[GIT PULL REQUEST] md updates for 4.3

From: Neil Brown
Date: Sat Sep 05 2015 - 05:37:26 EST



Please pull these updates. I've already merged with the 'block' tree
to resolve a few simple conflicts.

Thanks,
NeilBrown


The following changes since commit 1081230b748de8f03f37f80c53dfa89feda9b8de:

Merge branch 'for-4.3/core' of git://git.kernel.dk/linux-block (2015-09-02 13:10:25 -0700)

are available in the git repository at:

git://neil.brown.name/md tags/md/4.3

for you to fetch changes up to e89c6fdf9e0eb1b5a03574d4ca73e83eae8deb91:

Merge linux-block/for-4.3/core into md/for-linux (2015-09-05 11:08:32 +0200)

----------------------------------------------------------------
md updates for 4.3

- An assortment of little fixes, several for minor races only likely
to be hit during testing
- further cluster-md-raid1 development, not ready for real use yet.
- new RAID6 syndrome code for ARM NEON
- fix a race where a write can return before failure of one device
is properly recorded in metadata, so an immediate crash might result
in that write being lost.

----------------------------------------------------------------
Ard Biesheuvel (1):
md/raid6: delta syndrome for ARM NEON

Benjamin Randazzo (1):
md: simplify get_bitmap_file now that "file" is zeroed.

Guoqing Jiang (11):
md-cluster: use %pU to print UUIDs
md-cluster: split recover_slot for future code reuse
md-cluster: transfer the resync ownership to another node
md-cluster: fix deadlock issue on message lock
md-cluster: init completion within lockres_init
md-cluster: add the error check if failed to get dlm lock
md-cluster: init suspend_list and suspend_lock early in join
md-cluster: remove the unused sb_lock
md-cluster: add missed lockres_free
md-cluster: only call complete(&cinfo->completion) when node join cluster
md-cluster: Read the disk bitmap sb and check if it needs recovery

NeilBrown (20):
md/raid0: update queue parameter in a safer location.
md: Keep /proc/mdstat reporting recovery until fully DONE.
md: close some races between setting and checking sync_action.
md/raid5: consider updating reshape_position at start of reshape.
md/raid10: fix a few typos in comments
md/raid5: always set conf->prev_chunk_sectors and ->prev_algo
md/raid5: switch to use conf->chunk_sectors in place of mddev->chunk_sectors where possible
md/raid5: strengthen check on reshape_position at run.
md/raid5: remove incorrect "min_t()" when calculating writepos.
md: set MD_RECOVERY_RECOVER when starting a degraded array.
md: be careful when testing resync_max against curr_resync_completed.
md: sync sync_completed has correct value as recovery finishes.
md/raid5: handle possible race as reshape completes.
md: extend spinlock protection in register_md_cluster_operations
md-cluster: remove inappropriate try_module_get from join()
md/raid1: ensure device failure recorded before write request returns.
md/raid10: ensure device failure recorded before write request returns.
md/raid5: use bio_list for the list of bios to return.
md/raid5: ensure device failure recorded before write request returns.
Merge linux-block/for-4.3/core into md/for-linux

Sasha Levin (1):
md: setup safemode_timer before it's being used

Documentation/md-cluster.txt | 4 +-
drivers/md/md-cluster.c | 159 ++++++++++++++++++++++++++++---------------
drivers/md/md.c | 110 +++++++++++++++++++-----------
drivers/md/raid0.c | 75 ++++++++++----------
drivers/md/raid1.c | 30 +++++++-
drivers/md/raid1.h | 5 ++
drivers/md/raid10.c | 33 ++++++++-
drivers/md/raid10.h | 6 ++
drivers/md/raid5.c | 140 ++++++++++++++++++++++---------------
drivers/md/raid5.h | 5 +-
lib/raid6/neon.c | 13 +++-
lib/raid6/neon.uc | 46 +++++++++++++
12 files changed, 433 insertions(+), 193 deletions(-)

Attachment: signature.asc
Description: PGP signature