Changelog V5 -> V6:
- Fix oom when the memory load is high, by storing the delayed nodes into the
  root's radix tree, and letting btrfs inodes go.

Changelog V4 -> V5:
- Fix the race on adding the delayed node to the inode, which is spotted by
  Chris Mason.
- Merge Chris Mason's incremental patch into this patch.
- Fix deadlock between readdir() and memory fault, which is reported by
  Itaru Kitayama.

Changelog V3 -> V4:
- Fix nested lock, which is reported by Itaru Kitayama, by updating space cache
  inode in time.

Changelog V2 -> V3:
- Fix the race between the delayed worker and the task which does delayed items
  balance, which is reported by Tsutomu Itoh.
- Modify the patch address David Sterba's comment.
- Fix the bug of the cpu recursion spinlock, reported by Chris Mason

Changelog V1 -> V2:
- break up the global rb-tree, use a list to manage the delayed nodes,
  which is created for every directory and file, and used to manage the
  delayed directory name index items and the delayed inode item.
- introduce a worker to deal with the delayed nodes.

Compare with Ext3/4, the performance of file creation and deletion on btrfs
is very poor. the reason is that btrfs must do a lot of b+ tree insertions,
such as inode item, directory name item, directory name index and so on.

If we can do some delayed b+ tree insertion or deletion, we can improve the
performance, so we made this patch which implemented delayed directory name
index insertion/deletion and delayed inode update.

Implementation:
- introduce a delayed root object into the filesystem, that use two lists to
  manage the delayed nodes which are created for every file/directory.
  One is used to manage all the delayed nodes that have delayed items. And the
  other is used to manage the delayed nodes which is waiting to be dealt with
  by the work thread.
- Every delayed node has two rb-tree, one is used to manage the directory name
  index which is going to be inserted into b+ tree, and the other is used to
  manage the directory name index which is going to be deleted from b+ tree.
- introduce a worker to deal with the delayed operation. This worker is used
  to deal with the works of the delayed directory name index items insertion
  and deletion and the delayed inode update.
  When the delayed items is beyond the lower limit, we create works for some
  delayed nodes and insert them into the work queue of the worker, and then
  go back.
  When the delayed items is beyond the upper bound, we create works for all
  the delayed nodes that haven't been dealt with, and insert them into the work
  queue of the worker, and then wait for that the untreated items is below some
  threshold value.
- When we want to insert a directory name index into b+ tree, we just add the
  information into the delayed inserting rb-tree.
  And then we check the number of the delayed items and do delayed items
  balance. (The balance policy is above.)
- When we want to delete a directory name index from the b+ tree, we search it
  in the inserting rb-tree at first. If we look it up, just drop it. If not,
  add the key of it into the delayed deleting rb-tree.
  Similar to the delayed inserting rb-tree, we also check the number of the
  delayed items and do delayed items balance.
  (The same to inserting manipulation)
- When we want to update the metadata of some inode, we cached the data of the
  inode into the delayed node. the worker will flush it into the b+ tree after
  dealing with the delayed insertion and deletion.
- We will move the delayed node to the tail of the list after we access the
  delayed node, By this way, we can cache more delayed items and merge more
  inode updates.
- If we want to commit transaction, we will deal with all the delayed node.
- the delayed node will be freed when we free the btrfs inode.
- Before we log the inode items, we commit all the directory name index items
  and the delayed inode update.

I did a quick test by the benchmark tool[1] and found we can improve the
performance of file creation by ~15%, and file deletion by ~20%.

Before applying this patch:
Create files:
        Total files: 50000
        Total time: 1.096108
        Average time: 0.000022
Delete files:
        Total files: 50000
        Total time: 1.510403
        Average time: 0.000030

After applying this patch:
Create files:
        Total files: 50000
        Total time: 0.932899
        Average time: 0.000019
Delete files:
        Total files: 50000
        Total time: 1.215732
        Average time: 0.000024

[1] http://marc.info/?l=linux-btrfs&m=128212635122920&q=p3

Many thanks for Kitayama-san's help!
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Reviewed-by: NDavid Sterba <dave@jikos.cz>
Tested-by: NTsutomu Itoh <t-itoh@jp.fujitsu.com>
Tested-by: NItaru Kitayama <kitayama@cl.bb4u.ne.jp>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

Steps to reproduce the bug:

  - Call FS_IOC_SETLFAGS ioctl with flags=FS_COMPR_FL
  - Call FS_IOC_SETFLAGS ioctl with flags=0
  - Call FS_IOC_GETFLAGS ioctl, and you'll see FS_COMPR_FL is still set!
Signed-off-by: NLi Zefan <lizf@cn.fujitsu.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

As we've added per file compression/cow support.
Signed-off-by: NLi Zefan <lizf@cn.fujitsu.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

FS_COW_FL and FS_NOCOW_FL were newly introduced to control per file
COW in btrfs, but FS_NOCOW_FL is sufficient.

The fact is we don't have corresponding BTRFS_INODE_COW flag.

COW is default, and FS_NOCOW_FL can be used to switch off COW for
a single file.

If we mount btrfs with nodatacow, a newly created file will be set with
the FS_NOCOW_FL flag. So to turn on COW for it, we can just clear the
FS_NOCOW_FL flag.
Signed-off-by: NLi Zefan <lizf@cn.fujitsu.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

When a btrfs disk is created by mixed data & metadata option, it will have no
pure data or pure metadata space info.

In btrfs's for-linus branch, commit 78b1ea13838039cd88afdd62519b40b344d6c920
(Btrfs: fix OOPS of empty filesystem after balance) initializes space infos at
the very beginning.  The problem is this initialization does not take the mixed
case into account, which will cause btrfs will easily get into ENOSPC in mixed
case.
Signed-off-by: NLiu Bo <liubo2009@cn.fujitsu.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

If posix_acl_from_xattr() returns an error code, a negative address is
dereferenced causing an oops; fix by checking for error code first.
Signed-off-by: NDaniel J Blueman <daniel.blueman@gmail.com>
Reviewed-by: NJosef Bacik <josef@redhat.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

These changes were incorrectly fixed by codespell. They were now
manually corrected.
Signed-off-by: NLucas De Marchi <lucas.demarchi@profusion.mobi>

The error processing of several places is changed like setting the
error number only at the error.
Signed-off-by: NTsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

In btrfs_submit_direct_hook if the first btrfs_map_block fails we need to put
the orig_bio, not bio.
Signed-off-by: NJosef Bacik <josef@redhat.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

If our space cache is wrong, we do the right thing and free up everything that
we loaded, however we don't reset the total_bitmaps counter or the thresholds or
anything.  So in btrfs_remove_free_space_cache make sure to call free_bitmap()
if it's a bitmap, this will keep us from panicing when we check to make sure we
don't have too many bitmaps.  Thanks,
Signed-off-by: NJosef Bacik <josef@redhat.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

Since commit dc89e982, we've changed
to use a specific slab for alocation of free_space items.
Signed-off-by: NLi Zefan <lizf@cn.fujitsu.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

Signed-off-by: NDavid Sterba <dsterba@suse.cz>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

The check on the return value of kmalloc() is added to some places.
Signed-off-by: NTsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

the space cache use extent_readpages() to read free space information,
so we can not use GFP_KERNEL flag to allocate memory, or it may lead
to deadlock.
Signed-off-by: NItaru Kitayama <kitayama@cl.bb4u.ne.jp>
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

It is necessary to unlock mutex_lock before it return an error when
btrfs_alloc_path() fails.
Signed-off-by: NTsutomu Itoh <t-itoh@jp.fujitsu.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

The Btrfs submit bio threads have a small number of
threads responsible for pushing down bios we've collected
for a large number of devices.

Since we do all the bios for a single device at once,
we want to make sure we unplug and send down the bios
for each device as we're done processing them.

The new plugging API removed the btrfs code to
unplug while processing bios, this adds it back with
the new API.
Signed-off-by: NChris Mason <chris.mason@oracle.com>

The free space caching code was recently reworked to
cache all the pages it needed instead of using find_get_page everywhere.

One loop was missed though, so it ended up leaking pages.  This fixes
it to use our page array instead of find_get_page.
Signed-off-by: NChris Mason <chris.mason@oracle.com>

Everytime we try to allocate disk space we try and see if we can pre-emptively
allocate a chunk, but in the common case we don't allocate anything, so there is
no sense in taking the chunk_mutex at all.  So instead if we are allocating a
chunk, mark it in the space_info so we don't get two people trying to allocate
at the same time.  Thanks,
Signed-off-by: NJosef Bacik <josef@redhat.com>
Reviewed-by: NLiu Bo <liubo2009@cn.fujitsu.com>

A recent commit caches the extent state in end_bio_extent_readpage,
but the search it does should look for locked extents.  This
fixes things to make it more effective.
Signed-off-by: NChris Mason <chris.mason@oracle.com>

find_free_extent likes to allocate in contiguous clusters,
which makes writeback faster, especially on SSD storage.  As
the FS fragments, these clusters become harder to find and we have
to decide between allocating a new chunk to make more clusters
or giving up on the cluster to allocate from the free space
we have.

Right now it creates too many chunks, and you can end up with
a whole FS that is mostly empty metadata chunks.  This commit
changes the allocation code to be more strict and only
allocate new chunks when we've made good use of the chunks we
already have.
Signed-off-by: NChris Mason <chris.mason@oracle.com>

Call posix_acl_valid() to check if an acl is valid or not.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NLi Zefan <lizf@cn.fujitsu.com>

Link count of the inode is not decreased if btrfs_set_inode_index()
fails.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Singed-off-by: NLi Zefan <lizf@cn.fujitsu.com>

btrfs_next_leaf() can return -errno, and we should propagate
it to userspace.

This also simplifies how we walk the btree path.
Signed-off-by: NLi Zefan <lizf@cn.fujitsu.com>

btrfs_next_leaf() can return -errno, and we should propagate
it to userspace.

This also simplifies how we walk the btree path.
Signed-off-by: NLi Zefan <lizf@cn.fujitsu.com>

The extent_io code can take cached pointers into the extent state trees,
and these can make lookups much faster in common operations.  The
caching only happens when specific bits are set that prevent merging
and splitting of the extent state.

A help function was added to uncache the state, and it was testing
the same set of conditionals.  This can leak in very strange corner
cases where the lock bit goes away unexpectedly.

The uncaching should be unconditional.  Once we have a ref on the
extent we should always give it up.
Signed-off-by: NChris Mason <chris.mason@oracle.com>

In several places the sequence (set_extent_uptodate, unlock_extent) is used.
This leads to a duplicate lookup of the extent state. This patch lets
set_extent_uptodate return a cached extent_state which can be passed to
unlock_extent_cached.
The occurences of the above sequences are updated to use the cache. Only
end_bio_extent_readpage is updated that it first gets a cached state to
pass it to the readpage_end_io_hook as the prototype requested and is later
on being used for set/unlock.
Signed-off-by: NArne Jansen <sensille@gmx.net>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

I've been working on making our O_DIRECT latency not suck and I noticed we were
taking the trans_mutex in btrfs_end_transaction.  So to do this we convert
num_writers and use_count to atomic_t's and just decrement them in
btrfs_end_transaction.  Instead of deleting the transaction from the trans list
in put_transaction we do that in btrfs_commit_transaction() since that's the
only time it actually needs to be removed from the list.  Thanks,
Signed-off-by: NJosef Bacik <josef@redhat.com>

We create two subvolumes (meego_root and meego_home) in
btrfs root directory. And set meego_root as default mount
subvolume. After we remount btrfs, meego_root is mounted
to top directory by default. Then when we try to mount
meego_home (subvol=meego_home) to a subdirectory, it failed.
The problem is when default mount subvolume is set to
meego_root, we search meego_home in meego_root but can not find
it. So the solution is to add a new mount option (subvolrootid)
to specify subvol id of root and search subvol name in it. For
our case, now we can use "-o subvolrootid=0,subvol=meego_home)
to mount meego_home.

Detail information can be found in meego bugzilla:
https://bugs.meego.com/show_bug.cgi?id=15055Signed-off-by: NZhong, Xin <xin.zhong@intel.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

Fix address space annotation correct in ioctl.c.
Signed-off-by: NDaniel J Blueman <daniel.blueman@gmail.com>

 		       BTRFS_BLOCK_GROUP_SYSTEM,
@@ -2387,7 +2387,7 @@ long btrfs_ioctl_space_info(struct btrfs_root
*root, void __user *arg)
 		up_read(&info->groups_sem);
 	}

-	user_dest = (struct btrfs_ioctl_space_info *)
+	user_dest = (struct btrfs_ioctl_space_info __user *)
 		(arg + sizeof(struct btrfs_ioctl_space_args));

 	if (copy_to_user(user_dest, dest_orig, alloc_size))
Reviewed-by: NJosef Bacik <josef@redhat.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

Apparently it is ok to submit a read to an IDE device with the same target page
for different offsets.  This is what Windows does under qemu.  The problem is
under DIO we expect them to be different buffers for checksumming reasons, and
so this sort of thing will result in checksum errors, when in reality the file
is fine.  So when reading, check to make sure that all iov bases are different,
and if they aren't fall back to buffered mode, since that will work out right.
Thanks,
Signed-off-by: NJosef Bacik <josef@redhat.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

Fix data corruption caused by memcpy() usage on overlapping data.
I've observed it first when found out usermode linux crash on btrfs.

?all chain is the following:
------------[ cut here ]------------
WARNING: at /home/slyfox/linux-2.6/fs/btrfs/extent_io.c:3900 memcpy_extent_buffer+0x1a5/0x219()
Call Trace:
6fa39a58:  [<601b495e>] _raw_spin_unlock_irqrestore+0x18/0x1c
6fa39a68:  [<60029ad9>] warn_slowpath_common+0x59/0x70
6fa39aa8:  [<60029b05>] warn_slowpath_null+0x15/0x17
6fa39ab8:  [<600efc97>] memcpy_extent_buffer+0x1a5/0x219
6fa39b48:  [<600efd9f>] memmove_extent_buffer+0x94/0x208
6fa39bc8:  [<600becbf>] btrfs_del_items+0x214/0x473
6fa39c78:  [<600ce1b0>] btrfs_delete_one_dir_name+0x7c/0xda
6fa39cc8:  [<600dad6b>] __btrfs_unlink_inode+0xad/0x25d
6fa39d08:  [<600d7864>] btrfs_start_transaction+0xe/0x10
6fa39d48:  [<600dc9ff>] btrfs_unlink_inode+0x1b/0x3b
6fa39d78:  [<600e04bc>] btrfs_unlink+0x70/0xef
6fa39dc8:  [<6007f0d0>] vfs_unlink+0x58/0xa3
6fa39df8:  [<60080278>] do_unlinkat+0xd4/0x162
6fa39e48:  [<600517db>] call_rcu_sched+0xe/0x10
6fa39e58:  [<600452a8>] __put_cred+0x58/0x5a
6fa39e78:  [<6007446c>] sys_faccessat+0x154/0x166
6fa39ed8:  [<60080317>] sys_unlink+0x11/0x13
6fa39ee8:  [<60016b80>] handle_syscall+0x58/0x70
6fa39f08:  [<60021377>] userspace+0x2d4/0x381
6fa39fc8:  [<60014507>] fork_handler+0x62/0x69
---[ end trace 70b0ca2ef0266b93 ]---

http://www.mail-archive.com/linux-btrfs@vger.kernel.org/msg09302.htmlSigned-off-by: NSergei Trofimovich <slyfox@gentoo.org>
Reviewed-by: NJosef Bacik <josef@redhat.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

This patch fixes memory leaks in btrfs_new_inode().
Signed-off-by: NYoshinori Sano <yoshinori.sano@gmail.com>
Signed-off-by: NChris Mason <chris.mason@oracle.com>

Apparently it is ok to submit a read to an IDE device with the same target page
for different offsets.  This is what Windows does under qemu.  The problem is
under DIO we expect them to be different buffers for checksumming reasons, and
so this sort of thing will result in checksum errors, when in reality the file
is fine.  So when reading, check to make sure that all iov bases are different,
and if they aren't fall back to buffered mode, since that will work out right.
Thanks,
Signed-off-by: NJosef Bacik <josef@redhat.com>

In btrfs_get_block_direct we call btrfs_get_extent to lookup the extent for the
range that we are looking for.  If we don't find an extent, btrfs_get_extent
will insert a extent_map for that area and mark it as a hole.  So it does the
job of allocating a new extent map and inserting it into the io tree.  But if
we're creating a new extent we free it up and redo all of that work.  So instead
pass the em to btrfs_new_extent_direct(), and if it will work just allocate the
disk space and set it up properly and bypass the freeing/allocating of a new
extent map and the expensive operation of inserting the thing into the io_tree.
Thanks,
Signed-off-by: NJosef Bacik <josef@redhat.com>

When looking at our DIO performance Chris said that for small IO's doing the
async submit stuff tends to be more overhead than it's worth.  With this on top
of my other fixes I get about a 17-20% speedup doing a sequential dd with 4k
IO's.  Basically if we don't have to split the bio for the map length it's small
enough to be directly submitted, otherwise go back to the async submit.  Thanks,
Signed-off-by: NJosef Bacik <josef@redhat.com>

We have been unconditionally allocating a new bio and re-adding all pages from
our original bio to the new bio.  This is needed if our original bio is larger
than our stripe size, but if it is smaller than the stripe size then there is no
need to do this.  So check the map length and if we are under that then go ahead
and submit the original bio.  Thanks,
Signed-off-by: NJosef Bacik <josef@redhat.com>

In the DIO code we often don't update the i_disk_size because the i_size isn't
updated until after the DIO is completed, so basically we are allocating a path,
doing a search, and updating the inode item for no reason since nothing changed.
btrfs_ordered_update_i_size will return 1 if it didn't update i_disk_size, so
only run btrfs_update_inode if btrfs_ordered_update_i_size returns 0.  Thanks,
Signed-off-by: NJosef Bacik <josef@redhat.com>

Instead of calling kmap_atomic for every thing we set in the inode item, map the
entire inode item at the start and unmap it at the end.  This makes a sequential
dd of 400mb O_DIRECT something like 1% faster.  Thanks,
Signed-off-by: NJosef Bacik <josef@redhat.com>

I saw a lockup where we kept getting into this start transaction->commit
transaction loop because of enospce.  The fact is if we fail to make our
reservation, we've tried _everything_ several times, so we only need to try and
commit the transaction once, and if that doesn't work then we really are out of
space and need to just exit.  Thanks,
Signed-off-by: NJosef Bacik <josef@redhat.com>

Currently we don't handle running out of space in the cache, so to fix this we
keep track of how far in the cache we are.  Then we only dirty the pages if we
successfully modify all of them, otherwise if we have an error or run out of
space we can just drop them and not worry about the vm writing them out.
Thanks,

Tested-by Johannes Hirte <johannes.hirte@fem.tu-ilmenau.de>
Signed-off-by: NJosef Bacik <josef@redhat.com>