super.c 59.8 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",
52
	[FAULT_TRUNCATE]	= "truncate fail",
C
Chao Yu 已提交
53
	[FAULT_IO]		= "IO error",
54
	[FAULT_CHECKPOINT]	= "checkpoint error",
J
Jaegeuk Kim 已提交
55
};
S
Sheng Yong 已提交
56

57 58
static void f2fs_build_fault_attr(struct f2fs_sb_info *sbi,
						unsigned int rate)
S
Sheng Yong 已提交
59
{
60 61
	struct f2fs_fault_info *ffi = &sbi->fault_info;

S
Sheng Yong 已提交
62
	if (rate) {
63 64 65
		atomic_set(&ffi->inject_ops, 0);
		ffi->inject_rate = rate;
		ffi->inject_type = (1 << FAULT_MAX) - 1;
S
Sheng Yong 已提交
66
	} else {
67
		memset(ffi, 0, sizeof(struct f2fs_fault_info));
S
Sheng Yong 已提交
68 69
	}
}
70 71
#endif

72 73 74 75 76 77 78
/* 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
79
enum {
80
	Opt_gc_background,
J
Jaegeuk Kim 已提交
81
	Opt_disable_roll_forward,
82
	Opt_norecovery,
J
Jaegeuk Kim 已提交
83
	Opt_discard,
C
Chao Yu 已提交
84
	Opt_nodiscard,
J
Jaegeuk Kim 已提交
85
	Opt_noheap,
86
	Opt_heap,
87
	Opt_user_xattr,
J
Jaegeuk Kim 已提交
88
	Opt_nouser_xattr,
89
	Opt_acl,
J
Jaegeuk Kim 已提交
90 91 92
	Opt_noacl,
	Opt_active_logs,
	Opt_disable_ext_identify,
J
Jaegeuk Kim 已提交
93
	Opt_inline_xattr,
94
	Opt_noinline_xattr,
95
	Opt_inline_data,
96
	Opt_inline_dentry,
97
	Opt_noinline_dentry,
98
	Opt_flush_merge,
J
Jaegeuk Kim 已提交
99
	Opt_noflush_merge,
J
Jaegeuk Kim 已提交
100
	Opt_nobarrier,
101
	Opt_fastboot,
102
	Opt_extent_cache,
103
	Opt_noextent_cache,
W
Wanpeng Li 已提交
104
	Opt_noinline_data,
105
	Opt_data_flush,
106
	Opt_mode,
107
	Opt_io_size_bits,
108
	Opt_fault_injection,
J
Jaegeuk Kim 已提交
109 110
	Opt_lazytime,
	Opt_nolazytime,
J
Jaegeuk Kim 已提交
111 112 113 114
	Opt_err,
};

static match_table_t f2fs_tokens = {
115
	{Opt_gc_background, "background_gc=%s"},
J
Jaegeuk Kim 已提交
116
	{Opt_disable_roll_forward, "disable_roll_forward"},
117
	{Opt_norecovery, "norecovery"},
J
Jaegeuk Kim 已提交
118
	{Opt_discard, "discard"},
C
Chao Yu 已提交
119
	{Opt_nodiscard, "nodiscard"},
J
Jaegeuk Kim 已提交
120
	{Opt_noheap, "no_heap"},
121
	{Opt_heap, "heap"},
122
	{Opt_user_xattr, "user_xattr"},
J
Jaegeuk Kim 已提交
123
	{Opt_nouser_xattr, "nouser_xattr"},
124
	{Opt_acl, "acl"},
J
Jaegeuk Kim 已提交
125 126 127
	{Opt_noacl, "noacl"},
	{Opt_active_logs, "active_logs=%u"},
	{Opt_disable_ext_identify, "disable_ext_identify"},
J
Jaegeuk Kim 已提交
128
	{Opt_inline_xattr, "inline_xattr"},
129
	{Opt_noinline_xattr, "noinline_xattr"},
130
	{Opt_inline_data, "inline_data"},
131
	{Opt_inline_dentry, "inline_dentry"},
132
	{Opt_noinline_dentry, "noinline_dentry"},
133
	{Opt_flush_merge, "flush_merge"},
J
Jaegeuk Kim 已提交
134
	{Opt_noflush_merge, "noflush_merge"},
J
Jaegeuk Kim 已提交
135
	{Opt_nobarrier, "nobarrier"},
136
	{Opt_fastboot, "fastboot"},
137
	{Opt_extent_cache, "extent_cache"},
138
	{Opt_noextent_cache, "noextent_cache"},
W
Wanpeng Li 已提交
139
	{Opt_noinline_data, "noinline_data"},
140
	{Opt_data_flush, "data_flush"},
141
	{Opt_mode, "mode=%s"},
142
	{Opt_io_size_bits, "io_bits=%u"},
143
	{Opt_fault_injection, "fault_injection=%u"},
J
Jaegeuk Kim 已提交
144 145
	{Opt_lazytime, "lazytime"},
	{Opt_nolazytime, "nolazytime"},
J
Jaegeuk Kim 已提交
146 147 148
	{Opt_err, NULL},
};

149
/* Sysfs support for f2fs */
150 151 152
enum {
	GC_THREAD,	/* struct f2fs_gc_thread */
	SM_INFO,	/* struct f2fs_sm_info */
153
	DCC_INFO,	/* struct discard_cmd_control */
154
	NM_INFO,	/* struct f2fs_nm_info */
155
	F2FS_SBI,	/* struct f2fs_sb_info */
S
Sheng Yong 已提交
156 157 158 159
#ifdef CONFIG_F2FS_FAULT_INJECTION
	FAULT_INFO_RATE,	/* struct f2fs_fault_info */
	FAULT_INFO_TYPE,	/* struct f2fs_fault_info */
#endif
160 161
};

162 163 164 165 166
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);
167
	int struct_type;
168 169 170
	int offset;
};

171 172 173 174 175 176
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);
177 178
	else if (struct_type == DCC_INFO)
		return (unsigned char *)SM_I(sbi)->dcc_info;
179 180
	else if (struct_type == NM_INFO)
		return (unsigned char *)NM_I(sbi);
181 182
	else if (struct_type == F2FS_SBI)
		return (unsigned char *)sbi;
S
Sheng Yong 已提交
183 184 185
#ifdef CONFIG_F2FS_FAULT_INJECTION
	else if (struct_type == FAULT_INFO_RATE ||
					struct_type == FAULT_INFO_TYPE)
186
		return (unsigned char *)&sbi->fault_info;
S
Sheng Yong 已提交
187
#endif
188 189 190
	return NULL;
}

191 192 193 194 195 196 197 198 199 200 201 202 203
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)));
}

204 205 206
static ssize_t f2fs_sbi_show(struct f2fs_attr *a,
			struct f2fs_sb_info *sbi, char *buf)
{
207
	unsigned char *ptr = NULL;
208 209
	unsigned int *ui;

210 211
	ptr = __struct_ptr(sbi, a->struct_type);
	if (!ptr)
212 213
		return -EINVAL;

214
	ui = (unsigned int *)(ptr + a->offset);
215 216 217 218 219 220 221 222

	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)
{
223
	unsigned char *ptr;
224 225 226 227
	unsigned long t;
	unsigned int *ui;
	ssize_t ret;

228 229
	ptr = __struct_ptr(sbi, a->struct_type);
	if (!ptr)
230 231
		return -EINVAL;

232
	ui = (unsigned int *)(ptr + a->offset);
233 234 235 236

	ret = kstrtoul(skip_spaces(buf), 0, &t);
	if (ret < 0)
		return ret;
S
Sheng Yong 已提交
237 238 239 240
#ifdef CONFIG_F2FS_FAULT_INJECTION
	if (a->struct_type == FAULT_INFO_TYPE && t >= (1 << FAULT_MAX))
		return -EINVAL;
#endif
241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271
	*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);
}

272
#define F2FS_ATTR_OFFSET(_struct_type, _name, _mode, _show, _store, _offset) \
273 274 275 276
static struct f2fs_attr f2fs_attr_##_name = {			\
	.attr = {.name = __stringify(_name), .mode = _mode },	\
	.show	= _show,					\
	.store	= _store,					\
277 278
	.struct_type = _struct_type,				\
	.offset = _offset					\
279 280
}

281 282 283 284
#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))
285

286 287 288
#define F2FS_GENERAL_RO_ATTR(name) \
static struct f2fs_attr f2fs_attr_##name = __ATTR(name, 0444, name##_show, NULL)

289 290 291 292 293
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);
294
F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, max_small_discards, max_discards);
J
Jaegeuk Kim 已提交
295
F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, batched_trim_sections, trim_sections);
296 297
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);
298
F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_fsync_blocks, min_fsync_blocks);
299
F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_hot_blocks, min_hot_blocks);
300
F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, ram_thresh, ram_thresh);
C
Chao Yu 已提交
301
F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, ra_nid_pages, ra_nid_pages);
C
Chao Yu 已提交
302
F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, dirty_nats_ratio, dirty_nats_ratio);
303
F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, max_victim_search, max_victim_search);
304
F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, dir_level, dir_level);
305
F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, cp_interval, interval_time[CP_TIME]);
306
F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, idle_interval, interval_time[REQ_TIME]);
S
Sheng Yong 已提交
307 308 309 310
#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
311
F2FS_GENERAL_RO_ATTR(lifetime_write_kbytes);
312 313 314 315 316 317

#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),
318
	ATTR_LIST(gc_idle),
319
	ATTR_LIST(reclaim_segments),
320
	ATTR_LIST(max_small_discards),
J
Jaegeuk Kim 已提交
321
	ATTR_LIST(batched_trim_sections),
322 323
	ATTR_LIST(ipu_policy),
	ATTR_LIST(min_ipu_util),
324
	ATTR_LIST(min_fsync_blocks),
325
	ATTR_LIST(min_hot_blocks),
326
	ATTR_LIST(max_victim_search),
327
	ATTR_LIST(dir_level),
328
	ATTR_LIST(ram_thresh),
C
Chao Yu 已提交
329
	ATTR_LIST(ra_nid_pages),
C
Chao Yu 已提交
330
	ATTR_LIST(dirty_nats_ratio),
331
	ATTR_LIST(cp_interval),
332
	ATTR_LIST(idle_interval),
333 334 335 336
#ifdef CONFIG_F2FS_FAULT_INJECTION
	ATTR_LIST(inject_rate),
	ATTR_LIST(inject_type),
#endif
337
	ATTR_LIST(lifetime_write_kbytes),
338 339 340 341 342 343 344 345 346 347 348 349 350 351
	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,
};

352 353 354 355 356 357 358 359 360 361 362 363
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
364 365 366 367 368 369 370
static void init_once(void *foo)
{
	struct f2fs_inode_info *fi = (struct f2fs_inode_info *) foo;

	inode_init_once(&fi->vfs_inode);
}

371 372 373
static int parse_options(struct super_block *sb, char *options)
{
	struct f2fs_sb_info *sbi = F2FS_SB(sb);
374
	struct request_queue *q;
375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398
	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;
399
			if (strlen(name) == 2 && !strncmp(name, "on", 2)) {
400
				set_opt(sbi, BG_GC);
401 402
				clear_opt(sbi, FORCE_FG_GC);
			} else if (strlen(name) == 3 && !strncmp(name, "off", 3)) {
403
				clear_opt(sbi, BG_GC);
404 405 406 407 408
				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 {
409 410 411 412 413 414 415 416
				kfree(name);
				return -EINVAL;
			}
			kfree(name);
			break;
		case Opt_disable_roll_forward:
			set_opt(sbi, DISABLE_ROLL_FORWARD);
			break;
417 418 419 420 421 422
		case Opt_norecovery:
			/* this option mounts f2fs with ro */
			set_opt(sbi, DISABLE_ROLL_FORWARD);
			if (!f2fs_readonly(sb))
				return -EINVAL;
			break;
423
		case Opt_discard:
424 425 426
			q = bdev_get_queue(sb->s_bdev);
			if (blk_queue_discard(q)) {
				set_opt(sbi, DISCARD);
427
			} else if (!f2fs_sb_mounted_blkzoned(sb)) {
428 429 430 431
				f2fs_msg(sb, KERN_WARNING,
					"mounting with \"discard\" option, but "
					"the device does not support discard");
			}
432
			break;
C
Chao Yu 已提交
433
		case Opt_nodiscard:
434 435 436 437 438
			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
439
			clear_opt(sbi, DISCARD);
440
			break;
441 442 443
		case Opt_noheap:
			set_opt(sbi, NOHEAP);
			break;
444 445 446
		case Opt_heap:
			clear_opt(sbi, NOHEAP);
			break;
447
#ifdef CONFIG_F2FS_FS_XATTR
448 449 450
		case Opt_user_xattr:
			set_opt(sbi, XATTR_USER);
			break;
451 452 453
		case Opt_nouser_xattr:
			clear_opt(sbi, XATTR_USER);
			break;
J
Jaegeuk Kim 已提交
454 455 456
		case Opt_inline_xattr:
			set_opt(sbi, INLINE_XATTR);
			break;
457 458 459
		case Opt_noinline_xattr:
			clear_opt(sbi, INLINE_XATTR);
			break;
460
#else
461 462 463 464
		case Opt_user_xattr:
			f2fs_msg(sb, KERN_INFO,
				"user_xattr options not supported");
			break;
465 466 467 468
		case Opt_nouser_xattr:
			f2fs_msg(sb, KERN_INFO,
				"nouser_xattr options not supported");
			break;
J
Jaegeuk Kim 已提交
469 470 471 472
		case Opt_inline_xattr:
			f2fs_msg(sb, KERN_INFO,
				"inline_xattr options not supported");
			break;
473 474 475 476
		case Opt_noinline_xattr:
			f2fs_msg(sb, KERN_INFO,
				"noinline_xattr options not supported");
			break;
477 478
#endif
#ifdef CONFIG_F2FS_FS_POSIX_ACL
479 480 481
		case Opt_acl:
			set_opt(sbi, POSIX_ACL);
			break;
482 483 484 485
		case Opt_noacl:
			clear_opt(sbi, POSIX_ACL);
			break;
#else
486 487 488
		case Opt_acl:
			f2fs_msg(sb, KERN_INFO, "acl options not supported");
			break;
489 490 491 492 493 494 495 496 497 498 499 500 501 502
		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;
503 504 505
		case Opt_inline_data:
			set_opt(sbi, INLINE_DATA);
			break;
506 507 508
		case Opt_inline_dentry:
			set_opt(sbi, INLINE_DENTRY);
			break;
509 510 511
		case Opt_noinline_dentry:
			clear_opt(sbi, INLINE_DENTRY);
			break;
512 513 514
		case Opt_flush_merge:
			set_opt(sbi, FLUSH_MERGE);
			break;
J
Jaegeuk Kim 已提交
515 516 517
		case Opt_noflush_merge:
			clear_opt(sbi, FLUSH_MERGE);
			break;
J
Jaegeuk Kim 已提交
518 519 520
		case Opt_nobarrier:
			set_opt(sbi, NOBARRIER);
			break;
521 522 523
		case Opt_fastboot:
			set_opt(sbi, FASTBOOT);
			break;
524 525 526
		case Opt_extent_cache:
			set_opt(sbi, EXTENT_CACHE);
			break;
527 528 529
		case Opt_noextent_cache:
			clear_opt(sbi, EXTENT_CACHE);
			break;
W
Wanpeng Li 已提交
530 531 532
		case Opt_noinline_data:
			clear_opt(sbi, INLINE_DATA);
			break;
533 534 535
		case Opt_data_flush:
			set_opt(sbi, DATA_FLUSH);
			break;
536 537 538 539 540 541 542
		case Opt_mode:
			name = match_strdup(&args[0]);

			if (!name)
				return -ENOMEM;
			if (strlen(name) == 8 &&
					!strncmp(name, "adaptive", 8)) {
543 544 545 546 547 548 549
				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;
				}
550
				set_opt_mode(sbi, F2FS_MOUNT_ADAPTIVE);
551 552
			} else if (strlen(name) == 3 &&
					!strncmp(name, "lfs", 3)) {
553
				set_opt_mode(sbi, F2FS_MOUNT_LFS);
554 555 556 557 558 559
			} else {
				kfree(name);
				return -EINVAL;
			}
			kfree(name);
			break;
560 561 562 563 564 565 566 567 568 569 570
		case Opt_io_size_bits:
			if (args->from && match_int(args, &arg))
				return -EINVAL;
			if (arg > __ilog2_u32(BIO_MAX_PAGES)) {
				f2fs_msg(sb, KERN_WARNING,
					"Not support %d, larger than %d",
					1 << arg, BIO_MAX_PAGES);
				return -EINVAL;
			}
			sbi->write_io_size_bits = arg;
			break;
571 572 573 574
		case Opt_fault_injection:
			if (args->from && match_int(args, &arg))
				return -EINVAL;
#ifdef CONFIG_F2FS_FAULT_INJECTION
575
			f2fs_build_fault_attr(sbi, arg);
576
			set_opt(sbi, FAULT_INJECTION);
577 578 579 580 581
#else
			f2fs_msg(sb, KERN_INFO,
				"FAULT_INJECTION was not selected");
#endif
			break;
J
Jaegeuk Kim 已提交
582 583 584 585 586 587
		case Opt_lazytime:
			sb->s_flags |= MS_LAZYTIME;
			break;
		case Opt_nolazytime:
			sb->s_flags &= ~MS_LAZYTIME;
			break;
588 589 590 591 592 593 594
		default:
			f2fs_msg(sb, KERN_ERR,
				"Unrecognized mount option \"%s\" or missing value",
				p);
			return -EINVAL;
		}
	}
595 596 597 598 599 600 601

	if (F2FS_IO_SIZE_BITS(sbi) && !test_opt(sbi, LFS)) {
		f2fs_msg(sb, KERN_ERR,
				"Should set mode=lfs with %uKB-sized IO",
				F2FS_IO_SIZE_KB(sbi));
		return -EINVAL;
	}
602 603 604
	return 0;
}

J
Jaegeuk Kim 已提交
605 606 607 608
static struct inode *f2fs_alloc_inode(struct super_block *sb)
{
	struct f2fs_inode_info *fi;

609
	fi = kmem_cache_alloc(f2fs_inode_cachep, GFP_F2FS_ZERO);
J
Jaegeuk Kim 已提交
610 611 612 613 614
	if (!fi)
		return NULL;

	init_once((void *) fi);

M
Masanari Iida 已提交
615
	/* Initialize f2fs-specific inode info */
J
Jaegeuk Kim 已提交
616
	fi->vfs_inode.i_version = 1;
617
	atomic_set(&fi->dirty_pages, 0);
J
Jaegeuk Kim 已提交
618 619
	fi->i_current_depth = 1;
	fi->i_advise = 0;
620
	init_rwsem(&fi->i_sem);
621
	INIT_LIST_HEAD(&fi->dirty_list);
622
	INIT_LIST_HEAD(&fi->gdirty_list);
J
Jaegeuk Kim 已提交
623 624
	INIT_LIST_HEAD(&fi->inmem_pages);
	mutex_init(&fi->inmem_lock);
625 626
	init_rwsem(&fi->dio_rwsem[READ]);
	init_rwsem(&fi->dio_rwsem[WRITE]);
J
Jaegeuk Kim 已提交
627

628 629
	/* Will be used by directory only */
	fi->i_dir_level = F2FS_SB(sb)->dir_level;
J
Jaegeuk Kim 已提交
630 631 632
	return &fi->vfs_inode;
}

633 634
static int f2fs_drop_inode(struct inode *inode)
{
635
	int ret;
636 637 638 639 640 641 642
	/*
	 * 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)
	 */
643
	if ((!inode_unhashed(inode) && inode->i_state & I_SYNC)) {
644
		if (!inode->i_nlink && !is_bad_inode(inode)) {
J
Jaegeuk Kim 已提交
645 646
			/* to avoid evict_inode call simultaneously */
			atomic_inc(&inode->i_count);
647 648 649 650
			spin_unlock(&inode->i_lock);

			/* some remained atomic pages should discarded */
			if (f2fs_is_atomic_file(inode))
651
				drop_inmem_pages(inode);
652

J
Jaegeuk Kim 已提交
653 654 655
			/* should remain fi->extent_tree for writepage */
			f2fs_destroy_extent_node(inode);

656
			sb_start_intwrite(inode->i_sb);
657
			f2fs_i_size_write(inode, 0);
658 659

			if (F2FS_HAS_BLOCKS(inode))
660
				f2fs_truncate(inode);
661 662 663

			sb_end_intwrite(inode->i_sb);

664
			fscrypt_put_encryption_info(inode, NULL);
665
			spin_lock(&inode->i_lock);
J
Jaegeuk Kim 已提交
666
			atomic_dec(&inode->i_count);
667
		}
668
		trace_f2fs_drop_inode(inode, 0);
669
		return 0;
670
	}
671 672 673
	ret = generic_drop_inode(inode);
	trace_f2fs_drop_inode(inode, ret);
	return ret;
674 675
}

676
int f2fs_inode_dirtied(struct inode *inode, bool sync)
677
{
678
	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
679
	int ret = 0;
680 681 682

	spin_lock(&sbi->inode_lock[DIRTY_META]);
	if (is_inode_flag_set(inode, FI_DIRTY_INODE)) {
683 684 685 686
		ret = 1;
	} else {
		set_inode_flag(inode, FI_DIRTY_INODE);
		stat_inc_dirty_inode(sbi, DIRTY_META);
687
	}
688 689
	if (sync && list_empty(&F2FS_I(inode)->gdirty_list)) {
		list_add_tail(&F2FS_I(inode)->gdirty_list,
690
				&sbi->inode_list[DIRTY_META]);
691 692
		inc_page_count(sbi, F2FS_DIRTY_IMETA);
	}
693
	spin_unlock(&sbi->inode_lock[DIRTY_META]);
694
	return ret;
695 696 697 698 699 700 701 702 703 704 705
}

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;
	}
706 707 708 709
	if (!list_empty(&F2FS_I(inode)->gdirty_list)) {
		list_del_init(&F2FS_I(inode)->gdirty_list);
		dec_page_count(sbi, F2FS_DIRTY_IMETA);
	}
710
	clear_inode_flag(inode, FI_DIRTY_INODE);
711
	clear_inode_flag(inode, FI_AUTO_RECOVER);
712
	stat_dec_dirty_inode(F2FS_I_SB(inode), DIRTY_META);
713
	spin_unlock(&sbi->inode_lock[DIRTY_META]);
714 715
}

J
Jaegeuk Kim 已提交
716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734
/*
 * 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);

735
	f2fs_inode_dirtied(inode, false);
J
Jaegeuk Kim 已提交
736 737
}

J
Jaegeuk Kim 已提交
738 739 740 741 742 743
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));
}

744
static void f2fs_destroy_inode(struct inode *inode)
J
Jaegeuk Kim 已提交
745 746 747 748
{
	call_rcu(&inode->i_rcu, f2fs_i_callback);
}

749 750
static void destroy_percpu_info(struct f2fs_sb_info *sbi)
{
751
	percpu_counter_destroy(&sbi->alloc_valid_block_count);
752
	percpu_counter_destroy(&sbi->total_valid_inode_count);
753 754
}

J
Jaegeuk Kim 已提交
755 756 757 758 759 760 761 762 763 764 765 766 767
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
768 769 770 771
static void f2fs_put_super(struct super_block *sb)
{
	struct f2fs_sb_info *sbi = F2FS_SB(sb);

772 773
	if (sbi->s_proc) {
		remove_proc_entry("segment_info", sbi->s_proc);
774
		remove_proc_entry("segment_bits", sbi->s_proc);
775 776
		remove_proc_entry(sb->s_id, f2fs_proc_root);
	}
777
	kobject_del(&sbi->s_kobj);
778

J
Jaegeuk Kim 已提交
779 780
	stop_gc_thread(sbi);

781 782 783
	/* prevent remaining shrinker jobs */
	mutex_lock(&sbi->umount_mutex);

784 785 786 787 788
	/*
	 * 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.
	 */
789
	if (is_sbi_flag_set(sbi, SBI_IS_DIRTY) ||
790
			!is_set_ckpt_flags(sbi, CP_UMOUNT_FLAG)) {
791 792 793 794 795
		struct cp_control cpc = {
			.reason = CP_UMOUNT,
		};
		write_checkpoint(sbi, &cpc);
	}
J
Jaegeuk Kim 已提交
796

797 798 799
	/* be sure to wait for any on-going discard commands */
	f2fs_wait_discard_bio(sbi, NULL_ADDR);

J
Jaegeuk Kim 已提交
800 801 802
	/* write_checkpoint can update stat informaion */
	f2fs_destroy_stats(sbi);

803 804 805 806
	/*
	 * normally superblock is clean, so we need to release this.
	 * In addition, EIO will skip do checkpoint, we need this as well.
	 */
807
	release_ino_entry(sbi, true);
808

809 810 811
	f2fs_leave_shrinker(sbi);
	mutex_unlock(&sbi->umount_mutex);

812
	/* our cp_error case, we can wait for any writeback page */
813
	f2fs_flush_merged_bios(sbi);
814

J
Jaegeuk Kim 已提交
815 816 817 818 819 820 821 822
	iput(sbi->node_inode);
	iput(sbi->meta_inode);

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

	kfree(sbi->ckpt);
823 824
	kobject_put(&sbi->s_kobj);
	wait_for_completion(&sbi->s_kobj_unregister);
J
Jaegeuk Kim 已提交
825 826

	sb->s_fs_info = NULL;
K
Keith Mok 已提交
827 828
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
Y
Yunlei He 已提交
829
	kfree(sbi->raw_super);
830

J
Jaegeuk Kim 已提交
831
	destroy_device_list(sbi);
832
	mempool_destroy(sbi->write_io_dummy);
833
	destroy_percpu_info(sbi);
J
Jaegeuk Kim 已提交
834 835 836 837 838 839
	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
840
	int err = 0;
J
Jaegeuk Kim 已提交
841

842 843
	trace_f2fs_sync_fs(sb, sync);

844
	if (sync) {
845 846
		struct cp_control cpc;

847 848
		cpc.reason = __get_cp_reason(sbi);

849
		mutex_lock(&sbi->gc_mutex);
C
Chao Yu 已提交
850
		err = write_checkpoint(sbi, &cpc);
851 852
		mutex_unlock(&sbi->gc_mutex);
	}
853
	f2fs_trace_ios(NULL, 1);
J
Jaegeuk Kim 已提交
854

C
Chao Yu 已提交
855
	return err;
J
Jaegeuk Kim 已提交
856 857
}

858 859
static int f2fs_freeze(struct super_block *sb)
{
J
Jaegeuk Kim 已提交
860
	if (f2fs_readonly(sb))
861 862
		return 0;

863 864 865 866 867 868 869 870
	/* 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;
871 872 873 874 875 876 877
}

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

J
Jaegeuk Kim 已提交
878 879 880 881 882 883 884 885 886 887 888 889 890 891 892
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;
893
	buf->f_bfree = user_block_count - valid_user_blocks(sbi) + ovp_count;
J
Jaegeuk Kim 已提交
894 895
	buf->f_bavail = user_block_count - valid_user_blocks(sbi);

896
	buf->f_files = sbi->total_node_count - F2FS_RESERVED_NODE_NUM;
897 898
	buf->f_ffree = min(buf->f_files - valid_node_count(sbi),
							buf->f_bavail);
J
Jaegeuk Kim 已提交
899

900
	buf->f_namelen = F2FS_NAME_LEN;
J
Jaegeuk Kim 已提交
901 902 903 904 905 906 907 908 909 910
	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);

911 912 913 914 915 916
	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 {
917
		seq_printf(seq, ",background_gc=%s", "off");
918
	}
J
Jaegeuk Kim 已提交
919 920 921 922 923
	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))
924 925 926
		seq_puts(seq, ",no_heap");
	else
		seq_puts(seq, ",heap");
J
Jaegeuk Kim 已提交
927 928 929 930 931
#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
932 933
	if (test_opt(sbi, INLINE_XATTR))
		seq_puts(seq, ",inline_xattr");
934 935
	else
		seq_puts(seq, ",noinline_xattr");
J
Jaegeuk Kim 已提交
936 937 938 939 940 941 942 943
#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))
944
		seq_puts(seq, ",disable_ext_identify");
945 946
	if (test_opt(sbi, INLINE_DATA))
		seq_puts(seq, ",inline_data");
W
Wanpeng Li 已提交
947 948
	else
		seq_puts(seq, ",noinline_data");
949 950
	if (test_opt(sbi, INLINE_DENTRY))
		seq_puts(seq, ",inline_dentry");
951 952
	else
		seq_puts(seq, ",noinline_dentry");
953
	if (!f2fs_readonly(sbi->sb) && test_opt(sbi, FLUSH_MERGE))
954
		seq_puts(seq, ",flush_merge");
J
Jaegeuk Kim 已提交
955 956
	if (test_opt(sbi, NOBARRIER))
		seq_puts(seq, ",nobarrier");
957 958
	if (test_opt(sbi, FASTBOOT))
		seq_puts(seq, ",fastboot");
959 960
	if (test_opt(sbi, EXTENT_CACHE))
		seq_puts(seq, ",extent_cache");
961 962
	else
		seq_puts(seq, ",noextent_cache");
963 964
	if (test_opt(sbi, DATA_FLUSH))
		seq_puts(seq, ",data_flush");
965 966 967 968 969 970

	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
971
	seq_printf(seq, ",active_logs=%u", sbi->active_logs);
972 973
	if (F2FS_IO_SIZE_BITS(sbi))
		seq_printf(seq, ",io_size=%uKB", F2FS_IO_SIZE_KB(sbi));
974 975 976 977
#ifdef CONFIG_F2FS_FAULT_INJECTION
	if (test_opt(sbi, FAULT_INJECTION))
		seq_puts(seq, ",fault_injection");
#endif
J
Jaegeuk Kim 已提交
978 979 980 981

	return 0;
}

982 983 984 985
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
986 987
	unsigned int total_segs =
			le32_to_cpu(sbi->raw_super->segment_count_main);
988 989
	int i;

990 991 992
	seq_puts(seq, "format: segment_type|valid_blocks\n"
		"segment_type(0:HD, 1:WD, 2:CD, 3:HN, 4:WN, 5:CN)\n");

993
	for (i = 0; i < total_segs; i++) {
994 995 996
		struct seg_entry *se = get_seg_entry(sbi, i);

		if ((i % 10) == 0)
Y
Yunlei He 已提交
997
			seq_printf(seq, "%-10d", i);
998 999
		seq_printf(seq, "%d|%-3u", se->type,
					get_valid_blocks(sbi, i, 1));
1000 1001
		if ((i % 10) == 9 || i == (total_segs - 1))
			seq_putc(seq, '\n');
1002
		else
1003
			seq_putc(seq, ' ');
1004
	}
1005

1006 1007 1008
	return 0;
}

1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026
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++)
1027
			seq_printf(seq, " %.2x", se->cur_valid_map[j]);
1028 1029 1030 1031 1032
		seq_putc(seq, '\n');
	}
	return 0;
}

1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043
#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,					\
1044 1045
};

1046
F2FS_PROC_FILE_DEF(segment_info);
1047
F2FS_PROC_FILE_DEF(segment_bits);
1048

1049 1050 1051 1052 1053 1054
static void default_options(struct f2fs_sb_info *sbi)
{
	/* init some FS parameters */
	sbi->active_logs = NR_CURSEG_TYPE;

	set_opt(sbi, BG_GC);
C
Chao Yu 已提交
1055
	set_opt(sbi, INLINE_XATTR);
1056
	set_opt(sbi, INLINE_DATA);
1057
	set_opt(sbi, INLINE_DENTRY);
J
Jaegeuk Kim 已提交
1058
	set_opt(sbi, EXTENT_CACHE);
1059
	set_opt(sbi, NOHEAP);
J
Jaegeuk Kim 已提交
1060
	sbi->sb->s_flags |= MS_LAZYTIME;
J
Jaegeuk Kim 已提交
1061
	set_opt(sbi, FLUSH_MERGE);
1062
	if (f2fs_sb_mounted_blkzoned(sbi->sb)) {
1063 1064 1065 1066 1067
		set_opt_mode(sbi, F2FS_MOUNT_LFS);
		set_opt(sbi, DISCARD);
	} else {
		set_opt_mode(sbi, F2FS_MOUNT_ADAPTIVE);
	}
1068 1069 1070 1071 1072 1073 1074

#ifdef CONFIG_F2FS_FS_XATTR
	set_opt(sbi, XATTR_USER);
#endif
#ifdef CONFIG_F2FS_FS_POSIX_ACL
	set_opt(sbi, POSIX_ACL);
#endif
1075 1076 1077 1078

#ifdef CONFIG_F2FS_FAULT_INJECTION
	f2fs_build_fault_attr(sbi, 0);
#endif
1079 1080
}

1081 1082 1083 1084 1085
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;
1086 1087
	bool need_restart_gc = false;
	bool need_stop_gc = false;
1088
	bool no_extent_cache = !test_opt(sbi, EXTENT_CACHE);
1089 1090 1091
#ifdef CONFIG_F2FS_FAULT_INJECTION
	struct f2fs_fault_info ffi = sbi->fault_info;
#endif
1092 1093 1094 1095 1096 1097 1098 1099

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

1100 1101 1102 1103 1104 1105 1106 1107 1108
	/* 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);
	}

1109
	sbi->mount_opt.opt = 0;
1110
	default_options(sbi);
1111

1112 1113 1114 1115 1116 1117 1118
	/* parse mount options */
	err = parse_options(sb, data);
	if (err)
		goto restore_opts;

	/*
	 * Previous and new state of filesystem is RO,
1119
	 * so skip checking GC and FLUSH_MERGE conditions.
1120
	 */
1121
	if (f2fs_readonly(sb) && (*flags & MS_RDONLY))
1122 1123
		goto skip;

1124 1125 1126 1127 1128 1129 1130 1131
	/* 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;
	}

1132 1133 1134 1135 1136 1137 1138 1139
	/*
	 * 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);
1140
			need_restart_gc = true;
1141
		}
1142
	} else if (!sbi->gc_thread) {
1143 1144 1145
		err = start_gc_thread(sbi);
		if (err)
			goto restore_opts;
1146 1147 1148
		need_stop_gc = true;
	}

1149 1150 1151 1152 1153 1154 1155 1156 1157 1158
	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);
	}

1159 1160 1161 1162 1163
	/*
	 * 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)) {
1164 1165 1166
		clear_opt(sbi, FLUSH_MERGE);
		destroy_flush_cmd_control(sbi, false);
	} else {
1167 1168
		err = create_flush_cmd_control(sbi);
		if (err)
1169
			goto restore_gc;
1170 1171 1172
	}
skip:
	/* Update the POSIXACL Flag */
1173
	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
1174
		(test_opt(sbi, POSIX_ACL) ? MS_POSIXACL : 0);
1175

1176
	return 0;
1177 1178 1179 1180
restore_gc:
	if (need_restart_gc) {
		if (start_gc_thread(sbi))
			f2fs_msg(sbi->sb, KERN_WARNING,
A
arter97 已提交
1181
				"background gc thread has stopped");
1182 1183 1184
	} else if (need_stop_gc) {
		stop_gc_thread(sbi);
	}
1185 1186 1187
restore_opts:
	sbi->mount_opt = org_mount_opt;
	sbi->active_logs = active_logs;
1188 1189 1190
#ifdef CONFIG_F2FS_FAULT_INJECTION
	sbi->fault_info = ffi;
#endif
1191 1192 1193
	return err;
}

J
Jaegeuk Kim 已提交
1194 1195
static struct super_operations f2fs_sops = {
	.alloc_inode	= f2fs_alloc_inode,
1196
	.drop_inode	= f2fs_drop_inode,
J
Jaegeuk Kim 已提交
1197 1198
	.destroy_inode	= f2fs_destroy_inode,
	.write_inode	= f2fs_write_inode,
1199
	.dirty_inode	= f2fs_dirty_inode,
J
Jaegeuk Kim 已提交
1200 1201 1202 1203
	.show_options	= f2fs_show_options,
	.evict_inode	= f2fs_evict_inode,
	.put_super	= f2fs_put_super,
	.sync_fs	= f2fs_sync_fs,
1204 1205
	.freeze_fs	= f2fs_freeze,
	.unfreeze_fs	= f2fs_unfreeze,
J
Jaegeuk Kim 已提交
1206
	.statfs		= f2fs_statfs,
1207
	.remount_fs	= f2fs_remount,
J
Jaegeuk Kim 已提交
1208 1209
};

1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231
#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;
}

1232
static const struct fscrypt_operations f2fs_cryptops = {
1233
	.key_prefix	= "f2fs:",
1234 1235 1236 1237 1238 1239 1240
	.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
1241
static const struct fscrypt_operations f2fs_cryptops = {
1242 1243 1244 1245
	.is_encrypted	= f2fs_encrypted_inode,
};
#endif

J
Jaegeuk Kim 已提交
1246 1247 1248 1249 1250 1251
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;

1252
	if (check_nid_range(sbi, ino))
1253
		return ERR_PTR(-ESTALE);
J
Jaegeuk Kim 已提交
1254 1255 1256 1257 1258 1259 1260 1261 1262

	/*
	 * 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);
1263
	if (unlikely(generation && inode->i_generation != generation)) {
J
Jaegeuk Kim 已提交
1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290
		/* 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
1291
static loff_t max_file_blocks(void)
J
Jaegeuk Kim 已提交
1292
{
1293
	loff_t result = (DEF_ADDRS_PER_INODE - F2FS_INLINE_XATTR_ADDRS);
J
Jaegeuk Kim 已提交
1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309
	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;
}

1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320
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 */
1321
	return __sync_dirty_buffer(bh, REQ_PREFLUSH | REQ_FUA);
1322 1323
}

1324
static inline bool sanity_check_area_boundary(struct f2fs_sb_info *sbi,
1325
					struct buffer_head *bh)
1326
{
1327 1328
	struct f2fs_super_block *raw_super = (struct f2fs_super_block *)
					(bh->b_data + F2FS_SUPER_OFFSET);
1329
	struct super_block *sb = sbi->sb;
1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342
	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);
1343 1344 1345 1346
	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);
1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390

	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;
	}

1391
	if (main_end_blkaddr > seg_end_blkaddr) {
1392
		f2fs_msg(sb, KERN_INFO,
1393
			"Wrong MAIN_AREA boundary, start(%u) end(%u) block(%u)",
1394
			main_blkaddr,
1395 1396
			segment0_blkaddr +
				(segment_count << log_blocks_per_seg),
1397 1398
			segment_count_main << log_blocks_per_seg);
		return true;
1399 1400 1401 1402 1403 1404 1405 1406 1407
	} 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)) {
1408
			set_sbi_flag(sbi, SBI_NEED_SB_WRITE);
1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421
			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;
1422 1423 1424 1425
	}
	return false;
}

1426
static int sanity_check_raw_super(struct f2fs_sb_info *sbi,
1427
				struct buffer_head *bh)
J
Jaegeuk Kim 已提交
1428
{
1429 1430
	struct f2fs_super_block *raw_super = (struct f2fs_super_block *)
					(bh->b_data + F2FS_SUPER_OFFSET);
1431
	struct super_block *sb = sbi->sb;
J
Jaegeuk Kim 已提交
1432 1433
	unsigned int blocksize;

1434 1435 1436 1437
	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
1438
		return 1;
1439
	}
J
Jaegeuk Kim 已提交
1440

1441
	/* Currently, support only 4KB page cache size */
1442
	if (F2FS_BLKSIZE != PAGE_SIZE) {
1443
		f2fs_msg(sb, KERN_INFO,
1444
			"Invalid page_cache_size (%lu), supports only 4KB\n",
1445
			PAGE_SIZE);
1446 1447 1448
		return 1;
	}

J
Jaegeuk Kim 已提交
1449 1450
	/* Currently, support only 4KB block size */
	blocksize = 1 << le32_to_cpu(raw_super->log_blocksize);
1451
	if (blocksize != F2FS_BLKSIZE) {
1452 1453 1454
		f2fs_msg(sb, KERN_INFO,
			"Invalid blocksize (%u), supports only 4KB\n",
			blocksize);
J
Jaegeuk Kim 已提交
1455
		return 1;
1456
	}
1457

1458 1459 1460 1461 1462 1463 1464 1465
	/* 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
1466 1467 1468 1469 1470 1471 1472
	/* 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
1473
		return 1;
1474
	}
C
Chao Yu 已提交
1475 1476 1477 1478 1479 1480 1481
	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
1482
		return 1;
1483
	}
1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497

	/* 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 */
1498
	if (sanity_check_area_boundary(sbi, bh))
1499 1500
		return 1;

J
Jaegeuk Kim 已提交
1501 1502 1503
	return 0;
}

1504
int sanity_check_ckpt(struct f2fs_sb_info *sbi)
J
Jaegeuk Kim 已提交
1505 1506
{
	unsigned int total, fsmeta;
1507 1508
	struct f2fs_super_block *raw_super = F2FS_RAW_SUPER(sbi);
	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
J
Jaegeuk Kim 已提交
1509
	unsigned int ovp_segments, reserved_segments;
J
Jaegeuk Kim 已提交
1510 1511 1512 1513 1514 1515 1516 1517

	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);

1518
	if (unlikely(fsmeta >= total))
J
Jaegeuk Kim 已提交
1519
		return 1;
1520

J
Jaegeuk Kim 已提交
1521 1522 1523 1524 1525 1526 1527 1528 1529 1530
	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;
	}

1531
	if (unlikely(f2fs_cp_error(sbi))) {
1532 1533 1534
		f2fs_msg(sbi->sb, KERN_ERR, "A bug case: need to run fsck");
		return 1;
	}
J
Jaegeuk Kim 已提交
1535 1536 1537 1538 1539 1540
	return 0;
}

static void init_sb_info(struct f2fs_sb_info *sbi)
{
	struct f2fs_super_block *raw_super = sbi->raw_super;
1541
	int i;
J
Jaegeuk Kim 已提交
1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557

	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);
1558
	sbi->cur_victim_sec = NULL_SECNO;
1559
	sbi->max_victim_search = DEF_MAX_VICTIM_SEARCH;
J
Jaegeuk Kim 已提交
1560

1561
	sbi->dir_level = DEF_DIR_LEVEL;
1562
	sbi->interval_time[CP_TIME] = DEF_CP_INTERVAL;
1563
	sbi->interval_time[REQ_TIME] = DEF_IDLE_INTERVAL;
1564
	clear_sbi_flag(sbi, SBI_NEED_FSCK);
1565

1566 1567 1568
	for (i = 0; i < NR_COUNT_TYPE; i++)
		atomic_set(&sbi->nr_pages[i], 0);

1569 1570
	atomic_set(&sbi->wb_sync_req, 0);

1571 1572
	INIT_LIST_HEAD(&sbi->s_list);
	mutex_init(&sbi->umount_mutex);
1573 1574
	mutex_init(&sbi->wio_mutex[NODE]);
	mutex_init(&sbi->wio_mutex[DATA]);
1575
	spin_lock_init(&sbi->cp_lock);
J
Jaegeuk Kim 已提交
1576 1577
}

1578 1579
static int init_percpu_info(struct f2fs_sb_info *sbi)
{
1580
	int err;
1581

1582 1583 1584 1585 1586
	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,
1587
								GFP_KERNEL);
1588 1589
}

1590
#ifdef CONFIG_BLK_DEV_ZONED
J
Jaegeuk Kim 已提交
1591
static int init_blkz_info(struct f2fs_sb_info *sbi, int devi)
1592
{
J
Jaegeuk Kim 已提交
1593
	struct block_device *bdev = FDEV(devi).bdev;
1594 1595 1596 1597 1598 1599 1600 1601 1602 1603
	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
1604
	if (sbi->blocks_per_blkz && sbi->blocks_per_blkz !=
1605
				SECTOR_TO_BLOCK(bdev_zone_sectors(bdev)))
J
Jaegeuk Kim 已提交
1606
		return -EINVAL;
1607
	sbi->blocks_per_blkz = SECTOR_TO_BLOCK(bdev_zone_sectors(bdev));
J
Jaegeuk Kim 已提交
1608 1609 1610
	if (sbi->log_blocks_per_blkz && sbi->log_blocks_per_blkz !=
				__ilog2_u32(sbi->blocks_per_blkz))
		return -EINVAL;
1611
	sbi->log_blocks_per_blkz = __ilog2_u32(sbi->blocks_per_blkz);
J
Jaegeuk Kim 已提交
1612 1613
	FDEV(devi).nr_blkz = SECTOR_TO_BLOCK(nr_sectors) >>
					sbi->log_blocks_per_blkz;
1614
	if (nr_sectors & (bdev_zone_sectors(bdev) - 1))
J
Jaegeuk Kim 已提交
1615
		FDEV(devi).nr_blkz++;
1616

J
Jaegeuk Kim 已提交
1617 1618
	FDEV(devi).blkz_type = kmalloc(FDEV(devi).nr_blkz, GFP_KERNEL);
	if (!FDEV(devi).blkz_type)
1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642
		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
1643
			FDEV(devi).blkz_type[n] = zones[i].type;
1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654
			sector += zones[i].len;
			n++;
		}
	}

	kfree(zones);

	return err;
}
#endif

1655 1656
/*
 * Read f2fs raw super block.
1657 1658 1659
 * 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.
1660
 */
1661
static int read_raw_super_block(struct f2fs_sb_info *sbi,
1662
			struct f2fs_super_block **raw_super,
1663
			int *valid_super_block, int *recovery)
1664
{
1665
	struct super_block *sb = sbi->sb;
1666
	int block;
1667
	struct buffer_head *bh;
1668
	struct f2fs_super_block *super;
1669
	int err = 0;
1670

Y
Yunlei He 已提交
1671 1672 1673
	super = kzalloc(sizeof(struct f2fs_super_block), GFP_KERNEL);
	if (!super)
		return -ENOMEM;
1674 1675 1676 1677 1678

	for (block = 0; block < 2; block++) {
		bh = sb_bread(sb, block);
		if (!bh) {
			f2fs_msg(sb, KERN_ERR, "Unable to read %dth superblock",
1679
				block + 1);
1680 1681 1682
			err = -EIO;
			continue;
		}
1683

1684
		/* sanity checking of raw super */
1685
		if (sanity_check_raw_super(sbi, bh)) {
1686 1687 1688 1689 1690 1691 1692
			f2fs_msg(sb, KERN_ERR,
				"Can't find valid F2FS filesystem in %dth superblock",
				block + 1);
			err = -EINVAL;
			brelse(bh);
			continue;
		}
1693

1694
		if (!*raw_super) {
1695 1696
			memcpy(super, bh->b_data + F2FS_SUPER_OFFSET,
							sizeof(*super));
1697 1698 1699 1700
			*valid_super_block = block;
			*raw_super = super;
		}
		brelse(bh);
1701 1702
	}

1703 1704 1705
	/* Fail to read any one of the superblocks*/
	if (err < 0)
		*recovery = 1;
1706 1707

	/* No valid superblock */
1708
	if (!*raw_super)
Y
Yunlei He 已提交
1709
		kfree(super);
1710 1711
	else
		err = 0;
1712

1713
	return err;
1714 1715
}

1716
int f2fs_commit_super(struct f2fs_sb_info *sbi, bool recover)
J
Jaegeuk Kim 已提交
1717
{
J
Jaegeuk Kim 已提交
1718
	struct buffer_head *bh;
J
Jaegeuk Kim 已提交
1719 1720
	int err;

1721 1722 1723
	if ((recover && f2fs_readonly(sbi->sb)) ||
				bdev_read_only(sbi->sb->s_bdev)) {
		set_sbi_flag(sbi, SBI_NEED_SB_WRITE);
1724
		return -EROFS;
1725
	}
1726

1727 1728
	/* write back-up superblock first */
	bh = sb_getblk(sbi->sb, sbi->valid_super_block ? 0: 1);
J
Jaegeuk Kim 已提交
1729 1730
	if (!bh)
		return -EIO;
1731
	err = __f2fs_commit_super(bh, F2FS_RAW_SUPER(sbi));
J
Jaegeuk Kim 已提交
1732
	brelse(bh);
C
Chao Yu 已提交
1733 1734 1735

	/* if we are in recovery path, skip writing valid superblock */
	if (recover || err)
J
Jaegeuk Kim 已提交
1736
		return err;
J
Jaegeuk Kim 已提交
1737 1738

	/* write current valid superblock */
1739 1740 1741 1742 1743 1744
	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
1745 1746
}

J
Jaegeuk Kim 已提交
1747 1748 1749
static int f2fs_scan_devices(struct f2fs_sb_info *sbi)
{
	struct f2fs_super_block *raw_super = F2FS_RAW_SUPER(sbi);
1750
	unsigned int max_devices = MAX_DEVICES;
J
Jaegeuk Kim 已提交
1751 1752
	int i;

1753 1754 1755
	/* Initialize single device information */
	if (!RDEV(0).path[0]) {
		if (!bdev_is_zoned(sbi->sb->s_bdev))
J
Jaegeuk Kim 已提交
1756
			return 0;
1757 1758
		max_devices = 1;
	}
J
Jaegeuk Kim 已提交
1759

1760 1761 1762 1763 1764 1765 1766 1767
	/*
	 * Initialize multiple devices information, or single
	 * zoned block device information.
	 */
	sbi->devs = kcalloc(max_devices, sizeof(struct f2fs_dev_info),
				GFP_KERNEL);
	if (!sbi->devs)
		return -ENOMEM;
J
Jaegeuk Kim 已提交
1768

1769
	for (i = 0; i < max_devices; i++) {
J
Jaegeuk Kim 已提交
1770

1771 1772 1773 1774 1775 1776 1777
		if (i > 0 && !RDEV(i).path[0])
			break;

		if (max_devices == 1) {
			/* Single zoned block device mount */
			FDEV(0).bdev =
				blkdev_get_by_dev(sbi->sb->s_bdev->bd_dev,
J
Jaegeuk Kim 已提交
1778
					sbi->sb->s_mode, sbi->sb->s_type);
1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796
		} else {
			/* Multi-device mount */
			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,
J
Jaegeuk Kim 已提交
1797
					sbi->sb->s_mode, sbi->sb->s_type);
1798
		}
J
Jaegeuk Kim 已提交
1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817
		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;
			}
1818 1819
			if (max_devices == 1)
				break;
J
Jaegeuk Kim 已提交
1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835
			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);
	}
1836 1837
	f2fs_msg(sbi->sb, KERN_INFO,
			"IO Block Size: %8d KB", F2FS_IO_SIZE_KB(sbi));
J
Jaegeuk Kim 已提交
1838 1839 1840
	return 0;
}

J
Jaegeuk Kim 已提交
1841 1842 1843
static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
{
	struct f2fs_sb_info *sbi;
1844
	struct f2fs_super_block *raw_super;
J
Jaegeuk Kim 已提交
1845
	struct inode *root;
1846
	int err;
1847
	bool retry = true, need_fsck = false;
1848
	char *options = NULL;
1849
	int recovery, i, valid_super_block;
1850
	struct curseg_info *seg_i;
J
Jaegeuk Kim 已提交
1851

1852
try_onemore:
1853 1854
	err = -EINVAL;
	raw_super = NULL;
1855
	valid_super_block = -1;
1856 1857
	recovery = 0;

J
Jaegeuk Kim 已提交
1858 1859 1860 1861 1862
	/* allocate memory for f2fs-specific super block info */
	sbi = kzalloc(sizeof(struct f2fs_sb_info), GFP_KERNEL);
	if (!sbi)
		return -ENOMEM;

1863 1864
	sbi->sb = sb;

K
Keith Mok 已提交
1865 1866 1867 1868 1869 1870 1871 1872 1873
	/* 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;
	}

1874
	/* set a block size */
1875
	if (unlikely(!sb_set_blocksize(sb, F2FS_BLKSIZE))) {
1876
		f2fs_msg(sb, KERN_ERR, "unable to set blocksize");
J
Jaegeuk Kim 已提交
1877
		goto free_sbi;
1878
	}
J
Jaegeuk Kim 已提交
1879

1880
	err = read_raw_super_block(sbi, &raw_super, &valid_super_block,
1881
								&recovery);
1882 1883 1884
	if (err)
		goto free_sbi;

1885
	sb->s_fs_info = sbi;
1886 1887
	sbi->raw_super = raw_super;

1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899
	/*
	 * 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
1900
	default_options(sbi);
J
Jaegeuk Kim 已提交
1901
	/* parse mount options */
1902 1903 1904
	options = kstrdup((const char *)data, GFP_KERNEL);
	if (data && !options) {
		err = -ENOMEM;
J
Jaegeuk Kim 已提交
1905
		goto free_sb_buf;
1906 1907 1908 1909 1910
	}

	err = parse_options(sb, options);
	if (err)
		goto free_options;
J
Jaegeuk Kim 已提交
1911

C
Chao Yu 已提交
1912 1913 1914
	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
1915 1916 1917 1918
	sb->s_max_links = F2FS_LINK_MAX;
	get_random_bytes(&sbi->s_next_generation, sizeof(u32));

	sb->s_op = &f2fs_sops;
1919
	sb->s_cop = &f2fs_cryptops;
J
Jaegeuk Kim 已提交
1920 1921 1922 1923 1924 1925 1926 1927 1928
	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 */
1929
	sbi->valid_super_block = valid_super_block;
J
Jaegeuk Kim 已提交
1930 1931
	mutex_init(&sbi->gc_mutex);
	mutex_init(&sbi->cp_mutex);
1932
	init_rwsem(&sbi->node_write);
1933
	init_rwsem(&sbi->node_change);
1934 1935 1936

	/* disallow all the data/node/meta page writes */
	set_sbi_flag(sbi, SBI_POR_DOING);
J
Jaegeuk Kim 已提交
1937
	spin_lock_init(&sbi->stat_lock);
1938

1939
	init_rwsem(&sbi->read_io.io_rwsem);
J
Jaegeuk Kim 已提交
1940 1941 1942
	sbi->read_io.sbi = sbi;
	sbi->read_io.bio = NULL;
	for (i = 0; i < NR_PAGE_TYPE; i++) {
1943
		init_rwsem(&sbi->write_io[i].io_rwsem);
J
Jaegeuk Kim 已提交
1944 1945 1946
		sbi->write_io[i].sbi = sbi;
		sbi->write_io[i].bio = NULL;
	}
1947

1948
	init_rwsem(&sbi->cp_rwsem);
1949
	init_waitqueue_head(&sbi->cp_wait);
J
Jaegeuk Kim 已提交
1950 1951
	init_sb_info(sbi);

1952 1953 1954 1955
	err = init_percpu_info(sbi);
	if (err)
		goto free_options;

1956 1957
	if (F2FS_IO_SIZE(sbi) > 1) {
		sbi->write_io_dummy =
1958
			mempool_create_page_pool(2 * (F2FS_IO_SIZE(sbi) - 1), 0);
1959 1960 1961 1962
		if (!sbi->write_io_dummy)
			goto free_options;
	}

J
Jaegeuk Kim 已提交
1963 1964 1965
	/* get an inode for meta space */
	sbi->meta_inode = f2fs_iget(sb, F2FS_META_INO(sbi));
	if (IS_ERR(sbi->meta_inode)) {
1966
		f2fs_msg(sb, KERN_ERR, "Failed to read F2FS meta data inode");
J
Jaegeuk Kim 已提交
1967
		err = PTR_ERR(sbi->meta_inode);
1968
		goto free_io_dummy;
J
Jaegeuk Kim 已提交
1969 1970 1971
	}

	err = get_valid_checkpoint(sbi);
1972 1973
	if (err) {
		f2fs_msg(sb, KERN_ERR, "Failed to get valid F2FS checkpoint");
J
Jaegeuk Kim 已提交
1974
		goto free_meta_inode;
1975
	}
J
Jaegeuk Kim 已提交
1976

J
Jaegeuk Kim 已提交
1977 1978 1979 1980 1981 1982 1983
	/* 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
1984 1985
	sbi->total_valid_node_count =
				le32_to_cpu(sbi->ckpt->valid_node_count);
1986 1987
	percpu_counter_set(&sbi->total_valid_inode_count,
				le32_to_cpu(sbi->ckpt->valid_inode_count));
J
Jaegeuk Kim 已提交
1988 1989 1990 1991
	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;
1992

1993 1994 1995 1996
	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
1997

C
Chao Yu 已提交
1998 1999
	init_extent_cache_info(sbi);

J
Jaegeuk Kim 已提交
2000
	init_ino_entry_info(sbi);
J
Jaegeuk Kim 已提交
2001 2002 2003

	/* setup f2fs internal modules */
	err = build_segment_manager(sbi);
2004 2005 2006
	if (err) {
		f2fs_msg(sb, KERN_ERR,
			"Failed to initialize F2FS segment manager");
J
Jaegeuk Kim 已提交
2007
		goto free_sm;
2008
	}
J
Jaegeuk Kim 已提交
2009
	err = build_node_manager(sbi);
2010 2011 2012
	if (err) {
		f2fs_msg(sb, KERN_ERR,
			"Failed to initialize F2FS node manager");
J
Jaegeuk Kim 已提交
2013
		goto free_nm;
2014
	}
J
Jaegeuk Kim 已提交
2015

2016 2017 2018 2019 2020 2021 2022 2023 2024
	/* 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 =
2025
			le64_to_cpu(seg_i->journal->info.kbytes_written);
2026

J
Jaegeuk Kim 已提交
2027 2028 2029 2030 2031
	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)) {
2032
		f2fs_msg(sb, KERN_ERR, "Failed to read node inode");
J
Jaegeuk Kim 已提交
2033 2034 2035 2036
		err = PTR_ERR(sbi->node_inode);
		goto free_nm;
	}

2037 2038
	f2fs_join_shrinker(sbi);

2039 2040 2041 2042
	err = f2fs_build_stats(sbi);
	if (err)
		goto free_nm;

J
Jaegeuk Kim 已提交
2043
	/* if there are nt orphan nodes free them */
2044 2045 2046
	err = recover_orphan_inodes(sbi);
	if (err)
		goto free_node_inode;
J
Jaegeuk Kim 已提交
2047 2048 2049 2050

	/* read root inode and dentry */
	root = f2fs_iget(sb, F2FS_ROOT_INO(sbi));
	if (IS_ERR(root)) {
2051
		f2fs_msg(sb, KERN_ERR, "Failed to read root inode");
J
Jaegeuk Kim 已提交
2052 2053 2054
		err = PTR_ERR(root);
		goto free_node_inode;
	}
2055
	if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) {
2056
		iput(root);
2057
		err = -EINVAL;
2058
		goto free_node_inode;
2059
	}
J
Jaegeuk Kim 已提交
2060 2061 2062 2063 2064 2065 2066

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

2067 2068 2069
	if (f2fs_proc_root)
		sbi->s_proc = proc_mkdir(sb->s_id, f2fs_proc_root);

2070
	if (sbi->s_proc) {
2071 2072
		proc_create_data("segment_info", S_IRUGO, sbi->s_proc,
				 &f2fs_seq_segment_info_fops, sb);
2073 2074 2075
		proc_create_data("segment_bits", S_IRUGO, sbi->s_proc,
				 &f2fs_seq_segment_bits_fops, sb);
	}
2076

2077 2078 2079 2080 2081
	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)
2082
		goto free_proc;
2083

2084 2085
	/* recover fsynced data */
	if (!test_opt(sbi, DISABLE_ROLL_FORWARD)) {
2086 2087 2088 2089 2090
		/*
		 * 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) &&
2091
				!is_set_ckpt_flags(sbi, CP_UMOUNT_FLAG)) {
2092 2093 2094
			err = -EROFS;
			goto free_kobj;
		}
2095 2096 2097 2098

		if (need_fsck)
			set_sbi_flag(sbi, SBI_NEED_FSCK);

2099 2100 2101
		if (!retry)
			goto skip_recovery;

2102 2103
		err = recover_fsync_data(sbi, false);
		if (err < 0) {
2104
			need_fsck = true;
2105
			f2fs_msg(sb, KERN_ERR,
2106
				"Cannot recover all fsync data errno=%d", err);
2107 2108
			goto free_kobj;
		}
2109 2110 2111 2112 2113 2114 2115 2116 2117
	} 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;
		}
2118
	}
2119
skip_recovery:
2120 2121
	/* recover_fsync_data() cleared this already */
	clear_sbi_flag(sbi, SBI_POR_DOING);
2122

2123 2124 2125 2126
	/*
	 * If filesystem is not mounted as read-only then
	 * do start the gc_thread.
	 */
2127
	if (test_opt(sbi, BG_GC) && !f2fs_readonly(sb)) {
2128 2129 2130 2131 2132
		/* After POR, we can run background GC thread.*/
		err = start_gc_thread(sbi);
		if (err)
			goto free_kobj;
	}
2133
	kfree(options);
2134 2135

	/* recover broken superblock */
2136
	if (recovery) {
2137 2138
		err = f2fs_commit_super(sbi, true);
		f2fs_msg(sb, KERN_INFO,
2139
			"Try to recover %dth superblock, ret: %d",
2140
			sbi->valid_super_block ? 1 : 2, err);
2141 2142
	}

2143 2144
	f2fs_msg(sbi->sb, KERN_NOTICE, "Mounted with checkpoint version = %llx",
				cur_cp_version(F2FS_CKPT(sbi)));
2145
	f2fs_update_time(sbi, CP_TIME);
2146
	f2fs_update_time(sbi, REQ_TIME);
J
Jaegeuk Kim 已提交
2147
	return 0;
2148 2149

free_kobj:
2150
	f2fs_sync_inode_meta(sbi);
2151
	kobject_del(&sbi->s_kobj);
2152 2153
	kobject_put(&sbi->s_kobj);
	wait_for_completion(&sbi->s_kobj_unregister);
2154
free_proc:
2155 2156
	if (sbi->s_proc) {
		remove_proc_entry("segment_info", sbi->s_proc);
2157
		remove_proc_entry("segment_bits", sbi->s_proc);
2158 2159
		remove_proc_entry(sb->s_id, f2fs_proc_root);
	}
J
Jaegeuk Kim 已提交
2160 2161 2162 2163
free_root_inode:
	dput(sb->s_root);
	sb->s_root = NULL;
free_node_inode:
2164
	truncate_inode_pages_final(NODE_MAPPING(sbi));
2165
	mutex_lock(&sbi->umount_mutex);
2166
	release_ino_entry(sbi, true);
2167
	f2fs_leave_shrinker(sbi);
2168 2169 2170 2171 2172 2173 2174
	/*
	 * 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
2175
	iput(sbi->node_inode);
2176
	mutex_unlock(&sbi->umount_mutex);
2177
	f2fs_destroy_stats(sbi);
J
Jaegeuk Kim 已提交
2178 2179 2180 2181
free_nm:
	destroy_node_manager(sbi);
free_sm:
	destroy_segment_manager(sbi);
J
Jaegeuk Kim 已提交
2182 2183
free_devices:
	destroy_device_list(sbi);
J
Jaegeuk Kim 已提交
2184 2185 2186 2187
	kfree(sbi->ckpt);
free_meta_inode:
	make_bad_inode(sbi->meta_inode);
	iput(sbi->meta_inode);
2188 2189
free_io_dummy:
	mempool_destroy(sbi->write_io_dummy);
2190
free_options:
2191
	destroy_percpu_info(sbi);
2192
	kfree(options);
J
Jaegeuk Kim 已提交
2193
free_sb_buf:
Y
Yunlei He 已提交
2194
	kfree(raw_super);
J
Jaegeuk Kim 已提交
2195
free_sbi:
K
Keith Mok 已提交
2196 2197
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
J
Jaegeuk Kim 已提交
2198
	kfree(sbi);
2199 2200 2201

	/* give only one another chance */
	if (retry) {
2202
		retry = false;
2203 2204 2205
		shrink_dcache_sb(sb);
		goto try_onemore;
	}
J
Jaegeuk Kim 已提交
2206 2207 2208 2209 2210 2211 2212 2213 2214
	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);
}

2215 2216 2217
static void kill_f2fs_super(struct super_block *sb)
{
	if (sb->s_root)
2218
		set_sbi_flag(F2FS_SB(sb), SBI_IS_CLOSE);
2219 2220 2221
	kill_block_super(sb);
}

J
Jaegeuk Kim 已提交
2222 2223 2224 2225
static struct file_system_type f2fs_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "f2fs",
	.mount		= f2fs_mount,
2226
	.kill_sb	= kill_f2fs_super,
J
Jaegeuk Kim 已提交
2227 2228
	.fs_flags	= FS_REQUIRES_DEV,
};
2229
MODULE_ALIAS_FS("f2fs");
J
Jaegeuk Kim 已提交
2230

2231
static int __init init_inodecache(void)
J
Jaegeuk Kim 已提交
2232
{
2233 2234 2235
	f2fs_inode_cachep = kmem_cache_create("f2fs_inode_cache",
			sizeof(struct f2fs_inode_info), 0,
			SLAB_RECLAIM_ACCOUNT|SLAB_ACCOUNT, NULL);
2236
	if (!f2fs_inode_cachep)
J
Jaegeuk Kim 已提交
2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254
		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;

2255 2256
	f2fs_build_trace_ios();

J
Jaegeuk Kim 已提交
2257 2258 2259 2260 2261
	err = init_inodecache();
	if (err)
		goto fail;
	err = create_node_manager_caches();
	if (err)
2262
		goto free_inodecache;
2263
	err = create_segment_manager_caches();
J
Jaegeuk Kim 已提交
2264
	if (err)
2265
		goto free_node_manager_caches;
J
Jaegeuk Kim 已提交
2266 2267
	err = create_checkpoint_caches();
	if (err)
2268
		goto free_segment_manager_caches;
C
Chao Yu 已提交
2269 2270 2271
	err = create_extent_cache();
	if (err)
		goto free_checkpoint_caches;
2272
	f2fs_kset = kset_create_and_add("f2fs", NULL, fs_kobj);
2273 2274
	if (!f2fs_kset) {
		err = -ENOMEM;
C
Chao Yu 已提交
2275
		goto free_extent_cache;
2276
	}
2277
	err = register_shrinker(&f2fs_shrinker_info);
2278
	if (err)
2279
		goto free_kset;
2280 2281 2282 2283

	err = register_filesystem(&f2fs_fs_type);
	if (err)
		goto free_shrinker;
2284 2285 2286
	err = f2fs_create_root_stats();
	if (err)
		goto free_filesystem;
2287
	f2fs_proc_root = proc_mkdir("fs/f2fs", NULL);
2288 2289
	return 0;

2290 2291
free_filesystem:
	unregister_filesystem(&f2fs_fs_type);
2292 2293
free_shrinker:
	unregister_shrinker(&f2fs_shrinker_info);
2294 2295
free_kset:
	kset_unregister(f2fs_kset);
C
Chao Yu 已提交
2296 2297
free_extent_cache:
	destroy_extent_cache();
2298 2299
free_checkpoint_caches:
	destroy_checkpoint_caches();
2300 2301
free_segment_manager_caches:
	destroy_segment_manager_caches();
2302 2303 2304 2305
free_node_manager_caches:
	destroy_node_manager_caches();
free_inodecache:
	destroy_inodecache();
J
Jaegeuk Kim 已提交
2306 2307 2308 2309 2310 2311
fail:
	return err;
}

static void __exit exit_f2fs_fs(void)
{
2312
	remove_proc_entry("fs/f2fs", NULL);
2313
	f2fs_destroy_root_stats();
J
Jaegeuk Kim 已提交
2314
	unregister_filesystem(&f2fs_fs_type);
T
Tiezhu Yang 已提交
2315 2316
	unregister_shrinker(&f2fs_shrinker_info);
	kset_unregister(f2fs_kset);
W
Wanpeng Li 已提交
2317
	destroy_extent_cache();
J
Jaegeuk Kim 已提交
2318
	destroy_checkpoint_caches();
2319
	destroy_segment_manager_caches();
J
Jaegeuk Kim 已提交
2320 2321
	destroy_node_manager_caches();
	destroy_inodecache();
2322
	f2fs_destroy_trace_ios();
J
Jaegeuk Kim 已提交
2323 2324 2325 2326 2327 2328 2329 2330
}

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");
2331