This patch-set includes the following major enhancement patches.
o introduce large directory support
o introduce f2fs_issue_flush to merge redundant flush commands
o merge write IOs as much as possible aligned to the segment
o add sysfs entries to tune the f2fs configuration
o use radix_tree for the free_nid_list to reduce in-memory operations
o remove costly bit operations in f2fs_find_entry
o enhance the readahead flow for CP/NAT/SIT/SSA blocks
The other bug fixes are as follows.
o recover xattr node blocks correctly after sudden-power-cut
o fix to calculate the maximum number of node ids
o enhance to handle many error cases
And, there are a bunch of cleanups.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)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=eSW+
-----END PGP SIGNATURE-----
Merge tag 'for-f2fs-3.15' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs
Pull f2fs updates from Jaegeuk Kim:
"This patch-set includes the following major enhancement patches.
- introduce large directory support
- introduce f2fs_issue_flush to merge redundant flush commands
- merge write IOs as much as possible aligned to the segment
- add sysfs entries to tune the f2fs configuration
- use radix_tree for the free_nid_list to reduce in-memory operations
- remove costly bit operations in f2fs_find_entry
- enhance the readahead flow for CP/NAT/SIT/SSA blocks
The other bug fixes are as follows:
- recover xattr node blocks correctly after sudden-power-cut
- fix to calculate the maximum number of node ids
- enhance to handle many error cases
And, there are a bunch of cleanups"
* tag 'for-f2fs-3.15' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (62 commits)
f2fs: fix wrong statistics of inline data
f2fs: check the acl's validity before setting
f2fs: introduce f2fs_issue_flush to avoid redundant flush issue
f2fs: fix to cover io->bio with io_rwsem
f2fs: fix error path when fail to read inline data
f2fs: use list_for_each_entry{_safe} for simplyfying code
f2fs: avoid free slab cache under spinlock
f2fs: avoid unneeded lookup when xattr name length is too long
f2fs: avoid unnecessary bio submit when wait page writeback
f2fs: return -EIO when node id is not matched
f2fs: avoid RECLAIM_FS-ON-W warning
f2fs: skip unnecessary node writes during fsync
f2fs: introduce fi->i_sem to protect fi's info
f2fs: change reclaim rate in percentage
f2fs: add missing documentation for dir_level
f2fs: remove unnecessary threshold
f2fs: throttle the memory footprint with a sysfs entry
f2fs: avoid to drop nat entries due to the negative nr_shrink
f2fs: call f2fs_wait_on_page_writeback instead of native function
f2fs: introduce nr_pages_to_write for segment alignment
...
Reclaim will be leaving shadow entries in the page cache radix tree upon
evicting the real page. As those pages are found from the LRU, an
iput() can lead to the inode being freed concurrently. At this point,
reclaim must no longer install shadow pages because the inode freeing
code needs to ensure the page tree is really empty.
Add an address_space flag, AS_EXITING, that the inode freeing code sets
under the tree lock before doing the final truncate. Reclaim will check
for this flag before installing shadow pages.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Rik van Riel <riel@redhat.com>
Reviewed-by: Minchan Kim <minchan@kernel.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Bob Liu <bob.liu@oracle.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Greg Thelen <gthelen@google.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jan Kara <jack@suse.cz>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Luigi Semenzato <semenzato@google.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Metin Doslu <metin@citusdata.com>
Cc: Michel Lespinasse <walken@google.com>
Cc: Ozgun Erdogan <ozgun@citusdata.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Roman Gushchin <klamm@yandex-team.ru>
Cc: Ryan Mallon <rmallon@gmail.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch introduces an i_dir_level field to support large directory.
Previously, f2fs maintains multi-level hash tables to find a dentry quickly
from a bunch of chiild dentries in a directory, and the hash tables consist of
the following tree structure as below.
In Documentation/filesystems/f2fs.txt,
----------------------
A : bucket
B : block
N : MAX_DIR_HASH_DEPTH
----------------------
level #0 | A(2B)
|
level #1 | A(2B) - A(2B)
|
level #2 | A(2B) - A(2B) - A(2B) - A(2B)
. | . . . .
level #N/2 | A(2B) - A(2B) - A(2B) - A(2B) - A(2B) - ... - A(2B)
. | . . . .
level #N | A(4B) - A(4B) - A(4B) - A(4B) - A(4B) - ... - A(4B)
But, if we can guess that a directory will handle a number of child files,
we don't need to traverse the tree from level #0 to #N all the time.
Since the lower level tables contain relatively small number of dentries,
the miss ratio of the target dentry is likely to be high.
In order to avoid that, we can configure the hash tables sparsely from level #0
like this.
level #0 | A(2B) - A(2B) - A(2B) - A(2B)
level #1 | A(2B) - A(2B) - A(2B) - A(2B) - A(2B) - ... - A(2B)
. | . . . .
level #N/2 | A(2B) - A(2B) - A(2B) - A(2B) - A(2B) - ... - A(2B)
. | . . . .
level #N | A(4B) - A(4B) - A(4B) - A(4B) - A(4B) - ... - A(4B)
With this structure, we can skip the ineffective tree searches in lower level
hash tables.
This patch adds just a facility for this by introducing i_dir_level in
f2fs_inode.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
In order to make fs consistency, update_inode_page should not be failed all
the time. Otherwise, it is possible to lose some metadata in the inode like
a link count.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Fixed a variety of trivial checkpatch warnings. The only delta should
be some minor formatting on log strings that were split / too long.
Signed-off-by: Chris Fries <cfries@motorola.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
"boo sync" parameter is never referenced in f2fs_wait_on_page_writeback.
We should remove this parameter.
Signed-off-by: Yuan Zhong <yuan.mark.zhong@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch adds the number of inline_data files into the status information.
Note that the number is reset whenever the filesystem is newly mounted.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch introduces F2FS_INODE that returns struct f2fs_inode * from the inode
page.
By using this macro, we can remove unnecessary casting codes like below.
struct f2fs_inode *ri = &F2FS_NODE(inode_page)->i;
-> struct f2fs_inode *ri = F2FS_INODE(inode_page);
Reviewed-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
If you want to remove unnecessary BUG_ONs, you can just turn off F2FS_CHECK_FS
in your kernel config.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
The fs_locks is used to block other ops(ex, recovery) when doing checkpoint.
And each other operate routine(besides checkpoint) needs to acquire a fs_lock,
there is a terrible problem here, if these are too many concurrency threads acquiring
fs_lock, so that they will block each other and may lead to some performance problem,
but this is not the phenomenon we want to see.
Though there are some optimization patches introduced to enhance the usage of fs_lock,
but the thorough solution is using a *rw_sem* to replace the fs_lock.
Checkpoint routine takes write_sem, and other ops take read_sem, so that we can block
other ops(ex, recovery) when doing checkpoint, and other ops will not disturb each other,
this can avoid the problem described above completely.
Because of the weakness of rw_sem, the above change may introduce a potential problem
that the checkpoint thread might get starved if other threads are intensively locking
the read semaphore for I/O.(Pointed out by Xu Jin)
In order to avoid this, a wait_list is introduced, the appending read semaphore ops
will be dropped into the wait_list if checkpoint thread is waiting for write semaphore,
and will be waked up when checkpoint thread gives up write semaphore.
Thanks to Kim's previous review and test, and will be very glad to see other guys'
performance tests about this patch.
V2:
-fix the potential starvation problem.
-use more suitable func name suggested by Xu Jin.
Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
[Jaegeuk Kim: adjust minor coding standard]
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch adds basic inode flags for inline xattrs, F2FS_INLINE_XATTR,
and add a mount option, inline_xattr, which is enabled when xattr is set.
If the mount option is enabled, all the files are marked with the inline_xattrs
flag.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
In f2fs_write_inode, updating inode after f2fs_balance_fs is not
a optimized way in the case that f2fs_gc is performed ahead. The
inode page will be unnecessarily written out twice, one of which
is in f2fs_gc->...->sync_node_pages and the other is in
update_inode_page.
Let's update the inode page in prior to f2fs_balance_fs to avoid
this.
To reproduce it,
$ touch file (before this step, should make the device need f2fs_gc)
$ sync (or wait the bdi to write dirty inode)
Signed-off-by: Jin Xu <jinuxstyle@gmail.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch fixes a deadlock bug that occurs quite often when there are
concurrent write and fsync on a same file.
Following is the simplified call trace when tasks get hung.
fsync thread:
- f2fs_sync_file
...
- f2fs_write_data_pages
...
- update_extent_cache
...
- update_inode
- wait_on_page_writeback
bdi writeback thread
- __writeback_single_inode
- f2fs_write_data_pages
- mutex_lock(sbi->writepages)
The deadlock happens when the fsync thread waits on a inode page that has
been added to the f2fs' cached bio sbi->bio[NODE], and unfortunately,
no one else could be able to submit the cached bio to block layer for
writeback. This is because the fsync thread already hold a sbi->fs_lock and
the sbi->writepages lock, causing the bdi thread being blocked when attempt
to write data pages for the same inode. At the same time, f2fs_gc thread
does not notice the situation and could not help. Even the sync syscall
gets blocked.
To fix it, we could submit the cached bio first before waiting on a inode page
that is being written back.
Signed-off-by: Jin Xu <jinuxstyle@gmail.com>
[Jaegeuk Kim: add more cases to use f2fs_wait_on_page_writeback]
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Introduce help function F2FS_NODE() to simplify the conversion of node_page to
f2fs_node.
Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
If update_inode is called, we don't need to do write_inode.
So, let's use a *dirty* flag for each inode.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
While an orphan inode has zero link_count, f2fs_gc is able to select the inode
for foreground gc.
- f2fs_gc
- do_garbage_collect
- gc_data_segment
: f2fs_iget is failed
: get_valid_blocks() != 0, so that retry
--> here we got the infinite loop.
This patch resolved this issue.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Add tracepoints in f2fs for tracing the syncing
operations like filesystem sync, file sync enter/exit.
It will helf to trace the code under debugging scenarios.
Also add tracepoints for tracing the various inode operations
like building inode, eviction of inode, link/unlike of
inodes.
Signed-off-by: Namjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: Pankaj Kumar <pankaj.km@samsung.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
[Jaegeuk: combine and modify the tracepoint structures]
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
In the previous version, f2fs uses global locks according to the usage types,
such as directory operations, block allocation, block write, and so on.
Reference the following lock types in f2fs.h.
enum lock_type {
RENAME, /* for renaming operations */
DENTRY_OPS, /* for directory operations */
DATA_WRITE, /* for data write */
DATA_NEW, /* for data allocation */
DATA_TRUNC, /* for data truncate */
NODE_NEW, /* for node allocation */
NODE_TRUNC, /* for node truncate */
NODE_WRITE, /* for node write */
NR_LOCK_TYPE,
};
In that case, we lose the performance under the multi-threading environment,
since every types of operations must be conducted one at a time.
In order to address the problem, let's share the locks globally with a mutex
array regardless of any types.
So, let users grab a mutex and perform their jobs in parallel as much as
possbile.
For this, I propose a new global lock scheme as follows.
0. Data structure
- f2fs_sb_info -> mutex_lock[NR_GLOBAL_LOCKS]
- f2fs_sb_info -> node_write
1. mutex_lock_op(sbi)
- try to get an avaiable lock from the array.
- returns the index of the gottern lock variable.
2. mutex_unlock_op(sbi, index of the lock)
- unlock the given index of the lock.
3. mutex_lock_all(sbi)
- grab all the locks in the array before the checkpoint.
4. mutex_unlock_all(sbi)
- release all the locks in the array after checkpoint.
5. block_operations()
- call mutex_lock_all()
- sync_dirty_dir_inodes()
- grab node_write
- sync_node_pages()
Note that,
the pairs of mutex_lock_op()/mutex_unlock_op() and
mutex_lock_all()/mutex_unlock_all() should be used together.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch removes data_version check flow during the fsync call.
The original purpose for the use of data_version was to avoid writng inode
pages redundantly by the fsync calls repeatedly.
However, when user can modify file meta and then call fsync, we should not
skip fsync procedure.
So, let's remove this condition check and hope that user triggers in right
manner.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
In function check_nid_range, there is no need to trigger BUG_ON and make kernel stop.
Instead it could just check and indicate the inode number to be EINVAL.
Update the return path in do_read_inode to use the return from check_nid_range.
Signed-off-by: Namjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: Amit Sahrawat <a.sahrawat@samsung.com>
[Jaegeuk: replace BUG_ON with WARN_ON]
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
1. Background
Previously, if f2fs tries to move data blocks of an *evicting* inode during the
cleaning process, it stops the process incompletely and then restarts the whole
process, since it needs a locked inode to grab victim data pages in its address
space. In order to get a locked inode, iget_locked() by f2fs_iget() is normally
used, but, it waits if the inode is on freeing.
So, here is a deadlock scenario.
1. f2fs_evict_inode() <- inode "A"
2. f2fs_balance_fs()
3. f2fs_gc()
4. gc_data_segment()
5. f2fs_iget() <- inode "A" too!
If step #1 and #5 treat a same inode "A", step #5 would fall into deadlock since
the inode "A" is on freeing. In order to resolve this, f2fs_iget_nowait() which
skips __wait_on_freeing_inode() was introduced in step #5, and stops f2fs_gc()
to complete f2fs_evict_inode().
1. f2fs_evict_inode() <- inode "A"
2. f2fs_balance_fs()
3. f2fs_gc()
4. gc_data_segment()
5. f2fs_iget_nowait() <- inode "A", then stop f2fs_gc() w/ -ENOENT
2. Problem and Solution
In the above scenario, however, f2fs cannot finish f2fs_evict_inode() only if:
o there are not enough free sections, and
o f2fs_gc() tries to move data blocks of the *evicting* inode repeatedly.
So, the final solution is to use f2fs_iget() and remove f2fs_balance_fs() in
f2fs_evict_inode().
The f2fs_evict_inode() actually truncates all the data and node blocks, which
means that it doesn't produce any dirty node pages accordingly.
So, we don't need to do f2fs_balance_fs() in practical.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch supports ioctl FIFREEZE and FITHAW to snapshot filesystem.
Before calling f2fs_freeze, all writers would be suspended and sync_fs
would be completed. So no f2fs has to do something.
Just background gc operation should be skipped due to generate dirty
nodes and data until unfreeze.
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This patch stores inode->i_rdev into on-disk inode structure.
Alun reported that:
aspire tmp # mount -t f2fs /dev/sdb mnt
aspire tmp # mknod mnt/sda1 b 8 1
aspire tmp # mknod mnt/null c 1 3
aspire tmp # mknod mnt/console c 5 1
aspire tmp # ls -l mnt
total 2
crw-r--r-- 1 root root 5, 1 Jan 22 18:44 console
crw-r--r-- 1 root root 1, 3 Jan 22 18:44 null
brw-r--r-- 1 root root 8, 1 Jan 22 18:44 sda1
aspire tmp # umount mnt
aspire tmp # mount -t f2fs /dev/sdb mnt
aspire tmp # ls -l mnt
total 2
crw-r--r-- 1 root root 0, 0 Jan 22 18:44 console
crw-r--r-- 1 root root 0, 0 Jan 22 18:44 null
brw-r--r-- 1 root root 0, 0 Jan 22 18:44 sda1
In this report, f2fs lost the major/minor numbers of device files after umount.
The reason was revealed that f2fs does not store the inode->i_rdev to the
on-disk inode data structure.
So, as the other file systems do, f2fs also stores i_rdev into the i_addr fields
in on-disk inode structure without any on-disk layout changes.
Note that, this bug is limited to device files made by mknod().
Reported-and-Tested-by: Alun Jones <alun.linux@ty-penguin.org.uk>
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
The f2fs_balance_fs() is to check the number of free sections and decide whether
it needs to conduct cleaning or not. If there are not enough free sections, the
cleaning job should be started.
In order to control an amount of free sections even under high utilization, f2fs
should call f2fs_balance_fs at all the VFS interfaces that are able to produce
dirty pages.
This patch adds the function calls in the missing interfaces as follows.
1. f2fs_setxattr()
The f2fs_setxattr() produces dirty node pages so that we should call
f2fs_balance_fs() either likewise doing in other VFS interfaces such as
f2fs_lookup(), f2fs_mkdir(), and so on.
2. f2fs_sync_file()
We should guarantee serving free sections for syncing metadata during fsync.
Previously, there is no space check before triggering checkpoint and
sync_node_pages.
Therefore, if a bunch of fsync calls are triggered under 100% of FS utilization,
f2fs is able to be faced with no free sections, resulting in BUG_ON().
3. f2fs_sync_fs()
Before calling write_checkpoint(), we should guarantee that there are minimum
free sections.
4. f2fs_write_inode()
f2fs_write_inode() is also able to produce dirty node pages.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Ruslan reported that f2fs hangs with an infinite loop in f2fs_sync_file():
while (sync_node_pages(sbi, inode->i_ino, &wbc) == 0)
f2fs_write_inode(inode, NULL);
The reason was revealed that the cold flag is not set even thought this inode is
a normal file. Therefore, sync_node_pages() skips to write node blocks since it
only writes cold node blocks.
The cold flag is stored to the node_footer in node block, and whenever a new
node page is allocated, it is set according to its file type, file or directory.
But, after sudden-power-off, when recovering the inode page, f2fs doesn't recover
its cold flag.
So, let's assign the cold flag in more right places.
One more thing:
If f2fs_write_inode() returns an error due to whatever situations, there would
be no dirty node pages so that sync_node_pages() returns zero.
(i.e., zero means nothing was written.)
Reported-by: Ruslan N. Marchenko <me@ruff.mobi>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Previously, f2fs didn't track the parent inode number correctly which is stored
in each f2fs_inode. In the case of the following scenario, a bug can be occured.
Let's suppose there are one directory, "/b", and two files, "/a" and "/b/a".
- pino of "/a" is ROOT_INO.
- pino of "/b/a" is DIR_B_INO.
Then,
# sync
: The inode pages of "/a" and "/b/a" contain the parent inode numbers as
ROOT_INO and DIR_B_INO respectively.
# mv /a /b/a
: The parent inode number of "/a" should be changed to DIR_B_INO, but f2fs
didn't do that. Ref. f2fs_set_link().
In order to fix this clearly, I added i_pino in f2fs_inode_info, and whenever
it needs to be changed like in f2fs_add_link() and f2fs_set_link(), it is
updated temporarily in f2fs_inode_info.
And later, f2fs_write_inode() stores the latest information to the inode pages.
For power-off-recovery, f2fs_sync_file() triggers simply f2fs_write_inode().
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
As pointed out by Randy Dunlap, this patch removes all usage of "/**" for comment
blocks. Instead, just use "/*".
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
This adds core functions to get, read, write, and evict an inode.
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Linus Torvalds