[git pull] device mapper changes for 4.8
From: Mike Snitzer
Date: Tue Jul 26 2016 - 12:53:49 EST
Hi Linus,
As noted in my tag message, these DM changes for 4.8 build ontop of
Jens' block 'for-4.8/core' branch.
When pulling these DM change you will encounter a merge conflict with Jens'
'for-4.8/drivers' branch. Like Stephen Rothwell before me, I've resolved
this merge conflict and published a reference resolution; please see the
branch named 'dm-4.8-block-drivers-merge' on my linux-dm.git tree.
Also, there are conflicts once merged with the nvdimm tree, please see:
https://lkml.org/lkml/2016/7/22/41
The following changes since commit 72ef799b3f14f4cb4c56ba3af6e6bdcbae6df368:
block: do not merge requests without consulting with io scheduler (2016-07-20 21:35:12 -0600)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm tags/dm-4.8-changes
for you to fetch changes up to b5ab4a9ba5574430870391e93b125ada31217c4c:
dm: allow bio-based table to be upgraded to bio-based with DAX support (2016-07-20 23:49:52 -0400)
Please pull, thanks.
Mike
----------------------------------------------------------------
- initially based on Jens' 'for-4.8/core' (given all the flag churn) and
later merged with 'for-4.8/core' to pickup the QUEUE_FLAG_DAX commits
that DM depends on to provide its DAX support
- clean up the bio-based vs request-based DM core code by moving the
request-based DM core code out to dm-rq.[hc]
- reinstate bio-based support in the DM multipath target (done with the
idea that fast storage like NVMe over Fabrics could benefit) -- while
preserving support for request_fn and blk-mq request-based DM mpath
- SCSI and DM multipath persistent reservation fixes that were
coordinated with Martin Petersen.
- the DM raid target saw the most extensive change this cycle; it now
provides reshape and takeover support (by layering ontop of the
corresponding MD capabilities)
- DAX support for DM core and the linear, stripe and error targets
- A DM thin-provisioning block discard vs allocation race fix that
addresses potential for corruption
- A stable fix for DM verity-fec's block calculation during decode
- A few cleanups and fixes to DM core and various targets
----------------------------------------------------------------
Arnd Bergmann (1):
dm raid: don't use 'const' in function return
Bart Van Assche (2):
dm crypt: Fix sparse complaints
dm ioctl: Simplify parameter buffer management code
Christoph Hellwig (2):
sd: don't use the ALL_TG_PT bit for reservations
dm: call PR reserve/unreserve on each underlying device
Heinz Mauelshagen (44):
dm raid: rename variable 'ret' to 'r' to conform to other dm code
dm raid: use dm_arg_set API in constructor
dm raid: cleanup / provide infrastructure
dm raid: check constructor arguments for invalid raid level/argument combinations
dm raid: more use of flag testing wrappers
dm raid: use rt_is_raid*() in all appropriate checks
dm raid: introduce extended superblock and new raid types to support takeover/reshaping
dm raid: add new reshaping/raid10 format table line options to parameter parser
dm raid: enhance super_sync() to support new superblock members
dm raid: add raid level takeover support
dm raid: enhance status interface and fixup takeover/raid0
dm raid: inverse check for flags from invalid to valid flags
dm raid: add prerequisite functions and definitions for reshaping
dm raid: add reshaping support to the target
dm raid: update Documentation about reshaping/takeover/additonal RAID types
dm raid: support to change bitmap region size
dm raid: fix failed takeover/reshapes by keeping raid set frozen
dm raid: avoid superfluous memory barriers on static metadata
dm raid: prohibit to pass in both sync and nosync ctr flags
dm raid: the sync_page_io() metadata_op argument is bool
dm raid: reject too many write_mostly devices
dm raid: more restricting data_offset value checks
dm raid: remove bogus comment and fix comment typos
dm raid: enhance comments in takeover checks
dm raid: make rs_set_capacity to work on shrinking reshape
dm raid: fix rs_set_capacity on growing reshape
dm raid: clarify and fix recovery
dm raid: prohibit 'nosync' on new raid6 and reject resize during reshape
dm raid: fix raid10 device size error on out-of-place reshape
dm raid: add comments and fix typos
dm raid: fix new superblock/bitmap creation on disk addition
dm raid: also reject size change during recovery
dm raid: fix typo in write_mostly flag
dm raid: fix ctr memory leaks on error paths
dm raid: fix rebuild and catch bogus sync/resync flags
dm raid: fix rs_is_recovering() to allow for lvextend
dm raid: allow resize during recovery
dm raid: enhance reshape check and factor out reshape setup
dm raid: support delta_disks for raid1, fix table output
dm raid: use rs->raid_disks to avoid memory leaks on free
dm raid: use rdev_for_each in status
dm raid: change logical functions to actually return bool
dm raid: address checkpatch.pl complaints
dm raid: fix random optimal_io_size for raid0
Joe Thornber (2):
dm btree: fix a bug in dm_btree_find_next_single()
dm thin: fix a race condition between discarding and provisioning a block
Mike Snitzer (16):
dm: move request-based code out to dm-rq.[hc]
dm mpath: reinstate bio-based support
dm mpath: remove bio-based bloat from struct dm_mpath_io
dm mpath: add optional "queue_mode" feature
dm raid: tabify appropriate whitespace
dm raid: remove ti_error_* wrappers
dm raid: bump to v1.9.0 and make the extended SB feature flag reflect it
dm raid: rename _argname_by_flag to dm_raid_arg_name_by_flag
dm raid: add missing "dm-raid0" module alias
dm raid: rename _in_range to __within_range
dm raid: remove all the bitops wrappers
dm raid: rename functions that alloc and free struct raid_set
dm raid: various code cleanups
dm rq: check kthread_run return for .request_fn request-based DM
Merge remote-tracking branch 'jens/for-4.8/core' into dm-4.8
dm error: add DAX support
Sami Tolvanen (1):
dm verity fec: fix block calculation
Tahsin Erdogan (1):
dm: fix second blk_delay_queue() parameter to be in msec units not jiffies
Toshi Kani (5):
dm: add infrastructure for DAX support
dm linear: add DAX support
dm stripe: add DAX support
dm snap: add fake origin_direct_access
dm: allow bio-based table to be upgraded to bio-based with DAX support
Documentation/device-mapper/dm-raid.txt | 58 +-
drivers/md/Makefile | 3 +-
drivers/md/dm-builtin.c | 2 +-
drivers/md/dm-core.h | 149 ++
drivers/md/dm-crypt.c | 4 +-
drivers/md/dm-io.c | 2 +-
drivers/md/dm-ioctl.c | 31 +-
drivers/md/dm-kcopyd.c | 2 +-
drivers/md/dm-linear.c | 21 +-
drivers/md/dm-mpath.c | 354 +++-
drivers/md/dm-raid.c | 3073 +++++++++++++++++++++++++------
drivers/md/dm-rq.c | 970 ++++++++++
drivers/md/dm-rq.h | 64 +
drivers/md/dm-snap.c | 8 +
drivers/md/dm-stats.c | 2 +-
drivers/md/dm-stripe.c | 26 +-
drivers/md/dm-sysfs.c | 3 +-
drivers/md/dm-table.c | 114 +-
drivers/md/dm-target.c | 11 +-
drivers/md/dm-thin-metadata.c | 30 +
drivers/md/dm-thin-metadata.h | 3 +
drivers/md/dm-thin.c | 102 +-
drivers/md/dm-verity-fec.c | 4 +-
drivers/md/dm.c | 1499 +++------------
drivers/md/dm.h | 36 +-
drivers/md/persistent-data/dm-btree.c | 9 +-
drivers/scsi/sd.c | 3 +-
include/linux/device-mapper.h | 26 +
include/uapi/linux/dm-ioctl.h | 4 +-
29 files changed, 4638 insertions(+), 1975 deletions(-)
create mode 100644 drivers/md/dm-core.h
create mode 100644 drivers/md/dm-rq.c
create mode 100644 drivers/md/dm-rq.h