WSL2-Linux-Kernel/block
Shaohua Li 9ba52e5812 blk-mq: don't lose requests if a stopped queue restarts
Normally if driver is busy to dispatch a request the logic is like below:
block layer:					driver:
	__blk_mq_run_hw_queue
a.						blk_mq_stop_hw_queue
b.	rq add to ctx->dispatch

later:
1.						blk_mq_start_hw_queue
2.	__blk_mq_run_hw_queue

But it's possible step 1-2 runs between a and b. And since rq isn't in
ctx->dispatch yet, step 2 will not run rq. The rq might get lost if
there are no subsequent requests kick in.

Signed-off-by: Shaohua Li <shli@fb.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
2015-05-04 14:32:48 -06:00
..
partitions Merge branch 'for-3.20/core' of git://git.kernel.dk/linux-block 2015-02-12 14:13:23 -08:00
Kconfig block: Add T10 Protection Information functions 2014-09-27 09:14:59 -06:00
Kconfig.iosched
Makefile block: Add T10 Protection Information functions 2014-09-27 09:14:59 -06:00
bio-integrity.c block: fix regression where bio_integrity_process uses wrong bio_vec iterator 2014-12-02 08:15:21 -07:00
bio.c block: rewrite and split __bio_copy_iov() 2015-02-05 09:30:44 -07:00
blk-cgroup.c blkcg: remove blkcg->id 2014-09-08 09:55:37 -06:00
blk-cgroup.h blkcg: remove blkcg->id 2014-09-08 09:55:37 -06:00
blk-core.c block: destroy bdi before blockdev is unregistered. 2015-04-27 10:27:20 -06:00
blk-exec.c blk-mq: avoid infinite recursion with the FUA flag 2014-09-22 11:55:19 -06:00
blk-flush.c blk-mq: support per-distpatch_queue flush machinery 2014-09-25 15:22:45 -06:00
blk-integrity.c block: Don't merge requests if integrity flags differ 2014-09-27 09:14:57 -06:00
blk-ioc.c
blk-iopoll.c
blk-lib.c block: Quiesce zeroout wrapper 2015-02-05 10:14:54 -07:00
blk-map.c blk_rq_map_user(): use import_single_range() 2015-04-11 22:27:13 -04:00
blk-merge.c Fix bug in blk_rq_merge_ok 2015-03-20 08:50:41 -06:00
blk-mq-cpu.c
blk-mq-cpumap.c blk-mq: Use all available hardware queues 2014-12-09 09:08:21 -07:00
blk-mq-sysfs.c blk-mq: add blk_mq_init_allocated_queue and export blk_mq_register_disk 2015-03-13 08:26:53 -06:00
blk-mq-tag.c blkmq: Fix NULL pointer deref when all reserved tags in 2015-03-18 17:06:18 -06:00
blk-mq-tag.h blk-mq: add tag allocation policy 2015-01-23 14:18:00 -07:00
blk-mq.c blk-mq: don't lose requests if a stopped queue restarts 2015-05-04 14:32:48 -06:00
blk-mq.h blk-mq: release mq's kobjects in blk_release_queue() 2015-01-29 08:30:51 -08:00
blk-settings.c block: fix blk_stack_limits() regression due to lcm() change 2015-03-31 09:45:50 -06:00
blk-softirq.c
blk-sysfs.c block: destroy bdi before blockdev is unregistered. 2015-04-27 10:27:20 -06:00
blk-tag.c block: support different tag allocation policy 2015-01-23 14:15:46 -07:00
blk-throttle.c blk-throttle: check stats_cpu before reading it from sysfs 2015-02-20 22:11:58 -08:00
blk-timeout.c blk-mq: Allow requests to never expire 2015-01-08 08:59:01 -07:00
blk.h blk-mq: support per-distpatch_queue flush machinery 2014-09-25 15:22:45 -06:00
bounce.c block:bounce: fix call inc_|dec_zone_page_state on different pages confuse value of NR_BOUNCE 2015-04-27 09:24:07 -06:00
bsg-lib.c
bsg.c block: Simplify bsg complete all 2015-02-04 09:57:52 -07:00
cfq-iosched.c cfq-iosched: handle failure of cfq group allocation 2015-02-09 10:22:39 -07:00
cmdline-parser.c
compat_ioctl.c block, bdi: an active gendisk always has a request_queue associated with it 2014-09-08 10:00:35 -06:00
deadline-iosched.c
elevator.c elevator: fix double release of elevator module 2015-04-23 10:47:44 -06:00
genhd.c genhd: check for int overflow in disk_expand_part_tbl() 2014-11-19 13:09:07 -07:00
ioctl.c block: Add discard flag to blkdev_issue_zeroout() function 2015-01-21 10:41:46 -07:00
ioprio.c block: Fix computation of merged request priority 2014-10-31 08:30:43 -06:00
noop-iosched.c
partition-generic.c block: Fix dev_t minor allocation lifetime 2014-09-03 15:01:02 -06:00
scsi_ioctl.c sg_io(): use import_iovec() 2015-04-11 22:27:13 -04:00
t10-pi.c block: Add T10 Protection Information functions 2014-09-27 09:14:59 -06:00