super.c 57.2 KB
Newer Older
J
Jaegeuk Kim 已提交
1
/*
J
Jaegeuk Kim 已提交
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
 * fs/f2fs/super.c
 *
 * Copyright (c) 2012 Samsung Electronics Co., Ltd.
 *             http://www.samsung.com/
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 */
#include <linux/module.h>
#include <linux/init.h>
#include <linux/fs.h>
#include <linux/statfs.h>
#include <linux/buffer_head.h>
#include <linux/backing-dev.h>
#include <linux/kthread.h>
#include <linux/parser.h>
#include <linux/mount.h>
#include <linux/seq_file.h>
21
#include <linux/proc_fs.h>
J
Jaegeuk Kim 已提交
22 23
#include <linux/random.h>
#include <linux/exportfs.h>
24
#include <linux/blkdev.h>
J
Jaegeuk Kim 已提交
25
#include <linux/f2fs_fs.h>
26
#include <linux/sysfs.h>
J
Jaegeuk Kim 已提交
27 28 29

#include "f2fs.h"
#include "node.h"
30
#include "segment.h"
J
Jaegeuk Kim 已提交
31
#include "xattr.h"
32
#include "gc.h"
J
Jaegeuk Kim 已提交
33
#include "trace.h"
J
Jaegeuk Kim 已提交
34

35 36 37
#define CREATE_TRACE_POINTS
#include <trace/events/f2fs.h>

38
static struct proc_dir_entry *f2fs_proc_root;
J
Jaegeuk Kim 已提交
39
static struct kmem_cache *f2fs_inode_cachep;
40
static struct kset *f2fs_kset;
J
Jaegeuk Kim 已提交
41

42
#ifdef CONFIG_F2FS_FAULT_INJECTION
J
Jaegeuk Kim 已提交
43 44 45

char *fault_name[FAULT_MAX] = {
	[FAULT_KMALLOC]		= "kmalloc",
46
	[FAULT_PAGE_ALLOC]	= "page alloc",
J
Jaegeuk Kim 已提交
47 48 49 50
	[FAULT_ALLOC_NID]	= "alloc nid",
	[FAULT_ORPHAN]		= "orphan",
	[FAULT_BLOCK]		= "no more block",
	[FAULT_DIR_DEPTH]	= "too big dir depth",
51
	[FAULT_EVICT_INODE]	= "evict_inode fail",
C
Chao Yu 已提交
52
	[FAULT_IO]		= "IO error",
53
	[FAULT_CHECKPOINT]	= "checkpoint error",
J
Jaegeuk Kim 已提交
54
};
S
Sheng Yong 已提交
55

56 57
static void f2fs_build_fault_attr(struct f2fs_sb_info *sbi,
						unsigned int rate)
S
Sheng Yong 已提交
58
{
59 60
	struct f2fs_fault_info *ffi = &sbi->fault_info;

S
Sheng Yong 已提交
61
	if (rate) {
62 63 64
		atomic_set(&ffi->inject_ops, 0);
		ffi->inject_rate = rate;
		ffi->inject_type = (1 << FAULT_MAX) - 1;
S
Sheng Yong 已提交
65
	} else {
66
		memset(ffi, 0, sizeof(struct f2fs_fault_info));
S
Sheng Yong 已提交
67 68
	}
}
69 70
#endif

71 72 73 74 75 76 77
/* f2fs-wide shrinker description */
static struct shrinker f2fs_shrinker_info = {
	.scan_objects = f2fs_shrink_scan,
	.count_objects = f2fs_shrink_count,
	.seeks = DEFAULT_SEEKS,
};

J
Jaegeuk Kim 已提交
78
enum {
79
	Opt_gc_background,
J
Jaegeuk Kim 已提交
80
	Opt_disable_roll_forward,
81
	Opt_norecovery,
J
Jaegeuk Kim 已提交
82
	Opt_discard,
C
Chao Yu 已提交
83
	Opt_nodiscard,
J
Jaegeuk Kim 已提交
84
	Opt_noheap,
85
	Opt_user_xattr,
J
Jaegeuk Kim 已提交
86
	Opt_nouser_xattr,
87
	Opt_acl,
J
Jaegeuk Kim 已提交
88 89 90
	Opt_noacl,
	Opt_active_logs,
	Opt_disable_ext_identify,
J
Jaegeuk Kim 已提交
91
	Opt_inline_xattr,
92
	Opt_inline_data,
93
	Opt_inline_dentry,
94
	Opt_noinline_dentry,
95
	Opt_flush_merge,
J
Jaegeuk Kim 已提交
96
	Opt_noflush_merge,
J
Jaegeuk Kim 已提交
97
	Opt_nobarrier,
98
	Opt_fastboot,
99
	Opt_extent_cache,
100
	Opt_noextent_cache,
W
Wanpeng Li 已提交
101
	Opt_noinline_data,
102
	Opt_data_flush,
103
	Opt_mode,
104
	Opt_fault_injection,
J
Jaegeuk Kim 已提交
105 106
	Opt_lazytime,
	Opt_nolazytime,
J
Jaegeuk Kim 已提交
107 108 109 110
	Opt_err,
};

static match_table_t f2fs_tokens = {
111
	{Opt_gc_background, "background_gc=%s"},
J
Jaegeuk Kim 已提交
112
	{Opt_disable_roll_forward, "disable_roll_forward"},
113
	{Opt_norecovery, "norecovery"},
J
Jaegeuk Kim 已提交
114
	{Opt_discard, "discard"},
C
Chao Yu 已提交
115
	{Opt_nodiscard, "nodiscard"},
J
Jaegeuk Kim 已提交
116
	{Opt_noheap, "no_heap"},
117
	{Opt_user_xattr, "user_xattr"},
J
Jaegeuk Kim 已提交
118
	{Opt_nouser_xattr, "nouser_xattr"},
119
	{Opt_acl, "acl"},
J
Jaegeuk Kim 已提交
120 121 122
	{Opt_noacl, "noacl"},
	{Opt_active_logs, "active_logs=%u"},
	{Opt_disable_ext_identify, "disable_ext_identify"},
J
Jaegeuk Kim 已提交
123
	{Opt_inline_xattr, "inline_xattr"},
124
	{Opt_inline_data, "inline_data"},
125
	{Opt_inline_dentry, "inline_dentry"},
126
	{Opt_noinline_dentry, "noinline_dentry"},
127
	{Opt_flush_merge, "flush_merge"},
J
Jaegeuk Kim 已提交
128
	{Opt_noflush_merge, "noflush_merge"},
J
Jaegeuk Kim 已提交
129
	{Opt_nobarrier, "nobarrier"},
130
	{Opt_fastboot, "fastboot"},
131
	{Opt_extent_cache, "extent_cache"},
132
	{Opt_noextent_cache, "noextent_cache"},
W
Wanpeng Li 已提交
133
	{Opt_noinline_data, "noinline_data"},
134
	{Opt_data_flush, "data_flush"},
135
	{Opt_mode, "mode=%s"},
136
	{Opt_fault_injection, "fault_injection=%u"},
J
Jaegeuk Kim 已提交
137 138
	{Opt_lazytime, "lazytime"},
	{Opt_nolazytime, "nolazytime"},
J
Jaegeuk Kim 已提交
139 140 141
	{Opt_err, NULL},
};

142
/* Sysfs support for f2fs */
143 144 145
enum {
	GC_THREAD,	/* struct f2fs_gc_thread */
	SM_INFO,	/* struct f2fs_sm_info */
146
	NM_INFO,	/* struct f2fs_nm_info */
147
	F2FS_SBI,	/* struct f2fs_sb_info */
S
Sheng Yong 已提交
148 149 150 151
#ifdef CONFIG_F2FS_FAULT_INJECTION
	FAULT_INFO_RATE,	/* struct f2fs_fault_info */
	FAULT_INFO_TYPE,	/* struct f2fs_fault_info */
#endif
152 153
};

154 155 156 157 158
struct f2fs_attr {
	struct attribute attr;
	ssize_t (*show)(struct f2fs_attr *, struct f2fs_sb_info *, char *);
	ssize_t (*store)(struct f2fs_attr *, struct f2fs_sb_info *,
			 const char *, size_t);
159
	int struct_type;
160 161 162
	int offset;
};

163 164 165 166 167 168
static unsigned char *__struct_ptr(struct f2fs_sb_info *sbi, int struct_type)
{
	if (struct_type == GC_THREAD)
		return (unsigned char *)sbi->gc_thread;
	else if (struct_type == SM_INFO)
		return (unsigned char *)SM_I(sbi);
169 170
	else if (struct_type == NM_INFO)
		return (unsigned char *)NM_I(sbi);
171 172
	else if (struct_type == F2FS_SBI)
		return (unsigned char *)sbi;
S
Sheng Yong 已提交
173 174 175
#ifdef CONFIG_F2FS_FAULT_INJECTION
	else if (struct_type == FAULT_INFO_RATE ||
					struct_type == FAULT_INFO_TYPE)
176
		return (unsigned char *)&sbi->fault_info;
S
Sheng Yong 已提交
177
#endif
178 179 180
	return NULL;
}

181 182 183 184 185 186 187 188 189 190 191 192 193
static ssize_t lifetime_write_kbytes_show(struct f2fs_attr *a,
		struct f2fs_sb_info *sbi, char *buf)
{
	struct super_block *sb = sbi->sb;

	if (!sb->s_bdev->bd_part)
		return snprintf(buf, PAGE_SIZE, "0\n");

	return snprintf(buf, PAGE_SIZE, "%llu\n",
		(unsigned long long)(sbi->kbytes_written +
			BD_PART_WRITTEN(sbi)));
}

194 195 196
static ssize_t f2fs_sbi_show(struct f2fs_attr *a,
			struct f2fs_sb_info *sbi, char *buf)
{
197
	unsigned char *ptr = NULL;
198 199
	unsigned int *ui;

200 201
	ptr = __struct_ptr(sbi, a->struct_type);
	if (!ptr)
202 203
		return -EINVAL;

204
	ui = (unsigned int *)(ptr + a->offset);
205 206 207 208 209 210 211 212

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

static ssize_t f2fs_sbi_store(struct f2fs_attr *a,
			struct f2fs_sb_info *sbi,
			const char *buf, size_t count)
{
213
	unsigned char *ptr;
214 215 216 217
	unsigned long t;
	unsigned int *ui;
	ssize_t ret;

218 219
	ptr = __struct_ptr(sbi, a->struct_type);
	if (!ptr)
220 221
		return -EINVAL;

222
	ui = (unsigned int *)(ptr + a->offset);
223 224 225 226

	ret = kstrtoul(skip_spaces(buf), 0, &t);
	if (ret < 0)
		return ret;
S
Sheng Yong 已提交
227 228 229 230
#ifdef CONFIG_F2FS_FAULT_INJECTION
	if (a->struct_type == FAULT_INFO_TYPE && t >= (1 << FAULT_MAX))
		return -EINVAL;
#endif
231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261
	*ui = t;
	return count;
}

static ssize_t f2fs_attr_show(struct kobject *kobj,
				struct attribute *attr, char *buf)
{
	struct f2fs_sb_info *sbi = container_of(kobj, struct f2fs_sb_info,
								s_kobj);
	struct f2fs_attr *a = container_of(attr, struct f2fs_attr, attr);

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

static ssize_t f2fs_attr_store(struct kobject *kobj, struct attribute *attr,
						const char *buf, size_t len)
{
	struct f2fs_sb_info *sbi = container_of(kobj, struct f2fs_sb_info,
									s_kobj);
	struct f2fs_attr *a = container_of(attr, struct f2fs_attr, attr);

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

static void f2fs_sb_release(struct kobject *kobj)
{
	struct f2fs_sb_info *sbi = container_of(kobj, struct f2fs_sb_info,
								s_kobj);
	complete(&sbi->s_kobj_unregister);
}

262
#define F2FS_ATTR_OFFSET(_struct_type, _name, _mode, _show, _store, _offset) \
263 264 265 266
static struct f2fs_attr f2fs_attr_##_name = {			\
	.attr = {.name = __stringify(_name), .mode = _mode },	\
	.show	= _show,					\
	.store	= _store,					\
267 268
	.struct_type = _struct_type,				\
	.offset = _offset					\
269 270
}

271 272 273 274
#define F2FS_RW_ATTR(struct_type, struct_name, name, elname)	\
	F2FS_ATTR_OFFSET(struct_type, name, 0644,		\
		f2fs_sbi_show, f2fs_sbi_store,			\
		offsetof(struct struct_name, elname))
275

276 277 278
#define F2FS_GENERAL_RO_ATTR(name) \
static struct f2fs_attr f2fs_attr_##name = __ATTR(name, 0444, name##_show, NULL)

279 280 281 282 283
F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_min_sleep_time, min_sleep_time);
F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_max_sleep_time, max_sleep_time);
F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_no_gc_sleep_time, no_gc_sleep_time);
F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_idle, gc_idle);
F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, reclaim_segments, rec_prefree_segments);
284
F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, max_small_discards, max_discards);
J
Jaegeuk Kim 已提交
285
F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, batched_trim_sections, trim_sections);
286 287
F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, ipu_policy, ipu_policy);
F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_ipu_util, min_ipu_util);
288
F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_fsync_blocks, min_fsync_blocks);
289
F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, ram_thresh, ram_thresh);
C
Chao Yu 已提交
290
F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, ra_nid_pages, ra_nid_pages);
C
Chao Yu 已提交
291
F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, dirty_nats_ratio, dirty_nats_ratio);
292
F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, max_victim_search, max_victim_search);
293
F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, dir_level, dir_level);
294
F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, cp_interval, interval_time[CP_TIME]);
295
F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, idle_interval, interval_time[REQ_TIME]);
S
Sheng Yong 已提交
296 297 298 299
#ifdef CONFIG_F2FS_FAULT_INJECTION
F2FS_RW_ATTR(FAULT_INFO_RATE, f2fs_fault_info, inject_rate, inject_rate);
F2FS_RW_ATTR(FAULT_INFO_TYPE, f2fs_fault_info, inject_type, inject_type);
#endif
300
F2FS_GENERAL_RO_ATTR(lifetime_write_kbytes);
301 302 303 304 305 306

#define ATTR_LIST(name) (&f2fs_attr_##name.attr)
static struct attribute *f2fs_attrs[] = {
	ATTR_LIST(gc_min_sleep_time),
	ATTR_LIST(gc_max_sleep_time),
	ATTR_LIST(gc_no_gc_sleep_time),
307
	ATTR_LIST(gc_idle),
308
	ATTR_LIST(reclaim_segments),
309
	ATTR_LIST(max_small_discards),
J
Jaegeuk Kim 已提交
310
	ATTR_LIST(batched_trim_sections),
311 312
	ATTR_LIST(ipu_policy),
	ATTR_LIST(min_ipu_util),
313
	ATTR_LIST(min_fsync_blocks),
314
	ATTR_LIST(max_victim_search),
315
	ATTR_LIST(dir_level),
316
	ATTR_LIST(ram_thresh),
C
Chao Yu 已提交
317
	ATTR_LIST(ra_nid_pages),
C
Chao Yu 已提交
318
	ATTR_LIST(dirty_nats_ratio),
319
	ATTR_LIST(cp_interval),
320
	ATTR_LIST(idle_interval),
321 322 323 324
#ifdef CONFIG_F2FS_FAULT_INJECTION
	ATTR_LIST(inject_rate),
	ATTR_LIST(inject_type),
#endif
325
	ATTR_LIST(lifetime_write_kbytes),
326 327 328 329 330 331 332 333 334 335 336 337 338 339
	NULL,
};

static const struct sysfs_ops f2fs_attr_ops = {
	.show	= f2fs_attr_show,
	.store	= f2fs_attr_store,
};

static struct kobj_type f2fs_ktype = {
	.default_attrs	= f2fs_attrs,
	.sysfs_ops	= &f2fs_attr_ops,
	.release	= f2fs_sb_release,
};

340 341 342 343 344 345 346 347 348 349 350 351
void f2fs_msg(struct super_block *sb, const char *level, const char *fmt, ...)
{
	struct va_format vaf;
	va_list args;

	va_start(args, fmt);
	vaf.fmt = fmt;
	vaf.va = &args;
	printk("%sF2FS-fs (%s): %pV\n", level, sb->s_id, &vaf);
	va_end(args);
}

J
Jaegeuk Kim 已提交
352 353 354 355 356 357 358
static void init_once(void *foo)
{
	struct f2fs_inode_info *fi = (struct f2fs_inode_info *) foo;

	inode_init_once(&fi->vfs_inode);
}

359 360 361
static int parse_options(struct super_block *sb, char *options)
{
	struct f2fs_sb_info *sbi = F2FS_SB(sb);
362
	struct request_queue *q;
363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386
	substring_t args[MAX_OPT_ARGS];
	char *p, *name;
	int arg = 0;

	if (!options)
		return 0;

	while ((p = strsep(&options, ",")) != NULL) {
		int token;
		if (!*p)
			continue;
		/*
		 * Initialize args struct so we know whether arg was
		 * found; some options take optional arguments.
		 */
		args[0].to = args[0].from = NULL;
		token = match_token(p, f2fs_tokens, args);

		switch (token) {
		case Opt_gc_background:
			name = match_strdup(&args[0]);

			if (!name)
				return -ENOMEM;
387
			if (strlen(name) == 2 && !strncmp(name, "on", 2)) {
388
				set_opt(sbi, BG_GC);
389 390
				clear_opt(sbi, FORCE_FG_GC);
			} else if (strlen(name) == 3 && !strncmp(name, "off", 3)) {
391
				clear_opt(sbi, BG_GC);
392 393 394 395 396
				clear_opt(sbi, FORCE_FG_GC);
			} else if (strlen(name) == 4 && !strncmp(name, "sync", 4)) {
				set_opt(sbi, BG_GC);
				set_opt(sbi, FORCE_FG_GC);
			} else {
397 398 399 400 401 402 403 404
				kfree(name);
				return -EINVAL;
			}
			kfree(name);
			break;
		case Opt_disable_roll_forward:
			set_opt(sbi, DISABLE_ROLL_FORWARD);
			break;
405 406 407 408 409 410
		case Opt_norecovery:
			/* this option mounts f2fs with ro */
			set_opt(sbi, DISABLE_ROLL_FORWARD);
			if (!f2fs_readonly(sb))
				return -EINVAL;
			break;
411
		case Opt_discard:
412 413 414
			q = bdev_get_queue(sb->s_bdev);
			if (blk_queue_discard(q)) {
				set_opt(sbi, DISCARD);
415
			} else if (!f2fs_sb_mounted_blkzoned(sb)) {
416 417 418 419
				f2fs_msg(sb, KERN_WARNING,
					"mounting with \"discard\" option, but "
					"the device does not support discard");
			}
420
			break;
C
Chao Yu 已提交
421
		case Opt_nodiscard:
422 423 424 425 426
			if (f2fs_sb_mounted_blkzoned(sb)) {
				f2fs_msg(sb, KERN_WARNING,
					"discard is required for zoned block devices");
				return -EINVAL;
			}
C
Chao Yu 已提交
427
			clear_opt(sbi, DISCARD);
428
			break;
429 430 431 432
		case Opt_noheap:
			set_opt(sbi, NOHEAP);
			break;
#ifdef CONFIG_F2FS_FS_XATTR
433 434 435
		case Opt_user_xattr:
			set_opt(sbi, XATTR_USER);
			break;
436 437 438
		case Opt_nouser_xattr:
			clear_opt(sbi, XATTR_USER);
			break;
J
Jaegeuk Kim 已提交
439 440 441
		case Opt_inline_xattr:
			set_opt(sbi, INLINE_XATTR);
			break;
442
#else
443 444 445 446
		case Opt_user_xattr:
			f2fs_msg(sb, KERN_INFO,
				"user_xattr options not supported");
			break;
447 448 449 450
		case Opt_nouser_xattr:
			f2fs_msg(sb, KERN_INFO,
				"nouser_xattr options not supported");
			break;
J
Jaegeuk Kim 已提交
451 452 453 454
		case Opt_inline_xattr:
			f2fs_msg(sb, KERN_INFO,
				"inline_xattr options not supported");
			break;
455 456
#endif
#ifdef CONFIG_F2FS_FS_POSIX_ACL
457 458 459
		case Opt_acl:
			set_opt(sbi, POSIX_ACL);
			break;
460 461 462 463
		case Opt_noacl:
			clear_opt(sbi, POSIX_ACL);
			break;
#else
464 465 466
		case Opt_acl:
			f2fs_msg(sb, KERN_INFO, "acl options not supported");
			break;
467 468 469 470 471 472 473 474 475 476 477 478 479 480
		case Opt_noacl:
			f2fs_msg(sb, KERN_INFO, "noacl options not supported");
			break;
#endif
		case Opt_active_logs:
			if (args->from && match_int(args, &arg))
				return -EINVAL;
			if (arg != 2 && arg != 4 && arg != NR_CURSEG_TYPE)
				return -EINVAL;
			sbi->active_logs = arg;
			break;
		case Opt_disable_ext_identify:
			set_opt(sbi, DISABLE_EXT_IDENTIFY);
			break;
481 482 483
		case Opt_inline_data:
			set_opt(sbi, INLINE_DATA);
			break;
484 485 486
		case Opt_inline_dentry:
			set_opt(sbi, INLINE_DENTRY);
			break;
487 488 489
		case Opt_noinline_dentry:
			clear_opt(sbi, INLINE_DENTRY);
			break;
490 491 492
		case Opt_flush_merge:
			set_opt(sbi, FLUSH_MERGE);
			break;
J
Jaegeuk Kim 已提交
493 494 495
		case Opt_noflush_merge:
			clear_opt(sbi, FLUSH_MERGE);
			break;
J
Jaegeuk Kim 已提交
496 497 498
		case Opt_nobarrier:
			set_opt(sbi, NOBARRIER);
			break;
499 500 501
		case Opt_fastboot:
			set_opt(sbi, FASTBOOT);
			break;
502 503 504
		case Opt_extent_cache:
			set_opt(sbi, EXTENT_CACHE);
			break;
505 506 507
		case Opt_noextent_cache:
			clear_opt(sbi, EXTENT_CACHE);
			break;
W
Wanpeng Li 已提交
508 509 510
		case Opt_noinline_data:
			clear_opt(sbi, INLINE_DATA);
			break;
511 512 513
		case Opt_data_flush:
			set_opt(sbi, DATA_FLUSH);
			break;
514 515 516 517 518 519 520
		case Opt_mode:
			name = match_strdup(&args[0]);

			if (!name)
				return -ENOMEM;
			if (strlen(name) == 8 &&
					!strncmp(name, "adaptive", 8)) {
521 522 523 524 525 526 527
				if (f2fs_sb_mounted_blkzoned(sb)) {
					f2fs_msg(sb, KERN_WARNING,
						 "adaptive mode is not allowed with "
						 "zoned block device feature");
					kfree(name);
					return -EINVAL;
				}
528
				set_opt_mode(sbi, F2FS_MOUNT_ADAPTIVE);
529 530
			} else if (strlen(name) == 3 &&
					!strncmp(name, "lfs", 3)) {
531
				set_opt_mode(sbi, F2FS_MOUNT_LFS);
532 533 534 535 536 537
			} else {
				kfree(name);
				return -EINVAL;
			}
			kfree(name);
			break;
538 539 540 541
		case Opt_fault_injection:
			if (args->from && match_int(args, &arg))
				return -EINVAL;
#ifdef CONFIG_F2FS_FAULT_INJECTION
542
			f2fs_build_fault_attr(sbi, arg);
543 544 545 546 547
#else
			f2fs_msg(sb, KERN_INFO,
				"FAULT_INJECTION was not selected");
#endif
			break;
J
Jaegeuk Kim 已提交
548 549 550 551 552 553
		case Opt_lazytime:
			sb->s_flags |= MS_LAZYTIME;
			break;
		case Opt_nolazytime:
			sb->s_flags &= ~MS_LAZYTIME;
			break;
554 555 556 557 558 559 560 561 562 563
		default:
			f2fs_msg(sb, KERN_ERR,
				"Unrecognized mount option \"%s\" or missing value",
				p);
			return -EINVAL;
		}
	}
	return 0;
}

J
Jaegeuk Kim 已提交
564 565 566 567
static struct inode *f2fs_alloc_inode(struct super_block *sb)
{
	struct f2fs_inode_info *fi;

568
	fi = kmem_cache_alloc(f2fs_inode_cachep, GFP_F2FS_ZERO);
J
Jaegeuk Kim 已提交
569 570 571 572 573
	if (!fi)
		return NULL;

	init_once((void *) fi);

M
Masanari Iida 已提交
574
	/* Initialize f2fs-specific inode info */
J
Jaegeuk Kim 已提交
575
	fi->vfs_inode.i_version = 1;
576
	atomic_set(&fi->dirty_pages, 0);
J
Jaegeuk Kim 已提交
577 578
	fi->i_current_depth = 1;
	fi->i_advise = 0;
579
	init_rwsem(&fi->i_sem);
580
	INIT_LIST_HEAD(&fi->dirty_list);
581
	INIT_LIST_HEAD(&fi->gdirty_list);
J
Jaegeuk Kim 已提交
582 583
	INIT_LIST_HEAD(&fi->inmem_pages);
	mutex_init(&fi->inmem_lock);
584 585
	init_rwsem(&fi->dio_rwsem[READ]);
	init_rwsem(&fi->dio_rwsem[WRITE]);
J
Jaegeuk Kim 已提交
586

587 588
	/* Will be used by directory only */
	fi->i_dir_level = F2FS_SB(sb)->dir_level;
J
Jaegeuk Kim 已提交
589 590 591
	return &fi->vfs_inode;
}

592 593 594 595 596 597 598 599 600
static int f2fs_drop_inode(struct inode *inode)
{
	/*
	 * This is to avoid a deadlock condition like below.
	 * writeback_single_inode(inode)
	 *  - f2fs_write_data_page
	 *    - f2fs_gc -> iput -> evict
	 *       - inode_wait_for_writeback(inode)
	 */
601
	if ((!inode_unhashed(inode) && inode->i_state & I_SYNC)) {
602
		if (!inode->i_nlink && !is_bad_inode(inode)) {
J
Jaegeuk Kim 已提交
603 604
			/* to avoid evict_inode call simultaneously */
			atomic_inc(&inode->i_count);
605 606 607 608
			spin_unlock(&inode->i_lock);

			/* some remained atomic pages should discarded */
			if (f2fs_is_atomic_file(inode))
609
				drop_inmem_pages(inode);
610

J
Jaegeuk Kim 已提交
611 612 613
			/* should remain fi->extent_tree for writepage */
			f2fs_destroy_extent_node(inode);

614
			sb_start_intwrite(inode->i_sb);
615
			f2fs_i_size_write(inode, 0);
616 617

			if (F2FS_HAS_BLOCKS(inode))
618
				f2fs_truncate(inode);
619 620 621

			sb_end_intwrite(inode->i_sb);

622
			fscrypt_put_encryption_info(inode, NULL);
623
			spin_lock(&inode->i_lock);
J
Jaegeuk Kim 已提交
624
			atomic_dec(&inode->i_count);
625
		}
626
		return 0;
627
	}
628

629
	return generic_drop_inode(inode);
630 631
}

632
int f2fs_inode_dirtied(struct inode *inode, bool sync)
633
{
634
	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
635
	int ret = 0;
636 637 638

	spin_lock(&sbi->inode_lock[DIRTY_META]);
	if (is_inode_flag_set(inode, FI_DIRTY_INODE)) {
639 640 641 642
		ret = 1;
	} else {
		set_inode_flag(inode, FI_DIRTY_INODE);
		stat_inc_dirty_inode(sbi, DIRTY_META);
643
	}
644 645
	if (sync && list_empty(&F2FS_I(inode)->gdirty_list)) {
		list_add_tail(&F2FS_I(inode)->gdirty_list,
646
				&sbi->inode_list[DIRTY_META]);
647 648
		inc_page_count(sbi, F2FS_DIRTY_IMETA);
	}
649
	spin_unlock(&sbi->inode_lock[DIRTY_META]);
650
	return ret;
651 652 653 654 655 656 657 658 659 660 661
}

void f2fs_inode_synced(struct inode *inode)
{
	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);

	spin_lock(&sbi->inode_lock[DIRTY_META]);
	if (!is_inode_flag_set(inode, FI_DIRTY_INODE)) {
		spin_unlock(&sbi->inode_lock[DIRTY_META]);
		return;
	}
662 663 664 665
	if (!list_empty(&F2FS_I(inode)->gdirty_list)) {
		list_del_init(&F2FS_I(inode)->gdirty_list);
		dec_page_count(sbi, F2FS_DIRTY_IMETA);
	}
666
	clear_inode_flag(inode, FI_DIRTY_INODE);
667
	clear_inode_flag(inode, FI_AUTO_RECOVER);
668
	stat_dec_dirty_inode(F2FS_I_SB(inode), DIRTY_META);
669
	spin_unlock(&sbi->inode_lock[DIRTY_META]);
670 671
}

J
Jaegeuk Kim 已提交
672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690
/*
 * f2fs_dirty_inode() is called from __mark_inode_dirty()
 *
 * We should call set_dirty_inode to write the dirty inode through write_inode.
 */
static void f2fs_dirty_inode(struct inode *inode, int flags)
{
	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);

	if (inode->i_ino == F2FS_NODE_INO(sbi) ||
			inode->i_ino == F2FS_META_INO(sbi))
		return;

	if (flags == I_DIRTY_TIME)
		return;

	if (is_inode_flag_set(inode, FI_AUTO_RECOVER))
		clear_inode_flag(inode, FI_AUTO_RECOVER);

691
	f2fs_inode_dirtied(inode, false);
J
Jaegeuk Kim 已提交
692 693
}

J
Jaegeuk Kim 已提交
694 695 696 697 698 699
static void f2fs_i_callback(struct rcu_head *head)
{
	struct inode *inode = container_of(head, struct inode, i_rcu);
	kmem_cache_free(f2fs_inode_cachep, F2FS_I(inode));
}

700
static void f2fs_destroy_inode(struct inode *inode)
J
Jaegeuk Kim 已提交
701 702 703 704
{
	call_rcu(&inode->i_rcu, f2fs_i_callback);
}

705 706
static void destroy_percpu_info(struct f2fs_sb_info *sbi)
{
707
	percpu_counter_destroy(&sbi->alloc_valid_block_count);
708
	percpu_counter_destroy(&sbi->total_valid_inode_count);
709 710
}

J
Jaegeuk Kim 已提交
711 712 713 714 715 716 717 718 719 720 721 722 723
static void destroy_device_list(struct f2fs_sb_info *sbi)
{
	int i;

	for (i = 0; i < sbi->s_ndevs; i++) {
		blkdev_put(FDEV(i).bdev, FMODE_EXCL);
#ifdef CONFIG_BLK_DEV_ZONED
		kfree(FDEV(i).blkz_type);
#endif
	}
	kfree(sbi->devs);
}

J
Jaegeuk Kim 已提交
724 725 726 727
static void f2fs_put_super(struct super_block *sb)
{
	struct f2fs_sb_info *sbi = F2FS_SB(sb);

728 729
	if (sbi->s_proc) {
		remove_proc_entry("segment_info", sbi->s_proc);
730
		remove_proc_entry("segment_bits", sbi->s_proc);
731 732
		remove_proc_entry(sb->s_id, f2fs_proc_root);
	}
733
	kobject_del(&sbi->s_kobj);
734

J
Jaegeuk Kim 已提交
735 736
	stop_gc_thread(sbi);

737 738 739
	/* prevent remaining shrinker jobs */
	mutex_lock(&sbi->umount_mutex);

740 741 742 743 744
	/*
	 * We don't need to do checkpoint when superblock is clean.
	 * But, the previous checkpoint was not done by umount, it needs to do
	 * clean checkpoint again.
	 */
745
	if (is_sbi_flag_set(sbi, SBI_IS_DIRTY) ||
746
			!is_set_ckpt_flags(sbi, CP_UMOUNT_FLAG)) {
747 748 749 750 751
		struct cp_control cpc = {
			.reason = CP_UMOUNT,
		};
		write_checkpoint(sbi, &cpc);
	}
J
Jaegeuk Kim 已提交
752

J
Jaegeuk Kim 已提交
753 754 755
	/* write_checkpoint can update stat informaion */
	f2fs_destroy_stats(sbi);

756 757 758 759
	/*
	 * normally superblock is clean, so we need to release this.
	 * In addition, EIO will skip do checkpoint, we need this as well.
	 */
760
	release_ino_entry(sbi, true);
761

762 763 764
	f2fs_leave_shrinker(sbi);
	mutex_unlock(&sbi->umount_mutex);

765
	/* our cp_error case, we can wait for any writeback page */
766
	f2fs_flush_merged_bios(sbi);
767

J
Jaegeuk Kim 已提交
768 769 770 771 772 773 774 775
	iput(sbi->node_inode);
	iput(sbi->meta_inode);

	/* destroy f2fs internal modules */
	destroy_node_manager(sbi);
	destroy_segment_manager(sbi);

	kfree(sbi->ckpt);
776 777
	kobject_put(&sbi->s_kobj);
	wait_for_completion(&sbi->s_kobj_unregister);
J
Jaegeuk Kim 已提交
778 779

	sb->s_fs_info = NULL;
K
Keith Mok 已提交
780 781
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
Y
Yunlei He 已提交
782
	kfree(sbi->raw_super);
783

J
Jaegeuk Kim 已提交
784 785
	destroy_device_list(sbi);

786
	destroy_percpu_info(sbi);
J
Jaegeuk Kim 已提交
787 788 789 790 791 792
	kfree(sbi);
}

int f2fs_sync_fs(struct super_block *sb, int sync)
{
	struct f2fs_sb_info *sbi = F2FS_SB(sb);
C
Chao Yu 已提交
793
	int err = 0;
J
Jaegeuk Kim 已提交
794

795 796
	trace_f2fs_sync_fs(sb, sync);

797
	if (sync) {
798 799
		struct cp_control cpc;

800 801
		cpc.reason = __get_cp_reason(sbi);

802
		mutex_lock(&sbi->gc_mutex);
C
Chao Yu 已提交
803
		err = write_checkpoint(sbi, &cpc);
804 805
		mutex_unlock(&sbi->gc_mutex);
	}
806
	f2fs_trace_ios(NULL, 1);
J
Jaegeuk Kim 已提交
807

C
Chao Yu 已提交
808
	return err;
J
Jaegeuk Kim 已提交
809 810
}

811 812
static int f2fs_freeze(struct super_block *sb)
{
J
Jaegeuk Kim 已提交
813
	if (f2fs_readonly(sb))
814 815
		return 0;

816 817 818 819 820 821 822 823
	/* IO error happened before */
	if (unlikely(f2fs_cp_error(F2FS_SB(sb))))
		return -EIO;

	/* must be clean, since sync_filesystem() was already called */
	if (is_sbi_flag_set(F2FS_SB(sb), SBI_IS_DIRTY))
		return -EINVAL;
	return 0;
824 825 826 827 828 829 830
}

static int f2fs_unfreeze(struct super_block *sb)
{
	return 0;
}

J
Jaegeuk Kim 已提交
831 832 833 834 835 836 837 838 839 840 841 842 843 844 845
static int f2fs_statfs(struct dentry *dentry, struct kstatfs *buf)
{
	struct super_block *sb = dentry->d_sb;
	struct f2fs_sb_info *sbi = F2FS_SB(sb);
	u64 id = huge_encode_dev(sb->s_bdev->bd_dev);
	block_t total_count, user_block_count, start_count, ovp_count;

	total_count = le64_to_cpu(sbi->raw_super->block_count);
	user_block_count = sbi->user_block_count;
	start_count = le32_to_cpu(sbi->raw_super->segment0_blkaddr);
	ovp_count = SM_I(sbi)->ovp_segments << sbi->log_blocks_per_seg;
	buf->f_type = F2FS_SUPER_MAGIC;
	buf->f_bsize = sbi->blocksize;

	buf->f_blocks = total_count - start_count;
846
	buf->f_bfree = user_block_count - valid_user_blocks(sbi) + ovp_count;
J
Jaegeuk Kim 已提交
847 848
	buf->f_bavail = user_block_count - valid_user_blocks(sbi);

849
	buf->f_files = sbi->total_node_count - F2FS_RESERVED_NODE_NUM;
850 851
	buf->f_ffree = min(buf->f_files - valid_node_count(sbi),
							buf->f_bavail);
J
Jaegeuk Kim 已提交
852

853
	buf->f_namelen = F2FS_NAME_LEN;
J
Jaegeuk Kim 已提交
854 855 856 857 858 859 860 861 862 863
	buf->f_fsid.val[0] = (u32)id;
	buf->f_fsid.val[1] = (u32)(id >> 32);

	return 0;
}

static int f2fs_show_options(struct seq_file *seq, struct dentry *root)
{
	struct f2fs_sb_info *sbi = F2FS_SB(root->d_sb);

864 865 866 867 868 869
	if (!f2fs_readonly(sbi->sb) && test_opt(sbi, BG_GC)) {
		if (test_opt(sbi, FORCE_FG_GC))
			seq_printf(seq, ",background_gc=%s", "sync");
		else
			seq_printf(seq, ",background_gc=%s", "on");
	} else {
870
		seq_printf(seq, ",background_gc=%s", "off");
871
	}
J
Jaegeuk Kim 已提交
872 873 874 875 876 877 878 879 880 881 882
	if (test_opt(sbi, DISABLE_ROLL_FORWARD))
		seq_puts(seq, ",disable_roll_forward");
	if (test_opt(sbi, DISCARD))
		seq_puts(seq, ",discard");
	if (test_opt(sbi, NOHEAP))
		seq_puts(seq, ",no_heap_alloc");
#ifdef CONFIG_F2FS_FS_XATTR
	if (test_opt(sbi, XATTR_USER))
		seq_puts(seq, ",user_xattr");
	else
		seq_puts(seq, ",nouser_xattr");
J
Jaegeuk Kim 已提交
883 884
	if (test_opt(sbi, INLINE_XATTR))
		seq_puts(seq, ",inline_xattr");
J
Jaegeuk Kim 已提交
885 886 887 888 889 890 891 892
#endif
#ifdef CONFIG_F2FS_FS_POSIX_ACL
	if (test_opt(sbi, POSIX_ACL))
		seq_puts(seq, ",acl");
	else
		seq_puts(seq, ",noacl");
#endif
	if (test_opt(sbi, DISABLE_EXT_IDENTIFY))
893
		seq_puts(seq, ",disable_ext_identify");
894 895
	if (test_opt(sbi, INLINE_DATA))
		seq_puts(seq, ",inline_data");
W
Wanpeng Li 已提交
896 897
	else
		seq_puts(seq, ",noinline_data");
898 899
	if (test_opt(sbi, INLINE_DENTRY))
		seq_puts(seq, ",inline_dentry");
900 901
	else
		seq_puts(seq, ",noinline_dentry");
902
	if (!f2fs_readonly(sbi->sb) && test_opt(sbi, FLUSH_MERGE))
903
		seq_puts(seq, ",flush_merge");
J
Jaegeuk Kim 已提交
904 905
	if (test_opt(sbi, NOBARRIER))
		seq_puts(seq, ",nobarrier");
906 907
	if (test_opt(sbi, FASTBOOT))
		seq_puts(seq, ",fastboot");
908 909
	if (test_opt(sbi, EXTENT_CACHE))
		seq_puts(seq, ",extent_cache");
910 911
	else
		seq_puts(seq, ",noextent_cache");
912 913
	if (test_opt(sbi, DATA_FLUSH))
		seq_puts(seq, ",data_flush");
914 915 916 917 918 919

	seq_puts(seq, ",mode=");
	if (test_opt(sbi, ADAPTIVE))
		seq_puts(seq, "adaptive");
	else if (test_opt(sbi, LFS))
		seq_puts(seq, "lfs");
J
Jaegeuk Kim 已提交
920 921 922 923 924
	seq_printf(seq, ",active_logs=%u", sbi->active_logs);

	return 0;
}

925 926 927 928
static int segment_info_seq_show(struct seq_file *seq, void *offset)
{
	struct super_block *sb = seq->private;
	struct f2fs_sb_info *sbi = F2FS_SB(sb);
C
Chris Fries 已提交
929 930
	unsigned int total_segs =
			le32_to_cpu(sbi->raw_super->segment_count_main);
931 932
	int i;

933 934 935
	seq_puts(seq, "format: segment_type|valid_blocks\n"
		"segment_type(0:HD, 1:WD, 2:CD, 3:HN, 4:WN, 5:CN)\n");

936
	for (i = 0; i < total_segs; i++) {
937 938 939
		struct seg_entry *se = get_seg_entry(sbi, i);

		if ((i % 10) == 0)
Y
Yunlei He 已提交
940
			seq_printf(seq, "%-10d", i);
941 942
		seq_printf(seq, "%d|%-3u", se->type,
					get_valid_blocks(sbi, i, 1));
943 944
		if ((i % 10) == 9 || i == (total_segs - 1))
			seq_putc(seq, '\n');
945
		else
946
			seq_putc(seq, ' ');
947
	}
948

949 950 951
	return 0;
}

952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969
static int segment_bits_seq_show(struct seq_file *seq, void *offset)
{
	struct super_block *sb = seq->private;
	struct f2fs_sb_info *sbi = F2FS_SB(sb);
	unsigned int total_segs =
			le32_to_cpu(sbi->raw_super->segment_count_main);
	int i, j;

	seq_puts(seq, "format: segment_type|valid_blocks|bitmaps\n"
		"segment_type(0:HD, 1:WD, 2:CD, 3:HN, 4:WN, 5:CN)\n");

	for (i = 0; i < total_segs; i++) {
		struct seg_entry *se = get_seg_entry(sbi, i);

		seq_printf(seq, "%-10d", i);
		seq_printf(seq, "%d|%-3u|", se->type,
					get_valid_blocks(sbi, i, 1));
		for (j = 0; j < SIT_VBLOCK_MAP_SIZE; j++)
970
			seq_printf(seq, " %.2x", se->cur_valid_map[j]);
971 972 973 974 975
		seq_putc(seq, '\n');
	}
	return 0;
}

976 977 978 979 980 981 982 983 984 985 986
#define F2FS_PROC_FILE_DEF(_name)					\
static int _name##_open_fs(struct inode *inode, struct file *file)	\
{									\
	return single_open(file, _name##_seq_show, PDE_DATA(inode));	\
}									\
									\
static const struct file_operations f2fs_seq_##_name##_fops = {		\
	.open = _name##_open_fs,					\
	.read = seq_read,						\
	.llseek = seq_lseek,						\
	.release = single_release,					\
987 988
};

989
F2FS_PROC_FILE_DEF(segment_info);
990
F2FS_PROC_FILE_DEF(segment_bits);
991

992 993 994 995 996 997 998
static void default_options(struct f2fs_sb_info *sbi)
{
	/* init some FS parameters */
	sbi->active_logs = NR_CURSEG_TYPE;

	set_opt(sbi, BG_GC);
	set_opt(sbi, INLINE_DATA);
999
	set_opt(sbi, INLINE_DENTRY);
J
Jaegeuk Kim 已提交
1000
	set_opt(sbi, EXTENT_CACHE);
J
Jaegeuk Kim 已提交
1001
	sbi->sb->s_flags |= MS_LAZYTIME;
J
Jaegeuk Kim 已提交
1002
	set_opt(sbi, FLUSH_MERGE);
1003
	if (f2fs_sb_mounted_blkzoned(sbi->sb)) {
1004 1005 1006 1007 1008
		set_opt_mode(sbi, F2FS_MOUNT_LFS);
		set_opt(sbi, DISCARD);
	} else {
		set_opt_mode(sbi, F2FS_MOUNT_ADAPTIVE);
	}
1009 1010 1011 1012 1013 1014 1015

#ifdef CONFIG_F2FS_FS_XATTR
	set_opt(sbi, XATTR_USER);
#endif
#ifdef CONFIG_F2FS_FS_POSIX_ACL
	set_opt(sbi, POSIX_ACL);
#endif
1016 1017 1018 1019

#ifdef CONFIG_F2FS_FAULT_INJECTION
	f2fs_build_fault_attr(sbi, 0);
#endif
1020 1021
}

1022 1023 1024 1025 1026
static int f2fs_remount(struct super_block *sb, int *flags, char *data)
{
	struct f2fs_sb_info *sbi = F2FS_SB(sb);
	struct f2fs_mount_info org_mount_opt;
	int err, active_logs;
1027 1028
	bool need_restart_gc = false;
	bool need_stop_gc = false;
1029
	bool no_extent_cache = !test_opt(sbi, EXTENT_CACHE);
1030 1031 1032
#ifdef CONFIG_F2FS_FAULT_INJECTION
	struct f2fs_fault_info ffi = sbi->fault_info;
#endif
1033 1034 1035 1036 1037 1038 1039 1040

	/*
	 * Save the old mount options in case we
	 * need to restore them.
	 */
	org_mount_opt = sbi->mount_opt;
	active_logs = sbi->active_logs;

1041 1042 1043 1044 1045 1046 1047 1048 1049
	/* recover superblocks we couldn't write due to previous RO mount */
	if (!(*flags & MS_RDONLY) && is_sbi_flag_set(sbi, SBI_NEED_SB_WRITE)) {
		err = f2fs_commit_super(sbi, false);
		f2fs_msg(sb, KERN_INFO,
			"Try to recover all the superblocks, ret: %d", err);
		if (!err)
			clear_sbi_flag(sbi, SBI_NEED_SB_WRITE);
	}

1050
	sbi->mount_opt.opt = 0;
1051
	default_options(sbi);
1052

1053 1054 1055 1056 1057 1058 1059
	/* parse mount options */
	err = parse_options(sb, data);
	if (err)
		goto restore_opts;

	/*
	 * Previous and new state of filesystem is RO,
1060
	 * so skip checking GC and FLUSH_MERGE conditions.
1061
	 */
1062
	if (f2fs_readonly(sb) && (*flags & MS_RDONLY))
1063 1064
		goto skip;

1065 1066 1067 1068 1069 1070 1071 1072
	/* disallow enable/disable extent_cache dynamically */
	if (no_extent_cache == !!test_opt(sbi, EXTENT_CACHE)) {
		err = -EINVAL;
		f2fs_msg(sbi->sb, KERN_WARNING,
				"switch extent_cache option is not allowed");
		goto restore_opts;
	}

1073 1074 1075 1076 1077 1078 1079 1080
	/*
	 * We stop the GC thread if FS is mounted as RO
	 * or if background_gc = off is passed in mount
	 * option. Also sync the filesystem.
	 */
	if ((*flags & MS_RDONLY) || !test_opt(sbi, BG_GC)) {
		if (sbi->gc_thread) {
			stop_gc_thread(sbi);
1081
			need_restart_gc = true;
1082
		}
1083
	} else if (!sbi->gc_thread) {
1084 1085 1086
		err = start_gc_thread(sbi);
		if (err)
			goto restore_opts;
1087 1088 1089
		need_stop_gc = true;
	}

1090 1091 1092 1093 1094 1095 1096 1097 1098 1099
	if (*flags & MS_RDONLY) {
		writeback_inodes_sb(sb, WB_REASON_SYNC);
		sync_inodes_sb(sb);

		set_sbi_flag(sbi, SBI_IS_DIRTY);
		set_sbi_flag(sbi, SBI_IS_CLOSE);
		f2fs_sync_fs(sb, 1);
		clear_sbi_flag(sbi, SBI_IS_CLOSE);
	}

1100 1101 1102 1103 1104
	/*
	 * We stop issue flush thread if FS is mounted as RO
	 * or if flush_merge is not passed in mount option.
	 */
	if ((*flags & MS_RDONLY) || !test_opt(sbi, FLUSH_MERGE)) {
1105 1106 1107
		clear_opt(sbi, FLUSH_MERGE);
		destroy_flush_cmd_control(sbi, false);
	} else {
1108 1109
		err = create_flush_cmd_control(sbi);
		if (err)
1110
			goto restore_gc;
1111 1112 1113
	}
skip:
	/* Update the POSIXACL Flag */
1114
	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
1115
		(test_opt(sbi, POSIX_ACL) ? MS_POSIXACL : 0);
1116

1117
	return 0;
1118 1119 1120 1121
restore_gc:
	if (need_restart_gc) {
		if (start_gc_thread(sbi))
			f2fs_msg(sbi->sb, KERN_WARNING,
A
arter97 已提交
1122
				"background gc thread has stopped");
1123 1124 1125
	} else if (need_stop_gc) {
		stop_gc_thread(sbi);
	}
1126 1127 1128
restore_opts:
	sbi->mount_opt = org_mount_opt;
	sbi->active_logs = active_logs;
1129 1130 1131
#ifdef CONFIG_F2FS_FAULT_INJECTION
	sbi->fault_info = ffi;
#endif
1132 1133 1134
	return err;
}

J
Jaegeuk Kim 已提交
1135 1136
static struct super_operations f2fs_sops = {
	.alloc_inode	= f2fs_alloc_inode,
1137
	.drop_inode	= f2fs_drop_inode,
J
Jaegeuk Kim 已提交
1138 1139
	.destroy_inode	= f2fs_destroy_inode,
	.write_inode	= f2fs_write_inode,
1140
	.dirty_inode	= f2fs_dirty_inode,
J
Jaegeuk Kim 已提交
1141 1142 1143 1144
	.show_options	= f2fs_show_options,
	.evict_inode	= f2fs_evict_inode,
	.put_super	= f2fs_put_super,
	.sync_fs	= f2fs_sync_fs,
1145 1146
	.freeze_fs	= f2fs_freeze,
	.unfreeze_fs	= f2fs_unfreeze,
J
Jaegeuk Kim 已提交
1147
	.statfs		= f2fs_statfs,
1148
	.remount_fs	= f2fs_remount,
J
Jaegeuk Kim 已提交
1149 1150
};

1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172
#ifdef CONFIG_F2FS_FS_ENCRYPTION
static int f2fs_get_context(struct inode *inode, void *ctx, size_t len)
{
	return f2fs_getxattr(inode, F2FS_XATTR_INDEX_ENCRYPTION,
				F2FS_XATTR_NAME_ENCRYPTION_CONTEXT,
				ctx, len, NULL);
}

static int f2fs_set_context(struct inode *inode, const void *ctx, size_t len,
							void *fs_data)
{
	return f2fs_setxattr(inode, F2FS_XATTR_INDEX_ENCRYPTION,
				F2FS_XATTR_NAME_ENCRYPTION_CONTEXT,
				ctx, len, fs_data, XATTR_CREATE);
}

static unsigned f2fs_max_namelen(struct inode *inode)
{
	return S_ISLNK(inode->i_mode) ?
			inode->i_sb->s_blocksize : F2FS_NAME_LEN;
}

1173
static const struct fscrypt_operations f2fs_cryptops = {
1174
	.key_prefix	= "f2fs:",
1175 1176 1177 1178 1179 1180 1181
	.get_context	= f2fs_get_context,
	.set_context	= f2fs_set_context,
	.is_encrypted	= f2fs_encrypted_inode,
	.empty_dir	= f2fs_empty_dir,
	.max_namelen	= f2fs_max_namelen,
};
#else
1182
static const struct fscrypt_operations f2fs_cryptops = {
1183 1184 1185 1186
	.is_encrypted	= f2fs_encrypted_inode,
};
#endif

J
Jaegeuk Kim 已提交
1187 1188 1189 1190 1191 1192
static struct inode *f2fs_nfs_get_inode(struct super_block *sb,
		u64 ino, u32 generation)
{
	struct f2fs_sb_info *sbi = F2FS_SB(sb);
	struct inode *inode;

1193
	if (check_nid_range(sbi, ino))
1194
		return ERR_PTR(-ESTALE);
J
Jaegeuk Kim 已提交
1195 1196 1197 1198 1199 1200 1201 1202 1203

	/*
	 * f2fs_iget isn't quite right if the inode is currently unallocated!
	 * However f2fs_iget currently does appropriate checks to handle stale
	 * inodes so everything is OK.
	 */
	inode = f2fs_iget(sb, ino);
	if (IS_ERR(inode))
		return ERR_CAST(inode);
1204
	if (unlikely(generation && inode->i_generation != generation)) {
J
Jaegeuk Kim 已提交
1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231
		/* we didn't find the right inode.. */
		iput(inode);
		return ERR_PTR(-ESTALE);
	}
	return inode;
}

static struct dentry *f2fs_fh_to_dentry(struct super_block *sb, struct fid *fid,
		int fh_len, int fh_type)
{
	return generic_fh_to_dentry(sb, fid, fh_len, fh_type,
				    f2fs_nfs_get_inode);
}

static struct dentry *f2fs_fh_to_parent(struct super_block *sb, struct fid *fid,
		int fh_len, int fh_type)
{
	return generic_fh_to_parent(sb, fid, fh_len, fh_type,
				    f2fs_nfs_get_inode);
}

static const struct export_operations f2fs_export_ops = {
	.fh_to_dentry = f2fs_fh_to_dentry,
	.fh_to_parent = f2fs_fh_to_parent,
	.get_parent = f2fs_get_parent,
};

C
Chao Yu 已提交
1232
static loff_t max_file_blocks(void)
J
Jaegeuk Kim 已提交
1233
{
1234
	loff_t result = (DEF_ADDRS_PER_INODE - F2FS_INLINE_XATTR_ADDRS);
J
Jaegeuk Kim 已提交
1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250
	loff_t leaf_count = ADDRS_PER_BLOCK;

	/* two direct node blocks */
	result += (leaf_count * 2);

	/* two indirect node blocks */
	leaf_count *= NIDS_PER_BLOCK;
	result += (leaf_count * 2);

	/* one double indirect node block */
	leaf_count *= NIDS_PER_BLOCK;
	result += leaf_count;

	return result;
}

1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261
static int __f2fs_commit_super(struct buffer_head *bh,
			struct f2fs_super_block *super)
{
	lock_buffer(bh);
	if (super)
		memcpy(bh->b_data + F2FS_SUPER_OFFSET, super, sizeof(*super));
	set_buffer_uptodate(bh);
	set_buffer_dirty(bh);
	unlock_buffer(bh);

	/* it's rare case, we can do fua all the time */
1262
	return __sync_dirty_buffer(bh, REQ_PREFLUSH | REQ_FUA);
1263 1264
}

1265
static inline bool sanity_check_area_boundary(struct f2fs_sb_info *sbi,
1266
					struct buffer_head *bh)
1267
{
1268 1269
	struct f2fs_super_block *raw_super = (struct f2fs_super_block *)
					(bh->b_data + F2FS_SUPER_OFFSET);
1270
	struct super_block *sb = sbi->sb;
1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283
	u32 segment0_blkaddr = le32_to_cpu(raw_super->segment0_blkaddr);
	u32 cp_blkaddr = le32_to_cpu(raw_super->cp_blkaddr);
	u32 sit_blkaddr = le32_to_cpu(raw_super->sit_blkaddr);
	u32 nat_blkaddr = le32_to_cpu(raw_super->nat_blkaddr);
	u32 ssa_blkaddr = le32_to_cpu(raw_super->ssa_blkaddr);
	u32 main_blkaddr = le32_to_cpu(raw_super->main_blkaddr);
	u32 segment_count_ckpt = le32_to_cpu(raw_super->segment_count_ckpt);
	u32 segment_count_sit = le32_to_cpu(raw_super->segment_count_sit);
	u32 segment_count_nat = le32_to_cpu(raw_super->segment_count_nat);
	u32 segment_count_ssa = le32_to_cpu(raw_super->segment_count_ssa);
	u32 segment_count_main = le32_to_cpu(raw_super->segment_count_main);
	u32 segment_count = le32_to_cpu(raw_super->segment_count);
	u32 log_blocks_per_seg = le32_to_cpu(raw_super->log_blocks_per_seg);
1284 1285 1286 1287
	u64 main_end_blkaddr = main_blkaddr +
				(segment_count_main << log_blocks_per_seg);
	u64 seg_end_blkaddr = segment0_blkaddr +
				(segment_count << log_blocks_per_seg);
1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331

	if (segment0_blkaddr != cp_blkaddr) {
		f2fs_msg(sb, KERN_INFO,
			"Mismatch start address, segment0(%u) cp_blkaddr(%u)",
			segment0_blkaddr, cp_blkaddr);
		return true;
	}

	if (cp_blkaddr + (segment_count_ckpt << log_blocks_per_seg) !=
							sit_blkaddr) {
		f2fs_msg(sb, KERN_INFO,
			"Wrong CP boundary, start(%u) end(%u) blocks(%u)",
			cp_blkaddr, sit_blkaddr,
			segment_count_ckpt << log_blocks_per_seg);
		return true;
	}

	if (sit_blkaddr + (segment_count_sit << log_blocks_per_seg) !=
							nat_blkaddr) {
		f2fs_msg(sb, KERN_INFO,
			"Wrong SIT boundary, start(%u) end(%u) blocks(%u)",
			sit_blkaddr, nat_blkaddr,
			segment_count_sit << log_blocks_per_seg);
		return true;
	}

	if (nat_blkaddr + (segment_count_nat << log_blocks_per_seg) !=
							ssa_blkaddr) {
		f2fs_msg(sb, KERN_INFO,
			"Wrong NAT boundary, start(%u) end(%u) blocks(%u)",
			nat_blkaddr, ssa_blkaddr,
			segment_count_nat << log_blocks_per_seg);
		return true;
	}

	if (ssa_blkaddr + (segment_count_ssa << log_blocks_per_seg) !=
							main_blkaddr) {
		f2fs_msg(sb, KERN_INFO,
			"Wrong SSA boundary, start(%u) end(%u) blocks(%u)",
			ssa_blkaddr, main_blkaddr,
			segment_count_ssa << log_blocks_per_seg);
		return true;
	}

1332
	if (main_end_blkaddr > seg_end_blkaddr) {
1333
		f2fs_msg(sb, KERN_INFO,
1334
			"Wrong MAIN_AREA boundary, start(%u) end(%u) block(%u)",
1335
			main_blkaddr,
1336 1337
			segment0_blkaddr +
				(segment_count << log_blocks_per_seg),
1338 1339
			segment_count_main << log_blocks_per_seg);
		return true;
1340 1341 1342 1343 1344 1345 1346 1347 1348
	} else if (main_end_blkaddr < seg_end_blkaddr) {
		int err = 0;
		char *res;

		/* fix in-memory information all the time */
		raw_super->segment_count = cpu_to_le32((main_end_blkaddr -
				segment0_blkaddr) >> log_blocks_per_seg);

		if (f2fs_readonly(sb) || bdev_read_only(sb->s_bdev)) {
1349
			set_sbi_flag(sbi, SBI_NEED_SB_WRITE);
1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362
			res = "internally";
		} else {
			err = __f2fs_commit_super(bh, NULL);
			res = err ? "failed" : "done";
		}
		f2fs_msg(sb, KERN_INFO,
			"Fix alignment : %s, start(%u) end(%u) block(%u)",
			res, main_blkaddr,
			segment0_blkaddr +
				(segment_count << log_blocks_per_seg),
			segment_count_main << log_blocks_per_seg);
		if (err)
			return true;
1363 1364 1365 1366
	}
	return false;
}

1367
static int sanity_check_raw_super(struct f2fs_sb_info *sbi,
1368
				struct buffer_head *bh)
J
Jaegeuk Kim 已提交
1369
{
1370 1371
	struct f2fs_super_block *raw_super = (struct f2fs_super_block *)
					(bh->b_data + F2FS_SUPER_OFFSET);
1372
	struct super_block *sb = sbi->sb;
J
Jaegeuk Kim 已提交
1373 1374
	unsigned int blocksize;

1375 1376 1377 1378
	if (F2FS_SUPER_MAGIC != le32_to_cpu(raw_super->magic)) {
		f2fs_msg(sb, KERN_INFO,
			"Magic Mismatch, valid(0x%x) - read(0x%x)",
			F2FS_SUPER_MAGIC, le32_to_cpu(raw_super->magic));
J
Jaegeuk Kim 已提交
1379
		return 1;
1380
	}
J
Jaegeuk Kim 已提交
1381

1382
	/* Currently, support only 4KB page cache size */
1383
	if (F2FS_BLKSIZE != PAGE_SIZE) {
1384
		f2fs_msg(sb, KERN_INFO,
1385
			"Invalid page_cache_size (%lu), supports only 4KB\n",
1386
			PAGE_SIZE);
1387 1388 1389
		return 1;
	}

J
Jaegeuk Kim 已提交
1390 1391
	/* Currently, support only 4KB block size */
	blocksize = 1 << le32_to_cpu(raw_super->log_blocksize);
1392
	if (blocksize != F2FS_BLKSIZE) {
1393 1394 1395
		f2fs_msg(sb, KERN_INFO,
			"Invalid blocksize (%u), supports only 4KB\n",
			blocksize);
J
Jaegeuk Kim 已提交
1396
		return 1;
1397
	}
1398

1399 1400 1401 1402 1403 1404 1405 1406
	/* check log blocks per segment */
	if (le32_to_cpu(raw_super->log_blocks_per_seg) != 9) {
		f2fs_msg(sb, KERN_INFO,
			"Invalid log blocks per segment (%u)\n",
			le32_to_cpu(raw_super->log_blocks_per_seg));
		return 1;
	}

C
Chao Yu 已提交
1407 1408 1409 1410 1411 1412 1413
	/* Currently, support 512/1024/2048/4096 bytes sector size */
	if (le32_to_cpu(raw_super->log_sectorsize) >
				F2FS_MAX_LOG_SECTOR_SIZE ||
		le32_to_cpu(raw_super->log_sectorsize) <
				F2FS_MIN_LOG_SECTOR_SIZE) {
		f2fs_msg(sb, KERN_INFO, "Invalid log sectorsize (%u)",
			le32_to_cpu(raw_super->log_sectorsize));
J
Jaegeuk Kim 已提交
1414
		return 1;
1415
	}
C
Chao Yu 已提交
1416 1417 1418 1419 1420 1421 1422
	if (le32_to_cpu(raw_super->log_sectors_per_block) +
		le32_to_cpu(raw_super->log_sectorsize) !=
			F2FS_MAX_LOG_SECTOR_SIZE) {
		f2fs_msg(sb, KERN_INFO,
			"Invalid log sectors per block(%u) log sectorsize(%u)",
			le32_to_cpu(raw_super->log_sectors_per_block),
			le32_to_cpu(raw_super->log_sectorsize));
J
Jaegeuk Kim 已提交
1423
		return 1;
1424
	}
1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438

	/* check reserved ino info */
	if (le32_to_cpu(raw_super->node_ino) != 1 ||
		le32_to_cpu(raw_super->meta_ino) != 2 ||
		le32_to_cpu(raw_super->root_ino) != 3) {
		f2fs_msg(sb, KERN_INFO,
			"Invalid Fs Meta Ino: node(%u) meta(%u) root(%u)",
			le32_to_cpu(raw_super->node_ino),
			le32_to_cpu(raw_super->meta_ino),
			le32_to_cpu(raw_super->root_ino));
		return 1;
	}

	/* check CP/SIT/NAT/SSA/MAIN_AREA area boundary */
1439
	if (sanity_check_area_boundary(sbi, bh))
1440 1441
		return 1;

J
Jaegeuk Kim 已提交
1442 1443 1444
	return 0;
}

1445
int sanity_check_ckpt(struct f2fs_sb_info *sbi)
J
Jaegeuk Kim 已提交
1446 1447
{
	unsigned int total, fsmeta;
1448 1449
	struct f2fs_super_block *raw_super = F2FS_RAW_SUPER(sbi);
	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
J
Jaegeuk Kim 已提交
1450
	unsigned int ovp_segments, reserved_segments;
J
Jaegeuk Kim 已提交
1451 1452 1453 1454 1455 1456 1457 1458

	total = le32_to_cpu(raw_super->segment_count);
	fsmeta = le32_to_cpu(raw_super->segment_count_ckpt);
	fsmeta += le32_to_cpu(raw_super->segment_count_sit);
	fsmeta += le32_to_cpu(raw_super->segment_count_nat);
	fsmeta += le32_to_cpu(ckpt->rsvd_segment_count);
	fsmeta += le32_to_cpu(raw_super->segment_count_ssa);

1459
	if (unlikely(fsmeta >= total))
J
Jaegeuk Kim 已提交
1460
		return 1;
1461

J
Jaegeuk Kim 已提交
1462 1463 1464 1465 1466 1467 1468 1469 1470 1471
	ovp_segments = le32_to_cpu(ckpt->overprov_segment_count);
	reserved_segments = le32_to_cpu(ckpt->rsvd_segment_count);

	if (unlikely(fsmeta < F2FS_MIN_SEGMENTS ||
			ovp_segments == 0 || reserved_segments == 0)) {
		f2fs_msg(sbi->sb, KERN_ERR,
			"Wrong layout: check mkfs.f2fs version");
		return 1;
	}

1472
	if (unlikely(f2fs_cp_error(sbi))) {
1473 1474 1475
		f2fs_msg(sbi->sb, KERN_ERR, "A bug case: need to run fsck");
		return 1;
	}
J
Jaegeuk Kim 已提交
1476 1477 1478 1479 1480 1481
	return 0;
}

static void init_sb_info(struct f2fs_sb_info *sbi)
{
	struct f2fs_super_block *raw_super = sbi->raw_super;
1482
	int i;
J
Jaegeuk Kim 已提交
1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498

	sbi->log_sectors_per_block =
		le32_to_cpu(raw_super->log_sectors_per_block);
	sbi->log_blocksize = le32_to_cpu(raw_super->log_blocksize);
	sbi->blocksize = 1 << sbi->log_blocksize;
	sbi->log_blocks_per_seg = le32_to_cpu(raw_super->log_blocks_per_seg);
	sbi->blocks_per_seg = 1 << sbi->log_blocks_per_seg;
	sbi->segs_per_sec = le32_to_cpu(raw_super->segs_per_sec);
	sbi->secs_per_zone = le32_to_cpu(raw_super->secs_per_zone);
	sbi->total_sections = le32_to_cpu(raw_super->section_count);
	sbi->total_node_count =
		(le32_to_cpu(raw_super->segment_count_nat) / 2)
			* sbi->blocks_per_seg * NAT_ENTRY_PER_BLOCK;
	sbi->root_ino_num = le32_to_cpu(raw_super->root_ino);
	sbi->node_ino_num = le32_to_cpu(raw_super->node_ino);
	sbi->meta_ino_num = le32_to_cpu(raw_super->meta_ino);
1499
	sbi->cur_victim_sec = NULL_SECNO;
1500
	sbi->max_victim_search = DEF_MAX_VICTIM_SEARCH;
J
Jaegeuk Kim 已提交
1501

1502
	sbi->dir_level = DEF_DIR_LEVEL;
1503
	sbi->interval_time[CP_TIME] = DEF_CP_INTERVAL;
1504
	sbi->interval_time[REQ_TIME] = DEF_IDLE_INTERVAL;
1505
	clear_sbi_flag(sbi, SBI_NEED_FSCK);
1506

1507 1508 1509
	for (i = 0; i < NR_COUNT_TYPE; i++)
		atomic_set(&sbi->nr_pages[i], 0);

1510 1511
	INIT_LIST_HEAD(&sbi->s_list);
	mutex_init(&sbi->umount_mutex);
1512 1513
	mutex_init(&sbi->wio_mutex[NODE]);
	mutex_init(&sbi->wio_mutex[DATA]);
1514
	spin_lock_init(&sbi->cp_lock);
J
Jaegeuk Kim 已提交
1515 1516
}

1517 1518
static int init_percpu_info(struct f2fs_sb_info *sbi)
{
1519
	int err;
1520

1521 1522 1523 1524 1525
	err = percpu_counter_init(&sbi->alloc_valid_block_count, 0, GFP_KERNEL);
	if (err)
		return err;

	return percpu_counter_init(&sbi->total_valid_inode_count, 0,
1526
								GFP_KERNEL);
1527 1528
}

1529
#ifdef CONFIG_BLK_DEV_ZONED
J
Jaegeuk Kim 已提交
1530
static int init_blkz_info(struct f2fs_sb_info *sbi, int devi)
1531
{
J
Jaegeuk Kim 已提交
1532
	struct block_device *bdev = FDEV(devi).bdev;
1533 1534 1535 1536 1537 1538 1539 1540 1541 1542
	sector_t nr_sectors = bdev->bd_part->nr_sects;
	sector_t sector = 0;
	struct blk_zone *zones;
	unsigned int i, nr_zones;
	unsigned int n = 0;
	int err = -EIO;

	if (!f2fs_sb_mounted_blkzoned(sbi->sb))
		return 0;

J
Jaegeuk Kim 已提交
1543
	if (sbi->blocks_per_blkz && sbi->blocks_per_blkz !=
1544
				SECTOR_TO_BLOCK(bdev_zone_sectors(bdev)))
J
Jaegeuk Kim 已提交
1545
		return -EINVAL;
1546
	sbi->blocks_per_blkz = SECTOR_TO_BLOCK(bdev_zone_sectors(bdev));
J
Jaegeuk Kim 已提交
1547 1548 1549
	if (sbi->log_blocks_per_blkz && sbi->log_blocks_per_blkz !=
				__ilog2_u32(sbi->blocks_per_blkz))
		return -EINVAL;
1550
	sbi->log_blocks_per_blkz = __ilog2_u32(sbi->blocks_per_blkz);
J
Jaegeuk Kim 已提交
1551 1552
	FDEV(devi).nr_blkz = SECTOR_TO_BLOCK(nr_sectors) >>
					sbi->log_blocks_per_blkz;
1553
	if (nr_sectors & (bdev_zone_sectors(bdev) - 1))
J
Jaegeuk Kim 已提交
1554
		FDEV(devi).nr_blkz++;
1555

J
Jaegeuk Kim 已提交
1556 1557
	FDEV(devi).blkz_type = kmalloc(FDEV(devi).nr_blkz, GFP_KERNEL);
	if (!FDEV(devi).blkz_type)
1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581
		return -ENOMEM;

#define F2FS_REPORT_NR_ZONES   4096

	zones = kcalloc(F2FS_REPORT_NR_ZONES, sizeof(struct blk_zone),
			GFP_KERNEL);
	if (!zones)
		return -ENOMEM;

	/* Get block zones type */
	while (zones && sector < nr_sectors) {

		nr_zones = F2FS_REPORT_NR_ZONES;
		err = blkdev_report_zones(bdev, sector,
					  zones, &nr_zones,
					  GFP_KERNEL);
		if (err)
			break;
		if (!nr_zones) {
			err = -EIO;
			break;
		}

		for (i = 0; i < nr_zones; i++) {
J
Jaegeuk Kim 已提交
1582
			FDEV(devi).blkz_type[n] = zones[i].type;
1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593
			sector += zones[i].len;
			n++;
		}
	}

	kfree(zones);

	return err;
}
#endif

1594 1595
/*
 * Read f2fs raw super block.
1596 1597 1598
 * Because we have two copies of super block, so read both of them
 * to get the first valid one. If any one of them is broken, we pass
 * them recovery flag back to the caller.
1599
 */
1600
static int read_raw_super_block(struct f2fs_sb_info *sbi,
1601
			struct f2fs_super_block **raw_super,
1602
			int *valid_super_block, int *recovery)
1603
{
1604
	struct super_block *sb = sbi->sb;
1605
	int block;
1606
	struct buffer_head *bh;
1607
	struct f2fs_super_block *super;
1608
	int err = 0;
1609

Y
Yunlei He 已提交
1610 1611 1612
	super = kzalloc(sizeof(struct f2fs_super_block), GFP_KERNEL);
	if (!super)
		return -ENOMEM;
1613 1614 1615 1616 1617

	for (block = 0; block < 2; block++) {
		bh = sb_bread(sb, block);
		if (!bh) {
			f2fs_msg(sb, KERN_ERR, "Unable to read %dth superblock",
1618
				block + 1);
1619 1620 1621
			err = -EIO;
			continue;
		}
1622

1623
		/* sanity checking of raw super */
1624
		if (sanity_check_raw_super(sbi, bh)) {
1625 1626 1627 1628 1629 1630 1631
			f2fs_msg(sb, KERN_ERR,
				"Can't find valid F2FS filesystem in %dth superblock",
				block + 1);
			err = -EINVAL;
			brelse(bh);
			continue;
		}
1632

1633
		if (!*raw_super) {
1634 1635
			memcpy(super, bh->b_data + F2FS_SUPER_OFFSET,
							sizeof(*super));
1636 1637 1638 1639
			*valid_super_block = block;
			*raw_super = super;
		}
		brelse(bh);
1640 1641
	}

1642 1643 1644
	/* Fail to read any one of the superblocks*/
	if (err < 0)
		*recovery = 1;
1645 1646

	/* No valid superblock */
1647
	if (!*raw_super)
Y
Yunlei He 已提交
1648
		kfree(super);
1649 1650
	else
		err = 0;
1651

1652
	return err;
1653 1654
}

1655
int f2fs_commit_super(struct f2fs_sb_info *sbi, bool recover)
J
Jaegeuk Kim 已提交
1656
{
J
Jaegeuk Kim 已提交
1657
	struct buffer_head *bh;
J
Jaegeuk Kim 已提交
1658 1659
	int err;

1660 1661 1662
	if ((recover && f2fs_readonly(sbi->sb)) ||
				bdev_read_only(sbi->sb->s_bdev)) {
		set_sbi_flag(sbi, SBI_NEED_SB_WRITE);
1663
		return -EROFS;
1664
	}
1665

1666 1667
	/* write back-up superblock first */
	bh = sb_getblk(sbi->sb, sbi->valid_super_block ? 0: 1);
J
Jaegeuk Kim 已提交
1668 1669
	if (!bh)
		return -EIO;
1670
	err = __f2fs_commit_super(bh, F2FS_RAW_SUPER(sbi));
J
Jaegeuk Kim 已提交
1671
	brelse(bh);
C
Chao Yu 已提交
1672 1673 1674

	/* if we are in recovery path, skip writing valid superblock */
	if (recover || err)
J
Jaegeuk Kim 已提交
1675
		return err;
J
Jaegeuk Kim 已提交
1676 1677

	/* write current valid superblock */
1678 1679 1680 1681 1682 1683
	bh = sb_getblk(sbi->sb, sbi->valid_super_block);
	if (!bh)
		return -EIO;
	err = __f2fs_commit_super(bh, F2FS_RAW_SUPER(sbi));
	brelse(bh);
	return err;
J
Jaegeuk Kim 已提交
1684 1685
}

J
Jaegeuk Kim 已提交
1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756
static int f2fs_scan_devices(struct f2fs_sb_info *sbi)
{
	struct f2fs_super_block *raw_super = F2FS_RAW_SUPER(sbi);
	int i;

	for (i = 0; i < MAX_DEVICES; i++) {
		if (!RDEV(i).path[0])
			return 0;

		if (i == 0) {
			sbi->devs = kzalloc(sizeof(struct f2fs_dev_info) *
						MAX_DEVICES, GFP_KERNEL);
			if (!sbi->devs)
				return -ENOMEM;
		}

		memcpy(FDEV(i).path, RDEV(i).path, MAX_PATH_LEN);
		FDEV(i).total_segments = le32_to_cpu(RDEV(i).total_segments);
		if (i == 0) {
			FDEV(i).start_blk = 0;
			FDEV(i).end_blk = FDEV(i).start_blk +
				(FDEV(i).total_segments <<
				sbi->log_blocks_per_seg) - 1 +
				le32_to_cpu(raw_super->segment0_blkaddr);
		} else {
			FDEV(i).start_blk = FDEV(i - 1).end_blk + 1;
			FDEV(i).end_blk = FDEV(i).start_blk +
				(FDEV(i).total_segments <<
				sbi->log_blocks_per_seg) - 1;
		}

		FDEV(i).bdev = blkdev_get_by_path(FDEV(i).path,
					sbi->sb->s_mode, sbi->sb->s_type);
		if (IS_ERR(FDEV(i).bdev))
			return PTR_ERR(FDEV(i).bdev);

		/* to release errored devices */
		sbi->s_ndevs = i + 1;

#ifdef CONFIG_BLK_DEV_ZONED
		if (bdev_zoned_model(FDEV(i).bdev) == BLK_ZONED_HM &&
				!f2fs_sb_mounted_blkzoned(sbi->sb)) {
			f2fs_msg(sbi->sb, KERN_ERR,
				"Zoned block device feature not enabled\n");
			return -EINVAL;
		}
		if (bdev_zoned_model(FDEV(i).bdev) != BLK_ZONED_NONE) {
			if (init_blkz_info(sbi, i)) {
				f2fs_msg(sbi->sb, KERN_ERR,
					"Failed to initialize F2FS blkzone information");
				return -EINVAL;
			}
			f2fs_msg(sbi->sb, KERN_INFO,
				"Mount Device [%2d]: %20s, %8u, %8x - %8x (zone: %s)",
				i, FDEV(i).path,
				FDEV(i).total_segments,
				FDEV(i).start_blk, FDEV(i).end_blk,
				bdev_zoned_model(FDEV(i).bdev) == BLK_ZONED_HA ?
				"Host-aware" : "Host-managed");
			continue;
		}
#endif
		f2fs_msg(sbi->sb, KERN_INFO,
			"Mount Device [%2d]: %20s, %8u, %8x - %8x",
				i, FDEV(i).path,
				FDEV(i).total_segments,
				FDEV(i).start_blk, FDEV(i).end_blk);
	}
	return 0;
}

J
Jaegeuk Kim 已提交
1757 1758 1759
static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
{
	struct f2fs_sb_info *sbi;
1760
	struct f2fs_super_block *raw_super;
J
Jaegeuk Kim 已提交
1761
	struct inode *root;
1762
	int err;
1763
	bool retry = true, need_fsck = false;
1764
	char *options = NULL;
1765
	int recovery, i, valid_super_block;
1766
	struct curseg_info *seg_i;
J
Jaegeuk Kim 已提交
1767

1768
try_onemore:
1769 1770
	err = -EINVAL;
	raw_super = NULL;
1771
	valid_super_block = -1;
1772 1773
	recovery = 0;

J
Jaegeuk Kim 已提交
1774 1775 1776 1777 1778
	/* allocate memory for f2fs-specific super block info */
	sbi = kzalloc(sizeof(struct f2fs_sb_info), GFP_KERNEL);
	if (!sbi)
		return -ENOMEM;

1779 1780
	sbi->sb = sb;

K
Keith Mok 已提交
1781 1782 1783 1784 1785 1786 1787 1788 1789
	/* Load the checksum driver */
	sbi->s_chksum_driver = crypto_alloc_shash("crc32", 0, 0);
	if (IS_ERR(sbi->s_chksum_driver)) {
		f2fs_msg(sb, KERN_ERR, "Cannot load crc32 driver.");
		err = PTR_ERR(sbi->s_chksum_driver);
		sbi->s_chksum_driver = NULL;
		goto free_sbi;
	}

1790
	/* set a block size */
1791
	if (unlikely(!sb_set_blocksize(sb, F2FS_BLKSIZE))) {
1792
		f2fs_msg(sb, KERN_ERR, "unable to set blocksize");
J
Jaegeuk Kim 已提交
1793
		goto free_sbi;
1794
	}
J
Jaegeuk Kim 已提交
1795

1796
	err = read_raw_super_block(sbi, &raw_super, &valid_super_block,
1797
								&recovery);
1798 1799 1800
	if (err)
		goto free_sbi;

1801
	sb->s_fs_info = sbi;
1802 1803
	sbi->raw_super = raw_super;

1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815
	/*
	 * The BLKZONED feature indicates that the drive was formatted with
	 * zone alignment optimization. This is optional for host-aware
	 * devices, but mandatory for host-managed zoned block devices.
	 */
#ifndef CONFIG_BLK_DEV_ZONED
	if (f2fs_sb_mounted_blkzoned(sb)) {
		f2fs_msg(sb, KERN_ERR,
			 "Zoned block device support is not enabled\n");
		goto free_sb_buf;
	}
#endif
1816
	default_options(sbi);
J
Jaegeuk Kim 已提交
1817
	/* parse mount options */
1818 1819 1820
	options = kstrdup((const char *)data, GFP_KERNEL);
	if (data && !options) {
		err = -ENOMEM;
J
Jaegeuk Kim 已提交
1821
		goto free_sb_buf;
1822 1823 1824 1825 1826
	}

	err = parse_options(sb, options);
	if (err)
		goto free_options;
J
Jaegeuk Kim 已提交
1827

C
Chao Yu 已提交
1828 1829 1830
	sbi->max_file_blocks = max_file_blocks();
	sb->s_maxbytes = sbi->max_file_blocks <<
				le32_to_cpu(raw_super->log_blocksize);
J
Jaegeuk Kim 已提交
1831 1832 1833 1834
	sb->s_max_links = F2FS_LINK_MAX;
	get_random_bytes(&sbi->s_next_generation, sizeof(u32));

	sb->s_op = &f2fs_sops;
1835
	sb->s_cop = &f2fs_cryptops;
J
Jaegeuk Kim 已提交
1836 1837 1838 1839 1840 1841 1842 1843 1844
	sb->s_xattr = f2fs_xattr_handlers;
	sb->s_export_op = &f2fs_export_ops;
	sb->s_magic = F2FS_SUPER_MAGIC;
	sb->s_time_gran = 1;
	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
		(test_opt(sbi, POSIX_ACL) ? MS_POSIXACL : 0);
	memcpy(sb->s_uuid, raw_super->uuid, sizeof(raw_super->uuid));

	/* init f2fs-specific super block info */
1845
	sbi->valid_super_block = valid_super_block;
J
Jaegeuk Kim 已提交
1846 1847
	mutex_init(&sbi->gc_mutex);
	mutex_init(&sbi->cp_mutex);
1848
	init_rwsem(&sbi->node_write);
1849 1850 1851

	/* disallow all the data/node/meta page writes */
	set_sbi_flag(sbi, SBI_POR_DOING);
J
Jaegeuk Kim 已提交
1852
	spin_lock_init(&sbi->stat_lock);
1853

1854
	init_rwsem(&sbi->read_io.io_rwsem);
J
Jaegeuk Kim 已提交
1855 1856 1857
	sbi->read_io.sbi = sbi;
	sbi->read_io.bio = NULL;
	for (i = 0; i < NR_PAGE_TYPE; i++) {
1858
		init_rwsem(&sbi->write_io[i].io_rwsem);
J
Jaegeuk Kim 已提交
1859 1860 1861
		sbi->write_io[i].sbi = sbi;
		sbi->write_io[i].bio = NULL;
	}
1862

1863
	init_rwsem(&sbi->cp_rwsem);
1864
	init_waitqueue_head(&sbi->cp_wait);
J
Jaegeuk Kim 已提交
1865 1866
	init_sb_info(sbi);

1867 1868 1869 1870
	err = init_percpu_info(sbi);
	if (err)
		goto free_options;

J
Jaegeuk Kim 已提交
1871 1872 1873
	/* get an inode for meta space */
	sbi->meta_inode = f2fs_iget(sb, F2FS_META_INO(sbi));
	if (IS_ERR(sbi->meta_inode)) {
1874
		f2fs_msg(sb, KERN_ERR, "Failed to read F2FS meta data inode");
J
Jaegeuk Kim 已提交
1875
		err = PTR_ERR(sbi->meta_inode);
1876
		goto free_options;
J
Jaegeuk Kim 已提交
1877 1878 1879
	}

	err = get_valid_checkpoint(sbi);
1880 1881
	if (err) {
		f2fs_msg(sb, KERN_ERR, "Failed to get valid F2FS checkpoint");
J
Jaegeuk Kim 已提交
1882
		goto free_meta_inode;
1883
	}
J
Jaegeuk Kim 已提交
1884

J
Jaegeuk Kim 已提交
1885 1886 1887 1888 1889 1890 1891
	/* Initialize device list */
	err = f2fs_scan_devices(sbi);
	if (err) {
		f2fs_msg(sb, KERN_ERR, "Failed to find devices");
		goto free_devices;
	}

J
Jaegeuk Kim 已提交
1892 1893
	sbi->total_valid_node_count =
				le32_to_cpu(sbi->ckpt->valid_node_count);
1894 1895
	percpu_counter_set(&sbi->total_valid_inode_count,
				le32_to_cpu(sbi->ckpt->valid_inode_count));
J
Jaegeuk Kim 已提交
1896 1897 1898 1899
	sbi->user_block_count = le64_to_cpu(sbi->ckpt->user_block_count);
	sbi->total_valid_block_count =
				le64_to_cpu(sbi->ckpt->valid_block_count);
	sbi->last_valid_block_count = sbi->total_valid_block_count;
1900

1901 1902 1903 1904
	for (i = 0; i < NR_INODE_TYPE; i++) {
		INIT_LIST_HEAD(&sbi->inode_list[i]);
		spin_lock_init(&sbi->inode_lock[i]);
	}
J
Jaegeuk Kim 已提交
1905

C
Chao Yu 已提交
1906 1907
	init_extent_cache_info(sbi);

J
Jaegeuk Kim 已提交
1908
	init_ino_entry_info(sbi);
J
Jaegeuk Kim 已提交
1909 1910 1911

	/* setup f2fs internal modules */
	err = build_segment_manager(sbi);
1912 1913 1914
	if (err) {
		f2fs_msg(sb, KERN_ERR,
			"Failed to initialize F2FS segment manager");
J
Jaegeuk Kim 已提交
1915
		goto free_sm;
1916
	}
J
Jaegeuk Kim 已提交
1917
	err = build_node_manager(sbi);
1918 1919 1920
	if (err) {
		f2fs_msg(sb, KERN_ERR,
			"Failed to initialize F2FS node manager");
J
Jaegeuk Kim 已提交
1921
		goto free_nm;
1922
	}
J
Jaegeuk Kim 已提交
1923

1924 1925 1926 1927 1928 1929 1930 1931 1932
	/* For write statistics */
	if (sb->s_bdev->bd_part)
		sbi->sectors_written_start =
			(u64)part_stat_read(sb->s_bdev->bd_part, sectors[1]);

	/* Read accumulated write IO statistics if exists */
	seg_i = CURSEG_I(sbi, CURSEG_HOT_NODE);
	if (__exist_node_summaries(sbi))
		sbi->kbytes_written =
1933
			le64_to_cpu(seg_i->journal->info.kbytes_written);
1934

J
Jaegeuk Kim 已提交
1935 1936 1937 1938 1939
	build_gc_manager(sbi);

	/* get an inode for node space */
	sbi->node_inode = f2fs_iget(sb, F2FS_NODE_INO(sbi));
	if (IS_ERR(sbi->node_inode)) {
1940
		f2fs_msg(sb, KERN_ERR, "Failed to read node inode");
J
Jaegeuk Kim 已提交
1941 1942 1943 1944
		err = PTR_ERR(sbi->node_inode);
		goto free_nm;
	}

1945 1946
	f2fs_join_shrinker(sbi);

J
Jaegeuk Kim 已提交
1947
	/* if there are nt orphan nodes free them */
1948 1949 1950
	err = recover_orphan_inodes(sbi);
	if (err)
		goto free_node_inode;
J
Jaegeuk Kim 已提交
1951 1952 1953 1954

	/* read root inode and dentry */
	root = f2fs_iget(sb, F2FS_ROOT_INO(sbi));
	if (IS_ERR(root)) {
1955
		f2fs_msg(sb, KERN_ERR, "Failed to read root inode");
J
Jaegeuk Kim 已提交
1956 1957 1958
		err = PTR_ERR(root);
		goto free_node_inode;
	}
1959
	if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) {
1960
		iput(root);
1961
		err = -EINVAL;
1962
		goto free_node_inode;
1963
	}
J
Jaegeuk Kim 已提交
1964 1965 1966 1967 1968 1969 1970 1971 1972

	sb->s_root = d_make_root(root); /* allocate root dentry */
	if (!sb->s_root) {
		err = -ENOMEM;
		goto free_root_inode;
	}

	err = f2fs_build_stats(sbi);
	if (err)
1973
		goto free_root_inode;
J
Jaegeuk Kim 已提交
1974

1975 1976 1977
	if (f2fs_proc_root)
		sbi->s_proc = proc_mkdir(sb->s_id, f2fs_proc_root);

1978
	if (sbi->s_proc) {
1979 1980
		proc_create_data("segment_info", S_IRUGO, sbi->s_proc,
				 &f2fs_seq_segment_info_fops, sb);
1981 1982 1983
		proc_create_data("segment_bits", S_IRUGO, sbi->s_proc,
				 &f2fs_seq_segment_bits_fops, sb);
	}
1984

1985 1986 1987 1988 1989
	sbi->s_kobj.kset = f2fs_kset;
	init_completion(&sbi->s_kobj_unregister);
	err = kobject_init_and_add(&sbi->s_kobj, &f2fs_ktype, NULL,
							"%s", sb->s_id);
	if (err)
1990
		goto free_proc;
1991

1992 1993
	/* recover fsynced data */
	if (!test_opt(sbi, DISABLE_ROLL_FORWARD)) {
1994 1995 1996 1997 1998
		/*
		 * mount should be failed, when device has readonly mode, and
		 * previous checkpoint was not done by clean system shutdown.
		 */
		if (bdev_read_only(sb->s_bdev) &&
1999
				!is_set_ckpt_flags(sbi, CP_UMOUNT_FLAG)) {
2000 2001 2002
			err = -EROFS;
			goto free_kobj;
		}
2003 2004 2005 2006

		if (need_fsck)
			set_sbi_flag(sbi, SBI_NEED_FSCK);

2007 2008 2009
		if (!retry)
			goto skip_recovery;

2010 2011
		err = recover_fsync_data(sbi, false);
		if (err < 0) {
2012
			need_fsck = true;
2013
			f2fs_msg(sb, KERN_ERR,
2014
				"Cannot recover all fsync data errno=%d", err);
2015 2016
			goto free_kobj;
		}
2017 2018 2019 2020 2021 2022 2023 2024 2025
	} else {
		err = recover_fsync_data(sbi, true);

		if (!f2fs_readonly(sb) && err > 0) {
			err = -EINVAL;
			f2fs_msg(sb, KERN_ERR,
				"Need to recover fsync data");
			goto free_kobj;
		}
2026
	}
2027
skip_recovery:
2028 2029
	/* recover_fsync_data() cleared this already */
	clear_sbi_flag(sbi, SBI_POR_DOING);
2030

2031 2032 2033 2034
	/*
	 * If filesystem is not mounted as read-only then
	 * do start the gc_thread.
	 */
2035
	if (test_opt(sbi, BG_GC) && !f2fs_readonly(sb)) {
2036 2037 2038 2039 2040
		/* After POR, we can run background GC thread.*/
		err = start_gc_thread(sbi);
		if (err)
			goto free_kobj;
	}
2041
	kfree(options);
2042 2043

	/* recover broken superblock */
2044
	if (recovery) {
2045 2046
		err = f2fs_commit_super(sbi, true);
		f2fs_msg(sb, KERN_INFO,
2047
			"Try to recover %dth superblock, ret: %d",
2048
			sbi->valid_super_block ? 1 : 2, err);
2049 2050
	}

2051
	f2fs_update_time(sbi, CP_TIME);
2052
	f2fs_update_time(sbi, REQ_TIME);
J
Jaegeuk Kim 已提交
2053
	return 0;
2054 2055

free_kobj:
2056
	f2fs_sync_inode_meta(sbi);
2057
	kobject_del(&sbi->s_kobj);
2058 2059
	kobject_put(&sbi->s_kobj);
	wait_for_completion(&sbi->s_kobj_unregister);
2060
free_proc:
2061 2062
	if (sbi->s_proc) {
		remove_proc_entry("segment_info", sbi->s_proc);
2063
		remove_proc_entry("segment_bits", sbi->s_proc);
2064 2065 2066
		remove_proc_entry(sb->s_id, f2fs_proc_root);
	}
	f2fs_destroy_stats(sbi);
J
Jaegeuk Kim 已提交
2067 2068 2069 2070
free_root_inode:
	dput(sb->s_root);
	sb->s_root = NULL;
free_node_inode:
2071
	truncate_inode_pages_final(NODE_MAPPING(sbi));
2072
	mutex_lock(&sbi->umount_mutex);
2073
	release_ino_entry(sbi, true);
2074
	f2fs_leave_shrinker(sbi);
2075 2076 2077 2078 2079 2080 2081
	/*
	 * Some dirty meta pages can be produced by recover_orphan_inodes()
	 * failed by EIO. Then, iput(node_inode) can trigger balance_fs_bg()
	 * followed by write_checkpoint() through f2fs_write_node_pages(), which
	 * falls into an infinite loop in sync_meta_pages().
	 */
	truncate_inode_pages_final(META_MAPPING(sbi));
J
Jaegeuk Kim 已提交
2082
	iput(sbi->node_inode);
2083
	mutex_unlock(&sbi->umount_mutex);
J
Jaegeuk Kim 已提交
2084 2085 2086 2087
free_nm:
	destroy_node_manager(sbi);
free_sm:
	destroy_segment_manager(sbi);
J
Jaegeuk Kim 已提交
2088 2089
free_devices:
	destroy_device_list(sbi);
J
Jaegeuk Kim 已提交
2090 2091 2092 2093
	kfree(sbi->ckpt);
free_meta_inode:
	make_bad_inode(sbi->meta_inode);
	iput(sbi->meta_inode);
2094
free_options:
2095
	destroy_percpu_info(sbi);
2096
	kfree(options);
J
Jaegeuk Kim 已提交
2097
free_sb_buf:
Y
Yunlei He 已提交
2098
	kfree(raw_super);
J
Jaegeuk Kim 已提交
2099
free_sbi:
K
Keith Mok 已提交
2100 2101
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
J
Jaegeuk Kim 已提交
2102
	kfree(sbi);
2103 2104 2105

	/* give only one another chance */
	if (retry) {
2106
		retry = false;
2107 2108 2109
		shrink_dcache_sb(sb);
		goto try_onemore;
	}
J
Jaegeuk Kim 已提交
2110 2111 2112 2113 2114 2115 2116 2117 2118
	return err;
}

static struct dentry *f2fs_mount(struct file_system_type *fs_type, int flags,
			const char *dev_name, void *data)
{
	return mount_bdev(fs_type, flags, dev_name, data, f2fs_fill_super);
}

2119 2120 2121
static void kill_f2fs_super(struct super_block *sb)
{
	if (sb->s_root)
2122
		set_sbi_flag(F2FS_SB(sb), SBI_IS_CLOSE);
2123 2124 2125
	kill_block_super(sb);
}

J
Jaegeuk Kim 已提交
2126 2127 2128 2129
static struct file_system_type f2fs_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "f2fs",
	.mount		= f2fs_mount,
2130
	.kill_sb	= kill_f2fs_super,
J
Jaegeuk Kim 已提交
2131 2132
	.fs_flags	= FS_REQUIRES_DEV,
};
2133
MODULE_ALIAS_FS("f2fs");
J
Jaegeuk Kim 已提交
2134

2135
static int __init init_inodecache(void)
J
Jaegeuk Kim 已提交
2136
{
2137 2138 2139
	f2fs_inode_cachep = kmem_cache_create("f2fs_inode_cache",
			sizeof(struct f2fs_inode_info), 0,
			SLAB_RECLAIM_ACCOUNT|SLAB_ACCOUNT, NULL);
2140
	if (!f2fs_inode_cachep)
J
Jaegeuk Kim 已提交
2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158
		return -ENOMEM;
	return 0;
}

static void destroy_inodecache(void)
{
	/*
	 * Make sure all delayed rcu free inodes are flushed before we
	 * destroy cache.
	 */
	rcu_barrier();
	kmem_cache_destroy(f2fs_inode_cachep);
}

static int __init init_f2fs_fs(void)
{
	int err;

2159 2160
	f2fs_build_trace_ios();

J
Jaegeuk Kim 已提交
2161 2162 2163 2164 2165
	err = init_inodecache();
	if (err)
		goto fail;
	err = create_node_manager_caches();
	if (err)
2166
		goto free_inodecache;
2167
	err = create_segment_manager_caches();
J
Jaegeuk Kim 已提交
2168
	if (err)
2169
		goto free_node_manager_caches;
J
Jaegeuk Kim 已提交
2170 2171
	err = create_checkpoint_caches();
	if (err)
2172
		goto free_segment_manager_caches;
C
Chao Yu 已提交
2173 2174 2175
	err = create_extent_cache();
	if (err)
		goto free_checkpoint_caches;
2176
	f2fs_kset = kset_create_and_add("f2fs", NULL, fs_kobj);
2177 2178
	if (!f2fs_kset) {
		err = -ENOMEM;
C
Chao Yu 已提交
2179
		goto free_extent_cache;
2180
	}
2181
	err = register_shrinker(&f2fs_shrinker_info);
2182
	if (err)
2183
		goto free_kset;
2184 2185 2186 2187

	err = register_filesystem(&f2fs_fs_type);
	if (err)
		goto free_shrinker;
2188 2189 2190
	err = f2fs_create_root_stats();
	if (err)
		goto free_filesystem;
2191
	f2fs_proc_root = proc_mkdir("fs/f2fs", NULL);
2192 2193
	return 0;

2194 2195
free_filesystem:
	unregister_filesystem(&f2fs_fs_type);
2196 2197
free_shrinker:
	unregister_shrinker(&f2fs_shrinker_info);
2198 2199
free_kset:
	kset_unregister(f2fs_kset);
C
Chao Yu 已提交
2200 2201
free_extent_cache:
	destroy_extent_cache();
2202 2203
free_checkpoint_caches:
	destroy_checkpoint_caches();
2204 2205
free_segment_manager_caches:
	destroy_segment_manager_caches();
2206 2207 2208 2209
free_node_manager_caches:
	destroy_node_manager_caches();
free_inodecache:
	destroy_inodecache();
J
Jaegeuk Kim 已提交
2210 2211 2212 2213 2214 2215
fail:
	return err;
}

static void __exit exit_f2fs_fs(void)
{
2216
	remove_proc_entry("fs/f2fs", NULL);
2217
	f2fs_destroy_root_stats();
J
Jaegeuk Kim 已提交
2218
	unregister_filesystem(&f2fs_fs_type);
T
Tiezhu Yang 已提交
2219 2220
	unregister_shrinker(&f2fs_shrinker_info);
	kset_unregister(f2fs_kset);
W
Wanpeng Li 已提交
2221
	destroy_extent_cache();
J
Jaegeuk Kim 已提交
2222
	destroy_checkpoint_caches();
2223
	destroy_segment_manager_caches();
J
Jaegeuk Kim 已提交
2224 2225
	destroy_node_manager_caches();
	destroy_inodecache();
2226
	f2fs_destroy_trace_ios();
J
Jaegeuk Kim 已提交
2227 2228 2229 2230 2231 2232 2233 2234
}

module_init(init_f2fs_fs)
module_exit(exit_f2fs_fs)

MODULE_AUTHOR("Samsung Electronics's Praesto Team");
MODULE_DESCRIPTION("Flash Friendly File System");
MODULE_LICENSE("GPL");
2235