Commit Graph

39 Commits

Author SHA1 Message Date
Chao Yu 32f9bc25cb f2fs: support ->rename2()
Now new interface ->rename2() is added to VFS, here are related description:
https://lkml.org/lkml/2014/2/7/873
https://lkml.org/lkml/2014/2/7/758

This patch adds function f2fs_rename2() to support ->rename2() including
handling both RENAME_EXCHANGE and RENAME_NOREPLACE flag.

Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2014-07-25 08:14:08 -07:00
Chao Yu 1256010ab1 f2fs: reduce region of f2fs_lock_op covered for better concurrency
In our rename process, region of f2fs_lock_op covered is too big as some of the
code like f2fs_empty_dir/f2fs_find_entry are not needed to protect by this lock.

So in the extreme case like doing checkpoint when we rename old inode to exist
inode in a large directory could cause lower concurrency.

Let's reduce the region of f2fs_lock_op to fix this.

Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2014-07-09 14:04:25 -07:00
Jaegeuk Kim a014e037be f2fs: clean up an unused parameter and assignment
This patch cleans up simple unnecessary codes.

Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2014-07-09 14:04:25 -07:00
Jaegeuk Kim b97a9b5da8 f2fs: introduce f2fs_do_tmpfile for code consistency
This patch adds f2fs_do_tmpfile to eliminate the redundant init_inode_metadata
flow.
Throught this, we can provide the consistent lock usage, e.g., fi->i_sem,  and
this will enable better debugging stuffs.

Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2014-07-09 14:04:24 -07:00
Chao Yu 50732df02e f2fs: support ->tmpfile()
Add function f2fs_tmpfile() to support O_TMPFILE file creation, and modify logic
of init_inode_metadata to enable linkat temp file.

Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2014-07-09 14:04:24 -07:00
Jaegeuk Kim b2c0829912 f2fs: do checkpoint for the renamed inode
If an inode is renamed, it should be registered as file_lost_pino to conduct
checkpoint at f2fs_sync_file.
Otherwise, the inode cannot be recovered due to no dent_mark in the following
scenario.

Note that, this scenario is from xfstests/322.

1. create "a"
2. fsync "a"
3. rename "a" to "b"
4. fsync "b"
5. Sudden power-cut

After recovery is done, "b" should be seen.
However, the result shows "a", since the recovery procedure does not enter
recover_dentry due to no dent_mark.

The reason is like below.
- The nid of "a" is checkpointed during #2, f2fs_sync_file.
- The inode page for "b" produced by #3 is written without dent_mark by
sync_node_pages.

So, this patch fixes this bug by assinging file_lost_pino to the "a"'s inode.
If the pino is lost, f2fs_sync_file conducts checkpoint, and then recovers
the latest pino and its dentry information for further recovery.

Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2014-07-09 05:59:31 -07:00
Chao Yu dd4d961fe7 f2fs: release new entry page correctly in error path of f2fs_rename
This patch correct releasing code of new_page to avoid BUG_ON in error patch of
f2fs_rename.

Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
2014-07-09 05:59:11 -07:00
Chao Yu 70ff5dfeb6 f2fs: use inode_init_owner() to simplify codes
This patch uses exported inode_init_owner() to simplify codes in
f2fs_new_inode().

Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2014-05-08 18:23:21 +09:00
Chao Yu 48b230a583 f2fs: fix wrong statistics of inline data
If we remove a file that has inline data after mount, our statistics turns to
inaccurate.

cat /sys/kernel/debug/f2fs/status
  - Inline_data Inode: 4294967295

Let's add stat_inc_inline_inode() to stat inline info of the file when lookup.

Change log from v1:
 o stat in f2fs_lookup() instead of in do_read_inode() for excluding wrong stat.

Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2014-04-07 12:40:58 +09:00
Jaegeuk Kim d928bfbfe7 f2fs: introduce fi->i_sem to protect fi's info
This patch introduces fi->i_sem to protect fi's info that includes xattr_ver,
pino, i_nlink.
This enables to remove i_mutex during f2fs_sync_file, resulting in performance
improvement when a number of fsync calls are triggered from many concurrent
threads.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2014-03-20 22:10:11 +09:00
Linus Torvalds bf3d846b78 Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull vfs updates from Al Viro:
 "Assorted stuff; the biggest pile here is Christoph's ACL series.  Plus
  assorted cleanups and fixes all over the place...

  There will be another pile later this week"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (43 commits)
  __dentry_path() fixes
  vfs: Remove second variable named error in __dentry_path
  vfs: Is mounted should be testing mnt_ns for NULL or error.
  Fix race when checking i_size on direct i/o read
  hfsplus: remove can_set_xattr
  nfsd: use get_acl and ->set_acl
  fs: remove generic_acl
  nfs: use generic posix ACL infrastructure for v3 Posix ACLs
  gfs2: use generic posix ACL infrastructure
  jfs: use generic posix ACL infrastructure
  xfs: use generic posix ACL infrastructure
  reiserfs: use generic posix ACL infrastructure
  ocfs2: use generic posix ACL infrastructure
  jffs2: use generic posix ACL infrastructure
  hfsplus: use generic posix ACL infrastructure
  f2fs: use generic posix ACL infrastructure
  ext2/3/4: use generic posix ACL infrastructure
  btrfs: use generic posix ACL infrastructure
  fs: make posix_acl_create more useful
  fs: make posix_acl_chmod more useful
  ...
2014-01-28 08:38:04 -08:00
Christoph Hellwig a6dda0e63e f2fs: use generic posix ACL infrastructure
f2fs has some weird mode bit handling, so still using the old
chmod code for now.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2014-01-25 23:58:19 -05:00
Jaegeuk Kim a18ff06340 f2fs: call mark_inode_dirty to flush dirty pages
If a dentry page is updated, we should call mark_inode_dirty to add the inode
into the dirty list, so that its dentry pages are flushed to the disk.
Otherwise, the inode can be evicted without flush.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2014-01-22 18:40:34 +09:00
Chao Yu deead09009 f2fs: avoid to set wrong pino of inode when rename dir
When we rename a dir to new name which is not exist previous,
we will set pino of parent inode with ino of child inode in f2fs_set_link.
It destroy consistency of pino, it should be fixed.

Thanks for previous work of Shu Tan.

Signed-off-by: Shu Tan <shu.tan@samsung.com>
Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-12-23 10:42:51 +09:00
Jaegeuk Kim ccaaca2591 f2fs: fix writing incorrect orphan blocks
Previously, there was a erroneous scenario like below.
thread 1:                       thread 2:
 f2fs_unlink
  - acquire_orphan_inode
    : sbi->n_orphans++           write_checkpoint
                                 - block_operations
                                  : f2fs_lock_all
                                 - do_checkpoint
                                  : write orphan blocks with sbi->n_orphans
                                 - unblock_operations
  - f2fs_lock_op
  - release_orphan_inode
  - f2fs_unlock_op

During the checkpoint by thread 2, f2fs stores a wrong orphan block according
to the wrong sbi->n_orphans.
To avoid this, simply we should make cover acquire_orphan_inode too with
f2fs_lock_op.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-10-08 10:19:28 +09:00
Gu Zheng e479556bfd f2fs: use rw_sem instead of fs_lock(locks mutex)
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>
2013-10-07 11:33:05 +09:00
Gu Zheng 749ebfd174 f2fs: use strncasecmp() simplify the string comparison
Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-08-27 21:50:12 +09:00
Jaegeuk Kim 8cb8268809 f2fs: fix omitting to update inode page
The f2fs_set_link updates its parent inode number, so we should sync this to
the inode block.
Otherwise, the data can be lost after sudden-power-off.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-08-27 21:49:04 +09:00
Jaegeuk Kim cbd56e7d20 f2fs: fix handling orphan inodes
This patch fixes mishandling of the sbi->n_orphans variable.

If users request lots of f2fs_unlink(), check_orphan_space() could be contended.
In such the case, sbi->n_orphans can be read incorrectly so that f2fs_unlink()
would fall into the wrong state which results in the failure of
add_orphan_inode().

So, let's increment sbi->n_orphans virtually prior to the actual orphan inode
stuffs. After that, let's release sbi->n_orphans by calling release_orphan_inode
or remove_orphan_inode.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-30 15:17:03 +09:00
Jaegeuk Kim 1cd14cafc6 f2fs: update file name in the inode block during f2fs_rename
The error is reproducible by:
0. mkfs.f2fs /dev/sdb1 & mount
1. touch test1
2. touch test2
3. mv test1 test2
4. umount
5. dumpt.f2fs -i 4 /dev/sdb1

After this, when we retrieve the inode->i_name of test2 by dump.f2fs, we get
test1 instead of test2.
This is because f2fs didn't update the file name during the f2fs_rename.

So, this patch fixes that.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-07-30 15:17:03 +09:00
Jaegeuk Kim 354a3399dc f2fs: recover wrong pino after checkpoint during fsync
If a file is linked, f2fs loose its parent inode number so that fsync calls
for the linked file should do checkpoint all the time.
But, if we can recover its parent inode number after the checkpoint, we can
adjust roll-forward mechanism for the further fsync calls, which is able to
improve the fsync performance significatly.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-06-14 09:04:45 +09:00
Jaegeuk Kim 2d4d9fb591 f2fs: fix i_blocks translation on various types of files
Basically an inode manages the number of allocated blocks with inode->i_blocks
which is represented in a unit of sectors, not file system blocks.
But, f2fs has used i_blocks in a unit of file system blocks, and f2fs_getattr
translates it to the number of sectors when fstat is called.

However, previously f2fs_file_inode_operations only has this, so this patch adds
it to all the types of inode_operations.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-06-11 16:01:09 +09:00
Jaegeuk Kim 83d5d6f66b f2fs: cover cp_file information with ilock
If a file is linked with other files, it should be checkpointed at every fsync
calls.
For this, we use set_cp_file() with FADVISE_CP_BIT, but previously we didn't
cover the flag by the global lock.
This patch fixes that the inode page stores this correctly.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-05-28 15:03:06 +09:00
Jaegeuk Kim 6f6fd833e1 f2fs: use ihold
Use the following helper function committed by Al.

commit 7de9c6ee3e
Author: Al Viro <viro@zeniv.linux.org.uk>
Date:   Sat Oct 23 11:11:40 2010 -0400

    new helper: ihold()

...

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-05-28 15:03:04 +09:00
Jaegeuk Kim 93ff10d690 f2fs: should not make_bad_inode on f2fs_link failure
If -ENOSPC is met during f2fs_link, we should not make the inode as bad.
The inode is still alive.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-05-28 15:03:04 +09:00
Jaegeuk Kim 0a364af18f f2fs: remove unnecessary por_doing check
This por_doing check is totally not related to the recovery process.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-05-28 15:03:01 +09:00
Jaegeuk Kim 531ad7d58c f2fs: avoid deadlock during evict after f2fs_gc
o Deadlock case #1

Thread 1:
- writeback_sb_inodes
 - do_writepages
  - f2fs_write_data_pages
   - write_cache_pages
    - f2fs_write_data_page
     - f2fs_balance_fs
      - wait mutex_lock(gc_mutex)

Thread 2:
- f2fs_balance_fs
 - mutex_lock(gc_mutex)
 - f2fs_gc
  - f2fs_iget
   - wait iget_locked(inode->i_lock)

Thread 3:
- do_unlinkat
 - iput
  - lock(inode->i_lock)
   - evict
    - inode_wait_for_writeback

o Deadlock case #2

Thread 1:
- __writeback_single_inode
 : set I_SYNC
  - do_writepages
   - f2fs_write_data_page
    - f2fs_balance_fs
     - f2fs_gc
      - iput
       - evict
        - inode_wait_for_writeback(I_SYNC)

In order to avoid this, even though iput is called with the zero-reference
count, we need to stop the eviction procedure if the inode is on writeback.
So this patch links f2fs_drop_inode which checks the I_SYNC flag.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-05-08 19:54:08 +09:00
Jaegeuk Kim d70b4f53b9 f2fs: add a tracepoint on f2fs_new_inode
This can help when debugging the free nid allocation flows.

Reviewed-by: Namjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-04-29 10:52:01 +09:00
Namjae Jeon a2a4a7e4ab f2fs: add tracepoints for sync & inode operations
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>
2013-04-23 15:30:27 +09:00
Namjae Jeon e66509f03e f2fs: make is_multimedia_file code align with its name
The code conditions put inside the function is_multimedia_file are
reverse to the name i.e, we need to negate the return to actually
check if the file is a multimedia file. So, change the code and usage
path to align both the name and comparision conditions.

Signed-off-by: Namjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: Amit Sahrawat <a.sahrawat@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-04-23 08:56:21 +09:00
Jaegeuk Kim 399368372e f2fs: introduce a new global lock scheme
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>
2013-04-09 18:21:18 +09:00
Jaegeuk Kim 953a3e27e1 f2fs: fix to give correct parent inode number for roll forward
When we recover fsync'ed data after power-off-recovery, we should guarantee
that any parent inode number should be correct for each direct inode blocks.

So, let's make the following rules.

- The fsync should do checkpoint to all the inodes that were experienced hard
links.

- So, the only normal files can be recovered by roll-forward.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-03-27 09:16:25 +09:00
Jaegeuk Kim 5a20d339c7 f2fs: align f2fs maximum name length to linux based filesystem
The maximum filename length supported in linux is 255 characters.
So let's follow that.

Signed-off-by: Namjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: Amit Sahrawat <a.sahrawat@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2013-03-18 21:00:35 +09:00
Leon Romanovsky 9836b8b949 f2fs: unify string length declarations and usage
This patch is intended to unify string length declarations and usage.
There are number of calls to strlen which return size_t object.
The size of this object depends on compiler if it will be bigger,
equal or even smaller than an unsigned int

Signed-off-by: Leon Romanovsky <leon@leon.nu>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2012-12-28 11:27:53 +09:00
Jaegeuk Kim 1efef83202 f2fs: do f2fs_balance_fs in front of dir operations
In order to conserve free sections to deal with the worst-case scenarios, f2fs
should be able to freeze all the directory operations especially when there are
not enough free sections. The f2fs_balance_fs() is for this use.

When FS utilization becomes almost 100%, directory operations can be failed due
to -ENOSPC frequently, which produces some dirty node pages occasionally.

Previously, in such a case, f2fs_balance_fs() is not able to be triggered since
it is triggered only if the directory operation ends up with success.

So, this patch triggers f2fs_balance_fs() at first before handling directory
operations.

Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2012-12-26 10:39:52 +09:00
Namjae Jeon a0d42539e1 f2fs: make use of GFP_F2FS_ZERO for setting gfp_mask
Since, GFP_NOFS and __GFP_ZERO is being used to set gfp_mask.
We can instead make use of already predefined macro GFP_F2FS_ZERO.

Signed-off-by: Namjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: Amit Sahrawat <a.sahrawat@samsung.com>
2012-12-11 13:43:44 +09:00
Namjae Jeon 61412b64b9 f2fs: move error condition for mkdir at proper place
In function f2fs_mkdir, err is being initialized without even checking
if there was any error in new inode creation. So, instead check the
inode error and make use of error/return condition.

Signed-off-by: Namjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: Amit Sahrawat <a.sahrawat@samsung.com>
2012-12-11 13:43:44 +09:00
Jaegeuk Kim 0a8165d7c2 f2fs: adjust kernel coding style
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>
2012-12-11 13:43:42 +09:00
Jaegeuk Kim 57397d86c6 f2fs: add inode operations for special inodes
This adds inode operations for directory, symlink, and special inodes.

Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
2012-12-11 13:43:41 +09:00