Skip to content

L
linux
Switch branch/tag
  1. 01 Jul, 2013 7 commits
    • Jaegeuk Kim's avatar
      f2fs: fix to recover i_size from roll-forward · a1dd3c13
      Jaegeuk Kim authored 
      If user requests many data writes and fsync together, the last updated i_size
should be stored to the inode block consistently.

But, previous write_end just marks the inode as dirty and doesn't update its
metadata into its inode block.
After that, fsync just writes the inode block with newly updated data index
excluding inode metadata updates.

So, this patch introduces write_end in which updates inode block too when the
i_size is changed.
Signed-off-by: default avatarJaegeuk Kim <jaegeuk.kim@samsung.com>
      a1dd3c13
    • Gu Zheng's avatar
      f2fs: remove the unused argument "sbi" of func destroy_fsync_dnodes() · 5ebefc5b
      Gu Zheng authored 
      As destroy_fsync_dnodes() is a simple list-cleanup func, so delete the unused
and unrelated f2fs_sb_info argument of it.
Signed-off-by: default avatarGu Zheng <guz.fnst@cn.fujitsu.com>
Signed-off-by: default avatarJaegeuk Kim <jaegeuk.kim@samsung.com>
      5ebefc5b
    • Jaegeuk Kim's avatar
      f2fs: remove reusing any prefree segments · 763bfe1b
      Jaegeuk Kim authored 
      This patch removes check_prefree_segments initially designed to enhance the
performance by narrowing the range of LBA usage across the whole block device.

When allocating a new segment, previous f2fs tries to find proper prefree
segments, and then, if finds a segment, it reuses the segment for further
data or node block allocation.

However, I found that this was totally wrong approach since the prefree segments
have several data or node blocks that will be used by the roll-forward mechanism
operated after sudden-power-off.

Let's assume the following scenario.

/* write 8MB with fsync */
for (i = 0; i < 2048; i++) {
	offset = i * 4096;
	write(fd, offset, 4KB);
	fsync(fd);
}

In this case, naive segment allocation sequence will be like:
 data segment: x, x+1, x+2, x+3
 node segment: y, y+1, y+2, y+3.

But, if we can reuse prefree segments, the sequence can be like:
 data segment: x, x+1, y, y+1
 node segment: y, y+1, y+2, y+3.
Because, y, y+1, and y+2 became prefree segments one by one, and those are
reused by data allocation.

After conducting this workload, we should consider how to recover the latest
inode with its data.
If we reuse the prefree segments such as y or y+1, we lost the old node blocks
so that f2fs even cannot start roll-forward recovery.

Therefore, I suggest that we should remove reusing prefree segments.
Signed-off-by: default avatarJaegeuk Kim <jaegeuk.kim@samsung.com>
      763bfe1b
    • Gu Zheng's avatar
      f2fs: code cleanup and simplify in func {find/add}_gc_inode · 6cc4af56
      Gu Zheng authored 
      This patch simplifies list operations in find_gc_inode and add_gc_inode.
Just simple code cleanup.
Signed-off-by: default avatarGu Zheng <guz.fnst@cn.fujitsu.com>
[Jaegeuk Kim: add description]
Signed-off-by: default avatarJaegeuk Kim <jaegeuk.kim@samsung.com>
      6cc4af56
    • Namjae Jeon's avatar
      f2fs: optimize the init_dirty_segmap function · 8736fbf0
      Namjae Jeon authored 
      Optimize the while loop condition

Since this condition will always be true and while loop will
be terminated by the following condition in code:

if (segno >= TOTAL_SEGS(sbi))
    break;
Hence we can replace the while loop condition with while(1)
instead of always checking for segno to be less than Total segs.

Also we do not need to use TOTAL_SEGS() everytime. We can store
this value in a local variable since this value is constant.
Signed-off-by: default avatarNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: default avatarPankaj Kumar <pankaj.km@samsung.com>
Signed-off-by: default avatarJaegeuk Kim <jaegeuk.kim@samsung.com>
      8736fbf0
    • Jaegeuk Kim's avatar
      f2fs: fix an endian conversion bug detected by sparse · 060dd67b
      Jaegeuk Kim authored 
      This patch should fix the following bug reported by kbuild test robot.

fs/f2fs/recovery.c:233:33: sparse: incorrect type in assignment
(different base types)

parse warnings: (new ones prefixed by >>)

>> recovery.c:233: sparse: incorrect type in assignment (different base types)
   recovery.c:233:    expected unsigned int [unsigned] [assigned] ofs_in_node
   recovery.c:233:    got restricted __le16 [assigned] [usertype] ofs_in_node
>> recovery.c:238: sparse: incorrect type in assignment (different base types)
   recovery.c:238:    expected unsigned int [unsigned] ofs_in_node
   recovery.c:238:    got restricted __le16 [assigned] [usertype] ofs_in_node
Signed-off-by: default avatarJaegeuk Kim <jaegeuk.kim@samsung.com>
      060dd67b
    • Jaegeuk Kim's avatar
      f2fs: fix crc endian conversion · 7e586fa0
      Jaegeuk Kim authored 
      While calculating CRC for the checkpoint block, we use __u32, but when storing
the crc value to the disk, we use __le32.

Let's fix the inconsistency.
Reported-and-Tested-by: default avatarOded Gabbay <ogabbay@advaoptical.com>
Signed-off-by: default avatarJaegeuk Kim <jaegeuk.kim@samsung.com>
      7e586fa0
  3. 17 Jun, 2013 1 commit
    • Namjae Jeon's avatar
      f2fs: add remount_fs callback support · 696c018c
      Namjae Jeon authored 
      Add the f2fs_remount function call which will be used
during the filesystem remounting. This function
will help us to change the mount options specific to
f2fs.

Also modify the f2fs background_gc mount option, which
will allow the user to dynamically trun on/off the
garbage collection in f2fs based on the background_gc
value. If background_gc=on, Garbage collection will
be turned off & if background_gc=off, Garbage collection
will be truned on.

By default the garbage collection is on in f2fs.

Change Log:
v2: Incorporated the review comments by Gu Zheng.
    Removing the restore part for VFS flags
    Updating comments with proper flag conditions
    Display GC background option as ON/OFF
    Revised conditions to stop GC in case of remount

v1: Initial changes for adding remount_fs callback
support.

Cc: Gu Zheng <guz.fnst@cn.fujitsu.com>
Signed-off-by: default avatarNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: default avatarPankaj Kumar <pankaj.km@samsung.com>
Reviewed-by: default avatarGu Zheng <guz.fnst@cn.fujitsu.com>
[Jaegeuk Kim: change /** with /* for the coding style]
Signed-off-by: default avatarJaegeuk Kim <jaegeuk.kim@samsung.com>
      696c018c
  5. 14 Jun, 2013 7 commits
  7. 11 Jun, 2013 4 commits
    • Jaegeuk Kim's avatar
      f2fs: sync dir->i_size with its block allocation · 699489bb
      Jaegeuk Kim authored 
      If new dentry block is allocated and its i_size is updated, we should update
its inode block together in order to sync i_size and its block allocation.
Otherwise, we can loose additional dentry block due to the unconsistent i_size.

Errorneous Scenario
-------------------

In the recovery routine,
 - recovery_dentry
 | - __f2fs_add_link
 | | - get_new_data_page
 | | | - i_size_write(new_i_size)
 | | | - mark_inode_dirty_sync(dir)
 | | - update_parent_metadata
 | | | - mark_inode_dirty(dir)
 |
 - write_checkpoint
   - sync_dirty_dir_inodes
     - filemap_flush(dentry_blocks)
       - f2fs_write_data_page
         - skip to write the last dentry block due to index < i_size

In the above flow, new_i_size is not updated to its inode block so that the
last dentry block will be lost accordingly.
Signed-off-by: default avatarJaegeuk Kim <jaegeuk.kim@samsung.com>
      699489bb
    • Jaegeuk Kim's avatar
      f2fs: fix i_blocks translation on various types of files · 2d4d9fb5
      Jaegeuk Kim authored 
      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: default avatarJaegeuk Kim <jaegeuk.kim@samsung.com>
      2d4d9fb5
    • Gu Zheng's avatar
      f2fs: set sb->s_fs_info before calling parse_options() · 5fb08372
      Gu Zheng authored 
      In f2fs_fill_super(), set sb->s_fs_info before calling parse_options(), then we can get
f2fs_sb_info via F2FS_SB(sb) in parse_options().
So that the second argument "sbi" of func parse_options() is no longer needed.
Signed-off-by: default avatarGu Zheng <guz.fnst@cn.fujitsu.com>
Signed-off-by: default avatarJaegeuk Kim <jaegeuk.kim@samsung.com>
      5fb08372
    • Jaegeuk Kim's avatar
      f2fs: support xattr security labels · 8ae8f162
      Jaegeuk Kim authored 
      This patch adds the support of security labels for f2fs, which will be used
by Linus Security Models (LSMs).

Quote from http://en.wikipedia.org/wiki/Linux_Security_Modules:
"Linux Security Modules (LSM) is a framework that allows the Linux kernel to
support a variety of computer security models while avoiding favoritism toward
any single security implementation. The framework is licensed under the terms of
the GNU General Public License and is standard part of the Linux kernel since
Linux 2.6. AppArmor, SELinux, Smack and TOMOYO Linux are the currently accepted
modules in the official kernel.".
Signed-off-by: default avatarJaegeuk Kim <jaegeuk.kim@samsung.com>
      8ae8f162
  9. 07 Jun, 2013 1 commit
    • Jaegeuk Kim's avatar
      f2fs: fix iget/iput of dir during recovery · 5deb8267
      Jaegeuk Kim authored 
      It is possible that iput is skipped after iget during the recovery.

In recover_dentry(),
 dir = f2fs_iget();
 ...
 if (de && inode->i_ino == le32_to_cpu(de->ino))
	goto out;

In this case, this dir is not able to be added in dirty_dir_inode_list.
The actual linking is done only when set_page_dirty() is called.

So let's add this newly got inode into the list explicitly, and put it at the
end of the recovery routine.
Signed-off-by: default avatarJaegeuk Kim <jaegeuk.kim@samsung.com>
      5deb8267
  11. 06 Jun, 2013 1 commit
    • Namjae Jeon's avatar
      f2fs: reorganise the function get_victim_by_default · b2b3460a
      Namjae Jeon authored 
      Fix the function get_victim_by_default, where it checks
for the condition  that p.min_segno != NULL_SEGNO as
shown:

if (p.min_segno != NULL_SEGNO)
           goto got_it;

and if above condition is true then

got_it:
        if (p.min_segno != NULL_SEGNO) {

So this condition is being checked twice. Hence move the goto
statement after the if condition so that duplication of condition
check is avoided.

Also this function makes a call to get_max_cost() to compute
the max cost based on the f2fs_sbi_info and victim policy. Since
get_max_cost depends on on three parameters of victim_sel_policy
=> alloc_mode, gc_mode & ofs_unit, once this victim policy is
initialised, these value will not change till the execution
time of get_victim_by_default() & also f2fs_sbi_info structure
parameters will not change.

Hence making calls to get_max_cost() in while loop does not seems to
be a good point. Instead we can call it once in begining and store
the results in local variable, which later can serve our purpose
for comparing the cost with max cost inside the while loop.
Signed-off-by: default avatarNamjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: default avatarPankaj Kumar <pankaj.km@samsung.com>
Signed-off-by: default avatarJaegeuk Kim <jaegeuk.kim@samsung.com>
      b2b3460a
  13. 03 Jun, 2013 1 commit
  15. 28 May, 2013 18 commits
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7