This series adds READ_ONCE() for existing lockless reads of
jbd2_inode fields in jbd2 and filesystem callbacks used by ext4 and ocfs2.

This is based on Jan's suggestion in the review of the ext4 jinode
publication race fix. [1]

[1]: https://lore.kernel.org/all/4jxwogttddiaoqbstlgou5ox6zs27ngjjz5ukrxafm2z5ijxod@so4eqnykiegj/

Thanks,
Li

Li Chen (3):
  jbd2: use READ_ONCE for lockless jinode reads
  ext4: use READ_ONCE for lockless jinode reads
  ocfs2: use READ_ONCE for lockless jinode reads

 fs/ext4/inode.c       |  6 ++++--
 fs/ext4/super.c       | 13 ++++++++-----
 fs/jbd2/commit.c      | 39 ++++++++++++++++++++++++++++++++-------
 fs/jbd2/transaction.c |  2 +-
 fs/ocfs2/journal.c    |  7 +++++--
 5 files changed, 50 insertions(+), 17 deletions(-)

-- 
2.52.0