releases/4.14.60/f2fs-fix-to-don-t-trigger-writeback-during-recovery.patch

   1 From foo@baz Sat Jul 28 10:25:26 CEST 2018
   2 From: Chao Yu <yuchao0@huawei.com>
   3 Date: Sat, 26 May 2018 18:03:34 +0800
   4 Subject: f2fs: fix to don't trigger writeback during recovery
   5
   6 From: Chao Yu <yuchao0@huawei.com>
   7
   8 [ Upstream commit 64c74a7ab505ea40d1b3e5d02735ecab08ae1b14 ]
   9
  10 - f2fs_fill_super
  11  - recover_fsync_data
  12   - recover_data
  13    - del_fsync_inode
  14     - iput
  15      - iput_final
  16       - write_inode_now
  17        - f2fs_write_inode
  18         - f2fs_balance_fs
  19          - f2fs_balance_fs_bg
  20           - sync_dirty_inodes
  21
  22 With data_flush mount option, during recovery, in order to avoid entering
  23 above writeback flow, let's detect recovery status and do skip in
  24 f2fs_balance_fs_bg.
  25
  26 Signed-off-by: Chao Yu <yuchao0@huawei.com>
  27 Signed-off-by: Yunlei He <heyunlei@huawei.com>
  28 Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
  29 Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
  30 Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
  31 ---
  32  fs/f2fs/segment.c |    3 +++
  33  1 file changed, 3 insertions(+)
  34
  35 --- a/fs/f2fs/segment.c
  36 +++ b/fs/f2fs/segment.c
  37 @@ -435,6 +435,9 @@ void f2fs_balance_fs(struct f2fs_sb_info
  38
  39  void f2fs_balance_fs_bg(struct f2fs_sb_info *sbi)
  40  {
  41 +       if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
  42 +               return;
  43 +
  44         /* try to shrink extent cache when there is no enough memory */
  45         if (!available_free_memory(sbi, EXTENT_CACHE))
  46                 f2fs_shrink_extent_tree(sbi, EXTENT_CACHE_SHRINK_NUMBER);