super.c 53.2 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
				set_opt_mode(sbi, F2FS_MOUNT_ADAPTIVE);
522 523
			} else if (strlen(name) == 3 &&
					!strncmp(name, "lfs", 3)) {
524
				set_opt_mode(sbi, F2FS_MOUNT_LFS);
525 526 527 528 529 530
			} else {
				kfree(name);
				return -EINVAL;
			}
			kfree(name);
			break;
531 532 533 534
		case Opt_fault_injection:
			if (args->from && match_int(args, &arg))
				return -EINVAL;
#ifdef CONFIG_F2FS_FAULT_INJECTION
535
			f2fs_build_fault_attr(sbi, arg);
536 537 538 539 540
#else
			f2fs_msg(sb, KERN_INFO,
				"FAULT_INJECTION was not selected");
#endif
			break;
J
Jaegeuk Kim 已提交
541 542 543 544 545 546
		case Opt_lazytime:
			sb->s_flags |= MS_LAZYTIME;
			break;
		case Opt_nolazytime:
			sb->s_flags &= ~MS_LAZYTIME;
			break;
547 548 549 550 551 552 553 554 555 556
		default:
			f2fs_msg(sb, KERN_ERR,
				"Unrecognized mount option \"%s\" or missing value",
				p);
			return -EINVAL;
		}
	}
	return 0;
}

J
Jaegeuk Kim 已提交
557 558 559 560
static struct inode *f2fs_alloc_inode(struct super_block *sb)
{
	struct f2fs_inode_info *fi;

561
	fi = kmem_cache_alloc(f2fs_inode_cachep, GFP_F2FS_ZERO);
J
Jaegeuk Kim 已提交
562 563 564 565 566
	if (!fi)
		return NULL;

	init_once((void *) fi);

567 568 569 570 571
	if (percpu_counter_init(&fi->dirty_pages, 0, GFP_NOFS)) {
		kmem_cache_free(f2fs_inode_cachep, fi);
		return NULL;
	}

M
Masanari Iida 已提交
572
	/* Initialize f2fs-specific inode info */
J
Jaegeuk Kim 已提交
573 574 575
	fi->vfs_inode.i_version = 1;
	fi->i_current_depth = 1;
	fi->i_advise = 0;
576
	init_rwsem(&fi->i_sem);
577
	INIT_LIST_HEAD(&fi->dirty_list);
578
	INIT_LIST_HEAD(&fi->gdirty_list);
J
Jaegeuk Kim 已提交
579 580
	INIT_LIST_HEAD(&fi->inmem_pages);
	mutex_init(&fi->inmem_lock);
581 582
	init_rwsem(&fi->dio_rwsem[READ]);
	init_rwsem(&fi->dio_rwsem[WRITE]);
J
Jaegeuk Kim 已提交
583

584 585
	/* Will be used by directory only */
	fi->i_dir_level = F2FS_SB(sb)->dir_level;
J
Jaegeuk Kim 已提交
586 587 588
	return &fi->vfs_inode;
}

589 590 591 592 593 594 595 596 597
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)
	 */
598
	if ((!inode_unhashed(inode) && inode->i_state & I_SYNC)) {
599
		if (!inode->i_nlink && !is_bad_inode(inode)) {
J
Jaegeuk Kim 已提交
600 601
			/* to avoid evict_inode call simultaneously */
			atomic_inc(&inode->i_count);
602 603 604 605
			spin_unlock(&inode->i_lock);

			/* some remained atomic pages should discarded */
			if (f2fs_is_atomic_file(inode))
606
				drop_inmem_pages(inode);
607

J
Jaegeuk Kim 已提交
608 609 610
			/* should remain fi->extent_tree for writepage */
			f2fs_destroy_extent_node(inode);

611
			sb_start_intwrite(inode->i_sb);
612
			f2fs_i_size_write(inode, 0);
613 614

			if (F2FS_HAS_BLOCKS(inode))
615
				f2fs_truncate(inode);
616 617 618

			sb_end_intwrite(inode->i_sb);

619
			fscrypt_put_encryption_info(inode, NULL);
620
			spin_lock(&inode->i_lock);
J
Jaegeuk Kim 已提交
621
			atomic_dec(&inode->i_count);
622
		}
623
		return 0;
624
	}
625

626
	return generic_drop_inode(inode);
627 628
}

629
int f2fs_inode_dirtied(struct inode *inode, bool sync)
630
{
631
	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
632
	int ret = 0;
633 634 635

	spin_lock(&sbi->inode_lock[DIRTY_META]);
	if (is_inode_flag_set(inode, FI_DIRTY_INODE)) {
636 637 638 639
		ret = 1;
	} else {
		set_inode_flag(inode, FI_DIRTY_INODE);
		stat_inc_dirty_inode(sbi, DIRTY_META);
640
	}
641 642
	if (sync && list_empty(&F2FS_I(inode)->gdirty_list)) {
		list_add_tail(&F2FS_I(inode)->gdirty_list,
643
				&sbi->inode_list[DIRTY_META]);
644 645
		inc_page_count(sbi, F2FS_DIRTY_IMETA);
	}
646
	spin_unlock(&sbi->inode_lock[DIRTY_META]);
647
	return ret;
648 649 650 651 652 653 654 655 656 657 658
}

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;
	}
659 660 661 662
	if (!list_empty(&F2FS_I(inode)->gdirty_list)) {
		list_del_init(&F2FS_I(inode)->gdirty_list);
		dec_page_count(sbi, F2FS_DIRTY_IMETA);
	}
663
	clear_inode_flag(inode, FI_DIRTY_INODE);
664
	clear_inode_flag(inode, FI_AUTO_RECOVER);
665
	stat_dec_dirty_inode(F2FS_I_SB(inode), DIRTY_META);
666
	spin_unlock(&sbi->inode_lock[DIRTY_META]);
667 668
}

J
Jaegeuk Kim 已提交
669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687
/*
 * 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);

688
	f2fs_inode_dirtied(inode, false);
J
Jaegeuk Kim 已提交
689 690
}

J
Jaegeuk Kim 已提交
691 692 693 694 695 696
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));
}

697
static void f2fs_destroy_inode(struct inode *inode)
J
Jaegeuk Kim 已提交
698
{
699
	percpu_counter_destroy(&F2FS_I(inode)->dirty_pages);
J
Jaegeuk Kim 已提交
700 701 702
	call_rcu(&inode->i_rcu, f2fs_i_callback);
}

703 704
static void destroy_percpu_info(struct f2fs_sb_info *sbi)
{
705
	percpu_counter_destroy(&sbi->alloc_valid_block_count);
706
	percpu_counter_destroy(&sbi->total_valid_inode_count);
707 708
}

J
Jaegeuk Kim 已提交
709 710 711 712
static void f2fs_put_super(struct super_block *sb)
{
	struct f2fs_sb_info *sbi = F2FS_SB(sb);

713 714
	if (sbi->s_proc) {
		remove_proc_entry("segment_info", sbi->s_proc);
715
		remove_proc_entry("segment_bits", sbi->s_proc);
716 717
		remove_proc_entry(sb->s_id, f2fs_proc_root);
	}
718
	kobject_del(&sbi->s_kobj);
719

J
Jaegeuk Kim 已提交
720 721
	stop_gc_thread(sbi);

722 723 724
	/* prevent remaining shrinker jobs */
	mutex_lock(&sbi->umount_mutex);

725 726 727 728 729
	/*
	 * 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.
	 */
730
	if (is_sbi_flag_set(sbi, SBI_IS_DIRTY) ||
731
			!is_set_ckpt_flags(sbi, CP_UMOUNT_FLAG)) {
732 733 734 735 736
		struct cp_control cpc = {
			.reason = CP_UMOUNT,
		};
		write_checkpoint(sbi, &cpc);
	}
J
Jaegeuk Kim 已提交
737

J
Jaegeuk Kim 已提交
738 739 740
	/* write_checkpoint can update stat informaion */
	f2fs_destroy_stats(sbi);

741 742 743 744
	/*
	 * normally superblock is clean, so we need to release this.
	 * In addition, EIO will skip do checkpoint, we need this as well.
	 */
745
	release_ino_entry(sbi, true);
746

747 748 749
	f2fs_leave_shrinker(sbi);
	mutex_unlock(&sbi->umount_mutex);

750
	/* our cp_error case, we can wait for any writeback page */
751
	f2fs_flush_merged_bios(sbi);
752

J
Jaegeuk Kim 已提交
753 754 755 756 757 758 759 760
	iput(sbi->node_inode);
	iput(sbi->meta_inode);

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

	kfree(sbi->ckpt);
761 762
	kobject_put(&sbi->s_kobj);
	wait_for_completion(&sbi->s_kobj_unregister);
J
Jaegeuk Kim 已提交
763 764

	sb->s_fs_info = NULL;
K
Keith Mok 已提交
765 766
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
Y
Yunlei He 已提交
767
	kfree(sbi->raw_super);
768 769

	destroy_percpu_info(sbi);
J
Jaegeuk Kim 已提交
770 771 772 773 774 775
	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
776
	int err = 0;
J
Jaegeuk Kim 已提交
777

778 779
	trace_f2fs_sync_fs(sb, sync);

780
	if (sync) {
781 782
		struct cp_control cpc;

783 784
		cpc.reason = __get_cp_reason(sbi);

785
		mutex_lock(&sbi->gc_mutex);
C
Chao Yu 已提交
786
		err = write_checkpoint(sbi, &cpc);
787 788
		mutex_unlock(&sbi->gc_mutex);
	}
789
	f2fs_trace_ios(NULL, 1);
J
Jaegeuk Kim 已提交
790

C
Chao Yu 已提交
791
	return err;
J
Jaegeuk Kim 已提交
792 793
}

794 795 796 797
static int f2fs_freeze(struct super_block *sb)
{
	int err;

J
Jaegeuk Kim 已提交
798
	if (f2fs_readonly(sb))
799 800 801 802 803 804 805 806 807 808 809
		return 0;

	err = f2fs_sync_fs(sb, 1);
	return err;
}

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

J
Jaegeuk Kim 已提交
810 811 812 813 814 815 816 817 818 819 820 821 822 823 824
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;
825
	buf->f_bfree = user_block_count - valid_user_blocks(sbi) + ovp_count;
J
Jaegeuk Kim 已提交
826 827
	buf->f_bavail = user_block_count - valid_user_blocks(sbi);

828 829
	buf->f_files = sbi->total_node_count - F2FS_RESERVED_NODE_NUM;
	buf->f_ffree = buf->f_files - valid_inode_count(sbi);
J
Jaegeuk Kim 已提交
830

831
	buf->f_namelen = F2FS_NAME_LEN;
J
Jaegeuk Kim 已提交
832 833 834 835 836 837 838 839 840 841
	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);

842 843 844 845 846 847
	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 {
848
		seq_printf(seq, ",background_gc=%s", "off");
849
	}
J
Jaegeuk Kim 已提交
850 851 852 853 854 855 856 857 858 859 860
	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
861 862
	if (test_opt(sbi, INLINE_XATTR))
		seq_puts(seq, ",inline_xattr");
J
Jaegeuk Kim 已提交
863 864 865 866 867 868 869 870
#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))
871
		seq_puts(seq, ",disable_ext_identify");
872 873
	if (test_opt(sbi, INLINE_DATA))
		seq_puts(seq, ",inline_data");
W
Wanpeng Li 已提交
874 875
	else
		seq_puts(seq, ",noinline_data");
876 877
	if (test_opt(sbi, INLINE_DENTRY))
		seq_puts(seq, ",inline_dentry");
878 879
	else
		seq_puts(seq, ",noinline_dentry");
880
	if (!f2fs_readonly(sbi->sb) && test_opt(sbi, FLUSH_MERGE))
881
		seq_puts(seq, ",flush_merge");
J
Jaegeuk Kim 已提交
882 883
	if (test_opt(sbi, NOBARRIER))
		seq_puts(seq, ",nobarrier");
884 885
	if (test_opt(sbi, FASTBOOT))
		seq_puts(seq, ",fastboot");
886 887
	if (test_opt(sbi, EXTENT_CACHE))
		seq_puts(seq, ",extent_cache");
888 889
	else
		seq_puts(seq, ",noextent_cache");
890 891
	if (test_opt(sbi, DATA_FLUSH))
		seq_puts(seq, ",data_flush");
892 893 894 895 896 897

	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
898 899 900 901 902
	seq_printf(seq, ",active_logs=%u", sbi->active_logs);

	return 0;
}

903 904 905 906
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
907 908
	unsigned int total_segs =
			le32_to_cpu(sbi->raw_super->segment_count_main);
909 910
	int i;

911 912 913
	seq_puts(seq, "format: segment_type|valid_blocks\n"
		"segment_type(0:HD, 1:WD, 2:CD, 3:HN, 4:WN, 5:CN)\n");

914
	for (i = 0; i < total_segs; i++) {
915 916 917
		struct seg_entry *se = get_seg_entry(sbi, i);

		if ((i % 10) == 0)
Y
Yunlei He 已提交
918
			seq_printf(seq, "%-10d", i);
919 920
		seq_printf(seq, "%d|%-3u", se->type,
					get_valid_blocks(sbi, i, 1));
921 922
		if ((i % 10) == 9 || i == (total_segs - 1))
			seq_putc(seq, '\n');
923
		else
924
			seq_putc(seq, ' ');
925
	}
926

927 928 929
	return 0;
}

930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947
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++)
948
			seq_printf(seq, " %.2x", se->cur_valid_map[j]);
949 950 951 952 953
		seq_putc(seq, '\n');
	}
	return 0;
}

954 955 956 957 958 959 960 961 962 963 964
#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,					\
965 966
};

967
F2FS_PROC_FILE_DEF(segment_info);
968
F2FS_PROC_FILE_DEF(segment_bits);
969

970 971 972 973 974 975 976
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);
977
	set_opt(sbi, INLINE_DENTRY);
J
Jaegeuk Kim 已提交
978
	set_opt(sbi, EXTENT_CACHE);
J
Jaegeuk Kim 已提交
979
	sbi->sb->s_flags |= MS_LAZYTIME;
J
Jaegeuk Kim 已提交
980
	set_opt(sbi, FLUSH_MERGE);
981
	if (f2fs_sb_mounted_blkzoned(sbi->sb)) {
982 983 984 985 986
		set_opt_mode(sbi, F2FS_MOUNT_LFS);
		set_opt(sbi, DISCARD);
	} else {
		set_opt_mode(sbi, F2FS_MOUNT_ADAPTIVE);
	}
987 988 989 990 991 992 993

#ifdef CONFIG_F2FS_FS_XATTR
	set_opt(sbi, XATTR_USER);
#endif
#ifdef CONFIG_F2FS_FS_POSIX_ACL
	set_opt(sbi, POSIX_ACL);
#endif
994 995 996 997

#ifdef CONFIG_F2FS_FAULT_INJECTION
	f2fs_build_fault_attr(sbi, 0);
#endif
998 999
}

1000 1001 1002 1003 1004
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;
1005 1006
	bool need_restart_gc = false;
	bool need_stop_gc = false;
1007
	bool no_extent_cache = !test_opt(sbi, EXTENT_CACHE);
1008 1009 1010
#ifdef CONFIG_F2FS_FAULT_INJECTION
	struct f2fs_fault_info ffi = sbi->fault_info;
#endif
1011 1012 1013 1014 1015 1016 1017 1018

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

1019 1020 1021 1022 1023 1024 1025 1026 1027
	/* 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);
	}

1028
	sbi->mount_opt.opt = 0;
1029
	default_options(sbi);
1030

1031 1032 1033 1034 1035 1036 1037
	/* parse mount options */
	err = parse_options(sb, data);
	if (err)
		goto restore_opts;

	/*
	 * Previous and new state of filesystem is RO,
1038
	 * so skip checking GC and FLUSH_MERGE conditions.
1039
	 */
1040
	if (f2fs_readonly(sb) && (*flags & MS_RDONLY))
1041 1042
		goto skip;

1043 1044 1045 1046 1047 1048 1049 1050
	/* 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;
	}

1051 1052 1053 1054 1055 1056 1057 1058
	/*
	 * 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);
1059
			need_restart_gc = true;
1060
		}
1061
	} else if (!sbi->gc_thread) {
1062 1063 1064
		err = start_gc_thread(sbi);
		if (err)
			goto restore_opts;
1065 1066 1067
		need_stop_gc = true;
	}

1068 1069 1070 1071 1072 1073 1074 1075 1076 1077
	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);
	}

1078 1079 1080 1081 1082
	/*
	 * 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)) {
1083
		destroy_flush_cmd_control(sbi);
1084
	} else if (!SM_I(sbi)->cmd_control_info) {
1085 1086
		err = create_flush_cmd_control(sbi);
		if (err)
1087
			goto restore_gc;
1088 1089 1090
	}
skip:
	/* Update the POSIXACL Flag */
1091
	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
1092
		(test_opt(sbi, POSIX_ACL) ? MS_POSIXACL : 0);
1093

1094
	return 0;
1095 1096 1097 1098
restore_gc:
	if (need_restart_gc) {
		if (start_gc_thread(sbi))
			f2fs_msg(sbi->sb, KERN_WARNING,
A
arter97 已提交
1099
				"background gc thread has stopped");
1100 1101 1102
	} else if (need_stop_gc) {
		stop_gc_thread(sbi);
	}
1103 1104 1105
restore_opts:
	sbi->mount_opt = org_mount_opt;
	sbi->active_logs = active_logs;
1106 1107 1108
#ifdef CONFIG_F2FS_FAULT_INJECTION
	sbi->fault_info = ffi;
#endif
1109 1110 1111
	return err;
}

J
Jaegeuk Kim 已提交
1112 1113
static struct super_operations f2fs_sops = {
	.alloc_inode	= f2fs_alloc_inode,
1114
	.drop_inode	= f2fs_drop_inode,
J
Jaegeuk Kim 已提交
1115 1116
	.destroy_inode	= f2fs_destroy_inode,
	.write_inode	= f2fs_write_inode,
1117
	.dirty_inode	= f2fs_dirty_inode,
J
Jaegeuk Kim 已提交
1118 1119 1120 1121
	.show_options	= f2fs_show_options,
	.evict_inode	= f2fs_evict_inode,
	.put_super	= f2fs_put_super,
	.sync_fs	= f2fs_sync_fs,
1122 1123
	.freeze_fs	= f2fs_freeze,
	.unfreeze_fs	= f2fs_unfreeze,
J
Jaegeuk Kim 已提交
1124
	.statfs		= f2fs_statfs,
1125
	.remount_fs	= f2fs_remount,
J
Jaegeuk Kim 已提交
1126 1127
};

1128 1129 1130 1131 1132 1133 1134 1135
#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);
}

1136 1137 1138 1139 1140 1141
static int f2fs_key_prefix(struct inode *inode, u8 **key)
{
	*key = F2FS_I_SB(inode)->key_prefix;
	return F2FS_I_SB(inode)->key_prefix_size;
}

1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157
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;
}

static struct fscrypt_operations f2fs_cryptops = {
	.get_context	= f2fs_get_context,
1158
	.key_prefix	= f2fs_key_prefix,
1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169
	.set_context	= f2fs_set_context,
	.is_encrypted	= f2fs_encrypted_inode,
	.empty_dir	= f2fs_empty_dir,
	.max_namelen	= f2fs_max_namelen,
};
#else
static struct fscrypt_operations f2fs_cryptops = {
	.is_encrypted	= f2fs_encrypted_inode,
};
#endif

J
Jaegeuk Kim 已提交
1170 1171 1172 1173 1174 1175
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;

1176
	if (check_nid_range(sbi, ino))
1177
		return ERR_PTR(-ESTALE);
J
Jaegeuk Kim 已提交
1178 1179 1180 1181 1182 1183 1184 1185 1186

	/*
	 * 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);
1187
	if (unlikely(generation && inode->i_generation != generation)) {
J
Jaegeuk Kim 已提交
1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214
		/* 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
1215
static loff_t max_file_blocks(void)
J
Jaegeuk Kim 已提交
1216
{
1217
	loff_t result = (DEF_ADDRS_PER_INODE - F2FS_INLINE_XATTR_ADDRS);
J
Jaegeuk Kim 已提交
1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233
	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;
}

1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247
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 */
	return __sync_dirty_buffer(bh, WRITE_FLUSH_FUA);
}

1248
static inline bool sanity_check_area_boundary(struct f2fs_sb_info *sbi,
1249
					struct buffer_head *bh)
1250
{
1251 1252
	struct f2fs_super_block *raw_super = (struct f2fs_super_block *)
					(bh->b_data + F2FS_SUPER_OFFSET);
1253
	struct super_block *sb = sbi->sb;
1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266
	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);
1267 1268 1269 1270
	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);
1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 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

	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;
	}

1315
	if (main_end_blkaddr > seg_end_blkaddr) {
1316
		f2fs_msg(sb, KERN_INFO,
1317
			"Wrong MAIN_AREA boundary, start(%u) end(%u) block(%u)",
1318
			main_blkaddr,
1319 1320
			segment0_blkaddr +
				(segment_count << log_blocks_per_seg),
1321 1322
			segment_count_main << log_blocks_per_seg);
		return true;
1323 1324 1325 1326 1327 1328 1329 1330 1331
	} 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)) {
1332
			set_sbi_flag(sbi, SBI_NEED_SB_WRITE);
1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345
			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;
1346 1347 1348 1349
	}
	return false;
}

1350
static int sanity_check_raw_super(struct f2fs_sb_info *sbi,
1351
				struct buffer_head *bh)
J
Jaegeuk Kim 已提交
1352
{
1353 1354
	struct f2fs_super_block *raw_super = (struct f2fs_super_block *)
					(bh->b_data + F2FS_SUPER_OFFSET);
1355
	struct super_block *sb = sbi->sb;
J
Jaegeuk Kim 已提交
1356 1357
	unsigned int blocksize;

1358 1359 1360 1361
	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
1362
		return 1;
1363
	}
J
Jaegeuk Kim 已提交
1364

1365
	/* Currently, support only 4KB page cache size */
1366
	if (F2FS_BLKSIZE != PAGE_SIZE) {
1367
		f2fs_msg(sb, KERN_INFO,
1368
			"Invalid page_cache_size (%lu), supports only 4KB\n",
1369
			PAGE_SIZE);
1370 1371 1372
		return 1;
	}

J
Jaegeuk Kim 已提交
1373 1374
	/* Currently, support only 4KB block size */
	blocksize = 1 << le32_to_cpu(raw_super->log_blocksize);
1375
	if (blocksize != F2FS_BLKSIZE) {
1376 1377 1378
		f2fs_msg(sb, KERN_INFO,
			"Invalid blocksize (%u), supports only 4KB\n",
			blocksize);
J
Jaegeuk Kim 已提交
1379
		return 1;
1380
	}
1381

1382 1383 1384 1385 1386 1387 1388 1389
	/* 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
1390 1391 1392 1393 1394 1395 1396
	/* 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
1397
		return 1;
1398
	}
C
Chao Yu 已提交
1399 1400 1401 1402 1403 1404 1405
	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
1406
		return 1;
1407
	}
1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421

	/* 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 */
1422
	if (sanity_check_area_boundary(sbi, bh))
1423 1424
		return 1;

J
Jaegeuk Kim 已提交
1425 1426 1427
	return 0;
}

1428
int sanity_check_ckpt(struct f2fs_sb_info *sbi)
J
Jaegeuk Kim 已提交
1429 1430
{
	unsigned int total, fsmeta;
1431 1432
	struct f2fs_super_block *raw_super = F2FS_RAW_SUPER(sbi);
	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
J
Jaegeuk Kim 已提交
1433 1434 1435 1436 1437 1438 1439 1440

	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);

1441
	if (unlikely(fsmeta >= total))
J
Jaegeuk Kim 已提交
1442
		return 1;
1443

1444
	if (unlikely(f2fs_cp_error(sbi))) {
1445 1446 1447
		f2fs_msg(sbi->sb, KERN_ERR, "A bug case: need to run fsck");
		return 1;
	}
J
Jaegeuk Kim 已提交
1448 1449 1450 1451 1452 1453
	return 0;
}

static void init_sb_info(struct f2fs_sb_info *sbi)
{
	struct f2fs_super_block *raw_super = sbi->raw_super;
1454
	int i;
J
Jaegeuk Kim 已提交
1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470

	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);
1471
	sbi->cur_victim_sec = NULL_SECNO;
1472
	sbi->max_victim_search = DEF_MAX_VICTIM_SEARCH;
J
Jaegeuk Kim 已提交
1473

1474
	sbi->dir_level = DEF_DIR_LEVEL;
1475
	sbi->interval_time[CP_TIME] = DEF_CP_INTERVAL;
1476
	sbi->interval_time[REQ_TIME] = DEF_IDLE_INTERVAL;
1477
	clear_sbi_flag(sbi, SBI_NEED_FSCK);
1478

1479 1480 1481
	for (i = 0; i < NR_COUNT_TYPE; i++)
		atomic_set(&sbi->nr_pages[i], 0);

1482 1483
	INIT_LIST_HEAD(&sbi->s_list);
	mutex_init(&sbi->umount_mutex);
1484 1485
	mutex_init(&sbi->wio_mutex[NODE]);
	mutex_init(&sbi->wio_mutex[DATA]);
1486
	spin_lock_init(&sbi->cp_lock);
1487 1488 1489 1490 1491 1492

#ifdef CONFIG_F2FS_FS_ENCRYPTION
	memcpy(sbi->key_prefix, F2FS_KEY_DESC_PREFIX,
				F2FS_KEY_DESC_PREFIX_SIZE);
	sbi->key_prefix_size = F2FS_KEY_DESC_PREFIX_SIZE;
#endif
J
Jaegeuk Kim 已提交
1493 1494
}

1495 1496
static int init_percpu_info(struct f2fs_sb_info *sbi)
{
1497
	int err;
1498

1499 1500 1501 1502 1503
	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,
1504
								GFP_KERNEL);
1505 1506
}

1507 1508
/*
 * Read f2fs raw super block.
1509 1510 1511
 * 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.
1512
 */
1513
static int read_raw_super_block(struct f2fs_sb_info *sbi,
1514
			struct f2fs_super_block **raw_super,
1515
			int *valid_super_block, int *recovery)
1516
{
1517
	struct super_block *sb = sbi->sb;
1518
	int block;
1519
	struct buffer_head *bh;
1520
	struct f2fs_super_block *super;
1521
	int err = 0;
1522

Y
Yunlei He 已提交
1523 1524 1525
	super = kzalloc(sizeof(struct f2fs_super_block), GFP_KERNEL);
	if (!super)
		return -ENOMEM;
1526 1527 1528 1529 1530

	for (block = 0; block < 2; block++) {
		bh = sb_bread(sb, block);
		if (!bh) {
			f2fs_msg(sb, KERN_ERR, "Unable to read %dth superblock",
1531
				block + 1);
1532 1533 1534
			err = -EIO;
			continue;
		}
1535

1536
		/* sanity checking of raw super */
1537
		if (sanity_check_raw_super(sbi, bh)) {
1538 1539 1540 1541 1542 1543 1544
			f2fs_msg(sb, KERN_ERR,
				"Can't find valid F2FS filesystem in %dth superblock",
				block + 1);
			err = -EINVAL;
			brelse(bh);
			continue;
		}
1545

1546
		if (!*raw_super) {
1547 1548
			memcpy(super, bh->b_data + F2FS_SUPER_OFFSET,
							sizeof(*super));
1549 1550 1551 1552
			*valid_super_block = block;
			*raw_super = super;
		}
		brelse(bh);
1553 1554
	}

1555 1556 1557
	/* Fail to read any one of the superblocks*/
	if (err < 0)
		*recovery = 1;
1558 1559

	/* No valid superblock */
1560
	if (!*raw_super)
Y
Yunlei He 已提交
1561
		kfree(super);
1562 1563
	else
		err = 0;
1564

1565
	return err;
1566 1567
}

1568
int f2fs_commit_super(struct f2fs_sb_info *sbi, bool recover)
J
Jaegeuk Kim 已提交
1569
{
J
Jaegeuk Kim 已提交
1570
	struct buffer_head *bh;
J
Jaegeuk Kim 已提交
1571 1572
	int err;

1573 1574 1575
	if ((recover && f2fs_readonly(sbi->sb)) ||
				bdev_read_only(sbi->sb->s_bdev)) {
		set_sbi_flag(sbi, SBI_NEED_SB_WRITE);
1576
		return -EROFS;
1577
	}
1578

1579 1580
	/* write back-up superblock first */
	bh = sb_getblk(sbi->sb, sbi->valid_super_block ? 0: 1);
J
Jaegeuk Kim 已提交
1581 1582
	if (!bh)
		return -EIO;
1583
	err = __f2fs_commit_super(bh, F2FS_RAW_SUPER(sbi));
J
Jaegeuk Kim 已提交
1584
	brelse(bh);
C
Chao Yu 已提交
1585 1586 1587

	/* if we are in recovery path, skip writing valid superblock */
	if (recover || err)
J
Jaegeuk Kim 已提交
1588
		return err;
J
Jaegeuk Kim 已提交
1589 1590

	/* write current valid superblock */
1591 1592 1593 1594 1595 1596
	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
1597 1598
}

J
Jaegeuk Kim 已提交
1599 1600 1601
static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
{
	struct f2fs_sb_info *sbi;
1602
	struct f2fs_super_block *raw_super;
J
Jaegeuk Kim 已提交
1603
	struct inode *root;
1604
	int err;
1605
	bool retry = true, need_fsck = false;
1606
	char *options = NULL;
1607
	int recovery, i, valid_super_block;
1608
	struct curseg_info *seg_i;
J
Jaegeuk Kim 已提交
1609

1610
try_onemore:
1611 1612
	err = -EINVAL;
	raw_super = NULL;
1613
	valid_super_block = -1;
1614 1615
	recovery = 0;

J
Jaegeuk Kim 已提交
1616 1617 1618 1619 1620
	/* allocate memory for f2fs-specific super block info */
	sbi = kzalloc(sizeof(struct f2fs_sb_info), GFP_KERNEL);
	if (!sbi)
		return -ENOMEM;

1621 1622
	sbi->sb = sb;

K
Keith Mok 已提交
1623 1624 1625 1626 1627 1628 1629 1630 1631
	/* 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;
	}

1632
	/* set a block size */
1633
	if (unlikely(!sb_set_blocksize(sb, F2FS_BLKSIZE))) {
1634
		f2fs_msg(sb, KERN_ERR, "unable to set blocksize");
J
Jaegeuk Kim 已提交
1635
		goto free_sbi;
1636
	}
J
Jaegeuk Kim 已提交
1637

1638
	err = read_raw_super_block(sbi, &raw_super, &valid_super_block,
1639
								&recovery);
1640 1641 1642
	if (err)
		goto free_sbi;

1643
	sb->s_fs_info = sbi;
1644 1645
	sbi->raw_super = raw_super;

1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665
	/*
	 * 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;
	}
#else
	if (bdev_zoned_model(sb->s_bdev) == BLK_ZONED_HM &&
	    !f2fs_sb_mounted_blkzoned(sb)) {
		f2fs_msg(sb, KERN_ERR,
			 "Zoned block device feature not enabled\n");
		goto free_sb_buf;
	}
#endif

1666
	default_options(sbi);
J
Jaegeuk Kim 已提交
1667
	/* parse mount options */
1668 1669 1670
	options = kstrdup((const char *)data, GFP_KERNEL);
	if (data && !options) {
		err = -ENOMEM;
J
Jaegeuk Kim 已提交
1671
		goto free_sb_buf;
1672 1673 1674 1675 1676
	}

	err = parse_options(sb, options);
	if (err)
		goto free_options;
J
Jaegeuk Kim 已提交
1677

C
Chao Yu 已提交
1678 1679 1680
	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
1681 1682 1683 1684
	sb->s_max_links = F2FS_LINK_MAX;
	get_random_bytes(&sbi->s_next_generation, sizeof(u32));

	sb->s_op = &f2fs_sops;
1685
	sb->s_cop = &f2fs_cryptops;
J
Jaegeuk Kim 已提交
1686 1687 1688 1689 1690 1691 1692 1693 1694
	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 */
1695
	sbi->valid_super_block = valid_super_block;
J
Jaegeuk Kim 已提交
1696 1697
	mutex_init(&sbi->gc_mutex);
	mutex_init(&sbi->cp_mutex);
1698
	init_rwsem(&sbi->node_write);
1699 1700 1701

	/* disallow all the data/node/meta page writes */
	set_sbi_flag(sbi, SBI_POR_DOING);
J
Jaegeuk Kim 已提交
1702
	spin_lock_init(&sbi->stat_lock);
1703

1704
	init_rwsem(&sbi->read_io.io_rwsem);
J
Jaegeuk Kim 已提交
1705 1706 1707
	sbi->read_io.sbi = sbi;
	sbi->read_io.bio = NULL;
	for (i = 0; i < NR_PAGE_TYPE; i++) {
1708
		init_rwsem(&sbi->write_io[i].io_rwsem);
J
Jaegeuk Kim 已提交
1709 1710 1711
		sbi->write_io[i].sbi = sbi;
		sbi->write_io[i].bio = NULL;
	}
1712

1713
	init_rwsem(&sbi->cp_rwsem);
1714
	init_waitqueue_head(&sbi->cp_wait);
J
Jaegeuk Kim 已提交
1715 1716
	init_sb_info(sbi);

1717 1718 1719 1720
	err = init_percpu_info(sbi);
	if (err)
		goto free_options;

J
Jaegeuk Kim 已提交
1721 1722 1723
	/* get an inode for meta space */
	sbi->meta_inode = f2fs_iget(sb, F2FS_META_INO(sbi));
	if (IS_ERR(sbi->meta_inode)) {
1724
		f2fs_msg(sb, KERN_ERR, "Failed to read F2FS meta data inode");
J
Jaegeuk Kim 已提交
1725
		err = PTR_ERR(sbi->meta_inode);
1726
		goto free_options;
J
Jaegeuk Kim 已提交
1727 1728 1729
	}

	err = get_valid_checkpoint(sbi);
1730 1731
	if (err) {
		f2fs_msg(sb, KERN_ERR, "Failed to get valid F2FS checkpoint");
J
Jaegeuk Kim 已提交
1732
		goto free_meta_inode;
1733
	}
J
Jaegeuk Kim 已提交
1734 1735 1736

	sbi->total_valid_node_count =
				le32_to_cpu(sbi->ckpt->valid_node_count);
1737 1738
	percpu_counter_set(&sbi->total_valid_inode_count,
				le32_to_cpu(sbi->ckpt->valid_inode_count));
J
Jaegeuk Kim 已提交
1739 1740 1741 1742
	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;
1743

1744 1745 1746 1747
	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
1748

C
Chao Yu 已提交
1749 1750
	init_extent_cache_info(sbi);

J
Jaegeuk Kim 已提交
1751
	init_ino_entry_info(sbi);
J
Jaegeuk Kim 已提交
1752 1753 1754

	/* setup f2fs internal modules */
	err = build_segment_manager(sbi);
1755 1756 1757
	if (err) {
		f2fs_msg(sb, KERN_ERR,
			"Failed to initialize F2FS segment manager");
J
Jaegeuk Kim 已提交
1758
		goto free_sm;
1759
	}
J
Jaegeuk Kim 已提交
1760
	err = build_node_manager(sbi);
1761 1762 1763
	if (err) {
		f2fs_msg(sb, KERN_ERR,
			"Failed to initialize F2FS node manager");
J
Jaegeuk Kim 已提交
1764
		goto free_nm;
1765
	}
J
Jaegeuk Kim 已提交
1766

1767 1768 1769 1770 1771 1772 1773 1774 1775
	/* 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 =
1776
			le64_to_cpu(seg_i->journal->info.kbytes_written);
1777

J
Jaegeuk Kim 已提交
1778 1779 1780 1781 1782
	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)) {
1783
		f2fs_msg(sb, KERN_ERR, "Failed to read node inode");
J
Jaegeuk Kim 已提交
1784 1785 1786 1787
		err = PTR_ERR(sbi->node_inode);
		goto free_nm;
	}

1788 1789
	f2fs_join_shrinker(sbi);

J
Jaegeuk Kim 已提交
1790
	/* if there are nt orphan nodes free them */
1791 1792 1793
	err = recover_orphan_inodes(sbi);
	if (err)
		goto free_node_inode;
J
Jaegeuk Kim 已提交
1794 1795 1796 1797

	/* read root inode and dentry */
	root = f2fs_iget(sb, F2FS_ROOT_INO(sbi));
	if (IS_ERR(root)) {
1798
		f2fs_msg(sb, KERN_ERR, "Failed to read root inode");
J
Jaegeuk Kim 已提交
1799 1800 1801
		err = PTR_ERR(root);
		goto free_node_inode;
	}
1802
	if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) {
1803
		iput(root);
1804
		err = -EINVAL;
1805
		goto free_node_inode;
1806
	}
J
Jaegeuk Kim 已提交
1807 1808 1809 1810 1811 1812 1813 1814 1815

	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)
1816
		goto free_root_inode;
J
Jaegeuk Kim 已提交
1817

1818 1819 1820
	if (f2fs_proc_root)
		sbi->s_proc = proc_mkdir(sb->s_id, f2fs_proc_root);

1821
	if (sbi->s_proc) {
1822 1823
		proc_create_data("segment_info", S_IRUGO, sbi->s_proc,
				 &f2fs_seq_segment_info_fops, sb);
1824 1825 1826
		proc_create_data("segment_bits", S_IRUGO, sbi->s_proc,
				 &f2fs_seq_segment_bits_fops, sb);
	}
1827

1828 1829 1830 1831 1832
	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)
1833
		goto free_proc;
1834

1835 1836
	/* recover fsynced data */
	if (!test_opt(sbi, DISABLE_ROLL_FORWARD)) {
1837 1838 1839 1840 1841
		/*
		 * 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) &&
1842
				!is_set_ckpt_flags(sbi, CP_UMOUNT_FLAG)) {
1843 1844 1845
			err = -EROFS;
			goto free_kobj;
		}
1846 1847 1848 1849

		if (need_fsck)
			set_sbi_flag(sbi, SBI_NEED_FSCK);

1850 1851 1852
		if (!retry)
			goto skip_recovery;

1853 1854
		err = recover_fsync_data(sbi, false);
		if (err < 0) {
1855
			need_fsck = true;
1856
			f2fs_msg(sb, KERN_ERR,
1857
				"Cannot recover all fsync data errno=%d", err);
1858 1859
			goto free_kobj;
		}
1860 1861 1862 1863 1864 1865 1866 1867 1868
	} 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;
		}
1869
	}
1870
skip_recovery:
1871 1872
	/* recover_fsync_data() cleared this already */
	clear_sbi_flag(sbi, SBI_POR_DOING);
1873

1874 1875 1876 1877
	/*
	 * If filesystem is not mounted as read-only then
	 * do start the gc_thread.
	 */
1878
	if (test_opt(sbi, BG_GC) && !f2fs_readonly(sb)) {
1879 1880 1881 1882 1883
		/* After POR, we can run background GC thread.*/
		err = start_gc_thread(sbi);
		if (err)
			goto free_kobj;
	}
1884
	kfree(options);
1885 1886

	/* recover broken superblock */
1887
	if (recovery) {
1888 1889
		err = f2fs_commit_super(sbi, true);
		f2fs_msg(sb, KERN_INFO,
1890
			"Try to recover %dth superblock, ret: %d",
1891
			sbi->valid_super_block ? 1 : 2, err);
1892 1893
	}

1894
	f2fs_update_time(sbi, CP_TIME);
1895
	f2fs_update_time(sbi, REQ_TIME);
J
Jaegeuk Kim 已提交
1896
	return 0;
1897 1898

free_kobj:
1899
	f2fs_sync_inode_meta(sbi);
1900
	kobject_del(&sbi->s_kobj);
1901 1902
	kobject_put(&sbi->s_kobj);
	wait_for_completion(&sbi->s_kobj_unregister);
1903
free_proc:
1904 1905
	if (sbi->s_proc) {
		remove_proc_entry("segment_info", sbi->s_proc);
1906
		remove_proc_entry("segment_bits", sbi->s_proc);
1907 1908 1909
		remove_proc_entry(sb->s_id, f2fs_proc_root);
	}
	f2fs_destroy_stats(sbi);
J
Jaegeuk Kim 已提交
1910 1911 1912 1913
free_root_inode:
	dput(sb->s_root);
	sb->s_root = NULL;
free_node_inode:
1914
	truncate_inode_pages_final(NODE_MAPPING(sbi));
1915
	mutex_lock(&sbi->umount_mutex);
1916
	release_ino_entry(sbi, true);
1917
	f2fs_leave_shrinker(sbi);
1918 1919 1920 1921 1922 1923 1924
	/*
	 * 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
1925
	iput(sbi->node_inode);
1926
	mutex_unlock(&sbi->umount_mutex);
J
Jaegeuk Kim 已提交
1927 1928 1929 1930 1931 1932 1933 1934
free_nm:
	destroy_node_manager(sbi);
free_sm:
	destroy_segment_manager(sbi);
	kfree(sbi->ckpt);
free_meta_inode:
	make_bad_inode(sbi->meta_inode);
	iput(sbi->meta_inode);
1935
free_options:
1936
	destroy_percpu_info(sbi);
1937
	kfree(options);
J
Jaegeuk Kim 已提交
1938
free_sb_buf:
Y
Yunlei He 已提交
1939
	kfree(raw_super);
J
Jaegeuk Kim 已提交
1940
free_sbi:
K
Keith Mok 已提交
1941 1942
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
J
Jaegeuk Kim 已提交
1943
	kfree(sbi);
1944 1945 1946

	/* give only one another chance */
	if (retry) {
1947
		retry = false;
1948 1949 1950
		shrink_dcache_sb(sb);
		goto try_onemore;
	}
J
Jaegeuk Kim 已提交
1951 1952 1953 1954 1955 1956 1957 1958 1959
	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);
}

1960 1961 1962
static void kill_f2fs_super(struct super_block *sb)
{
	if (sb->s_root)
1963
		set_sbi_flag(F2FS_SB(sb), SBI_IS_CLOSE);
1964 1965 1966
	kill_block_super(sb);
}

J
Jaegeuk Kim 已提交
1967 1968 1969 1970
static struct file_system_type f2fs_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "f2fs",
	.mount		= f2fs_mount,
1971
	.kill_sb	= kill_f2fs_super,
J
Jaegeuk Kim 已提交
1972 1973
	.fs_flags	= FS_REQUIRES_DEV,
};
1974
MODULE_ALIAS_FS("f2fs");
J
Jaegeuk Kim 已提交
1975

1976
static int __init init_inodecache(void)
J
Jaegeuk Kim 已提交
1977
{
1978 1979 1980
	f2fs_inode_cachep = kmem_cache_create("f2fs_inode_cache",
			sizeof(struct f2fs_inode_info), 0,
			SLAB_RECLAIM_ACCOUNT|SLAB_ACCOUNT, NULL);
1981
	if (!f2fs_inode_cachep)
J
Jaegeuk Kim 已提交
1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
		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;

2000 2001
	f2fs_build_trace_ios();

J
Jaegeuk Kim 已提交
2002 2003 2004 2005 2006
	err = init_inodecache();
	if (err)
		goto fail;
	err = create_node_manager_caches();
	if (err)
2007
		goto free_inodecache;
2008
	err = create_segment_manager_caches();
J
Jaegeuk Kim 已提交
2009
	if (err)
2010
		goto free_node_manager_caches;
J
Jaegeuk Kim 已提交
2011 2012
	err = create_checkpoint_caches();
	if (err)
2013
		goto free_segment_manager_caches;
C
Chao Yu 已提交
2014 2015 2016
	err = create_extent_cache();
	if (err)
		goto free_checkpoint_caches;
2017
	f2fs_kset = kset_create_and_add("f2fs", NULL, fs_kobj);
2018 2019
	if (!f2fs_kset) {
		err = -ENOMEM;
C
Chao Yu 已提交
2020
		goto free_extent_cache;
2021
	}
2022
	err = register_shrinker(&f2fs_shrinker_info);
2023
	if (err)
2024
		goto free_kset;
2025 2026 2027 2028

	err = register_filesystem(&f2fs_fs_type);
	if (err)
		goto free_shrinker;
2029 2030 2031
	err = f2fs_create_root_stats();
	if (err)
		goto free_filesystem;
2032
	f2fs_proc_root = proc_mkdir("fs/f2fs", NULL);
2033 2034
	return 0;

2035 2036
free_filesystem:
	unregister_filesystem(&f2fs_fs_type);
2037 2038
free_shrinker:
	unregister_shrinker(&f2fs_shrinker_info);
2039 2040
free_kset:
	kset_unregister(f2fs_kset);
C
Chao Yu 已提交
2041 2042
free_extent_cache:
	destroy_extent_cache();
2043 2044
free_checkpoint_caches:
	destroy_checkpoint_caches();
2045 2046
free_segment_manager_caches:
	destroy_segment_manager_caches();
2047 2048 2049 2050
free_node_manager_caches:
	destroy_node_manager_caches();
free_inodecache:
	destroy_inodecache();
J
Jaegeuk Kim 已提交
2051 2052 2053 2054 2055 2056
fail:
	return err;
}

static void __exit exit_f2fs_fs(void)
{
2057
	remove_proc_entry("fs/f2fs", NULL);
2058
	f2fs_destroy_root_stats();
J
Jaegeuk Kim 已提交
2059
	unregister_filesystem(&f2fs_fs_type);
T
Tiezhu Yang 已提交
2060 2061
	unregister_shrinker(&f2fs_shrinker_info);
	kset_unregister(f2fs_kset);
W
Wanpeng Li 已提交
2062
	destroy_extent_cache();
J
Jaegeuk Kim 已提交
2063
	destroy_checkpoint_caches();
2064
	destroy_segment_manager_caches();
J
Jaegeuk Kim 已提交
2065 2066
	destroy_node_manager_caches();
	destroy_inodecache();
2067
	f2fs_destroy_trace_ios();
J
Jaegeuk Kim 已提交
2068 2069 2070 2071 2072 2073 2074 2075
}

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");