[GIT PULL] More bcachefs fixes for 6.7

[Kent Overstreet]

Hi Linus, more bcachefs fixes for 6.7. Nothing terribly exciting, the
bug squashing continues...

Cheers,
Kent

----------------------------------------------------------------

The following changes since commit 33cc938e65a98f1d29d0a18403dbbee050dcad9a:

  Linux 6.7-rc4 (2023-12-03 18:52:56 +0900)

are available in the Git repository at:

  https://evilpiepirate.org/git/bcachefs.git tags/bcachefs-2023-12-10

for you to fetch changes up to a66ff26b0f31189e413a87065c25949c359e4bef:

  bcachefs: Close journal entry if necessary when flushing all pins (2023-12-10 16:53:46 -0500)

----------------------------------------------------------------
More bcachefs bugfixes for 6.7:

 - Fix a rare emergency shutdown path bug: dropping journal pins after
   the filesystem has mostly been torn down is not what we want.
 - Fix some concurrency issues with the btree write buffer and journal
   replay by not using the btree write buffer until journal replay is
   finished
 - A fixup from the prior patch to kill journal pre-reservations: at the
   start of the btree update path, where previously we took a
   pre-reservation, we do at least want to check the journal watermark.
 - Fix a race between dropping device metadata and btree node writes,
   which would re-add a pointer to a device that had just been dropped
 - Fix one of the SCRU lock warnings, in
   bch2_compression_stats_to_text().
 - Partial fix for a rare transaction paths overflow, when indirect
   extents had been split by background tasks, by not running certain
   triggers when they're not needed.
 - Fix for creating a snapshot with implicit source in a subdirectory of
   the containing subvolume
 - Don't unfreeze when we're emergency read-only
 - Fix for rebalance spinning trying to compress unwritten extentns
 - Another deleted_inodes fix, for directories
 - Fix a rare deadlock (usually just an unecessary wait) when flushing
   the journal with an open journal entry.

----------------------------------------------------------------
Brian Foster (1):
      bcachefs: don't attempt rw on unfreeze when shutdown

Daniel Hill (1):
      bcachefs: rebalance shouldn't attempt to compress unwritten extents

Kent Overstreet (10):
      bcachefs: Don't drop journal pins in exit path
      bcachefs; Don't use btree write buffer until journal replay is finished
      bcachefs: Fix a journal deadlock in replay
      bcachefs: Fix bch2_extent_drop_ptrs() call
      bcachefs: Convert compression_stats to for_each_btree_key2
      bcachefs: Don't run indirect extent trigger unless inserting/deleting
      bcachefs: Fix creating snapshot with implict source
      bcachefs: Fix deleted inode check for dirs
      bcachefs: Fix uninitialized var in bch2_journal_replay()
      bcachefs: Close journal entry if necessary when flushing all pins

 fs/bcachefs/btree_cache.c           |  8 +++-----
 fs/bcachefs/btree_io.c              |  4 ++--
 fs/bcachefs/btree_io.h              |  3 ---
 fs/bcachefs/btree_key_cache.c       |  2 --
 fs/bcachefs/btree_update.c          | 16 ++++++++++++++++
 fs/bcachefs/btree_update_interior.c | 11 +++++++++++
 fs/bcachefs/data_update.c           |  4 ++--
 fs/bcachefs/dirent.c                | 19 +++++++++++--------
 fs/bcachefs/dirent.h                |  1 +
 fs/bcachefs/extents.c               |  3 ++-
 fs/bcachefs/fs-ioctl.c              |  2 +-
 fs/bcachefs/fs.c                    |  3 +++
 fs/bcachefs/inode.c                 | 15 ++++++++++-----
 fs/bcachefs/journal.c               |  8 ++++----
 fs/bcachefs/journal.h               |  1 +
 fs/bcachefs/journal_io.c            |  1 +
 fs/bcachefs/journal_reclaim.c       |  3 +++
 fs/bcachefs/recovery.c              |  2 +-
 fs/bcachefs/reflink.c               |  8 ++++++++
 fs/bcachefs/sysfs.c                 |  8 ++++----
 20 files changed, 84 insertions(+), 38 deletions(-)