Hi Linus,
This is the main pull request for block IO related changes for the 4.16
kernel. Nothing major in this pull request, but a good amount of
improvements and fixes all over the map. This pull request contains:
- BFQ improvements, fixes, and cleanups from Angelo, Chiara, and Paolo.
- Support for SMR zones for deadline and mq-deadline from Damien and
Christoph.
- Set of fixes for bcache by way of Michael Lyle, including fixes from
himself, Kent, Rui, Tang, and Coly.
- Series from Matias for lightnvm with fixes from Hans Holmberg, Javier,
and Matias. Mostly centered around pblk, and the removing rrpc 1.2 in
preparation for supporting 2.0.
- A couple of NVMe pull requests from Christoph. Nothing major in here,
just fixes and cleanups, and support for command tracing from
Johannes.
- Support for blk-throttle for tracking reads and writes separately.
From Joseph Qi. A few cleanups/fixes also for blk-throttle from
Weiping.
- Series from Mike Snitzer that enables dm to register its queue more
logically, something that's alwways been problematic on dm since it's
a stacked device.
- Series from Ming cleaning up some of the bio accessor use, in
preparation for supporting multipage bvecs.
- Various fixes from Ming closing up holes around queue mapping and
quiescing.
- BSD partition fix from Richard Narron, fixing a problem where we can't
mount newer (10/11) FreeBSD partitions.
- Series from Tejun reworking blk-mq timeout handling. The previous
scheme relied on atomic bits, but it had races where we would think a
request had timed out if it to reused at the wrong time.
- null_blk now supports faking timeouts, to enable us to better exercise
and test that functionality separately. From me.
- Kill the separate atomic poll bit in the request struct. After this,
we don't use the atomic bits on blk-mq anymore at all. From me.
- sgl_alloc/free helpers from Bart.
- Heavily contended tag case scalability improvement from me.
- Various little fixes and cleanups from Arnd, Bart, Corentin, Douglas,
Eryu, Goldwyn, and myself.
Note that you'll get a single merge conflict when pulling this in, due
to the change:
commit 4ccafe032005e9b96acbef2e389a4de5b1254add
Author: Jens Axboe <axboe@xxxxxxxxx>
Date: Wed Dec 20 13:13:58 2017 -0700
block: unalign call_single_data in struct request
that went into your tree after I had forked off for-4.16/block. The
resolution is trivial, just ensure that the moved 'csd' struct member
retains the type of struct __call_single_data, not call_single_data_t.
Please pull!
git://git.kernel.dk/linux-block.git for-4.16/block
----------------------------------------------------------------
Angelo Ruocco (2):
block, bfq: check low_latency flag in bfq_bfqq_save_state()
block, bfq: remove superfluous check in queue-merging setup
Arnd Bergmann (3):
DAC960: split up ioctl function to reduce stack size
null_blk: remove explicit 'select FAULT_INJECTION'
blkcg: simplify statistic accumulation code
Bart Van Assche (20):
pktcdvd: Fix pkt_setup_dev() error path
pktcdvd: Fix a recently introduced NULL pointer dereference
lib/scatterlist: Introduce sgl_alloc() and sgl_free()
crypto: scompress - use sgl_alloc() and sgl_free()
nvmet/fc: Use sgl_alloc() and sgl_free()
nvmet/rdma: Use sgl_alloc() and sgl_free()
target: Use sgl_alloc_order() and sgl_free()
blk-mq: Fix spelling in a source code comment
block: Fix kernel-doc warnings reported when building with W=1
blk-mq: Explain when 'active_queues' is decremented
blk-mq: Add locking annotations to hctx_lock() and hctx_unlock()
blk-mq: Reduce the number of if-statements in blk_mq_mark_tag_wait()
block: Fix __bio_integrity_endio() documentation
block: Unexport elv_register_queue() and elv_unregister_queue()
block: Document scheduler modification locking requirements
block: Protect less code with sysfs_lock in blk_{un,}register_queue()
lib/scatterlist: Fix chaining support in sgl_alloc_order()
blk-mq: Rename blk_mq_request_direct_issue() into blk_mq_request_issue_directly()
blk-mq: Avoid that blk_mq_delay_run_hw_queue() introduces unintended delays
block: Remove kblockd_schedule_delayed_work{,_on}()
Chiara Bruschi (1):
block, bfq: fix occurrences of request finish method's old name
Christoph Hellwig (5):
block: introduce zoned block devices zone write locking
genirq/affinity: assign vectors to all possible CPUs
blk-mq: simplify queue mapping & schedule with each possisble CPU
nvme-pci: clean up CMB initialization
nvme-pci: clean up SMBSZ bit definitions
Coly Li (2):
bcache: reduce cache_set devices iteration by devices_max_used
bcache: fix misleading error message in bch_count_io_errors()
Corentin Labbe (1):
block: remove smart1,2.h
Damien Le Moal (4):
mq-deadline: Introduce dispatch helpers
mq-deadline: Introduce zone locking support
deadline-iosched: Introduce dispatch helpers
deadline-iosched: Introduce zone locking support
Douglas Gilbert (1):
blk_rq_map_user_iov: fix error override
Eryu Guan (1):
blk-mq-debugfs: don't allow write on attributes with seq_operations set
Goldwyn Rodrigues (1):
block: Set BIO_TRACE_COMPLETION on new bio during split
Hans Holmberg (5):
lightnvm: pblk: refactor emeta consistency check
lightnvm: pblk: rename sync_point to flush_point
lightnvm: pblk: clear flush point on completed writes
lightnvm: pblk: prevent premature sync point resets
lightnvm: pblk: remove pblk_gc_stop
Ilya Dryomov (2):
block: fail op_is_write() requests to read-only partitions
block: add bdev_read_only() checks to common helpers
Israel Rukshin (3):
nvmet: fix error flow in nvmet_alloc_ctrl()
nvmet: rearrange nvmet_ctrl_free()
nvme: fix subsystem multiple controllers support check
James Smart (7):
nvme_fcloop: fix abort race condition
nvme_fcloop: disassocate local port structs
nvme_fcloop: rework to remove xxx_IN_ISR feature flags
nvme_fcloop: refactor host/target io job access
nvmet-fc: cleanup nvmet add_port/remove_port
nvme-fc: fix rogue admin cmds stalling teardown
nvme-fc: correct hang in nvme_ns_remove()
Javier González (13):
lightnvm: remove unnecessary field from nvm_rq
lightnvm: refactor target type lookup
lightnvm: guarantee target unique name across devs.
lightnvm: pblk: compress and reorder helper functions
lightnvm: pblk: remove pblk_for_each_lun helper
lightnvm: pblk: use exact free block counter in RL
lightnvm: set target over-provision on create ioctl
lightnvm: pblk: ignore high ecc errors on recovery
lightnvm: pblk: do not log recovery read errors
lightnvm: pblk: ensure kthread alloc. before kicking it
lightnvm: pblk: free write buffer on init failure
lightnvm: pblk: print instance name on instance info
lightnvm: pblk: add iostat support
Jens Axboe (17):
blk-mq: improve heavily contended tag case
mq-deadline: make it clear that __dd_dispatch_request() works on all hw queues
Merge branch 'nvme-4.16' of git://git.infradead.org/nvme into for-4.16/block
blk-mq: move hctx lock/unlock into a helper
blk-mq: silence false positive warnings in hctx_unlock()
bfq-iosched: don't call bfqg_and_blkg_put for !CONFIG_BFQ_GROUP_IOSCHED
null_blk: wire up timeouts
null_blk: add option for managing IO timeouts
blk-mq: add a few missing debugfs RQF_ flags
block: remove REQ_ATOM_POLL_SLEPT
block: add accessors for setting/querying request deadline
block: convert REQ_ATOM_COMPLETE to stealing rq->__deadline bit
block: rearrange a few request fields for better cache layout
blk-mq: add missing RQF_STARTED to debugfs
blk-mq: fix bad clear of RQF_MQ_INFLIGHT in blk_mq_ct_ctx_init()
Merge branch 'nvme-4.16' of git://git.infradead.org/nvme into for-4.16/block
Merge branch 'nvme-4.16' of git://git.infradead.org/nvme into for-4.16/block
Jianchao Wang (2):
nvme-pci: fix NULL pointer reference in nvme_alloc_ns
nvme-pci: introduce RECONNECTING state to mark initializing procedure
Johannes Thumshirn (4):
bsg: use pr_debug instead of hand crafted macros
nvme: don't free uuid pointer before printing it
nvme: add tracepoint for nvme_setup_cmd
nvme: add tracepoint for nvme_complete_rq
Joseph Qi (1):
blk-throttle: track read and write request individually
Keith Busch (7):
nvme: Add more command status translation
nvme/multipath: Consult blk_status_t for failover
block: Provide blk_status_t decoding for path errors
nvme/multipath: Use blk_path_error
dm mpath: Use blk_path_error
nvme-pci: Fix queue double allocations
nvme-pci: Suspend queues after deleting them
Kent Overstreet (2):
bcache: Fix, improve efficiency of closure_sync()
bcache: mark closure_sync() __sched
Liu Bo (1):
blk-mq: remove confusing comment of blk_mq_sched_dispatch_requests
Matias Bjørling (7):
null_blk: remove lightnvm support
lightnvm: remove rrpc
lightnvm: use internal pblk methods
lightnvm: remove hybrid ocssd 1.2 support
lightnvm: remove lower page tables
lightnvm: make geometry structures 2.0 ready
lightnvm: pblk: refactor pblk_ppa_comp function
Max Gurtovoy (2):
nvme: modify the debug level for setting shutdown timeout
nvme-rdma: remove redundant boolean for inline_data
Michael Lyle (4):
bcache: writeback: properly order backing device IO
bcache: allow quick writeback when backing idle
bcache: fix writeback target calc on large devices
bcache: closures: move control bits one bit right
Mike Snitzer (6):
block: only bdi_unregister() in del_gendisk() if !GENHD_FL_HIDDEN
block: properly protect the 'queue' kobj in blk_unregister_queue
block: allow gendisk's request_queue registration to be deferred
dm: fix incomplete request_queue initialization
blk-mq: factor out a few helpers from __blk_mq_try_issue_directly
blk-mq-sched: remove unused 'can_block' arg from blk_mq_sched_insert_request
Ming Lei (24):
block: introduce bio helpers for converting to multipage bvec
block: convert to bio_first_bvec_all & bio_first_page_all
fs: convert to bio_last_bvec_all()
block: bounce: avoid direct access to bvec table
block: bounce: don't access bio->bi_io_vec in copy_to_high_bio_irq
dm: limit the max bio size as BIO_MAX_PAGES * PAGE_SIZE
bcache: comment on direct access to bvec table
block: move bio_alloc_pages() to bcache
btrfs: avoid access to .bi_vcnt directly
btrfs: avoid accessing bvec table directly for a cloned bio
dm-crypt: don't clear bvec->bv_page in crypt_free_buffer_pages()
blk-merge: compute bio->bi_seg_front_size efficiently
block: blk-merge: try to make front segments in full size
block: blk-merge: remove unnecessary check
blk-mq: quiesce queue before freeing queue
blk-mq: quiesce queue during switching io sched and updating nr_requests
blk-mq: avoid to map CPU into stale hw queue
blk-mq: fix race between updating nr_hw_queues and switching io sched
blk-mq: fix kernel oops in blk_mq_tag_idle()
Revert "block: blk-merge: try to make front segments in full size"
blk-mq: make sure hctx->next_cpu is set correctly
blk-mq: turn WARN_ON in __blk_mq_run_hw_queue into printk
blk-mq: improve DM's blk-mq IO merging via blk_insert_cloned_request feedback
blk-mq: don't dispatch request in blk_mq_request_direct_issue if queue is busy
Minwoo Im (2):
nvme-pci: remove an unnecessary initialization in HMB code
nvme: fix comment typos in nvme_create_io_queues
Nitzan Carmi (1):
nvme: take refcount on transport module
Paolo Bonzini (1):
block: silently forbid sending any ioctl to a partition
Paolo Valente (9):
block, bfq: increase threshold to deem I/O as random
block, bfq: add missing rq_pos_tree update on rq removal
block, bfq: let a queue be merged only shortly after starting I/O
block, bfq: consider also past I/O in soft real-time detection
block, bfq: remove batches of confusing ifdefs
block, bfq: put async queues for root bfq groups too
block, bfq: release oom-queue ref to root group on exit
block, bfq: limit tags for writes and async I/O
block, bfq: limit sectors served with interactive weight raising
Richard Narron (1):
partitions/msdos: Unable to mount UFS 44bsd partitions
Roland Dreier (1):
nvme-fabrics: fix memory leak when parsing host ID option
Roy Shterman (2):
nvme-fabrics: protect against module unload during create_ctrl
nvme: host delete_work and reset_work on separate workqueues
Rui Hua (1):
bcache: ret IOERR when read meets metadata error
Sagi Grimberg (7):
nvmet-rdma: removed queue cleanup from module exit
nvmet-rdma: lowering log level for chatty debug messages
nvmet: lower log level for each queue creation
nvme-pci: don't open-code nvme_reset_ctrl
nvme-pci: serialize pci resets
nvme-pci: allocate device queues storage space at probe
nvmet: release a ns reference in nvmet_req_uninit if needed
Tang Junhui (3):
bcache: stop writeback thread after detaching
bcache: segregate flash only volume write streams
bcache: fix wrong return value in bch_debug_init()
Tejun Heo (7):
blk-mq: protect completion path with RCU
blk-mq: replace timeout synchronization with a RCU and generation based scheme
blk-mq: use blk_mq_rq_state() instead of testing REQ_ATOM_COMPLETE
blk-mq: make blk_abort_request() trigger timeout path
blk-mq: remove REQ_ATOM_COMPLETE usages from blk-mq
blk-mq: remove REQ_ATOM_STARTED
blk-mq: rename blk_mq_hw_ctx->queue_rq_srcu to ->srcu
Tina Ruchandani (1):
aoe: use ktime_t instead of timeval
Vasyl Gomonovych (1):
bcache: Use PTR_ERR_OR_ZERO()
Wang Long (1):
writeback: update comment in inode_io_list_move_locked
Zhai Zhaoxuan (1):
bcache: fix unmatched generic_end_io_acct() & generic_start_io_acct()
weiping zhang (2):
blk-throttle: export io_serviced_recursive, io_service_bytes_recursive
blk-throttle: use queue_is_rq_based
block/bfq-cgroup.c | 7 +-
block/bfq-iosched.c | 529 ++++++++---
block/bfq-iosched.h | 19 +
block/bfq-wf2q.c | 7 +
block/bio-integrity.c | 1 -
block/bio.c | 30 +-
block/blk-core.c | 87 +-
block/blk-exec.c | 2 +-
block/blk-lib.c | 12 +
block/blk-map.c | 4 +-
block/blk-merge.c | 13 +-
block/blk-mq-debugfs.c | 22 +-
block/blk-mq-sched.c | 3 +-
block/blk-mq-sched.h | 2 +-
block/blk-mq-sysfs.c | 9 +-
block/blk-mq-tag.c | 13 +-
block/blk-mq.c | 667 ++++++++-----
block/blk-mq.h | 52 +-
block/blk-sysfs.c | 47 +-
block/blk-throttle.c | 146 +--
block/blk-timeout.c | 26 +-
block/blk-zoned.c | 42 +
block/blk.h | 46 +-
block/bounce.c | 33 +-
block/bsg-lib.c | 3 +-
block/bsg.c | 40 +-
block/deadline-iosched.c | 114 ++-
block/elevator.c | 12 +-
block/genhd.c | 23 +-
block/mq-deadline.c | 141 ++-
block/partitions/msdos.c | 4 +-
block/scsi_ioctl.c | 34 +-
crypto/Kconfig | 1 +
crypto/scompress.c | 51 +-
drivers/block/DAC960.c | 160 ++--
drivers/block/Kconfig | 4 +
drivers/block/aoe/aoe.h | 3 +-
drivers/block/aoe/aoecmd.c | 48 +-
drivers/block/drbd/drbd_bitmap.c | 2 +-
drivers/block/null_blk.c | 290 ++----
drivers/block/pktcdvd.c | 12 +-
drivers/block/smart1,2.h | 278 ------
drivers/block/zram/zram_drv.c | 2 +-
drivers/lightnvm/Kconfig | 7 -
drivers/lightnvm/Makefile | 1 -
drivers/lightnvm/core.c | 462 ++++-----
drivers/lightnvm/pblk-cache.c | 5 +
drivers/lightnvm/pblk-core.c | 55 +-
drivers/lightnvm/pblk-gc.c | 23 +-
drivers/lightnvm/pblk-init.c | 104 +-
drivers/lightnvm/pblk-map.c | 2 +-
drivers/lightnvm/pblk-rb.c | 111 ++-
drivers/lightnvm/pblk-read.c | 35 +-
drivers/lightnvm/pblk-recovery.c | 43 +-
drivers/lightnvm/pblk-rl.c | 54 +-
drivers/lightnvm/pblk-sysfs.c | 15 +-
drivers/lightnvm/pblk-write.c | 23 +-
drivers/lightnvm/pblk.h | 163 ++--
drivers/lightnvm/rrpc.c | 1625 --------------------------------
drivers/lightnvm/rrpc.h | 290 ------
drivers/md/bcache/alloc.c | 19 +-
drivers/md/bcache/bcache.h | 24 +-
drivers/md/bcache/btree.c | 10 +-
drivers/md/bcache/closure.c | 47 +-
drivers/md/bcache/closure.h | 60 +-
drivers/md/bcache/debug.c | 7 +-
drivers/md/bcache/io.c | 13 +-
drivers/md/bcache/movinggc.c | 2 +-
drivers/md/bcache/request.c | 29 +-
drivers/md/bcache/super.c | 27 +-
drivers/md/bcache/util.c | 34 +
drivers/md/bcache/util.h | 1 +
drivers/md/bcache/writeback.c | 203 +++-
drivers/md/bcache/writeback.h | 12 +-
drivers/md/dm-crypt.c | 1 -
drivers/md/dm-mpath.c | 19 +-
drivers/md/dm-rq.c | 28 +-
drivers/md/dm.c | 21 +-
drivers/nvme/host/Makefile | 4 +
drivers/nvme/host/core.c | 134 ++-
drivers/nvme/host/fabrics.c | 22 +-
drivers/nvme/host/fabrics.h | 2 +
drivers/nvme/host/fc.c | 7 +
drivers/nvme/host/lightnvm.c | 185 +---
drivers/nvme/host/multipath.c | 44 +-
drivers/nvme/host/nvme.h | 9 +-
drivers/nvme/host/pci.c | 216 ++---
drivers/nvme/host/rdma.c | 6 +-
drivers/nvme/host/trace.c | 130 +++
drivers/nvme/host/trace.h | 165 ++++
drivers/nvme/target/Kconfig | 2 +
drivers/nvme/target/core.c | 14 +-
drivers/nvme/target/fabrics-cmd.c | 2 +-
drivers/nvme/target/fc.c | 60 +-
drivers/nvme/target/fcloop.c | 244 ++++-
drivers/nvme/target/loop.c | 3 +-
drivers/nvme/target/rdma.c | 83 +-
drivers/target/Kconfig | 1 +
drivers/target/target_core_transport.c | 46 +-
fs/btrfs/compression.c | 4 +-
fs/btrfs/extent_io.c | 11 +-
fs/btrfs/extent_io.h | 2 +-
fs/btrfs/inode.c | 8 +-
fs/buffer.c | 2 +-
fs/f2fs/data.c | 2 +-
fs/fs-writeback.c | 2 +-
include/linux/bio.h | 24 +-
include/linux/blk-cgroup.h | 8 +-
include/linux/blk-mq.h | 3 +-
include/linux/blk_types.h | 28 +
include/linux/blkdev.h | 172 +++-
include/linux/bvec.h | 9 +
include/linux/elevator.h | 2 -
include/linux/genhd.h | 5 +
include/linux/lightnvm.h | 125 +--
include/linux/nvme.h | 22 +-
include/linux/scatterlist.h | 11 +
include/uapi/linux/lightnvm.h | 9 +
kernel/irq/affinity.c | 30 +-
kernel/power/swap.c | 2 +-
lib/Kconfig | 4 +
lib/sbitmap.c | 2 +-
lib/scatterlist.c | 127 +++
mm/page_io.c | 4 +-
124 files changed, 3884 insertions(+), 4729 deletions(-)
delete mode 100644 drivers/block/smart1,2.h
delete mode 100644 drivers/lightnvm/rrpc.c
delete mode 100644 drivers/lightnvm/rrpc.h
create mode 100644 drivers/nvme/host/trace.c
create mode 100644 drivers/nvme/host/trace.h
--
Jens Axboe
