WSL2-Linux-Kernel/fs/btrfs
Filipe Manana 259c4b96d7 btrfs: stop doing unnecessary log updates during a rename
During a rename, we call __btrfs_unlink_inode(), which will call
btrfs_del_inode_ref_in_log() and btrfs_del_dir_entries_in_log(), in order
to remove an inode reference and a directory entry from the log. These
are necessary when __btrfs_unlink_inode() is called from the unlink path,
but not necessary when it's called from a rename context, because:

1) For the btrfs_del_inode_ref_in_log() call, it's pointless to delete the
   inode reference related to the old name, because later in the rename
   path we call btrfs_log_new_name(), which will drop all inode references
   from the log and copy all inode references from the subvolume tree to
   the log tree. So we are doing one unnecessary btree operation which
   adds additional latency and lock contention in case there are other
   tasks accessing the log tree;

2) For the btrfs_del_dir_entries_in_log() call, we are now doing the
   equivalent at btrfs_log_new_name() since the previous patch in the
   series, that has the subject "btrfs: avoid logging all directory
   changes during renames". In fact, having __btrfs_unlink_inode() call
   this function not only adds additional latency and lock contention due
   to the extra btree operation, but also can make btrfs_log_new_name()
   unnecessarily log a range item to track the deletion of the old name,
   since it has no way to known that the directory entry related to the
   old name was previously logged and already deleted by
   __btrfs_unlink_inode() through its call to
   btrfs_del_dir_entries_in_log().

So skip those calls at __btrfs_unlink_inode() when we are doing a rename.
Skipping them also allows us now to reduce the duration of time we are
pinning a log transaction during renames, which is always beneficial as
it's not delaying so much other tasks trying to sync the log tree, in
particular we end up not holding the log transaction pinned while adding
the new name (adding inode ref, directory entry, etc).

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

  1/5 btrfs: add helper to delete a dir entry from a log tree
  2/5 btrfs: pass the dentry to btrfs_log_new_name() instead of the inode
  3/5 btrfs: avoid logging all directory changes during renames
  4/5 btrfs: stop doing unnecessary log updates during a rename
  5/5 btrfs: avoid inode logging during rename and link when possible

Just like the previous patch in the series, "btrfs: avoid logging all
directory changes during renames", the following script mimics part of
what a package installation/upgrade with zypper does, which is basically
renaming a lot of files, in some directory under /usr, to a name with a
suffix of "-RPMDELETE":

  $ cat test.sh
  #!/bin/bash

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

  NUM_FILES=10000

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

  mkdir $MNT/testdir

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

  sync

  # Do some change to testdir and fsync it.
  echo -n > $MNT/testdir/file_$((NUM_FILES + 1))
  xfs_io -c "fsync" $MNT/testdir

  echo "Renaming $NUM_FILES files..."
  start=$(date +%s%N)
  for ((i = 1; i <= $NUM_FILES; i++)); do
      mv $MNT/testdir/file_$i $MNT/testdir/file_$i-RPMDELETE
  done
  end=$(date +%s%N)

  dur=$(( (end - start) / 1000000 ))
  echo "Renames took $dur milliseconds"

  umount $MNT

Testing this change on box a using a non-debug kernel (Debian's default
kernel config) gave the following results:

NUM_FILES=10000, before patchset:                   27399 ms
NUM_FILES=10000, after patches 1/5 to 3/5 applied:   9093 ms (-66.8%)
NUM_FILES=10000, after patches 1/5 to 4/5 applied:   9016 ms (-67.1%)

NUM_FILES=5000, before patchset:                     9241 ms
NUM_FILES=5000, after patches 1/5 to 3/5 applied:    4642 ms (-49.8%)
NUM_FILES=5000, after patches 1/5 to 4/5 applied:    4553 ms (-50.7%)

NUM_FILES=2000, before patchset:                     2550 ms
NUM_FILES=2000, after patches 1/5 to 3/5 applied:    1788 ms (-29.9%)
NUM_FILES=2000, after patches 1/5 to 4/5 applied:    1767 ms (-30.7%)

NUM_FILES=1000, before patchset:                     1088 ms
NUM_FILES=1000, after patches 1/5 to 3/5 applied:     905 ms (-16.9%)
NUM_FILES=1000, after patches 1/5 to 4/5 applied:     883 ms (-18.8%)

The next patch in the series (5/5), also contains dbench results after
applying to whole patchset.

Link: https://bugzilla.opensuse.org/show_bug.cgi?id=1193549
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
2022-03-14 13:13:47 +01:00
..
tests btrfs: selftests: dump extent io tree if extent-io-tree test failed 2022-01-07 14:18:27 +01:00
Kconfig btrfs: use generic Kconfig option for 256kB page size limit 2022-01-20 08:52:55 +02:00
Makefile btrfs: remove reada infrastructure 2022-01-07 14:18:26 +01:00
acl.c
async-thread.c btrfs: fix memory ordering between normal and ordered work functions 2021-11-16 16:50:23 +01:00
async-thread.h
backref.c btrfs: stop accessing ->extent_root directly 2022-01-03 15:09:49 +01:00
backref.h
block-group.c btrfs: skip reserved bytes warning on unmount after log cleanup failure 2022-01-31 16:06:50 +01:00
block-group.h btrfs: fix deadlock between chunk allocation and chunk btree modifications 2021-10-26 19:08:07 +02:00
block-rsv.c btrfs: reserve extra space for the free space tree 2022-01-07 14:18:25 +01:00
block-rsv.h btrfs: init root block_rsv at init root time 2022-01-03 15:09:48 +01:00
btrfs_inode.h btrfs: put initial index value of a directory in a constant 2022-03-14 13:13:46 +01:00
check-integrity.c btrfs: check-integrity: stop storing the block device name in btrfsic_dev_state 2021-10-26 19:08:07 +02:00
check-integrity.h
compression.c btrfs: remove unnecessary parameter type from compression_decompress_bio 2022-01-07 14:18:27 +01:00
compression.h btrfs: determine stripe boundary at bio allocation time in btrfs_submit_compressed_write 2021-10-26 19:08:04 +02:00
ctree.c btrfs: refactor unlock_up 2022-01-07 14:18:26 +01:00
ctree.h btrfs: do not start relocation until in progress drops are done 2022-03-02 16:52:39 +01:00
delalloc-space.c btrfs: change root to fs_info for btrfs_reserve_metadata_bytes 2022-01-03 15:09:45 +01:00
delalloc-space.h
delayed-inode.c btrfs: add an inode-item.h 2022-01-07 14:18:23 +01:00
delayed-inode.h
delayed-ref.c btrfs: reserve extra space for the free space tree 2022-01-07 14:18:25 +01:00
delayed-ref.h btrfs: make btrfs_ref::real_root optional 2021-10-26 19:08:06 +02:00
dev-replace.c btrfs: add device major-minor info in the struct btrfs_device 2022-03-14 13:13:47 +01:00
dev-replace.h
dir-item.c btrfs: drop the _nr from the item helpers 2022-01-03 15:09:43 +01:00
discard.c
discard.h
disk-io.c btrfs: do not start relocation until in progress drops are done 2022-03-02 16:52:39 +01:00
disk-io.h btrfs: track the csum, extent, and free space trees in a rb tree 2022-01-03 15:09:50 +01:00
export.c
export.h
extent-io-tree.h
extent-tree.c btrfs: zoned: remove redundant initialization of to_add 2022-03-14 13:13:47 +01:00
extent_io.c for-5.17-rc6-tag 2022-03-06 12:19:36 -08:00
extent_io.h btrfs: cleanup for extent_write_locked_range() 2021-10-26 19:08:04 +02:00
extent_map.c btrfs: defrag: don't use merged extent map for their generation check 2022-02-23 17:43:13 +01:00
extent_map.h btrfs: defrag: don't use merged extent map for their generation check 2022-02-23 17:43:13 +01:00
file-item.c btrfs: stop accessing ->csum_root directly 2022-01-03 15:09:49 +01:00
file.c btrfs: reduce extent threshold for autodefrag 2022-02-24 16:11:28 +01:00
free-space-cache.c btrfs: add inode to truncate control 2022-01-07 14:18:24 +01:00
free-space-cache.h btrfs: change name and type of private member of btrfs_free_space_ctl 2022-01-03 15:09:50 +01:00
free-space-tree.c btrfs: track the csum, extent, and free space trees in a rb tree 2022-01-03 15:09:50 +01:00
free-space-tree.h
inode-item.c btrfs: make should_throttle loop local in btrfs_truncate_inode_items 2022-01-07 14:18:25 +01:00
inode-item.h btrfs: add inode to truncate control 2022-01-07 14:18:24 +01:00
inode.c btrfs: stop doing unnecessary log updates during a rename 2022-03-14 13:13:47 +01:00
ioctl.c btrfs: reuse existing inode from btrfs_ioctl 2022-03-14 13:13:46 +01:00
locking.c
locking.h
lzo.c btrfs: prevent copying too big compressed lzo segment 2022-02-15 19:59:09 +01:00
misc.h
ordered-data.c
ordered-data.h
orphan.c
print-tree.c btrfs: drop the _nr from the item helpers 2022-01-03 15:09:43 +01:00
print-tree.h
props.c btrfs: change root to fs_info for btrfs_reserve_metadata_bytes 2022-01-03 15:09:45 +01:00
props.h
qgroup.c btrfs: move QUOTA_ENABLED check to rescan_should_stop from btrfs_qgroup_rescan_worker 2022-03-14 13:13:47 +01:00
qgroup.h
raid56.c btrfs: remove btrfs_raid_bio::fs_info member 2021-10-26 19:08:03 +02:00
raid56.h btrfs: remove btrfs_raid_bio::fs_info member 2021-10-26 19:08:03 +02:00
rcu-string.h
ref-verify.c btrfs: stop accessing ->extent_root directly 2022-01-03 15:09:49 +01:00
ref-verify.h
reflink.c btrfs: drop the _nr from the item helpers 2022-01-03 15:09:43 +01:00
reflink.h
relocation.c btrfs: do not start relocation until in progress drops are done 2022-03-02 16:52:39 +01:00
root-tree.c btrfs: do not start relocation until in progress drops are done 2022-03-02 16:52:39 +01:00
scrub.c btrfs: scrub: remove redundant initialization of increment 2022-03-14 13:13:47 +01:00
send.c btrfs: send: remove redundant ret variable in fs_path_copy 2022-03-14 13:13:47 +01:00
send.h btrfs: reuse existing inode from btrfs_ioctl 2022-03-14 13:13:46 +01:00
space-info.c btrfs: fix argument list that the kdoc format and script verified 2022-01-07 14:18:27 +01:00
space-info.h btrfs: change root to fs_info for btrfs_reserve_metadata_bytes 2022-01-03 15:09:45 +01:00
struct-funcs.c
subpage.c btrfs: subpage: fix a wrong check on subpage->writers 2022-03-02 16:51:39 +01:00
subpage.h btrfs: rework page locking in __extent_writepage() 2021-10-26 19:08:05 +02:00
super.c btrfs: match stale devices by dev_t 2022-03-14 13:13:47 +01:00
sysfs.c btrfs: sysfs: add devinfo/fsid to retrieve actual fsid from the device 2022-01-07 14:18:25 +01:00
sysfs.h
transaction.c btrfs: fix relocation crash due to premature return from btrfs_commit_transaction() 2022-03-02 16:52:46 +01:00
transaction.h btrfs: do not start relocation until in progress drops are done 2022-03-02 16:52:39 +01:00
tree-checker.c btrfs: tree-checker: use u64 for item data end to avoid overflow 2022-03-02 16:52:32 +01:00
tree-checker.h
tree-defrag.c btrfs: remove unnecessary extent root check in btrfs_defrag_leaves 2022-01-03 15:09:48 +01:00
tree-log.c btrfs: stop doing unnecessary log updates during a rename 2022-03-14 13:13:47 +01:00
tree-log.h btrfs: avoid logging all directory changes during renames 2022-03-14 13:13:47 +01:00
tree-mod-log.c
tree-mod-log.h
ulist.c
ulist.h
uuid-tree.c btrfs: drop the _nr from the item helpers 2022-01-03 15:09:43 +01:00
verity.c btrfs: drop the _nr from the item helpers 2022-01-03 15:09:43 +01:00
volumes.c btrfs: cleanup temporary variables when finding rotational device status 2022-03-14 13:13:47 +01:00
volumes.h btrfs: add device major-minor info in the struct btrfs_device 2022-03-14 13:13:47 +01:00
xattr.c btrfs: drop the _nr from the item helpers 2022-01-03 15:09:43 +01:00
xattr.h
zlib.c Revert "btrfs: compression: drop kmap/kunmap from zlib" 2021-10-29 13:03:05 +02:00
zoned.c btrfs: zoned: fix chunk allocation condition for zoned allocator 2022-01-07 14:18:26 +01:00
zoned.h btrfs: zoned: fix chunk allocation condition for zoned allocator 2022-01-07 14:18:26 +01:00
zstd.c lib: zstd: Add kernel-specific API 2021-11-08 16:55:21 -08:00