Skip to content

L
linux-davinci
Switch branch/tag
  1. 04 Jan, 2009 1 commit
  3. 06 Jan, 2009 3 commits
    • Theodore Ts'o's avatar
      ext4: Remove code to create the journal inode · c3191067
      Theodore Ts'o authored 
      This code has been obsolete in quite some time, since the supported
method for adding a journal inode is to use tune2fs (or to creating
new filesystem with a journal via mke2fs or mkfs.ext4).
Signed-off-by: default avatar"Theodore Ts'o" <tytso@mit.edu>
      c3191067
    • Toshiyuki Okajima's avatar
      ext4: provide function to release metadata pages under memory pressure · c39a7f84
      Toshiyuki Okajima authored 
      Pages in the page cache belonging to ext4 data files are released via
the ext4_releasepage() function specified in the ext4 inode's
address_space_ops.  However, metadata blocks (such as indirect blocks,
directory blocks, etc) are managed via the block device
address_space_ops, and they can not be released by
try_to_free_buffers() if they have a journal head attached to them.

To address this, we supply a release_metadata function which calls
jbd2_journal_try_to_free_buffers() function to free the metadata, and
which is called by the block device's blkdev_releasepage() function.
Signed-off-by: default avatarToshiyuki Okajima <toshi.okajima@jp.fujitsu.com>
Signed-off-by: default avatar"Theodore Ts'o" <tytso@mit.edu>
Cc: linux-fsdevel@vger.kernel.org
      c39a7f84
    • Toshiyuki Okajima's avatar
      ext3: provide function to release metadata pages under memory pressure · 6b082b53
      Toshiyuki Okajima authored 
      Pages in the page cache belonging to ext3 data files are released via
the ext3_releasepage() function specified in the ext3 inode's
address_space_ops.  However, metadata blocks (such as indirect blocks,
directory blocks, etc) are managed via the block device
address_space_ops, and they can not be released by
try_to_free_buffers() if they have a journal head attached to them.

To address this, we supply a try_to_free_pages() function which calls
journal_try_to_free_buffers() function to free the metadata, and which
is called by the block device's blkdev_releasepage() function.
Signed-off-by: default avatarToshiyuki Okajima <toshi.okajima@jp.fujitsu.com>
Signed-off-by: default avatar"Theodore Ts'o" <tytso@mit.edu>
Cc: linux-fsdevel@vger.kernel.org
      6b082b53
  5. 03 Jan, 2009 1 commit
  7. 06 Jan, 2009 6 commits
  9. 04 Jan, 2009 1 commit
  11. 06 Jan, 2009 3 commits
    • Aneesh Kumar K.V's avatar
      ext4: Use high 16 bits of the block group descriptor's free counts fields · 560671a0
      Aneesh Kumar K.V authored 
      Rename the lower bits with suffix _lo and add helper
to access the values. Also rename bg_itable_unused_hi
to bg_pad as in e2fsprogs.
Signed-off-by: default avatarAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: default avatar"Theodore Ts'o" <tytso@mit.edu>
      560671a0
    • Aneesh Kumar K.V's avatar
      ext4: Fix race between read_block_bitmap() and mark_diskspace_used() · e8134b27
      Aneesh Kumar K.V authored 
      We need to make sure we update the block bitmap and clear
EXT4_BG_BLOCK_UNINIT flag with sb_bgl_lock held, since
ext4_read_block_bitmap() looks at EXT4_BG_BLOCK_UNINIT to decide
whether to initialize the block bitmap each time it is called
(introduced by commit c806e68f), and this can race with block
allocations in ext4_mb_mark_diskspace_used().

ext4_read_block_bitmap does:

spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group));
if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
	ext4_init_block_bitmap(sb, bh, block_group, desc);

Now on the block allocation side we do

mb_set_bits(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group), bitmap_bh->b_data,
			ac->ac_b_ex.fe_start, ac->ac_b_ex.fe_len);
....
spin_lock(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group));
if (gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
	gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT);

ie on allocation we update the bitmap then we take the sb_bgl_lock
and clear the EXT4_BG_BLOCK_UNINIT flag. What can happen is a
parallel ext4_read_block_bitmap can zero out the bitmap in between
the above mb_set_bits and spin_lock(sb_bg_lock..)

The race results in below user visible errors
EXT4-fs error (device sdb1): ext4_mb_release_inode_pa: free 100, pa_free 105
EXT4-fs error (device sdb1): mb_free_blocks: double-free of inode 0's block ..
Signed-off-by: default avatarAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: default avatar"Theodore Ts'o" <tytso@mit.edu>
Cc: stable@kernel.org
      e8134b27
    • Aneesh Kumar K.V's avatar
      ext4: fix BUG when calling ext4_error with locked block group · 5d1b1b3f
      Aneesh Kumar K.V authored 
      The mballoc code likes to call ext4_error while it is holding locked
block groups.  This can causes a scheduling in atomic context BUG.  We
can't just unlock the block group and relock it after/if ext4_error
returns since that might result in race conditions in the case where
the filesystem is set to continue after finding errors.
Signed-off-by: default avatarAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: default avatar"Theodore Ts'o" <tytso@mit.edu>
      5d1b1b3f
  13. 24 Nov, 2008 1 commit
    • Aneesh Kumar K.V's avatar
      ext4: Fix lockdep recursive locking warning · b7be019e
      Aneesh Kumar K.V authored 
      In ext4_mb_init_group(), if the filesystem block size is less than
PAGE_SIZE/2, the code tries to grab alloc_sem for multiple block
groups in a loop.  We need to allow for this by using
down_write_nested() and passing in the loop index as a lock subclass
number.  This works because no other code path needs to take multiple
alloc_sem's.  Note that lockdep will fail for filesystem blocksize
smaller than to PAGE_SIZE/16k.  (e.g., a 1k filesystem blocksize with
a 32k page size, or a 2k filesystem blocksize with a 64k blocksize,
etc.)
Signed-off-by: default avatarAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: default avatar"Theodore Ts'o" <tytso@mit.edu>
      b7be019e
  15. 06 Jan, 2009 1 commit
  17. 06 Nov, 2008 1 commit
  19. 25 Nov, 2008 1 commit
  21. 06 Jan, 2009 2 commits
  23. 22 Nov, 2008 1 commit
  25. 05 Nov, 2008 2 commits
  27. 06 Jan, 2009 1 commit
  29. 04 Nov, 2008 1 commit
  31. 04 Jan, 2009 1 commit
  33. 17 Dec, 2008 1 commit
  35. 26 Nov, 2008 1 commit
    • Josef Bacik's avatar
      jbd2: improve jbd2 fsync batching · e07f7183
      Josef Bacik authored 
      This patch removes the static sleep time in favor of a more self
optimizing approach where we measure the average amount of time it
takes to commit a transaction to disk and the ammount of time a
transaction has been running.  If somebody does a sync write or an
fsync() traditionally we would sleep for 1 jiffies, which depending on
the value of HZ could be a significant amount of time compared to how
long it takes to commit a transaction to the underlying storage.  With
this patch instead of sleeping for a jiffie, we check to see if the
amount of time this transaction has been running is less than the
average commit time, and if it is we sleep for the delta using
schedule_hrtimeout to give us a higher precision sleep time.  This
greatly benefits high end storage where you could end up sleeping for
longer than it takes to commit the transaction and therefore sitting
idle instead of allowing the transaction to be committed by keeping
the sleep time to a minimum so you are sure to always be doing
something.
Signed-off-by: default avatarJosef Bacik <jbacik@redhat.com>
Signed-off-by: default avatar"Theodore Ts'o" <tytso@mit.edu>
      e07f7183
  37. 06 Jan, 2009 3 commits
    • Aneesh Kumar K.V's avatar
      ext4: Don't overwrite allocation_context ac_status · 032115fc
      Aneesh Kumar K.V authored 
      We can call ext4_mb_check_limits even after successfully allocating
the requested blocks.  In that case, make sure we don't overwrite
ac_status if it already has the status AC_STATUS_FOUND.  This fixes
the lockdep warning:

=============================================
[ INFO: possible recursive locking detected ]
2.6.28-rc6-autokern1 #1
---------------------------------------------
fsstress/11948 is trying to acquire lock:
 (&meta_group_info[i]->alloc_sem){----}, at: [<c04d9a49>] ext4_mb_load_buddy+0x9f/0x278
.....

stack backtrace:
.....
 [<c04db974>] ext4_mb_regular_allocator+0xbb5/0xd44
.....

but task is already holding lock:
 (&meta_group_info[i]->alloc_sem){----}, at: [<c04d9a49>] ext4_mb_load_buddy+0x9f/0x278
Signed-off-by: default avatarAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: default avatar"Theodore Ts'o" <tytso@mit.edu>
Cc: stable@kernel.org
      032115fc
    • Theodore Ts'o's avatar
      ext4: remove extraneous newlines from calls to ext4_error() and ext4_warning() · fde4d95a
      Theodore Ts'o authored 
      This removes annoying blank syslog entries emitted by ext4_error() or
ext4_warning(), since these functions add their own newline.
Signed-off-by: default avatarNick Warne <nick@ukfsn.org>
Signed-off-by: default avatar"Theodore Ts'o" <tytso@mit.edu>
      fde4d95a
    • Theodore Ts'o's avatar
      jbd2: Add barrier not supported test to journal_wait_on_commit_record · fd98496f
      Theodore Ts'o authored 
      Xen doesn't report that barriers are not supported until buffer I/O is
reported as completed, instead of when the buffer I/O is submitted.
Add a check and a fallback codepath to journal_wait_on_commit_record()
to detect this case, so that attempts to mount ext4 filesystems on
LVM/devicemapper devices on Xen guests don't blow up with an "Aborting
journal on device XXX"; "Remounting filesystem read-only" error.

Thanks to Andreas Sundstrom for reporting this issue.
Signed-off-by: default avatar"Theodore Ts'o" <tytso@mit.edu>
Cc: stable@kernel.org
      fd98496f
  39. 07 Jan, 2009 1 commit
    • Frank Mayhar's avatar
      ext4: Allow ext4 to run without a journal · 0390131b
      Frank Mayhar authored 
      A few weeks ago I posted a patch for discussion that allowed ext4 to run
without a journal.  Since that time I've integrated the excellent
comments from Andreas and fixed several serious bugs.  We're currently
running with this patch and generating some performance numbers against
both ext2 (with backported reservations code) and ext4 with and without
a journal.  It just so happens that running without a journal is
slightly faster for most everything.

We did
	iozone -T -t 4 s 2g -r 256k -T -I -i0 -i1 -i2

which creates 4 threads, each of which create and do reads and writes on
a 2G file, with a buffer size of 256K, using O_DIRECT for all file opens
to bypass the page cache.  Results:

                     ext2        ext4, default   ext4, no journal
  initial writes   13.0 MB/s        15.4 MB/s          15.7 MB/s
  rewrites         13.1 MB/s        15.6 MB/s          15.9 MB/s
  reads            15.2 MB/s        16.9 MB/s          17.2 MB/s
  re-reads         15.3 MB/s        16.9 MB/s          17.2 MB/s
  random readers    5.6 MB/s         5.6 MB/s           5.7 MB/s
  random writers    5.1 MB/s         5.3 MB/s           5.4 MB/s 

So it seems that, so far, this was a useful exercise.
Signed-off-by: default avatarFrank Mayhar <fmayhar@google.com>
Signed-off-by: default avatar"Theodore Ts'o" <tytso@mit.edu>
      0390131b
  41. 17 Dec, 2008 1 commit
  43. 27 Nov, 2008 1 commit
  45. 26 Nov, 2008 1 commit
  47. 25 Nov, 2008 2 commits
  49. 06 Jan, 2009 2 commits