This patch enables injecting sit entry. print_raw_sit_entry_info() is
added to show values of the sit entry.
The meanings of options are:
* sit: means sit entry is injected, its argument chooses which sit pack
to be injected, where 0 means the current valid sit is choosen
automatically.
* blk: is the block address of the sit entry.
The members could be injected in cp contains:
* vblocks: sit entry vblocks
* valid_map: sit entry valid_map
* mtime: sit entry mtime
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Sheng Yong <shengyong@oppo.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch enables injecting nat entry. print_raw_nat_entry_info() is
added to show values of the nat entry.
The meanings of options are:
* nat: means nat entry is injected, its argument chooses which nat pack
to be injected, where 0 means the current valid nat is choosen
automatically.
* nid: is the nid of the nat entry
The members could be injected in cp contains:
* version: nat entry version
* ino: nat entry ino
* block_addr: nat entry block_addr
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Sheng Yong <shengyong@oppo.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch enables injecting checkpoint. To archive this, a helper
write_raw_cp_blocks() is added to write the first and last blocks
of a specific cp. And print_ckpt_info() is exported to show new
checkpoint info.
The meanings of options are:
* cp: means cp is injected, its argument chooses which cp pack to be
injected, where 0 means the current valid cp is choosen
automatically.
The members could be injected in cp contains:
* checkpoint_ver: checkpoint version
* ckpt_flags: checkpoint flags
* cur_node_segno: cur_node_segno array
* cur_node_blkoff: cur_node_blkoff array
* cur_data_segno: cur_data_segno array
* cur_data_blkoff: cur_data_blkoff array
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Sheng Yong <shengyong@oppo.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch enables injecting super block.
The meanings of options are:
* sb: means sb is injected, its argument chooses which sb pack to be
injected, where 0 means the current valid sb is choosen
automatically.
* mb: indicates which member in sb is injected, its argument is the
member name.
* idx: is used when the mb is an array, its argument is the index of
the array.
* val: is the new value to be set when the type of mb is a number.
* str: is the new value to be set when the type of mb is a string.
The members could be injected in sb contains:
* magic: magic number
* s_stop_reason: s_stop_reason array
* s_errors: s_errors array
* devs.path: path in devs array
Signed-off-by: Sheng Yong <shengyong@oppo.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch introduces a new tool inject.f2fs to modify metadata or data
(directory entry) of f2fs image offline flexibly.
With inject.f2fs, it is easier to generate a corrupted f2fs image, which
can help verify fsck or reproduce userspace behaviors of some a fault.
If option `--dry-run' is used, nothing really gets changed, and that
could be used to get the value of a specified field.
The following patches enable injecting each part of f2fs.
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Sheng Yong <shengyong@oppo.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Remove redundant function definition of is_sit_bitmap_set() and
export it.
Signed-off-by: Sheng Yong <shengyong@oppo.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
For zoned model, the first device (devices[0]) is not zoned device,
whose zoned_model is not F2FS_ZONED_HM. Let's skip it and check write
pointer of left devices continuously.
Signed-off-by: Sheng Yong <shengyong@oppo.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
If a superblock failed in sanity check, it should be fixed. This patch
add a new state `sb_invalid' to tell fsck needs to update superblock
at the end of all checkings.
This patch also cleans up force_stop, abnormal_stop, fs_errors and
sb_invalid by merging them into an `invalid_sb' flags, and each of
them is indicated using one bit.
Signed-off-by: Sheng Yong <shengyong@oppo.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Introduce a new help addrs_per_page() to wrap common code
from addrs_per_inode() and addrs_per_block() for cleanup.
Signed-off-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This patch fixes some cases in where we missed to consider
endianness when we access/update inode.i_flags.
Signed-off-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Quoted from manual of chattr:
"
CHATTR(1)
...
ATTRIBUTES
A file with the 'd' attribute set is not a candidate for backup when
the dump(8) program is run.
"
Once we set F2FS_NODUMP_FL flag to inode, do not allow dumping info
from it.
Signed-off-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
BLKFINISHZONE may not defined in old blkzoned.h, and cause build error.
Signed-off-by: Wu Bo <bo.wu@vivo.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Previously, dumped symlinks would always create regular files instead.
This allows symlinks to be dumped as symlinks with the -L option.
The i_name field's name may not be the same as the actual name from the
dirent, so we use the dirent name when available.
Currently hardlinks aren't detected, so print a warning if we notice a
nondirectory with a link count over 1.
Signed-off-by: Daniel Rosenberg <drosen@google.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
When using the f2fs_io tool to set_verity, it will fail as follows:
unisc:/data # ./f2fs_io set_verity file
FS_IOC_ENABLE_VERITY: Inappropriate ioctl for device
this is because commit: 95ae251fe828 ("f2fs: add fs-verity support"),
the passed parameters do not match the latest kernel version.
After patch:
unisoc:/data # ./f2fs_io set_verity file
Set fsverity bit to file
unisoc:/data # ./f2fs_io getflags file
get a flag on file ret=0, flags=verity
Fixes: 95ae251fe828 ("f2fs: add fs-verity support")
Signed-off-by: Xiuhong Wang <xiuhong.wang@unisoc.com>
Signed-off-by: Zhiguo Niu <zhiguo.niu@unisoc.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
If input exceeds int range, data will be truncated, such as lseek:
unisoc:/data # ./f2fs_io lseek set 3221225000 file
returned offset=2147483647
The offset is truncated.
After patch:
unisoc:/data # ./f2fs_io lseek set 3221225000 file
returned offset=3221225000
Signed-off-by: Xiuhong Wang <xiuhong.wang@unisoc.com>
Signed-off-by: Zhiguo Niu <zhiguo.niu@unisoc.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
The code for dumping files was ignoring i_extra_isize, leading to
invalid data for some inlined files after dumping. This swaps inode
dumping to use the common inline_data_addr function.
Signed-off-by: Daniel Rosenberg <drosen@google.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Add missing usage for directory dumping functionality of dump.f2fs
as below:
-r dump out from the root inode
-f do not prompt before dumping
-y alias for -f
-o dump inodes to the given path
-P preserve mode/owner/group for dumped inode
Cc: Daniel Rosenberg <drosen@google.com>
Signed-off-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Xattrs for files with inline data were being skipped. This dumps those,
as well as xattrs for folders.
Signed-off-by: Daniel Rosenberg <drosen@google.com>
Reviewed-by: Daeho Jeong <daehojeong@google.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This adds the ability to dump folders as well as files. Folders are
dumped recursively. Additionally, dumped files/folders may be directed
to a folder specified by -o [path] instead of ./lost_found. The -r flag
will dump the entire fs from the root inode. -f or -y will skip the
prompt before dumping, and -P will preserve the mode/owner info for the
created file/folder.
Signed-off-by: Daniel Rosenberg <drosen@google.com>
Reviewed-by: Daeho Jeong <daehojeong@google.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
For multiple device, each device should be aligned to zone size, instead
of aligning the total size.
Signed-off-by: Sheng Yong <shengyong@oppo.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Add missing description for readonly feature in manual of mkfs.f2fs.
Signed-off-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Send reset commands to only current zones and finish the others.
Signed-off-by: Daeho Jeong <daehojeong@google.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Support get_advise command to get i_advise field value and info
in file.
For example:
f2fs_io get_advise /mnt/f2fs/foo.so
i_advise=0x11, advise_type: cold keep_size
Signed-off-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
The reserved segments should be aligned to the section boundary.
Reviewed-by: Daeho Jeong <daehojeong@google.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
This fixes some android build failures due to the missing permission when
checking the loop device. Until we get a better solution, let's ignore
the error with warnings.
Reviewed-by: Huang Jianan <huangjianan@xiaomi.com>
Reviewed-by: Daeho Jeong <daehojeong@google.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Add a write_advice command with same behavior as existing write command
with the exception that data should be treated as hot or cold.
Signed-off-by: Konstantin Vyshetsky <vkon@google.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Add a new stop cp reason STOP_CP_REASON_NO_SEGMENT for keeping
consistent with kernel codes.
Signed-off-by: Zhiguo Niu <zhiguo.niu@unisoc.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Added lseek command to support lseek() for SEEK_DATA and SEEK_HOLE.
Signed-off-by: Daeho Jeong <daehojeong@google.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
When we turn on sparse_mode, stat() will be failed, as the file will be
created afterwards. Let's fix.
Fixes: 14197d546b ("f2fs-tools: fix to check loop device")
Reviewed-by: Daeho Jeong <daehojeong@google.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Make to allocate logs after conventional area for HM zoned devices to
spare them for file pinning support.
Signed-off-by: Daeho Jeong <daehojeong@google.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
If sparse files is set along with multiple devices, we initialize
sparse file data multiple times without freeing the previously allocated
data. This skips the initialization for subsequent devices, as sparse
file mode currently only deals with device 0 anyways.
Signed-off-by: Daniel Rosenberg <drosen@google.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Since we may not know the block size when initializing sparse files, we
should assume that the sparse file's blocksize is correct.
Signed-off-by: Daniel Rosenberg <drosen@google.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
generic/019 50s ... _check_generic_filesystem: filesystem on /dev/vdc is inconsistent
(see /media/fstests/results//generic/019.full for details)
[FSCK] Max image size: 16196 MB, Free space: 188 MB
[FSCK] Unreachable nat entries [Ok..] [0x0]
[FSCK] SIT valid block bitmap checking [Ok..]
[FSCK] Hard link checking for regular file [Ok..] [0x166]
[FSCK] valid_block_count matching with CP [Ok..] [0x3ecfe7]
[FSCK] valid_node_count matching with CP (de lookup) [Ok..] [0x4c79]
[FSCK] valid_node_count matching with CP (nat lookup) [Ok..] [0x4c79]
[FSCK] valid_inode_count matched with CP [Ok..] [0xb46]
[FSCK] free segment_count matched with CP [Ok..] [0x9d]
[FSCK] next block offset is free [Ok..]
[FSCK] fixing SIT types
[FSCK] other corrupted bugs [Ok..]
Error: Could not conduct fsync!!!
Generic/019 will trigger fsync() on a clean image, but it will fail
due to simulated failure on disk, result in testcase failure.
Let's add c.need_fsync to record dirty status of image, and only trigger
fsync() when there is dirty data in image.
Signed-off-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
As Kane Ch'in reported in bugzilla [1]
I am using some Debian VMs with f2fs root partition for development.
My host machine crashed for some reason and this caused the f2fs
partitions in the VMs to become corrupted. I tried to boot from Debian
Live and repair the partitions but failed.
do_record_fsync_data: [node] ino = 24573, nid = 0, blkaddr = 2063580
recover_data: ino = 24573, nid = 0, recorded = 0, err = 0
do_record_fsync_data: [node] ino = 471286, nid = 0, blkaddr = 2063581
recover_data: ino = 471286, nid = 0, recorded = 0, err = 0
[ASSERT] (do_record_fsync_data:3475) 0
During do_record_fsync_data(), if dnode in warm node chain is valid in
SIT table, it's better to continue checking rather than triggering
assert().
[1] https://bugzilla.kernel.org/show_bug.cgi?id=218349
Reported-by: Kane Ch'in <qinfd2023@lzu.edu.cn>
Signed-off-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Call path:
fsck_chk_orphan_node
fsck_chk_node_blk
fsck_chk_inode_blk
'F2FS_FT_ORPHAN' will pass to fsck_chk_inode_blk(). If the orphan inode
is a DIR, it will be wrongly fixed.
Fixes: 8fd836f ("fsck: clear unexpected casefold flags")
Signed-off-by: Wu Bo <bo.wu@vivo.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Describes the block size option -b. This sets the block size, which is
restricted to powers of 2. The linux kernel currently requires that the
block size be equal to the page size to mount.
Signed-off-by: Daniel Rosenberg <drosen@google.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Do not finishing zones for current zones.
Signed-off-by: Daeho Jeong <daehojeong@google.com>
Fixes: 06a25b021d ("f2fs-tools: make six open zone check resilient")
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
