linux.git/fs/ocfs2/alloc.c, branch v3.19

ocfs2: reflink: fix slow unlink for refcounted file

2014-12-19T03:08:11+00:00

When running ocfs2 test suite multiple nodes reflink stress test, for a
4 nodes cluster, every unlink() for refcounted file needs about 700s.

The slow unlink is caused by the contention of refcount tree lock since
all nodes are unlink files using the same refcount tree.  When the
unlinking file have many extents(over 1600 in our test), most of the
extents has refcounted flag set.  In ocfs2_commit_truncate(), it will
execute the following call trace for every extents.  This means it needs
get and released refcount tree lock about 1600 times.  And when several
nodes are do this at the same time, the performance will be very low.

  ocfs2_remove_btree_range()
  --  ocfs2_lock_refcount_tree()
  ----  ocfs2_refcount_lock()
  ------  __ocfs2_cluster_lock()

ocfs2_refcount_lock() is costly, move it to ocfs2_commit_truncate() to
do lock/unlock once can improve a lot performance.

Signed-off-by: Junxiao Bi 
Cc: Wengang 
Reviewed-by: Mark Fasheh 
Cc: Joel Becker 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

ocfs2: correctly check the return value of ocfs2_search_extent_list

2014-08-07T01:01:13+00:00

ocfs2_search_extent_list may return -1, so we should check the return
value in ocfs2_split_and_insert, otherwise it may cause array index out of
bound.

And ocfs2_search_extent_list can only return value less than
el->l_next_free_rec, so check if it is equal or larger than
le16_to_cpu(el->l_next_free_rec) is meaningless.

Signed-off-by: Yingtai Xie 
Signed-off-by: Joseph Qi 
Cc: Joel Becker 
Cc: Mark Fasheh 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

ocfs2: fix umount hang while shutting down truncate log

2014-06-04T23:53:54+00:00

Revert commit 75f82eaa502c ("ocfs2: fix NULL pointer dereference when
dismount and ocfs2rec simultaneously") because it may cause a umount
hang while shutting down the truncate log.

fix NULL pointer dereference when dismount and ocfs2rec simultaneously

The situation is as followes:
ocfs2_dismout_volume
-> ocfs2_recovery_exit
  -> free osb->recovery_map
-> ocfs2_truncate_shutdown
  -> lock global bitmap inode
    -> ocfs2_wait_for_recovery
	  -> check whether osb->recovery_map->rm_used is zero

Because osb->recovery_map is already freed, rm_used can be any other
values, so it may yield umount hang.

To prevent NULL pointer dereference while getting sys_root_inode, we use
a osb_tl_disable flag to disable schedule osb_truncate_log_wq after
truncate log shutdown.

Signed-off-by: joyce.xue 
Cc: Mark Fasheh 
Cc: Joel Becker 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

ocfs2: call ocfs2_update_inode_fsync_trans when updating any inode

2014-04-03T23:20:56+00:00

Ensure that ocfs2_update_inode_fsync_trans() is called any time we touch
an inode in a given transaction.  This is a follow-on to the previous
patch to reduce lock contention and deadlocking during an fsync
operation.

Signed-off-by: Darrick J. Wong 
Cc: Mark Fasheh 
Cc: Joel Becker 
Cc: Wengang 
Cc: Greg Marsden 
Cc: Srinivas Eeda 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

ocfs2: improve fsync efficiency and fix deadlock between aio_write and sync_file

2014-04-03T23:20:53+00:00

Currently, ocfs2_sync_file grabs i_mutex and forces the current journal
transaction to complete.  This isn't terribly efficient, since sync_file
really only needs to wait for the last transaction involving that inode
to complete, and this doesn't require i_mutex.

Therefore, implement the necessary bits to track the newest tid
associated with an inode, and teach sync_file to wait for that instead
of waiting for everything in the journal to commit.  Furthermore, only
issue the flush request to the drive if jbd2 hasn't already done so.

This also eliminates the deadlock between ocfs2_file_aio_write() and
ocfs2_sync_file().  aio_write takes i_mutex then calls
ocfs2_aiodio_wait() to wait for unaligned dio writes to finish.
However, if that dio completion involves calling fsync, then we can get
into trouble when some ocfs2_sync_file tries to take i_mutex.

Signed-off-by: Darrick J. Wong 
Reviewed-by: Mark Fasheh 
Cc: Joel Becker 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

ocfs2: fix issue that ocfs2_setattr() does not deal with new_i_size==i_size

2014-02-11T00:01:43+00:00

The issue scenario is as following:

- Create a small file and fallocate a large disk space for a file with
  FALLOC_FL_KEEP_SIZE option.

- ftruncate the file back to the original size again.  but the disk free
  space is not changed back.  This is a real bug that be fixed in this
  patch.

In order to solve the issue above, we modified ocfs2_setattr(), if
attr->ia_size != i_size_read(inode), It calls ocfs2_truncate_file(), and
truncate disk space to attr->ia_size.

Signed-off-by: Younger Liu 
Reviewed-by: Jie Liu 
Tested-by: Jie Liu 
Cc: Joel Becker 
Reviewed-by: Mark Fasheh 
Cc: Sunil Mushran 
Reviewed-by: Jensen 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

ocfs2: free allocated clusters if error occurs after ocfs2_claim_clusters

2014-02-06T21:48:51+00:00

Even if using the same jbd2 handle, we cannot rollback a transaction.
So once some error occurs after successfully allocating clusters, the
allocated clusters will never be used and it means they are lost.  For
example, call ocfs2_claim_clusters successfully when expanding a file,
but failed in ocfs2_insert_extent.  So we need free the allocated
clusters if they are not used indeed.

Signed-off-by: Zongxun Wang 
Signed-off-by: Joseph Qi 
Acked-by: Joel Becker 
Cc: Mark Fasheh 
Cc: Li Zefan 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

ocfs2: return EINVAL if the given range to discard is less than block size

2014-01-22T00:19:42+00:00

For FITRIM ioctl(2), we should not keep silence if the given range
length ls less than a block size as there is no data blocks would be
discareded.  Hence it should return EINVAL instead.  This issue can be
verified via xfstests/generic/288 which is used for FITRIM argument
handling tests.

Signed-off-by: Jie Liu 
Cc: Mark Fasheh 
Cc: Joel Becker 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

ocfs2: return ENOMEM when sb_getblk() fails

2013-11-13T03:09:00+00:00

The only reason for sb_getblk() failing is if it can't allocate the
buffer_head.  So return ENOMEM instead when it fails.

[joseph.qi@huawei.com: ocfs2_symlink_get_block() and ocfs2_read_blocks_sync() and ocfs2_read_blocks() need the same change]
Signed-off-by: Rui Xiang 
Reviewed-by: Jie Liu 
Reviewed-by: Mark Fasheh 
Cc: Joel Becker 
Cc: Joseph Qi 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

ocfs2: fix mutex_unlock and possible memory leak in ocfs2_remove_btree_range

2013-07-03T23:07:23+00:00

In ocfs2_remove_btree_range, when calling ocfs2_lock_refcount_tree and
ocfs2_prepare_refcount_change_for_del failed, it goes to out and then
tries to call mutex_unlock without mutex_lock before.  And when calling
ocfs2_reserve_blocks_for_rec_trunc failed, it should free ref_tree
before return.

Signed-off-by: Joseph Qi 
Reviewed-by: Jie Liu 
Cc: Joel Becker 
Cc: Mark Fasheh 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds