super.c 48.9 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
S
Sheng Yong 已提交
43
struct f2fs_fault_info f2fs_fault;
J
Jaegeuk Kim 已提交
44 45 46

char *fault_name[FAULT_MAX] = {
	[FAULT_KMALLOC]		= "kmalloc",
47
	[FAULT_PAGE_ALLOC]	= "page alloc",
J
Jaegeuk Kim 已提交
48 49 50 51
	[FAULT_ALLOC_NID]	= "alloc nid",
	[FAULT_ORPHAN]		= "orphan",
	[FAULT_BLOCK]		= "no more block",
	[FAULT_DIR_DEPTH]	= "too big dir depth",
J
Jaegeuk Kim 已提交
52
};
S
Sheng Yong 已提交
53 54 55 56 57 58 59 60 61 62 63

static void f2fs_build_fault_attr(unsigned int rate)
{
	if (rate) {
		atomic_set(&f2fs_fault.inject_ops, 0);
		f2fs_fault.inject_rate = rate;
		f2fs_fault.inject_type = (1 << FAULT_MAX) - 1;
	} else {
		memset(&f2fs_fault, 0, sizeof(struct f2fs_fault_info));
	}
}
64 65
#endif

66 67 68 69 70 71 72
/* 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
73
enum {
74
	Opt_gc_background,
J
Jaegeuk Kim 已提交
75
	Opt_disable_roll_forward,
76
	Opt_norecovery,
J
Jaegeuk Kim 已提交
77 78
	Opt_discard,
	Opt_noheap,
79
	Opt_user_xattr,
J
Jaegeuk Kim 已提交
80
	Opt_nouser_xattr,
81
	Opt_acl,
J
Jaegeuk Kim 已提交
82 83 84
	Opt_noacl,
	Opt_active_logs,
	Opt_disable_ext_identify,
J
Jaegeuk Kim 已提交
85
	Opt_inline_xattr,
86
	Opt_inline_data,
87
	Opt_inline_dentry,
88
	Opt_flush_merge,
J
Jaegeuk Kim 已提交
89
	Opt_nobarrier,
90
	Opt_fastboot,
91
	Opt_extent_cache,
92
	Opt_noextent_cache,
W
Wanpeng Li 已提交
93
	Opt_noinline_data,
94
	Opt_data_flush,
95
	Opt_fault_injection,
J
Jaegeuk Kim 已提交
96 97 98 99
	Opt_err,
};

static match_table_t f2fs_tokens = {
100
	{Opt_gc_background, "background_gc=%s"},
J
Jaegeuk Kim 已提交
101
	{Opt_disable_roll_forward, "disable_roll_forward"},
102
	{Opt_norecovery, "norecovery"},
J
Jaegeuk Kim 已提交
103 104
	{Opt_discard, "discard"},
	{Opt_noheap, "no_heap"},
105
	{Opt_user_xattr, "user_xattr"},
J
Jaegeuk Kim 已提交
106
	{Opt_nouser_xattr, "nouser_xattr"},
107
	{Opt_acl, "acl"},
J
Jaegeuk Kim 已提交
108 109 110
	{Opt_noacl, "noacl"},
	{Opt_active_logs, "active_logs=%u"},
	{Opt_disable_ext_identify, "disable_ext_identify"},
J
Jaegeuk Kim 已提交
111
	{Opt_inline_xattr, "inline_xattr"},
112
	{Opt_inline_data, "inline_data"},
113
	{Opt_inline_dentry, "inline_dentry"},
114
	{Opt_flush_merge, "flush_merge"},
J
Jaegeuk Kim 已提交
115
	{Opt_nobarrier, "nobarrier"},
116
	{Opt_fastboot, "fastboot"},
117
	{Opt_extent_cache, "extent_cache"},
118
	{Opt_noextent_cache, "noextent_cache"},
W
Wanpeng Li 已提交
119
	{Opt_noinline_data, "noinline_data"},
120
	{Opt_data_flush, "data_flush"},
121
	{Opt_fault_injection, "fault_injection=%u"},
J
Jaegeuk Kim 已提交
122 123 124
	{Opt_err, NULL},
};

125
/* Sysfs support for f2fs */
126 127 128
enum {
	GC_THREAD,	/* struct f2fs_gc_thread */
	SM_INFO,	/* struct f2fs_sm_info */
129
	NM_INFO,	/* struct f2fs_nm_info */
130
	F2FS_SBI,	/* struct f2fs_sb_info */
S
Sheng Yong 已提交
131 132 133 134
#ifdef CONFIG_F2FS_FAULT_INJECTION
	FAULT_INFO_RATE,	/* struct f2fs_fault_info */
	FAULT_INFO_TYPE,	/* struct f2fs_fault_info */
#endif
135 136
};

137 138 139 140 141
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);
142
	int struct_type;
143 144 145
	int offset;
};

146 147 148 149 150 151
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);
152 153
	else if (struct_type == NM_INFO)
		return (unsigned char *)NM_I(sbi);
154 155
	else if (struct_type == F2FS_SBI)
		return (unsigned char *)sbi;
S
Sheng Yong 已提交
156 157 158 159 160
#ifdef CONFIG_F2FS_FAULT_INJECTION
	else if (struct_type == FAULT_INFO_RATE ||
					struct_type == FAULT_INFO_TYPE)
		return (unsigned char *)&f2fs_fault;
#endif
161 162 163
	return NULL;
}

164 165 166 167 168 169 170 171 172 173 174 175 176
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)));
}

177 178 179
static ssize_t f2fs_sbi_show(struct f2fs_attr *a,
			struct f2fs_sb_info *sbi, char *buf)
{
180
	unsigned char *ptr = NULL;
181 182
	unsigned int *ui;

183 184
	ptr = __struct_ptr(sbi, a->struct_type);
	if (!ptr)
185 186
		return -EINVAL;

187
	ui = (unsigned int *)(ptr + a->offset);
188 189 190 191 192 193 194 195

	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)
{
196
	unsigned char *ptr;
197 198 199 200
	unsigned long t;
	unsigned int *ui;
	ssize_t ret;

201 202
	ptr = __struct_ptr(sbi, a->struct_type);
	if (!ptr)
203 204
		return -EINVAL;

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

	ret = kstrtoul(skip_spaces(buf), 0, &t);
	if (ret < 0)
		return ret;
S
Sheng Yong 已提交
210 211 212 213
#ifdef CONFIG_F2FS_FAULT_INJECTION
	if (a->struct_type == FAULT_INFO_TYPE && t >= (1 << FAULT_MAX))
		return -EINVAL;
#endif
214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244
	*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);
}

245
#define F2FS_ATTR_OFFSET(_struct_type, _name, _mode, _show, _store, _offset) \
246 247 248 249
static struct f2fs_attr f2fs_attr_##_name = {			\
	.attr = {.name = __stringify(_name), .mode = _mode },	\
	.show	= _show,					\
	.store	= _store,					\
250 251
	.struct_type = _struct_type,				\
	.offset = _offset					\
252 253
}

254 255 256 257
#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))
258

259 260 261
#define F2FS_GENERAL_RO_ATTR(name) \
static struct f2fs_attr f2fs_attr_##name = __ATTR(name, 0444, name##_show, NULL)

262 263 264 265 266
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);
267
F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, max_small_discards, max_discards);
J
Jaegeuk Kim 已提交
268
F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, batched_trim_sections, trim_sections);
269 270
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);
271
F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_fsync_blocks, min_fsync_blocks);
272
F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, ram_thresh, ram_thresh);
C
Chao Yu 已提交
273
F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, ra_nid_pages, ra_nid_pages);
C
Chao Yu 已提交
274
F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, dirty_nats_ratio, dirty_nats_ratio);
275
F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, max_victim_search, max_victim_search);
276
F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, dir_level, dir_level);
277
F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, cp_interval, interval_time[CP_TIME]);
278
F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, idle_interval, interval_time[REQ_TIME]);
S
Sheng Yong 已提交
279 280 281 282
#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
283
F2FS_GENERAL_RO_ATTR(lifetime_write_kbytes);
284 285 286 287 288 289

#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),
290
	ATTR_LIST(gc_idle),
291
	ATTR_LIST(reclaim_segments),
292
	ATTR_LIST(max_small_discards),
J
Jaegeuk Kim 已提交
293
	ATTR_LIST(batched_trim_sections),
294 295
	ATTR_LIST(ipu_policy),
	ATTR_LIST(min_ipu_util),
296
	ATTR_LIST(min_fsync_blocks),
297
	ATTR_LIST(max_victim_search),
298
	ATTR_LIST(dir_level),
299
	ATTR_LIST(ram_thresh),
C
Chao Yu 已提交
300
	ATTR_LIST(ra_nid_pages),
C
Chao Yu 已提交
301
	ATTR_LIST(dirty_nats_ratio),
302
	ATTR_LIST(cp_interval),
303
	ATTR_LIST(idle_interval),
304
	ATTR_LIST(lifetime_write_kbytes),
305 306 307 308 309 310 311 312 313 314 315 316 317 318
	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,
};

S
Sheng Yong 已提交
319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334
#ifdef CONFIG_F2FS_FAULT_INJECTION
/* sysfs for f2fs fault injection */
static struct kobject f2fs_fault_inject;

static struct attribute *f2fs_fault_attrs[] = {
	ATTR_LIST(inject_rate),
	ATTR_LIST(inject_type),
	NULL
};

static struct kobj_type f2fs_fault_ktype = {
	.default_attrs	= f2fs_fault_attrs,
	.sysfs_ops	= &f2fs_attr_ops,
};
#endif

335 336 337 338 339 340 341 342 343 344 345 346
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
347 348 349 350 351 352 353
static void init_once(void *foo)
{
	struct f2fs_inode_info *fi = (struct f2fs_inode_info *) foo;

	inode_init_once(&fi->vfs_inode);
}

354 355 356
static int parse_options(struct super_block *sb, char *options)
{
	struct f2fs_sb_info *sbi = F2FS_SB(sb);
357
	struct request_queue *q;
358 359 360 361
	substring_t args[MAX_OPT_ARGS];
	char *p, *name;
	int arg = 0;

362
#ifdef CONFIG_F2FS_FAULT_INJECTION
S
Sheng Yong 已提交
363
	f2fs_build_fault_attr(0);
364
#endif
S
Sheng Yong 已提交
365

366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385
	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;
386
			if (strlen(name) == 2 && !strncmp(name, "on", 2)) {
387
				set_opt(sbi, BG_GC);
388 389
				clear_opt(sbi, FORCE_FG_GC);
			} else if (strlen(name) == 3 && !strncmp(name, "off", 3)) {
390
				clear_opt(sbi, BG_GC);
391 392 393 394 395
				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 {
396 397 398 399 400 401 402 403
				kfree(name);
				return -EINVAL;
			}
			kfree(name);
			break;
		case Opt_disable_roll_forward:
			set_opt(sbi, DISABLE_ROLL_FORWARD);
			break;
404 405 406 407 408 409
		case Opt_norecovery:
			/* this option mounts f2fs with ro */
			set_opt(sbi, DISABLE_ROLL_FORWARD);
			if (!f2fs_readonly(sb))
				return -EINVAL;
			break;
410
		case Opt_discard:
411 412 413 414 415 416 417 418
			q = bdev_get_queue(sb->s_bdev);
			if (blk_queue_discard(q)) {
				set_opt(sbi, DISCARD);
			} else {
				f2fs_msg(sb, KERN_WARNING,
					"mounting with \"discard\" option, but "
					"the device does not support discard");
			}
419 420 421 422 423
			break;
		case Opt_noheap:
			set_opt(sbi, NOHEAP);
			break;
#ifdef CONFIG_F2FS_FS_XATTR
424 425 426
		case Opt_user_xattr:
			set_opt(sbi, XATTR_USER);
			break;
427 428 429
		case Opt_nouser_xattr:
			clear_opt(sbi, XATTR_USER);
			break;
J
Jaegeuk Kim 已提交
430 431 432
		case Opt_inline_xattr:
			set_opt(sbi, INLINE_XATTR);
			break;
433
#else
434 435 436 437
		case Opt_user_xattr:
			f2fs_msg(sb, KERN_INFO,
				"user_xattr options not supported");
			break;
438 439 440 441
		case Opt_nouser_xattr:
			f2fs_msg(sb, KERN_INFO,
				"nouser_xattr options not supported");
			break;
J
Jaegeuk Kim 已提交
442 443 444 445
		case Opt_inline_xattr:
			f2fs_msg(sb, KERN_INFO,
				"inline_xattr options not supported");
			break;
446 447
#endif
#ifdef CONFIG_F2FS_FS_POSIX_ACL
448 449 450
		case Opt_acl:
			set_opt(sbi, POSIX_ACL);
			break;
451 452 453 454
		case Opt_noacl:
			clear_opt(sbi, POSIX_ACL);
			break;
#else
455 456 457
		case Opt_acl:
			f2fs_msg(sb, KERN_INFO, "acl options not supported");
			break;
458 459 460 461 462 463 464 465 466 467 468 469 470 471
		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;
472 473 474
		case Opt_inline_data:
			set_opt(sbi, INLINE_DATA);
			break;
475 476 477
		case Opt_inline_dentry:
			set_opt(sbi, INLINE_DENTRY);
			break;
478 479 480
		case Opt_flush_merge:
			set_opt(sbi, FLUSH_MERGE);
			break;
J
Jaegeuk Kim 已提交
481 482 483
		case Opt_nobarrier:
			set_opt(sbi, NOBARRIER);
			break;
484 485 486
		case Opt_fastboot:
			set_opt(sbi, FASTBOOT);
			break;
487 488 489
		case Opt_extent_cache:
			set_opt(sbi, EXTENT_CACHE);
			break;
490 491 492
		case Opt_noextent_cache:
			clear_opt(sbi, EXTENT_CACHE);
			break;
W
Wanpeng Li 已提交
493 494 495
		case Opt_noinline_data:
			clear_opt(sbi, INLINE_DATA);
			break;
496 497 498
		case Opt_data_flush:
			set_opt(sbi, DATA_FLUSH);
			break;
499 500 501 502
		case Opt_fault_injection:
			if (args->from && match_int(args, &arg))
				return -EINVAL;
#ifdef CONFIG_F2FS_FAULT_INJECTION
S
Sheng Yong 已提交
503
			f2fs_build_fault_attr(arg);
504 505 506 507 508
#else
			f2fs_msg(sb, KERN_INFO,
				"FAULT_INJECTION was not selected");
#endif
			break;
509 510 511 512 513 514 515 516 517 518
		default:
			f2fs_msg(sb, KERN_ERR,
				"Unrecognized mount option \"%s\" or missing value",
				p);
			return -EINVAL;
		}
	}
	return 0;
}

J
Jaegeuk Kim 已提交
519 520 521 522
static struct inode *f2fs_alloc_inode(struct super_block *sb)
{
	struct f2fs_inode_info *fi;

523
	fi = kmem_cache_alloc(f2fs_inode_cachep, GFP_F2FS_ZERO);
J
Jaegeuk Kim 已提交
524 525 526 527 528
	if (!fi)
		return NULL;

	init_once((void *) fi);

M
Masanari Iida 已提交
529
	/* Initialize f2fs-specific inode info */
J
Jaegeuk Kim 已提交
530
	fi->vfs_inode.i_version = 1;
531
	atomic_set(&fi->dirty_pages, 0);
J
Jaegeuk Kim 已提交
532 533
	fi->i_current_depth = 1;
	fi->i_advise = 0;
534
	init_rwsem(&fi->i_sem);
535
	INIT_LIST_HEAD(&fi->dirty_list);
J
Jaegeuk Kim 已提交
536 537
	INIT_LIST_HEAD(&fi->inmem_pages);
	mutex_init(&fi->inmem_lock);
J
Jaegeuk Kim 已提交
538 539 540

	set_inode_flag(fi, FI_NEW_INODE);

J
Jaegeuk Kim 已提交
541 542 543
	if (test_opt(F2FS_SB(sb), INLINE_XATTR))
		set_inode_flag(fi, FI_INLINE_XATTR);

544 545
	/* Will be used by directory only */
	fi->i_dir_level = F2FS_SB(sb)->dir_level;
J
Jaegeuk Kim 已提交
546 547 548
	return &fi->vfs_inode;
}

549 550 551 552 553 554 555 556 557
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)
	 */
558 559
	if (!inode_unhashed(inode) && inode->i_state & I_SYNC) {
		if (!inode->i_nlink && !is_bad_inode(inode)) {
J
Jaegeuk Kim 已提交
560 561
			/* to avoid evict_inode call simultaneously */
			atomic_inc(&inode->i_count);
562 563 564 565
			spin_unlock(&inode->i_lock);

			/* some remained atomic pages should discarded */
			if (f2fs_is_atomic_file(inode))
566
				drop_inmem_pages(inode);
567

J
Jaegeuk Kim 已提交
568 569 570
			/* should remain fi->extent_tree for writepage */
			f2fs_destroy_extent_node(inode);

571 572 573 574
			sb_start_intwrite(inode->i_sb);
			i_size_write(inode, 0);

			if (F2FS_HAS_BLOCKS(inode))
575
				f2fs_truncate(inode, true);
576 577 578

			sb_end_intwrite(inode->i_sb);

579
			fscrypt_put_encryption_info(inode, NULL);
580
			spin_lock(&inode->i_lock);
J
Jaegeuk Kim 已提交
581
			atomic_dec(&inode->i_count);
582
		}
583
		return 0;
584
	}
585 586 587
	return generic_drop_inode(inode);
}

588 589 590 591 592 593 594 595 596 597
/*
 * 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)
{
	set_inode_flag(F2FS_I(inode), FI_DIRTY_INODE);
}

J
Jaegeuk Kim 已提交
598 599 600 601 602 603
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));
}

604
static void f2fs_destroy_inode(struct inode *inode)
J
Jaegeuk Kim 已提交
605 606 607 608 609 610 611 612
{
	call_rcu(&inode->i_rcu, f2fs_i_callback);
}

static void f2fs_put_super(struct super_block *sb)
{
	struct f2fs_sb_info *sbi = F2FS_SB(sb);

613 614
	if (sbi->s_proc) {
		remove_proc_entry("segment_info", sbi->s_proc);
615
		remove_proc_entry("segment_bits", sbi->s_proc);
616 617
		remove_proc_entry(sb->s_id, f2fs_proc_root);
	}
618
	kobject_del(&sbi->s_kobj);
619

J
Jaegeuk Kim 已提交
620 621
	stop_gc_thread(sbi);

622 623 624
	/* prevent remaining shrinker jobs */
	mutex_lock(&sbi->umount_mutex);

625 626 627 628 629
	/*
	 * 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.
	 */
630
	if (is_sbi_flag_set(sbi, SBI_IS_DIRTY) ||
631
			!is_set_ckpt_flags(F2FS_CKPT(sbi), CP_UMOUNT_FLAG)) {
632 633 634 635 636
		struct cp_control cpc = {
			.reason = CP_UMOUNT,
		};
		write_checkpoint(sbi, &cpc);
	}
J
Jaegeuk Kim 已提交
637

J
Jaegeuk Kim 已提交
638 639 640
	/* write_checkpoint can update stat informaion */
	f2fs_destroy_stats(sbi);

641 642 643 644
	/*
	 * normally superblock is clean, so we need to release this.
	 * In addition, EIO will skip do checkpoint, we need this as well.
	 */
645
	release_ino_entry(sbi, true);
646
	release_discard_addrs(sbi);
647

648 649 650
	f2fs_leave_shrinker(sbi);
	mutex_unlock(&sbi->umount_mutex);

651
	/* our cp_error case, we can wait for any writeback page */
652
	f2fs_flush_merged_bios(sbi);
653

J
Jaegeuk Kim 已提交
654 655 656 657 658 659 660 661
	iput(sbi->node_inode);
	iput(sbi->meta_inode);

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

	kfree(sbi->ckpt);
662 663
	kobject_put(&sbi->s_kobj);
	wait_for_completion(&sbi->s_kobj_unregister);
J
Jaegeuk Kim 已提交
664 665

	sb->s_fs_info = NULL;
K
Keith Mok 已提交
666 667
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
Y
Yunlei He 已提交
668
	kfree(sbi->raw_super);
J
Jaegeuk Kim 已提交
669 670 671 672 673 674
	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
675
	int err = 0;
J
Jaegeuk Kim 已提交
676

677 678
	trace_f2fs_sync_fs(sb, sync);

679
	if (sync) {
680 681
		struct cp_control cpc;

682 683
		cpc.reason = __get_cp_reason(sbi);

684
		mutex_lock(&sbi->gc_mutex);
C
Chao Yu 已提交
685
		err = write_checkpoint(sbi, &cpc);
686 687
		mutex_unlock(&sbi->gc_mutex);
	}
688
	f2fs_trace_ios(NULL, 1);
J
Jaegeuk Kim 已提交
689

C
Chao Yu 已提交
690
	return err;
J
Jaegeuk Kim 已提交
691 692
}

693 694 695 696
static int f2fs_freeze(struct super_block *sb)
{
	int err;

J
Jaegeuk Kim 已提交
697
	if (f2fs_readonly(sb))
698 699 700 701 702 703 704 705 706 707 708
		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
709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726
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;
	buf->f_bfree = buf->f_blocks - valid_user_blocks(sbi) - ovp_count;
	buf->f_bavail = user_block_count - valid_user_blocks(sbi);

727 728
	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
729

730
	buf->f_namelen = F2FS_NAME_LEN;
J
Jaegeuk Kim 已提交
731 732 733 734 735 736 737 738 739 740
	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);

741 742 743 744 745 746
	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 {
747
		seq_printf(seq, ",background_gc=%s", "off");
748
	}
J
Jaegeuk Kim 已提交
749 750 751 752 753 754 755 756 757 758 759
	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
760 761
	if (test_opt(sbi, INLINE_XATTR))
		seq_puts(seq, ",inline_xattr");
J
Jaegeuk Kim 已提交
762 763 764 765 766 767 768 769
#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))
770
		seq_puts(seq, ",disable_ext_identify");
771 772
	if (test_opt(sbi, INLINE_DATA))
		seq_puts(seq, ",inline_data");
W
Wanpeng Li 已提交
773 774
	else
		seq_puts(seq, ",noinline_data");
775 776
	if (test_opt(sbi, INLINE_DENTRY))
		seq_puts(seq, ",inline_dentry");
777
	if (!f2fs_readonly(sbi->sb) && test_opt(sbi, FLUSH_MERGE))
778
		seq_puts(seq, ",flush_merge");
J
Jaegeuk Kim 已提交
779 780
	if (test_opt(sbi, NOBARRIER))
		seq_puts(seq, ",nobarrier");
781 782
	if (test_opt(sbi, FASTBOOT))
		seq_puts(seq, ",fastboot");
783 784
	if (test_opt(sbi, EXTENT_CACHE))
		seq_puts(seq, ",extent_cache");
785 786
	else
		seq_puts(seq, ",noextent_cache");
787 788
	if (test_opt(sbi, DATA_FLUSH))
		seq_puts(seq, ",data_flush");
J
Jaegeuk Kim 已提交
789 790 791 792 793
	seq_printf(seq, ",active_logs=%u", sbi->active_logs);

	return 0;
}

794 795 796 797
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
798 799
	unsigned int total_segs =
			le32_to_cpu(sbi->raw_super->segment_count_main);
800 801
	int i;

802 803 804
	seq_puts(seq, "format: segment_type|valid_blocks\n"
		"segment_type(0:HD, 1:WD, 2:CD, 3:HN, 4:WN, 5:CN)\n");

805
	for (i = 0; i < total_segs; i++) {
806 807 808
		struct seg_entry *se = get_seg_entry(sbi, i);

		if ((i % 10) == 0)
Y
Yunlei He 已提交
809
			seq_printf(seq, "%-10d", i);
810 811
		seq_printf(seq, "%d|%-3u", se->type,
					get_valid_blocks(sbi, i, 1));
812 813
		if ((i % 10) == 9 || i == (total_segs - 1))
			seq_putc(seq, '\n');
814
		else
815
			seq_putc(seq, ' ');
816
	}
817

818 819 820
	return 0;
}

821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844
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++)
			seq_printf(seq, "%x ", se->cur_valid_map[j]);
		seq_putc(seq, '\n');
	}
	return 0;
}

845 846 847 848 849 850 851 852 853 854 855 856
#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 = {		\
	.owner = THIS_MODULE,						\
	.open = _name##_open_fs,					\
	.read = seq_read,						\
	.llseek = seq_lseek,						\
	.release = single_release,					\
857 858
};

859
F2FS_PROC_FILE_DEF(segment_info);
860
F2FS_PROC_FILE_DEF(segment_bits);
861

862 863 864 865 866 867 868
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);
J
Jaegeuk Kim 已提交
869
	set_opt(sbi, EXTENT_CACHE);
870 871 872 873 874 875 876 877 878

#ifdef CONFIG_F2FS_FS_XATTR
	set_opt(sbi, XATTR_USER);
#endif
#ifdef CONFIG_F2FS_FS_POSIX_ACL
	set_opt(sbi, POSIX_ACL);
#endif
}

879 880 881 882 883
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;
884 885
	bool need_restart_gc = false;
	bool need_stop_gc = false;
886
	bool no_extent_cache = !test_opt(sbi, EXTENT_CACHE);
887 888 889 890 891 892 893 894

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

895 896 897 898 899 900 901 902 903
	/* 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);
	}

904
	sbi->mount_opt.opt = 0;
905
	default_options(sbi);
906

907 908 909 910 911 912 913
	/* parse mount options */
	err = parse_options(sb, data);
	if (err)
		goto restore_opts;

	/*
	 * Previous and new state of filesystem is RO,
914
	 * so skip checking GC and FLUSH_MERGE conditions.
915
	 */
916
	if (f2fs_readonly(sb) && (*flags & MS_RDONLY))
917 918
		goto skip;

919 920 921 922 923 924 925 926
	/* 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;
	}

927 928 929 930 931 932 933 934
	/*
	 * 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);
935
			need_restart_gc = true;
936
		}
937
	} else if (!sbi->gc_thread) {
938 939 940
		err = start_gc_thread(sbi);
		if (err)
			goto restore_opts;
941 942 943
		need_stop_gc = true;
	}

944 945 946 947 948 949 950 951 952 953
	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);
	}

954 955 956 957 958
	/*
	 * 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)) {
959
		destroy_flush_cmd_control(sbi);
960
	} else if (!SM_I(sbi)->cmd_control_info) {
961 962
		err = create_flush_cmd_control(sbi);
		if (err)
963
			goto restore_gc;
964 965 966
	}
skip:
	/* Update the POSIXACL Flag */
967
	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
968
		(test_opt(sbi, POSIX_ACL) ? MS_POSIXACL : 0);
969

970
	return 0;
971 972 973 974
restore_gc:
	if (need_restart_gc) {
		if (start_gc_thread(sbi))
			f2fs_msg(sbi->sb, KERN_WARNING,
A
arter97 已提交
975
				"background gc thread has stopped");
976 977 978
	} else if (need_stop_gc) {
		stop_gc_thread(sbi);
	}
979 980 981 982 983 984
restore_opts:
	sbi->mount_opt = org_mount_opt;
	sbi->active_logs = active_logs;
	return err;
}

J
Jaegeuk Kim 已提交
985 986
static struct super_operations f2fs_sops = {
	.alloc_inode	= f2fs_alloc_inode,
987
	.drop_inode	= f2fs_drop_inode,
J
Jaegeuk Kim 已提交
988 989
	.destroy_inode	= f2fs_destroy_inode,
	.write_inode	= f2fs_write_inode,
990
	.dirty_inode	= f2fs_dirty_inode,
J
Jaegeuk Kim 已提交
991 992 993 994
	.show_options	= f2fs_show_options,
	.evict_inode	= f2fs_evict_inode,
	.put_super	= f2fs_put_super,
	.sync_fs	= f2fs_sync_fs,
995 996
	.freeze_fs	= f2fs_freeze,
	.unfreeze_fs	= f2fs_unfreeze,
J
Jaegeuk Kim 已提交
997
	.statfs		= f2fs_statfs,
998
	.remount_fs	= f2fs_remount,
J
Jaegeuk Kim 已提交
999 1000
};

1001 1002 1003 1004 1005 1006 1007 1008
#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);
}

1009 1010 1011 1012 1013 1014
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;
}

1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030
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,
1031
	.key_prefix	= f2fs_key_prefix,
1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042
	.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
1043 1044 1045 1046 1047 1048
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;

1049
	if (check_nid_range(sbi, ino))
1050
		return ERR_PTR(-ESTALE);
J
Jaegeuk Kim 已提交
1051 1052 1053 1054 1055 1056 1057 1058 1059

	/*
	 * 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);
1060
	if (unlikely(generation && inode->i_generation != generation)) {
J
Jaegeuk Kim 已提交
1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087
		/* 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
1088
static loff_t max_file_blocks(void)
J
Jaegeuk Kim 已提交
1089
{
1090
	loff_t result = (DEF_ADDRS_PER_INODE - F2FS_INLINE_XATTR_ADDRS);
J
Jaegeuk Kim 已提交
1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106
	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;
}

1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120
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);
}

1121
static inline bool sanity_check_area_boundary(struct f2fs_sb_info *sbi,
1122
					struct buffer_head *bh)
1123
{
1124 1125
	struct f2fs_super_block *raw_super = (struct f2fs_super_block *)
					(bh->b_data + F2FS_SUPER_OFFSET);
1126
	struct super_block *sb = sbi->sb;
1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139
	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);
1140 1141 1142 1143
	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);
1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187

	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;
	}

1188
	if (main_end_blkaddr > seg_end_blkaddr) {
1189
		f2fs_msg(sb, KERN_INFO,
1190
			"Wrong MAIN_AREA boundary, start(%u) end(%u) block(%u)",
1191
			main_blkaddr,
1192 1193
			segment0_blkaddr +
				(segment_count << log_blocks_per_seg),
1194 1195
			segment_count_main << log_blocks_per_seg);
		return true;
1196 1197 1198 1199 1200 1201 1202 1203 1204
	} 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)) {
1205
			set_sbi_flag(sbi, SBI_NEED_SB_WRITE);
1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218
			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;
1219 1220 1221 1222
	}
	return false;
}

1223
static int sanity_check_raw_super(struct f2fs_sb_info *sbi,
1224
				struct buffer_head *bh)
J
Jaegeuk Kim 已提交
1225
{
1226 1227
	struct f2fs_super_block *raw_super = (struct f2fs_super_block *)
					(bh->b_data + F2FS_SUPER_OFFSET);
1228
	struct super_block *sb = sbi->sb;
J
Jaegeuk Kim 已提交
1229 1230
	unsigned int blocksize;

1231 1232 1233 1234
	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
1235
		return 1;
1236
	}
J
Jaegeuk Kim 已提交
1237

1238
	/* Currently, support only 4KB page cache size */
1239
	if (F2FS_BLKSIZE != PAGE_SIZE) {
1240
		f2fs_msg(sb, KERN_INFO,
1241
			"Invalid page_cache_size (%lu), supports only 4KB\n",
1242
			PAGE_SIZE);
1243 1244 1245
		return 1;
	}

J
Jaegeuk Kim 已提交
1246 1247
	/* Currently, support only 4KB block size */
	blocksize = 1 << le32_to_cpu(raw_super->log_blocksize);
1248
	if (blocksize != F2FS_BLKSIZE) {
1249 1250 1251
		f2fs_msg(sb, KERN_INFO,
			"Invalid blocksize (%u), supports only 4KB\n",
			blocksize);
J
Jaegeuk Kim 已提交
1252
		return 1;
1253
	}
1254

1255 1256 1257 1258 1259 1260 1261 1262
	/* 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
1263 1264 1265 1266 1267 1268 1269
	/* 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
1270
		return 1;
1271
	}
C
Chao Yu 已提交
1272 1273 1274 1275 1276 1277 1278
	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
1279
		return 1;
1280
	}
1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294

	/* 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 */
1295
	if (sanity_check_area_boundary(sbi, bh))
1296 1297
		return 1;

J
Jaegeuk Kim 已提交
1298 1299 1300
	return 0;
}

1301
int sanity_check_ckpt(struct f2fs_sb_info *sbi)
J
Jaegeuk Kim 已提交
1302 1303
{
	unsigned int total, fsmeta;
1304 1305
	struct f2fs_super_block *raw_super = F2FS_RAW_SUPER(sbi);
	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
J
Jaegeuk Kim 已提交
1306 1307 1308 1309 1310 1311 1312 1313

	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);

1314
	if (unlikely(fsmeta >= total))
J
Jaegeuk Kim 已提交
1315
		return 1;
1316

1317
	if (unlikely(f2fs_cp_error(sbi))) {
1318 1319 1320
		f2fs_msg(sbi->sb, KERN_ERR, "A bug case: need to run fsck");
		return 1;
	}
J
Jaegeuk Kim 已提交
1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343
	return 0;
}

static void init_sb_info(struct f2fs_sb_info *sbi)
{
	struct f2fs_super_block *raw_super = sbi->raw_super;
	int i;

	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);
1344
	sbi->cur_victim_sec = NULL_SECNO;
1345
	sbi->max_victim_search = DEF_MAX_VICTIM_SEARCH;
J
Jaegeuk Kim 已提交
1346 1347 1348

	for (i = 0; i < NR_COUNT_TYPE; i++)
		atomic_set(&sbi->nr_pages[i], 0);
1349 1350

	sbi->dir_level = DEF_DIR_LEVEL;
1351
	sbi->interval_time[CP_TIME] = DEF_CP_INTERVAL;
1352
	sbi->interval_time[REQ_TIME] = DEF_IDLE_INTERVAL;
1353
	clear_sbi_flag(sbi, SBI_NEED_FSCK);
1354 1355 1356

	INIT_LIST_HEAD(&sbi->s_list);
	mutex_init(&sbi->umount_mutex);
1357 1358 1359 1360 1361 1362

#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
1363 1364
}

1365 1366
/*
 * Read f2fs raw super block.
1367 1368 1369
 * 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.
1370
 */
1371
static int read_raw_super_block(struct f2fs_sb_info *sbi,
1372
			struct f2fs_super_block **raw_super,
1373
			int *valid_super_block, int *recovery)
1374
{
1375
	struct super_block *sb = sbi->sb;
1376
	int block;
1377
	struct buffer_head *bh;
1378
	struct f2fs_super_block *super;
1379
	int err = 0;
1380

Y
Yunlei He 已提交
1381 1382 1383
	super = kzalloc(sizeof(struct f2fs_super_block), GFP_KERNEL);
	if (!super)
		return -ENOMEM;
1384 1385 1386 1387 1388

	for (block = 0; block < 2; block++) {
		bh = sb_bread(sb, block);
		if (!bh) {
			f2fs_msg(sb, KERN_ERR, "Unable to read %dth superblock",
1389
				block + 1);
1390 1391 1392
			err = -EIO;
			continue;
		}
1393

1394
		/* sanity checking of raw super */
1395
		if (sanity_check_raw_super(sbi, bh)) {
1396 1397 1398 1399 1400 1401 1402
			f2fs_msg(sb, KERN_ERR,
				"Can't find valid F2FS filesystem in %dth superblock",
				block + 1);
			err = -EINVAL;
			brelse(bh);
			continue;
		}
1403

1404
		if (!*raw_super) {
1405 1406
			memcpy(super, bh->b_data + F2FS_SUPER_OFFSET,
							sizeof(*super));
1407 1408 1409 1410
			*valid_super_block = block;
			*raw_super = super;
		}
		brelse(bh);
1411 1412
	}

1413 1414 1415
	/* Fail to read any one of the superblocks*/
	if (err < 0)
		*recovery = 1;
1416 1417

	/* No valid superblock */
1418
	if (!*raw_super)
Y
Yunlei He 已提交
1419
		kfree(super);
1420 1421
	else
		err = 0;
1422

1423
	return err;
1424 1425
}

1426
int f2fs_commit_super(struct f2fs_sb_info *sbi, bool recover)
J
Jaegeuk Kim 已提交
1427
{
J
Jaegeuk Kim 已提交
1428
	struct buffer_head *bh;
J
Jaegeuk Kim 已提交
1429 1430
	int err;

1431 1432 1433
	if ((recover && f2fs_readonly(sbi->sb)) ||
				bdev_read_only(sbi->sb->s_bdev)) {
		set_sbi_flag(sbi, SBI_NEED_SB_WRITE);
1434
		return -EROFS;
1435
	}
1436

1437 1438
	/* write back-up superblock first */
	bh = sb_getblk(sbi->sb, sbi->valid_super_block ? 0: 1);
J
Jaegeuk Kim 已提交
1439 1440
	if (!bh)
		return -EIO;
1441
	err = __f2fs_commit_super(bh, F2FS_RAW_SUPER(sbi));
J
Jaegeuk Kim 已提交
1442
	brelse(bh);
C
Chao Yu 已提交
1443 1444 1445

	/* if we are in recovery path, skip writing valid superblock */
	if (recover || err)
J
Jaegeuk Kim 已提交
1446
		return err;
J
Jaegeuk Kim 已提交
1447 1448

	/* write current valid superblock */
1449 1450 1451 1452 1453 1454
	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
1455 1456
}

J
Jaegeuk Kim 已提交
1457 1458 1459
static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
{
	struct f2fs_sb_info *sbi;
1460
	struct f2fs_super_block *raw_super;
J
Jaegeuk Kim 已提交
1461
	struct inode *root;
1462
	int err;
1463
	bool retry = true, need_fsck = false;
1464
	char *options = NULL;
1465
	int recovery, i, valid_super_block;
1466
	struct curseg_info *seg_i;
J
Jaegeuk Kim 已提交
1467

1468
try_onemore:
1469 1470
	err = -EINVAL;
	raw_super = NULL;
1471
	valid_super_block = -1;
1472 1473
	recovery = 0;

J
Jaegeuk Kim 已提交
1474 1475 1476 1477 1478
	/* allocate memory for f2fs-specific super block info */
	sbi = kzalloc(sizeof(struct f2fs_sb_info), GFP_KERNEL);
	if (!sbi)
		return -ENOMEM;

1479 1480
	sbi->sb = sb;

K
Keith Mok 已提交
1481 1482 1483 1484 1485 1486 1487 1488 1489
	/* 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;
	}

1490
	/* set a block size */
1491
	if (unlikely(!sb_set_blocksize(sb, F2FS_BLKSIZE))) {
1492
		f2fs_msg(sb, KERN_ERR, "unable to set blocksize");
J
Jaegeuk Kim 已提交
1493
		goto free_sbi;
1494
	}
J
Jaegeuk Kim 已提交
1495

1496
	err = read_raw_super_block(sbi, &raw_super, &valid_super_block,
1497
								&recovery);
1498 1499 1500
	if (err)
		goto free_sbi;

1501
	sb->s_fs_info = sbi;
1502
	default_options(sbi);
J
Jaegeuk Kim 已提交
1503
	/* parse mount options */
1504 1505 1506
	options = kstrdup((const char *)data, GFP_KERNEL);
	if (data && !options) {
		err = -ENOMEM;
J
Jaegeuk Kim 已提交
1507
		goto free_sb_buf;
1508 1509 1510 1511 1512
	}

	err = parse_options(sb, options);
	if (err)
		goto free_options;
J
Jaegeuk Kim 已提交
1513

C
Chao Yu 已提交
1514 1515 1516
	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
1517 1518 1519 1520
	sb->s_max_links = F2FS_LINK_MAX;
	get_random_bytes(&sbi->s_next_generation, sizeof(u32));

	sb->s_op = &f2fs_sops;
1521
	sb->s_cop = &f2fs_cryptops;
J
Jaegeuk Kim 已提交
1522 1523 1524 1525 1526 1527 1528 1529 1530 1531
	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 */
	sbi->raw_super = raw_super;
1532
	sbi->valid_super_block = valid_super_block;
J
Jaegeuk Kim 已提交
1533
	mutex_init(&sbi->gc_mutex);
1534
	mutex_init(&sbi->writepages);
J
Jaegeuk Kim 已提交
1535
	mutex_init(&sbi->cp_mutex);
1536
	init_rwsem(&sbi->node_write);
1537 1538 1539

	/* disallow all the data/node/meta page writes */
	set_sbi_flag(sbi, SBI_POR_DOING);
J
Jaegeuk Kim 已提交
1540
	spin_lock_init(&sbi->stat_lock);
1541

1542
	init_rwsem(&sbi->read_io.io_rwsem);
J
Jaegeuk Kim 已提交
1543 1544 1545
	sbi->read_io.sbi = sbi;
	sbi->read_io.bio = NULL;
	for (i = 0; i < NR_PAGE_TYPE; i++) {
1546
		init_rwsem(&sbi->write_io[i].io_rwsem);
J
Jaegeuk Kim 已提交
1547 1548 1549
		sbi->write_io[i].sbi = sbi;
		sbi->write_io[i].bio = NULL;
	}
1550

1551
	init_rwsem(&sbi->cp_rwsem);
1552
	init_waitqueue_head(&sbi->cp_wait);
J
Jaegeuk Kim 已提交
1553 1554 1555 1556 1557
	init_sb_info(sbi);

	/* get an inode for meta space */
	sbi->meta_inode = f2fs_iget(sb, F2FS_META_INO(sbi));
	if (IS_ERR(sbi->meta_inode)) {
1558
		f2fs_msg(sb, KERN_ERR, "Failed to read F2FS meta data inode");
J
Jaegeuk Kim 已提交
1559
		err = PTR_ERR(sbi->meta_inode);
1560
		goto free_options;
J
Jaegeuk Kim 已提交
1561 1562 1563
	}

	err = get_valid_checkpoint(sbi);
1564 1565
	if (err) {
		f2fs_msg(sb, KERN_ERR, "Failed to get valid F2FS checkpoint");
J
Jaegeuk Kim 已提交
1566
		goto free_meta_inode;
1567
	}
J
Jaegeuk Kim 已提交
1568 1569 1570 1571 1572 1573 1574 1575 1576 1577

	sbi->total_valid_node_count =
				le32_to_cpu(sbi->ckpt->valid_node_count);
	sbi->total_valid_inode_count =
				le32_to_cpu(sbi->ckpt->valid_inode_count);
	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;
	sbi->alloc_valid_block_count = 0;
1578 1579 1580 1581
	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
1582

C
Chao Yu 已提交
1583 1584
	init_extent_cache_info(sbi);

J
Jaegeuk Kim 已提交
1585
	init_ino_entry_info(sbi);
J
Jaegeuk Kim 已提交
1586 1587 1588

	/* setup f2fs internal modules */
	err = build_segment_manager(sbi);
1589 1590 1591
	if (err) {
		f2fs_msg(sb, KERN_ERR,
			"Failed to initialize F2FS segment manager");
J
Jaegeuk Kim 已提交
1592
		goto free_sm;
1593
	}
J
Jaegeuk Kim 已提交
1594
	err = build_node_manager(sbi);
1595 1596 1597
	if (err) {
		f2fs_msg(sb, KERN_ERR,
			"Failed to initialize F2FS node manager");
J
Jaegeuk Kim 已提交
1598
		goto free_nm;
1599
	}
J
Jaegeuk Kim 已提交
1600

1601 1602 1603 1604 1605 1606 1607 1608 1609
	/* 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 =
1610
			le64_to_cpu(seg_i->journal->info.kbytes_written);
1611

J
Jaegeuk Kim 已提交
1612 1613 1614 1615 1616
	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)) {
1617
		f2fs_msg(sb, KERN_ERR, "Failed to read node inode");
J
Jaegeuk Kim 已提交
1618 1619 1620 1621
		err = PTR_ERR(sbi->node_inode);
		goto free_nm;
	}

1622 1623
	f2fs_join_shrinker(sbi);

J
Jaegeuk Kim 已提交
1624
	/* if there are nt orphan nodes free them */
1625 1626 1627
	err = recover_orphan_inodes(sbi);
	if (err)
		goto free_node_inode;
J
Jaegeuk Kim 已提交
1628 1629 1630 1631

	/* read root inode and dentry */
	root = f2fs_iget(sb, F2FS_ROOT_INO(sbi));
	if (IS_ERR(root)) {
1632
		f2fs_msg(sb, KERN_ERR, "Failed to read root inode");
J
Jaegeuk Kim 已提交
1633 1634 1635
		err = PTR_ERR(root);
		goto free_node_inode;
	}
1636
	if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) {
1637
		iput(root);
1638
		err = -EINVAL;
1639
		goto free_node_inode;
1640
	}
J
Jaegeuk Kim 已提交
1641 1642 1643 1644 1645 1646 1647 1648 1649

	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)
1650
		goto free_root_inode;
J
Jaegeuk Kim 已提交
1651

1652 1653 1654
	if (f2fs_proc_root)
		sbi->s_proc = proc_mkdir(sb->s_id, f2fs_proc_root);

1655
	if (sbi->s_proc) {
1656 1657
		proc_create_data("segment_info", S_IRUGO, sbi->s_proc,
				 &f2fs_seq_segment_info_fops, sb);
1658 1659 1660
		proc_create_data("segment_bits", S_IRUGO, sbi->s_proc,
				 &f2fs_seq_segment_bits_fops, sb);
	}
1661

1662 1663 1664 1665 1666
	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)
1667
		goto free_proc;
1668

1669 1670
	/* recover fsynced data */
	if (!test_opt(sbi, DISABLE_ROLL_FORWARD)) {
1671 1672 1673 1674 1675 1676 1677 1678 1679
		/*
		 * 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) &&
				!is_set_ckpt_flags(sbi->ckpt, CP_UMOUNT_FLAG)) {
			err = -EROFS;
			goto free_kobj;
		}
1680 1681 1682 1683

		if (need_fsck)
			set_sbi_flag(sbi, SBI_NEED_FSCK);

1684 1685
		err = recover_fsync_data(sbi, false);
		if (err < 0) {
1686
			need_fsck = true;
1687
			f2fs_msg(sb, KERN_ERR,
1688
				"Cannot recover all fsync data errno=%d", err);
1689 1690
			goto free_kobj;
		}
1691 1692 1693 1694 1695 1696 1697 1698 1699
	} 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;
		}
1700
	}
1701

1702 1703
	/* recover_fsync_data() cleared this already */
	clear_sbi_flag(sbi, SBI_POR_DOING);
1704

1705 1706 1707 1708
	/*
	 * If filesystem is not mounted as read-only then
	 * do start the gc_thread.
	 */
1709
	if (test_opt(sbi, BG_GC) && !f2fs_readonly(sb)) {
1710 1711 1712 1713 1714
		/* After POR, we can run background GC thread.*/
		err = start_gc_thread(sbi);
		if (err)
			goto free_kobj;
	}
1715
	kfree(options);
1716 1717

	/* recover broken superblock */
1718
	if (recovery) {
1719 1720
		err = f2fs_commit_super(sbi, true);
		f2fs_msg(sb, KERN_INFO,
1721
			"Try to recover %dth superblock, ret: %d",
1722
			sbi->valid_super_block ? 1 : 2, err);
1723 1724
	}

1725
	f2fs_update_time(sbi, CP_TIME);
1726
	f2fs_update_time(sbi, REQ_TIME);
J
Jaegeuk Kim 已提交
1727
	return 0;
1728 1729 1730

free_kobj:
	kobject_del(&sbi->s_kobj);
1731 1732
	kobject_put(&sbi->s_kobj);
	wait_for_completion(&sbi->s_kobj_unregister);
1733
free_proc:
1734 1735
	if (sbi->s_proc) {
		remove_proc_entry("segment_info", sbi->s_proc);
1736
		remove_proc_entry("segment_bits", sbi->s_proc);
1737 1738 1739
		remove_proc_entry(sb->s_id, f2fs_proc_root);
	}
	f2fs_destroy_stats(sbi);
J
Jaegeuk Kim 已提交
1740 1741 1742 1743
free_root_inode:
	dput(sb->s_root);
	sb->s_root = NULL;
free_node_inode:
1744 1745
	mutex_lock(&sbi->umount_mutex);
	f2fs_leave_shrinker(sbi);
J
Jaegeuk Kim 已提交
1746
	iput(sbi->node_inode);
1747
	mutex_unlock(&sbi->umount_mutex);
J
Jaegeuk Kim 已提交
1748 1749 1750 1751 1752 1753 1754 1755
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);
1756 1757
free_options:
	kfree(options);
J
Jaegeuk Kim 已提交
1758
free_sb_buf:
Y
Yunlei He 已提交
1759
	kfree(raw_super);
J
Jaegeuk Kim 已提交
1760
free_sbi:
K
Keith Mok 已提交
1761 1762
	if (sbi->s_chksum_driver)
		crypto_free_shash(sbi->s_chksum_driver);
J
Jaegeuk Kim 已提交
1763
	kfree(sbi);
1764 1765 1766

	/* give only one another chance */
	if (retry) {
1767
		retry = false;
1768 1769 1770
		shrink_dcache_sb(sb);
		goto try_onemore;
	}
J
Jaegeuk Kim 已提交
1771 1772 1773 1774 1775 1776 1777 1778 1779
	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);
}

1780 1781 1782
static void kill_f2fs_super(struct super_block *sb)
{
	if (sb->s_root)
1783
		set_sbi_flag(F2FS_SB(sb), SBI_IS_CLOSE);
1784 1785 1786
	kill_block_super(sb);
}

J
Jaegeuk Kim 已提交
1787 1788 1789 1790
static struct file_system_type f2fs_fs_type = {
	.owner		= THIS_MODULE,
	.name		= "f2fs",
	.mount		= f2fs_mount,
1791
	.kill_sb	= kill_f2fs_super,
J
Jaegeuk Kim 已提交
1792 1793
	.fs_flags	= FS_REQUIRES_DEV,
};
1794
MODULE_ALIAS_FS("f2fs");
J
Jaegeuk Kim 已提交
1795

1796
static int __init init_inodecache(void)
J
Jaegeuk Kim 已提交
1797
{
1798 1799 1800
	f2fs_inode_cachep = kmem_cache_create("f2fs_inode_cache",
			sizeof(struct f2fs_inode_info), 0,
			SLAB_RECLAIM_ACCOUNT|SLAB_ACCOUNT, NULL);
1801
	if (!f2fs_inode_cachep)
J
Jaegeuk Kim 已提交
1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819
		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;

1820 1821
	f2fs_build_trace_ios();

J
Jaegeuk Kim 已提交
1822 1823 1824 1825 1826
	err = init_inodecache();
	if (err)
		goto fail;
	err = create_node_manager_caches();
	if (err)
1827
		goto free_inodecache;
1828
	err = create_segment_manager_caches();
J
Jaegeuk Kim 已提交
1829
	if (err)
1830
		goto free_node_manager_caches;
J
Jaegeuk Kim 已提交
1831 1832
	err = create_checkpoint_caches();
	if (err)
1833
		goto free_segment_manager_caches;
C
Chao Yu 已提交
1834 1835 1836
	err = create_extent_cache();
	if (err)
		goto free_checkpoint_caches;
1837
	f2fs_kset = kset_create_and_add("f2fs", NULL, fs_kobj);
1838 1839
	if (!f2fs_kset) {
		err = -ENOMEM;
C
Chao Yu 已提交
1840
		goto free_extent_cache;
1841
	}
S
Sheng Yong 已提交
1842 1843 1844 1845 1846 1847 1848 1849 1850 1851
#ifdef CONFIG_F2FS_FAULT_INJECTION
	f2fs_fault_inject.kset = f2fs_kset;
	f2fs_build_fault_attr(0);
	err = kobject_init_and_add(&f2fs_fault_inject, &f2fs_fault_ktype,
				NULL, "fault_injection");
	if (err) {
		f2fs_fault_inject.kset = NULL;
		goto free_kset;
	}
#endif
1852
	err = register_shrinker(&f2fs_shrinker_info);
1853
	if (err)
1854
		goto free_kset;
1855 1856 1857 1858

	err = register_filesystem(&f2fs_fs_type);
	if (err)
		goto free_shrinker;
1859 1860 1861
	err = f2fs_create_root_stats();
	if (err)
		goto free_filesystem;
1862
	f2fs_proc_root = proc_mkdir("fs/f2fs", NULL);
1863 1864
	return 0;

1865 1866
free_filesystem:
	unregister_filesystem(&f2fs_fs_type);
1867 1868
free_shrinker:
	unregister_shrinker(&f2fs_shrinker_info);
1869
free_kset:
S
Sheng Yong 已提交
1870 1871 1872 1873
#ifdef CONFIG_F2FS_FAULT_INJECTION
	if (f2fs_fault_inject.kset)
		kobject_put(&f2fs_fault_inject);
#endif
1874
	kset_unregister(f2fs_kset);
C
Chao Yu 已提交
1875 1876
free_extent_cache:
	destroy_extent_cache();
1877 1878
free_checkpoint_caches:
	destroy_checkpoint_caches();
1879 1880
free_segment_manager_caches:
	destroy_segment_manager_caches();
1881 1882 1883 1884
free_node_manager_caches:
	destroy_node_manager_caches();
free_inodecache:
	destroy_inodecache();
J
Jaegeuk Kim 已提交
1885 1886 1887 1888 1889 1890
fail:
	return err;
}

static void __exit exit_f2fs_fs(void)
{
1891
	remove_proc_entry("fs/f2fs", NULL);
1892
	f2fs_destroy_root_stats();
1893
	unregister_shrinker(&f2fs_shrinker_info);
J
Jaegeuk Kim 已提交
1894
	unregister_filesystem(&f2fs_fs_type);
W
Wanpeng Li 已提交
1895
	destroy_extent_cache();
J
Jaegeuk Kim 已提交
1896
	destroy_checkpoint_caches();
1897
	destroy_segment_manager_caches();
J
Jaegeuk Kim 已提交
1898 1899
	destroy_node_manager_caches();
	destroy_inodecache();
S
Sheng Yong 已提交
1900 1901 1902
#ifdef CONFIG_F2FS_FAULT_INJECTION
	kobject_put(&f2fs_fault_inject);
#endif
1903
	kset_unregister(f2fs_kset);
1904
	f2fs_destroy_trace_ios();
J
Jaegeuk Kim 已提交
1905 1906 1907 1908 1909 1910 1911 1912
}

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");