[PATCH v2 0/7] Improve zoned (SMR) HDD write throughput

[Damien Le Moal <dlemoal@kernel.org>]

Jens,

This patch series cleans up the zone write plugging code and introduces
the ability to issue all write BIOs from a single context (a kthread)
instead of allowing multiple zones to be written at the same time using
a per zone work. As shown in patch 6, raw block device tests and XFS
tests with an SMR HDD show that this can significantly increase write
throughput (up to 40% over the current zone write plugging).

Changes from v1:
 - Reworked patch 1 to remove the fragile use of refcount_read. Zone
   write plug removal is now more solid by being controlled with
   refcount_dec_and_test().
 - Changed patch 2 to keep the refcount increase before scheduling the
   BIO work but drop that reference if scheduling the work did nothing.
 - Dropped the patch "block: remove BLK_ZONE_WPLUG_UNHASHED" as the
   patch 1 changes make it unnecessary.
 - Modified patch 5 to have the zoned_qd1_writes attribute visible only
   for zoned block devices.
 - Added review tags

Damien Le Moal (7):
  block: fix zone write plug removal
  block: fix zone write plugs refcount handling in
    disk_zone_wplug_schedule_bio_work()
  block: remove disk_zone_is_full()
  block: rename struct gendisk zone_wplugs_lock field
  block: allow submitting all zone writes from a single context
  block: default to QD=1 writes for blk-mq rotational zoned devices
  Documentation: ABI: stable: document the zoned_qd1_writes attribute

 Documentation/ABI/stable/sysfs-block |  15 ++
 block/blk-mq-debugfs.c               |   1 +
 block/blk-sysfs.c                    |  46 +++-
 block/blk-zoned.c                    | 368 +++++++++++++++++----------
 include/linux/blkdev.h               |  10 +-
 5 files changed, 308 insertions(+), 132 deletions(-)

-- 
2.53.0