXT4_SB(sb)->s_qf_names[i]) {
			int ret = ext4_quota_on_mount(sb, i);
2219
			if (ret < 0)
2220 2221 2222
				ext4_msg(sb, KERN_ERR,
					"Cannot turn on journaled "
					"quota: error %d", ret);
2223 2224 2225 2226 2227 2228 2229
		}
	}
#endif

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

2230 2231
		inode = ext4_orphan_get(sb, le32_to_cpu(es->s_last_orphan));
		if (IS_ERR(inode)) {
2232 2233 2234 2235
			es->s_last_orphan = 0;
			break;
		}

2236
		list_add(&EXT4_I(inode)->i_orphan, &EXT4_SB(sb)->s_orphan);
2237
		dquot_initialize(inode);
2238
		if (inode->i_nlink) {
2239 2240 2241 2242
			if (test_opt(sb, DEBUG))
				ext4_msg(sb, KERN_DEBUG,
					"%s: truncating inode %lu to %lld bytes",
					__func__, inode->i_ino, inode->i_size);
2243
			jbd_debug(2, "truncating inode %lu to %lld bytes\n",
2244
				  inode->i_ino, inode->i_size);
2245
			mutex_lock(&inode->i_mutex);
2246
			truncate_inode_pages(inode->i_mapping, inode->i_size);
2247
			ext4_truncate(inode);
2248
			mutex_unlock(&inode->i_mutex);
2249 2250
			nr_truncates++;
		} else {
2251 2252 2253 2254
			if (test_opt(sb, DEBUG))
				ext4_msg(sb, KERN_DEBUG,
					"%s: deleting unreferenced inode %lu",
					__func__, inode->i_ino);
2255 2256 2257 2258 2259 2260 2261
			jbd_debug(2, "deleting unreferenced inode %lu\n",
				  inode->i_ino);
			nr_orphans++;
		}
		iput(inode);  /* The delete magic happens here! */
	}

2262
#define PLURAL(x) (x), ((x) == 1) ? "" : "s"
2263 2264

	if (nr_orphans)
2265 2266
		ext4_msg(sb, KERN_INFO, "%d orphan inode%s deleted",
		       PLURAL(nr_orphans));
2267
	if (nr_truncates)
2268 2269
		ext4_msg(sb, KERN_INFO, "%d truncate%s cleaned up",
		       PLURAL(nr_truncates));
2270 2271
#ifdef CONFIG_QUOTA
	/* Turn quotas off */
J
Jan Kara 已提交
2272
	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
2273
		if (sb_dqopt(sb)->files[i])
2274
			dquot_quota_off(sb, i);
2275 2276 2277 2278
	}
#endif
	sb->s_flags = s_flags; /* Restore MS_RDONLY status */
}
2279

2280 2281 2282 2283 2284 2285 2286
/*
 * 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.
 *
2287 2288 2289 2290 2291 2292
 * 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.
 *
2293 2294
 * Note, this does *not* consider any metadata overhead for vfs i_blocks.
 */
2295
static loff_t ext4_max_size(int blkbits, int has_huge_files)
2296 2297 2298 2299 2300
{
	loff_t res;
	loff_t upper_limit = MAX_LFS_FILESIZE;

	/* small i_blocks in vfs inode? */
2301
	if (!has_huge_files || sizeof(blkcnt_t) < sizeof(u64)) {
2302
		/*
2303
		 * CONFIG_LBDAF is not enabled implies the inode
2304 2305 2306 2307 2308 2309 2310 2311 2312 2313
		 * 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;
	}

2314 2315 2316 2317 2318 2319
	/*
	 * 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;
2320 2321 2322 2323 2324 2325 2326 2327
	res <<= blkbits;

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

	return res;
}
2328 2329

/*
2330
 * Maximal bitmap file size.  There is a direct, and {,double-,triple-}indirect
2331 2332
 * 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.
2333
 */
2334
static loff_t ext4_max_bitmap_size(int bits, int has_huge_files)
2335
{
2336
	loff_t res = EXT4_NDIR_BLOCKS;
2337 2338
	int meta_blocks;
	loff_t upper_limit;
2339 2340 2341 2342 2343 2344
	/* 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.
2345 2346
	 */

2347
	if (!has_huge_files || sizeof(blkcnt_t) < sizeof(u64)) {
2348
		/*
2349
		 * !has_huge_files or CONFIG_LBDAF not enabled implies that
2350 2351
		 * the inode i_block field represents total file blocks in
		 * 2^32 512-byte sectors == size of vfs inode i_blocks * 8
2352 2353 2354 2355 2356 2357 2358
		 */
		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
2359 2360 2361 2362 2363 2364
		/*
		 * 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
		 */
2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377
		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;
2378 2379 2380 2381 2382 2383 2384

	res += 1LL << (bits-2);
	res += 1LL << (2*(bits-2));
	res += 1LL << (3*(bits-2));
	res <<= bits;
	if (res > upper_limit)
		res = upper_limit;
2385 2386 2387 2388

	if (res > MAX_LFS_FILESIZE)
		res = MAX_LFS_FILESIZE;

2389 2390 2391
	return res;
}

2392
static ext4_fsblk_t descriptor_loc(struct super_block *sb,
2393
				   ext4_fsblk_t logical_sb_block, int nr)
2394
{
2395
	struct ext4_sb_info *sbi = EXT4_SB(sb);
2396
	ext4_group_t bg, first_meta_bg;
2397 2398 2399 2400
	int has_super = 0;

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

2401
	if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_META_BG) ||
2402
	    nr < first_meta_bg)
2403
		return logical_sb_block + nr + 1;
2404
	bg = sbi->s_desc_per_block * nr;
2405
	if (ext4_bg_has_super(sb, bg))
2406
		has_super = 1;
2407

2408 2409 2410 2411 2412 2413 2414 2415 2416 2417
	/*
	 * 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++;

2418
	return (has_super + ext4_group_first_block_no(sb, bg));
2419 2420
}

2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436
/**
 * 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
2437
	int ret;
2438 2439

	if (sbi->s_stripe && sbi->s_stripe <= sbi->s_blocks_per_group)
D
Dan Ehrenberg 已提交
2440 2441 2442 2443 2444 2445 2446
		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;
2447

D
Dan Ehrenberg 已提交
2448 2449 2450 2451 2452 2453
	/*
	 * 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;
2454

D
Dan Ehrenberg 已提交
2455
	return ret;
2456
}
2457

T
Theodore Ts'o 已提交
2458 2459 2460 2461 2462
/* sysfs supprt */

struct ext4_attr {
	struct attribute attr;
	ssize_t (*show)(struct ext4_attr *, struct ext4_sb_info *, char *);
2463
	ssize_t (*store)(struct ext4_attr *, struct ext4_sb_info *,
T
Theodore Ts'o 已提交
2464
			 const char *, size_t);
2465 2466 2467 2468
	union {
		int offset;
		int deprecated_val;
	} u;
T
Theodore Ts'o 已提交
2469 2470
};

L
Lukas Czerner 已提交
2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481
static int parse_strtoull(const char *buf,
		unsigned long long max, unsigned long long *value)
{
	int ret;

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

T
Theodore Ts'o 已提交
2482 2483 2484 2485 2486
static ssize_t delayed_allocation_blocks_show(struct ext4_attr *a,
					      struct ext4_sb_info *sbi,
					      char *buf)
{
	return snprintf(buf, PAGE_SIZE, "%llu\n",
2487 2488
		(s64) EXT4_C2B(sbi,
			percpu_counter_sum(&sbi->s_dirtyclusters_counter)));
T
Theodore Ts'o 已提交
2489 2490 2491 2492 2493 2494 2495
}

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

2496 2497
	if (!sb->s_bdev->bd_part)
		return snprintf(buf, PAGE_SIZE, "0\n");
T
Theodore Ts'o 已提交
2498 2499 2500 2501 2502 2503 2504 2505 2506 2507
	return snprintf(buf, PAGE_SIZE, "%lu\n",
			(part_stat_read(sb->s_bdev->bd_part, sectors[1]) -
			 sbi->s_sectors_written_start) >> 1);
}

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

2508 2509
	if (!sb->s_bdev->bd_part)
		return snprintf(buf, PAGE_SIZE, "0\n");
T
Theodore Ts'o 已提交
2510
	return snprintf(buf, PAGE_SIZE, "%llu\n",
2511
			(unsigned long long)(sbi->s_kbytes_written +
T
Theodore Ts'o 已提交
2512
			((part_stat_read(sb->s_bdev->bd_part, sectors[1]) -
2513
			  EXT4_SB(sb)->s_sectors_written_start) >> 1)));
T
Theodore Ts'o 已提交
2514 2515 2516 2517 2518 2519 2520
}

static ssize_t inode_readahead_blks_store(struct ext4_attr *a,
					  struct ext4_sb_info *sbi,
					  const char *buf, size_t count)
{
	unsigned long t;
2521
	int ret;
T
Theodore Ts'o 已提交
2522

2523 2524 2525
	ret = kstrtoul(skip_spaces(buf), 0, &t);
	if (ret)
		return ret;
T
Theodore Ts'o 已提交
2526

2527
	if (t && (!is_power_of_2(t) || t > 0x40000000))
T
Theodore Ts'o 已提交
2528 2529 2530 2531 2532 2533 2534
		return -EINVAL;

	sbi->s_inode_readahead_blks = t;
	return count;
}

static ssize_t sbi_ui_show(struct ext4_attr *a,
2535
			   struct ext4_sb_info *sbi, char *buf)
T
Theodore Ts'o 已提交
2536
{
2537
	unsigned int *ui = (unsigned int *) (((char *) sbi) + a->u.offset);
T
Theodore Ts'o 已提交
2538 2539 2540 2541 2542 2543 2544 2545

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

static ssize_t sbi_ui_store(struct ext4_attr *a,
			    struct ext4_sb_info *sbi,
			    const char *buf, size_t count)
{
2546
	unsigned int *ui = (unsigned int *) (((char *) sbi) + a->u.offset);
T
Theodore Ts'o 已提交
2547
	unsigned long t;
2548
	int ret;
T
Theodore Ts'o 已提交
2549

2550 2551 2552
	ret = kstrtoul(skip_spaces(buf), 0, &t);
	if (ret)
		return ret;
T
Theodore Ts'o 已提交
2553 2554 2555 2556
	*ui = t;
	return count;
}

2557 2558 2559 2560 2561 2562 2563 2564 2565 2566
static ssize_t es_ui_show(struct ext4_attr *a,
			   struct ext4_sb_info *sbi, char *buf)
{

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

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

L
Lukas Czerner 已提交
2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587
static ssize_t reserved_clusters_show(struct ext4_attr *a,
				  struct ext4_sb_info *sbi, char *buf)
{
	return snprintf(buf, PAGE_SIZE, "%llu\n",
		(unsigned long long) atomic64_read(&sbi->s_resv_clusters));
}

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

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

	return ret ? ret : count;
}

2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604
static ssize_t trigger_test_error(struct ext4_attr *a,
				  struct ext4_sb_info *sbi,
				  const char *buf, size_t count)
{
	int len = count;

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

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

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

2605 2606 2607 2608 2609 2610
static ssize_t sbi_deprecated_show(struct ext4_attr *a,
				   struct ext4_sb_info *sbi, char *buf)
{
	return snprintf(buf, PAGE_SIZE, "%d\n", a->u.deprecated_val);
}

T
Theodore Ts'o 已提交
2611 2612 2613 2614 2615
#define EXT4_ATTR_OFFSET(_name,_mode,_show,_store,_elname) \
static struct ext4_attr ext4_attr_##_name = {			\
	.attr = {.name = __stringify(_name), .mode = _mode },	\
	.show	= _show,					\
	.store	= _store,					\
2616 2617 2618
	.u = {							\
		.offset = offsetof(struct ext4_sb_info, _elname),\
	},							\
T
Theodore Ts'o 已提交
2619
}
2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630

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

T
Theodore Ts'o 已提交
2631 2632 2633
#define EXT4_ATTR(name, mode, show, store) \
static struct ext4_attr ext4_attr_##name = __ATTR(name, mode, show, store)

2634
#define EXT4_INFO_ATTR(name) EXT4_ATTR(name, 0444, NULL, NULL)
T
Theodore Ts'o 已提交
2635 2636
#define EXT4_RO_ATTR(name) EXT4_ATTR(name, 0444, name##_show, NULL)
#define EXT4_RW_ATTR(name) EXT4_ATTR(name, 0644, name##_show, name##_store)
2637 2638 2639

#define EXT4_RO_ATTR_ES_UI(name, elname)	\
	EXT4_ATTR_OFFSET_ES(name, 0444, es_ui_show, NULL, elname)
T
Theodore Ts'o 已提交
2640 2641
#define EXT4_RW_ATTR_SBI_UI(name, elname)	\
	EXT4_ATTR_OFFSET(name, 0644, sbi_ui_show, sbi_ui_store, elname)
2642

T
Theodore Ts'o 已提交
2643
#define ATTR_LIST(name) &ext4_attr_##name.attr
2644 2645 2646 2647 2648 2649 2650 2651
#define EXT4_DEPRECATED_ATTR(_name, _val)	\
static struct ext4_attr ext4_attr_##_name = {			\
	.attr = {.name = __stringify(_name), .mode = 0444 },	\
	.show	= sbi_deprecated_show,				\
	.u = {							\
		.deprecated_val = _val,				\
	},							\
}
T
Theodore Ts'o 已提交
2652 2653 2654 2655

EXT4_RO_ATTR(delayed_allocation_blocks);
EXT4_RO_ATTR(session_write_kbytes);
EXT4_RO_ATTR(lifetime_write_kbytes);
L
Lukas Czerner 已提交
2656
EXT4_RW_ATTR(reserved_clusters);
T
Theodore Ts'o 已提交
2657 2658
EXT4_ATTR_OFFSET(inode_readahead_blks, 0644, sbi_ui_show,
		 inode_readahead_blks_store, s_inode_readahead_blks);
2659
EXT4_RW_ATTR_SBI_UI(inode_goal, s_inode_goal);
T
Theodore Ts'o 已提交
2660 2661 2662 2663 2664 2665
EXT4_RW_ATTR_SBI_UI(mb_stats, s_mb_stats);
EXT4_RW_ATTR_SBI_UI(mb_max_to_scan, s_mb_max_to_scan);
EXT4_RW_ATTR_SBI_UI(mb_min_to_scan, s_mb_min_to_scan);
EXT4_RW_ATTR_SBI_UI(mb_order2_req, s_mb_order2_reqs);
EXT4_RW_ATTR_SBI_UI(mb_stream_req, s_mb_stream_request);
EXT4_RW_ATTR_SBI_UI(mb_group_prealloc, s_mb_group_prealloc);
2666
EXT4_DEPRECATED_ATTR(max_writeback_mb_bump, 128);
2667
EXT4_RW_ATTR_SBI_UI(extent_max_zeroout_kb, s_extent_max_zeroout_kb);
2668
EXT4_ATTR(trigger_fs_error, 0200, NULL, trigger_test_error);
2669 2670 2671 2672 2673 2674
EXT4_RW_ATTR_SBI_UI(err_ratelimit_interval_ms, s_err_ratelimit_state.interval);
EXT4_RW_ATTR_SBI_UI(err_ratelimit_burst, s_err_ratelimit_state.burst);
EXT4_RW_ATTR_SBI_UI(warning_ratelimit_interval_ms, s_warning_ratelimit_state.interval);
EXT4_RW_ATTR_SBI_UI(warning_ratelimit_burst, s_warning_ratelimit_state.burst);
EXT4_RW_ATTR_SBI_UI(msg_ratelimit_interval_ms, s_msg_ratelimit_state.interval);
EXT4_RW_ATTR_SBI_UI(msg_ratelimit_burst, s_msg_ratelimit_state.burst);
2675 2676 2677
EXT4_RO_ATTR_ES_UI(errors_count, s_error_count);
EXT4_RO_ATTR_ES_UI(first_error_time, s_first_error_time);
EXT4_RO_ATTR_ES_UI(last_error_time, s_last_error_time);
T
Theodore Ts'o 已提交
2678 2679 2680 2681 2682

static struct attribute *ext4_attrs[] = {
	ATTR_LIST(delayed_allocation_blocks),
	ATTR_LIST(session_write_kbytes),
	ATTR_LIST(lifetime_write_kbytes),
L
Lukas Czerner 已提交
2683
	ATTR_LIST(reserved_clusters),
T
Theodore Ts'o 已提交
2684
	ATTR_LIST(inode_readahead_blks),
2685
	ATTR_LIST(inode_goal),
T
Theodore Ts'o 已提交
2686 2687 2688 2689 2690 2691
	ATTR_LIST(mb_stats),
	ATTR_LIST(mb_max_to_scan),
	ATTR_LIST(mb_min_to_scan),
	ATTR_LIST(mb_order2_req),
	ATTR_LIST(mb_stream_req),
	ATTR_LIST(mb_group_prealloc),
2692
	ATTR_LIST(max_writeback_mb_bump),
2693
	ATTR_LIST(extent_max_zeroout_kb),
2694
	ATTR_LIST(trigger_fs_error),
2695 2696 2697 2698 2699 2700
	ATTR_LIST(err_ratelimit_interval_ms),
	ATTR_LIST(err_ratelimit_burst),
	ATTR_LIST(warning_ratelimit_interval_ms),
	ATTR_LIST(warning_ratelimit_burst),
	ATTR_LIST(msg_ratelimit_interval_ms),
	ATTR_LIST(msg_ratelimit_burst),
2701 2702 2703
	ATTR_LIST(errors_count),
	ATTR_LIST(first_error_time),
	ATTR_LIST(last_error_time),
T
Theodore Ts'o 已提交
2704 2705 2706
	NULL,
};

2707 2708
/* Features this copy of ext4 supports */
EXT4_INFO_ATTR(lazy_itable_init);
2709
EXT4_INFO_ATTR(batched_discard);
2710
EXT4_INFO_ATTR(meta_bg_resize);
2711
EXT4_INFO_ATTR(encryption);
2712 2713 2714

static struct attribute *ext4_feat_attrs[] = {
	ATTR_LIST(lazy_itable_init),
2715
	ATTR_LIST(batched_discard),
2716
	ATTR_LIST(meta_bg_resize),
2717
	ATTR_LIST(encryption),
2718 2719 2720
	NULL,
};

T
Theodore Ts'o 已提交
2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748
static ssize_t ext4_attr_show(struct kobject *kobj,
			      struct attribute *attr, char *buf)
{
	struct ext4_sb_info *sbi = container_of(kobj, struct ext4_sb_info,
						s_kobj);
	struct ext4_attr *a = container_of(attr, struct ext4_attr, attr);

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

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

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

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

2749
static const struct sysfs_ops ext4_attr_ops = {
T
Theodore Ts'o 已提交
2750 2751 2752 2753 2754 2755 2756 2757 2758 2759
	.show	= ext4_attr_show,
	.store	= ext4_attr_store,
};

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

2760 2761 2762 2763 2764
static void ext4_feat_release(struct kobject *kobj)
{
	complete(&ext4_feat->f_kobj_unregister);
}

2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780
static ssize_t ext4_feat_show(struct kobject *kobj,
			      struct attribute *attr, char *buf)
{
	return snprintf(buf, PAGE_SIZE, "supported\n");
}

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

2781 2782
static struct kobj_type ext4_feat_ktype = {
	.default_attrs	= ext4_feat_attrs,
2783
	.sysfs_ops	= &ext4_feat_ops,
2784 2785 2786
	.release	= ext4_feat_release,
};

2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806
/*
 * Check whether this filesystem can be mounted based on
 * the features present and the RDONLY/RDWR mount requested.
 * Returns 1 if this filesystem can be mounted as requested,
 * 0 if it cannot be.
 */
static int ext4_feature_set_ok(struct super_block *sb, int readonly)
{
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, ~EXT4_FEATURE_INCOMPAT_SUPP)) {
		ext4_msg(sb, KERN_ERR,
			"Couldn't mount because of "
			"unsupported optional features (%x)",
			(le32_to_cpu(EXT4_SB(sb)->s_es->s_feature_incompat) &
			~EXT4_FEATURE_INCOMPAT_SUPP));
		return 0;
	}

	if (readonly)
		return 1;

D
Darrick J. Wong 已提交
2807 2808 2809 2810 2811 2812
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_READONLY)) {
		ext4_msg(sb, KERN_INFO, "filesystem is read-only");
		sb->s_flags |= MS_RDONLY;
		return 1;
	}

2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832
	/* Check that feature set is OK for a read-write mount */
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, ~EXT4_FEATURE_RO_COMPAT_SUPP)) {
		ext4_msg(sb, KERN_ERR, "couldn't mount RDWR because of "
			 "unsupported optional features (%x)",
			 (le32_to_cpu(EXT4_SB(sb)->s_es->s_feature_ro_compat) &
				~EXT4_FEATURE_RO_COMPAT_SUPP));
		return 0;
	}
	/*
	 * Large file size enabled file system can only be mounted
	 * read-write on 32-bit systems if kernel is built with CONFIG_LBDAF
	 */
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_HUGE_FILE)) {
		if (sizeof(blkcnt_t) < sizeof(u64)) {
			ext4_msg(sb, KERN_ERR, "Filesystem with huge files "
				 "cannot be mounted RDWR without "
				 "CONFIG_LBDAF");
			return 0;
		}
	}
2833 2834 2835 2836 2837 2838 2839
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_BIGALLOC) &&
	    !EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS)) {
		ext4_msg(sb, KERN_ERR,
			 "Can't support bigalloc feature without "
			 "extents feature\n");
		return 0;
	}
2840 2841 2842 2843 2844 2845 2846 2847 2848 2849

#ifndef CONFIG_QUOTA
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_QUOTA) &&
	    !readonly) {
		ext4_msg(sb, KERN_ERR,
			 "Filesystem with quota feature cannot be mounted RDWR "
			 "without CONFIG_QUOTA");
		return 0;
	}
#endif  /* CONFIG_QUOTA */
2850 2851 2852
	return 1;
}

2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866
/*
 * 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)
2867 2868
		/* fsck newer than v1.41.13 is needed to clean this condition. */
		ext4_msg(sb, KERN_NOTICE, "error count since last fsck: %u",
2869 2870
			 le32_to_cpu(es->s_error_count));
	if (es->s_first_error_time) {
2871
		printk(KERN_NOTICE "EXT4-fs (%s): initial error at time %u: %.*s:%d",
2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884
		       sb->s_id, le32_to_cpu(es->s_first_error_time),
		       (int) sizeof(es->s_first_error_func),
		       es->s_first_error_func,
		       le32_to_cpu(es->s_first_error_line));
		if (es->s_first_error_ino)
			printk(": inode %u",
			       le32_to_cpu(es->s_first_error_ino));
		if (es->s_first_error_block)
			printk(": block %llu", (unsigned long long)
			       le64_to_cpu(es->s_first_error_block));
		printk("\n");
	}
	if (es->s_last_error_time) {
2885
		printk(KERN_NOTICE "EXT4-fs (%s): last error at time %u: %.*s:%d",
2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900
		       sb->s_id, le32_to_cpu(es->s_last_error_time),
		       (int) sizeof(es->s_last_error_func),
		       es->s_last_error_func,
		       le32_to_cpu(es->s_last_error_line));
		if (es->s_last_error_ino)
			printk(": inode %u",
			       le32_to_cpu(es->s_last_error_ino));
		if (es->s_last_error_block)
			printk(": block %llu", (unsigned long long)
			       le64_to_cpu(es->s_last_error_block));
		printk("\n");
	}
	mod_timer(&sbi->s_err_report, jiffies + 24*60*60*HZ);  /* Once a day */
}

2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912
/* 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;

2913
	sb_start_write(sb);
2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924
	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;
	}

2925
	if (group >= ngroups)
2926 2927 2928 2929 2930 2931 2932
		ret = 1;

	if (!ret) {
		timeout = jiffies;
		ret = ext4_init_inode_table(sb, group,
					    elr->lr_timeout ? 0 : 1);
		if (elr->lr_timeout == 0) {
2933 2934
			timeout = (jiffies - timeout) *
				  elr->lr_sbi->s_li_wait_mult;
2935 2936 2937 2938 2939
			elr->lr_timeout = timeout;
		}
		elr->lr_next_sched = jiffies + elr->lr_timeout;
		elr->lr_next_group = group + 1;
	}
2940
	sb_end_write(sb);
2941 2942 2943 2944 2945 2946

	return ret;
}

/*
 * Remove lr_request from the list_request and free the
2947
 * request structure. Should be called with li_list_mtx held
2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964
 */
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)
{
2965 2966 2967
	mutex_lock(&ext4_li_mtx);
	if (!ext4_li_info) {
		mutex_unlock(&ext4_li_mtx);
2968
		return;
2969
	}
2970 2971

	mutex_lock(&ext4_li_info->li_list_mtx);
2972
	ext4_remove_li_request(EXT4_SB(sb)->s_li_request);
2973
	mutex_unlock(&ext4_li_info->li_list_mtx);
2974
	mutex_unlock(&ext4_li_mtx);
2975 2976
}

2977 2978
static struct task_struct *ext4_lazyinit_task;

2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992
/*
 * 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;
2993
	unsigned long next_wakeup, cur;
2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010

	BUG_ON(NULL == eli);

cont_thread:
	while (true) {
		next_wakeup = MAX_JIFFY_OFFSET;

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

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

3011 3012 3013 3014 3015 3016
			if (time_after_eq(jiffies, elr->lr_next_sched)) {
				if (ext4_run_li_request(elr) != 0) {
					/* error, remove the lazy_init job */
					ext4_remove_li_request(elr);
					continue;
				}
3017 3018 3019 3020 3021 3022 3023
			}

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

3024
		try_to_freeze();
3025

3026 3027
		cur = jiffies;
		if ((time_after_eq(cur, next_wakeup)) ||
3028
		    (MAX_JIFFY_OFFSET == next_wakeup)) {
3029 3030 3031 3032
			cond_resched();
			continue;
		}

3033 3034
		schedule_timeout_interruptible(next_wakeup - cur);

3035 3036 3037 3038
		if (kthread_should_stop()) {
			ext4_clear_request_list();
			goto exit_thread;
		}
3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080
	}

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)
{
3081 3082 3083 3084
	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);
3085 3086 3087
		ext4_clear_request_list();
		kfree(ext4_li_info);
		ext4_li_info = NULL;
3088
		printk(KERN_CRIT "EXT4-fs: error %d creating inode table "
3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156
				 "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.
	 */
3157 3158
	elr->lr_next_sched = jiffies + (prandom_u32() %
				(EXT4_DEF_LI_MAX_START_DELAY * HZ));
3159 3160 3161
	return elr;
}

3162 3163
int ext4_register_li_request(struct super_block *sb,
			     ext4_group_t first_not_zeroed)
3164 3165
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
3166
	struct ext4_li_request *elr = NULL;
3167
	ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count;
3168
	int ret = 0;
3169

3170
	mutex_lock(&ext4_li_mtx);
3171 3172 3173 3174 3175 3176
	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;
3177
		goto out;
3178
	}
3179 3180 3181

	if (first_not_zeroed == ngroups ||
	    (sb->s_flags & MS_RDONLY) ||
3182
	    !test_opt(sb, INIT_INODE_TABLE))
3183
		goto out;
3184 3185

	elr = ext4_li_request_new(sb, first_not_zeroed);
3186 3187 3188 3189
	if (!elr) {
		ret = -ENOMEM;
		goto out;
	}
3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201

	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;
3202 3203 3204 3205 3206 3207
	/*
	 * 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;
3208 3209 3210 3211 3212 3213 3214

	if (!(ext4_li_info->li_state & EXT4_LAZYINIT_RUNNING)) {
		ret = ext4_run_lazyinit_thread();
		if (ret)
			goto out;
	}
out:
3215 3216
	mutex_unlock(&ext4_li_mtx);
	if (ret)
3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230
		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.
	 */
3231
	if (!ext4_li_info || !ext4_lazyinit_task)
3232 3233
		return;

3234
	kthread_stop(ext4_lazyinit_task);
3235 3236
}

3237 3238 3239 3240 3241 3242
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);

3243
	if (ext4_has_metadata_csum(sb)) {
3244
		/* journal checksum v3 */
3245
		compat = 0;
3246
		incompat = JBD2_FEATURE_INCOMPAT_CSUM_V3;
3247 3248 3249 3250 3251 3252
	} else {
		/* journal checksum v1 */
		compat = JBD2_FEATURE_COMPAT_CHECKSUM;
		incompat = 0;
	}

3253 3254 3255 3256
	jbd2_journal_clear_features(sbi->s_journal,
			JBD2_FEATURE_COMPAT_CHECKSUM, 0,
			JBD2_FEATURE_INCOMPAT_CSUM_V3 |
			JBD2_FEATURE_INCOMPAT_CSUM_V2);
3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268
	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 {
3269 3270
		jbd2_journal_clear_features(sbi->s_journal, 0, 0,
				JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT);
3271 3272 3273 3274 3275
	}

	return ret;
}

3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299
/*
 * 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;

3300 3301 3302 3303
	if (!EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_BIGALLOC))
		return (ext4_bg_has_super(sb, grp) + ext4_bg_num_gdb(sb, grp) +
			sbi->s_itb_per_group + 2);

3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352
	first_block = le32_to_cpu(sbi->s_es->s_first_data_block) +
		(grp * EXT4_BLOCKS_PER_GROUP(sb));
	last_block = first_block + EXT4_BLOCKS_PER_GROUP(sb) - 1;
	for (i = 0; i < ngroups; i++) {
		gdp = ext4_get_group_desc(sb, i, NULL);
		b = ext4_block_bitmap(sb, gdp);
		if (b >= first_block && b <= last_block) {
			ext4_set_bit(EXT4_B2C(sbi, b - first_block), buf);
			count++;
		}
		b = ext4_inode_bitmap(sb, gdp);
		if (b >= first_block && b <= last_block) {
			ext4_set_bit(EXT4_B2C(sbi, b - first_block), buf);
			count++;
		}
		b = ext4_inode_table(sb, gdp);
		if (b >= first_block && b + sbi->s_itb_per_group <= last_block)
			for (j = 0; j < sbi->s_itb_per_group; j++, b++) {
				int c = EXT4_B2C(sbi, b - first_block);
				ext4_set_bit(c, buf);
				count++;
			}
		if (i != grp)
			continue;
		s = 0;
		if (ext4_bg_has_super(sb, grp)) {
			ext4_set_bit(s++, buf);
			count++;
		}
		for (j = ext4_bg_num_gdb(sb, grp); j > 0; j--) {
			ext4_set_bit(EXT4_B2C(sbi, s++), buf);
			count++;
		}
	}
	if (!count)
		return 0;
	return EXT4_CLUSTERS_PER_GROUP(sb) -
		ext4_count_free(buf, EXT4_CLUSTERS_PER_GROUP(sb) / 8);
}

/*
 * Compute the overhead and stash it in sbi->s_overhead
 */
int ext4_calculate_overhead(struct super_block *sb)
{
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_super_block *es = sbi->s_es;
	ext4_group_t i, ngroups = ext4_get_groups_count(sb);
	ext4_fsblk_t overhead = 0;
3353
	char *buf = (char *) get_zeroed_page(GFP_NOFS);
3354 3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380

	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();
	}
3381 3382
	/* Add the internal journal blocks as well */
	if (sbi->s_journal && !sbi->journal_bdev)
3383
		overhead += EXT4_NUM_B2C(sbi, sbi->s_journal->j_maxlen);
3384

3385 3386 3387 3388 3389 3390
	sbi->s_overhead = overhead;
	smp_wmb();
	free_page((unsigned long) buf);
	return 0;
}

L
Lukas Czerner 已提交
3391

3392
static ext4_fsblk_t ext4_calculate_resv_clusters(struct super_block *sb)
L
Lukas Czerner 已提交
3393 3394 3395
{
	ext4_fsblk_t resv_clusters;

3396 3397 3398 3399 3400 3401 3402 3403
	/*
	 * There's no need to reserve anything when we aren't using extents.
	 * The space estimates are exact, there are no unwritten extents,
	 * hole punching doesn't need new metadata... This is needed especially
	 * to keep ext2/3 backward compatibility.
	 */
	if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS))
		return 0;
L
Lukas Czerner 已提交
3404 3405 3406 3407
	/*
	 * 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
3408
	 * unwritten extents in delalloc path. In most cases such
L
Lukas Czerner 已提交
3409 3410 3411
	 * allocation would require 1, or 2 blocks, higher numbers are
	 * very rare.
	 */
3412 3413
	resv_clusters = ext4_blocks_count(EXT4_SB(sb)->s_es) >>
			EXT4_SB(sb)->s_cluster_bits;
L
Lukas Czerner 已提交
3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433

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

	return resv_clusters;
}


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

	if (count >= clusters)
		return -EINVAL;

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

3434
static int ext4_fill_super(struct super_block *sb, void *data, int silent)
3435
{
3436
	char *orig_data = kstrdup(data, GFP_KERNEL);
3437
	struct buffer_head *bh;
3438 3439 3440 3441
	struct ext4_super_block *es = NULL;
	struct ext4_sb_info *sbi;
	ext4_fsblk_t block;
	ext4_fsblk_t sb_block = get_sb_block(&data);
3442
	ext4_fsblk_t logical_sb_block;
3443 3444 3445 3446
	unsigned long offset = 0;
	unsigned long journal_devnum = 0;
	unsigned long def_mount_opts;
	struct inode *root;
3447
	char *cp;
3448
	const char *descr;
3449
	int ret = -ENOMEM;
3450
	int blocksize, clustersize;
3451 3452
	unsigned int db_count;
	unsigned int i;
3453
	int needs_recovery, has_huge_files, has_bigalloc;
L
Laurent Vivier 已提交
3454
	__u64 blocks_count;
3455
	int err = 0;
3456
	unsigned int journal_ioprio = DEFAULT_JOURNAL_IOPRIO;
3457
	ext4_group_t first_not_zeroed;
3458 3459 3460

	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
	if (!sbi)
3461
		goto out_free_orig;
3462 3463 3464 3465 3466

	sbi->s_blockgroup_lock =
		kzalloc(sizeof(struct blockgroup_lock), GFP_KERNEL);
	if (!sbi->s_blockgroup_lock) {
		kfree(sbi);
3467
		goto out_free_orig;
3468
	}
3469
	sb->s_fs_info = sbi;
3470
	sbi->s_sb = sb;
3471
	sbi->s_inode_readahead_blks = EXT4_DEF_INODE_READAHEAD_BLKS;
M
Miklos Szeredi 已提交
3472
	sbi->s_sb_block = sb_block;
3473 3474 3475
	if (sb->s_bdev->bd_part)
		sbi->s_sectors_written_start =
			part_stat_read(sb->s_bdev->bd_part, sectors[1]);
3476

3477 3478 3479 3480
	/* Cleanup superblock name */
	for (cp = sb->s_id; (cp = strchr(cp, '/'));)
		*cp = '!';

3481
	/* -EINVAL is default */
3482
	ret = -EINVAL;
3483
	blocksize = sb_min_blocksize(sb, EXT4_MIN_BLOCK_SIZE);
3484
	if (!blocksize) {
3485
		ext4_msg(sb, KERN_ERR, "unable to set blocksize");
3486 3487 3488 3489
		goto out_fail;
	}

	/*
3490
	 * The ext4 superblock will not be buffer aligned for other than 1kB
3491 3492
	 * block sizes.  We need to calculate the offset from buffer start.
	 */
3493
	if (blocksize != EXT4_MIN_BLOCK_SIZE) {
3494 3495
		logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE;
		offset = do_div(logical_sb_block, blocksize);
3496
	} else {
3497
		logical_sb_block = sb_block;
3498 3499
	}

3500
	if (!(bh = sb_bread_unmovable(sb, logical_sb_block))) {
3501
		ext4_msg(sb, KERN_ERR, "unable to read superblock");
3502 3503 3504 3505
		goto out_fail;
	}
	/*
	 * Note: s_es must be initialized as soon as possible because
3506
	 *       some ext4 macro-instructions depend on its value
3507
	 */
3508
	es = (struct ext4_super_block *) (bh->b_data + offset);
3509 3510
	sbi->s_es = es;
	sb->s_magic = le16_to_cpu(es->s_magic);
3511 3512
	if (sb->s_magic != EXT4_SUPER_MAGIC)
		goto cantfind_ext4;
T
Theodore Ts'o 已提交
3513
	sbi->s_kbytes_written = le64_to_cpu(es->s_kbytes_written);
3514

3515 3516 3517 3518
	/* Warn if metadata_csum and gdt_csum are both set. */
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
				       EXT4_FEATURE_RO_COMPAT_METADATA_CSUM) &&
	    EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_GDT_CSUM))
3519
		ext4_warning(sb, "metadata_csum and uninit_bg are "
3520 3521
			     "redundant flags; please run fsck.");

3522 3523 3524 3525 3526 3527 3528 3529
	/* 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;
	}

3530 3531 3532 3533 3534 3535 3536 3537 3538 3539 3540 3541
	/* Load the checksum driver */
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
				       EXT4_FEATURE_RO_COMPAT_METADATA_CSUM)) {
		sbi->s_chksum_driver = crypto_alloc_shash("crc32c", 0, 0);
		if (IS_ERR(sbi->s_chksum_driver)) {
			ext4_msg(sb, KERN_ERR, "Cannot load crc32c driver.");
			ret = PTR_ERR(sbi->s_chksum_driver);
			sbi->s_chksum_driver = NULL;
			goto failed_mount;
		}
	}

3542 3543 3544 3545 3546 3547 3548 3549 3550
	/* Check superblock checksum */
	if (!ext4_superblock_csum_verify(sb, es)) {
		ext4_msg(sb, KERN_ERR, "VFS: Found ext4 filesystem with "
			 "invalid superblock checksum.  Run e2fsck?");
		silent = 1;
		goto cantfind_ext4;
	}

	/* Precompute checksum seed for all metadata */
3551
	if (ext4_has_metadata_csum(sb))
3552 3553 3554
		sbi->s_csum_seed = ext4_chksum(sbi, ~0, es->s_uuid,
					       sizeof(es->s_uuid));

3555 3556
	/* Set defaults before we parse the mount options */
	def_mount_opts = le32_to_cpu(es->s_default_mount_opts);
3557
	set_opt(sb, INIT_INODE_TABLE);
3558
	if (def_mount_opts & EXT4_DEFM_DEBUG)
3559
		set_opt(sb, DEBUG);
3560
	if (def_mount_opts & EXT4_DEFM_BSDGROUPS)
3561
		set_opt(sb, GRPID);
3562
	if (def_mount_opts & EXT4_DEFM_UID16)
3563
		set_opt(sb, NO_UID32);
3564 3565
	/* xattr user namespace & acls are now defaulted on */
	set_opt(sb, XATTR_USER);
T
Theodore Ts'o 已提交
3566
#ifdef CONFIG_EXT4_FS_POSIX_ACL
3567
	set_opt(sb, POSIX_ACL);
3568
#endif
3569 3570 3571 3572
	/* don't forget to enable journal_csum when metadata_csum is enabled. */
	if (ext4_has_metadata_csum(sb))
		set_opt(sb, JOURNAL_CHECKSUM);

3573
	if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_DATA)
3574
		set_opt(sb, JOURNAL_DATA);
3575
	else if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_ORDERED)
3576
		set_opt(sb, ORDERED_DATA);
3577
	else if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_WBACK)
3578
		set_opt(sb, WRITEBACK_DATA);
3579 3580

	if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_PANIC)
3581
		set_opt(sb, ERRORS_PANIC);
3582
	else if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_CONTINUE)
3583
		set_opt(sb, ERRORS_CONT);
3584
	else
3585
		set_opt(sb, ERRORS_RO);
3586 3587
	/* block_validity enabled by default; disable with noblock_validity */
	set_opt(sb, BLOCK_VALIDITY);
3588
	if (def_mount_opts & EXT4_DEFM_DISCARD)
3589
		set_opt(sb, DISCARD);
3590

3591 3592
	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));
3593 3594 3595
	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;
3596

3597
	if ((def_mount_opts & EXT4_DEFM_NOBARRIER) == 0)
3598
		set_opt(sb, BARRIER);
3599

3600 3601 3602 3603
	/*
	 * enable delayed allocation by default
	 * Use -o nodelalloc to turn it off
	 */
3604
	if (!IS_EXT3_SB(sb) && !IS_EXT2_SB(sb) &&
3605
	    ((def_mount_opts & EXT4_DEFM_NODELALLOC) == 0))
3606
		set_opt(sb, DELALLOC);
3607

3608 3609 3610 3611 3612 3613
	/*
	 * 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;

3614
	if (!parse_options((char *) sbi->s_es->s_mount_opts, sb,
3615
			   &journal_devnum, &journal_ioprio, 0)) {
3616 3617 3618 3619
		ext4_msg(sb, KERN_WARNING,
			 "failed to parse options in superblock: %s",
			 sbi->s_es->s_mount_opts);
	}
3620
	sbi->s_def_mount_opt = sbi->s_mount_opt;
3621
	if (!parse_options((char *) data, sb, &journal_devnum,
3622
			   &journal_ioprio, 0))
3623 3624
		goto failed_mount;

3625 3626 3627 3628 3629 3630 3631 3632 3633 3634 3635
	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 "
3636
				 "both data=journal and dioread_nolock");
3637 3638
			goto failed_mount;
		}
R
Ross Zwisler 已提交
3639 3640 3641 3642 3643
		if (test_opt(sb, DAX)) {
			ext4_msg(sb, KERN_ERR, "can't mount with "
				 "both data=journal and dax");
			goto failed_mount;
		}
3644 3645 3646 3647
		if (test_opt(sb, DELALLOC))
			clear_opt(sb, DELALLOC);
	}

3648
	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
3649
		(test_opt(sb, POSIX_ACL) ? MS_POSIXACL : 0);
3650

3651 3652 3653 3654
	if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV &&
	    (EXT4_HAS_COMPAT_FEATURE(sb, ~0U) ||
	     EXT4_HAS_RO_COMPAT_FEATURE(sb, ~0U) ||
	     EXT4_HAS_INCOMPAT_FEATURE(sb, ~0U)))
3655 3656 3657
		ext4_msg(sb, KERN_WARNING,
		       "feature flags set on rev 0 fs, "
		       "running e2fsck is recommended");
3658

3659 3660 3661 3662 3663 3664 3665 3666 3667 3668
	if (es->s_creator_os == cpu_to_le32(EXT4_OS_HURD)) {
		set_opt2(sb, HURD_COMPAT);
		if (EXT4_HAS_INCOMPAT_FEATURE(sb,
					      EXT4_FEATURE_INCOMPAT_64BIT)) {
			ext4_msg(sb, KERN_ERR,
				 "The Hurd can't support 64-bit file systems");
			goto failed_mount;
		}
	}

3669 3670 3671 3672 3673 3674 3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690
	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;
		}
	}

3691 3692 3693 3694 3695
	/*
	 * 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.
	 */
3696
	if (!ext4_feature_set_ok(sb, (sb->s_flags & MS_RDONLY)))
3697
		goto failed_mount;
3698

J
Jan Kara 已提交
3699
	blocksize = BLOCK_SIZE << le32_to_cpu(es->s_log_block_size);
3700 3701
	if (blocksize < EXT4_MIN_BLOCK_SIZE ||
	    blocksize > EXT4_MAX_BLOCK_SIZE) {
3702 3703
		ext4_msg(sb, KERN_ERR,
		       "Unsupported filesystem blocksize %d", blocksize);
3704 3705 3706
		goto failed_mount;
	}

R
Ross Zwisler 已提交
3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719
	if (sbi->s_mount_opt & EXT4_MOUNT_DAX) {
		if (blocksize != PAGE_SIZE) {
			ext4_msg(sb, KERN_ERR,
					"error: unsupported blocksize for dax");
			goto failed_mount;
		}
		if (!sb->s_bdev->bd_disk->fops->direct_access) {
			ext4_msg(sb, KERN_ERR,
					"error: device does not support dax");
			goto failed_mount;
		}
	}

3720 3721 3722 3723 3724 3725 3726
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_ENCRYPT) &&
	    es->s_encryption_level) {
		ext4_msg(sb, KERN_ERR, "Unsupported encryption level %d",
			 es->s_encryption_level);
		goto failed_mount;
	}

3727
	if (sb->s_blocksize != blocksize) {
3728 3729
		/* Validate the filesystem blocksize */
		if (!sb_set_blocksize(sb, blocksize)) {
3730
			ext4_msg(sb, KERN_ERR, "bad block size %d",
3731
					blocksize);
3732 3733 3734
			goto failed_mount;
		}

3735
		brelse(bh);
3736 3737
		logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE;
		offset = do_div(logical_sb_block, blocksize);
3738
		bh = sb_bread_unmovable(sb, logical_sb_block);
3739
		if (!bh) {
3740 3741
			ext4_msg(sb, KERN_ERR,
			       "Can't read superblock on 2nd try");
3742 3743
			goto failed_mount;
		}
3744
		es = (struct ext4_super_block *)(bh->b_data + offset);
3745
		sbi->s_es = es;
3746
		if (es->s_magic != cpu_to_le16(EXT4_SUPER_MAGIC)) {
3747 3748
			ext4_msg(sb, KERN_ERR,
			       "Magic mismatch, very weird!");
3749 3750 3751 3752
			goto failed_mount;
		}
	}

3753 3754
	has_huge_files = EXT4_HAS_RO_COMPAT_FEATURE(sb,
				EXT4_FEATURE_RO_COMPAT_HUGE_FILE);
3755 3756 3757
	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);
3758

3759 3760 3761
	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;
3762 3763 3764
	} else {
		sbi->s_inode_size = le16_to_cpu(es->s_inode_size);
		sbi->s_first_ino = le32_to_cpu(es->s_first_ino);
3765
		if ((sbi->s_inode_size < EXT4_GOOD_OLD_INODE_SIZE) ||
V
Vignesh Babu 已提交
3766
		    (!is_power_of_2(sbi->s_inode_size)) ||
3767
		    (sbi->s_inode_size > blocksize)) {
3768 3769
			ext4_msg(sb, KERN_ERR,
			       "unsupported inode size: %d",
3770
			       sbi->s_inode_size);
3771 3772
			goto failed_mount;
		}
K
Kalpak Shah 已提交
3773 3774
		if (sbi->s_inode_size > EXT4_GOOD_OLD_INODE_SIZE)
			sb->s_time_gran = 1 << (EXT4_EPOCH_BITS - 2);
3775
	}
3776

3777 3778
	sbi->s_desc_size = le16_to_cpu(es->s_desc_size);
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_64BIT)) {
3779
		if (sbi->s_desc_size < EXT4_MIN_DESC_SIZE_64BIT ||
3780
		    sbi->s_desc_size > EXT4_MAX_DESC_SIZE ||
V
vignesh babu 已提交
3781
		    !is_power_of_2(sbi->s_desc_size)) {
3782 3783
			ext4_msg(sb, KERN_ERR,
			       "unsupported descriptor size %lu",
3784 3785 3786 3787 3788
			       sbi->s_desc_size);
			goto failed_mount;
		}
	} else
		sbi->s_desc_size = EXT4_MIN_DESC_SIZE;
3789

3790 3791
	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);
3792
	if (EXT4_INODE_SIZE(sb) == 0 || EXT4_INODES_PER_GROUP(sb) == 0)
3793
		goto cantfind_ext4;
3794

3795
	sbi->s_inodes_per_block = blocksize / EXT4_INODE_SIZE(sb);
3796
	if (sbi->s_inodes_per_block == 0)
3797
		goto cantfind_ext4;
3798 3799
	sbi->s_itb_per_group = sbi->s_inodes_per_group /
					sbi->s_inodes_per_block;
3800
	sbi->s_desc_per_block = blocksize / EXT4_DESC_SIZE(sb);
3801 3802
	sbi->s_sbh = bh;
	sbi->s_mount_state = le16_to_cpu(es->s_state);
3803 3804
	sbi->s_addr_per_block_bits = ilog2(EXT4_ADDR_PER_BLOCK(sb));
	sbi->s_desc_per_block_bits = ilog2(EXT4_DESC_PER_BLOCK(sb));
3805

3806
	for (i = 0; i < 4; i++)
3807 3808
		sbi->s_hash_seed[i] = le32_to_cpu(es->s_hash_seed[i]);
	sbi->s_def_hash_version = es->s_def_hash_version;
3809 3810 3811 3812 3813
	if (EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_DIR_INDEX)) {
		i = le32_to_cpu(es->s_flags);
		if (i & EXT2_FLAGS_UNSIGNED_HASH)
			sbi->s_hash_unsigned = 3;
		else if ((i & EXT2_FLAGS_SIGNED_HASH) == 0) {
3814
#ifdef __CHAR_UNSIGNED__
3815 3816 3817 3818
			if (!(sb->s_flags & MS_RDONLY))
				es->s_flags |=
					cpu_to_le32(EXT2_FLAGS_UNSIGNED_HASH);
			sbi->s_hash_unsigned = 3;
3819
#else
3820 3821 3822
			if (!(sb->s_flags & MS_RDONLY))
				es->s_flags |=
					cpu_to_le32(EXT2_FLAGS_SIGNED_HASH);
3823
#endif
3824
		}
3825
	}
3826

3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870
	/* Handle clustersize */
	clustersize = BLOCK_SIZE << le32_to_cpu(es->s_log_cluster_size);
	has_bigalloc = EXT4_HAS_RO_COMPAT_FEATURE(sb,
				EXT4_FEATURE_RO_COMPAT_BIGALLOC);
	if (has_bigalloc) {
		if (clustersize < blocksize) {
			ext4_msg(sb, KERN_ERR,
				 "cluster size (%d) smaller than "
				 "block size (%d)", clustersize, blocksize);
			goto failed_mount;
		}
		sbi->s_cluster_bits = le32_to_cpu(es->s_log_cluster_size) -
			le32_to_cpu(es->s_log_block_size);
		sbi->s_clusters_per_group =
			le32_to_cpu(es->s_clusters_per_group);
		if (sbi->s_clusters_per_group > blocksize * 8) {
			ext4_msg(sb, KERN_ERR,
				 "#clusters per group too big: %lu",
				 sbi->s_clusters_per_group);
			goto failed_mount;
		}
		if (sbi->s_blocks_per_group !=
		    (sbi->s_clusters_per_group * (clustersize / blocksize))) {
			ext4_msg(sb, KERN_ERR, "blocks per group (%lu) and "
				 "clusters per group (%lu) inconsistent",
				 sbi->s_blocks_per_group,
				 sbi->s_clusters_per_group);
			goto failed_mount;
		}
	} else {
		if (clustersize != blocksize) {
			ext4_warning(sb, "fragment/cluster size (%d) != "
				     "block size (%d)", clustersize,
				     blocksize);
			clustersize = blocksize;
		}
		if (sbi->s_blocks_per_group > blocksize * 8) {
			ext4_msg(sb, KERN_ERR,
				 "#blocks per group too big: %lu",
				 sbi->s_blocks_per_group);
			goto failed_mount;
		}
		sbi->s_clusters_per_group = sbi->s_blocks_per_group;
		sbi->s_cluster_bits = 0;
3871
	}
3872 3873
	sbi->s_cluster_ratio = clustersize / blocksize;

3874
	if (sbi->s_inodes_per_group > blocksize * 8) {
3875 3876
		ext4_msg(sb, KERN_ERR,
		       "#inodes per group too big: %lu",
3877
		       sbi->s_inodes_per_group);
3878 3879 3880
		goto failed_mount;
	}

3881 3882 3883 3884
	/* Do we have standard group size of clustersize * 8 blocks ? */
	if (sbi->s_blocks_per_group == clustersize << 3)
		set_opt2(sb, STD_GROUP_SIZE);

3885 3886 3887 3888
	/*
	 * Test whether we have more sectors than will fit in sector_t,
	 * and whether the max offset is addressable by the page cache.
	 */
3889
	err = generic_check_addressable(sb->s_blocksize_bits,
3890
					ext4_blocks_count(es));
3891
	if (err) {
3892
		ext4_msg(sb, KERN_ERR, "filesystem"
3893
			 " too large to mount safely on this system");
3894
		if (sizeof(sector_t) < 8)
3895
			ext4_msg(sb, KERN_WARNING, "CONFIG_LBDAF not enabled");
3896 3897 3898
		goto failed_mount;
	}

3899 3900
	if (EXT4_BLOCKS_PER_GROUP(sb) == 0)
		goto cantfind_ext4;
3901

3902 3903 3904
	/* 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) {
3905 3906
		ext4_msg(sb, KERN_WARNING, "bad geometry: block count %llu "
		       "exceeds size of device (%llu blocks)",
3907 3908 3909 3910
		       ext4_blocks_count(es), blocks_count);
		goto failed_mount;
	}

3911 3912 3913 3914 3915
	/*
	 * 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)) {
3916
		ext4_msg(sb, KERN_WARNING, "bad geometry: first data "
3917 3918 3919
			 "block %u is beyond end of filesystem (%llu)",
			 le32_to_cpu(es->s_first_data_block),
			 ext4_blocks_count(es));
3920 3921
		goto failed_mount;
	}
L
Laurent Vivier 已提交
3922 3923 3924 3925
	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));
3926
	if (blocks_count > ((uint64_t)1<<32) - EXT4_DESC_PER_BLOCK(sb)) {
3927
		ext4_msg(sb, KERN_WARNING, "groups count too large: %u "
3928
		       "(block count %llu, first data block %u, "
3929
		       "blocks per group %lu)", sbi->s_groups_count,
3930 3931 3932 3933 3934
		       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
3935
	sbi->s_groups_count = blocks_count;
3936 3937
	sbi->s_blockfile_groups = min_t(ext4_group_t, sbi->s_groups_count,
			(EXT4_MAX_BLOCK_FILE_PHYS / EXT4_BLOCKS_PER_GROUP(sb)));
3938 3939
	db_count = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) /
		   EXT4_DESC_PER_BLOCK(sb);
3940 3941 3942
	sbi->s_group_desc = ext4_kvmalloc(db_count *
					  sizeof(struct buffer_head *),
					  GFP_KERNEL);
3943
	if (sbi->s_group_desc == NULL) {
3944
		ext4_msg(sb, KERN_ERR, "not enough memory");
3945
		ret = -ENOMEM;
3946 3947 3948
		goto failed_mount;
	}

3949 3950
	if (ext4_proc_root)
		sbi->s_proc = proc_mkdir(sb->s_id, ext4_proc_root);
3951

3952 3953 3954 3955
	if (sbi->s_proc)
		proc_create_data("options", S_IRUGO, sbi->s_proc,
				 &ext4_seq_options_fops, sb);

3956
	bgl_lock_init(sbi->s_blockgroup_lock);
3957 3958

	for (i = 0; i < db_count; i++) {
3959
		block = descriptor_loc(sb, logical_sb_block, i);
3960
		sbi->s_group_desc[i] = sb_bread_unmovable(sb, block);
3961
		if (!sbi->s_group_desc[i]) {
3962 3963
			ext4_msg(sb, KERN_ERR,
			       "can't read group descriptor %d", i);
3964 3965 3966 3967
			db_count = i;
			goto failed_mount2;
		}
	}
3968
	if (!ext4_check_descriptors(sb, &first_not_zeroed)) {
3969
		ext4_msg(sb, KERN_ERR, "group descriptors corrupted!");
3970
		goto failed_mount2;
3971
	}
3972

3973
	sbi->s_gdb_count = db_count;
3974 3975 3976
	get_random_bytes(&sbi->s_next_generation, sizeof(u32));
	spin_lock_init(&sbi->s_next_gen_lock);

3977 3978
	setup_timer(&sbi->s_err_report, print_daily_error_info,
		(unsigned long) sb);
3979

3980
	/* Register extent status tree shrinker */
3981
	if (ext4_es_register_shrinker(sbi))
3982 3983
		goto failed_mount3;

3984
	sbi->s_stripe = ext4_get_stripe_size(sbi);
3985
	sbi->s_extent_max_zeroout_kb = 32;
3986

3987 3988 3989
	/*
	 * set up enough so that it can read an inode
	 */
3990
	sb->s_op = &ext4_sops;
3991 3992
	sb->s_export_op = &ext4_export_ops;
	sb->s_xattr = ext4_xattr_handlers;
3993
#ifdef CONFIG_QUOTA
3994
	sb->dq_op = &ext4_quota_operations;
3995
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_QUOTA))
3996
		sb->s_qcop = &dquot_quotactl_sysfile_ops;
3997 3998
	else
		sb->s_qcop = &ext4_qctl_operations;
J
Jan Kara 已提交
3999
	sb->s_quota_types = QTYPE_MASK_USR | QTYPE_MASK_GRP;
4000
#endif
4001 4002
	memcpy(sb->s_uuid, es->s_uuid, sizeof(es->s_uuid));

4003
	INIT_LIST_HEAD(&sbi->s_orphan); /* unlinked but open files */
4004
	mutex_init(&sbi->s_orphan_lock);
4005 4006 4007 4008

	sb->s_root = NULL;

	needs_recovery = (es->s_last_orphan != 0 ||
4009 4010
			  EXT4_HAS_INCOMPAT_FEATURE(sb,
				    EXT4_FEATURE_INCOMPAT_RECOVER));
4011

4012 4013 4014
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_MMP) &&
	    !(sb->s_flags & MS_RDONLY))
		if (ext4_multi_mount_protect(sb, le64_to_cpu(es->s_mmp_block)))
4015
			goto failed_mount3a;
4016

4017 4018 4019 4020 4021
	/*
	 * The first inode we look at is the journal inode.  Don't try
	 * root first: it may be modified in the journal!
	 */
	if (!test_opt(sb, NOLOAD) &&
4022 4023
	    EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL)) {
		if (ext4_load_journal(sb, es, journal_devnum))
4024
			goto failed_mount3a;
4025 4026
	} else if (test_opt(sb, NOLOAD) && !(sb->s_flags & MS_RDONLY) &&
	      EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER)) {
4027 4028
		ext4_msg(sb, KERN_ERR, "required journal recovery "
		       "suppressed and not mounted read-only");
4029
		goto failed_mount_wq;
4030
	} else {
4031
		clear_opt(sb, DATA_FLAGS);
4032 4033 4034
		sbi->s_journal = NULL;
		needs_recovery = 0;
		goto no_journal;
4035 4036
	}

4037
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_64BIT) &&
4038 4039
	    !jbd2_journal_set_features(EXT4_SB(sb)->s_journal, 0, 0,
				       JBD2_FEATURE_INCOMPAT_64BIT)) {
4040
		ext4_msg(sb, KERN_ERR, "Failed to set 64-bit journal feature");
4041
		goto failed_mount_wq;
4042 4043
	}

4044 4045 4046 4047
	if (!set_journal_csum_feature_set(sb)) {
		ext4_msg(sb, KERN_ERR, "Failed to set journal checksum "
			 "feature set");
		goto failed_mount_wq;
4048
	}
4049

4050 4051 4052 4053 4054
	/* 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
4055 4056 4057
		 * capabilities: ORDERED_DATA if the journal can
		 * cope, else JOURNAL_DATA
		 */
4058 4059
		if (jbd2_journal_check_available_features
		    (sbi->s_journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE))
4060
			set_opt(sb, ORDERED_DATA);
4061
		else
4062
			set_opt(sb, JOURNAL_DATA);
4063 4064
		break;

4065 4066
	case EXT4_MOUNT_ORDERED_DATA:
	case EXT4_MOUNT_WRITEBACK_DATA:
4067 4068
		if (!jbd2_journal_check_available_features
		    (sbi->s_journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE)) {
4069 4070
			ext4_msg(sb, KERN_ERR, "Journal does not support "
			       "requested data journaling mode");
4071
			goto failed_mount_wq;
4072 4073 4074 4075
		}
	default:
		break;
	}
4076
	set_task_ioprio(sbi->s_journal->j_task, journal_ioprio);
4077

B
Bobi Jam 已提交
4078 4079
	sbi->s_journal->j_commit_callback = ext4_journal_commit_callback;

4080
no_journal:
4081 4082 4083 4084 4085 4086 4087 4088
	if (ext4_mballoc_ready) {
		sbi->s_mb_cache = ext4_xattr_create_cache(sb->s_id);
		if (!sbi->s_mb_cache) {
			ext4_msg(sb, KERN_ERR, "Failed to create an mb_cache");
			goto failed_mount_wq;
		}
	}

4089 4090 4091 4092 4093 4094 4095 4096 4097
	if ((DUMMY_ENCRYPTION_ENABLED(sbi) ||
	     EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_ENCRYPT)) &&
	    (blocksize != PAGE_CACHE_SIZE)) {
		ext4_msg(sb, KERN_ERR,
			 "Unsupported blocksize for fs encryption");
		goto failed_mount_wq;
	}

	if (DUMMY_ENCRYPTION_ENABLED(sbi) &&
4098 4099 4100 4101 4102 4103
	    !(sb->s_flags & MS_RDONLY) &&
	    !EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_ENCRYPT)) {
		EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_ENCRYPT);
		ext4_commit_super(sb, 1);
	}

4104 4105 4106 4107 4108 4109 4110
	/*
	 * 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 {
4111 4112
		err = ext4_calculate_overhead(sb);
		if (err)
4113 4114 4115
			goto failed_mount_wq;
	}

T
Tejun Heo 已提交
4116 4117 4118 4119
	/*
	 * The maximum number of concurrent works can be high and
	 * concurrency isn't really necessary.  Limit it to 1.
	 */
4120 4121 4122 4123
	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");
4124
		ret = -ENOMEM;
4125 4126 4127
		goto failed_mount4;
	}

4128
	/*
4129
	 * The jbd2_journal_load will have done any necessary log recovery,
4130 4131 4132
	 * so we can safely mount the rest of the filesystem now.
	 */

4133 4134
	root = ext4_iget(sb, EXT4_ROOT_INO);
	if (IS_ERR(root)) {
4135
		ext4_msg(sb, KERN_ERR, "get root inode failed");
4136
		ret = PTR_ERR(root);
4137
		root = NULL;
4138 4139 4140
		goto failed_mount4;
	}
	if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) {
4141
		ext4_msg(sb, KERN_ERR, "corrupt root inode, run e2fsck");
A
Al Viro 已提交
4142
		iput(root);
4143 4144
		goto failed_mount4;
	}
4145
	sb->s_root = d_make_root(root);
4146
	if (!sb->s_root) {
4147
		ext4_msg(sb, KERN_ERR, "get root dentry failed");
4148 4149 4150
		ret = -ENOMEM;
		goto failed_mount4;
	}
4151

4152 4153
	if (ext4_setup_super(sb, es, sb->s_flags & MS_RDONLY))
		sb->s_flags |= MS_RDONLY;
K
Kalpak Shah 已提交
4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170 4171 4172 4173 4174 4175

	/* determine the minimum size of new large inodes, if present */
	if (sbi->s_inode_size > EXT4_GOOD_OLD_INODE_SIZE) {
		sbi->s_want_extra_isize = sizeof(struct ext4_inode) -
						     EXT4_GOOD_OLD_INODE_SIZE;
		if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
				       EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE)) {
			if (sbi->s_want_extra_isize <
			    le16_to_cpu(es->s_want_extra_isize))
				sbi->s_want_extra_isize =
					le16_to_cpu(es->s_want_extra_isize);
			if (sbi->s_want_extra_isize <
			    le16_to_cpu(es->s_min_extra_isize))
				sbi->s_want_extra_isize =
					le16_to_cpu(es->s_min_extra_isize);
		}
	}
	/* Check if enough inode space is available */
	if (EXT4_GOOD_OLD_INODE_SIZE + sbi->s_want_extra_isize >
							sbi->s_inode_size) {
		sbi->s_want_extra_isize = sizeof(struct ext4_inode) -
						       EXT4_GOOD_OLD_INODE_SIZE;
4176 4177
		ext4_msg(sb, KERN_INFO, "required extra inode space not"
			 "available");
K
Kalpak Shah 已提交
4178 4179
	}

4180
	err = ext4_reserve_clusters(sbi, ext4_calculate_resv_clusters(sb));
L
Lukas Czerner 已提交
4181 4182
	if (err) {
		ext4_msg(sb, KERN_ERR, "failed to reserve %llu clusters for "
4183
			 "reserved pool", ext4_calculate_resv_clusters(sb));
4184
		goto failed_mount4a;
L
Lukas Czerner 已提交
4185 4186
	}

4187 4188
	err = ext4_setup_system_zone(sb);
	if (err) {
4189
		ext4_msg(sb, KERN_ERR, "failed to initialize system "
4190
			 "zone (%d)", err);
4191 4192 4193 4194 4195 4196 4197 4198
		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);
4199
		goto failed_mount5;
4200 4201
	}

4202 4203 4204
	block = ext4_count_free_clusters(sb);
	ext4_free_blocks_count_set(sbi->s_es, 
				   EXT4_C2B(sbi, block));
4205 4206
	err = percpu_counter_init(&sbi->s_freeclusters_counter, block,
				  GFP_KERNEL);
4207 4208 4209
	if (!err) {
		unsigned long freei = ext4_count_free_inodes(sb);
		sbi->s_es->s_free_inodes_count = cpu_to_le32(freei);
4210 4211
		err = percpu_counter_init(&sbi->s_freeinodes_counter, freei,
					  GFP_KERNEL);
4212 4213 4214
	}
	if (!err)
		err = percpu_counter_init(&sbi->s_dirs_counter,
4215
					  ext4_count_dirs(sb), GFP_KERNEL);
4216
	if (!err)
4217 4218
		err = percpu_counter_init(&sbi->s_dirtyclusters_counter, 0,
					  GFP_KERNEL);
4219 4220 4221 4222 4223 4224 4225 4226 4227 4228 4229 4230 4231
	if (err) {
		ext4_msg(sb, KERN_ERR, "insufficient memory");
		goto failed_mount6;
	}

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

4232 4233
	err = ext4_register_li_request(sb, first_not_zeroed);
	if (err)
4234
		goto failed_mount6;
4235

T
Theodore Ts'o 已提交
4236 4237 4238 4239
	sbi->s_kobj.kset = ext4_kset;
	init_completion(&sbi->s_kobj_unregister);
	err = kobject_init_and_add(&sbi->s_kobj, &ext4_ktype, NULL,
				   "%s", sb->s_id);
4240 4241
	if (err)
		goto failed_mount7;
T
Theodore Ts'o 已提交
4242

4243 4244 4245 4246 4247 4248 4249 4250 4251 4252
#ifdef CONFIG_QUOTA
	/* Enable quota usage during mount. */
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_QUOTA) &&
	    !(sb->s_flags & MS_RDONLY)) {
		err = ext4_enable_quotas(sb);
		if (err)
			goto failed_mount8;
	}
#endif  /* CONFIG_QUOTA */

4253 4254 4255
	EXT4_SB(sb)->s_mount_state |= EXT4_ORPHAN_FS;
	ext4_orphan_cleanup(sb, es);
	EXT4_SB(sb)->s_mount_state &= ~EXT4_ORPHAN_FS;
4256
	if (needs_recovery) {
4257
		ext4_msg(sb, KERN_INFO, "recovery complete");
4258 4259 4260 4261 4262 4263 4264 4265 4266 4267 4268 4269
		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";

4270 4271 4272 4273 4274 4275 4276 4277
	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");
	}

4278
	ext4_msg(sb, KERN_INFO, "mounted filesystem with%s. "
4279 4280
		 "Opts: %s%s%s", descr, sbi->s_es->s_mount_opts,
		 *sbi->s_es->s_mount_opts ? "; " : "", orig_data);
4281

4282 4283
	if (es->s_error_count)
		mod_timer(&sbi->s_err_report, jiffies + 300*HZ); /* 5 minutes */
4284

4285 4286 4287 4288 4289
	/* 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);

4290
	kfree(orig_data);
4291 4292
	return 0;

4293
cantfind_ext4:
4294
	if (!silent)
4295
		ext4_msg(sb, KERN_ERR, "VFS: Can't find ext4 filesystem");
4296 4297
	goto failed_mount;

4298 4299 4300 4301
#ifdef CONFIG_QUOTA
failed_mount8:
	kobject_del(&sbi->s_kobj);
#endif
4302 4303 4304
failed_mount7:
	ext4_unregister_li_request(sb);
failed_mount6:
4305
	ext4_mb_release(sb);
4306
	if (sbi->s_flex_groups)
A
Al Viro 已提交
4307
		kvfree(sbi->s_flex_groups);
4308 4309 4310 4311
	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);
4312
failed_mount5:
4313 4314 4315
	ext4_ext_release(sb);
	ext4_release_system_zone(sb);
failed_mount4a:
A
Al Viro 已提交
4316
	dput(sb->s_root);
4317
	sb->s_root = NULL;
A
Al Viro 已提交
4318
failed_mount4:
4319
	ext4_msg(sb, KERN_ERR, "mount failed");
4320 4321
	if (EXT4_SB(sb)->rsv_conversion_wq)
		destroy_workqueue(EXT4_SB(sb)->rsv_conversion_wq);
4322
failed_mount_wq:
4323 4324 4325 4326
	if (sbi->s_journal) {
		jbd2_journal_destroy(sbi->s_journal);
		sbi->s_journal = NULL;
	}
4327
failed_mount3a:
4328
	ext4_es_unregister_shrinker(sbi);
4329
failed_mount3:
4330
	del_timer_sync(&sbi->s_err_report);
4331 4332
	if (sbi->s_mmp_tsk)
		kthread_stop(sbi->s_mmp_tsk);
4333 4334 4335
failed_mount2:
	for (i = 0; i < db_count; i++)
		brelse(sbi->s_group_desc[i]);
A
Al Viro 已提交
4336
	kvfree(sbi->s_group_desc);
4337
failed_mount:
4338 4339
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
4340
	if (sbi->s_proc) {
4341
		remove_proc_entry("options", sbi->s_proc);
4342
		remove_proc_entry(sb->s_id, ext4_proc_root);
4343
	}
4344
#ifdef CONFIG_QUOTA
J
Jan Kara 已提交
4345
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
4346 4347
		kfree(sbi->s_qf_names[i]);
#endif
4348
	ext4_blkdev_remove(sbi);
4349 4350 4351
	brelse(bh);
out_fail:
	sb->s_fs_info = NULL;
4352
	kfree(sbi->s_blockgroup_lock);
4353
	kfree(sbi);
4354
out_free_orig:
4355
	kfree(orig_data);
4356
	return err ? err : ret;
4357 4358 4359 4360 4361 4362 4363
}

/*
 * 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.
 */
4364
static void ext4_init_journal_params(struct super_block *sb, journal_t *journal)
4365
{
4366
	struct ext4_sb_info *sbi = EXT4_SB(sb);
4367

4368 4369 4370
	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;
4371

4372
	write_lock(&journal->j_state_lock);
4373
	if (test_opt(sb, BARRIER))
4374
		journal->j_flags |= JBD2_BARRIER;
4375
	else
4376
		journal->j_flags &= ~JBD2_BARRIER;
4377 4378 4379 4380
	if (test_opt(sb, DATA_ERR_ABORT))
		journal->j_flags |= JBD2_ABORT_ON_SYNCDATA_ERR;
	else
		journal->j_flags &= ~JBD2_ABORT_ON_SYNCDATA_ERR;
4381
	write_unlock(&journal->j_state_lock);
4382 4383
}

4384
static journal_t *ext4_get_journal(struct super_block *sb,
4385 4386 4387 4388 4389
				   unsigned int journal_inum)
{
	struct inode *journal_inode;
	journal_t *journal;

4390 4391
	BUG_ON(!EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL));

4392 4393 4394 4395
	/* First, test for the existence of a valid inode on disk.  Bad
	 * things happen if we iget() an unused inode, as the subsequent
	 * iput() will try to delete it. */

4396 4397
	journal_inode = ext4_iget(sb, journal_inum);
	if (IS_ERR(journal_inode)) {
4398
		ext4_msg(sb, KERN_ERR, "no journal found");
4399 4400 4401 4402 4403
		return NULL;
	}
	if (!journal_inode->i_nlink) {
		make_bad_inode(journal_inode);
		iput(journal_inode);
4404
		ext4_msg(sb, KERN_ERR, "journal inode is deleted");
4405 4406 4407
		return NULL;
	}

4408
	jbd_debug(2, "Journal inode found at %p: %lld bytes\n",
4409
		  journal_inode, journal_inode->i_size);
4410
	if (!S_ISREG(journal_inode->i_mode)) {
4411
		ext4_msg(sb, KERN_ERR, "invalid journal inode");
4412 4413 4414 4415
		iput(journal_inode);
		return NULL;
	}

4416
	journal = jbd2_journal_init_inode(journal_inode);
4417
	if (!journal) {
4418
		ext4_msg(sb, KERN_ERR, "Could not load journal inode");
4419 4420 4421 4422
		iput(journal_inode);
		return NULL;
	}
	journal->j_private = sb;
4423
	ext4_init_journal_params(sb, journal);
4424 4425 4426
	return journal;
}

4427
static journal_t *ext4_get_dev_journal(struct super_block *sb,
4428 4429
				       dev_t j_dev)
{
4430
	struct buffer_head *bh;
4431
	journal_t *journal;
4432 4433
	ext4_fsblk_t start;
	ext4_fsblk_t len;
4434
	int hblock, blocksize;
4435
	ext4_fsblk_t sb_block;
4436
	unsigned long offset;
4437
	struct ext4_super_block *es;
4438 4439
	struct block_device *bdev;

4440 4441
	BUG_ON(!EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL));

4442
	bdev = ext4_blkdev_get(j_dev, sb);
4443 4444 4445 4446
	if (bdev == NULL)
		return NULL;

	blocksize = sb->s_blocksize;
4447
	hblock = bdev_logical_block_size(bdev);
4448
	if (blocksize < hblock) {
4449 4450
		ext4_msg(sb, KERN_ERR,
			"blocksize too small for journal device");
4451 4452 4453
		goto out_bdev;
	}

4454 4455
	sb_block = EXT4_MIN_BLOCK_SIZE / blocksize;
	offset = EXT4_MIN_BLOCK_SIZE % blocksize;
4456 4457
	set_blocksize(bdev, blocksize);
	if (!(bh = __bread(bdev, sb_block, blocksize))) {
4458 4459
		ext4_msg(sb, KERN_ERR, "couldn't read superblock of "
		       "external journal");
4460 4461 4462
		goto out_bdev;
	}

4463
	es = (struct ext4_super_block *) (bh->b_data + offset);
4464
	if ((le16_to_cpu(es->s_magic) != EXT4_SUPER_MAGIC) ||
4465
	    !(le32_to_cpu(es->s_feature_incompat) &
4466
	      EXT4_FEATURE_INCOMPAT_JOURNAL_DEV)) {
4467 4468
		ext4_msg(sb, KERN_ERR, "external journal has "
					"bad superblock");
4469 4470 4471 4472
		brelse(bh);
		goto out_bdev;
	}

4473 4474 4475 4476 4477 4478 4479 4480 4481
	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;
	}

4482
	if (memcmp(EXT4_SB(sb)->s_es->s_journal_uuid, es->s_uuid, 16)) {
4483
		ext4_msg(sb, KERN_ERR, "journal UUID does not match");
4484 4485 4486 4487
		brelse(bh);
		goto out_bdev;
	}

L
Laurent Vivier 已提交
4488
	len = ext4_blocks_count(es);
4489 4490 4491
	start = sb_block + 1;
	brelse(bh);	/* we're done with the superblock */

4492
	journal = jbd2_journal_init_dev(bdev, sb->s_bdev,
4493 4494
					start, len, blocksize);
	if (!journal) {
4495
		ext4_msg(sb, KERN_ERR, "failed to create device journal");
4496 4497 4498
		goto out_bdev;
	}
	journal->j_private = sb;
4499
	ll_rw_block(READ | REQ_META | REQ_PRIO, 1, &journal->j_sb_buffer);
4500 4501
	wait_on_buffer(journal->j_sb_buffer);
	if (!buffer_uptodate(journal->j_sb_buffer)) {
4502
		ext4_msg(sb, KERN_ERR, "I/O error on journal device");
4503 4504 4505
		goto out_journal;
	}
	if (be32_to_cpu(journal->j_superblock->s_nr_users) != 1) {
4506 4507
		ext4_msg(sb, KERN_ERR, "External journal has more than one "
					"user (unsupported) - %d",
4508 4509 4510
			be32_to_cpu(journal->j_superblock->s_nr_users));
		goto out_journal;
	}
4511 4512
	EXT4_SB(sb)->journal_bdev = bdev;
	ext4_init_journal_params(sb, journal);
4513
	return journal;
4514

4515
out_journal:
4516
	jbd2_journal_destroy(journal);
4517
out_bdev:
4518
	ext4_blkdev_put(bdev);
4519 4520 4521
	return NULL;
}

4522 4523
static int ext4_load_journal(struct super_block *sb,
			     struct ext4_super_block *es,
4524 4525 4526 4527 4528 4529 4530 4531
			     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;

4532 4533
	BUG_ON(!EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL));

4534 4535
	if (journal_devnum &&
	    journal_devnum != le32_to_cpu(es->s_journal_dev)) {
4536 4537
		ext4_msg(sb, KERN_INFO, "external journal device major/minor "
			"numbers have changed");
4538 4539 4540 4541 4542 4543 4544 4545 4546 4547 4548
		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.
	 */
4549
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER)) {
4550
		if (sb->s_flags & MS_RDONLY) {
4551 4552
			ext4_msg(sb, KERN_INFO, "INFO: recovery "
					"required on readonly filesystem");
4553
			if (really_read_only) {
4554 4555
				ext4_msg(sb, KERN_ERR, "write access "
					"unavailable, cannot proceed");
4556 4557
				return -EROFS;
			}
4558 4559
			ext4_msg(sb, KERN_INFO, "write access will "
			       "be enabled during recovery");
4560 4561 4562 4563
		}
	}

	if (journal_inum && journal_dev) {
4564 4565
		ext4_msg(sb, KERN_ERR, "filesystem has both journal "
		       "and inode journals!");
4566 4567 4568 4569
		return -EINVAL;
	}

	if (journal_inum) {
4570
		if (!(journal = ext4_get_journal(sb, journal_inum)))
4571 4572
			return -EINVAL;
	} else {
4573
		if (!(journal = ext4_get_dev_journal(sb, journal_dev)))
4574 4575 4576
			return -EINVAL;
	}

4577
	if (!(journal->j_flags & JBD2_BARRIER))
4578
		ext4_msg(sb, KERN_INFO, "barriers disabled");
4579

4580
	if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER))
4581
		err = jbd2_journal_wipe(journal, !really_read_only);
4582 4583 4584 4585 4586
	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);
4587
		err = jbd2_journal_load(journal);
4588 4589 4590 4591 4592
		if (save)
			memcpy(((char *) es) + EXT4_S_ERR_START,
			       save, EXT4_S_ERR_LEN);
		kfree(save);
	}
4593 4594

	if (err) {
4595
		ext4_msg(sb, KERN_ERR, "error loading journal");
4596
		jbd2_journal_destroy(journal);
4597 4598 4599
		return err;
	}

4600 4601
	EXT4_SB(sb)->s_journal = journal;
	ext4_clear_journal_err(sb, es);
4602

4603
	if (!really_read_only && journal_devnum &&
4604 4605 4606 4607
	    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. */
4608
		ext4_commit_super(sb, 1);
4609 4610 4611 4612 4613
	}

	return 0;
}

4614
static int ext4_commit_super(struct super_block *sb, int sync)
4615
{
4616
	struct ext4_super_block *es = EXT4_SB(sb)->s_es;
4617
	struct buffer_head *sbh = EXT4_SB(sb)->s_sbh;
4618
	int error = 0;
4619

4620
	if (!sbh)
4621
		return error;
4622 4623 4624 4625 4626 4627 4628 4629 4630
	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.
		 */
4631 4632
		ext4_msg(sb, KERN_ERR, "previous I/O error to "
		       "superblock detected");
4633 4634 4635
		clear_buffer_write_io_error(sbh);
		set_buffer_uptodate(sbh);
	}
4636 4637 4638 4639 4640 4641 4642 4643 4644 4645 4646 4647
	/*
	 * 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());
4648 4649 4650
	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
4651 4652
			    ((part_stat_read(sb->s_bdev->bd_part, sectors[1]) -
			      EXT4_SB(sb)->s_sectors_written_start) >> 1));
4653 4654 4655
	else
		es->s_kbytes_written =
			cpu_to_le64(EXT4_SB(sb)->s_kbytes_written);
4656 4657
	if (percpu_counter_initialized(&EXT4_SB(sb)->s_freeclusters_counter))
		ext4_free_blocks_count_set(es,
4658 4659
			EXT4_C2B(EXT4_SB(sb), percpu_counter_sum_positive(
				&EXT4_SB(sb)->s_freeclusters_counter)));
4660 4661 4662
	if (percpu_counter_initialized(&EXT4_SB(sb)->s_freeinodes_counter))
		es->s_free_inodes_count =
			cpu_to_le32(percpu_counter_sum_positive(
4663
				&EXT4_SB(sb)->s_freeinodes_counter));
4664
	BUFFER_TRACE(sbh, "marking dirty");
4665
	ext4_superblock_csum_set(sb);
4666
	mark_buffer_dirty(sbh);
4667
	if (sync) {
4668 4669 4670 4671 4672 4673
		error = sync_dirty_buffer(sbh);
		if (error)
			return error;

		error = buffer_write_io_error(sbh);
		if (error) {
4674 4675
			ext4_msg(sb, KERN_ERR, "I/O error while writing "
			       "superblock");
4676 4677 4678 4679
			clear_buffer_write_io_error(sbh);
			set_buffer_uptodate(sbh);
		}
	}
4680
	return error;
4681 4682 4683 4684 4685 4686 4687
}

/*
 * 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.
 */
4688 4689
static void ext4_mark_recovery_complete(struct super_block *sb,
					struct ext4_super_block *es)
4690
{
4691
	journal_t *journal = EXT4_SB(sb)->s_journal;
4692

4693 4694 4695 4696
	if (!EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL)) {
		BUG_ON(journal != NULL);
		return;
	}
4697
	jbd2_journal_lock_updates(journal);
4698 4699 4700
	if (jbd2_journal_flush(journal) < 0)
		goto out;

4701
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER) &&
4702
	    sb->s_flags & MS_RDONLY) {
4703
		EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
4704
		ext4_commit_super(sb, 1);
4705
	}
4706 4707

out:
4708
	jbd2_journal_unlock_updates(journal);
4709 4710 4711 4712 4713 4714 4715
}

/*
 * 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.
 */
4716 4717
static void ext4_clear_journal_err(struct super_block *sb,
				   struct ext4_super_block *es)
4718 4719 4720 4721 4722
{
	journal_t *journal;
	int j_errno;
	const char *errstr;

4723 4724
	BUG_ON(!EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL));

4725
	journal = EXT4_SB(sb)->s_journal;
4726 4727 4728

	/*
	 * Now check for any error status which may have been recorded in the
4729
	 * journal by a prior ext4_error() or ext4_abort()
4730 4731
	 */

4732
	j_errno = jbd2_journal_errno(journal);
4733 4734 4735
	if (j_errno) {
		char nbuf[16];

4736
		errstr = ext4_decode_error(sb, j_errno, nbuf);
4737
		ext4_warning(sb, "Filesystem error recorded "
4738
			     "from previous mount: %s", errstr);
4739
		ext4_warning(sb, "Marking fs in need of filesystem check.");
4740

4741 4742
		EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS;
		es->s_state |= cpu_to_le16(EXT4_ERROR_FS);
4743
		ext4_commit_super(sb, 1);
4744

4745
		jbd2_journal_clear_err(journal);
4746
		jbd2_journal_update_sb_errno(journal);
4747 4748 4749 4750 4751 4752 4753
	}
}

/*
 * Force the running and committing transactions to commit,
 * and wait on the commit.
 */
4754
int ext4_force_commit(struct super_block *sb)
4755 4756 4757 4758 4759 4760
{
	journal_t *journal;

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

4761
	journal = EXT4_SB(sb)->s_journal;
4762
	return ext4_journal_force_commit(journal);
4763 4764
}

4765
static int ext4_sync_fs(struct super_block *sb, int wait)
4766
{
4767
	int ret = 0;
4768
	tid_t target;
4769
	bool needs_barrier = false;
4770
	struct ext4_sb_info *sbi = EXT4_SB(sb);
4771

4772
	trace_ext4_sync_fs(sb, wait);
4773
	flush_workqueue(sbi->rsv_conversion_wq);
4774 4775 4776 4777 4778
	/*
	 * Writeback quota in non-journalled quota case - journalled quota has
	 * no dirty dquots
	 */
	dquot_writeback_dquots(sb, -1);
4779 4780 4781 4782 4783
	/*
	 * 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.
	 */
4784 4785 4786 4787 4788 4789 4790 4791 4792 4793 4794 4795
	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))
4796 4797 4798 4799 4800 4801
		needs_barrier = true;
	if (needs_barrier) {
		int err;
		err = blkdev_issue_flush(sb->s_bdev, GFP_KERNEL, NULL);
		if (!ret)
			ret = err;
4802
	}
4803 4804 4805 4806

	return ret;
}

4807 4808 4809
/*
 * 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.
4810 4811
 *
 * Note that only this function cannot bring a filesystem to be in a clean
4812 4813
 * state independently. It relies on upper layer to stop all data & metadata
 * modifications.
4814
 */
4815
static int ext4_freeze(struct super_block *sb)
4816
{
4817 4818
	int error = 0;
	journal_t *journal;
4819

4820 4821
	if (sb->s_flags & MS_RDONLY)
		return 0;
4822

4823
	journal = EXT4_SB(sb)->s_journal;
4824

4825 4826 4827
	if (journal) {
		/* Now we set up the journal barrier. */
		jbd2_journal_lock_updates(journal);
4828

4829 4830 4831 4832 4833 4834 4835 4836
		/*
		 * Don't clear the needs_recovery flag if we failed to
		 * flush the journal.
		 */
		error = jbd2_journal_flush(journal);
		if (error < 0)
			goto out;
	}
4837 4838 4839 4840

	/* Journal blocked and flushed, clear needs_recovery flag. */
	EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
	error = ext4_commit_super(sb, 1);
4841
out:
4842 4843 4844
	if (journal)
		/* we rely on upper layer to stop further updates */
		jbd2_journal_unlock_updates(journal);
4845
	return error;
4846 4847 4848 4849 4850 4851
}

/*
 * 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.
 */
4852
static int ext4_unfreeze(struct super_block *sb)
4853
{
4854 4855 4856 4857 4858 4859
	if (sb->s_flags & MS_RDONLY)
		return 0;

	/* Reset the needs_recovery flag before the fs is unlocked. */
	EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
	ext4_commit_super(sb, 1);
4860
	return 0;
4861 4862
}

4863 4864 4865 4866 4867
/*
 * Structure to save mount options for ext4_remount's benefit
 */
struct ext4_mount_options {
	unsigned long s_mount_opt;
4868
	unsigned long s_mount_opt2;
4869 4870
	kuid_t s_resuid;
	kgid_t s_resgid;
4871 4872 4873 4874
	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
4875
	char *s_qf_names[EXT4_MAXQUOTAS];
4876 4877 4878
#endif
};

4879
static int ext4_remount(struct super_block *sb, int *flags, char *data)
4880
{
4881
	struct ext4_super_block *es;
4882
	struct ext4_sb_info *sbi = EXT4_SB(sb);
4883
	unsigned long old_sb_flags;
4884
	struct ext4_mount_options old_opts;
4885
	int enable_quota = 0;
4886
	ext4_group_t g;
4887
	unsigned int journal_ioprio = DEFAULT_JOURNAL_IOPRIO;
4888
	int err = 0;
4889
#ifdef CONFIG_QUOTA
4890
	int i, j;
4891
#endif
4892
	char *orig_data = kstrdup(data, GFP_KERNEL);
4893 4894 4895 4896

	/* Store the original options */
	old_sb_flags = sb->s_flags;
	old_opts.s_mount_opt = sbi->s_mount_opt;
4897
	old_opts.s_mount_opt2 = sbi->s_mount_opt2;
4898 4899 4900
	old_opts.s_resuid = sbi->s_resuid;
	old_opts.s_resgid = sbi->s_resgid;
	old_opts.s_commit_interval = sbi->s_commit_interval;
4901 4902
	old_opts.s_min_batch_time = sbi->s_min_batch_time;
	old_opts.s_max_batch_time = sbi->s_max_batch_time;
4903 4904
#ifdef CONFIG_QUOTA
	old_opts.s_jquota_fmt = sbi->s_jquota_fmt;
J
Jan Kara 已提交
4905
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
4906 4907 4908 4909 4910 4911
		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]);
4912
				kfree(orig_data);
4913 4914 4915 4916
				return -ENOMEM;
			}
		} else
			old_opts.s_qf_names[i] = NULL;
4917
#endif
4918 4919
	if (sbi->s_journal && sbi->s_journal->j_task->io_context)
		journal_ioprio = sbi->s_journal->j_task->io_context->ioprio;
4920

4921
	if (!parse_options(data, sb, NULL, &journal_ioprio, 1)) {
4922 4923 4924 4925
		err = -EINVAL;
		goto restore_opts;
	}

4926
	if ((old_opts.s_mount_opt & EXT4_MOUNT_JOURNAL_CHECKSUM) ^
4927 4928
	    test_opt(sb, JOURNAL_CHECKSUM)) {
		ext4_msg(sb, KERN_ERR, "changing journal_checksum "
4929 4930
			 "during remount not supported; ignoring");
		sbi->s_mount_opt ^= EXT4_MOUNT_JOURNAL_CHECKSUM;
4931 4932
	}

4933 4934 4935 4936 4937 4938 4939 4940 4941 4942 4943 4944 4945
	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
4946 4947 4948 4949 4950 4951 4952 4953 4954 4955 4956 4957
		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;
4958 4959
	}

4960
	if (sbi->s_mount_flags & EXT4_MF_FS_ABORTED)
4961
		ext4_abort(sb, "Abort forced by user");
4962 4963

	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
4964
		(test_opt(sb, POSIX_ACL) ? MS_POSIXACL : 0);
4965 4966 4967

	es = sbi->s_es;

4968
	if (sbi->s_journal) {
4969
		ext4_init_journal_params(sb, sbi->s_journal);
4970 4971
		set_task_ioprio(sbi->s_journal->j_task, journal_ioprio);
	}
4972

4973
	if ((*flags & MS_RDONLY) != (sb->s_flags & MS_RDONLY)) {
4974
		if (sbi->s_mount_flags & EXT4_MF_FS_ABORTED) {
4975 4976 4977 4978 4979
			err = -EROFS;
			goto restore_opts;
		}

		if (*flags & MS_RDONLY) {
4980 4981 4982
			err = sync_filesystem(sb);
			if (err < 0)
				goto restore_opts;
4983 4984
			err = dquot_suspend(sb, -1);
			if (err < 0)
4985 4986
				goto restore_opts;

4987 4988 4989 4990 4991 4992 4993 4994 4995 4996 4997
			/*
			 * 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.
			 */
4998 4999
			if (!(es->s_state & cpu_to_le16(EXT4_VALID_FS)) &&
			    (sbi->s_mount_state & EXT4_VALID_FS))
5000 5001
				es->s_state = cpu_to_le16(sbi->s_mount_state);

5002
			if (sbi->s_journal)
5003
				ext4_mark_recovery_complete(sb, es);
5004
		} else {
5005
			/* Make sure we can mount this feature set readwrite */
D
Darrick J. Wong 已提交
5006 5007 5008
			if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
					EXT4_FEATURE_RO_COMPAT_READONLY) ||
			    !ext4_feature_set_ok(sb, 0)) {
5009 5010 5011
				err = -EROFS;
				goto restore_opts;
			}
5012 5013
			/*
			 * Make sure the group descriptor checksums
5014
			 * are sane.  If they aren't, refuse to remount r/w.
5015 5016 5017 5018 5019
			 */
			for (g = 0; g < sbi->s_groups_count; g++) {
				struct ext4_group_desc *gdp =
					ext4_get_group_desc(sb, g, NULL);

5020
				if (!ext4_group_desc_csum_verify(sb, g, gdp)) {
5021 5022
					ext4_msg(sb, KERN_ERR,
	       "ext4_remount: Checksum for group %u failed (%u!=%u)",
5023 5024 5025 5026 5027 5028 5029
		g, le16_to_cpu(ext4_group_desc_csum(sbi, g, gdp)),
					       le16_to_cpu(gdp->bg_checksum));
					err = -EINVAL;
					goto restore_opts;
				}
			}

5030 5031 5032 5033 5034 5035
			/*
			 * 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) {
5036
				ext4_msg(sb, KERN_WARNING, "Couldn't "
5037 5038
				       "remount RDWR because of unprocessed "
				       "orphan inode list.  Please "
5039
				       "umount/remount instead");
5040 5041 5042 5043
				err = -EINVAL;
				goto restore_opts;
			}

5044 5045 5046 5047 5048 5049
			/*
			 * 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.)
			 */
5050 5051
			if (sbi->s_journal)
				ext4_clear_journal_err(sb, es);
5052
			sbi->s_mount_state = le16_to_cpu(es->s_state);
5053
			if (!ext4_setup_super(sb, es, 0))
5054
				sb->s_flags &= ~MS_RDONLY;
5055 5056 5057 5058 5059 5060 5061
			if (EXT4_HAS_INCOMPAT_FEATURE(sb,
						     EXT4_FEATURE_INCOMPAT_MMP))
				if (ext4_multi_mount_protect(sb,
						le64_to_cpu(es->s_mmp_block))) {
					err = -EROFS;
					goto restore_opts;
				}
5062
			enable_quota = 1;
5063 5064
		}
	}
5065 5066 5067 5068 5069 5070 5071 5072 5073 5074 5075 5076 5077

	/*
	 * 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);
	}

5078
	ext4_setup_system_zone(sb);
5079
	if (sbi->s_journal == NULL && !(old_sb_flags & MS_RDONLY))
5080
		ext4_commit_super(sb, 1);
5081

5082 5083
#ifdef CONFIG_QUOTA
	/* Release old quota file names */
J
Jan Kara 已提交
5084
	for (i = 0; i < EXT4_MAXQUOTAS; i++)
5085
		kfree(old_opts.s_qf_names[i]);
5086 5087 5088 5089 5090 5091
	if (enable_quota) {
		if (sb_any_quota_suspended(sb))
			dquot_resume(sb, -1);
		else if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
					EXT4_FEATURE_RO_COMPAT_QUOTA)) {
			err = ext4_enable_quotas(sb);
5092
			if (err)
5093 5094 5095
				goto restore_opts;
		}
	}
5096
#endif
5097

5098
	*flags = (*flags & ~MS_LAZYTIME) | (sb->s_flags & MS_LAZYTIME);
5099 5100
	ext4_msg(sb, KERN_INFO, "re-mounted. Opts: %s", orig_data);
	kfree(orig_data);
5101
	return 0;
5102

5103 5104 5105
restore_opts:
	sb->s_flags = old_sb_flags;
	sbi->s_mount_opt = old_opts.s_mount_opt;
5106
	sbi->s_mount_opt2 = old_opts.s_mount_opt2;
5107 5108 5109
	sbi->s_resuid = old_opts.s_resuid;
	sbi->s_resgid = old_opts.s_resgid;
	sbi->s_commit_interval = old_opts.s_commit_interval;
5110 5111
	sbi->s_min_batch_time = old_opts.s_min_batch_time;
	sbi->s_max_batch_time = old_opts.s_max_batch_time;
5112 5113
#ifdef CONFIG_QUOTA
	sbi->s_jquota_fmt = old_opts.s_jquota_fmt;
J
Jan Kara 已提交
5114
	for (i = 0; i < EXT4_MAXQUOTAS; i++) {
5115
		kfree(sbi->s_qf_names[i]);
5116 5117 5118
		sbi->s_qf_names[i] = old_opts.s_qf_names[i];
	}
#endif
5119
	kfree(orig_data);
5120 5121 5122
	return err;
}

5123
static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf)
5124 5125
{
	struct super_block *sb = dentry->d_sb;
5126 5127
	struct ext4_sb_info *sbi = EXT4_SB(sb);
	struct ext4_super_block *es = sbi->s_es;
L
Lukas Czerner 已提交
5128
	ext4_fsblk_t overhead = 0, resv_blocks;
P
Pekka Enberg 已提交
5129
	u64 fsid;
5130
	s64 bfree;
L
Lukas Czerner 已提交
5131
	resv_blocks = EXT4_C2B(sbi, atomic64_read(&sbi->s_resv_clusters));
5132

5133 5134
	if (!test_opt(sb, MINIX_DF))
		overhead = sbi->s_overhead;
5135

5136
	buf->f_type = EXT4_SUPER_MAGIC;
5137
	buf->f_bsize = sb->s_blocksize;
5138
	buf->f_blocks = ext4_blocks_count(es) - EXT4_C2B(sbi, overhead);
5139 5140
	bfree = percpu_counter_sum_positive(&sbi->s_freeclusters_counter) -
		percpu_counter_sum_positive(&sbi->s_dirtyclusters_counter);
5141
	/* prevent underflow in case that few free space is available */
5142
	buf->f_bfree = EXT4_C2B(sbi, max_t(s64, bfree, 0));
L
Lukas Czerner 已提交
5143 5144 5145
	buf->f_bavail = buf->f_bfree -
			(ext4_r_blocks_count(es) + resv_blocks);
	if (buf->f_bfree < (ext4_r_blocks_count(es) + resv_blocks))
5146 5147
		buf->f_bavail = 0;
	buf->f_files = le32_to_cpu(es->s_inodes_count);
P
Peter Zijlstra 已提交
5148
	buf->f_ffree = percpu_counter_sum_positive(&sbi->s_freeinodes_counter);
5149
	buf->f_namelen = EXT4_NAME_LEN;
P
Pekka Enberg 已提交
5150 5151 5152 5153
	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;
5154

5155 5156 5157
	return 0;
}

5158 5159
/* Helper function for writing quotas on sync - we need to start transaction
 * before quota file is locked for write. Otherwise the are possible deadlocks:
5160
 * Process 1                         Process 2
5161
 * ext4_create()                     quota_sync()
5162
 *   jbd2_journal_start()                  write_dquot()
5163
 *   dquot_initialize()                         down(dqio_mutex)
5164
 *     down(dqio_mutex)                    jbd2_journal_start()
5165 5166 5167 5168 5169 5170 5171
 *
 */

#ifdef CONFIG_QUOTA

static inline struct inode *dquot_to_inode(struct dquot *dquot)
{
5172
	return sb_dqopt(dquot->dq_sb)->files[dquot->dq_id.type];
5173 5174
}

5175
static int ext4_write_dquot(struct dquot *dquot)
5176 5177 5178 5179 5180 5181
{
	int ret, err;
	handle_t *handle;
	struct inode *inode;

	inode = dquot_to_inode(dquot);
5182
	handle = ext4_journal_start(inode, EXT4_HT_QUOTA,
5183
				    EXT4_QUOTA_TRANS_BLOCKS(dquot->dq_sb));
5184 5185 5186
	if (IS_ERR(handle))
		return PTR_ERR(handle);
	ret = dquot_commit(dquot);
5187
	err = ext4_journal_stop(handle);
5188 5189 5190 5191 5192
	if (!ret)
		ret = err;
	return ret;
}

5193
static int ext4_acquire_dquot(struct dquot *dquot)
5194 5195 5196 5197
{
	int ret, err;
	handle_t *handle;

5198
	handle = ext4_journal_start(dquot_to_inode(dquot), EXT4_HT_QUOTA,
5199
				    EXT4_QUOTA_INIT_BLOCKS(dquot->dq_sb));
5200 5201 5202
	if (IS_ERR(handle))
		return PTR_ERR(handle);
	ret = dquot_acquire(dquot);
5203
	err = ext4_journal_stop(handle);
5204 5205 5206 5207 5208
	if (!ret)
		ret = err;
	return ret;
}

5209
static int ext4_release_dquot(struct dquot *dquot)
5210 5211 5212 5213
{
	int ret, err;
	handle_t *handle;

5214
	handle = ext4_journal_start(dquot_to_inode(dquot), EXT4_HT_QUOTA,
5215
				    EXT4_QUOTA_DEL_BLOCKS(dquot->dq_sb));
J
Jan Kara 已提交
5216 5217 5218
	if (IS_ERR(handle)) {
		/* Release dquot anyway to avoid endless cycle in dqput() */
		dquot_release(dquot);
5219
		return PTR_ERR(handle);
J
Jan Kara 已提交
5220
	}
5221
	ret = dquot_release(dquot);
5222
	err = ext4_journal_stop(handle);
5223 5224 5225 5226 5227
	if (!ret)
		ret = err;
	return ret;
}

5228
static int ext4_mark_dquot_dirty(struct dquot *dquot)
5229
{
5230 5231 5232
	struct super_block *sb = dquot->dq_sb;
	struct ext4_sb_info *sbi = EXT4_SB(sb);

5233
	/* Are we journaling quotas? */
5234 5235
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_QUOTA) ||
	    sbi->s_qf_names[USRQUOTA] || sbi->s_qf_names[GRPQUOTA]) {
5236
		dquot_mark_dquot_dirty(dquot);
5237
		return ext4_write_dquot(dquot);
5238 5239 5240 5241 5242
	} else {
		return dquot_mark_dquot_dirty(dquot);
	}
}

5243
static int ext4_write_info(struct super_block *sb, int type)
5244 5245 5246 5247 5248
{
	int ret, err;
	handle_t *handle;

	/* Data block + inode block */
5249
	handle = ext4_journal_start(d_inode(sb->s_root), EXT4_HT_QUOTA, 2);
5250 5251 5252
	if (IS_ERR(handle))
		return PTR_ERR(handle);
	ret = dquot_commit_info(sb, type);
5253
	err = ext4_journal_stop(handle);
5254 5255 5256 5257 5258 5259 5260 5261 5262
	if (!ret)
		ret = err;
	return ret;
}

/*
 * Turn on quotas during mount time - we need to find
 * the quota file and such...
 */
5263
static int ext4_quota_on_mount(struct super_block *sb, int type)
5264
{
5265 5266
	return dquot_quota_on_mount(sb, EXT4_SB(sb)->s_qf_names[type],
					EXT4_SB(sb)->s_jquota_fmt, type);
5267 5268 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
	return dquot_quota_on(sb, type, format_id, path);
5310 5311
}

5312 5313 5314 5315 5316
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
5317
	unsigned long qf_inums[EXT4_MAXQUOTAS] = {
5318 5319 5320 5321 5322 5323 5324 5325 5326 5327 5328 5329 5330 5331 5332
		le32_to_cpu(EXT4_SB(sb)->s_es->s_usr_quota_inum),
		le32_to_cpu(EXT4_SB(sb)->s_es->s_grp_quota_inum)
	};

	BUG_ON(!EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_QUOTA));

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

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

J
Jan Kara 已提交
5333 5334
	/* Don't account quota for quota files to avoid recursion */
	qf_inode->i_flags |= S_NOQUOTA;
5335 5336 5337 5338 5339 5340 5341 5342 5343 5344
	err = dquot_enable(qf_inode, type, format_id, flags);
	iput(qf_inode);

	return err;
}

/* Enable usage tracking for all quota types. */
static int ext4_enable_quotas(struct super_block *sb)
{
	int type, err = 0;
J
Jan Kara 已提交
5345
	unsigned long qf_inums[EXT4_MAXQUOTAS] = {
5346 5347 5348 5349 5350
		le32_to_cpu(EXT4_SB(sb)->s_es->s_usr_quota_inum),
		le32_to_cpu(EXT4_SB(sb)->s_es->s_grp_quota_inum)
	};

	sb_dqopt(sb)->flags |= DQUOT_QUOTA_SYS_FILE;
J
Jan Kara 已提交
5351
	for (type = 0; type < EXT4_MAXQUOTAS; type++) {
5352 5353 5354 5355 5356
		if (qf_inums[type]) {
			err = ext4_quota_enable(sb, type, QFMT_VFS_V1,
						DQUOT_USAGE_ENABLED);
			if (err) {
				ext4_warning(sb,
5357 5358 5359
					"Failed to enable quota tracking "
					"(type=%d, err=%d). Please run "
					"e2fsck to fix.", type, err);
5360 5361 5362 5363 5364 5365 5366
				return err;
			}
		}
	}
	return 0;
}

5367 5368
static int ext4_quota_off(struct super_block *sb, int type)
{
5369 5370 5371
	struct inode *inode = sb_dqopt(sb)->files[type];
	handle_t *handle;

5372 5373 5374
	/* Force all delayed allocation blocks to be allocated.
	 * Caller already holds s_umount sem */
	if (test_opt(sb, DELALLOC))
5375 5376
		sync_filesystem(sb);

5377 5378 5379
	if (!inode)
		goto out;

5380 5381
	/* Update modification times of quota files when userspace can
	 * start looking at them */
5382
	handle = ext4_journal_start(inode, EXT4_HT_QUOTA, 1);
5383 5384 5385 5386 5387 5388 5389
	if (IS_ERR(handle))
		goto out;
	inode->i_mtime = inode->i_ctime = CURRENT_TIME;
	ext4_mark_inode_dirty(handle, inode);
	ext4_journal_stop(handle);

out:
5390 5391 5392
	return dquot_quota_off(sb, type);
}

5393 5394
/* 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
5395
 * itself serializes the operations (and no one else should touch the files)
5396
 * we don't have to be afraid of races */
5397
static ssize_t ext4_quota_read(struct super_block *sb, int type, char *data,
5398 5399 5400
			       size_t len, loff_t off)
{
	struct inode *inode = sb_dqopt(sb)->files[type];
A
Aneesh Kumar K.V 已提交
5401
	ext4_lblk_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
5402 5403 5404 5405 5406 5407 5408 5409 5410 5411 5412 5413 5414 5415
	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;
5416 5417 5418
		bh = ext4_bread(NULL, inode, blk, 0);
		if (IS_ERR(bh))
			return PTR_ERR(bh);
5419 5420 5421 5422 5423 5424 5425 5426 5427 5428 5429 5430 5431 5432 5433
		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) */
5434
static ssize_t ext4_quota_write(struct super_block *sb, int type,
5435 5436 5437
				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
5438
	ext4_lblk_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
5439
	int err, offset = off & (sb->s_blocksize - 1);
5440 5441 5442
	struct buffer_head *bh;
	handle_t *handle = journal_current_handle();

5443
	if (EXT4_SB(sb)->s_journal && !handle) {
5444 5445
		ext4_msg(sb, KERN_WARNING, "Quota write (off=%llu, len=%llu)"
			" cancelled because transaction is not started",
J
Jan Kara 已提交
5446 5447 5448
			(unsigned long long)off, (unsigned long long)len);
		return -EIO;
	}
5449 5450 5451 5452 5453 5454 5455 5456 5457 5458 5459
	/*
	 * 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;
	}

5460 5461 5462
	bh = ext4_bread(handle, inode, blk, 1);
	if (IS_ERR(bh))
		return PTR_ERR(bh);
5463 5464
	if (!bh)
		goto out;
5465
	BUFFER_TRACE(bh, "get write access");
5466 5467 5468
	err = ext4_journal_get_write_access(handle, bh);
	if (err) {
		brelse(bh);
5469
		return err;
5470
	}
5471 5472 5473 5474
	lock_buffer(bh);
	memcpy(bh->b_data+offset, data, len);
	flush_dcache_page(bh->b_page);
	unlock_buffer(bh);
5475
	err = ext4_handle_dirty_metadata(handle, NULL, bh);
5476
	brelse(bh);
5477
out:
5478 5479
	if (inode->i_size < off + len) {
		i_size_write(inode, off + len);
5480
		EXT4_I(inode)->i_disksize = inode->i_size;
5481
		ext4_mark_inode_dirty(handle, inode);
5482
	}
5483
	return len;
5484 5485 5486 5487
}

#endif

A
Al Viro 已提交
5488 5489
static struct dentry *ext4_mount(struct file_system_type *fs_type, int flags,
		       const char *dev_name, void *data)
5490
{
A
Al Viro 已提交
5491
	return mount_bdev(fs_type, flags, dev_name, data, ext4_fill_super);
5492 5493
}

5494
#if !defined(CONFIG_EXT2_FS) && !defined(CONFIG_EXT2_FS_MODULE) && defined(CONFIG_EXT4_USE_FOR_EXT23)
5495 5496 5497 5498 5499 5500 5501 5502 5503 5504 5505 5506
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);
}
5507 5508 5509 5510 5511 5512 5513 5514 5515 5516 5517

static inline int ext2_feature_set_ok(struct super_block *sb)
{
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, ~EXT2_FEATURE_INCOMPAT_SUPP))
		return 0;
	if (sb->s_flags & MS_RDONLY)
		return 1;
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, ~EXT2_FEATURE_RO_COMPAT_SUPP))
		return 0;
	return 1;
}
5518 5519 5520
#else
static inline void register_as_ext2(void) { }
static inline void unregister_as_ext2(void) { }
5521
static inline int ext2_feature_set_ok(struct super_block *sb) { return 0; }
5522 5523
#endif

5524
#if !defined(CONFIG_EXT3_FS) && !defined(CONFIG_EXT3_FS_MODULE) && defined(CONFIG_EXT4_USE_FOR_EXT23)
5525 5526 5527 5528 5529 5530 5531 5532 5533 5534 5535 5536
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);
}
5537 5538 5539 5540 5541 5542 5543 5544 5545 5546 5547 5548 5549

static inline int ext3_feature_set_ok(struct super_block *sb)
{
	if (EXT4_HAS_INCOMPAT_FEATURE(sb, ~EXT3_FEATURE_INCOMPAT_SUPP))
		return 0;
	if (!EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL))
		return 0;
	if (sb->s_flags & MS_RDONLY)
		return 1;
	if (EXT4_HAS_RO_COMPAT_FEATURE(sb, ~EXT3_FEATURE_RO_COMPAT_SUPP))
		return 0;
	return 1;
}
5550 5551 5552
#else
static inline void register_as_ext3(void) { }
static inline void unregister_as_ext3(void) { }
5553
static inline int ext3_feature_set_ok(struct super_block *sb) { return 0; }
5554 5555
#endif

T
Theodore Ts'o 已提交
5556 5557 5558
static struct file_system_type ext4_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "ext4",
A
Al Viro 已提交
5559
	.mount		= ext4_mount,
T
Theodore Ts'o 已提交
5560 5561 5562
	.kill_sb	= kill_block_super,
	.fs_flags	= FS_REQUIRES_DEV,
};
5563
MODULE_ALIAS_FS("ext4");
T
Theodore Ts'o 已提交
5564

5565
static int __init ext4_init_feat_adverts(void)
5566 5567 5568 5569 5570 5571 5572 5573 5574 5575 5576 5577 5578 5579 5580 5581 5582 5583 5584 5585 5586 5587 5588
{
	struct ext4_features *ef;
	int ret = -ENOMEM;

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

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

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

5589 5590 5591 5592 5593 5594 5595
static void ext4_exit_feat_adverts(void)
{
	kobject_put(&ext4_feat->f_kobj);
	wait_for_completion(&ext4_feat->f_kobj_unregister);
	kfree(ext4_feat);
}

5596 5597 5598 5599
/* Shared across all ext4 file systems */
wait_queue_head_t ext4__ioend_wq[EXT4_WQ_HASH_SZ];
struct mutex ext4__aio_mutex[EXT4_WQ_HASH_SZ];

5600
static int __init ext4_init_fs(void)
5601
{
5602
	int i, err;
5603

5604 5605 5606
	ext4_li_info = NULL;
	mutex_init(&ext4_li_mtx);

5607
	/* Build-time check for flags consistency */
5608
	ext4_check_flag_values();
5609 5610 5611 5612 5613 5614

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

5615
	err = ext4_init_es();
5616 5617
	if (err)
		return err;
5618 5619 5620 5621 5622

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

5623
	err = ext4_init_system_zone();
5624
	if (err)
5625
		goto out6;
T
Theodore Ts'o 已提交
5626
	ext4_kset = kset_create_and_add("ext4", NULL, fs_kobj);
5627 5628
	if (!ext4_kset) {
		err = -ENOMEM;
T
Theodore Ts'o 已提交
5629
		goto out5;
5630
	}
5631
	ext4_proc_root = proc_mkdir("fs/ext4", NULL);
5632 5633

	err = ext4_init_feat_adverts();
T
Theodore Ts'o 已提交
5634 5635
	if (err)
		goto out4;
5636

5637
	err = ext4_init_mballoc();
5638 5639
	if (err)
		goto out2;
5640 5641
	else
		ext4_mballoc_ready = 1;
5642 5643 5644
	err = init_inodecache();
	if (err)
		goto out1;
5645
	register_as_ext3();
5646
	register_as_ext2();
T
Theodore Ts'o 已提交
5647
	err = register_filesystem(&ext4_fs_type);
5648 5649
	if (err)
		goto out;
5650

5651 5652
	return 0;
out:
5653 5654
	unregister_as_ext2();
	unregister_as_ext3();
5655 5656
	destroy_inodecache();
out1:
5657
	ext4_mballoc_ready = 0;
5658
	ext4_exit_mballoc();
5659
out2:
5660
	ext4_exit_feat_adverts();
T
Theodore Ts'o 已提交
5661
out4:
5662 5663
	if (ext4_proc_root)
		remove_proc_entry("fs/ext4", NULL);
5664
	kset_unregister(ext4_kset);
5665
out5:
5666
	ext4_exit_system_zone();
5667
out6:
5668
	ext4_exit_pageio();
5669 5670 5671
out7:
	ext4_exit_es();

5672 5673 5674
	return err;
}

5675
static void __exit ext4_exit_fs(void)
5676
{
5677
	ext4_exit_crypto();
5678
	ext4_destroy_lazyinit_thread();
5679 5680
	unregister_as_ext2();
	unregister_as_ext3();
T
Theodore Ts'o 已提交
5681
	unregister_filesystem(&ext4_fs_type);
5682
	destroy_inodecache();
5683
	ext4_exit_mballoc();
5684
	ext4_exit_feat_adverts();
5685
	remove_proc_entry("fs/ext4", NULL);
T
Theodore Ts'o 已提交
5686
	kset_unregister(ext4_kset);
5687 5688
	ext4_exit_system_zone();
	ext4_exit_pageio();
5689
	ext4_exit_es();
5690 5691 5692
}

MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others");
5693
MODULE_DESCRIPTION("Fourth Extended Filesystem");
5694
MODULE_LICENSE("GPL");
5695 5696
module_init(ext4_init_fs)
module_exit(ext4_exit_fs)