f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
/**
|
|
|
|
* mount.c
|
|
|
|
*
|
|
|
|
* Copyright (c) 2013 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 "fsck.h"
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
#include "node.h"
|
2017-11-03 04:25:12 +00:00
|
|
|
#include "xattr.h"
|
2014-12-13 21:55:59 +00:00
|
|
|
#include <locale.h>
|
fsck: Check write pointer consistency of open zones
On sudden f2fs shutdown, write pointers of zoned block devices can go
further but f2fs meta data keeps current segments at positions before the
write operations. After remounting the f2fs, this inconsistency causes
write operations not at write pointers and "Unaligned write command"
error is reported.
To avoid the error, have f2fs.fsck check consistency of write pointers
of open zones that current segments point to. Compare each current
segment's position and the write pointer position of the open zone. If
inconsistency is found and 'fix_on' flag is set, assign a new zone to the
current segment and check the newly assigned zone has write pointer at
the zone start. Leave the original zone as is to keep data recorded in
it.
To care about fsync data, refer each seg_entry's ckpt_valid_map to get
the last valid block in the zone. If the last valid block is beyond the
current segments position, fsync data exits in the zone. In case fsync
data exists, do not assign a new zone to the current segment not to lose
the fsync data. It is expected that the kernel replay the fsync data and
fix the write pointer inconsistency at mount time.
Also check consistency between write pointer of the zone the current
segment points to with valid block maps of the zone. If the last valid
block is beyond the write pointer position, report to indicate a bug. If
'fix_on' flag is set, assign a new zone to the current segment.
When inconsistencies are found, turn on 'bug_on' flag in fsck_verify() to
ask users to fix them or not. When inconsistencies get fixed, turn on
'force' flag in fsck_verify() to enforce fixes in following checks.
This check and fix is done twice. The first is done at the beginning of
do_fsck() function so that other fixes can reflect the current segment
modification. The second is done in fsck_verify() to reflect updated meta
data by other fixes.
Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-11-28 07:59:29 +00:00
|
|
|
#include <stdbool.h>
|
2017-11-03 04:25:12 +00:00
|
|
|
#ifdef HAVE_LINUX_POSIX_ACL_H
|
|
|
|
#include <linux/posix_acl.h>
|
|
|
|
#endif
|
|
|
|
#ifdef HAVE_SYS_ACL_H
|
|
|
|
#include <sys/acl.h>
|
|
|
|
#endif
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2017-11-14 18:53:32 +00:00
|
|
|
#ifndef ACL_UNDEFINED_TAG
|
|
|
|
#define ACL_UNDEFINED_TAG (0x00)
|
|
|
|
#define ACL_USER_OBJ (0x01)
|
|
|
|
#define ACL_USER (0x02)
|
|
|
|
#define ACL_GROUP_OBJ (0x04)
|
|
|
|
#define ACL_GROUP (0x08)
|
|
|
|
#define ACL_MASK (0x10)
|
|
|
|
#define ACL_OTHER (0x20)
|
|
|
|
#endif
|
|
|
|
|
2016-11-02 10:17:18 +00:00
|
|
|
u32 get_free_segments(struct f2fs_sb_info *sbi)
|
2015-12-17 02:31:38 +00:00
|
|
|
{
|
|
|
|
u32 i, free_segs = 0;
|
|
|
|
|
|
|
|
for (i = 0; i < TOTAL_SEGS(sbi); i++) {
|
|
|
|
struct seg_entry *se = get_seg_entry(sbi, i);
|
|
|
|
|
2018-06-04 08:14:36 +00:00
|
|
|
if (se->valid_blocks == 0x0 && !IS_CUR_SEGNO(sbi, i))
|
2015-12-17 02:31:38 +00:00
|
|
|
free_segs++;
|
|
|
|
}
|
|
|
|
return free_segs;
|
|
|
|
}
|
|
|
|
|
2015-12-10 00:18:44 +00:00
|
|
|
void update_free_segments(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
|
|
|
char *progress = "-*|*-";
|
|
|
|
static int i = 0;
|
|
|
|
|
2017-11-30 01:21:12 +00:00
|
|
|
if (c.dbg_lv)
|
|
|
|
return;
|
|
|
|
|
2015-12-10 00:18:44 +00:00
|
|
|
MSG(0, "\r [ %c ] Free segments: 0x%x", progress[i % 5], get_free_segments(sbi));
|
|
|
|
fflush(stdout);
|
|
|
|
i++;
|
|
|
|
}
|
|
|
|
|
2017-11-03 04:25:12 +00:00
|
|
|
#if defined(HAVE_LINUX_POSIX_ACL_H) || defined(HAVE_SYS_ACL_H)
|
2019-04-24 17:57:30 +00:00
|
|
|
static void print_acl(const u8 *value, int size)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
2019-04-24 17:57:30 +00:00
|
|
|
const struct f2fs_acl_header *hdr = (struct f2fs_acl_header *)value;
|
|
|
|
const struct f2fs_acl_entry *entry = (struct f2fs_acl_entry *)(hdr + 1);
|
|
|
|
const u8 *end = value + size;
|
2017-11-03 04:25:12 +00:00
|
|
|
int i, count;
|
|
|
|
|
|
|
|
if (hdr->a_version != cpu_to_le32(F2FS_ACL_VERSION)) {
|
|
|
|
MSG(0, "Invalid ACL version [0x%x : 0x%x]\n",
|
|
|
|
le32_to_cpu(hdr->a_version), F2FS_ACL_VERSION);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
count = f2fs_acl_count(size);
|
|
|
|
if (count <= 0) {
|
|
|
|
MSG(0, "Invalid ACL value size %d\n", size);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
for (i = 0; i < count; i++) {
|
2019-04-24 17:57:30 +00:00
|
|
|
if ((u8 *)entry > end) {
|
2017-11-03 04:25:12 +00:00
|
|
|
MSG(0, "Invalid ACL entries count %d\n", count);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
switch (le16_to_cpu(entry->e_tag)) {
|
|
|
|
case ACL_USER_OBJ:
|
|
|
|
case ACL_GROUP_OBJ:
|
|
|
|
case ACL_MASK:
|
|
|
|
case ACL_OTHER:
|
|
|
|
MSG(0, "tag:0x%x perm:0x%x\n",
|
|
|
|
le16_to_cpu(entry->e_tag),
|
|
|
|
le16_to_cpu(entry->e_perm));
|
|
|
|
entry = (struct f2fs_acl_entry *)((char *)entry +
|
|
|
|
sizeof(struct f2fs_acl_entry_short));
|
|
|
|
break;
|
|
|
|
case ACL_USER:
|
|
|
|
MSG(0, "tag:0x%x perm:0x%x uid:%u\n",
|
|
|
|
le16_to_cpu(entry->e_tag),
|
|
|
|
le16_to_cpu(entry->e_perm),
|
|
|
|
le32_to_cpu(entry->e_id));
|
|
|
|
entry = (struct f2fs_acl_entry *)((char *)entry +
|
|
|
|
sizeof(struct f2fs_acl_entry));
|
|
|
|
break;
|
|
|
|
case ACL_GROUP:
|
|
|
|
MSG(0, "tag:0x%x perm:0x%x gid:%u\n",
|
|
|
|
le16_to_cpu(entry->e_tag),
|
|
|
|
le16_to_cpu(entry->e_perm),
|
|
|
|
le32_to_cpu(entry->e_id));
|
|
|
|
entry = (struct f2fs_acl_entry *)((char *)entry +
|
|
|
|
sizeof(struct f2fs_acl_entry));
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
MSG(0, "Unknown ACL tag 0x%x\n",
|
|
|
|
le16_to_cpu(entry->e_tag));
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
2019-04-24 17:57:30 +00:00
|
|
|
#endif /* HAVE_LINUX_POSIX_ACL_H || HAVE_SYS_ACL_H */
|
2017-11-03 04:25:12 +00:00
|
|
|
|
2019-04-24 17:57:30 +00:00
|
|
|
static void print_xattr_entry(const struct f2fs_xattr_entry *ent)
|
2017-11-03 04:25:12 +00:00
|
|
|
{
|
2019-04-24 17:57:30 +00:00
|
|
|
const u8 *value = (const u8 *)&ent->e_name[ent->e_name_len];
|
|
|
|
const int size = le16_to_cpu(ent->e_value_size);
|
|
|
|
const struct fscrypt_context *ctx;
|
2017-11-03 04:25:12 +00:00
|
|
|
int i;
|
|
|
|
|
|
|
|
MSG(0, "\nxattr: e_name_index:%d e_name:", ent->e_name_index);
|
2019-04-24 17:57:30 +00:00
|
|
|
for (i = 0; i < ent->e_name_len; i++)
|
2017-11-03 04:25:12 +00:00
|
|
|
MSG(0, "%c", ent->e_name[i]);
|
|
|
|
MSG(0, " e_name_len:%d e_value_size:%d e_value:\n",
|
2019-04-24 17:57:30 +00:00
|
|
|
ent->e_name_len, size);
|
2017-11-03 04:25:12 +00:00
|
|
|
|
|
|
|
switch (ent->e_name_index) {
|
2019-04-24 17:57:30 +00:00
|
|
|
#if defined(HAVE_LINUX_POSIX_ACL_H) || defined(HAVE_SYS_ACL_H)
|
2017-11-03 04:25:12 +00:00
|
|
|
case F2FS_XATTR_INDEX_POSIX_ACL_ACCESS:
|
|
|
|
case F2FS_XATTR_INDEX_POSIX_ACL_DEFAULT:
|
2019-04-24 17:57:30 +00:00
|
|
|
print_acl(value, size);
|
|
|
|
return;
|
|
|
|
#endif
|
2017-11-03 04:25:12 +00:00
|
|
|
case F2FS_XATTR_INDEX_ENCRYPTION:
|
2019-04-24 17:57:30 +00:00
|
|
|
ctx = (const struct fscrypt_context *)value;
|
|
|
|
if (size != sizeof(*ctx) ||
|
|
|
|
ctx->format != FS_ENCRYPTION_CONTEXT_FORMAT_V1)
|
|
|
|
break;
|
2017-11-03 04:25:12 +00:00
|
|
|
MSG(0, "format: %d\n", ctx->format);
|
|
|
|
MSG(0, "contents_encryption_mode: 0x%x\n", ctx->contents_encryption_mode);
|
|
|
|
MSG(0, "filenames_encryption_mode: 0x%x\n", ctx->filenames_encryption_mode);
|
|
|
|
MSG(0, "flags: 0x%x\n", ctx->flags);
|
|
|
|
MSG(0, "master_key_descriptor: ");
|
|
|
|
for (i = 0; i < FS_KEY_DESCRIPTOR_SIZE; i++)
|
|
|
|
MSG(0, "%02X", ctx->master_key_descriptor[i]);
|
|
|
|
MSG(0, "\nnonce: ");
|
|
|
|
for (i = 0; i < FS_KEY_DERIVATION_NONCE_SIZE; i++)
|
|
|
|
MSG(0, "%02X", ctx->nonce[i]);
|
|
|
|
MSG(0, "\n");
|
2019-04-24 17:57:30 +00:00
|
|
|
return;
|
2017-11-03 04:25:12 +00:00
|
|
|
}
|
2019-04-24 17:57:30 +00:00
|
|
|
for (i = 0; i < size; i++)
|
|
|
|
MSG(0, "%02X", value[i]);
|
|
|
|
MSG(0, "\n");
|
2017-11-03 04:25:12 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
void print_inode_info(struct f2fs_sb_info *sbi,
|
|
|
|
struct f2fs_node *node, int name)
|
|
|
|
{
|
|
|
|
struct f2fs_inode *inode = &node->i;
|
|
|
|
void *xattr_addr;
|
|
|
|
struct f2fs_xattr_entry *ent;
|
2019-04-24 17:59:09 +00:00
|
|
|
char en[F2FS_PRINT_NAMELEN];
|
2014-05-14 00:02:55 +00:00
|
|
|
unsigned int i = 0;
|
2018-07-10 14:47:38 +00:00
|
|
|
u32 namelen = le32_to_cpu(inode->i_namelen);
|
2017-03-17 13:56:19 +00:00
|
|
|
int enc_name = file_enc_name(inode);
|
2020-03-16 03:05:04 +00:00
|
|
|
int ofs = get_extra_isize(node);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2019-04-24 17:59:09 +00:00
|
|
|
pretty_print_filename(inode->i_name, namelen, en, enc_name);
|
|
|
|
if (name && en[0]) {
|
2016-10-24 02:16:03 +00:00
|
|
|
MSG(0, " - File name : %s%s\n", en,
|
2017-03-17 13:56:19 +00:00
|
|
|
enc_name ? " <encrypted>" : "");
|
2014-12-13 21:55:59 +00:00
|
|
|
setlocale(LC_ALL, "");
|
|
|
|
MSG(0, " - File size : %'llu (bytes)\n",
|
|
|
|
le64_to_cpu(inode->i_size));
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
DISP_u32(inode, i_mode);
|
2015-04-24 01:40:52 +00:00
|
|
|
DISP_u32(inode, i_advise);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
DISP_u32(inode, i_uid);
|
|
|
|
DISP_u32(inode, i_gid);
|
|
|
|
DISP_u32(inode, i_links);
|
|
|
|
DISP_u64(inode, i_size);
|
|
|
|
DISP_u64(inode, i_blocks);
|
|
|
|
|
|
|
|
DISP_u64(inode, i_atime);
|
|
|
|
DISP_u32(inode, i_atime_nsec);
|
|
|
|
DISP_u64(inode, i_ctime);
|
|
|
|
DISP_u32(inode, i_ctime_nsec);
|
|
|
|
DISP_u64(inode, i_mtime);
|
|
|
|
DISP_u32(inode, i_mtime_nsec);
|
|
|
|
|
|
|
|
DISP_u32(inode, i_generation);
|
|
|
|
DISP_u32(inode, i_current_depth);
|
|
|
|
DISP_u32(inode, i_xattr_nid);
|
|
|
|
DISP_u32(inode, i_flags);
|
2014-10-19 06:51:17 +00:00
|
|
|
DISP_u32(inode, i_inline);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
DISP_u32(inode, i_pino);
|
2016-07-15 12:01:19 +00:00
|
|
|
DISP_u32(inode, i_dir_level);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2019-04-24 17:59:09 +00:00
|
|
|
if (en[0]) {
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
DISP_u32(inode, i_namelen);
|
2016-10-24 02:16:03 +00:00
|
|
|
printf("%-30s\t\t[%s]\n", "i_name", en);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
printf("i_ext: fofs:%x blkaddr:%x len:%x\n",
|
2016-07-15 12:01:20 +00:00
|
|
|
le32_to_cpu(inode->i_ext.fofs),
|
|
|
|
le32_to_cpu(inode->i_ext.blk_addr),
|
|
|
|
le32_to_cpu(inode->i_ext.len));
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2017-11-16 11:50:55 +00:00
|
|
|
if (c.feature & cpu_to_le32(F2FS_FEATURE_EXTRA_ATTR)) {
|
|
|
|
DISP_u16(inode, i_extra_isize);
|
|
|
|
if (c.feature & cpu_to_le32(F2FS_FEATURE_FLEXIBLE_INLINE_XATTR))
|
|
|
|
DISP_u16(inode, i_inline_xattr_size);
|
|
|
|
if (c.feature & cpu_to_le32(F2FS_FEATURE_PRJQUOTA))
|
|
|
|
DISP_u32(inode, i_projid);
|
|
|
|
if (c.feature & cpu_to_le32(F2FS_FEATURE_INODE_CHKSUM))
|
|
|
|
DISP_u32(inode, i_inode_checksum);
|
2018-01-25 06:58:45 +00:00
|
|
|
if (c.feature & cpu_to_le32(F2FS_FEATURE_INODE_CRTIME)) {
|
|
|
|
DISP_u64(inode, i_crtime);
|
|
|
|
DISP_u32(inode, i_crtime_nsec);
|
|
|
|
}
|
2019-03-25 13:19:35 +00:00
|
|
|
if (c.feature & cpu_to_le32(F2FS_FEATURE_COMPRESSION)) {
|
|
|
|
DISP_u64(inode, i_compr_blocks);
|
|
|
|
DISP_u32(inode, i_compress_algrithm);
|
|
|
|
DISP_u32(inode, i_log_cluster_size);
|
|
|
|
DISP_u32(inode, i_padding);
|
|
|
|
}
|
2017-11-16 11:50:55 +00:00
|
|
|
}
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2020-03-16 03:05:04 +00:00
|
|
|
for (i = 0; i < ADDRS_PER_INODE(inode); i++) {
|
2020-04-15 03:26:48 +00:00
|
|
|
block_t blkaddr;
|
2019-03-25 13:20:33 +00:00
|
|
|
char *flag = "";
|
2017-07-26 14:49:57 +00:00
|
|
|
|
2020-04-15 03:26:48 +00:00
|
|
|
if (i + ofs >= DEF_ADDRS_PER_INODE)
|
|
|
|
break;
|
|
|
|
|
|
|
|
blkaddr = le32_to_cpu(inode->i_addr[i + ofs]);
|
|
|
|
|
2019-03-25 13:20:33 +00:00
|
|
|
if (blkaddr == 0x0)
|
|
|
|
continue;
|
|
|
|
if (blkaddr == COMPRESS_ADDR)
|
|
|
|
flag = "cluster flag";
|
|
|
|
else if (blkaddr == NEW_ADDR)
|
|
|
|
flag = "reserved flag";
|
2020-03-16 03:05:04 +00:00
|
|
|
printf("i_addr[0x%x] %-16s\t\t[0x%8x : %u]\n", i + ofs, flag,
|
2019-03-25 13:20:33 +00:00
|
|
|
blkaddr, blkaddr);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
DISP_u32(inode, i_nid[0]); /* direct */
|
|
|
|
DISP_u32(inode, i_nid[1]); /* direct */
|
|
|
|
DISP_u32(inode, i_nid[2]); /* indirect */
|
|
|
|
DISP_u32(inode, i_nid[3]); /* indirect */
|
|
|
|
DISP_u32(inode, i_nid[4]); /* double indirect */
|
|
|
|
|
2017-11-03 04:25:12 +00:00
|
|
|
xattr_addr = read_all_xattrs(sbi, node);
|
2020-04-07 10:01:07 +00:00
|
|
|
if (xattr_addr) {
|
|
|
|
list_for_each_xattr(ent, xattr_addr) {
|
|
|
|
print_xattr_entry(ent);
|
|
|
|
}
|
|
|
|
free(xattr_addr);
|
2017-11-03 04:25:12 +00:00
|
|
|
}
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
printf("\n");
|
|
|
|
}
|
|
|
|
|
2017-11-03 04:25:12 +00:00
|
|
|
void print_node_info(struct f2fs_sb_info *sbi,
|
|
|
|
struct f2fs_node *node_block, int verbose)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
|
|
|
nid_t ino = le32_to_cpu(node_block->footer.ino);
|
|
|
|
nid_t nid = le32_to_cpu(node_block->footer.nid);
|
|
|
|
/* Is this inode? */
|
|
|
|
if (ino == nid) {
|
2016-10-13 23:04:52 +00:00
|
|
|
DBG(verbose, "Node ID [0x%x:%u] is inode\n", nid, nid);
|
2017-11-03 04:25:12 +00:00
|
|
|
print_inode_info(sbi, node_block, verbose);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
} else {
|
2013-07-18 02:20:05 +00:00
|
|
|
int i;
|
|
|
|
u32 *dump_blk = (u32 *)node_block;
|
2016-10-13 23:04:52 +00:00
|
|
|
DBG(verbose,
|
|
|
|
"Node ID [0x%x:%u] is direct node or indirect node.\n",
|
2014-08-28 00:13:23 +00:00
|
|
|
nid, nid);
|
2019-03-25 13:19:35 +00:00
|
|
|
for (i = 0; i < DEF_ADDRS_PER_BLOCK; i++)
|
2016-10-13 23:04:52 +00:00
|
|
|
MSG(verbose, "[%d]\t\t\t[0x%8x : %d]\n",
|
2014-08-28 00:13:23 +00:00
|
|
|
i, dump_blk[i], dump_blk[i]);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2015-12-14 22:29:41 +00:00
|
|
|
static void DISP_label(u_int16_t *name)
|
|
|
|
{
|
|
|
|
char buffer[MAX_VOLUME_NAME];
|
|
|
|
|
|
|
|
utf16_to_utf8(buffer, name, MAX_VOLUME_NAME, MAX_VOLUME_NAME);
|
|
|
|
printf("%-30s" "\t\t[%s]\n", "volum_name", buffer);
|
|
|
|
}
|
|
|
|
|
2015-12-09 01:33:23 +00:00
|
|
|
void print_raw_sb_info(struct f2fs_super_block *sb)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
2016-09-17 01:41:00 +00:00
|
|
|
if (!c.dbg_lv)
|
2013-07-30 07:39:04 +00:00
|
|
|
return;
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
printf("\n");
|
|
|
|
printf("+--------------------------------------------------------+\n");
|
|
|
|
printf("| Super block |\n");
|
|
|
|
printf("+--------------------------------------------------------+\n");
|
|
|
|
|
|
|
|
DISP_u32(sb, magic);
|
|
|
|
DISP_u32(sb, major_ver);
|
2015-12-14 22:29:41 +00:00
|
|
|
|
|
|
|
DISP_label(sb->volume_name);
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
DISP_u32(sb, minor_ver);
|
|
|
|
DISP_u32(sb, log_sectorsize);
|
|
|
|
DISP_u32(sb, log_sectors_per_block);
|
|
|
|
|
|
|
|
DISP_u32(sb, log_blocksize);
|
|
|
|
DISP_u32(sb, log_blocks_per_seg);
|
|
|
|
DISP_u32(sb, segs_per_sec);
|
|
|
|
DISP_u32(sb, secs_per_zone);
|
|
|
|
DISP_u32(sb, checksum_offset);
|
|
|
|
DISP_u64(sb, block_count);
|
|
|
|
|
|
|
|
DISP_u32(sb, section_count);
|
|
|
|
DISP_u32(sb, segment_count);
|
|
|
|
DISP_u32(sb, segment_count_ckpt);
|
|
|
|
DISP_u32(sb, segment_count_sit);
|
|
|
|
DISP_u32(sb, segment_count_nat);
|
|
|
|
|
|
|
|
DISP_u32(sb, segment_count_ssa);
|
|
|
|
DISP_u32(sb, segment_count_main);
|
|
|
|
DISP_u32(sb, segment0_blkaddr);
|
|
|
|
|
|
|
|
DISP_u32(sb, cp_blkaddr);
|
|
|
|
DISP_u32(sb, sit_blkaddr);
|
|
|
|
DISP_u32(sb, nat_blkaddr);
|
|
|
|
DISP_u32(sb, ssa_blkaddr);
|
|
|
|
DISP_u32(sb, main_blkaddr);
|
|
|
|
|
|
|
|
DISP_u32(sb, root_ino);
|
|
|
|
DISP_u32(sb, node_ino);
|
|
|
|
DISP_u32(sb, meta_ino);
|
2014-05-12 13:03:46 +00:00
|
|
|
DISP_u32(sb, cp_payload);
|
2018-09-28 12:25:59 +00:00
|
|
|
DISP_u32(sb, crc);
|
2018-06-08 05:11:48 +00:00
|
|
|
DISP("%-.256s", sb, version);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
printf("\n");
|
|
|
|
}
|
|
|
|
|
|
|
|
void print_ckpt_info(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
|
|
|
struct f2fs_checkpoint *cp = F2FS_CKPT(sbi);
|
|
|
|
|
2016-09-17 01:41:00 +00:00
|
|
|
if (!c.dbg_lv)
|
2013-07-30 07:39:04 +00:00
|
|
|
return;
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
printf("\n");
|
|
|
|
printf("+--------------------------------------------------------+\n");
|
|
|
|
printf("| Checkpoint |\n");
|
|
|
|
printf("+--------------------------------------------------------+\n");
|
|
|
|
|
|
|
|
DISP_u64(cp, checkpoint_ver);
|
|
|
|
DISP_u64(cp, user_block_count);
|
|
|
|
DISP_u64(cp, valid_block_count);
|
|
|
|
DISP_u32(cp, rsvd_segment_count);
|
|
|
|
DISP_u32(cp, overprov_segment_count);
|
|
|
|
DISP_u32(cp, free_segment_count);
|
|
|
|
|
|
|
|
DISP_u32(cp, alloc_type[CURSEG_HOT_NODE]);
|
|
|
|
DISP_u32(cp, alloc_type[CURSEG_WARM_NODE]);
|
|
|
|
DISP_u32(cp, alloc_type[CURSEG_COLD_NODE]);
|
|
|
|
DISP_u32(cp, cur_node_segno[0]);
|
|
|
|
DISP_u32(cp, cur_node_segno[1]);
|
|
|
|
DISP_u32(cp, cur_node_segno[2]);
|
|
|
|
|
|
|
|
DISP_u32(cp, cur_node_blkoff[0]);
|
|
|
|
DISP_u32(cp, cur_node_blkoff[1]);
|
|
|
|
DISP_u32(cp, cur_node_blkoff[2]);
|
|
|
|
|
|
|
|
|
|
|
|
DISP_u32(cp, alloc_type[CURSEG_HOT_DATA]);
|
|
|
|
DISP_u32(cp, alloc_type[CURSEG_WARM_DATA]);
|
|
|
|
DISP_u32(cp, alloc_type[CURSEG_COLD_DATA]);
|
|
|
|
DISP_u32(cp, cur_data_segno[0]);
|
|
|
|
DISP_u32(cp, cur_data_segno[1]);
|
|
|
|
DISP_u32(cp, cur_data_segno[2]);
|
|
|
|
|
|
|
|
DISP_u32(cp, cur_data_blkoff[0]);
|
|
|
|
DISP_u32(cp, cur_data_blkoff[1]);
|
|
|
|
DISP_u32(cp, cur_data_blkoff[2]);
|
|
|
|
|
|
|
|
DISP_u32(cp, ckpt_flags);
|
|
|
|
DISP_u32(cp, cp_pack_total_block_count);
|
|
|
|
DISP_u32(cp, cp_pack_start_sum);
|
|
|
|
DISP_u32(cp, valid_node_count);
|
|
|
|
DISP_u32(cp, valid_inode_count);
|
|
|
|
DISP_u32(cp, next_free_nid);
|
|
|
|
DISP_u32(cp, sit_ver_bitmap_bytesize);
|
|
|
|
DISP_u32(cp, nat_ver_bitmap_bytesize);
|
|
|
|
DISP_u32(cp, checksum_offset);
|
|
|
|
DISP_u64(cp, elapsed_time);
|
|
|
|
|
|
|
|
DISP_u32(cp, sit_nat_version_bitmap[0]);
|
|
|
|
printf("\n\n");
|
|
|
|
}
|
|
|
|
|
2015-03-20 23:57:47 +00:00
|
|
|
void print_cp_state(u32 flag)
|
|
|
|
{
|
|
|
|
MSG(0, "Info: checkpoint state = %x : ", flag);
|
2018-09-19 01:28:37 +00:00
|
|
|
if (flag & CP_QUOTA_NEED_FSCK_FLAG)
|
|
|
|
MSG(0, "%s", " quota_need_fsck");
|
2018-06-21 01:20:32 +00:00
|
|
|
if (flag & CP_LARGE_NAT_BITMAP_FLAG)
|
|
|
|
MSG(0, "%s", " large_nat_bitmap");
|
2018-01-20 21:55:14 +00:00
|
|
|
if (flag & CP_NOCRC_RECOVERY_FLAG)
|
|
|
|
MSG(0, "%s", " allow_nocrc");
|
|
|
|
if (flag & CP_TRIMMED_FLAG)
|
|
|
|
MSG(0, "%s", " trimmed");
|
|
|
|
if (flag & CP_NAT_BITS_FLAG)
|
|
|
|
MSG(0, "%s", " nat_bits");
|
|
|
|
if (flag & CP_CRC_RECOVERY_FLAG)
|
|
|
|
MSG(0, "%s", " crc");
|
|
|
|
if (flag & CP_FASTBOOT_FLAG)
|
|
|
|
MSG(0, "%s", " fastboot");
|
2015-03-20 23:57:47 +00:00
|
|
|
if (flag & CP_FSCK_FLAG)
|
|
|
|
MSG(0, "%s", " fsck");
|
|
|
|
if (flag & CP_ERROR_FLAG)
|
|
|
|
MSG(0, "%s", " error");
|
|
|
|
if (flag & CP_COMPACT_SUM_FLAG)
|
|
|
|
MSG(0, "%s", " compacted_summary");
|
|
|
|
if (flag & CP_ORPHAN_PRESENT_FLAG)
|
|
|
|
MSG(0, "%s", " orphan_inodes");
|
2018-09-20 23:16:07 +00:00
|
|
|
if (flag & CP_DISABLED_FLAG)
|
|
|
|
MSG(0, "%s", " disabled");
|
2020-03-24 09:12:50 +00:00
|
|
|
if (flag & CP_RESIZEFS_FLAG)
|
|
|
|
MSG(0, "%s", " resizefs");
|
2015-03-20 23:57:47 +00:00
|
|
|
if (flag & CP_UMOUNT_FLAG)
|
|
|
|
MSG(0, "%s", " unmount");
|
|
|
|
else
|
|
|
|
MSG(0, "%s", " sudden-power-off");
|
|
|
|
MSG(0, "\n");
|
|
|
|
}
|
|
|
|
|
2015-04-23 23:42:29 +00:00
|
|
|
void print_sb_state(struct f2fs_super_block *sb)
|
|
|
|
{
|
|
|
|
__le32 f = sb->feature;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
MSG(0, "Info: superblock features = %x : ", f);
|
|
|
|
if (f & cpu_to_le32(F2FS_FEATURE_ENCRYPT)) {
|
|
|
|
MSG(0, "%s", " encrypt");
|
|
|
|
}
|
2018-01-11 03:33:47 +00:00
|
|
|
if (f & cpu_to_le32(F2FS_FEATURE_VERITY)) {
|
|
|
|
MSG(0, "%s", " verity");
|
|
|
|
}
|
2016-10-28 07:56:59 +00:00
|
|
|
if (f & cpu_to_le32(F2FS_FEATURE_BLKZONED)) {
|
2017-11-16 11:50:58 +00:00
|
|
|
MSG(0, "%s", " blkzoned");
|
2016-06-13 16:36:38 +00:00
|
|
|
}
|
2017-07-26 14:49:57 +00:00
|
|
|
if (f & cpu_to_le32(F2FS_FEATURE_EXTRA_ATTR)) {
|
2017-11-16 11:50:58 +00:00
|
|
|
MSG(0, "%s", " extra_attr");
|
2017-07-26 14:49:57 +00:00
|
|
|
}
|
2017-07-26 14:54:18 +00:00
|
|
|
if (f & cpu_to_le32(F2FS_FEATURE_PRJQUOTA)) {
|
2017-11-16 11:50:58 +00:00
|
|
|
MSG(0, "%s", " project_quota");
|
2017-07-26 14:54:18 +00:00
|
|
|
}
|
2017-07-26 15:01:35 +00:00
|
|
|
if (f & cpu_to_le32(F2FS_FEATURE_INODE_CHKSUM)) {
|
2017-11-16 11:50:58 +00:00
|
|
|
MSG(0, "%s", " inode_checksum");
|
2017-07-26 15:01:35 +00:00
|
|
|
}
|
2017-09-04 10:58:29 +00:00
|
|
|
if (f & cpu_to_le32(F2FS_FEATURE_FLEXIBLE_INLINE_XATTR)) {
|
2017-11-16 11:50:58 +00:00
|
|
|
MSG(0, "%s", " flexible_inline_xattr");
|
2017-09-04 10:58:29 +00:00
|
|
|
}
|
2017-10-30 23:21:09 +00:00
|
|
|
if (f & cpu_to_le32(F2FS_FEATURE_QUOTA_INO)) {
|
2017-11-16 11:50:58 +00:00
|
|
|
MSG(0, "%s", " quota_ino");
|
2017-10-30 23:21:09 +00:00
|
|
|
}
|
2018-01-25 06:58:45 +00:00
|
|
|
if (f & cpu_to_le32(F2FS_FEATURE_INODE_CRTIME)) {
|
|
|
|
MSG(0, "%s", " inode_crtime");
|
|
|
|
}
|
2018-03-06 03:39:39 +00:00
|
|
|
if (f & cpu_to_le32(F2FS_FEATURE_LOST_FOUND)) {
|
|
|
|
MSG(0, "%s", " lost_found");
|
|
|
|
}
|
2018-09-28 12:25:59 +00:00
|
|
|
if (f & cpu_to_le32(F2FS_FEATURE_SB_CHKSUM)) {
|
|
|
|
MSG(0, "%s", " sb_checksum");
|
|
|
|
}
|
2019-07-11 20:45:41 +00:00
|
|
|
if (f & cpu_to_le32(F2FS_FEATURE_CASEFOLD)) {
|
|
|
|
MSG(0, "%s", " casefold");
|
|
|
|
}
|
2019-03-25 13:19:35 +00:00
|
|
|
if (f & cpu_to_le32(F2FS_FEATURE_COMPRESSION)) {
|
|
|
|
MSG(0, "%s", " compression");
|
|
|
|
}
|
2015-04-23 23:42:29 +00:00
|
|
|
MSG(0, "\n");
|
|
|
|
MSG(0, "Info: superblock encrypt level = %d, salt = ",
|
|
|
|
sb->encryption_level);
|
|
|
|
for (i = 0; i < 16; i++)
|
|
|
|
MSG(0, "%02x", sb->encrypt_pw_salt[i]);
|
|
|
|
MSG(0, "\n");
|
|
|
|
}
|
|
|
|
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
static inline bool is_valid_data_blkaddr(block_t blkaddr)
|
|
|
|
{
|
2019-03-25 13:19:35 +00:00
|
|
|
if (blkaddr == NEW_ADDR || blkaddr == NULL_ADDR ||
|
|
|
|
blkaddr == COMPRESS_ADDR)
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
return 0;
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2019-08-09 10:52:57 +00:00
|
|
|
bool f2fs_is_valid_blkaddr(struct f2fs_sb_info *sbi,
|
|
|
|
block_t blkaddr, int type)
|
|
|
|
{
|
|
|
|
switch (type) {
|
|
|
|
case META_NAT:
|
|
|
|
break;
|
|
|
|
case META_SIT:
|
|
|
|
if (blkaddr >= SIT_BLK_CNT(sbi))
|
|
|
|
return 0;
|
|
|
|
break;
|
|
|
|
case META_SSA:
|
|
|
|
if (blkaddr >= MAIN_BLKADDR(sbi) ||
|
|
|
|
blkaddr < SM_I(sbi)->ssa_blkaddr)
|
|
|
|
return 0;
|
|
|
|
break;
|
|
|
|
case META_CP:
|
|
|
|
if (blkaddr >= SIT_I(sbi)->sit_base_addr ||
|
|
|
|
blkaddr < __start_cp_addr(sbi))
|
|
|
|
return 0;
|
|
|
|
break;
|
|
|
|
case META_POR:
|
|
|
|
if (blkaddr >= MAX_BLKADDR(sbi) ||
|
|
|
|
blkaddr < MAIN_BLKADDR(sbi))
|
|
|
|
return 0;
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
ASSERT(0);
|
|
|
|
}
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline block_t current_sit_addr(struct f2fs_sb_info *sbi,
|
|
|
|
unsigned int start);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Readahead CP/NAT/SIT/SSA pages
|
|
|
|
*/
|
|
|
|
int f2fs_ra_meta_pages(struct f2fs_sb_info *sbi, block_t start, int nrpages,
|
|
|
|
int type)
|
|
|
|
{
|
|
|
|
block_t blkno = start;
|
|
|
|
block_t blkaddr, start_blk = 0, len = 0;
|
|
|
|
|
|
|
|
for (; nrpages-- > 0; blkno++) {
|
|
|
|
|
|
|
|
if (!f2fs_is_valid_blkaddr(sbi, blkno, type))
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
switch (type) {
|
|
|
|
case META_NAT:
|
|
|
|
if (blkno >= NAT_BLOCK_OFFSET(NM_I(sbi)->max_nid))
|
|
|
|
blkno = 0;
|
|
|
|
/* get nat block addr */
|
|
|
|
blkaddr = current_nat_addr(sbi,
|
|
|
|
blkno * NAT_ENTRY_PER_BLOCK, NULL);
|
|
|
|
break;
|
|
|
|
case META_SIT:
|
|
|
|
/* get sit block addr */
|
|
|
|
blkaddr = current_sit_addr(sbi,
|
|
|
|
blkno * SIT_ENTRY_PER_BLOCK);
|
|
|
|
break;
|
|
|
|
case META_SSA:
|
|
|
|
case META_CP:
|
|
|
|
case META_POR:
|
|
|
|
blkaddr = blkno;
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
ASSERT(0);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!len) {
|
|
|
|
start_blk = blkaddr;
|
|
|
|
len = 1;
|
|
|
|
} else if (start_blk + len == blkaddr) {
|
|
|
|
len++;
|
|
|
|
} else {
|
|
|
|
dev_readahead(start_blk << F2FS_BLKSIZE_BITS,
|
|
|
|
len << F2FS_BLKSIZE_BITS);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
out:
|
|
|
|
if (len)
|
|
|
|
dev_readahead(start_blk << F2FS_BLKSIZE_BITS,
|
|
|
|
len << F2FS_BLKSIZE_BITS);
|
|
|
|
return blkno - start;
|
|
|
|
}
|
|
|
|
|
2018-09-28 12:25:58 +00:00
|
|
|
void update_superblock(struct f2fs_super_block *sb, int sb_mask)
|
|
|
|
{
|
|
|
|
int addr, ret;
|
|
|
|
u_int8_t *buf;
|
2018-09-28 12:25:59 +00:00
|
|
|
u32 old_crc, new_crc;
|
2018-09-28 12:25:58 +00:00
|
|
|
|
|
|
|
buf = calloc(BLOCK_SZ, 1);
|
|
|
|
ASSERT(buf);
|
|
|
|
|
2018-09-28 12:25:59 +00:00
|
|
|
if (get_sb(feature) & F2FS_FEATURE_SB_CHKSUM) {
|
|
|
|
old_crc = get_sb(crc);
|
|
|
|
new_crc = f2fs_cal_crc32(F2FS_SUPER_MAGIC, sb,
|
|
|
|
SB_CHKSUM_OFFSET);
|
|
|
|
set_sb(crc, new_crc);
|
|
|
|
MSG(1, "Info: SB CRC is updated (0x%x -> 0x%x)\n",
|
|
|
|
old_crc, new_crc);
|
|
|
|
}
|
|
|
|
|
2018-09-28 12:25:58 +00:00
|
|
|
memcpy(buf + F2FS_SUPER_OFFSET, sb, sizeof(*sb));
|
|
|
|
for (addr = SB0_ADDR; addr < SB_MAX_ADDR; addr++) {
|
|
|
|
if (SB_MASK(addr) & sb_mask) {
|
|
|
|
ret = dev_write_block(buf, addr);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
free(buf);
|
|
|
|
DBG(0, "Info: Done to update superblock\n");
|
|
|
|
}
|
|
|
|
|
2016-03-23 18:50:39 +00:00
|
|
|
static inline int sanity_check_area_boundary(struct f2fs_super_block *sb,
|
2018-09-28 12:25:58 +00:00
|
|
|
enum SB_ADDR sb_addr)
|
2016-03-23 18:50:39 +00:00
|
|
|
{
|
|
|
|
u32 segment0_blkaddr = get_sb(segment0_blkaddr);
|
|
|
|
u32 cp_blkaddr = get_sb(cp_blkaddr);
|
|
|
|
u32 sit_blkaddr = get_sb(sit_blkaddr);
|
|
|
|
u32 nat_blkaddr = get_sb(nat_blkaddr);
|
|
|
|
u32 ssa_blkaddr = get_sb(ssa_blkaddr);
|
|
|
|
u32 main_blkaddr = get_sb(main_blkaddr);
|
|
|
|
u32 segment_count_ckpt = get_sb(segment_count_ckpt);
|
|
|
|
u32 segment_count_sit = get_sb(segment_count_sit);
|
|
|
|
u32 segment_count_nat = get_sb(segment_count_nat);
|
|
|
|
u32 segment_count_ssa = get_sb(segment_count_ssa);
|
|
|
|
u32 segment_count_main = get_sb(segment_count_main);
|
|
|
|
u32 segment_count = get_sb(segment_count);
|
|
|
|
u32 log_blocks_per_seg = get_sb(log_blocks_per_seg);
|
|
|
|
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);
|
|
|
|
|
|
|
|
if (segment0_blkaddr != cp_blkaddr) {
|
|
|
|
MSG(0, "\tMismatch segment0(%u) cp_blkaddr(%u)\n",
|
|
|
|
segment0_blkaddr, cp_blkaddr);
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (cp_blkaddr + (segment_count_ckpt << log_blocks_per_seg) !=
|
|
|
|
sit_blkaddr) {
|
|
|
|
MSG(0, "\tWrong CP boundary, start(%u) end(%u) blocks(%u)\n",
|
|
|
|
cp_blkaddr, sit_blkaddr,
|
|
|
|
segment_count_ckpt << log_blocks_per_seg);
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (sit_blkaddr + (segment_count_sit << log_blocks_per_seg) !=
|
|
|
|
nat_blkaddr) {
|
|
|
|
MSG(0, "\tWrong SIT boundary, start(%u) end(%u) blocks(%u)\n",
|
|
|
|
sit_blkaddr, nat_blkaddr,
|
|
|
|
segment_count_sit << log_blocks_per_seg);
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (nat_blkaddr + (segment_count_nat << log_blocks_per_seg) !=
|
|
|
|
ssa_blkaddr) {
|
|
|
|
MSG(0, "\tWrong NAT boundary, start(%u) end(%u) blocks(%u)\n",
|
|
|
|
nat_blkaddr, ssa_blkaddr,
|
|
|
|
segment_count_nat << log_blocks_per_seg);
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (ssa_blkaddr + (segment_count_ssa << log_blocks_per_seg) !=
|
|
|
|
main_blkaddr) {
|
|
|
|
MSG(0, "\tWrong SSA boundary, start(%u) end(%u) blocks(%u)\n",
|
|
|
|
ssa_blkaddr, main_blkaddr,
|
|
|
|
segment_count_ssa << log_blocks_per_seg);
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (main_end_blkaddr > seg_end_blkaddr) {
|
|
|
|
MSG(0, "\tWrong MAIN_AREA, start(%u) end(%u) block(%u)\n",
|
|
|
|
main_blkaddr,
|
|
|
|
segment0_blkaddr +
|
|
|
|
(segment_count << log_blocks_per_seg),
|
|
|
|
segment_count_main << log_blocks_per_seg);
|
|
|
|
return -1;
|
|
|
|
} else if (main_end_blkaddr < seg_end_blkaddr) {
|
|
|
|
set_sb(segment_count, (main_end_blkaddr -
|
|
|
|
segment0_blkaddr) >> log_blocks_per_seg);
|
|
|
|
|
2018-09-28 12:25:58 +00:00
|
|
|
update_superblock(sb, SB_MASK(sb_addr));
|
|
|
|
MSG(0, "Info: Fix alignment: start(%u) end(%u) block(%u)\n",
|
2016-03-23 18:50:39 +00:00
|
|
|
main_blkaddr,
|
|
|
|
segment0_blkaddr +
|
|
|
|
(segment_count << log_blocks_per_seg),
|
|
|
|
segment_count_main << log_blocks_per_seg);
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2018-09-28 12:25:59 +00:00
|
|
|
static int verify_sb_chksum(struct f2fs_super_block *sb)
|
|
|
|
{
|
|
|
|
if (SB_CHKSUM_OFFSET != get_sb(checksum_offset)) {
|
|
|
|
MSG(0, "\tInvalid SB CRC offset: %u\n",
|
|
|
|
get_sb(checksum_offset));
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
if (f2fs_crc_valid(get_sb(crc), sb,
|
|
|
|
get_sb(checksum_offset))) {
|
|
|
|
MSG(0, "\tInvalid SB CRC: 0x%x\n", get_sb(crc));
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2018-09-28 12:25:58 +00:00
|
|
|
int sanity_check_raw_super(struct f2fs_super_block *sb, enum SB_ADDR sb_addr)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
|
|
|
unsigned int blocksize;
|
2019-06-13 08:10:51 +00:00
|
|
|
unsigned int segment_count, segs_per_sec, secs_per_zone;
|
|
|
|
unsigned int total_sections, blocks_per_seg;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2018-09-28 12:25:59 +00:00
|
|
|
if ((get_sb(feature) & F2FS_FEATURE_SB_CHKSUM) &&
|
|
|
|
verify_sb_chksum(sb))
|
|
|
|
return -1;
|
|
|
|
|
2019-06-13 08:10:51 +00:00
|
|
|
if (F2FS_SUPER_MAGIC != get_sb(magic)) {
|
|
|
|
MSG(0, "Magic Mismatch, valid(0x%x) - read(0x%x)\n",
|
|
|
|
F2FS_SUPER_MAGIC, get_sb(magic));
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
return -1;
|
2019-06-13 08:10:51 +00:00
|
|
|
}
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2019-06-13 08:10:51 +00:00
|
|
|
if (F2FS_BLKSIZE != PAGE_CACHE_SIZE) {
|
|
|
|
MSG(0, "Invalid page_cache_size (%d), supports only 4KB\n",
|
|
|
|
PAGE_CACHE_SIZE);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
return -1;
|
2019-06-13 08:10:51 +00:00
|
|
|
}
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2015-11-22 04:02:18 +00:00
|
|
|
blocksize = 1 << get_sb(log_blocksize);
|
2019-06-13 08:10:51 +00:00
|
|
|
if (F2FS_BLKSIZE != blocksize) {
|
|
|
|
MSG(0, "Invalid blocksize (%u), supports only 4KB\n",
|
|
|
|
blocksize);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
return -1;
|
2019-06-13 08:10:51 +00:00
|
|
|
}
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2016-03-23 18:50:39 +00:00
|
|
|
/* check log blocks per segment */
|
2019-06-13 08:10:51 +00:00
|
|
|
if (get_sb(log_blocks_per_seg) != 9) {
|
|
|
|
MSG(0, "Invalid log blocks per segment (%u)\n",
|
|
|
|
get_sb(log_blocks_per_seg));
|
2016-03-23 18:50:39 +00:00
|
|
|
return -1;
|
2019-06-13 08:10:51 +00:00
|
|
|
}
|
2016-03-23 18:50:39 +00:00
|
|
|
|
|
|
|
/* Currently, support 512/1024/2048/4096 bytes sector size */
|
2015-11-22 04:02:18 +00:00
|
|
|
if (get_sb(log_sectorsize) > F2FS_MAX_LOG_SECTOR_SIZE ||
|
2019-06-13 08:10:51 +00:00
|
|
|
get_sb(log_sectorsize) < F2FS_MIN_LOG_SECTOR_SIZE) {
|
|
|
|
MSG(0, "Invalid log sectorsize (%u)\n", get_sb(log_sectorsize));
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
return -1;
|
2019-06-13 08:10:51 +00:00
|
|
|
}
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2015-11-22 04:02:18 +00:00
|
|
|
if (get_sb(log_sectors_per_block) + get_sb(log_sectorsize) !=
|
2019-06-13 08:10:51 +00:00
|
|
|
F2FS_MAX_LOG_SECTOR_SIZE) {
|
|
|
|
MSG(0, "Invalid log sectors per block(%u) log sectorsize(%u)\n",
|
|
|
|
get_sb(log_sectors_per_block),
|
|
|
|
get_sb(log_sectorsize));
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
return -1;
|
2019-06-13 08:10:51 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
segment_count = get_sb(segment_count);
|
|
|
|
segs_per_sec = get_sb(segs_per_sec);
|
|
|
|
secs_per_zone = get_sb(secs_per_zone);
|
|
|
|
total_sections = get_sb(section_count);
|
|
|
|
|
|
|
|
/* blocks_per_seg should be 512, given the above check */
|
|
|
|
blocks_per_seg = 1 << get_sb(log_blocks_per_seg);
|
|
|
|
|
|
|
|
if (segment_count > F2FS_MAX_SEGMENT ||
|
|
|
|
segment_count < F2FS_MIN_SEGMENTS) {
|
|
|
|
MSG(0, "\tInvalid segment count (%u)\n", segment_count);
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (total_sections > segment_count ||
|
|
|
|
total_sections < F2FS_MIN_SEGMENTS ||
|
|
|
|
segs_per_sec > segment_count || !segs_per_sec) {
|
|
|
|
MSG(0, "\tInvalid segment/section count (%u, %u x %u)\n",
|
|
|
|
segment_count, total_sections, segs_per_sec);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
if ((segment_count / segs_per_sec) < total_sections) {
|
|
|
|
MSG(0, "Small segment_count (%u < %u * %u)\n",
|
|
|
|
segment_count, segs_per_sec, total_sections);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (segment_count > (get_sb(block_count) >> 9)) {
|
|
|
|
MSG(0, "Wrong segment_count / block_count (%u > %llu)\n",
|
|
|
|
segment_count, get_sb(block_count));
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2019-09-23 04:24:59 +00:00
|
|
|
if (sb->devs[0].path[0]) {
|
|
|
|
unsigned int dev_segs = le32_to_cpu(sb->devs[0].total_segments);
|
|
|
|
int i = 1;
|
|
|
|
|
|
|
|
while (i < MAX_DEVICES && sb->devs[i].path[0]) {
|
|
|
|
dev_segs += le32_to_cpu(sb->devs[i].total_segments);
|
|
|
|
i++;
|
|
|
|
}
|
|
|
|
if (segment_count != dev_segs) {
|
|
|
|
MSG(0, "Segment count (%u) mismatch with total segments from devices (%u)",
|
|
|
|
segment_count, dev_segs);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2019-06-13 08:10:51 +00:00
|
|
|
if (secs_per_zone > total_sections || !secs_per_zone) {
|
|
|
|
MSG(0, "Wrong secs_per_zone / total_sections (%u, %u)\n",
|
|
|
|
secs_per_zone, total_sections);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
if (get_sb(extension_count) > F2FS_MAX_EXTENSION ||
|
|
|
|
sb->hot_ext_count > F2FS_MAX_EXTENSION ||
|
|
|
|
get_sb(extension_count) +
|
|
|
|
sb->hot_ext_count > F2FS_MAX_EXTENSION) {
|
|
|
|
MSG(0, "Corrupted extension count (%u + %u > %u)\n",
|
|
|
|
get_sb(extension_count),
|
|
|
|
sb->hot_ext_count,
|
|
|
|
F2FS_MAX_EXTENSION);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (get_sb(cp_payload) > (blocks_per_seg - F2FS_CP_PACKS)) {
|
|
|
|
MSG(0, "Insane cp_payload (%u > %u)\n",
|
|
|
|
get_sb(cp_payload), blocks_per_seg - F2FS_CP_PACKS);
|
|
|
|
return 1;
|
|
|
|
}
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2016-03-23 18:50:39 +00:00
|
|
|
/* check reserved ino info */
|
|
|
|
if (get_sb(node_ino) != 1 || get_sb(meta_ino) != 2 ||
|
2019-06-13 08:10:51 +00:00
|
|
|
get_sb(root_ino) != 3) {
|
|
|
|
MSG(0, "Invalid Fs Meta Ino: node(%u) meta(%u) root(%u)\n",
|
|
|
|
get_sb(node_ino), get_sb(meta_ino), get_sb(root_ino));
|
2016-03-23 18:50:39 +00:00
|
|
|
return -1;
|
2019-06-13 08:10:51 +00:00
|
|
|
}
|
2016-03-23 18:50:39 +00:00
|
|
|
|
2016-10-28 07:57:04 +00:00
|
|
|
/* Check zoned block device feature */
|
2016-11-02 00:23:40 +00:00
|
|
|
if (c.devices[0].zoned_model == F2FS_ZONED_HM &&
|
2016-10-28 07:57:04 +00:00
|
|
|
!(sb->feature & cpu_to_le32(F2FS_FEATURE_BLKZONED))) {
|
|
|
|
MSG(0, "\tMissing zoned block device feature\n");
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
2018-09-28 12:25:58 +00:00
|
|
|
if (sanity_check_area_boundary(sb, sb_addr))
|
2016-03-23 18:50:39 +00:00
|
|
|
return -1;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2018-09-28 12:25:58 +00:00
|
|
|
int validate_super_block(struct f2fs_sb_info *sbi, enum SB_ADDR sb_addr)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
2017-11-30 04:34:37 +00:00
|
|
|
char buf[F2FS_BLKSIZE];
|
2014-11-07 03:34:40 +00:00
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
sbi->raw_super = malloc(sizeof(struct f2fs_super_block));
|
2018-11-26 13:36:00 +00:00
|
|
|
if (!sbi->raw_super)
|
|
|
|
return -ENOMEM;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2018-09-28 12:25:58 +00:00
|
|
|
if (dev_read_block(buf, sb_addr))
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
return -1;
|
|
|
|
|
2017-11-30 04:34:37 +00:00
|
|
|
memcpy(sbi->raw_super, buf + F2FS_SUPER_OFFSET,
|
|
|
|
sizeof(struct f2fs_super_block));
|
|
|
|
|
2018-09-28 12:25:58 +00:00
|
|
|
if (!sanity_check_raw_super(sbi->raw_super, sb_addr)) {
|
2014-11-07 03:34:40 +00:00
|
|
|
/* get kernel version */
|
2016-09-17 01:41:00 +00:00
|
|
|
if (c.kd >= 0) {
|
|
|
|
dev_read_version(c.version, 0, VERSION_LEN);
|
|
|
|
get_kernel_version(c.version);
|
2014-11-07 03:34:40 +00:00
|
|
|
} else {
|
2018-05-01 20:54:25 +00:00
|
|
|
get_kernel_uname_version(c.version);
|
2014-11-07 03:34:40 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* build sb version */
|
2016-09-17 01:41:00 +00:00
|
|
|
memcpy(c.sb_version, sbi->raw_super->version, VERSION_LEN);
|
|
|
|
get_kernel_version(c.sb_version);
|
|
|
|
memcpy(c.init_version, sbi->raw_super->init_version, VERSION_LEN);
|
|
|
|
get_kernel_version(c.init_version);
|
2014-11-07 03:34:40 +00:00
|
|
|
|
2016-09-17 01:41:00 +00:00
|
|
|
MSG(0, "Info: MKFS version\n \"%s\"\n", c.init_version);
|
2014-11-07 03:34:40 +00:00
|
|
|
MSG(0, "Info: FSCK version\n from \"%s\"\n to \"%s\"\n",
|
2016-09-17 01:41:00 +00:00
|
|
|
c.sb_version, c.version);
|
2019-10-14 17:10:31 +00:00
|
|
|
if (!c.no_kernel_check &&
|
|
|
|
memcmp(c.sb_version, c.version, VERSION_LEN)) {
|
2014-11-07 03:34:40 +00:00
|
|
|
memcpy(sbi->raw_super->version,
|
2016-09-17 01:41:00 +00:00
|
|
|
c.version, VERSION_LEN);
|
2018-09-28 12:25:58 +00:00
|
|
|
update_superblock(sbi->raw_super, SB_MASK(sb_addr));
|
2014-11-07 03:34:40 +00:00
|
|
|
|
2016-09-17 01:41:00 +00:00
|
|
|
c.auto_fix = 0;
|
|
|
|
c.fix_on = 1;
|
2014-11-07 03:34:40 +00:00
|
|
|
}
|
2015-04-23 23:42:29 +00:00
|
|
|
print_sb_state(sbi->raw_super);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
return 0;
|
2014-11-07 03:34:40 +00:00
|
|
|
}
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
free(sbi->raw_super);
|
2016-01-21 01:13:14 +00:00
|
|
|
sbi->raw_super = NULL;
|
2018-09-28 12:25:58 +00:00
|
|
|
MSG(0, "\tCan't find a valid F2FS superblock at 0x%x\n", sb_addr);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
|
|
|
int init_sb_info(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
2015-11-22 04:02:18 +00:00
|
|
|
struct f2fs_super_block *sb = F2FS_RAW_SUPER(sbi);
|
2015-03-21 00:07:48 +00:00
|
|
|
u64 total_sectors;
|
2016-11-02 00:23:40 +00:00
|
|
|
int i;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2015-11-22 04:02:18 +00:00
|
|
|
sbi->log_sectors_per_block = get_sb(log_sectors_per_block);
|
|
|
|
sbi->log_blocksize = get_sb(log_blocksize);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
sbi->blocksize = 1 << sbi->log_blocksize;
|
2015-11-22 04:02:18 +00:00
|
|
|
sbi->log_blocks_per_seg = get_sb(log_blocks_per_seg);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
sbi->blocks_per_seg = 1 << sbi->log_blocks_per_seg;
|
2015-11-22 04:02:18 +00:00
|
|
|
sbi->segs_per_sec = get_sb(segs_per_sec);
|
|
|
|
sbi->secs_per_zone = get_sb(secs_per_zone);
|
|
|
|
sbi->total_sections = get_sb(section_count);
|
|
|
|
sbi->total_node_count = (get_sb(segment_count_nat) / 2) *
|
|
|
|
sbi->blocks_per_seg * NAT_ENTRY_PER_BLOCK;
|
|
|
|
sbi->root_ino_num = get_sb(root_ino);
|
|
|
|
sbi->node_ino_num = get_sb(node_ino);
|
|
|
|
sbi->meta_ino_num = get_sb(meta_ino);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
sbi->cur_victim_sec = NULL_SEGNO;
|
2015-03-21 00:07:48 +00:00
|
|
|
|
2016-11-02 00:23:40 +00:00
|
|
|
for (i = 0; i < MAX_DEVICES; i++) {
|
|
|
|
if (!sb->devs[i].path[0])
|
|
|
|
break;
|
|
|
|
|
|
|
|
if (i) {
|
|
|
|
c.devices[i].path = strdup((char *)sb->devs[i].path);
|
|
|
|
if (get_device_info(i))
|
|
|
|
ASSERT(0);
|
|
|
|
} else {
|
|
|
|
ASSERT(!strcmp((char *)sb->devs[i].path,
|
|
|
|
(char *)c.devices[i].path));
|
|
|
|
}
|
|
|
|
|
|
|
|
c.devices[i].total_segments =
|
|
|
|
le32_to_cpu(sb->devs[i].total_segments);
|
|
|
|
if (i)
|
|
|
|
c.devices[i].start_blkaddr =
|
|
|
|
c.devices[i - 1].end_blkaddr + 1;
|
|
|
|
c.devices[i].end_blkaddr = c.devices[i].start_blkaddr +
|
|
|
|
c.devices[i].total_segments *
|
|
|
|
c.blks_per_seg - 1;
|
|
|
|
if (i == 0)
|
|
|
|
c.devices[i].end_blkaddr += get_sb(segment0_blkaddr);
|
|
|
|
|
|
|
|
c.ndevs = i + 1;
|
|
|
|
MSG(0, "Info: Device[%d] : %s blkaddr = %"PRIx64"--%"PRIx64"\n",
|
|
|
|
i, c.devices[i].path,
|
|
|
|
c.devices[i].start_blkaddr,
|
|
|
|
c.devices[i].end_blkaddr);
|
|
|
|
}
|
|
|
|
|
2015-11-22 04:02:18 +00:00
|
|
|
total_sectors = get_sb(block_count) << sbi->log_sectors_per_block;
|
2015-03-21 00:07:48 +00:00
|
|
|
MSG(0, "Info: total FS sectors = %"PRIu64" (%"PRIu64" MB)\n",
|
2016-02-27 02:43:08 +00:00
|
|
|
total_sectors, total_sectors >>
|
|
|
|
(20 - get_sb(log_sectorsize)));
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2019-05-14 09:33:39 +00:00
|
|
|
static int verify_checksum_chksum(struct f2fs_checkpoint *cp)
|
|
|
|
{
|
|
|
|
unsigned int chksum_offset = get_cp(checksum_offset);
|
|
|
|
unsigned int crc, cal_crc;
|
|
|
|
|
f2fs-tools: relocate chksum_offset for large_nat_bitmap feature
For large_nat_bitmap feature, there is a design flaw:
Previous:
struct f2fs_checkpoint layout:
+--------------------------+ 0x0000
| checkpoint_ver |
| ...... |
| checksum_offset |------+
| ...... | |
| sit_nat_version_bitmap[] |<-----|-------+
| ...... | | |
| checksum_value |<-----+ |
+--------------------------+ 0x1000 |
| | nat_bitmap + sit_bitmap
| payload blocks | |
| | |
+--------------------------|<-------------+
Obviously, if nat_bitmap size + sit_bitmap size is larger than
MAX_BITMAP_SIZE_IN_CKPT, nat_bitmap or sit_bitmap may overlap
checkpoint checksum's position, once checkpoint() is triggered
from kernel, nat or sit bitmap will be damaged by checksum field.
In order to fix this, let's relocate checksum_value's position
to the head of sit_nat_version_bitmap as below, then nat/sit
bitmap and chksum value update will become safe.
After:
struct f2fs_checkpoint layout:
+--------------------------+ 0x0000
| checkpoint_ver |
| ...... |
| checksum_offset |------+
| ...... | |
| sit_nat_version_bitmap[] |<-----+
| ...... |<-------------+
| | |
+--------------------------+ 0x1000 |
| | nat_bitmap + sit_bitmap
| payload blocks | |
| | |
+--------------------------|<-------------+
Related report and discussion:
https://sourceforge.net/p/linux-f2fs/mailman/message/36642346/
In addition, during writing checkpoint, if large_nat_bitmap feature is
enabled, we need to set CP_LARGE_NAT_BITMAP_FLAG flag in checkpoint.
Reported-by: Park Ju Hyung <qkrwngud825@gmail.com>
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-05-14 09:33:40 +00:00
|
|
|
if (chksum_offset < CP_MIN_CHKSUM_OFFSET ||
|
|
|
|
chksum_offset > CP_CHKSUM_OFFSET) {
|
2019-05-14 09:33:39 +00:00
|
|
|
MSG(0, "\tInvalid CP CRC offset: %u\n", chksum_offset);
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
crc = le32_to_cpu(*(__le32 *)((unsigned char *)cp + chksum_offset));
|
|
|
|
cal_crc = f2fs_checkpoint_chksum(cp);
|
|
|
|
if (cal_crc != crc) {
|
|
|
|
MSG(0, "\tInvalid CP CRC: offset:%u, crc:0x%x, calc:0x%x\n",
|
|
|
|
chksum_offset, crc, cal_crc);
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2019-06-13 08:09:03 +00:00
|
|
|
static void *get_checkpoint_version(block_t cp_addr)
|
|
|
|
{
|
|
|
|
void *cp_page;
|
|
|
|
|
|
|
|
cp_page = malloc(PAGE_SIZE);
|
|
|
|
ASSERT(cp_page);
|
|
|
|
|
|
|
|
if (dev_read_block(cp_page, cp_addr) < 0)
|
|
|
|
ASSERT(0);
|
|
|
|
|
|
|
|
if (verify_checksum_chksum((struct f2fs_checkpoint *)cp_page))
|
|
|
|
goto out;
|
|
|
|
return cp_page;
|
|
|
|
out:
|
|
|
|
free(cp_page);
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2014-08-28 00:13:23 +00:00
|
|
|
void *validate_checkpoint(struct f2fs_sb_info *sbi, block_t cp_addr,
|
|
|
|
unsigned long long *version)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
|
|
|
void *cp_page_1, *cp_page_2;
|
2015-11-22 04:02:18 +00:00
|
|
|
struct f2fs_checkpoint *cp;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
unsigned long long cur_version = 0, pre_version = 0;
|
|
|
|
|
|
|
|
/* Read the 1st cp block in this CP pack */
|
2019-06-13 08:09:03 +00:00
|
|
|
cp_page_1 = get_checkpoint_version(cp_addr);
|
|
|
|
if (!cp_page_1)
|
|
|
|
return NULL;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2015-11-22 04:02:18 +00:00
|
|
|
cp = (struct f2fs_checkpoint *)cp_page_1;
|
2018-08-03 04:09:05 +00:00
|
|
|
if (get_cp(cp_pack_total_block_count) > sbi->blocks_per_seg)
|
|
|
|
goto invalid_cp1;
|
|
|
|
|
2015-11-22 04:02:18 +00:00
|
|
|
pre_version = get_cp(checkpoint_ver);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
/* Read the 2nd cp block in this CP pack */
|
2015-11-22 04:02:18 +00:00
|
|
|
cp_addr += get_cp(cp_pack_total_block_count) - 1;
|
2019-06-13 08:09:03 +00:00
|
|
|
cp_page_2 = get_checkpoint_version(cp_addr);
|
|
|
|
if (!cp_page_2)
|
|
|
|
goto invalid_cp1;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2015-11-22 04:02:18 +00:00
|
|
|
cp = (struct f2fs_checkpoint *)cp_page_2;
|
|
|
|
cur_version = get_cp(checkpoint_ver);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
if (cur_version == pre_version) {
|
|
|
|
*version = cur_version;
|
|
|
|
free(cp_page_2);
|
|
|
|
return cp_page_1;
|
|
|
|
}
|
|
|
|
|
|
|
|
free(cp_page_2);
|
|
|
|
invalid_cp1:
|
|
|
|
free(cp_page_1);
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
int get_valid_checkpoint(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
2015-11-22 04:02:18 +00:00
|
|
|
struct f2fs_super_block *sb = F2FS_RAW_SUPER(sbi);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
void *cp1, *cp2, *cur_page;
|
|
|
|
unsigned long blk_size = sbi->blocksize;
|
2015-04-10 17:36:06 +00:00
|
|
|
unsigned long long cp1_version = 0, cp2_version = 0, version;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
unsigned long long cp_start_blk_no;
|
2017-04-27 19:12:22 +00:00
|
|
|
unsigned int cp_payload, cp_blks;
|
2014-08-28 21:55:07 +00:00
|
|
|
int ret;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2017-04-27 19:12:22 +00:00
|
|
|
cp_payload = get_sb(cp_payload);
|
|
|
|
if (cp_payload > F2FS_BLK_ALIGN(MAX_SIT_BITMAP_SIZE))
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
cp_blks = 1 + cp_payload;
|
2014-05-12 13:03:46 +00:00
|
|
|
sbi->ckpt = malloc(cp_blks * blk_size);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
if (!sbi->ckpt)
|
|
|
|
return -ENOMEM;
|
|
|
|
/*
|
|
|
|
* Finding out valid cp block involves read both
|
|
|
|
* sets( cp pack1 and cp pack 2)
|
|
|
|
*/
|
2015-11-22 04:02:18 +00:00
|
|
|
cp_start_blk_no = get_sb(cp_blkaddr);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
cp1 = validate_checkpoint(sbi, cp_start_blk_no, &cp1_version);
|
|
|
|
|
|
|
|
/* The second checkpoint pack should start at the next segment */
|
2015-11-22 04:02:18 +00:00
|
|
|
cp_start_blk_no += 1 << get_sb(log_blocks_per_seg);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
cp2 = validate_checkpoint(sbi, cp_start_blk_no, &cp2_version);
|
|
|
|
|
|
|
|
if (cp1 && cp2) {
|
2014-08-28 21:55:07 +00:00
|
|
|
if (ver_after(cp2_version, cp1_version)) {
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
cur_page = cp2;
|
2014-08-28 21:55:07 +00:00
|
|
|
sbi->cur_cp = 2;
|
2015-04-10 17:36:06 +00:00
|
|
|
version = cp2_version;
|
2014-08-28 21:55:07 +00:00
|
|
|
} else {
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
cur_page = cp1;
|
2014-08-28 21:55:07 +00:00
|
|
|
sbi->cur_cp = 1;
|
2015-04-10 17:36:06 +00:00
|
|
|
version = cp1_version;
|
2014-08-28 21:55:07 +00:00
|
|
|
}
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
} else if (cp1) {
|
|
|
|
cur_page = cp1;
|
2014-08-28 21:55:07 +00:00
|
|
|
sbi->cur_cp = 1;
|
2015-04-10 17:36:06 +00:00
|
|
|
version = cp1_version;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
} else if (cp2) {
|
|
|
|
cur_page = cp2;
|
2014-08-28 21:55:07 +00:00
|
|
|
sbi->cur_cp = 2;
|
2015-04-10 17:36:06 +00:00
|
|
|
version = cp2_version;
|
2015-12-28 11:33:22 +00:00
|
|
|
} else
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
goto fail_no_cp;
|
|
|
|
|
2015-04-29 00:51:26 +00:00
|
|
|
MSG(0, "Info: CKPT version = %llx\n", version);
|
2015-04-10 17:36:06 +00:00
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
memcpy(sbi->ckpt, cur_page, blk_size);
|
|
|
|
|
2014-05-12 13:03:46 +00:00
|
|
|
if (cp_blks > 1) {
|
2014-06-30 20:57:17 +00:00
|
|
|
unsigned int i;
|
2014-05-12 13:03:46 +00:00
|
|
|
unsigned long long cp_blk_no;
|
|
|
|
|
2015-11-22 04:02:18 +00:00
|
|
|
cp_blk_no = get_sb(cp_blkaddr);
|
2014-05-12 13:03:46 +00:00
|
|
|
if (cur_page == cp2)
|
2015-11-22 04:02:18 +00:00
|
|
|
cp_blk_no += 1 << get_sb(log_blocks_per_seg);
|
|
|
|
|
2014-05-12 13:03:46 +00:00
|
|
|
/* copy sit bitmap */
|
|
|
|
for (i = 1; i < cp_blks; i++) {
|
|
|
|
unsigned char *ckpt = (unsigned char *)sbi->ckpt;
|
2014-08-28 21:55:07 +00:00
|
|
|
ret = dev_read_block(cur_page, cp_blk_no + i);
|
|
|
|
ASSERT(ret >= 0);
|
2014-05-12 13:03:46 +00:00
|
|
|
memcpy(ckpt + i * blk_size, cur_page, blk_size);
|
|
|
|
}
|
|
|
|
}
|
2015-12-28 11:33:22 +00:00
|
|
|
if (cp1)
|
|
|
|
free(cp1);
|
|
|
|
if (cp2)
|
|
|
|
free(cp2);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
fail_no_cp:
|
|
|
|
free(sbi->ckpt);
|
2016-03-23 18:19:20 +00:00
|
|
|
sbi->ckpt = NULL;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
2020-03-24 09:12:50 +00:00
|
|
|
/*
|
|
|
|
* For a return value of 1, caller should further check for c.fix_on state
|
|
|
|
* and take appropriate action.
|
|
|
|
*/
|
|
|
|
static int f2fs_should_proceed(struct f2fs_super_block *sb, u32 flag)
|
|
|
|
{
|
|
|
|
if (!c.fix_on && (c.auto_fix || c.preen_mode)) {
|
|
|
|
if (flag & CP_FSCK_FLAG ||
|
|
|
|
flag & CP_QUOTA_NEED_FSCK_FLAG ||
|
|
|
|
(exist_qf_ino(sb) && (flag & CP_ERROR_FLAG))) {
|
|
|
|
c.fix_on = 1;
|
|
|
|
} else if (!c.preen_mode) {
|
|
|
|
print_cp_state(flag);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
int sanity_check_ckpt(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
|
|
|
unsigned int total, fsmeta;
|
2015-11-22 04:02:18 +00:00
|
|
|
struct f2fs_super_block *sb = F2FS_RAW_SUPER(sbi);
|
|
|
|
struct f2fs_checkpoint *cp = F2FS_CKPT(sbi);
|
2020-03-24 09:12:50 +00:00
|
|
|
unsigned int flag = get_cp(ckpt_flags);
|
2018-08-03 04:09:05 +00:00
|
|
|
unsigned int ovp_segments, reserved_segments;
|
|
|
|
unsigned int main_segs, blocks_per_seg;
|
|
|
|
unsigned int sit_segs, nat_segs;
|
|
|
|
unsigned int sit_bitmap_size, nat_bitmap_size;
|
|
|
|
unsigned int log_blocks_per_seg;
|
|
|
|
unsigned int segment_count_main;
|
|
|
|
unsigned int cp_pack_start_sum, cp_payload;
|
|
|
|
block_t user_block_count;
|
|
|
|
int i;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2015-11-22 04:02:18 +00:00
|
|
|
total = get_sb(segment_count);
|
|
|
|
fsmeta = get_sb(segment_count_ckpt);
|
2018-08-03 04:09:05 +00:00
|
|
|
sit_segs = get_sb(segment_count_sit);
|
|
|
|
fsmeta += sit_segs;
|
|
|
|
nat_segs = get_sb(segment_count_nat);
|
|
|
|
fsmeta += nat_segs;
|
2015-11-22 04:02:18 +00:00
|
|
|
fsmeta += get_cp(rsvd_segment_count);
|
|
|
|
fsmeta += get_sb(segment_count_ssa);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
if (fsmeta >= total)
|
|
|
|
return 1;
|
|
|
|
|
2018-08-03 04:09:05 +00:00
|
|
|
ovp_segments = get_cp(overprov_segment_count);
|
|
|
|
reserved_segments = get_cp(rsvd_segment_count);
|
|
|
|
|
|
|
|
if (fsmeta < F2FS_MIN_SEGMENT || ovp_segments == 0 ||
|
|
|
|
reserved_segments == 0) {
|
|
|
|
MSG(0, "\tWrong layout: check mkfs.f2fs version\n");
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
user_block_count = get_cp(user_block_count);
|
|
|
|
segment_count_main = get_sb(segment_count_main);
|
|
|
|
log_blocks_per_seg = get_sb(log_blocks_per_seg);
|
|
|
|
if (!user_block_count || user_block_count >=
|
|
|
|
segment_count_main << log_blocks_per_seg) {
|
2020-03-24 09:12:50 +00:00
|
|
|
ASSERT_MSG("\tWrong user_block_count(%u)\n", user_block_count);
|
|
|
|
|
|
|
|
if (!f2fs_should_proceed(sb, flag))
|
|
|
|
return 1;
|
|
|
|
if (!c.fix_on)
|
|
|
|
return 1;
|
|
|
|
|
|
|
|
if (flag & (CP_FSCK_FLAG | CP_RESIZEFS_FLAG)) {
|
|
|
|
u32 valid_user_block_cnt;
|
|
|
|
u32 seg_cnt_main = get_sb(segment_count) -
|
|
|
|
(get_sb(segment_count_ckpt) +
|
|
|
|
get_sb(segment_count_sit) +
|
|
|
|
get_sb(segment_count_nat) +
|
|
|
|
get_sb(segment_count_ssa));
|
|
|
|
|
|
|
|
/* validate segment_count_main in sb first */
|
|
|
|
if (seg_cnt_main != get_sb(segment_count_main)) {
|
|
|
|
MSG(0, "Inconsistent segment_cnt_main %u in sb\n",
|
|
|
|
segment_count_main << log_blocks_per_seg);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
valid_user_block_cnt = ((get_sb(segment_count_main) -
|
|
|
|
get_cp(overprov_segment_count)) * c.blks_per_seg);
|
|
|
|
MSG(0, "Info: Fix wrong user_block_count in CP: (%u) -> (%u)\n",
|
|
|
|
user_block_count, valid_user_block_cnt);
|
|
|
|
set_cp(user_block_count, valid_user_block_cnt);
|
|
|
|
c.bug_on = 1;
|
|
|
|
}
|
2018-08-03 04:09:05 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
main_segs = get_sb(segment_count_main);
|
|
|
|
blocks_per_seg = sbi->blocks_per_seg;
|
|
|
|
|
|
|
|
for (i = 0; i < NR_CURSEG_NODE_TYPE; i++) {
|
|
|
|
if (get_cp(cur_node_segno[i]) >= main_segs ||
|
|
|
|
get_cp(cur_node_blkoff[i]) >= blocks_per_seg)
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
for (i = 0; i < NR_CURSEG_DATA_TYPE; i++) {
|
|
|
|
if (get_cp(cur_data_segno[i]) >= main_segs ||
|
|
|
|
get_cp(cur_data_blkoff[i]) >= blocks_per_seg)
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
sit_bitmap_size = get_cp(sit_ver_bitmap_bytesize);
|
|
|
|
nat_bitmap_size = get_cp(nat_ver_bitmap_bytesize);
|
|
|
|
|
|
|
|
if (sit_bitmap_size != ((sit_segs / 2) << log_blocks_per_seg) / 8 ||
|
|
|
|
nat_bitmap_size != ((nat_segs / 2) << log_blocks_per_seg) / 8) {
|
|
|
|
MSG(0, "\tWrong bitmap size: sit(%u), nat(%u)\n",
|
|
|
|
sit_bitmap_size, nat_bitmap_size);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
cp_pack_start_sum = __start_sum_addr(sbi);
|
2018-09-28 12:26:00 +00:00
|
|
|
cp_payload = __cp_payload(sbi);
|
2018-08-03 04:09:05 +00:00
|
|
|
if (cp_pack_start_sum < cp_payload + 1 ||
|
|
|
|
cp_pack_start_sum > blocks_per_seg - 1 -
|
|
|
|
NR_CURSEG_TYPE) {
|
2018-09-28 12:26:00 +00:00
|
|
|
MSG(0, "\tWrong cp_pack_start_sum(%u) or cp_payload(%u)\n",
|
|
|
|
cp_pack_start_sum, cp_payload);
|
|
|
|
if ((get_sb(feature) & F2FS_FEATURE_SB_CHKSUM))
|
|
|
|
return 1;
|
|
|
|
set_sb(cp_payload, cp_pack_start_sum - 1);
|
|
|
|
update_superblock(sb, SB_MASK_ALL);
|
2018-08-03 04:09:05 +00:00
|
|
|
}
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2018-07-02 09:22:41 +00:00
|
|
|
pgoff_t current_nat_addr(struct f2fs_sb_info *sbi, nid_t start, int *pack)
|
2015-12-10 00:18:44 +00:00
|
|
|
{
|
|
|
|
struct f2fs_nm_info *nm_i = NM_I(sbi);
|
|
|
|
pgoff_t block_off;
|
|
|
|
pgoff_t block_addr;
|
|
|
|
int seg_off;
|
|
|
|
|
|
|
|
block_off = NAT_BLOCK_OFFSET(start);
|
|
|
|
seg_off = block_off >> sbi->log_blocks_per_seg;
|
|
|
|
|
|
|
|
block_addr = (pgoff_t)(nm_i->nat_blkaddr +
|
|
|
|
(seg_off << sbi->log_blocks_per_seg << 1) +
|
|
|
|
(block_off & ((1 << sbi->log_blocks_per_seg) -1)));
|
2018-07-02 09:22:41 +00:00
|
|
|
if (pack)
|
|
|
|
*pack = 1;
|
2015-12-10 00:18:44 +00:00
|
|
|
|
2018-07-02 09:22:41 +00:00
|
|
|
if (f2fs_test_bit(block_off, nm_i->nat_bitmap)) {
|
2015-12-10 00:18:44 +00:00
|
|
|
block_addr += sbi->blocks_per_seg;
|
2018-07-02 09:22:41 +00:00
|
|
|
if (pack)
|
|
|
|
*pack = 2;
|
|
|
|
}
|
2015-12-10 00:18:44 +00:00
|
|
|
|
|
|
|
return block_addr;
|
|
|
|
}
|
|
|
|
|
2020-06-19 09:52:12 +00:00
|
|
|
/* will not init nid_bitmap from nat */
|
|
|
|
static int f2fs_early_init_nid_bitmap(struct f2fs_sb_info *sbi)
|
2015-12-10 00:18:44 +00:00
|
|
|
{
|
|
|
|
struct f2fs_nm_info *nm_i = NM_I(sbi);
|
|
|
|
int nid_bitmap_size = (nm_i->max_nid + BITS_PER_BYTE - 1) / BITS_PER_BYTE;
|
|
|
|
struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
|
|
|
|
struct f2fs_summary_block *sum = curseg->sum_blk;
|
|
|
|
struct f2fs_journal *journal = &sum->journal;
|
|
|
|
nid_t nid;
|
|
|
|
int i;
|
|
|
|
|
2017-10-30 23:21:09 +00:00
|
|
|
if (!(c.func == SLOAD || c.func == FSCK))
|
2015-12-10 00:18:44 +00:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
nm_i->nid_bitmap = (char *)calloc(nid_bitmap_size, 1);
|
|
|
|
if (!nm_i->nid_bitmap)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
/* arbitrarily set 0 bit */
|
|
|
|
f2fs_set_bit(0, nm_i->nid_bitmap);
|
|
|
|
|
2019-01-14 01:33:08 +00:00
|
|
|
if (nats_in_cursum(journal) > NAT_JOURNAL_ENTRIES) {
|
|
|
|
MSG(0, "\tError: f2fs_init_nid_bitmap truncate n_nats(%u) to "
|
|
|
|
"NAT_JOURNAL_ENTRIES(%lu)\n",
|
|
|
|
nats_in_cursum(journal), NAT_JOURNAL_ENTRIES);
|
|
|
|
journal->n_nats = cpu_to_le16(NAT_JOURNAL_ENTRIES);
|
2019-05-24 06:09:47 +00:00
|
|
|
c.fix_on = 1;
|
2019-01-14 01:33:08 +00:00
|
|
|
}
|
|
|
|
|
2015-12-10 00:18:44 +00:00
|
|
|
for (i = 0; i < nats_in_cursum(journal); i++) {
|
|
|
|
block_t addr;
|
|
|
|
|
|
|
|
addr = le32_to_cpu(nat_in_journal(journal, i).block_addr);
|
2019-01-14 01:33:08 +00:00
|
|
|
if (!IS_VALID_BLK_ADDR(sbi, addr)) {
|
|
|
|
MSG(0, "\tError: f2fs_init_nid_bitmap: addr(%u) is invalid!!!\n", addr);
|
|
|
|
journal->n_nats = cpu_to_le16(i);
|
2019-05-24 06:09:47 +00:00
|
|
|
c.fix_on = 1;
|
2019-01-14 01:33:08 +00:00
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
2015-12-10 00:18:44 +00:00
|
|
|
nid = le32_to_cpu(nid_in_journal(journal, i));
|
2019-01-14 01:33:08 +00:00
|
|
|
if (!IS_VALID_NID(sbi, nid)) {
|
|
|
|
MSG(0, "\tError: f2fs_init_nid_bitmap: nid(%u) is invalid!!!\n", nid);
|
|
|
|
journal->n_nats = cpu_to_le16(i);
|
2019-05-24 06:09:47 +00:00
|
|
|
c.fix_on = 1;
|
2019-01-14 01:33:08 +00:00
|
|
|
continue;
|
|
|
|
}
|
2015-12-10 00:18:44 +00:00
|
|
|
if (addr != NULL_ADDR)
|
|
|
|
f2fs_set_bit(nid, nm_i->nid_bitmap);
|
|
|
|
}
|
2020-06-19 09:52:12 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* will init nid_bitmap from nat */
|
|
|
|
static int f2fs_late_init_nid_bitmap(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
|
|
|
struct f2fs_nm_info *nm_i = NM_I(sbi);
|
|
|
|
struct f2fs_nat_block *nat_block;
|
|
|
|
block_t start_blk;
|
|
|
|
nid_t nid;
|
|
|
|
|
|
|
|
if (!(c.func == SLOAD || c.func == FSCK))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
nat_block = malloc(F2FS_BLKSIZE);
|
|
|
|
if (!nat_block) {
|
|
|
|
free(nm_i->nid_bitmap);
|
|
|
|
return -ENOMEM;
|
|
|
|
}
|
|
|
|
|
|
|
|
f2fs_ra_meta_pages(sbi, 0, NAT_BLOCK_OFFSET(nm_i->max_nid),
|
|
|
|
META_NAT);
|
|
|
|
for (nid = 0; nid < nm_i->max_nid; nid++) {
|
|
|
|
if (!(nid % NAT_ENTRY_PER_BLOCK)) {
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
start_blk = current_nat_addr(sbi, nid, NULL);
|
|
|
|
ret = dev_read_block(nat_block, start_blk);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (nat_block->entries[nid % NAT_ENTRY_PER_BLOCK].block_addr)
|
|
|
|
f2fs_set_bit(nid, nm_i->nid_bitmap);
|
|
|
|
}
|
|
|
|
|
2018-04-20 18:40:18 +00:00
|
|
|
free(nat_block);
|
2015-12-10 00:18:44 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2017-02-09 02:55:43 +00:00
|
|
|
u32 update_nat_bits_flags(struct f2fs_super_block *sb,
|
|
|
|
struct f2fs_checkpoint *cp, u32 flags)
|
|
|
|
{
|
|
|
|
u_int32_t nat_bits_bytes, nat_bits_blocks;
|
|
|
|
|
|
|
|
nat_bits_bytes = get_sb(segment_count_nat) << 5;
|
|
|
|
nat_bits_blocks = F2FS_BYTES_TO_BLK((nat_bits_bytes << 1) + 8 +
|
|
|
|
F2FS_BLKSIZE - 1);
|
|
|
|
if (get_cp(cp_pack_total_block_count) <=
|
|
|
|
(1 << get_sb(log_blocks_per_seg)) - nat_bits_blocks)
|
|
|
|
flags |= CP_NAT_BITS_FLAG;
|
|
|
|
else
|
|
|
|
flags &= (~CP_NAT_BITS_FLAG);
|
|
|
|
|
|
|
|
return flags;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* should call flush_journal_entries() bfore this */
|
|
|
|
void write_nat_bits(struct f2fs_sb_info *sbi,
|
|
|
|
struct f2fs_super_block *sb, struct f2fs_checkpoint *cp, int set)
|
|
|
|
{
|
|
|
|
struct f2fs_nm_info *nm_i = NM_I(sbi);
|
|
|
|
u_int32_t nat_blocks = get_sb(segment_count_nat) <<
|
|
|
|
(get_sb(log_blocks_per_seg) - 1);
|
|
|
|
u_int32_t nat_bits_bytes = nat_blocks >> 3;
|
|
|
|
u_int32_t nat_bits_blocks = F2FS_BYTES_TO_BLK((nat_bits_bytes << 1) +
|
|
|
|
8 + F2FS_BLKSIZE - 1);
|
|
|
|
unsigned char *nat_bits, *full_nat_bits, *empty_nat_bits;
|
|
|
|
struct f2fs_nat_block *nat_block;
|
|
|
|
u_int32_t i, j;
|
|
|
|
block_t blkaddr;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
nat_bits = calloc(F2FS_BLKSIZE, nat_bits_blocks);
|
|
|
|
ASSERT(nat_bits);
|
|
|
|
|
|
|
|
nat_block = malloc(F2FS_BLKSIZE);
|
|
|
|
ASSERT(nat_block);
|
|
|
|
|
|
|
|
full_nat_bits = nat_bits + 8;
|
|
|
|
empty_nat_bits = full_nat_bits + nat_bits_bytes;
|
|
|
|
|
|
|
|
memset(full_nat_bits, 0, nat_bits_bytes);
|
|
|
|
memset(empty_nat_bits, 0, nat_bits_bytes);
|
|
|
|
|
|
|
|
for (i = 0; i < nat_blocks; i++) {
|
|
|
|
int seg_off = i >> get_sb(log_blocks_per_seg);
|
|
|
|
int valid = 0;
|
|
|
|
|
|
|
|
blkaddr = (pgoff_t)(get_sb(nat_blkaddr) +
|
|
|
|
(seg_off << get_sb(log_blocks_per_seg) << 1) +
|
|
|
|
(i & ((1 << get_sb(log_blocks_per_seg)) - 1)));
|
|
|
|
|
2018-03-23 01:57:54 +00:00
|
|
|
/*
|
|
|
|
* Should consider new nat_blocks is larger than old
|
|
|
|
* nm_i->nat_blocks, since nm_i->nat_bitmap is based on
|
|
|
|
* old one.
|
|
|
|
*/
|
|
|
|
if (i < nm_i->nat_blocks && f2fs_test_bit(i, nm_i->nat_bitmap))
|
2017-02-09 02:55:43 +00:00
|
|
|
blkaddr += (1 << get_sb(log_blocks_per_seg));
|
|
|
|
|
|
|
|
ret = dev_read_block(nat_block, blkaddr);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
|
|
|
|
for (j = 0; j < NAT_ENTRY_PER_BLOCK; j++) {
|
|
|
|
if ((i == 0 && j == 0) ||
|
|
|
|
nat_block->entries[j].block_addr != NULL_ADDR)
|
|
|
|
valid++;
|
|
|
|
}
|
|
|
|
if (valid == 0)
|
|
|
|
test_and_set_bit_le(i, empty_nat_bits);
|
|
|
|
else if (valid == NAT_ENTRY_PER_BLOCK)
|
|
|
|
test_and_set_bit_le(i, full_nat_bits);
|
|
|
|
}
|
|
|
|
*(__le64 *)nat_bits = get_cp_crc(cp);
|
|
|
|
free(nat_block);
|
|
|
|
|
|
|
|
blkaddr = get_sb(segment0_blkaddr) + (set <<
|
|
|
|
get_sb(log_blocks_per_seg)) - nat_bits_blocks;
|
|
|
|
|
|
|
|
DBG(1, "\tWriting NAT bits pages, at offset 0x%08x\n", blkaddr);
|
|
|
|
|
|
|
|
for (i = 0; i < nat_bits_blocks; i++) {
|
|
|
|
if (dev_write_block(nat_bits + i * F2FS_BLKSIZE, blkaddr + i))
|
|
|
|
ASSERT_MSG("\tError: write NAT bits to disk!!!\n");
|
|
|
|
}
|
|
|
|
MSG(0, "Info: Write valid nat_bits in checkpoint\n");
|
|
|
|
|
|
|
|
free(nat_bits);
|
|
|
|
}
|
|
|
|
|
2018-06-17 17:56:49 +00:00
|
|
|
static int check_nat_bits(struct f2fs_sb_info *sbi,
|
|
|
|
struct f2fs_super_block *sb, struct f2fs_checkpoint *cp)
|
|
|
|
{
|
|
|
|
struct f2fs_nm_info *nm_i = NM_I(sbi);
|
|
|
|
u_int32_t nat_blocks = get_sb(segment_count_nat) <<
|
|
|
|
(get_sb(log_blocks_per_seg) - 1);
|
|
|
|
u_int32_t nat_bits_bytes = nat_blocks >> 3;
|
|
|
|
u_int32_t nat_bits_blocks = F2FS_BYTES_TO_BLK((nat_bits_bytes << 1) +
|
|
|
|
8 + F2FS_BLKSIZE - 1);
|
|
|
|
unsigned char *nat_bits, *full_nat_bits, *empty_nat_bits;
|
|
|
|
struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
|
|
|
|
struct f2fs_journal *journal = &curseg->sum_blk->journal;
|
|
|
|
u_int32_t i, j;
|
|
|
|
block_t blkaddr;
|
|
|
|
int err = 0;
|
|
|
|
|
|
|
|
nat_bits = calloc(F2FS_BLKSIZE, nat_bits_blocks);
|
|
|
|
ASSERT(nat_bits);
|
|
|
|
|
|
|
|
full_nat_bits = nat_bits + 8;
|
|
|
|
empty_nat_bits = full_nat_bits + nat_bits_bytes;
|
|
|
|
|
|
|
|
blkaddr = get_sb(segment0_blkaddr) + (sbi->cur_cp <<
|
|
|
|
get_sb(log_blocks_per_seg)) - nat_bits_blocks;
|
|
|
|
|
|
|
|
for (i = 0; i < nat_bits_blocks; i++) {
|
|
|
|
if (dev_read_block(nat_bits + i * F2FS_BLKSIZE, blkaddr + i))
|
|
|
|
ASSERT_MSG("\tError: read NAT bits to disk!!!\n");
|
|
|
|
}
|
|
|
|
|
|
|
|
if (*(__le64 *)nat_bits != get_cp_crc(cp) || nats_in_cursum(journal)) {
|
|
|
|
/*
|
|
|
|
* if there is a journal, f2fs was not shutdown cleanly. Let's
|
|
|
|
* flush them with nat_bits.
|
|
|
|
*/
|
|
|
|
if (c.fix_on)
|
|
|
|
err = -1;
|
|
|
|
/* Otherwise, kernel will disable nat_bits */
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
for (i = 0; i < nat_blocks; i++) {
|
|
|
|
u_int32_t start_nid = i * NAT_ENTRY_PER_BLOCK;
|
|
|
|
u_int32_t valid = 0;
|
|
|
|
int empty = test_bit_le(i, empty_nat_bits);
|
|
|
|
int full = test_bit_le(i, full_nat_bits);
|
|
|
|
|
|
|
|
for (j = 0; j < NAT_ENTRY_PER_BLOCK; j++) {
|
|
|
|
if (f2fs_test_bit(start_nid + j, nm_i->nid_bitmap))
|
|
|
|
valid++;
|
|
|
|
}
|
|
|
|
if (valid == 0) {
|
|
|
|
if (!empty || full) {
|
|
|
|
err = -1;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
} else if (valid == NAT_ENTRY_PER_BLOCK) {
|
|
|
|
if (empty || !full) {
|
|
|
|
err = -1;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
if (empty || full) {
|
|
|
|
err = -1;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
out:
|
|
|
|
free(nat_bits);
|
|
|
|
if (!err) {
|
|
|
|
MSG(0, "Info: Checked valid nat_bits in checkpoint\n");
|
|
|
|
} else {
|
2018-11-28 23:11:28 +00:00
|
|
|
c.bug_nat_bits = 1;
|
2018-06-17 17:56:49 +00:00
|
|
|
MSG(0, "Info: Corrupted valid nat_bits in checkpoint\n");
|
|
|
|
}
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
int init_node_manager(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
2015-11-22 04:02:18 +00:00
|
|
|
struct f2fs_super_block *sb = F2FS_RAW_SUPER(sbi);
|
|
|
|
struct f2fs_checkpoint *cp = F2FS_CKPT(sbi);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
struct f2fs_nm_info *nm_i = NM_I(sbi);
|
|
|
|
unsigned char *version_bitmap;
|
2018-03-23 01:57:54 +00:00
|
|
|
unsigned int nat_segs;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2015-11-22 04:02:18 +00:00
|
|
|
nm_i->nat_blkaddr = get_sb(nat_blkaddr);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
/* segment_count_nat includes pair segment so divide to 2. */
|
2015-11-22 04:02:18 +00:00
|
|
|
nat_segs = get_sb(segment_count_nat) >> 1;
|
2018-03-23 01:57:54 +00:00
|
|
|
nm_i->nat_blocks = nat_segs << get_sb(log_blocks_per_seg);
|
|
|
|
nm_i->max_nid = NAT_ENTRY_PER_BLOCK * nm_i->nat_blocks;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
nm_i->fcnt = 0;
|
|
|
|
nm_i->nat_cnt = 0;
|
2015-11-22 04:02:18 +00:00
|
|
|
nm_i->init_scan_nid = get_cp(next_free_nid);
|
|
|
|
nm_i->next_scan_nid = get_cp(next_free_nid);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
nm_i->bitmap_size = __bitmap_size(sbi, NAT_BITMAP);
|
|
|
|
|
|
|
|
nm_i->nat_bitmap = malloc(nm_i->bitmap_size);
|
|
|
|
if (!nm_i->nat_bitmap)
|
|
|
|
return -ENOMEM;
|
|
|
|
version_bitmap = __bitmap_ptr(sbi, NAT_BITMAP);
|
|
|
|
if (!version_bitmap)
|
|
|
|
return -EFAULT;
|
|
|
|
|
|
|
|
/* copy version bitmap */
|
|
|
|
memcpy(nm_i->nat_bitmap, version_bitmap, nm_i->bitmap_size);
|
2020-06-19 09:52:12 +00:00
|
|
|
return f2fs_early_init_nid_bitmap(sbi);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
int build_node_manager(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
|
|
|
int err;
|
|
|
|
sbi->nm_info = malloc(sizeof(struct f2fs_nm_info));
|
|
|
|
if (!sbi->nm_info)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
err = init_node_manager(sbi);
|
|
|
|
if (err)
|
|
|
|
return err;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
int build_sit_info(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
2015-11-22 04:02:18 +00:00
|
|
|
struct f2fs_super_block *sb = F2FS_RAW_SUPER(sbi);
|
|
|
|
struct f2fs_checkpoint *cp = F2FS_CKPT(sbi);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
struct sit_info *sit_i;
|
2019-08-09 10:52:53 +00:00
|
|
|
unsigned int sit_segs;
|
|
|
|
int start;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
char *src_bitmap, *dst_bitmap;
|
2019-08-09 10:52:55 +00:00
|
|
|
unsigned char *bitmap;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
unsigned int bitmap_size;
|
|
|
|
|
|
|
|
sit_i = malloc(sizeof(struct sit_info));
|
2018-10-11 22:21:35 +00:00
|
|
|
if (!sit_i) {
|
|
|
|
MSG(1, "\tError: Malloc failed for build_sit_info!\n");
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
return -ENOMEM;
|
2018-10-11 22:21:35 +00:00
|
|
|
}
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
SM_I(sbi)->sit_info = sit_i;
|
|
|
|
|
|
|
|
sit_i->sentries = calloc(TOTAL_SEGS(sbi) * sizeof(struct seg_entry), 1);
|
2018-10-11 22:21:35 +00:00
|
|
|
if (!sit_i->sentries) {
|
|
|
|
MSG(1, "\tError: Calloc failed for build_sit_info!\n");
|
|
|
|
goto free_sit_info;
|
|
|
|
}
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2019-08-09 10:52:55 +00:00
|
|
|
bitmap_size = TOTAL_SEGS(sbi) * SIT_VBLOCK_MAP_SIZE;
|
|
|
|
|
2019-11-28 07:59:28 +00:00
|
|
|
if (need_fsync_data_record(sbi))
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
bitmap_size += bitmap_size;
|
|
|
|
|
2019-08-09 10:52:55 +00:00
|
|
|
sit_i->bitmap = calloc(bitmap_size, 1);
|
|
|
|
if (!sit_i->bitmap) {
|
|
|
|
MSG(1, "\tError: Calloc failed for build_sit_info!!\n");
|
|
|
|
goto free_sentries;
|
|
|
|
}
|
|
|
|
|
|
|
|
bitmap = sit_i->bitmap;
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
for (start = 0; start < TOTAL_SEGS(sbi); start++) {
|
2019-08-09 10:52:55 +00:00
|
|
|
sit_i->sentries[start].cur_valid_map = bitmap;
|
|
|
|
bitmap += SIT_VBLOCK_MAP_SIZE;
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
|
2019-11-28 07:59:28 +00:00
|
|
|
if (need_fsync_data_record(sbi)) {
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
sit_i->sentries[start].ckpt_valid_map = bitmap;
|
|
|
|
bitmap += SIT_VBLOCK_MAP_SIZE;
|
|
|
|
}
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
|
2015-11-22 04:02:18 +00:00
|
|
|
sit_segs = get_sb(segment_count_sit) >> 1;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
bitmap_size = __bitmap_size(sbi, SIT_BITMAP);
|
|
|
|
src_bitmap = __bitmap_ptr(sbi, SIT_BITMAP);
|
|
|
|
|
|
|
|
dst_bitmap = malloc(bitmap_size);
|
2018-10-11 22:21:35 +00:00
|
|
|
if (!dst_bitmap) {
|
|
|
|
MSG(1, "\tError: Malloc failed for build_sit_info!!\n");
|
|
|
|
goto free_validity_maps;
|
|
|
|
}
|
2018-11-26 13:36:00 +00:00
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
memcpy(dst_bitmap, src_bitmap, bitmap_size);
|
|
|
|
|
2015-11-22 04:02:18 +00:00
|
|
|
sit_i->sit_base_addr = get_sb(sit_blkaddr);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
sit_i->sit_blocks = sit_segs << sbi->log_blocks_per_seg;
|
2015-11-22 04:02:18 +00:00
|
|
|
sit_i->written_valid_blocks = get_cp(valid_block_count);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
sit_i->sit_bitmap = dst_bitmap;
|
|
|
|
sit_i->bitmap_size = bitmap_size;
|
|
|
|
sit_i->dirty_sentries = 0;
|
|
|
|
sit_i->sents_per_block = SIT_ENTRY_PER_BLOCK;
|
2015-11-22 04:02:18 +00:00
|
|
|
sit_i->elapsed_time = get_cp(elapsed_time);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
return 0;
|
2018-10-11 22:21:35 +00:00
|
|
|
|
|
|
|
free_validity_maps:
|
2019-08-09 10:52:55 +00:00
|
|
|
free(sit_i->bitmap);
|
|
|
|
free_sentries:
|
2018-10-11 22:21:35 +00:00
|
|
|
free(sit_i->sentries);
|
|
|
|
free_sit_info:
|
|
|
|
free(sit_i);
|
|
|
|
|
|
|
|
return -ENOMEM;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
|
2014-05-14 00:02:55 +00:00
|
|
|
void reset_curseg(struct f2fs_sb_info *sbi, int type)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
|
|
|
struct curseg_info *curseg = CURSEG_I(sbi, type);
|
2014-08-28 20:49:04 +00:00
|
|
|
struct summary_footer *sum_footer;
|
2014-08-28 21:55:07 +00:00
|
|
|
struct seg_entry *se;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2014-08-28 20:49:04 +00:00
|
|
|
sum_footer = &(curseg->sum_blk->footer);
|
|
|
|
memset(sum_footer, 0, sizeof(struct summary_footer));
|
|
|
|
if (IS_DATASEG(type))
|
|
|
|
SET_SUM_TYPE(sum_footer, SUM_TYPE_DATA);
|
|
|
|
if (IS_NODESEG(type))
|
|
|
|
SET_SUM_TYPE(sum_footer, SUM_TYPE_NODE);
|
2014-08-28 21:55:07 +00:00
|
|
|
se = get_seg_entry(sbi, curseg->segno);
|
|
|
|
se->type = type;
|
2018-06-13 01:51:49 +00:00
|
|
|
se->dirty = 1;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
|
2014-08-28 20:49:04 +00:00
|
|
|
static void read_compacted_summaries(struct f2fs_sb_info *sbi)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
|
|
|
struct curseg_info *curseg;
|
2014-08-28 20:49:04 +00:00
|
|
|
unsigned int i, j, offset;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
block_t start;
|
|
|
|
char *kaddr;
|
2014-08-28 20:49:04 +00:00
|
|
|
int ret;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
start = start_sum_block(sbi);
|
|
|
|
|
|
|
|
kaddr = (char *)malloc(PAGE_SIZE);
|
2018-11-26 13:36:00 +00:00
|
|
|
ASSERT(kaddr);
|
|
|
|
|
2014-08-28 20:49:04 +00:00
|
|
|
ret = dev_read_block(kaddr, start++);
|
|
|
|
ASSERT(ret >= 0);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
|
2016-03-11 08:43:54 +00:00
|
|
|
memcpy(&curseg->sum_blk->journal.n_nats, kaddr, SUM_JOURNAL_SIZE);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
curseg = CURSEG_I(sbi, CURSEG_COLD_DATA);
|
2016-03-11 08:43:54 +00:00
|
|
|
memcpy(&curseg->sum_blk->journal.n_sits, kaddr + SUM_JOURNAL_SIZE,
|
2014-08-28 00:13:23 +00:00
|
|
|
SUM_JOURNAL_SIZE);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
offset = 2 * SUM_JOURNAL_SIZE;
|
|
|
|
for (i = CURSEG_HOT_DATA; i <= CURSEG_COLD_DATA; i++) {
|
|
|
|
unsigned short blk_off;
|
2014-08-28 23:55:45 +00:00
|
|
|
struct curseg_info *curseg = CURSEG_I(sbi, i);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2014-05-14 00:02:55 +00:00
|
|
|
reset_curseg(sbi, i);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
if (curseg->alloc_type == SSR)
|
|
|
|
blk_off = sbi->blocks_per_seg;
|
2014-08-28 23:55:45 +00:00
|
|
|
else
|
|
|
|
blk_off = curseg->next_blkoff;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2017-04-27 18:29:22 +00:00
|
|
|
ASSERT(blk_off <= ENTRIES_IN_SUM);
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
for (j = 0; j < blk_off; j++) {
|
|
|
|
struct f2fs_summary *s;
|
|
|
|
s = (struct f2fs_summary *)(kaddr + offset);
|
|
|
|
curseg->sum_blk->entries[j] = *s;
|
|
|
|
offset += SUMMARY_SIZE;
|
2014-08-28 00:13:23 +00:00
|
|
|
if (offset + SUMMARY_SIZE <=
|
|
|
|
PAGE_CACHE_SIZE - SUM_FOOTER_SIZE)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
continue;
|
|
|
|
memset(kaddr, 0, PAGE_SIZE);
|
2014-08-28 20:49:04 +00:00
|
|
|
ret = dev_read_block(kaddr, start++);
|
|
|
|
ASSERT(ret >= 0);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
offset = 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
free(kaddr);
|
|
|
|
}
|
|
|
|
|
2014-08-28 20:49:04 +00:00
|
|
|
static void restore_node_summary(struct f2fs_sb_info *sbi,
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
unsigned int segno, struct f2fs_summary_block *sum_blk)
|
|
|
|
{
|
|
|
|
struct f2fs_node *node_blk;
|
|
|
|
struct f2fs_summary *sum_entry;
|
|
|
|
block_t addr;
|
2014-05-14 00:02:55 +00:00
|
|
|
unsigned int i;
|
2014-08-28 20:49:04 +00:00
|
|
|
int ret;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2014-08-28 20:49:04 +00:00
|
|
|
node_blk = malloc(F2FS_BLKSIZE);
|
|
|
|
ASSERT(node_blk);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
/* scan the node segment */
|
|
|
|
addr = START_BLOCK(sbi, segno);
|
|
|
|
sum_entry = &sum_blk->entries[0];
|
|
|
|
|
|
|
|
for (i = 0; i < sbi->blocks_per_seg; i++, sum_entry++) {
|
2014-08-28 20:49:04 +00:00
|
|
|
ret = dev_read_block(node_blk, addr);
|
|
|
|
ASSERT(ret >= 0);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
sum_entry->nid = node_blk->footer.nid;
|
|
|
|
addr++;
|
|
|
|
}
|
2014-08-28 20:49:04 +00:00
|
|
|
free(node_blk);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
|
2014-08-28 20:49:04 +00:00
|
|
|
static void read_normal_summaries(struct f2fs_sb_info *sbi, int type)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
2015-11-22 04:02:18 +00:00
|
|
|
struct f2fs_checkpoint *cp = F2FS_CKPT(sbi);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
struct f2fs_summary_block *sum_blk;
|
|
|
|
struct curseg_info *curseg;
|
|
|
|
unsigned int segno = 0;
|
|
|
|
block_t blk_addr = 0;
|
2014-08-28 20:49:04 +00:00
|
|
|
int ret;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
if (IS_DATASEG(type)) {
|
2015-11-22 04:02:18 +00:00
|
|
|
segno = get_cp(cur_data_segno[type]);
|
|
|
|
if (is_set_ckpt_flags(cp, CP_UMOUNT_FLAG))
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
blk_addr = sum_blk_addr(sbi, NR_CURSEG_TYPE, type);
|
|
|
|
else
|
|
|
|
blk_addr = sum_blk_addr(sbi, NR_CURSEG_DATA_TYPE, type);
|
|
|
|
} else {
|
2015-11-22 04:02:18 +00:00
|
|
|
segno = get_cp(cur_node_segno[type - CURSEG_HOT_NODE]);
|
|
|
|
if (is_set_ckpt_flags(cp, CP_UMOUNT_FLAG))
|
2014-08-28 00:13:23 +00:00
|
|
|
blk_addr = sum_blk_addr(sbi, NR_CURSEG_NODE_TYPE,
|
|
|
|
type - CURSEG_HOT_NODE);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
else
|
|
|
|
blk_addr = GET_SUM_BLKADDR(sbi, segno);
|
|
|
|
}
|
|
|
|
|
|
|
|
sum_blk = (struct f2fs_summary_block *)malloc(PAGE_SIZE);
|
2018-11-26 13:36:00 +00:00
|
|
|
ASSERT(sum_blk);
|
|
|
|
|
2014-08-28 20:49:04 +00:00
|
|
|
ret = dev_read_block(sum_blk, blk_addr);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
|
2015-11-22 04:02:18 +00:00
|
|
|
if (IS_NODESEG(type) && !is_set_ckpt_flags(cp, CP_UMOUNT_FLAG))
|
2014-08-28 20:49:04 +00:00
|
|
|
restore_node_summary(sbi, segno, sum_blk);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
curseg = CURSEG_I(sbi, type);
|
|
|
|
memcpy(curseg->sum_blk, sum_blk, PAGE_CACHE_SIZE);
|
2014-05-14 00:02:55 +00:00
|
|
|
reset_curseg(sbi, type);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
free(sum_blk);
|
|
|
|
}
|
|
|
|
|
2015-10-23 18:45:36 +00:00
|
|
|
void update_sum_entry(struct f2fs_sb_info *sbi, block_t blk_addr,
|
|
|
|
struct f2fs_summary *sum)
|
|
|
|
{
|
|
|
|
struct f2fs_summary_block *sum_blk;
|
|
|
|
u32 segno, offset;
|
|
|
|
int type, ret;
|
|
|
|
struct seg_entry *se;
|
|
|
|
|
|
|
|
segno = GET_SEGNO(sbi, blk_addr);
|
|
|
|
offset = OFFSET_IN_SEG(sbi, blk_addr);
|
|
|
|
|
|
|
|
se = get_seg_entry(sbi, segno);
|
|
|
|
|
|
|
|
sum_blk = get_sum_block(sbi, segno, &type);
|
|
|
|
memcpy(&sum_blk->entries[offset], sum, sizeof(*sum));
|
|
|
|
sum_blk->footer.entry_type = IS_NODESEG(se->type) ? SUM_TYPE_NODE :
|
|
|
|
SUM_TYPE_DATA;
|
|
|
|
|
2015-12-19 01:49:29 +00:00
|
|
|
/* write SSA all the time */
|
2017-03-17 11:08:21 +00:00
|
|
|
ret = dev_write_block(sum_blk, GET_SUM_BLKADDR(sbi, segno));
|
|
|
|
ASSERT(ret >= 0);
|
2015-12-19 01:49:29 +00:00
|
|
|
|
|
|
|
if (type == SEG_TYPE_NODE || type == SEG_TYPE_DATA ||
|
|
|
|
type == SEG_TYPE_MAX)
|
|
|
|
free(sum_blk);
|
2015-10-23 18:45:36 +00:00
|
|
|
}
|
|
|
|
|
2014-08-28 20:49:04 +00:00
|
|
|
static void restore_curseg_summaries(struct f2fs_sb_info *sbi)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
|
|
|
int type = CURSEG_HOT_DATA;
|
|
|
|
|
|
|
|
if (is_set_ckpt_flags(F2FS_CKPT(sbi), CP_COMPACT_SUM_FLAG)) {
|
2014-08-28 20:49:04 +00:00
|
|
|
read_compacted_summaries(sbi);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
type = CURSEG_HOT_NODE;
|
|
|
|
}
|
|
|
|
|
2014-08-28 20:49:04 +00:00
|
|
|
for (; type <= CURSEG_COLD_NODE; type++)
|
|
|
|
read_normal_summaries(sbi, type);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
|
2018-10-11 22:21:35 +00:00
|
|
|
static int build_curseg(struct f2fs_sb_info *sbi)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
2015-11-22 04:02:18 +00:00
|
|
|
struct f2fs_checkpoint *cp = F2FS_CKPT(sbi);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
struct curseg_info *array;
|
2014-08-28 23:55:45 +00:00
|
|
|
unsigned short blk_off;
|
|
|
|
unsigned int segno;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
int i;
|
|
|
|
|
|
|
|
array = malloc(sizeof(*array) * NR_CURSEG_TYPE);
|
2018-10-11 22:21:35 +00:00
|
|
|
if (!array) {
|
|
|
|
MSG(1, "\tError: Malloc failed for build_curseg!\n");
|
|
|
|
return -ENOMEM;
|
|
|
|
}
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
SM_I(sbi)->curseg_array = array;
|
|
|
|
|
|
|
|
for (i = 0; i < NR_CURSEG_TYPE; i++) {
|
2020-07-15 19:08:05 +00:00
|
|
|
array[i].sum_blk = calloc(PAGE_CACHE_SIZE, 1);
|
2018-10-11 22:21:35 +00:00
|
|
|
if (!array[i].sum_blk) {
|
2020-07-15 19:08:05 +00:00
|
|
|
MSG(1, "\tError: Calloc failed for build_curseg!!\n");
|
2018-10-11 22:21:35 +00:00
|
|
|
goto seg_cleanup;
|
|
|
|
}
|
|
|
|
|
2014-08-28 23:55:45 +00:00
|
|
|
if (i <= CURSEG_COLD_DATA) {
|
2015-11-22 04:02:18 +00:00
|
|
|
blk_off = get_cp(cur_data_blkoff[i]);
|
|
|
|
segno = get_cp(cur_data_segno[i]);
|
2014-08-28 23:55:45 +00:00
|
|
|
}
|
|
|
|
if (i > CURSEG_COLD_DATA) {
|
2015-11-22 04:02:18 +00:00
|
|
|
blk_off = get_cp(cur_node_blkoff[i - CURSEG_HOT_NODE]);
|
|
|
|
segno = get_cp(cur_node_segno[i - CURSEG_HOT_NODE]);
|
2014-08-28 23:55:45 +00:00
|
|
|
}
|
2017-04-27 22:45:28 +00:00
|
|
|
ASSERT(segno < TOTAL_SEGS(sbi));
|
|
|
|
ASSERT(blk_off < DEFAULT_BLOCKS_PER_SEGMENT);
|
|
|
|
|
2014-08-28 23:55:45 +00:00
|
|
|
array[i].segno = segno;
|
|
|
|
array[i].zone = GET_ZONENO_FROM_SEGNO(sbi, segno);
|
|
|
|
array[i].next_segno = NULL_SEGNO;
|
|
|
|
array[i].next_blkoff = blk_off;
|
2015-11-22 04:02:18 +00:00
|
|
|
array[i].alloc_type = cp->alloc_type[i];
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
2014-08-28 20:49:04 +00:00
|
|
|
restore_curseg_summaries(sbi);
|
2018-10-11 22:21:35 +00:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
seg_cleanup:
|
|
|
|
for(--i ; i >=0; --i)
|
|
|
|
free(array[i].sum_blk);
|
|
|
|
free(array);
|
|
|
|
|
|
|
|
return -ENOMEM;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
|
2015-11-22 03:13:42 +00:00
|
|
|
static inline void check_seg_range(struct f2fs_sb_info *sbi, unsigned int segno)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
|
|
|
unsigned int end_segno = SM_I(sbi)->segment_count - 1;
|
|
|
|
ASSERT(segno <= end_segno);
|
|
|
|
}
|
|
|
|
|
2019-08-09 10:52:56 +00:00
|
|
|
static inline block_t current_sit_addr(struct f2fs_sb_info *sbi,
|
|
|
|
unsigned int segno)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
|
|
|
struct sit_info *sit_i = SIT_I(sbi);
|
|
|
|
unsigned int offset = SIT_BLOCK_OFFSET(sit_i, segno);
|
|
|
|
block_t blk_addr = sit_i->sit_base_addr + offset;
|
|
|
|
|
|
|
|
check_seg_range(sbi, segno);
|
|
|
|
|
|
|
|
/* calculate sit block address */
|
|
|
|
if (f2fs_test_bit(offset, sit_i->sit_bitmap))
|
|
|
|
blk_addr += sit_i->sit_blocks;
|
|
|
|
|
2019-08-09 10:52:56 +00:00
|
|
|
return blk_addr;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
|
2019-08-09 10:52:56 +00:00
|
|
|
void get_current_sit_page(struct f2fs_sb_info *sbi,
|
2014-08-28 21:55:07 +00:00
|
|
|
unsigned int segno, struct f2fs_sit_block *sit_blk)
|
|
|
|
{
|
2019-08-09 10:52:56 +00:00
|
|
|
block_t blk_addr = current_sit_addr(sbi, segno);
|
2014-08-28 21:55:07 +00:00
|
|
|
|
2019-08-09 10:52:56 +00:00
|
|
|
ASSERT(dev_read_block(sit_blk, blk_addr) >= 0);
|
|
|
|
}
|
2014-08-28 21:55:07 +00:00
|
|
|
|
2019-08-09 10:52:56 +00:00
|
|
|
void rewrite_current_sit_page(struct f2fs_sb_info *sbi,
|
|
|
|
unsigned int segno, struct f2fs_sit_block *sit_blk)
|
|
|
|
{
|
|
|
|
block_t blk_addr = current_sit_addr(sbi, segno);
|
|
|
|
|
|
|
|
ASSERT(dev_write_block(sit_blk, blk_addr) >= 0);
|
2014-08-28 21:55:07 +00:00
|
|
|
}
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
void check_block_count(struct f2fs_sb_info *sbi,
|
2014-05-14 00:02:55 +00:00
|
|
|
unsigned int segno, struct f2fs_sit_entry *raw_sit)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
|
|
|
struct f2fs_sm_info *sm_info = SM_I(sbi);
|
|
|
|
unsigned int end_segno = sm_info->segment_count - 1;
|
|
|
|
int valid_blocks = 0;
|
2014-05-14 00:02:55 +00:00
|
|
|
unsigned int i;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
/* check segment usage */
|
2014-08-28 00:15:55 +00:00
|
|
|
if (GET_SIT_VBLOCKS(raw_sit) > sbi->blocks_per_seg)
|
|
|
|
ASSERT_MSG("Invalid SIT vblocks: segno=0x%x, %u",
|
|
|
|
segno, GET_SIT_VBLOCKS(raw_sit));
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
/* check boundary of a given segment number */
|
2014-08-28 00:15:55 +00:00
|
|
|
if (segno > end_segno)
|
|
|
|
ASSERT_MSG("Invalid SEGNO: 0x%x", segno);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
/* check bitmap with valid block count */
|
2014-08-28 00:15:55 +00:00
|
|
|
for (i = 0; i < SIT_VBLOCK_MAP_SIZE; i++)
|
|
|
|
valid_blocks += get_bits_in_byte(raw_sit->valid_map[i]);
|
|
|
|
|
|
|
|
if (GET_SIT_VBLOCKS(raw_sit) != valid_blocks)
|
|
|
|
ASSERT_MSG("Wrong SIT valid blocks: segno=0x%x, %u vs. %u",
|
|
|
|
segno, GET_SIT_VBLOCKS(raw_sit), valid_blocks);
|
2014-08-28 21:55:07 +00:00
|
|
|
|
|
|
|
if (GET_SIT_TYPE(raw_sit) >= NO_CHECK_TYPE)
|
|
|
|
ASSERT_MSG("Wrong SIT type: segno=0x%x, %u",
|
|
|
|
segno, GET_SIT_TYPE(raw_sit));
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
void __seg_info_from_raw_sit(struct seg_entry *se,
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
struct f2fs_sit_entry *raw_sit)
|
|
|
|
{
|
|
|
|
se->valid_blocks = GET_SIT_VBLOCKS(raw_sit);
|
|
|
|
memcpy(se->cur_valid_map, raw_sit->valid_map, SIT_VBLOCK_MAP_SIZE);
|
|
|
|
se->type = GET_SIT_TYPE(raw_sit);
|
2014-11-13 21:30:24 +00:00
|
|
|
se->orig_type = GET_SIT_TYPE(raw_sit);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
se->mtime = le64_to_cpu(raw_sit->mtime);
|
|
|
|
}
|
|
|
|
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
void seg_info_from_raw_sit(struct f2fs_sb_info *sbi, struct seg_entry *se,
|
|
|
|
struct f2fs_sit_entry *raw_sit)
|
|
|
|
{
|
|
|
|
__seg_info_from_raw_sit(se, raw_sit);
|
|
|
|
|
2019-11-28 07:59:28 +00:00
|
|
|
if (!need_fsync_data_record(sbi))
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
return;
|
|
|
|
se->ckpt_valid_blocks = se->valid_blocks;
|
|
|
|
memcpy(se->ckpt_valid_map, se->cur_valid_map, SIT_VBLOCK_MAP_SIZE);
|
|
|
|
se->ckpt_type = se->type;
|
|
|
|
}
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
struct seg_entry *get_seg_entry(struct f2fs_sb_info *sbi,
|
|
|
|
unsigned int segno)
|
|
|
|
{
|
|
|
|
struct sit_info *sit_i = SIT_I(sbi);
|
|
|
|
return &sit_i->sentries[segno];
|
|
|
|
}
|
|
|
|
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
unsigned short get_seg_vblocks(struct f2fs_sb_info *sbi, struct seg_entry *se)
|
|
|
|
{
|
2019-11-28 07:59:28 +00:00
|
|
|
if (!need_fsync_data_record(sbi))
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
return se->valid_blocks;
|
|
|
|
else
|
|
|
|
return se->ckpt_valid_blocks;
|
|
|
|
}
|
|
|
|
|
|
|
|
unsigned char *get_seg_bitmap(struct f2fs_sb_info *sbi, struct seg_entry *se)
|
|
|
|
{
|
2019-11-28 07:59:28 +00:00
|
|
|
if (!need_fsync_data_record(sbi))
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
return se->cur_valid_map;
|
|
|
|
else
|
|
|
|
return se->ckpt_valid_map;
|
|
|
|
}
|
|
|
|
|
|
|
|
unsigned char get_seg_type(struct f2fs_sb_info *sbi, struct seg_entry *se)
|
|
|
|
{
|
2019-11-28 07:59:28 +00:00
|
|
|
if (!need_fsync_data_record(sbi))
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
return se->type;
|
|
|
|
else
|
|
|
|
return se->ckpt_type;
|
|
|
|
}
|
|
|
|
|
2015-03-27 04:27:56 +00:00
|
|
|
struct f2fs_summary_block *get_sum_block(struct f2fs_sb_info *sbi,
|
|
|
|
unsigned int segno, int *ret_type)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
2015-11-22 04:02:18 +00:00
|
|
|
struct f2fs_checkpoint *cp = F2FS_CKPT(sbi);
|
2015-03-27 04:27:56 +00:00
|
|
|
struct f2fs_summary_block *sum_blk;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
struct curseg_info *curseg;
|
|
|
|
int type, ret;
|
|
|
|
u64 ssa_blk;
|
|
|
|
|
2015-03-27 04:27:56 +00:00
|
|
|
*ret_type= SEG_TYPE_MAX;
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
ssa_blk = GET_SUM_BLKADDR(sbi, segno);
|
|
|
|
for (type = 0; type < NR_CURSEG_NODE_TYPE; type++) {
|
2015-11-22 04:02:18 +00:00
|
|
|
if (segno == get_cp(cur_node_segno[type])) {
|
2014-02-05 16:09:36 +00:00
|
|
|
curseg = CURSEG_I(sbi, CURSEG_HOT_NODE + type);
|
2014-08-28 20:49:04 +00:00
|
|
|
if (!IS_SUM_NODE_SEG(curseg->sum_blk->footer)) {
|
|
|
|
ASSERT_MSG("segno [0x%x] indicates a data "
|
|
|
|
"segment, but should be node",
|
|
|
|
segno);
|
2015-03-27 04:27:56 +00:00
|
|
|
*ret_type = -SEG_TYPE_CUR_NODE;
|
|
|
|
} else {
|
|
|
|
*ret_type = SEG_TYPE_CUR_NODE;
|
2014-08-28 20:49:04 +00:00
|
|
|
}
|
2015-03-27 04:27:56 +00:00
|
|
|
return curseg->sum_blk;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
for (type = 0; type < NR_CURSEG_DATA_TYPE; type++) {
|
2015-11-22 04:02:18 +00:00
|
|
|
if (segno == get_cp(cur_data_segno[type])) {
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
curseg = CURSEG_I(sbi, type);
|
2014-08-28 20:49:04 +00:00
|
|
|
if (IS_SUM_NODE_SEG(curseg->sum_blk->footer)) {
|
|
|
|
ASSERT_MSG("segno [0x%x] indicates a node "
|
|
|
|
"segment, but should be data",
|
|
|
|
segno);
|
2015-03-27 04:27:56 +00:00
|
|
|
*ret_type = -SEG_TYPE_CUR_DATA;
|
|
|
|
} else {
|
|
|
|
*ret_type = SEG_TYPE_CUR_DATA;
|
2014-08-28 20:49:04 +00:00
|
|
|
}
|
2015-03-27 04:27:56 +00:00
|
|
|
return curseg->sum_blk;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2015-03-27 04:27:56 +00:00
|
|
|
sum_blk = calloc(BLOCK_SZ, 1);
|
|
|
|
ASSERT(sum_blk);
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
ret = dev_read_block(sum_blk, ssa_blk);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
|
|
|
|
if (IS_SUM_NODE_SEG(sum_blk->footer))
|
2015-03-27 04:27:56 +00:00
|
|
|
*ret_type = SEG_TYPE_NODE;
|
|
|
|
else if (IS_SUM_DATA_SEG(sum_blk->footer))
|
|
|
|
*ret_type = SEG_TYPE_DATA;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2015-03-27 04:27:56 +00:00
|
|
|
return sum_blk;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
|
2014-08-28 00:13:23 +00:00
|
|
|
int get_sum_entry(struct f2fs_sb_info *sbi, u32 blk_addr,
|
|
|
|
struct f2fs_summary *sum_entry)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
|
|
|
struct f2fs_summary_block *sum_blk;
|
|
|
|
u32 segno, offset;
|
2015-03-27 04:27:56 +00:00
|
|
|
int type;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
segno = GET_SEGNO(sbi, blk_addr);
|
|
|
|
offset = OFFSET_IN_SEG(sbi, blk_addr);
|
|
|
|
|
2015-03-27 04:27:56 +00:00
|
|
|
sum_blk = get_sum_block(sbi, segno, &type);
|
2014-08-28 00:13:23 +00:00
|
|
|
memcpy(sum_entry, &(sum_blk->entries[offset]),
|
|
|
|
sizeof(struct f2fs_summary));
|
2015-03-27 04:27:56 +00:00
|
|
|
if (type == SEG_TYPE_NODE || type == SEG_TYPE_DATA ||
|
|
|
|
type == SEG_TYPE_MAX)
|
|
|
|
free(sum_blk);
|
|
|
|
return type;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
|
2014-08-27 23:39:23 +00:00
|
|
|
static void get_nat_entry(struct f2fs_sb_info *sbi, nid_t nid,
|
|
|
|
struct f2fs_nat_entry *raw_nat)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
|
|
|
struct f2fs_nat_block *nat_block;
|
|
|
|
pgoff_t block_addr;
|
2016-07-15 12:01:25 +00:00
|
|
|
int entry_off;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
int ret;
|
|
|
|
|
|
|
|
if (lookup_nat_in_journal(sbi, nid, raw_nat) >= 0)
|
2014-08-27 23:39:23 +00:00
|
|
|
return;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
nat_block = (struct f2fs_nat_block *)calloc(BLOCK_SZ, 1);
|
2016-11-04 09:34:11 +00:00
|
|
|
ASSERT(nat_block);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
entry_off = nid % NAT_ENTRY_PER_BLOCK;
|
2018-07-02 09:22:41 +00:00
|
|
|
block_addr = current_nat_addr(sbi, nid, NULL);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
ret = dev_read_block(nat_block, block_addr);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
|
2014-08-27 23:39:23 +00:00
|
|
|
memcpy(raw_nat, &nat_block->entries[entry_off],
|
|
|
|
sizeof(struct f2fs_nat_entry));
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
free(nat_block);
|
|
|
|
}
|
|
|
|
|
2015-10-23 18:45:36 +00:00
|
|
|
void update_data_blkaddr(struct f2fs_sb_info *sbi, nid_t nid,
|
|
|
|
u16 ofs_in_node, block_t newaddr)
|
|
|
|
{
|
|
|
|
struct f2fs_node *node_blk = NULL;
|
|
|
|
struct node_info ni;
|
|
|
|
block_t oldaddr, startaddr, endaddr;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
node_blk = (struct f2fs_node *)calloc(BLOCK_SZ, 1);
|
2016-11-04 09:34:11 +00:00
|
|
|
ASSERT(node_blk);
|
2015-10-23 18:45:36 +00:00
|
|
|
|
|
|
|
get_node_info(sbi, nid, &ni);
|
|
|
|
|
|
|
|
/* read node_block */
|
|
|
|
ret = dev_read_block(node_blk, ni.blk_addr);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
|
|
|
|
/* check its block address */
|
|
|
|
if (node_blk->footer.nid == node_blk->footer.ino) {
|
2017-07-26 14:49:57 +00:00
|
|
|
int ofs = get_extra_isize(node_blk);
|
|
|
|
|
|
|
|
oldaddr = le32_to_cpu(node_blk->i.i_addr[ofs + ofs_in_node]);
|
|
|
|
node_blk->i.i_addr[ofs + ofs_in_node] = cpu_to_le32(newaddr);
|
2019-05-24 07:28:11 +00:00
|
|
|
ret = write_inode(node_blk, ni.blk_addr);
|
|
|
|
ASSERT(ret >= 0);
|
2015-10-23 18:45:36 +00:00
|
|
|
} else {
|
|
|
|
oldaddr = le32_to_cpu(node_blk->dn.addr[ofs_in_node]);
|
|
|
|
node_blk->dn.addr[ofs_in_node] = cpu_to_le32(newaddr);
|
2019-05-24 07:28:11 +00:00
|
|
|
ret = dev_write_block(node_blk, ni.blk_addr);
|
|
|
|
ASSERT(ret >= 0);
|
2015-10-23 18:45:36 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* check extent cache entry */
|
|
|
|
if (node_blk->footer.nid != node_blk->footer.ino) {
|
|
|
|
get_node_info(sbi, le32_to_cpu(node_blk->footer.ino), &ni);
|
|
|
|
|
|
|
|
/* read inode block */
|
|
|
|
ret = dev_read_block(node_blk, ni.blk_addr);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
}
|
|
|
|
|
|
|
|
startaddr = le32_to_cpu(node_blk->i.i_ext.blk_addr);
|
|
|
|
endaddr = startaddr + le32_to_cpu(node_blk->i.i_ext.len);
|
|
|
|
if (oldaddr >= startaddr && oldaddr < endaddr) {
|
|
|
|
node_blk->i.i_ext.len = 0;
|
|
|
|
|
|
|
|
/* update inode block */
|
2019-05-24 07:28:11 +00:00
|
|
|
ASSERT(write_inode(node_blk, ni.blk_addr) >= 0);
|
2015-10-23 18:45:36 +00:00
|
|
|
}
|
|
|
|
free(node_blk);
|
|
|
|
}
|
|
|
|
|
2015-12-10 00:18:44 +00:00
|
|
|
void update_nat_blkaddr(struct f2fs_sb_info *sbi, nid_t ino,
|
|
|
|
nid_t nid, block_t newaddr)
|
2015-10-23 18:45:36 +00:00
|
|
|
{
|
|
|
|
struct f2fs_nat_block *nat_block;
|
|
|
|
pgoff_t block_addr;
|
2016-07-15 12:01:25 +00:00
|
|
|
int entry_off;
|
2015-10-23 18:45:36 +00:00
|
|
|
int ret;
|
|
|
|
|
|
|
|
nat_block = (struct f2fs_nat_block *)calloc(BLOCK_SZ, 1);
|
2016-11-04 09:34:11 +00:00
|
|
|
ASSERT(nat_block);
|
2015-10-23 18:45:36 +00:00
|
|
|
|
|
|
|
entry_off = nid % NAT_ENTRY_PER_BLOCK;
|
2018-07-02 09:22:41 +00:00
|
|
|
block_addr = current_nat_addr(sbi, nid, NULL);
|
2015-10-23 18:45:36 +00:00
|
|
|
|
|
|
|
ret = dev_read_block(nat_block, block_addr);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
|
2015-12-10 00:18:44 +00:00
|
|
|
if (ino)
|
|
|
|
nat_block->entries[entry_off].ino = cpu_to_le32(ino);
|
2015-10-23 18:45:36 +00:00
|
|
|
nat_block->entries[entry_off].block_addr = cpu_to_le32(newaddr);
|
2018-03-06 03:39:40 +00:00
|
|
|
if (c.func == FSCK)
|
|
|
|
F2FS_FSCK(sbi)->entries[nid] = nat_block->entries[entry_off];
|
2015-10-23 18:45:36 +00:00
|
|
|
|
|
|
|
ret = dev_write_block(nat_block, block_addr);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
free(nat_block);
|
|
|
|
}
|
|
|
|
|
2014-08-27 23:39:23 +00:00
|
|
|
void get_node_info(struct f2fs_sb_info *sbi, nid_t nid, struct node_info *ni)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
|
|
|
struct f2fs_nat_entry raw_nat;
|
2018-01-22 12:03:46 +00:00
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
ni->nid = nid;
|
2019-12-10 16:42:20 +00:00
|
|
|
if (c.func == FSCK && F2FS_FSCK(sbi)->nr_nat_entries) {
|
2018-01-22 12:03:46 +00:00
|
|
|
node_info_from_raw_nat(ni, &(F2FS_FSCK(sbi)->entries[nid]));
|
2018-02-23 03:18:00 +00:00
|
|
|
if (ni->blk_addr)
|
|
|
|
return;
|
|
|
|
/* nat entry is not cached, read it */
|
2018-01-22 12:03:46 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
get_nat_entry(sbi, nid, &raw_nat);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
node_info_from_raw_nat(ni, &raw_nat);
|
|
|
|
}
|
|
|
|
|
2018-10-11 22:21:35 +00:00
|
|
|
static int build_sit_entries(struct f2fs_sb_info *sbi)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
|
|
|
struct sit_info *sit_i = SIT_I(sbi);
|
|
|
|
struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_COLD_DATA);
|
2016-03-11 08:43:54 +00:00
|
|
|
struct f2fs_journal *journal = &curseg->sum_blk->journal;
|
2018-01-29 03:22:14 +00:00
|
|
|
struct f2fs_sit_block *sit_blk;
|
2016-09-24 04:29:17 +00:00
|
|
|
struct seg_entry *se;
|
|
|
|
struct f2fs_sit_entry sit;
|
2019-08-09 10:52:57 +00:00
|
|
|
int sit_blk_cnt = SIT_BLK_CNT(sbi);
|
|
|
|
unsigned int i, segno, end;
|
|
|
|
unsigned int readed, start_blk = 0;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2018-01-29 03:22:14 +00:00
|
|
|
sit_blk = calloc(BLOCK_SZ, 1);
|
2018-10-11 22:21:35 +00:00
|
|
|
if (!sit_blk) {
|
|
|
|
MSG(1, "\tError: Calloc failed for build_sit_entries!\n");
|
|
|
|
return -ENOMEM;
|
|
|
|
}
|
|
|
|
|
2019-08-09 10:52:57 +00:00
|
|
|
do {
|
|
|
|
readed = f2fs_ra_meta_pages(sbi, start_blk, MAX_RA_BLOCKS,
|
|
|
|
META_SIT);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2019-08-09 10:52:57 +00:00
|
|
|
segno = start_blk * sit_i->sents_per_block;
|
|
|
|
end = (start_blk + readed) * sit_i->sents_per_block;
|
|
|
|
|
|
|
|
for (; segno < end && segno < TOTAL_SEGS(sbi); segno++) {
|
|
|
|
se = &sit_i->sentries[segno];
|
|
|
|
|
|
|
|
get_current_sit_page(sbi, segno, sit_blk);
|
|
|
|
sit = sit_blk->entries[SIT_ENTRY_OFFSET(sit_i, segno)];
|
|
|
|
|
|
|
|
check_block_count(sbi, segno, &sit);
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
seg_info_from_raw_sit(sbi, se, &sit);
|
2019-08-09 10:52:57 +00:00
|
|
|
}
|
|
|
|
start_blk += readed;
|
|
|
|
} while (start_blk < sit_blk_cnt);
|
2016-09-24 04:29:17 +00:00
|
|
|
|
|
|
|
|
2018-01-29 03:22:14 +00:00
|
|
|
free(sit_blk);
|
2019-05-24 08:52:41 +00:00
|
|
|
|
|
|
|
if (sits_in_cursum(journal) > SIT_JOURNAL_ENTRIES) {
|
|
|
|
MSG(0, "\tError: build_sit_entries truncate n_sits(%u) to "
|
|
|
|
"SIT_JOURNAL_ENTRIES(%lu)\n",
|
|
|
|
sits_in_cursum(journal), SIT_JOURNAL_ENTRIES);
|
|
|
|
journal->n_sits = cpu_to_le16(SIT_JOURNAL_ENTRIES);
|
|
|
|
c.fix_on = 1;
|
|
|
|
}
|
|
|
|
|
2016-09-24 04:29:17 +00:00
|
|
|
for (i = 0; i < sits_in_cursum(journal); i++) {
|
|
|
|
segno = le32_to_cpu(segno_in_journal(journal, i));
|
2019-05-24 08:52:41 +00:00
|
|
|
|
|
|
|
if (segno >= TOTAL_SEGS(sbi)) {
|
|
|
|
MSG(0, "\tError: build_sit_entries: segno(%u) is invalid!!!\n", segno);
|
|
|
|
journal->n_sits = cpu_to_le16(i);
|
|
|
|
c.fix_on = 1;
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
2016-09-24 04:29:17 +00:00
|
|
|
se = &sit_i->sentries[segno];
|
|
|
|
sit = sit_in_journal(journal, i);
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
check_block_count(sbi, segno, &sit);
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
seg_info_from_raw_sit(sbi, se, &sit);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
2018-10-11 22:21:35 +00:00
|
|
|
return 0;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
|
2020-07-01 08:23:29 +00:00
|
|
|
static int early_build_segment_manager(struct f2fs_sb_info *sbi)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
2015-11-22 04:02:18 +00:00
|
|
|
struct f2fs_super_block *sb = F2FS_RAW_SUPER(sbi);
|
|
|
|
struct f2fs_checkpoint *cp = F2FS_CKPT(sbi);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
struct f2fs_sm_info *sm_info;
|
|
|
|
|
|
|
|
sm_info = malloc(sizeof(struct f2fs_sm_info));
|
2018-10-11 22:21:35 +00:00
|
|
|
if (!sm_info) {
|
|
|
|
MSG(1, "\tError: Malloc failed for build_segment_manager!\n");
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
return -ENOMEM;
|
2018-10-11 22:21:35 +00:00
|
|
|
}
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
/* init sm info */
|
|
|
|
sbi->sm_info = sm_info;
|
2015-11-22 04:02:18 +00:00
|
|
|
sm_info->seg0_blkaddr = get_sb(segment0_blkaddr);
|
|
|
|
sm_info->main_blkaddr = get_sb(main_blkaddr);
|
|
|
|
sm_info->segment_count = get_sb(segment_count);
|
|
|
|
sm_info->reserved_segments = get_cp(rsvd_segment_count);
|
|
|
|
sm_info->ovp_segments = get_cp(overprov_segment_count);
|
|
|
|
sm_info->main_segments = get_sb(segment_count_main);
|
|
|
|
sm_info->ssa_blkaddr = get_sb(ssa_blkaddr);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2020-07-01 08:23:29 +00:00
|
|
|
if (build_sit_info(sbi) || build_curseg(sbi)) {
|
2018-10-11 22:21:35 +00:00
|
|
|
free(sm_info);
|
|
|
|
return -ENOMEM;
|
|
|
|
}
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2020-07-01 08:23:29 +00:00
|
|
|
static int late_build_segment_manager(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
|
|
|
if (sbi->seg_manager_done)
|
|
|
|
return 1; /* this function was already called */
|
|
|
|
|
|
|
|
sbi->seg_manager_done = true;
|
|
|
|
if (build_sit_entries(sbi)) {
|
|
|
|
free (sbi->sm_info);
|
|
|
|
return -ENOMEM;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2014-08-28 00:15:55 +00:00
|
|
|
void build_sit_area_bitmap(struct f2fs_sb_info *sbi)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
|
|
|
struct f2fs_fsck *fsck = F2FS_FSCK(sbi);
|
|
|
|
struct f2fs_sm_info *sm_i = SM_I(sbi);
|
2014-05-14 00:02:55 +00:00
|
|
|
unsigned int segno = 0;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
char *ptr = NULL;
|
|
|
|
u32 sum_vblocks = 0;
|
|
|
|
u32 free_segs = 0;
|
|
|
|
struct seg_entry *se;
|
|
|
|
|
|
|
|
fsck->sit_area_bitmap_sz = sm_i->main_segments * SIT_VBLOCK_MAP_SIZE;
|
|
|
|
fsck->sit_area_bitmap = calloc(1, fsck->sit_area_bitmap_sz);
|
2016-11-04 09:34:11 +00:00
|
|
|
ASSERT(fsck->sit_area_bitmap);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
ptr = fsck->sit_area_bitmap;
|
|
|
|
|
|
|
|
ASSERT(fsck->sit_area_bitmap_sz == fsck->main_area_bitmap_sz);
|
|
|
|
|
2014-08-28 00:15:55 +00:00
|
|
|
for (segno = 0; segno < TOTAL_SEGS(sbi); segno++) {
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
se = get_seg_entry(sbi, segno);
|
|
|
|
|
|
|
|
memcpy(ptr, se->cur_valid_map, SIT_VBLOCK_MAP_SIZE);
|
|
|
|
ptr += SIT_VBLOCK_MAP_SIZE;
|
|
|
|
|
|
|
|
if (se->valid_blocks == 0x0) {
|
2017-01-19 03:03:40 +00:00
|
|
|
if (le32_to_cpu(sbi->ckpt->cur_node_segno[0]) == segno ||
|
|
|
|
le32_to_cpu(sbi->ckpt->cur_data_segno[0]) == segno ||
|
|
|
|
le32_to_cpu(sbi->ckpt->cur_node_segno[1]) == segno ||
|
|
|
|
le32_to_cpu(sbi->ckpt->cur_data_segno[1]) == segno ||
|
|
|
|
le32_to_cpu(sbi->ckpt->cur_node_segno[2]) == segno ||
|
|
|
|
le32_to_cpu(sbi->ckpt->cur_data_segno[2]) == segno) {
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
continue;
|
|
|
|
} else {
|
|
|
|
free_segs++;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
sum_vblocks += se->valid_blocks;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
fsck->chk.sit_valid_blocks = sum_vblocks;
|
|
|
|
fsck->chk.sit_free_segs = free_segs;
|
|
|
|
|
2014-08-28 00:13:23 +00:00
|
|
|
DBG(1, "Blocks [0x%x : %d] Free Segs [0x%x : %d]\n\n",
|
|
|
|
sum_vblocks, sum_vblocks,
|
2013-07-30 07:39:04 +00:00
|
|
|
free_segs, free_segs);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
|
2014-08-28 21:55:07 +00:00
|
|
|
void rewrite_sit_area_bitmap(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
|
|
|
struct f2fs_fsck *fsck = F2FS_FSCK(sbi);
|
|
|
|
struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_COLD_DATA);
|
|
|
|
struct sit_info *sit_i = SIT_I(sbi);
|
2018-01-29 03:22:14 +00:00
|
|
|
struct f2fs_sit_block *sit_blk;
|
2014-08-28 21:55:07 +00:00
|
|
|
unsigned int segno = 0;
|
|
|
|
struct f2fs_summary_block *sum = curseg->sum_blk;
|
|
|
|
char *ptr = NULL;
|
|
|
|
|
2018-01-29 03:22:14 +00:00
|
|
|
sit_blk = calloc(BLOCK_SZ, 1);
|
|
|
|
ASSERT(sit_blk);
|
2014-08-28 21:55:07 +00:00
|
|
|
/* remove sit journal */
|
2016-03-11 08:43:54 +00:00
|
|
|
sum->journal.n_sits = 0;
|
2014-08-28 21:55:07 +00:00
|
|
|
|
|
|
|
ptr = fsck->main_area_bitmap;
|
|
|
|
|
|
|
|
for (segno = 0; segno < TOTAL_SEGS(sbi); segno++) {
|
|
|
|
struct f2fs_sit_entry *sit;
|
|
|
|
struct seg_entry *se;
|
|
|
|
u16 valid_blocks = 0;
|
|
|
|
u16 type;
|
|
|
|
int i;
|
|
|
|
|
2018-01-29 03:22:14 +00:00
|
|
|
get_current_sit_page(sbi, segno, sit_blk);
|
2014-08-28 21:55:07 +00:00
|
|
|
sit = &sit_blk->entries[SIT_ENTRY_OFFSET(sit_i, segno)];
|
|
|
|
memcpy(sit->valid_map, ptr, SIT_VBLOCK_MAP_SIZE);
|
|
|
|
|
|
|
|
/* update valid block count */
|
|
|
|
for (i = 0; i < SIT_VBLOCK_MAP_SIZE; i++)
|
|
|
|
valid_blocks += get_bits_in_byte(sit->valid_map[i]);
|
|
|
|
|
|
|
|
se = get_seg_entry(sbi, segno);
|
2016-11-02 10:17:18 +00:00
|
|
|
memcpy(se->cur_valid_map, ptr, SIT_VBLOCK_MAP_SIZE);
|
|
|
|
se->valid_blocks = valid_blocks;
|
2014-08-28 21:55:07 +00:00
|
|
|
type = se->type;
|
|
|
|
if (type >= NO_CHECK_TYPE) {
|
2014-11-06 04:25:49 +00:00
|
|
|
ASSERT_MSG("Invalide type and valid blocks=%x,%x",
|
|
|
|
segno, valid_blocks);
|
2014-08-28 21:55:07 +00:00
|
|
|
type = 0;
|
|
|
|
}
|
|
|
|
sit->vblocks = cpu_to_le16((type << SIT_VBLOCKS_SHIFT) |
|
|
|
|
valid_blocks);
|
|
|
|
rewrite_current_sit_page(sbi, segno, sit_blk);
|
|
|
|
|
|
|
|
ptr += SIT_VBLOCK_MAP_SIZE;
|
|
|
|
}
|
2018-01-29 03:22:14 +00:00
|
|
|
|
|
|
|
free(sit_blk);
|
2014-08-28 21:55:07 +00:00
|
|
|
}
|
|
|
|
|
2016-11-04 09:34:10 +00:00
|
|
|
static int flush_sit_journal_entries(struct f2fs_sb_info *sbi)
|
2015-10-23 18:45:36 +00:00
|
|
|
{
|
|
|
|
struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_COLD_DATA);
|
2016-03-11 08:43:54 +00:00
|
|
|
struct f2fs_journal *journal = &curseg->sum_blk->journal;
|
2015-10-23 18:45:36 +00:00
|
|
|
struct sit_info *sit_i = SIT_I(sbi);
|
2018-01-29 03:22:14 +00:00
|
|
|
struct f2fs_sit_block *sit_blk;
|
2015-10-23 18:45:36 +00:00
|
|
|
unsigned int segno;
|
|
|
|
int i;
|
|
|
|
|
2018-01-29 03:22:14 +00:00
|
|
|
sit_blk = calloc(BLOCK_SZ, 1);
|
|
|
|
ASSERT(sit_blk);
|
2016-03-11 08:43:54 +00:00
|
|
|
for (i = 0; i < sits_in_cursum(journal); i++) {
|
2015-10-23 18:45:36 +00:00
|
|
|
struct f2fs_sit_entry *sit;
|
|
|
|
struct seg_entry *se;
|
|
|
|
|
2016-03-11 08:43:54 +00:00
|
|
|
segno = segno_in_journal(journal, i);
|
2015-10-23 18:45:36 +00:00
|
|
|
se = get_seg_entry(sbi, segno);
|
|
|
|
|
2018-01-29 03:22:14 +00:00
|
|
|
get_current_sit_page(sbi, segno, sit_blk);
|
2015-10-23 18:45:36 +00:00
|
|
|
sit = &sit_blk->entries[SIT_ENTRY_OFFSET(sit_i, segno)];
|
|
|
|
|
|
|
|
memcpy(sit->valid_map, se->cur_valid_map, SIT_VBLOCK_MAP_SIZE);
|
|
|
|
sit->vblocks = cpu_to_le16((se->type << SIT_VBLOCKS_SHIFT) |
|
|
|
|
se->valid_blocks);
|
|
|
|
sit->mtime = cpu_to_le64(se->mtime);
|
|
|
|
|
|
|
|
rewrite_current_sit_page(sbi, segno, sit_blk);
|
|
|
|
}
|
2016-11-04 09:34:10 +00:00
|
|
|
|
2018-01-29 03:22:14 +00:00
|
|
|
free(sit_blk);
|
2016-03-11 08:43:54 +00:00
|
|
|
journal->n_sits = 0;
|
2016-11-04 09:34:10 +00:00
|
|
|
return i;
|
2015-10-23 18:45:36 +00:00
|
|
|
}
|
|
|
|
|
2016-11-04 09:34:10 +00:00
|
|
|
static int flush_nat_journal_entries(struct f2fs_sb_info *sbi)
|
2015-10-23 18:45:36 +00:00
|
|
|
{
|
|
|
|
struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
|
2016-03-11 08:43:54 +00:00
|
|
|
struct f2fs_journal *journal = &curseg->sum_blk->journal;
|
2015-10-23 18:45:36 +00:00
|
|
|
struct f2fs_nat_block *nat_block;
|
|
|
|
pgoff_t block_addr;
|
2016-07-15 12:01:25 +00:00
|
|
|
int entry_off;
|
2015-10-23 18:45:36 +00:00
|
|
|
nid_t nid;
|
|
|
|
int ret;
|
|
|
|
int i = 0;
|
|
|
|
|
2016-11-04 09:34:11 +00:00
|
|
|
nat_block = (struct f2fs_nat_block *)calloc(BLOCK_SZ, 1);
|
|
|
|
ASSERT(nat_block);
|
2015-10-23 18:45:36 +00:00
|
|
|
next:
|
2016-03-11 08:43:54 +00:00
|
|
|
if (i >= nats_in_cursum(journal)) {
|
2016-11-04 09:34:11 +00:00
|
|
|
free(nat_block);
|
2016-03-11 08:43:54 +00:00
|
|
|
journal->n_nats = 0;
|
2016-11-04 09:34:10 +00:00
|
|
|
return i;
|
2015-10-23 18:45:36 +00:00
|
|
|
}
|
|
|
|
|
2016-03-11 08:43:54 +00:00
|
|
|
nid = le32_to_cpu(nid_in_journal(journal, i));
|
2015-10-23 18:45:36 +00:00
|
|
|
|
|
|
|
entry_off = nid % NAT_ENTRY_PER_BLOCK;
|
2018-07-02 09:22:41 +00:00
|
|
|
block_addr = current_nat_addr(sbi, nid, NULL);
|
2015-10-23 18:45:36 +00:00
|
|
|
|
|
|
|
ret = dev_read_block(nat_block, block_addr);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
|
2016-03-11 08:43:54 +00:00
|
|
|
memcpy(&nat_block->entries[entry_off], &nat_in_journal(journal, i),
|
2015-10-23 18:45:36 +00:00
|
|
|
sizeof(struct f2fs_nat_entry));
|
|
|
|
|
|
|
|
ret = dev_write_block(nat_block, block_addr);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
i++;
|
|
|
|
goto next;
|
|
|
|
}
|
|
|
|
|
|
|
|
void flush_journal_entries(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
2016-11-04 09:34:10 +00:00
|
|
|
int n_nats = flush_nat_journal_entries(sbi);
|
|
|
|
int n_sits = flush_sit_journal_entries(sbi);
|
|
|
|
|
|
|
|
if (n_nats || n_sits)
|
2019-07-17 01:28:51 +00:00
|
|
|
write_checkpoints(sbi);
|
2015-10-23 18:45:36 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
void flush_sit_entries(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
|
|
|
struct sit_info *sit_i = SIT_I(sbi);
|
2018-01-29 03:22:14 +00:00
|
|
|
struct f2fs_sit_block *sit_blk;
|
2015-10-23 18:45:36 +00:00
|
|
|
unsigned int segno = 0;
|
|
|
|
|
2018-01-29 03:22:14 +00:00
|
|
|
sit_blk = calloc(BLOCK_SZ, 1);
|
|
|
|
ASSERT(sit_blk);
|
2015-10-23 18:45:36 +00:00
|
|
|
/* update free segments */
|
|
|
|
for (segno = 0; segno < TOTAL_SEGS(sbi); segno++) {
|
|
|
|
struct f2fs_sit_entry *sit;
|
|
|
|
struct seg_entry *se;
|
|
|
|
|
|
|
|
se = get_seg_entry(sbi, segno);
|
|
|
|
|
|
|
|
if (!se->dirty)
|
|
|
|
continue;
|
|
|
|
|
2018-01-29 03:22:14 +00:00
|
|
|
get_current_sit_page(sbi, segno, sit_blk);
|
2015-10-23 18:45:36 +00:00
|
|
|
sit = &sit_blk->entries[SIT_ENTRY_OFFSET(sit_i, segno)];
|
|
|
|
memcpy(sit->valid_map, se->cur_valid_map, SIT_VBLOCK_MAP_SIZE);
|
|
|
|
sit->vblocks = cpu_to_le16((se->type << SIT_VBLOCKS_SHIFT) |
|
|
|
|
se->valid_blocks);
|
|
|
|
rewrite_current_sit_page(sbi, segno, sit_blk);
|
|
|
|
}
|
|
|
|
|
2018-01-29 03:22:14 +00:00
|
|
|
free(sit_blk);
|
2015-10-23 18:45:36 +00:00
|
|
|
}
|
|
|
|
|
2019-05-16 12:40:43 +00:00
|
|
|
int relocate_curseg_offset(struct f2fs_sb_info *sbi, int type)
|
|
|
|
{
|
|
|
|
struct curseg_info *curseg = CURSEG_I(sbi, type);
|
|
|
|
struct seg_entry *se = get_seg_entry(sbi, curseg->segno);
|
|
|
|
unsigned int i;
|
|
|
|
|
2019-11-28 07:59:26 +00:00
|
|
|
if (c.zoned_model == F2FS_ZONED_HM)
|
|
|
|
return -EINVAL;
|
|
|
|
|
2019-05-16 12:40:43 +00:00
|
|
|
for (i = 0; i < sbi->blocks_per_seg; i++) {
|
|
|
|
if (!f2fs_test_bit(i, (const char *)se->cur_valid_map))
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (i == sbi->blocks_per_seg)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
DBG(1, "Update curseg[%d].next_blkoff %u -> %u, alloc_type %s -> SSR\n",
|
|
|
|
type, curseg->next_blkoff, i,
|
|
|
|
curseg->alloc_type == LFS ? "LFS" : "SSR");
|
|
|
|
|
|
|
|
curseg->next_blkoff = i;
|
|
|
|
curseg->alloc_type = SSR;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2019-08-09 10:52:59 +00:00
|
|
|
void set_section_type(struct f2fs_sb_info *sbi, unsigned int segno, int type)
|
|
|
|
{
|
|
|
|
unsigned int i;
|
|
|
|
|
|
|
|
if (sbi->segs_per_sec == 1)
|
|
|
|
return;
|
|
|
|
|
|
|
|
for (i = 0; i < sbi->segs_per_sec; i++) {
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
struct seg_entry *se = get_seg_entry(sbi, segno + i);
|
|
|
|
|
2019-08-09 10:52:59 +00:00
|
|
|
se->type = type;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
fsck: Check write pointer consistency of open zones
On sudden f2fs shutdown, write pointers of zoned block devices can go
further but f2fs meta data keeps current segments at positions before the
write operations. After remounting the f2fs, this inconsistency causes
write operations not at write pointers and "Unaligned write command"
error is reported.
To avoid the error, have f2fs.fsck check consistency of write pointers
of open zones that current segments point to. Compare each current
segment's position and the write pointer position of the open zone. If
inconsistency is found and 'fix_on' flag is set, assign a new zone to the
current segment and check the newly assigned zone has write pointer at
the zone start. Leave the original zone as is to keep data recorded in
it.
To care about fsync data, refer each seg_entry's ckpt_valid_map to get
the last valid block in the zone. If the last valid block is beyond the
current segments position, fsync data exits in the zone. In case fsync
data exists, do not assign a new zone to the current segment not to lose
the fsync data. It is expected that the kernel replay the fsync data and
fix the write pointer inconsistency at mount time.
Also check consistency between write pointer of the zone the current
segment points to with valid block maps of the zone. If the last valid
block is beyond the write pointer position, report to indicate a bug. If
'fix_on' flag is set, assign a new zone to the current segment.
When inconsistencies are found, turn on 'bug_on' flag in fsck_verify() to
ask users to fix them or not. When inconsistencies get fixed, turn on
'force' flag in fsck_verify() to enforce fixes in following checks.
This check and fix is done twice. The first is done at the beginning of
do_fsck() function so that other fixes can reflect the current segment
modification. The second is done in fsck_verify() to reflect updated meta
data by other fixes.
Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-11-28 07:59:29 +00:00
|
|
|
#ifdef HAVE_LINUX_BLKZONED_H
|
|
|
|
|
|
|
|
static bool write_pointer_at_zone_start(struct f2fs_sb_info *sbi,
|
|
|
|
unsigned int zone_segno)
|
|
|
|
{
|
|
|
|
u_int64_t sector;
|
|
|
|
struct blk_zone blkz;
|
|
|
|
block_t block = START_BLOCK(sbi, zone_segno);
|
|
|
|
int log_sectors_per_block = sbi->log_blocksize - SECTOR_SHIFT;
|
|
|
|
int ret, j;
|
|
|
|
|
|
|
|
if (c.zoned_model != F2FS_ZONED_HM)
|
|
|
|
return true;
|
|
|
|
|
|
|
|
for (j = 0; j < MAX_DEVICES; j++) {
|
|
|
|
if (!c.devices[j].path)
|
|
|
|
break;
|
|
|
|
if (c.devices[j].start_blkaddr <= block &&
|
|
|
|
block <= c.devices[j].end_blkaddr)
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (j >= MAX_DEVICES)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
sector = (block - c.devices[j].start_blkaddr) << log_sectors_per_block;
|
|
|
|
ret = f2fs_report_zone(j, sector, &blkz);
|
|
|
|
if (ret)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
if (blk_zone_type(&blkz) != BLK_ZONE_TYPE_SEQWRITE_REQ)
|
|
|
|
return true;
|
|
|
|
|
|
|
|
return blk_zone_sector(&blkz) == blk_zone_wp_sector(&blkz);
|
|
|
|
}
|
|
|
|
|
|
|
|
#else
|
|
|
|
|
2019-12-19 01:26:03 +00:00
|
|
|
static bool write_pointer_at_zone_start(struct f2fs_sb_info *UNUSED(sbi),
|
|
|
|
unsigned int UNUSED(zone_segno))
|
fsck: Check write pointer consistency of open zones
On sudden f2fs shutdown, write pointers of zoned block devices can go
further but f2fs meta data keeps current segments at positions before the
write operations. After remounting the f2fs, this inconsistency causes
write operations not at write pointers and "Unaligned write command"
error is reported.
To avoid the error, have f2fs.fsck check consistency of write pointers
of open zones that current segments point to. Compare each current
segment's position and the write pointer position of the open zone. If
inconsistency is found and 'fix_on' flag is set, assign a new zone to the
current segment and check the newly assigned zone has write pointer at
the zone start. Leave the original zone as is to keep data recorded in
it.
To care about fsync data, refer each seg_entry's ckpt_valid_map to get
the last valid block in the zone. If the last valid block is beyond the
current segments position, fsync data exits in the zone. In case fsync
data exists, do not assign a new zone to the current segment not to lose
the fsync data. It is expected that the kernel replay the fsync data and
fix the write pointer inconsistency at mount time.
Also check consistency between write pointer of the zone the current
segment points to with valid block maps of the zone. If the last valid
block is beyond the write pointer position, report to indicate a bug. If
'fix_on' flag is set, assign a new zone to the current segment.
When inconsistencies are found, turn on 'bug_on' flag in fsck_verify() to
ask users to fix them or not. When inconsistencies get fixed, turn on
'force' flag in fsck_verify() to enforce fixes in following checks.
This check and fix is done twice. The first is done at the beginning of
do_fsck() function so that other fixes can reflect the current segment
modification. The second is done in fsck_verify() to reflect updated meta
data by other fixes.
Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-11-28 07:59:29 +00:00
|
|
|
{
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
#endif
|
|
|
|
|
2019-11-28 07:59:26 +00:00
|
|
|
int find_next_free_block(struct f2fs_sb_info *sbi, u64 *to, int left,
|
|
|
|
int want_type, bool new_sec)
|
2015-12-10 04:30:41 +00:00
|
|
|
{
|
2016-11-03 01:00:48 +00:00
|
|
|
struct f2fs_super_block *sb = F2FS_RAW_SUPER(sbi);
|
2015-12-10 04:30:41 +00:00
|
|
|
struct seg_entry *se;
|
|
|
|
u32 segno;
|
2018-01-04 01:51:03 +00:00
|
|
|
u32 offset;
|
2016-11-02 12:55:56 +00:00
|
|
|
int not_enough = 0;
|
2016-11-03 01:00:48 +00:00
|
|
|
u64 end_blkaddr = (get_sb(segment_count_main) <<
|
|
|
|
get_sb(log_blocks_per_seg)) + get_sb(main_blkaddr);
|
2015-12-10 04:30:41 +00:00
|
|
|
|
2018-05-09 02:48:33 +00:00
|
|
|
if (*to > 0)
|
|
|
|
*to -= left;
|
2015-12-17 02:31:38 +00:00
|
|
|
if (get_free_segments(sbi) <= SM_I(sbi)->reserved_segments + 1)
|
2016-11-02 12:55:56 +00:00
|
|
|
not_enough = 1;
|
2015-12-17 02:31:38 +00:00
|
|
|
|
2016-11-03 01:00:48 +00:00
|
|
|
while (*to >= SM_I(sbi)->main_blkaddr && *to < end_blkaddr) {
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
unsigned short vblocks;
|
|
|
|
unsigned char *bitmap;
|
|
|
|
unsigned char type;
|
|
|
|
|
2015-12-10 04:30:41 +00:00
|
|
|
segno = GET_SEGNO(sbi, *to);
|
|
|
|
offset = OFFSET_IN_SEG(sbi, *to);
|
|
|
|
|
|
|
|
se = get_seg_entry(sbi, segno);
|
|
|
|
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
vblocks = get_seg_vblocks(sbi, se);
|
|
|
|
bitmap = get_seg_bitmap(sbi, se);
|
|
|
|
type = get_seg_type(sbi, se);
|
|
|
|
|
|
|
|
if (vblocks == sbi->blocks_per_seg ||
|
2018-06-04 08:14:36 +00:00
|
|
|
IS_CUR_SEGNO(sbi, segno)) {
|
2015-12-17 23:37:24 +00:00
|
|
|
*to = left ? START_BLOCK(sbi, segno) - 1:
|
|
|
|
START_BLOCK(sbi, segno + 1);
|
|
|
|
continue;
|
|
|
|
}
|
2016-11-02 12:55:56 +00:00
|
|
|
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
if (vblocks == 0 && not_enough) {
|
2016-11-02 12:55:56 +00:00
|
|
|
*to = left ? START_BLOCK(sbi, segno) - 1:
|
|
|
|
START_BLOCK(sbi, segno + 1);
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
if (vblocks == 0 && !(segno % sbi->segs_per_sec)) {
|
2015-12-10 04:30:41 +00:00
|
|
|
struct seg_entry *se2;
|
2016-01-13 19:51:33 +00:00
|
|
|
unsigned int i;
|
2015-12-10 04:30:41 +00:00
|
|
|
|
2016-10-24 02:16:04 +00:00
|
|
|
for (i = 1; i < sbi->segs_per_sec; i++) {
|
2015-12-10 04:30:41 +00:00
|
|
|
se2 = get_seg_entry(sbi, segno + i);
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
if (get_seg_vblocks(sbi, se2))
|
2015-12-10 04:30:41 +00:00
|
|
|
break;
|
|
|
|
}
|
2019-08-09 10:52:59 +00:00
|
|
|
|
fsck: Check write pointer consistency of open zones
On sudden f2fs shutdown, write pointers of zoned block devices can go
further but f2fs meta data keeps current segments at positions before the
write operations. After remounting the f2fs, this inconsistency causes
write operations not at write pointers and "Unaligned write command"
error is reported.
To avoid the error, have f2fs.fsck check consistency of write pointers
of open zones that current segments point to. Compare each current
segment's position and the write pointer position of the open zone. If
inconsistency is found and 'fix_on' flag is set, assign a new zone to the
current segment and check the newly assigned zone has write pointer at
the zone start. Leave the original zone as is to keep data recorded in
it.
To care about fsync data, refer each seg_entry's ckpt_valid_map to get
the last valid block in the zone. If the last valid block is beyond the
current segments position, fsync data exits in the zone. In case fsync
data exists, do not assign a new zone to the current segment not to lose
the fsync data. It is expected that the kernel replay the fsync data and
fix the write pointer inconsistency at mount time.
Also check consistency between write pointer of the zone the current
segment points to with valid block maps of the zone. If the last valid
block is beyond the write pointer position, report to indicate a bug. If
'fix_on' flag is set, assign a new zone to the current segment.
When inconsistencies are found, turn on 'bug_on' flag in fsck_verify() to
ask users to fix them or not. When inconsistencies get fixed, turn on
'force' flag in fsck_verify() to enforce fixes in following checks.
This check and fix is done twice. The first is done at the beginning of
do_fsck() function so that other fixes can reflect the current segment
modification. The second is done in fsck_verify() to reflect updated meta
data by other fixes.
Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-11-28 07:59:29 +00:00
|
|
|
if (i == sbi->segs_per_sec &&
|
|
|
|
write_pointer_at_zone_start(sbi, segno)) {
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
set_section_type(sbi, segno, want_type);
|
2015-12-10 04:30:41 +00:00
|
|
|
return 0;
|
2019-08-09 10:52:59 +00:00
|
|
|
}
|
2015-12-10 04:30:41 +00:00
|
|
|
}
|
|
|
|
|
2019-11-28 07:59:26 +00:00
|
|
|
if (type == want_type && !new_sec &&
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
!f2fs_test_bit(offset, (const char *)bitmap))
|
2015-12-10 04:30:41 +00:00
|
|
|
return 0;
|
2015-12-17 23:37:24 +00:00
|
|
|
|
2015-12-10 04:30:41 +00:00
|
|
|
*to = left ? *to - 1: *to + 1;
|
|
|
|
}
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
2019-11-28 07:59:27 +00:00
|
|
|
static void move_one_curseg_info(struct f2fs_sb_info *sbi, u64 from, int left,
|
|
|
|
int i)
|
2015-12-10 04:30:41 +00:00
|
|
|
{
|
2019-11-28 07:59:27 +00:00
|
|
|
struct curseg_info *curseg = CURSEG_I(sbi, i);
|
|
|
|
struct f2fs_summary_block buf;
|
|
|
|
u32 old_segno;
|
|
|
|
u64 ssa_blk, to;
|
|
|
|
int ret;
|
2015-12-10 04:30:41 +00:00
|
|
|
|
2019-11-28 07:59:27 +00:00
|
|
|
/* update original SSA too */
|
|
|
|
ssa_blk = GET_SUM_BLKADDR(sbi, curseg->segno);
|
|
|
|
ret = dev_write_block(curseg->sum_blk, ssa_blk);
|
|
|
|
ASSERT(ret >= 0);
|
2015-12-10 04:30:41 +00:00
|
|
|
|
2019-11-28 07:59:27 +00:00
|
|
|
to = from;
|
|
|
|
ret = find_next_free_block(sbi, &to, left, i,
|
|
|
|
c.zoned_model == F2FS_ZONED_HM);
|
|
|
|
ASSERT(ret == 0);
|
2015-12-10 04:30:41 +00:00
|
|
|
|
2019-11-28 07:59:27 +00:00
|
|
|
old_segno = curseg->segno;
|
|
|
|
curseg->segno = GET_SEGNO(sbi, to);
|
|
|
|
curseg->next_blkoff = OFFSET_IN_SEG(sbi, to);
|
|
|
|
curseg->alloc_type = c.zoned_model == F2FS_ZONED_HM ? LFS : SSR;
|
2015-12-10 04:30:41 +00:00
|
|
|
|
2019-11-28 07:59:27 +00:00
|
|
|
/* update new segno */
|
|
|
|
ssa_blk = GET_SUM_BLKADDR(sbi, curseg->segno);
|
|
|
|
ret = dev_read_block(&buf, ssa_blk);
|
|
|
|
ASSERT(ret >= 0);
|
2015-12-10 04:30:41 +00:00
|
|
|
|
2019-11-28 07:59:27 +00:00
|
|
|
memcpy(curseg->sum_blk, &buf, SUM_ENTRIES_SIZE);
|
2015-12-10 04:30:41 +00:00
|
|
|
|
2019-11-28 07:59:27 +00:00
|
|
|
/* update se->types */
|
|
|
|
reset_curseg(sbi, i);
|
2015-12-20 07:17:11 +00:00
|
|
|
|
2019-11-28 07:59:27 +00:00
|
|
|
FIX_MSG("Move curseg[%d] %x -> %x after %"PRIx64"\n",
|
|
|
|
i, old_segno, curseg->segno, from);
|
|
|
|
}
|
2015-12-10 04:30:41 +00:00
|
|
|
|
2019-11-28 07:59:27 +00:00
|
|
|
void move_curseg_info(struct f2fs_sb_info *sbi, u64 from, int left)
|
|
|
|
{
|
|
|
|
int i;
|
|
|
|
|
|
|
|
/* update summary blocks having nullified journal entries */
|
|
|
|
for (i = 0; i < NO_CHECK_TYPE; i++)
|
|
|
|
move_one_curseg_info(sbi, from, left, i);
|
2015-12-10 04:30:41 +00:00
|
|
|
}
|
|
|
|
|
2019-05-16 12:40:43 +00:00
|
|
|
void update_curseg_info(struct f2fs_sb_info *sbi, int type)
|
|
|
|
{
|
|
|
|
if (!relocate_curseg_offset(sbi, type))
|
|
|
|
return;
|
2019-11-28 07:59:27 +00:00
|
|
|
move_one_curseg_info(sbi, SM_I(sbi)->main_blkaddr, 0, type);
|
2019-05-16 12:40:43 +00:00
|
|
|
}
|
|
|
|
|
2015-12-17 01:43:34 +00:00
|
|
|
void zero_journal_entries(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for (i = 0; i < NO_CHECK_TYPE; i++)
|
2016-03-11 08:43:54 +00:00
|
|
|
CURSEG_I(sbi, i)->sum_blk->journal.n_nats = 0;
|
2015-12-17 01:43:34 +00:00
|
|
|
}
|
|
|
|
|
2015-12-10 04:30:41 +00:00
|
|
|
void write_curseg_info(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
|
|
|
struct f2fs_checkpoint *cp = F2FS_CKPT(sbi);
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for (i = 0; i < NO_CHECK_TYPE; i++) {
|
|
|
|
cp->alloc_type[i] = CURSEG_I(sbi, i)->alloc_type;
|
|
|
|
if (i < CURSEG_HOT_NODE) {
|
|
|
|
set_cp(cur_data_segno[i], CURSEG_I(sbi, i)->segno);
|
|
|
|
set_cp(cur_data_blkoff[i],
|
|
|
|
CURSEG_I(sbi, i)->next_blkoff);
|
|
|
|
} else {
|
|
|
|
int n = i - CURSEG_HOT_NODE;
|
|
|
|
|
|
|
|
set_cp(cur_node_segno[n], CURSEG_I(sbi, i)->segno);
|
|
|
|
set_cp(cur_node_blkoff[n],
|
|
|
|
CURSEG_I(sbi, i)->next_blkoff);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2014-08-28 00:13:23 +00:00
|
|
|
int lookup_nat_in_journal(struct f2fs_sb_info *sbi, u32 nid,
|
|
|
|
struct f2fs_nat_entry *raw_nat)
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
{
|
|
|
|
struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
|
2016-03-11 08:43:54 +00:00
|
|
|
struct f2fs_journal *journal = &curseg->sum_blk->journal;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
int i = 0;
|
|
|
|
|
2016-03-11 08:43:54 +00:00
|
|
|
for (i = 0; i < nats_in_cursum(journal); i++) {
|
|
|
|
if (le32_to_cpu(nid_in_journal(journal, i)) == nid) {
|
|
|
|
memcpy(raw_nat, &nat_in_journal(journal, i),
|
2014-08-28 00:13:23 +00:00
|
|
|
sizeof(struct f2fs_nat_entry));
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
DBG(3, "==> Found nid [0x%x] in nat cache\n", nid);
|
|
|
|
return i;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
2014-08-28 21:55:07 +00:00
|
|
|
void nullify_nat_entry(struct f2fs_sb_info *sbi, u32 nid)
|
|
|
|
{
|
|
|
|
struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
|
2016-03-11 08:43:54 +00:00
|
|
|
struct f2fs_journal *journal = &curseg->sum_blk->journal;
|
2014-08-28 21:55:07 +00:00
|
|
|
struct f2fs_nat_block *nat_block;
|
|
|
|
pgoff_t block_addr;
|
2016-07-15 12:01:25 +00:00
|
|
|
int entry_off;
|
2014-08-28 21:55:07 +00:00
|
|
|
int ret;
|
|
|
|
int i = 0;
|
|
|
|
|
|
|
|
/* check in journal */
|
2016-03-11 08:43:54 +00:00
|
|
|
for (i = 0; i < nats_in_cursum(journal); i++) {
|
|
|
|
if (le32_to_cpu(nid_in_journal(journal, i)) == nid) {
|
|
|
|
memset(&nat_in_journal(journal, i), 0,
|
2014-08-28 21:55:07 +00:00
|
|
|
sizeof(struct f2fs_nat_entry));
|
2017-11-02 03:56:12 +00:00
|
|
|
FIX_MSG("Remove nid [0x%x] in nat journal", nid);
|
2014-08-28 21:55:07 +00:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
nat_block = (struct f2fs_nat_block *)calloc(BLOCK_SZ, 1);
|
2016-11-04 09:34:11 +00:00
|
|
|
ASSERT(nat_block);
|
2014-08-28 21:55:07 +00:00
|
|
|
|
|
|
|
entry_off = nid % NAT_ENTRY_PER_BLOCK;
|
2018-07-02 09:22:41 +00:00
|
|
|
block_addr = current_nat_addr(sbi, nid, NULL);
|
2014-08-28 21:55:07 +00:00
|
|
|
|
|
|
|
ret = dev_read_block(nat_block, block_addr);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
|
2017-11-02 03:56:05 +00:00
|
|
|
if (nid == F2FS_NODE_INO(sbi) || nid == F2FS_META_INO(sbi)) {
|
|
|
|
FIX_MSG("nid [0x%x] block_addr= 0x%x -> 0x1", nid,
|
|
|
|
le32_to_cpu(nat_block->entries[entry_off].block_addr));
|
|
|
|
nat_block->entries[entry_off].block_addr = cpu_to_le32(0x1);
|
|
|
|
} else {
|
|
|
|
memset(&nat_block->entries[entry_off], 0,
|
2014-08-28 21:55:07 +00:00
|
|
|
sizeof(struct f2fs_nat_entry));
|
2017-11-02 03:56:12 +00:00
|
|
|
FIX_MSG("Remove nid [0x%x] in NAT", nid);
|
2017-11-02 03:56:05 +00:00
|
|
|
}
|
2014-08-28 21:55:07 +00:00
|
|
|
|
|
|
|
ret = dev_write_block(nat_block, block_addr);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
free(nat_block);
|
|
|
|
}
|
|
|
|
|
2019-07-17 01:28:51 +00:00
|
|
|
void duplicate_checkpoint(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
|
|
|
struct f2fs_super_block *sb = F2FS_RAW_SUPER(sbi);
|
|
|
|
unsigned long long dst, src;
|
|
|
|
void *buf;
|
|
|
|
unsigned int seg_size = 1 << get_sb(log_blocks_per_seg);
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
if (sbi->cp_backuped)
|
|
|
|
return;
|
|
|
|
|
|
|
|
buf = malloc(F2FS_BLKSIZE * seg_size);
|
|
|
|
ASSERT(buf);
|
|
|
|
|
|
|
|
if (sbi->cur_cp == 1) {
|
|
|
|
src = get_sb(cp_blkaddr);
|
|
|
|
dst = src + seg_size;
|
|
|
|
} else {
|
|
|
|
dst = get_sb(cp_blkaddr);
|
|
|
|
src = dst + seg_size;
|
|
|
|
}
|
|
|
|
|
|
|
|
ret = dev_read(buf, src << F2FS_BLKSIZE_BITS,
|
|
|
|
seg_size << F2FS_BLKSIZE_BITS);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
|
|
|
|
ret = dev_write(buf, dst << F2FS_BLKSIZE_BITS,
|
|
|
|
seg_size << F2FS_BLKSIZE_BITS);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
|
|
|
|
free(buf);
|
|
|
|
|
|
|
|
ret = f2fs_fsync_device();
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
|
|
|
|
sbi->cp_backuped = 1;
|
|
|
|
|
|
|
|
MSG(0, "Info: Duplicate valid checkpoint to mirror position "
|
|
|
|
"%llu -> %llu\n", src, dst);
|
|
|
|
}
|
|
|
|
|
2015-10-23 18:45:36 +00:00
|
|
|
void write_checkpoint(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
|
|
|
struct f2fs_checkpoint *cp = F2FS_CKPT(sbi);
|
|
|
|
struct f2fs_super_block *sb = F2FS_RAW_SUPER(sbi);
|
|
|
|
block_t orphan_blks = 0;
|
|
|
|
unsigned long long cp_blk_no;
|
|
|
|
u32 flags = CP_UMOUNT_FLAG;
|
|
|
|
int i, ret;
|
|
|
|
u_int32_t crc = 0;
|
|
|
|
|
|
|
|
if (is_set_ckpt_flags(cp, CP_ORPHAN_PRESENT_FLAG)) {
|
|
|
|
orphan_blks = __start_sum_addr(sbi) - 1;
|
|
|
|
flags |= CP_ORPHAN_PRESENT_FLAG;
|
|
|
|
}
|
2019-01-25 04:38:10 +00:00
|
|
|
if (is_set_ckpt_flags(cp, CP_TRIMMED_FLAG))
|
|
|
|
flags |= CP_TRIMMED_FLAG;
|
2018-09-20 23:16:07 +00:00
|
|
|
if (is_set_ckpt_flags(cp, CP_DISABLED_FLAG))
|
|
|
|
flags |= CP_DISABLED_FLAG;
|
f2fs-tools: relocate chksum_offset for large_nat_bitmap feature
For large_nat_bitmap feature, there is a design flaw:
Previous:
struct f2fs_checkpoint layout:
+--------------------------+ 0x0000
| checkpoint_ver |
| ...... |
| checksum_offset |------+
| ...... | |
| sit_nat_version_bitmap[] |<-----|-------+
| ...... | | |
| checksum_value |<-----+ |
+--------------------------+ 0x1000 |
| | nat_bitmap + sit_bitmap
| payload blocks | |
| | |
+--------------------------|<-------------+
Obviously, if nat_bitmap size + sit_bitmap size is larger than
MAX_BITMAP_SIZE_IN_CKPT, nat_bitmap or sit_bitmap may overlap
checkpoint checksum's position, once checkpoint() is triggered
from kernel, nat or sit bitmap will be damaged by checksum field.
In order to fix this, let's relocate checksum_value's position
to the head of sit_nat_version_bitmap as below, then nat/sit
bitmap and chksum value update will become safe.
After:
struct f2fs_checkpoint layout:
+--------------------------+ 0x0000
| checkpoint_ver |
| ...... |
| checksum_offset |------+
| ...... | |
| sit_nat_version_bitmap[] |<-----+
| ...... |<-------------+
| | |
+--------------------------+ 0x1000 |
| | nat_bitmap + sit_bitmap
| payload blocks | |
| | |
+--------------------------|<-------------+
Related report and discussion:
https://sourceforge.net/p/linux-f2fs/mailman/message/36642346/
In addition, during writing checkpoint, if large_nat_bitmap feature is
enabled, we need to set CP_LARGE_NAT_BITMAP_FLAG flag in checkpoint.
Reported-by: Park Ju Hyung <qkrwngud825@gmail.com>
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-05-14 09:33:40 +00:00
|
|
|
if (is_set_ckpt_flags(cp, CP_LARGE_NAT_BITMAP_FLAG)) {
|
|
|
|
flags |= CP_LARGE_NAT_BITMAP_FLAG;
|
|
|
|
set_cp(checksum_offset, CP_MIN_CHKSUM_OFFSET);
|
|
|
|
} else {
|
|
|
|
set_cp(checksum_offset, CP_CHKSUM_OFFSET);
|
|
|
|
}
|
2015-10-23 18:45:36 +00:00
|
|
|
|
2015-12-17 02:31:38 +00:00
|
|
|
set_cp(free_segment_count, get_free_segments(sbi));
|
2019-08-05 09:44:06 +00:00
|
|
|
if (c.func == FSCK) {
|
|
|
|
struct f2fs_fsck *fsck = F2FS_FSCK(sbi);
|
|
|
|
|
|
|
|
set_cp(valid_block_count, fsck->chk.valid_blk_cnt);
|
|
|
|
set_cp(valid_node_count, fsck->chk.valid_node_cnt);
|
|
|
|
set_cp(valid_inode_count, fsck->chk.valid_inode_cnt);
|
|
|
|
} else {
|
|
|
|
set_cp(valid_block_count, sbi->total_valid_block_count);
|
|
|
|
set_cp(valid_node_count, sbi->total_valid_node_count);
|
|
|
|
set_cp(valid_inode_count, sbi->total_valid_inode_count);
|
|
|
|
}
|
2015-10-23 18:45:36 +00:00
|
|
|
set_cp(cp_pack_total_block_count, 8 + orphan_blks + get_sb(cp_payload));
|
|
|
|
|
2017-02-09 02:55:43 +00:00
|
|
|
flags = update_nat_bits_flags(sb, cp, flags);
|
|
|
|
set_cp(ckpt_flags, flags);
|
|
|
|
|
2019-05-14 09:33:39 +00:00
|
|
|
crc = f2fs_checkpoint_chksum(cp);
|
|
|
|
*((__le32 *)((unsigned char *)cp + get_cp(checksum_offset))) =
|
|
|
|
cpu_to_le32(crc);
|
2015-10-23 18:45:36 +00:00
|
|
|
|
|
|
|
cp_blk_no = get_sb(cp_blkaddr);
|
|
|
|
if (sbi->cur_cp == 2)
|
|
|
|
cp_blk_no += 1 << get_sb(log_blocks_per_seg);
|
|
|
|
|
|
|
|
/* write the first cp */
|
|
|
|
ret = dev_write_block(cp, cp_blk_no++);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
|
|
|
|
/* skip payload */
|
|
|
|
cp_blk_no += get_sb(cp_payload);
|
|
|
|
/* skip orphan blocks */
|
|
|
|
cp_blk_no += orphan_blks;
|
|
|
|
|
|
|
|
/* update summary blocks having nullified journal entries */
|
|
|
|
for (i = 0; i < NO_CHECK_TYPE; i++) {
|
|
|
|
struct curseg_info *curseg = CURSEG_I(sbi, i);
|
|
|
|
u64 ssa_blk;
|
|
|
|
|
|
|
|
ret = dev_write_block(curseg->sum_blk, cp_blk_no++);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
|
|
|
|
/* update original SSA too */
|
|
|
|
ssa_blk = GET_SUM_BLKADDR(sbi, curseg->segno);
|
|
|
|
ret = dev_write_block(curseg->sum_blk, ssa_blk);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
}
|
|
|
|
|
2017-02-09 02:55:43 +00:00
|
|
|
/* Write nat bits */
|
|
|
|
if (flags & CP_NAT_BITS_FLAG)
|
|
|
|
write_nat_bits(sbi, sb, cp, sbi->cur_cp);
|
2017-08-23 04:13:04 +00:00
|
|
|
|
|
|
|
/* in case of sudden power off */
|
2017-11-24 09:33:39 +00:00
|
|
|
ret = f2fs_fsync_device();
|
|
|
|
ASSERT(ret >= 0);
|
2017-08-23 04:13:04 +00:00
|
|
|
|
|
|
|
/* write the last cp */
|
|
|
|
ret = dev_write_block(cp, cp_blk_no++);
|
|
|
|
ASSERT(ret >= 0);
|
2019-05-24 08:26:40 +00:00
|
|
|
|
|
|
|
ret = f2fs_fsync_device();
|
|
|
|
ASSERT(ret >= 0);
|
2015-10-23 18:45:36 +00:00
|
|
|
}
|
|
|
|
|
2019-07-17 01:28:51 +00:00
|
|
|
void write_checkpoints(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
|
|
|
/* copy valid checkpoint to its mirror position */
|
|
|
|
duplicate_checkpoint(sbi);
|
|
|
|
|
|
|
|
/* repair checkpoint at CP #0 position */
|
|
|
|
sbi->cur_cp = 1;
|
|
|
|
write_checkpoint(sbi);
|
|
|
|
}
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
void build_nat_area_bitmap(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
2016-11-08 00:48:10 +00:00
|
|
|
struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
|
|
|
|
struct f2fs_journal *journal = &curseg->sum_blk->journal;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
struct f2fs_fsck *fsck = F2FS_FSCK(sbi);
|
2015-11-22 04:02:18 +00:00
|
|
|
struct f2fs_super_block *sb = F2FS_RAW_SUPER(sbi);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
struct f2fs_nm_info *nm_i = NM_I(sbi);
|
|
|
|
struct f2fs_nat_block *nat_block;
|
2016-11-08 00:48:10 +00:00
|
|
|
struct node_info ni;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
u32 nid, nr_nat_blks;
|
|
|
|
pgoff_t block_off;
|
|
|
|
pgoff_t block_addr;
|
|
|
|
int seg_off;
|
2014-05-14 00:02:55 +00:00
|
|
|
int ret;
|
|
|
|
unsigned int i;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
nat_block = (struct f2fs_nat_block *)calloc(BLOCK_SZ, 1);
|
2014-08-28 21:55:07 +00:00
|
|
|
ASSERT(nat_block);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
/* Alloc & build nat entry bitmap */
|
2015-11-22 04:02:18 +00:00
|
|
|
nr_nat_blks = (get_sb(segment_count_nat) / 2) <<
|
|
|
|
sbi->log_blocks_per_seg;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
fsck->nr_nat_entries = nr_nat_blks * NAT_ENTRY_PER_BLOCK;
|
|
|
|
fsck->nat_area_bitmap_sz = (fsck->nr_nat_entries + 7) / 8;
|
|
|
|
fsck->nat_area_bitmap = calloc(fsck->nat_area_bitmap_sz, 1);
|
2016-11-04 09:34:11 +00:00
|
|
|
ASSERT(fsck->nat_area_bitmap);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2016-03-14 06:16:55 +00:00
|
|
|
fsck->entries = calloc(sizeof(struct f2fs_nat_entry),
|
|
|
|
fsck->nr_nat_entries);
|
|
|
|
ASSERT(fsck->entries);
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
for (block_off = 0; block_off < nr_nat_blks; block_off++) {
|
|
|
|
|
|
|
|
seg_off = block_off >> sbi->log_blocks_per_seg;
|
|
|
|
block_addr = (pgoff_t)(nm_i->nat_blkaddr +
|
2014-08-28 00:13:23 +00:00
|
|
|
(seg_off << sbi->log_blocks_per_seg << 1) +
|
|
|
|
(block_off & ((1 << sbi->log_blocks_per_seg) - 1)));
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
if (f2fs_test_bit(block_off, nm_i->nat_bitmap))
|
|
|
|
block_addr += sbi->blocks_per_seg;
|
|
|
|
|
|
|
|
ret = dev_read_block(nat_block, block_addr);
|
|
|
|
ASSERT(ret >= 0);
|
|
|
|
|
|
|
|
nid = block_off * NAT_ENTRY_PER_BLOCK;
|
|
|
|
for (i = 0; i < NAT_ENTRY_PER_BLOCK; i++) {
|
|
|
|
ni.nid = nid + i;
|
|
|
|
|
2014-08-28 00:06:17 +00:00
|
|
|
if ((nid + i) == F2FS_NODE_INO(sbi) ||
|
|
|
|
(nid + i) == F2FS_META_INO(sbi)) {
|
2017-11-02 03:56:05 +00:00
|
|
|
/*
|
|
|
|
* block_addr of node/meta inode should be 0x1.
|
|
|
|
* Set this bit, and fsck_verify will fix it.
|
|
|
|
*/
|
2016-03-03 02:32:31 +00:00
|
|
|
if (le32_to_cpu(nat_block->entries[i].block_addr) != 0x1) {
|
2017-11-02 03:56:05 +00:00
|
|
|
ASSERT_MSG("\tError: ino[0x%x] block_addr[0x%x] is invalid\n",
|
2016-03-03 02:32:31 +00:00
|
|
|
nid + i, le32_to_cpu(nat_block->entries[i].block_addr));
|
2017-11-02 03:56:05 +00:00
|
|
|
f2fs_set_bit(nid + i, fsck->nat_area_bitmap);
|
2016-03-03 02:32:31 +00:00
|
|
|
}
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
2016-11-08 00:48:10 +00:00
|
|
|
node_info_from_raw_nat(&ni, &nat_block->entries[i]);
|
|
|
|
if (ni.blk_addr == 0x0)
|
|
|
|
continue;
|
|
|
|
if (ni.ino == 0x0) {
|
|
|
|
ASSERT_MSG("\tError: ino[0x%8x] or blk_addr[0x%16x]"
|
|
|
|
" is invalid\n", ni.ino, ni.blk_addr);
|
|
|
|
}
|
|
|
|
if (ni.ino == (nid + i)) {
|
|
|
|
fsck->nat_valid_inode_cnt++;
|
|
|
|
DBG(3, "ino[0x%8x] maybe is inode\n", ni.ino);
|
|
|
|
}
|
|
|
|
if (nid + i == 0) {
|
|
|
|
/*
|
|
|
|
* nat entry [0] must be null. If
|
|
|
|
* it is corrupted, set its bit in
|
|
|
|
* nat_area_bitmap, fsck_verify will
|
|
|
|
* nullify it
|
|
|
|
*/
|
|
|
|
ASSERT_MSG("Invalid nat entry[0]: "
|
|
|
|
"blk_addr[0x%x]\n", ni.blk_addr);
|
|
|
|
fsck->chk.valid_nat_entry_cnt--;
|
|
|
|
}
|
2016-03-14 06:16:55 +00:00
|
|
|
|
2016-11-08 00:48:10 +00:00
|
|
|
DBG(3, "nid[0x%8x] addr[0x%16x] ino[0x%8x]\n",
|
|
|
|
nid + i, ni.blk_addr, ni.ino);
|
|
|
|
f2fs_set_bit(nid + i, fsck->nat_area_bitmap);
|
|
|
|
fsck->chk.valid_nat_entry_cnt++;
|
|
|
|
|
|
|
|
fsck->entries[nid + i] = nat_block->entries[i];
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Traverse nat journal, update the corresponding entries */
|
|
|
|
for (i = 0; i < nats_in_cursum(journal); i++) {
|
|
|
|
struct f2fs_nat_entry raw_nat;
|
|
|
|
nid = le32_to_cpu(nid_in_journal(journal, i));
|
|
|
|
ni.nid = nid;
|
2014-08-28 00:13:23 +00:00
|
|
|
|
2016-11-08 00:48:10 +00:00
|
|
|
DBG(3, "==> Found nid [0x%x] in nat cache, update it\n", nid);
|
2016-03-14 06:16:55 +00:00
|
|
|
|
2016-11-08 00:48:10 +00:00
|
|
|
/* Clear the original bit and count */
|
|
|
|
if (fsck->entries[nid].block_addr != 0x0) {
|
|
|
|
fsck->chk.valid_nat_entry_cnt--;
|
|
|
|
f2fs_clear_bit(nid, fsck->nat_area_bitmap);
|
|
|
|
if (fsck->entries[nid].ino == nid)
|
|
|
|
fsck->nat_valid_inode_cnt--;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Use nat entries in journal */
|
|
|
|
memcpy(&raw_nat, &nat_in_journal(journal, i),
|
|
|
|
sizeof(struct f2fs_nat_entry));
|
|
|
|
node_info_from_raw_nat(&ni, &raw_nat);
|
|
|
|
if (ni.blk_addr != 0x0) {
|
|
|
|
if (ni.ino == 0x0)
|
|
|
|
ASSERT_MSG("\tError: ino[0x%8x] or blk_addr[0x%16x]"
|
|
|
|
" is invalid\n", ni.ino, ni.blk_addr);
|
|
|
|
if (ni.ino == nid) {
|
|
|
|
fsck->nat_valid_inode_cnt++;
|
|
|
|
DBG(3, "ino[0x%8x] maybe is inode\n", ni.ino);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
2016-11-08 00:48:10 +00:00
|
|
|
f2fs_set_bit(nid, fsck->nat_area_bitmap);
|
|
|
|
fsck->chk.valid_nat_entry_cnt++;
|
|
|
|
DBG(3, "nid[0x%x] in nat cache\n", nid);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
2016-11-08 00:48:10 +00:00
|
|
|
fsck->entries[nid] = raw_nat;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
free(nat_block);
|
|
|
|
|
2013-07-30 07:39:04 +00:00
|
|
|
DBG(1, "valid nat entries (block_addr != 0x0) [0x%8x : %u]\n",
|
2014-08-28 00:13:23 +00:00
|
|
|
fsck->chk.valid_nat_entry_cnt,
|
|
|
|
fsck->chk.valid_nat_entry_cnt);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
|
2016-05-18 10:00:37 +00:00
|
|
|
static int check_sector_size(struct f2fs_super_block *sb)
|
|
|
|
{
|
|
|
|
u_int32_t log_sectorsize, log_sectors_per_block;
|
|
|
|
|
2016-09-17 01:41:00 +00:00
|
|
|
log_sectorsize = log_base_2(c.sector_size);
|
|
|
|
log_sectors_per_block = log_base_2(c.sectors_per_blk);
|
2016-05-18 10:00:37 +00:00
|
|
|
|
2016-09-29 22:00:07 +00:00
|
|
|
if (log_sectorsize == get_sb(log_sectorsize) &&
|
|
|
|
log_sectors_per_block == get_sb(log_sectors_per_block))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
set_sb(log_sectorsize, log_sectorsize);
|
|
|
|
set_sb(log_sectors_per_block, log_sectors_per_block);
|
2016-05-18 10:00:37 +00:00
|
|
|
|
2018-09-28 12:25:58 +00:00
|
|
|
update_superblock(sb, SB_MASK_ALL);
|
2016-05-18 10:00:37 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2019-07-11 20:45:41 +00:00
|
|
|
static int tune_sb_features(struct f2fs_sb_info *sbi)
|
2018-04-19 20:53:31 +00:00
|
|
|
{
|
|
|
|
int sb_changed = 0;
|
|
|
|
struct f2fs_super_block *sb = F2FS_RAW_SUPER(sbi);
|
|
|
|
|
|
|
|
if (!(sb->feature & cpu_to_le32(F2FS_FEATURE_ENCRYPT)) &&
|
|
|
|
c.feature & cpu_to_le32(F2FS_FEATURE_ENCRYPT)) {
|
|
|
|
sb->feature |= cpu_to_le32(F2FS_FEATURE_ENCRYPT);
|
|
|
|
MSG(0, "Info: Set Encryption feature\n");
|
|
|
|
sb_changed = 1;
|
|
|
|
}
|
2019-07-11 20:45:41 +00:00
|
|
|
if (!(sb->feature & cpu_to_le32(F2FS_FEATURE_CASEFOLD)) &&
|
|
|
|
c.feature & cpu_to_le32(F2FS_FEATURE_CASEFOLD)) {
|
|
|
|
if (!c.s_encoding) {
|
|
|
|
ERR_MSG("ERROR: Must specify encoding to enable casefolding.\n");
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
sb->feature |= cpu_to_le32(F2FS_FEATURE_CASEFOLD);
|
|
|
|
MSG(0, "Info: Set Casefold feature\n");
|
|
|
|
sb_changed = 1;
|
|
|
|
}
|
2018-04-19 20:53:31 +00:00
|
|
|
/* TODO: quota needs to allocate inode numbers */
|
|
|
|
|
|
|
|
c.feature = sb->feature;
|
|
|
|
if (!sb_changed)
|
2019-07-11 20:45:41 +00:00
|
|
|
return 0;
|
2018-04-19 20:53:31 +00:00
|
|
|
|
2018-09-28 12:25:58 +00:00
|
|
|
update_superblock(sb, SB_MASK_ALL);
|
2019-07-11 20:45:41 +00:00
|
|
|
return 0;
|
2018-04-19 20:53:31 +00:00
|
|
|
}
|
|
|
|
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
static struct fsync_inode_entry *get_fsync_inode(struct list_head *head,
|
|
|
|
nid_t ino)
|
|
|
|
{
|
|
|
|
struct fsync_inode_entry *entry;
|
|
|
|
|
|
|
|
list_for_each_entry(entry, head, list)
|
|
|
|
if (entry->ino == ino)
|
|
|
|
return entry;
|
|
|
|
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct fsync_inode_entry *add_fsync_inode(struct list_head *head,
|
|
|
|
nid_t ino)
|
|
|
|
{
|
|
|
|
struct fsync_inode_entry *entry;
|
|
|
|
|
|
|
|
entry = calloc(sizeof(struct fsync_inode_entry), 1);
|
|
|
|
if (!entry)
|
|
|
|
return NULL;
|
|
|
|
entry->ino = ino;
|
|
|
|
list_add_tail(&entry->list, head);
|
|
|
|
return entry;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void del_fsync_inode(struct fsync_inode_entry *entry)
|
|
|
|
{
|
|
|
|
list_del(&entry->list);
|
|
|
|
free(entry);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void destroy_fsync_dnodes(struct list_head *head)
|
|
|
|
{
|
|
|
|
struct fsync_inode_entry *entry, *tmp;
|
|
|
|
|
|
|
|
list_for_each_entry_safe(entry, tmp, head, list)
|
|
|
|
del_fsync_inode(entry);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int find_fsync_inode(struct f2fs_sb_info *sbi, struct list_head *head)
|
|
|
|
{
|
|
|
|
struct curseg_info *curseg;
|
|
|
|
struct f2fs_node *node_blk;
|
|
|
|
block_t blkaddr;
|
|
|
|
unsigned int loop_cnt = 0;
|
|
|
|
unsigned int free_blocks = TOTAL_SEGS(sbi) * sbi->blocks_per_seg -
|
|
|
|
sbi->total_valid_block_count;
|
|
|
|
int err = 0;
|
|
|
|
|
|
|
|
/* get node pages in the current segment */
|
|
|
|
curseg = CURSEG_I(sbi, CURSEG_WARM_NODE);
|
|
|
|
blkaddr = NEXT_FREE_BLKADDR(sbi, curseg);
|
|
|
|
|
|
|
|
node_blk = calloc(F2FS_BLKSIZE, 1);
|
|
|
|
ASSERT(node_blk);
|
|
|
|
|
|
|
|
while (1) {
|
|
|
|
struct fsync_inode_entry *entry;
|
|
|
|
|
|
|
|
if (!f2fs_is_valid_blkaddr(sbi, blkaddr, META_POR))
|
|
|
|
break;
|
|
|
|
|
|
|
|
err = dev_read_block(node_blk, blkaddr);
|
|
|
|
if (err)
|
|
|
|
break;
|
|
|
|
|
|
|
|
if (!is_recoverable_dnode(sbi, node_blk))
|
|
|
|
break;
|
|
|
|
|
|
|
|
if (!is_fsync_dnode(node_blk))
|
|
|
|
goto next;
|
|
|
|
|
|
|
|
entry = get_fsync_inode(head, ino_of_node(node_blk));
|
|
|
|
if (!entry) {
|
|
|
|
entry = add_fsync_inode(head, ino_of_node(node_blk));
|
|
|
|
if (!entry) {
|
|
|
|
err = -1;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
entry->blkaddr = blkaddr;
|
|
|
|
|
|
|
|
if (IS_INODE(node_blk) && is_dent_dnode(node_blk))
|
|
|
|
entry->last_dentry = blkaddr;
|
|
|
|
next:
|
|
|
|
/* sanity check in order to detect looped node chain */
|
|
|
|
if (++loop_cnt >= free_blocks ||
|
|
|
|
blkaddr == next_blkaddr_of_node(node_blk)) {
|
|
|
|
MSG(0, "\tdetect looped node chain, blkaddr:%u, next:%u\n",
|
|
|
|
blkaddr,
|
|
|
|
next_blkaddr_of_node(node_blk));
|
|
|
|
err = -1;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
blkaddr = next_blkaddr_of_node(node_blk);
|
|
|
|
}
|
|
|
|
|
|
|
|
free(node_blk);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int do_record_fsync_data(struct f2fs_sb_info *sbi,
|
|
|
|
struct f2fs_node *node_blk,
|
|
|
|
block_t blkaddr)
|
|
|
|
{
|
|
|
|
unsigned int segno, offset;
|
|
|
|
struct seg_entry *se;
|
|
|
|
unsigned int ofs_in_node = 0;
|
|
|
|
unsigned int start, end;
|
|
|
|
int err = 0, recorded = 0;
|
|
|
|
|
|
|
|
segno = GET_SEGNO(sbi, blkaddr);
|
|
|
|
se = get_seg_entry(sbi, segno);
|
|
|
|
offset = OFFSET_IN_SEG(sbi, blkaddr);
|
|
|
|
|
|
|
|
if (f2fs_test_bit(offset, (char *)se->cur_valid_map)) {
|
|
|
|
ASSERT(0);
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
if (f2fs_test_bit(offset, (char *)se->ckpt_valid_map)) {
|
|
|
|
ASSERT(0);
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!se->ckpt_valid_blocks)
|
|
|
|
se->ckpt_type = CURSEG_WARM_NODE;
|
|
|
|
|
|
|
|
se->ckpt_valid_blocks++;
|
|
|
|
f2fs_set_bit(offset, (char *)se->ckpt_valid_map);
|
|
|
|
|
|
|
|
MSG(1, "do_record_fsync_data: [node] ino = %u, nid = %u, blkaddr = %u\n",
|
|
|
|
ino_of_node(node_blk), ofs_of_node(node_blk), blkaddr);
|
|
|
|
|
|
|
|
/* inline data */
|
|
|
|
if (IS_INODE(node_blk) && (node_blk->i.i_inline & F2FS_INLINE_DATA))
|
|
|
|
return 0;
|
|
|
|
/* xattr node */
|
|
|
|
if (ofs_of_node(node_blk) == XATTR_NODE_OFFSET)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
/* step 3: recover data indices */
|
|
|
|
start = start_bidx_of_node(ofs_of_node(node_blk), node_blk);
|
2019-03-25 13:19:35 +00:00
|
|
|
end = start + ADDRS_PER_PAGE(sbi, node_blk, NULL);
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
|
|
|
|
for (; start < end; start++, ofs_in_node++) {
|
|
|
|
blkaddr = datablock_addr(node_blk, ofs_in_node);
|
|
|
|
|
|
|
|
if (!is_valid_data_blkaddr(blkaddr))
|
|
|
|
continue;
|
|
|
|
|
|
|
|
if (!f2fs_is_valid_blkaddr(sbi, blkaddr, META_POR)) {
|
|
|
|
err = -1;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
segno = GET_SEGNO(sbi, blkaddr);
|
|
|
|
se = get_seg_entry(sbi, segno);
|
|
|
|
offset = OFFSET_IN_SEG(sbi, blkaddr);
|
|
|
|
|
|
|
|
if (f2fs_test_bit(offset, (char *)se->cur_valid_map))
|
|
|
|
continue;
|
|
|
|
if (f2fs_test_bit(offset, (char *)se->ckpt_valid_map))
|
|
|
|
continue;
|
|
|
|
|
|
|
|
if (!se->ckpt_valid_blocks)
|
|
|
|
se->ckpt_type = CURSEG_WARM_DATA;
|
|
|
|
|
|
|
|
se->ckpt_valid_blocks++;
|
|
|
|
f2fs_set_bit(offset, (char *)se->ckpt_valid_map);
|
|
|
|
|
|
|
|
MSG(1, "do_record_fsync_data: [data] ino = %u, nid = %u, blkaddr = %u\n",
|
|
|
|
ino_of_node(node_blk), ofs_of_node(node_blk), blkaddr);
|
|
|
|
|
|
|
|
recorded++;
|
|
|
|
}
|
|
|
|
out:
|
|
|
|
MSG(1, "recover_data: ino = %u, nid = %u, recorded = %d, err = %d\n",
|
|
|
|
ino_of_node(node_blk), ofs_of_node(node_blk),
|
|
|
|
recorded, err);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int traverse_dnodes(struct f2fs_sb_info *sbi,
|
|
|
|
struct list_head *inode_list)
|
|
|
|
{
|
|
|
|
struct curseg_info *curseg;
|
|
|
|
struct f2fs_node *node_blk;
|
|
|
|
block_t blkaddr;
|
|
|
|
int err = 0;
|
|
|
|
|
|
|
|
/* get node pages in the current segment */
|
|
|
|
curseg = CURSEG_I(sbi, CURSEG_WARM_NODE);
|
|
|
|
blkaddr = NEXT_FREE_BLKADDR(sbi, curseg);
|
|
|
|
|
|
|
|
node_blk = calloc(F2FS_BLKSIZE, 1);
|
|
|
|
ASSERT(node_blk);
|
|
|
|
|
|
|
|
while (1) {
|
|
|
|
struct fsync_inode_entry *entry;
|
|
|
|
|
|
|
|
if (!f2fs_is_valid_blkaddr(sbi, blkaddr, META_POR))
|
|
|
|
break;
|
|
|
|
|
|
|
|
err = dev_read_block(node_blk, blkaddr);
|
|
|
|
if (err)
|
|
|
|
break;
|
|
|
|
|
|
|
|
if (!is_recoverable_dnode(sbi, node_blk))
|
|
|
|
break;
|
|
|
|
|
|
|
|
entry = get_fsync_inode(inode_list,
|
|
|
|
ino_of_node(node_blk));
|
|
|
|
if (!entry)
|
|
|
|
goto next;
|
|
|
|
|
|
|
|
err = do_record_fsync_data(sbi, node_blk, blkaddr);
|
|
|
|
if (err)
|
|
|
|
break;
|
|
|
|
|
|
|
|
if (entry->blkaddr == blkaddr)
|
|
|
|
del_fsync_inode(entry);
|
|
|
|
next:
|
|
|
|
blkaddr = next_blkaddr_of_node(node_blk);
|
|
|
|
}
|
|
|
|
|
|
|
|
free(node_blk);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int record_fsync_data(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
|
|
|
struct list_head inode_list = LIST_HEAD_INIT(inode_list);
|
|
|
|
int ret;
|
|
|
|
|
2019-11-28 07:59:28 +00:00
|
|
|
if (!need_fsync_data_record(sbi))
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
ret = find_fsync_inode(sbi, &inode_list);
|
|
|
|
if (ret)
|
|
|
|
goto out;
|
|
|
|
|
2020-07-01 08:23:29 +00:00
|
|
|
ret = late_build_segment_manager(sbi);
|
|
|
|
if (ret < 0) {
|
|
|
|
ERR_MSG("late_build_segment_manager failed\n");
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
ret = traverse_dnodes(sbi, &inode_list);
|
|
|
|
out:
|
|
|
|
destroy_fsync_dnodes(&inode_list);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
int f2fs_do_mount(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
2015-11-22 04:02:18 +00:00
|
|
|
struct f2fs_checkpoint *cp = NULL;
|
2017-02-09 02:55:43 +00:00
|
|
|
struct f2fs_super_block *sb = NULL;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
int ret;
|
2014-09-03 01:07:35 +00:00
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
sbi->active_logs = NR_CURSEG_TYPE;
|
2018-09-28 12:25:58 +00:00
|
|
|
ret = validate_super_block(sbi, SB0_ADDR);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
if (ret) {
|
2018-09-28 12:25:58 +00:00
|
|
|
ret = validate_super_block(sbi, SB1_ADDR);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
if (ret)
|
|
|
|
return -1;
|
|
|
|
}
|
2017-02-09 02:55:43 +00:00
|
|
|
sb = F2FS_RAW_SUPER(sbi);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2017-02-09 02:55:43 +00:00
|
|
|
ret = check_sector_size(sb);
|
2016-05-18 10:00:37 +00:00
|
|
|
if (ret)
|
|
|
|
return -1;
|
|
|
|
|
2017-02-09 02:55:43 +00:00
|
|
|
print_raw_sb_info(sb);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
init_sb_info(sbi);
|
|
|
|
|
|
|
|
ret = get_valid_checkpoint(sbi);
|
|
|
|
if (ret) {
|
|
|
|
ERR_MSG("Can't find valid checkpoint\n");
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
2020-03-24 09:12:50 +00:00
|
|
|
c.bug_on = 0;
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
if (sanity_check_ckpt(sbi)) {
|
|
|
|
ERR_MSG("Checkpoint is polluted\n");
|
|
|
|
return -1;
|
|
|
|
}
|
2016-01-15 04:36:20 +00:00
|
|
|
cp = F2FS_CKPT(sbi);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
2019-08-09 10:53:00 +00:00
|
|
|
if (c.func != FSCK && c.func != DUMP &&
|
|
|
|
!is_set_ckpt_flags(F2FS_CKPT(sbi), CP_UMOUNT_FLAG)) {
|
|
|
|
ERR_MSG("Mount unclean image to replay log first\n");
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
print_ckpt_info(sbi);
|
|
|
|
|
2018-09-19 01:28:37 +00:00
|
|
|
if (c.quota_fix) {
|
|
|
|
if (get_cp(ckpt_flags) & CP_QUOTA_NEED_FSCK_FLAG)
|
|
|
|
c.fix_on = 1;
|
|
|
|
}
|
|
|
|
|
2019-07-11 20:45:41 +00:00
|
|
|
if (tune_sb_features(sbi))
|
|
|
|
return -1;
|
2017-07-26 15:01:35 +00:00
|
|
|
|
|
|
|
/* precompute checksum seed for metadata */
|
|
|
|
if (c.feature & cpu_to_le32(F2FS_FEATURE_INODE_CHKSUM))
|
|
|
|
c.chksum_seed = f2fs_cal_crc32(~0, sb->uuid, sizeof(sb->uuid));
|
2014-09-03 01:07:35 +00:00
|
|
|
|
2015-11-22 04:02:18 +00:00
|
|
|
sbi->total_valid_node_count = get_cp(valid_node_count);
|
|
|
|
sbi->total_valid_inode_count = get_cp(valid_inode_count);
|
|
|
|
sbi->user_block_count = get_cp(user_block_count);
|
|
|
|
sbi->total_valid_block_count = get_cp(valid_block_count);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
sbi->last_valid_block_count = sbi->total_valid_block_count;
|
|
|
|
sbi->alloc_valid_block_count = 0;
|
|
|
|
|
2020-07-01 08:23:29 +00:00
|
|
|
if (early_build_segment_manager(sbi)) {
|
|
|
|
ERR_MSG("early_build_segment_manager failed\n");
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (build_node_manager(sbi)) {
|
2016-09-29 10:38:35 +00:00
|
|
|
ERR_MSG("build_node_manager failed\n");
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
f2fs-tools: fix to skip block allocation for fsynced data
Previously, we don't allow block allocation on unclean umounted image,
result in failing to repair quota system file.
In this patch, we port most recovery codes from kernel to userspace
tools, so that on unclean image, during fsck initialization, we will
record all data/node block address we may recover in kernel, and
then during allocation of quota file repair, we can skip those blocks
to avoid block use conflict.
Eventually, if free space is enough, we can repair the quota system
file on an unclean umounted image.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: remove unnecessary parameter]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2019-08-14 08:48:55 +00:00
|
|
|
if (record_fsync_data(sbi)) {
|
|
|
|
ERR_MSG("record_fsync_data failed\n");
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
2020-03-24 09:12:50 +00:00
|
|
|
if (!f2fs_should_proceed(sb, get_cp(ckpt_flags)))
|
|
|
|
return 1;
|
2019-05-24 06:09:47 +00:00
|
|
|
|
2020-07-01 08:23:29 +00:00
|
|
|
if (late_build_segment_manager(sbi) < 0) {
|
|
|
|
ERR_MSG("late_build_segment_manager failed\n");
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
2020-06-19 09:52:12 +00:00
|
|
|
if (f2fs_late_init_nid_bitmap(sbi)) {
|
|
|
|
ERR_MSG("f2fs_late_init_nid_bitmap failed\n");
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
2017-02-09 02:55:43 +00:00
|
|
|
/* Check nat_bits */
|
2018-06-17 17:56:49 +00:00
|
|
|
if (c.func == FSCK && is_set_ckpt_flags(cp, CP_NAT_BITS_FLAG)) {
|
|
|
|
if (check_nat_bits(sbi, sb, cp) && c.fix_on)
|
2017-02-09 02:55:43 +00:00
|
|
|
write_nat_bits(sbi, sb, cp, sbi->cur_cp);
|
|
|
|
}
|
2014-09-03 01:07:35 +00:00
|
|
|
return 0;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
void f2fs_do_umount(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
|
|
|
struct sit_info *sit_i = SIT_I(sbi);
|
|
|
|
struct f2fs_sm_info *sm_i = SM_I(sbi);
|
|
|
|
struct f2fs_nm_info *nm_i = NM_I(sbi);
|
2014-05-14 00:02:55 +00:00
|
|
|
unsigned int i;
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
|
|
|
|
/* free nm_info */
|
2017-10-30 23:21:09 +00:00
|
|
|
if (c.func == SLOAD || c.func == FSCK)
|
2015-12-10 00:18:44 +00:00
|
|
|
free(nm_i->nid_bitmap);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
free(nm_i->nat_bitmap);
|
|
|
|
free(sbi->nm_info);
|
|
|
|
|
|
|
|
/* free sit_info */
|
2019-08-09 10:52:55 +00:00
|
|
|
free(sit_i->bitmap);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
free(sit_i->sit_bitmap);
|
2019-08-09 10:52:54 +00:00
|
|
|
free(sit_i->sentries);
|
f2fs-tools: add fsck.f2fs and dump.f2fs
fsck.f2fs checks file system consistency, but does not repair a broken
file system yet.
dump.f2fs shows the information of a specific inode and makes dump file
of SSA and SIT.
f2fs checks file system consistency as follows:
o When data about used area and its metadata are identical,
f2fs is considered consistent. To verify such consistency, we use
three bitmaps: nat_area_bitmap, sit_area_bitmap, and main_area_bitmap.
First, each bit in nat_area_bitmap corresponds to a nid in NAT.
Second, each bit in sit_area_bitmap corresponds to a valid block in a
segment. This bitmap is same to the total valid_map of f2fs_sit_entries
in SIT.
Last, each bit in main_area_bitmap corresponds to a block in main area
except meta area.
After a consistency check of each block, we set or clear the
corresponding bit of each bitmap.
From the root node, we start consistency check. The verified
information varies according to block type.
1. NODE
- Read information of node block from NAT
- Check if block address is allocated using node info.
- Check if the type of f2fs_summary related to nid in SSA is NODE.
- Update the corresponding bit in nat_area_bitmap.
- Update the corresponding bit in sit_area_bitmap.
- Set the corresponding bit in main_area_bitmap to 1.
- Then, read node block. According to its attribute, explore
inode/direct node/indirect node/double indirect node
recursively.
- If it is an inode block, we also check its xattr and hard link.
2. DATA
- Check if the type of f2fs_summary related to nid in SSA is DATA.
- Set the corresponding bits of sit_area_bitmap and
main_area_bitmap to visited
- If it is a dentry block, traverse each dentries that may be
regular
file or directory. At this time, it will check inode block again.
Finally, we verify whether
- every nat_area_bitmap is visited
- any unreachable hard link exists
- values of sit_area_bitmap and main_area_bitmap are identical
- total_valid_block_count/node_count/inode_count are correct
Usage:
o fsck.f2fs
# fsck.f2fs /dev/sdx
options:
-d debug level [default:0]
o dump.f2fs
# dump.f2fs -i [ino] /dev/sdx
# dump.f2fs -s 0~-1 /dev/sdx (SIT dump)
# dump.f2fs -a 0~-1 /dev/sdx (SSA dump)
options:
-d debug level [default:0]
-i inode no (hex)
-s [SIT dump segno from #1~#2 (decimal), for all 0~-1]
-a [SSA dump segno from #1~#2 (decimal), for all 0~-1]
Note: To use dump.f2fs, please run make install or ln -s fsck.f2fs
dump.f2fs
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Byoung Geun Kim <bgbg.kim@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-04 08:11:32 +00:00
|
|
|
free(sm_i->sit_info);
|
|
|
|
|
|
|
|
/* free sm_info */
|
|
|
|
for (i = 0; i < NR_CURSEG_TYPE; i++)
|
|
|
|
free(sm_i->curseg_array[i].sum_blk);
|
|
|
|
|
|
|
|
free(sm_i->curseg_array);
|
|
|
|
free(sbi->sm_info);
|
|
|
|
|
|
|
|
free(sbi->ckpt);
|
|
|
|
free(sbi->raw_super);
|
|
|
|
}
|
2019-05-06 08:58:06 +00:00
|
|
|
|
|
|
|
#ifdef WITH_ANDROID
|
|
|
|
int f2fs_sparse_initialize_meta(struct f2fs_sb_info *sbi)
|
|
|
|
{
|
|
|
|
struct f2fs_super_block *sb = sbi->raw_super;
|
|
|
|
u_int32_t sit_seg_count, sit_size;
|
|
|
|
u_int32_t nat_seg_count, nat_size;
|
|
|
|
u_int64_t sit_seg_addr, nat_seg_addr, payload_addr;
|
|
|
|
u_int32_t seg_size = 1 << get_sb(log_blocks_per_seg);
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
if (!c.sparse_mode)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
sit_seg_addr = get_sb(sit_blkaddr);
|
|
|
|
sit_seg_count = get_sb(segment_count_sit);
|
|
|
|
sit_size = sit_seg_count * seg_size;
|
|
|
|
|
|
|
|
DBG(1, "\tSparse: filling sit area at block offset: 0x%08"PRIx64" len: %u\n",
|
|
|
|
sit_seg_addr, sit_size);
|
|
|
|
ret = dev_fill(NULL, sit_seg_addr * F2FS_BLKSIZE,
|
|
|
|
sit_size * F2FS_BLKSIZE);
|
|
|
|
if (ret) {
|
|
|
|
MSG(1, "\tError: While zeroing out the sit area "
|
|
|
|
"on disk!!!\n");
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
nat_seg_addr = get_sb(nat_blkaddr);
|
|
|
|
nat_seg_count = get_sb(segment_count_nat);
|
|
|
|
nat_size = nat_seg_count * seg_size;
|
|
|
|
|
|
|
|
DBG(1, "\tSparse: filling nat area at block offset 0x%08"PRIx64" len: %u\n",
|
|
|
|
nat_seg_addr, nat_size);
|
|
|
|
ret = dev_fill(NULL, nat_seg_addr * F2FS_BLKSIZE,
|
|
|
|
nat_size * F2FS_BLKSIZE);
|
|
|
|
if (ret) {
|
|
|
|
MSG(1, "\tError: While zeroing out the nat area "
|
|
|
|
"on disk!!!\n");
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
payload_addr = get_sb(segment0_blkaddr) + 1;
|
|
|
|
|
|
|
|
DBG(1, "\tSparse: filling bitmap area at block offset 0x%08"PRIx64" len: %u\n",
|
|
|
|
payload_addr, get_sb(cp_payload));
|
|
|
|
ret = dev_fill(NULL, payload_addr * F2FS_BLKSIZE,
|
|
|
|
get_sb(cp_payload) * F2FS_BLKSIZE);
|
|
|
|
if (ret) {
|
|
|
|
MSG(1, "\tError: While zeroing out the nat/sit bitmap area "
|
|
|
|
"on disk!!!\n");
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
payload_addr += seg_size;
|
|
|
|
|
|
|
|
DBG(1, "\tSparse: filling bitmap area at block offset 0x%08"PRIx64" len: %u\n",
|
|
|
|
payload_addr, get_sb(cp_payload));
|
|
|
|
ret = dev_fill(NULL, payload_addr * F2FS_BLKSIZE,
|
|
|
|
get_sb(cp_payload) * F2FS_BLKSIZE);
|
|
|
|
if (ret) {
|
|
|
|
MSG(1, "\tError: While zeroing out the nat/sit bitmap area "
|
|
|
|
"on disk!!!\n");
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
#else
|
|
|
|
int f2fs_sparse_initialize_meta(struct f2fs_sb_info *sbi) { return 0; }
|
|
|
|
#endif
|