acrn-kernel/block
Jens Axboe 492e0aba08 block: treat poll queue enter similarly to timeouts
commit 33391eecd6 upstream.

We ran into an issue where a production workload would randomly grind to
a halt and not continue until the pending IO had timed out. This turned
out to be a complicated interaction between queue freezing and polled
IO:

1) You have an application that does polled IO. At any point in time,
   there may be polled IO pending.

2) You have a monitoring application that issues a passthrough command,
   which is marked with side effects such that it needs to freeze the
   queue.

3) Passthrough command is started, which calls blk_freeze_queue_start()
   on the device. At this point the queue is marked frozen, and any
   attempt to enter the queue will fail (for non-blocking) or block.

4) Now the driver calls blk_mq_freeze_queue_wait(), which will return
   when the queue is quiesced and pending IO has completed.

5) The pending IO is polled IO, but any attempt to poll IO through the
   normal iocb_bio_iopoll() -> bio_poll() will fail when it gets to
   bio_queue_enter() as the queue is frozen. Rather than poll and
   complete IO, the polling threads will sit in a tight loop attempting
   to poll, but failing to enter the queue to do so.

The end result is that progress for either application will be stalled
until all pending polled IO has timed out. This causes obvious huge
latency issues for the application doing polled IO, but also long delays
for passthrough command.

Fix this by treating queue enter for polled IO just like we do for
timeouts. This allows quick quiesce of the queue as we still poll and
complete this IO, while still disallowing queueing up new IO.

Reviewed-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2024-02-16 19:06:31 +01:00
..
partitions block: Move checking GENHD_FL_NO_PART to bdev_add_partition() 2024-01-31 16:17:11 -08:00
Kconfig
Kconfig.iosched
Makefile
badblocks.c
bdev.c block: update the stable_writes flag in bdev_add 2024-01-10 17:10:32 +01:00
bfq-cgroup.c
bfq-iosched.c block, bfq: Fix division by zero error on zero wsum 2023-05-24 17:32:38 +01:00
bfq-iosched.h
bfq-wf2q.c
bio-integrity.c block: factor out a bvec_set_page helper 2023-09-23 11:11:08 +02:00
bio.c block: prevent an integer overflow in bvec_try_merge_hw_page 2024-02-05 20:12:53 +00:00
blk-cgroup-fc-appid.c
blk-cgroup-rwstat.c
blk-cgroup-rwstat.h
blk-cgroup.c blk-cgroup: bypass blkcg_deactivate_policy after destroying 2023-12-20 17:00:21 +01:00
blk-cgroup.h
blk-core.c block: treat poll queue enter similarly to timeouts 2024-02-16 19:06:31 +01:00
blk-crypto-fallback.c blk-crypto: dynamically allocate fallback profile 2023-08-23 17:52:39 +02:00
blk-crypto-internal.h blk-mq: release crypto keyslot before reporting I/O complete 2023-05-11 23:03:00 +09:00
blk-crypto-profile.c blk-crypto: use dynamic lock class for blk_crypto_profile::lock 2023-07-23 13:49:21 +02:00
blk-crypto-sysfs.c
blk-crypto.c blk-crypto: make blk_crypto_evict_key() more robust 2023-05-11 23:03:01 +09:00
blk-flush.c
blk-ia-ranges.c
blk-integrity.c
blk-ioc.c
blk-iocost.c blk-iocost: Fix an UBSAN shift-out-of-bounds warning 2024-02-16 19:06:29 +01:00
blk-iolatency.c
blk-ioprio.c
blk-ioprio.h
blk-lib.c
blk-map.c block: fix bio-cache for passthru IO 2023-06-05 09:26:21 +02:00
blk-merge.c blk-mq: release crypto keyslot before reporting I/O complete 2023-05-11 23:03:00 +09:00
blk-mq-cpumap.c
blk-mq-debugfs-zoned.c
blk-mq-debugfs.c blk-mq: fix potential io hang by wrong 'wake_batch' 2023-07-19 16:20:55 +02:00
blk-mq-debugfs.h
blk-mq-pci.c
blk-mq-rdma.c
blk-mq-sched.c
blk-mq-sched.h
blk-mq-sysfs.c
blk-mq-tag.c blk-mq: fix potential io hang by wrong 'wake_batch' 2023-07-19 16:20:55 +02:00
blk-mq-tag.h
blk-mq-virtio.c
blk-mq.c blk-mq: fix IO hang from sbitmap wakeup race 2024-02-05 20:12:59 +00:00
blk-mq.h blk-mq: fix potential io hang by wrong 'wake_batch' 2023-07-19 16:20:55 +02:00
blk-pm.c
blk-pm.h
blk-rq-qos.c
blk-rq-qos.h
blk-settings.c block: make BLK_DEF_MAX_SECTORS unsigned 2024-01-25 15:27:30 -08:00
blk-stat.c blk-stat: fix QUEUE_FLAG_STATS clear 2023-05-11 23:03:00 +09:00
blk-stat.h
blk-sysfs.c block: fix use-after-free of q->q_usage_counter 2023-10-10 22:00:37 +02:00
blk-throttle.c blk-throttle: fix lockdep warning of "cgroup_mutex or RCU read lock required!" 2023-12-20 17:00:21 +01:00
blk-throttle.h
blk-timeout.c
blk-wbt.c
blk-wbt.h
blk-zoned.c
blk.h
bounce.c
bsg-lib.c
bsg.c
disk-events.c block: increment diskseq on all media change events 2023-07-19 16:21:47 +02:00
elevator.c
elevator.h
fops.c block: Don't invalidate pagecache for invalid falloc modes 2024-01-10 17:10:20 +01:00
genhd.c block: add check of 'minors' and 'first_minor' in device_add_disk() 2024-01-25 15:27:28 -08:00
holder.c
ioctl.c block: Move checking GENHD_FL_NO_PART to bdev_add_partition() 2024-01-31 16:17:11 -08:00
ioprio.c
kyber-iosched.c
mq-deadline.c block/mq-deadline: use correct way to throttling write requests 2023-09-13 09:42:42 +02:00
opal_proto.h
sed-opal.c
t10-pi.c