Hi,
This series originally aim to fix ext4_handle_error() and ext4_abort()
cannot panic because of we invoke __jbd2_journal_abort_hard() when we
failed to submit commit record without setting JBD2_REC_ERR flag.
I add patch 1 and patch 4 to switch to use jbd2_journal_abort() and do
some cleanup job at this iteration as Jan suggested. I also add patch 3
to fix missing updating ESHUTDOWN problem in commit 818d276ceb8 "ext4:
Add the journal checksum feature", please revirew this series and give
some suggestions.
Changes since v2:
- Fix spelling mistakes in the first patch.
- Keep JBD2_REC_ERR and remove the last place that invoke
jbd2_journal_abort() with 0 errno and the corresponding logic in
__journal_abort_soft().
- Fix missing updating errno in the jbd2 sb after jbd2 shutdown abort.
Thanks,
Yi.
zhangyi (F) (4):
jbd2: switch to use jbd2_journal_abort() when failed to submit the
commit record
ext4, jbd2: ensure panic when aborting with zero errno
jbd2: make sure ESHUTDOWN to be recorded in the journal superblock
jbd2: clean __jbd2_journal_abort_hard() and __journal_abort_soft()
fs/jbd2/checkpoint.c | 2 +-
fs/jbd2/commit.c | 4 +-
fs/jbd2/journal.c | 111 ++++++++++++++++---------------------------
include/linux/jbd2.h | 1 -
4 files changed, 45 insertions(+), 73 deletions(-)
--
2.17.2
