path: root/btrfs-fix-autodefrag-on-5.16.9.patch

From 6c67e14b140aba83be3aee93961ade179dbc2473 Mon Sep 17 00:00:00 2001
From: Scott B <arglebargle@arglebargle.dev>
Date: Fri, 11 Feb 2022 23:52:12 -0800
Subject: [PATCH] btrfs fix autodefrag on 5.16.9
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Squashed commit of the following:

commit 7af5a9b695e62bdb82b55cb255c448e3af3ac587
Author: Qu Wenruo <wqu@suse.com>
Date:   Fri Feb 11 14:41:43 2022 +0800

    btrfs: defrag: make btrfs_defrag_file() to report accurate number of defragged sectors

    Previously rework btrfs_defrag_file() can only report the number of
    sectors from the first run of defrag_collect_targets().

    This number is not accurate as if holes are punched after the first
    defrag_collect_targets() call, we will not choose to defrag the holes.

    Originally this is to avoid passing @sectors_defragged to every involved
    functions.

    But now since we have btrfs_defrag_ctrl, there is no need to do such
    inaccurate accounting, just update btrfs_defrag_ctrl::sectors_defragged
    after a successful defrag_one_locked_target() call.

    Signed-off-by: Qu Wenruo <wqu@suse.com>
    Reviewed-by: Filipe Manana <fdmanana@suse.com>

commit 7d6ad9ac62135f86c190f4ccf1ea1e8bb2e13480
Author: Qu Wenruo <wqu@suse.com>
Date:   Fri Feb 11 14:41:42 2022 +0800

    btrfs: defrag: use btrfs_defrag_ctrl to replace btrfs_ioctl_defrag_range_args for btrfs_defrag_file()

    This brings the following benefits:

    - No more strange range->start update to indicate last scanned bytenr
      We have btrfs_defrag_ctrl::last_scanned (exclusive) for it directly.

    - No more return value to indicate defragged sectors
      Now btrfs_defrag_file() will just return 0 if no error happened.
      And btrfs_defrag_ctrl::sectors_defragged will show that value.

    - Less parameters to carry around
      Now most defrag_* functions only need to fetch its policy parameters
      from btrfs_defrag_ctrl directly.

    Signed-off-by: Qu Wenruo <wqu@suse.com>

commit b19878cde4728eeb3b5e017a6718ffd9e263c1a2
Author: Qu Wenruo <wqu@suse.com>
Date:   Fri Feb 11 14:41:41 2022 +0800

    btrfs: defrag: introduce btrfs_defrag_ctrl structure for later usage

    Currently btrfs_defrag_file() accepts not only
    btrfs_ioctl_defrag_range_args but also other parameters like @newer_than
    and @max_sectors_to_defrag for extra policies.

    Those extra values are hidden from defrag ioctl and even caused bugs in
    the past due to different behaviors based on those extra values.

    Here we introduce a new structure, btrfs_defrag_ctrl, to include:

    - all members in btrfs_ioctl_defrag_range_args

    - @max_sectors_to_defrag and @newer_than

    - Extra values which callers of btrfs_defrag_file() may care
      Like @sectors_defragged and @last_scanned.

    With the new structure, also introduce a new helper,
    btrfs_defrag_ioctl_args_to_ctrl() to:

    - Do extra sanity check on @compress and @flags

    - Do range alignment when possible

    - Set default values.

    Signed-off-by: Qu Wenruo <wqu@suse.com>
    Reviewed-by: Filipe Manana <fdmanana@suse.com>

commit a510f6c16dbfead2fcf0b04489d676d16851ba9e
Author: Qu Wenruo <wqu@suse.com>
Date:   Fri Feb 11 14:41:39 2022 +0800

    btrfs: defrag: allow defrag_one_cluster() to skip large extent which is not a target

    In the rework of btrfs_defrag_file(), we always call
    defrag_one_cluster() and increase the offset by cluster size, which is
    only 256K.

    But there are cases where we have a large extent (e.g. 128M) which
    doesn't need to be defragged at all.

    Before the refactor, we can directly skip the range, but now we have to
    scan that extent map again and again until the cluster moves after the
    non-target extent.

    Fix the problem by allow defrag_one_cluster() to increase
    btrfs_defrag_ctrl::last_scanned to the end of an extent, if and only if
    the last extent of the cluster is not a target.

    The test script looks like this:

    	mkfs.btrfs -f $dev > /dev/null

    	mount $dev $mnt

    	# As btrfs ioctl uses 32M as extent_threshold
    	xfs_io -f -c "pwrite 0 64M" $mnt/file1
    	sync
    	# Some fragemented range to defrag
    	xfs_io -s -c "pwrite 65548k 4k" \
    		  -c "pwrite 65544k 4k" \
    		  -c "pwrite 65540k 4k" \
    		  -c "pwrite 65536k 4k" \
    		  $mnt/file1
    	sync

    	echo "=== before ==="
    	xfs_io -c "fiemap -v" $mnt/file1
    	echo "=== after ==="
    	btrfs fi defrag $mnt/file1
    	sync
    	xfs_io -c "fiemap -v" $mnt/file1
    	umount $mnt

    With extra ftrace put into defrag_one_cluster(), before the patch it
    would result tons of loops:

    (As defrag_one_cluster() is inlined, the function name is its caller)

      btrfs-126062  [005] .....  4682.816026: btrfs_defrag_file: r/i=5/257 start=0 len=262144
      btrfs-126062  [005] .....  4682.816027: btrfs_defrag_file: r/i=5/257 start=262144 len=262144
      btrfs-126062  [005] .....  4682.816028: btrfs_defrag_file: r/i=5/257 start=524288 len=262144
      btrfs-126062  [005] .....  4682.816028: btrfs_defrag_file: r/i=5/257 start=786432 len=262144
      btrfs-126062  [005] .....  4682.816028: btrfs_defrag_file: r/i=5/257 start=1048576 len=262144
      ...
      btrfs-126062  [005] .....  4682.816043: btrfs_defrag_file: r/i=5/257 start=67108864 len=262144

    But with this patch there will be just one loop, then directly to the
    end of the extent:

      btrfs-130471  [014] .....  5434.029558: defrag_one_cluster: r/i=5/257 start=0 len=262144
      btrfs-130471  [014] .....  5434.029559: defrag_one_cluster: r/i=5/257 start=67108864 len=16384

    Cc: stable@vger.kernel.org # 5.16
    Signed-off-by: Qu Wenruo <wqu@suse.com>
    Reviewed-by: Filipe Manana <fdmanana@suse.com>

commit 3f2d69fc4a7a4ce3f389b9e84fa3c830f6a8b5c5
Author: Qu Wenruo <wqu@suse.com>
Date:   Fri Feb 11 14:41:40 2022 +0800

    btrfs: uapi: introduce BTRFS_DEFRAG_RANGE_MASK for later sanity check

    And since we're here, replace the hardcoded bit flags (1, 2) with
    (1UL << 0) and (1UL << 1), respectively.

    Signed-off-by: Qu Wenruo <wqu@suse.com>
    Reviewed-by: Filipe Manana <fdmanana@suse.com>

commit b6c665523425451af94eb3f044d4474c81f94b1e
Author: Qu Wenruo <wqu@suse.com>
Date:   Fri Jan 28 15:21:22 2022 +0800

    btrfs: defrag: remove an ambiguous condition for rejection

    From the very beginning of btrfs defrag, there is a check to reject
    extents which meet both conditions:

    - Physically adjacent

      We may want to defrag physically adjacent extents to reduce the number
      of extents or the size of subvolume tree.

    - Larger than 128K

      This may be there for compressed extents, but unfortunately 128K is
      exactly the max capacity for compressed extents.
      And the check is > 128K, thus it never rejects compressed extents.

      Furthermore, the compressed extent capacity bug is fixed by previous
      patch, there is no reason for that check anymore.

    The original check has a very small ranges to reject (the target extent
    size is > 128K, and default extent threshold is 256K), and for
    compressed extent it doesn't work at all.

    So it's better just to remove the rejection, and allow us to defrag
    physically adjacent extents.

    Reviewed-by: Filipe Manana <fdmanana@suse.com>
    Signed-off-by: Qu Wenruo <wqu@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit cb53ba48a2b6c9126d128f301b4ed8085dcbce7b
Author: Qu Wenruo <wqu@suse.com>
Date:   Fri Jan 28 15:21:21 2022 +0800

    btrfs: defrag: don't defrag extents which is already at its max capacity

    [BUG]
    For compressed extents, defrag ioctl will always try to defrag any
    compressed extents, wasting not only IO but also CPU time to
    compress/decompress:

       mkfs.btrfs -f $DEV
       mount -o compress $DEV $MNT
       xfs_io -f -c "pwrite -S 0xab 0 128K" $MNT/foobar
       sync
       xfs_io -f -c "pwrite -S 0xcd 128K 128K" $MNT/foobar
       sync
       echo "=== before ==="
       xfs_io -c "fiemap -v" $MNT/foobar
       btrfs filesystem defrag $MNT/foobar
       sync
       echo "=== after ==="
       xfs_io -c "fiemap -v" $MNT/foobar

    Then it shows the 2 128K extents just get CoW for no extra benefit, with
    extra IO/CPU spent:

        === before ===
        /mnt/btrfs/file1:
         EXT: FILE-OFFSET      BLOCK-RANGE      TOTAL FLAGS
           0: [0..255]:        26624..26879       256   0x8
           1: [256..511]:      26632..26887       256   0x9
        === after ===
        /mnt/btrfs/file1:
         EXT: FILE-OFFSET      BLOCK-RANGE      TOTAL FLAGS
           0: [0..255]:        26640..26895       256   0x8
           1: [256..511]:      26648..26903       256   0x9

    This affects not only v5.16 (after the defrag rework), but also v5.15
    (before the defrag rework).

    [CAUSE]
    >From the very beginning, btrfs defrag never checks if one extent is
    already at its max capacity (128K for compressed extents, 128M
    otherwise).

    And the default extent size threshold is 256K, which is already beyond
    the compressed extent max size.

    This means, by default btrfs defrag ioctl will mark all compressed
    extent which is not adjacent to a hole/preallocated range for defrag.

    [FIX]
    Introduce a helper to grab the maximum extent size, and then in
    defrag_collect_targets() and defrag_check_next_extent(), reject extents
    which are already at their max capacity.

    Reported-by: Filipe Manana <fdmanana@suse.com>
    Reviewed-by: Filipe Manana <fdmanana@suse.com>
    Signed-off-by: Qu Wenruo <wqu@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit e5cf566d32d6f7b244d87ab0c0797b43d54b4c37
Author: Qu Wenruo <wqu@suse.com>
Date:   Fri Jan 28 15:21:20 2022 +0800

    btrfs: defrag: don't try to merge regular extents with preallocated extents

    [BUG]
    With older kernels (before v5.16), btrfs will defrag preallocated extents.
    While with newer kernels (v5.16 and newer) btrfs will not defrag
    preallocated extents, but it will defrag the extent just before the
    preallocated extent, even it's just a single sector.

    This can be exposed by the following small script:

    	mkfs.btrfs -f $dev > /dev/null

    	mount $dev $mnt
    	xfs_io -f -c "pwrite 0 4k" -c sync -c "falloc 4k 16K" $mnt/file
    	xfs_io -c "fiemap -v" $mnt/file
    	btrfs fi defrag $mnt/file
    	sync
    	xfs_io -c "fiemap -v" $mnt/file

    The output looks like this on older kernels:

    /mnt/btrfs/file:
     EXT: FILE-OFFSET      BLOCK-RANGE      TOTAL FLAGS
       0: [0..7]:          26624..26631         8   0x0
       1: [8..39]:         26632..26663        32 0x801
    /mnt/btrfs/file:
     EXT: FILE-OFFSET      BLOCK-RANGE      TOTAL FLAGS
       0: [0..39]:         26664..26703        40   0x1

    Which defrags the single sector along with the preallocated extent, and
    replace them with an regular extent into a new location (caused by data
    COW).
    This wastes most of the data IO just for the preallocated range.

    On the other hand, v5.16 is slightly better:

    /mnt/btrfs/file:
     EXT: FILE-OFFSET      BLOCK-RANGE      TOTAL FLAGS
       0: [0..7]:          26624..26631         8   0x0
       1: [8..39]:         26632..26663        32 0x801
    /mnt/btrfs/file:
     EXT: FILE-OFFSET      BLOCK-RANGE      TOTAL FLAGS
       0: [0..7]:          26664..26671         8   0x0
       1: [8..39]:         26632..26663        32 0x801

    The preallocated range is not defragged, but the sector before it still
    gets defragged, which has no need for it.

    [CAUSE]
    One of the function reused by the old and new behavior is
    defrag_check_next_extent(), it will determine if we should defrag
    current extent by checking the next one.

    It only checks if the next extent is a hole or inlined, but it doesn't
    check if it's preallocated.

    On the other hand, out of the function, both old and new kernel will
    reject preallocated extents.

    Such inconsistent behavior causes above behavior.

    [FIX]
    - Also check if next extent is preallocated
      If so, don't defrag current extent.

    - Add comments for each branch why we reject the extent

    This will reduce the IO caused by defrag ioctl and autodefrag.

    Reviewed-by: Filipe Manana <fdmanana@suse.com>
    Signed-off-by: Qu Wenruo <wqu@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit f9649fa5a78f6e27a8bb4ec026efe3b4c1d64bc8
Author: Sidong Yang <realwakka@gmail.com>
Date:   Sun Feb 6 12:52:48 2022 +0000

    btrfs: qgroup: remove duplicated check in adding qgroup relations

    Removes duplicated check when adding qgroup relations.
    btrfs_add_qgroup_relations function adds relations by calling
    add_relation_rb(). add_relation_rb() checks that member/parentid exists
    in current qgroup_tree. But it already checked before calling the
    function. It seems that we don't need to double check.

    Add new function __add_relation_rb() that adds relations with
    qgroup structures and makes old function use the new one. And it makes
    btrfs_add_qgroup_relation() function work without double checks by
    calling the new function.

    Signed-off-by: Sidong Yang <realwakka@gmail.com>
    Reviewed-by: David Sterba <dsterba@suse.com>
    [ add comments ]
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 843263d23fb3348d562a0e410d3e8e552e829ef3
Author: Dāvis Mosāns <davispuh@gmail.com>
Date:   Wed Feb 2 23:44:54 2022 +0200

    btrfs: add lzo workspace buffer length constants

    It makes it more readable for length checking and is be used repeatedly.

    Signed-off-by: Dāvis Mosāns <davispuh@gmail.com>
    Reviewed-by: David Sterba <dsterba@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit b1794acfaaef72cc21c3ec3f92d63b1da0842f54
Author: Dāvis Mosāns <davispuh@gmail.com>
Date:   Wed Feb 2 23:44:55 2022 +0200

    btrfs: prevent copying too big compressed lzo segment

    Compressed length can be corrupted to be a lot larger than memory
    we have allocated for buffer.
    This will cause memcpy in copy_compressed_segment to write outside
    of allocated memory.

    This mostly results in stuck read syscall but sometimes when using
    btrfs send can get #GP

      kernel: general protection fault, probably for non-canonical address 0x841551d5c1000: 0000 [#1] PREEMPT SMP NOPTI
      kernel: CPU: 17 PID: 264 Comm: kworker/u256:7 Tainted: P           OE     5.17.0-rc2-1 #12
      kernel: Workqueue: btrfs-endio btrfs_work_helper [btrfs]
      kernel: RIP: 0010:lzo_decompress_bio (./include/linux/fortify-string.h:225 fs/btrfs/lzo.c:322 fs/btrfs/lzo.c:394) btrfs
      Code starting with the faulting instruction
      ===========================================
         0:*  48 8b 06                mov    (%rsi),%rax              <-- trapping instruction
         3:   48 8d 79 08             lea    0x8(%rcx),%rdi
         7:   48 83 e7 f8             and    $0xfffffffffffffff8,%rdi
         b:   48 89 01                mov    %rax,(%rcx)
         e:   44 89 f0                mov    %r14d,%eax
        11:   48 8b 54 06 f8          mov    -0x8(%rsi,%rax,1),%rdx
      kernel: RSP: 0018:ffffb110812efd50 EFLAGS: 00010212
      kernel: RAX: 0000000000001000 RBX: 000000009ca264c8 RCX: ffff98996e6d8ff8
      kernel: RDX: 0000000000000064 RSI: 000841551d5c1000 RDI: ffffffff9500435d
      kernel: RBP: ffff989a3be856c0 R08: 0000000000000000 R09: 0000000000000000
      kernel: R10: 0000000000000000 R11: 0000000000001000 R12: ffff98996e6d8000
      kernel: R13: 0000000000000008 R14: 0000000000001000 R15: 000841551d5c1000
      kernel: FS:  0000000000000000(0000) GS:ffff98a09d640000(0000) knlGS:0000000000000000
      kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
      kernel: CR2: 00001e9f984d9ea8 CR3: 000000014971a000 CR4: 00000000003506e0
      kernel: Call Trace:
      kernel:  <TASK>
      kernel: end_compressed_bio_read (fs/btrfs/compression.c:104 fs/btrfs/compression.c:1363 fs/btrfs/compression.c:323) btrfs
      kernel: end_workqueue_fn (fs/btrfs/disk-io.c:1923) btrfs
      kernel: btrfs_work_helper (fs/btrfs/async-thread.c:326) btrfs
      kernel: process_one_work (./arch/x86/include/asm/jump_label.h:27 ./include/linux/jump_label.h:212 ./include/trace/events/workqueue.h:108 kernel/workqueue.c:2312)
      kernel: worker_thread (./include/linux/list.h:292 kernel/workqueue.c:2455)
      kernel: ? process_one_work (kernel/workqueue.c:2397)
      kernel: kthread (kernel/kthread.c:377)
      kernel: ? kthread_complete_and_exit (kernel/kthread.c:332)
      kernel: ret_from_fork (arch/x86/entry/entry_64.S:301)
      kernel:  </TASK>

    CC: stable@vger.kernel.org # 4.9+
    Signed-off-by: Dāvis Mosāns <davispuh@gmail.com>
    Reviewed-by: David Sterba <dsterba@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 09ad560a431c83f3741f1e545924c7fbb8957dd4
Author: Dāvis Mosāns <davispuh@gmail.com>
Date:   Sat Feb 5 20:48:23 2022 +0200

    btrfs: send: in case of IO error log it

    Currently if we get IO error while doing send then we abort without
    logging information about which file caused issue.  So log it to help
    with debugging.

    CC: stable@vger.kernel.org # 4.9+
    Signed-off-by: Dāvis Mosāns <davispuh@gmail.com>
    Reviewed-by: David Sterba <dsterba@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 43b4cb906eef17f9a7ca8e660f3e9e44176082f6
Author: Filipe Manana <fdmanana@suse.com>
Date:   Wed Feb 2 15:26:09 2022 +0000

    btrfs: get rid of warning on transaction commit when using flushoncommit

    When using the flushoncommit mount option, during almost every transaction
    commit we trigger a warning from __writeback_inodes_sb_nr():

      $ cat fs/fs-writeback.c:
      (...)
      static void __writeback_inodes_sb_nr(struct super_block *sb, ...
      {
            (...)
            WARN_ON(!rwsem_is_locked(&sb->s_umount));
            (...)
      }
      (...)

    The trace produced in dmesg looks like the following:

      [947.473890] WARNING: CPU: 5 PID: 930 at fs/fs-writeback.c:2610 __writeback_inodes_sb_nr+0x7e/0xb3
      [947.481623] Modules linked in: nfsd nls_cp437 cifs asn1_decoder cifs_arc4 fscache cifs_md4 ipmi_ssif
      [947.489571] CPU: 5 PID: 930 Comm: btrfs-transacti Not tainted 95.16.3-srb-asrock-00001-g36437ad63879 #186
      [947.497969] RIP: 0010:__writeback_inodes_sb_nr+0x7e/0xb3
      [947.502097] Code: 24 10 4c 89 44 24 18 c6 (...)
      [947.519760] RSP: 0018:ffffc90000777e10 EFLAGS: 00010246
      [947.523818] RAX: 0000000000000000 RBX: 0000000000963300 RCX: 0000000000000000
      [947.529765] RDX: 0000000000000000 RSI: 000000000000fa51 RDI: ffffc90000777e50
      [947.535740] RBP: ffff888101628a90 R08: ffff888100955800 R09: ffff888100956000
      [947.541701] R10: 0000000000000002 R11: 0000000000000001 R12: ffff888100963488
      [947.547645] R13: ffff888100963000 R14: ffff888112fb7200 R15: ffff888100963460
      [947.553621] FS:  0000000000000000(0000) GS:ffff88841fd40000(0000) knlGS:0000000000000000
      [947.560537] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
      [947.565122] CR2: 0000000008be50c4 CR3: 000000000220c000 CR4: 00000000001006e0
      [947.571072] Call Trace:
      [947.572354]  <TASK>
      [947.573266]  btrfs_commit_transaction+0x1f1/0x998
      [947.576785]  ? start_transaction+0x3ab/0x44e
      [947.579867]  ? schedule_timeout+0x8a/0xdd
      [947.582716]  transaction_kthread+0xe9/0x156
      [947.585721]  ? btrfs_cleanup_transaction.isra.0+0x407/0x407
      [947.590104]  kthread+0x131/0x139
      [947.592168]  ? set_kthread_struct+0x32/0x32
      [947.595174]  ret_from_fork+0x22/0x30
      [947.597561]  </TASK>
      [947.598553] ---[ end trace 644721052755541c ]---

    This is because we started using writeback_inodes_sb() to flush delalloc
    when committing a transaction (when using -o flushoncommit), in order to
    avoid deadlocks with filesystem freeze operations. This change was made
    by commit ce8ea7cc6eb313 ("btrfs: don't call btrfs_start_delalloc_roots
    in flushoncommit"). After that change we started producing that warning,
    and every now and then a user reports this since the warning happens too
    often, it spams dmesg/syslog, and a user is unsure if this reflects any
    problem that might compromise the filesystem's reliability.

    We can not just lock the sb->s_umount semaphore before calling
    writeback_inodes_sb(), because that would at least deadlock with
    filesystem freezing, since at fs/super.c:freeze_super() sync_filesystem()
    is called while we are holding that semaphore in write mode, and that can
    trigger a transaction commit, resulting in a deadlock. It would also
    trigger the same type of deadlock in the unmount path. Possibly, it could
    also introduce some other locking dependencies that lockdep would report.

    To fix this call try_to_writeback_inodes_sb() instead of
    writeback_inodes_sb(), because that will try to read lock sb->s_umount
    and then will only call writeback_inodes_sb() if it was able to lock it.
    This is fine because the cases where it can't read lock sb->s_umount
    are during a filesystem unmount or during a filesystem freeze - in those
    cases sb->s_umount is write locked and sync_filesystem() is called, which
    calls writeback_inodes_sb(). In other words, in all cases where we can't
    take a read lock on sb->s_umount, writeback is already being triggered
    elsewhere.

    An alternative would be to call btrfs_start_delalloc_roots() with a
    number of pages different from LONG_MAX, for example matching the number
    of delalloc bytes we currently have, in which case we would end up
    starting all delalloc with filemap_fdatawrite_wbc() and not with an
    async flush via filemap_flush() - that is only possible after the rather
    recent commit e076ab2a2ca70a ("btrfs: shrink delalloc pages instead of
    full inodes"). However that creates a whole new can of worms due to new
    lock dependencies, which lockdep complains, like for example:

    [ 8948.247280] ======================================================
    [ 8948.247823] WARNING: possible circular locking dependency detected
    [ 8948.248353] 5.17.0-rc1-btrfs-next-111 #1 Not tainted
    [ 8948.248786] ------------------------------------------------------
    [ 8948.249320] kworker/u16:18/933570 is trying to acquire lock:
    [ 8948.249812] ffff9b3de1591690 (sb_internal#2){.+.+}-{0:0}, at: find_free_extent+0x141e/0x1590 [btrfs]
    [ 8948.250638]
                   but task is already holding lock:
    [ 8948.251140] ffff9b3e09c717d8 (&root->delalloc_mutex){+.+.}-{3:3}, at: start_delalloc_inodes+0x78/0x400 [btrfs]
    [ 8948.252018]
                   which lock already depends on the new lock.

    [ 8948.252710]
                   the existing dependency chain (in reverse order) is:
    [ 8948.253343]
                   -> #2 (&root->delalloc_mutex){+.+.}-{3:3}:
    [ 8948.253950]        __mutex_lock+0x90/0x900
    [ 8948.254354]        start_delalloc_inodes+0x78/0x400 [btrfs]
    [ 8948.254859]        btrfs_start_delalloc_roots+0x194/0x2a0 [btrfs]
    [ 8948.255408]        btrfs_commit_transaction+0x32f/0xc00 [btrfs]
    [ 8948.255942]        btrfs_mksubvol+0x380/0x570 [btrfs]
    [ 8948.256406]        btrfs_mksnapshot+0x81/0xb0 [btrfs]
    [ 8948.256870]        __btrfs_ioctl_snap_create+0x17f/0x190 [btrfs]
    [ 8948.257413]        btrfs_ioctl_snap_create_v2+0xbb/0x140 [btrfs]
    [ 8948.257961]        btrfs_ioctl+0x1196/0x3630 [btrfs]
    [ 8948.258418]        __x64_sys_ioctl+0x83/0xb0
    [ 8948.258793]        do_syscall_64+0x3b/0xc0
    [ 8948.259146]        entry_SYSCALL_64_after_hwframe+0x44/0xae
    [ 8948.259709]
                   -> #1 (&fs_info->delalloc_root_mutex){+.+.}-{3:3}:
    [ 8948.260330]        __mutex_lock+0x90/0x900
    [ 8948.260692]        btrfs_start_delalloc_roots+0x97/0x2a0 [btrfs]
    [ 8948.261234]        btrfs_commit_transaction+0x32f/0xc00 [btrfs]
    [ 8948.261766]        btrfs_set_free_space_cache_v1_active+0x38/0x60 [btrfs]
    [ 8948.262379]        btrfs_start_pre_rw_mount+0x119/0x180 [btrfs]
    [ 8948.262909]        open_ctree+0x1511/0x171e [btrfs]
    [ 8948.263359]        btrfs_mount_root.cold+0x12/0xde [btrfs]
    [ 8948.263863]        legacy_get_tree+0x30/0x50
    [ 8948.264242]        vfs_get_tree+0x28/0xc0
    [ 8948.264594]        vfs_kern_mount.part.0+0x71/0xb0
    [ 8948.265017]        btrfs_mount+0x11d/0x3a0 [btrfs]
    [ 8948.265462]        legacy_get_tree+0x30/0x50
    [ 8948.265851]        vfs_get_tree+0x28/0xc0
    [ 8948.266203]        path_mount+0x2d4/0xbe0
    [ 8948.266554]        __x64_sys_mount+0x103/0x140
    [ 8948.266940]        do_syscall_64+0x3b/0xc0
    [ 8948.267300]        entry_SYSCALL_64_after_hwframe+0x44/0xae
    [ 8948.267790]
                   -> #0 (sb_internal#2){.+.+}-{0:0}:
    [ 8948.268322]        __lock_acquire+0x12e8/0x2260
    [ 8948.268733]        lock_acquire+0xd7/0x310
    [ 8948.269092]        start_transaction+0x44c/0x6e0 [btrfs]
    [ 8948.269591]        find_free_extent+0x141e/0x1590 [btrfs]
    [ 8948.270087]        btrfs_reserve_extent+0x14b/0x280 [btrfs]
    [ 8948.270588]        cow_file_range+0x17e/0x490 [btrfs]
    [ 8948.271051]        btrfs_run_delalloc_range+0x345/0x7a0 [btrfs]
    [ 8948.271586]        writepage_delalloc+0xb5/0x170 [btrfs]
    [ 8948.272071]        __extent_writepage+0x156/0x3c0 [btrfs]
    [ 8948.272579]        extent_write_cache_pages+0x263/0x460 [btrfs]
    [ 8948.273113]        extent_writepages+0x76/0x130 [btrfs]
    [ 8948.273573]        do_writepages+0xd2/0x1c0
    [ 8948.273942]        filemap_fdatawrite_wbc+0x68/0x90
    [ 8948.274371]        start_delalloc_inodes+0x17f/0x400 [btrfs]
    [ 8948.274876]        btrfs_start_delalloc_roots+0x194/0x2a0 [btrfs]
    [ 8948.275417]        flush_space+0x1f2/0x630 [btrfs]
    [ 8948.275863]        btrfs_async_reclaim_data_space+0x108/0x1b0 [btrfs]
    [ 8948.276438]        process_one_work+0x252/0x5a0
    [ 8948.276829]        worker_thread+0x55/0x3b0
    [ 8948.277189]        kthread+0xf2/0x120
    [ 8948.277506]        ret_from_fork+0x22/0x30
    [ 8948.277868]
                   other info that might help us debug this:

    [ 8948.278548] Chain exists of:
                     sb_internal#2 --> &fs_info->delalloc_root_mutex --> &root->delalloc_mutex

    [ 8948.279601]  Possible unsafe locking scenario:

    [ 8948.280102]        CPU0                    CPU1
    [ 8948.280508]        ----                    ----
    [ 8948.280915]   lock(&root->delalloc_mutex);
    [ 8948.281271]                                lock(&fs_info->delalloc_root_mutex);
    [ 8948.281915]                                lock(&root->delalloc_mutex);
    [ 8948.282487]   lock(sb_internal#2);
    [ 8948.282800]
                    *** DEADLOCK ***

    [ 8948.283333] 4 locks held by kworker/u16:18/933570:
    [ 8948.283750]  #0: ffff9b3dc00a9d48 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x1d2/0x5a0
    [ 8948.284609]  #1: ffffa90349dafe70 ((work_completion)(&fs_info->async_data_reclaim_work)){+.+.}-{0:0}, at: process_one_work+0x1d2/0x5a0
    [ 8948.285637]  #2: ffff9b3e14db5040 (&fs_info->delalloc_root_mutex){+.+.}-{3:3}, at: btrfs_start_delalloc_roots+0x97/0x2a0 [btrfs]
    [ 8948.286674]  #3: ffff9b3e09c717d8 (&root->delalloc_mutex){+.+.}-{3:3}, at: start_delalloc_inodes+0x78/0x400 [btrfs]
    [ 8948.287596]
                  stack backtrace:
    [ 8948.287975] CPU: 3 PID: 933570 Comm: kworker/u16:18 Not tainted 5.17.0-rc1-btrfs-next-111 #1
    [ 8948.288677] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qemu.org 04/01/2014
    [ 8948.289649] Workqueue: events_unbound btrfs_async_reclaim_data_space [btrfs]
    [ 8948.290298] Call Trace:
    [ 8948.290517]  <TASK>
    [ 8948.290700]  dump_stack_lvl+0x59/0x73
    [ 8948.291026]  check_noncircular+0xf3/0x110
    [ 8948.291375]  ? start_transaction+0x228/0x6e0 [btrfs]
    [ 8948.291826]  __lock_acquire+0x12e8/0x2260
    [ 8948.292241]  lock_acquire+0xd7/0x310
    [ 8948.292714]  ? find_free_extent+0x141e/0x1590 [btrfs]
    [ 8948.293241]  ? lock_is_held_type+0xea/0x140
    [ 8948.293601]  start_transaction+0x44c/0x6e0 [btrfs]
    [ 8948.294055]  ? find_free_extent+0x141e/0x1590 [btrfs]
    [ 8948.294518]  find_free_extent+0x141e/0x1590 [btrfs]
    [ 8948.294957]  ? _raw_spin_unlock+0x29/0x40
    [ 8948.295312]  ? btrfs_get_alloc_profile+0x124/0x290 [btrfs]
    [ 8948.295813]  btrfs_reserve_extent+0x14b/0x280 [btrfs]
    [ 8948.296270]  cow_file_range+0x17e/0x490 [btrfs]
    [ 8948.296691]  btrfs_run_delalloc_range+0x345/0x7a0 [btrfs]
    [ 8948.297175]  ? find_lock_delalloc_range+0x247/0x270 [btrfs]
    [ 8948.297678]  writepage_delalloc+0xb5/0x170 [btrfs]
    [ 8948.298123]  __extent_writepage+0x156/0x3c0 [btrfs]
    [ 8948.298570]  extent_write_cache_pages+0x263/0x460 [btrfs]
    [ 8948.299061]  extent_writepages+0x76/0x130 [btrfs]
    [ 8948.299495]  do_writepages+0xd2/0x1c0
    [ 8948.299817]  ? sched_clock_cpu+0xd/0x110
    [ 8948.300160]  ? lock_release+0x155/0x4a0
    [ 8948.300494]  filemap_fdatawrite_wbc+0x68/0x90
    [ 8948.300874]  ? do_raw_spin_unlock+0x4b/0xa0
    [ 8948.301243]  start_delalloc_inodes+0x17f/0x400 [btrfs]
    [ 8948.301706]  ? lock_release+0x155/0x4a0
    [ 8948.302055]  btrfs_start_delalloc_roots+0x194/0x2a0 [btrfs]
    [ 8948.302564]  flush_space+0x1f2/0x630 [btrfs]
    [ 8948.302970]  btrfs_async_reclaim_data_space+0x108/0x1b0 [btrfs]
    [ 8948.303510]  process_one_work+0x252/0x5a0
    [ 8948.303860]  ? process_one_work+0x5a0/0x5a0
    [ 8948.304221]  worker_thread+0x55/0x3b0
    [ 8948.304543]  ? process_one_work+0x5a0/0x5a0
    [ 8948.304904]  kthread+0xf2/0x120
    [ 8948.305184]  ? kthread_complete_and_exit+0x20/0x20
    [ 8948.305598]  ret_from_fork+0x22/0x30
    [ 8948.305921]  </TASK>

    It all comes from the fact that btrfs_start_delalloc_roots() takes the
    delalloc_root_mutex, in the transaction commit path we are holding a
    read lock on one of the superblock's freeze semaphores (via
    sb_start_intwrite()), the async reclaim task can also do a call to
    btrfs_start_delalloc_roots(), which ends up triggering writeback with
    calls to filemap_fdatawrite_wbc(), resulting in extent allocation which
    in turn can call btrfs_start_transaction(), which will result in taking
    the freeze semaphore via sb_start_intwrite(), forming a nasty dependency
    on all those locks which can be taken in different orders by different
    code paths.

    So just adopt the simple approach of calling try_to_writeback_inodes_sb()
    at btrfs_start_delalloc_flush().

    Link: https://lore.kernel.org/linux-btrfs/20220130005258.GA7465@cuci.nl/
    Link: https://lore.kernel.org/linux-btrfs/43acc426-d683-d1b6-729d-c6bc4a2fff4d@gmail.com/
    Link: https://lore.kernel.org/linux-btrfs/6833930a-08d7-6fbc-0141-eb9cdfd6bb4d@gmail.com/
    Link: https://lore.kernel.org/linux-btrfs/20190322041731.GF16651@hungrycats.org/
    Reviewed-by: Omar Sandoval <osandov@fb.com>
    Signed-off-by: Filipe Manana <fdmanana@suse.com>
    [ add more link reports ]
    Signed-off-by: David Sterba <dsterba@suse.com>

commit b805349fbdc9a47199d96bc193f64b9399ec6761
Author: Qu Wenruo <wqu@suse.com>
Date:   Tue Feb 8 14:54:05 2022 +0800

    btrfs: defrag: don't try to defrag extents which are under writeback

    Once we start writeback (have called btrfs_run_delalloc_range()), we
    allocate an extent, create an extent map point to that extent, with a
    generation of (u64)-1, created the ordered extent and then clear the
    DELALLOC bit from the range in the inode's io tree.

    Such extent map can pass the first call of defrag_collect_targets(), as
    its generation is (u64)-1, meets any possible minimal generation check.
    And the range will not have DELALLOC bit, also passing the DELALLOC bit
    check.

    It will only be re-checked in the second call of
    defrag_collect_targets(), which will wait for writeback.

    But at that stage we have already spent our time waiting for some IO we
    may or may not want to defrag.

    Let's reject such extents early so we won't waste our time.

    Reviewed-by: Filipe Manana <fdmanana@suse.com>
    Signed-off-by: Qu Wenruo <wqu@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit e8a7717c71287a11dc81098199d7116d6a4b6006
Author: Qu Wenruo <wqu@suse.com>
Date:   Tue Feb 8 13:31:19 2022 +0800

    btrfs: populate extent_map::generation when reading from disk

    When btrfs_get_extent() tries to get some file extent from disk, it
    never populates extent_map::generation, leaving the value to be 0.

    On the other hand, for extent map generated by IO, it will get its
    generation properly set at finish_ordered_io()

     finish_ordered_io()
     |- unpin_extent_cache(gen = trans->transid)
        |- em->generation = gen;

    [CAUSE]
    Since extent_map::generation is mostly used by fsync code, and for fsync
    they only care about modified extents, which all have their
    em::generation > 0.

    Thus it's fine to not populate em read from disk for fsync.

    [CORNER CASE]
    However autodefrag also relies on em::generation to determine if one
    extent needs to be defragged.

    This unpopulated extent_map::generation can prevent the following
    autodefrag case from working:

    	mkfs.btrfs -f $dev
    	mount $dev $mnt -o autodefrag

    	# initial write to queue the inode for autodefrag
    	xfs_io -f -c "pwrite 0 4k" $mnt/file
    	sync

    	# Real fragmented write
    	xfs_io -f -s -c "pwrite -b 4096 0 32k" $mnt/file
    	sync
    	echo "=== before autodefrag ==="
    	xfs_io -c "fiemap -v" $mnt/file

    	# Drop cache to force em to be read from disk
    	echo 3 > /proc/sys/vm/drop_caches
    	mount -o remount,commit=1 $mnt
    	sleep 3
    	sync

    	echo "=== After autodefrag ==="
    	xfs_io -c "fiemap -v" $mnt/file
    	umount $mnt

    The result looks like this:

      === before autodefrag ===
      /mnt/btrfs/file:
       EXT: FILE-OFFSET      BLOCK-RANGE      TOTAL FLAGS
         0: [0..15]:         26672..26687        16   0x0
         1: [16..31]:        26656..26671        16   0x0
         2: [32..47]:        26640..26655        16   0x0
         3: [48..63]:        26624..26639        16   0x1
      === After autodefrag ===
      /mnt/btrfs/file:
       EXT: FILE-OFFSET      BLOCK-RANGE      TOTAL FLAGS
         0: [0..15]:         26672..26687        16   0x0
         1: [16..31]:        26656..26671        16   0x0
         2: [32..47]:        26640..26655        16   0x0
         3: [48..63]:        26624..26639        16   0x1

    This fragmented 32K will not be defragged by autodefrag.

    [FIX]
    To make things less weird, just populate extent_map::generation when
    reading file extents from disk.

    This would make above fragmented extents to be properly defragged:

      == before autodefrag ===
      /mnt/btrfs/file:
       EXT: FILE-OFFSET      BLOCK-RANGE      TOTAL FLAGS
         0: [0..15]:         26672..26687        16   0x0
         1: [16..31]:        26656..26671        16   0x0
         2: [32..47]:        26640..26655        16   0x0
         3: [48..63]:        26624..26639        16   0x1
      === After autodefrag ===
      /mnt/btrfs/file:
       EXT: FILE-OFFSET      BLOCK-RANGE      TOTAL FLAGS
         0: [0..63]:         26688..26751        64   0x1

    Reviewed-by: Filipe Manana <fdmanana@suse.com>
    Signed-off-by: Qu Wenruo <wqu@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit df6d916f35305a85d6636256fbc9708a78df7465
Author: Filipe Manana <fdmanana@suse.com>
Date:   Thu Feb 3 14:55:50 2022 +0000

    btrfs: prepare extents to be logged before locking a log tree path

    When we want to log an extent, in the fast fsync path, we obtain a path
    to the leaf that will hold the file extent item either through a deletion
    search, via btrfs_drop_extents(), or through an insertion search using
    btrfs_insert_empty_item(). After that we fill the file extent item's
    fields one by one directly on the leaf.

    Instead of doing that, we could prepare the file extent item before
    obtaining a btree path, and then copy the prepared extent item with a
    single operation once we get the path. This helps avoid some contention
    on the log tree, since we are holding write locks for longer than
    necessary, especially in the case where the path is obtained via
    btrfs_drop_extents() through a deletion search, which always keeps a
    write lock on the nodes at levels 1 and 2 (besides the leaf).

    This change does that, we prepare the file extent item that is going to
    be inserted before acquiring a path, and then copy it into a leaf using
    a single copy operation once we get a path.

    This change if part of a patchset that is comprised of the following
    patches:

      1/6 btrfs: remove unnecessary leaf free space checks when pushing items
      2/6 btrfs: avoid unnecessary COW of leaves when deleting items from a leaf
      3/6 btrfs: avoid unnecessary computation when deleting items from a leaf
      4/6 btrfs: remove constraint on number of visited leaves when replacing extents
      5/6 btrfs: remove useless path release in the fast fsync path
      6/6 btrfs: prepare extents to be logged before locking a log tree path

    The following test was run to measure the impact of the whole patchset:

      $ cat test.sh
      #!/bin/bash

      DEV=/dev/sdi
      MNT=/mnt/sdi
      MOUNT_OPTIONS="-o ssd"
      MKFS_OPTIONS="-R free-space-tree -O no-holes"

      NUM_JOBS=8
      FILE_SIZE=128M
      RUN_TIME=200

      cat <<EOF > /tmp/fio-job.ini
      [writers]
      rw=randwrite
      fsync=1
      fallocate=none
      group_reporting=1
      direct=0
      bssplit=4k/20:8k/20:16k/20:32k/10:64k/10:128k/5:256k/5:512k/5:1m/5
      ioengine=sync
      filesize=$FILE_SIZE
      runtime=$RUN_TIME
      time_based
      directory=$MNT
      numjobs=$NUM_JOBS
      thread
      EOF

      echo "performance" | \
          tee /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor

      echo
      echo "Using config:"
      echo
      cat /tmp/fio-job.ini
      echo

      umount $MNT &> /dev/null
      mkfs.btrfs -f $MKFS_OPTIONS $DEV
      mount $MOUNT_OPTIONS $DEV $MNT

      fio /tmp/fio-job.ini

      umount $MNT

    The test ran inside a VM (8 cores, 32G of RAM) with the target disk
    mapping to a raw NVMe device, and using a non-debug kernel config
    (Debian's default config).

    Before the patchset:

    WRITE: bw=116MiB/s (122MB/s), 116MiB/s-116MiB/s (122MB/s-122MB/s), io=22.7GiB (24.4GB), run=200013-200013msec

    After the patchset:

    WRITE: bw=125MiB/s (131MB/s), 125MiB/s-125MiB/s (131MB/s-131MB/s), io=24.3GiB (26.1GB), run=200007-200007msec

    A 7.8% gain on throughput and +7.0% more IO done in the same period of
    time (200 seconds).

    Signed-off-by: Filipe Manana <fdmanana@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 901ebe7172fa1fd03e4cc43d9d5f6a191d2e6428
Author: Filipe Manana <fdmanana@suse.com>
Date:   Thu Feb 3 14:55:49 2022 +0000

    btrfs: remove useless path release in the fast fsync path

    There's no point in calling btrfs_release_path() after finishing the loop
    that logs the modified extents, since log_one_extent() returns with the
    path released. In case the list of extents is empty, the path is already
    released, so there's no need for that case as well.
    So just remove that unnecessary btrfs_release_path() call.

    This change if part of a patchset that is comprised of the following
    patches:

      1/6 btrfs: remove unnecessary leaf free space checks when pushing items
      2/6 btrfs: avoid unnecessary COW of leaves when deleting items from a leaf
      3/6 btrfs: avoid unnecessary computation when deleting items from a leaf
      4/6 btrfs: remove constraint on number of visited leaves when replacing extents
      5/6 btrfs: remove useless path release in the fast fsync path
      6/6 btrfs: prepare extents to be logged before locking a log tree path

    The last patch in the series has some performance test result in its
    changelog.

    Signed-off-by: Filipe Manana <fdmanana@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 12d0362b8dad82e240e37aa43f7d344f7206c009
Author: Filipe Manana <fdmanana@suse.com>
Date:   Thu Feb 3 14:55:48 2022 +0000

    btrfs: remove constraint on number of visited leaves when replacing extents

    At btrfs_drop_extents(), we try to replace a range of file extent items
    with a new file extent in a single btree search, to avoid the need to do
    a search for deletion, followed by a path release and followed by yet
    another search for insertion.

    When I originally added that optimization, in commit 1acae57b161ef1
    ("Btrfs: faster file extent item replace operations"), I left a constraint
    to do the fast replace only if we visited a single leaf. That was because
    in the most common case we find all file extent items that need to be
    deleted (or trimmed) in a single leaf, however it can work for other
    common cases like when we need to delete a few file extent items located
    at the end of a leaf and a few more located at the beginning of the next
    leaf. The key for the new file extent item is greater than the key of
    any deleted or trimmed file extent item from previous leaves, so we are
    fine to use the last leaf that we found as long as we are holding a
    write lock on it - even if the new key ends up at slot 0, as if that's
    the case, the btree search has obtained a write lock on any upper nodes
    that need to have a key pointer updated.

    So removed the constraint that limits the optimization to the case where
    we visited only a single leaf.

    This change if part of a patchset that is comprised of the following
    patches:

      1/6 btrfs: remove unnecessary leaf free space checks when pushing items
      2/6 btrfs: avoid unnecessary COW of leaves when deleting items from a leaf
      3/6 btrfs: avoid unnecessary computation when deleting items from a leaf
      4/6 btrfs: remove constraint on number of visited leaves when replacing extents
      5/6 btrfs: remove useless path release in the fast fsync path
      6/6 btrfs: prepare extents to be logged before locking a log tree path

    The last patch in the series has some performance test result in its
    changelog.

    Signed-off-by: Filipe Manana <fdmanana@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 967aa565ee63760ea7f1c00be743f8e24ee83aa6
Author: Filipe Manana <fdmanana@suse.com>
Date:   Thu Feb 3 14:55:47 2022 +0000

    btrfs: avoid unnecessary computation when deleting items from a leaf

    When deleting items from a leaf, we always compute the sum of the data
    sizes of the items that are going to be deleted. However we only use
    that sum when the last item to delete is behind the last item in the
    leaf. This unnecessarily wastes CPU time when we are deleting either
    the whole leaf or from some slot > 0 up to the last item in the leaf,
    and both of these cases are common (e.g. truncation operation, either
    as a result of truncate(2) or when logging inodes, deleting checksums
    after removing a large enough extent, etc).

    So compute only the sum of the data sizes if the last item to be
    deleted does not match the last item in the leaf.

    This change if part of a patchset that is comprised of the following
    patches:

      1/6 btrfs: remove unnecessary leaf free space checks when pushing items
      2/6 btrfs: avoid unnecessary COW of leaves when deleting items from a leaf
      3/6 btrfs: avoid unnecessary computation when deleting items from a leaf
      4/6 btrfs: remove constraint on number of visited leaves when replacing extents
      5/6 btrfs: remove useless path release in the fast fsync path
      6/6 btrfs: prepare extents to be logged before locking a log tree path

    The last patch in the series has some performance test result in its
    changelog.

    Signed-off-by: Filipe Manana <fdmanana@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 4fd3e4a94b70dfab57ce617a2c8196a77e8cc29d
Author: Filipe Manana <fdmanana@suse.com>
Date:   Thu Feb 3 14:55:46 2022 +0000

    btrfs: avoid unnecessary COW of leaves when deleting items from a leaf

    When we delete items from a leaf, if we end up with more than two thirds
    of unused leaf space, we try to delete the leaf by moving all its items
    into its left and right neighbour leaves. Sometimes that is not possible
    because there is not enough free space in the left and right leaves, and
    in that case we end up not deleting our leaf.

    The way we are doing this is not ideal and can be improved in the
    following ways:

    1) When we call push_leaf_left(), we pass a value of 1 byte to the data
       size parameter of push_leaf_left(). This is not realistic value because
       no item can have a size less than 25 bytes, which is the size of struct
       btrfs_item. This means that means that if the left leaf has not enough
       free space to push any item, we end up COWing it even if we end up not
       changing its content at all.

       COWing that leaf means allocating a new metadata extent, marking it
       dirty and doing more IO when committing a transaction or when syncing a
       log tree. For a log tree case, it's particularly more important to
       avoid the useless COW operation, as more IO can imply a higher latency
       for an fsync operation.

       So instead of passing 1 as the minimum data size for push_leaf_left(),
       pass the size of the first item in our leaf, as we don't want to COW
       the left leaf if we can't at least push the first item of our leaf;

    2) When we call push_leaf_right(), we also pass a value of 1 byte as the
       data size parameter of push_leaf_right(). Like the previous case, it
       will also result in COWing the right leaf even if we are not able to
       move any items into it, since there can't be any item with a size
       smaller than 25 bytes (the size of struct btrfs_item).

       So instead of passing 1 as the minimum data size to push_leaf_right(),
       pass a size that corresponds to the sum of the size of all the
       remaining items in our leaf. We are not interested in moving less than
       that, because if we do, we are not able to delete our leaf and we have
       COWed the right leaf for nothing. Plus, moving only some of the items
       of our leaf, it means an even less balanced tree.

       Just like the previous case, we want to avoid the useless COW of the
       right leaf, this way we don't have to spend time allocating one new
       metadata extent, and doing more IO when committing a transaction or
       syncing a log tree. For the log tree case it's specially more important
       because more IO can result in a higher latency for a fsync operation.

    So adjust the minimum data size passed to push_leaf_left() and
    push_leaf_right() as mentioned above.

    This change if part of a patchset that is comprised of the following
    patches:

      1/6 btrfs: remove unnecessary leaf free space checks when pushing items
      2/6 btrfs: avoid unnecessary COW of leaves when deleting items from a leaf
      3/6 btrfs: avoid unnecessary computation when deleting items from a leaf
      4/6 btrfs: remove constraint on number of visited leaves when replacing extents
      5/6 btrfs: remove useless path release in the fast fsync path
      6/6 btrfs: prepare extents to be logged before locking a log tree path

    Not being able to delete a leaf that became less than 1/3 full after
    deleting items from it is actually common. For example, for the fio test
    mentioned in the changelog of patch 6/6, we are only able to delete a
    leaf at btrfs_del_items() about 5.3% of the time, due to its left and
    right neighbour leaves not having enough free space to push all the
    remaining items into them.

    The last patch in the series has some performance test result in its
    changelog.

    Signed-off-by: Filipe Manana <fdmanana@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 1b956d9978687f9d463a39c0d66a5eab958b9f3a
Author: Filipe Manana <fdmanana@suse.com>
Date:   Thu Feb 3 14:55:45 2022 +0000

    btrfs: remove unnecessary leaf free space checks when pushing items

    When trying to push items from a leaf into its left and right neighbours,
    we lock the left or right leaf, check if it has the required minimum free
    space, COW the leaf and then check again if it has the minimum required
    free space. This second check is pointless:

    1) Most and foremost because it's not needed. We have a write lock on the
       leaf and on its parent node, so no one can come in and change either
       the pre-COW or post-COW version of the leaf for the whole duration of
       the push_leaf_left() and push_leaf_right() calls;

    2) The call to btrfs_leaf_free_space() is not trivial, it has a fair
       amount of arithmetic operations and access to fields in the leaf's
       header and items, so it's not very cheap.

    So remove the duplicated free space checks.

    This change if part of a patchset that is comprised of the following
    patches:

      1/6 btrfs: remove unnecessary leaf free space checks when pushing items
      2/6 btrfs: avoid unnecessary COW of leaves when deleting items from a leaf
      3/6 btrfs: avoid unnecessary computation when deleting items from a leaf
      4/6 btrfs: remove constraint on number of visited leaves when replacing extents
      5/6 btrfs: remove useless path release in the fast fsync path
      6/6 btrfs: prepare extents to be logged before locking a log tree path

    The last patch in the series has some performance test result in its
    changelog.

    Signed-off-by: Filipe Manana <fdmanana@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 784a4d85814dea9bf1096e864d9368d032c0112a
Author: David Sterba <dsterba@suse.com>
Date:   Tue Feb 1 15:42:07 2022 +0100

    btrfs: replace BUILD_BUG_ON by static_assert

    The static_assert introduced in 6bab69c65013 ("build_bug.h: add wrapper
    for _Static_assert") has been supported by compilers for a long time
    (gcc 4.6, clang 3.0) and can be used in header files. We don't need to
    put BUILD_BUG_ON to random functions but rather keep it next to the
    definition.

    The exception here is the UAPI header btrfs_tree.h that could be
    potentially included by userspace code and the static assert is not
    defined (nor used in any other header).

    Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 2e0a36dc72d2a11fad03763dbd2ac4da106cfd1f
Author: Qu Wenruo <wqu@suse.com>
Date:   Sun Jan 30 20:53:15 2022 +0800

    btrfs: don't hold CPU for too long when defragging a file

    There is a user report about "btrfs filesystem defrag" causing 120s
    timeout problem.

    For btrfs_defrag_file() it will iterate all file extents if called from
    defrag ioctl, thus it can take a long time.

    There is no reason not to release the CPU during such a long operation.

    Add cond_resched() after defragged one cluster.

    CC: stable@vger.kernel.org # 5.16
    Link: https://lore.kernel.org/linux-btrfs/10e51417-2203-f0a4-2021-86c8511cc367@gmx.com
    Signed-off-by: Qu Wenruo <wqu@suse.com>
    Reviewed-by: David Sterba <dsterba@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit cb792362499dacbdd3986d10ad109d0efd875eab
Author: Josef Bacik <josef@toxicpanda.com>
Date:   Fri Nov 5 16:45:28 2021 -0400

    btrfs: rework async transaction committing

    Currently we do this awful thing where we get another ref on a trans
    handle, async off that handle and commit the transaction from that work.
    Because we do this we have to mess with current->journal_info and the
    freeze counting stuff.

    We already have an async thing to kick for the transaction commit, the
    transaction kthread.  Replace this work struct with a flag on the
    fs_info to tell the kthread to go ahead and commit even if it's before
    our timeout.  Then we can drastically simplify the async transaction
    commit path.

    Note: this can be simplified and functionality based on the pending
    operation COMMIT.

    Signed-off-by: Josef Bacik <josef@toxicpanda.com>
    [ add note ]
    Reviewed-by: David Sterba <dsterba@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 42786287b2e3443e3a3c90d7305fdc9b4287f00b
Author: Nikolay Borisov <nborisov@suse.com>
Date:   Tue Nov 23 09:23:42 2021 +0200

    btrfs: eliminate if in main loop in tree_search_offset

    Reshuffle the code inside the first loop of tree_search_offset so that
    one if() is eliminated and the becomes more linear.

    Reviewed-by: Josef Bacik <josef@toxicpanda.com>
    Signed-off-by: Nikolay Borisov <nborisov@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit e8bbc534dbd1d0b0d17bf68d7615e644513e652a
Author: Qu Wenruo <wqu@suse.com>
Date:   Fri Nov 19 14:19:33 2021 +0800

    btrfs: don't check stripe length if the profile is not stripe based

    [BUG]
    When debugging calc_bio_boundaries(), I found that even for RAID1
    metadata, we're following stripe length to calculate stripe boundary.

      # mkfs.btrfs -m raid1 -d raid1 /dev/test/scratch[12]
      # mount /dev/test/scratch /mnt/btrfs
      # xfs_io -f -c "pwrite 0 64K" /mnt/btrfs/file
      # umount

    Above very basic operations will make calc_bio_boundaries() to report
    the following result:

      submit_extent_page: r/i=1/1 file_offset=22036480 len_to_stripe_boundary=49152
      submit_extent_page: r/i=1/1 file_offset=30474240 len_to_stripe_boundary=65536
      ...
      submit_extent_page: r/i=1/1 file_offset=30523392 len_to_stripe_boundary=16384
      submit_extent_page: r/i=1/1 file_offset=30457856 len_to_stripe_boundary=16384
      submit_extent_page: r/i=5/257 file_offset=0 len_to_stripe_boundary=65536
      submit_extent_page: r/i=5/257 file_offset=65536 len_to_stripe_boundary=65536
      submit_extent_page: r/i=1/1 file_offset=30490624 len_to_stripe_boundary=49152
      submit_extent_page: r/i=1/1 file_offset=30507008 len_to_stripe_boundary=32768

    Where "r/i" is the rootid and inode, 1/1 means they metadata.
    The remaining names match the member used in kernel.

    Even all data/metadata are using RAID1, we're still following stripe
    length.

    [CAUSE]
    This behavior is caused by a wrong condition in btrfs_get_io_geometry():

    	if (map->type & BTRFS_BLOCK_GROUP_PROFILE_MASK) {
    		/* Fill using stripe_len */
    		len = min_t(u64, em->len - offset, max_len);
    	} else {
    		len = em->len - offset;
    	}

    This means, only for SINGLE we will not follow stripe_len.

    However for profiles like RAID1*, DUP, they don't need to bother
    stripe_len.

    This can lead to unnecessary bio split for RAID1*/DUP profiles, and can
    even be a blockage for future zoned RAID support.

    [FIX]
    Introduce one single-use macro, BTRFS_BLOCK_GROUP_STRIPE_MASK, and
    change the condition to only calculate the length using stripe length
    for stripe based profiles.

    Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
    Reviewed-by: Anand Jain <anand.jain@oracle.com>
    Signed-off-by: Qu Wenruo <wqu@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit b8cb209bdc55ae144881b2ae67dd36941813f970
Author: Nikolay Borisov <nborisov@suse.com>
Date:   Mon Nov 22 17:16:46 2021 +0200

    btrfs: get next entry in tree_search_offset before doing checks

    This is a small optimisation since the currently 'entry' is already
    checked in the if () {} else if {} construct above the loop. In essence
    the first iteration of the final while loop is redundant. To eliminate
    this extra check simply get the next entry at the beginning of the loop.

    Reviewed-by: Josef Bacik <josef@toxicpanda.com>
    Signed-off-by: Nikolay Borisov <nborisov@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 0fe871af1b0056f912e654d5455486c9f76b0c5e
Author: Josef Bacik <josef@toxicpanda.com>
Date:   Thu Nov 18 16:33:15 2021 -0500

    btrfs: index free space entries on size

    Currently we index free space on offset only, because usually we have a
    hint from the allocator that we want to honor for locality reasons.
    However if we fail to use this hint we have to go back to a brute force
    search through the free space entries to find a large enough extent.

    With sufficiently fragmented free space this becomes quite expensive, as
    we have to linearly search all of the free space entries to find if we
    have a part that's long enough.

    To fix this add a cached rb tree to index based on free space entry
    bytes.  This will allow us to quickly look up the largest chunk in the
    free space tree for this block group, and stop searching once we've
    found an entry that is too small to satisfy our allocation.  We simply
    choose to use this tree if we're searching from the beginning of the
    block group, as we know we do not care about locality at that point.

    I wrote an allocator test that creates a 10TiB ram backed null block
    device and then fallocates random files until the file system is full.
    I think go through and delete all of the odd files.  Then I spawn 8
    threads that fallocate 64MiB files (1/2 our extent size cap) until the
    file system is full again.  I use bcc's funclatency to measure the
    latency of find_free_extent.  The baseline results are

         nsecs               : count     distribution
             0 -> 1          : 0        |                                        |
             2 -> 3          : 0        |                                        |
             4 -> 7          : 0        |                                        |
             8 -> 15         : 0        |                                        |
            16 -> 31         : 0        |                                        |
            32 -> 63         : 0        |                                        |
            64 -> 127        : 0        |                                        |
           128 -> 255        : 0        |                                        |
           256 -> 511        : 10356    |****                                    |
           512 -> 1023       : 58242    |*************************               |
          1024 -> 2047       : 74418    |********************************        |
          2048 -> 4095       : 90393    |****************************************|
          4096 -> 8191       : 79119    |***********************************     |
          8192 -> 16383      : 35614    |***************                         |
         16384 -> 32767      : 13418    |*****                                   |
         32768 -> 65535      : 12811    |*****                                   |
         65536 -> 131071     : 17090    |*******                                 |
        131072 -> 262143     : 26465    |***********                             |
        262144 -> 524287     : 40179    |*****************                       |
        524288 -> 1048575    : 55469    |************************                |
       1048576 -> 2097151    : 48807    |*********************                   |
       2097152 -> 4194303    : 26744    |***********                             |
       4194304 -> 8388607    : 35351    |***************                         |
       8388608 -> 16777215   : 13918    |******                                  |
      16777216 -> 33554431   : 21       |                                        |

    avg = 908079 nsecs, total: 580889071441 nsecs, count: 639690

    And the patch results are

         nsecs               : count     distribution
             0 -> 1          : 0        |                                        |
             2 -> 3          : 0        |                                        |
             4 -> 7          : 0        |                                        |
             8 -> 15         : 0        |                                        |
            16 -> 31         : 0        |                                        |
            32 -> 63         : 0        |                                        |
            64 -> 127        : 0        |                                        |
           128 -> 255        : 0        |                                        |
           256 -> 511        : 6883     |**                                      |
           512 -> 1023       : 54346    |*********************                   |
          1024 -> 2047       : 79170    |********************************        |
          2048 -> 4095       : 98890    |****************************************|
          4096 -> 8191       : 81911    |*********************************       |
          8192 -> 16383      : 27075    |**********                              |
         16384 -> 32767      : 14668    |*****                                   |
         32768 -> 65535      : 13251    |*****                                   |
         65536 -> 131071     : 15340    |******                                  |
        131072 -> 262143     : 26715    |**********                              |
        262144 -> 524287     : 43274    |*****************                       |
        524288 -> 1048575    : 53870    |*********************                   |
       1048576 -> 2097151    : 55368    |**********************                  |
       2097152 -> 4194303    : 41036    |****************                        |
       4194304 -> 8388607    : 24927    |**********                              |
       8388608 -> 16777215   : 33       |                                        |
      16777216 -> 33554431   : 9        |                                        |

    avg = 623599 nsecs, total: 397259314759 nsecs, count: 637042

    There's a little variation in the amount of calls done because of timing
    of the threads with metadata requirements, but the avg, total, and
    count's are relatively consistent between runs (usually within 2-5% of
    each other).  As you can see here we have around a 30% decrease in
    average latency with a 30% decrease in overall time spent in
    find_free_extent.

    Reviewed-by: Filipe Manana <fdmanana@suse.com>
    Signed-off-by: Josef Bacik <josef@toxicpanda.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit db7a62b00a8ff463466ceb7c68728d8bfcc2d65d
Author: Josef Bacik <josef@toxicpanda.com>
Date:   Thu Nov 18 16:33:14 2021 -0500

    btrfs: only use ->max_extent_size if it is set in the bitmap

    While adding self tests for my space index change I was hitting a
    problem where the space indexed tree wasn't returning the expected
    ->max_extent_size.  This is because we will skip searching any entry
    that doesn't have ->bytes >= the amount of bytes we want.  However we'll
    still set the max_extent_size based on that entry.  The problem is if we
    don't search the bitmap we won't have ->max_extent_size set properly, so
    we can't really trust it.

    This doesn't really result in a problem per-se, it can just result in us
    not finding contiguous area that may exist.  Fix the max_extent_size
    helper to return ->bytes if ->max_extent_size isn't set, and add a big
    comment explaining why we're doing this.

    Signed-off-by: Josef Bacik <josef@toxicpanda.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 1796c46d66a36d108c3d13292dd47020dd4e02e1
Author: Josef Bacik <josef@toxicpanda.com>
Date:   Thu Oct 21 14:58:37 2021 -0400

    btrfs: rename btrfs_item_end_nr to btrfs_item_data_end

    The name btrfs_item_end_nr() is a bit of a misnomer, as it's actually
    the offset of the end of the data the item points to.  In fact all of
    the helpers that we use btrfs_item_end_nr() use data in their name, like
    BTRFS_LEAF_DATA_SIZE() and leaf_data().  Rename to btrfs_item_data_end()
    to make it clear what this helper is giving us.

    Signed-off-by: Josef Bacik <josef@toxicpanda.com>
    Reviewed-by: David Sterba <dsterba@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 89b1779e5b64958f44df12119436d01621a6f87a
Author: Josef Bacik <josef@toxicpanda.com>
Date:   Thu Oct 21 14:58:36 2021 -0400

    btrfs: remove the btrfs_item_end() helper

    We're only using btrfs_item_end() from btrfs_item_end_nr(), so this can
    be collapsed.

    Signed-off-by: Josef Bacik <josef@toxicpanda.com>
    Reviewed-by: David Sterba <dsterba@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 8b3c0a73f1c9ae9506fb95f8581d35432f2bb8e8
Author: Josef Bacik <josef@toxicpanda.com>
Date:   Thu Oct 21 14:58:35 2021 -0400

    btrfs: drop the _nr from the item helpers

    Now that all call sites are using the slot number to modify item values,
    rename the SETGET helpers to raw_item_*(), and then rework the _nr()
    helpers to be the btrfs_item_*() btrfs_set_item_*() helpers, and then
    rename all of the callers to the new helpers.

    Signed-off-by: Josef Bacik <josef@toxicpanda.com>
    Reviewed-by: David Sterba <dsterba@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit cb7d27e5e1205b9ca8f512a48f6772b2eb2b84f9
Author: Josef Bacik <josef@toxicpanda.com>
Date:   Thu Oct 21 14:58:34 2021 -0400

    btrfs: introduce item_nr token variant helpers

    The last remaining place where we have the pattern of

    	item = btrfs_item_nr(slot)
    	<do something with the item>

    are the token helpers.  Handle this by introducing token helpers that
    will do the btrfs_item_nr() work inside of the helper itself, and then
    convert all users of the btrfs_item token helpers to the new _nr()
    variants.

    Signed-off-by: Josef Bacik <josef@toxicpanda.com>
    Reviewed-by: David Sterba <dsterba@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 18ba83dcd99b9619cfc1a246cfb84b1c9b530097
Author: Josef Bacik <josef@toxicpanda.com>
Date:   Thu Oct 21 14:58:33 2021 -0400

    btrfs: make btrfs_file_extent_inline_item_len take a slot

    Instead of getting the btrfs_item for this, simply pass in the slot of
    the item and then use the btrfs_item_size_nr() helper inside of
    btrfs_file_extent_inline_item_len().

    Signed-off-by: Josef Bacik <josef@toxicpanda.com>
    Reviewed-by: David Sterba <dsterba@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 898ecfdac6314335d8135f741414b0b6867ce1ab
Author: Josef Bacik <josef@toxicpanda.com>
Date:   Thu Oct 21 14:58:32 2021 -0400

    btrfs: add btrfs_set_item_*_nr() helpers

    We have the pattern of

    	item = btrfs_item_nr(slot);
    	btrfs_set_item_*(leaf, item);

    in a bunch of places in our code.  Fix this by adding
    btrfs_set_item_*_nr() helpers which will do the appropriate work, and
    replace those calls with

    	btrfs_set_item_*_nr(leaf, slot);

    Signed-off-by: Josef Bacik <josef@toxicpanda.com>
    Reviewed-by: David Sterba <dsterba@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 35a9546592ae8d0f3348cd0f776a12dbf3b81aa8
Author: Josef Bacik <josef@toxicpanda.com>
Date:   Thu Oct 21 14:58:31 2021 -0400

    btrfs: use btrfs_item_size_nr/btrfs_item_offset_nr everywhere

    We have this pattern in a lot of places

    	item = btrfs_item_nr(slot);
    	btrfs_item_size(leaf, item);

    when we could simply use

    	btrfs_item_size(leaf, slot);

    Fix all callers of btrfs_item_size() and btrfs_item_offset() to use the
    _nr variation of the helpers.

    Reviewed-by: Qu Wenruo <wqu@suse.com>
    Signed-off-by: Josef Bacik <josef@toxicpanda.com>
    Reviewed-by: David Sterba <dsterba@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 8e279d85c2ac593b8caf53f3ac72d0b7047d96f5
Author: Filipe Manana <fdmanana@suse.com>
Date:   Mon Oct 25 17:31:54 2021 +0100

    btrfs: remove no longer needed logic for replaying directory deletes

    Now that we log only dir index keys when logging a directory, we no longer
    need to deal with dir item keys in the log replay code for replaying
    directory deletes. This is also true for the case when we replay a log
    tree created by a kernel that still logs dir items.

    So remove the remaining code of the replay of directory deletes algorithm
    that deals with dir item keys.

    Reviewed-by: Josef Bacik <josef@toxicpanda.com>
    Signed-off-by: Filipe Manana <fdmanana@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit b8ac7a8b1dd9bba3b30bd034d754bb4932c4a970
Author: Filipe Manana <fdmanana@suse.com>
Date:   Mon Oct 25 17:31:53 2021 +0100

    btrfs: only copy dir index keys when logging a directory

    Currently, when logging a directory, we copy both dir items and dir index
    items from the fs/subvolume tree to the log tree. Both items have exactly
    the same data (same struct btrfs_dir_item), the difference lies in the key
    values, where a dir index key contains the index number of a directory
    entry while the dir item key does not, as it's used for doing fast lookups
    of an entry by name, while the former is used for sorting entries when
    listing a directory.

    We can exploit that and log only the dir index items, since they contain
    all the information needed to correctly add, replace and delete directory
    entries when replaying a log tree. Logging only the dir index items is
    also backward and forward compatible: an unpatched kernel (without this
    change) can correctly replay a log tree generated by a patched kernel
    (with this patch), and a patched kernel can correctly replay a log tree
    generated by an unpatched kernel.

    The backward compatibility is ensured because:

    1) For inserting a new dentry: a dentry is only inserted when we find a
       new dir index key - we can only insert if we know the dir index offset,
       which is encoded in the dir index key's offset;

    2) For deleting dentries: during log replay, before adding or replacing
       dentries, we first replay dentry deletions. Whenever we find a dir item
       key or a dir index key in the subvolume/fs tree that is not logged in
       a range for which the log tree is authoritative, we do the unlink of
       the dentry, which removes both the existing dir item key and the dir
       index key. Therefore logging just dir index keys is enough to ensure
       dentry deletions are correctly replayed;

    3) For dentry replacements: they work when we log only dir index keys
       and this is mostly due to a combination of 1) and 2). If we replace a
       dentry with name "foobar" to point from inode A to inode B, then we
       know the dir index key for the new dentry is different from the old
       one, as it has an index number (key offset) larger than the old one.
       This results in replaying a deletion, through replay_dir_deletes(),
       that causes the old dentry to be removed, both the dir item key and
       the dir index key, as mentioned at 2). Then when processing the new
       dir index key, we add the new dentry, adding both a new dir item key
       and a new index key pointing to inode B, as stated in 1).

    The forward compatibility, the ability for a patched kernel to replay a
    log created by an older, unpatched kernel, comes from the changes required
    for making sure we are able to replay a log that only contains dir index
    keys - we simply ignore every dir item key we find.

    So modify directory logging to log only dir index items, and modify the
    log replay process to ignore dir item keys, from log trees created by an
    unpatched kernel, and process only with dir index keys. This reduces the
    amount of logged metadata by about half, and therefore the time spent
    logging or fsyncing large directories (less CPU time and less IO).

    The following test script was used to measure this change:

       #!/bin/bash

       DEV=/dev/nvme0n1
       MNT=/mnt/nvme0n1

       NUM_NEW_FILES=1000000
       NUM_FILE_DELETES=10000

       mkfs.btrfs -f $DEV
       mount -o ssd $DEV $MNT

       mkdir $MNT/testdir

       for ((i = 1; i <= $NUM_NEW_FILES; i++)); do
               echo -n > $MNT/testdir/file_$i
       done

       start=$(date +%s%N)
       xfs_io -c "fsync" $MNT/testdir
       end=$(date +%s%N)

       dur=$(( (end - start) / 1000000 ))
       echo "dir fsync took $dur ms after adding $NUM_NEW_FILES files"

       # sync to force transaction commit and wipeout the log.
       sync

       del_inc=$(( $NUM_NEW_FILES / $NUM_FILE_DELETES ))
       for ((i = 1; i <= $NUM_NEW_FILES; i += $del_inc)); do
               rm -f $MNT/testdir/file_$i
       done

       start=$(date +%s%N)
       xfs_io -c "fsync" $MNT/testdir
       end=$(date +%s%N)

       dur=$(( (end - start) / 1000000 ))
       echo "dir fsync took $dur ms after deleting $NUM_FILE_DELETES files"
       echo

       umount $MNT

    The tests were run on a physical machine, with a non-debug kernel (Debian's
    default kernel config), for different values of $NUM_NEW_FILES and
    $NUM_FILE_DELETES, and the results were the following:

    ** Before patch, NUM_NEW_FILES = 1 000 000, NUM_DELETE_FILES = 10 000 **

    dir fsync took 8412 ms after adding 1000000 files
    dir fsync took 500 ms after deleting 10000 files

    ** After patch, NUM_NEW_FILES = 1 000 000, NUM_DELETE_FILES = 10 000 **

    dir fsync took 4252 ms after adding 1000000 files   (-49.5%)
    dir fsync took 269 ms after deleting 10000 files    (-46.2%)

    ** Before patch, NUM_NEW_FILES = 100 000, NUM_DELETE_FILES = 1 000 **

    dir fsync took 745 ms after adding 100000 files
    dir fsync took 59 ms after deleting 1000 files

    ** After patch, NUM_NEW_FILES = 100 000, NUM_DELETE_FILES = 1 000 **

    dir fsync took 404 ms after adding 100000 files   (-45.8%)
    dir fsync took 31 ms after deleting 1000 files    (-47.5%)

    ** Before patch, NUM_NEW_FILES = 10 000, NUM_DELETE_FILES = 1 000 **

    dir fsync took 67 ms after adding 10000 files
    dir fsync took 9 ms after deleting 1000 files

    ** After patch, NUM_NEW_FILES = 10 000, NUM_DELETE_FILES = 1 000 **

    dir fsync took 36 ms after adding 10000 files   (-46.3%)
    dir fsync took 5 ms after deleting 1000 files   (-44.4%)

    ** Before patch, NUM_NEW_FILES = 1 000, NUM_DELETE_FILES = 100 **

    dir fsync took 9 ms after adding 1000 files
    dir fsync took 4 ms after deleting 100 files

    ** After patch, NUM_NEW_FILES = 1 000, NUM_DELETE_FILES = 100 **

    dir fsync took 7 ms after adding 1000 files     (-22.2%)
    dir fsync took 3 ms after deleting 100 files    (-25.0%)

    Reviewed-by: Josef Bacik <josef@toxicpanda.com>
    Signed-off-by: Filipe Manana <fdmanana@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 793440833c94fa896424e2ceef71d376a2ae2454
Author: Nikolay Borisov <nborisov@suse.com>
Date:   Thu Oct 14 10:03:11 2021 +0300

    btrfs: remove spurious unlock/lock of unused_bgs_lock

    Since both unused block groups and reclaim bgs lists are protected by
    unused_bgs_lock then free them in the same critical section without
    doing an extra unlock/lock pair.

    Reviewed-by: Josef Bacik <josef@toxicpanda.com>
    Signed-off-by: Nikolay Borisov <nborisov@suse.com>
    Reviewed-by: David Sterba <dsterba@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>

commit 9f4889dbcf40db16d5cfd02dae54143ecfcf036a
Author: Filipe Manana <fdmanana@suse.com>
Date:   Thu Oct 28 16:03:41 2021 +0100

    btrfs: fix ENOSPC failure when attempting direct IO write into NOCOW range

    When doing a direct IO write against a file range that either has
    preallocated extents in that range or has regular extents and the file
    has the NOCOW attribute set, the write fails with -ENOSPC when all of
    the following conditions are met:

    1) There are no data blocks groups with enough free space matching
       the size of the write;

    2) There's not enough unallocated space for allocating a new data block
       group;

    3) The extents in the target file range are not shared, neither through
       snapshots nor through reflinks.

    This is wrong because a NOCOW write can be done in such case, and in fact
    it's possible to do it using a buffered IO write, since when failing to
    allocate data space, the buffered IO path checks if a NOCOW write is
    possible.

    The failure in direct IO write path comes from the fact that early on,
    at btrfs_dio_iomap_begin(), we try to allocate data space for the write
    and if it that fails we return the error and stop - we never check if we
    can do NOCOW. But later, at btrfs_get_blocks_direct_write(), we check
    if we can do a NOCOW write into the range, or a subset of the range, and
    then release the previously reserved data space.

    Fix this by doing the data reservation only if needed, when we must COW,
    at btrfs_get_blocks_direct_write() instead of doing it at
    btrfs_dio_iomap_begin(). This also simplifies a bit the logic and removes
    the inneficiency of doing unnecessary data reservations.

    The following example test script reproduces the problem:

      $ cat dio-nocow-enospc.sh
      #!/bin/bash

      DEV=/dev/sdj
      MNT=/mnt/sdj

      # Use a small fixed size (1G) filesystem so that it's quick to fill
      # it up.
      # Make sure the mixed block groups feature is not enabled because we
      # later want to not have more space available for allocating data
      # extents but still have enough metadata space free for the file writes.
      mkfs.btrfs -f -b $((1024 * 1024 * 1024)) -O ^mixed-bg $DEV
      mount $DEV $MNT

      # Create our test file with the NOCOW attribute set.
      touch $MNT/foobar
      chattr +C $MNT/foobar

      # Now fill in all unallocated space with data for our test file.
      # This will allocate a data block group that will be full and leave
      # no (or a very small amount of) unallocated space in the device, so
      # that it will not be possible to allocate a new block group later.
      echo
      echo "Creating test file with initial data..."
      xfs_io -c "pwrite -S 0xab -b 1M 0 900M" $MNT/foobar

      # Now try a direct IO write against file range [0, 10M[.
      # This should succeed since this is a NOCOW file and an extent for the
      # range was previously allocated.
      echo
      echo "Trying direct IO write over allocated space..."
      xfs_io -d -c "pwrite -S 0xcd -b 10M 0 10M" $MNT/foobar

      umount $MNT

    When running the test:

      $ ./dio-nocow-enospc.sh
      (...)

      Creating test file with initial data...
      wrote 943718400/943718400 bytes at offset 0
      900 MiB, 900 ops; 0:00:01.43 (625.526 MiB/sec and 625.5265 ops/sec)

      Trying direct IO write over allocated space...
      pwrite: No space left on device

    A test case for fstests will follow, testing both this direct IO write
    scenario as well as the buffered IO write scenario to make it less likely
    to get future regressions on the buffered IO case.

    Reviewed-by: Josef Bacik <josef@toxicpanda.com>
    Signed-off-by: Filipe Manana <fdmanana@suse.com>
    Signed-off-by: David Sterba <dsterba@suse.com>
---
 fs/btrfs/backref.c                   |  16 +-
 fs/btrfs/block-group.c               |   2 -
 fs/btrfs/btrfs_inode.h               |  18 +-
 fs/btrfs/compression.h               |   2 +
 fs/btrfs/ctree.c                     | 208 ++++-----
 fs/btrfs/ctree.h                     |  96 ++--
 fs/btrfs/dev-replace.c               |   4 +-
 fs/btrfs/dir-item.c                  |  12 +-
 fs/btrfs/disk-io.c                   |   3 +-
 fs/btrfs/extent-tree.c               |  14 +-
 fs/btrfs/file-item.c                 |  25 +-
 fs/btrfs/file.c                      |  26 +-
 fs/btrfs/free-space-cache.c          | 177 ++++++--
 fs/btrfs/free-space-cache.h          |   2 +
 fs/btrfs/inode-item.c                |  14 +-
 fs/btrfs/inode.c                     | 172 +++----
 fs/btrfs/ioctl.c                     | 258 +++++++----
 fs/btrfs/lzo.c                       |  20 +-
 fs/btrfs/print-tree.c                |   8 +-
 fs/btrfs/props.c                     |   2 +-
 fs/btrfs/qgroup.c                    |  41 +-
 fs/btrfs/ref-verify.c                |   2 +-
 fs/btrfs/reflink.c                   |   2 +-
 fs/btrfs/relocation.c                |   2 +-
 fs/btrfs/root-tree.c                 |   4 +-
 fs/btrfs/scrub.c                     |   2 +-
 fs/btrfs/send.c                      |  22 +-
 fs/btrfs/sysfs.c                     |  10 +-
 fs/btrfs/tests/extent-buffer-tests.c |  17 +-
 fs/btrfs/transaction.c               |  76 +---
 fs/btrfs/transaction.h               |   2 +-
 fs/btrfs/tree-checker.c              |  56 +--
 fs/btrfs/tree-log.c                  | 656 ++++++++++++---------------
 fs/btrfs/uuid-tree.c                 |  10 +-
 fs/btrfs/verity.c                    |   2 +-
 fs/btrfs/volumes.c                   |  13 +-
 fs/btrfs/xattr.c                     |   8 +-
 include/uapi/linux/btrfs.h           |   6 +-
 include/uapi/linux/btrfs_tree.h      |   4 +-
 39 files changed, 1062 insertions(+), 952 deletions(-)

diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c
index 8b090c40daf7..3b0c4bed242e 100644
--- a/fs/btrfs/backref.c
+++ b/fs/btrfs/backref.c
@@ -950,7 +950,7 @@ static int add_inline_refs(const struct btrfs_fs_info *fs_info,
 	leaf = path->nodes[0];
 	slot = path->slots[0];
 
-	item_size = btrfs_item_size_nr(leaf, slot);
+	item_size = btrfs_item_size(leaf, slot);
 	BUG_ON(item_size < sizeof(*ei));
 
 	ei = btrfs_item_ptr(leaf, slot, struct btrfs_extent_item);
@@ -1792,7 +1792,7 @@ int extent_from_logical(struct btrfs_fs_info *fs_info, u64 logical,
 	}
 
 	eb = path->nodes[0];
-	item_size = btrfs_item_size_nr(eb, path->slots[0]);
+	item_size = btrfs_item_size(eb, path->slots[0]);
 	BUG_ON(item_size < sizeof(*ei));
 
 	ei = btrfs_item_ptr(eb, path->slots[0], struct btrfs_extent_item);
@@ -2071,7 +2071,6 @@ static int iterate_inode_refs(u64 inum, struct btrfs_root *fs_root,
 	u64 parent = 0;
 	int found = 0;
 	struct extent_buffer *eb;
-	struct btrfs_item *item;
 	struct btrfs_inode_ref *iref;
 	struct btrfs_key found_key;
 
@@ -2097,10 +2096,9 @@ static int iterate_inode_refs(u64 inum, struct btrfs_root *fs_root,
 		}
 		btrfs_release_path(path);
 
-		item = btrfs_item_nr(slot);
 		iref = btrfs_item_ptr(eb, slot, struct btrfs_inode_ref);
 
-		for (cur = 0; cur < btrfs_item_size(eb, item); cur += len) {
+		for (cur = 0; cur < btrfs_item_size(eb, slot); cur += len) {
 			name_len = btrfs_inode_ref_name_len(eb, iref);
 			/* path must be released before calling iterate()! */
 			btrfs_debug(fs_root->fs_info,
@@ -2156,7 +2154,7 @@ static int iterate_inode_extrefs(u64 inum, struct btrfs_root *fs_root,
 		}
 		btrfs_release_path(path);
 
-		item_size = btrfs_item_size_nr(eb, slot);
+		item_size = btrfs_item_size(eb, slot);
 		ptr = btrfs_item_ptr_offset(eb, slot);
 		cur_offset = 0;
 
@@ -2377,7 +2375,7 @@ int btrfs_backref_iter_start(struct btrfs_backref_iter *iter, u64 bytenr)
 	iter->item_ptr = (u32)btrfs_item_ptr_offset(path->nodes[0],
 						    path->slots[0]);
 	iter->end_ptr = (u32)(iter->item_ptr +
-			btrfs_item_size_nr(path->nodes[0], path->slots[0]));
+			btrfs_item_size(path->nodes[0], path->slots[0]));
 	ei = btrfs_item_ptr(path->nodes[0], path->slots[0],
 			    struct btrfs_extent_item);
 
@@ -2417,7 +2415,7 @@ int btrfs_backref_iter_start(struct btrfs_backref_iter *iter, u64 bytenr)
 		iter->cur_ptr = (u32)btrfs_item_ptr_offset(path->nodes[0],
 							   path->slots[0]);
 		iter->item_ptr = iter->cur_ptr;
-		iter->end_ptr = (u32)(iter->item_ptr + btrfs_item_size_nr(
+		iter->end_ptr = (u32)(iter->item_ptr + btrfs_item_size(
 				      path->nodes[0], path->slots[0]));
 	}
 
@@ -2482,7 +2480,7 @@ int btrfs_backref_iter_next(struct btrfs_backref_iter *iter)
 	iter->item_ptr = (u32)btrfs_item_ptr_offset(path->nodes[0],
 					path->slots[0]);
 	iter->cur_ptr = iter->item_ptr;
-	iter->end_ptr = iter->item_ptr + (u32)btrfs_item_size_nr(path->nodes[0],
+	iter->end_ptr = iter->item_ptr + (u32)btrfs_item_size(path->nodes[0],
 						path->slots[0]);
 	return 0;
 }
diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c
index b67c965725ea..27da1dfbd626 100644
--- a/fs/btrfs/block-group.c
+++ b/fs/btrfs/block-group.c
@@ -3924,9 +3924,7 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 		list_del_init(&block_group->bg_list);
 		btrfs_put_block_group(block_group);
 	}
-	spin_unlock(&info->unused_bgs_lock);
 
-	spin_lock(&info->unused_bgs_lock);
 	while (!list_empty(&info->reclaim_bgs)) {
 		block_group = list_first_entry(&info->reclaim_bgs,
 					       struct btrfs_block_group,
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index ab2a4a52e0bb..b3e46aabc3d8 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -138,19 +138,11 @@ struct btrfs_inode {
 	/* a local copy of root's last_log_commit */
 	int last_log_commit;
 
-	union {
-		/*
-		 * Total number of bytes pending delalloc, used by stat to
-		 * calculate the real block usage of the file. This is used
-		 * only for files.
-		 */
-		u64 delalloc_bytes;
-		/*
-		 * The offset of the last dir item key that was logged.
-		 * This is used only for directories.
-		 */
-		u64 last_dir_item_offset;
-	};
+	/*
+	 * Total number of bytes pending delalloc, used by stat to calculate the
+	 * real block usage of the file. This is used only for files.
+	 */
+	u64 delalloc_bytes;
 
 	union {
 		/*
diff --git a/fs/btrfs/compression.h b/fs/btrfs/compression.h
index 56eef0821e3e..7dbd14caab01 100644
--- a/fs/btrfs/compression.h
+++ b/fs/btrfs/compression.h
@@ -22,6 +22,8 @@ struct btrfs_inode;
 
 /* Maximum length of compressed data stored on disk */
 #define BTRFS_MAX_COMPRESSED		(SZ_128K)
+static_assert((BTRFS_MAX_COMPRESSED % PAGE_SIZE) == 0);
+
 /* Maximum size of data before compression */
 #define BTRFS_MAX_UNCOMPRESSED		(SZ_128K)
 
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 35660791e084..5ca7a535d109 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -2627,19 +2627,14 @@ static noinline int split_node(struct btrfs_trans_handle *trans,
  */
 static int leaf_space_used(struct extent_buffer *l, int start, int nr)
 {
-	struct btrfs_item *start_item;
-	struct btrfs_item *end_item;
 	int data_len;
 	int nritems = btrfs_header_nritems(l);
 	int end = min(nritems, start + nr) - 1;
 
 	if (!nr)
 		return 0;
-	start_item = btrfs_item_nr(start);
-	end_item = btrfs_item_nr(end);
-	data_len = btrfs_item_offset(l, start_item) +
-		   btrfs_item_size(l, start_item);
-	data_len = data_len - btrfs_item_offset(l, end_item);
+	data_len = btrfs_item_offset(l, start) + btrfs_item_size(l, start);
+	data_len = data_len - btrfs_item_offset(l, end);
 	data_len += sizeof(struct btrfs_item) * nr;
 	WARN_ON(data_len < 0);
 	return data_len;
@@ -2686,7 +2681,6 @@ static noinline int __push_leaf_right(struct btrfs_path *path,
 	u32 i;
 	int push_space = 0;
 	int push_items = 0;
-	struct btrfs_item *item;
 	u32 nr;
 	u32 right_nritems;
 	u32 data_end;
@@ -2703,8 +2697,6 @@ static noinline int __push_leaf_right(struct btrfs_path *path,
 	slot = path->slots[1];
 	i = left_nritems - 1;
 	while (i >= nr) {
-		item = btrfs_item_nr(i);
-
 		if (!empty && push_items > 0) {
 			if (path->slots[0] > i)
 				break;
@@ -2719,12 +2711,13 @@ static noinline int __push_leaf_right(struct btrfs_path *path,
 		if (path->slots[0] == i)
 			push_space += data_size;
 
-		this_item_size = btrfs_item_size(left, item);
-		if (this_item_size + sizeof(*item) + push_space > free_space)
+		this_item_size = btrfs_item_size(left, i);
+		if (this_item_size + sizeof(struct btrfs_item) +
+		    push_space > free_space)
 			break;
 
 		push_items++;
-		push_space += this_item_size + sizeof(*item);
+		push_space += this_item_size + sizeof(struct btrfs_item);
 		if (i == 0)
 			break;
 		i--;
@@ -2738,7 +2731,7 @@ static noinline int __push_leaf_right(struct btrfs_path *path,
 	/* push left to right */
 	right_nritems = btrfs_header_nritems(right);
 
-	push_space = btrfs_item_end_nr(left, left_nritems - push_items);
+	push_space = btrfs_item_data_end(left, left_nritems - push_items);
 	push_space -= leaf_data_end(left);
 
 	/* make room in the right data area */
@@ -2769,9 +2762,8 @@ static noinline int __push_leaf_right(struct btrfs_path *path,
 	btrfs_set_header_nritems(right, right_nritems);
 	push_space = BTRFS_LEAF_DATA_SIZE(fs_info);
 	for (i = 0; i < right_nritems; i++) {
-		item = btrfs_item_nr(i);
-		push_space -= btrfs_token_item_size(&token, item);
-		btrfs_set_token_item_offset(&token, item, push_space);
+		push_space -= btrfs_token_item_size(&token, i);
+		btrfs_set_token_item_offset(&token, i, push_space);
 	}
 
 	left_nritems -= push_items;
@@ -2856,16 +2848,11 @@ static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root
 	if (free_space < data_size)
 		goto out_unlock;
 
-	/* cow and double check */
 	ret = btrfs_cow_block(trans, root, right, upper,
 			      slot + 1, &right, BTRFS_NESTING_RIGHT_COW);
 	if (ret)
 		goto out_unlock;
 
-	free_space = btrfs_leaf_free_space(right);
-	if (free_space < data_size)
-		goto out_unlock;
-
 	left_nritems = btrfs_header_nritems(left);
 	if (left_nritems == 0)
 		goto out_unlock;
@@ -2916,7 +2903,6 @@ static noinline int __push_leaf_left(struct btrfs_path *path, int data_size,
 	int i;
 	int push_space = 0;
 	int push_items = 0;
-	struct btrfs_item *item;
 	u32 old_left_nritems;
 	u32 nr;
 	int ret = 0;
@@ -2930,8 +2916,6 @@ static noinline int __push_leaf_left(struct btrfs_path *path, int data_size,
 		nr = min(right_nritems - 1, max_slot);
 
 	for (i = 0; i < nr; i++) {
-		item = btrfs_item_nr(i);
-
 		if (!empty && push_items > 0) {
 			if (path->slots[0] < i)
 				break;
@@ -2946,12 +2930,13 @@ static noinline int __push_leaf_left(struct btrfs_path *path, int data_size,
 		if (path->slots[0] == i)
 			push_space += data_size;
 
-		this_item_size = btrfs_item_size(right, item);
-		if (this_item_size + sizeof(*item) + push_space > free_space)
+		this_item_size = btrfs_item_size(right, i);
+		if (this_item_size + sizeof(struct btrfs_item) + push_space >
+		    free_space)
 			break;
 
 		push_items++;
-		push_space += this_item_size + sizeof(*item);
+		push_space += this_item_size + sizeof(struct btrfs_item);
 	}
 
 	if (push_items == 0) {
@@ -2967,25 +2952,23 @@ static noinline int __push_leaf_left(struct btrfs_path *path, int data_size,
 			   push_items * sizeof(struct btrfs_item));
 
 	push_space = BTRFS_LEAF_DATA_SIZE(fs_info) -
-		     btrfs_item_offset_nr(right, push_items - 1);
+		     btrfs_item_offset(right, push_items - 1);
 
 	copy_extent_buffer(left, right, BTRFS_LEAF_DATA_OFFSET +
 		     leaf_data_end(left) - push_space,
 		     BTRFS_LEAF_DATA_OFFSET +
-		     btrfs_item_offset_nr(right, push_items - 1),
+		     btrfs_item_offset(right, push_items - 1),
 		     push_space);
 	old_left_nritems = btrfs_header_nritems(left);
 	BUG_ON(old_left_nritems <= 0);
 
 	btrfs_init_map_token(&token, left);
-	old_left_item_size = btrfs_item_offset_nr(left, old_left_nritems - 1);
+	old_left_item_size = btrfs_item_offset(left, old_left_nritems - 1);
 	for (i = old_left_nritems; i < old_left_nritems + push_items; i++) {
 		u32 ioff;
 
-		item = btrfs_item_nr(i);
-
-		ioff = btrfs_token_item_offset(&token, item);
-		btrfs_set_token_item_offset(&token, item,
+		ioff = btrfs_token_item_offset(&token, i);
+		btrfs_set_token_item_offset(&token, i,
 		      ioff - (BTRFS_LEAF_DATA_SIZE(fs_info) - old_left_item_size));
 	}
 	btrfs_set_header_nritems(left, old_left_nritems + push_items);
@@ -2996,7 +2979,7 @@ static noinline int __push_leaf_left(struct btrfs_path *path, int data_size,
 		       right_nritems);
 
 	if (push_items < right_nritems) {
-		push_space = btrfs_item_offset_nr(right, push_items - 1) -
+		push_space = btrfs_item_offset(right, push_items - 1) -
 						  leaf_data_end(right);
 		memmove_extent_buffer(right, BTRFS_LEAF_DATA_OFFSET +
 				      BTRFS_LEAF_DATA_SIZE(fs_info) - push_space,
@@ -3014,10 +2997,8 @@ static noinline int __push_leaf_left(struct btrfs_path *path, int data_size,
 	btrfs_set_header_nritems(right, right_nritems);
 	push_space = BTRFS_LEAF_DATA_SIZE(fs_info);
 	for (i = 0; i < right_nritems; i++) {
-		item = btrfs_item_nr(i);
-
-		push_space = push_space - btrfs_token_item_size(&token, item);
-		btrfs_set_token_item_offset(&token, item, push_space);
+		push_space = push_space - btrfs_token_item_size(&token, i);
+		btrfs_set_token_item_offset(&token, i, push_space);
 	}
 
 	btrfs_mark_buffer_dirty(left);
@@ -3096,7 +3077,6 @@ static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root
 		goto out;
 	}
 
-	/* cow and double check */
 	ret = btrfs_cow_block(trans, root, left,
 			      path->nodes[1], slot - 1, &left,
 			      BTRFS_NESTING_LEFT_COW);
@@ -3107,12 +3087,6 @@ static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root
 		goto out;
 	}
 
-	free_space = btrfs_leaf_free_space(left);
-	if (free_space < data_size) {
-		ret = 1;
-		goto out;
-	}
-
 	if (check_sibling_keys(left, right)) {
 		ret = -EUCLEAN;
 		goto out;
@@ -3145,7 +3119,7 @@ static noinline void copy_for_split(struct btrfs_trans_handle *trans,
 
 	nritems = nritems - mid;
 	btrfs_set_header_nritems(right, nritems);
-	data_copy_size = btrfs_item_end_nr(l, mid) - leaf_data_end(l);
+	data_copy_size = btrfs_item_data_end(l, mid) - leaf_data_end(l);
 
 	copy_extent_buffer(right, l, btrfs_item_nr_offset(0),
 			   btrfs_item_nr_offset(mid),
@@ -3156,15 +3130,14 @@ static noinline void copy_for_split(struct btrfs_trans_handle *trans,
 		     data_copy_size, BTRFS_LEAF_DATA_OFFSET +
 		     leaf_data_end(l), data_copy_size);
 
-	rt_data_off = BTRFS_LEAF_DATA_SIZE(fs_info) - btrfs_item_end_nr(l, mid);
+	rt_data_off = BTRFS_LEAF_DATA_SIZE(fs_info) - btrfs_item_data_end(l, mid);
 
 	btrfs_init_map_token(&token, right);
 	for (i = 0; i < nritems; i++) {
-		struct btrfs_item *item = btrfs_item_nr(i);
 		u32 ioff;
 
-		ioff = btrfs_token_item_offset(&token, item);
-		btrfs_set_token_item_offset(&token, item, ioff + rt_data_off);
+		ioff = btrfs_token_item_offset(&token, i);
+		btrfs_set_token_item_offset(&token, i, ioff + rt_data_off);
 	}
 
 	btrfs_set_header_nritems(l, mid);
@@ -3280,7 +3253,7 @@ static noinline int split_leaf(struct btrfs_trans_handle *trans,
 
 	l = path->nodes[0];
 	slot = path->slots[0];
-	if (extend && data_size + btrfs_item_size_nr(l, slot) +
+	if (extend && data_size + btrfs_item_size(l, slot) +
 	    sizeof(struct btrfs_item) > BTRFS_LEAF_DATA_SIZE(fs_info))
 		return -EOVERFLOW;
 
@@ -3449,7 +3422,7 @@ static noinline int setup_leaf_for_split(struct btrfs_trans_handle *trans,
 	if (btrfs_leaf_free_space(leaf) >= ins_len)
 		return 0;
 
-	item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+	item_size = btrfs_item_size(leaf, path->slots[0]);
 	if (key.type == BTRFS_EXTENT_DATA_KEY) {
 		fi = btrfs_item_ptr(leaf, path->slots[0],
 				    struct btrfs_file_extent_item);
@@ -3469,7 +3442,7 @@ static noinline int setup_leaf_for_split(struct btrfs_trans_handle *trans,
 	ret = -EAGAIN;
 	leaf = path->nodes[0];
 	/* if our item isn't there, return now */
-	if (item_size != btrfs_item_size_nr(leaf, path->slots[0]))
+	if (item_size != btrfs_item_size(leaf, path->slots[0]))
 		goto err;
 
 	/* the leaf has  changed, it now has room.  return now */
@@ -3500,9 +3473,7 @@ static noinline int split_item(struct btrfs_path *path,
 			       unsigned long split_offset)
 {
 	struct extent_buffer *leaf;
-	struct btrfs_item *item;
-	struct btrfs_item *new_item;
-	int slot;
+	int orig_slot, slot;
 	char *buf;
 	u32 nritems;
 	u32 item_size;
@@ -3512,9 +3483,9 @@ static noinline int split_item(struct btrfs_path *path,
 	leaf = path->nodes[0];
 	BUG_ON(btrfs_leaf_free_space(leaf) < sizeof(struct btrfs_item));
 
-	item = btrfs_item_nr(path->slots[0]);
-	orig_offset = btrfs_item_offset(leaf, item);
-	item_size = btrfs_item_size(leaf, item);
+	orig_slot = path->slots[0];
+	orig_offset = btrfs_item_offset(leaf, path->slots[0]);
+	item_size = btrfs_item_size(leaf, path->slots[0]);
 
 	buf = kmalloc(item_size, GFP_NOFS);
 	if (!buf)
@@ -3535,14 +3506,12 @@ static noinline int split_item(struct btrfs_path *path,
 	btrfs_cpu_key_to_disk(&disk_key, new_key);
 	btrfs_set_item_key(leaf, &disk_key, slot);
 
-	new_item = btrfs_item_nr(slot);
+	btrfs_set_item_offset(leaf, slot, orig_offset);
+	btrfs_set_item_size(leaf, slot, item_size - split_offset);
 
-	btrfs_set_item_offset(leaf, new_item, orig_offset);
-	btrfs_set_item_size(leaf, new_item, item_size - split_offset);
-
-	btrfs_set_item_offset(leaf, item,
-			      orig_offset + item_size - split_offset);
-	btrfs_set_item_size(leaf, item, split_offset);
+	btrfs_set_item_offset(leaf, orig_slot,
+				 orig_offset + item_size - split_offset);
+	btrfs_set_item_size(leaf, orig_slot, split_offset);
 
 	btrfs_set_header_nritems(leaf, nritems + 1);
 
@@ -3603,7 +3572,6 @@ void btrfs_truncate_item(struct btrfs_path *path, u32 new_size, int from_end)
 {
 	int slot;
 	struct extent_buffer *leaf;
-	struct btrfs_item *item;
 	u32 nritems;
 	unsigned int data_end;
 	unsigned int old_data_start;
@@ -3615,14 +3583,14 @@ void btrfs_truncate_item(struct btrfs_path *path, u32 new_size, int from_end)
 	leaf = path->nodes[0];
 	slot = path->slots[0];
 
-	old_size = btrfs_item_size_nr(leaf, slot);
+	old_size = btrfs_item_size(leaf, slot);
 	if (old_size == new_size)
 		return;
 
 	nritems = btrfs_header_nritems(leaf);
 	data_end = leaf_data_end(leaf);
 
-	old_data_start = btrfs_item_offset_nr(leaf, slot);
+	old_data_start = btrfs_item_offset(leaf, slot);
 
 	size_diff = old_size - new_size;
 
@@ -3636,10 +3604,9 @@ void btrfs_truncate_item(struct btrfs_path *path, u32 new_size, int from_end)
 	btrfs_init_map_token(&token, leaf);
 	for (i = slot; i < nritems; i++) {
 		u32 ioff;
-		item = btrfs_item_nr(i);
 
-		ioff = btrfs_token_item_offset(&token, item);
-		btrfs_set_token_item_offset(&token, item, ioff + size_diff);
+		ioff = btrfs_token_item_offset(&token, i);
+		btrfs_set_token_item_offset(&token, i, ioff + size_diff);
 	}
 
 	/* shift the data */
@@ -3682,8 +3649,7 @@ void btrfs_truncate_item(struct btrfs_path *path, u32 new_size, int from_end)
 			fixup_low_keys(path, &disk_key, 1);
 	}
 
-	item = btrfs_item_nr(slot);
-	btrfs_set_item_size(leaf, item, new_size);
+	btrfs_set_item_size(leaf, slot, new_size);
 	btrfs_mark_buffer_dirty(leaf);
 
 	if (btrfs_leaf_free_space(leaf) < 0) {
@@ -3699,7 +3665,6 @@ void btrfs_extend_item(struct btrfs_path *path, u32 data_size)
 {
 	int slot;
 	struct extent_buffer *leaf;
-	struct btrfs_item *item;
 	u32 nritems;
 	unsigned int data_end;
 	unsigned int old_data;
@@ -3717,7 +3682,7 @@ void btrfs_extend_item(struct btrfs_path *path, u32 data_size)
 		BUG();
 	}
 	slot = path->slots[0];
-	old_data = btrfs_item_end_nr(leaf, slot);
+	old_data = btrfs_item_data_end(leaf, slot);
 
 	BUG_ON(slot < 0);
 	if (slot >= nritems) {
@@ -3734,10 +3699,9 @@ void btrfs_extend_item(struct btrfs_path *path, u32 data_size)
 	btrfs_init_map_token(&token, leaf);
 	for (i = slot; i < nritems; i++) {
 		u32 ioff;
-		item = btrfs_item_nr(i);
 
-		ioff = btrfs_token_item_offset(&token, item);
-		btrfs_set_token_item_offset(&token, item, ioff - data_size);
+		ioff = btrfs_token_item_offset(&token, i);
+		btrfs_set_token_item_offset(&token, i, ioff - data_size);
 	}
 
 	/* shift the data */
@@ -3746,9 +3710,8 @@ void btrfs_extend_item(struct btrfs_path *path, u32 data_size)
 		      data_end, old_data - data_end);
 
 	data_end = old_data;
-	old_size = btrfs_item_size_nr(leaf, slot);
-	item = btrfs_item_nr(slot);
-	btrfs_set_item_size(leaf, item, old_size + data_size);
+	old_size = btrfs_item_size(leaf, slot);
+	btrfs_set_item_size(leaf, slot, old_size + data_size);
 	btrfs_mark_buffer_dirty(leaf);
 
 	if (btrfs_leaf_free_space(leaf) < 0) {
@@ -3770,7 +3733,6 @@ static void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *p
 				   const struct btrfs_item_batch *batch)
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
-	struct btrfs_item *item;
 	int i;
 	u32 nritems;
 	unsigned int data_end;
@@ -3807,7 +3769,7 @@ static void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *p
 
 	btrfs_init_map_token(&token, leaf);
 	if (slot != nritems) {
-		unsigned int old_data = btrfs_item_end_nr(leaf, slot);
+		unsigned int old_data = btrfs_item_data_end(leaf, slot);
 
 		if (old_data < data_end) {
 			btrfs_print_leaf(leaf);
@@ -3823,10 +3785,9 @@ static void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *p
 		for (i = slot; i < nritems; i++) {
 			u32 ioff;
 
-			item = btrfs_item_nr(i);
-			ioff = btrfs_token_item_offset(&token, item);
-			btrfs_set_token_item_offset(&token, item,
-						    ioff - batch->total_data_size);
+			ioff = btrfs_token_item_offset(&token, i);
+			btrfs_set_token_item_offset(&token, i,
+						       ioff - batch->total_data_size);
 		}
 		/* shift the items */
 		memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot + batch->nr),
@@ -3845,10 +3806,9 @@ static void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *p
 	for (i = 0; i < batch->nr; i++) {
 		btrfs_cpu_key_to_disk(&disk_key, &batch->keys[i]);
 		btrfs_set_item_key(leaf, &disk_key, slot + i);
-		item = btrfs_item_nr(slot + i);
 		data_end -= batch->data_sizes[i];
-		btrfs_set_token_item_offset(&token, item, data_end);
-		btrfs_set_token_item_size(&token, item, batch->data_sizes[i]);
+		btrfs_set_token_item_offset(&token, slot + i, data_end);
+		btrfs_set_token_item_size(&token, slot + i, batch->data_sizes[i]);
 	}
 
 	btrfs_set_header_nritems(leaf, nritems + batch->nr);
@@ -3955,7 +3915,7 @@ int btrfs_duplicate_item(struct btrfs_trans_handle *trans,
 	u32 item_size;
 
 	leaf = path->nodes[0];
-	item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+	item_size = btrfs_item_size(leaf, path->slots[0]);
 	ret = setup_leaf_for_split(trans, root, path,
 				   item_size + sizeof(struct btrfs_item));
 	if (ret)
@@ -4056,25 +4016,22 @@ int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 {
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	struct extent_buffer *leaf;
-	struct btrfs_item *item;
-	u32 last_off;
-	u32 dsize = 0;
 	int ret = 0;
 	int wret;
-	int i;
 	u32 nritems;
 
 	leaf = path->nodes[0];
-	last_off = btrfs_item_offset_nr(leaf, slot + nr - 1);
-
-	for (i = 0; i < nr; i++)
-		dsize += btrfs_item_size_nr(leaf, slot + i);
-
 	nritems = btrfs_header_nritems(leaf);
 
 	if (slot + nr != nritems) {
-		int data_end = leaf_data_end(leaf);
+		const u32 last_off = btrfs_item_offset(leaf, slot + nr - 1);
+		const int data_end = leaf_data_end(leaf);
 		struct btrfs_map_token token;
+		u32 dsize = 0;
+		int i;
+
+		for (i = 0; i < nr; i++)
+			dsize += btrfs_item_size(leaf, slot + i);
 
 		memmove_extent_buffer(leaf, BTRFS_LEAF_DATA_OFFSET +
 			      data_end + dsize,
@@ -4085,9 +4042,8 @@ int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 		for (i = slot + nr; i < nritems; i++) {
 			u32 ioff;
 
-			item = btrfs_item_nr(i);
-			ioff = btrfs_token_item_offset(&token, item);
-			btrfs_set_token_item_offset(&token, item, ioff + dsize);
+			ioff = btrfs_token_item_offset(&token, i);
+			btrfs_set_token_item_offset(&token, i, ioff + dsize);
 		}
 
 		memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot),
@@ -4115,24 +4071,50 @@ int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 			fixup_low_keys(path, &disk_key, 1);
 		}
 
-		/* delete the leaf if it is mostly empty */
+		/*
+		 * Try to delete the leaf if it is mostly empty. We do this by
+		 * trying to move all its items into its left and right neighbours.
+		 * If we can't move all the items, then we don't delete it - it's
+		 * not ideal, but future insertions might fill the leaf with more
+		 * items, or items from other leaves might be moved later into our
+		 * leaf due to deletions on those leaves.
+		 */
 		if (used < BTRFS_LEAF_DATA_SIZE(fs_info) / 3) {
+			u32 min_push_space;
+
 			/* push_leaf_left fixes the path.
 			 * make sure the path still points to our leaf
 			 * for possible call to del_ptr below
 			 */
 			slot = path->slots[1];
 			atomic_inc(&leaf->refs);
-
-			wret = push_leaf_left(trans, root, path, 1, 1,
-					      1, (u32)-1);
+			/*
+			 * We want to be able to at least push one item to the
+			 * left neighbour leaf, and that's the first item.
+			 */
+			min_push_space = sizeof(struct btrfs_item) +
+				btrfs_item_size(leaf, 0);
+			wret = push_leaf_left(trans, root, path, 0,
+					      min_push_space, 1, (u32)-1);
 			if (wret < 0 && wret != -ENOSPC)
 				ret = wret;
 
 			if (path->nodes[0] == leaf &&
 			    btrfs_header_nritems(leaf)) {
-				wret = push_leaf_right(trans, root, path, 1,
-						       1, 1, 0);
+				/*
+				 * If we were not able to push all items from our
+				 * leaf to its left neighbour, then attempt to
+				 * either push all the remaining items to the
+				 * right neighbour or none. There's no advantage
+				 * in pushing only some items, instead of all, as
+				 * it's pointless to end up with a leaf having
+				 * too few items while the neighbours can be full
+				 * or nearly full.
+				 */
+				nritems = btrfs_header_nritems(leaf);
+				min_push_space = leaf_space_used(leaf, 0, nritems);
+				wret = push_leaf_right(trans, root, path, 0,
+						       min_push_space, 1, 0);
 				if (wret < 0 && wret != -ENOSPC)
 					ret = wret;
 			}
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 5fe5eccb3c87..223e9d9e1b8b 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -601,6 +601,9 @@ enum {
 	/* Indicate whether there are any tree modification log users */
 	BTRFS_FS_TREE_MOD_LOG_USERS,
 
+	/* Indicate that we want the transaction kthread to commit right now. */
+	BTRFS_FS_COMMIT_TRANS,
+
 #if BITS_PER_LONG == 32
 	/* Indicate if we have error/warn message printed on 32bit systems */
 	BTRFS_FS_32BIT_ERROR,
@@ -1603,25 +1606,25 @@ DECLARE_BTRFS_SETGET_BITS(64)
 static inline u##bits btrfs_##name(const struct extent_buffer *eb,	\
 				   const type *s)			\
 {									\
-	BUILD_BUG_ON(sizeof(u##bits) != sizeof(((type *)0))->member);	\
+	static_assert(sizeof(u##bits) == sizeof(((type *)0))->member);	\
 	return btrfs_get_##bits(eb, s, offsetof(type, member));		\
 }									\
 static inline void btrfs_set_##name(const struct extent_buffer *eb, type *s, \
 				    u##bits val)			\
 {									\
-	BUILD_BUG_ON(sizeof(u##bits) != sizeof(((type *)0))->member);	\
+	static_assert(sizeof(u##bits) == sizeof(((type *)0))->member);	\
 	btrfs_set_##bits(eb, s, offsetof(type, member), val);		\
 }									\
 static inline u##bits btrfs_token_##name(struct btrfs_map_token *token,	\
 					 const type *s)			\
 {									\
-	BUILD_BUG_ON(sizeof(u##bits) != sizeof(((type *)0))->member);	\
+	static_assert(sizeof(u##bits) == sizeof(((type *)0))->member);	\
 	return btrfs_get_token_##bits(token, s, offsetof(type, member));\
 }									\
 static inline void btrfs_set_token_##name(struct btrfs_map_token *token,\
 					  type *s, u##bits val)		\
 {									\
-	BUILD_BUG_ON(sizeof(u##bits) != sizeof(((type *)0))->member);	\
+	static_assert(sizeof(u##bits) == sizeof(((type *)0))->member);	\
 	btrfs_set_token_##bits(token, s, offsetof(type, member), val);	\
 }
 
@@ -1652,8 +1655,8 @@ static inline void btrfs_set_##name(type *s, u##bits val)		\
 static inline u64 btrfs_device_total_bytes(const struct extent_buffer *eb,
 					   struct btrfs_dev_item *s)
 {
-	BUILD_BUG_ON(sizeof(u64) !=
-		     sizeof(((struct btrfs_dev_item *)0))->total_bytes);
+	static_assert(sizeof(u64) ==
+		      sizeof(((struct btrfs_dev_item *)0))->total_bytes);
 	return btrfs_get_64(eb, s, offsetof(struct btrfs_dev_item,
 					    total_bytes));
 }
@@ -1661,8 +1664,8 @@ static inline void btrfs_set_device_total_bytes(const struct extent_buffer *eb,
 						struct btrfs_dev_item *s,
 						u64 val)
 {
-	BUILD_BUG_ON(sizeof(u64) !=
-		     sizeof(((struct btrfs_dev_item *)0))->total_bytes);
+	static_assert(sizeof(u64) ==
+		      sizeof(((struct btrfs_dev_item *)0))->total_bytes);
 	WARN_ON(!IS_ALIGNED(val, eb->fs_info->sectorsize));
 	btrfs_set_64(eb, s, offsetof(struct btrfs_dev_item, total_bytes), val);
 }
@@ -1960,8 +1963,8 @@ static inline void btrfs_set_node_key(const struct extent_buffer *eb,
 }
 
 /* struct btrfs_item */
-BTRFS_SETGET_FUNCS(item_offset, struct btrfs_item, offset, 32);
-BTRFS_SETGET_FUNCS(item_size, struct btrfs_item, size, 32);
+BTRFS_SETGET_FUNCS(raw_item_offset, struct btrfs_item, offset, 32);
+BTRFS_SETGET_FUNCS(raw_item_size, struct btrfs_item, size, 32);
 BTRFS_SETGET_STACK_FUNCS(stack_item_offset, struct btrfs_item, offset, 32);
 BTRFS_SETGET_STACK_FUNCS(stack_item_size, struct btrfs_item, size, 32);
 
@@ -1976,25 +1979,36 @@ static inline struct btrfs_item *btrfs_item_nr(int nr)
 	return (struct btrfs_item *)btrfs_item_nr_offset(nr);
 }
 
-static inline u32 btrfs_item_end(const struct extent_buffer *eb,
-				 struct btrfs_item *item)
-{
-	return btrfs_item_offset(eb, item) + btrfs_item_size(eb, item);
+#define BTRFS_ITEM_SETGET_FUNCS(member)						\
+static inline u32 btrfs_item_##member(const struct extent_buffer *eb,		\
+				      int slot)					\
+{										\
+	return btrfs_raw_item_##member(eb, btrfs_item_nr(slot));		\
+}										\
+static inline void btrfs_set_item_##member(const struct extent_buffer *eb,	\
+					   int slot, u32 val)			\
+{										\
+	btrfs_set_raw_item_##member(eb, btrfs_item_nr(slot), val);		\
+}										\
+static inline u32 btrfs_token_item_##member(struct btrfs_map_token *token,	\
+					    int slot)				\
+{										\
+	struct btrfs_item *item = btrfs_item_nr(slot);				\
+	return btrfs_token_raw_item_##member(token, item);			\
+}										\
+static inline void btrfs_set_token_item_##member(struct btrfs_map_token *token,	\
+						 int slot, u32 val)		\
+{										\
+	struct btrfs_item *item = btrfs_item_nr(slot);				\
+	btrfs_set_token_raw_item_##member(token, item, val);			\
 }
 
-static inline u32 btrfs_item_end_nr(const struct extent_buffer *eb, int nr)
-{
-	return btrfs_item_end(eb, btrfs_item_nr(nr));
-}
+BTRFS_ITEM_SETGET_FUNCS(offset)
+BTRFS_ITEM_SETGET_FUNCS(size);
 
-static inline u32 btrfs_item_offset_nr(const struct extent_buffer *eb, int nr)
+static inline u32 btrfs_item_data_end(const struct extent_buffer *eb, int nr)
 {
-	return btrfs_item_offset(eb, btrfs_item_nr(nr));
-}
-
-static inline u32 btrfs_item_size_nr(const struct extent_buffer *eb, int nr)
-{
-	return btrfs_item_size(eb, btrfs_item_nr(nr));
+	return btrfs_item_offset(eb, nr) + btrfs_item_size(eb, nr);
 }
 
 static inline void btrfs_item_key(const struct extent_buffer *eb,
@@ -2463,7 +2477,7 @@ static inline unsigned int leaf_data_end(const struct extent_buffer *leaf)
 
 	if (nr == 0)
 		return BTRFS_LEAF_DATA_SIZE(leaf->fs_info);
-	return btrfs_item_offset_nr(leaf, nr - 1);
+	return btrfs_item_offset(leaf, nr - 1);
 }
 
 /* struct btrfs_file_extent_item */
@@ -2522,9 +2536,9 @@ BTRFS_SETGET_FUNCS(file_extent_other_encoding, struct btrfs_file_extent_item,
  */
 static inline u32 btrfs_file_extent_inline_item_len(
 						const struct extent_buffer *eb,
-						struct btrfs_item *e)
+						int nr)
 {
-	return btrfs_item_size(eb, e) - BTRFS_FILE_EXTENT_INLINE_DATA_START;
+	return btrfs_item_size(eb, nr) - BTRFS_FILE_EXTENT_INLINE_DATA_START;
 }
 
 /* btrfs_qgroup_status_item */
@@ -2616,11 +2630,11 @@ BTRFS_SETGET_STACK_FUNCS(stack_dev_replace_cursor_right,
 /* helper function to cast into the data area of the leaf. */
 #define btrfs_item_ptr(leaf, slot, type) \
 	((type *)(BTRFS_LEAF_DATA_OFFSET + \
-	btrfs_item_offset_nr(leaf, slot)))
+	btrfs_item_offset(leaf, slot)))
 
 #define btrfs_item_ptr_offset(leaf, slot) \
 	((unsigned long)(BTRFS_LEAF_DATA_OFFSET + \
-	btrfs_item_offset_nr(leaf, slot)))
+	btrfs_item_offset(leaf, slot)))
 
 static inline u32 btrfs_crc32c(u32 crc, const void *address, unsigned length)
 {
@@ -3297,9 +3311,27 @@ int btrfs_fileattr_set(struct user_namespace *mnt_userns,
 int btrfs_ioctl_get_supported_features(void __user *arg);
 void btrfs_sync_inode_flags_to_i_flags(struct inode *inode);
 int __pure btrfs_is_empty_uuid(u8 *uuid);
+
+struct btrfs_defrag_ctrl {
+	/* Input, read-only fields */
+	u64	start;
+	u64	len;
+	u32	extent_thresh;
+	u64	newer_than;
+	u64	max_sectors_to_defrag;
+	u8	compress;
+	u8	flags;
+
+	/* Output fields */
+	u64	sectors_defragged;
+	u64	last_scanned;	/* Exclusive bytenr */
+};
+int btrfs_defrag_ioctl_args_to_ctrl(struct btrfs_fs_info *fs_info,
+				    struct btrfs_ioctl_defrag_range_args *args,
+				    struct btrfs_defrag_ctrl *ctrl,
+				    u64 max_sectors_to_defrag, u64 newer_than);
 int btrfs_defrag_file(struct inode *inode, struct file_ra_state *ra,
-		      struct btrfs_ioctl_defrag_range_args *range,
-		      u64 newer_than, unsigned long max_to_defrag);
+		      struct btrfs_defrag_ctrl *ctrl);
 void btrfs_get_block_group_info(struct list_head *groups_list,
 				struct btrfs_ioctl_space_info *space);
 void btrfs_update_ioctl_balance_args(struct btrfs_fs_info *fs_info,
diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c
index e0238dd5f2f2..66fa61cb3f23 100644
--- a/fs/btrfs/dev-replace.c
+++ b/fs/btrfs/dev-replace.c
@@ -128,7 +128,7 @@ int btrfs_init_dev_replace(struct btrfs_fs_info *fs_info)
 	}
 	slot = path->slots[0];
 	eb = path->nodes[0];
-	item_size = btrfs_item_size_nr(eb, slot);
+	item_size = btrfs_item_size(eb, slot);
 	ptr = btrfs_item_ptr(eb, slot, struct btrfs_dev_replace_item);
 
 	if (item_size != sizeof(struct btrfs_dev_replace_item)) {
@@ -381,7 +381,7 @@ int btrfs_run_dev_replace(struct btrfs_trans_handle *trans)
 	}
 
 	if (ret == 0 &&
-	    btrfs_item_size_nr(path->nodes[0], path->slots[0]) < sizeof(*ptr)) {
+	    btrfs_item_size(path->nodes[0], path->slots[0]) < sizeof(*ptr)) {
 		/*
 		 * need to delete old one and insert a new one.
 		 * Since no attempt is made to recover any old state, if the
diff --git a/fs/btrfs/dir-item.c b/fs/btrfs/dir-item.c
index 7721ce0c0604..3b532bab0755 100644
--- a/fs/btrfs/dir-item.c
+++ b/fs/btrfs/dir-item.c
@@ -27,7 +27,6 @@ static struct btrfs_dir_item *insert_with_overflow(struct btrfs_trans_handle
 	struct btrfs_fs_info *fs_info = root->fs_info;
 	int ret;
 	char *ptr;
-	struct btrfs_item *item;
 	struct extent_buffer *leaf;
 
 	ret = btrfs_insert_empty_item(trans, root, path, cpu_key, data_size);
@@ -41,10 +40,9 @@ static struct btrfs_dir_item *insert_with_overflow(struct btrfs_trans_handle
 		return ERR_PTR(ret);
 	WARN_ON(ret > 0);
 	leaf = path->nodes[0];
-	item = btrfs_item_nr(path->slots[0]);
 	ptr = btrfs_item_ptr(leaf, path->slots[0], char);
-	BUG_ON(data_size > btrfs_item_size(leaf, item));
-	ptr += btrfs_item_size(leaf, item) - data_size;
+	ASSERT(data_size <= btrfs_item_size(leaf, path->slots[0]));
+	ptr += btrfs_item_size(leaf, path->slots[0]) - data_size;
 	return (struct btrfs_dir_item *)ptr;
 }
 
@@ -271,7 +269,7 @@ int btrfs_check_dir_item_collision(struct btrfs_root *root, u64 dir,
 	data_size = sizeof(*di) + name_len;
 	leaf = path->nodes[0];
 	slot = path->slots[0];
-	if (data_size + btrfs_item_size_nr(leaf, slot) +
+	if (data_size + btrfs_item_size(leaf, slot) +
 	    sizeof(struct btrfs_item) > BTRFS_LEAF_DATA_SIZE(root->fs_info)) {
 		ret = -EOVERFLOW;
 	} else {
@@ -409,7 +407,7 @@ struct btrfs_dir_item *btrfs_match_dir_item_name(struct btrfs_fs_info *fs_info,
 	leaf = path->nodes[0];
 	dir_item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_dir_item);
 
-	total_len = btrfs_item_size_nr(leaf, path->slots[0]);
+	total_len = btrfs_item_size(leaf, path->slots[0]);
 	while (cur < total_len) {
 		this_len = sizeof(*dir_item) +
 			btrfs_dir_name_len(leaf, dir_item) +
@@ -445,7 +443,7 @@ int btrfs_delete_one_dir_name(struct btrfs_trans_handle *trans,
 	leaf = path->nodes[0];
 	sub_item_len = sizeof(*di) + btrfs_dir_name_len(leaf, di) +
 		btrfs_dir_data_len(leaf, di);
-	item_len = btrfs_item_size_nr(leaf, path->slots[0]);
+	item_len = btrfs_item_size(leaf, path->slots[0]);
 	if (sub_item_len == item_len) {
 		ret = btrfs_del_item(trans, root, path);
 	} else {
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 5f0a879c1043..e4275da0572c 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -1935,7 +1935,8 @@ static int transaction_kthread(void *arg)
 		}
 
 		delta = ktime_get_seconds() - cur->start_time;
-		if (cur->state < TRANS_STATE_COMMIT_START &&
+		if (!test_and_clear_bit(BTRFS_FS_COMMIT_TRANS, &fs_info->flags) &&
+		    cur->state < TRANS_STATE_COMMIT_START &&
 		    delta < fs_info->commit_interval) {
 			spin_unlock(&fs_info->trans_lock);
 			delay -= msecs_to_jiffies((delta - 1) * 1000);
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 7b4ee1b2d5d8..91ca32c9459a 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -171,7 +171,7 @@ int btrfs_lookup_extent_info(struct btrfs_trans_handle *trans,
 
 	if (ret == 0) {
 		leaf = path->nodes[0];
-		item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+		item_size = btrfs_item_size(leaf, path->slots[0]);
 		if (item_size >= sizeof(*ei)) {
 			ei = btrfs_item_ptr(leaf, path->slots[0],
 					    struct btrfs_extent_item);
@@ -865,7 +865,7 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
 	}
 
 	leaf = path->nodes[0];
-	item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+	item_size = btrfs_item_size(leaf, path->slots[0]);
 	if (unlikely(item_size < sizeof(*ei))) {
 		err = -EINVAL;
 		btrfs_print_v0_err(fs_info);
@@ -1007,7 +1007,7 @@ void setup_inline_extent_backref(struct btrfs_fs_info *fs_info,
 		__run_delayed_extent_op(extent_op, leaf, ei);
 
 	ptr = (unsigned long)ei + item_offset;
-	end = (unsigned long)ei + btrfs_item_size_nr(leaf, path->slots[0]);
+	end = (unsigned long)ei + btrfs_item_size(leaf, path->slots[0]);
 	if (ptr < end - size)
 		memmove_extent_buffer(leaf, ptr + size, ptr,
 				      end - size - ptr);
@@ -1119,7 +1119,7 @@ void update_inline_extent_backref(struct btrfs_path *path,
 	} else {
 		*last_ref = 1;
 		size =  btrfs_extent_inline_ref_size(type);
-		item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+		item_size = btrfs_item_size(leaf, path->slots[0]);
 		ptr = (unsigned long)iref;
 		end = (unsigned long)ei + item_size;
 		if (ptr + size < end)
@@ -1634,7 +1634,7 @@ static int run_delayed_extent_op(struct btrfs_trans_handle *trans,
 	}
 
 	leaf = path->nodes[0];
-	item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+	item_size = btrfs_item_size(leaf, path->slots[0]);
 
 	if (unlikely(item_size < sizeof(*ei))) {
 		err = -EINVAL;
@@ -2316,7 +2316,7 @@ static noinline int check_committed_ref(struct btrfs_root *root,
 		goto out;
 
 	ret = 1;
-	item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+	item_size = btrfs_item_size(leaf, path->slots[0]);
 	ei = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_extent_item);
 
 	/* If extent item has more than 1 inline ref then it's shared */
@@ -3068,7 +3068,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 	}
 
 	leaf = path->nodes[0];
-	item_size = btrfs_item_size_nr(leaf, extent_slot);
+	item_size = btrfs_item_size(leaf, extent_slot);
 	if (unlikely(item_size < sizeof(*ei))) {
 		ret = -EINVAL;
 		btrfs_print_v0_err(info);
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index d1cbb64a78f3..107d6557ebc3 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -208,7 +208,7 @@ btrfs_lookup_csum(struct btrfs_trans_handle *trans,
 
 		csum_offset = (bytenr - found_key.offset) >>
 				fs_info->sectorsize_bits;
-		csums_in_item = btrfs_item_size_nr(leaf, path->slots[0]);
+		csums_in_item = btrfs_item_size(leaf, path->slots[0]);
 		csums_in_item /= csum_size;
 
 		if (csum_offset == csums_in_item) {
@@ -274,7 +274,7 @@ static int search_csum_tree(struct btrfs_fs_info *fs_info,
 		item = btrfs_item_ptr(path->nodes[0], path->slots[0],
 				      struct btrfs_csum_item);
 		btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
-		itemsize = btrfs_item_size_nr(path->nodes[0], path->slots[0]);
+		itemsize = btrfs_item_size(path->nodes[0], path->slots[0]);
 
 		csum_start = key.offset;
 		csum_len = (itemsize / csum_size) * sectorsize;
@@ -291,7 +291,7 @@ static int search_csum_tree(struct btrfs_fs_info *fs_info,
 		goto out;
 	}
 	btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
-	itemsize = btrfs_item_size_nr(path->nodes[0], path->slots[0]);
+	itemsize = btrfs_item_size(path->nodes[0], path->slots[0]);
 
 	csum_start = key.offset;
 	csum_len = (itemsize / csum_size) * sectorsize;
@@ -534,7 +534,7 @@ int btrfs_lookup_csums_range(struct btrfs_root *root, u64 start, u64 end,
 		    key.type == BTRFS_EXTENT_CSUM_KEY) {
 			offset = (start - key.offset) >> fs_info->sectorsize_bits;
 			if (offset * csum_size <
-			    btrfs_item_size_nr(leaf, path->slots[0] - 1))
+			    btrfs_item_size(leaf, path->slots[0] - 1))
 				path->slots[0]--;
 		}
 	}
@@ -559,7 +559,7 @@ int btrfs_lookup_csums_range(struct btrfs_root *root, u64 start, u64 end,
 		if (key.offset > start)
 			start = key.offset;
 
-		size = btrfs_item_size_nr(leaf, path->slots[0]);
+		size = btrfs_item_size(leaf, path->slots[0]);
 		csum_end = key.offset + (size / csum_size) * fs_info->sectorsize;
 		if (csum_end <= start) {
 			path->slots[0]++;
@@ -750,7 +750,7 @@ static noinline void truncate_one_csum(struct btrfs_fs_info *fs_info,
 	u32 blocksize_bits = fs_info->sectorsize_bits;
 
 	leaf = path->nodes[0];
-	csum_end = btrfs_item_size_nr(leaf, path->slots[0]) / csum_size;
+	csum_end = btrfs_item_size(leaf, path->slots[0]) / csum_size;
 	csum_end <<= blocksize_bits;
 	csum_end += key->offset;
 
@@ -834,7 +834,7 @@ int btrfs_del_csums(struct btrfs_trans_handle *trans,
 		if (key.offset >= end_byte)
 			break;
 
-		csum_end = btrfs_item_size_nr(leaf, path->slots[0]) / csum_size;
+		csum_end = btrfs_item_size(leaf, path->slots[0]) / csum_size;
 		csum_end <<= blocksize_bits;
 		csum_end += key.offset;
 
@@ -1002,7 +1002,7 @@ int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans,
 		item_end = btrfs_item_ptr(leaf, path->slots[0],
 					  struct btrfs_csum_item);
 		item_end = (struct btrfs_csum_item *)((char *)item_end +
-			   btrfs_item_size_nr(leaf, path->slots[0]));
+			   btrfs_item_size(leaf, path->slots[0]));
 		goto found;
 	}
 	ret = PTR_ERR(item);
@@ -1013,7 +1013,7 @@ int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans,
 		u32 item_size;
 		/* we found one, but it isn't big enough yet */
 		leaf = path->nodes[0];
-		item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+		item_size = btrfs_item_size(leaf, path->slots[0]);
 		if ((item_size / csum_size) >=
 		    MAX_CSUM_ITEMS(fs_info, csum_size)) {
 			/* already at max size, make a new one */
@@ -1070,7 +1070,7 @@ int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans,
 	}
 
 extend_csum:
-	if (csum_offset == btrfs_item_size_nr(leaf, path->slots[0]) /
+	if (csum_offset == btrfs_item_size(leaf, path->slots[0]) /
 	    csum_size) {
 		int extend_nr;
 		u64 tmp;
@@ -1125,7 +1125,7 @@ int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans,
 		diff = min(diff,
 			   MAX_CSUM_ITEMS(fs_info, csum_size) * csum_size);
 
-		diff = diff - btrfs_item_size_nr(leaf, path->slots[0]);
+		diff = diff - btrfs_item_size(leaf, path->slots[0]);
 		diff = min_t(u32, btrfs_leaf_free_space(leaf), diff);
 		diff /= csum_size;
 		diff *= csum_size;
@@ -1162,7 +1162,7 @@ int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans,
 csum:
 	item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_csum_item);
 	item_end = (struct btrfs_csum_item *)((unsigned char *)item +
-				      btrfs_item_size_nr(leaf, path->slots[0]));
+				      btrfs_item_size(leaf, path->slots[0]));
 	item = (struct btrfs_csum_item *)((unsigned char *)item +
 					  csum_offset * csum_size);
 found:
@@ -1208,6 +1208,7 @@ void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode,
 	extent_start = key.offset;
 	extent_end = btrfs_file_extent_end(path);
 	em->ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi);
+	em->generation = btrfs_file_extent_generation(leaf, fi);
 	if (type == BTRFS_FILE_EXTENT_REG ||
 	    type == BTRFS_FILE_EXTENT_PREALLOC) {
 		em->start = extent_start;
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 11204dbbe053..12e63be6a35b 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -277,8 +277,7 @@ static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info,
 {
 	struct btrfs_root *inode_root;
 	struct inode *inode;
-	struct btrfs_ioctl_defrag_range_args range;
-	int num_defrag;
+	struct btrfs_defrag_ctrl ctrl = {0};
 	int ret;
 
 	/* get the inode */
@@ -297,21 +296,23 @@ static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info,
 
 	/* do a chunk of defrag */
 	clear_bit(BTRFS_INODE_IN_DEFRAG, &BTRFS_I(inode)->runtime_flags);
-	memset(&range, 0, sizeof(range));
-	range.len = (u64)-1;
-	range.start = defrag->last_offset;
+	ctrl.len = (u64)-1;
+	ctrl.start = defrag->last_offset;
+	ctrl.newer_than = defrag->transid;
+	ctrl.max_sectors_to_defrag = BTRFS_DEFRAG_BATCH;
 
 	sb_start_write(fs_info->sb);
-	num_defrag = btrfs_defrag_file(inode, NULL, &range, defrag->transid,
-				       BTRFS_DEFRAG_BATCH);
+	ret = btrfs_defrag_file(inode, NULL, &ctrl);
 	sb_end_write(fs_info->sb);
+	if (ret < 0)
+		goto out;
 	/*
 	 * if we filled the whole defrag batch, there
 	 * must be more work to do.  Queue this defrag
 	 * again
 	 */
-	if (num_defrag == BTRFS_DEFRAG_BATCH) {
-		defrag->last_offset = range.start;
+	if (ctrl.sectors_defragged == BTRFS_DEFRAG_BATCH) {
+		defrag->last_offset = ctrl.last_scanned;
 		btrfs_requeue_inode_defrag(BTRFS_I(inode), defrag);
 	} else if (defrag->last_offset && !defrag->cycled) {
 		/*
@@ -325,7 +326,7 @@ static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info,
 	} else {
 		kmem_cache_free(btrfs_inode_defrag_cachep, defrag);
 	}
-
+out:
 	iput(inode);
 	return 0;
 cleanup:
@@ -718,7 +719,6 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans,
 	int modify_tree = -1;
 	int update_refs;
 	int found = 0;
-	int leafs_visited = 0;
 	struct btrfs_path *path = args->path;
 
 	args->bytes_found = 0;
@@ -756,7 +756,6 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans,
 				path->slots[0]--;
 		}
 		ret = 0;
-		leafs_visited++;
 next_slot:
 		leaf = path->nodes[0];
 		if (path->slots[0] >= btrfs_header_nritems(leaf)) {
@@ -768,7 +767,6 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans,
 				ret = 0;
 				break;
 			}
-			leafs_visited++;
 			leaf = path->nodes[0];
 			recow = 1;
 		}
@@ -1014,7 +1012,7 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans,
 	 * which case it unlocked our path, so check path->locks[0] matches a
 	 * write lock.
 	 */
-	if (!ret && args->replace_extent && leafs_visited == 1 &&
+	if (!ret && args->replace_extent &&
 	    path->locks[0] == BTRFS_WRITE_LOCK &&
 	    btrfs_leaf_free_space(leaf) >=
 	    sizeof(struct btrfs_item) + args->extent_item_size) {
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index f3fee88c8ee0..a45017b12185 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -1580,6 +1580,50 @@ static int tree_insert_offset(struct rb_root *root, u64 offset,
 	return 0;
 }
 
+/*
+ * This is a little subtle.  We *only* have ->max_extent_size set if we actually
+ * searched through the bitmap and figured out the largest ->max_extent_size,
+ * otherwise it's 0.  In the case that it's 0 we don't want to tell the
+ * allocator the wrong thing, we want to use the actual real max_extent_size
+ * we've found already if it's larger, or we want to use ->bytes.
+ *
+ * This matters because find_free_space() will skip entries who's ->bytes is
+ * less than the required bytes.  So if we didn't search down this bitmap, we
+ * may pick some previous entry that has a smaller ->max_extent_size than we
+ * have.  For example, assume we have two entries, one that has
+ * ->max_extent_size set to 4K and ->bytes set to 1M.  A second entry hasn't set
+ * ->max_extent_size yet, has ->bytes set to 8K and it's contiguous.  We will
+ *  call into find_free_space(), and return with max_extent_size == 4K, because
+ *  that first bitmap entry had ->max_extent_size set, but the second one did
+ *  not.  If instead we returned 8K we'd come in searching for 8K, and find the
+ *  8K contiguous range.
+ *
+ *  Consider the other case, we have 2 8K chunks in that second entry and still
+ *  don't have ->max_extent_size set.  We'll return 16K, and the next time the
+ *  allocator comes in it'll fully search our second bitmap, and this time it'll
+ *  get an uptodate value of 8K as the maximum chunk size.  Then we'll get the
+ *  right allocation the next loop through.
+ */
+static inline u64 get_max_extent_size(const struct btrfs_free_space *entry)
+{
+	if (entry->bitmap && entry->max_extent_size)
+		return entry->max_extent_size;
+	return entry->bytes;
+}
+
+/*
+ * We want the largest entry to be leftmost, so this is inverted from what you'd
+ * normally expect.
+ */
+static bool entry_less(struct rb_node *node, const struct rb_node *parent)
+{
+	const struct btrfs_free_space *entry, *exist;
+
+	entry = rb_entry(node, struct btrfs_free_space, bytes_index);
+	exist = rb_entry(parent, struct btrfs_free_space, bytes_index);
+	return get_max_extent_size(exist) < get_max_extent_size(entry);
+}
+
 /*
  * searches the tree for the given offset.
  *
@@ -1592,15 +1636,10 @@ tree_search_offset(struct btrfs_free_space_ctl *ctl,
 		   u64 offset, int bitmap_only, int fuzzy)
 {
 	struct rb_node *n = ctl->free_space_offset.rb_node;
-	struct btrfs_free_space *entry, *prev = NULL;
+	struct btrfs_free_space *entry = NULL, *prev = NULL;
 
 	/* find entry that is closest to the 'offset' */
-	while (1) {
-		if (!n) {
-			entry = NULL;
-			break;
-		}
-
+	while (n) {
 		entry = rb_entry(n, struct btrfs_free_space, offset_index);
 		prev = entry;
 
@@ -1610,6 +1649,8 @@ tree_search_offset(struct btrfs_free_space_ctl *ctl,
 			n = n->rb_right;
 		else
 			break;
+
+		entry = NULL;
 	}
 
 	if (bitmap_only) {
@@ -1686,6 +1727,10 @@ tree_search_offset(struct btrfs_free_space_ctl *ctl,
 		return NULL;
 
 	while (1) {
+		n = rb_next(&entry->offset_index);
+		if (!n)
+			return NULL;
+		entry = rb_entry(n, struct btrfs_free_space, offset_index);
 		if (entry->bitmap) {
 			if (entry->offset + BITS_PER_BITMAP *
 			    ctl->unit > offset)
@@ -1694,11 +1739,6 @@ tree_search_offset(struct btrfs_free_space_ctl *ctl,
 			if (entry->offset + entry->bytes > offset)
 				break;
 		}
-
-		n = rb_next(&entry->offset_index);
-		if (!n)
-			return NULL;
-		entry = rb_entry(n, struct btrfs_free_space, offset_index);
 	}
 	return entry;
 }
@@ -1708,6 +1748,7 @@ __unlink_free_space(struct btrfs_free_space_ctl *ctl,
 		    struct btrfs_free_space *info)
 {
 	rb_erase(&info->offset_index, &ctl->free_space_offset);
+	rb_erase_cached(&info->bytes_index, &ctl->free_space_bytes);
 	ctl->free_extents--;
 
 	if (!info->bitmap && !btrfs_free_space_trimmed(info)) {
@@ -1734,6 +1775,8 @@ static int link_free_space(struct btrfs_free_space_ctl *ctl,
 	if (ret)
 		return ret;
 
+	rb_add_cached(&info->bytes_index, &ctl->free_space_bytes, entry_less);
+
 	if (!info->bitmap && !btrfs_free_space_trimmed(info)) {
 		ctl->discardable_extents[BTRFS_STAT_CURR]++;
 		ctl->discardable_bytes[BTRFS_STAT_CURR] += info->bytes;
@@ -1744,6 +1787,22 @@ static int link_free_space(struct btrfs_free_space_ctl *ctl,
 	return ret;
 }
 
+static void relink_bitmap_entry(struct btrfs_free_space_ctl *ctl,
+				struct btrfs_free_space *info)
+{
+	ASSERT(info->bitmap);
+
+	/*
+	 * If our entry is empty it's because we're on a cluster and we don't
+	 * want to re-link it into our ctl bytes index.
+	 */
+	if (RB_EMPTY_NODE(&info->bytes_index))
+		return;
+
+	rb_erase_cached(&info->bytes_index, &ctl->free_space_bytes);
+	rb_add_cached(&info->bytes_index, &ctl->free_space_bytes, entry_less);
+}
+
 static inline void __bitmap_clear_bits(struct btrfs_free_space_ctl *ctl,
 				       struct btrfs_free_space *info,
 				       u64 offset, u64 bytes)
@@ -1762,6 +1821,8 @@ static inline void __bitmap_clear_bits(struct btrfs_free_space_ctl *ctl,
 	if (info->max_extent_size > ctl->unit)
 		info->max_extent_size = 0;
 
+	relink_bitmap_entry(ctl, info);
+
 	if (start && test_bit(start - 1, info->bitmap))
 		extent_delta++;
 
@@ -1797,9 +1858,16 @@ static void bitmap_set_bits(struct btrfs_free_space_ctl *ctl,
 
 	bitmap_set(info->bitmap, start, count);
 
+	/*
+	 * We set some bytes, we have no idea what the max extent size is
+	 * anymore.
+	 */
+	info->max_extent_size = 0;
 	info->bytes += bytes;
 	ctl->free_space += bytes;
 
+	relink_bitmap_entry(ctl, info);
+
 	if (start && test_bit(start - 1, info->bitmap))
 		extent_delta--;
 
@@ -1867,20 +1935,14 @@ static int search_bitmap(struct btrfs_free_space_ctl *ctl,
 
 	*bytes = (u64)(max_bits) * ctl->unit;
 	bitmap_info->max_extent_size = *bytes;
+	relink_bitmap_entry(ctl, bitmap_info);
 	return -1;
 }
 
-static inline u64 get_max_extent_size(struct btrfs_free_space *entry)
-{
-	if (entry->bitmap)
-		return entry->max_extent_size;
-	return entry->bytes;
-}
-
 /* Cache the size of the max extent in bytes */
 static struct btrfs_free_space *
 find_free_space(struct btrfs_free_space_ctl *ctl, u64 *offset, u64 *bytes,
-		unsigned long align, u64 *max_extent_size)
+		unsigned long align, u64 *max_extent_size, bool use_bytes_index)
 {
 	struct btrfs_free_space *entry;
 	struct rb_node *node;
@@ -1890,16 +1952,38 @@ find_free_space(struct btrfs_free_space_ctl *ctl, u64 *offset, u64 *bytes,
 
 	if (!ctl->free_space_offset.rb_node)
 		goto out;
+again:
+	if (use_bytes_index) {
+		node = rb_first_cached(&ctl->free_space_bytes);
+	} else {
+		entry = tree_search_offset(ctl, offset_to_bitmap(ctl, *offset),
+					   0, 1);
+		if (!entry)
+			goto out;
+		node = &entry->offset_index;
+	}
 
-	entry = tree_search_offset(ctl, offset_to_bitmap(ctl, *offset), 0, 1);
-	if (!entry)
-		goto out;
+	for (; node; node = rb_next(node)) {
+		if (use_bytes_index)
+			entry = rb_entry(node, struct btrfs_free_space,
+					 bytes_index);
+		else
+			entry = rb_entry(node, struct btrfs_free_space,
+					 offset_index);
 
-	for (node = &entry->offset_index; node; node = rb_next(node)) {
-		entry = rb_entry(node, struct btrfs_free_space, offset_index);
+		/*
+		 * If we are using the bytes index then all subsequent entries
+		 * in this tree are going to be < bytes, so simply set the max
+		 * extent size and exit the loop.
+		 *
+		 * If we're using the offset index then we need to keep going
+		 * through the rest of the tree.
+		 */
 		if (entry->bytes < *bytes) {
 			*max_extent_size = max(get_max_extent_size(entry),
 					       *max_extent_size);
+			if (use_bytes_index)
+				break;
 			continue;
 		}
 
@@ -1916,6 +2000,13 @@ find_free_space(struct btrfs_free_space_ctl *ctl, u64 *offset, u64 *bytes,
 			tmp = entry->offset;
 		}
 
+		/*
+		 * We don't break here if we're using the bytes index because we
+		 * may have another entry that has the correct alignment that is
+		 * the right size, so we don't want to miss that possibility.
+		 * At worst this adds another loop through the logic, but if we
+		 * broke here we could prematurely ENOSPC.
+		 */
 		if (entry->bytes < *bytes + align_off) {
 			*max_extent_size = max(get_max_extent_size(entry),
 					       *max_extent_size);
@@ -1923,6 +2014,7 @@ find_free_space(struct btrfs_free_space_ctl *ctl, u64 *offset, u64 *bytes,
 		}
 
 		if (entry->bitmap) {
+			struct rb_node *old_next = rb_next(node);
 			u64 size = *bytes;
 
 			ret = search_bitmap(ctl, entry, &tmp, &size, true);
@@ -1935,6 +2027,15 @@ find_free_space(struct btrfs_free_space_ctl *ctl, u64 *offset, u64 *bytes,
 					max(get_max_extent_size(entry),
 					    *max_extent_size);
 			}
+
+			/*
+			 * The bitmap may have gotten re-arranged in the space
+			 * index here because the max_extent_size may have been
+			 * updated.  Start from the beginning again if this
+			 * happened.
+			 */
+			if (use_bytes_index && old_next != rb_next(node))
+				goto again;
 			continue;
 		}
 
@@ -2083,12 +2184,6 @@ static u64 add_bytes_to_bitmap(struct btrfs_free_space_ctl *ctl,
 
 	bitmap_set_bits(ctl, info, offset, bytes_to_set);
 
-	/*
-	 * We set some bytes, we have no idea what the max extent size is
-	 * anymore.
-	 */
-	info->max_extent_size = 0;
-
 	return bytes_to_set;
 
 }
@@ -2486,6 +2581,7 @@ int __btrfs_add_free_space(struct btrfs_fs_info *fs_info,
 	info->bytes = bytes;
 	info->trim_state = trim_state;
 	RB_CLEAR_NODE(&info->offset_index);
+	RB_CLEAR_NODE(&info->bytes_index);
 
 	spin_lock(&ctl->tree_lock);
 
@@ -2799,6 +2895,7 @@ void btrfs_init_free_space_ctl(struct btrfs_block_group *block_group,
 	ctl->start = block_group->start;
 	ctl->private = block_group;
 	ctl->op = &free_space_op;
+	ctl->free_space_bytes = RB_ROOT_CACHED;
 	INIT_LIST_HEAD(&ctl->trimming_ranges);
 	mutex_init(&ctl->cache_writeout_mutex);
 
@@ -2864,6 +2961,8 @@ static void __btrfs_return_cluster_to_free_space(
 		}
 		tree_insert_offset(&ctl->free_space_offset,
 				   entry->offset, &entry->offset_index, bitmap);
+		rb_add_cached(&entry->bytes_index, &ctl->free_space_bytes,
+			      entry_less);
 	}
 	cluster->root = RB_ROOT;
 	spin_unlock(&cluster->lock);
@@ -2965,12 +3064,14 @@ u64 btrfs_find_space_for_alloc(struct btrfs_block_group *block_group,
 	u64 align_gap = 0;
 	u64 align_gap_len = 0;
 	enum btrfs_trim_state align_gap_trim_state = BTRFS_TRIM_STATE_UNTRIMMED;
+	bool use_bytes_index = (offset == block_group->start);
 
 	ASSERT(!btrfs_is_zoned(block_group->fs_info));
 
 	spin_lock(&ctl->tree_lock);
 	entry = find_free_space(ctl, &offset, &bytes_search,
-				block_group->full_stripe_len, max_extent_size);
+				block_group->full_stripe_len, max_extent_size,
+				use_bytes_index);
 	if (!entry)
 		goto out;
 
@@ -3254,6 +3355,17 @@ static int btrfs_bitmap_cluster(struct btrfs_block_group *block_group,
 
 	cluster->window_start = start * ctl->unit + entry->offset;
 	rb_erase(&entry->offset_index, &ctl->free_space_offset);
+	rb_erase_cached(&entry->bytes_index, &ctl->free_space_bytes);
+
+	/*
+	 * We need to know if we're currently on the normal space index when we
+	 * manipulate the bitmap so that we know we need to remove and re-insert
+	 * it into the space_index tree.  Clear the bytes_index node here so the
+	 * bitmap manipulation helpers know not to mess with the space_index
+	 * until this bitmap entry is added back into the normal cache.
+	 */
+	RB_CLEAR_NODE(&entry->bytes_index);
+
 	ret = tree_insert_offset(&cluster->root, entry->offset,
 				 &entry->offset_index, 1);
 	ASSERT(!ret); /* -EEXIST; Logic error */
@@ -3344,6 +3456,7 @@ setup_cluster_no_bitmap(struct btrfs_block_group *block_group,
 			continue;
 
 		rb_erase(&entry->offset_index, &ctl->free_space_offset);
+		rb_erase_cached(&entry->bytes_index, &ctl->free_space_bytes);
 		ret = tree_insert_offset(&cluster->root, entry->offset,
 					 &entry->offset_index, 0);
 		total_size += entry->bytes;
diff --git a/fs/btrfs/free-space-cache.h b/fs/btrfs/free-space-cache.h
index 1f23088d43f9..dd982d204d2d 100644
--- a/fs/btrfs/free-space-cache.h
+++ b/fs/btrfs/free-space-cache.h
@@ -22,6 +22,7 @@ enum btrfs_trim_state {
 
 struct btrfs_free_space {
 	struct rb_node offset_index;
+	struct rb_node bytes_index;
 	u64 offset;
 	u64 bytes;
 	u64 max_extent_size;
@@ -45,6 +46,7 @@ static inline bool btrfs_free_space_trimming_bitmap(
 struct btrfs_free_space_ctl {
 	spinlock_t tree_lock;
 	struct rb_root free_space_offset;
+	struct rb_root_cached free_space_bytes;
 	u64 free_space;
 	int extents_thresh;
 	int free_extents;
diff --git a/fs/btrfs/inode-item.c b/fs/btrfs/inode-item.c
index 37f36ffdaf6b..56755ce9a907 100644
--- a/fs/btrfs/inode-item.c
+++ b/fs/btrfs/inode-item.c
@@ -19,7 +19,7 @@ struct btrfs_inode_ref *btrfs_find_name_in_backref(struct extent_buffer *leaf,
 	u32 cur_offset = 0;
 	int len;
 
-	item_size = btrfs_item_size_nr(leaf, slot);
+	item_size = btrfs_item_size(leaf, slot);
 	ptr = btrfs_item_ptr_offset(leaf, slot);
 	while (cur_offset < item_size) {
 		ref = (struct btrfs_inode_ref *)(ptr + cur_offset);
@@ -45,7 +45,7 @@ struct btrfs_inode_extref *btrfs_find_name_in_ext_backref(
 	u32 cur_offset = 0;
 	int ref_name_len;
 
-	item_size = btrfs_item_size_nr(leaf, slot);
+	item_size = btrfs_item_size(leaf, slot);
 	ptr = btrfs_item_ptr_offset(leaf, slot);
 
 	/*
@@ -139,7 +139,7 @@ static int btrfs_del_inode_extref(struct btrfs_trans_handle *trans,
 	}
 
 	leaf = path->nodes[0];
-	item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+	item_size = btrfs_item_size(leaf, path->slots[0]);
 	if (index)
 		*index = btrfs_inode_extref_index(leaf, extref);
 
@@ -208,7 +208,7 @@ int btrfs_del_inode_ref(struct btrfs_trans_handle *trans,
 		goto out;
 	}
 	leaf = path->nodes[0];
-	item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+	item_size = btrfs_item_size(leaf, path->slots[0]);
 
 	if (index)
 		*index = btrfs_inode_ref_index(leaf, ref);
@@ -256,7 +256,6 @@ static int btrfs_insert_inode_extref(struct btrfs_trans_handle *trans,
 	struct btrfs_path *path;
 	struct btrfs_key key;
 	struct extent_buffer *leaf;
-	struct btrfs_item *item;
 
 	key.objectid = inode_objectid;
 	key.type = BTRFS_INODE_EXTREF_KEY;
@@ -282,9 +281,8 @@ static int btrfs_insert_inode_extref(struct btrfs_trans_handle *trans,
 		goto out;
 
 	leaf = path->nodes[0];
-	item = btrfs_item_nr(path->slots[0]);
 	ptr = (unsigned long)btrfs_item_ptr(leaf, path->slots[0], char);
-	ptr += btrfs_item_size(leaf, item) - ins_len;
+	ptr += btrfs_item_size(leaf, path->slots[0]) - ins_len;
 	extref = (struct btrfs_inode_extref *)ptr;
 
 	btrfs_set_inode_extref_name_len(path->nodes[0], extref, name_len);
@@ -332,7 +330,7 @@ int btrfs_insert_inode_ref(struct btrfs_trans_handle *trans,
 		if (ref)
 			goto out;
 
-		old_size = btrfs_item_size_nr(path->nodes[0], path->slots[0]);
+		old_size = btrfs_item_size(path->nodes[0], path->slots[0]);
 		btrfs_extend_item(path, ins_len);
 		ref = btrfs_item_ptr(path->nodes[0], path->slots[0],
 				     struct btrfs_inode_ref);
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 39a674543461..0ed8cc6afa37 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -61,8 +61,6 @@ struct btrfs_iget_args {
 };
 
 struct btrfs_dio_data {
-	u64 reserve;
-	loff_t length;
 	ssize_t submitted;
 	struct extent_changeset *data_reserved;
 };
@@ -625,7 +623,6 @@ static noinline int compress_file_range(struct async_chunk *async_chunk)
 again:
 	will_compress = 0;
 	nr_pages = (end >> PAGE_SHIFT) - (start >> PAGE_SHIFT) + 1;
-	BUILD_BUG_ON((BTRFS_MAX_COMPRESSED % PAGE_SIZE) != 0);
 	nr_pages = min_t(unsigned long, nr_pages,
 			BTRFS_MAX_COMPRESSED / PAGE_SIZE);
 
@@ -5950,21 +5947,17 @@ static struct inode *new_simple_dir(struct super_block *s,
 	return inode;
 }
 
+static_assert(BTRFS_FT_UNKNOWN == FT_UNKNOWN);
+static_assert(BTRFS_FT_REG_FILE == FT_REG_FILE);
+static_assert(BTRFS_FT_DIR == FT_DIR);
+static_assert(BTRFS_FT_CHRDEV == FT_CHRDEV);
+static_assert(BTRFS_FT_BLKDEV == FT_BLKDEV);
+static_assert(BTRFS_FT_FIFO == FT_FIFO);
+static_assert(BTRFS_FT_SOCK == FT_SOCK);
+static_assert(BTRFS_FT_SYMLINK == FT_SYMLINK);
+
 static inline u8 btrfs_inode_type(struct inode *inode)
 {
-	/*
-	 * Compile-time asserts that generic FT_* types still match
-	 * BTRFS_FT_* types
-	 */
-	BUILD_BUG_ON(BTRFS_FT_UNKNOWN != FT_UNKNOWN);
-	BUILD_BUG_ON(BTRFS_FT_REG_FILE != FT_REG_FILE);
-	BUILD_BUG_ON(BTRFS_FT_DIR != FT_DIR);
-	BUILD_BUG_ON(BTRFS_FT_CHRDEV != FT_CHRDEV);
-	BUILD_BUG_ON(BTRFS_FT_BLKDEV != FT_BLKDEV);
-	BUILD_BUG_ON(BTRFS_FT_FIFO != FT_FIFO);
-	BUILD_BUG_ON(BTRFS_FT_SOCK != FT_SOCK);
-	BUILD_BUG_ON(BTRFS_FT_SYMLINK != FT_SYMLINK);
-
 	return fs_umode_to_ftype(inode->i_mode);
 }
 
@@ -6998,8 +6991,7 @@ static noinline int uncompress_inline(struct btrfs_path *path,
 	WARN_ON(pg_offset != 0);
 	compress_type = btrfs_file_extent_compression(leaf, item);
 	max_size = btrfs_file_extent_ram_bytes(leaf, item);
-	inline_size = btrfs_file_extent_inline_item_len(leaf,
-					btrfs_item_nr(path->slots[0]));
+	inline_size = btrfs_file_extent_inline_item_len(leaf, path->slots[0]);
 	tmp = kmalloc(inline_size, GFP_NOFS);
 	if (!tmp)
 		return -ENOMEM;
@@ -7773,6 +7765,10 @@ static int btrfs_get_blocks_direct_write(struct extent_map **map,
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct extent_map *em = *map;
+	int type;
+	u64 block_start, orig_start, orig_block_len, ram_bytes;
+	bool can_nocow = false;
+	bool space_reserved = false;
 	int ret = 0;
 
 	/*
@@ -7787,9 +7783,6 @@ static int btrfs_get_blocks_direct_write(struct extent_map **map,
 	if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags) ||
 	    ((BTRFS_I(inode)->flags & BTRFS_INODE_NODATACOW) &&
 	     em->block_start != EXTENT_MAP_HOLE)) {
-		int type;
-		u64 block_start, orig_start, orig_block_len, ram_bytes;
-
 		if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
 			type = BTRFS_ORDERED_PREALLOC;
 		else
@@ -7799,53 +7792,92 @@ static int btrfs_get_blocks_direct_write(struct extent_map **map,
 
 		if (can_nocow_extent(inode, start, &len, &orig_start,
 				     &orig_block_len, &ram_bytes, false) == 1 &&
-		    btrfs_inc_nocow_writers(fs_info, block_start)) {
-			struct extent_map *em2;
+		    btrfs_inc_nocow_writers(fs_info, block_start))
+			can_nocow = true;
+	}
 
-			em2 = btrfs_create_dio_extent(BTRFS_I(inode), start, len,
-						      orig_start, block_start,
-						      len, orig_block_len,
-						      ram_bytes, type);
+	if (can_nocow) {
+		struct extent_map *em2;
+
+		/* We can NOCOW, so only need to reserve metadata space. */
+		ret = btrfs_delalloc_reserve_metadata(BTRFS_I(inode), len);
+		if (ret < 0) {
+			/* Our caller expects us to free the input extent map. */
+			free_extent_map(em);
+			*map = NULL;
 			btrfs_dec_nocow_writers(fs_info, block_start);
-			if (type == BTRFS_ORDERED_PREALLOC) {
-				free_extent_map(em);
-				*map = em = em2;
-			}
-
-			if (em2 && IS_ERR(em2)) {
-				ret = PTR_ERR(em2);
-				goto out;
-			}
-			/*
-			 * For inode marked NODATACOW or extent marked PREALLOC,
-			 * use the existing or preallocated extent, so does not
-			 * need to adjust btrfs_space_info's bytes_may_use.
-			 */
-			btrfs_free_reserved_data_space_noquota(fs_info, len);
-			goto skip_cow;
+			goto out;
 		}
+		space_reserved = true;
+
+		em2 = btrfs_create_dio_extent(BTRFS_I(inode), start, len,
+					      orig_start, block_start,
+					      len, orig_block_len,
+					      ram_bytes, type);
+		btrfs_dec_nocow_writers(fs_info, block_start);
+		if (type == BTRFS_ORDERED_PREALLOC) {
+			free_extent_map(em);
+			*map = em = em2;
+		}
+
+		if (IS_ERR(em2)) {
+			ret = PTR_ERR(em2);
+			goto out;
+		}
+	} else {
+		const u64 prev_len = len;
+
+		/* Our caller expects us to free the input extent map. */
+		free_extent_map(em);
+		*map = NULL;
+
+		/* We have to COW, so need to reserve metadata and data space. */
+		ret = btrfs_delalloc_reserve_space(BTRFS_I(inode),
+						   &dio_data->data_reserved,
+						   start, len);
+		if (ret < 0)
+			goto out;
+		space_reserved = true;
+
+		em = btrfs_new_extent_direct(BTRFS_I(inode), start, len);
+		if (IS_ERR(em)) {
+			ret = PTR_ERR(em);
+			goto out;
+		}
+		*map = em;
+		len = min(len, em->len - (start - em->start));
+		if (len < prev_len)
+			btrfs_delalloc_release_space(BTRFS_I(inode),
+						     dio_data->data_reserved,
+						     start + len, prev_len - len,
+						     true);
 	}
 
-	/* this will cow the extent */
-	free_extent_map(em);
-	*map = em = btrfs_new_extent_direct(BTRFS_I(inode), start, len);
-	if (IS_ERR(em)) {
-		ret = PTR_ERR(em);
-		goto out;
-	}
+	/*
+	 * We have created our ordered extent, so we can now release our reservation
+	 * for an outstanding extent.
+	 */
+	btrfs_delalloc_release_extents(BTRFS_I(inode), len);
 
-	len = min(len, em->len - (start - em->start));
-
-skip_cow:
 	/*
 	 * Need to update the i_size under the extent lock so buffered
 	 * readers will get the updated i_size when we unlock.
 	 */
 	if (start + len > i_size_read(inode))
 		i_size_write(inode, start + len);
-
-	dio_data->reserve -= len;
 out:
+	if (ret && space_reserved) {
+		btrfs_delalloc_release_extents(BTRFS_I(inode), len);
+		if (can_nocow) {
+			btrfs_delalloc_release_metadata(BTRFS_I(inode), len, true);
+		} else {
+			btrfs_delalloc_release_space(BTRFS_I(inode),
+						     dio_data->data_reserved,
+						     start, len, true);
+			extent_changeset_free(dio_data->data_reserved);
+			dio_data->data_reserved = NULL;
+		}
+	}
 	return ret;
 }
 
@@ -7887,18 +7919,6 @@ static int btrfs_dio_iomap_begin(struct inode *inode, loff_t start,
 	if (!dio_data)
 		return -ENOMEM;
 
-	dio_data->length = length;
-	if (write) {
-		dio_data->reserve = round_up(length, fs_info->sectorsize);
-		ret = btrfs_delalloc_reserve_space(BTRFS_I(inode),
-				&dio_data->data_reserved,
-				start, dio_data->reserve);
-		if (ret) {
-			extent_changeset_free(dio_data->data_reserved);
-			kfree(dio_data);
-			return ret;
-		}
-	}
 	iomap->private = dio_data;
 
 
@@ -7991,14 +8011,8 @@ static int btrfs_dio_iomap_begin(struct inode *inode, loff_t start,
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, lockend,
 			     &cached_state);
 err:
-	if (dio_data) {
-		btrfs_delalloc_release_space(BTRFS_I(inode),
-				dio_data->data_reserved, start,
-				dio_data->reserve, true);
-		btrfs_delalloc_release_extents(BTRFS_I(inode), dio_data->reserve);
-		extent_changeset_free(dio_data->data_reserved);
-		kfree(dio_data);
-	}
+	kfree(dio_data);
+
 	return ret;
 }
 
@@ -8028,14 +8042,8 @@ static int btrfs_dio_iomap_end(struct inode *inode, loff_t pos, loff_t length,
 		ret = -ENOTBLK;
 	}
 
-	if (write) {
-		if (dio_data->reserve)
-			btrfs_delalloc_release_space(BTRFS_I(inode),
-					dio_data->data_reserved, pos,
-					dio_data->reserve, true);
-		btrfs_delalloc_release_extents(BTRFS_I(inode), dio_data->length);
+	if (write)
 		extent_changeset_free(dio_data->data_reserved);
-	}
 out:
 	kfree(dio_data);
 	iomap->private = NULL;
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 48e03e176f31..5de240144273 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -1020,23 +1020,37 @@ static struct extent_map *defrag_lookup_extent(struct inode *inode, u64 start,
 	return em;
 }
 
+static u32 get_extent_max_capacity(const struct extent_map *em)
+{
+	if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags))
+		return BTRFS_MAX_COMPRESSED;
+	return BTRFS_MAX_EXTENT_SIZE;
+}
+
 static bool defrag_check_next_extent(struct inode *inode, struct extent_map *em,
 				     bool locked)
 {
 	struct extent_map *next;
-	bool ret = true;
+	bool ret = false;
 
 	/* this is the last extent */
 	if (em->start + em->len >= i_size_read(inode))
-		return false;
+		return ret;
 
 	next = defrag_lookup_extent(inode, em->start + em->len, locked);
+	/* No more em or hole */
 	if (!next || next->block_start >= EXTENT_MAP_LAST_BYTE)
-		ret = false;
-	else if ((em->block_start + em->block_len == next->block_start) &&
-		 (em->block_len > SZ_128K && next->block_len > SZ_128K))
-		ret = false;
-
+		goto out;
+	if (test_bit(EXTENT_FLAG_PREALLOC, &next->flags))
+		goto out;
+	/*
+	 * If the next extent is at its max capcity, defragging current extent
+	 * makes no sense, as the total number of extents won't change.
+	 */
+	if (next->len >= get_extent_max_capacity(em))
+		goto out;
+	ret = true;
+out:
 	free_extent_map(next);
 	return ret;
 }
@@ -1146,22 +1160,21 @@ struct defrag_target_range {
 /*
  * Collect all valid target extents.
  *
+ * @ctrl:	   extra defrag policy control
  * @start:	   file offset to lookup
  * @len:	   length to lookup
- * @extent_thresh: file extent size threshold, any extent size >= this value
- *		   will be ignored
- * @newer_than:    only defrag extents newer than this value
- * @do_compress:   whether the defrag is doing compression
- *		   if true, @extent_thresh will be ignored and all regular
- *		   file extents meeting @newer_than will be targets.
  * @locked:	   if the range has already held extent lock
  * @target_list:   list of targets file extents
+ *
+ * Will update ctrl::last_scanned.
  */
 static int defrag_collect_targets(struct btrfs_inode *inode,
-				  u64 start, u64 len, u32 extent_thresh,
-				  u64 newer_than, bool do_compress,
-				  bool locked, struct list_head *target_list)
+				  struct btrfs_defrag_ctrl *ctrl,
+				  u64 start, u32 len, bool locked,
+				  struct list_head *target_list)
 {
+	bool do_compress = ctrl->flags & BTRFS_DEFRAG_RANGE_COMPRESS;
+	bool last_is_target = false;
 	u64 cur = start;
 	int ret = 0;
 
@@ -1171,6 +1184,7 @@ static int defrag_collect_targets(struct btrfs_inode *inode,
 		bool next_mergeable = true;
 		u64 range_len;
 
+		last_is_target = false;
 		em = defrag_lookup_extent(&inode->vfs_inode, cur, locked);
 		if (!em)
 			break;
@@ -1181,7 +1195,11 @@ static int defrag_collect_targets(struct btrfs_inode *inode,
 			goto next;
 
 		/* Skip older extent */
-		if (em->generation < newer_than)
+		if (em->generation < ctrl->newer_than)
+			goto next;
+
+		/* This em is under writeback, no need to defrag */
+		if (em->generation == (u64)-1)
 			goto next;
 
 		/*
@@ -1221,7 +1239,14 @@ static int defrag_collect_targets(struct btrfs_inode *inode,
 			goto add;
 
 		/* Skip too large extent */
-		if (range_len >= extent_thresh)
+		if (range_len >= ctrl->extent_thresh)
+			goto next;
+
+		/*
+		 * Skip extents already at its max capacity, this is mostly for
+		 * compressed extents, which max cap is only 128K.
+		 */
+		if (em->len >= get_extent_max_capacity(em))
 			goto next;
 
 		next_mergeable = defrag_check_next_extent(&inode->vfs_inode, em,
@@ -1242,6 +1267,7 @@ static int defrag_collect_targets(struct btrfs_inode *inode,
 		}
 
 add:
+		last_is_target = true;
 		range_len = min(extent_map_end(em), start + len) - cur;
 		/*
 		 * This one is a good target, check if it can be merged into
@@ -1285,10 +1311,27 @@ static int defrag_collect_targets(struct btrfs_inode *inode,
 			kfree(entry);
 		}
 	}
+	if (!ret) {
+		/*
+		 * If the last extent is not a target, the caller can skip to
+		 * the end of that extent.
+		 * Otherwise, we can only go the end of the spcified range.
+		 *
+		 * And we may got a range smaller than current
+		 * ctrl->last_scanned (e.g. executed in the defrag_one_range
+		 * call), so we have to ensure we didn't decrease
+		 * ctrl->last_scanned.
+		 */
+		if (!last_is_target)
+			ctrl->last_scanned = max(cur, ctrl->last_scanned);
+		else
+			ctrl->last_scanned = max(start + len, ctrl->last_scanned);
+	}
 	return ret;
 }
 
 #define CLUSTER_SIZE	(SZ_256K)
+static_assert(IS_ALIGNED(CLUSTER_SIZE, PAGE_SIZE));
 
 /*
  * Defrag one contiguous target range.
@@ -1342,8 +1385,8 @@ static int defrag_one_locked_target(struct btrfs_inode *inode,
 	return ret;
 }
 
-static int defrag_one_range(struct btrfs_inode *inode, u64 start, u32 len,
-			    u32 extent_thresh, u64 newer_than, bool do_compress)
+static int defrag_one_range(struct btrfs_inode *inode,
+			    struct btrfs_defrag_ctrl *ctrl, u64 start, u32 len)
 {
 	struct extent_state *cached_state = NULL;
 	struct defrag_target_range *entry;
@@ -1387,8 +1430,7 @@ static int defrag_one_range(struct btrfs_inode *inode, u64 start, u32 len,
 	 * And this time we have extent locked already, pass @locked = true
 	 * so that we won't relock the extent range and cause deadlock.
 	 */
-	ret = defrag_collect_targets(inode, start, len, extent_thresh,
-				     newer_than, do_compress, true,
+	ret = defrag_collect_targets(inode, ctrl, start, len, true,
 				     &target_list);
 	if (ret < 0)
 		goto unlock_extent;
@@ -1398,6 +1440,8 @@ static int defrag_one_range(struct btrfs_inode *inode, u64 start, u32 len,
 					       &cached_state);
 		if (ret < 0)
 			break;
+		ctrl->sectors_defragged += entry->len >>
+					   inode->root->fs_info->sectorsize_bits;
 	}
 
 	list_for_each_entry_safe(entry, tmp, &target_list, list) {
@@ -1419,12 +1463,17 @@ static int defrag_one_range(struct btrfs_inode *inode, u64 start, u32 len,
 	return ret;
 }
 
+/*
+ * Return <0 for error.
+ * Return >0 if we hit the ctrl->max_sectors_to_defrag limit
+ * Return 0 if we finished the range without error.
+ *
+ * For >= 0 case, ctrl->last_scanned and ctrl->sectors_defragged will be updated.
+ */
 static int defrag_one_cluster(struct btrfs_inode *inode,
 			      struct file_ra_state *ra,
-			      u64 start, u32 len, u32 extent_thresh,
-			      u64 newer_than, bool do_compress,
-			      unsigned long *sectors_defragged,
-			      unsigned long max_sectors)
+			      struct btrfs_defrag_ctrl *ctrl,
+			      u64 start, u32 len)
 {
 	const u32 sectorsize = inode->root->fs_info->sectorsize;
 	struct defrag_target_range *entry;
@@ -1432,9 +1481,7 @@ static int defrag_one_cluster(struct btrfs_inode *inode,
 	LIST_HEAD(target_list);
 	int ret;
 
-	BUILD_BUG_ON(!IS_ALIGNED(CLUSTER_SIZE, PAGE_SIZE));
-	ret = defrag_collect_targets(inode, start, len, extent_thresh,
-				     newer_than, do_compress, false,
+	ret = defrag_collect_targets(inode, ctrl, start, len, false,
 				     &target_list);
 	if (ret < 0)
 		goto out;
@@ -1443,32 +1490,25 @@ static int defrag_one_cluster(struct btrfs_inode *inode,
 		u32 range_len = entry->len;
 
 		/* Reached or beyond the limit */
-		if (max_sectors && *sectors_defragged >= max_sectors) {
+		if (ctrl->max_sectors_to_defrag &&
+		    ctrl->sectors_defragged >= ctrl->max_sectors_to_defrag) {
 			ret = 1;
 			break;
 		}
 
-		if (max_sectors)
+		if (ctrl->max_sectors_to_defrag)
 			range_len = min_t(u32, range_len,
-				(max_sectors - *sectors_defragged) * sectorsize);
+				(ctrl->max_sectors_to_defrag -
+				 ctrl->sectors_defragged) * sectorsize);
 
 		if (ra)
 			page_cache_sync_readahead(inode->vfs_inode.i_mapping,
 				ra, NULL, entry->start >> PAGE_SHIFT,
 				((entry->start + range_len - 1) >> PAGE_SHIFT) -
 				(entry->start >> PAGE_SHIFT) + 1);
-		/*
-		 * Here we may not defrag any range if holes are punched before
-		 * we locked the pages.
-		 * But that's fine, it only affects the @sectors_defragged
-		 * accounting.
-		 */
-		ret = defrag_one_range(inode, entry->start, range_len,
-				       extent_thresh, newer_than, do_compress);
+		ret = defrag_one_range(inode, ctrl, entry->start, range_len);
 		if (ret < 0)
 			break;
-		*sectors_defragged += range_len >>
-				      inode->root->fs_info->sectorsize_bits;
 	}
 out:
 	list_for_each_entry_safe(entry, tmp, &target_list, list) {
@@ -1478,64 +1518,93 @@ static int defrag_one_cluster(struct btrfs_inode *inode,
 	return ret;
 }
 
+/*
+ * Convert the old ioctl format to the new btrfs_defrag_ctrl structure.
+ *
+ * Will also do basic tasks like setting default values and sanity checks.
+ */
+int btrfs_defrag_ioctl_args_to_ctrl(struct btrfs_fs_info *fs_info,
+				    struct btrfs_ioctl_defrag_range_args *args,
+				    struct btrfs_defrag_ctrl *ctrl,
+				    u64 max_sectors_to_defrag, u64 newer_than)
+{
+	u64 range_end;
+
+	if (args->flags & ~BTRFS_DEFRAG_RANGE_FLAGS_MASK)
+		return -EOPNOTSUPP;
+	if (args->compress_type >= BTRFS_NR_COMPRESS_TYPES)
+		return -EOPNOTSUPP;
+
+	ctrl->start = round_down(args->start, fs_info->sectorsize);
+	/*
+	 * If @len does not overflow with @start nor is -1, align the length.
+	 * Otherwise set it to (u64)-1 so later btrfs_defrag_file() will
+	 * determine the length using isize.
+	 */
+	if (!check_add_overflow(args->start, args->len, &range_end) &&
+	    args->len != (u64)-1)
+		ctrl->len = round_up(range_end, fs_info->sectorsize) -
+			    ctrl->start;
+	else
+		ctrl->len = -1;
+	ctrl->flags = args->flags;
+	ctrl->compress = args->compress_type;
+	if (args->extent_thresh == 0)
+		ctrl->extent_thresh = SZ_256K;
+	else
+		ctrl->extent_thresh = args->extent_thresh;
+	ctrl->newer_than = newer_than;
+	ctrl->last_scanned = 0;
+	ctrl->sectors_defragged = 0;
+	return 0;
+}
+
 /*
  * Entry point to file defragmentation.
  *
  * @inode:	   inode to be defragged
  * @ra:		   readahead state (can be NUL)
- * @range:	   defrag options including range and flags
- * @newer_than:	   minimum transid to defrag
- * @max_to_defrag: max number of sectors to be defragged, if 0, the whole inode
- *		   will be defragged.
+ * @ctrl:	   defrag options including range and various policy parameters
  *
  * Return <0 for error.
- * Return >=0 for the number of sectors defragged, and range->start will be updated
- * to indicate the file offset where next defrag should be started at.
- * (Mostly for autodefrag, which sets @max_to_defrag thus we may exit early without
- *  defragging all the range).
+ * Return 0 if the defrag is done without error, ctrl->last_scanned and
+ * ctrl->sectors_defragged will be updated.
  */
 int btrfs_defrag_file(struct inode *inode, struct file_ra_state *ra,
-		      struct btrfs_ioctl_defrag_range_args *range,
-		      u64 newer_than, unsigned long max_to_defrag)
+		      struct btrfs_defrag_ctrl *ctrl)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	unsigned long sectors_defragged = 0;
 	u64 isize = i_size_read(inode);
 	u64 cur;
 	u64 last_byte;
-	bool do_compress = range->flags & BTRFS_DEFRAG_RANGE_COMPRESS;
+	bool do_compress = ctrl->flags & BTRFS_DEFRAG_RANGE_COMPRESS;
 	bool ra_allocated = false;
-	int compress_type = BTRFS_COMPRESS_ZLIB;
 	int ret = 0;
-	u32 extent_thresh = range->extent_thresh;
 	pgoff_t start_index;
 
 	if (isize == 0)
 		return 0;
 
-	if (range->start >= isize)
+	if (ctrl->start >= isize)
 		return -EINVAL;
 
-	if (do_compress) {
-		if (range->compress_type >= BTRFS_NR_COMPRESS_TYPES)
-			return -EINVAL;
-		if (range->compress_type)
-			compress_type = range->compress_type;
-	}
+	if (do_compress)
+		ASSERT(ctrl->compress < BTRFS_NR_COMPRESS_TYPES);
 
-	if (extent_thresh == 0)
-		extent_thresh = SZ_256K;
+	if (ctrl->extent_thresh == 0)
+		ctrl->extent_thresh = SZ_256K;
 
-	if (range->start + range->len > range->start) {
+	if (ctrl->start + ctrl->len > ctrl->start) {
 		/* Got a specific range */
-		last_byte = min(isize, range->start + range->len);
+		last_byte = min(isize, ctrl->start + ctrl->len);
 	} else {
 		/* Defrag until file end */
 		last_byte = isize;
 	}
 
 	/* Align the range */
-	cur = round_down(range->start, fs_info->sectorsize);
+	cur = round_down(ctrl->start, fs_info->sectorsize);
+	ctrl->last_scanned = cur;
 	last_byte = round_up(last_byte, fs_info->sectorsize) - 1;
 
 	/*
@@ -1559,12 +1628,9 @@ int btrfs_defrag_file(struct inode *inode, struct file_ra_state *ra,
 		inode->i_mapping->writeback_index = start_index;
 
 	while (cur < last_byte) {
-		const unsigned long prev_sectors_defragged = sectors_defragged;
+		const unsigned long prev_sectors_defragged = ctrl->sectors_defragged;
 		u64 cluster_end;
 
-		/* The cluster size 256K should always be page aligned */
-		BUILD_BUG_ON(!IS_ALIGNED(CLUSTER_SIZE, PAGE_SIZE));
-
 		if (btrfs_defrag_cancelled(fs_info)) {
 			ret = -EAGAIN;
 			break;
@@ -1586,48 +1652,41 @@ int btrfs_defrag_file(struct inode *inode, struct file_ra_state *ra,
 			break;
 		}
 		if (do_compress)
-			BTRFS_I(inode)->defrag_compress = compress_type;
-		ret = defrag_one_cluster(BTRFS_I(inode), ra, cur,
-				cluster_end + 1 - cur, extent_thresh,
-				newer_than, do_compress,
-				&sectors_defragged, max_to_defrag);
+			BTRFS_I(inode)->defrag_compress = ctrl->compress;
+		ret = defrag_one_cluster(BTRFS_I(inode), ra, ctrl, cur,
+				cluster_end + 1 - cur);
 
-		if (sectors_defragged > prev_sectors_defragged)
+		if (ctrl->sectors_defragged > prev_sectors_defragged)
 			balance_dirty_pages_ratelimited(inode->i_mapping);
 
 		btrfs_inode_unlock(inode, 0);
 		if (ret < 0)
 			break;
-		cur = cluster_end + 1;
+		cur = max(cluster_end + 1, ctrl->last_scanned);
 		if (ret > 0) {
 			ret = 0;
 			break;
 		}
+		cond_resched();
 	}
 
 	if (ra_allocated)
 		kfree(ra);
-	/*
-	 * Update range.start for autodefrag, this will indicate where to start
-	 * in next run.
-	 */
-	range->start = cur;
-	if (sectors_defragged) {
+	if (ctrl->sectors_defragged) {
 		/*
 		 * We have defragged some sectors, for compression case they
 		 * need to be written back immediately.
 		 */
-		if (range->flags & BTRFS_DEFRAG_RANGE_START_IO) {
+		if (ctrl->flags & BTRFS_DEFRAG_RANGE_START_IO) {
 			filemap_flush(inode->i_mapping);
 			if (test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT,
 				     &BTRFS_I(inode)->runtime_flags))
 				filemap_flush(inode->i_mapping);
 		}
-		if (range->compress_type == BTRFS_COMPRESS_LZO)
+		if (ctrl->compress == BTRFS_COMPRESS_LZO)
 			btrfs_set_fs_incompat(fs_info, COMPRESS_LZO);
-		else if (range->compress_type == BTRFS_COMPRESS_ZSTD)
+		else if (ctrl->compress == BTRFS_COMPRESS_ZSTD)
 			btrfs_set_fs_incompat(fs_info, COMPRESS_ZSTD);
-		ret = sectors_defragged;
 	}
 	if (do_compress) {
 		btrfs_inode_lock(inode, 0);
@@ -2147,7 +2206,7 @@ static noinline int copy_to_sk(struct btrfs_path *path,
 
 	for (i = slot; i < nritems; i++) {
 		item_off = btrfs_item_ptr_offset(leaf, i);
-		item_len = btrfs_item_size_nr(leaf, i);
+		item_len = btrfs_item_size(leaf, i);
 
 		btrfs_item_key_to_cpu(leaf, key, i);
 		if (!key_in_sk(key, sk))
@@ -2601,7 +2660,7 @@ static int btrfs_search_path_in_tree_user(struct user_namespace *mnt_userns,
 	btrfs_item_key_to_cpu(leaf, &key, slot);
 
 	item_off = btrfs_item_ptr_offset(leaf, slot);
-	item_len = btrfs_item_size_nr(leaf, slot);
+	item_len = btrfs_item_size(leaf, slot);
 	/* Check if dirid in ROOT_REF corresponds to passed dirid */
 	rref = btrfs_item_ptr(leaf, slot, struct btrfs_root_ref);
 	if (args->dirid != btrfs_root_ref_dirid(leaf, rref)) {
@@ -2803,7 +2862,7 @@ static int btrfs_ioctl_get_subvol_info(struct file *file, void __user *argp)
 
 			item_off = btrfs_item_ptr_offset(leaf, slot)
 					+ sizeof(struct btrfs_root_ref);
-			item_len = btrfs_item_size_nr(leaf, slot)
+			item_len = btrfs_item_size(leaf, slot)
 					- sizeof(struct btrfs_root_ref);
 			read_extent_buffer(leaf, subvol_info->name,
 					   item_off, item_len);
@@ -3148,6 +3207,7 @@ static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
 	struct inode *inode = file_inode(file);
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_ioctl_defrag_range_args range = {0};
+	struct btrfs_defrag_ctrl ctrl = {0};
 	int ret;
 
 	ret = mnt_want_write_file(file);
@@ -3193,8 +3253,11 @@ static int btrfs_ioctl_defrag(struct file *file, void __user *argp)
 			/* the rest are all set to zero by kzalloc */
 			range.len = (u64)-1;
 		}
-		ret = btrfs_defrag_file(file_inode(file), &file->f_ra,
-					&range, BTRFS_OLDEST_GENERATION, 0);
+		ret = btrfs_defrag_ioctl_args_to_ctrl(root->fs_info, &range,
+						      &ctrl, 0, BTRFS_OLDEST_GENERATION);
+		if (ret < 0)
+			break;
+		ret = btrfs_defrag_file(file_inode(file), &file->f_ra, &ctrl);
 		if (ret > 0)
 			ret = 0;
 		break;
@@ -3683,7 +3746,6 @@ static noinline long btrfs_ioctl_start_sync(struct btrfs_root *root,
 {
 	struct btrfs_trans_handle *trans;
 	u64 transid;
-	int ret;
 
 	trans = btrfs_attach_transaction_barrier(root);
 	if (IS_ERR(trans)) {
@@ -3695,11 +3757,7 @@ static noinline long btrfs_ioctl_start_sync(struct btrfs_root *root,
 		goto out;
 	}
 	transid = trans->transid;
-	ret = btrfs_commit_transaction_async(trans);
-	if (ret) {
-		btrfs_end_transaction(trans);
-		return ret;
-	}
+	btrfs_commit_transaction_async(trans);
 out:
 	if (argp)
 		if (copy_to_user(argp, &transid, sizeof(transid)))
diff --git a/fs/btrfs/lzo.c b/fs/btrfs/lzo.c
index 0fb90cbe7669..430ad36b8b08 100644
--- a/fs/btrfs/lzo.c
+++ b/fs/btrfs/lzo.c
@@ -55,6 +55,9 @@
  * 0x1000   | SegHdr N+1| Data payload N+1 ...                |
  */
 
+#define WORKSPACE_BUF_LENGTH	(lzo1x_worst_compress(PAGE_SIZE))
+#define WORKSPACE_CBUF_LENGTH	(lzo1x_worst_compress(PAGE_SIZE))
+
 struct workspace {
 	void *mem;
 	void *buf;	/* where decompressed data goes */
@@ -83,8 +86,8 @@ struct list_head *lzo_alloc_workspace(unsigned int level)
 		return ERR_PTR(-ENOMEM);
 
 	workspace->mem = kvmalloc(LZO1X_MEM_COMPRESS, GFP_KERNEL);
-	workspace->buf = kvmalloc(lzo1x_worst_compress(PAGE_SIZE), GFP_KERNEL);
-	workspace->cbuf = kvmalloc(lzo1x_worst_compress(PAGE_SIZE), GFP_KERNEL);
+	workspace->buf = kvmalloc(WORKSPACE_BUF_LENGTH, GFP_KERNEL);
+	workspace->cbuf = kvmalloc(WORKSPACE_CBUF_LENGTH, GFP_KERNEL);
 	if (!workspace->mem || !workspace->buf || !workspace->cbuf)
 		goto fail;
 
@@ -380,6 +383,17 @@ int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
 		kunmap(cur_page);
 		cur_in += LZO_LEN;
 
+		if (seg_len > WORKSPACE_CBUF_LENGTH) {
+			/*
+			 * seg_len shouldn't be larger than we have allocated
+			 * for workspace->cbuf
+			 */
+			btrfs_err(fs_info, "unexpectedly large lzo segment len %u",
+					seg_len);
+			ret = -EIO;
+			goto out;
+		}
+
 		/* Copy the compressed segment payload into workspace */
 		copy_compressed_segment(cb, workspace->cbuf, seg_len, &cur_in);
 
@@ -422,7 +436,7 @@ int lzo_decompress(struct list_head *ws, unsigned char *data_in,
 	struct workspace *workspace = list_entry(ws, struct workspace, list);
 	size_t in_len;
 	size_t out_len;
-	size_t max_segment_len = lzo1x_worst_compress(PAGE_SIZE);
+	size_t max_segment_len = WORKSPACE_BUF_LENGTH;
 	int ret = 0;
 	char *kaddr;
 	unsigned long bytes;
diff --git a/fs/btrfs/print-tree.c b/fs/btrfs/print-tree.c
index aae1027bd76a..0775ae9f4419 100644
--- a/fs/btrfs/print-tree.c
+++ b/fs/btrfs/print-tree.c
@@ -85,7 +85,7 @@ static void print_extent_item(struct extent_buffer *eb, int slot, int type)
 	struct btrfs_disk_key key;
 	unsigned long end;
 	unsigned long ptr;
-	u32 item_size = btrfs_item_size_nr(eb, slot);
+	u32 item_size = btrfs_item_size(eb, slot);
 	u64 flags;
 	u64 offset;
 	int ref_index = 0;
@@ -200,7 +200,6 @@ void btrfs_print_leaf(struct extent_buffer *l)
 	struct btrfs_fs_info *fs_info;
 	int i;
 	u32 type, nr;
-	struct btrfs_item *item;
 	struct btrfs_root_item *ri;
 	struct btrfs_dir_item *di;
 	struct btrfs_inode_item *ii;
@@ -224,12 +223,11 @@ void btrfs_print_leaf(struct extent_buffer *l)
 		   btrfs_leaf_free_space(l), btrfs_header_owner(l));
 	print_eb_refs_lock(l);
 	for (i = 0 ; i < nr ; i++) {
-		item = btrfs_item_nr(i);
 		btrfs_item_key_to_cpu(l, &key, i);
 		type = key.type;
 		pr_info("\titem %d key (%llu %u %llu) itemoff %d itemsize %d\n",
 			i, key.objectid, type, key.offset,
-			btrfs_item_offset(l, item), btrfs_item_size(l, item));
+			btrfs_item_offset(l, i), btrfs_item_size(l, i));
 		switch (type) {
 		case BTRFS_INODE_ITEM_KEY:
 			ii = btrfs_item_ptr(l, i, struct btrfs_inode_item);
@@ -347,7 +345,7 @@ void btrfs_print_leaf(struct extent_buffer *l)
 		case BTRFS_UUID_KEY_SUBVOL:
 		case BTRFS_UUID_KEY_RECEIVED_SUBVOL:
 			print_uuid_item(l, btrfs_item_ptr_offset(l, i),
-					btrfs_item_size_nr(l, i));
+					btrfs_item_size(l, i));
 			break;
 		}
 	}
diff --git a/fs/btrfs/props.c b/fs/btrfs/props.c
index b1cb5a8c2999..a978676aa627 100644
--- a/fs/btrfs/props.c
+++ b/fs/btrfs/props.c
@@ -158,7 +158,7 @@ static int iterate_object_props(struct btrfs_root *root,
 
 		di = btrfs_item_ptr(leaf, slot, struct btrfs_dir_item);
 		cur = 0;
-		total_len = btrfs_item_size_nr(leaf, slot);
+		total_len = btrfs_item_size(leaf, slot);
 
 		while (cur < total_len) {
 			u32 name_len = btrfs_dir_name_len(leaf, di);
diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index 26134b7476a2..3712cd5fdbfe 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -258,16 +258,19 @@ static int del_qgroup_rb(struct btrfs_fs_info *fs_info, u64 qgroupid)
 	return 0;
 }
 
-/* must be called with qgroup_lock held */
-static int add_relation_rb(struct btrfs_fs_info *fs_info,
-			   u64 memberid, u64 parentid)
+/*
+ * Add relation specified by two qgroups.
+ *
+ * Must be called with qgroup_lock held.
+ *
+ * Return: 0        on success
+ *         -ENOENT  if one of the qgroups is NULL
+ *         <0       other errors
+ */
+static int __add_relation_rb(struct btrfs_qgroup *member, struct btrfs_qgroup *parent)
 {
-	struct btrfs_qgroup *member;
-	struct btrfs_qgroup *parent;
 	struct btrfs_qgroup_list *list;
 
-	member = find_qgroup_rb(fs_info, memberid);
-	parent = find_qgroup_rb(fs_info, parentid);
 	if (!member || !parent)
 		return -ENOENT;
 
@@ -283,7 +286,27 @@ static int add_relation_rb(struct btrfs_fs_info *fs_info,
 	return 0;
 }
 
-/* must be called with qgroup_lock held */
+/*
+ * Add relation specified by two qgoup ids.
+ *
+ * Must be called with qgroup_lock held.
+ *
+ * Return: 0        on success
+ *         -ENOENT  if one of the ids does not exist
+ *         <0       other errors
+ */
+static int add_relation_rb(struct btrfs_fs_info *fs_info, u64 memberid, u64 parentid)
+{
+	struct btrfs_qgroup *member;
+	struct btrfs_qgroup *parent;
+
+	member = find_qgroup_rb(fs_info, memberid);
+	parent = find_qgroup_rb(fs_info, parentid);
+
+	return __add_relation_rb(member, parent);
+}
+
+/* Must be called with qgroup_lock held */
 static int del_relation_rb(struct btrfs_fs_info *fs_info,
 			   u64 memberid, u64 parentid)
 {
@@ -1444,7 +1467,7 @@ int btrfs_add_qgroup_relation(struct btrfs_trans_handle *trans, u64 src,
 	}
 
 	spin_lock(&fs_info->qgroup_lock);
-	ret = add_relation_rb(fs_info, src, dst);
+	ret = __add_relation_rb(member, parent);
 	if (ret < 0) {
 		spin_unlock(&fs_info->qgroup_lock);
 		goto out;
diff --git a/fs/btrfs/ref-verify.c b/fs/btrfs/ref-verify.c
index e2b9f8616501..f34130d90dee 100644
--- a/fs/btrfs/ref-verify.c
+++ b/fs/btrfs/ref-verify.c
@@ -435,7 +435,7 @@ static int process_extent_item(struct btrfs_fs_info *fs_info,
 	struct btrfs_extent_data_ref *dref;
 	struct btrfs_shared_data_ref *sref;
 	struct extent_buffer *leaf = path->nodes[0];
-	u32 item_size = btrfs_item_size_nr(leaf, slot);
+	u32 item_size = btrfs_item_size(leaf, slot);
 	unsigned long end, ptr;
 	u64 offset, flags, count;
 	int type, ret;
diff --git a/fs/btrfs/reflink.c b/fs/btrfs/reflink.c
index e0f93b357548..a3930da4eb3f 100644
--- a/fs/btrfs/reflink.c
+++ b/fs/btrfs/reflink.c
@@ -439,7 +439,7 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
 			break;
 		}
 		next_key_min_offset = key.offset + datal;
-		size = btrfs_item_size_nr(leaf, slot);
+		size = btrfs_item_size(leaf, slot);
 		read_extent_buffer(leaf, buf, btrfs_item_ptr_offset(leaf, slot),
 				   size);
 
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 33a0ee7ac590..ee0a0efc7efd 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -3149,7 +3149,7 @@ static int add_tree_block(struct reloc_control *rc,
 	u64 owner = 0;
 
 	eb =  path->nodes[0];
-	item_size = btrfs_item_size_nr(eb, path->slots[0]);
+	item_size = btrfs_item_size(eb, path->slots[0]);
 
 	if (extent_key->type == BTRFS_METADATA_ITEM_KEY ||
 	    item_size >= sizeof(*ei) + sizeof(*bi)) {
diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c
index d20166336557..3297368aa359 100644
--- a/fs/btrfs/root-tree.c
+++ b/fs/btrfs/root-tree.c
@@ -25,7 +25,7 @@ static void btrfs_read_root_item(struct extent_buffer *eb, int slot,
 	u32 len;
 	int need_reset = 0;
 
-	len = btrfs_item_size_nr(eb, slot);
+	len = btrfs_item_size(eb, slot);
 	read_extent_buffer(eb, item, btrfs_item_ptr_offset(eb, slot),
 			   min_t(u32, len, sizeof(*item)));
 	if (len < sizeof(*item))
@@ -146,7 +146,7 @@ int btrfs_update_root(struct btrfs_trans_handle *trans, struct btrfs_root
 	l = path->nodes[0];
 	slot = path->slots[0];
 	ptr = btrfs_item_ptr_offset(l, slot);
-	old_len = btrfs_item_size_nr(l, slot);
+	old_len = btrfs_item_size(l, slot);
 
 	/*
 	 * If this is the first time we update the root item which originated
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index 8f6ceea33969..d175c5ab1134 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -758,7 +758,7 @@ static void scrub_print_warning(const char *errstr, struct scrub_block *sblock)
 
 	eb = path->nodes[0];
 	ei = btrfs_item_ptr(eb, path->slots[0], struct btrfs_extent_item);
-	item_size = btrfs_item_size_nr(eb, path->slots[0]);
+	item_size = btrfs_item_size(eb, path->slots[0]);
 
 	if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) {
 		do {
diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index 040324d71118..93b9fe2dca67 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -898,7 +898,6 @@ static int iterate_inode_ref(struct btrfs_root *root, struct btrfs_path *path,
 			     iterate_inode_ref_t iterate, void *ctx)
 {
 	struct extent_buffer *eb = path->nodes[0];
-	struct btrfs_item *item;
 	struct btrfs_inode_ref *iref;
 	struct btrfs_inode_extref *extref;
 	struct btrfs_path *tmp_path;
@@ -930,12 +929,11 @@ static int iterate_inode_ref(struct btrfs_root *root, struct btrfs_path *path,
 	if (found_key->type == BTRFS_INODE_REF_KEY) {
 		ptr = (unsigned long)btrfs_item_ptr(eb, slot,
 						    struct btrfs_inode_ref);
-		item = btrfs_item_nr(slot);
-		total = btrfs_item_size(eb, item);
+		total = btrfs_item_size(eb, slot);
 		elem_size = sizeof(*iref);
 	} else {
 		ptr = btrfs_item_ptr_offset(eb, slot);
-		total = btrfs_item_size_nr(eb, slot);
+		total = btrfs_item_size(eb, slot);
 		elem_size = sizeof(*extref);
 	}
 
@@ -1018,7 +1016,6 @@ static int iterate_dir_item(struct btrfs_root *root, struct btrfs_path *path,
 {
 	int ret = 0;
 	struct extent_buffer *eb;
-	struct btrfs_item *item;
 	struct btrfs_dir_item *di;
 	struct btrfs_key di_key;
 	char *buf = NULL;
@@ -1047,11 +1044,10 @@ static int iterate_dir_item(struct btrfs_root *root, struct btrfs_path *path,
 
 	eb = path->nodes[0];
 	slot = path->slots[0];
-	item = btrfs_item_nr(slot);
 	di = btrfs_item_ptr(eb, slot, struct btrfs_dir_item);
 	cur = 0;
 	len = 0;
-	total = btrfs_item_size(eb, item);
+	total = btrfs_item_size(eb, slot);
 
 	num = 0;
 	while (cur < total) {
@@ -3622,7 +3618,7 @@ static int is_ancestor(struct btrfs_root *root,
 		    key.type != BTRFS_INODE_EXTREF_KEY)
 			break;
 
-		item_size = btrfs_item_size_nr(leaf, slot);
+		item_size = btrfs_item_size(leaf, slot);
 		while (cur_offset < item_size) {
 			u64 parent;
 			u64 parent_gen;
@@ -4983,6 +4979,10 @@ static int put_file_data(struct send_ctx *sctx, u64 offset, u32 len)
 			lock_page(page);
 			if (!PageUptodate(page)) {
 				unlock_page(page);
+				btrfs_err(fs_info,
+			"send: IO error at offset %llu for inode %llu root %llu",
+					page_offset(page), sctx->cur_ino,
+					sctx->send_root->root_key.objectid);
 				put_page(page);
 				ret = -EIO;
 				break;
@@ -6566,7 +6566,7 @@ static int compare_refs(struct send_ctx *sctx, struct btrfs_path *path,
 	}
 
 	leaf = path->nodes[0];
-	item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+	item_size = btrfs_item_size(leaf, path->slots[0]);
 	ptr = btrfs_item_ptr_offset(leaf, path->slots[0]);
 	while (cur_offset < item_size) {
 		extref = (struct btrfs_inode_extref *)(ptr +
@@ -6791,8 +6791,8 @@ static int tree_compare_item(struct btrfs_path *left_path,
 	int len1, len2;
 	unsigned long off1, off2;
 
-	len1 = btrfs_item_size_nr(left_path->nodes[0], left_path->slots[0]);
-	len2 = btrfs_item_size_nr(right_path->nodes[0], right_path->slots[0]);
+	len1 = btrfs_item_size(left_path->nodes[0], left_path->slots[0]);
+	len2 = btrfs_item_size(right_path->nodes[0], right_path->slots[0]);
 	if (len1 != len2)
 		return 1;
 
diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c
index f9eff3b0f77c..836a20fdfca1 100644
--- a/fs/btrfs/sysfs.c
+++ b/fs/btrfs/sysfs.c
@@ -1104,6 +1104,11 @@ static inline struct btrfs_fs_info *to_fs_info(struct kobject *kobj)
 static char btrfs_unknown_feature_names[FEAT_MAX][NUM_FEATURE_BITS][BTRFS_FEATURE_NAME_MAX];
 static struct btrfs_feature_attr btrfs_feature_attrs[FEAT_MAX][NUM_FEATURE_BITS];
 
+static_assert(ARRAY_SIZE(btrfs_unknown_feature_names) ==
+	      ARRAY_SIZE(btrfs_feature_attrs));
+static_assert(ARRAY_SIZE(btrfs_unknown_feature_names[0]) ==
+	      ARRAY_SIZE(btrfs_feature_attrs[0]));
+
 static const u64 supported_feature_masks[FEAT_MAX] = {
 	[FEAT_COMPAT]    = BTRFS_FEATURE_COMPAT_SUPP,
 	[FEAT_COMPAT_RO] = BTRFS_FEATURE_COMPAT_RO_SUPP,
@@ -1272,11 +1277,6 @@ static void init_feature_attrs(void)
 	struct btrfs_feature_attr *fa;
 	int set, i;
 
-	BUILD_BUG_ON(ARRAY_SIZE(btrfs_unknown_feature_names) !=
-		     ARRAY_SIZE(btrfs_feature_attrs));
-	BUILD_BUG_ON(ARRAY_SIZE(btrfs_unknown_feature_names[0]) !=
-		     ARRAY_SIZE(btrfs_feature_attrs[0]));
-
 	memset(btrfs_feature_attrs, 0, sizeof(btrfs_feature_attrs));
 	memset(btrfs_unknown_feature_names, 0,
 	       sizeof(btrfs_unknown_feature_names));
diff --git a/fs/btrfs/tests/extent-buffer-tests.c b/fs/btrfs/tests/extent-buffer-tests.c
index 2a95f7224e18..51a8b075c259 100644
--- a/fs/btrfs/tests/extent-buffer-tests.c
+++ b/fs/btrfs/tests/extent-buffer-tests.c
@@ -15,7 +15,6 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 	struct btrfs_path *path = NULL;
 	struct btrfs_root *root = NULL;
 	struct extent_buffer *eb;
-	struct btrfs_item *item;
 	char *value = "mary had a little lamb";
 	char *split1 = "mary had a little";
 	char *split2 = " lamb";
@@ -61,7 +60,6 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 	key.offset = 0;
 
 	btrfs_setup_item_for_insert(root, path, &key, value_len);
-	item = btrfs_item_nr(0);
 	write_extent_buffer(eb, value, btrfs_item_ptr_offset(eb, 0),
 			    value_len);
 
@@ -90,8 +88,7 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 		goto out;
 	}
 
-	item = btrfs_item_nr(0);
-	if (btrfs_item_size(eb, item) != strlen(split1)) {
+	if (btrfs_item_size(eb, 0) != strlen(split1)) {
 		test_err("invalid len in the first split");
 		ret = -EINVAL;
 		goto out;
@@ -115,8 +112,7 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 		goto out;
 	}
 
-	item = btrfs_item_nr(1);
-	if (btrfs_item_size(eb, item) != strlen(split2)) {
+	if (btrfs_item_size(eb, 1) != strlen(split2)) {
 		test_err("invalid len in the second split");
 		ret = -EINVAL;
 		goto out;
@@ -147,8 +143,7 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 		goto out;
 	}
 
-	item = btrfs_item_nr(0);
-	if (btrfs_item_size(eb, item) != strlen(split3)) {
+	if (btrfs_item_size(eb, 0) != strlen(split3)) {
 		test_err("invalid len in the first split");
 		ret = -EINVAL;
 		goto out;
@@ -171,8 +166,7 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 		goto out;
 	}
 
-	item = btrfs_item_nr(1);
-	if (btrfs_item_size(eb, item) != strlen(split4)) {
+	if (btrfs_item_size(eb, 1) != strlen(split4)) {
 		test_err("invalid len in the second split");
 		ret = -EINVAL;
 		goto out;
@@ -195,8 +189,7 @@ static int test_btrfs_split_item(u32 sectorsize, u32 nodesize)
 		goto out;
 	}
 
-	item = btrfs_item_nr(2);
-	if (btrfs_item_size(eb, item) != strlen(split2)) {
+	if (btrfs_item_size(eb, 2) != strlen(split2)) {
 		test_err("invalid len in the second split");
 		ret = -EINVAL;
 		goto out;
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index 27b93a6c41bb..f3c094af9283 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -1861,50 +1861,14 @@ int btrfs_transaction_blocked(struct btrfs_fs_info *info)
 	return ret;
 }
 
-/*
- * commit transactions asynchronously. once btrfs_commit_transaction_async
- * returns, any subsequent transaction will not be allowed to join.
- */
-struct btrfs_async_commit {
-	struct btrfs_trans_handle *newtrans;
-	struct work_struct work;
-};
-
-static void do_async_commit(struct work_struct *work)
-{
-	struct btrfs_async_commit *ac =
-		container_of(work, struct btrfs_async_commit, work);
-
-	/*
-	 * We've got freeze protection passed with the transaction.
-	 * Tell lockdep about it.
-	 */
-	if (ac->newtrans->type & __TRANS_FREEZABLE)
-		__sb_writers_acquired(ac->newtrans->fs_info->sb, SB_FREEZE_FS);
-
-	current->journal_info = ac->newtrans;
-
-	btrfs_commit_transaction(ac->newtrans);
-	kfree(ac);
-}
-
-int btrfs_commit_transaction_async(struct btrfs_trans_handle *trans)
+void btrfs_commit_transaction_async(struct btrfs_trans_handle *trans)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	struct btrfs_async_commit *ac;
 	struct btrfs_transaction *cur_trans;
 
-	ac = kmalloc(sizeof(*ac), GFP_NOFS);
-	if (!ac)
-		return -ENOMEM;
-
-	INIT_WORK(&ac->work, do_async_commit);
-	ac->newtrans = btrfs_join_transaction(trans->root);
-	if (IS_ERR(ac->newtrans)) {
-		int err = PTR_ERR(ac->newtrans);
-		kfree(ac);
-		return err;
-	}
+	/* Kick the transaction kthread. */
+	set_bit(BTRFS_FS_COMMIT_TRANS, &fs_info->flags);
+	wake_up_process(fs_info->transaction_kthread);
 
 	/* take transaction reference */
 	cur_trans = trans->transaction;
@@ -1912,14 +1876,6 @@ int btrfs_commit_transaction_async(struct btrfs_trans_handle *trans)
 
 	btrfs_end_transaction(trans);
 
-	/*
-	 * Tell lockdep we've released the freeze rwsem, since the
-	 * async commit thread will be the one to unlock it.
-	 */
-	if (ac->newtrans->type & __TRANS_FREEZABLE)
-		__sb_writers_release(fs_info->sb, SB_FREEZE_FS);
-
-	schedule_work(&ac->work);
 	/*
 	 * Wait for the current transaction commit to start and block
 	 * subsequent transaction joins
@@ -1927,14 +1883,9 @@ int btrfs_commit_transaction_async(struct btrfs_trans_handle *trans)
 	wait_event(fs_info->transaction_blocked_wait,
 		   cur_trans->state >= TRANS_STATE_COMMIT_START ||
 		   TRANS_ABORTED(cur_trans));
-	if (current->journal_info == trans)
-		current->journal_info = NULL;
-
 	btrfs_put_transaction(cur_trans);
-	return 0;
 }
 
-
 static void cleanup_transaction(struct btrfs_trans_handle *trans, int err)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
@@ -2013,16 +1964,24 @@ static void btrfs_cleanup_pending_block_groups(struct btrfs_trans_handle *trans)
 static inline int btrfs_start_delalloc_flush(struct btrfs_fs_info *fs_info)
 {
 	/*
-	 * We use writeback_inodes_sb here because if we used
+	 * We use try_to_writeback_inodes_sb() here because if we used
 	 * btrfs_start_delalloc_roots we would deadlock with fs freeze.
 	 * Currently are holding the fs freeze lock, if we do an async flush
 	 * we'll do btrfs_join_transaction() and deadlock because we need to
 	 * wait for the fs freeze lock.  Using the direct flushing we benefit
 	 * from already being in a transaction and our join_transaction doesn't
 	 * have to re-take the fs freeze lock.
+	 *
+	 * Note that try_to_writeback_inodes_sb() will only trigger writeback
+	 * if it can read lock sb->s_umount. It will always be able to lock it,
+	 * except when the filesystem is being unmounted or being frozen, but in
+	 * those cases sync_filesystem() is called, which results in calling
+	 * writeback_inodes_sb() while holding a write lock on sb->s_umount.
+	 * Note that we don't call writeback_inodes_sb() directly, because it
+	 * will emit a warning if sb->s_umount is not locked.
 	 */
 	if (btrfs_test_opt(fs_info, FLUSHONCOMMIT))
-		writeback_inodes_sb(fs_info->sb, WB_REASON_SYNC);
+		try_to_writeback_inodes_sb(fs_info->sb, WB_REASON_SYNC);
 	return 0;
 }
 
@@ -2224,6 +2183,13 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)
 	wait_event(cur_trans->writer_wait,
 		   atomic_read(&cur_trans->num_writers) == 1);
 
+	/*
+	 * We've started the commit, clear the flag in case we were triggered to
+	 * do an async commit but somebody else started before the transaction
+	 * kthread could do the work.
+	 */
+	clear_bit(BTRFS_FS_COMMIT_TRANS, &fs_info->flags);
+
 	if (TRANS_ABORTED(cur_trans)) {
 		ret = cur_trans->aborted;
 		goto scrub_continue;
diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h
index eba07b8119bb..d0705485f5c8 100644
--- a/fs/btrfs/transaction.h
+++ b/fs/btrfs/transaction.h
@@ -219,7 +219,7 @@ void btrfs_add_dead_root(struct btrfs_root *root);
 int btrfs_defrag_root(struct btrfs_root *root);
 int btrfs_clean_one_deleted_snapshot(struct btrfs_root *root);
 int btrfs_commit_transaction(struct btrfs_trans_handle *trans);
-int btrfs_commit_transaction_async(struct btrfs_trans_handle *trans);
+void btrfs_commit_transaction_async(struct btrfs_trans_handle *trans);
 int btrfs_end_transaction_throttle(struct btrfs_trans_handle *trans);
 bool btrfs_should_end_transaction(struct btrfs_trans_handle *trans);
 void btrfs_throttle(struct btrfs_fs_info *fs_info);
diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c
index 7733e8ac0a69..72e1c942197d 100644
--- a/fs/btrfs/tree-checker.c
+++ b/fs/btrfs/tree-checker.c
@@ -202,7 +202,7 @@ static int check_extent_data_item(struct extent_buffer *leaf,
 	struct btrfs_fs_info *fs_info = leaf->fs_info;
 	struct btrfs_file_extent_item *fi;
 	u32 sectorsize = fs_info->sectorsize;
-	u32 item_size = btrfs_item_size_nr(leaf, slot);
+	u32 item_size = btrfs_item_size(leaf, slot);
 	u64 extent_end;
 
 	if (unlikely(!IS_ALIGNED(key->offset, sectorsize))) {
@@ -354,17 +354,17 @@ static int check_csum_item(struct extent_buffer *leaf, struct btrfs_key *key,
 			key->offset, sectorsize);
 		return -EUCLEAN;
 	}
-	if (unlikely(!IS_ALIGNED(btrfs_item_size_nr(leaf, slot), csumsize))) {
+	if (unlikely(!IS_ALIGNED(btrfs_item_size(leaf, slot), csumsize))) {
 		generic_err(leaf, slot,
 	"unaligned item size for csum item, have %u should be aligned to %u",
-			btrfs_item_size_nr(leaf, slot), csumsize);
+			btrfs_item_size(leaf, slot), csumsize);
 		return -EUCLEAN;
 	}
 	if (slot > 0 && prev_key->type == BTRFS_EXTENT_CSUM_KEY) {
 		u64 prev_csum_end;
 		u32 prev_item_size;
 
-		prev_item_size = btrfs_item_size_nr(leaf, slot - 1);
+		prev_item_size = btrfs_item_size(leaf, slot - 1);
 		prev_csum_end = (prev_item_size / csumsize) * sectorsize;
 		prev_csum_end += prev_key->offset;
 		if (unlikely(prev_csum_end > key->offset)) {
@@ -483,7 +483,7 @@ static int check_dir_item(struct extent_buffer *leaf,
 {
 	struct btrfs_fs_info *fs_info = leaf->fs_info;
 	struct btrfs_dir_item *di;
-	u32 item_size = btrfs_item_size_nr(leaf, slot);
+	u32 item_size = btrfs_item_size(leaf, slot);
 	u32 cur = 0;
 
 	if (unlikely(!check_prev_ino(leaf, key, slot, prev_key)))
@@ -640,7 +640,7 @@ static int check_block_group_item(struct extent_buffer *leaf,
 				  struct btrfs_key *key, int slot)
 {
 	struct btrfs_block_group_item bgi;
-	u32 item_size = btrfs_item_size_nr(leaf, slot);
+	u32 item_size = btrfs_item_size(leaf, slot);
 	u64 flags;
 	u64 type;
 
@@ -912,10 +912,10 @@ static int check_leaf_chunk_item(struct extent_buffer *leaf,
 {
 	int num_stripes;
 
-	if (unlikely(btrfs_item_size_nr(leaf, slot) < sizeof(struct btrfs_chunk))) {
+	if (unlikely(btrfs_item_size(leaf, slot) < sizeof(struct btrfs_chunk))) {
 		chunk_err(leaf, chunk, key->offset,
 			"invalid chunk item size: have %u expect [%zu, %u)",
-			btrfs_item_size_nr(leaf, slot),
+			btrfs_item_size(leaf, slot),
 			sizeof(struct btrfs_chunk),
 			BTRFS_LEAF_DATA_SIZE(leaf->fs_info));
 		return -EUCLEAN;
@@ -927,10 +927,10 @@ static int check_leaf_chunk_item(struct extent_buffer *leaf,
 		goto out;
 
 	if (unlikely(btrfs_chunk_item_size(num_stripes) !=
-		     btrfs_item_size_nr(leaf, slot))) {
+		     btrfs_item_size(leaf, slot))) {
 		chunk_err(leaf, chunk, key->offset,
 			"invalid chunk item size: have %u expect %lu",
-			btrfs_item_size_nr(leaf, slot),
+			btrfs_item_size(leaf, slot),
 			btrfs_chunk_item_size(num_stripes));
 		return -EUCLEAN;
 	}
@@ -1095,12 +1095,12 @@ static int check_root_item(struct extent_buffer *leaf, struct btrfs_key *key,
 	if (unlikely(ret < 0))
 		return ret;
 
-	if (unlikely(btrfs_item_size_nr(leaf, slot) != sizeof(ri) &&
-		     btrfs_item_size_nr(leaf, slot) !=
+	if (unlikely(btrfs_item_size(leaf, slot) != sizeof(ri) &&
+		     btrfs_item_size(leaf, slot) !=
 		     btrfs_legacy_root_item_size())) {
 		generic_err(leaf, slot,
 			    "invalid root item size, have %u expect %zu or %u",
-			    btrfs_item_size_nr(leaf, slot), sizeof(ri),
+			    btrfs_item_size(leaf, slot), sizeof(ri),
 			    btrfs_legacy_root_item_size());
 		return -EUCLEAN;
 	}
@@ -1111,7 +1111,7 @@ static int check_root_item(struct extent_buffer *leaf, struct btrfs_key *key,
 	 * And since we allow geneartion_v2 as 0, it will still pass the check.
 	 */
 	read_extent_buffer(leaf, &ri, btrfs_item_ptr_offset(leaf, slot),
-			   btrfs_item_size_nr(leaf, slot));
+			   btrfs_item_size(leaf, slot));
 
 	/* Generation related */
 	if (unlikely(btrfs_root_generation(&ri) >
@@ -1208,7 +1208,7 @@ static int check_extent_item(struct extent_buffer *leaf,
 	bool is_tree_block = false;
 	unsigned long ptr;	/* Current pointer inside inline refs */
 	unsigned long end;	/* Extent item end */
-	const u32 item_size = btrfs_item_size_nr(leaf, slot);
+	const u32 item_size = btrfs_item_size(leaf, slot);
 	u64 flags;
 	u64 generation;
 	u64 total_refs;		/* Total refs in btrfs_extent_item */
@@ -1432,10 +1432,10 @@ static int check_simple_keyed_refs(struct extent_buffer *leaf,
 	if (key->type == BTRFS_SHARED_DATA_REF_KEY)
 		expect_item_size = sizeof(struct btrfs_shared_data_ref);
 
-	if (unlikely(btrfs_item_size_nr(leaf, slot) != expect_item_size)) {
+	if (unlikely(btrfs_item_size(leaf, slot) != expect_item_size)) {
 		generic_err(leaf, slot,
 		"invalid item size, have %u expect %u for key type %u",
-			    btrfs_item_size_nr(leaf, slot),
+			    btrfs_item_size(leaf, slot),
 			    expect_item_size, key->type);
 		return -EUCLEAN;
 	}
@@ -1460,12 +1460,12 @@ static int check_extent_data_ref(struct extent_buffer *leaf,
 {
 	struct btrfs_extent_data_ref *dref;
 	unsigned long ptr = btrfs_item_ptr_offset(leaf, slot);
-	const unsigned long end = ptr + btrfs_item_size_nr(leaf, slot);
+	const unsigned long end = ptr + btrfs_item_size(leaf, slot);
 
-	if (unlikely(btrfs_item_size_nr(leaf, slot) % sizeof(*dref) != 0)) {
+	if (unlikely(btrfs_item_size(leaf, slot) % sizeof(*dref) != 0)) {
 		generic_err(leaf, slot,
 	"invalid item size, have %u expect aligned to %zu for key type %u",
-			    btrfs_item_size_nr(leaf, slot),
+			    btrfs_item_size(leaf, slot),
 			    sizeof(*dref), key->type);
 		return -EUCLEAN;
 	}
@@ -1507,16 +1507,16 @@ static int check_inode_ref(struct extent_buffer *leaf,
 	if (unlikely(!check_prev_ino(leaf, key, slot, prev_key)))
 		return -EUCLEAN;
 	/* namelen can't be 0, so item_size == sizeof() is also invalid */
-	if (unlikely(btrfs_item_size_nr(leaf, slot) <= sizeof(*iref))) {
+	if (unlikely(btrfs_item_size(leaf, slot) <= sizeof(*iref))) {
 		inode_ref_err(leaf, slot,
 			"invalid item size, have %u expect (%zu, %u)",
-			btrfs_item_size_nr(leaf, slot),
+			btrfs_item_size(leaf, slot),
 			sizeof(*iref), BTRFS_LEAF_DATA_SIZE(leaf->fs_info));
 		return -EUCLEAN;
 	}
 
 	ptr = btrfs_item_ptr_offset(leaf, slot);
-	end = ptr + btrfs_item_size_nr(leaf, slot);
+	end = ptr + btrfs_item_size(leaf, slot);
 	while (ptr < end) {
 		u16 namelen;
 
@@ -1689,12 +1689,12 @@ static int check_leaf(struct extent_buffer *leaf, bool check_item_data)
 		if (slot == 0)
 			item_end_expected = BTRFS_LEAF_DATA_SIZE(fs_info);
 		else
-			item_end_expected = btrfs_item_offset_nr(leaf,
+			item_end_expected = btrfs_item_offset(leaf,
 								 slot - 1);
-		if (unlikely(btrfs_item_end_nr(leaf, slot) != item_end_expected)) {
+		if (unlikely(btrfs_item_data_end(leaf, slot) != item_end_expected)) {
 			generic_err(leaf, slot,
 				"unexpected item end, have %u expect %u",
-				btrfs_item_end_nr(leaf, slot),
+				btrfs_item_data_end(leaf, slot),
 				item_end_expected);
 			return -EUCLEAN;
 		}
@@ -1704,11 +1704,11 @@ static int check_leaf(struct extent_buffer *leaf, bool check_item_data)
 		 * just in case all the items are consistent to each other, but
 		 * all point outside of the leaf.
 		 */
-		if (unlikely(btrfs_item_end_nr(leaf, slot) >
+		if (unlikely(btrfs_item_data_end(leaf, slot) >
 			     BTRFS_LEAF_DATA_SIZE(fs_info))) {
 			generic_err(leaf, slot,
 			"slot end outside of leaf, have %u expect range [0, %u]",
-				btrfs_item_end_nr(leaf, slot),
+				btrfs_item_data_end(leaf, slot),
 				BTRFS_LEAF_DATA_SIZE(fs_info));
 			return -EUCLEAN;
 		}
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 6993dcdba6f1..cc3a8d8a3841 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -386,7 +386,7 @@ static int do_overwrite_item(struct btrfs_trans_handle *trans,
 	if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID)
 		overwrite_root = 1;
 
-	item_size = btrfs_item_size_nr(eb, slot);
+	item_size = btrfs_item_size(eb, slot);
 	src_ptr = btrfs_item_ptr_offset(eb, slot);
 
 	/* Our caller must have done a search for the key for us. */
@@ -409,7 +409,7 @@ static int do_overwrite_item(struct btrfs_trans_handle *trans,
 	if (ret == 0) {
 		char *src_copy;
 		char *dst_copy;
-		u32 dst_size = btrfs_item_size_nr(path->nodes[0],
+		u32 dst_size = btrfs_item_size(path->nodes[0],
 						  path->slots[0]);
 		if (dst_size != item_size)
 			goto insert;
@@ -503,7 +503,7 @@ static int do_overwrite_item(struct btrfs_trans_handle *trans,
 	/* make sure any existing item is the correct size */
 	if (ret == -EEXIST || ret == -EOVERFLOW) {
 		u32 found_size;
-		found_size = btrfs_item_size_nr(path->nodes[0],
+		found_size = btrfs_item_size(path->nodes[0],
 						path->slots[0]);
 		if (found_size > item_size)
 			btrfs_truncate_item(path, item_size, 1);
@@ -1096,7 +1096,7 @@ static inline int __add_inode_ref(struct btrfs_trans_handle *trans,
 		 * otherwise they must be unlinked as a conflict
 		 */
 		ptr = btrfs_item_ptr_offset(leaf, path->slots[0]);
-		ptr_end = ptr + btrfs_item_size_nr(leaf, path->slots[0]);
+		ptr_end = ptr + btrfs_item_size(leaf, path->slots[0]);
 		while (ptr < ptr_end) {
 			victim_ref = (struct btrfs_inode_ref *)ptr;
 			victim_name_len = btrfs_inode_ref_name_len(leaf,
@@ -1155,7 +1155,7 @@ static inline int __add_inode_ref(struct btrfs_trans_handle *trans,
 
 		leaf = path->nodes[0];
 
-		item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+		item_size = btrfs_item_size(leaf, path->slots[0]);
 		base = btrfs_item_ptr_offset(leaf, path->slots[0]);
 
 		while (cur_offset < item_size) {
@@ -1318,7 +1318,7 @@ static int unlink_old_inode_refs(struct btrfs_trans_handle *trans,
 
 	eb = path->nodes[0];
 	ref_ptr = btrfs_item_ptr_offset(eb, path->slots[0]);
-	ref_end = ref_ptr + btrfs_item_size_nr(eb, path->slots[0]);
+	ref_end = ref_ptr + btrfs_item_size(eb, path->slots[0]);
 	while (ref_ptr < ref_end) {
 		char *name = NULL;
 		int namelen;
@@ -1504,7 +1504,7 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans,
 	int ref_struct_size;
 
 	ref_ptr = btrfs_item_ptr_offset(eb, slot);
-	ref_end = ref_ptr + btrfs_item_size_nr(eb, slot);
+	ref_end = ref_ptr + btrfs_item_size(eb, slot);
 
 	if (key->type == BTRFS_INODE_EXTREF_KEY) {
 		struct btrfs_inode_extref *r;
@@ -1678,7 +1678,7 @@ static int count_inode_extrefs(struct btrfs_root *root,
 			break;
 
 		leaf = path->nodes[0];
-		item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+		item_size = btrfs_item_size(leaf, path->slots[0]);
 		ptr = btrfs_item_ptr_offset(leaf, path->slots[0]);
 		cur_offset = 0;
 
@@ -1732,7 +1732,7 @@ static int count_inode_refs(struct btrfs_root *root,
 		    key.type != BTRFS_INODE_REF_KEY)
 			break;
 		ptr = btrfs_item_ptr_offset(path->nodes[0], path->slots[0]);
-		ptr_end = ptr + btrfs_item_size_nr(path->nodes[0],
+		ptr_end = ptr + btrfs_item_size(path->nodes[0],
 						   path->slots[0]);
 		while (ptr < ptr_end) {
 			struct btrfs_inode_ref *ref;
@@ -1950,6 +1950,34 @@ static noinline int insert_one_name(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
+static int delete_conflicting_dir_entry(struct btrfs_trans_handle *trans,
+					struct btrfs_inode *dir,
+					struct btrfs_path *path,
+					struct btrfs_dir_item *dst_di,
+					const struct btrfs_key *log_key,
+					u8 log_type,
+					bool exists)
+{
+	struct btrfs_key found_key;
+
+	btrfs_dir_item_key_to_cpu(path->nodes[0], dst_di, &found_key);
+	/* The existing dentry points to the same inode, don't delete it. */
+	if (found_key.objectid == log_key->objectid &&
+	    found_key.type == log_key->type &&
+	    found_key.offset == log_key->offset &&
+	    btrfs_dir_type(path->nodes[0], dst_di) == log_type)
+		return 1;
+
+	/*
+	 * Don't drop the conflicting directory entry if the inode for the new
+	 * entry doesn't exist.
+	 */
+	if (!exists)
+		return 0;
+
+	return drop_one_dir_item(trans, path, dir, dst_di);
+}
+
 /*
  * take a single entry in a log directory item and replay it into
  * the subvolume.
@@ -1975,14 +2003,17 @@ static noinline int replay_one_name(struct btrfs_trans_handle *trans,
 {
 	char *name;
 	int name_len;
-	struct btrfs_dir_item *dst_di;
-	struct btrfs_key found_key;
+	struct btrfs_dir_item *dir_dst_di;
+	struct btrfs_dir_item *index_dst_di;
+	bool dir_dst_matches = false;
+	bool index_dst_matches = false;
 	struct btrfs_key log_key;
+	struct btrfs_key search_key;
 	struct inode *dir;
 	u8 log_type;
 	bool exists;
 	int ret;
-	bool update_size = (key->type == BTRFS_DIR_INDEX_KEY);
+	bool update_size = true;
 	bool name_added = false;
 
 	dir = read_one_inode(root, key->objectid);
@@ -2008,76 +2039,53 @@ static noinline int replay_one_name(struct btrfs_trans_handle *trans,
 	exists = (ret == 0);
 	ret = 0;
 
-	if (key->type == BTRFS_DIR_ITEM_KEY) {
-		dst_di = btrfs_lookup_dir_item(trans, root, path, key->objectid,
-				       name, name_len, 1);
-	} else if (key->type == BTRFS_DIR_INDEX_KEY) {
-		dst_di = btrfs_lookup_dir_index_item(trans, root, path,
-						     key->objectid,
-						     key->offset, name,
-						     name_len, 1);
-	} else {
-		/* Corruption */
-		ret = -EINVAL;
+	dir_dst_di = btrfs_lookup_dir_item(trans, root, path, key->objectid,
+					   name, name_len, 1);
+	if (IS_ERR(dir_dst_di)) {
+		ret = PTR_ERR(dir_dst_di);
 		goto out;
-	}
-
-	if (IS_ERR(dst_di)) {
-		ret = PTR_ERR(dst_di);
-		goto out;
-	} else if (!dst_di) {
-		/* we need a sequence number to insert, so we only
-		 * do inserts for the BTRFS_DIR_INDEX_KEY types
-		 */
-		if (key->type != BTRFS_DIR_INDEX_KEY)
+	} else if (dir_dst_di) {
+		ret = delete_conflicting_dir_entry(trans, BTRFS_I(dir), path,
+						   dir_dst_di, &log_key, log_type,
+						   exists);
+		if (ret < 0)
 			goto out;
-		goto insert;
+		dir_dst_matches = (ret == 1);
 	}
 
-	btrfs_dir_item_key_to_cpu(path->nodes[0], dst_di, &found_key);
-	/* the existing item matches the logged item */
-	if (found_key.objectid == log_key.objectid &&
-	    found_key.type == log_key.type &&
-	    found_key.offset == log_key.offset &&
-	    btrfs_dir_type(path->nodes[0], dst_di) == log_type) {
+	btrfs_release_path(path);
+
+	index_dst_di = btrfs_lookup_dir_index_item(trans, root, path,
+						   key->objectid, key->offset,
+						   name, name_len, 1);
+	if (IS_ERR(index_dst_di)) {
+		ret = PTR_ERR(index_dst_di);
+		goto out;
+	} else if (index_dst_di) {
+		ret = delete_conflicting_dir_entry(trans, BTRFS_I(dir), path,
+						   index_dst_di, &log_key,
+						   log_type, exists);
+		if (ret < 0)
+			goto out;
+		index_dst_matches = (ret == 1);
+	}
+
+	btrfs_release_path(path);
+
+	if (dir_dst_matches && index_dst_matches) {
+		ret = 0;
 		update_size = false;
 		goto out;
 	}
 
-	/*
-	 * don't drop the conflicting directory entry if the inode
-	 * for the new entry doesn't exist
-	 */
-	if (!exists)
-		goto out;
-
-	ret = drop_one_dir_item(trans, path, BTRFS_I(dir), dst_di);
-	if (ret)
-		goto out;
-
-	if (key->type == BTRFS_DIR_INDEX_KEY)
-		goto insert;
-out:
-	btrfs_release_path(path);
-	if (!ret && update_size) {
-		btrfs_i_size_write(BTRFS_I(dir), dir->i_size + name_len * 2);
-		ret = btrfs_update_inode(trans, root, BTRFS_I(dir));
-	}
-	kfree(name);
-	iput(dir);
-	if (!ret && name_added)
-		ret = 1;
-	return ret;
-
-insert:
 	/*
 	 * Check if the inode reference exists in the log for the given name,
 	 * inode and parent inode
 	 */
-	found_key.objectid = log_key.objectid;
-	found_key.type = BTRFS_INODE_REF_KEY;
-	found_key.offset = key->objectid;
-	ret = backref_in_log(root->log_root, &found_key, 0, name, name_len);
+	search_key.objectid = log_key.objectid;
+	search_key.type = BTRFS_INODE_REF_KEY;
+	search_key.offset = key->objectid;
+	ret = backref_in_log(root->log_root, &search_key, 0, name, name_len);
 	if (ret < 0) {
 	        goto out;
 	} else if (ret) {
@@ -2087,10 +2095,10 @@ static noinline int replay_one_name(struct btrfs_trans_handle *trans,
 	        goto out;
 	}
 
-	found_key.objectid = log_key.objectid;
-	found_key.type = BTRFS_INODE_EXTREF_KEY;
-	found_key.offset = key->objectid;
-	ret = backref_in_log(root->log_root, &found_key, key->objectid, name,
+	search_key.objectid = log_key.objectid;
+	search_key.type = BTRFS_INODE_EXTREF_KEY;
+	search_key.offset = key->objectid;
+	ret = backref_in_log(root->log_root, &search_key, key->objectid, name,
 			     name_len);
 	if (ret < 0) {
 		goto out;
@@ -2109,87 +2117,76 @@ static noinline int replay_one_name(struct btrfs_trans_handle *trans,
 		name_added = true;
 	update_size = false;
 	ret = 0;
-	goto out;
+
+out:
+	if (!ret && update_size) {
+		btrfs_i_size_write(BTRFS_I(dir), dir->i_size + name_len * 2);
+		ret = btrfs_update_inode(trans, root, BTRFS_I(dir));
+	}
+	kfree(name);
+	iput(dir);
+	if (!ret && name_added)
+		ret = 1;
+	return ret;
 }
 
-/*
- * find all the names in a directory item and reconcile them into
- * the subvolume.  Only BTRFS_DIR_ITEM_KEY types will have more than
- * one name in a directory item, but the same code gets used for
- * both directory index types
- */
+/* Replay one dir item from a BTRFS_DIR_INDEX_KEY key. */
 static noinline int replay_one_dir_item(struct btrfs_trans_handle *trans,
 					struct btrfs_root *root,
 					struct btrfs_path *path,
 					struct extent_buffer *eb, int slot,
 					struct btrfs_key *key)
 {
-	int ret = 0;
-	u32 item_size = btrfs_item_size_nr(eb, slot);
+	int ret;
 	struct btrfs_dir_item *di;
-	int name_len;
-	unsigned long ptr;
-	unsigned long ptr_end;
-	struct btrfs_path *fixup_path = NULL;
 
-	ptr = btrfs_item_ptr_offset(eb, slot);
-	ptr_end = ptr + item_size;
-	while (ptr < ptr_end) {
-		di = (struct btrfs_dir_item *)ptr;
-		name_len = btrfs_dir_name_len(eb, di);
-		ret = replay_one_name(trans, root, path, eb, di, key);
-		if (ret < 0)
-			break;
-		ptr = (unsigned long)(di + 1);
-		ptr += name_len;
+	/* We only log dir index keys, which only contain a single dir item. */
+	ASSERT(key->type == BTRFS_DIR_INDEX_KEY);
 
-		/*
-		 * If this entry refers to a non-directory (directories can not
-		 * have a link count > 1) and it was added in the transaction
-		 * that was not committed, make sure we fixup the link count of
-		 * the inode it the entry points to. Otherwise something like
-		 * the following would result in a directory pointing to an
-		 * inode with a wrong link that does not account for this dir
-		 * entry:
-		 *
-		 * mkdir testdir
-		 * touch testdir/foo
-		 * touch testdir/bar
-		 * sync
-		 *
-		 * ln testdir/bar testdir/bar_link
-		 * ln testdir/foo testdir/foo_link
-		 * xfs_io -c "fsync" testdir/bar
-		 *
-		 * <power failure>
-		 *
-		 * mount fs, log replay happens
-		 *
-		 * File foo would remain with a link count of 1 when it has two
-		 * entries pointing to it in the directory testdir. This would
-		 * make it impossible to ever delete the parent directory has
-		 * it would result in stale dentries that can never be deleted.
-		 */
-		if (ret == 1 && btrfs_dir_type(eb, di) != BTRFS_FT_DIR) {
-			struct btrfs_key di_key;
+	di = btrfs_item_ptr(eb, slot, struct btrfs_dir_item);
+	ret = replay_one_name(trans, root, path, eb, di, key);
+	if (ret < 0)
+		return ret;
 
-			if (!fixup_path) {
-				fixup_path = btrfs_alloc_path();
-				if (!fixup_path) {
-					ret = -ENOMEM;
-					break;
-				}
-			}
+	/*
+	 * If this entry refers to a non-directory (directories can not have a
+	 * link count > 1) and it was added in the transaction that was not
+	 * committed, make sure we fixup the link count of the inode the entry
+	 * points to. Otherwise something like the following would result in a
+	 * directory pointing to an inode with a wrong link that does not account
+	 * for this dir entry:
+	 *
+	 * mkdir testdir
+	 * touch testdir/foo
+	 * touch testdir/bar
+	 * sync
+	 *
+	 * ln testdir/bar testdir/bar_link
+	 * ln testdir/foo testdir/foo_link
+	 * xfs_io -c "fsync" testdir/bar
+	 *
+	 * <power failure>
+	 *
+	 * mount fs, log replay happens
+	 *
+	 * File foo would remain with a link count of 1 when it has two entries
+	 * pointing to it in the directory testdir. This would make it impossible
+	 * to ever delete the parent directory has it would result in stale
+	 * dentries that can never be deleted.
+	 */
+	if (ret == 1 && btrfs_dir_type(eb, di) != BTRFS_FT_DIR) {
+		struct btrfs_path *fixup_path;
+		struct btrfs_key di_key;
 
-			btrfs_dir_item_key_to_cpu(eb, di, &di_key);
-			ret = link_to_fixup_dir(trans, root, fixup_path,
-						di_key.objectid);
-			if (ret)
-				break;
-		}
-		ret = 0;
+		fixup_path = btrfs_alloc_path();
+		if (!fixup_path)
+			return -ENOMEM;
+
+		btrfs_dir_item_key_to_cpu(eb, di, &di_key);
+		ret = link_to_fixup_dir(trans, root, fixup_path, di_key.objectid);
+		btrfs_free_path(fixup_path);
 	}
-	btrfs_free_path(fixup_path);
+
 	return ret;
 }
 
@@ -2206,7 +2203,7 @@ static noinline int replay_one_dir_item(struct btrfs_trans_handle *trans,
  */
 static noinline int find_dir_range(struct btrfs_root *root,
 				   struct btrfs_path *path,
-				   u64 dirid, int key_type,
+				   u64 dirid,
 				   u64 *start_ret, u64 *end_ret)
 {
 	struct btrfs_key key;
@@ -2219,7 +2216,7 @@ static noinline int find_dir_range(struct btrfs_root *root,
 		return 1;
 
 	key.objectid = dirid;
-	key.type = key_type;
+	key.type = BTRFS_DIR_LOG_INDEX_KEY;
 	key.offset = *start_ret;
 
 	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
@@ -2233,7 +2230,7 @@ static noinline int find_dir_range(struct btrfs_root *root,
 	if (ret != 0)
 		btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
 
-	if (key.type != key_type || key.objectid != dirid) {
+	if (key.type != BTRFS_DIR_LOG_INDEX_KEY || key.objectid != dirid) {
 		ret = 1;
 		goto next;
 	}
@@ -2260,7 +2257,7 @@ static noinline int find_dir_range(struct btrfs_root *root,
 
 	btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
 
-	if (key.type != key_type || key.objectid != dirid) {
+	if (key.type != BTRFS_DIR_LOG_INDEX_KEY || key.objectid != dirid) {
 		ret = 1;
 		goto out;
 	}
@@ -2291,95 +2288,82 @@ static noinline int check_item_in_log(struct btrfs_trans_handle *trans,
 	int ret;
 	struct extent_buffer *eb;
 	int slot;
-	u32 item_size;
 	struct btrfs_dir_item *di;
-	struct btrfs_dir_item *log_di;
 	int name_len;
-	unsigned long ptr;
-	unsigned long ptr_end;
 	char *name;
-	struct inode *inode;
+	struct inode *inode = NULL;
 	struct btrfs_key location;
 
-again:
+	/*
+	 * Currenly we only log dir index keys. Even if we replay a log created
+	 * by an older kernel that logged both dir index and dir item keys, all
+	 * we need to do is process the dir index keys, we (and our caller) can
+	 * safely ignore dir item keys (key type BTRFS_DIR_ITEM_KEY).
+	 */
+	ASSERT(dir_key->type == BTRFS_DIR_INDEX_KEY);
+
 	eb = path->nodes[0];
 	slot = path->slots[0];
-	item_size = btrfs_item_size_nr(eb, slot);
-	ptr = btrfs_item_ptr_offset(eb, slot);
-	ptr_end = ptr + item_size;
-	while (ptr < ptr_end) {
-		di = (struct btrfs_dir_item *)ptr;
-		name_len = btrfs_dir_name_len(eb, di);
-		name = kmalloc(name_len, GFP_NOFS);
-		if (!name) {
-			ret = -ENOMEM;
-			goto out;
-		}
-		read_extent_buffer(eb, name, (unsigned long)(di + 1),
-				  name_len);
-		log_di = NULL;
-		if (log && dir_key->type == BTRFS_DIR_ITEM_KEY) {
-			log_di = btrfs_lookup_dir_item(trans, log, log_path,
-						       dir_key->objectid,
-						       name, name_len, 0);
-		} else if (log && dir_key->type == BTRFS_DIR_INDEX_KEY) {
-			log_di = btrfs_lookup_dir_index_item(trans, log,
-						     log_path,
+	di = btrfs_item_ptr(eb, slot, struct btrfs_dir_item);
+	name_len = btrfs_dir_name_len(eb, di);
+	name = kmalloc(name_len, GFP_NOFS);
+	if (!name) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	read_extent_buffer(eb, name, (unsigned long)(di + 1), name_len);
+
+	if (log) {
+		struct btrfs_dir_item *log_di;
+
+		log_di = btrfs_lookup_dir_index_item(trans, log, log_path,
 						     dir_key->objectid,
 						     dir_key->offset,
 						     name, name_len, 0);
-		}
-		if (!log_di) {
-			btrfs_dir_item_key_to_cpu(eb, di, &location);
-			btrfs_release_path(path);
-			btrfs_release_path(log_path);
-			inode = read_one_inode(root, location.objectid);
-			if (!inode) {
-				kfree(name);
-				return -EIO;
-			}
-
-			ret = link_to_fixup_dir(trans, root,
-						path, location.objectid);
-			if (ret) {
-				kfree(name);
-				iput(inode);
-				goto out;
-			}
-
-			inc_nlink(inode);
-			ret = btrfs_unlink_inode(trans, BTRFS_I(dir),
-					BTRFS_I(inode), name, name_len);
-			if (!ret)
-				ret = btrfs_run_delayed_items(trans);
-			kfree(name);
-			iput(inode);
-			if (ret)
-				goto out;
-
-			/* there might still be more names under this key
-			 * check and repeat if required
-			 */
-			ret = btrfs_search_slot(NULL, root, dir_key, path,
-						0, 0);
-			if (ret == 0)
-				goto again;
+		if (IS_ERR(log_di)) {
+			ret = PTR_ERR(log_di);
+			goto out;
+		} else if (log_di) {
+			/* The dentry exists in the log, we have nothing to do. */
 			ret = 0;
 			goto out;
-		} else if (IS_ERR(log_di)) {
-			kfree(name);
-			return PTR_ERR(log_di);
 		}
-		btrfs_release_path(log_path);
-		kfree(name);
-
-		ptr = (unsigned long)(di + 1);
-		ptr += name_len;
 	}
-	ret = 0;
+
+	btrfs_dir_item_key_to_cpu(eb, di, &location);
+	btrfs_release_path(path);
+	btrfs_release_path(log_path);
+	inode = read_one_inode(root, location.objectid);
+	if (!inode) {
+		ret = -EIO;
+		goto out;
+	}
+
+	ret = link_to_fixup_dir(trans, root, path, location.objectid);
+	if (ret)
+		goto out;
+
+	inc_nlink(inode);
+	ret = btrfs_unlink_inode(trans, BTRFS_I(dir), BTRFS_I(inode), name,
+				 name_len);
+	if (ret)
+		goto out;
+
+	ret = btrfs_run_delayed_items(trans);
+	if (ret)
+		goto out;
+
+	/*
+	 * Unlike dir item keys, dir index keys can only have one name (entry) in
+	 * them, as there are no key collisions since each key has a unique offset
+	 * (an index number), so we're done.
+	 */
 out:
 	btrfs_release_path(path);
 	btrfs_release_path(log_path);
+	kfree(name);
+	iput(inode);
 	return ret;
 }
 
@@ -2422,7 +2406,7 @@ static int replay_xattr_deletes(struct btrfs_trans_handle *trans,
 		}
 
 		di = btrfs_item_ptr(path->nodes[0], i, struct btrfs_dir_item);
-		total_size = btrfs_item_size_nr(path->nodes[0], i);
+		total_size = btrfs_item_size(path->nodes[0], i);
 		cur = 0;
 		while (cur < total_size) {
 			u16 name_len = btrfs_dir_name_len(path->nodes[0], di);
@@ -2499,7 +2483,6 @@ static noinline int replay_dir_deletes(struct btrfs_trans_handle *trans,
 {
 	u64 range_start;
 	u64 range_end;
-	int key_type = BTRFS_DIR_LOG_ITEM_KEY;
 	int ret = 0;
 	struct btrfs_key dir_key;
 	struct btrfs_key found_key;
@@ -2507,7 +2490,7 @@ static noinline int replay_dir_deletes(struct btrfs_trans_handle *trans,
 	struct inode *dir;
 
 	dir_key.objectid = dirid;
-	dir_key.type = BTRFS_DIR_ITEM_KEY;
+	dir_key.type = BTRFS_DIR_INDEX_KEY;
 	log_path = btrfs_alloc_path();
 	if (!log_path)
 		return -ENOMEM;
@@ -2521,14 +2504,14 @@ static noinline int replay_dir_deletes(struct btrfs_trans_handle *trans,
 		btrfs_free_path(log_path);
 		return 0;
 	}
-again:
+
 	range_start = 0;
 	range_end = 0;
 	while (1) {
 		if (del_all)
 			range_end = (u64)-1;
 		else {
-			ret = find_dir_range(log, path, dirid, key_type,
+			ret = find_dir_range(log, path, dirid,
 					     &range_start, &range_end);
 			if (ret < 0)
 				goto out;
@@ -2555,8 +2538,10 @@ static noinline int replay_dir_deletes(struct btrfs_trans_handle *trans,
 			btrfs_item_key_to_cpu(path->nodes[0], &found_key,
 					      path->slots[0]);
 			if (found_key.objectid != dirid ||
-			    found_key.type != dir_key.type)
-				goto next_type;
+			    found_key.type != dir_key.type) {
+				ret = 0;
+				goto out;
+			}
 
 			if (found_key.offset > range_end)
 				break;
@@ -2575,15 +2560,7 @@ static noinline int replay_dir_deletes(struct btrfs_trans_handle *trans,
 			break;
 		range_start = range_end + 1;
 	}
-
-next_type:
 	ret = 0;
-	if (key_type == BTRFS_DIR_LOG_ITEM_KEY) {
-		key_type = BTRFS_DIR_LOG_INDEX_KEY;
-		dir_key.type = BTRFS_DIR_INDEX_KEY;
-		btrfs_release_path(path);
-		goto again;
-	}
 out:
 	btrfs_release_path(path);
 	btrfs_free_path(log_path);
@@ -2743,12 +2720,13 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb,
 						eb, i, &key);
 			if (ret)
 				break;
-		} else if (key.type == BTRFS_DIR_ITEM_KEY) {
-			ret = replay_one_dir_item(wc->trans, root, path,
-						  eb, i, &key);
-			if (ret)
-				break;
 		}
+		/*
+		 * We don't log BTRFS_DIR_ITEM_KEY keys anymore, only the
+		 * BTRFS_DIR_INDEX_KEY items which we use to derive the
+		 * BTRFS_DIR_ITEM_KEY items. If we are replaying a log from an
+		 * older kernel with such keys, ignore them.
+		 */
 	}
 	btrfs_free_path(path);
 	return ret;
@@ -3551,20 +3529,10 @@ void btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans,
 		goto out_unlock;
 	}
 
-	di = btrfs_lookup_dir_item(trans, log, path, dir_ino,
-				   name, name_len, -1);
-	if (IS_ERR(di)) {
-		err = PTR_ERR(di);
-		goto fail;
-	}
-	if (di) {
-		ret = btrfs_delete_one_dir_name(trans, log, path, di);
-		if (ret) {
-			err = ret;
-			goto fail;
-		}
-	}
-	btrfs_release_path(path);
+	/*
+	 * We only log dir index items of a directory, so we don't need to look
+	 * for dir item keys.
+	 */
 	di = btrfs_lookup_dir_index_item(trans, log, path, dir_ino,
 					 index, name, name_len, -1);
 	if (IS_ERR(di)) {
@@ -3628,7 +3596,7 @@ void btrfs_del_inode_ref_in_log(struct btrfs_trans_handle *trans,
 static noinline int insert_dir_log_key(struct btrfs_trans_handle *trans,
 				       struct btrfs_root *log,
 				       struct btrfs_path *path,
-				       int key_type, u64 dirid,
+				       u64 dirid,
 				       u64 first_offset, u64 last_offset)
 {
 	int ret;
@@ -3637,10 +3605,7 @@ static noinline int insert_dir_log_key(struct btrfs_trans_handle *trans,
 
 	key.objectid = dirid;
 	key.offset = first_offset;
-	if (key_type == BTRFS_DIR_ITEM_KEY)
-		key.type = BTRFS_DIR_LOG_ITEM_KEY;
-	else
-		key.type = BTRFS_DIR_LOG_INDEX_KEY;
+	key.type = BTRFS_DIR_LOG_INDEX_KEY;
 	ret = btrfs_insert_empty_item(trans, log, path, &key, sizeof(*item));
 	if (ret)
 		return ret;
@@ -3675,7 +3640,7 @@ static int flush_dir_items_batch(struct btrfs_trans_handle *trans,
 
 	if (count == 1) {
 		btrfs_item_key_to_cpu(src, &key, start_slot);
-		item_size = btrfs_item_size_nr(src, start_slot);
+		item_size = btrfs_item_size(src, start_slot);
 		batch.keys = &key;
 		batch.data_sizes = &item_size;
 		batch.total_data_size = item_size;
@@ -3698,7 +3663,7 @@ static int flush_dir_items_batch(struct btrfs_trans_handle *trans,
 			const int slot = start_slot + i;
 
 			btrfs_item_key_to_cpu(src, &ins_keys[i], slot);
-			ins_sizes[i] = btrfs_item_size_nr(src, slot);
+			ins_sizes[i] = btrfs_item_size(src, slot);
 			batch.total_data_size += ins_sizes[i];
 		}
 	}
@@ -3732,7 +3697,6 @@ static int process_dir_items_leaf(struct btrfs_trans_handle *trans,
 				  struct btrfs_inode *inode,
 				  struct btrfs_path *path,
 				  struct btrfs_path *dst_path,
-				  int key_type,
 				  struct btrfs_log_ctx *ctx)
 {
 	struct btrfs_root *log = inode->root->log_root;
@@ -3740,24 +3704,18 @@ static int process_dir_items_leaf(struct btrfs_trans_handle *trans,
 	const int nritems = btrfs_header_nritems(src);
 	const u64 ino = btrfs_ino(inode);
 	const bool inode_logged_before = inode_logged(trans, inode);
-	u64 last_logged_key_offset;
 	bool last_found = false;
 	int batch_start = 0;
 	int batch_size = 0;
 	int i;
 
-	if (key_type == BTRFS_DIR_ITEM_KEY)
-		last_logged_key_offset = inode->last_dir_item_offset;
-	else
-		last_logged_key_offset = inode->last_dir_index_offset;
-
 	for (i = path->slots[0]; i < nritems; i++) {
 		struct btrfs_key key;
 		int ret;
 
 		btrfs_item_key_to_cpu(src, &key, i);
 
-		if (key.objectid != ino || key.type != key_type) {
+		if (key.objectid != ino || key.type != BTRFS_DIR_INDEX_KEY) {
 			last_found = true;
 			break;
 		}
@@ -3806,7 +3764,7 @@ static int process_dir_items_leaf(struct btrfs_trans_handle *trans,
 		 * we logged is in the log tree, saving time and avoiding adding
 		 * contention on the log tree.
 		 */
-		if (key.offset > last_logged_key_offset)
+		if (key.offset > inode->last_dir_index_offset)
 			goto add_to_batch;
 		/*
 		 * Check if the key was already logged before. If not we can add
@@ -3865,7 +3823,7 @@ static int process_dir_items_leaf(struct btrfs_trans_handle *trans,
 static noinline int log_dir_items(struct btrfs_trans_handle *trans,
 			  struct btrfs_inode *inode,
 			  struct btrfs_path *path,
-			  struct btrfs_path *dst_path, int key_type,
+			  struct btrfs_path *dst_path,
 			  struct btrfs_log_ctx *ctx,
 			  u64 min_offset, u64 *last_offset_ret)
 {
@@ -3879,7 +3837,7 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans,
 	u64 ino = btrfs_ino(inode);
 
 	min_key.objectid = ino;
-	min_key.type = key_type;
+	min_key.type = BTRFS_DIR_INDEX_KEY;
 	min_key.offset = min_offset;
 
 	ret = btrfs_search_forward(root, &min_key, path, trans->transid);
@@ -3888,9 +3846,10 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans,
 	 * we didn't find anything from this transaction, see if there
 	 * is anything at all
 	 */
-	if (ret != 0 || min_key.objectid != ino || min_key.type != key_type) {
+	if (ret != 0 || min_key.objectid != ino ||
+	    min_key.type != BTRFS_DIR_INDEX_KEY) {
 		min_key.objectid = ino;
-		min_key.type = key_type;
+		min_key.type = BTRFS_DIR_INDEX_KEY;
 		min_key.offset = (u64)-1;
 		btrfs_release_path(path);
 		ret = btrfs_search_slot(NULL, root, &min_key, path, 0, 0);
@@ -3898,7 +3857,7 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans,
 			btrfs_release_path(path);
 			return ret;
 		}
-		ret = btrfs_previous_item(root, path, ino, key_type);
+		ret = btrfs_previous_item(root, path, ino, BTRFS_DIR_INDEX_KEY);
 
 		/* if ret == 0 there are items for this type,
 		 * create a range to tell us the last key of this type.
@@ -3909,18 +3868,18 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans,
 			struct btrfs_key tmp;
 			btrfs_item_key_to_cpu(path->nodes[0], &tmp,
 					      path->slots[0]);
-			if (key_type == tmp.type)
+			if (tmp.type == BTRFS_DIR_INDEX_KEY)
 				first_offset = max(min_offset, tmp.offset) + 1;
 		}
 		goto done;
 	}
 
 	/* go backward to find any previous key */
-	ret = btrfs_previous_item(root, path, ino, key_type);
+	ret = btrfs_previous_item(root, path, ino, BTRFS_DIR_INDEX_KEY);
 	if (ret == 0) {
 		struct btrfs_key tmp;
 		btrfs_item_key_to_cpu(path->nodes[0], &tmp, path->slots[0]);
-		if (key_type == tmp.type) {
+		if (tmp.type == BTRFS_DIR_INDEX_KEY) {
 			first_offset = tmp.offset;
 			ret = overwrite_item(trans, log, dst_path,
 					     path->nodes[0], path->slots[0],
@@ -3951,8 +3910,7 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans,
 	 * from our directory
 	 */
 	while (1) {
-		ret = process_dir_items_leaf(trans, inode, path, dst_path,
-					     key_type, ctx);
+		ret = process_dir_items_leaf(trans, inode, path, dst_path, ctx);
 		if (ret != 0) {
 			if (ret < 0)
 				err = ret;
@@ -3973,7 +3931,7 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans,
 			goto done;
 		}
 		btrfs_item_key_to_cpu(path->nodes[0], &min_key, path->slots[0]);
-		if (min_key.objectid != ino || min_key.type != key_type) {
+		if (min_key.objectid != ino || min_key.type != BTRFS_DIR_INDEX_KEY) {
 			last_offset = (u64)-1;
 			goto done;
 		}
@@ -4004,8 +3962,8 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans,
 		 * insert the log range keys to indicate where the log
 		 * is valid
 		 */
-		ret = insert_dir_log_key(trans, log, path, key_type,
-					 ino, first_offset, last_offset);
+		ret = insert_dir_log_key(trans, log, path, ino, first_offset,
+					 last_offset);
 		if (ret)
 			err = ret;
 	}
@@ -4033,35 +3991,28 @@ static noinline int log_directory_changes(struct btrfs_trans_handle *trans,
 	u64 min_key;
 	u64 max_key;
 	int ret;
-	int key_type = BTRFS_DIR_ITEM_KEY;
 
 	/*
 	 * If this is the first time we are being logged in the current
 	 * transaction, or we were logged before but the inode was evicted and
-	 * reloaded later, in which case its logged_trans is 0, reset the values
-	 * of the last logged key offsets. Note that we don't use the helper
+	 * reloaded later, in which case its logged_trans is 0, reset the value
+	 * of the last logged key offset. Note that we don't use the helper
 	 * function inode_logged() here - that is because the function returns
 	 * true after an inode eviction, assuming the worst case as it can not
 	 * know for sure if the inode was logged before. So we can not skip key
 	 * searches in the case the inode was evicted, because it may not have
 	 * been logged in this transaction and may have been logged in a past
-	 * transaction, so we need to reset the last dir item and index offsets
-	 * to (u64)-1.
+	 * transaction, so we need to reset the last dir index offset to (u64)-1.
 	 */
-	if (inode->logged_trans != trans->transid) {
-		inode->last_dir_item_offset = (u64)-1;
+	if (inode->logged_trans != trans->transid)
 		inode->last_dir_index_offset = (u64)-1;
-	}
-again:
+
 	min_key = 0;
 	max_key = 0;
-	if (key_type == BTRFS_DIR_ITEM_KEY)
-		ctx->last_dir_item_offset = inode->last_dir_item_offset;
-	else
-		ctx->last_dir_item_offset = inode->last_dir_index_offset;
+	ctx->last_dir_item_offset = inode->last_dir_index_offset;
 
 	while (1) {
-		ret = log_dir_items(trans, inode, path, dst_path, key_type,
+		ret = log_dir_items(trans, inode, path, dst_path,
 				ctx, min_key, &max_key);
 		if (ret)
 			return ret;
@@ -4070,13 +4021,8 @@ static noinline int log_directory_changes(struct btrfs_trans_handle *trans,
 		min_key = max_key + 1;
 	}
 
-	if (key_type == BTRFS_DIR_ITEM_KEY) {
-		inode->last_dir_item_offset = ctx->last_dir_item_offset;
-		key_type = BTRFS_DIR_INDEX_KEY;
-		goto again;
-	} else {
-		inode->last_dir_index_offset = ctx->last_dir_item_offset;
-	}
+	inode->last_dir_index_offset = ctx->last_dir_item_offset;
+
 	return 0;
 }
 
@@ -4350,7 +4296,7 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
 	batch.nr = nr;
 
 	for (i = 0; i < nr; i++) {
-		ins_sizes[i] = btrfs_item_size_nr(src, i + start_slot);
+		ins_sizes[i] = btrfs_item_size(src, i + start_slot);
 		batch.total_data_size += ins_sizes[i];
 		btrfs_item_key_to_cpu(src, ins_keys + i, i + start_slot);
 	}
@@ -4573,14 +4519,34 @@ static int log_one_extent(struct btrfs_trans_handle *trans,
 {
 	struct btrfs_drop_extents_args drop_args = { 0 };
 	struct btrfs_root *log = inode->root->log_root;
-	struct btrfs_file_extent_item *fi;
+	struct btrfs_file_extent_item fi = { 0 };
 	struct extent_buffer *leaf;
-	struct btrfs_map_token token;
 	struct btrfs_key key;
 	u64 extent_offset = em->start - em->orig_start;
 	u64 block_len;
 	int ret;
 
+	btrfs_set_stack_file_extent_generation(&fi, trans->transid);
+	if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
+		btrfs_set_stack_file_extent_type(&fi, BTRFS_FILE_EXTENT_PREALLOC);
+	else
+		btrfs_set_stack_file_extent_type(&fi, BTRFS_FILE_EXTENT_REG);
+
+	block_len = max(em->block_len, em->orig_block_len);
+	if (em->compress_type != BTRFS_COMPRESS_NONE) {
+		btrfs_set_stack_file_extent_disk_bytenr(&fi, em->block_start);
+		btrfs_set_stack_file_extent_disk_num_bytes(&fi, block_len);
+	} else if (em->block_start < EXTENT_MAP_LAST_BYTE) {
+		btrfs_set_stack_file_extent_disk_bytenr(&fi, em->block_start -
+							extent_offset);
+		btrfs_set_stack_file_extent_disk_num_bytes(&fi, block_len);
+	}
+
+	btrfs_set_stack_file_extent_offset(&fi, extent_offset);
+	btrfs_set_stack_file_extent_num_bytes(&fi, em->len);
+	btrfs_set_stack_file_extent_ram_bytes(&fi, em->ram_bytes);
+	btrfs_set_stack_file_extent_compression(&fi, em->compress_type);
+
 	ret = log_extent_csums(trans, inode, log, em, ctx);
 	if (ret)
 		return ret;
@@ -4599,7 +4565,7 @@ static int log_one_extent(struct btrfs_trans_handle *trans,
 		drop_args.start = em->start;
 		drop_args.end = em->start + em->len;
 		drop_args.replace_extent = true;
-		drop_args.extent_item_size = sizeof(*fi);
+		drop_args.extent_item_size = sizeof(fi);
 		ret = btrfs_drop_extents(trans, log, inode, &drop_args);
 		if (ret)
 			return ret;
@@ -4611,44 +4577,14 @@ static int log_one_extent(struct btrfs_trans_handle *trans,
 		key.offset = em->start;
 
 		ret = btrfs_insert_empty_item(trans, log, path, &key,
-					      sizeof(*fi));
+					      sizeof(fi));
 		if (ret)
 			return ret;
 	}
 	leaf = path->nodes[0];
-	btrfs_init_map_token(&token, leaf);
-	fi = btrfs_item_ptr(leaf, path->slots[0],
-			    struct btrfs_file_extent_item);
-
-	btrfs_set_token_file_extent_generation(&token, fi, trans->transid);
-	if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
-		btrfs_set_token_file_extent_type(&token, fi,
-						 BTRFS_FILE_EXTENT_PREALLOC);
-	else
-		btrfs_set_token_file_extent_type(&token, fi,
-						 BTRFS_FILE_EXTENT_REG);
-
-	block_len = max(em->block_len, em->orig_block_len);
-	if (em->compress_type != BTRFS_COMPRESS_NONE) {
-		btrfs_set_token_file_extent_disk_bytenr(&token, fi,
-							em->block_start);
-		btrfs_set_token_file_extent_disk_num_bytes(&token, fi, block_len);
-	} else if (em->block_start < EXTENT_MAP_LAST_BYTE) {
-		btrfs_set_token_file_extent_disk_bytenr(&token, fi,
-							em->block_start -
-							extent_offset);
-		btrfs_set_token_file_extent_disk_num_bytes(&token, fi, block_len);
-	} else {
-		btrfs_set_token_file_extent_disk_bytenr(&token, fi, 0);
-		btrfs_set_token_file_extent_disk_num_bytes(&token, fi, 0);
-	}
-
-	btrfs_set_token_file_extent_offset(&token, fi, extent_offset);
-	btrfs_set_token_file_extent_num_bytes(&token, fi, em->len);
-	btrfs_set_token_file_extent_ram_bytes(&token, fi, em->ram_bytes);
-	btrfs_set_token_file_extent_compression(&token, fi, em->compress_type);
-	btrfs_set_token_file_extent_encryption(&token, fi, 0);
-	btrfs_set_token_file_extent_other_encoding(&token, fi, 0);
+	write_extent_buffer(leaf, &fi,
+			    btrfs_item_ptr_offset(leaf, path->slots[0]),
+			    sizeof(fi));
 	btrfs_mark_buffer_dirty(leaf);
 
 	btrfs_release_path(path);
@@ -4862,7 +4798,6 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
 	WARN_ON(!list_empty(&extents));
 	write_unlock(&tree->lock);
 
-	btrfs_release_path(path);
 	if (!ret)
 		ret = btrfs_log_prealloc_extents(trans, inode, path);
 	if (ret)
@@ -5166,7 +5101,7 @@ static int btrfs_check_ref_name_override(struct extent_buffer *eb,
 	struct btrfs_path *search_path;
 	char *name = NULL;
 	u32 name_len = 0;
-	u32 item_size = btrfs_item_size_nr(eb, slot);
+	u32 item_size = btrfs_item_size(eb, slot);
 	u32 cur_offset = 0;
 	unsigned long ptr = btrfs_item_ptr_offset(eb, slot);
 
@@ -5899,18 +5834,12 @@ struct btrfs_dir_list {
  *    link_to_fixup_dir());
  *
  * 2) For directories we log with a mode of LOG_INODE_ALL. It's possible that
- *    while logging the inode's items new items with keys BTRFS_DIR_ITEM_KEY and
- *    BTRFS_DIR_INDEX_KEY are added to fs/subvol tree and the logged inode item
+ *    while logging the inode's items new index items (key type
+ *    BTRFS_DIR_INDEX_KEY) are added to fs/subvol tree and the logged inode item
  *    has a size that doesn't match the sum of the lengths of all the logged
- *    names. This does not result in a problem because if a dir_item key is
- *    logged but its matching dir_index key is not logged, at log replay time we
- *    don't use it to replay the respective name (see replay_one_name()). On the
- *    other hand if only the dir_index key ends up being logged, the respective
- *    name is added to the fs/subvol tree with both the dir_item and dir_index
- *    keys created (see replay_one_name()).
- *    The directory's inode item with a wrong i_size is not a problem as well,
- *    since we don't use it at log replay time to set the i_size in the inode
- *    item of the fs/subvol tree (see overwrite_item()).
+ *    names - this is ok, not a problem, because at log replay time we set the
+ *    directory's i_size to the correct value (see replay_one_name() and
+ *    do_overwrite_item()).
  */
 static int log_new_dir_dentries(struct btrfs_trans_handle *trans,
 				struct btrfs_root *root,
@@ -5956,7 +5885,7 @@ static int log_new_dir_dentries(struct btrfs_trans_handle *trans,
 			goto next_dir_inode;
 
 		min_key.objectid = dir_elem->ino;
-		min_key.type = BTRFS_DIR_ITEM_KEY;
+		min_key.type = BTRFS_DIR_INDEX_KEY;
 		min_key.offset = 0;
 again:
 		btrfs_release_path(path);
@@ -5981,7 +5910,7 @@ static int log_new_dir_dentries(struct btrfs_trans_handle *trans,
 
 			btrfs_item_key_to_cpu(leaf, &min_key, i);
 			if (min_key.objectid != dir_elem->ino ||
-			    min_key.type != BTRFS_DIR_ITEM_KEY)
+			    min_key.type != BTRFS_DIR_INDEX_KEY)
 				goto next_dir_inode;
 
 			di = btrfs_item_ptr(leaf, i, struct btrfs_dir_item);
@@ -6093,7 +6022,7 @@ static int btrfs_log_all_parents(struct btrfs_trans_handle *trans,
 		if (key.objectid != ino || key.type > BTRFS_INODE_EXTREF_KEY)
 			break;
 
-		item_size = btrfs_item_size_nr(leaf, slot);
+		item_size = btrfs_item_size(leaf, slot);
 		ptr = btrfs_item_ptr_offset(leaf, slot);
 		while (cur_offset < item_size) {
 			struct btrfs_key inode_key;
@@ -6795,15 +6724,14 @@ void btrfs_log_new_name(struct btrfs_trans_handle *trans,
 	 * was previously logged, make sure the next log attempt on the directory
 	 * is not skipped and logs the inode again. This is because the log may
 	 * not currently be authoritative for a range including the old
-	 * BTRFS_DIR_ITEM_KEY and BTRFS_DIR_INDEX_KEY keys, so we want to make
-	 * sure after a log replay we do not end up with both the new and old
-	 * dentries around (in case the inode is a directory we would have a
-	 * directory with two hard links and 2 inode references for different
-	 * parents). The next log attempt of old_dir will happen at
-	 * btrfs_log_all_parents(), called through btrfs_log_inode_parent()
-	 * below, because we have previously set inode->last_unlink_trans to the
-	 * current transaction ID, either here or at btrfs_record_unlink_dir() in
-	 * case inode is a directory.
+	 * BTRFS_DIR_INDEX_KEY key, so we want to make sure after a log replay we
+	 * do not end up with both the new and old dentries around (in case the
+	 * inode is a directory we would have a directory with two hard links and
+	 * 2 inode references for different parents). The next log attempt of
+	 * old_dir will happen at btrfs_log_all_parents(), called through
+	 * btrfs_log_inode_parent() below, because we have previously set
+	 * inode->last_unlink_trans to the current transaction ID, either here or
+	 * at btrfs_record_unlink_dir() in case the inode is a directory.
 	 */
 	if (old_dir)
 		old_dir->logged_trans = 0;
diff --git a/fs/btrfs/uuid-tree.c b/fs/btrfs/uuid-tree.c
index 74023c8a783f..b458452a1aaf 100644
--- a/fs/btrfs/uuid-tree.c
+++ b/fs/btrfs/uuid-tree.c
@@ -52,7 +52,7 @@ static int btrfs_uuid_tree_lookup(struct btrfs_root *uuid_root, u8 *uuid,
 
 	eb = path->nodes[0];
 	slot = path->slots[0];
-	item_size = btrfs_item_size_nr(eb, slot);
+	item_size = btrfs_item_size(eb, slot);
 	offset = btrfs_item_ptr_offset(eb, slot);
 	ret = -ENOENT;
 
@@ -125,7 +125,7 @@ int btrfs_uuid_tree_add(struct btrfs_trans_handle *trans, u8 *uuid, u8 type,
 		eb = path->nodes[0];
 		slot = path->slots[0];
 		offset = btrfs_item_ptr_offset(eb, slot);
-		offset += btrfs_item_size_nr(eb, slot) - sizeof(subid_le);
+		offset += btrfs_item_size(eb, slot) - sizeof(subid_le);
 	} else {
 		btrfs_warn(fs_info,
 			   "insert uuid item failed %d (0x%016llx, 0x%016llx) type %u!",
@@ -186,7 +186,7 @@ int btrfs_uuid_tree_remove(struct btrfs_trans_handle *trans, u8 *uuid, u8 type,
 	eb = path->nodes[0];
 	slot = path->slots[0];
 	offset = btrfs_item_ptr_offset(eb, slot);
-	item_size = btrfs_item_size_nr(eb, slot);
+	item_size = btrfs_item_size(eb, slot);
 	if (!IS_ALIGNED(item_size, sizeof(u64))) {
 		btrfs_warn(fs_info, "uuid item with illegal size %lu!",
 			   (unsigned long)item_size);
@@ -208,7 +208,7 @@ int btrfs_uuid_tree_remove(struct btrfs_trans_handle *trans, u8 *uuid, u8 type,
 		goto out;
 	}
 
-	item_size = btrfs_item_size_nr(eb, slot);
+	item_size = btrfs_item_size(eb, slot);
 	if (item_size == sizeof(subid)) {
 		ret = btrfs_del_item(trans, uuid_root, path);
 		goto out;
@@ -331,7 +331,7 @@ int btrfs_uuid_tree_iterate(struct btrfs_fs_info *fs_info)
 			goto skip;
 
 		offset = btrfs_item_ptr_offset(leaf, slot);
-		item_size = btrfs_item_size_nr(leaf, slot);
+		item_size = btrfs_item_size(leaf, slot);
 		if (!IS_ALIGNED(item_size, sizeof(u64))) {
 			btrfs_warn(fs_info,
 				   "uuid item with illegal size %lu!",
diff --git a/fs/btrfs/verity.c b/fs/btrfs/verity.c
index 4968535dfff0..90eb5c2830a9 100644
--- a/fs/btrfs/verity.c
+++ b/fs/btrfs/verity.c
@@ -333,7 +333,7 @@ static int read_key_bytes(struct btrfs_inode *inode, u8 key_type, u64 offset,
 		if (key.objectid != btrfs_ino(inode) || key.type != key_type)
 			break;
 
-		item_end = btrfs_item_size_nr(leaf, path->slots[0]) + key.offset;
+		item_end = btrfs_item_size(leaf, path->slots[0]) + key.offset;
 
 		if (copied > 0) {
 			/*
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 42391d4aeb11..5f4ac1a2e1f3 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -34,6 +34,10 @@
 #include "discard.h"
 #include "zoned.h"
 
+#define BTRFS_BLOCK_GROUP_STRIPE_MASK	(BTRFS_BLOCK_GROUP_RAID0 | \
+					 BTRFS_BLOCK_GROUP_RAID10 | \
+					 BTRFS_BLOCK_GROUP_RAID56_MASK)
+
 const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
 	[BTRFS_RAID_RAID10] = {
 		.sub_stripes	= 2,
@@ -4643,7 +4647,7 @@ int btrfs_uuid_scan_kthread(void *data)
 
 		eb = path->nodes[0];
 		slot = path->slots[0];
-		item_size = btrfs_item_size_nr(eb, slot);
+		item_size = btrfs_item_size(eb, slot);
 		if (item_size < sizeof(root_item))
 			goto skip;
 
@@ -6314,7 +6318,8 @@ int btrfs_get_io_geometry(struct btrfs_fs_info *fs_info, struct extent_map *em,
 	stripe_offset = offset - stripe_offset;
 	data_stripes = nr_data_stripes(map);
 
-	if (map->type & BTRFS_BLOCK_GROUP_PROFILE_MASK) {
+	/* Only stripe based profiles needs to check against stripe length. */
+	if (map->type & BTRFS_BLOCK_GROUP_STRIPE_MASK) {
 		u64 max_len = stripe_len - stripe_offset;
 
 		/*
@@ -7730,7 +7735,7 @@ static int btrfs_device_init_dev_stats(struct btrfs_device *device,
 	}
 	slot = path->slots[0];
 	eb = path->nodes[0];
-	item_size = btrfs_item_size_nr(eb, slot);
+	item_size = btrfs_item_size(eb, slot);
 
 	ptr = btrfs_item_ptr(eb, slot, struct btrfs_dev_stats_item);
 
@@ -7808,7 +7813,7 @@ static int update_dev_stat_item(struct btrfs_trans_handle *trans,
 	}
 
 	if (ret == 0 &&
-	    btrfs_item_size_nr(path->nodes[0], path->slots[0]) < sizeof(*ptr)) {
+	    btrfs_item_size(path->nodes[0], path->slots[0]) < sizeof(*ptr)) {
 		/* need to delete old one and insert a new one */
 		ret = btrfs_del_item(trans, dev_root, path);
 		if (ret != 0) {
diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c
index 2837b4c8424d..99abf41b89b9 100644
--- a/fs/btrfs/xattr.c
+++ b/fs/btrfs/xattr.c
@@ -168,9 +168,8 @@ int btrfs_setxattr(struct btrfs_trans_handle *trans, struct inode *inode,
 		const int slot = path->slots[0];
 		struct extent_buffer *leaf = path->nodes[0];
 		const u16 old_data_len = btrfs_dir_data_len(leaf, di);
-		const u32 item_size = btrfs_item_size_nr(leaf, slot);
+		const u32 item_size = btrfs_item_size(leaf, slot);
 		const u32 data_size = sizeof(*di) + name_len + size;
-		struct btrfs_item *item;
 		unsigned long data_ptr;
 		char *ptr;
 
@@ -196,9 +195,8 @@ int btrfs_setxattr(struct btrfs_trans_handle *trans, struct inode *inode,
 			btrfs_extend_item(path, data_size);
 		}
 
-		item = btrfs_item_nr(slot);
 		ptr = btrfs_item_ptr(leaf, slot, char);
-		ptr += btrfs_item_size(leaf, item) - data_size;
+		ptr += btrfs_item_size(leaf, slot) - data_size;
 		di = (struct btrfs_dir_item *)ptr;
 		btrfs_set_dir_data_len(leaf, di, size);
 		data_ptr = ((unsigned long)(di + 1)) + name_len;
@@ -335,7 +333,7 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size)
 			goto next_item;
 
 		di = btrfs_item_ptr(leaf, slot, struct btrfs_dir_item);
-		item_size = btrfs_item_size_nr(leaf, slot);
+		item_size = btrfs_item_size(leaf, slot);
 		cur = 0;
 		while (cur < item_size) {
 			u16 name_len = btrfs_dir_name_len(leaf, di);
diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h
index 738619994e26..012a71ab5d8e 100644
--- a/include/uapi/linux/btrfs.h
+++ b/include/uapi/linux/btrfs.h
@@ -575,8 +575,10 @@ struct btrfs_ioctl_clone_range_args {
  * Used by:
  * struct btrfs_ioctl_defrag_range_args.flags
  */
-#define BTRFS_DEFRAG_RANGE_COMPRESS 1
-#define BTRFS_DEFRAG_RANGE_START_IO 2
+#define BTRFS_DEFRAG_RANGE_COMPRESS	(1UL << 0)
+#define BTRFS_DEFRAG_RANGE_START_IO	(1UL << 1)
+#define BTRFS_DEFRAG_RANGE_FLAGS_MASK	(BTRFS_DEFRAG_RANGE_COMPRESS |\
+					 BTRFS_DEFRAG_RANGE_START_IO)
 struct btrfs_ioctl_defrag_range_args {
 	/* start of the defrag operation */
 	__u64 start;
diff --git a/include/uapi/linux/btrfs_tree.h b/include/uapi/linux/btrfs_tree.h
index e1c4c732aaba..5416f1f1a77a 100644
--- a/include/uapi/linux/btrfs_tree.h
+++ b/include/uapi/linux/btrfs_tree.h
@@ -146,7 +146,9 @@
 
 /*
  * dir items are the name -> inode pointers in a directory.  There is one
- * for every name in a directory.
+ * for every name in a directory.  BTRFS_DIR_LOG_ITEM_KEY is no longer used
+ * but it's still defined here for documentation purposes and to help avoid
+ * having its numerical value reused in the future.
  */
 #define BTRFS_DIR_LOG_ITEM_KEY  60
 #define BTRFS_DIR_LOG_INDEX_KEY 72
-- 
2.35.1